Hi folks,
I've attached a tarball containing the patch series that adds
metadata CRC support to xfsprogs. It is still very much a work in
progress, but is sufficient to start running xfstests on CRC enable
filesystems.
THere is much missing functionality, especially with respect to
directly/attribute modifications. Only mkfs and xfs_repair have been
updated to explicitly support CRC on directory/sttribute blocks,
though all the other tools will stil run the result might be a
little unpredictable.
The patch series is as follows:
Dave Chinner (26):
xfsprogs: sync code to current kernel code
xfsprogs: update libxfs to 3.9-rc1 + xfsdev
xfsprogs: add CRC32c infrastructure
xfsprogs: updata libxlog to current kernel code
libxfs: add crc format changes to generic btrees
xfsprogs: add crc format chagnes to ag headers
libxfs: change quota buffer formats
libxfs: add version 3 inode support
libxfs: add support for crc headers on remote symlinks
xfs: add CRC checks to block format directory blocks
xfs: add CRC checking to dir2 free blocks
xfs: add CRC checking to dir2 data blocks
xfs: add CRC checking to dir2 leaf blocks
xfs: shortform directory offsets change for dir3 format
xfs: add CRCs to dir2/da node blocks
xfs: add CRCs to attr leaf blocks
xfs: split remote attribute code out
xfs: add CRC protection to remote attributes
xfs: add buffer types to directory and attribute buffers
xfs: buffer type overruns blf_flags field
xfs: add CRC checks to the superblock
xfs: implement extended feature masks
xfsprogs: introduce CRC support into mkfs.xfs
xfsprogs: add crc format support to repair
xfsprogs: add crc format support to db
xfs_repair: update for dir/attr crc format changes.
Diffstat is below [1].
The changes are roughly:
- sync to current kernel code
- apply userspace converted version of kernel patches
- update xfsprogs utilities to be aware of CRCs
Much of the last 4 patches needs to be split up and redirected back
to the patches that introduce the bugs that are fixed or
conversions that are missed.
Repair needs support for remote attribute and symlink block CRC
support, as well as verifying the new feature masks. mkfs needs to
handle thenew feature masks correctly, too. xfs_db needs to be able
to display all crc enabled formats, and needs to ensure all
read-only functions work properly on CRC enabled filesystems. I
think that metadump will also need some attention. So there's plenty
of work yet to do, though it is already usable.
For non-CRC filesystems, the utilities appear to work without
regressions in xfstests. Most of the testing I have done is on
non-CRC filesytsems.
For CRC enabled filesystems, xfstests will run for a short while
before the kernel bugs out with an attribute corruption. Over the
next week I hope to get the kernel code to the point where it will
run most of xfstests with CRCs enabled:
$ sudo MKFS_OPTIONS="-m crc=1" ./check -g auto
FSTYP -- xfs (debug)
PLATFORM -- Linux/x86_64 test-4 3.9.0-rc4-dgc+
MKFS_OPTIONS -- -f -m crc=1 /dev/vdb
MOUNT_OPTIONS -- /dev/vdb /mnt/scr
generic/001 4s
generic/002 0s
generic/005 0s
generic/006 1s
generic/007 1s
generic/010 [not run] dbtest was not built for this platform
generic/011 1s
generic/013 12s
generic/014 3s
generic/015 0s
generic/020
Message from syslogd@test-4 at Apr 13 00:35:45 ...
kernel:[ 451.268820] XFS: Assertion failed: ichdr->firstused >= ichdr->count * sizeof(xfs_attr_leaf_entry_t) + xfs_attr3_leaf_hdr_size(leaf), file: fs/xfs/xfs_attr_leaf.c, line: 1427
$
Finally, I'm sending this update out as a quilt series style tarball
rather than email patches as some of the patches are too large for
the list and will get blocked. Hence you get patches and a series
file that I've just exported from my current guilt tree.
Obligatory warning: If you enable CRCs, the code will eat your
data. Use this only only for testing with data you don't care about.
Comments, flames and testing results all welcome.
Cheers,
Dave.
[1] diffstat:
copy/xfs_copy.c | 5 +-
db/agf.c | 3 +
db/agfl.c | 16 +
db/agfl.h | 2 +
db/agi.c | 3 +
db/attr.c | 20 +-
db/btblock.c | 145 +
db/btblock.h | 10 +
db/check.c | 55 +-
db/dir.c | 8 +-
db/dir.h | 6 +
db/dir2.c | 16 +-
db/dir2.h | 32 +
db/dir2sf.c | 24 +-
db/dirshort.c | 1 +
db/field.c | 16 +
db/field.h | 8 +
db/freesp.c | 11 +-
db/init.c | 4 +
db/inode.c | 27 +
db/inode.h | 3 +
db/metadump.c | 27 +-
db/sb.c | 40 +-
db/type.c | 35 +-
db/type.h | 3 +-
estimate/xfs_estimate.c | 3 +
fsr/xfs_fsr.c | 2 +-
include/Makefile | 13 +-
include/libxfs.h | 222 +-
include/libxlog.h | 34 +-
include/linux.h | 1 +
include/platform_defs.h.in | 5 +
include/swab.h | 38 +
include/xfs.h | 1 +
include/xfs_ag.h | 81 +-
include/xfs_alloc.h | 57 +-
include/xfs_alloc_btree.h | 29 +-
include/xfs_arch.h | 16 +
include/xfs_attr_leaf.h | 135 +-
include/xfs_attr_remote.h | 52 +
include/xfs_bmap.h | 325 +-
include/xfs_bmap_btree.h | 21 +-
include/xfs_btree.h | 87 +-
include/xfs_buf_item.h | 117 +-
include/xfs_cksum.h | 63 +
include/xfs_da_btree.h | 185 +-
include/xfs_dinode.h | 47 +-
include/xfs_dir1.h | 13 +
include/xfs_dir2.h | 95 +-
include/xfs_dir2_block.h | 92 -
include/xfs_dir2_data.h | 184 -
include/xfs_dir2_format.h | 836 +++++
include/xfs_dir2_leaf.h | 253 --
include/xfs_dir2_node.h | 100 -
include/xfs_dir2_sf.h | 171 -
include/xfs_fs.h | 38 +-
include/xfs_ialloc.h | 17 +-
include/xfs_ialloc_btree.h | 12 +-
include/xfs_inode.h | 263 +-
include/xfs_inode_item.h | 22 +-
include/xfs_inum.h | 16 -
include/xfs_log.h | 31 +-
include/xfs_log_priv.h | 101 +-
include/xfs_mount.h | 48 +-
include/xfs_quota.h | 58 +-
include/xfs_rtalloc.h | 4 +-
include/xfs_sb.h | 172 +-
include/xfs_symlink.h | 45 +
include/xfs_trace.h | 69 +-
include/xfs_trans.h | 88 +-
include/xfs_types.h | 25 +-
libxfs/Makefile | 26 +-
libxfs/crc32.c | 1036 ++++++
libxfs/crc32defs.h | 72 +
libxfs/gen_crc32table.c | 144 +
libxfs/init.c | 26 +-
libxfs/logitem.c | 20 +-
libxfs/rdwr.c | 522 ++-
libxfs/trans.c | 55 +-
libxfs/util.c | 230 +-
libxfs/xfs.h | 147 +-
libxfs/xfs_alloc.c | 694 ++--
libxfs/xfs_alloc_btree.c | 140 +-
libxfs/xfs_attr.c | 546 +--
libxfs/xfs_attr_leaf.c | 1858 ++++++----
libxfs/xfs_attr_remote.c | 527 +++
libxfs/xfs_bmap.c | 8950 ++++++++++++++++++++++++----------------------
libxfs/xfs_bmap_btree.c | 124 +-
libxfs/xfs_btree.c | 415 ++-
libxfs/xfs_da_btree.c | 2598 ++++++++------
libxfs/xfs_dir2.c | 149 +-
libxfs/xfs_dir2_block.c | 806 +++--
libxfs/xfs_dir2_data.c | 604 ++--
libxfs/xfs_dir2_leaf.c | 1342 ++++---
libxfs/xfs_dir2_node.c | 1455 ++++----
libxfs/xfs_dir2_priv.h | 160 +
libxfs/xfs_dir2_sf.c | 326 +-
libxfs/xfs_ialloc.c | 632 ++--
libxfs/xfs_ialloc_btree.c | 87 +-
libxfs/xfs_inode.c | 505 ++-
libxfs/xfs_mount.c | 259 +-
libxfs/xfs_rtalloc.c | 77 +-
libxfs/xfs_symlink.c | 154 +
libxfs/xfs_trans.c | 9 +-
libxlog/xfs_log_recover.c | 299 +-
logprint/log_copy.c | 2 +-
logprint/log_dump.c | 2 +-
logprint/log_misc.c | 9 +-
logprint/log_print_all.c | 7 +-
logprint/log_print_trans.c | 22 +-
logprint/logprint.c | 8 +-
logprint/logprint.h | 10 +-
mdrestore/Makefile | 2 +-
mkfs/maxtrres.c | 4 +-
mkfs/proto.c | 14 +-
mkfs/xfs_mkfs.c | 162 +-
mkfs/xfs_mkfs.h | 12 +-
repair/agheader.c | 36 +-
repair/attr_repair.c | 63 +-
repair/dino_chunks.c | 13 +-
repair/dinode.c | 208 +-
repair/dir.c | 96 +-
repair/dir2.c | 497 +--
repair/dir2.h | 28 +-
repair/phase2.c | 13 +-
repair/phase3.c | 2 +-
repair/phase5.c | 157 +-
repair/phase6.c | 327 +-
repair/prefetch.c | 11 +-
repair/rt.c | 4 +-
repair/scan.c | 166 +-
repair/scan.h | 12 +-
repair/versions.c | 2 +-
repair/xfs_repair.c | 9 +-
134 files changed, 19036 insertions(+), 12663 deletions(-)
create mode 100644 include/xfs_attr_remote.h
create mode 100644 include/xfs_cksum.h
create mode 100644 include/xfs_dir1.h
delete mode 100644 include/xfs_dir2_block.h
delete mode 100644 include/xfs_dir2_data.h
create mode 100644 include/xfs_dir2_format.h
delete mode 100644 include/xfs_dir2_leaf.h
delete mode 100644 include/xfs_dir2_node.h
delete mode 100644 include/xfs_dir2_sf.h
create mode 100644 include/xfs_symlink.h
create mode 100644 libxfs/crc32.c
create mode 100644 libxfs/crc32defs.h
create mode 100644 libxfs/gen_crc32table.c
create mode 100644 libxfs/xfs_attr_remote.c
create mode 100644 libxfs/xfs_dir2_priv.h
create mode 100644 libxfs/xfs_symlink.c
--
Dave Chinner
david@xxxxxxxxxxxxx
xfsprogs-crc-patchset-v2/ 0000755 0001750 0001750 00000000000 12132011610 014237 5 ustar dave dave xfsprogs-crc-patchset-v2/series 0000644 0001750 0001750 00000001436 12132011546 015470 0 ustar dave dave repair-fix-extent-count
logprint-agi-dump
xfsprogs-kernel-3.8-update
xfsprogs-kernel-3.9-rc1-dev-update
xfsprogs-crc-infrastructure
xfsprogs-libxlog-kernel-sync
btree-format-changes
ag-hdr-format-changes
quota-format-changes
inode-v3
remote-symlink-crc
xfs-add-crc-to-dir2-block-2
xfs-add-crc-to-dir2-free-2
xfs-add-crc-to-dir2-data-2
xfs-add-crc-to-dir2-leaf-3
dir2-sf
xfs-add-crc-to-da-node
xfs-add-crc-to-attr-leaf
xfs-attr-rmt-move
xfs-add-crc-to-remote-attr
xfs-da-blf-types
xfs-trans-type-blf-change
xfs__add_crc_checks_to_the_superblock-2
xfs-sb-v5-feature-masks
xfsprogs-mkfs-crc-support
xfsprogs-repair-crc-support
xfsprogs-db-crc-support
xfsprogs-repair-dir-crc-support
debug
bpm-progs-deb
xfsprogs-freespace
spaceman-trim
spaceman-prealloc-flush
spaceman-ag-control
xfsprogs-freespace-2
xfsprogs-crc-patchset-v2/repair-fix-extent-count 0000600 0001750 0001750 00000004320 12132011546 020662 0 ustar dave dave xfs_repair: validate on-disk extent count better
From: Dave Chinner <dchinner@xxxxxxxxxx>
When scanning a btree format inode, we trust the extent count to be
in range. However, values of the range 2^31 <= cnt < 2^32 are
invalid and can cause problems with signed range checks. This
results in assert failures which validating the extent count such
as:
xfs_repair: dinode.c:768: process_bmbt_reclist_int: Assertion `i < *numrecs' failed.
Validate the extent count is at least within the positive range of a
singed 32 bit integer before using it.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
repair/dinode.c | 25 +++++++++++++++++++++++--
1 file changed, 23 insertions(+), 2 deletions(-)
diff --git a/repair/dinode.c b/repair/dinode.c
index 5a2da39..239bb7b 100644
--- a/repair/dinode.c
+++ b/repair/dinode.c
@@ -1293,7 +1293,7 @@ process_exinode(
xfs_bmbt_rec_t *rp;
xfs_dfiloff_t first_key;
xfs_dfiloff_t last_key;
- int numrecs;
+ int32_t numrecs;
int ret;
lino = XFS_AGINO_TO_INO(mp, agno, ino);
@@ -1302,6 +1302,15 @@ process_exinode(
numrecs = XFS_DFORK_NEXTENTS(dip, whichfork);
/*
+ * We've already decided on the maximum number of extents on the inode,
+ * and numrecs may be corrupt. Hence make sure we only allow numrecs to
+ * be in the range of valid on-disk numbers, which is:
+ * 0 < numrecs < 2^31 - 1
+ */
+ if (numrecs < 0)
+ numrecs = *nex;
+
+ /*
* XXX - if we were going to fix up the btree record,
* we'd do it right here. For now, if there's a problem,
* we'll bail out and presumably clear the inode.
@@ -2038,11 +2047,23 @@ process_inode_data_fork(
{
xfs_ino_t lino = XFS_AGINO_TO_INO(mp, agno, ino);
int err = 0;
+ int nex;
+
+ /*
+ * extent count on disk is only valid for positive values. The kernel
+ * uses negative values in memory. hence if we see negative numbers
+ * here, trash it!
+ */
+ nex = be32_to_cpu(dino->di_nextents);
+ if (nex < 0)
+ *nextents = 1;
+ else
+ *nextents = nex;
- *nextents = be32_to_cpu(dino->di_nextents);
if (*nextents > be64_to_cpu(dino->di_nblocks))
*nextents = 1;
+
if (dino->di_format != XFS_DINODE_FMT_LOCAL && type != XR_INO_RTDATA)
*dblkmap = blkmap_alloc(*nextents, XFS_DATA_FORK);
*nextents = 0;
xfsprogs-crc-patchset-v2/logprint-agi-dump 0000600 0001750 0001750 00000002412 12132011546 017520 0 ustar dave dave ---
logprint/log_print_all.c | 19 +++++++++++++++++++
1 file changed, 19 insertions(+)
diff --git a/logprint/log_print_all.c b/logprint/log_print_all.c
index 5b2d942..70b0905 100644
--- a/logprint/log_print_all.c
+++ b/logprint/log_print_all.c
@@ -122,6 +122,7 @@ xlog_recover_print_buffer(
be32_to_cpu(*(__be32 *)(p+56)),
be32_to_cpu(*(__be32 *)(p+60)));
} else if (be32_to_cpu(*(__be32 *)p) == XFS_AGI_MAGIC) {
+ int bucket, buckets;
agi = (xfs_agi_t *)p;
printf(_(" AGI Buffer: (XAGI)\n"));
if (!print_buffer)
@@ -137,6 +138,24 @@ xlog_recover_print_buffer(
be32_to_cpu(agi->agi_level),
be32_to_cpu(agi->agi_freecount),
be32_to_cpu(agi->agi_newino));
+ if (len == 128) {
+ buckets = 17;
+ } else if (len == 256) {
+ buckets = 32 + 17;
+ } else {
+ buckets = XFS_AGI_UNLINKED_BUCKETS;
+ }
+ for (bucket = 0; bucket < buckets;) {
+ int col;
+ printf(_("bucket[%d - %d]: "), bucket, bucket+3);
+ for (col = 0; col < 4; col++, bucket++) {
+ if (bucket < buckets) {
+ printf("0x%x ",
+ be32_to_cpu(agi->agi_unlinked[bucket]));
+ }
+ }
+ printf("\n");
+ }
} else if (be32_to_cpu(*(__be32 *)p) == XFS_AGF_MAGIC) {
agf = (xfs_agf_t *)p;
printf(_(" AGF Buffer: (XAGF)\n"));
xfsprogs-crc-patchset-v2/xfsprogs-kernel-3.8-update 0000600 0001750 0001750 00003124456 12132011546 021120 0 ustar dave dave xfsprogs: sync code to current kernel code
From: Dave Chinner <dchinner@xxxxxxxxxx>
Update code base to match kernel 3.8-rc2 code base.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
diff --git a/db/check.c b/db/check.c
index e601e0a..5921779 100644
--- a/db/check.c
+++ b/db/check.c
@@ -31,6 +31,8 @@
#include "output.h"
#include "init.h"
#include "malloc.h"
+#include "dir.h"
+#include "dir2.h"
typedef enum {
IS_USER_QUOTA, IS_PROJECT_QUOTA, IS_GROUP_QUOTA,
@@ -2223,7 +2225,7 @@ process_data_dir_v2(
bf = data->hdr.bestfree;
ptr = (char *)data->u;
if (be32_to_cpu(block->hdr.magic) == XFS_DIR2_BLOCK_MAGIC) {
- btp = xfs_dir2_block_tail_p(mp, block);
+ btp = xfs_dir2_block_tail_p(mp, &block->hdr);
lep = xfs_dir2_block_leaf_p(btp);
endptr = (char *)lep;
if (endptr <= ptr || endptr > (char *)btp) {
@@ -2840,7 +2842,7 @@ process_inode(
break;
}
if (ic) {
- dqprid = xfs_get_projid(idic); /* dquot ID is u32 */
+ dqprid = xfs_get_projid(&idic); /* dquot ID is u32 */
quota_add(&dqprid, &idic.di_gid, &idic.di_uid,
0, bc, ic, rc);
}
@@ -3166,7 +3168,7 @@ process_leaf_node_dir_v2_free(
error++;
return;
}
- maxent = XFS_DIR2_MAX_FREE_BESTS(mp);
+ maxent = xfs_dir2_free_max_bests(mp);
if (be32_to_cpu(free->hdr.firstdb) != xfs_dir2_da_to_db(mp,
dabno - mp->m_dirfreeblk) * maxent) {
if (!sflag || v)
@@ -3643,10 +3645,10 @@ process_sf_dir_v2(
if (v)
dbprintf(_("dir %lld entry . %lld\n"), id->ino, id->ino);
(*dot)++;
- sfe = xfs_dir2_sf_firstentry(sf);
+ sfe = xfs_dir2_sf_firstentry(&sf->hdr);
offset = XFS_DIR2_DATA_FIRST_OFFSET;
for (i = sf->hdr.count - 1, i8 = 0; i >= 0; i--) {
- if ((__psint_t)sfe + xfs_dir2_sf_entsize_byentry(sf, sfe) -
+ if ((__psint_t)sfe + xfs_dir2_sf_entsize(&sf->hdr,sfe->namelen) -
(__psint_t)sf > be64_to_cpu(dip->di_size)) {
if (!sflag)
dbprintf(_("dir %llu bad size in entry at %d\n"),
@@ -3655,7 +3657,7 @@ process_sf_dir_v2(
error++;
break;
}
- lino = xfs_dir2_sf_get_inumber(sf, xfs_dir2_sf_inumberp(sfe));
+ lino = xfs_dir2_sfe_get_ino(&sf->hdr, sfe);
if (lino > XFS_DIR2_MAX_SHORT_INUM)
i8++;
cid = find_inode(lino, 1);
@@ -3685,8 +3687,8 @@ process_sf_dir_v2(
}
offset =
xfs_dir2_sf_get_offset(sfe) +
- xfs_dir2_data_entsize(sfe->namelen);
- sfe = xfs_dir2_sf_nextentry(sf, sfe);
+ xfs_dir2_sf_entsize(&sf->hdr, sfe->namelen);
+ sfe = xfs_dir2_sf_nextentry(&sf->hdr, sfe);
}
if (i < 0 && (__psint_t)sfe - (__psint_t)sf !=
be64_to_cpu(dip->di_size)) {
@@ -3702,7 +3704,7 @@ process_sf_dir_v2(
dbprintf(_("dir %llu offsets too high\n"), id->ino);
error++;
}
- lino = xfs_dir2_sf_get_inumber(sf, &sf->hdr.parent);
+ lino = xfs_dir2_sf_get_parent_ino(&sf->hdr);
if (lino > XFS_DIR2_MAX_SHORT_INUM)
i8++;
cid = find_inode(lino, 1);
diff --git a/db/dir.h b/db/dir.h
index 67e6689..5be82bf 100644
--- a/db/dir.h
+++ b/db/dir.h
@@ -16,6 +16,12 @@
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*/
+/*
+ * only db understands v1 directory format, so include them
+ * directly here
+ */
+#include <xfs_dir1.h>
+
extern const field_t dir_flds[];
extern const field_t dir_hfld[];
extern const field_t dir_blkinfo_flds[];
diff --git a/db/dir2.c b/db/dir2.c
index 51c598b..be58639 100644
--- a/db/dir2.c
+++ b/db/dir2.c
@@ -192,7 +192,7 @@ dir2_block_leaf_count(
block = obj;
if (be32_to_cpu(block->hdr.magic) != XFS_DIR2_BLOCK_MAGIC)
return 0;
- btp = xfs_dir2_block_tail_p(mp, block);
+ btp = xfs_dir2_block_tail_p(mp, &block->hdr);
return be32_to_cpu(btp->count);
}
@@ -210,7 +210,7 @@ dir2_block_leaf_offset(
ASSERT(startoff == 0);
block = obj;
ASSERT(be32_to_cpu(block->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
- btp = xfs_dir2_block_tail_p(mp, block);
+ btp = xfs_dir2_block_tail_p(mp, &block->hdr);
lep = xfs_dir2_block_leaf_p(btp) + idx;
return bitize((int)((char *)lep - (char *)block));
}
@@ -242,7 +242,7 @@ dir2_block_tail_offset(
ASSERT(idx == 0);
block = obj;
ASSERT(be32_to_cpu(block->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
- btp = xfs_dir2_block_tail_p(mp, block);
+ btp = xfs_dir2_block_tail_p(mp, &block->hdr);
return bitize((int)((char *)btp - (char *)block));
}
@@ -264,7 +264,7 @@ dir2_block_u_count(
block = obj;
if (be32_to_cpu(block->hdr.magic) != XFS_DIR2_BLOCK_MAGIC)
return 0;
- btp = xfs_dir2_block_tail_p(mp, block);
+ btp = xfs_dir2_block_tail_p(mp, &block->hdr);
ptr = (char *)block->u;
endptr = (char *)xfs_dir2_block_leaf_p(btp);
for (i = 0; ptr < endptr; i++) {
@@ -297,7 +297,7 @@ dir2_block_u_offset(
ASSERT(startoff == 0);
block = obj;
ASSERT(be32_to_cpu(block->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
- btp = xfs_dir2_block_tail_p(mp, block);
+ btp = xfs_dir2_block_tail_p(mp, &block->hdr);
ptr = (char *)block->u;
endptr = (char *)xfs_dir2_block_leaf_p(btp);
for (i = 0; i < idx; i++) {
diff --git a/db/dir2.h b/db/dir2.h
index dcf4787..4c2847b 100644
--- a/db/dir2.h
+++ b/db/dir2.h
@@ -27,5 +27,37 @@ extern const field_t dir2_leaf_entry_flds[];
extern const field_t dir2_leaf_hdr_flds[];
extern const field_t dir2_leaf_tail_flds[];
+/*
+ * generic dir2 structures used by xfs_db
+ */
+typedef union {
+ xfs_dir2_data_entry_t entry;
+ xfs_dir2_data_unused_t unused;
+} xfs_dir2_data_union_t;
+
+typedef struct xfs_dir2_data {
+ xfs_dir2_data_hdr_t hdr; /* magic XFS_DIR2_DATA_MAGIC */
+ xfs_dir2_data_union_t u[1];
+} xfs_dir2_data_t;
+
+typedef struct xfs_dir2_block {
+ xfs_dir2_data_hdr_t hdr; /* magic XFS_DIR2_BLOCK_MAGIC */
+ xfs_dir2_data_union_t u[1];
+ xfs_dir2_leaf_entry_t leaf[1];
+ xfs_dir2_block_tail_t tail;
+} xfs_dir2_block_t;
+
+typedef struct xfs_dir2_sf {
+ xfs_dir2_sf_hdr_t hdr; /* shortform header */
+ xfs_dir2_sf_entry_t list[1]; /* shortform entries */
+} xfs_dir2_sf_t;
+
+static inline xfs_dir2_inou_t *xfs_dir2_sf_inumberp(xfs_dir2_sf_entry_t *sfep)
+{
+ return (xfs_dir2_inou_t *)&(sfep)->name[(sfep)->namelen];
+}
+
extern int dir2_data_union_size(void *obj, int startoff, int idx);
extern int dir2_size(void *obj, int startoff, int idx);
+
+
diff --git a/db/dir2sf.c b/db/dir2sf.c
index b2db088..92f8a66 100644
--- a/db/dir2sf.c
+++ b/db/dir2sf.c
@@ -22,6 +22,7 @@
#include "fprint.h"
#include "field.h"
#include "bit.h"
+#include "dir2.h"
#include "dir2sf.h"
static int dir2_inou_i4_count(void *obj, int startoff);
@@ -149,10 +150,10 @@ dir2_sf_entry_size(
ASSERT(bitoffs(startoff) == 0);
sf = (xfs_dir2_sf_t *)((char *)obj + byteize(startoff));
- e = xfs_dir2_sf_firstentry(sf);
+ e = xfs_dir2_sf_firstentry(&sf->hdr);
for (i = 0; i < idx; i++)
- e = xfs_dir2_sf_nextentry(sf, e);
- return bitize((int)xfs_dir2_sf_entsize_byentry(sf, e));
+ e = xfs_dir2_sf_nextentry(&sf->hdr, e);
+ return bitize((int)xfs_dir2_sf_entsize(&sf->hdr, e->namelen));
}
/*ARGSUSED*/
@@ -194,9 +195,9 @@ dir2_sf_list_offset(
ASSERT(bitoffs(startoff) == 0);
sf = (xfs_dir2_sf_t *)((char *)obj + byteize(startoff));
- e = xfs_dir2_sf_firstentry(sf);
+ e = xfs_dir2_sf_firstentry(&sf->hdr);
for (i = 0; i < idx; i++)
- e = xfs_dir2_sf_nextentry(sf, e);
+ e = xfs_dir2_sf_nextentry(&sf->hdr, e);
return bitize((int)((char *)e - (char *)sf));
}
@@ -214,8 +215,8 @@ dir2sf_size(
ASSERT(bitoffs(startoff) == 0);
ASSERT(idx == 0);
sf = (xfs_dir2_sf_t *)((char *)obj + byteize(startoff));
- e = xfs_dir2_sf_firstentry(sf);
+ e = xfs_dir2_sf_firstentry(&sf->hdr);
for (i = 0; i < sf->hdr.count; i++)
- e = xfs_dir2_sf_nextentry(sf, e);
+ e = xfs_dir2_sf_nextentry(&sf->hdr, e);
return bitize((int)((char *)e - (char *)sf));
}
diff --git a/db/dirshort.c b/db/dirshort.c
index b0d3b6b..2b23c28 100644
--- a/db/dirshort.c
+++ b/db/dirshort.c
@@ -22,6 +22,7 @@
#include "fprint.h"
#include "field.h"
#include "bit.h"
+#include "dir.h"
#include "dirshort.h"
static int dir_sf_entry_name_count(void *obj, int startoff);
diff --git a/db/metadump.c b/db/metadump.c
index 9f15d9e..5739f86 100644
--- a/db/metadump.c
+++ b/db/metadump.c
@@ -26,6 +26,10 @@
#include "init.h"
#include "sig.h"
#include "xfs_metadump.h"
+#include "fprint.h"
+#include "faddr.h"
+#include "field.h"
+#include "dir2.h"
#define DEFAULT_MAX_EXT_SIZE 1000
@@ -916,7 +920,7 @@ obfuscate_sf_dir(
(long long)cur_ino);
}
- sfep = xfs_dir2_sf_firstentry(sfp);
+ sfep = xfs_dir2_sf_firstentry(&sfp->hdr);
for (i = 0; (i < sfp->hdr.count) &&
((char *)sfep - (char *)sfp < ino_dir_size); i++) {
@@ -935,7 +939,7 @@ obfuscate_sf_dir(
namelen = ino_dir_size - ((char *)&sfep->name[0] -
(char *)sfp);
} else if ((char *)sfep - (char *)sfp +
- xfs_dir2_sf_entsize_byentry(sfp, sfep) >
+ xfs_dir2_sf_entsize(&sfp->hdr, sfep->namelen) >
ino_dir_size) {
if (show_warnings)
print_warning("entry length in dir inode %llu "
@@ -946,12 +950,11 @@ obfuscate_sf_dir(
(char *)sfp);
}
- generate_obfuscated_name(xfs_dir2_sf_get_inumber(sfp,
- xfs_dir2_sf_inumberp(sfep)), namelen,
- &sfep->name[0]);
+ generate_obfuscated_name(xfs_dir2_sfe_get_ino(&sfp->hdr, sfep),
+ namelen, &sfep->name[0]);
sfep = (xfs_dir2_sf_entry_t *)((char *)sfep +
- xfs_dir2_sf_entsize_byname(sfp, namelen));
+ xfs_dir2_sf_entsize(&sfp->hdr, namelen));
}
}
@@ -1107,9 +1110,10 @@ obfuscate_dir_data_blocks(
if (is_block_format) {
xfs_dir2_leaf_entry_t *blp;
xfs_dir2_block_tail_t *btp;
+ xfs_dir2_block_t *blk;
- btp = xfs_dir2_block_tail_p(mp,
- (xfs_dir2_block_t *)block);
+ blk = (xfs_dir2_block_t *)block;
+ btp = xfs_dir2_block_tail_p(mp, &blk->hdr);
blp = xfs_dir2_block_leaf_p(btp);
if ((char *)blp > (char *)btp)
blp = (xfs_dir2_leaf_entry_t *)btp;
diff --git a/estimate/xfs_estimate.c b/estimate/xfs_estimate.c
index 54a72e9..a8a3b99 100644
--- a/estimate/xfs_estimate.c
+++ b/estimate/xfs_estimate.c
@@ -18,8 +18,11 @@
/*
* Estimate space of an XFS filesystem
+ *
+ * XXX: assumes dirv1 format.
*/
#include <xfs/libxfs.h>
+#include <xfs_dir1.h>
#include <sys/stat.h>
#include <ftw.h>
diff --git a/fsr/xfs_fsr.c b/fsr/xfs_fsr.c
index 3d017ca..0471c55 100644
--- a/fsr/xfs_fsr.c
+++ b/fsr/xfs_fsr.c
@@ -16,8 +16,8 @@
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*/
-#include <xfs/xfs.h>
#include <libxfs.h>
+#include <xfs/xfs.h>
#include <xfs/xfs_types.h>
#include <xfs/jdm.h>
#include <xfs/xfs_dfrag.h>
diff --git a/include/Makefile b/include/Makefile
index 7ad6ddb..b94d305 100644
--- a/include/Makefile
+++ b/include/Makefile
@@ -24,15 +24,14 @@ QAHFILES = libxfs.h libxlog.h \
xfs_ag.h xfs_alloc.h xfs_alloc_btree.h xfs_arch.h xfs_attr_leaf.h \
xfs_attr_sf.h xfs_bit.h xfs_bmap.h xfs_bmap_btree.h xfs_btree.h \
xfs_btree_trace.h xfs_buf_item.h xfs_da_btree.h xfs_dinode.h \
- xfs_dir2.h xfs_dir2_block.h xfs_dir2_data.h xfs_dir2_leaf.h \
- xfs_dir2_node.h xfs_dir2_sf.h xfs_dir_leaf.h xfs_dir_sf.h \
+ xfs_dir2.h xfs_dir2_format.h xfs_dir1.h xfs_dir_leaf.h xfs_dir_sf.h \
xfs_extfree_item.h xfs_ialloc.h xfs_ialloc_btree.h \
xfs_inode.h xfs_inode_item.h xfs_inum.h \
xfs_log.h xfs_log_priv.h xfs_log_recover.h xfs_metadump.h \
xfs_mount.h xfs_quota.h xfs_rtalloc.h xfs_sb.h xfs_trace.h \
- xfs_trans.h xfs_trans_space.h xfs_types.h xfs_dfrag.h
+ xfs_trans.h xfs_trans_space.h xfs_dfrag.h
-HFILES = handle.h jdm.h xqm.h xfs.h xfs_fs.h
+HFILES = handle.h jdm.h xqm.h xfs.h xfs_fs.h xfs_types.h
HFILES += $(PKG_PLATFORM).h
PHFILES = darwin.h freebsd.h irix.h linux.h gnukfreebsd.h
DKHFILES = volume.h fstyp.h dvh.h
diff --git a/include/libxfs.h b/include/libxfs.h
index 9574f0b..0e2fc5f 100644
--- a/include/libxfs.h
+++ b/include/libxfs.h
@@ -33,8 +33,8 @@
#include <xfs/swab.h>
#include <xfs/atomic.h>
-#include <xfs/xfs_fs.h>
#include <xfs/xfs_types.h>
+#include <xfs/xfs_fs.h>
#include <xfs/xfs_arch.h>
#include <xfs/xfs_bit.h>
#include <xfs/xfs_inum.h>
@@ -46,8 +46,6 @@
#include <xfs/xfs_bmap_btree.h>
#include <xfs/xfs_alloc_btree.h>
#include <xfs/xfs_ialloc_btree.h>
-#include <xfs/xfs_dir_sf.h>
-#include <xfs/xfs_dir2_sf.h>
#include <xfs/xfs_attr_sf.h>
#include <xfs/xfs_dinode.h>
#include <xfs/xfs_inode.h>
@@ -136,7 +134,6 @@ extern int libxfs_log_clear (dev_t, xfs_daddr_t, uint, uuid_t *,
extern int libxfs_log_header (xfs_caddr_t, uuid_t *, int, int, int,
libxfs_get_block_t *, void *);
-
/*
* Define a user-level mount structure with all we need
* in order to make use of the numerous XFS_* macros.
@@ -200,6 +197,7 @@ typedef struct xfs_mount {
xfs_dablk_t m_dirfreeblk; /* blockno of dirfreeindex v2 */
} xfs_mount_t;
+
#define LIBXFS_MOUNT_ROOTINOS 0x0001
#define LIBXFS_MOUNT_DEBUGGER 0x0002
#define LIBXFS_MOUNT_32BITINODES 0x0004
@@ -219,11 +217,32 @@ extern void libxfs_rtmount_destroy (xfs_mount_t *);
/*
* Simple I/O interface
*/
+typedef struct xfs_buftarg {
+ struct xfs_mount *bt_mount;
+ dev_t dev;
+} xfs_buftarg_t;
+
+#define XB_PAGES 2
+
+struct xfs_buf_map {
+ xfs_daddr_t bm_bn; /* block number for I/O */
+ int bm_len; /* size of I/O */
+};
+
+#define DEFINE_SINGLE_BUF_MAP(map, blkno, numblk) \
+ struct xfs_buf_map (map) = { .bm_bn = (blkno), .bm_len = (numblk) };
+
+struct xfs_buf_ops {
+ void (*verify_read)(struct xfs_buf *);
+ void (*verify_write)(struct xfs_buf *);
+};
+
typedef struct xfs_buf {
struct cache_node b_node;
unsigned int b_flags;
xfs_daddr_t b_blkno;
unsigned b_bcount;
+ unsigned int b_length;
dev_t b_dev;
pthread_mutex_t b_lock;
pthread_t b_holder;
@@ -231,8 +250,13 @@ typedef struct xfs_buf {
void *b_fsprivate;
void *b_fsprivate2;
void *b_fsprivate3;
- char *b_addr;
+ void *b_addr;
int b_error;
+ const struct xfs_buf_ops *b_ops;
+ struct xfs_buftarg *b_target;
+ struct xfs_perag *b_pag;
+ struct xfs_buf_map *b_map;
+ int b_nmaps;
#ifdef XFS_BUF_TRACING
struct list_head b_lock_list;
const char *b_func;
@@ -245,10 +269,11 @@ enum xfs_buf_flags_t { /* b_flags bits */
LIBXFS_B_EXIT = 0x0001, /* ==LIBXFS_EXIT_ON_FAILURE */
LIBXFS_B_DIRTY = 0x0002, /* buffer has been modified */
LIBXFS_B_STALE = 0x0004, /* buffer marked as invalid */
- LIBXFS_B_UPTODATE = 0x0008 /* buffer is sync'd to disk */
+ LIBXFS_B_UPTODATE = 0x0008, /* buffer is sync'd to disk */
+ LIBXFS_B_DISCONTIG = 0x0010, /* discontiguous buffer */
};
-#define XFS_BUF_PTR(bp) ((bp)->b_addr)
+#define XFS_BUF_PTR(bp) ((char *)(bp)->b_addr)
#define xfs_buf_offset(bp, offset) (XFS_BUF_PTR(bp) + (offset))
#define XFS_BUF_ADDR(bp) ((bp)->b_blkno)
#define XFS_BUF_SIZE(bp) ((bp)->b_bcount)
@@ -275,6 +300,8 @@ enum xfs_buf_flags_t { /* b_flags bits */
(pri))
#define XFS_BUF_PRIORITY(bp) (cache_node_get_priority( \
(struct cache_node *)(bp)))
+#define xfs_buf_set_ref(bp,ref) ((void) 0)
+#define xfs_buf_ioerror(bp,err) (bp)->b_error = (err);
/* Buffer Cache Interfaces */
@@ -288,23 +315,34 @@ extern struct cache_operations libxfs_bcache_operations;
#define libxfs_readbuf(dev, daddr, len, flags) \
libxfs_trace_readbuf(__FUNCTION__, __FILE__, __LINE__, \
(dev), (daddr), (len), (flags))
+#define libxfs_readbuf_map(dev, map, nmaps, flags) \
+ libxfs_trace_readbuf_map(__FUNCTION__, __FILE__, __LINE__, \
+ (dev), (map), (nmaps), (flags))
#define libxfs_writebuf(buf, flags) \
libxfs_trace_writebuf(__FUNCTION__, __FILE__, __LINE__, \
(buf), (flags))
#define libxfs_getbuf(dev, daddr, len) \
libxfs_trace_getbuf(__FUNCTION__, __FILE__, __LINE__, \
(dev), (daddr), (len))
+#define libxfs_getbuf_map(dev, map, nmaps) \
+ libxfs_trace_getbuf_map(__FUNCTION__, __FILE__, __LINE__, \
+ (dev), (map), (nmaps))
#define libxfs_getbuf_flags(dev, daddr, len, flags) \
- libxfs_trace_getbuf(__FUNCTION__, __FILE__, __LINE__, \
+ libxfs_trace_getbuf_flags(__FUNCTION__, __FILE__, __LINE__, \
(dev), (daddr), (len), (flags))
#define libxfs_putbuf(buf) \
libxfs_trace_putbuf(__FUNCTION__, __FILE__, __LINE__, (buf))
extern xfs_buf_t *libxfs_trace_readbuf(const char *, const char *, int,
dev_t, xfs_daddr_t, int, int);
+extern xfs_buf_t *libxfs_trace_readbuf_map(const char *, const char *, int,
+ dev_t, struct xfs_buf_map *, int, int);
extern int libxfs_trace_writebuf(const char *, const char *, int,
xfs_buf_t *, int);
-extern xfs_buf_t *libxfs_trace_getbuf(const char *, const char *, int, dev_t, xfs_daddr_t, int);
+extern xfs_buf_t *libxfs_trace_getbuf(const char *, const char *, int,
+ dev_t, xfs_daddr_t, int);
+extern xfs_buf_t *libxfs_trace_getbuf_map(const char *, const char *, int,
+ dev_t, struct xfs_buf_map *, int);
extern xfs_buf_t *libxfs_trace_getbuf_flags(const char *, const char *, int,
dev_t, xfs_daddr_t, int, unsigned int);
extern void libxfs_trace_putbuf (const char *, const char *, int,
@@ -313,8 +351,10 @@ extern void libxfs_trace_putbuf (const char *, const char *, int,
#else
extern xfs_buf_t *libxfs_readbuf(dev_t, xfs_daddr_t, int, int);
+extern xfs_buf_t *libxfs_readbuf_map(dev_t, struct xfs_buf_map *, int, int);
extern int libxfs_writebuf(xfs_buf_t *, int);
extern xfs_buf_t *libxfs_getbuf(dev_t, xfs_daddr_t, int);
+extern xfs_buf_t *libxfs_getbuf_map(dev_t, struct xfs_buf_map *, int);
extern xfs_buf_t *libxfs_getbuf_flags(dev_t, xfs_daddr_t, int, unsigned int);
extern void libxfs_putbuf (xfs_buf_t *);
@@ -358,6 +398,7 @@ typedef struct xfs_inode_log_item {
xfs_log_item_t ili_item; /* common portion */
struct xfs_inode *ili_inode; /* inode pointer */
unsigned short ili_flags; /* misc flags */
+ unsigned int ili_fields; /* fields to be logged */
unsigned int ili_last_fields; /* fields when flushed*/
xfs_inode_log_format_t ili_format; /* logged structure */
int ili_lock_flags;
@@ -409,11 +450,50 @@ extern void libxfs_trans_bjoin (xfs_trans_t *, struct xfs_buf *);
extern void libxfs_trans_bhold (xfs_trans_t *, struct xfs_buf *);
extern void libxfs_trans_log_buf (xfs_trans_t *, struct xfs_buf *,
uint, uint);
+/*
extern xfs_buf_t *libxfs_trans_get_buf (xfs_trans_t *, dev_t,
xfs_daddr_t, int, uint);
extern int libxfs_trans_read_buf (xfs_mount_t *, xfs_trans_t *, dev_t,
xfs_daddr_t, int, uint, struct xfs_buf **);
-
+*/
+
+struct xfs_buf *libxfs_trans_get_buf_map(struct xfs_trans *tp, dev_t dev,
+ struct xfs_buf_map *map, int nmaps,
+ uint flags);
+
+static inline struct xfs_buf *
+libxfs_trans_get_buf(
+ struct xfs_trans *tp,
+ dev_t dev,
+ xfs_daddr_t blkno,
+ int numblks,
+ uint flags)
+{
+ DEFINE_SINGLE_BUF_MAP(map, blkno, numblks);
+ return libxfs_trans_get_buf_map(tp, dev, &map, 1, flags);
+}
+
+int libxfs_trans_read_buf_map(struct xfs_mount *mp,
+ struct xfs_trans *tp, dev_t dev,
+ struct xfs_buf_map *map, int nmaps,
+ uint flags, struct xfs_buf **bpp,
+ const struct xfs_buf_ops *ops);
+
+static inline int
+libxfs_trans_read_buf(
+ struct xfs_mount *mp,
+ struct xfs_trans *tp,
+ dev_t dev,
+ xfs_daddr_t blkno,
+ int numblks,
+ uint flags,
+ struct xfs_buf **bpp,
+ const struct xfs_buf_ops *ops)
+{
+ DEFINE_SINGLE_BUF_MAP(map, blkno, numblks);
+ return libxfs_trans_read_buf_map(mp, tp, dev, &map, 1,
+ flags, bpp, ops);
+}
/*
* Inode interface
@@ -438,6 +518,27 @@ typedef struct xfs_inode {
#define LIBXFS_ATTR_CREATE 0x0010 /* create, but fail if attr exists */
#define LIBXFS_ATTR_REPLACE 0x0020 /* set, but fail if attr not exists */
+/*
+ * Project quota id helpers (previously projid was 16bit only and using two
+ * 16bit values to hold new 32bit projid was chosen to retain compatibility with
+ * "old" filesystems).
+ *
+ * Copied here from xfs_inode.h because it has to be defined after the struct
+ * xfs_inode...
+ */
+static inline prid_t
+xfs_get_projid(struct xfs_icdinode *id)
+{
+ return (prid_t)id->di_projid_hi << 16 | id->di_projid_lo;
+}
+
+static inline void
+xfs_set_projid(struct xfs_icdinode *id, prid_t projid)
+{
+ id->di_projid_hi = (__uint16_t) (projid >> 16);
+ id->di_projid_lo = (__uint16_t) (projid & 0xffff);
+}
+
typedef struct cred {
uid_t cr_uid;
gid_t cr_gid;
@@ -451,8 +552,6 @@ extern void libxfs_trans_inode_alloc_buf (xfs_trans_t *, xfs_buf_t *);
extern void libxfs_trans_ichgtime(struct xfs_trans *,
struct xfs_inode *, int);
extern int libxfs_iflush_int (xfs_inode_t *, xfs_buf_t *);
-extern int libxfs_iread (xfs_mount_t *, xfs_trans_t *, xfs_ino_t,
- xfs_inode_t *, xfs_daddr_t);
/* Inode Cache Interfaces */
extern struct cache *libxfs_icache;
@@ -462,14 +561,7 @@ extern int libxfs_iget (xfs_mount_t *, xfs_trans_t *, xfs_ino_t,
uint, xfs_inode_t **, xfs_daddr_t);
extern void libxfs_iput (xfs_inode_t *, uint);
-extern int xfs_imap_to_bp(xfs_mount_t *mp, xfs_trans_t *tp, struct xfs_imap *imap,
- xfs_buf_t **bpp, uint buf_flags, uint iget_flags);
-
-#include <xfs/xfs_dir_leaf.h> /* dirv1 support in db & repair */
-#include <xfs/xfs_dir2_data.h>
-#include <xfs/xfs_dir2_leaf.h>
-#include <xfs/xfs_dir2_block.h>
-#include <xfs/xfs_dir2_node.h>
+#include <xfs/xfs_dir2_format.h>
/* Shared utility routines */
extern unsigned int libxfs_log2_roundup(unsigned int i);
@@ -478,11 +570,6 @@ extern int libxfs_alloc_file_space (xfs_inode_t *, xfs_off_t,
xfs_off_t, int, int);
extern int libxfs_bmap_finish(xfs_trans_t **, xfs_bmap_free_t *, int *);
-extern void libxfs_da_bjoin (xfs_trans_t *, xfs_dabuf_t *);
-extern void libxfs_da_bhold (xfs_trans_t *, xfs_dabuf_t *);
-extern int libxfs_da_read_bufr(xfs_trans_t *, xfs_inode_t *, xfs_dablk_t,
- xfs_daddr_t, xfs_dabuf_t **, int);
-
extern void libxfs_fs_repair_cmn_err(int, struct xfs_mount *, char *, ...);
extern void libxfs_fs_cmn_err(int, struct xfs_mount *, char *, ...);
@@ -536,7 +623,8 @@ void xfs_bmbt_disk_get_all(xfs_bmbt_rec_t *r, xfs_bmbt_irec_t *s);
/* xfs_bmap.h */
#define libxfs_bmap_cancel xfs_bmap_cancel
#define libxfs_bmap_last_offset xfs_bmap_last_offset
-#define libxfs_bmapi xfs_bmapi
+#define libxfs_bmapi_write xfs_bmapi_write
+#define libxfs_bmapi_read xfs_bmapi_read
#define libxfs_bunmapi xfs_bunmapi
/* xfs_bmap_btree.h */
@@ -546,6 +634,7 @@ void xfs_bmbt_disk_get_all(xfs_bmbt_rec_t *r, xfs_bmbt_irec_t *s);
#define libxfs_da_brelse xfs_da_brelse
#define libxfs_da_hashname xfs_da_hashname
#define libxfs_da_shrink_inode xfs_da_shrink_inode
+#define libxfs_da_read_buf xfs_da_read_buf
/* xfs_dir2.h */
#define libxfs_dir_createname xfs_dir_createname
@@ -566,6 +655,8 @@ void xfs_bmbt_disk_get_all(xfs_bmbt_rec_t *r, xfs_bmbt_irec_t *s);
/* xfs_inode.h */
#define libxfs_dinode_from_disk xfs_dinode_from_disk
#define libxfs_dinode_to_disk xfs_dinode_to_disk
+void xfs_dinode_from_disk(struct xfs_icdinode *,
+ struct xfs_dinode *);
#define libxfs_idata_realloc xfs_idata_realloc
#define libxfs_idestroy_fork xfs_idestroy_fork
diff --git a/include/linux.h b/include/linux.h
index e0bcba2..5bb91cd 100644
--- a/include/linux.h
+++ b/include/linux.h
@@ -27,6 +27,7 @@
#include <malloc.h>
#include <getopt.h>
#include <endian.h>
+#include <stdbool.h>
static __inline__ int xfsctl(const char *path, int fd, int cmd, void *p)
{
diff --git a/include/platform_defs.h.in b/include/platform_defs.h.in
index 4e1e0c4..217d6c0 100644
--- a/include/platform_defs.h.in
+++ b/include/platform_defs.h.in
@@ -34,6 +34,7 @@
#include <ctype.h>
#include <sys/types.h>
#include <limits.h>
+#include <stdbool.h>
#undef HAVE___U32
#ifdef HAVE___U32
@@ -57,6 +58,10 @@ typedef signed long long int __s64;
#define __force
#endif
+typedef __u16 __bitwise __le16;
+typedef __u32 __bitwise __le32;
+typedef __u64 __bitwise __le64;
+
typedef __u16 __bitwise __be16;
typedef __u32 __bitwise __be32;
typedef __u64 __bitwise __be64;
diff --git a/include/swab.h b/include/swab.h
index 2684aa7..3de44d8 100644
--- a/include/swab.h
+++ b/include/swab.h
@@ -153,4 +153,42 @@ static __inline__ void __swab64s(__u64 *addr)
(__extension__ ({__arch__swab64s(addr);}));
}
+static inline __uint16_t get_unaligned_be16(void *p)
+{
+ __uint8_t *__p = p;
+ return __p[0] << 8 | __p[1];
+}
+
+static inline __uint32_t get_unaligned_be32(void *p)
+{
+ __uint8_t *__p = p;
+ return __p[0] << 24 | __p[1] << 16 | __p[2] << 8 | __p[3];
+}
+
+static inline __uint64_t get_unaligned_be64(void *p)
+{
+ return (__uint64_t)get_unaligned_be32(p) << 32 |
+ get_unaligned_be32(p + 4);
+}
+
+static inline void put_unaligned_be16(__uint16_t val, void *p)
+{
+ __uint8_t *__p = p;
+ *__p++ = val >> 8;
+ *__p++ = val;
+}
+
+static inline void put_unaligned_be32(__uint32_t val, void *p)
+{
+ __uint8_t *__p = p;
+ put_unaligned_be16(val >> 16, __p);
+ put_unaligned_be16(val, __p + 2);
+}
+
+static inline void put_unaligned_be64(__uint64_t val, void *p)
+{
+ put_unaligned_be32(val >> 32, p);
+ put_unaligned_be32(val, p + 4);
+}
+
#endif /* SWAB_H */
diff --git a/include/xfs.h b/include/xfs.h
index 147a5a0..e32c4a8 100644
--- a/include/xfs.h
+++ b/include/xfs.h
@@ -34,6 +34,7 @@
#define __XFS_H__
#include <xfs/platform_defs.h>
+#include <xfs/xfs_types.h>
#include <xfs/xfs_fs.h>
#endif /* __XFS_H__ */
diff --git a/include/xfs_ag.h b/include/xfs_ag.h
index 5adce91..f2aeedb 100644
--- a/include/xfs_ag.h
+++ b/include/xfs_ag.h
@@ -103,11 +103,13 @@ typedef struct xfs_agf {
/* disk block (xfs_daddr_t) in the AG */
#define XFS_AGF_DADDR(mp) ((xfs_daddr_t)(1 << (mp)->m_sectbb_log))
#define XFS_AGF_BLOCK(mp) XFS_HDR_BLOCK(mp, XFS_AGF_DADDR(mp))
-#define XFS_BUF_TO_AGF(bp) ((xfs_agf_t *)XFS_BUF_PTR(bp))
+#define XFS_BUF_TO_AGF(bp) ((xfs_agf_t *)((bp)->b_addr))
extern int xfs_read_agf(struct xfs_mount *mp, struct xfs_trans *tp,
xfs_agnumber_t agno, int flags, struct xfs_buf **bpp);
+extern const struct xfs_buf_ops xfs_agf_buf_ops;
+
/*
* Size of the unlinked inode hash table in the agi.
*/
@@ -156,11 +158,13 @@ typedef struct xfs_agi {
/* disk block (xfs_daddr_t) in the AG */
#define XFS_AGI_DADDR(mp) ((xfs_daddr_t)(2 << (mp)->m_sectbb_log))
#define XFS_AGI_BLOCK(mp) XFS_HDR_BLOCK(mp, XFS_AGI_DADDR(mp))
-#define XFS_BUF_TO_AGI(bp) ((xfs_agi_t *)XFS_BUF_PTR(bp))
+#define XFS_BUF_TO_AGI(bp) ((xfs_agi_t *)((bp)->b_addr))
extern int xfs_read_agi(struct xfs_mount *mp, struct xfs_trans *tp,
xfs_agnumber_t agno, struct xfs_buf **bpp);
+extern const struct xfs_buf_ops xfs_agi_buf_ops;
+
/*
* The third a.g. block contains the a.g. freelist, an array
* of block pointers to blocks owned by the allocation btree code.
@@ -168,31 +172,13 @@ extern int xfs_read_agi(struct xfs_mount *mp, struct xfs_trans *tp,
#define XFS_AGFL_DADDR(mp) ((xfs_daddr_t)(3 << (mp)->m_sectbb_log))
#define XFS_AGFL_BLOCK(mp) XFS_HDR_BLOCK(mp, XFS_AGFL_DADDR(mp))
#define XFS_AGFL_SIZE(mp) ((mp)->m_sb.sb_sectsize / sizeof(xfs_agblock_t))
-#define XFS_BUF_TO_AGFL(bp) ((xfs_agfl_t *)XFS_BUF_PTR(bp))
+#define XFS_BUF_TO_AGFL(bp) ((xfs_agfl_t *)((bp)->b_addr))
typedef struct xfs_agfl {
__be32 agfl_bno[1]; /* actually XFS_AGFL_SIZE(mp) */
} xfs_agfl_t;
/*
- * Busy block/extent entry. Indexed by a rbtree in perag to mark blocks that
- * have been freed but whose transactions aren't committed to disk yet.
- *
- * Note that we use the transaction ID to record the transaction, not the
- * transaction structure itself. See xfs_alloc_busy_insert() for details.
- */
-struct xfs_busy_extent {
-#ifdef __KERNEL__
- struct rb_node rb_node; /* ag by-bno indexed search tree */
-#endif
- struct list_head list; /* transaction busy extent list */
- xfs_agnumber_t agno;
- xfs_agblock_t bno;
- xfs_extlen_t length;
- xlog_tid_t tid; /* transaction that created this */
-};
-
-/*
* Per-ag incore structure, copies of information in agf and agi,
* to improve the performance of allocation group selection.
*/
@@ -251,6 +237,7 @@ typedef struct xfs_perag {
#define XFS_ICI_NO_TAG (-1) /* special flag for an untagged lookup
in xfs_inode_ag_iterator */
#define XFS_ICI_RECLAIM_TAG 0 /* inode is to be reclaimed */
+#define XFS_ICI_EOFBLOCKS_TAG 1 /* inode has blocks beyond EOF */
#define XFS_AG_MAXLEVELS(mp) ((mp)->m_ag_maxlevels)
#define XFS_MIN_FREELIST_RAW(bl,cl,mp) \
diff --git a/include/xfs_alloc.h b/include/xfs_alloc.h
index 895009a..99d0a61 100644
--- a/include/xfs_alloc.h
+++ b/include/xfs_alloc.h
@@ -19,10 +19,12 @@
#define __XFS_ALLOC_H__
struct xfs_buf;
+struct xfs_btree_cur;
struct xfs_mount;
struct xfs_perag;
struct xfs_trans;
-struct xfs_busy_extent;
+
+extern struct workqueue_struct *xfs_alloc_wq;
/*
* Freespace allocation types. Argument to xfs_alloc_[v]extent.
@@ -74,6 +76,22 @@ typedef unsigned int xfs_alloctype_t;
#define XFS_ALLOC_SET_ASIDE(mp) (4 + ((mp)->m_sb.sb_agcount * 4))
/*
+ * When deciding how much space to allocate out of an AG, we limit the
+ * allocation maximum size to the size the AG. However, we cannot use all the
+ * blocks in the AG - some are permanently used by metadata. These
+ * blocks are generally:
+ * - the AG superblock, AGF, AGI and AGFL
+ * - the AGF (bno and cnt) and AGI btree root blocks
+ * - 4 blocks on the AGFL according to XFS_ALLOC_SET_ASIDE() limits
+ *
+ * The AG headers are sector sized, so the amount of space they take up is
+ * dependent on filesystem geometry. The others are all single blocks.
+ */
+#define XFS_ALLOC_AG_MAX_USABLE(mp) \
+ ((mp)->m_sb.sb_agblocks - XFS_BB_TO_FSB(mp, XFS_FSS_TO_BB(mp, 4)) - 7)
+
+
+/*
* Argument structure for xfs_alloc routines.
* This is turned into a structure to avoid having 20 arguments passed
* down several levels of the stack.
@@ -117,19 +135,6 @@ xfs_extlen_t
xfs_alloc_longest_free_extent(struct xfs_mount *mp,
struct xfs_perag *pag);
-#ifdef __KERNEL__
-
-void
-xfs_alloc_busy_insert(xfs_trans_t *tp,
- xfs_agnumber_t agno,
- xfs_agblock_t bno,
- xfs_extlen_t len);
-
-void
-xfs_alloc_busy_clear(struct xfs_mount *mp, struct xfs_busy_extent *busyp);
-
-#endif /* __KERNEL__ */
-
/*
* Compute and fill in value of m_ag_maxlevels.
*/
@@ -205,4 +210,28 @@ xfs_free_extent(
xfs_fsblock_t bno, /* starting block number of extent */
xfs_extlen_t len); /* length of extent */
+int /* error */
+xfs_alloc_lookup_le(
+ struct xfs_btree_cur *cur, /* btree cursor */
+ xfs_agblock_t bno, /* starting block of extent */
+ xfs_extlen_t len, /* length of extent */
+ int *stat); /* success/failure */
+
+int /* error */
+xfs_alloc_lookup_ge(
+ struct xfs_btree_cur *cur, /* btree cursor */
+ xfs_agblock_t bno, /* starting block of extent */
+ xfs_extlen_t len, /* length of extent */
+ int *stat); /* success/failure */
+
+int /* error */
+xfs_alloc_get_rec(
+ struct xfs_btree_cur *cur, /* btree cursor */
+ xfs_agblock_t *bno, /* output: starting block of extent */
+ xfs_extlen_t *len, /* output: length of extent */
+ int *stat); /* output: success/failure */
+
+extern const struct xfs_buf_ops xfs_agf_buf_ops;
+extern const struct xfs_buf_ops xfs_agfl_buf_ops;
+
#endif /* __XFS_ALLOC_H__ */
diff --git a/include/xfs_alloc_btree.h b/include/xfs_alloc_btree.h
index a6caa00..7e89a2b 100644
--- a/include/xfs_alloc_btree.h
+++ b/include/xfs_alloc_btree.h
@@ -51,20 +51,6 @@ typedef struct xfs_alloc_rec_incore {
typedef __be32 xfs_alloc_ptr_t;
/*
- * Minimum and maximum blocksize and sectorsize.
- * The blocksize upper limit is pretty much arbitrary.
- * The sectorsize upper limit is due to sizeof(sb_sectsize).
- */
-#define XFS_MIN_BLOCKSIZE_LOG 9 /* i.e. 512 bytes */
-#define XFS_MAX_BLOCKSIZE_LOG 16 /* i.e. 65536 bytes */
-#define XFS_MIN_BLOCKSIZE (1 << XFS_MIN_BLOCKSIZE_LOG)
-#define XFS_MAX_BLOCKSIZE (1 << XFS_MAX_BLOCKSIZE_LOG)
-#define XFS_MIN_SECTORSIZE_LOG 9 /* i.e. 512 bytes */
-#define XFS_MAX_SECTORSIZE_LOG 15 /* i.e. 32768 bytes */
-#define XFS_MIN_SECTORSIZE (1 << XFS_MIN_SECTORSIZE_LOG)
-#define XFS_MAX_SECTORSIZE (1 << XFS_MAX_SECTORSIZE_LOG)
-
-/*
* Block numbers in the AG:
* SB is sector 0, AGF is sector 1, AGI is sector 2, AGFL is sector 3.
*/
@@ -107,4 +93,6 @@ extern struct xfs_btree_cur *xfs_allocbt_init_cursor(struct xfs_mount *,
xfs_agnumber_t, xfs_btnum_t);
extern int xfs_allocbt_maxrecs(struct xfs_mount *, int, int);
+extern const struct xfs_buf_ops xfs_allocbt_buf_ops;
+
#endif /* __XFS_ALLOC_BTREE_H__ */
diff --git a/include/xfs_attr_leaf.h b/include/xfs_attr_leaf.h
index 9c7d22f..77de139 100644
--- a/include/xfs_attr_leaf.h
+++ b/include/xfs_attr_leaf.h
@@ -31,7 +31,6 @@
struct attrlist;
struct attrlist_cursor_kern;
struct xfs_attr_list_context;
-struct xfs_dabuf;
struct xfs_da_args;
struct xfs_da_state;
struct xfs_da_state_blk;
@@ -215,7 +214,7 @@ int xfs_attr_shortform_getvalue(struct xfs_da_args *args);
int xfs_attr_shortform_to_leaf(struct xfs_da_args *args);
int xfs_attr_shortform_remove(struct xfs_da_args *args);
int xfs_attr_shortform_list(struct xfs_attr_list_context *context);
-int xfs_attr_shortform_allfit(struct xfs_dabuf *bp, struct xfs_inode *dp);
+int xfs_attr_shortform_allfit(struct xfs_buf *bp, struct xfs_inode *dp);
int xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes);
@@ -223,7 +222,7 @@ int xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes);
* Internal routines when attribute fork size == XFS_LBSIZE(mp).
*/
int xfs_attr_leaf_to_node(struct xfs_da_args *args);
-int xfs_attr_leaf_to_shortform(struct xfs_dabuf *bp,
+int xfs_attr_leaf_to_shortform(struct xfs_buf *bp,
struct xfs_da_args *args, int forkoff);
int xfs_attr_leaf_clearflag(struct xfs_da_args *args);
int xfs_attr_leaf_setflag(struct xfs_da_args *args);
@@ -235,14 +234,14 @@ int xfs_attr_leaf_flipflags(xfs_da_args_t *args);
int xfs_attr_leaf_split(struct xfs_da_state *state,
struct xfs_da_state_blk *oldblk,
struct xfs_da_state_blk *newblk);
-int xfs_attr_leaf_lookup_int(struct xfs_dabuf *leaf,
+int xfs_attr_leaf_lookup_int(struct xfs_buf *leaf,
struct xfs_da_args *args);
-int xfs_attr_leaf_getvalue(struct xfs_dabuf *bp, struct xfs_da_args *args);
-int xfs_attr_leaf_add(struct xfs_dabuf *leaf_buffer,
+int xfs_attr_leaf_getvalue(struct xfs_buf *bp, struct xfs_da_args *args);
+int xfs_attr_leaf_add(struct xfs_buf *leaf_buffer,
struct xfs_da_args *args);
-int xfs_attr_leaf_remove(struct xfs_dabuf *leaf_buffer,
+int xfs_attr_leaf_remove(struct xfs_buf *leaf_buffer,
struct xfs_da_args *args);
-int xfs_attr_leaf_list_int(struct xfs_dabuf *bp,
+int xfs_attr_leaf_list_int(struct xfs_buf *bp,
struct xfs_attr_list_context *context);
/*
@@ -257,9 +256,15 @@ int xfs_attr_root_inactive(struct xfs_trans **trans, struct xfs_inode *dp);
/*
* Utility routines.
*/
-xfs_dahash_t xfs_attr_leaf_lasthash(struct xfs_dabuf *bp, int *count);
-int xfs_attr_leaf_order(struct xfs_dabuf *leaf1_bp,
- struct xfs_dabuf *leaf2_bp);
+xfs_dahash_t xfs_attr_leaf_lasthash(struct xfs_buf *bp, int *count);
+int xfs_attr_leaf_order(struct xfs_buf *leaf1_bp,
+ struct xfs_buf *leaf2_bp);
int xfs_attr_leaf_newentsize(int namelen, int valuelen, int blocksize,
int *local);
+int xfs_attr_leaf_read(struct xfs_trans *tp, struct xfs_inode *dp,
+ xfs_dablk_t bno, xfs_daddr_t mappedbno,
+ struct xfs_buf **bpp);
+
+extern const struct xfs_buf_ops xfs_attr_leaf_buf_ops;
+
#endif /* __XFS_ATTR_LEAF_H__ */
diff --git a/include/xfs_bmap.h b/include/xfs_bmap.h
index 3651191..de451a2 100644
--- a/include/xfs_bmap.h
+++ b/include/xfs_bmap.h
@@ -62,36 +62,32 @@ typedef struct xfs_bmap_free
#define XFS_BMAP_MAX_NMAP 4
/*
- * Flags for xfs_bmapi
+ * Flags for xfs_bmapi_*
*/
-#define XFS_BMAPI_WRITE 0x001 /* write operation: allocate space */
-#define XFS_BMAPI_DELAY 0x002 /* delayed write operation */
-#define XFS_BMAPI_ENTIRE 0x004 /* return entire extent, not trimmed */
-#define XFS_BMAPI_METADATA 0x008 /* mapping metadata not user data */
-#define XFS_BMAPI_ATTRFORK 0x010 /* use attribute fork not data */
-#define XFS_BMAPI_RSVBLOCKS 0x020 /* OK to alloc. reserved data blocks */
-#define XFS_BMAPI_PREALLOC 0x040 /* preallocation op: unwritten space */
-#define XFS_BMAPI_IGSTATE 0x080 /* Ignore state - */
+#define XFS_BMAPI_ENTIRE 0x001 /* return entire extent, not trimmed */
+#define XFS_BMAPI_METADATA 0x002 /* mapping metadata not user data */
+#define XFS_BMAPI_ATTRFORK 0x004 /* use attribute fork not data */
+#define XFS_BMAPI_PREALLOC 0x008 /* preallocation op: unwritten space */
+#define XFS_BMAPI_IGSTATE 0x010 /* Ignore state - */
/* combine contig. space */
-#define XFS_BMAPI_CONTIG 0x100 /* must allocate only one extent */
+#define XFS_BMAPI_CONTIG 0x020 /* must allocate only one extent */
/*
* unwritten extent conversion - this needs write cache flushing and no additional
* allocation alignments. When specified with XFS_BMAPI_PREALLOC it converts
* from written to unwritten, otherwise convert from unwritten to written.
*/
-#define XFS_BMAPI_CONVERT 0x200
+#define XFS_BMAPI_CONVERT 0x040
+#define XFS_BMAPI_STACK_SWITCH 0x080
#define XFS_BMAPI_FLAGS \
- { XFS_BMAPI_WRITE, "WRITE" }, \
- { XFS_BMAPI_DELAY, "DELAY" }, \
{ XFS_BMAPI_ENTIRE, "ENTIRE" }, \
{ XFS_BMAPI_METADATA, "METADATA" }, \
{ XFS_BMAPI_ATTRFORK, "ATTRFORK" }, \
- { XFS_BMAPI_RSVBLOCKS, "RSVBLOCKS" }, \
{ XFS_BMAPI_PREALLOC, "PREALLOC" }, \
{ XFS_BMAPI_IGSTATE, "IGSTATE" }, \
{ XFS_BMAPI_CONTIG, "CONTIG" }, \
- { XFS_BMAPI_CONVERT, "CONVERT" }
+ { XFS_BMAPI_CONVERT, "CONVERT" }, \
+ { XFS_BMAPI_STACK_SWITCH, "STACK_SWITCH" }
static inline int xfs_bmapi_aflag(int w)
@@ -115,23 +111,32 @@ static inline void xfs_bmap_init(xfs_bmap_free_t *flp, xfs_fsblock_t *fbp)
* Argument structure for xfs_bmap_alloc.
*/
typedef struct xfs_bmalloca {
- xfs_fsblock_t firstblock; /* i/o first block allocated */
- xfs_fsblock_t rval; /* starting block of new extent */
- xfs_fileoff_t off; /* offset in file filling in */
+ xfs_fsblock_t *firstblock; /* i/o first block allocated */
+ struct xfs_bmap_free *flist; /* bmap freelist */
struct xfs_trans *tp; /* transaction pointer */
struct xfs_inode *ip; /* incore inode pointer */
- struct xfs_bmbt_irec *prevp; /* extent before the new one */
- struct xfs_bmbt_irec *gotp; /* extent after, or delayed */
- xfs_extlen_t alen; /* i/o length asked/allocated */
+ struct xfs_bmbt_irec prev; /* extent before the new one */
+ struct xfs_bmbt_irec got; /* extent after, or delayed */
+
+ xfs_fileoff_t offset; /* offset in file filling in */
+ xfs_extlen_t length; /* i/o length asked/allocated */
+ xfs_fsblock_t blkno; /* starting block of new extent */
+
+ struct xfs_btree_cur *cur; /* btree cursor */
+ xfs_extnum_t idx; /* current extent index */
+ int nallocs;/* number of extents alloc'd */
+ int logflags;/* flags for transaction logging */
+
xfs_extlen_t total; /* total blocks needed for xaction */
xfs_extlen_t minlen; /* minimum allocation size (blocks) */
xfs_extlen_t minleft; /* amount must be left after alloc */
char eof; /* set if allocating past last extent */
char wasdel; /* replacing a delayed allocation */
char userdata;/* set if is user data */
- char low; /* low on space, using seq'l ags */
char aeof; /* allocated space at eof */
char conv; /* overwriting unwritten extents */
+ int flags;
+ int stack_switch;
} xfs_bmalloca_t;
/*
@@ -154,251 +159,65 @@ typedef struct xfs_bmalloca {
{ BMAP_RIGHT_FILLING, "RF" }, \
{ BMAP_ATTRFORK, "ATTR" }
-/*
- * Add bmap trace insert entries for all the contents of the extent list.
- *
- * Quite excessive tracing. Only do this for debug builds.
- */
#if defined(__KERNEL) && defined(DEBUG)
-void
-xfs_bmap_trace_exlist(
- struct xfs_inode *ip, /* incore inode pointer */
- xfs_extnum_t cnt, /* count of entries in list */
- int whichfork,
- unsigned long caller_ip); /* data or attr fork */
+void xfs_bmap_trace_exlist(struct xfs_inode *ip, xfs_extnum_t cnt,
+ int whichfork, unsigned long caller_ip);
#define XFS_BMAP_TRACE_EXLIST(ip,c,w) \
xfs_bmap_trace_exlist(ip,c,w, _THIS_IP_)
#else
#define XFS_BMAP_TRACE_EXLIST(ip,c,w)
#endif
-/*
- * Convert inode from non-attributed to attributed.
- * Must not be in a transaction, ip must not be locked.
- */
-int /* error code */
-xfs_bmap_add_attrfork(
- struct xfs_inode *ip, /* incore inode pointer */
- int size, /* space needed for new attribute */
- int rsvd); /* flag for reserved block allocation */
-
-/*
- * Add the extent to the list of extents to be free at transaction end.
- * The list is maintained sorted (by block number).
- */
-void
-xfs_bmap_add_free(
- xfs_fsblock_t bno, /* fs block number of extent */
- xfs_filblks_t len, /* length of extent */
- xfs_bmap_free_t *flist, /* list of extents */
- struct xfs_mount *mp); /* mount point structure */
-
-/*
- * Routine to clean up the free list data structure when
- * an error occurs during a transaction.
- */
-void
-xfs_bmap_cancel(
- xfs_bmap_free_t *flist); /* free list to clean up */
-
-/*
- * Compute and fill in the value of the maximum depth of a bmap btree
- * in this filesystem. Done once, during mount.
- */
-void
-xfs_bmap_compute_maxlevels(
- struct xfs_mount *mp, /* file system mount structure */
- int whichfork); /* data or attr fork */
-
-/*
- * Returns the file-relative block number of the first unused block in the file.
- * This is the lowest-address hole if the file has holes, else the first block
- * past the end of file.
- */
-int /* error */
-xfs_bmap_first_unused(
- struct xfs_trans *tp, /* transaction pointer */
- struct xfs_inode *ip, /* incore inode */
- xfs_extlen_t len, /* size of hole to find */
- xfs_fileoff_t *unused, /* unused block num */
- int whichfork); /* data or attr fork */
-
-/*
- * Returns the file-relative block number of the last block + 1 before
- * last_block (input value) in the file.
- * This is not based on i_size, it is based on the extent list.
- * Returns 0 for local files, as they do not have an extent list.
- */
-int /* error */
-xfs_bmap_last_before(
- struct xfs_trans *tp, /* transaction pointer */
- struct xfs_inode *ip, /* incore inode */
- xfs_fileoff_t *last_block, /* last block */
- int whichfork); /* data or attr fork */
-
-/*
- * Returns the file-relative block number of the first block past eof in
- * the file. This is not based on i_size, it is based on the extent list.
- * Returns 0 for local files, as they do not have an extent list.
- */
-int /* error */
-xfs_bmap_last_offset(
- struct xfs_trans *tp, /* transaction pointer */
- struct xfs_inode *ip, /* incore inode */
- xfs_fileoff_t *unused, /* last block num */
- int whichfork); /* data or attr fork */
-
-/*
- * Returns whether the selected fork of the inode has exactly one
- * block or not. For the data fork we check this matches di_size,
- * implying the file's range is 0..bsize-1.
- */
-int
-xfs_bmap_one_block(
- struct xfs_inode *ip, /* incore inode */
- int whichfork); /* data or attr fork */
-
-/*
- * Read in the extents to iu_extents.
- * All inode fields are set up by caller, we just traverse the btree
- * and copy the records in.
- */
-int /* error */
-xfs_bmap_read_extents(
- struct xfs_trans *tp, /* transaction pointer */
- struct xfs_inode *ip, /* incore inode */
- int whichfork); /* data or attr fork */
-
-/*
- * Map file blocks to filesystem blocks.
- * File range is given by the bno/len pair.
- * Adds blocks to file if a write ("flags & XFS_BMAPI_WRITE" set)
- * into a hole or past eof.
- * Only allocates blocks from a single allocation group,
- * to avoid locking problems.
- * The returned value in "firstblock" from the first call in a transaction
- * must be remembered and presented to subsequent calls in "firstblock".
- * An upper bound for the number of blocks to be allocated is supplied to
- * the first call in "total"; if no allocation group has that many free
- * blocks then the call will fail (return NULLFSBLOCK in "firstblock").
- */
-int /* error */
-xfs_bmapi(
- struct xfs_trans *tp, /* transaction pointer */
- struct xfs_inode *ip, /* incore inode */
- xfs_fileoff_t bno, /* starting file offs. mapped */
- xfs_filblks_t len, /* length to map in file */
- int flags, /* XFS_BMAPI_... */
- xfs_fsblock_t *firstblock, /* first allocated block
- controls a.g. for allocs */
- xfs_extlen_t total, /* total blocks needed */
- struct xfs_bmbt_irec *mval, /* output: map values */
- int *nmap, /* i/o: mval size/count */
- xfs_bmap_free_t *flist); /* i/o: list extents to free */
-
-/*
- * Map file blocks to filesystem blocks, simple version.
- * One block only, read-only.
- * For flags, only the XFS_BMAPI_ATTRFORK flag is examined.
- * For the other flag values, the effect is as if XFS_BMAPI_METADATA
- * was set and all the others were clear.
- */
-int /* error */
-xfs_bmapi_single(
- struct xfs_trans *tp, /* transaction pointer */
- struct xfs_inode *ip, /* incore inode */
- int whichfork, /* data or attr fork */
- xfs_fsblock_t *fsb, /* output: mapped block */
- xfs_fileoff_t bno); /* starting file offs. mapped */
-
-/*
- * Unmap (remove) blocks from a file.
- * If nexts is nonzero then the number of extents to remove is limited to
- * that value. If not all extents in the block range can be removed then
- * *done is set.
- */
-int /* error */
-xfs_bunmapi(
- struct xfs_trans *tp, /* transaction pointer */
- struct xfs_inode *ip, /* incore inode */
- xfs_fileoff_t bno, /* starting offset to unmap */
- xfs_filblks_t len, /* length to unmap in file */
- int flags, /* XFS_BMAPI_... */
- xfs_extnum_t nexts, /* number of extents max */
- xfs_fsblock_t *firstblock, /* first allocated block
- controls a.g. for allocs */
- xfs_bmap_free_t *flist, /* i/o: list extents to free */
- int *done); /* set if not done yet */
-
-/*
- * Check an extent list, which has just been read, for
- * any bit in the extent flag field.
- */
-int
-xfs_check_nostate_extents(
- struct xfs_ifork *ifp,
- xfs_extnum_t idx,
- xfs_extnum_t num);
-
-uint
-xfs_default_attroffset(
- struct xfs_inode *ip);
+int xfs_bmap_add_attrfork(struct xfs_inode *ip, int size, int rsvd);
+void xfs_bmap_add_free(xfs_fsblock_t bno, xfs_filblks_t len,
+ struct xfs_bmap_free *flist, struct xfs_mount *mp);
+void xfs_bmap_cancel(struct xfs_bmap_free *flist);
+void xfs_bmap_compute_maxlevels(struct xfs_mount *mp, int whichfork);
+int xfs_bmap_first_unused(struct xfs_trans *tp, struct xfs_inode *ip,
+ xfs_extlen_t len, xfs_fileoff_t *unused, int whichfork);
+int xfs_bmap_last_before(struct xfs_trans *tp, struct xfs_inode *ip,
+ xfs_fileoff_t *last_block, int whichfork);
+int xfs_bmap_last_offset(struct xfs_trans *tp, struct xfs_inode *ip,
+ xfs_fileoff_t *unused, int whichfork);
+int xfs_bmap_one_block(struct xfs_inode *ip, int whichfork);
+int xfs_bmap_read_extents(struct xfs_trans *tp, struct xfs_inode *ip,
+ int whichfork);
+int xfs_bmapi_read(struct xfs_inode *ip, xfs_fileoff_t bno,
+ xfs_filblks_t len, struct xfs_bmbt_irec *mval,
+ int *nmap, int flags);
+int xfs_bmapi_delay(struct xfs_inode *ip, xfs_fileoff_t bno,
+ xfs_filblks_t len, struct xfs_bmbt_irec *mval,
+ int *nmap, int flags);
+int xfs_bmapi_write(struct xfs_trans *tp, struct xfs_inode *ip,
+ xfs_fileoff_t bno, xfs_filblks_t len, int flags,
+ xfs_fsblock_t *firstblock, xfs_extlen_t total,
+ struct xfs_bmbt_irec *mval, int *nmap,
+ struct xfs_bmap_free *flist);
+int xfs_bunmapi(struct xfs_trans *tp, struct xfs_inode *ip,
+ xfs_fileoff_t bno, xfs_filblks_t len, int flags,
+ xfs_extnum_t nexts, xfs_fsblock_t *firstblock,
+ struct xfs_bmap_free *flist, int *done);
+int xfs_check_nostate_extents(struct xfs_ifork *ifp, xfs_extnum_t idx,
+ xfs_extnum_t num);
+uint xfs_default_attroffset(struct xfs_inode *ip);
#ifdef __KERNEL__
-
-/*
- * Routine to be called at transaction's end by xfs_bmapi, xfs_bunmapi
- * caller. Frees all the extents that need freeing, which must be done
- * last due to locking considerations.
- *
- * Return 1 if the given transaction was committed and a new one allocated,
- * and 0 otherwise.
- */
-int /* error */
-xfs_bmap_finish(
- struct xfs_trans **tp, /* transaction pointer addr */
- xfs_bmap_free_t *flist, /* i/o: list extents to free */
- int *committed); /* xact committed or not */
-
/* bmap to userspace formatter - copy to user & advance pointer */
typedef int (*xfs_bmap_format_t)(void **, struct getbmapx *, int *);
-/*
- * Get inode's extents as described in bmv, and format for output.
- */
-int /* error code */
-xfs_getbmap(
- xfs_inode_t *ip,
- struct getbmapx *bmv, /* user bmap structure */
- xfs_bmap_format_t formatter, /* format to user */
- void *arg); /* formatter arg */
+int xfs_bmap_finish(struct xfs_trans **tp, struct xfs_bmap_free *flist,
+ int *committed);
+int xfs_getbmap(struct xfs_inode *ip, struct getbmapx *bmv,
+ xfs_bmap_format_t formatter, void *arg);
+int xfs_bmap_eof(struct xfs_inode *ip, xfs_fileoff_t endoff,
+ int whichfork, int *eof);
+int xfs_bmap_count_blocks(struct xfs_trans *tp, struct xfs_inode *ip,
+ int whichfork, int *count);
+int xfs_bmap_punch_delalloc_range(struct xfs_inode *ip,
+ xfs_fileoff_t start_fsb, xfs_fileoff_t length);
-/*
- * Check if the endoff is outside the last extent. If so the caller will grow
- * the allocation to a stripe unit boundary
- */
-int
-xfs_bmap_eof(
- struct xfs_inode *ip,
- xfs_fileoff_t endoff,
- int whichfork,
- int *eof);
-
-/*
- * Count fsblocks of the given fork.
- */
-int
-xfs_bmap_count_blocks(
- xfs_trans_t *tp,
- struct xfs_inode *ip,
- int whichfork,
- int *count);
+xfs_daddr_t xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb);
-int
-xfs_bmap_punch_delalloc_range(
- struct xfs_inode *ip,
- xfs_fileoff_t start_fsb,
- xfs_fileoff_t length);
#endif /* __KERNEL__ */
#endif /* __XFS_BMAP_H__ */
diff --git a/include/xfs_bmap_btree.h b/include/xfs_bmap_btree.h
index 0e66c4e..88469ca 100644
--- a/include/xfs_bmap_btree.h
+++ b/include/xfs_bmap_btree.h
@@ -236,5 +236,6 @@ extern int xfs_bmbt_maxrecs(struct xfs_mount *, int blocklen, int leaf);
extern struct xfs_btree_cur *xfs_bmbt_init_cursor(struct xfs_mount *,
struct xfs_trans *, struct xfs_inode *, int);
+extern const struct xfs_buf_ops xfs_bmbt_buf_ops;
#endif /* __XFS_BMAP_BTREE_H__ */
diff --git a/include/xfs_btree.h b/include/xfs_btree.h
index 82fafc6..be1eb23 100644
--- a/include/xfs_btree.h
+++ b/include/xfs_btree.h
@@ -188,6 +188,8 @@ struct xfs_btree_ops {
__int64_t (*key_diff)(struct xfs_btree_cur *cur,
union xfs_btree_key *key);
+ const struct xfs_buf_ops *buf_ops;
+
#ifdef DEBUG
/* check that k1 is lower than k2 */
int (*keys_inorder)(struct xfs_btree_cur *cur,
@@ -281,7 +283,7 @@ typedef struct xfs_btree_cur
/*
* Convert from buffer to btree block header.
*/
-#define XFS_BUF_TO_BLOCK(bp) ((struct xfs_btree_block *)XFS_BUF_PTR(bp))
+#define XFS_BUF_TO_BLOCK(bp) ((struct xfs_btree_block *)((bp)->b_addr))
/*
@@ -374,7 +376,8 @@ xfs_btree_read_bufl(
xfs_fsblock_t fsbno, /* file system block number */
uint lock, /* lock flags for read_buf */
struct xfs_buf **bpp, /* buffer for fsbno */
- int refval);/* ref count value for buffer */
+ int refval, /* ref count value for buffer */
+ const struct xfs_buf_ops *ops);
/*
* Read-ahead the block, don't wait for it, don't return a buffer.
@@ -384,7 +387,8 @@ void /* error */
xfs_btree_reada_bufl(
struct xfs_mount *mp, /* file system mount point */
xfs_fsblock_t fsbno, /* file system block number */
- xfs_extlen_t count); /* count of filesystem blocks */
+ xfs_extlen_t count, /* count of filesystem blocks */
+ const struct xfs_buf_ops *ops);
/*
* Read-ahead the block, don't wait for it, don't return a buffer.
@@ -395,8 +399,20 @@ xfs_btree_reada_bufs(
struct xfs_mount *mp, /* file system mount point */
xfs_agnumber_t agno, /* allocation group number */
xfs_agblock_t agbno, /* allocation group block number */
- xfs_extlen_t count); /* count of filesystem blocks */
+ xfs_extlen_t count, /* count of filesystem blocks */
+ const struct xfs_buf_ops *ops);
+/*
+ * Initialise a new btree block header
+ */
+void
+xfs_btree_init_block(
+ struct xfs_mount *mp,
+ struct xfs_buf *bp,
+ __u32 magic,
+ __u16 level,
+ __u16 numrecs,
+ unsigned int flags);
/*
* Common btree core entry points.
diff --git a/include/xfs_buf_item.h b/include/xfs_buf_item.h
index a5efba9..ee36c88 100644
--- a/include/xfs_buf_item.h
+++ b/include/xfs_buf_item.h
@@ -21,23 +21,6 @@
extern kmem_zone_t *xfs_buf_item_zone;
/*
- * This is the structure used to lay out a buf log item in the
- * log. The data map describes which 128 byte chunks of the buffer
- * have been logged.
- * For 6.2 and beyond, this is XFS_LI_BUF. We use this to log everything.
- */
-typedef struct xfs_buf_log_format {
- unsigned short blf_type; /* buf log item type indicator */
- unsigned short blf_size; /* size of this item */
- ushort blf_flags; /* misc state */
- ushort blf_len; /* number of blocks in this buf */
- __int64_t blf_blkno; /* starting blkno of this buf */
- unsigned int blf_map_size; /* size of data bitmap in words */
- unsigned int blf_data_map[1];/* variable size bitmap of */
- /* regions of buffer in this item */
-} xfs_buf_log_format_t;
-
-/*
* This flag indicates that the buffer contains on disk inodes
* and requires special recovery handling.
*/
@@ -61,6 +44,23 @@ typedef struct xfs_buf_log_format {
#define NBWORD (NBBY * sizeof(unsigned int))
/*
+ * This is the structure used to lay out a buf log item in the
+ * log. The data map describes which 128 byte chunks of the buffer
+ * have been logged.
+ */
+#define XFS_BLF_DATAMAP_SIZE ((XFS_MAX_BLOCKSIZE / XFS_BLF_CHUNK) / NBWORD)
+
+typedef struct xfs_buf_log_format {
+ unsigned short blf_type; /* buf log item type indicator */
+ unsigned short blf_size; /* size of this item */
+ ushort blf_flags; /* misc state */
+ ushort blf_len; /* number of blocks in this buf */
+ __int64_t blf_blkno; /* starting blkno of this buf */
+ unsigned int blf_map_size; /* used size of data bitmap in words */
+ unsigned int blf_data_map[XFS_BLF_DATAMAP_SIZE]; /* dirty bitmap */
+} xfs_buf_log_format_t;
+
+/*
* buf log item flags
*/
#define XFS_BLI_HOLD 0x01
@@ -69,7 +69,7 @@ typedef struct xfs_buf_log_format {
#define XFS_BLI_LOGGED 0x08
#define XFS_BLI_INODE_ALLOC_BUF 0x10
#define XFS_BLI_STALE_INODE 0x20
-#define XFS_BLI_INODE_BUF 0x40
+#define XFS_BLI_INODE_BUF 0x40
#define XFS_BLI_FLAGS \
{ XFS_BLI_HOLD, "HOLD" }, \
@@ -80,6 +80,7 @@ typedef struct xfs_buf_log_format {
{ XFS_BLI_STALE_INODE, "STALE_INODE" }, \
{ XFS_BLI_INODE_BUF, "INODE_BUF" }
+
#ifdef __KERNEL__
struct xfs_buf;
@@ -97,11 +98,9 @@ typedef struct xfs_buf_log_item {
unsigned int bli_flags; /* misc flags */
unsigned int bli_recur; /* lock recursion count */
atomic_t bli_refcount; /* cnt of tp refs */
-#ifdef XFS_TRANS_DEBUG
- char *bli_orig; /* original buffer copy */
- char *bli_logged; /* bytes logged (bitmap) */
-#endif
- xfs_buf_log_format_t bli_format; /* in-log header */
+ int bli_format_count; /* count of headers */
+ struct xfs_buf_log_format *bli_formats; /* array of in-log header ptrs */
+ struct xfs_buf_log_format __bli_format; /* embedded in-log header */
} xfs_buf_log_item_t;
void xfs_buf_item_init(struct xfs_buf *, struct xfs_mount *);
@@ -114,16 +113,6 @@ void xfs_buf_attach_iodone(struct xfs_buf *,
void xfs_buf_iodone_callbacks(struct xfs_buf *);
void xfs_buf_iodone(struct xfs_buf *, struct xfs_log_item *);
-#ifdef XFS_TRANS_DEBUG
-void
-xfs_buf_item_flush_log_debug(
- struct xfs_buf *bp,
- uint first,
- uint last);
-#else
-#define xfs_buf_item_flush_log_debug(bp, first, last)
-#endif
-
#endif /* __KERNEL__ */
#endif /* __XFS_BUF_ITEM_H__ */
diff --git a/include/xfs_da_btree.h b/include/xfs_da_btree.h
index fe9f5a8..ee5170c 100644
--- a/include/xfs_da_btree.h
+++ b/include/xfs_da_btree.h
@@ -18,7 +18,6 @@
#ifndef __XFS_DA_BTREE_H__
#define __XFS_DA_BTREE_H__
-struct xfs_buf;
struct xfs_bmap_free;
struct xfs_inode;
struct xfs_mount;
@@ -32,7 +31,7 @@ struct zone;
/*
* This structure is common to both leaf nodes and non-leaf nodes in the Btree.
*
- * Is is used to manage a doubly linked list of all blocks at the same
+ * It is used to manage a doubly linked list of all blocks at the same
* level in the Btree, and to identify which type of block this is.
*/
#define XFS_DA_NODE_MAGIC 0xfebe /* magic number: non-leaf blocks */
@@ -133,35 +132,6 @@ typedef struct xfs_da_args {
{ XFS_DA_OP_CILOOKUP, "CILOOKUP" }
/*
- * Structure to describe buffer(s) for a block.
- * This is needed in the directory version 2 format case, when
- * multiple non-contiguous fsblocks might be needed to cover one
- * logical directory block.
- * If the buffer count is 1 then the data pointer points to the
- * same place as the b_addr field for the buffer, else to kmem_alloced memory.
- */
-typedef struct xfs_dabuf {
- int nbuf; /* number of buffer pointers present */
- short dirty; /* data needs to be copied back */
- short bbcount; /* how large is data in bbs */
- void *data; /* pointer for buffers' data */
-#ifdef XFS_DABUF_DEBUG
- inst_t *ra; /* return address of caller to make */
- struct xfs_dabuf *next; /* next in global chain */
- struct xfs_dabuf *prev; /* previous in global chain */
- struct xfs_buftarg *target; /* device for buffer */
- xfs_daddr_t blkno; /* daddr first in bps[0] */
-#endif
- struct xfs_buf *bps[1]; /* actually nbuf of these */
-} xfs_dabuf_t;
-#define XFS_DA_BUF_SIZE(n) \
- (sizeof(xfs_dabuf_t) + sizeof(struct xfs_buf *) * ((n) - 1))
-
-#ifdef XFS_DABUF_DEBUG
-extern xfs_dabuf_t *xfs_dabuf_global_list;
-#endif
-
-/*
* Storage for holding state during Btree searches and split/join ops.
*
* Only need space for 5 intermediate nodes. With a minimum of 62-way
@@ -169,7 +139,7 @@ extern xfs_dabuf_t *xfs_dabuf_global_list;
* which is slightly more than enough.
*/
typedef struct xfs_da_state_blk {
- xfs_dabuf_t *bp; /* buffer containing block */
+ struct xfs_buf *bp; /* buffer containing block */
xfs_dablk_t blkno; /* filesystem blkno of buffer */
xfs_daddr_t disk_blkno; /* on-disk blkno (in BBs) of buffer */
int index; /* relevant index into block */
@@ -222,7 +192,7 @@ struct xfs_nameops {
* Routines used for growing the Btree.
*/
int xfs_da_node_create(xfs_da_args_t *args, xfs_dablk_t blkno, int level,
- xfs_dabuf_t **bpp, int whichfork);
+ struct xfs_buf **bpp, int whichfork);
int xfs_da_split(xfs_da_state_t *state);
/*
@@ -243,21 +213,28 @@ int xfs_da_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path,
*/
int xfs_da_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
xfs_da_state_blk_t *new_blk);
+int xfs_da_node_read(struct xfs_trans *tp, struct xfs_inode *dp,
+ xfs_dablk_t bno, xfs_daddr_t mappedbno,
+ struct xfs_buf **bpp, int which_fork);
/*
* Utility routines.
*/
int xfs_da_grow_inode(xfs_da_args_t *args, xfs_dablk_t *new_blkno);
+int xfs_da_grow_inode_int(struct xfs_da_args *args, xfs_fileoff_t *bno,
+ int count);
int xfs_da_get_buf(struct xfs_trans *trans, struct xfs_inode *dp,
xfs_dablk_t bno, xfs_daddr_t mappedbno,
- xfs_dabuf_t **bp, int whichfork);
+ struct xfs_buf **bp, int whichfork);
int xfs_da_read_buf(struct xfs_trans *trans, struct xfs_inode *dp,
xfs_dablk_t bno, xfs_daddr_t mappedbno,
- xfs_dabuf_t **bpp, int whichfork);
+ struct xfs_buf **bpp, int whichfork,
+ const struct xfs_buf_ops *ops);
xfs_daddr_t xfs_da_reada_buf(struct xfs_trans *trans, struct xfs_inode *dp,
- xfs_dablk_t bno, int whichfork);
+ xfs_dablk_t bno, xfs_daddr_t mapped_bno,
+ int whichfork, const struct xfs_buf_ops *ops);
int xfs_da_shrink_inode(xfs_da_args_t *args, xfs_dablk_t dead_blkno,
- xfs_dabuf_t *dead_buf);
+ struct xfs_buf *dead_buf);
uint xfs_da_hashname(const __uint8_t *name_string, int name_length);
enum xfs_dacmp xfs_da_compname(struct xfs_da_args *args,
@@ -267,15 +244,7 @@ enum xfs_dacmp xfs_da_compname(struct xfs_da_args *args,
xfs_da_state_t *xfs_da_state_alloc(void);
void xfs_da_state_free(xfs_da_state_t *state);
-void xfs_da_buf_done(xfs_dabuf_t *dabuf);
-void xfs_da_log_buf(struct xfs_trans *tp, xfs_dabuf_t *dabuf, uint first,
- uint last);
-void xfs_da_brelse(struct xfs_trans *tp, xfs_dabuf_t *dabuf);
-void xfs_da_binval(struct xfs_trans *tp, xfs_dabuf_t *dabuf);
-xfs_daddr_t xfs_da_blkno(xfs_dabuf_t *dabuf);
-
extern struct kmem_zone *xfs_da_state_zone;
-extern struct kmem_zone *xfs_dabuf_zone;
extern const struct xfs_nameops xfs_default_nameops;
#endif /* __XFS_DA_BTREE_H__ */
diff --git a/include/xfs_dinode.h b/include/xfs_dinode.h
index dffba9b..1d9643b 100644
--- a/include/xfs_dinode.h
+++ b/include/xfs_dinode.h
@@ -33,7 +33,7 @@ typedef struct xfs_timestamp {
* variable size the leftover area split into a data and an attribute fork.
* The format of the data and attribute fork depends on the format of the
* inode as indicated by di_format and di_aformat. To access the data and
- * attribute use the XFS_DFORK_PTR, XFS_DFORK_DPTR, and XFS_DFORK_PTR macros
+ * attribute use the XFS_DFORK_DPTR, XFS_DFORK_APTR, and XFS_DFORK_PTR macros
* below.
*
* There is a very similar struct icdinode in xfs_inode which matches the
@@ -148,7 +148,7 @@ typedef enum xfs_dinode_fmt {
be32_to_cpu((dip)->di_nextents) : \
be16_to_cpu((dip)->di_anextents))
-#define XFS_BUF_TO_DINODE(bp) ((xfs_dinode_t *)XFS_BUF_PTR(bp))
+#define XFS_BUF_TO_DINODE(bp) ((xfs_dinode_t *)((bp)->b_addr))
/*
* For block and character special files the 32bit dev_t is stored at the
diff --git a/include/xfs_dir1.h b/include/xfs_dir1.h
new file mode 100644
index 0000000..d83537c
--- /dev/null
+++ b/include/xfs_dir1.h
@@ -0,0 +1,13 @@
+#ifndef __XFS_DIRV1_H
+#define __XFS_DIRV1_H 1
+
+#if XFS_BIG_INUMS
+typedef xfs_ino_t xfs_intino_t;
+#else
+typedef __uint32_t xfs_intino_t;
+#endif
+
+#include <xfs_dir_sf.h>
+#include <xfs_dir_leaf.h>
+
+#endif /* __XFS_DIRV1_H */
diff --git a/include/xfs_dir2.h b/include/xfs_dir2.h
index 74a3b10..2dd0cc0 100644
--- a/include/xfs_dir2.h
+++ b/include/xfs_dir2.h
@@ -16,49 +16,23 @@
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*/
#ifndef __XFS_DIR2_H__
-#define __XFS_DIR2_H__
+#define __XFS_DIR2_H__
-struct uio;
-struct xfs_dabuf;
-struct xfs_da_args;
-struct xfs_dir2_put_args;
struct xfs_bmap_free;
+struct xfs_da_args;
struct xfs_inode;
struct xfs_mount;
struct xfs_trans;
+struct xfs_dir2_sf_hdr;
+struct xfs_dir2_sf_entry;
+struct xfs_dir2_data_hdr;
+struct xfs_dir2_data_entry;
+struct xfs_dir2_data_unused;
+struct xfs_dir2_data_free;
-/*
- * Directory version 2.
- * There are 4 possible formats:
- * shortform
- * single block - data with embedded leaf at the end
- * multiple data blocks, single leaf+freeindex block
- * data blocks, node&leaf blocks (btree), freeindex blocks
- *
- * The shortform format is in xfs_dir2_sf.h.
- * The single block format is in xfs_dir2_block.h.
- * The data block format is in xfs_dir2_data.h.
- * The leaf and freeindex block formats are in xfs_dir2_leaf.h.
- * Node blocks are the same as the other version, in xfs_da_btree.h.
- */
-
-/*
- * Byte offset in data block and shortform entry.
- */
-typedef __uint16_t xfs_dir2_data_off_t;
-#define NULLDATAOFF 0xffffU
typedef uint xfs_dir2_data_aoff_t; /* argument form */
-
-/*
- * Directory block number (logical dirblk in file)
- */
typedef __uint32_t xfs_dir2_db_t;
-/*
- * Byte offset in a directory.
- */
-typedef xfs_off_t xfs_dir2_off_t;
-
extern struct xfs_name xfs_name_dotdot;
/*
@@ -86,21 +60,44 @@ extern int xfs_dir_replace(struct xfs_trans *tp, struct xfs_inode *dp,
struct xfs_bmap_free *flist, xfs_extlen_t tot);
extern int xfs_dir_canenter(struct xfs_trans *tp, struct xfs_inode *dp,
struct xfs_name *name, uint resblks);
-extern int xfs_dir_ino_validate(struct xfs_mount *mp, xfs_ino_t ino);
/*
- * Utility routines for v2 directories.
+ * Direct call from the bmap code, bypassing the generic directory layer.
*/
-extern int xfs_dir2_grow_inode(struct xfs_da_args *args, int space,
- xfs_dir2_db_t *dbp);
-extern int xfs_dir2_isblock(struct xfs_trans *tp, struct xfs_inode *dp,
- int *vp);
-extern int xfs_dir2_isleaf(struct xfs_trans *tp, struct xfs_inode *dp,
- int *vp);
+extern int xfs_dir2_sf_to_block(struct xfs_da_args *args);
+
+/*
+ * used by db
+ */
+extern xfs_ino_t xfs_dir2_sf_get_parent_ino(struct xfs_dir2_sf_hdr *sfp);
+extern void xfs_dir2_sf_put_parent_ino(struct xfs_dir2_sf_hdr *sfp,
+ xfs_ino_t ino);
+extern xfs_ino_t xfs_dir2_sfe_get_ino(struct xfs_dir2_sf_hdr *sfp,
+ struct xfs_dir2_sf_entry *sfep);
+extern void xfs_dir2_sfe_put_ino( struct xfs_dir2_sf_hdr *,
+ struct xfs_dir2_sf_entry *sfep, xfs_ino_t ino);
+
+extern int xfs_dir2_isblock(struct xfs_trans *tp, struct xfs_inode *dp, int *r);
+extern int xfs_dir2_isleaf(struct xfs_trans *tp, struct xfs_inode *dp, int *r);
extern int xfs_dir2_shrink_inode(struct xfs_da_args *args, xfs_dir2_db_t db,
- struct xfs_dabuf *bp);
+ struct xfs_buf *bp);
+
+extern void xfs_dir2_data_freescan(struct xfs_mount *mp,
+ struct xfs_dir2_data_hdr *hdr, int *loghead);
+extern void xfs_dir2_data_log_entry(struct xfs_trans *tp, struct xfs_buf *bp,
+ struct xfs_dir2_data_entry *dep);
+extern void xfs_dir2_data_log_header(struct xfs_trans *tp,
+ struct xfs_buf *bp);
+extern void xfs_dir2_data_log_unused(struct xfs_trans *tp, struct xfs_buf *bp,
+ struct xfs_dir2_data_unused *dup);
+extern void xfs_dir2_data_make_free(struct xfs_trans *tp, struct xfs_buf *bp,
+ xfs_dir2_data_aoff_t offset, xfs_dir2_data_aoff_t len,
+ int *needlogp, int *needscanp);
+extern void xfs_dir2_data_use_free(struct xfs_trans *tp, struct xfs_buf *bp,
+ struct xfs_dir2_data_unused *dup, xfs_dir2_data_aoff_t offset,
+ xfs_dir2_data_aoff_t len, int *needlogp, int *needscanp);
-extern int xfs_dir_cilookup_result(struct xfs_da_args *args,
- const unsigned char *name, int len);
+extern struct xfs_dir2_data_free *xfs_dir2_data_freefind(
+ struct xfs_dir2_data_hdr *hdr, struct xfs_dir2_data_unused *dup);
#endif /* __XFS_DIR2_H__ */
diff --git a/include/xfs_dir2_block.h b/include/xfs_dir2_block.h
deleted file mode 100644
index 10e6896..0000000
--- a/include/xfs_dir2_block.h
+++ /dev/null
@@ -1,92 +0,0 @@
-/*
- * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc.
- * All Rights Reserved.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it would be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write the Free Software Foundation,
- * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
- */
-#ifndef __XFS_DIR2_BLOCK_H__
-#define __XFS_DIR2_BLOCK_H__
-
-/*
- * xfs_dir2_block.h
- * Directory version 2, single block format structures
- */
-
-struct uio;
-struct xfs_dabuf;
-struct xfs_da_args;
-struct xfs_dir2_data_hdr;
-struct xfs_dir2_leaf_entry;
-struct xfs_inode;
-struct xfs_mount;
-struct xfs_trans;
-
-/*
- * The single block format is as follows:
- * xfs_dir2_data_hdr_t structure
- * xfs_dir2_data_entry_t and xfs_dir2_data_unused_t structures
- * xfs_dir2_leaf_entry_t structures
- * xfs_dir2_block_tail_t structure
- */
-
-#define XFS_DIR2_BLOCK_MAGIC 0x58443242 /* XD2B: for one block dirs */
-
-typedef struct xfs_dir2_block_tail {
- __be32 count; /* count of leaf entries */
- __be32 stale; /* count of stale lf entries */
-} xfs_dir2_block_tail_t;
-
-/*
- * Generic single-block structure, for xfs_db.
- */
-typedef struct xfs_dir2_block {
- xfs_dir2_data_hdr_t hdr; /* magic XFS_DIR2_BLOCK_MAGIC */
- xfs_dir2_data_union_t u[1];
- xfs_dir2_leaf_entry_t leaf[1];
- xfs_dir2_block_tail_t tail;
-} xfs_dir2_block_t;
-
-/*
- * Pointer to the leaf header embedded in a data block (1-block format)
- */
-static inline xfs_dir2_block_tail_t *
-xfs_dir2_block_tail_p(struct xfs_mount *mp, xfs_dir2_block_t *block)
-{
- return (((xfs_dir2_block_tail_t *)
- ((char *)(block) + (mp)->m_dirblksize)) - 1);
-}
-
-/*
- * Pointer to the leaf entries embedded in a data block (1-block format)
- */
-static inline struct xfs_dir2_leaf_entry *
-xfs_dir2_block_leaf_p(xfs_dir2_block_tail_t *btp)
-{
- return ((struct xfs_dir2_leaf_entry *)btp) - be32_to_cpu(btp->count);
-}
-
-/*
- * Function declarations.
- */
-extern int xfs_dir2_block_addname(struct xfs_da_args *args);
-extern int xfs_dir2_block_getdents(struct xfs_inode *dp, void *dirent,
- xfs_off_t *offset, filldir_t filldir);
-extern int xfs_dir2_block_lookup(struct xfs_da_args *args);
-extern int xfs_dir2_block_removename(struct xfs_da_args *args);
-extern int xfs_dir2_block_replace(struct xfs_da_args *args);
-extern int xfs_dir2_leaf_to_block(struct xfs_da_args *args,
- struct xfs_dabuf *lbp, struct xfs_dabuf *dbp);
-extern int xfs_dir2_sf_to_block(struct xfs_da_args *args);
-
-#endif /* __XFS_DIR2_BLOCK_H__ */
diff --git a/include/xfs_dir2_data.h b/include/xfs_dir2_data.h
deleted file mode 100644
index efbc290..0000000
--- a/include/xfs_dir2_data.h
+++ /dev/null
@@ -1,184 +0,0 @@
-/*
- * Copyright (c) 2000,2005 Silicon Graphics, Inc.
- * All Rights Reserved.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it would be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write the Free Software Foundation,
- * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
- */
-#ifndef __XFS_DIR2_DATA_H__
-#define __XFS_DIR2_DATA_H__
-
-/*
- * Directory format 2, data block structures.
- */
-
-struct xfs_dabuf;
-struct xfs_da_args;
-struct xfs_inode;
-struct xfs_trans;
-
-/*
- * Constants.
- */
-#define XFS_DIR2_DATA_MAGIC 0x58443244 /* XD2D: for multiblock dirs */
-#define XFS_DIR2_DATA_ALIGN_LOG 3 /* i.e., 8 bytes */
-#define XFS_DIR2_DATA_ALIGN (1 << XFS_DIR2_DATA_ALIGN_LOG)
-#define XFS_DIR2_DATA_FREE_TAG 0xffff
-#define XFS_DIR2_DATA_FD_COUNT 3
-
-/*
- * Directory address space divided into sections,
- * spaces separated by 32GB.
- */
-#define XFS_DIR2_SPACE_SIZE (1ULL << (32 + XFS_DIR2_DATA_ALIGN_LOG))
-#define XFS_DIR2_DATA_SPACE 0
-#define XFS_DIR2_DATA_OFFSET (XFS_DIR2_DATA_SPACE * XFS_DIR2_SPACE_SIZE)
-#define XFS_DIR2_DATA_FIRSTDB(mp) \
- xfs_dir2_byte_to_db(mp, XFS_DIR2_DATA_OFFSET)
-
-/*
- * Offsets of . and .. in data space (always block 0)
- */
-#define XFS_DIR2_DATA_DOT_OFFSET \
- ((xfs_dir2_data_aoff_t)sizeof(xfs_dir2_data_hdr_t))
-#define XFS_DIR2_DATA_DOTDOT_OFFSET \
- (XFS_DIR2_DATA_DOT_OFFSET + xfs_dir2_data_entsize(1))
-#define XFS_DIR2_DATA_FIRST_OFFSET \
- (XFS_DIR2_DATA_DOTDOT_OFFSET + xfs_dir2_data_entsize(2))
-
-/*
- * Structures.
- */
-
-/*
- * Describe a free area in the data block.
- * The freespace will be formatted as a xfs_dir2_data_unused_t.
- */
-typedef struct xfs_dir2_data_free {
- __be16 offset; /* start of freespace */
- __be16 length; /* length of freespace */
-} xfs_dir2_data_free_t;
-
-/*
- * Header for the data blocks.
- * Always at the beginning of a directory-sized block.
- * The code knows that XFS_DIR2_DATA_FD_COUNT is 3.
- */
-typedef struct xfs_dir2_data_hdr {
- __be32 magic; /* XFS_DIR2_DATA_MAGIC */
- /* or XFS_DIR2_BLOCK_MAGIC */
- xfs_dir2_data_free_t bestfree[XFS_DIR2_DATA_FD_COUNT];
-} xfs_dir2_data_hdr_t;
-
-/*
- * Active entry in a data block. Aligned to 8 bytes.
- * Tag appears as the last 2 bytes.
- */
-typedef struct xfs_dir2_data_entry {
- __be64 inumber; /* inode number */
- __u8 namelen; /* name length */
- __u8 name[1]; /* name bytes, no null */
- /* variable offset */
- __be16 tag; /* starting offset of us */
-} xfs_dir2_data_entry_t;
-
-/*
- * Unused entry in a data block. Aligned to 8 bytes.
- * Tag appears as the last 2 bytes.
- */
-typedef struct xfs_dir2_data_unused {
- __be16 freetag; /* XFS_DIR2_DATA_FREE_TAG */
- __be16 length; /* total free length */
- /* variable offset */
- __be16 tag; /* starting offset of us */
-} xfs_dir2_data_unused_t;
-
-typedef union {
- xfs_dir2_data_entry_t entry;
- xfs_dir2_data_unused_t unused;
-} xfs_dir2_data_union_t;
-
-/*
- * Generic data block structure, for xfs_db.
- */
-typedef struct xfs_dir2_data {
- xfs_dir2_data_hdr_t hdr; /* magic XFS_DIR2_DATA_MAGIC */
- xfs_dir2_data_union_t u[1];
-} xfs_dir2_data_t;
-
-/*
- * Macros.
- */
-
-/*
- * Size of a data entry.
- */
-static inline int xfs_dir2_data_entsize(int n)
-{
- return (int)roundup(offsetof(xfs_dir2_data_entry_t, name[0]) + (n) + \
- (uint)sizeof(xfs_dir2_data_off_t), XFS_DIR2_DATA_ALIGN);
-}
-
-/*
- * Pointer to an entry's tag word.
- */
-static inline __be16 *
-xfs_dir2_data_entry_tag_p(xfs_dir2_data_entry_t *dep)
-{
- return (__be16 *)((char *)dep +
- xfs_dir2_data_entsize(dep->namelen) - sizeof(__be16));
-}
-
-/*
- * Pointer to a freespace's tag word.
- */
-static inline __be16 *
-xfs_dir2_data_unused_tag_p(xfs_dir2_data_unused_t *dup)
-{
- return (__be16 *)((char *)dup +
- be16_to_cpu(dup->length) - sizeof(__be16));
-}
-
-/*
- * Function declarations.
- */
-#ifdef DEBUG
-extern void xfs_dir2_data_check(struct xfs_inode *dp, struct xfs_dabuf *bp);
-#else
-#define xfs_dir2_data_check(dp,bp)
-#endif
-extern xfs_dir2_data_free_t *xfs_dir2_data_freefind(xfs_dir2_data_t *d,
- xfs_dir2_data_unused_t *dup);
-extern xfs_dir2_data_free_t *xfs_dir2_data_freeinsert(xfs_dir2_data_t *d,
- xfs_dir2_data_unused_t *dup, int *loghead);
-extern void xfs_dir2_data_freescan(struct xfs_mount *mp, xfs_dir2_data_t *d,
- int *loghead);
-extern int xfs_dir2_data_init(struct xfs_da_args *args, xfs_dir2_db_t blkno,
- struct xfs_dabuf **bpp);
-extern void xfs_dir2_data_log_entry(struct xfs_trans *tp, struct xfs_dabuf *bp,
- xfs_dir2_data_entry_t *dep);
-extern void xfs_dir2_data_log_header(struct xfs_trans *tp,
- struct xfs_dabuf *bp);
-extern void xfs_dir2_data_log_unused(struct xfs_trans *tp, struct xfs_dabuf *bp,
- xfs_dir2_data_unused_t *dup);
-extern void xfs_dir2_data_make_free(struct xfs_trans *tp, struct xfs_dabuf *bp,
- xfs_dir2_data_aoff_t offset,
- xfs_dir2_data_aoff_t len, int *needlogp,
- int *needscanp);
-extern void xfs_dir2_data_use_free(struct xfs_trans *tp, struct xfs_dabuf *bp,
- xfs_dir2_data_unused_t *dup,
- xfs_dir2_data_aoff_t offset,
- xfs_dir2_data_aoff_t len, int *needlogp,
- int *needscanp);
-
-#endif /* __XFS_DIR2_DATA_H__ */
diff --git a/include/xfs_dir2_format.h b/include/xfs_dir2_format.h
new file mode 100644
index 0000000..0727098
--- /dev/null
+++ b/include/xfs_dir2_format.h
@@ -0,0 +1,597 @@
+/*
+ * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+#ifndef __XFS_DIR2_FORMAT_H__
+#define __XFS_DIR2_FORMAT_H__
+
+/*
+ * Directory version 2.
+ *
+ * There are 4 possible formats:
+ * - shortform - embedded into the inode
+ * - single block - data with embedded leaf at the end
+ * - multiple data blocks, single leaf+freeindex block
+ * - data blocks, node and leaf blocks (btree), freeindex blocks
+ *
+ * Note: many node blocks structures and constants are shared with the attr
+ * code and defined in xfs_da_btree.h.
+ */
+
+#define XFS_DIR2_BLOCK_MAGIC 0x58443242 /* XD2B: single block dirs */
+#define XFS_DIR2_DATA_MAGIC 0x58443244 /* XD2D: multiblock dirs */
+#define XFS_DIR2_FREE_MAGIC 0x58443246 /* XD2F: free index blocks */
+
+/*
+ * Byte offset in data block and shortform entry.
+ */
+typedef __uint16_t xfs_dir2_data_off_t;
+#define NULLDATAOFF 0xffffU
+typedef uint xfs_dir2_data_aoff_t; /* argument form */
+
+/*
+ * Normalized offset (in a data block) of the entry, really xfs_dir2_data_off_t.
+ * Only need 16 bits, this is the byte offset into the single block form.
+ */
+typedef struct { __uint8_t i[2]; } __arch_pack xfs_dir2_sf_off_t;
+
+/*
+ * Offset in data space of a data entry.
+ */
+typedef __uint32_t xfs_dir2_dataptr_t;
+#define XFS_DIR2_MAX_DATAPTR ((xfs_dir2_dataptr_t)0xffffffff)
+#define XFS_DIR2_NULL_DATAPTR ((xfs_dir2_dataptr_t)0)
+
+/*
+ * Byte offset in a directory.
+ */
+typedef xfs_off_t xfs_dir2_off_t;
+
+/*
+ * Directory block number (logical dirblk in file)
+ */
+typedef __uint32_t xfs_dir2_db_t;
+
+/*
+ * Inode number stored as 8 8-bit values.
+ */
+typedef struct { __uint8_t i[8]; } xfs_dir2_ino8_t;
+
+/*
+ * Inode number stored as 4 8-bit values.
+ * Works a lot of the time, when all the inode numbers in a directory
+ * fit in 32 bits.
+ */
+typedef struct { __uint8_t i[4]; } xfs_dir2_ino4_t;
+
+typedef union {
+ xfs_dir2_ino8_t i8;
+ xfs_dir2_ino4_t i4;
+} xfs_dir2_inou_t;
+#define XFS_DIR2_MAX_SHORT_INUM ((xfs_ino_t)0xffffffffULL)
+
+/*
+ * Directory layout when stored internal to an inode.
+ *
+ * Small directories are packed as tightly as possible so as to fit into the
+ * literal area of the inode. These "shortform" directories consist of a
+ * single xfs_dir2_sf_hdr header followed by zero or more xfs_dir2_sf_entry
+ * structures. Due the different inode number storage size and the variable
+ * length name field in the xfs_dir2_sf_entry all these structure are
+ * variable length, and the accessors in this file should be used to iterate
+ * over them.
+ */
+typedef struct xfs_dir2_sf_hdr {
+ __uint8_t count; /* count of entries */
+ __uint8_t i8count; /* count of 8-byte inode #s */
+ xfs_dir2_inou_t parent; /* parent dir inode number */
+} __arch_pack xfs_dir2_sf_hdr_t;
+
+typedef struct xfs_dir2_sf_entry {
+ __u8 namelen; /* actual name length */
+ xfs_dir2_sf_off_t offset; /* saved offset */
+ __u8 name[]; /* name, variable size */
+ /*
+ * A xfs_dir2_ino8_t or xfs_dir2_ino4_t follows here, at a
+ * variable offset after the name.
+ */
+} __arch_pack xfs_dir2_sf_entry_t;
+
+static inline int xfs_dir2_sf_hdr_size(int i8count)
+{
+ return sizeof(struct xfs_dir2_sf_hdr) -
+ (i8count == 0) *
+ (sizeof(xfs_dir2_ino8_t) - sizeof(xfs_dir2_ino4_t));
+}
+
+static inline xfs_dir2_data_aoff_t
+xfs_dir2_sf_get_offset(xfs_dir2_sf_entry_t *sfep)
+{
+ return get_unaligned_be16(&sfep->offset.i);
+}
+
+static inline void
+xfs_dir2_sf_put_offset(xfs_dir2_sf_entry_t *sfep, xfs_dir2_data_aoff_t off)
+{
+ put_unaligned_be16(off, &sfep->offset.i);
+}
+
+static inline int
+xfs_dir2_sf_entsize(struct xfs_dir2_sf_hdr *hdr, int len)
+{
+ return sizeof(struct xfs_dir2_sf_entry) + /* namelen + offset */
+ len + /* name */
+ (hdr->i8count ? /* ino */
+ sizeof(xfs_dir2_ino8_t) :
+ sizeof(xfs_dir2_ino4_t));
+}
+
+static inline struct xfs_dir2_sf_entry *
+xfs_dir2_sf_firstentry(struct xfs_dir2_sf_hdr *hdr)
+{
+ return (struct xfs_dir2_sf_entry *)
+ ((char *)hdr + xfs_dir2_sf_hdr_size(hdr->i8count));
+}
+
+static inline struct xfs_dir2_sf_entry *
+xfs_dir2_sf_nextentry(struct xfs_dir2_sf_hdr *hdr,
+ struct xfs_dir2_sf_entry *sfep)
+{
+ return (struct xfs_dir2_sf_entry *)
+ ((char *)sfep + xfs_dir2_sf_entsize(hdr, sfep->namelen));
+}
+
+
+/*
+ * Data block structures.
+ *
+ * A pure data block looks like the following drawing on disk:
+ *
+ * +-------------------------------------------------+
+ * | xfs_dir2_data_hdr_t |
+ * +-------------------------------------------------+
+ * | xfs_dir2_data_entry_t OR xfs_dir2_data_unused_t |
+ * | xfs_dir2_data_entry_t OR xfs_dir2_data_unused_t |
+ * | xfs_dir2_data_entry_t OR xfs_dir2_data_unused_t |
+ * | ... |
+ * +-------------------------------------------------+
+ * | unused space |
+ * +-------------------------------------------------+
+ *
+ * As all the entries are variable size structures the accessors below should
+ * be used to iterate over them.
+ *
+ * In addition to the pure data blocks for the data and node formats,
+ * most structures are also used for the combined data/freespace "block"
+ * format below.
+ */
+
+#define XFS_DIR2_DATA_ALIGN_LOG 3 /* i.e., 8 bytes */
+#define XFS_DIR2_DATA_ALIGN (1 << XFS_DIR2_DATA_ALIGN_LOG)
+#define XFS_DIR2_DATA_FREE_TAG 0xffff
+#define XFS_DIR2_DATA_FD_COUNT 3
+
+/*
+ * Directory address space divided into sections,
+ * spaces separated by 32GB.
+ */
+#define XFS_DIR2_SPACE_SIZE (1ULL << (32 + XFS_DIR2_DATA_ALIGN_LOG))
+#define XFS_DIR2_DATA_SPACE 0
+#define XFS_DIR2_DATA_OFFSET (XFS_DIR2_DATA_SPACE * XFS_DIR2_SPACE_SIZE)
+#define XFS_DIR2_DATA_FIRSTDB(mp) \
+ xfs_dir2_byte_to_db(mp, XFS_DIR2_DATA_OFFSET)
+
+/*
+ * Offsets of . and .. in data space (always block 0)
+ */
+#define XFS_DIR2_DATA_DOT_OFFSET \
+ ((xfs_dir2_data_aoff_t)sizeof(struct xfs_dir2_data_hdr))
+#define XFS_DIR2_DATA_DOTDOT_OFFSET \
+ (XFS_DIR2_DATA_DOT_OFFSET + xfs_dir2_data_entsize(1))
+#define XFS_DIR2_DATA_FIRST_OFFSET \
+ (XFS_DIR2_DATA_DOTDOT_OFFSET + xfs_dir2_data_entsize(2))
+
+/*
+ * Describe a free area in the data block.
+ *
+ * The freespace will be formatted as a xfs_dir2_data_unused_t.
+ */
+typedef struct xfs_dir2_data_free {
+ __be16 offset; /* start of freespace */
+ __be16 length; /* length of freespace */
+} xfs_dir2_data_free_t;
+
+/*
+ * Header for the data blocks.
+ *
+ * The code knows that XFS_DIR2_DATA_FD_COUNT is 3.
+ */
+typedef struct xfs_dir2_data_hdr {
+ __be32 magic; /* XFS_DIR2_DATA_MAGIC or */
+ /* XFS_DIR2_BLOCK_MAGIC */
+ xfs_dir2_data_free_t bestfree[XFS_DIR2_DATA_FD_COUNT];
+} xfs_dir2_data_hdr_t;
+
+/*
+ * Active entry in a data block.
+ *
+ * Aligned to 8 bytes. After the variable length name field there is a
+ * 2 byte tag field, which can be accessed using xfs_dir2_data_entry_tag_p.
+ */
+typedef struct xfs_dir2_data_entry {
+ __be64 inumber; /* inode number */
+ __u8 namelen; /* name length */
+ __u8 name[]; /* name bytes, no null */
+ /* __be16 tag; */ /* starting offset of us */
+} xfs_dir2_data_entry_t;
+
+/*
+ * Unused entry in a data block.
+ *
+ * Aligned to 8 bytes. Tag appears as the last 2 bytes and must be accessed
+ * using xfs_dir2_data_unused_tag_p.
+ */
+typedef struct xfs_dir2_data_unused {
+ __be16 freetag; /* XFS_DIR2_DATA_FREE_TAG */
+ __be16 length; /* total free length */
+ /* variable offset */
+ __be16 tag; /* starting offset of us */
+} xfs_dir2_data_unused_t;
+
+/*
+ * Size of a data entry.
+ */
+static inline int xfs_dir2_data_entsize(int n)
+{
+ return (int)roundup(offsetof(struct xfs_dir2_data_entry, name[0]) + n +
+ (uint)sizeof(xfs_dir2_data_off_t), XFS_DIR2_DATA_ALIGN);
+}
+
+/*
+ * Pointer to an entry's tag word.
+ */
+static inline __be16 *
+xfs_dir2_data_entry_tag_p(struct xfs_dir2_data_entry *dep)
+{
+ return (__be16 *)((char *)dep +
+ xfs_dir2_data_entsize(dep->namelen) - sizeof(__be16));
+}
+
+/*
+ * Pointer to a freespace's tag word.
+ */
+static inline __be16 *
+xfs_dir2_data_unused_tag_p(struct xfs_dir2_data_unused *dup)
+{
+ return (__be16 *)((char *)dup +
+ be16_to_cpu(dup->length) - sizeof(__be16));
+}
+
+/*
+ * Leaf block structures.
+ *
+ * A pure leaf block looks like the following drawing on disk:
+ *
+ * +---------------------------+
+ * | xfs_dir2_leaf_hdr_t |
+ * +---------------------------+
+ * | xfs_dir2_leaf_entry_t |
+ * | xfs_dir2_leaf_entry_t |
+ * | xfs_dir2_leaf_entry_t |
+ * | xfs_dir2_leaf_entry_t |
+ * | ... |
+ * +---------------------------+
+ * | xfs_dir2_data_off_t |
+ * | xfs_dir2_data_off_t |
+ * | xfs_dir2_data_off_t |
+ * | ... |
+ * +---------------------------+
+ * | xfs_dir2_leaf_tail_t |
+ * +---------------------------+
+ *
+ * The xfs_dir2_data_off_t members (bests) and tail are at the end of the block
+ * for single-leaf (magic = XFS_DIR2_LEAF1_MAGIC) blocks only, but not present
+ * for directories with separate leaf nodes and free space blocks
+ * (magic = XFS_DIR2_LEAFN_MAGIC).
+ *
+ * As all the entries are variable size structures the accessors below should
+ * be used to iterate over them.
+ */
+
+/*
+ * Offset of the leaf/node space. First block in this space
+ * is the btree root.
+ */
+#define XFS_DIR2_LEAF_SPACE 1
+#define XFS_DIR2_LEAF_OFFSET (XFS_DIR2_LEAF_SPACE * XFS_DIR2_SPACE_SIZE)
+#define XFS_DIR2_LEAF_FIRSTDB(mp) \
+ xfs_dir2_byte_to_db(mp, XFS_DIR2_LEAF_OFFSET)
+
+/*
+ * Leaf block header.
+ */
+typedef struct xfs_dir2_leaf_hdr {
+ xfs_da_blkinfo_t info; /* header for da routines */
+ __be16 count; /* count of entries */
+ __be16 stale; /* count of stale entries */
+} xfs_dir2_leaf_hdr_t;
+
+/*
+ * Leaf block entry.
+ */
+typedef struct xfs_dir2_leaf_entry {
+ __be32 hashval; /* hash value of name */
+ __be32 address; /* address of data entry */
+} xfs_dir2_leaf_entry_t;
+
+/*
+ * Leaf block tail.
+ */
+typedef struct xfs_dir2_leaf_tail {
+ __be32 bestcount;
+} xfs_dir2_leaf_tail_t;
+
+/*
+ * Leaf block.
+ */
+typedef struct xfs_dir2_leaf {
+ xfs_dir2_leaf_hdr_t hdr; /* leaf header */
+ xfs_dir2_leaf_entry_t ents[]; /* entries */
+} xfs_dir2_leaf_t;
+
+/*
+ * DB blocks here are logical directory block numbers, not filesystem blocks.
+ */
+
+static inline int xfs_dir2_max_leaf_ents(struct xfs_mount *mp)
+{
+ return (mp->m_dirblksize - (uint)sizeof(struct xfs_dir2_leaf_hdr)) /
+ (uint)sizeof(struct xfs_dir2_leaf_entry);
+}
+
+/*
+ * Get address of the bestcount field in the single-leaf block.
+ */
+static inline struct xfs_dir2_leaf_tail *
+xfs_dir2_leaf_tail_p(struct xfs_mount *mp, struct xfs_dir2_leaf *lp)
+{
+ return (struct xfs_dir2_leaf_tail *)
+ ((char *)lp + mp->m_dirblksize -
+ sizeof(struct xfs_dir2_leaf_tail));
+}
+
+/*
+ * Get address of the bests array in the single-leaf block.
+ */
+static inline __be16 *
+xfs_dir2_leaf_bests_p(struct xfs_dir2_leaf_tail *ltp)
+{
+ return (__be16 *)ltp - be32_to_cpu(ltp->bestcount);
+}
+
+/*
+ * Convert dataptr to byte in file space
+ */
+static inline xfs_dir2_off_t
+xfs_dir2_dataptr_to_byte(struct xfs_mount *mp, xfs_dir2_dataptr_t dp)
+{
+ return (xfs_dir2_off_t)dp << XFS_DIR2_DATA_ALIGN_LOG;
+}
+
+/*
+ * Convert byte in file space to dataptr. It had better be aligned.
+ */
+static inline xfs_dir2_dataptr_t
+xfs_dir2_byte_to_dataptr(struct xfs_mount *mp, xfs_dir2_off_t by)
+{
+ return (xfs_dir2_dataptr_t)(by >> XFS_DIR2_DATA_ALIGN_LOG);
+}
+
+/*
+ * Convert byte in space to (DB) block
+ */
+static inline xfs_dir2_db_t
+xfs_dir2_byte_to_db(struct xfs_mount *mp, xfs_dir2_off_t by)
+{
+ return (xfs_dir2_db_t)
+ (by >> (mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog));
+}
+
+/*
+ * Convert dataptr to a block number
+ */
+static inline xfs_dir2_db_t
+xfs_dir2_dataptr_to_db(struct xfs_mount *mp, xfs_dir2_dataptr_t dp)
+{
+ return xfs_dir2_byte_to_db(mp, xfs_dir2_dataptr_to_byte(mp, dp));
+}
+
+/*
+ * Convert byte in space to offset in a block
+ */
+static inline xfs_dir2_data_aoff_t
+xfs_dir2_byte_to_off(struct xfs_mount *mp, xfs_dir2_off_t by)
+{
+ return (xfs_dir2_data_aoff_t)(by &
+ ((1 << (mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog)) - 1));
+}
+
+/*
+ * Convert dataptr to a byte offset in a block
+ */
+static inline xfs_dir2_data_aoff_t
+xfs_dir2_dataptr_to_off(struct xfs_mount *mp, xfs_dir2_dataptr_t dp)
+{
+ return xfs_dir2_byte_to_off(mp, xfs_dir2_dataptr_to_byte(mp, dp));
+}
+
+/*
+ * Convert block and offset to byte in space
+ */
+static inline xfs_dir2_off_t
+xfs_dir2_db_off_to_byte(struct xfs_mount *mp, xfs_dir2_db_t db,
+ xfs_dir2_data_aoff_t o)
+{
+ return ((xfs_dir2_off_t)db <<
+ (mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog)) + o;
+}
+
+/*
+ * Convert block (DB) to block (dablk)
+ */
+static inline xfs_dablk_t
+xfs_dir2_db_to_da(struct xfs_mount *mp, xfs_dir2_db_t db)
+{
+ return (xfs_dablk_t)(db << mp->m_sb.sb_dirblklog);
+}
+
+/*
+ * Convert byte in space to (DA) block
+ */
+static inline xfs_dablk_t
+xfs_dir2_byte_to_da(struct xfs_mount *mp, xfs_dir2_off_t by)
+{
+ return xfs_dir2_db_to_da(mp, xfs_dir2_byte_to_db(mp, by));
+}
+
+/*
+ * Convert block and offset to dataptr
+ */
+static inline xfs_dir2_dataptr_t
+xfs_dir2_db_off_to_dataptr(struct xfs_mount *mp, xfs_dir2_db_t db,
+ xfs_dir2_data_aoff_t o)
+{
+ return xfs_dir2_byte_to_dataptr(mp, xfs_dir2_db_off_to_byte(mp, db, o));
+}
+
+/*
+ * Convert block (dablk) to block (DB)
+ */
+static inline xfs_dir2_db_t
+xfs_dir2_da_to_db(struct xfs_mount *mp, xfs_dablk_t da)
+{
+ return (xfs_dir2_db_t)(da >> mp->m_sb.sb_dirblklog);
+}
+
+/*
+ * Convert block (dablk) to byte offset in space
+ */
+static inline xfs_dir2_off_t
+xfs_dir2_da_to_byte(struct xfs_mount *mp, xfs_dablk_t da)
+{
+ return xfs_dir2_db_off_to_byte(mp, xfs_dir2_da_to_db(mp, da), 0);
+}
+
+/*
+ * Free space block defintions for the node format.
+ */
+
+/*
+ * Offset of the freespace index.
+ */
+#define XFS_DIR2_FREE_SPACE 2
+#define XFS_DIR2_FREE_OFFSET (XFS_DIR2_FREE_SPACE * XFS_DIR2_SPACE_SIZE)
+#define XFS_DIR2_FREE_FIRSTDB(mp) \
+ xfs_dir2_byte_to_db(mp, XFS_DIR2_FREE_OFFSET)
+
+typedef struct xfs_dir2_free_hdr {
+ __be32 magic; /* XFS_DIR2_FREE_MAGIC */
+ __be32 firstdb; /* db of first entry */
+ __be32 nvalid; /* count of valid entries */
+ __be32 nused; /* count of used entries */
+} xfs_dir2_free_hdr_t;
+
+typedef struct xfs_dir2_free {
+ xfs_dir2_free_hdr_t hdr; /* block header */
+ __be16 bests[]; /* best free counts */
+ /* unused entries are -1 */
+} xfs_dir2_free_t;
+
+static inline int xfs_dir2_free_max_bests(struct xfs_mount *mp)
+{
+ return (mp->m_dirblksize - sizeof(struct xfs_dir2_free_hdr)) /
+ sizeof(xfs_dir2_data_off_t);
+}
+
+/*
+ * Convert data space db to the corresponding free db.
+ */
+static inline xfs_dir2_db_t
+xfs_dir2_db_to_fdb(struct xfs_mount *mp, xfs_dir2_db_t db)
+{
+ return XFS_DIR2_FREE_FIRSTDB(mp) + db / xfs_dir2_free_max_bests(mp);
+}
+
+/*
+ * Convert data space db to the corresponding index in a free db.
+ */
+static inline int
+xfs_dir2_db_to_fdindex(struct xfs_mount *mp, xfs_dir2_db_t db)
+{
+ return db % xfs_dir2_free_max_bests(mp);
+}
+
+/*
+ * Single block format.
+ *
+ * The single block format looks like the following drawing on disk:
+ *
+ * +-------------------------------------------------+
+ * | xfs_dir2_data_hdr_t |
+ * +-------------------------------------------------+
+ * | xfs_dir2_data_entry_t OR xfs_dir2_data_unused_t |
+ * | xfs_dir2_data_entry_t OR xfs_dir2_data_unused_t |
+ * | xfs_dir2_data_entry_t OR xfs_dir2_data_unused_t :
+ * | ... |
+ * +-------------------------------------------------+
+ * | unused space |
+ * +-------------------------------------------------+
+ * | ... |
+ * | xfs_dir2_leaf_entry_t |
+ * | xfs_dir2_leaf_entry_t |
+ * +-------------------------------------------------+
+ * | xfs_dir2_block_tail_t |
+ * +-------------------------------------------------+
+ *
+ * As all the entries are variable size structures the accessors below should
+ * be used to iterate over them.
+ */
+
+typedef struct xfs_dir2_block_tail {
+ __be32 count; /* count of leaf entries */
+ __be32 stale; /* count of stale lf entries */
+} xfs_dir2_block_tail_t;
+
+/*
+ * Pointer to the leaf header embedded in a data block (1-block format)
+ */
+static inline struct xfs_dir2_block_tail *
+xfs_dir2_block_tail_p(struct xfs_mount *mp, struct xfs_dir2_data_hdr *hdr)
+{
+ return ((struct xfs_dir2_block_tail *)
+ ((char *)hdr + mp->m_dirblksize)) - 1;
+}
+
+/*
+ * Pointer to the leaf entries embedded in a data block (1-block format)
+ */
+static inline struct xfs_dir2_leaf_entry *
+xfs_dir2_block_leaf_p(struct xfs_dir2_block_tail *btp)
+{
+ return ((struct xfs_dir2_leaf_entry *)btp) - be32_to_cpu(btp->count);
+}
+
+#endif /* __XFS_DIR2_FORMAT_H__ */
diff --git a/include/xfs_dir2_leaf.h b/include/xfs_dir2_leaf.h
deleted file mode 100644
index 6c9539f..0000000
--- a/include/xfs_dir2_leaf.h
+++ /dev/null
@@ -1,253 +0,0 @@
-/*
- * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc.
- * All Rights Reserved.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it would be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write the Free Software Foundation,
- * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
- */
-#ifndef __XFS_DIR2_LEAF_H__
-#define __XFS_DIR2_LEAF_H__
-
-struct uio;
-struct xfs_dabuf;
-struct xfs_da_args;
-struct xfs_inode;
-struct xfs_mount;
-struct xfs_trans;
-
-/*
- * Offset of the leaf/node space. First block in this space
- * is the btree root.
- */
-#define XFS_DIR2_LEAF_SPACE 1
-#define XFS_DIR2_LEAF_OFFSET (XFS_DIR2_LEAF_SPACE * XFS_DIR2_SPACE_SIZE)
-#define XFS_DIR2_LEAF_FIRSTDB(mp) \
- xfs_dir2_byte_to_db(mp, XFS_DIR2_LEAF_OFFSET)
-
-/*
- * Offset in data space of a data entry.
- */
-typedef __uint32_t xfs_dir2_dataptr_t;
-#define XFS_DIR2_MAX_DATAPTR ((xfs_dir2_dataptr_t)0xffffffff)
-#define XFS_DIR2_NULL_DATAPTR ((xfs_dir2_dataptr_t)0)
-
-/*
- * Leaf block header.
- */
-typedef struct xfs_dir2_leaf_hdr {
- xfs_da_blkinfo_t info; /* header for da routines */
- __be16 count; /* count of entries */
- __be16 stale; /* count of stale entries */
-} xfs_dir2_leaf_hdr_t;
-
-/*
- * Leaf block entry.
- */
-typedef struct xfs_dir2_leaf_entry {
- __be32 hashval; /* hash value of name */
- __be32 address; /* address of data entry */
-} xfs_dir2_leaf_entry_t;
-
-/*
- * Leaf block tail.
- */
-typedef struct xfs_dir2_leaf_tail {
- __be32 bestcount;
-} xfs_dir2_leaf_tail_t;
-
-/*
- * Leaf block.
- * bests and tail are at the end of the block for single-leaf only
- * (magic = XFS_DIR2_LEAF1_MAGIC not XFS_DIR2_LEAFN_MAGIC).
- */
-typedef struct xfs_dir2_leaf {
- xfs_dir2_leaf_hdr_t hdr; /* leaf header */
- xfs_dir2_leaf_entry_t ents[1]; /* entries */
- /* ... */
- xfs_dir2_data_off_t bests[1]; /* best free counts */
- xfs_dir2_leaf_tail_t tail; /* leaf tail */
-} xfs_dir2_leaf_t;
-
-/*
- * DB blocks here are logical directory block numbers, not filesystem blocks.
- */
-
-static inline int xfs_dir2_max_leaf_ents(struct xfs_mount *mp)
-{
- return (int)(((mp)->m_dirblksize - (uint)sizeof(xfs_dir2_leaf_hdr_t)) /
- (uint)sizeof(xfs_dir2_leaf_entry_t));
-}
-
-/*
- * Get address of the bestcount field in the single-leaf block.
- */
-static inline xfs_dir2_leaf_tail_t *
-xfs_dir2_leaf_tail_p(struct xfs_mount *mp, xfs_dir2_leaf_t *lp)
-{
- return (xfs_dir2_leaf_tail_t *)
- ((char *)(lp) + (mp)->m_dirblksize -
- (uint)sizeof(xfs_dir2_leaf_tail_t));
-}
-
-/*
- * Get address of the bests array in the single-leaf block.
- */
-static inline __be16 *
-xfs_dir2_leaf_bests_p(xfs_dir2_leaf_tail_t *ltp)
-{
- return (__be16 *)ltp - be32_to_cpu(ltp->bestcount);
-}
-
-/*
- * Convert dataptr to byte in file space
- */
-static inline xfs_dir2_off_t
-xfs_dir2_dataptr_to_byte(struct xfs_mount *mp, xfs_dir2_dataptr_t dp)
-{
- return (xfs_dir2_off_t)(dp) << XFS_DIR2_DATA_ALIGN_LOG;
-}
-
-/*
- * Convert byte in file space to dataptr. It had better be aligned.
- */
-static inline xfs_dir2_dataptr_t
-xfs_dir2_byte_to_dataptr(struct xfs_mount *mp, xfs_dir2_off_t by)
-{
- return (xfs_dir2_dataptr_t)((by) >> XFS_DIR2_DATA_ALIGN_LOG);
-}
-
-/*
- * Convert byte in space to (DB) block
- */
-static inline xfs_dir2_db_t
-xfs_dir2_byte_to_db(struct xfs_mount *mp, xfs_dir2_off_t by)
-{
- return (xfs_dir2_db_t)((by) >> \
- ((mp)->m_sb.sb_blocklog + (mp)->m_sb.sb_dirblklog));
-}
-
-/*
- * Convert dataptr to a block number
- */
-static inline xfs_dir2_db_t
-xfs_dir2_dataptr_to_db(struct xfs_mount *mp, xfs_dir2_dataptr_t dp)
-{
- return xfs_dir2_byte_to_db(mp, xfs_dir2_dataptr_to_byte(mp, dp));
-}
-
-/*
- * Convert byte in space to offset in a block
- */
-static inline xfs_dir2_data_aoff_t
-xfs_dir2_byte_to_off(struct xfs_mount *mp, xfs_dir2_off_t by)
-{
- return (xfs_dir2_data_aoff_t)((by) & \
- ((1 << ((mp)->m_sb.sb_blocklog + (mp)->m_sb.sb_dirblklog)) - 1));
-}
-
-/*
- * Convert dataptr to a byte offset in a block
- */
-static inline xfs_dir2_data_aoff_t
-xfs_dir2_dataptr_to_off(struct xfs_mount *mp, xfs_dir2_dataptr_t dp)
-{
- return xfs_dir2_byte_to_off(mp, xfs_dir2_dataptr_to_byte(mp, dp));
-}
-
-/*
- * Convert block and offset to byte in space
- */
-static inline xfs_dir2_off_t
-xfs_dir2_db_off_to_byte(struct xfs_mount *mp, xfs_dir2_db_t db,
- xfs_dir2_data_aoff_t o)
-{
- return ((xfs_dir2_off_t)(db) << \
- ((mp)->m_sb.sb_blocklog + (mp)->m_sb.sb_dirblklog)) + (o);
-}
-
-/*
- * Convert block (DB) to block (dablk)
- */
-static inline xfs_dablk_t
-xfs_dir2_db_to_da(struct xfs_mount *mp, xfs_dir2_db_t db)
-{
- return (xfs_dablk_t)((db) << (mp)->m_sb.sb_dirblklog);
-}
-
-/*
- * Convert byte in space to (DA) block
- */
-static inline xfs_dablk_t
-xfs_dir2_byte_to_da(struct xfs_mount *mp, xfs_dir2_off_t by)
-{
- return xfs_dir2_db_to_da(mp, xfs_dir2_byte_to_db(mp, by));
-}
-
-/*
- * Convert block and offset to dataptr
- */
-static inline xfs_dir2_dataptr_t
-xfs_dir2_db_off_to_dataptr(struct xfs_mount *mp, xfs_dir2_db_t db,
- xfs_dir2_data_aoff_t o)
-{
- return xfs_dir2_byte_to_dataptr(mp, xfs_dir2_db_off_to_byte(mp, db, o));
-}
-
-/*
- * Convert block (dablk) to block (DB)
- */
-static inline xfs_dir2_db_t
-xfs_dir2_da_to_db(struct xfs_mount *mp, xfs_dablk_t da)
-{
- return (xfs_dir2_db_t)((da) >> (mp)->m_sb.sb_dirblklog);
-}
-
-/*
- * Convert block (dablk) to byte offset in space
- */
-static inline xfs_dir2_off_t
-xfs_dir2_da_to_byte(struct xfs_mount *mp, xfs_dablk_t da)
-{
- return xfs_dir2_db_off_to_byte(mp, xfs_dir2_da_to_db(mp, da), 0);
-}
-
-/*
- * Function declarations.
- */
-extern int xfs_dir2_block_to_leaf(struct xfs_da_args *args,
- struct xfs_dabuf *dbp);
-extern int xfs_dir2_leaf_addname(struct xfs_da_args *args);
-extern void xfs_dir2_leaf_compact(struct xfs_da_args *args,
- struct xfs_dabuf *bp);
-extern void xfs_dir2_leaf_compact_x1(struct xfs_dabuf *bp, int *indexp,
- int *lowstalep, int *highstalep,
- int *lowlogp, int *highlogp);
-extern int xfs_dir2_leaf_getdents(struct xfs_inode *dp, void *dirent,
- size_t bufsize, xfs_off_t *offset,
- filldir_t filldir);
-extern int xfs_dir2_leaf_init(struct xfs_da_args *args, xfs_dir2_db_t bno,
- struct xfs_dabuf **bpp, int magic);
-extern void xfs_dir2_leaf_log_ents(struct xfs_trans *tp, struct xfs_dabuf *bp,
- int first, int last);
-extern void xfs_dir2_leaf_log_header(struct xfs_trans *tp,
- struct xfs_dabuf *bp);
-extern int xfs_dir2_leaf_lookup(struct xfs_da_args *args);
-extern int xfs_dir2_leaf_removename(struct xfs_da_args *args);
-extern int xfs_dir2_leaf_replace(struct xfs_da_args *args);
-extern int xfs_dir2_leaf_search_hash(struct xfs_da_args *args,
- struct xfs_dabuf *lbp);
-extern int xfs_dir2_leaf_trim_data(struct xfs_da_args *args,
- struct xfs_dabuf *lbp, xfs_dir2_db_t db);
-extern int xfs_dir2_node_to_leaf(struct xfs_da_state *state);
-
-#endif /* __XFS_DIR2_LEAF_H__ */
diff --git a/include/xfs_dir2_node.h b/include/xfs_dir2_node.h
deleted file mode 100644
index 82dfe71..0000000
--- a/include/xfs_dir2_node.h
+++ /dev/null
@@ -1,100 +0,0 @@
-/*
- * Copyright (c) 2000,2005 Silicon Graphics, Inc.
- * All Rights Reserved.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it would be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write the Free Software Foundation,
- * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
- */
-#ifndef __XFS_DIR2_NODE_H__
-#define __XFS_DIR2_NODE_H__
-
-/*
- * Directory version 2, btree node format structures
- */
-
-struct uio;
-struct xfs_dabuf;
-struct xfs_da_args;
-struct xfs_da_state;
-struct xfs_da_state_blk;
-struct xfs_inode;
-struct xfs_trans;
-
-/*
- * Offset of the freespace index.
- */
-#define XFS_DIR2_FREE_SPACE 2
-#define XFS_DIR2_FREE_OFFSET (XFS_DIR2_FREE_SPACE * XFS_DIR2_SPACE_SIZE)
-#define XFS_DIR2_FREE_FIRSTDB(mp) \
- xfs_dir2_byte_to_db(mp, XFS_DIR2_FREE_OFFSET)
-
-#define XFS_DIR2_FREE_MAGIC 0x58443246 /* XD2F */
-
-typedef struct xfs_dir2_free_hdr {
- __be32 magic; /* XFS_DIR2_FREE_MAGIC */
- __be32 firstdb; /* db of first entry */
- __be32 nvalid; /* count of valid entries */
- __be32 nused; /* count of used entries */
-} xfs_dir2_free_hdr_t;
-
-typedef struct xfs_dir2_free {
- xfs_dir2_free_hdr_t hdr; /* block header */
- __be16 bests[1]; /* best free counts */
- /* unused entries are -1 */
-} xfs_dir2_free_t;
-
-#define XFS_DIR2_MAX_FREE_BESTS(mp) \
- (((mp)->m_dirblksize - (uint)sizeof(xfs_dir2_free_hdr_t)) / \
- (uint)sizeof(xfs_dir2_data_off_t))
-
-/*
- * Convert data space db to the corresponding free db.
- */
-static inline xfs_dir2_db_t
-xfs_dir2_db_to_fdb(struct xfs_mount *mp, xfs_dir2_db_t db)
-{
- return (XFS_DIR2_FREE_FIRSTDB(mp) + (db) / XFS_DIR2_MAX_FREE_BESTS(mp));
-}
-
-/*
- * Convert data space db to the corresponding index in a free db.
- */
-static inline int
-xfs_dir2_db_to_fdindex(struct xfs_mount *mp, xfs_dir2_db_t db)
-{
- return ((db) % XFS_DIR2_MAX_FREE_BESTS(mp));
-}
-
-extern int xfs_dir2_leaf_to_node(struct xfs_da_args *args,
- struct xfs_dabuf *lbp);
-extern xfs_dahash_t xfs_dir2_leafn_lasthash(struct xfs_dabuf *bp, int *count);
-extern int xfs_dir2_leafn_lookup_int(struct xfs_dabuf *bp,
- struct xfs_da_args *args, int *indexp,
- struct xfs_da_state *state);
-extern int xfs_dir2_leafn_order(struct xfs_dabuf *leaf1_bp,
- struct xfs_dabuf *leaf2_bp);
-extern int xfs_dir2_leafn_split(struct xfs_da_state *state,
- struct xfs_da_state_blk *oldblk,
- struct xfs_da_state_blk *newblk);
-extern int xfs_dir2_leafn_toosmall(struct xfs_da_state *state, int *action);
-extern void xfs_dir2_leafn_unbalance(struct xfs_da_state *state,
- struct xfs_da_state_blk *drop_blk,
- struct xfs_da_state_blk *save_blk);
-extern int xfs_dir2_node_addname(struct xfs_da_args *args);
-extern int xfs_dir2_node_lookup(struct xfs_da_args *args);
-extern int xfs_dir2_node_removename(struct xfs_da_args *args);
-extern int xfs_dir2_node_replace(struct xfs_da_args *args);
-extern int xfs_dir2_node_trim_free(struct xfs_da_args *args, xfs_fileoff_t fo,
- int *rvalp);
-
-#endif /* __XFS_DIR2_NODE_H__ */
diff --git a/include/xfs_dir2_sf.h b/include/xfs_dir2_sf.h
deleted file mode 100644
index 6ac44b5..0000000
--- a/include/xfs_dir2_sf.h
+++ /dev/null
@@ -1,171 +0,0 @@
-/*
- * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc.
- * All Rights Reserved.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it would be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write the Free Software Foundation,
- * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
- */
-#ifndef __XFS_DIR2_SF_H__
-#define __XFS_DIR2_SF_H__
-
-/*
- * Directory layout when stored internal to an inode.
- *
- * Small directories are packed as tightly as possible so as to
- * fit into the literal area of the inode.
- */
-
-struct uio;
-struct xfs_dabuf;
-struct xfs_da_args;
-struct xfs_dir2_block;
-struct xfs_inode;
-struct xfs_mount;
-struct xfs_trans;
-
-/*
- * Inode number stored as 8 8-bit values.
- */
-typedef struct { __uint8_t i[8]; } xfs_dir2_ino8_t;
-
-/*
- * Inode number stored as 4 8-bit values.
- * Works a lot of the time, when all the inode numbers in a directory
- * fit in 32 bits.
- */
-typedef struct { __uint8_t i[4]; } xfs_dir2_ino4_t;
-
-typedef union {
- xfs_dir2_ino8_t i8;
- xfs_dir2_ino4_t i4;
-} xfs_dir2_inou_t;
-#define XFS_DIR2_MAX_SHORT_INUM ((xfs_ino_t)0xffffffffULL)
-
-/*
- * Normalized offset (in a data block) of the entry, really xfs_dir2_data_off_t.
- * Only need 16 bits, this is the byte offset into the single block form.
- */
-typedef struct { __uint8_t i[2]; } __arch_pack xfs_dir2_sf_off_t;
-
-/*
- * The parent directory has a dedicated field, and the self-pointer must
- * be calculated on the fly.
- *
- * Entries are packed toward the top as tightly as possible. The header
- * and the elements must be memcpy'd out into a work area to get correct
- * alignment for the inode number fields.
- */
-typedef struct xfs_dir2_sf_hdr {
- __uint8_t count; /* count of entries */
- __uint8_t i8count; /* count of 8-byte inode #s */
- xfs_dir2_inou_t parent; /* parent dir inode number */
-} __arch_pack xfs_dir2_sf_hdr_t;
-
-typedef struct xfs_dir2_sf_entry {
- __uint8_t namelen; /* actual name length */
- xfs_dir2_sf_off_t offset; /* saved offset */
- __uint8_t name[1]; /* name, variable size */
- xfs_dir2_inou_t inumber; /* inode number, var. offset */
-} __arch_pack xfs_dir2_sf_entry_t;
-
-typedef struct xfs_dir2_sf {
- xfs_dir2_sf_hdr_t hdr; /* shortform header */
- xfs_dir2_sf_entry_t list[1]; /* shortform entries */
-} xfs_dir2_sf_t;
-
-static inline int xfs_dir2_sf_hdr_size(int i8count)
-{
- return ((uint)sizeof(xfs_dir2_sf_hdr_t) - \
- ((i8count) == 0) * \
- ((uint)sizeof(xfs_dir2_ino8_t) - (uint)sizeof(xfs_dir2_ino4_t)));
-}
-
-static inline xfs_dir2_inou_t *xfs_dir2_sf_inumberp(xfs_dir2_sf_entry_t *sfep)
-{
- return (xfs_dir2_inou_t *)&(sfep)->name[(sfep)->namelen];
-}
-
-static inline xfs_intino_t
-xfs_dir2_sf_get_inumber(xfs_dir2_sf_t *sfp, xfs_dir2_inou_t *from)
-{
- return ((sfp)->hdr.i8count == 0 ? \
- (xfs_intino_t)XFS_GET_DIR_INO4((from)->i4) : \
- (xfs_intino_t)XFS_GET_DIR_INO8((from)->i8));
-}
-
-static inline void xfs_dir2_sf_put_inumber(xfs_dir2_sf_t *sfp, xfs_ino_t *from,
- xfs_dir2_inou_t *to)
-{
- if ((sfp)->hdr.i8count == 0)
- XFS_PUT_DIR_INO4(*(from), (to)->i4);
- else
- XFS_PUT_DIR_INO8(*(from), (to)->i8);
-}
-
-static inline xfs_dir2_data_aoff_t
-xfs_dir2_sf_get_offset(xfs_dir2_sf_entry_t *sfep)
-{
- return INT_GET_UNALIGNED_16_BE(&(sfep)->offset.i);
-}
-
-static inline void
-xfs_dir2_sf_put_offset(xfs_dir2_sf_entry_t *sfep, xfs_dir2_data_aoff_t off)
-{
- INT_SET_UNALIGNED_16_BE(&(sfep)->offset.i, off);
-}
-
-static inline int xfs_dir2_sf_entsize_byname(xfs_dir2_sf_t *sfp, int len)
-{
- return ((uint)sizeof(xfs_dir2_sf_entry_t) - 1 + (len) - \
- ((sfp)->hdr.i8count == 0) * \
- ((uint)sizeof(xfs_dir2_ino8_t) - (uint)sizeof(xfs_dir2_ino4_t)));
-}
-
-static inline int
-xfs_dir2_sf_entsize_byentry(xfs_dir2_sf_t *sfp, xfs_dir2_sf_entry_t *sfep)
-{
- return ((uint)sizeof(xfs_dir2_sf_entry_t) - 1 + (sfep)->namelen - \
- ((sfp)->hdr.i8count == 0) * \
- ((uint)sizeof(xfs_dir2_ino8_t) - (uint)sizeof(xfs_dir2_ino4_t)));
-}
-
-static inline xfs_dir2_sf_entry_t *xfs_dir2_sf_firstentry(xfs_dir2_sf_t *sfp)
-{
- return ((xfs_dir2_sf_entry_t *) \
- ((char *)(sfp) + xfs_dir2_sf_hdr_size(sfp->hdr.i8count)));
-}
-
-static inline xfs_dir2_sf_entry_t *
-xfs_dir2_sf_nextentry(xfs_dir2_sf_t *sfp, xfs_dir2_sf_entry_t *sfep)
-{
- return ((xfs_dir2_sf_entry_t *) \
- ((char *)(sfep) + xfs_dir2_sf_entsize_byentry(sfp,sfep)));
-}
-
-/*
- * Functions.
- */
-extern int xfs_dir2_block_sfsize(struct xfs_inode *dp,
- struct xfs_dir2_block *block,
- xfs_dir2_sf_hdr_t *sfhp);
-extern int xfs_dir2_block_to_sf(struct xfs_da_args *args, struct xfs_dabuf *bp,
- int size, xfs_dir2_sf_hdr_t *sfhp);
-extern int xfs_dir2_sf_addname(struct xfs_da_args *args);
-extern int xfs_dir2_sf_create(struct xfs_da_args *args, xfs_ino_t pino);
-extern int xfs_dir2_sf_getdents(struct xfs_inode *dp, void *dirent,
- xfs_off_t *offset, filldir_t filldir);
-extern int xfs_dir2_sf_lookup(struct xfs_da_args *args);
-extern int xfs_dir2_sf_removename(struct xfs_da_args *args);
-extern int xfs_dir2_sf_replace(struct xfs_da_args *args);
-
-#endif /* __XFS_DIR2_SF_H__ */
diff --git a/include/xfs_fs.h b/include/xfs_fs.h
index faac5af..1cc1aa0 100644
--- a/include/xfs_fs.h
+++ b/include/xfs_fs.h
@@ -233,7 +233,8 @@ typedef struct xfs_fsop_resblks {
#define XFS_FSOP_GEOM_FLAGS_LOGV2 0x0100 /* log format version 2 */
#define XFS_FSOP_GEOM_FLAGS_SECTOR 0x0200 /* sector sizes >1BB */
#define XFS_FSOP_GEOM_FLAGS_ATTR2 0x0400 /* inline attributes rework */
-#define XFS_FSOP_GEOM_FLAGS_DIRV2CI 0x1000 /* ASCII only CI names */
+#define XFS_FSOP_GEOM_FLAGS_PROJID32 0x0800 /* 32-bit project IDs */
+#define XFS_FSOP_GEOM_FLAGS_DIRV2CI 0x1000 /* ASCII only CI names */
#define XFS_FSOP_GEOM_FLAGS_LAZYSB 0x4000 /* lazy superblock counters */
@@ -249,6 +250,11 @@ typedef struct xfs_fsop_resblks {
#define XFS_MAX_LOG_BYTES \
((2 * 1024 * 1024 * 1024ULL) - XFS_MIN_LOG_BYTES)
+/* Used for sanity checks on superblock */
+#define XFS_MAX_DBLOCKS(s) ((xfs_drfsbno_t)(s)->sb_agcount * (s)->sb_agblocks)
+#define XFS_MIN_DBLOCKS(s) ((xfs_drfsbno_t)((s)->sb_agcount - 1) * \
+ (s)->sb_agblocks + XFS_MIN_AG_BLOCKS)
+
/*
* Structures for XFS_IOC_FSGROWFSDATA, XFS_IOC_FSGROWFSLOG & XFS_IOC_FSGROWFSRT
*/
@@ -334,6 +340,35 @@ typedef struct xfs_error_injection {
/*
+ * Speculative preallocation trimming.
+ */
+#define XFS_EOFBLOCKS_VERSION 1
+struct xfs_eofblocks {
+ __u32 eof_version;
+ __u32 eof_flags;
+ uid_t eof_uid;
+ gid_t eof_gid;
+ prid_t eof_prid;
+ __u32 pad32;
+ __u64 eof_min_file_size;
+ __u64 pad64[12];
+};
+
+/* eof_flags values */
+#define XFS_EOF_FLAGS_SYNC (1 << 0) /* sync/wait mode scan */
+#define XFS_EOF_FLAGS_UID (1 << 1) /* filter by uid */
+#define XFS_EOF_FLAGS_GID (1 << 2) /* filter by gid */
+#define XFS_EOF_FLAGS_PRID (1 << 3) /* filter by project id */
+#define XFS_EOF_FLAGS_MINFILESIZE (1 << 4) /* filter by min file size */
+#define XFS_EOF_FLAGS_VALID \
+ (XFS_EOF_FLAGS_SYNC | \
+ XFS_EOF_FLAGS_UID | \
+ XFS_EOF_FLAGS_GID | \
+ XFS_EOF_FLAGS_PRID | \
+ XFS_EOF_FLAGS_MINFILESIZE)
+
+
+/*
* The user-level Handle Request interface structure.
*/
typedef struct xfs_fsop_handlereq {
@@ -451,6 +486,7 @@ typedef struct xfs_handle {
/* XFS_IOC_GETBIOSIZE ---- deprecated 47 */
#define XFS_IOC_GETBMAPX _IOWR('X', 56, struct getbmap)
#define XFS_IOC_ZERO_RANGE _IOW ('X', 57, struct xfs_flock64)
+#define XFS_IOC_FREE_EOFBLOCKS _IOR ('X', 58, struct xfs_eofblocks)
/*
* ioctl commands that replace IRIX syssgi()'s
diff --git a/include/xfs_ialloc.h b/include/xfs_ialloc.h
index bb53854..c8da3df 100644
--- a/include/xfs_ialloc.h
+++ b/include/xfs_ialloc.h
@@ -47,15 +47,6 @@ xfs_make_iptr(struct xfs_mount *mp, struct xfs_buf *b, int o)
}
/*
- * Find a free (set) bit in the inode bitmask.
- */
-static inline int xfs_ialloc_find_free(xfs_inofree_t *fp)
-{
- return xfs_lowbit64(*fp);
-}
-
-
-/*
* Allocate an inode on disk.
* Mode is used to tell whether the new inode will need space, and whether
* it is a directory.
@@ -81,11 +72,9 @@ int /* error */
xfs_dialloc(
struct xfs_trans *tp, /* transaction pointer */
xfs_ino_t parent, /* parent inode (directory) */
- mode_t mode, /* mode bits for new inode */
+ umode_t mode, /* mode bits for new inode */
int okalloc, /* ok to allocate more space */
struct xfs_buf **agbp, /* buf for a.g. inode header */
- boolean_t *alloc_done, /* an allocation was done to replenish
- the free inodes */
xfs_ino_t *inop); /* inode number allocated */
/*
@@ -158,7 +147,9 @@ int xfs_inobt_lookup(struct xfs_btree_cur *cur, xfs_agino_t ino,
/*
* Get the data from the pointed-to record.
*/
-extern int xfs_inobt_get_rec(struct xfs_btree_cur *cur,
+int xfs_inobt_get_rec(struct xfs_btree_cur *cur,
xfs_inobt_rec_incore_t *rec, int *stat);
+extern const struct xfs_buf_ops xfs_agi_buf_ops;
+
#endif /* __XFS_IALLOC_H__ */
diff --git a/include/xfs_ialloc_btree.h b/include/xfs_ialloc_btree.h
index f782ad0..25c0239 100644
--- a/include/xfs_ialloc_btree.h
+++ b/include/xfs_ialloc_btree.h
@@ -109,4 +109,6 @@ extern struct xfs_btree_cur *xfs_inobt_init_cursor(struct xfs_mount *,
struct xfs_trans *, struct xfs_buf *, xfs_agnumber_t);
extern int xfs_inobt_maxrecs(struct xfs_mount *, int, int);
+extern const struct xfs_buf_ops xfs_inobt_buf_ops;
+
#endif /* __XFS_IALLOC_BTREE_H__ */
diff --git a/include/xfs_inode.h b/include/xfs_inode.h
index ca56544..437b3af 100644
--- a/include/xfs_inode.h
+++ b/include/xfs_inode.h
@@ -66,8 +66,6 @@ typedef struct xfs_ifork {
struct xfs_btree_block *if_broot; /* file's incore btree root */
short if_broot_bytes; /* bytes allocated for root */
unsigned char if_flags; /* per-fork flags */
- unsigned char if_ext_max; /* max # of extent records */
- xfs_extnum_t if_lastex; /* last if_extents used */
union {
xfs_bmbt_rec_host_t *if_extents;/* linear map file exts */
xfs_ext_irec_t *if_ext_irec; /* irec map file exts */
@@ -111,7 +109,7 @@ struct xfs_imap {
* Generally, we do not want to hold the i_rlock while holding the
* i_ilock. Hierarchy is i_iolock followed by i_rlock.
*
- * xfs_iptr_t contains all the inode fields upto and including the
+ * xfs_iptr_t contains all the inode fields up to and including the
* i_mnext and i_mprev fields, it is used as a marker in the inode
* chain off the mount structure by xfs_sync calls.
*/
@@ -207,29 +205,12 @@ typedef struct xfs_icdinode {
((w) == XFS_DATA_FORK ? \
((ip)->i_d.di_nextents = (n)) : \
((ip)->i_d.di_anextents = (n)))
+#define XFS_IFORK_MAXEXT(ip, w) \
+ (XFS_IFORK_SIZE(ip, w) / sizeof(xfs_bmbt_rec_t))
-/*
- * Project quota id helpers (previously projid was 16bit only
- * and using two 16bit values to hold new 32bit projid was choosen
- * to retain compatibility with "old" filesystems).
- */
-static inline __uint32_t
-xfs_get_projid(struct xfs_icdinode i_d)
-{
- return (__uint32_t)i_d.di_projid_hi << 16 | i_d.di_projid_lo;
-}
-
-static inline void
-xfs_set_projid(struct xfs_icdinode *i_d,
- __uint32_t projid)
-{
- i_d->di_projid_hi = (__uint16_t) (projid >> 16);
- i_d->di_projid_lo = (__uint16_t) (projid & 0xffff);
-}
#ifdef __KERNEL__
-struct bhv_desc;
struct xfs_buf;
struct xfs_bmap_free;
struct xfs_bmbt_irec;
@@ -238,12 +219,6 @@ struct xfs_mount;
struct xfs_trans;
struct xfs_dquot;
-typedef struct dm_attrs_s {
- __uint32_t da_dmevmask; /* DMIG event mask */
- __uint16_t da_dmstate; /* DMIG state info */
- __uint16_t da_pad; /* DMIG extra padding */
-} dm_attrs_t;
-
typedef struct xfs_inode {
/* Inode linking and identification information. */
struct xfs_mount *i_mount; /* fs mount struct ptr */
@@ -259,32 +234,21 @@ typedef struct xfs_inode {
xfs_ifork_t i_df; /* data fork */
/* Transaction and locking information. */
- struct xfs_trans *i_transp; /* ptr to owning transaction*/
struct xfs_inode_log_item *i_itemp; /* logging information */
mrlock_t i_lock; /* inode lock */
mrlock_t i_iolock; /* inode IO lock */
- struct completion i_flush; /* inode flush completion q */
atomic_t i_pincount; /* inode pin count */
- wait_queue_head_t i_ipin_wait; /* inode pinning wait queue */
spinlock_t i_flags_lock; /* inode i_flags lock */
/* Miscellaneous state. */
- unsigned short i_flags; /* see defined flags below */
- unsigned char i_update_core; /* timestamps/size is dirty */
+ unsigned long i_flags; /* see defined flags below */
unsigned int i_delayed_blks; /* count of delay alloc blks */
xfs_icdinode_t i_d; /* most of ondisk inode */
- xfs_fsize_t i_size; /* in-memory size */
- xfs_fsize_t i_new_size; /* size when write completes */
- atomic_t i_iocount; /* outstanding I/O count */
-
/* VFS inode */
struct inode i_vnode; /* embedded VFS inode */
} xfs_inode_t;
-#define XFS_ISIZE(ip) (((ip)->i_d.di_mode & S_IFMT) == S_IFREG) ? \
- (ip)->i_size : (ip)->i_d.di_size;
-
/* Convert from vfs inode to xfs inode */
static inline struct xfs_inode *XFS_I(struct inode *inode)
{
@@ -298,6 +262,32 @@ static inline struct inode *VFS_I(struct xfs_inode *ip)
}
/*
+ * For regular files we only update the on-disk filesize when actually
+ * writing data back to disk. Until then only the copy in the VFS inode
+ * is uptodate.
+ */
+static inline xfs_fsize_t XFS_ISIZE(struct xfs_inode *ip)
+{
+ if (S_ISREG(ip->i_d.di_mode))
+ return i_size_read(VFS_I(ip));
+ return ip->i_d.di_size;
+}
+
+/*
+ * If this I/O goes past the on-disk inode size update it unless it would
+ * be past the current in-core inode size.
+ */
+static inline xfs_fsize_t
+xfs_new_eof(struct xfs_inode *ip, xfs_fsize_t new_size)
+{
+ xfs_fsize_t i_size = i_size_read(VFS_I(ip));
+
+ if (new_size > i_size)
+ new_size = i_size;
+ return new_size > ip->i_d.di_size ? new_size : 0;
+}
+
+/*
* i_flags helper functions
*/
static inline void
@@ -351,9 +341,22 @@ xfs_iflags_test_and_clear(xfs_inode_t *ip, unsigned short flags)
return ret;
}
+static inline int
+xfs_iflags_test_and_set(xfs_inode_t *ip, unsigned short flags)
+{
+ int ret;
+
+ spin_lock(&ip->i_flags_lock);
+ ret = ip->i_flags & flags;
+ if (!ret)
+ ip->i_flags |= flags;
+ spin_unlock(&ip->i_flags_lock);
+ return ret;
+}
+
/*
* Project quota id helpers (previously projid was 16bit only
- * and using two 16bit values to hold new 32bit projid was choosen
+ * and using two 16bit values to hold new 32bit projid was chosen
* to retain compatibility with "old" filesystems).
*/
static inline prid_t
@@ -371,35 +374,58 @@ xfs_set_projid(struct xfs_inode *ip,
}
/*
- * Manage the i_flush queue embedded in the inode. This completion
- * queue synchronizes processes attempting to flush the in-core
- * inode back to disk.
+ * In-core inode flags.
+ */
+#define XFS_IRECLAIM (1 << 0) /* started reclaiming this inode */
+#define XFS_ISTALE (1 << 1) /* inode has been staled */
+#define XFS_IRECLAIMABLE (1 << 2) /* inode can be reclaimed */
+#define XFS_INEW (1 << 3) /* inode has just been allocated */
+#define XFS_IFILESTREAM (1 << 4) /* inode is in a filestream dir. */
+#define XFS_ITRUNCATED (1 << 5) /* truncated down so flush-on-close */
+#define XFS_IDIRTY_RELEASE (1 << 6) /* dirty release already seen */
+#define __XFS_IFLOCK_BIT 7 /* inode is being flushed right now */
+#define XFS_IFLOCK (1 << __XFS_IFLOCK_BIT)
+#define __XFS_IPINNED_BIT 8 /* wakeup key for zero pin count */
+#define XFS_IPINNED (1 << __XFS_IPINNED_BIT)
+#define XFS_IDONTCACHE (1 << 9) /* don't cache the inode long term */
+
+/*
+ * Per-lifetime flags need to be reset when re-using a reclaimable inode during
+ * inode lookup. This prevents unintended behaviour on the new inode from
+ * ocurring.
*/
-static inline void xfs_iflock(xfs_inode_t *ip)
+#define XFS_IRECLAIM_RESET_FLAGS \
+ (XFS_IRECLAIMABLE | XFS_IRECLAIM | \
+ XFS_IDIRTY_RELEASE | XFS_ITRUNCATED | \
+ XFS_IFILESTREAM);
+
+/*
+ * Synchronize processes attempting to flush the in-core inode back to disk.
+ */
+
+extern void __xfs_iflock(struct xfs_inode *ip);
+
+static inline int xfs_iflock_nowait(struct xfs_inode *ip)
{
- wait_for_completion(&ip->i_flush);
+ return !xfs_iflags_test_and_set(ip, XFS_IFLOCK);
}
-static inline int xfs_iflock_nowait(xfs_inode_t *ip)
+static inline void xfs_iflock(struct xfs_inode *ip)
{
- return try_wait_for_completion(&ip->i_flush);
+ if (!xfs_iflock_nowait(ip))
+ __xfs_iflock(ip);
}
-static inline void xfs_ifunlock(xfs_inode_t *ip)
+static inline void xfs_ifunlock(struct xfs_inode *ip)
{
- complete(&ip->i_flush);
+ xfs_iflags_clear(ip, XFS_IFLOCK);
+ wake_up_bit(&ip->i_flags, __XFS_IFLOCK_BIT);
}
-/*
- * In-core inode flags.
- */
-#define XFS_IRECLAIM 0x0001 /* started reclaiming this inode */
-#define XFS_ISTALE 0x0002 /* inode has been staled */
-#define XFS_IRECLAIMABLE 0x0004 /* inode can be reclaimed */
-#define XFS_INEW 0x0008 /* inode has just been allocated */
-#define XFS_IFILESTREAM 0x0010 /* inode is in a filestream directory */
-#define XFS_ITRUNCATED 0x0020 /* truncated down so flush-on-close */
-#define XFS_IDIRTY_RELEASE 0x0040 /* dirty release already seen */
+static inline int xfs_isiflocked(struct xfs_inode *ip)
+{
+ return xfs_iflags_test(ip, XFS_IFLOCK);
+}
/*
* Flags for inode locking.
@@ -410,7 +436,6 @@ static inline void xfs_ifunlock(xfs_inode_t *ip)
#define XFS_IOLOCK_SHARED (1<<1)
#define XFS_ILOCK_EXCL (1<<2)
#define XFS_ILOCK_SHARED (1<<3)
-#define XFS_IUNLOCK_NONOTIFY (1<<4)
#define XFS_LOCK_MASK (XFS_IOLOCK_EXCL | XFS_IOLOCK_SHARED \
| XFS_ILOCK_EXCL | XFS_ILOCK_SHARED)
@@ -419,35 +444,41 @@ static inline void xfs_ifunlock(xfs_inode_t *ip)
{ XFS_IOLOCK_EXCL, "IOLOCK_EXCL" }, \
{ XFS_IOLOCK_SHARED, "IOLOCK_SHARED" }, \
{ XFS_ILOCK_EXCL, "ILOCK_EXCL" }, \
- { XFS_ILOCK_SHARED, "ILOCK_SHARED" }, \
- { XFS_IUNLOCK_NONOTIFY, "IUNLOCK_NONOTIFY" }
+ { XFS_ILOCK_SHARED, "ILOCK_SHARED" }
/*
* Flags for lockdep annotations.
*
- * XFS_I[O]LOCK_PARENT - for operations that require locking two inodes
- * (ie directory operations that require locking a directory inode and
- * an entry inode). The first inode gets locked with this flag so it
- * gets a lockdep subclass of 1 and the second lock will have a lockdep
- * subclass of 0.
+ * XFS_LOCK_PARENT - for directory operations that require locking a
+ * parent directory inode and a child entry inode. The parent gets locked
+ * with this flag so it gets a lockdep subclass of 1 and the child entry
+ * lock will have a lockdep subclass of 0.
+ *
+ * XFS_LOCK_RTBITMAP/XFS_LOCK_RTSUM - the realtime device bitmap and summary
+ * inodes do not participate in the normal lock order, and thus have their
+ * own subclasses.
*
* XFS_LOCK_INUMORDER - for locking several inodes at the some time
* with xfs_lock_inodes(). This flag is used as the starting subclass
* and each subsequent lock acquired will increment the subclass by one.
- * So the first lock acquired will have a lockdep subclass of 2, the
- * second lock will have a lockdep subclass of 3, and so on. It is
+ * So the first lock acquired will have a lockdep subclass of 4, the
+ * second lock will have a lockdep subclass of 5, and so on. It is
* the responsibility of the class builder to shift this to the correct
* portion of the lock_mode lockdep mask.
*/
#define XFS_LOCK_PARENT 1
-#define XFS_LOCK_INUMORDER 2
+#define XFS_LOCK_RTBITMAP 2
+#define XFS_LOCK_RTSUM 3
+#define XFS_LOCK_INUMORDER 4
#define XFS_IOLOCK_SHIFT 16
#define XFS_IOLOCK_PARENT (XFS_LOCK_PARENT << XFS_IOLOCK_SHIFT)
#define XFS_ILOCK_SHIFT 24
#define XFS_ILOCK_PARENT (XFS_LOCK_PARENT << XFS_ILOCK_SHIFT)
+#define XFS_ILOCK_RTBITMAP (XFS_LOCK_RTBITMAP << XFS_ILOCK_SHIFT)
+#define XFS_ILOCK_RTSUM (XFS_LOCK_RTSUM << XFS_ILOCK_SHIFT)
#define XFS_IOLOCK_DEP_MASK 0x00ff0000
#define XFS_ILOCK_DEP_MASK 0xff000000
@@ -456,18 +487,6 @@ static inline void xfs_ifunlock(xfs_inode_t *ip)
#define XFS_IOLOCK_DEP(flags) (((flags) & XFS_IOLOCK_DEP_MASK) >> XFS_IOLOCK_SHIFT)
#define XFS_ILOCK_DEP(flags) (((flags) & XFS_ILOCK_DEP_MASK) >> XFS_ILOCK_SHIFT)
-extern struct lock_class_key xfs_iolock_reclaimable;
-
-/*
- * Flags for xfs_itruncate_start().
- */
-#define XFS_ITRUNC_DEFINITE 0x1
-#define XFS_ITRUNC_MAYBE 0x2
-
-#define XFS_ITRUNC_FLAGS \
- { XFS_ITRUNC_DEFINITE, "DEFINITE" }, \
- { XFS_ITRUNC_MAYBE, "MAYBE" }
-
/*
* For multiple groups support: if S_ISGID bit is set in the parent
* directory, group of new file is set to that of the parent, and
@@ -477,11 +496,10 @@ extern struct lock_class_key xfs_iolock_reclaimable;
(((pip)->i_mount->m_flags & XFS_MOUNT_GRPID) || \
((pip)->i_d.di_mode & S_ISGID))
+
/*
- * xfs_iget.c prototypes.
+ * xfs_inode.c prototypes.
*/
-int xfs_iget(struct xfs_mount *, struct xfs_trans *, xfs_ino_t,
- uint, uint, xfs_inode_t **);
void xfs_ilock(xfs_inode_t *, uint);
int xfs_ilock_nowait(xfs_inode_t *, uint);
void xfs_iunlock(xfs_inode_t *, uint);
@@ -489,33 +507,25 @@ void xfs_ilock_demote(xfs_inode_t *, uint);
int xfs_isilocked(xfs_inode_t *, uint);
uint xfs_ilock_map_shared(xfs_inode_t *);
void xfs_iunlock_map_shared(xfs_inode_t *, uint);
-void xfs_inode_free(struct xfs_inode *ip);
-
-/*
- * xfs_inode.c prototypes.
- */
-int xfs_ialloc(struct xfs_trans *, xfs_inode_t *, mode_t,
+int xfs_ialloc(struct xfs_trans *, xfs_inode_t *, umode_t,
xfs_nlink_t, xfs_dev_t, prid_t, int,
- struct xfs_buf **, boolean_t *, xfs_inode_t **);
+ struct xfs_buf **, xfs_inode_t **);
uint xfs_ip2xflags(struct xfs_inode *);
uint xfs_dic2xflags(struct xfs_dinode *);
int xfs_ifree(struct xfs_trans *, xfs_inode_t *,
struct xfs_bmap_free *);
-int xfs_itruncate_start(xfs_inode_t *, uint, xfs_fsize_t);
-int xfs_itruncate_finish(struct xfs_trans **, xfs_inode_t *,
- xfs_fsize_t, int, int);
+int xfs_itruncate_extents(struct xfs_trans **, struct xfs_inode *,
+ int, xfs_fsize_t);
int xfs_iunlink(struct xfs_trans *, xfs_inode_t *);
void xfs_iext_realloc(xfs_inode_t *, int, int);
void xfs_iunpin_wait(xfs_inode_t *);
-int xfs_iflush(xfs_inode_t *, uint);
+int xfs_iflush(struct xfs_inode *, struct xfs_buf **);
void xfs_lock_inodes(xfs_inode_t **, int, uint);
void xfs_lock_two_inodes(xfs_inode_t *, xfs_inode_t *, uint);
-void xfs_synchronize_times(xfs_inode_t *);
-void xfs_mark_inode_dirty(xfs_inode_t *);
-void xfs_mark_inode_dirty_sync(xfs_inode_t *);
+xfs_extlen_t xfs_get_extsz_hint(struct xfs_inode *ip);
#define IHOLD(ip) \
do { \
@@ -537,19 +547,15 @@ do { \
*/
#define XFS_IGET_CREATE 0x1
#define XFS_IGET_UNTRUSTED 0x2
+#define XFS_IGET_DONTCACHE 0x4
-int xfs_inotobp(struct xfs_mount *, struct xfs_trans *,
- xfs_ino_t, struct xfs_dinode **,
- struct xfs_buf **, int *, uint);
-int xfs_itobp(struct xfs_mount *, struct xfs_trans *,
- struct xfs_inode *, struct xfs_dinode **,
- struct xfs_buf **, uint);
+int xfs_imap_to_bp(struct xfs_mount *, struct xfs_trans *,
+ struct xfs_imap *, struct xfs_dinode **,
+ struct xfs_buf **, uint, uint);
int xfs_iread(struct xfs_mount *, struct xfs_trans *,
struct xfs_inode *, uint);
void xfs_dinode_to_disk(struct xfs_dinode *,
struct xfs_icdinode *);
-void xfs_dinode_from_disk(struct xfs_icdinode *,
- struct xfs_dinode *);
void xfs_idestroy_fork(struct xfs_inode *, int);
void xfs_idata_realloc(struct xfs_inode *, int, int);
void xfs_iroot_realloc(struct xfs_inode *, int, int);
@@ -579,16 +585,10 @@ void xfs_iext_irec_compact(xfs_ifork_t *);
void xfs_iext_irec_compact_pages(xfs_ifork_t *);
void xfs_iext_irec_compact_full(xfs_ifork_t *);
void xfs_iext_irec_update_extoffs(xfs_ifork_t *, int, int);
+bool xfs_can_free_eofblocks(struct xfs_inode *, bool);
#define xfs_ipincount(ip) ((unsigned int) atomic_read(&ip->i_pincount))
-#ifdef DEBUG
-void xfs_isize_check(struct xfs_mount *, struct xfs_inode *,
- xfs_fsize_t);
-#else /* DEBUG */
-#define xfs_isize_check(mp, ip, isize)
-#endif /* DEBUG */
-
#if defined(DEBUG)
void xfs_inobp_check(struct xfs_mount *, struct xfs_buf *);
#else
@@ -598,5 +598,6 @@ void xfs_inobp_check(struct xfs_mount *, struct xfs_buf *);
extern struct kmem_zone *xfs_ifork_zone;
extern struct kmem_zone *xfs_inode_zone;
extern struct kmem_zone *xfs_ili_zone;
+extern const struct xfs_buf_ops xfs_inode_buf_ops;
#endif /* __XFS_INODE_H__ */
diff --git a/include/xfs_inode_item.h b/include/xfs_inode_item.h
index d3dee61..779812f 100644
--- a/include/xfs_inode_item.h
+++ b/include/xfs_inode_item.h
@@ -86,6 +86,15 @@ typedef struct xfs_inode_log_format_64 {
#define XFS_ILOG_AEXT 0x080 /* log i_af.if_extents */
#define XFS_ILOG_ABROOT 0x100 /* log i_af.i_broot */
+
+/*
+ * The timestamps are dirty, but not necessarily anything else in the inode
+ * core. Unlike the other fields above this one must never make it to disk
+ * in the ilf_fields of the inode_log_format, but is purely store in-memory in
+ * ili_fields in the inode_log_item.
+ */
+#define XFS_ILOG_TIMESTAMP 0x4000
+
#define XFS_ILOG_NONCORE (XFS_ILOG_DDATA | XFS_ILOG_DEXT | \
XFS_ILOG_DBROOT | XFS_ILOG_DEV | \
XFS_ILOG_UUID | XFS_ILOG_ADATA | \
@@ -101,7 +110,7 @@ typedef struct xfs_inode_log_format_64 {
XFS_ILOG_DEXT | XFS_ILOG_DBROOT | \
XFS_ILOG_DEV | XFS_ILOG_UUID | \
XFS_ILOG_ADATA | XFS_ILOG_AEXT | \
- XFS_ILOG_ABROOT)
+ XFS_ILOG_ABROOT | XFS_ILOG_TIMESTAMP)
static inline int xfs_ilog_fbroot(int w)
{
@@ -134,30 +143,25 @@ typedef struct xfs_inode_log_item {
unsigned short ili_lock_flags; /* lock flags */
unsigned short ili_logged; /* flushed logged data */
unsigned int ili_last_fields; /* fields when flushed */
+ unsigned int ili_fields; /* fields to be logged */
struct xfs_bmbt_rec *ili_extents_buf; /* array of logged
data exts */
struct xfs_bmbt_rec *ili_aextents_buf; /* array of logged
attr exts */
-#ifdef XFS_TRANS_DEBUG
- int ili_root_size;
- char *ili_orig_root;
-#endif
xfs_inode_log_format_t ili_format; /* logged structure */
} xfs_inode_log_item_t;
static inline int xfs_inode_clean(xfs_inode_t *ip)
{
- return (!ip->i_itemp ||
- !(ip->i_itemp->ili_format.ilf_fields & XFS_ILOG_ALL)) &&
- !ip->i_update_core;
+ return !ip->i_itemp || !(ip->i_itemp->ili_fields & XFS_ILOG_ALL);
}
extern void xfs_inode_item_init(struct xfs_inode *, struct xfs_mount *);
extern void xfs_inode_item_destroy(struct xfs_inode *);
extern void xfs_iflush_done(struct xfs_buf *, struct xfs_log_item *);
extern void xfs_istale_done(struct xfs_buf *, struct xfs_log_item *);
-extern void xfs_iflush_abort(struct xfs_inode *);
+extern void xfs_iflush_abort(struct xfs_inode *, bool);
extern int xfs_inode_item_format_convert(xfs_log_iovec_t *,
xfs_inode_log_format_t *);
diff --git a/include/xfs_inum.h b/include/xfs_inum.h
index b8e4ee4..90efdaf 100644
--- a/include/xfs_inum.h
+++ b/include/xfs_inum.h
@@ -26,22 +26,6 @@
* high agno_log-agblklog-inopblog bits - 0
*/
-typedef __uint32_t xfs_agino_t; /* within allocation grp inode number */
-
-/*
- * Useful inode bits for this kernel.
- * Used in some places where having 64-bits in the 32-bit kernels
- * costs too much.
- */
-#if XFS_BIG_INUMS
-typedef xfs_ino_t xfs_intino_t;
-#else
-typedef __uint32_t xfs_intino_t;
-#endif
-
-#define NULLFSINO ((xfs_ino_t)-1)
-#define NULLAGINO ((xfs_agino_t)-1)
-
struct xfs_mount;
#define XFS_INO_MASK(k) (__uint32_t)((1ULL << (k)) - 1)
diff --git a/include/xfs_log.h b/include/xfs_log.h
index 916eb7d..5caee96 100644
--- a/include/xfs_log.h
+++ b/include/xfs_log.h
@@ -53,15 +53,6 @@ static inline xfs_lsn_t _lsn_cmp(xfs_lsn_t lsn1, xfs_lsn_t lsn2)
#define XFS_LOG_REL_PERM_RESERV 0x1
/*
- * Flags to xfs_log_reserve()
- *
- * XFS_LOG_PERM_RESERV: Permanent reservation. When writes are
- * performed against this type of reservation, the reservation
- * is not decreased. Long running transactions should use this.
- */
-#define XFS_LOG_PERM_RESERV 0x2
-
-/*
* Flags to xfs_log_force()
*
* XFS_LOG_SYNC: Synchronous force in-core log to disk
@@ -137,7 +128,7 @@ struct xfs_trans;
void xfs_log_item_init(struct xfs_mount *mp,
struct xfs_log_item *item,
int type,
- struct xfs_item_ops *ops);
+ const struct xfs_item_ops *ops);
xfs_lsn_t xfs_log_done(struct xfs_mount *mp,
struct xlog_ticket *ticket,
@@ -160,8 +151,9 @@ int xfs_log_mount(struct xfs_mount *mp,
xfs_daddr_t start_block,
int num_bblocks);
int xfs_log_mount_finish(struct xfs_mount *mp);
-void xfs_log_move_tail(struct xfs_mount *mp,
- xfs_lsn_t tail_lsn);
+xfs_lsn_t xlog_assign_tail_lsn(struct xfs_mount *mp);
+xfs_lsn_t xlog_assign_tail_lsn_locked(struct xfs_mount *mp);
+void xfs_log_space_wake(struct xfs_mount *mp);
int xfs_log_notify(struct xfs_mount *mp,
struct xlog_in_core *iclog,
xfs_log_callback_t *callback_entry);
@@ -172,13 +164,9 @@ int xfs_log_reserve(struct xfs_mount *mp,
int count,
struct xlog_ticket **ticket,
__uint8_t clientid,
- uint flags,
+ bool permanent,
uint t_type);
-int xfs_log_write(struct xfs_mount *mp,
- xfs_log_iovec_t region[],
- int nentries,
- struct xlog_ticket *ticket,
- xfs_lsn_t *start_lsn);
+int xfs_log_regrant(struct xfs_mount *mp, struct xlog_ticket *tic);
int xfs_log_unmount_write(struct xfs_mount *mp);
void xfs_log_unmount(struct xfs_mount *mp);
int xfs_log_force_umount(struct xfs_mount *mp, int logerror);
@@ -189,12 +177,13 @@ void xlog_iodone(struct xfs_buf *);
struct xlog_ticket *xfs_log_ticket_get(struct xlog_ticket *ticket);
void xfs_log_ticket_put(struct xlog_ticket *ticket);
-xlog_tid_t xfs_log_get_trans_ident(struct xfs_trans *tp);
-
int xfs_log_commit_cil(struct xfs_mount *mp, struct xfs_trans *tp,
- struct xfs_log_vec *log_vector,
xfs_lsn_t *commit_lsn, int flags);
bool xfs_log_item_in_current_chkpt(struct xfs_log_item *lip);
+void xfs_log_work_queue(struct xfs_mount *mp);
+void xfs_log_worker(struct work_struct *work);
+void xfs_log_quiesce(struct xfs_mount *mp);
+
#endif
#endif /* __XFS_LOG_H__ */
diff --git a/include/xfs_log_priv.h b/include/xfs_log_priv.h
index d5f8be8..16d8d12 100644
--- a/include/xfs_log_priv.h
+++ b/include/xfs_log_priv.h
@@ -19,7 +19,7 @@
#define __XFS_LOG_PRIV_H__
struct xfs_buf;
-struct log;
+struct xlog;
struct xlog_ticket;
struct xfs_mount;
@@ -87,10 +87,6 @@ static inline uint xlog_get_client_id(__be32 i)
return be32_to_cpu(i) >> 24;
}
-#define xlog_panic(args...) cmn_err(CE_PANIC, ## args)
-#define xlog_exit(args...) cmn_err(CE_PANIC, ## args)
-#define xlog_warn(args...) cmn_err(CE_WARN, ## args)
-
/*
* In core log state
*/
@@ -143,11 +139,13 @@ static inline uint xlog_get_client_id(__be32 i)
/*
* Flags for log structure
*/
-#define XLOG_CHKSUM_MISMATCH 0x1 /* used only during recovery */
#define XLOG_ACTIVE_RECOVERY 0x2 /* in the middle of recovery */
#define XLOG_RECOVERY_NEEDED 0x4 /* log was recovered */
#define XLOG_IO_ERROR 0x8 /* log hit an I/O error, and being
shutdown */
+#define XLOG_TAIL_WARN 0x10 /* log tail verify warning issued */
+
+typedef __uint32_t xlog_tid_t;
#ifdef __KERNEL__
/*
@@ -240,8 +238,8 @@ typedef struct xlog_res {
} xlog_res_t;
typedef struct xlog_ticket {
- wait_queue_head_t t_wait; /* ticket wait queue */
struct list_head t_queue; /* reserve/write queue */
+ struct task_struct *t_task; /* task that owns this ticket */
xlog_tid_t t_tid; /* transaction identifier : 4 */
atomic_t t_ref; /* ticket reference count : 4 */
int t_curr_res; /* current reservation in bytes : 4 */
@@ -292,7 +290,7 @@ typedef struct xlog_rec_header {
__be32 h_len; /* len in bytes; should be 64-bit aligned: 4 */
__be64 h_lsn; /* lsn of this LR : 8 */
__be64 h_tail_lsn; /* lsn of 1st LR w/ buffers not committed: 8 */
- __be32 h_chksum; /* may not be used; non-zero if used : 4 */
+ __le32 h_crc; /* crc of log record : 4 */
__be32 h_prev_block; /* block number to previous LR : 4 */
__be32 h_num_logops; /* number of log operations in this LR : 4 */
__be32 h_cycle_data[XLOG_HEADER_CYCLE_SIZE / BBSIZE];
@@ -353,7 +351,7 @@ typedef struct xlog_in_core {
struct xlog_in_core *ic_next;
struct xlog_in_core *ic_prev;
struct xfs_buf *ic_bp;
- struct log *ic_log;
+ struct xlog *ic_log;
int ic_size;
int ic_offset;
int ic_bwritecnt;
@@ -410,7 +408,7 @@ struct xfs_cil_ctx {
* operations almost as efficient as the old logging methods.
*/
struct xfs_cil {
- struct log *xc_log;
+ struct xlog *xc_log;
struct list_head xc_cil;
spinlock_t xc_cil_lock;
struct xfs_cil_ctx *xc_ctx;
@@ -418,6 +416,8 @@ struct xfs_cil {
struct list_head xc_committing;
wait_queue_head_t xc_commit_wait;
xfs_lsn_t xc_current_sequence;
+ struct work_struct xc_push_work;
+ xfs_lsn_t xc_push_seq;
};
/*
@@ -471,12 +471,22 @@ struct xfs_cil {
#define XLOG_CIL_HARD_SPACE_LIMIT(log) (3 * (log->l_logsize >> 4))
/*
+ * ticket grant locks, queues and accounting have their own cachlines
+ * as these are quite hot and can be operated on concurrently.
+ */
+struct xlog_grant_head {
+ spinlock_t lock ____cacheline_aligned_in_smp;
+ struct list_head waiters;
+ atomic64_t grant;
+};
+
+/*
* The reservation head lsn is not made up of a cycle number and block number.
* Instead, it uses a cycle number and byte number. Logs don't expect to
* overflow 31 bits worth of byte offset, so using a byte number will mean
* that round off problems won't occur when releasing partial reservations.
*/
-typedef struct log {
+struct xlog {
/* The following fields don't need locking */
struct xfs_mount *l_mp; /* mount point */
struct xfs_ail *l_ailp; /* AIL log is working with */
@@ -484,6 +494,7 @@ typedef struct log {
struct xfs_buf *l_xbuf; /* extra buffer for log
* wrapping */
struct xfs_buftarg *l_targ; /* buftarg of log */
+ struct delayed_work l_work; /* background flush work */
uint l_flags;
uint l_quotaoffs_flag; /* XFS_DQ_*, for QUOTAOFFs */
struct list_head *l_buf_cancel_table;
@@ -521,24 +532,15 @@ typedef struct log {
/* lsn of 1st LR with unflushed * buffers */
atomic64_t l_tail_lsn ____cacheline_aligned_in_smp;
- /*
- * ticket grant locks, queues and accounting have their own cachlines
- * as these are quite hot and can be operated on concurrently.
- */
- spinlock_t l_grant_reserve_lock ____cacheline_aligned_in_smp;
- struct list_head l_reserveq;
- atomic64_t l_grant_reserve_head;
-
- spinlock_t l_grant_write_lock ____cacheline_aligned_in_smp;
- struct list_head l_writeq;
- atomic64_t l_grant_write_head;
+ struct xlog_grant_head l_reserve_head;
+ struct xlog_grant_head l_write_head;
/* The following field are used for debugging; need to hold icloglock */
#ifdef DEBUG
char *l_iclog_bak[XLOG_MAX_ICLOGS];
#endif
-} xlog_t;
+};
#define XLOG_BUF_CANCEL_BUCKET(log, blkno) \
((log)->l_buf_cancel_table + ((__uint64_t)blkno % XLOG_BC_TABLE_SIZE))
@@ -546,15 +548,25 @@ typedef struct log {
#define XLOG_FORCED_SHUTDOWN(log) ((log)->l_flags & XLOG_IO_ERROR)
/* common routines */
-extern xfs_lsn_t xlog_assign_tail_lsn(struct xfs_mount *mp);
-extern int xlog_recover(xlog_t *log);
-extern int xlog_recover_finish(xlog_t *log);
-extern void xlog_pack_data(xlog_t *log, xlog_in_core_t *iclog, int);
+extern int
+xlog_recover(
+ struct xlog *log);
+extern int
+xlog_recover_finish(
+ struct xlog *log);
+
+extern __le32 xlog_cksum(struct xlog *log, struct xlog_rec_header *rhead,
+ char *dp, int size);
extern kmem_zone_t *xfs_log_ticket_zone;
-struct xlog_ticket *xlog_ticket_alloc(struct log *log, int unit_bytes,
- int count, char client, uint xflags,
- int alloc_flags);
+struct xlog_ticket *
+xlog_ticket_alloc(
+ struct xlog *log,
+ int unit_bytes,
+ int count,
+ char client,
+ bool permanent,
+ xfs_km_flags_t alloc_flags);
static inline void
@@ -566,15 +578,20 @@ xlog_write_adv_cnt(void **ptr, int *len, int *off, size_t bytes)
}
void xlog_print_tic_res(struct xfs_mount *mp, struct xlog_ticket *ticket);
-int xlog_write(struct log *log, struct xfs_log_vec *log_vector,
- struct xlog_ticket *tic, xfs_lsn_t *start_lsn,
- xlog_in_core_t **commit_iclog, uint flags);
+int
+xlog_write(
+ struct xlog *log,
+ struct xfs_log_vec *log_vector,
+ struct xlog_ticket *tic,
+ xfs_lsn_t *start_lsn,
+ struct xlog_in_core **commit_iclog,
+ uint flags);
/*
* When we crack an atomic LSN, we sample it first so that the value will not
* change while we are cracking it into the component values. This means we
* will always get consistent component values to work from. This should always
- * be used to smaple and crack LSNs taht are stored and updated in atomic
+ * be used to sample and crack LSNs that are stored and updated in atomic
* variables.
*/
static inline void
@@ -628,17 +645,23 @@ xlog_assign_grant_head(atomic64_t *head, int cycle, int space)
/*
* Committed Item List interfaces
*/
-int xlog_cil_init(struct log *log);
-void xlog_cil_init_post_recovery(struct log *log);
-void xlog_cil_destroy(struct log *log);
+int
+xlog_cil_init(struct xlog *log);
+void
+xlog_cil_init_post_recovery(struct xlog *log);
+void
+xlog_cil_destroy(struct xlog *log);
/*
* CIL force routines
*/
-xfs_lsn_t xlog_cil_force_lsn(struct log *log, xfs_lsn_t sequence);
+xfs_lsn_t
+xlog_cil_force_lsn(
+ struct xlog *log,
+ xfs_lsn_t sequence);
static inline void
-xlog_cil_force(struct log *log)
+xlog_cil_force(struct xlog *log)
{
xlog_cil_force_lsn(log, log->l_cilp->xc_current_sequence);
}
diff --git a/include/xfs_mount.h b/include/xfs_mount.h
index 9be9a05..c267379 100644
--- a/include/xfs_mount.h
+++ b/include/xfs_mount.h
@@ -52,15 +52,8 @@ typedef struct xfs_trans_reservations {
#else /* __KERNEL__ */
-#include "xfs_sync.h"
-
-struct log;
-struct xfs_mount_args;
+struct xlog;
struct xfs_inode;
-struct xfs_bmbt_irec;
-struct xfs_bmap_free;
-struct xfs_extdelta;
-struct xfs_swapext;
struct xfs_mru_cache;
struct xfs_nameops;
struct xfs_ail;
@@ -134,7 +127,7 @@ typedef struct xfs_mount {
uint m_readio_blocks; /* min read size blocks */
uint m_writeio_log; /* min write size log bytes */
uint m_writeio_blocks; /* min write size blocks */
- struct log *m_log; /* log specific stuff */
+ struct xlog *m_log; /* log specific stuff */
int m_logbufs; /* number of log buffers */
int m_logbsize; /* size of each log buffer */
uint m_rsumlevels; /* rt summary levels */
@@ -177,7 +170,6 @@ typedef struct xfs_mount {
uint m_qflags; /* quota status flags */
xfs_trans_reservations_t m_reservations;/* precomputed res values */
__uint64_t m_maxicount; /* maximum inode count */
- __uint64_t m_maxioffset; /* maximum inode offset */
__uint64_t m_resblks; /* total reserved blocks */
__uint64_t m_resblks_avail;/* available reserved blocks */
__uint64_t m_resblks_save; /* reserved blks @ remount,ro */
@@ -204,17 +196,21 @@ typedef struct xfs_mount {
struct mutex m_icsb_mutex; /* balancer sync lock */
#endif
struct xfs_mru_cache *m_filestream; /* per-mount filestream data */
- struct task_struct *m_sync_task; /* generalised sync thread */
- xfs_sync_work_t m_sync_work; /* work item for VFS_SYNC */
- struct list_head m_sync_list; /* sync thread work item list */
- spinlock_t m_sync_lock; /* work item list lock */
- int m_sync_seq; /* sync thread generation no. */
- wait_queue_head_t m_wait_single_sync_task;
+ struct delayed_work m_reclaim_work; /* background inode reclaim */
+ struct delayed_work m_eofblocks_work; /* background eof blocks
+ trimming */
__int64_t m_update_flags; /* sb flags we need to update
on the next remount,rw */
struct shrinker m_inode_shrink; /* inode reclaim shrinker */
int64_t m_low_space[XFS_LOWSP_MAX];
/* low free space thresholds */
+
+ struct workqueue_struct *m_data_workqueue;
+ struct workqueue_struct *m_unwritten_workqueue;
+ struct workqueue_struct *m_cil_workqueue;
+ struct workqueue_struct *m_reclaim_workqueue;
+ struct workqueue_struct *m_log_workqueue;
+ struct workqueue_struct *m_eofblocks_workqueue;
} xfs_mount_t;
/*
@@ -223,11 +219,11 @@ typedef struct xfs_mount {
#define XFS_MOUNT_WSYNC (1ULL << 0) /* for nfs - all metadata ops
must be synchronous except
for space allocations */
-#define XFS_MOUNT_DELAYLOG (1ULL << 1) /* delayed logging is enabled */
#define XFS_MOUNT_WAS_CLEAN (1ULL << 3)
#define XFS_MOUNT_FS_SHUTDOWN (1ULL << 4) /* atomic stop of all filesystem
operations, typically for
disk errors in metadata */
+#define XFS_MOUNT_DISCARD (1ULL << 5) /* discard unused blocks */
#define XFS_MOUNT_RETERR (1ULL << 6) /* return alignment errors to
user */
#define XFS_MOUNT_NOALIGN (1ULL << 7) /* turn off stripe alignment
@@ -297,8 +293,6 @@ xfs_preferred_iosize(xfs_mount_t *mp)
PAGE_CACHE_SIZE));
}
-#define XFS_MAXIOFFSET(mp) ((mp)->m_maxioffset)
-
#define XFS_LAST_UNMOUNT_WAS_CLEAN(mp) \
((mp)->m_flags & XFS_MOUNT_WAS_CLEAN)
#define XFS_FORCED_SHUTDOWN(mp) ((mp)->m_flags & XFS_MOUNT_FS_SHUTDOWN)
@@ -314,9 +308,6 @@ void xfs_do_force_shutdown(struct xfs_mount *mp, int flags, char *fname,
#define SHUTDOWN_REMOTE_REQ 0x0010 /* shutdown came from remote cell */
#define SHUTDOWN_DEVICE_REQ 0x0020 /* failed all paths to the device */
-#define xfs_test_for_freeze(mp) ((mp)->m_super->s_frozen)
-#define xfs_wait_for_freeze(mp,l) vfs_check_frozen((mp)->m_super, (l))
-
/*
* Flags for xfs_mountfs
*/
@@ -366,12 +357,11 @@ typedef struct xfs_mod_sb {
int64_t msb_delta; /* Change to make to specified field */
} xfs_mod_sb_t;
-extern int xfs_log_sbcount(xfs_mount_t *, uint);
+extern int xfs_log_sbcount(xfs_mount_t *);
extern __uint64_t xfs_default_resblks(xfs_mount_t *mp);
extern int xfs_mountfs(xfs_mount_t *mp);
extern void xfs_unmountfs(xfs_mount_t *);
-extern int xfs_unmountfs_writesb(xfs_mount_t *);
extern int xfs_mod_incore_sb(xfs_mount_t *, xfs_sb_field_t, int64_t, int);
extern int xfs_mod_incore_sb_batch(xfs_mount_t *, xfs_mod_sb_t *,
uint, int);
@@ -402,4 +392,6 @@ extern int xfs_initialize_perag(struct xfs_mount *, xfs_agnumber_t,
extern void xfs_sb_from_disk(struct xfs_sb *, struct xfs_dsb *);
extern void xfs_sb_to_disk(struct xfs_dsb *, struct xfs_sb *, __int64_t);
+extern const struct xfs_buf_ops xfs_sb_buf_ops;
+
#endif /* __XFS_MOUNT_H__ */
diff --git a/include/xfs_quota.h b/include/xfs_quota.h
index 5d1f57d..b50ec5b 100644
--- a/include/xfs_quota.h
+++ b/include/xfs_quota.h
@@ -87,8 +87,7 @@ typedef struct xfs_dqblk {
#define XFS_DQ_PROJ 0x0002 /* project quota */
#define XFS_DQ_GROUP 0x0004 /* a group quota */
#define XFS_DQ_DIRTY 0x0008 /* dquot is dirty */
-#define XFS_DQ_WANT 0x0010 /* for lookup/reclaim race */
-#define XFS_DQ_INACTIVE 0x0020 /* dq off mplist & hashlist */
+#define XFS_DQ_FREEING 0x0010 /* dquot is beeing torn down */
#define XFS_DQ_ALLTYPES (XFS_DQ_USER|XFS_DQ_PROJ|XFS_DQ_GROUP)
@@ -97,8 +96,7 @@ typedef struct xfs_dqblk {
{ XFS_DQ_PROJ, "PROJ" }, \
{ XFS_DQ_GROUP, "GROUP" }, \
{ XFS_DQ_DIRTY, "DIRTY" }, \
- { XFS_DQ_WANT, "WANT" }, \
- { XFS_DQ_INACTIVE, "INACTIVE" }
+ { XFS_DQ_FREEING, "FREEING" }
/*
* In the worst case, when both user and group quotas are on,
@@ -176,6 +174,8 @@ typedef struct xfs_qoff_logformat {
#define XFS_UQUOTA_ACTIVE 0x0100 /* uquotas are being turned off */
#define XFS_PQUOTA_ACTIVE 0x0200 /* pquotas are being turned off */
#define XFS_GQUOTA_ACTIVE 0x0400 /* gquotas are being turned off */
+#define XFS_ALL_QUOTA_ACTIVE \
+ (XFS_UQUOTA_ACTIVE | XFS_PQUOTA_ACTIVE | XFS_GQUOTA_ACTIVE)
/*
* Checking XFS_IS_*QUOTA_ON() while holding any inode lock guarantees
@@ -199,7 +199,6 @@ typedef struct xfs_qoff_logformat {
#define XFS_QMOPT_UQUOTA 0x0000004 /* user dquot requested */
#define XFS_QMOPT_PQUOTA 0x0000008 /* project dquot requested */
#define XFS_QMOPT_FORCE_RES 0x0000010 /* ignore quota limits */
-#define XFS_QMOPT_DQSUSER 0x0000020 /* don't cache super users dquot */
#define XFS_QMOPT_SBVERSION 0x0000040 /* change superblock version num */
#define XFS_QMOPT_DOWARN 0x0000400 /* increase warning cnt if needed */
#define XFS_QMOPT_DQREPAIR 0x0001000 /* repair dquot if damaged */
@@ -299,11 +298,6 @@ typedef struct xfs_dqtrx {
long qt_delrtb_delta; /* delayed RT blk count changes */
} xfs_dqtrx_t;
-extern int xfs_qm_dqcheck(xfs_disk_dquot_t *, xfs_dqid_t, uint, uint, char *);
-extern int xfs_mount_reset_sbqflags(struct xfs_mount *);
-
-#endif /* __KERNEL__ */
-
#ifdef CONFIG_XFS_QUOTA
extern void xfs_trans_dup_dqinfo(struct xfs_trans *, struct xfs_trans *);
extern void xfs_trans_free_dqinfo(struct xfs_trans *);
@@ -331,25 +325,36 @@ extern int xfs_qm_dqattach_locked(struct xfs_inode *, uint);
extern void xfs_qm_dqdetach(struct xfs_inode *);
extern void xfs_qm_dqrele(struct xfs_dquot *);
extern void xfs_qm_statvfs(struct xfs_inode *, struct kstatfs *);
-extern int xfs_qm_sync(struct xfs_mount *, int);
extern int xfs_qm_newmount(struct xfs_mount *, uint *, uint *);
extern void xfs_qm_mount_quotas(struct xfs_mount *);
extern void xfs_qm_unmount(struct xfs_mount *);
extern void xfs_qm_unmount_quotas(struct xfs_mount *);
#else
-#define xfs_qm_vop_dqalloc(ip, uid, gid, prid, flags, udqp, gdqp) ({ \
- *(udqp) = NULL; \
- *(gdqp) = NULL; \
- 0; \
-})
+static inline int
+xfs_qm_vop_dqalloc(struct xfs_inode *ip, uid_t uid, gid_t gid, prid_t prid,
+ uint flags, struct xfs_dquot **udqp, struct xfs_dquot **gdqp)
+{
+ *udqp = NULL;
+ *gdqp = NULL;
+ return 0;
+}
#define xfs_trans_dup_dqinfo(tp, tp2)
#define xfs_trans_free_dqinfo(tp)
#define xfs_trans_mod_dquot_byino(tp, ip, fields, delta)
#define xfs_trans_apply_dquot_deltas(tp)
#define xfs_trans_unreserve_and_mod_dquots(tp)
-#define xfs_trans_reserve_quota_nblks(tp, ip, blks, inos, flg) (0)
-#define xfs_trans_reserve_quota_bydquots(tp, mp, uqp, gqp, blks, inos, flg) (0)
+static inline int xfs_trans_reserve_quota_nblks(struct xfs_trans *tp,
+ struct xfs_inode *ip, long nblks, long ninos, uint flags)
+{
+ return 0;
+}
+static inline int xfs_trans_reserve_quota_bydquots(struct xfs_trans *tp,
+ struct xfs_mount *mp, struct xfs_dquot *udqp,
+ struct xfs_dquot *gdqp, long nblks, long nions, uint flags)
+{
+ return 0;
+}
#define xfs_qm_vop_create_dqattach(tp, ip, u, g)
#define xfs_qm_vop_rename_dqattach(it) (0)
#define xfs_qm_vop_chown(tp, ip, old, new) (NULL)
@@ -359,7 +364,6 @@ extern void xfs_qm_unmount_quotas(struct xfs_mount *);
#define xfs_qm_dqdetach(ip)
#define xfs_qm_dqrele(d)
#define xfs_qm_statvfs(ip, s)
-#define xfs_qm_sync(mp, flags) (0)
#define xfs_qm_newmount(mp, a, b) (0)
#define xfs_qm_mount_quotas(mp)
#define xfs_qm_unmount(mp)
@@ -372,4 +376,9 @@ extern void xfs_qm_unmount_quotas(struct xfs_mount *);
xfs_trans_reserve_quota_bydquots(tp, mp, ud, gd, nb, ni, \
f | XFS_QMOPT_RES_REGBLKS)
+extern int xfs_qm_dqcheck(struct xfs_mount *, xfs_disk_dquot_t *,
+ xfs_dqid_t, uint, uint, char *);
+extern int xfs_mount_reset_sbqflags(struct xfs_mount *);
+
+#endif /* __KERNEL__ */
#endif /* __XFS_QUOTA_H__ */
diff --git a/include/xfs_rtalloc.h b/include/xfs_rtalloc.h
index ff614c2..f7f3a35 100644
--- a/include/xfs_rtalloc.h
+++ b/include/xfs_rtalloc.h
@@ -47,7 +47,7 @@ struct xfs_trans;
#define XFS_SUMOFFSTOBLOCK(mp,s) \
(((s) * (uint)sizeof(xfs_suminfo_t)) >> (mp)->m_sb.sb_blocklog)
#define XFS_SUMPTR(mp,bp,so) \
- ((xfs_suminfo_t *)((char *)XFS_BUF_PTR(bp) + \
+ ((xfs_suminfo_t *)((bp)->b_addr + \
(((so) * (uint)sizeof(xfs_suminfo_t)) & XFS_BLOCKMASK(mp))))
#define XFS_BITTOBLOCK(mp,bi) ((bi) >> (mp)->m_blkbit_log)
@@ -154,7 +154,7 @@ xfs_rtmount_init(
if (mp->m_sb.sb_rblocks == 0)
return 0;
- cmn_err(CE_WARN, "XFS: Not built with CONFIG_XFS_RT");
+ xfs_warn(mp, "Not built with CONFIG_XFS_RT");
return ENOSYS;
}
# define xfs_rtmount_inodes(m) (((mp)->m_sb.sb_rblocks == 0)? 0 : (ENOSYS))
diff --git a/include/xfs_sb.h b/include/xfs_sb.h
index 5dcc2d7..6a7f8b0 100644
--- a/include/xfs_sb.h
+++ b/include/xfs_sb.h
@@ -81,6 +81,7 @@ struct xfs_mount;
#define XFS_SB_VERSION2_ATTR2BIT 0x00000008 /* Inline attr rework */
#define XFS_SB_VERSION2_PARENTBIT 0x00000010 /* parent pointers */
#define XFS_SB_VERSION2_PROJID32BIT 0x00000080 /* 32 bit project id */
+#define XFS_SB_VERSION2_CRCBIT 0x00000100 /* metadata CRCs */
#define XFS_SB_VERSION2_OKREALFBITS \
(XFS_SB_VERSION2_LAZYSBCOUNTBIT | \
@@ -510,13 +511,19 @@ static inline void xfs_sb_version_addprojid32bit(xfs_sb_t *sbp)
sbp->sb_bad_features2 |= XFS_SB_VERSION2_PROJID32BIT;
}
+static inline int xfs_sb_version_hascrc(xfs_sb_t *sbp)
+{
+ return (xfs_sb_version_hasmorebits(sbp) &&
+ (sbp->sb_features2 & XFS_SB_VERSION2_CRCBIT));
+}
+
/*
* end of superblock version macros
*/
#define XFS_SB_DADDR ((xfs_daddr_t)0) /* daddr in filesystem/ag */
#define XFS_SB_BLOCK(mp) XFS_HDR_BLOCK(mp, XFS_SB_DADDR)
-#define XFS_BUF_TO_SBP(bp) ((xfs_dsb_t *)XFS_BUF_PTR(bp))
+#define XFS_BUF_TO_SBP(bp) ((xfs_dsb_t *)((bp)->b_addr))
#define XFS_HDR_BLOCK(mp,d) ((xfs_agblock_t)XFS_BB_TO_FSBT(mp,d))
#define XFS_DADDR_TO_FSB(mp,d) XFS_AGB_TO_FSB(mp, \
@@ -536,7 +543,6 @@ static inline void xfs_sb_version_addprojid32bit(xfs_sb_t *sbp)
#define XFS_BB_TO_FSB(mp,bb) \
(((bb) + (XFS_FSB_TO_BB(mp,1) - 1)) >> (mp)->m_blkbb_log)
#define XFS_BB_TO_FSBT(mp,bb) ((bb) >> (mp)->m_blkbb_log)
-#define XFS_BB_FSB_OFFSET(mp,bb) ((bb) & ((mp)->m_bsize - 1))
/*
* File system block to byte conversions.
diff --git a/include/xfs_trace.h b/include/xfs_trace.h
index 1abcf28..5f32097 100644
--- a/include/xfs_trace.h
+++ b/include/xfs_trace.h
@@ -26,11 +26,14 @@
#define trace_xfs_alloc_near_greater(a) ((void) 0)
#define trace_xfs_alloc_near_lesser(a) ((void) 0)
#define trace_xfs_alloc_near_error(a) ((void) 0)
+#define trace_xfs_alloc_near_noentry(a) ((void) 0)
+#define trace_xfs_alloc_near_busy(a) ((void) 0)
#define trace_xfs_alloc_size_neither(a) ((void) 0)
#define trace_xfs_alloc_size_noentry(a) ((void) 0)
#define trace_xfs_alloc_size_nominleft(a) ((void) 0)
#define trace_xfs_alloc_size_done(a) ((void) 0)
#define trace_xfs_alloc_size_error(a) ((void) 0)
+#define trace_xfs_alloc_size_busy(a) ((void) 0)
#define trace_xfs_alloc_small_freelist(a) ((void) 0)
#define trace_xfs_alloc_small_notenough(a) ((void) 0)
#define trace_xfs_alloc_small_done(a) ((void) 0)
@@ -91,12 +94,74 @@
#define trace_xfs_dir2_sf_toino4(a) ((void) 0)
#define trace_xfs_dir2_sf_toino8(a) ((void) 0)
+#define trace_xfs_da_node_create(a) ((void) 0)
+#define trace_xfs_da_split(a) ((void) 0)
+#define trace_xfs_attr_leaf_split_before(a) ((void) 0)
+#define trace_xfs_attr_leaf_split_after(a) ((void) 0)
+#define trace_xfs_da_root_split(a) ((void) 0)
+#define trace_xfs_da_node_split(a) ((void) 0)
+#define trace_xfs_da_node_rebalance(a) ((void) 0)
+#define trace_xfs_da_node_add(a) ((void) 0)
+#define trace_xfs_da_join(a) ((void) 0)
+#define trace_xfs_da_root_join(a) ((void) 0)
+#define trace_xfs_da_node_toosmall(a) ((void) 0)
+#define trace_xfs_da_fixhashpath(a) ((void) 0)
+#define trace_xfs_da_node_remove(a) ((void) 0)
+#define trace_xfs_da_node_unbalance(a) ((void) 0)
+#define trace_xfs_da_link_before(a) ((void) 0)
+#define trace_xfs_da_link_after(a) ((void) 0)
+#define trace_xfs_da_unlink_back(a) ((void) 0)
+#define trace_xfs_da_unlink_forward(a) ((void) 0)
+#define trace_xfs_da_path_shift(a) ((void) 0)
+#define trace_xfs_da_grow_inode(a) ((void) 0)
+#define trace_xfs_da_swap_lastblock(a) ((void) 0)
+#define trace_xfs_da_shrink_inode(a) ((void) 0)
+
+#define trace_xfs_attr_sf_create(a) ((void) 0)
+#define trace_xfs_attr_sf_add(a) ((void) 0)
+#define trace_xfs_attr_sf_remove(a) ((void) 0)
+#define trace_xfs_attr_sf_lookup(a) ((void) 0)
+#define trace_xfs_attr_sf_to_leaf(a) ((void) 0)
+#define trace_xfs_attr_leaf_to_sf(a) ((void) 0)
+#define trace_xfs_attr_leaf_to_node(a) ((void) 0)
+#define trace_xfs_attr_leaf_create(a) ((void) 0)
+#define trace_xfs_attr_leaf_split(a) ((void) 0)
+#define trace_xfs_attr_leaf_add_old(a) ((void) 0)
+#define trace_xfs_attr_leaf_add_new(a) ((void) 0)
+#define trace_xfs_attr_leaf_add(a) ((void) 0)
+#define trace_xfs_attr_leaf_add_work(a) ((void) 0)
+#define trace_xfs_attr_leaf_compact(a) ((void) 0)
+#define trace_xfs_attr_leaf_rebalance(a) ((void) 0)
+#define trace_xfs_attr_leaf_toosmall(a) ((void) 0)
+#define trace_xfs_attr_leaf_remove(a) ((void) 0)
+#define trace_xfs_attr_leaf_unbalance(a) ((void) 0)
+#define trace_xfs_attr_leaf_lookup(a) ((void) 0)
+#define trace_xfs_attr_leaf_clearflag(a) ((void) 0)
+#define trace_xfs_attr_leaf_setflag(a) ((void) 0)
+#define trace_xfs_attr_leaf_flipflags(a) ((void) 0)
+
+#define trace_xfs_attr_sf_addname(a) ((void) 0)
+#define trace_xfs_attr_leaf_addname(a) ((void) 0)
+#define trace_xfs_attr_leaf_replace(a) ((void) 0)
+#define trace_xfs_attr_leaf_removename(a) ((void) 0)
+#define trace_xfs_attr_leaf_get(a) ((void) 0)
+#define trace_xfs_attr_node_addname(a) ((void) 0)
+#define trace_xfs_attr_node_replace(a) ((void) 0)
+#define trace_xfs_attr_node_removename(a) ((void) 0)
+#define trace_xfs_attr_fillstate(a) ((void) 0)
+#define trace_xfs_attr_refillstate(a) ((void) 0)
+#define trace_xfs_attr_node_get(a) ((void) 0)
+#define trace_xfs_attr_rmtval_get(a) ((void) 0)
+#define trace_xfs_attr_rmtval_set(a) ((void) 0)
+#define trace_xfs_attr_rmtval_remove(a) ((void) 0)
+
#define trace_xfs_bmap_pre_update(a,b,c,d) ((void) 0)
#define trace_xfs_bmap_post_update(a,b,c,d) ((void) 0)
#define trace_xfs_extlist(a,b,c,d) ((void) 0)
#define trace_xfs_bunmap(a,b,c,d,e) ((void) 0)
-#define trace_xfs_perag_get(a,b,c,d) ((void) 0)
-#define trace_xfs_perag_put(a,b,c,d) ((void) 0)
+/* set c = c to avoid unused var warnings */
+#define trace_xfs_perag_get(a,b,c,d) ((c) = (c))
+#define trace_xfs_perag_put(a,b,c,d) ((c) = (c))
#endif /* __TRACE_H__ */
diff --git a/include/xfs_trans.h b/include/xfs_trans.h
index 2870308..acf1381 100644
--- a/include/xfs_trans.h
+++ b/include/xfs_trans.h
@@ -163,9 +163,8 @@ typedef struct xfs_trans_header {
*/
struct xfs_log_item_desc {
struct xfs_log_item *lid_item;
- ushort lid_size;
- unsigned char lid_flags;
struct list_head lid_trans;
+ unsigned char lid_flags;
};
#define XFS_LID_DIRTY 0x1
@@ -180,6 +179,8 @@ struct xfs_log_item_desc {
#define XFS_TRANS_SYNC 0x08 /* make commit synchronous */
#define XFS_TRANS_DQ_DIRTY 0x10 /* at least one dquot in trx dirty */
#define XFS_TRANS_RESERVE 0x20 /* OK to use reserved data blocks */
+#define XFS_TRANS_FREEZE_PROT 0x40 /* Transaction has elevated writer
+ count in superblock */
/*
* Values for call flags parameter.
@@ -323,7 +324,7 @@ typedef struct xfs_log_item {
struct xfs_log_item *);
/* buffer item iodone */
/* callback func */
- struct xfs_item_ops *li_ops; /* function list */
+ const struct xfs_item_ops *li_ops; /* function list */
/* delayed logging */
struct list_head li_cil; /* CIL pointers */
@@ -338,37 +339,33 @@ typedef struct xfs_log_item {
{ XFS_LI_IN_AIL, "IN_AIL" }, \
{ XFS_LI_ABORTED, "ABORTED" }
-typedef struct xfs_item_ops {
+struct xfs_item_ops {
uint (*iop_size)(xfs_log_item_t *);
void (*iop_format)(xfs_log_item_t *, struct xfs_log_iovec *);
void (*iop_pin)(xfs_log_item_t *);
void (*iop_unpin)(xfs_log_item_t *, int remove);
- uint (*iop_trylock)(xfs_log_item_t *);
+ uint (*iop_push)(struct xfs_log_item *, struct list_head *);
void (*iop_unlock)(xfs_log_item_t *);
xfs_lsn_t (*iop_committed)(xfs_log_item_t *, xfs_lsn_t);
- void (*iop_push)(xfs_log_item_t *);
- void (*iop_pushbuf)(xfs_log_item_t *);
void (*iop_committing)(xfs_log_item_t *, xfs_lsn_t);
-} xfs_item_ops_t;
+};
#define IOP_SIZE(ip) (*(ip)->li_ops->iop_size)(ip)
#define IOP_FORMAT(ip,vp) (*(ip)->li_ops->iop_format)(ip, vp)
#define IOP_PIN(ip) (*(ip)->li_ops->iop_pin)(ip)
#define IOP_UNPIN(ip, remove) (*(ip)->li_ops->iop_unpin)(ip, remove)
-#define IOP_TRYLOCK(ip) (*(ip)->li_ops->iop_trylock)(ip)
+#define IOP_PUSH(ip, list) (*(ip)->li_ops->iop_push)(ip, list)
#define IOP_UNLOCK(ip) (*(ip)->li_ops->iop_unlock)(ip)
#define IOP_COMMITTED(ip, lsn) (*(ip)->li_ops->iop_committed)(ip, lsn)
-#define IOP_PUSH(ip) (*(ip)->li_ops->iop_push)(ip)
-#define IOP_PUSHBUF(ip) (*(ip)->li_ops->iop_pushbuf)(ip)
#define IOP_COMMITTING(ip, lsn) (*(ip)->li_ops->iop_committing)(ip, lsn)
/*
- * Return values for the IOP_TRYLOCK() routines.
+ * Return values for the IOP_PUSH() routines.
*/
-#define XFS_ITEM_SUCCESS 0
-#define XFS_ITEM_PINNED 1
-#define XFS_ITEM_LOCKED 2
-#define XFS_ITEM_PUSHBUF 3
+#define XFS_ITEM_SUCCESS 0
+#define XFS_ITEM_PINNED 1
+#define XFS_ITEM_LOCKED 2
+#define XFS_ITEM_FLUSHING 3
/*
* This is the type of function which can be given to xfs_trans_callback()
@@ -445,16 +442,53 @@ typedef struct xfs_trans {
* XFS transaction mechanism exported interfaces.
*/
xfs_trans_t *xfs_trans_alloc(struct xfs_mount *, uint);
-xfs_trans_t *_xfs_trans_alloc(struct xfs_mount *, uint, uint);
+xfs_trans_t *_xfs_trans_alloc(struct xfs_mount *, uint, xfs_km_flags_t);
xfs_trans_t *xfs_trans_dup(xfs_trans_t *);
int xfs_trans_reserve(xfs_trans_t *, uint, uint, uint,
uint, uint);
void xfs_trans_mod_sb(xfs_trans_t *, uint, int64_t);
-struct xfs_buf *xfs_trans_get_buf(xfs_trans_t *, struct xfs_buftarg *, xfs_daddr_t,
- int, uint);
-int xfs_trans_read_buf(struct xfs_mount *, xfs_trans_t *,
- struct xfs_buftarg *, xfs_daddr_t, int, uint,
- struct xfs_buf **);
+
+struct xfs_buf *xfs_trans_get_buf_map(struct xfs_trans *tp,
+ struct xfs_buftarg *target,
+ struct xfs_buf_map *map, int nmaps,
+ uint flags);
+
+static inline struct xfs_buf *
+xfs_trans_get_buf(
+ struct xfs_trans *tp,
+ struct xfs_buftarg *target,
+ xfs_daddr_t blkno,
+ int numblks,
+ uint flags)
+{
+ DEFINE_SINGLE_BUF_MAP(map, blkno, numblks);
+ return xfs_trans_get_buf_map(tp, target, &map, 1, flags);
+}
+
+int xfs_trans_read_buf_map(struct xfs_mount *mp,
+ struct xfs_trans *tp,
+ struct xfs_buftarg *target,
+ struct xfs_buf_map *map, int nmaps,
+ xfs_buf_flags_t flags,
+ struct xfs_buf **bpp,
+ const struct xfs_buf_ops *ops);
+
+static inline int
+xfs_trans_read_buf(
+ struct xfs_mount *mp,
+ struct xfs_trans *tp,
+ struct xfs_buftarg *target,
+ xfs_daddr_t blkno,
+ int numblks,
+ xfs_buf_flags_t flags,
+ struct xfs_buf **bpp,
+ const struct xfs_buf_ops *ops)
+{
+ DEFINE_SINGLE_BUF_MAP(map, blkno, numblks);
+ return xfs_trans_read_buf_map(mp, tp, target, &map, 1,
+ flags, bpp, ops);
+}
+
struct xfs_buf *xfs_trans_getsb(xfs_trans_t *, struct xfs_mount *, int);
void xfs_trans_brelse(xfs_trans_t *, struct xfs_buf *);
@@ -466,11 +500,8 @@ void xfs_trans_inode_buf(xfs_trans_t *, struct xfs_buf *);
void xfs_trans_stale_inode_buf(xfs_trans_t *, struct xfs_buf *);
void xfs_trans_dquot_buf(xfs_trans_t *, struct xfs_buf *, uint);
void xfs_trans_inode_alloc_buf(xfs_trans_t *, struct xfs_buf *);
-int xfs_trans_iget(struct xfs_mount *, xfs_trans_t *,
- xfs_ino_t , uint, uint, struct xfs_inode **);
void xfs_trans_ichgtime(struct xfs_trans *, struct xfs_inode *, int);
-void xfs_trans_ijoin_ref(struct xfs_trans *, struct xfs_inode *, uint);
-void xfs_trans_ijoin(struct xfs_trans *, struct xfs_inode *);
+void xfs_trans_ijoin(struct xfs_trans *, struct xfs_inode *, uint);
void xfs_trans_log_buf(xfs_trans_t *, struct xfs_buf *, uint, uint);
void xfs_trans_log_inode(xfs_trans_t *, struct xfs_inode *, uint);
struct xfs_efi_log_item *xfs_trans_get_efi(xfs_trans_t *, uint);
@@ -486,10 +517,7 @@ void xfs_trans_log_efd_extent(xfs_trans_t *,
struct xfs_efd_log_item *,
xfs_fsblock_t,
xfs_extlen_t);
-int _xfs_trans_commit(xfs_trans_t *,
- uint flags,
- int *);
-#define xfs_trans_commit(tp, flags) _xfs_trans_commit(tp, flags, NULL)
+int xfs_trans_commit(xfs_trans_t *, uint flags);
void xfs_trans_cancel(xfs_trans_t *, int);
int xfs_trans_ail_init(struct xfs_mount *);
void xfs_trans_ail_destroy(struct xfs_mount *);
diff --git a/include/xfs_types.h b/include/xfs_types.h
index 26d1867..e9bd5c3 100644
--- a/include/xfs_types.h
+++ b/include/xfs_types.h
@@ -33,7 +33,6 @@ typedef signed long long int __int64_t;
typedef unsigned long long int __uint64_t;
typedef enum { B_FALSE,B_TRUE } boolean_t;
-typedef __uint32_t prid_t; /* project ID */
typedef __uint32_t inst_t; /* an instruction */
typedef __s64 xfs_off_t; /* <file offset> type */
@@ -56,7 +55,10 @@ typedef __uint64_t __psunsigned_t;
#endif /* __KERNEL__ */
+typedef __uint32_t prid_t; /* project ID */
+
typedef __uint32_t xfs_agblock_t; /* blockno in alloc. group */
+typedef __uint32_t xfs_agino_t; /* inode # within allocation grp */
typedef __uint32_t xfs_extlen_t; /* extent length in blocks */
typedef __uint32_t xfs_agnumber_t; /* allocation group number */
typedef __int32_t xfs_extnum_t; /* # of extents in a file */
@@ -73,8 +75,6 @@ typedef __int32_t xfs_tid_t; /* transaction identifier */
typedef __uint32_t xfs_dablk_t; /* dir/attr block number (in file) */
typedef __uint32_t xfs_dahash_t; /* dir/attr hash value */
-typedef __uint32_t xlog_tid_t; /* transaction ID type */
-
/*
* These types are 64 bits on disk but are either 32 or 64 bits in memory.
* Disk based types:
@@ -103,6 +103,7 @@ typedef __uint64_t xfs_fileoff_t; /* block number in a file */
typedef __int64_t xfs_sfiloff_t; /* signed block number in a file */
typedef __uint64_t xfs_filblks_t; /* number of blocks in a file */
+
/*
* Null values for the types.
*/
@@ -122,6 +123,9 @@ typedef __uint64_t xfs_filblks_t; /* number of blocks in a file */
#define NULLCOMMITLSN ((xfs_lsn_t)-1)
+#define NULLFSINO ((xfs_ino_t)-1)
+#define NULLAGINO ((xfs_agino_t)-1)
+
/*
* Max values for extlen, extnum, aextnum.
*/
@@ -130,6 +134,20 @@ typedef __uint64_t xfs_filblks_t; /* number of blocks in a file */
#define MAXAEXTNUM ((xfs_aextnum_t)0x7fff) /* signed short */
/*
+ * Minimum and maximum blocksize and sectorsize.
+ * The blocksize upper limit is pretty much arbitrary.
+ * The sectorsize upper limit is due to sizeof(sb_sectsize).
+ */
+#define XFS_MIN_BLOCKSIZE_LOG 9 /* i.e. 512 bytes */
+#define XFS_MAX_BLOCKSIZE_LOG 16 /* i.e. 65536 bytes */
+#define XFS_MIN_BLOCKSIZE (1 << XFS_MIN_BLOCKSIZE_LOG)
+#define XFS_MAX_BLOCKSIZE (1 << XFS_MAX_BLOCKSIZE_LOG)
+#define XFS_MIN_SECTORSIZE_LOG 9 /* i.e. 512 bytes */
+#define XFS_MAX_SECTORSIZE_LOG 15 /* i.e. 32768 bytes */
+#define XFS_MIN_SECTORSIZE (1 << XFS_MIN_SECTORSIZE_LOG)
+#define XFS_MAX_SECTORSIZE (1 << XFS_MAX_SECTORSIZE_LOG)
+
+/*
* Min numbers of data/attr fork btree root pointers.
*/
#define MINDBTPTRS 3
diff --git a/libxfs/Makefile b/libxfs/Makefile
index b3fd85c..dc9b22c 100644
--- a/libxfs/Makefile
+++ b/libxfs/Makefile
@@ -10,7 +10,7 @@ LT_CURRENT = 0
LT_REVISION = 0
LT_AGE = 0
-HFILES = xfs.h init.h
+HFILES = xfs.h init.h xfs_dir2_priv.h
CFILES = cache.c init.c kmem.c logitem.c radix-tree.c rdwr.c trans.c util.c \
xfs_alloc.c xfs_ialloc.c xfs_inode.c xfs_btree.c xfs_alloc_btree.c \
xfs_ialloc_btree.c xfs_bmap_btree.c xfs_da_btree.c \
diff --git a/libxfs/init.c b/libxfs/init.c
index 08fc584..1b11fb0 100644
--- a/libxfs/init.c
+++ b/libxfs/init.c
@@ -371,7 +371,6 @@ manage_zones(int release)
extern kmem_zone_t *xfs_ili_zone;
extern kmem_zone_t *xfs_inode_zone;
extern kmem_zone_t *xfs_ifork_zone;
- extern kmem_zone_t *xfs_dabuf_zone;
extern kmem_zone_t *xfs_buf_item_zone;
extern kmem_zone_t *xfs_da_state_zone;
extern kmem_zone_t *xfs_btree_cur_zone;
@@ -383,7 +382,6 @@ manage_zones(int release)
kmem_free(xfs_buf_zone);
kmem_free(xfs_inode_zone);
kmem_free(xfs_ifork_zone);
- kmem_free(xfs_dabuf_zone);
kmem_free(xfs_buf_item_zone);
kmem_free(xfs_da_state_zone);
kmem_free(xfs_btree_cur_zone);
@@ -395,7 +393,6 @@ manage_zones(int release)
xfs_buf_zone = kmem_zone_init(sizeof(xfs_buf_t), "xfs_buffer");
xfs_inode_zone = kmem_zone_init(sizeof(xfs_inode_t), "xfs_inode");
xfs_ifork_zone = kmem_zone_init(sizeof(xfs_ifork_t), "xfs_ifork");
- xfs_dabuf_zone = kmem_zone_init(sizeof(xfs_dabuf_t), "xfs_dabuf");
xfs_ili_zone = kmem_zone_init(
sizeof(xfs_inode_log_item_t), "xfs_inode_log_item");
xfs_buf_item_zone = kmem_zone_init(
diff --git a/libxfs/logitem.c b/libxfs/logitem.c
index 116d3ed..84e4c14 100644
--- a/libxfs/logitem.c
+++ b/libxfs/logitem.c
@@ -32,21 +32,27 @@ kmem_zone_t *xfs_ili_zone; /* inode log item zone */
xfs_buf_t *
xfs_trans_buf_item_match(
xfs_trans_t *tp,
- xfs_buftarg_t *target,
- xfs_daddr_t blkno,
- int len)
+ dev_t dev,
+ struct xfs_buf_map *map,
+ int nmaps)
{
struct xfs_log_item_desc *lidp;
struct xfs_buf_log_item *blip;
+ int len = 0;
+ int i;
+
+ for (i = 0; i < nmaps; i++)
+ len += map[i].bm_len;
- len = BBTOB(len);
list_for_each_entry(lidp, &tp->t_items, lid_trans) {
blip = (struct xfs_buf_log_item *)lidp->lid_item;
if (blip->bli_item.li_type == XFS_LI_BUF &&
- XFS_BUF_TARGET(blip->bli_buf) == target->dev &&
- XFS_BUF_ADDR(blip->bli_buf) == blkno &&
- XFS_BUF_COUNT(blip->bli_buf) == len)
+ blip->bli_buf->b_dev == dev &&
+ XFS_BUF_ADDR(blip->bli_buf) == map[0].bm_bn &&
+ blip->bli_buf->b_bcount == BBTOB(len)) {
+ ASSERT(blip->bli_buf->b_map_count == nmaps);
return blip->bli_buf;
+ }
}
return NULL;
diff --git a/libxfs/rdwr.c b/libxfs/rdwr.c
index 432a1af..e75edd0 100644
--- a/libxfs/rdwr.c
+++ b/libxfs/rdwr.c
@@ -159,7 +159,7 @@ libxfs_log_header(
head->h_len = cpu_to_be32(sunit - BBSIZE);
else
head->h_len = cpu_to_be32(20);
- head->h_chksum = cpu_to_be32(0);
+ head->h_crc = cpu_to_be32(0);
head->h_prev_block = cpu_to_be32(-1);
head->h_num_logops = cpu_to_be32(1);
head->h_cycle_data[0] = cpu_to_be32(0xb0c0d0d0);
@@ -193,72 +193,86 @@ libxfs_log_header(
#ifdef XFS_BUF_TRACING
#undef libxfs_readbuf
+#undef libxfs_readbuf_map
#undef libxfs_writebuf
#undef libxfs_getbuf
+#undef libxfs_getbuf_map
#undef libxfs_getbuf_flags
#undef libxfs_putbuf
-xfs_buf_t *libxfs_readbuf(dev_t, xfs_daddr_t, int, int);
+xfs_buf_t *libxfs_readbuf(dev_t, xfs_daddr_t, int, int);
+xfs_buf_t *libxfs_readbuf_map(dev_t, struct xfs_buf_map *, int, int);
int libxfs_writebuf(xfs_buf_t *, int);
-xfs_buf_t *libxfs_getbuf(dev_t, xfs_daddr_t, int);
+xfs_buf_t *libxfs_getbuf(dev_t, xfs_daddr_t, int);
+xfs_buf_t *libxfs_getbuf_map(dev_t, struct xfs_buf_map *, int);
+xfs_buf_t *libxfs_getbuf_flags(dev_t, xfs_daddr_t, int, unsigned int);
void libxfs_putbuf (xfs_buf_t *);
+#define __add_trace(bp, func, file, line) \
+do { \
+ if (bp) { \
+ (bp)->b_func = (func); \
+ (bp)->b_file = (file); \
+ (bp)->b_line = (line); \
+ } \
+} while (0)
+
xfs_buf_t *
-libxfs_trace_readbuf(const char *func, const char *file, int line, dev_t dev, xfs_daddr_t blkno, int len, int flags)
+libxfs_trace_readbuf(const char *func, const char *file, int line,
+ dev_t dev, xfs_daddr_t blkno, int len, int flags)
{
xfs_buf_t *bp = libxfs_readbuf(dev, blkno, len, flags);
+ __add_trace(bp, func, file, line);
+ return bp;
+}
- if (bp){
- bp->b_func = func;
- bp->b_file = file;
- bp->b_line = line;
- }
-
+xfs_buf_t *
+libxfs_trace_readbuf_map(const char *func, const char *file, int line,
+ dev_t dev, struct xfs_buf_map *map, int nmaps, int flags)
+{
+ xfs_buf_t *bp = libxfs_readbuf_map(dev, map, nmaps, flags);
+ __add_trace(bp, func, file, line);
return bp;
}
int
libxfs_trace_writebuf(const char *func, const char *file, int line, xfs_buf_t *bp, int flags)
{
- bp->b_func = func;
- bp->b_file = file;
- bp->b_line = line;
-
+ __add_trace(bp, func, file, line);
return libxfs_writebuf(bp, flags);
}
xfs_buf_t *
-libxfs_trace_getbuf(const char *func, const char *file, int line, dev_t device, xfs_daddr_t blkno, int len)
+libxfs_trace_getbuf(const char *func, const char *file, int line,
+ dev_t device, xfs_daddr_t blkno, int len)
{
xfs_buf_t *bp = libxfs_getbuf(device, blkno, len);
+ __add_trace(bp, func, file, line);
+ return bp;
+}
- bp->b_func = func;
- bp->b_file = file;
- bp->b_line = line;
-
+xfs_buf_t *
+libxfs_trace_getbuf_map(const char *func, const char *file, int line,
+ dev_t device, struct xfs_buf_map *map, int nmaps)
+{
+ xfs_buf_t *bp = libxfs_getbuf_map(device, map, nmaps);
+ __add_trace(bp, func, file, line);
return bp;
}
xfs_buf_t *
libxfs_trace_getbuf_flags(const char *func, const char *file, int line,
- dev_t device, xfs_daddr_t blkno, int len, unsigned long flags)
+ dev_t device, xfs_daddr_t blkno, int len, unsigned int flags)
{
- xfs_buf_t *bp = libxfs_getbuf(device, blkno, len, flags);
-
- bp->b_func = func;
- bp->b_file = file;
- bp->b_line = line;
-
+ xfs_buf_t *bp = libxfs_getbuf_flags(device, blkno, len, flags);
+ __add_trace(bp, func, file, line);
return bp;
}
void
libxfs_trace_putbuf(const char *func, const char *file, int line, xfs_buf_t *bp)
{
- bp->b_func = func;
- bp->b_file = file;
- bp->b_line = line;
-
+ __add_trace(bp, func, file, line);
libxfs_putbuf(bp);
}
@@ -279,30 +293,40 @@ static struct cache_mru xfs_buf_freelist =
{{&xfs_buf_freelist.cm_list, &xfs_buf_freelist.cm_list},
0, PTHREAD_MUTEX_INITIALIZER };
-typedef struct {
- dev_t device;
- xfs_daddr_t blkno;
- unsigned int bblen;
-} xfs_bufkey_t;
+/*
+ * The bufkey is used to pass the new buffer information to the cache object
+ * allocation routine. Because discontiguous buffers need to pass different
+ * information, we need fields to pass that information. However, because the
+ * blkno and bblen is needed for the initial cache entry lookup (i.e. for
+ * bcompare) the fact that the map/nmaps is non-null to switch to discontiguous
+ * buffer initialisation instead of a contiguous buffer.
+ */
+struct xfs_bufkey {
+ dev_t device;
+ xfs_daddr_t blkno;
+ unsigned int bblen;
+ struct xfs_buf_map *map;
+ int nmaps;
+};
static unsigned int
libxfs_bhash(cache_key_t key, unsigned int hashsize)
{
- return (((unsigned int)((xfs_bufkey_t *)key)->blkno) >> 5) % hashsize;
+ return (((unsigned int)((struct xfs_bufkey *)key)->blkno) >> 5) % hashsize;
}
static int
libxfs_bcompare(struct cache_node *node, cache_key_t key)
{
- xfs_buf_t *bp = (xfs_buf_t *)node;
- xfs_bufkey_t *bkey = (xfs_bufkey_t *)key;
+ struct xfs_buf *bp = (struct xfs_buf *)node;
+ struct xfs_bufkey *bkey = (struct xfs_bufkey *)key;
#ifdef IO_BCOMPARE_CHECK
if (bp->b_dev == bkey->device &&
bp->b_blkno == bkey->blkno &&
bp->b_bcount != BBTOB(bkey->bblen))
fprintf(stderr, "%lx: Badness in key lookup (length)\n"
- "bp=(bno %llu, len %u bytes) key=(bno %llu, len %u bytes)\n",
+ "bp=(bno 0x%llx, len %u bytes) key=(bno 0x%llx, len %u bytes)\n",
pthread_self(),
(unsigned long long)bp->b_blkno, (int)bp->b_bcount,
(unsigned long long)bkey->blkno, BBTOB(bkey->bblen));
@@ -322,11 +346,12 @@ libxfs_bprint(xfs_buf_t *bp)
}
static void
-libxfs_initbuf(xfs_buf_t *bp, dev_t device, xfs_daddr_t bno, unsigned int bytes)
+__initbuf(xfs_buf_t *bp, dev_t device, xfs_daddr_t bno, unsigned int bytes)
{
bp->b_flags = 0;
bp->b_blkno = bno;
bp->b_bcount = bytes;
+ bp->b_length = BTOBB(bytes);
bp->b_dev = device;
bp->b_error = 0;
if (!bp->b_addr)
@@ -346,11 +371,44 @@ libxfs_initbuf(xfs_buf_t *bp, dev_t device, xfs_daddr_t bno, unsigned int bytes)
bp->b_recur = 0;
}
+static void
+libxfs_initbuf(xfs_buf_t *bp, dev_t device, xfs_daddr_t bno, unsigned int bytes)
+{
+ __initbuf(bp, device, bno, bytes);
+}
+
+static void
+libxfs_initbuf_map(xfs_buf_t *bp, dev_t device, struct xfs_buf_map *map, int nmaps)
+{
+ unsigned int bytes = 0;
+ int i;
+
+ bytes = sizeof(struct xfs_buf_map) * nmaps;
+ bp->b_map = malloc(bytes);
+ if (!bp->b_map) {
+ fprintf(stderr,
+ _("%s: %s can't malloc %u bytes: %s\n"),
+ progname, __FUNCTION__, bytes,
+ strerror(errno));
+ exit(1);
+ }
+ bp->b_nmaps = nmaps;
+
+ bytes = 0;
+ for ( i = 0; i < nmaps; i++) {
+ bp->b_map[i].bm_bn = map[i].bm_bn;
+ bp->b_map[i].bm_len = map[i].bm_len;
+ bytes += BBTOB(map[i].bm_len);
+ }
+
+ __initbuf(bp, device, map[0].bm_bn, bytes);
+ bp->b_flags |= LIBXFS_B_DISCONTIG;
+}
+
xfs_buf_t *
-libxfs_getbufr(dev_t device, xfs_daddr_t blkno, int bblen)
+__libxfs_getbufr(int blen)
{
xfs_buf_t *bp;
- int blen = BBTOB(bblen);
/*
* first look for a buffer that can be used as-is,
@@ -372,15 +430,27 @@ libxfs_getbufr(dev_t device, xfs_daddr_t blkno, int bblen)
list_del_init(&bp->b_node.cn_mru);
free(bp->b_addr);
bp->b_addr = NULL;
+ free(bp->b_map);
+ bp->b_map = NULL;
}
} else
bp = kmem_zone_zalloc(xfs_buf_zone, 0);
pthread_mutex_unlock(&xfs_buf_freelist.cm_mutex);
- if (bp != NULL)
+ return bp;
+}
+
+xfs_buf_t *
+libxfs_getbufr(dev_t device, xfs_daddr_t blkno, int bblen)
+{
+ xfs_buf_t *bp;
+ int blen = BBTOB(bblen);
+
+ bp =__libxfs_getbufr(blen);
+ if (bp)
libxfs_initbuf(bp, device, blkno, blen);
#ifdef IO_DEBUG
- printf("%lx: %s: allocated %u bytes buffer, key=%llu(%llu), %p\n",
+ printf("%lx: %s: allocated %u bytes buffer, key=0x%llx(0x%llx), %p\n",
pthread_self(), __FUNCTION__, blen,
(long long)LIBXFS_BBTOOFF64(blkno), (long long)blkno, bp);
#endif
@@ -388,6 +458,38 @@ libxfs_getbufr(dev_t device, xfs_daddr_t blkno, int bblen)
return bp;
}
+xfs_buf_t *
+libxfs_getbufr_map(dev_t device, xfs_daddr_t blkno, int bblen,
+ struct xfs_buf_map *map, int nmaps)
+{
+ xfs_buf_t *bp;
+ int blen = BBTOB(bblen);
+
+ if (!map || !nmaps) {
+ fprintf(stderr,
+ _("%s: %s invalid map %p or nmaps %d\n"),
+ progname, __FUNCTION__, map, nmaps);
+ exit(1);
+ }
+
+ if (blkno != map[0].bm_bn) {
+ fprintf(stderr,
+ _("%s: %s map blkno %lx doesn't match key %lx\n"),
+ progname, __FUNCTION__, map[0].bm_bn, blkno);
+ exit(1);
+ }
+
+ bp =__libxfs_getbufr(blen);
+ if (bp)
+ libxfs_initbuf_map(bp, device, map, nmaps);
+#ifdef IO_DEBUG
+ printf("%lx: %s: allocated %u bytes buffer, key=0x%llx(0x%llx), %p\n",
+ pthread_self(), __FUNCTION__, blen,
+ (long long)LIBXFS_BBTOOFF64(blkno), (long long)blkno, bp);
+#endif
+
+ return bp;
+}
#ifdef XFS_BUF_TRACING
struct list_head lock_buf_list = {&lock_buf_list, &lock_buf_list};
@@ -396,18 +498,12 @@ int lock_buf_count = 0;
extern int use_xfs_buf_lock;
-struct xfs_buf *
-libxfs_getbuf_flags(dev_t device, xfs_daddr_t blkno, int len, unsigned int flags)
+static struct xfs_buf *
+__cache_lookup(struct xfs_bufkey *key, unsigned int flags)
{
- xfs_buf_t *bp;
- xfs_bufkey_t key;
- int miss;
-
- key.device = device;
- key.blkno = blkno;
- key.bblen = len;
+ struct xfs_buf *bp;
- miss = cache_node_get(libxfs_bcache, &key, (struct cache_node **)&bp);
+ cache_node_get(libxfs_bcache, key, (struct cache_node **)&bp);
if (!bp)
return NULL;
@@ -423,7 +519,7 @@ libxfs_getbuf_flags(dev_t device, xfs_daddr_t blkno, int len, unsigned int flags
if (pthread_equal(bp->b_holder, pthread_self())) {
fprintf(stderr,
_("Warning: recursive buffer locking at block %" PRIu64 " detected\n"),
- blkno);
+ key->blkno);
bp->b_recur++;
return bp;
} else {
@@ -444,9 +540,9 @@ libxfs_getbuf_flags(dev_t device, xfs_daddr_t blkno, int len, unsigned int flags
pthread_mutex_unlock(&libxfs_bcache->c_mutex);
#endif
#ifdef IO_DEBUG
- printf("%lx %s: %s buffer %p for bno = %llu\n",
- pthread_self(), __FUNCTION__, miss ? "miss" : "hit",
- bp, (long long)LIBXFS_BBTOOFF64(blkno));
+ printf("%lx %s: hit buffer %p for bno = 0x%llx/0x%llx\n",
+ pthread_self(), __FUNCTION__,
+ bp, bp->b_bn, (long long)LIBXFS_BBTOOFF64(key->blkno));
#endif
return bp;
@@ -456,11 +552,40 @@ out_put:
}
struct xfs_buf *
+libxfs_getbuf_flags(dev_t device, xfs_daddr_t blkno, int len, unsigned int flags)
+{
+ struct xfs_bufkey key = {0};
+
+ key.device = device;
+ key.blkno = blkno;
+ key.bblen = len;
+
+ return __cache_lookup(&key, flags);
+}
+
+struct xfs_buf *
libxfs_getbuf(dev_t device, xfs_daddr_t blkno, int len)
{
return libxfs_getbuf_flags(device, blkno, len, 0);
}
+struct xfs_buf *
+libxfs_getbuf_map(dev_t device, struct xfs_buf_map *map, int nmaps)
+{
+ struct xfs_bufkey key = {0};
+ int i;
+
+ key.device = device;
+ key.blkno = map[0].bm_bn;
+ for (i = 0; i < nmaps; i++) {
+ key.bblen += map[i].bm_len;
+ }
+ key.map = map;
+ key.nmaps = nmaps;
+
+ return __cache_lookup(&key, 0);
+}
+
void
libxfs_putbuf(xfs_buf_t *bp)
{
@@ -485,7 +610,7 @@ libxfs_putbuf(xfs_buf_t *bp)
void
libxfs_purgebuf(xfs_buf_t *bp)
{
- xfs_bufkey_t key;
+ struct xfs_bufkey key = {0};
key.device = bp->b_dev;
key.blkno = bp->b_blkno;
@@ -497,47 +622,62 @@ libxfs_purgebuf(xfs_buf_t *bp)
static struct cache_node *
libxfs_balloc(cache_key_t key)
{
- xfs_bufkey_t *bufkey = (xfs_bufkey_t *)key;
+ struct xfs_bufkey *bufkey = (struct xfs_bufkey *)key;
+ if (bufkey->map)
+ return (struct cache_node *)
+ libxfs_getbufr_map(bufkey->device,
+ bufkey->blkno, bufkey->bblen,
+ bufkey->map, bufkey->nmaps);
return (struct cache_node *)libxfs_getbufr(bufkey->device,
- bufkey->blkno, bufkey->bblen);
+ bufkey->blkno, bufkey->bblen);
}
-int
-libxfs_readbufr(dev_t dev, xfs_daddr_t blkno, xfs_buf_t *bp, int len, int flags)
+
+static int
+__read_buf(int fd, void *buf, int len, off64_t offset, int flags)
{
- int fd = libxfs_device_to_fd(dev);
- int bytes = BBTOB(len);
- int error;
int sts;
- ASSERT(BBTOB(len) <= bp->b_bcount);
-
- sts = pread64(fd, bp->b_addr, bytes, LIBXFS_BBTOOFF64(blkno));
+ sts = pread64(fd, buf, len, offset);
if (sts < 0) {
- error = errno;
+ int error = errno;
fprintf(stderr, _("%s: read failed: %s\n"),
progname, strerror(error));
if (flags & LIBXFS_EXIT_ON_FAILURE)
exit(1);
return error;
- } else if (sts != bytes) {
+ } else if (sts != len) {
fprintf(stderr, _("%s: error - read only %d of %d bytes\n"),
- progname, sts, bytes);
+ progname, sts, len);
if (flags & LIBXFS_EXIT_ON_FAILURE)
exit(1);
return EIO;
}
-#ifdef IO_DEBUG
- printf("%lx: %s: read %u bytes, blkno=%llu(%llu), %p\n",
- pthread_self(), __FUNCTION__, bytes,
- (long long)LIBXFS_BBTOOFF64(blkno), (long long)blkno, bp);
-#endif
- if (bp->b_dev == dev &&
+ return 0;
+}
+
+int
+libxfs_readbufr(dev_t dev, xfs_daddr_t blkno, xfs_buf_t *bp, int len, int flags)
+{
+ int fd = libxfs_device_to_fd(dev);
+ int bytes = BBTOB(len);
+ int error;
+
+ ASSERT(BBTOB(len) <= bp->b_bcount);
+
+ error = __read_buf(fd, bp->b_addr, bytes, LIBXFS_BBTOOFF64(blkno), flags);
+ if (!error &&
+ bp->b_dev == dev &&
bp->b_blkno == blkno &&
bp->b_bcount == bytes)
bp->b_flags |= LIBXFS_B_UPTODATE;
- return 0;
+#ifdef IO_DEBUG
+ printf("%lx: %s: read %u bytes, error %d, blkno=0x%llx(0x%llx), %p\n",
+ pthread_self(), __FUNCTION__, bytes, error,
+ (long long)LIBXFS_BBTOOFF64(blkno), (long long)blkno, bp);
+#endif
+ return error;
}
xfs_buf_t *
@@ -555,37 +695,113 @@ libxfs_readbuf(dev_t dev, xfs_daddr_t blkno, int len, int flags)
return bp;
}
-int
-libxfs_writebufr(xfs_buf_t *bp)
+struct xfs_buf *
+libxfs_readbuf_map(dev_t dev, struct xfs_buf_map *map, int nmaps, int flags)
+{
+ xfs_buf_t *bp;
+ int error = 0;
+ int fd;
+ int i;
+ char *buf;
+
+ if (nmaps == 1)
+ return libxfs_readbuf(dev, map[0].bm_bn, map[0].bm_len, flags);
+
+ bp = libxfs_getbuf_map(dev, map, nmaps);
+ if (!bp || (bp->b_flags & (LIBXFS_B_UPTODATE|LIBXFS_B_DIRTY)))
+ return bp;
+
+ ASSERT(bp->b_nmaps = nmaps);
+
+ fd = libxfs_device_to_fd(dev);
+ buf = bp->b_addr;
+ for (i = 0; i < bp->b_nmaps; i++) {
+ off64_t offset = LIBXFS_BBTOOFF64(bp->b_map[i].bm_bn);
+ int len = BBTOB(bp->b_map[i].bm_len);
+
+ ASSERT(bp->b_map[i].bm_bn == map[i].bm_bn);
+ ASSERT(bp->b_map[i].bm_len == map[i].bm_len);
+
+ error = __read_buf(fd, buf, len, offset, flags);
+ if (error) {
+ bp->b_error = error;
+ break;
+ }
+ buf += len;
+ offset += len;
+ }
+
+ if (!error)
+ bp->b_flags |= LIBXFS_B_UPTODATE;
+#ifdef IO_DEBUG
+ printf("%lx: %s: read %lu bytes, error %d, blkno=%llu(%llu), %p\n",
+ pthread_self(), __FUNCTION__, buf - (char *)bp->b_addr, error,
+ (long long)LIBXFS_BBTOOFF64(bp->b_blkno), (long long)bp->b_blkno, bp);
+#endif
+ return bp;
+}
+
+static int
+__write_buf(int fd, void *buf, int len, off64_t offset, int flags)
{
int sts;
- int fd = libxfs_device_to_fd(bp->b_dev);
- int error;
- sts = pwrite64(fd, bp->b_addr, bp->b_bcount, LIBXFS_BBTOOFF64(bp->b_blkno));
+ sts = pwrite64(fd, buf, len, offset);
if (sts < 0) {
- error = errno;
+ int error = errno;
fprintf(stderr, _("%s: pwrite64 failed: %s\n"),
progname, strerror(error));
- if (bp->b_flags & LIBXFS_B_EXIT)
+ if (flags & LIBXFS_B_EXIT)
exit(1);
return error;
- } else if (sts != bp->b_bcount) {
- fprintf(stderr, _("%s: error - wrote only %d of %d bytes\n"),
- progname, sts, bp->b_bcount);
- if (bp->b_flags & LIBXFS_B_EXIT)
+ } else if (sts != len) {
+ fprintf(stderr, _("%s: error - pwrite64 only %d of %d bytes\n"),
+ progname, sts, len);
+ if (flags & LIBXFS_B_EXIT)
exit(1);
return EIO;
}
+ return 0;
+}
+
+int
+libxfs_writebufr(xfs_buf_t *bp)
+{
+ int fd = libxfs_device_to_fd(bp->b_dev);
+ int error = 0;
+
+ if (!(bp->b_flags & LIBXFS_B_DISCONTIG)) {
+ error = __write_buf(fd, bp->b_addr, bp->b_bcount,
+ LIBXFS_BBTOOFF64(bp->b_blkno), bp->b_flags);
+ } else {
+ int i;
+ char *buf = bp->b_addr;
+
+ for (i = 0; i < bp->b_nmaps; i++) {
+ off64_t offset = LIBXFS_BBTOOFF64(bp->b_map[i].bm_bn);
+ int len = BBTOB(bp->b_map[i].bm_len);
+
+ error = __write_buf(fd, buf, len, offset, bp->b_flags);
+ if (error) {
+ bp->b_error = error;
+ break;
+ }
+ buf += len;
+ offset += len;
+ }
+ }
+
#ifdef IO_DEBUG
printf("%lx: %s: wrote %u bytes, blkno=%llu(%llu), %p\n",
pthread_self(), __FUNCTION__, bp->b_bcount,
(long long)LIBXFS_BBTOOFF64(bp->b_blkno),
(long long)bp->b_blkno, bp);
#endif
- bp->b_flags |= LIBXFS_B_UPTODATE;
- bp->b_flags &= ~(LIBXFS_B_DIRTY | LIBXFS_B_EXIT);
- return 0;
+ if (!error) {
+ bp->b_flags |= LIBXFS_B_UPTODATE;
+ bp->b_flags &= ~(LIBXFS_B_DIRTY | LIBXFS_B_EXIT);
+ }
+ return error;
}
int
@@ -609,7 +825,7 @@ libxfs_iomove(xfs_buf_t *bp, uint boff, int len, void *data, int flags)
#ifdef IO_DEBUG
if (boff + len > bp->b_bcount) {
printf("Badness, iomove out of range!\n"
- "bp=(bno %llu, bytes %u) range=(boff %u, bytes %u)\n",
+ "bp=(bno 0x%llx, bytes %u) range=(boff %u, bytes %u)\n",
(long long)bp->b_blkno, bp->b_bcount, boff, len);
abort();
}
@@ -742,7 +958,10 @@ libxfs_iget(xfs_mount_t *mp, xfs_trans_t *tp, xfs_ino_t ino, uint lock_flags,
fprintf(stderr, "%s: allocated inode, ino=%llu(%llu), %p\n",
__FUNCTION__, (unsigned long long)ino, bno, ip);
#endif
- if ((error = libxfs_iread(mp, tp, ino, ip, bno))) {
+ ip->i_ino = ino;
+ ip->i_mount = mp;
+ error = xfs_iread(mp, tp, ip, bno);
+ if (error) {
cache_node_purge(libxfs_icache, &ino,
(struct cache_node *)ip);
ip = NULL;
diff --git a/libxfs/trans.c b/libxfs/trans.c
index a745d51..7cb3c8c 100644
--- a/libxfs/trans.c
+++ b/libxfs/trans.c
@@ -252,7 +252,7 @@ xfs_trans_log_inode(
* this coordination mechanism.
*/
flags |= ip->i_itemp->ili_last_fields;
- ip->i_itemp->ili_format.ilf_fields |= flags;
+ ip->i_itemp->ili_fields |= flags;
}
/*
@@ -338,7 +338,7 @@ libxfs_trans_binval(
if (bip->bli_flags & XFS_BLI_STALE)
return;
XFS_BUF_UNDELAYWRITE(bp);
- XFS_BUF_STALE(bp);
+ xfs_buf_stale(bp);
bip->bli_flags |= XFS_BLI_STALE;
bip->bli_flags &= ~XFS_BLI_DIRTY;
bip->bli_format.blf_flags &= ~XFS_BLF_INODE_BUF;
@@ -383,22 +383,20 @@ libxfs_trans_bhold(
}
xfs_buf_t *
-libxfs_trans_get_buf(
+libxfs_trans_get_buf_map(
xfs_trans_t *tp,
dev_t dev,
- xfs_daddr_t d,
- int len,
+ struct xfs_buf_map *map,
+ int nmaps,
uint f)
{
xfs_buf_t *bp;
xfs_buf_log_item_t *bip;
- xfs_buftarg_t bdev;
if (tp == NULL)
- return libxfs_getbuf(dev, d, len);
+ return libxfs_getbuf_map(dev, map, nmaps);
- bdev.dev = dev;
- bp = xfs_trans_buf_item_match(tp, &bdev, d, len);
+ bp = xfs_trans_buf_item_match(tp, dev, map, nmaps);
if (bp != NULL) {
ASSERT(XFS_BUF_FSPRIVATE2(bp, xfs_trans_t *) == tp);
bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *);
@@ -407,7 +405,7 @@ libxfs_trans_get_buf(
return bp;
}
- bp = libxfs_getbuf(dev, d, len);
+ bp = libxfs_getbuf_map(dev, map, nmaps);
if (bp == NULL)
return NULL;
#ifdef XACT_DEBUG
@@ -432,15 +430,13 @@ libxfs_trans_getsb(
{
xfs_buf_t *bp;
xfs_buf_log_item_t *bip;
- xfs_buftarg_t bdev;
- int len;
+ int len = XFS_FSS_TO_BB(mp, 1);
+ DEFINE_SINGLE_BUF_MAP(map, XFS_SB_DADDR, len);
if (tp == NULL)
return libxfs_getsb(mp, flags);
- bdev.dev = mp->m_dev;
- len = XFS_FSS_TO_BB(mp, 1);
- bp = xfs_trans_buf_item_match(tp, &bdev, XFS_SB_DADDR, len);
+ bp = xfs_trans_buf_item_match(tp, mp->m_dev, &map, 1);
if (bp != NULL) {
ASSERT(XFS_BUF_FSPRIVATE2(bp, xfs_trans_t *) == tp);
bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *);
@@ -465,24 +461,24 @@ libxfs_trans_getsb(
}
int
-libxfs_trans_read_buf(
+libxfs_trans_read_buf_map(
xfs_mount_t *mp,
xfs_trans_t *tp,
dev_t dev,
- xfs_daddr_t blkno,
- int len,
+ struct xfs_buf_map *map,
+ int nmaps,
uint flags,
- xfs_buf_t **bpp)
+ xfs_buf_t **bpp,
+ const struct xfs_buf_ops *ops)
{
xfs_buf_t *bp;
xfs_buf_log_item_t *bip;
- xfs_buftarg_t bdev;
int error;
*bpp = NULL;
if (tp == NULL) {
- bp = libxfs_readbuf(dev, blkno, len, flags);
+ bp = libxfs_readbuf_map(dev, map, nmaps, flags);
if (!bp) {
return (flags & XBF_TRYLOCK) ?
EAGAIN : XFS_ERROR(ENOMEM);
@@ -492,8 +488,7 @@ libxfs_trans_read_buf(
goto done;
}
- bdev.dev = dev;
- bp = xfs_trans_buf_item_match(tp, &bdev, blkno, len);
+ bp = xfs_trans_buf_item_match(tp, dev, map, nmaps);
if (bp != NULL) {
ASSERT(XFS_BUF_FSPRIVATE2(bp, xfs_trans_t *) == tp);
ASSERT(XFS_BUF_FSPRIVATE(bp, void *) != NULL);
@@ -502,7 +497,7 @@ libxfs_trans_read_buf(
goto done;
}
- bp = libxfs_readbuf(dev, blkno, len, flags);
+ bp = libxfs_readbuf_map(dev, map, nmaps, flags);
if (!bp) {
return (flags & XBF_TRYLOCK) ?
EAGAIN : XFS_ERROR(ENOMEM);
@@ -588,7 +583,7 @@ inode_item_done(
mp = iip->ili_item.li_mountp;
ASSERT(ip != NULL);
- if (!(iip->ili_format.ilf_fields & XFS_ILOG_ALL)) {
+ if (!(iip->ili_fields & XFS_ILOG_ALL)) {
ip->i_transp = NULL; /* disassociate from transaction */
iip->ili_flags = 0; /* reset all flags */
goto ili_done;
@@ -597,9 +592,9 @@ inode_item_done(
/*
* Get the buffer containing the on-disk inode.
*/
- error = xfs_itobp(mp, NULL, ip, &dip, &bp, 0);
+ error = xfs_imap_to_bp(mp, NULL, &ip->i_imap, &dip, &bp, 0, 0);
if (error) {
- fprintf(stderr, _("%s: warning - itobp failed (%d)\n"),
+ fprintf(stderr, _("%s: warning - imap_to_bp failed (%d)\n"),
progname, error);
goto ili_done;
}
@@ -674,6 +669,7 @@ trans_committed(
struct xfs_log_item *lip = lidp->lid_item;
xfs_trans_del_item(lip);
+
if (lip->li_type == XFS_LI_BUF)
buf_item_done((xfs_buf_log_item_t *)lip);
else if (lip->li_type == XFS_LI_INODE)
diff --git a/libxfs/util.c b/libxfs/util.c
index bffbac0..2ad4bfd 100644
--- a/libxfs/util.c
+++ b/libxfs/util.c
@@ -50,130 +50,6 @@ libxfs_trans_ichgtime(
}
/*
- * Given a mount structure and an inode number, return a pointer
- * to a newly allocated in-core inode coresponding to the given
- * inode number.
- *
- * Initialize the inode's attributes and extent pointers if it
- * already has them (it will not if the inode has no links).
- *
- * NOTE: this has slightly different behaviour to the kernel in
- * that this version requires the already allocated *ip being
- * passed in while the kernel version does the allocation and
- * returns it in **ip.
- */
-int
-libxfs_iread(
- xfs_mount_t *mp,
- xfs_trans_t *tp,
- xfs_ino_t ino,
- xfs_inode_t *ip,
- xfs_daddr_t bno)
-{
- xfs_buf_t *bp;
- xfs_dinode_t *dip;
- int error;
-
- ip->i_ino = ino;
- ip->i_mount = mp;
-
- /*
- * Fill in the location information in the in-core inode.
- */
- error = xfs_imap(mp, tp, ip->i_ino, &ip->i_imap, 0);
- if (error)
- return error;
-
- /*
- * Get pointers to the on-disk inode and the buffer containing it.
- */
- error = xfs_imap_to_bp(mp, tp, &ip->i_imap, &bp, XBF_LOCK, 0);
- if (error)
- return error;
- dip = (xfs_dinode_t *)xfs_buf_offset(bp, ip->i_imap.im_boffset);
-
- /*
- * If we got something that isn't an inode it means someone
- * (nfs or dmi) has a stale handle.
- */
- if (be16_to_cpu(dip->di_magic) != XFS_DINODE_MAGIC) {
- xfs_trans_brelse(tp, bp);
- return EINVAL;
- }
-
- /*
- * If the on-disk inode is already linked to a directory
- * entry, copy all of the inode into the in-core inode.
- * xfs_iformat() handles copying in the inode format
- * specific information.
- * Otherwise, just get the truly permanent information.
- */
- if (dip->di_mode) {
- xfs_dinode_from_disk(&ip->i_d, dip);
- error = xfs_iformat(ip, dip);
- if (error) {
- xfs_trans_brelse(tp, bp);
- return error;
- }
- } else {
- ip->i_d.di_magic = be16_to_cpu(dip->di_magic);
- ip->i_d.di_version = dip->di_version;
- ip->i_d.di_gen = be32_to_cpu(dip->di_gen);
- ip->i_d.di_flushiter = be16_to_cpu(dip->di_flushiter);
- /*
- * Make sure to pull in the mode here as well in
- * case the inode is released without being used.
- * This ensures that xfs_inactive() will see that
- * the inode is already free and not try to mess
- * with the uninitialized part of it.
- */
- ip->i_d.di_mode = 0;
- /*
- * Initialize the per-fork minima and maxima for a new
- * inode here. xfs_iformat will do it for old inodes.
- */
- ip->i_df.if_ext_max =
- XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t);
- }
-
- /*
- * The inode format changed when we moved the link count and
- * made it 32 bits long. If this is an old format inode,
- * convert it in memory to look like a new one. If it gets
- * flushed to disk we will convert back before flushing or
- * logging it. We zero out the new projid_lo/hi field and the old link
- * count field. We'll handle clearing the pad field (the remains
- * of the old uuid field) when we actually convert the inode to
- * the new format. We don't change the version number so that we
- * can distinguish this from a real new format inode.
- */
- if (ip->i_d.di_version == 1) {
- ip->i_d.di_nlink = ip->i_d.di_onlink;
- ip->i_d.di_onlink = 0;
- xfs_set_projid(&ip->i_d, 0);
- }
-
- ip->i_delayed_blks = 0;
- ip->i_size = ip->i_d.di_size;
-
- /*
- * Use xfs_trans_brelse() to release the buffer containing the
- * on-disk inode, because it was acquired with xfs_trans_read_buf()
- * in xfs_itobp() above. If tp is NULL, this is just a normal
- * brelse(). If we're within a transaction, then xfs_trans_brelse()
- * will only release the buffer if it is not dirty within the
- * transaction. It will be OK to release the buffer in this case,
- * because inodes on disk are never destroyed and we will be
- * locking the new in-core inode before putting it in the hash
- * table where other processes can find it. Thus we don't have
- * to worry about the inode being changed just because we released
- * the buffer.
- */
- xfs_trans_brelse(tp, bp);
- return 0;
-}
-
-/*
* Allocate an inode on disk and return a copy of its in-core version.
* Set mode, nlink, and rdev appropriately within the inode.
* The uid and gid for the inode are set according to the contents of
@@ -193,7 +69,6 @@ libxfs_ialloc(
struct fsxattr *fsx,
int okalloc,
xfs_buf_t **ialloc_context,
- boolean_t *call_again,
xfs_inode_t **ipp)
{
xfs_ino_t ino;
@@ -206,10 +81,10 @@ libxfs_ialloc(
* the on-disk inode to be allocated.
*/
error = xfs_dialloc(tp, pip ? pip->i_ino : 0, mode, okalloc,
- ialloc_context, call_again, &ino);
+ ialloc_context, &ino);
if (error != 0)
return error;
- if (*call_again || ino == NULLFSINO) {
+ if (*ialloc_context || ino == NULLFSINO) {
*ipp = NULL;
return 0;
}
@@ -455,7 +330,7 @@ libxfs_iflush_int(xfs_inode_t *ip, xfs_buf_t *bp)
dip->di_onlink = 0;
memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad));
memset(&(dip->di_pad[0]), 0, sizeof(dip->di_pad));
- ASSERT(xfs_get_projid(ip->i_d) == 0);
+ ASSERT(xfs_get_projid(&ip->i_d) == 0);
}
}
@@ -560,7 +435,7 @@ libxfs_alloc_file_space(
error = 0;
imapp = &imaps[0];
reccount = 1;
- xfs_bmapi_flags = XFS_BMAPI_WRITE | (alloc_type ? XFS_BMAPI_PREALLOC : 0);
+ xfs_bmapi_flags = alloc_type ? XFS_BMAPI_PREALLOC : 0;
mp = ip->i_mount;
startoffset_fsb = XFS_B_TO_FSBT(mp, offset);
allocatesize_fsb = XFS_B_TO_FSB(mp, count);
@@ -578,7 +453,7 @@ libxfs_alloc_file_space(
xfs_trans_ihold(tp, ip);
xfs_bmap_init(&free_list, &firstfsb);
- error = xfs_bmapi(tp, ip, startoffset_fsb, allocatesize_fsb,
+ error = xfs_bmapi_write(tp, ip, startoffset_fsb, allocatesize_fsb,
xfs_bmapi_flags, &firstfsb, 0, imapp,
&reccount, &free_list);
@@ -617,56 +492,6 @@ libxfs_log2_roundup(unsigned int i)
}
/*
- * Get a buffer for the dir/attr block, fill in the contents.
- * Don't check magic number, the caller will (it's xfs_repair).
- *
- * Originally from xfs_da_btree.c in the kernel, but only used
- * in userspace so it now resides here.
- */
-int
-libxfs_da_read_bufr(
- xfs_trans_t *trans,
- xfs_inode_t *dp,
- xfs_dablk_t bno,
- xfs_daddr_t mappedbno,
- xfs_dabuf_t **bpp,
- int whichfork)
-{
- return xfs_da_do_buf(trans, dp, bno, &mappedbno, bpp, whichfork, 2,
- (inst_t *)__return_address);
-}
-
-/*
- * Hold dabuf at transaction commit.
- *
- * Originally from xfs_da_btree.c in the kernel, but only used
- * in userspace so it now resides here.
- */
-void
-libxfs_da_bhold(xfs_trans_t *tp, xfs_dabuf_t *dabuf)
-{
- int i;
-
- for (i = 0; i < dabuf->nbuf; i++)
- xfs_trans_bhold(tp, dabuf->bps[i]);
-}
-
-/*
- * Join dabuf to transaction.
- *
- * Originally from xfs_da_btree.c in the kernel, but only used
- * in userspace so it now resides here.
- */
-void
-libxfs_da_bjoin(xfs_trans_t *tp, xfs_dabuf_t *dabuf)
-{
- int i;
-
- for (i = 0; i < dabuf->nbuf; i++)
- xfs_trans_bjoin(tp, dabuf->bps[i]);
-}
-
-/*
* Wrapper around call to libxfs_ialloc. Takes care of committing and
* allocating a new transaction as needed.
*
@@ -684,21 +509,25 @@ libxfs_inode_alloc(
struct fsxattr *fsx,
xfs_inode_t **ipp)
{
- boolean_t call_again;
int i;
xfs_buf_t *ialloc_context;
xfs_inode_t *ip;
xfs_trans_t *ntp;
int error;
- call_again = B_FALSE;
ialloc_context = (xfs_buf_t *)0;
error = libxfs_ialloc(*tp, pip, mode, nlink, rdev, cr, fsx,
- 1, &ialloc_context, &call_again, &ip);
- if (error)
+ 1, &ialloc_context, &ip);
+ if (error) {
+ *ipp = NULL;
return error;
+ }
+ if (!ialloc_context && !ip) {
+ *ipp = NULL;
+ return XFS_ERROR(ENOSPC);
+ }
- if (call_again) {
+ if (ialloc_context) {
xfs_trans_bhold(*tp, ialloc_context);
ntp = xfs_trans_dup(*tp);
xfs_trans_commit(*tp, 0);
@@ -710,8 +539,7 @@ libxfs_inode_alloc(
}
xfs_trans_bjoin(*tp, ialloc_context);
error = libxfs_ialloc(*tp, pip, mode, nlink, rdev, cr,
- fsx, 1, &ialloc_context,
- &call_again, &ip);
+ fsx, 1, &ialloc_context, &ip);
if (!ip)
error = ENOSPC;
if (error)
diff --git a/libxfs/xfs.h b/libxfs/xfs.h
index b3f8378..6213284 100644
--- a/libxfs/xfs.h
+++ b/libxfs/xfs.h
@@ -43,14 +43,21 @@
*/
#include <xfs/libxfs.h>
+#include "xfs_dir2_priv.h"
-typedef struct { dev_t dev; } xfs_buftarg_t;
+#undef ASSERT
+#define ASSERT(ex) assert(ex)
-typedef __uint32_t uint_t;
+typedef __uint32_t uint_t;
typedef __uint32_t inst_t; /* an instruction */
+#ifndef EWRONGFS
+#define EWRONGFS EINVAL
+#endif
-#define m_ddev_targp m_dev
+#define m_ddev_targp m_dev
+#define m_logdev_targp m_logdev
+#define m_rtdev_targp m_rtdev
#define xfs_error_level 0
#define STATIC static
@@ -64,10 +71,11 @@ typedef __uint32_t inst_t; /* an instruction */
#define IHOLD(ip) ((void) 0)
-#define XFS_CORRUPTION_ERROR(e,l,mp,m) ((void) 0)
+/* stop unused var warnings by assigning mp to itself */
+#define XFS_CORRUPTION_ERROR(e,l,mp,m) do { (mp) = (mp); } while (0)
+#define XFS_ERROR_REPORT(e,l,mp) do { (mp) = (mp); } while (0)
#define XFS_QM_DQATTACH(mp,ip,flags) 0
#define XFS_ERROR(e) (e)
-#define XFS_ERROR_REPORT(e,l,mp) ((void) 0)
#define XFS_ERRLEVEL_LOW 1
#define XFS_FORCED_SHUTDOWN(mp) 0
#define XFS_ILOCK_EXCL 0
@@ -124,35 +132,6 @@ static inline int __do_div(unsigned long long *n, unsigned base)
({ type __x = (x); type __y = (y); __x > __y ? __x: __y; })
-static inline __uint32_t __get_unaligned_be32(const __uint8_t *p)
-{
- return p[0] << 24 | p[1] << 16 | p[2] << 8 | p[3];
-}
-
-static inline __uint64_t get_unaligned_be64(void *p)
-{
- return (__uint64_t)__get_unaligned_be32(p) << 32 |
- __get_unaligned_be32(p + 4);
-}
-
-static inline void __put_unaligned_be16(__uint16_t val, __uint8_t *p)
-{
- *p++ = val >> 8;
- *p++ = val;
-}
-
-static inline void __put_unaligned_be32(__uint32_t val, __uint8_t *p)
-{
- __put_unaligned_be16(val >> 16, p);
- __put_unaligned_be16(val, p + 2);
-}
-
-static inline void put_unaligned_be64(__uint64_t val, void *p)
-{
- __put_unaligned_be32(val >> 32, p);
- __put_unaligned_be32(val, p + 4);
-}
-
static inline __attribute__((const))
int is_power_of_2(unsigned long n)
@@ -191,10 +170,11 @@ roundup_pow_of_two(uint v)
#define XBF_LOCK XFS_BUF_LOCK
#define XBF_TRYLOCK XFS_BUF_TRYLOCK
#define XBF_DONT_BLOCK 0
+#define XBF_UNMAPPED 0
#define XFS_BUF_GETERROR(bp) 0
#define XFS_BUF_DONE(bp) ((bp)->b_flags |= LIBXFS_B_UPTODATE)
#define XFS_BUF_ISDONE(bp) ((bp)->b_flags & LIBXFS_B_UPTODATE)
-#define XFS_BUF_STALE(bp) ((bp)->b_flags |= LIBXFS_B_STALE)
+#define xfs_buf_stale(bp) ((bp)->b_flags |= LIBXFS_B_STALE)
#define XFS_BUF_UNDELAYWRITE(bp) ((bp)->b_flags &= ~LIBXFS_B_DIRTY)
#define XFS_BUF_SET_VTYPE(a,b) ((void) 0)
#define XFS_BUF_SET_VTYPE_REF(a,b,c) ((void) 0)
@@ -260,6 +240,8 @@ roundup_pow_of_two(uint v)
#define xfs_trans_log_inode libxfs_trans_log_inode
#define xfs_trans_mod_sb libxfs_trans_mod_sb
#define xfs_trans_read_buf libxfs_trans_read_buf
+#define xfs_trans_read_buf_map libxfs_trans_read_buf_map
+#define xfs_trans_get_buf_map libxfs_trans_get_buf_map
#define xfs_trans_reserve libxfs_trans_reserve
#define xfs_trans_get_block_res(tp) 1
@@ -268,12 +250,16 @@ roundup_pow_of_two(uint v)
#define xfs_trans_agflist_delta(tp, d)
#define xfs_trans_agbtree_delta(tp, d)
-#define xfs_buf_readahead(a,b,c) ((void) 0) /* no readahead */
-#define xfs_btree_reada_bufl(m,fsb,c) ((void) 0)
-#define xfs_btree_reada_bufs(m,fsb,c,x) ((void) 0)
-#define xfs_buftrace(x,y) ((void) 0) /* debug only */
+#define xfs_buf_readahead(a,b,c,ops) ((void) 0) /* no readahead */
+#define xfs_buf_readahead_map(a,b,c,ops) ((void) 0) /* no readahead */
+#define xfs_btree_reada_bufl(m,fsb,c,ops) ((void) 0)
+#define xfs_btree_reada_bufs(m,fsb,c,x,ops) ((void) 0)
+#define xfs_buftrace(x,y) ((void) 0) /* debug only */
#define xfs_cmn_err(tag,level,mp,fmt,args...) cmn_err(level,fmt, ## args)
+#define xfs_warn(mp,fmt,args...) cmn_err(CE_WARN,fmt, ## args)
+#define xfs_alert(mp,fmt,args...) cmn_err(CE_ALERT,fmt, ## args)
+#define xfs_alert_tag(mp,tag,fmt,args...) cmn_err(CE_ALERT,fmt, ## args)
#define xfs_dir2_trace_args(where, args) ((void) 0)
#define xfs_dir2_trace_args_b(where, args, bp) ((void) 0)
@@ -289,15 +275,28 @@ roundup_pow_of_two(uint v)
#define xfs_initialize_perag_icache(pag) ((void) 0)
#define xfs_ilock(ip,mode) ((void) 0)
+#define xfs_ilock_nowait(ip,mode) ((void) 0)
+#define xfs_ilock_demote(ip,mode) ((void) 0)
#define xfs_iunlock(ip,mode) ((void) 0)
+#define xfs_ilock_map_shared(ip,mode) ((void) 0)
+#define xfs_iunlock_map_shared(ip,mode) ((void) 0)
+#define __xfs_flock(ip) ((void) 0)
/* space allocation */
-#define xfs_alloc_busy_search(tp,ag,b,len) 0
+#define xfs_extent_busy_reuse(mp,ag,bno,len,user) ((void) 0)
+#define xfs_extent_busy_insert(tp,ag,bno,len,flags) ((void) 0)
+#define xfs_extent_busy_trim(args,fbno,flen,bno,len) \
+do { \
+ *(bno) = (fbno); \
+ *(len) = (flen); \
+} while (0)
+
/* avoid unused variable warning */
#define xfs_alloc_busy_insert(tp,ag,b,len) ({ \
xfs_agnumber_t __foo = ag; \
__foo = 0; \
})
+
#define xfs_rotorstep 1
#define xfs_bmap_rtalloc(a) (ENOSYS)
#define xfs_rtpick_extent(mp,tp,len,p) (ENOSYS)
@@ -306,6 +305,15 @@ roundup_pow_of_two(uint v)
#define xfs_filestream_lookup_ag(ip) (0)
#define xfs_filestream_new_ag(ip,ag) (0)
+#define xfs_log_force(mp,flags) ((void) 0)
+#define XFS_LOG_SYNC 1
+
+/* quota bits */
+#define xfs_trans_mod_dquot_byino(t,i,f,d) ((void) 0)
+#define xfs_trans_reserve_quota_nblks(t,i,b,n,f) (0)
+#define xfs_trans_unreserve_quota_nblks(t,i,b,n,f) ((void) 0)
+#define xfs_qm_dqattach(i,f) (0)
+
/*
* Prototypes for kernel static functions that are aren't in their
* associated header files
@@ -318,14 +326,20 @@ int xfs_attr_rmtval_get(struct xfs_da_args *);
void xfs_bmap_del_free(xfs_bmap_free_t *, xfs_bmap_free_item_t *,
xfs_bmap_free_item_t *);
-/* xfs_da_btree.c */
-int xfs_da_do_buf(xfs_trans_t *, xfs_inode_t *, xfs_dablk_t, xfs_daddr_t *,
- xfs_dabuf_t **, int, int, inst_t *);
-
/* xfs_inode.c */
void xfs_iflush_fork(xfs_inode_t *, xfs_dinode_t *, xfs_inode_log_item_t *,
int, xfs_buf_t *);
-int xfs_iformat(xfs_inode_t *, xfs_dinode_t *);
+/*
+ * For regular files we only update the on-disk filesize when actually
+ * writing data back to disk. Until then only the copy in the VFS inode
+ * is uptodate.
+ */
+static inline xfs_fsize_t XFS_ISIZE(struct xfs_inode *ip)
+{
+ if (S_ISREG(ip->i_d.di_mode))
+ return ip->i_size;
+ return ip->i_d.di_size;
+}
/* xfs_mount.c */
int xfs_initialize_perag_data(xfs_mount_t *, xfs_agnumber_t);
@@ -348,8 +362,8 @@ void xfs_buf_item_init (xfs_buf_t *, xfs_mount_t *);
void xfs_buf_item_log (xfs_buf_log_item_t *, uint, uint);
/* xfs_trans_buf.c */
-xfs_buf_t *xfs_trans_buf_item_match (xfs_trans_t *, xfs_buftarg_t *,
- xfs_daddr_t, int);
+xfs_buf_t *xfs_trans_buf_item_match(xfs_trans_t *, dev_t,
+ struct xfs_buf_map *, int);
/* local source files */
int xfs_mod_incore_sb(xfs_mount_t *, xfs_sb_field_t, int64_t, int);
diff --git a/libxfs/xfs_alloc.c b/libxfs/xfs_alloc.c
index a76512d..61cdc6c 100644
--- a/libxfs/xfs_alloc.c
+++ b/libxfs/xfs_alloc.c
@@ -22,19 +22,11 @@
#define XFSA_FIXUP_BNO_OK 1
#define XFSA_FIXUP_CNT_OK 2
-/*
- * Prototypes for per-ag allocation routines
- */
-
STATIC int xfs_alloc_ag_vextent_exact(xfs_alloc_arg_t *);
STATIC int xfs_alloc_ag_vextent_near(xfs_alloc_arg_t *);
STATIC int xfs_alloc_ag_vextent_size(xfs_alloc_arg_t *);
STATIC int xfs_alloc_ag_vextent_small(xfs_alloc_arg_t *,
- xfs_btree_cur_t *, xfs_agblock_t *, xfs_extlen_t *, int *);
-
-/*
- * Internal functions.
- */
+ xfs_btree_cur_t *, xfs_agblock_t *, xfs_extlen_t *, int *);
/*
* Lookup the record equal to [bno, len] in the btree given by cur.
@@ -55,7 +47,7 @@ xfs_alloc_lookup_eq(
* Lookup the first record greater than or equal to [bno, len]
* in the btree given by cur.
*/
-STATIC int /* error */
+int /* error */
xfs_alloc_lookup_ge(
struct xfs_btree_cur *cur, /* btree cursor */
xfs_agblock_t bno, /* starting block of extent */
@@ -71,7 +63,7 @@ xfs_alloc_lookup_ge(
* Lookup the first record less than or equal to [bno, len]
* in the btree given by cur.
*/
-STATIC int /* error */
+int /* error */
xfs_alloc_lookup_le(
struct xfs_btree_cur *cur, /* btree cursor */
xfs_agblock_t bno, /* starting block of extent */
@@ -104,7 +96,7 @@ xfs_alloc_update(
/*
* Get the data from the pointed-to record.
*/
-STATIC int /* error */
+int /* error */
xfs_alloc_get_rec(
struct xfs_btree_cur *cur, /* btree cursor */
xfs_agblock_t *bno, /* output: starting block of extent */
@@ -128,27 +120,28 @@ xfs_alloc_get_rec(
*/
STATIC void
xfs_alloc_compute_aligned(
+ xfs_alloc_arg_t *args, /* allocation argument structure */
xfs_agblock_t foundbno, /* starting block in found extent */
xfs_extlen_t foundlen, /* length in found extent */
- xfs_extlen_t alignment, /* alignment for allocation */
- xfs_extlen_t minlen, /* minimum length for allocation */
xfs_agblock_t *resbno, /* result block number */
xfs_extlen_t *reslen) /* result length */
{
xfs_agblock_t bno;
- xfs_extlen_t diff;
xfs_extlen_t len;
- if (alignment > 1 && foundlen >= minlen) {
- bno = roundup(foundbno, alignment);
- diff = bno - foundbno;
- len = diff >= foundlen ? 0 : foundlen - diff;
+ /* Trim busy sections out of found extent */
+ xfs_extent_busy_trim(args, foundbno, foundlen, &bno, &len);
+
+ if (args->alignment > 1 && len >= args->minlen) {
+ xfs_agblock_t aligned_bno = roundup(bno, args->alignment);
+ xfs_extlen_t diff = aligned_bno - bno;
+
+ *resbno = aligned_bno;
+ *reslen = diff >= len ? 0 : len - diff;
} else {
- bno = foundbno;
- len = foundlen;
+ *resbno = bno;
+ *reslen = len;
}
- *resbno = bno;
- *reslen = len;
}
/*
@@ -262,7 +255,6 @@ xfs_alloc_fix_minleft(
return 1;
agf = XFS_BUF_TO_AGF(args->agbp);
diff = be32_to_cpu(agf->agf_freeblks)
- + be32_to_cpu(agf->agf_flcount)
- args->len - args->minleft;
if (diff >= 0)
return 1;
@@ -418,6 +410,60 @@ xfs_alloc_fixup_trees(
return 0;
}
+static void
+xfs_agfl_verify(
+ struct xfs_buf *bp)
+{
+#ifdef WHEN_CRCS_COME_ALONG
+ /*
+ * we cannot actually do any verification of the AGFL because mkfs does
+ * not initialise the AGFL to zero or NULL. Hence the only valid part of
+ * the AGFL is what the AGF says is active. We can't get to the AGF, so
+ * we can't verify just those entries are valid.
+ *
+ * This problem goes away when the CRC format change comes along as that
+ * requires the AGFL to be initialised by mkfs. At that point, we can
+ * verify the blocks in the agfl -active or not- lie within the bounds
+ * of the AG. Until then, just leave this check ifdef'd out.
+ */
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_agfl *agfl = XFS_BUF_TO_AGFL(bp);
+ int agfl_ok = 1;
+
+ int i;
+
+ for (i = 0; i < XFS_AGFL_SIZE(mp); i++) {
+ if (be32_to_cpu(agfl->agfl_bno[i]) == NULLAGBLOCK ||
+ be32_to_cpu(agfl->agfl_bno[i]) >= mp->m_sb.sb_agblocks)
+ agfl_ok = 0;
+ }
+
+ if (!agfl_ok) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, agfl);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+#endif
+}
+
+static void
+xfs_agfl_write_verify(
+ struct xfs_buf *bp)
+{
+ xfs_agfl_verify(bp);
+}
+
+static void
+xfs_agfl_read_verify(
+ struct xfs_buf *bp)
+{
+ xfs_agfl_verify(bp);
+}
+
+const struct xfs_buf_ops xfs_agfl_buf_ops = {
+ .verify_read = xfs_agfl_read_verify,
+ .verify_write = xfs_agfl_write_verify,
+};
+
/*
* Read in the allocation group free block array.
*/
@@ -435,16 +481,36 @@ xfs_alloc_read_agfl(
error = xfs_trans_read_buf(
mp, tp, mp->m_ddev_targp,
XFS_AG_DADDR(mp, agno, XFS_AGFL_DADDR(mp)),
- XFS_FSS_TO_BB(mp, 1), 0, &bp);
+ XFS_FSS_TO_BB(mp, 1), 0, &bp, &xfs_agfl_buf_ops);
if (error)
return error;
- ASSERT(bp);
- ASSERT(!XFS_BUF_GETERROR(bp));
- XFS_BUF_SET_VTYPE_REF(bp, B_FS_AGFL, XFS_AGFL_REF);
+ ASSERT(!xfs_buf_geterror(bp));
+ xfs_buf_set_ref(bp, XFS_AGFL_REF);
*bpp = bp;
return 0;
}
+STATIC int
+xfs_alloc_update_counters(
+ struct xfs_trans *tp,
+ struct xfs_perag *pag,
+ struct xfs_buf *agbp,
+ long len)
+{
+ struct xfs_agf *agf = XFS_BUF_TO_AGF(agbp);
+
+ pag->pagf_freeblks += len;
+ be32_add_cpu(&agf->agf_freeblks, len);
+
+ xfs_trans_agblocks_delta(tp, len);
+ if (unlikely(be32_to_cpu(agf->agf_freeblks) >
+ be32_to_cpu(agf->agf_length)))
+ return EFSCORRUPTED;
+
+ xfs_alloc_log_agf(tp, agbp, XFS_AGF_FREEBLKS);
+ return 0;
+}
+
/*
* Allocation group level functions.
*/
@@ -486,49 +552,36 @@ xfs_alloc_ag_vextent(
ASSERT(0);
/* NOTREACHED */
}
- if (error)
+
+ if (error || args->agbno == NULLAGBLOCK)
return error;
- /*
- * If the allocation worked, need to change the agf structure
- * (and log it), and the superblock.
- */
- if (args->agbno != NULLAGBLOCK) {
- xfs_agf_t *agf; /* allocation group freelist header */
- long slen = (long)args->len;
- ASSERT(args->len >= args->minlen && args->len <= args->maxlen);
- ASSERT(!(args->wasfromfl) || !args->isfl);
- ASSERT(args->agbno % args->alignment == 0);
- if (!(args->wasfromfl)) {
-
- agf = XFS_BUF_TO_AGF(args->agbp);
- be32_add_cpu(&agf->agf_freeblks, -(args->len));
- xfs_trans_agblocks_delta(args->tp,
- -((long)(args->len)));
- args->pag->pagf_freeblks -= args->len;
- ASSERT(be32_to_cpu(agf->agf_freeblks) <=
- be32_to_cpu(agf->agf_length));
- xfs_alloc_log_agf(args->tp, args->agbp,
- XFS_AGF_FREEBLKS);
- /*
- * Search the busylist for these blocks and mark the
- * transaction as synchronous if blocks are found. This
- * avoids the need to block due to a synchronous log
- * force to ensure correct ordering as the synchronous
- * transaction will guarantee that for us.
- */
- if (xfs_alloc_busy_search(args->mp, args->agno,
- args->agbno, args->len))
- xfs_trans_set_sync(args->tp);
- }
- if (!args->isfl)
- xfs_trans_mod_sb(args->tp,
- args->wasdel ? XFS_TRANS_SB_RES_FDBLOCKS :
- XFS_TRANS_SB_FDBLOCKS, -slen);
- XFS_STATS_INC(xs_allocx);
- XFS_STATS_ADD(xs_allocb, args->len);
+ ASSERT(args->len >= args->minlen);
+ ASSERT(args->len <= args->maxlen);
+ ASSERT(!args->wasfromfl || !args->isfl);
+ ASSERT(args->agbno % args->alignment == 0);
+
+ if (!args->wasfromfl) {
+ error = xfs_alloc_update_counters(args->tp, args->pag,
+ args->agbp,
+ -((long)(args->len)));
+ if (error)
+ return error;
+
+ ASSERT(!xfs_extent_busy_search(args->mp, args->agno,
+ args->agbno, args->len));
}
- return 0;
+
+ if (!args->isfl) {
+ xfs_trans_mod_sb(args->tp, args->wasdel ?
+ XFS_TRANS_SB_RES_FDBLOCKS :
+ XFS_TRANS_SB_FDBLOCKS,
+ -((long)(args->len)));
+ }
+
+ XFS_STATS_INC(xs_allocx);
+ XFS_STATS_ADD(xs_allocb, args->len);
+ return error;
}
/*
@@ -543,17 +596,16 @@ xfs_alloc_ag_vextent_exact(
{
xfs_btree_cur_t *bno_cur;/* by block-number btree cursor */
xfs_btree_cur_t *cnt_cur;/* by count btree cursor */
- xfs_agblock_t end; /* end of allocated extent */
int error;
xfs_agblock_t fbno; /* start block of found extent */
- xfs_agblock_t fend; /* end block of found extent */
xfs_extlen_t flen; /* length of found extent */
+ xfs_agblock_t tbno; /* start block of trimmed extent */
+ xfs_extlen_t tlen; /* length of trimmed extent */
+ xfs_agblock_t tend; /* end block of trimmed extent */
int i; /* success/failure of operation */
- xfs_agblock_t maxend; /* end of maximal extent */
- xfs_agblock_t minend; /* end of minimal extent */
- xfs_extlen_t rlen; /* length of returned extent */
ASSERT(args->alignment == 1);
+
/*
* Allocate/initialize a cursor for the by-number freespace btree.
*/
@@ -579,14 +631,22 @@ xfs_alloc_ag_vextent_exact(
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
ASSERT(fbno <= args->agbno);
- minend = args->agbno + args->minlen;
- maxend = args->agbno + args->maxlen;
- fend = fbno + flen;
/*
- * Give up if the freespace isn't long enough for the minimum request.
+ * Check for overlapping busy extents.
+ */
+ xfs_extent_busy_trim(args, fbno, flen, &tbno, &tlen);
+
+ /*
+ * Give up if the start of the extent is busy, or the freespace isn't
+ * long enough for the minimum request.
*/
- if (fend < minend)
+ if (tbno > args->agbno)
+ goto not_found;
+ if (tlen < args->minlen)
+ goto not_found;
+ tend = tbno + tlen;
+ if (tend < args->agbno + args->minlen)
goto not_found;
/*
@@ -595,18 +655,16 @@ xfs_alloc_ag_vextent_exact(
*
* Fix the length according to mod and prod if given.
*/
- end = XFS_AGBLOCK_MIN(fend, maxend);
- args->len = end - args->agbno;
+ args->len = XFS_AGBLOCK_MIN(tend, args->agbno + args->maxlen)
+ - args->agbno;
xfs_alloc_fix_len(args);
if (!xfs_alloc_fix_minleft(args))
goto not_found;
- rlen = args->len;
- ASSERT(args->agbno + rlen <= fend);
- end = args->agbno + rlen;
+ ASSERT(args->agbno + args->len <= tend);
/*
- * We are allocating agbno for rlen [agbno .. end]
+ * We are allocating agbno for args->len
* Allocate/initialize a cursor for the by-size btree.
*/
cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
@@ -619,8 +677,10 @@ xfs_alloc_ag_vextent_exact(
xfs_btree_del_cursor(cnt_cur, XFS_BTREE_ERROR);
goto error0;
}
+
xfs_btree_del_cursor(bno_cur, XFS_BTREE_NOERROR);
xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
+
args->wasfromfl = 0;
trace_xfs_alloc_exact_done(args);
return 0;
@@ -649,11 +709,11 @@ xfs_alloc_find_best_extent(
struct xfs_btree_cur **scur, /* searching cursor */
xfs_agblock_t gdiff, /* difference for search comparison */
xfs_agblock_t *sbno, /* extent found by search */
- xfs_extlen_t *slen,
- xfs_extlen_t *slena, /* aligned length */
+ xfs_extlen_t *slen, /* extent length */
+ xfs_agblock_t *sbnoa, /* aligned extent found by search */
+ xfs_extlen_t *slena, /* aligned extent length */
int dir) /* 0 = search right, 1 = search left */
{
- xfs_agblock_t bno;
xfs_agblock_t new;
xfs_agblock_t sdiff;
int error;
@@ -671,17 +731,16 @@ xfs_alloc_find_best_extent(
if (error)
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
- xfs_alloc_compute_aligned(*sbno, *slen, args->alignment,
- args->minlen, &bno, slena);
+ xfs_alloc_compute_aligned(args, *sbno, *slen, sbnoa, slena);
/*
* The good extent is closer than this one.
*/
if (!dir) {
- if (bno >= args->agbno + gdiff)
+ if (*sbnoa >= args->agbno + gdiff)
goto out_use_good;
} else {
- if (bno <= args->agbno - gdiff)
+ if (*sbnoa <= args->agbno - gdiff)
goto out_use_good;
}
@@ -693,8 +752,8 @@ xfs_alloc_find_best_extent(
xfs_alloc_fix_len(args);
sdiff = xfs_alloc_compute_diff(args->agbno, args->len,
- args->alignment, *sbno,
- *slen, &new);
+ args->alignment, *sbnoa,
+ *slena, &new);
/*
* Choose closer size and invalidate other cursor.
@@ -744,7 +803,7 @@ xfs_alloc_ag_vextent_near(
xfs_agblock_t gtbnoa; /* aligned ... */
xfs_extlen_t gtdiff; /* difference to right side entry */
xfs_extlen_t gtlen; /* length of right side entry */
- xfs_extlen_t gtlena = 0; /* aligned ... */
+ xfs_extlen_t gtlena; /* aligned ... */
xfs_agblock_t gtnew; /* useful start bno of right side */
int error; /* error code */
int i; /* result code, temporary */
@@ -753,9 +812,10 @@ xfs_alloc_ag_vextent_near(
xfs_agblock_t ltbnoa; /* aligned ... */
xfs_extlen_t ltdiff; /* difference to left side entry */
xfs_extlen_t ltlen; /* length of left side entry */
- xfs_extlen_t ltlena = 0; /* aligned ... */
+ xfs_extlen_t ltlena; /* aligned ... */
xfs_agblock_t ltnew; /* useful start bno of left side */
xfs_extlen_t rlen; /* length of returned extent */
+ int forced = 0;
#if defined(DEBUG) && defined(__KERNEL__)
/*
* Randomly don't execute the first algorithm.
@@ -764,13 +824,20 @@ xfs_alloc_ag_vextent_near(
dofirst = random32() & 1;
#endif
+
+restart:
+ bno_cur_lt = NULL;
+ bno_cur_gt = NULL;
+ ltlen = 0;
+ gtlena = 0;
+ ltlena = 0;
+
/*
* Get a cursor for the by-size btree.
*/
cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
args->agno, XFS_BTNUM_CNT);
- ltlen = 0;
- bno_cur_lt = bno_cur_gt = NULL;
+
/*
* See if there are any free extents as big as maxlen.
*/
@@ -786,11 +853,13 @@ xfs_alloc_ag_vextent_near(
goto error0;
if (i == 0 || ltlen == 0) {
xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
+ trace_xfs_alloc_near_noentry(args);
return 0;
}
ASSERT(i == 1);
}
args->wasfromfl = 0;
+
/*
* First algorithm.
* If the requested extent is large wrt the freespaces available
@@ -844,8 +913,8 @@ xfs_alloc_ag_vextent_near(
if ((error = xfs_alloc_get_rec(cnt_cur, <bno, <len, &i)))
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
- xfs_alloc_compute_aligned(ltbno, ltlen, args->alignment,
- args->minlen, <bnoa, <lena);
+ xfs_alloc_compute_aligned(args, ltbno, ltlen,
+ <bnoa, <lena);
if (ltlena < args->minlen)
continue;
args->len = XFS_EXTLEN_MIN(ltlena, args->maxlen);
@@ -854,7 +923,7 @@ xfs_alloc_ag_vextent_near(
if (args->len < blen)
continue;
ltdiff = xfs_alloc_compute_diff(args->agbno, args->len,
- args->alignment, ltbno, ltlen, <new);
+ args->alignment, ltbnoa, ltlena, <new);
if (ltnew != NULLAGBLOCK &&
(args->len > blen || ltdiff < bdiff)) {
bdiff = ltdiff;
@@ -965,8 +1034,8 @@ xfs_alloc_ag_vextent_near(
if ((error = xfs_alloc_get_rec(bno_cur_lt, <bno, <len, &i)))
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
- xfs_alloc_compute_aligned(ltbno, ltlen, args->alignment,
- args->minlen, <bnoa, <lena);
+ xfs_alloc_compute_aligned(args, ltbno, ltlen,
+ <bnoa, <lena);
if (ltlena >= args->minlen)
break;
if ((error = xfs_btree_decrement(bno_cur_lt, 0, &i)))
@@ -981,8 +1050,8 @@ xfs_alloc_ag_vextent_near(
if ((error = xfs_alloc_get_rec(bno_cur_gt, >bno, >len, &i)))
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
- xfs_alloc_compute_aligned(gtbno, gtlen, args->alignment,
- args->minlen, >bnoa, >lena);
+ xfs_alloc_compute_aligned(args, gtbno, gtlen,
+ >bnoa, >lena);
if (gtlena >= args->minlen)
break;
if ((error = xfs_btree_increment(bno_cur_gt, 0, &i)))
@@ -1005,13 +1074,13 @@ xfs_alloc_ag_vextent_near(
*/
args->len = XFS_EXTLEN_MIN(ltlena, args->maxlen);
xfs_alloc_fix_len(args);
- rlen = args->len;
ltdiff = xfs_alloc_compute_diff(args->agbno, args->len,
- args->alignment, ltbno, ltlen, <new);
+ args->alignment, ltbnoa, ltlena, <new);
error = xfs_alloc_find_best_extent(args,
&bno_cur_lt, &bno_cur_gt,
- ltdiff, >bno, >len, >lena,
+ ltdiff, >bno, >len,
+ >bnoa, >lena,
0 /* search right */);
} else {
ASSERT(gtlena >= args->minlen);
@@ -1022,11 +1091,12 @@ xfs_alloc_ag_vextent_near(
args->len = XFS_EXTLEN_MIN(gtlena, args->maxlen);
xfs_alloc_fix_len(args);
gtdiff = xfs_alloc_compute_diff(args->agbno, args->len,
- args->alignment, gtbno, gtlen, >new);
+ args->alignment, gtbnoa, gtlena, >new);
error = xfs_alloc_find_best_extent(args,
&bno_cur_gt, &bno_cur_lt,
- gtdiff, <bno, <len, <lena,
+ gtdiff, <bno, <len,
+ <bnoa, <lena,
1 /* search left */);
}
@@ -1038,6 +1108,13 @@ xfs_alloc_ag_vextent_near(
* If we couldn't get anything, give up.
*/
if (bno_cur_lt == NULL && bno_cur_gt == NULL) {
+ xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
+
+ if (!forced++) {
+ trace_xfs_alloc_near_busy(args);
+ xfs_log_force(args->mp, XFS_LOG_SYNC);
+ goto restart;
+ }
trace_xfs_alloc_size_neither(args);
args->agbno = NULLAGBLOCK;
return 0;
@@ -1072,12 +1149,13 @@ xfs_alloc_ag_vextent_near(
return 0;
}
rlen = args->len;
- (void)xfs_alloc_compute_diff(args->agbno, rlen, args->alignment, ltbno,
- ltlen, <new);
+ (void)xfs_alloc_compute_diff(args->agbno, rlen, args->alignment,
+ ltbnoa, ltlena, <new);
ASSERT(ltnew >= ltbno);
- ASSERT(ltnew + rlen <= ltbno + ltlen);
+ ASSERT(ltnew + rlen <= ltbnoa + ltlena);
ASSERT(ltnew + rlen <= be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length));
args->agbno = ltnew;
+
if ((error = xfs_alloc_fixup_trees(cnt_cur, bno_cur_lt, ltbno, ltlen,
ltnew, rlen, XFSA_FIXUP_BNO_OK)))
goto error0;
@@ -1120,26 +1198,35 @@ xfs_alloc_ag_vextent_size(
int i; /* temp status variable */
xfs_agblock_t rbno; /* returned block number */
xfs_extlen_t rlen; /* length of returned extent */
+ int forced = 0;
+restart:
/*
* Allocate and initialize a cursor for the by-size btree.
*/
cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
args->agno, XFS_BTNUM_CNT);
bno_cur = NULL;
+
/*
* Look for an entry >= maxlen+alignment-1 blocks.
*/
if ((error = xfs_alloc_lookup_ge(cnt_cur, 0,
args->maxlen + args->alignment - 1, &i)))
goto error0;
+
/*
- * If none, then pick up the last entry in the tree unless the
- * tree is empty.
+ * If none or we have busy extents that we cannot allocate from, then
+ * we have to settle for a smaller extent. In the case that there are
+ * no large extents, this will return the last entry in the tree unless
+ * the tree is empty. In the case that there are only busy large
+ * extents, this will return the largest small extent unless there
+ * are no smaller extents available.
*/
- if (!i) {
- if ((error = xfs_alloc_ag_vextent_small(args, cnt_cur, &fbno,
- &flen, &i)))
+ if (!i || forced > 1) {
+ error = xfs_alloc_ag_vextent_small(args, cnt_cur,
+ &fbno, &flen, &i);
+ if (error)
goto error0;
if (i == 0 || flen == 0) {
xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
@@ -1147,23 +1234,56 @@ xfs_alloc_ag_vextent_size(
return 0;
}
ASSERT(i == 1);
+ xfs_alloc_compute_aligned(args, fbno, flen, &rbno, &rlen);
+ } else {
+ /*
+ * Search for a non-busy extent that is large enough.
+ * If we are at low space, don't check, or if we fall of
+ * the end of the btree, turn off the busy check and
+ * restart.
+ */
+ for (;;) {
+ error = xfs_alloc_get_rec(cnt_cur, &fbno, &flen, &i);
+ if (error)
+ goto error0;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
+
+ xfs_alloc_compute_aligned(args, fbno, flen,
+ &rbno, &rlen);
+
+ if (rlen >= args->maxlen)
+ break;
+
+ error = xfs_btree_increment(cnt_cur, 0, &i);
+ if (error)
+ goto error0;
+ if (i == 0) {
+ /*
+ * Our only valid extents must have been busy.
+ * Make it unbusy by forcing the log out and
+ * retrying. If we've been here before, forcing
+ * the log isn't making the extents available,
+ * which means they have probably been freed in
+ * this transaction. In that case, we have to
+ * give up on them and we'll attempt a minlen
+ * allocation the next time around.
+ */
+ xfs_btree_del_cursor(cnt_cur,
+ XFS_BTREE_NOERROR);
+ trace_xfs_alloc_size_busy(args);
+ if (!forced++)
+ xfs_log_force(args->mp, XFS_LOG_SYNC);
+ goto restart;
+ }
+ }
}
- /*
- * There's a freespace as big as maxlen+alignment-1, get it.
- */
- else {
- if ((error = xfs_alloc_get_rec(cnt_cur, &fbno, &flen, &i)))
- goto error0;
- XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
- }
+
/*
* In the first case above, we got the last entry in the
* by-size btree. Now we check to see if the space hits maxlen
* once aligned; if not, we search left for something better.
* This can't happen in the second case above.
*/
- xfs_alloc_compute_aligned(fbno, flen, args->alignment, args->minlen,
- &rbno, &rlen);
rlen = XFS_EXTLEN_MIN(args->maxlen, rlen);
XFS_WANT_CORRUPTED_GOTO(rlen == 0 ||
(rlen <= flen && rbno + rlen <= fbno + flen), error0);
@@ -1188,8 +1308,8 @@ xfs_alloc_ag_vextent_size(
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
if (flen < bestrlen)
break;
- xfs_alloc_compute_aligned(fbno, flen, args->alignment,
- args->minlen, &rbno, &rlen);
+ xfs_alloc_compute_aligned(args, fbno, flen,
+ &rbno, &rlen);
rlen = XFS_EXTLEN_MIN(args->maxlen, rlen);
XFS_WANT_CORRUPTED_GOTO(rlen == 0 ||
(rlen <= flen && rbno + rlen <= fbno + flen),
@@ -1217,13 +1337,19 @@ xfs_alloc_ag_vextent_size(
* Fix up the length.
*/
args->len = rlen;
- xfs_alloc_fix_len(args);
- if (rlen < args->minlen || !xfs_alloc_fix_minleft(args)) {
- xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
- trace_xfs_alloc_size_nominleft(args);
- args->agbno = NULLAGBLOCK;
- return 0;
+ if (rlen < args->minlen) {
+ if (!forced++) {
+ xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
+ trace_xfs_alloc_size_busy(args);
+ xfs_log_force(args->mp, XFS_LOG_SYNC);
+ goto restart;
+ }
+ goto out_nominleft;
}
+ xfs_alloc_fix_len(args);
+
+ if (!xfs_alloc_fix_minleft(args))
+ goto out_nominleft;
rlen = args->len;
XFS_WANT_CORRUPTED_GOTO(rlen <= flen, error0);
/*
@@ -1253,6 +1379,12 @@ error0:
if (bno_cur)
xfs_btree_del_cursor(bno_cur, XFS_BTREE_ERROR);
return error;
+
+out_nominleft:
+ xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
+ trace_xfs_alloc_size_nominleft(args);
+ args->agbno = NULLAGBLOCK;
+ return 0;
}
/*
@@ -1292,6 +1424,9 @@ xfs_alloc_ag_vextent_small(
if (error)
goto error0;
if (fbno != NULLAGBLOCK) {
+ xfs_extent_busy_reuse(args->mp, args->agno, fbno, 1,
+ args->userdata);
+
if (args->userdata) {
xfs_buf_t *bp;
@@ -1367,6 +1502,7 @@ xfs_free_ag_extent(
xfs_mount_t *mp; /* mount point struct for filesystem */
xfs_agblock_t nbno; /* new starting block of freespace */
xfs_extlen_t nlen; /* new length of freespace */
+ xfs_perag_t *pag; /* per allocation group data */
mp = tp->t_mountp;
/*
@@ -1565,45 +1701,23 @@ xfs_free_ag_extent(
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
cnt_cur = NULL;
+
/*
* Update the freespace totals in the ag and superblock.
*/
- {
- xfs_agf_t *agf;
- xfs_perag_t *pag; /* per allocation group data */
-
- pag = xfs_perag_get(mp, agno);
- pag->pagf_freeblks += len;
- xfs_perag_put(pag);
-
- agf = XFS_BUF_TO_AGF(agbp);
- be32_add_cpu(&agf->agf_freeblks, len);
- xfs_trans_agblocks_delta(tp, len);
- XFS_WANT_CORRUPTED_GOTO(
- be32_to_cpu(agf->agf_freeblks) <=
- be32_to_cpu(agf->agf_length),
- error0);
- xfs_alloc_log_agf(tp, agbp, XFS_AGF_FREEBLKS);
- if (!isfl)
- xfs_trans_mod_sb(tp, XFS_TRANS_SB_FDBLOCKS, (long)len);
- XFS_STATS_INC(xs_freex);
- XFS_STATS_ADD(xs_freeb, len);
- }
+ pag = xfs_perag_get(mp, agno);
+ error = xfs_alloc_update_counters(tp, pag, agbp, len);
+ xfs_perag_put(pag);
+ if (error)
+ goto error0;
+
+ if (!isfl)
+ xfs_trans_mod_sb(tp, XFS_TRANS_SB_FDBLOCKS, (long)len);
+ XFS_STATS_INC(xs_freex);
+ XFS_STATS_ADD(xs_freeb, len);
trace_xfs_free_extent(mp, agno, bno, len, isfl, haveleft, haveright);
- /*
- * Since blocks move to the free list without the coordination
- * used in xfs_bmap_finish, we can't allow block to be available
- * for reallocation and non-transaction writing (user data)
- * until we know that the transaction that moved it to the free
- * list is permanently on disk. We track the blocks by declaring
- * these blocks as "busy"; the busy list is maintained on a per-ag
- * basis and each transaction records which entries should be removed
- * when the iclog commits to disk. If a busy block is allocated,
- * the iclog is pushed up to the LSN that freed the block.
- */
- xfs_alloc_busy_insert(tp, agno, bno, len);
return 0;
error0:
@@ -1788,12 +1902,11 @@ xfs_alloc_fix_freelist(
/*
* Initialize the args structure.
*/
+ memset(&targs, 0, sizeof(targs));
targs.tp = tp;
targs.mp = mp;
targs.agbp = agbp;
targs.agno = args->agno;
- targs.mod = targs.minleft = targs.wasdel = targs.userdata =
- targs.minalignslop = 0;
targs.alignment = targs.minlen = targs.prod = targs.isfl = 1;
targs.type = XFS_ALLOCTYPE_THIS_AG;
targs.pag = pag;
@@ -1900,21 +2013,6 @@ xfs_alloc_get_freelist(
xfs_alloc_log_agf(tp, agbp, logflags);
*bnop = bno;
- /*
- * As blocks are freed, they are added to the per-ag busy list and
- * remain there until the freeing transaction is committed to disk.
- * Now that we have allocated blocks, this list must be searched to see
- * if a block is being reused. If one is, then the freeing transaction
- * must be pushed to disk before this transaction.
- *
- * We do this by setting the current transaction to a sync transaction
- * which guarantees that the freeing transaction is on disk before this
- * transaction. This is done instead of a synchronous log force here so
- * that we don't sit and wait with the AGF locked in the transaction
- * during the log force.
- */
- if (xfs_alloc_busy_search(mp, be32_to_cpu(agf->agf_seqno), bno, 1))
- xfs_trans_set_sync(tp);
return 0;
}
@@ -2027,6 +2125,63 @@ xfs_alloc_put_freelist(
return 0;
}
+static void
+xfs_agf_verify(
+ struct xfs_buf *bp)
+ {
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_agf *agf;
+ int agf_ok;
+
+ agf = XFS_BUF_TO_AGF(bp);
+
+ agf_ok = agf->agf_magicnum == cpu_to_be32(XFS_AGF_MAGIC) &&
+ XFS_AGF_GOOD_VERSION(be32_to_cpu(agf->agf_versionnum)) &&
+ be32_to_cpu(agf->agf_freeblks) <= be32_to_cpu(agf->agf_length) &&
+ be32_to_cpu(agf->agf_flfirst) < XFS_AGFL_SIZE(mp) &&
+ be32_to_cpu(agf->agf_fllast) < XFS_AGFL_SIZE(mp) &&
+ be32_to_cpu(agf->agf_flcount) <= XFS_AGFL_SIZE(mp);
+
+ /*
+ * during growfs operations, the perag is not fully initialised,
+ * so we can't use it for any useful checking. growfs ensures we can't
+ * use it by using uncached buffers that don't have the perag attached
+ * so we can detect and avoid this problem.
+ */
+ if (bp->b_pag)
+ agf_ok = agf_ok && be32_to_cpu(agf->agf_seqno) ==
+ bp->b_pag->pag_agno;
+
+ if (xfs_sb_version_haslazysbcount(&mp->m_sb))
+ agf_ok = agf_ok && be32_to_cpu(agf->agf_btreeblks) <=
+ be32_to_cpu(agf->agf_length);
+
+ if (unlikely(XFS_TEST_ERROR(!agf_ok, mp, XFS_ERRTAG_ALLOC_READ_AGF,
+ XFS_RANDOM_ALLOC_READ_AGF))) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, agf);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+}
+
+static void
+xfs_agf_read_verify(
+ struct xfs_buf *bp)
+{
+ xfs_agf_verify(bp);
+}
+
+static void
+xfs_agf_write_verify(
+ struct xfs_buf *bp)
+{
+ xfs_agf_verify(bp);
+}
+
+const struct xfs_buf_ops xfs_agf_buf_ops = {
+ .verify_read = xfs_agf_read_verify,
+ .verify_write = xfs_agf_write_verify,
+};
+
/*
* Read in the allocation group header (free/alloc section).
*/
@@ -2038,45 +2193,20 @@ xfs_read_agf(
int flags, /* XFS_BUF_ */
struct xfs_buf **bpp) /* buffer for the ag freelist header */
{
- struct xfs_agf *agf; /* ag freelist header */
- int agf_ok; /* set if agf is consistent */
int error;
ASSERT(agno != NULLAGNUMBER);
error = xfs_trans_read_buf(
mp, tp, mp->m_ddev_targp,
XFS_AG_DADDR(mp, agno, XFS_AGF_DADDR(mp)),
- XFS_FSS_TO_BB(mp, 1), flags, bpp);
+ XFS_FSS_TO_BB(mp, 1), flags, bpp, &xfs_agf_buf_ops);
if (error)
return error;
if (!*bpp)
return 0;
- ASSERT(!XFS_BUF_GETERROR(*bpp));
- agf = XFS_BUF_TO_AGF(*bpp);
-
- /*
- * Validate the magic number of the agf block.
- */
- agf_ok =
- be32_to_cpu(agf->agf_magicnum) == XFS_AGF_MAGIC &&
- XFS_AGF_GOOD_VERSION(be32_to_cpu(agf->agf_versionnum)) &&
- be32_to_cpu(agf->agf_freeblks) <= be32_to_cpu(agf->agf_length) &&
- be32_to_cpu(agf->agf_flfirst) < XFS_AGFL_SIZE(mp) &&
- be32_to_cpu(agf->agf_fllast) < XFS_AGFL_SIZE(mp) &&
- be32_to_cpu(agf->agf_flcount) <= XFS_AGFL_SIZE(mp) &&
- be32_to_cpu(agf->agf_seqno) == agno;
- if (xfs_sb_version_haslazysbcount(&mp->m_sb))
- agf_ok = agf_ok && be32_to_cpu(agf->agf_btreeblks) <=
- be32_to_cpu(agf->agf_length);
- if (unlikely(XFS_TEST_ERROR(!agf_ok, mp, XFS_ERRTAG_ALLOC_READ_AGF,
- XFS_RANDOM_ALLOC_READ_AGF))) {
- XFS_CORRUPTION_ERROR("xfs_alloc_read_agf",
- XFS_ERRLEVEL_LOW, mp, agf);
- xfs_trans_brelse(tp, *bpp);
- return XFS_ERROR(EFSCORRUPTED);
- }
- XFS_BUF_SET_VTYPE_REF(*bpp, B_FS_AGF, XFS_AGF_REF);
+ ASSERT(!(*bpp)->b_error);
+ xfs_buf_set_ref(*bpp, XFS_AGF_REF);
return 0;
}
@@ -2104,7 +2234,7 @@ xfs_alloc_read_agf(
return error;
if (!*bpp)
return 0;
- ASSERT(!XFS_BUF_GETERROR(*bpp));
+ ASSERT(!(*bpp)->b_error);
agf = XFS_BUF_TO_AGF(*bpp);
pag = xfs_perag_get(mp, agno);
@@ -2371,18 +2501,36 @@ xfs_free_extent(
memset(&args, 0, sizeof(xfs_alloc_arg_t));
args.tp = tp;
args.mp = tp->t_mountp;
+
+ /*
+ * validate that the block number is legal - the enables us to detect
+ * and handle a silent filesystem corruption rather than crashing.
+ */
args.agno = XFS_FSB_TO_AGNO(args.mp, bno);
- ASSERT(args.agno < args.mp->m_sb.sb_agcount);
+ if (args.agno >= args.mp->m_sb.sb_agcount)
+ return EFSCORRUPTED;
+
args.agbno = XFS_FSB_TO_AGBNO(args.mp, bno);
+ if (args.agbno >= args.mp->m_sb.sb_agblocks)
+ return EFSCORRUPTED;
+
args.pag = xfs_perag_get(args.mp, args.agno);
- if ((error = xfs_alloc_fix_freelist(&args, XFS_ALLOC_FLAG_FREEING)))
+ ASSERT(args.pag);
+
+ error = xfs_alloc_fix_freelist(&args, XFS_ALLOC_FLAG_FREEING);
+ if (error)
goto error0;
-#ifdef DEBUG
- ASSERT(args.agbp != NULL);
- ASSERT((args.agbno + len) <=
- be32_to_cpu(XFS_BUF_TO_AGF(args.agbp)->agf_length));
-#endif
+
+ /* validate the extent size is legal now we have the agf locked */
+ if (args.agbno + len >
+ be32_to_cpu(XFS_BUF_TO_AGF(args.agbp)->agf_length)) {
+ error = EFSCORRUPTED;
+ goto error0;
+ }
+
error = xfs_free_ag_extent(tp, args.agbp, args.agno, args.agbno, len, 0);
+ if (!error)
+ xfs_extent_busy_insert(tp, args.agno, args.agbno, len, 0);
error0:
xfs_perag_put(args.pag);
return error;
diff --git a/libxfs/xfs_alloc_btree.c b/libxfs/xfs_alloc_btree.c
index b782d9d..dc9ed48 100644
--- a/libxfs/xfs_alloc_btree.c
+++ b/libxfs/xfs_alloc_btree.c
@@ -75,6 +75,8 @@ xfs_allocbt_alloc_block(
return 0;
}
+ xfs_extent_busy_reuse(cur->bc_mp, cur->bc_private.a.agno, bno, 1, false);
+
xfs_trans_agbtree_delta(cur->bc_tp, 1);
new->s = cpu_to_be32(bno);
@@ -98,19 +100,11 @@ xfs_allocbt_free_block(
if (error)
return error;
- /*
- * Since blocks move to the free list without the coordination used in
- * xfs_bmap_finish, we can't allow block to be available for
- * reallocation and non-transaction writing (user data) until we know
- * that the transaction that moved it to the free list is permanently
- * on disk. We track the blocks by declaring these blocks as "busy";
- * the busy list is maintained on a per-ag basis and each transaction
- * records which entries should be removed when the iclog commits to
- * disk. If a busy block is allocated, the iclog is pushed up to the
- * LSN that freed the block.
- */
- xfs_alloc_busy_insert(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1);
+ xfs_extent_busy_insert(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1,
+ XFS_EXTENT_BUSY_SKIP_DISCARD);
xfs_trans_agbtree_delta(cur->bc_tp, -1);
+
+ xfs_trans_binval(cur->bc_tp, bp);
return 0;
}
@@ -260,6 +254,82 @@ xfs_allocbt_key_diff(
return (__int64_t)be32_to_cpu(kp->ar_startblock) - rec->ar_startblock;
}
+static void
+xfs_allocbt_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
+ struct xfs_perag *pag = bp->b_pag;
+ unsigned int level;
+ int sblock_ok; /* block passes checks */
+
+ /*
+ * magic number and level verification
+ *
+ * During growfs operations, we can't verify the exact level as the
+ * perag is not fully initialised and hence not attached to the buffer.
+ * In this case, check against the maximum tree depth.
+ */
+ level = be16_to_cpu(block->bb_level);
+ switch (cpu_to_be32(block->bb_magic)) {
+ case XFS_ABTB_MAGIC:
+ if (pag)
+ sblock_ok = level < pag->pagf_levels[XFS_BTNUM_BNOi];
+ else
+ sblock_ok = level < mp->m_ag_maxlevels;
+ break;
+ case XFS_ABTC_MAGIC:
+ if (pag)
+ sblock_ok = level < pag->pagf_levels[XFS_BTNUM_CNTi];
+ else
+ sblock_ok = level < mp->m_ag_maxlevels;
+ break;
+ default:
+ sblock_ok = 0;
+ break;
+ }
+
+ /* numrecs verification */
+ sblock_ok = sblock_ok &&
+ be16_to_cpu(block->bb_numrecs) <= mp->m_alloc_mxr[level != 0];
+
+ /* sibling pointer verification */
+ sblock_ok = sblock_ok &&
+ (block->bb_u.s.bb_leftsib == cpu_to_be32(NULLAGBLOCK) ||
+ be32_to_cpu(block->bb_u.s.bb_leftsib) < mp->m_sb.sb_agblocks) &&
+ block->bb_u.s.bb_leftsib &&
+ (block->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK) ||
+ be32_to_cpu(block->bb_u.s.bb_rightsib) < mp->m_sb.sb_agblocks) &&
+ block->bb_u.s.bb_rightsib;
+
+ if (!sblock_ok) {
+ trace_xfs_btree_corrupt(bp, _RET_IP_);
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, block);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+}
+
+static void
+xfs_allocbt_read_verify(
+ struct xfs_buf *bp)
+{
+ xfs_allocbt_verify(bp);
+}
+
+static void
+xfs_allocbt_write_verify(
+ struct xfs_buf *bp)
+{
+ xfs_allocbt_verify(bp);
+}
+
+const struct xfs_buf_ops xfs_allocbt_buf_ops = {
+ .verify_read = xfs_allocbt_read_verify,
+ .verify_write = xfs_allocbt_write_verify,
+};
+
+
#ifdef DEBUG
STATIC int
xfs_allocbt_keys_inorder(
@@ -381,7 +451,7 @@ static const struct xfs_btree_ops xfs_allocbt_ops = {
.init_rec_from_cur = xfs_allocbt_init_rec_from_cur,
.init_ptr_from_cur = xfs_allocbt_init_ptr_from_cur,
.key_diff = xfs_allocbt_key_diff,
-
+ .buf_ops = &xfs_allocbt_buf_ops,
#ifdef DEBUG
.keys_inorder = xfs_allocbt_keys_inorder,
.recs_inorder = xfs_allocbt_recs_inorder,
@@ -415,13 +485,16 @@ xfs_allocbt_init_cursor(
cur->bc_tp = tp;
cur->bc_mp = mp;
- cur->bc_nlevels = be32_to_cpu(agf->agf_levels[btnum]);
cur->bc_btnum = btnum;
cur->bc_blocklog = mp->m_sb.sb_blocklog;
-
cur->bc_ops = &xfs_allocbt_ops;
- if (btnum == XFS_BTNUM_CNT)
+
+ if (btnum == XFS_BTNUM_CNT) {
+ cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]);
cur->bc_flags = XFS_BTREE_LASTREC_UPDATE;
+ } else {
+ cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]);
+ }
cur->bc_private.a.agbp = agbp;
cur->bc_private.a.agno = agno;
diff --git a/libxfs/xfs_attr.c b/libxfs/xfs_attr.c
index eec7f8f..42546a9 100644
--- a/libxfs/xfs_attr.c
+++ b/libxfs/xfs_attr.c
@@ -295,8 +295,7 @@ xfs_attr_set_int(
return (error);
}
- xfs_trans_ijoin(args.trans, dp, XFS_ILOCK_EXCL);
- xfs_trans_ihold(args.trans, dp);
+ xfs_trans_ijoin(args.trans, dp, 0);
/*
* If the attribute list is non-existent or a shortform list,
@@ -365,10 +364,8 @@ xfs_attr_set_int(
* bmap_finish() may have committed the last trans and started
* a new one. We need the inode to be in all transactions.
*/
- if (committed) {
- xfs_trans_ijoin(args.trans, dp, XFS_ILOCK_EXCL);
- xfs_trans_ihold(args.trans, dp);
- }
+ if (committed)
+ xfs_trans_ijoin(args.trans, dp, 0);
/*
* Commit the leaf transformation. We'll need another (linked)
@@ -469,6 +466,13 @@ xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags)
args.whichfork = XFS_ATTR_FORK;
/*
+ * we have no control over the attribute names that userspace passes us
+ * to remove, so we have to allow the name lookup prior to attribute
+ * removal to fail.
+ */
+ args.op_flags = XFS_DA_OP_OKNOENT;
+
+ /*
* Attach the dquots to the inode.
*/
error = xfs_qm_dqattach(dp, 0);
@@ -509,8 +513,7 @@ xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags)
* No need to make quota reservations here. We expect to release some
* blocks not allocate in the common case.
*/
- xfs_trans_ijoin(args.trans, dp, XFS_ILOCK_EXCL);
- xfs_trans_ihold(args.trans, dp);
+ xfs_trans_ijoin(args.trans, dp, 0);
/*
* Decide on what work routines to call based on the inode size.
@@ -603,6 +606,8 @@ xfs_attr_shortform_addname(xfs_da_args_t *args)
{
int newsize, forkoff, retval;
+ trace_xfs_attr_sf_addname(args);
+
retval = xfs_attr_shortform_lookup(args);
if ((args->flags & ATTR_REPLACE) && (retval == ENOATTR)) {
return(retval);
@@ -643,19 +648,19 @@ STATIC int
xfs_attr_leaf_addname(xfs_da_args_t *args)
{
xfs_inode_t *dp;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
int retval, error, committed, forkoff;
+ trace_xfs_attr_leaf_addname(args);
+
/*
* Read the (only) block in the attribute list in.
*/
dp = args->dp;
args->blkno = 0;
- error = xfs_da_read_buf(args->trans, args->dp, args->blkno, -1, &bp,
- XFS_ATTR_FORK);
+ error = xfs_attr_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
if (error)
- return(error);
- ASSERT(bp != NULL);
+ return error;
/*
* Look up the given attribute in the leaf block. Figure out if
@@ -663,13 +668,16 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
*/
retval = xfs_attr_leaf_lookup_int(bp, args);
if ((args->flags & ATTR_REPLACE) && (retval == ENOATTR)) {
- xfs_da_brelse(args->trans, bp);
+ xfs_trans_brelse(args->trans, bp);
return(retval);
} else if (retval == EEXIST) {
if (args->flags & ATTR_CREATE) { /* pure create op */
- xfs_da_brelse(args->trans, bp);
+ xfs_trans_brelse(args->trans, bp);
return(retval);
}
+
+ trace_xfs_attr_leaf_replace(args);
+
args->op_flags |= XFS_DA_OP_RENAME; /* an atomic rename */
args->blkno2 = args->blkno; /* set 2nd entry info*/
args->index2 = args->index;
@@ -682,7 +690,6 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
* if required.
*/
retval = xfs_attr_leaf_add(bp, args);
- xfs_da_buf_done(bp);
if (retval == ENOSPC) {
/*
* Promote the attribute list to the Btree format, then
@@ -706,10 +713,8 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
* bmap_finish() may have committed the last trans and started
* a new one. We need the inode to be in all transactions.
*/
- if (committed) {
- xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
- xfs_trans_ihold(args->trans, dp);
- }
+ if (committed)
+ xfs_trans_ijoin(args->trans, dp, 0);
/*
* Commit the current trans (including the inode) and start
@@ -779,12 +784,12 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
* Read in the block containing the "old" attr, then
* remove the "old" attr from that block (neat, huh!)
*/
- error = xfs_da_read_buf(args->trans, args->dp, args->blkno, -1,
- &bp, XFS_ATTR_FORK);
+ error = xfs_attr_leaf_read(args->trans, args->dp, args->blkno,
+ -1, &bp);
if (error)
- return(error);
- ASSERT(bp != NULL);
- (void)xfs_attr_leaf_remove(bp, args);
+ return error;
+
+ xfs_attr_leaf_remove(bp, args);
/*
* If the result is small enough, shrink it all into the inode.
@@ -810,12 +815,9 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
* and started a new one. We need the inode to be
* in all transactions.
*/
- if (committed) {
- xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
- xfs_trans_ihold(args->trans, dp);
- }
- } else
- xfs_da_buf_done(bp);
+ if (committed)
+ xfs_trans_ijoin(args->trans, dp, 0);
+ }
/*
* Commit the remove and start the next trans in series.
@@ -841,28 +843,27 @@ STATIC int
xfs_attr_leaf_removename(xfs_da_args_t *args)
{
xfs_inode_t *dp;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
int error, committed, forkoff;
+ trace_xfs_attr_leaf_removename(args);
+
/*
* Remove the attribute.
*/
dp = args->dp;
args->blkno = 0;
- error = xfs_da_read_buf(args->trans, args->dp, args->blkno, -1, &bp,
- XFS_ATTR_FORK);
- if (error) {
- return(error);
- }
+ error = xfs_attr_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+ if (error)
+ return error;
- ASSERT(bp != NULL);
error = xfs_attr_leaf_lookup_int(bp, args);
if (error == ENOATTR) {
- xfs_da_brelse(args->trans, bp);
+ xfs_trans_brelse(args->trans, bp);
return(error);
}
- (void)xfs_attr_leaf_remove(bp, args);
+ xfs_attr_leaf_remove(bp, args);
/*
* If the result is small enough, shrink it all into the inode.
@@ -886,12 +887,9 @@ xfs_attr_leaf_removename(xfs_da_args_t *args)
* bmap_finish() may have committed the last trans and started
* a new one. We need the inode to be in all transactions.
*/
- if (committed) {
- xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
- xfs_trans_ihold(args->trans, dp);
- }
- } else
- xfs_da_buf_done(bp);
+ if (committed)
+ xfs_trans_ijoin(args->trans, dp, 0);
+ }
return(0);
}
@@ -904,23 +902,23 @@ xfs_attr_leaf_removename(xfs_da_args_t *args)
STATIC int
xfs_attr_leaf_get(xfs_da_args_t *args)
{
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
int error;
+ trace_xfs_attr_leaf_get(args);
+
args->blkno = 0;
- error = xfs_da_read_buf(args->trans, args->dp, args->blkno, -1, &bp,
- XFS_ATTR_FORK);
+ error = xfs_attr_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
if (error)
- return(error);
- ASSERT(bp != NULL);
+ return error;
error = xfs_attr_leaf_lookup_int(bp, args);
if (error != EEXIST) {
- xfs_da_brelse(args->trans, bp);
+ xfs_trans_brelse(args->trans, bp);
return(error);
}
error = xfs_attr_leaf_getvalue(bp, args);
- xfs_da_brelse(args->trans, bp);
+ xfs_trans_brelse(args->trans, bp);
if (!error && (args->rmtblkno > 0) && !(args->flags & ATTR_KERNOVAL)) {
error = xfs_attr_rmtval_get(args);
}
@@ -950,6 +948,8 @@ xfs_attr_node_addname(xfs_da_args_t *args)
xfs_mount_t *mp;
int committed, retval, error;
+ trace_xfs_attr_node_addname(args);
+
/*
* Fill in bucket of arguments/results/context to carry around.
*/
@@ -976,6 +976,9 @@ restart:
} else if (retval == EEXIST) {
if (args->flags & ATTR_CREATE)
goto out;
+
+ trace_xfs_attr_node_replace(args);
+
args->op_flags |= XFS_DA_OP_RENAME; /* atomic rename op */
args->blkno2 = args->blkno; /* set 2nd entry info*/
args->index2 = args->index;
@@ -1013,10 +1016,8 @@ restart:
* and started a new one. We need the inode to be
* in all transactions.
*/
- if (committed) {
- xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
- xfs_trans_ihold(args->trans, dp);
- }
+ if (committed)
+ xfs_trans_ijoin(args->trans, dp, 0);
/*
* Commit the node conversion and start the next
@@ -1052,10 +1053,8 @@ restart:
* bmap_finish() may have committed the last trans and started
* a new one. We need the inode to be in all transactions.
*/
- if (committed) {
- xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
- xfs_trans_ihold(args->trans, dp);
- }
+ if (committed)
+ xfs_trans_ijoin(args->trans, dp, 0);
} else {
/*
* Addition succeeded, update Btree hashvals.
@@ -1166,10 +1165,8 @@ restart:
* and started a new one. We need the inode to be
* in all transactions.
*/
- if (committed) {
- xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
- xfs_trans_ihold(args->trans, dp);
- }
+ if (committed)
+ xfs_trans_ijoin(args->trans, dp, 0);
}
/*
@@ -1210,9 +1207,11 @@ xfs_attr_node_removename(xfs_da_args_t *args)
xfs_da_state_t *state;
xfs_da_state_blk_t *blk;
xfs_inode_t *dp;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
int retval, error, committed, forkoff;
+ trace_xfs_attr_node_removename(args);
+
/*
* Tie a string around our finger to remind us where we are.
*/
@@ -1300,10 +1299,8 @@ xfs_attr_node_removename(xfs_da_args_t *args)
* bmap_finish() may have committed the last trans and started
* a new one. We need the inode to be in all transactions.
*/
- if (committed) {
- xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
- xfs_trans_ihold(args->trans, dp);
- }
+ if (committed)
+ xfs_trans_ijoin(args->trans, dp, 0);
/*
* Commit the Btree join operation and start a new trans.
@@ -1322,16 +1319,11 @@ xfs_attr_node_removename(xfs_da_args_t *args)
*/
ASSERT(state->path.active == 1);
ASSERT(state->path.blk[0].bp);
- xfs_da_buf_done(state->path.blk[0].bp);
state->path.blk[0].bp = NULL;
- error = xfs_da_read_buf(args->trans, args->dp, 0, -1, &bp,
- XFS_ATTR_FORK);
+ error = xfs_attr_leaf_read(args->trans, args->dp, 0, -1, &bp);
if (error)
goto out;
- ASSERT(be16_to_cpu(((xfs_attr_leafblock_t *)
- bp->data)->hdr.info.magic)
- == XFS_ATTR_LEAF_MAGIC);
if ((forkoff = xfs_attr_shortform_allfit(bp, dp))) {
xfs_bmap_init(args->flist, args->firstblock);
@@ -1354,12 +1346,10 @@ xfs_attr_node_removename(xfs_da_args_t *args)
* and started a new one. We need the inode to be
* in all transactions.
*/
- if (committed) {
- xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
- xfs_trans_ihold(args->trans, dp);
- }
+ if (committed)
+ xfs_trans_ijoin(args->trans, dp, 0);
} else
- xfs_da_brelse(args->trans, bp);
+ xfs_trans_brelse(args->trans, bp);
}
error = 0;
@@ -1381,6 +1371,8 @@ xfs_attr_fillstate(xfs_da_state_t *state)
xfs_da_state_blk_t *blk;
int level;
+ trace_xfs_attr_fillstate(state->args);
+
/*
* Roll down the "path" in the state structure, storing the on-disk
* block number for those buffers in the "path".
@@ -1389,8 +1381,7 @@ xfs_attr_fillstate(xfs_da_state_t *state)
ASSERT((path->active >= 0) && (path->active < XFS_DA_NODE_MAXDEPTH));
for (blk = path->blk, level = 0; level < path->active; blk++, level++) {
if (blk->bp) {
- blk->disk_blkno = xfs_da_blkno(blk->bp);
- xfs_da_buf_done(blk->bp);
+ blk->disk_blkno = XFS_BUF_ADDR(blk->bp);
blk->bp = NULL;
} else {
blk->disk_blkno = 0;
@@ -1405,8 +1396,7 @@ xfs_attr_fillstate(xfs_da_state_t *state)
ASSERT((path->active >= 0) && (path->active < XFS_DA_NODE_MAXDEPTH));
for (blk = path->blk, level = 0; level < path->active; blk++, level++) {
if (blk->bp) {
- blk->disk_blkno = xfs_da_blkno(blk->bp);
- xfs_da_buf_done(blk->bp);
+ blk->disk_blkno = XFS_BUF_ADDR(blk->bp);
blk->bp = NULL;
} else {
blk->disk_blkno = 0;
@@ -1429,6 +1419,8 @@ xfs_attr_refillstate(xfs_da_state_t *state)
xfs_da_state_blk_t *blk;
int level, error;
+ trace_xfs_attr_refillstate(state->args);
+
/*
* Roll down the "path" in the state structure, storing the on-disk
* block number for those buffers in the "path".
@@ -1437,7 +1429,7 @@ xfs_attr_refillstate(xfs_da_state_t *state)
ASSERT((path->active >= 0) && (path->active < XFS_DA_NODE_MAXDEPTH));
for (blk = path->blk, level = 0; level < path->active; blk++, level++) {
if (blk->disk_blkno) {
- error = xfs_da_read_buf(state->args->trans,
+ error = xfs_da_node_read(state->args->trans,
state->args->dp,
blk->blkno, blk->disk_blkno,
&blk->bp, XFS_ATTR_FORK);
@@ -1456,7 +1448,7 @@ xfs_attr_refillstate(xfs_da_state_t *state)
ASSERT((path->active >= 0) && (path->active < XFS_DA_NODE_MAXDEPTH));
for (blk = path->blk, level = 0; level < path->active; blk++, level++) {
if (blk->disk_blkno) {
- error = xfs_da_read_buf(state->args->trans,
+ error = xfs_da_node_read(state->args->trans,
state->args->dp,
blk->blkno, blk->disk_blkno,
&blk->bp, XFS_ATTR_FORK);
@@ -1485,6 +1477,8 @@ xfs_attr_node_get(xfs_da_args_t *args)
int error, retval;
int i;
+ trace_xfs_attr_node_get(args);
+
state = xfs_da_state_alloc();
state->args = args;
state->mp = args->dp->i_mount;
@@ -1516,7 +1510,7 @@ xfs_attr_node_get(xfs_da_args_t *args)
* If not in a transaction, we have to release all the buffers.
*/
for (i = 0; i < state->path.active; i++) {
- xfs_da_brelse(args->trans, state->path.blk[i].bp);
+ xfs_trans_brelse(args->trans, state->path.blk[i].bp);
state->path.blk[i].bp = NULL;
}
@@ -1543,6 +1537,8 @@ xfs_attr_rmtval_get(xfs_da_args_t *args)
int nmap, error, tmp, valuelen, blkcnt, i;
xfs_dablk_t lblkno;
+ trace_xfs_attr_rmtval_get(args);
+
ASSERT(!(args->flags & ATTR_KERNOVAL));
mp = args->dp->i_mount;
@@ -1551,10 +1547,9 @@ xfs_attr_rmtval_get(xfs_da_args_t *args)
lblkno = args->rmtblkno;
while (valuelen > 0) {
nmap = ATTR_RMTVALUE_MAPSIZE;
- error = xfs_bmapi(args->trans, args->dp, (xfs_fileoff_t)lblkno,
- args->rmtblkcnt,
- XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,
- NULL, 0, map, &nmap, NULL);
+ error = xfs_bmapi_read(args->dp, (xfs_fileoff_t)lblkno,
+ args->rmtblkcnt, map, &nmap,
+ XFS_BMAPI_ATTRFORK);
if (error)
return(error);
ASSERT(nmap >= 1);
@@ -1564,14 +1559,12 @@ xfs_attr_rmtval_get(xfs_da_args_t *args)
(map[i].br_startblock != HOLESTARTBLOCK));
dblkno = XFS_FSB_TO_DADDR(mp, map[i].br_startblock);
blkcnt = XFS_FSB_TO_BB(mp, map[i].br_blockcount);
- error = xfs_read_buf(mp, mp->m_ddev_targp, dblkno,
- blkcnt, XBF_LOCK | XBF_DONT_BLOCK,
- &bp);
+ error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp,
+ dblkno, blkcnt, 0, &bp, NULL);
if (error)
return(error);
- tmp = (valuelen < XFS_BUF_SIZE(bp))
- ? valuelen : XFS_BUF_SIZE(bp);
+ tmp = min_t(int, valuelen, BBTOB(bp->b_length));
xfs_buf_iomove(bp, 0, tmp, dst, XBRW_READ);
xfs_buf_relse(bp);
dst += tmp;
@@ -1601,6 +1594,8 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
xfs_dablk_t lblkno;
int blkcnt, valuelen, nmap, error, tmp, committed;
+ trace_xfs_attr_rmtval_set(args);
+
dp = args->dp;
mp = dp->i_mount;
src = args->value;
@@ -1628,10 +1623,9 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
*/
xfs_bmap_init(args->flist, args->firstblock);
nmap = 1;
- error = xfs_bmapi(args->trans, dp, (xfs_fileoff_t)lblkno,
+ error = xfs_bmapi_write(args->trans, dp, (xfs_fileoff_t)lblkno,
blkcnt,
- XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA |
- XFS_BMAPI_WRITE,
+ XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,
args->firstblock, args->total, &map, &nmap,
args->flist);
if (!error) {
@@ -1649,10 +1643,8 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
* bmap_finish() may have committed the last trans and started
* a new one. We need the inode to be in all transactions.
*/
- if (committed) {
- xfs_trans_ijoin(args->trans, dp, XFS_ILOCK_EXCL);
- xfs_trans_ihold(args->trans, dp);
- }
+ if (committed)
+ xfs_trans_ijoin(args->trans, dp, 0);
ASSERT(nmap == 1);
ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
@@ -1677,19 +1669,18 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
lblkno = args->rmtblkno;
valuelen = args->valuelen;
while (valuelen > 0) {
+ int buflen;
+
/*
* Try to remember where we decided to put the value.
*/
xfs_bmap_init(args->flist, args->firstblock);
nmap = 1;
- error = xfs_bmapi(NULL, dp, (xfs_fileoff_t)lblkno,
- args->rmtblkcnt,
- XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,
- args->firstblock, 0, &map, &nmap,
- NULL);
- if (error) {
+ error = xfs_bmapi_read(dp, (xfs_fileoff_t)lblkno,
+ args->rmtblkcnt, &map, &nmap,
+ XFS_BMAPI_ATTRFORK);
+ if (error)
return(error);
- }
ASSERT(nmap == 1);
ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
(map.br_startblock != HOLESTARTBLOCK));
@@ -1697,19 +1688,20 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
dblkno = XFS_FSB_TO_DADDR(mp, map.br_startblock),
blkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount);
- bp = xfs_buf_get(mp->m_ddev_targp, dblkno, blkcnt,
- XBF_LOCK | XBF_DONT_BLOCK);
- ASSERT(bp);
- ASSERT(!XFS_BUF_GETERROR(bp));
+ bp = xfs_buf_get(mp->m_ddev_targp, dblkno, blkcnt, 0);
+ if (!bp)
+ return ENOMEM;
- tmp = (valuelen < XFS_BUF_SIZE(bp)) ? valuelen :
- XFS_BUF_SIZE(bp);
+ buflen = BBTOB(bp->b_length);
+ tmp = min_t(int, valuelen, buflen);
xfs_buf_iomove(bp, 0, tmp, src, XBRW_WRITE);
- if (tmp < XFS_BUF_SIZE(bp))
- xfs_buf_zero(bp, tmp, XFS_BUF_SIZE(bp) - tmp);
- if ((error = xfs_bwrite(mp, bp))) {/* GROT: NOTE: synchronous write */
- return (error);
- }
+ if (tmp < buflen)
+ xfs_buf_zero(bp, tmp, buflen - tmp);
+
+ error = xfs_bwrite(mp, bp); /* GROT: NOTE: synchronous write */
+ xfs_buf_relse(bp);
+ if (error)
+ return error;
src += tmp;
valuelen -= tmp;
@@ -1733,6 +1725,8 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args)
xfs_dablk_t lblkno;
int valuelen, blkcnt, nmap, error, done, committed;
+ trace_xfs_attr_rmtval_remove(args);
+
mp = args->dp->i_mount;
/*
@@ -1745,16 +1739,12 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args)
/*
* Try to remember where we decided to put the value.
*/
- xfs_bmap_init(args->flist, args->firstblock);
nmap = 1;
- error = xfs_bmapi(NULL, args->dp, (xfs_fileoff_t)lblkno,
- args->rmtblkcnt,
- XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,
- args->firstblock, 0, &map, &nmap,
- args->flist);
- if (error) {
+ error = xfs_bmapi_read(args->dp, (xfs_fileoff_t)lblkno,
+ args->rmtblkcnt, &map, &nmap,
+ XFS_BMAPI_ATTRFORK);
+ if (error)
return(error);
- }
ASSERT(nmap == 1);
ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
(map.br_startblock != HOLESTARTBLOCK));
@@ -1767,8 +1757,7 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args)
*/
bp = xfs_incore(mp->m_ddev_targp, dblkno, blkcnt, XBF_TRYLOCK);
if (bp) {
- XFS_BUF_STALE(bp);
- XFS_BUF_UNDELAYWRITE(bp);
+ xfs_buf_stale(bp);
xfs_buf_relse(bp);
bp = NULL;
}
@@ -1805,10 +1794,8 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args)
* bmap_finish() may have committed the last trans and started
* a new one. We need the inode to be in all transactions.
*/
- if (committed) {
- xfs_trans_ijoin(args->trans, args->dp, XFS_ILOCK_EXCL);
- xfs_trans_ihold(args->trans, args->dp);
- }
+ if (committed)
+ xfs_trans_ijoin(args->trans, args->dp, 0);
/*
* Close out trans and start the next one in the chain.
diff --git a/libxfs/xfs_attr_leaf.c b/libxfs/xfs_attr_leaf.c
index d1f7a20..824c122 100644
--- a/libxfs/xfs_attr_leaf.c
+++ b/libxfs/xfs_attr_leaf.c
@@ -32,10 +32,11 @@
* Routines used for growing the Btree.
*/
STATIC int xfs_attr_leaf_create(xfs_da_args_t *args, xfs_dablk_t which_block,
- xfs_dabuf_t **bpp);
-STATIC int xfs_attr_leaf_add_work(xfs_dabuf_t *leaf_buffer, xfs_da_args_t *args,
- int freemap_index);
-STATIC void xfs_attr_leaf_compact(xfs_trans_t *trans, xfs_dabuf_t *leaf_buffer);
+ struct xfs_buf **bpp);
+STATIC int xfs_attr_leaf_add_work(struct xfs_buf *leaf_buffer,
+ xfs_da_args_t *args, int freemap_index);
+STATIC void xfs_attr_leaf_compact(struct xfs_da_args *args,
+ struct xfs_buf *leaf_buffer);
STATIC void xfs_attr_leaf_rebalance(xfs_da_state_t *state,
xfs_da_state_blk_t *blk1,
xfs_da_state_blk_t *blk2);
@@ -56,6 +57,52 @@ STATIC void xfs_attr_leaf_moveents(xfs_attr_leafblock_t *src_leaf,
xfs_mount_t *mp);
STATIC int xfs_attr_leaf_entsize(xfs_attr_leafblock_t *leaf, int index);
+static void
+xfs_attr_leaf_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_attr_leaf_hdr *hdr = bp->b_addr;
+ int block_ok = 0;
+
+ block_ok = hdr->info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC);
+ if (!block_ok) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, hdr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+}
+
+static void
+xfs_attr_leaf_read_verify(
+ struct xfs_buf *bp)
+{
+ xfs_attr_leaf_verify(bp);
+}
+
+static void
+xfs_attr_leaf_write_verify(
+ struct xfs_buf *bp)
+{
+ xfs_attr_leaf_verify(bp);
+}
+
+const struct xfs_buf_ops xfs_attr_leaf_buf_ops = {
+ .verify_read = xfs_attr_leaf_read_verify,
+ .verify_write = xfs_attr_leaf_write_verify,
+};
+
+int
+xfs_attr_leaf_read(
+ struct xfs_trans *tp,
+ struct xfs_inode *dp,
+ xfs_dablk_t bno,
+ xfs_daddr_t mappedbno,
+ struct xfs_buf **bpp)
+{
+ return xfs_da_read_buf(tp, dp, bno, mappedbno, bpp,
+ XFS_ATTR_FORK, &xfs_attr_leaf_buf_ops);
+}
+
/*========================================================================
* Namespace helper routines
*========================================================================*/
@@ -78,6 +125,7 @@ xfs_attr_namesp_match(int arg_flags, int ondisk_flags)
/*
* Query whether the requested number of additional bytes of extended
* attribute space will be able to fit inline.
+ *
* Returns zero if not, else the di_forkoff fork offset to be used in the
* literal area for attribute data once the new bytes have been added.
*
@@ -90,7 +138,7 @@ xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes)
int offset;
int minforkoff; /* lower limit on valid forkoff locations */
int maxforkoff; /* upper limit on valid forkoff locations */
- int dsize;
+ int dsize;
xfs_mount_t *mp = dp->i_mount;
offset = (XFS_LITINO(mp) - bytes) >> 3; /* rounded down */
@@ -104,47 +152,60 @@ xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes)
return (offset >= minforkoff) ? minforkoff : 0;
}
- if (!(mp->m_flags & XFS_MOUNT_ATTR2)) {
- if (bytes <= XFS_IFORK_ASIZE(dp))
- return dp->i_d.di_forkoff;
+ /*
+ * If the requested numbers of bytes is smaller or equal to the
+ * current attribute fork size we can always proceed.
+ *
+ * Note that if_bytes in the data fork might actually be larger than
+ * the current data fork size is due to delalloc extents. In that
+ * case either the extent count will go down when they are converted
+ * to real extents, or the delalloc conversion will take care of the
+ * literal area rebalancing.
+ */
+ if (bytes <= XFS_IFORK_ASIZE(dp))
+ return dp->i_d.di_forkoff;
+
+ /*
+ * For attr2 we can try to move the forkoff if there is space in the
+ * literal area, but for the old format we are done if there is no
+ * space in the fixed attribute fork.
+ */
+ if (!(mp->m_flags & XFS_MOUNT_ATTR2))
return 0;
- }
dsize = dp->i_df.if_bytes;
-
+
switch (dp->i_d.di_format) {
case XFS_DINODE_FMT_EXTENTS:
- /*
+ /*
* If there is no attr fork and the data fork is extents,
- * determine if creating the default attr fork will result
- * in the extents form migrating to btree. If so, the
- * minimum offset only needs to be the space required for
+ * determine if creating the default attr fork will result
+ * in the extents form migrating to btree. If so, the
+ * minimum offset only needs to be the space required for
* the btree root.
- */
+ */
if (!dp->i_d.di_forkoff && dp->i_df.if_bytes >
xfs_default_attroffset(dp))
dsize = XFS_BMDR_SPACE_CALC(MINDBTPTRS);
break;
-
case XFS_DINODE_FMT_BTREE:
/*
- * If have data btree then keep forkoff if we have one,
- * otherwise we are adding a new attr, so then we set
- * minforkoff to where the btree root can finish so we have
+ * If we have a data btree then keep forkoff if we have one,
+ * otherwise we are adding a new attr, so then we set
+ * minforkoff to where the btree root can finish so we have
* plenty of room for attrs
*/
if (dp->i_d.di_forkoff) {
- if (offset < dp->i_d.di_forkoff)
+ if (offset < dp->i_d.di_forkoff)
return 0;
- else
- return dp->i_d.di_forkoff;
- } else
- dsize = XFS_BMAP_BROOT_SPACE(dp->i_df.if_broot);
+ return dp->i_d.di_forkoff;
+ }
+ dsize = XFS_BMAP_BROOT_SPACE(dp->i_df.if_broot);
break;
}
-
- /*
- * A data fork btree root must have space for at least
+
+ /*
+ * A data fork btree root must have space for at least
* MINDBTPTRS key/ptr pairs if the data fork is small or empty.
*/
minforkoff = MAX(dsize, XFS_BMDR_SPACE_CALC(MINDBTPTRS));
@@ -154,10 +215,10 @@ xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes)
maxforkoff = XFS_LITINO(mp) - XFS_BMDR_SPACE_CALC(MINABTPTRS);
maxforkoff = maxforkoff >> 3; /* rounded down */
- if (offset >= minforkoff && offset < maxforkoff)
- return offset;
if (offset >= maxforkoff)
return maxforkoff;
+ if (offset >= minforkoff)
+ return offset;
return 0;
}
@@ -189,6 +250,8 @@ xfs_attr_shortform_create(xfs_da_args_t *args)
xfs_inode_t *dp;
xfs_ifork_t *ifp;
+ trace_xfs_attr_sf_create(args);
+
dp = args->dp;
ASSERT(dp != NULL);
ifp = dp->i_afp;
@@ -222,13 +285,11 @@ xfs_attr_shortform_add(xfs_da_args_t *args, int forkoff)
xfs_inode_t *dp;
xfs_ifork_t *ifp;
+ trace_xfs_attr_sf_add(args);
+
dp = args->dp;
mp = dp->i_mount;
dp->i_d.di_forkoff = forkoff;
- dp->i_df.if_ext_max =
- XFS_IFORK_DSIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t);
- dp->i_afp->if_ext_max =
- XFS_IFORK_ASIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t);
ifp = dp->i_afp;
ASSERT(ifp->if_flags & XFS_IFINLINE);
@@ -280,7 +341,6 @@ xfs_attr_fork_reset(
ASSERT(ip->i_d.di_anextents == 0);
ASSERT(ip->i_afp == NULL);
- ip->i_df.if_ext_max = XFS_IFORK_DSIZE(ip) / sizeof(xfs_bmbt_rec_t);
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
}
@@ -296,6 +356,8 @@ xfs_attr_shortform_remove(xfs_da_args_t *args)
xfs_mount_t *mp;
xfs_inode_t *dp;
+ trace_xfs_attr_sf_remove(args);
+
dp = args->dp;
mp = dp->i_mount;
base = sizeof(xfs_attr_sf_hdr_t);
@@ -343,10 +405,6 @@ xfs_attr_shortform_remove(xfs_da_args_t *args)
(args->op_flags & XFS_DA_OP_ADDNAME) ||
!(mp->m_flags & XFS_MOUNT_ATTR2) ||
dp->i_d.di_format == XFS_DINODE_FMT_BTREE);
- dp->i_afp->if_ext_max =
- XFS_IFORK_ASIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t);
- dp->i_df.if_ext_max =
- XFS_IFORK_DSIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t);
xfs_trans_log_inode(args->trans, dp,
XFS_ILOG_CORE | XFS_ILOG_ADATA);
}
@@ -368,6 +426,8 @@ xfs_attr_shortform_lookup(xfs_da_args_t *args)
int i;
xfs_ifork_t *ifp;
+ trace_xfs_attr_sf_lookup(args);
+
ifp = args->dp->i_afp;
ASSERT(ifp->if_flags & XFS_IFINLINE);
sf = (xfs_attr_shortform_t *)ifp->if_u1.if_data;
@@ -436,9 +496,11 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
char *tmpbuffer;
int error, i, size;
xfs_dablk_t blkno;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
xfs_ifork_t *ifp;
+ trace_xfs_attr_sf_to_leaf(args);
+
dp = args->dp;
ifp = dp->i_afp;
sf = (xfs_attr_shortform_t *)ifp->if_u1.if_data;
@@ -504,8 +566,6 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
error = 0;
out:
- if(bp)
- xfs_da_buf_done(bp);
kmem_free(tmpbuffer);
return(error);
}
@@ -515,15 +575,17 @@ out:
* a shortform attribute list.
*/
int
-xfs_attr_shortform_allfit(xfs_dabuf_t *bp, xfs_inode_t *dp)
+xfs_attr_shortform_allfit(
+ struct xfs_buf *bp,
+ struct xfs_inode *dp)
{
xfs_attr_leafblock_t *leaf;
xfs_attr_leaf_entry_t *entry;
xfs_attr_leaf_name_local_t *name_loc;
int bytes, i;
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ leaf = bp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
entry = &leaf->entries[0];
bytes = sizeof(struct xfs_attr_sf_hdr);
@@ -552,7 +614,10 @@ xfs_attr_shortform_allfit(xfs_dabuf_t *bp, xfs_inode_t *dp)
* Convert a leaf attribute list to shortform attribute list
*/
int
-xfs_attr_leaf_to_shortform(xfs_dabuf_t *bp, xfs_da_args_t *args, int forkoff)
+xfs_attr_leaf_to_shortform(
+ struct xfs_buf *bp,
+ xfs_da_args_t *args,
+ int forkoff)
{
xfs_attr_leafblock_t *leaf;
xfs_attr_leaf_entry_t *entry;
@@ -562,15 +627,17 @@ xfs_attr_leaf_to_shortform(xfs_dabuf_t *bp, xfs_da_args_t *args, int forkoff)
char *tmpbuffer;
int error, i;
+ trace_xfs_attr_leaf_to_sf(args);
+
dp = args->dp;
tmpbuffer = kmem_alloc(XFS_LBSIZE(dp->i_mount), KM_SLEEP);
ASSERT(tmpbuffer != NULL);
ASSERT(bp != NULL);
- memcpy(tmpbuffer, bp->data, XFS_LBSIZE(dp->i_mount));
+ memcpy(tmpbuffer, bp->b_addr, XFS_LBSIZE(dp->i_mount));
leaf = (xfs_attr_leafblock_t *)tmpbuffer;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
- memset(bp->data, 0, XFS_LBSIZE(dp->i_mount));
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
+ memset(bp->b_addr, 0, XFS_LBSIZE(dp->i_mount));
/*
* Clean out the prior contents of the attribute list.
@@ -631,30 +698,30 @@ xfs_attr_leaf_to_node(xfs_da_args_t *args)
xfs_attr_leafblock_t *leaf;
xfs_da_intnode_t *node;
xfs_inode_t *dp;
- xfs_dabuf_t *bp1, *bp2;
+ struct xfs_buf *bp1, *bp2;
xfs_dablk_t blkno;
int error;
+ trace_xfs_attr_leaf_to_node(args);
+
dp = args->dp;
bp1 = bp2 = NULL;
error = xfs_da_grow_inode(args, &blkno);
if (error)
goto out;
- error = xfs_da_read_buf(args->trans, args->dp, 0, -1, &bp1,
- XFS_ATTR_FORK);
+ error = xfs_attr_leaf_read(args->trans, args->dp, 0, -1, &bp1);
if (error)
goto out;
- ASSERT(bp1 != NULL);
+
bp2 = NULL;
error = xfs_da_get_buf(args->trans, args->dp, blkno, -1, &bp2,
XFS_ATTR_FORK);
if (error)
goto out;
- ASSERT(bp2 != NULL);
- memcpy(bp2->data, bp1->data, XFS_LBSIZE(dp->i_mount));
- xfs_da_buf_done(bp1);
+ bp2->b_ops = bp1->b_ops;
+ memcpy(bp2->b_addr, bp1->b_addr, XFS_LBSIZE(dp->i_mount));
bp1 = NULL;
- xfs_da_log_buf(args->trans, bp2, 0, XFS_LBSIZE(dp->i_mount) - 1);
+ xfs_trans_log_buf(args->trans, bp2, 0, XFS_LBSIZE(dp->i_mount) - 1);
/*
* Set up the new root node.
@@ -662,21 +729,17 @@ xfs_attr_leaf_to_node(xfs_da_args_t *args)
error = xfs_da_node_create(args, 0, 1, &bp1, XFS_ATTR_FORK);
if (error)
goto out;
- node = bp1->data;
- leaf = bp2->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ node = bp1->b_addr;
+ leaf = bp2->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
/* both on-disk, don't endian-flip twice */
node->btree[0].hashval =
leaf->entries[be16_to_cpu(leaf->hdr.count)-1 ].hashval;
node->btree[0].before = cpu_to_be32(blkno);
node->hdr.count = cpu_to_be16(1);
- xfs_da_log_buf(args->trans, bp1, 0, XFS_LBSIZE(dp->i_mount) - 1);
+ xfs_trans_log_buf(args->trans, bp1, 0, XFS_LBSIZE(dp->i_mount) - 1);
error = 0;
out:
- if (bp1)
- xfs_da_buf_done(bp1);
- if (bp2)
- xfs_da_buf_done(bp2);
return(error);
}
@@ -690,22 +753,27 @@ out:
* or a leaf in a node attribute list.
*/
STATIC int
-xfs_attr_leaf_create(xfs_da_args_t *args, xfs_dablk_t blkno, xfs_dabuf_t **bpp)
+xfs_attr_leaf_create(
+ xfs_da_args_t *args,
+ xfs_dablk_t blkno,
+ struct xfs_buf **bpp)
{
xfs_attr_leafblock_t *leaf;
xfs_attr_leaf_hdr_t *hdr;
xfs_inode_t *dp;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
int error;
+ trace_xfs_attr_leaf_create(args);
+
dp = args->dp;
ASSERT(dp != NULL);
error = xfs_da_get_buf(args->trans, args->dp, blkno, -1, &bp,
XFS_ATTR_FORK);
if (error)
return(error);
- ASSERT(bp != NULL);
- leaf = bp->data;
+ bp->b_ops = &xfs_attr_leaf_buf_ops;
+ leaf = bp->b_addr;
memset((char *)leaf, 0, XFS_LBSIZE(dp->i_mount));
hdr = &leaf->hdr;
hdr->info.magic = cpu_to_be16(XFS_ATTR_LEAF_MAGIC);
@@ -719,7 +787,7 @@ xfs_attr_leaf_create(xfs_da_args_t *args, xfs_dablk_t blkno, xfs_dabuf_t **bpp)
hdr->freemap[0].size = cpu_to_be16(be16_to_cpu(hdr->firstused) -
sizeof(xfs_attr_leaf_hdr_t));
- xfs_da_log_buf(args->trans, bp, 0, XFS_LBSIZE(dp->i_mount) - 1);
+ xfs_trans_log_buf(args->trans, bp, 0, XFS_LBSIZE(dp->i_mount) - 1);
*bpp = bp;
return(0);
@@ -735,6 +803,8 @@ xfs_attr_leaf_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
xfs_dablk_t blkno;
int error;
+ trace_xfs_attr_leaf_split(state->args);
+
/*
* Allocate space for a new leaf node.
*/
@@ -764,10 +834,13 @@ xfs_attr_leaf_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
*
* Insert the "new" entry in the correct block.
*/
- if (state->inleaf)
+ if (state->inleaf) {
+ trace_xfs_attr_leaf_add_old(state->args);
error = xfs_attr_leaf_add(oldblk->bp, state->args);
- else
+ } else {
+ trace_xfs_attr_leaf_add_new(state->args);
error = xfs_attr_leaf_add(newblk->bp, state->args);
+ }
/*
* Update last hashval in each block since we added the name.
@@ -781,15 +854,19 @@ xfs_attr_leaf_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
* Add a name to the leaf attribute list structure.
*/
int
-xfs_attr_leaf_add(xfs_dabuf_t *bp, xfs_da_args_t *args)
+xfs_attr_leaf_add(
+ struct xfs_buf *bp,
+ struct xfs_da_args *args)
{
xfs_attr_leafblock_t *leaf;
xfs_attr_leaf_hdr_t *hdr;
xfs_attr_leaf_map_t *map;
int tablesize, entsize, sum, tmp, i;
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ trace_xfs_attr_leaf_add(args);
+
+ leaf = bp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
ASSERT((args->index >= 0)
&& (args->index <= be16_to_cpu(leaf->hdr.count)));
hdr = &leaf->hdr;
@@ -833,7 +910,7 @@ xfs_attr_leaf_add(xfs_dabuf_t *bp, xfs_da_args_t *args)
* Compact the entries to coalesce free space.
* This may change the hdr->count via dropping INCOMPLETE entries.
*/
- xfs_attr_leaf_compact(args->trans, bp);
+ xfs_attr_leaf_compact(args, bp);
/*
* After compaction, the block is guaranteed to have only one
@@ -850,7 +927,10 @@ xfs_attr_leaf_add(xfs_dabuf_t *bp, xfs_da_args_t *args)
* Add a name to a leaf attribute list structure.
*/
STATIC int
-xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex)
+xfs_attr_leaf_add_work(
+ struct xfs_buf *bp,
+ xfs_da_args_t *args,
+ int mapindex)
{
xfs_attr_leafblock_t *leaf;
xfs_attr_leaf_hdr_t *hdr;
@@ -861,8 +941,10 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex)
xfs_mount_t *mp;
int tmp, i;
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ trace_xfs_attr_leaf_add_work(args);
+
+ leaf = bp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
hdr = &leaf->hdr;
ASSERT((mapindex >= 0) && (mapindex < XFS_ATTR_LEAF_MAPSIZE));
ASSERT((args->index >= 0) && (args->index <= be16_to_cpu(hdr->count)));
@@ -875,7 +957,7 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex)
tmp = be16_to_cpu(hdr->count) - args->index;
tmp *= sizeof(xfs_attr_leaf_entry_t);
memmove((char *)(entry+1), (char *)entry, tmp);
- xfs_da_log_buf(args->trans, bp,
+ xfs_trans_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, entry, tmp + sizeof(*entry)));
}
be16_add_cpu(&hdr->count, 1);
@@ -907,7 +989,7 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex)
args->index2++;
}
}
- xfs_da_log_buf(args->trans, bp,
+ xfs_trans_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, entry, sizeof(*entry)));
ASSERT((args->index == 0) ||
(be32_to_cpu(entry->hashval) >= be32_to_cpu((entry-1)->hashval)));
@@ -915,8 +997,6 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex)
(be32_to_cpu(entry->hashval) <= be32_to_cpu((entry+1)->hashval)));
/*
- * Copy the attribute name and value into the new space.
- *
* For "remote" attribute values, simply note that we need to
* allocate space for the "remote" value. We can't actually
* allocate the extents in this transaction, and we can't decide
@@ -941,7 +1021,7 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex)
args->rmtblkno = 1;
args->rmtblkcnt = XFS_B_TO_FSB(mp, args->valuelen);
}
- xfs_da_log_buf(args->trans, bp,
+ xfs_trans_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, xfs_attr_leaf_name(leaf, args->index),
xfs_attr_leaf_entsize(leaf, args->index)));
@@ -965,7 +1045,7 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex)
}
}
be16_add_cpu(&hdr->usedbytes, xfs_attr_leaf_entsize(leaf, args->index));
- xfs_da_log_buf(args->trans, bp,
+ xfs_trans_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, hdr, sizeof(*hdr)));
return(0);
}
@@ -974,24 +1054,28 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex)
* Garbage collect a leaf attribute list block by copying it to a new buffer.
*/
STATIC void
-xfs_attr_leaf_compact(xfs_trans_t *trans, xfs_dabuf_t *bp)
+xfs_attr_leaf_compact(
+ struct xfs_da_args *args,
+ struct xfs_buf *bp)
{
- xfs_attr_leafblock_t *leaf_s, *leaf_d;
- xfs_attr_leaf_hdr_t *hdr_s, *hdr_d;
- xfs_mount_t *mp;
- char *tmpbuffer;
+ xfs_attr_leafblock_t *leaf_s, *leaf_d;
+ xfs_attr_leaf_hdr_t *hdr_s, *hdr_d;
+ struct xfs_trans *trans = args->trans;
+ struct xfs_mount *mp = trans->t_mountp;
+ char *tmpbuffer;
+
+ trace_xfs_attr_leaf_compact(args);
- mp = trans->t_mountp;
tmpbuffer = kmem_alloc(XFS_LBSIZE(mp), KM_SLEEP);
ASSERT(tmpbuffer != NULL);
- memcpy(tmpbuffer, bp->data, XFS_LBSIZE(mp));
- memset(bp->data, 0, XFS_LBSIZE(mp));
+ memcpy(tmpbuffer, bp->b_addr, XFS_LBSIZE(mp));
+ memset(bp->b_addr, 0, XFS_LBSIZE(mp));
/*
* Copy basic information
*/
leaf_s = (xfs_attr_leafblock_t *)tmpbuffer;
- leaf_d = bp->data;
+ leaf_d = bp->b_addr;
hdr_s = &leaf_s->hdr;
hdr_d = &leaf_d->hdr;
hdr_d->info = hdr_s->info; /* struct copy */
@@ -1014,7 +1098,7 @@ xfs_attr_leaf_compact(xfs_trans_t *trans, xfs_dabuf_t *bp)
*/
xfs_attr_leaf_moveents(leaf_s, 0, leaf_d, 0,
be16_to_cpu(hdr_s->count), mp);
- xfs_da_log_buf(trans, bp, 0, XFS_LBSIZE(mp) - 1);
+ xfs_trans_log_buf(trans, bp, 0, XFS_LBSIZE(mp) - 1);
kmem_free(tmpbuffer);
}
@@ -1046,12 +1130,15 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
*/
ASSERT(blk1->magic == XFS_ATTR_LEAF_MAGIC);
ASSERT(blk2->magic == XFS_ATTR_LEAF_MAGIC);
- leaf1 = blk1->bp->data;
- leaf2 = blk2->bp->data;
- ASSERT(be16_to_cpu(leaf1->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
- ASSERT(be16_to_cpu(leaf2->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ leaf1 = blk1->bp->b_addr;
+ leaf2 = blk2->bp->b_addr;
+ ASSERT(leaf1->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
+ ASSERT(leaf2->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
+ ASSERT(leaf2->hdr.count == 0);
args = state->args;
+ trace_xfs_attr_leaf_rebalance(args);
+
/*
* Check ordering of blocks, reverse if it makes things simpler.
*
@@ -1063,8 +1150,8 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
tmp_blk = blk1;
blk1 = blk2;
blk2 = tmp_blk;
- leaf1 = blk1->bp->data;
- leaf2 = blk2->bp->data;
+ leaf1 = blk1->bp->b_addr;
+ leaf2 = blk2->bp->b_addr;
swap = 1;
}
hdr1 = &leaf1->hdr;
@@ -1101,9 +1188,8 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
max = be16_to_cpu(hdr2->firstused)
- sizeof(xfs_attr_leaf_hdr_t);
max -= be16_to_cpu(hdr2->count) * sizeof(xfs_attr_leaf_entry_t);
- if (space > max) {
- xfs_attr_leaf_compact(args->trans, blk2->bp);
- }
+ if (space > max)
+ xfs_attr_leaf_compact(args, blk2->bp);
/*
* Move high entries from leaf1 to low end of leaf2.
@@ -1111,13 +1197,14 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
xfs_attr_leaf_moveents(leaf1, be16_to_cpu(hdr1->count) - count,
leaf2, 0, count, state->mp);
- xfs_da_log_buf(args->trans, blk1->bp, 0, state->blocksize-1);
- xfs_da_log_buf(args->trans, blk2->bp, 0, state->blocksize-1);
+ xfs_trans_log_buf(args->trans, blk1->bp, 0, state->blocksize-1);
+ xfs_trans_log_buf(args->trans, blk2->bp, 0, state->blocksize-1);
} else if (count > be16_to_cpu(hdr1->count)) {
/*
* I assert that since all callers pass in an empty
* second buffer, this code should never execute.
*/
+ ASSERT(0);
/*
* Figure the total bytes to be added to the destination leaf.
@@ -1133,9 +1220,8 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
max = be16_to_cpu(hdr1->firstused)
- sizeof(xfs_attr_leaf_hdr_t);
max -= be16_to_cpu(hdr1->count) * sizeof(xfs_attr_leaf_entry_t);
- if (space > max) {
- xfs_attr_leaf_compact(args->trans, blk1->bp);
- }
+ if (space > max)
+ xfs_attr_leaf_compact(args, blk1->bp);
/*
* Move low entries from leaf2 to high end of leaf1.
@@ -1143,8 +1229,8 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
xfs_attr_leaf_moveents(leaf2, 0, leaf1,
be16_to_cpu(hdr1->count), count, state->mp);
- xfs_da_log_buf(args->trans, blk1->bp, 0, state->blocksize-1);
- xfs_da_log_buf(args->trans, blk2->bp, 0, state->blocksize-1);
+ xfs_trans_log_buf(args->trans, blk1->bp, 0, state->blocksize-1);
+ xfs_trans_log_buf(args->trans, blk2->bp, 0, state->blocksize-1);
}
/*
@@ -1179,10 +1265,24 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
args->index2 = 0;
args->blkno2 = blk2->blkno;
} else {
+ /*
+ * On a double leaf split, the original attr location
+ * is already stored in blkno2/index2, so don't
+ * overwrite it overwise we corrupt the tree.
+ */
blk2->index = blk1->index
- be16_to_cpu(leaf1->hdr.count);
- args->index = args->index2 = blk2->index;
- args->blkno = args->blkno2 = blk2->blkno;
+ args->index = blk2->index;
+ args->blkno = blk2->blkno;
+ if (!state->extravalid) {
+ /*
+ * set the new attr location to match the old
+ * one and let the higher level split code
+ * decide where in the leaf to place it.
+ */
+ args->index2 = blk2->index;
+ args->blkno2 = blk2->blkno;
+ }
}
} else {
ASSERT(state->inleaf == 1);
@@ -1213,8 +1313,8 @@ xfs_attr_leaf_figure_balance(xfs_da_state_t *state,
/*
* Set up environment.
*/
- leaf1 = blk1->bp->data;
- leaf2 = blk2->bp->data;
+ leaf1 = blk1->bp->b_addr;
+ leaf2 = blk2->bp->b_addr;
hdr1 = &leaf1->hdr;
hdr2 = &leaf2->hdr;
foundit = 0;
@@ -1316,7 +1416,9 @@ xfs_attr_leaf_toosmall(xfs_da_state_t *state, int *action)
xfs_da_blkinfo_t *info;
int count, bytes, forward, error, retval, i;
xfs_dablk_t blkno;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
+
+ trace_xfs_attr_leaf_toosmall(state->args);
/*
* Check for the degenerate case of the block being over 50% full.
@@ -1324,8 +1426,8 @@ xfs_attr_leaf_toosmall(xfs_da_state_t *state, int *action)
* to coalesce with a sibling.
*/
blk = &state->path.blk[ state->path.active-1 ];
- info = blk->bp->data;
- ASSERT(be16_to_cpu(info->magic) == XFS_ATTR_LEAF_MAGIC);
+ info = blk->bp->b_addr;
+ ASSERT(info->magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
leaf = (xfs_attr_leafblock_t *)info;
count = be16_to_cpu(leaf->hdr.count);
bytes = sizeof(xfs_attr_leaf_hdr_t) +
@@ -1377,23 +1479,21 @@ xfs_attr_leaf_toosmall(xfs_da_state_t *state, int *action)
blkno = be32_to_cpu(info->back);
if (blkno == 0)
continue;
- error = xfs_da_read_buf(state->args->trans, state->args->dp,
- blkno, -1, &bp, XFS_ATTR_FORK);
+ error = xfs_attr_leaf_read(state->args->trans, state->args->dp,
+ blkno, -1, &bp);
if (error)
return(error);
- ASSERT(bp != NULL);
leaf = (xfs_attr_leafblock_t *)info;
count = be16_to_cpu(leaf->hdr.count);
bytes = state->blocksize - (state->blocksize>>2);
bytes -= be16_to_cpu(leaf->hdr.usedbytes);
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ leaf = bp->b_addr;
count += be16_to_cpu(leaf->hdr.count);
bytes -= be16_to_cpu(leaf->hdr.usedbytes);
bytes -= count * sizeof(xfs_attr_leaf_entry_t);
bytes -= sizeof(xfs_attr_leaf_hdr_t);
- xfs_da_brelse(state->args->trans, bp);
+ xfs_trans_brelse(state->args->trans, bp);
if (bytes >= 0)
break; /* fits with at least 25% to spare */
}
@@ -1431,7 +1531,9 @@ xfs_attr_leaf_toosmall(xfs_da_state_t *state, int *action)
* If two leaves are 37% full, when combined they will leave 25% free.
*/
int
-xfs_attr_leaf_remove(xfs_dabuf_t *bp, xfs_da_args_t *args)
+xfs_attr_leaf_remove(
+ struct xfs_buf *bp,
+ xfs_da_args_t *args)
{
xfs_attr_leafblock_t *leaf;
xfs_attr_leaf_hdr_t *hdr;
@@ -1441,8 +1543,10 @@ xfs_attr_leaf_remove(xfs_dabuf_t *bp, xfs_da_args_t *args)
int tablesize, tmp, i;
xfs_mount_t *mp;
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ trace_xfs_attr_leaf_remove(args);
+
+ leaf = bp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
hdr = &leaf->hdr;
mp = args->trans->t_mountp;
ASSERT((be16_to_cpu(hdr->count) > 0)
@@ -1534,7 +1638,7 @@ xfs_attr_leaf_remove(xfs_dabuf_t *bp, xfs_da_args_t *args)
*/
memset(xfs_attr_leaf_name(leaf, args->index), 0, entsize);
be16_add_cpu(&hdr->usedbytes, -entsize);
- xfs_da_log_buf(args->trans, bp,
+ xfs_trans_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, xfs_attr_leaf_name(leaf, args->index),
entsize));
@@ -1542,7 +1646,7 @@ xfs_attr_leaf_remove(xfs_dabuf_t *bp, xfs_da_args_t *args)
* sizeof(xfs_attr_leaf_entry_t);
memmove((char *)entry, (char *)(entry+1), tmp);
be16_add_cpu(&hdr->count, -1);
- xfs_da_log_buf(args->trans, bp,
+ xfs_trans_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, entry, tmp + sizeof(*entry)));
entry = &leaf->entries[be16_to_cpu(hdr->count)];
memset((char *)entry, 0, sizeof(xfs_attr_leaf_entry_t));
@@ -1572,7 +1676,7 @@ xfs_attr_leaf_remove(xfs_dabuf_t *bp, xfs_da_args_t *args)
} else {
hdr->holes = 1; /* mark as needing compaction */
}
- xfs_da_log_buf(args->trans, bp,
+ xfs_trans_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, hdr, sizeof(*hdr)));
/*
@@ -1597,16 +1701,18 @@ xfs_attr_leaf_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
xfs_mount_t *mp;
char *tmpbuffer;
+ trace_xfs_attr_leaf_unbalance(state->args);
+
/*
* Set up environment.
*/
mp = state->mp;
ASSERT(drop_blk->magic == XFS_ATTR_LEAF_MAGIC);
ASSERT(save_blk->magic == XFS_ATTR_LEAF_MAGIC);
- drop_leaf = drop_blk->bp->data;
- save_leaf = save_blk->bp->data;
- ASSERT(be16_to_cpu(drop_leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
- ASSERT(be16_to_cpu(save_leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ drop_leaf = drop_blk->bp->b_addr;
+ save_leaf = save_blk->bp->b_addr;
+ ASSERT(drop_leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
+ ASSERT(save_leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
drop_hdr = &drop_leaf->hdr;
save_hdr = &save_leaf->hdr;
@@ -1669,7 +1775,7 @@ xfs_attr_leaf_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
kmem_free(tmpbuffer);
}
- xfs_da_log_buf(state->args->trans, save_blk->bp, 0,
+ xfs_trans_log_buf(state->args->trans, save_blk->bp, 0,
state->blocksize - 1);
/*
@@ -1697,7 +1803,9 @@ xfs_attr_leaf_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
* Don't change the args->value unless we find the attribute.
*/
int
-xfs_attr_leaf_lookup_int(xfs_dabuf_t *bp, xfs_da_args_t *args)
+xfs_attr_leaf_lookup_int(
+ struct xfs_buf *bp,
+ xfs_da_args_t *args)
{
xfs_attr_leafblock_t *leaf;
xfs_attr_leaf_entry_t *entry;
@@ -1706,8 +1814,10 @@ xfs_attr_leaf_lookup_int(xfs_dabuf_t *bp, xfs_da_args_t *args)
int probe, span;
xfs_dahash_t hashval;
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ trace_xfs_attr_leaf_lookup(args);
+
+ leaf = bp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
ASSERT(be16_to_cpu(leaf->hdr.count)
< (XFS_LBSIZE(args->dp->i_mount)/8));
@@ -1802,7 +1912,9 @@ xfs_attr_leaf_lookup_int(xfs_dabuf_t *bp, xfs_da_args_t *args)
* list structure.
*/
int
-xfs_attr_leaf_getvalue(xfs_dabuf_t *bp, xfs_da_args_t *args)
+xfs_attr_leaf_getvalue(
+ struct xfs_buf *bp,
+ xfs_da_args_t *args)
{
int valuelen;
xfs_attr_leafblock_t *leaf;
@@ -1810,8 +1922,8 @@ xfs_attr_leaf_getvalue(xfs_dabuf_t *bp, xfs_da_args_t *args)
xfs_attr_leaf_name_local_t *name_loc;
xfs_attr_leaf_name_remote_t *name_rmt;
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ leaf = bp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
ASSERT(be16_to_cpu(leaf->hdr.count)
< (XFS_LBSIZE(args->dp->i_mount)/8));
ASSERT(args->index < be16_to_cpu(leaf->hdr.count));
@@ -1879,8 +1991,8 @@ xfs_attr_leaf_moveents(xfs_attr_leafblock_t *leaf_s, int start_s,
/*
* Set up environment.
*/
- ASSERT(be16_to_cpu(leaf_s->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
- ASSERT(be16_to_cpu(leaf_d->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(leaf_s->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
+ ASSERT(leaf_d->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
hdr_s = &leaf_s->hdr;
hdr_d = &leaf_d->hdr;
ASSERT((be16_to_cpu(hdr_s->count) > 0) &&
@@ -2008,14 +2120,16 @@ xfs_attr_leaf_moveents(xfs_attr_leafblock_t *leaf_s, int start_s,
* Return 0 unless leaf2 should go before leaf1.
*/
int
-xfs_attr_leaf_order(xfs_dabuf_t *leaf1_bp, xfs_dabuf_t *leaf2_bp)
+xfs_attr_leaf_order(
+ struct xfs_buf *leaf1_bp,
+ struct xfs_buf *leaf2_bp)
{
xfs_attr_leafblock_t *leaf1, *leaf2;
- leaf1 = leaf1_bp->data;
- leaf2 = leaf2_bp->data;
- ASSERT((be16_to_cpu(leaf1->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC) &&
- (be16_to_cpu(leaf2->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC));
+ leaf1 = leaf1_bp->b_addr;
+ leaf2 = leaf2_bp->b_addr;
+ ASSERT((leaf1->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC)) &&
+ (leaf2->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC)));
if ((be16_to_cpu(leaf1->hdr.count) > 0) &&
(be16_to_cpu(leaf2->hdr.count) > 0) &&
((be32_to_cpu(leaf2->entries[0].hashval) <
@@ -2033,12 +2147,14 @@ xfs_attr_leaf_order(xfs_dabuf_t *leaf1_bp, xfs_dabuf_t *leaf2_bp)
* Pick up the last hashvalue from a leaf block.
*/
xfs_dahash_t
-xfs_attr_leaf_lasthash(xfs_dabuf_t *bp, int *count)
+xfs_attr_leaf_lasthash(
+ struct xfs_buf *bp,
+ int *count)
{
xfs_attr_leafblock_t *leaf;
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ leaf = bp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
if (count)
*count = be16_to_cpu(leaf->hdr.count);
if (!leaf->hdr.count)
@@ -2057,7 +2173,7 @@ xfs_attr_leaf_entsize(xfs_attr_leafblock_t *leaf, int index)
xfs_attr_leaf_name_remote_t *name_rmt;
int size;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
if (leaf->entries[index].flags & XFS_ATTR_LOCAL) {
name_loc = xfs_attr_leaf_name_local(leaf, index);
size = xfs_attr_leaf_entsize_local(name_loc->namelen,
@@ -2107,7 +2223,7 @@ xfs_attr_leaf_clearflag(xfs_da_args_t *args)
xfs_attr_leafblock_t *leaf;
xfs_attr_leaf_entry_t *entry;
xfs_attr_leaf_name_remote_t *name_rmt;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
int error;
#ifdef DEBUG
xfs_attr_leaf_name_local_t *name_loc;
@@ -2115,18 +2231,15 @@ xfs_attr_leaf_clearflag(xfs_da_args_t *args)
char *name;
#endif /* DEBUG */
+ trace_xfs_attr_leaf_clearflag(args);
/*
* Set up the operation.
*/
- error = xfs_da_read_buf(args->trans, args->dp, args->blkno, -1, &bp,
- XFS_ATTR_FORK);
- if (error) {
+ error = xfs_attr_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+ if (error)
return(error);
- }
- ASSERT(bp != NULL);
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ leaf = bp->b_addr;
ASSERT(args->index < be16_to_cpu(leaf->hdr.count));
ASSERT(args->index >= 0);
entry = &leaf->entries[ args->index ];
@@ -2148,7 +2261,7 @@ xfs_attr_leaf_clearflag(xfs_da_args_t *args)
#endif /* DEBUG */
entry->flags &= ~XFS_ATTR_INCOMPLETE;
- xfs_da_log_buf(args->trans, bp,
+ xfs_trans_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, entry, sizeof(*entry)));
if (args->rmtblkno) {
@@ -2156,10 +2269,9 @@ xfs_attr_leaf_clearflag(xfs_da_args_t *args)
name_rmt = xfs_attr_leaf_name_remote(leaf, args->index);
name_rmt->valueblk = cpu_to_be32(args->rmtblkno);
name_rmt->valuelen = cpu_to_be32(args->valuelen);
- xfs_da_log_buf(args->trans, bp,
+ xfs_trans_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, name_rmt, sizeof(*name_rmt)));
}
- xfs_da_buf_done(bp);
/*
* Commit the flag value change and start the next trans in series.
@@ -2176,37 +2288,34 @@ xfs_attr_leaf_setflag(xfs_da_args_t *args)
xfs_attr_leafblock_t *leaf;
xfs_attr_leaf_entry_t *entry;
xfs_attr_leaf_name_remote_t *name_rmt;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
int error;
+ trace_xfs_attr_leaf_setflag(args);
+
/*
* Set up the operation.
*/
- error = xfs_da_read_buf(args->trans, args->dp, args->blkno, -1, &bp,
- XFS_ATTR_FORK);
- if (error) {
+ error = xfs_attr_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+ if (error)
return(error);
- }
- ASSERT(bp != NULL);
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ leaf = bp->b_addr;
ASSERT(args->index < be16_to_cpu(leaf->hdr.count));
ASSERT(args->index >= 0);
entry = &leaf->entries[ args->index ];
ASSERT((entry->flags & XFS_ATTR_INCOMPLETE) == 0);
entry->flags |= XFS_ATTR_INCOMPLETE;
- xfs_da_log_buf(args->trans, bp,
+ xfs_trans_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, entry, sizeof(*entry)));
if ((entry->flags & XFS_ATTR_LOCAL) == 0) {
name_rmt = xfs_attr_leaf_name_remote(leaf, args->index);
name_rmt->valueblk = 0;
name_rmt->valuelen = 0;
- xfs_da_log_buf(args->trans, bp,
+ xfs_trans_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, name_rmt, sizeof(*name_rmt)));
}
- xfs_da_buf_done(bp);
/*
* Commit the flag value change and start the next trans in series.
@@ -2227,7 +2336,7 @@ xfs_attr_leaf_flipflags(xfs_da_args_t *args)
xfs_attr_leafblock_t *leaf1, *leaf2;
xfs_attr_leaf_entry_t *entry1, *entry2;
xfs_attr_leaf_name_remote_t *name_rmt;
- xfs_dabuf_t *bp1, *bp2;
+ struct xfs_buf *bp1, *bp2;
int error;
#ifdef DEBUG
xfs_attr_leaf_name_local_t *name_loc;
@@ -2235,38 +2344,33 @@ xfs_attr_leaf_flipflags(xfs_da_args_t *args)
char *name1, *name2;
#endif /* DEBUG */
+ trace_xfs_attr_leaf_flipflags(args);
+
/*
* Read the block containing the "old" attr
*/
- error = xfs_da_read_buf(args->trans, args->dp, args->blkno, -1, &bp1,
- XFS_ATTR_FORK);
- if (error) {
- return(error);
- }
- ASSERT(bp1 != NULL);
+ error = xfs_attr_leaf_read(args->trans, args->dp, args->blkno, -1, &bp1);
+ if (error)
+ return error;
/*
* Read the block containing the "new" attr, if it is different
*/
if (args->blkno2 != args->blkno) {
- error = xfs_da_read_buf(args->trans, args->dp, args->blkno2,
- -1, &bp2, XFS_ATTR_FORK);
- if (error) {
- return(error);
- }
- ASSERT(bp2 != NULL);
+ error = xfs_attr_leaf_read(args->trans, args->dp, args->blkno2,
+ -1, &bp2);
+ if (error)
+ return error;
} else {
bp2 = bp1;
}
- leaf1 = bp1->data;
- ASSERT(be16_to_cpu(leaf1->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ leaf1 = bp1->b_addr;
ASSERT(args->index < be16_to_cpu(leaf1->hdr.count));
ASSERT(args->index >= 0);
entry1 = &leaf1->entries[ args->index ];
- leaf2 = bp2->data;
- ASSERT(be16_to_cpu(leaf2->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ leaf2 = bp2->b_addr;
ASSERT(args->index2 < be16_to_cpu(leaf2->hdr.count));
ASSERT(args->index2 >= 0);
entry2 = &leaf2->entries[ args->index2 ];
@@ -2299,30 +2403,27 @@ xfs_attr_leaf_flipflags(xfs_da_args_t *args)
ASSERT((entry2->flags & XFS_ATTR_INCOMPLETE) == 0);
entry1->flags &= ~XFS_ATTR_INCOMPLETE;
- xfs_da_log_buf(args->trans, bp1,
+ xfs_trans_log_buf(args->trans, bp1,
XFS_DA_LOGRANGE(leaf1, entry1, sizeof(*entry1)));
if (args->rmtblkno) {
ASSERT((entry1->flags & XFS_ATTR_LOCAL) == 0);
name_rmt = xfs_attr_leaf_name_remote(leaf1, args->index);
name_rmt->valueblk = cpu_to_be32(args->rmtblkno);
name_rmt->valuelen = cpu_to_be32(args->valuelen);
- xfs_da_log_buf(args->trans, bp1,
+ xfs_trans_log_buf(args->trans, bp1,
XFS_DA_LOGRANGE(leaf1, name_rmt, sizeof(*name_rmt)));
}
entry2->flags |= XFS_ATTR_INCOMPLETE;
- xfs_da_log_buf(args->trans, bp2,
+ xfs_trans_log_buf(args->trans, bp2,
XFS_DA_LOGRANGE(leaf2, entry2, sizeof(*entry2)));
if ((entry2->flags & XFS_ATTR_LOCAL) == 0) {
name_rmt = xfs_attr_leaf_name_remote(leaf2, args->index2);
name_rmt->valueblk = 0;
name_rmt->valuelen = 0;
- xfs_da_log_buf(args->trans, bp2,
+ xfs_trans_log_buf(args->trans, bp2,
XFS_DA_LOGRANGE(leaf2, name_rmt, sizeof(*name_rmt)));
}
- xfs_da_buf_done(bp1);
- if (bp1 != bp2)
- xfs_da_buf_done(bp2);
/*
* Commit the flag value change and start the next trans in series.
diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index 5a626b0..b328a0b 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -18,10 +18,6 @@
#include <xfs.h>
-#ifdef DEBUG
-STATIC void
-xfs_bmap_check_leaf_extents(xfs_btree_cur_t *cur, xfs_inode_t *ip, int whichfork);
-#endif
kmem_zone_t *xfs_bmap_free_item_zone;
@@ -29,6 +25,16 @@ kmem_zone_t *xfs_bmap_free_item_zone;
* Prototypes for internal bmap routines.
*/
+#ifdef DEBUG
+STATIC void
+xfs_bmap_check_leaf_extents(
+ struct xfs_btree_cur *cur,
+ struct xfs_inode *ip,
+ int whichfork);
+#else
+#define xfs_bmap_check_leaf_extents(cur, ip, whichfork) do { } while (0)
+#endif
+
/*
* Called from xfs_bmap_add_attrfork to handle extents format files.
@@ -53,75 +59,6 @@ xfs_bmap_add_attrfork_local(
int *flags); /* inode logging flags */
/*
- * Called by xfs_bmapi to update file extent records and the btree
- * after allocating space (or doing a delayed allocation).
- */
-STATIC int /* error */
-xfs_bmap_add_extent(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_extnum_t idx, /* extent number to update/insert */
- xfs_btree_cur_t **curp, /* if *curp is null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to add to file extents */
- xfs_fsblock_t *first, /* pointer to firstblock variable */
- xfs_bmap_free_t *flist, /* list of extents to be freed */
- int *logflagsp, /* inode logging flags */
- int whichfork, /* data or attr fork */
- int rsvd); /* OK to allocate reserved blocks */
-
-/*
- * Called by xfs_bmap_add_extent to handle cases converting a delayed
- * allocation to a real allocation.
- */
-STATIC int /* error */
-xfs_bmap_add_extent_delay_real(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_extnum_t idx, /* extent number to update/insert */
- xfs_btree_cur_t **curp, /* if *curp is null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to add to file extents */
- xfs_filblks_t *dnew, /* new delayed-alloc indirect blocks */
- xfs_fsblock_t *first, /* pointer to firstblock variable */
- xfs_bmap_free_t *flist, /* list of extents to be freed */
- int *logflagsp, /* inode logging flags */
- int rsvd); /* OK to allocate reserved blocks */
-
-/*
- * Called by xfs_bmap_add_extent to handle cases converting a hole
- * to a delayed allocation.
- */
-STATIC int /* error */
-xfs_bmap_add_extent_hole_delay(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_extnum_t idx, /* extent number to update/insert */
- xfs_bmbt_irec_t *new, /* new data to add to file extents */
- int *logflagsp,/* inode logging flags */
- int rsvd); /* OK to allocate reserved blocks */
-
-/*
- * Called by xfs_bmap_add_extent to handle cases converting a hole
- * to a real allocation.
- */
-STATIC int /* error */
-xfs_bmap_add_extent_hole_real(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_extnum_t idx, /* extent number to update/insert */
- xfs_btree_cur_t *cur, /* if null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to add to file extents */
- int *logflagsp, /* inode logging flags */
- int whichfork); /* data or attr fork */
-
-/*
- * Called by xfs_bmap_add_extent to handle cases converting an unwritten
- * allocation to a real allocation or vice versa.
- */
-STATIC int /* error */
-xfs_bmap_add_extent_unwritten_real(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_extnum_t idx, /* extent number to update/insert */
- xfs_btree_cur_t **curp, /* if *curp is null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to add to file extents */
- int *logflagsp); /* inode logging flags */
-
-/*
* xfs_bmap_alloc is called by xfs_bmapi to allocate an extent for a file.
* It figures out where to ask the underlying allocator to put the new extent.
*/
@@ -144,22 +81,6 @@ xfs_bmap_btree_to_extents(
int whichfork); /* data or attr fork */
/*
- * Called by xfs_bmapi to update file extent records and the btree
- * after removing space (or undoing a delayed allocation).
- */
-STATIC int /* error */
-xfs_bmap_del_extent(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_trans_t *tp, /* current trans pointer */
- xfs_extnum_t idx, /* extent number to update/insert */
- xfs_bmap_free_t *flist, /* list of extents to be freed */
- xfs_btree_cur_t *cur, /* if null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to add to file extents */
- int *logflagsp,/* inode logging flags */
- int whichfork, /* data or attr fork */
- int rsvd); /* OK to allocate reserved blocks */
-
-/*
* Convert an extents-format file into a btree-format file.
* The new file will have a root block (in the inode) and a single child block.
*/
@@ -189,19 +110,6 @@ xfs_bmap_local_to_extents(
int whichfork); /* data or attr fork */
/*
- * Check the last inode extent to determine whether this allocation will result
- * in blocks being allocated at the end of the file. When we allocate new data
- * blocks at the end of the file which do not start at the previous data block,
- * we will try to align the new blocks at stripe unit boundaries.
- */
-STATIC int /* error */
-xfs_bmap_isaeof(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_fileoff_t off, /* file offset in fsblocks */
- int whichfork, /* data or attribute fork */
- char *aeof); /* return value */
-
-/*
* Compute the worst-case number of indirect blocks that will be used
* for ip's delayed extent of length "len".
*/
@@ -283,7 +191,27 @@ xfs_bmbt_lookup_ge(
}
/*
-* Update the record referred to by cur to the value given
+ * Check if the inode needs to be converted to btree format.
+ */
+static inline bool xfs_bmap_needs_btree(struct xfs_inode *ip, int whichfork)
+{
+ return XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS &&
+ XFS_IFORK_NEXTENTS(ip, whichfork) >
+ XFS_IFORK_MAXEXT(ip, whichfork);
+}
+
+/*
+ * Check if the inode should be converted to extent format.
+ */
+static inline bool xfs_bmap_wants_extents(struct xfs_inode *ip, int whichfork)
+{
+ return XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE &&
+ XFS_IFORK_NEXTENTS(ip, whichfork) <=
+ XFS_IFORK_MAXEXT(ip, whichfork);
+}
+
+/*
+ * Update the record referred to by cur to the value given
* by [off, bno, len, state].
* This either works (return 0) or gets an EFSCORRUPTED error.
*/
@@ -388,7 +316,7 @@ xfs_bmap_add_attrfork_local(
if (ip->i_df.if_bytes <= XFS_IFORK_DSIZE(ip))
return 0;
- if ((ip->i_d.di_mode & S_IFMT) == S_IFDIR) {
+ if (S_ISDIR(ip->i_d.di_mode)) {
mp = ip->i_mount;
memset(&dargs, 0, sizeof(dargs));
dargs.dp = ip;
@@ -405,188 +333,13 @@ xfs_bmap_add_attrfork_local(
}
/*
- * Called by xfs_bmapi to update file extent records and the btree
- * after allocating space (or doing a delayed allocation).
- */
-STATIC int /* error */
-xfs_bmap_add_extent(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_extnum_t idx, /* extent number to update/insert */
- xfs_btree_cur_t **curp, /* if *curp is null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to add to file extents */
- xfs_fsblock_t *first, /* pointer to firstblock variable */
- xfs_bmap_free_t *flist, /* list of extents to be freed */
- int *logflagsp, /* inode logging flags */
- int whichfork, /* data or attr fork */
- int rsvd) /* OK to use reserved data blocks */
-{
- xfs_btree_cur_t *cur; /* btree cursor or null */
- xfs_filblks_t da_new; /* new count del alloc blocks used */
- xfs_filblks_t da_old; /* old count del alloc blocks used */
- int error; /* error return value */
- xfs_ifork_t *ifp; /* inode fork ptr */
- int logflags; /* returned value */
- xfs_extnum_t nextents; /* number of extents in file now */
-
- XFS_STATS_INC(xs_add_exlist);
- cur = *curp;
- ifp = XFS_IFORK_PTR(ip, whichfork);
- nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
- ASSERT(idx <= nextents);
- da_old = da_new = 0;
- error = 0;
- /*
- * This is the first extent added to a new/empty file.
- * Special case this one, so other routines get to assume there are
- * already extents in the list.
- */
- if (nextents == 0) {
- xfs_iext_insert(ip, 0, 1, new,
- whichfork == XFS_ATTR_FORK ? BMAP_ATTRFORK : 0);
-
- ASSERT(cur == NULL);
- ifp->if_lastex = 0;
- if (!isnullstartblock(new->br_startblock)) {
- XFS_IFORK_NEXT_SET(ip, whichfork, 1);
- logflags = XFS_ILOG_CORE | xfs_ilog_fext(whichfork);
- } else
- logflags = 0;
- }
- /*
- * Any kind of new delayed allocation goes here.
- */
- else if (isnullstartblock(new->br_startblock)) {
- if (cur)
- ASSERT((cur->bc_private.b.flags &
- XFS_BTCUR_BPRV_WASDEL) == 0);
- if ((error = xfs_bmap_add_extent_hole_delay(ip, idx, new,
- &logflags, rsvd)))
- goto done;
- }
- /*
- * Real allocation off the end of the file.
- */
- else if (idx == nextents) {
- if (cur)
- ASSERT((cur->bc_private.b.flags &
- XFS_BTCUR_BPRV_WASDEL) == 0);
- if ((error = xfs_bmap_add_extent_hole_real(ip, idx, cur, new,
- &logflags, whichfork)))
- goto done;
- } else {
- xfs_bmbt_irec_t prev; /* old extent at offset idx */
-
- /*
- * Get the record referred to by idx.
- */
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx), &prev);
- /*
- * If it's a real allocation record, and the new allocation ends
- * after the start of the referred to record, then we're filling
- * in a delayed or unwritten allocation with a real one, or
- * converting real back to unwritten.
- */
- if (!isnullstartblock(new->br_startblock) &&
- new->br_startoff + new->br_blockcount > prev.br_startoff) {
- if (prev.br_state != XFS_EXT_UNWRITTEN &&
- isnullstartblock(prev.br_startblock)) {
- da_old = startblockval(prev.br_startblock);
- if (cur)
- ASSERT(cur->bc_private.b.flags &
- XFS_BTCUR_BPRV_WASDEL);
- if ((error = xfs_bmap_add_extent_delay_real(ip,
- idx, &cur, new, &da_new, first, flist,
- &logflags, rsvd)))
- goto done;
- } else if (new->br_state == XFS_EXT_NORM) {
- ASSERT(new->br_state == XFS_EXT_NORM);
- if ((error = xfs_bmap_add_extent_unwritten_real(
- ip, idx, &cur, new, &logflags)))
- goto done;
- } else {
- ASSERT(new->br_state == XFS_EXT_UNWRITTEN);
- if ((error = xfs_bmap_add_extent_unwritten_real(
- ip, idx, &cur, new, &logflags)))
- goto done;
- }
- ASSERT(*curp == cur || *curp == NULL);
- }
- /*
- * Otherwise we're filling in a hole with an allocation.
- */
- else {
- if (cur)
- ASSERT((cur->bc_private.b.flags &
- XFS_BTCUR_BPRV_WASDEL) == 0);
- if ((error = xfs_bmap_add_extent_hole_real(ip, idx, cur,
- new, &logflags, whichfork)))
- goto done;
- }
- }
-
- ASSERT(*curp == cur || *curp == NULL);
- /*
- * Convert to a btree if necessary.
- */
- if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS &&
- XFS_IFORK_NEXTENTS(ip, whichfork) > ifp->if_ext_max) {
- int tmp_logflags; /* partial log flag return val */
-
- ASSERT(cur == NULL);
- error = xfs_bmap_extents_to_btree(ip->i_transp, ip, first,
- flist, &cur, da_old > 0, &tmp_logflags, whichfork);
- logflags |= tmp_logflags;
- if (error)
- goto done;
- }
- /*
- * Adjust for changes in reserved delayed indirect blocks.
- * Nothing to do for disk quotas here.
- */
- if (da_old || da_new) {
- xfs_filblks_t nblks;
-
- nblks = da_new;
- if (cur)
- nblks += cur->bc_private.b.allocated;
- ASSERT(nblks <= da_old);
- if (nblks < da_old)
- xfs_icsb_modify_counters(ip->i_mount, XFS_SBS_FDBLOCKS,
- (int64_t)(da_old - nblks), rsvd);
- }
- /*
- * Clear out the allocated field, done with it now in any case.
- */
- if (cur) {
- cur->bc_private.b.allocated = 0;
- *curp = cur;
- }
-done:
-#ifdef DEBUG
- if (!error)
- xfs_bmap_check_leaf_extents(*curp, ip, whichfork);
-#endif
- *logflagsp = logflags;
- return error;
-}
-
-/*
- * Called by xfs_bmap_add_extent to handle cases converting a delayed
- * allocation to a real allocation.
+ * Convert a delayed allocation to a real allocation.
*/
STATIC int /* error */
xfs_bmap_add_extent_delay_real(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_extnum_t idx, /* extent number to update/insert */
- xfs_btree_cur_t **curp, /* if *curp is null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to add to file extents */
- xfs_filblks_t *dnew, /* new delayed-alloc indirect blocks */
- xfs_fsblock_t *first, /* pointer to firstblock variable */
- xfs_bmap_free_t *flist, /* list of extents to be freed */
- int *logflagsp, /* inode logging flags */
- int rsvd) /* OK to use reserved data block allocation */
+ struct xfs_bmalloca *bma)
{
- xfs_btree_cur_t *cur; /* btree cursor */
+ struct xfs_bmbt_irec *new = &bma->got;
int diff; /* temp value */
xfs_bmbt_rec_host_t *ep; /* extent entry for idx */
int error; /* error return value */
@@ -597,10 +350,22 @@ xfs_bmap_add_extent_delay_real(
/* left is 0, right is 1, prev is 2 */
int rval=0; /* return value (logging flags) */
int state = 0;/* state bits, accessed thru macros */
- xfs_filblks_t temp=0; /* value for dnew calculations */
- xfs_filblks_t temp2=0;/* value for dnew calculations */
+ xfs_filblks_t da_new; /* new count del alloc blocks used */
+ xfs_filblks_t da_old; /* old count del alloc blocks used */
+ xfs_filblks_t temp=0; /* value for da_new calculations */
+ xfs_filblks_t temp2=0;/* value for da_new calculations */
int tmp_rval; /* partial logging flags */
+ ifp = XFS_IFORK_PTR(bma->ip, XFS_DATA_FORK);
+
+ ASSERT(bma->idx >= 0);
+ ASSERT(bma->idx <= ifp->if_bytes / sizeof(struct xfs_bmbt_rec));
+ ASSERT(!isnullstartblock(new->br_startblock));
+ ASSERT(!bma->cur ||
+ (bma->cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL));
+
+ XFS_STATS_INC(xs_add_exlist);
+
#define LEFT r[0]
#define RIGHT r[1]
#define PREV r[2]
@@ -608,14 +373,15 @@ xfs_bmap_add_extent_delay_real(
/*
* Set up a bunch of variables to make the tests simpler.
*/
- cur = *curp;
- ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
- ep = xfs_iext_get_ext(ifp, idx);
+ ep = xfs_iext_get_ext(ifp, bma->idx);
xfs_bmbt_get_all(ep, &PREV);
new_endoff = new->br_startoff + new->br_blockcount;
ASSERT(PREV.br_startoff <= new->br_startoff);
ASSERT(PREV.br_startoff + PREV.br_blockcount >= new_endoff);
+ da_old = startblockval(PREV.br_startblock);
+ da_new = 0;
+
/*
* Set flags determining what part of the previous delayed allocation
* extent is being replaced by a real allocation.
@@ -629,9 +395,9 @@ xfs_bmap_add_extent_delay_real(
* Check and set flags if this segment has a left neighbor.
* Don't set contiguous if the combined extent would be too large.
*/
- if (idx > 0) {
+ if (bma->idx > 0) {
state |= BMAP_LEFT_VALID;
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx - 1), &LEFT);
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx - 1), &LEFT);
if (isnullstartblock(LEFT.br_startblock))
state |= BMAP_LEFT_DELAY;
@@ -649,9 +415,9 @@ xfs_bmap_add_extent_delay_real(
* Don't set contiguous if the combined extent would be too large.
* Also check for all-three-contiguous being too large.
*/
- if (idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) {
+ if (bma->idx < bma->ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) {
state |= BMAP_RIGHT_VALID;
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx + 1), &RIGHT);
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx + 1), &RIGHT);
if (isnullstartblock(RIGHT.br_startblock))
state |= BMAP_RIGHT_DELAY;
@@ -682,38 +448,41 @@ xfs_bmap_add_extent_delay_real(
* Filling in all of a previously delayed allocation extent.
* The left and right neighbors are both contiguous with new.
*/
- trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
+ bma->idx--;
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, bma->idx),
LEFT.br_blockcount + PREV.br_blockcount +
RIGHT.br_blockcount);
- trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
- xfs_iext_remove(ip, idx, 2, state);
- ip->i_df.if_lastex = idx - 1;
- ip->i_d.di_nextents--;
- if (cur == NULL)
+ xfs_iext_remove(bma->ip, bma->idx + 1, 2, state);
+ bma->ip->i_d.di_nextents--;
+ if (bma->cur == NULL)
rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
else {
rval = XFS_ILOG_CORE;
- if ((error = xfs_bmbt_lookup_eq(cur, RIGHT.br_startoff,
+ error = xfs_bmbt_lookup_eq(bma->cur, RIGHT.br_startoff,
RIGHT.br_startblock,
- RIGHT.br_blockcount, &i)))
+ RIGHT.br_blockcount, &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_btree_delete(cur, &i)))
+ error = xfs_btree_delete(bma->cur, &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_btree_decrement(cur, 0, &i)))
+ error = xfs_btree_decrement(bma->cur, 0, &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_bmbt_update(cur, LEFT.br_startoff,
+ error = xfs_bmbt_update(bma->cur, LEFT.br_startoff,
LEFT.br_startblock,
LEFT.br_blockcount +
PREV.br_blockcount +
- RIGHT.br_blockcount, LEFT.br_state)))
+ RIGHT.br_blockcount, LEFT.br_state);
+ if (error)
goto done;
}
- *dnew = 0;
break;
case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG:
@@ -721,29 +490,31 @@ xfs_bmap_add_extent_delay_real(
* Filling in all of a previously delayed allocation extent.
* The left neighbor is contiguous, the right is not.
*/
- trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
+ bma->idx--;
+
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, bma->idx),
LEFT.br_blockcount + PREV.br_blockcount);
- trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
- ip->i_df.if_lastex = idx - 1;
- xfs_iext_remove(ip, idx, 1, state);
- if (cur == NULL)
+ xfs_iext_remove(bma->ip, bma->idx + 1, 1, state);
+ if (bma->cur == NULL)
rval = XFS_ILOG_DEXT;
else {
rval = 0;
- if ((error = xfs_bmbt_lookup_eq(cur, LEFT.br_startoff,
+ error = xfs_bmbt_lookup_eq(bma->cur, LEFT.br_startoff,
LEFT.br_startblock, LEFT.br_blockcount,
- &i)))
+ &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_bmbt_update(cur, LEFT.br_startoff,
+ error = xfs_bmbt_update(bma->cur, LEFT.br_startoff,
LEFT.br_startblock,
LEFT.br_blockcount +
- PREV.br_blockcount, LEFT.br_state)))
+ PREV.br_blockcount, LEFT.br_state);
+ if (error)
goto done;
}
- *dnew = 0;
break;
case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
@@ -751,30 +522,30 @@ xfs_bmap_add_extent_delay_real(
* Filling in all of a previously delayed allocation extent.
* The right neighbor is contiguous, the left is not.
*/
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
xfs_bmbt_set_startblock(ep, new->br_startblock);
xfs_bmbt_set_blockcount(ep,
PREV.br_blockcount + RIGHT.br_blockcount);
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
- ip->i_df.if_lastex = idx;
- xfs_iext_remove(ip, idx + 1, 1, state);
- if (cur == NULL)
+ xfs_iext_remove(bma->ip, bma->idx + 1, 1, state);
+ if (bma->cur == NULL)
rval = XFS_ILOG_DEXT;
else {
rval = 0;
- if ((error = xfs_bmbt_lookup_eq(cur, RIGHT.br_startoff,
+ error = xfs_bmbt_lookup_eq(bma->cur, RIGHT.br_startoff,
RIGHT.br_startblock,
- RIGHT.br_blockcount, &i)))
+ RIGHT.br_blockcount, &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_bmbt_update(cur, PREV.br_startoff,
+ error = xfs_bmbt_update(bma->cur, PREV.br_startoff,
new->br_startblock,
PREV.br_blockcount +
- RIGHT.br_blockcount, PREV.br_state)))
+ RIGHT.br_blockcount, PREV.br_state);
+ if (error)
goto done;
}
- *dnew = 0;
break;
case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING:
@@ -783,27 +554,27 @@ xfs_bmap_add_extent_delay_real(
* Neither the left nor right neighbors are contiguous with
* the new one.
*/
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
xfs_bmbt_set_startblock(ep, new->br_startblock);
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
- ip->i_df.if_lastex = idx;
- ip->i_d.di_nextents++;
- if (cur == NULL)
+ bma->ip->i_d.di_nextents++;
+ if (bma->cur == NULL)
rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
else {
rval = XFS_ILOG_CORE;
- if ((error = xfs_bmbt_lookup_eq(cur, new->br_startoff,
+ error = xfs_bmbt_lookup_eq(bma->cur, new->br_startoff,
new->br_startblock, new->br_blockcount,
- &i)))
+ &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 0, done);
- cur->bc_rec.b.br_state = XFS_EXT_NORM;
- if ((error = xfs_btree_insert(cur, &i)))
+ bma->cur->bc_rec.b.br_state = XFS_EXT_NORM;
+ error = xfs_btree_insert(bma->cur, &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 1, done);
}
- *dnew = 0;
break;
case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG:
@@ -811,38 +582,40 @@ xfs_bmap_add_extent_delay_real(
* Filling in the first part of a previous delayed allocation.
* The left neighbor is contiguous.
*/
- trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx - 1, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, bma->idx - 1),
LEFT.br_blockcount + new->br_blockcount);
xfs_bmbt_set_startoff(ep,
PREV.br_startoff + new->br_blockcount);
- trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(bma->ip, bma->idx - 1, state, _THIS_IP_);
temp = PREV.br_blockcount - new->br_blockcount;
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
xfs_bmbt_set_blockcount(ep, temp);
- ip->i_df.if_lastex = idx - 1;
- if (cur == NULL)
+ if (bma->cur == NULL)
rval = XFS_ILOG_DEXT;
else {
rval = 0;
- if ((error = xfs_bmbt_lookup_eq(cur, LEFT.br_startoff,
+ error = xfs_bmbt_lookup_eq(bma->cur, LEFT.br_startoff,
LEFT.br_startblock, LEFT.br_blockcount,
- &i)))
+ &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_bmbt_update(cur, LEFT.br_startoff,
+ error = xfs_bmbt_update(bma->cur, LEFT.br_startoff,
LEFT.br_startblock,
LEFT.br_blockcount +
new->br_blockcount,
- LEFT.br_state)))
+ LEFT.br_state);
+ if (error)
goto done;
}
- temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
+ da_new = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(bma->ip, temp),
startblockval(PREV.br_startblock));
- xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
- *dnew = temp;
+ xfs_bmbt_set_startblock(ep, nullstartblock(da_new));
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+
+ bma->idx--;
break;
case BMAP_LEFT_FILLING:
@@ -850,43 +623,43 @@ xfs_bmap_add_extent_delay_real(
* Filling in the first part of a previous delayed allocation.
* The left neighbor is not contiguous.
*/
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
xfs_bmbt_set_startoff(ep, new_endoff);
temp = PREV.br_blockcount - new->br_blockcount;
xfs_bmbt_set_blockcount(ep, temp);
- xfs_iext_insert(ip, idx, 1, new, state);
- ip->i_df.if_lastex = idx;
- ip->i_d.di_nextents++;
- if (cur == NULL)
+ xfs_iext_insert(bma->ip, bma->idx, 1, new, state);
+ bma->ip->i_d.di_nextents++;
+ if (bma->cur == NULL)
rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
else {
rval = XFS_ILOG_CORE;
- if ((error = xfs_bmbt_lookup_eq(cur, new->br_startoff,
+ error = xfs_bmbt_lookup_eq(bma->cur, new->br_startoff,
new->br_startblock, new->br_blockcount,
- &i)))
+ &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 0, done);
- cur->bc_rec.b.br_state = XFS_EXT_NORM;
- if ((error = xfs_btree_insert(cur, &i)))
+ bma->cur->bc_rec.b.br_state = XFS_EXT_NORM;
+ error = xfs_btree_insert(bma->cur, &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 1, done);
}
- if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
- ip->i_d.di_nextents > ip->i_df.if_ext_max) {
- error = xfs_bmap_extents_to_btree(ip->i_transp, ip,
- first, flist, &cur, 1, &tmp_rval,
- XFS_DATA_FORK);
+
+ if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
+ error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
+ bma->firstblock, bma->flist,
+ &bma->cur, 1, &tmp_rval, XFS_DATA_FORK);
rval |= tmp_rval;
if (error)
goto done;
}
- temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
+ da_new = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(bma->ip, temp),
startblockval(PREV.br_startblock) -
- (cur ? cur->bc_private.b.allocated : 0));
- ep = xfs_iext_get_ext(ifp, idx + 1);
- xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
- trace_xfs_bmap_post_update(ip, idx + 1, state, _THIS_IP_);
- *dnew = temp;
+ (bma->cur ? bma->cur->bc_private.b.allocated : 0));
+ ep = xfs_iext_get_ext(ifp, bma->idx + 1);
+ xfs_bmbt_set_startblock(ep, nullstartblock(da_new));
+ trace_xfs_bmap_post_update(bma->ip, bma->idx + 1, state, _THIS_IP_);
break;
case BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
@@ -895,36 +668,39 @@ xfs_bmap_add_extent_delay_real(
* The right neighbor is contiguous with the new allocation.
*/
temp = PREV.br_blockcount - new->br_blockcount;
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
- trace_xfs_bmap_pre_update(ip, idx + 1, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx + 1, state, _THIS_IP_);
xfs_bmbt_set_blockcount(ep, temp);
- xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, idx + 1),
+ xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, bma->idx + 1),
new->br_startoff, new->br_startblock,
new->br_blockcount + RIGHT.br_blockcount,
RIGHT.br_state);
- trace_xfs_bmap_post_update(ip, idx + 1, state, _THIS_IP_);
- ip->i_df.if_lastex = idx + 1;
- if (cur == NULL)
+ trace_xfs_bmap_post_update(bma->ip, bma->idx + 1, state, _THIS_IP_);
+ if (bma->cur == NULL)
rval = XFS_ILOG_DEXT;
else {
rval = 0;
- if ((error = xfs_bmbt_lookup_eq(cur, RIGHT.br_startoff,
+ error = xfs_bmbt_lookup_eq(bma->cur, RIGHT.br_startoff,
RIGHT.br_startblock,
- RIGHT.br_blockcount, &i)))
+ RIGHT.br_blockcount, &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_bmbt_update(cur, new->br_startoff,
+ error = xfs_bmbt_update(bma->cur, new->br_startoff,
new->br_startblock,
new->br_blockcount +
RIGHT.br_blockcount,
- RIGHT.br_state)))
+ RIGHT.br_state);
+ if (error)
goto done;
}
- temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
+
+ da_new = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(bma->ip, temp),
startblockval(PREV.br_startblock));
- xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
- *dnew = temp;
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
+ xfs_bmbt_set_startblock(ep, nullstartblock(da_new));
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+
+ bma->idx++;
break;
case BMAP_RIGHT_FILLING:
@@ -933,41 +709,43 @@ xfs_bmap_add_extent_delay_real(
* The right neighbor is not contiguous.
*/
temp = PREV.br_blockcount - new->br_blockcount;
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
xfs_bmbt_set_blockcount(ep, temp);
- xfs_iext_insert(ip, idx + 1, 1, new, state);
- ip->i_df.if_lastex = idx + 1;
- ip->i_d.di_nextents++;
- if (cur == NULL)
+ xfs_iext_insert(bma->ip, bma->idx + 1, 1, new, state);
+ bma->ip->i_d.di_nextents++;
+ if (bma->cur == NULL)
rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
else {
rval = XFS_ILOG_CORE;
- if ((error = xfs_bmbt_lookup_eq(cur, new->br_startoff,
+ error = xfs_bmbt_lookup_eq(bma->cur, new->br_startoff,
new->br_startblock, new->br_blockcount,
- &i)))
+ &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 0, done);
- cur->bc_rec.b.br_state = XFS_EXT_NORM;
- if ((error = xfs_btree_insert(cur, &i)))
+ bma->cur->bc_rec.b.br_state = XFS_EXT_NORM;
+ error = xfs_btree_insert(bma->cur, &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 1, done);
}
- if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
- ip->i_d.di_nextents > ip->i_df.if_ext_max) {
- error = xfs_bmap_extents_to_btree(ip->i_transp, ip,
- first, flist, &cur, 1, &tmp_rval,
- XFS_DATA_FORK);
+
+ if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
+ error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
+ bma->firstblock, bma->flist, &bma->cur, 1,
+ &tmp_rval, XFS_DATA_FORK);
rval |= tmp_rval;
if (error)
goto done;
}
- temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
+ da_new = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(bma->ip, temp),
startblockval(PREV.br_startblock) -
- (cur ? cur->bc_private.b.allocated : 0));
- ep = xfs_iext_get_ext(ifp, idx);
- xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
- *dnew = temp;
+ (bma->cur ? bma->cur->bc_private.b.allocated : 0));
+ ep = xfs_iext_get_ext(ifp, bma->idx);
+ xfs_bmbt_set_startblock(ep, nullstartblock(da_new));
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+
+ bma->idx++;
break;
case 0:
@@ -975,82 +753,83 @@ xfs_bmap_add_extent_delay_real(
* Filling in the middle part of a previous delayed allocation.
* Contiguity is impossible here.
* This case is avoided almost all the time.
+ *
+ * We start with a delayed allocation:
+ *
+ * +ddddddddddddddddddddddddddddddddddddddddddddddddddddddd+
+ * PREV @ idx
+ *
+ * and we are allocating:
+ * +rrrrrrrrrrrrrrrrr+
+ * new
+ *
+ * and we set it up for insertion as:
+ * +ddddddddddddddddddd+rrrrrrrrrrrrrrrrr+ddddddddddddddddd+
+ * new
+ * PREV @ idx LEFT RIGHT
+ * inserted at idx + 1
*/
temp = new->br_startoff - PREV.br_startoff;
- trace_xfs_bmap_pre_update(ip, idx, 0, _THIS_IP_);
- xfs_bmbt_set_blockcount(ep, temp);
- r[0] = *new;
- r[1].br_state = PREV.br_state;
- r[1].br_startblock = 0;
- r[1].br_startoff = new_endoff;
temp2 = PREV.br_startoff + PREV.br_blockcount - new_endoff;
- r[1].br_blockcount = temp2;
- xfs_iext_insert(ip, idx + 1, 2, &r[0], state);
- ip->i_df.if_lastex = idx + 1;
- ip->i_d.di_nextents++;
- if (cur == NULL)
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, 0, _THIS_IP_);
+ xfs_bmbt_set_blockcount(ep, temp); /* truncate PREV */
+ LEFT = *new;
+ RIGHT.br_state = PREV.br_state;
+ RIGHT.br_startblock = nullstartblock(
+ (int)xfs_bmap_worst_indlen(bma->ip, temp2));
+ RIGHT.br_startoff = new_endoff;
+ RIGHT.br_blockcount = temp2;
+ /* insert LEFT (r[0]) and RIGHT (r[1]) at the same time */
+ xfs_iext_insert(bma->ip, bma->idx + 1, 2, &LEFT, state);
+ bma->ip->i_d.di_nextents++;
+ if (bma->cur == NULL)
rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
else {
rval = XFS_ILOG_CORE;
- if ((error = xfs_bmbt_lookup_eq(cur, new->br_startoff,
+ error = xfs_bmbt_lookup_eq(bma->cur, new->br_startoff,
new->br_startblock, new->br_blockcount,
- &i)))
+ &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 0, done);
- cur->bc_rec.b.br_state = XFS_EXT_NORM;
- if ((error = xfs_btree_insert(cur, &i)))
+ bma->cur->bc_rec.b.br_state = XFS_EXT_NORM;
+ error = xfs_btree_insert(bma->cur, &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 1, done);
}
- if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
- ip->i_d.di_nextents > ip->i_df.if_ext_max) {
- error = xfs_bmap_extents_to_btree(ip->i_transp, ip,
- first, flist, &cur, 1, &tmp_rval,
- XFS_DATA_FORK);
+
+ if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
+ error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
+ bma->firstblock, bma->flist, &bma->cur,
+ 1, &tmp_rval, XFS_DATA_FORK);
rval |= tmp_rval;
if (error)
goto done;
}
- temp = xfs_bmap_worst_indlen(ip, temp);
- temp2 = xfs_bmap_worst_indlen(ip, temp2);
+ temp = xfs_bmap_worst_indlen(bma->ip, temp);
+ temp2 = xfs_bmap_worst_indlen(bma->ip, temp2);
diff = (int)(temp + temp2 - startblockval(PREV.br_startblock) -
- (cur ? cur->bc_private.b.allocated : 0));
- if (diff > 0 &&
- xfs_icsb_modify_counters(ip->i_mount, XFS_SBS_FDBLOCKS,
- -((int64_t)diff), rsvd)) {
- /*
- * Ick gross gag me with a spoon.
- */
- ASSERT(0); /* want to see if this ever happens! */
- while (diff > 0) {
- if (temp) {
- temp--;
- diff--;
- if (!diff ||
- !xfs_icsb_modify_counters(ip->i_mount,
- XFS_SBS_FDBLOCKS,
- -((int64_t)diff), rsvd))
- break;
- }
- if (temp2) {
- temp2--;
- diff--;
- if (!diff ||
- !xfs_icsb_modify_counters(ip->i_mount,
- XFS_SBS_FDBLOCKS,
- -((int64_t)diff), rsvd))
- break;
- }
- }
+ (bma->cur ? bma->cur->bc_private.b.allocated : 0));
+ if (diff > 0) {
+ error = xfs_icsb_modify_counters(bma->ip->i_mount,
+ XFS_SBS_FDBLOCKS,
+ -((int64_t)diff), 0);
+ ASSERT(!error);
+ if (error)
+ goto done;
}
- ep = xfs_iext_get_ext(ifp, idx);
+
+ ep = xfs_iext_get_ext(ifp, bma->idx);
xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
- trace_xfs_bmap_pre_update(ip, idx + 2, state, _THIS_IP_);
- xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, idx + 2),
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx + 2, state, _THIS_IP_);
+ xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, bma->idx + 2),
nullstartblock((int)temp2));
- trace_xfs_bmap_post_update(ip, idx + 2, state, _THIS_IP_);
- *dnew = temp + temp2;
+ trace_xfs_bmap_post_update(bma->ip, bma->idx + 2, state, _THIS_IP_);
+
+ bma->idx++;
+ da_new = temp + temp2;
break;
case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
@@ -1065,9 +844,39 @@ xfs_bmap_add_extent_delay_real(
*/
ASSERT(0);
}
- *curp = cur;
+
+ /* convert to a btree if necessary */
+ if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
+ int tmp_logflags; /* partial log flag return val */
+
+ ASSERT(bma->cur == NULL);
+ error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
+ bma->firstblock, bma->flist, &bma->cur,
+ da_old > 0, &tmp_logflags, XFS_DATA_FORK);
+ bma->logflags |= tmp_logflags;
+ if (error)
+ goto done;
+ }
+
+ /* adjust for changes in reserved delayed indirect blocks */
+ if (da_old || da_new) {
+ temp = da_new;
+ if (bma->cur)
+ temp += bma->cur->bc_private.b.allocated;
+ ASSERT(temp <= da_old);
+ if (temp < da_old)
+ xfs_icsb_modify_counters(bma->ip->i_mount,
+ XFS_SBS_FDBLOCKS,
+ (int64_t)(da_old - temp), 0);
+ }
+
+ /* clear out the allocated field, done with it now in any case. */
+ if (bma->cur)
+ bma->cur->bc_private.b.allocated = 0;
+
+ xfs_bmap_check_leaf_extents(bma->cur, bma->ip, XFS_DATA_FORK);
done:
- *logflagsp = rval;
+ bma->logflags |= rval;
return error;
#undef LEFT
#undef RIGHT
@@ -1075,15 +884,17 @@ done:
}
/*
- * Called by xfs_bmap_add_extent to handle cases converting an unwritten
- * allocation to a real allocation or vice versa.
+ * Convert an unwritten allocation to a real allocation or vice versa.
*/
STATIC int /* error */
xfs_bmap_add_extent_unwritten_real(
+ struct xfs_trans *tp,
xfs_inode_t *ip, /* incore inode pointer */
- xfs_extnum_t idx, /* extent number to update/insert */
+ xfs_extnum_t *idx, /* extent number to update/insert */
xfs_btree_cur_t **curp, /* if *curp is null, not a btree */
xfs_bmbt_irec_t *new, /* new data to add to file extents */
+ xfs_fsblock_t *first, /* pointer to firstblock variable */
+ xfs_bmap_free_t *flist, /* list of extents to be freed */
int *logflagsp) /* inode logging flags */
{
xfs_btree_cur_t *cur; /* btree cursor */
@@ -1099,16 +910,26 @@ xfs_bmap_add_extent_unwritten_real(
int rval=0; /* return value (logging flags) */
int state = 0;/* state bits, accessed thru macros */
+ *logflagsp = 0;
+
+ cur = *curp;
+ ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
+
+ ASSERT(*idx >= 0);
+ ASSERT(*idx <= ifp->if_bytes / sizeof(struct xfs_bmbt_rec));
+ ASSERT(!isnullstartblock(new->br_startblock));
+
+ XFS_STATS_INC(xs_add_exlist);
+
#define LEFT r[0]
#define RIGHT r[1]
#define PREV r[2]
+
/*
* Set up a bunch of variables to make the tests simpler.
*/
error = 0;
- cur = *curp;
- ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
- ep = xfs_iext_get_ext(ifp, idx);
+ ep = xfs_iext_get_ext(ifp, *idx);
xfs_bmbt_get_all(ep, &PREV);
newext = new->br_state;
oldext = (newext == XFS_EXT_UNWRITTEN) ?
@@ -1131,9 +952,9 @@ xfs_bmap_add_extent_unwritten_real(
* Check and set flags if this segment has a left neighbor.
* Don't set contiguous if the combined extent would be too large.
*/
- if (idx > 0) {
+ if (*idx > 0) {
state |= BMAP_LEFT_VALID;
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx - 1), &LEFT);
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx - 1), &LEFT);
if (isnullstartblock(LEFT.br_startblock))
state |= BMAP_LEFT_DELAY;
@@ -1151,9 +972,9 @@ xfs_bmap_add_extent_unwritten_real(
* Don't set contiguous if the combined extent would be too large.
* Also check for all-three-contiguous being too large.
*/
- if (idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) {
+ if (*idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) {
state |= BMAP_RIGHT_VALID;
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx + 1), &RIGHT);
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx + 1), &RIGHT);
if (isnullstartblock(RIGHT.br_startblock))
state |= BMAP_RIGHT_DELAY;
}
@@ -1182,14 +1003,15 @@ xfs_bmap_add_extent_unwritten_real(
* Setting all of a previous oldext extent to newext.
* The left and right neighbors are both contiguous with new.
*/
- trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
+ --*idx;
+
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx),
LEFT.br_blockcount + PREV.br_blockcount +
RIGHT.br_blockcount);
- trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- xfs_iext_remove(ip, idx, 2, state);
- ip->i_df.if_lastex = idx - 1;
+ xfs_iext_remove(ip, *idx + 1, 2, state);
ip->i_d.di_nextents -= 2;
if (cur == NULL)
rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
@@ -1225,13 +1047,14 @@ xfs_bmap_add_extent_unwritten_real(
* Setting all of a previous oldext extent to newext.
* The left neighbor is contiguous, the right is not.
*/
- trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
+ --*idx;
+
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx),
LEFT.br_blockcount + PREV.br_blockcount);
- trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- ip->i_df.if_lastex = idx - 1;
- xfs_iext_remove(ip, idx, 1, state);
+ xfs_iext_remove(ip, *idx + 1, 1, state);
ip->i_d.di_nextents--;
if (cur == NULL)
rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
@@ -1261,13 +1084,12 @@ xfs_bmap_add_extent_unwritten_real(
* Setting all of a previous oldext extent to newext.
* The right neighbor is contiguous, the left is not.
*/
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
xfs_bmbt_set_blockcount(ep,
PREV.br_blockcount + RIGHT.br_blockcount);
xfs_bmbt_set_state(ep, newext);
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
- ip->i_df.if_lastex = idx;
- xfs_iext_remove(ip, idx + 1, 1, state);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
+ xfs_iext_remove(ip, *idx + 1, 1, state);
ip->i_d.di_nextents--;
if (cur == NULL)
rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
@@ -1298,11 +1120,10 @@ xfs_bmap_add_extent_unwritten_real(
* Neither the left nor right neighbors are contiguous with
* the new one.
*/
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
xfs_bmbt_set_state(ep, newext);
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- ip->i_df.if_lastex = idx;
if (cur == NULL)
rval = XFS_ILOG_DEXT;
else {
@@ -1324,21 +1145,22 @@ xfs_bmap_add_extent_unwritten_real(
* Setting the first part of a previous oldext extent to newext.
* The left neighbor is contiguous.
*/
- trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
+ trace_xfs_bmap_pre_update(ip, *idx - 1, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx - 1),
LEFT.br_blockcount + new->br_blockcount);
xfs_bmbt_set_startoff(ep,
PREV.br_startoff + new->br_blockcount);
- trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(ip, *idx - 1, state, _THIS_IP_);
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
xfs_bmbt_set_startblock(ep,
new->br_startblock + new->br_blockcount);
xfs_bmbt_set_blockcount(ep,
PREV.br_blockcount - new->br_blockcount);
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
+
+ --*idx;
- ip->i_df.if_lastex = idx - 1;
if (cur == NULL)
rval = XFS_ILOG_DEXT;
else {
@@ -1356,10 +1178,11 @@ xfs_bmap_add_extent_unwritten_real(
goto done;
if ((error = xfs_btree_decrement(cur, 0, &i)))
goto done;
- if (xfs_bmbt_update(cur, LEFT.br_startoff,
+ error = xfs_bmbt_update(cur, LEFT.br_startoff,
LEFT.br_startblock,
LEFT.br_blockcount + new->br_blockcount,
- LEFT.br_state))
+ LEFT.br_state);
+ if (error)
goto done;
}
break;
@@ -1369,17 +1192,16 @@ xfs_bmap_add_extent_unwritten_real(
* Setting the first part of a previous oldext extent to newext.
* The left neighbor is not contiguous.
*/
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
ASSERT(ep && xfs_bmbt_get_state(ep) == oldext);
xfs_bmbt_set_startoff(ep, new_endoff);
xfs_bmbt_set_blockcount(ep,
PREV.br_blockcount - new->br_blockcount);
xfs_bmbt_set_startblock(ep,
new->br_startblock + new->br_blockcount);
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- xfs_iext_insert(ip, idx, 1, new, state);
- ip->i_df.if_lastex = idx;
+ xfs_iext_insert(ip, *idx, 1, new, state);
ip->i_d.di_nextents++;
if (cur == NULL)
rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
@@ -1408,17 +1230,19 @@ xfs_bmap_add_extent_unwritten_real(
* Setting the last part of a previous oldext extent to newext.
* The right neighbor is contiguous with the new allocation.
*/
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
- trace_xfs_bmap_pre_update(ip, idx + 1, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
xfs_bmbt_set_blockcount(ep,
PREV.br_blockcount - new->br_blockcount);
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
- xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, idx + 1),
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
+
+ ++*idx;
+
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, *idx),
new->br_startoff, new->br_startblock,
new->br_blockcount + RIGHT.br_blockcount, newext);
- trace_xfs_bmap_post_update(ip, idx + 1, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- ip->i_df.if_lastex = idx + 1;
if (cur == NULL)
rval = XFS_ILOG_DEXT;
else {
@@ -1448,13 +1272,14 @@ xfs_bmap_add_extent_unwritten_real(
* Setting the last part of a previous oldext extent to newext.
* The right neighbor is not contiguous.
*/
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
xfs_bmbt_set_blockcount(ep,
PREV.br_blockcount - new->br_blockcount);
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
+
+ ++*idx;
+ xfs_iext_insert(ip, *idx, 1, new, state);
- xfs_iext_insert(ip, idx + 1, 1, new, state);
- ip->i_df.if_lastex = idx + 1;
ip->i_d.di_nextents++;
if (cur == NULL)
rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
@@ -1488,10 +1313,10 @@ xfs_bmap_add_extent_unwritten_real(
* newext. Contiguity is impossible here.
* One extent becomes three extents.
*/
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
xfs_bmbt_set_blockcount(ep,
new->br_startoff - PREV.br_startoff);
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
r[0] = *new;
r[1].br_startoff = new_endoff;
@@ -1499,8 +1324,10 @@ xfs_bmap_add_extent_unwritten_real(
PREV.br_startoff + PREV.br_blockcount - new_endoff;
r[1].br_startblock = new->br_startblock + new->br_blockcount;
r[1].br_state = oldext;
- xfs_iext_insert(ip, idx + 1, 2, &r[0], state);
- ip->i_df.if_lastex = idx + 1;
+
+ ++*idx;
+ xfs_iext_insert(ip, *idx, 2, &r[0], state);
+
ip->i_d.di_nextents += 2;
if (cur == NULL)
rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
@@ -1553,9 +1380,28 @@ xfs_bmap_add_extent_unwritten_real(
*/
ASSERT(0);
}
- *curp = cur;
+
+ /* convert to a btree if necessary */
+ if (xfs_bmap_needs_btree(ip, XFS_DATA_FORK)) {
+ int tmp_logflags; /* partial log flag return val */
+
+ ASSERT(cur == NULL);
+ error = xfs_bmap_extents_to_btree(tp, ip, first, flist, &cur,
+ 0, &tmp_logflags, XFS_DATA_FORK);
+ *logflagsp |= tmp_logflags;
+ if (error)
+ goto done;
+ }
+
+ /* clear out the allocated field, done with it now in any case. */
+ if (cur) {
+ cur->bc_private.b.allocated = 0;
+ *curp = cur;
+ }
+
+ xfs_bmap_check_leaf_extents(*curp, ip, XFS_DATA_FORK);
done:
- *logflagsp = rval;
+ *logflagsp |= rval;
return error;
#undef LEFT
#undef RIGHT
@@ -1563,19 +1409,14 @@ done:
}
/*
- * Called by xfs_bmap_add_extent to handle cases converting a hole
- * to a delayed allocation.
+ * Convert a hole to a delayed allocation.
*/
-/*ARGSUSED*/
-STATIC int /* error */
+STATIC void
xfs_bmap_add_extent_hole_delay(
xfs_inode_t *ip, /* incore inode pointer */
- xfs_extnum_t idx, /* extent number to update/insert */
- xfs_bmbt_irec_t *new, /* new data to add to file extents */
- int *logflagsp, /* inode logging flags */
- int rsvd) /* OK to allocate reserved blocks */
+ xfs_extnum_t *idx, /* extent number to update/insert */
+ xfs_bmbt_irec_t *new) /* new data to add to file extents */
{
- xfs_bmbt_rec_host_t *ep; /* extent record for idx */
xfs_ifork_t *ifp; /* inode fork pointer */
xfs_bmbt_irec_t left; /* left neighbor extent entry */
xfs_filblks_t newlen=0; /* new indirect size */
@@ -1585,16 +1426,15 @@ xfs_bmap_add_extent_hole_delay(
xfs_filblks_t temp=0; /* temp for indirect calculations */
ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
- ep = xfs_iext_get_ext(ifp, idx);
state = 0;
ASSERT(isnullstartblock(new->br_startblock));
/*
* Check and set flags if this segment has a left neighbor
*/
- if (idx > 0) {
+ if (*idx > 0) {
state |= BMAP_LEFT_VALID;
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx - 1), &left);
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx - 1), &left);
if (isnullstartblock(left.br_startblock))
state |= BMAP_LEFT_DELAY;
@@ -1604,9 +1444,9 @@ xfs_bmap_add_extent_hole_delay(
* Check and set flags if the current (right) segment exists.
* If it doesn't exist, we're converting the hole at end-of-file.
*/
- if (idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t)) {
+ if (*idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t)) {
state |= BMAP_RIGHT_VALID;
- xfs_bmbt_get_all(ep, &right);
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx), &right);
if (isnullstartblock(right.br_startblock))
state |= BMAP_RIGHT_DELAY;
@@ -1639,21 +1479,21 @@ xfs_bmap_add_extent_hole_delay(
* on the left and on the right.
* Merge all three into a single extent record.
*/
+ --*idx;
temp = left.br_blockcount + new->br_blockcount +
right.br_blockcount;
- trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1), temp);
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx), temp);
oldlen = startblockval(left.br_startblock) +
startblockval(new->br_startblock) +
startblockval(right.br_startblock);
newlen = xfs_bmap_worst_indlen(ip, temp);
- xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, idx - 1),
+ xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, *idx),
nullstartblock((int)newlen));
- trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- xfs_iext_remove(ip, idx, 1, state);
- ip->i_df.if_lastex = idx - 1;
+ xfs_iext_remove(ip, *idx + 1, 1, state);
break;
case BMAP_LEFT_CONTIG:
@@ -1662,17 +1502,17 @@ xfs_bmap_add_extent_hole_delay(
* on the left.
* Merge the new allocation with the left neighbor.
*/
+ --*idx;
temp = left.br_blockcount + new->br_blockcount;
- trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1), temp);
+
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx), temp);
oldlen = startblockval(left.br_startblock) +
startblockval(new->br_startblock);
newlen = xfs_bmap_worst_indlen(ip, temp);
- xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, idx - 1),
+ xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, *idx),
nullstartblock((int)newlen));
- trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
-
- ip->i_df.if_lastex = idx - 1;
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
break;
case BMAP_RIGHT_CONTIG:
@@ -1681,16 +1521,15 @@ xfs_bmap_add_extent_hole_delay(
* on the right.
* Merge the new allocation with the right neighbor.
*/
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
temp = new->br_blockcount + right.br_blockcount;
oldlen = startblockval(new->br_startblock) +
startblockval(right.br_startblock);
newlen = xfs_bmap_worst_indlen(ip, temp);
- xfs_bmbt_set_allf(ep, new->br_startoff,
+ xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, *idx),
+ new->br_startoff,
nullstartblock((int)newlen), temp, right.br_state);
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
-
- ip->i_df.if_lastex = idx;
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
break;
case 0:
@@ -1700,36 +1539,28 @@ xfs_bmap_add_extent_hole_delay(
* Insert a new entry.
*/
oldlen = newlen = 0;
- xfs_iext_insert(ip, idx, 1, new, state);
- ip->i_df.if_lastex = idx;
+ xfs_iext_insert(ip, *idx, 1, new, state);
break;
}
if (oldlen != newlen) {
ASSERT(oldlen > newlen);
xfs_icsb_modify_counters(ip->i_mount, XFS_SBS_FDBLOCKS,
- (int64_t)(oldlen - newlen), rsvd);
+ (int64_t)(oldlen - newlen), 0);
/*
* Nothing to do for disk quota accounting here.
*/
}
- *logflagsp = 0;
- return 0;
}
/*
- * Called by xfs_bmap_add_extent to handle cases converting a hole
- * to a real allocation.
+ * Convert a hole to a real allocation.
*/
STATIC int /* error */
xfs_bmap_add_extent_hole_real(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_extnum_t idx, /* extent number to update/insert */
- xfs_btree_cur_t *cur, /* if null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to add to file extents */
- int *logflagsp, /* inode logging flags */
- int whichfork) /* data or attr fork */
+ struct xfs_bmalloca *bma,
+ int whichfork)
{
- xfs_bmbt_rec_host_t *ep; /* pointer to extent entry ins. point */
+ struct xfs_bmbt_irec *new = &bma->got;
int error; /* error return value */
int i; /* temp state */
xfs_ifork_t *ifp; /* inode fork pointer */
@@ -1738,20 +1569,26 @@ xfs_bmap_add_extent_hole_real(
int rval=0; /* return value (logging flags) */
int state; /* state bits, accessed thru macros */
- ifp = XFS_IFORK_PTR(ip, whichfork);
- ASSERT(idx <= ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t));
- ep = xfs_iext_get_ext(ifp, idx);
- state = 0;
+ ifp = XFS_IFORK_PTR(bma->ip, whichfork);
+
+ ASSERT(bma->idx >= 0);
+ ASSERT(bma->idx <= ifp->if_bytes / sizeof(struct xfs_bmbt_rec));
+ ASSERT(!isnullstartblock(new->br_startblock));
+ ASSERT(!bma->cur ||
+ !(bma->cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL));
+
+ XFS_STATS_INC(xs_add_exlist);
+ state = 0;
if (whichfork == XFS_ATTR_FORK)
state |= BMAP_ATTRFORK;
/*
* Check and set flags if this segment has a left neighbor.
*/
- if (idx > 0) {
+ if (bma->idx > 0) {
state |= BMAP_LEFT_VALID;
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx - 1), &left);
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx - 1), &left);
if (isnullstartblock(left.br_startblock))
state |= BMAP_LEFT_DELAY;
}
@@ -1760,9 +1597,9 @@ xfs_bmap_add_extent_hole_real(
* Check and set flags if this segment has a current value.
* Not true if we're inserting into the "hole" at eof.
*/
- if (idx < ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t)) {
+ if (bma->idx < ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t)) {
state |= BMAP_RIGHT_VALID;
- xfs_bmbt_get_all(ep, &right);
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx), &right);
if (isnullstartblock(right.br_startblock))
state |= BMAP_RIGHT_DELAY;
}
@@ -1799,38 +1636,42 @@ xfs_bmap_add_extent_hole_real(
* left and on the right.
* Merge all three into a single extent record.
*/
- trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
+ --bma->idx;
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, bma->idx),
left.br_blockcount + new->br_blockcount +
right.br_blockcount);
- trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
- xfs_iext_remove(ip, idx, 1, state);
- ifp->if_lastex = idx - 1;
- XFS_IFORK_NEXT_SET(ip, whichfork,
- XFS_IFORK_NEXTENTS(ip, whichfork) - 1);
- if (cur == NULL) {
+ xfs_iext_remove(bma->ip, bma->idx + 1, 1, state);
+
+ XFS_IFORK_NEXT_SET(bma->ip, whichfork,
+ XFS_IFORK_NEXTENTS(bma->ip, whichfork) - 1);
+ if (bma->cur == NULL) {
rval = XFS_ILOG_CORE | xfs_ilog_fext(whichfork);
} else {
rval = XFS_ILOG_CORE;
- if ((error = xfs_bmbt_lookup_eq(cur,
- right.br_startoff,
- right.br_startblock,
- right.br_blockcount, &i)))
+ error = xfs_bmbt_lookup_eq(bma->cur, right.br_startoff,
+ right.br_startblock, right.br_blockcount,
+ &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_btree_delete(cur, &i)))
+ error = xfs_btree_delete(bma->cur, &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_btree_decrement(cur, 0, &i)))
+ error = xfs_btree_decrement(bma->cur, 0, &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_bmbt_update(cur, left.br_startoff,
+ error = xfs_bmbt_update(bma->cur, left.br_startoff,
left.br_startblock,
left.br_blockcount +
new->br_blockcount +
right.br_blockcount,
- left.br_state)))
+ left.br_state);
+ if (error)
goto done;
}
break;
@@ -1841,27 +1682,28 @@ xfs_bmap_add_extent_hole_real(
* on the left.
* Merge the new allocation with the left neighbor.
*/
- trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
+ --bma->idx;
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, bma->idx),
left.br_blockcount + new->br_blockcount);
- trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
- ifp->if_lastex = idx - 1;
- if (cur == NULL) {
+ if (bma->cur == NULL) {
rval = xfs_ilog_fext(whichfork);
} else {
rval = 0;
- if ((error = xfs_bmbt_lookup_eq(cur,
- left.br_startoff,
- left.br_startblock,
- left.br_blockcount, &i)))
+ error = xfs_bmbt_lookup_eq(bma->cur, left.br_startoff,
+ left.br_startblock, left.br_blockcount,
+ &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_bmbt_update(cur, left.br_startoff,
+ error = xfs_bmbt_update(bma->cur, left.br_startoff,
left.br_startblock,
left.br_blockcount +
new->br_blockcount,
- left.br_state)))
+ left.br_state);
+ if (error)
goto done;
}
break;
@@ -1872,28 +1714,30 @@ xfs_bmap_add_extent_hole_real(
* on the right.
* Merge the new allocation with the right neighbor.
*/
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
- xfs_bmbt_set_allf(ep, new->br_startoff, new->br_startblock,
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
+ xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, bma->idx),
+ new->br_startoff, new->br_startblock,
new->br_blockcount + right.br_blockcount,
right.br_state);
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
- ifp->if_lastex = idx;
- if (cur == NULL) {
+ if (bma->cur == NULL) {
rval = xfs_ilog_fext(whichfork);
} else {
rval = 0;
- if ((error = xfs_bmbt_lookup_eq(cur,
+ error = xfs_bmbt_lookup_eq(bma->cur,
right.br_startoff,
right.br_startblock,
- right.br_blockcount, &i)))
+ right.br_blockcount, &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_bmbt_update(cur, new->br_startoff,
+ error = xfs_bmbt_update(bma->cur, new->br_startoff,
new->br_startblock,
new->br_blockcount +
right.br_blockcount,
- right.br_state)))
+ right.br_state);
+ if (error)
goto done;
}
break;
@@ -1904,29 +1748,49 @@ xfs_bmap_add_extent_hole_real(
* real allocation.
* Insert a new entry.
*/
- xfs_iext_insert(ip, idx, 1, new, state);
- ifp->if_lastex = idx;
- XFS_IFORK_NEXT_SET(ip, whichfork,
- XFS_IFORK_NEXTENTS(ip, whichfork) + 1);
- if (cur == NULL) {
+ xfs_iext_insert(bma->ip, bma->idx, 1, new, state);
+ XFS_IFORK_NEXT_SET(bma->ip, whichfork,
+ XFS_IFORK_NEXTENTS(bma->ip, whichfork) + 1);
+ if (bma->cur == NULL) {
rval = XFS_ILOG_CORE | xfs_ilog_fext(whichfork);
} else {
rval = XFS_ILOG_CORE;
- if ((error = xfs_bmbt_lookup_eq(cur,
+ error = xfs_bmbt_lookup_eq(bma->cur,
new->br_startoff,
new->br_startblock,
- new->br_blockcount, &i)))
+ new->br_blockcount, &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 0, done);
- cur->bc_rec.b.br_state = new->br_state;
- if ((error = xfs_btree_insert(cur, &i)))
+ bma->cur->bc_rec.b.br_state = new->br_state;
+ error = xfs_btree_insert(bma->cur, &i);
+ if (error)
goto done;
XFS_WANT_CORRUPTED_GOTO(i == 1, done);
}
break;
}
+
+ /* convert to a btree if necessary */
+ if (xfs_bmap_needs_btree(bma->ip, whichfork)) {
+ int tmp_logflags; /* partial log flag return val */
+
+ ASSERT(bma->cur == NULL);
+ error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
+ bma->firstblock, bma->flist, &bma->cur,
+ 0, &tmp_logflags, whichfork);
+ bma->logflags |= tmp_logflags;
+ if (error)
+ goto done;
+ }
+
+ /* clear out the allocated field, done with it now in any case. */
+ if (bma->cur)
+ bma->cur->bc_private.b.allocated = 0;
+
+ xfs_bmap_check_leaf_extents(bma->cur, bma->ip, whichfork);
done:
- *logflagsp = rval;
+ bma->logflags |= rval;
return error;
}
@@ -2113,26 +1977,26 @@ xfs_bmap_adjacent(
XFS_FSB_TO_AGBNO(mp, x) < mp->m_sb.sb_agblocks)
mp = ap->ip->i_mount;
- nullfb = ap->firstblock == NULLFSBLOCK;
+ nullfb = *ap->firstblock == NULLFSBLOCK;
rt = XFS_IS_REALTIME_INODE(ap->ip) && ap->userdata;
- fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp, ap->firstblock);
+ fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp, *ap->firstblock);
/*
* If allocating at eof, and there's a previous real block,
* try to use its last block as our starting point.
*/
- if (ap->eof && ap->prevp->br_startoff != NULLFILEOFF &&
- !isnullstartblock(ap->prevp->br_startblock) &&
- ISVALID(ap->prevp->br_startblock + ap->prevp->br_blockcount,
- ap->prevp->br_startblock)) {
- ap->rval = ap->prevp->br_startblock + ap->prevp->br_blockcount;
+ if (ap->eof && ap->prev.br_startoff != NULLFILEOFF &&
+ !isnullstartblock(ap->prev.br_startblock) &&
+ ISVALID(ap->prev.br_startblock + ap->prev.br_blockcount,
+ ap->prev.br_startblock)) {
+ ap->blkno = ap->prev.br_startblock + ap->prev.br_blockcount;
/*
* Adjust for the gap between prevp and us.
*/
- adjust = ap->off -
- (ap->prevp->br_startoff + ap->prevp->br_blockcount);
+ adjust = ap->offset -
+ (ap->prev.br_startoff + ap->prev.br_blockcount);
if (adjust &&
- ISVALID(ap->rval + adjust, ap->prevp->br_startblock))
- ap->rval += adjust;
+ ISVALID(ap->blkno + adjust, ap->prev.br_startblock))
+ ap->blkno += adjust;
}
/*
* If not at eof, then compare the two neighbor blocks.
@@ -2149,17 +2013,17 @@ xfs_bmap_adjacent(
* If there's a previous (left) block, select a requested
* start block based on it.
*/
- if (ap->prevp->br_startoff != NULLFILEOFF &&
- !isnullstartblock(ap->prevp->br_startblock) &&
- (prevbno = ap->prevp->br_startblock +
- ap->prevp->br_blockcount) &&
- ISVALID(prevbno, ap->prevp->br_startblock)) {
+ if (ap->prev.br_startoff != NULLFILEOFF &&
+ !isnullstartblock(ap->prev.br_startblock) &&
+ (prevbno = ap->prev.br_startblock +
+ ap->prev.br_blockcount) &&
+ ISVALID(prevbno, ap->prev.br_startblock)) {
/*
* Calculate gap to end of previous block.
*/
- adjust = prevdiff = ap->off -
- (ap->prevp->br_startoff +
- ap->prevp->br_blockcount);
+ adjust = prevdiff = ap->offset -
+ (ap->prev.br_startoff +
+ ap->prev.br_blockcount);
/*
* Figure the startblock based on the previous block's
* end and the gap size.
@@ -2168,9 +2032,9 @@ xfs_bmap_adjacent(
* allocating, or using it gives us an invalid block
* number, then just use the end of the previous block.
*/
- if (prevdiff <= XFS_ALLOC_GAP_UNITS * ap->alen &&
+ if (prevdiff <= XFS_ALLOC_GAP_UNITS * ap->length &&
ISVALID(prevbno + prevdiff,
- ap->prevp->br_startblock))
+ ap->prev.br_startblock))
prevbno += adjust;
else
prevdiff += adjust;
@@ -2191,16 +2055,16 @@ xfs_bmap_adjacent(
* If there's a following (right) block, select a requested
* start block based on it.
*/
- if (!isnullstartblock(ap->gotp->br_startblock)) {
+ if (!isnullstartblock(ap->got.br_startblock)) {
/*
* Calculate gap to start of next block.
*/
- adjust = gotdiff = ap->gotp->br_startoff - ap->off;
+ adjust = gotdiff = ap->got.br_startoff - ap->offset;
/*
* Figure the startblock based on the next block's
* start and the gap size.
*/
- gotbno = ap->gotp->br_startblock;
+ gotbno = ap->got.br_startblock;
/*
* Heuristic!
* If the gap is large relative to the piece we're
@@ -2208,12 +2072,12 @@ xfs_bmap_adjacent(
* number, then just use the start of the next block
* offset by our length.
*/
- if (gotdiff <= XFS_ALLOC_GAP_UNITS * ap->alen &&
+ if (gotdiff <= XFS_ALLOC_GAP_UNITS * ap->length &&
ISVALID(gotbno - gotdiff, gotbno))
gotbno -= adjust;
- else if (ISVALID(gotbno - ap->alen, gotbno)) {
- gotbno -= ap->alen;
- gotdiff += adjust - ap->alen;
+ else if (ISVALID(gotbno - ap->length, gotbno)) {
+ gotbno -= ap->length;
+ gotdiff += adjust - ap->length;
} else
gotdiff += adjust;
/*
@@ -2234,11 +2098,11 @@ xfs_bmap_adjacent(
* one, else ap->rval is already set (to 0 or the inode block).
*/
if (prevbno != NULLFSBLOCK && gotbno != NULLFSBLOCK)
- ap->rval = prevdiff <= gotdiff ? prevbno : gotbno;
+ ap->blkno = prevdiff <= gotdiff ? prevbno : gotbno;
else if (prevbno != NULLFSBLOCK)
- ap->rval = prevbno;
+ ap->blkno = prevbno;
else if (gotbno != NULLFSBLOCK)
- ap->rval = gotbno;
+ ap->blkno = gotbno;
}
#undef ISVALID
}
@@ -2271,7 +2135,7 @@ xfs_bmap_btalloc_nullfb(
startag = ag = 0;
pag = xfs_perag_get(mp, ag);
- while (*blen < ap->alen) {
+ while (*blen < args->maxlen) {
if (!pag->pagf_init) {
error = xfs_alloc_pagf_init(mp, args->tp, ag,
XFS_ALLOC_FLAG_TRYLOCK);
@@ -2293,7 +2157,7 @@ xfs_bmap_btalloc_nullfb(
notinit = 1;
if (xfs_inode_is_filestream(ap->ip)) {
- if (*blen >= ap->alen)
+ if (*blen >= args->maxlen)
break;
if (ap->userdata) {
@@ -2339,21 +2203,21 @@ xfs_bmap_btalloc_nullfb(
* If the best seen length is less than the request
* length, use the best as the minimum.
*/
- else if (*blen < ap->alen)
+ else if (*blen < args->maxlen)
args->minlen = *blen;
/*
- * Otherwise we've seen an extent as big as alen,
+ * Otherwise we've seen an extent as big as maxlen,
* use that as the minimum.
*/
else
- args->minlen = ap->alen;
+ args->minlen = args->maxlen;
/*
* set the failure fallback case to look in the selected
* AG as the stream may have moved.
*/
if (xfs_inode_is_filestream(ap->ip))
- ap->rval = args->fsbno = XFS_AGB_TO_FSB(mp, ag, 0);
+ ap->blkno = args->fsbno = XFS_AGB_TO_FSB(mp, ag, 0);
return 0;
}
@@ -2375,53 +2239,58 @@ xfs_bmap_btalloc(
int tryagain;
int error;
+ ASSERT(ap->length);
+
mp = ap->ip->i_mount;
align = ap->userdata ? xfs_get_extsz_hint(ap->ip) : 0;
if (unlikely(align)) {
- error = xfs_bmap_extsize_align(mp, ap->gotp, ap->prevp,
+ error = xfs_bmap_extsize_align(mp, &ap->got, &ap->prev,
align, 0, ap->eof, 0, ap->conv,
- &ap->off, &ap->alen);
+ &ap->offset, &ap->length);
ASSERT(!error);
- ASSERT(ap->alen);
+ ASSERT(ap->length);
}
- nullfb = ap->firstblock == NULLFSBLOCK;
- fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp, ap->firstblock);
+ nullfb = *ap->firstblock == NULLFSBLOCK;
+ fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp, *ap->firstblock);
if (nullfb) {
if (ap->userdata && xfs_inode_is_filestream(ap->ip)) {
ag = xfs_filestream_lookup_ag(ap->ip);
ag = (ag != NULLAGNUMBER) ? ag : 0;
- ap->rval = XFS_AGB_TO_FSB(mp, ag, 0);
+ ap->blkno = XFS_AGB_TO_FSB(mp, ag, 0);
} else {
- ap->rval = XFS_INO_TO_FSB(mp, ap->ip->i_ino);
+ ap->blkno = XFS_INO_TO_FSB(mp, ap->ip->i_ino);
}
} else
- ap->rval = ap->firstblock;
+ ap->blkno = *ap->firstblock;
xfs_bmap_adjacent(ap);
/*
- * If allowed, use ap->rval; otherwise must use firstblock since
+ * If allowed, use ap->blkno; otherwise must use firstblock since
* it's in the right allocation group.
*/
- if (nullfb || XFS_FSB_TO_AGNO(mp, ap->rval) == fb_agno)
+ if (nullfb || XFS_FSB_TO_AGNO(mp, ap->blkno) == fb_agno)
;
else
- ap->rval = ap->firstblock;
+ ap->blkno = *ap->firstblock;
/*
* Normal allocation, done through xfs_alloc_vextent.
*/
tryagain = isaligned = 0;
+ memset(&args, 0, sizeof(args));
args.tp = ap->tp;
args.mp = mp;
- args.fsbno = ap->rval;
- args.maxlen = MIN(ap->alen, mp->m_sb.sb_agblocks);
- args.firstblock = ap->firstblock;
+ args.fsbno = ap->blkno;
+
+ /* Trim the allocation back to the maximum an AG can fit. */
+ args.maxlen = MIN(ap->length, XFS_ALLOC_AG_MAX_USABLE(mp));
+ args.firstblock = *ap->firstblock;
blen = 0;
if (nullfb) {
error = xfs_bmap_btalloc_nullfb(ap, &args, &blen);
if (error)
return error;
- } else if (ap->low) {
+ } else if (ap->flist->xbf_low) {
if (xfs_inode_is_filestream(ap->ip))
args.type = XFS_ALLOCTYPE_FIRST_AG;
else
@@ -2435,14 +2304,14 @@ xfs_bmap_btalloc(
/* apply extent size hints if obtained earlier */
if (unlikely(align)) {
args.prod = align;
- if ((args.mod = (xfs_extlen_t)do_mod(ap->off, args.prod)))
+ if ((args.mod = (xfs_extlen_t)do_mod(ap->offset, args.prod)))
args.mod = (xfs_extlen_t)(args.prod - args.mod);
} else if (mp->m_sb.sb_blocksize >= PAGE_CACHE_SIZE) {
args.prod = 1;
args.mod = 0;
} else {
args.prod = PAGE_CACHE_SIZE >> mp->m_sb.sb_blocklog;
- if ((args.mod = (xfs_extlen_t)(do_mod(ap->off, args.prod))))
+ if ((args.mod = (xfs_extlen_t)(do_mod(ap->offset, args.prod))))
args.mod = (xfs_extlen_t)(args.prod - args.mod);
}
/*
@@ -2454,15 +2323,15 @@ xfs_bmap_btalloc(
* is >= the stripe unit and the allocation offset is
* at the end of file.
*/
- if (!ap->low && ap->aeof) {
- if (!ap->off) {
+ if (!ap->flist->xbf_low && ap->aeof) {
+ if (!ap->offset) {
args.alignment = mp->m_dalign;
atype = args.type;
isaligned = 1;
/*
* Adjust for alignment
*/
- if (blen > args.alignment && blen <= ap->alen)
+ if (blen > args.alignment && blen <= args.maxlen)
args.minlen = blen - args.alignment;
args.minalignslop = 0;
} else {
@@ -2481,7 +2350,7 @@ xfs_bmap_btalloc(
* of minlen+alignment+slop doesn't go up
* between the calls.
*/
- if (blen > mp->m_dalign && blen <= ap->alen)
+ if (blen > mp->m_dalign && blen <= args.maxlen)
nextminlen = blen - mp->m_dalign;
else
nextminlen = args.minlen;
@@ -2508,7 +2377,7 @@ xfs_bmap_btalloc(
* turned on.
*/
args.type = atype;
- args.fsbno = ap->rval;
+ args.fsbno = ap->blkno;
args.alignment = mp->m_dalign;
args.minlen = nextminlen;
args.minalignslop = 0;
@@ -2522,7 +2391,7 @@ xfs_bmap_btalloc(
* try again.
*/
args.type = atype;
- args.fsbno = ap->rval;
+ args.fsbno = ap->blkno;
args.alignment = 0;
if ((error = xfs_alloc_vextent(&args)))
return error;
@@ -2531,7 +2400,7 @@ xfs_bmap_btalloc(
args.minlen > ap->minlen) {
args.minlen = ap->minlen;
args.type = XFS_ALLOCTYPE_START_BNO;
- args.fsbno = ap->rval;
+ args.fsbno = ap->blkno;
if ((error = xfs_alloc_vextent(&args)))
return error;
}
@@ -2542,13 +2411,26 @@ xfs_bmap_btalloc(
args.minleft = 0;
if ((error = xfs_alloc_vextent(&args)))
return error;
- ap->low = 1;
+ ap->flist->xbf_low = 1;
}
if (args.fsbno != NULLFSBLOCK) {
- ap->firstblock = ap->rval = args.fsbno;
+ /*
+ * check the allocation happened at the same or higher AG than
+ * the first block that was allocated.
+ */
+ ASSERT(*ap->firstblock == NULLFSBLOCK ||
+ XFS_FSB_TO_AGNO(mp, *ap->firstblock) ==
+ XFS_FSB_TO_AGNO(mp, args.fsbno) ||
+ (ap->flist->xbf_low &&
+ XFS_FSB_TO_AGNO(mp, *ap->firstblock) <
+ XFS_FSB_TO_AGNO(mp, args.fsbno)));
+
+ ap->blkno = args.fsbno;
+ if (*ap->firstblock == NULLFSBLOCK)
+ *ap->firstblock = args.fsbno;
ASSERT(nullfb || fb_agno == args.agno ||
- (ap->low && fb_agno < args.agno));
- ap->alen = args.len;
+ (ap->flist->xbf_low && fb_agno < args.agno));
+ ap->length = args.len;
ap->ip->i_d.di_nblocks += args.len;
xfs_trans_log_inode(ap->tp, ap->ip, XFS_ILOG_CORE);
if (ap->wasdel)
@@ -2562,8 +2444,8 @@ xfs_bmap_btalloc(
XFS_TRANS_DQ_BCOUNT,
(long) args.len);
} else {
- ap->rval = NULLFSBLOCK;
- ap->alen = 0;
+ ap->blkno = NULLFSBLOCK;
+ ap->length = 0;
}
return 0;
}
@@ -2620,8 +2502,9 @@ xfs_bmap_btree_to_extents(
if ((error = xfs_btree_check_lptr(cur, cbno, 1)))
return error;
#endif
- if ((error = xfs_btree_read_bufl(mp, tp, cbno, 0, &cbp,
- XFS_BMAP_BTREE_REF)))
+ error = xfs_btree_read_bufl(mp, tp, cbno, 0, &cbp, XFS_BMAP_BTREE_REF,
+ &xfs_bmbt_buf_ops);
+ if (error)
return error;
cblock = XFS_BUF_TO_BLOCK(cbp);
if ((error = xfs_btree_check_block(cur, cblock, 0, cbp)))
@@ -2648,13 +2531,12 @@ STATIC int /* error */
xfs_bmap_del_extent(
xfs_inode_t *ip, /* incore inode pointer */
xfs_trans_t *tp, /* current transaction pointer */
- xfs_extnum_t idx, /* extent number to update/delete */
+ xfs_extnum_t *idx, /* extent number to update/delete */
xfs_bmap_free_t *flist, /* list of extents to be freed */
xfs_btree_cur_t *cur, /* if null, not a btree */
xfs_bmbt_irec_t *del, /* data to remove from extents */
int *logflagsp, /* inode logging flags */
- int whichfork, /* data or attr fork */
- int rsvd) /* OK to allocate reserved blocks */
+ int whichfork) /* data or attr fork */
{
xfs_filblks_t da_new; /* new delay-alloc indirect blocks */
xfs_filblks_t da_old; /* old delay-alloc indirect blocks */
@@ -2685,10 +2567,10 @@ xfs_bmap_del_extent(
mp = ip->i_mount;
ifp = XFS_IFORK_PTR(ip, whichfork);
- ASSERT((idx >= 0) && (idx < ifp->if_bytes /
+ ASSERT((*idx >= 0) && (*idx < ifp->if_bytes /
(uint)sizeof(xfs_bmbt_rec_t)));
ASSERT(del->br_blockcount > 0);
- ep = xfs_iext_get_ext(ifp, idx);
+ ep = xfs_iext_get_ext(ifp, *idx);
xfs_bmbt_get_all(ep, &got);
ASSERT(got.br_startoff <= del->br_startoff);
del_endoff = del->br_startoff + del->br_blockcount;
@@ -2719,8 +2601,8 @@ xfs_bmap_del_extent(
len = del->br_blockcount;
do_div(bno, mp->m_sb.sb_rextsize);
do_div(len, mp->m_sb.sb_rextsize);
- if ((error = xfs_rtfree_extent(ip->i_transp, bno,
- (xfs_extlen_t)len)))
+ error = xfs_rtfree_extent(tp, bno, (xfs_extlen_t)len);
+ if (error)
goto done;
do_fx = 0;
nblks = len * mp->m_sb.sb_rextsize;
@@ -2762,11 +2644,12 @@ xfs_bmap_del_extent(
/*
* Matches the whole extent. Delete the entry.
*/
- xfs_iext_remove(ip, idx, 1,
+ xfs_iext_remove(ip, *idx, 1,
whichfork == XFS_ATTR_FORK ? BMAP_ATTRFORK : 0);
- ifp->if_lastex = idx;
+ --*idx;
if (delay)
break;
+
XFS_IFORK_NEXT_SET(ip, whichfork,
XFS_IFORK_NEXTENTS(ip, whichfork) - 1);
flags |= XFS_ILOG_CORE;
@@ -2783,21 +2666,20 @@ xfs_bmap_del_extent(
/*
* Deleting the first part of the extent.
*/
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
xfs_bmbt_set_startoff(ep, del_endoff);
temp = got.br_blockcount - del->br_blockcount;
xfs_bmbt_set_blockcount(ep, temp);
- ifp->if_lastex = idx;
if (delay) {
temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
da_old);
xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
da_new = temp;
break;
}
xfs_bmbt_set_startblock(ep, del_endblock);
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
if (!cur) {
flags |= xfs_ilog_fext(whichfork);
break;
@@ -2813,18 +2695,17 @@ xfs_bmap_del_extent(
* Deleting the last part of the extent.
*/
temp = got.br_blockcount - del->br_blockcount;
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
xfs_bmbt_set_blockcount(ep, temp);
- ifp->if_lastex = idx;
if (delay) {
temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
da_old);
xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
da_new = temp;
break;
}
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
if (!cur) {
flags |= xfs_ilog_fext(whichfork);
break;
@@ -2841,7 +2722,7 @@ xfs_bmap_del_extent(
* Deleting the middle of the extent.
*/
temp = del->br_startoff - got.br_startoff;
- trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
xfs_bmbt_set_blockcount(ep, temp);
new.br_startoff = del_endoff;
temp2 = got_endoff - del_endoff;
@@ -2928,9 +2809,9 @@ xfs_bmap_del_extent(
}
}
}
- trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
- xfs_iext_insert(ip, idx + 1, 1, &new, state);
- ifp->if_lastex = idx + 1;
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
+ xfs_iext_insert(ip, *idx + 1, 1, &new, state);
+ ++*idx;
break;
}
/*
@@ -2957,7 +2838,7 @@ xfs_bmap_del_extent(
ASSERT(da_old >= da_new);
if (da_old > da_new) {
xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS,
- (int64_t)(da_old - da_new), rsvd);
+ (int64_t)(da_old - da_new), 0);
}
done:
*logflagsp = flags;
@@ -3014,8 +2895,7 @@ xfs_bmap_extents_to_btree(
ifp = XFS_IFORK_PTR(ip, whichfork);
ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS);
- ASSERT(ifp->if_ext_max ==
- XFS_IFORK_SIZE(ip, whichfork) / (uint)sizeof(xfs_bmbt_rec_t));
+
/*
* Make space in the inode incore.
*/
@@ -3044,6 +2924,7 @@ xfs_bmap_extents_to_btree(
* Convert to a btree with two levels, one record in root.
*/
XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_BTREE);
+ memset(&args, 0, sizeof(args));
args.tp = tp;
args.mp = mp;
args.firstblock = *firstblock;
@@ -3058,8 +2939,6 @@ xfs_bmap_extents_to_btree(
args.fsbno = *firstblock;
}
args.minlen = args.maxlen = args.prod = 1;
- args.total = args.minleft = args.alignment = args.mod = args.isfl =
- args.minalignslop = 0;
args.wasdel = wasdel;
*logflagsp = 0;
if ((error = xfs_alloc_vextent(&args))) {
@@ -3083,6 +2962,7 @@ xfs_bmap_extents_to_btree(
/*
* Fill in the child block.
*/
+ abp->b_ops = &xfs_bmbt_buf_ops;
ablock = XFS_BUF_TO_BLOCK(abp);
ablock->bb_magic = cpu_to_be32(XFS_BMAP_MAGIC);
ablock->bb_level = 0;
@@ -3161,13 +3041,8 @@ xfs_bmap_forkoff_reset(
ip->i_d.di_format != XFS_DINODE_FMT_BTREE) {
uint dfl_forkoff = xfs_default_attroffset(ip) >> 3;
- if (dfl_forkoff > ip->i_d.di_forkoff) {
+ if (dfl_forkoff > ip->i_d.di_forkoff)
ip->i_d.di_forkoff = dfl_forkoff;
- ip->i_df.if_ext_max =
- XFS_IFORK_DSIZE(ip) / sizeof(xfs_bmbt_rec_t);
- ip->i_afp->if_ext_max =
- XFS_IFORK_ASIZE(ip) / sizeof(xfs_bmbt_rec_t);
- }
}
}
@@ -3194,8 +3069,7 @@ xfs_bmap_local_to_extents(
* We don't want to deal with the case of keeping inode data inline yet.
* So sending the data fork of a regular inode is invalid.
*/
- ASSERT(!((ip->i_d.di_mode & S_IFMT) == S_IFREG &&
- whichfork == XFS_DATA_FORK));
+ ASSERT(!(S_ISREG(ip->i_d.di_mode) && whichfork == XFS_DATA_FORK));
ifp = XFS_IFORK_PTR(ip, whichfork);
ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL);
flags = 0;
@@ -3205,6 +3079,7 @@ xfs_bmap_local_to_extents(
xfs_buf_t *bp; /* buffer for extent block */
xfs_bmbt_rec_host_t *ep;/* extent record pointer */
+ memset(&args, 0, sizeof(args));
args.tp = tp;
args.mp = ip->i_mount;
args.firstblock = *firstblock;
@@ -3222,8 +3097,6 @@ xfs_bmap_local_to_extents(
args.type = XFS_ALLOCTYPE_NEAR_BNO;
}
args.total = total;
- args.mod = args.minleft = args.alignment = args.wasdel =
- args.isfl = args.minalignslop = 0;
args.minlen = args.maxlen = args.prod = 1;
if ((error = xfs_alloc_vextent(&args)))
goto done;
@@ -3234,8 +3107,8 @@ xfs_bmap_local_to_extents(
ASSERT(args.len == 1);
*firstblock = args.fsbno;
bp = xfs_btree_get_bufl(args.mp, tp, args.fsbno, 0);
- memcpy((char *)XFS_BUF_PTR(bp), ifp->if_u1.if_data,
- ifp->if_bytes);
+ bp->b_ops = &xfs_bmbt_buf_ops;
+ memcpy(bp->b_addr, ifp->if_u1.if_data, ifp->if_bytes);
xfs_trans_log_buf(tp, bp, 0, ifp->if_bytes - 1);
xfs_bmap_forkoff_reset(args.mp, ip, whichfork);
xfs_idata_realloc(ip, -ifp->if_bytes, whichfork);
@@ -3341,7 +3214,7 @@ xfs_bmap_search_extents(
if (unlikely(!(gotp->br_startblock) && (*lastxp != NULLEXTNUM) &&
!(XFS_IS_REALTIME_INODE(ip) && fork == XFS_DATA_FORK))) {
- xfs_cmn_err(XFS_PTAG_FSBLOCK_ZERO, CE_ALERT, ip->i_mount,
+ xfs_alert_tag(ip->i_mount, XFS_PTAG_FSBLOCK_ZERO,
"Access to block zero in inode %llu "
"start_block: %llx start_off: %llx "
"blkcnt: %llx extent-state: %x lastx: %x\n",
@@ -3409,8 +3282,6 @@ xfs_bmap_add_attrfork(
int error; /* error return value */
ASSERT(XFS_IFORK_Q(ip) == 0);
- ASSERT(ip->i_df.if_ext_max ==
- XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t));
mp = ip->i_mount;
ASSERT(!XFS_NOT_DQATTACHED(mp, ip));
@@ -3441,7 +3312,7 @@ xfs_bmap_add_attrfork(
}
ASSERT(ip->i_d.di_anextents == 0);
- xfs_trans_ijoin_ref(tp, ip, XFS_ILOCK_EXCL);
+ xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
switch (ip->i_d.di_format) {
@@ -3465,12 +3336,9 @@ xfs_bmap_add_attrfork(
error = XFS_ERROR(EINVAL);
goto error1;
}
- ip->i_df.if_ext_max =
- XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t);
+
ASSERT(ip->i_afp == NULL);
ip->i_afp = kmem_zone_zalloc(xfs_ifork_zone, KM_SLEEP);
- ip->i_afp->if_ext_max =
- XFS_IFORK_ASIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t);
ip->i_afp->if_flags = XFS_IFEXTENTS;
logflags = 0;
xfs_bmap_init(&flist, &firstblock);
@@ -3514,20 +3382,17 @@ xfs_bmap_add_attrfork(
} else
spin_unlock(&mp->m_sb_lock);
}
- if ((error = xfs_bmap_finish(&tp, &flist, &committed)))
+
+ error = xfs_bmap_finish(&tp, &flist, &committed);
+ if (error)
goto error2;
- error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
- ASSERT(ip->i_df.if_ext_max ==
- XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t));
- return error;
+ return xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
error2:
xfs_bmap_cancel(&flist);
error1:
xfs_iunlock(ip, XFS_ILOCK_EXCL);
error0:
xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT);
- ASSERT(ip->i_df.if_ext_max ==
- XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t));
return error;
}
@@ -3753,42 +3618,122 @@ xfs_bmap_last_before(
return 0;
}
+STATIC int
+xfs_bmap_last_extent(
+ struct xfs_trans *tp,
+ struct xfs_inode *ip,
+ int whichfork,
+ struct xfs_bmbt_irec *rec,
+ int *is_empty)
+{
+ struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork);
+ int error;
+ int nextents;
+
+ if (!(ifp->if_flags & XFS_IFEXTENTS)) {
+ error = xfs_iread_extents(tp, ip, whichfork);
+ if (error)
+ return error;
+ }
+
+ nextents = ifp->if_bytes / sizeof(xfs_bmbt_rec_t);
+ if (nextents == 0) {
+ *is_empty = 1;
+ return 0;
+ }
+
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, nextents - 1), rec);
+ *is_empty = 0;
+ return 0;
+}
+
+/*
+ * Check the last inode extent to determine whether this allocation will result
+ * in blocks being allocated at the end of the file. When we allocate new data
+ * blocks at the end of the file which do not start at the previous data block,
+ * we will try to align the new blocks at stripe unit boundaries.
+ *
+ * Returns 0 in bma->aeof if the file (fork) is empty as any new write will be
+ * at, or past the EOF.
+ */
+STATIC int
+xfs_bmap_isaeof(
+ struct xfs_bmalloca *bma,
+ int whichfork)
+{
+ struct xfs_bmbt_irec rec;
+ int is_empty;
+ int error;
+
+ bma->aeof = 0;
+ error = xfs_bmap_last_extent(NULL, bma->ip, whichfork, &rec,
+ &is_empty);
+ if (error || is_empty)
+ return error;
+
+ /*
+ * Check if we are allocation or past the last extent, or at least into
+ * the last delayed allocated extent.
+ */
+ bma->aeof = bma->offset >= rec.br_startoff + rec.br_blockcount ||
+ (bma->offset >= rec.br_startoff &&
+ isnullstartblock(rec.br_startblock));
+ return 0;
+}
+
+/*
+ * Check if the endoff is outside the last extent. If so the caller will grow
+ * the allocation to a stripe unit boundary. All offsets are considered outside
+ * the end of file for an empty fork, so 1 is returned in *eof in that case.
+ */
+int
+xfs_bmap_eof(
+ struct xfs_inode *ip,
+ xfs_fileoff_t endoff,
+ int whichfork,
+ int *eof)
+{
+ struct xfs_bmbt_irec rec;
+ int error;
+
+ error = xfs_bmap_last_extent(NULL, ip, whichfork, &rec, eof);
+ if (error || *eof)
+ return error;
+
+ *eof = endoff >= rec.br_startoff + rec.br_blockcount;
+ return 0;
+}
+
/*
* Returns the file-relative block number of the first block past eof in
* the file. This is not based on i_size, it is based on the extent records.
* Returns 0 for local files, as they do not have extent records.
*/
-int /* error */
+int
xfs_bmap_last_offset(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode */
- xfs_fileoff_t *last_block, /* last block */
- int whichfork) /* data or attr fork */
+ struct xfs_trans *tp,
+ struct xfs_inode *ip,
+ xfs_fileoff_t *last_block,
+ int whichfork)
{
- xfs_bmbt_rec_host_t *ep; /* pointer to last extent */
- int error; /* error return value */
- xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_extnum_t nextents; /* number of extent entries */
+ struct xfs_bmbt_irec rec;
+ int is_empty;
+ int error;
+
+ *last_block = 0;
+
+ if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL)
+ return 0;
if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE &&
- XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
- XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_LOCAL)
+ XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS)
return XFS_ERROR(EIO);
- if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL) {
- *last_block = 0;
- return 0;
- }
- ifp = XFS_IFORK_PTR(ip, whichfork);
- if (!(ifp->if_flags & XFS_IFEXTENTS) &&
- (error = xfs_iread_extents(tp, ip, whichfork)))
+
+ error = xfs_bmap_last_extent(NULL, ip, whichfork, &rec, &is_empty);
+ if (error || is_empty)
return error;
- nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
- if (!nextents) {
- *last_block = 0;
- return 0;
- }
- ep = xfs_iext_get_ext(ifp, nextents - 1);
- *last_block = xfs_bmbt_get_startoff(ep) + xfs_bmbt_get_blockcount(ep);
+
+ *last_block = rec.br_startoff + rec.br_blockcount;
return 0;
}
@@ -3808,11 +3753,8 @@ xfs_bmap_one_block(
xfs_bmbt_irec_t s; /* internal version of extent */
#ifndef DEBUG
- if (whichfork == XFS_DATA_FORK) {
- return ((ip->i_d.di_mode & S_IFMT) == S_IFREG) ?
- (ip->i_size == ip->i_mount->m_sb.sb_blocksize) :
- (ip->i_d.di_size == ip->i_mount->m_sb.sb_blocksize);
- }
+ if (whichfork == XFS_DATA_FORK)
+ return XFS_ISIZE(ip) == ip->i_mount->m_sb.sb_blocksize;
#endif /* !DEBUG */
if (XFS_IFORK_NEXTENTS(ip, whichfork) != 1)
return 0;
@@ -3824,7 +3766,7 @@ xfs_bmap_one_block(
xfs_bmbt_get_all(ep, &s);
rval = s.br_startoff == 0 && s.br_blockcount == 1;
if (rval && whichfork == XFS_DATA_FORK)
- ASSERT(ip->i_size == ip->i_mount->m_sb.sb_blocksize);
+ ASSERT(XFS_ISIZE(ip) == ip->i_mount->m_sb.sb_blocksize);
return rval;
}
@@ -3836,7 +3778,7 @@ xfs_bmap_sanity_check(
{
struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
- if (be32_to_cpu(block->bb_magic) != XFS_BMAP_MAGIC ||
+ if (block->bb_magic != cpu_to_be32(XFS_BMAP_MAGIC) ||
be16_to_cpu(block->bb_level) != level ||
be16_to_cpu(block->bb_numrecs) == 0 ||
be16_to_cpu(block->bb_numrecs) > mp->m_bmap_dmxr[level != 0])
@@ -3890,8 +3832,9 @@ xfs_bmap_read_extents(
* pointer (leftmost) at each level.
*/
while (level-- > 0) {
- if ((error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp,
- XFS_BMAP_BTREE_REF)))
+ error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp,
+ XFS_BMAP_BTREE_REF, &xfs_bmbt_buf_ops);
+ if (error)
return error;
block = XFS_BUF_TO_BLOCK(bp);
XFS_WANT_CORRUPTED_GOTO(
@@ -3918,16 +3861,14 @@ xfs_bmap_read_extents(
xfs_extnum_t num_recs;
xfs_extnum_t start;
-
num_recs = xfs_btree_get_numrecs(block);
if (unlikely(i + num_recs > room)) {
ASSERT(i + num_recs <= room);
- xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount,
+ xfs_warn(ip->i_mount,
"corrupt dinode %Lu, (btree extents).",
(unsigned long long) ip->i_ino);
- XFS_ERROR_REPORT("xfs_bmap_read_extents(1)",
- XFS_ERRLEVEL_LOW,
- ip->i_mount);
+ XFS_CORRUPTION_ERROR("xfs_bmap_read_extents(1)",
+ XFS_ERRLEVEL_LOW, ip->i_mount, block);
goto error0;
}
XFS_WANT_CORRUPTED_GOTO(
@@ -3938,7 +3879,8 @@ xfs_bmap_read_extents(
*/
nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
if (nextbno != NULLFSBLOCK)
- xfs_btree_reada_bufl(mp, nextbno, 1);
+ xfs_btree_reada_bufl(mp, nextbno, 1,
+ &xfs_bmbt_buf_ops);
/*
* Copy records into the extent records.
*/
@@ -3970,8 +3912,9 @@ xfs_bmap_read_extents(
*/
if (bno == NULLFSBLOCK)
break;
- if ((error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp,
- XFS_BMAP_BTREE_REF)))
+ error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp,
+ XFS_BMAP_BTREE_REF, &xfs_bmbt_buf_ops);
+ if (error)
return error;
block = XFS_BUF_TO_BLOCK(bp);
}
@@ -4042,9 +3985,8 @@ xfs_bmap_validate_ret(
ASSERT(i == 0 ||
mval[i - 1].br_startoff + mval[i - 1].br_blockcount ==
mval[i].br_startoff);
- if ((flags & XFS_BMAPI_WRITE) && !(flags & XFS_BMAPI_DELAY))
- ASSERT(mval[i].br_startblock != DELAYSTARTBLOCK &&
- mval[i].br_startblock != HOLESTARTBLOCK);
+ ASSERT(mval[i].br_startblock != DELAYSTARTBLOCK &&
+ mval[i].br_startblock != HOLESTARTBLOCK);
ASSERT(mval[i].br_state == XFS_EXT_NORM ||
mval[i].br_state == XFS_EXT_UNWRITTEN);
}
@@ -4053,395 +3995,166 @@ xfs_bmap_validate_ret(
/*
- * Map file blocks to filesystem blocks.
- * File range is given by the bno/len pair.
- * Adds blocks to file if a write ("flags & XFS_BMAPI_WRITE" set)
- * into a hole or past eof.
- * Only allocates blocks from a single allocation group,
- * to avoid locking problems.
- * The returned value in "firstblock" from the first call in a transaction
- * must be remembered and presented to subsequent calls in "firstblock".
- * An upper bound for the number of blocks to be allocated is supplied to
- * the first call in "total"; if no allocation group has that many free
- * blocks then the call will fail (return NULLFSBLOCK in "firstblock").
+ * Trim the returned map to the required bounds
*/
-int /* error */
-xfs_bmapi(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode */
- xfs_fileoff_t bno, /* starting file offs. mapped */
- xfs_filblks_t len, /* length to map in file */
- int flags, /* XFS_BMAPI_... */
- xfs_fsblock_t *firstblock, /* first allocated block
- controls a.g. for allocs */
- xfs_extlen_t total, /* total blocks needed */
- xfs_bmbt_irec_t *mval, /* output: map values */
- int *nmap, /* i/o: mval size/count */
- xfs_bmap_free_t *flist) /* i/o: list extents to free */
+STATIC void
+xfs_bmapi_trim_map(
+ struct xfs_bmbt_irec *mval,
+ struct xfs_bmbt_irec *got,
+ xfs_fileoff_t *bno,
+ xfs_filblks_t len,
+ xfs_fileoff_t obno,
+ xfs_fileoff_t end,
+ int n,
+ int flags)
{
- xfs_fsblock_t abno; /* allocated block number */
- xfs_extlen_t alen; /* allocated extent length */
- xfs_fileoff_t aoff; /* allocated file offset */
- xfs_bmalloca_t bma = { 0 }; /* args for xfs_bmap_alloc */
- xfs_btree_cur_t *cur; /* bmap btree cursor */
- xfs_fileoff_t end; /* end of mapped file region */
- int eof; /* we've hit the end of extents */
- xfs_bmbt_rec_host_t *ep; /* extent record pointer */
- int error; /* error return */
- xfs_bmbt_irec_t got; /* current file extent record */
- xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_extlen_t indlen; /* indirect blocks length */
- xfs_extnum_t lastx; /* last useful extent number */
- int logflags; /* flags for transaction logging */
- xfs_extlen_t minleft; /* min blocks left after allocation */
- xfs_extlen_t minlen; /* min allocation size */
- xfs_mount_t *mp; /* xfs mount structure */
- int n; /* current extent index */
- int nallocs; /* number of extents alloc'd */
- xfs_extnum_t nextents; /* number of extents in file */
- xfs_fileoff_t obno; /* old block number (offset) */
- xfs_bmbt_irec_t prev; /* previous file extent record */
- int tmp_logflags; /* temp flags holder */
- int whichfork; /* data or attr fork */
- char inhole; /* current location is hole in file */
- char wasdelay; /* old extent was delayed */
- char wr; /* this is a write request */
- char rt; /* this is a realtime file */
-#ifdef DEBUG
- xfs_fileoff_t orig_bno; /* original block number value */
- int orig_flags; /* original flags arg value */
- xfs_filblks_t orig_len; /* original value of len arg */
- xfs_bmbt_irec_t *orig_mval; /* original value of mval */
- int orig_nmap; /* original value of *nmap */
-
- orig_bno = bno;
- orig_len = len;
- orig_flags = flags;
- orig_mval = mval;
- orig_nmap = *nmap;
-#endif
- ASSERT(*nmap >= 1);
- ASSERT(*nmap <= XFS_BMAP_MAX_NMAP || !(flags & XFS_BMAPI_WRITE));
- whichfork = (flags & XFS_BMAPI_ATTRFORK) ?
- XFS_ATTR_FORK : XFS_DATA_FORK;
- mp = ip->i_mount;
- if (unlikely(XFS_TEST_ERROR(
- (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
- XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE &&
- XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_LOCAL),
- mp, XFS_ERRTAG_BMAPIFORMAT, XFS_RANDOM_BMAPIFORMAT))) {
- XFS_ERROR_REPORT("xfs_bmapi", XFS_ERRLEVEL_LOW, mp);
- return XFS_ERROR(EFSCORRUPTED);
+ if ((flags & XFS_BMAPI_ENTIRE) ||
+ got->br_startoff + got->br_blockcount <= obno) {
+ *mval = *got;
+ if (isnullstartblock(got->br_startblock))
+ mval->br_startblock = DELAYSTARTBLOCK;
+ return;
}
- if (XFS_FORCED_SHUTDOWN(mp))
- return XFS_ERROR(EIO);
- rt = (whichfork == XFS_DATA_FORK) && XFS_IS_REALTIME_INODE(ip);
- ifp = XFS_IFORK_PTR(ip, whichfork);
- ASSERT(ifp->if_ext_max ==
- XFS_IFORK_SIZE(ip, whichfork) / (uint)sizeof(xfs_bmbt_rec_t));
- if ((wr = (flags & XFS_BMAPI_WRITE)) != 0)
- XFS_STATS_INC(xs_blk_mapw);
+
+ if (obno > *bno)
+ *bno = obno;
+ ASSERT((*bno >= obno) || (n == 0));
+ ASSERT(*bno < end);
+ mval->br_startoff = *bno;
+ if (isnullstartblock(got->br_startblock))
+ mval->br_startblock = DELAYSTARTBLOCK;
else
- XFS_STATS_INC(xs_blk_mapr);
+ mval->br_startblock = got->br_startblock +
+ (*bno - got->br_startoff);
/*
- * IGSTATE flag is used to combine extents which
- * differ only due to the state of the extents.
- * This technique is used from xfs_getbmap()
- * when the caller does not wish to see the
- * separation (which is the default).
- *
- * This technique is also used when writing a
- * buffer which has been partially written,
- * (usually by being flushed during a chunkread),
- * to ensure one write takes place. This also
- * prevents a change in the xfs inode extents at
- * this time, intentionally. This change occurs
- * on completion of the write operation, in
- * xfs_strat_comp(), where the xfs_bmapi() call
- * is transactioned, and the extents combined.
+ * Return the minimum of what we got and what we asked for for
+ * the length. We can use the len variable here because it is
+ * modified below and we could have been there before coming
+ * here if the first part of the allocation didn't overlap what
+ * was asked for.
*/
- if ((flags & XFS_BMAPI_IGSTATE) && wr) /* if writing unwritten space */
- wr = 0; /* no allocations are allowed */
- ASSERT(wr || !(flags & XFS_BMAPI_DELAY));
- logflags = 0;
- nallocs = 0;
- cur = NULL;
- if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL) {
- ASSERT(wr && tp);
- if ((error = xfs_bmap_local_to_extents(tp, ip,
- firstblock, total, &logflags, whichfork)))
- goto error0;
- }
- if (wr && *firstblock == NULLFSBLOCK) {
- if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE)
- minleft = be16_to_cpu(ifp->if_broot->bb_level) + 1;
- else
- minleft = 1;
- } else
- minleft = 0;
- if (!(ifp->if_flags & XFS_IFEXTENTS) &&
- (error = xfs_iread_extents(tp, ip, whichfork)))
- goto error0;
- ep = xfs_bmap_search_extents(ip, bno, whichfork, &eof, &lastx, &got,
- &prev);
- nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
- n = 0;
- end = bno + len;
- obno = bno;
- bma.ip = NULL;
+ mval->br_blockcount = XFS_FILBLKS_MIN(end - *bno,
+ got->br_blockcount - (*bno - got->br_startoff));
+ mval->br_state = got->br_state;
+ ASSERT(mval->br_blockcount <= len);
+ return;
+}
- while (bno < end && n < *nmap) {
- /*
- * Reading past eof, act as though there's a hole
- * up to end.
- */
- if (eof && !wr)
- got.br_startoff = end;
- inhole = eof || got.br_startoff > bno;
- wasdelay = wr && !inhole && !(flags & XFS_BMAPI_DELAY) &&
- isnullstartblock(got.br_startblock);
- /*
- * First, deal with the hole before the allocated space
- * that we found, if any.
- */
- if (wr && (inhole || wasdelay)) {
- /*
- * For the wasdelay case, we could also just
- * allocate the stuff asked for in this bmap call
- * but that wouldn't be as good.
- */
- if (wasdelay) {
- alen = (xfs_extlen_t)got.br_blockcount;
- aoff = got.br_startoff;
- if (lastx != NULLEXTNUM && lastx) {
- ep = xfs_iext_get_ext(ifp, lastx - 1);
- xfs_bmbt_get_all(ep, &prev);
- }
- } else {
- alen = (xfs_extlen_t)
- XFS_FILBLKS_MIN(len, MAXEXTLEN);
- if (!eof)
- alen = (xfs_extlen_t)
- XFS_FILBLKS_MIN(alen,
- got.br_startoff - bno);
- aoff = bno;
- }
- minlen = (flags & XFS_BMAPI_CONTIG) ? alen : 1;
- if (flags & XFS_BMAPI_DELAY) {
- xfs_extlen_t extsz;
-
- /* Figure out the extent size, adjust alen */
- extsz = xfs_get_extsz_hint(ip);
- if (extsz) {
- error = xfs_bmap_extsize_align(mp,
- &got, &prev, extsz,
- rt, eof,
- flags&XFS_BMAPI_DELAY,
- flags&XFS_BMAPI_CONVERT,
- &aoff, &alen);
- ASSERT(!error);
- }
+/*
+ * Update and validate the extent map to return
+ */
+STATIC void
+xfs_bmapi_update_map(
+ struct xfs_bmbt_irec **map,
+ xfs_fileoff_t *bno,
+ xfs_filblks_t *len,
+ xfs_fileoff_t obno,
+ xfs_fileoff_t end,
+ int *n,
+ int flags)
+{
+ xfs_bmbt_irec_t *mval = *map;
+
+ ASSERT((flags & XFS_BMAPI_ENTIRE) ||
+ ((mval->br_startoff + mval->br_blockcount) <= end));
+ ASSERT((flags & XFS_BMAPI_ENTIRE) || (mval->br_blockcount <= *len) ||
+ (mval->br_startoff < obno));
+
+ *bno = mval->br_startoff + mval->br_blockcount;
+ *len = end - *bno;
+ if (*n > 0 && mval->br_startoff == mval[-1].br_startoff) {
+ /* update previous map with new information */
+ ASSERT(mval->br_startblock == mval[-1].br_startblock);
+ ASSERT(mval->br_blockcount > mval[-1].br_blockcount);
+ ASSERT(mval->br_state == mval[-1].br_state);
+ mval[-1].br_blockcount = mval->br_blockcount;
+ mval[-1].br_state = mval->br_state;
+ } else if (*n > 0 && mval->br_startblock != DELAYSTARTBLOCK &&
+ mval[-1].br_startblock != DELAYSTARTBLOCK &&
+ mval[-1].br_startblock != HOLESTARTBLOCK &&
+ mval->br_startblock == mval[-1].br_startblock +
+ mval[-1].br_blockcount &&
+ ((flags & XFS_BMAPI_IGSTATE) ||
+ mval[-1].br_state == mval->br_state)) {
+ ASSERT(mval->br_startoff ==
+ mval[-1].br_startoff + mval[-1].br_blockcount);
+ mval[-1].br_blockcount += mval->br_blockcount;
+ } else if (*n > 0 &&
+ mval->br_startblock == DELAYSTARTBLOCK &&
+ mval[-1].br_startblock == DELAYSTARTBLOCK &&
+ mval->br_startoff ==
+ mval[-1].br_startoff + mval[-1].br_blockcount) {
+ mval[-1].br_blockcount += mval->br_blockcount;
+ mval[-1].br_state = mval->br_state;
+ } else if (!((*n == 0) &&
+ ((mval->br_startoff + mval->br_blockcount) <=
+ obno))) {
+ mval++;
+ (*n)++;
+ }
+ *map = mval;
+}
- if (rt)
- extsz = alen / mp->m_sb.sb_rextsize;
+/*
+ * Map file blocks to filesystem blocks without allocation.
+ */
+int
+xfs_bmapi_read(
+ struct xfs_inode *ip,
+ xfs_fileoff_t bno,
+ xfs_filblks_t len,
+ struct xfs_bmbt_irec *mval,
+ int *nmap,
+ int flags)
+{
+ struct xfs_mount *mp = ip->i_mount;
+ struct xfs_ifork *ifp;
+ struct xfs_bmbt_irec got;
+ struct xfs_bmbt_irec prev;
+ xfs_fileoff_t obno;
+ xfs_fileoff_t end;
+ xfs_extnum_t lastx;
+ int error;
+ int eof;
+ int n = 0;
+ int whichfork = (flags & XFS_BMAPI_ATTRFORK) ?
+ XFS_ATTR_FORK : XFS_DATA_FORK;
- /*
- * Make a transaction-less quota reservation for
- * delayed allocation blocks. This number gets
- * adjusted later. We return if we haven't
- * allocated blocks already inside this loop.
- */
- error = xfs_trans_reserve_quota_nblks(
- NULL, ip, (long)alen, 0,
- rt ? XFS_QMOPT_RES_RTBLKS :
- XFS_QMOPT_RES_REGBLKS);
- if (error) {
- if (n == 0) {
- *nmap = 0;
- ASSERT(cur == NULL);
- return error;
- }
- break;
- }
+ ASSERT(*nmap >= 1);
+ ASSERT(!(flags & ~(XFS_BMAPI_ATTRFORK|XFS_BMAPI_ENTIRE|
+ XFS_BMAPI_IGSTATE)));
- /*
- * Split changing sb for alen and indlen since
- * they could be coming from different places.
- */
- indlen = (xfs_extlen_t)
- xfs_bmap_worst_indlen(ip, alen);
- ASSERT(indlen > 0);
-
- if (rt) {
- error = xfs_mod_incore_sb(mp,
- XFS_SBS_FREXTENTS,
- -((int64_t)extsz), (flags &
- XFS_BMAPI_RSVBLOCKS));
- } else {
- error = xfs_icsb_modify_counters(mp,
- XFS_SBS_FDBLOCKS,
- -((int64_t)alen), (flags &
- XFS_BMAPI_RSVBLOCKS));
- }
- if (!error) {
- error = xfs_icsb_modify_counters(mp,
- XFS_SBS_FDBLOCKS,
- -((int64_t)indlen), (flags &
- XFS_BMAPI_RSVBLOCKS));
- if (error && rt)
- xfs_mod_incore_sb(mp,
- XFS_SBS_FREXTENTS,
- (int64_t)extsz, (flags &
- XFS_BMAPI_RSVBLOCKS));
- else if (error)
- xfs_icsb_modify_counters(mp,
- XFS_SBS_FDBLOCKS,
- (int64_t)alen, (flags &
- XFS_BMAPI_RSVBLOCKS));
- }
+ if (unlikely(XFS_TEST_ERROR(
+ (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
+ XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
+ mp, XFS_ERRTAG_BMAPIFORMAT, XFS_RANDOM_BMAPIFORMAT))) {
+ XFS_ERROR_REPORT("xfs_bmapi_read", XFS_ERRLEVEL_LOW, mp);
+ return XFS_ERROR(EFSCORRUPTED);
+ }
- if (error) {
- if (XFS_IS_QUOTA_ON(mp))
- /* unreserve the blocks now */
- (void)
- xfs_trans_unreserve_quota_nblks(
- NULL, ip,
- (long)alen, 0, rt ?
- XFS_QMOPT_RES_RTBLKS :
- XFS_QMOPT_RES_REGBLKS);
- break;
- }
+ if (XFS_FORCED_SHUTDOWN(mp))
+ return XFS_ERROR(EIO);
- ip->i_delayed_blks += alen;
- abno = nullstartblock(indlen);
- } else {
- /*
- * If first time, allocate and fill in
- * once-only bma fields.
- */
- if (bma.ip == NULL) {
- bma.tp = tp;
- bma.ip = ip;
- bma.prevp = &prev;
- bma.gotp = &got;
- bma.total = total;
- bma.userdata = 0;
- }
- /* Indicate if this is the first user data
- * in the file, or just any user data.
- */
- if (!(flags & XFS_BMAPI_METADATA)) {
- bma.userdata = (aoff == 0) ?
- XFS_ALLOC_INITIAL_USER_DATA :
- XFS_ALLOC_USERDATA;
- }
- /*
- * Fill in changeable bma fields.
- */
- bma.eof = eof;
- bma.firstblock = *firstblock;
- bma.alen = alen;
- bma.off = aoff;
- bma.conv = !!(flags & XFS_BMAPI_CONVERT);
- bma.wasdel = wasdelay;
- bma.minlen = minlen;
- bma.low = flist->xbf_low;
- bma.minleft = minleft;
- /*
- * Only want to do the alignment at the
- * eof if it is userdata and allocation length
- * is larger than a stripe unit.
- */
- if (mp->m_dalign && alen >= mp->m_dalign &&
- (!(flags & XFS_BMAPI_METADATA)) &&
- (whichfork == XFS_DATA_FORK)) {
- if ((error = xfs_bmap_isaeof(ip, aoff,
- whichfork, &bma.aeof)))
- goto error0;
- } else
- bma.aeof = 0;
- /*
- * Call allocator.
- */
- if ((error = xfs_bmap_alloc(&bma)))
- goto error0;
- /*
- * Copy out result fields.
- */
- abno = bma.rval;
- if ((flist->xbf_low = bma.low))
- minleft = 0;
- alen = bma.alen;
- aoff = bma.off;
- ASSERT(*firstblock == NULLFSBLOCK ||
- XFS_FSB_TO_AGNO(mp, *firstblock) ==
- XFS_FSB_TO_AGNO(mp, bma.firstblock) ||
- (flist->xbf_low &&
- XFS_FSB_TO_AGNO(mp, *firstblock) <
- XFS_FSB_TO_AGNO(mp, bma.firstblock)));
- *firstblock = bma.firstblock;
- if (cur)
- cur->bc_private.b.firstblock =
- *firstblock;
- if (abno == NULLFSBLOCK)
- break;
- if ((ifp->if_flags & XFS_IFBROOT) && !cur) {
- cur = xfs_bmbt_init_cursor(mp, tp,
- ip, whichfork);
- cur->bc_private.b.firstblock =
- *firstblock;
- cur->bc_private.b.flist = flist;
- }
- /*
- * Bump the number of extents we've allocated
- * in this call.
- */
- nallocs++;
- }
- if (cur)
- cur->bc_private.b.flags =
- wasdelay ? XFS_BTCUR_BPRV_WASDEL : 0;
- got.br_startoff = aoff;
- got.br_startblock = abno;
- got.br_blockcount = alen;
- got.br_state = XFS_EXT_NORM; /* assume normal */
- /*
- * Determine state of extent, and the filesystem.
- * A wasdelay extent has been initialized, so
- * shouldn't be flagged as unwritten.
- */
- if (wr && xfs_sb_version_hasextflgbit(&mp->m_sb)) {
- if (!wasdelay && (flags & XFS_BMAPI_PREALLOC))
- got.br_state = XFS_EXT_UNWRITTEN;
- }
- error = xfs_bmap_add_extent(ip, lastx, &cur, &got,
- firstblock, flist, &tmp_logflags,
- whichfork, (flags & XFS_BMAPI_RSVBLOCKS));
- logflags |= tmp_logflags;
- if (error)
- goto error0;
- lastx = ifp->if_lastex;
- ep = xfs_iext_get_ext(ifp, lastx);
- nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
- xfs_bmbt_get_all(ep, &got);
- ASSERT(got.br_startoff <= aoff);
- ASSERT(got.br_startoff + got.br_blockcount >=
- aoff + alen);
-#ifdef DEBUG
- if (flags & XFS_BMAPI_DELAY) {
- ASSERT(isnullstartblock(got.br_startblock));
- ASSERT(startblockval(got.br_startblock) > 0);
- }
- ASSERT(got.br_state == XFS_EXT_NORM ||
- got.br_state == XFS_EXT_UNWRITTEN);
-#endif
- /*
- * Fall down into the found allocated space case.
- */
- } else if (inhole) {
- /*
- * Reading in a hole.
- */
+ XFS_STATS_INC(xs_blk_mapr);
+
+ ifp = XFS_IFORK_PTR(ip, whichfork);
+
+ if (!(ifp->if_flags & XFS_IFEXTENTS)) {
+ error = xfs_iread_extents(NULL, ip, whichfork);
+ if (error)
+ return error;
+ }
+
+ xfs_bmap_search_extents(ip, bno, whichfork, &eof, &lastx, &got, &prev);
+ end = bno + len;
+ obno = bno;
+
+ while (bno < end && n < *nmap) {
+ /* Reading past eof, act as though there's a hole up to end. */
+ if (eof)
+ got.br_startoff = end;
+ if (got.br_startoff > bno) {
+ /* Reading in a hole. */
mval->br_startoff = bno;
mval->br_startblock = HOLESTARTBLOCK;
mval->br_blockcount =
@@ -4453,193 +4166,624 @@ xfs_bmapi(
n++;
continue;
}
+
+ /* set up the extent map to return. */
+ xfs_bmapi_trim_map(mval, &got, &bno, len, obno, end, n, flags);
+ xfs_bmapi_update_map(&mval, &bno, &len, obno, end, &n, flags);
+
+ /* If we're done, stop now. */
+ if (bno >= end || n >= *nmap)
+ break;
+
+ /* Else go on to the next record. */
+ if (++lastx < ifp->if_bytes / sizeof(xfs_bmbt_rec_t))
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, lastx), &got);
+ else
+ eof = 1;
+ }
+ *nmap = n;
+ return 0;
+}
+
+STATIC int
+xfs_bmapi_reserve_delalloc(
+ struct xfs_inode *ip,
+ xfs_fileoff_t aoff,
+ xfs_filblks_t len,
+ struct xfs_bmbt_irec *got,
+ struct xfs_bmbt_irec *prev,
+ xfs_extnum_t *lastx,
+ int eof)
+{
+ struct xfs_mount *mp = ip->i_mount;
+ struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
+ xfs_extlen_t alen;
+ xfs_extlen_t indlen;
+ char rt = XFS_IS_REALTIME_INODE(ip);
+ xfs_extlen_t extsz;
+ int error;
+
+ alen = XFS_FILBLKS_MIN(len, MAXEXTLEN);
+ if (!eof)
+ alen = XFS_FILBLKS_MIN(alen, got->br_startoff - aoff);
+
+ /* Figure out the extent size, adjust alen */
+ extsz = xfs_get_extsz_hint(ip);
+ if (extsz) {
/*
- * Then deal with the allocated space we found.
+ * Make sure we don't exceed a single extent length when we
+ * align the extent by reducing length we are going to
+ * allocate by the maximum amount extent size aligment may
+ * require.
*/
- ASSERT(ep != NULL);
- if (!(flags & XFS_BMAPI_ENTIRE) &&
- (got.br_startoff + got.br_blockcount > obno)) {
- if (obno > bno)
- bno = obno;
- ASSERT((bno >= obno) || (n == 0));
- ASSERT(bno < end);
- mval->br_startoff = bno;
- if (isnullstartblock(got.br_startblock)) {
- ASSERT(!wr || (flags & XFS_BMAPI_DELAY));
- mval->br_startblock = DELAYSTARTBLOCK;
- } else
- mval->br_startblock =
- got.br_startblock +
- (bno - got.br_startoff);
- /*
- * Return the minimum of what we got and what we
- * asked for for the length. We can use the len
- * variable here because it is modified below
- * and we could have been there before coming
- * here if the first part of the allocation
- * didn't overlap what was asked for.
- */
- mval->br_blockcount =
- XFS_FILBLKS_MIN(end - bno, got.br_blockcount -
- (bno - got.br_startoff));
- mval->br_state = got.br_state;
- ASSERT(mval->br_blockcount <= len);
- } else {
- *mval = got;
- if (isnullstartblock(mval->br_startblock)) {
- ASSERT(!wr || (flags & XFS_BMAPI_DELAY));
- mval->br_startblock = DELAYSTARTBLOCK;
+ alen = XFS_FILBLKS_MIN(len, MAXEXTLEN - (2 * extsz - 1));
+ error = xfs_bmap_extsize_align(mp, got, prev, extsz, rt, eof,
+ 1, 0, &aoff, &alen);
+ ASSERT(!error);
+ }
+
+ if (rt)
+ extsz = alen / mp->m_sb.sb_rextsize;
+
+ /*
+ * Make a transaction-less quota reservation for delayed allocation
+ * blocks. This number gets adjusted later. We return if we haven't
+ * allocated blocks already inside this loop.
+ */
+ error = xfs_trans_reserve_quota_nblks(NULL, ip, (long)alen, 0,
+ rt ? XFS_QMOPT_RES_RTBLKS : XFS_QMOPT_RES_REGBLKS);
+ if (error)
+ return error;
+
+ /*
+ * Split changing sb for alen and indlen since they could be coming
+ * from different places.
+ */
+ indlen = (xfs_extlen_t)xfs_bmap_worst_indlen(ip, alen);
+ ASSERT(indlen > 0);
+
+ if (rt) {
+ error = xfs_mod_incore_sb(mp, XFS_SBS_FREXTENTS,
+ -((int64_t)extsz), 0);
+ } else {
+ error = xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS,
+ -((int64_t)alen), 0);
+ }
+
+ if (error)
+ goto out_unreserve_quota;
+
+ error = xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS,
+ -((int64_t)indlen), 0);
+ if (error)
+ goto out_unreserve_blocks;
+
+
+ ip->i_delayed_blks += alen;
+
+ got->br_startoff = aoff;
+ got->br_startblock = nullstartblock(indlen);
+ got->br_blockcount = alen;
+ got->br_state = XFS_EXT_NORM;
+ xfs_bmap_add_extent_hole_delay(ip, lastx, got);
+
+ /*
+ * Update our extent pointer, given that xfs_bmap_add_extent_hole_delay
+ * might have merged it into one of the neighbouring ones.
+ */
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *lastx), got);
+
+ ASSERT(got->br_startoff <= aoff);
+ ASSERT(got->br_startoff + got->br_blockcount >= aoff + alen);
+ ASSERT(isnullstartblock(got->br_startblock));
+ ASSERT(got->br_state == XFS_EXT_NORM);
+ return 0;
+
+out_unreserve_blocks:
+ if (rt)
+ xfs_mod_incore_sb(mp, XFS_SBS_FREXTENTS, extsz, 0);
+ else
+ xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS, alen, 0);
+out_unreserve_quota:
+ if (XFS_IS_QUOTA_ON(mp))
+ xfs_trans_unreserve_quota_nblks(NULL, ip, (long)alen, 0, rt ?
+ XFS_QMOPT_RES_RTBLKS : XFS_QMOPT_RES_REGBLKS);
+ return error;
+}
+
+/*
+ * Map file blocks to filesystem blocks, adding delayed allocations as needed.
+ */
+int
+xfs_bmapi_delay(
+ struct xfs_inode *ip, /* incore inode */
+ xfs_fileoff_t bno, /* starting file offs. mapped */
+ xfs_filblks_t len, /* length to map in file */
+ struct xfs_bmbt_irec *mval, /* output: map values */
+ int *nmap, /* i/o: mval size/count */
+ int flags) /* XFS_BMAPI_... */
+{
+ struct xfs_mount *mp = ip->i_mount;
+ struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
+ struct xfs_bmbt_irec got; /* current file extent record */
+ struct xfs_bmbt_irec prev; /* previous file extent record */
+ xfs_fileoff_t obno; /* old block number (offset) */
+ xfs_fileoff_t end; /* end of mapped file region */
+ xfs_extnum_t lastx; /* last useful extent number */
+ int eof; /* we've hit the end of extents */
+ int n = 0; /* current extent index */
+ int error = 0;
+
+ ASSERT(*nmap >= 1);
+ ASSERT(*nmap <= XFS_BMAP_MAX_NMAP);
+ ASSERT(!(flags & ~XFS_BMAPI_ENTIRE));
+
+ if (unlikely(XFS_TEST_ERROR(
+ (XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) != XFS_DINODE_FMT_EXTENTS &&
+ XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) != XFS_DINODE_FMT_BTREE),
+ mp, XFS_ERRTAG_BMAPIFORMAT, XFS_RANDOM_BMAPIFORMAT))) {
+ XFS_ERROR_REPORT("xfs_bmapi_delay", XFS_ERRLEVEL_LOW, mp);
+ return XFS_ERROR(EFSCORRUPTED);
+ }
+
+ if (XFS_FORCED_SHUTDOWN(mp))
+ return XFS_ERROR(EIO);
+
+ XFS_STATS_INC(xs_blk_mapw);
+
+ if (!(ifp->if_flags & XFS_IFEXTENTS)) {
+ error = xfs_iread_extents(NULL, ip, XFS_DATA_FORK);
+ if (error)
+ return error;
+ }
+
+ xfs_bmap_search_extents(ip, bno, XFS_DATA_FORK, &eof, &lastx, &got, &prev);
+ end = bno + len;
+ obno = bno;
+
+ while (bno < end && n < *nmap) {
+ if (eof || got.br_startoff > bno) {
+ error = xfs_bmapi_reserve_delalloc(ip, bno, len, &got,
+ &prev, &lastx, eof);
+ if (error) {
+ if (n == 0) {
+ *nmap = 0;
+ return error;
+ }
+ break;
}
}
+ /* set up the extent map to return. */
+ xfs_bmapi_trim_map(mval, &got, &bno, len, obno, end, n, flags);
+ xfs_bmapi_update_map(&mval, &bno, &len, obno, end, &n, flags);
+
+ /* If we're done, stop now. */
+ if (bno >= end || n >= *nmap)
+ break;
+
+ /* Else go on to the next record. */
+ prev = got;
+ if (++lastx < ifp->if_bytes / sizeof(xfs_bmbt_rec_t))
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, lastx), &got);
+ else
+ eof = 1;
+ }
+
+ *nmap = n;
+ return 0;
+}
+
+
+STATIC int
+xfs_bmapi_allocate(
+ struct xfs_bmalloca *bma)
+{
+ struct xfs_mount *mp = bma->ip->i_mount;
+ int whichfork = (bma->flags & XFS_BMAPI_ATTRFORK) ?
+ XFS_ATTR_FORK : XFS_DATA_FORK;
+ struct xfs_ifork *ifp = XFS_IFORK_PTR(bma->ip, whichfork);
+ int tmp_logflags = 0;
+ int error;
+
+ ASSERT(bma->length > 0);
+
+ /*
+ * For the wasdelay case, we could also just allocate the stuff asked
+ * for in this bmap call but that wouldn't be as good.
+ */
+ if (bma->wasdel) {
+ bma->length = (xfs_extlen_t)bma->got.br_blockcount;
+ bma->offset = bma->got.br_startoff;
+ if (bma->idx != NULLEXTNUM && bma->idx) {
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx - 1),
+ &bma->prev);
+ }
+ } else {
+ bma->length = XFS_FILBLKS_MIN(bma->length, MAXEXTLEN);
+ if (!bma->eof)
+ bma->length = XFS_FILBLKS_MIN(bma->length,
+ bma->got.br_startoff - bma->offset);
+ }
+
+ /*
+ * Indicate if this is the first user data in the file, or just any
+ * user data.
+ */
+ if (!(bma->flags & XFS_BMAPI_METADATA)) {
+ bma->userdata = (bma->offset == 0) ?
+ XFS_ALLOC_INITIAL_USER_DATA : XFS_ALLOC_USERDATA;
+ }
+
+ bma->minlen = (bma->flags & XFS_BMAPI_CONTIG) ? bma->length : 1;
+
+ /*
+ * Only want to do the alignment at the eof if it is userdata and
+ * allocation length is larger than a stripe unit.
+ */
+ if (mp->m_dalign && bma->length >= mp->m_dalign &&
+ !(bma->flags & XFS_BMAPI_METADATA) && whichfork == XFS_DATA_FORK) {
+ error = xfs_bmap_isaeof(bma, whichfork);
+ if (error)
+ return error;
+ }
+
+ if (bma->flags & XFS_BMAPI_STACK_SWITCH)
+ bma->stack_switch = 1;
+
+ error = xfs_bmap_alloc(bma);
+ if (error)
+ return error;
+
+ if (bma->flist->xbf_low)
+ bma->minleft = 0;
+ if (bma->cur)
+ bma->cur->bc_private.b.firstblock = *bma->firstblock;
+ if (bma->blkno == NULLFSBLOCK)
+ return 0;
+ if ((ifp->if_flags & XFS_IFBROOT) && !bma->cur) {
+ bma->cur = xfs_bmbt_init_cursor(mp, bma->tp, bma->ip, whichfork);
+ bma->cur->bc_private.b.firstblock = *bma->firstblock;
+ bma->cur->bc_private.b.flist = bma->flist;
+ }
+ /*
+ * Bump the number of extents we've allocated
+ * in this call.
+ */
+ bma->nallocs++;
+
+ if (bma->cur)
+ bma->cur->bc_private.b.flags =
+ bma->wasdel ? XFS_BTCUR_BPRV_WASDEL : 0;
+
+ bma->got.br_startoff = bma->offset;
+ bma->got.br_startblock = bma->blkno;
+ bma->got.br_blockcount = bma->length;
+ bma->got.br_state = XFS_EXT_NORM;
+
+ /*
+ * A wasdelay extent has been initialized, so shouldn't be flagged
+ * as unwritten.
+ */
+ if (!bma->wasdel && (bma->flags & XFS_BMAPI_PREALLOC) &&
+ xfs_sb_version_hasextflgbit(&mp->m_sb))
+ bma->got.br_state = XFS_EXT_UNWRITTEN;
+
+ if (bma->wasdel)
+ error = xfs_bmap_add_extent_delay_real(bma);
+ else
+ error = xfs_bmap_add_extent_hole_real(bma, whichfork);
+
+ bma->logflags |= tmp_logflags;
+ if (error)
+ return error;
+
+ /*
+ * Update our extent pointer, given that xfs_bmap_add_extent_delay_real
+ * or xfs_bmap_add_extent_hole_real might have merged it into one of
+ * the neighbouring ones.
+ */
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx), &bma->got);
+
+ ASSERT(bma->got.br_startoff <= bma->offset);
+ ASSERT(bma->got.br_startoff + bma->got.br_blockcount >=
+ bma->offset + bma->length);
+ ASSERT(bma->got.br_state == XFS_EXT_NORM ||
+ bma->got.br_state == XFS_EXT_UNWRITTEN);
+ return 0;
+}
+
+
+STATIC int
+xfs_bmapi_convert_unwritten(
+ struct xfs_bmalloca *bma,
+ struct xfs_bmbt_irec *mval,
+ xfs_filblks_t len,
+ int flags)
+{
+ int whichfork = (flags & XFS_BMAPI_ATTRFORK) ?
+ XFS_ATTR_FORK : XFS_DATA_FORK;
+ struct xfs_ifork *ifp = XFS_IFORK_PTR(bma->ip, whichfork);
+ int tmp_logflags = 0;
+ int error;
+
+ /* check if we need to do unwritten->real conversion */
+ if (mval->br_state == XFS_EXT_UNWRITTEN &&
+ (flags & XFS_BMAPI_PREALLOC))
+ return 0;
+
+ /* check if we need to do real->unwritten conversion */
+ if (mval->br_state == XFS_EXT_NORM &&
+ (flags & (XFS_BMAPI_PREALLOC | XFS_BMAPI_CONVERT)) !=
+ (XFS_BMAPI_PREALLOC | XFS_BMAPI_CONVERT))
+ return 0;
+
+ /*
+ * Modify (by adding) the state flag, if writing.
+ */
+ ASSERT(mval->br_blockcount <= len);
+ if ((ifp->if_flags & XFS_IFBROOT) && !bma->cur) {
+ bma->cur = xfs_bmbt_init_cursor(bma->ip->i_mount, bma->tp,
+ bma->ip, whichfork);
+ bma->cur->bc_private.b.firstblock = *bma->firstblock;
+ bma->cur->bc_private.b.flist = bma->flist;
+ }
+ mval->br_state = (mval->br_state == XFS_EXT_UNWRITTEN)
+ ? XFS_EXT_NORM : XFS_EXT_UNWRITTEN;
+
+ error = xfs_bmap_add_extent_unwritten_real(bma->tp, bma->ip, &bma->idx,
+ &bma->cur, mval, bma->firstblock, bma->flist,
+ &tmp_logflags);
+ bma->logflags |= tmp_logflags;
+ if (error)
+ return error;
+
+ /*
+ * Update our extent pointer, given that
+ * xfs_bmap_add_extent_unwritten_real might have merged it into one
+ * of the neighbouring ones.
+ */
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx), &bma->got);
+
+ /*
+ * We may have combined previously unwritten space with written space,
+ * so generate another request.
+ */
+ if (mval->br_blockcount < len)
+ return EAGAIN;
+ return 0;
+}
+
+/*
+ * Map file blocks to filesystem blocks, and allocate blocks or convert the
+ * extent state if necessary. Details behaviour is controlled by the flags
+ * parameter. Only allocates blocks from a single allocation group, to avoid
+ * locking problems.
+ *
+ * The returned value in "firstblock" from the first call in a transaction
+ * must be remembered and presented to subsequent calls in "firstblock".
+ * An upper bound for the number of blocks to be allocated is supplied to
+ * the first call in "total"; if no allocation group has that many free
+ * blocks then the call will fail (return NULLFSBLOCK in "firstblock").
+ */
+int
+xfs_bmapi_write(
+ struct xfs_trans *tp, /* transaction pointer */
+ struct xfs_inode *ip, /* incore inode */
+ xfs_fileoff_t bno, /* starting file offs. mapped */
+ xfs_filblks_t len, /* length to map in file */
+ int flags, /* XFS_BMAPI_... */
+ xfs_fsblock_t *firstblock, /* first allocated block
+ controls a.g. for allocs */
+ xfs_extlen_t total, /* total blocks needed */
+ struct xfs_bmbt_irec *mval, /* output: map values */
+ int *nmap, /* i/o: mval size/count */
+ struct xfs_bmap_free *flist) /* i/o: list extents to free */
+{
+ struct xfs_mount *mp = ip->i_mount;
+ struct xfs_ifork *ifp;
+ struct xfs_bmalloca bma = { 0 }; /* args for xfs_bmap_alloc */
+ xfs_fileoff_t end; /* end of mapped file region */
+ int eof; /* after the end of extents */
+ int error; /* error return */
+ int n; /* current extent index */
+ xfs_fileoff_t obno; /* old block number (offset) */
+ int whichfork; /* data or attr fork */
+ char inhole; /* current location is hole in file */
+ char wasdelay; /* old extent was delayed */
+
+#ifdef DEBUG
+ xfs_fileoff_t orig_bno; /* original block number value */
+ int orig_flags; /* original flags arg value */
+ xfs_filblks_t orig_len; /* original value of len arg */
+ struct xfs_bmbt_irec *orig_mval; /* original value of mval */
+ int orig_nmap; /* original value of *nmap */
+
+ orig_bno = bno;
+ orig_len = len;
+ orig_flags = flags;
+ orig_mval = mval;
+ orig_nmap = *nmap;
+#endif
+
+ ASSERT(*nmap >= 1);
+ ASSERT(*nmap <= XFS_BMAP_MAX_NMAP);
+ ASSERT(!(flags & XFS_BMAPI_IGSTATE));
+ ASSERT(tp != NULL);
+ ASSERT(len > 0);
+
+ whichfork = (flags & XFS_BMAPI_ATTRFORK) ?
+ XFS_ATTR_FORK : XFS_DATA_FORK;
+
+ if (unlikely(XFS_TEST_ERROR(
+ (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
+ XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE &&
+ XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_LOCAL),
+ mp, XFS_ERRTAG_BMAPIFORMAT, XFS_RANDOM_BMAPIFORMAT))) {
+ XFS_ERROR_REPORT("xfs_bmapi_write", XFS_ERRLEVEL_LOW, mp);
+ return XFS_ERROR(EFSCORRUPTED);
+ }
+
+ if (XFS_FORCED_SHUTDOWN(mp))
+ return XFS_ERROR(EIO);
+
+ ifp = XFS_IFORK_PTR(ip, whichfork);
+
+ XFS_STATS_INC(xs_blk_mapw);
+
+ if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL) {
+ error = xfs_bmap_local_to_extents(tp, ip, firstblock, total,
+ &bma.logflags, whichfork);
+ if (error)
+ goto error0;
+ }
+
+ if (*firstblock == NULLFSBLOCK) {
+ if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE)
+ bma.minleft = be16_to_cpu(ifp->if_broot->bb_level) + 1;
+ else
+ bma.minleft = 1;
+ } else {
+ bma.minleft = 0;
+ }
+
+ if (!(ifp->if_flags & XFS_IFEXTENTS)) {
+ error = xfs_iread_extents(tp, ip, whichfork);
+ if (error)
+ goto error0;
+ }
+
+ xfs_bmap_search_extents(ip, bno, whichfork, &eof, &bma.idx, &bma.got,
+ &bma.prev);
+ n = 0;
+ end = bno + len;
+ obno = bno;
+
+ bma.tp = tp;
+ bma.ip = ip;
+ bma.total = total;
+ bma.userdata = 0;
+ bma.flist = flist;
+ bma.firstblock = firstblock;
+
+ while (bno < end && n < *nmap) {
+ inhole = eof || bma.got.br_startoff > bno;
+ wasdelay = !inhole && isnullstartblock(bma.got.br_startblock);
+
/*
- * Check if writing previously allocated but
- * unwritten extents.
+ * First, deal with the hole before the allocated space
+ * that we found, if any.
*/
- if (wr &&
- ((mval->br_state == XFS_EXT_UNWRITTEN &&
- ((flags & (XFS_BMAPI_PREALLOC|XFS_BMAPI_DELAY)) == 0)) ||
- (mval->br_state == XFS_EXT_NORM &&
- ((flags & (XFS_BMAPI_PREALLOC|XFS_BMAPI_CONVERT)) ==
- (XFS_BMAPI_PREALLOC|XFS_BMAPI_CONVERT))))) {
+ if (inhole || wasdelay) {
+ bma.eof = eof;
+ bma.conv = !!(flags & XFS_BMAPI_CONVERT);
+ bma.wasdel = wasdelay;
+ bma.offset = bno;
+ bma.flags = flags;
+
/*
- * Modify (by adding) the state flag, if writing.
+ * There's a 32/64 bit type mismatch between the
+ * allocation length request (which can be 64 bits in
+ * length) and the bma length request, which is
+ * xfs_extlen_t and therefore 32 bits. Hence we have to
+ * check for 32-bit overflows and handle them here.
*/
- ASSERT(mval->br_blockcount <= len);
- if ((ifp->if_flags & XFS_IFBROOT) && !cur) {
- cur = xfs_bmbt_init_cursor(mp,
- tp, ip, whichfork);
- cur->bc_private.b.firstblock =
- *firstblock;
- cur->bc_private.b.flist = flist;
- }
- mval->br_state = (mval->br_state == XFS_EXT_UNWRITTEN)
- ? XFS_EXT_NORM
- : XFS_EXT_UNWRITTEN;
- error = xfs_bmap_add_extent(ip, lastx, &cur, mval,
- firstblock, flist, &tmp_logflags,
- whichfork, (flags & XFS_BMAPI_RSVBLOCKS));
- logflags |= tmp_logflags;
+ if (len > (xfs_filblks_t)MAXEXTLEN)
+ bma.length = MAXEXTLEN;
+ else
+ bma.length = len;
+
+ ASSERT(len > 0);
+ ASSERT(bma.length > 0);
+ error = xfs_bmapi_allocate(&bma);
if (error)
goto error0;
- lastx = ifp->if_lastex;
- ep = xfs_iext_get_ext(ifp, lastx);
- nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
- xfs_bmbt_get_all(ep, &got);
- /*
- * We may have combined previously unwritten
- * space with written space, so generate
- * another request.
- */
- if (mval->br_blockcount < len)
- continue;
+ if (bma.blkno == NULLFSBLOCK)
+ break;
}
- ASSERT((flags & XFS_BMAPI_ENTIRE) ||
- ((mval->br_startoff + mval->br_blockcount) <= end));
- ASSERT((flags & XFS_BMAPI_ENTIRE) ||
- (mval->br_blockcount <= len) ||
- (mval->br_startoff < obno));
- bno = mval->br_startoff + mval->br_blockcount;
- len = end - bno;
- if (n > 0 && mval->br_startoff == mval[-1].br_startoff) {
- ASSERT(mval->br_startblock == mval[-1].br_startblock);
- ASSERT(mval->br_blockcount > mval[-1].br_blockcount);
- ASSERT(mval->br_state == mval[-1].br_state);
- mval[-1].br_blockcount = mval->br_blockcount;
- mval[-1].br_state = mval->br_state;
- } else if (n > 0 && mval->br_startblock != DELAYSTARTBLOCK &&
- mval[-1].br_startblock != DELAYSTARTBLOCK &&
- mval[-1].br_startblock != HOLESTARTBLOCK &&
- mval->br_startblock ==
- mval[-1].br_startblock + mval[-1].br_blockcount &&
- ((flags & XFS_BMAPI_IGSTATE) ||
- mval[-1].br_state == mval->br_state)) {
- ASSERT(mval->br_startoff ==
- mval[-1].br_startoff + mval[-1].br_blockcount);
- mval[-1].br_blockcount += mval->br_blockcount;
- } else if (n > 0 &&
- mval->br_startblock == DELAYSTARTBLOCK &&
- mval[-1].br_startblock == DELAYSTARTBLOCK &&
- mval->br_startoff ==
- mval[-1].br_startoff + mval[-1].br_blockcount) {
- mval[-1].br_blockcount += mval->br_blockcount;
- mval[-1].br_state = mval->br_state;
- } else if (!((n == 0) &&
- ((mval->br_startoff + mval->br_blockcount) <=
- obno))) {
- mval++;
- n++;
- }
+ /* Deal with the allocated space we found. */
+ xfs_bmapi_trim_map(mval, &bma.got, &bno, len, obno,
+ end, n, flags);
+
+ /* Execute unwritten extent conversion if necessary */
+ error = xfs_bmapi_convert_unwritten(&bma, mval, len, flags);
+ if (error == EAGAIN)
+ continue;
+ if (error)
+ goto error0;
+
+ /* update the extent map to return */
+ xfs_bmapi_update_map(&mval, &bno, &len, obno, end, &n, flags);
+
/*
* If we're done, stop now. Stop when we've allocated
* XFS_BMAP_MAX_NMAP extents no matter what. Otherwise
* the transaction may get too big.
*/
- if (bno >= end || n >= *nmap || nallocs >= *nmap)
+ if (bno >= end || n >= *nmap || bma.nallocs >= *nmap)
break;
- /*
- * Else go on to the next record.
- */
- ep = xfs_iext_get_ext(ifp, ++lastx);
- prev = got;
- if (lastx >= nextents)
+
+ /* Else go on to the next record. */
+ bma.prev = bma.got;
+ if (++bma.idx < ifp->if_bytes / sizeof(xfs_bmbt_rec_t)) {
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma.idx),
+ &bma.got);
+ } else
eof = 1;
- else
- xfs_bmbt_get_all(ep, &got);
}
- ifp->if_lastex = lastx;
*nmap = n;
+
/*
* Transform from btree to extents, give it cur.
*/
- if (tp && XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE &&
- XFS_IFORK_NEXTENTS(ip, whichfork) <= ifp->if_ext_max) {
- ASSERT(wr && cur);
- error = xfs_bmap_btree_to_extents(tp, ip, cur,
+ if (xfs_bmap_wants_extents(ip, whichfork)) {
+ int tmp_logflags = 0;
+
+ ASSERT(bma.cur);
+ error = xfs_bmap_btree_to_extents(tp, ip, bma.cur,
&tmp_logflags, whichfork);
- logflags |= tmp_logflags;
+ bma.logflags |= tmp_logflags;
if (error)
goto error0;
}
- ASSERT(ifp->if_ext_max ==
- XFS_IFORK_SIZE(ip, whichfork) / (uint)sizeof(xfs_bmbt_rec_t));
+
ASSERT(XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE ||
- XFS_IFORK_NEXTENTS(ip, whichfork) > ifp->if_ext_max);
+ XFS_IFORK_NEXTENTS(ip, whichfork) >
+ XFS_IFORK_MAXEXT(ip, whichfork));
error = 0;
error0:
/*
* Log everything. Do this after conversion, there's no point in
* logging the extent records if we've converted to btree format.
*/
- if ((logflags & xfs_ilog_fext(whichfork)) &&
+ if ((bma.logflags & xfs_ilog_fext(whichfork)) &&
XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS)
- logflags &= ~xfs_ilog_fext(whichfork);
- else if ((logflags & xfs_ilog_fbroot(whichfork)) &&
+ bma.logflags &= ~xfs_ilog_fext(whichfork);
+ else if ((bma.logflags & xfs_ilog_fbroot(whichfork)) &&
XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)
- logflags &= ~xfs_ilog_fbroot(whichfork);
+ bma.logflags &= ~xfs_ilog_fbroot(whichfork);
/*
* Log whatever the flags say, even if error. Otherwise we might miss
* detecting a case where the data is changed, there's an error,
* and it's not logged so we don't shutdown when we should.
*/
- if (logflags) {
- ASSERT(tp && wr);
- xfs_trans_log_inode(tp, ip, logflags);
- }
- if (cur) {
+ if (bma.logflags)
+ xfs_trans_log_inode(tp, ip, bma.logflags);
+
+ if (bma.cur) {
if (!error) {
ASSERT(*firstblock == NULLFSBLOCK ||
XFS_FSB_TO_AGNO(mp, *firstblock) ==
XFS_FSB_TO_AGNO(mp,
- cur->bc_private.b.firstblock) ||
+ bma.cur->bc_private.b.firstblock) ||
(flist->xbf_low &&
XFS_FSB_TO_AGNO(mp, *firstblock) <
XFS_FSB_TO_AGNO(mp,
- cur->bc_private.b.firstblock)));
- *firstblock = cur->bc_private.b.firstblock;
+ bma.cur->bc_private.b.firstblock)));
+ *firstblock = bma.cur->bc_private.b.firstblock;
}
- xfs_btree_del_cursor(cur,
+ xfs_btree_del_cursor(bma.cur,
error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR);
}
if (!error)
@@ -4649,59 +4793,6 @@ error0:
}
/*
- * Map file blocks to filesystem blocks, simple version.
- * One block (extent) only, read-only.
- * For flags, only the XFS_BMAPI_ATTRFORK flag is examined.
- * For the other flag values, the effect is as if XFS_BMAPI_METADATA
- * was set and all the others were clear.
- */
-int /* error */
-xfs_bmapi_single(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode */
- int whichfork, /* data or attr fork */
- xfs_fsblock_t *fsb, /* output: mapped block */
- xfs_fileoff_t bno) /* starting file offs. mapped */
-{
- int eof; /* we've hit the end of extents */
- int error; /* error return */
- xfs_bmbt_irec_t got; /* current file extent record */
- xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_extnum_t lastx; /* last useful extent number */
- xfs_bmbt_irec_t prev; /* previous file extent record */
-
- ifp = XFS_IFORK_PTR(ip, whichfork);
- if (unlikely(
- XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE &&
- XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS)) {
- XFS_ERROR_REPORT("xfs_bmapi_single", XFS_ERRLEVEL_LOW,
- ip->i_mount);
- return XFS_ERROR(EFSCORRUPTED);
- }
- if (XFS_FORCED_SHUTDOWN(ip->i_mount))
- return XFS_ERROR(EIO);
- XFS_STATS_INC(xs_blk_mapr);
- if (!(ifp->if_flags & XFS_IFEXTENTS) &&
- (error = xfs_iread_extents(tp, ip, whichfork)))
- return error;
- (void)xfs_bmap_search_extents(ip, bno, whichfork, &eof, &lastx, &got,
- &prev);
- /*
- * Reading past eof, act as though there's a hole
- * up to end.
- */
- if (eof || got.br_startoff > bno) {
- *fsb = NULLFSBLOCK;
- return 0;
- }
- ASSERT(!isnullstartblock(got.br_startblock));
- ASSERT(bno < got.br_startoff + got.br_blockcount);
- *fsb = got.br_startblock + (bno - got.br_startoff);
- ifp->if_lastex = lastx;
- return 0;
-}
-
-/*
* Unmap (remove) blocks from a file.
* If nexts is nonzero then the number of extents to remove is limited to
* that value. If not all extents in the block range can be removed then
@@ -4739,7 +4830,6 @@ xfs_bunmapi(
int tmp_logflags; /* partial logging flags */
int wasdel; /* was a delayed alloc extent */
int whichfork; /* data or attribute fork */
- int rsvd; /* OK to allocate reserved blocks */
xfs_fsblock_t sum;
trace_xfs_bunmap(ip, bno, len, flags, _RET_IP_);
@@ -4757,11 +4847,10 @@ xfs_bunmapi(
mp = ip->i_mount;
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
- rsvd = (flags & XFS_BMAPI_RSVBLOCKS) != 0;
+
ASSERT(len > 0);
ASSERT(nexts >= 0);
- ASSERT(ifp->if_ext_max ==
- XFS_IFORK_SIZE(ip, whichfork) / (uint)sizeof(xfs_bmbt_rec_t));
+
if (!(ifp->if_flags & XFS_IFEXTENTS) &&
(error = xfs_iread_extents(tp, ip, whichfork)))
return error;
@@ -4795,6 +4884,15 @@ xfs_bunmapi(
cur->bc_private.b.flags = 0;
} else
cur = NULL;
+
+ if (isrt) {
+ /*
+ * Synchronize by locking the bitmap inode.
+ */
+ xfs_ilock(mp->m_rbmip, XFS_ILOCK_EXCL);
+ xfs_trans_ijoin(tp, mp->m_rbmip, XFS_ILOCK_EXCL);
+ }
+
extno = 0;
while (bno != (xfs_fileoff_t)-1 && bno >= start && lastx >= 0 &&
(nexts == 0 || extno < nexts)) {
@@ -4873,9 +4971,9 @@ xfs_bunmapi(
del.br_blockcount = mod;
}
del.br_state = XFS_EXT_UNWRITTEN;
- error = xfs_bmap_add_extent(ip, lastx, &cur, &del,
- firstblock, flist, &logflags,
- XFS_DATA_FORK, 0);
+ error = xfs_bmap_add_extent_unwritten_real(tp, ip,
+ &lastx, &cur, &del, firstblock, flist,
+ &logflags);
if (error)
goto error0;
goto nodelete;
@@ -4901,9 +4999,12 @@ xfs_bunmapi(
*/
ASSERT(bno >= del.br_blockcount);
bno -= del.br_blockcount;
- if (bno < got.br_startoff) {
- if (--lastx >= 0)
- xfs_bmbt_get_all(--ep, &got);
+ if (got.br_startoff > bno) {
+ if (--lastx >= 0) {
+ ep = xfs_iext_get_ext(ifp,
+ lastx);
+ xfs_bmbt_get_all(ep, &got);
+ }
}
continue;
} else if (del.br_state == XFS_EXT_UNWRITTEN) {
@@ -4927,18 +5028,19 @@ xfs_bunmapi(
prev.br_startoff = start;
}
prev.br_state = XFS_EXT_UNWRITTEN;
- error = xfs_bmap_add_extent(ip, lastx - 1, &cur,
- &prev, firstblock, flist, &logflags,
- XFS_DATA_FORK, 0);
+ lastx--;
+ error = xfs_bmap_add_extent_unwritten_real(tp,
+ ip, &lastx, &cur, &prev,
+ firstblock, flist, &logflags);
if (error)
goto error0;
goto nodelete;
} else {
ASSERT(del.br_state == XFS_EXT_NORM);
del.br_state = XFS_EXT_UNWRITTEN;
- error = xfs_bmap_add_extent(ip, lastx, &cur,
- &del, firstblock, flist, &logflags,
- XFS_DATA_FORK, 0);
+ error = xfs_bmap_add_extent_unwritten_real(tp,
+ ip, &lastx, &cur, &del,
+ firstblock, flist, &logflags);
if (error)
goto error0;
goto nodelete;
@@ -4953,13 +5055,13 @@ xfs_bunmapi(
rtexts = XFS_FSB_TO_B(mp, del.br_blockcount);
do_div(rtexts, mp->m_sb.sb_rextsize);
xfs_mod_incore_sb(mp, XFS_SBS_FREXTENTS,
- (int64_t)rtexts, rsvd);
+ (int64_t)rtexts, 0);
(void)xfs_trans_reserve_quota_nblks(NULL,
ip, -((long)del.br_blockcount), 0,
XFS_QMOPT_RES_RTBLKS);
} else {
xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS,
- (int64_t)del.br_blockcount, rsvd);
+ (int64_t)del.br_blockcount, 0);
(void)xfs_trans_reserve_quota_nblks(NULL,
ip, -((long)del.br_blockcount), 0,
XFS_QMOPT_RES_REGBLKS);
@@ -4983,46 +5085,43 @@ xfs_bunmapi(
*/
if (!wasdel && xfs_trans_get_block_res(tp) == 0 &&
XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS &&
- XFS_IFORK_NEXTENTS(ip, whichfork) >= ifp->if_ext_max &&
+ XFS_IFORK_NEXTENTS(ip, whichfork) >= /* Note the >= */
+ XFS_IFORK_MAXEXT(ip, whichfork) &&
del.br_startoff > got.br_startoff &&
del.br_startoff + del.br_blockcount <
got.br_startoff + got.br_blockcount) {
error = XFS_ERROR(ENOSPC);
goto error0;
}
- error = xfs_bmap_del_extent(ip, tp, lastx, flist, cur, &del,
- &tmp_logflags, whichfork, rsvd);
+ error = xfs_bmap_del_extent(ip, tp, &lastx, flist, cur, &del,
+ &tmp_logflags, whichfork);
logflags |= tmp_logflags;
if (error)
goto error0;
bno = del.br_startoff - 1;
nodelete:
- lastx = ifp->if_lastex;
/*
* If not done go on to the next (previous) record.
- * Reset ep in case the extents array was re-alloced.
*/
- ep = xfs_iext_get_ext(ifp, lastx);
if (bno != (xfs_fileoff_t)-1 && bno >= start) {
- if (lastx >= XFS_IFORK_NEXTENTS(ip, whichfork) ||
- xfs_bmbt_get_startoff(ep) > bno) {
- if (--lastx >= 0)
- ep = xfs_iext_get_ext(ifp, lastx);
- }
- if (lastx >= 0)
+ if (lastx >= 0) {
+ ep = xfs_iext_get_ext(ifp, lastx);
+ if (xfs_bmbt_get_startoff(ep) > bno) {
+ if (--lastx >= 0)
+ ep = xfs_iext_get_ext(ifp,
+ lastx);
+ }
xfs_bmbt_get_all(ep, &got);
+ }
extno++;
}
}
- ifp->if_lastex = lastx;
*done = bno == (xfs_fileoff_t)-1 || bno < start || lastx < 0;
- ASSERT(ifp->if_ext_max ==
- XFS_IFORK_SIZE(ip, whichfork) / (uint)sizeof(xfs_bmbt_rec_t));
+
/*
* Convert to a btree if necessary.
*/
- if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS &&
- XFS_IFORK_NEXTENTS(ip, whichfork) > ifp->if_ext_max) {
+ if (xfs_bmap_needs_btree(ip, whichfork)) {
ASSERT(cur == NULL);
error = xfs_bmap_extents_to_btree(tp, ip, firstblock, flist,
&cur, 0, &tmp_logflags, whichfork);
@@ -5033,8 +5132,7 @@ nodelete:
/*
* transform from btree to extents, give it cur
*/
- else if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE &&
- XFS_IFORK_NEXTENTS(ip, whichfork) <= ifp->if_ext_max) {
+ else if (xfs_bmap_wants_extents(ip, whichfork)) {
ASSERT(cur != NULL);
error = xfs_bmap_btree_to_extents(tp, ip, cur, &tmp_logflags,
whichfork);
@@ -5045,8 +5143,6 @@ nodelete:
/*
* transform from extents to local?
*/
- ASSERT(ifp->if_ext_max ==
- XFS_IFORK_SIZE(ip, whichfork) / (uint)sizeof(xfs_bmbt_rec_t));
error = 0;
error0:
/*
@@ -5076,88 +5172,252 @@ error0:
return error;
}
+#ifdef DEBUG
+STATIC struct xfs_buf *
+xfs_bmap_get_bp(
+ struct xfs_btree_cur *cur,
+ xfs_fsblock_t bno)
+{
+ struct xfs_log_item_desc *lidp;
+ int i;
+
+ if (!cur)
+ return NULL;
+
+ for (i = 0; i < XFS_BTREE_MAXLEVELS; i++) {
+ if (!cur->bc_bufs[i])
+ break;
+ if (XFS_BUF_ADDR(cur->bc_bufs[i]) == bno)
+ return cur->bc_bufs[i];
+ }
+
+ /* Chase down all the log items to see if the bp is there */
+ list_for_each_entry(lidp, &cur->bc_tp->t_items, lid_trans) {
+ struct xfs_buf_log_item *bip;
+ bip = (struct xfs_buf_log_item *)lidp->lid_item;
+ if (bip->bli_item.li_type == XFS_LI_BUF &&
+ XFS_BUF_ADDR(bip->bli_buf) == bno)
+ return bip->bli_buf;
+ }
+
+ return NULL;
+}
+
+STATIC void
+xfs_check_block(
+ struct xfs_btree_block *block,
+ xfs_mount_t *mp,
+ int root,
+ short sz)
+{
+ int i, j, dmxr;
+ __be64 *pp, *thispa; /* pointer to block address */
+ xfs_bmbt_key_t *prevp, *keyp;
+
+ ASSERT(be16_to_cpu(block->bb_level) > 0);
+
+ prevp = NULL;
+ for( i = 1; i <= xfs_btree_get_numrecs(block); i++) {
+ dmxr = mp->m_bmap_dmxr[0];
+ keyp = XFS_BMBT_KEY_ADDR(mp, block, i);
+
+ if (prevp) {
+ ASSERT(be64_to_cpu(prevp->br_startoff) <
+ be64_to_cpu(keyp->br_startoff));
+ }
+ prevp = keyp;
+
+ /*
+ * Compare the block numbers to see if there are dups.
+ */
+ if (root)
+ pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, i, sz);
+ else
+ pp = XFS_BMBT_PTR_ADDR(mp, block, i, dmxr);
+
+ for (j = i+1; j <= be16_to_cpu(block->bb_numrecs); j++) {
+ if (root)
+ thispa = XFS_BMAP_BROOT_PTR_ADDR(mp, block, j, sz);
+ else
+ thispa = XFS_BMBT_PTR_ADDR(mp, block, j, dmxr);
+ if (*thispa == *pp) {
+ xfs_warn(mp, "%s: thispa(%d) == pp(%d) %Ld",
+ __func__, j, i,
+ (unsigned long long)be64_to_cpu(*thispa));
+ panic("%s: ptrs are equal in node\n",
+ __func__);
+ }
+ }
+ }
+}
+
/*
- * Check the last inode extent to determine whether this allocation will result
- * in blocks being allocated at the end of the file. When we allocate new data
- * blocks at the end of the file which do not start at the previous data block,
- * we will try to align the new blocks at stripe unit boundaries.
+ * Check that the extents for the inode ip are in the right order in all
+ * btree leaves.
*/
-STATIC int /* error */
-xfs_bmap_isaeof(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_fileoff_t off, /* file offset in fsblocks */
- int whichfork, /* data or attribute fork */
- char *aeof) /* return value */
+
+STATIC void
+xfs_bmap_check_leaf_extents(
+ xfs_btree_cur_t *cur, /* btree cursor or null */
+ xfs_inode_t *ip, /* incore inode pointer */
+ int whichfork) /* data or attr fork */
{
- int error; /* error return value */
- xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_bmbt_rec_host_t *lastrec; /* extent record pointer */
- xfs_extnum_t nextents; /* number of file extents */
- xfs_bmbt_irec_t s; /* expanded extent record */
+ struct xfs_btree_block *block; /* current btree block */
+ xfs_fsblock_t bno; /* block # of "block" */
+ xfs_buf_t *bp; /* buffer for "block" */
+ int error; /* error return value */
+ xfs_extnum_t i=0, j; /* index into the extents list */
+ xfs_ifork_t *ifp; /* fork structure */
+ int level; /* btree level, for checking */
+ xfs_mount_t *mp; /* file system mount structure */
+ __be64 *pp; /* pointer to block address */
+ xfs_bmbt_rec_t *ep; /* pointer to current extent */
+ xfs_bmbt_rec_t last = {0, 0}; /* last extent in prev block */
+ xfs_bmbt_rec_t *nextp; /* pointer to next extent */
+ int bp_release = 0;
- ASSERT(whichfork == XFS_DATA_FORK);
- ifp = XFS_IFORK_PTR(ip, whichfork);
- if (!(ifp->if_flags & XFS_IFEXTENTS) &&
- (error = xfs_iread_extents(NULL, ip, whichfork)))
- return error;
- nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
- if (nextents == 0) {
- *aeof = 1;
- return 0;
+ if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE) {
+ return;
}
+
+ bno = NULLFSBLOCK;
+ mp = ip->i_mount;
+ ifp = XFS_IFORK_PTR(ip, whichfork);
+ block = ifp->if_broot;
/*
- * Go to the last extent
+ * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out.
*/
- lastrec = xfs_iext_get_ext(ifp, nextents - 1);
- xfs_bmbt_get_all(lastrec, &s);
+ level = be16_to_cpu(block->bb_level);
+ ASSERT(level > 0);
+ xfs_check_block(block, mp, 1, ifp->if_broot_bytes);
+ pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
+ bno = be64_to_cpu(*pp);
+
+ ASSERT(bno != NULLDFSBNO);
+ ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount);
+ ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks);
+
/*
- * Check we are allocating in the last extent (for delayed allocations)
- * or past the last extent for non-delayed allocations.
+ * Go down the tree until leaf level is reached, following the first
+ * pointer (leftmost) at each level.
*/
- *aeof = (off >= s.br_startoff &&
- off < s.br_startoff + s.br_blockcount &&
- isnullstartblock(s.br_startblock)) ||
- off >= s.br_startoff + s.br_blockcount;
- return 0;
-}
+ while (level-- > 0) {
+ /* See if buf is in cur first */
+ bp_release = 0;
+ bp = xfs_bmap_get_bp(cur, XFS_FSB_TO_DADDR(mp, bno));
+ if (!bp) {
+ bp_release = 1;
+ error = xfs_btree_read_bufl(mp, NULL, bno, 0, &bp,
+ XFS_BMAP_BTREE_REF,
+ &xfs_bmbt_buf_ops);
+ if (error)
+ goto error_norelse;
+ }
+ block = XFS_BUF_TO_BLOCK(bp);
+ XFS_WANT_CORRUPTED_GOTO(
+ xfs_bmap_sanity_check(mp, bp, level),
+ error0);
+ if (level == 0)
+ break;
-/*
- * Check if the endoff is outside the last extent. If so the caller will grow
- * the allocation to a stripe unit boundary.
- */
-int /* error */
-xfs_bmap_eof(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_fileoff_t endoff, /* file offset in fsblocks */
- int whichfork, /* data or attribute fork */
- int *eof) /* result value */
-{
- xfs_fsblock_t blockcount; /* extent block count */
- int error; /* error return value */
- xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_bmbt_rec_host_t *lastrec; /* extent record pointer */
- xfs_extnum_t nextents; /* number of file extents */
- xfs_fileoff_t startoff; /* extent starting file offset */
+ /*
+ * Check this block for basic sanity (increasing keys and
+ * no duplicate blocks).
+ */
- ASSERT(whichfork == XFS_DATA_FORK);
- ifp = XFS_IFORK_PTR(ip, whichfork);
- if (!(ifp->if_flags & XFS_IFEXTENTS) &&
- (error = xfs_iread_extents(NULL, ip, whichfork)))
- return error;
- nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
- if (nextents == 0) {
- *eof = 1;
- return 0;
+ xfs_check_block(block, mp, 0, 0);
+ pp = XFS_BMBT_PTR_ADDR(mp, block, 1, mp->m_bmap_dmxr[1]);
+ bno = be64_to_cpu(*pp);
+ XFS_WANT_CORRUPTED_GOTO(XFS_FSB_SANITY_CHECK(mp, bno), error0);
+ if (bp_release) {
+ bp_release = 0;
+ xfs_trans_brelse(NULL, bp);
+ }
}
+
+ /*
+ * Here with bp and block set to the leftmost leaf node in the tree.
+ */
+ i = 0;
+
/*
- * Go to the last extent
+ * Loop over all leaf nodes checking that all extents are in the right order.
*/
- lastrec = xfs_iext_get_ext(ifp, nextents - 1);
- startoff = xfs_bmbt_get_startoff(lastrec);
- blockcount = xfs_bmbt_get_blockcount(lastrec);
- *eof = endoff >= startoff + blockcount;
- return 0;
+ for (;;) {
+ xfs_fsblock_t nextbno;
+ xfs_extnum_t num_recs;
+
+
+ num_recs = xfs_btree_get_numrecs(block);
+
+ /*
+ * Read-ahead the next leaf block, if any.
+ */
+
+ nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
+
+ /*
+ * Check all the extents to make sure they are OK.
+ * If we had a previous block, the last entry should
+ * conform with the first entry in this one.
+ */
+
+ ep = XFS_BMBT_REC_ADDR(mp, block, 1);
+ if (i) {
+ ASSERT(xfs_bmbt_disk_get_startoff(&last) +
+ xfs_bmbt_disk_get_blockcount(&last) <=
+ xfs_bmbt_disk_get_startoff(ep));
+ }
+ for (j = 1; j < num_recs; j++) {
+ nextp = XFS_BMBT_REC_ADDR(mp, block, j + 1);
+ ASSERT(xfs_bmbt_disk_get_startoff(ep) +
+ xfs_bmbt_disk_get_blockcount(ep) <=
+ xfs_bmbt_disk_get_startoff(nextp));
+ ep = nextp;
+ }
+
+ last = *ep;
+ i += num_recs;
+ if (bp_release) {
+ bp_release = 0;
+ xfs_trans_brelse(NULL, bp);
+ }
+ bno = nextbno;
+ /*
+ * If we've reached the end, stop.
+ */
+ if (bno == NULLFSBLOCK)
+ break;
+
+ bp_release = 0;
+ bp = xfs_bmap_get_bp(cur, XFS_FSB_TO_DADDR(mp, bno));
+ if (!bp) {
+ bp_release = 1;
+ error = xfs_btree_read_bufl(mp, NULL, bno, 0, &bp,
+ XFS_BMAP_BTREE_REF,
+ &xfs_bmbt_buf_ops);
+ if (error)
+ goto error_norelse;
+ }
+ block = XFS_BUF_TO_BLOCK(bp);
+ }
+ if (bp_release) {
+ bp_release = 0;
+ xfs_trans_brelse(NULL, bp);
+ }
+ return;
+
+error0:
+ xfs_warn(mp, "%s: at error0", __func__);
+ if (bp_release)
+ xfs_trans_brelse(NULL, bp);
+error_norelse:
+ xfs_warn(mp, "%s: BAD after btree leaves for %d extents",
+ __func__, i);
+ panic("%s: CORRUPTED BTREE OR SOMETHING", __func__);
+ return;
}
+#endif
/*
* Count fsblocks of the given fork.
@@ -5229,7 +5489,9 @@ xfs_bmap_count_tree(
struct xfs_btree_block *block, *nextblock;
int numrecs;
- if ((error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp, XFS_BMAP_BTREE_REF)))
+ error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp, XFS_BMAP_BTREE_REF,
+ &xfs_bmbt_buf_ops);
+ if (error)
return error;
*count += 1;
block = XFS_BUF_TO_BLOCK(bp);
@@ -5238,8 +5500,10 @@ xfs_bmap_count_tree(
/* Not at node above leaves, count this level of nodes */
nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
while (nextbno != NULLFSBLOCK) {
- if ((error = xfs_btree_read_bufl(mp, tp, nextbno,
- 0, &nbp, XFS_BMAP_BTREE_REF)))
+ error = xfs_btree_read_bufl(mp, tp, nextbno, 0, &nbp,
+ XFS_BMAP_BTREE_REF,
+ &xfs_bmbt_buf_ops);
+ if (error)
return error;
*count += 1;
nextblock = XFS_BUF_TO_BLOCK(nbp);
@@ -5268,8 +5532,10 @@ xfs_bmap_count_tree(
if (nextbno == NULLFSBLOCK)
break;
bno = nextbno;
- if ((error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp,
- XFS_BMAP_BTREE_REF)))
+ error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp,
+ XFS_BMAP_BTREE_REF,
+ &xfs_bmbt_buf_ops);
+ if (error)
return error;
*count += 1;
block = XFS_BUF_TO_BLOCK(bp);
@@ -5315,3 +5581,16 @@ xfs_bmap_disk_count_leaves(
*count += xfs_bmbt_disk_get_blockcount(frp);
}
}
+
+/*
+ * Convert the given file system block to a disk block. We have to treat it
+ * differently based on whether the file is a real time file or not, because the
+ * bmap code does.
+ */
+xfs_daddr_t
+xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb)
+{
+ return (XFS_IS_REALTIME_INODE(ip) ? \
+ (xfs_daddr_t)XFS_FSB_TO_BB((ip)->i_mount, (fsb)) : \
+ XFS_FSB_TO_DADDR((ip)->i_mount, (fsb)));
+}
diff --git a/libxfs/xfs_bmap_btree.c b/libxfs/xfs_bmap_btree.c
index ff51fdd..836f52f 100644
--- a/libxfs/xfs_bmap_btree.c
+++ b/libxfs/xfs_bmap_btree.c
@@ -403,10 +403,10 @@ xfs_bmbt_to_bmdr(
xfs_bmbt_key_t *tkp;
__be64 *tpp;
- ASSERT(be32_to_cpu(rblock->bb_magic) == XFS_BMAP_MAGIC);
- ASSERT(be64_to_cpu(rblock->bb_u.l.bb_leftsib) == NULLDFSBNO);
- ASSERT(be64_to_cpu(rblock->bb_u.l.bb_rightsib) == NULLDFSBNO);
- ASSERT(be16_to_cpu(rblock->bb_level) > 0);
+ ASSERT(rblock->bb_magic == cpu_to_be32(XFS_BMAP_MAGIC));
+ ASSERT(rblock->bb_u.l.bb_leftsib == cpu_to_be64(NULLDFSBNO));
+ ASSERT(rblock->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO));
+ ASSERT(rblock->bb_level != 0);
dblock->bb_level = rblock->bb_level;
dblock->bb_numrecs = rblock->bb_numrecs;
dmxr = xfs_bmdr_maxrecs(mp, dblocklen, 0);
@@ -687,6 +687,67 @@ xfs_bmbt_key_diff(
cur->bc_rec.b.br_startoff;
}
+static void
+xfs_bmbt_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
+ unsigned int level;
+ int lblock_ok; /* block passes checks */
+
+ /* magic number and level verification.
+ *
+ * We don't know waht fork we belong to, so just verify that the level
+ * is less than the maximum of the two. Later checks will be more
+ * precise.
+ */
+ level = be16_to_cpu(block->bb_level);
+ lblock_ok = block->bb_magic == cpu_to_be32(XFS_BMAP_MAGIC) &&
+ level < MAX(mp->m_bm_maxlevels[0], mp->m_bm_maxlevels[1]);
+
+ /* numrecs verification */
+ lblock_ok = lblock_ok &&
+ be16_to_cpu(block->bb_numrecs) <= mp->m_bmap_dmxr[level != 0];
+
+ /* sibling pointer verification */
+ lblock_ok = lblock_ok &&
+ block->bb_u.l.bb_leftsib &&
+ (block->bb_u.l.bb_leftsib == cpu_to_be64(NULLDFSBNO) ||
+ XFS_FSB_SANITY_CHECK(mp,
+ be64_to_cpu(block->bb_u.l.bb_leftsib))) &&
+ block->bb_u.l.bb_rightsib &&
+ (block->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO) ||
+ XFS_FSB_SANITY_CHECK(mp,
+ be64_to_cpu(block->bb_u.l.bb_rightsib)));
+
+ if (!lblock_ok) {
+ trace_xfs_btree_corrupt(bp, _RET_IP_);
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, block);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+}
+
+static void
+xfs_bmbt_read_verify(
+ struct xfs_buf *bp)
+{
+ xfs_bmbt_verify(bp);
+}
+
+static void
+xfs_bmbt_write_verify(
+ struct xfs_buf *bp)
+{
+ xfs_bmbt_verify(bp);
+}
+
+const struct xfs_buf_ops xfs_bmbt_buf_ops = {
+ .verify_read = xfs_bmbt_read_verify,
+ .verify_write = xfs_bmbt_write_verify,
+};
+
+
#ifdef DEBUG
STATIC int
xfs_bmbt_keys_inorder(
@@ -815,7 +876,7 @@ static const struct xfs_btree_ops xfs_bmbt_ops = {
.init_rec_from_cur = xfs_bmbt_init_rec_from_cur,
.init_ptr_from_cur = xfs_bmbt_init_ptr_from_cur,
.key_diff = xfs_bmbt_key_diff,
-
+ .buf_ops = &xfs_bmbt_buf_ops,
#ifdef DEBUG
.keys_inorder = xfs_bmbt_keys_inorder,
.recs_inorder = xfs_bmbt_recs_inorder,
diff --git a/libxfs/xfs_btree.c b/libxfs/xfs_btree.c
index 02854db..c35269b 100644
--- a/libxfs/xfs_btree.c
+++ b/libxfs/xfs_btree.c
@@ -48,11 +48,11 @@ xfs_btree_check_lblock(
be16_to_cpu(block->bb_numrecs) <=
cur->bc_ops->get_maxrecs(cur, level) &&
block->bb_u.l.bb_leftsib &&
- (be64_to_cpu(block->bb_u.l.bb_leftsib) == NULLDFSBNO ||
+ (block->bb_u.l.bb_leftsib == cpu_to_be64(NULLDFSBNO) ||
XFS_FSB_SANITY_CHECK(mp,
be64_to_cpu(block->bb_u.l.bb_leftsib))) &&
block->bb_u.l.bb_rightsib &&
- (be64_to_cpu(block->bb_u.l.bb_rightsib) == NULLDFSBNO ||
+ (block->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO) ||
XFS_FSB_SANITY_CHECK(mp,
be64_to_cpu(block->bb_u.l.bb_rightsib)));
if (unlikely(XFS_TEST_ERROR(!lblock_ok, mp,
@@ -87,10 +87,10 @@ xfs_btree_check_sblock(
be16_to_cpu(block->bb_level) == level &&
be16_to_cpu(block->bb_numrecs) <=
cur->bc_ops->get_maxrecs(cur, level) &&
- (be32_to_cpu(block->bb_u.s.bb_leftsib) == NULLAGBLOCK ||
+ (block->bb_u.s.bb_leftsib == cpu_to_be32(NULLAGBLOCK) ||
be32_to_cpu(block->bb_u.s.bb_leftsib) < agflen) &&
block->bb_u.s.bb_leftsib &&
- (be32_to_cpu(block->bb_u.s.bb_rightsib) == NULLAGBLOCK ||
+ (block->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK) ||
be32_to_cpu(block->bb_u.s.bb_rightsib) < agflen) &&
block->bb_u.s.bb_rightsib;
if (unlikely(XFS_TEST_ERROR(!sblock_ok, cur->bc_mp,
@@ -250,16 +250,19 @@ xfs_btree_dup_cursor(
for (i = 0; i < new->bc_nlevels; i++) {
new->bc_ptrs[i] = cur->bc_ptrs[i];
new->bc_ra[i] = cur->bc_ra[i];
- if ((bp = cur->bc_bufs[i])) {
- if ((error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
- XFS_BUF_ADDR(bp), mp->m_bsize, 0, &bp))) {
+ bp = cur->bc_bufs[i];
+ if (bp) {
+ error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
+ XFS_BUF_ADDR(bp), mp->m_bsize,
+ 0, &bp,
+ cur->bc_ops->buf_ops);
+ if (error) {
xfs_btree_del_cursor(new, error);
*ncur = NULL;
return error;
}
new->bc_bufs[i] = bp;
- ASSERT(bp);
- ASSERT(!XFS_BUF_GETERROR(bp));
+ ASSERT(!xfs_buf_geterror(bp));
} else
new->bc_bufs[i] = NULL;
}
@@ -450,8 +453,7 @@ xfs_btree_get_bufl(
ASSERT(fsbno != NULLFSBLOCK);
d = XFS_FSB_TO_DADDR(mp, fsbno);
bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, d, mp->m_bsize, lock);
- ASSERT(bp);
- ASSERT(!XFS_BUF_GETERROR(bp));
+ ASSERT(!xfs_buf_geterror(bp));
return bp;
}
@@ -474,8 +476,7 @@ xfs_btree_get_bufs(
ASSERT(agbno != NULLAGBLOCK);
d = XFS_AGB_TO_DADDR(mp, agno, agbno);
bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, d, mp->m_bsize, lock);
- ASSERT(bp);
- ASSERT(!XFS_BUF_GETERROR(bp));
+ ASSERT(!xfs_buf_geterror(bp));
return bp;
}
@@ -493,9 +494,9 @@ xfs_btree_islastblock(
block = xfs_btree_get_block(cur, level, &bp);
xfs_btree_check_block(cur, block, level, bp);
if (cur->bc_flags & XFS_BTREE_LONG_PTRS)
- return be64_to_cpu(block->bb_u.l.bb_rightsib) == NULLDFSBNO;
+ return block->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO);
else
- return be32_to_cpu(block->bb_u.s.bb_rightsib) == NULLAGBLOCK;
+ return block->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK);
}
/*
@@ -596,69 +597,29 @@ xfs_btree_offsets(
* Get a buffer for the block, return it read in.
* Long-form addressing.
*/
-int /* error */
+int
xfs_btree_read_bufl(
- xfs_mount_t *mp, /* file system mount point */
- xfs_trans_t *tp, /* transaction pointer */
- xfs_fsblock_t fsbno, /* file system block number */
- uint lock, /* lock flags for read_buf */
- xfs_buf_t **bpp, /* buffer for fsbno */
- int refval) /* ref count value for buffer */
-{
- xfs_buf_t *bp; /* return value */
+ struct xfs_mount *mp, /* file system mount point */
+ struct xfs_trans *tp, /* transaction pointer */
+ xfs_fsblock_t fsbno, /* file system block number */
+ uint lock, /* lock flags for read_buf */
+ struct xfs_buf **bpp, /* buffer for fsbno */
+ int refval, /* ref count value for buffer */
+ const struct xfs_buf_ops *ops)
+{
+ struct xfs_buf *bp; /* return value */
xfs_daddr_t d; /* real disk block address */
- int error;
+ int error;
ASSERT(fsbno != NULLFSBLOCK);
d = XFS_FSB_TO_DADDR(mp, fsbno);
- if ((error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, d,
- mp->m_bsize, lock, &bp))) {
- return error;
- }
- ASSERT(!bp || !XFS_BUF_GETERROR(bp));
- if (bp != NULL) {
- XFS_BUF_SET_VTYPE_REF(bp, B_FS_MAP, refval);
- }
- *bpp = bp;
- return 0;
-}
-
-/*
- * Get a buffer for the block, return it read in.
- * Short-form addressing.
- */
-int /* error */
-xfs_btree_read_bufs(
- xfs_mount_t *mp, /* file system mount point */
- xfs_trans_t *tp, /* transaction pointer */
- xfs_agnumber_t agno, /* allocation group number */
- xfs_agblock_t agbno, /* allocation group block number */
- uint lock, /* lock flags for read_buf */
- xfs_buf_t **bpp, /* buffer for agno/agbno */
- int refval) /* ref count value for buffer */
-{
- xfs_buf_t *bp; /* return value */
- xfs_daddr_t d; /* real disk block address */
- int error;
-
- ASSERT(agno != NULLAGNUMBER);
- ASSERT(agbno != NULLAGBLOCK);
- d = XFS_AGB_TO_DADDR(mp, agno, agbno);
- if ((error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, d,
- mp->m_bsize, lock, &bp))) {
+ error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, d,
+ mp->m_bsize, lock, &bp, ops);
+ if (error)
return error;
- }
- ASSERT(!bp || !XFS_BUF_GETERROR(bp));
- if (bp != NULL) {
- switch (refval) {
- case XFS_ALLOC_BTREE_REF:
- XFS_BUF_SET_VTYPE_REF(bp, B_FS_MAP, refval);
- break;
- case XFS_INO_BTREE_REF:
- XFS_BUF_SET_VTYPE_REF(bp, B_FS_INOMAP, refval);
- break;
- }
- }
+ ASSERT(!xfs_buf_geterror(bp));
+ if (bp)
+ xfs_buf_set_ref(bp, refval);
*bpp = bp;
return 0;
}
@@ -674,12 +635,14 @@ xfs_btree_readahead_lblock(
xfs_dfsbno_t right = be64_to_cpu(block->bb_u.l.bb_rightsib);
if ((lr & XFS_BTCUR_LEFTRA) && left != NULLDFSBNO) {
- xfs_btree_reada_bufl(cur->bc_mp, left, 1);
+ xfs_btree_reada_bufl(cur->bc_mp, left, 1,
+ cur->bc_ops->buf_ops);
rval++;
}
if ((lr & XFS_BTCUR_RIGHTRA) && right != NULLDFSBNO) {
- xfs_btree_reada_bufl(cur->bc_mp, right, 1);
+ xfs_btree_reada_bufl(cur->bc_mp, right, 1,
+ cur->bc_ops->buf_ops);
rval++;
}
@@ -699,13 +662,13 @@ xfs_btree_readahead_sblock(
if ((lr & XFS_BTCUR_LEFTRA) && left != NULLAGBLOCK) {
xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.a.agno,
- left, 1);
+ left, 1, cur->bc_ops->buf_ops);
rval++;
}
if ((lr & XFS_BTCUR_RIGHTRA) && right != NULLAGBLOCK) {
xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.a.agno,
- right, 1);
+ right, 1, cur->bc_ops->buf_ops);
rval++;
}
@@ -762,14 +725,14 @@ xfs_btree_setbuf(
b = XFS_BUF_TO_BLOCK(bp);
if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
- if (be64_to_cpu(b->bb_u.l.bb_leftsib) == NULLDFSBNO)
+ if (b->bb_u.l.bb_leftsib == cpu_to_be64(NULLDFSBNO))
cur->bc_ra[lev] |= XFS_BTCUR_LEFTRA;
- if (be64_to_cpu(b->bb_u.l.bb_rightsib) == NULLDFSBNO)
+ if (b->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO))
cur->bc_ra[lev] |= XFS_BTCUR_RIGHTRA;
} else {
- if (be32_to_cpu(b->bb_u.s.bb_leftsib) == NULLAGBLOCK)
+ if (b->bb_u.s.bb_leftsib == cpu_to_be32(NULLAGBLOCK))
cur->bc_ra[lev] |= XFS_BTCUR_LEFTRA;
- if (be32_to_cpu(b->bb_u.s.bb_rightsib) == NULLAGBLOCK)
+ if (b->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK))
cur->bc_ra[lev] |= XFS_BTCUR_RIGHTRA;
}
}
@@ -780,9 +743,9 @@ xfs_btree_ptr_is_null(
union xfs_btree_ptr *ptr)
{
if (cur->bc_flags & XFS_BTREE_LONG_PTRS)
- return be64_to_cpu(ptr->l) == NULLDFSBNO;
+ return ptr->l == cpu_to_be64(NULLDFSBNO);
else
- return be32_to_cpu(ptr->s) == NULLAGBLOCK;
+ return ptr->s == cpu_to_be32(NULLAGBLOCK);
}
STATIC void
@@ -843,18 +806,22 @@ xfs_btree_set_sibling(
}
}
-STATIC void
+void
xfs_btree_init_block(
- struct xfs_btree_cur *cur,
- int level,
- int numrecs,
- struct xfs_btree_block *new) /* new block */
+ struct xfs_mount *mp,
+ struct xfs_buf *bp,
+ __u32 magic,
+ __u16 level,
+ __u16 numrecs,
+ unsigned int flags)
{
- new->bb_magic = cpu_to_be32(xfs_magics[cur->bc_btnum]);
+ struct xfs_btree_block *new = XFS_BUF_TO_BLOCK(bp);
+
+ new->bb_magic = cpu_to_be32(magic);
new->bb_level = cpu_to_be16(level);
new->bb_numrecs = cpu_to_be16(numrecs);
- if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
+ if (flags & XFS_BTREE_LONG_PTRS) {
new->bb_u.l.bb_leftsib = cpu_to_be64(NULLDFSBNO);
new->bb_u.l.bb_rightsib = cpu_to_be64(NULLDFSBNO);
} else {
@@ -863,6 +830,17 @@ xfs_btree_init_block(
}
}
+STATIC void
+xfs_btree_init_block_cur(
+ struct xfs_btree_cur *cur,
+ int level,
+ int numrecs,
+ struct xfs_buf *bp)
+{
+ xfs_btree_init_block(cur->bc_mp, bp, xfs_magics[cur->bc_btnum],
+ level, numrecs, cur->bc_flags);
+}
+
/*
* Return true if ptr is the last record in the btree and
* we need to track updateѕ to this record. The decision
@@ -908,12 +886,12 @@ xfs_btree_ptr_to_daddr(
union xfs_btree_ptr *ptr)
{
if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
- ASSERT(be64_to_cpu(ptr->l) != NULLDFSBNO);
+ ASSERT(ptr->l != cpu_to_be64(NULLDFSBNO));
return XFS_FSB_TO_DADDR(cur->bc_mp, be64_to_cpu(ptr->l));
} else {
ASSERT(cur->bc_private.a.agno != NULLAGNUMBER);
- ASSERT(be32_to_cpu(ptr->s) != NULLAGBLOCK);
+ ASSERT(ptr->s != cpu_to_be32(NULLAGBLOCK));
return XFS_AGB_TO_DADDR(cur->bc_mp, cur->bc_private.a.agno,
be32_to_cpu(ptr->s));
@@ -928,13 +906,13 @@ xfs_btree_set_refs(
switch (cur->bc_btnum) {
case XFS_BTNUM_BNO:
case XFS_BTNUM_CNT:
- XFS_BUF_SET_VTYPE_REF(bp, B_FS_MAP, XFS_ALLOC_BTREE_REF);
+ xfs_buf_set_ref(bp, XFS_ALLOC_BTREE_REF);
break;
case XFS_BTNUM_INO:
- XFS_BUF_SET_VTYPE_REF(bp, B_FS_INOMAP, XFS_INO_BTREE_REF);
+ xfs_buf_set_ref(bp, XFS_INO_BTREE_REF);
break;
case XFS_BTNUM_BMAP:
- XFS_BUF_SET_VTYPE_REF(bp, B_FS_MAP, XFS_BMAP_BTREE_REF);
+ xfs_buf_set_ref(bp, XFS_BMAP_BTREE_REF);
break;
default:
ASSERT(0);
@@ -959,9 +937,10 @@ xfs_btree_get_buf_block(
*bpp = xfs_trans_get_buf(cur->bc_tp, mp->m_ddev_targp, d,
mp->m_bsize, flags);
- ASSERT(*bpp);
- ASSERT(!XFS_BUF_GETERROR(*bpp));
+ if (!*bpp)
+ return ENOMEM;
+ (*bpp)->b_ops = cur->bc_ops->buf_ops;
*block = XFS_BUF_TO_BLOCK(*bpp);
return 0;
}
@@ -988,20 +967,15 @@ xfs_btree_read_buf_block(
d = xfs_btree_ptr_to_daddr(cur, ptr);
error = xfs_trans_read_buf(mp, cur->bc_tp, mp->m_ddev_targp, d,
- mp->m_bsize, flags, bpp);
+ mp->m_bsize, flags, bpp,
+ cur->bc_ops->buf_ops);
if (error)
return error;
- ASSERT(*bpp != NULL);
- ASSERT(!XFS_BUF_GETERROR(*bpp));
-
+ ASSERT(!xfs_buf_geterror(*bpp));
xfs_btree_set_refs(cur, *bpp);
*block = XFS_BUF_TO_BLOCK(*bpp);
-
- error = xfs_btree_check_block(cur, *block, level, *bpp);
- if (error)
- xfs_trans_brelse(cur->bc_tp, *bpp);
- return error;
+ return 0;
}
/*
@@ -2174,7 +2148,7 @@ xfs_btree_split(
goto error0;
/* Fill in the btree header for the new right block. */
- xfs_btree_init_block(cur, xfs_btree_get_level(left), 0, right);
+ xfs_btree_init_block_cur(cur, xfs_btree_get_level(left), 0, rbp);
/*
* Split the entries between the old and the new block evenly.
@@ -2483,7 +2457,7 @@ xfs_btree_new_root(
nptr = 2;
}
/* Fill in the new block's btree header and log it. */
- xfs_btree_init_block(cur, cur->bc_nlevels, 2, new);
+ xfs_btree_init_block_cur(cur, cur->bc_nlevels, 2, nbp);
xfs_btree_log_block(cur, nbp, XFS_BB_ALL_BITS);
ASSERT(!xfs_btree_ptr_is_null(cur, &lptr) &&
!xfs_btree_ptr_is_null(cur, &rptr));
diff --git a/libxfs/xfs_da_btree.c b/libxfs/xfs_da_btree.c
index 3eb34d6..a31d353 100644
--- a/libxfs/xfs_da_btree.c
+++ b/libxfs/xfs_da_btree.c
@@ -62,14 +62,92 @@ STATIC void xfs_da_node_unbalance(xfs_da_state_t *state,
/*
* Utility routines.
*/
-STATIC uint xfs_da_node_lasthash(xfs_dabuf_t *bp, int *count);
-STATIC int xfs_da_node_order(xfs_dabuf_t *node1_bp, xfs_dabuf_t *node2_bp);
-STATIC xfs_dabuf_t *xfs_da_buf_make(int nbuf, xfs_buf_t **bps, inst_t *ra);
+STATIC uint xfs_da_node_lasthash(struct xfs_buf *bp, int *count);
+STATIC int xfs_da_node_order(struct xfs_buf *node1_bp,
+ struct xfs_buf *node2_bp);
STATIC int xfs_da_blk_unlink(xfs_da_state_t *state,
xfs_da_state_blk_t *drop_blk,
xfs_da_state_blk_t *save_blk);
STATIC void xfs_da_state_kill_altpath(xfs_da_state_t *state);
+static void
+xfs_da_node_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_da_node_hdr *hdr = bp->b_addr;
+ int block_ok = 0;
+
+ block_ok = hdr->info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC);
+ block_ok = block_ok &&
+ be16_to_cpu(hdr->level) > 0 &&
+ be16_to_cpu(hdr->count) > 0 ;
+ if (!block_ok) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, hdr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+
+}
+
+static void
+xfs_da_node_write_verify(
+ struct xfs_buf *bp)
+{
+ xfs_da_node_verify(bp);
+}
+
+/*
+ * leaf/node format detection on trees is sketchy, so a node read can be done on
+ * leaf level blocks when detection identifies the tree as a node format tree
+ * incorrectly. In this case, we need to swap the verifier to match the correct
+ * format of the block being read.
+ */
+static void
+xfs_da_node_read_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_da_blkinfo *info = bp->b_addr;
+
+ switch (be16_to_cpu(info->magic)) {
+ case XFS_DA_NODE_MAGIC:
+ xfs_da_node_verify(bp);
+ break;
+ case XFS_ATTR_LEAF_MAGIC:
+ bp->b_ops = &xfs_attr_leaf_buf_ops;
+ bp->b_ops->verify_read(bp);
+ return;
+ case XFS_DIR2_LEAFN_MAGIC:
+ bp->b_ops = &xfs_dir2_leafn_buf_ops;
+ bp->b_ops->verify_read(bp);
+ return;
+ default:
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
+ mp, info);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ break;
+ }
+}
+
+const struct xfs_buf_ops xfs_da_node_buf_ops = {
+ .verify_read = xfs_da_node_read_verify,
+ .verify_write = xfs_da_node_write_verify,
+};
+
+
+int
+xfs_da_node_read(
+ struct xfs_trans *tp,
+ struct xfs_inode *dp,
+ xfs_dablk_t bno,
+ xfs_daddr_t mappedbno,
+ struct xfs_buf **bpp,
+ int which_fork)
+{
+ return xfs_da_read_buf(tp, dp, bno, mappedbno, bpp,
+ which_fork, &xfs_da_node_buf_ops);
+}
+
/*========================================================================
* Routines used for growing the Btree.
*========================================================================*/
@@ -79,19 +157,21 @@ STATIC void xfs_da_state_kill_altpath(xfs_da_state_t *state);
*/
int
xfs_da_node_create(xfs_da_args_t *args, xfs_dablk_t blkno, int level,
- xfs_dabuf_t **bpp, int whichfork)
+ struct xfs_buf **bpp, int whichfork)
{
xfs_da_intnode_t *node;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
int error;
xfs_trans_t *tp;
+ trace_xfs_da_node_create(args);
+
tp = args->trans;
error = xfs_da_get_buf(tp, args->dp, blkno, -1, &bp, whichfork);
if (error)
return(error);
ASSERT(bp != NULL);
- node = bp->data;
+ node = bp->b_addr;
node->hdr.info.forw = 0;
node->hdr.info.back = 0;
node->hdr.info.magic = cpu_to_be16(XFS_DA_NODE_MAGIC);
@@ -99,9 +179,10 @@ xfs_da_node_create(xfs_da_args_t *args, xfs_dablk_t blkno, int level,
node->hdr.count = 0;
node->hdr.level = cpu_to_be16(level);
- xfs_da_log_buf(tp, bp,
+ xfs_trans_log_buf(tp, bp,
XFS_DA_LOGRANGE(node, &node->hdr, sizeof(node->hdr)));
+ bp->b_ops = &xfs_da_node_buf_ops;
*bpp = bp;
return(0);
}
@@ -115,9 +196,11 @@ xfs_da_split(xfs_da_state_t *state)
{
xfs_da_state_blk_t *oldblk, *newblk, *addblk;
xfs_da_intnode_t *node;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
int max, action, error, i;
+ trace_xfs_da_split(state->args);
+
/*
* Walk back up the tree splitting/inserting/adjusting as necessary.
* If we need to insert and there isn't room, split the node, then
@@ -156,10 +239,12 @@ xfs_da_split(xfs_da_state_t *state)
state->extravalid = 1;
if (state->inleaf) {
state->extraafter = 0; /* before newblk */
+ trace_xfs_attr_leaf_split_before(state->args);
error = xfs_attr_leaf_split(state, oldblk,
&state->extrablk);
} else {
state->extraafter = 1; /* after newblk */
+ trace_xfs_attr_leaf_split_after(state->args);
error = xfs_attr_leaf_split(state, newblk,
&state->extrablk);
}
@@ -176,7 +261,6 @@ xfs_da_split(xfs_da_state_t *state)
case XFS_DA_NODE_MAGIC:
error = xfs_da_node_split(state, oldblk, newblk, addblk,
max - i, &action);
- xfs_da_buf_done(addblk->bp);
addblk->bp = NULL;
if (error)
return(error); /* GROT: dir is inconsistent */
@@ -194,13 +278,6 @@ xfs_da_split(xfs_da_state_t *state)
* Update the btree to show the new hashval for this child.
*/
xfs_da_fixhashpath(state, &state->path);
- /*
- * If we won't need this block again, it's getting dropped
- * from the active path by the loop control, so we need
- * to mark it done now.
- */
- if (i > 0 || !addblk)
- xfs_da_buf_done(oldblk->bp);
}
if (!addblk)
return(0);
@@ -212,8 +289,6 @@ xfs_da_split(xfs_da_state_t *state)
oldblk = &state->path.blk[0];
error = xfs_da_root_split(state, oldblk, addblk);
if (error) {
- xfs_da_buf_done(oldblk->bp);
- xfs_da_buf_done(addblk->bp);
addblk->bp = NULL;
return(error); /* GROT: dir is inconsistent */
}
@@ -225,7 +300,7 @@ xfs_da_split(xfs_da_state_t *state)
* and the original block 0 could be at any position in the list.
*/
- node = oldblk->bp->data;
+ node = oldblk->bp->b_addr;
if (node->hdr.info.forw) {
if (be32_to_cpu(node->hdr.info.forw) == addblk->blkno) {
bp = addblk->bp;
@@ -233,13 +308,13 @@ xfs_da_split(xfs_da_state_t *state)
ASSERT(state->extravalid);
bp = state->extrablk.bp;
}
- node = bp->data;
+ node = bp->b_addr;
node->hdr.info.back = cpu_to_be32(oldblk->blkno);
- xfs_da_log_buf(state->args->trans, bp,
+ xfs_trans_log_buf(state->args->trans, bp,
XFS_DA_LOGRANGE(node, &node->hdr.info,
sizeof(node->hdr.info)));
}
- node = oldblk->bp->data;
+ node = oldblk->bp->b_addr;
if (node->hdr.info.back) {
if (be32_to_cpu(node->hdr.info.back) == addblk->blkno) {
bp = addblk->bp;
@@ -247,14 +322,12 @@ xfs_da_split(xfs_da_state_t *state)
ASSERT(state->extravalid);
bp = state->extrablk.bp;
}
- node = bp->data;
+ node = bp->b_addr;
node->hdr.info.forw = cpu_to_be32(oldblk->blkno);
- xfs_da_log_buf(state->args->trans, bp,
+ xfs_trans_log_buf(state->args->trans, bp,
XFS_DA_LOGRANGE(node, &node->hdr.info,
sizeof(node->hdr.info)));
}
- xfs_da_buf_done(oldblk->bp);
- xfs_da_buf_done(addblk->bp);
addblk->bp = NULL;
return(0);
}
@@ -271,13 +344,15 @@ xfs_da_root_split(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
xfs_da_intnode_t *node, *oldroot;
xfs_da_args_t *args;
xfs_dablk_t blkno;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
int error, size;
xfs_inode_t *dp;
xfs_trans_t *tp;
xfs_mount_t *mp;
xfs_dir2_leaf_t *leaf;
+ trace_xfs_da_root_split(state->args);
+
/*
* Copy the existing (incorrect) block from the root node position
* to a free space somewhere.
@@ -294,20 +369,21 @@ xfs_da_root_split(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
if (error)
return(error);
ASSERT(bp != NULL);
- node = bp->data;
- oldroot = blk1->bp->data;
- if (be16_to_cpu(oldroot->hdr.info.magic) == XFS_DA_NODE_MAGIC) {
+ node = bp->b_addr;
+ oldroot = blk1->bp->b_addr;
+ if (oldroot->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC)) {
size = (int)((char *)&oldroot->btree[be16_to_cpu(oldroot->hdr.count)] -
(char *)oldroot);
} else {
- ASSERT(be16_to_cpu(oldroot->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
+ ASSERT(oldroot->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
leaf = (xfs_dir2_leaf_t *)oldroot;
size = (int)((char *)&leaf->ents[be16_to_cpu(leaf->hdr.count)] -
(char *)leaf);
}
memcpy(node, oldroot, size);
- xfs_da_log_buf(tp, bp, 0, size - 1);
- xfs_da_buf_done(blk1->bp);
+ xfs_trans_log_buf(tp, bp, 0, size - 1);
+
+ bp->b_ops = blk1->bp->b_ops;
blk1->bp = bp;
blk1->blkno = blkno;
@@ -319,7 +395,7 @@ xfs_da_root_split(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
be16_to_cpu(node->hdr.level) + 1, &bp, args->whichfork);
if (error)
return(error);
- node = bp->data;
+ node = bp->b_addr;
node->btree[0].hashval = cpu_to_be32(blk1->hashval);
node->btree[0].before = cpu_to_be32(blk1->blkno);
node->btree[1].hashval = cpu_to_be32(blk2->hashval);
@@ -327,7 +403,7 @@ xfs_da_root_split(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
node->hdr.count = cpu_to_be16(2);
#ifdef DEBUG
- if (be16_to_cpu(oldroot->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC) {
+ if (oldroot->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC)) {
ASSERT(blk1->blkno >= mp->m_dirleafblk &&
blk1->blkno < mp->m_dirfreeblk);
ASSERT(blk2->blkno >= mp->m_dirleafblk &&
@@ -336,10 +412,9 @@ xfs_da_root_split(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
#endif
/* Header is already logged by xfs_da_node_create */
- xfs_da_log_buf(tp, bp,
+ xfs_trans_log_buf(tp, bp,
XFS_DA_LOGRANGE(node, node->btree,
sizeof(xfs_da_node_entry_t) * 2));
- xfs_da_buf_done(bp);
return(0);
}
@@ -358,8 +433,10 @@ xfs_da_node_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
int newcount, error;
int useextra;
- node = oldblk->bp->data;
- ASSERT(be16_to_cpu(node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
+ trace_xfs_da_node_split(state->args);
+
+ node = oldblk->bp->b_addr;
+ ASSERT(node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
/*
* With V2 dirs the extra block is data or freespace.
@@ -405,7 +482,7 @@ xfs_da_node_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
*
* If we had double-split op below us, then add the extra block too.
*/
- node = oldblk->bp->data;
+ node = oldblk->bp->b_addr;
if (oldblk->index <= be16_to_cpu(node->hdr.count)) {
oldblk->index++;
xfs_da_node_add(state, oldblk, addblk);
@@ -444,8 +521,10 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
int count, tmp;
xfs_trans_t *tp;
- node1 = blk1->bp->data;
- node2 = blk2->bp->data;
+ trace_xfs_da_node_rebalance(state->args);
+
+ node1 = blk1->bp->b_addr;
+ node2 = blk2->bp->b_addr;
/*
* Figure out how many entries need to move, and in which direction.
* Swap the nodes around if that makes it simpler.
@@ -458,8 +537,8 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
node1 = node2;
node2 = tmpnode;
}
- ASSERT(be16_to_cpu(node1->hdr.info.magic) == XFS_DA_NODE_MAGIC);
- ASSERT(be16_to_cpu(node2->hdr.info.magic) == XFS_DA_NODE_MAGIC);
+ ASSERT(node1->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
+ ASSERT(node2->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
count = (be16_to_cpu(node1->hdr.count) - be16_to_cpu(node2->hdr.count)) / 2;
if (count == 0)
return;
@@ -499,7 +578,7 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
btree_d = &node1->btree[be16_to_cpu(node1->hdr.count)];
memcpy(btree_d, btree_s, tmp);
be16_add_cpu(&node1->hdr.count, count);
- xfs_da_log_buf(tp, blk1->bp,
+ xfs_trans_log_buf(tp, blk1->bp,
XFS_DA_LOGRANGE(node1, btree_d, tmp));
/*
@@ -516,9 +595,9 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
/*
* Log header of node 1 and all current bits of node 2.
*/
- xfs_da_log_buf(tp, blk1->bp,
+ xfs_trans_log_buf(tp, blk1->bp,
XFS_DA_LOGRANGE(node1, &node1->hdr, sizeof(node1->hdr)));
- xfs_da_log_buf(tp, blk2->bp,
+ xfs_trans_log_buf(tp, blk2->bp,
XFS_DA_LOGRANGE(node2, &node2->hdr,
sizeof(node2->hdr) +
sizeof(node2->btree[0]) * be16_to_cpu(node2->hdr.count)));
@@ -527,8 +606,8 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
* Record the last hashval from each block for upward propagation.
* (note: don't use the swapped node pointers)
*/
- node1 = blk1->bp->data;
- node2 = blk2->bp->data;
+ node1 = blk1->bp->b_addr;
+ node2 = blk2->bp->b_addr;
blk1->hashval = be32_to_cpu(node1->btree[be16_to_cpu(node1->hdr.count)-1].hashval);
blk2->hashval = be32_to_cpu(node2->btree[be16_to_cpu(node2->hdr.count)-1].hashval);
@@ -552,8 +631,10 @@ xfs_da_node_add(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
xfs_da_node_entry_t *btree;
int tmp;
- node = oldblk->bp->data;
- ASSERT(be16_to_cpu(node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
+ trace_xfs_da_node_add(state->args);
+
+ node = oldblk->bp->b_addr;
+ ASSERT(node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
ASSERT((oldblk->index >= 0) && (oldblk->index <= be16_to_cpu(node->hdr.count)));
ASSERT(newblk->blkno != 0);
if (state->args->whichfork == XFS_DATA_FORK)
@@ -571,10 +652,10 @@ xfs_da_node_add(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
}
btree->hashval = cpu_to_be32(newblk->hashval);
btree->before = cpu_to_be32(newblk->blkno);
- xfs_da_log_buf(state->args->trans, oldblk->bp,
+ xfs_trans_log_buf(state->args->trans, oldblk->bp,
XFS_DA_LOGRANGE(node, btree, tmp + sizeof(*btree)));
be16_add_cpu(&node->hdr.count, 1);
- xfs_da_log_buf(state->args->trans, oldblk->bp,
+ xfs_trans_log_buf(state->args->trans, oldblk->bp,
XFS_DA_LOGRANGE(node, &node->hdr, sizeof(node->hdr)));
/*
@@ -597,6 +678,8 @@ xfs_da_join(xfs_da_state_t *state)
xfs_da_state_blk_t *drop_blk, *save_blk;
int action, error;
+ trace_xfs_da_join(state->args);
+
action = 0;
drop_blk = &state->path.blk[ state->path.active-1 ];
save_blk = &state->altpath.blk[ state->path.active-1 ];
@@ -670,6 +753,24 @@ xfs_da_join(xfs_da_state_t *state)
return(error);
}
+#ifdef DEBUG
+static void
+xfs_da_blkinfo_onlychild_validate(struct xfs_da_blkinfo *blkinfo, __u16 level)
+{
+ __be16 magic = blkinfo->magic;
+
+ if (level == 1) {
+ ASSERT(magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
+ magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
+ } else
+ ASSERT(magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
+ ASSERT(!blkinfo->forw);
+ ASSERT(!blkinfo->back);
+}
+#else /* !DEBUG */
+#define xfs_da_blkinfo_onlychild_validate(blkinfo, level)
+#endif /* !DEBUG */
+
/*
* We have only one entry in the root. Copy the only remaining child of
* the old root to block 0 as the new root node.
@@ -678,18 +779,18 @@ STATIC int
xfs_da_root_join(xfs_da_state_t *state, xfs_da_state_blk_t *root_blk)
{
xfs_da_intnode_t *oldroot;
- /* REFERENCED */
- xfs_da_blkinfo_t *blkinfo;
xfs_da_args_t *args;
xfs_dablk_t child;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
int error;
+ trace_xfs_da_root_join(state->args);
+
args = state->args;
ASSERT(args != NULL);
ASSERT(root_blk->magic == XFS_DA_NODE_MAGIC);
- oldroot = root_blk->bp->data;
- ASSERT(be16_to_cpu(oldroot->hdr.info.magic) == XFS_DA_NODE_MAGIC);
+ oldroot = root_blk->bp->b_addr;
+ ASSERT(oldroot->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
ASSERT(!oldroot->hdr.info.forw);
ASSERT(!oldroot->hdr.info.back);
@@ -705,22 +806,23 @@ xfs_da_root_join(xfs_da_state_t *state, xfs_da_state_blk_t *root_blk)
*/
child = be32_to_cpu(oldroot->btree[0].before);
ASSERT(child != 0);
- error = xfs_da_read_buf(args->trans, args->dp, child, -1, &bp,
+ error = xfs_da_node_read(args->trans, args->dp, child, -1, &bp,
args->whichfork);
if (error)
return(error);
ASSERT(bp != NULL);
- blkinfo = bp->data;
- if (be16_to_cpu(oldroot->hdr.level) == 1) {
- ASSERT(be16_to_cpu(blkinfo->magic) == XFS_DIR2_LEAFN_MAGIC ||
- be16_to_cpu(blkinfo->magic) == XFS_ATTR_LEAF_MAGIC);
- } else {
- ASSERT(be16_to_cpu(blkinfo->magic) == XFS_DA_NODE_MAGIC);
- }
- ASSERT(!blkinfo->forw);
- ASSERT(!blkinfo->back);
- memcpy(root_blk->bp->data, bp->data, state->blocksize);
- xfs_da_log_buf(args->trans, root_blk->bp, 0, state->blocksize - 1);
+ xfs_da_blkinfo_onlychild_validate(bp->b_addr,
+ be16_to_cpu(oldroot->hdr.level));
+
+ /*
+ * This could be copying a leaf back into the root block in the case of
+ * there only being a single leaf block left in the tree. Hence we have
+ * to update the b_ops pointer as well to match the buffer type change
+ * that could occur.
+ */
+ memcpy(root_blk->bp->b_addr, bp->b_addr, state->blocksize);
+ root_blk->bp->b_ops = bp->b_ops;
+ xfs_trans_log_buf(args->trans, root_blk->bp, 0, state->blocksize - 1);
error = xfs_da_shrink_inode(args, child, bp);
return(error);
}
@@ -742,7 +844,9 @@ xfs_da_node_toosmall(xfs_da_state_t *state, int *action)
xfs_da_blkinfo_t *info;
int count, forward, error, retval, i;
xfs_dablk_t blkno;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
+
+ trace_xfs_da_node_toosmall(state->args);
/*
* Check for the degenerate case of the block being over 50% full.
@@ -750,8 +854,8 @@ xfs_da_node_toosmall(xfs_da_state_t *state, int *action)
* to coalesce with a sibling.
*/
blk = &state->path.blk[ state->path.active-1 ];
- info = blk->bp->data;
- ASSERT(be16_to_cpu(info->magic) == XFS_DA_NODE_MAGIC);
+ info = blk->bp->b_addr;
+ ASSERT(info->magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
node = (xfs_da_intnode_t *)info;
count = be16_to_cpu(node->hdr.count);
if (count > (state->node_ents >> 1)) {
@@ -800,7 +904,7 @@ xfs_da_node_toosmall(xfs_da_state_t *state, int *action)
blkno = be32_to_cpu(info->back);
if (blkno == 0)
continue;
- error = xfs_da_read_buf(state->args->trans, state->args->dp,
+ error = xfs_da_node_read(state->args->trans, state->args->dp,
blkno, -1, &bp, state->args->whichfork);
if (error)
return(error);
@@ -810,10 +914,10 @@ xfs_da_node_toosmall(xfs_da_state_t *state, int *action)
count = state->node_ents;
count -= state->node_ents >> 2;
count -= be16_to_cpu(node->hdr.count);
- node = bp->data;
- ASSERT(be16_to_cpu(node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
+ node = bp->b_addr;
+ ASSERT(node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
count -= be16_to_cpu(node->hdr.count);
- xfs_da_brelse(state->args->trans, bp);
+ xfs_trans_brelse(state->args->trans, bp);
if (count >= 0)
break; /* fits with at least 25% to spare */
}
@@ -865,6 +969,8 @@ xfs_da_fixhashpath(xfs_da_state_t *state, xfs_da_state_path_t *path)
xfs_dahash_t lasthash=0;
int level, count;
+ trace_xfs_da_fixhashpath(state->args);
+
level = path->active-1;
blk = &path->blk[ level ];
switch (blk->magic) {
@@ -885,14 +991,14 @@ xfs_da_fixhashpath(xfs_da_state_t *state, xfs_da_state_path_t *path)
break;
}
for (blk--, level--; level >= 0; blk--, level--) {
- node = blk->bp->data;
- ASSERT(be16_to_cpu(node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
+ node = blk->bp->b_addr;
+ ASSERT(node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
btree = &node->btree[ blk->index ];
if (be32_to_cpu(btree->hashval) == lasthash)
break;
blk->hashval = lasthash;
btree->hashval = cpu_to_be32(lasthash);
- xfs_da_log_buf(state->args->trans, blk->bp,
+ xfs_trans_log_buf(state->args->trans, blk->bp,
XFS_DA_LOGRANGE(node, btree, sizeof(*btree)));
lasthash = be32_to_cpu(node->btree[be16_to_cpu(node->hdr.count)-1].hashval);
@@ -909,7 +1015,9 @@ xfs_da_node_remove(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk)
xfs_da_node_entry_t *btree;
int tmp;
- node = drop_blk->bp->data;
+ trace_xfs_da_node_remove(state->args);
+
+ node = drop_blk->bp->b_addr;
ASSERT(drop_blk->index < be16_to_cpu(node->hdr.count));
ASSERT(drop_blk->index >= 0);
@@ -921,15 +1029,15 @@ xfs_da_node_remove(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk)
tmp = be16_to_cpu(node->hdr.count) - drop_blk->index - 1;
tmp *= (uint)sizeof(xfs_da_node_entry_t);
memmove(btree, btree + 1, tmp);
- xfs_da_log_buf(state->args->trans, drop_blk->bp,
+ xfs_trans_log_buf(state->args->trans, drop_blk->bp,
XFS_DA_LOGRANGE(node, btree, tmp));
btree = &node->btree[be16_to_cpu(node->hdr.count)-1];
}
memset((char *)btree, 0, sizeof(xfs_da_node_entry_t));
- xfs_da_log_buf(state->args->trans, drop_blk->bp,
+ xfs_trans_log_buf(state->args->trans, drop_blk->bp,
XFS_DA_LOGRANGE(node, btree, sizeof(*btree)));
be16_add_cpu(&node->hdr.count, -1);
- xfs_da_log_buf(state->args->trans, drop_blk->bp,
+ xfs_trans_log_buf(state->args->trans, drop_blk->bp,
XFS_DA_LOGRANGE(node, &node->hdr, sizeof(node->hdr)));
/*
@@ -952,10 +1060,12 @@ xfs_da_node_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
int tmp;
xfs_trans_t *tp;
- drop_node = drop_blk->bp->data;
- save_node = save_blk->bp->data;
- ASSERT(be16_to_cpu(drop_node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
- ASSERT(be16_to_cpu(save_node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
+ trace_xfs_da_node_unbalance(state->args);
+
+ drop_node = drop_blk->bp->b_addr;
+ save_node = save_blk->bp->b_addr;
+ ASSERT(drop_node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
+ ASSERT(save_node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
tp = state->args->trans;
/*
@@ -970,13 +1080,13 @@ xfs_da_node_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
tmp = be16_to_cpu(save_node->hdr.count) * (uint)sizeof(xfs_da_node_entry_t);
memmove(btree, &save_node->btree[0], tmp);
btree = &save_node->btree[0];
- xfs_da_log_buf(tp, save_blk->bp,
+ xfs_trans_log_buf(tp, save_blk->bp,
XFS_DA_LOGRANGE(save_node, btree,
(be16_to_cpu(save_node->hdr.count) + be16_to_cpu(drop_node->hdr.count)) *
sizeof(xfs_da_node_entry_t)));
} else {
btree = &save_node->btree[be16_to_cpu(save_node->hdr.count)];
- xfs_da_log_buf(tp, save_blk->bp,
+ xfs_trans_log_buf(tp, save_blk->bp,
XFS_DA_LOGRANGE(save_node, btree,
be16_to_cpu(drop_node->hdr.count) *
sizeof(xfs_da_node_entry_t)));
@@ -989,7 +1099,7 @@ xfs_da_node_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
memcpy(btree, &drop_node->btree[0], tmp);
be16_add_cpu(&save_node->hdr.count, be16_to_cpu(drop_node->hdr.count));
- xfs_da_log_buf(tp, save_blk->bp,
+ xfs_trans_log_buf(tp, save_blk->bp,
XFS_DA_LOGRANGE(save_node, &save_node->hdr,
sizeof(save_node->hdr)));
@@ -1040,14 +1150,14 @@ xfs_da_node_lookup_int(xfs_da_state_t *state, int *result)
* Read the next node down in the tree.
*/
blk->blkno = blkno;
- error = xfs_da_read_buf(args->trans, args->dp, blkno,
+ error = xfs_da_node_read(args->trans, args->dp, blkno,
-1, &blk->bp, args->whichfork);
if (error) {
blk->blkno = 0;
state->path.active--;
return(error);
}
- curr = blk->bp->data;
+ curr = blk->bp->b_addr;
blk->magic = be16_to_cpu(curr->magic);
ASSERT(blk->magic == XFS_DA_NODE_MAGIC ||
blk->magic == XFS_DIR2_LEAFN_MAGIC ||
@@ -1057,7 +1167,7 @@ xfs_da_node_lookup_int(xfs_da_state_t *state, int *result)
* Search an intermediate node for a match.
*/
if (blk->magic == XFS_DA_NODE_MAGIC) {
- node = blk->bp->data;
+ node = blk->bp->b_addr;
max = be16_to_cpu(node->hdr.count);
blk->hashval = be32_to_cpu(node->btree[max-1].hashval);
@@ -1163,15 +1273,15 @@ xfs_da_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
xfs_da_blkinfo_t *old_info, *new_info, *tmp_info;
xfs_da_args_t *args;
int before=0, error;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
/*
* Set up environment.
*/
args = state->args;
ASSERT(args != NULL);
- old_info = old_blk->bp->data;
- new_info = new_blk->bp->data;
+ old_info = old_blk->bp->b_addr;
+ new_info = new_blk->bp->b_addr;
ASSERT(old_blk->magic == XFS_DA_NODE_MAGIC ||
old_blk->magic == XFS_DIR2_LEAFN_MAGIC ||
old_blk->magic == XFS_ATTR_LEAF_MAGIC);
@@ -1198,48 +1308,48 @@ xfs_da_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
/*
* Link new block in before existing block.
*/
+ trace_xfs_da_link_before(args);
new_info->forw = cpu_to_be32(old_blk->blkno);
new_info->back = old_info->back;
if (old_info->back) {
- error = xfs_da_read_buf(args->trans, args->dp,
+ error = xfs_da_node_read(args->trans, args->dp,
be32_to_cpu(old_info->back),
-1, &bp, args->whichfork);
if (error)
return(error);
ASSERT(bp != NULL);
- tmp_info = bp->data;
+ tmp_info = bp->b_addr;
ASSERT(be16_to_cpu(tmp_info->magic) == be16_to_cpu(old_info->magic));
ASSERT(be32_to_cpu(tmp_info->forw) == old_blk->blkno);
tmp_info->forw = cpu_to_be32(new_blk->blkno);
- xfs_da_log_buf(args->trans, bp, 0, sizeof(*tmp_info)-1);
- xfs_da_buf_done(bp);
+ xfs_trans_log_buf(args->trans, bp, 0, sizeof(*tmp_info)-1);
}
old_info->back = cpu_to_be32(new_blk->blkno);
} else {
/*
* Link new block in after existing block.
*/
+ trace_xfs_da_link_after(args);
new_info->forw = old_info->forw;
new_info->back = cpu_to_be32(old_blk->blkno);
if (old_info->forw) {
- error = xfs_da_read_buf(args->trans, args->dp,
+ error = xfs_da_node_read(args->trans, args->dp,
be32_to_cpu(old_info->forw),
-1, &bp, args->whichfork);
if (error)
return(error);
ASSERT(bp != NULL);
- tmp_info = bp->data;
+ tmp_info = bp->b_addr;
ASSERT(tmp_info->magic == old_info->magic);
ASSERT(be32_to_cpu(tmp_info->back) == old_blk->blkno);
tmp_info->back = cpu_to_be32(new_blk->blkno);
- xfs_da_log_buf(args->trans, bp, 0, sizeof(*tmp_info)-1);
- xfs_da_buf_done(bp);
+ xfs_trans_log_buf(args->trans, bp, 0, sizeof(*tmp_info)-1);
}
old_info->forw = cpu_to_be32(new_blk->blkno);
}
- xfs_da_log_buf(args->trans, old_blk->bp, 0, sizeof(*tmp_info) - 1);
- xfs_da_log_buf(args->trans, new_blk->bp, 0, sizeof(*tmp_info) - 1);
+ xfs_trans_log_buf(args->trans, old_blk->bp, 0, sizeof(*tmp_info) - 1);
+ xfs_trans_log_buf(args->trans, new_blk->bp, 0, sizeof(*tmp_info) - 1);
return(0);
}
@@ -1247,14 +1357,16 @@ xfs_da_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
* Compare two intermediate nodes for "order".
*/
STATIC int
-xfs_da_node_order(xfs_dabuf_t *node1_bp, xfs_dabuf_t *node2_bp)
+xfs_da_node_order(
+ struct xfs_buf *node1_bp,
+ struct xfs_buf *node2_bp)
{
xfs_da_intnode_t *node1, *node2;
- node1 = node1_bp->data;
- node2 = node2_bp->data;
- ASSERT((be16_to_cpu(node1->hdr.info.magic) == XFS_DA_NODE_MAGIC) &&
- (be16_to_cpu(node2->hdr.info.magic) == XFS_DA_NODE_MAGIC));
+ node1 = node1_bp->b_addr;
+ node2 = node2_bp->b_addr;
+ ASSERT(node1->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC) &&
+ node2->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
if ((be16_to_cpu(node1->hdr.count) > 0) && (be16_to_cpu(node2->hdr.count) > 0) &&
((be32_to_cpu(node2->btree[0].hashval) <
be32_to_cpu(node1->btree[0].hashval)) ||
@@ -1269,12 +1381,14 @@ xfs_da_node_order(xfs_dabuf_t *node1_bp, xfs_dabuf_t *node2_bp)
* Pick up the last hashvalue from an intermediate node.
*/
STATIC uint
-xfs_da_node_lasthash(xfs_dabuf_t *bp, int *count)
+xfs_da_node_lasthash(
+ struct xfs_buf *bp,
+ int *count)
{
xfs_da_intnode_t *node;
- node = bp->data;
- ASSERT(be16_to_cpu(node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
+ node = bp->b_addr;
+ ASSERT(node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
if (count)
*count = be16_to_cpu(node->hdr.count);
if (!node->hdr.count)
@@ -1291,7 +1405,7 @@ xfs_da_blk_unlink(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
{
xfs_da_blkinfo_t *drop_info, *save_info, *tmp_info;
xfs_da_args_t *args;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
int error;
/*
@@ -1299,8 +1413,8 @@ xfs_da_blk_unlink(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
*/
args = state->args;
ASSERT(args != NULL);
- save_info = save_blk->bp->data;
- drop_info = drop_blk->bp->data;
+ save_info = save_blk->bp->b_addr;
+ drop_info = drop_blk->bp->b_addr;
ASSERT(save_blk->magic == XFS_DA_NODE_MAGIC ||
save_blk->magic == XFS_DIR2_LEAFN_MAGIC ||
save_blk->magic == XFS_ATTR_LEAF_MAGIC);
@@ -1316,42 +1430,42 @@ xfs_da_blk_unlink(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
* Unlink the leaf block from the doubly linked chain of leaves.
*/
if (be32_to_cpu(save_info->back) == drop_blk->blkno) {
+ trace_xfs_da_unlink_back(args);
save_info->back = drop_info->back;
if (drop_info->back) {
- error = xfs_da_read_buf(args->trans, args->dp,
+ error = xfs_da_node_read(args->trans, args->dp,
be32_to_cpu(drop_info->back),
-1, &bp, args->whichfork);
if (error)
return(error);
ASSERT(bp != NULL);
- tmp_info = bp->data;
+ tmp_info = bp->b_addr;
ASSERT(tmp_info->magic == save_info->magic);
ASSERT(be32_to_cpu(tmp_info->forw) == drop_blk->blkno);
tmp_info->forw = cpu_to_be32(save_blk->blkno);
- xfs_da_log_buf(args->trans, bp, 0,
+ xfs_trans_log_buf(args->trans, bp, 0,
sizeof(*tmp_info) - 1);
- xfs_da_buf_done(bp);
}
} else {
+ trace_xfs_da_unlink_forward(args);
save_info->forw = drop_info->forw;
if (drop_info->forw) {
- error = xfs_da_read_buf(args->trans, args->dp,
+ error = xfs_da_node_read(args->trans, args->dp,
be32_to_cpu(drop_info->forw),
-1, &bp, args->whichfork);
if (error)
return(error);
ASSERT(bp != NULL);
- tmp_info = bp->data;
+ tmp_info = bp->b_addr;
ASSERT(tmp_info->magic == save_info->magic);
ASSERT(be32_to_cpu(tmp_info->back) == drop_blk->blkno);
tmp_info->back = cpu_to_be32(save_blk->blkno);
- xfs_da_log_buf(args->trans, bp, 0,
+ xfs_trans_log_buf(args->trans, bp, 0,
sizeof(*tmp_info) - 1);
- xfs_da_buf_done(bp);
}
}
- xfs_da_log_buf(args->trans, save_blk->bp, 0, sizeof(*save_info) - 1);
+ xfs_trans_log_buf(args->trans, save_blk->bp, 0, sizeof(*save_info) - 1);
return(0);
}
@@ -1374,6 +1488,8 @@ xfs_da_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path,
xfs_dablk_t blkno=0;
int level, error;
+ trace_xfs_da_path_shift(state->args);
+
/*
* Roll up the Btree looking for the first block where our
* current index is not at the edge of the block. Note that
@@ -1386,8 +1502,8 @@ xfs_da_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path,
level = (path->active-1) - 1; /* skip bottom layer in path */
for (blk = &path->blk[level]; level >= 0; blk--, level--) {
ASSERT(blk->bp != NULL);
- node = blk->bp->data;
- ASSERT(be16_to_cpu(node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
+ node = blk->bp->b_addr;
+ ASSERT(node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
if (forward && (blk->index < be16_to_cpu(node->hdr.count)-1)) {
blk->index++;
blkno = be32_to_cpu(node->btree[blk->index].before);
@@ -1414,21 +1530,21 @@ xfs_da_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path,
* (if it's dirty, trans won't actually let go)
*/
if (release)
- xfs_da_brelse(args->trans, blk->bp);
+ xfs_trans_brelse(args->trans, blk->bp);
/*
* Read the next child block.
*/
blk->blkno = blkno;
- error = xfs_da_read_buf(args->trans, args->dp, blkno, -1,
- &blk->bp, args->whichfork);
+ error = xfs_da_node_read(args->trans, args->dp, blkno, -1,
+ &blk->bp, args->whichfork);
if (error)
return(error);
ASSERT(blk->bp != NULL);
- info = blk->bp->data;
- ASSERT(be16_to_cpu(info->magic) == XFS_DA_NODE_MAGIC ||
- be16_to_cpu(info->magic) == XFS_DIR2_LEAFN_MAGIC ||
- be16_to_cpu(info->magic) == XFS_ATTR_LEAF_MAGIC);
+ info = blk->bp->b_addr;
+ ASSERT(info->magic == cpu_to_be16(XFS_DA_NODE_MAGIC) ||
+ info->magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
+ info->magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
blk->magic = be16_to_cpu(info->magic);
if (blk->magic == XFS_DA_NODE_MAGIC) {
node = (xfs_da_intnode_t *)info;
@@ -1521,79 +1637,60 @@ const struct xfs_nameops xfs_default_nameops = {
.compname = xfs_da_compname
};
-/*
- * Add a block to the btree ahead of the file.
- * Return the new block number to the caller.
- */
int
-xfs_da_grow_inode(xfs_da_args_t *args, xfs_dablk_t *new_blkno)
+xfs_da_grow_inode_int(
+ struct xfs_da_args *args,
+ xfs_fileoff_t *bno,
+ int count)
{
- xfs_fileoff_t bno, b;
- xfs_bmbt_irec_t map;
- xfs_bmbt_irec_t *mapp;
- xfs_inode_t *dp;
- int nmap, error, w, count, c, got, i, mapi;
- xfs_trans_t *tp;
- xfs_mount_t *mp;
- xfs_drfsbno_t nblks;
-
- dp = args->dp;
- mp = dp->i_mount;
- w = args->whichfork;
- tp = args->trans;
- nblks = dp->i_d.di_nblocks;
+ struct xfs_trans *tp = args->trans;
+ struct xfs_inode *dp = args->dp;
+ int w = args->whichfork;
+ xfs_drfsbno_t nblks = dp->i_d.di_nblocks;
+ struct xfs_bmbt_irec map, *mapp;
+ int nmap, error, got, i, mapi;
/*
- * For new directories adjust the file offset and block count.
- */
- if (w == XFS_DATA_FORK) {
- bno = mp->m_dirleafblk;
- count = mp->m_dirblkfsbs;
- } else {
- bno = 0;
- count = 1;
- }
- /*
* Find a spot in the file space to put the new block.
*/
- if ((error = xfs_bmap_first_unused(tp, dp, count, &bno, w)))
+ error = xfs_bmap_first_unused(tp, dp, count, bno, w);
+ if (error)
return error;
- if (w == XFS_DATA_FORK)
- ASSERT(bno >= mp->m_dirleafblk && bno < mp->m_dirfreeblk);
+
/*
* Try mapping it in one filesystem block.
*/
nmap = 1;
ASSERT(args->firstblock != NULL);
- if ((error = xfs_bmapi(tp, dp, bno, count,
- xfs_bmapi_aflag(w)|XFS_BMAPI_WRITE|XFS_BMAPI_METADATA|
- XFS_BMAPI_CONTIG,
+ error = xfs_bmapi_write(tp, dp, *bno, count,
+ xfs_bmapi_aflag(w)|XFS_BMAPI_METADATA|XFS_BMAPI_CONTIG,
args->firstblock, args->total, &map, &nmap,
- args->flist))) {
+ args->flist);
+ if (error)
return error;
- }
+
ASSERT(nmap <= 1);
if (nmap == 1) {
mapp = ↦
mapi = 1;
- }
- /*
- * If we didn't get it and the block might work if fragmented,
- * try without the CONTIG flag. Loop until we get it all.
- */
- else if (nmap == 0 && count > 1) {
+ } else if (nmap == 0 && count > 1) {
+ xfs_fileoff_t b;
+ int c;
+
+ /*
+ * If we didn't get it and the block might work if fragmented,
+ * try without the CONTIG flag. Loop until we get it all.
+ */
mapp = kmem_alloc(sizeof(*mapp) * count, KM_SLEEP);
- for (b = bno, mapi = 0; b < bno + count; ) {
+ for (b = *bno, mapi = 0; b < *bno + count; ) {
nmap = MIN(XFS_BMAP_MAX_NMAP, count);
- c = (int)(bno + count - b);
- if ((error = xfs_bmapi(tp, dp, b, c,
- xfs_bmapi_aflag(w)|XFS_BMAPI_WRITE|
- XFS_BMAPI_METADATA,
+ c = (int)(*bno + count - b);
+ error = xfs_bmapi_write(tp, dp, b, c,
+ xfs_bmapi_aflag(w)|XFS_BMAPI_METADATA,
args->firstblock, args->total,
- &mapp[mapi], &nmap, args->flist))) {
- kmem_free(mapp);
- return error;
- }
+ &mapp[mapi], &nmap, args->flist);
+ if (error)
+ goto out_free_map;
if (nmap < 1)
break;
mapi += nmap;
@@ -1604,24 +1701,55 @@ xfs_da_grow_inode(xfs_da_args_t *args, xfs_dablk_t *new_blkno)
mapi = 0;
mapp = NULL;
}
+
/*
* Count the blocks we got, make sure it matches the total.
*/
for (i = 0, got = 0; i < mapi; i++)
got += mapp[i].br_blockcount;
- if (got != count || mapp[0].br_startoff != bno ||
+ if (got != count || mapp[0].br_startoff != *bno ||
mapp[mapi - 1].br_startoff + mapp[mapi - 1].br_blockcount !=
- bno + count) {
- if (mapp != &map)
- kmem_free(mapp);
- return XFS_ERROR(ENOSPC);
+ *bno + count) {
+ error = XFS_ERROR(ENOSPC);
+ goto out_free_map;
}
- if (mapp != &map)
- kmem_free(mapp);
+
/* account for newly allocated blocks in reserved blocks total */
args->total -= dp->i_d.di_nblocks - nblks;
- *new_blkno = (xfs_dablk_t)bno;
- return 0;
+
+out_free_map:
+ if (mapp != &map)
+ kmem_free(mapp);
+ return error;
+}
+
+/*
+ * Add a block to the btree ahead of the file.
+ * Return the new block number to the caller.
+ */
+int
+xfs_da_grow_inode(
+ struct xfs_da_args *args,
+ xfs_dablk_t *new_blkno)
+{
+ xfs_fileoff_t bno;
+ int count;
+ int error;
+
+ trace_xfs_da_grow_inode(args);
+
+ if (args->whichfork == XFS_DATA_FORK) {
+ bno = args->dp->i_mount->m_dirleafblk;
+ count = args->dp->i_mount->m_dirblkfsbs;
+ } else {
+ bno = 0;
+ count = 1;
+ }
+
+ error = xfs_da_grow_inode_int(args, &bno, count);
+ if (!error)
+ *new_blkno = (xfs_dablk_t)bno;
+ return error;
}
/*
@@ -1633,11 +1761,13 @@ xfs_da_grow_inode(xfs_da_args_t *args, xfs_dablk_t *new_blkno)
* a bmap btree split to do that.
*/
STATIC int
-xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
- xfs_dabuf_t **dead_bufp)
+xfs_da_swap_lastblock(
+ xfs_da_args_t *args,
+ xfs_dablk_t *dead_blknop,
+ struct xfs_buf **dead_bufp)
{
xfs_dablk_t dead_blkno, last_blkno, sib_blkno, par_blkno;
- xfs_dabuf_t *dead_buf, *last_buf, *sib_buf, *par_buf;
+ struct xfs_buf *dead_buf, *last_buf, *sib_buf, *par_buf;
xfs_fileoff_t lastoff;
xfs_inode_t *ip;
xfs_trans_t *tp;
@@ -1648,6 +1778,8 @@ xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
xfs_dir2_leaf_t *dead_leaf2;
xfs_dahash_t dead_hash;
+ trace_xfs_da_swap_lastblock(args);
+
dead_buf = *dead_bufp;
dead_blkno = *dead_blknop;
tp = args->trans;
@@ -1668,23 +1800,24 @@ xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
* Read the last block in the btree space.
*/
last_blkno = (xfs_dablk_t)lastoff - mp->m_dirblkfsbs;
- if ((error = xfs_da_read_buf(tp, ip, last_blkno, -1, &last_buf, w)))
+ error = xfs_da_node_read(tp, ip, last_blkno, -1, &last_buf, w);
+ if (error)
return error;
/*
* Copy the last block into the dead buffer and log it.
*/
- memcpy(dead_buf->data, last_buf->data, mp->m_dirblksize);
- xfs_da_log_buf(tp, dead_buf, 0, mp->m_dirblksize - 1);
- dead_info = dead_buf->data;
+ memcpy(dead_buf->b_addr, last_buf->b_addr, mp->m_dirblksize);
+ xfs_trans_log_buf(tp, dead_buf, 0, mp->m_dirblksize - 1);
+ dead_info = dead_buf->b_addr;
/*
* Get values from the moved block.
*/
- if (be16_to_cpu(dead_info->magic) == XFS_DIR2_LEAFN_MAGIC) {
+ if (dead_info->magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC)) {
dead_leaf2 = (xfs_dir2_leaf_t *)dead_info;
dead_level = 0;
dead_hash = be32_to_cpu(dead_leaf2->ents[be16_to_cpu(dead_leaf2->hdr.count) - 1].hashval);
} else {
- ASSERT(be16_to_cpu(dead_info->magic) == XFS_DA_NODE_MAGIC);
+ ASSERT(dead_info->magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
dead_node = (xfs_da_intnode_t *)dead_info;
dead_level = be16_to_cpu(dead_node->hdr.level);
dead_hash = be32_to_cpu(dead_node->btree[be16_to_cpu(dead_node->hdr.count) - 1].hashval);
@@ -1694,9 +1827,10 @@ xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
* If the moved block has a left sibling, fix up the pointers.
*/
if ((sib_blkno = be32_to_cpu(dead_info->back))) {
- if ((error = xfs_da_read_buf(tp, ip, sib_blkno, -1, &sib_buf, w)))
+ error = xfs_da_node_read(tp, ip, sib_blkno, -1, &sib_buf, w);
+ if (error)
goto done;
- sib_info = sib_buf->data;
+ sib_info = sib_buf->b_addr;
if (unlikely(
be32_to_cpu(sib_info->forw) != last_blkno ||
sib_info->magic != dead_info->magic)) {
@@ -1706,19 +1840,19 @@ xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
goto done;
}
sib_info->forw = cpu_to_be32(dead_blkno);
- xfs_da_log_buf(tp, sib_buf,
+ xfs_trans_log_buf(tp, sib_buf,
XFS_DA_LOGRANGE(sib_info, &sib_info->forw,
sizeof(sib_info->forw)));
- xfs_da_buf_done(sib_buf);
sib_buf = NULL;
}
/*
* If the moved block has a right sibling, fix up the pointers.
*/
if ((sib_blkno = be32_to_cpu(dead_info->forw))) {
- if ((error = xfs_da_read_buf(tp, ip, sib_blkno, -1, &sib_buf, w)))
+ error = xfs_da_node_read(tp, ip, sib_blkno, -1, &sib_buf, w);
+ if (error)
goto done;
- sib_info = sib_buf->data;
+ sib_info = sib_buf->b_addr;
if (unlikely(
be32_to_cpu(sib_info->back) != last_blkno ||
sib_info->magic != dead_info->magic)) {
@@ -1728,10 +1862,9 @@ xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
goto done;
}
sib_info->back = cpu_to_be32(dead_blkno);
- xfs_da_log_buf(tp, sib_buf,
+ xfs_trans_log_buf(tp, sib_buf,
XFS_DA_LOGRANGE(sib_info, &sib_info->back,
sizeof(sib_info->back)));
- xfs_da_buf_done(sib_buf);
sib_buf = NULL;
}
par_blkno = mp->m_dirleafblk;
@@ -1740,11 +1873,12 @@ xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
* Walk down the tree looking for the parent of the moved block.
*/
for (;;) {
- if ((error = xfs_da_read_buf(tp, ip, par_blkno, -1, &par_buf, w)))
+ error = xfs_da_node_read(tp, ip, par_blkno, -1, &par_buf, w);
+ if (error)
goto done;
- par_node = par_buf->data;
- if (unlikely(
- be16_to_cpu(par_node->hdr.info.magic) != XFS_DA_NODE_MAGIC ||
+ par_node = par_buf->b_addr;
+ if (unlikely(par_node->hdr.info.magic !=
+ cpu_to_be16(XFS_DA_NODE_MAGIC) ||
(level >= 0 && level != be16_to_cpu(par_node->hdr.level) + 1))) {
XFS_ERROR_REPORT("xfs_da_swap_lastblock(4)",
XFS_ERRLEVEL_LOW, mp);
@@ -1766,7 +1900,7 @@ xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
par_blkno = be32_to_cpu(par_node->btree[entno].before);
if (level == dead_level + 1)
break;
- xfs_da_brelse(tp, par_buf);
+ xfs_trans_brelse(tp, par_buf);
par_buf = NULL;
}
/*
@@ -1782,7 +1916,7 @@ xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
if (entno < be16_to_cpu(par_node->hdr.count))
break;
par_blkno = be32_to_cpu(par_node->hdr.info.forw);
- xfs_da_brelse(tp, par_buf);
+ xfs_trans_brelse(tp, par_buf);
par_buf = NULL;
if (unlikely(par_blkno == 0)) {
XFS_ERROR_REPORT("xfs_da_swap_lastblock(6)",
@@ -1790,12 +1924,13 @@ xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
error = XFS_ERROR(EFSCORRUPTED);
goto done;
}
- if ((error = xfs_da_read_buf(tp, ip, par_blkno, -1, &par_buf, w)))
+ error = xfs_da_node_read(tp, ip, par_blkno, -1, &par_buf, w);
+ if (error)
goto done;
- par_node = par_buf->data;
+ par_node = par_buf->b_addr;
if (unlikely(
be16_to_cpu(par_node->hdr.level) != level ||
- be16_to_cpu(par_node->hdr.info.magic) != XFS_DA_NODE_MAGIC)) {
+ par_node->hdr.info.magic != cpu_to_be16(XFS_DA_NODE_MAGIC))) {
XFS_ERROR_REPORT("xfs_da_swap_lastblock(7)",
XFS_ERRLEVEL_LOW, mp);
error = XFS_ERROR(EFSCORRUPTED);
@@ -1807,20 +1942,18 @@ xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
* Update the parent entry pointing to the moved block.
*/
par_node->btree[entno].before = cpu_to_be32(dead_blkno);
- xfs_da_log_buf(tp, par_buf,
+ xfs_trans_log_buf(tp, par_buf,
XFS_DA_LOGRANGE(par_node, &par_node->btree[entno].before,
sizeof(par_node->btree[entno].before)));
- xfs_da_buf_done(par_buf);
- xfs_da_buf_done(dead_buf);
*dead_blknop = last_blkno;
*dead_bufp = last_buf;
return 0;
done:
if (par_buf)
- xfs_da_brelse(tp, par_buf);
+ xfs_trans_brelse(tp, par_buf);
if (sib_buf)
- xfs_da_brelse(tp, sib_buf);
- xfs_da_brelse(tp, last_buf);
+ xfs_trans_brelse(tp, sib_buf);
+ xfs_trans_brelse(tp, last_buf);
return error;
}
@@ -1828,14 +1961,18 @@ done:
* Remove a btree block from a directory or attribute.
*/
int
-xfs_da_shrink_inode(xfs_da_args_t *args, xfs_dablk_t dead_blkno,
- xfs_dabuf_t *dead_buf)
+xfs_da_shrink_inode(
+ xfs_da_args_t *args,
+ xfs_dablk_t dead_blkno,
+ struct xfs_buf *dead_buf)
{
xfs_inode_t *dp;
int done, error, w, count;
xfs_trans_t *tp;
xfs_mount_t *mp;
+ trace_xfs_da_shrink_inode(args);
+
dp = args->dp;
w = args->whichfork;
tp = args->trans;
@@ -1862,7 +1999,7 @@ xfs_da_shrink_inode(xfs_da_args_t *args, xfs_dablk_t dead_blkno,
break;
}
}
- xfs_da_binval(tp, dead_buf);
+ xfs_trans_binval(tp, dead_buf);
return error;
}
@@ -1894,36 +2031,75 @@ xfs_da_map_covers_blocks(
}
/*
- * Make a dabuf.
- * Used for get_buf, read_buf, read_bufr, and reada_buf.
+ * Convert a struct xfs_bmbt_irec to a struct xfs_buf_map.
+ *
+ * For the single map case, it is assumed that the caller has provided a pointer
+ * to a valid xfs_buf_map. For the multiple map case, this function will
+ * allocate the xfs_buf_map to hold all the maps and replace the caller's single
+ * map pointer with the allocated map.
*/
-int
-xfs_da_do_buf(
- xfs_trans_t *trans,
- xfs_inode_t *dp,
- xfs_dablk_t bno,
- xfs_daddr_t *mappedbnop,
- xfs_dabuf_t **bpp,
- int whichfork,
- int caller,
- inst_t *ra)
+static int
+xfs_buf_map_from_irec(
+ struct xfs_mount *mp,
+ struct xfs_buf_map **mapp,
+ unsigned int *nmaps,
+ struct xfs_bmbt_irec *irecs,
+ unsigned int nirecs)
{
- xfs_buf_t *bp = NULL;
- xfs_buf_t **bplist;
- int error=0;
- int i;
- xfs_bmbt_irec_t map;
- xfs_bmbt_irec_t *mapp;
- xfs_daddr_t mappedbno;
- xfs_mount_t *mp;
- int nbplist=0;
- int nfsb;
- int nmap;
- xfs_dabuf_t *rbp;
+ struct xfs_buf_map *map;
+ int i;
+
+ ASSERT(*nmaps == 1);
+ ASSERT(nirecs >= 1);
+
+ if (nirecs > 1) {
+ map = kmem_zalloc(nirecs * sizeof(struct xfs_buf_map), KM_SLEEP);
+ if (!map)
+ return ENOMEM;
+ *mapp = map;
+ }
+
+ *nmaps = nirecs;
+ map = *mapp;
+ for (i = 0; i < *nmaps; i++) {
+ ASSERT(irecs[i].br_startblock != DELAYSTARTBLOCK &&
+ irecs[i].br_startblock != HOLESTARTBLOCK);
+ map[i].bm_bn = XFS_FSB_TO_DADDR(mp, irecs[i].br_startblock);
+ map[i].bm_len = XFS_FSB_TO_BB(mp, irecs[i].br_blockcount);
+ }
+ return 0;
+}
+
+/*
+ * Map the block we are given ready for reading. There are three possible return
+ * values:
+ * -1 - will be returned if we land in a hole and mappedbno == -2 so the
+ * caller knows not to execute a subsequent read.
+ * 0 - if we mapped the block successfully
+ * >0 - positive error number if there was an error.
+ */
+static int
+xfs_dabuf_map(
+ struct xfs_trans *trans,
+ struct xfs_inode *dp,
+ xfs_dablk_t bno,
+ xfs_daddr_t mappedbno,
+ int whichfork,
+ struct xfs_buf_map **map,
+ int *nmaps)
+{
+ struct xfs_mount *mp = dp->i_mount;
+ int nfsb;
+ int error = 0;
+ struct xfs_bmbt_irec irec;
+ struct xfs_bmbt_irec *irecs = &irec;
+ int nirecs;
+
+ ASSERT(map && *map);
+ ASSERT(*nmaps == 1);
- mp = dp->i_mount;
nfsb = (whichfork == XFS_DATA_FORK) ? mp->m_dirblkfsbs : 1;
- mappedbno = *mappedbnop;
+
/*
* Caller doesn't have a mapping. -2 means don't complain
* if we land in a hole.
@@ -1932,139 +2108,154 @@ xfs_da_do_buf(
/*
* Optimize the one-block case.
*/
- if (nfsb == 1) {
- xfs_fsblock_t fsb;
+ if (nfsb != 1)
+ irecs = kmem_zalloc(sizeof(irec) * nfsb, KM_SLEEP);
- if ((error =
- xfs_bmapi_single(trans, dp, whichfork, &fsb,
- (xfs_fileoff_t)bno))) {
- return error;
- }
- mapp = ↦
- if (fsb == NULLFSBLOCK) {
- nmap = 0;
- } else {
- map.br_startblock = fsb;
- map.br_startoff = (xfs_fileoff_t)bno;
- map.br_blockcount = 1;
- nmap = 1;
- }
- } else {
- mapp = kmem_alloc(sizeof(*mapp) * nfsb, KM_SLEEP);
- nmap = nfsb;
- if ((error = xfs_bmapi(trans, dp, (xfs_fileoff_t)bno,
- nfsb,
- XFS_BMAPI_METADATA |
- xfs_bmapi_aflag(whichfork),
- NULL, 0, mapp, &nmap, NULL)))
- goto exit0;
- }
+ nirecs = nfsb;
+ error = xfs_bmapi_read(dp, (xfs_fileoff_t)bno, nfsb, irecs,
+ &nirecs, xfs_bmapi_aflag(whichfork));
+ if (error)
+ goto out;
} else {
- map.br_startblock = XFS_DADDR_TO_FSB(mp, mappedbno);
- map.br_startoff = (xfs_fileoff_t)bno;
- map.br_blockcount = nfsb;
- mapp = ↦
- nmap = 1;
+ irecs->br_startblock = XFS_DADDR_TO_FSB(mp, mappedbno);
+ irecs->br_startoff = (xfs_fileoff_t)bno;
+ irecs->br_blockcount = nfsb;
+ irecs->br_state = 0;
+ nirecs = 1;
}
- if (!xfs_da_map_covers_blocks(nmap, mapp, bno, nfsb)) {
- error = mappedbno == -2 ? 0 : XFS_ERROR(EFSCORRUPTED);
+
+ if (!xfs_da_map_covers_blocks(nirecs, irecs, bno, nfsb)) {
+ error = mappedbno == -2 ? -1 : XFS_ERROR(EFSCORRUPTED);
if (unlikely(error == EFSCORRUPTED)) {
if (xfs_error_level >= XFS_ERRLEVEL_LOW) {
- cmn_err(CE_ALERT, "xfs_da_do_buf: bno %lld\n",
- (long long)bno);
- cmn_err(CE_ALERT, "dir: inode %lld\n",
+ int i;
+ xfs_alert(mp, "%s: bno %lld dir: inode %lld",
+ __func__, (long long)bno,
(long long)dp->i_ino);
- for (i = 0; i < nmap; i++) {
- cmn_err(CE_ALERT,
- "[%02d] br_startoff %lld br_startblock %lld br_blockcount %lld br_state %d\n",
+ for (i = 0; i < *nmaps; i++) {
+ xfs_alert(mp,
+"[%02d] br_startoff %lld br_startblock %lld br_blockcount %lld br_state %d",
i,
- (long long)mapp[i].br_startoff,
- (long long)mapp[i].br_startblock,
- (long long)mapp[i].br_blockcount,
- mapp[i].br_state);
+ (long long)irecs[i].br_startoff,
+ (long long)irecs[i].br_startblock,
+ (long long)irecs[i].br_blockcount,
+ irecs[i].br_state);
}
}
XFS_ERROR_REPORT("xfs_da_do_buf(1)",
XFS_ERRLEVEL_LOW, mp);
}
- goto exit0;
+ goto out;
}
- if (caller != 3 && nmap > 1) {
- bplist = kmem_alloc(sizeof(*bplist) * nmap, KM_SLEEP);
- nbplist = 0;
- } else
- bplist = NULL;
- /*
- * Turn the mapping(s) into buffer(s).
- */
- for (i = 0; i < nmap; i++) {
- int nmapped;
-
- mappedbno = XFS_FSB_TO_DADDR(mp, mapp[i].br_startblock);
- if (i == 0)
- *mappedbnop = mappedbno;
- nmapped = (int)XFS_FSB_TO_BB(mp, mapp[i].br_blockcount);
- switch (caller) {
- case 0:
- bp = xfs_trans_get_buf(trans, mp->m_ddev_targp,
- mappedbno, nmapped, 0);
- error = bp ? XFS_BUF_GETERROR(bp) : XFS_ERROR(EIO);
- break;
- case 1:
- case 2:
- bp = NULL;
- error = xfs_trans_read_buf(mp, trans, mp->m_ddev_targp,
- mappedbno, nmapped, 0, &bp);
- break;
- case 3:
- xfs_buf_readahead(mp->m_ddev_targp, mappedbno, nmapped);
+ error = xfs_buf_map_from_irec(mp, map, nmaps, irecs, nirecs);
+out:
+ if (irecs != &irec)
+ kmem_free(irecs);
+ return error;
+}
+
+/*
+ * Get a buffer for the dir/attr block.
+ */
+int
+xfs_da_get_buf(
+ struct xfs_trans *trans,
+ struct xfs_inode *dp,
+ xfs_dablk_t bno,
+ xfs_daddr_t mappedbno,
+ struct xfs_buf **bpp,
+ int whichfork)
+{
+ struct xfs_buf *bp;
+ struct xfs_buf_map map;
+ struct xfs_buf_map *mapp;
+ int nmap;
+ int error;
+
+ *bpp = NULL;
+ mapp = ↦
+ nmap = 1;
+ error = xfs_dabuf_map(trans, dp, bno, mappedbno, whichfork,
+ &mapp, &nmap);
+ if (error) {
+ /* mapping a hole is not an error, but we don't continue */
+ if (error == -1)
error = 0;
- bp = NULL;
- break;
- }
- if (error) {
- if (bp)
- xfs_trans_brelse(trans, bp);
- goto exit1;
- }
- if (!bp)
- continue;
- if (caller == 1) {
- if (whichfork == XFS_ATTR_FORK) {
- XFS_BUF_SET_VTYPE_REF(bp, B_FS_ATTR_BTREE,
- XFS_ATTR_BTREE_REF);
- } else {
- XFS_BUF_SET_VTYPE_REF(bp, B_FS_DIR_BTREE,
- XFS_DIR_BTREE_REF);
- }
- }
- if (bplist) {
- bplist[nbplist++] = bp;
- }
+ goto out_free;
}
- /*
- * Build a dabuf structure.
- */
- if (bplist) {
- rbp = xfs_da_buf_make(nbplist, bplist, ra);
- } else if (bp)
- rbp = xfs_da_buf_make(1, &bp, ra);
+
+ bp = xfs_trans_get_buf_map(trans, dp->i_mount->m_ddev_targp,
+ mapp, nmap, 0);
+ error = bp ? bp->b_error : XFS_ERROR(EIO);
+ if (error) {
+ xfs_trans_brelse(trans, bp);
+ goto out_free;
+ }
+
+ *bpp = bp;
+
+out_free:
+ if (mapp != &map)
+ kmem_free(mapp);
+
+ return error;
+}
+
+/*
+ * Get a buffer for the dir/attr block, fill in the contents.
+ */
+int
+xfs_da_read_buf(
+ struct xfs_trans *trans,
+ struct xfs_inode *dp,
+ xfs_dablk_t bno,
+ xfs_daddr_t mappedbno,
+ struct xfs_buf **bpp,
+ int whichfork,
+ const struct xfs_buf_ops *ops)
+{
+ struct xfs_buf *bp;
+ struct xfs_buf_map map;
+ struct xfs_buf_map *mapp;
+ int nmap;
+ int error;
+
+ *bpp = NULL;
+ mapp = ↦
+ nmap = 1;
+ error = xfs_dabuf_map(trans, dp, bno, mappedbno, whichfork,
+ &mapp, &nmap);
+ if (error) {
+ /* mapping a hole is not an error, but we don't continue */
+ if (error == -1)
+ error = 0;
+ goto out_free;
+ }
+
+ error = xfs_trans_read_buf_map(dp->i_mount, trans,
+ dp->i_mount->m_ddev_targp,
+ mapp, nmap, 0, &bp, ops);
+ if (error)
+ goto out_free;
+
+ if (whichfork == XFS_ATTR_FORK)
+ xfs_buf_set_ref(bp, XFS_ATTR_BTREE_REF);
else
- rbp = NULL;
+ xfs_buf_set_ref(bp, XFS_DIR_BTREE_REF);
+
/*
- * For read_buf, check the magic number.
+ * This verification code will be moved to a CRC verification callback
+ * function so just leave it here unchanged until then.
*/
- if (caller == 1) {
- xfs_dir2_data_t *data;
- xfs_dir2_free_t *free;
- xfs_da_blkinfo_t *info;
+ {
+ xfs_dir2_data_hdr_t *hdr = bp->b_addr;
+ xfs_dir2_free_t *free = bp->b_addr;
+ xfs_da_blkinfo_t *info = bp->b_addr;
uint magic, magic1;
+ struct xfs_mount *mp = dp->i_mount;
- info = rbp->data;
- data = rbp->data;
- free = rbp->data;
magic = be16_to_cpu(info->magic);
- magic1 = be32_to_cpu(data->hdr.magic);
+ magic1 = be32_to_cpu(hdr->magic);
if (unlikely(
XFS_TEST_ERROR((magic != XFS_DA_NODE_MAGIC) &&
(magic != XFS_ATTR_LEAF_MAGIC) &&
@@ -2072,71 +2263,23 @@ xfs_da_do_buf(
(magic != XFS_DIR2_LEAFN_MAGIC) &&
(magic1 != XFS_DIR2_BLOCK_MAGIC) &&
(magic1 != XFS_DIR2_DATA_MAGIC) &&
- (be32_to_cpu(free->hdr.magic) != XFS_DIR2_FREE_MAGIC),
+ (free->hdr.magic != cpu_to_be32(XFS_DIR2_FREE_MAGIC)),
mp, XFS_ERRTAG_DA_READ_BUF,
XFS_RANDOM_DA_READ_BUF))) {
- trace_xfs_da_btree_corrupt(rbp->bps[0], _RET_IP_);
+ trace_xfs_da_btree_corrupt(bp, _RET_IP_);
XFS_CORRUPTION_ERROR("xfs_da_do_buf(2)",
XFS_ERRLEVEL_LOW, mp, info);
error = XFS_ERROR(EFSCORRUPTED);
- xfs_da_brelse(trans, rbp);
- nbplist = 0;
- goto exit1;
+ xfs_trans_brelse(trans, bp);
+ goto out_free;
}
}
- if (bplist) {
- kmem_free(bplist);
- }
- if (mapp != &map) {
- kmem_free(mapp);
- }
- if (bpp)
- *bpp = rbp;
- return 0;
-exit1:
- if (bplist) {
- for (i = 0; i < nbplist; i++)
- xfs_trans_brelse(trans, bplist[i]);
- kmem_free(bplist);
- }
-exit0:
+ *bpp = bp;
+out_free:
if (mapp != &map)
kmem_free(mapp);
- if (bpp)
- *bpp = NULL;
- return error;
-}
-/*
- * Get a buffer for the dir/attr block.
- */
-int
-xfs_da_get_buf(
- xfs_trans_t *trans,
- xfs_inode_t *dp,
- xfs_dablk_t bno,
- xfs_daddr_t mappedbno,
- xfs_dabuf_t **bpp,
- int whichfork)
-{
- return xfs_da_do_buf(trans, dp, bno, &mappedbno, bpp, whichfork, 0,
- (inst_t *)__return_address);
-}
-
-/*
- * Get a buffer for the dir/attr block, fill in the contents.
- */
-int
-xfs_da_read_buf(
- xfs_trans_t *trans,
- xfs_inode_t *dp,
- xfs_dablk_t bno,
- xfs_daddr_t mappedbno,
- xfs_dabuf_t **bpp,
- int whichfork)
-{
- return xfs_da_do_buf(trans, dp, bno, &mappedbno, bpp, whichfork, 1,
- (inst_t *)__return_address);
+ return error;
}
/*
@@ -2144,23 +2287,42 @@ xfs_da_read_buf(
*/
xfs_daddr_t
xfs_da_reada_buf(
- xfs_trans_t *trans,
- xfs_inode_t *dp,
- xfs_dablk_t bno,
- int whichfork)
+ struct xfs_trans *trans,
+ struct xfs_inode *dp,
+ xfs_dablk_t bno,
+ xfs_daddr_t mappedbno,
+ int whichfork,
+ const struct xfs_buf_ops *ops)
{
- xfs_daddr_t rval;
+ struct xfs_buf_map map;
+ struct xfs_buf_map *mapp;
+ int nmap;
+ int error;
- rval = -1;
- if (xfs_da_do_buf(trans, dp, bno, &rval, NULL, whichfork, 3,
- (inst_t *)__return_address))
+ mapp = ↦
+ nmap = 1;
+ error = xfs_dabuf_map(trans, dp, bno, mappedbno, whichfork,
+ &mapp, &nmap);
+ if (error) {
+ /* mapping a hole is not an error, but we don't continue */
+ if (error == -1)
+ error = 0;
+ goto out_free;
+ }
+
+ mappedbno = mapp[0].bm_bn;
+ xfs_buf_readahead_map(dp->i_mount->m_ddev_targp, mapp, nmap, ops);
+
+out_free:
+ if (mapp != &map)
+ kmem_free(mapp);
+
+ if (error)
return -1;
- else
- return rval;
+ return mappedbno;
}
kmem_zone_t *xfs_da_state_zone; /* anchor for state struct zone */
-kmem_zone_t *xfs_dabuf_zone; /* dabuf zone */
/*
* Allocate a dir-state structure.
@@ -2180,13 +2342,8 @@ xfs_da_state_kill_altpath(xfs_da_state_t *state)
{
int i;
- for (i = 0; i < state->altpath.active; i++) {
- if (state->altpath.blk[i].bp) {
- if (state->altpath.blk[i].bp != state->path.blk[i].bp)
- xfs_da_buf_done(state->altpath.blk[i].bp);
- state->altpath.blk[i].bp = NULL;
- }
- }
+ for (i = 0; i < state->altpath.active; i++)
+ state->altpath.blk[i].bp = NULL;
state->altpath.active = 0;
}
@@ -2196,244 +2353,9 @@ xfs_da_state_kill_altpath(xfs_da_state_t *state)
void
xfs_da_state_free(xfs_da_state_t *state)
{
- int i;
-
xfs_da_state_kill_altpath(state);
- for (i = 0; i < state->path.active; i++) {
- if (state->path.blk[i].bp)
- xfs_da_buf_done(state->path.blk[i].bp);
- }
- if (state->extravalid && state->extrablk.bp)
- xfs_da_buf_done(state->extrablk.bp);
#ifdef DEBUG
memset((char *)state, 0, sizeof(*state));
#endif /* DEBUG */
kmem_zone_free(xfs_da_state_zone, state);
}
-
-#ifdef XFS_DABUF_DEBUG
-xfs_dabuf_t *xfs_dabuf_global_list;
-static DEFINE_SPINLOCK(xfs_dabuf_global_lock);
-#endif
-
-/*
- * Create a dabuf.
- */
-/* ARGSUSED */
-STATIC xfs_dabuf_t *
-xfs_da_buf_make(int nbuf, xfs_buf_t **bps, inst_t *ra)
-{
- xfs_buf_t *bp;
- xfs_dabuf_t *dabuf;
- int i;
- int off;
-
- if (nbuf == 1)
- dabuf = kmem_zone_alloc(xfs_dabuf_zone, KM_NOFS);
- else
- dabuf = kmem_alloc(XFS_DA_BUF_SIZE(nbuf), KM_NOFS);
- dabuf->dirty = 0;
-#ifdef XFS_DABUF_DEBUG
- dabuf->ra = ra;
- dabuf->target = XFS_BUF_TARGET(bps[0]);
- dabuf->blkno = XFS_BUF_ADDR(bps[0]);
-#endif
- if (nbuf == 1) {
- dabuf->nbuf = 1;
- bp = bps[0];
- dabuf->bbcount = (short)BTOBB(XFS_BUF_COUNT(bp));
- dabuf->data = XFS_BUF_PTR(bp);
- dabuf->bps[0] = bp;
- } else {
- dabuf->nbuf = nbuf;
- for (i = 0, dabuf->bbcount = 0; i < nbuf; i++) {
- dabuf->bps[i] = bp = bps[i];
- dabuf->bbcount += BTOBB(XFS_BUF_COUNT(bp));
- }
- dabuf->data = kmem_alloc(BBTOB(dabuf->bbcount), KM_SLEEP);
- for (i = off = 0; i < nbuf; i++, off += XFS_BUF_COUNT(bp)) {
- bp = bps[i];
- memcpy((char *)dabuf->data + off, XFS_BUF_PTR(bp),
- XFS_BUF_COUNT(bp));
- }
- }
-#ifdef XFS_DABUF_DEBUG
- {
- xfs_dabuf_t *p;
-
- spin_lock(&xfs_dabuf_global_lock);
- for (p = xfs_dabuf_global_list; p; p = p->next) {
- ASSERT(p->blkno != dabuf->blkno ||
- p->target != dabuf->target);
- }
- dabuf->prev = NULL;
- if (xfs_dabuf_global_list)
- xfs_dabuf_global_list->prev = dabuf;
- dabuf->next = xfs_dabuf_global_list;
- xfs_dabuf_global_list = dabuf;
- spin_unlock(&xfs_dabuf_global_lock);
- }
-#endif
- return dabuf;
-}
-
-/*
- * Un-dirty a dabuf.
- */
-STATIC void
-xfs_da_buf_clean(xfs_dabuf_t *dabuf)
-{
- xfs_buf_t *bp;
- int i;
- int off;
-
- if (dabuf->dirty) {
- ASSERT(dabuf->nbuf > 1);
- dabuf->dirty = 0;
- for (i = off = 0; i < dabuf->nbuf;
- i++, off += XFS_BUF_COUNT(bp)) {
- bp = dabuf->bps[i];
- memcpy(XFS_BUF_PTR(bp), (char *)dabuf->data + off,
- XFS_BUF_COUNT(bp));
- }
- }
-}
-
-/*
- * Release a dabuf.
- */
-void
-xfs_da_buf_done(xfs_dabuf_t *dabuf)
-{
- ASSERT(dabuf);
- ASSERT(dabuf->nbuf && dabuf->data && dabuf->bbcount && dabuf->bps[0]);
- if (dabuf->dirty)
- xfs_da_buf_clean(dabuf);
- if (dabuf->nbuf > 1)
- kmem_free(dabuf->data);
-#ifdef XFS_DABUF_DEBUG
- {
- spin_lock(&xfs_dabuf_global_lock);
- if (dabuf->prev)
- dabuf->prev->next = dabuf->next;
- else
- xfs_dabuf_global_list = dabuf->next;
- if (dabuf->next)
- dabuf->next->prev = dabuf->prev;
- spin_unlock(&xfs_dabuf_global_lock);
- }
- memset(dabuf, 0, XFS_DA_BUF_SIZE(dabuf->nbuf));
-#endif
- if (dabuf->nbuf == 1)
- kmem_zone_free(xfs_dabuf_zone, dabuf);
- else
- kmem_free(dabuf);
-}
-
-/*
- * Log transaction from a dabuf.
- */
-void
-xfs_da_log_buf(xfs_trans_t *tp, xfs_dabuf_t *dabuf, uint first, uint last)
-{
- xfs_buf_t *bp;
- uint f;
- int i;
- uint l;
- int off;
-
- ASSERT(dabuf->nbuf && dabuf->data && dabuf->bbcount && dabuf->bps[0]);
- if (dabuf->nbuf == 1) {
- ASSERT(dabuf->data == (void *)XFS_BUF_PTR(dabuf->bps[0]));
- xfs_trans_log_buf(tp, dabuf->bps[0], first, last);
- return;
- }
- dabuf->dirty = 1;
- ASSERT(first <= last);
- for (i = off = 0; i < dabuf->nbuf; i++, off += XFS_BUF_COUNT(bp)) {
- bp = dabuf->bps[i];
- f = off;
- l = f + XFS_BUF_COUNT(bp) - 1;
- if (f < first)
- f = first;
- if (l > last)
- l = last;
- if (f <= l)
- xfs_trans_log_buf(tp, bp, f - off, l - off);
- /*
- * B_DONE is set by xfs_trans_log buf.
- * If we don't set it on a new buffer (get not read)
- * then if we don't put anything in the buffer it won't
- * be set, and at commit it it released into the cache,
- * and then a read will fail.
- */
- else if (!(XFS_BUF_ISDONE(bp)))
- XFS_BUF_DONE(bp);
- }
- ASSERT(last < off);
-}
-
-/*
- * Release dabuf from a transaction.
- * Have to free up the dabuf before the buffers are released,
- * since the synchronization on the dabuf is really the lock on the buffer.
- */
-void
-xfs_da_brelse(xfs_trans_t *tp, xfs_dabuf_t *dabuf)
-{
- xfs_buf_t *bp;
- xfs_buf_t **bplist;
- int i;
- int nbuf;
-
- ASSERT(dabuf->nbuf && dabuf->data && dabuf->bbcount && dabuf->bps[0]);
- if ((nbuf = dabuf->nbuf) == 1) {
- bplist = &bp;
- bp = dabuf->bps[0];
- } else {
- bplist = kmem_alloc(nbuf * sizeof(*bplist), KM_SLEEP);
- memcpy(bplist, dabuf->bps, nbuf * sizeof(*bplist));
- }
- xfs_da_buf_done(dabuf);
- for (i = 0; i < nbuf; i++)
- xfs_trans_brelse(tp, bplist[i]);
- if (bplist != &bp)
- kmem_free(bplist);
-}
-
-/*
- * Invalidate dabuf from a transaction.
- */
-void
-xfs_da_binval(xfs_trans_t *tp, xfs_dabuf_t *dabuf)
-{
- xfs_buf_t *bp;
- xfs_buf_t **bplist;
- int i;
- int nbuf;
-
- ASSERT(dabuf->nbuf && dabuf->data && dabuf->bbcount && dabuf->bps[0]);
- if ((nbuf = dabuf->nbuf) == 1) {
- bplist = &bp;
- bp = dabuf->bps[0];
- } else {
- bplist = kmem_alloc(nbuf * sizeof(*bplist), KM_SLEEP);
- memcpy(bplist, dabuf->bps, nbuf * sizeof(*bplist));
- }
- xfs_da_buf_done(dabuf);
- for (i = 0; i < nbuf; i++)
- xfs_trans_binval(tp, bplist[i]);
- if (bplist != &bp)
- kmem_free(bplist);
-}
-
-/*
- * Get the first daddr from a dabuf.
- */
-xfs_daddr_t
-xfs_da_blkno(xfs_dabuf_t *dabuf)
-{
- ASSERT(dabuf->nbuf);
- ASSERT(dabuf->data);
- return XFS_BUF_ADDR(dabuf->bps[0]);
-}
diff --git a/libxfs/xfs_dir2.c b/libxfs/xfs_dir2.c
index d475118..6a4027f 100644
--- a/libxfs/xfs_dir2.c
+++ b/libxfs/xfs_dir2.c
@@ -98,15 +98,15 @@ int
xfs_dir_isempty(
xfs_inode_t *dp)
{
- xfs_dir2_sf_t *sfp;
+ xfs_dir2_sf_hdr_t *sfp;
- ASSERT((dp->i_d.di_mode & S_IFMT) == S_IFDIR);
+ ASSERT(S_ISDIR(dp->i_d.di_mode));
if (dp->i_d.di_size == 0) /* might happen during shutdown. */
return 1;
if (dp->i_d.di_size > XFS_IFORK_DSIZE(dp))
return 0;
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
- return !sfp->hdr.count;
+ sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
+ return !sfp->count;
}
/*
@@ -135,7 +135,7 @@ xfs_dir_ino_validate(
XFS_AGINO_TO_INO(mp, agno, agino) == ino;
if (unlikely(XFS_TEST_ERROR(!ino_ok, mp, XFS_ERRTAG_DIR_INO_VALIDATE,
XFS_RANDOM_DIR_INO_VALIDATE))) {
- xfs_fs_cmn_err(CE_WARN, mp, "Invalid inode number 0x%Lx",
+ xfs_warn(mp, "Invalid inode number 0x%Lx",
(unsigned long long) ino);
XFS_ERROR_REPORT("xfs_dir_ino_validate", XFS_ERRLEVEL_LOW, mp);
return XFS_ERROR(EFSCORRUPTED);
@@ -158,7 +158,7 @@ xfs_dir_init(
memset((char *)&args, 0, sizeof(args));
args.dp = dp;
args.trans = tp;
- ASSERT((dp->i_d.di_mode & S_IFMT) == S_IFDIR);
+ ASSERT(S_ISDIR(dp->i_d.di_mode));
if ((error = xfs_dir_ino_validate(tp->t_mountp, pdp->i_ino)))
return error;
return xfs_dir2_sf_create(&args, pdp->i_ino);
@@ -181,7 +181,7 @@ xfs_dir_createname(
int rval;
int v; /* type-checking value */
- ASSERT((dp->i_d.di_mode & S_IFMT) == S_IFDIR);
+ ASSERT(S_ISDIR(dp->i_d.di_mode));
if ((rval = xfs_dir_ino_validate(tp->t_mountp, inum)))
return rval;
XFS_STATS_INC(xs_dir_create);
@@ -257,7 +257,7 @@ xfs_dir_lookup(
int rval;
int v; /* type-checking value */
- ASSERT((dp->i_d.di_mode & S_IFMT) == S_IFDIR);
+ ASSERT(S_ISDIR(dp->i_d.di_mode));
XFS_STATS_INC(xs_dir_lookup);
memset(&args, 0, sizeof(xfs_da_args_t));
@@ -312,7 +312,7 @@ xfs_dir_removename(
int rval;
int v; /* type-checking value */
- ASSERT((dp->i_d.di_mode & S_IFMT) == S_IFDIR);
+ ASSERT(S_ISDIR(dp->i_d.di_mode));
XFS_STATS_INC(xs_dir_remove);
memset(&args, 0, sizeof(xfs_da_args_t));
@@ -359,7 +359,7 @@ xfs_dir_replace(
int rval;
int v; /* type-checking value */
- ASSERT((dp->i_d.di_mode & S_IFMT) == S_IFDIR);
+ ASSERT(S_ISDIR(dp->i_d.di_mode));
if ((rval = xfs_dir_ino_validate(tp->t_mountp, inum)))
return rval;
@@ -397,129 +397,34 @@ xfs_dir_replace(
/*
* Add a block to the directory.
- * This routine is for data and free blocks, not leaf/node blocks
- * which are handled by xfs_da_grow_inode.
+ *
+ * This routine is for data and free blocks, not leaf/node blocks which are
+ * handled by xfs_da_grow_inode.
*/
int
xfs_dir2_grow_inode(
- xfs_da_args_t *args,
- int space, /* v2 dir's space XFS_DIR2_xxx_SPACE */
- xfs_dir2_db_t *dbp) /* out: block number added */
+ struct xfs_da_args *args,
+ int space, /* v2 dir's space XFS_DIR2_xxx_SPACE */
+ xfs_dir2_db_t *dbp) /* out: block number added */
{
- xfs_fileoff_t bno; /* directory offset of new block */
- int count; /* count of filesystem blocks */
- xfs_inode_t *dp; /* incore directory inode */
- int error;
- int got; /* blocks actually mapped */
- int i;
- xfs_bmbt_irec_t map; /* single structure for bmap */
- int mapi; /* mapping index */
- xfs_bmbt_irec_t *mapp; /* bmap mapping structure(s) */
- xfs_mount_t *mp;
- int nmap; /* number of bmap entries */
- xfs_trans_t *tp;
- xfs_drfsbno_t nblks;
+ struct xfs_inode *dp = args->dp;
+ struct xfs_mount *mp = dp->i_mount;
+ xfs_fileoff_t bno; /* directory offset of new block */
+ int count; /* count of filesystem blocks */
+ int error;
trace_xfs_dir2_grow_inode(args, space);
- dp = args->dp;
- tp = args->trans;
- mp = dp->i_mount;
- nblks = dp->i_d.di_nblocks;
/*
* Set lowest possible block in the space requested.
*/
bno = XFS_B_TO_FSBT(mp, space * XFS_DIR2_SPACE_SIZE);
count = mp->m_dirblkfsbs;
- /*
- * Find the first hole for our block.
- */
- if ((error = xfs_bmap_first_unused(tp, dp, count, &bno, XFS_DATA_FORK)))
- return error;
- nmap = 1;
- ASSERT(args->firstblock != NULL);
- /*
- * Try mapping the new block contiguously (one extent).
- */
- if ((error = xfs_bmapi(tp, dp, bno, count,
- XFS_BMAPI_WRITE|XFS_BMAPI_METADATA|XFS_BMAPI_CONTIG,
- args->firstblock, args->total, &map, &nmap,
- args->flist)))
- return error;
- ASSERT(nmap <= 1);
- if (nmap == 1) {
- mapp = ↦
- mapi = 1;
- }
- /*
- * Didn't work and this is a multiple-fsb directory block.
- * Try again with contiguous flag turned on.
- */
- else if (nmap == 0 && count > 1) {
- xfs_fileoff_t b; /* current file offset */
- /*
- * Space for maximum number of mappings.
- */
- mapp = kmem_alloc(sizeof(*mapp) * count, KM_SLEEP);
- /*
- * Iterate until we get to the end of our block.
- */
- for (b = bno, mapi = 0; b < bno + count; ) {
- int c; /* current fsb count */
-
- /*
- * Can't map more than MAX_NMAP at once.
- */
- nmap = MIN(XFS_BMAP_MAX_NMAP, count);
- c = (int)(bno + count - b);
- if ((error = xfs_bmapi(tp, dp, b, c,
- XFS_BMAPI_WRITE|XFS_BMAPI_METADATA,
- args->firstblock, args->total,
- &mapp[mapi], &nmap, args->flist))) {
- kmem_free(mapp);
- return error;
- }
- if (nmap < 1)
- break;
- /*
- * Add this bunch into our table, go to the next offset.
- */
- mapi += nmap;
- b = mapp[mapi - 1].br_startoff +
- mapp[mapi - 1].br_blockcount;
- }
- }
- /*
- * Didn't work.
- */
- else {
- mapi = 0;
- mapp = NULL;
- }
- /*
- * See how many fsb's we got.
- */
- for (i = 0, got = 0; i < mapi; i++)
- got += mapp[i].br_blockcount;
- /*
- * Didn't get enough fsb's, or the first/last block's are wrong.
- */
- if (got != count || mapp[0].br_startoff != bno ||
- mapp[mapi - 1].br_startoff + mapp[mapi - 1].br_blockcount !=
- bno + count) {
- if (mapp != &map)
- kmem_free(mapp);
- return XFS_ERROR(ENOSPC);
- }
- /*
- * Done with the temporary mapping table.
- */
- if (mapp != &map)
- kmem_free(mapp);
+ error = xfs_da_grow_inode_int(args, &bno, count);
+ if (error)
+ return error;
- /* account for newly allocated blocks in reserved blocks total */
- args->total -= dp->i_d.di_nblocks - nblks;
*dbp = xfs_dir2_da_to_db(mp, (xfs_dablk_t)bno);
/*
@@ -531,7 +436,7 @@ xfs_dir2_grow_inode(
size = XFS_FSB_TO_B(mp, bno + count);
if (size > dp->i_d.di_size) {
dp->i_d.di_size = size;
- xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE);
+ xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE);
}
}
return 0;
@@ -588,7 +493,7 @@ int
xfs_dir2_shrink_inode(
xfs_da_args_t *args,
xfs_dir2_db_t db,
- xfs_dabuf_t *bp)
+ struct xfs_buf *bp)
{
xfs_fileoff_t bno; /* directory file offset */
xfs_dablk_t da; /* directory file offset */
@@ -630,7 +535,7 @@ xfs_dir2_shrink_inode(
/*
* Invalidate the buffer from the transaction.
*/
- xfs_da_binval(tp, bp);
+ xfs_trans_binval(tp, bp);
/*
* If it's not a data block, we're done.
*/
diff --git a/libxfs/xfs_dir2_block.c b/libxfs/xfs_dir2_block.c
index b614ea6..7397faa 100644
--- a/libxfs/xfs_dir2_block.c
+++ b/libxfs/xfs_dir2_block.c
@@ -21,10 +21,10 @@
/*
* Local function prototypes.
*/
-static void xfs_dir2_block_log_leaf(xfs_trans_t *tp, xfs_dabuf_t *bp, int first,
- int last);
-static void xfs_dir2_block_log_tail(xfs_trans_t *tp, xfs_dabuf_t *bp);
-static int xfs_dir2_block_lookup_int(xfs_da_args_t *args, xfs_dabuf_t **bpp,
+static void xfs_dir2_block_log_leaf(xfs_trans_t *tp, struct xfs_buf *bp,
+ int first, int last);
+static void xfs_dir2_block_log_tail(xfs_trans_t *tp, struct xfs_buf *bp);
+static int xfs_dir2_block_lookup_int(xfs_da_args_t *args, struct xfs_buf **bpp,
int *entno);
static int xfs_dir2_block_sort(const void *a, const void *b);
@@ -40,6 +40,214 @@ xfs_dir_startup(void)
xfs_dir_hash_dotdot = xfs_da_hashname((unsigned char *)"..", 2);
}
+static void
+xfs_dir2_block_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_dir2_data_hdr *hdr = bp->b_addr;
+ int block_ok = 0;
+
+ block_ok = hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC);
+ block_ok = block_ok && __xfs_dir2_data_check(NULL, bp) == 0;
+
+ if (!block_ok) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, hdr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+}
+
+static void
+xfs_dir2_block_read_verify(
+ struct xfs_buf *bp)
+{
+ xfs_dir2_block_verify(bp);
+}
+
+static void
+xfs_dir2_block_write_verify(
+ struct xfs_buf *bp)
+{
+ xfs_dir2_block_verify(bp);
+}
+
+const struct xfs_buf_ops xfs_dir2_block_buf_ops = {
+ .verify_read = xfs_dir2_block_read_verify,
+ .verify_write = xfs_dir2_block_write_verify,
+};
+
+static int
+xfs_dir2_block_read(
+ struct xfs_trans *tp,
+ struct xfs_inode *dp,
+ struct xfs_buf **bpp)
+{
+ struct xfs_mount *mp = dp->i_mount;
+
+ return xfs_da_read_buf(tp, dp, mp->m_dirdatablk, -1, bpp,
+ XFS_DATA_FORK, &xfs_dir2_block_buf_ops);
+}
+
+static void
+xfs_dir2_block_need_space(
+ struct xfs_dir2_data_hdr *hdr,
+ struct xfs_dir2_block_tail *btp,
+ struct xfs_dir2_leaf_entry *blp,
+ __be16 **tagpp,
+ struct xfs_dir2_data_unused **dupp,
+ struct xfs_dir2_data_unused **enddupp,
+ int *compact,
+ int len)
+{
+ struct xfs_dir2_data_free *bf;
+ __be16 *tagp = NULL;
+ struct xfs_dir2_data_unused *dup = NULL;
+ struct xfs_dir2_data_unused *enddup = NULL;
+
+ *compact = 0;
+ bf = hdr->bestfree;
+
+ /*
+ * If there are stale entries we'll use one for the leaf.
+ */
+ if (btp->stale) {
+ if (be16_to_cpu(bf[0].length) >= len) {
+ /*
+ * The biggest entry enough to avoid compaction.
+ */
+ dup = (xfs_dir2_data_unused_t *)
+ ((char *)hdr + be16_to_cpu(bf[0].offset));
+ goto out;
+ }
+
+ /*
+ * Will need to compact to make this work.
+ * Tag just before the first leaf entry.
+ */
+ *compact = 1;
+ tagp = (__be16 *)blp - 1;
+
+ /* Data object just before the first leaf entry. */
+ dup = (xfs_dir2_data_unused_t *)((char *)hdr + be16_to_cpu(*tagp));
+
+ /*
+ * If it's not free then the data will go where the
+ * leaf data starts now, if it works at all.
+ */
+ if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
+ if (be16_to_cpu(dup->length) + (be32_to_cpu(btp->stale) - 1) *
+ (uint)sizeof(*blp) < len)
+ dup = NULL;
+ } else if ((be32_to_cpu(btp->stale) - 1) * (uint)sizeof(*blp) < len)
+ dup = NULL;
+ else
+ dup = (xfs_dir2_data_unused_t *)blp;
+ goto out;
+ }
+
+ /*
+ * no stale entries, so just use free space.
+ * Tag just before the first leaf entry.
+ */
+ tagp = (__be16 *)blp - 1;
+
+ /* Data object just before the first leaf entry. */
+ enddup = (xfs_dir2_data_unused_t *)((char *)hdr + be16_to_cpu(*tagp));
+
+ /*
+ * If it's not free then can't do this add without cleaning up:
+ * the space before the first leaf entry needs to be free so it
+ * can be expanded to hold the pointer to the new entry.
+ */
+ if (be16_to_cpu(enddup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
+ /*
+ * Check out the biggest freespace and see if it's the same one.
+ */
+ dup = (xfs_dir2_data_unused_t *)
+ ((char *)hdr + be16_to_cpu(bf[0].offset));
+ if (dup != enddup) {
+ /*
+ * Not the same free entry, just check its length.
+ */
+ if (be16_to_cpu(dup->length) < len)
+ dup = NULL;
+ goto out;
+ }
+
+ /*
+ * It is the biggest freespace, can it hold the leaf too?
+ */
+ if (be16_to_cpu(dup->length) < len + (uint)sizeof(*blp)) {
+ /*
+ * Yes, use the second-largest entry instead if it works.
+ */
+ if (be16_to_cpu(bf[1].length) >= len)
+ dup = (xfs_dir2_data_unused_t *)
+ ((char *)hdr + be16_to_cpu(bf[1].offset));
+ else
+ dup = NULL;
+ }
+ }
+out:
+ *tagpp = tagp;
+ *dupp = dup;
+ *enddupp = enddup;
+}
+
+/*
+ * compact the leaf entries.
+ * Leave the highest-numbered stale entry stale.
+ * XXX should be the one closest to mid but mid is not yet computed.
+ */
+static void
+xfs_dir2_block_compact(
+ struct xfs_trans *tp,
+ struct xfs_buf *bp,
+ struct xfs_dir2_data_hdr *hdr,
+ struct xfs_dir2_block_tail *btp,
+ struct xfs_dir2_leaf_entry *blp,
+ int *needlog,
+ int *lfloghigh,
+ int *lfloglow)
+{
+ int fromidx; /* source leaf index */
+ int toidx; /* target leaf index */
+ int needscan = 0;
+ int highstale; /* high stale index */
+
+ fromidx = toidx = be32_to_cpu(btp->count) - 1;
+ highstale = *lfloghigh = -1;
+ for (; fromidx >= 0; fromidx--) {
+ if (blp[fromidx].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR)) {
+ if (highstale == -1)
+ highstale = toidx;
+ else {
+ if (*lfloghigh == -1)
+ *lfloghigh = toidx;
+ continue;
+ }
+ }
+ if (fromidx < toidx)
+ blp[toidx] = blp[fromidx];
+ toidx--;
+ }
+ *lfloglow = toidx + 1 - (be32_to_cpu(btp->stale) - 1);
+ *lfloghigh -= be32_to_cpu(btp->stale) - 1;
+ be32_add_cpu(&btp->count, -(be32_to_cpu(btp->stale) - 1));
+ xfs_dir2_data_make_free(tp, bp,
+ (xfs_dir2_data_aoff_t)((char *)blp - (char *)hdr),
+ (xfs_dir2_data_aoff_t)((be32_to_cpu(btp->stale) - 1) * sizeof(*blp)),
+ needlog, &needscan);
+ blp += be32_to_cpu(btp->stale) - 1;
+ btp->stale = cpu_to_be32(1);
+ /*
+ * If we now need to rebuild the bestfree map, do so.
+ * This needs to happen before the next call to use_free.
+ */
+ if (needscan)
+ xfs_dir2_data_freescan(tp->t_mountp, hdr, needlog);
+}
+
/*
* Add an entry to a block directory.
*/
@@ -47,10 +255,9 @@ int /* error */
xfs_dir2_block_addname(
xfs_da_args_t *args) /* directory op arguments */
{
- xfs_dir2_data_free_t *bf; /* bestfree table in block */
- xfs_dir2_block_t *block; /* directory block structure */
+ xfs_dir2_data_hdr_t *hdr; /* block header */
xfs_dir2_leaf_entry_t *blp; /* block leaf entries */
- xfs_dabuf_t *bp; /* buffer for block */
+ struct xfs_buf *bp; /* buffer for block */
xfs_dir2_block_tail_t *btp; /* block tail */
int compact; /* need to compact leaf ents */
xfs_dir2_data_entry_t *dep; /* block data entry */
@@ -78,203 +285,72 @@ xfs_dir2_block_addname(
dp = args->dp;
tp = args->trans;
mp = dp->i_mount;
- /*
- * Read the (one and only) directory block into dabuf bp.
- */
- if ((error =
- xfs_da_read_buf(tp, dp, mp->m_dirdatablk, -1, &bp, XFS_DATA_FORK))) {
+
+ /* Read the (one and only) directory block into bp. */
+ error = xfs_dir2_block_read(tp, dp, &bp);
+ if (error)
return error;
- }
- ASSERT(bp != NULL);
- block = bp->data;
- /*
- * Check the magic number, corrupted if wrong.
- */
- if (unlikely(be32_to_cpu(block->hdr.magic) != XFS_DIR2_BLOCK_MAGIC)) {
- XFS_CORRUPTION_ERROR("xfs_dir2_block_addname",
- XFS_ERRLEVEL_LOW, mp, block);
- xfs_da_brelse(tp, bp);
- return XFS_ERROR(EFSCORRUPTED);
- }
+
len = xfs_dir2_data_entsize(args->namelen);
+
/*
* Set up pointers to parts of the block.
*/
- bf = block->hdr.bestfree;
- btp = xfs_dir2_block_tail_p(mp, block);
+ hdr = bp->b_addr;
+ btp = xfs_dir2_block_tail_p(mp, hdr);
blp = xfs_dir2_block_leaf_p(btp);
+
/*
- * No stale entries? Need space for entry and new leaf.
- */
- if (!btp->stale) {
- /*
- * Tag just before the first leaf entry.
- */
- tagp = (__be16 *)blp - 1;
- /*
- * Data object just before the first leaf entry.
- */
- enddup = (xfs_dir2_data_unused_t *)((char *)block + be16_to_cpu(*tagp));
- /*
- * If it's not free then can't do this add without cleaning up:
- * the space before the first leaf entry needs to be free so it
- * can be expanded to hold the pointer to the new entry.
- */
- if (be16_to_cpu(enddup->freetag) != XFS_DIR2_DATA_FREE_TAG)
- dup = enddup = NULL;
- /*
- * Check out the biggest freespace and see if it's the same one.
- */
- else {
- dup = (xfs_dir2_data_unused_t *)
- ((char *)block + be16_to_cpu(bf[0].offset));
- if (dup == enddup) {
- /*
- * It is the biggest freespace, is it too small
- * to hold the new leaf too?
- */
- if (be16_to_cpu(dup->length) < len + (uint)sizeof(*blp)) {
- /*
- * Yes, we use the second-largest
- * entry instead if it works.
- */
- if (be16_to_cpu(bf[1].length) >= len)
- dup = (xfs_dir2_data_unused_t *)
- ((char *)block +
- be16_to_cpu(bf[1].offset));
- else
- dup = NULL;
- }
- } else {
- /*
- * Not the same free entry,
- * just check its length.
- */
- if (be16_to_cpu(dup->length) < len) {
- dup = NULL;
- }
- }
- }
- compact = 0;
- }
- /*
- * If there are stale entries we'll use one for the leaf.
- * Is the biggest entry enough to avoid compaction?
+ * Find out if we can reuse stale entries or whether we need extra
+ * space for entry and new leaf.
*/
- else if (be16_to_cpu(bf[0].length) >= len) {
- dup = (xfs_dir2_data_unused_t *)
- ((char *)block + be16_to_cpu(bf[0].offset));
- compact = 0;
- }
+ xfs_dir2_block_need_space(hdr, btp, blp, &tagp, &dup,
+ &enddup, &compact, len);
+
/*
- * Will need to compact to make this work.
+ * Done everything we need for a space check now.
*/
- else {
- /*
- * Tag just before the first leaf entry.
- */
- tagp = (__be16 *)blp - 1;
- /*
- * Data object just before the first leaf entry.
- */
- dup = (xfs_dir2_data_unused_t *)((char *)block + be16_to_cpu(*tagp));
- /*
- * If it's not free then the data will go where the
- * leaf data starts now, if it works at all.
- */
- if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
- if (be16_to_cpu(dup->length) + (be32_to_cpu(btp->stale) - 1) *
- (uint)sizeof(*blp) < len)
- dup = NULL;
- } else if ((be32_to_cpu(btp->stale) - 1) * (uint)sizeof(*blp) < len)
- dup = NULL;
- else
- dup = (xfs_dir2_data_unused_t *)blp;
- compact = 1;
+ if (args->op_flags & XFS_DA_OP_JUSTCHECK) {
+ xfs_trans_brelse(tp, bp);
+ if (!dup)
+ return XFS_ERROR(ENOSPC);
+ return 0;
}
- /*
- * If this isn't a real add, we're done with the buffer.
- */
- if (args->op_flags & XFS_DA_OP_JUSTCHECK)
- xfs_da_brelse(tp, bp);
+
/*
* If we don't have space for the new entry & leaf ...
*/
if (!dup) {
- /*
- * Not trying to actually do anything, or don't have
- * a space reservation: return no-space.
- */
- if ((args->op_flags & XFS_DA_OP_JUSTCHECK) || args->total == 0)
+ /* Don't have a space reservation: return no-space. */
+ if (args->total == 0)
return XFS_ERROR(ENOSPC);
/*
* Convert to the next larger format.
* Then add the new entry in that format.
*/
error = xfs_dir2_block_to_leaf(args, bp);
- xfs_da_buf_done(bp);
if (error)
return error;
return xfs_dir2_leaf_addname(args);
}
- /*
- * Just checking, and it would work, so say so.
- */
- if (args->op_flags & XFS_DA_OP_JUSTCHECK)
- return 0;
+
needlog = needscan = 0;
+
/*
* If need to compact the leaf entries, do it now.
- * Leave the highest-numbered stale entry stale.
- * XXX should be the one closest to mid but mid is not yet computed.
*/
- if (compact) {
- int fromidx; /* source leaf index */
- int toidx; /* target leaf index */
-
- for (fromidx = toidx = be32_to_cpu(btp->count) - 1,
- highstale = lfloghigh = -1;
- fromidx >= 0;
- fromidx--) {
- if (be32_to_cpu(blp[fromidx].address) == XFS_DIR2_NULL_DATAPTR) {
- if (highstale == -1)
- highstale = toidx;
- else {
- if (lfloghigh == -1)
- lfloghigh = toidx;
- continue;
- }
- }
- if (fromidx < toidx)
- blp[toidx] = blp[fromidx];
- toidx--;
- }
- lfloglow = toidx + 1 - (be32_to_cpu(btp->stale) - 1);
- lfloghigh -= be32_to_cpu(btp->stale) - 1;
- be32_add_cpu(&btp->count, -(be32_to_cpu(btp->stale) - 1));
- xfs_dir2_data_make_free(tp, bp,
- (xfs_dir2_data_aoff_t)((char *)blp - (char *)block),
- (xfs_dir2_data_aoff_t)((be32_to_cpu(btp->stale) - 1) * sizeof(*blp)),
- &needlog, &needscan);
- blp += be32_to_cpu(btp->stale) - 1;
- btp->stale = cpu_to_be32(1);
+ if (compact)
+ xfs_dir2_block_compact(tp, bp, hdr, btp, blp, &needlog,
+ &lfloghigh, &lfloglow);
+ else if (btp->stale) {
/*
- * If we now need to rebuild the bestfree map, do so.
- * This needs to happen before the next call to use_free.
+ * Set leaf logging boundaries to impossible state.
+ * For the no-stale case they're set explicitly.
*/
- if (needscan) {
- xfs_dir2_data_freescan(mp, (xfs_dir2_data_t *)block, &needlog);
- needscan = 0;
- }
- }
- /*
- * Set leaf logging boundaries to impossible state.
- * For the no-stale case they're set explicitly.
- */
- else if (btp->stale) {
lfloglow = be32_to_cpu(btp->count);
lfloghigh = -1;
}
+
/*
* Find the slot that's first lower than our hash value, -1 if none.
*/
@@ -299,7 +375,7 @@ xfs_dir2_block_addname(
*/
xfs_dir2_data_use_free(tp, bp, enddup,
(xfs_dir2_data_aoff_t)
- ((char *)enddup - (char *)block + be16_to_cpu(enddup->length) -
+ ((char *)enddup - (char *)hdr + be16_to_cpu(enddup->length) -
sizeof(*blp)),
(xfs_dir2_data_aoff_t)sizeof(*blp),
&needlog, &needscan);
@@ -312,8 +388,7 @@ xfs_dir2_block_addname(
* This needs to happen before the next call to use_free.
*/
if (needscan) {
- xfs_dir2_data_freescan(mp, (xfs_dir2_data_t *)block,
- &needlog);
+ xfs_dir2_data_freescan(mp, hdr, &needlog);
needscan = 0;
}
/*
@@ -334,12 +409,14 @@ xfs_dir2_block_addname(
else {
for (lowstale = mid;
lowstale >= 0 &&
- be32_to_cpu(blp[lowstale].address) != XFS_DIR2_NULL_DATAPTR;
+ blp[lowstale].address !=
+ cpu_to_be32(XFS_DIR2_NULL_DATAPTR);
lowstale--)
continue;
for (highstale = mid + 1;
highstale < be32_to_cpu(btp->count) &&
- be32_to_cpu(blp[highstale].address) != XFS_DIR2_NULL_DATAPTR &&
+ blp[highstale].address !=
+ cpu_to_be32(XFS_DIR2_NULL_DATAPTR) &&
(lowstale < 0 || mid - lowstale > highstale - mid);
highstale++)
continue;
@@ -378,13 +455,13 @@ xfs_dir2_block_addname(
*/
blp[mid].hashval = cpu_to_be32(args->hashval);
blp[mid].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp,
- (char *)dep - (char *)block));
+ (char *)dep - (char *)hdr));
xfs_dir2_block_log_leaf(tp, bp, lfloglow, lfloghigh);
/*
* Mark space for the data entry used.
*/
xfs_dir2_data_use_free(tp, bp, dup,
- (xfs_dir2_data_aoff_t)((char *)dup - (char *)block),
+ (xfs_dir2_data_aoff_t)((char *)dup - (char *)hdr),
(xfs_dir2_data_aoff_t)len, &needlog, &needscan);
/*
* Create the new data entry.
@@ -393,18 +470,17 @@ xfs_dir2_block_addname(
dep->namelen = args->namelen;
memcpy(dep->name, args->name, args->namelen);
tagp = xfs_dir2_data_entry_tag_p(dep);
- *tagp = cpu_to_be16((char *)dep - (char *)block);
+ *tagp = cpu_to_be16((char *)dep - (char *)hdr);
/*
* Clean up the bestfree array and log the header, tail, and entry.
*/
if (needscan)
- xfs_dir2_data_freescan(mp, (xfs_dir2_data_t *)block, &needlog);
+ xfs_dir2_data_freescan(mp, hdr, &needlog);
if (needlog)
xfs_dir2_data_log_header(tp, bp);
xfs_dir2_block_log_tail(tp, bp);
xfs_dir2_data_log_entry(tp, bp, dep);
xfs_dir2_data_check(dp, bp);
- xfs_da_buf_done(bp);
return 0;
}
@@ -414,21 +490,18 @@ xfs_dir2_block_addname(
static void
xfs_dir2_block_log_leaf(
xfs_trans_t *tp, /* transaction structure */
- xfs_dabuf_t *bp, /* block buffer */
+ struct xfs_buf *bp, /* block buffer */
int first, /* index of first logged leaf */
int last) /* index of last logged leaf */
{
- xfs_dir2_block_t *block; /* directory block structure */
- xfs_dir2_leaf_entry_t *blp; /* block leaf entries */
- xfs_dir2_block_tail_t *btp; /* block tail */
- xfs_mount_t *mp; /* filesystem mount point */
+ xfs_dir2_data_hdr_t *hdr = bp->b_addr;
+ xfs_dir2_leaf_entry_t *blp;
+ xfs_dir2_block_tail_t *btp;
- mp = tp->t_mountp;
- block = bp->data;
- btp = xfs_dir2_block_tail_p(mp, block);
+ btp = xfs_dir2_block_tail_p(tp->t_mountp, hdr);
blp = xfs_dir2_block_leaf_p(btp);
- xfs_da_log_buf(tp, bp, (uint)((char *)&blp[first] - (char *)block),
- (uint)((char *)&blp[last + 1] - (char *)block - 1));
+ xfs_trans_log_buf(tp, bp, (uint)((char *)&blp[first] - (char *)hdr),
+ (uint)((char *)&blp[last + 1] - (char *)hdr - 1));
}
/*
@@ -437,17 +510,14 @@ xfs_dir2_block_log_leaf(
static void
xfs_dir2_block_log_tail(
xfs_trans_t *tp, /* transaction structure */
- xfs_dabuf_t *bp) /* block buffer */
+ struct xfs_buf *bp) /* block buffer */
{
- xfs_dir2_block_t *block; /* directory block structure */
- xfs_dir2_block_tail_t *btp; /* block tail */
- xfs_mount_t *mp; /* filesystem mount point */
+ xfs_dir2_data_hdr_t *hdr = bp->b_addr;
+ xfs_dir2_block_tail_t *btp;
- mp = tp->t_mountp;
- block = bp->data;
- btp = xfs_dir2_block_tail_p(mp, block);
- xfs_da_log_buf(tp, bp, (uint)((char *)btp - (char *)block),
- (uint)((char *)(btp + 1) - (char *)block - 1));
+ btp = xfs_dir2_block_tail_p(tp->t_mountp, hdr);
+ xfs_trans_log_buf(tp, bp, (uint)((char *)btp - (char *)hdr),
+ (uint)((char *)(btp + 1) - (char *)hdr - 1));
}
/*
@@ -458,9 +528,9 @@ int /* error */
xfs_dir2_block_lookup(
xfs_da_args_t *args) /* dir lookup arguments */
{
- xfs_dir2_block_t *block; /* block structure */
+ xfs_dir2_data_hdr_t *hdr; /* block header */
xfs_dir2_leaf_entry_t *blp; /* block leaf entries */
- xfs_dabuf_t *bp; /* block buffer */
+ struct xfs_buf *bp; /* block buffer */
xfs_dir2_block_tail_t *btp; /* block tail */
xfs_dir2_data_entry_t *dep; /* block data entry */
xfs_inode_t *dp; /* incore inode */
@@ -478,21 +548,21 @@ xfs_dir2_block_lookup(
return error;
dp = args->dp;
mp = dp->i_mount;
- block = bp->data;
+ hdr = bp->b_addr;
xfs_dir2_data_check(dp, bp);
- btp = xfs_dir2_block_tail_p(mp, block);
+ btp = xfs_dir2_block_tail_p(mp, hdr);
blp = xfs_dir2_block_leaf_p(btp);
/*
* Get the offset from the leaf entry, to point to the data.
*/
- dep = (xfs_dir2_data_entry_t *)((char *)block +
+ dep = (xfs_dir2_data_entry_t *)((char *)hdr +
xfs_dir2_dataptr_to_off(mp, be32_to_cpu(blp[ent].address)));
/*
* Fill in inode number, CI name if appropriate, release the block.
*/
args->inumber = be64_to_cpu(dep->inumber);
error = xfs_dir_cilookup_result(args, dep->name, dep->namelen);
- xfs_da_brelse(args->trans, bp);
+ xfs_trans_brelse(args->trans, bp);
return XFS_ERROR(error);
}
@@ -502,13 +572,13 @@ xfs_dir2_block_lookup(
static int /* error */
xfs_dir2_block_lookup_int(
xfs_da_args_t *args, /* dir lookup arguments */
- xfs_dabuf_t **bpp, /* returned block buffer */
+ struct xfs_buf **bpp, /* returned block buffer */
int *entno) /* returned entry number */
{
xfs_dir2_dataptr_t addr; /* data entry address */
- xfs_dir2_block_t *block; /* block structure */
+ xfs_dir2_data_hdr_t *hdr; /* block header */
xfs_dir2_leaf_entry_t *blp; /* block leaf entries */
- xfs_dabuf_t *bp; /* block buffer */
+ struct xfs_buf *bp; /* block buffer */
xfs_dir2_block_tail_t *btp; /* block tail */
xfs_dir2_data_entry_t *dep; /* block data entry */
xfs_inode_t *dp; /* incore inode */
@@ -524,17 +594,14 @@ xfs_dir2_block_lookup_int(
dp = args->dp;
tp = args->trans;
mp = dp->i_mount;
- /*
- * Read the buffer, return error if we can't get it.
- */
- if ((error =
- xfs_da_read_buf(tp, dp, mp->m_dirdatablk, -1, &bp, XFS_DATA_FORK))) {
+
+ error = xfs_dir2_block_read(tp, dp, &bp);
+ if (error)
return error;
- }
- ASSERT(bp != NULL);
- block = bp->data;
+
+ hdr = bp->b_addr;
xfs_dir2_data_check(dp, bp);
- btp = xfs_dir2_block_tail_p(mp, block);
+ btp = xfs_dir2_block_tail_p(mp, hdr);
blp = xfs_dir2_block_leaf_p(btp);
/*
* Loop doing a binary search for our hash value.
@@ -551,7 +618,7 @@ xfs_dir2_block_lookup_int(
high = mid - 1;
if (low > high) {
ASSERT(args->op_flags & XFS_DA_OP_OKNOENT);
- xfs_da_brelse(tp, bp);
+ xfs_trans_brelse(tp, bp);
return XFS_ERROR(ENOENT);
}
}
@@ -572,7 +639,7 @@ xfs_dir2_block_lookup_int(
* Get pointer to the entry from the leaf.
*/
dep = (xfs_dir2_data_entry_t *)
- ((char *)block + xfs_dir2_dataptr_to_off(mp, addr));
+ ((char *)hdr + xfs_dir2_dataptr_to_off(mp, addr));
/*
* Compare name and if it's an exact match, return the index
* and buffer. If it's the first case-insensitive match, store
@@ -599,7 +666,7 @@ xfs_dir2_block_lookup_int(
/*
* No match, release the buffer and return ENOENT.
*/
- xfs_da_brelse(tp, bp);
+ xfs_trans_brelse(tp, bp);
return XFS_ERROR(ENOENT);
}
@@ -611,9 +678,9 @@ int /* error */
xfs_dir2_block_removename(
xfs_da_args_t *args) /* directory operation args */
{
- xfs_dir2_block_t *block; /* block structure */
+ xfs_dir2_data_hdr_t *hdr; /* block header */
xfs_dir2_leaf_entry_t *blp; /* block leaf pointer */
- xfs_dabuf_t *bp; /* block buffer */
+ struct xfs_buf *bp; /* block buffer */
xfs_dir2_block_tail_t *btp; /* block tail */
xfs_dir2_data_entry_t *dep; /* block data entry */
xfs_inode_t *dp; /* incore inode */
@@ -638,20 +705,20 @@ xfs_dir2_block_removename(
dp = args->dp;
tp = args->trans;
mp = dp->i_mount;
- block = bp->data;
- btp = xfs_dir2_block_tail_p(mp, block);
+ hdr = bp->b_addr;
+ btp = xfs_dir2_block_tail_p(mp, hdr);
blp = xfs_dir2_block_leaf_p(btp);
/*
* Point to the data entry using the leaf entry.
*/
dep = (xfs_dir2_data_entry_t *)
- ((char *)block + xfs_dir2_dataptr_to_off(mp, be32_to_cpu(blp[ent].address)));
+ ((char *)hdr + xfs_dir2_dataptr_to_off(mp, be32_to_cpu(blp[ent].address)));
/*
* Mark the data entry's space free.
*/
needlog = needscan = 0;
xfs_dir2_data_make_free(tp, bp,
- (xfs_dir2_data_aoff_t)((char *)dep - (char *)block),
+ (xfs_dir2_data_aoff_t)((char *)dep - (char *)hdr),
xfs_dir2_data_entsize(dep->namelen), &needlog, &needscan);
/*
* Fix up the block tail.
@@ -667,18 +734,17 @@ xfs_dir2_block_removename(
* Fix up bestfree, log the header if necessary.
*/
if (needscan)
- xfs_dir2_data_freescan(mp, (xfs_dir2_data_t *)block, &needlog);
+ xfs_dir2_data_freescan(mp, hdr, &needlog);
if (needlog)
xfs_dir2_data_log_header(tp, bp);
xfs_dir2_data_check(dp, bp);
/*
* See if the size as a shortform is good enough.
*/
- if ((size = xfs_dir2_block_sfsize(dp, block, &sfh)) >
- XFS_IFORK_DSIZE(dp)) {
- xfs_da_buf_done(bp);
+ size = xfs_dir2_block_sfsize(dp, hdr, &sfh);
+ if (size > XFS_IFORK_DSIZE(dp))
return 0;
- }
+
/*
* If it works, do the conversion.
*/
@@ -693,9 +759,9 @@ int /* error */
xfs_dir2_block_replace(
xfs_da_args_t *args) /* directory operation args */
{
- xfs_dir2_block_t *block; /* block structure */
+ xfs_dir2_data_hdr_t *hdr; /* block header */
xfs_dir2_leaf_entry_t *blp; /* block leaf entries */
- xfs_dabuf_t *bp; /* block buffer */
+ struct xfs_buf *bp; /* block buffer */
xfs_dir2_block_tail_t *btp; /* block tail */
xfs_dir2_data_entry_t *dep; /* block data entry */
xfs_inode_t *dp; /* incore inode */
@@ -714,14 +780,14 @@ xfs_dir2_block_replace(
}
dp = args->dp;
mp = dp->i_mount;
- block = bp->data;
- btp = xfs_dir2_block_tail_p(mp, block);
+ hdr = bp->b_addr;
+ btp = xfs_dir2_block_tail_p(mp, hdr);
blp = xfs_dir2_block_leaf_p(btp);
/*
* Point to the data entry we need to change.
*/
dep = (xfs_dir2_data_entry_t *)
- ((char *)block + xfs_dir2_dataptr_to_off(mp, be32_to_cpu(blp[ent].address)));
+ ((char *)hdr + xfs_dir2_dataptr_to_off(mp, be32_to_cpu(blp[ent].address)));
ASSERT(be64_to_cpu(dep->inumber) != args->inumber);
/*
* Change the inode number to the new value.
@@ -729,7 +795,6 @@ xfs_dir2_block_replace(
dep->inumber = cpu_to_be64(args->inumber);
xfs_dir2_data_log_entry(args->trans, bp, dep);
xfs_dir2_data_check(dp, bp);
- xfs_da_buf_done(bp);
return 0;
}
@@ -756,11 +821,11 @@ xfs_dir2_block_sort(
int /* error */
xfs_dir2_leaf_to_block(
xfs_da_args_t *args, /* operation arguments */
- xfs_dabuf_t *lbp, /* leaf buffer */
- xfs_dabuf_t *dbp) /* data buffer */
+ struct xfs_buf *lbp, /* leaf buffer */
+ struct xfs_buf *dbp) /* data buffer */
{
__be16 *bestsp; /* leaf bests table */
- xfs_dir2_block_t *block; /* block structure */
+ xfs_dir2_data_hdr_t *hdr; /* block header */
xfs_dir2_block_tail_t *btp; /* block tail */
xfs_inode_t *dp; /* incore directory inode */
xfs_dir2_data_unused_t *dup; /* unused data entry */
@@ -783,8 +848,8 @@ xfs_dir2_leaf_to_block(
dp = args->dp;
tp = args->trans;
mp = dp->i_mount;
- leaf = lbp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAF1_MAGIC);
+ leaf = lbp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC));
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
/*
* If there are data blocks other than the first one, take this
@@ -795,48 +860,46 @@ xfs_dir2_leaf_to_block(
while (dp->i_d.di_size > mp->m_dirblksize) {
bestsp = xfs_dir2_leaf_bests_p(ltp);
if (be16_to_cpu(bestsp[be32_to_cpu(ltp->bestcount) - 1]) ==
- mp->m_dirblksize - (uint)sizeof(block->hdr)) {
+ mp->m_dirblksize - (uint)sizeof(*hdr)) {
if ((error =
xfs_dir2_leaf_trim_data(args, lbp,
(xfs_dir2_db_t)(be32_to_cpu(ltp->bestcount) - 1))))
- goto out;
- } else {
- error = 0;
- goto out;
- }
+ return error;
+ } else
+ return 0;
}
/*
* Read the data block if we don't already have it, give up if it fails.
*/
- if (dbp == NULL &&
- (error = xfs_da_read_buf(tp, dp, mp->m_dirdatablk, -1, &dbp,
- XFS_DATA_FORK))) {
- goto out;
+ if (!dbp) {
+ error = xfs_dir2_data_read(tp, dp, mp->m_dirdatablk, -1, &dbp);
+ if (error)
+ return error;
}
- block = dbp->data;
- ASSERT(be32_to_cpu(block->hdr.magic) == XFS_DIR2_DATA_MAGIC);
+ hdr = dbp->b_addr;
+ ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC));
/*
* Size of the "leaf" area in the block.
*/
- size = (uint)sizeof(block->tail) +
+ size = (uint)sizeof(xfs_dir2_block_tail_t) +
(uint)sizeof(*lep) * (be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale));
/*
* Look at the last data entry.
*/
- tagp = (__be16 *)((char *)block + mp->m_dirblksize) - 1;
- dup = (xfs_dir2_data_unused_t *)((char *)block + be16_to_cpu(*tagp));
+ tagp = (__be16 *)((char *)hdr + mp->m_dirblksize) - 1;
+ dup = (xfs_dir2_data_unused_t *)((char *)hdr + be16_to_cpu(*tagp));
/*
* If it's not free or is too short we can't do it.
*/
if (be16_to_cpu(dup->freetag) != XFS_DIR2_DATA_FREE_TAG ||
- be16_to_cpu(dup->length) < size) {
- error = 0;
- goto out;
- }
+ be16_to_cpu(dup->length) < size)
+ return 0;
+
/*
* Start converting it to block form.
*/
- block->hdr.magic = cpu_to_be32(XFS_DIR2_BLOCK_MAGIC);
+ dbp->b_ops = &xfs_dir2_block_buf_ops;
+ hdr->magic = cpu_to_be32(XFS_DIR2_BLOCK_MAGIC);
needlog = 1;
needscan = 0;
/*
@@ -847,7 +910,7 @@ xfs_dir2_leaf_to_block(
/*
* Initialize the block tail.
*/
- btp = xfs_dir2_block_tail_p(mp, block);
+ btp = xfs_dir2_block_tail_p(mp, hdr);
btp->count = cpu_to_be32(be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale));
btp->stale = 0;
xfs_dir2_block_log_tail(tp, dbp);
@@ -856,7 +919,8 @@ xfs_dir2_leaf_to_block(
*/
lep = xfs_dir2_block_leaf_p(btp);
for (from = to = 0; from < be16_to_cpu(leaf->hdr.count); from++) {
- if (be32_to_cpu(leaf->ents[from].address) == XFS_DIR2_NULL_DATAPTR)
+ if (leaf->ents[from].address ==
+ cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
continue;
lep[to++] = leaf->ents[from];
}
@@ -866,32 +930,24 @@ xfs_dir2_leaf_to_block(
* Scan the bestfree if we need it and log the data block header.
*/
if (needscan)
- xfs_dir2_data_freescan(mp, (xfs_dir2_data_t *)block, &needlog);
+ xfs_dir2_data_freescan(mp, hdr, &needlog);
if (needlog)
xfs_dir2_data_log_header(tp, dbp);
/*
* Pitch the old leaf block.
*/
error = xfs_da_shrink_inode(args, mp->m_dirleafblk, lbp);
- lbp = NULL;
- if (error) {
- goto out;
- }
+ if (error)
+ return error;
+
/*
* Now see if the resulting block can be shrunken to shortform.
*/
- if ((size = xfs_dir2_block_sfsize(dp, block, &sfh)) >
- XFS_IFORK_DSIZE(dp)) {
- error = 0;
- goto out;
- }
+ size = xfs_dir2_block_sfsize(dp, hdr, &sfh);
+ if (size > XFS_IFORK_DSIZE(dp))
+ return 0;
+
return xfs_dir2_block_to_sf(args, dbp, size, &sfh);
-out:
- if (lbp)
- xfs_da_buf_done(lbp);
- if (dbp)
- xfs_da_buf_done(dbp);
- return error;
}
/*
@@ -902,12 +958,10 @@ xfs_dir2_sf_to_block(
xfs_da_args_t *args) /* operation arguments */
{
xfs_dir2_db_t blkno; /* dir-relative block # (0) */
- xfs_dir2_block_t *block; /* block structure */
+ xfs_dir2_data_hdr_t *hdr; /* block header */
xfs_dir2_leaf_entry_t *blp; /* block leaf entries */
- xfs_dabuf_t *bp; /* block buffer */
+ struct xfs_buf *bp; /* block buffer */
xfs_dir2_block_tail_t *btp; /* block tail pointer */
- char *buf; /* sf buffer */
- int buf_len;
xfs_dir2_data_entry_t *dep; /* data entry pointer */
xfs_inode_t *dp; /* incore directory inode */
int dummy; /* trash */
@@ -921,7 +975,8 @@ xfs_dir2_sf_to_block(
int newoffset; /* offset from current entry */
int offset; /* target block offset */
xfs_dir2_sf_entry_t *sfep; /* sf entry pointer */
- xfs_dir2_sf_t *sfp; /* shortform structure */
+ xfs_dir2_sf_hdr_t *oldsfp; /* old shortform header */
+ xfs_dir2_sf_hdr_t *sfp; /* shortform header */
__be16 *tagp; /* end of data entry */
xfs_trans_t *tp; /* transaction pointer */
struct xfs_name name;
@@ -939,32 +994,30 @@ xfs_dir2_sf_to_block(
ASSERT(XFS_FORCED_SHUTDOWN(mp));
return XFS_ERROR(EIO);
}
+
+ oldsfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
+
ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
ASSERT(dp->i_df.if_u1.if_data != NULL);
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
- ASSERT(dp->i_d.di_size >= xfs_dir2_sf_hdr_size(sfp->hdr.i8count));
+ ASSERT(dp->i_d.di_size >= xfs_dir2_sf_hdr_size(oldsfp->i8count));
+
/*
- * Copy the directory into the stack buffer.
+ * Copy the directory into a temporary buffer.
* Then pitch the incore inode data so we can make extents.
*/
+ sfp = kmem_alloc(dp->i_df.if_bytes, KM_SLEEP);
+ memcpy(sfp, oldsfp, dp->i_df.if_bytes);
- buf_len = dp->i_df.if_bytes;
- buf = kmem_alloc(buf_len, KM_SLEEP);
-
- memcpy(buf, sfp, buf_len);
- xfs_idata_realloc(dp, -buf_len, XFS_DATA_FORK);
+ xfs_idata_realloc(dp, -dp->i_df.if_bytes, XFS_DATA_FORK);
dp->i_d.di_size = 0;
xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE);
- /*
- * Reset pointer - old sfp is gone.
- */
- sfp = (xfs_dir2_sf_t *)buf;
+
/*
* Add block 0 to the inode.
*/
error = xfs_dir2_grow_inode(args, XFS_DIR2_DATA_SPACE, &blkno);
if (error) {
- kmem_free(buf);
+ kmem_free(sfp);
return error;
}
/*
@@ -972,21 +1025,22 @@ xfs_dir2_sf_to_block(
*/
error = xfs_dir2_data_init(args, blkno, &bp);
if (error) {
- kmem_free(buf);
+ kmem_free(sfp);
return error;
}
- block = bp->data;
- block->hdr.magic = cpu_to_be32(XFS_DIR2_BLOCK_MAGIC);
+ bp->b_ops = &xfs_dir2_block_buf_ops;
+ hdr = bp->b_addr;
+ hdr->magic = cpu_to_be32(XFS_DIR2_BLOCK_MAGIC);
/*
* Compute size of block "tail" area.
*/
i = (uint)sizeof(*btp) +
- (sfp->hdr.count + 2) * (uint)sizeof(xfs_dir2_leaf_entry_t);
+ (sfp->count + 2) * (uint)sizeof(xfs_dir2_leaf_entry_t);
/*
* The whole thing is initialized to free by the init routine.
* Say we're using the leaf and tail area.
*/
- dup = (xfs_dir2_data_unused_t *)block->u;
+ dup = (xfs_dir2_data_unused_t *)(hdr + 1);
needlog = needscan = 0;
xfs_dir2_data_use_free(tp, bp, dup, mp->m_dirblksize - i, i, &needlog,
&needscan);
@@ -994,50 +1048,51 @@ xfs_dir2_sf_to_block(
/*
* Fill in the tail.
*/
- btp = xfs_dir2_block_tail_p(mp, block);
- btp->count = cpu_to_be32(sfp->hdr.count + 2); /* ., .. */
+ btp = xfs_dir2_block_tail_p(mp, hdr);
+ btp->count = cpu_to_be32(sfp->count + 2); /* ., .. */
btp->stale = 0;
blp = xfs_dir2_block_leaf_p(btp);
- endoffset = (uint)((char *)blp - (char *)block);
+ endoffset = (uint)((char *)blp - (char *)hdr);
/*
* Remove the freespace, we'll manage it.
*/
xfs_dir2_data_use_free(tp, bp, dup,
- (xfs_dir2_data_aoff_t)((char *)dup - (char *)block),
+ (xfs_dir2_data_aoff_t)((char *)dup - (char *)hdr),
be16_to_cpu(dup->length), &needlog, &needscan);
/*
* Create entry for .
*/
dep = (xfs_dir2_data_entry_t *)
- ((char *)block + XFS_DIR2_DATA_DOT_OFFSET);
+ ((char *)hdr + XFS_DIR2_DATA_DOT_OFFSET);
dep->inumber = cpu_to_be64(dp->i_ino);
dep->namelen = 1;
dep->name[0] = '.';
tagp = xfs_dir2_data_entry_tag_p(dep);
- *tagp = cpu_to_be16((char *)dep - (char *)block);
+ *tagp = cpu_to_be16((char *)dep - (char *)hdr);
xfs_dir2_data_log_entry(tp, bp, dep);
blp[0].hashval = cpu_to_be32(xfs_dir_hash_dot);
blp[0].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp,
- (char *)dep - (char *)block));
+ (char *)dep - (char *)hdr));
/*
* Create entry for ..
*/
dep = (xfs_dir2_data_entry_t *)
- ((char *)block + XFS_DIR2_DATA_DOTDOT_OFFSET);
- dep->inumber = cpu_to_be64(xfs_dir2_sf_get_inumber(sfp, &sfp->hdr.parent));
+ ((char *)hdr + XFS_DIR2_DATA_DOTDOT_OFFSET);
+ dep->inumber = cpu_to_be64(xfs_dir2_sf_get_parent_ino(sfp));
dep->namelen = 2;
dep->name[0] = dep->name[1] = '.';
tagp = xfs_dir2_data_entry_tag_p(dep);
- *tagp = cpu_to_be16((char *)dep - (char *)block);
+ *tagp = cpu_to_be16((char *)dep - (char *)hdr);
xfs_dir2_data_log_entry(tp, bp, dep);
blp[1].hashval = cpu_to_be32(xfs_dir_hash_dotdot);
blp[1].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp,
- (char *)dep - (char *)block));
+ (char *)dep - (char *)hdr));
offset = XFS_DIR2_DATA_FIRST_OFFSET;
/*
* Loop over existing entries, stuff them in.
*/
- if ((i = 0) == sfp->hdr.count)
+ i = 0;
+ if (!sfp->count)
sfep = NULL;
else
sfep = xfs_dir2_sf_firstentry(sfp);
@@ -1057,43 +1112,40 @@ xfs_dir2_sf_to_block(
* There should be a hole here, make one.
*/
if (offset < newoffset) {
- dup = (xfs_dir2_data_unused_t *)
- ((char *)block + offset);
+ dup = (xfs_dir2_data_unused_t *)((char *)hdr + offset);
dup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
dup->length = cpu_to_be16(newoffset - offset);
*xfs_dir2_data_unused_tag_p(dup) = cpu_to_be16(
- ((char *)dup - (char *)block));
+ ((char *)dup - (char *)hdr));
xfs_dir2_data_log_unused(tp, bp, dup);
- (void)xfs_dir2_data_freeinsert((xfs_dir2_data_t *)block,
- dup, &dummy);
+ xfs_dir2_data_freeinsert(hdr, dup, &dummy);
offset += be16_to_cpu(dup->length);
continue;
}
/*
* Copy a real entry.
*/
- dep = (xfs_dir2_data_entry_t *)((char *)block + newoffset);
- dep->inumber = cpu_to_be64(xfs_dir2_sf_get_inumber(sfp,
- xfs_dir2_sf_inumberp(sfep)));
+ dep = (xfs_dir2_data_entry_t *)((char *)hdr + newoffset);
+ dep->inumber = cpu_to_be64(xfs_dir2_sfe_get_ino(sfp, sfep));
dep->namelen = sfep->namelen;
memcpy(dep->name, sfep->name, dep->namelen);
tagp = xfs_dir2_data_entry_tag_p(dep);
- *tagp = cpu_to_be16((char *)dep - (char *)block);
+ *tagp = cpu_to_be16((char *)dep - (char *)hdr);
xfs_dir2_data_log_entry(tp, bp, dep);
name.name = sfep->name;
name.len = sfep->namelen;
blp[2 + i].hashval = cpu_to_be32(mp->m_dirnameops->
hashname(&name));
blp[2 + i].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp,
- (char *)dep - (char *)block));
- offset = (int)((char *)(tagp + 1) - (char *)block);
- if (++i == sfp->hdr.count)
+ (char *)dep - (char *)hdr));
+ offset = (int)((char *)(tagp + 1) - (char *)hdr);
+ if (++i == sfp->count)
sfep = NULL;
else
sfep = xfs_dir2_sf_nextentry(sfp, sfep);
}
/* Done with the temporary buffer */
- kmem_free(buf);
+ kmem_free(sfp);
/*
* Sort the leaf entries by hash value.
*/
@@ -1106,6 +1158,5 @@ xfs_dir2_sf_to_block(
xfs_dir2_block_log_leaf(tp, bp, 0, be32_to_cpu(btp->count) - 1);
xfs_dir2_block_log_tail(tp, bp);
xfs_dir2_data_check(dp, bp);
- xfs_da_buf_done(bp);
return 0;
}
diff --git a/libxfs/xfs_dir2_data.c b/libxfs/xfs_dir2_data.c
index d89b5b1..eb86739 100644
--- a/libxfs/xfs_dir2_data.c
+++ b/libxfs/xfs_dir2_data.c
@@ -18,23 +18,21 @@
#include <xfs.h>
-
-#ifdef DEBUG
/*
* Check the consistency of the data block.
* The input can also be a block-format directory.
- * Pop an assert if we find anything bad.
+ * Return 0 is the buffer is good, otherwise an error.
*/
-void
-xfs_dir2_data_check(
- xfs_inode_t *dp, /* incore inode pointer */
- xfs_dabuf_t *bp) /* data block's buffer */
+int
+__xfs_dir2_data_check(
+ struct xfs_inode *dp, /* incore inode pointer */
+ struct xfs_buf *bp) /* data block's buffer */
{
xfs_dir2_dataptr_t addr; /* addr for leaf lookup */
xfs_dir2_data_free_t *bf; /* bestfree table */
xfs_dir2_block_tail_t *btp=NULL; /* block tail */
int count; /* count of entries found */
- xfs_dir2_data_t *d; /* data block pointer */
+ xfs_dir2_data_hdr_t *hdr; /* data block header */
xfs_dir2_data_entry_t *dep; /* data entry */
xfs_dir2_data_free_t *dfp; /* bestfree entry */
xfs_dir2_data_unused_t *dup; /* unused entry */
@@ -49,36 +47,46 @@ xfs_dir2_data_check(
int stale; /* count of stale leaves */
struct xfs_name name;
- mp = dp->i_mount;
- d = bp->data;
- ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
- be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
- bf = d->hdr.bestfree;
- p = (char *)d->u;
- if (be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC) {
- btp = xfs_dir2_block_tail_p(mp, (xfs_dir2_block_t *)d);
+ mp = bp->b_target->bt_mount;
+ hdr = bp->b_addr;
+ bf = hdr->bestfree;
+ p = (char *)(hdr + 1);
+
+ switch (be32_to_cpu(hdr->magic)) {
+ case XFS_DIR2_BLOCK_MAGIC:
+ btp = xfs_dir2_block_tail_p(mp, hdr);
lep = xfs_dir2_block_leaf_p(btp);
endp = (char *)lep;
- } else
- endp = (char *)d + mp->m_dirblksize;
+ break;
+ case XFS_DIR2_DATA_MAGIC:
+ endp = (char *)hdr + mp->m_dirblksize;
+ break;
+ default:
+ XFS_ERROR_REPORT("Bad Magic", XFS_ERRLEVEL_LOW, mp);
+ return EFSCORRUPTED;
+ }
+
count = lastfree = freeseen = 0;
/*
* Account for zero bestfree entries.
*/
if (!bf[0].length) {
- ASSERT(!bf[0].offset);
+ XFS_WANT_CORRUPTED_RETURN(!bf[0].offset);
freeseen |= 1 << 0;
}
if (!bf[1].length) {
- ASSERT(!bf[1].offset);
+ XFS_WANT_CORRUPTED_RETURN(!bf[1].offset);
freeseen |= 1 << 1;
}
if (!bf[2].length) {
- ASSERT(!bf[2].offset);
+ XFS_WANT_CORRUPTED_RETURN(!bf[2].offset);
freeseen |= 1 << 2;
}
- ASSERT(be16_to_cpu(bf[0].length) >= be16_to_cpu(bf[1].length));
- ASSERT(be16_to_cpu(bf[1].length) >= be16_to_cpu(bf[2].length));
+
+ XFS_WANT_CORRUPTED_RETURN(be16_to_cpu(bf[0].length) >=
+ be16_to_cpu(bf[1].length));
+ XFS_WANT_CORRUPTED_RETURN(be16_to_cpu(bf[1].length) >=
+ be16_to_cpu(bf[2].length));
/*
* Loop over the data/unused entries.
*/
@@ -90,17 +98,20 @@ xfs_dir2_data_check(
* doesn't need to be there.
*/
if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
- ASSERT(lastfree == 0);
- ASSERT(be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup)) ==
- (char *)dup - (char *)d);
- dfp = xfs_dir2_data_freefind(d, dup);
+ XFS_WANT_CORRUPTED_RETURN(lastfree == 0);
+ XFS_WANT_CORRUPTED_RETURN(
+ be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup)) ==
+ (char *)dup - (char *)hdr);
+ dfp = xfs_dir2_data_freefind(hdr, dup);
if (dfp) {
i = (int)(dfp - bf);
- ASSERT((freeseen & (1 << i)) == 0);
+ XFS_WANT_CORRUPTED_RETURN(
+ (freeseen & (1 << i)) == 0);
freeseen |= 1 << i;
} else {
- ASSERT(be16_to_cpu(dup->length) <=
- be16_to_cpu(bf[2].length));
+ XFS_WANT_CORRUPTED_RETURN(
+ be16_to_cpu(dup->length) <=
+ be16_to_cpu(bf[2].length));
}
p += be16_to_cpu(dup->length);
lastfree = 1;
@@ -113,16 +124,18 @@ xfs_dir2_data_check(
* The linear search is crude but this is DEBUG code.
*/
dep = (xfs_dir2_data_entry_t *)p;
- ASSERT(dep->namelen != 0);
- ASSERT(xfs_dir_ino_validate(mp, be64_to_cpu(dep->inumber)) == 0);
- ASSERT(be16_to_cpu(*xfs_dir2_data_entry_tag_p(dep)) ==
- (char *)dep - (char *)d);
+ XFS_WANT_CORRUPTED_RETURN(dep->namelen != 0);
+ XFS_WANT_CORRUPTED_RETURN(
+ !xfs_dir_ino_validate(mp, be64_to_cpu(dep->inumber)));
+ XFS_WANT_CORRUPTED_RETURN(
+ be16_to_cpu(*xfs_dir2_data_entry_tag_p(dep)) ==
+ (char *)dep - (char *)hdr);
count++;
lastfree = 0;
- if (be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC) {
+ if (hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC)) {
addr = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk,
(xfs_dir2_data_aoff_t)
- ((char *)dep - (char *)d));
+ ((char *)dep - (char *)hdr));
name.name = dep->name;
name.len = dep->namelen;
hash = mp->m_dirnameops->hashname(&name);
@@ -131,26 +144,122 @@ xfs_dir2_data_check(
be32_to_cpu(lep[i].hashval) == hash)
break;
}
- ASSERT(i < be32_to_cpu(btp->count));
+ XFS_WANT_CORRUPTED_RETURN(i < be32_to_cpu(btp->count));
}
p += xfs_dir2_data_entsize(dep->namelen);
}
/*
* Need to have seen all the entries and all the bestfree slots.
*/
- ASSERT(freeseen == 7);
- if (be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC) {
+ XFS_WANT_CORRUPTED_RETURN(freeseen == 7);
+ if (hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC)) {
for (i = stale = 0; i < be32_to_cpu(btp->count); i++) {
- if (be32_to_cpu(lep[i].address) == XFS_DIR2_NULL_DATAPTR)
+ if (lep[i].address ==
+ cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
stale++;
if (i > 0)
- ASSERT(be32_to_cpu(lep[i].hashval) >= be32_to_cpu(lep[i - 1].hashval));
+ XFS_WANT_CORRUPTED_RETURN(
+ be32_to_cpu(lep[i].hashval) >=
+ be32_to_cpu(lep[i - 1].hashval));
}
- ASSERT(count == be32_to_cpu(btp->count) - be32_to_cpu(btp->stale));
- ASSERT(stale == be32_to_cpu(btp->stale));
+ XFS_WANT_CORRUPTED_RETURN(count ==
+ be32_to_cpu(btp->count) - be32_to_cpu(btp->stale));
+ XFS_WANT_CORRUPTED_RETURN(stale == be32_to_cpu(btp->stale));
+ }
+ return 0;
+}
+
+static void
+xfs_dir2_data_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_dir2_data_hdr *hdr = bp->b_addr;
+ int block_ok = 0;
+
+ block_ok = hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC);
+ block_ok = block_ok && __xfs_dir2_data_check(NULL, bp) == 0;
+
+ if (!block_ok) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, hdr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+}
+
+/*
+ * Readahead of the first block of the directory when it is opened is completely
+ * oblivious to the format of the directory. Hence we can either get a block
+ * format buffer or a data format buffer on readahead.
+ */
+static void
+xfs_dir2_data_reada_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_dir2_data_hdr *hdr = bp->b_addr;
+
+ switch (be32_to_cpu(hdr->magic)) {
+ case XFS_DIR2_BLOCK_MAGIC:
+ bp->b_ops = &xfs_dir2_block_buf_ops;
+ bp->b_ops->verify_read(bp);
+ return;
+ case XFS_DIR2_DATA_MAGIC:
+ xfs_dir2_data_verify(bp);
+ return;
+ default:
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, hdr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ break;
}
}
-#endif
+
+static void
+xfs_dir2_data_read_verify(
+ struct xfs_buf *bp)
+{
+ xfs_dir2_data_verify(bp);
+}
+
+static void
+xfs_dir2_data_write_verify(
+ struct xfs_buf *bp)
+{
+ xfs_dir2_data_verify(bp);
+}
+
+const struct xfs_buf_ops xfs_dir2_data_buf_ops = {
+ .verify_read = xfs_dir2_data_read_verify,
+ .verify_write = xfs_dir2_data_write_verify,
+};
+
+static const struct xfs_buf_ops xfs_dir2_data_reada_buf_ops = {
+ .verify_read = xfs_dir2_data_reada_verify,
+ .verify_write = xfs_dir2_data_write_verify,
+};
+
+
+int
+xfs_dir2_data_read(
+ struct xfs_trans *tp,
+ struct xfs_inode *dp,
+ xfs_dablk_t bno,
+ xfs_daddr_t mapped_bno,
+ struct xfs_buf **bpp)
+{
+ return xfs_da_read_buf(tp, dp, bno, mapped_bno, bpp,
+ XFS_DATA_FORK, &xfs_dir2_data_buf_ops);
+}
+
+int
+xfs_dir2_data_readahead(
+ struct xfs_trans *tp,
+ struct xfs_inode *dp,
+ xfs_dablk_t bno,
+ xfs_daddr_t mapped_bno)
+{
+ return xfs_da_reada_buf(tp, dp, bno, mapped_bno,
+ XFS_DATA_FORK, &xfs_dir2_data_reada_buf_ops);
+}
/*
* Given a data block and an unused entry from that block,
@@ -158,7 +267,7 @@ xfs_dir2_data_check(
*/
xfs_dir2_data_free_t *
xfs_dir2_data_freefind(
- xfs_dir2_data_t *d, /* data block */
+ xfs_dir2_data_hdr_t *hdr, /* data block */
xfs_dir2_data_unused_t *dup) /* data unused entry */
{
xfs_dir2_data_free_t *dfp; /* bestfree entry */
@@ -168,17 +277,17 @@ xfs_dir2_data_freefind(
int seenzero; /* saw a 0 bestfree entry */
#endif
- off = (xfs_dir2_data_aoff_t)((char *)dup - (char *)d);
+ off = (xfs_dir2_data_aoff_t)((char *)dup - (char *)hdr);
#if defined(DEBUG) && defined(__KERNEL__)
/*
* Validate some consistency in the bestfree table.
* Check order, non-overlapping entries, and if we find the
* one we're looking for it has to be exact.
*/
- ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
- be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
- for (dfp = &d->hdr.bestfree[0], seenzero = matched = 0;
- dfp < &d->hdr.bestfree[XFS_DIR2_DATA_FD_COUNT];
+ ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
+ for (dfp = &hdr->bestfree[0], seenzero = matched = 0;
+ dfp < &hdr->bestfree[XFS_DIR2_DATA_FD_COUNT];
dfp++) {
if (!dfp->offset) {
ASSERT(!dfp->length);
@@ -194,7 +303,7 @@ xfs_dir2_data_freefind(
else
ASSERT(be16_to_cpu(dfp->offset) + be16_to_cpu(dfp->length) <= off);
ASSERT(matched || be16_to_cpu(dfp->length) >= be16_to_cpu(dup->length));
- if (dfp > &d->hdr.bestfree[0])
+ if (dfp > &hdr->bestfree[0])
ASSERT(be16_to_cpu(dfp[-1].length) >= be16_to_cpu(dfp[0].length));
}
#endif
@@ -203,13 +312,13 @@ xfs_dir2_data_freefind(
* it can't be there since they're sorted.
*/
if (be16_to_cpu(dup->length) <
- be16_to_cpu(d->hdr.bestfree[XFS_DIR2_DATA_FD_COUNT - 1].length))
+ be16_to_cpu(hdr->bestfree[XFS_DIR2_DATA_FD_COUNT - 1].length))
return NULL;
/*
* Look at the three bestfree entries for our guy.
*/
- for (dfp = &d->hdr.bestfree[0];
- dfp < &d->hdr.bestfree[XFS_DIR2_DATA_FD_COUNT];
+ for (dfp = &hdr->bestfree[0];
+ dfp < &hdr->bestfree[XFS_DIR2_DATA_FD_COUNT];
dfp++) {
if (!dfp->offset)
return NULL;
@@ -227,7 +336,7 @@ xfs_dir2_data_freefind(
*/
xfs_dir2_data_free_t * /* entry inserted */
xfs_dir2_data_freeinsert(
- xfs_dir2_data_t *d, /* data block pointer */
+ xfs_dir2_data_hdr_t *hdr, /* data block pointer */
xfs_dir2_data_unused_t *dup, /* unused space */
int *loghead) /* log the data header (out) */
{
@@ -235,12 +344,13 @@ xfs_dir2_data_freeinsert(
xfs_dir2_data_free_t new; /* new bestfree entry */
#ifdef __KERNEL__
- ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
- be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
+ ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
#endif
- dfp = d->hdr.bestfree;
+ dfp = hdr->bestfree;
new.length = dup->length;
- new.offset = cpu_to_be16((char *)dup - (char *)d);
+ new.offset = cpu_to_be16((char *)dup - (char *)hdr);
+
/*
* Insert at position 0, 1, or 2; or not at all.
*/
@@ -270,36 +380,36 @@ xfs_dir2_data_freeinsert(
*/
STATIC void
xfs_dir2_data_freeremove(
- xfs_dir2_data_t *d, /* data block pointer */
+ xfs_dir2_data_hdr_t *hdr, /* data block header */
xfs_dir2_data_free_t *dfp, /* bestfree entry pointer */
int *loghead) /* out: log data header */
{
#ifdef __KERNEL__
- ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
- be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
+ ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
#endif
/*
* It's the first entry, slide the next 2 up.
*/
- if (dfp == &d->hdr.bestfree[0]) {
- d->hdr.bestfree[0] = d->hdr.bestfree[1];
- d->hdr.bestfree[1] = d->hdr.bestfree[2];
+ if (dfp == &hdr->bestfree[0]) {
+ hdr->bestfree[0] = hdr->bestfree[1];
+ hdr->bestfree[1] = hdr->bestfree[2];
}
/*
* It's the second entry, slide the 3rd entry up.
*/
- else if (dfp == &d->hdr.bestfree[1])
- d->hdr.bestfree[1] = d->hdr.bestfree[2];
+ else if (dfp == &hdr->bestfree[1])
+ hdr->bestfree[1] = hdr->bestfree[2];
/*
* Must be the last entry.
*/
else
- ASSERT(dfp == &d->hdr.bestfree[2]);
+ ASSERT(dfp == &hdr->bestfree[2]);
/*
* Clear the 3rd entry, must be zero now.
*/
- d->hdr.bestfree[2].length = 0;
- d->hdr.bestfree[2].offset = 0;
+ hdr->bestfree[2].length = 0;
+ hdr->bestfree[2].offset = 0;
*loghead = 1;
}
@@ -309,7 +419,7 @@ xfs_dir2_data_freeremove(
void
xfs_dir2_data_freescan(
xfs_mount_t *mp, /* filesystem mount point */
- xfs_dir2_data_t *d, /* data block pointer */
+ xfs_dir2_data_hdr_t *hdr, /* data block header */
int *loghead) /* out: log data header */
{
xfs_dir2_block_tail_t *btp; /* block tail */
@@ -319,23 +429,23 @@ xfs_dir2_data_freescan(
char *p; /* current entry pointer */
#ifdef __KERNEL__
- ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
- be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
+ ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
#endif
/*
* Start by clearing the table.
*/
- memset(d->hdr.bestfree, 0, sizeof(d->hdr.bestfree));
+ memset(hdr->bestfree, 0, sizeof(hdr->bestfree));
*loghead = 1;
/*
* Set up pointers.
*/
- p = (char *)d->u;
- if (be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC) {
- btp = xfs_dir2_block_tail_p(mp, (xfs_dir2_block_t *)d);
+ p = (char *)(hdr + 1);
+ if (hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC)) {
+ btp = xfs_dir2_block_tail_p(mp, hdr);
endp = (char *)xfs_dir2_block_leaf_p(btp);
} else
- endp = (char *)d + mp->m_dirblksize;
+ endp = (char *)hdr + mp->m_dirblksize;
/*
* Loop over the block's entries.
*/
@@ -345,9 +455,9 @@ xfs_dir2_data_freescan(
* If it's a free entry, insert it.
*/
if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
- ASSERT((char *)dup - (char *)d ==
+ ASSERT((char *)dup - (char *)hdr ==
be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup)));
- xfs_dir2_data_freeinsert(d, dup, loghead);
+ xfs_dir2_data_freeinsert(hdr, dup, loghead);
p += be16_to_cpu(dup->length);
}
/*
@@ -355,7 +465,7 @@ xfs_dir2_data_freescan(
*/
else {
dep = (xfs_dir2_data_entry_t *)p;
- ASSERT((char *)dep - (char *)d ==
+ ASSERT((char *)dep - (char *)hdr ==
be16_to_cpu(*xfs_dir2_data_entry_tag_p(dep)));
p += xfs_dir2_data_entsize(dep->namelen);
}
@@ -370,10 +480,10 @@ int /* error */
xfs_dir2_data_init(
xfs_da_args_t *args, /* directory operation args */
xfs_dir2_db_t blkno, /* logical dir block number */
- xfs_dabuf_t **bpp) /* output block buffer */
+ struct xfs_buf **bpp) /* output block buffer */
{
- xfs_dabuf_t *bp; /* block buffer */
- xfs_dir2_data_t *d; /* pointer to block */
+ struct xfs_buf *bp; /* block buffer */
+ xfs_dir2_data_hdr_t *hdr; /* data block header */
xfs_inode_t *dp; /* incore directory inode */
xfs_dir2_data_unused_t *dup; /* unused entry pointer */
int error; /* error return value */
@@ -390,30 +500,31 @@ xfs_dir2_data_init(
*/
error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(mp, blkno), -1, &bp,
XFS_DATA_FORK);
- if (error) {
+ if (error)
return error;
- }
- ASSERT(bp != NULL);
+ bp->b_ops = &xfs_dir2_data_buf_ops;
+
/*
* Initialize the header.
*/
- d = bp->data;
- d->hdr.magic = cpu_to_be32(XFS_DIR2_DATA_MAGIC);
- d->hdr.bestfree[0].offset = cpu_to_be16(sizeof(d->hdr));
+ hdr = bp->b_addr;
+ hdr->magic = cpu_to_be32(XFS_DIR2_DATA_MAGIC);
+ hdr->bestfree[0].offset = cpu_to_be16(sizeof(*hdr));
for (i = 1; i < XFS_DIR2_DATA_FD_COUNT; i++) {
- d->hdr.bestfree[i].length = 0;
- d->hdr.bestfree[i].offset = 0;
+ hdr->bestfree[i].length = 0;
+ hdr->bestfree[i].offset = 0;
}
+
/*
* Set up an unused entry for the block's body.
*/
- dup = &d->u[0].unused;
+ dup = (xfs_dir2_data_unused_t *)(hdr + 1);
dup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
- t=mp->m_dirblksize - (uint)sizeof(d->hdr);
- d->hdr.bestfree[0].length = cpu_to_be16(t);
+ t = mp->m_dirblksize - (uint)sizeof(*hdr);
+ hdr->bestfree[0].length = cpu_to_be16(t);
dup->length = cpu_to_be16(t);
- *xfs_dir2_data_unused_tag_p(dup) = cpu_to_be16((char *)dup - (char *)d);
+ *xfs_dir2_data_unused_tag_p(dup) = cpu_to_be16((char *)dup - (char *)hdr);
/*
* Log it and return it.
*/
@@ -428,18 +539,18 @@ xfs_dir2_data_init(
*/
void
xfs_dir2_data_log_entry(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_dabuf_t *bp, /* block buffer */
+ struct xfs_trans *tp,
+ struct xfs_buf *bp,
xfs_dir2_data_entry_t *dep) /* data entry pointer */
{
- xfs_dir2_data_t *d; /* data block pointer */
+ xfs_dir2_data_hdr_t *hdr = bp->b_addr;
+
+ ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
- d = bp->data;
- ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
- be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
- xfs_da_log_buf(tp, bp, (uint)((char *)dep - (char *)d),
+ xfs_trans_log_buf(tp, bp, (uint)((char *)dep - (char *)hdr),
(uint)((char *)(xfs_dir2_data_entry_tag_p(dep) + 1) -
- (char *)d - 1));
+ (char *)hdr - 1));
}
/*
@@ -447,16 +558,15 @@ xfs_dir2_data_log_entry(
*/
void
xfs_dir2_data_log_header(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_dabuf_t *bp) /* block buffer */
+ struct xfs_trans *tp,
+ struct xfs_buf *bp)
{
- xfs_dir2_data_t *d; /* data block pointer */
+ xfs_dir2_data_hdr_t *hdr = bp->b_addr;
+
+ ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
- d = bp->data;
- ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
- be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
- xfs_da_log_buf(tp, bp, (uint)((char *)&d->hdr - (char *)d),
- (uint)(sizeof(d->hdr) - 1));
+ xfs_trans_log_buf(tp, bp, 0, sizeof(*hdr) - 1);
}
/*
@@ -464,27 +574,27 @@ xfs_dir2_data_log_header(
*/
void
xfs_dir2_data_log_unused(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_dabuf_t *bp, /* block buffer */
+ struct xfs_trans *tp,
+ struct xfs_buf *bp,
xfs_dir2_data_unused_t *dup) /* data unused pointer */
{
- xfs_dir2_data_t *d; /* data block pointer */
+ xfs_dir2_data_hdr_t *hdr = bp->b_addr;
+
+ ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
- d = bp->data;
- ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
- be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
/*
* Log the first part of the unused entry.
*/
- xfs_da_log_buf(tp, bp, (uint)((char *)dup - (char *)d),
+ xfs_trans_log_buf(tp, bp, (uint)((char *)dup - (char *)hdr),
(uint)((char *)&dup->length + sizeof(dup->length) -
- 1 - (char *)d));
+ 1 - (char *)hdr));
/*
* Log the end (tag) of the unused entry.
*/
- xfs_da_log_buf(tp, bp,
- (uint)((char *)xfs_dir2_data_unused_tag_p(dup) - (char *)d),
- (uint)((char *)xfs_dir2_data_unused_tag_p(dup) - (char *)d +
+ xfs_trans_log_buf(tp, bp,
+ (uint)((char *)xfs_dir2_data_unused_tag_p(dup) - (char *)hdr),
+ (uint)((char *)xfs_dir2_data_unused_tag_p(dup) - (char *)hdr +
sizeof(xfs_dir2_data_off_t) - 1));
}
@@ -494,14 +604,14 @@ xfs_dir2_data_log_unused(
*/
void
xfs_dir2_data_make_free(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_dabuf_t *bp, /* block buffer */
+ struct xfs_trans *tp,
+ struct xfs_buf *bp,
xfs_dir2_data_aoff_t offset, /* starting byte offset */
xfs_dir2_data_aoff_t len, /* length in bytes */
int *needlogp, /* out: log header */
int *needscanp) /* out: regen bestfree */
{
- xfs_dir2_data_t *d; /* data block pointer */
+ xfs_dir2_data_hdr_t *hdr; /* data block pointer */
xfs_dir2_data_free_t *dfp; /* bestfree pointer */
char *endptr; /* end of data area */
xfs_mount_t *mp; /* filesystem mount point */
@@ -511,28 +621,29 @@ xfs_dir2_data_make_free(
xfs_dir2_data_unused_t *prevdup; /* unused entry before us */
mp = tp->t_mountp;
- d = bp->data;
+ hdr = bp->b_addr;
+
/*
* Figure out where the end of the data area is.
*/
- if (be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC)
- endptr = (char *)d + mp->m_dirblksize;
+ if (hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC))
+ endptr = (char *)hdr + mp->m_dirblksize;
else {
xfs_dir2_block_tail_t *btp; /* block tail */
- ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
- btp = xfs_dir2_block_tail_p(mp, (xfs_dir2_block_t *)d);
+ ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
+ btp = xfs_dir2_block_tail_p(mp, hdr);
endptr = (char *)xfs_dir2_block_leaf_p(btp);
}
/*
* If this isn't the start of the block, then back up to
* the previous entry and see if it's free.
*/
- if (offset > sizeof(d->hdr)) {
+ if (offset > sizeof(*hdr)) {
__be16 *tagp; /* tag just before us */
- tagp = (__be16 *)((char *)d + offset) - 1;
- prevdup = (xfs_dir2_data_unused_t *)((char *)d + be16_to_cpu(*tagp));
+ tagp = (__be16 *)((char *)hdr + offset) - 1;
+ prevdup = (xfs_dir2_data_unused_t *)((char *)hdr + be16_to_cpu(*tagp));
if (be16_to_cpu(prevdup->freetag) != XFS_DIR2_DATA_FREE_TAG)
prevdup = NULL;
} else
@@ -541,9 +652,9 @@ xfs_dir2_data_make_free(
* If this isn't the end of the block, see if the entry after
* us is free.
*/
- if ((char *)d + offset + len < endptr) {
+ if ((char *)hdr + offset + len < endptr) {
postdup =
- (xfs_dir2_data_unused_t *)((char *)d + offset + len);
+ (xfs_dir2_data_unused_t *)((char *)hdr + offset + len);
if (be16_to_cpu(postdup->freetag) != XFS_DIR2_DATA_FREE_TAG)
postdup = NULL;
} else
@@ -560,21 +671,21 @@ xfs_dir2_data_make_free(
/*
* See if prevdup and/or postdup are in bestfree table.
*/
- dfp = xfs_dir2_data_freefind(d, prevdup);
- dfp2 = xfs_dir2_data_freefind(d, postdup);
+ dfp = xfs_dir2_data_freefind(hdr, prevdup);
+ dfp2 = xfs_dir2_data_freefind(hdr, postdup);
/*
* We need a rescan unless there are exactly 2 free entries
* namely our two. Then we know what's happening, otherwise
* since the third bestfree is there, there might be more
* entries.
*/
- needscan = (d->hdr.bestfree[2].length != 0);
+ needscan = (hdr->bestfree[2].length != 0);
/*
* Fix up the new big freespace.
*/
be16_add_cpu(&prevdup->length, len + be16_to_cpu(postdup->length));
*xfs_dir2_data_unused_tag_p(prevdup) =
- cpu_to_be16((char *)prevdup - (char *)d);
+ cpu_to_be16((char *)prevdup - (char *)hdr);
xfs_dir2_data_log_unused(tp, bp, prevdup);
if (!needscan) {
/*
@@ -584,18 +695,18 @@ xfs_dir2_data_make_free(
* Remove entry 1 first then entry 0.
*/
ASSERT(dfp && dfp2);
- if (dfp == &d->hdr.bestfree[1]) {
- dfp = &d->hdr.bestfree[0];
+ if (dfp == &hdr->bestfree[1]) {
+ dfp = &hdr->bestfree[0];
ASSERT(dfp2 == dfp);
- dfp2 = &d->hdr.bestfree[1];
+ dfp2 = &hdr->bestfree[1];
}
- xfs_dir2_data_freeremove(d, dfp2, needlogp);
- xfs_dir2_data_freeremove(d, dfp, needlogp);
+ xfs_dir2_data_freeremove(hdr, dfp2, needlogp);
+ xfs_dir2_data_freeremove(hdr, dfp, needlogp);
/*
* Now insert the new entry.
*/
- dfp = xfs_dir2_data_freeinsert(d, prevdup, needlogp);
- ASSERT(dfp == &d->hdr.bestfree[0]);
+ dfp = xfs_dir2_data_freeinsert(hdr, prevdup, needlogp);
+ ASSERT(dfp == &hdr->bestfree[0]);
ASSERT(dfp->length == prevdup->length);
ASSERT(!dfp[1].length);
ASSERT(!dfp[2].length);
@@ -605,10 +716,10 @@ xfs_dir2_data_make_free(
* The entry before us is free, merge with it.
*/
else if (prevdup) {
- dfp = xfs_dir2_data_freefind(d, prevdup);
+ dfp = xfs_dir2_data_freefind(hdr, prevdup);
be16_add_cpu(&prevdup->length, len);
*xfs_dir2_data_unused_tag_p(prevdup) =
- cpu_to_be16((char *)prevdup - (char *)d);
+ cpu_to_be16((char *)prevdup - (char *)hdr);
xfs_dir2_data_log_unused(tp, bp, prevdup);
/*
* If the previous entry was in the table, the new entry
@@ -616,27 +727,27 @@ xfs_dir2_data_make_free(
* the old one and add the new one.
*/
if (dfp) {
- xfs_dir2_data_freeremove(d, dfp, needlogp);
- (void)xfs_dir2_data_freeinsert(d, prevdup, needlogp);
+ xfs_dir2_data_freeremove(hdr, dfp, needlogp);
+ xfs_dir2_data_freeinsert(hdr, prevdup, needlogp);
}
/*
* Otherwise we need a scan if the new entry is big enough.
*/
else {
needscan = be16_to_cpu(prevdup->length) >
- be16_to_cpu(d->hdr.bestfree[2].length);
+ be16_to_cpu(hdr->bestfree[2].length);
}
}
/*
* The following entry is free, merge with it.
*/
else if (postdup) {
- dfp = xfs_dir2_data_freefind(d, postdup);
- newdup = (xfs_dir2_data_unused_t *)((char *)d + offset);
+ dfp = xfs_dir2_data_freefind(hdr, postdup);
+ newdup = (xfs_dir2_data_unused_t *)((char *)hdr + offset);
newdup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
newdup->length = cpu_to_be16(len + be16_to_cpu(postdup->length));
*xfs_dir2_data_unused_tag_p(newdup) =
- cpu_to_be16((char *)newdup - (char *)d);
+ cpu_to_be16((char *)newdup - (char *)hdr);
xfs_dir2_data_log_unused(tp, bp, newdup);
/*
* If the following entry was in the table, the new entry
@@ -644,28 +755,28 @@ xfs_dir2_data_make_free(
* the old one and add the new one.
*/
if (dfp) {
- xfs_dir2_data_freeremove(d, dfp, needlogp);
- (void)xfs_dir2_data_freeinsert(d, newdup, needlogp);
+ xfs_dir2_data_freeremove(hdr, dfp, needlogp);
+ xfs_dir2_data_freeinsert(hdr, newdup, needlogp);
}
/*
* Otherwise we need a scan if the new entry is big enough.
*/
else {
needscan = be16_to_cpu(newdup->length) >
- be16_to_cpu(d->hdr.bestfree[2].length);
+ be16_to_cpu(hdr->bestfree[2].length);
}
}
/*
* Neither neighbor is free. Make a new entry.
*/
else {
- newdup = (xfs_dir2_data_unused_t *)((char *)d + offset);
+ newdup = (xfs_dir2_data_unused_t *)((char *)hdr + offset);
newdup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
newdup->length = cpu_to_be16(len);
*xfs_dir2_data_unused_tag_p(newdup) =
- cpu_to_be16((char *)newdup - (char *)d);
+ cpu_to_be16((char *)newdup - (char *)hdr);
xfs_dir2_data_log_unused(tp, bp, newdup);
- (void)xfs_dir2_data_freeinsert(d, newdup, needlogp);
+ xfs_dir2_data_freeinsert(hdr, newdup, needlogp);
}
*needscanp = needscan;
}
@@ -675,15 +786,15 @@ xfs_dir2_data_make_free(
*/
void
xfs_dir2_data_use_free(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_dabuf_t *bp, /* data block buffer */
+ struct xfs_trans *tp,
+ struct xfs_buf *bp,
xfs_dir2_data_unused_t *dup, /* unused entry */
xfs_dir2_data_aoff_t offset, /* starting offset to use */
xfs_dir2_data_aoff_t len, /* length to use */
int *needlogp, /* out: need to log header */
int *needscanp) /* out: need regen bestfree */
{
- xfs_dir2_data_t *d; /* data block */
+ xfs_dir2_data_hdr_t *hdr; /* data block header */
xfs_dir2_data_free_t *dfp; /* bestfree pointer */
int matchback; /* matches end of freespace */
int matchfront; /* matches start of freespace */
@@ -692,24 +803,24 @@ xfs_dir2_data_use_free(
xfs_dir2_data_unused_t *newdup2; /* another new unused entry */
int oldlen; /* old unused entry's length */
- d = bp->data;
- ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
- be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
+ hdr = bp->b_addr;
+ ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
ASSERT(be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG);
- ASSERT(offset >= (char *)dup - (char *)d);
- ASSERT(offset + len <= (char *)dup + be16_to_cpu(dup->length) - (char *)d);
- ASSERT((char *)dup - (char *)d == be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup)));
+ ASSERT(offset >= (char *)dup - (char *)hdr);
+ ASSERT(offset + len <= (char *)dup + be16_to_cpu(dup->length) - (char *)hdr);
+ ASSERT((char *)dup - (char *)hdr == be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup)));
/*
* Look up the entry in the bestfree table.
*/
- dfp = xfs_dir2_data_freefind(d, dup);
+ dfp = xfs_dir2_data_freefind(hdr, dup);
oldlen = be16_to_cpu(dup->length);
- ASSERT(dfp || oldlen <= be16_to_cpu(d->hdr.bestfree[2].length));
+ ASSERT(dfp || oldlen <= be16_to_cpu(hdr->bestfree[2].length));
/*
* Check for alignment with front and back of the entry.
*/
- matchfront = (char *)dup - (char *)d == offset;
- matchback = (char *)dup + oldlen - (char *)d == offset + len;
+ matchfront = (char *)dup - (char *)hdr == offset;
+ matchback = (char *)dup + oldlen - (char *)hdr == offset + len;
ASSERT(*needscanp == 0);
needscan = 0;
/*
@@ -718,9 +829,9 @@ xfs_dir2_data_use_free(
*/
if (matchfront && matchback) {
if (dfp) {
- needscan = (d->hdr.bestfree[2].offset != 0);
+ needscan = (hdr->bestfree[2].offset != 0);
if (!needscan)
- xfs_dir2_data_freeremove(d, dfp, needlogp);
+ xfs_dir2_data_freeremove(hdr, dfp, needlogp);
}
}
/*
@@ -728,27 +839,27 @@ xfs_dir2_data_use_free(
* Make a new entry with the remaining freespace.
*/
else if (matchfront) {
- newdup = (xfs_dir2_data_unused_t *)((char *)d + offset + len);
+ newdup = (xfs_dir2_data_unused_t *)((char *)hdr + offset + len);
newdup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
newdup->length = cpu_to_be16(oldlen - len);
*xfs_dir2_data_unused_tag_p(newdup) =
- cpu_to_be16((char *)newdup - (char *)d);
+ cpu_to_be16((char *)newdup - (char *)hdr);
xfs_dir2_data_log_unused(tp, bp, newdup);
/*
* If it was in the table, remove it and add the new one.
*/
if (dfp) {
- xfs_dir2_data_freeremove(d, dfp, needlogp);
- dfp = xfs_dir2_data_freeinsert(d, newdup, needlogp);
+ xfs_dir2_data_freeremove(hdr, dfp, needlogp);
+ dfp = xfs_dir2_data_freeinsert(hdr, newdup, needlogp);
ASSERT(dfp != NULL);
ASSERT(dfp->length == newdup->length);
- ASSERT(be16_to_cpu(dfp->offset) == (char *)newdup - (char *)d);
+ ASSERT(be16_to_cpu(dfp->offset) == (char *)newdup - (char *)hdr);
/*
* If we got inserted at the last slot,
* that means we don't know if there was a better
* choice for the last slot, or not. Rescan.
*/
- needscan = dfp == &d->hdr.bestfree[2];
+ needscan = dfp == &hdr->bestfree[2];
}
}
/*
@@ -757,25 +868,25 @@ xfs_dir2_data_use_free(
*/
else if (matchback) {
newdup = dup;
- newdup->length = cpu_to_be16(((char *)d + offset) - (char *)newdup);
+ newdup->length = cpu_to_be16(((char *)hdr + offset) - (char *)newdup);
*xfs_dir2_data_unused_tag_p(newdup) =
- cpu_to_be16((char *)newdup - (char *)d);
+ cpu_to_be16((char *)newdup - (char *)hdr);
xfs_dir2_data_log_unused(tp, bp, newdup);
/*
* If it was in the table, remove it and add the new one.
*/
if (dfp) {
- xfs_dir2_data_freeremove(d, dfp, needlogp);
- dfp = xfs_dir2_data_freeinsert(d, newdup, needlogp);
+ xfs_dir2_data_freeremove(hdr, dfp, needlogp);
+ dfp = xfs_dir2_data_freeinsert(hdr, newdup, needlogp);
ASSERT(dfp != NULL);
ASSERT(dfp->length == newdup->length);
- ASSERT(be16_to_cpu(dfp->offset) == (char *)newdup - (char *)d);
+ ASSERT(be16_to_cpu(dfp->offset) == (char *)newdup - (char *)hdr);
/*
* If we got inserted at the last slot,
* that means we don't know if there was a better
* choice for the last slot, or not. Rescan.
*/
- needscan = dfp == &d->hdr.bestfree[2];
+ needscan = dfp == &hdr->bestfree[2];
}
}
/*
@@ -784,15 +895,15 @@ xfs_dir2_data_use_free(
*/
else {
newdup = dup;
- newdup->length = cpu_to_be16(((char *)d + offset) - (char *)newdup);
+ newdup->length = cpu_to_be16(((char *)hdr + offset) - (char *)newdup);
*xfs_dir2_data_unused_tag_p(newdup) =
- cpu_to_be16((char *)newdup - (char *)d);
+ cpu_to_be16((char *)newdup - (char *)hdr);
xfs_dir2_data_log_unused(tp, bp, newdup);
- newdup2 = (xfs_dir2_data_unused_t *)((char *)d + offset + len);
+ newdup2 = (xfs_dir2_data_unused_t *)((char *)hdr + offset + len);
newdup2->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
newdup2->length = cpu_to_be16(oldlen - len - be16_to_cpu(newdup->length));
*xfs_dir2_data_unused_tag_p(newdup2) =
- cpu_to_be16((char *)newdup2 - (char *)d);
+ cpu_to_be16((char *)newdup2 - (char *)hdr);
xfs_dir2_data_log_unused(tp, bp, newdup2);
/*
* If the old entry was in the table, we need to scan
@@ -803,13 +914,12 @@ xfs_dir2_data_use_free(
* the 2 new will work.
*/
if (dfp) {
- needscan = (d->hdr.bestfree[2].length != 0);
+ needscan = (hdr->bestfree[2].length != 0);
if (!needscan) {
- xfs_dir2_data_freeremove(d, dfp, needlogp);
- (void)xfs_dir2_data_freeinsert(d, newdup,
- needlogp);
- (void)xfs_dir2_data_freeinsert(d, newdup2,
- needlogp);
+ xfs_dir2_data_freeremove(hdr, dfp, needlogp);
+ xfs_dir2_data_freeinsert(hdr, newdup, needlogp);
+ xfs_dir2_data_freeinsert(hdr, newdup2,
+ needlogp);
}
}
}
diff --git a/libxfs/xfs_dir2_leaf.c b/libxfs/xfs_dir2_leaf.c
index 9ce2320..d303813 100644
--- a/libxfs/xfs_dir2_leaf.c
+++ b/libxfs/xfs_dir2_leaf.c
@@ -22,16 +22,93 @@
* Local function declarations.
*/
#ifdef DEBUG
-static void xfs_dir2_leaf_check(xfs_inode_t *dp, xfs_dabuf_t *bp);
+static void xfs_dir2_leaf_check(struct xfs_inode *dp, struct xfs_buf *bp);
#else
#define xfs_dir2_leaf_check(dp, bp)
#endif
-static int xfs_dir2_leaf_lookup_int(xfs_da_args_t *args, xfs_dabuf_t **lbpp,
- int *indexp, xfs_dabuf_t **dbpp);
-static void xfs_dir2_leaf_log_bests(struct xfs_trans *tp, struct xfs_dabuf *bp,
+static int xfs_dir2_leaf_lookup_int(xfs_da_args_t *args, struct xfs_buf **lbpp,
+ int *indexp, struct xfs_buf **dbpp);
+static void xfs_dir2_leaf_log_bests(struct xfs_trans *tp, struct xfs_buf *bp,
int first, int last);
-static void xfs_dir2_leaf_log_tail(struct xfs_trans *tp, struct xfs_dabuf *bp);
+static void xfs_dir2_leaf_log_tail(struct xfs_trans *tp, struct xfs_buf *bp);
+static void
+xfs_dir2_leaf_verify(
+ struct xfs_buf *bp,
+ __be16 magic)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_dir2_leaf_hdr *hdr = bp->b_addr;
+ int block_ok = 0;
+
+ block_ok = hdr->info.magic == magic;
+ if (!block_ok) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, hdr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+}
+
+static void
+xfs_dir2_leaf1_read_verify(
+ struct xfs_buf *bp)
+{
+ xfs_dir2_leaf_verify(bp, cpu_to_be16(XFS_DIR2_LEAF1_MAGIC));
+}
+
+static void
+xfs_dir2_leaf1_write_verify(
+ struct xfs_buf *bp)
+{
+ xfs_dir2_leaf_verify(bp, cpu_to_be16(XFS_DIR2_LEAF1_MAGIC));
+}
+
+void
+xfs_dir2_leafn_read_verify(
+ struct xfs_buf *bp)
+{
+ xfs_dir2_leaf_verify(bp, cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
+}
+
+void
+xfs_dir2_leafn_write_verify(
+ struct xfs_buf *bp)
+{
+ xfs_dir2_leaf_verify(bp, cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
+}
+
+static const struct xfs_buf_ops xfs_dir2_leaf1_buf_ops = {
+ .verify_read = xfs_dir2_leaf1_read_verify,
+ .verify_write = xfs_dir2_leaf1_write_verify,
+};
+
+const struct xfs_buf_ops xfs_dir2_leafn_buf_ops = {
+ .verify_read = xfs_dir2_leafn_read_verify,
+ .verify_write = xfs_dir2_leafn_write_verify,
+};
+
+static int
+xfs_dir2_leaf_read(
+ struct xfs_trans *tp,
+ struct xfs_inode *dp,
+ xfs_dablk_t fbno,
+ xfs_daddr_t mappedbno,
+ struct xfs_buf **bpp)
+{
+ return xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
+ XFS_DATA_FORK, &xfs_dir2_leaf1_buf_ops);
+}
+
+int
+xfs_dir2_leafn_read(
+ struct xfs_trans *tp,
+ struct xfs_inode *dp,
+ xfs_dablk_t fbno,
+ xfs_daddr_t mappedbno,
+ struct xfs_buf **bpp)
+{
+ return xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
+ XFS_DATA_FORK, &xfs_dir2_leafn_buf_ops);
+}
/*
* Convert a block form directory to a leaf form directory.
@@ -39,16 +116,16 @@ static void xfs_dir2_leaf_log_tail(struct xfs_trans *tp, struct xfs_dabuf *bp);
int /* error */
xfs_dir2_block_to_leaf(
xfs_da_args_t *args, /* operation arguments */
- xfs_dabuf_t *dbp) /* input block's buffer */
+ struct xfs_buf *dbp) /* input block's buffer */
{
__be16 *bestsp; /* leaf's bestsp entries */
xfs_dablk_t blkno; /* leaf block's bno */
- xfs_dir2_block_t *block; /* block structure */
+ xfs_dir2_data_hdr_t *hdr; /* block header */
xfs_dir2_leaf_entry_t *blp; /* block's leaf entries */
xfs_dir2_block_tail_t *btp; /* block's tail */
xfs_inode_t *dp; /* incore directory inode */
int error; /* error return code */
- xfs_dabuf_t *lbp; /* leaf block's buffer */
+ struct xfs_buf *lbp; /* leaf block's buffer */
xfs_dir2_db_t ldb; /* leaf block's bno */
xfs_dir2_leaf_t *leaf; /* leaf structure */
xfs_dir2_leaf_tail_t *ltp; /* leaf's tail */
@@ -79,10 +156,10 @@ xfs_dir2_block_to_leaf(
return error;
}
ASSERT(lbp != NULL);
- leaf = lbp->data;
- block = dbp->data;
+ leaf = lbp->b_addr;
+ hdr = dbp->b_addr;
xfs_dir2_data_check(dp, dbp);
- btp = xfs_dir2_block_tail_p(mp, block);
+ btp = xfs_dir2_block_tail_p(mp, hdr);
blp = xfs_dir2_block_leaf_p(btp);
/*
* Set the counts in the leaf header.
@@ -102,23 +179,24 @@ xfs_dir2_block_to_leaf(
* tail be free.
*/
xfs_dir2_data_make_free(tp, dbp,
- (xfs_dir2_data_aoff_t)((char *)blp - (char *)block),
- (xfs_dir2_data_aoff_t)((char *)block + mp->m_dirblksize -
+ (xfs_dir2_data_aoff_t)((char *)blp - (char *)hdr),
+ (xfs_dir2_data_aoff_t)((char *)hdr + mp->m_dirblksize -
(char *)blp),
&needlog, &needscan);
/*
* Fix up the block header, make it a data block.
*/
- block->hdr.magic = cpu_to_be32(XFS_DIR2_DATA_MAGIC);
+ dbp->b_ops = &xfs_dir2_data_buf_ops;
+ hdr->magic = cpu_to_be32(XFS_DIR2_DATA_MAGIC);
if (needscan)
- xfs_dir2_data_freescan(mp, (xfs_dir2_data_t *)block, &needlog);
+ xfs_dir2_data_freescan(mp, hdr, &needlog);
/*
* Set up leaf tail and bests table.
*/
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
ltp->bestcount = cpu_to_be32(1);
bestsp = xfs_dir2_leaf_bests_p(ltp);
- bestsp[0] = block->hdr.bestfree[0].length;
+ bestsp[0] = hdr->bestfree[0].length;
/*
* Log the data header and leaf bests table.
*/
@@ -127,10 +205,134 @@ xfs_dir2_block_to_leaf(
xfs_dir2_leaf_check(dp, lbp);
xfs_dir2_data_check(dp, dbp);
xfs_dir2_leaf_log_bests(tp, lbp, 0, 0);
- xfs_da_buf_done(lbp);
return 0;
}
+STATIC void
+xfs_dir2_leaf_find_stale(
+ struct xfs_dir2_leaf *leaf,
+ int index,
+ int *lowstale,
+ int *highstale)
+{
+ /*
+ * Find the first stale entry before our index, if any.
+ */
+ for (*lowstale = index - 1; *lowstale >= 0; --*lowstale) {
+ if (leaf->ents[*lowstale].address ==
+ cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
+ break;
+ }
+
+ /*
+ * Find the first stale entry at or after our index, if any.
+ * Stop if the result would require moving more entries than using
+ * lowstale.
+ */
+ for (*highstale = index;
+ *highstale < be16_to_cpu(leaf->hdr.count);
+ ++*highstale) {
+ if (leaf->ents[*highstale].address ==
+ cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
+ break;
+ if (*lowstale >= 0 && index - *lowstale <= *highstale - index)
+ break;
+ }
+}
+
+struct xfs_dir2_leaf_entry *
+xfs_dir2_leaf_find_entry(
+ xfs_dir2_leaf_t *leaf, /* leaf structure */
+ int index, /* leaf table position */
+ int compact, /* need to compact leaves */
+ int lowstale, /* index of prev stale leaf */
+ int highstale, /* index of next stale leaf */
+ int *lfloglow, /* low leaf logging index */
+ int *lfloghigh) /* high leaf logging index */
+{
+ if (!leaf->hdr.stale) {
+ xfs_dir2_leaf_entry_t *lep; /* leaf entry table pointer */
+
+ /*
+ * Now we need to make room to insert the leaf entry.
+ *
+ * If there are no stale entries, just insert a hole at index.
+ */
+ lep = &leaf->ents[index];
+ if (index < be16_to_cpu(leaf->hdr.count))
+ memmove(lep + 1, lep,
+ (be16_to_cpu(leaf->hdr.count) - index) *
+ sizeof(*lep));
+
+ /*
+ * Record low and high logging indices for the leaf.
+ */
+ *lfloglow = index;
+ *lfloghigh = be16_to_cpu(leaf->hdr.count);
+ be16_add_cpu(&leaf->hdr.count, 1);
+ return lep;
+ }
+
+ /*
+ * There are stale entries.
+ *
+ * We will use one of them for the new entry. It's probably not at
+ * the right location, so we'll have to shift some up or down first.
+ *
+ * If we didn't compact before, we need to find the nearest stale
+ * entries before and after our insertion point.
+ */
+ if (compact == 0)
+ xfs_dir2_leaf_find_stale(leaf, index, &lowstale, &highstale);
+
+ /*
+ * If the low one is better, use it.
+ */
+ if (lowstale >= 0 &&
+ (highstale == be16_to_cpu(leaf->hdr.count) ||
+ index - lowstale - 1 < highstale - index)) {
+ ASSERT(index - lowstale - 1 >= 0);
+ ASSERT(leaf->ents[lowstale].address ==
+ cpu_to_be32(XFS_DIR2_NULL_DATAPTR));
+
+ /*
+ * Copy entries up to cover the stale entry and make room
+ * for the new entry.
+ */
+ if (index - lowstale - 1 > 0) {
+ memmove(&leaf->ents[lowstale],
+ &leaf->ents[lowstale + 1],
+ (index - lowstale - 1) *
+ sizeof(xfs_dir2_leaf_entry_t));
+ }
+ *lfloglow = MIN(lowstale, *lfloglow);
+ *lfloghigh = MAX(index - 1, *lfloghigh);
+ be16_add_cpu(&leaf->hdr.stale, -1);
+ return &leaf->ents[index - 1];
+ }
+
+ /*
+ * The high one is better, so use that one.
+ */
+ ASSERT(highstale - index >= 0);
+ ASSERT(leaf->ents[highstale].address ==
+ cpu_to_be32(XFS_DIR2_NULL_DATAPTR));
+
+ /*
+ * Copy entries down to cover the stale entry and make room for the
+ * new entry.
+ */
+ if (highstale - index > 0) {
+ memmove(&leaf->ents[index + 1],
+ &leaf->ents[index],
+ (highstale - index) * sizeof(xfs_dir2_leaf_entry_t));
+ }
+ *lfloglow = MIN(index, *lfloglow);
+ *lfloghigh = MAX(highstale, *lfloghigh);
+ be16_add_cpu(&leaf->hdr.stale, -1);
+ return &leaf->ents[index];
+}
+
/*
* Add an entry to a leaf form directory.
*/
@@ -140,8 +342,8 @@ xfs_dir2_leaf_addname(
{
__be16 *bestsp; /* freespace table in leaf */
int compact; /* need to compact leaves */
- xfs_dir2_data_t *data; /* data block structure */
- xfs_dabuf_t *dbp; /* data block buffer */
+ xfs_dir2_data_hdr_t *hdr; /* data block header */
+ struct xfs_buf *dbp; /* data block buffer */
xfs_dir2_data_entry_t *dep; /* data block entry */
xfs_inode_t *dp; /* incore directory inode */
xfs_dir2_data_unused_t *dup; /* data unused entry */
@@ -150,7 +352,7 @@ xfs_dir2_leaf_addname(
int highstale; /* index of next stale leaf */
int i; /* temporary, index */
int index; /* leaf table position */
- xfs_dabuf_t *lbp; /* leaf's buffer */
+ struct xfs_buf *lbp; /* leaf's buffer */
xfs_dir2_leaf_t *leaf; /* leaf structure */
int length; /* length of new entry */
xfs_dir2_leaf_entry_t *lep; /* leaf entry table pointer */
@@ -171,15 +373,11 @@ xfs_dir2_leaf_addname(
dp = args->dp;
tp = args->trans;
mp = dp->i_mount;
- /*
- * Read the leaf block.
- */
- error = xfs_da_read_buf(tp, dp, mp->m_dirleafblk, -1, &lbp,
- XFS_DATA_FORK);
- if (error) {
+
+ error = xfs_dir2_leaf_read(tp, dp, mp->m_dirleafblk, -1, &lbp);
+ if (error)
return error;
- }
- ASSERT(lbp != NULL);
+
/*
* Look up the entry by hash value and name.
* We know it's not there, our caller has already done a lookup.
@@ -187,7 +385,7 @@ xfs_dir2_leaf_addname(
* But if there are dup hash values the index is of the first of those.
*/
index = xfs_dir2_leaf_search_hash(args, lbp);
- leaf = lbp->data;
+ leaf = lbp->b_addr;
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
bestsp = xfs_dir2_leaf_bests_p(ltp);
length = xfs_dir2_data_entsize(args->namelen);
@@ -204,7 +402,7 @@ xfs_dir2_leaf_addname(
continue;
i = xfs_dir2_dataptr_to_db(mp, be32_to_cpu(lep->address));
ASSERT(i < be32_to_cpu(ltp->bestcount));
- ASSERT(be16_to_cpu(bestsp[i]) != NULLDATAOFF);
+ ASSERT(bestsp[i] != cpu_to_be16(NULLDATAOFF));
if (be16_to_cpu(bestsp[i]) >= length) {
use_block = i;
break;
@@ -218,7 +416,8 @@ xfs_dir2_leaf_addname(
/*
* Remember a block we see that's missing.
*/
- if (be16_to_cpu(bestsp[i]) == NULLDATAOFF && use_block == -1)
+ if (bestsp[i] == cpu_to_be16(NULLDATAOFF) &&
+ use_block == -1)
use_block = i;
else if (be16_to_cpu(bestsp[i]) >= length) {
use_block = i;
@@ -229,14 +428,17 @@ xfs_dir2_leaf_addname(
/*
* How many bytes do we need in the leaf block?
*/
- needbytes =
- (leaf->hdr.stale ? 0 : (uint)sizeof(leaf->ents[0])) +
- (use_block != -1 ? 0 : (uint)sizeof(leaf->bests[0]));
+ needbytes = 0;
+ if (!leaf->hdr.stale)
+ needbytes += sizeof(xfs_dir2_leaf_entry_t);
+ if (use_block == -1)
+ needbytes += sizeof(xfs_dir2_data_off_t);
+
/*
* Now kill use_block if it refers to a missing block, so we
* can use it as an indication of allocation needed.
*/
- if (use_block != -1 && be16_to_cpu(bestsp[use_block]) == NULLDATAOFF)
+ if (use_block != -1 && bestsp[use_block] == cpu_to_be16(NULLDATAOFF))
use_block = -1;
/*
* If we don't have enough free bytes but we can make enough
@@ -257,14 +459,13 @@ xfs_dir2_leaf_addname(
*/
if ((args->op_flags & XFS_DA_OP_JUSTCHECK) ||
args->total == 0) {
- xfs_da_brelse(tp, lbp);
+ xfs_trans_brelse(tp, lbp);
return XFS_ERROR(ENOSPC);
}
/*
* Convert to node form.
*/
error = xfs_dir2_leaf_to_node(args, lbp);
- xfs_da_buf_done(lbp);
if (error)
return error;
/*
@@ -282,7 +483,7 @@ xfs_dir2_leaf_addname(
* a new data block.
*/
if (args->op_flags & XFS_DA_OP_JUSTCHECK) {
- xfs_da_brelse(tp, lbp);
+ xfs_trans_brelse(tp, lbp);
return use_block == -1 ? XFS_ERROR(ENOSPC) : 0;
}
/*
@@ -290,7 +491,7 @@ xfs_dir2_leaf_addname(
* changed anything.
*/
if (args->total == 0 && use_block == -1) {
- xfs_da_brelse(tp, lbp);
+ xfs_trans_brelse(tp, lbp);
return XFS_ERROR(ENOSPC);
}
/*
@@ -321,14 +522,14 @@ xfs_dir2_leaf_addname(
*/
if ((error = xfs_dir2_grow_inode(args, XFS_DIR2_DATA_SPACE,
&use_block))) {
- xfs_da_brelse(tp, lbp);
+ xfs_trans_brelse(tp, lbp);
return error;
}
/*
* Initialize the block.
*/
if ((error = xfs_dir2_data_init(args, use_block, &dbp))) {
- xfs_da_brelse(tp, lbp);
+ xfs_trans_brelse(tp, lbp);
return error;
}
/*
@@ -348,37 +549,36 @@ xfs_dir2_leaf_addname(
*/
else
xfs_dir2_leaf_log_bests(tp, lbp, use_block, use_block);
- data = dbp->data;
- bestsp[use_block] = data->hdr.bestfree[0].length;
+ hdr = dbp->b_addr;
+ bestsp[use_block] = hdr->bestfree[0].length;
grown = 1;
- }
- /*
- * Already had space in some data block.
- * Just read that one in.
- */
- else {
- if ((error =
- xfs_da_read_buf(tp, dp, xfs_dir2_db_to_da(mp, use_block),
- -1, &dbp, XFS_DATA_FORK))) {
- xfs_da_brelse(tp, lbp);
+ } else {
+ /*
+ * Already had space in some data block.
+ * Just read that one in.
+ */
+ error = xfs_dir2_data_read(tp, dp,
+ xfs_dir2_db_to_da(mp, use_block),
+ -1, &dbp);
+ if (error) {
+ xfs_trans_brelse(tp, lbp);
return error;
}
- data = dbp->data;
+ hdr = dbp->b_addr;
grown = 0;
}
- xfs_dir2_data_check(dp, dbp);
/*
* Point to the biggest freespace in our data block.
*/
dup = (xfs_dir2_data_unused_t *)
- ((char *)data + be16_to_cpu(data->hdr.bestfree[0].offset));
+ ((char *)hdr + be16_to_cpu(hdr->bestfree[0].offset));
ASSERT(be16_to_cpu(dup->length) >= length);
needscan = needlog = 0;
/*
* Mark the initial part of our freespace in use for the new entry.
*/
xfs_dir2_data_use_free(tp, dbp, dup,
- (xfs_dir2_data_aoff_t)((char *)dup - (char *)data), length,
+ (xfs_dir2_data_aoff_t)((char *)dup - (char *)hdr), length,
&needlog, &needscan);
/*
* Initialize our new entry (at last).
@@ -388,12 +588,12 @@ xfs_dir2_leaf_addname(
dep->namelen = args->namelen;
memcpy(dep->name, args->name, dep->namelen);
tagp = xfs_dir2_data_entry_tag_p(dep);
- *tagp = cpu_to_be16((char *)dep - (char *)data);
+ *tagp = cpu_to_be16((char *)dep - (char *)hdr);
/*
* Need to scan fix up the bestfree table.
*/
if (needscan)
- xfs_dir2_data_freescan(mp, data, &needlog);
+ xfs_dir2_data_freescan(mp, hdr, &needlog);
/*
* Need to log the data block's header.
*/
@@ -404,107 +604,15 @@ xfs_dir2_leaf_addname(
* If the bests table needs to be changed, do it.
* Log the change unless we've already done that.
*/
- if (be16_to_cpu(bestsp[use_block]) != be16_to_cpu(data->hdr.bestfree[0].length)) {
- bestsp[use_block] = data->hdr.bestfree[0].length;
+ if (be16_to_cpu(bestsp[use_block]) != be16_to_cpu(hdr->bestfree[0].length)) {
+ bestsp[use_block] = hdr->bestfree[0].length;
if (!grown)
xfs_dir2_leaf_log_bests(tp, lbp, use_block, use_block);
}
- /*
- * Now we need to make room to insert the leaf entry.
- * If there are no stale entries, we just insert a hole at index.
- */
- if (!leaf->hdr.stale) {
- /*
- * lep is still good as the index leaf entry.
- */
- if (index < be16_to_cpu(leaf->hdr.count))
- memmove(lep + 1, lep,
- (be16_to_cpu(leaf->hdr.count) - index) * sizeof(*lep));
- /*
- * Record low and high logging indices for the leaf.
- */
- lfloglow = index;
- lfloghigh = be16_to_cpu(leaf->hdr.count);
- be16_add_cpu(&leaf->hdr.count, 1);
- }
- /*
- * There are stale entries.
- * We will use one of them for the new entry.
- * It's probably not at the right location, so we'll have to
- * shift some up or down first.
- */
- else {
- /*
- * If we didn't compact before, we need to find the nearest
- * stale entries before and after our insertion point.
- */
- if (compact == 0) {
- /*
- * Find the first stale entry before the insertion
- * point, if any.
- */
- for (lowstale = index - 1;
- lowstale >= 0 &&
- be32_to_cpu(leaf->ents[lowstale].address) !=
- XFS_DIR2_NULL_DATAPTR;
- lowstale--)
- continue;
- /*
- * Find the next stale entry at or after the insertion
- * point, if any. Stop if we go so far that the
- * lowstale entry would be better.
- */
- for (highstale = index;
- highstale < be16_to_cpu(leaf->hdr.count) &&
- be32_to_cpu(leaf->ents[highstale].address) !=
- XFS_DIR2_NULL_DATAPTR &&
- (lowstale < 0 ||
- index - lowstale - 1 >= highstale - index);
- highstale++)
- continue;
- }
- /*
- * If the low one is better, use it.
- */
- if (lowstale >= 0 &&
- (highstale == be16_to_cpu(leaf->hdr.count) ||
- index - lowstale - 1 < highstale - index)) {
- ASSERT(index - lowstale - 1 >= 0);
- ASSERT(be32_to_cpu(leaf->ents[lowstale].address) ==
- XFS_DIR2_NULL_DATAPTR);
- /*
- * Copy entries up to cover the stale entry
- * and make room for the new entry.
- */
- if (index - lowstale - 1 > 0)
- memmove(&leaf->ents[lowstale],
- &leaf->ents[lowstale + 1],
- (index - lowstale - 1) * sizeof(*lep));
- lep = &leaf->ents[index - 1];
- lfloglow = MIN(lowstale, lfloglow);
- lfloghigh = MAX(index - 1, lfloghigh);
- }
- /*
- * The high one is better, so use that one.
- */
- else {
- ASSERT(highstale - index >= 0);
- ASSERT(be32_to_cpu(leaf->ents[highstale].address) ==
- XFS_DIR2_NULL_DATAPTR);
- /*
- * Copy entries down to cover the stale entry
- * and make room for the new entry.
- */
- if (highstale - index > 0)
- memmove(&leaf->ents[index + 1],
- &leaf->ents[index],
- (highstale - index) * sizeof(*lep));
- lep = &leaf->ents[index];
- lfloglow = MIN(index, lfloglow);
- lfloghigh = MAX(highstale, lfloghigh);
- }
- be16_add_cpu(&leaf->hdr.stale, -1);
- }
+
+ lep = xfs_dir2_leaf_find_entry(leaf, index, compact, lowstale,
+ highstale, &lfloglow, &lfloghigh);
+
/*
* Fill in the new leaf entry.
*/
@@ -517,9 +625,7 @@ xfs_dir2_leaf_addname(
xfs_dir2_leaf_log_header(tp, lbp);
xfs_dir2_leaf_log_ents(tp, lbp, lfloglow, lfloghigh);
xfs_dir2_leaf_check(dp, lbp);
- xfs_da_buf_done(lbp);
xfs_dir2_data_check(dp, dbp);
- xfs_da_buf_done(dbp);
return 0;
}
@@ -530,8 +636,8 @@ xfs_dir2_leaf_addname(
*/
STATIC void
xfs_dir2_leaf_check(
- xfs_inode_t *dp, /* incore directory inode */
- xfs_dabuf_t *bp) /* leaf's buffer */
+ struct xfs_inode *dp, /* incore directory inode */
+ struct xfs_buf *bp) /* leaf's buffer */
{
int i; /* leaf index */
xfs_dir2_leaf_t *leaf; /* leaf structure */
@@ -539,9 +645,9 @@ xfs_dir2_leaf_check(
xfs_mount_t *mp; /* filesystem mount point */
int stale; /* count of stale leaves */
- leaf = bp->data;
+ leaf = bp->b_addr;
mp = dp->i_mount;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAF1_MAGIC);
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC));
/*
* This value is not restrictive enough.
* Should factor in the size of the bests table as well.
@@ -561,7 +667,7 @@ xfs_dir2_leaf_check(
if (i + 1 < be16_to_cpu(leaf->hdr.count))
ASSERT(be32_to_cpu(leaf->ents[i].hashval) <=
be32_to_cpu(leaf->ents[i + 1].hashval));
- if (be32_to_cpu(leaf->ents[i].address) == XFS_DIR2_NULL_DATAPTR)
+ if (leaf->ents[i].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
stale++;
}
ASSERT(be16_to_cpu(leaf->hdr.stale) == stale);
@@ -575,14 +681,14 @@ xfs_dir2_leaf_check(
void
xfs_dir2_leaf_compact(
xfs_da_args_t *args, /* operation arguments */
- xfs_dabuf_t *bp) /* leaf buffer */
+ struct xfs_buf *bp) /* leaf buffer */
{
int from; /* source leaf index */
xfs_dir2_leaf_t *leaf; /* leaf structure */
int loglow; /* first leaf entry to log */
int to; /* target leaf index */
- leaf = bp->data;
+ leaf = bp->b_addr;
if (!leaf->hdr.stale) {
return;
}
@@ -590,7 +696,8 @@ xfs_dir2_leaf_compact(
* Compress out the stale entries in place.
*/
for (from = to = 0, loglow = -1; from < be16_to_cpu(leaf->hdr.count); from++) {
- if (be32_to_cpu(leaf->ents[from].address) == XFS_DIR2_NULL_DATAPTR)
+ if (leaf->ents[from].address ==
+ cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
continue;
/*
* Only actually copy the entries that are different.
@@ -623,7 +730,7 @@ xfs_dir2_leaf_compact(
*/
void
xfs_dir2_leaf_compact_x1(
- xfs_dabuf_t *bp, /* leaf buffer */
+ struct xfs_buf *bp, /* leaf buffer */
int *indexp, /* insertion index */
int *lowstalep, /* out: stale entry before us */
int *highstalep, /* out: stale entry after us */
@@ -639,27 +746,12 @@ xfs_dir2_leaf_compact_x1(
int newindex=0; /* new insertion index */
int to; /* destination copy index */
- leaf = bp->data;
+ leaf = bp->b_addr;
ASSERT(be16_to_cpu(leaf->hdr.stale) > 1);
index = *indexp;
- /*
- * Find the first stale entry before our index, if any.
- */
- for (lowstale = index - 1;
- lowstale >= 0 &&
- be32_to_cpu(leaf->ents[lowstale].address) != XFS_DIR2_NULL_DATAPTR;
- lowstale--)
- continue;
- /*
- * Find the first stale entry at or after our index, if any.
- * Stop if the answer would be worse than lowstale.
- */
- for (highstale = index;
- highstale < be16_to_cpu(leaf->hdr.count) &&
- be32_to_cpu(leaf->ents[highstale].address) != XFS_DIR2_NULL_DATAPTR &&
- (lowstale < 0 || index - lowstale > highstale - index);
- highstale++)
- continue;
+
+ xfs_dir2_leaf_find_stale(leaf, index, &lowstale, &highstale);
+
/*
* Pick the better of lowstale and highstale.
*/
@@ -680,7 +772,8 @@ xfs_dir2_leaf_compact_x1(
if (index == from)
newindex = to;
if (from != keepstale &&
- be32_to_cpu(leaf->ents[from].address) == XFS_DIR2_NULL_DATAPTR) {
+ leaf->ents[from].address ==
+ cpu_to_be32(XFS_DIR2_NULL_DATAPTR)) {
if (from == to)
*lowlogp = to;
continue;
@@ -730,10 +823,10 @@ int
xfs_dir2_leaf_init(
xfs_da_args_t *args, /* operation arguments */
xfs_dir2_db_t bno, /* directory block number */
- xfs_dabuf_t **bpp, /* out: leaf buffer */
+ struct xfs_buf **bpp, /* out: leaf buffer */
int magic) /* magic number for block */
{
- xfs_dabuf_t *bp; /* leaf buffer */
+ struct xfs_buf *bp; /* leaf buffer */
xfs_inode_t *dp; /* incore directory inode */
int error; /* error return code */
xfs_dir2_leaf_t *leaf; /* leaf structure */
@@ -751,15 +844,14 @@ xfs_dir2_leaf_init(
* Get the buffer for the block.
*/
error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(mp, bno), -1, &bp,
- XFS_DATA_FORK);
- if (error) {
+ XFS_DATA_FORK);
+ if (error)
return error;
- }
- ASSERT(bp != NULL);
- leaf = bp->data;
+
/*
* Initialize the header.
*/
+ leaf = bp->b_addr;
leaf->hdr.info.magic = cpu_to_be16(magic);
leaf->hdr.info.forw = 0;
leaf->hdr.info.back = 0;
@@ -772,10 +864,12 @@ xfs_dir2_leaf_init(
* the block.
*/
if (magic == XFS_DIR2_LEAF1_MAGIC) {
+ bp->b_ops = &xfs_dir2_leaf1_buf_ops;
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
ltp->bestcount = 0;
xfs_dir2_leaf_log_tail(tp, bp);
- }
+ } else
+ bp->b_ops = &xfs_dir2_leafn_buf_ops;
*bpp = bp;
return 0;
}
@@ -786,7 +880,7 @@ xfs_dir2_leaf_init(
static void
xfs_dir2_leaf_log_bests(
xfs_trans_t *tp, /* transaction pointer */
- xfs_dabuf_t *bp, /* leaf buffer */
+ struct xfs_buf *bp, /* leaf buffer */
int first, /* first entry to log */
int last) /* last entry to log */
{
@@ -795,12 +889,12 @@ xfs_dir2_leaf_log_bests(
xfs_dir2_leaf_t *leaf; /* leaf structure */
xfs_dir2_leaf_tail_t *ltp; /* leaf tail structure */
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAF1_MAGIC);
+ leaf = bp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC));
ltp = xfs_dir2_leaf_tail_p(tp->t_mountp, leaf);
firstb = xfs_dir2_leaf_bests_p(ltp) + first;
lastb = xfs_dir2_leaf_bests_p(ltp) + last;
- xfs_da_log_buf(tp, bp, (uint)((char *)firstb - (char *)leaf),
+ xfs_trans_log_buf(tp, bp, (uint)((char *)firstb - (char *)leaf),
(uint)((char *)lastb - (char *)leaf + sizeof(*lastb) - 1));
}
@@ -810,7 +904,7 @@ xfs_dir2_leaf_log_bests(
void
xfs_dir2_leaf_log_ents(
xfs_trans_t *tp, /* transaction pointer */
- xfs_dabuf_t *bp, /* leaf buffer */
+ struct xfs_buf *bp, /* leaf buffer */
int first, /* first entry to log */
int last) /* last entry to log */
{
@@ -818,12 +912,12 @@ xfs_dir2_leaf_log_ents(
xfs_dir2_leaf_entry_t *lastlep; /* pointer to last entry */
xfs_dir2_leaf_t *leaf; /* leaf structure */
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAF1_MAGIC ||
- be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
+ leaf = bp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC) ||
+ leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
firstlep = &leaf->ents[first];
lastlep = &leaf->ents[last];
- xfs_da_log_buf(tp, bp, (uint)((char *)firstlep - (char *)leaf),
+ xfs_trans_log_buf(tp, bp, (uint)((char *)firstlep - (char *)leaf),
(uint)((char *)lastlep - (char *)leaf + sizeof(*lastlep) - 1));
}
@@ -832,15 +926,15 @@ xfs_dir2_leaf_log_ents(
*/
void
xfs_dir2_leaf_log_header(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_dabuf_t *bp) /* leaf buffer */
+ struct xfs_trans *tp,
+ struct xfs_buf *bp)
{
xfs_dir2_leaf_t *leaf; /* leaf structure */
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAF1_MAGIC ||
- be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
- xfs_da_log_buf(tp, bp, (uint)((char *)&leaf->hdr - (char *)leaf),
+ leaf = bp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC) ||
+ leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
+ xfs_trans_log_buf(tp, bp, (uint)((char *)&leaf->hdr - (char *)leaf),
(uint)(sizeof(leaf->hdr) - 1));
}
@@ -849,18 +943,18 @@ xfs_dir2_leaf_log_header(
*/
STATIC void
xfs_dir2_leaf_log_tail(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_dabuf_t *bp) /* leaf buffer */
+ struct xfs_trans *tp,
+ struct xfs_buf *bp)
{
xfs_dir2_leaf_t *leaf; /* leaf structure */
xfs_dir2_leaf_tail_t *ltp; /* leaf tail structure */
xfs_mount_t *mp; /* filesystem mount point */
mp = tp->t_mountp;
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAF1_MAGIC);
+ leaf = bp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC));
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
- xfs_da_log_buf(tp, bp, (uint)((char *)ltp - (char *)leaf),
+ xfs_trans_log_buf(tp, bp, (uint)((char *)ltp - (char *)leaf),
(uint)(mp->m_dirblksize - 1));
}
@@ -873,12 +967,12 @@ int
xfs_dir2_leaf_lookup(
xfs_da_args_t *args) /* operation arguments */
{
- xfs_dabuf_t *dbp; /* data block buffer */
+ struct xfs_buf *dbp; /* data block buffer */
xfs_dir2_data_entry_t *dep; /* data block entry */
xfs_inode_t *dp; /* incore directory inode */
int error; /* error return code */
int index; /* found entry index */
- xfs_dabuf_t *lbp; /* leaf buffer */
+ struct xfs_buf *lbp; /* leaf buffer */
xfs_dir2_leaf_t *leaf; /* leaf structure */
xfs_dir2_leaf_entry_t *lep; /* leaf entry */
xfs_trans_t *tp; /* transaction pointer */
@@ -894,7 +988,7 @@ xfs_dir2_leaf_lookup(
tp = args->trans;
dp = args->dp;
xfs_dir2_leaf_check(dp, lbp);
- leaf = lbp->data;
+ leaf = lbp->b_addr;
/*
* Get to the leaf entry and contained data entry address.
*/
@@ -903,15 +997,15 @@ xfs_dir2_leaf_lookup(
* Point to the data entry.
*/
dep = (xfs_dir2_data_entry_t *)
- ((char *)dbp->data +
+ ((char *)dbp->b_addr +
xfs_dir2_dataptr_to_off(dp->i_mount, be32_to_cpu(lep->address)));
/*
* Return the found inode number & CI name if appropriate
*/
args->inumber = be64_to_cpu(dep->inumber);
error = xfs_dir_cilookup_result(args, dep->name, dep->namelen);
- xfs_da_brelse(tp, dbp);
- xfs_da_brelse(tp, lbp);
+ xfs_trans_brelse(tp, dbp);
+ xfs_trans_brelse(tp, lbp);
return XFS_ERROR(error);
}
@@ -924,17 +1018,17 @@ xfs_dir2_leaf_lookup(
static int /* error */
xfs_dir2_leaf_lookup_int(
xfs_da_args_t *args, /* operation arguments */
- xfs_dabuf_t **lbpp, /* out: leaf buffer */
+ struct xfs_buf **lbpp, /* out: leaf buffer */
int *indexp, /* out: index in leaf block */
- xfs_dabuf_t **dbpp) /* out: data buffer */
+ struct xfs_buf **dbpp) /* out: data buffer */
{
xfs_dir2_db_t curdb = -1; /* current data block number */
- xfs_dabuf_t *dbp = NULL; /* data buffer */
+ struct xfs_buf *dbp = NULL; /* data buffer */
xfs_dir2_data_entry_t *dep; /* data entry */
xfs_inode_t *dp; /* incore directory inode */
int error; /* error return code */
int index; /* index in leaf block */
- xfs_dabuf_t *lbp; /* leaf buffer */
+ struct xfs_buf *lbp; /* leaf buffer */
xfs_dir2_leaf_entry_t *lep; /* leaf entry */
xfs_dir2_leaf_t *leaf; /* leaf structure */
xfs_mount_t *mp; /* filesystem mount point */
@@ -946,15 +1040,13 @@ xfs_dir2_leaf_lookup_int(
dp = args->dp;
tp = args->trans;
mp = dp->i_mount;
- /*
- * Read the leaf block into the buffer.
- */
- error = xfs_da_read_buf(tp, dp, mp->m_dirleafblk, -1, &lbp,
- XFS_DATA_FORK);
+
+ error = xfs_dir2_leaf_read(tp, dp, mp->m_dirleafblk, -1, &lbp);
if (error)
return error;
+
*lbpp = lbp;
- leaf = lbp->data;
+ leaf = lbp->b_addr;
xfs_dir2_leaf_check(dp, lbp);
/*
* Look for the first leaf entry with our hash value.
@@ -982,21 +1074,20 @@ xfs_dir2_leaf_lookup_int(
*/
if (newdb != curdb) {
if (dbp)
- xfs_da_brelse(tp, dbp);
- error = xfs_da_read_buf(tp, dp,
- xfs_dir2_db_to_da(mp, newdb),
- -1, &dbp, XFS_DATA_FORK);
+ xfs_trans_brelse(tp, dbp);
+ error = xfs_dir2_data_read(tp, dp,
+ xfs_dir2_db_to_da(mp, newdb),
+ -1, &dbp);
if (error) {
- xfs_da_brelse(tp, lbp);
+ xfs_trans_brelse(tp, lbp);
return error;
}
- xfs_dir2_data_check(dp, dbp);
curdb = newdb;
}
/*
* Point to the data entry.
*/
- dep = (xfs_dir2_data_entry_t *)((char *)dbp->data +
+ dep = (xfs_dir2_data_entry_t *)((char *)dbp->b_addr +
xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address)));
/*
* Compare name and if it's an exact match, return the index
@@ -1024,12 +1115,12 @@ xfs_dir2_leaf_lookup_int(
if (args->cmpresult == XFS_CMP_CASE) {
ASSERT(cidb != -1);
if (cidb != curdb) {
- xfs_da_brelse(tp, dbp);
- error = xfs_da_read_buf(tp, dp,
- xfs_dir2_db_to_da(mp, cidb),
- -1, &dbp, XFS_DATA_FORK);
+ xfs_trans_brelse(tp, dbp);
+ error = xfs_dir2_data_read(tp, dp,
+ xfs_dir2_db_to_da(mp, cidb),
+ -1, &dbp);
if (error) {
- xfs_da_brelse(tp, lbp);
+ xfs_trans_brelse(tp, lbp);
return error;
}
}
@@ -1041,8 +1132,8 @@ xfs_dir2_leaf_lookup_int(
*/
ASSERT(cidb == -1);
if (dbp)
- xfs_da_brelse(tp, dbp);
- xfs_da_brelse(tp, lbp);
+ xfs_trans_brelse(tp, dbp);
+ xfs_trans_brelse(tp, lbp);
return XFS_ERROR(ENOENT);
}
@@ -1054,15 +1145,15 @@ xfs_dir2_leaf_removename(
xfs_da_args_t *args) /* operation arguments */
{
__be16 *bestsp; /* leaf block best freespace */
- xfs_dir2_data_t *data; /* data block structure */
+ xfs_dir2_data_hdr_t *hdr; /* data block header */
xfs_dir2_db_t db; /* data block number */
- xfs_dabuf_t *dbp; /* data block buffer */
+ struct xfs_buf *dbp; /* data block buffer */
xfs_dir2_data_entry_t *dep; /* data entry structure */
xfs_inode_t *dp; /* incore directory inode */
int error; /* error return code */
xfs_dir2_db_t i; /* temporary data block # */
int index; /* index into leaf entries */
- xfs_dabuf_t *lbp; /* leaf buffer */
+ struct xfs_buf *lbp; /* leaf buffer */
xfs_dir2_leaf_t *leaf; /* leaf structure */
xfs_dir2_leaf_entry_t *lep; /* leaf entry */
xfs_dir2_leaf_tail_t *ltp; /* leaf tail structure */
@@ -1083,8 +1174,8 @@ xfs_dir2_leaf_removename(
dp = args->dp;
tp = args->trans;
mp = dp->i_mount;
- leaf = lbp->data;
- data = dbp->data;
+ leaf = lbp->b_addr;
+ hdr = dbp->b_addr;
xfs_dir2_data_check(dp, dbp);
/*
* Point to the leaf entry, use that to point to the data entry.
@@ -1092,9 +1183,9 @@ xfs_dir2_leaf_removename(
lep = &leaf->ents[index];
db = xfs_dir2_dataptr_to_db(mp, be32_to_cpu(lep->address));
dep = (xfs_dir2_data_entry_t *)
- ((char *)data + xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address)));
+ ((char *)hdr + xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address)));
needscan = needlog = 0;
- oldbest = be16_to_cpu(data->hdr.bestfree[0].length);
+ oldbest = be16_to_cpu(hdr->bestfree[0].length);
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
bestsp = xfs_dir2_leaf_bests_p(ltp);
ASSERT(be16_to_cpu(bestsp[db]) == oldbest);
@@ -1102,7 +1193,7 @@ xfs_dir2_leaf_removename(
* Mark the former data entry unused.
*/
xfs_dir2_data_make_free(tp, dbp,
- (xfs_dir2_data_aoff_t)((char *)dep - (char *)data),
+ (xfs_dir2_data_aoff_t)((char *)dep - (char *)hdr),
xfs_dir2_data_entsize(dep->namelen), &needlog, &needscan);
/*
* We just mark the leaf entry stale by putting a null in it.
@@ -1116,23 +1207,23 @@ xfs_dir2_leaf_removename(
* log the data block header if necessary.
*/
if (needscan)
- xfs_dir2_data_freescan(mp, data, &needlog);
+ xfs_dir2_data_freescan(mp, hdr, &needlog);
if (needlog)
xfs_dir2_data_log_header(tp, dbp);
/*
* If the longest freespace in the data block has changed,
* put the new value in the bests table and log that.
*/
- if (be16_to_cpu(data->hdr.bestfree[0].length) != oldbest) {
- bestsp[db] = data->hdr.bestfree[0].length;
+ if (be16_to_cpu(hdr->bestfree[0].length) != oldbest) {
+ bestsp[db] = hdr->bestfree[0].length;
xfs_dir2_leaf_log_bests(tp, lbp, db, db);
}
xfs_dir2_data_check(dp, dbp);
/*
* If the data block is now empty then get rid of the data block.
*/
- if (be16_to_cpu(data->hdr.bestfree[0].length) ==
- mp->m_dirblksize - (uint)sizeof(data->hdr)) {
+ if (be16_to_cpu(hdr->bestfree[0].length) ==
+ mp->m_dirblksize - (uint)sizeof(*hdr)) {
ASSERT(db != mp->m_dirdatablk);
if ((error = xfs_dir2_shrink_inode(args, db, dbp))) {
/*
@@ -1141,12 +1232,9 @@ xfs_dir2_leaf_removename(
* Just go on, returning success, leaving the
* empty block in place.
*/
- if (error == ENOSPC && args->total == 0) {
- xfs_da_buf_done(dbp);
+ if (error == ENOSPC && args->total == 0)
error = 0;
- }
xfs_dir2_leaf_check(dp, lbp);
- xfs_da_buf_done(lbp);
return error;
}
dbp = NULL;
@@ -1159,7 +1247,7 @@ xfs_dir2_leaf_removename(
* Look for the last active entry (i).
*/
for (i = db - 1; i > 0; i--) {
- if (be16_to_cpu(bestsp[i]) != NULLDATAOFF)
+ if (bestsp[i] != cpu_to_be16(NULLDATAOFF))
break;
}
/*
@@ -1177,10 +1265,9 @@ xfs_dir2_leaf_removename(
/*
* If the data block was not the first one, drop it.
*/
- else if (db != mp->m_dirdatablk && dbp != NULL) {
- xfs_da_buf_done(dbp);
+ else if (db != mp->m_dirdatablk)
dbp = NULL;
- }
+
xfs_dir2_leaf_check(dp, lbp);
/*
* See if we can convert to block form.
@@ -1195,12 +1282,12 @@ int /* error */
xfs_dir2_leaf_replace(
xfs_da_args_t *args) /* operation arguments */
{
- xfs_dabuf_t *dbp; /* data block buffer */
+ struct xfs_buf *dbp; /* data block buffer */
xfs_dir2_data_entry_t *dep; /* data block entry */
xfs_inode_t *dp; /* incore directory inode */
int error; /* error return code */
int index; /* index of leaf entry */
- xfs_dabuf_t *lbp; /* leaf buffer */
+ struct xfs_buf *lbp; /* leaf buffer */
xfs_dir2_leaf_t *leaf; /* leaf structure */
xfs_dir2_leaf_entry_t *lep; /* leaf entry */
xfs_trans_t *tp; /* transaction pointer */
@@ -1214,7 +1301,7 @@ xfs_dir2_leaf_replace(
return error;
}
dp = args->dp;
- leaf = lbp->data;
+ leaf = lbp->b_addr;
/*
* Point to the leaf entry, get data address from it.
*/
@@ -1223,7 +1310,7 @@ xfs_dir2_leaf_replace(
* Point to the data entry.
*/
dep = (xfs_dir2_data_entry_t *)
- ((char *)dbp->data +
+ ((char *)dbp->b_addr +
xfs_dir2_dataptr_to_off(dp->i_mount, be32_to_cpu(lep->address)));
ASSERT(args->inumber != be64_to_cpu(dep->inumber));
/*
@@ -1232,9 +1319,8 @@ xfs_dir2_leaf_replace(
dep->inumber = cpu_to_be64(args->inumber);
tp = args->trans;
xfs_dir2_data_log_entry(tp, dbp, dep);
- xfs_da_buf_done(dbp);
xfs_dir2_leaf_check(dp, lbp);
- xfs_da_brelse(tp, lbp);
+ xfs_trans_brelse(tp, lbp);
return 0;
}
@@ -1246,7 +1332,7 @@ xfs_dir2_leaf_replace(
int /* index value */
xfs_dir2_leaf_search_hash(
xfs_da_args_t *args, /* operation arguments */
- xfs_dabuf_t *lbp) /* leaf buffer */
+ struct xfs_buf *lbp) /* leaf buffer */
{
xfs_dahash_t hash=0; /* hash from this entry */
xfs_dahash_t hashwant; /* hash value looking for */
@@ -1256,7 +1342,7 @@ xfs_dir2_leaf_search_hash(
xfs_dir2_leaf_entry_t *lep; /* leaf entry */
int mid=0; /* current leaf index */
- leaf = lbp->data;
+ leaf = lbp->b_addr;
#ifndef __KERNEL__
if (!leaf->hdr.count)
return 0;
@@ -1299,14 +1385,11 @@ xfs_dir2_leaf_search_hash(
int /* error */
xfs_dir2_leaf_trim_data(
xfs_da_args_t *args, /* operation arguments */
- xfs_dabuf_t *lbp, /* leaf buffer */
+ struct xfs_buf *lbp, /* leaf buffer */
xfs_dir2_db_t db) /* data block number */
{
__be16 *bestsp; /* leaf bests table */
-#ifdef DEBUG
- xfs_dir2_data_t *data; /* data block structure */
-#endif
- xfs_dabuf_t *dbp; /* data block buffer */
+ struct xfs_buf *dbp; /* data block buffer */
xfs_inode_t *dp; /* incore directory inode */
int error; /* error return value */
xfs_dir2_leaf_t *leaf; /* leaf structure */
@@ -1320,30 +1403,30 @@ xfs_dir2_leaf_trim_data(
/*
* Read the offending data block. We need its buffer.
*/
- if ((error = xfs_da_read_buf(tp, dp, xfs_dir2_db_to_da(mp, db), -1, &dbp,
- XFS_DATA_FORK))) {
+ error = xfs_dir2_data_read(tp, dp, xfs_dir2_db_to_da(mp, db), -1, &dbp);
+ if (error)
return error;
- }
-#ifdef DEBUG
- data = dbp->data;
- ASSERT(be32_to_cpu(data->hdr.magic) == XFS_DIR2_DATA_MAGIC);
-#endif
- /* this seems to be an error
- * data is only valid if DEBUG is defined?
- * RMC 09/08/1999
- */
- leaf = lbp->data;
+ leaf = lbp->b_addr;
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
- ASSERT(be16_to_cpu(data->hdr.bestfree[0].length) ==
- mp->m_dirblksize - (uint)sizeof(data->hdr));
+
+#ifdef DEBUG
+{
+ struct xfs_dir2_data_hdr *hdr = dbp->b_addr;
+
+ ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC));
+ ASSERT(be16_to_cpu(hdr->bestfree[0].length) ==
+ mp->m_dirblksize - (uint)sizeof(*hdr));
ASSERT(db == be32_to_cpu(ltp->bestcount) - 1);
+}
+#endif
+
/*
* Get rid of the data block.
*/
if ((error = xfs_dir2_shrink_inode(args, db, dbp))) {
ASSERT(error != ENOSPC);
- xfs_da_brelse(tp, dbp);
+ xfs_trans_brelse(tp, dbp);
return error;
}
/*
@@ -1357,6 +1440,20 @@ xfs_dir2_leaf_trim_data(
return 0;
}
+static inline size_t
+xfs_dir2_leaf_size(
+ struct xfs_dir2_leaf_hdr *hdr,
+ int counts)
+{
+ int entries;
+
+ entries = be16_to_cpu(hdr->count) - be16_to_cpu(hdr->stale);
+ return sizeof(xfs_dir2_leaf_hdr_t) +
+ entries * sizeof(xfs_dir2_leaf_entry_t) +
+ counts * sizeof(xfs_dir2_data_off_t) +
+ sizeof(xfs_dir2_leaf_tail_t);
+}
+
/*
* Convert node form directory to leaf form directory.
* The root of the node form dir needs to already be a LEAFN block.
@@ -1369,10 +1466,10 @@ xfs_dir2_node_to_leaf(
xfs_da_args_t *args; /* operation arguments */
xfs_inode_t *dp; /* incore directory inode */
int error; /* error return code */
- xfs_dabuf_t *fbp; /* buffer for freespace block */
+ struct xfs_buf *fbp; /* buffer for freespace block */
xfs_fileoff_t fo; /* freespace file offset */
xfs_dir2_free_t *free; /* freespace structure */
- xfs_dabuf_t *lbp; /* buffer for leaf block */
+ struct xfs_buf *lbp; /* buffer for leaf block */
xfs_dir2_leaf_tail_t *ltp; /* tail of leaf structure */
xfs_dir2_leaf_t *leaf; /* leaf structure */
xfs_mount_t *mp; /* filesystem mount point */
@@ -1426,30 +1523,28 @@ xfs_dir2_node_to_leaf(
if (XFS_FSB_TO_B(mp, fo) > XFS_DIR2_LEAF_OFFSET + mp->m_dirblksize)
return 0;
lbp = state->path.blk[0].bp;
- leaf = lbp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
+ leaf = lbp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
/*
* Read the freespace block.
*/
- if ((error = xfs_da_read_buf(tp, dp, mp->m_dirfreeblk, -1, &fbp,
- XFS_DATA_FORK))) {
+ error = xfs_dir2_free_read(tp, dp, mp->m_dirfreeblk, &fbp);
+ if (error)
return error;
- }
- free = fbp->data;
- ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
+ free = fbp->b_addr;
+ ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
ASSERT(!free->hdr.firstdb);
+
/*
* Now see if the leafn and free data will fit in a leaf1.
* If not, release the buffer and give up.
*/
- if ((uint)sizeof(leaf->hdr) +
- (be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale)) * (uint)sizeof(leaf->ents[0]) +
- be32_to_cpu(free->hdr.nvalid) * (uint)sizeof(leaf->bests[0]) +
- (uint)sizeof(leaf->tail) >
- mp->m_dirblksize) {
- xfs_da_brelse(tp, fbp);
+ if (xfs_dir2_leaf_size(&leaf->hdr, be32_to_cpu(free->hdr.nvalid)) >
+ mp->m_dirblksize) {
+ xfs_trans_brelse(tp, fbp);
return 0;
}
+
/*
* If the leaf has any stale entries in it, compress them out.
* The compact routine will log the header.
@@ -1458,7 +1553,10 @@ xfs_dir2_node_to_leaf(
xfs_dir2_leaf_compact(args, lbp);
else
xfs_dir2_leaf_log_header(tp, lbp);
+
+ lbp->b_ops = &xfs_dir2_leaf1_buf_ops;
leaf->hdr.info.magic = cpu_to_be16(XFS_DIR2_LEAF1_MAGIC);
+
/*
* Set up the leaf tail from the freespace block.
*/
@@ -1468,7 +1566,7 @@ xfs_dir2_node_to_leaf(
* Set up the leaf bests table.
*/
memcpy(xfs_dir2_leaf_bests_p(ltp), free->bests,
- be32_to_cpu(ltp->bestcount) * sizeof(leaf->bests[0]));
+ be32_to_cpu(ltp->bestcount) * sizeof(xfs_dir2_data_off_t));
xfs_dir2_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
xfs_dir2_leaf_log_tail(tp, lbp);
xfs_dir2_leaf_check(dp, lbp);
diff --git a/libxfs/xfs_dir2_node.c b/libxfs/xfs_dir2_node.c
index e4e20d6..649f677 100644
--- a/libxfs/xfs_dir2_node.c
+++ b/libxfs/xfs_dir2_node.c
@@ -21,40 +21,108 @@
/*
* Function declarations.
*/
-static void xfs_dir2_free_log_header(xfs_trans_t *tp, xfs_dabuf_t *bp);
-static int xfs_dir2_leafn_add(xfs_dabuf_t *bp, xfs_da_args_t *args, int index);
+static int xfs_dir2_leafn_add(struct xfs_buf *bp, xfs_da_args_t *args,
+ int index);
#ifdef DEBUG
-static void xfs_dir2_leafn_check(xfs_inode_t *dp, xfs_dabuf_t *bp);
+static void xfs_dir2_leafn_check(struct xfs_inode *dp, struct xfs_buf *bp);
#else
#define xfs_dir2_leafn_check(dp, bp)
#endif
-static void xfs_dir2_leafn_moveents(xfs_da_args_t *args, xfs_dabuf_t *bp_s,
- int start_s, xfs_dabuf_t *bp_d, int start_d,
- int count);
+static void xfs_dir2_leafn_moveents(xfs_da_args_t *args, struct xfs_buf *bp_s,
+ int start_s, struct xfs_buf *bp_d,
+ int start_d, int count);
static void xfs_dir2_leafn_rebalance(xfs_da_state_t *state,
xfs_da_state_blk_t *blk1,
xfs_da_state_blk_t *blk2);
-static int xfs_dir2_leafn_remove(xfs_da_args_t *args, xfs_dabuf_t *bp,
+static int xfs_dir2_leafn_remove(xfs_da_args_t *args, struct xfs_buf *bp,
int index, xfs_da_state_blk_t *dblk,
int *rval);
static int xfs_dir2_node_addname_int(xfs_da_args_t *args,
xfs_da_state_blk_t *fblk);
+static void
+xfs_dir2_free_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_dir2_free_hdr *hdr = bp->b_addr;
+ int block_ok = 0;
+
+ block_ok = hdr->magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC);
+ if (!block_ok) {
+ XFS_CORRUPTION_ERROR("xfs_dir2_free_verify magic",
+ XFS_ERRLEVEL_LOW, mp, hdr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+}
+
+static void
+xfs_dir2_free_read_verify(
+ struct xfs_buf *bp)
+{
+ xfs_dir2_free_verify(bp);
+}
+
+static void
+xfs_dir2_free_write_verify(
+ struct xfs_buf *bp)
+{
+ xfs_dir2_free_verify(bp);
+}
+
+static const struct xfs_buf_ops xfs_dir2_free_buf_ops = {
+ .verify_read = xfs_dir2_free_read_verify,
+ .verify_write = xfs_dir2_free_write_verify,
+};
+
+
+static int
+__xfs_dir2_free_read(
+ struct xfs_trans *tp,
+ struct xfs_inode *dp,
+ xfs_dablk_t fbno,
+ xfs_daddr_t mappedbno,
+ struct xfs_buf **bpp)
+{
+ return xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
+ XFS_DATA_FORK, &xfs_dir2_free_buf_ops);
+}
+
+int
+xfs_dir2_free_read(
+ struct xfs_trans *tp,
+ struct xfs_inode *dp,
+ xfs_dablk_t fbno,
+ struct xfs_buf **bpp)
+{
+ return __xfs_dir2_free_read(tp, dp, fbno, -1, bpp);
+}
+
+static int
+xfs_dir2_free_try_read(
+ struct xfs_trans *tp,
+ struct xfs_inode *dp,
+ xfs_dablk_t fbno,
+ struct xfs_buf **bpp)
+{
+ return __xfs_dir2_free_read(tp, dp, fbno, -2, bpp);
+}
+
/*
* Log entries from a freespace block.
*/
STATIC void
xfs_dir2_free_log_bests(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_dabuf_t *bp, /* freespace buffer */
+ struct xfs_trans *tp,
+ struct xfs_buf *bp,
int first, /* first entry to log */
int last) /* last entry to log */
{
xfs_dir2_free_t *free; /* freespace structure */
- free = bp->data;
- ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
- xfs_da_log_buf(tp, bp,
+ free = bp->b_addr;
+ ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
+ xfs_trans_log_buf(tp, bp,
(uint)((char *)&free->bests[first] - (char *)free),
(uint)((char *)&free->bests[last] - (char *)free +
sizeof(free->bests[0]) - 1));
@@ -65,14 +133,14 @@ xfs_dir2_free_log_bests(
*/
static void
xfs_dir2_free_log_header(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_dabuf_t *bp) /* freespace buffer */
+ struct xfs_trans *tp,
+ struct xfs_buf *bp)
{
xfs_dir2_free_t *free; /* freespace structure */
- free = bp->data;
- ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
- xfs_da_log_buf(tp, bp, (uint)((char *)&free->hdr - (char *)free),
+ free = bp->b_addr;
+ ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
+ xfs_trans_log_buf(tp, bp, (uint)((char *)&free->hdr - (char *)free),
(uint)(sizeof(xfs_dir2_free_hdr_t) - 1));
}
@@ -84,11 +152,11 @@ xfs_dir2_free_log_header(
int /* error */
xfs_dir2_leaf_to_node(
xfs_da_args_t *args, /* operation arguments */
- xfs_dabuf_t *lbp) /* leaf buffer */
+ struct xfs_buf *lbp) /* leaf buffer */
{
xfs_inode_t *dp; /* incore directory inode */
int error; /* error return value */
- xfs_dabuf_t *fbp; /* freespace buffer */
+ struct xfs_buf *fbp; /* freespace buffer */
xfs_dir2_db_t fdb; /* freespace block number */
xfs_dir2_free_t *free; /* freespace structure */
__be16 *from; /* pointer to freespace entry */
@@ -116,13 +184,14 @@ xfs_dir2_leaf_to_node(
/*
* Get the buffer for the new freespace block.
*/
- if ((error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(mp, fdb), -1, &fbp,
- XFS_DATA_FORK))) {
+ error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(mp, fdb), -1, &fbp,
+ XFS_DATA_FORK);
+ if (error)
return error;
- }
- ASSERT(fbp != NULL);
- free = fbp->data;
- leaf = lbp->data;
+ fbp->b_ops = &xfs_dir2_free_buf_ops;
+
+ free = fbp->b_addr;
+ leaf = lbp->b_addr;
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
/*
* Initialize the freespace block header.
@@ -142,14 +211,16 @@ xfs_dir2_leaf_to_node(
*to = cpu_to_be16(off);
}
free->hdr.nused = cpu_to_be32(n);
+
+ lbp->b_ops = &xfs_dir2_leafn_buf_ops;
leaf->hdr.info.magic = cpu_to_be16(XFS_DIR2_LEAFN_MAGIC);
+
/*
* Log everything.
*/
xfs_dir2_leaf_log_header(tp, lbp);
xfs_dir2_free_log_header(tp, fbp);
xfs_dir2_free_log_bests(tp, fbp, 0, be32_to_cpu(free->hdr.nvalid) - 1);
- xfs_da_buf_done(fbp);
xfs_dir2_leafn_check(dp, lbp);
return 0;
}
@@ -160,7 +231,7 @@ xfs_dir2_leaf_to_node(
*/
static int /* error */
xfs_dir2_leafn_add(
- xfs_dabuf_t *bp, /* leaf buffer */
+ struct xfs_buf *bp, /* leaf buffer */
xfs_da_args_t *args, /* operation arguments */
int index) /* insertion pt for new entry */
{
@@ -180,7 +251,7 @@ xfs_dir2_leafn_add(
dp = args->dp;
mp = dp->i_mount;
tp = args->trans;
- leaf = bp->data;
+ leaf = bp->b_addr;
/*
* Quick check just to make sure we are not going to index
@@ -224,89 +295,13 @@ xfs_dir2_leafn_add(
lfloglow = be16_to_cpu(leaf->hdr.count);
lfloghigh = -1;
}
- /*
- * No stale entries, just insert a space for the new entry.
- */
- if (!leaf->hdr.stale) {
- lep = &leaf->ents[index];
- if (index < be16_to_cpu(leaf->hdr.count))
- memmove(lep + 1, lep,
- (be16_to_cpu(leaf->hdr.count) - index) * sizeof(*lep));
- lfloglow = index;
- lfloghigh = be16_to_cpu(leaf->hdr.count);
- be16_add_cpu(&leaf->hdr.count, 1);
- }
- /*
- * There are stale entries. We'll use one for the new entry.
- */
- else {
- /*
- * If we didn't do a compact then we need to figure out
- * which stale entry will be used.
- */
- if (compact == 0) {
- /*
- * Find first stale entry before our insertion point.
- */
- for (lowstale = index - 1;
- lowstale >= 0 &&
- be32_to_cpu(leaf->ents[lowstale].address) !=
- XFS_DIR2_NULL_DATAPTR;
- lowstale--)
- continue;
- /*
- * Find next stale entry after insertion point.
- * Stop looking if the answer would be worse than
- * lowstale already found.
- */
- for (highstale = index;
- highstale < be16_to_cpu(leaf->hdr.count) &&
- be32_to_cpu(leaf->ents[highstale].address) !=
- XFS_DIR2_NULL_DATAPTR &&
- (lowstale < 0 ||
- index - lowstale - 1 >= highstale - index);
- highstale++)
- continue;
- }
- /*
- * Using the low stale entry.
- * Shift entries up toward the stale slot.
- */
- if (lowstale >= 0 &&
- (highstale == be16_to_cpu(leaf->hdr.count) ||
- index - lowstale - 1 < highstale - index)) {
- ASSERT(be32_to_cpu(leaf->ents[lowstale].address) ==
- XFS_DIR2_NULL_DATAPTR);
- ASSERT(index - lowstale - 1 >= 0);
- if (index - lowstale - 1 > 0)
- memmove(&leaf->ents[lowstale],
- &leaf->ents[lowstale + 1],
- (index - lowstale - 1) * sizeof(*lep));
- lep = &leaf->ents[index - 1];
- lfloglow = MIN(lowstale, lfloglow);
- lfloghigh = MAX(index - 1, lfloghigh);
- }
- /*
- * Using the high stale entry.
- * Shift entries down toward the stale slot.
- */
- else {
- ASSERT(be32_to_cpu(leaf->ents[highstale].address) ==
- XFS_DIR2_NULL_DATAPTR);
- ASSERT(highstale - index >= 0);
- if (highstale - index > 0)
- memmove(&leaf->ents[index + 1],
- &leaf->ents[index],
- (highstale - index) * sizeof(*lep));
- lep = &leaf->ents[index];
- lfloglow = MIN(index, lfloglow);
- lfloghigh = MAX(highstale, lfloghigh);
- }
- be16_add_cpu(&leaf->hdr.stale, -1);
- }
+
/*
* Insert the new entry, log everything.
*/
+ lep = xfs_dir2_leaf_find_entry(leaf, index, compact, lowstale,
+ highstale, &lfloglow, &lfloghigh);
+
lep->hashval = cpu_to_be32(args->hashval);
lep->address = cpu_to_be32(xfs_dir2_db_off_to_dataptr(mp,
args->blkno, args->index));
@@ -322,24 +317,24 @@ xfs_dir2_leafn_add(
*/
void
xfs_dir2_leafn_check(
- xfs_inode_t *dp, /* incore directory inode */
- xfs_dabuf_t *bp) /* leaf buffer */
+ struct xfs_inode *dp,
+ struct xfs_buf *bp)
{
int i; /* leaf index */
xfs_dir2_leaf_t *leaf; /* leaf structure */
xfs_mount_t *mp; /* filesystem mount point */
int stale; /* count of stale leaves */
- leaf = bp->data;
+ leaf = bp->b_addr;
mp = dp->i_mount;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
ASSERT(be16_to_cpu(leaf->hdr.count) <= xfs_dir2_max_leaf_ents(mp));
for (i = stale = 0; i < be16_to_cpu(leaf->hdr.count); i++) {
if (i + 1 < be16_to_cpu(leaf->hdr.count)) {
ASSERT(be32_to_cpu(leaf->ents[i].hashval) <=
be32_to_cpu(leaf->ents[i + 1].hashval));
}
- if (be32_to_cpu(leaf->ents[i].address) == XFS_DIR2_NULL_DATAPTR)
+ if (leaf->ents[i].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
stale++;
}
ASSERT(be16_to_cpu(leaf->hdr.stale) == stale);
@@ -352,13 +347,13 @@ xfs_dir2_leafn_check(
*/
xfs_dahash_t /* hash value */
xfs_dir2_leafn_lasthash(
- xfs_dabuf_t *bp, /* leaf buffer */
+ struct xfs_buf *bp, /* leaf buffer */
int *count) /* count of entries in leaf */
{
xfs_dir2_leaf_t *leaf; /* leaf structure */
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
+ leaf = bp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
if (count)
*count = be16_to_cpu(leaf->hdr.count);
if (!leaf->hdr.count)
@@ -372,12 +367,12 @@ xfs_dir2_leafn_lasthash(
*/
STATIC int
xfs_dir2_leafn_lookup_for_addname(
- xfs_dabuf_t *bp, /* leaf buffer */
+ struct xfs_buf *bp, /* leaf buffer */
xfs_da_args_t *args, /* operation arguments */
int *indexp, /* out: leaf entry index */
xfs_da_state_t *state) /* state to fill in */
{
- xfs_dabuf_t *curbp = NULL; /* current data/free buffer */
+ struct xfs_buf *curbp = NULL; /* current data/free buffer */
xfs_dir2_db_t curdb = -1; /* current data block number */
xfs_dir2_db_t curfdb = -1; /* current free block number */
xfs_inode_t *dp; /* incore directory inode */
@@ -396,8 +391,8 @@ xfs_dir2_leafn_lookup_for_addname(
dp = args->dp;
tp = args->trans;
mp = dp->i_mount;
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
+ leaf = bp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
#ifdef __KERNEL__
ASSERT(be16_to_cpu(leaf->hdr.count) > 0);
#endif
@@ -413,8 +408,8 @@ xfs_dir2_leafn_lookup_for_addname(
/* If so, it's a free block buffer, get the block number. */
curbp = state->extrablk.bp;
curfdb = state->extrablk.blkno;
- free = curbp->data;
- ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
+ free = curbp->b_addr;
+ ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
}
length = xfs_dir2_data_entsize(args->namelen);
/*
@@ -455,20 +450,18 @@ xfs_dir2_leafn_lookup_for_addname(
* If we had one before, drop it.
*/
if (curbp)
- xfs_da_brelse(tp, curbp);
- /*
- * Read the free block.
- */
- error = xfs_da_read_buf(tp, dp,
+ xfs_trans_brelse(tp, curbp);
+
+ error = xfs_dir2_free_read(tp, dp,
xfs_dir2_db_to_da(mp, newfdb),
- -1, &curbp, XFS_DATA_FORK);
+ &curbp);
if (error)
return error;
- free = curbp->data;
+ free = curbp->b_addr;
ASSERT(be32_to_cpu(free->hdr.magic) ==
XFS_DIR2_FREE_MAGIC);
ASSERT((be32_to_cpu(free->hdr.firstdb) %
- XFS_DIR2_MAX_FREE_BESTS(mp)) == 0);
+ xfs_dir2_free_max_bests(mp)) == 0);
ASSERT(be32_to_cpu(free->hdr.firstdb) <= curdb);
ASSERT(curdb < be32_to_cpu(free->hdr.firstdb) +
be32_to_cpu(free->hdr.nvalid));
@@ -480,11 +473,12 @@ xfs_dir2_leafn_lookup_for_addname(
/*
* If it has room, return it.
*/
- if (unlikely(be16_to_cpu(free->bests[fi]) == NULLDATAOFF)) {
+ if (unlikely(free->bests[fi] ==
+ cpu_to_be16(NULLDATAOFF))) {
XFS_ERROR_REPORT("xfs_dir2_leafn_lookup_int",
XFS_ERRLEVEL_LOW, mp);
if (curfdb != newfdb)
- xfs_da_brelse(tp, curbp);
+ xfs_trans_brelse(tp, curbp);
return XFS_ERROR(EFSCORRUPTED);
}
curfdb = newfdb;
@@ -519,12 +513,12 @@ out:
*/
STATIC int
xfs_dir2_leafn_lookup_for_entry(
- xfs_dabuf_t *bp, /* leaf buffer */
+ struct xfs_buf *bp, /* leaf buffer */
xfs_da_args_t *args, /* operation arguments */
int *indexp, /* out: leaf entry index */
xfs_da_state_t *state) /* state to fill in */
{
- xfs_dabuf_t *curbp = NULL; /* current data/free buffer */
+ struct xfs_buf *curbp = NULL; /* current data/free buffer */
xfs_dir2_db_t curdb = -1; /* current data block number */
xfs_dir2_data_entry_t *dep; /* data block entry */
xfs_inode_t *dp; /* incore directory inode */
@@ -540,8 +534,8 @@ xfs_dir2_leafn_lookup_for_entry(
dp = args->dp;
tp = args->trans;
mp = dp->i_mount;
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
+ leaf = bp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
#ifdef __KERNEL__
ASSERT(be16_to_cpu(leaf->hdr.count) > 0);
#endif
@@ -585,7 +579,7 @@ xfs_dir2_leafn_lookup_for_entry(
*/
if (curbp && (args->cmpresult == XFS_CMP_DIFFERENT ||
curdb != state->extrablk.blkno))
- xfs_da_brelse(tp, curbp);
+ xfs_trans_brelse(tp, curbp);
/*
* If needing the block that is saved with a CI match,
* use it otherwise read in the new data block.
@@ -595,9 +589,9 @@ xfs_dir2_leafn_lookup_for_entry(
ASSERT(state->extravalid);
curbp = state->extrablk.bp;
} else {
- error = xfs_da_read_buf(tp, dp,
+ error = xfs_dir2_data_read(tp, dp,
xfs_dir2_db_to_da(mp, newdb),
- -1, &curbp, XFS_DATA_FORK);
+ -1, &curbp);
if (error)
return error;
}
@@ -607,7 +601,7 @@ xfs_dir2_leafn_lookup_for_entry(
/*
* Point to the data entry.
*/
- dep = (xfs_dir2_data_entry_t *)((char *)curbp->data +
+ dep = (xfs_dir2_data_entry_t *)((char *)curbp->b_addr +
xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address)));
/*
* Compare the entry and if it's an exact match, return
@@ -619,7 +613,7 @@ xfs_dir2_leafn_lookup_for_entry(
/* If there is a CI match block, drop it */
if (args->cmpresult != XFS_CMP_DIFFERENT &&
curdb != state->extrablk.blkno)
- xfs_da_brelse(tp, state->extrablk.bp);
+ xfs_trans_brelse(tp, state->extrablk.bp);
args->cmpresult = cmp;
args->inumber = be64_to_cpu(dep->inumber);
*indexp = index;
@@ -627,8 +621,9 @@ xfs_dir2_leafn_lookup_for_entry(
state->extrablk.bp = curbp;
state->extrablk.blkno = curdb;
state->extrablk.index = (int)((char *)dep -
- (char *)curbp->data);
+ (char *)curbp->b_addr);
state->extrablk.magic = XFS_DIR2_DATA_MAGIC;
+ curbp->b_ops = &xfs_dir2_data_buf_ops;
if (cmp == XFS_CMP_EXACT)
return XFS_ERROR(EEXIST);
}
@@ -643,10 +638,11 @@ xfs_dir2_leafn_lookup_for_entry(
state->extrablk.index = -1;
state->extrablk.blkno = curdb;
state->extrablk.magic = XFS_DIR2_DATA_MAGIC;
+ curbp->b_ops = &xfs_dir2_data_buf_ops;
} else {
/* If the curbp is not the CI match block, drop it */
if (state->extrablk.bp != curbp)
- xfs_da_brelse(tp, curbp);
+ xfs_trans_brelse(tp, curbp);
}
} else {
state->extravalid = 0;
@@ -662,7 +658,7 @@ xfs_dir2_leafn_lookup_for_entry(
*/
int
xfs_dir2_leafn_lookup_int(
- xfs_dabuf_t *bp, /* leaf buffer */
+ struct xfs_buf *bp, /* leaf buffer */
xfs_da_args_t *args, /* operation arguments */
int *indexp, /* out: leaf entry index */
xfs_da_state_t *state) /* state to fill in */
@@ -680,9 +676,9 @@ xfs_dir2_leafn_lookup_int(
static void
xfs_dir2_leafn_moveents(
xfs_da_args_t *args, /* operation arguments */
- xfs_dabuf_t *bp_s, /* source leaf buffer */
+ struct xfs_buf *bp_s, /* source leaf buffer */
int start_s, /* source leaf index */
- xfs_dabuf_t *bp_d, /* destination leaf buffer */
+ struct xfs_buf *bp_d, /* destination leaf buffer */
int start_d, /* destination leaf index */
int count) /* count of leaves to copy */
{
@@ -700,8 +696,8 @@ xfs_dir2_leafn_moveents(
return;
}
tp = args->trans;
- leaf_s = bp_s->data;
- leaf_d = bp_d->data;
+ leaf_s = bp_s->b_addr;
+ leaf_d = bp_d->b_addr;
/*
* If the destination index is not the end of the current
* destination leaf entries, open up a hole in the destination
@@ -722,7 +718,8 @@ xfs_dir2_leafn_moveents(
int i; /* temp leaf index */
for (i = start_s, stale = 0; i < start_s + count; i++) {
- if (be32_to_cpu(leaf_s->ents[i].address) == XFS_DIR2_NULL_DATAPTR)
+ if (leaf_s->ents[i].address ==
+ cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
stale++;
}
} else
@@ -761,16 +758,16 @@ xfs_dir2_leafn_moveents(
*/
int /* sort order */
xfs_dir2_leafn_order(
- xfs_dabuf_t *leaf1_bp, /* leaf1 buffer */
- xfs_dabuf_t *leaf2_bp) /* leaf2 buffer */
+ struct xfs_buf *leaf1_bp, /* leaf1 buffer */
+ struct xfs_buf *leaf2_bp) /* leaf2 buffer */
{
xfs_dir2_leaf_t *leaf1; /* leaf1 structure */
xfs_dir2_leaf_t *leaf2; /* leaf2 structure */
- leaf1 = leaf1_bp->data;
- leaf2 = leaf2_bp->data;
- ASSERT(be16_to_cpu(leaf1->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
- ASSERT(be16_to_cpu(leaf2->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
+ leaf1 = leaf1_bp->b_addr;
+ leaf2 = leaf2_bp->b_addr;
+ ASSERT(leaf1->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
+ ASSERT(leaf2->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
if (be16_to_cpu(leaf1->hdr.count) > 0 &&
be16_to_cpu(leaf2->hdr.count) > 0 &&
(be32_to_cpu(leaf2->ents[0].hashval) < be32_to_cpu(leaf1->ents[0].hashval) ||
@@ -816,8 +813,8 @@ xfs_dir2_leafn_rebalance(
blk1 = blk2;
blk2 = tmp;
}
- leaf1 = blk1->bp->data;
- leaf2 = blk2->bp->data;
+ leaf1 = blk1->bp->b_addr;
+ leaf2 = blk2->bp->b_addr;
oldsum = be16_to_cpu(leaf1->hdr.count) + be16_to_cpu(leaf2->hdr.count);
#ifdef DEBUG
oldstale = be16_to_cpu(leaf1->hdr.stale) + be16_to_cpu(leaf2->hdr.stale);
@@ -879,11 +876,81 @@ xfs_dir2_leafn_rebalance(
if(blk2->index < 0) {
state->inleaf = 1;
blk2->index = 0;
- cmn_err(CE_ALERT,
- "xfs_dir2_leafn_rebalance: picked the wrong leaf? reverting original leaf: "
- "blk1->index %d\n",
- blk1->index);
+ xfs_alert(args->dp->i_mount,
+ "%s: picked the wrong leaf? reverting original leaf: blk1->index %d\n",
+ __func__, blk1->index);
+ }
+}
+
+static int
+xfs_dir2_data_block_free(
+ xfs_da_args_t *args,
+ struct xfs_dir2_data_hdr *hdr,
+ struct xfs_dir2_free *free,
+ xfs_dir2_db_t fdb,
+ int findex,
+ struct xfs_buf *fbp,
+ int longest)
+{
+ struct xfs_trans *tp = args->trans;
+ int logfree = 0;
+
+ if (!hdr) {
+ /* One less used entry in the free table. */
+ be32_add_cpu(&free->hdr.nused, -1);
+ xfs_dir2_free_log_header(tp, fbp);
+
+ /*
+ * If this was the last entry in the table, we can trim the
+ * table size back. There might be other entries at the end
+ * referring to non-existent data blocks, get those too.
+ */
+ if (findex == be32_to_cpu(free->hdr.nvalid) - 1) {
+ int i; /* free entry index */
+
+ for (i = findex - 1; i >= 0; i--) {
+ if (free->bests[i] != cpu_to_be16(NULLDATAOFF))
+ break;
+ }
+ free->hdr.nvalid = cpu_to_be32(i + 1);
+ logfree = 0;
+ } else {
+ /* Not the last entry, just punch it out. */
+ free->bests[findex] = cpu_to_be16(NULLDATAOFF);
+ logfree = 1;
+ }
+ /*
+ * If there are no useful entries left in the block,
+ * get rid of the block if we can.
+ */
+ if (!free->hdr.nused) {
+ int error;
+
+ error = xfs_dir2_shrink_inode(args, fdb, fbp);
+ if (error == 0) {
+ fbp = NULL;
+ logfree = 0;
+ } else if (error != ENOSPC || args->total != 0)
+ return error;
+ /*
+ * It's possible to get ENOSPC if there is no
+ * space reservation. In this case some one
+ * else will eventually get rid of this block.
+ */
+ }
+ } else {
+ /*
+ * Data block is not empty, just set the free entry to the new
+ * value.
+ */
+ free->bests[findex] = cpu_to_be16(longest);
+ logfree = 1;
}
+
+ /* Log the free entry that changed, unless we got rid of it. */
+ if (logfree)
+ xfs_dir2_free_log_bests(tp, fbp, findex, findex);
+ return 0;
}
/*
@@ -894,14 +961,14 @@ xfs_dir2_leafn_rebalance(
static int /* error */
xfs_dir2_leafn_remove(
xfs_da_args_t *args, /* operation arguments */
- xfs_dabuf_t *bp, /* leaf buffer */
+ struct xfs_buf *bp, /* leaf buffer */
int index, /* leaf entry index */
xfs_da_state_blk_t *dblk, /* data block */
int *rval) /* resulting block needs join */
{
- xfs_dir2_data_t *data; /* data block structure */
+ xfs_dir2_data_hdr_t *hdr; /* data block header */
xfs_dir2_db_t db; /* data block number */
- xfs_dabuf_t *dbp; /* data block buffer */
+ struct xfs_buf *dbp; /* data block buffer */
xfs_dir2_data_entry_t *dep; /* data block entry */
xfs_inode_t *dp; /* incore directory inode */
xfs_dir2_leaf_t *leaf; /* leaf structure */
@@ -918,8 +985,8 @@ xfs_dir2_leafn_remove(
dp = args->dp;
tp = args->trans;
mp = dp->i_mount;
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
+ leaf = bp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
/*
* Point to the entry we're removing.
*/
@@ -944,9 +1011,9 @@ xfs_dir2_leafn_remove(
* in the data block in case it changes.
*/
dbp = dblk->bp;
- data = dbp->data;
- dep = (xfs_dir2_data_entry_t *)((char *)data + off);
- longest = be16_to_cpu(data->hdr.bestfree[0].length);
+ hdr = dbp->b_addr;
+ dep = (xfs_dir2_data_entry_t *)((char *)hdr + off);
+ longest = be16_to_cpu(hdr->bestfree[0].length);
needlog = needscan = 0;
xfs_dir2_data_make_free(tp, dbp, off,
xfs_dir2_data_entsize(dep->namelen), &needlog, &needscan);
@@ -955,7 +1022,7 @@ xfs_dir2_leafn_remove(
* Log the data block header if needed.
*/
if (needscan)
- xfs_dir2_data_freescan(mp, data, &needlog);
+ xfs_dir2_data_freescan(mp, hdr, &needlog);
if (needlog)
xfs_dir2_data_log_header(tp, dbp);
xfs_dir2_data_check(dp, dbp);
@@ -963,126 +1030,63 @@ xfs_dir2_leafn_remove(
* If the longest data block freespace changes, need to update
* the corresponding freeblock entry.
*/
- if (longest < be16_to_cpu(data->hdr.bestfree[0].length)) {
+ if (longest < be16_to_cpu(hdr->bestfree[0].length)) {
int error; /* error return value */
- xfs_dabuf_t *fbp; /* freeblock buffer */
+ struct xfs_buf *fbp; /* freeblock buffer */
xfs_dir2_db_t fdb; /* freeblock block number */
int findex; /* index in freeblock entries */
xfs_dir2_free_t *free; /* freeblock structure */
- int logfree; /* need to log free entry */
/*
* Convert the data block number to a free block,
* read in the free block.
*/
fdb = xfs_dir2_db_to_fdb(mp, db);
- if ((error = xfs_da_read_buf(tp, dp, xfs_dir2_db_to_da(mp, fdb),
- -1, &fbp, XFS_DATA_FORK))) {
+ error = xfs_dir2_free_read(tp, dp, xfs_dir2_db_to_da(mp, fdb),
+ &fbp);
+ if (error)
return error;
- }
- free = fbp->data;
- ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
+ free = fbp->b_addr;
+ ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
ASSERT(be32_to_cpu(free->hdr.firstdb) ==
- XFS_DIR2_MAX_FREE_BESTS(mp) *
+ xfs_dir2_free_max_bests(mp) *
(fdb - XFS_DIR2_FREE_FIRSTDB(mp)));
/*
* Calculate which entry we need to fix.
*/
findex = xfs_dir2_db_to_fdindex(mp, db);
- longest = be16_to_cpu(data->hdr.bestfree[0].length);
+ longest = be16_to_cpu(hdr->bestfree[0].length);
/*
* If the data block is now empty we can get rid of it
* (usually).
*/
- if (longest == mp->m_dirblksize - (uint)sizeof(data->hdr)) {
+ if (longest == mp->m_dirblksize - (uint)sizeof(*hdr)) {
/*
* Try to punch out the data block.
*/
error = xfs_dir2_shrink_inode(args, db, dbp);
if (error == 0) {
dblk->bp = NULL;
- data = NULL;
+ hdr = NULL;
}
/*
* We can get ENOSPC if there's no space reservation.
* In this case just drop the buffer and some one else
* will eventually get rid of the empty block.
*/
- else if (error == ENOSPC && args->total == 0)
- xfs_da_buf_done(dbp);
- else
+ else if (!(error == ENOSPC && args->total == 0))
return error;
}
/*
* If we got rid of the data block, we can eliminate that entry
* in the free block.
*/
- if (data == NULL) {
- /*
- * One less used entry in the free table.
- */
- be32_add_cpu(&free->hdr.nused, -1);
- xfs_dir2_free_log_header(tp, fbp);
- /*
- * If this was the last entry in the table, we can
- * trim the table size back. There might be other
- * entries at the end referring to non-existent
- * data blocks, get those too.
- */
- if (findex == be32_to_cpu(free->hdr.nvalid) - 1) {
- int i; /* free entry index */
-
- for (i = findex - 1;
- i >= 0 && be16_to_cpu(free->bests[i]) == NULLDATAOFF;
- i--)
- continue;
- free->hdr.nvalid = cpu_to_be32(i + 1);
- logfree = 0;
- }
- /*
- * Not the last entry, just punch it out.
- */
- else {
- free->bests[findex] = cpu_to_be16(NULLDATAOFF);
- logfree = 1;
- }
- /*
- * If there are no useful entries left in the block,
- * get rid of the block if we can.
- */
- if (!free->hdr.nused) {
- error = xfs_dir2_shrink_inode(args, fdb, fbp);
- if (error == 0) {
- fbp = NULL;
- logfree = 0;
- } else if (error != ENOSPC || args->total != 0)
- return error;
- /*
- * It's possible to get ENOSPC if there is no
- * space reservation. In this case some one
- * else will eventually get rid of this block.
- */
- }
- }
- /*
- * Data block is not empty, just set the free entry to
- * the new value.
- */
- else {
- free->bests[findex] = cpu_to_be16(longest);
- logfree = 1;
- }
- /*
- * Log the free entry that changed, unless we got rid of it.
- */
- if (logfree)
- xfs_dir2_free_log_bests(tp, fbp, findex, findex);
- /*
- * Drop the buffer if we still have it.
- */
- if (fbp)
- xfs_da_buf_done(fbp);
+ error = xfs_dir2_data_block_free(args, hdr, free,
+ fdb, findex, fbp, longest);
+ if (error)
+ return error;
}
+
xfs_dir2_leafn_check(dp, bp);
/*
* Return indication of whether this leaf block is empty enough
@@ -1173,7 +1177,7 @@ xfs_dir2_leafn_toosmall(
{
xfs_da_state_blk_t *blk; /* leaf block */
xfs_dablk_t blkno; /* leaf block number */
- xfs_dabuf_t *bp; /* leaf buffer */
+ struct xfs_buf *bp; /* leaf buffer */
int bytes; /* bytes in use */
int count; /* leaf live entry count */
int error; /* error return value */
@@ -1189,8 +1193,8 @@ xfs_dir2_leafn_toosmall(
* to coalesce with a sibling.
*/
blk = &state->path.blk[state->path.active - 1];
- info = blk->bp->data;
- ASSERT(be16_to_cpu(info->magic) == XFS_DIR2_LEAFN_MAGIC);
+ info = blk->bp->b_addr;
+ ASSERT(info->magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
leaf = (xfs_dir2_leaf_t *)info;
count = be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale);
bytes = (uint)sizeof(leaf->hdr) + count * (uint)sizeof(leaf->ents[0]);
@@ -1236,20 +1240,19 @@ xfs_dir2_leafn_toosmall(
/*
* Read the sibling leaf block.
*/
- if ((error =
- xfs_da_read_buf(state->args->trans, state->args->dp, blkno,
- -1, &bp, XFS_DATA_FORK))) {
+ error = xfs_dir2_leafn_read(state->args->trans, state->args->dp,
+ blkno, -1, &bp);
+ if (error)
return error;
- }
- ASSERT(bp != NULL);
+
/*
* Count bytes in the two blocks combined.
*/
leaf = (xfs_dir2_leaf_t *)info;
count = be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale);
bytes = state->blocksize - (state->blocksize >> 2);
- leaf = bp->data;
- ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
+ leaf = bp->b_addr;
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
count += be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale);
bytes -= count * (uint)sizeof(leaf->ents[0]);
/*
@@ -1257,7 +1260,7 @@ xfs_dir2_leafn_toosmall(
*/
if (bytes >= 0)
break;
- xfs_da_brelse(state->args->trans, bp);
+ xfs_trans_brelse(state->args->trans, bp);
}
/*
* Didn't like either block, give up.
@@ -1266,11 +1269,7 @@ xfs_dir2_leafn_toosmall(
*action = 0;
return 0;
}
- /*
- * Done with the sibling leaf block here, drop the dabuf
- * so path_shift can get it.
- */
- xfs_da_buf_done(bp);
+
/*
* Make altpath point to the block we want to keep (the lower
* numbered block) and path point to the block we want to drop.
@@ -1306,10 +1305,10 @@ xfs_dir2_leafn_unbalance(
args = state->args;
ASSERT(drop_blk->magic == XFS_DIR2_LEAFN_MAGIC);
ASSERT(save_blk->magic == XFS_DIR2_LEAFN_MAGIC);
- drop_leaf = drop_blk->bp->data;
- save_leaf = save_blk->bp->data;
- ASSERT(be16_to_cpu(drop_leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
- ASSERT(be16_to_cpu(save_leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
+ drop_leaf = drop_blk->bp->b_addr;
+ save_leaf = save_blk->bp->b_addr;
+ ASSERT(drop_leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
+ ASSERT(save_leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
/*
* If there are any stale leaf entries, take this opportunity
* to purge them.
@@ -1413,15 +1412,15 @@ xfs_dir2_node_addname_int(
xfs_da_args_t *args, /* operation arguments */
xfs_da_state_blk_t *fblk) /* optional freespace block */
{
- xfs_dir2_data_t *data; /* data block structure */
+ xfs_dir2_data_hdr_t *hdr; /* data block header */
xfs_dir2_db_t dbno; /* data block number */
- xfs_dabuf_t *dbp; /* data block buffer */
+ struct xfs_buf *dbp; /* data block buffer */
xfs_dir2_data_entry_t *dep; /* data entry pointer */
xfs_inode_t *dp; /* incore directory inode */
xfs_dir2_data_unused_t *dup; /* data unused entry pointer */
int error; /* error return value */
xfs_dir2_db_t fbno; /* freespace block number */
- xfs_dabuf_t *fbp; /* freespace buffer */
+ struct xfs_buf *fbp; /* freespace buffer */
int findex; /* freespace entry index */
xfs_dir2_free_t *free=NULL; /* freespace block structure */
xfs_dir2_db_t ifbno; /* initial freespace block no */
@@ -1449,8 +1448,8 @@ xfs_dir2_node_addname_int(
* Remember initial freespace block number.
*/
ifbno = fblk->blkno;
- free = fbp->data;
- ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
+ free = fbp->b_addr;
+ ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
findex = fblk->index;
/*
* This means the free entry showed that the data block had
@@ -1525,16 +1524,15 @@ xfs_dir2_node_addname_int(
* This should be really rare, so there's no reason
* to avoid it.
*/
- if ((error = xfs_da_read_buf(tp, dp,
- xfs_dir2_db_to_da(mp, fbno), -2, &fbp,
- XFS_DATA_FORK))) {
+ error = xfs_dir2_free_try_read(tp, dp,
+ xfs_dir2_db_to_da(mp, fbno),
+ &fbp);
+ if (error)
return error;
- }
- if (unlikely(fbp == NULL)) {
+ if (!fbp)
continue;
- }
- free = fbp->data;
- ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
+ free = fbp->b_addr;
+ ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
findex = 0;
}
/*
@@ -1551,7 +1549,7 @@ xfs_dir2_node_addname_int(
/*
* Drop the block.
*/
- xfs_da_brelse(tp, fbp);
+ xfs_trans_brelse(tp, fbp);
fbp = NULL;
if (fblk && fblk->bp)
fblk->bp = NULL;
@@ -1566,36 +1564,23 @@ xfs_dir2_node_addname_int(
/*
* Not allowed to allocate, return failure.
*/
- if ((args->op_flags & XFS_DA_OP_JUSTCHECK) ||
- args->total == 0) {
- /*
- * Drop the freespace buffer unless it came from our
- * caller.
- */
- if ((fblk == NULL || fblk->bp == NULL) && fbp != NULL)
- xfs_da_buf_done(fbp);
+ if ((args->op_flags & XFS_DA_OP_JUSTCHECK) || args->total == 0)
return XFS_ERROR(ENOSPC);
- }
+
/*
* Allocate and initialize the new data block.
*/
if (unlikely((error = xfs_dir2_grow_inode(args,
XFS_DIR2_DATA_SPACE,
&dbno)) ||
- (error = xfs_dir2_data_init(args, dbno, &dbp)))) {
- /*
- * Drop the freespace buffer unless it came from our
- * caller.
- */
- if ((fblk == NULL || fblk->bp == NULL) && fbp != NULL)
- xfs_da_buf_done(fbp);
+ (error = xfs_dir2_data_init(args, dbno, &dbp))))
return error;
- }
+
/*
* If (somehow) we have a freespace block, get rid of it.
*/
if (fbp)
- xfs_da_brelse(tp, fbp);
+ xfs_trans_brelse(tp, fbp);
if (fblk && fblk->bp)
fblk->bp = NULL;
@@ -1604,12 +1589,12 @@ xfs_dir2_node_addname_int(
* that was just allocated.
*/
fbno = xfs_dir2_db_to_fdb(mp, dbno);
- if (unlikely(error = xfs_da_read_buf(tp, dp,
- xfs_dir2_db_to_da(mp, fbno), -2, &fbp,
- XFS_DATA_FORK))) {
- xfs_da_buf_done(dbp);
+ error = xfs_dir2_free_try_read(tp, dp,
+ xfs_dir2_db_to_da(mp, fbno),
+ &fbp);
+ if (error)
return error;
- }
+
/*
* If there wasn't a freespace block, the read will
* return a NULL fbp. Allocate and initialize a new one.
@@ -1621,26 +1606,22 @@ xfs_dir2_node_addname_int(
}
if (unlikely(xfs_dir2_db_to_fdb(mp, dbno) != fbno)) {
- cmn_err(CE_ALERT,
- "xfs_dir2_node_addname_int: dir ino "
- "%llu needed freesp block %lld for\n"
- " data block %lld, got %lld\n"
- " ifbno %llu lastfbno %d\n",
- (unsigned long long)dp->i_ino,
+ xfs_alert(mp,
+ "%s: dir ino %llu needed freesp block %lld for\n"
+ " data block %lld, got %lld ifbno %llu lastfbno %d",
+ __func__, (unsigned long long)dp->i_ino,
(long long)xfs_dir2_db_to_fdb(mp, dbno),
(long long)dbno, (long long)fbno,
(unsigned long long)ifbno, lastfbno);
if (fblk) {
- cmn_err(CE_ALERT,
- " fblk 0x%p blkno %llu "
- "index %d magic 0x%x\n",
+ xfs_alert(mp,
+ " fblk 0x%p blkno %llu index %d magic 0x%x",
fblk,
(unsigned long long)fblk->blkno,
fblk->index,
fblk->magic);
} else {
- cmn_err(CE_ALERT,
- " ... fblk is NULL\n");
+ xfs_alert(mp, " ... fblk is NULL");
}
XFS_ERROR_REPORT("xfs_dir2_node_addname_int",
XFS_ERRLEVEL_LOW, mp);
@@ -1650,27 +1631,27 @@ xfs_dir2_node_addname_int(
/*
* Get a buffer for the new block.
*/
- if ((error = xfs_da_get_buf(tp, dp,
- xfs_dir2_db_to_da(mp, fbno),
- -1, &fbp, XFS_DATA_FORK))) {
+ error = xfs_da_get_buf(tp, dp,
+ xfs_dir2_db_to_da(mp, fbno),
+ -1, &fbp, XFS_DATA_FORK);
+ if (error)
return error;
- }
- ASSERT(fbp != NULL);
+ fbp->b_ops = &xfs_dir2_free_buf_ops;
/*
* Initialize the new block to be empty, and remember
* its first slot as our empty slot.
*/
- free = fbp->data;
+ free = fbp->b_addr;
free->hdr.magic = cpu_to_be32(XFS_DIR2_FREE_MAGIC);
free->hdr.firstdb = cpu_to_be32(
(fbno - XFS_DIR2_FREE_FIRSTDB(mp)) *
- XFS_DIR2_MAX_FREE_BESTS(mp));
+ xfs_dir2_free_max_bests(mp));
free->hdr.nvalid = 0;
free->hdr.nused = 0;
} else {
- free = fbp->data;
- ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
+ free = fbp->b_addr;
+ ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
}
/*
@@ -1682,7 +1663,7 @@ xfs_dir2_node_addname_int(
* freespace block, extend that table.
*/
if (findex >= be32_to_cpu(free->hdr.nvalid)) {
- ASSERT(findex < XFS_DIR2_MAX_FREE_BESTS(mp));
+ ASSERT(findex < xfs_dir2_free_max_bests(mp));
free->hdr.nvalid = cpu_to_be32(findex + 1);
/*
* Tag new entry so nused will go up.
@@ -1693,7 +1674,7 @@ xfs_dir2_node_addname_int(
* If this entry was for an empty data block
* (this should always be true) then update the header.
*/
- if (be16_to_cpu(free->bests[findex]) == NULLDATAOFF) {
+ if (free->bests[findex] == cpu_to_be16(NULLDATAOFF)) {
be32_add_cpu(&free->hdr.nused, 1);
xfs_dir2_free_log_header(tp, fbp);
}
@@ -1702,8 +1683,8 @@ xfs_dir2_node_addname_int(
* We haven't allocated the data entry yet so this will
* change again.
*/
- data = dbp->data;
- free->bests[findex] = data->hdr.bestfree[0].length;
+ hdr = dbp->b_addr;
+ free->bests[findex] = hdr->bestfree[0].length;
logfree = 1;
}
/*
@@ -1713,36 +1694,31 @@ xfs_dir2_node_addname_int(
/*
* If just checking, we succeeded.
*/
- if (args->op_flags & XFS_DA_OP_JUSTCHECK) {
- if ((fblk == NULL || fblk->bp == NULL) && fbp != NULL)
- xfs_da_buf_done(fbp);
+ if (args->op_flags & XFS_DA_OP_JUSTCHECK)
return 0;
- }
+
/*
* Read the data block in.
*/
- if (unlikely(
- error = xfs_da_read_buf(tp, dp, xfs_dir2_db_to_da(mp, dbno),
- -1, &dbp, XFS_DATA_FORK))) {
- if ((fblk == NULL || fblk->bp == NULL) && fbp != NULL)
- xfs_da_buf_done(fbp);
+ error = xfs_dir2_data_read(tp, dp, xfs_dir2_db_to_da(mp, dbno),
+ -1, &dbp);
+ if (error)
return error;
- }
- data = dbp->data;
+ hdr = dbp->b_addr;
logfree = 0;
}
- ASSERT(be16_to_cpu(data->hdr.bestfree[0].length) >= length);
+ ASSERT(be16_to_cpu(hdr->bestfree[0].length) >= length);
/*
* Point to the existing unused space.
*/
dup = (xfs_dir2_data_unused_t *)
- ((char *)data + be16_to_cpu(data->hdr.bestfree[0].offset));
+ ((char *)hdr + be16_to_cpu(hdr->bestfree[0].offset));
needscan = needlog = 0;
/*
* Mark the first part of the unused space, inuse for us.
*/
xfs_dir2_data_use_free(tp, dbp, dup,
- (xfs_dir2_data_aoff_t)((char *)dup - (char *)data), length,
+ (xfs_dir2_data_aoff_t)((char *)dup - (char *)hdr), length,
&needlog, &needscan);
/*
* Fill in the new entry and log it.
@@ -1752,13 +1728,13 @@ xfs_dir2_node_addname_int(
dep->namelen = args->namelen;
memcpy(dep->name, args->name, dep->namelen);
tagp = xfs_dir2_data_entry_tag_p(dep);
- *tagp = cpu_to_be16((char *)dep - (char *)data);
+ *tagp = cpu_to_be16((char *)dep - (char *)hdr);
xfs_dir2_data_log_entry(tp, dbp, dep);
/*
* Rescan the block for bestfree if needed.
*/
if (needscan)
- xfs_dir2_data_freescan(mp, data, &needlog);
+ xfs_dir2_data_freescan(mp, hdr, &needlog);
/*
* Log the data block header if needed.
*/
@@ -1767,8 +1743,8 @@ xfs_dir2_node_addname_int(
/*
* If the freespace entry is now wrong, update it.
*/
- if (be16_to_cpu(free->bests[findex]) != be16_to_cpu(data->hdr.bestfree[0].length)) {
- free->bests[findex] = data->hdr.bestfree[0].length;
+ if (be16_to_cpu(free->bests[findex]) != be16_to_cpu(hdr->bestfree[0].length)) {
+ free->bests[findex] = hdr->bestfree[0].length;
logfree = 1;
}
/*
@@ -1777,16 +1753,10 @@ xfs_dir2_node_addname_int(
if (logfree)
xfs_dir2_free_log_bests(tp, fbp, findex, findex);
/*
- * If the caller didn't hand us the freespace block, drop it.
- */
- if ((fblk == NULL || fblk->bp == NULL) && fbp != NULL)
- xfs_da_buf_done(fbp);
- /*
* Return the data block and offset in args, then drop the data block.
*/
args->blkno = (xfs_dablk_t)dbno;
args->index = be16_to_cpu(*tagp);
- xfs_da_buf_done(dbp);
return 0;
}
@@ -1824,22 +1794,23 @@ xfs_dir2_node_lookup(
/* If a CI match, dup the actual name and return EEXIST */
xfs_dir2_data_entry_t *dep;
- dep = (xfs_dir2_data_entry_t *)((char *)state->extrablk.bp->
- data + state->extrablk.index);
+ dep = (xfs_dir2_data_entry_t *)
+ ((char *)state->extrablk.bp->b_addr +
+ state->extrablk.index);
rval = xfs_dir_cilookup_result(args, dep->name, dep->namelen);
}
/*
* Release the btree blocks and leaf block.
*/
for (i = 0; i < state->path.active; i++) {
- xfs_da_brelse(args->trans, state->path.blk[i].bp);
+ xfs_trans_brelse(args->trans, state->path.blk[i].bp);
state->path.blk[i].bp = NULL;
}
/*
* Release the data block if we have it.
*/
if (state->extravalid && state->extrablk.bp) {
- xfs_da_brelse(args->trans, state->extrablk.bp);
+ xfs_trans_brelse(args->trans, state->extrablk.bp);
state->extrablk.bp = NULL;
}
xfs_da_state_free(state);
@@ -1918,7 +1889,7 @@ xfs_dir2_node_replace(
xfs_da_args_t *args) /* operation arguments */
{
xfs_da_state_blk_t *blk; /* leaf block */
- xfs_dir2_data_t *data; /* data block structure */
+ xfs_dir2_data_hdr_t *hdr; /* data block header */
xfs_dir2_data_entry_t *dep; /* data entry changed */
int error; /* error return value */
int i; /* btree level */
@@ -1956,16 +1927,16 @@ xfs_dir2_node_replace(
*/
blk = &state->path.blk[state->path.active - 1];
ASSERT(blk->magic == XFS_DIR2_LEAFN_MAGIC);
- leaf = blk->bp->data;
+ leaf = blk->bp->b_addr;
lep = &leaf->ents[blk->index];
ASSERT(state->extravalid);
/*
* Point to the data entry.
*/
- data = state->extrablk.bp->data;
- ASSERT(be32_to_cpu(data->hdr.magic) == XFS_DIR2_DATA_MAGIC);
+ hdr = state->extrablk.bp->b_addr;
+ ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC));
dep = (xfs_dir2_data_entry_t *)
- ((char *)data +
+ ((char *)hdr +
xfs_dir2_dataptr_to_off(state->mp, be32_to_cpu(lep->address)));
ASSERT(inum != be64_to_cpu(dep->inumber));
/*
@@ -1979,14 +1950,14 @@ xfs_dir2_node_replace(
* Didn't find it, and we're holding a data block. Drop it.
*/
else if (state->extravalid) {
- xfs_da_brelse(args->trans, state->extrablk.bp);
+ xfs_trans_brelse(args->trans, state->extrablk.bp);
state->extrablk.bp = NULL;
}
/*
* Release all the buffers in the cursor.
*/
for (i = 0; i < state->path.active; i++) {
- xfs_da_brelse(args->trans, state->path.blk[i].bp);
+ xfs_trans_brelse(args->trans, state->path.blk[i].bp);
state->path.blk[i].bp = NULL;
}
xfs_da_state_free(state);
@@ -2003,7 +1974,7 @@ xfs_dir2_node_trim_free(
xfs_fileoff_t fo, /* free block number */
int *rvalp) /* out: did something */
{
- xfs_dabuf_t *bp; /* freespace buffer */
+ struct xfs_buf *bp; /* freespace buffer */
xfs_inode_t *dp; /* incore directory inode */
int error; /* error return code */
xfs_dir2_free_t *free; /* freespace structure */
@@ -2016,25 +1987,22 @@ xfs_dir2_node_trim_free(
/*
* Read the freespace block.
*/
- if (unlikely(error = xfs_da_read_buf(tp, dp, (xfs_dablk_t)fo, -2, &bp,
- XFS_DATA_FORK))) {
+ error = xfs_dir2_free_try_read(tp, dp, fo, &bp);
+ if (error)
return error;
- }
-
/*
* There can be holes in freespace. If fo is a hole, there's
* nothing to do.
*/
- if (bp == NULL) {
+ if (!bp)
return 0;
- }
- free = bp->data;
- ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
+ free = bp->b_addr;
+ ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
/*
* If there are used entries, there's nothing to do.
*/
if (be32_to_cpu(free->hdr.nused) > 0) {
- xfs_da_brelse(tp, bp);
+ xfs_trans_brelse(tp, bp);
*rvalp = 0;
return 0;
}
@@ -2050,7 +2018,7 @@ xfs_dir2_node_trim_free(
* pieces. This is the last block of an extent.
*/
ASSERT(error != ENOSPC);
- xfs_da_brelse(tp, bp);
+ xfs_trans_brelse(tp, bp);
return error;
}
/*
diff --git a/libxfs/xfs_dir2_priv.h b/libxfs/xfs_dir2_priv.h
new file mode 100644
index 0000000..7da79f6
--- /dev/null
+++ b/libxfs/xfs_dir2_priv.h
@@ -0,0 +1,152 @@
+/*
+ * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+#ifndef __XFS_DIR2_PRIV_H__
+#define __XFS_DIR2_PRIV_H__
+
+/* xfs_dir2.c */
+extern int xfs_dir_ino_validate(struct xfs_mount *mp, xfs_ino_t ino);
+extern int xfs_dir2_isblock(struct xfs_trans *tp, struct xfs_inode *dp, int *r);
+extern int xfs_dir2_isleaf(struct xfs_trans *tp, struct xfs_inode *dp, int *r);
+extern int xfs_dir2_grow_inode(struct xfs_da_args *args, int space,
+ xfs_dir2_db_t *dbp);
+extern int xfs_dir2_shrink_inode(struct xfs_da_args *args, xfs_dir2_db_t db,
+ struct xfs_buf *bp);
+extern int xfs_dir_cilookup_result(struct xfs_da_args *args,
+ const unsigned char *name, int len);
+
+/* xfs_dir2_block.c */
+extern const struct xfs_buf_ops xfs_dir2_block_buf_ops;
+
+extern int xfs_dir2_block_addname(struct xfs_da_args *args);
+extern int xfs_dir2_block_getdents(struct xfs_inode *dp, void *dirent,
+ xfs_off_t *offset, filldir_t filldir);
+extern int xfs_dir2_block_lookup(struct xfs_da_args *args);
+extern int xfs_dir2_block_removename(struct xfs_da_args *args);
+extern int xfs_dir2_block_replace(struct xfs_da_args *args);
+extern int xfs_dir2_leaf_to_block(struct xfs_da_args *args,
+ struct xfs_buf *lbp, struct xfs_buf *dbp);
+
+/* xfs_dir2_data.c */
+#ifdef DEBUG
+#define xfs_dir2_data_check(dp,bp) __xfs_dir2_data_check(dp, bp);
+#else
+#define xfs_dir2_data_check(dp,bp)
+#endif
+
+extern const struct xfs_buf_ops xfs_dir2_data_buf_ops;
+
+extern int __xfs_dir2_data_check(struct xfs_inode *dp, struct xfs_buf *bp);
+extern int xfs_dir2_data_read(struct xfs_trans *tp, struct xfs_inode *dp,
+ xfs_dablk_t bno, xfs_daddr_t mapped_bno, struct xfs_buf **bpp);
+extern int xfs_dir2_data_readahead(struct xfs_trans *tp, struct xfs_inode *dp,
+ xfs_dablk_t bno, xfs_daddr_t mapped_bno);
+
+extern struct xfs_dir2_data_free *
+xfs_dir2_data_freeinsert(struct xfs_dir2_data_hdr *hdr,
+ struct xfs_dir2_data_unused *dup, int *loghead);
+extern void xfs_dir2_data_freescan(struct xfs_mount *mp,
+ struct xfs_dir2_data_hdr *hdr, int *loghead);
+extern int xfs_dir2_data_init(struct xfs_da_args *args, xfs_dir2_db_t blkno,
+ struct xfs_buf **bpp);
+extern void xfs_dir2_data_log_entry(struct xfs_trans *tp, struct xfs_buf *bp,
+ struct xfs_dir2_data_entry *dep);
+extern void xfs_dir2_data_log_header(struct xfs_trans *tp,
+ struct xfs_buf *bp);
+extern void xfs_dir2_data_log_unused(struct xfs_trans *tp, struct xfs_buf *bp,
+ struct xfs_dir2_data_unused *dup);
+extern void xfs_dir2_data_make_free(struct xfs_trans *tp, struct xfs_buf *bp,
+ xfs_dir2_data_aoff_t offset, xfs_dir2_data_aoff_t len,
+ int *needlogp, int *needscanp);
+extern void xfs_dir2_data_use_free(struct xfs_trans *tp, struct xfs_buf *bp,
+ struct xfs_dir2_data_unused *dup, xfs_dir2_data_aoff_t offset,
+ xfs_dir2_data_aoff_t len, int *needlogp, int *needscanp);
+
+/* xfs_dir2_leaf.c */
+extern const struct xfs_buf_ops xfs_dir2_leafn_buf_ops;
+
+extern int xfs_dir2_leafn_read(struct xfs_trans *tp, struct xfs_inode *dp,
+ xfs_dablk_t fbno, xfs_daddr_t mappedbno, struct xfs_buf **bpp);
+extern int xfs_dir2_block_to_leaf(struct xfs_da_args *args,
+ struct xfs_buf *dbp);
+extern int xfs_dir2_leaf_addname(struct xfs_da_args *args);
+extern void xfs_dir2_leaf_compact(struct xfs_da_args *args,
+ struct xfs_buf *bp);
+extern void xfs_dir2_leaf_compact_x1(struct xfs_buf *bp, int *indexp,
+ int *lowstalep, int *highstalep, int *lowlogp, int *highlogp);
+extern int xfs_dir2_leaf_getdents(struct xfs_inode *dp, void *dirent,
+ size_t bufsize, xfs_off_t *offset, filldir_t filldir);
+extern int xfs_dir2_leaf_init(struct xfs_da_args *args, xfs_dir2_db_t bno,
+ struct xfs_buf **bpp, int magic);
+extern void xfs_dir2_leaf_log_ents(struct xfs_trans *tp, struct xfs_buf *bp,
+ int first, int last);
+extern void xfs_dir2_leaf_log_header(struct xfs_trans *tp,
+ struct xfs_buf *bp);
+extern int xfs_dir2_leaf_lookup(struct xfs_da_args *args);
+extern int xfs_dir2_leaf_removename(struct xfs_da_args *args);
+extern int xfs_dir2_leaf_replace(struct xfs_da_args *args);
+extern int xfs_dir2_leaf_search_hash(struct xfs_da_args *args,
+ struct xfs_buf *lbp);
+extern int xfs_dir2_leaf_trim_data(struct xfs_da_args *args,
+ struct xfs_buf *lbp, xfs_dir2_db_t db);
+extern struct xfs_dir2_leaf_entry *
+xfs_dir2_leaf_find_entry(struct xfs_dir2_leaf *leaf, int index, int compact,
+ int lowstale, int highstale,
+ int *lfloglow, int *lfloghigh);
+extern int xfs_dir2_node_to_leaf(struct xfs_da_state *state);
+
+/* xfs_dir2_node.c */
+extern int xfs_dir2_leaf_to_node(struct xfs_da_args *args,
+ struct xfs_buf *lbp);
+extern xfs_dahash_t xfs_dir2_leafn_lasthash(struct xfs_buf *bp, int *count);
+extern int xfs_dir2_leafn_lookup_int(struct xfs_buf *bp,
+ struct xfs_da_args *args, int *indexp,
+ struct xfs_da_state *state);
+extern int xfs_dir2_leafn_order(struct xfs_buf *leaf1_bp,
+ struct xfs_buf *leaf2_bp);
+extern int xfs_dir2_leafn_split(struct xfs_da_state *state,
+ struct xfs_da_state_blk *oldblk, struct xfs_da_state_blk *newblk);
+extern int xfs_dir2_leafn_toosmall(struct xfs_da_state *state, int *action);
+extern void xfs_dir2_leafn_unbalance(struct xfs_da_state *state,
+ struct xfs_da_state_blk *drop_blk,
+ struct xfs_da_state_blk *save_blk);
+extern int xfs_dir2_node_addname(struct xfs_da_args *args);
+extern int xfs_dir2_node_lookup(struct xfs_da_args *args);
+extern int xfs_dir2_node_removename(struct xfs_da_args *args);
+extern int xfs_dir2_node_replace(struct xfs_da_args *args);
+extern int xfs_dir2_node_trim_free(struct xfs_da_args *args, xfs_fileoff_t fo,
+ int *rvalp);
+extern int xfs_dir2_free_read(struct xfs_trans *tp, struct xfs_inode *dp,
+ xfs_dablk_t fbno, struct xfs_buf **bpp);
+
+/* xfs_dir2_sf.c */
+extern xfs_ino_t xfs_dir2_sf_get_parent_ino(struct xfs_dir2_sf_hdr *sfp);
+extern xfs_ino_t xfs_dir2_sfe_get_ino(struct xfs_dir2_sf_hdr *sfp,
+ struct xfs_dir2_sf_entry *sfep);
+extern int xfs_dir2_block_sfsize(struct xfs_inode *dp,
+ struct xfs_dir2_data_hdr *block, struct xfs_dir2_sf_hdr *sfhp);
+extern int xfs_dir2_block_to_sf(struct xfs_da_args *args, struct xfs_buf *bp,
+ int size, xfs_dir2_sf_hdr_t *sfhp);
+extern int xfs_dir2_sf_addname(struct xfs_da_args *args);
+extern int xfs_dir2_sf_create(struct xfs_da_args *args, xfs_ino_t pino);
+extern int xfs_dir2_sf_getdents(struct xfs_inode *dp, void *dirent,
+ xfs_off_t *offset, filldir_t filldir);
+extern int xfs_dir2_sf_lookup(struct xfs_da_args *args);
+extern int xfs_dir2_sf_removename(struct xfs_da_args *args);
+extern int xfs_dir2_sf_replace(struct xfs_da_args *args);
+
+#endif /* __XFS_DIR2_PRIV_H__ */
diff --git a/libxfs/xfs_dir2_sf.c b/libxfs/xfs_dir2_sf.c
index 6b5e6d4..a96be76 100644
--- a/libxfs/xfs_dir2_sf.c
+++ b/libxfs/xfs_dir2_sf.c
@@ -41,6 +41,82 @@ static void xfs_dir2_sf_toino8(xfs_da_args_t *args);
#endif /* XFS_BIG_INUMS */
/*
+ * Inode numbers in short-form directories can come in two versions,
+ * either 4 bytes or 8 bytes wide. These helpers deal with the
+ * two forms transparently by looking at the headers i8count field.
+ *
+ * For 64-bit inode number the most significant byte must be zero.
+ */
+static xfs_ino_t
+xfs_dir2_sf_get_ino(
+ struct xfs_dir2_sf_hdr *hdr,
+ xfs_dir2_inou_t *from)
+{
+ if (hdr->i8count)
+ return get_unaligned_be64(&from->i8.i) & 0x00ffffffffffffffULL;
+ else
+ return get_unaligned_be32(&from->i4.i);
+}
+
+static void
+xfs_dir2_sf_put_ino(
+ struct xfs_dir2_sf_hdr *hdr,
+ xfs_dir2_inou_t *to,
+ xfs_ino_t ino)
+{
+ ASSERT((ino & 0xff00000000000000ULL) == 0);
+
+ if (hdr->i8count)
+ put_unaligned_be64(ino, &to->i8.i);
+ else
+ put_unaligned_be32(ino, &to->i4.i);
+}
+
+xfs_ino_t
+xfs_dir2_sf_get_parent_ino(
+ struct xfs_dir2_sf_hdr *hdr)
+{
+ return xfs_dir2_sf_get_ino(hdr, &hdr->parent);
+}
+
+void
+xfs_dir2_sf_put_parent_ino(
+ struct xfs_dir2_sf_hdr *hdr,
+ xfs_ino_t ino)
+{
+ xfs_dir2_sf_put_ino(hdr, &hdr->parent, ino);
+}
+
+/*
+ * In short-form directory entries the inode numbers are stored at variable
+ * offset behind the entry name. The inode numbers may only be accessed
+ * through the helpers below.
+ */
+static xfs_dir2_inou_t *
+xfs_dir2_sfe_inop(
+ struct xfs_dir2_sf_entry *sfep)
+{
+ return (xfs_dir2_inou_t *)&sfep->name[sfep->namelen];
+}
+
+xfs_ino_t
+xfs_dir2_sfe_get_ino(
+ struct xfs_dir2_sf_hdr *hdr,
+ struct xfs_dir2_sf_entry *sfep)
+{
+ return xfs_dir2_sf_get_ino(hdr, xfs_dir2_sfe_inop(sfep));
+}
+
+void
+xfs_dir2_sfe_put_ino(
+ struct xfs_dir2_sf_hdr *hdr,
+ struct xfs_dir2_sf_entry *sfep,
+ xfs_ino_t ino)
+{
+ xfs_dir2_sf_put_ino(hdr, xfs_dir2_sfe_inop(sfep), ino);
+}
+
+/*
* Given a block directory (dp/block), calculate its size as a shortform (sf)
* directory and a header for the sf directory, if it will fit it the
* space currently present in the inode. If it won't fit, the output
@@ -49,7 +125,7 @@ static void xfs_dir2_sf_toino8(xfs_da_args_t *args);
int /* size for sf form */
xfs_dir2_block_sfsize(
xfs_inode_t *dp, /* incore inode pointer */
- xfs_dir2_block_t *block, /* block directory data */
+ xfs_dir2_data_hdr_t *hdr, /* block directory data */
xfs_dir2_sf_hdr_t *sfhp) /* output: header for sf form */
{
xfs_dir2_dataptr_t addr; /* data entry address */
@@ -69,7 +145,7 @@ xfs_dir2_block_sfsize(
mp = dp->i_mount;
count = i8count = namelen = 0;
- btp = xfs_dir2_block_tail_p(mp, block);
+ btp = xfs_dir2_block_tail_p(mp, hdr);
blp = xfs_dir2_block_leaf_p(btp);
/*
@@ -82,7 +158,7 @@ xfs_dir2_block_sfsize(
* Calculate the pointer to the entry at hand.
*/
dep = (xfs_dir2_data_entry_t *)
- ((char *)block + xfs_dir2_dataptr_to_off(mp, addr));
+ ((char *)hdr + xfs_dir2_dataptr_to_off(mp, addr));
/*
* Detect . and .., so we can special-case them.
* . is not included in sf directories.
@@ -119,7 +195,7 @@ xfs_dir2_block_sfsize(
*/
sfhp->count = count;
sfhp->i8count = i8count;
- xfs_dir2_sf_put_inumber((xfs_dir2_sf_t *)sfhp, &parent, &sfhp->parent);
+ xfs_dir2_sf_put_parent_ino(sfhp, parent);
return size;
}
@@ -130,11 +206,11 @@ xfs_dir2_block_sfsize(
int /* error */
xfs_dir2_block_to_sf(
xfs_da_args_t *args, /* operation arguments */
- xfs_dabuf_t *bp, /* block buffer */
+ struct xfs_buf *bp,
int size, /* shortform directory size */
xfs_dir2_sf_hdr_t *sfhp) /* shortform directory hdr */
{
- xfs_dir2_block_t *block; /* block structure */
+ xfs_dir2_data_hdr_t *hdr; /* block header */
xfs_dir2_block_tail_t *btp; /* block tail pointer */
xfs_dir2_data_entry_t *dep; /* data entry pointer */
xfs_inode_t *dp; /* incore directory inode */
@@ -145,8 +221,7 @@ xfs_dir2_block_to_sf(
xfs_mount_t *mp; /* filesystem mount point */
char *ptr; /* current data pointer */
xfs_dir2_sf_entry_t *sfep; /* shortform entry */
- xfs_dir2_sf_t *sfp; /* shortform structure */
- xfs_ino_t temp;
+ xfs_dir2_sf_hdr_t *sfp; /* shortform directory header */
trace_xfs_dir2_block_to_sf(args);
@@ -157,13 +232,14 @@ xfs_dir2_block_to_sf(
* Make a copy of the block data, so we can shrink the inode
* and add local data.
*/
- block = kmem_alloc(mp->m_dirblksize, KM_SLEEP);
- memcpy(block, bp->data, mp->m_dirblksize);
+ hdr = kmem_alloc(mp->m_dirblksize, KM_SLEEP);
+ memcpy(hdr, bp->b_addr, mp->m_dirblksize);
logflags = XFS_ILOG_CORE;
if ((error = xfs_dir2_shrink_inode(args, mp->m_dirdatablk, bp))) {
ASSERT(error != ENOSPC);
goto out;
}
+
/*
* The buffer is now unconditionally gone, whether
* xfs_dir2_shrink_inode worked or not.
@@ -179,14 +255,14 @@ xfs_dir2_block_to_sf(
/*
* Copy the header into the newly allocate local space.
*/
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
+ sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
memcpy(sfp, sfhp, xfs_dir2_sf_hdr_size(sfhp->i8count));
dp->i_d.di_size = size;
/*
* Set up to loop over the block's entries.
*/
- btp = xfs_dir2_block_tail_p(mp, block);
- ptr = (char *)block->u;
+ btp = xfs_dir2_block_tail_p(mp, hdr);
+ ptr = (char *)(hdr + 1);
endptr = (char *)xfs_dir2_block_leaf_p(btp);
sfep = xfs_dir2_sf_firstentry(sfp);
/*
@@ -214,7 +290,7 @@ xfs_dir2_block_to_sf(
else if (dep->namelen == 2 &&
dep->name[0] == '.' && dep->name[1] == '.')
ASSERT(be64_to_cpu(dep->inumber) ==
- xfs_dir2_sf_get_inumber(sfp, &sfp->hdr.parent));
+ xfs_dir2_sf_get_parent_ino(sfp));
/*
* Normal entry, copy it into shortform.
*/
@@ -222,11 +298,11 @@ xfs_dir2_block_to_sf(
sfep->namelen = dep->namelen;
xfs_dir2_sf_put_offset(sfep,
(xfs_dir2_data_aoff_t)
- ((char *)dep - (char *)block));
+ ((char *)dep - (char *)hdr));
memcpy(sfep->name, dep->name, dep->namelen);
- temp = be64_to_cpu(dep->inumber);
- xfs_dir2_sf_put_inumber(sfp, &temp,
- xfs_dir2_sf_inumberp(sfep));
+ xfs_dir2_sfe_put_ino(sfp, sfep,
+ be64_to_cpu(dep->inumber));
+
sfep = xfs_dir2_sf_nextentry(sfp, sfep);
}
ptr += xfs_dir2_data_entsize(dep->namelen);
@@ -235,7 +311,7 @@ xfs_dir2_block_to_sf(
xfs_dir2_sf_check(args);
out:
xfs_trans_log_inode(args->trans, dp, logflags);
- kmem_free(block);
+ kmem_free(hdr);
return error;
}
@@ -258,7 +334,7 @@ xfs_dir2_sf_addname(
xfs_dir2_data_aoff_t offset = 0; /* offset for new entry */
int old_isize; /* di_size before adding name */
int pick; /* which algorithm to use */
- xfs_dir2_sf_t *sfp; /* shortform structure */
+ xfs_dir2_sf_hdr_t *sfp; /* shortform structure */
xfs_dir2_sf_entry_t *sfep = NULL; /* shortform entry */
trace_xfs_dir2_sf_addname(args);
@@ -275,19 +351,19 @@ xfs_dir2_sf_addname(
}
ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
ASSERT(dp->i_df.if_u1.if_data != NULL);
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
- ASSERT(dp->i_d.di_size >= xfs_dir2_sf_hdr_size(sfp->hdr.i8count));
+ sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
+ ASSERT(dp->i_d.di_size >= xfs_dir2_sf_hdr_size(sfp->i8count));
/*
* Compute entry (and change in) size.
*/
- add_entsize = xfs_dir2_sf_entsize_byname(sfp, args->namelen);
+ add_entsize = xfs_dir2_sf_entsize(sfp, args->namelen);
incr_isize = add_entsize;
objchange = 0;
#if XFS_BIG_INUMS
/*
* Do we have to change to 8 byte inodes?
*/
- if (args->inumber > XFS_DIR2_MAX_SHORT_INUM && sfp->hdr.i8count == 0) {
+ if (args->inumber > XFS_DIR2_MAX_SHORT_INUM && sfp->i8count == 0) {
/*
* Yes, adjust the entry size and the total size.
*/
@@ -295,7 +371,7 @@ xfs_dir2_sf_addname(
(uint)sizeof(xfs_dir2_ino8_t) -
(uint)sizeof(xfs_dir2_ino4_t);
incr_isize +=
- (sfp->hdr.count + 2) *
+ (sfp->count + 2) *
((uint)sizeof(xfs_dir2_ino8_t) -
(uint)sizeof(xfs_dir2_ino4_t));
objchange = 1;
@@ -365,21 +441,21 @@ xfs_dir2_sf_addname_easy(
{
int byteoff; /* byte offset in sf dir */
xfs_inode_t *dp; /* incore directory inode */
- xfs_dir2_sf_t *sfp; /* shortform structure */
+ xfs_dir2_sf_hdr_t *sfp; /* shortform structure */
dp = args->dp;
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
+ sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
byteoff = (int)((char *)sfep - (char *)sfp);
/*
* Grow the in-inode space.
*/
- xfs_idata_realloc(dp, xfs_dir2_sf_entsize_byname(sfp, args->namelen),
+ xfs_idata_realloc(dp, xfs_dir2_sf_entsize(sfp, args->namelen),
XFS_DATA_FORK);
/*
* Need to set up again due to realloc of the inode data.
*/
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
+ sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
sfep = (xfs_dir2_sf_entry_t *)((char *)sfp + byteoff);
/*
* Fill in the new entry.
@@ -387,15 +463,14 @@ xfs_dir2_sf_addname_easy(
sfep->namelen = args->namelen;
xfs_dir2_sf_put_offset(sfep, offset);
memcpy(sfep->name, args->name, sfep->namelen);
- xfs_dir2_sf_put_inumber(sfp, &args->inumber,
- xfs_dir2_sf_inumberp(sfep));
+ xfs_dir2_sfe_put_ino(sfp, sfep, args->inumber);
/*
* Update the header and inode.
*/
- sfp->hdr.count++;
+ sfp->count++;
#if XFS_BIG_INUMS
if (args->inumber > XFS_DIR2_MAX_SHORT_INUM)
- sfp->hdr.i8count++;
+ sfp->i8count++;
#endif
dp->i_d.di_size = new_isize;
xfs_dir2_sf_check(args);
@@ -425,19 +500,19 @@ xfs_dir2_sf_addname_hard(
xfs_dir2_data_aoff_t offset; /* current offset value */
int old_isize; /* previous di_size */
xfs_dir2_sf_entry_t *oldsfep; /* entry in original dir */
- xfs_dir2_sf_t *oldsfp; /* original shortform dir */
+ xfs_dir2_sf_hdr_t *oldsfp; /* original shortform dir */
xfs_dir2_sf_entry_t *sfep; /* entry in new dir */
- xfs_dir2_sf_t *sfp; /* new shortform dir */
+ xfs_dir2_sf_hdr_t *sfp; /* new shortform dir */
/*
* Copy the old directory to the stack buffer.
*/
dp = args->dp;
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
+ sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
old_isize = (int)dp->i_d.di_size;
buf = kmem_alloc(old_isize, KM_SLEEP);
- oldsfp = (xfs_dir2_sf_t *)buf;
+ oldsfp = (xfs_dir2_sf_hdr_t *)buf;
memcpy(oldsfp, sfp, old_isize);
/*
* Loop over the old directory finding the place we're going
@@ -466,7 +541,7 @@ xfs_dir2_sf_addname_hard(
/*
* Reset the pointer since the buffer was reallocated.
*/
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
+ sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
/*
* Copy the first part of the directory, including the header.
*/
@@ -479,12 +554,11 @@ xfs_dir2_sf_addname_hard(
sfep->namelen = args->namelen;
xfs_dir2_sf_put_offset(sfep, offset);
memcpy(sfep->name, args->name, sfep->namelen);
- xfs_dir2_sf_put_inumber(sfp, &args->inumber,
- xfs_dir2_sf_inumberp(sfep));
- sfp->hdr.count++;
+ xfs_dir2_sfe_put_ino(sfp, sfep, args->inumber);
+ sfp->count++;
#if XFS_BIG_INUMS
if (args->inumber > XFS_DIR2_MAX_SHORT_INUM && !objchange)
- sfp->hdr.i8count++;
+ sfp->i8count++;
#endif
/*
* If there's more left to copy, do that.
@@ -518,14 +592,14 @@ xfs_dir2_sf_addname_pick(
xfs_mount_t *mp; /* filesystem mount point */
xfs_dir2_data_aoff_t offset; /* data block offset */
xfs_dir2_sf_entry_t *sfep; /* shortform entry */
- xfs_dir2_sf_t *sfp; /* shortform structure */
+ xfs_dir2_sf_hdr_t *sfp; /* shortform structure */
int size; /* entry's data size */
int used; /* data bytes used */
dp = args->dp;
mp = dp->i_mount;
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
+ sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
size = xfs_dir2_data_entsize(args->namelen);
offset = XFS_DIR2_DATA_FIRST_OFFSET;
sfep = xfs_dir2_sf_firstentry(sfp);
@@ -535,7 +609,7 @@ xfs_dir2_sf_addname_pick(
* Keep track of data offset and whether we've seen a place
* to insert the new entry.
*/
- for (i = 0; i < sfp->hdr.count; i++) {
+ for (i = 0; i < sfp->count; i++) {
if (!holefit)
holefit = offset + size <= xfs_dir2_sf_get_offset(sfep);
offset = xfs_dir2_sf_get_offset(sfep) +
@@ -547,7 +621,7 @@ xfs_dir2_sf_addname_pick(
* was a data block (block form directory).
*/
used = offset +
- (sfp->hdr.count + 3) * (uint)sizeof(xfs_dir2_leaf_entry_t) +
+ (sfp->count + 3) * (uint)sizeof(xfs_dir2_leaf_entry_t) +
(uint)sizeof(xfs_dir2_block_tail_t);
/*
* If it won't fit in a block form then we can't insert it,
@@ -593,30 +667,30 @@ xfs_dir2_sf_check(
xfs_ino_t ino; /* entry inode number */
int offset; /* data offset */
xfs_dir2_sf_entry_t *sfep; /* shortform dir entry */
- xfs_dir2_sf_t *sfp; /* shortform structure */
+ xfs_dir2_sf_hdr_t *sfp; /* shortform structure */
dp = args->dp;
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
+ sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
offset = XFS_DIR2_DATA_FIRST_OFFSET;
- ino = xfs_dir2_sf_get_inumber(sfp, &sfp->hdr.parent);
+ ino = xfs_dir2_sf_get_parent_ino(sfp);
i8count = ino > XFS_DIR2_MAX_SHORT_INUM;
for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp);
- i < sfp->hdr.count;
+ i < sfp->count;
i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) {
ASSERT(xfs_dir2_sf_get_offset(sfep) >= offset);
- ino = xfs_dir2_sf_get_inumber(sfp, xfs_dir2_sf_inumberp(sfep));
+ ino = xfs_dir2_sfe_get_ino(sfp, sfep);
i8count += ino > XFS_DIR2_MAX_SHORT_INUM;
offset =
xfs_dir2_sf_get_offset(sfep) +
xfs_dir2_data_entsize(sfep->namelen);
}
- ASSERT(i8count == sfp->hdr.i8count);
+ ASSERT(i8count == sfp->i8count);
ASSERT(XFS_BIG_INUMS || i8count == 0);
ASSERT((char *)sfep - (char *)sfp == dp->i_d.di_size);
ASSERT(offset +
- (sfp->hdr.count + 2) * (uint)sizeof(xfs_dir2_leaf_entry_t) +
+ (sfp->count + 2) * (uint)sizeof(xfs_dir2_leaf_entry_t) +
(uint)sizeof(xfs_dir2_block_tail_t) <=
dp->i_mount->m_dirblksize);
}
@@ -632,7 +706,7 @@ xfs_dir2_sf_create(
{
xfs_inode_t *dp; /* incore directory inode */
int i8count; /* parent inode is an 8-byte number */
- xfs_dir2_sf_t *sfp; /* shortform structure */
+ xfs_dir2_sf_hdr_t *sfp; /* shortform structure */
int size; /* directory size */
trace_xfs_dir2_sf_create(args);
@@ -662,13 +736,13 @@ xfs_dir2_sf_create(
/*
* Fill in the header,
*/
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
- sfp->hdr.i8count = i8count;
+ sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
+ sfp->i8count = i8count;
/*
* Now can put in the inode number, since i8count is set.
*/
- xfs_dir2_sf_put_inumber(sfp, &pino, &sfp->hdr.parent);
- sfp->hdr.count = 0;
+ xfs_dir2_sf_put_parent_ino(sfp, pino);
+ sfp->count = 0;
dp->i_d.di_size = size;
xfs_dir2_sf_check(args);
xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_DDATA);
@@ -687,7 +761,7 @@ xfs_dir2_sf_lookup(
int i; /* entry index */
int error;
xfs_dir2_sf_entry_t *sfep; /* shortform directory entry */
- xfs_dir2_sf_t *sfp; /* shortform structure */
+ xfs_dir2_sf_hdr_t *sfp; /* shortform structure */
enum xfs_dacmp cmp; /* comparison result */
xfs_dir2_sf_entry_t *ci_sfep; /* case-insens. entry */
@@ -706,8 +780,8 @@ xfs_dir2_sf_lookup(
}
ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
ASSERT(dp->i_df.if_u1.if_data != NULL);
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
- ASSERT(dp->i_d.di_size >= xfs_dir2_sf_hdr_size(sfp->hdr.i8count));
+ sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
+ ASSERT(dp->i_d.di_size >= xfs_dir2_sf_hdr_size(sfp->i8count));
/*
* Special case for .
*/
@@ -721,7 +795,7 @@ xfs_dir2_sf_lookup(
*/
if (args->namelen == 2 &&
args->name[0] == '.' && args->name[1] == '.') {
- args->inumber = xfs_dir2_sf_get_inumber(sfp, &sfp->hdr.parent);
+ args->inumber = xfs_dir2_sf_get_parent_ino(sfp);
args->cmpresult = XFS_CMP_EXACT;
return XFS_ERROR(EEXIST);
}
@@ -729,7 +803,7 @@ xfs_dir2_sf_lookup(
* Loop over all the entries trying to match ours.
*/
ci_sfep = NULL;
- for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->hdr.count;
+ for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->count;
i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) {
/*
* Compare name and if it's an exact match, return the inode
@@ -740,8 +814,7 @@ xfs_dir2_sf_lookup(
sfep->namelen);
if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) {
args->cmpresult = cmp;
- args->inumber = xfs_dir2_sf_get_inumber(sfp,
- xfs_dir2_sf_inumberp(sfep));
+ args->inumber = xfs_dir2_sfe_get_ino(sfp, sfep);
if (cmp == XFS_CMP_EXACT)
return XFS_ERROR(EEXIST);
ci_sfep = sfep;
@@ -773,7 +846,7 @@ xfs_dir2_sf_removename(
int newsize; /* new inode size */
int oldsize; /* old inode size */
xfs_dir2_sf_entry_t *sfep; /* shortform directory entry */
- xfs_dir2_sf_t *sfp; /* shortform structure */
+ xfs_dir2_sf_hdr_t *sfp; /* shortform structure */
trace_xfs_dir2_sf_removename(args);
@@ -790,32 +863,31 @@ xfs_dir2_sf_removename(
}
ASSERT(dp->i_df.if_bytes == oldsize);
ASSERT(dp->i_df.if_u1.if_data != NULL);
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
- ASSERT(oldsize >= xfs_dir2_sf_hdr_size(sfp->hdr.i8count));
+ sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
+ ASSERT(oldsize >= xfs_dir2_sf_hdr_size(sfp->i8count));
/*
* Loop over the old directory entries.
* Find the one we're deleting.
*/
- for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->hdr.count;
+ for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->count;
i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) {
if (xfs_da_compname(args, sfep->name, sfep->namelen) ==
XFS_CMP_EXACT) {
- ASSERT(xfs_dir2_sf_get_inumber(sfp,
- xfs_dir2_sf_inumberp(sfep)) ==
- args->inumber);
+ ASSERT(xfs_dir2_sfe_get_ino(sfp, sfep) ==
+ args->inumber);
break;
}
}
/*
* Didn't find it.
*/
- if (i == sfp->hdr.count)
+ if (i == sfp->count)
return XFS_ERROR(ENOENT);
/*
* Calculate sizes.
*/
byteoff = (int)((char *)sfep - (char *)sfp);
- entsize = xfs_dir2_sf_entsize_byname(sfp, args->namelen);
+ entsize = xfs_dir2_sf_entsize(sfp, args->namelen);
newsize = oldsize - entsize;
/*
* Copy the part if any after the removed entry, sliding it down.
@@ -826,22 +898,22 @@ xfs_dir2_sf_removename(
/*
* Fix up the header and file size.
*/
- sfp->hdr.count--;
+ sfp->count--;
dp->i_d.di_size = newsize;
/*
* Reallocate, making it smaller.
*/
xfs_idata_realloc(dp, newsize - oldsize, XFS_DATA_FORK);
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
+ sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
#if XFS_BIG_INUMS
/*
* Are we changing inode number size?
*/
if (args->inumber > XFS_DIR2_MAX_SHORT_INUM) {
- if (sfp->hdr.i8count == 1)
+ if (sfp->i8count == 1)
xfs_dir2_sf_toino4(args);
else
- sfp->hdr.i8count--;
+ sfp->i8count--;
}
#endif
xfs_dir2_sf_check(args);
@@ -865,7 +937,7 @@ xfs_dir2_sf_replace(
int i8elevated; /* sf_toino8 set i8count=1 */
#endif
xfs_dir2_sf_entry_t *sfep; /* shortform directory entry */
- xfs_dir2_sf_t *sfp; /* shortform structure */
+ xfs_dir2_sf_hdr_t *sfp; /* shortform structure */
trace_xfs_dir2_sf_replace(args);
@@ -881,19 +953,19 @@ xfs_dir2_sf_replace(
}
ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
ASSERT(dp->i_df.if_u1.if_data != NULL);
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
- ASSERT(dp->i_d.di_size >= xfs_dir2_sf_hdr_size(sfp->hdr.i8count));
+ sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
+ ASSERT(dp->i_d.di_size >= xfs_dir2_sf_hdr_size(sfp->i8count));
#if XFS_BIG_INUMS
/*
* New inode number is large, and need to convert to 8-byte inodes.
*/
- if (args->inumber > XFS_DIR2_MAX_SHORT_INUM && sfp->hdr.i8count == 0) {
+ if (args->inumber > XFS_DIR2_MAX_SHORT_INUM && sfp->i8count == 0) {
int error; /* error return value */
int newsize; /* new inode size */
newsize =
dp->i_df.if_bytes +
- (sfp->hdr.count + 1) *
+ (sfp->count + 1) *
((uint)sizeof(xfs_dir2_ino8_t) -
(uint)sizeof(xfs_dir2_ino4_t));
/*
@@ -911,7 +983,7 @@ xfs_dir2_sf_replace(
*/
xfs_dir2_sf_toino8(args);
i8elevated = 1;
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
+ sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
} else
i8elevated = 0;
#endif
@@ -922,34 +994,32 @@ xfs_dir2_sf_replace(
if (args->namelen == 2 &&
args->name[0] == '.' && args->name[1] == '.') {
#if XFS_BIG_INUMS || defined(DEBUG)
- ino = xfs_dir2_sf_get_inumber(sfp, &sfp->hdr.parent);
+ ino = xfs_dir2_sf_get_parent_ino(sfp);
ASSERT(args->inumber != ino);
#endif
- xfs_dir2_sf_put_inumber(sfp, &args->inumber, &sfp->hdr.parent);
+ xfs_dir2_sf_put_parent_ino(sfp, args->inumber);
}
/*
* Normal entry, look for the name.
*/
else {
for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp);
- i < sfp->hdr.count;
+ i < sfp->count;
i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) {
if (xfs_da_compname(args, sfep->name, sfep->namelen) ==
XFS_CMP_EXACT) {
#if XFS_BIG_INUMS || defined(DEBUG)
- ino = xfs_dir2_sf_get_inumber(sfp,
- xfs_dir2_sf_inumberp(sfep));
+ ino = xfs_dir2_sfe_get_ino(sfp, sfep);
ASSERT(args->inumber != ino);
#endif
- xfs_dir2_sf_put_inumber(sfp, &args->inumber,
- xfs_dir2_sf_inumberp(sfep));
+ xfs_dir2_sfe_put_ino(sfp, sfep, args->inumber);
break;
}
}
/*
* Didn't find it.
*/
- if (i == sfp->hdr.count) {
+ if (i == sfp->count) {
ASSERT(args->op_flags & XFS_DA_OP_OKNOENT);
#if XFS_BIG_INUMS
if (i8elevated)
@@ -967,10 +1037,10 @@ xfs_dir2_sf_replace(
/*
* And the old count was one, so need to convert to small.
*/
- if (sfp->hdr.i8count == 1)
+ if (sfp->i8count == 1)
xfs_dir2_sf_toino4(args);
else
- sfp->hdr.i8count--;
+ sfp->i8count--;
}
/*
* See if the old number was small, the new number is large.
@@ -981,9 +1051,9 @@ xfs_dir2_sf_replace(
* add to the i8count unless we just converted to 8-byte
* inodes (which does an implied i8count = 1)
*/
- ASSERT(sfp->hdr.i8count != 0);
+ ASSERT(sfp->i8count != 0);
if (!i8elevated)
- sfp->hdr.i8count++;
+ sfp->i8count++;
}
#endif
xfs_dir2_sf_check(args);
@@ -1003,13 +1073,12 @@ xfs_dir2_sf_toino4(
char *buf; /* old dir's buffer */
xfs_inode_t *dp; /* incore directory inode */
int i; /* entry index */
- xfs_ino_t ino; /* entry inode number */
int newsize; /* new inode size */
xfs_dir2_sf_entry_t *oldsfep; /* old sf entry */
- xfs_dir2_sf_t *oldsfp; /* old sf directory */
+ xfs_dir2_sf_hdr_t *oldsfp; /* old sf directory */
int oldsize; /* old inode size */
xfs_dir2_sf_entry_t *sfep; /* new sf entry */
- xfs_dir2_sf_t *sfp; /* new sf directory */
+ xfs_dir2_sf_hdr_t *sfp; /* new sf directory */
trace_xfs_dir2_sf_toino4(args);
@@ -1022,44 +1091,42 @@ xfs_dir2_sf_toino4(
*/
oldsize = dp->i_df.if_bytes;
buf = kmem_alloc(oldsize, KM_SLEEP);
- oldsfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
- ASSERT(oldsfp->hdr.i8count == 1);
+ oldsfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
+ ASSERT(oldsfp->i8count == 1);
memcpy(buf, oldsfp, oldsize);
/*
* Compute the new inode size.
*/
newsize =
oldsize -
- (oldsfp->hdr.count + 1) *
+ (oldsfp->count + 1) *
((uint)sizeof(xfs_dir2_ino8_t) - (uint)sizeof(xfs_dir2_ino4_t));
xfs_idata_realloc(dp, -oldsize, XFS_DATA_FORK);
xfs_idata_realloc(dp, newsize, XFS_DATA_FORK);
/*
* Reset our pointers, the data has moved.
*/
- oldsfp = (xfs_dir2_sf_t *)buf;
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
+ oldsfp = (xfs_dir2_sf_hdr_t *)buf;
+ sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
/*
* Fill in the new header.
*/
- sfp->hdr.count = oldsfp->hdr.count;
- sfp->hdr.i8count = 0;
- ino = xfs_dir2_sf_get_inumber(oldsfp, &oldsfp->hdr.parent);
- xfs_dir2_sf_put_inumber(sfp, &ino, &sfp->hdr.parent);
+ sfp->count = oldsfp->count;
+ sfp->i8count = 0;
+ xfs_dir2_sf_put_parent_ino(sfp, xfs_dir2_sf_get_parent_ino(oldsfp));
/*
* Copy the entries field by field.
*/
for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp),
oldsfep = xfs_dir2_sf_firstentry(oldsfp);
- i < sfp->hdr.count;
+ i < sfp->count;
i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep),
oldsfep = xfs_dir2_sf_nextentry(oldsfp, oldsfep)) {
sfep->namelen = oldsfep->namelen;
sfep->offset = oldsfep->offset;
memcpy(sfep->name, oldsfep->name, sfep->namelen);
- ino = xfs_dir2_sf_get_inumber(oldsfp,
- xfs_dir2_sf_inumberp(oldsfep));
- xfs_dir2_sf_put_inumber(sfp, &ino, xfs_dir2_sf_inumberp(sfep));
+ xfs_dir2_sfe_put_ino(sfp, sfep,
+ xfs_dir2_sfe_get_ino(oldsfp, oldsfep));
}
/*
* Clean up the inode.
@@ -1081,13 +1148,12 @@ xfs_dir2_sf_toino8(
char *buf; /* old dir's buffer */
xfs_inode_t *dp; /* incore directory inode */
int i; /* entry index */
- xfs_ino_t ino; /* entry inode number */
int newsize; /* new inode size */
xfs_dir2_sf_entry_t *oldsfep; /* old sf entry */
- xfs_dir2_sf_t *oldsfp; /* old sf directory */
+ xfs_dir2_sf_hdr_t *oldsfp; /* old sf directory */
int oldsize; /* old inode size */
xfs_dir2_sf_entry_t *sfep; /* new sf entry */
- xfs_dir2_sf_t *sfp; /* new sf directory */
+ xfs_dir2_sf_hdr_t *sfp; /* new sf directory */
trace_xfs_dir2_sf_toino8(args);
@@ -1100,44 +1166,42 @@ xfs_dir2_sf_toino8(
*/
oldsize = dp->i_df.if_bytes;
buf = kmem_alloc(oldsize, KM_SLEEP);
- oldsfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
- ASSERT(oldsfp->hdr.i8count == 0);
+ oldsfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
+ ASSERT(oldsfp->i8count == 0);
memcpy(buf, oldsfp, oldsize);
/*
* Compute the new inode size.
*/
newsize =
oldsize +
- (oldsfp->hdr.count + 1) *
+ (oldsfp->count + 1) *
((uint)sizeof(xfs_dir2_ino8_t) - (uint)sizeof(xfs_dir2_ino4_t));
xfs_idata_realloc(dp, -oldsize, XFS_DATA_FORK);
xfs_idata_realloc(dp, newsize, XFS_DATA_FORK);
/*
* Reset our pointers, the data has moved.
*/
- oldsfp = (xfs_dir2_sf_t *)buf;
- sfp = (xfs_dir2_sf_t *)dp->i_df.if_u1.if_data;
+ oldsfp = (xfs_dir2_sf_hdr_t *)buf;
+ sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
/*
* Fill in the new header.
*/
- sfp->hdr.count = oldsfp->hdr.count;
- sfp->hdr.i8count = 1;
- ino = xfs_dir2_sf_get_inumber(oldsfp, &oldsfp->hdr.parent);
- xfs_dir2_sf_put_inumber(sfp, &ino, &sfp->hdr.parent);
+ sfp->count = oldsfp->count;
+ sfp->i8count = 1;
+ xfs_dir2_sf_put_parent_ino(sfp, xfs_dir2_sf_get_parent_ino(oldsfp));
/*
* Copy the entries field by field.
*/
for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp),
oldsfep = xfs_dir2_sf_firstentry(oldsfp);
- i < sfp->hdr.count;
+ i < sfp->count;
i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep),
oldsfep = xfs_dir2_sf_nextentry(oldsfp, oldsfep)) {
sfep->namelen = oldsfep->namelen;
sfep->offset = oldsfep->offset;
memcpy(sfep->name, oldsfep->name, sfep->namelen);
- ino = xfs_dir2_sf_get_inumber(oldsfp,
- xfs_dir2_sf_inumberp(oldsfep));
- xfs_dir2_sf_put_inumber(sfp, &ino, xfs_dir2_sf_inumberp(sfep));
+ xfs_dir2_sfe_put_ino(sfp, sfep,
+ xfs_dir2_sfe_get_ino(oldsfp, oldsfep));
}
/*
* Clean up the inode.
diff --git a/libxfs/xfs_ialloc.c b/libxfs/xfs_ialloc.c
index 1fcafb6..529d92d 100644
--- a/libxfs/xfs_ialloc.c
+++ b/libxfs/xfs_ialloc.c
@@ -131,7 +131,7 @@ xfs_check_agi_freecount(
/*
* Initialise a new set of inodes.
*/
-STATIC void
+STATIC int
xfs_ialloc_inode_init(
struct xfs_mount *mp,
struct xfs_trans *tp,
@@ -182,10 +182,9 @@ xfs_ialloc_inode_init(
d = XFS_AGB_TO_DADDR(mp, agno, agbno + (j * blks_per_cluster));
fbuf = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
mp->m_bsize * blks_per_cluster,
- XBF_LOCK);
- ASSERT(fbuf);
- ASSERT(!XFS_BUF_GETERROR(fbuf));
-
+ XBF_UNMAPPED);
+ if (!fbuf)
+ return ENOMEM;
/*
* Initialize all inodes in this buffer and then log them.
*
@@ -193,6 +192,7 @@ xfs_ialloc_inode_init(
* to log a whole cluster of inodes instead of all the
* individual transactions causing a lot of log traffic.
*/
+ fbuf->b_ops = &xfs_inode_buf_ops;
xfs_buf_zero(fbuf, 0, ninodes << mp->m_sb.sb_inodelog);
for (i = 0; i < ninodes; i++) {
int ioffset = i << mp->m_sb.sb_inodelog;
@@ -207,6 +207,7 @@ xfs_ialloc_inode_init(
}
xfs_trans_inode_alloc_buf(tp, fbuf);
}
+ return 0;
}
/*
@@ -232,6 +233,7 @@ xfs_ialloc_ag_alloc(
/* boundary */
struct xfs_perag *pag;
+ memset(&args, 0, sizeof(args));
args.tp = tp;
args.mp = tp->t_mountp;
@@ -258,8 +260,6 @@ xfs_ialloc_ag_alloc(
(args.agbno < be32_to_cpu(agi->agi_length)))) {
args.fsbno = XFS_AGB_TO_FSB(args.mp, agno, args.agbno);
args.type = XFS_ALLOCTYPE_THIS_BNO;
- args.mod = args.total = args.wasdel = args.isfl =
- args.userdata = args.minalignslop = 0;
args.prod = 1;
/*
@@ -312,8 +312,6 @@ xfs_ialloc_ag_alloc(
* Allocate a fixed-size extent of inodes.
*/
args.type = XFS_ALLOCTYPE_NEAR_BNO;
- args.mod = args.total = args.wasdel = args.isfl =
- args.userdata = args.minalignslop = 0;
args.prod = 1;
/*
* Allow space for the inode btree to split.
@@ -351,9 +349,11 @@ xfs_ialloc_ag_alloc(
* rather than a linear progression to prevent the next generation
* number from being easily guessable.
*/
- xfs_ialloc_inode_init(args.mp, tp, agno, args.agbno, args.len,
- random32());
+ error = xfs_ialloc_inode_init(args.mp, tp, agno, args.agbno,
+ args.len, random32());
+ if (error)
+ return error;
/*
* Convert the results.
*/
@@ -411,7 +411,7 @@ xfs_ialloc_next_ag(
spin_lock(&mp->m_agirotor_lock);
agno = mp->m_agirotor;
- if (++mp->m_agirotor == mp->m_maxagi)
+ if (++mp->m_agirotor >= mp->m_maxagi)
mp->m_agirotor = 0;
spin_unlock(&mp->m_agirotor_lock);
@@ -422,14 +422,13 @@ xfs_ialloc_next_ag(
* Select an allocation group to look for a free inode in, based on the parent
* inode and then mode. Return the allocation group buffer.
*/
-STATIC xfs_buf_t * /* allocation group buffer */
+STATIC xfs_agnumber_t
xfs_ialloc_ag_select(
xfs_trans_t *tp, /* transaction pointer */
xfs_ino_t parent, /* parent directory inode number */
- mode_t mode, /* bits set to indicate file type */
+ umode_t mode, /* bits set to indicate file type */
int okalloc) /* ok to allocate more space */
{
- xfs_buf_t *agbp; /* allocation group header buffer */
xfs_agnumber_t agcount; /* number of ag's in the filesystem */
xfs_agnumber_t agno; /* current ag number */
int flags; /* alloc buffer locking flags */
@@ -439,6 +438,7 @@ xfs_ialloc_ag_select(
int needspace; /* file mode implies space allocated */
xfs_perag_t *pag; /* per allocation group data */
xfs_agnumber_t pagno; /* parent (starting) ag number */
+ int error;
/*
* Files of these types need at least one block if length > 0
@@ -454,7 +454,9 @@ xfs_ialloc_ag_select(
if (pagno >= agcount)
pagno = 0;
}
+
ASSERT(pagno < agcount);
+
/*
* Loop through allocation groups, looking for one with a little
* free space in it. Note we don't look for free inodes, exactly.
@@ -466,51 +468,45 @@ xfs_ialloc_ag_select(
flags = XFS_ALLOC_FLAG_TRYLOCK;
for (;;) {
pag = xfs_perag_get(mp, agno);
+ if (!pag->pagi_inodeok) {
+ xfs_ialloc_next_ag(mp);
+ goto nextag;
+ }
+
if (!pag->pagi_init) {
- if (xfs_ialloc_read_agi(mp, tp, agno, &agbp)) {
- agbp = NULL;
+ error = xfs_ialloc_pagi_init(mp, tp, agno);
+ if (error)
goto nextag;
- }
- } else
- agbp = NULL;
+ }
- if (!pag->pagi_inodeok) {
- xfs_ialloc_next_ag(mp);
- goto unlock_nextag;
+ if (pag->pagi_freecount) {
+ xfs_perag_put(pag);
+ return agno;
}
- /*
- * Is there enough free space for the file plus a block
- * of inodes (if we need to allocate some)?
- */
- ineed = pag->pagi_freecount ? 0 : XFS_IALLOC_BLOCKS(mp);
- if (ineed && !pag->pagf_init) {
- if (agbp == NULL &&
- xfs_ialloc_read_agi(mp, tp, agno, &agbp)) {
- agbp = NULL;
+ if (!okalloc)
+ goto nextag;
+
+ if (!pag->pagf_init) {
+ error = xfs_alloc_pagf_init(mp, tp, agno, flags);
+ if (error)
goto nextag;
- }
- (void)xfs_alloc_pagf_init(mp, tp, agno, flags);
}
- if (!ineed || pag->pagf_init) {
- if (ineed && !(longest = pag->pagf_longest))
- longest = pag->pagf_flcount > 0;
- if (!ineed ||
- (pag->pagf_freeblks >= needspace + ineed &&
- longest >= ineed &&
- okalloc)) {
- if (agbp == NULL &&
- xfs_ialloc_read_agi(mp, tp, agno, &agbp)) {
- agbp = NULL;
- goto nextag;
- }
- xfs_perag_put(pag);
- return agbp;
- }
+
+ /*
+ * Is there enough free space for the file plus a block of
+ * inodes? (if we need to allocate some)?
+ */
+ ineed = XFS_IALLOC_BLOCKS(mp);
+ longest = pag->pagf_longest;
+ if (!longest)
+ longest = pag->pagf_flcount > 0;
+
+ if (pag->pagf_freeblks >= needspace + ineed &&
+ longest >= ineed) {
+ xfs_perag_put(pag);
+ return agno;
}
-unlock_nextag:
- if (agbp)
- xfs_trans_brelse(tp, agbp);
nextag:
xfs_perag_put(pag);
/*
@@ -518,13 +514,13 @@ nextag:
* down.
*/
if (XFS_FORCED_SHUTDOWN(mp))
- return NULL;
+ return NULLAGNUMBER;
agno++;
if (agno >= agcount)
agno = 0;
if (agno == pagno) {
if (flags == 0)
- return NULL;
+ return NULLAGNUMBER;
flags = 0;
}
}
@@ -587,188 +583,39 @@ xfs_ialloc_get_rec(
}
/*
- * Visible inode allocation functions.
- */
-
-/*
- * Allocate an inode on disk.
- * Mode is used to tell whether the new inode will need space, and whether
- * it is a directory.
+ * Allocate an inode.
*
- * The arguments IO_agbp and alloc_done are defined to work within
- * the constraint of one allocation per transaction.
- * xfs_dialloc() is designed to be called twice if it has to do an
- * allocation to make more free inodes. On the first call,
- * IO_agbp should be set to NULL. If an inode is available,
- * i.e., xfs_dialloc() did not need to do an allocation, an inode
- * number is returned. In this case, IO_agbp would be set to the
- * current ag_buf and alloc_done set to false.
- * If an allocation needed to be done, xfs_dialloc would return
- * the current ag_buf in IO_agbp and set alloc_done to true.
- * The caller should then commit the current transaction, allocate a new
- * transaction, and call xfs_dialloc() again, passing in the previous
- * value of IO_agbp. IO_agbp should be held across the transactions.
- * Since the agbp is locked across the two calls, the second call is
- * guaranteed to have a free inode available.
- *
- * Once we successfully pick an inode its number is returned and the
- * on-disk data structures are updated. The inode itself is not read
- * in, since doing so would break ordering constraints with xfs_reclaim.
+ * The caller selected an AG for us, and made sure that free inodes are
+ * available.
*/
-int
-xfs_dialloc(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_ino_t parent, /* parent inode (directory) */
- mode_t mode, /* mode bits for new inode */
- int okalloc, /* ok to allocate more space */
- xfs_buf_t **IO_agbp, /* in/out ag header's buffer */
- boolean_t *alloc_done, /* true if we needed to replenish
- inode freelist */
- xfs_ino_t *inop) /* inode number allocated */
+STATIC int
+xfs_dialloc_ag(
+ struct xfs_trans *tp,
+ struct xfs_buf *agbp,
+ xfs_ino_t parent,
+ xfs_ino_t *inop)
{
- xfs_agnumber_t agcount; /* number of allocation groups */
- xfs_buf_t *agbp; /* allocation group header's buffer */
- xfs_agnumber_t agno; /* allocation group number */
- xfs_agi_t *agi; /* allocation group header structure */
- xfs_btree_cur_t *cur; /* inode allocation btree cursor */
- int error; /* error return value */
- int i; /* result code */
- int ialloced; /* inode allocation status */
- int noroom = 0; /* no space for inode blk allocation */
- xfs_ino_t ino; /* fs-relative inode to be returned */
- /* REFERENCED */
- int j; /* result code */
- xfs_mount_t *mp; /* file system mount structure */
- int offset; /* index of inode in chunk */
- xfs_agino_t pagino; /* parent's AG relative inode # */
- xfs_agnumber_t pagno; /* parent's AG number */
- xfs_inobt_rec_incore_t rec; /* inode allocation record */
- xfs_agnumber_t tagno; /* testing allocation group number */
- xfs_btree_cur_t *tcur; /* temp cursor */
- xfs_inobt_rec_incore_t trec; /* temp inode allocation record */
- struct xfs_perag *pag;
-
-
- if (*IO_agbp == NULL) {
- /*
- * We do not have an agbp, so select an initial allocation
- * group for inode allocation.
- */
- agbp = xfs_ialloc_ag_select(tp, parent, mode, okalloc);
- /*
- * Couldn't find an allocation group satisfying the
- * criteria, give up.
- */
- if (!agbp) {
- *inop = NULLFSINO;
- return 0;
- }
- agi = XFS_BUF_TO_AGI(agbp);
- ASSERT(be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC);
- } else {
- /*
- * Continue where we left off before. In this case, we
- * know that the allocation group has free inodes.
- */
- agbp = *IO_agbp;
- agi = XFS_BUF_TO_AGI(agbp);
- ASSERT(be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC);
- ASSERT(be32_to_cpu(agi->agi_freecount) > 0);
- }
- mp = tp->t_mountp;
- agcount = mp->m_sb.sb_agcount;
- agno = be32_to_cpu(agi->agi_seqno);
- tagno = agno;
- pagno = XFS_INO_TO_AGNO(mp, parent);
- pagino = XFS_INO_TO_AGINO(mp, parent);
-
- /*
- * If we have already hit the ceiling of inode blocks then clear
- * okalloc so we scan all available agi structures for a free
- * inode.
- */
-
- if (mp->m_maxicount &&
- mp->m_sb.sb_icount + XFS_IALLOC_INODES(mp) > mp->m_maxicount) {
- noroom = 1;
- okalloc = 0;
- }
+ struct xfs_mount *mp = tp->t_mountp;
+ struct xfs_agi *agi = XFS_BUF_TO_AGI(agbp);
+ xfs_agnumber_t agno = be32_to_cpu(agi->agi_seqno);
+ xfs_agnumber_t pagno = XFS_INO_TO_AGNO(mp, parent);
+ xfs_agino_t pagino = XFS_INO_TO_AGINO(mp, parent);
+ struct xfs_perag *pag;
+ struct xfs_btree_cur *cur, *tcur;
+ struct xfs_inobt_rec_incore rec, trec;
+ xfs_ino_t ino;
+ int error;
+ int offset;
+ int i, j;
- /*
- * Loop until we find an allocation group that either has free inodes
- * or in which we can allocate some inodes. Iterate through the
- * allocation groups upward, wrapping at the end.
- */
- *alloc_done = B_FALSE;
- while (!agi->agi_freecount) {
- /*
- * Don't do anything if we're not supposed to allocate
- * any blocks, just go on to the next ag.
- */
- if (okalloc) {
- /*
- * Try to allocate some new inodes in the allocation
- * group.
- */
- if ((error = xfs_ialloc_ag_alloc(tp, agbp, &ialloced))) {
- xfs_trans_brelse(tp, agbp);
- if (error == ENOSPC) {
- *inop = NULLFSINO;
- return 0;
- } else
- return error;
- }
- if (ialloced) {
- /*
- * We successfully allocated some inodes, return
- * the current context to the caller so that it
- * can commit the current transaction and call
- * us again where we left off.
- */
- ASSERT(be32_to_cpu(agi->agi_freecount) > 0);
- *alloc_done = B_TRUE;
- *IO_agbp = agbp;
- *inop = NULLFSINO;
- return 0;
- }
- }
- /*
- * If it failed, give up on this ag.
- */
- xfs_trans_brelse(tp, agbp);
- /*
- * Go on to the next ag: get its ag header.
- */
-nextag:
- if (++tagno == agcount)
- tagno = 0;
- if (tagno == agno) {
- *inop = NULLFSINO;
- return noroom ? ENOSPC : 0;
- }
- pag = xfs_perag_get(mp, tagno);
- if (pag->pagi_inodeok == 0) {
- xfs_perag_put(pag);
- goto nextag;
- }
- error = xfs_ialloc_read_agi(mp, tp, tagno, &agbp);
- xfs_perag_put(pag);
- if (error)
- goto nextag;
- agi = XFS_BUF_TO_AGI(agbp);
- ASSERT(be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC);
- }
- /*
- * Here with an allocation group that has a free inode.
- * Reset agno since we may have chosen a new ag in the
- * loop above.
- */
- agno = tagno;
- *IO_agbp = NULL;
pag = xfs_perag_get(mp, agno);
+ ASSERT(pag->pagi_init);
+ ASSERT(pag->pagi_inodeok);
+ ASSERT(pag->pagi_freecount > 0);
+
restart_pagno:
- cur = xfs_inobt_init_cursor(mp, tp, agbp, be32_to_cpu(agi->agi_seqno));
+ cur = xfs_inobt_init_cursor(mp, tp, agbp, agno);
/*
* If pagino is 0 (this is the root inode allocation) use newino.
* This must work because we've just allocated some.
@@ -925,7 +772,7 @@ nextag:
* See if the most recently allocated block has any free.
*/
newino:
- if (be32_to_cpu(agi->agi_newino) != NULLAGINO) {
+ if (agi->agi_newino != cpu_to_be32(NULLAGINO)) {
error = xfs_inobt_lookup(cur, be32_to_cpu(agi->agi_newino),
XFS_LOOKUP_EQ, &i);
if (error)
@@ -968,7 +815,7 @@ newino:
}
alloc_inode:
- offset = xfs_ialloc_find_free(&rec.ir_free);
+ offset = xfs_lowbit64(rec.ir_free);
ASSERT(offset >= 0);
ASSERT(offset < XFS_INODES_PER_CHUNK);
ASSERT((XFS_AGINO_TO_OFFSET(mp, rec.ir_startino) %
@@ -1000,6 +847,165 @@ error0:
return error;
}
+/*
+ * Allocate an inode on disk.
+ *
+ * Mode is used to tell whether the new inode will need space, and whether it
+ * is a directory.
+ *
+ * This function is designed to be called twice if it has to do an allocation
+ * to make more free inodes. On the first call, *IO_agbp should be set to NULL.
+ * If an inode is available without having to performn an allocation, an inode
+ * number is returned. In this case, *IO_agbp is set to NULL. If an allocation
+ * needs to be done, xfs_dialloc returns the current AGI buffer in *IO_agbp.
+ * The caller should then commit the current transaction, allocate a
+ * new transaction, and call xfs_dialloc() again, passing in the previous value
+ * of *IO_agbp. IO_agbp should be held across the transactions. Since the AGI
+ * buffer is locked across the two calls, the second call is guaranteed to have
+ * a free inode available.
+ *
+ * Once we successfully pick an inode its number is returned and the on-disk
+ * data structures are updated. The inode itself is not read in, since doing so
+ * would break ordering constraints with xfs_reclaim.
+ */
+int
+xfs_dialloc(
+ struct xfs_trans *tp,
+ xfs_ino_t parent,
+ umode_t mode,
+ int okalloc,
+ struct xfs_buf **IO_agbp,
+ xfs_ino_t *inop)
+{
+ struct xfs_mount *mp = tp->t_mountp;
+ struct xfs_buf *agbp;
+ xfs_agnumber_t agno;
+ int error;
+ int ialloced;
+ int noroom = 0;
+ xfs_agnumber_t start_agno;
+ struct xfs_perag *pag;
+
+ if (*IO_agbp) {
+ /*
+ * If the caller passes in a pointer to the AGI buffer,
+ * continue where we left off before. In this case, we
+ * know that the allocation group has free inodes.
+ */
+ agbp = *IO_agbp;
+ goto out_alloc;
+ }
+
+ /*
+ * We do not have an agbp, so select an initial allocation
+ * group for inode allocation.
+ */
+ start_agno = xfs_ialloc_ag_select(tp, parent, mode, okalloc);
+ if (start_agno == NULLAGNUMBER) {
+ *inop = NULLFSINO;
+ return 0;
+ }
+
+ /*
+ * If we have already hit the ceiling of inode blocks then clear
+ * okalloc so we scan all available agi structures for a free
+ * inode.
+ */
+ if (mp->m_maxicount &&
+ mp->m_sb.sb_icount + XFS_IALLOC_INODES(mp) > mp->m_maxicount) {
+ noroom = 1;
+ okalloc = 0;
+ }
+
+ /*
+ * Loop until we find an allocation group that either has free inodes
+ * or in which we can allocate some inodes. Iterate through the
+ * allocation groups upward, wrapping at the end.
+ */
+ agno = start_agno;
+ for (;;) {
+ pag = xfs_perag_get(mp, agno);
+ if (!pag->pagi_inodeok) {
+ xfs_ialloc_next_ag(mp);
+ goto nextag;
+ }
+
+ if (!pag->pagi_init) {
+ error = xfs_ialloc_pagi_init(mp, tp, agno);
+ if (error)
+ goto out_error;
+ }
+
+ /*
+ * Do a first racy fast path check if this AG is usable.
+ */
+ if (!pag->pagi_freecount && !okalloc)
+ goto nextag;
+
+ /*
+ * Then read in the AGI buffer and recheck with the AGI buffer
+ * lock held.
+ */
+ error = xfs_ialloc_read_agi(mp, tp, agno, &agbp);
+ if (error)
+ goto out_error;
+
+ if (pag->pagi_freecount) {
+ xfs_perag_put(pag);
+ goto out_alloc;
+ }
+
+ if (!okalloc)
+ goto nextag_relse_buffer;
+
+
+ error = xfs_ialloc_ag_alloc(tp, agbp, &ialloced);
+ if (error) {
+ xfs_trans_brelse(tp, agbp);
+
+ if (error != ENOSPC)
+ goto out_error;
+
+ xfs_perag_put(pag);
+ *inop = NULLFSINO;
+ return 0;
+ }
+
+ if (ialloced) {
+ /*
+ * We successfully allocated some inodes, return
+ * the current context to the caller so that it
+ * can commit the current transaction and call
+ * us again where we left off.
+ */
+ ASSERT(pag->pagi_freecount > 0);
+ xfs_perag_put(pag);
+
+ *IO_agbp = agbp;
+ *inop = NULLFSINO;
+ return 0;
+ }
+
+nextag_relse_buffer:
+ xfs_trans_brelse(tp, agbp);
+nextag:
+ xfs_perag_put(pag);
+ if (++agno == mp->m_sb.sb_agcount)
+ agno = 0;
+ if (agno == start_agno) {
+ *inop = NULLFSINO;
+ return noroom ? ENOSPC : 0;
+ }
+ }
+
+out_alloc:
+ *IO_agbp = NULL;
+ return xfs_dialloc_ag(tp, agbp, parent, inop);
+out_error:
+ xfs_perag_put(pag);
+ return XFS_ERROR(error);
+}
+
STATIC int
xfs_imap_lookup(
struct xfs_mount *mp,
@@ -1019,10 +1025,9 @@ xfs_imap_lookup(
error = xfs_ialloc_read_agi(mp, tp, agno, &agbp);
if (error) {
- xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
- "xfs_ialloc_read_agi() returned "
- "error %d, agno %d",
- error, agno);
+ xfs_alert(mp,
+ "%s: xfs_ialloc_read_agi() returned error %d, agno %d",
+ __func__, error, agno);
return error;
}
@@ -1100,24 +1105,21 @@ xfs_imap(
if (flags & XFS_IGET_UNTRUSTED)
return XFS_ERROR(EINVAL);
if (agno >= mp->m_sb.sb_agcount) {
- xfs_fs_cmn_err(CE_ALERT, mp,
- "xfs_imap: agno (%d) >= "
- "mp->m_sb.sb_agcount (%d)",
- agno, mp->m_sb.sb_agcount);
+ xfs_alert(mp,
+ "%s: agno (%d) >= mp->m_sb.sb_agcount (%d)",
+ __func__, agno, mp->m_sb.sb_agcount);
}
if (agbno >= mp->m_sb.sb_agblocks) {
- xfs_fs_cmn_err(CE_ALERT, mp,
- "xfs_imap: agbno (0x%llx) >= "
- "mp->m_sb.sb_agblocks (0x%lx)",
- (unsigned long long) agbno,
- (unsigned long) mp->m_sb.sb_agblocks);
+ xfs_alert(mp,
+ "%s: agbno (0x%llx) >= mp->m_sb.sb_agblocks (0x%lx)",
+ __func__, (unsigned long long)agbno,
+ (unsigned long)mp->m_sb.sb_agblocks);
}
if (ino != XFS_AGINO_TO_INO(mp, agno, agino)) {
- xfs_fs_cmn_err(CE_ALERT, mp,
- "xfs_imap: ino (0x%llx) != "
- "XFS_AGINO_TO_INO(mp, agno, agino) "
- "(0x%llx)",
- ino, XFS_AGINO_TO_INO(mp, agno, agino));
+ xfs_alert(mp,
+ "%s: ino (0x%llx) != XFS_AGINO_TO_INO() (0x%llx)",
+ __func__, ino,
+ XFS_AGINO_TO_INO(mp, agno, agino));
}
xfs_stack_trace();
#endif /* DEBUG */
@@ -1189,10 +1191,9 @@ out_map:
*/
if ((imap->im_blkno + imap->im_len) >
XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks)) {
- xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
- "(imap->im_blkno (0x%llx) + imap->im_len (0x%llx)) > "
- " XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks) (0x%llx)",
- (unsigned long long) imap->im_blkno,
+ xfs_alert(mp,
+ "%s: (im_blkno (0x%llx) + im_len (0x%llx)) > sb_dblocks (0x%llx)",
+ __func__, (unsigned long long) imap->im_blkno,
(unsigned long long) imap->im_len,
XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks));
return XFS_ERROR(EINVAL);
@@ -1253,7 +1254,7 @@ xfs_ialloc_log_agi(
xfs_agi_t *agi; /* allocation group header */
agi = XFS_BUF_TO_AGI(bp);
- ASSERT(be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC);
+ ASSERT(agi->agi_magicnum == cpu_to_be32(XFS_AGI_MAGIC));
#endif
/*
* Compute byte offsets for the first and last fields.
@@ -1279,6 +1280,57 @@ xfs_check_agi_unlinked(
#define xfs_check_agi_unlinked(agi)
#endif
+static void
+xfs_agi_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_agi *agi = XFS_BUF_TO_AGI(bp);
+ int agi_ok;
+
+ /*
+ * Validate the magic number of the agi block.
+ */
+ agi_ok = agi->agi_magicnum == cpu_to_be32(XFS_AGI_MAGIC) &&
+ XFS_AGI_GOOD_VERSION(be32_to_cpu(agi->agi_versionnum));
+
+ /*
+ * during growfs operations, the perag is not fully initialised,
+ * so we can't use it for any useful checking. growfs ensures we can't
+ * use it by using uncached buffers that don't have the perag attached
+ * so we can detect and avoid this problem.
+ */
+ if (bp->b_pag)
+ agi_ok = agi_ok && be32_to_cpu(agi->agi_seqno) ==
+ bp->b_pag->pag_agno;
+
+ if (unlikely(XFS_TEST_ERROR(!agi_ok, mp, XFS_ERRTAG_IALLOC_READ_AGI,
+ XFS_RANDOM_IALLOC_READ_AGI))) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, agi);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+ xfs_check_agi_unlinked(agi);
+}
+
+static void
+xfs_agi_read_verify(
+ struct xfs_buf *bp)
+{
+ xfs_agi_verify(bp);
+}
+
+static void
+xfs_agi_write_verify(
+ struct xfs_buf *bp)
+{
+ xfs_agi_verify(bp);
+}
+
+const struct xfs_buf_ops xfs_agi_buf_ops = {
+ .verify_read = xfs_agi_read_verify,
+ .verify_write = xfs_agi_write_verify,
+};
+
/*
* Read in the allocation group header (inode allocation section)
*/
@@ -1289,38 +1341,18 @@ xfs_read_agi(
xfs_agnumber_t agno, /* allocation group number */
struct xfs_buf **bpp) /* allocation group hdr buf */
{
- struct xfs_agi *agi; /* allocation group header */
- int agi_ok; /* agi is consistent */
int error;
ASSERT(agno != NULLAGNUMBER);
error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp)),
- XFS_FSS_TO_BB(mp, 1), 0, bpp);
+ XFS_FSS_TO_BB(mp, 1), 0, bpp, &xfs_agi_buf_ops);
if (error)
return error;
- ASSERT(*bpp && !XFS_BUF_GETERROR(*bpp));
- agi = XFS_BUF_TO_AGI(*bpp);
-
- /*
- * Validate the magic number of the agi block.
- */
- agi_ok = be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC &&
- XFS_AGI_GOOD_VERSION(be32_to_cpu(agi->agi_versionnum)) &&
- be32_to_cpu(agi->agi_seqno) == agno;
- if (unlikely(XFS_TEST_ERROR(!agi_ok, mp, XFS_ERRTAG_IALLOC_READ_AGI,
- XFS_RANDOM_IALLOC_READ_AGI))) {
- XFS_CORRUPTION_ERROR("xfs_read_agi", XFS_ERRLEVEL_LOW,
- mp, agi);
- xfs_trans_brelse(tp, *bpp);
- return XFS_ERROR(EFSCORRUPTED);
- }
-
- XFS_BUF_SET_VTYPE_REF(*bpp, B_FS_AGI, XFS_AGI_REF);
-
- xfs_check_agi_unlinked(agi);
+ ASSERT(!xfs_buf_geterror(*bpp));
+ xfs_buf_set_ref(*bpp, XFS_AGI_REF);
return 0;
}
diff --git a/libxfs/xfs_ialloc_btree.c b/libxfs/xfs_ialloc_btree.c
index 35dd96f..0bc24cc 100644
--- a/libxfs/xfs_ialloc_btree.c
+++ b/libxfs/xfs_ialloc_btree.c
@@ -163,6 +163,59 @@ xfs_inobt_key_diff(
cur->bc_rec.i.ir_startino;
}
+void
+xfs_inobt_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
+ unsigned int level;
+ int sblock_ok; /* block passes checks */
+
+ /* magic number and level verification */
+ level = be16_to_cpu(block->bb_level);
+ sblock_ok = block->bb_magic == cpu_to_be32(XFS_IBT_MAGIC) &&
+ level < mp->m_in_maxlevels;
+
+ /* numrecs verification */
+ sblock_ok = sblock_ok &&
+ be16_to_cpu(block->bb_numrecs) <= mp->m_inobt_mxr[level != 0];
+
+ /* sibling pointer verification */
+ sblock_ok = sblock_ok &&
+ (block->bb_u.s.bb_leftsib == cpu_to_be32(NULLAGBLOCK) ||
+ be32_to_cpu(block->bb_u.s.bb_leftsib) < mp->m_sb.sb_agblocks) &&
+ block->bb_u.s.bb_leftsib &&
+ (block->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK) ||
+ be32_to_cpu(block->bb_u.s.bb_rightsib) < mp->m_sb.sb_agblocks) &&
+ block->bb_u.s.bb_rightsib;
+
+ if (!sblock_ok) {
+ trace_xfs_btree_corrupt(bp, _RET_IP_);
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, block);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+}
+
+static void
+xfs_inobt_read_verify(
+ struct xfs_buf *bp)
+{
+ xfs_inobt_verify(bp);
+}
+
+static void
+xfs_inobt_write_verify(
+ struct xfs_buf *bp)
+{
+ xfs_inobt_verify(bp);
+}
+
+const struct xfs_buf_ops xfs_inobt_buf_ops = {
+ .verify_read = xfs_inobt_read_verify,
+ .verify_write = xfs_inobt_write_verify,
+};
+
#ifdef DEBUG
STATIC int
xfs_inobt_keys_inorder(
@@ -266,7 +319,7 @@ static const struct xfs_btree_ops xfs_inobt_ops = {
.init_rec_from_cur = xfs_inobt_init_rec_from_cur,
.init_ptr_from_cur = xfs_inobt_init_ptr_from_cur,
.key_diff = xfs_inobt_key_diff,
-
+ .buf_ops = &xfs_inobt_buf_ops,
#ifdef DEBUG
.keys_inorder = xfs_inobt_keys_inorder,
.recs_inorder = xfs_inobt_recs_inorder,
diff --git a/libxfs/xfs_inode.c b/libxfs/xfs_inode.c
index e4474fd..2970f46 100644
--- a/libxfs/xfs_inode.c
+++ b/libxfs/xfs_inode.c
@@ -21,6 +21,12 @@
kmem_zone_t *xfs_ifork_zone;
kmem_zone_t *xfs_inode_zone;
+/*
+ * Used in xfs_itruncate_extents(). This is the maximum number of extents
+ * freed from a file in a single transaction.
+ */
+#define XFS_ITRUNC_MAX_EXTENTS 2
+
STATIC int xfs_iformat_local(xfs_inode_t *, xfs_dinode_t *, int, int);
STATIC int xfs_iformat_extents(xfs_inode_t *, xfs_dinode_t *, int);
STATIC int xfs_iformat_btree(xfs_inode_t *, xfs_dinode_t *, int);
@@ -73,8 +79,8 @@ xfs_inobp_check(
dip = (xfs_dinode_t *)xfs_buf_offset(bp,
i * mp->m_sb.sb_inodesize);
if (!dip->di_next_unlinked) {
- xfs_fs_cmn_err(CE_ALERT, mp,
- "Detected a bogus zero next_unlinked field in incore inode buffer 0x%p. About to pop an ASSERT.",
+ xfs_alert(mp,
+ "Detected bogus zero next_unlinked field in incore inode buffer 0x%p.",
bp);
ASSERT(dip->di_next_unlinked);
}
@@ -82,176 +88,108 @@ xfs_inobp_check(
}
#endif
-/*
- * Find the buffer associated with the given inode map
- * We do basic validation checks on the buffer once it has been
- * retrieved from disk.
- */
-int
-xfs_imap_to_bp(
- xfs_mount_t *mp,
- xfs_trans_t *tp,
- struct xfs_imap *imap,
- xfs_buf_t **bpp,
- uint buf_flags,
- uint iget_flags)
+static void
+xfs_inode_buf_verify(
+ struct xfs_buf *bp)
{
- int error;
+ struct xfs_mount *mp = bp->b_target->bt_mount;
int i;
int ni;
- xfs_buf_t *bp;
-
- error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, imap->im_blkno,
- (int)imap->im_len, buf_flags, &bp);
- if (error) {
- if (error != EAGAIN) {
- cmn_err(CE_WARN,
- "xfs_imap_to_bp: xfs_trans_read_buf()returned "
- "an error %d on %s. Returning error.",
- error, mp->m_fsname);
- } else {
- ASSERT(buf_flags & XBF_TRYLOCK);
- }
- return error;
- }
/*
* Validate the magic number and version of every inode in the buffer
- * (if DEBUG kernel) or the first inode in the buffer, otherwise.
*/
-#ifdef DEBUG
- ni = BBTOB(imap->im_len) >> mp->m_sb.sb_inodelog;
-#else /* usual case */
- ni = 1;
-#endif
-
+ ni = XFS_BB_TO_FSB(mp, bp->b_length) * mp->m_sb.sb_inopblock;
for (i = 0; i < ni; i++) {
int di_ok;
xfs_dinode_t *dip;
- dip = (xfs_dinode_t *)xfs_buf_offset(bp,
+ dip = (struct xfs_dinode *)xfs_buf_offset(bp,
(i << mp->m_sb.sb_inodelog));
- di_ok = be16_to_cpu(dip->di_magic) == XFS_DINODE_MAGIC &&
+ di_ok = dip->di_magic == cpu_to_be16(XFS_DINODE_MAGIC) &&
XFS_DINODE_GOOD_VERSION(dip->di_version);
if (unlikely(XFS_TEST_ERROR(!di_ok, mp,
XFS_ERRTAG_ITOBP_INOTOBP,
XFS_RANDOM_ITOBP_INOTOBP))) {
- if (iget_flags & XFS_IGET_UNTRUSTED) {
- xfs_trans_brelse(tp, bp);
- return XFS_ERROR(EINVAL);
- }
- XFS_CORRUPTION_ERROR("xfs_imap_to_bp",
- XFS_ERRLEVEL_HIGH, mp, dip);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_HIGH,
+ mp, dip);
#ifdef DEBUG
- cmn_err(CE_PANIC,
- "Device %s - bad inode magic/vsn "
- "daddr %lld #%d (magic=%x)",
- XFS_BUFTARG_NAME(mp->m_ddev_targp),
- (unsigned long long)imap->im_blkno, i,
+ xfs_emerg(mp,
+ "bad inode magic/vsn daddr %lld #%d (magic=%x)",
+ (unsigned long long)bp->b_bn, i,
be16_to_cpu(dip->di_magic));
+ ASSERT(0);
#endif
- xfs_trans_brelse(tp, bp);
- return XFS_ERROR(EFSCORRUPTED);
}
}
-
xfs_inobp_check(mp, bp);
+}
- /*
- * Mark the buffer as an inode buffer now that it looks good
- */
- XFS_BUF_SET_VTYPE(bp, B_FS_INO);
- *bpp = bp;
- return 0;
+static void
+xfs_inode_buf_read_verify(
+ struct xfs_buf *bp)
+{
+ xfs_inode_buf_verify(bp);
}
-/*
- * This routine is called to map an inode number within a file
- * system to the buffer containing the on-disk version of the
- * inode. It returns a pointer to the buffer containing the
- * on-disk inode in the bpp parameter, and in the dip parameter
- * it returns a pointer to the on-disk inode within that buffer.
- *
- * If a non-zero error is returned, then the contents of bpp and
- * dipp are undefined.
- *
- * Use xfs_imap() to determine the size and location of the
- * buffer to read from disk.
- */
-int
-xfs_inotobp(
- xfs_mount_t *mp,
- xfs_trans_t *tp,
- xfs_ino_t ino,
- xfs_dinode_t **dipp,
- xfs_buf_t **bpp,
- int *offset,
- uint imap_flags)
+static void
+xfs_inode_buf_write_verify(
+ struct xfs_buf *bp)
{
- struct xfs_imap imap;
- xfs_buf_t *bp;
- int error;
-
- imap.im_blkno = 0;
- error = xfs_imap(mp, tp, ino, &imap, imap_flags);
- if (error)
- return error;
-
- error = xfs_imap_to_bp(mp, tp, &imap, &bp, XBF_LOCK, imap_flags);
- if (error)
- return error;
-
- *dipp = (xfs_dinode_t *)xfs_buf_offset(bp, imap.im_boffset);
- *bpp = bp;
- *offset = imap.im_boffset;
- return 0;
+ xfs_inode_buf_verify(bp);
}
+const struct xfs_buf_ops xfs_inode_buf_ops = {
+ .verify_read = xfs_inode_buf_read_verify,
+ .verify_write = xfs_inode_buf_write_verify,
+};
+
/*
- * This routine is called to map an inode to the buffer containing
- * the on-disk version of the inode. It returns a pointer to the
- * buffer containing the on-disk inode in the bpp parameter, and in
- * the dip parameter it returns a pointer to the on-disk inode within
- * that buffer.
+ * This routine is called to map an inode to the buffer containing the on-disk
+ * version of the inode. It returns a pointer to the buffer containing the
+ * on-disk inode in the bpp parameter, and in the dipp parameter it returns a
+ * pointer to the on-disk inode within that buffer.
*
- * If a non-zero error is returned, then the contents of bpp and
- * dipp are undefined.
- *
- * The inode is expected to already been mapped to its buffer and read
- * in once, thus we can use the mapping information stored in the inode
- * rather than calling xfs_imap(). This allows us to avoid the overhead
- * of looking at the inode btree for small block file systems
- * (see xfs_imap()).
+ * If a non-zero error is returned, then the contents of bpp and dipp are
+ * undefined.
*/
int
-xfs_itobp(
- xfs_mount_t *mp,
- xfs_trans_t *tp,
- xfs_inode_t *ip,
- xfs_dinode_t **dipp,
- xfs_buf_t **bpp,
- uint buf_flags)
+xfs_imap_to_bp(
+ struct xfs_mount *mp,
+ struct xfs_trans *tp,
+ struct xfs_imap *imap,
+ struct xfs_dinode **dipp,
+ struct xfs_buf **bpp,
+ uint buf_flags,
+ uint iget_flags)
{
- xfs_buf_t *bp;
- int error;
+ struct xfs_buf *bp;
+ int error;
- ASSERT(ip->i_imap.im_blkno != 0);
+ buf_flags |= XBF_UNMAPPED;
+ error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, imap->im_blkno,
+ (int)imap->im_len, buf_flags, &bp,
+ &xfs_inode_buf_ops);
+ if (error) {
+ if (error == EAGAIN) {
+ ASSERT(buf_flags & XBF_TRYLOCK);
+ return error;
+ }
- error = xfs_imap_to_bp(mp, tp, &ip->i_imap, &bp, buf_flags, 0);
- if (error)
- return error;
+ if (error == EFSCORRUPTED &&
+ (iget_flags & XFS_IGET_UNTRUSTED))
+ return XFS_ERROR(EINVAL);
- if (!bp) {
- ASSERT(buf_flags & XBF_TRYLOCK);
- ASSERT(tp == NULL);
- *bpp = NULL;
- return EAGAIN;
+ xfs_warn(mp, "%s: xfs_trans_read_buf() returned error %d.",
+ __func__, error);
+ return error;
}
- *dipp = (xfs_dinode_t *)xfs_buf_offset(bp, ip->i_imap.im_boffset);
*bpp = bp;
+ *dipp = (struct xfs_dinode *)xfs_buf_offset(bp, imap->im_boffset);
return 0;
}
@@ -264,23 +202,20 @@ xfs_itobp(
* brought in-core. The rest will be in-lined in if_extents when it
* is first referenced (see xfs_iread_extents()).
*/
-int
+STATIC int
xfs_iformat(
xfs_inode_t *ip,
xfs_dinode_t *dip)
{
xfs_attr_shortform_t *atp;
int size;
- int error;
+ int error = 0;
xfs_fsize_t di_size;
- ip->i_df.if_ext_max =
- XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t);
- error = 0;
if (unlikely(be32_to_cpu(dip->di_nextents) +
be16_to_cpu(dip->di_anextents) >
be64_to_cpu(dip->di_nblocks))) {
- xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount,
+ xfs_warn(ip->i_mount,
"corrupt dinode %Lu, extent total = %d, nblocks = %Lu.",
(unsigned long long)ip->i_ino,
(int)(be32_to_cpu(dip->di_nextents) +
@@ -293,8 +228,7 @@ xfs_iformat(
}
if (unlikely(dip->di_forkoff > ip->i_mount->m_sb.sb_inodesize)) {
- xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount,
- "corrupt dinode %Lu, forkoff = 0x%x.",
+ xfs_warn(ip->i_mount, "corrupt dinode %Lu, forkoff = 0x%x.",
(unsigned long long)ip->i_ino,
dip->di_forkoff);
XFS_CORRUPTION_ERROR("xfs_iformat(2)", XFS_ERRLEVEL_LOW,
@@ -304,7 +238,7 @@ xfs_iformat(
if (unlikely((ip->i_d.di_flags & XFS_DIFLAG_REALTIME) &&
!ip->i_mount->m_rtdev)) {
- xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount,
+ xfs_warn(ip->i_mount,
"corrupt dinode %Lu, has realtime flag set.",
ip->i_ino);
XFS_CORRUPTION_ERROR("xfs_iformat(realtime)",
@@ -323,7 +257,6 @@ xfs_iformat(
return XFS_ERROR(EFSCORRUPTED);
}
ip->i_d.di_size = 0;
- ip->i_size = 0;
ip->i_df.if_u2.if_rdev = xfs_dinode_get_rdev(dip);
break;
@@ -335,10 +268,9 @@ xfs_iformat(
/*
* no local regular files yet
*/
- if (unlikely((be16_to_cpu(dip->di_mode) & S_IFMT) == S_IFREG)) {
- xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount,
- "corrupt inode %Lu "
- "(local format for regular file).",
+ if (unlikely(S_ISREG(be16_to_cpu(dip->di_mode)))) {
+ xfs_warn(ip->i_mount,
+ "corrupt inode %Lu (local format for regular file).",
(unsigned long long) ip->i_ino);
XFS_CORRUPTION_ERROR("xfs_iformat(4)",
XFS_ERRLEVEL_LOW,
@@ -348,9 +280,8 @@ xfs_iformat(
di_size = be64_to_cpu(dip->di_size);
if (unlikely(di_size > XFS_DFORK_DSIZE(dip, ip->i_mount))) {
- xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount,
- "corrupt inode %Lu "
- "(bad size %Ld for local inode).",
+ xfs_warn(ip->i_mount,
+ "corrupt inode %Lu (bad size %Ld for local inode).",
(unsigned long long) ip->i_ino,
(long long) di_size);
XFS_CORRUPTION_ERROR("xfs_iformat(5)",
@@ -384,19 +315,18 @@ xfs_iformat(
}
if (!XFS_DFORK_Q(dip))
return 0;
+
ASSERT(ip->i_afp == NULL);
ip->i_afp = kmem_zone_zalloc(xfs_ifork_zone, KM_SLEEP | KM_NOFS);
- ip->i_afp->if_ext_max =
- XFS_IFORK_ASIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t);
+
switch (dip->di_aformat) {
case XFS_DINODE_FMT_LOCAL:
atp = (xfs_attr_shortform_t *)XFS_DFORK_APTR(dip);
size = be16_to_cpu(atp->hdr.totsize);
if (unlikely(size < sizeof(struct xfs_attr_sf_hdr))) {
- xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount,
- "corrupt inode %Lu "
- "(bad attr fork size %Ld).",
+ xfs_warn(ip->i_mount,
+ "corrupt inode %Lu (bad attr fork size %Ld).",
(unsigned long long) ip->i_ino,
(long long) size);
XFS_CORRUPTION_ERROR("xfs_iformat(8)",
@@ -451,9 +381,8 @@ xfs_iformat_local(
* kmem_alloc() or memcpy() below.
*/
if (unlikely(size > XFS_DFORK_SIZE(dip, ip->i_mount, whichfork))) {
- xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount,
- "corrupt inode %Lu "
- "(bad size %d for local fork, size = %d).",
+ xfs_warn(ip->i_mount,
+ "corrupt inode %Lu (bad size %d for local fork, size = %d).",
(unsigned long long) ip->i_ino, size,
XFS_DFORK_SIZE(dip, ip->i_mount, whichfork));
XFS_CORRUPTION_ERROR("xfs_iformat_local", XFS_ERRLEVEL_LOW,
@@ -510,8 +439,7 @@ xfs_iformat_extents(
* kmem_alloc() or memcpy() below.
*/
if (unlikely(size < 0 || size > XFS_DFORK_SIZE(dip, ip->i_mount, whichfork))) {
- xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount,
- "corrupt inode %Lu ((a)extents = %d).",
+ xfs_warn(ip->i_mount, "corrupt inode %Lu ((a)extents = %d).",
(unsigned long long) ip->i_ino, nex);
XFS_CORRUPTION_ERROR("xfs_iformat_extents(1)", XFS_ERRLEVEL_LOW,
ip->i_mount, dip);
@@ -582,15 +510,15 @@ xfs_iformat_btree(
* or the number of extents is greater than the number of
* blocks.
*/
- if (unlikely(XFS_IFORK_NEXTENTS(ip, whichfork) <= ifp->if_ext_max
- || XFS_BMDR_SPACE_CALC(nrecs) >
- XFS_DFORK_SIZE(dip, ip->i_mount, whichfork)
- || XFS_IFORK_NEXTENTS(ip, whichfork) > ip->i_d.di_nblocks)) {
- xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount,
- "corrupt inode %Lu (btree).",
+ if (unlikely(XFS_IFORK_NEXTENTS(ip, whichfork) <=
+ XFS_IFORK_MAXEXT(ip, whichfork) ||
+ XFS_BMDR_SPACE_CALC(nrecs) >
+ XFS_DFORK_SIZE(dip, ip->i_mount, whichfork) ||
+ XFS_IFORK_NEXTENTS(ip, whichfork) > ip->i_d.di_nblocks)) {
+ xfs_warn(ip->i_mount, "corrupt inode %Lu (btree).",
(unsigned long long) ip->i_ino);
- XFS_ERROR_REPORT("xfs_iformat_btree", XFS_ERRLEVEL_LOW,
- ip->i_mount);
+ XFS_CORRUPTION_ERROR("xfs_iformat_btree", XFS_ERRLEVEL_LOW,
+ ip->i_mount, dip);
return XFS_ERROR(EFSCORRUPTED);
}
@@ -683,6 +611,124 @@ xfs_dinode_to_disk(
}
/*
+ * Read the disk inode attributes into the in-core inode structure.
+ */
+int
+xfs_iread(
+ xfs_mount_t *mp,
+ xfs_trans_t *tp,
+ xfs_inode_t *ip,
+ uint iget_flags)
+{
+ xfs_buf_t *bp;
+ xfs_dinode_t *dip;
+ int error;
+
+ /*
+ * Fill in the location information in the in-core inode.
+ */
+ error = xfs_imap(mp, tp, ip->i_ino, &ip->i_imap, iget_flags);
+ if (error)
+ return error;
+
+ /*
+ * Get pointers to the on-disk inode and the buffer containing it.
+ */
+ error = xfs_imap_to_bp(mp, tp, &ip->i_imap, &dip, &bp, 0, iget_flags);
+ if (error)
+ return error;
+
+ /*
+ * If we got something that isn't an inode it means someone
+ * (nfs or dmi) has a stale handle.
+ */
+ if (dip->di_magic != cpu_to_be16(XFS_DINODE_MAGIC)) {
+#ifdef DEBUG
+ xfs_alert(mp,
+ "%s: dip->di_magic (0x%x) != XFS_DINODE_MAGIC (0x%x)",
+ __func__, be16_to_cpu(dip->di_magic), XFS_DINODE_MAGIC);
+#endif /* DEBUG */
+ error = XFS_ERROR(EINVAL);
+ goto out_brelse;
+ }
+
+ /*
+ * If the on-disk inode is already linked to a directory
+ * entry, copy all of the inode into the in-core inode.
+ * xfs_iformat() handles copying in the inode format
+ * specific information.
+ * Otherwise, just get the truly permanent information.
+ */
+ if (dip->di_mode) {
+ xfs_dinode_from_disk(&ip->i_d, dip);
+ error = xfs_iformat(ip, dip);
+ if (error) {
+#ifdef DEBUG
+ xfs_alert(mp, "%s: xfs_iformat() returned error %d",
+ __func__, error);
+#endif /* DEBUG */
+ goto out_brelse;
+ }
+ } else {
+ ip->i_d.di_magic = be16_to_cpu(dip->di_magic);
+ ip->i_d.di_version = dip->di_version;
+ ip->i_d.di_gen = be32_to_cpu(dip->di_gen);
+ ip->i_d.di_flushiter = be16_to_cpu(dip->di_flushiter);
+ /*
+ * Make sure to pull in the mode here as well in
+ * case the inode is released without being used.
+ * This ensures that xfs_inactive() will see that
+ * the inode is already free and not try to mess
+ * with the uninitialized part of it.
+ */
+ ip->i_d.di_mode = 0;
+ }
+
+ /*
+ * The inode format changed when we moved the link count and
+ * made it 32 bits long. If this is an old format inode,
+ * convert it in memory to look like a new one. If it gets
+ * flushed to disk we will convert back before flushing or
+ * logging it. We zero out the new projid field and the old link
+ * count field. We'll handle clearing the pad field (the remains
+ * of the old uuid field) when we actually convert the inode to
+ * the new format. We don't change the version number so that we
+ * can distinguish this from a real new format inode.
+ */
+ if (ip->i_d.di_version == 1) {
+ ip->i_d.di_nlink = ip->i_d.di_onlink;
+ ip->i_d.di_onlink = 0;
+ xfs_set_projid(&ip->i_d, 0);
+ }
+
+ ip->i_delayed_blks = 0;
+
+ /*
+ * Mark the buffer containing the inode as something to keep
+ * around for a while. This helps to keep recently accessed
+ * meta-data in-core longer.
+ */
+ xfs_buf_set_ref(bp, XFS_INO_REF);
+
+ /*
+ * Use xfs_trans_brelse() to release the buffer containing the
+ * on-disk inode, because it was acquired with xfs_trans_read_buf()
+ * in xfs_imap_to_bp() above. If tp is NULL, this is just a normal
+ * brelse(). If we're within a transaction, then xfs_trans_brelse()
+ * will only release the buffer if it is not dirty within the
+ * transaction. It will be OK to release the buffer in this case,
+ * because inodes on disk are never destroyed and we will be
+ * locking the new in-core inode before putting it in the hash
+ * table where other processes can find it. Thus we don't have
+ * to worry about the inode being changed just because we released
+ * the buffer.
+ */
+ out_brelse:
+ xfs_trans_brelse(tp, bp);
+ return error;
+}
+
+/*
* Read in extents from a btree-format inode.
* Allocate and fill in if_extents. Real work is done in xfs_bmap.c.
*/
@@ -707,7 +753,6 @@ xfs_iread_extents(
/*
* We know that the size is valid (it's checked in iformat_btree)
*/
- ifp->if_lastex = NULLEXTNUM;
ifp->if_bytes = ifp->if_real_bytes = 0;
ifp->if_flags |= XFS_IFEXTENTS;
xfs_iext_add(ifp, 0, nextents);
@@ -1067,9 +1112,6 @@ xfs_iflush_fork(
char *cp;
xfs_ifork_t *ifp;
xfs_mount_t *mp;
-#ifdef XFS_TRANS_DEBUG
- int first;
-#endif
static const short brootflag[2] =
{ XFS_ILOG_DBROOT, XFS_ILOG_ABROOT };
static const short dataflag[2] =
@@ -1092,7 +1134,7 @@ xfs_iflush_fork(
mp = ip->i_mount;
switch (XFS_IFORK_FORMAT(ip, whichfork)) {
case XFS_DINODE_FMT_LOCAL:
- if ((iip->ili_format.ilf_fields & dataflag[whichfork]) &&
+ if ((iip->ili_fields & dataflag[whichfork]) &&
(ifp->if_bytes > 0)) {
ASSERT(ifp->if_u1.if_data != NULL);
ASSERT(ifp->if_bytes <= XFS_IFORK_SIZE(ip, whichfork));
@@ -1102,13 +1144,10 @@ xfs_iflush_fork(
case XFS_DINODE_FMT_EXTENTS:
ASSERT((ifp->if_flags & XFS_IFEXTENTS) ||
- !(iip->ili_format.ilf_fields & extflag[whichfork]));
- ASSERT((xfs_iext_get_ext(ifp, 0) != NULL) ||
- (ifp->if_bytes == 0));
- ASSERT((xfs_iext_get_ext(ifp, 0) == NULL) ||
- (ifp->if_bytes > 0));
- if ((iip->ili_format.ilf_fields & extflag[whichfork]) &&
+ !(iip->ili_fields & extflag[whichfork]));
+ if ((iip->ili_fields & extflag[whichfork]) &&
(ifp->if_bytes > 0)) {
+ ASSERT(xfs_iext_get_ext(ifp, 0));
ASSERT(XFS_IFORK_NEXTENTS(ip, whichfork) > 0);
(void)xfs_iextents_copy(ip, (xfs_bmbt_rec_t *)cp,
whichfork);
@@ -1116,7 +1155,7 @@ xfs_iflush_fork(
break;
case XFS_DINODE_FMT_BTREE:
- if ((iip->ili_format.ilf_fields & brootflag[whichfork]) &&
+ if ((iip->ili_fields & brootflag[whichfork]) &&
(ifp->if_broot_bytes > 0)) {
ASSERT(ifp->if_broot != NULL);
ASSERT(ifp->if_broot_bytes <=
@@ -1129,14 +1168,14 @@ xfs_iflush_fork(
break;
case XFS_DINODE_FMT_DEV:
- if (iip->ili_format.ilf_fields & XFS_ILOG_DEV) {
+ if (iip->ili_fields & XFS_ILOG_DEV) {
ASSERT(whichfork == XFS_DATA_FORK);
xfs_dinode_put_rdev(dip, ip->i_df.if_u2.if_rdev);
}
break;
case XFS_DINODE_FMT_UUID:
- if (iip->ili_format.ilf_fields & XFS_ILOG_UUID) {
+ if (iip->ili_fields & XFS_ILOG_UUID) {
ASSERT(whichfork == XFS_DATA_FORK);
memcpy(XFS_DFORK_DPTR(dip),
&ip->i_df.if_u2.if_uuid,
@@ -1159,6 +1198,8 @@ xfs_iext_get_ext(
xfs_extnum_t idx) /* index of target extent */
{
ASSERT(idx >= 0);
+ ASSERT(idx < ifp->if_bytes / sizeof(xfs_bmbt_rec_t));
+
if ((ifp->if_flags & XFS_IFEXTIREC) && (idx == 0)) {
return ifp->if_u1.if_ext_irec->er_extbuf;
} else if (ifp->if_flags & XFS_IFEXTIREC) {
@@ -1238,7 +1279,6 @@ xfs_iext_add(
}
ifp->if_u1.if_extents = ifp->if_u2.if_inline_ext;
ifp->if_real_bytes = 0;
- ifp->if_lastex = nextents + ext_diff;
}
/*
* Otherwise use a linear (direct) extent list.
@@ -1933,8 +1973,10 @@ xfs_iext_idx_to_irec(
xfs_extnum_t page_idx = *idxp; /* extent index in target list */
ASSERT(ifp->if_flags & XFS_IFEXTIREC);
- ASSERT(page_idx >= 0 && page_idx <=
- ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t));
+ ASSERT(page_idx >= 0);
+ ASSERT(page_idx <= ifp->if_bytes / sizeof(xfs_bmbt_rec_t));
+ ASSERT(page_idx < ifp->if_bytes / sizeof(xfs_bmbt_rec_t) || realloc);
+
nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
erp_idx = 0;
low = 0;
diff --git a/libxfs/xfs_mount.c b/libxfs/xfs_mount.c
index 32d2255..a9155b3 100644
--- a/libxfs/xfs_mount.c
+++ b/libxfs/xfs_mount.c
@@ -90,8 +90,8 @@ xfs_perag_get(struct xfs_mount *mp, xfs_agnumber_t agno)
ASSERT(atomic_read(&pag->pag_ref) >= 0);
ref = atomic_inc_return(&pag->pag_ref);
}
- trace_xfs_perag_get(mp, agno, ref, _RET_IP_);
rcu_read_unlock();
+ trace_xfs_perag_get(mp, agno, ref, _RET_IP_);
return pag;
}
@@ -105,6 +105,114 @@ xfs_perag_put(struct xfs_perag *pag)
trace_xfs_perag_put(pag->pag_mount, pag->pag_agno, ref, _RET_IP_);
}
+/*
+ * Check the validity of the SB found.
+ */
+STATIC int
+xfs_mount_validate_sb(
+ xfs_mount_t *mp,
+ xfs_sb_t *sbp,
+ bool check_inprogress)
+{
+
+ /*
+ * If the log device and data device have the
+ * same device number, the log is internal.
+ * Consequently, the sb_logstart should be non-zero. If
+ * we have a zero sb_logstart in this case, we may be trying to mount
+ * a volume filesystem in a non-volume manner.
+ */
+ if (sbp->sb_magicnum != XFS_SB_MAGIC) {
+ xfs_warn(mp, "bad magic number");
+ return XFS_ERROR(EWRONGFS);
+ }
+
+ if (!xfs_sb_good_version(sbp)) {
+ xfs_warn(mp, "bad version");
+ return XFS_ERROR(EWRONGFS);
+ }
+
+ if (unlikely(
+ sbp->sb_logstart == 0 && mp->m_logdev == mp->m_dev)) {
+ xfs_warn(mp,
+ "filesystem is marked as having an external log; "
+ "specify logdev on the mount command line.");
+ return XFS_ERROR(EINVAL);
+ }
+
+ if (unlikely(
+ sbp->sb_logstart != 0 && mp->m_logdev != mp->m_dev)) {
+ xfs_warn(mp,
+ "filesystem is marked as having an internal log; "
+ "do not specify logdev on the mount command line.");
+ return XFS_ERROR(EINVAL);
+ }
+
+ /*
+ * More sanity checking. Most of these were stolen directly from
+ * xfs_repair.
+ */
+ if (unlikely(
+ sbp->sb_agcount <= 0 ||
+ sbp->sb_sectsize < XFS_MIN_SECTORSIZE ||
+ sbp->sb_sectsize > XFS_MAX_SECTORSIZE ||
+ sbp->sb_sectlog < XFS_MIN_SECTORSIZE_LOG ||
+ sbp->sb_sectlog > XFS_MAX_SECTORSIZE_LOG ||
+ sbp->sb_sectsize != (1 << sbp->sb_sectlog) ||
+ sbp->sb_blocksize < XFS_MIN_BLOCKSIZE ||
+ sbp->sb_blocksize > XFS_MAX_BLOCKSIZE ||
+ sbp->sb_blocklog < XFS_MIN_BLOCKSIZE_LOG ||
+ sbp->sb_blocklog > XFS_MAX_BLOCKSIZE_LOG ||
+ sbp->sb_blocksize != (1 << sbp->sb_blocklog) ||
+ sbp->sb_inodesize < XFS_DINODE_MIN_SIZE ||
+ sbp->sb_inodesize > XFS_DINODE_MAX_SIZE ||
+ sbp->sb_inodelog < XFS_DINODE_MIN_LOG ||
+ sbp->sb_inodelog > XFS_DINODE_MAX_LOG ||
+ sbp->sb_inodesize != (1 << sbp->sb_inodelog) ||
+ (sbp->sb_blocklog - sbp->sb_inodelog != sbp->sb_inopblog) ||
+ (sbp->sb_rextsize * sbp->sb_blocksize > XFS_MAX_RTEXTSIZE) ||
+ (sbp->sb_rextsize * sbp->sb_blocksize < XFS_MIN_RTEXTSIZE) ||
+ (sbp->sb_imax_pct > 100 /* zero sb_imax_pct is valid */) ||
+ sbp->sb_dblocks == 0 ||
+ sbp->sb_dblocks > XFS_MAX_DBLOCKS(sbp) ||
+ sbp->sb_dblocks < XFS_MIN_DBLOCKS(sbp))) {
+ XFS_CORRUPTION_ERROR("SB sanity check failed",
+ XFS_ERRLEVEL_LOW, mp, sbp);
+ return XFS_ERROR(EFSCORRUPTED);
+ }
+
+ /*
+ * Currently only very few inode sizes are supported.
+ */
+ switch (sbp->sb_inodesize) {
+ case 256:
+ case 512:
+ case 1024:
+ case 2048:
+ break;
+ default:
+ xfs_warn(mp, "inode size of %d bytes not supported",
+ sbp->sb_inodesize);
+ return XFS_ERROR(ENOSYS);
+ }
+
+
+ if (check_inprogress && sbp->sb_inprogress) {
+ xfs_warn(mp, "Offline file system operation in progress!");
+ return XFS_ERROR(EFSCORRUPTED);
+ }
+
+ /*
+ * Version 1 directory format has never worked on Linux.
+ */
+ if (unlikely(!xfs_sb_version_hasdirv2(sbp))) {
+ xfs_warn(mp, "file system using version 1 directory format");
+ return XFS_ERROR(ENOSYS);
+ }
+
+ return 0;
+}
+
void
xfs_sb_from_disk(
xfs_sb_t *to,
@@ -211,6 +319,72 @@ xfs_sb_to_disk(
}
}
+static void
+xfs_sb_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_sb sb;
+ int error;
+
+ xfs_sb_from_disk(&sb, XFS_BUF_TO_SBP(bp));
+
+ /*
+ * Only check the in progress field for the primary superblock as
+ * mkfs.xfs doesn't clear it from secondary superblocks.
+ */
+ error = xfs_mount_validate_sb(mp, &sb, bp->b_blkno == XFS_SB_DADDR);
+ if (error)
+ xfs_buf_ioerror(bp, error);
+}
+
+static void
+xfs_sb_read_verify(
+ struct xfs_buf *bp)
+{
+ xfs_sb_verify(bp);
+}
+
+/*
+ * We may be probed for a filesystem match, so we may not want to emit
+ * messages when the superblock buffer is not actually an XFS superblock.
+ * If we find an XFS superblock, the run a normal, noisy mount because we are
+ * really going to mount it and want to know about errors.
+ */
+static void
+xfs_sb_quiet_read_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_sb sb;
+
+ xfs_sb_from_disk(&sb, XFS_BUF_TO_SBP(bp));
+
+ if (sb.sb_magicnum == XFS_SB_MAGIC) {
+ /* XFS filesystem, verify noisily! */
+ xfs_sb_read_verify(bp);
+ return;
+ }
+ /* quietly fail */
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+}
+
+static void
+xfs_sb_write_verify(
+ struct xfs_buf *bp)
+{
+ xfs_sb_verify(bp);
+}
+
+const struct xfs_buf_ops xfs_sb_buf_ops = {
+ .verify_read = xfs_sb_read_verify,
+ .verify_write = xfs_sb_write_verify,
+};
+
+static const struct xfs_buf_ops xfs_sb_quiet_buf_ops = {
+ .verify_read = xfs_sb_quiet_read_verify,
+ .verify_write = xfs_sb_write_verify,
+};
+
/*
* xfs_mount_common
*
diff --git a/libxfs/xfs_rtalloc.c b/libxfs/xfs_rtalloc.c
index 4fbdaa9..1de85fd 100644
--- a/libxfs/xfs_rtalloc.c
+++ b/libxfs/xfs_rtalloc.c
@@ -49,34 +49,24 @@ xfs_rtbuf_get(
xfs_buf_t **bpp) /* output: buffer for the block */
{
xfs_buf_t *bp; /* block buffer, result */
- xfs_daddr_t d; /* disk addr of block */
- int error; /* error value */
- xfs_fsblock_t fsb; /* fs block number for block */
xfs_inode_t *ip; /* bitmap or summary inode */
+ xfs_bmbt_irec_t map;
+ int nmap = 1;
+ int error; /* error value */
ip = issum ? mp->m_rsumip : mp->m_rbmip;
- /*
- * Map from the file offset (block) and inode number to the
- * file system block.
- */
- error = xfs_bmapi_single(tp, ip, XFS_DATA_FORK, &fsb, block);
- if (error) {
+
+ error = xfs_bmapi_read(ip, block, 1, &map, &nmap, XFS_DATA_FORK);
+ if (error)
return error;
- }
- ASSERT(fsb != NULLFSBLOCK);
- /*
- * Convert to disk address for buffer cache.
- */
- d = XFS_FSB_TO_DADDR(mp, fsb);
- /*
- * Read the buffer.
- */
- error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, d,
- mp->m_bsize, 0, &bp);
- if (error) {
+
+ ASSERT(map.br_startblock != NULLFSBLOCK);
+ error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
+ XFS_FSB_TO_DADDR(mp, map.br_startblock),
+ mp->m_bsize, 0, &bp, NULL);
+ if (error)
return error;
- }
- ASSERT(bp && !XFS_BUF_GETERROR(bp));
+ ASSERT(!xfs_buf_geterror(bp));
*bpp = bp;
return 0;
}
@@ -115,7 +105,7 @@ xfs_rtfind_back(
if (error) {
return error;
}
- bufp = (xfs_rtword_t *)XFS_BUF_PTR(bp);
+ bufp = bp->b_addr;
/*
* Get the first word's index & point to it.
*/
@@ -167,7 +157,7 @@ xfs_rtfind_back(
if (error) {
return error;
}
- bufp = (xfs_rtword_t *)XFS_BUF_PTR(bp);
+ bufp = bp->b_addr;
word = XFS_BLOCKWMASK(mp);
b = &bufp[word];
} else {
@@ -213,7 +203,7 @@ xfs_rtfind_back(
if (error) {
return error;
}
- bufp = (xfs_rtword_t *)XFS_BUF_PTR(bp);
+ bufp = bp->b_addr;
word = XFS_BLOCKWMASK(mp);
b = &bufp[word];
} else {
@@ -290,7 +280,7 @@ xfs_rtfind_forw(
if (error) {
return error;
}
- bufp = (xfs_rtword_t *)XFS_BUF_PTR(bp);
+ bufp = bp->b_addr;
/*
* Get the first word's index & point to it.
*/
@@ -341,7 +331,7 @@ xfs_rtfind_forw(
if (error) {
return error;
}
- b = bufp = (xfs_rtword_t *)XFS_BUF_PTR(bp);
+ b = bufp = bp->b_addr;
word = 0;
} else {
/*
@@ -386,7 +376,7 @@ xfs_rtfind_forw(
if (error) {
return error;
}
- b = bufp = (xfs_rtword_t *)XFS_BUF_PTR(bp);
+ b = bufp = bp->b_addr;
word = 0;
} else {
/*
@@ -537,7 +527,7 @@ xfs_rtmodify_range(
if (error) {
return error;
}
- bufp = (xfs_rtword_t *)XFS_BUF_PTR(bp);
+ bufp = bp->b_addr;
/*
* Compute the starting word's address, and starting bit.
*/
@@ -582,7 +572,7 @@ xfs_rtmodify_range(
if (error) {
return error;
}
- first = b = bufp = (xfs_rtword_t *)XFS_BUF_PTR(bp);
+ first = b = bufp = bp->b_addr;
word = 0;
} else {
/*
@@ -622,7 +612,7 @@ xfs_rtmodify_range(
if (error) {
return error;
}
- first = b = bufp = (xfs_rtword_t *)XFS_BUF_PTR(bp);
+ first = b = bufp = bp->b_addr;
word = 0;
} else {
/*
@@ -720,8 +710,8 @@ xfs_rtmodify_summary(
*/
sp = XFS_SUMPTR(mp, bp, so);
*sp += delta;
- xfs_trans_log_buf(tp, bp, (uint)((char *)sp - (char *)XFS_BUF_PTR(bp)),
- (uint)((char *)sp - (char *)XFS_BUF_PTR(bp) + sizeof(*sp) - 1));
+ xfs_trans_log_buf(tp, bp, (uint)((char *)sp - (char *)bp->b_addr),
+ (uint)((char *)sp - (char *)bp->b_addr + sizeof(*sp) - 1));
return 0;
}
@@ -736,18 +726,15 @@ xfs_rtfree_extent(
xfs_extlen_t len) /* length of extent freed */
{
int error; /* error value */
- xfs_inode_t *ip; /* bitmap file inode */
xfs_mount_t *mp; /* file system mount structure */
xfs_fsblock_t sb; /* summary file block number */
xfs_buf_t *sumbp; /* summary file block buffer */
mp = tp->t_mountp;
- /*
- * Synchronize by locking the bitmap inode.
- */
- if ((error = xfs_trans_iget(mp, tp, mp->m_sb.sb_rbmino, 0,
- XFS_ILOCK_EXCL, &ip)))
- return error;
+
+ ASSERT(mp->m_rbmip->i_itemp != NULL);
+ ASSERT(xfs_isilocked(mp->m_rbmip, XFS_ILOCK_EXCL));
+
#if defined(__KERNEL__) && defined(DEBUG)
/*
* Check to see that this whole range is currently allocated.
@@ -780,10 +767,10 @@ xfs_rtfree_extent(
*/
if (tp->t_frextents_delta + mp->m_sb.sb_frextents ==
mp->m_sb.sb_rextents) {
- if (!(ip->i_d.di_flags & XFS_DIFLAG_NEWRTBM))
- ip->i_d.di_flags |= XFS_DIFLAG_NEWRTBM;
- *(__uint64_t *)&ip->i_d.di_atime = 0;
- xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+ if (!(mp->m_rbmip->i_d.di_flags & XFS_DIFLAG_NEWRTBM))
+ mp->m_rbmip->i_d.di_flags |= XFS_DIFLAG_NEWRTBM;
+ *(__uint64_t *)&mp->m_rbmip->i_d.di_atime = 0;
+ xfs_trans_log_inode(tp, mp->m_rbmip, XFS_ILOG_CORE);
}
return 0;
}
diff --git a/libxfs/xfs_trans.c b/libxfs/xfs_trans.c
index 7249196..bdd0ebc 100644
--- a/libxfs/xfs_trans.c
+++ b/libxfs/xfs_trans.c
@@ -583,14 +583,13 @@ xfs_trans_add_item(
{
struct xfs_log_item_desc *lidp;
- ASSERT(lip->li_mountp = tp->t_mountp);
- ASSERT(lip->li_ailp = tp->t_mountp->m_ail);
+ ASSERT(lip->li_mountp == tp->t_mountp);
+ ASSERT(lip->li_ailp == tp->t_mountp->m_ail);
lidp = kmem_zone_zalloc(xfs_log_item_desc_zone, KM_SLEEP | KM_NOFS);
lidp->lid_item = lip;
lidp->lid_flags = 0;
- lidp->lid_size = 0;
list_add_tail(&lidp->lid_trans, &tp->t_items);
lip->li_desc = lidp;
@@ -673,8 +672,6 @@ xfs_trans_roll(
if (error)
return error;
- xfs_trans_ijoin(trans, dp, XFS_ILOCK_EXCL);
- xfs_trans_ihold(trans, dp);
+ xfs_trans_ijoin(trans, dp, 0);
return 0;
}
-
diff --git a/logprint/log_misc.c b/logprint/log_misc.c
index 2e3dbdf..8e72788 100644
--- a/logprint/log_misc.c
+++ b/logprint/log_misc.c
@@ -17,6 +17,7 @@
*/
#include "logprint.h"
+#include <xfs_dir1.h>
#define CLEARED_BLKS (-5)
#define ZEROED_LOG (-4)
@@ -1039,7 +1040,7 @@ xlog_print_rec_head(xlog_rec_header_t *head, int *len)
}
/* check for cleared blocks written by xlog_clear_stale_blocks() */
- if (!head->h_len && !head->h_chksum && !head->h_prev_block &&
+ if (!head->h_len && !head->h_crc && !head->h_prev_block &&
!head->h_num_logops && !head->h_size)
return CLEARED_BLKS;
diff --git a/logprint/log_print_all.c b/logprint/log_print_all.c
index 69d6e71..cbc51d9 100644
--- a/logprint/log_print_all.c
+++ b/logprint/log_print_all.c
@@ -243,7 +243,7 @@ xlog_recover_print_inode_core(
(di->di_magic>>8) & 0xff, di->di_magic & 0xff,
di->di_mode, di->di_version, di->di_format, di->di_onlink);
printf(_(" uid:%d gid:%d nlink:%d projid:%u\n"),
- di->di_uid, di->di_gid, di->di_nlink, xfs_get_projid(*di));
+ di->di_uid, di->di_gid, di->di_nlink, xfs_get_projid(di));
printf(_(" atime:%d mtime:%d ctime:%d\n"),
di->di_atime.t_sec, di->di_mtime.t_sec, di->di_ctime.t_sec);
printf(_(" flushiter:%d\n"), di->di_flushiter);
diff --git a/mkfs/proto.c b/mkfs/proto.c
index 3021028..56eed31 100644
--- a/mkfs/proto.c
+++ b/mkfs/proto.c
@@ -243,7 +243,7 @@ newfile(
} else if (len > 0) {
nb = XFS_B_TO_FSB(mp, len);
nmap = 1;
- error = libxfs_bmapi(tp, ip, 0, nb, XFS_BMAPI_WRITE, first, nb,
+ error = libxfs_bmapi_write(tp, ip, 0, nb, 0, first, nb,
&map, &nmap, flist);
if (error) {
fail(_("error allocating space for a file"), error);
@@ -667,9 +667,9 @@ rtinit(
xfs_bmap_init(&flist, &first);
while (bno < mp->m_sb.sb_rbmblocks) {
nmap = XFS_BMAP_MAX_NMAP;
- error = libxfs_bmapi(tp, rbmip, bno,
+ error = libxfs_bmapi_write(tp, rbmip, bno,
(xfs_extlen_t)(mp->m_sb.sb_rbmblocks - bno),
- XFS_BMAPI_WRITE, &first, mp->m_sb.sb_rbmblocks,
+ 0, &first, mp->m_sb.sb_rbmblocks,
map, &nmap, &flist);
if (error) {
fail(_("Allocation of the realtime bitmap failed"),
@@ -704,9 +704,9 @@ rtinit(
xfs_bmap_init(&flist, &first);
while (bno < nsumblocks) {
nmap = XFS_BMAP_MAX_NMAP;
- error = libxfs_bmapi(tp, rsumip, bno,
+ error = libxfs_bmapi_write(tp, rsumip, bno,
(xfs_extlen_t)(nsumblocks - bno),
- XFS_BMAPI_WRITE, &first, nsumblocks,
+ 0, &first, nsumblocks,
map, &nmap, &flist);
if (error) {
fail(_("Allocation of the realtime summary failed"),
diff --git a/repair/dir.c b/repair/dir.c
index 01c8f10..daca17d 100644
--- a/repair/dir.c
+++ b/repair/dir.c
@@ -17,6 +17,7 @@
*/
#include <libxfs.h>
+#include <xfs_dir1.h>
#include "avl.h"
#include "globals.h"
#include "agheader.h"
diff --git a/repair/dir2.c b/repair/dir2.c
index 7a614a8..c455c83 100644
--- a/repair/dir2.c
+++ b/repair/dir2.c
@@ -79,182 +79,36 @@ dir2_is_badino(
* Multibuffer handling.
* V2 directory blocks can be noncontiguous, needing multiple buffers.
*/
-static xfs_dabuf_t *
+static struct xfs_buf *
da_read_buf(
xfs_mount_t *mp,
int nex,
bmap_ext_t *bmp)
{
- xfs_buf_t *bp;
- xfs_buf_t *bparray[4];
- xfs_buf_t **bplist;
- xfs_dabuf_t *dabuf;
+#define MAP_ARRAY_SZ 4
+ struct xfs_buf_map map_array[MAP_ARRAY_SZ];
+ struct xfs_buf_map *map;
+ struct xfs_buf *bp;
int i;
- int off;
- if (nex > (sizeof(bparray)/sizeof(xfs_buf_t *))) {
- bplist = calloc(nex, sizeof(*bplist));
- if (bplist == NULL) {
+ if (nex > MAP_ARRAY_SZ) {
+ map = calloc(nex, sizeof(*map));
+ if (map == NULL) {
do_error(_("couldn't malloc dir2 buffer list\n"));
exit(1);
}
- }
- else {
+ } else {
/* common case avoids calloc/free */
- bplist = bparray;
+ map = map_array;
}
for (i = 0; i < nex; i++) {
- pftrace("about to read off %llu (len = %d)",
- (long long)XFS_FSB_TO_DADDR(mp, bmp[i].startblock),
- XFS_FSB_TO_BB(mp, bmp[i].blockcount));
-
- bplist[i] = libxfs_readbuf(mp->m_dev,
- XFS_FSB_TO_DADDR(mp, bmp[i].startblock),
- XFS_FSB_TO_BB(mp, bmp[i].blockcount), 0);
- if (!bplist[i]) {
- nex = i;
- goto failed;
- }
-
- pftrace("readbuf %p (%llu, %d)", bplist[i],
- (long long)XFS_BUF_ADDR(bplist[i]),
- XFS_BUF_COUNT(bplist[i]));
- }
- dabuf = malloc(XFS_DA_BUF_SIZE(nex));
- if (dabuf == NULL) {
- do_error(_("couldn't malloc dir2 buffer header\n"));
- exit(1);
- }
- dabuf->dirty = 0;
- dabuf->nbuf = nex;
- if (nex == 1) {
- bp = bplist[0];
- dabuf->bbcount = (short)BTOBB(XFS_BUF_COUNT(bp));
- dabuf->data = XFS_BUF_PTR(bp);
- dabuf->bps[0] = bp;
- } else {
- for (i = 0, dabuf->bbcount = 0; i < nex; i++) {
- dabuf->bps[i] = bp = bplist[i];
- dabuf->bbcount += BTOBB(XFS_BUF_COUNT(bp));
- }
- dabuf->data = malloc(BBTOB(dabuf->bbcount));
- if (dabuf->data == NULL) {
- do_error(_("couldn't malloc dir2 buffer data\n"));
- exit(1);
- }
- for (i = off = 0; i < nex; i++, off += XFS_BUF_COUNT(bp)) {
- bp = bplist[i];
- memmove((char *)dabuf->data + off, XFS_BUF_PTR(bp),
- XFS_BUF_COUNT(bp));
- }
- }
- if (bplist != bparray)
- free(bplist);
- return dabuf;
-failed:
- for (i = 0; i < nex; i++)
- libxfs_putbuf(bplist[i]);
- if (bplist != bparray)
- free(bplist);
- return NULL;
-}
-
-static void
-da_buf_clean(
- xfs_dabuf_t *dabuf)
-{
- xfs_buf_t *bp;
- int i;
- int off;
-
- if (dabuf->dirty) {
- dabuf->dirty = 0;
- for (i=off=0; i < dabuf->nbuf; i++, off += XFS_BUF_COUNT(bp)) {
- bp = dabuf->bps[i];
- memmove(XFS_BUF_PTR(bp), (char *)dabuf->data + off,
- XFS_BUF_COUNT(bp));
- }
- }
-}
-
-static void
-da_buf_done(
- xfs_dabuf_t *dabuf)
-{
- da_buf_clean(dabuf);
- if (dabuf->nbuf > 1)
- free(dabuf->data);
- free(dabuf);
-}
-
-static int
-da_bwrite(
- xfs_mount_t *mp,
- xfs_dabuf_t *dabuf)
-{
- xfs_buf_t *bp;
- xfs_buf_t **bplist;
- int e;
- int error;
- int i;
- int nbuf;
- int off;
-
- if ((nbuf = dabuf->nbuf) == 1) {
- bplist = &bp;
- bp = dabuf->bps[0];
- } else {
- bplist = malloc(nbuf * sizeof(*bplist));
- if (bplist == NULL) {
- do_error(_("couldn't malloc dir2 buffer list\n"));
- exit(1);
- }
- memmove(bplist, dabuf->bps, nbuf * sizeof(*bplist));
- for (i = off = 0; i < nbuf; i++, off += XFS_BUF_COUNT(bp)) {
- bp = bplist[i];
- memmove(XFS_BUF_PTR(bp), (char *)dabuf->data + off,
- XFS_BUF_COUNT(bp));
- }
- }
- da_buf_done(dabuf);
- for (i = error = 0; i < nbuf; i++) {
- e = libxfs_writebuf(bplist[i], 0);
- if (e)
- error = e;
- }
- if (bplist != &bp)
- free(bplist);
- return error;
-}
-
-static void
-da_brelse(
- xfs_dabuf_t *dabuf)
-{
- xfs_buf_t *bp;
- xfs_buf_t **bplist;
- int i;
- int nbuf;
-
- if ((nbuf = dabuf->nbuf) == 1) {
- bplist = &bp;
- bp = dabuf->bps[0];
- } else {
- bplist = malloc(nbuf * sizeof(*bplist));
- if (bplist == NULL) {
- do_error(_("couldn't malloc dir2 buffer list\n"));
- exit(1);
- }
- memmove(bplist, dabuf->bps, nbuf * sizeof(*bplist));
- }
- da_buf_done(dabuf);
- for (i = 0; i < nbuf; i++) {
- pftrace("putbuf %p (%llu)", bplist[i],
- (long long)XFS_BUF_ADDR(bplist[i]));
- libxfs_putbuf(bplist[i]);
+ map[i].bm_bn = XFS_FSB_TO_DADDR(mp, bmp[i].startblock);
+ map[i].bm_len = XFS_FSB_TO_BB(mp, bmp[i].blockcount);
}
- if (bplist != &bp)
- free(bplist);
+ bp = libxfs_readbuf_map(mp->m_dev, map, nex, 0);
+ if (map != map_array)
+ free(map);
+ return bp;
}
/*
@@ -270,7 +124,7 @@ traverse_int_dir2block(xfs_mount_t *mp,
{
bmap_ext_t *bmp;
xfs_dablk_t bno;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
int i;
int nex;
xfs_da_blkinfo_t *info;
@@ -307,7 +161,7 @@ _("can't read block %u for directory inode %" PRIu64 "\n"),
goto error_out;
}
- info = bp->data;
+ info = bp->b_addr;
if (be16_to_cpu(info->magic) == XFS_DIR2_LEAFN_MAGIC) {
if ( i != -1 ) {
@@ -316,10 +170,10 @@ _("found non-root LEAFN node in inode %" PRIu64 " bno = %u\n"),
da_cursor->ino, bno);
}
*rbno = 0;
- da_brelse(bp);
+ libxfs_putbuf(bp);
return(1);
} else if (be16_to_cpu(info->magic) != XFS_DA_NODE_MAGIC) {
- da_brelse(bp);
+ libxfs_putbuf(bp);
do_warn(
_("bad dir magic number 0x%x in inode %" PRIu64 " bno = %u\n"),
be16_to_cpu(info->magic),
@@ -328,7 +182,7 @@ _("bad dir magic number 0x%x in inode %" PRIu64 " bno = %u\n"),
}
node = (xfs_da_intnode_t*)info;
if (be16_to_cpu(node->hdr.count) > mp->m_dir_node_ents) {
- da_brelse(bp);
+ libxfs_putbuf(bp);
do_warn(
_("bad record count in inode %" PRIu64 ", count = %d, max = %d\n"), da_cursor->ino,
be16_to_cpu(node->hdr.count),
@@ -344,7 +198,7 @@ _("bad record count in inode %" PRIu64 ", count = %d, max = %d\n"), da_cursor->i
do_warn(
_("bad header depth for directory inode %" PRIu64 "\n"),
da_cursor->ino);
- da_brelse(bp);
+ libxfs_putbuf(bp);
i = -1;
goto error_out;
}
@@ -355,7 +209,7 @@ _("bad header depth for directory inode %" PRIu64 "\n"),
do_warn(
_("bad directory btree for directory inode %" PRIu64 "\n"),
da_cursor->ino);
- da_brelse(bp);
+ libxfs_putbuf(bp);
goto error_out;
}
}
@@ -380,7 +234,7 @@ _("bad directory btree for directory inode %" PRIu64 "\n"),
error_out:
while (i > 1 && i <= da_cursor->active) {
- da_brelse(da_cursor->level[i].bp);
+ libxfs_putbuf(da_cursor->level[i].bp);
i++;
}
@@ -409,7 +263,7 @@ release_dir2_cursor_int(xfs_mount_t *mp,
}
ASSERT(error != 0);
- da_brelse(cursor->level[level].bp);
+ libxfs_putbuf(cursor->level[level].bp);
cursor->level[level].bp = NULL;
}
@@ -458,7 +312,7 @@ verify_final_dir2_path(xfs_mount_t *mp,
* in the block which should be the final (rightmost) entry
*/
entry = cursor->level[this_level].index;
- node = (xfs_da_intnode_t *)(cursor->level[this_level].bp->data);
+ node = (xfs_da_intnode_t *)(cursor->level[this_level].bp->b_addr);
/*
* check internal block consistency on this level -- ensure
* that all entries are used, encountered and expected hashvals
@@ -530,9 +384,9 @@ _("would correct bad hashval in non-leaf dir block\n"
(cursor->level[this_level].dirty && !no_modify));
if (cursor->level[this_level].dirty && !no_modify)
- da_bwrite(mp, cursor->level[this_level].bp);
+ libxfs_writebuf(cursor->level[this_level].bp, 0);
else
- da_brelse(cursor->level[this_level].bp);
+ libxfs_putbuf(cursor->level[this_level].bp);
cursor->level[this_level].bp = NULL;
@@ -598,7 +452,7 @@ verify_dir2_path(xfs_mount_t *mp,
xfs_da_intnode_t *node;
xfs_da_intnode_t *newnode;
xfs_dablk_t dabno;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
int bad;
int entry;
int this_level = p_level + 1;
@@ -611,7 +465,7 @@ verify_dir2_path(xfs_mount_t *mp,
* should be processed now in this level.
*/
entry = cursor->level[this_level].index;
- node = cursor->level[this_level].bp->data;
+ node = cursor->level[this_level].bp->b_addr;
/*
* if this block is out of entries, validate this
@@ -665,7 +519,7 @@ _("can't read block %u for directory inode %" PRIu64 "\n"),
return(1);
}
- newnode = bp->data;
+ newnode = bp->b_addr;
/*
* verify magic number and back pointer, sanity-check
* entry count, verify level
@@ -700,7 +554,7 @@ _("bad level %d in block %u for directory inode %" PRIu64 "\n"),
bad++;
}
if (bad) {
- da_brelse(bp);
+ libxfs_putbuf(bp);
return(1);
}
/*
@@ -711,9 +565,9 @@ _("bad level %d in block %u for directory inode %" PRIu64 "\n"),
(cursor->level[this_level].dirty && !no_modify));
if (cursor->level[this_level].dirty && !no_modify)
- da_bwrite(mp, cursor->level[this_level].bp);
+ libxfs_writebuf(cursor->level[this_level].bp, 0);
else
- da_brelse(cursor->level[this_level].bp);
+ libxfs_putbuf(cursor->level[this_level].bp);
cursor->level[this_level].bp = bp;
cursor->level[this_level].dirty = 0;
cursor->level[this_level].bno = dabno;
@@ -785,21 +639,19 @@ process_sf_dir2_fixi8(
memmove(oldsfp, newsfp, oldsize);
newsfp->hdr.count = oldsfp->hdr.count;
newsfp->hdr.i8count = 0;
- ino = xfs_dir2_sf_get_inumber(oldsfp, &oldsfp->hdr.parent);
- xfs_dir2_sf_put_inumber(newsfp, &ino, &newsfp->hdr.parent);
- oldsfep = xfs_dir2_sf_firstentry(oldsfp);
- newsfep = xfs_dir2_sf_firstentry(newsfp);
+ ino = xfs_dir2_sf_get_parent_ino(&sfp->hdr);
+ xfs_dir2_sf_put_parent_ino(&newsfp->hdr, ino);
+ oldsfep = xfs_dir2_sf_firstentry(&oldsfp->hdr);
+ newsfep = xfs_dir2_sf_firstentry(&newsfp->hdr);
while ((int)((char *)oldsfep - (char *)oldsfp) < oldsize) {
newsfep->namelen = oldsfep->namelen;
xfs_dir2_sf_put_offset(newsfep,
xfs_dir2_sf_get_offset(oldsfep));
memmove(newsfep->name, oldsfep->name, newsfep->namelen);
- ino = xfs_dir2_sf_get_inumber(oldsfp,
- xfs_dir2_sf_inumberp(oldsfep));
- xfs_dir2_sf_put_inumber(newsfp, &ino,
- xfs_dir2_sf_inumberp(newsfep));
- oldsfep = xfs_dir2_sf_nextentry(oldsfp, oldsfep);
- newsfep = xfs_dir2_sf_nextentry(newsfp, newsfep);
+ ino = xfs_dir2_sfe_get_ino(&oldsfp->hdr, oldsfep);
+ xfs_dir2_sfe_put_ino(&newsfp->hdr, newsfep, ino);
+ oldsfep = xfs_dir2_sf_nextentry(&oldsfp->hdr, oldsfep);
+ newsfep = xfs_dir2_sf_nextentry(&newsfp->hdr, newsfep);
}
*next_sfep = newsfep;
free(oldsfp);
@@ -818,13 +670,13 @@ process_sf_dir2_fixoff(
xfs_dir2_sf_t *sfp;
sfp = (xfs_dir2_sf_t *)XFS_DFORK_DPTR(dip);
- sfep = xfs_dir2_sf_firstentry(sfp);
+ sfep = xfs_dir2_sf_firstentry(&sfp->hdr);
offset = XFS_DIR2_DATA_FIRST_OFFSET;
for (i = 0; i < sfp->hdr.count; i++) {
xfs_dir2_sf_put_offset(sfep, offset);
offset += xfs_dir2_data_entsize(sfep->namelen);
- sfep = xfs_dir2_sf_nextentry(sfp, sfep);
+ sfep = xfs_dir2_sf_nextentry(&sfp->hdr, sfep);
}
}
@@ -881,13 +733,12 @@ process_sf_dir2(
/*
* Initialize i8 based on size of parent inode number.
*/
- i8 = (xfs_dir2_sf_get_inumber(sfp, &sfp->hdr.parent)
- > XFS_DIR2_MAX_SHORT_INUM);
+ i8 = (xfs_dir2_sf_get_parent_ino(&sfp->hdr) > XFS_DIR2_MAX_SHORT_INUM);
/*
* check for bad entry count
*/
- if (num_entries * xfs_dir2_sf_entsize_byname(sfp, 1) +
+ if (num_entries * xfs_dir2_sf_entsize(&sfp->hdr, 1) +
xfs_dir2_sf_hdr_size(0) > max_size || num_entries == 0)
num_entries = 0xFF;
@@ -895,7 +746,7 @@ process_sf_dir2(
* run through entries, stop at first bad entry, don't need
* to check for .. since that's encoded in its own field
*/
- sfep = next_sfep = xfs_dir2_sf_firstentry(sfp);
+ sfep = next_sfep = xfs_dir2_sf_firstentry(&sfp->hdr);
for (i = 0;
i < num_entries && ino_dir_size > (char *)next_sfep - (char *)sfp;
i++) {
@@ -903,7 +754,7 @@ process_sf_dir2(
sfep = next_sfep;
junkit = 0;
bad_sfnamelen = 0;
- lino = xfs_dir2_sf_get_inumber(sfp, xfs_dir2_sf_inumberp(sfep));
+ lino = xfs_dir2_sfe_get_ino(&sfp->hdr, sfep);
/*
* if entry points to self, junk it since only '.' or '..'
* should do that and shortform dirs don't contain either
@@ -1017,7 +868,7 @@ _("zero length entry in shortform dir %" PRIu64 ""),
break;
}
} else if ((__psint_t) sfep - (__psint_t) sfp +
- xfs_dir2_sf_entsize_byentry(sfp, sfep)
+ xfs_dir2_sf_entsize(&sfp->hdr, sfep->namelen)
> ino_dir_size) {
bad_sfnamelen = 1;
@@ -1105,8 +956,8 @@ _("entry contains offset out of order in shortform dir %" PRIu64 "\n"),
name[namelen] = '\0';
if (!no_modify) {
- tmp_elen =
- xfs_dir2_sf_entsize_byentry(sfp, sfep);
+ tmp_elen = xfs_dir2_sf_entsize(&sfp->hdr,
+ sfep->namelen);
be64_add_cpu(&dip->di_size, -tmp_elen);
ino_dir_size -= tmp_elen;
@@ -1158,11 +1009,9 @@ _("would have junked entry \"%s\" in directory inode %" PRIu64 "\n"),
*/
next_sfep = (tmp_sfep == NULL)
? (xfs_dir2_sf_entry_t *) ((__psint_t) sfep
- + ((!bad_sfnamelen)
- ? xfs_dir2_sf_entsize_byentry(sfp,
- sfep)
- : xfs_dir2_sf_entsize_byname(sfp,
- namelen)))
+ + ((!bad_sfnamelen)
+ ? xfs_dir2_sf_entsize(&sfp->hdr, sfep->namelen)
+ : xfs_dir2_sf_entsize(&sfp->hdr, namelen)))
: tmp_sfep;
}
@@ -1242,7 +1091,7 @@ _("corrected entry offsets in directory %" PRIu64 "\n"),
/*
* check parent (..) entry
*/
- *parent = xfs_dir2_sf_get_inumber(sfp, &sfp->hdr.parent);
+ *parent = xfs_dir2_sf_get_parent_ino(&sfp->hdr);
/*
* if parent entry is bogus, null it out. we'll fix it later .
@@ -1256,7 +1105,7 @@ _("bogus .. inode number (%" PRIu64 ") in directory inode %" PRIu64 ", "),
if (!no_modify) {
do_warn(_("clearing inode number\n"));
- xfs_dir2_sf_put_inumber(sfp, &zero, &sfp->hdr.parent);
+ xfs_dir2_sf_put_parent_ino(&sfp->hdr, zero);
*dino_dirty = 1;
*repair = 1;
} else {
@@ -1271,7 +1120,7 @@ _("bogus .. inode number (%" PRIu64 ") in directory inode %" PRIu64 ", "),
_("corrected root directory %" PRIu64 " .. entry, was %" PRIu64 ", now %" PRIu64 "\n"),
ino, *parent, ino);
*parent = ino;
- xfs_dir2_sf_put_inumber(sfp, parent, &sfp->hdr.parent);
+ xfs_dir2_sf_put_parent_ino(&sfp->hdr, ino);
*dino_dirty = 1;
*repair = 1;
} else {
@@ -1291,7 +1140,7 @@ _("bad .. entry in directory inode %" PRIu64 ", points to self, "),
if (!no_modify) {
do_warn(_("clearing inode number\n"));
- xfs_dir2_sf_put_inumber(sfp, &zero, &sfp->hdr.parent);
+ xfs_dir2_sf_put_parent_ino(&sfp->hdr, zero);
*dino_dirty = 1;
*repair = 1;
} else {
@@ -1314,11 +1163,12 @@ process_dir2_data(
int ino_discovery,
char *dirname, /* directory pathname */
xfs_ino_t *parent, /* out - NULLFSINO if entry not exist */
- xfs_dabuf_t *bp,
+ struct xfs_buf *bp,
int *dot, /* out - 1 if there is a dot, else 0 */
int *dotdot, /* out - 1 if there's a dotdot, else 0 */
xfs_dablk_t da_bno,
- char *endptr)
+ char *endptr,
+ int *dirty)
{
int badbest;
xfs_dir2_data_free_t *bf;
@@ -1338,7 +1188,7 @@ process_dir2_data(
char *ptr;
xfs_ino_t ent_ino;
- d = bp->data;
+ d = bp->b_addr;
bf = d->hdr.bestfree;
ptr = (char *)d->u;
badbest = lastfree = freeseen = 0;
@@ -1372,7 +1222,7 @@ process_dir2_data(
(char *)dup - (char *)d)
break;
badbest |= lastfree != 0;
- dfp = xfs_dir2_data_freefind(d, dup);
+ dfp = xfs_dir2_data_freefind(&d->hdr, dup);
if (dfp) {
i = dfp - bf;
badbest |= (freeseen & (1 << i)) != 0;
@@ -1515,7 +1365,7 @@ _("\tclearing inode number in entry at offset %" PRIdPTR "...\n"),
(intptr_t)ptr - (intptr_t)d);
dep->inumber = cpu_to_be64(BADFSINO);
ent_ino = BADFSINO;
- bp->dirty = 1;
+ *dirty = 1;
} else {
do_warn(
_("\twould clear inode number in entry at offset %" PRIdPTR "...\n"),
@@ -1541,7 +1391,7 @@ _("entry at block %u offset %" PRIdPTR " in directory inode %" PRIu64 " has ille
*/
if (!no_modify && ent_ino == BADFSINO) {
dep->name[0] = '/';
- bp->dirty = 1;
+ *dirty = 1;
junkit = 0;
}
/*
@@ -1577,7 +1427,7 @@ _("bad .. entry in root directory inode %" PRIu64 ", was %" PRIu64 ": "),
if (!no_modify) {
do_warn(_("correcting\n"));
dep->inumber = cpu_to_be64(ino);
- bp->dirty = 1;
+ *dirty = 1;
} else {
do_warn(_("would correct\n"));
}
@@ -1609,7 +1459,7 @@ _("bad . entry in directory inode %" PRIu64 ", was %" PRIu64 ": "),
if (!no_modify) {
do_warn(_("correcting\n"));
dep->inumber = cpu_to_be64(ino);
- bp->dirty = 1;
+ *dirty = 1;
} else {
do_warn(_("would correct\n"));
}
@@ -1636,7 +1486,7 @@ _("entry \"%*.*s\" in directory inode %" PRIu64 " points to self: "),
if (junkit) {
if (!no_modify) {
dep->name[0] = '/';
- bp->dirty = 1;
+ *dirty = 1;
do_warn(_("clearing entry\n"));
} else {
do_warn(_("would clear entry\n"));
@@ -1656,8 +1506,8 @@ _("bad bestfree table in block %u in directory inode %" PRIu64 ": "),
da_bno, ino);
if (!no_modify) {
do_warn(_("repairing table\n"));
- libxfs_dir2_data_freescan(mp, d, &i);
- bp->dirty = 1;
+ libxfs_dir2_data_freescan(mp, &d->hdr, &i);
+ *dirty = 1;
} else {
do_warn(_("would repair table\n"));
}
@@ -1686,11 +1536,12 @@ process_block_dir2(
xfs_dir2_block_t *block;
xfs_dir2_leaf_entry_t *blp;
bmap_ext_t *bmp;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
xfs_dir2_block_tail_t *btp;
int nex;
int rval;
bmap_ext_t lbmp;
+ int dirty = 0;
*repair = *dot = *dotdot = 0;
*parent = NULLFSINO;
@@ -1713,7 +1564,7 @@ _("can't read block %u for directory inode %" PRIu64 "\n"),
/*
* Verify the block
*/
- block = bp->data;
+ block = bp->b_addr;
if (be32_to_cpu(block->hdr.magic) != XFS_DIR2_BLOCK_MAGIC)
do_warn(
_("bad directory block magic # %#x in block %u for directory inode %" PRIu64 "\n"),
@@ -1722,7 +1573,7 @@ _("bad directory block magic # %#x in block %u for directory inode %" PRIu64 "\n
* process the data area
* this also checks & fixes the bestfree
*/
- btp = xfs_dir2_block_tail_p(mp, block);
+ btp = xfs_dir2_block_tail_p(mp, &block->hdr);
blp = xfs_dir2_block_leaf_p(btp);
/*
* Don't let this go past the end of the block.
@@ -1730,12 +1581,12 @@ _("bad directory block magic # %#x in block %u for directory inode %" PRIu64 "\n
if ((char *)blp > (char *)btp)
blp = (xfs_dir2_leaf_entry_t *)btp;
rval = process_dir2_data(mp, ino, dip, ino_discovery, dirname, parent,
- bp, dot, dotdot, mp->m_dirdatablk, (char *)blp);
- if (bp->dirty && !no_modify) {
+ bp, dot, dotdot, mp->m_dirdatablk, (char *)blp, &dirty);
+ if (dirty && !no_modify) {
*repair = 1;
- da_bwrite(mp, bp);
+ libxfs_writebuf(bp, 0);
} else
- da_brelse(bp);
+ libxfs_putbuf(bp);
return rval;
}
@@ -1794,7 +1645,7 @@ process_leaf_level_dir2(
int *repair)
{
bmap_ext_t *bmp;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
int buf_dirty;
xfs_dahash_t current_hashval;
xfs_dablk_t da_bno;
@@ -1839,7 +1690,7 @@ _("can't read file block %u for directory inode %" PRIu64 "\n"),
da_bno, ino);
goto error_out;
}
- leaf = bp->data;
+ leaf = bp->b_addr;
/*
* Check magic number for leaf directory btree block.
*/
@@ -1849,7 +1700,7 @@ _("can't read file block %u for directory inode %" PRIu64 "\n"),
_("bad directory leaf magic # %#x for directory inode %" PRIu64 " block %u\n"),
be16_to_cpu(leaf->hdr.info.magic),
ino, da_bno);
- da_brelse(bp);
+ libxfs_putbuf(bp);
goto error_out;
}
buf_dirty = 0;
@@ -1859,7 +1710,7 @@ _("bad directory leaf magic # %#x for directory inode %" PRIu64 " block %u\n"),
*/
if (process_leaf_block_dir2(mp, leaf, da_bno, ino,
current_hashval, &greatest_hashval)) {
- da_brelse(bp);
+ libxfs_putbuf(bp);
goto error_out;
}
/*
@@ -1878,14 +1729,14 @@ _("bad directory leaf magic # %#x for directory inode %" PRIu64 " block %u\n"),
do_warn(
_("bad sibling back pointer for block %u in directory inode %" PRIu64 "\n"),
da_bno, ino);
- da_brelse(bp);
+ libxfs_putbuf(bp);
goto error_out;
}
prev_bno = da_bno;
da_bno = be32_to_cpu(leaf->hdr.info.forw);
if (da_bno != 0) {
if (verify_dir2_path(mp, da_cursor, 0)) {
- da_brelse(bp);
+ libxfs_putbuf(bp);
goto error_out;
}
}
@@ -1893,9 +1744,9 @@ _("bad sibling back pointer for block %u in directory inode %" PRIu64 "\n"),
ASSERT(buf_dirty == 0 || (buf_dirty && !no_modify));
if (buf_dirty && !no_modify) {
*repair = 1;
- da_bwrite(mp, bp);
+ libxfs_writebuf(bp, 0);
} else
- da_brelse(bp);
+ libxfs_putbuf(bp);
} while (da_bno != 0);
if (verify_final_dir2_path(mp, da_cursor, 0)) {
/*
@@ -1988,7 +1839,7 @@ process_leaf_node_dir2(
int isnode) /* node directory not leaf */
{
bmap_ext_t *bmp;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
xfs_dir2_data_t *data;
xfs_dfiloff_t dbno;
int good;
@@ -1997,6 +1848,7 @@ process_leaf_node_dir2(
int nex;
int t;
bmap_ext_t lbmp;
+ int dirty = 0;
*repair = *dot = *dotdot = good = 0;
*parent = NULLFSINO;
@@ -2023,21 +1875,21 @@ _("can't read block %" PRIu64 " for directory inode %" PRIu64 "\n"),
dbno, ino);
continue;
}
- data = bp->data;
+ data = bp->b_addr;
if (be32_to_cpu(data->hdr.magic) != XFS_DIR2_DATA_MAGIC)
do_warn(
_("bad directory block magic # %#x in block %" PRIu64 " for directory inode %" PRIu64 "\n"),
be32_to_cpu(data->hdr.magic), dbno, ino);
i = process_dir2_data(mp, ino, dip, ino_discovery, dirname,
parent, bp, dot, dotdot, (xfs_dablk_t)dbno,
- (char *)data + mp->m_dirblksize);
+ (char *)data + mp->m_dirblksize, &dirty);
if (i == 0)
good++;
- if (bp->dirty && !no_modify) {
+ if (dirty && !no_modify) {
*repair = 1;
- da_bwrite(mp, bp);
+ libxfs_writebuf(bp, 0);
} else
- da_brelse(bp);
+ libxfs_putbuf(bp);
}
if (good == 0)
return 1;
diff --git a/repair/dir2.h b/repair/dir2.h
index a88579f..63e35d4 100644
--- a/repair/dir2.h
+++ b/repair/dir2.h
@@ -23,6 +23,32 @@ struct blkmap;
struct bmap_ext;
/*
+ * generic dir2 structures used by xfs_repair.
+ * XXX: shared with xfsdb
+ */
+typedef union {
+ xfs_dir2_data_entry_t entry;
+ xfs_dir2_data_unused_t unused;
+} xfs_dir2_data_union_t;
+
+typedef struct xfs_dir2_data {
+ xfs_dir2_data_hdr_t hdr; /* magic XFS_DIR2_DATA_MAGIC */
+ xfs_dir2_data_union_t u[1];
+} xfs_dir2_data_t;
+
+typedef struct xfs_dir2_block {
+ xfs_dir2_data_hdr_t hdr; /* magic XFS_DIR2_BLOCK_MAGIC */
+ xfs_dir2_data_union_t u[1];
+ xfs_dir2_leaf_entry_t leaf[1];
+ xfs_dir2_block_tail_t tail;
+} xfs_dir2_block_t;
+
+typedef struct xfs_dir2_sf {
+ xfs_dir2_sf_hdr_t hdr; /* shortform header */
+ xfs_dir2_sf_entry_t list[1]; /* shortform entries */
+} xfs_dir2_sf_t;
+
+/*
* the cursor gets passed up and down the da btree processing
* routines. The interior block processing routines use the
* cursor to determine if the pointers to and from the preceding
@@ -42,7 +68,7 @@ struct bmap_ext;
* Currently, we just trash it.
*/
typedef struct dir2_level_state {
- xfs_dabuf_t *bp; /* block bp */
+ xfs_buf_t *bp; /* block bp */
xfs_dablk_t bno; /* file block number */
xfs_dahash_t hashval; /* last verified hashval */
int index; /* current index in block */
diff --git a/repair/phase6.c b/repair/phase6.c
index cbe0b35..00f70a2 100644
--- a/repair/phase6.c
+++ b/repair/phase6.c
@@ -17,6 +17,7 @@
*/
#include <libxfs.h>
+#include <xfs_dir1.h>
#include "avl.h"
#include "globals.h"
#include "agheader.h"
@@ -484,9 +485,9 @@ mk_rbmino(xfs_mount_t *mp)
xfs_bmap_init(&flist, &first);
while (bno < mp->m_sb.sb_rbmblocks) {
nmap = XFS_BMAP_MAX_NMAP;
- error = libxfs_bmapi(tp, ip, bno,
+ error = libxfs_bmapi_write(tp, ip, bno,
(xfs_extlen_t)(mp->m_sb.sb_rbmblocks - bno),
- XFS_BMAPI_WRITE, &first, mp->m_sb.sb_rbmblocks,
+ 0, &first, mp->m_sb.sb_rbmblocks,
map, &nmap, &flist);
if (error) {
do_error(
@@ -542,7 +543,7 @@ fill_rbmino(xfs_mount_t *mp)
* fill the file one block at a time
*/
nmap = 1;
- error = libxfs_bmapi(tp, ip, bno, 1, XFS_BMAPI_WRITE,
+ error = libxfs_bmapi_write(tp, ip, bno, 1, 0,
&first, 1, &map, &nmap, NULL);
if (error || nmap != 1) {
do_error(
@@ -555,7 +556,7 @@ fill_rbmino(xfs_mount_t *mp)
error = libxfs_trans_read_buf(
mp, tp, mp->m_dev,
XFS_FSB_TO_DADDR(mp, map.br_startblock),
- XFS_FSB_TO_BB(mp, 1), 1, &bp);
+ XFS_FSB_TO_BB(mp, 1), 1, &bp, NULL);
if (error) {
do_warn(
@@ -611,7 +612,7 @@ fill_rsumino(xfs_mount_t *mp)
* fill the file one block at a time
*/
nmap = 1;
- error = libxfs_bmapi(tp, ip, bno, 1, XFS_BMAPI_WRITE,
+ error = libxfs_bmapi_write(tp, ip, bno, 1, 0,
&first, 1, &map, &nmap, NULL);
if (error || nmap != 1) {
do_error(
@@ -624,7 +625,7 @@ fill_rsumino(xfs_mount_t *mp)
error = libxfs_trans_read_buf(
mp, tp, mp->m_dev,
XFS_FSB_TO_DADDR(mp, map.br_startblock),
- XFS_FSB_TO_BB(mp, 1), 1, &bp);
+ XFS_FSB_TO_BB(mp, 1), 1, &bp, NULL);
if (error) {
do_warn(
@@ -723,10 +724,9 @@ mk_rsumino(xfs_mount_t *mp)
xfs_bmap_init(&flist, &first);
while (bno < nsumblocks) {
nmap = XFS_BMAP_MAX_NMAP;
- error = libxfs_bmapi(tp, ip, bno,
+ error = libxfs_bmapi_write(tp, ip, bno,
(xfs_extlen_t)(nsumblocks - bno),
- XFS_BMAPI_WRITE, &first, nsumblocks,
- map, &nmap, &flist);
+ 0, &first, nsumblocks, map, &nmap, &flist);
if (error) {
do_error(
_("couldn't allocate realtime summary inode, error = %d\n"),
@@ -1139,7 +1139,6 @@ map_first_dblock_fsbno(xfs_mount_t *mp,
xfs_inode_t *ip,
xfs_dablk_t *bno)
{
- xfs_fsblock_t fblock;
xfs_da_intnode_t *node;
xfs_buf_t *bp;
xfs_dablk_t da_bno;
@@ -1159,13 +1158,11 @@ map_first_dblock_fsbno(xfs_mount_t *mp,
*bno = 0;
i = -1;
node = NULL;
- fblock = NULLFSBLOCK;
ftype = _("dir");
nmap = 1;
- error = libxfs_bmapi(NULL, ip, (xfs_fileoff_t) da_bno, 1,
- XFS_BMAPI_METADATA, &fblock, 0,
- &map, &nmap, NULL);
+ error = libxfs_bmapi_read(ip, (xfs_fileoff_t) da_bno, 1,
+ &map, &nmap, 0);
if (error || nmap != 1) {
if (!no_modify)
do_error(
@@ -1235,9 +1232,8 @@ _("bad dir/attr magic number in inode %" PRIu64 ", file bno = %u, fsbno = %" PRI
bp = NULL;
nmap = 1;
- error = libxfs_bmapi(NULL, ip, (xfs_fileoff_t) da_bno, 1,
- XFS_BMAPI_METADATA, &fblock, 0,
- &map, &nmap, NULL);
+ error = libxfs_bmapi_read(ip, (xfs_fileoff_t) da_bno, 1,
+ &map, &nmap, 0);
if (error || nmap != 1) {
if (!no_modify)
do_error(
@@ -1549,7 +1545,6 @@ longform_dir_entry_check(xfs_mount_t *mp,
xfs_dir_leafblock_t *leaf;
xfs_buf_t *bp;
xfs_dfsbno_t fsbno;
- xfs_fsblock_t fblock;
xfs_dablk_t da_bno;
int dirty;
int nmap;
@@ -1559,7 +1554,6 @@ longform_dir_entry_check(xfs_mount_t *mp,
char *ftype;
da_bno = 0;
- fblock = NULLFSBLOCK;
*need_dot = 1;
ftype = _("dir");
@@ -1621,9 +1615,8 @@ _("bad magic # (0x%x) for dir ino %" PRIu64 " leaf block (bno %u fsbno %" PRIu64
if (da_bno != 0) {
nmap = 1;
- error = libxfs_bmapi(NULL, ip, (xfs_fileoff_t)da_bno, 1,
- XFS_BMAPI_METADATA, &fblock, 0,
- &map, &nmap, NULL);
+ error = libxfs_bmapi_read(ip, (xfs_fileoff_t)da_bno, 1,
+ &map, &nmap, 0);
if (error || nmap != 1) {
if (!no_modify)
do_error(
@@ -1790,7 +1783,7 @@ dir2_kill_block(
xfs_mount_t *mp,
xfs_inode_t *ip,
xfs_dablk_t da_bno,
- xfs_dabuf_t *bp)
+ struct xfs_buf *bp)
{
xfs_da_args_t args;
int committed;
@@ -1808,7 +1801,7 @@ dir2_kill_block(
res_failed(error);
libxfs_trans_ijoin(tp, ip, 0);
libxfs_trans_ihold(tp, ip);
- libxfs_da_bjoin(tp, bp);
+ libxfs_trans_bjoin(tp, bp);
memset(&args, 0, sizeof(args));
xfs_bmap_init(&flist, &firstblock);
args.dp = ip;
@@ -1840,7 +1833,7 @@ longform_dir2_entry_check_data(
int *need_dot,
ino_tree_node_t *current_irec,
int current_ino_offset,
- xfs_dabuf_t **bpp,
+ struct xfs_buf **bpp,
dir_hash_tab_t *hashtab,
freetab_t **freetabp,
xfs_dablk_t da_bno,
@@ -1848,7 +1841,7 @@ longform_dir2_entry_check_data(
{
xfs_dir2_dataptr_t addr;
xfs_dir2_leaf_entry_t *blp;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
xfs_dir2_block_tail_t *btp;
int committed;
xfs_dir2_data_t *d;
@@ -1877,14 +1870,14 @@ longform_dir2_entry_check_data(
int wantmagic;
bp = *bpp;
- d = bp->data;
+ d = bp->b_addr;
ptr = (char *)d->u;
nbad = 0;
needscan = needlog = 0;
junkit = 0;
freetab = *freetabp;
if (isblock) {
- btp = xfs_dir2_block_tail_p(mp, (xfs_dir2_block_t *)d);
+ btp = xfs_dir2_block_tail_p(mp, (struct xfs_dir2_data_hdr *)d);
blp = xfs_dir2_block_leaf_p(btp);
endptr = (char *)blp;
if (endptr > (char *)btp)
@@ -1972,7 +1965,7 @@ longform_dir2_entry_check_data(
dir2_kill_block(mp, ip, da_bno, bp);
} else {
do_warn(_("would junk block\n"));
- libxfs_da_brelse(NULL, bp);
+ libxfs_putbuf(bp);
}
freetab->ents[db].v = NULLDATAOFF;
*bpp = NULL;
@@ -1990,8 +1983,8 @@ longform_dir2_entry_check_data(
res_failed(error);
libxfs_trans_ijoin(tp, ip, 0);
libxfs_trans_ihold(tp, ip);
- libxfs_da_bjoin(tp, bp);
- libxfs_da_bhold(tp, bp);
+ libxfs_trans_bjoin(tp, bp);
+ libxfs_trans_bhold(tp, bp);
xfs_bmap_init(&flist, &firstblock);
if (be32_to_cpu(d->hdr.magic) != wantmagic) {
do_warn(
@@ -2256,7 +2249,7 @@ _("entry \"%s\" in dir inode %" PRIu64 " inconsistent with .. value (%" PRIu64 "
}
*num_illegal += nbad;
if (needscan)
- libxfs_dir2_data_freescan(mp, d, &needlog);
+ libxfs_dir2_data_freescan(mp, &d->hdr, &needlog);
if (needlog)
libxfs_dir2_data_log_header(tp, bp);
libxfs_bmap_finish(&tp, &flist, &committed);
@@ -2277,7 +2270,7 @@ longform_dir2_check_leaf(
{
int badtail;
__be16 *bestsp;
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
xfs_dablk_t da_bno;
int i;
xfs_dir2_leaf_t *leaf;
@@ -2285,13 +2278,13 @@ longform_dir2_check_leaf(
int seeval;
da_bno = mp->m_dirleafblk;
- if (libxfs_da_read_bufr(NULL, ip, da_bno, -1, &bp, XFS_DATA_FORK)) {
+ if (libxfs_da_read_buf(NULL, ip, da_bno, -1, &bp, XFS_DATA_FORK, NULL)) {
do_error(
_("can't read block %u for directory inode %" PRIu64 "\n"),
da_bno, ip->i_ino);
/* NOTREACHED */
}
- leaf = bp->data;
+ leaf = bp->b_addr;
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
bestsp = xfs_dir2_leaf_bests_p(ltp);
if (be16_to_cpu(leaf->hdr.info.magic) != XFS_DIR2_LEAF1_MAGIC ||
@@ -2299,21 +2292,21 @@ longform_dir2_check_leaf(
be32_to_cpu(leaf->hdr.info.back) ||
be16_to_cpu(leaf->hdr.count) <
be16_to_cpu(leaf->hdr.stale) ||
- be16_to_cpu(leaf->hdr.count) >
+ be16_to_cpu(leaf->hdr.count) >
xfs_dir2_max_leaf_ents(mp) ||
- (char *)&leaf->ents[be16_to_cpu(
+ (char *)&leaf->ents[be16_to_cpu(
leaf->hdr.count)] > (char *)bestsp) {
do_warn(
_("leaf block %u for directory inode %" PRIu64 " bad header\n"),
da_bno, ip->i_ino);
- libxfs_da_brelse(NULL, bp);
+ libxfs_putbuf(bp);
return 1;
}
seeval = dir_hash_see_all(hashtab, leaf->ents,
be16_to_cpu(leaf->hdr.count),
be16_to_cpu(leaf->hdr.stale));
if (dir_hash_check(hashtab, ip, seeval)) {
- libxfs_da_brelse(NULL, bp);
+ libxfs_putbuf(bp);
return 1;
}
badtail = freetab->nents != be32_to_cpu(ltp->bestcount);
@@ -2325,10 +2318,10 @@ longform_dir2_check_leaf(
do_warn(
_("leaf block %u for directory inode %" PRIu64 " bad tail\n"),
da_bno, ip->i_ino);
- libxfs_da_brelse(NULL, bp);
+ libxfs_putbuf(bp);
return 1;
}
- libxfs_da_brelse(NULL, bp);
+ libxfs_putbuf(bp);
return 0;
}
@@ -2343,7 +2336,7 @@ longform_dir2_check_node(
dir_hash_tab_t *hashtab,
freetab_t *freetab)
{
- xfs_dabuf_t *bp;
+ struct xfs_buf *bp;
xfs_dablk_t da_bno;
xfs_dir2_db_t fdb;
xfs_dir2_free_t *free;
@@ -2359,25 +2352,25 @@ longform_dir2_check_node(
next_da_bno = da_bno + mp->m_dirblkfsbs - 1;
if (bmap_next_offset(NULL, ip, &next_da_bno, XFS_DATA_FORK))
break;
- if (libxfs_da_read_bufr(NULL, ip, da_bno, -1, &bp,
- XFS_DATA_FORK)) {
+ if (libxfs_da_read_buf(NULL, ip, da_bno, -1, &bp,
+ XFS_DATA_FORK, NULL)) {
do_warn(
_("can't read leaf block %u for directory inode %" PRIu64 "\n"),
da_bno, ip->i_ino);
return 1;
}
- leaf = bp->data;
+ leaf = bp->b_addr;
if (be16_to_cpu(leaf->hdr.info.magic) != XFS_DIR2_LEAFN_MAGIC) {
if (be16_to_cpu(leaf->hdr.info.magic) ==
XFS_DA_NODE_MAGIC) {
- libxfs_da_brelse(NULL, bp);
+ libxfs_putbuf(bp);
continue;
}
do_warn(
_("unknown magic number %#x for block %u in directory inode %" PRIu64 "\n"),
be16_to_cpu(leaf->hdr.info.magic),
da_bno, ip->i_ino);
- libxfs_da_brelse(NULL, bp);
+ libxfs_putbuf(bp);
return 1;
}
if (be16_to_cpu(leaf->hdr.count) > xfs_dir2_max_leaf_ents(mp) ||
@@ -2386,13 +2379,13 @@ longform_dir2_check_node(
do_warn(
_("leaf block %u for directory inode %" PRIu64 " bad header\n"),
da_bno, ip->i_ino);
- libxfs_da_brelse(NULL, bp);
+ libxfs_putbuf(bp);
return 1;
}
seeval = dir_hash_see_all(hashtab, leaf->ents,
be16_to_cpu(leaf->hdr.count),
be16_to_cpu(leaf->hdr.stale));
- libxfs_da_brelse(NULL, bp);
+ libxfs_putbuf(bp);
if (seeval != DIR_HASH_CK_OK)
return 1;
}
@@ -2405,25 +2398,25 @@ longform_dir2_check_node(
next_da_bno = da_bno + mp->m_dirblkfsbs - 1;
if (bmap_next_offset(NULL, ip, &next_da_bno, XFS_DATA_FORK))
break;
- if (libxfs_da_read_bufr(NULL, ip, da_bno, -1, &bp,
- XFS_DATA_FORK)) {
+ if (libxfs_da_read_buf(NULL, ip, da_bno, -1, &bp,
+ XFS_DATA_FORK, NULL)) {
do_warn(
_("can't read freespace block %u for directory inode %" PRIu64 "\n"),
da_bno, ip->i_ino);
return 1;
}
- free = bp->data;
+ free = bp->b_addr;
fdb = xfs_dir2_da_to_db(mp, da_bno);
if (be32_to_cpu(free->hdr.magic) != XFS_DIR2_FREE_MAGIC ||
be32_to_cpu(free->hdr.firstdb) !=
(fdb - XFS_DIR2_FREE_FIRSTDB(mp)) *
- XFS_DIR2_MAX_FREE_BESTS(mp) ||
+ xfs_dir2_free_max_bests(mp) ||
be32_to_cpu(free->hdr.nvalid) <
be32_to_cpu(free->hdr.nused)) {
do_warn(
_("free block %u for directory inode %" PRIu64 " bad header\n"),
da_bno, ip->i_ino);
- libxfs_da_brelse(NULL, bp);
+ libxfs_putbuf(bp);
return 1;
}
for (i = used = 0; i < be32_to_cpu(free->hdr.nvalid); i++) {
@@ -2431,11 +2424,11 @@ longform_dir2_check_node(
freetab->nents ||
freetab->ents[i + be32_to_cpu(
free->hdr.firstdb)].v !=
- be16_to_cpu(free->bests[i])) {
+ be16_to_cpu(free->bests[i])) {
do_warn(
_("free block %u entry %i for directory ino %" PRIu64 " bad\n"),
da_bno, i, ip->i_ino);
- libxfs_da_brelse(NULL, bp);
+ libxfs_putbuf(bp);
return 1;
}
used += be16_to_cpu(free->bests[i]) != NULLDATAOFF;
@@ -2445,10 +2438,10 @@ longform_dir2_check_node(
do_warn(
_("free block %u for directory inode %" PRIu64 " bad nused\n"),
da_bno, ip->i_ino);
- libxfs_da_brelse(NULL, bp);
+ libxfs_putbuf(bp);
return 1;
}
- libxfs_da_brelse(NULL, bp);
+ libxfs_putbuf(bp);
}
for (i = 0; i < freetab->nents; i++) {
if ((freetab->ents[i].s == 0) &&
@@ -2478,7 +2471,7 @@ longform_dir2_entry_check(xfs_mount_t *mp,
dir_hash_tab_t *hashtab)
{
xfs_dir2_block_t *block;
- xfs_dabuf_t **bplist;
+ struct xfs_buf **bplist;
xfs_dablk_t da_bno;
freetab_t *freetab;
int num_bps;
@@ -2505,7 +2498,7 @@ longform_dir2_entry_check(xfs_mount_t *mp,
freetab->ents[i].s = 0;
}
num_bps = freetab->naents;
- bplist = calloc(num_bps, sizeof(xfs_dabuf_t*));
+ bplist = calloc(num_bps, sizeof(struct xfs_buf*));
/* is this a block, leaf, or node directory? */
libxfs_dir2_isblock(NULL, ip, &isblock);
libxfs_dir2_isleaf(NULL, ip, &isleaf);
@@ -2521,14 +2514,14 @@ longform_dir2_entry_check(xfs_mount_t *mp,
if (db >= num_bps) {
/* more data blocks than expected */
num_bps = db + 1;
- bplist = realloc(bplist, num_bps * sizeof(xfs_dabuf_t*));
+ bplist = realloc(bplist, num_bps * sizeof(struct xfs_buf*));
if (!bplist)
do_error(
_("realloc failed in longform_dir2_entry_check (%zu bytes)\n"),
- num_bps * sizeof(xfs_dabuf_t*));
+ num_bps * sizeof(struct xfs_buf*));
}
- if (libxfs_da_read_bufr(NULL, ip, da_bno, -1, &bplist[db],
- XFS_DATA_FORK)) {
+ if (libxfs_da_read_buf(NULL, ip, da_bno, -1, &bplist[db],
+ XFS_DATA_FORK, NULL)) {
do_warn(
_("can't read data block %u for directory inode %" PRIu64 "\n"),
da_bno, ino);
@@ -2547,8 +2540,8 @@ longform_dir2_entry_check(xfs_mount_t *mp,
xfs_dir2_block_tail_t *btp;
xfs_dir2_leaf_entry_t *blp;
- block = bplist[0]->data;
- btp = xfs_dir2_block_tail_p(mp, block);
+ block = bplist[0]->b_addr;
+ btp = xfs_dir2_block_tail_p(mp, &block->hdr);
blp = xfs_dir2_block_leaf_p(btp);
seeval = dir_hash_see_all(hashtab, blp,
be32_to_cpu(btp->count),
@@ -2567,14 +2560,14 @@ longform_dir2_entry_check(xfs_mount_t *mp,
dir_hash_dup_names(hashtab);
for (i = 0; i < freetab->naents; i++)
if (bplist[i])
- libxfs_da_brelse(NULL, bplist[i]);
+ libxfs_putbuf(bplist[i]);
longform_dir2_rebuild(mp, ino, ip, irec, ino_offset, hashtab);
*num_illegal = 0;
*need_dot = 0;
} else {
for (i = 0; i < freetab->naents; i++)
if (bplist[i])
- libxfs_da_brelse(NULL, bplist[i]);
+ libxfs_putbuf(bplist[i]);
}
free(bplist);
@@ -2931,7 +2924,7 @@ shortform_dir2_entry_check(xfs_mount_t *mp,
do_warn(
_("setting .. in sf dir inode %" PRIu64 " to %" PRIu64 "\n"),
ino, parent);
- xfs_dir2_sf_put_inumber(sfp, &parent, &sfp->hdr.parent);
+ xfs_dir2_sf_put_parent_ino(&sfp->hdr, parent);
*ino_dirty = 1;
}
return;
@@ -2948,15 +2941,14 @@ shortform_dir2_entry_check(xfs_mount_t *mp,
/*
* Initialise i8 counter -- the parent inode number counts as well.
*/
- i8 = (xfs_dir2_sf_get_inumber(sfp, &sfp->hdr.parent) >
- XFS_DIR2_MAX_SHORT_INUM);
+ i8 = xfs_dir2_sf_get_parent_ino(&sfp->hdr) > XFS_DIR2_MAX_SHORT_INUM;
/*
* now run through entries, stop at first bad entry, don't need
* to skip over '..' since that's encoded in its own field and
* no need to worry about '.' since it doesn't exist.
*/
- sfep = next_sfep = xfs_dir2_sf_firstentry(sfp);
+ sfep = next_sfep = xfs_dir2_sf_firstentry(&sfp->hdr);
for (i = 0; i < sfp->hdr.count && max_size >
(__psint_t)next_sfep - (__psint_t)sfp;
@@ -2965,7 +2957,7 @@ shortform_dir2_entry_check(xfs_mount_t *mp,
bad_sfnamelen = 0;
tmp_sfep = NULL;
- lino = xfs_dir2_sf_get_inumber(sfp, xfs_dir2_sf_inumberp(sfep));
+ lino = xfs_dir2_sfe_get_ino(&sfp->hdr, sfep);
namelen = sfep->namelen;
@@ -2994,7 +2986,7 @@ shortform_dir2_entry_check(xfs_mount_t *mp,
break;
}
} else if (no_modify && (__psint_t) sfep - (__psint_t) sfp +
- + xfs_dir2_sf_entsize_byentry(sfp, sfep)
+ + xfs_dir2_sf_entsize(&sfp->hdr, sfep->namelen)
> ip->i_d.di_size) {
bad_sfnamelen = 1;
@@ -3024,7 +3016,7 @@ shortform_dir2_entry_check(xfs_mount_t *mp,
if (no_modify && verify_inum(mp, lino)) {
next_sfep = (xfs_dir2_sf_entry_t *)((__psint_t)sfep +
- xfs_dir2_sf_entsize_byentry(sfp, sfep));
+ xfs_dir2_sf_entsize(&sfp->hdr, sfep->namelen));
continue;
}
@@ -3075,8 +3067,8 @@ shortform_dir2_entry_check(xfs_mount_t *mp,
* check for duplicate names in directory.
*/
if (!dir_hash_add(mp, hashtab, (xfs_dir2_dataptr_t)
- (sfep - xfs_dir2_sf_firstentry(sfp)),
- lino, sfep->namelen, sfep->name)) {
+ (sfep - xfs_dir2_sf_firstentry(&sfp->hdr)),
+ lino, sfep->namelen, sfep->name)) {
do_warn(
_("entry \"%s\" (ino %" PRIu64 ") in dir %" PRIu64 " is a duplicate name"),
fname, lino, ino);
@@ -3132,7 +3124,8 @@ do_junkit:
if (lino == orphanage_ino)
orphanage_ino = 0;
if (!no_modify) {
- tmp_elen = xfs_dir2_sf_entsize_byentry(sfp, sfep);
+ tmp_elen = xfs_dir2_sf_entsize(&sfp->hdr,
+ sfep->namelen);
tmp_sfep = (xfs_dir2_sf_entry_t *)
((__psint_t) sfep + tmp_elen);
tmp_len = max_size - ((__psint_t) tmp_sfep
@@ -3183,9 +3176,9 @@ do_junkit:
next_sfep = (tmp_sfep == NULL)
? (xfs_dir2_sf_entry_t *) ((__psint_t) sfep
- + ((!bad_sfnamelen)
- ? xfs_dir2_sf_entsize_byentry(sfp, sfep)
- : xfs_dir2_sf_entsize_byname(sfp, namelen)))
+ + ((!bad_sfnamelen)
+ ? xfs_dir2_sf_entsize(&sfp->hdr, sfep->namelen)
+ : xfs_dir2_sf_entsize(&sfp->hdr, namelen)))
: tmp_sfep;
}
xfsprogs-crc-patchset-v2/xfsprogs-kernel-3.9-rc1-dev-update 0000600 0001750 0001750 00001107275 12132011546 022356 0 ustar dave dave xfsprogs: update libxfs to 3.9-rc1 + xfsdev
From: Dave Chinner <dchinner@xxxxxxxxxx>
Update the code from 3.8-rc2 to 3.9-rc1 and the current xfs-dev tree
(to pick up xfs_bmap.c rework).
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
db/check.c | 3 +-
db/metadump.c | 3 +-
include/libxfs.h | 2 +-
include/xfs_dinode.h | 6 +-
include/xfs_inode.h | 6 +-
include/xfs_mount.h | 7 +-
include/xfs_types.h | 1 -
libxfs/xfs.h | 4 +-
libxfs/xfs_alloc.c | 2 +-
libxfs/xfs_attr.c | 1 +
libxfs/xfs_attr_leaf.c | 6 +-
libxfs/xfs_bmap.c | 8388 +++++++++++++++++++++++------------------------
libxfs/xfs_dir2_block.c | 6 +-
libxfs/xfs_ialloc.c | 6 +-
libxfs/xfs_inode.c | 1 +
libxfs/xfs_mount.c | 2 +-
repair/dinode.c | 8 +-
repair/prefetch.c | 2 +-
18 files changed, 4218 insertions(+), 4236 deletions(-)
diff --git a/db/check.c b/db/check.c
index f1cec8f..1dfef74 100644
--- a/db/check.c
+++ b/db/check.c
@@ -2715,7 +2715,8 @@ process_inode(
error++;
return;
}
- if ((unsigned int)XFS_DFORK_ASIZE(dip, mp) >= XFS_LITINO(mp)) {
+ if ((unsigned int)XFS_DFORK_ASIZE(dip, mp) >=
+ XFS_LITINO(mp, idic.di_version)) {
if (v)
dbprintf(_("bad fork offset %d for inode %lld\n"),
idic.di_forkoff, id->ino);
diff --git a/db/metadump.c b/db/metadump.c
index 5739f86..0635e7b 100644
--- a/db/metadump.c
+++ b/db/metadump.c
@@ -1637,7 +1637,8 @@ process_inode(
nametable_clear();
/* copy extended attributes if they exist and forkoff is valid */
- if (success && XFS_DFORK_DSIZE(dip, mp) < XFS_LITINO(mp)) {
+ if (success &&
+ XFS_DFORK_DSIZE(dip, mp) < XFS_LITINO(mp, dip->di_version)) {
attr_data.remote_val_count = 0;
switch (dip->di_aformat) {
case XFS_DINODE_FMT_LOCAL:
diff --git a/include/libxfs.h b/include/libxfs.h
index 0e2fc5f..b60bcf5 100644
--- a/include/libxfs.h
+++ b/include/libxfs.h
@@ -610,7 +610,7 @@ int libxfs_attr_remove(struct xfs_inode *, const unsigned char *, int);
/* xfs_bmap.c */
xfs_bmbt_rec_host_t *xfs_bmap_search_extents(xfs_inode_t *, xfs_fileoff_t,
int, int *, xfs_extnum_t *, xfs_bmbt_irec_t *,
- xfs_bmbt_irec_t *);
+ xfs_bmbt_irec_t *);
void xfs_bmbt_disk_get_all(xfs_bmbt_rec_t *r, xfs_bmbt_irec_t *s);
/* xfs_attr_leaf.h */
diff --git a/include/xfs_dinode.h b/include/xfs_dinode.h
index 1d9643b..88a3368 100644
--- a/include/xfs_dinode.h
+++ b/include/xfs_dinode.h
@@ -104,7 +104,7 @@ typedef enum xfs_dinode_fmt {
/*
* Inode size for given fs.
*/
-#define XFS_LITINO(mp) \
+#define XFS_LITINO(mp, version) \
((int)(((mp)->m_sb.sb_inodesize) - sizeof(struct xfs_dinode)))
#define XFS_BROOT_SIZE_ADJ \
@@ -119,10 +119,10 @@ typedef enum xfs_dinode_fmt {
#define XFS_DFORK_DSIZE(dip,mp) \
(XFS_DFORK_Q(dip) ? \
XFS_DFORK_BOFF(dip) : \
- XFS_LITINO(mp))
+ XFS_LITINO(mp, (dip)->di_version))
#define XFS_DFORK_ASIZE(dip,mp) \
(XFS_DFORK_Q(dip) ? \
- XFS_LITINO(mp) - XFS_DFORK_BOFF(dip) : \
+ XFS_LITINO(mp, (dip)->di_version) - XFS_DFORK_BOFF(dip) : \
0)
#define XFS_DFORK_SIZE(dip,mp,w) \
((w) == XFS_DATA_FORK ? \
diff --git a/include/xfs_inode.h b/include/xfs_inode.h
index 437b3af..4733f85 100644
--- a/include/xfs_inode.h
+++ b/include/xfs_inode.h
@@ -180,10 +180,11 @@ typedef struct xfs_icdinode {
#define XFS_IFORK_DSIZE(ip) \
(XFS_IFORK_Q(ip) ? \
XFS_IFORK_BOFF(ip) : \
- XFS_LITINO((ip)->i_mount))
+ XFS_LITINO((ip)->i_mount, (ip)->i_d.di_version))
#define XFS_IFORK_ASIZE(ip) \
(XFS_IFORK_Q(ip) ? \
- XFS_LITINO((ip)->i_mount) - XFS_IFORK_BOFF(ip) : \
+ XFS_LITINO((ip)->i_mount, (ip)->i_d.di_version) - \
+ XFS_IFORK_BOFF(ip) : \
0)
#define XFS_IFORK_SIZE(ip,w) \
((w) == XFS_DATA_FORK ? \
@@ -419,6 +420,7 @@ static inline void xfs_iflock(struct xfs_inode *ip)
static inline void xfs_ifunlock(struct xfs_inode *ip)
{
xfs_iflags_clear(ip, XFS_IFLOCK);
+ smp_mb();
wake_up_bit(&ip->i_flags, __XFS_IFLOCK_BIT);
}
diff --git a/include/xfs_mount.h b/include/xfs_mount.h
index c267379..28bbf46 100644
--- a/include/xfs_mount.h
+++ b/include/xfs_mount.h
@@ -41,6 +41,12 @@ typedef struct xfs_trans_reservations {
uint tr_growrtalloc; /* grow realtime allocations */
uint tr_growrtzero; /* grow realtime zeroing */
uint tr_growrtfree; /* grow realtime freeing */
+ uint tr_qm_sbchange; /* change quota flags */
+ uint tr_qm_setqlim; /* adjust quota limits */
+ uint tr_qm_dqalloc; /* allocate quota on disk */
+ uint tr_qm_quotaoff; /* turn quota off */
+ uint tr_qm_equotaoff;/* end of turn quota off */
+ uint tr_sb; /* modify superblock */
} xfs_trans_reservations_t;
#ifndef __KERNEL__
@@ -201,7 +207,6 @@ typedef struct xfs_mount {
trimming */
__int64_t m_update_flags; /* sb flags we need to update
on the next remount,rw */
- struct shrinker m_inode_shrink; /* inode reclaim shrinker */
int64_t m_low_space[XFS_LOWSP_MAX];
/* low free space thresholds */
diff --git a/include/xfs_types.h b/include/xfs_types.h
index e9bd5c3..dd6bf71 100644
--- a/include/xfs_types.h
+++ b/include/xfs_types.h
@@ -32,7 +32,6 @@ typedef unsigned int __uint32_t;
typedef signed long long int __int64_t;
typedef unsigned long long int __uint64_t;
-typedef enum { B_FALSE,B_TRUE } boolean_t;
typedef __uint32_t inst_t; /* an instruction */
typedef __s64 xfs_off_t; /* <file offset> type */
diff --git a/libxfs/xfs.h b/libxfs/xfs.h
index 6213284..9fbe261 100644
--- a/libxfs/xfs.h
+++ b/libxfs/xfs.h
@@ -107,10 +107,10 @@ typedef __uint32_t inst_t; /* an instruction */
#define rcu_read_unlock() ((void) 0)
/*
- * random32 is used for di_gen inode allocation, it must be zero for libxfs
+ * prandom_u32 is used for di_gen inode allocation, it must be zero for libxfs
* or all sorts of badness can occur!
*/
-#define random32() 0
+#define prandom_u32() 0
#define PAGE_CACHE_SIZE getpagesize()
diff --git a/libxfs/xfs_alloc.c b/libxfs/xfs_alloc.c
index 61cdc6c..e59fdac 100644
--- a/libxfs/xfs_alloc.c
+++ b/libxfs/xfs_alloc.c
@@ -822,7 +822,7 @@ xfs_alloc_ag_vextent_near(
*/
int dofirst; /* set to do first algorithm */
- dofirst = random32() & 1;
+ dofirst = prandom_u32() & 1;
#endif
restart:
diff --git a/libxfs/xfs_attr.c b/libxfs/xfs_attr.c
index 42546a9..2adf92b 100644
--- a/libxfs/xfs_attr.c
+++ b/libxfs/xfs_attr.c
@@ -275,6 +275,7 @@ xfs_attr_set_int(
if (rsvd)
args.trans->t_flags |= XFS_TRANS_RESERVE;
+
error = xfs_trans_reserve(args.trans, args.total,
XFS_ATTRSETM_LOG_RES(mp) +
XFS_ATTRSETRT_LOG_RES(mp) * args.total,
diff --git a/libxfs/xfs_attr_leaf.c b/libxfs/xfs_attr_leaf.c
index 824c122..426130f 100644
--- a/libxfs/xfs_attr_leaf.c
+++ b/libxfs/xfs_attr_leaf.c
@@ -141,7 +141,8 @@ xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes)
int dsize;
xfs_mount_t *mp = dp->i_mount;
- offset = (XFS_LITINO(mp) - bytes) >> 3; /* rounded down */
+ /* rounded down */
+ offset = (XFS_LITINO(mp, dp->i_d.di_version) - bytes) >> 3;
switch (dp->i_d.di_format) {
case XFS_DINODE_FMT_DEV:
@@ -212,7 +213,8 @@ xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes)
minforkoff = roundup(minforkoff, 8) >> 3;
/* attr fork btree root can have at least this many key/ptr pairs */
- maxforkoff = XFS_LITINO(mp) - XFS_BMDR_SPACE_CALC(MINABTPTRS);
+ maxforkoff = XFS_LITINO(mp, dp->i_d.di_version) -
+ XFS_BMDR_SPACE_CALC(MINABTPTRS);
maxforkoff = maxforkoff >> 3; /* rounded down */
if (offset >= maxforkoff)
diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index b328a0b..c8232a9 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -15,152 +15,77 @@
* along with this program; if not, write the Free Software Foundation,
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*/
-
#include <xfs.h>
-
kmem_zone_t *xfs_bmap_free_item_zone;
/*
- * Prototypes for internal bmap routines.
- */
-
-#ifdef DEBUG
-STATIC void
-xfs_bmap_check_leaf_extents(
- struct xfs_btree_cur *cur,
- struct xfs_inode *ip,
- int whichfork);
-#else
-#define xfs_bmap_check_leaf_extents(cur, ip, whichfork) do { } while (0)
-#endif
-
-
-/*
- * Called from xfs_bmap_add_attrfork to handle extents format files.
- */
-STATIC int /* error */
-xfs_bmap_add_attrfork_extents(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_fsblock_t *firstblock, /* first block allocated */
- xfs_bmap_free_t *flist, /* blocks to free at commit */
- int *flags); /* inode logging flags */
-
-/*
- * Called from xfs_bmap_add_attrfork to handle local format files.
- */
-STATIC int /* error */
-xfs_bmap_add_attrfork_local(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_fsblock_t *firstblock, /* first block allocated */
- xfs_bmap_free_t *flist, /* blocks to free at commit */
- int *flags); /* inode logging flags */
-
-/*
- * xfs_bmap_alloc is called by xfs_bmapi to allocate an extent for a file.
- * It figures out where to ask the underlying allocator to put the new extent.
- */
-STATIC int /* error */
-xfs_bmap_alloc(
- xfs_bmalloca_t *ap); /* bmap alloc argument struct */
-
-/*
- * Transform a btree format file with only one leaf node, where the
- * extents list will fit in the inode, into an extents format file.
- * Since the file extents are already in-core, all we have to do is
- * give up the space for the btree root and pitch the leaf block.
- */
-STATIC int /* error */
-xfs_bmap_btree_to_extents(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_btree_cur_t *cur, /* btree cursor */
- int *logflagsp, /* inode logging flags */
- int whichfork); /* data or attr fork */
-
-/*
- * Convert an extents-format file into a btree-format file.
- * The new file will have a root block (in the inode) and a single child block.
- */
-STATIC int /* error */
-xfs_bmap_extents_to_btree(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_fsblock_t *firstblock, /* first-block-allocated */
- xfs_bmap_free_t *flist, /* blocks freed in xaction */
- xfs_btree_cur_t **curp, /* cursor returned to caller */
- int wasdel, /* converting a delayed alloc */
- int *logflagsp, /* inode logging flags */
- int whichfork); /* data or attr fork */
-
-/*
- * Convert a local file to an extents file.
- * This code is sort of bogus, since the file data needs to get
- * logged so it won't be lost. The bmap-level manipulations are ok, though.
- */
-STATIC int /* error */
-xfs_bmap_local_to_extents(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_fsblock_t *firstblock, /* first block allocated in xaction */
- xfs_extlen_t total, /* total blocks needed by transaction */
- int *logflagsp, /* inode logging flags */
- int whichfork); /* data or attr fork */
-
-/*
- * Compute the worst-case number of indirect blocks that will be used
- * for ip's delayed extent of length "len".
+ * Miscellaneous helper functions
*/
-STATIC xfs_filblks_t
-xfs_bmap_worst_indlen(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_filblks_t len); /* delayed extent length */
-#ifdef DEBUG
/*
- * Perform various validation checks on the values being returned
- * from xfs_bmapi().
+ * Compute and fill in the value of the maximum depth of a bmap btree
+ * in this filesystem. Done once, during mount.
*/
-STATIC void
-xfs_bmap_validate_ret(
- xfs_fileoff_t bno,
- xfs_filblks_t len,
- int flags,
- xfs_bmbt_irec_t *mval,
- int nmap,
- int ret_nmap);
-#else
-#define xfs_bmap_validate_ret(bno,len,flags,mval,onmap,nmap)
-#endif /* DEBUG */
-
-STATIC int
-xfs_bmap_count_tree(
- xfs_mount_t *mp,
- xfs_trans_t *tp,
- xfs_ifork_t *ifp,
- xfs_fsblock_t blockno,
- int levelin,
- int *count);
-
-STATIC void
-xfs_bmap_count_leaves(
- xfs_ifork_t *ifp,
- xfs_extnum_t idx,
- int numrecs,
- int *count);
+void
+xfs_bmap_compute_maxlevels(
+ xfs_mount_t *mp, /* file system mount structure */
+ int whichfork) /* data or attr fork */
+{
+ int level; /* btree level */
+ uint maxblocks; /* max blocks at this level */
+ uint maxleafents; /* max leaf entries possible */
+ int maxrootrecs; /* max records in root block */
+ int minleafrecs; /* min records in leaf block */
+ int minnoderecs; /* min records in node block */
+ int sz; /* root block size */
-STATIC void
-xfs_bmap_disk_count_leaves(
- struct xfs_mount *mp,
- struct xfs_btree_block *block,
- int numrecs,
- int *count);
+ /*
+ * The maximum number of extents in a file, hence the maximum
+ * number of leaf entries, is controlled by the type of di_nextents
+ * (a signed 32-bit number, xfs_extnum_t), or by di_anextents
+ * (a signed 16-bit number, xfs_aextnum_t).
+ *
+ * Note that we can no longer assume that if we are in ATTR1 that
+ * the fork offset of all the inodes will be
+ * (xfs_default_attroffset(ip) >> 3) because we could have mounted
+ * with ATTR2 and then mounted back with ATTR1, keeping the
+ * di_forkoff's fixed but probably at various positions. Therefore,
+ * for both ATTR1 and ATTR2 we have to assume the worst case scenario
+ * of a minimum size available.
+ */
+ if (whichfork == XFS_DATA_FORK) {
+ maxleafents = MAXEXTNUM;
+ sz = XFS_BMDR_SPACE_CALC(MINDBTPTRS);
+ } else {
+ maxleafents = MAXAEXTNUM;
+ sz = XFS_BMDR_SPACE_CALC(MINABTPTRS);
+ }
+ maxrootrecs = xfs_bmdr_maxrecs(mp, sz, 0);
+ minleafrecs = mp->m_bmap_dmnr[0];
+ minnoderecs = mp->m_bmap_dmnr[1];
+ maxblocks = (maxleafents + minleafrecs - 1) / minleafrecs;
+ for (level = 1; maxblocks > 1; level++) {
+ if (maxblocks <= maxrootrecs)
+ maxblocks = 1;
+ else
+ maxblocks = (maxblocks + minnoderecs - 1) / minnoderecs;
+ }
+ mp->m_bm_maxlevels[whichfork] = level;
+}
/*
- * Bmap internal routines.
+ * Convert the given file system block to a disk block. We have to treat it
+ * differently based on whether the file is a real time file or not, because the
+ * bmap code does.
*/
+xfs_daddr_t
+xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb)
+{
+ return (XFS_IS_REALTIME_INODE(ip) ? \
+ (xfs_daddr_t)XFS_FSB_TO_BB((ip)->i_mount, (fsb)) : \
+ XFS_FSB_TO_DADDR((ip)->i_mount, (fsb)));
+}
STATIC int /* error */
xfs_bmbt_lookup_eq(
@@ -230,3769 +155,3881 @@ xfs_bmbt_update(
}
/*
- * Called from xfs_bmap_add_attrfork to handle btree format files.
+ * Compute the worst-case number of indirect blocks that will be used
+ * for ip's delayed extent of length "len".
*/
-STATIC int /* error */
-xfs_bmap_add_attrfork_btree(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_fsblock_t *firstblock, /* first block allocated */
- xfs_bmap_free_t *flist, /* blocks to free at commit */
- int *flags) /* inode logging flags */
+STATIC xfs_filblks_t
+xfs_bmap_worst_indlen(
+ xfs_inode_t *ip, /* incore inode pointer */
+ xfs_filblks_t len) /* delayed extent length */
{
- xfs_btree_cur_t *cur; /* btree cursor */
- int error; /* error return value */
- xfs_mount_t *mp; /* file system mount struct */
- int stat; /* newroot status */
+ int level; /* btree level number */
+ int maxrecs; /* maximum record count at this level */
+ xfs_mount_t *mp; /* mount structure */
+ xfs_filblks_t rval; /* return value */
mp = ip->i_mount;
- if (ip->i_df.if_broot_bytes <= XFS_IFORK_DSIZE(ip))
- *flags |= XFS_ILOG_DBROOT;
- else {
- cur = xfs_bmbt_init_cursor(mp, tp, ip, XFS_DATA_FORK);
- cur->bc_private.b.flist = flist;
- cur->bc_private.b.firstblock = *firstblock;
- if ((error = xfs_bmbt_lookup_ge(cur, 0, 0, 0, &stat)))
- goto error0;
- /* must be at least one entry */
- XFS_WANT_CORRUPTED_GOTO(stat == 1, error0);
- if ((error = xfs_btree_new_iroot(cur, flags, &stat)))
- goto error0;
- if (stat == 0) {
- xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
- return XFS_ERROR(ENOSPC);
- }
- *firstblock = cur->bc_private.b.firstblock;
- cur->bc_private.b.allocated = 0;
- xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
+ maxrecs = mp->m_bmap_dmxr[0];
+ for (level = 0, rval = 0;
+ level < XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK);
+ level++) {
+ len += maxrecs - 1;
+ do_div(len, maxrecs);
+ rval += len;
+ if (len == 1)
+ return rval + XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK) -
+ level - 1;
+ if (level == 0)
+ maxrecs = mp->m_bmap_dmxr[1];
}
- return 0;
-error0:
- xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
- return error;
+ return rval;
}
/*
- * Called from xfs_bmap_add_attrfork to handle extents format files.
+ * Calculate the default attribute fork offset for newly created inodes.
*/
-STATIC int /* error */
-xfs_bmap_add_attrfork_extents(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_fsblock_t *firstblock, /* first block allocated */
- xfs_bmap_free_t *flist, /* blocks to free at commit */
- int *flags) /* inode logging flags */
+uint
+xfs_default_attroffset(
+ struct xfs_inode *ip)
{
- xfs_btree_cur_t *cur; /* bmap btree cursor */
- int error; /* error return value */
+ struct xfs_mount *mp = ip->i_mount;
+ uint offset;
- if (ip->i_d.di_nextents * sizeof(xfs_bmbt_rec_t) <= XFS_IFORK_DSIZE(ip))
- return 0;
- cur = NULL;
- error = xfs_bmap_extents_to_btree(tp, ip, firstblock, flist, &cur, 0,
- flags, XFS_DATA_FORK);
- if (cur) {
- cur->bc_private.b.allocated = 0;
- xfs_btree_del_cursor(cur,
- error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR);
+ if (mp->m_sb.sb_inodesize == 256) {
+ offset = XFS_LITINO(mp, ip->i_d.di_version) -
+ XFS_BMDR_SPACE_CALC(MINABTPTRS);
+ } else {
+ offset = XFS_BMDR_SPACE_CALC(6 * MINABTPTRS);
}
- return error;
+
+ ASSERT(offset < XFS_LITINO(mp, ip->i_d.di_version));
+ return offset;
}
/*
- * Called from xfs_bmap_add_attrfork to handle local format files.
+ * Helper routine to reset inode di_forkoff field when switching
+ * attribute fork from local to extent format - we reset it where
+ * possible to make space available for inline data fork extents.
*/
-STATIC int /* error */
-xfs_bmap_add_attrfork_local(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_fsblock_t *firstblock, /* first block allocated */
- xfs_bmap_free_t *flist, /* blocks to free at commit */
- int *flags) /* inode logging flags */
+STATIC void
+xfs_bmap_forkoff_reset(
+ xfs_mount_t *mp,
+ xfs_inode_t *ip,
+ int whichfork)
{
- xfs_da_args_t dargs; /* args for dir/attr code */
- int error; /* error return value */
- xfs_mount_t *mp; /* mount structure pointer */
+ if (whichfork == XFS_ATTR_FORK &&
+ ip->i_d.di_format != XFS_DINODE_FMT_DEV &&
+ ip->i_d.di_format != XFS_DINODE_FMT_UUID &&
+ ip->i_d.di_format != XFS_DINODE_FMT_BTREE) {
+ uint dfl_forkoff = xfs_default_attroffset(ip) >> 3;
- if (ip->i_df.if_bytes <= XFS_IFORK_DSIZE(ip))
- return 0;
- if (S_ISDIR(ip->i_d.di_mode)) {
- mp = ip->i_mount;
- memset(&dargs, 0, sizeof(dargs));
- dargs.dp = ip;
- dargs.firstblock = firstblock;
- dargs.flist = flist;
- dargs.total = mp->m_dirblkfsbs;
- dargs.whichfork = XFS_DATA_FORK;
- dargs.trans = tp;
- error = xfs_dir2_sf_to_block(&dargs);
- } else
- error = xfs_bmap_local_to_extents(tp, ip, firstblock, 1, flags,
- XFS_DATA_FORK);
- return error;
+ if (dfl_forkoff > ip->i_d.di_forkoff)
+ ip->i_d.di_forkoff = dfl_forkoff;
+ }
}
/*
- * Convert a delayed allocation to a real allocation.
+ * Extent tree block counting routines.
*/
-STATIC int /* error */
-xfs_bmap_add_extent_delay_real(
- struct xfs_bmalloca *bma)
+
+/*
+ * Count leaf blocks given a range of extent records.
+ */
+STATIC void
+xfs_bmap_count_leaves(
+ xfs_ifork_t *ifp,
+ xfs_extnum_t idx,
+ int numrecs,
+ int *count)
{
- struct xfs_bmbt_irec *new = &bma->got;
- int diff; /* temp value */
- xfs_bmbt_rec_host_t *ep; /* extent entry for idx */
- int error; /* error return value */
- int i; /* temp state */
- xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_fileoff_t new_endoff; /* end offset of new entry */
- xfs_bmbt_irec_t r[3]; /* neighbor extent entries */
- /* left is 0, right is 1, prev is 2 */
- int rval=0; /* return value (logging flags) */
- int state = 0;/* state bits, accessed thru macros */
- xfs_filblks_t da_new; /* new count del alloc blocks used */
- xfs_filblks_t da_old; /* old count del alloc blocks used */
- xfs_filblks_t temp=0; /* value for da_new calculations */
- xfs_filblks_t temp2=0;/* value for da_new calculations */
- int tmp_rval; /* partial logging flags */
+ int b;
- ifp = XFS_IFORK_PTR(bma->ip, XFS_DATA_FORK);
+ for (b = 0; b < numrecs; b++) {
+ xfs_bmbt_rec_host_t *frp = xfs_iext_get_ext(ifp, idx + b);
+ *count += xfs_bmbt_get_blockcount(frp);
+ }
+}
- ASSERT(bma->idx >= 0);
- ASSERT(bma->idx <= ifp->if_bytes / sizeof(struct xfs_bmbt_rec));
- ASSERT(!isnullstartblock(new->br_startblock));
- ASSERT(!bma->cur ||
- (bma->cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL));
+/*
+ * Count leaf blocks given a range of extent records originally
+ * in btree format.
+ */
+STATIC void
+xfs_bmap_disk_count_leaves(
+ struct xfs_mount *mp,
+ struct xfs_btree_block *block,
+ int numrecs,
+ int *count)
+{
+ int b;
+ xfs_bmbt_rec_t *frp;
- XFS_STATS_INC(xs_add_exlist);
+ for (b = 1; b <= numrecs; b++) {
+ frp = XFS_BMBT_REC_ADDR(mp, block, b);
+ *count += xfs_bmbt_disk_get_blockcount(frp);
+ }
+}
-#define LEFT r[0]
-#define RIGHT r[1]
-#define PREV r[2]
+/*
+ * Recursively walks each level of a btree
+ * to count total fsblocks is use.
+ */
+STATIC int /* error */
+xfs_bmap_count_tree(
+ xfs_mount_t *mp, /* file system mount point */
+ xfs_trans_t *tp, /* transaction pointer */
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ xfs_fsblock_t blockno, /* file system block number */
+ int levelin, /* level in btree */
+ int *count) /* Count of blocks */
+{
+ int error;
+ xfs_buf_t *bp, *nbp;
+ int level = levelin;
+ __be64 *pp;
+ xfs_fsblock_t bno = blockno;
+ xfs_fsblock_t nextbno;
+ struct xfs_btree_block *block, *nextblock;
+ int numrecs;
- /*
- * Set up a bunch of variables to make the tests simpler.
- */
- ep = xfs_iext_get_ext(ifp, bma->idx);
- xfs_bmbt_get_all(ep, &PREV);
- new_endoff = new->br_startoff + new->br_blockcount;
- ASSERT(PREV.br_startoff <= new->br_startoff);
- ASSERT(PREV.br_startoff + PREV.br_blockcount >= new_endoff);
+ error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp, XFS_BMAP_BTREE_REF,
+ &xfs_bmbt_buf_ops);
+ if (error)
+ return error;
+ *count += 1;
+ block = XFS_BUF_TO_BLOCK(bp);
- da_old = startblockval(PREV.br_startblock);
- da_new = 0;
+ if (--level) {
+ /* Not at node above leaves, count this level of nodes */
+ nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
+ while (nextbno != NULLFSBLOCK) {
+ error = xfs_btree_read_bufl(mp, tp, nextbno, 0, &nbp,
+ XFS_BMAP_BTREE_REF,
+ &xfs_bmbt_buf_ops);
+ if (error)
+ return error;
+ *count += 1;
+ nextblock = XFS_BUF_TO_BLOCK(nbp);
+ nextbno = be64_to_cpu(nextblock->bb_u.l.bb_rightsib);
+ xfs_trans_brelse(tp, nbp);
+ }
- /*
- * Set flags determining what part of the previous delayed allocation
- * extent is being replaced by a real allocation.
- */
- if (PREV.br_startoff == new->br_startoff)
- state |= BMAP_LEFT_FILLING;
- if (PREV.br_startoff + PREV.br_blockcount == new_endoff)
- state |= BMAP_RIGHT_FILLING;
+ /* Dive to the next level */
+ pp = XFS_BMBT_PTR_ADDR(mp, block, 1, mp->m_bmap_dmxr[1]);
+ bno = be64_to_cpu(*pp);
+ if (unlikely((error =
+ xfs_bmap_count_tree(mp, tp, ifp, bno, level, count)) < 0)) {
+ xfs_trans_brelse(tp, bp);
+ XFS_ERROR_REPORT("xfs_bmap_count_tree(1)",
+ XFS_ERRLEVEL_LOW, mp);
+ return XFS_ERROR(EFSCORRUPTED);
+ }
+ xfs_trans_brelse(tp, bp);
+ } else {
+ /* count all level 1 nodes and their leaves */
+ for (;;) {
+ nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
+ numrecs = be16_to_cpu(block->bb_numrecs);
+ xfs_bmap_disk_count_leaves(mp, block, numrecs, count);
+ xfs_trans_brelse(tp, bp);
+ if (nextbno == NULLFSBLOCK)
+ break;
+ bno = nextbno;
+ error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp,
+ XFS_BMAP_BTREE_REF,
+ &xfs_bmbt_buf_ops);
+ if (error)
+ return error;
+ *count += 1;
+ block = XFS_BUF_TO_BLOCK(bp);
+ }
+ }
+ return 0;
+}
- /*
- * Check and set flags if this segment has a left neighbor.
- * Don't set contiguous if the combined extent would be too large.
- */
- if (bma->idx > 0) {
- state |= BMAP_LEFT_VALID;
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx - 1), &LEFT);
+/*
+ * Count fsblocks of the given fork.
+ */
+int /* error */
+xfs_bmap_count_blocks(
+ xfs_trans_t *tp, /* transaction pointer */
+ xfs_inode_t *ip, /* incore inode */
+ int whichfork, /* data or attr fork */
+ int *count) /* out: count of blocks */
+{
+ struct xfs_btree_block *block; /* current btree block */
+ xfs_fsblock_t bno; /* block # of "block" */
+ xfs_ifork_t *ifp; /* fork structure */
+ int level; /* btree level, for checking */
+ xfs_mount_t *mp; /* file system mount structure */
+ __be64 *pp; /* pointer to block address */
- if (isnullstartblock(LEFT.br_startblock))
- state |= BMAP_LEFT_DELAY;
+ bno = NULLFSBLOCK;
+ mp = ip->i_mount;
+ ifp = XFS_IFORK_PTR(ip, whichfork);
+ if ( XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS ) {
+ xfs_bmap_count_leaves(ifp, 0,
+ ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t),
+ count);
+ return 0;
}
- if ((state & BMAP_LEFT_VALID) && !(state & BMAP_LEFT_DELAY) &&
- LEFT.br_startoff + LEFT.br_blockcount == new->br_startoff &&
- LEFT.br_startblock + LEFT.br_blockcount == new->br_startblock &&
- LEFT.br_state == new->br_state &&
- LEFT.br_blockcount + new->br_blockcount <= MAXEXTLEN)
- state |= BMAP_LEFT_CONTIG;
-
/*
- * Check and set flags if this segment has a right neighbor.
- * Don't set contiguous if the combined extent would be too large.
- * Also check for all-three-contiguous being too large.
+ * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out.
*/
- if (bma->idx < bma->ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) {
- state |= BMAP_RIGHT_VALID;
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx + 1), &RIGHT);
+ block = ifp->if_broot;
+ level = be16_to_cpu(block->bb_level);
+ ASSERT(level > 0);
+ pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
+ bno = be64_to_cpu(*pp);
+ ASSERT(bno != NULLDFSBNO);
+ ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount);
+ ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks);
- if (isnullstartblock(RIGHT.br_startblock))
- state |= BMAP_RIGHT_DELAY;
+ if (unlikely(xfs_bmap_count_tree(mp, tp, ifp, bno, level, count) < 0)) {
+ XFS_ERROR_REPORT("xfs_bmap_count_blocks(2)", XFS_ERRLEVEL_LOW,
+ mp);
+ return XFS_ERROR(EFSCORRUPTED);
}
- if ((state & BMAP_RIGHT_VALID) && !(state & BMAP_RIGHT_DELAY) &&
- new_endoff == RIGHT.br_startoff &&
- new->br_startblock + new->br_blockcount == RIGHT.br_startblock &&
- new->br_state == RIGHT.br_state &&
- new->br_blockcount + RIGHT.br_blockcount <= MAXEXTLEN &&
- ((state & (BMAP_LEFT_CONTIG | BMAP_LEFT_FILLING |
- BMAP_RIGHT_FILLING)) !=
- (BMAP_LEFT_CONTIG | BMAP_LEFT_FILLING |
- BMAP_RIGHT_FILLING) ||
- LEFT.br_blockcount + new->br_blockcount + RIGHT.br_blockcount
- <= MAXEXTLEN))
- state |= BMAP_RIGHT_CONTIG;
+ return 0;
+}
- error = 0;
- /*
- * Switch out based on the FILLING and CONTIG state bits.
- */
- switch (state & (BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG |
- BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG)) {
- case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG |
- BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
- /*
- * Filling in all of a previously delayed allocation extent.
- * The left and right neighbors are both contiguous with new.
- */
- bma->idx--;
- trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, bma->idx),
- LEFT.br_blockcount + PREV.br_blockcount +
- RIGHT.br_blockcount);
- trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+/*
+ * Debug/sanity checking code
+ */
- xfs_iext_remove(bma->ip, bma->idx + 1, 2, state);
- bma->ip->i_d.di_nextents--;
- if (bma->cur == NULL)
- rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
- else {
- rval = XFS_ILOG_CORE;
- error = xfs_bmbt_lookup_eq(bma->cur, RIGHT.br_startoff,
- RIGHT.br_startblock,
- RIGHT.br_blockcount, &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- error = xfs_btree_delete(bma->cur, &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- error = xfs_btree_decrement(bma->cur, 0, &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- error = xfs_bmbt_update(bma->cur, LEFT.br_startoff,
- LEFT.br_startblock,
- LEFT.br_blockcount +
- PREV.br_blockcount +
- RIGHT.br_blockcount, LEFT.br_state);
- if (error)
- goto done;
- }
- break;
+STATIC int
+xfs_bmap_sanity_check(
+ struct xfs_mount *mp,
+ struct xfs_buf *bp,
+ int level)
+{
+ struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
- case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG:
- /*
- * Filling in all of a previously delayed allocation extent.
- * The left neighbor is contiguous, the right is not.
- */
- bma->idx--;
+ if (block->bb_magic != cpu_to_be32(XFS_BMAP_MAGIC) ||
+ be16_to_cpu(block->bb_level) != level ||
+ be16_to_cpu(block->bb_numrecs) == 0 ||
+ be16_to_cpu(block->bb_numrecs) > mp->m_bmap_dmxr[level != 0])
+ return 0;
+ return 1;
+}
- trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, bma->idx),
- LEFT.br_blockcount + PREV.br_blockcount);
- trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+#ifdef DEBUG
+STATIC struct xfs_buf *
+xfs_bmap_get_bp(
+ struct xfs_btree_cur *cur,
+ xfs_fsblock_t bno)
+{
+ struct xfs_log_item_desc *lidp;
+ int i;
- xfs_iext_remove(bma->ip, bma->idx + 1, 1, state);
- if (bma->cur == NULL)
- rval = XFS_ILOG_DEXT;
- else {
- rval = 0;
- error = xfs_bmbt_lookup_eq(bma->cur, LEFT.br_startoff,
- LEFT.br_startblock, LEFT.br_blockcount,
- &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- error = xfs_bmbt_update(bma->cur, LEFT.br_startoff,
- LEFT.br_startblock,
- LEFT.br_blockcount +
- PREV.br_blockcount, LEFT.br_state);
- if (error)
- goto done;
- }
- break;
+ if (!cur)
+ return NULL;
- case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
- /*
- * Filling in all of a previously delayed allocation extent.
- * The right neighbor is contiguous, the left is not.
- */
- trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
- xfs_bmbt_set_startblock(ep, new->br_startblock);
- xfs_bmbt_set_blockcount(ep,
- PREV.br_blockcount + RIGHT.br_blockcount);
- trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+ for (i = 0; i < XFS_BTREE_MAXLEVELS; i++) {
+ if (!cur->bc_bufs[i])
+ break;
+ if (XFS_BUF_ADDR(cur->bc_bufs[i]) == bno)
+ return cur->bc_bufs[i];
+ }
- xfs_iext_remove(bma->ip, bma->idx + 1, 1, state);
- if (bma->cur == NULL)
- rval = XFS_ILOG_DEXT;
- else {
- rval = 0;
- error = xfs_bmbt_lookup_eq(bma->cur, RIGHT.br_startoff,
- RIGHT.br_startblock,
- RIGHT.br_blockcount, &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- error = xfs_bmbt_update(bma->cur, PREV.br_startoff,
- new->br_startblock,
- PREV.br_blockcount +
- RIGHT.br_blockcount, PREV.br_state);
- if (error)
- goto done;
- }
- break;
+ /* Chase down all the log items to see if the bp is there */
+ list_for_each_entry(lidp, &cur->bc_tp->t_items, lid_trans) {
+ struct xfs_buf_log_item *bip;
+ bip = (struct xfs_buf_log_item *)lidp->lid_item;
+ if (bip->bli_item.li_type == XFS_LI_BUF &&
+ XFS_BUF_ADDR(bip->bli_buf) == bno)
+ return bip->bli_buf;
+ }
- case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING:
- /*
- * Filling in all of a previously delayed allocation extent.
- * Neither the left nor right neighbors are contiguous with
- * the new one.
- */
- trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
- xfs_bmbt_set_startblock(ep, new->br_startblock);
- trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+ return NULL;
+}
- bma->ip->i_d.di_nextents++;
- if (bma->cur == NULL)
- rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
- else {
- rval = XFS_ILOG_CORE;
- error = xfs_bmbt_lookup_eq(bma->cur, new->br_startoff,
- new->br_startblock, new->br_blockcount,
- &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 0, done);
- bma->cur->bc_rec.b.br_state = XFS_EXT_NORM;
- error = xfs_btree_insert(bma->cur, &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- }
- break;
+STATIC void
+xfs_check_block(
+ struct xfs_btree_block *block,
+ xfs_mount_t *mp,
+ int root,
+ short sz)
+{
+ int i, j, dmxr;
+ __be64 *pp, *thispa; /* pointer to block address */
+ xfs_bmbt_key_t *prevp, *keyp;
- case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG:
- /*
- * Filling in the first part of a previous delayed allocation.
- * The left neighbor is contiguous.
- */
- trace_xfs_bmap_pre_update(bma->ip, bma->idx - 1, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, bma->idx - 1),
- LEFT.br_blockcount + new->br_blockcount);
- xfs_bmbt_set_startoff(ep,
- PREV.br_startoff + new->br_blockcount);
- trace_xfs_bmap_post_update(bma->ip, bma->idx - 1, state, _THIS_IP_);
+ ASSERT(be16_to_cpu(block->bb_level) > 0);
- temp = PREV.br_blockcount - new->br_blockcount;
- trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(ep, temp);
- if (bma->cur == NULL)
- rval = XFS_ILOG_DEXT;
- else {
- rval = 0;
- error = xfs_bmbt_lookup_eq(bma->cur, LEFT.br_startoff,
- LEFT.br_startblock, LEFT.br_blockcount,
- &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- error = xfs_bmbt_update(bma->cur, LEFT.br_startoff,
- LEFT.br_startblock,
- LEFT.br_blockcount +
- new->br_blockcount,
- LEFT.br_state);
- if (error)
- goto done;
- }
- da_new = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(bma->ip, temp),
- startblockval(PREV.br_startblock));
- xfs_bmbt_set_startblock(ep, nullstartblock(da_new));
- trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+ prevp = NULL;
+ for( i = 1; i <= xfs_btree_get_numrecs(block); i++) {
+ dmxr = mp->m_bmap_dmxr[0];
+ keyp = XFS_BMBT_KEY_ADDR(mp, block, i);
- bma->idx--;
- break;
+ if (prevp) {
+ ASSERT(be64_to_cpu(prevp->br_startoff) <
+ be64_to_cpu(keyp->br_startoff));
+ }
+ prevp = keyp;
- case BMAP_LEFT_FILLING:
/*
- * Filling in the first part of a previous delayed allocation.
- * The left neighbor is not contiguous.
+ * Compare the block numbers to see if there are dups.
*/
- trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
- xfs_bmbt_set_startoff(ep, new_endoff);
- temp = PREV.br_blockcount - new->br_blockcount;
- xfs_bmbt_set_blockcount(ep, temp);
- xfs_iext_insert(bma->ip, bma->idx, 1, new, state);
- bma->ip->i_d.di_nextents++;
- if (bma->cur == NULL)
- rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
- else {
- rval = XFS_ILOG_CORE;
- error = xfs_bmbt_lookup_eq(bma->cur, new->br_startoff,
- new->br_startblock, new->br_blockcount,
- &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 0, done);
- bma->cur->bc_rec.b.br_state = XFS_EXT_NORM;
- error = xfs_btree_insert(bma->cur, &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ if (root)
+ pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, i, sz);
+ else
+ pp = XFS_BMBT_PTR_ADDR(mp, block, i, dmxr);
+
+ for (j = i+1; j <= be16_to_cpu(block->bb_numrecs); j++) {
+ if (root)
+ thispa = XFS_BMAP_BROOT_PTR_ADDR(mp, block, j, sz);
+ else
+ thispa = XFS_BMBT_PTR_ADDR(mp, block, j, dmxr);
+ if (*thispa == *pp) {
+ xfs_warn(mp, "%s: thispa(%d) == pp(%d) %Ld",
+ __func__, j, i,
+ (unsigned long long)be64_to_cpu(*thispa));
+ panic("%s: ptrs are equal in node\n",
+ __func__);
+ }
}
+ }
+}
- if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
- error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
- bma->firstblock, bma->flist,
- &bma->cur, 1, &tmp_rval, XFS_DATA_FORK);
- rval |= tmp_rval;
+/*
+ * Check that the extents for the inode ip are in the right order in all
+ * btree leaves.
+ */
+
+STATIC void
+xfs_bmap_check_leaf_extents(
+ xfs_btree_cur_t *cur, /* btree cursor or null */
+ xfs_inode_t *ip, /* incore inode pointer */
+ int whichfork) /* data or attr fork */
+{
+ struct xfs_btree_block *block; /* current btree block */
+ xfs_fsblock_t bno; /* block # of "block" */
+ xfs_buf_t *bp; /* buffer for "block" */
+ int error; /* error return value */
+ xfs_extnum_t i=0, j; /* index into the extents list */
+ xfs_ifork_t *ifp; /* fork structure */
+ int level; /* btree level, for checking */
+ xfs_mount_t *mp; /* file system mount structure */
+ __be64 *pp; /* pointer to block address */
+ xfs_bmbt_rec_t *ep; /* pointer to current extent */
+ xfs_bmbt_rec_t last = {0, 0}; /* last extent in prev block */
+ xfs_bmbt_rec_t *nextp; /* pointer to next extent */
+ int bp_release = 0;
+
+ if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE) {
+ return;
+ }
+
+ bno = NULLFSBLOCK;
+ mp = ip->i_mount;
+ ifp = XFS_IFORK_PTR(ip, whichfork);
+ block = ifp->if_broot;
+ /*
+ * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out.
+ */
+ level = be16_to_cpu(block->bb_level);
+ ASSERT(level > 0);
+ xfs_check_block(block, mp, 1, ifp->if_broot_bytes);
+ pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
+ bno = be64_to_cpu(*pp);
+
+ ASSERT(bno != NULLDFSBNO);
+ ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount);
+ ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks);
+
+ /*
+ * Go down the tree until leaf level is reached, following the first
+ * pointer (leftmost) at each level.
+ */
+ while (level-- > 0) {
+ /* See if buf is in cur first */
+ bp_release = 0;
+ bp = xfs_bmap_get_bp(cur, XFS_FSB_TO_DADDR(mp, bno));
+ if (!bp) {
+ bp_release = 1;
+ error = xfs_btree_read_bufl(mp, NULL, bno, 0, &bp,
+ XFS_BMAP_BTREE_REF,
+ &xfs_bmbt_buf_ops);
if (error)
- goto done;
+ goto error_norelse;
}
- da_new = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(bma->ip, temp),
- startblockval(PREV.br_startblock) -
- (bma->cur ? bma->cur->bc_private.b.allocated : 0));
- ep = xfs_iext_get_ext(ifp, bma->idx + 1);
- xfs_bmbt_set_startblock(ep, nullstartblock(da_new));
- trace_xfs_bmap_post_update(bma->ip, bma->idx + 1, state, _THIS_IP_);
- break;
+ block = XFS_BUF_TO_BLOCK(bp);
+ XFS_WANT_CORRUPTED_GOTO(
+ xfs_bmap_sanity_check(mp, bp, level),
+ error0);
+ if (level == 0)
+ break;
- case BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
/*
- * Filling in the last part of a previous delayed allocation.
- * The right neighbor is contiguous with the new allocation.
+ * Check this block for basic sanity (increasing keys and
+ * no duplicate blocks).
*/
- temp = PREV.br_blockcount - new->br_blockcount;
- trace_xfs_bmap_pre_update(bma->ip, bma->idx + 1, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(ep, temp);
- xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, bma->idx + 1),
- new->br_startoff, new->br_startblock,
- new->br_blockcount + RIGHT.br_blockcount,
- RIGHT.br_state);
- trace_xfs_bmap_post_update(bma->ip, bma->idx + 1, state, _THIS_IP_);
- if (bma->cur == NULL)
- rval = XFS_ILOG_DEXT;
- else {
- rval = 0;
- error = xfs_bmbt_lookup_eq(bma->cur, RIGHT.br_startoff,
- RIGHT.br_startblock,
- RIGHT.br_blockcount, &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- error = xfs_bmbt_update(bma->cur, new->br_startoff,
- new->br_startblock,
- new->br_blockcount +
- RIGHT.br_blockcount,
- RIGHT.br_state);
- if (error)
- goto done;
+
+ xfs_check_block(block, mp, 0, 0);
+ pp = XFS_BMBT_PTR_ADDR(mp, block, 1, mp->m_bmap_dmxr[1]);
+ bno = be64_to_cpu(*pp);
+ XFS_WANT_CORRUPTED_GOTO(XFS_FSB_SANITY_CHECK(mp, bno), error0);
+ if (bp_release) {
+ bp_release = 0;
+ xfs_trans_brelse(NULL, bp);
}
+ }
- da_new = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(bma->ip, temp),
- startblockval(PREV.br_startblock));
- trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
- xfs_bmbt_set_startblock(ep, nullstartblock(da_new));
- trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+ /*
+ * Here with bp and block set to the leftmost leaf node in the tree.
+ */
+ i = 0;
- bma->idx++;
- break;
+ /*
+ * Loop over all leaf nodes checking that all extents are in the right order.
+ */
+ for (;;) {
+ xfs_fsblock_t nextbno;
+ xfs_extnum_t num_recs;
+
+
+ num_recs = xfs_btree_get_numrecs(block);
- case BMAP_RIGHT_FILLING:
/*
- * Filling in the last part of a previous delayed allocation.
- * The right neighbor is not contiguous.
+ * Read-ahead the next leaf block, if any.
*/
- temp = PREV.br_blockcount - new->br_blockcount;
- trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(ep, temp);
- xfs_iext_insert(bma->ip, bma->idx + 1, 1, new, state);
- bma->ip->i_d.di_nextents++;
- if (bma->cur == NULL)
- rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
- else {
- rval = XFS_ILOG_CORE;
- error = xfs_bmbt_lookup_eq(bma->cur, new->br_startoff,
- new->br_startblock, new->br_blockcount,
- &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 0, done);
- bma->cur->bc_rec.b.br_state = XFS_EXT_NORM;
- error = xfs_btree_insert(bma->cur, &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- }
- if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
- error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
- bma->firstblock, bma->flist, &bma->cur, 1,
- &tmp_rval, XFS_DATA_FORK);
- rval |= tmp_rval;
- if (error)
- goto done;
- }
- da_new = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(bma->ip, temp),
- startblockval(PREV.br_startblock) -
- (bma->cur ? bma->cur->bc_private.b.allocated : 0));
- ep = xfs_iext_get_ext(ifp, bma->idx);
- xfs_bmbt_set_startblock(ep, nullstartblock(da_new));
- trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+ nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
- bma->idx++;
- break;
+ /*
+ * Check all the extents to make sure they are OK.
+ * If we had a previous block, the last entry should
+ * conform with the first entry in this one.
+ */
- case 0:
+ ep = XFS_BMBT_REC_ADDR(mp, block, 1);
+ if (i) {
+ ASSERT(xfs_bmbt_disk_get_startoff(&last) +
+ xfs_bmbt_disk_get_blockcount(&last) <=
+ xfs_bmbt_disk_get_startoff(ep));
+ }
+ for (j = 1; j < num_recs; j++) {
+ nextp = XFS_BMBT_REC_ADDR(mp, block, j + 1);
+ ASSERT(xfs_bmbt_disk_get_startoff(ep) +
+ xfs_bmbt_disk_get_blockcount(ep) <=
+ xfs_bmbt_disk_get_startoff(nextp));
+ ep = nextp;
+ }
+
+ last = *ep;
+ i += num_recs;
+ if (bp_release) {
+ bp_release = 0;
+ xfs_trans_brelse(NULL, bp);
+ }
+ bno = nextbno;
/*
- * Filling in the middle part of a previous delayed allocation.
- * Contiguity is impossible here.
- * This case is avoided almost all the time.
- *
- * We start with a delayed allocation:
- *
- * +ddddddddddddddddddddddddddddddddddddddddddddddddddddddd+
- * PREV @ idx
- *
- * and we are allocating:
- * +rrrrrrrrrrrrrrrrr+
- * new
- *
- * and we set it up for insertion as:
- * +ddddddddddddddddddd+rrrrrrrrrrrrrrrrr+ddddddddddddddddd+
- * new
- * PREV @ idx LEFT RIGHT
- * inserted at idx + 1
+ * If we've reached the end, stop.
*/
- temp = new->br_startoff - PREV.br_startoff;
- temp2 = PREV.br_startoff + PREV.br_blockcount - new_endoff;
- trace_xfs_bmap_pre_update(bma->ip, bma->idx, 0, _THIS_IP_);
- xfs_bmbt_set_blockcount(ep, temp); /* truncate PREV */
- LEFT = *new;
- RIGHT.br_state = PREV.br_state;
- RIGHT.br_startblock = nullstartblock(
- (int)xfs_bmap_worst_indlen(bma->ip, temp2));
- RIGHT.br_startoff = new_endoff;
- RIGHT.br_blockcount = temp2;
- /* insert LEFT (r[0]) and RIGHT (r[1]) at the same time */
- xfs_iext_insert(bma->ip, bma->idx + 1, 2, &LEFT, state);
- bma->ip->i_d.di_nextents++;
- if (bma->cur == NULL)
- rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
- else {
- rval = XFS_ILOG_CORE;
- error = xfs_bmbt_lookup_eq(bma->cur, new->br_startoff,
- new->br_startblock, new->br_blockcount,
- &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 0, done);
- bma->cur->bc_rec.b.br_state = XFS_EXT_NORM;
- error = xfs_btree_insert(bma->cur, &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- }
+ if (bno == NULLFSBLOCK)
+ break;
- if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
- error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
- bma->firstblock, bma->flist, &bma->cur,
- 1, &tmp_rval, XFS_DATA_FORK);
- rval |= tmp_rval;
- if (error)
- goto done;
- }
- temp = xfs_bmap_worst_indlen(bma->ip, temp);
- temp2 = xfs_bmap_worst_indlen(bma->ip, temp2);
- diff = (int)(temp + temp2 - startblockval(PREV.br_startblock) -
- (bma->cur ? bma->cur->bc_private.b.allocated : 0));
- if (diff > 0) {
- error = xfs_icsb_modify_counters(bma->ip->i_mount,
- XFS_SBS_FDBLOCKS,
- -((int64_t)diff), 0);
- ASSERT(!error);
+ bp_release = 0;
+ bp = xfs_bmap_get_bp(cur, XFS_FSB_TO_DADDR(mp, bno));
+ if (!bp) {
+ bp_release = 1;
+ error = xfs_btree_read_bufl(mp, NULL, bno, 0, &bp,
+ XFS_BMAP_BTREE_REF,
+ &xfs_bmbt_buf_ops);
if (error)
- goto done;
+ goto error_norelse;
}
-
- ep = xfs_iext_get_ext(ifp, bma->idx);
- xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
- trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
- trace_xfs_bmap_pre_update(bma->ip, bma->idx + 2, state, _THIS_IP_);
- xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, bma->idx + 2),
- nullstartblock((int)temp2));
- trace_xfs_bmap_post_update(bma->ip, bma->idx + 2, state, _THIS_IP_);
-
- bma->idx++;
- da_new = temp + temp2;
- break;
-
- case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
- case BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
- case BMAP_LEFT_FILLING | BMAP_RIGHT_CONTIG:
- case BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG:
- case BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
- case BMAP_LEFT_CONTIG:
- case BMAP_RIGHT_CONTIG:
- /*
- * These cases are all impossible.
- */
- ASSERT(0);
- }
-
- /* convert to a btree if necessary */
- if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
- int tmp_logflags; /* partial log flag return val */
-
- ASSERT(bma->cur == NULL);
- error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
- bma->firstblock, bma->flist, &bma->cur,
- da_old > 0, &tmp_logflags, XFS_DATA_FORK);
- bma->logflags |= tmp_logflags;
- if (error)
- goto done;
+ block = XFS_BUF_TO_BLOCK(bp);
}
-
- /* adjust for changes in reserved delayed indirect blocks */
- if (da_old || da_new) {
- temp = da_new;
- if (bma->cur)
- temp += bma->cur->bc_private.b.allocated;
- ASSERT(temp <= da_old);
- if (temp < da_old)
- xfs_icsb_modify_counters(bma->ip->i_mount,
- XFS_SBS_FDBLOCKS,
- (int64_t)(da_old - temp), 0);
+ if (bp_release) {
+ bp_release = 0;
+ xfs_trans_brelse(NULL, bp);
}
+ return;
- /* clear out the allocated field, done with it now in any case. */
- if (bma->cur)
- bma->cur->bc_private.b.allocated = 0;
-
- xfs_bmap_check_leaf_extents(bma->cur, bma->ip, XFS_DATA_FORK);
-done:
- bma->logflags |= rval;
- return error;
-#undef LEFT
-#undef RIGHT
-#undef PREV
+error0:
+ xfs_warn(mp, "%s: at error0", __func__);
+ if (bp_release)
+ xfs_trans_brelse(NULL, bp);
+error_norelse:
+ xfs_warn(mp, "%s: BAD after btree leaves for %d extents",
+ __func__, i);
+ panic("%s: CORRUPTED BTREE OR SOMETHING", __func__);
+ return;
}
/*
- * Convert an unwritten allocation to a real allocation or vice versa.
+ * Add bmap trace insert entries for all the contents of the extent records.
*/
-STATIC int /* error */
-xfs_bmap_add_extent_unwritten_real(
- struct xfs_trans *tp,
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_extnum_t *idx, /* extent number to update/insert */
- xfs_btree_cur_t **curp, /* if *curp is null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to add to file extents */
- xfs_fsblock_t *first, /* pointer to firstblock variable */
- xfs_bmap_free_t *flist, /* list of extents to be freed */
- int *logflagsp) /* inode logging flags */
+void
+xfs_bmap_trace_exlist(
+ xfs_inode_t *ip, /* incore inode pointer */
+ xfs_extnum_t cnt, /* count of entries in the list */
+ int whichfork, /* data or attr fork */
+ unsigned long caller_ip)
{
- xfs_btree_cur_t *cur; /* btree cursor */
- xfs_bmbt_rec_host_t *ep; /* extent entry for idx */
- int error; /* error return value */
- int i; /* temp state */
- xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_fileoff_t new_endoff; /* end offset of new entry */
- xfs_exntst_t newext; /* new extent state */
- xfs_exntst_t oldext; /* old extent state */
- xfs_bmbt_irec_t r[3]; /* neighbor extent entries */
- /* left is 0, right is 1, prev is 2 */
- int rval=0; /* return value (logging flags) */
- int state = 0;/* state bits, accessed thru macros */
-
- *logflagsp = 0;
+ xfs_extnum_t idx; /* extent record index */
+ xfs_ifork_t *ifp; /* inode fork pointer */
+ int state = 0;
- cur = *curp;
- ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
+ if (whichfork == XFS_ATTR_FORK)
+ state |= BMAP_ATTRFORK;
- ASSERT(*idx >= 0);
- ASSERT(*idx <= ifp->if_bytes / sizeof(struct xfs_bmbt_rec));
- ASSERT(!isnullstartblock(new->br_startblock));
+ ifp = XFS_IFORK_PTR(ip, whichfork);
+ ASSERT(cnt == (ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t)));
+ for (idx = 0; idx < cnt; idx++)
+ trace_xfs_extlist(ip, idx, whichfork, caller_ip);
+}
- XFS_STATS_INC(xs_add_exlist);
+/*
+ * Validate that the bmbt_irecs being returned from bmapi are valid
+ * given the callers original parameters. Specifically check the
+ * ranges of the returned irecs to ensure that they only extent beyond
+ * the given parameters if the XFS_BMAPI_ENTIRE flag was set.
+ */
+STATIC void
+xfs_bmap_validate_ret(
+ xfs_fileoff_t bno,
+ xfs_filblks_t len,
+ int flags,
+ xfs_bmbt_irec_t *mval,
+ int nmap,
+ int ret_nmap)
+{
+ int i; /* index to map values */
-#define LEFT r[0]
-#define RIGHT r[1]
-#define PREV r[2]
+ ASSERT(ret_nmap <= nmap);
- /*
- * Set up a bunch of variables to make the tests simpler.
- */
- error = 0;
- ep = xfs_iext_get_ext(ifp, *idx);
- xfs_bmbt_get_all(ep, &PREV);
- newext = new->br_state;
- oldext = (newext == XFS_EXT_UNWRITTEN) ?
- XFS_EXT_NORM : XFS_EXT_UNWRITTEN;
- ASSERT(PREV.br_state == oldext);
- new_endoff = new->br_startoff + new->br_blockcount;
- ASSERT(PREV.br_startoff <= new->br_startoff);
- ASSERT(PREV.br_startoff + PREV.br_blockcount >= new_endoff);
+ for (i = 0; i < ret_nmap; i++) {
+ ASSERT(mval[i].br_blockcount > 0);
+ if (!(flags & XFS_BMAPI_ENTIRE)) {
+ ASSERT(mval[i].br_startoff >= bno);
+ ASSERT(mval[i].br_blockcount <= len);
+ ASSERT(mval[i].br_startoff + mval[i].br_blockcount <=
+ bno + len);
+ } else {
+ ASSERT(mval[i].br_startoff < bno + len);
+ ASSERT(mval[i].br_startoff + mval[i].br_blockcount >
+ bno);
+ }
+ ASSERT(i == 0 ||
+ mval[i - 1].br_startoff + mval[i - 1].br_blockcount ==
+ mval[i].br_startoff);
+ ASSERT(mval[i].br_startblock != DELAYSTARTBLOCK &&
+ mval[i].br_startblock != HOLESTARTBLOCK);
+ ASSERT(mval[i].br_state == XFS_EXT_NORM ||
+ mval[i].br_state == XFS_EXT_UNWRITTEN);
+ }
+}
- /*
- * Set flags determining what part of the previous oldext allocation
- * extent is being replaced by a newext allocation.
- */
- if (PREV.br_startoff == new->br_startoff)
- state |= BMAP_LEFT_FILLING;
- if (PREV.br_startoff + PREV.br_blockcount == new_endoff)
- state |= BMAP_RIGHT_FILLING;
+#else
+#define xfs_bmap_check_leaf_extents(cur, ip, whichfork) do { } while (0)
+#define xfs_bmap_validate_ret(bno,len,flags,mval,onmap,nmap)
+#endif /* DEBUG */
- /*
- * Check and set flags if this segment has a left neighbor.
- * Don't set contiguous if the combined extent would be too large.
- */
- if (*idx > 0) {
- state |= BMAP_LEFT_VALID;
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx - 1), &LEFT);
+/*
+ * bmap free list manipulation functions
+ */
- if (isnullstartblock(LEFT.br_startblock))
- state |= BMAP_LEFT_DELAY;
- }
-
- if ((state & BMAP_LEFT_VALID) && !(state & BMAP_LEFT_DELAY) &&
- LEFT.br_startoff + LEFT.br_blockcount == new->br_startoff &&
- LEFT.br_startblock + LEFT.br_blockcount == new->br_startblock &&
- LEFT.br_state == newext &&
- LEFT.br_blockcount + new->br_blockcount <= MAXEXTLEN)
- state |= BMAP_LEFT_CONTIG;
+/*
+ * Add the extent to the list of extents to be free at transaction end.
+ * The list is maintained sorted (by block number).
+ */
+void
+xfs_bmap_add_free(
+ xfs_fsblock_t bno, /* fs block number of extent */
+ xfs_filblks_t len, /* length of extent */
+ xfs_bmap_free_t *flist, /* list of extents */
+ xfs_mount_t *mp) /* mount point structure */
+{
+ xfs_bmap_free_item_t *cur; /* current (next) element */
+ xfs_bmap_free_item_t *new; /* new element */
+ xfs_bmap_free_item_t *prev; /* previous element */
+#ifdef DEBUG
+ xfs_agnumber_t agno;
+ xfs_agblock_t agbno;
- /*
- * Check and set flags if this segment has a right neighbor.
- * Don't set contiguous if the combined extent would be too large.
- * Also check for all-three-contiguous being too large.
- */
- if (*idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) {
- state |= BMAP_RIGHT_VALID;
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx + 1), &RIGHT);
- if (isnullstartblock(RIGHT.br_startblock))
- state |= BMAP_RIGHT_DELAY;
+ ASSERT(bno != NULLFSBLOCK);
+ ASSERT(len > 0);
+ ASSERT(len <= MAXEXTLEN);
+ ASSERT(!isnullstartblock(bno));
+ agno = XFS_FSB_TO_AGNO(mp, bno);
+ agbno = XFS_FSB_TO_AGBNO(mp, bno);
+ ASSERT(agno < mp->m_sb.sb_agcount);
+ ASSERT(agbno < mp->m_sb.sb_agblocks);
+ ASSERT(len < mp->m_sb.sb_agblocks);
+ ASSERT(agbno + len <= mp->m_sb.sb_agblocks);
+#endif
+ ASSERT(xfs_bmap_free_item_zone != NULL);
+ new = kmem_zone_alloc(xfs_bmap_free_item_zone, KM_SLEEP);
+ new->xbfi_startblock = bno;
+ new->xbfi_blockcount = (xfs_extlen_t)len;
+ for (prev = NULL, cur = flist->xbf_first;
+ cur != NULL;
+ prev = cur, cur = cur->xbfi_next) {
+ if (cur->xbfi_startblock >= bno)
+ break;
}
+ if (prev)
+ prev->xbfi_next = new;
+ else
+ flist->xbf_first = new;
+ new->xbfi_next = cur;
+ flist->xbf_count++;
+}
- if ((state & BMAP_RIGHT_VALID) && !(state & BMAP_RIGHT_DELAY) &&
- new_endoff == RIGHT.br_startoff &&
- new->br_startblock + new->br_blockcount == RIGHT.br_startblock &&
- newext == RIGHT.br_state &&
- new->br_blockcount + RIGHT.br_blockcount <= MAXEXTLEN &&
- ((state & (BMAP_LEFT_CONTIG | BMAP_LEFT_FILLING |
- BMAP_RIGHT_FILLING)) !=
- (BMAP_LEFT_CONTIG | BMAP_LEFT_FILLING |
- BMAP_RIGHT_FILLING) ||
- LEFT.br_blockcount + new->br_blockcount + RIGHT.br_blockcount
- <= MAXEXTLEN))
- state |= BMAP_RIGHT_CONTIG;
+/*
+ * Remove the entry "free" from the free item list. Prev points to the
+ * previous entry, unless "free" is the head of the list.
+ */
+void
+xfs_bmap_del_free(
+ xfs_bmap_free_t *flist, /* free item list header */
+ xfs_bmap_free_item_t *prev, /* previous item on list, if any */
+ xfs_bmap_free_item_t *free) /* list item to be freed */
+{
+ if (prev)
+ prev->xbfi_next = free->xbfi_next;
+ else
+ flist->xbf_first = free->xbfi_next;
+ flist->xbf_count--;
+ kmem_zone_free(xfs_bmap_free_item_zone, free);
+}
- /*
- * Switch out based on the FILLING and CONTIG state bits.
- */
- switch (state & (BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG |
- BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG)) {
- case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG |
- BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
- /*
- * Setting all of a previous oldext extent to newext.
- * The left and right neighbors are both contiguous with new.
- */
- --*idx;
- trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx),
- LEFT.br_blockcount + PREV.br_blockcount +
- RIGHT.br_blockcount);
- trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
+/*
+ * Free up any items left in the list.
+ */
+void
+xfs_bmap_cancel(
+ xfs_bmap_free_t *flist) /* list of bmap_free_items */
+{
+ xfs_bmap_free_item_t *free; /* free list item */
+ xfs_bmap_free_item_t *next;
- xfs_iext_remove(ip, *idx + 1, 2, state);
- ip->i_d.di_nextents -= 2;
- if (cur == NULL)
- rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
- else {
- rval = XFS_ILOG_CORE;
- if ((error = xfs_bmbt_lookup_eq(cur, RIGHT.br_startoff,
- RIGHT.br_startblock,
- RIGHT.br_blockcount, &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_btree_delete(cur, &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_btree_decrement(cur, 0, &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_btree_delete(cur, &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_btree_decrement(cur, 0, &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_bmbt_update(cur, LEFT.br_startoff,
- LEFT.br_startblock,
- LEFT.br_blockcount + PREV.br_blockcount +
- RIGHT.br_blockcount, LEFT.br_state)))
- goto done;
- }
- break;
+ if (flist->xbf_count == 0)
+ return;
+ ASSERT(flist->xbf_first != NULL);
+ for (free = flist->xbf_first; free; free = next) {
+ next = free->xbfi_next;
+ xfs_bmap_del_free(flist, NULL, free);
+ }
+ ASSERT(flist->xbf_count == 0);
+}
- case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG:
- /*
- * Setting all of a previous oldext extent to newext.
- * The left neighbor is contiguous, the right is not.
- */
- --*idx;
+/*
+ * Inode fork format manipulation functions
+ */
- trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx),
- LEFT.br_blockcount + PREV.br_blockcount);
- trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
+/*
+ * Transform a btree format file with only one leaf node, where the
+ * extents list will fit in the inode, into an extents format file.
+ * Since the file extents are already in-core, all we have to do is
+ * give up the space for the btree root and pitch the leaf block.
+ */
+STATIC int /* error */
+xfs_bmap_btree_to_extents(
+ xfs_trans_t *tp, /* transaction pointer */
+ xfs_inode_t *ip, /* incore inode pointer */
+ xfs_btree_cur_t *cur, /* btree cursor */
+ int *logflagsp, /* inode logging flags */
+ int whichfork) /* data or attr fork */
+{
+ /* REFERENCED */
+ struct xfs_btree_block *cblock;/* child btree block */
+ xfs_fsblock_t cbno; /* child block number */
+ xfs_buf_t *cbp; /* child block's buffer */
+ int error; /* error return value */
+ xfs_ifork_t *ifp; /* inode fork data */
+ xfs_mount_t *mp; /* mount point structure */
+ __be64 *pp; /* ptr to block address */
+ struct xfs_btree_block *rblock;/* root btree block */
- xfs_iext_remove(ip, *idx + 1, 1, state);
- ip->i_d.di_nextents--;
- if (cur == NULL)
- rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
- else {
- rval = XFS_ILOG_CORE;
- if ((error = xfs_bmbt_lookup_eq(cur, PREV.br_startoff,
- PREV.br_startblock, PREV.br_blockcount,
- &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_btree_delete(cur, &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_btree_decrement(cur, 0, &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_bmbt_update(cur, LEFT.br_startoff,
- LEFT.br_startblock,
- LEFT.br_blockcount + PREV.br_blockcount,
- LEFT.br_state)))
- goto done;
- }
- break;
-
- case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
- /*
- * Setting all of a previous oldext extent to newext.
- * The right neighbor is contiguous, the left is not.
- */
- trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(ep,
- PREV.br_blockcount + RIGHT.br_blockcount);
- xfs_bmbt_set_state(ep, newext);
- trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- xfs_iext_remove(ip, *idx + 1, 1, state);
- ip->i_d.di_nextents--;
- if (cur == NULL)
- rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
- else {
- rval = XFS_ILOG_CORE;
- if ((error = xfs_bmbt_lookup_eq(cur, RIGHT.br_startoff,
- RIGHT.br_startblock,
- RIGHT.br_blockcount, &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_btree_delete(cur, &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_btree_decrement(cur, 0, &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_bmbt_update(cur, new->br_startoff,
- new->br_startblock,
- new->br_blockcount + RIGHT.br_blockcount,
- newext)))
- goto done;
- }
- break;
-
- case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING:
- /*
- * Setting all of a previous oldext extent to newext.
- * Neither the left nor right neighbors are contiguous with
- * the new one.
- */
- trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
- xfs_bmbt_set_state(ep, newext);
- trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
-
- if (cur == NULL)
- rval = XFS_ILOG_DEXT;
- else {
- rval = 0;
- if ((error = xfs_bmbt_lookup_eq(cur, new->br_startoff,
- new->br_startblock, new->br_blockcount,
- &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_bmbt_update(cur, new->br_startoff,
- new->br_startblock, new->br_blockcount,
- newext)))
- goto done;
- }
- break;
-
- case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG:
- /*
- * Setting the first part of a previous oldext extent to newext.
- * The left neighbor is contiguous.
- */
- trace_xfs_bmap_pre_update(ip, *idx - 1, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx - 1),
- LEFT.br_blockcount + new->br_blockcount);
- xfs_bmbt_set_startoff(ep,
- PREV.br_startoff + new->br_blockcount);
- trace_xfs_bmap_post_update(ip, *idx - 1, state, _THIS_IP_);
-
- trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
- xfs_bmbt_set_startblock(ep,
- new->br_startblock + new->br_blockcount);
- xfs_bmbt_set_blockcount(ep,
- PREV.br_blockcount - new->br_blockcount);
- trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
-
- --*idx;
-
- if (cur == NULL)
- rval = XFS_ILOG_DEXT;
- else {
- rval = 0;
- if ((error = xfs_bmbt_lookup_eq(cur, PREV.br_startoff,
- PREV.br_startblock, PREV.br_blockcount,
- &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_bmbt_update(cur,
- PREV.br_startoff + new->br_blockcount,
- PREV.br_startblock + new->br_blockcount,
- PREV.br_blockcount - new->br_blockcount,
- oldext)))
- goto done;
- if ((error = xfs_btree_decrement(cur, 0, &i)))
- goto done;
- error = xfs_bmbt_update(cur, LEFT.br_startoff,
- LEFT.br_startblock,
- LEFT.br_blockcount + new->br_blockcount,
- LEFT.br_state);
- if (error)
- goto done;
- }
- break;
-
- case BMAP_LEFT_FILLING:
- /*
- * Setting the first part of a previous oldext extent to newext.
- * The left neighbor is not contiguous.
- */
- trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
- ASSERT(ep && xfs_bmbt_get_state(ep) == oldext);
- xfs_bmbt_set_startoff(ep, new_endoff);
- xfs_bmbt_set_blockcount(ep,
- PREV.br_blockcount - new->br_blockcount);
- xfs_bmbt_set_startblock(ep,
- new->br_startblock + new->br_blockcount);
- trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
-
- xfs_iext_insert(ip, *idx, 1, new, state);
- ip->i_d.di_nextents++;
- if (cur == NULL)
- rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
- else {
- rval = XFS_ILOG_CORE;
- if ((error = xfs_bmbt_lookup_eq(cur, PREV.br_startoff,
- PREV.br_startblock, PREV.br_blockcount,
- &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_bmbt_update(cur,
- PREV.br_startoff + new->br_blockcount,
- PREV.br_startblock + new->br_blockcount,
- PREV.br_blockcount - new->br_blockcount,
- oldext)))
- goto done;
- cur->bc_rec.b = *new;
- if ((error = xfs_btree_insert(cur, &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- }
- break;
-
- case BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
- /*
- * Setting the last part of a previous oldext extent to newext.
- * The right neighbor is contiguous with the new allocation.
- */
- trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(ep,
- PREV.br_blockcount - new->br_blockcount);
- trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
-
- ++*idx;
-
- trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
- xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, *idx),
- new->br_startoff, new->br_startblock,
- new->br_blockcount + RIGHT.br_blockcount, newext);
- trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
-
- if (cur == NULL)
- rval = XFS_ILOG_DEXT;
- else {
- rval = 0;
- if ((error = xfs_bmbt_lookup_eq(cur, PREV.br_startoff,
- PREV.br_startblock,
- PREV.br_blockcount, &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_bmbt_update(cur, PREV.br_startoff,
- PREV.br_startblock,
- PREV.br_blockcount - new->br_blockcount,
- oldext)))
- goto done;
- if ((error = xfs_btree_increment(cur, 0, &i)))
- goto done;
- if ((error = xfs_bmbt_update(cur, new->br_startoff,
- new->br_startblock,
- new->br_blockcount + RIGHT.br_blockcount,
- newext)))
- goto done;
- }
- break;
-
- case BMAP_RIGHT_FILLING:
- /*
- * Setting the last part of a previous oldext extent to newext.
- * The right neighbor is not contiguous.
- */
- trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(ep,
- PREV.br_blockcount - new->br_blockcount);
- trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
-
- ++*idx;
- xfs_iext_insert(ip, *idx, 1, new, state);
-
- ip->i_d.di_nextents++;
- if (cur == NULL)
- rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
- else {
- rval = XFS_ILOG_CORE;
- if ((error = xfs_bmbt_lookup_eq(cur, PREV.br_startoff,
- PREV.br_startblock, PREV.br_blockcount,
- &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- if ((error = xfs_bmbt_update(cur, PREV.br_startoff,
- PREV.br_startblock,
- PREV.br_blockcount - new->br_blockcount,
- oldext)))
- goto done;
- if ((error = xfs_bmbt_lookup_eq(cur, new->br_startoff,
- new->br_startblock, new->br_blockcount,
- &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 0, done);
- cur->bc_rec.b.br_state = XFS_EXT_NORM;
- if ((error = xfs_btree_insert(cur, &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- }
- break;
-
- case 0:
- /*
- * Setting the middle part of a previous oldext extent to
- * newext. Contiguity is impossible here.
- * One extent becomes three extents.
- */
- trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(ep,
- new->br_startoff - PREV.br_startoff);
- trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
-
- r[0] = *new;
- r[1].br_startoff = new_endoff;
- r[1].br_blockcount =
- PREV.br_startoff + PREV.br_blockcount - new_endoff;
- r[1].br_startblock = new->br_startblock + new->br_blockcount;
- r[1].br_state = oldext;
-
- ++*idx;
- xfs_iext_insert(ip, *idx, 2, &r[0], state);
-
- ip->i_d.di_nextents += 2;
- if (cur == NULL)
- rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
- else {
- rval = XFS_ILOG_CORE;
- if ((error = xfs_bmbt_lookup_eq(cur, PREV.br_startoff,
- PREV.br_startblock, PREV.br_blockcount,
- &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- /* new right extent - oldext */
- if ((error = xfs_bmbt_update(cur, r[1].br_startoff,
- r[1].br_startblock, r[1].br_blockcount,
- r[1].br_state)))
- goto done;
- /* new left extent - oldext */
- cur->bc_rec.b = PREV;
- cur->bc_rec.b.br_blockcount =
- new->br_startoff - PREV.br_startoff;
- if ((error = xfs_btree_insert(cur, &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- /*
- * Reset the cursor to the position of the new extent
- * we are about to insert as we can't trust it after
- * the previous insert.
- */
- if ((error = xfs_bmbt_lookup_eq(cur, new->br_startoff,
- new->br_startblock, new->br_blockcount,
- &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 0, done);
- /* new middle extent - newext */
- cur->bc_rec.b.br_state = new->br_state;
- if ((error = xfs_btree_insert(cur, &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- }
- break;
-
- case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
- case BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
- case BMAP_LEFT_FILLING | BMAP_RIGHT_CONTIG:
- case BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG:
- case BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
- case BMAP_LEFT_CONTIG:
- case BMAP_RIGHT_CONTIG:
- /*
- * These cases are all impossible.
- */
- ASSERT(0);
- }
-
- /* convert to a btree if necessary */
- if (xfs_bmap_needs_btree(ip, XFS_DATA_FORK)) {
- int tmp_logflags; /* partial log flag return val */
-
- ASSERT(cur == NULL);
- error = xfs_bmap_extents_to_btree(tp, ip, first, flist, &cur,
- 0, &tmp_logflags, XFS_DATA_FORK);
- *logflagsp |= tmp_logflags;
- if (error)
- goto done;
- }
-
- /* clear out the allocated field, done with it now in any case. */
- if (cur) {
- cur->bc_private.b.allocated = 0;
- *curp = cur;
- }
-
- xfs_bmap_check_leaf_extents(*curp, ip, XFS_DATA_FORK);
-done:
- *logflagsp |= rval;
- return error;
-#undef LEFT
-#undef RIGHT
-#undef PREV
-}
+ mp = ip->i_mount;
+ ifp = XFS_IFORK_PTR(ip, whichfork);
+ ASSERT(ifp->if_flags & XFS_IFEXTENTS);
+ ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE);
+ rblock = ifp->if_broot;
+ ASSERT(be16_to_cpu(rblock->bb_level) == 1);
+ ASSERT(be16_to_cpu(rblock->bb_numrecs) == 1);
+ ASSERT(xfs_bmbt_maxrecs(mp, ifp->if_broot_bytes, 0) == 1);
+ pp = XFS_BMAP_BROOT_PTR_ADDR(mp, rblock, 1, ifp->if_broot_bytes);
+ cbno = be64_to_cpu(*pp);
+ *logflagsp = 0;
+#ifdef DEBUG
+ if ((error = xfs_btree_check_lptr(cur, cbno, 1)))
+ return error;
+#endif
+ error = xfs_btree_read_bufl(mp, tp, cbno, 0, &cbp, XFS_BMAP_BTREE_REF,
+ &xfs_bmbt_buf_ops);
+ if (error)
+ return error;
+ cblock = XFS_BUF_TO_BLOCK(cbp);
+ if ((error = xfs_btree_check_block(cur, cblock, 0, cbp)))
+ return error;
+ xfs_bmap_add_free(cbno, 1, cur->bc_private.b.flist, mp);
+ ip->i_d.di_nblocks--;
+ xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_BCOUNT, -1L);
+ xfs_trans_binval(tp, cbp);
+ if (cur->bc_bufs[0] == cbp)
+ cur->bc_bufs[0] = NULL;
+ xfs_iroot_realloc(ip, -1, whichfork);
+ ASSERT(ifp->if_broot == NULL);
+ ASSERT((ifp->if_flags & XFS_IFBROOT) == 0);
+ XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_EXTENTS);
+ *logflagsp = XFS_ILOG_CORE | xfs_ilog_fext(whichfork);
+ return 0;
+}
/*
- * Convert a hole to a delayed allocation.
+ * Convert an extents-format file into a btree-format file.
+ * The new file will have a root block (in the inode) and a single child block.
*/
-STATIC void
-xfs_bmap_add_extent_hole_delay(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_extnum_t *idx, /* extent number to update/insert */
- xfs_bmbt_irec_t *new) /* new data to add to file extents */
+STATIC int /* error */
+xfs_bmap_extents_to_btree(
+ xfs_trans_t *tp, /* transaction pointer */
+ xfs_inode_t *ip, /* incore inode pointer */
+ xfs_fsblock_t *firstblock, /* first-block-allocated */
+ xfs_bmap_free_t *flist, /* blocks freed in xaction */
+ xfs_btree_cur_t **curp, /* cursor returned to caller */
+ int wasdel, /* converting a delayed alloc */
+ int *logflagsp, /* inode logging flags */
+ int whichfork) /* data or attr fork */
{
- xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_bmbt_irec_t left; /* left neighbor extent entry */
- xfs_filblks_t newlen=0; /* new indirect size */
- xfs_filblks_t oldlen=0; /* old indirect size */
- xfs_bmbt_irec_t right; /* right neighbor extent entry */
- int state; /* state bits, accessed thru macros */
- xfs_filblks_t temp=0; /* temp for indirect calculations */
+ struct xfs_btree_block *ablock; /* allocated (child) bt block */
+ xfs_buf_t *abp; /* buffer for ablock */
+ xfs_alloc_arg_t args; /* allocation arguments */
+ xfs_bmbt_rec_t *arp; /* child record pointer */
+ struct xfs_btree_block *block; /* btree root block */
+ xfs_btree_cur_t *cur; /* bmap btree cursor */
+ xfs_bmbt_rec_host_t *ep; /* extent record pointer */
+ int error; /* error return value */
+ xfs_extnum_t i, cnt; /* extent record index */
+ xfs_ifork_t *ifp; /* inode fork pointer */
+ xfs_bmbt_key_t *kp; /* root block key pointer */
+ xfs_mount_t *mp; /* mount structure */
+ xfs_extnum_t nextents; /* number of file extents */
+ xfs_bmbt_ptr_t *pp; /* root block address pointer */
- ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
- state = 0;
- ASSERT(isnullstartblock(new->br_startblock));
+ ifp = XFS_IFORK_PTR(ip, whichfork);
+ ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS);
/*
- * Check and set flags if this segment has a left neighbor
+ * Make space in the inode incore.
*/
- if (*idx > 0) {
- state |= BMAP_LEFT_VALID;
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx - 1), &left);
-
- if (isnullstartblock(left.br_startblock))
- state |= BMAP_LEFT_DELAY;
- }
+ xfs_iroot_realloc(ip, 1, whichfork);
+ ifp->if_flags |= XFS_IFBROOT;
/*
- * Check and set flags if the current (right) segment exists.
- * If it doesn't exist, we're converting the hole at end-of-file.
+ * Fill in the root.
*/
- if (*idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t)) {
- state |= BMAP_RIGHT_VALID;
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx), &right);
+ block = ifp->if_broot;
+ block->bb_magic = cpu_to_be32(XFS_BMAP_MAGIC);
+ block->bb_level = cpu_to_be16(1);
+ block->bb_numrecs = cpu_to_be16(1);
+ block->bb_u.l.bb_leftsib = cpu_to_be64(NULLDFSBNO);
+ block->bb_u.l.bb_rightsib = cpu_to_be64(NULLDFSBNO);
- if (isnullstartblock(right.br_startblock))
- state |= BMAP_RIGHT_DELAY;
+ /*
+ * Need a cursor. Can't allocate until bb_level is filled in.
+ */
+ mp = ip->i_mount;
+ cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
+ cur->bc_private.b.firstblock = *firstblock;
+ cur->bc_private.b.flist = flist;
+ cur->bc_private.b.flags = wasdel ? XFS_BTCUR_BPRV_WASDEL : 0;
+ /*
+ * Convert to a btree with two levels, one record in root.
+ */
+ XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_BTREE);
+ memset(&args, 0, sizeof(args));
+ args.tp = tp;
+ args.mp = mp;
+ args.firstblock = *firstblock;
+ if (*firstblock == NULLFSBLOCK) {
+ args.type = XFS_ALLOCTYPE_START_BNO;
+ args.fsbno = XFS_INO_TO_FSB(mp, ip->i_ino);
+ } else if (flist->xbf_low) {
+ args.type = XFS_ALLOCTYPE_START_BNO;
+ args.fsbno = *firstblock;
+ } else {
+ args.type = XFS_ALLOCTYPE_NEAR_BNO;
+ args.fsbno = *firstblock;
+ }
+ args.minlen = args.maxlen = args.prod = 1;
+ args.wasdel = wasdel;
+ *logflagsp = 0;
+ if ((error = xfs_alloc_vextent(&args))) {
+ xfs_iroot_realloc(ip, -1, whichfork);
+ xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
+ return error;
}
-
/*
- * Set contiguity flags on the left and right neighbors.
- * Don't let extents get too large, even if the pieces are contiguous.
+ * Allocation can't fail, the space was reserved.
*/
- if ((state & BMAP_LEFT_VALID) && (state & BMAP_LEFT_DELAY) &&
- left.br_startoff + left.br_blockcount == new->br_startoff &&
- left.br_blockcount + new->br_blockcount <= MAXEXTLEN)
- state |= BMAP_LEFT_CONTIG;
-
- if ((state & BMAP_RIGHT_VALID) && (state & BMAP_RIGHT_DELAY) &&
- new->br_startoff + new->br_blockcount == right.br_startoff &&
- new->br_blockcount + right.br_blockcount <= MAXEXTLEN &&
- (!(state & BMAP_LEFT_CONTIG) ||
- (left.br_blockcount + new->br_blockcount +
- right.br_blockcount <= MAXEXTLEN)))
- state |= BMAP_RIGHT_CONTIG;
-
+ ASSERT(args.fsbno != NULLFSBLOCK);
+ ASSERT(*firstblock == NULLFSBLOCK ||
+ args.agno == XFS_FSB_TO_AGNO(mp, *firstblock) ||
+ (flist->xbf_low &&
+ args.agno > XFS_FSB_TO_AGNO(mp, *firstblock)));
+ *firstblock = cur->bc_private.b.firstblock = args.fsbno;
+ cur->bc_private.b.allocated++;
+ ip->i_d.di_nblocks++;
+ xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_BCOUNT, 1L);
+ abp = xfs_btree_get_bufl(mp, tp, args.fsbno, 0);
/*
- * Switch out based on the contiguity flags.
+ * Fill in the child block.
*/
- switch (state & (BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG)) {
- case BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
- /*
- * New allocation is contiguous with delayed allocations
- * on the left and on the right.
- * Merge all three into a single extent record.
- */
- --*idx;
- temp = left.br_blockcount + new->br_blockcount +
- right.br_blockcount;
-
- trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx), temp);
- oldlen = startblockval(left.br_startblock) +
- startblockval(new->br_startblock) +
- startblockval(right.br_startblock);
- newlen = xfs_bmap_worst_indlen(ip, temp);
- xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, *idx),
- nullstartblock((int)newlen));
- trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
+ abp->b_ops = &xfs_bmbt_buf_ops;
+ ablock = XFS_BUF_TO_BLOCK(abp);
+ ablock->bb_magic = cpu_to_be32(XFS_BMAP_MAGIC);
+ ablock->bb_level = 0;
+ ablock->bb_u.l.bb_leftsib = cpu_to_be64(NULLDFSBNO);
+ ablock->bb_u.l.bb_rightsib = cpu_to_be64(NULLDFSBNO);
+ arp = XFS_BMBT_REC_ADDR(mp, ablock, 1);
+ nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
+ for (cnt = i = 0; i < nextents; i++) {
+ ep = xfs_iext_get_ext(ifp, i);
+ if (!isnullstartblock(xfs_bmbt_get_startblock(ep))) {
+ arp->l0 = cpu_to_be64(ep->l0);
+ arp->l1 = cpu_to_be64(ep->l1);
+ arp++; cnt++;
+ }
+ }
+ ASSERT(cnt == XFS_IFORK_NEXTENTS(ip, whichfork));
+ xfs_btree_set_numrecs(ablock, cnt);
- xfs_iext_remove(ip, *idx + 1, 1, state);
- break;
+ /*
+ * Fill in the root key and pointer.
+ */
+ kp = XFS_BMBT_KEY_ADDR(mp, block, 1);
+ arp = XFS_BMBT_REC_ADDR(mp, ablock, 1);
+ kp->br_startoff = cpu_to_be64(xfs_bmbt_disk_get_startoff(arp));
+ pp = XFS_BMBT_PTR_ADDR(mp, block, 1, xfs_bmbt_get_maxrecs(cur,
+ be16_to_cpu(block->bb_level)));
+ *pp = cpu_to_be64(args.fsbno);
- case BMAP_LEFT_CONTIG:
- /*
- * New allocation is contiguous with a delayed allocation
- * on the left.
- * Merge the new allocation with the left neighbor.
- */
- --*idx;
- temp = left.br_blockcount + new->br_blockcount;
+ /*
+ * Do all this logging at the end so that
+ * the root is at the right level.
+ */
+ xfs_btree_log_block(cur, abp, XFS_BB_ALL_BITS);
+ xfs_btree_log_recs(cur, abp, 1, be16_to_cpu(ablock->bb_numrecs));
+ ASSERT(*curp == NULL);
+ *curp = cur;
+ *logflagsp = XFS_ILOG_CORE | xfs_ilog_fbroot(whichfork);
+ return 0;
+}
- trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx), temp);
- oldlen = startblockval(left.br_startblock) +
- startblockval(new->br_startblock);
- newlen = xfs_bmap_worst_indlen(ip, temp);
- xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, *idx),
- nullstartblock((int)newlen));
- trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- break;
+/*
+ * Convert a local file to an extents file.
+ * This code is out of bounds for data forks of regular files,
+ * since the file data needs to get logged so things will stay consistent.
+ * (The bmap-level manipulations are ok, though).
+ */
+STATIC int /* error */
+xfs_bmap_local_to_extents(
+ xfs_trans_t *tp, /* transaction pointer */
+ xfs_inode_t *ip, /* incore inode pointer */
+ xfs_fsblock_t *firstblock, /* first block allocated in xaction */
+ xfs_extlen_t total, /* total blocks needed by transaction */
+ int *logflagsp, /* inode logging flags */
+ int whichfork,
+ void (*init_fn)(struct xfs_buf *bp,
+ struct xfs_inode *ip,
+ struct xfs_ifork *ifp))
+{
+ int error; /* error return value */
+ int flags; /* logging flags returned */
+ xfs_ifork_t *ifp; /* inode fork pointer */
- case BMAP_RIGHT_CONTIG:
- /*
- * New allocation is contiguous with a delayed allocation
- * on the right.
- * Merge the new allocation with the right neighbor.
- */
- trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
- temp = new->br_blockcount + right.br_blockcount;
- oldlen = startblockval(new->br_startblock) +
- startblockval(right.br_startblock);
- newlen = xfs_bmap_worst_indlen(ip, temp);
- xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, *idx),
- new->br_startoff,
- nullstartblock((int)newlen), temp, right.br_state);
- trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- break;
+ /*
+ * We don't want to deal with the case of keeping inode data inline yet.
+ * So sending the data fork of a regular inode is invalid.
+ */
+ ASSERT(!(S_ISREG(ip->i_d.di_mode) && whichfork == XFS_DATA_FORK));
+ ifp = XFS_IFORK_PTR(ip, whichfork);
+ ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL);
+ flags = 0;
+ error = 0;
+ if (ifp->if_bytes) {
+ xfs_alloc_arg_t args; /* allocation arguments */
+ xfs_buf_t *bp; /* buffer for extent block */
+ xfs_bmbt_rec_host_t *ep;/* extent record pointer */
- case 0:
- /*
- * New allocation is not contiguous with another
- * delayed allocation.
- * Insert a new entry.
- */
- oldlen = newlen = 0;
- xfs_iext_insert(ip, *idx, 1, new, state);
- break;
- }
- if (oldlen != newlen) {
- ASSERT(oldlen > newlen);
- xfs_icsb_modify_counters(ip->i_mount, XFS_SBS_FDBLOCKS,
- (int64_t)(oldlen - newlen), 0);
+ ASSERT((ifp->if_flags &
+ (XFS_IFINLINE|XFS_IFEXTENTS|XFS_IFEXTIREC)) == XFS_IFINLINE);
+ memset(&args, 0, sizeof(args));
+ args.tp = tp;
+ args.mp = ip->i_mount;
+ args.firstblock = *firstblock;
/*
- * Nothing to do for disk quota accounting here.
+ * Allocate a block. We know we need only one, since the
+ * file currently fits in an inode.
*/
+ if (*firstblock == NULLFSBLOCK) {
+ args.fsbno = XFS_INO_TO_FSB(args.mp, ip->i_ino);
+ args.type = XFS_ALLOCTYPE_START_BNO;
+ } else {
+ args.fsbno = *firstblock;
+ args.type = XFS_ALLOCTYPE_NEAR_BNO;
+ }
+ args.total = total;
+ args.minlen = args.maxlen = args.prod = 1;
+ error = xfs_alloc_vextent(&args);
+ if (error)
+ goto done;
+
+ /* Can't fail, the space was reserved. */
+ ASSERT(args.fsbno != NULLFSBLOCK);
+ ASSERT(args.len == 1);
+ *firstblock = args.fsbno;
+ bp = xfs_btree_get_bufl(args.mp, tp, args.fsbno, 0);
+
+ /* initialise the block and copy the data */
+ init_fn(bp, ip, ifp);
+
+ /* account for the change in fork size and log everything */
+ xfs_trans_log_buf(tp, bp, 0, ifp->if_bytes - 1);
+ xfs_bmap_forkoff_reset(args.mp, ip, whichfork);
+ xfs_idata_realloc(ip, -ifp->if_bytes, whichfork);
+ xfs_iext_add(ifp, 0, 1);
+ ep = xfs_iext_get_ext(ifp, 0);
+ xfs_bmbt_set_allf(ep, 0, args.fsbno, 1, XFS_EXT_NORM);
+ trace_xfs_bmap_post_update(ip, 0,
+ whichfork == XFS_ATTR_FORK ? BMAP_ATTRFORK : 0,
+ _THIS_IP_);
+ XFS_IFORK_NEXT_SET(ip, whichfork, 1);
+ ip->i_d.di_nblocks = 1;
+ xfs_trans_mod_dquot_byino(tp, ip,
+ XFS_TRANS_DQ_BCOUNT, 1L);
+ flags |= xfs_ilog_fext(whichfork);
+ } else {
+ ASSERT(XFS_IFORK_NEXTENTS(ip, whichfork) == 0);
+ xfs_bmap_forkoff_reset(ip->i_mount, ip, whichfork);
}
+ ifp->if_flags &= ~XFS_IFINLINE;
+ ifp->if_flags |= XFS_IFEXTENTS;
+ XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_EXTENTS);
+ flags |= XFS_ILOG_CORE;
+done:
+ *logflagsp = flags;
+ return error;
}
/*
- * Convert a hole to a real allocation.
+ * Called from xfs_bmap_add_attrfork to handle btree format files.
*/
-STATIC int /* error */
-xfs_bmap_add_extent_hole_real(
- struct xfs_bmalloca *bma,
- int whichfork)
+STATIC int /* error */
+xfs_bmap_add_attrfork_btree(
+ xfs_trans_t *tp, /* transaction pointer */
+ xfs_inode_t *ip, /* incore inode pointer */
+ xfs_fsblock_t *firstblock, /* first block allocated */
+ xfs_bmap_free_t *flist, /* blocks to free at commit */
+ int *flags) /* inode logging flags */
{
- struct xfs_bmbt_irec *new = &bma->got;
- int error; /* error return value */
- int i; /* temp state */
- xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_bmbt_irec_t left; /* left neighbor extent entry */
- xfs_bmbt_irec_t right; /* right neighbor extent entry */
- int rval=0; /* return value (logging flags) */
- int state; /* state bits, accessed thru macros */
-
- ifp = XFS_IFORK_PTR(bma->ip, whichfork);
-
- ASSERT(bma->idx >= 0);
- ASSERT(bma->idx <= ifp->if_bytes / sizeof(struct xfs_bmbt_rec));
- ASSERT(!isnullstartblock(new->br_startblock));
- ASSERT(!bma->cur ||
- !(bma->cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL));
-
- XFS_STATS_INC(xs_add_exlist);
-
- state = 0;
- if (whichfork == XFS_ATTR_FORK)
- state |= BMAP_ATTRFORK;
+ xfs_btree_cur_t *cur; /* btree cursor */
+ int error; /* error return value */
+ xfs_mount_t *mp; /* file system mount struct */
+ int stat; /* newroot status */
- /*
- * Check and set flags if this segment has a left neighbor.
- */
- if (bma->idx > 0) {
- state |= BMAP_LEFT_VALID;
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx - 1), &left);
- if (isnullstartblock(left.br_startblock))
- state |= BMAP_LEFT_DELAY;
+ mp = ip->i_mount;
+ if (ip->i_df.if_broot_bytes <= XFS_IFORK_DSIZE(ip))
+ *flags |= XFS_ILOG_DBROOT;
+ else {
+ cur = xfs_bmbt_init_cursor(mp, tp, ip, XFS_DATA_FORK);
+ cur->bc_private.b.flist = flist;
+ cur->bc_private.b.firstblock = *firstblock;
+ if ((error = xfs_bmbt_lookup_ge(cur, 0, 0, 0, &stat)))
+ goto error0;
+ /* must be at least one entry */
+ XFS_WANT_CORRUPTED_GOTO(stat == 1, error0);
+ if ((error = xfs_btree_new_iroot(cur, flags, &stat)))
+ goto error0;
+ if (stat == 0) {
+ xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
+ return XFS_ERROR(ENOSPC);
+ }
+ *firstblock = cur->bc_private.b.firstblock;
+ cur->bc_private.b.allocated = 0;
+ xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
}
+ return 0;
+error0:
+ xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
+ return error;
+}
- /*
- * Check and set flags if this segment has a current value.
- * Not true if we're inserting into the "hole" at eof.
- */
- if (bma->idx < ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t)) {
- state |= BMAP_RIGHT_VALID;
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx), &right);
- if (isnullstartblock(right.br_startblock))
- state |= BMAP_RIGHT_DELAY;
+/*
+ * Called from xfs_bmap_add_attrfork to handle extents format files.
+ */
+STATIC int /* error */
+xfs_bmap_add_attrfork_extents(
+ xfs_trans_t *tp, /* transaction pointer */
+ xfs_inode_t *ip, /* incore inode pointer */
+ xfs_fsblock_t *firstblock, /* first block allocated */
+ xfs_bmap_free_t *flist, /* blocks to free at commit */
+ int *flags) /* inode logging flags */
+{
+ xfs_btree_cur_t *cur; /* bmap btree cursor */
+ int error; /* error return value */
+
+ if (ip->i_d.di_nextents * sizeof(xfs_bmbt_rec_t) <= XFS_IFORK_DSIZE(ip))
+ return 0;
+ cur = NULL;
+ error = xfs_bmap_extents_to_btree(tp, ip, firstblock, flist, &cur, 0,
+ flags, XFS_DATA_FORK);
+ if (cur) {
+ cur->bc_private.b.allocated = 0;
+ xfs_btree_del_cursor(cur,
+ error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR);
}
+ return error;
+}
- /*
- * We're inserting a real allocation between "left" and "right".
- * Set the contiguity flags. Don't let extents get too large.
- */
- if ((state & BMAP_LEFT_VALID) && !(state & BMAP_LEFT_DELAY) &&
- left.br_startoff + left.br_blockcount == new->br_startoff &&
- left.br_startblock + left.br_blockcount == new->br_startblock &&
- left.br_state == new->br_state &&
- left.br_blockcount + new->br_blockcount <= MAXEXTLEN)
- state |= BMAP_LEFT_CONTIG;
+/*
+ * Block initialisation functions for local to extent format conversion.
+ * As these get more complex, they will be moved to the relevant files,
+ * but for now they are too simple to worry about.
+ */
+STATIC void
+xfs_bmap_local_to_extents_init_fn(
+ struct xfs_buf *bp,
+ struct xfs_inode *ip,
+ struct xfs_ifork *ifp)
+{
+ bp->b_ops = &xfs_bmbt_buf_ops;
+ memcpy(bp->b_addr, ifp->if_u1.if_data, ifp->if_bytes);
+}
- if ((state & BMAP_RIGHT_VALID) && !(state & BMAP_RIGHT_DELAY) &&
- new->br_startoff + new->br_blockcount == right.br_startoff &&
- new->br_startblock + new->br_blockcount == right.br_startblock &&
- new->br_state == right.br_state &&
- new->br_blockcount + right.br_blockcount <= MAXEXTLEN &&
- (!(state & BMAP_LEFT_CONTIG) ||
- left.br_blockcount + new->br_blockcount +
- right.br_blockcount <= MAXEXTLEN))
- state |= BMAP_RIGHT_CONTIG;
+STATIC void
+xfs_symlink_local_to_remote(
+ struct xfs_buf *bp,
+ struct xfs_inode *ip,
+ struct xfs_ifork *ifp)
+{
+ /* remote symlink blocks are not verifiable until CRCs come along */
+ bp->b_ops = NULL;
+ memcpy(bp->b_addr, ifp->if_u1.if_data, ifp->if_bytes);
+}
- error = 0;
- /*
- * Select which case we're in here, and implement it.
- */
- switch (state & (BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG)) {
- case BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
- /*
- * New allocation is contiguous with real allocations on the
- * left and on the right.
- * Merge all three into a single extent record.
- */
- --bma->idx;
- trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, bma->idx),
- left.br_blockcount + new->br_blockcount +
- right.br_blockcount);
- trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+/*
+ * Called from xfs_bmap_add_attrfork to handle local format files. Each
+ * different data fork content type needs a different callout to do the
+ * conversion. Some are basic and only require special block initialisation
+ * callouts for the data formating, others (directories) are so specialised they
+ * handle everything themselves.
+ *
+ * XXX (dgc): investigate whether directory conversion can use the generic
+ * formatting callout. It should be possible - it's just a very complex
+ * formatter. it would also require passing the transaction through to the init
+ * function.
+ */
+STATIC int /* error */
+xfs_bmap_add_attrfork_local(
+ xfs_trans_t *tp, /* transaction pointer */
+ xfs_inode_t *ip, /* incore inode pointer */
+ xfs_fsblock_t *firstblock, /* first block allocated */
+ xfs_bmap_free_t *flist, /* blocks to free at commit */
+ int *flags) /* inode logging flags */
+{
+ xfs_da_args_t dargs; /* args for dir/attr code */
- xfs_iext_remove(bma->ip, bma->idx + 1, 1, state);
+ if (ip->i_df.if_bytes <= XFS_IFORK_DSIZE(ip))
+ return 0;
- XFS_IFORK_NEXT_SET(bma->ip, whichfork,
- XFS_IFORK_NEXTENTS(bma->ip, whichfork) - 1);
- if (bma->cur == NULL) {
- rval = XFS_ILOG_CORE | xfs_ilog_fext(whichfork);
- } else {
- rval = XFS_ILOG_CORE;
- error = xfs_bmbt_lookup_eq(bma->cur, right.br_startoff,
- right.br_startblock, right.br_blockcount,
- &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- error = xfs_btree_delete(bma->cur, &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- error = xfs_btree_decrement(bma->cur, 0, &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- error = xfs_bmbt_update(bma->cur, left.br_startoff,
- left.br_startblock,
- left.br_blockcount +
- new->br_blockcount +
- right.br_blockcount,
- left.br_state);
- if (error)
- goto done;
- }
- break;
+ if (S_ISDIR(ip->i_d.di_mode)) {
+ memset(&dargs, 0, sizeof(dargs));
+ dargs.dp = ip;
+ dargs.firstblock = firstblock;
+ dargs.flist = flist;
+ dargs.total = ip->i_mount->m_dirblkfsbs;
+ dargs.whichfork = XFS_DATA_FORK;
+ dargs.trans = tp;
+ return xfs_dir2_sf_to_block(&dargs);
+ }
- case BMAP_LEFT_CONTIG:
- /*
- * New allocation is contiguous with a real allocation
- * on the left.
- * Merge the new allocation with the left neighbor.
- */
- --bma->idx;
- trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, bma->idx),
- left.br_blockcount + new->br_blockcount);
- trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+ if (S_ISLNK(ip->i_d.di_mode))
+ return xfs_bmap_local_to_extents(tp, ip, firstblock, 1,
+ flags, XFS_DATA_FORK,
+ xfs_symlink_local_to_remote);
- if (bma->cur == NULL) {
- rval = xfs_ilog_fext(whichfork);
- } else {
- rval = 0;
- error = xfs_bmbt_lookup_eq(bma->cur, left.br_startoff,
- left.br_startblock, left.br_blockcount,
- &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- error = xfs_bmbt_update(bma->cur, left.br_startoff,
- left.br_startblock,
- left.br_blockcount +
- new->br_blockcount,
- left.br_state);
- if (error)
- goto done;
- }
- break;
+ return xfs_bmap_local_to_extents(tp, ip, firstblock, 1, flags,
+ XFS_DATA_FORK,
+ xfs_bmap_local_to_extents_init_fn);
+}
- case BMAP_RIGHT_CONTIG:
- /*
- * New allocation is contiguous with a real allocation
- * on the right.
- * Merge the new allocation with the right neighbor.
- */
- trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
- xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, bma->idx),
- new->br_startoff, new->br_startblock,
- new->br_blockcount + right.br_blockcount,
- right.br_state);
- trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+/*
+ * Convert inode from non-attributed to attributed.
+ * Must not be in a transaction, ip must not be locked.
+ */
+int /* error code */
+xfs_bmap_add_attrfork(
+ xfs_inode_t *ip, /* incore inode pointer */
+ int size, /* space new attribute needs */
+ int rsvd) /* xact may use reserved blks */
+{
+ xfs_fsblock_t firstblock; /* 1st block/ag allocated */
+ xfs_bmap_free_t flist; /* freed extent records */
+ xfs_mount_t *mp; /* mount structure */
+ xfs_trans_t *tp; /* transaction pointer */
+ int blks; /* space reservation */
+ int version = 1; /* superblock attr version */
+ int committed; /* xaction was committed */
+ int logflags; /* logging flags */
+ int error; /* error return value */
- if (bma->cur == NULL) {
- rval = xfs_ilog_fext(whichfork);
- } else {
- rval = 0;
- error = xfs_bmbt_lookup_eq(bma->cur,
- right.br_startoff,
- right.br_startblock,
- right.br_blockcount, &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- error = xfs_bmbt_update(bma->cur, new->br_startoff,
- new->br_startblock,
- new->br_blockcount +
- right.br_blockcount,
- right.br_state);
- if (error)
- goto done;
- }
- break;
+ ASSERT(XFS_IFORK_Q(ip) == 0);
- case 0:
+ mp = ip->i_mount;
+ ASSERT(!XFS_NOT_DQATTACHED(mp, ip));
+ tp = xfs_trans_alloc(mp, XFS_TRANS_ADDAFORK);
+ blks = XFS_ADDAFORK_SPACE_RES(mp);
+ if (rsvd)
+ tp->t_flags |= XFS_TRANS_RESERVE;
+ if ((error = xfs_trans_reserve(tp, blks, XFS_ADDAFORK_LOG_RES(mp), 0,
+ XFS_TRANS_PERM_LOG_RES, XFS_ADDAFORK_LOG_COUNT)))
+ goto error0;
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ error = xfs_trans_reserve_quota_nblks(tp, ip, blks, 0, rsvd ?
+ XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_FORCE_RES :
+ XFS_QMOPT_RES_REGBLKS);
+ if (error) {
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES);
+ return error;
+ }
+ if (XFS_IFORK_Q(ip))
+ goto error1;
+ if (ip->i_d.di_aformat != XFS_DINODE_FMT_EXTENTS) {
/*
- * New allocation is not contiguous with another
- * real allocation.
- * Insert a new entry.
+ * For inodes coming from pre-6.2 filesystems.
*/
- xfs_iext_insert(bma->ip, bma->idx, 1, new, state);
- XFS_IFORK_NEXT_SET(bma->ip, whichfork,
- XFS_IFORK_NEXTENTS(bma->ip, whichfork) + 1);
- if (bma->cur == NULL) {
- rval = XFS_ILOG_CORE | xfs_ilog_fext(whichfork);
- } else {
- rval = XFS_ILOG_CORE;
- error = xfs_bmbt_lookup_eq(bma->cur,
- new->br_startoff,
- new->br_startblock,
- new->br_blockcount, &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 0, done);
- bma->cur->bc_rec.b.br_state = new->br_state;
- error = xfs_btree_insert(bma->cur, &i);
- if (error)
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- }
- break;
+ ASSERT(ip->i_d.di_aformat == 0);
+ ip->i_d.di_aformat = XFS_DINODE_FMT_EXTENTS;
}
+ ASSERT(ip->i_d.di_anextents == 0);
- /* convert to a btree if necessary */
- if (xfs_bmap_needs_btree(bma->ip, whichfork)) {
- int tmp_logflags; /* partial log flag return val */
+ xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
+ xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
- ASSERT(bma->cur == NULL);
- error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
- bma->firstblock, bma->flist, &bma->cur,
- 0, &tmp_logflags, whichfork);
- bma->logflags |= tmp_logflags;
- if (error)
- goto done;
+ switch (ip->i_d.di_format) {
+ case XFS_DINODE_FMT_DEV:
+ ip->i_d.di_forkoff = roundup(sizeof(xfs_dev_t), 8) >> 3;
+ break;
+ case XFS_DINODE_FMT_UUID:
+ ip->i_d.di_forkoff = roundup(sizeof(uuid_t), 8) >> 3;
+ break;
+ case XFS_DINODE_FMT_LOCAL:
+ case XFS_DINODE_FMT_EXTENTS:
+ case XFS_DINODE_FMT_BTREE:
+ ip->i_d.di_forkoff = xfs_attr_shortform_bytesfit(ip, size);
+ if (!ip->i_d.di_forkoff)
+ ip->i_d.di_forkoff = xfs_default_attroffset(ip) >> 3;
+ else if (mp->m_flags & XFS_MOUNT_ATTR2)
+ version = 2;
+ break;
+ default:
+ ASSERT(0);
+ error = XFS_ERROR(EINVAL);
+ goto error1;
}
- /* clear out the allocated field, done with it now in any case. */
- if (bma->cur)
- bma->cur->bc_private.b.allocated = 0;
+ ASSERT(ip->i_afp == NULL);
+ ip->i_afp = kmem_zone_zalloc(xfs_ifork_zone, KM_SLEEP);
+ ip->i_afp->if_flags = XFS_IFEXTENTS;
+ logflags = 0;
+ xfs_bmap_init(&flist, &firstblock);
+ switch (ip->i_d.di_format) {
+ case XFS_DINODE_FMT_LOCAL:
+ error = xfs_bmap_add_attrfork_local(tp, ip, &firstblock, &flist,
+ &logflags);
+ break;
+ case XFS_DINODE_FMT_EXTENTS:
+ error = xfs_bmap_add_attrfork_extents(tp, ip, &firstblock,
+ &flist, &logflags);
+ break;
+ case XFS_DINODE_FMT_BTREE:
+ error = xfs_bmap_add_attrfork_btree(tp, ip, &firstblock, &flist,
+ &logflags);
+ break;
+ default:
+ error = 0;
+ break;
+ }
+ if (logflags)
+ xfs_trans_log_inode(tp, ip, logflags);
+ if (error)
+ goto error2;
+ if (!xfs_sb_version_hasattr(&mp->m_sb) ||
+ (!xfs_sb_version_hasattr2(&mp->m_sb) && version == 2)) {
+ __int64_t sbfields = 0;
- xfs_bmap_check_leaf_extents(bma->cur, bma->ip, whichfork);
-done:
- bma->logflags |= rval;
+ spin_lock(&mp->m_sb_lock);
+ if (!xfs_sb_version_hasattr(&mp->m_sb)) {
+ xfs_sb_version_addattr(&mp->m_sb);
+ sbfields |= XFS_SB_VERSIONNUM;
+ }
+ if (!xfs_sb_version_hasattr2(&mp->m_sb) && version == 2) {
+ xfs_sb_version_addattr2(&mp->m_sb);
+ sbfields |= (XFS_SB_VERSIONNUM | XFS_SB_FEATURES2);
+ }
+ if (sbfields) {
+ spin_unlock(&mp->m_sb_lock);
+ xfs_mod_sb(tp, sbfields);
+ } else
+ spin_unlock(&mp->m_sb_lock);
+ }
+
+ error = xfs_bmap_finish(&tp, &flist, &committed);
+ if (error)
+ goto error2;
+ return xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
+error2:
+ xfs_bmap_cancel(&flist);
+error1:
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+error0:
+ xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT);
return error;
}
/*
- * Adjust the size of the new extent based on di_extsize and rt extsize.
- */
-STATIC int
-xfs_bmap_extsize_align(
- xfs_mount_t *mp,
- xfs_bmbt_irec_t *gotp, /* next extent pointer */
- xfs_bmbt_irec_t *prevp, /* previous extent pointer */
- xfs_extlen_t extsz, /* align to this extent size */
- int rt, /* is this a realtime inode? */
- int eof, /* is extent at end-of-file? */
- int delay, /* creating delalloc extent? */
- int convert, /* overwriting unwritten extent? */
- xfs_fileoff_t *offp, /* in/out: aligned offset */
- xfs_extlen_t *lenp) /* in/out: aligned length */
-{
- xfs_fileoff_t orig_off; /* original offset */
- xfs_extlen_t orig_alen; /* original length */
- xfs_fileoff_t orig_end; /* original off+len */
- xfs_fileoff_t nexto; /* next file offset */
- xfs_fileoff_t prevo; /* previous file offset */
- xfs_fileoff_t align_off; /* temp for offset */
- xfs_extlen_t align_alen; /* temp for length */
- xfs_extlen_t temp; /* temp for calculations */
-
- if (convert)
- return 0;
-
- orig_off = align_off = *offp;
- orig_alen = align_alen = *lenp;
- orig_end = orig_off + orig_alen;
+ * Internal and external extent tree search functions.
+ */
- /*
- * If this request overlaps an existing extent, then don't
- * attempt to perform any additional alignment.
- */
- if (!delay && !eof &&
- (orig_off >= gotp->br_startoff) &&
- (orig_end <= gotp->br_startoff + gotp->br_blockcount)) {
- return 0;
- }
+/*
+ * Read in the extents to if_extents.
+ * All inode fields are set up by caller, we just traverse the btree
+ * and copy the records in. If the file system cannot contain unwritten
+ * extents, the records are checked for no "state" flags.
+ */
+int /* error */
+xfs_bmap_read_extents(
+ xfs_trans_t *tp, /* transaction pointer */
+ xfs_inode_t *ip, /* incore inode */
+ int whichfork) /* data or attr fork */
+{
+ struct xfs_btree_block *block; /* current btree block */
+ xfs_fsblock_t bno; /* block # of "block" */
+ xfs_buf_t *bp; /* buffer for "block" */
+ int error; /* error return value */
+ xfs_exntfmt_t exntf; /* XFS_EXTFMT_NOSTATE, if checking */
+ xfs_extnum_t i, j; /* index into the extents list */
+ xfs_ifork_t *ifp; /* fork structure */
+ int level; /* btree level, for checking */
+ xfs_mount_t *mp; /* file system mount structure */
+ __be64 *pp; /* pointer to block address */
+ /* REFERENCED */
+ xfs_extnum_t room; /* number of entries there's room for */
+ bno = NULLFSBLOCK;
+ mp = ip->i_mount;
+ ifp = XFS_IFORK_PTR(ip, whichfork);
+ exntf = (whichfork != XFS_DATA_FORK) ? XFS_EXTFMT_NOSTATE :
+ XFS_EXTFMT_INODE(ip);
+ block = ifp->if_broot;
/*
- * If the file offset is unaligned vs. the extent size
- * we need to align it. This will be possible unless
- * the file was previously written with a kernel that didn't
- * perform this alignment, or if a truncate shot us in the
- * foot.
+ * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out.
*/
- temp = do_mod(orig_off, extsz);
- if (temp) {
- align_alen += temp;
- align_off -= temp;
- }
+ level = be16_to_cpu(block->bb_level);
+ ASSERT(level > 0);
+ pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
+ bno = be64_to_cpu(*pp);
+ ASSERT(bno != NULLDFSBNO);
+ ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount);
+ ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks);
/*
- * Same adjustment for the end of the requested area.
+ * Go down the tree until leaf level is reached, following the first
+ * pointer (leftmost) at each level.
*/
- if ((temp = (align_alen % extsz))) {
- align_alen += extsz - temp;
+ while (level-- > 0) {
+ error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp,
+ XFS_BMAP_BTREE_REF, &xfs_bmbt_buf_ops);
+ if (error)
+ return error;
+ block = XFS_BUF_TO_BLOCK(bp);
+ XFS_WANT_CORRUPTED_GOTO(
+ xfs_bmap_sanity_check(mp, bp, level),
+ error0);
+ if (level == 0)
+ break;
+ pp = XFS_BMBT_PTR_ADDR(mp, block, 1, mp->m_bmap_dmxr[1]);
+ bno = be64_to_cpu(*pp);
+ XFS_WANT_CORRUPTED_GOTO(XFS_FSB_SANITY_CHECK(mp, bno), error0);
+ xfs_trans_brelse(tp, bp);
}
/*
- * If the previous block overlaps with this proposed allocation
- * then move the start forward without adjusting the length.
- */
- if (prevp->br_startoff != NULLFILEOFF) {
- if (prevp->br_startblock == HOLESTARTBLOCK)
- prevo = prevp->br_startoff;
- else
- prevo = prevp->br_startoff + prevp->br_blockcount;
- } else
- prevo = 0;
- if (align_off != orig_off && align_off < prevo)
- align_off = prevo;
- /*
- * If the next block overlaps with this proposed allocation
- * then move the start back without adjusting the length,
- * but not before offset 0.
- * This may of course make the start overlap previous block,
- * and if we hit the offset 0 limit then the next block
- * can still overlap too.
+ * Here with bp and block set to the leftmost leaf node in the tree.
*/
- if (!eof && gotp->br_startoff != NULLFILEOFF) {
- if ((delay && gotp->br_startblock == HOLESTARTBLOCK) ||
- (!delay && gotp->br_startblock == DELAYSTARTBLOCK))
- nexto = gotp->br_startoff + gotp->br_blockcount;
- else
- nexto = gotp->br_startoff;
- } else
- nexto = NULLFILEOFF;
- if (!eof &&
- align_off + align_alen != orig_end &&
- align_off + align_alen > nexto)
- align_off = nexto > align_alen ? nexto - align_alen : 0;
+ room = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
+ i = 0;
/*
- * If we're now overlapping the next or previous extent that
- * means we can't fit an extsz piece in this hole. Just move
- * the start forward to the first valid spot and set
- * the length so we hit the end.
+ * Loop over all leaf nodes. Copy information to the extent records.
*/
- if (align_off != orig_off && align_off < prevo)
- align_off = prevo;
- if (align_off + align_alen != orig_end &&
- align_off + align_alen > nexto &&
- nexto != NULLFILEOFF) {
- ASSERT(nexto > prevo);
- align_alen = nexto - align_off;
- }
+ for (;;) {
+ xfs_bmbt_rec_t *frp;
+ xfs_fsblock_t nextbno;
+ xfs_extnum_t num_recs;
+ xfs_extnum_t start;
- /*
- * If realtime, and the result isn't a multiple of the realtime
- * extent size we need to remove blocks until it is.
- */
- if (rt && (temp = (align_alen % mp->m_sb.sb_rextsize))) {
- /*
- * We're not covering the original request, or
- * we won't be able to once we fix the length.
- */
- if (orig_off < align_off ||
- orig_end > align_off + align_alen ||
- align_alen - temp < orig_alen)
- return XFS_ERROR(EINVAL);
- /*
- * Try to fix it by moving the start up.
- */
- if (align_off + temp <= orig_off) {
- align_alen -= temp;
- align_off += temp;
+ num_recs = xfs_btree_get_numrecs(block);
+ if (unlikely(i + num_recs > room)) {
+ ASSERT(i + num_recs <= room);
+ xfs_warn(ip->i_mount,
+ "corrupt dinode %Lu, (btree extents).",
+ (unsigned long long) ip->i_ino);
+ XFS_CORRUPTION_ERROR("xfs_bmap_read_extents(1)",
+ XFS_ERRLEVEL_LOW, ip->i_mount, block);
+ goto error0;
}
+ XFS_WANT_CORRUPTED_GOTO(
+ xfs_bmap_sanity_check(mp, bp, 0),
+ error0);
/*
- * Try to fix it by moving the end in.
+ * Read-ahead the next leaf block, if any.
*/
- else if (align_off + align_alen - temp >= orig_end)
- align_alen -= temp;
+ nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
+ if (nextbno != NULLFSBLOCK)
+ xfs_btree_reada_bufl(mp, nextbno, 1,
+ &xfs_bmbt_buf_ops);
/*
- * Set the start to the minimum then trim the length.
+ * Copy records into the extent records.
*/
- else {
- align_alen -= orig_off - align_off;
- align_off = orig_off;
- align_alen -= align_alen % mp->m_sb.sb_rextsize;
+ frp = XFS_BMBT_REC_ADDR(mp, block, 1);
+ start = i;
+ for (j = 0; j < num_recs; j++, i++, frp++) {
+ xfs_bmbt_rec_host_t *trp = xfs_iext_get_ext(ifp, i);
+ trp->l0 = be64_to_cpu(frp->l0);
+ trp->l1 = be64_to_cpu(frp->l1);
+ }
+ if (exntf == XFS_EXTFMT_NOSTATE) {
+ /*
+ * Check all attribute bmap btree records and
+ * any "older" data bmap btree records for a
+ * set bit in the "extent flag" position.
+ */
+ if (unlikely(xfs_check_nostate_extents(ifp,
+ start, num_recs))) {
+ XFS_ERROR_REPORT("xfs_bmap_read_extents(2)",
+ XFS_ERRLEVEL_LOW,
+ ip->i_mount);
+ goto error0;
+ }
}
+ xfs_trans_brelse(tp, bp);
+ bno = nextbno;
/*
- * Result doesn't cover the request, fail it.
+ * If we've reached the end, stop.
*/
- if (orig_off < align_off || orig_end > align_off + align_alen)
- return XFS_ERROR(EINVAL);
- } else {
- ASSERT(orig_off >= align_off);
- ASSERT(orig_end <= align_off + align_alen);
+ if (bno == NULLFSBLOCK)
+ break;
+ error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp,
+ XFS_BMAP_BTREE_REF, &xfs_bmbt_buf_ops);
+ if (error)
+ return error;
+ block = XFS_BUF_TO_BLOCK(bp);
}
+ ASSERT(i == (ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t)));
+ ASSERT(i == XFS_IFORK_NEXTENTS(ip, whichfork));
+ XFS_BMAP_TRACE_EXLIST(ip, i, whichfork);
+ return 0;
+error0:
+ xfs_trans_brelse(tp, bp);
+ return XFS_ERROR(EFSCORRUPTED);
+}
-#ifdef DEBUG
- if (!eof && gotp->br_startoff != NULLFILEOFF)
- ASSERT(align_off + align_alen <= gotp->br_startoff);
- if (prevp->br_startoff != NULLFILEOFF)
- ASSERT(align_off >= prevp->br_startoff + prevp->br_blockcount);
+
+/*
+ * Search the extent records for the entry containing block bno.
+ * If bno lies in a hole, point to the next entry. If bno lies
+ * past eof, *eofp will be set, and *prevp will contain the last
+ * entry (null if none). Else, *lastxp will be set to the index
+ * of the found entry; *gotp will contain the entry.
+ */
+STATIC xfs_bmbt_rec_host_t * /* pointer to found extent entry */
+xfs_bmap_search_multi_extents(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ xfs_fileoff_t bno, /* block number searched for */
+ int *eofp, /* out: end of file found */
+ xfs_extnum_t *lastxp, /* out: last extent index */
+ xfs_bmbt_irec_t *gotp, /* out: extent entry found */
+ xfs_bmbt_irec_t *prevp) /* out: previous extent entry found */
+{
+ xfs_bmbt_rec_host_t *ep; /* extent record pointer */
+ xfs_extnum_t lastx; /* last extent index */
+
+ /*
+ * Initialize the extent entry structure to catch access to
+ * uninitialized br_startblock field.
+ */
+ gotp->br_startoff = 0xffa5a5a5a5a5a5a5LL;
+ gotp->br_blockcount = 0xa55a5a5a5a5a5a5aLL;
+ gotp->br_state = XFS_EXT_INVALID;
+#if XFS_BIG_BLKNOS
+ gotp->br_startblock = 0xffffa5a5a5a5a5a5LL;
+#else
+ gotp->br_startblock = 0xffffa5a5;
#endif
+ prevp->br_startoff = NULLFILEOFF;
- *lenp = align_alen;
- *offp = align_off;
- return 0;
+ ep = xfs_iext_bno_to_ext(ifp, bno, &lastx);
+ if (lastx > 0) {
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, lastx - 1), prevp);
+ }
+ if (lastx < (ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t))) {
+ xfs_bmbt_get_all(ep, gotp);
+ *eofp = 0;
+ } else {
+ if (lastx > 0) {
+ *gotp = *prevp;
+ }
+ *eofp = 1;
+ ep = NULL;
+ }
+ *lastxp = lastx;
+ return ep;
}
-#define XFS_ALLOC_GAP_UNITS 4
-
-STATIC void
-xfs_bmap_adjacent(
- xfs_bmalloca_t *ap) /* bmap alloc argument struct */
+/*
+ * Search the extents list for the inode, for the extent containing bno.
+ * If bno lies in a hole, point to the next entry. If bno lies past eof,
+ * *eofp will be set, and *prevp will contain the last entry (null if none).
+ * Else, *lastxp will be set to the index of the found
+ * entry; *gotp will contain the entry.
+ */
+xfs_bmbt_rec_host_t * /* pointer to found extent entry */
+xfs_bmap_search_extents(
+ xfs_inode_t *ip, /* incore inode pointer */
+ xfs_fileoff_t bno, /* block number searched for */
+ int fork, /* data or attr fork */
+ int *eofp, /* out: end of file found */
+ xfs_extnum_t *lastxp, /* out: last extent index */
+ xfs_bmbt_irec_t *gotp, /* out: extent entry found */
+ xfs_bmbt_irec_t *prevp) /* out: previous extent entry found */
{
- xfs_fsblock_t adjust; /* adjustment to block numbers */
- xfs_agnumber_t fb_agno; /* ag number of ap->firstblock */
- xfs_mount_t *mp; /* mount point structure */
- int nullfb; /* true if ap->firstblock isn't set */
- int rt; /* true if inode is realtime */
+ xfs_ifork_t *ifp; /* inode fork pointer */
+ xfs_bmbt_rec_host_t *ep; /* extent record pointer */
-#define ISVALID(x,y) \
- (rt ? \
- (x) < mp->m_sb.sb_rblocks : \
- XFS_FSB_TO_AGNO(mp, x) == XFS_FSB_TO_AGNO(mp, y) && \
- XFS_FSB_TO_AGNO(mp, x) < mp->m_sb.sb_agcount && \
- XFS_FSB_TO_AGBNO(mp, x) < mp->m_sb.sb_agblocks)
+ XFS_STATS_INC(xs_look_exlist);
+ ifp = XFS_IFORK_PTR(ip, fork);
- mp = ap->ip->i_mount;
- nullfb = *ap->firstblock == NULLFSBLOCK;
- rt = XFS_IS_REALTIME_INODE(ap->ip) && ap->userdata;
- fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp, *ap->firstblock);
- /*
- * If allocating at eof, and there's a previous real block,
- * try to use its last block as our starting point.
- */
- if (ap->eof && ap->prev.br_startoff != NULLFILEOFF &&
- !isnullstartblock(ap->prev.br_startblock) &&
- ISVALID(ap->prev.br_startblock + ap->prev.br_blockcount,
- ap->prev.br_startblock)) {
- ap->blkno = ap->prev.br_startblock + ap->prev.br_blockcount;
- /*
- * Adjust for the gap between prevp and us.
- */
- adjust = ap->offset -
- (ap->prev.br_startoff + ap->prev.br_blockcount);
- if (adjust &&
- ISVALID(ap->blkno + adjust, ap->prev.br_startblock))
- ap->blkno += adjust;
+ ep = xfs_bmap_search_multi_extents(ifp, bno, eofp, lastxp, gotp, prevp);
+
+ if (unlikely(!(gotp->br_startblock) && (*lastxp != NULLEXTNUM) &&
+ !(XFS_IS_REALTIME_INODE(ip) && fork == XFS_DATA_FORK))) {
+ xfs_alert_tag(ip->i_mount, XFS_PTAG_FSBLOCK_ZERO,
+ "Access to block zero in inode %llu "
+ "start_block: %llx start_off: %llx "
+ "blkcnt: %llx extent-state: %x lastx: %x\n",
+ (unsigned long long)ip->i_ino,
+ (unsigned long long)gotp->br_startblock,
+ (unsigned long long)gotp->br_startoff,
+ (unsigned long long)gotp->br_blockcount,
+ gotp->br_state, *lastxp);
+ *lastxp = NULLEXTNUM;
+ *eofp = 1;
+ return NULL;
}
- /*
- * If not at eof, then compare the two neighbor blocks.
- * Figure out whether either one gives us a good starting point,
- * and pick the better one.
- */
- else if (!ap->eof) {
- xfs_fsblock_t gotbno; /* right side block number */
- xfs_fsblock_t gotdiff=0; /* right side difference */
- xfs_fsblock_t prevbno; /* left side block number */
- xfs_fsblock_t prevdiff=0; /* left side difference */
+ return ep;
+}
+/*
+ * Returns the file-relative block number of the first unused block(s)
+ * in the file with at least "len" logically contiguous blocks free.
+ * This is the lowest-address hole if the file has holes, else the first block
+ * past the end of file.
+ * Return 0 if the file is currently local (in-inode).
+ */
+int /* error */
+xfs_bmap_first_unused(
+ xfs_trans_t *tp, /* transaction pointer */
+ xfs_inode_t *ip, /* incore inode */
+ xfs_extlen_t len, /* size of hole to find */
+ xfs_fileoff_t *first_unused, /* unused block */
+ int whichfork) /* data or attr fork */
+{
+ int error; /* error return value */
+ int idx; /* extent record index */
+ xfs_ifork_t *ifp; /* inode fork pointer */
+ xfs_fileoff_t lastaddr; /* last block number seen */
+ xfs_fileoff_t lowest; /* lowest useful block */
+ xfs_fileoff_t max; /* starting useful block */
+ xfs_fileoff_t off; /* offset for this block */
+ xfs_extnum_t nextents; /* number of extent entries */
+
+ ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE ||
+ XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS ||
+ XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL);
+ if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL) {
+ *first_unused = 0;
+ return 0;
+ }
+ ifp = XFS_IFORK_PTR(ip, whichfork);
+ if (!(ifp->if_flags & XFS_IFEXTENTS) &&
+ (error = xfs_iread_extents(tp, ip, whichfork)))
+ return error;
+ lowest = *first_unused;
+ nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
+ for (idx = 0, lastaddr = 0, max = lowest; idx < nextents; idx++) {
+ xfs_bmbt_rec_host_t *ep = xfs_iext_get_ext(ifp, idx);
+ off = xfs_bmbt_get_startoff(ep);
/*
- * If there's a previous (left) block, select a requested
- * start block based on it.
- */
- if (ap->prev.br_startoff != NULLFILEOFF &&
- !isnullstartblock(ap->prev.br_startblock) &&
- (prevbno = ap->prev.br_startblock +
- ap->prev.br_blockcount) &&
- ISVALID(prevbno, ap->prev.br_startblock)) {
- /*
- * Calculate gap to end of previous block.
- */
- adjust = prevdiff = ap->offset -
- (ap->prev.br_startoff +
- ap->prev.br_blockcount);
- /*
- * Figure the startblock based on the previous block's
- * end and the gap size.
- * Heuristic!
- * If the gap is large relative to the piece we're
- * allocating, or using it gives us an invalid block
- * number, then just use the end of the previous block.
- */
- if (prevdiff <= XFS_ALLOC_GAP_UNITS * ap->length &&
- ISVALID(prevbno + prevdiff,
- ap->prev.br_startblock))
- prevbno += adjust;
- else
- prevdiff += adjust;
- /*
- * If the firstblock forbids it, can't use it,
- * must use default.
- */
- if (!rt && !nullfb &&
- XFS_FSB_TO_AGNO(mp, prevbno) != fb_agno)
- prevbno = NULLFSBLOCK;
- }
- /*
- * No previous block or can't follow it, just default.
- */
- else
- prevbno = NULLFSBLOCK;
- /*
- * If there's a following (right) block, select a requested
- * start block based on it.
+ * See if the hole before this extent will work.
*/
- if (!isnullstartblock(ap->got.br_startblock)) {
- /*
- * Calculate gap to start of next block.
- */
- adjust = gotdiff = ap->got.br_startoff - ap->offset;
- /*
- * Figure the startblock based on the next block's
- * start and the gap size.
- */
- gotbno = ap->got.br_startblock;
- /*
- * Heuristic!
- * If the gap is large relative to the piece we're
- * allocating, or using it gives us an invalid block
- * number, then just use the start of the next block
- * offset by our length.
- */
- if (gotdiff <= XFS_ALLOC_GAP_UNITS * ap->length &&
- ISVALID(gotbno - gotdiff, gotbno))
- gotbno -= adjust;
- else if (ISVALID(gotbno - ap->length, gotbno)) {
- gotbno -= ap->length;
- gotdiff += adjust - ap->length;
- } else
- gotdiff += adjust;
- /*
- * If the firstblock forbids it, can't use it,
- * must use default.
- */
- if (!rt && !nullfb &&
- XFS_FSB_TO_AGNO(mp, gotbno) != fb_agno)
- gotbno = NULLFSBLOCK;
+ if (off >= lowest + len && off - max >= len) {
+ *first_unused = max;
+ return 0;
}
- /*
- * No next block, just default.
- */
+ lastaddr = off + xfs_bmbt_get_blockcount(ep);
+ max = XFS_FILEOFF_MAX(lastaddr, lowest);
+ }
+ *first_unused = max;
+ return 0;
+}
+
+/*
+ * Returns the file-relative block number of the last block + 1 before
+ * last_block (input value) in the file.
+ * This is not based on i_size, it is based on the extent records.
+ * Returns 0 for local files, as they do not have extent records.
+ */
+int /* error */
+xfs_bmap_last_before(
+ xfs_trans_t *tp, /* transaction pointer */
+ xfs_inode_t *ip, /* incore inode */
+ xfs_fileoff_t *last_block, /* last block */
+ int whichfork) /* data or attr fork */
+{
+ xfs_fileoff_t bno; /* input file offset */
+ int eof; /* hit end of file */
+ xfs_bmbt_rec_host_t *ep; /* pointer to last extent */
+ int error; /* error return value */
+ xfs_bmbt_irec_t got; /* current extent value */
+ xfs_ifork_t *ifp; /* inode fork pointer */
+ xfs_extnum_t lastx; /* last extent used */
+ xfs_bmbt_irec_t prev; /* previous extent value */
+
+ if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE &&
+ XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
+ XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_LOCAL)
+ return XFS_ERROR(EIO);
+ if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL) {
+ *last_block = 0;
+ return 0;
+ }
+ ifp = XFS_IFORK_PTR(ip, whichfork);
+ if (!(ifp->if_flags & XFS_IFEXTENTS) &&
+ (error = xfs_iread_extents(tp, ip, whichfork)))
+ return error;
+ bno = *last_block - 1;
+ ep = xfs_bmap_search_extents(ip, bno, whichfork, &eof, &lastx, &got,
+ &prev);
+ if (eof || xfs_bmbt_get_startoff(ep) > bno) {
+ if (prev.br_startoff == NULLFILEOFF)
+ *last_block = 0;
else
- gotbno = NULLFSBLOCK;
- /*
- * If both valid, pick the better one, else the only good
- * one, else ap->rval is already set (to 0 or the inode block).
- */
- if (prevbno != NULLFSBLOCK && gotbno != NULLFSBLOCK)
- ap->blkno = prevdiff <= gotdiff ? prevbno : gotbno;
- else if (prevbno != NULLFSBLOCK)
- ap->blkno = prevbno;
- else if (gotbno != NULLFSBLOCK)
- ap->blkno = gotbno;
+ *last_block = prev.br_startoff + prev.br_blockcount;
}
-#undef ISVALID
+ /*
+ * Otherwise *last_block is already the right answer.
+ */
+ return 0;
}
STATIC int
-xfs_bmap_btalloc_nullfb(
- struct xfs_bmalloca *ap,
- struct xfs_alloc_arg *args,
- xfs_extlen_t *blen)
+xfs_bmap_last_extent(
+ struct xfs_trans *tp,
+ struct xfs_inode *ip,
+ int whichfork,
+ struct xfs_bmbt_irec *rec,
+ int *is_empty)
{
- struct xfs_mount *mp = ap->ip->i_mount;
- struct xfs_perag *pag;
- xfs_agnumber_t ag, startag;
- int notinit = 0;
+ struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork);
int error;
+ int nextents;
- if (ap->userdata && xfs_inode_is_filestream(ap->ip))
- args->type = XFS_ALLOCTYPE_NEAR_BNO;
- else
- args->type = XFS_ALLOCTYPE_START_BNO;
- args->total = ap->total;
+ if (!(ifp->if_flags & XFS_IFEXTENTS)) {
+ error = xfs_iread_extents(tp, ip, whichfork);
+ if (error)
+ return error;
+ }
+
+ nextents = ifp->if_bytes / sizeof(xfs_bmbt_rec_t);
+ if (nextents == 0) {
+ *is_empty = 1;
+ return 0;
+ }
+
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, nextents - 1), rec);
+ *is_empty = 0;
+ return 0;
+}
+
+/*
+ * Check the last inode extent to determine whether this allocation will result
+ * in blocks being allocated at the end of the file. When we allocate new data
+ * blocks at the end of the file which do not start at the previous data block,
+ * we will try to align the new blocks at stripe unit boundaries.
+ *
+ * Returns 0 in bma->aeof if the file (fork) is empty as any new write will be
+ * at, or past the EOF.
+ */
+STATIC int
+xfs_bmap_isaeof(
+ struct xfs_bmalloca *bma,
+ int whichfork)
+{
+ struct xfs_bmbt_irec rec;
+ int is_empty;
+ int error;
+
+ bma->aeof = 0;
+ error = xfs_bmap_last_extent(NULL, bma->ip, whichfork, &rec,
+ &is_empty);
+ if (error || is_empty)
+ return error;
/*
- * Search for an allocation group with a single extent large enough
- * for the request. If one isn't found, then adjust the minimum
- * allocation size to the largest space found.
+ * Check if we are allocation or past the last extent, or at least into
+ * the last delayed allocated extent.
*/
- startag = ag = XFS_FSB_TO_AGNO(mp, args->fsbno);
- if (startag == NULLAGNUMBER)
- startag = ag = 0;
+ bma->aeof = bma->offset >= rec.br_startoff + rec.br_blockcount ||
+ (bma->offset >= rec.br_startoff &&
+ isnullstartblock(rec.br_startblock));
+ return 0;
+}
- pag = xfs_perag_get(mp, ag);
- while (*blen < args->maxlen) {
- if (!pag->pagf_init) {
- error = xfs_alloc_pagf_init(mp, args->tp, ag,
- XFS_ALLOC_FLAG_TRYLOCK);
- if (error) {
- xfs_perag_put(pag);
- return error;
- }
- }
+/*
+ * Check if the endoff is outside the last extent. If so the caller will grow
+ * the allocation to a stripe unit boundary. All offsets are considered outside
+ * the end of file for an empty fork, so 1 is returned in *eof in that case.
+ */
+int
+xfs_bmap_eof(
+ struct xfs_inode *ip,
+ xfs_fileoff_t endoff,
+ int whichfork,
+ int *eof)
+{
+ struct xfs_bmbt_irec rec;
+ int error;
- /*
- * See xfs_alloc_fix_freelist...
- */
- if (pag->pagf_init) {
- xfs_extlen_t longest;
- longest = xfs_alloc_longest_free_extent(mp, pag);
- if (*blen < longest)
- *blen = longest;
- } else
- notinit = 1;
+ error = xfs_bmap_last_extent(NULL, ip, whichfork, &rec, eof);
+ if (error || *eof)
+ return error;
- if (xfs_inode_is_filestream(ap->ip)) {
- if (*blen >= args->maxlen)
- break;
+ *eof = endoff >= rec.br_startoff + rec.br_blockcount;
+ return 0;
+}
- if (ap->userdata) {
- /*
- * If startag is an invalid AG, we've
- * come here once before and
- * xfs_filestream_new_ag picked the
- * best currently available.
- *
- * Don't continue looping, since we
- * could loop forever.
- */
- if (startag == NULLAGNUMBER)
- break;
+/*
+ * Returns the file-relative block number of the first block past eof in
+ * the file. This is not based on i_size, it is based on the extent records.
+ * Returns 0 for local files, as they do not have extent records.
+ */
+int
+xfs_bmap_last_offset(
+ struct xfs_trans *tp,
+ struct xfs_inode *ip,
+ xfs_fileoff_t *last_block,
+ int whichfork)
+{
+ struct xfs_bmbt_irec rec;
+ int is_empty;
+ int error;
+
+ *last_block = 0;
+
+ if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL)
+ return 0;
+
+ if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE &&
+ XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS)
+ return XFS_ERROR(EIO);
+
+ error = xfs_bmap_last_extent(NULL, ip, whichfork, &rec, &is_empty);
+ if (error || is_empty)
+ return error;
+
+ *last_block = rec.br_startoff + rec.br_blockcount;
+ return 0;
+}
+
+/*
+ * Returns whether the selected fork of the inode has exactly one
+ * block or not. For the data fork we check this matches di_size,
+ * implying the file's range is 0..bsize-1.
+ */
+int /* 1=>1 block, 0=>otherwise */
+xfs_bmap_one_block(
+ xfs_inode_t *ip, /* incore inode */
+ int whichfork) /* data or attr fork */
+{
+ xfs_bmbt_rec_host_t *ep; /* ptr to fork's extent */
+ xfs_ifork_t *ifp; /* inode fork pointer */
+ int rval; /* return value */
+ xfs_bmbt_irec_t s; /* internal version of extent */
+
+#ifndef DEBUG
+ if (whichfork == XFS_DATA_FORK)
+ return XFS_ISIZE(ip) == ip->i_mount->m_sb.sb_blocksize;
+#endif /* !DEBUG */
+ if (XFS_IFORK_NEXTENTS(ip, whichfork) != 1)
+ return 0;
+ if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS)
+ return 0;
+ ifp = XFS_IFORK_PTR(ip, whichfork);
+ ASSERT(ifp->if_flags & XFS_IFEXTENTS);
+ ep = xfs_iext_get_ext(ifp, 0);
+ xfs_bmbt_get_all(ep, &s);
+ rval = s.br_startoff == 0 && s.br_blockcount == 1;
+ if (rval && whichfork == XFS_DATA_FORK)
+ ASSERT(XFS_ISIZE(ip) == ip->i_mount->m_sb.sb_blocksize);
+ return rval;
+}
+
+/*
+ * Extent tree manipulation functions used during allocation.
+ */
+
+/*
+ * Convert a delayed allocation to a real allocation.
+ */
+STATIC int /* error */
+xfs_bmap_add_extent_delay_real(
+ struct xfs_bmalloca *bma)
+{
+ struct xfs_bmbt_irec *new = &bma->got;
+ int diff; /* temp value */
+ xfs_bmbt_rec_host_t *ep; /* extent entry for idx */
+ int error; /* error return value */
+ int i; /* temp state */
+ xfs_ifork_t *ifp; /* inode fork pointer */
+ xfs_fileoff_t new_endoff; /* end offset of new entry */
+ xfs_bmbt_irec_t r[3]; /* neighbor extent entries */
+ /* left is 0, right is 1, prev is 2 */
+ int rval=0; /* return value (logging flags) */
+ int state = 0;/* state bits, accessed thru macros */
+ xfs_filblks_t da_new; /* new count del alloc blocks used */
+ xfs_filblks_t da_old; /* old count del alloc blocks used */
+ xfs_filblks_t temp=0; /* value for da_new calculations */
+ xfs_filblks_t temp2=0;/* value for da_new calculations */
+ int tmp_rval; /* partial logging flags */
+
+ ifp = XFS_IFORK_PTR(bma->ip, XFS_DATA_FORK);
- error = xfs_filestream_new_ag(ap, &ag);
- xfs_perag_put(pag);
- if (error)
- return error;
+ ASSERT(bma->idx >= 0);
+ ASSERT(bma->idx <= ifp->if_bytes / sizeof(struct xfs_bmbt_rec));
+ ASSERT(!isnullstartblock(new->br_startblock));
+ ASSERT(!bma->cur ||
+ (bma->cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL));
- /* loop again to set 'blen'*/
- startag = NULLAGNUMBER;
- pag = xfs_perag_get(mp, ag);
- continue;
- }
- }
- if (++ag == mp->m_sb.sb_agcount)
- ag = 0;
- if (ag == startag)
- break;
- xfs_perag_put(pag);
- pag = xfs_perag_get(mp, ag);
- }
- xfs_perag_put(pag);
+ XFS_STATS_INC(xs_add_exlist);
+
+#define LEFT r[0]
+#define RIGHT r[1]
+#define PREV r[2]
/*
- * Since the above loop did a BUF_TRYLOCK, it is
- * possible that there is space for this request.
+ * Set up a bunch of variables to make the tests simpler.
*/
- if (notinit || *blen < ap->minlen)
- args->minlen = ap->minlen;
+ ep = xfs_iext_get_ext(ifp, bma->idx);
+ xfs_bmbt_get_all(ep, &PREV);
+ new_endoff = new->br_startoff + new->br_blockcount;
+ ASSERT(PREV.br_startoff <= new->br_startoff);
+ ASSERT(PREV.br_startoff + PREV.br_blockcount >= new_endoff);
+
+ da_old = startblockval(PREV.br_startblock);
+ da_new = 0;
+
/*
- * If the best seen length is less than the request
- * length, use the best as the minimum.
+ * Set flags determining what part of the previous delayed allocation
+ * extent is being replaced by a real allocation.
*/
- else if (*blen < args->maxlen)
- args->minlen = *blen;
+ if (PREV.br_startoff == new->br_startoff)
+ state |= BMAP_LEFT_FILLING;
+ if (PREV.br_startoff + PREV.br_blockcount == new_endoff)
+ state |= BMAP_RIGHT_FILLING;
+
/*
- * Otherwise we've seen an extent as big as maxlen,
- * use that as the minimum.
+ * Check and set flags if this segment has a left neighbor.
+ * Don't set contiguous if the combined extent would be too large.
*/
- else
- args->minlen = args->maxlen;
+ if (bma->idx > 0) {
+ state |= BMAP_LEFT_VALID;
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx - 1), &LEFT);
+
+ if (isnullstartblock(LEFT.br_startblock))
+ state |= BMAP_LEFT_DELAY;
+ }
+
+ if ((state & BMAP_LEFT_VALID) && !(state & BMAP_LEFT_DELAY) &&
+ LEFT.br_startoff + LEFT.br_blockcount == new->br_startoff &&
+ LEFT.br_startblock + LEFT.br_blockcount == new->br_startblock &&
+ LEFT.br_state == new->br_state &&
+ LEFT.br_blockcount + new->br_blockcount <= MAXEXTLEN)
+ state |= BMAP_LEFT_CONTIG;
/*
- * set the failure fallback case to look in the selected
- * AG as the stream may have moved.
+ * Check and set flags if this segment has a right neighbor.
+ * Don't set contiguous if the combined extent would be too large.
+ * Also check for all-three-contiguous being too large.
*/
- if (xfs_inode_is_filestream(ap->ip))
- ap->blkno = args->fsbno = XFS_AGB_TO_FSB(mp, ag, 0);
+ if (bma->idx < bma->ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) {
+ state |= BMAP_RIGHT_VALID;
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx + 1), &RIGHT);
- return 0;
-}
+ if (isnullstartblock(RIGHT.br_startblock))
+ state |= BMAP_RIGHT_DELAY;
+ }
-STATIC int
-xfs_bmap_btalloc(
- xfs_bmalloca_t *ap) /* bmap alloc argument struct */
-{
- xfs_mount_t *mp; /* mount point structure */
- xfs_alloctype_t atype = 0; /* type for allocation routines */
- xfs_extlen_t align; /* minimum allocation alignment */
- xfs_agnumber_t fb_agno; /* ag number of ap->firstblock */
- xfs_agnumber_t ag;
- xfs_alloc_arg_t args;
- xfs_extlen_t blen;
- xfs_extlen_t nextminlen = 0;
- int nullfb; /* true if ap->firstblock isn't set */
- int isaligned;
- int tryagain;
- int error;
+ if ((state & BMAP_RIGHT_VALID) && !(state & BMAP_RIGHT_DELAY) &&
+ new_endoff == RIGHT.br_startoff &&
+ new->br_startblock + new->br_blockcount == RIGHT.br_startblock &&
+ new->br_state == RIGHT.br_state &&
+ new->br_blockcount + RIGHT.br_blockcount <= MAXEXTLEN &&
+ ((state & (BMAP_LEFT_CONTIG | BMAP_LEFT_FILLING |
+ BMAP_RIGHT_FILLING)) !=
+ (BMAP_LEFT_CONTIG | BMAP_LEFT_FILLING |
+ BMAP_RIGHT_FILLING) ||
+ LEFT.br_blockcount + new->br_blockcount + RIGHT.br_blockcount
+ <= MAXEXTLEN))
+ state |= BMAP_RIGHT_CONTIG;
- ASSERT(ap->length);
+ error = 0;
+ /*
+ * Switch out based on the FILLING and CONTIG state bits.
+ */
+ switch (state & (BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG |
+ BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG)) {
+ case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG |
+ BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
+ /*
+ * Filling in all of a previously delayed allocation extent.
+ * The left and right neighbors are both contiguous with new.
+ */
+ bma->idx--;
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, bma->idx),
+ LEFT.br_blockcount + PREV.br_blockcount +
+ RIGHT.br_blockcount);
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
- mp = ap->ip->i_mount;
- align = ap->userdata ? xfs_get_extsz_hint(ap->ip) : 0;
- if (unlikely(align)) {
- error = xfs_bmap_extsize_align(mp, &ap->got, &ap->prev,
- align, 0, ap->eof, 0, ap->conv,
- &ap->offset, &ap->length);
- ASSERT(!error);
- ASSERT(ap->length);
- }
- nullfb = *ap->firstblock == NULLFSBLOCK;
- fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp, *ap->firstblock);
- if (nullfb) {
- if (ap->userdata && xfs_inode_is_filestream(ap->ip)) {
- ag = xfs_filestream_lookup_ag(ap->ip);
- ag = (ag != NULLAGNUMBER) ? ag : 0;
- ap->blkno = XFS_AGB_TO_FSB(mp, ag, 0);
- } else {
- ap->blkno = XFS_INO_TO_FSB(mp, ap->ip->i_ino);
+ xfs_iext_remove(bma->ip, bma->idx + 1, 2, state);
+ bma->ip->i_d.di_nextents--;
+ if (bma->cur == NULL)
+ rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
+ else {
+ rval = XFS_ILOG_CORE;
+ error = xfs_bmbt_lookup_eq(bma->cur, RIGHT.br_startoff,
+ RIGHT.br_startblock,
+ RIGHT.br_blockcount, &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ error = xfs_btree_delete(bma->cur, &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ error = xfs_btree_decrement(bma->cur, 0, &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ error = xfs_bmbt_update(bma->cur, LEFT.br_startoff,
+ LEFT.br_startblock,
+ LEFT.br_blockcount +
+ PREV.br_blockcount +
+ RIGHT.br_blockcount, LEFT.br_state);
+ if (error)
+ goto done;
}
- } else
- ap->blkno = *ap->firstblock;
+ break;
+
+ case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG:
+ /*
+ * Filling in all of a previously delayed allocation extent.
+ * The left neighbor is contiguous, the right is not.
+ */
+ bma->idx--;
+
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, bma->idx),
+ LEFT.br_blockcount + PREV.br_blockcount);
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+
+ xfs_iext_remove(bma->ip, bma->idx + 1, 1, state);
+ if (bma->cur == NULL)
+ rval = XFS_ILOG_DEXT;
+ else {
+ rval = 0;
+ error = xfs_bmbt_lookup_eq(bma->cur, LEFT.br_startoff,
+ LEFT.br_startblock, LEFT.br_blockcount,
+ &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ error = xfs_bmbt_update(bma->cur, LEFT.br_startoff,
+ LEFT.br_startblock,
+ LEFT.br_blockcount +
+ PREV.br_blockcount, LEFT.br_state);
+ if (error)
+ goto done;
+ }
+ break;
+
+ case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
+ /*
+ * Filling in all of a previously delayed allocation extent.
+ * The right neighbor is contiguous, the left is not.
+ */
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
+ xfs_bmbt_set_startblock(ep, new->br_startblock);
+ xfs_bmbt_set_blockcount(ep,
+ PREV.br_blockcount + RIGHT.br_blockcount);
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+
+ xfs_iext_remove(bma->ip, bma->idx + 1, 1, state);
+ if (bma->cur == NULL)
+ rval = XFS_ILOG_DEXT;
+ else {
+ rval = 0;
+ error = xfs_bmbt_lookup_eq(bma->cur, RIGHT.br_startoff,
+ RIGHT.br_startblock,
+ RIGHT.br_blockcount, &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ error = xfs_bmbt_update(bma->cur, PREV.br_startoff,
+ new->br_startblock,
+ PREV.br_blockcount +
+ RIGHT.br_blockcount, PREV.br_state);
+ if (error)
+ goto done;
+ }
+ break;
+
+ case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING:
+ /*
+ * Filling in all of a previously delayed allocation extent.
+ * Neither the left nor right neighbors are contiguous with
+ * the new one.
+ */
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
+ xfs_bmbt_set_startblock(ep, new->br_startblock);
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
- xfs_bmap_adjacent(ap);
+ bma->ip->i_d.di_nextents++;
+ if (bma->cur == NULL)
+ rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
+ else {
+ rval = XFS_ILOG_CORE;
+ error = xfs_bmbt_lookup_eq(bma->cur, new->br_startoff,
+ new->br_startblock, new->br_blockcount,
+ &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 0, done);
+ bma->cur->bc_rec.b.br_state = XFS_EXT_NORM;
+ error = xfs_btree_insert(bma->cur, &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ }
+ break;
- /*
- * If allowed, use ap->blkno; otherwise must use firstblock since
- * it's in the right allocation group.
- */
- if (nullfb || XFS_FSB_TO_AGNO(mp, ap->blkno) == fb_agno)
- ;
- else
- ap->blkno = *ap->firstblock;
- /*
- * Normal allocation, done through xfs_alloc_vextent.
- */
- tryagain = isaligned = 0;
- memset(&args, 0, sizeof(args));
- args.tp = ap->tp;
- args.mp = mp;
- args.fsbno = ap->blkno;
+ case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG:
+ /*
+ * Filling in the first part of a previous delayed allocation.
+ * The left neighbor is contiguous.
+ */
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx - 1, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, bma->idx - 1),
+ LEFT.br_blockcount + new->br_blockcount);
+ xfs_bmbt_set_startoff(ep,
+ PREV.br_startoff + new->br_blockcount);
+ trace_xfs_bmap_post_update(bma->ip, bma->idx - 1, state, _THIS_IP_);
- /* Trim the allocation back to the maximum an AG can fit. */
- args.maxlen = MIN(ap->length, XFS_ALLOC_AG_MAX_USABLE(mp));
- args.firstblock = *ap->firstblock;
- blen = 0;
- if (nullfb) {
- error = xfs_bmap_btalloc_nullfb(ap, &args, &blen);
- if (error)
- return error;
- } else if (ap->flist->xbf_low) {
- if (xfs_inode_is_filestream(ap->ip))
- args.type = XFS_ALLOCTYPE_FIRST_AG;
- else
- args.type = XFS_ALLOCTYPE_START_BNO;
- args.total = args.minlen = ap->minlen;
- } else {
- args.type = XFS_ALLOCTYPE_NEAR_BNO;
- args.total = ap->total;
- args.minlen = ap->minlen;
- }
- /* apply extent size hints if obtained earlier */
- if (unlikely(align)) {
- args.prod = align;
- if ((args.mod = (xfs_extlen_t)do_mod(ap->offset, args.prod)))
- args.mod = (xfs_extlen_t)(args.prod - args.mod);
- } else if (mp->m_sb.sb_blocksize >= PAGE_CACHE_SIZE) {
- args.prod = 1;
- args.mod = 0;
- } else {
- args.prod = PAGE_CACHE_SIZE >> mp->m_sb.sb_blocklog;
- if ((args.mod = (xfs_extlen_t)(do_mod(ap->offset, args.prod))))
- args.mod = (xfs_extlen_t)(args.prod - args.mod);
- }
- /*
- * If we are not low on available data blocks, and the
- * underlying logical volume manager is a stripe, and
- * the file offset is zero then try to allocate data
- * blocks on stripe unit boundary.
- * NOTE: ap->aeof is only set if the allocation length
- * is >= the stripe unit and the allocation offset is
- * at the end of file.
- */
- if (!ap->flist->xbf_low && ap->aeof) {
- if (!ap->offset) {
- args.alignment = mp->m_dalign;
- atype = args.type;
- isaligned = 1;
- /*
- * Adjust for alignment
- */
- if (blen > args.alignment && blen <= args.maxlen)
- args.minlen = blen - args.alignment;
- args.minalignslop = 0;
- } else {
- /*
- * First try an exact bno allocation.
- * If it fails then do a near or start bno
- * allocation with alignment turned on.
- */
- atype = args.type;
- tryagain = 1;
- args.type = XFS_ALLOCTYPE_THIS_BNO;
- args.alignment = 1;
- /*
- * Compute the minlen+alignment for the
- * next case. Set slop so that the value
- * of minlen+alignment+slop doesn't go up
- * between the calls.
- */
- if (blen > mp->m_dalign && blen <= args.maxlen)
- nextminlen = blen - mp->m_dalign;
- else
- nextminlen = args.minlen;
- if (nextminlen + mp->m_dalign > args.minlen + 1)
- args.minalignslop =
- nextminlen + mp->m_dalign -
- args.minlen - 1;
- else
- args.minalignslop = 0;
+ temp = PREV.br_blockcount - new->br_blockcount;
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(ep, temp);
+ if (bma->cur == NULL)
+ rval = XFS_ILOG_DEXT;
+ else {
+ rval = 0;
+ error = xfs_bmbt_lookup_eq(bma->cur, LEFT.br_startoff,
+ LEFT.br_startblock, LEFT.br_blockcount,
+ &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ error = xfs_bmbt_update(bma->cur, LEFT.br_startoff,
+ LEFT.br_startblock,
+ LEFT.br_blockcount +
+ new->br_blockcount,
+ LEFT.br_state);
+ if (error)
+ goto done;
}
- } else {
- args.alignment = 1;
- args.minalignslop = 0;
- }
- args.minleft = ap->minleft;
- args.wasdel = ap->wasdel;
- args.isfl = 0;
- args.userdata = ap->userdata;
- if ((error = xfs_alloc_vextent(&args)))
- return error;
- if (tryagain && args.fsbno == NULLFSBLOCK) {
+ da_new = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(bma->ip, temp),
+ startblockval(PREV.br_startblock));
+ xfs_bmbt_set_startblock(ep, nullstartblock(da_new));
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+
+ bma->idx--;
+ break;
+
+ case BMAP_LEFT_FILLING:
/*
- * Exact allocation failed. Now try with alignment
- * turned on.
+ * Filling in the first part of a previous delayed allocation.
+ * The left neighbor is not contiguous.
*/
- args.type = atype;
- args.fsbno = ap->blkno;
- args.alignment = mp->m_dalign;
- args.minlen = nextminlen;
- args.minalignslop = 0;
- isaligned = 1;
- if ((error = xfs_alloc_vextent(&args)))
- return error;
- }
- if (isaligned && args.fsbno == NULLFSBLOCK) {
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
+ xfs_bmbt_set_startoff(ep, new_endoff);
+ temp = PREV.br_blockcount - new->br_blockcount;
+ xfs_bmbt_set_blockcount(ep, temp);
+ xfs_iext_insert(bma->ip, bma->idx, 1, new, state);
+ bma->ip->i_d.di_nextents++;
+ if (bma->cur == NULL)
+ rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
+ else {
+ rval = XFS_ILOG_CORE;
+ error = xfs_bmbt_lookup_eq(bma->cur, new->br_startoff,
+ new->br_startblock, new->br_blockcount,
+ &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 0, done);
+ bma->cur->bc_rec.b.br_state = XFS_EXT_NORM;
+ error = xfs_btree_insert(bma->cur, &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ }
+
+ if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
+ error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
+ bma->firstblock, bma->flist,
+ &bma->cur, 1, &tmp_rval, XFS_DATA_FORK);
+ rval |= tmp_rval;
+ if (error)
+ goto done;
+ }
+ da_new = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(bma->ip, temp),
+ startblockval(PREV.br_startblock) -
+ (bma->cur ? bma->cur->bc_private.b.allocated : 0));
+ ep = xfs_iext_get_ext(ifp, bma->idx + 1);
+ xfs_bmbt_set_startblock(ep, nullstartblock(da_new));
+ trace_xfs_bmap_post_update(bma->ip, bma->idx + 1, state, _THIS_IP_);
+ break;
+
+ case BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
/*
- * allocation failed, so turn off alignment and
- * try again.
+ * Filling in the last part of a previous delayed allocation.
+ * The right neighbor is contiguous with the new allocation.
*/
- args.type = atype;
- args.fsbno = ap->blkno;
- args.alignment = 0;
- if ((error = xfs_alloc_vextent(&args)))
- return error;
- }
- if (args.fsbno == NULLFSBLOCK && nullfb &&
- args.minlen > ap->minlen) {
- args.minlen = ap->minlen;
- args.type = XFS_ALLOCTYPE_START_BNO;
- args.fsbno = ap->blkno;
- if ((error = xfs_alloc_vextent(&args)))
- return error;
- }
- if (args.fsbno == NULLFSBLOCK && nullfb) {
- args.fsbno = 0;
- args.type = XFS_ALLOCTYPE_FIRST_AG;
- args.total = ap->minlen;
- args.minleft = 0;
- if ((error = xfs_alloc_vextent(&args)))
- return error;
- ap->flist->xbf_low = 1;
- }
- if (args.fsbno != NULLFSBLOCK) {
+ temp = PREV.br_blockcount - new->br_blockcount;
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx + 1, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(ep, temp);
+ xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, bma->idx + 1),
+ new->br_startoff, new->br_startblock,
+ new->br_blockcount + RIGHT.br_blockcount,
+ RIGHT.br_state);
+ trace_xfs_bmap_post_update(bma->ip, bma->idx + 1, state, _THIS_IP_);
+ if (bma->cur == NULL)
+ rval = XFS_ILOG_DEXT;
+ else {
+ rval = 0;
+ error = xfs_bmbt_lookup_eq(bma->cur, RIGHT.br_startoff,
+ RIGHT.br_startblock,
+ RIGHT.br_blockcount, &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ error = xfs_bmbt_update(bma->cur, new->br_startoff,
+ new->br_startblock,
+ new->br_blockcount +
+ RIGHT.br_blockcount,
+ RIGHT.br_state);
+ if (error)
+ goto done;
+ }
+
+ da_new = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(bma->ip, temp),
+ startblockval(PREV.br_startblock));
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
+ xfs_bmbt_set_startblock(ep, nullstartblock(da_new));
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+
+ bma->idx++;
+ break;
+
+ case BMAP_RIGHT_FILLING:
/*
- * check the allocation happened at the same or higher AG than
- * the first block that was allocated.
+ * Filling in the last part of a previous delayed allocation.
+ * The right neighbor is not contiguous.
*/
- ASSERT(*ap->firstblock == NULLFSBLOCK ||
- XFS_FSB_TO_AGNO(mp, *ap->firstblock) ==
- XFS_FSB_TO_AGNO(mp, args.fsbno) ||
- (ap->flist->xbf_low &&
- XFS_FSB_TO_AGNO(mp, *ap->firstblock) <
- XFS_FSB_TO_AGNO(mp, args.fsbno)));
+ temp = PREV.br_blockcount - new->br_blockcount;
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(ep, temp);
+ xfs_iext_insert(bma->ip, bma->idx + 1, 1, new, state);
+ bma->ip->i_d.di_nextents++;
+ if (bma->cur == NULL)
+ rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
+ else {
+ rval = XFS_ILOG_CORE;
+ error = xfs_bmbt_lookup_eq(bma->cur, new->br_startoff,
+ new->br_startblock, new->br_blockcount,
+ &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 0, done);
+ bma->cur->bc_rec.b.br_state = XFS_EXT_NORM;
+ error = xfs_btree_insert(bma->cur, &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ }
- ap->blkno = args.fsbno;
- if (*ap->firstblock == NULLFSBLOCK)
- *ap->firstblock = args.fsbno;
- ASSERT(nullfb || fb_agno == args.agno ||
- (ap->flist->xbf_low && fb_agno < args.agno));
- ap->length = args.len;
- ap->ip->i_d.di_nblocks += args.len;
- xfs_trans_log_inode(ap->tp, ap->ip, XFS_ILOG_CORE);
- if (ap->wasdel)
- ap->ip->i_delayed_blks -= args.len;
+ if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
+ error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
+ bma->firstblock, bma->flist, &bma->cur, 1,
+ &tmp_rval, XFS_DATA_FORK);
+ rval |= tmp_rval;
+ if (error)
+ goto done;
+ }
+ da_new = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(bma->ip, temp),
+ startblockval(PREV.br_startblock) -
+ (bma->cur ? bma->cur->bc_private.b.allocated : 0));
+ ep = xfs_iext_get_ext(ifp, bma->idx);
+ xfs_bmbt_set_startblock(ep, nullstartblock(da_new));
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+
+ bma->idx++;
+ break;
+
+ case 0:
/*
- * Adjust the disk quota also. This was reserved
- * earlier.
+ * Filling in the middle part of a previous delayed allocation.
+ * Contiguity is impossible here.
+ * This case is avoided almost all the time.
+ *
+ * We start with a delayed allocation:
+ *
+ * +ddddddddddddddddddddddddddddddddddddddddddddddddddddddd+
+ * PREV @ idx
+ *
+ * and we are allocating:
+ * +rrrrrrrrrrrrrrrrr+
+ * new
+ *
+ * and we set it up for insertion as:
+ * +ddddddddddddddddddd+rrrrrrrrrrrrrrrrr+ddddddddddddddddd+
+ * new
+ * PREV @ idx LEFT RIGHT
+ * inserted at idx + 1
*/
- xfs_trans_mod_dquot_byino(ap->tp, ap->ip,
- ap->wasdel ? XFS_TRANS_DQ_DELBCOUNT :
- XFS_TRANS_DQ_BCOUNT,
- (long) args.len);
- } else {
- ap->blkno = NULLFSBLOCK;
- ap->length = 0;
- }
- return 0;
-}
+ temp = new->br_startoff - PREV.br_startoff;
+ temp2 = PREV.br_startoff + PREV.br_blockcount - new_endoff;
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, 0, _THIS_IP_);
+ xfs_bmbt_set_blockcount(ep, temp); /* truncate PREV */
+ LEFT = *new;
+ RIGHT.br_state = PREV.br_state;
+ RIGHT.br_startblock = nullstartblock(
+ (int)xfs_bmap_worst_indlen(bma->ip, temp2));
+ RIGHT.br_startoff = new_endoff;
+ RIGHT.br_blockcount = temp2;
+ /* insert LEFT (r[0]) and RIGHT (r[1]) at the same time */
+ xfs_iext_insert(bma->ip, bma->idx + 1, 2, &LEFT, state);
+ bma->ip->i_d.di_nextents++;
+ if (bma->cur == NULL)
+ rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
+ else {
+ rval = XFS_ILOG_CORE;
+ error = xfs_bmbt_lookup_eq(bma->cur, new->br_startoff,
+ new->br_startblock, new->br_blockcount,
+ &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 0, done);
+ bma->cur->bc_rec.b.br_state = XFS_EXT_NORM;
+ error = xfs_btree_insert(bma->cur, &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ }
-/*
- * xfs_bmap_alloc is called by xfs_bmapi to allocate an extent for a file.
- * It figures out where to ask the underlying allocator to put the new extent.
- */
-STATIC int
-xfs_bmap_alloc(
- xfs_bmalloca_t *ap) /* bmap alloc argument struct */
-{
- if (XFS_IS_REALTIME_INODE(ap->ip) && ap->userdata)
- return xfs_bmap_rtalloc(ap);
- return xfs_bmap_btalloc(ap);
-}
+ if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
+ error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
+ bma->firstblock, bma->flist, &bma->cur,
+ 1, &tmp_rval, XFS_DATA_FORK);
+ rval |= tmp_rval;
+ if (error)
+ goto done;
+ }
+ temp = xfs_bmap_worst_indlen(bma->ip, temp);
+ temp2 = xfs_bmap_worst_indlen(bma->ip, temp2);
+ diff = (int)(temp + temp2 - startblockval(PREV.br_startblock) -
+ (bma->cur ? bma->cur->bc_private.b.allocated : 0));
+ if (diff > 0) {
+ error = xfs_icsb_modify_counters(bma->ip->i_mount,
+ XFS_SBS_FDBLOCKS,
+ -((int64_t)diff), 0);
+ ASSERT(!error);
+ if (error)
+ goto done;
+ }
-/*
- * Transform a btree format file with only one leaf node, where the
- * extents list will fit in the inode, into an extents format file.
- * Since the file extents are already in-core, all we have to do is
- * give up the space for the btree root and pitch the leaf block.
- */
-STATIC int /* error */
-xfs_bmap_btree_to_extents(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_btree_cur_t *cur, /* btree cursor */
- int *logflagsp, /* inode logging flags */
- int whichfork) /* data or attr fork */
-{
- /* REFERENCED */
- struct xfs_btree_block *cblock;/* child btree block */
- xfs_fsblock_t cbno; /* child block number */
- xfs_buf_t *cbp; /* child block's buffer */
- int error; /* error return value */
- xfs_ifork_t *ifp; /* inode fork data */
- xfs_mount_t *mp; /* mount point structure */
- __be64 *pp; /* ptr to block address */
- struct xfs_btree_block *rblock;/* root btree block */
+ ep = xfs_iext_get_ext(ifp, bma->idx);
+ xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx + 2, state, _THIS_IP_);
+ xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, bma->idx + 2),
+ nullstartblock((int)temp2));
+ trace_xfs_bmap_post_update(bma->ip, bma->idx + 2, state, _THIS_IP_);
- mp = ip->i_mount;
- ifp = XFS_IFORK_PTR(ip, whichfork);
- ASSERT(ifp->if_flags & XFS_IFEXTENTS);
- ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE);
- rblock = ifp->if_broot;
- ASSERT(be16_to_cpu(rblock->bb_level) == 1);
- ASSERT(be16_to_cpu(rblock->bb_numrecs) == 1);
- ASSERT(xfs_bmbt_maxrecs(mp, ifp->if_broot_bytes, 0) == 1);
- pp = XFS_BMAP_BROOT_PTR_ADDR(mp, rblock, 1, ifp->if_broot_bytes);
- cbno = be64_to_cpu(*pp);
- *logflagsp = 0;
-#ifdef DEBUG
- if ((error = xfs_btree_check_lptr(cur, cbno, 1)))
- return error;
-#endif
- error = xfs_btree_read_bufl(mp, tp, cbno, 0, &cbp, XFS_BMAP_BTREE_REF,
- &xfs_bmbt_buf_ops);
- if (error)
- return error;
- cblock = XFS_BUF_TO_BLOCK(cbp);
- if ((error = xfs_btree_check_block(cur, cblock, 0, cbp)))
- return error;
- xfs_bmap_add_free(cbno, 1, cur->bc_private.b.flist, mp);
- ip->i_d.di_nblocks--;
- xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_BCOUNT, -1L);
- xfs_trans_binval(tp, cbp);
- if (cur->bc_bufs[0] == cbp)
- cur->bc_bufs[0] = NULL;
- xfs_iroot_realloc(ip, -1, whichfork);
- ASSERT(ifp->if_broot == NULL);
- ASSERT((ifp->if_flags & XFS_IFBROOT) == 0);
- XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_EXTENTS);
- *logflagsp = XFS_ILOG_CORE | xfs_ilog_fext(whichfork);
- return 0;
+ bma->idx++;
+ da_new = temp + temp2;
+ break;
+
+ case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
+ case BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
+ case BMAP_LEFT_FILLING | BMAP_RIGHT_CONTIG:
+ case BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG:
+ case BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
+ case BMAP_LEFT_CONTIG:
+ case BMAP_RIGHT_CONTIG:
+ /*
+ * These cases are all impossible.
+ */
+ ASSERT(0);
+ }
+
+ /* convert to a btree if necessary */
+ if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
+ int tmp_logflags; /* partial log flag return val */
+
+ ASSERT(bma->cur == NULL);
+ error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
+ bma->firstblock, bma->flist, &bma->cur,
+ da_old > 0, &tmp_logflags, XFS_DATA_FORK);
+ bma->logflags |= tmp_logflags;
+ if (error)
+ goto done;
+ }
+
+ /* adjust for changes in reserved delayed indirect blocks */
+ if (da_old || da_new) {
+ temp = da_new;
+ if (bma->cur)
+ temp += bma->cur->bc_private.b.allocated;
+ ASSERT(temp <= da_old);
+ if (temp < da_old)
+ xfs_icsb_modify_counters(bma->ip->i_mount,
+ XFS_SBS_FDBLOCKS,
+ (int64_t)(da_old - temp), 0);
+ }
+
+ /* clear out the allocated field, done with it now in any case. */
+ if (bma->cur)
+ bma->cur->bc_private.b.allocated = 0;
+
+ xfs_bmap_check_leaf_extents(bma->cur, bma->ip, XFS_DATA_FORK);
+done:
+ bma->logflags |= rval;
+ return error;
+#undef LEFT
+#undef RIGHT
+#undef PREV
}
/*
- * Called by xfs_bmapi to update file extent records and the btree
- * after removing space (or undoing a delayed allocation).
+ * Convert an unwritten allocation to a real allocation or vice versa.
*/
STATIC int /* error */
-xfs_bmap_del_extent(
+xfs_bmap_add_extent_unwritten_real(
+ struct xfs_trans *tp,
xfs_inode_t *ip, /* incore inode pointer */
- xfs_trans_t *tp, /* current transaction pointer */
- xfs_extnum_t *idx, /* extent number to update/delete */
+ xfs_extnum_t *idx, /* extent number to update/insert */
+ xfs_btree_cur_t **curp, /* if *curp is null, not a btree */
+ xfs_bmbt_irec_t *new, /* new data to add to file extents */
+ xfs_fsblock_t *first, /* pointer to firstblock variable */
xfs_bmap_free_t *flist, /* list of extents to be freed */
- xfs_btree_cur_t *cur, /* if null, not a btree */
- xfs_bmbt_irec_t *del, /* data to remove from extents */
- int *logflagsp, /* inode logging flags */
- int whichfork) /* data or attr fork */
+ int *logflagsp) /* inode logging flags */
{
- xfs_filblks_t da_new; /* new delay-alloc indirect blocks */
- xfs_filblks_t da_old; /* old delay-alloc indirect blocks */
- xfs_fsblock_t del_endblock=0; /* first block past del */
- xfs_fileoff_t del_endoff; /* first offset past del */
- int delay; /* current block is delayed allocated */
- int do_fx; /* free extent at end of routine */
- xfs_bmbt_rec_host_t *ep; /* current extent entry pointer */
+ xfs_btree_cur_t *cur; /* btree cursor */
+ xfs_bmbt_rec_host_t *ep; /* extent entry for idx */
int error; /* error return value */
- int flags; /* inode logging flags */
- xfs_bmbt_irec_t got; /* current extent entry */
- xfs_fileoff_t got_endoff; /* first offset past got */
int i; /* temp state */
xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_mount_t *mp; /* mount structure */
- xfs_filblks_t nblks; /* quota/sb block count */
- xfs_bmbt_irec_t new; /* new record to be inserted */
- /* REFERENCED */
- uint qfield; /* quota field to update */
- xfs_filblks_t temp; /* for indirect length calculations */
- xfs_filblks_t temp2; /* for indirect length calculations */
- int state = 0;
+ xfs_fileoff_t new_endoff; /* end offset of new entry */
+ xfs_exntst_t newext; /* new extent state */
+ xfs_exntst_t oldext; /* old extent state */
+ xfs_bmbt_irec_t r[3]; /* neighbor extent entries */
+ /* left is 0, right is 1, prev is 2 */
+ int rval=0; /* return value (logging flags) */
+ int state = 0;/* state bits, accessed thru macros */
- XFS_STATS_INC(xs_del_exlist);
+ *logflagsp = 0;
- if (whichfork == XFS_ATTR_FORK)
- state |= BMAP_ATTRFORK;
+ cur = *curp;
+ ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
- mp = ip->i_mount;
- ifp = XFS_IFORK_PTR(ip, whichfork);
- ASSERT((*idx >= 0) && (*idx < ifp->if_bytes /
- (uint)sizeof(xfs_bmbt_rec_t)));
- ASSERT(del->br_blockcount > 0);
- ep = xfs_iext_get_ext(ifp, *idx);
- xfs_bmbt_get_all(ep, &got);
- ASSERT(got.br_startoff <= del->br_startoff);
- del_endoff = del->br_startoff + del->br_blockcount;
- got_endoff = got.br_startoff + got.br_blockcount;
- ASSERT(got_endoff >= del_endoff);
- delay = isnullstartblock(got.br_startblock);
- ASSERT(isnullstartblock(del->br_startblock) == delay);
- flags = 0;
- qfield = 0;
+ ASSERT(*idx >= 0);
+ ASSERT(*idx <= ifp->if_bytes / sizeof(struct xfs_bmbt_rec));
+ ASSERT(!isnullstartblock(new->br_startblock));
+
+ XFS_STATS_INC(xs_add_exlist);
+
+#define LEFT r[0]
+#define RIGHT r[1]
+#define PREV r[2]
+
+ /*
+ * Set up a bunch of variables to make the tests simpler.
+ */
error = 0;
+ ep = xfs_iext_get_ext(ifp, *idx);
+ xfs_bmbt_get_all(ep, &PREV);
+ newext = new->br_state;
+ oldext = (newext == XFS_EXT_UNWRITTEN) ?
+ XFS_EXT_NORM : XFS_EXT_UNWRITTEN;
+ ASSERT(PREV.br_state == oldext);
+ new_endoff = new->br_startoff + new->br_blockcount;
+ ASSERT(PREV.br_startoff <= new->br_startoff);
+ ASSERT(PREV.br_startoff + PREV.br_blockcount >= new_endoff);
+
/*
- * If deleting a real allocation, must free up the disk space.
+ * Set flags determining what part of the previous oldext allocation
+ * extent is being replaced by a newext allocation.
*/
- if (!delay) {
- flags = XFS_ILOG_CORE;
- /*
- * Realtime allocation. Free it and record di_nblocks update.
- */
- if (whichfork == XFS_DATA_FORK && XFS_IS_REALTIME_INODE(ip)) {
- xfs_fsblock_t bno;
- xfs_filblks_t len;
+ if (PREV.br_startoff == new->br_startoff)
+ state |= BMAP_LEFT_FILLING;
+ if (PREV.br_startoff + PREV.br_blockcount == new_endoff)
+ state |= BMAP_RIGHT_FILLING;
- ASSERT(do_mod(del->br_blockcount,
- mp->m_sb.sb_rextsize) == 0);
- ASSERT(do_mod(del->br_startblock,
- mp->m_sb.sb_rextsize) == 0);
- bno = del->br_startblock;
- len = del->br_blockcount;
- do_div(bno, mp->m_sb.sb_rextsize);
- do_div(len, mp->m_sb.sb_rextsize);
- error = xfs_rtfree_extent(tp, bno, (xfs_extlen_t)len);
- if (error)
- goto done;
- do_fx = 0;
- nblks = len * mp->m_sb.sb_rextsize;
- qfield = XFS_TRANS_DQ_RTBCOUNT;
- }
- /*
- * Ordinary allocation.
- */
- else {
- do_fx = 1;
- nblks = del->br_blockcount;
- qfield = XFS_TRANS_DQ_BCOUNT;
- }
- /*
- * Set up del_endblock and cur for later.
- */
- del_endblock = del->br_startblock + del->br_blockcount;
- if (cur) {
- if ((error = xfs_bmbt_lookup_eq(cur, got.br_startoff,
- got.br_startblock, got.br_blockcount,
- &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- }
- da_old = da_new = 0;
- } else {
- da_old = startblockval(got.br_startblock);
- da_new = 0;
- nblks = 0;
- do_fx = 0;
+ /*
+ * Check and set flags if this segment has a left neighbor.
+ * Don't set contiguous if the combined extent would be too large.
+ */
+ if (*idx > 0) {
+ state |= BMAP_LEFT_VALID;
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx - 1), &LEFT);
+
+ if (isnullstartblock(LEFT.br_startblock))
+ state |= BMAP_LEFT_DELAY;
}
+
+ if ((state & BMAP_LEFT_VALID) && !(state & BMAP_LEFT_DELAY) &&
+ LEFT.br_startoff + LEFT.br_blockcount == new->br_startoff &&
+ LEFT.br_startblock + LEFT.br_blockcount == new->br_startblock &&
+ LEFT.br_state == newext &&
+ LEFT.br_blockcount + new->br_blockcount <= MAXEXTLEN)
+ state |= BMAP_LEFT_CONTIG;
+
/*
- * Set flag value to use in switch statement.
- * Left-contig is 2, right-contig is 1.
+ * Check and set flags if this segment has a right neighbor.
+ * Don't set contiguous if the combined extent would be too large.
+ * Also check for all-three-contiguous being too large.
*/
- switch (((got.br_startoff == del->br_startoff) << 1) |
- (got_endoff == del_endoff)) {
- case 3:
- /*
- * Matches the whole extent. Delete the entry.
- */
- xfs_iext_remove(ip, *idx, 1,
- whichfork == XFS_ATTR_FORK ? BMAP_ATTRFORK : 0);
- --*idx;
- if (delay)
- break;
+ if (*idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) {
+ state |= BMAP_RIGHT_VALID;
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx + 1), &RIGHT);
+ if (isnullstartblock(RIGHT.br_startblock))
+ state |= BMAP_RIGHT_DELAY;
+ }
- XFS_IFORK_NEXT_SET(ip, whichfork,
- XFS_IFORK_NEXTENTS(ip, whichfork) - 1);
- flags |= XFS_ILOG_CORE;
- if (!cur) {
- flags |= xfs_ilog_fext(whichfork);
- break;
- }
- if ((error = xfs_btree_delete(cur, &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- break;
+ if ((state & BMAP_RIGHT_VALID) && !(state & BMAP_RIGHT_DELAY) &&
+ new_endoff == RIGHT.br_startoff &&
+ new->br_startblock + new->br_blockcount == RIGHT.br_startblock &&
+ newext == RIGHT.br_state &&
+ new->br_blockcount + RIGHT.br_blockcount <= MAXEXTLEN &&
+ ((state & (BMAP_LEFT_CONTIG | BMAP_LEFT_FILLING |
+ BMAP_RIGHT_FILLING)) !=
+ (BMAP_LEFT_CONTIG | BMAP_LEFT_FILLING |
+ BMAP_RIGHT_FILLING) ||
+ LEFT.br_blockcount + new->br_blockcount + RIGHT.br_blockcount
+ <= MAXEXTLEN))
+ state |= BMAP_RIGHT_CONTIG;
- case 2:
+ /*
+ * Switch out based on the FILLING and CONTIG state bits.
+ */
+ switch (state & (BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG |
+ BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG)) {
+ case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG |
+ BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
/*
- * Deleting the first part of the extent.
+ * Setting all of a previous oldext extent to newext.
+ * The left and right neighbors are both contiguous with new.
*/
+ --*idx;
+
trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
- xfs_bmbt_set_startoff(ep, del_endoff);
- temp = got.br_blockcount - del->br_blockcount;
- xfs_bmbt_set_blockcount(ep, temp);
- if (delay) {
- temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
- da_old);
- xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
- trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- da_new = temp;
- break;
- }
- xfs_bmbt_set_startblock(ep, del_endblock);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx),
+ LEFT.br_blockcount + PREV.br_blockcount +
+ RIGHT.br_blockcount);
trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- if (!cur) {
- flags |= xfs_ilog_fext(whichfork);
- break;
+
+ xfs_iext_remove(ip, *idx + 1, 2, state);
+ ip->i_d.di_nextents -= 2;
+ if (cur == NULL)
+ rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
+ else {
+ rval = XFS_ILOG_CORE;
+ if ((error = xfs_bmbt_lookup_eq(cur, RIGHT.br_startoff,
+ RIGHT.br_startblock,
+ RIGHT.br_blockcount, &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ if ((error = xfs_btree_delete(cur, &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ if ((error = xfs_btree_decrement(cur, 0, &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ if ((error = xfs_btree_delete(cur, &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ if ((error = xfs_btree_decrement(cur, 0, &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ if ((error = xfs_bmbt_update(cur, LEFT.br_startoff,
+ LEFT.br_startblock,
+ LEFT.br_blockcount + PREV.br_blockcount +
+ RIGHT.br_blockcount, LEFT.br_state)))
+ goto done;
}
- if ((error = xfs_bmbt_update(cur, del_endoff, del_endblock,
- got.br_blockcount - del->br_blockcount,
- got.br_state)))
- goto done;
break;
- case 1:
+ case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG:
/*
- * Deleting the last part of the extent.
+ * Setting all of a previous oldext extent to newext.
+ * The left neighbor is contiguous, the right is not.
*/
- temp = got.br_blockcount - del->br_blockcount;
+ --*idx;
+
trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(ep, temp);
- if (delay) {
- temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
- da_old);
- xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
- trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- da_new = temp;
- break;
- }
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx),
+ LEFT.br_blockcount + PREV.br_blockcount);
trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- if (!cur) {
- flags |= xfs_ilog_fext(whichfork);
- break;
+
+ xfs_iext_remove(ip, *idx + 1, 1, state);
+ ip->i_d.di_nextents--;
+ if (cur == NULL)
+ rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
+ else {
+ rval = XFS_ILOG_CORE;
+ if ((error = xfs_bmbt_lookup_eq(cur, PREV.br_startoff,
+ PREV.br_startblock, PREV.br_blockcount,
+ &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ if ((error = xfs_btree_delete(cur, &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ if ((error = xfs_btree_decrement(cur, 0, &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ if ((error = xfs_bmbt_update(cur, LEFT.br_startoff,
+ LEFT.br_startblock,
+ LEFT.br_blockcount + PREV.br_blockcount,
+ LEFT.br_state)))
+ goto done;
}
- if ((error = xfs_bmbt_update(cur, got.br_startoff,
- got.br_startblock,
- got.br_blockcount - del->br_blockcount,
- got.br_state)))
- goto done;
break;
- case 0:
+ case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
/*
- * Deleting the middle of the extent.
+ * Setting all of a previous oldext extent to newext.
+ * The right neighbor is contiguous, the left is not.
*/
- temp = del->br_startoff - got.br_startoff;
trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
- xfs_bmbt_set_blockcount(ep, temp);
- new.br_startoff = del_endoff;
- temp2 = got_endoff - del_endoff;
- new.br_blockcount = temp2;
- new.br_state = got.br_state;
- if (!delay) {
- new.br_startblock = del_endblock;
- flags |= XFS_ILOG_CORE;
- if (cur) {
- if ((error = xfs_bmbt_update(cur,
- got.br_startoff,
- got.br_startblock, temp,
- got.br_state)))
- goto done;
- if ((error = xfs_btree_increment(cur, 0, &i)))
- goto done;
- cur->bc_rec.b = new;
- error = xfs_btree_insert(cur, &i);
- if (error && error != ENOSPC)
- goto done;
- /*
- * If get no-space back from btree insert,
- * it tried a split, and we have a zero
- * block reservation.
- * Fix up our state and return the error.
- */
- if (error == ENOSPC) {
- /*
- * Reset the cursor, don't trust
- * it after any insert operation.
- */
- if ((error = xfs_bmbt_lookup_eq(cur,
- got.br_startoff,
- got.br_startblock,
- temp, &i)))
- goto done;
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- /*
- * Update the btree record back
- * to the original value.
- */
- if ((error = xfs_bmbt_update(cur,
- got.br_startoff,
- got.br_startblock,
- got.br_blockcount,
- got.br_state)))
- goto done;
- /*
- * Reset the extent record back
- * to the original value.
- */
- xfs_bmbt_set_blockcount(ep,
- got.br_blockcount);
- flags = 0;
- error = XFS_ERROR(ENOSPC);
- goto done;
- }
- XFS_WANT_CORRUPTED_GOTO(i == 1, done);
- } else
- flags |= xfs_ilog_fext(whichfork);
- XFS_IFORK_NEXT_SET(ip, whichfork,
- XFS_IFORK_NEXTENTS(ip, whichfork) + 1);
- } else {
- ASSERT(whichfork == XFS_DATA_FORK);
- temp = xfs_bmap_worst_indlen(ip, temp);
- xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
- temp2 = xfs_bmap_worst_indlen(ip, temp2);
- new.br_startblock = nullstartblock((int)temp2);
- da_new = temp + temp2;
- while (da_new > da_old) {
- if (temp) {
- temp--;
- da_new--;
- xfs_bmbt_set_startblock(ep,
- nullstartblock((int)temp));
- }
- if (da_new == da_old)
- break;
- if (temp2) {
- temp2--;
- da_new--;
- new.br_startblock =
- nullstartblock((int)temp2);
- }
- }
- }
+ xfs_bmbt_set_blockcount(ep,
+ PREV.br_blockcount + RIGHT.br_blockcount);
+ xfs_bmbt_set_state(ep, newext);
trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- xfs_iext_insert(ip, *idx + 1, 1, &new, state);
- ++*idx;
- break;
- }
- /*
- * If we need to, add to list of extents to delete.
- */
- if (do_fx)
- xfs_bmap_add_free(del->br_startblock, del->br_blockcount, flist,
- mp);
- /*
- * Adjust inode # blocks in the file.
- */
- if (nblks)
- ip->i_d.di_nblocks -= nblks;
- /*
- * Adjust quota data.
- */
- if (qfield)
- xfs_trans_mod_dquot_byino(tp, ip, qfield, (long)-nblks);
-
- /*
- * Account for change in delayed indirect blocks.
- * Nothing to do for disk quota accounting here.
- */
- ASSERT(da_old >= da_new);
- if (da_old > da_new) {
- xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS,
- (int64_t)(da_old - da_new), 0);
- }
-done:
- *logflagsp = flags;
- return error;
-}
+ xfs_iext_remove(ip, *idx + 1, 1, state);
+ ip->i_d.di_nextents--;
+ if (cur == NULL)
+ rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
+ else {
+ rval = XFS_ILOG_CORE;
+ if ((error = xfs_bmbt_lookup_eq(cur, RIGHT.br_startoff,
+ RIGHT.br_startblock,
+ RIGHT.br_blockcount, &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ if ((error = xfs_btree_delete(cur, &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ if ((error = xfs_btree_decrement(cur, 0, &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ if ((error = xfs_bmbt_update(cur, new->br_startoff,
+ new->br_startblock,
+ new->br_blockcount + RIGHT.br_blockcount,
+ newext)))
+ goto done;
+ }
+ break;
-/*
- * Remove the entry "free" from the free item list. Prev points to the
- * previous entry, unless "free" is the head of the list.
- */
-void
-xfs_bmap_del_free(
- xfs_bmap_free_t *flist, /* free item list header */
- xfs_bmap_free_item_t *prev, /* previous item on list, if any */
- xfs_bmap_free_item_t *free) /* list item to be freed */
-{
- if (prev)
- prev->xbfi_next = free->xbfi_next;
- else
- flist->xbf_first = free->xbfi_next;
- flist->xbf_count--;
- kmem_zone_free(xfs_bmap_free_item_zone, free);
-}
+ case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING:
+ /*
+ * Setting all of a previous oldext extent to newext.
+ * Neither the left nor right neighbors are contiguous with
+ * the new one.
+ */
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ xfs_bmbt_set_state(ep, newext);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
-/*
- * Convert an extents-format file into a btree-format file.
- * The new file will have a root block (in the inode) and a single child block.
- */
-STATIC int /* error */
-xfs_bmap_extents_to_btree(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_fsblock_t *firstblock, /* first-block-allocated */
- xfs_bmap_free_t *flist, /* blocks freed in xaction */
- xfs_btree_cur_t **curp, /* cursor returned to caller */
- int wasdel, /* converting a delayed alloc */
- int *logflagsp, /* inode logging flags */
- int whichfork) /* data or attr fork */
-{
- struct xfs_btree_block *ablock; /* allocated (child) bt block */
- xfs_buf_t *abp; /* buffer for ablock */
- xfs_alloc_arg_t args; /* allocation arguments */
- xfs_bmbt_rec_t *arp; /* child record pointer */
- struct xfs_btree_block *block; /* btree root block */
- xfs_btree_cur_t *cur; /* bmap btree cursor */
- xfs_bmbt_rec_host_t *ep; /* extent record pointer */
- int error; /* error return value */
- xfs_extnum_t i, cnt; /* extent record index */
- xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_bmbt_key_t *kp; /* root block key pointer */
- xfs_mount_t *mp; /* mount structure */
- xfs_extnum_t nextents; /* number of file extents */
- xfs_bmbt_ptr_t *pp; /* root block address pointer */
+ if (cur == NULL)
+ rval = XFS_ILOG_DEXT;
+ else {
+ rval = 0;
+ if ((error = xfs_bmbt_lookup_eq(cur, new->br_startoff,
+ new->br_startblock, new->br_blockcount,
+ &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ if ((error = xfs_bmbt_update(cur, new->br_startoff,
+ new->br_startblock, new->br_blockcount,
+ newext)))
+ goto done;
+ }
+ break;
- ifp = XFS_IFORK_PTR(ip, whichfork);
- ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS);
+ case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG:
+ /*
+ * Setting the first part of a previous oldext extent to newext.
+ * The left neighbor is contiguous.
+ */
+ trace_xfs_bmap_pre_update(ip, *idx - 1, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx - 1),
+ LEFT.br_blockcount + new->br_blockcount);
+ xfs_bmbt_set_startoff(ep,
+ PREV.br_startoff + new->br_blockcount);
+ trace_xfs_bmap_post_update(ip, *idx - 1, state, _THIS_IP_);
- /*
- * Make space in the inode incore.
- */
- xfs_iroot_realloc(ip, 1, whichfork);
- ifp->if_flags |= XFS_IFBROOT;
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ xfs_bmbt_set_startblock(ep,
+ new->br_startblock + new->br_blockcount);
+ xfs_bmbt_set_blockcount(ep,
+ PREV.br_blockcount - new->br_blockcount);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- /*
- * Fill in the root.
- */
- block = ifp->if_broot;
- block->bb_magic = cpu_to_be32(XFS_BMAP_MAGIC);
- block->bb_level = cpu_to_be16(1);
- block->bb_numrecs = cpu_to_be16(1);
- block->bb_u.l.bb_leftsib = cpu_to_be64(NULLDFSBNO);
- block->bb_u.l.bb_rightsib = cpu_to_be64(NULLDFSBNO);
+ --*idx;
- /*
- * Need a cursor. Can't allocate until bb_level is filled in.
- */
- mp = ip->i_mount;
- cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
- cur->bc_private.b.firstblock = *firstblock;
- cur->bc_private.b.flist = flist;
- cur->bc_private.b.flags = wasdel ? XFS_BTCUR_BPRV_WASDEL : 0;
- /*
- * Convert to a btree with two levels, one record in root.
- */
- XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_BTREE);
- memset(&args, 0, sizeof(args));
- args.tp = tp;
- args.mp = mp;
- args.firstblock = *firstblock;
- if (*firstblock == NULLFSBLOCK) {
- args.type = XFS_ALLOCTYPE_START_BNO;
- args.fsbno = XFS_INO_TO_FSB(mp, ip->i_ino);
- } else if (flist->xbf_low) {
- args.type = XFS_ALLOCTYPE_START_BNO;
- args.fsbno = *firstblock;
- } else {
- args.type = XFS_ALLOCTYPE_NEAR_BNO;
- args.fsbno = *firstblock;
- }
- args.minlen = args.maxlen = args.prod = 1;
- args.wasdel = wasdel;
- *logflagsp = 0;
- if ((error = xfs_alloc_vextent(&args))) {
- xfs_iroot_realloc(ip, -1, whichfork);
- xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
- return error;
- }
- /*
- * Allocation can't fail, the space was reserved.
- */
- ASSERT(args.fsbno != NULLFSBLOCK);
- ASSERT(*firstblock == NULLFSBLOCK ||
- args.agno == XFS_FSB_TO_AGNO(mp, *firstblock) ||
- (flist->xbf_low &&
- args.agno > XFS_FSB_TO_AGNO(mp, *firstblock)));
- *firstblock = cur->bc_private.b.firstblock = args.fsbno;
- cur->bc_private.b.allocated++;
- ip->i_d.di_nblocks++;
- xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_BCOUNT, 1L);
- abp = xfs_btree_get_bufl(mp, tp, args.fsbno, 0);
- /*
- * Fill in the child block.
- */
- abp->b_ops = &xfs_bmbt_buf_ops;
- ablock = XFS_BUF_TO_BLOCK(abp);
- ablock->bb_magic = cpu_to_be32(XFS_BMAP_MAGIC);
- ablock->bb_level = 0;
- ablock->bb_u.l.bb_leftsib = cpu_to_be64(NULLDFSBNO);
- ablock->bb_u.l.bb_rightsib = cpu_to_be64(NULLDFSBNO);
- arp = XFS_BMBT_REC_ADDR(mp, ablock, 1);
- nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
- for (cnt = i = 0; i < nextents; i++) {
- ep = xfs_iext_get_ext(ifp, i);
- if (!isnullstartblock(xfs_bmbt_get_startblock(ep))) {
- arp->l0 = cpu_to_be64(ep->l0);
- arp->l1 = cpu_to_be64(ep->l1);
- arp++; cnt++;
+ if (cur == NULL)
+ rval = XFS_ILOG_DEXT;
+ else {
+ rval = 0;
+ if ((error = xfs_bmbt_lookup_eq(cur, PREV.br_startoff,
+ PREV.br_startblock, PREV.br_blockcount,
+ &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ if ((error = xfs_bmbt_update(cur,
+ PREV.br_startoff + new->br_blockcount,
+ PREV.br_startblock + new->br_blockcount,
+ PREV.br_blockcount - new->br_blockcount,
+ oldext)))
+ goto done;
+ if ((error = xfs_btree_decrement(cur, 0, &i)))
+ goto done;
+ error = xfs_bmbt_update(cur, LEFT.br_startoff,
+ LEFT.br_startblock,
+ LEFT.br_blockcount + new->br_blockcount,
+ LEFT.br_state);
+ if (error)
+ goto done;
}
- }
- ASSERT(cnt == XFS_IFORK_NEXTENTS(ip, whichfork));
- xfs_btree_set_numrecs(ablock, cnt);
-
- /*
- * Fill in the root key and pointer.
- */
- kp = XFS_BMBT_KEY_ADDR(mp, block, 1);
- arp = XFS_BMBT_REC_ADDR(mp, ablock, 1);
- kp->br_startoff = cpu_to_be64(xfs_bmbt_disk_get_startoff(arp));
- pp = XFS_BMBT_PTR_ADDR(mp, block, 1, xfs_bmbt_get_maxrecs(cur,
- be16_to_cpu(block->bb_level)));
- *pp = cpu_to_be64(args.fsbno);
+ break;
- /*
- * Do all this logging at the end so that
- * the root is at the right level.
- */
- xfs_btree_log_block(cur, abp, XFS_BB_ALL_BITS);
- xfs_btree_log_recs(cur, abp, 1, be16_to_cpu(ablock->bb_numrecs));
- ASSERT(*curp == NULL);
- *curp = cur;
- *logflagsp = XFS_ILOG_CORE | xfs_ilog_fbroot(whichfork);
- return 0;
-}
+ case BMAP_LEFT_FILLING:
+ /*
+ * Setting the first part of a previous oldext extent to newext.
+ * The left neighbor is not contiguous.
+ */
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ ASSERT(ep && xfs_bmbt_get_state(ep) == oldext);
+ xfs_bmbt_set_startoff(ep, new_endoff);
+ xfs_bmbt_set_blockcount(ep,
+ PREV.br_blockcount - new->br_blockcount);
+ xfs_bmbt_set_startblock(ep,
+ new->br_startblock + new->br_blockcount);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
-/*
- * Calculate the default attribute fork offset for newly created inodes.
- */
-uint
-xfs_default_attroffset(
- struct xfs_inode *ip)
-{
- struct xfs_mount *mp = ip->i_mount;
- uint offset;
+ xfs_iext_insert(ip, *idx, 1, new, state);
+ ip->i_d.di_nextents++;
+ if (cur == NULL)
+ rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
+ else {
+ rval = XFS_ILOG_CORE;
+ if ((error = xfs_bmbt_lookup_eq(cur, PREV.br_startoff,
+ PREV.br_startblock, PREV.br_blockcount,
+ &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ if ((error = xfs_bmbt_update(cur,
+ PREV.br_startoff + new->br_blockcount,
+ PREV.br_startblock + new->br_blockcount,
+ PREV.br_blockcount - new->br_blockcount,
+ oldext)))
+ goto done;
+ cur->bc_rec.b = *new;
+ if ((error = xfs_btree_insert(cur, &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ }
+ break;
- if (mp->m_sb.sb_inodesize == 256) {
- offset = XFS_LITINO(mp) -
- XFS_BMDR_SPACE_CALC(MINABTPTRS);
- } else {
- offset = XFS_BMDR_SPACE_CALC(6 * MINABTPTRS);
- }
+ case BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
+ /*
+ * Setting the last part of a previous oldext extent to newext.
+ * The right neighbor is contiguous with the new allocation.
+ */
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(ep,
+ PREV.br_blockcount - new->br_blockcount);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- ASSERT(offset < XFS_LITINO(mp));
- return offset;
-}
+ ++*idx;
-/*
- * Helper routine to reset inode di_forkoff field when switching
- * attribute fork from local to extent format - we reset it where
- * possible to make space available for inline data fork extents.
- */
-STATIC void
-xfs_bmap_forkoff_reset(
- xfs_mount_t *mp,
- xfs_inode_t *ip,
- int whichfork)
-{
- if (whichfork == XFS_ATTR_FORK &&
- ip->i_d.di_format != XFS_DINODE_FMT_DEV &&
- ip->i_d.di_format != XFS_DINODE_FMT_UUID &&
- ip->i_d.di_format != XFS_DINODE_FMT_BTREE) {
- uint dfl_forkoff = xfs_default_attroffset(ip) >> 3;
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, *idx),
+ new->br_startoff, new->br_startblock,
+ new->br_blockcount + RIGHT.br_blockcount, newext);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- if (dfl_forkoff > ip->i_d.di_forkoff)
- ip->i_d.di_forkoff = dfl_forkoff;
- }
-}
+ if (cur == NULL)
+ rval = XFS_ILOG_DEXT;
+ else {
+ rval = 0;
+ if ((error = xfs_bmbt_lookup_eq(cur, PREV.br_startoff,
+ PREV.br_startblock,
+ PREV.br_blockcount, &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ if ((error = xfs_bmbt_update(cur, PREV.br_startoff,
+ PREV.br_startblock,
+ PREV.br_blockcount - new->br_blockcount,
+ oldext)))
+ goto done;
+ if ((error = xfs_btree_increment(cur, 0, &i)))
+ goto done;
+ if ((error = xfs_bmbt_update(cur, new->br_startoff,
+ new->br_startblock,
+ new->br_blockcount + RIGHT.br_blockcount,
+ newext)))
+ goto done;
+ }
+ break;
-/*
- * Convert a local file to an extents file.
- * This code is out of bounds for data forks of regular files,
- * since the file data needs to get logged so things will stay consistent.
- * (The bmap-level manipulations are ok, though).
- */
-STATIC int /* error */
-xfs_bmap_local_to_extents(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_fsblock_t *firstblock, /* first block allocated in xaction */
- xfs_extlen_t total, /* total blocks needed by transaction */
- int *logflagsp, /* inode logging flags */
- int whichfork) /* data or attr fork */
-{
- int error; /* error return value */
- int flags; /* logging flags returned */
- xfs_ifork_t *ifp; /* inode fork pointer */
+ case BMAP_RIGHT_FILLING:
+ /*
+ * Setting the last part of a previous oldext extent to newext.
+ * The right neighbor is not contiguous.
+ */
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(ep,
+ PREV.br_blockcount - new->br_blockcount);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
- /*
- * We don't want to deal with the case of keeping inode data inline yet.
- * So sending the data fork of a regular inode is invalid.
- */
- ASSERT(!(S_ISREG(ip->i_d.di_mode) && whichfork == XFS_DATA_FORK));
- ifp = XFS_IFORK_PTR(ip, whichfork);
- ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL);
- flags = 0;
- error = 0;
- if (ifp->if_bytes) {
- xfs_alloc_arg_t args; /* allocation arguments */
- xfs_buf_t *bp; /* buffer for extent block */
- xfs_bmbt_rec_host_t *ep;/* extent record pointer */
+ ++*idx;
+ xfs_iext_insert(ip, *idx, 1, new, state);
- memset(&args, 0, sizeof(args));
- args.tp = tp;
- args.mp = ip->i_mount;
- args.firstblock = *firstblock;
- ASSERT((ifp->if_flags &
- (XFS_IFINLINE|XFS_IFEXTENTS|XFS_IFEXTIREC)) == XFS_IFINLINE);
+ ip->i_d.di_nextents++;
+ if (cur == NULL)
+ rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
+ else {
+ rval = XFS_ILOG_CORE;
+ if ((error = xfs_bmbt_lookup_eq(cur, PREV.br_startoff,
+ PREV.br_startblock, PREV.br_blockcount,
+ &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ if ((error = xfs_bmbt_update(cur, PREV.br_startoff,
+ PREV.br_startblock,
+ PREV.br_blockcount - new->br_blockcount,
+ oldext)))
+ goto done;
+ if ((error = xfs_bmbt_lookup_eq(cur, new->br_startoff,
+ new->br_startblock, new->br_blockcount,
+ &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 0, done);
+ cur->bc_rec.b.br_state = XFS_EXT_NORM;
+ if ((error = xfs_btree_insert(cur, &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ }
+ break;
+
+ case 0:
/*
- * Allocate a block. We know we need only one, since the
- * file currently fits in an inode.
+ * Setting the middle part of a previous oldext extent to
+ * newext. Contiguity is impossible here.
+ * One extent becomes three extents.
*/
- if (*firstblock == NULLFSBLOCK) {
- args.fsbno = XFS_INO_TO_FSB(args.mp, ip->i_ino);
- args.type = XFS_ALLOCTYPE_START_BNO;
- } else {
- args.fsbno = *firstblock;
- args.type = XFS_ALLOCTYPE_NEAR_BNO;
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(ep,
+ new->br_startoff - PREV.br_startoff);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
+
+ r[0] = *new;
+ r[1].br_startoff = new_endoff;
+ r[1].br_blockcount =
+ PREV.br_startoff + PREV.br_blockcount - new_endoff;
+ r[1].br_startblock = new->br_startblock + new->br_blockcount;
+ r[1].br_state = oldext;
+
+ ++*idx;
+ xfs_iext_insert(ip, *idx, 2, &r[0], state);
+
+ ip->i_d.di_nextents += 2;
+ if (cur == NULL)
+ rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
+ else {
+ rval = XFS_ILOG_CORE;
+ if ((error = xfs_bmbt_lookup_eq(cur, PREV.br_startoff,
+ PREV.br_startblock, PREV.br_blockcount,
+ &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ /* new right extent - oldext */
+ if ((error = xfs_bmbt_update(cur, r[1].br_startoff,
+ r[1].br_startblock, r[1].br_blockcount,
+ r[1].br_state)))
+ goto done;
+ /* new left extent - oldext */
+ cur->bc_rec.b = PREV;
+ cur->bc_rec.b.br_blockcount =
+ new->br_startoff - PREV.br_startoff;
+ if ((error = xfs_btree_insert(cur, &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ /*
+ * Reset the cursor to the position of the new extent
+ * we are about to insert as we can't trust it after
+ * the previous insert.
+ */
+ if ((error = xfs_bmbt_lookup_eq(cur, new->br_startoff,
+ new->br_startblock, new->br_blockcount,
+ &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 0, done);
+ /* new middle extent - newext */
+ cur->bc_rec.b.br_state = new->br_state;
+ if ((error = xfs_btree_insert(cur, &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
}
- args.total = total;
- args.minlen = args.maxlen = args.prod = 1;
- if ((error = xfs_alloc_vextent(&args)))
- goto done;
+ break;
+
+ case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
+ case BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
+ case BMAP_LEFT_FILLING | BMAP_RIGHT_CONTIG:
+ case BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG:
+ case BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
+ case BMAP_LEFT_CONTIG:
+ case BMAP_RIGHT_CONTIG:
/*
- * Can't fail, the space was reserved.
+ * These cases are all impossible.
*/
- ASSERT(args.fsbno != NULLFSBLOCK);
- ASSERT(args.len == 1);
- *firstblock = args.fsbno;
- bp = xfs_btree_get_bufl(args.mp, tp, args.fsbno, 0);
- bp->b_ops = &xfs_bmbt_buf_ops;
- memcpy(bp->b_addr, ifp->if_u1.if_data, ifp->if_bytes);
- xfs_trans_log_buf(tp, bp, 0, ifp->if_bytes - 1);
- xfs_bmap_forkoff_reset(args.mp, ip, whichfork);
- xfs_idata_realloc(ip, -ifp->if_bytes, whichfork);
- xfs_iext_add(ifp, 0, 1);
- ep = xfs_iext_get_ext(ifp, 0);
- xfs_bmbt_set_allf(ep, 0, args.fsbno, 1, XFS_EXT_NORM);
- trace_xfs_bmap_post_update(ip, 0,
- whichfork == XFS_ATTR_FORK ? BMAP_ATTRFORK : 0,
- _THIS_IP_);
- XFS_IFORK_NEXT_SET(ip, whichfork, 1);
- ip->i_d.di_nblocks = 1;
- xfs_trans_mod_dquot_byino(tp, ip,
- XFS_TRANS_DQ_BCOUNT, 1L);
- flags |= xfs_ilog_fext(whichfork);
- } else {
- ASSERT(XFS_IFORK_NEXTENTS(ip, whichfork) == 0);
- xfs_bmap_forkoff_reset(ip->i_mount, ip, whichfork);
+ ASSERT(0);
+ }
+
+ /* convert to a btree if necessary */
+ if (xfs_bmap_needs_btree(ip, XFS_DATA_FORK)) {
+ int tmp_logflags; /* partial log flag return val */
+
+ ASSERT(cur == NULL);
+ error = xfs_bmap_extents_to_btree(tp, ip, first, flist, &cur,
+ 0, &tmp_logflags, XFS_DATA_FORK);
+ *logflagsp |= tmp_logflags;
+ if (error)
+ goto done;
}
- ifp->if_flags &= ~XFS_IFINLINE;
- ifp->if_flags |= XFS_IFEXTENTS;
- XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_EXTENTS);
- flags |= XFS_ILOG_CORE;
+
+ /* clear out the allocated field, done with it now in any case. */
+ if (cur) {
+ cur->bc_private.b.allocated = 0;
+ *curp = cur;
+ }
+
+ xfs_bmap_check_leaf_extents(*curp, ip, XFS_DATA_FORK);
done:
- *logflagsp = flags;
+ *logflagsp |= rval;
return error;
+#undef LEFT
+#undef RIGHT
+#undef PREV
}
/*
- * Search the extent records for the entry containing block bno.
- * If bno lies in a hole, point to the next entry. If bno lies
- * past eof, *eofp will be set, and *prevp will contain the last
- * entry (null if none). Else, *lastxp will be set to the index
- * of the found entry; *gotp will contain the entry.
+ * Convert a hole to a delayed allocation.
*/
-STATIC xfs_bmbt_rec_host_t * /* pointer to found extent entry */
-xfs_bmap_search_multi_extents(
- xfs_ifork_t *ifp, /* inode fork pointer */
- xfs_fileoff_t bno, /* block number searched for */
- int *eofp, /* out: end of file found */
- xfs_extnum_t *lastxp, /* out: last extent index */
- xfs_bmbt_irec_t *gotp, /* out: extent entry found */
- xfs_bmbt_irec_t *prevp) /* out: previous extent entry found */
+STATIC void
+xfs_bmap_add_extent_hole_delay(
+ xfs_inode_t *ip, /* incore inode pointer */
+ xfs_extnum_t *idx, /* extent number to update/insert */
+ xfs_bmbt_irec_t *new) /* new data to add to file extents */
{
- xfs_bmbt_rec_host_t *ep; /* extent record pointer */
- xfs_extnum_t lastx; /* last extent index */
+ xfs_ifork_t *ifp; /* inode fork pointer */
+ xfs_bmbt_irec_t left; /* left neighbor extent entry */
+ xfs_filblks_t newlen=0; /* new indirect size */
+ xfs_filblks_t oldlen=0; /* old indirect size */
+ xfs_bmbt_irec_t right; /* right neighbor extent entry */
+ int state; /* state bits, accessed thru macros */
+ xfs_filblks_t temp=0; /* temp for indirect calculations */
+
+ ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
+ state = 0;
+ ASSERT(isnullstartblock(new->br_startblock));
/*
- * Initialize the extent entry structure to catch access to
- * uninitialized br_startblock field.
+ * Check and set flags if this segment has a left neighbor
*/
- gotp->br_startoff = 0xffa5a5a5a5a5a5a5LL;
- gotp->br_blockcount = 0xa55a5a5a5a5a5a5aLL;
- gotp->br_state = XFS_EXT_INVALID;
-#if XFS_BIG_BLKNOS
- gotp->br_startblock = 0xffffa5a5a5a5a5a5LL;
-#else
- gotp->br_startblock = 0xffffa5a5;
-#endif
- prevp->br_startoff = NULLFILEOFF;
+ if (*idx > 0) {
+ state |= BMAP_LEFT_VALID;
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx - 1), &left);
- ep = xfs_iext_bno_to_ext(ifp, bno, &lastx);
- if (lastx > 0) {
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp, lastx - 1), prevp);
+ if (isnullstartblock(left.br_startblock))
+ state |= BMAP_LEFT_DELAY;
}
- if (lastx < (ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t))) {
- xfs_bmbt_get_all(ep, gotp);
- *eofp = 0;
- } else {
- if (lastx > 0) {
- *gotp = *prevp;
- }
- *eofp = 1;
- ep = NULL;
+
+ /*
+ * Check and set flags if the current (right) segment exists.
+ * If it doesn't exist, we're converting the hole at end-of-file.
+ */
+ if (*idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t)) {
+ state |= BMAP_RIGHT_VALID;
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx), &right);
+
+ if (isnullstartblock(right.br_startblock))
+ state |= BMAP_RIGHT_DELAY;
+ }
+
+ /*
+ * Set contiguity flags on the left and right neighbors.
+ * Don't let extents get too large, even if the pieces are contiguous.
+ */
+ if ((state & BMAP_LEFT_VALID) && (state & BMAP_LEFT_DELAY) &&
+ left.br_startoff + left.br_blockcount == new->br_startoff &&
+ left.br_blockcount + new->br_blockcount <= MAXEXTLEN)
+ state |= BMAP_LEFT_CONTIG;
+
+ if ((state & BMAP_RIGHT_VALID) && (state & BMAP_RIGHT_DELAY) &&
+ new->br_startoff + new->br_blockcount == right.br_startoff &&
+ new->br_blockcount + right.br_blockcount <= MAXEXTLEN &&
+ (!(state & BMAP_LEFT_CONTIG) ||
+ (left.br_blockcount + new->br_blockcount +
+ right.br_blockcount <= MAXEXTLEN)))
+ state |= BMAP_RIGHT_CONTIG;
+
+ /*
+ * Switch out based on the contiguity flags.
+ */
+ switch (state & (BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG)) {
+ case BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
+ /*
+ * New allocation is contiguous with delayed allocations
+ * on the left and on the right.
+ * Merge all three into a single extent record.
+ */
+ --*idx;
+ temp = left.br_blockcount + new->br_blockcount +
+ right.br_blockcount;
+
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx), temp);
+ oldlen = startblockval(left.br_startblock) +
+ startblockval(new->br_startblock) +
+ startblockval(right.br_startblock);
+ newlen = xfs_bmap_worst_indlen(ip, temp);
+ xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, *idx),
+ nullstartblock((int)newlen));
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
+
+ xfs_iext_remove(ip, *idx + 1, 1, state);
+ break;
+
+ case BMAP_LEFT_CONTIG:
+ /*
+ * New allocation is contiguous with a delayed allocation
+ * on the left.
+ * Merge the new allocation with the left neighbor.
+ */
+ --*idx;
+ temp = left.br_blockcount + new->br_blockcount;
+
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx), temp);
+ oldlen = startblockval(left.br_startblock) +
+ startblockval(new->br_startblock);
+ newlen = xfs_bmap_worst_indlen(ip, temp);
+ xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, *idx),
+ nullstartblock((int)newlen));
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
+ break;
+
+ case BMAP_RIGHT_CONTIG:
+ /*
+ * New allocation is contiguous with a delayed allocation
+ * on the right.
+ * Merge the new allocation with the right neighbor.
+ */
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ temp = new->br_blockcount + right.br_blockcount;
+ oldlen = startblockval(new->br_startblock) +
+ startblockval(right.br_startblock);
+ newlen = xfs_bmap_worst_indlen(ip, temp);
+ xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, *idx),
+ new->br_startoff,
+ nullstartblock((int)newlen), temp, right.br_state);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
+ break;
+
+ case 0:
+ /*
+ * New allocation is not contiguous with another
+ * delayed allocation.
+ * Insert a new entry.
+ */
+ oldlen = newlen = 0;
+ xfs_iext_insert(ip, *idx, 1, new, state);
+ break;
+ }
+ if (oldlen != newlen) {
+ ASSERT(oldlen > newlen);
+ xfs_icsb_modify_counters(ip->i_mount, XFS_SBS_FDBLOCKS,
+ (int64_t)(oldlen - newlen), 0);
+ /*
+ * Nothing to do for disk quota accounting here.
+ */
}
- *lastxp = lastx;
- return ep;
}
/*
- * Search the extents list for the inode, for the extent containing bno.
- * If bno lies in a hole, point to the next entry. If bno lies past eof,
- * *eofp will be set, and *prevp will contain the last entry (null if none).
- * Else, *lastxp will be set to the index of the found
- * entry; *gotp will contain the entry.
+ * Convert a hole to a real allocation.
*/
-xfs_bmbt_rec_host_t * /* pointer to found extent entry */
-xfs_bmap_search_extents(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_fileoff_t bno, /* block number searched for */
- int fork, /* data or attr fork */
- int *eofp, /* out: end of file found */
- xfs_extnum_t *lastxp, /* out: last extent index */
- xfs_bmbt_irec_t *gotp, /* out: extent entry found */
- xfs_bmbt_irec_t *prevp) /* out: previous extent entry found */
+STATIC int /* error */
+xfs_bmap_add_extent_hole_real(
+ struct xfs_bmalloca *bma,
+ int whichfork)
{
- xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_bmbt_rec_host_t *ep; /* extent record pointer */
+ struct xfs_bmbt_irec *new = &bma->got;
+ int error; /* error return value */
+ int i; /* temp state */
+ xfs_ifork_t *ifp; /* inode fork pointer */
+ xfs_bmbt_irec_t left; /* left neighbor extent entry */
+ xfs_bmbt_irec_t right; /* right neighbor extent entry */
+ int rval=0; /* return value (logging flags) */
+ int state; /* state bits, accessed thru macros */
- XFS_STATS_INC(xs_look_exlist);
- ifp = XFS_IFORK_PTR(ip, fork);
+ ifp = XFS_IFORK_PTR(bma->ip, whichfork);
- ep = xfs_bmap_search_multi_extents(ifp, bno, eofp, lastxp, gotp, prevp);
+ ASSERT(bma->idx >= 0);
+ ASSERT(bma->idx <= ifp->if_bytes / sizeof(struct xfs_bmbt_rec));
+ ASSERT(!isnullstartblock(new->br_startblock));
+ ASSERT(!bma->cur ||
+ !(bma->cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL));
- if (unlikely(!(gotp->br_startblock) && (*lastxp != NULLEXTNUM) &&
- !(XFS_IS_REALTIME_INODE(ip) && fork == XFS_DATA_FORK))) {
- xfs_alert_tag(ip->i_mount, XFS_PTAG_FSBLOCK_ZERO,
- "Access to block zero in inode %llu "
- "start_block: %llx start_off: %llx "
- "blkcnt: %llx extent-state: %x lastx: %x\n",
- (unsigned long long)ip->i_ino,
- (unsigned long long)gotp->br_startblock,
- (unsigned long long)gotp->br_startoff,
- (unsigned long long)gotp->br_blockcount,
- gotp->br_state, *lastxp);
- *lastxp = NULLEXTNUM;
- *eofp = 1;
- return NULL;
+ XFS_STATS_INC(xs_add_exlist);
+
+ state = 0;
+ if (whichfork == XFS_ATTR_FORK)
+ state |= BMAP_ATTRFORK;
+
+ /*
+ * Check and set flags if this segment has a left neighbor.
+ */
+ if (bma->idx > 0) {
+ state |= BMAP_LEFT_VALID;
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx - 1), &left);
+ if (isnullstartblock(left.br_startblock))
+ state |= BMAP_LEFT_DELAY;
}
- return ep;
-}
-
-/*
- * Compute the worst-case number of indirect blocks that will be used
- * for ip's delayed extent of length "len".
- */
-STATIC xfs_filblks_t
-xfs_bmap_worst_indlen(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_filblks_t len) /* delayed extent length */
-{
- int level; /* btree level number */
- int maxrecs; /* maximum record count at this level */
- xfs_mount_t *mp; /* mount structure */
- xfs_filblks_t rval; /* return value */
- mp = ip->i_mount;
- maxrecs = mp->m_bmap_dmxr[0];
- for (level = 0, rval = 0;
- level < XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK);
- level++) {
- len += maxrecs - 1;
- do_div(len, maxrecs);
- rval += len;
- if (len == 1)
- return rval + XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK) -
- level - 1;
- if (level == 0)
- maxrecs = mp->m_bmap_dmxr[1];
+ /*
+ * Check and set flags if this segment has a current value.
+ * Not true if we're inserting into the "hole" at eof.
+ */
+ if (bma->idx < ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t)) {
+ state |= BMAP_RIGHT_VALID;
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx), &right);
+ if (isnullstartblock(right.br_startblock))
+ state |= BMAP_RIGHT_DELAY;
}
- return rval;
-}
-/*
- * Convert inode from non-attributed to attributed.
- * Must not be in a transaction, ip must not be locked.
- */
-int /* error code */
-xfs_bmap_add_attrfork(
- xfs_inode_t *ip, /* incore inode pointer */
- int size, /* space new attribute needs */
- int rsvd) /* xact may use reserved blks */
-{
- xfs_fsblock_t firstblock; /* 1st block/ag allocated */
- xfs_bmap_free_t flist; /* freed extent records */
- xfs_mount_t *mp; /* mount structure */
- xfs_trans_t *tp; /* transaction pointer */
- int blks; /* space reservation */
- int version = 1; /* superblock attr version */
- int committed; /* xaction was committed */
- int logflags; /* logging flags */
- int error; /* error return value */
+ /*
+ * We're inserting a real allocation between "left" and "right".
+ * Set the contiguity flags. Don't let extents get too large.
+ */
+ if ((state & BMAP_LEFT_VALID) && !(state & BMAP_LEFT_DELAY) &&
+ left.br_startoff + left.br_blockcount == new->br_startoff &&
+ left.br_startblock + left.br_blockcount == new->br_startblock &&
+ left.br_state == new->br_state &&
+ left.br_blockcount + new->br_blockcount <= MAXEXTLEN)
+ state |= BMAP_LEFT_CONTIG;
- ASSERT(XFS_IFORK_Q(ip) == 0);
+ if ((state & BMAP_RIGHT_VALID) && !(state & BMAP_RIGHT_DELAY) &&
+ new->br_startoff + new->br_blockcount == right.br_startoff &&
+ new->br_startblock + new->br_blockcount == right.br_startblock &&
+ new->br_state == right.br_state &&
+ new->br_blockcount + right.br_blockcount <= MAXEXTLEN &&
+ (!(state & BMAP_LEFT_CONTIG) ||
+ left.br_blockcount + new->br_blockcount +
+ right.br_blockcount <= MAXEXTLEN))
+ state |= BMAP_RIGHT_CONTIG;
- mp = ip->i_mount;
- ASSERT(!XFS_NOT_DQATTACHED(mp, ip));
- tp = xfs_trans_alloc(mp, XFS_TRANS_ADDAFORK);
- blks = XFS_ADDAFORK_SPACE_RES(mp);
- if (rsvd)
- tp->t_flags |= XFS_TRANS_RESERVE;
- if ((error = xfs_trans_reserve(tp, blks, XFS_ADDAFORK_LOG_RES(mp), 0,
- XFS_TRANS_PERM_LOG_RES, XFS_ADDAFORK_LOG_COUNT)))
- goto error0;
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- error = xfs_trans_reserve_quota_nblks(tp, ip, blks, 0, rsvd ?
- XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_FORCE_RES :
- XFS_QMOPT_RES_REGBLKS);
- if (error) {
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES);
- return error;
- }
- if (XFS_IFORK_Q(ip))
- goto error1;
- if (ip->i_d.di_aformat != XFS_DINODE_FMT_EXTENTS) {
+ error = 0;
+ /*
+ * Select which case we're in here, and implement it.
+ */
+ switch (state & (BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG)) {
+ case BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
/*
- * For inodes coming from pre-6.2 filesystems.
+ * New allocation is contiguous with real allocations on the
+ * left and on the right.
+ * Merge all three into a single extent record.
*/
- ASSERT(ip->i_d.di_aformat == 0);
- ip->i_d.di_aformat = XFS_DINODE_FMT_EXTENTS;
- }
- ASSERT(ip->i_d.di_anextents == 0);
+ --bma->idx;
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, bma->idx),
+ left.br_blockcount + new->br_blockcount +
+ right.br_blockcount);
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
- xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
- xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+ xfs_iext_remove(bma->ip, bma->idx + 1, 1, state);
- switch (ip->i_d.di_format) {
- case XFS_DINODE_FMT_DEV:
- ip->i_d.di_forkoff = roundup(sizeof(xfs_dev_t), 8) >> 3;
+ XFS_IFORK_NEXT_SET(bma->ip, whichfork,
+ XFS_IFORK_NEXTENTS(bma->ip, whichfork) - 1);
+ if (bma->cur == NULL) {
+ rval = XFS_ILOG_CORE | xfs_ilog_fext(whichfork);
+ } else {
+ rval = XFS_ILOG_CORE;
+ error = xfs_bmbt_lookup_eq(bma->cur, right.br_startoff,
+ right.br_startblock, right.br_blockcount,
+ &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ error = xfs_btree_delete(bma->cur, &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ error = xfs_btree_decrement(bma->cur, 0, &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ error = xfs_bmbt_update(bma->cur, left.br_startoff,
+ left.br_startblock,
+ left.br_blockcount +
+ new->br_blockcount +
+ right.br_blockcount,
+ left.br_state);
+ if (error)
+ goto done;
+ }
break;
- case XFS_DINODE_FMT_UUID:
- ip->i_d.di_forkoff = roundup(sizeof(uuid_t), 8) >> 3;
+
+ case BMAP_LEFT_CONTIG:
+ /*
+ * New allocation is contiguous with a real allocation
+ * on the left.
+ * Merge the new allocation with the left neighbor.
+ */
+ --bma->idx;
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, bma->idx),
+ left.br_blockcount + new->br_blockcount);
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+
+ if (bma->cur == NULL) {
+ rval = xfs_ilog_fext(whichfork);
+ } else {
+ rval = 0;
+ error = xfs_bmbt_lookup_eq(bma->cur, left.br_startoff,
+ left.br_startblock, left.br_blockcount,
+ &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ error = xfs_bmbt_update(bma->cur, left.br_startoff,
+ left.br_startblock,
+ left.br_blockcount +
+ new->br_blockcount,
+ left.br_state);
+ if (error)
+ goto done;
+ }
break;
- case XFS_DINODE_FMT_LOCAL:
- case XFS_DINODE_FMT_EXTENTS:
- case XFS_DINODE_FMT_BTREE:
- ip->i_d.di_forkoff = xfs_attr_shortform_bytesfit(ip, size);
- if (!ip->i_d.di_forkoff)
- ip->i_d.di_forkoff = xfs_default_attroffset(ip) >> 3;
- else if (mp->m_flags & XFS_MOUNT_ATTR2)
- version = 2;
+
+ case BMAP_RIGHT_CONTIG:
+ /*
+ * New allocation is contiguous with a real allocation
+ * on the right.
+ * Merge the new allocation with the right neighbor.
+ */
+ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
+ xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, bma->idx),
+ new->br_startoff, new->br_startblock,
+ new->br_blockcount + right.br_blockcount,
+ right.br_state);
+ trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+
+ if (bma->cur == NULL) {
+ rval = xfs_ilog_fext(whichfork);
+ } else {
+ rval = 0;
+ error = xfs_bmbt_lookup_eq(bma->cur,
+ right.br_startoff,
+ right.br_startblock,
+ right.br_blockcount, &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ error = xfs_bmbt_update(bma->cur, new->br_startoff,
+ new->br_startblock,
+ new->br_blockcount +
+ right.br_blockcount,
+ right.br_state);
+ if (error)
+ goto done;
+ }
+ break;
+
+ case 0:
+ /*
+ * New allocation is not contiguous with another
+ * real allocation.
+ * Insert a new entry.
+ */
+ xfs_iext_insert(bma->ip, bma->idx, 1, new, state);
+ XFS_IFORK_NEXT_SET(bma->ip, whichfork,
+ XFS_IFORK_NEXTENTS(bma->ip, whichfork) + 1);
+ if (bma->cur == NULL) {
+ rval = XFS_ILOG_CORE | xfs_ilog_fext(whichfork);
+ } else {
+ rval = XFS_ILOG_CORE;
+ error = xfs_bmbt_lookup_eq(bma->cur,
+ new->br_startoff,
+ new->br_startblock,
+ new->br_blockcount, &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 0, done);
+ bma->cur->bc_rec.b.br_state = new->br_state;
+ error = xfs_btree_insert(bma->cur, &i);
+ if (error)
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ }
break;
- default:
- ASSERT(0);
- error = XFS_ERROR(EINVAL);
- goto error1;
}
- ASSERT(ip->i_afp == NULL);
- ip->i_afp = kmem_zone_zalloc(xfs_ifork_zone, KM_SLEEP);
- ip->i_afp->if_flags = XFS_IFEXTENTS;
- logflags = 0;
- xfs_bmap_init(&flist, &firstblock);
- switch (ip->i_d.di_format) {
- case XFS_DINODE_FMT_LOCAL:
- error = xfs_bmap_add_attrfork_local(tp, ip, &firstblock, &flist,
- &logflags);
- break;
- case XFS_DINODE_FMT_EXTENTS:
- error = xfs_bmap_add_attrfork_extents(tp, ip, &firstblock,
- &flist, &logflags);
- break;
- case XFS_DINODE_FMT_BTREE:
- error = xfs_bmap_add_attrfork_btree(tp, ip, &firstblock, &flist,
- &logflags);
- break;
- default:
- error = 0;
- break;
- }
- if (logflags)
- xfs_trans_log_inode(tp, ip, logflags);
- if (error)
- goto error2;
- if (!xfs_sb_version_hasattr(&mp->m_sb) ||
- (!xfs_sb_version_hasattr2(&mp->m_sb) && version == 2)) {
- __int64_t sbfields = 0;
+ /* convert to a btree if necessary */
+ if (xfs_bmap_needs_btree(bma->ip, whichfork)) {
+ int tmp_logflags; /* partial log flag return val */
- spin_lock(&mp->m_sb_lock);
- if (!xfs_sb_version_hasattr(&mp->m_sb)) {
- xfs_sb_version_addattr(&mp->m_sb);
- sbfields |= XFS_SB_VERSIONNUM;
- }
- if (!xfs_sb_version_hasattr2(&mp->m_sb) && version == 2) {
- xfs_sb_version_addattr2(&mp->m_sb);
- sbfields |= (XFS_SB_VERSIONNUM | XFS_SB_FEATURES2);
- }
- if (sbfields) {
- spin_unlock(&mp->m_sb_lock);
- xfs_mod_sb(tp, sbfields);
- } else
- spin_unlock(&mp->m_sb_lock);
+ ASSERT(bma->cur == NULL);
+ error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
+ bma->firstblock, bma->flist, &bma->cur,
+ 0, &tmp_logflags, whichfork);
+ bma->logflags |= tmp_logflags;
+ if (error)
+ goto done;
}
- error = xfs_bmap_finish(&tp, &flist, &committed);
- if (error)
- goto error2;
- return xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
-error2:
- xfs_bmap_cancel(&flist);
-error1:
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
-error0:
- xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT);
+ /* clear out the allocated field, done with it now in any case. */
+ if (bma->cur)
+ bma->cur->bc_private.b.allocated = 0;
+
+ xfs_bmap_check_leaf_extents(bma->cur, bma->ip, whichfork);
+done:
+ bma->logflags |= rval;
return error;
}
/*
- * Add the extent to the list of extents to be free at transaction end.
- * The list is maintained sorted (by block number).
+ * Functions used in the extent read, allocate and remove paths
*/
-/* ARGSUSED */
-void
-xfs_bmap_add_free(
- xfs_fsblock_t bno, /* fs block number of extent */
- xfs_filblks_t len, /* length of extent */
- xfs_bmap_free_t *flist, /* list of extents */
- xfs_mount_t *mp) /* mount point structure */
-{
- xfs_bmap_free_item_t *cur; /* current (next) element */
- xfs_bmap_free_item_t *new; /* new element */
- xfs_bmap_free_item_t *prev; /* previous element */
-#ifdef DEBUG
- xfs_agnumber_t agno;
- xfs_agblock_t agbno;
-
- ASSERT(bno != NULLFSBLOCK);
- ASSERT(len > 0);
- ASSERT(len <= MAXEXTLEN);
- ASSERT(!isnullstartblock(bno));
- agno = XFS_FSB_TO_AGNO(mp, bno);
- agbno = XFS_FSB_TO_AGBNO(mp, bno);
- ASSERT(agno < mp->m_sb.sb_agcount);
- ASSERT(agbno < mp->m_sb.sb_agblocks);
- ASSERT(len < mp->m_sb.sb_agblocks);
- ASSERT(agbno + len <= mp->m_sb.sb_agblocks);
-#endif
- ASSERT(xfs_bmap_free_item_zone != NULL);
- new = kmem_zone_alloc(xfs_bmap_free_item_zone, KM_SLEEP);
- new->xbfi_startblock = bno;
- new->xbfi_blockcount = (xfs_extlen_t)len;
- for (prev = NULL, cur = flist->xbf_first;
- cur != NULL;
- prev = cur, cur = cur->xbfi_next) {
- if (cur->xbfi_startblock >= bno)
- break;
- }
- if (prev)
- prev->xbfi_next = new;
- else
- flist->xbf_first = new;
- new->xbfi_next = cur;
- flist->xbf_count++;
-}
/*
- * Compute and fill in the value of the maximum depth of a bmap btree
- * in this filesystem. Done once, during mount.
+ * Adjust the size of the new extent based on di_extsize and rt extsize.
*/
-void
-xfs_bmap_compute_maxlevels(
- xfs_mount_t *mp, /* file system mount structure */
- int whichfork) /* data or attr fork */
+STATIC int
+xfs_bmap_extsize_align(
+ xfs_mount_t *mp,
+ xfs_bmbt_irec_t *gotp, /* next extent pointer */
+ xfs_bmbt_irec_t *prevp, /* previous extent pointer */
+ xfs_extlen_t extsz, /* align to this extent size */
+ int rt, /* is this a realtime inode? */
+ int eof, /* is extent at end-of-file? */
+ int delay, /* creating delalloc extent? */
+ int convert, /* overwriting unwritten extent? */
+ xfs_fileoff_t *offp, /* in/out: aligned offset */
+ xfs_extlen_t *lenp) /* in/out: aligned length */
{
- int level; /* btree level */
- uint maxblocks; /* max blocks at this level */
- uint maxleafents; /* max leaf entries possible */
- int maxrootrecs; /* max records in root block */
- int minleafrecs; /* min records in leaf block */
- int minnoderecs; /* min records in node block */
- int sz; /* root block size */
+ xfs_fileoff_t orig_off; /* original offset */
+ xfs_extlen_t orig_alen; /* original length */
+ xfs_fileoff_t orig_end; /* original off+len */
+ xfs_fileoff_t nexto; /* next file offset */
+ xfs_fileoff_t prevo; /* previous file offset */
+ xfs_fileoff_t align_off; /* temp for offset */
+ xfs_extlen_t align_alen; /* temp for length */
+ xfs_extlen_t temp; /* temp for calculations */
+
+ if (convert)
+ return 0;
+
+ orig_off = align_off = *offp;
+ orig_alen = align_alen = *lenp;
+ orig_end = orig_off + orig_alen;
/*
- * The maximum number of extents in a file, hence the maximum
- * number of leaf entries, is controlled by the type of di_nextents
- * (a signed 32-bit number, xfs_extnum_t), or by di_anextents
- * (a signed 16-bit number, xfs_aextnum_t).
- *
- * Note that we can no longer assume that if we are in ATTR1 that
- * the fork offset of all the inodes will be
- * (xfs_default_attroffset(ip) >> 3) because we could have mounted
- * with ATTR2 and then mounted back with ATTR1, keeping the
- * di_forkoff's fixed but probably at various positions. Therefore,
- * for both ATTR1 and ATTR2 we have to assume the worst case scenario
- * of a minimum size available.
+ * If this request overlaps an existing extent, then don't
+ * attempt to perform any additional alignment.
*/
- if (whichfork == XFS_DATA_FORK) {
- maxleafents = MAXEXTNUM;
- sz = XFS_BMDR_SPACE_CALC(MINDBTPTRS);
- } else {
- maxleafents = MAXAEXTNUM;
- sz = XFS_BMDR_SPACE_CALC(MINABTPTRS);
- }
- maxrootrecs = xfs_bmdr_maxrecs(mp, sz, 0);
- minleafrecs = mp->m_bmap_dmnr[0];
- minnoderecs = mp->m_bmap_dmnr[1];
- maxblocks = (maxleafents + minleafrecs - 1) / minleafrecs;
- for (level = 1; maxblocks > 1; level++) {
- if (maxblocks <= maxrootrecs)
- maxblocks = 1;
- else
- maxblocks = (maxblocks + minnoderecs - 1) / minnoderecs;
+ if (!delay && !eof &&
+ (orig_off >= gotp->br_startoff) &&
+ (orig_end <= gotp->br_startoff + gotp->br_blockcount)) {
+ return 0;
}
- mp->m_bm_maxlevels[whichfork] = level;
-}
-
-/*
- * Free up any items left in the list.
- */
-void
-xfs_bmap_cancel(
- xfs_bmap_free_t *flist) /* list of bmap_free_items */
-{
- xfs_bmap_free_item_t *free; /* free list item */
- xfs_bmap_free_item_t *next;
- if (flist->xbf_count == 0)
- return;
- ASSERT(flist->xbf_first != NULL);
- for (free = flist->xbf_first; free; free = next) {
- next = free->xbfi_next;
- xfs_bmap_del_free(flist, NULL, free);
+ /*
+ * If the file offset is unaligned vs. the extent size
+ * we need to align it. This will be possible unless
+ * the file was previously written with a kernel that didn't
+ * perform this alignment, or if a truncate shot us in the
+ * foot.
+ */
+ temp = do_mod(orig_off, extsz);
+ if (temp) {
+ align_alen += temp;
+ align_off -= temp;
+ }
+ /*
+ * Same adjustment for the end of the requested area.
+ */
+ if ((temp = (align_alen % extsz))) {
+ align_alen += extsz - temp;
+ }
+ /*
+ * If the previous block overlaps with this proposed allocation
+ * then move the start forward without adjusting the length.
+ */
+ if (prevp->br_startoff != NULLFILEOFF) {
+ if (prevp->br_startblock == HOLESTARTBLOCK)
+ prevo = prevp->br_startoff;
+ else
+ prevo = prevp->br_startoff + prevp->br_blockcount;
+ } else
+ prevo = 0;
+ if (align_off != orig_off && align_off < prevo)
+ align_off = prevo;
+ /*
+ * If the next block overlaps with this proposed allocation
+ * then move the start back without adjusting the length,
+ * but not before offset 0.
+ * This may of course make the start overlap previous block,
+ * and if we hit the offset 0 limit then the next block
+ * can still overlap too.
+ */
+ if (!eof && gotp->br_startoff != NULLFILEOFF) {
+ if ((delay && gotp->br_startblock == HOLESTARTBLOCK) ||
+ (!delay && gotp->br_startblock == DELAYSTARTBLOCK))
+ nexto = gotp->br_startoff + gotp->br_blockcount;
+ else
+ nexto = gotp->br_startoff;
+ } else
+ nexto = NULLFILEOFF;
+ if (!eof &&
+ align_off + align_alen != orig_end &&
+ align_off + align_alen > nexto)
+ align_off = nexto > align_alen ? nexto - align_alen : 0;
+ /*
+ * If we're now overlapping the next or previous extent that
+ * means we can't fit an extsz piece in this hole. Just move
+ * the start forward to the first valid spot and set
+ * the length so we hit the end.
+ */
+ if (align_off != orig_off && align_off < prevo)
+ align_off = prevo;
+ if (align_off + align_alen != orig_end &&
+ align_off + align_alen > nexto &&
+ nexto != NULLFILEOFF) {
+ ASSERT(nexto > prevo);
+ align_alen = nexto - align_off;
}
- ASSERT(flist->xbf_count == 0);
-}
-
-/*
- * Returns the file-relative block number of the first unused block(s)
- * in the file with at least "len" logically contiguous blocks free.
- * This is the lowest-address hole if the file has holes, else the first block
- * past the end of file.
- * Return 0 if the file is currently local (in-inode).
- */
-int /* error */
-xfs_bmap_first_unused(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode */
- xfs_extlen_t len, /* size of hole to find */
- xfs_fileoff_t *first_unused, /* unused block */
- int whichfork) /* data or attr fork */
-{
- int error; /* error return value */
- int idx; /* extent record index */
- xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_fileoff_t lastaddr; /* last block number seen */
- xfs_fileoff_t lowest; /* lowest useful block */
- xfs_fileoff_t max; /* starting useful block */
- xfs_fileoff_t off; /* offset for this block */
- xfs_extnum_t nextents; /* number of extent entries */
- ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE ||
- XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS ||
- XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL);
- if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL) {
- *first_unused = 0;
- return 0;
- }
- ifp = XFS_IFORK_PTR(ip, whichfork);
- if (!(ifp->if_flags & XFS_IFEXTENTS) &&
- (error = xfs_iread_extents(tp, ip, whichfork)))
- return error;
- lowest = *first_unused;
- nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
- for (idx = 0, lastaddr = 0, max = lowest; idx < nextents; idx++) {
- xfs_bmbt_rec_host_t *ep = xfs_iext_get_ext(ifp, idx);
- off = xfs_bmbt_get_startoff(ep);
+ /*
+ * If realtime, and the result isn't a multiple of the realtime
+ * extent size we need to remove blocks until it is.
+ */
+ if (rt && (temp = (align_alen % mp->m_sb.sb_rextsize))) {
/*
- * See if the hole before this extent will work.
+ * We're not covering the original request, or
+ * we won't be able to once we fix the length.
*/
- if (off >= lowest + len && off - max >= len) {
- *first_unused = max;
- return 0;
+ if (orig_off < align_off ||
+ orig_end > align_off + align_alen ||
+ align_alen - temp < orig_alen)
+ return XFS_ERROR(EINVAL);
+ /*
+ * Try to fix it by moving the start up.
+ */
+ if (align_off + temp <= orig_off) {
+ align_alen -= temp;
+ align_off += temp;
}
- lastaddr = off + xfs_bmbt_get_blockcount(ep);
- max = XFS_FILEOFF_MAX(lastaddr, lowest);
+ /*
+ * Try to fix it by moving the end in.
+ */
+ else if (align_off + align_alen - temp >= orig_end)
+ align_alen -= temp;
+ /*
+ * Set the start to the minimum then trim the length.
+ */
+ else {
+ align_alen -= orig_off - align_off;
+ align_off = orig_off;
+ align_alen -= align_alen % mp->m_sb.sb_rextsize;
+ }
+ /*
+ * Result doesn't cover the request, fail it.
+ */
+ if (orig_off < align_off || orig_end > align_off + align_alen)
+ return XFS_ERROR(EINVAL);
+ } else {
+ ASSERT(orig_off >= align_off);
+ ASSERT(orig_end <= align_off + align_alen);
}
- *first_unused = max;
+
+#ifdef DEBUG
+ if (!eof && gotp->br_startoff != NULLFILEOFF)
+ ASSERT(align_off + align_alen <= gotp->br_startoff);
+ if (prevp->br_startoff != NULLFILEOFF)
+ ASSERT(align_off >= prevp->br_startoff + prevp->br_blockcount);
+#endif
+
+ *lenp = align_alen;
+ *offp = align_off;
return 0;
}
-/*
- * Returns the file-relative block number of the last block + 1 before
- * last_block (input value) in the file.
- * This is not based on i_size, it is based on the extent records.
- * Returns 0 for local files, as they do not have extent records.
- */
-int /* error */
-xfs_bmap_last_before(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode */
- xfs_fileoff_t *last_block, /* last block */
- int whichfork) /* data or attr fork */
+#define XFS_ALLOC_GAP_UNITS 4
+
+STATIC void
+xfs_bmap_adjacent(
+ xfs_bmalloca_t *ap) /* bmap alloc argument struct */
{
- xfs_fileoff_t bno; /* input file offset */
- int eof; /* hit end of file */
- xfs_bmbt_rec_host_t *ep; /* pointer to last extent */
- int error; /* error return value */
- xfs_bmbt_irec_t got; /* current extent value */
- xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_extnum_t lastx; /* last extent used */
- xfs_bmbt_irec_t prev; /* previous extent value */
+ xfs_fsblock_t adjust; /* adjustment to block numbers */
+ xfs_agnumber_t fb_agno; /* ag number of ap->firstblock */
+ xfs_mount_t *mp; /* mount point structure */
+ int nullfb; /* true if ap->firstblock isn't set */
+ int rt; /* true if inode is realtime */
- if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE &&
- XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
- XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_LOCAL)
- return XFS_ERROR(EIO);
- if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL) {
- *last_block = 0;
- return 0;
+#define ISVALID(x,y) \
+ (rt ? \
+ (x) < mp->m_sb.sb_rblocks : \
+ XFS_FSB_TO_AGNO(mp, x) == XFS_FSB_TO_AGNO(mp, y) && \
+ XFS_FSB_TO_AGNO(mp, x) < mp->m_sb.sb_agcount && \
+ XFS_FSB_TO_AGBNO(mp, x) < mp->m_sb.sb_agblocks)
+
+ mp = ap->ip->i_mount;
+ nullfb = *ap->firstblock == NULLFSBLOCK;
+ rt = XFS_IS_REALTIME_INODE(ap->ip) && ap->userdata;
+ fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp, *ap->firstblock);
+ /*
+ * If allocating at eof, and there's a previous real block,
+ * try to use its last block as our starting point.
+ */
+ if (ap->eof && ap->prev.br_startoff != NULLFILEOFF &&
+ !isnullstartblock(ap->prev.br_startblock) &&
+ ISVALID(ap->prev.br_startblock + ap->prev.br_blockcount,
+ ap->prev.br_startblock)) {
+ ap->blkno = ap->prev.br_startblock + ap->prev.br_blockcount;
+ /*
+ * Adjust for the gap between prevp and us.
+ */
+ adjust = ap->offset -
+ (ap->prev.br_startoff + ap->prev.br_blockcount);
+ if (adjust &&
+ ISVALID(ap->blkno + adjust, ap->prev.br_startblock))
+ ap->blkno += adjust;
}
- ifp = XFS_IFORK_PTR(ip, whichfork);
- if (!(ifp->if_flags & XFS_IFEXTENTS) &&
- (error = xfs_iread_extents(tp, ip, whichfork)))
- return error;
- bno = *last_block - 1;
- ep = xfs_bmap_search_extents(ip, bno, whichfork, &eof, &lastx, &got,
- &prev);
- if (eof || xfs_bmbt_get_startoff(ep) > bno) {
- if (prev.br_startoff == NULLFILEOFF)
- *last_block = 0;
+ /*
+ * If not at eof, then compare the two neighbor blocks.
+ * Figure out whether either one gives us a good starting point,
+ * and pick the better one.
+ */
+ else if (!ap->eof) {
+ xfs_fsblock_t gotbno; /* right side block number */
+ xfs_fsblock_t gotdiff=0; /* right side difference */
+ xfs_fsblock_t prevbno; /* left side block number */
+ xfs_fsblock_t prevdiff=0; /* left side difference */
+
+ /*
+ * If there's a previous (left) block, select a requested
+ * start block based on it.
+ */
+ if (ap->prev.br_startoff != NULLFILEOFF &&
+ !isnullstartblock(ap->prev.br_startblock) &&
+ (prevbno = ap->prev.br_startblock +
+ ap->prev.br_blockcount) &&
+ ISVALID(prevbno, ap->prev.br_startblock)) {
+ /*
+ * Calculate gap to end of previous block.
+ */
+ adjust = prevdiff = ap->offset -
+ (ap->prev.br_startoff +
+ ap->prev.br_blockcount);
+ /*
+ * Figure the startblock based on the previous block's
+ * end and the gap size.
+ * Heuristic!
+ * If the gap is large relative to the piece we're
+ * allocating, or using it gives us an invalid block
+ * number, then just use the end of the previous block.
+ */
+ if (prevdiff <= XFS_ALLOC_GAP_UNITS * ap->length &&
+ ISVALID(prevbno + prevdiff,
+ ap->prev.br_startblock))
+ prevbno += adjust;
+ else
+ prevdiff += adjust;
+ /*
+ * If the firstblock forbids it, can't use it,
+ * must use default.
+ */
+ if (!rt && !nullfb &&
+ XFS_FSB_TO_AGNO(mp, prevbno) != fb_agno)
+ prevbno = NULLFSBLOCK;
+ }
+ /*
+ * No previous block or can't follow it, just default.
+ */
+ else
+ prevbno = NULLFSBLOCK;
+ /*
+ * If there's a following (right) block, select a requested
+ * start block based on it.
+ */
+ if (!isnullstartblock(ap->got.br_startblock)) {
+ /*
+ * Calculate gap to start of next block.
+ */
+ adjust = gotdiff = ap->got.br_startoff - ap->offset;
+ /*
+ * Figure the startblock based on the next block's
+ * start and the gap size.
+ */
+ gotbno = ap->got.br_startblock;
+ /*
+ * Heuristic!
+ * If the gap is large relative to the piece we're
+ * allocating, or using it gives us an invalid block
+ * number, then just use the start of the next block
+ * offset by our length.
+ */
+ if (gotdiff <= XFS_ALLOC_GAP_UNITS * ap->length &&
+ ISVALID(gotbno - gotdiff, gotbno))
+ gotbno -= adjust;
+ else if (ISVALID(gotbno - ap->length, gotbno)) {
+ gotbno -= ap->length;
+ gotdiff += adjust - ap->length;
+ } else
+ gotdiff += adjust;
+ /*
+ * If the firstblock forbids it, can't use it,
+ * must use default.
+ */
+ if (!rt && !nullfb &&
+ XFS_FSB_TO_AGNO(mp, gotbno) != fb_agno)
+ gotbno = NULLFSBLOCK;
+ }
+ /*
+ * No next block, just default.
+ */
else
- *last_block = prev.br_startoff + prev.br_blockcount;
- }
- /*
- * Otherwise *last_block is already the right answer.
- */
- return 0;
-}
-
-STATIC int
-xfs_bmap_last_extent(
- struct xfs_trans *tp,
- struct xfs_inode *ip,
- int whichfork,
- struct xfs_bmbt_irec *rec,
- int *is_empty)
-{
- struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork);
- int error;
- int nextents;
-
- if (!(ifp->if_flags & XFS_IFEXTENTS)) {
- error = xfs_iread_extents(tp, ip, whichfork);
- if (error)
- return error;
- }
-
- nextents = ifp->if_bytes / sizeof(xfs_bmbt_rec_t);
- if (nextents == 0) {
- *is_empty = 1;
- return 0;
+ gotbno = NULLFSBLOCK;
+ /*
+ * If both valid, pick the better one, else the only good
+ * one, else ap->blkno is already set (to 0 or the inode block).
+ */
+ if (prevbno != NULLFSBLOCK && gotbno != NULLFSBLOCK)
+ ap->blkno = prevdiff <= gotdiff ? prevbno : gotbno;
+ else if (prevbno != NULLFSBLOCK)
+ ap->blkno = prevbno;
+ else if (gotbno != NULLFSBLOCK)
+ ap->blkno = gotbno;
}
-
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp, nextents - 1), rec);
- *is_empty = 0;
- return 0;
+#undef ISVALID
}
-/*
- * Check the last inode extent to determine whether this allocation will result
- * in blocks being allocated at the end of the file. When we allocate new data
- * blocks at the end of the file which do not start at the previous data block,
- * we will try to align the new blocks at stripe unit boundaries.
- *
- * Returns 0 in bma->aeof if the file (fork) is empty as any new write will be
- * at, or past the EOF.
- */
STATIC int
-xfs_bmap_isaeof(
- struct xfs_bmalloca *bma,
- int whichfork)
+xfs_bmap_btalloc_nullfb(
+ struct xfs_bmalloca *ap,
+ struct xfs_alloc_arg *args,
+ xfs_extlen_t *blen)
{
- struct xfs_bmbt_irec rec;
- int is_empty;
+ struct xfs_mount *mp = ap->ip->i_mount;
+ struct xfs_perag *pag;
+ xfs_agnumber_t ag, startag;
+ int notinit = 0;
int error;
- bma->aeof = 0;
- error = xfs_bmap_last_extent(NULL, bma->ip, whichfork, &rec,
- &is_empty);
- if (error || is_empty)
- return error;
+ if (ap->userdata && xfs_inode_is_filestream(ap->ip))
+ args->type = XFS_ALLOCTYPE_NEAR_BNO;
+ else
+ args->type = XFS_ALLOCTYPE_START_BNO;
+ args->total = ap->total;
/*
- * Check if we are allocation or past the last extent, or at least into
- * the last delayed allocated extent.
+ * Search for an allocation group with a single extent large enough
+ * for the request. If one isn't found, then adjust the minimum
+ * allocation size to the largest space found.
*/
- bma->aeof = bma->offset >= rec.br_startoff + rec.br_blockcount ||
- (bma->offset >= rec.br_startoff &&
- isnullstartblock(rec.br_startblock));
- return 0;
-}
-
-/*
- * Check if the endoff is outside the last extent. If so the caller will grow
- * the allocation to a stripe unit boundary. All offsets are considered outside
- * the end of file for an empty fork, so 1 is returned in *eof in that case.
- */
-int
-xfs_bmap_eof(
- struct xfs_inode *ip,
- xfs_fileoff_t endoff,
- int whichfork,
- int *eof)
-{
- struct xfs_bmbt_irec rec;
- int error;
-
- error = xfs_bmap_last_extent(NULL, ip, whichfork, &rec, eof);
- if (error || *eof)
- return error;
-
- *eof = endoff >= rec.br_startoff + rec.br_blockcount;
- return 0;
-}
+ startag = ag = XFS_FSB_TO_AGNO(mp, args->fsbno);
+ if (startag == NULLAGNUMBER)
+ startag = ag = 0;
-/*
- * Returns the file-relative block number of the first block past eof in
- * the file. This is not based on i_size, it is based on the extent records.
- * Returns 0 for local files, as they do not have extent records.
- */
-int
-xfs_bmap_last_offset(
- struct xfs_trans *tp,
- struct xfs_inode *ip,
- xfs_fileoff_t *last_block,
- int whichfork)
-{
- struct xfs_bmbt_irec rec;
- int is_empty;
- int error;
+ pag = xfs_perag_get(mp, ag);
+ while (*blen < args->maxlen) {
+ if (!pag->pagf_init) {
+ error = xfs_alloc_pagf_init(mp, args->tp, ag,
+ XFS_ALLOC_FLAG_TRYLOCK);
+ if (error) {
+ xfs_perag_put(pag);
+ return error;
+ }
+ }
- *last_block = 0;
+ /*
+ * See xfs_alloc_fix_freelist...
+ */
+ if (pag->pagf_init) {
+ xfs_extlen_t longest;
+ longest = xfs_alloc_longest_free_extent(mp, pag);
+ if (*blen < longest)
+ *blen = longest;
+ } else
+ notinit = 1;
- if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL)
- return 0;
+ if (xfs_inode_is_filestream(ap->ip)) {
+ if (*blen >= args->maxlen)
+ break;
- if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE &&
- XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS)
- return XFS_ERROR(EIO);
+ if (ap->userdata) {
+ /*
+ * If startag is an invalid AG, we've
+ * come here once before and
+ * xfs_filestream_new_ag picked the
+ * best currently available.
+ *
+ * Don't continue looping, since we
+ * could loop forever.
+ */
+ if (startag == NULLAGNUMBER)
+ break;
- error = xfs_bmap_last_extent(NULL, ip, whichfork, &rec, &is_empty);
- if (error || is_empty)
- return error;
+ error = xfs_filestream_new_ag(ap, &ag);
+ xfs_perag_put(pag);
+ if (error)
+ return error;
- *last_block = rec.br_startoff + rec.br_blockcount;
- return 0;
-}
+ /* loop again to set 'blen'*/
+ startag = NULLAGNUMBER;
+ pag = xfs_perag_get(mp, ag);
+ continue;
+ }
+ }
+ if (++ag == mp->m_sb.sb_agcount)
+ ag = 0;
+ if (ag == startag)
+ break;
+ xfs_perag_put(pag);
+ pag = xfs_perag_get(mp, ag);
+ }
+ xfs_perag_put(pag);
-/*
- * Returns whether the selected fork of the inode has exactly one
- * block or not. For the data fork we check this matches di_size,
- * implying the file's range is 0..bsize-1.
- */
-int /* 1=>1 block, 0=>otherwise */
-xfs_bmap_one_block(
- xfs_inode_t *ip, /* incore inode */
- int whichfork) /* data or attr fork */
-{
- xfs_bmbt_rec_host_t *ep; /* ptr to fork's extent */
- xfs_ifork_t *ifp; /* inode fork pointer */
- int rval; /* return value */
- xfs_bmbt_irec_t s; /* internal version of extent */
+ /*
+ * Since the above loop did a BUF_TRYLOCK, it is
+ * possible that there is space for this request.
+ */
+ if (notinit || *blen < ap->minlen)
+ args->minlen = ap->minlen;
+ /*
+ * If the best seen length is less than the request
+ * length, use the best as the minimum.
+ */
+ else if (*blen < args->maxlen)
+ args->minlen = *blen;
+ /*
+ * Otherwise we've seen an extent as big as maxlen,
+ * use that as the minimum.
+ */
+ else
+ args->minlen = args->maxlen;
-#ifndef DEBUG
- if (whichfork == XFS_DATA_FORK)
- return XFS_ISIZE(ip) == ip->i_mount->m_sb.sb_blocksize;
-#endif /* !DEBUG */
- if (XFS_IFORK_NEXTENTS(ip, whichfork) != 1)
- return 0;
- if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS)
- return 0;
- ifp = XFS_IFORK_PTR(ip, whichfork);
- ASSERT(ifp->if_flags & XFS_IFEXTENTS);
- ep = xfs_iext_get_ext(ifp, 0);
- xfs_bmbt_get_all(ep, &s);
- rval = s.br_startoff == 0 && s.br_blockcount == 1;
- if (rval && whichfork == XFS_DATA_FORK)
- ASSERT(XFS_ISIZE(ip) == ip->i_mount->m_sb.sb_blocksize);
- return rval;
+ /*
+ * set the failure fallback case to look in the selected
+ * AG as the stream may have moved.
+ */
+ if (xfs_inode_is_filestream(ap->ip))
+ ap->blkno = args->fsbno = XFS_AGB_TO_FSB(mp, ag, 0);
+
+ return 0;
}
STATIC int
-xfs_bmap_sanity_check(
- struct xfs_mount *mp,
- struct xfs_buf *bp,
- int level)
+xfs_bmap_btalloc(
+ xfs_bmalloca_t *ap) /* bmap alloc argument struct */
{
- struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
+ xfs_mount_t *mp; /* mount point structure */
+ xfs_alloctype_t atype = 0; /* type for allocation routines */
+ xfs_extlen_t align; /* minimum allocation alignment */
+ xfs_agnumber_t fb_agno; /* ag number of ap->firstblock */
+ xfs_agnumber_t ag;
+ xfs_alloc_arg_t args;
+ xfs_extlen_t blen;
+ xfs_extlen_t nextminlen = 0;
+ int nullfb; /* true if ap->firstblock isn't set */
+ int isaligned;
+ int tryagain;
+ int error;
- if (block->bb_magic != cpu_to_be32(XFS_BMAP_MAGIC) ||
- be16_to_cpu(block->bb_level) != level ||
- be16_to_cpu(block->bb_numrecs) == 0 ||
- be16_to_cpu(block->bb_numrecs) > mp->m_bmap_dmxr[level != 0])
- return 0;
- return 1;
-}
+ ASSERT(ap->length);
-/*
- * Read in the extents to if_extents.
- * All inode fields are set up by caller, we just traverse the btree
- * and copy the records in. If the file system cannot contain unwritten
- * extents, the records are checked for no "state" flags.
- */
-int /* error */
-xfs_bmap_read_extents(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode */
- int whichfork) /* data or attr fork */
-{
- struct xfs_btree_block *block; /* current btree block */
- xfs_fsblock_t bno; /* block # of "block" */
- xfs_buf_t *bp; /* buffer for "block" */
- int error; /* error return value */
- xfs_exntfmt_t exntf; /* XFS_EXTFMT_NOSTATE, if checking */
- xfs_extnum_t i, j; /* index into the extents list */
- xfs_ifork_t *ifp; /* fork structure */
- int level; /* btree level, for checking */
- xfs_mount_t *mp; /* file system mount structure */
- __be64 *pp; /* pointer to block address */
- /* REFERENCED */
- xfs_extnum_t room; /* number of entries there's room for */
+ mp = ap->ip->i_mount;
+ align = ap->userdata ? xfs_get_extsz_hint(ap->ip) : 0;
+ if (unlikely(align)) {
+ error = xfs_bmap_extsize_align(mp, &ap->got, &ap->prev,
+ align, 0, ap->eof, 0, ap->conv,
+ &ap->offset, &ap->length);
+ ASSERT(!error);
+ ASSERT(ap->length);
+ }
+ nullfb = *ap->firstblock == NULLFSBLOCK;
+ fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp, *ap->firstblock);
+ if (nullfb) {
+ if (ap->userdata && xfs_inode_is_filestream(ap->ip)) {
+ ag = xfs_filestream_lookup_ag(ap->ip);
+ ag = (ag != NULLAGNUMBER) ? ag : 0;
+ ap->blkno = XFS_AGB_TO_FSB(mp, ag, 0);
+ } else {
+ ap->blkno = XFS_INO_TO_FSB(mp, ap->ip->i_ino);
+ }
+ } else
+ ap->blkno = *ap->firstblock;
+
+ xfs_bmap_adjacent(ap);
- bno = NULLFSBLOCK;
- mp = ip->i_mount;
- ifp = XFS_IFORK_PTR(ip, whichfork);
- exntf = (whichfork != XFS_DATA_FORK) ? XFS_EXTFMT_NOSTATE :
- XFS_EXTFMT_INODE(ip);
- block = ifp->if_broot;
/*
- * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out.
+ * If allowed, use ap->blkno; otherwise must use firstblock since
+ * it's in the right allocation group.
*/
- level = be16_to_cpu(block->bb_level);
- ASSERT(level > 0);
- pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
- bno = be64_to_cpu(*pp);
- ASSERT(bno != NULLDFSBNO);
- ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount);
- ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks);
+ if (nullfb || XFS_FSB_TO_AGNO(mp, ap->blkno) == fb_agno)
+ ;
+ else
+ ap->blkno = *ap->firstblock;
/*
- * Go down the tree until leaf level is reached, following the first
- * pointer (leftmost) at each level.
+ * Normal allocation, done through xfs_alloc_vextent.
*/
- while (level-- > 0) {
- error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp,
- XFS_BMAP_BTREE_REF, &xfs_bmbt_buf_ops);
+ tryagain = isaligned = 0;
+ memset(&args, 0, sizeof(args));
+ args.tp = ap->tp;
+ args.mp = mp;
+ args.fsbno = ap->blkno;
+
+ /* Trim the allocation back to the maximum an AG can fit. */
+ args.maxlen = MIN(ap->length, XFS_ALLOC_AG_MAX_USABLE(mp));
+ args.firstblock = *ap->firstblock;
+ blen = 0;
+ if (nullfb) {
+ error = xfs_bmap_btalloc_nullfb(ap, &args, &blen);
if (error)
return error;
- block = XFS_BUF_TO_BLOCK(bp);
- XFS_WANT_CORRUPTED_GOTO(
- xfs_bmap_sanity_check(mp, bp, level),
- error0);
- if (level == 0)
- break;
- pp = XFS_BMBT_PTR_ADDR(mp, block, 1, mp->m_bmap_dmxr[1]);
- bno = be64_to_cpu(*pp);
- XFS_WANT_CORRUPTED_GOTO(XFS_FSB_SANITY_CHECK(mp, bno), error0);
- xfs_trans_brelse(tp, bp);
+ } else if (ap->flist->xbf_low) {
+ if (xfs_inode_is_filestream(ap->ip))
+ args.type = XFS_ALLOCTYPE_FIRST_AG;
+ else
+ args.type = XFS_ALLOCTYPE_START_BNO;
+ args.total = args.minlen = ap->minlen;
+ } else {
+ args.type = XFS_ALLOCTYPE_NEAR_BNO;
+ args.total = ap->total;
+ args.minlen = ap->minlen;
+ }
+ /* apply extent size hints if obtained earlier */
+ if (unlikely(align)) {
+ args.prod = align;
+ if ((args.mod = (xfs_extlen_t)do_mod(ap->offset, args.prod)))
+ args.mod = (xfs_extlen_t)(args.prod - args.mod);
+ } else if (mp->m_sb.sb_blocksize >= PAGE_CACHE_SIZE) {
+ args.prod = 1;
+ args.mod = 0;
+ } else {
+ args.prod = PAGE_CACHE_SIZE >> mp->m_sb.sb_blocklog;
+ if ((args.mod = (xfs_extlen_t)(do_mod(ap->offset, args.prod))))
+ args.mod = (xfs_extlen_t)(args.prod - args.mod);
}
/*
- * Here with bp and block set to the leftmost leaf node in the tree.
- */
- room = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
- i = 0;
- /*
- * Loop over all leaf nodes. Copy information to the extent records.
+ * If we are not low on available data blocks, and the
+ * underlying logical volume manager is a stripe, and
+ * the file offset is zero then try to allocate data
+ * blocks on stripe unit boundary.
+ * NOTE: ap->aeof is only set if the allocation length
+ * is >= the stripe unit and the allocation offset is
+ * at the end of file.
*/
- for (;;) {
- xfs_bmbt_rec_t *frp;
- xfs_fsblock_t nextbno;
- xfs_extnum_t num_recs;
- xfs_extnum_t start;
-
- num_recs = xfs_btree_get_numrecs(block);
- if (unlikely(i + num_recs > room)) {
- ASSERT(i + num_recs <= room);
- xfs_warn(ip->i_mount,
- "corrupt dinode %Lu, (btree extents).",
- (unsigned long long) ip->i_ino);
- XFS_CORRUPTION_ERROR("xfs_bmap_read_extents(1)",
- XFS_ERRLEVEL_LOW, ip->i_mount, block);
- goto error0;
+ if (!ap->flist->xbf_low && ap->aeof) {
+ if (!ap->offset) {
+ args.alignment = mp->m_dalign;
+ atype = args.type;
+ isaligned = 1;
+ /*
+ * Adjust for alignment
+ */
+ if (blen > args.alignment && blen <= args.maxlen)
+ args.minlen = blen - args.alignment;
+ args.minalignslop = 0;
+ } else {
+ /*
+ * First try an exact bno allocation.
+ * If it fails then do a near or start bno
+ * allocation with alignment turned on.
+ */
+ atype = args.type;
+ tryagain = 1;
+ args.type = XFS_ALLOCTYPE_THIS_BNO;
+ args.alignment = 1;
+ /*
+ * Compute the minlen+alignment for the
+ * next case. Set slop so that the value
+ * of minlen+alignment+slop doesn't go up
+ * between the calls.
+ */
+ if (blen > mp->m_dalign && blen <= args.maxlen)
+ nextminlen = blen - mp->m_dalign;
+ else
+ nextminlen = args.minlen;
+ if (nextminlen + mp->m_dalign > args.minlen + 1)
+ args.minalignslop =
+ nextminlen + mp->m_dalign -
+ args.minlen - 1;
+ else
+ args.minalignslop = 0;
}
- XFS_WANT_CORRUPTED_GOTO(
- xfs_bmap_sanity_check(mp, bp, 0),
- error0);
+ } else {
+ args.alignment = 1;
+ args.minalignslop = 0;
+ }
+ args.minleft = ap->minleft;
+ args.wasdel = ap->wasdel;
+ args.isfl = 0;
+ args.userdata = ap->userdata;
+ if ((error = xfs_alloc_vextent(&args)))
+ return error;
+ if (tryagain && args.fsbno == NULLFSBLOCK) {
+ /*
+ * Exact allocation failed. Now try with alignment
+ * turned on.
+ */
+ args.type = atype;
+ args.fsbno = ap->blkno;
+ args.alignment = mp->m_dalign;
+ args.minlen = nextminlen;
+ args.minalignslop = 0;
+ isaligned = 1;
+ if ((error = xfs_alloc_vextent(&args)))
+ return error;
+ }
+ if (isaligned && args.fsbno == NULLFSBLOCK) {
/*
- * Read-ahead the next leaf block, if any.
+ * allocation failed, so turn off alignment and
+ * try again.
*/
- nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
- if (nextbno != NULLFSBLOCK)
- xfs_btree_reada_bufl(mp, nextbno, 1,
- &xfs_bmbt_buf_ops);
+ args.type = atype;
+ args.fsbno = ap->blkno;
+ args.alignment = 0;
+ if ((error = xfs_alloc_vextent(&args)))
+ return error;
+ }
+ if (args.fsbno == NULLFSBLOCK && nullfb &&
+ args.minlen > ap->minlen) {
+ args.minlen = ap->minlen;
+ args.type = XFS_ALLOCTYPE_START_BNO;
+ args.fsbno = ap->blkno;
+ if ((error = xfs_alloc_vextent(&args)))
+ return error;
+ }
+ if (args.fsbno == NULLFSBLOCK && nullfb) {
+ args.fsbno = 0;
+ args.type = XFS_ALLOCTYPE_FIRST_AG;
+ args.total = ap->minlen;
+ args.minleft = 0;
+ if ((error = xfs_alloc_vextent(&args)))
+ return error;
+ ap->flist->xbf_low = 1;
+ }
+ if (args.fsbno != NULLFSBLOCK) {
/*
- * Copy records into the extent records.
+ * check the allocation happened at the same or higher AG than
+ * the first block that was allocated.
*/
- frp = XFS_BMBT_REC_ADDR(mp, block, 1);
- start = i;
- for (j = 0; j < num_recs; j++, i++, frp++) {
- xfs_bmbt_rec_host_t *trp = xfs_iext_get_ext(ifp, i);
- trp->l0 = be64_to_cpu(frp->l0);
- trp->l1 = be64_to_cpu(frp->l1);
- }
- if (exntf == XFS_EXTFMT_NOSTATE) {
- /*
- * Check all attribute bmap btree records and
- * any "older" data bmap btree records for a
- * set bit in the "extent flag" position.
- */
- if (unlikely(xfs_check_nostate_extents(ifp,
- start, num_recs))) {
- XFS_ERROR_REPORT("xfs_bmap_read_extents(2)",
- XFS_ERRLEVEL_LOW,
- ip->i_mount);
- goto error0;
- }
- }
- xfs_trans_brelse(tp, bp);
- bno = nextbno;
+ ASSERT(*ap->firstblock == NULLFSBLOCK ||
+ XFS_FSB_TO_AGNO(mp, *ap->firstblock) ==
+ XFS_FSB_TO_AGNO(mp, args.fsbno) ||
+ (ap->flist->xbf_low &&
+ XFS_FSB_TO_AGNO(mp, *ap->firstblock) <
+ XFS_FSB_TO_AGNO(mp, args.fsbno)));
+
+ ap->blkno = args.fsbno;
+ if (*ap->firstblock == NULLFSBLOCK)
+ *ap->firstblock = args.fsbno;
+ ASSERT(nullfb || fb_agno == args.agno ||
+ (ap->flist->xbf_low && fb_agno < args.agno));
+ ap->length = args.len;
+ ap->ip->i_d.di_nblocks += args.len;
+ xfs_trans_log_inode(ap->tp, ap->ip, XFS_ILOG_CORE);
+ if (ap->wasdel)
+ ap->ip->i_delayed_blks -= args.len;
/*
- * If we've reached the end, stop.
+ * Adjust the disk quota also. This was reserved
+ * earlier.
*/
- if (bno == NULLFSBLOCK)
- break;
- error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp,
- XFS_BMAP_BTREE_REF, &xfs_bmbt_buf_ops);
- if (error)
- return error;
- block = XFS_BUF_TO_BLOCK(bp);
+ xfs_trans_mod_dquot_byino(ap->tp, ap->ip,
+ ap->wasdel ? XFS_TRANS_DQ_DELBCOUNT :
+ XFS_TRANS_DQ_BCOUNT,
+ (long) args.len);
+ } else {
+ ap->blkno = NULLFSBLOCK;
+ ap->length = 0;
}
- ASSERT(i == (ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t)));
- ASSERT(i == XFS_IFORK_NEXTENTS(ip, whichfork));
- XFS_BMAP_TRACE_EXLIST(ip, i, whichfork);
return 0;
-error0:
- xfs_trans_brelse(tp, bp);
- return XFS_ERROR(EFSCORRUPTED);
-}
-
-#ifdef DEBUG
-/*
- * Add bmap trace insert entries for all the contents of the extent records.
- */
-void
-xfs_bmap_trace_exlist(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_extnum_t cnt, /* count of entries in the list */
- int whichfork, /* data or attr fork */
- unsigned long caller_ip)
-{
- xfs_extnum_t idx; /* extent record index */
- xfs_ifork_t *ifp; /* inode fork pointer */
- int state = 0;
-
- if (whichfork == XFS_ATTR_FORK)
- state |= BMAP_ATTRFORK;
-
- ifp = XFS_IFORK_PTR(ip, whichfork);
- ASSERT(cnt == (ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t)));
- for (idx = 0; idx < cnt; idx++)
- trace_xfs_extlist(ip, idx, whichfork, caller_ip);
}
/*
- * Validate that the bmbt_irecs being returned from bmapi are valid
- * given the callers original parameters. Specifically check the
- * ranges of the returned irecs to ensure that they only extent beyond
- * the given parameters if the XFS_BMAPI_ENTIRE flag was set.
+ * xfs_bmap_alloc is called by xfs_bmapi to allocate an extent for a file.
+ * It figures out where to ask the underlying allocator to put the new extent.
*/
-STATIC void
-xfs_bmap_validate_ret(
- xfs_fileoff_t bno,
- xfs_filblks_t len,
- int flags,
- xfs_bmbt_irec_t *mval,
- int nmap,
- int ret_nmap)
+STATIC int
+xfs_bmap_alloc(
+ xfs_bmalloca_t *ap) /* bmap alloc argument struct */
{
- int i; /* index to map values */
-
- ASSERT(ret_nmap <= nmap);
-
- for (i = 0; i < ret_nmap; i++) {
- ASSERT(mval[i].br_blockcount > 0);
- if (!(flags & XFS_BMAPI_ENTIRE)) {
- ASSERT(mval[i].br_startoff >= bno);
- ASSERT(mval[i].br_blockcount <= len);
- ASSERT(mval[i].br_startoff + mval[i].br_blockcount <=
- bno + len);
- } else {
- ASSERT(mval[i].br_startoff < bno + len);
- ASSERT(mval[i].br_startoff + mval[i].br_blockcount >
- bno);
- }
- ASSERT(i == 0 ||
- mval[i - 1].br_startoff + mval[i - 1].br_blockcount ==
- mval[i].br_startoff);
- ASSERT(mval[i].br_startblock != DELAYSTARTBLOCK &&
- mval[i].br_startblock != HOLESTARTBLOCK);
- ASSERT(mval[i].br_state == XFS_EXT_NORM ||
- mval[i].br_state == XFS_EXT_UNWRITTEN);
- }
+ if (XFS_IS_REALTIME_INODE(ap->ip) && ap->userdata)
+ return xfs_bmap_rtalloc(ap);
+ return xfs_bmap_btalloc(ap);
}
-#endif /* DEBUG */
-
/*
* Trim the returned map to the required bounds
@@ -4385,9 +4422,12 @@ xfs_bmapi_allocate(
struct xfs_ifork *ifp = XFS_IFORK_PTR(bma->ip, whichfork);
int tmp_logflags = 0;
int error;
+ int rt;
ASSERT(bma->length > 0);
+ rt = (whichfork == XFS_DATA_FORK) && XFS_IS_REALTIME_INODE(bma->ip);
+
/*
* For the wasdelay case, we could also just allocate the stuff asked
* for in this bmap call but that wouldn't be as good.
@@ -4428,9 +4468,6 @@ xfs_bmapi_allocate(
return error;
}
- if (bma->flags & XFS_BMAPI_STACK_SWITCH)
- bma->stack_switch = 1;
-
error = xfs_bmap_alloc(bma);
if (error)
return error;
@@ -4493,7 +4530,6 @@ xfs_bmapi_allocate(
return 0;
}
-
STATIC int
xfs_bmapi_convert_unwritten(
struct xfs_bmalloca *bma,
@@ -4632,8 +4668,32 @@ xfs_bmapi_write(
XFS_STATS_INC(xs_blk_mapw);
if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL) {
+ /*
+ * XXX (dgc): This assumes we are only called for inodes that
+ * contain content neutral data in local format. Anything that
+ * contains caller-specific data in local format that needs
+ * transformation to move to a block format needs to do the
+ * conversion to extent format itself.
+ *
+ * Directory data forks and attribute forks handle this
+ * themselves, but with the addition of metadata verifiers every
+ * data fork in local format now contains caller specific data
+ * and as such conversion through this function is likely to be
+ * broken.
+ *
+ * The only likely user of this branch is for remote symlinks,
+ * but we cannot overwrite the data fork contents of the symlink
+ * (EEXIST occurs higher up the stack) and so it will never go
+ * from local format to extent format here. Hence I don't think
+ * this branch is ever executed intentionally and we should
+ * consider removing it and asserting that xfs_bmapi_write()
+ * cannot be called directly on local format forks. i.e. callers
+ * are completely responsible for local to extent format
+ * conversion, not xfs_bmapi_write().
+ */
error = xfs_bmap_local_to_extents(tp, ip, firstblock, total,
- &bma.logflags, whichfork);
+ &bma.logflags, whichfork,
+ xfs_bmap_local_to_extents_init_fn);
if (error)
goto error0;
}
@@ -4666,6 +4726,9 @@ xfs_bmapi_write(
bma.flist = flist;
bma.firstblock = firstblock;
+ if (flags & XFS_BMAPI_STACK_SWITCH)
+ bma.stack_switch = 1;
+
while (bno < end && n < *nmap) {
inhole = eof || bma.got.br_startoff > bno;
wasdelay = !inhole && isnullstartblock(bma.got.br_startblock);
@@ -4793,804 +4856,703 @@ error0:
}
/*
- * Unmap (remove) blocks from a file.
- * If nexts is nonzero then the number of extents to remove is limited to
- * that value. If not all extents in the block range can be removed then
- * *done is set.
+ * Called by xfs_bmapi to update file extent records and the btree
+ * after removing space (or undoing a delayed allocation).
*/
-int /* error */
-xfs_bunmapi(
- xfs_trans_t *tp, /* transaction pointer */
- struct xfs_inode *ip, /* incore inode */
- xfs_fileoff_t bno, /* starting offset to unmap */
- xfs_filblks_t len, /* length to unmap in file */
- int flags, /* misc flags */
- xfs_extnum_t nexts, /* number of extents max */
- xfs_fsblock_t *firstblock, /* first allocated block
- controls a.g. for allocs */
- xfs_bmap_free_t *flist, /* i/o: list extents to free */
- int *done) /* set if not done yet */
+STATIC int /* error */
+xfs_bmap_del_extent(
+ xfs_inode_t *ip, /* incore inode pointer */
+ xfs_trans_t *tp, /* current transaction pointer */
+ xfs_extnum_t *idx, /* extent number to update/delete */
+ xfs_bmap_free_t *flist, /* list of extents to be freed */
+ xfs_btree_cur_t *cur, /* if null, not a btree */
+ xfs_bmbt_irec_t *del, /* data to remove from extents */
+ int *logflagsp, /* inode logging flags */
+ int whichfork) /* data or attr fork */
{
- xfs_btree_cur_t *cur; /* bmap btree cursor */
- xfs_bmbt_irec_t del; /* extent being deleted */
- int eof; /* is deleting at eof */
- xfs_bmbt_rec_host_t *ep; /* extent record pointer */
- int error; /* error return value */
- xfs_extnum_t extno; /* extent number in list */
- xfs_bmbt_irec_t got; /* current extent record */
- xfs_ifork_t *ifp; /* inode fork pointer */
- int isrt; /* freeing in rt area */
- xfs_extnum_t lastx; /* last extent index used */
- int logflags; /* transaction logging flags */
- xfs_extlen_t mod; /* rt extent offset */
- xfs_mount_t *mp; /* mount structure */
- xfs_extnum_t nextents; /* number of file extents */
- xfs_bmbt_irec_t prev; /* previous extent record */
- xfs_fileoff_t start; /* first file offset deleted */
- int tmp_logflags; /* partial logging flags */
- int wasdel; /* was a delayed alloc extent */
- int whichfork; /* data or attribute fork */
- xfs_fsblock_t sum;
-
- trace_xfs_bunmap(ip, bno, len, flags, _RET_IP_);
-
- whichfork = (flags & XFS_BMAPI_ATTRFORK) ?
- XFS_ATTR_FORK : XFS_DATA_FORK;
- ifp = XFS_IFORK_PTR(ip, whichfork);
- if (unlikely(
- XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
- XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)) {
- XFS_ERROR_REPORT("xfs_bunmapi", XFS_ERRLEVEL_LOW,
- ip->i_mount);
- return XFS_ERROR(EFSCORRUPTED);
- }
- mp = ip->i_mount;
- if (XFS_FORCED_SHUTDOWN(mp))
- return XFS_ERROR(EIO);
+ xfs_filblks_t da_new; /* new delay-alloc indirect blocks */
+ xfs_filblks_t da_old; /* old delay-alloc indirect blocks */
+ xfs_fsblock_t del_endblock=0; /* first block past del */
+ xfs_fileoff_t del_endoff; /* first offset past del */
+ int delay; /* current block is delayed allocated */
+ int do_fx; /* free extent at end of routine */
+ xfs_bmbt_rec_host_t *ep; /* current extent entry pointer */
+ int error; /* error return value */
+ int flags; /* inode logging flags */
+ xfs_bmbt_irec_t got; /* current extent entry */
+ xfs_fileoff_t got_endoff; /* first offset past got */
+ int i; /* temp state */
+ xfs_ifork_t *ifp; /* inode fork pointer */
+ xfs_mount_t *mp; /* mount structure */
+ xfs_filblks_t nblks; /* quota/sb block count */
+ xfs_bmbt_irec_t new; /* new record to be inserted */
+ /* REFERENCED */
+ uint qfield; /* quota field to update */
+ xfs_filblks_t temp; /* for indirect length calculations */
+ xfs_filblks_t temp2; /* for indirect length calculations */
+ int state = 0;
- ASSERT(len > 0);
- ASSERT(nexts >= 0);
+ XFS_STATS_INC(xs_del_exlist);
- if (!(ifp->if_flags & XFS_IFEXTENTS) &&
- (error = xfs_iread_extents(tp, ip, whichfork)))
- return error;
- nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
- if (nextents == 0) {
- *done = 1;
- return 0;
- }
- XFS_STATS_INC(xs_blk_unmap);
- isrt = (whichfork == XFS_DATA_FORK) && XFS_IS_REALTIME_INODE(ip);
- start = bno;
- bno = start + len - 1;
- ep = xfs_bmap_search_extents(ip, bno, whichfork, &eof, &lastx, &got,
- &prev);
+ if (whichfork == XFS_ATTR_FORK)
+ state |= BMAP_ATTRFORK;
+ mp = ip->i_mount;
+ ifp = XFS_IFORK_PTR(ip, whichfork);
+ ASSERT((*idx >= 0) && (*idx < ifp->if_bytes /
+ (uint)sizeof(xfs_bmbt_rec_t)));
+ ASSERT(del->br_blockcount > 0);
+ ep = xfs_iext_get_ext(ifp, *idx);
+ xfs_bmbt_get_all(ep, &got);
+ ASSERT(got.br_startoff <= del->br_startoff);
+ del_endoff = del->br_startoff + del->br_blockcount;
+ got_endoff = got.br_startoff + got.br_blockcount;
+ ASSERT(got_endoff >= del_endoff);
+ delay = isnullstartblock(got.br_startblock);
+ ASSERT(isnullstartblock(del->br_startblock) == delay);
+ flags = 0;
+ qfield = 0;
+ error = 0;
/*
- * Check to see if the given block number is past the end of the
- * file, back up to the last block if so...
+ * If deleting a real allocation, must free up the disk space.
*/
- if (eof) {
- ep = xfs_iext_get_ext(ifp, --lastx);
- xfs_bmbt_get_all(ep, &got);
- bno = got.br_startoff + got.br_blockcount - 1;
- }
- logflags = 0;
- if (ifp->if_flags & XFS_IFBROOT) {
- ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE);
- cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
- cur->bc_private.b.firstblock = *firstblock;
- cur->bc_private.b.flist = flist;
- cur->bc_private.b.flags = 0;
- } else
- cur = NULL;
-
- if (isrt) {
+ if (!delay) {
+ flags = XFS_ILOG_CORE;
/*
- * Synchronize by locking the bitmap inode.
+ * Realtime allocation. Free it and record di_nblocks update.
*/
- xfs_ilock(mp->m_rbmip, XFS_ILOCK_EXCL);
- xfs_trans_ijoin(tp, mp->m_rbmip, XFS_ILOCK_EXCL);
- }
+ if (whichfork == XFS_DATA_FORK && XFS_IS_REALTIME_INODE(ip)) {
+ xfs_fsblock_t bno;
+ xfs_filblks_t len;
- extno = 0;
- while (bno != (xfs_fileoff_t)-1 && bno >= start && lastx >= 0 &&
- (nexts == 0 || extno < nexts)) {
- /*
- * Is the found extent after a hole in which bno lives?
- * Just back up to the previous extent, if so.
- */
- if (got.br_startoff > bno) {
- if (--lastx < 0)
- break;
- ep = xfs_iext_get_ext(ifp, lastx);
- xfs_bmbt_get_all(ep, &got);
+ ASSERT(do_mod(del->br_blockcount,
+ mp->m_sb.sb_rextsize) == 0);
+ ASSERT(do_mod(del->br_startblock,
+ mp->m_sb.sb_rextsize) == 0);
+ bno = del->br_startblock;
+ len = del->br_blockcount;
+ do_div(bno, mp->m_sb.sb_rextsize);
+ do_div(len, mp->m_sb.sb_rextsize);
+ error = xfs_rtfree_extent(tp, bno, (xfs_extlen_t)len);
+ if (error)
+ goto done;
+ do_fx = 0;
+ nblks = len * mp->m_sb.sb_rextsize;
+ qfield = XFS_TRANS_DQ_RTBCOUNT;
}
/*
- * Is the last block of this extent before the range
- * we're supposed to delete? If so, we're done.
+ * Ordinary allocation.
*/
- bno = XFS_FILEOFF_MIN(bno,
- got.br_startoff + got.br_blockcount - 1);
- if (bno < start)
- break;
+ else {
+ do_fx = 1;
+ nblks = del->br_blockcount;
+ qfield = XFS_TRANS_DQ_BCOUNT;
+ }
/*
- * Then deal with the (possibly delayed) allocated space
- * we found.
+ * Set up del_endblock and cur for later.
*/
- ASSERT(ep != NULL);
- del = got;
- wasdel = isnullstartblock(del.br_startblock);
- if (got.br_startoff < start) {
- del.br_startoff = start;
- del.br_blockcount -= start - got.br_startoff;
- if (!wasdel)
- del.br_startblock += start - got.br_startoff;
- }
- if (del.br_startoff + del.br_blockcount > bno + 1)
- del.br_blockcount = bno + 1 - del.br_startoff;
- sum = del.br_startblock + del.br_blockcount;
- if (isrt &&
- (mod = do_mod(sum, mp->m_sb.sb_rextsize))) {
- /*
- * Realtime extent not lined up at the end.
- * The extent could have been split into written
- * and unwritten pieces, or we could just be
- * unmapping part of it. But we can't really
- * get rid of part of a realtime extent.
- */
- if (del.br_state == XFS_EXT_UNWRITTEN ||
- !xfs_sb_version_hasextflgbit(&mp->m_sb)) {
- /*
- * This piece is unwritten, or we're not
- * using unwritten extents. Skip over it.
- */
- ASSERT(bno >= mod);
- bno -= mod > del.br_blockcount ?
- del.br_blockcount : mod;
- if (bno < got.br_startoff) {
- if (--lastx >= 0)
- xfs_bmbt_get_all(xfs_iext_get_ext(
- ifp, lastx), &got);
- }
- continue;
- }
- /*
- * It's written, turn it unwritten.
- * This is better than zeroing it.
- */
- ASSERT(del.br_state == XFS_EXT_NORM);
- ASSERT(xfs_trans_get_block_res(tp) > 0);
- /*
- * If this spans a realtime extent boundary,
- * chop it back to the start of the one we end at.
- */
- if (del.br_blockcount > mod) {
- del.br_startoff += del.br_blockcount - mod;
- del.br_startblock += del.br_blockcount - mod;
- del.br_blockcount = mod;
- }
- del.br_state = XFS_EXT_UNWRITTEN;
- error = xfs_bmap_add_extent_unwritten_real(tp, ip,
- &lastx, &cur, &del, firstblock, flist,
- &logflags);
- if (error)
- goto error0;
- goto nodelete;
+ del_endblock = del->br_startblock + del->br_blockcount;
+ if (cur) {
+ if ((error = xfs_bmbt_lookup_eq(cur, got.br_startoff,
+ got.br_startblock, got.br_blockcount,
+ &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
}
- if (isrt && (mod = do_mod(del.br_startblock, mp->m_sb.sb_rextsize))) {
- /*
- * Realtime extent is lined up at the end but not
- * at the front. We'll get rid of full extents if
- * we can.
- */
- mod = mp->m_sb.sb_rextsize - mod;
- if (del.br_blockcount > mod) {
- del.br_blockcount -= mod;
- del.br_startoff += mod;
- del.br_startblock += mod;
- } else if ((del.br_startoff == start &&
- (del.br_state == XFS_EXT_UNWRITTEN ||
- xfs_trans_get_block_res(tp) == 0)) ||
- !xfs_sb_version_hasextflgbit(&mp->m_sb)) {
- /*
- * Can't make it unwritten. There isn't
- * a full extent here so just skip it.
- */
- ASSERT(bno >= del.br_blockcount);
- bno -= del.br_blockcount;
- if (got.br_startoff > bno) {
- if (--lastx >= 0) {
- ep = xfs_iext_get_ext(ifp,
- lastx);
- xfs_bmbt_get_all(ep, &got);
- }
- }
- continue;
- } else if (del.br_state == XFS_EXT_UNWRITTEN) {
- /*
- * This one is already unwritten.
- * It must have a written left neighbor.
- * Unwrite the killed part of that one and
- * try again.
- */
- ASSERT(lastx > 0);
- xfs_bmbt_get_all(xfs_iext_get_ext(ifp,
- lastx - 1), &prev);
- ASSERT(prev.br_state == XFS_EXT_NORM);
- ASSERT(!isnullstartblock(prev.br_startblock));
- ASSERT(del.br_startblock ==
- prev.br_startblock + prev.br_blockcount);
- if (prev.br_startoff < start) {
- mod = start - prev.br_startoff;
- prev.br_blockcount -= mod;
- prev.br_startblock += mod;
- prev.br_startoff = start;
- }
- prev.br_state = XFS_EXT_UNWRITTEN;
- lastx--;
- error = xfs_bmap_add_extent_unwritten_real(tp,
- ip, &lastx, &cur, &prev,
- firstblock, flist, &logflags);
- if (error)
- goto error0;
- goto nodelete;
- } else {
- ASSERT(del.br_state == XFS_EXT_NORM);
- del.br_state = XFS_EXT_UNWRITTEN;
- error = xfs_bmap_add_extent_unwritten_real(tp,
- ip, &lastx, &cur, &del,
- firstblock, flist, &logflags);
- if (error)
- goto error0;
- goto nodelete;
- }
+ da_old = da_new = 0;
+ } else {
+ da_old = startblockval(got.br_startblock);
+ da_new = 0;
+ nblks = 0;
+ do_fx = 0;
+ }
+ /*
+ * Set flag value to use in switch statement.
+ * Left-contig is 2, right-contig is 1.
+ */
+ switch (((got.br_startoff == del->br_startoff) << 1) |
+ (got_endoff == del_endoff)) {
+ case 3:
+ /*
+ * Matches the whole extent. Delete the entry.
+ */
+ xfs_iext_remove(ip, *idx, 1,
+ whichfork == XFS_ATTR_FORK ? BMAP_ATTRFORK : 0);
+ --*idx;
+ if (delay)
+ break;
+
+ XFS_IFORK_NEXT_SET(ip, whichfork,
+ XFS_IFORK_NEXTENTS(ip, whichfork) - 1);
+ flags |= XFS_ILOG_CORE;
+ if (!cur) {
+ flags |= xfs_ilog_fext(whichfork);
+ break;
}
- if (wasdel) {
- ASSERT(startblockval(del.br_startblock) > 0);
- /* Update realtime/data freespace, unreserve quota */
- if (isrt) {
- xfs_filblks_t rtexts;
+ if ((error = xfs_btree_delete(cur, &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ break;
- rtexts = XFS_FSB_TO_B(mp, del.br_blockcount);
- do_div(rtexts, mp->m_sb.sb_rextsize);
- xfs_mod_incore_sb(mp, XFS_SBS_FREXTENTS,
- (int64_t)rtexts, 0);
- (void)xfs_trans_reserve_quota_nblks(NULL,
- ip, -((long)del.br_blockcount), 0,
- XFS_QMOPT_RES_RTBLKS);
- } else {
- xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS,
- (int64_t)del.br_blockcount, 0);
- (void)xfs_trans_reserve_quota_nblks(NULL,
- ip, -((long)del.br_blockcount), 0,
- XFS_QMOPT_RES_REGBLKS);
- }
- ip->i_delayed_blks -= del.br_blockcount;
- if (cur)
- cur->bc_private.b.flags |=
- XFS_BTCUR_BPRV_WASDEL;
- } else if (cur)
- cur->bc_private.b.flags &= ~XFS_BTCUR_BPRV_WASDEL;
+ case 2:
/*
- * If it's the case where the directory code is running
- * with no block reservation, and the deleted block is in
- * the middle of its extent, and the resulting insert
- * of an extent would cause transformation to btree format,
- * then reject it. The calling code will then swap
- * blocks around instead.
- * We have to do this now, rather than waiting for the
- * conversion to btree format, since the transaction
- * will be dirty.
+ * Deleting the first part of the extent.
*/
- if (!wasdel && xfs_trans_get_block_res(tp) == 0 &&
- XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS &&
- XFS_IFORK_NEXTENTS(ip, whichfork) >= /* Note the >= */
- XFS_IFORK_MAXEXT(ip, whichfork) &&
- del.br_startoff > got.br_startoff &&
- del.br_startoff + del.br_blockcount <
- got.br_startoff + got.br_blockcount) {
- error = XFS_ERROR(ENOSPC);
- goto error0;
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ xfs_bmbt_set_startoff(ep, del_endoff);
+ temp = got.br_blockcount - del->br_blockcount;
+ xfs_bmbt_set_blockcount(ep, temp);
+ if (delay) {
+ temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
+ da_old);
+ xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
+ da_new = temp;
+ break;
}
- error = xfs_bmap_del_extent(ip, tp, &lastx, flist, cur, &del,
- &tmp_logflags, whichfork);
- logflags |= tmp_logflags;
- if (error)
- goto error0;
- bno = del.br_startoff - 1;
-nodelete:
+ xfs_bmbt_set_startblock(ep, del_endblock);
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
+ if (!cur) {
+ flags |= xfs_ilog_fext(whichfork);
+ break;
+ }
+ if ((error = xfs_bmbt_update(cur, del_endoff, del_endblock,
+ got.br_blockcount - del->br_blockcount,
+ got.br_state)))
+ goto done;
+ break;
+
+ case 1:
/*
- * If not done go on to the next (previous) record.
+ * Deleting the last part of the extent.
*/
- if (bno != (xfs_fileoff_t)-1 && bno >= start) {
- if (lastx >= 0) {
- ep = xfs_iext_get_ext(ifp, lastx);
- if (xfs_bmbt_get_startoff(ep) > bno) {
- if (--lastx >= 0)
- ep = xfs_iext_get_ext(ifp,
- lastx);
+ temp = got.br_blockcount - del->br_blockcount;
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(ep, temp);
+ if (delay) {
+ temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
+ da_old);
+ xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
+ da_new = temp;
+ break;
+ }
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
+ if (!cur) {
+ flags |= xfs_ilog_fext(whichfork);
+ break;
+ }
+ if ((error = xfs_bmbt_update(cur, got.br_startoff,
+ got.br_startblock,
+ got.br_blockcount - del->br_blockcount,
+ got.br_state)))
+ goto done;
+ break;
+
+ case 0:
+ /*
+ * Deleting the middle of the extent.
+ */
+ temp = del->br_startoff - got.br_startoff;
+ trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+ xfs_bmbt_set_blockcount(ep, temp);
+ new.br_startoff = del_endoff;
+ temp2 = got_endoff - del_endoff;
+ new.br_blockcount = temp2;
+ new.br_state = got.br_state;
+ if (!delay) {
+ new.br_startblock = del_endblock;
+ flags |= XFS_ILOG_CORE;
+ if (cur) {
+ if ((error = xfs_bmbt_update(cur,
+ got.br_startoff,
+ got.br_startblock, temp,
+ got.br_state)))
+ goto done;
+ if ((error = xfs_btree_increment(cur, 0, &i)))
+ goto done;
+ cur->bc_rec.b = new;
+ error = xfs_btree_insert(cur, &i);
+ if (error && error != ENOSPC)
+ goto done;
+ /*
+ * If get no-space back from btree insert,
+ * it tried a split, and we have a zero
+ * block reservation.
+ * Fix up our state and return the error.
+ */
+ if (error == ENOSPC) {
+ /*
+ * Reset the cursor, don't trust
+ * it after any insert operation.
+ */
+ if ((error = xfs_bmbt_lookup_eq(cur,
+ got.br_startoff,
+ got.br_startblock,
+ temp, &i)))
+ goto done;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ /*
+ * Update the btree record back
+ * to the original value.
+ */
+ if ((error = xfs_bmbt_update(cur,
+ got.br_startoff,
+ got.br_startblock,
+ got.br_blockcount,
+ got.br_state)))
+ goto done;
+ /*
+ * Reset the extent record back
+ * to the original value.
+ */
+ xfs_bmbt_set_blockcount(ep,
+ got.br_blockcount);
+ flags = 0;
+ error = XFS_ERROR(ENOSPC);
+ goto done;
+ }
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ } else
+ flags |= xfs_ilog_fext(whichfork);
+ XFS_IFORK_NEXT_SET(ip, whichfork,
+ XFS_IFORK_NEXTENTS(ip, whichfork) + 1);
+ } else {
+ ASSERT(whichfork == XFS_DATA_FORK);
+ temp = xfs_bmap_worst_indlen(ip, temp);
+ xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
+ temp2 = xfs_bmap_worst_indlen(ip, temp2);
+ new.br_startblock = nullstartblock((int)temp2);
+ da_new = temp + temp2;
+ while (da_new > da_old) {
+ if (temp) {
+ temp--;
+ da_new--;
+ xfs_bmbt_set_startblock(ep,
+ nullstartblock((int)temp));
+ }
+ if (da_new == da_old)
+ break;
+ if (temp2) {
+ temp2--;
+ da_new--;
+ new.br_startblock =
+ nullstartblock((int)temp2);
}
- xfs_bmbt_get_all(ep, &got);
}
- extno++;
}
- }
- *done = bno == (xfs_fileoff_t)-1 || bno < start || lastx < 0;
-
- /*
- * Convert to a btree if necessary.
- */
- if (xfs_bmap_needs_btree(ip, whichfork)) {
- ASSERT(cur == NULL);
- error = xfs_bmap_extents_to_btree(tp, ip, firstblock, flist,
- &cur, 0, &tmp_logflags, whichfork);
- logflags |= tmp_logflags;
- if (error)
- goto error0;
+ trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
+ xfs_iext_insert(ip, *idx + 1, 1, &new, state);
+ ++*idx;
+ break;
}
/*
- * transform from btree to extents, give it cur
+ * If we need to, add to list of extents to delete.
*/
- else if (xfs_bmap_wants_extents(ip, whichfork)) {
- ASSERT(cur != NULL);
- error = xfs_bmap_btree_to_extents(tp, ip, cur, &tmp_logflags,
- whichfork);
- logflags |= tmp_logflags;
- if (error)
- goto error0;
- }
+ if (do_fx)
+ xfs_bmap_add_free(del->br_startblock, del->br_blockcount, flist,
+ mp);
/*
- * transform from extents to local?
+ * Adjust inode # blocks in the file.
*/
- error = 0;
-error0:
+ if (nblks)
+ ip->i_d.di_nblocks -= nblks;
/*
- * Log everything. Do this after conversion, there's no point in
- * logging the extent records if we've converted to btree format.
+ * Adjust quota data.
*/
- if ((logflags & xfs_ilog_fext(whichfork)) &&
- XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS)
- logflags &= ~xfs_ilog_fext(whichfork);
- else if ((logflags & xfs_ilog_fbroot(whichfork)) &&
- XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)
- logflags &= ~xfs_ilog_fbroot(whichfork);
+ if (qfield)
+ xfs_trans_mod_dquot_byino(tp, ip, qfield, (long)-nblks);
+
/*
- * Log inode even in the error case, if the transaction
- * is dirty we'll need to shut down the filesystem.
+ * Account for change in delayed indirect blocks.
+ * Nothing to do for disk quota accounting here.
*/
- if (logflags)
- xfs_trans_log_inode(tp, ip, logflags);
- if (cur) {
- if (!error) {
- *firstblock = cur->bc_private.b.firstblock;
- cur->bc_private.b.allocated = 0;
- }
- xfs_btree_del_cursor(cur,
- error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR);
+ ASSERT(da_old >= da_new);
+ if (da_old > da_new) {
+ xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS,
+ (int64_t)(da_old - da_new), 0);
}
+done:
+ *logflagsp = flags;
return error;
}
-#ifdef DEBUG
-STATIC struct xfs_buf *
-xfs_bmap_get_bp(
- struct xfs_btree_cur *cur,
- xfs_fsblock_t bno)
-{
- struct xfs_log_item_desc *lidp;
- int i;
-
- if (!cur)
- return NULL;
-
- for (i = 0; i < XFS_BTREE_MAXLEVELS; i++) {
- if (!cur->bc_bufs[i])
- break;
- if (XFS_BUF_ADDR(cur->bc_bufs[i]) == bno)
- return cur->bc_bufs[i];
- }
-
- /* Chase down all the log items to see if the bp is there */
- list_for_each_entry(lidp, &cur->bc_tp->t_items, lid_trans) {
- struct xfs_buf_log_item *bip;
- bip = (struct xfs_buf_log_item *)lidp->lid_item;
- if (bip->bli_item.li_type == XFS_LI_BUF &&
- XFS_BUF_ADDR(bip->bli_buf) == bno)
- return bip->bli_buf;
- }
-
- return NULL;
-}
-
-STATIC void
-xfs_check_block(
- struct xfs_btree_block *block,
- xfs_mount_t *mp,
- int root,
- short sz)
-{
- int i, j, dmxr;
- __be64 *pp, *thispa; /* pointer to block address */
- xfs_bmbt_key_t *prevp, *keyp;
-
- ASSERT(be16_to_cpu(block->bb_level) > 0);
-
- prevp = NULL;
- for( i = 1; i <= xfs_btree_get_numrecs(block); i++) {
- dmxr = mp->m_bmap_dmxr[0];
- keyp = XFS_BMBT_KEY_ADDR(mp, block, i);
-
- if (prevp) {
- ASSERT(be64_to_cpu(prevp->br_startoff) <
- be64_to_cpu(keyp->br_startoff));
- }
- prevp = keyp;
-
- /*
- * Compare the block numbers to see if there are dups.
- */
- if (root)
- pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, i, sz);
- else
- pp = XFS_BMBT_PTR_ADDR(mp, block, i, dmxr);
-
- for (j = i+1; j <= be16_to_cpu(block->bb_numrecs); j++) {
- if (root)
- thispa = XFS_BMAP_BROOT_PTR_ADDR(mp, block, j, sz);
- else
- thispa = XFS_BMBT_PTR_ADDR(mp, block, j, dmxr);
- if (*thispa == *pp) {
- xfs_warn(mp, "%s: thispa(%d) == pp(%d) %Ld",
- __func__, j, i,
- (unsigned long long)be64_to_cpu(*thispa));
- panic("%s: ptrs are equal in node\n",
- __func__);
- }
- }
- }
-}
-
/*
- * Check that the extents for the inode ip are in the right order in all
- * btree leaves.
+ * Unmap (remove) blocks from a file.
+ * If nexts is nonzero then the number of extents to remove is limited to
+ * that value. If not all extents in the block range can be removed then
+ * *done is set.
*/
-
-STATIC void
-xfs_bmap_check_leaf_extents(
- xfs_btree_cur_t *cur, /* btree cursor or null */
- xfs_inode_t *ip, /* incore inode pointer */
- int whichfork) /* data or attr fork */
+int /* error */
+xfs_bunmapi(
+ xfs_trans_t *tp, /* transaction pointer */
+ struct xfs_inode *ip, /* incore inode */
+ xfs_fileoff_t bno, /* starting offset to unmap */
+ xfs_filblks_t len, /* length to unmap in file */
+ int flags, /* misc flags */
+ xfs_extnum_t nexts, /* number of extents max */
+ xfs_fsblock_t *firstblock, /* first allocated block
+ controls a.g. for allocs */
+ xfs_bmap_free_t *flist, /* i/o: list extents to free */
+ int *done) /* set if not done yet */
{
- struct xfs_btree_block *block; /* current btree block */
- xfs_fsblock_t bno; /* block # of "block" */
- xfs_buf_t *bp; /* buffer for "block" */
- int error; /* error return value */
- xfs_extnum_t i=0, j; /* index into the extents list */
- xfs_ifork_t *ifp; /* fork structure */
- int level; /* btree level, for checking */
- xfs_mount_t *mp; /* file system mount structure */
- __be64 *pp; /* pointer to block address */
- xfs_bmbt_rec_t *ep; /* pointer to current extent */
- xfs_bmbt_rec_t last = {0, 0}; /* last extent in prev block */
- xfs_bmbt_rec_t *nextp; /* pointer to next extent */
- int bp_release = 0;
+ xfs_btree_cur_t *cur; /* bmap btree cursor */
+ xfs_bmbt_irec_t del; /* extent being deleted */
+ int eof; /* is deleting at eof */
+ xfs_bmbt_rec_host_t *ep; /* extent record pointer */
+ int error; /* error return value */
+ xfs_extnum_t extno; /* extent number in list */
+ xfs_bmbt_irec_t got; /* current extent record */
+ xfs_ifork_t *ifp; /* inode fork pointer */
+ int isrt; /* freeing in rt area */
+ xfs_extnum_t lastx; /* last extent index used */
+ int logflags; /* transaction logging flags */
+ xfs_extlen_t mod; /* rt extent offset */
+ xfs_mount_t *mp; /* mount structure */
+ xfs_extnum_t nextents; /* number of file extents */
+ xfs_bmbt_irec_t prev; /* previous extent record */
+ xfs_fileoff_t start; /* first file offset deleted */
+ int tmp_logflags; /* partial logging flags */
+ int wasdel; /* was a delayed alloc extent */
+ int whichfork; /* data or attribute fork */
+ xfs_fsblock_t sum;
- if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE) {
- return;
- }
+ trace_xfs_bunmap(ip, bno, len, flags, _RET_IP_);
- bno = NULLFSBLOCK;
- mp = ip->i_mount;
+ whichfork = (flags & XFS_BMAPI_ATTRFORK) ?
+ XFS_ATTR_FORK : XFS_DATA_FORK;
ifp = XFS_IFORK_PTR(ip, whichfork);
- block = ifp->if_broot;
- /*
- * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out.
- */
- level = be16_to_cpu(block->bb_level);
- ASSERT(level > 0);
- xfs_check_block(block, mp, 1, ifp->if_broot_bytes);
- pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
- bno = be64_to_cpu(*pp);
-
- ASSERT(bno != NULLDFSBNO);
- ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount);
- ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks);
-
- /*
- * Go down the tree until leaf level is reached, following the first
- * pointer (leftmost) at each level.
- */
- while (level-- > 0) {
- /* See if buf is in cur first */
- bp_release = 0;
- bp = xfs_bmap_get_bp(cur, XFS_FSB_TO_DADDR(mp, bno));
- if (!bp) {
- bp_release = 1;
- error = xfs_btree_read_bufl(mp, NULL, bno, 0, &bp,
- XFS_BMAP_BTREE_REF,
- &xfs_bmbt_buf_ops);
- if (error)
- goto error_norelse;
- }
- block = XFS_BUF_TO_BLOCK(bp);
- XFS_WANT_CORRUPTED_GOTO(
- xfs_bmap_sanity_check(mp, bp, level),
- error0);
- if (level == 0)
- break;
+ if (unlikely(
+ XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
+ XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)) {
+ XFS_ERROR_REPORT("xfs_bunmapi", XFS_ERRLEVEL_LOW,
+ ip->i_mount);
+ return XFS_ERROR(EFSCORRUPTED);
+ }
+ mp = ip->i_mount;
+ if (XFS_FORCED_SHUTDOWN(mp))
+ return XFS_ERROR(EIO);
- /*
- * Check this block for basic sanity (increasing keys and
- * no duplicate blocks).
- */
+ ASSERT(len > 0);
+ ASSERT(nexts >= 0);
- xfs_check_block(block, mp, 0, 0);
- pp = XFS_BMBT_PTR_ADDR(mp, block, 1, mp->m_bmap_dmxr[1]);
- bno = be64_to_cpu(*pp);
- XFS_WANT_CORRUPTED_GOTO(XFS_FSB_SANITY_CHECK(mp, bno), error0);
- if (bp_release) {
- bp_release = 0;
- xfs_trans_brelse(NULL, bp);
- }
+ if (!(ifp->if_flags & XFS_IFEXTENTS) &&
+ (error = xfs_iread_extents(tp, ip, whichfork)))
+ return error;
+ nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
+ if (nextents == 0) {
+ *done = 1;
+ return 0;
}
+ XFS_STATS_INC(xs_blk_unmap);
+ isrt = (whichfork == XFS_DATA_FORK) && XFS_IS_REALTIME_INODE(ip);
+ start = bno;
+ bno = start + len - 1;
+ ep = xfs_bmap_search_extents(ip, bno, whichfork, &eof, &lastx, &got,
+ &prev);
/*
- * Here with bp and block set to the leftmost leaf node in the tree.
- */
- i = 0;
-
- /*
- * Loop over all leaf nodes checking that all extents are in the right order.
+ * Check to see if the given block number is past the end of the
+ * file, back up to the last block if so...
*/
- for (;;) {
- xfs_fsblock_t nextbno;
- xfs_extnum_t num_recs;
-
-
- num_recs = xfs_btree_get_numrecs(block);
+ if (eof) {
+ ep = xfs_iext_get_ext(ifp, --lastx);
+ xfs_bmbt_get_all(ep, &got);
+ bno = got.br_startoff + got.br_blockcount - 1;
+ }
+ logflags = 0;
+ if (ifp->if_flags & XFS_IFBROOT) {
+ ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE);
+ cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
+ cur->bc_private.b.firstblock = *firstblock;
+ cur->bc_private.b.flist = flist;
+ cur->bc_private.b.flags = 0;
+ } else
+ cur = NULL;
+ if (isrt) {
/*
- * Read-ahead the next leaf block, if any.
+ * Synchronize by locking the bitmap inode.
*/
+ xfs_ilock(mp->m_rbmip, XFS_ILOCK_EXCL);
+ xfs_trans_ijoin(tp, mp->m_rbmip, XFS_ILOCK_EXCL);
+ }
- nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
-
+ extno = 0;
+ while (bno != (xfs_fileoff_t)-1 && bno >= start && lastx >= 0 &&
+ (nexts == 0 || extno < nexts)) {
/*
- * Check all the extents to make sure they are OK.
- * If we had a previous block, the last entry should
- * conform with the first entry in this one.
+ * Is the found extent after a hole in which bno lives?
+ * Just back up to the previous extent, if so.
*/
-
- ep = XFS_BMBT_REC_ADDR(mp, block, 1);
- if (i) {
- ASSERT(xfs_bmbt_disk_get_startoff(&last) +
- xfs_bmbt_disk_get_blockcount(&last) <=
- xfs_bmbt_disk_get_startoff(ep));
+ if (got.br_startoff > bno) {
+ if (--lastx < 0)
+ break;
+ ep = xfs_iext_get_ext(ifp, lastx);
+ xfs_bmbt_get_all(ep, &got);
}
- for (j = 1; j < num_recs; j++) {
- nextp = XFS_BMBT_REC_ADDR(mp, block, j + 1);
- ASSERT(xfs_bmbt_disk_get_startoff(ep) +
- xfs_bmbt_disk_get_blockcount(ep) <=
- xfs_bmbt_disk_get_startoff(nextp));
- ep = nextp;
+ /*
+ * Is the last block of this extent before the range
+ * we're supposed to delete? If so, we're done.
+ */
+ bno = XFS_FILEOFF_MIN(bno,
+ got.br_startoff + got.br_blockcount - 1);
+ if (bno < start)
+ break;
+ /*
+ * Then deal with the (possibly delayed) allocated space
+ * we found.
+ */
+ ASSERT(ep != NULL);
+ del = got;
+ wasdel = isnullstartblock(del.br_startblock);
+ if (got.br_startoff < start) {
+ del.br_startoff = start;
+ del.br_blockcount -= start - got.br_startoff;
+ if (!wasdel)
+ del.br_startblock += start - got.br_startoff;
+ }
+ if (del.br_startoff + del.br_blockcount > bno + 1)
+ del.br_blockcount = bno + 1 - del.br_startoff;
+ sum = del.br_startblock + del.br_blockcount;
+ if (isrt &&
+ (mod = do_mod(sum, mp->m_sb.sb_rextsize))) {
+ /*
+ * Realtime extent not lined up at the end.
+ * The extent could have been split into written
+ * and unwritten pieces, or we could just be
+ * unmapping part of it. But we can't really
+ * get rid of part of a realtime extent.
+ */
+ if (del.br_state == XFS_EXT_UNWRITTEN ||
+ !xfs_sb_version_hasextflgbit(&mp->m_sb)) {
+ /*
+ * This piece is unwritten, or we're not
+ * using unwritten extents. Skip over it.
+ */
+ ASSERT(bno >= mod);
+ bno -= mod > del.br_blockcount ?
+ del.br_blockcount : mod;
+ if (bno < got.br_startoff) {
+ if (--lastx >= 0)
+ xfs_bmbt_get_all(xfs_iext_get_ext(
+ ifp, lastx), &got);
+ }
+ continue;
+ }
+ /*
+ * It's written, turn it unwritten.
+ * This is better than zeroing it.
+ */
+ ASSERT(del.br_state == XFS_EXT_NORM);
+ ASSERT(xfs_trans_get_block_res(tp) > 0);
+ /*
+ * If this spans a realtime extent boundary,
+ * chop it back to the start of the one we end at.
+ */
+ if (del.br_blockcount > mod) {
+ del.br_startoff += del.br_blockcount - mod;
+ del.br_startblock += del.br_blockcount - mod;
+ del.br_blockcount = mod;
+ }
+ del.br_state = XFS_EXT_UNWRITTEN;
+ error = xfs_bmap_add_extent_unwritten_real(tp, ip,
+ &lastx, &cur, &del, firstblock, flist,
+ &logflags);
+ if (error)
+ goto error0;
+ goto nodelete;
}
+ if (isrt && (mod = do_mod(del.br_startblock, mp->m_sb.sb_rextsize))) {
+ /*
+ * Realtime extent is lined up at the end but not
+ * at the front. We'll get rid of full extents if
+ * we can.
+ */
+ mod = mp->m_sb.sb_rextsize - mod;
+ if (del.br_blockcount > mod) {
+ del.br_blockcount -= mod;
+ del.br_startoff += mod;
+ del.br_startblock += mod;
+ } else if ((del.br_startoff == start &&
+ (del.br_state == XFS_EXT_UNWRITTEN ||
+ xfs_trans_get_block_res(tp) == 0)) ||
+ !xfs_sb_version_hasextflgbit(&mp->m_sb)) {
+ /*
+ * Can't make it unwritten. There isn't
+ * a full extent here so just skip it.
+ */
+ ASSERT(bno >= del.br_blockcount);
+ bno -= del.br_blockcount;
+ if (got.br_startoff > bno) {
+ if (--lastx >= 0) {
+ ep = xfs_iext_get_ext(ifp,
+ lastx);
+ xfs_bmbt_get_all(ep, &got);
+ }
+ }
+ continue;
+ } else if (del.br_state == XFS_EXT_UNWRITTEN) {
+ /*
+ * This one is already unwritten.
+ * It must have a written left neighbor.
+ * Unwrite the killed part of that one and
+ * try again.
+ */
+ ASSERT(lastx > 0);
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp,
+ lastx - 1), &prev);
+ ASSERT(prev.br_state == XFS_EXT_NORM);
+ ASSERT(!isnullstartblock(prev.br_startblock));
+ ASSERT(del.br_startblock ==
+ prev.br_startblock + prev.br_blockcount);
+ if (prev.br_startoff < start) {
+ mod = start - prev.br_startoff;
+ prev.br_blockcount -= mod;
+ prev.br_startblock += mod;
+ prev.br_startoff = start;
+ }
+ prev.br_state = XFS_EXT_UNWRITTEN;
+ lastx--;
+ error = xfs_bmap_add_extent_unwritten_real(tp,
+ ip, &lastx, &cur, &prev,
+ firstblock, flist, &logflags);
+ if (error)
+ goto error0;
+ goto nodelete;
+ } else {
+ ASSERT(del.br_state == XFS_EXT_NORM);
+ del.br_state = XFS_EXT_UNWRITTEN;
+ error = xfs_bmap_add_extent_unwritten_real(tp,
+ ip, &lastx, &cur, &del,
+ firstblock, flist, &logflags);
+ if (error)
+ goto error0;
+ goto nodelete;
+ }
+ }
+ if (wasdel) {
+ ASSERT(startblockval(del.br_startblock) > 0);
+ /* Update realtime/data freespace, unreserve quota */
+ if (isrt) {
+ xfs_filblks_t rtexts;
- last = *ep;
- i += num_recs;
- if (bp_release) {
- bp_release = 0;
- xfs_trans_brelse(NULL, bp);
+ rtexts = XFS_FSB_TO_B(mp, del.br_blockcount);
+ do_div(rtexts, mp->m_sb.sb_rextsize);
+ xfs_mod_incore_sb(mp, XFS_SBS_FREXTENTS,
+ (int64_t)rtexts, 0);
+ (void)xfs_trans_reserve_quota_nblks(NULL,
+ ip, -((long)del.br_blockcount), 0,
+ XFS_QMOPT_RES_RTBLKS);
+ } else {
+ xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS,
+ (int64_t)del.br_blockcount, 0);
+ (void)xfs_trans_reserve_quota_nblks(NULL,
+ ip, -((long)del.br_blockcount), 0,
+ XFS_QMOPT_RES_REGBLKS);
+ }
+ ip->i_delayed_blks -= del.br_blockcount;
+ if (cur)
+ cur->bc_private.b.flags |=
+ XFS_BTCUR_BPRV_WASDEL;
+ } else if (cur)
+ cur->bc_private.b.flags &= ~XFS_BTCUR_BPRV_WASDEL;
+ /*
+ * If it's the case where the directory code is running
+ * with no block reservation, and the deleted block is in
+ * the middle of its extent, and the resulting insert
+ * of an extent would cause transformation to btree format,
+ * then reject it. The calling code will then swap
+ * blocks around instead.
+ * We have to do this now, rather than waiting for the
+ * conversion to btree format, since the transaction
+ * will be dirty.
+ */
+ if (!wasdel && xfs_trans_get_block_res(tp) == 0 &&
+ XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS &&
+ XFS_IFORK_NEXTENTS(ip, whichfork) >= /* Note the >= */
+ XFS_IFORK_MAXEXT(ip, whichfork) &&
+ del.br_startoff > got.br_startoff &&
+ del.br_startoff + del.br_blockcount <
+ got.br_startoff + got.br_blockcount) {
+ error = XFS_ERROR(ENOSPC);
+ goto error0;
}
- bno = nextbno;
+ error = xfs_bmap_del_extent(ip, tp, &lastx, flist, cur, &del,
+ &tmp_logflags, whichfork);
+ logflags |= tmp_logflags;
+ if (error)
+ goto error0;
+ bno = del.br_startoff - 1;
+nodelete:
/*
- * If we've reached the end, stop.
+ * If not done go on to the next (previous) record.
*/
- if (bno == NULLFSBLOCK)
- break;
-
- bp_release = 0;
- bp = xfs_bmap_get_bp(cur, XFS_FSB_TO_DADDR(mp, bno));
- if (!bp) {
- bp_release = 1;
- error = xfs_btree_read_bufl(mp, NULL, bno, 0, &bp,
- XFS_BMAP_BTREE_REF,
- &xfs_bmbt_buf_ops);
- if (error)
- goto error_norelse;
+ if (bno != (xfs_fileoff_t)-1 && bno >= start) {
+ if (lastx >= 0) {
+ ep = xfs_iext_get_ext(ifp, lastx);
+ if (xfs_bmbt_get_startoff(ep) > bno) {
+ if (--lastx >= 0)
+ ep = xfs_iext_get_ext(ifp,
+ lastx);
+ }
+ xfs_bmbt_get_all(ep, &got);
+ }
+ extno++;
}
- block = XFS_BUF_TO_BLOCK(bp);
}
- if (bp_release) {
- bp_release = 0;
- xfs_trans_brelse(NULL, bp);
- }
- return;
-
-error0:
- xfs_warn(mp, "%s: at error0", __func__);
- if (bp_release)
- xfs_trans_brelse(NULL, bp);
-error_norelse:
- xfs_warn(mp, "%s: BAD after btree leaves for %d extents",
- __func__, i);
- panic("%s: CORRUPTED BTREE OR SOMETHING", __func__);
- return;
-}
-#endif
-
-/*
- * Count fsblocks of the given fork.
- */
-int /* error */
-xfs_bmap_count_blocks(
- xfs_trans_t *tp, /* transaction pointer */
- xfs_inode_t *ip, /* incore inode */
- int whichfork, /* data or attr fork */
- int *count) /* out: count of blocks */
-{
- struct xfs_btree_block *block; /* current btree block */
- xfs_fsblock_t bno; /* block # of "block" */
- xfs_ifork_t *ifp; /* fork structure */
- int level; /* btree level, for checking */
- xfs_mount_t *mp; /* file system mount structure */
- __be64 *pp; /* pointer to block address */
+ *done = bno == (xfs_fileoff_t)-1 || bno < start || lastx < 0;
- bno = NULLFSBLOCK;
- mp = ip->i_mount;
- ifp = XFS_IFORK_PTR(ip, whichfork);
- if ( XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS ) {
- xfs_bmap_count_leaves(ifp, 0,
- ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t),
- count);
- return 0;
+ /*
+ * Convert to a btree if necessary.
+ */
+ if (xfs_bmap_needs_btree(ip, whichfork)) {
+ ASSERT(cur == NULL);
+ error = xfs_bmap_extents_to_btree(tp, ip, firstblock, flist,
+ &cur, 0, &tmp_logflags, whichfork);
+ logflags |= tmp_logflags;
+ if (error)
+ goto error0;
}
-
/*
- * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out.
+ * transform from btree to extents, give it cur
*/
- block = ifp->if_broot;
- level = be16_to_cpu(block->bb_level);
- ASSERT(level > 0);
- pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
- bno = be64_to_cpu(*pp);
- ASSERT(bno != NULLDFSBNO);
- ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount);
- ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks);
-
- if (unlikely(xfs_bmap_count_tree(mp, tp, ifp, bno, level, count) < 0)) {
- XFS_ERROR_REPORT("xfs_bmap_count_blocks(2)", XFS_ERRLEVEL_LOW,
- mp);
- return XFS_ERROR(EFSCORRUPTED);
+ else if (xfs_bmap_wants_extents(ip, whichfork)) {
+ ASSERT(cur != NULL);
+ error = xfs_bmap_btree_to_extents(tp, ip, cur, &tmp_logflags,
+ whichfork);
+ logflags |= tmp_logflags;
+ if (error)
+ goto error0;
}
-
- return 0;
-}
-
-/*
- * Recursively walks each level of a btree
- * to count total fsblocks is use.
- */
-STATIC int /* error */
-xfs_bmap_count_tree(
- xfs_mount_t *mp, /* file system mount point */
- xfs_trans_t *tp, /* transaction pointer */
- xfs_ifork_t *ifp, /* inode fork pointer */
- xfs_fsblock_t blockno, /* file system block number */
- int levelin, /* level in btree */
- int *count) /* Count of blocks */
-{
- int error;
- xfs_buf_t *bp, *nbp;
- int level = levelin;
- __be64 *pp;
- xfs_fsblock_t bno = blockno;
- xfs_fsblock_t nextbno;
- struct xfs_btree_block *block, *nextblock;
- int numrecs;
-
- error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp, XFS_BMAP_BTREE_REF,
- &xfs_bmbt_buf_ops);
- if (error)
- return error;
- *count += 1;
- block = XFS_BUF_TO_BLOCK(bp);
-
- if (--level) {
- /* Not at node above leaves, count this level of nodes */
- nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
- while (nextbno != NULLFSBLOCK) {
- error = xfs_btree_read_bufl(mp, tp, nextbno, 0, &nbp,
- XFS_BMAP_BTREE_REF,
- &xfs_bmbt_buf_ops);
- if (error)
- return error;
- *count += 1;
- nextblock = XFS_BUF_TO_BLOCK(nbp);
- nextbno = be64_to_cpu(nextblock->bb_u.l.bb_rightsib);
- xfs_trans_brelse(tp, nbp);
- }
-
- /* Dive to the next level */
- pp = XFS_BMBT_PTR_ADDR(mp, block, 1, mp->m_bmap_dmxr[1]);
- bno = be64_to_cpu(*pp);
- if (unlikely((error =
- xfs_bmap_count_tree(mp, tp, ifp, bno, level, count)) < 0)) {
- xfs_trans_brelse(tp, bp);
- XFS_ERROR_REPORT("xfs_bmap_count_tree(1)",
- XFS_ERRLEVEL_LOW, mp);
- return XFS_ERROR(EFSCORRUPTED);
- }
- xfs_trans_brelse(tp, bp);
- } else {
- /* count all level 1 nodes and their leaves */
- for (;;) {
- nextbno = be64_to_cpu(block->bb_u.l.bb_rightsib);
- numrecs = be16_to_cpu(block->bb_numrecs);
- xfs_bmap_disk_count_leaves(mp, block, numrecs, count);
- xfs_trans_brelse(tp, bp);
- if (nextbno == NULLFSBLOCK)
- break;
- bno = nextbno;
- error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp,
- XFS_BMAP_BTREE_REF,
- &xfs_bmbt_buf_ops);
- if (error)
- return error;
- *count += 1;
- block = XFS_BUF_TO_BLOCK(bp);
+ /*
+ * transform from extents to local?
+ */
+ error = 0;
+error0:
+ /*
+ * Log everything. Do this after conversion, there's no point in
+ * logging the extent records if we've converted to btree format.
+ */
+ if ((logflags & xfs_ilog_fext(whichfork)) &&
+ XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS)
+ logflags &= ~xfs_ilog_fext(whichfork);
+ else if ((logflags & xfs_ilog_fbroot(whichfork)) &&
+ XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)
+ logflags &= ~xfs_ilog_fbroot(whichfork);
+ /*
+ * Log inode even in the error case, if the transaction
+ * is dirty we'll need to shut down the filesystem.
+ */
+ if (logflags)
+ xfs_trans_log_inode(tp, ip, logflags);
+ if (cur) {
+ if (!error) {
+ *firstblock = cur->bc_private.b.firstblock;
+ cur->bc_private.b.allocated = 0;
}
+ xfs_btree_del_cursor(cur,
+ error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR);
}
- return 0;
-}
-
-/*
- * Count leaf blocks given a range of extent records.
- */
-STATIC void
-xfs_bmap_count_leaves(
- xfs_ifork_t *ifp,
- xfs_extnum_t idx,
- int numrecs,
- int *count)
-{
- int b;
-
- for (b = 0; b < numrecs; b++) {
- xfs_bmbt_rec_host_t *frp = xfs_iext_get_ext(ifp, idx + b);
- *count += xfs_bmbt_get_blockcount(frp);
- }
-}
-
-/*
- * Count leaf blocks given a range of extent records originally
- * in btree format.
- */
-STATIC void
-xfs_bmap_disk_count_leaves(
- struct xfs_mount *mp,
- struct xfs_btree_block *block,
- int numrecs,
- int *count)
-{
- int b;
- xfs_bmbt_rec_t *frp;
-
- for (b = 1; b <= numrecs; b++) {
- frp = XFS_BMBT_REC_ADDR(mp, block, b);
- *count += xfs_bmbt_disk_get_blockcount(frp);
- }
-}
-
-/*
- * Convert the given file system block to a disk block. We have to treat it
- * differently based on whether the file is a real time file or not, because the
- * bmap code does.
- */
-xfs_daddr_t
-xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb)
-{
- return (XFS_IS_REALTIME_INODE(ip) ? \
- (xfs_daddr_t)XFS_FSB_TO_BB((ip)->i_mount, (fsb)) : \
- XFS_FSB_TO_DADDR((ip)->i_mount, (fsb)));
+ return error;
}
diff --git a/libxfs/xfs_dir2_block.c b/libxfs/xfs_dir2_block.c
index 7397faa..2a99dea 100644
--- a/libxfs/xfs_dir2_block.c
+++ b/libxfs/xfs_dir2_block.c
@@ -339,10 +339,12 @@ xfs_dir2_block_addname(
/*
* If need to compact the leaf entries, do it now.
*/
- if (compact)
+ if (compact) {
xfs_dir2_block_compact(tp, bp, hdr, btp, blp, &needlog,
&lfloghigh, &lfloglow);
- else if (btp->stale) {
+ /* recalculate blp post-compaction */
+ blp = xfs_dir2_block_leaf_p(btp);
+ } else if (btp->stale) {
/*
* Set leaf logging boundaries to impossible state.
* For the no-stale case they're set explicitly.
diff --git a/libxfs/xfs_ialloc.c b/libxfs/xfs_ialloc.c
index 529d92d..f0322c9 100644
--- a/libxfs/xfs_ialloc.c
+++ b/libxfs/xfs_ialloc.c
@@ -260,6 +260,8 @@ xfs_ialloc_ag_alloc(
(args.agbno < be32_to_cpu(agi->agi_length)))) {
args.fsbno = XFS_AGB_TO_FSB(args.mp, agno, args.agbno);
args.type = XFS_ALLOCTYPE_THIS_BNO;
+ args.mod = args.total = args.wasdel = args.isfl =
+ args.userdata = args.minalignslop = 0;
args.prod = 1;
/*
@@ -312,6 +314,8 @@ xfs_ialloc_ag_alloc(
* Allocate a fixed-size extent of inodes.
*/
args.type = XFS_ALLOCTYPE_NEAR_BNO;
+ args.mod = args.total = args.wasdel = args.isfl =
+ args.userdata = args.minalignslop = 0;
args.prod = 1;
/*
* Allow space for the inode btree to split.
@@ -350,7 +354,7 @@ xfs_ialloc_ag_alloc(
* number from being easily guessable.
*/
error = xfs_ialloc_inode_init(args.mp, tp, agno, args.agbno,
- args.len, random32());
+ args.len, prandom_u32());
if (error)
return error;
diff --git a/libxfs/xfs_inode.c b/libxfs/xfs_inode.c
index 2970f46..3cf2423 100644
--- a/libxfs/xfs_inode.c
+++ b/libxfs/xfs_inode.c
@@ -1112,6 +1112,7 @@ xfs_iflush_fork(
char *cp;
xfs_ifork_t *ifp;
xfs_mount_t *mp;
+
static const short brootflag[2] =
{ XFS_ILOG_DBROOT, XFS_ILOG_ABROOT };
static const short dataflag[2] =
diff --git a/libxfs/xfs_mount.c b/libxfs/xfs_mount.c
index a9155b3..b7514fb 100644
--- a/libxfs/xfs_mount.c
+++ b/libxfs/xfs_mount.c
@@ -365,7 +365,7 @@ xfs_sb_quiet_read_verify(
return;
}
/* quietly fail */
- xfs_buf_ioerror(bp, EFSCORRUPTED);
+ xfs_buf_ioerror(bp, EWRONGFS);
}
static void
diff --git a/repair/dinode.c b/repair/dinode.c
index 5a2da39..f564216 100644
--- a/repair/dinode.c
+++ b/repair/dinode.c
@@ -250,7 +250,8 @@ clear_dinode(xfs_mount_t *mp, xfs_dinode_t *dino, xfs_ino_t ino_num)
/* and clear the forks */
if (dirty && !no_modify)
- memset(XFS_DFORK_DPTR(dino), 0, XFS_LITINO(mp));
+ memset(XFS_DFORK_DPTR(dino), 0,
+ XFS_LITINO(mp, dino->di_version));
return(dirty);
}
@@ -1919,11 +1920,12 @@ _("bad attr fork offset %d in dev inode %" PRIu64 ", should be %d\n"),
case XFS_DINODE_FMT_LOCAL: /* fall through ... */
case XFS_DINODE_FMT_EXTENTS: /* fall through ... */
case XFS_DINODE_FMT_BTREE:
- if (dino->di_forkoff >= (XFS_LITINO(mp) >> 3)) {
+ if (dino->di_forkoff >=
+ (XFS_LITINO(mp, dino->di_version) >> 3)) {
do_warn(
_("bad attr fork offset %d in inode %" PRIu64 ", max=%d\n"),
dino->di_forkoff, lino,
- XFS_LITINO(mp) >> 3);
+ XFS_LITINO(mp, dino->di_version) >> 3);
return 1;
}
break;
diff --git a/repair/prefetch.c b/repair/prefetch.c
index c0b1002..e0eff96 100644
--- a/repair/prefetch.c
+++ b/repair/prefetch.c
@@ -374,7 +374,7 @@ pf_read_inode_dirs(
continue;
if ((dino->di_forkoff != 0) &&
- (dino->di_forkoff >= (XFS_LITINO(mp) >> 3)))
+ (dino->di_forkoff >= XFS_LITINO(mp, dino->di_version) >> 3))
continue;
switch (dino->di_format) {
xfsprogs-crc-patchset-v2/xfsprogs-crc-infrastructure 0000600 0001750 0001750 00000155746 12132011546 021702 0 ustar dave dave xfsprogs: add CRC32c infrastructure
From: Dave Chinner <dchinner@xxxxxxxxxx>
Pull the generic crc32(c) code from the kernel and add it to libxfs.
Modify it to build in the libxfs environment, and drop the bigendian
CRC version as it is unused by XFS, which uses the little endian
version so that it can be hardware accelerated using native
instructions on x86-64 CPUs.
Also wire up the self-test code in the crc32 module to the build
infrastructure and make passing the self test a build dependency.
This prevents xfsprogs from being built on platforms that the CRC
algorithm does not work on and hence ensures the tools do not write
bad CRCs to disk as a result of a broken calculation.
Also pull the XFS CRC helper functions across in preparation for
using the CRC functions in libxfs.
XXX: something in the CRC table generation breaks the debian package
build. It fails to build libxfs as a dependency of mkfs.xfs. Works
fine outside the debian build environment, so I'm not sure what the
issue is yet. Most likely it is the execution path of the
gen_crc32table binary that is built...
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
include/Makefile | 6 +-
include/libxfs.h | 10 +
include/xfs_arch.h | 16 +
include/xfs_cksum.h | 63 +++
libxfs/Makefile | 26 +-
libxfs/crc32.c | 1036 +++++++++++++++++++++++++++++++++++++++++++++++
libxfs/crc32defs.h | 72 ++++
libxfs/gen_crc32table.c | 144 +++++++
8 files changed, 1367 insertions(+), 6 deletions(-)
diff --git a/include/Makefile b/include/Makefile
index de2c4f0..9ed5077 100644
--- a/include/Makefile
+++ b/include/Makefile
@@ -23,9 +23,9 @@ QAHFILES = libxfs.h libxlog.h \
swab.h \
xfs_ag.h xfs_alloc.h xfs_alloc_btree.h xfs_arch.h xfs_attr_leaf.h \
xfs_attr_sf.h xfs_bit.h xfs_bmap.h xfs_bmap_btree.h xfs_btree.h \
- xfs_btree_trace.h xfs_buf_item.h xfs_da_btree.h xfs_dinode.h \
- xfs_dir2.h xfs_dir2_format.h xfs_dir1.h xfs_dir_leaf.h xfs_dir_sf.h \
- xfs_extfree_item.h xfs_ialloc.h xfs_ialloc_btree.h \
+ xfs_btree_trace.h xfs_buf_item.h xfs_cksum.h xfs_da_btree.h \
+ xfs_dinode.h xfs_dir2.h xfs_dir2_format.h xfs_dir1.h xfs_dir_leaf.h \
+ xfs_dir_sf.h xfs_extfree_item.h xfs_ialloc.h xfs_ialloc_btree.h \
xfs_inode.h xfs_inode_item.h xfs_inum.h \
xfs_log.h xfs_log_priv.h xfs_log_recover.h xfs_metadump.h \
xfs_mount.h xfs_quota.h xfs_rtalloc.h xfs_sb.h xfs_trace.h \
diff --git a/include/libxfs.h b/include/libxfs.h
index 0e2fc5f..e671391 100644
--- a/include/libxfs.h
+++ b/include/libxfs.h
@@ -668,4 +668,14 @@ void xfs_dinode_from_disk(struct xfs_icdinode *,
/* xfs_rtalloc.c */
int libxfs_rtfree_extent(struct xfs_trans *, xfs_rtblock_t, xfs_extlen_t);
+/* CRC wrappers */
+
+extern uint32_t crc32_le(uint32_t crc, unsigned char const *p, size_t len);
+extern uint32_t crc32c_le(uint32_t crc, unsigned char const *p, size_t len);
+
+#define crc32(c,p,l) crc32_le((c),(unsigned char const *)(p),(l))
+#define crc32c(c,p,l) crc32c_le((c),(unsigned char const *)(p),(l))
+
+#include <xfs/xfs_cksum.h>
+
#endif /* __LIBXFS_H__ */
diff --git a/include/xfs_arch.h b/include/xfs_arch.h
index 0902249..d5e591a 100644
--- a/include/xfs_arch.h
+++ b/include/xfs_arch.h
@@ -47,6 +47,14 @@
#define be16_to_cpu(val) ((__force __u16)(__be16)(val))
#define be32_to_cpu(val) ((__force __u32)(__be32)(val))
#define be64_to_cpu(val) ((__force __u64)(__be64)(val))
+
+#define cpu_to_le32(val) ((__force __be32)__swab32((__u32)(val)))
+#define le32_to_cpu(val) (__swab32((__force __u32)(__le32)(val)))
+
+#define __constant_cpu_to_le32(val) \
+ ((__force __le32)___constant_swab32((__u32)(val)))
+#define __constant_cpu_to_be32(val) \
+ ((__force __be32)(__u32)(val))
#else
#define cpu_to_be16(val) ((__force __be16)__swab16((__u16)(val)))
#define cpu_to_be32(val) ((__force __be32)__swab32((__u32)(val)))
@@ -54,6 +62,14 @@
#define be16_to_cpu(val) (__swab16((__force __u16)(__be16)(val)))
#define be32_to_cpu(val) (__swab32((__force __u32)(__be32)(val)))
#define be64_to_cpu(val) (__swab64((__force __u64)(__be64)(val)))
+
+#define cpu_to_le32(val) ((__force __le32)(__u32)(val))
+#define le32_to_cpu(val) ((__force __u32)(__le32)(val))
+
+#define __constant_cpu_to_le32(val) \
+ ((__force __le32)(__u32)(val))
+#define __constant_cpu_to_be32(val) \
+ ((__force __be32)___constant_swab32((__u32)(val)))
#endif
static inline void be16_add_cpu(__be16 *a, __s16 b)
diff --git a/include/xfs_cksum.h b/include/xfs_cksum.h
new file mode 100644
index 0000000..fad1676
--- /dev/null
+++ b/include/xfs_cksum.h
@@ -0,0 +1,63 @@
+#ifndef _XFS_CKSUM_H
+#define _XFS_CKSUM_H 1
+
+#define XFS_CRC_SEED (~(__uint32_t)0)
+
+/*
+ * Calculate the intermediate checksum for a buffer that has the CRC field
+ * inside it. The offset of the 32bit crc fields is passed as the
+ * cksum_offset parameter.
+ */
+static inline __uint32_t
+xfs_start_cksum(char *buffer, size_t length, unsigned long cksum_offset)
+{
+ __uint32_t zero = 0;
+ __uint32_t crc;
+
+ /* Calculate CRC up to the checksum. */
+ crc = crc32c(XFS_CRC_SEED, buffer, cksum_offset);
+
+ /* Skip checksum field */
+ crc = crc32c(crc, &zero, sizeof(__u32));
+
+ /* Calculate the rest of the CRC. */
+ return crc32c(crc, &buffer[cksum_offset + sizeof(__be32)],
+ length - (cksum_offset + sizeof(__be32)));
+}
+
+/*
+ * Convert the intermediate checksum to the final ondisk format.
+ *
+ * The CRC32c calculation uses LE format even on BE machines, but returns the
+ * result in host endian format. Hence we need to byte swap it back to LE format
+ * so that it is consistent on disk.
+ */
+static inline __le32
+xfs_end_cksum(__uint32_t crc)
+{
+ return ~cpu_to_le32(crc);
+}
+
+/*
+ * Helper to generate the checksum for a buffer.
+ */
+static inline void
+xfs_update_cksum(char *buffer, size_t length, unsigned long cksum_offset)
+{
+ __uint32_t crc = xfs_start_cksum(buffer, length, cksum_offset);
+
+ *(__le32 *)(buffer + cksum_offset) = xfs_end_cksum(crc);
+}
+
+/*
+ * Helper to verify the checksum for a buffer.
+ */
+static inline int
+xfs_verify_cksum(char *buffer, size_t length, unsigned long cksum_offset)
+{
+ __uint32_t crc = xfs_start_cksum(buffer, length, cksum_offset);
+
+ return *(__le32 *)(buffer + cksum_offset) == xfs_end_cksum(crc);
+}
+
+#endif /* _XFS_CKSUM_H */
diff --git a/libxfs/Makefile b/libxfs/Makefile
index dc9b22c..28f71c8 100644
--- a/libxfs/Makefile
+++ b/libxfs/Makefile
@@ -10,17 +10,19 @@ LT_CURRENT = 0
LT_REVISION = 0
LT_AGE = 0
-HFILES = xfs.h init.h xfs_dir2_priv.h
+HFILES = xfs.h init.h xfs_dir2_priv.h crc32defs.h crc32table.h
CFILES = cache.c init.c kmem.c logitem.c radix-tree.c rdwr.c trans.c util.c \
xfs_alloc.c xfs_ialloc.c xfs_inode.c xfs_btree.c xfs_alloc_btree.c \
xfs_ialloc_btree.c xfs_bmap_btree.c xfs_da_btree.c \
xfs_dir2.c xfs_dir2_leaf.c xfs_attr_leaf.c xfs_dir2_block.c \
xfs_dir2_node.c xfs_dir2_data.c xfs_dir2_sf.c xfs_bmap.c \
- xfs_mount.c xfs_rtalloc.c xfs_trans.c xfs_attr.c
+ xfs_mount.c xfs_rtalloc.c xfs_trans.c xfs_attr.c \
+ crc32.c
CFILES += $(PKG_PLATFORM).c
PCFILES = darwin.c freebsd.c irix.c linux.c
LSRCFILES = $(shell echo $(PCFILES) | sed -e "s/$(PKG_PLATFORM).c//g")
+LSRCFILES += gen_crc32table.c
#
# Tracing flags:
@@ -38,7 +40,25 @@ LTLIBS = $(LIBPTHREAD) $(LIBRT)
# don't try linking xfs_repair with a debug libxfs.
DEBUG = -DNDEBUG
-default: ltdepend $(LTLIBRARY)
+LDIRT = gen_crc32table crc32table.h crc32selftest
+
+default: crc32selftest ltdepend $(LTLIBRARY)
+
+crc32table.h: gen_crc32table.c
+ @echo " [CC] gen_crc32table"
+ $(Q) $(CC) $(CFLAGS) -o gen_crc32table $<
+ @echo " [GENERATE] $@"
+ $(Q) ./gen_crc32table > crc32table.h
+
+# The selftest binary will return an error if it fails. This is made a
+# dependency of the build process so that we refuse to build the tools on broken
+# systems/architectures. Hence we make sure that xfsprogs will never use a
+# busted CRC calculation at build time and hence avoid putting bad CRCs down on
+# disk.
+crc32selftest: gen_crc32table.c crc32table.h crc32.c
+ @echo " [TEST] CRC32"
+ $(Q) $(CC) $(CFLAGS) -D CRC32_SELFTEST=1 crc32.c -o $@
+ $(Q) ./$@
include $(BUILDRULES)
diff --git a/libxfs/crc32.c b/libxfs/crc32.c
new file mode 100644
index 0000000..1c0d958
--- /dev/null
+++ b/libxfs/crc32.c
@@ -0,0 +1,1036 @@
+/*
+ * Aug 8, 2011 Bob Pearson with help from Joakim Tjernlund and George Spelvin
+ * cleaned up code to current version of sparse and added the slicing-by-8
+ * algorithm to the closely similar existing slicing-by-4 algorithm.
+ *
+ * Oct 15, 2000 Matt Domsch <Matt_Domsch@xxxxxxxx>
+ * Nicer crc32 functions/docs submitted by linux@xxxxxxxxxxx. Thanks!
+ * Code was from the public domain, copyright abandoned. Code was
+ * subsequently included in the kernel, thus was re-licensed under the
+ * GNU GPL v2.
+ *
+ * Oct 12, 2000 Matt Domsch <Matt_Domsch@xxxxxxxx>
+ * Same crc32 function was used in 5 other places in the kernel.
+ * I made one version, and deleted the others.
+ * There are various incantations of crc32(). Some use a seed of 0 or ~0.
+ * Some xor at the end with ~0. The generic crc32() function takes
+ * seed as an argument, and doesn't xor at the end. Then individual
+ * users can do whatever they need.
+ * drivers/net/smc9194.c uses seed ~0, doesn't xor with ~0.
+ * fs/jffs2 uses seed 0, doesn't xor with ~0.
+ * fs/partitions/efi.c uses seed ~0, xor's with ~0.
+ *
+ * This source code is licensed under the GNU General Public License,
+ * Version 2. See the file COPYING for more details.
+ */
+
+/* see: Documentation/crc32.txt for a description of algorithms */
+
+/*
+ * lifted from the 3.8-rc2 kernel source for xfsprogs. Killed CONFIG_X86
+ * specific bits for just the generic algorithm. Also removed the big endian
+ * version of the algorithm as XFS only uses the little endian CRC version to
+ * match the hardware acceleration available on Intel CPUs.
+ */
+
+#include <libxfs.h>
+#include "crc32defs.h"
+
+/* types specifc to this file */
+typedef __u8 u8;
+typedef __u16 u16;
+typedef __u32 u32;
+typedef __u32 u64;
+#define __pure
+
+#if CRC_LE_BITS > 8
+# define tole(x) ((__force u32) __constant_cpu_to_le32(x))
+#else
+# define tole(x) (x)
+#endif
+
+#if CRC_BE_BITS > 8
+# define tobe(x) ((__force u32) __constant_cpu_to_be32(x))
+#else
+# define tobe(x) (x)
+#endif
+
+#include "crc32table.h"
+
+#if CRC_LE_BITS > 8 || CRC_BE_BITS > 8
+
+/* implements slicing-by-4 or slicing-by-8 algorithm */
+static inline u32
+crc32_body(u32 crc, unsigned char const *buf, size_t len, const u32 (*tab)[256])
+{
+# ifdef __LITTLE_ENDIAN
+# define DO_CRC(x) crc = t0[(crc ^ (x)) & 255] ^ (crc >> 8)
+# define DO_CRC4 (t3[(q) & 255] ^ t2[(q >> 8) & 255] ^ \
+ t1[(q >> 16) & 255] ^ t0[(q >> 24) & 255])
+# define DO_CRC8 (t7[(q) & 255] ^ t6[(q >> 8) & 255] ^ \
+ t5[(q >> 16) & 255] ^ t4[(q >> 24) & 255])
+# else
+# define DO_CRC(x) crc = t0[((crc >> 24) ^ (x)) & 255] ^ (crc << 8)
+# define DO_CRC4 (t0[(q) & 255] ^ t1[(q >> 8) & 255] ^ \
+ t2[(q >> 16) & 255] ^ t3[(q >> 24) & 255])
+# define DO_CRC8 (t4[(q) & 255] ^ t5[(q >> 8) & 255] ^ \
+ t6[(q >> 16) & 255] ^ t7[(q >> 24) & 255])
+# endif
+ const u32 *b;
+ size_t rem_len;
+ const u32 *t0=tab[0], *t1=tab[1], *t2=tab[2], *t3=tab[3];
+# if CRC_LE_BITS != 32
+ const u32 *t4 = tab[4], *t5 = tab[5], *t6 = tab[6], *t7 = tab[7];
+# endif
+ u32 q;
+
+ /* Align it */
+ if (((long)buf & 3) && len) {
+ do {
+ DO_CRC(*buf++);
+ } while ((--len) && ((long)buf)&3);
+ }
+
+# if CRC_LE_BITS == 32
+ rem_len = len & 3;
+ len = len >> 2;
+# else
+ rem_len = len & 7;
+ len = len >> 3;
+# endif
+
+ b = (const u32 *)buf;
+ for (--b; len; --len) {
+ q = crc ^ *++b; /* use pre increment for speed */
+# if CRC_LE_BITS == 32
+ crc = DO_CRC4;
+# else
+ crc = DO_CRC8;
+ q = *++b;
+ crc ^= DO_CRC4;
+# endif
+ }
+ len = rem_len;
+ /* And the last few bytes */
+ if (len) {
+ u8 *p = (u8 *)(b + 1) - 1;
+ do {
+ DO_CRC(*++p); /* use pre increment for speed */
+ } while (--len);
+ }
+ return crc;
+#undef DO_CRC
+#undef DO_CRC4
+#undef DO_CRC8
+}
+#endif
+
+/**
+ * crc32_le() - Calculate bitwise little-endian Ethernet AUTODIN II CRC32
+ * @crc: seed value for computation. ~0 for Ethernet, sometimes 0 for
+ * other uses, or the previous crc32 value if computing incrementally.
+ * @p: pointer to buffer over which CRC is run
+ * @len: length of buffer @p
+ */
+static inline u32 __pure crc32_le_generic(u32 crc, unsigned char const *p,
+ size_t len, const u32 (*tab)[256],
+ u32 polynomial)
+{
+#if CRC_LE_BITS == 1
+ int i;
+ while (len--) {
+ crc ^= *p++;
+ for (i = 0; i < 8; i++)
+ crc = (crc >> 1) ^ ((crc & 1) ? polynomial : 0);
+ }
+# elif CRC_LE_BITS == 2
+ while (len--) {
+ crc ^= *p++;
+ crc = (crc >> 2) ^ tab[0][crc & 3];
+ crc = (crc >> 2) ^ tab[0][crc & 3];
+ crc = (crc >> 2) ^ tab[0][crc & 3];
+ crc = (crc >> 2) ^ tab[0][crc & 3];
+ }
+# elif CRC_LE_BITS == 4
+ while (len--) {
+ crc ^= *p++;
+ crc = (crc >> 4) ^ tab[0][crc & 15];
+ crc = (crc >> 4) ^ tab[0][crc & 15];
+ }
+# elif CRC_LE_BITS == 8
+ /* aka Sarwate algorithm */
+ while (len--) {
+ crc ^= *p++;
+ crc = (crc >> 8) ^ tab[0][crc & 255];
+ }
+# else
+ crc = (__force u32) cpu_to_le32(crc);
+ crc = crc32_body(crc, p, len, tab);
+ crc = le32_to_cpu((__force __le32)crc);
+#endif
+ return crc;
+}
+
+#if CRC_LE_BITS == 1
+u32 __pure crc32_le(u32 crc, unsigned char const *p, size_t len)
+{
+ return crc32_le_generic(crc, p, len, NULL, CRCPOLY_LE);
+}
+u32 __pure crc32c_le(u32 crc, unsigned char const *p, size_t len)
+{
+ return crc32_le_generic(crc, p, len, NULL, CRC32C_POLY_LE);
+}
+#else
+u32 __pure crc32_le(u32 crc, unsigned char const *p, size_t len)
+{
+ return crc32_le_generic(crc, p, len,
+ (const u32 (*)[256])crc32table_le, CRCPOLY_LE);
+}
+u32 __pure crc32c_le(u32 crc, unsigned char const *p, size_t len)
+{
+ return crc32_le_generic(crc, p, len,
+ (const u32 (*)[256])crc32ctable_le, CRC32C_POLY_LE);
+}
+#endif
+
+
+#ifdef CRC32_SELFTEST
+
+/* 4096 random bytes */
+static u8 __attribute__((__aligned__(8))) test_buf[] =
+{
+ 0x5b, 0x85, 0x21, 0xcb, 0x09, 0x68, 0x7d, 0x30,
+ 0xc7, 0x69, 0xd7, 0x30, 0x92, 0xde, 0x59, 0xe4,
+ 0xc9, 0x6e, 0x8b, 0xdb, 0x98, 0x6b, 0xaa, 0x60,
+ 0xa8, 0xb5, 0xbc, 0x6c, 0xa9, 0xb1, 0x5b, 0x2c,
+ 0xea, 0xb4, 0x92, 0x6a, 0x3f, 0x79, 0x91, 0xe4,
+ 0xe9, 0x70, 0x51, 0x8c, 0x7f, 0x95, 0x6f, 0x1a,
+ 0x56, 0xa1, 0x5c, 0x27, 0x03, 0x67, 0x9f, 0x3a,
+ 0xe2, 0x31, 0x11, 0x29, 0x6b, 0x98, 0xfc, 0xc4,
+ 0x53, 0x24, 0xc5, 0x8b, 0xce, 0x47, 0xb2, 0xb9,
+ 0x32, 0xcb, 0xc1, 0xd0, 0x03, 0x57, 0x4e, 0xd4,
+ 0xe9, 0x3c, 0xa1, 0x63, 0xcf, 0x12, 0x0e, 0xca,
+ 0xe1, 0x13, 0xd1, 0x93, 0xa6, 0x88, 0x5c, 0x61,
+ 0x5b, 0xbb, 0xf0, 0x19, 0x46, 0xb4, 0xcf, 0x9e,
+ 0xb6, 0x6b, 0x4c, 0x3a, 0xcf, 0x60, 0xf9, 0x7a,
+ 0x8d, 0x07, 0x63, 0xdb, 0x40, 0xe9, 0x0b, 0x6f,
+ 0xad, 0x97, 0xf1, 0xed, 0xd0, 0x1e, 0x26, 0xfd,
+ 0xbf, 0xb7, 0xc8, 0x04, 0x94, 0xf8, 0x8b, 0x8c,
+ 0xf1, 0xab, 0x7a, 0xd4, 0xdd, 0xf3, 0xe8, 0x88,
+ 0xc3, 0xed, 0x17, 0x8a, 0x9b, 0x40, 0x0d, 0x53,
+ 0x62, 0x12, 0x03, 0x5f, 0x1b, 0x35, 0x32, 0x1f,
+ 0xb4, 0x7b, 0x93, 0x78, 0x0d, 0xdb, 0xce, 0xa4,
+ 0xc0, 0x47, 0xd5, 0xbf, 0x68, 0xe8, 0x5d, 0x74,
+ 0x8f, 0x8e, 0x75, 0x1c, 0xb2, 0x4f, 0x9a, 0x60,
+ 0xd1, 0xbe, 0x10, 0xf4, 0x5c, 0xa1, 0x53, 0x09,
+ 0xa5, 0xe0, 0x09, 0x54, 0x85, 0x5c, 0xdc, 0x07,
+ 0xe7, 0x21, 0x69, 0x7b, 0x8a, 0xfd, 0x90, 0xf1,
+ 0x22, 0xd0, 0xb4, 0x36, 0x28, 0xe6, 0xb8, 0x0f,
+ 0x39, 0xde, 0xc8, 0xf3, 0x86, 0x60, 0x34, 0xd2,
+ 0x5e, 0xdf, 0xfd, 0xcf, 0x0f, 0xa9, 0x65, 0xf0,
+ 0xd5, 0x4d, 0x96, 0x40, 0xe3, 0xdf, 0x3f, 0x95,
+ 0x5a, 0x39, 0x19, 0x93, 0xf4, 0x75, 0xce, 0x22,
+ 0x00, 0x1c, 0x93, 0xe2, 0x03, 0x66, 0xf4, 0x93,
+ 0x73, 0x86, 0x81, 0x8e, 0x29, 0x44, 0x48, 0x86,
+ 0x61, 0x7c, 0x48, 0xa3, 0x43, 0xd2, 0x9c, 0x8d,
+ 0xd4, 0x95, 0xdd, 0xe1, 0x22, 0x89, 0x3a, 0x40,
+ 0x4c, 0x1b, 0x8a, 0x04, 0xa8, 0x09, 0x69, 0x8b,
+ 0xea, 0xc6, 0x55, 0x8e, 0x57, 0xe6, 0x64, 0x35,
+ 0xf0, 0xc7, 0x16, 0x9f, 0x5d, 0x5e, 0x86, 0x40,
+ 0x46, 0xbb, 0xe5, 0x45, 0x88, 0xfe, 0xc9, 0x63,
+ 0x15, 0xfb, 0xf5, 0xbd, 0x71, 0x61, 0xeb, 0x7b,
+ 0x78, 0x70, 0x07, 0x31, 0x03, 0x9f, 0xb2, 0xc8,
+ 0xa7, 0xab, 0x47, 0xfd, 0xdf, 0xa0, 0x78, 0x72,
+ 0xa4, 0x2a, 0xe4, 0xb6, 0xba, 0xc0, 0x1e, 0x86,
+ 0x71, 0xe6, 0x3d, 0x18, 0x37, 0x70, 0xe6, 0xff,
+ 0xe0, 0xbc, 0x0b, 0x22, 0xa0, 0x1f, 0xd3, 0xed,
+ 0xa2, 0x55, 0x39, 0xab, 0xa8, 0x13, 0x73, 0x7c,
+ 0x3f, 0xb2, 0xd6, 0x19, 0xac, 0xff, 0x99, 0xed,
+ 0xe8, 0xe6, 0xa6, 0x22, 0xe3, 0x9c, 0xf1, 0x30,
+ 0xdc, 0x01, 0x0a, 0x56, 0xfa, 0xe4, 0xc9, 0x99,
+ 0xdd, 0xa8, 0xd8, 0xda, 0x35, 0x51, 0x73, 0xb4,
+ 0x40, 0x86, 0x85, 0xdb, 0x5c, 0xd5, 0x85, 0x80,
+ 0x14, 0x9c, 0xfd, 0x98, 0xa9, 0x82, 0xc5, 0x37,
+ 0xff, 0x32, 0x5d, 0xd0, 0x0b, 0xfa, 0xdc, 0x04,
+ 0x5e, 0x09, 0xd2, 0xca, 0x17, 0x4b, 0x1a, 0x8e,
+ 0x15, 0xe1, 0xcc, 0x4e, 0x52, 0x88, 0x35, 0xbd,
+ 0x48, 0xfe, 0x15, 0xa0, 0x91, 0xfd, 0x7e, 0x6c,
+ 0x0e, 0x5d, 0x79, 0x1b, 0x81, 0x79, 0xd2, 0x09,
+ 0x34, 0x70, 0x3d, 0x81, 0xec, 0xf6, 0x24, 0xbb,
+ 0xfb, 0xf1, 0x7b, 0xdf, 0x54, 0xea, 0x80, 0x9b,
+ 0xc7, 0x99, 0x9e, 0xbd, 0x16, 0x78, 0x12, 0x53,
+ 0x5e, 0x01, 0xa7, 0x4e, 0xbd, 0x67, 0xe1, 0x9b,
+ 0x4c, 0x0e, 0x61, 0x45, 0x97, 0xd2, 0xf0, 0x0f,
+ 0xfe, 0x15, 0x08, 0xb7, 0x11, 0x4c, 0xe7, 0xff,
+ 0x81, 0x53, 0xff, 0x91, 0x25, 0x38, 0x7e, 0x40,
+ 0x94, 0xe5, 0xe0, 0xad, 0xe6, 0xd9, 0x79, 0xb6,
+ 0x92, 0xc9, 0xfc, 0xde, 0xc3, 0x1a, 0x23, 0xbb,
+ 0xdd, 0xc8, 0x51, 0x0c, 0x3a, 0x72, 0xfa, 0x73,
+ 0x6f, 0xb7, 0xee, 0x61, 0x39, 0x03, 0x01, 0x3f,
+ 0x7f, 0x94, 0x2e, 0x2e, 0xba, 0x3a, 0xbb, 0xb4,
+ 0xfa, 0x6a, 0x17, 0xfe, 0xea, 0xef, 0x5e, 0x66,
+ 0x97, 0x3f, 0x32, 0x3d, 0xd7, 0x3e, 0xb1, 0xf1,
+ 0x6c, 0x14, 0x4c, 0xfd, 0x37, 0xd3, 0x38, 0x80,
+ 0xfb, 0xde, 0xa6, 0x24, 0x1e, 0xc8, 0xca, 0x7f,
+ 0x3a, 0x93, 0xd8, 0x8b, 0x18, 0x13, 0xb2, 0xe5,
+ 0xe4, 0x93, 0x05, 0x53, 0x4f, 0x84, 0x66, 0xa7,
+ 0x58, 0x5c, 0x7b, 0x86, 0x52, 0x6d, 0x0d, 0xce,
+ 0xa4, 0x30, 0x7d, 0xb6, 0x18, 0x9f, 0xeb, 0xff,
+ 0x22, 0xbb, 0x72, 0x29, 0xb9, 0x44, 0x0b, 0x48,
+ 0x1e, 0x84, 0x71, 0x81, 0xe3, 0x6d, 0x73, 0x26,
+ 0x92, 0xb4, 0x4d, 0x2a, 0x29, 0xb8, 0x1f, 0x72,
+ 0xed, 0xd0, 0xe1, 0x64, 0x77, 0xea, 0x8e, 0x88,
+ 0x0f, 0xef, 0x3f, 0xb1, 0x3b, 0xad, 0xf9, 0xc9,
+ 0x8b, 0xd0, 0xac, 0xc6, 0xcc, 0xa9, 0x40, 0xcc,
+ 0x76, 0xf6, 0x3b, 0x53, 0xb5, 0x88, 0xcb, 0xc8,
+ 0x37, 0xf1, 0xa2, 0xba, 0x23, 0x15, 0x99, 0x09,
+ 0xcc, 0xe7, 0x7a, 0x3b, 0x37, 0xf7, 0x58, 0xc8,
+ 0x46, 0x8c, 0x2b, 0x2f, 0x4e, 0x0e, 0xa6, 0x5c,
+ 0xea, 0x85, 0x55, 0xba, 0x02, 0x0e, 0x0e, 0x48,
+ 0xbc, 0xe1, 0xb1, 0x01, 0x35, 0x79, 0x13, 0x3d,
+ 0x1b, 0xc0, 0x53, 0x68, 0x11, 0xe7, 0x95, 0x0f,
+ 0x9d, 0x3f, 0x4c, 0x47, 0x7b, 0x4d, 0x1c, 0xae,
+ 0x50, 0x9b, 0xcb, 0xdd, 0x05, 0x8d, 0x9a, 0x97,
+ 0xfd, 0x8c, 0xef, 0x0c, 0x1d, 0x67, 0x73, 0xa8,
+ 0x28, 0x36, 0xd5, 0xb6, 0x92, 0x33, 0x40, 0x75,
+ 0x0b, 0x51, 0xc3, 0x64, 0xba, 0x1d, 0xc2, 0xcc,
+ 0xee, 0x7d, 0x54, 0x0f, 0x27, 0x69, 0xa7, 0x27,
+ 0x63, 0x30, 0x29, 0xd9, 0xc8, 0x84, 0xd8, 0xdf,
+ 0x9f, 0x68, 0x8d, 0x04, 0xca, 0xa6, 0xc5, 0xc7,
+ 0x7a, 0x5c, 0xc8, 0xd1, 0xcb, 0x4a, 0xec, 0xd0,
+ 0xd8, 0x20, 0x69, 0xc5, 0x17, 0xcd, 0x78, 0xc8,
+ 0x75, 0x23, 0x30, 0x69, 0xc9, 0xd4, 0xea, 0x5c,
+ 0x4f, 0x6b, 0x86, 0x3f, 0x8b, 0xfe, 0xee, 0x44,
+ 0xc9, 0x7c, 0xb7, 0xdd, 0x3e, 0xe5, 0xec, 0x54,
+ 0x03, 0x3e, 0xaa, 0x82, 0xc6, 0xdf, 0xb2, 0x38,
+ 0x0e, 0x5d, 0xb3, 0x88, 0xd9, 0xd3, 0x69, 0x5f,
+ 0x8f, 0x70, 0x8a, 0x7e, 0x11, 0xd9, 0x1e, 0x7b,
+ 0x38, 0xf1, 0x42, 0x1a, 0xc0, 0x35, 0xf5, 0xc7,
+ 0x36, 0x85, 0xf5, 0xf7, 0xb8, 0x7e, 0xc7, 0xef,
+ 0x18, 0xf1, 0x63, 0xd6, 0x7a, 0xc6, 0xc9, 0x0e,
+ 0x4d, 0x69, 0x4f, 0x84, 0xef, 0x26, 0x41, 0x0c,
+ 0xec, 0xc7, 0xe0, 0x7e, 0x3c, 0x67, 0x01, 0x4c,
+ 0x62, 0x1a, 0x20, 0x6f, 0xee, 0x47, 0x4d, 0xc0,
+ 0x99, 0x13, 0x8d, 0x91, 0x4a, 0x26, 0xd4, 0x37,
+ 0x28, 0x90, 0x58, 0x75, 0x66, 0x2b, 0x0a, 0xdf,
+ 0xda, 0xee, 0x92, 0x25, 0x90, 0x62, 0x39, 0x9e,
+ 0x44, 0x98, 0xad, 0xc1, 0x88, 0xed, 0xe4, 0xb4,
+ 0xaf, 0xf5, 0x8c, 0x9b, 0x48, 0x4d, 0x56, 0x60,
+ 0x97, 0x0f, 0x61, 0x59, 0x9e, 0xa6, 0x27, 0xfe,
+ 0xc1, 0x91, 0x15, 0x38, 0xb8, 0x0f, 0xae, 0x61,
+ 0x7d, 0x26, 0x13, 0x5a, 0x73, 0xff, 0x1c, 0xa3,
+ 0x61, 0x04, 0x58, 0x48, 0x55, 0x44, 0x11, 0xfe,
+ 0x15, 0xca, 0xc3, 0xbd, 0xca, 0xc5, 0xb4, 0x40,
+ 0x5d, 0x1b, 0x7f, 0x39, 0xb5, 0x9c, 0x35, 0xec,
+ 0x61, 0x15, 0x32, 0x32, 0xb8, 0x4e, 0x40, 0x9f,
+ 0x17, 0x1f, 0x0a, 0x4d, 0xa9, 0x91, 0xef, 0xb7,
+ 0xb0, 0xeb, 0xc2, 0x83, 0x9a, 0x6c, 0xd2, 0x79,
+ 0x43, 0x78, 0x5e, 0x2f, 0xe5, 0xdd, 0x1a, 0x3c,
+ 0x45, 0xab, 0x29, 0x40, 0x3a, 0x37, 0x5b, 0x6f,
+ 0xd7, 0xfc, 0x48, 0x64, 0x3c, 0x49, 0xfb, 0x21,
+ 0xbe, 0xc3, 0xff, 0x07, 0xfb, 0x17, 0xe9, 0xc9,
+ 0x0c, 0x4c, 0x5c, 0x15, 0x9e, 0x8e, 0x22, 0x30,
+ 0x0a, 0xde, 0x48, 0x7f, 0xdb, 0x0d, 0xd1, 0x2b,
+ 0x87, 0x38, 0x9e, 0xcc, 0x5a, 0x01, 0x16, 0xee,
+ 0x75, 0x49, 0x0d, 0x30, 0x01, 0x34, 0x6a, 0xb6,
+ 0x9a, 0x5a, 0x2a, 0xec, 0xbb, 0x48, 0xac, 0xd3,
+ 0x77, 0x83, 0xd8, 0x08, 0x86, 0x4f, 0x48, 0x09,
+ 0x29, 0x41, 0x79, 0xa1, 0x03, 0x12, 0xc4, 0xcd,
+ 0x90, 0x55, 0x47, 0x66, 0x74, 0x9a, 0xcc, 0x4f,
+ 0x35, 0x8c, 0xd6, 0x98, 0xef, 0xeb, 0x45, 0xb9,
+ 0x9a, 0x26, 0x2f, 0x39, 0xa5, 0x70, 0x6d, 0xfc,
+ 0xb4, 0x51, 0xee, 0xf4, 0x9c, 0xe7, 0x38, 0x59,
+ 0xad, 0xf4, 0xbc, 0x46, 0xff, 0x46, 0x8e, 0x60,
+ 0x9c, 0xa3, 0x60, 0x1d, 0xf8, 0x26, 0x72, 0xf5,
+ 0x72, 0x9d, 0x68, 0x80, 0x04, 0xf6, 0x0b, 0xa1,
+ 0x0a, 0xd5, 0xa7, 0x82, 0x3a, 0x3e, 0x47, 0xa8,
+ 0x5a, 0xde, 0x59, 0x4f, 0x7b, 0x07, 0xb3, 0xe9,
+ 0x24, 0x19, 0x3d, 0x34, 0x05, 0xec, 0xf1, 0xab,
+ 0x6e, 0x64, 0x8f, 0xd3, 0xe6, 0x41, 0x86, 0x80,
+ 0x70, 0xe3, 0x8d, 0x60, 0x9c, 0x34, 0x25, 0x01,
+ 0x07, 0x4d, 0x19, 0x41, 0x4e, 0x3d, 0x5c, 0x7e,
+ 0xa8, 0xf5, 0xcc, 0xd5, 0x7b, 0xe2, 0x7d, 0x3d,
+ 0x49, 0x86, 0x7d, 0x07, 0xb7, 0x10, 0xe3, 0x35,
+ 0xb8, 0x84, 0x6d, 0x76, 0xab, 0x17, 0xc6, 0x38,
+ 0xb4, 0xd3, 0x28, 0x57, 0xad, 0xd3, 0x88, 0x5a,
+ 0xda, 0xea, 0xc8, 0x94, 0xcc, 0x37, 0x19, 0xac,
+ 0x9c, 0x9f, 0x4b, 0x00, 0x15, 0xc0, 0xc8, 0xca,
+ 0x1f, 0x15, 0xaa, 0xe0, 0xdb, 0xf9, 0x2f, 0x57,
+ 0x1b, 0x24, 0xc7, 0x6f, 0x76, 0x29, 0xfb, 0xed,
+ 0x25, 0x0d, 0xc0, 0xfe, 0xbd, 0x5a, 0xbf, 0x20,
+ 0x08, 0x51, 0x05, 0xec, 0x71, 0xa3, 0xbf, 0xef,
+ 0x5e, 0x99, 0x75, 0xdb, 0x3c, 0x5f, 0x9a, 0x8c,
+ 0xbb, 0x19, 0x5c, 0x0e, 0x93, 0x19, 0xf8, 0x6a,
+ 0xbc, 0xf2, 0x12, 0x54, 0x2f, 0xcb, 0x28, 0x64,
+ 0x88, 0xb3, 0x92, 0x0d, 0x96, 0xd1, 0xa6, 0xe4,
+ 0x1f, 0xf1, 0x4d, 0xa4, 0xab, 0x1c, 0xee, 0x54,
+ 0xf2, 0xad, 0x29, 0x6d, 0x32, 0x37, 0xb2, 0x16,
+ 0x77, 0x5c, 0xdc, 0x2e, 0x54, 0xec, 0x75, 0x26,
+ 0xc6, 0x36, 0xd9, 0x17, 0x2c, 0xf1, 0x7a, 0xdc,
+ 0x4b, 0xf1, 0xe2, 0xd9, 0x95, 0xba, 0xac, 0x87,
+ 0xc1, 0xf3, 0x8e, 0x58, 0x08, 0xd8, 0x87, 0x60,
+ 0xc9, 0xee, 0x6a, 0xde, 0xa4, 0xd2, 0xfc, 0x0d,
+ 0xe5, 0x36, 0xc4, 0x5c, 0x52, 0xb3, 0x07, 0x54,
+ 0x65, 0x24, 0xc1, 0xb1, 0xd1, 0xb1, 0x53, 0x13,
+ 0x31, 0x79, 0x7f, 0x05, 0x76, 0xeb, 0x37, 0x59,
+ 0x15, 0x2b, 0xd1, 0x3f, 0xac, 0x08, 0x97, 0xeb,
+ 0x91, 0x98, 0xdf, 0x6c, 0x09, 0x0d, 0x04, 0x9f,
+ 0xdc, 0x3b, 0x0e, 0x60, 0x68, 0x47, 0x23, 0x15,
+ 0x16, 0xc6, 0x0b, 0x35, 0xf8, 0x77, 0xa2, 0x78,
+ 0x50, 0xd4, 0x64, 0x22, 0x33, 0xff, 0xfb, 0x93,
+ 0x71, 0x46, 0x50, 0x39, 0x1b, 0x9c, 0xea, 0x4e,
+ 0x8d, 0x0c, 0x37, 0xe5, 0x5c, 0x51, 0x3a, 0x31,
+ 0xb2, 0x85, 0x84, 0x3f, 0x41, 0xee, 0xa2, 0xc1,
+ 0xc6, 0x13, 0x3b, 0x54, 0x28, 0xd2, 0x18, 0x37,
+ 0xcc, 0x46, 0x9f, 0x6a, 0x91, 0x3d, 0x5a, 0x15,
+ 0x3c, 0x89, 0xa3, 0x61, 0x06, 0x7d, 0x2e, 0x78,
+ 0xbe, 0x7d, 0x40, 0xba, 0x2f, 0x95, 0xb1, 0x2f,
+ 0x87, 0x3b, 0x8a, 0xbe, 0x6a, 0xf4, 0xc2, 0x31,
+ 0x74, 0xee, 0x91, 0xe0, 0x23, 0xaa, 0x5d, 0x7f,
+ 0xdd, 0xf0, 0x44, 0x8c, 0x0b, 0x59, 0x2b, 0xfc,
+ 0x48, 0x3a, 0xdf, 0x07, 0x05, 0x38, 0x6c, 0xc9,
+ 0xeb, 0x18, 0x24, 0x68, 0x8d, 0x58, 0x98, 0xd3,
+ 0x31, 0xa3, 0xe4, 0x70, 0x59, 0xb1, 0x21, 0xbe,
+ 0x7e, 0x65, 0x7d, 0xb8, 0x04, 0xab, 0xf6, 0xe4,
+ 0xd7, 0xda, 0xec, 0x09, 0x8f, 0xda, 0x6d, 0x24,
+ 0x07, 0xcc, 0x29, 0x17, 0x05, 0x78, 0x1a, 0xc1,
+ 0xb1, 0xce, 0xfc, 0xaa, 0x2d, 0xe7, 0xcc, 0x85,
+ 0x84, 0x84, 0x03, 0x2a, 0x0c, 0x3f, 0xa9, 0xf8,
+ 0xfd, 0x84, 0x53, 0x59, 0x5c, 0xf0, 0xd4, 0x09,
+ 0xf0, 0xd2, 0x6c, 0x32, 0x03, 0xb0, 0xa0, 0x8c,
+ 0x52, 0xeb, 0x23, 0x91, 0x88, 0x43, 0x13, 0x46,
+ 0xf6, 0x1e, 0xb4, 0x1b, 0xf5, 0x8e, 0x3a, 0xb5,
+ 0x3d, 0x00, 0xf6, 0xe5, 0x08, 0x3d, 0x5f, 0x39,
+ 0xd3, 0x21, 0x69, 0xbc, 0x03, 0x22, 0x3a, 0xd2,
+ 0x5c, 0x84, 0xf8, 0x15, 0xc4, 0x80, 0x0b, 0xbc,
+ 0x29, 0x3c, 0xf3, 0x95, 0x98, 0xcd, 0x8f, 0x35,
+ 0xbc, 0xa5, 0x3e, 0xfc, 0xd4, 0x13, 0x9e, 0xde,
+ 0x4f, 0xce, 0x71, 0x9d, 0x09, 0xad, 0xf2, 0x80,
+ 0x6b, 0x65, 0x7f, 0x03, 0x00, 0x14, 0x7c, 0x15,
+ 0x85, 0x40, 0x6d, 0x70, 0xea, 0xdc, 0xb3, 0x63,
+ 0x35, 0x4f, 0x4d, 0xe0, 0xd9, 0xd5, 0x3c, 0x58,
+ 0x56, 0x23, 0x80, 0xe2, 0x36, 0xdd, 0x75, 0x1d,
+ 0x94, 0x11, 0x41, 0x8e, 0xe0, 0x81, 0x8e, 0xcf,
+ 0xe0, 0xe5, 0xf6, 0xde, 0xd1, 0xe7, 0x04, 0x12,
+ 0x79, 0x92, 0x2b, 0x71, 0x2a, 0x79, 0x8b, 0x7c,
+ 0x44, 0x79, 0x16, 0x30, 0x4e, 0xf4, 0xf6, 0x9b,
+ 0xb7, 0x40, 0xa3, 0x5a, 0xa7, 0x69, 0x3e, 0xc1,
+ 0x3a, 0x04, 0xd0, 0x88, 0xa0, 0x3b, 0xdd, 0xc6,
+ 0x9e, 0x7e, 0x1e, 0x1e, 0x8f, 0x44, 0xf7, 0x73,
+ 0x67, 0x1e, 0x1a, 0x78, 0xfa, 0x62, 0xf4, 0xa9,
+ 0xa8, 0xc6, 0x5b, 0xb8, 0xfa, 0x06, 0x7d, 0x5e,
+ 0x38, 0x1c, 0x9a, 0x39, 0xe9, 0x39, 0x98, 0x22,
+ 0x0b, 0xa7, 0xac, 0x0b, 0xf3, 0xbc, 0xf1, 0xeb,
+ 0x8c, 0x81, 0xe3, 0x48, 0x8a, 0xed, 0x42, 0xc2,
+ 0x38, 0xcf, 0x3e, 0xda, 0xd2, 0x89, 0x8d, 0x9c,
+ 0x53, 0xb5, 0x2f, 0x41, 0x01, 0x26, 0x84, 0x9c,
+ 0xa3, 0x56, 0xf6, 0x49, 0xc7, 0xd4, 0x9f, 0x93,
+ 0x1b, 0x96, 0x49, 0x5e, 0xad, 0xb3, 0x84, 0x1f,
+ 0x3c, 0xa4, 0xe0, 0x9b, 0xd1, 0x90, 0xbc, 0x38,
+ 0x6c, 0xdd, 0x95, 0x4d, 0x9d, 0xb1, 0x71, 0x57,
+ 0x2d, 0x34, 0xe8, 0xb8, 0x42, 0xc7, 0x99, 0x03,
+ 0xc7, 0x07, 0x30, 0x65, 0x91, 0x55, 0xd5, 0x90,
+ 0x70, 0x97, 0x37, 0x68, 0xd4, 0x11, 0xf9, 0xe8,
+ 0xce, 0xec, 0xdc, 0x34, 0xd5, 0xd3, 0xb7, 0xc4,
+ 0xb8, 0x97, 0x05, 0x92, 0xad, 0xf8, 0xe2, 0x36,
+ 0x64, 0x41, 0xc9, 0xc5, 0x41, 0x77, 0x52, 0xd7,
+ 0x2c, 0xa5, 0x24, 0x2f, 0xd9, 0x34, 0x0b, 0x47,
+ 0x35, 0xa7, 0x28, 0x8b, 0xc5, 0xcd, 0xe9, 0x46,
+ 0xac, 0x39, 0x94, 0x3c, 0x10, 0xc6, 0x29, 0x73,
+ 0x0e, 0x0e, 0x5d, 0xe0, 0x71, 0x03, 0x8a, 0x72,
+ 0x0e, 0x26, 0xb0, 0x7d, 0x84, 0xed, 0x95, 0x23,
+ 0x49, 0x5a, 0x45, 0x83, 0x45, 0x60, 0x11, 0x4a,
+ 0x46, 0x31, 0xd4, 0xd8, 0x16, 0x54, 0x98, 0x58,
+ 0xed, 0x6d, 0xcc, 0x5d, 0xd6, 0x50, 0x61, 0x9f,
+ 0x9d, 0xc5, 0x3e, 0x9d, 0x32, 0x47, 0xde, 0x96,
+ 0xe1, 0x5d, 0xd8, 0xf8, 0xb4, 0x69, 0x6f, 0xb9,
+ 0x15, 0x90, 0x57, 0x7a, 0xf6, 0xad, 0xb0, 0x5b,
+ 0xf5, 0xa6, 0x36, 0x94, 0xfd, 0x84, 0xce, 0x1c,
+ 0x0f, 0x4b, 0xd0, 0xc2, 0x5b, 0x6b, 0x56, 0xef,
+ 0x73, 0x93, 0x0b, 0xc3, 0xee, 0xd9, 0xcf, 0xd3,
+ 0xa4, 0x22, 0x58, 0xcd, 0x50, 0x6e, 0x65, 0xf4,
+ 0xe9, 0xb7, 0x71, 0xaf, 0x4b, 0xb3, 0xb6, 0x2f,
+ 0x0f, 0x0e, 0x3b, 0xc9, 0x85, 0x14, 0xf5, 0x17,
+ 0xe8, 0x7a, 0x3a, 0xbf, 0x5f, 0x5e, 0xf8, 0x18,
+ 0x48, 0xa6, 0x72, 0xab, 0x06, 0x95, 0xe9, 0xc8,
+ 0xa7, 0xf4, 0x32, 0x44, 0x04, 0x0c, 0x84, 0x98,
+ 0x73, 0xe3, 0x89, 0x8d, 0x5f, 0x7e, 0x4a, 0x42,
+ 0x8f, 0xc5, 0x28, 0xb1, 0x82, 0xef, 0x1c, 0x97,
+ 0x31, 0x3b, 0x4d, 0xe0, 0x0e, 0x10, 0x10, 0x97,
+ 0x93, 0x49, 0x78, 0x2f, 0x0d, 0x86, 0x8b, 0xa1,
+ 0x53, 0xa9, 0x81, 0x20, 0x79, 0xe7, 0x07, 0x77,
+ 0xb6, 0xac, 0x5e, 0xd2, 0x05, 0xcd, 0xe9, 0xdb,
+ 0x8a, 0x94, 0x82, 0x8a, 0x23, 0xb9, 0x3d, 0x1c,
+ 0xa9, 0x7d, 0x72, 0x4a, 0xed, 0x33, 0xa3, 0xdb,
+ 0x21, 0xa7, 0x86, 0x33, 0x45, 0xa5, 0xaa, 0x56,
+ 0x45, 0xb5, 0x83, 0x29, 0x40, 0x47, 0x79, 0x04,
+ 0x6e, 0xb9, 0x95, 0xd0, 0x81, 0x77, 0x2d, 0x48,
+ 0x1e, 0xfe, 0xc3, 0xc2, 0x1e, 0xe5, 0xf2, 0xbe,
+ 0xfd, 0x3b, 0x94, 0x9f, 0xc4, 0xc4, 0x26, 0x9d,
+ 0xe4, 0x66, 0x1e, 0x19, 0xee, 0x6c, 0x79, 0x97,
+ 0x11, 0x31, 0x4b, 0x0d, 0x01, 0xcb, 0xde, 0xa8,
+ 0xf6, 0x6d, 0x7c, 0x39, 0x46, 0x4e, 0x7e, 0x3f,
+ 0x94, 0x17, 0xdf, 0xa1, 0x7d, 0xd9, 0x1c, 0x8e,
+ 0xbc, 0x7d, 0x33, 0x7d, 0xe3, 0x12, 0x40, 0xca,
+ 0xab, 0x37, 0x11, 0x46, 0xd4, 0xae, 0xef, 0x44,
+ 0xa2, 0xb3, 0x6a, 0x66, 0x0e, 0x0c, 0x90, 0x7f,
+ 0xdf, 0x5c, 0x66, 0x5f, 0xf2, 0x94, 0x9f, 0xa6,
+ 0x73, 0x4f, 0xeb, 0x0d, 0xad, 0xbf, 0xc0, 0x63,
+ 0x5c, 0xdc, 0x46, 0x51, 0xe8, 0x8e, 0x90, 0x19,
+ 0xa8, 0xa4, 0x3c, 0x91, 0x79, 0xfa, 0x7e, 0x58,
+ 0x85, 0x13, 0x55, 0xc5, 0x19, 0x82, 0x37, 0x1b,
+ 0x0a, 0x02, 0x1f, 0x99, 0x6b, 0x18, 0xf1, 0x28,
+ 0x08, 0xa2, 0x73, 0xb8, 0x0f, 0x2e, 0xcd, 0xbf,
+ 0xf3, 0x86, 0x7f, 0xea, 0xef, 0xd0, 0xbb, 0xa6,
+ 0x21, 0xdf, 0x49, 0x73, 0x51, 0xcc, 0x36, 0xd3,
+ 0x3e, 0xa0, 0xf8, 0x44, 0xdf, 0xd3, 0xa6, 0xbe,
+ 0x8a, 0xd4, 0x57, 0xdd, 0x72, 0x94, 0x61, 0x0f,
+ 0x82, 0xd1, 0x07, 0xb8, 0x7c, 0x18, 0x83, 0xdf,
+ 0x3a, 0xe5, 0x50, 0x6a, 0x82, 0x20, 0xac, 0xa9,
+ 0xa8, 0xff, 0xd9, 0xf3, 0x77, 0x33, 0x5a, 0x9e,
+ 0x7f, 0x6d, 0xfe, 0x5d, 0x33, 0x41, 0x42, 0xe7,
+ 0x6c, 0x19, 0xe0, 0x44, 0x8a, 0x15, 0xf6, 0x70,
+ 0x98, 0xb7, 0x68, 0x4d, 0xfa, 0x97, 0x39, 0xb0,
+ 0x8e, 0xe8, 0x84, 0x8b, 0x75, 0x30, 0xb7, 0x7d,
+ 0x92, 0x69, 0x20, 0x9c, 0x81, 0xfb, 0x4b, 0xf4,
+ 0x01, 0x50, 0xeb, 0xce, 0x0c, 0x1c, 0x6c, 0xb5,
+ 0x4a, 0xd7, 0x27, 0x0c, 0xce, 0xbb, 0xe5, 0x85,
+ 0xf0, 0xb6, 0xee, 0xd5, 0x70, 0xdd, 0x3b, 0xfc,
+ 0xd4, 0x99, 0xf1, 0x33, 0xdd, 0x8b, 0xc4, 0x2f,
+ 0xae, 0xab, 0x74, 0x96, 0x32, 0xc7, 0x4c, 0x56,
+ 0x3c, 0x89, 0x0f, 0x96, 0x0b, 0x42, 0xc0, 0xcb,
+ 0xee, 0x0f, 0x0b, 0x8c, 0xfb, 0x7e, 0x47, 0x7b,
+ 0x64, 0x48, 0xfd, 0xb2, 0x00, 0x80, 0x89, 0xa5,
+ 0x13, 0x55, 0x62, 0xfc, 0x8f, 0xe2, 0x42, 0x03,
+ 0xb7, 0x4e, 0x2a, 0x79, 0xb4, 0x82, 0xea, 0x23,
+ 0x49, 0xda, 0xaf, 0x52, 0x63, 0x1e, 0x60, 0x03,
+ 0x89, 0x06, 0x44, 0x46, 0x08, 0xc3, 0xc4, 0x87,
+ 0x70, 0x2e, 0xda, 0x94, 0xad, 0x6b, 0xe0, 0xe4,
+ 0xd1, 0x8a, 0x06, 0xc2, 0xa8, 0xc0, 0xa7, 0x43,
+ 0x3c, 0x47, 0x52, 0x0e, 0xc3, 0x77, 0x81, 0x11,
+ 0x67, 0x0e, 0xa0, 0x70, 0x04, 0x47, 0x29, 0x40,
+ 0x86, 0x0d, 0x34, 0x56, 0xa7, 0xc9, 0x35, 0x59,
+ 0x68, 0xdc, 0x93, 0x81, 0x70, 0xee, 0x86, 0xd9,
+ 0x80, 0x06, 0x40, 0x4f, 0x1a, 0x0d, 0x40, 0x30,
+ 0x0b, 0xcb, 0x96, 0x47, 0xc1, 0xb7, 0x52, 0xfd,
+ 0x56, 0xe0, 0x72, 0x4b, 0xfb, 0xbd, 0x92, 0x45,
+ 0x61, 0x71, 0xc2, 0x33, 0x11, 0xbf, 0x52, 0x83,
+ 0x79, 0x26, 0xe0, 0x49, 0x6b, 0xb7, 0x05, 0x8b,
+ 0xe8, 0x0e, 0x87, 0x31, 0xd7, 0x9d, 0x8a, 0xf5,
+ 0xc0, 0x5f, 0x2e, 0x58, 0x4a, 0xdb, 0x11, 0xb3,
+ 0x6c, 0x30, 0x2a, 0x46, 0x19, 0xe3, 0x27, 0x84,
+ 0x1f, 0x63, 0x6e, 0xf6, 0x57, 0xc7, 0xc9, 0xd8,
+ 0x5e, 0xba, 0xb3, 0x87, 0xd5, 0x83, 0x26, 0x34,
+ 0x21, 0x9e, 0x65, 0xde, 0x42, 0xd3, 0xbe, 0x7b,
+ 0xbc, 0x91, 0x71, 0x44, 0x4d, 0x99, 0x3b, 0x31,
+ 0xe5, 0x3f, 0x11, 0x4e, 0x7f, 0x13, 0x51, 0x3b,
+ 0xae, 0x79, 0xc9, 0xd3, 0x81, 0x8e, 0x25, 0x40,
+ 0x10, 0xfc, 0x07, 0x1e, 0xf9, 0x7b, 0x9a, 0x4b,
+ 0x6c, 0xe3, 0xb3, 0xad, 0x1a, 0x0a, 0xdd, 0x9e,
+ 0x59, 0x0c, 0xa2, 0xcd, 0xae, 0x48, 0x4a, 0x38,
+ 0x5b, 0x47, 0x41, 0x94, 0x65, 0x6b, 0xbb, 0xeb,
+ 0x5b, 0xe3, 0xaf, 0x07, 0x5b, 0xd4, 0x4a, 0xa2,
+ 0xc9, 0x5d, 0x2f, 0x64, 0x03, 0xd7, 0x3a, 0x2c,
+ 0x6e, 0xce, 0x76, 0x95, 0xb4, 0xb3, 0xc0, 0xf1,
+ 0xe2, 0x45, 0x73, 0x7a, 0x5c, 0xab, 0xc1, 0xfc,
+ 0x02, 0x8d, 0x81, 0x29, 0xb3, 0xac, 0x07, 0xec,
+ 0x40, 0x7d, 0x45, 0xd9, 0x7a, 0x59, 0xee, 0x34,
+ 0xf0, 0xe9, 0xd5, 0x7b, 0x96, 0xb1, 0x3d, 0x95,
+ 0xcc, 0x86, 0xb5, 0xb6, 0x04, 0x2d, 0xb5, 0x92,
+ 0x7e, 0x76, 0xf4, 0x06, 0xa9, 0xa3, 0x12, 0x0f,
+ 0xb1, 0xaf, 0x26, 0xba, 0x7c, 0xfc, 0x7e, 0x1c,
+ 0xbc, 0x2c, 0x49, 0x97, 0x53, 0x60, 0x13, 0x0b,
+ 0xa6, 0x61, 0x83, 0x89, 0x42, 0xd4, 0x17, 0x0c,
+ 0x6c, 0x26, 0x52, 0xc3, 0xb3, 0xd4, 0x67, 0xf5,
+ 0xe3, 0x04, 0xb7, 0xf4, 0xcb, 0x80, 0xb8, 0xcb,
+ 0x77, 0x56, 0x3e, 0xaa, 0x57, 0x54, 0xee, 0xb4,
+ 0x2c, 0x67, 0xcf, 0xf2, 0xdc, 0xbe, 0x55, 0xf9,
+ 0x43, 0x1f, 0x6e, 0x22, 0x97, 0x67, 0x7f, 0xc4,
+ 0xef, 0xb1, 0x26, 0x31, 0x1e, 0x27, 0xdf, 0x41,
+ 0x80, 0x47, 0x6c, 0xe2, 0xfa, 0xa9, 0x8c, 0x2a,
+ 0xf6, 0xf2, 0xab, 0xf0, 0x15, 0xda, 0x6c, 0xc8,
+ 0xfe, 0xb5, 0x23, 0xde, 0xa9, 0x05, 0x3f, 0x06,
+ 0x54, 0x4c, 0xcd, 0xe1, 0xab, 0xfc, 0x0e, 0x62,
+ 0x33, 0x31, 0x73, 0x2c, 0x76, 0xcb, 0xb4, 0x47,
+ 0x1e, 0x20, 0xad, 0xd8, 0xf2, 0x31, 0xdd, 0xc4,
+ 0x8b, 0x0c, 0x77, 0xbe, 0xe1, 0x8b, 0x26, 0x00,
+ 0x02, 0x58, 0xd6, 0x8d, 0xef, 0xad, 0x74, 0x67,
+ 0xab, 0x3f, 0xef, 0xcb, 0x6f, 0xb0, 0xcc, 0x81,
+ 0x44, 0x4c, 0xaf, 0xe9, 0x49, 0x4f, 0xdb, 0xa0,
+ 0x25, 0xa4, 0xf0, 0x89, 0xf1, 0xbe, 0xd8, 0x10,
+ 0xff, 0xb1, 0x3b, 0x4b, 0xfa, 0x98, 0xf5, 0x79,
+ 0x6d, 0x1e, 0x69, 0x4d, 0x57, 0xb1, 0xc8, 0x19,
+ 0x1b, 0xbd, 0x1e, 0x8c, 0x84, 0xb7, 0x7b, 0xe8,
+ 0xd2, 0x2d, 0x09, 0x41, 0x41, 0x37, 0x3d, 0xb1,
+ 0x6f, 0x26, 0x5d, 0x71, 0x16, 0x3d, 0xb7, 0x83,
+ 0x27, 0x2c, 0xa7, 0xb6, 0x50, 0xbd, 0x91, 0x86,
+ 0xab, 0x24, 0xa1, 0x38, 0xfd, 0xea, 0x71, 0x55,
+ 0x7e, 0x9a, 0x07, 0x77, 0x4b, 0xfa, 0x61, 0x66,
+ 0x20, 0x1e, 0x28, 0x95, 0x18, 0x1b, 0xa4, 0xa0,
+ 0xfd, 0xc0, 0x89, 0x72, 0x43, 0xd9, 0x3b, 0x49,
+ 0x5a, 0x3f, 0x9d, 0xbf, 0xdb, 0xb4, 0x46, 0xea,
+ 0x42, 0x01, 0x77, 0x23, 0x68, 0x95, 0xb6, 0x24,
+ 0xb3, 0xa8, 0x6c, 0x28, 0x3b, 0x11, 0x40, 0x7e,
+ 0x18, 0x65, 0x6d, 0xd8, 0x24, 0x42, 0x7d, 0x88,
+ 0xc0, 0x52, 0xd9, 0x05, 0xe4, 0x95, 0x90, 0x87,
+ 0x8c, 0xf4, 0xd0, 0x6b, 0xb9, 0x83, 0x99, 0x34,
+ 0x6d, 0xfe, 0x54, 0x40, 0x94, 0x52, 0x21, 0x4f,
+ 0x14, 0x25, 0xc5, 0xd6, 0x5e, 0x95, 0xdc, 0x0a,
+ 0x2b, 0x89, 0x20, 0x11, 0x84, 0x48, 0xd6, 0x3a,
+ 0xcd, 0x5c, 0x24, 0xad, 0x62, 0xe3, 0xb1, 0x93,
+ 0x25, 0x8d, 0xcd, 0x7e, 0xfc, 0x27, 0xa3, 0x37,
+ 0xfd, 0x84, 0xfc, 0x1b, 0xb2, 0xf1, 0x27, 0x38,
+ 0x5a, 0xb7, 0xfc, 0xf2, 0xfa, 0x95, 0x66, 0xd4,
+ 0xfb, 0xba, 0xa7, 0xd7, 0xa3, 0x72, 0x69, 0x48,
+ 0x48, 0x8c, 0xeb, 0x28, 0x89, 0xfe, 0x33, 0x65,
+ 0x5a, 0x36, 0x01, 0x7e, 0x06, 0x79, 0x0a, 0x09,
+ 0x3b, 0x74, 0x11, 0x9a, 0x6e, 0xbf, 0xd4, 0x9e,
+ 0x58, 0x90, 0x49, 0x4f, 0x4d, 0x08, 0xd4, 0xe5,
+ 0x4a, 0x09, 0x21, 0xef, 0x8b, 0xb8, 0x74, 0x3b,
+ 0x91, 0xdd, 0x36, 0x85, 0x60, 0x2d, 0xfa, 0xd4,
+ 0x45, 0x7b, 0x45, 0x53, 0xf5, 0x47, 0x87, 0x7e,
+ 0xa6, 0x37, 0xc8, 0x78, 0x7a, 0x68, 0x9d, 0x8d,
+ 0x65, 0x2c, 0x0e, 0x91, 0x5c, 0xa2, 0x60, 0xf0,
+ 0x8e, 0x3f, 0xe9, 0x1a, 0xcd, 0xaa, 0xe7, 0xd5,
+ 0x77, 0x18, 0xaf, 0xc9, 0xbc, 0x18, 0xea, 0x48,
+ 0x1b, 0xfb, 0x22, 0x48, 0x70, 0x16, 0x29, 0x9e,
+ 0x5b, 0xc1, 0x2c, 0x66, 0x23, 0xbc, 0xf0, 0x1f,
+ 0xef, 0xaf, 0xe4, 0xd6, 0x04, 0x19, 0x82, 0x7a,
+ 0x0b, 0xba, 0x4b, 0x46, 0xb1, 0x6a, 0x85, 0x5d,
+ 0xb4, 0x73, 0xd6, 0x21, 0xa1, 0x71, 0x60, 0x14,
+ 0xee, 0x0a, 0x77, 0xc4, 0x66, 0x2e, 0xf9, 0x69,
+ 0x30, 0xaf, 0x41, 0x0b, 0xc8, 0x83, 0x3c, 0x53,
+ 0x99, 0x19, 0x27, 0x46, 0xf7, 0x41, 0x6e, 0x56,
+ 0xdc, 0x94, 0x28, 0x67, 0x4e, 0xb7, 0x25, 0x48,
+ 0x8a, 0xc2, 0xe0, 0x60, 0x96, 0xcc, 0x18, 0xf4,
+ 0x84, 0xdd, 0xa7, 0x5e, 0x3e, 0x05, 0x0b, 0x26,
+ 0x26, 0xb2, 0x5c, 0x1f, 0x57, 0x1a, 0x04, 0x7e,
+ 0x6a, 0xe3, 0x2f, 0xb4, 0x35, 0xb6, 0x38, 0x40,
+ 0x40, 0xcd, 0x6f, 0x87, 0x2e, 0xef, 0xa3, 0xd7,
+ 0xa9, 0xc2, 0xe8, 0x0d, 0x27, 0xdf, 0x44, 0x62,
+ 0x99, 0xa0, 0xfc, 0xcf, 0x81, 0x78, 0xcb, 0xfe,
+ 0xe5, 0xa0, 0x03, 0x4e, 0x6c, 0xd7, 0xf4, 0xaf,
+ 0x7a, 0xbb, 0x61, 0x82, 0xfe, 0x71, 0x89, 0xb2,
+ 0x22, 0x7c, 0x8e, 0x83, 0x04, 0xce, 0xf6, 0x5d,
+ 0x84, 0x8f, 0x95, 0x6a, 0x7f, 0xad, 0xfd, 0x32,
+ 0x9c, 0x5e, 0xe4, 0x9c, 0x89, 0x60, 0x54, 0xaa,
+ 0x96, 0x72, 0xd2, 0xd7, 0x36, 0x85, 0xa9, 0x45,
+ 0xd2, 0x2a, 0xa1, 0x81, 0x49, 0x6f, 0x7e, 0x04,
+ 0xfa, 0xe2, 0xfe, 0x90, 0x26, 0x77, 0x5a, 0x33,
+ 0xb8, 0x04, 0x9a, 0x7a, 0xe6, 0x4c, 0x4f, 0xad,
+ 0x72, 0x96, 0x08, 0x28, 0x58, 0x13, 0xf8, 0xc4,
+ 0x1c, 0xf0, 0xc3, 0x45, 0x95, 0x49, 0x20, 0x8c,
+ 0x9f, 0x39, 0x70, 0xe1, 0x77, 0xfe, 0xd5, 0x4b,
+ 0xaf, 0x86, 0xda, 0xef, 0x22, 0x06, 0x83, 0x36,
+ 0x29, 0x12, 0x11, 0x40, 0xbc, 0x3b, 0x86, 0xaa,
+ 0xaa, 0x65, 0x60, 0xc3, 0x80, 0xca, 0xed, 0xa9,
+ 0xf3, 0xb0, 0x79, 0x96, 0xa2, 0x55, 0x27, 0x28,
+ 0x55, 0x73, 0x26, 0xa5, 0x50, 0xea, 0x92, 0x4b,
+ 0x3c, 0x5c, 0x82, 0x33, 0xf0, 0x01, 0x3f, 0x03,
+ 0xc1, 0x08, 0x05, 0xbf, 0x98, 0xf4, 0x9b, 0x6d,
+ 0xa5, 0xa8, 0xb4, 0x82, 0x0c, 0x06, 0xfa, 0xff,
+ 0x2d, 0x08, 0xf3, 0x05, 0x4f, 0x57, 0x2a, 0x39,
+ 0xd4, 0x83, 0x0d, 0x75, 0x51, 0xd8, 0x5b, 0x1b,
+ 0xd3, 0x51, 0x5a, 0x32, 0x2a, 0x9b, 0x32, 0xb2,
+ 0xf2, 0xa4, 0x96, 0x12, 0xf2, 0xae, 0x40, 0x34,
+ 0x67, 0xa8, 0xf5, 0x44, 0xd5, 0x35, 0x53, 0xfe,
+ 0xa3, 0x60, 0x96, 0x63, 0x0f, 0x1f, 0x6e, 0xb0,
+ 0x5a, 0x42, 0xa6, 0xfc, 0x51, 0x0b, 0x60, 0x27,
+ 0xbc, 0x06, 0x71, 0xed, 0x65, 0x5b, 0x23, 0x86,
+ 0x4a, 0x07, 0x3b, 0x22, 0x07, 0x46, 0xe6, 0x90,
+ 0x3e, 0xf3, 0x25, 0x50, 0x1b, 0x4c, 0x7f, 0x03,
+ 0x08, 0xa8, 0x36, 0x6b, 0x87, 0xe5, 0xe3, 0xdb,
+ 0x9a, 0x38, 0x83, 0xff, 0x9f, 0x1a, 0x9f, 0x57,
+ 0xa4, 0x2a, 0xf6, 0x37, 0xbc, 0x1a, 0xff, 0xc9,
+ 0x1e, 0x35, 0x0c, 0xc3, 0x7c, 0xa3, 0xb2, 0xe5,
+ 0xd2, 0xc6, 0xb4, 0x57, 0x47, 0xe4, 0x32, 0x16,
+ 0x6d, 0xa9, 0xae, 0x64, 0xe6, 0x2d, 0x8d, 0xc5,
+ 0x8d, 0x50, 0x8e, 0xe8, 0x1a, 0x22, 0x34, 0x2a,
+ 0xd9, 0xeb, 0x51, 0x90, 0x4a, 0xb1, 0x41, 0x7d,
+ 0x64, 0xf9, 0xb9, 0x0d, 0xf6, 0x23, 0x33, 0xb0,
+ 0x33, 0xf4, 0xf7, 0x3f, 0x27, 0x84, 0xc6, 0x0f,
+ 0x54, 0xa5, 0xc0, 0x2e, 0xec, 0x0b, 0x3a, 0x48,
+ 0x6e, 0x80, 0x35, 0x81, 0x43, 0x9b, 0x90, 0xb1,
+ 0xd0, 0x2b, 0xea, 0x21, 0xdc, 0xda, 0x5b, 0x09,
+ 0xf4, 0xcc, 0x10, 0xb4, 0xc7, 0xfe, 0x79, 0x51,
+ 0xc3, 0xc5, 0xac, 0x88, 0x74, 0x84, 0x0b, 0x4b,
+ 0xca, 0x79, 0x16, 0x29, 0xfb, 0x69, 0x54, 0xdf,
+ 0x41, 0x7e, 0xe9, 0xc7, 0x8e, 0xea, 0xa5, 0xfe,
+ 0xfc, 0x76, 0x0e, 0x90, 0xc4, 0x92, 0x38, 0xad,
+ 0x7b, 0x48, 0xe6, 0x6e, 0xf7, 0x21, 0xfd, 0x4e,
+ 0x93, 0x0a, 0x7b, 0x41, 0x83, 0x68, 0xfb, 0x57,
+ 0x51, 0x76, 0x34, 0xa9, 0x6c, 0x00, 0xaa, 0x4f,
+ 0x66, 0x65, 0x98, 0x4a, 0x4f, 0xa3, 0xa0, 0xef,
+ 0x69, 0x3f, 0xe3, 0x1c, 0x92, 0x8c, 0xfd, 0xd8,
+ 0xe8, 0xde, 0x7c, 0x7f, 0x3e, 0x84, 0x8e, 0x69,
+ 0x3c, 0xf1, 0xf2, 0x05, 0x46, 0xdc, 0x2f, 0x9d,
+ 0x5e, 0x6e, 0x4c, 0xfb, 0xb5, 0x99, 0x2a, 0x59,
+ 0x63, 0xc1, 0x34, 0xbc, 0x57, 0xc0, 0x0d, 0xb9,
+ 0x61, 0x25, 0xf3, 0x33, 0x23, 0x51, 0xb6, 0x0d,
+ 0x07, 0xa6, 0xab, 0x94, 0x4a, 0xb7, 0x2a, 0xea,
+ 0xee, 0xac, 0xa3, 0xc3, 0x04, 0x8b, 0x0e, 0x56,
+ 0xfe, 0x44, 0xa7, 0x39, 0xe2, 0xed, 0xed, 0xb4,
+ 0x22, 0x2b, 0xac, 0x12, 0x32, 0x28, 0x91, 0xd8,
+ 0xa5, 0xab, 0xff, 0x5f, 0xe0, 0x4b, 0xda, 0x78,
+ 0x17, 0xda, 0xf1, 0x01, 0x5b, 0xcd, 0xe2, 0x5f,
+ 0x50, 0x45, 0x73, 0x2b, 0xe4, 0x76, 0x77, 0xf4,
+ 0x64, 0x1d, 0x43, 0xfb, 0x84, 0x7a, 0xea, 0x91,
+ 0xae, 0xf9, 0x9e, 0xb7, 0xb4, 0xb0, 0x91, 0x5f,
+ 0x16, 0x35, 0x9a, 0x11, 0xb8, 0xc7, 0xc1, 0x8c,
+ 0xc6, 0x10, 0x8d, 0x2f, 0x63, 0x4a, 0xa7, 0x57,
+ 0x3a, 0x51, 0xd6, 0x32, 0x2d, 0x64, 0x72, 0xd4,
+ 0x66, 0xdc, 0x10, 0xa6, 0x67, 0xd6, 0x04, 0x23,
+ 0x9d, 0x0a, 0x11, 0x77, 0xdd, 0x37, 0x94, 0x17,
+ 0x3c, 0xbf, 0x8b, 0x65, 0xb0, 0x2e, 0x5e, 0x66,
+ 0x47, 0x64, 0xac, 0xdd, 0xf0, 0x84, 0xfd, 0x39,
+ 0xfa, 0x15, 0x5d, 0xef, 0xae, 0xca, 0xc1, 0x36,
+ 0xa7, 0x5c, 0xbf, 0xc7, 0x08, 0xc2, 0x66, 0x00,
+ 0x74, 0x74, 0x4e, 0x27, 0x3f, 0x55, 0x8a, 0xb7,
+ 0x38, 0x66, 0x83, 0x6d, 0xcf, 0x99, 0x9e, 0x60,
+ 0x8f, 0xdd, 0x2e, 0x62, 0x22, 0x0e, 0xef, 0x0c,
+ 0x98, 0xa7, 0x85, 0x74, 0x3b, 0x9d, 0xec, 0x9e,
+ 0xa9, 0x19, 0x72, 0xa5, 0x7f, 0x2c, 0x39, 0xb7,
+ 0x7d, 0xb7, 0xf1, 0x12, 0x65, 0x27, 0x4b, 0x5a,
+ 0xde, 0x17, 0xfe, 0xad, 0x44, 0xf3, 0x20, 0x4d,
+ 0xfd, 0xe4, 0x1f, 0xb5, 0x81, 0xb0, 0x36, 0x37,
+ 0x08, 0x6f, 0xc3, 0x0c, 0xe9, 0x85, 0x98, 0x82,
+ 0xa9, 0x62, 0x0c, 0xc4, 0x97, 0xc0, 0x50, 0xc8,
+ 0xa7, 0x3c, 0x50, 0x9f, 0x43, 0xb9, 0xcd, 0x5e,
+ 0x4d, 0xfa, 0x1c, 0x4b, 0x0b, 0xa9, 0x98, 0x85,
+ 0x38, 0x92, 0xac, 0x8d, 0xe4, 0xad, 0x9b, 0x98,
+ 0xab, 0xd9, 0x38, 0xac, 0x62, 0x52, 0xa3, 0x22,
+ 0x63, 0x0f, 0xbf, 0x95, 0x48, 0xdf, 0x69, 0xe7,
+ 0x8b, 0x33, 0xd5, 0xb2, 0xbd, 0x05, 0x49, 0x49,
+ 0x9d, 0x57, 0x73, 0x19, 0x33, 0xae, 0xfa, 0x33,
+ 0xf1, 0x19, 0xa8, 0x80, 0xce, 0x04, 0x9f, 0xbc,
+ 0x1d, 0x65, 0x82, 0x1b, 0xe5, 0x3a, 0x51, 0xc8,
+ 0x1c, 0x21, 0xe3, 0x5d, 0xf3, 0x7d, 0x9b, 0x2f,
+ 0x2c, 0x1d, 0x4a, 0x7f, 0x9b, 0x68, 0x35, 0xa3,
+ 0xb2, 0x50, 0xf7, 0x62, 0x79, 0xcd, 0xf4, 0x98,
+ 0x4f, 0xe5, 0x63, 0x7c, 0x3e, 0x45, 0x31, 0x8c,
+ 0x16, 0xa0, 0x12, 0xc8, 0x58, 0xce, 0x39, 0xa6,
+ 0xbc, 0x54, 0xdb, 0xc5, 0xe0, 0xd5, 0xba, 0xbc,
+ 0xb9, 0x04, 0xf4, 0x8d, 0xe8, 0x2f, 0x15, 0x9d,
+};
+
+/* 100 test cases */
+static struct crc_test {
+ u32 crc; /* random starting crc */
+ u32 start; /* random 6 bit offset in buf */
+ u32 length; /* random 11 bit length of test */
+ u32 crc_le; /* expected crc32_le result */
+ u32 crc_be; /* expected crc32_be result */
+ u32 crc32c_le; /* expected crc32c_le result */
+} test[] =
+{
+ {0x674bf11d, 0x00000038, 0x00000542, 0x0af6d466, 0xd8b6e4c1,
+ 0xf6e93d6c},
+ {0x35c672c6, 0x0000003a, 0x000001aa, 0xc6d3dfba, 0x28aaf3ad,
+ 0x0fe92aca},
+ {0x496da28e, 0x00000039, 0x000005af, 0xd933660f, 0x5d57e81f,
+ 0x52e1ebb8},
+ {0x09a9b90e, 0x00000027, 0x000001f8, 0xb45fe007, 0xf45fca9a,
+ 0x0798af9a},
+ {0xdc97e5a9, 0x00000025, 0x000003b6, 0xf81a3562, 0xe0126ba2,
+ 0x18eb3152},
+ {0x47c58900, 0x0000000a, 0x000000b9, 0x8e58eccf, 0xf3afc793,
+ 0xd00d08c7},
+ {0x292561e8, 0x0000000c, 0x00000403, 0xa2ba8aaf, 0x0b797aed,
+ 0x8ba966bc},
+ {0x415037f6, 0x00000003, 0x00000676, 0xa17d52e8, 0x7f0fdf35,
+ 0x11d694a2},
+ {0x3466e707, 0x00000026, 0x00000042, 0x258319be, 0x75c484a2,
+ 0x6ab3208d},
+ {0xafd1281b, 0x00000023, 0x000002ee, 0x4428eaf8, 0x06c7ad10,
+ 0xba4603c5},
+ {0xd3857b18, 0x00000028, 0x000004a2, 0x5c430821, 0xb062b7cb,
+ 0xe6071c6f},
+ {0x1d825a8f, 0x0000002b, 0x0000050b, 0xd2c45f0c, 0xd68634e0,
+ 0x179ec30a},
+ {0x5033e3bc, 0x0000000b, 0x00000078, 0xa3ea4113, 0xac6d31fb,
+ 0x0903beb8},
+ {0x94f1fb5e, 0x0000000f, 0x000003a2, 0xfbfc50b1, 0x3cfe50ed,
+ 0x6a7cb4fa},
+ {0xc9a0fe14, 0x00000009, 0x00000473, 0x5fb61894, 0x87070591,
+ 0xdb535801},
+ {0x88a034b1, 0x0000001c, 0x000005ad, 0xc1b16053, 0x46f95c67,
+ 0x92bed597},
+ {0xf0f72239, 0x00000020, 0x0000026d, 0xa6fa58f3, 0xf8c2c1dd,
+ 0x192a3f1b},
+ {0xcc20a5e3, 0x0000003b, 0x0000067a, 0x7740185a, 0x308b979a,
+ 0xccbaec1a},
+ {0xce589c95, 0x0000002b, 0x00000641, 0xd055e987, 0x40aae25b,
+ 0x7eabae4d},
+ {0x78edc885, 0x00000035, 0x000005be, 0xa39cb14b, 0x035b0d1f,
+ 0x28c72982},
+ {0x9d40a377, 0x0000003b, 0x00000038, 0x1f47ccd2, 0x197fbc9d,
+ 0xc3cd4d18},
+ {0x703d0e01, 0x0000003c, 0x000006f1, 0x88735e7c, 0xfed57c5a,
+ 0xbca8f0e7},
+ {0x776bf505, 0x0000000f, 0x000005b2, 0x5cc4fc01, 0xf32efb97,
+ 0x713f60b3},
+ {0x4a3e7854, 0x00000027, 0x000004b8, 0x8d923c82, 0x0cbfb4a2,
+ 0xebd08fd5},
+ {0x209172dd, 0x0000003b, 0x00000356, 0xb89e9c2b, 0xd7868138,
+ 0x64406c59},
+ {0x3ba4cc5b, 0x0000002f, 0x00000203, 0xe51601a9, 0x5b2a1032,
+ 0x7421890e},
+ {0xfc62f297, 0x00000000, 0x00000079, 0x71a8e1a2, 0x5d88685f,
+ 0xe9347603},
+ {0x64280b8b, 0x00000016, 0x000007ab, 0x0fa7a30c, 0xda3a455f,
+ 0x1bef9060},
+ {0x97dd724b, 0x00000033, 0x000007ad, 0x5788b2f4, 0xd7326d32,
+ 0x34720072},
+ {0x61394b52, 0x00000035, 0x00000571, 0xc66525f1, 0xcabe7fef,
+ 0x48310f59},
+ {0x29b4faff, 0x00000024, 0x0000006e, 0xca13751e, 0x993648e0,
+ 0x783a4213},
+ {0x29bfb1dc, 0x0000000b, 0x00000244, 0x436c43f7, 0x429f7a59,
+ 0x9e8efd41},
+ {0x86ae934b, 0x00000035, 0x00000104, 0x0760ec93, 0x9cf7d0f4,
+ 0xfc3d34a5},
+ {0xc4c1024e, 0x0000002e, 0x000006b1, 0x6516a3ec, 0x19321f9c,
+ 0x17a52ae2},
+ {0x3287a80a, 0x00000026, 0x00000496, 0x0b257eb1, 0x754ebd51,
+ 0x886d935a},
+ {0xa4db423e, 0x00000023, 0x0000045d, 0x9b3a66dc, 0x873e9f11,
+ 0xeaaeaeb2},
+ {0x7a1078df, 0x00000015, 0x0000014a, 0x8c2484c5, 0x6a628659,
+ 0x8e900a4b},
+ {0x6048bd5b, 0x00000006, 0x0000006a, 0x897e3559, 0xac9961af,
+ 0xd74662b1},
+ {0xd8f9ea20, 0x0000003d, 0x00000277, 0x60eb905b, 0xed2aaf99,
+ 0xd26752ba},
+ {0xea5ec3b4, 0x0000002a, 0x000004fe, 0x869965dc, 0x6c1f833b,
+ 0x8b1fcd62},
+ {0x2dfb005d, 0x00000016, 0x00000345, 0x6a3b117e, 0xf05e8521,
+ 0xf54342fe},
+ {0x5a214ade, 0x00000020, 0x000005b6, 0x467f70be, 0xcb22ccd3,
+ 0x5b95b988},
+ {0xf0ab9cca, 0x00000032, 0x00000515, 0xed223df3, 0x7f3ef01d,
+ 0x2e1176be},
+ {0x91b444f9, 0x0000002e, 0x000007f8, 0x84e9a983, 0x5676756f,
+ 0x66120546},
+ {0x1b5d2ddb, 0x0000002e, 0x0000012c, 0xba638c4c, 0x3f42047b,
+ 0xf256a5cc},
+ {0xd824d1bb, 0x0000003a, 0x000007b5, 0x6288653b, 0x3a3ebea0,
+ 0x4af1dd69},
+ {0x0470180c, 0x00000034, 0x000001f0, 0x9d5b80d6, 0x3de08195,
+ 0x56f0a04a},
+ {0xffaa3a3f, 0x00000036, 0x00000299, 0xf3a82ab8, 0x53e0c13d,
+ 0x74f6b6b2},
+ {0x6406cfeb, 0x00000023, 0x00000600, 0xa920b8e8, 0xe4e2acf4,
+ 0x085951fd},
+ {0xb24aaa38, 0x0000003e, 0x000004a1, 0x657cc328, 0x5077b2c3,
+ 0xc65387eb},
+ {0x58b2ab7c, 0x00000039, 0x000002b4, 0x3a17ee7e, 0x9dcb3643,
+ 0x1ca9257b},
+ {0x3db85970, 0x00000006, 0x000002b6, 0x95268b59, 0xb9812c10,
+ 0xfd196d76},
+ {0x857830c5, 0x00000003, 0x00000590, 0x4ef439d5, 0xf042161d,
+ 0x5ef88339},
+ {0xe1fcd978, 0x0000003e, 0x000007d8, 0xae8d8699, 0xce0a1ef5,
+ 0x2c3714d9},
+ {0xb982a768, 0x00000016, 0x000006e0, 0x62fad3df, 0x5f8a067b,
+ 0x58576548},
+ {0x1d581ce8, 0x0000001e, 0x0000058b, 0xf0f5da53, 0x26e39eee,
+ 0xfd7c57de},
+ {0x2456719b, 0x00000025, 0x00000503, 0x4296ac64, 0xd50e4c14,
+ 0xd5fedd59},
+ {0xfae6d8f2, 0x00000000, 0x0000055d, 0x057fdf2e, 0x2a31391a,
+ 0x1cc3b17b},
+ {0xcba828e3, 0x00000039, 0x000002ce, 0xe3f22351, 0x8f00877b,
+ 0x270eed73},
+ {0x13d25952, 0x0000000a, 0x0000072d, 0x76d4b4cc, 0x5eb67ec3,
+ 0x91ecbb11},
+ {0x0342be3f, 0x00000015, 0x00000599, 0xec75d9f1, 0x9d4d2826,
+ 0x05ed8d0c},
+ {0xeaa344e0, 0x00000014, 0x000004d8, 0x72a4c981, 0x2064ea06,
+ 0x0b09ad5b},
+ {0xbbb52021, 0x0000003b, 0x00000272, 0x04af99fc, 0xaf042d35,
+ 0xf8d511fb},
+ {0xb66384dc, 0x0000001d, 0x000007fc, 0xd7629116, 0x782bd801,
+ 0x5ad832cc},
+ {0x616c01b6, 0x00000022, 0x000002c8, 0x5b1dab30, 0x783ce7d2,
+ 0x1214d196},
+ {0xce2bdaad, 0x00000016, 0x0000062a, 0x932535c8, 0x3f02926d,
+ 0x5747218a},
+ {0x00fe84d7, 0x00000005, 0x00000205, 0x850e50aa, 0x753d649c,
+ 0xde8f14de},
+ {0xbebdcb4c, 0x00000006, 0x0000055d, 0xbeaa37a2, 0x2d8c9eba,
+ 0x3563b7b9},
+ {0xd8b1a02a, 0x00000010, 0x00000387, 0x5017d2fc, 0x503541a5,
+ 0x071475d0},
+ {0x3b96cad2, 0x00000036, 0x00000347, 0x1d2372ae, 0x926cd90b,
+ 0x54c79d60},
+ {0xc94c1ed7, 0x00000005, 0x0000038b, 0x9e9fdb22, 0x144a9178,
+ 0x4c53eee6},
+ {0x1aad454e, 0x00000025, 0x000002b2, 0xc3f6315c, 0x5c7a35b3,
+ 0x10137a3c},
+ {0xa4fec9a6, 0x00000000, 0x000006d6, 0x90be5080, 0xa4107605,
+ 0xaa9d6c73},
+ {0x1bbe71e2, 0x0000001f, 0x000002fd, 0x4e504c3b, 0x284ccaf1,
+ 0xb63d23e7},
+ {0x4201c7e4, 0x00000002, 0x000002b7, 0x7822e3f9, 0x0cc912a9,
+ 0x7f53e9cf},
+ {0x23fddc96, 0x00000003, 0x00000627, 0x8a385125, 0x07767e78,
+ 0x13c1cd83},
+ {0xd82ba25c, 0x00000016, 0x0000063e, 0x98e4148a, 0x283330c9,
+ 0x49ff5867},
+ {0x786f2032, 0x0000002d, 0x0000060f, 0xf201600a, 0xf561bfcd,
+ 0x8467f211},
+ {0xfebe4e1f, 0x0000002a, 0x000004f2, 0x95e51961, 0xfd80dcab,
+ 0x3f9683b2},
+ {0x1a6e0a39, 0x00000008, 0x00000672, 0x8af6c2a5, 0x78dd84cb,
+ 0x76a3f874},
+ {0x56000ab8, 0x0000000e, 0x000000e5, 0x36bacb8f, 0x22ee1f77,
+ 0x863b702f},
+ {0x4717fe0c, 0x00000000, 0x000006ec, 0x8439f342, 0x5c8e03da,
+ 0xdc6c58ff},
+ {0xd5d5d68e, 0x0000003c, 0x000003a3, 0x46fff083, 0x177d1b39,
+ 0x0622cc95},
+ {0xc25dd6c6, 0x00000024, 0x000006c0, 0x5ceb8eb4, 0x892b0d16,
+ 0xe85605cd},
+ {0xe9b11300, 0x00000023, 0x00000683, 0x07a5d59a, 0x6c6a3208,
+ 0x31da5f06},
+ {0x95cd285e, 0x00000001, 0x00000047, 0x7b3a4368, 0x0202c07e,
+ 0xa1f2e784},
+ {0xd9245a25, 0x0000001e, 0x000003a6, 0xd33c1841, 0x1936c0d5,
+ 0xb07cc616},
+ {0x103279db, 0x00000006, 0x0000039b, 0xca09b8a0, 0x77d62892,
+ 0xbf943b6c},
+ {0x1cba3172, 0x00000027, 0x000001c8, 0xcb377194, 0xebe682db,
+ 0x2c01af1c},
+ {0x8f613739, 0x0000000c, 0x000001df, 0xb4b0bc87, 0x7710bd43,
+ 0x0fe5f56d},
+ {0x1c6aa90d, 0x0000001b, 0x0000053c, 0x70559245, 0xda7894ac,
+ 0xf8943b2d},
+ {0xaabe5b93, 0x0000003d, 0x00000715, 0xcdbf42fa, 0x0c3b99e7,
+ 0xe4d89272},
+ {0xf15dd038, 0x00000006, 0x000006db, 0x6e104aea, 0x8d5967f2,
+ 0x7c2f6bbb},
+ {0x584dd49c, 0x00000020, 0x000007bc, 0x36b6cfd6, 0xad4e23b2,
+ 0xabbf388b},
+ {0x5d8c9506, 0x00000020, 0x00000470, 0x4c62378e, 0x31d92640,
+ 0x1dca1f4e},
+ {0xb80d17b0, 0x00000032, 0x00000346, 0x22a5bb88, 0x9a7ec89f,
+ 0x5c170e23},
+ {0xdaf0592e, 0x00000023, 0x000007b0, 0x3cab3f99, 0x9b1fdd99,
+ 0xc0e9d672},
+ {0x4793cc85, 0x0000000d, 0x00000706, 0xe82e04f6, 0xed3db6b7,
+ 0xc18bdc86},
+ {0x82ebf64e, 0x00000009, 0x000007c3, 0x69d590a9, 0x9efa8499,
+ 0xa874fcdd},
+ {0xb18a0319, 0x00000026, 0x000007db, 0x1cf98dcc, 0x8fa9ad6a,
+ 0x9dc0bb48},
+};
+
+static int crc32c_test(void)
+{
+ int i;
+ int errors = 0;
+ int bytes = 0;
+ struct timeval start, stop;
+ uint64_t usec;
+
+ /* keep static to prevent cache warming code from
+ * getting eliminated by the compiler */
+ static u32 crc;
+
+ /* pre-warm the cache */
+ for (i = 0; i < 100; i++) {
+ bytes += 2*test[i].length;
+
+ crc ^= crc32c_le(test[i].crc, test_buf +
+ test[i].start, test[i].length);
+ }
+
+ gettimeofday(&start, NULL);
+ for (i = 0; i < 100; i++) {
+ if (test[i].crc32c_le != crc32c_le(test[i].crc, test_buf +
+ test[i].start, test[i].length))
+ errors++;
+ }
+ gettimeofday(&stop, NULL);
+
+ usec = stop.tv_usec - start.tv_usec +
+ 1000000 * (stop.tv_sec - start.tv_sec);
+
+ if (errors)
+ printf("crc32c: %d self tests failed\n", errors);
+ else {
+ printf("crc32c: tests passed, %d bytes in %" PRIu64 " usec\n",
+ bytes, usec);
+ }
+
+ return errors;
+}
+
+static int crc32_test(void)
+{
+ int i;
+ int errors = 0;
+ int bytes = 0;
+ struct timeval start, stop;
+ uint64_t usec;
+
+ /* keep static to prevent cache warming code from
+ * getting eliminated by the compiler */
+ static u32 crc;
+
+ /* pre-warm the cache */
+ for (i = 0; i < 100; i++) {
+ bytes += 2*test[i].length;
+
+ crc ^= crc32_le(test[i].crc, test_buf +
+ test[i].start, test[i].length);
+
+#if 0 /* not used */
+ crc ^= crc32_be(test[i].crc, test_buf +
+ test[i].start, test[i].length);
+#endif
+ }
+
+ gettimeofday(&start, NULL);
+ for (i = 0; i < 100; i++) {
+ if (test[i].crc_le != crc32_le(test[i].crc, test_buf +
+ test[i].start, test[i].length))
+ errors++;
+
+#if 0 /* not used */
+ if (test[i].crc_be != crc32_be(test[i].crc, test_buf +
+ test[i].start, test[i].length))
+ errors++;
+#endif
+ }
+ gettimeofday(&stop, NULL);
+
+ usec = stop.tv_usec - start.tv_usec +
+ 1000000000 * (stop.tv_sec - start.tv_sec);
+
+ if (errors)
+ printf("crc32: %d self tests failed\n", errors);
+ else {
+ printf("crc32: tests passed, %d bytes in %" PRIu64 " usec\n",
+ bytes, usec);
+ }
+
+ return errors;
+}
+/*
+ * make sure we always return 0 for a successful test run, and non-zero for a
+ * failed run. The build infrastructure is looking for this information to
+ * determine whether to allow the build to proceed.
+ */
+int main(int argc, char **argv)
+{
+ int errors;
+
+ printf("CRC_LE_BITS = %d\n", CRC_LE_BITS);
+
+ errors = crc32_test();
+ errors += crc32c_test();
+
+ return errors != 0;
+}
+#endif /* CRC32_SELFTEST */
diff --git a/libxfs/crc32defs.h b/libxfs/crc32defs.h
new file mode 100644
index 0000000..64cba2c
--- /dev/null
+++ b/libxfs/crc32defs.h
@@ -0,0 +1,72 @@
+/*
+ * There are multiple 16-bit CRC polynomials in common use, but this is
+ * *the* standard CRC-32 polynomial, first popularized by Ethernet.
+ * x^32+x^26+x^23+x^22+x^16+x^12+x^11+x^10+x^8+x^7+x^5+x^4+x^2+x^1+x^0
+ */
+#define CRCPOLY_LE 0xedb88320
+#define CRCPOLY_BE 0x04c11db7
+
+/*
+ * This is the CRC32c polynomial, as outlined by Castagnoli.
+ * x^32+x^28+x^27+x^26+x^25+x^23+x^22+x^20+x^19+x^18+x^14+x^13+x^11+x^10+x^9+
+ * x^8+x^6+x^0
+ */
+#define CRC32C_POLY_LE 0x82F63B78
+
+/* Try to choose an implementation variant via Kconfig */
+#ifdef CONFIG_CRC32_SLICEBY8
+# define CRC_LE_BITS 64
+# define CRC_BE_BITS 64
+#endif
+#ifdef CONFIG_CRC32_SLICEBY4
+# define CRC_LE_BITS 32
+# define CRC_BE_BITS 32
+#endif
+#ifdef CONFIG_CRC32_SARWATE
+# define CRC_LE_BITS 8
+# define CRC_BE_BITS 8
+#endif
+#ifdef CONFIG_CRC32_BIT
+# define CRC_LE_BITS 1
+# define CRC_BE_BITS 1
+#endif
+
+/*
+ * How many bits at a time to use. Valid values are 1, 2, 4, 8, 32 and 64.
+ * For less performance-sensitive, use 4 or 8 to save table size.
+ * For larger systems choose same as CPU architecture as default.
+ * This works well on X86_64, SPARC64 systems. This may require some
+ * elaboration after experiments with other architectures.
+ */
+#ifndef CRC_LE_BITS
+# ifdef CONFIG_64BIT
+# define CRC_LE_BITS 64
+# else
+# define CRC_LE_BITS 32
+# endif
+#endif
+#ifndef CRC_BE_BITS
+# ifdef CONFIG_64BIT
+# define CRC_BE_BITS 64
+# else
+# define CRC_BE_BITS 32
+# endif
+#endif
+
+/*
+ * Little-endian CRC computation. Used with serial bit streams sent
+ * lsbit-first. Be sure to use cpu_to_le32() to append the computed CRC.
+ */
+#if CRC_LE_BITS > 64 || CRC_LE_BITS < 1 || CRC_LE_BITS == 16 || \
+ CRC_LE_BITS & CRC_LE_BITS-1
+# error "CRC_LE_BITS must be one of {1, 2, 4, 8, 32, 64}"
+#endif
+
+/*
+ * Big-endian CRC computation. Used with serial bit streams sent
+ * msbit-first. Be sure to use cpu_to_be32() to append the computed CRC.
+ */
+#if CRC_BE_BITS > 64 || CRC_BE_BITS < 1 || CRC_BE_BITS == 16 || \
+ CRC_BE_BITS & CRC_BE_BITS-1
+# error "CRC_BE_BITS must be one of {1, 2, 4, 8, 32, 64}"
+#endif
diff --git a/libxfs/gen_crc32table.c b/libxfs/gen_crc32table.c
new file mode 100644
index 0000000..574a2d1
--- /dev/null
+++ b/libxfs/gen_crc32table.c
@@ -0,0 +1,144 @@
+#include <stdio.h>
+#include "crc32defs.h"
+#include <inttypes.h>
+
+#define ENTRIES_PER_LINE 4
+
+#if CRC_LE_BITS > 8
+# define LE_TABLE_ROWS (CRC_LE_BITS/8)
+# define LE_TABLE_SIZE 256
+#else
+# define LE_TABLE_ROWS 1
+# define LE_TABLE_SIZE (1 << CRC_LE_BITS)
+#endif
+
+#if CRC_BE_BITS > 8
+# define BE_TABLE_ROWS (CRC_BE_BITS/8)
+# define BE_TABLE_SIZE 256
+#else
+# define BE_TABLE_ROWS 1
+# define BE_TABLE_SIZE (1 << CRC_BE_BITS)
+#endif
+
+static uint32_t crc32table_le[LE_TABLE_ROWS][256];
+static uint32_t crc32ctable_le[LE_TABLE_ROWS][256];
+
+/*
+ * big endian ordered CRC not used by XFS.
+static uint32_t crc32table_be[BE_TABLE_ROWS][256];
+ */
+
+/**
+ * crc32init_le() - allocate and initialize LE table data
+ *
+ * crc is the crc of the byte i; other entries are filled in based on the
+ * fact that crctable[i^j] = crctable[i] ^ crctable[j].
+ *
+ */
+static void crc32init_le_generic(const uint32_t polynomial,
+ uint32_t (*tab)[256])
+{
+ unsigned i, j;
+ uint32_t crc = 1;
+
+ tab[0][0] = 0;
+
+ for (i = LE_TABLE_SIZE >> 1; i; i >>= 1) {
+ crc = (crc >> 1) ^ ((crc & 1) ? polynomial : 0);
+ for (j = 0; j < LE_TABLE_SIZE; j += 2 * i)
+ tab[0][i + j] = crc ^ tab[0][j];
+ }
+ for (i = 0; i < LE_TABLE_SIZE; i++) {
+ crc = tab[0][i];
+ for (j = 1; j < LE_TABLE_ROWS; j++) {
+ crc = tab[0][crc & 0xff] ^ (crc >> 8);
+ tab[j][i] = crc;
+ }
+ }
+}
+
+static void crc32init_le(void)
+{
+ crc32init_le_generic(CRCPOLY_LE, crc32table_le);
+}
+
+static void crc32cinit_le(void)
+{
+ crc32init_le_generic(CRC32C_POLY_LE, crc32ctable_le);
+}
+
+/**
+ * crc32init_be() - allocate and initialize BE table data
+ */
+#if 0 /* not used */
+static void crc32init_be(void)
+{
+ unsigned i, j;
+ uint32_t crc = 0x80000000;
+
+ crc32table_be[0][0] = 0;
+
+ for (i = 1; i < BE_TABLE_SIZE; i <<= 1) {
+ crc = (crc << 1) ^ ((crc & 0x80000000) ? CRCPOLY_BE : 0);
+ for (j = 0; j < i; j++)
+ crc32table_be[0][i + j] = crc ^ crc32table_be[0][j];
+ }
+ for (i = 0; i < BE_TABLE_SIZE; i++) {
+ crc = crc32table_be[0][i];
+ for (j = 1; j < BE_TABLE_ROWS; j++) {
+ crc = crc32table_be[0][(crc >> 24) & 0xff] ^ (crc << 8);
+ crc32table_be[j][i] = crc;
+ }
+ }
+}
+#endif
+
+static void output_table(uint32_t (*table)[256], int rows, int len, char *trans)
+{
+ int i, j;
+
+ for (j = 0 ; j < rows; j++) {
+ printf("{");
+ for (i = 0; i < len - 1; i++) {
+ if (i % ENTRIES_PER_LINE == 0)
+ printf("\n");
+ printf("%s(0x%8.8xL), ", trans, table[j][i]);
+ }
+ printf("%s(0x%8.8xL)},\n", trans, table[j][len - 1]);
+ }
+}
+
+int main(int argc, char** argv)
+{
+ printf("/* this file is generated - do not edit */\n\n");
+
+ if (CRC_LE_BITS > 1) {
+ crc32init_le();
+ printf("static u32 crc32table_le[%d][%d] = {",
+ LE_TABLE_ROWS, LE_TABLE_SIZE);
+ output_table(crc32table_le, LE_TABLE_ROWS,
+ LE_TABLE_SIZE, "tole");
+ printf("};\n");
+ }
+
+#if 0 /* not used by xfsprogs */
+ if (CRC_BE_BITS > 1) {
+ crc32init_be();
+ printf("static u32 crc32table_be[%d][%d] = {",
+ BE_TABLE_ROWS, BE_TABLE_SIZE);
+ output_table(crc32table_be, LE_TABLE_ROWS,
+ BE_TABLE_SIZE, "tobe");
+ printf("};\n");
+ }
+#endif
+ if (CRC_LE_BITS > 1) {
+ crc32cinit_le();
+ printf("static u32 crc32ctable_le[%d][%d] = {",
+ LE_TABLE_ROWS, LE_TABLE_SIZE);
+ output_table(crc32ctable_le, LE_TABLE_ROWS,
+ LE_TABLE_SIZE, "tole");
+ printf("};\n");
+ }
+
+ return 0;
+}
xfsprogs-crc-patchset-v2/xfsprogs-libxlog-kernel-sync 0000600 0001750 0001750 00000066352 12132011546 021737 0 ustar dave dave xfsprogs: updata libxlog to current kernel code
From: Dave Chinner <dchinner@xxxxxxxxxx>
Update the log recovery code to match the current 3.8-rc2 kernel code.
Note: while this introduces CRC validation infrastructure, it is
currently short-circuited as we cannot validate the CRC from
userspace because we do not know what the size of the log buffer was
that wrote the header. This information is not written into the log
header anywhere, so we have no way of working out the correct number
of extra headers that need to be summed. This is also a problem for
the kernel code, and needs fixing.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
db/sb.c | 7 +-
include/libxlog.h | 32 ++---
libxlog/xfs_log_recover.c | 301 +++++++++++++++++++++++++++-----------------
logprint/log_copy.c | 2 +-
logprint/log_dump.c | 2 +-
logprint/log_misc.c | 4 +-
logprint/log_print_all.c | 2 +-
logprint/log_print_trans.c | 4 +-
logprint/logprint.c | 2 +-
logprint/logprint.h | 10 +-
repair/phase2.c | 7 +-
11 files changed, 229 insertions(+), 144 deletions(-)
diff --git a/db/sb.c b/db/sb.c
index 21f38c5..d83db9c 100644
--- a/db/sb.c
+++ b/db/sb.c
@@ -205,12 +205,15 @@ get_sb(xfs_agnumber_t agno, xfs_sb_t *sb)
}
/* workaround craziness in the xlog routines */
-int xlog_recover_do_trans(xlog_t *log, xlog_recover_t *t, int p) { return 0; }
+int xlog_recover_do_trans(struct xlog *log, xlog_recover_t *t, int p)
+{
+ return 0;
+}
int
sb_logcheck(void)
{
- xlog_t log;
+ struct xlog log;
xfs_daddr_t head_blk, tail_blk;
if (mp->m_sb.sb_logstart) {
diff --git a/include/libxlog.h b/include/libxlog.h
index d1142ab..36ede59 100644
--- a/include/libxlog.h
+++ b/include/libxlog.h
@@ -24,7 +24,7 @@
* xlog_t that we actually need to get our work done, avoiding
* the need to define any exotic kernel types in userland.
*/
-typedef struct log {
+struct xlog {
xfs_lsn_t l_tail_lsn; /* lsn of 1st LR w/ unflush buffers */
xfs_lsn_t l_last_sync_lsn;/* lsn of last LR on disk */
xfs_mount_t *l_mp; /* mount point */
@@ -45,7 +45,7 @@ typedef struct log {
uint l_sectbb_mask; /* sector size (in BBs)
* alignment mask */
int l_sectBBsize; /* size of log sector in 512 byte chunks */
-} xlog_t;
+};
#include <xfs/xfs_log_recover.h>
#include <xfs/xfs_buf_item.h>
@@ -76,6 +76,10 @@ typedef union {
#define unlikely(x) (x)
#define min(a,b) ((a) < (b) ? (a) : (b))
+#define xfs_warn(mp,fmt,args...) cmn_err(CE_WARN,fmt, ## args)
+#define xfs_alert(mp,fmt,args...) cmn_err(CE_ALERT,fmt, ## args)
+#define xfs_hex_dump(d,n) ((void) 0)
+
extern void xlog_warn(char *fmt,...);
extern void xlog_exit(char *fmt,...);
extern void xlog_panic(char *fmt,...);
@@ -88,34 +92,34 @@ extern int print_record_header;
/* libxfs parameters */
extern libxfs_init_t x;
-extern struct xfs_buf *xlog_get_bp(xlog_t *, int);
+extern struct xfs_buf *xlog_get_bp(struct xlog *, int);
extern void xlog_put_bp(struct xfs_buf *);
-extern int xlog_bread(xlog_t *log, xfs_daddr_t blk_no, int nbblks,
+extern int xlog_bread(struct xlog *log, xfs_daddr_t blk_no, int nbblks,
xfs_buf_t *bp, xfs_caddr_t *offset);
-extern int xlog_bread_noalign(xlog_t *log, xfs_daddr_t blk_no, int nbblks,
- xfs_buf_t *bp);
+extern int xlog_bread_noalign(struct xlog *log, xfs_daddr_t blk_no,
+ int nbblks, xfs_buf_t *bp);
-extern int xlog_find_zeroed(xlog_t *log, xfs_daddr_t *blk_no);
-extern int xlog_find_cycle_start(xlog_t *log, xfs_buf_t *bp,
+extern int xlog_find_zeroed(struct xlog *log, xfs_daddr_t *blk_no);
+extern int xlog_find_cycle_start(struct xlog *log, xfs_buf_t *bp,
xfs_daddr_t first_blk, xfs_daddr_t *last_blk,
uint cycle);
-extern int xlog_find_tail(xlog_t *log, xfs_daddr_t *head_blk,
+extern int xlog_find_tail(struct xlog *log, xfs_daddr_t *head_blk,
xfs_daddr_t *tail_blk);
-extern int xlog_test_footer(xlog_t *log);
-extern int xlog_recover(xlog_t *log, int readonly);
+extern int xlog_test_footer(struct xlog *log);
+extern int xlog_recover(struct xlog *log, int readonly);
extern void xlog_recover_print_data(xfs_caddr_t p, int len);
extern void xlog_recover_print_logitem(xlog_recover_item_t *item);
extern void xlog_recover_print_trans_head(xlog_recover_t *tr);
-extern int xlog_print_find_oldest(xlog_t *log, xfs_daddr_t *last_blk);
+extern int xlog_print_find_oldest(struct xlog *log, xfs_daddr_t *last_blk);
/* for transactional view */
extern void xlog_recover_print_trans_head(xlog_recover_t *tr);
extern void xlog_recover_print_trans(xlog_recover_t *trans,
struct list_head *itemq, int print);
-extern int xlog_do_recovery_pass(xlog_t *log, xfs_daddr_t head_blk,
+extern int xlog_do_recovery_pass(struct xlog *log, xfs_daddr_t head_blk,
xfs_daddr_t tail_blk, int pass);
-extern int xlog_recover_do_trans(xlog_t *log, xlog_recover_t *trans,
+extern int xlog_recover_do_trans(struct xlog *log, xlog_recover_t *trans,
int pass);
extern int xlog_header_check_recover(xfs_mount_t *mp,
xlog_rec_header_t *head);
diff --git a/libxlog/xfs_log_recover.c b/libxlog/xfs_log_recover.c
index 23fe6fd..ad53e86 100644
--- a/libxlog/xfs_log_recover.c
+++ b/libxlog/xfs_log_recover.c
@@ -18,10 +18,11 @@
#include <xfs/libxlog.h>
-#define xlog_unpack_data_checksum(rhead, dp, log) ((void)0)
-#define xlog_clear_stale_blocks(log, tail_lsn) (0)
#define xfs_readonly_buftarg(buftarg) (0)
+/* avoid set-but-unused var warning. gcc is not very bright. */
+#define xlog_clear_stale_blocks(log, taillsn) ((taillsn) = (taillsn))
+
/*
* Verify the given count of basic blocks is valid number of blocks
@@ -31,7 +32,7 @@
static inline int
xlog_buf_bbcount_valid(
- xlog_t *log,
+ struct xlog *log,
int bbcount)
{
return bbcount > 0 && bbcount <= log->l_logBBsize;
@@ -44,11 +45,11 @@ xlog_buf_bbcount_valid(
*/
xfs_buf_t *
xlog_get_bp(
- xlog_t *log,
+ struct xlog *log,
int nbblks)
{
if (!xlog_buf_bbcount_valid(log, nbblks)) {
- xlog_warn("XFS: Invalid block length (0x%x) given for buffer",
+ xfs_warn(log->l_mp, "Invalid block length (0x%x) for buffer",
nbblks);
XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_HIGH, log->l_mp);
return NULL;
@@ -57,7 +58,7 @@ xlog_get_bp(
/*
* We do log I/O in units of log sectors (a power-of-2
* multiple of the basic block size), so we round up the
- * requested size to acommodate the basic blocks required
+ * requested size to accommodate the basic blocks required
* for complete log sectors.
*
* In addition, the buffer may be used for a non-sector-
@@ -68,12 +69,11 @@ xlog_get_bp(
* an issue. Nor will this be a problem if the log I/O is
* done in basic blocks (sector size 1). But otherwise we
* extend the buffer by one extra log sector to ensure
- * there's space to accomodate this possiblility.
+ * there's space to accommodate this possibility.
*/
if (nbblks > 1 && log->l_sectBBsize > 1)
nbblks += log->l_sectBBsize;
- if (log->l_sectBBsize)
- nbblks = round_up(nbblks, log->l_sectBBsize);
+ nbblks = round_up(nbblks, log->l_sectBBsize);
return libxfs_getbufr(log->l_dev, (xfs_daddr_t)-1, nbblks);
}
@@ -91,57 +91,54 @@ xlog_put_bp(
*/
STATIC xfs_caddr_t
xlog_align(
- xlog_t *log,
+ struct xlog *log,
xfs_daddr_t blk_no,
int nbblks,
- xfs_buf_t *bp)
+ struct xfs_buf *bp)
{
- xfs_daddr_t offset = 0;
-
- if (log->l_sectBBsize)
- offset = blk_no & ((xfs_daddr_t)log->l_sectBBsize - 1);
+ xfs_daddr_t offset = blk_no & ((xfs_daddr_t)log->l_sectBBsize - 1);
- ASSERT(BBTOB(offset + nbblks) <= XFS_BUF_SIZE(bp));
- return XFS_BUF_PTR(bp) + BBTOB(offset);
+ ASSERT(offset + nbblks <= bp->b_length);
+ return bp->b_addr + BBTOB(offset);
}
+
/*
* nbblks should be uint, but oh well. Just want to catch that 32-bit length.
*/
int
xlog_bread_noalign(
- xlog_t *log,
+ struct xlog *log,
xfs_daddr_t blk_no,
int nbblks,
- xfs_buf_t *bp)
+ struct xfs_buf *bp)
{
if (!xlog_buf_bbcount_valid(log, nbblks)) {
- xlog_warn("XFS: Invalid block length (0x%x) given for buffer",
+ xfs_warn(log->l_mp, "Invalid block length (0x%x) for buffer",
nbblks);
XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_HIGH, log->l_mp);
return EFSCORRUPTED;
}
- if (log->l_sectBBsize > 1) {
- blk_no = round_down(blk_no, log->l_sectBBsize);
- nbblks = round_up(nbblks, log->l_sectBBsize);
- }
+ blk_no = round_down(blk_no, log->l_sectBBsize);
+ nbblks = round_up(nbblks, log->l_sectBBsize);
ASSERT(nbblks > 0);
ASSERT(BBTOB(nbblks) <= XFS_BUF_SIZE(bp));
XFS_BUF_SET_ADDR(bp, log->l_logBBstart + blk_no);
XFS_BUF_SET_COUNT(bp, BBTOB(nbblks));
+ bp->b_error = 0;
return libxfs_readbufr(log->l_dev, XFS_BUF_ADDR(bp), bp, nbblks, 0);
}
int
xlog_bread(
- xlog_t *log,
+ struct xlog *log,
xfs_daddr_t blk_no,
int nbblks,
- xfs_buf_t *bp,
+ struct xfs_buf *bp,
xfs_caddr_t *offset)
{
int error;
@@ -155,6 +152,35 @@ xlog_bread(
}
/*
+ * Read at an offset into the buffer. Returns with the buffer in it's original
+ * state regardless of the result of the read.
+ */
+STATIC int
+xlog_bread_offset(
+ struct xlog *log,
+ xfs_daddr_t blk_no, /* block to read from */
+ int nbblks, /* blocks to read */
+ struct xfs_buf *bp,
+ xfs_caddr_t offset)
+{
+ xfs_caddr_t orig_offset = bp->b_addr;
+ int orig_len = bp->b_bcount;
+ int error, error2;
+
+ error = XFS_BUF_SET_PTR(bp, offset, BBTOB(nbblks));
+ if (error)
+ return error;
+
+ error = xlog_bread_noalign(log, blk_no, nbblks, bp);
+
+ /* must reset buffer pointer even on error */
+ error2 = XFS_BUF_SET_PTR(bp, orig_offset, orig_len);
+ if (error)
+ return error;
+ return error2;
+}
+
+/*
* This routine finds (to an approximation) the first block in the physical
* log which contains the given cycle. It uses a binary search algorithm.
* Note that the algorithm can not be perfect because the disk will not
@@ -162,8 +188,8 @@ xlog_bread(
*/
int
xlog_find_cycle_start(
- xlog_t *log,
- xfs_buf_t *bp,
+ struct xlog *log,
+ struct xfs_buf *bp,
xfs_daddr_t first_blk,
xfs_daddr_t *last_blk,
uint cycle)
@@ -205,7 +231,7 @@ xlog_find_cycle_start(
*/
STATIC int
xlog_find_verify_cycle(
- xlog_t *log,
+ struct xlog *log,
xfs_daddr_t start_blk,
int nbblks,
uint stop_on_cycle_no,
@@ -225,9 +251,11 @@ xlog_find_verify_cycle(
* a log sector, or we're out of luck.
*/
bufblks = 1 << ffs(nbblks);
+ while (bufblks > log->l_logBBsize)
+ bufblks >>= 1;
while (!(bp = xlog_get_bp(log, bufblks))) {
bufblks >>= 1;
- if (bufblks < MAX(log->l_sectBBsize, 1))
+ if (bufblks < log->l_sectBBsize)
return ENOMEM;
}
@@ -272,7 +300,7 @@ out:
*/
STATIC int
xlog_find_verify_log_record(
- xlog_t *log,
+ struct xlog *log,
xfs_daddr_t start_blk,
xfs_daddr_t *last_blk,
int extra_bblks)
@@ -302,8 +330,8 @@ xlog_find_verify_log_record(
for (i = (*last_blk) - 1; i >= 0; i--) {
if (i < start_blk) {
/* valid log record not found */
- xlog_warn(
- "XFS: Log inconsistent (didn't find previous header)");
+ xfs_warn(log->l_mp,
+ "Log inconsistent (didn't find previous header)");
ASSERT(0);
error = XFS_ERROR(EIO);
goto out;
@@ -317,7 +345,7 @@ xlog_find_verify_log_record(
head = (xlog_rec_header_t *)offset;
- if (XLOG_HEADER_MAGIC_NUM == be32_to_cpu(head->h_magicno))
+ if (head->h_magicno == cpu_to_be32(XLOG_HEADER_MAGIC_NUM))
break;
if (!smallmem)
@@ -382,7 +410,7 @@ out:
*/
STATIC int
xlog_find_head(
- xlog_t *log,
+ struct xlog *log,
xfs_daddr_t *return_head_blk)
{
xfs_buf_t *bp;
@@ -403,12 +431,12 @@ xlog_find_head(
* mkfs etc write a dummy unmount record to a fresh
* log so we can store the uuid in there
*/
- xlog_warn("XFS: totally zeroed log");
+ xfs_warn(log->l_mp, "totally zeroed log");
}
return 0;
} else if (error) {
- xlog_warn("XFS: empty log check failed");
+ xfs_warn(log->l_mp, "empty log check failed");
return error;
}
@@ -631,7 +659,7 @@ validate_head:
xlog_put_bp(bp);
if (error)
- xlog_warn("XFS: failed to find log head");
+ xfs_warn(log->l_mp, "failed to find log head");
return error;
}
@@ -653,7 +681,7 @@ validate_head:
*/
int
xlog_find_tail(
- xlog_t *log,
+ struct xlog *log,
xfs_daddr_t *head_blk,
xfs_daddr_t *tail_blk)
{
@@ -699,7 +727,7 @@ xlog_find_tail(
if (error)
goto done;
- if (XLOG_HEADER_MAGIC_NUM == be32_to_cpu(*(__be32 *)offset)) {
+ if (*(__be32 *)offset == cpu_to_be32(XLOG_HEADER_MAGIC_NUM)) {
found = 1;
break;
}
@@ -716,15 +744,15 @@ xlog_find_tail(
if (error)
goto done;
- if (XLOG_HEADER_MAGIC_NUM ==
- be32_to_cpu(*(__be32 *)offset)) {
+ if (*(__be32 *)offset ==
+ cpu_to_be32(XLOG_HEADER_MAGIC_NUM)) {
found = 2;
break;
}
}
}
if (!found) {
- xlog_warn("XFS: xlog_find_tail: couldn't find sync record");
+ xfs_warn(log->l_mp, "%s: couldn't find sync record", __func__);
ASSERT(0);
return XFS_ERROR(EIO);
}
@@ -750,9 +778,9 @@ xlog_find_tail(
log->l_curr_cycle++;
atomic64_set(&log->l_tail_lsn, be64_to_cpu(rhead->h_tail_lsn));
atomic64_set(&log->l_last_sync_lsn, be64_to_cpu(rhead->h_lsn));
- xlog_assign_grant_head(&log->l_grant_reserve_head, log->l_curr_cycle,
+ xlog_assign_grant_head(&log->l_reserve_head.grant, log->l_curr_cycle,
BBTOB(log->l_curr_block));
- xlog_assign_grant_head(&log->l_grant_write_head, log->l_curr_cycle,
+ xlog_assign_grant_head(&log->l_write_head.grant, log->l_curr_cycle,
BBTOB(log->l_curr_block));
/*
@@ -840,7 +868,7 @@ done:
xlog_put_bp(bp);
if (error)
- xlog_warn("XFS: failed to locate log tail");
+ xfs_warn(log->l_mp, "failed to locate log tail");
return error;
}
@@ -862,7 +890,7 @@ done:
*/
int
xlog_find_zeroed(
- xlog_t *log,
+ struct xlog *log,
xfs_daddr_t *blk_no)
{
xfs_buf_t *bp;
@@ -904,7 +932,8 @@ xlog_find_zeroed(
* the first block must be 1. If it's not, maybe we're
* not looking at a log... Bail out.
*/
- xlog_warn("XFS: Log inconsistent or not a log (last==0, first!=1)");
+ xfs_warn(log->l_mp,
+ "Log inconsistent or not a log (last==0, first!=1)");
return XFS_ERROR(EINVAL);
}
@@ -1002,8 +1031,8 @@ xlog_recover_add_item(
STATIC int
xlog_recover_add_to_cont_trans(
- struct log *log,
- xlog_recover_t *trans,
+ struct xlog *log,
+ struct xlog_recover *trans,
xfs_caddr_t dp,
int len)
{
@@ -1025,7 +1054,7 @@ xlog_recover_add_to_cont_trans(
old_ptr = item->ri_buf[item->ri_cnt-1].i_addr;
old_len = item->ri_buf[item->ri_cnt-1].i_len;
- ptr = kmem_realloc(old_ptr, len+old_len, old_len, 0u);
+ ptr = kmem_realloc(old_ptr, len+old_len, old_len, KM_SLEEP);
memcpy(&ptr[old_len], dp, len); /* d, s, l */
item->ri_buf[item->ri_cnt-1].i_len += len;
item->ri_buf[item->ri_cnt-1].i_addr = ptr;
@@ -1048,8 +1077,8 @@ xlog_recover_add_to_cont_trans(
*/
STATIC int
xlog_recover_add_to_trans(
- struct log *log,
- xlog_recover_t *trans,
+ struct xlog *log,
+ struct xlog_recover *trans,
xfs_caddr_t dp,
int len)
{
@@ -1062,8 +1091,8 @@ xlog_recover_add_to_trans(
if (list_empty(&trans->r_itemq)) {
/* we need to catch log corruptions here */
if (*(uint *)dp != XFS_TRANS_HEADER_MAGIC) {
- xlog_warn("XFS: xlog_recover_add_to_trans: "
- "bad header magic number");
+ xfs_warn(log->l_mp, "%s: bad header magic number",
+ __func__);
ASSERT(0);
return XFS_ERROR(EIO);
}
@@ -1090,8 +1119,8 @@ xlog_recover_add_to_trans(
if (item->ri_total == 0) { /* first region to be added */
if (in_f->ilf_size == 0 ||
in_f->ilf_size > XLOG_MAX_REGIONS_IN_ITEM) {
- xlog_warn(
- "XFS: bad number of regions (%d) in inode log format",
+ xfs_warn(log->l_mp,
+ "bad number of regions (%d) in inode log format",
in_f->ilf_size);
ASSERT(0);
return XFS_ERROR(EIO);
@@ -1144,7 +1173,7 @@ xlog_recover_free_trans(
*/
STATIC int
xlog_recover_commit_trans(
- struct log *log,
+ struct xlog *log,
struct xlog_recover *trans,
int pass)
{
@@ -1163,7 +1192,7 @@ xlog_recover_unmount_trans(
xlog_recover_t *trans)
{
/* Do nothing now */
- xlog_warn("XFS: xlog_recover_unmount_trans: Unmount LR");
+ xfs_warn(log->l_mp, "%s: Unmount LR", __func__);
return 0;
}
@@ -1178,9 +1207,9 @@ xlog_recover_unmount_trans(
*/
STATIC int
xlog_recover_process_data(
- xlog_t *log,
+ struct xlog *log,
struct hlist_head rhash[],
- xlog_rec_header_t *rhead,
+ struct xlog_rec_header *rhead,
xfs_caddr_t dp,
int pass)
{
@@ -1206,8 +1235,8 @@ xlog_recover_process_data(
dp += sizeof(xlog_op_header_t);
if (ohead->oh_clientid != XFS_TRANSACTION &&
ohead->oh_clientid != XFS_LOG) {
- xlog_warn(
- "XFS: xlog_recover_process_data: bad clientid");
+ xfs_warn(log->l_mp, "%s: bad clientid 0x%x",
+ __func__, ohead->oh_clientid);
ASSERT(0);
return (XFS_ERROR(EIO));
}
@@ -1220,8 +1249,8 @@ xlog_recover_process_data(
be64_to_cpu(rhead->h_lsn));
} else {
if (dp + be32_to_cpu(ohead->oh_len) > lp) {
- xlog_warn(
- "XFS: xlog_recover_process_data: bad length");
+ xfs_warn(log->l_mp, "%s: bad length 0x%x",
+ __func__, be32_to_cpu(ohead->oh_len));
return (XFS_ERROR(EIO));
}
flags = ohead->oh_flags & ~XLOG_END_TRANS;
@@ -1241,8 +1270,8 @@ xlog_recover_process_data(
be32_to_cpu(ohead->oh_len));
break;
case XLOG_START_TRANS:
- xlog_warn(
- "XFS: xlog_recover_process_data: bad transaction");
+ xfs_warn(log->l_mp, "%s: bad transaction",
+ __func__);
ASSERT(0);
error = XFS_ERROR(EIO);
break;
@@ -1252,8 +1281,8 @@ xlog_recover_process_data(
dp, be32_to_cpu(ohead->oh_len));
break;
default:
- xlog_warn(
- "XFS: xlog_recover_process_data: bad flag");
+ xfs_warn(log->l_mp, "%s: bad flag 0x%x",
+ __func__, flags);
ASSERT(0);
error = XFS_ERROR(EIO);
break;
@@ -1267,13 +1296,64 @@ xlog_recover_process_data(
return 0;
}
-STATIC void
+/*
+ * Upack the log buffer data and crc check it. If the check fails, issue a
+ * warning if and only if the CRC in the header is non-zero. This makes the
+ * check an advisory warning, and the zero CRC check will prevent failure
+ * warnings from being emitted when upgrading the kernel from one that does not
+ * add CRCs by default.
+ *
+ * When filesystems are CRC enabled, this CRC mismatch becomes a fatal log
+ * corruption failure
+ *
+ * XXX: we cannot calculate the CRC here without knowing the number of extra
+ * headers taht were CRC'd in log write. This information is derived from the
+ * in-core log buffer size, and is not written to the log! Hence for now we
+ * simple *cannot* verify the CRCs, and so we short circuit it.
+ */
+#define xlog_cksum(l,r,dp,len) ((r)->h_crc)
+STATIC int
+xlog_unpack_data_crc(
+ struct xlog_rec_header *rhead,
+ xfs_caddr_t dp,
+ struct xlog *log)
+{
+ __le32 crc;
+
+ crc = xlog_cksum(log, rhead, dp, be32_to_cpu(rhead->h_len));
+ if (crc != rhead->h_crc) {
+ if (rhead->h_crc || xfs_sb_version_hascrc(&log->l_mp->m_sb)) {
+ xfs_alert(log->l_mp,
+ "log record CRC mismatch: found 0x%x, expected 0x%x.\n",
+ le32_to_cpu(rhead->h_crc),
+ le32_to_cpu(crc));
+ xfs_hex_dump(dp, 32);
+ }
+
+ /*
+ * If we've detected a log record corruption, then we can't
+ * recover past this point. Abort recovery if we are enforcing
+ * CRC protection by punting an error back up the stack.
+ */
+ if (xfs_sb_version_hascrc(&log->l_mp->m_sb))
+ return EFSCORRUPTED;
+ }
+
+ return 0;
+}
+
+STATIC int
xlog_unpack_data(
- xlog_rec_header_t *rhead,
+ struct xlog_rec_header *rhead,
xfs_caddr_t dp,
- xlog_t *log)
+ struct xlog *log)
{
int i, j, k;
+ int error;
+
+ error = xlog_unpack_data_crc(rhead, dp, log);
+ if (error)
+ return error;
for (i = 0; i < BTOBB(be32_to_cpu(rhead->h_len)) &&
i < (XLOG_HEADER_CYCLE_SIZE / BBSIZE); i++) {
@@ -1290,17 +1370,19 @@ xlog_unpack_data(
dp += BBSIZE;
}
}
+
+ return 0;
}
STATIC int
xlog_valid_rec_header(
- xlog_t *log,
- xlog_rec_header_t *rhead,
+ struct xlog *log,
+ struct xlog_rec_header *rhead,
xfs_daddr_t blkno)
{
int hlen;
- if (unlikely(be32_to_cpu(rhead->h_magicno) != XLOG_HEADER_MAGIC_NUM)) {
+ if (unlikely(rhead->h_magicno != cpu_to_be32(XLOG_HEADER_MAGIC_NUM))) {
XFS_ERROR_REPORT("xlog_valid_rec_header(1)",
XFS_ERRLEVEL_LOW, log->l_mp);
return XFS_ERROR(EFSCORRUPTED);
@@ -1308,7 +1390,7 @@ xlog_valid_rec_header(
if (unlikely(
(!rhead->h_version ||
(be32_to_cpu(rhead->h_version) & (~XLOG_VERSION_OKBITS))))) {
- xlog_warn("XFS: %s: unrecognised log version (%d).",
+ xfs_warn(log->l_mp, "%s: unrecognised log version (%d).",
__func__, be32_to_cpu(rhead->h_version));
return XFS_ERROR(EIO);
}
@@ -1338,7 +1420,7 @@ xlog_valid_rec_header(
*/
int
xlog_do_recovery_pass(
- xlog_t *log,
+ struct xlog *log,
xfs_daddr_t head_blk,
xfs_daddr_t tail_blk,
int pass)
@@ -1421,9 +1503,13 @@ xlog_do_recovery_pass(
if (error)
goto bread_err2;
- xlog_unpack_data(rhead, offset, log);
- if ((error = xlog_recover_process_data(log,
- rhash, rhead, offset, pass)))
+ error = xlog_unpack_data(rhead, offset, log);
+ if (error)
+ goto bread_err2;
+
+ error = xlog_recover_process_data(log,
+ rhash, rhead, offset, pass);
+ if (error)
goto bread_err2;
blk_no += bblks + hblks;
}
@@ -1438,7 +1524,7 @@ xlog_do_recovery_pass(
/*
* Check for header wrapping around physical end-of-log
*/
- offset = XFS_BUF_PTR(hbp);
+ offset = hbp->b_addr;
split_hblks = 0;
wrapped_hblks = 0;
if (blk_no + hblks <= log->l_logBBsize) {
@@ -1474,19 +1560,9 @@ xlog_do_recovery_pass(
* - order is important.
*/
wrapped_hblks = hblks - split_hblks;
- error = XFS_BUF_SET_PTR(hbp,
- offset + BBTOB(split_hblks),
- BBTOB(hblks - split_hblks));
- if (error)
- goto bread_err2;
-
- error = xlog_bread_noalign(log, 0,
- wrapped_hblks, hbp);
- if (error)
- goto bread_err2;
-
- error = XFS_BUF_SET_PTR(hbp, offset,
- BBTOB(hblks));
+ error = xlog_bread_offset(log, 0,
+ wrapped_hblks, hbp,
+ offset + BBTOB(split_hblks));
if (error)
goto bread_err2;
}
@@ -1508,7 +1584,7 @@ xlog_do_recovery_pass(
} else {
/* This log record is split across the
* physical end of log */
- offset = XFS_BUF_PTR(dbp);
+ offset = dbp->b_addr;
split_bblks = 0;
if (blk_no != log->l_logBBsize) {
/* some data is before the physical
@@ -1537,25 +1613,20 @@ xlog_do_recovery_pass(
* _first_, then the log start (LR header end)
* - order is important.
*/
- error = XFS_BUF_SET_PTR(dbp,
- offset + BBTOB(split_bblks),
- BBTOB(bblks - split_bblks));
+ error = xlog_bread_offset(log, 0,
+ bblks - split_bblks, dbp,
+ offset + BBTOB(split_bblks));
if (error)
goto bread_err2;
+ }
- error = xlog_bread_noalign(log, wrapped_hblks,
- bblks - split_bblks,
- dbp);
- if (error)
- goto bread_err2;
+ error = xlog_unpack_data(rhead, offset, log);
+ if (error)
+ goto bread_err2;
- error = XFS_BUF_SET_PTR(dbp, offset, h_size);
- if (error)
- goto bread_err2;
- }
- xlog_unpack_data(rhead, offset, log);
- if ((error = xlog_recover_process_data(log, rhash,
- rhead, offset, pass)))
+ error = xlog_recover_process_data(log, rhash,
+ rhead, offset, pass);
+ if (error)
goto bread_err2;
blk_no += bblks;
}
@@ -1580,9 +1651,13 @@ xlog_do_recovery_pass(
if (error)
goto bread_err2;
- xlog_unpack_data(rhead, offset, log);
- if ((error = xlog_recover_process_data(log, rhash,
- rhead, offset, pass)))
+ error = xlog_unpack_data(rhead, offset, log);
+ if (error)
+ goto bread_err2;
+
+ error = xlog_recover_process_data(log, rhash,
+ rhead, offset, pass);
+ if (error)
goto bread_err2;
blk_no += bblks + hblks;
}
diff --git a/logprint/log_copy.c b/logprint/log_copy.c
index fd36317..f6ef0e9 100644
--- a/logprint/log_copy.c
+++ b/logprint/log_copy.c
@@ -24,7 +24,7 @@
void
xfs_log_copy(
- xlog_t *log,
+ struct xlog *log,
int fd,
char *filename)
{
diff --git a/logprint/log_dump.c b/logprint/log_dump.c
index 1975b1b..a333077 100644
--- a/logprint/log_dump.c
+++ b/logprint/log_dump.c
@@ -24,7 +24,7 @@
void
xfs_log_dump(
- xlog_t *log,
+ struct xlog *log,
int fd,
int print_block_start)
{
diff --git a/logprint/log_misc.c b/logprint/log_misc.c
index c480795..4277a48 100644
--- a/logprint/log_misc.c
+++ b/logprint/log_misc.c
@@ -816,7 +816,7 @@ xlog_print_trans_dquot(xfs_caddr_t *ptr, int len, int *i, int num_ops)
*/
void
-xlog_print_lseek(xlog_t *log, int fd, xfs_daddr_t blkno, int whence)
+xlog_print_lseek(struct xlog *log, int fd, xfs_daddr_t blkno, int whence)
{
#define BBTOOFF64(bbs) (((xfs_off_t)(bbs)) << BBSHIFT)
xfs_off_t offset;
@@ -1265,7 +1265,7 @@ xlog_print_extended_headers(
/*
* This code is gross and needs to be rewritten.
*/
-void xfs_log_print(xlog_t *log,
+void xfs_log_print(struct xlog *log,
int fd,
int print_block_start)
{
diff --git a/logprint/log_print_all.c b/logprint/log_print_all.c
index cbc51d9..44fb4c4 100644
--- a/logprint/log_print_all.c
+++ b/logprint/log_print_all.c
@@ -23,7 +23,7 @@
*/
int
xlog_print_find_oldest(
- struct log *log,
+ struct xlog *log,
xfs_daddr_t *last_blk)
{
xfs_buf_t *bp;
diff --git a/logprint/log_print_trans.c b/logprint/log_print_trans.c
index 7405772..86e1c42 100644
--- a/logprint/log_print_trans.c
+++ b/logprint/log_print_trans.c
@@ -30,7 +30,7 @@ xlog_recover_print_trans_head(
int
xlog_recover_do_trans(
- xlog_t *log,
+ struct xlog *log,
xlog_recover_t *trans,
int pass)
{
@@ -40,7 +40,7 @@ xlog_recover_do_trans(
void
xfs_log_print_trans(
- xlog_t *log,
+ struct xlog *log,
int print_block_start)
{
xfs_daddr_t head_blk, tail_blk;
diff --git a/logprint/logprint.c b/logprint/logprint.c
index 482fd4e..2a01780 100644
--- a/logprint/logprint.c
+++ b/logprint/logprint.c
@@ -128,7 +128,7 @@ main(int argc, char **argv)
int c;
int logfd;
char *copy_file = NULL;
- xlog_t log = {0};
+ struct xlog log = {0};
xfs_mount_t mount;
setlocale(LC_ALL, "");
diff --git a/logprint/logprint.h b/logprint/logprint.h
index df4cea3..933c9e6 100644
--- a/logprint/logprint.h
+++ b/logprint/logprint.h
@@ -34,12 +34,12 @@ extern int print_no_print;
/* exports */
extern char *trans_type[];
-extern void xlog_print_lseek(xlog_t *, int, xfs_daddr_t, int);
+extern void xlog_print_lseek(struct xlog *, int, xfs_daddr_t, int);
-extern void xfs_log_copy(xlog_t *, int, char *);
-extern void xfs_log_dump(xlog_t *, int, int);
-extern void xfs_log_print(xlog_t *, int, int);
-extern void xfs_log_print_trans(xlog_t *, int);
+extern void xfs_log_copy(struct xlog *, int, char *);
+extern void xfs_log_dump(struct xlog *, int, int);
+extern void xfs_log_print(struct xlog *, int, int);
+extern void xfs_log_print_trans(struct xlog *, int);
extern void print_xlog_record_line(void);
extern void print_xlog_op_line(void);
diff --git a/repair/phase2.c b/repair/phase2.c
index 9fd591c..23b457a 100644
--- a/repair/phase2.c
+++ b/repair/phase2.c
@@ -29,13 +29,16 @@
void set_mp(xfs_mount_t *mpp);
/* workaround craziness in the xlog routines */
-int xlog_recover_do_trans(xlog_t *log, xlog_recover_t *t, int p) { return 0; }
+int xlog_recover_do_trans(struct xlog *log, xlog_recover_t *t, int p)
+{
+ return 0;
+}
static void
zero_log(xfs_mount_t *mp)
{
int error;
- xlog_t log;
+ struct xlog log;
xfs_daddr_t head_blk, tail_blk;
dev_t logdev = (mp->m_sb.sb_logstart == 0) ? x.logdev : x.ddev;
xfsprogs-crc-patchset-v2/btree-format-changes 0000600 0001750 0001750 00000153721 12132011546 020170 0 ustar dave dave libxfs: add crc format changes to generic btrees
From: Dave Chinner <dchinner@xxxxxxxxxx>
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
include/libxfs.h | 15 +--
include/xfs_alloc_btree.h | 13 ++-
include/xfs_bmap_btree.h | 20 ++--
include/xfs_btree.h | 60 +++++++++--
include/xfs_buf_item.h | 24 ++++-
include/xfs_dinode.h | 4 +-
include/xfs_ialloc_btree.h | 10 +-
include/xfs_trans.h | 2 +
libxfs/rdwr.c | 24 ++---
libxfs/xfs.h | 4 +
libxfs/xfs_alloc_btree.c | 99 +++++++++++------
libxfs/xfs_attr_leaf.c | 2 +-
libxfs/xfs_bmap.c | 49 ++++++---
libxfs/xfs_bmap_btree.c | 107 ++++++++++++------
libxfs/xfs_btree.c | 257 ++++++++++++++++++++++++++++++++++++--------
libxfs/xfs_ialloc_btree.c | 80 +++++++++-----
libxfs/xfs_inode.c | 33 +++---
libxfs/xfs_mount.c | 2 +-
mdrestore/Makefile | 2 +-
19 files changed, 587 insertions(+), 220 deletions(-)
diff --git a/include/libxfs.h b/include/libxfs.h
index b6e83f4..a4564fd 100644
--- a/include/libxfs.h
+++ b/include/libxfs.h
@@ -240,14 +240,14 @@ struct xfs_buf_ops {
typedef struct xfs_buf {
struct cache_node b_node;
unsigned int b_flags;
- xfs_daddr_t b_blkno;
+ xfs_daddr_t b_bn;
unsigned b_bcount;
unsigned int b_length;
dev_t b_dev;
pthread_mutex_t b_lock;
pthread_t b_holder;
unsigned int b_recur;
- void *b_fsprivate;
+ void *b_fspriv;
void *b_fsprivate2;
void *b_fsprivate3;
void *b_addr;
@@ -273,9 +273,11 @@ enum xfs_buf_flags_t { /* b_flags bits */
LIBXFS_B_DISCONTIG = 0x0010, /* discontiguous buffer */
};
+#define XFS_BUF_DADDR_NULL ((xfs_daddr_t) (-1LL))
+
#define XFS_BUF_PTR(bp) ((char *)(bp)->b_addr)
#define xfs_buf_offset(bp, offset) (XFS_BUF_PTR(bp) + (offset))
-#define XFS_BUF_ADDR(bp) ((bp)->b_blkno)
+#define XFS_BUF_ADDR(bp) ((bp)->b_bn)
#define XFS_BUF_SIZE(bp) ((bp)->b_bcount)
#define XFS_BUF_COUNT(bp) ((bp)->b_bcount)
#define XFS_BUF_TARGET(bp) ((bp)->b_dev)
@@ -284,11 +286,11 @@ enum xfs_buf_flags_t { /* b_flags bits */
XFS_BUF_SET_COUNT(bp,cnt); \
})
-#define XFS_BUF_SET_ADDR(bp,blk) ((bp)->b_blkno = (blk))
+#define XFS_BUF_SET_ADDR(bp,blk) ((bp)->b_bn = (blk))
#define XFS_BUF_SET_COUNT(bp,cnt) ((bp)->b_bcount = (cnt))
-#define XFS_BUF_FSPRIVATE(bp,type) ((type)(bp)->b_fsprivate)
-#define XFS_BUF_SET_FSPRIVATE(bp,val) (bp)->b_fsprivate = (void *)(val)
+#define XFS_BUF_FSPRIVATE(bp,type) ((type)(bp)->b_fspriv)
+#define XFS_BUF_SET_FSPRIVATE(bp,val) (bp)->b_fspriv = (void *)(val)
#define XFS_BUF_FSPRIVATE2(bp,type) ((type)(bp)->b_fsprivate2)
#define XFS_BUF_SET_FSPRIVATE2(bp,val) (bp)->b_fsprivate2 = (void *)(val)
#define XFS_BUF_FSPRIVATE3(bp,type) ((type)(bp)->b_fsprivate3)
@@ -392,6 +394,7 @@ typedef struct xfs_log_item {
struct xfs_log_item_desc *li_desc; /* ptr to current desc*/
struct xfs_mount *li_mountp; /* ptr to fs mount */
uint li_type; /* item type */
+ xfs_lsn_t li_lsn;
} xfs_log_item_t;
typedef struct xfs_inode_log_item {
diff --git a/include/xfs_alloc_btree.h b/include/xfs_alloc_btree.h
index 7e89a2b..70c3ea0 100644
--- a/include/xfs_alloc_btree.h
+++ b/include/xfs_alloc_btree.h
@@ -31,8 +31,10 @@ struct xfs_mount;
* by blockcount and blockno. All blocks look the same to make the code
* simpler; if we have time later, we'll make the optimizations.
*/
-#define XFS_ABTB_MAGIC 0x41425442 /* 'ABTB' for bno tree */
-#define XFS_ABTC_MAGIC 0x41425443 /* 'ABTC' for cnt tree */
+#define XFS_ABTB_MAGIC 0x41425442 /* 'ABTB' for bno tree */
+#define XFS_ABTB_CRC_MAGIC 0x41423342 /* 'AB3B' */
+#define XFS_ABTC_MAGIC 0x41425443 /* 'ABTC' for cnt tree */
+#define XFS_ABTC_CRC_MAGIC 0x41423343 /* 'AB3C' */
/*
* Data record/key structure
@@ -59,10 +61,11 @@ typedef __be32 xfs_alloc_ptr_t;
/*
* Btree block header size depends on a superblock flag.
- *
- * (not quite yet, but soon)
*/
-#define XFS_ALLOC_BLOCK_LEN(mp) XFS_BTREE_SBLOCK_LEN
+#define XFS_ALLOC_BLOCK_LEN(mp) \
+ (xfs_sb_version_hascrc(&((mp)->m_sb)) ? \
+ XFS_BTREE_SBLOCK_LEN + XFS_BTREE_CRCBLOCK_ADD : \
+ XFS_BTREE_SBLOCK_LEN)
/*
* Record, key, and pointer address macros for btree blocks.
diff --git a/include/xfs_bmap_btree.h b/include/xfs_bmap_btree.h
index 88469ca..8a28b89 100644
--- a/include/xfs_bmap_btree.h
+++ b/include/xfs_bmap_btree.h
@@ -18,7 +18,8 @@
#ifndef __XFS_BMAP_BTREE_H__
#define __XFS_BMAP_BTREE_H__
-#define XFS_BMAP_MAGIC 0x424d4150 /* 'BMAP' */
+#define XFS_BMAP_MAGIC 0x424d4150 /* 'BMAP' */
+#define XFS_BMAP_CRC_MAGIC 0x424d4133 /* 'BMA3' */
struct xfs_btree_cur;
struct xfs_btree_block;
@@ -136,10 +137,11 @@ typedef __be64 xfs_bmbt_ptr_t, xfs_bmdr_ptr_t;
/*
* Btree block header size depends on a superblock flag.
- *
- * (not quite yet, but soon)
*/
-#define XFS_BMBT_BLOCK_LEN(mp) XFS_BTREE_LBLOCK_LEN
+#define XFS_BMBT_BLOCK_LEN(mp) \
+ (xfs_sb_version_hascrc(&((mp)->m_sb)) ? \
+ XFS_BTREE_LBLOCK_LEN + XFS_BTREE_CRCBLOCK_ADD : \
+ XFS_BTREE_LBLOCK_LEN)
#define XFS_BMBT_REC_ADDR(mp, block, index) \
((xfs_bmbt_rec_t *) \
@@ -186,12 +188,12 @@ typedef __be64 xfs_bmbt_ptr_t, xfs_bmdr_ptr_t;
#define XFS_BMAP_BROOT_PTR_ADDR(mp, bb, i, sz) \
XFS_BMBT_PTR_ADDR(mp, bb, i, xfs_bmbt_maxrecs(mp, sz, 0))
-#define XFS_BMAP_BROOT_SPACE_CALC(nrecs) \
- (int)(XFS_BTREE_LBLOCK_LEN + \
+#define XFS_BMAP_BROOT_SPACE_CALC(mp, nrecs) \
+ (int)(XFS_BMBT_BLOCK_LEN(mp) + \
((nrecs) * (sizeof(xfs_bmbt_key_t) + sizeof(xfs_bmbt_ptr_t))))
-#define XFS_BMAP_BROOT_SPACE(bb) \
- (XFS_BMAP_BROOT_SPACE_CALC(be16_to_cpu((bb)->bb_numrecs)))
+#define XFS_BMAP_BROOT_SPACE(mp, bb) \
+ (XFS_BMAP_BROOT_SPACE_CALC(mp, be16_to_cpu((bb)->bb_numrecs)))
#define XFS_BMDR_SPACE_CALC(nrecs) \
(int)(sizeof(xfs_bmdr_block_t) + \
((nrecs) * (sizeof(xfs_bmbt_key_t) + sizeof(xfs_bmbt_ptr_t))))
@@ -204,7 +206,7 @@ typedef __be64 xfs_bmbt_ptr_t, xfs_bmdr_ptr_t;
/*
* Prototypes for xfs_bmap.c to call.
*/
-extern void xfs_bmdr_to_bmbt(struct xfs_mount *, xfs_bmdr_block_t *, int,
+extern void xfs_bmdr_to_bmbt(struct xfs_inode *, xfs_bmdr_block_t *, int,
struct xfs_btree_block *, int);
extern void xfs_bmbt_get_all(xfs_bmbt_rec_host_t *r, xfs_bmbt_irec_t *s);
extern xfs_filblks_t xfs_bmbt_get_blockcount(xfs_bmbt_rec_host_t *r);
diff --git a/include/xfs_btree.h b/include/xfs_btree.h
index be1eb23..02f89d8 100644
--- a/include/xfs_btree.h
+++ b/include/xfs_btree.h
@@ -42,11 +42,15 @@ extern kmem_zone_t *xfs_btree_cur_zone;
* Generic btree header.
*
* This is a combination of the actual format used on disk for short and long
- * format btrees. The first three fields are shared by both format, but
- * the pointers are different and should be used with care.
+ * format btrees. The first three fields are shared by both format, but the
+ * pointers are different and should be used with care.
*
- * To get the size of the actual short or long form headers please use
- * the size macros below. Never use sizeof(xfs_btree_block).
+ * To get the size of the actual short or long form headers please use the size
+ * macros below. Never use sizeof(xfs_btree_block).
+ *
+ * The blkno, crc, lsn, owner and uuid fields are only available in filesystems
+ * with the crc feature bit, and all accesses to them must be conditional on
+ * that flag.
*/
struct xfs_btree_block {
__be32 bb_magic; /* magic number for block type */
@@ -56,16 +60,35 @@ struct xfs_btree_block {
struct {
__be32 bb_leftsib;
__be32 bb_rightsib;
+
+ __be64 bb_blkno;
+ __be64 bb_lsn;
+ uuid_t bb_uuid;
+ __be32 bb_owner;
+ __le32 bb_crc;
} s; /* short form pointers */
struct {
__be64 bb_leftsib;
__be64 bb_rightsib;
+
+ __be64 bb_blkno;
+ __be64 bb_lsn;
+ uuid_t bb_uuid;
+ __be64 bb_owner;
+ __le32 bb_crc;
+ __be32 bb_pad; /* padding for alignment */
} l; /* long form pointers */
} bb_u; /* rest */
};
#define XFS_BTREE_SBLOCK_LEN 16 /* size of a short form block */
#define XFS_BTREE_LBLOCK_LEN 24 /* size of a long form block */
+#define XFS_BTREE_CRCBLOCK_ADD 32 /* size of blkno + crc + uuid */
+
+#define XFS_BTREE_SBLOCK_CRC_OFF \
+ offsetof(struct xfs_btree_block, bb_u.s.bb_crc)
+#define XFS_BTREE_LBLOCK_CRC_OFF \
+ offsetof(struct xfs_btree_block, bb_u.l.bb_crc)
/*
@@ -101,13 +124,11 @@ union xfs_btree_rec {
#define XFS_BB_NUMRECS 0x04
#define XFS_BB_LEFTSIB 0x08
#define XFS_BB_RIGHTSIB 0x10
+#define XFS_BB_BLKNO 0x20
#define XFS_BB_NUM_BITS 5
#define XFS_BB_ALL_BITS ((1 << XFS_BB_NUM_BITS) - 1)
-
-/*
- * Magic numbers for btree blocks.
- */
-extern const __uint32_t xfs_magics[];
+#define XFS_BB_NUM_BITS_CRC 8
+#define XFS_BB_ALL_BITS_CRC ((1 << XFS_BB_NUM_BITS_CRC) - 1)
/*
* Generic stats interface
@@ -275,6 +296,7 @@ typedef struct xfs_btree_cur
#define XFS_BTREE_LONG_PTRS (1<<0) /* pointers are 64bits long */
#define XFS_BTREE_ROOT_IN_INODE (1<<1) /* root may be variable size */
#define XFS_BTREE_LASTREC_UPDATE (1<<2) /* track last rec externally */
+#define XFS_BTREE_CRC_BLOCKS (1<<3) /* uses extended btree blocks */
#define XFS_BTREE_NOERROR 0
@@ -412,8 +434,20 @@ xfs_btree_init_block(
__u32 magic,
__u16 level,
__u16 numrecs,
+ __u64 owner,
unsigned int flags);
+void
+xfs_btree_init_block_int(
+ struct xfs_mount *mp,
+ struct xfs_btree_block *buf,
+ xfs_daddr_t blkno,
+ __u32 magic,
+ __u16 level,
+ __u16 numrecs,
+ __u64 owner,
+ unsigned int flags);
+
/*
* Common btree core entry points.
*/
@@ -427,6 +461,14 @@ int xfs_btree_delete(struct xfs_btree_cur *, int *);
int xfs_btree_get_rec(struct xfs_btree_cur *, union xfs_btree_rec **, int *);
/*
+ * btree block CRC helpers
+ */
+void xfs_btree_lblock_calc_crc(struct xfs_buf *);
+bool xfs_btree_lblock_verify_crc(struct xfs_buf *);
+void xfs_btree_sblock_calc_crc(struct xfs_buf *);
+bool xfs_btree_sblock_verify_crc(struct xfs_buf *);
+
+/*
* Internal btree helpers also used by xfs_bmap.c.
*/
void xfs_btree_log_block(struct xfs_btree_cur *, struct xfs_buf *, int);
diff --git a/include/xfs_buf_item.h b/include/xfs_buf_item.h
index ee36c88..101ef83 100644
--- a/include/xfs_buf_item.h
+++ b/include/xfs_buf_item.h
@@ -24,19 +24,33 @@ extern kmem_zone_t *xfs_buf_item_zone;
* This flag indicates that the buffer contains on disk inodes
* and requires special recovery handling.
*/
-#define XFS_BLF_INODE_BUF 0x1
+#define XFS_BLF_INODE_BUF (1<<0)
/*
* This flag indicates that the buffer should not be replayed
* during recovery because its blocks are being freed.
*/
-#define XFS_BLF_CANCEL 0x2
+#define XFS_BLF_CANCEL (1<<1)
+
/*
* This flag indicates that the buffer contains on disk
* user or group dquots and may require special recovery handling.
*/
-#define XFS_BLF_UDQUOT_BUF 0x4
-#define XFS_BLF_PDQUOT_BUF 0x8
-#define XFS_BLF_GDQUOT_BUF 0x10
+#define XFS_BLF_UDQUOT_BUF (1<<2)
+#define XFS_BLF_PDQUOT_BUF (1<<3)
+#define XFS_BLF_GDQUOT_BUF (1<<4)
+
+/*
+ * all buffers now need flags to tell recovery where the magic number
+ * is so that it can verify and calculate the CRCs on the buffer correctly
+ * once the changes have been replayed into the buffer.
+ */
+#define XFS_BLF_BTREE_BUF (1<<5)
+
+#define XFS_BLF_TYPE_MASK \
+ (XFS_BLF_UDQUOT_BUF | \
+ XFS_BLF_PDQUOT_BUF | \
+ XFS_BLF_GDQUOT_BUF | \
+ XFS_BLF_BTREE_BUF)
#define XFS_BLF_CHUNK 128
#define XFS_BLF_SHIFT 7
diff --git a/include/xfs_dinode.h b/include/xfs_dinode.h
index 88a3368..6b5bd17 100644
--- a/include/xfs_dinode.h
+++ b/include/xfs_dinode.h
@@ -107,8 +107,8 @@ typedef enum xfs_dinode_fmt {
#define XFS_LITINO(mp, version) \
((int)(((mp)->m_sb.sb_inodesize) - sizeof(struct xfs_dinode)))
-#define XFS_BROOT_SIZE_ADJ \
- (XFS_BTREE_LBLOCK_LEN - sizeof(xfs_bmdr_block_t))
+#define XFS_BROOT_SIZE_ADJ(ip) \
+ (XFS_BMBT_BLOCK_LEN((ip)->i_mount) - sizeof(xfs_bmdr_block_t))
/*
* Inode data & attribute fork sizes, per inode.
diff --git a/include/xfs_ialloc_btree.h b/include/xfs_ialloc_btree.h
index 25c0239..a1bfa7a 100644
--- a/include/xfs_ialloc_btree.h
+++ b/include/xfs_ialloc_btree.h
@@ -29,7 +29,8 @@ struct xfs_mount;
/*
* There is a btree for the inode map per allocation group.
*/
-#define XFS_IBT_MAGIC 0x49414254 /* 'IABT' */
+#define XFS_IBT_MAGIC 0x49414254 /* 'IABT' */
+#define XFS_IBT_CRC_MAGIC 0x49414233 /* 'IAB3' */
typedef __uint64_t xfs_inofree_t;
#define XFS_INODES_PER_CHUNK (NBBY * sizeof(xfs_inofree_t))
@@ -76,10 +77,11 @@ typedef __be32 xfs_inobt_ptr_t;
/*
* Btree block header size depends on a superblock flag.
- *
- * (not quite yet, but soon)
*/
-#define XFS_INOBT_BLOCK_LEN(mp) XFS_BTREE_SBLOCK_LEN
+#define XFS_INOBT_BLOCK_LEN(mp) \
+ (xfs_sb_version_hascrc(&((mp)->m_sb)) ? \
+ XFS_BTREE_SBLOCK_LEN + XFS_BTREE_CRCBLOCK_ADD : \
+ XFS_BTREE_SBLOCK_LEN)
/*
* Record, key, and pointer address macros for btree blocks.
diff --git a/include/xfs_trans.h b/include/xfs_trans.h
index acf1381..a9bd826 100644
--- a/include/xfs_trans.h
+++ b/include/xfs_trans.h
@@ -500,6 +500,8 @@ void xfs_trans_inode_buf(xfs_trans_t *, struct xfs_buf *);
void xfs_trans_stale_inode_buf(xfs_trans_t *, struct xfs_buf *);
void xfs_trans_dquot_buf(xfs_trans_t *, struct xfs_buf *, uint);
void xfs_trans_inode_alloc_buf(xfs_trans_t *, struct xfs_buf *);
+void xfs_trans_buf_set_type(struct xfs_trans *, struct xfs_buf *,
+ uint);
void xfs_trans_ichgtime(struct xfs_trans *, struct xfs_inode *, int);
void xfs_trans_ijoin(struct xfs_trans *, struct xfs_inode *, uint);
void xfs_trans_log_buf(xfs_trans_t *, struct xfs_buf *, uint, uint);
diff --git a/libxfs/rdwr.c b/libxfs/rdwr.c
index e75edd0..e9cc7b1 100644
--- a/libxfs/rdwr.c
+++ b/libxfs/rdwr.c
@@ -323,17 +323,17 @@ libxfs_bcompare(struct cache_node *node, cache_key_t key)
#ifdef IO_BCOMPARE_CHECK
if (bp->b_dev == bkey->device &&
- bp->b_blkno == bkey->blkno &&
+ bp->b_bn == bkey->blkno &&
bp->b_bcount != BBTOB(bkey->bblen))
fprintf(stderr, "%lx: Badness in key lookup (length)\n"
"bp=(bno 0x%llx, len %u bytes) key=(bno 0x%llx, len %u bytes)\n",
pthread_self(),
- (unsigned long long)bp->b_blkno, (int)bp->b_bcount,
+ (unsigned long long)bp->b_bn, (int)bp->b_bcount,
(unsigned long long)bkey->blkno, BBTOB(bkey->bblen));
#endif
return (bp->b_dev == bkey->device &&
- bp->b_blkno == bkey->blkno &&
+ bp->b_bn == bkey->blkno &&
bp->b_bcount == BBTOB(bkey->bblen));
}
@@ -341,7 +341,7 @@ void
libxfs_bprint(xfs_buf_t *bp)
{
fprintf(stderr, "Buffer 0x%p blkno=%llu bytes=%u flags=0x%x count=%u\n",
- bp, (unsigned long long)bp->b_blkno, (unsigned)bp->b_bcount,
+ bp, (unsigned long long)bp->b_bn, (unsigned)bp->b_bcount,
bp->b_flags, bp->b_node.cn_count);
}
@@ -349,7 +349,7 @@ static void
__initbuf(xfs_buf_t *bp, dev_t device, xfs_daddr_t bno, unsigned int bytes)
{
bp->b_flags = 0;
- bp->b_blkno = bno;
+ bp->b_bn = bno;
bp->b_bcount = bytes;
bp->b_length = BTOBB(bytes);
bp->b_dev = device;
@@ -613,7 +613,7 @@ libxfs_purgebuf(xfs_buf_t *bp)
struct xfs_bufkey key = {0};
key.device = bp->b_dev;
- key.blkno = bp->b_blkno;
+ key.blkno = bp->b_bn;
key.bblen = bp->b_bcount >> BBSHIFT;
cache_node_purge(libxfs_bcache, &key, (struct cache_node *)bp);
@@ -669,7 +669,7 @@ libxfs_readbufr(dev_t dev, xfs_daddr_t blkno, xfs_buf_t *bp, int len, int flags)
error = __read_buf(fd, bp->b_addr, bytes, LIBXFS_BBTOOFF64(blkno), flags);
if (!error &&
bp->b_dev == dev &&
- bp->b_blkno == blkno &&
+ bp->b_bn == blkno &&
bp->b_bcount == bytes)
bp->b_flags |= LIBXFS_B_UPTODATE;
#ifdef IO_DEBUG
@@ -736,7 +736,7 @@ libxfs_readbuf_map(dev_t dev, struct xfs_buf_map *map, int nmaps, int flags)
#ifdef IO_DEBUG
printf("%lx: %s: read %lu bytes, error %d, blkno=%llu(%llu), %p\n",
pthread_self(), __FUNCTION__, buf - (char *)bp->b_addr, error,
- (long long)LIBXFS_BBTOOFF64(bp->b_blkno), (long long)bp->b_blkno, bp);
+ (long long)LIBXFS_BBTOOFF64(bp->b_bn), (long long)bp->b_bn, bp);
#endif
return bp;
}
@@ -772,7 +772,7 @@ libxfs_writebufr(xfs_buf_t *bp)
if (!(bp->b_flags & LIBXFS_B_DISCONTIG)) {
error = __write_buf(fd, bp->b_addr, bp->b_bcount,
- LIBXFS_BBTOOFF64(bp->b_blkno), bp->b_flags);
+ LIBXFS_BBTOOFF64(bp->b_bn), bp->b_flags);
} else {
int i;
char *buf = bp->b_addr;
@@ -794,8 +794,8 @@ libxfs_writebufr(xfs_buf_t *bp)
#ifdef IO_DEBUG
printf("%lx: %s: wrote %u bytes, blkno=%llu(%llu), %p\n",
pthread_self(), __FUNCTION__, bp->b_bcount,
- (long long)LIBXFS_BBTOOFF64(bp->b_blkno),
- (long long)bp->b_blkno, bp);
+ (long long)LIBXFS_BBTOOFF64(bp->b_bn),
+ (long long)bp->b_bn, bp);
#endif
if (!error) {
bp->b_flags |= LIBXFS_B_UPTODATE;
@@ -826,7 +826,7 @@ libxfs_iomove(xfs_buf_t *bp, uint boff, int len, void *data, int flags)
if (boff + len > bp->b_bcount) {
printf("Badness, iomove out of range!\n"
"bp=(bno 0x%llx, bytes %u) range=(boff %u, bytes %u)\n",
- (long long)bp->b_blkno, bp->b_bcount, boff, len);
+ (long long)bp->b_bn, bp->b_bcount, boff, len);
abort();
}
#endif
diff --git a/libxfs/xfs.h b/libxfs/xfs.h
index 9fbe261..b3b45bb 100644
--- a/libxfs/xfs.h
+++ b/libxfs/xfs.h
@@ -249,6 +249,7 @@ roundup_pow_of_two(uint v)
#define xfs_trans_agblocks_delta(tp, d)
#define xfs_trans_agflist_delta(tp, d)
#define xfs_trans_agbtree_delta(tp, d)
+#define xfs_trans_buf_set_type(tp, bp, t)
#define xfs_buf_readahead(a,b,c,ops) ((void) 0) /* no readahead */
#define xfs_buf_readahead_map(a,b,c,ops) ((void) 0) /* no readahead */
@@ -314,6 +315,9 @@ do { \
#define xfs_trans_unreserve_quota_nblks(t,i,b,n,f) ((void) 0)
#define xfs_qm_dqattach(i,f) (0)
+#define uuid_copy(s,d) platform_uuid_copy((s),(d))
+#define uuid_equal(s,d) (platform_uuid_compare((s),(d)) == 0)
+
/*
* Prototypes for kernel static functions that are aren't in their
* associated header files
diff --git a/libxfs/xfs_alloc_btree.c b/libxfs/xfs_alloc_btree.c
index dc9ed48..949c96d 100644
--- a/libxfs/xfs_alloc_btree.c
+++ b/libxfs/xfs_alloc_btree.c
@@ -254,7 +254,7 @@ xfs_allocbt_key_diff(
return (__int64_t)be32_to_cpu(kp->ar_startblock) - rec->ar_startblock;
}
-static void
+static bool
xfs_allocbt_verify(
struct xfs_buf *bp)
{
@@ -262,66 +262,98 @@ xfs_allocbt_verify(
struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
struct xfs_perag *pag = bp->b_pag;
unsigned int level;
- int sblock_ok; /* block passes checks */
/*
* magic number and level verification
*
- * During growfs operations, we can't verify the exact level as the
- * perag is not fully initialised and hence not attached to the buffer.
- * In this case, check against the maximum tree depth.
+ * During growfs operations, we can't verify the exact level or owner as
+ * the perag is not fully initialised and hence not attached to the
+ * buffer. In this case, check against the maximum tree depth.
*/
level = be16_to_cpu(block->bb_level);
switch (cpu_to_be32(block->bb_magic)) {
+ case XFS_ABTB_CRC_MAGIC:
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return false;
+ if (!uuid_equal(&block->bb_u.s.bb_uuid, &mp->m_sb.sb_uuid))
+ return false;
+ if (block->bb_u.s.bb_blkno != cpu_to_be64(bp->b_bn))
+ return false;
+ if (pag &&
+ be32_to_cpu(block->bb_u.s.bb_owner) != pag->pag_agno)
+ return false;
+ /* fall through */
case XFS_ABTB_MAGIC:
- if (pag)
- sblock_ok = level < pag->pagf_levels[XFS_BTNUM_BNOi];
- else
- sblock_ok = level < mp->m_ag_maxlevels;
+ if (pag) {
+ if (level >= pag->pagf_levels[XFS_BTNUM_BNOi])
+ return false;
+ } else if (level >= mp->m_ag_maxlevels)
+ return false;
break;
+ case XFS_ABTC_CRC_MAGIC:
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return false;
+ if (!uuid_equal(&block->bb_u.s.bb_uuid, &mp->m_sb.sb_uuid))
+ return false;
+ if (block->bb_u.s.bb_blkno != cpu_to_be64(bp->b_bn))
+ return false;
+ if (pag &&
+ be32_to_cpu(block->bb_u.s.bb_owner) != pag->pag_agno)
+ return false;
+ /* fall through */
case XFS_ABTC_MAGIC:
- if (pag)
- sblock_ok = level < pag->pagf_levels[XFS_BTNUM_CNTi];
- else
- sblock_ok = level < mp->m_ag_maxlevels;
+ if (pag) {
+ if (level >= pag->pagf_levels[XFS_BTNUM_CNTi])
+ return false;
+ } else if (level >= mp->m_ag_maxlevels)
+ return false;
break;
default:
- sblock_ok = 0;
- break;
+ return false;
}
/* numrecs verification */
- sblock_ok = sblock_ok &&
- be16_to_cpu(block->bb_numrecs) <= mp->m_alloc_mxr[level != 0];
+ if (be16_to_cpu(block->bb_numrecs) > mp->m_alloc_mxr[level != 0])
+ return false;
/* sibling pointer verification */
- sblock_ok = sblock_ok &&
- (block->bb_u.s.bb_leftsib == cpu_to_be32(NULLAGBLOCK) ||
- be32_to_cpu(block->bb_u.s.bb_leftsib) < mp->m_sb.sb_agblocks) &&
- block->bb_u.s.bb_leftsib &&
- (block->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK) ||
- be32_to_cpu(block->bb_u.s.bb_rightsib) < mp->m_sb.sb_agblocks) &&
- block->bb_u.s.bb_rightsib;
-
- if (!sblock_ok) {
- trace_xfs_btree_corrupt(bp, _RET_IP_);
- XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, block);
- xfs_buf_ioerror(bp, EFSCORRUPTED);
- }
+ if (!block->bb_u.s.bb_leftsib ||
+ (be32_to_cpu(block->bb_u.s.bb_leftsib) >= mp->m_sb.sb_agblocks &&
+ block->bb_u.s.bb_leftsib != cpu_to_be32(NULLAGBLOCK)))
+ return false;
+ if (!block->bb_u.s.bb_rightsib ||
+ (be32_to_cpu(block->bb_u.s.bb_rightsib) >= mp->m_sb.sb_agblocks &&
+ block->bb_u.s.bb_rightsib != cpu_to_be32(NULLAGBLOCK)))
+ return false;
+
+ return true;
}
static void
xfs_allocbt_read_verify(
struct xfs_buf *bp)
{
- xfs_allocbt_verify(bp);
+ if (!(xfs_btree_sblock_verify_crc(bp) &&
+ xfs_allocbt_verify(bp))) {
+ trace_xfs_btree_corrupt(bp, _RET_IP_);
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
+ bp->b_target->bt_mount, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
}
static void
xfs_allocbt_write_verify(
struct xfs_buf *bp)
{
- xfs_allocbt_verify(bp);
+ if (!xfs_allocbt_verify(bp)) {
+ trace_xfs_btree_corrupt(bp, _RET_IP_);
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
+ bp->b_target->bt_mount, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+ xfs_btree_sblock_calc_crc(bp);
+
}
const struct xfs_buf_ops xfs_allocbt_buf_ops = {
@@ -499,6 +531,9 @@ xfs_allocbt_init_cursor(
cur->bc_private.a.agbp = agbp;
cur->bc_private.a.agno = agno;
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
+
return cur;
}
diff --git a/libxfs/xfs_attr_leaf.c b/libxfs/xfs_attr_leaf.c
index 426130f..85cb31d 100644
--- a/libxfs/xfs_attr_leaf.c
+++ b/libxfs/xfs_attr_leaf.c
@@ -201,7 +201,7 @@ xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes)
return 0;
return dp->i_d.di_forkoff;
}
- dsize = XFS_BMAP_BROOT_SPACE(dp->i_df.if_broot);
+ dsize = XFS_BMAP_BROOT_SPACE(mp, dp->i_df.if_broot);
break;
}
diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index c8232a9..5e736a5 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -407,11 +407,15 @@ xfs_bmap_sanity_check(
{
struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
- if (block->bb_magic != cpu_to_be32(XFS_BMAP_MAGIC) ||
- be16_to_cpu(block->bb_level) != level ||
+ if (block->bb_magic != cpu_to_be32(XFS_BMAP_CRC_MAGIC) &&
+ block->bb_magic != cpu_to_be32(XFS_BMAP_MAGIC))
+ return 0;
+
+ if (be16_to_cpu(block->bb_level) != level ||
be16_to_cpu(block->bb_numrecs) == 0 ||
be16_to_cpu(block->bb_numrecs) > mp->m_bmap_dmxr[level != 0])
return 0;
+
return 1;
}
@@ -914,6 +918,7 @@ xfs_bmap_extents_to_btree(
xfs_extnum_t nextents; /* number of file extents */
xfs_bmbt_ptr_t *pp; /* root block address pointer */
+ mp = ip->i_mount;
ifp = XFS_IFORK_PTR(ip, whichfork);
ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS);
@@ -927,16 +932,18 @@ xfs_bmap_extents_to_btree(
* Fill in the root.
*/
block = ifp->if_broot;
- block->bb_magic = cpu_to_be32(XFS_BMAP_MAGIC);
- block->bb_level = cpu_to_be16(1);
- block->bb_numrecs = cpu_to_be16(1);
- block->bb_u.l.bb_leftsib = cpu_to_be64(NULLDFSBNO);
- block->bb_u.l.bb_rightsib = cpu_to_be64(NULLDFSBNO);
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ xfs_btree_init_block_int(mp, block, XFS_BUF_DADDR_NULL,
+ XFS_BMAP_CRC_MAGIC, 1, 1, ip->i_ino,
+ XFS_BTREE_LONG_PTRS | XFS_BTREE_CRC_BLOCKS);
+ else
+ xfs_btree_init_block_int(mp, block, XFS_BUF_DADDR_NULL,
+ XFS_BMAP_MAGIC, 1, 1, ip->i_ino,
+ XFS_BTREE_LONG_PTRS);
/*
* Need a cursor. Can't allocate until bb_level is filled in.
*/
- mp = ip->i_mount;
cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
cur->bc_private.b.firstblock = *firstblock;
cur->bc_private.b.flist = flist;
@@ -985,10 +992,15 @@ xfs_bmap_extents_to_btree(
*/
abp->b_ops = &xfs_bmbt_buf_ops;
ablock = XFS_BUF_TO_BLOCK(abp);
- ablock->bb_magic = cpu_to_be32(XFS_BMAP_MAGIC);
- ablock->bb_level = 0;
- ablock->bb_u.l.bb_leftsib = cpu_to_be64(NULLDFSBNO);
- ablock->bb_u.l.bb_rightsib = cpu_to_be64(NULLDFSBNO);
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ xfs_btree_init_block_int(mp, ablock, abp->b_bn,
+ XFS_BMAP_CRC_MAGIC, 0, 0, ip->i_ino,
+ XFS_BTREE_LONG_PTRS | XFS_BTREE_CRC_BLOCKS);
+ else
+ xfs_btree_init_block_int(mp, ablock, abp->b_bn,
+ XFS_BMAP_MAGIC, 0, 0, ip->i_ino,
+ XFS_BTREE_LONG_PTRS);
+
arp = XFS_BMBT_REC_ADDR(mp, ablock, 1);
nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
for (cnt = i = 0; i < nextents; i++) {
@@ -1016,8 +1028,8 @@ xfs_bmap_extents_to_btree(
* Do all this logging at the end so that
* the root is at the right level.
*/
- xfs_btree_log_block(cur, abp, XFS_BB_ALL_BITS);
xfs_btree_log_recs(cur, abp, 1, be16_to_cpu(ablock->bb_numrecs));
+ xfs_btree_log_block(cur, abp, XFS_BB_ALL_BITS);
ASSERT(*curp == NULL);
*curp = cur;
*logflagsp = XFS_ILOG_CORE | xfs_ilog_fbroot(whichfork);
@@ -1038,7 +1050,8 @@ xfs_bmap_local_to_extents(
xfs_extlen_t total, /* total blocks needed by transaction */
int *logflagsp, /* inode logging flags */
int whichfork,
- void (*init_fn)(struct xfs_buf *bp,
+ void (*init_fn)(struct xfs_trans *tp,
+ struct xfs_buf *bp,
struct xfs_inode *ip,
struct xfs_ifork *ifp))
{
@@ -1090,7 +1103,7 @@ xfs_bmap_local_to_extents(
bp = xfs_btree_get_bufl(args.mp, tp, args.fsbno, 0);
/* initialise the block and copy the data */
- init_fn(bp, ip, ifp);
+ init_fn(tp, bp, ip, ifp);
/* account for the change in fork size and log everything */
xfs_trans_log_buf(tp, bp, 0, ifp->if_bytes - 1);
@@ -1197,16 +1210,19 @@ xfs_bmap_add_attrfork_extents(
*/
STATIC void
xfs_bmap_local_to_extents_init_fn(
+ struct xfs_trans *tp,
struct xfs_buf *bp,
struct xfs_inode *ip,
struct xfs_ifork *ifp)
{
bp->b_ops = &xfs_bmbt_buf_ops;
memcpy(bp->b_addr, ifp->if_u1.if_data, ifp->if_bytes);
+ xfs_trans_buf_set_type(tp, bp, XFS_BLF_BTREE_BUF);
}
STATIC void
xfs_symlink_local_to_remote(
+ struct xfs_trans *tp,
struct xfs_buf *bp,
struct xfs_inode *ip,
struct xfs_ifork *ifp)
@@ -1225,8 +1241,7 @@ xfs_symlink_local_to_remote(
*
* XXX (dgc): investigate whether directory conversion can use the generic
* formatting callout. It should be possible - it's just a very complex
- * formatter. it would also require passing the transaction through to the init
- * function.
+ * formatter.
*/
STATIC int /* error */
xfs_bmap_add_attrfork_local(
diff --git a/libxfs/xfs_bmap_btree.c b/libxfs/xfs_bmap_btree.c
index 836f52f..473db4a 100644
--- a/libxfs/xfs_bmap_btree.c
+++ b/libxfs/xfs_bmap_btree.c
@@ -38,24 +38,31 @@ xfs_extent_state(
*/
void
xfs_bmdr_to_bmbt(
- struct xfs_mount *mp,
+ struct xfs_inode *ip,
xfs_bmdr_block_t *dblock,
int dblocklen,
struct xfs_btree_block *rblock,
int rblocklen)
{
+ struct xfs_mount *mp = ip->i_mount;
int dmxr;
xfs_bmbt_key_t *fkp;
__be64 *fpp;
xfs_bmbt_key_t *tkp;
__be64 *tpp;
- rblock->bb_magic = cpu_to_be32(XFS_BMAP_MAGIC);
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ xfs_btree_init_block_int(mp, rblock, XFS_BUF_DADDR_NULL,
+ XFS_BMAP_CRC_MAGIC, 0, 0, ip->i_ino,
+ XFS_BTREE_LONG_PTRS | XFS_BTREE_CRC_BLOCKS);
+ else
+ xfs_btree_init_block_int(mp, rblock, XFS_BUF_DADDR_NULL,
+ XFS_BMAP_MAGIC, 0, 0, ip->i_ino,
+ XFS_BTREE_LONG_PTRS);
+
rblock->bb_level = dblock->bb_level;
ASSERT(be16_to_cpu(rblock->bb_level) > 0);
rblock->bb_numrecs = dblock->bb_numrecs;
- rblock->bb_u.l.bb_leftsib = cpu_to_be64(NULLDFSBNO);
- rblock->bb_u.l.bb_rightsib = cpu_to_be64(NULLDFSBNO);
dmxr = xfs_bmdr_maxrecs(mp, dblocklen, 0);
fkp = XFS_BMDR_KEY_ADDR(dblock, 1);
tkp = XFS_BMBT_KEY_ADDR(mp, rblock, 1);
@@ -403,7 +410,13 @@ xfs_bmbt_to_bmdr(
xfs_bmbt_key_t *tkp;
__be64 *tpp;
- ASSERT(rblock->bb_magic == cpu_to_be32(XFS_BMAP_MAGIC));
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ ASSERT(rblock->bb_magic == cpu_to_be32(XFS_BMAP_CRC_MAGIC));
+ ASSERT(uuid_equal(&rblock->bb_u.l.bb_uuid, &mp->m_sb.sb_uuid));
+ ASSERT(rblock->bb_u.l.bb_blkno ==
+ cpu_to_be64(XFS_BUF_DADDR_NULL));
+ } else
+ ASSERT(rblock->bb_magic == cpu_to_be32(XFS_BMAP_MAGIC));
ASSERT(rblock->bb_u.l.bb_leftsib == cpu_to_be64(NULLDFSBNO));
ASSERT(rblock->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO));
ASSERT(rblock->bb_level != 0);
@@ -687,45 +700,59 @@ xfs_bmbt_key_diff(
cur->bc_rec.b.br_startoff;
}
-static void
+static bool
xfs_bmbt_verify(
struct xfs_buf *bp)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
unsigned int level;
- int lblock_ok; /* block passes checks */
- /* magic number and level verification.
+ switch (be32_to_cpu(block->bb_magic)) {
+ case XFS_BMAP_CRC_MAGIC:
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return false;
+ if (!uuid_equal(&block->bb_u.l.bb_uuid, &mp->m_sb.sb_uuid))
+ return false;
+ if (block->bb_u.l.bb_blkno != cpu_to_be64(bp->b_bn))
+ return false;
+ /*
+ * XXX: need a better way of verifying the owner here. Right now
+ * just make sure there has been one set.
+ */
+ if (be64_to_cpu(block->bb_u.l.bb_owner) == 0)
+ return false;
+ /* fall through */
+ case XFS_BMAP_MAGIC:
+ break;
+ default:
+ return false;
+ }
+
+ /*
+ * numrecs and level verification.
*
- * We don't know waht fork we belong to, so just verify that the level
+ * We don't know what fork we belong to, so just verify that the level
* is less than the maximum of the two. Later checks will be more
* precise.
*/
level = be16_to_cpu(block->bb_level);
- lblock_ok = block->bb_magic == cpu_to_be32(XFS_BMAP_MAGIC) &&
- level < MAX(mp->m_bm_maxlevels[0], mp->m_bm_maxlevels[1]);
-
- /* numrecs verification */
- lblock_ok = lblock_ok &&
- be16_to_cpu(block->bb_numrecs) <= mp->m_bmap_dmxr[level != 0];
+ if (level > MAX(mp->m_bm_maxlevels[0], mp->m_bm_maxlevels[1]))
+ return false;
+ if (be16_to_cpu(block->bb_numrecs) > mp->m_bmap_dmxr[level != 0])
+ return false;
/* sibling pointer verification */
- lblock_ok = lblock_ok &&
- block->bb_u.l.bb_leftsib &&
- (block->bb_u.l.bb_leftsib == cpu_to_be64(NULLDFSBNO) ||
- XFS_FSB_SANITY_CHECK(mp,
- be64_to_cpu(block->bb_u.l.bb_leftsib))) &&
- block->bb_u.l.bb_rightsib &&
- (block->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO) ||
- XFS_FSB_SANITY_CHECK(mp,
- be64_to_cpu(block->bb_u.l.bb_rightsib)));
-
- if (!lblock_ok) {
- trace_xfs_btree_corrupt(bp, _RET_IP_);
- XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, block);
- xfs_buf_ioerror(bp, EFSCORRUPTED);
- }
+ if (!block->bb_u.l.bb_leftsib ||
+ (block->bb_u.l.bb_leftsib != cpu_to_be64(NULLDFSBNO) &&
+ !XFS_FSB_SANITY_CHECK(mp, be64_to_cpu(block->bb_u.l.bb_leftsib))))
+ return false;
+ if (!block->bb_u.l.bb_rightsib ||
+ (block->bb_u.l.bb_rightsib != cpu_to_be64(NULLDFSBNO) &&
+ !XFS_FSB_SANITY_CHECK(mp, be64_to_cpu(block->bb_u.l.bb_rightsib))))
+ return false;
+
+ return true;
}
static void
@@ -733,13 +760,29 @@ xfs_bmbt_read_verify(
struct xfs_buf *bp)
{
xfs_bmbt_verify(bp);
+ if (!(xfs_btree_lblock_verify_crc(bp) &&
+ xfs_bmbt_verify(bp))) {
+ trace_xfs_btree_corrupt(bp, _RET_IP_);
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
+ bp->b_target->bt_mount, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+
}
static void
xfs_bmbt_write_verify(
struct xfs_buf *bp)
{
- xfs_bmbt_verify(bp);
+ if (!xfs_bmbt_verify(bp)) {
+ xfs_warn(bp->b_target->bt_mount, "bmbt daddr 0x%llx failed", bp->b_bn);
+ trace_xfs_btree_corrupt(bp, _RET_IP_);
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
+ bp->b_target->bt_mount, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ return;
+ }
+ xfs_btree_lblock_calc_crc(bp);
}
const struct xfs_buf_ops xfs_bmbt_buf_ops = {
@@ -913,6 +956,8 @@ xfs_bmbt_init_cursor(
cur->bc_ops = &xfs_bmbt_ops;
cur->bc_flags = XFS_BTREE_LONG_PTRS | XFS_BTREE_ROOT_IN_INODE;
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
cur->bc_private.b.forksize = XFS_IFORK_SIZE(ip, whichfork);
cur->bc_private.b.ip = ip;
diff --git a/libxfs/xfs_btree.c b/libxfs/xfs_btree.c
index c35269b..a7c19e9 100644
--- a/libxfs/xfs_btree.c
+++ b/libxfs/xfs_btree.c
@@ -26,9 +26,13 @@ kmem_zone_t *xfs_btree_cur_zone;
/*
* Btree magic numbers.
*/
-const __uint32_t xfs_magics[XFS_BTNUM_MAX] = {
- XFS_ABTB_MAGIC, XFS_ABTC_MAGIC, XFS_BMAP_MAGIC, XFS_IBT_MAGIC
+static const __uint32_t xfs_magics[2][XFS_BTNUM_MAX] = {
+ { XFS_ABTB_MAGIC, XFS_ABTC_MAGIC, XFS_BMAP_MAGIC, XFS_IBT_MAGIC },
+ { XFS_ABTB_CRC_MAGIC, XFS_ABTC_CRC_MAGIC,
+ XFS_BMAP_CRC_MAGIC, XFS_IBT_CRC_MAGIC }
};
+#define xfs_btree_magic(cur) \
+ xfs_magics[!!((cur)->bc_flags & XFS_BTREE_CRC_BLOCKS)][cur->bc_btnum]
STATIC int /* error (0 or EFSCORRUPTED) */
@@ -38,30 +42,38 @@ xfs_btree_check_lblock(
int level, /* level of the btree block */
struct xfs_buf *bp) /* buffer for block, if any */
{
- int lblock_ok; /* block passes checks */
+ int lblock_ok = 1; /* block passes checks */
struct xfs_mount *mp; /* file system mount point */
mp = cur->bc_mp;
- lblock_ok =
- be32_to_cpu(block->bb_magic) == xfs_magics[cur->bc_btnum] &&
+
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ lblock_ok = lblock_ok &&
+ uuid_equal(&block->bb_u.l.bb_uuid, &mp->m_sb.sb_uuid) &&
+ block->bb_u.l.bb_blkno == cpu_to_be64(
+ bp ? bp->b_bn : XFS_BUF_DADDR_NULL);
+ }
+
+ lblock_ok = lblock_ok &&
+ be32_to_cpu(block->bb_magic) == xfs_btree_magic(cur) &&
be16_to_cpu(block->bb_level) == level &&
be16_to_cpu(block->bb_numrecs) <=
cur->bc_ops->get_maxrecs(cur, level) &&
block->bb_u.l.bb_leftsib &&
(block->bb_u.l.bb_leftsib == cpu_to_be64(NULLDFSBNO) ||
XFS_FSB_SANITY_CHECK(mp,
- be64_to_cpu(block->bb_u.l.bb_leftsib))) &&
+ be64_to_cpu(block->bb_u.l.bb_leftsib))) &&
block->bb_u.l.bb_rightsib &&
(block->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO) ||
XFS_FSB_SANITY_CHECK(mp,
- be64_to_cpu(block->bb_u.l.bb_rightsib)));
+ be64_to_cpu(block->bb_u.l.bb_rightsib)));
+
if (unlikely(XFS_TEST_ERROR(!lblock_ok, mp,
XFS_ERRTAG_BTREE_CHECK_LBLOCK,
XFS_RANDOM_BTREE_CHECK_LBLOCK))) {
if (bp)
trace_xfs_btree_corrupt(bp, _RET_IP_);
- XFS_ERROR_REPORT("xfs_btree_check_lblock", XFS_ERRLEVEL_LOW,
- mp);
+ XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
return XFS_ERROR(EFSCORRUPTED);
}
return 0;
@@ -74,16 +86,26 @@ xfs_btree_check_sblock(
int level, /* level of the btree block */
struct xfs_buf *bp) /* buffer containing block */
{
+ struct xfs_mount *mp; /* file system mount point */
struct xfs_buf *agbp; /* buffer for ag. freespace struct */
struct xfs_agf *agf; /* ag. freespace structure */
xfs_agblock_t agflen; /* native ag. freespace length */
- int sblock_ok; /* block passes checks */
+ int sblock_ok = 1; /* block passes checks */
+ mp = cur->bc_mp;
agbp = cur->bc_private.a.agbp;
agf = XFS_BUF_TO_AGF(agbp);
agflen = be32_to_cpu(agf->agf_length);
- sblock_ok =
- be32_to_cpu(block->bb_magic) == xfs_magics[cur->bc_btnum] &&
+
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ sblock_ok = sblock_ok &&
+ uuid_equal(&block->bb_u.s.bb_uuid, &mp->m_sb.sb_uuid) &&
+ block->bb_u.s.bb_blkno == cpu_to_be64(
+ bp ? bp->b_bn : XFS_BUF_DADDR_NULL);
+ }
+
+ sblock_ok = sblock_ok &&
+ be32_to_cpu(block->bb_magic) == xfs_btree_magic(cur) &&
be16_to_cpu(block->bb_level) == level &&
be16_to_cpu(block->bb_numrecs) <=
cur->bc_ops->get_maxrecs(cur, level) &&
@@ -93,13 +115,13 @@ xfs_btree_check_sblock(
(block->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK) ||
be32_to_cpu(block->bb_u.s.bb_rightsib) < agflen) &&
block->bb_u.s.bb_rightsib;
- if (unlikely(XFS_TEST_ERROR(!sblock_ok, cur->bc_mp,
+
+ if (unlikely(XFS_TEST_ERROR(!sblock_ok, mp,
XFS_ERRTAG_BTREE_CHECK_SBLOCK,
XFS_RANDOM_BTREE_CHECK_SBLOCK))) {
if (bp)
trace_xfs_btree_corrupt(bp, _RET_IP_);
- XFS_CORRUPTION_ERROR("xfs_btree_check_sblock",
- XFS_ERRLEVEL_LOW, cur->bc_mp, block);
+ XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
return XFS_ERROR(EFSCORRUPTED);
}
return 0;
@@ -178,6 +200,72 @@ xfs_btree_check_ptr(
#endif
/*
+ * Calculate CRC on the whole btree block and stuff it into the
+ * long-form btree header.
+ *
+ * Prior to calculting the CRC, pull the LSN out of the buffer log item and put
+ * it into the buffer so recovery knows what the last modifcation was that made
+ * it to disk.
+ */
+void
+xfs_btree_lblock_calc_crc(
+ struct xfs_buf *bp)
+{
+ struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
+ struct xfs_buf_log_item *bip = bp->b_fspriv;
+
+ if (!xfs_sb_version_hascrc(&bp->b_target->bt_mount->m_sb))
+ return;
+ if (bip)
+ block->bb_u.l.bb_lsn = cpu_to_be64(bip->bli_item.li_lsn);
+ xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length),
+ XFS_BTREE_LBLOCK_CRC_OFF);
+}
+
+bool
+xfs_btree_lblock_verify_crc(
+ struct xfs_buf *bp)
+{
+ if (xfs_sb_version_hascrc(&bp->b_target->bt_mount->m_sb))
+ return xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
+ XFS_BTREE_LBLOCK_CRC_OFF);
+ return true;
+}
+
+/*
+ * Calculate CRC on the whole btree block and stuff it into the
+ * short-form btree header.
+ *
+ * Prior to calculting the CRC, pull the LSN out of the buffer log item and put
+ * it into the buffer so recovery knows what the last modifcation was that made
+ * it to disk.
+ */
+void
+xfs_btree_sblock_calc_crc(
+ struct xfs_buf *bp)
+{
+ struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
+ struct xfs_buf_log_item *bip = bp->b_fspriv;
+
+ if (!xfs_sb_version_hascrc(&bp->b_target->bt_mount->m_sb))
+ return;
+ if (bip)
+ block->bb_u.s.bb_lsn = cpu_to_be64(bip->bli_item.li_lsn);
+ xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length),
+ XFS_BTREE_SBLOCK_CRC_OFF);
+}
+
+bool
+xfs_btree_sblock_verify_crc(
+ struct xfs_buf *bp)
+{
+ if (xfs_sb_version_hascrc(&bp->b_target->bt_mount->m_sb))
+ return xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
+ XFS_BTREE_SBLOCK_CRC_OFF);
+ return true;
+}
+
+/*
* Delete the btree cursor.
*/
void
@@ -261,10 +349,8 @@ xfs_btree_dup_cursor(
*ncur = NULL;
return error;
}
- new->bc_bufs[i] = bp;
- ASSERT(!xfs_buf_geterror(bp));
- } else
- new->bc_bufs[i] = NULL;
+ }
+ new->bc_bufs[i] = bp;
}
*ncur = new;
return 0;
@@ -305,9 +391,17 @@ xfs_btree_dup_cursor(
*/
static inline size_t xfs_btree_block_len(struct xfs_btree_cur *cur)
{
- return (cur->bc_flags & XFS_BTREE_LONG_PTRS) ?
- XFS_BTREE_LBLOCK_LEN :
- XFS_BTREE_SBLOCK_LEN;
+ size_t len;
+
+ if (cur->bc_flags & XFS_BTREE_LONG_PTRS)
+ len = XFS_BTREE_LBLOCK_LEN;
+ else
+ len = XFS_BTREE_SBLOCK_LEN;
+
+ if (cur->bc_flags & XFS_BTREE_CRC_BLOCKS)
+ len += XFS_BTREE_CRCBLOCK_ADD;
+
+ return len;
}
/*
@@ -807,43 +901,85 @@ xfs_btree_set_sibling(
}
void
+xfs_btree_init_block_int(
+ struct xfs_mount *mp,
+ struct xfs_btree_block *buf,
+ xfs_daddr_t blkno,
+ __u32 magic,
+ __u16 level,
+ __u16 numrecs,
+ __u64 owner,
+ unsigned int flags)
+{
+ buf->bb_magic = cpu_to_be32(magic);
+ buf->bb_level = cpu_to_be16(level);
+ buf->bb_numrecs = cpu_to_be16(numrecs);
+
+ if (flags & XFS_BTREE_LONG_PTRS) {
+ buf->bb_u.l.bb_leftsib = cpu_to_be64(NULLDFSBNO);
+ buf->bb_u.l.bb_rightsib = cpu_to_be64(NULLDFSBNO);
+ if (flags & XFS_BTREE_CRC_BLOCKS) {
+ buf->bb_u.l.bb_blkno = cpu_to_be64(blkno);
+ buf->bb_u.l.bb_owner = cpu_to_be64(owner);
+ uuid_copy(&buf->bb_u.l.bb_uuid, &mp->m_sb.sb_uuid);
+ buf->bb_u.l.bb_pad = 0;
+ }
+ } else {
+ /* owner is a 32 bit value on short blocks */
+ __u32 __owner = (__u32)owner;
+
+ buf->bb_u.s.bb_leftsib = cpu_to_be32(NULLAGBLOCK);
+ buf->bb_u.s.bb_rightsib = cpu_to_be32(NULLAGBLOCK);
+ if (flags & XFS_BTREE_CRC_BLOCKS) {
+ buf->bb_u.s.bb_blkno = cpu_to_be64(blkno);
+ buf->bb_u.s.bb_owner = cpu_to_be32(__owner);
+ uuid_copy(&buf->bb_u.s.bb_uuid, &mp->m_sb.sb_uuid);
+ }
+ }
+}
+
+void
xfs_btree_init_block(
struct xfs_mount *mp,
struct xfs_buf *bp,
__u32 magic,
__u16 level,
__u16 numrecs,
+ __u64 owner,
unsigned int flags)
{
- struct xfs_btree_block *new = XFS_BUF_TO_BLOCK(bp);
-
- new->bb_magic = cpu_to_be32(magic);
- new->bb_level = cpu_to_be16(level);
- new->bb_numrecs = cpu_to_be16(numrecs);
-
- if (flags & XFS_BTREE_LONG_PTRS) {
- new->bb_u.l.bb_leftsib = cpu_to_be64(NULLDFSBNO);
- new->bb_u.l.bb_rightsib = cpu_to_be64(NULLDFSBNO);
- } else {
- new->bb_u.s.bb_leftsib = cpu_to_be32(NULLAGBLOCK);
- new->bb_u.s.bb_rightsib = cpu_to_be32(NULLAGBLOCK);
- }
+ xfs_btree_init_block_int(mp, XFS_BUF_TO_BLOCK(bp), bp->b_bn,
+ magic, level, numrecs, owner, flags);
}
STATIC void
xfs_btree_init_block_cur(
struct xfs_btree_cur *cur,
+ struct xfs_buf *bp,
int level,
- int numrecs,
- struct xfs_buf *bp)
+ int numrecs)
{
- xfs_btree_init_block(cur->bc_mp, bp, xfs_magics[cur->bc_btnum],
- level, numrecs, cur->bc_flags);
+ __u64 owner;
+
+ /*
+ * we can pull the owner from the cursor right now as the different
+ * owners align directly with the pointer size of the btree. This may
+ * change in future, but is safe for current users of the generic btree
+ * code.
+ */
+ if (cur->bc_flags & XFS_BTREE_LONG_PTRS)
+ owner = cur->bc_private.b.ip->i_ino;
+ else
+ owner = cur->bc_private.a.agno;
+
+ xfs_btree_init_block_int(cur->bc_mp, XFS_BUF_TO_BLOCK(bp), bp->b_bn,
+ xfs_btree_magic(cur), level, numrecs,
+ owner, cur->bc_flags);
}
/*
* Return true if ptr is the last record in the btree and
- * we need to track updateѕ to this record. The decision
+ * we need to track updates to this record. The decision
* will be further refined in the update_lastrec method.
*/
STATIC int
@@ -1091,6 +1227,7 @@ xfs_btree_log_keys(
XFS_BTREE_TRACE_ARGBII(cur, bp, first, last);
if (bp) {
+ xfs_trans_buf_set_type(cur->bc_tp, bp, XFS_BLF_BTREE_BUF);
xfs_trans_log_buf(cur->bc_tp, bp,
xfs_btree_key_offset(cur, first),
xfs_btree_key_offset(cur, last + 1) - 1);
@@ -1115,6 +1252,7 @@ xfs_btree_log_recs(
XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY);
XFS_BTREE_TRACE_ARGBII(cur, bp, first, last);
+ xfs_trans_buf_set_type(cur->bc_tp, bp, XFS_BLF_BTREE_BUF);
xfs_trans_log_buf(cur->bc_tp, bp,
xfs_btree_rec_offset(cur, first),
xfs_btree_rec_offset(cur, last + 1) - 1);
@@ -1139,6 +1277,7 @@ xfs_btree_log_ptrs(
struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
int level = xfs_btree_get_level(block);
+ xfs_trans_buf_set_type(cur->bc_tp, bp, XFS_BLF_BTREE_BUF);
xfs_trans_log_buf(cur->bc_tp, bp,
xfs_btree_ptr_offset(cur, first, level),
xfs_btree_ptr_offset(cur, last + 1, level) - 1);
@@ -1167,7 +1306,12 @@ xfs_btree_log_block(
offsetof(struct xfs_btree_block, bb_numrecs),
offsetof(struct xfs_btree_block, bb_u.s.bb_leftsib),
offsetof(struct xfs_btree_block, bb_u.s.bb_rightsib),
- XFS_BTREE_SBLOCK_LEN
+ offsetof(struct xfs_btree_block, bb_u.s.bb_blkno),
+ offsetof(struct xfs_btree_block, bb_u.s.bb_lsn),
+ offsetof(struct xfs_btree_block, bb_u.s.bb_uuid),
+ offsetof(struct xfs_btree_block, bb_u.s.bb_owner),
+ offsetof(struct xfs_btree_block, bb_u.s.bb_crc),
+ XFS_BTREE_SBLOCK_LEN + XFS_BTREE_CRCBLOCK_ADD
};
static const short loffsets[] = { /* table of offsets (long) */
offsetof(struct xfs_btree_block, bb_magic),
@@ -1175,17 +1319,40 @@ xfs_btree_log_block(
offsetof(struct xfs_btree_block, bb_numrecs),
offsetof(struct xfs_btree_block, bb_u.l.bb_leftsib),
offsetof(struct xfs_btree_block, bb_u.l.bb_rightsib),
- XFS_BTREE_LBLOCK_LEN
+ offsetof(struct xfs_btree_block, bb_u.l.bb_blkno),
+ offsetof(struct xfs_btree_block, bb_u.l.bb_lsn),
+ offsetof(struct xfs_btree_block, bb_u.l.bb_uuid),
+ offsetof(struct xfs_btree_block, bb_u.l.bb_owner),
+ offsetof(struct xfs_btree_block, bb_u.l.bb_crc),
+ offsetof(struct xfs_btree_block, bb_u.l.bb_pad),
+ XFS_BTREE_LBLOCK_LEN + XFS_BTREE_CRCBLOCK_ADD
};
XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY);
XFS_BTREE_TRACE_ARGBI(cur, bp, fields);
if (bp) {
+ int nbits;
+
+ if (cur->bc_flags & XFS_BTREE_CRC_BLOCKS) {
+ /*
+ * We don't log the CRC when updating a btree
+ * block but instead recreate it during log
+ * recovery. As the log buffers have checksums
+ * of their this is safe and avoids logging a crc
+ * update in a lot of places.
+ */
+ if (fields == XFS_BB_ALL_BITS)
+ fields = XFS_BB_ALL_BITS_CRC;
+ nbits = XFS_BB_NUM_BITS_CRC;
+ } else {
+ nbits = XFS_BB_NUM_BITS;
+ }
xfs_btree_offsets(fields,
(cur->bc_flags & XFS_BTREE_LONG_PTRS) ?
loffsets : soffsets,
- XFS_BB_NUM_BITS, &first, &last);
+ nbits, &first, &last);
+ xfs_trans_buf_set_type(cur->bc_tp, bp, XFS_BLF_BTREE_BUF);
xfs_trans_log_buf(cur->bc_tp, bp, first, last);
} else {
xfs_trans_log_inode(cur->bc_tp, cur->bc_private.b.ip,
@@ -2148,7 +2315,7 @@ xfs_btree_split(
goto error0;
/* Fill in the btree header for the new right block. */
- xfs_btree_init_block_cur(cur, xfs_btree_get_level(left), 0, rbp);
+ xfs_btree_init_block_cur(cur, rbp, xfs_btree_get_level(left), 0);
/*
* Split the entries between the old and the new block evenly.
@@ -2457,7 +2624,7 @@ xfs_btree_new_root(
nptr = 2;
}
/* Fill in the new block's btree header and log it. */
- xfs_btree_init_block_cur(cur, cur->bc_nlevels, 2, nbp);
+ xfs_btree_init_block_cur(cur, nbp, cur->bc_nlevels, 2);
xfs_btree_log_block(cur, nbp, XFS_BB_ALL_BITS);
ASSERT(!xfs_btree_ptr_is_null(cur, &lptr) &&
!xfs_btree_ptr_is_null(cur, &rptr));
diff --git a/libxfs/xfs_ialloc_btree.c b/libxfs/xfs_ialloc_btree.c
index 0bc24cc..ee036bf 100644
--- a/libxfs/xfs_ialloc_btree.c
+++ b/libxfs/xfs_ialloc_btree.c
@@ -163,52 +163,82 @@ xfs_inobt_key_diff(
cur->bc_rec.i.ir_startino;
}
-void
+static int
xfs_inobt_verify(
struct xfs_buf *bp)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
+ struct xfs_perag *pag = bp->b_pag;
unsigned int level;
- int sblock_ok; /* block passes checks */
- /* magic number and level verification */
- level = be16_to_cpu(block->bb_level);
- sblock_ok = block->bb_magic == cpu_to_be32(XFS_IBT_MAGIC) &&
- level < mp->m_in_maxlevels;
+ /*
+ * During growfs operations, we can't verify the exact owner as the
+ * perag is not fully initialised and hence not attached to the buffer.
+ */
+ switch (be32_to_cpu(block->bb_magic)) {
+ case XFS_IBT_CRC_MAGIC:
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return false;
+ if (!uuid_equal(&block->bb_u.s.bb_uuid, &mp->m_sb.sb_uuid))
+ return false;
+ if (block->bb_u.s.bb_blkno != cpu_to_be64(bp->b_bn))
+ return false;
+ if (pag &&
+ be32_to_cpu(block->bb_u.s.bb_owner) != pag->pag_agno)
+ return false;
+ /* fall through */
+ case XFS_IBT_MAGIC:
+ break;
+ default:
+ return 0;
+ }
- /* numrecs verification */
- sblock_ok = sblock_ok &&
- be16_to_cpu(block->bb_numrecs) <= mp->m_inobt_mxr[level != 0];
+ /* numrecs and level verification */
+ level = be16_to_cpu(block->bb_level);
+ if (level >= mp->m_in_maxlevels)
+ return false;
+ if (be16_to_cpu(block->bb_numrecs) > mp->m_inobt_mxr[level != 0])
+ return false;
/* sibling pointer verification */
- sblock_ok = sblock_ok &&
- (block->bb_u.s.bb_leftsib == cpu_to_be32(NULLAGBLOCK) ||
- be32_to_cpu(block->bb_u.s.bb_leftsib) < mp->m_sb.sb_agblocks) &&
- block->bb_u.s.bb_leftsib &&
- (block->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK) ||
- be32_to_cpu(block->bb_u.s.bb_rightsib) < mp->m_sb.sb_agblocks) &&
- block->bb_u.s.bb_rightsib;
-
- if (!sblock_ok) {
- trace_xfs_btree_corrupt(bp, _RET_IP_);
- XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, block);
- xfs_buf_ioerror(bp, EFSCORRUPTED);
- }
+ if (!block->bb_u.s.bb_leftsib ||
+ (be32_to_cpu(block->bb_u.s.bb_leftsib) >= mp->m_sb.sb_agblocks &&
+ block->bb_u.s.bb_leftsib != cpu_to_be32(NULLAGBLOCK)))
+ return false;
+ if (!block->bb_u.s.bb_rightsib ||
+ (be32_to_cpu(block->bb_u.s.bb_rightsib) >= mp->m_sb.sb_agblocks &&
+ block->bb_u.s.bb_rightsib != cpu_to_be32(NULLAGBLOCK)))
+ return false;
+
+ return true;
}
static void
xfs_inobt_read_verify(
struct xfs_buf *bp)
{
- xfs_inobt_verify(bp);
+ if (!(xfs_btree_sblock_verify_crc(bp) &&
+ xfs_inobt_verify(bp))) {
+ trace_xfs_btree_corrupt(bp, _RET_IP_);
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
+ bp->b_target->bt_mount, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
}
static void
xfs_inobt_write_verify(
struct xfs_buf *bp)
{
- xfs_inobt_verify(bp);
+ if (!xfs_inobt_verify(bp)) {
+ trace_xfs_btree_corrupt(bp, _RET_IP_);
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
+ bp->b_target->bt_mount, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+ xfs_btree_sblock_calc_crc(bp);
+
}
const struct xfs_buf_ops xfs_inobt_buf_ops = {
@@ -355,6 +385,8 @@ xfs_inobt_init_cursor(
cur->bc_blocklog = mp->m_sb.sb_blocklog;
cur->bc_ops = &xfs_inobt_ops;
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
cur->bc_private.a.agbp = agbp;
cur->bc_private.a.agno = agno;
diff --git a/libxfs/xfs_inode.c b/libxfs/xfs_inode.c
index 3cf2423..f9f792c 100644
--- a/libxfs/xfs_inode.c
+++ b/libxfs/xfs_inode.c
@@ -492,6 +492,7 @@ xfs_iformat_btree(
xfs_dinode_t *dip,
int whichfork)
{
+ struct xfs_mount *mp = ip->i_mount;
xfs_bmdr_block_t *dfp;
xfs_ifork_t *ifp;
/* REFERENCED */
@@ -500,7 +501,7 @@ xfs_iformat_btree(
ifp = XFS_IFORK_PTR(ip, whichfork);
dfp = (xfs_bmdr_block_t *)XFS_DFORK_PTR(dip, whichfork);
- size = XFS_BMAP_BROOT_SPACE(dfp);
+ size = XFS_BMAP_BROOT_SPACE(mp, dfp);
nrecs = be16_to_cpu(dfp->bb_numrecs);
/*
@@ -511,14 +512,14 @@ xfs_iformat_btree(
* blocks.
*/
if (unlikely(XFS_IFORK_NEXTENTS(ip, whichfork) <=
- XFS_IFORK_MAXEXT(ip, whichfork) ||
+ XFS_IFORK_MAXEXT(ip, whichfork) ||
XFS_BMDR_SPACE_CALC(nrecs) >
- XFS_DFORK_SIZE(dip, ip->i_mount, whichfork) ||
+ XFS_DFORK_SIZE(dip, mp, whichfork) ||
XFS_IFORK_NEXTENTS(ip, whichfork) > ip->i_d.di_nblocks)) {
- xfs_warn(ip->i_mount, "corrupt inode %Lu (btree).",
- (unsigned long long) ip->i_ino);
+ xfs_warn(mp, "corrupt inode %Lu (btree).",
+ (unsigned long long) ip->i_ino);
XFS_CORRUPTION_ERROR("xfs_iformat_btree", XFS_ERRLEVEL_LOW,
- ip->i_mount, dip);
+ mp, dip);
return XFS_ERROR(EFSCORRUPTED);
}
@@ -529,8 +530,7 @@ xfs_iformat_btree(
* Copy and convert from the on-disk structure
* to the in-memory structure.
*/
- xfs_bmdr_to_bmbt(ip->i_mount, dfp,
- XFS_DFORK_SIZE(dip, ip->i_mount, whichfork),
+ xfs_bmdr_to_bmbt(ip, dfp, XFS_DFORK_SIZE(dip, ip->i_mount, whichfork),
ifp->if_broot, size);
ifp->if_flags &= ~XFS_IFEXTENTS;
ifp->if_flags |= XFS_IFBROOT;
@@ -813,7 +813,7 @@ xfs_iroot_realloc(
* allocate it now and get out.
*/
if (ifp->if_broot_bytes == 0) {
- new_size = (size_t)XFS_BMAP_BROOT_SPACE_CALC(rec_diff);
+ new_size = XFS_BMAP_BROOT_SPACE_CALC(mp, rec_diff);
ifp->if_broot = kmem_alloc(new_size, KM_SLEEP | KM_NOFS);
ifp->if_broot_bytes = (int)new_size;
return;
@@ -827,9 +827,9 @@ xfs_iroot_realloc(
*/
cur_max = xfs_bmbt_maxrecs(mp, ifp->if_broot_bytes, 0);
new_max = cur_max + rec_diff;
- new_size = (size_t)XFS_BMAP_BROOT_SPACE_CALC(new_max);
+ new_size = XFS_BMAP_BROOT_SPACE_CALC(mp, new_max);
ifp->if_broot = kmem_realloc(ifp->if_broot, new_size,
- (size_t)XFS_BMAP_BROOT_SPACE_CALC(cur_max), /* old size */
+ XFS_BMAP_BROOT_SPACE_CALC(mp, cur_max),
KM_SLEEP | KM_NOFS);
op = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, ifp->if_broot, 1,
ifp->if_broot_bytes);
@@ -837,7 +837,7 @@ xfs_iroot_realloc(
(int)new_size);
ifp->if_broot_bytes = (int)new_size;
ASSERT(ifp->if_broot_bytes <=
- XFS_IFORK_SIZE(ip, whichfork) + XFS_BROOT_SIZE_ADJ);
+ XFS_IFORK_SIZE(ip, whichfork) + XFS_BROOT_SIZE_ADJ(ip));
memmove(np, op, cur_max * (uint)sizeof(xfs_dfsbno_t));
return;
}
@@ -852,7 +852,7 @@ xfs_iroot_realloc(
new_max = cur_max + rec_diff;
ASSERT(new_max >= 0);
if (new_max > 0)
- new_size = (size_t)XFS_BMAP_BROOT_SPACE_CALC(new_max);
+ new_size = XFS_BMAP_BROOT_SPACE_CALC(mp, new_max);
else
new_size = 0;
if (new_size > 0) {
@@ -860,7 +860,8 @@ xfs_iroot_realloc(
/*
* First copy over the btree block header.
*/
- memcpy(new_broot, ifp->if_broot, XFS_BTREE_LBLOCK_LEN);
+ memcpy(new_broot, ifp->if_broot,
+ XFS_BMBT_BLOCK_LEN(ip->i_mount));
} else {
new_broot = NULL;
ifp->if_flags &= ~XFS_IFBROOT;
@@ -890,7 +891,7 @@ xfs_iroot_realloc(
ifp->if_broot = new_broot;
ifp->if_broot_bytes = (int)new_size;
ASSERT(ifp->if_broot_bytes <=
- XFS_IFORK_SIZE(ip, whichfork) + XFS_BROOT_SIZE_ADJ);
+ XFS_IFORK_SIZE(ip, whichfork) + XFS_BROOT_SIZE_ADJ(ip));
return;
}
@@ -1161,7 +1162,7 @@ xfs_iflush_fork(
ASSERT(ifp->if_broot != NULL);
ASSERT(ifp->if_broot_bytes <=
(XFS_IFORK_SIZE(ip, whichfork) +
- XFS_BROOT_SIZE_ADJ));
+ XFS_BROOT_SIZE_ADJ(ip)));
xfs_bmbt_to_bmdr(mp, ifp->if_broot, ifp->if_broot_bytes,
(xfs_bmdr_block_t *)cp,
XFS_DFORK_SIZE(dip, mp, whichfork));
diff --git a/libxfs/xfs_mount.c b/libxfs/xfs_mount.c
index b7514fb..7ab3519 100644
--- a/libxfs/xfs_mount.c
+++ b/libxfs/xfs_mount.c
@@ -333,7 +333,7 @@ xfs_sb_verify(
* Only check the in progress field for the primary superblock as
* mkfs.xfs doesn't clear it from secondary superblocks.
*/
- error = xfs_mount_validate_sb(mp, &sb, bp->b_blkno == XFS_SB_DADDR);
+ error = xfs_mount_validate_sb(mp, &sb, bp->b_bn == XFS_SB_DADDR);
if (error)
xfs_buf_ioerror(bp, error);
}
diff --git a/mdrestore/Makefile b/mdrestore/Makefile
index ca2d1a0..5171306 100644
--- a/mdrestore/Makefile
+++ b/mdrestore/Makefile
@@ -8,7 +8,7 @@ include $(TOPDIR)/include/builddefs
LTCOMMAND = xfs_mdrestore
CFILES = xfs_mdrestore.c
-LLDLIBS = $(LIBXFS) $(LIBRT) $(LIBPTHREAD)
+LLDLIBS = $(LIBXFS) $(LIBRT) $(LIBPTHREAD) $(LIBUUID)
LTDEPENDENCIES = $(LIBXFS)
LLDFLAGS = -static
xfsprogs-crc-patchset-v2/ag-hdr-format-changes 0000600 0001750 0001750 00000043045 12132011546 020226 0 ustar dave dave xfsprogs: add crc format chagnes to ag headers
From: Dave Chinner <dchinner@xxxxxxxxxx>
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
include/xfs_ag.h | 54 ++++-
include/xfs_buf_item.h | 8 +-
libxfs/xfs_alloc.c | 197 ++++++++++-----
libxfs/xfs_bmap.c | 625 +++++++++++++++++++++++++++++++++++++++++++++++-
libxfs/xfs_ialloc.c | 55 +++--
5 files changed, 851 insertions(+), 88 deletions(-)
diff --git a/include/xfs_ag.h b/include/xfs_ag.h
index f2aeedb..1e0fa34 100644
--- a/include/xfs_ag.h
+++ b/include/xfs_ag.h
@@ -30,6 +30,7 @@ struct xfs_trans;
#define XFS_AGF_MAGIC 0x58414746 /* 'XAGF' */
#define XFS_AGI_MAGIC 0x58414749 /* 'XAGI' */
+#define XFS_AGFL_MAGIC 0x5841464c /* 'XAFL' */
#define XFS_AGF_VERSION 1
#define XFS_AGI_VERSION 1
@@ -63,12 +64,29 @@ typedef struct xfs_agf {
__be32 agf_spare0; /* spare field */
__be32 agf_levels[XFS_BTNUM_AGF]; /* btree levels */
__be32 agf_spare1; /* spare field */
+
__be32 agf_flfirst; /* first freelist block's index */
__be32 agf_fllast; /* last freelist block's index */
__be32 agf_flcount; /* count of blocks in freelist */
__be32 agf_freeblks; /* total free blocks */
+
__be32 agf_longest; /* longest free space */
__be32 agf_btreeblks; /* # of blocks held in AGF btrees */
+ uuid_t agf_uuid; /* uuid of filesystem */
+
+ /*
+ * reserve some contiguous space for future logged fields before we add
+ * the unlogged fields. This makes the range logging via flags and
+ * structure offsets much simpler.
+ */
+ __be64 agf_spare64[16];
+
+ /* unlogged fields, written during buffer writeback. */
+ __be64 agf_lsn; /* last write sequence */
+ __be32 agf_crc; /* crc of agf sector */
+ __be32 agf_spare2;
+
+ /* structure must be padded to 64 bit alignment */
} xfs_agf_t;
#define XFS_AGF_MAGICNUM 0x00000001
@@ -83,6 +101,7 @@ typedef struct xfs_agf {
#define XFS_AGF_FREEBLKS 0x00000200
#define XFS_AGF_LONGEST 0x00000400
#define XFS_AGF_BTREEBLKS 0x00000800
+#define XFS_AGF_UUID 0x00001000
#define XFS_AGF_NUM_BITS 12
#define XFS_AGF_ALL_BITS ((1 << XFS_AGF_NUM_BITS) - 1)
@@ -98,7 +117,8 @@ typedef struct xfs_agf {
{ XFS_AGF_FLCOUNT, "FLCOUNT" }, \
{ XFS_AGF_FREEBLKS, "FREEBLKS" }, \
{ XFS_AGF_LONGEST, "LONGEST" }, \
- { XFS_AGF_BTREEBLKS, "BTREEBLKS" }
+ { XFS_AGF_BTREEBLKS, "BTREEBLKS" }, \
+ { XFS_AGF_UUID, "UUID" }
/* disk block (xfs_daddr_t) in the AG */
#define XFS_AGF_DADDR(mp) ((xfs_daddr_t)(1 << (mp)->m_sectbb_log))
@@ -132,6 +152,7 @@ typedef struct xfs_agi {
__be32 agi_root; /* root of inode btree */
__be32 agi_level; /* levels in inode btree */
__be32 agi_freecount; /* number of free inodes */
+
__be32 agi_newino; /* new inode just allocated */
__be32 agi_dirino; /* last directory inode chunk */
/*
@@ -139,6 +160,13 @@ typedef struct xfs_agi {
* still being referenced.
*/
__be32 agi_unlinked[XFS_AGI_UNLINKED_BUCKETS];
+
+ uuid_t agi_uuid; /* uuid of filesystem */
+ __be32 agi_crc; /* crc of agi sector */
+ __be32 agi_pad32;
+ __be64 agi_lsn; /* last write sequence */
+
+ /* structure must be padded to 64 bit alignment */
} xfs_agi_t;
#define XFS_AGI_MAGICNUM 0x00000001
@@ -171,11 +199,31 @@ extern const struct xfs_buf_ops xfs_agi_buf_ops;
*/
#define XFS_AGFL_DADDR(mp) ((xfs_daddr_t)(3 << (mp)->m_sectbb_log))
#define XFS_AGFL_BLOCK(mp) XFS_HDR_BLOCK(mp, XFS_AGFL_DADDR(mp))
-#define XFS_AGFL_SIZE(mp) ((mp)->m_sb.sb_sectsize / sizeof(xfs_agblock_t))
#define XFS_BUF_TO_AGFL(bp) ((xfs_agfl_t *)((bp)->b_addr))
+#define XFS_BUF_TO_AGFL_BNO(mp, bp) \
+ (xfs_sb_version_hascrc(&((mp)->m_sb)) ? \
+ &(XFS_BUF_TO_AGFL(bp)->agfl_bno[0]) : \
+ (__be32 *)(bp)->b_addr)
+
+/*
+ * Size of the AGFL. For CRC-enabled filesystes we steal a couple of
+ * slots in the beginning of the block for a proper header with the
+ * location information and CRC.
+ */
+#define XFS_AGFL_SIZE(mp) \
+ (((mp)->m_sb.sb_sectsize - \
+ (xfs_sb_version_hascrc(&((mp)->m_sb)) ? \
+ sizeof(struct xfs_agfl) : 0)) / \
+ sizeof(xfs_agblock_t))
+
typedef struct xfs_agfl {
- __be32 agfl_bno[1]; /* actually XFS_AGFL_SIZE(mp) */
+ __be32 agfl_magicnum;
+ __be32 agfl_seqno;
+ uuid_t agfl_uuid;
+ __be64 agfl_lsn;
+ __be32 agfl_crc;
+ __be32 agfl_bno[]; /* actually XFS_AGFL_SIZE(mp) */
} xfs_agfl_t;
/*
diff --git a/include/xfs_buf_item.h b/include/xfs_buf_item.h
index 101ef83..c256606 100644
--- a/include/xfs_buf_item.h
+++ b/include/xfs_buf_item.h
@@ -45,12 +45,18 @@ extern kmem_zone_t *xfs_buf_item_zone;
* once the changes have been replayed into the buffer.
*/
#define XFS_BLF_BTREE_BUF (1<<5)
+#define XFS_BLF_AGF_BUF (1<<6)
+#define XFS_BLF_AGFL_BUF (1<<7)
+#define XFS_BLF_AGI_BUF (1<<8)
#define XFS_BLF_TYPE_MASK \
(XFS_BLF_UDQUOT_BUF | \
XFS_BLF_PDQUOT_BUF | \
XFS_BLF_GDQUOT_BUF | \
- XFS_BLF_BTREE_BUF)
+ XFS_BLF_BTREE_BUF | \
+ XFS_BLF_AGF_BUF | \
+ XFS_BLF_AGFL_BUF | \
+ XFS_BLF_AGI_BUF)
#define XFS_BLF_CHUNK 128
#define XFS_BLF_SHIFT 7
diff --git a/libxfs/xfs_alloc.c b/libxfs/xfs_alloc.c
index e59fdac..30fc5f4 100644
--- a/libxfs/xfs_alloc.c
+++ b/libxfs/xfs_alloc.c
@@ -410,53 +410,84 @@ xfs_alloc_fixup_trees(
return 0;
}
-static void
+static bool
xfs_agfl_verify(
struct xfs_buf *bp)
{
-#ifdef WHEN_CRCS_COME_ALONG
- /*
- * we cannot actually do any verification of the AGFL because mkfs does
- * not initialise the AGFL to zero or NULL. Hence the only valid part of
- * the AGFL is what the AGF says is active. We can't get to the AGF, so
- * we can't verify just those entries are valid.
- *
- * This problem goes away when the CRC format change comes along as that
- * requires the AGFL to be initialised by mkfs. At that point, we can
- * verify the blocks in the agfl -active or not- lie within the bounds
- * of the AG. Until then, just leave this check ifdef'd out.
- */
struct xfs_mount *mp = bp->b_target->bt_mount;
struct xfs_agfl *agfl = XFS_BUF_TO_AGFL(bp);
- int agfl_ok = 1;
-
int i;
+ if (!uuid_equal(&agfl->agfl_uuid, &mp->m_sb.sb_uuid))
+ return false;
+ if (be32_to_cpu(agfl->agfl_magicnum) != XFS_AGFL_MAGIC)
+ return false;
+ /*
+ * during growfs operations, the perag is not fully initialised,
+ * so we can't use it for any useful checking. growfs ensures we can't
+ * use it by using uncached buffers that don't have the perag attached
+ * so we can detect and avoid this problem.
+ */
+ if (bp->b_pag && be32_to_cpu(agfl->agfl_seqno) != bp->b_pag->pag_agno)
+ return false;
+
for (i = 0; i < XFS_AGFL_SIZE(mp); i++) {
- if (be32_to_cpu(agfl->agfl_bno[i]) == NULLAGBLOCK ||
+ if (be32_to_cpu(agfl->agfl_bno[i]) != NULLAGBLOCK &&
be32_to_cpu(agfl->agfl_bno[i]) >= mp->m_sb.sb_agblocks)
- agfl_ok = 0;
+ return false;
}
+ return true;
+}
+
+static void
+xfs_agfl_read_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ int agfl_ok = 1;
+
+ /*
+ * There is no verification of non-crc AGFLs because mkfs does not
+ * initialise the AGFL to zero or NULL. Hence the only valid part of the
+ * AGFL is what the AGF says is active. We can't get to the AGF, so we
+ * can't verify just those entries are valid.
+ */
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return;
+
+ agfl_ok = xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
+ offsetof(struct xfs_agfl, agfl_crc));
+
+ agfl_ok = agfl_ok && xfs_agfl_verify(bp);
if (!agfl_ok) {
- XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, agfl);
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
xfs_buf_ioerror(bp, EFSCORRUPTED);
}
-#endif
}
static void
xfs_agfl_write_verify(
struct xfs_buf *bp)
{
- xfs_agfl_verify(bp);
-}
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_buf_log_item *bip = bp->b_fspriv;
-static void
-xfs_agfl_read_verify(
- struct xfs_buf *bp)
-{
- xfs_agfl_verify(bp);
+ /* no verification of non-crc AGFLs */
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return;
+
+ if (!xfs_agfl_verify(bp)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ return;
+ }
+
+ if (bip)
+ XFS_BUF_TO_AGFL(bp)->agfl_lsn = cpu_to_be64(bip->bli_item.li_lsn);
+
+ xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length),
+ offsetof(struct xfs_agfl, agfl_crc));
}
const struct xfs_buf_ops xfs_agfl_buf_ops = {
@@ -1964,18 +1995,18 @@ xfs_alloc_get_freelist(
int btreeblk) /* destination is a AGF btree */
{
xfs_agf_t *agf; /* a.g. freespace structure */
- xfs_agfl_t *agfl; /* a.g. freelist structure */
xfs_buf_t *agflbp;/* buffer for a.g. freelist structure */
xfs_agblock_t bno; /* block number returned */
+ __be32 *agfl_bno;
int error;
int logflags;
- xfs_mount_t *mp; /* mount structure */
+ xfs_mount_t *mp = tp->t_mountp;
xfs_perag_t *pag; /* per allocation group data */
- agf = XFS_BUF_TO_AGF(agbp);
/*
* Freelist is empty, give up.
*/
+ agf = XFS_BUF_TO_AGF(agbp);
if (!agf->agf_flcount) {
*bnop = NULLAGBLOCK;
return 0;
@@ -1983,15 +2014,17 @@ xfs_alloc_get_freelist(
/*
* Read the array of free blocks.
*/
- mp = tp->t_mountp;
- if ((error = xfs_alloc_read_agfl(mp, tp,
- be32_to_cpu(agf->agf_seqno), &agflbp)))
+ error = xfs_alloc_read_agfl(mp, tp, be32_to_cpu(agf->agf_seqno),
+ &agflbp);
+ if (error)
return error;
- agfl = XFS_BUF_TO_AGFL(agflbp);
+
+
/*
* Get the block number and update the data structures.
*/
- bno = be32_to_cpu(agfl->agfl_bno[be32_to_cpu(agf->agf_flfirst)]);
+ agfl_bno = XFS_BUF_TO_AGFL_BNO(mp, agflbp);
+ bno = be32_to_cpu(agfl_bno[be32_to_cpu(agf->agf_flfirst)]);
be32_add_cpu(&agf->agf_flfirst, 1);
xfs_trans_brelse(tp, agflbp);
if (be32_to_cpu(agf->agf_flfirst) == XFS_AGFL_SIZE(mp))
@@ -2040,11 +2073,14 @@ xfs_alloc_log_agf(
offsetof(xfs_agf_t, agf_freeblks),
offsetof(xfs_agf_t, agf_longest),
offsetof(xfs_agf_t, agf_btreeblks),
+ offsetof(xfs_agf_t, agf_uuid),
sizeof(xfs_agf_t)
};
trace_xfs_agf(tp->t_mountp, XFS_BUF_TO_AGF(bp), fields, _RET_IP_);
+ xfs_trans_buf_set_type(tp, bp, XFS_BLF_AGF_BUF);
+
xfs_btree_offsets(fields, offsets, XFS_AGF_NUM_BITS, &first, &last);
xfs_trans_log_buf(tp, bp, (uint)first, (uint)last);
}
@@ -2081,12 +2117,13 @@ xfs_alloc_put_freelist(
int btreeblk) /* block came from a AGF btree */
{
xfs_agf_t *agf; /* a.g. freespace structure */
- xfs_agfl_t *agfl; /* a.g. free block array */
__be32 *blockp;/* pointer to array entry */
int error;
int logflags;
xfs_mount_t *mp; /* mount structure */
xfs_perag_t *pag; /* per allocation group data */
+ __be32 *agfl_bno;
+ int startoff;
agf = XFS_BUF_TO_AGF(agbp);
mp = tp->t_mountp;
@@ -2094,7 +2131,6 @@ xfs_alloc_put_freelist(
if (!agflbp && (error = xfs_alloc_read_agfl(mp, tp,
be32_to_cpu(agf->agf_seqno), &agflbp)))
return error;
- agfl = XFS_BUF_TO_AGFL(agflbp);
be32_add_cpu(&agf->agf_fllast, 1);
if (be32_to_cpu(agf->agf_fllast) == XFS_AGFL_SIZE(mp))
agf->agf_fllast = 0;
@@ -2115,32 +2151,38 @@ xfs_alloc_put_freelist(
xfs_alloc_log_agf(tp, agbp, logflags);
ASSERT(be32_to_cpu(agf->agf_flcount) <= XFS_AGFL_SIZE(mp));
- blockp = &agfl->agfl_bno[be32_to_cpu(agf->agf_fllast)];
+
+ agfl_bno = XFS_BUF_TO_AGFL_BNO(mp, agflbp);
+ blockp = &agfl_bno[be32_to_cpu(agf->agf_fllast)];
*blockp = cpu_to_be32(bno);
+ startoff = (char *)blockp - (char *)agflbp->b_addr;
+
xfs_alloc_log_agf(tp, agbp, logflags);
- xfs_trans_log_buf(tp, agflbp,
- (int)((xfs_caddr_t)blockp - (xfs_caddr_t)agfl),
- (int)((xfs_caddr_t)blockp - (xfs_caddr_t)agfl +
- sizeof(xfs_agblock_t) - 1));
+
+ xfs_trans_buf_set_type(tp, agflbp, XFS_BLF_AGFL_BUF);
+ xfs_trans_log_buf(tp, agflbp, startoff,
+ startoff + sizeof(xfs_agblock_t) - 1);
return 0;
}
-static void
+static bool
xfs_agf_verify(
+ struct xfs_mount *mp,
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
- struct xfs_agf *agf;
- int agf_ok;
+ struct xfs_agf *agf = XFS_BUF_TO_AGF(bp);
- agf = XFS_BUF_TO_AGF(bp);
+ if (xfs_sb_version_hascrc(&mp->m_sb) &&
+ !uuid_equal(&agf->agf_uuid, &mp->m_sb.sb_uuid))
+ return false;
- agf_ok = agf->agf_magicnum == cpu_to_be32(XFS_AGF_MAGIC) &&
- XFS_AGF_GOOD_VERSION(be32_to_cpu(agf->agf_versionnum)) &&
- be32_to_cpu(agf->agf_freeblks) <= be32_to_cpu(agf->agf_length) &&
- be32_to_cpu(agf->agf_flfirst) < XFS_AGFL_SIZE(mp) &&
- be32_to_cpu(agf->agf_fllast) < XFS_AGFL_SIZE(mp) &&
- be32_to_cpu(agf->agf_flcount) <= XFS_AGFL_SIZE(mp);
+ if (!(agf->agf_magicnum == cpu_to_be32(XFS_AGF_MAGIC) &&
+ XFS_AGF_GOOD_VERSION(be32_to_cpu(agf->agf_versionnum)) &&
+ be32_to_cpu(agf->agf_freeblks) <= be32_to_cpu(agf->agf_length) &&
+ be32_to_cpu(agf->agf_flfirst) < XFS_AGFL_SIZE(mp) &&
+ be32_to_cpu(agf->agf_fllast) < XFS_AGFL_SIZE(mp) &&
+ be32_to_cpu(agf->agf_flcount) <= XFS_AGFL_SIZE(mp)))
+ return false;
/*
* during growfs operations, the perag is not fully initialised,
@@ -2148,33 +2190,58 @@ xfs_agf_verify(
* use it by using uncached buffers that don't have the perag attached
* so we can detect and avoid this problem.
*/
- if (bp->b_pag)
- agf_ok = agf_ok && be32_to_cpu(agf->agf_seqno) ==
- bp->b_pag->pag_agno;
+ if (bp->b_pag && be32_to_cpu(agf->agf_seqno) != bp->b_pag->pag_agno)
+ return false;
- if (xfs_sb_version_haslazysbcount(&mp->m_sb))
- agf_ok = agf_ok && be32_to_cpu(agf->agf_btreeblks) <=
- be32_to_cpu(agf->agf_length);
+ if (xfs_sb_version_haslazysbcount(&mp->m_sb) &&
+ be32_to_cpu(agf->agf_btreeblks) > be32_to_cpu(agf->agf_length))
+ return false;
+
+ return true;;
- if (unlikely(XFS_TEST_ERROR(!agf_ok, mp, XFS_ERRTAG_ALLOC_READ_AGF,
- XFS_RANDOM_ALLOC_READ_AGF))) {
- XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, agf);
- xfs_buf_ioerror(bp, EFSCORRUPTED);
- }
}
static void
xfs_agf_read_verify(
struct xfs_buf *bp)
{
- xfs_agf_verify(bp);
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ int agf_ok = 1;
+
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ agf_ok = xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
+ offsetof(struct xfs_agf, agf_crc));
+
+ agf_ok = agf_ok && xfs_agf_verify(mp, bp);
+
+ if (unlikely(XFS_TEST_ERROR(!agf_ok, mp, XFS_ERRTAG_ALLOC_READ_AGF,
+ XFS_RANDOM_ALLOC_READ_AGF))) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
}
static void
xfs_agf_write_verify(
struct xfs_buf *bp)
{
- xfs_agf_verify(bp);
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_buf_log_item *bip = bp->b_fspriv;
+
+ if (!xfs_agf_verify(mp, bp)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ return;
+ }
+
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return;
+
+ if (bip)
+ XFS_BUF_TO_AGF(bp)->agf_lsn = cpu_to_be64(bip->bli_item.li_lsn);
+
+ xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length),
+ offsetof(struct xfs_agf, agf_crc));
}
const struct xfs_buf_ops xfs_agf_buf_ops = {
diff --git a/libxfs/xfs_ialloc.c b/libxfs/xfs_ialloc.c
index f0322c9..feb4a4e 100644
--- a/libxfs/xfs_ialloc.c
+++ b/libxfs/xfs_ialloc.c
@@ -1267,6 +1267,7 @@ xfs_ialloc_log_agi(
/*
* Log the allocation group inode header buffer.
*/
+ xfs_trans_buf_set_type(tp, bp, XFS_BLF_AGI_BUF);
xfs_trans_log_buf(tp, bp, first, last);
}
@@ -1284,19 +1285,23 @@ xfs_check_agi_unlinked(
#define xfs_check_agi_unlinked(agi)
#endif
-static void
+static bool
xfs_agi_verify(
struct xfs_buf *bp)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
struct xfs_agi *agi = XFS_BUF_TO_AGI(bp);
- int agi_ok;
+ if (xfs_sb_version_hascrc(&mp->m_sb) &&
+ !uuid_equal(&agi->agi_uuid, &mp->m_sb.sb_uuid))
+ return false;
/*
* Validate the magic number of the agi block.
*/
- agi_ok = agi->agi_magicnum == cpu_to_be32(XFS_AGI_MAGIC) &&
- XFS_AGI_GOOD_VERSION(be32_to_cpu(agi->agi_versionnum));
+ if (agi->agi_magicnum != cpu_to_be32(XFS_AGI_MAGIC))
+ return false;
+ if (!XFS_AGI_GOOD_VERSION(be32_to_cpu(agi->agi_versionnum)))
+ return false;
/*
* during growfs operations, the perag is not fully initialised,
@@ -1304,30 +1309,52 @@ xfs_agi_verify(
* use it by using uncached buffers that don't have the perag attached
* so we can detect and avoid this problem.
*/
- if (bp->b_pag)
- agi_ok = agi_ok && be32_to_cpu(agi->agi_seqno) ==
- bp->b_pag->pag_agno;
+ if (bp->b_pag && be32_to_cpu(agi->agi_seqno) != bp->b_pag->pag_agno)
+ return false;
- if (unlikely(XFS_TEST_ERROR(!agi_ok, mp, XFS_ERRTAG_IALLOC_READ_AGI,
- XFS_RANDOM_IALLOC_READ_AGI))) {
- XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, agi);
- xfs_buf_ioerror(bp, EFSCORRUPTED);
- }
xfs_check_agi_unlinked(agi);
+ return true;
}
static void
xfs_agi_read_verify(
struct xfs_buf *bp)
{
- xfs_agi_verify(bp);
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ int agi_ok = 1;
+
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ agi_ok = xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
+ offsetof(struct xfs_agi, agi_crc));
+ agi_ok = agi_ok && xfs_agi_verify(bp);
+
+ if (unlikely(XFS_TEST_ERROR(!agi_ok, mp, XFS_ERRTAG_IALLOC_READ_AGI,
+ XFS_RANDOM_IALLOC_READ_AGI))) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
}
static void
xfs_agi_write_verify(
struct xfs_buf *bp)
{
- xfs_agi_verify(bp);
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_buf_log_item *bip = bp->b_fspriv;
+
+ if (!xfs_agi_verify(bp)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ return;
+ }
+
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return;
+
+ if (bip)
+ XFS_BUF_TO_AGI(bp)->agi_lsn = cpu_to_be64(bip->bli_item.li_lsn);
+ xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length),
+ offsetof(struct xfs_agi, agi_crc));
}
const struct xfs_buf_ops xfs_agi_buf_ops = {
xfsprogs-crc-patchset-v2/quota-format-changes 0000600 0001750 0001750 00000002173 12132011546 020212 0 ustar dave dave libxfs: change quota buffer formats
From: Dave Chinner <dchinner@xxxxxxxxxx>
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
include/xfs_quota.h | 11 ++++++++++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/include/xfs_quota.h b/include/xfs_quota.h
index b50ec5b..c61e31c 100644
--- a/include/xfs_quota.h
+++ b/include/xfs_quota.h
@@ -77,7 +77,14 @@ typedef struct xfs_disk_dquot {
*/
typedef struct xfs_dqblk {
xfs_disk_dquot_t dd_diskdq; /* portion that lives incore as well */
- char dd_fill[32]; /* filling for posterity */
+ char dd_fill[4]; /* filling for posterity */
+
+ /*
+ * These two are only present on filesystems with the CRC bits set.
+ */
+ __be32 dd_crc; /* checksum */
+ __be64 dd_lsn; /* last modification in log */
+ uuid_t dd_uuid; /* location information */
} xfs_dqblk_t;
/*
@@ -380,5 +387,7 @@ extern int xfs_qm_dqcheck(struct xfs_mount *, xfs_disk_dquot_t *,
xfs_dqid_t, uint, uint, char *);
extern int xfs_mount_reset_sbqflags(struct xfs_mount *);
+extern const struct xfs_buf_ops xfs_dquot_buf_ops;
+
#endif /* __KERNEL__ */
#endif /* __XFS_QUOTA_H__ */
xfsprogs-crc-patchset-v2/inode-v3 0000600 0001750 0001750 00000050210 12132011546 015604 0 ustar dave dave libxfs: add version 3 inode support
From: Dave Chinner <dchinner@xxxxxxxxxx>
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
Header from folded patch 'debug':
xfs_quota: fix report command parsing
From: Dave Chinner <dchinner@xxxxxxxxxx>
The report command line needs to be parsed as a whole not as
individual elements - report_f() is set up to do this correctly.
When treated as non-global command line, the report function is
called once for each command line arg, resulting in reports being
issued multiple times.
Set the command to be a global command so that it is only called
once.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
db/dir2sf.c | 9 +++--
include/xfs_buf_item.h | 4 +-
include/xfs_dinode.h | 33 +++++++++++++++--
include/xfs_inode.h | 26 +++++++++++++
libxfs/trans.c | 1 +
libxfs/util.c | 30 ++++++++++++++-
libxfs/xfs_ialloc.c | 25 ++++++++++++-
libxfs/xfs_inode.c | 91 ++++++++++++++++++++++++++++++++++++++++------
logprint/log_misc.c | 2 +-
logprint/log_print_all.c | 3 +-
repair/phase6.c | 63 +++++++++++++++++++++++++++++---
11 files changed, 257 insertions(+), 30 deletions(-)
diff --git a/db/dir2sf.c b/db/dir2sf.c
index 92f8a66..271e08a 100644
--- a/db/dir2sf.c
+++ b/db/dir2sf.c
@@ -74,10 +74,11 @@ dir2_inou_i4_count(
void *obj,
int startoff)
{
+ struct xfs_dinode *dip = obj;
xfs_dir2_sf_t *sf;
ASSERT(bitoffs(startoff) == 0);
- sf = (xfs_dir2_sf_t *)XFS_DFORK_DPTR(obj);
+ sf = (xfs_dir2_sf_t *)XFS_DFORK_DPTR(dip);
return sf->hdr.i8count == 0;
}
@@ -87,10 +88,11 @@ dir2_inou_i8_count(
void *obj,
int startoff)
{
+ struct xfs_dinode *dip = obj;
xfs_dir2_sf_t *sf;
ASSERT(bitoffs(startoff) == 0);
- sf = (xfs_dir2_sf_t *)XFS_DFORK_DPTR(obj);
+ sf = (xfs_dir2_sf_t *)XFS_DFORK_DPTR(dip);
return sf->hdr.i8count != 0;
}
@@ -101,11 +103,12 @@ dir2_inou_size(
int startoff,
int idx)
{
+ struct xfs_dinode *dip = obj;
xfs_dir2_sf_t *sf;
ASSERT(bitoffs(startoff) == 0);
ASSERT(idx == 0);
- sf = (xfs_dir2_sf_t *)XFS_DFORK_DPTR(obj);
+ sf = (xfs_dir2_sf_t *)XFS_DFORK_DPTR(dip);
return bitize(sf->hdr.i8count ?
(uint)sizeof(xfs_dir2_ino8_t) :
(uint)sizeof(xfs_dir2_ino4_t));
diff --git a/include/xfs_buf_item.h b/include/xfs_buf_item.h
index c256606..abae8c8 100644
--- a/include/xfs_buf_item.h
+++ b/include/xfs_buf_item.h
@@ -48,6 +48,7 @@ extern kmem_zone_t *xfs_buf_item_zone;
#define XFS_BLF_AGF_BUF (1<<6)
#define XFS_BLF_AGFL_BUF (1<<7)
#define XFS_BLF_AGI_BUF (1<<8)
+#define XFS_BLF_DINO_BUF (1<<9)
#define XFS_BLF_TYPE_MASK \
(XFS_BLF_UDQUOT_BUF | \
@@ -56,7 +57,8 @@ extern kmem_zone_t *xfs_buf_item_zone;
XFS_BLF_BTREE_BUF | \
XFS_BLF_AGF_BUF | \
XFS_BLF_AGFL_BUF | \
- XFS_BLF_AGI_BUF)
+ XFS_BLF_AGI_BUF | \
+ XFS_BLF_DINO_BUF)
#define XFS_BLF_CHUNK 128
#define XFS_BLF_SHIFT 7
diff --git a/include/xfs_dinode.h b/include/xfs_dinode.h
index 6b5bd17..f7a0e95 100644
--- a/include/xfs_dinode.h
+++ b/include/xfs_dinode.h
@@ -19,7 +19,7 @@
#define __XFS_DINODE_H__
#define XFS_DINODE_MAGIC 0x494e /* 'IN' */
-#define XFS_DINODE_GOOD_VERSION(v) (((v) == 1 || (v) == 2))
+#define XFS_DINODE_GOOD_VERSION(v) ((v) >= 1 && (v) <= 3)
typedef struct xfs_timestamp {
__be32 t_sec; /* timestamp seconds */
@@ -70,11 +70,36 @@ typedef struct xfs_dinode {
/* di_next_unlinked is the only non-core field in the old dinode */
__be32 di_next_unlinked;/* agi unlinked list ptr */
-} __attribute__((packed)) xfs_dinode_t;
+
+ /* start of the extended dinode, writable fields */
+ __le32 di_crc; /* CRC of the inode */
+ __be64 di_changecount; /* number of attribute changes */
+ __be64 di_lsn; /* flush sequence */
+ __be64 di_flags2; /* more random flags */
+ __u8 di_pad2[16]; /* more padding for future expansion */
+
+ /* fields only written to during inode creation */
+ xfs_timestamp_t di_crtime; /* time created */
+ __be64 di_ino; /* inode number */
+ uuid_t di_uuid; /* UUID of the filesystem */
+
+ /* structure must be padded to 64 bit alignment */
+} xfs_dinode_t;
#define DI_MAX_FLUSH 0xffff
/*
+ * Size of the core inode on disk. Version 1 and 2 inodes have
+ * the same size, but version 3 has grown a few additional fields.
+ */
+static inline uint xfs_dinode_size(int version)
+{
+ if (version == 3)
+ return sizeof(struct xfs_dinode);
+ return offsetof(struct xfs_dinode, di_crc);
+}
+
+/*
* The 32 bit link count in the inode theoretically maxes out at UINT_MAX.
* Since the pathconf interface is signed, we use 2^31 - 1 instead.
* The old inode format had a 16 bit link count, so its maximum is USHRT_MAX.
@@ -105,7 +130,7 @@ typedef enum xfs_dinode_fmt {
* Inode size for given fs.
*/
#define XFS_LITINO(mp, version) \
- ((int)(((mp)->m_sb.sb_inodesize) - sizeof(struct xfs_dinode)))
+ ((int)(((mp)->m_sb.sb_inodesize) - xfs_dinode_size(version)))
#define XFS_BROOT_SIZE_ADJ(ip) \
(XFS_BMBT_BLOCK_LEN((ip)->i_mount) - sizeof(xfs_bmdr_block_t))
@@ -133,7 +158,7 @@ typedef enum xfs_dinode_fmt {
* Return pointers to the data or attribute forks.
*/
#define XFS_DFORK_DPTR(dip) \
- ((char *)(dip) + sizeof(struct xfs_dinode))
+ ((char *)dip + xfs_dinode_size(dip->di_version))
#define XFS_DFORK_APTR(dip) \
(XFS_DFORK_DPTR(dip) + XFS_DFORK_BOFF(dip))
#define XFS_DFORK_PTR(dip,w) \
diff --git a/include/xfs_inode.h b/include/xfs_inode.h
index 4733f85..cc14743 100644
--- a/include/xfs_inode.h
+++ b/include/xfs_inode.h
@@ -150,13 +150,38 @@ typedef struct xfs_icdinode {
__uint16_t di_dmstate; /* DMIG state info */
__uint16_t di_flags; /* random flags, XFS_DIFLAG_... */
__uint32_t di_gen; /* generation number */
+
+ /* di_next_unlinked is the only non-core field in the old dinode */
+ __be32 di_next_unlinked;/* agi unlinked list ptr */
+
+ /* start of the extended dinode, writable fields */
+ __uint32_t di_crc; /* CRC of the inode */
+ __uint64_t di_changecount; /* number of attribute changes */
+ xfs_lsn_t di_lsn; /* flush sequence */
+ __uint64_t di_flags2; /* more random flags */
+ __uint8_t di_pad2[16]; /* more padding for future expansion */
+
+ /* fields only written to during inode creation */
+ xfs_ictimestamp_t di_crtime; /* time created */
+ xfs_ino_t di_ino; /* inode number */
+ uuid_t di_uuid; /* UUID of the filesystem */
+
+ /* structure must be padded to 64 bit alignment */
} xfs_icdinode_t;
+static inline uint xfs_icdinode_size(struct xfs_icdinode *dicp)
+{
+ if (dicp->di_version == 3)
+ return sizeof(struct xfs_icdinode);
+ return offsetof(struct xfs_icdinode, di_next_unlinked);
+}
+
/*
* Flags for xfs_ichgtime().
*/
#define XFS_ICHGTIME_MOD 0x1 /* data fork modification timestamp */
#define XFS_ICHGTIME_CHG 0x2 /* inode field change timestamp */
+#define XFS_ICHGTIME_CREATE 0x4 /* inode create timestamp */
/*
* Per-fork incore inode flags.
@@ -556,6 +581,7 @@ int xfs_imap_to_bp(struct xfs_mount *, struct xfs_trans *,
struct xfs_buf **, uint, uint);
int xfs_iread(struct xfs_mount *, struct xfs_trans *,
struct xfs_inode *, uint);
+void xfs_dinode_calc_crc(struct xfs_mount *, struct xfs_dinode *);
void xfs_dinode_to_disk(struct xfs_dinode *,
struct xfs_icdinode *);
void xfs_idestroy_fork(struct xfs_inode *, int);
diff --git a/libxfs/trans.c b/libxfs/trans.c
index 7cb3c8c..619aad1 100644
--- a/libxfs/trans.c
+++ b/libxfs/trans.c
@@ -218,6 +218,7 @@ libxfs_trans_inode_alloc_buf(
ASSERT(XFS_BUF_FSPRIVATE(bp, void *) != NULL);
bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *);
bip->bli_flags |= XFS_BLI_INODE_ALLOC_BUF;
+ xfs_trans_buf_set_type(tp, bp, XFS_BLF_DINO_BUF);
}
/*
diff --git a/libxfs/util.c b/libxfs/util.c
index 2ad4bfd..abe16cf 100644
--- a/libxfs/util.c
+++ b/libxfs/util.c
@@ -47,6 +47,10 @@ libxfs_trans_ichgtime(
ip->i_d.di_ctime.t_sec = (__int32_t)tv.tv_sec;
ip->i_d.di_ctime.t_nsec = (__int32_t)tv.tv_nsec;
}
+ if (flags & XFS_ICHGTIME_CREATE) {
+ ip->i_d.di_crtime.t_sec = (__int32_t)tv.tv_sec;
+ ip->i_d.di_crtime.t_nsec = (__int32_t)tv.tv_nsec;
+ }
}
/*
@@ -75,6 +79,7 @@ libxfs_ialloc(
xfs_inode_t *ip;
uint flags;
int error;
+ int times;
/*
* Call the space management code to pick
@@ -103,6 +108,7 @@ libxfs_ialloc(
ip->i_d.di_gid = cr->cr_gid;
xfs_set_projid(&ip->i_d, pip ? 0 : fsx->fsx_projid);
memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad));
+ xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG | XFS_ICHGTIME_MOD);
/*
* If the superblock version is up to where we support new format
@@ -128,7 +134,6 @@ libxfs_ialloc(
ip->i_d.di_size = 0;
ip->i_d.di_nextents = 0;
ASSERT(ip->i_d.di_nblocks == 0);
- xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG|XFS_ICHGTIME_MOD);
/*
* di_gen will have been taken care of in xfs_iread.
*/
@@ -136,6 +141,18 @@ libxfs_ialloc(
ip->i_d.di_dmevmask = 0;
ip->i_d.di_dmstate = 0;
ip->i_d.di_flags = pip ? 0 : fsx->fsx_xflags;
+
+ if (ip->i_d.di_version == 3) {
+ ASSERT(ip->i_d.di_ino == ino);
+ ASSERT(uuid_equal(&ip->i_d.di_uuid, &mp->m_sb.sb_uuid));
+ ip->i_d.di_crc = 0;
+ ip->i_d.di_changecount = 1;
+ ip->i_d.di_lsn = 0;
+ ip->i_d.di_flags2 = 0;
+ memset(&(ip->i_d.di_pad2[0]), 0, sizeof(ip->i_d.di_pad2));
+ ip->i_d.di_crtime = ip->i_d.di_mtime;
+ }
+
flags = XFS_ILOG_CORE;
switch (mode & S_IFMT) {
case S_IFIFO:
@@ -295,6 +312,10 @@ libxfs_iflush_int(xfs_inode_t *ip, xfs_buf_t *bp)
ASSERT(ip->i_d.di_nextents+ip->i_d.di_anextents <= ip->i_d.di_nblocks);
ASSERT(ip->i_d.di_forkoff <= mp->m_sb.sb_inodesize);
+ /* bump the change count on v3 inodes */
+ if (ip->i_d.di_version == 3)
+ ip->i_d.di_changecount++;
+
/*
* Copy the dirty parts of the inode into the on-disk
* inode. We always copy out the core of the inode,
@@ -338,6 +359,13 @@ libxfs_iflush_int(xfs_inode_t *ip, xfs_buf_t *bp)
if (XFS_IFORK_Q(ip))
xfs_iflush_fork(ip, dip, iip, XFS_ATTR_FORK, bp);
+ /* update the lsn in the on disk inode if required */
+ if (ip->i_d.di_version == 3)
+ dip->di_lsn = cpu_to_be64(iip->ili_item.li_lsn);
+
+ /* generate the checksum. */
+ xfs_dinode_calc_crc(mp, dip);
+
return 0;
}
diff --git a/libxfs/xfs_ialloc.c b/libxfs/xfs_ialloc.c
index feb4a4e..008c4ea 100644
--- a/libxfs/xfs_ialloc.c
+++ b/libxfs/xfs_ialloc.c
@@ -146,6 +146,7 @@ xfs_ialloc_inode_init(
int version;
int i, j;
xfs_daddr_t d;
+ xfs_ino_t ino = 0;
/*
* Loop over the new block(s), filling in the inodes.
@@ -169,8 +170,20 @@ xfs_ialloc_inode_init(
* the new inode format, then use the new inode version. Otherwise
* use the old version so that old kernels will continue to be
* able to use the file system.
+ *
+ * For v3 inodes, we also need to write the inode number into the inode,
+ * so calculate the first inode number of the chunk here as
+ * XFS_OFFBNO_TO_AGINO() only works on filesystem block boundaries, not
+ * cluster boundaries and so cannot be used in the cluster buffer loop
+ * below.
*/
- if (xfs_sb_version_hasnlink(&mp->m_sb))
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ version = 3;
+ ino = XFS_AGINO_TO_INO(mp, agno,
+ XFS_OFFBNO_TO_AGINO(mp, agbno, 0));
+ } else if (xfs_sb_version_hasnlink(&mp->m_sb))
+ version = 3;
+ else if (xfs_sb_version_hasnlink(&mp->m_sb))
version = 2;
else
version = 1;
@@ -196,13 +209,21 @@ xfs_ialloc_inode_init(
xfs_buf_zero(fbuf, 0, ninodes << mp->m_sb.sb_inodelog);
for (i = 0; i < ninodes; i++) {
int ioffset = i << mp->m_sb.sb_inodelog;
- uint isize = sizeof(struct xfs_dinode);
+ uint isize = xfs_dinode_size(version);
free = xfs_make_iptr(mp, fbuf, i);
free->di_magic = cpu_to_be16(XFS_DINODE_MAGIC);
free->di_version = version;
free->di_gen = cpu_to_be32(gen);
free->di_next_unlinked = cpu_to_be32(NULLAGINO);
+
+ if (version == 3) {
+ free->di_ino = cpu_to_be64(ino);
+ ino++;
+ uuid_copy(&free->di_uuid, &mp->m_sb.sb_uuid);
+ xfs_dinode_calc_crc(mp, free);
+ }
+
xfs_trans_log_buf(tp, fbuf, ioffset, ioffset + isize - 1);
}
xfs_trans_inode_alloc_buf(tp, fbuf);
diff --git a/libxfs/xfs_inode.c b/libxfs/xfs_inode.c
index f9f792c..d6513b9 100644
--- a/libxfs/xfs_inode.c
+++ b/libxfs/xfs_inode.c
@@ -572,6 +572,17 @@ xfs_dinode_from_disk(
to->di_dmstate = be16_to_cpu(from->di_dmstate);
to->di_flags = be16_to_cpu(from->di_flags);
to->di_gen = be32_to_cpu(from->di_gen);
+
+ if (to->di_version == 3) {
+ to->di_changecount = be64_to_cpu(from->di_changecount);
+ to->di_crtime.t_sec = be32_to_cpu(from->di_crtime.t_sec);
+ to->di_crtime.t_nsec = be32_to_cpu(from->di_crtime.t_nsec);
+ to->di_flags2 = be64_to_cpu(from->di_flags2);
+ to->di_ino = be64_to_cpu(from->di_ino);
+ to->di_lsn = be64_to_cpu(from->di_lsn);
+ memcpy(to->di_pad2, from->di_pad2, sizeof(to->di_pad2));
+ platform_uuid_copy(&to->di_uuid, &from->di_uuid);
+ }
}
void
@@ -608,6 +619,58 @@ xfs_dinode_to_disk(
to->di_dmstate = cpu_to_be16(from->di_dmstate);
to->di_flags = cpu_to_be16(from->di_flags);
to->di_gen = cpu_to_be32(from->di_gen);
+
+ if (from->di_version == 3) {
+ to->di_changecount = cpu_to_be64(from->di_changecount);
+ to->di_crtime.t_sec = cpu_to_be32(from->di_crtime.t_sec);
+ to->di_crtime.t_nsec = cpu_to_be32(from->di_crtime.t_nsec);
+ to->di_flags2 = cpu_to_be64(from->di_flags2);
+ to->di_ino = cpu_to_be64(from->di_ino);
+ to->di_lsn = cpu_to_be64(from->di_lsn);
+ memcpy(to->di_pad2, from->di_pad2, sizeof(to->di_pad2));
+ platform_uuid_copy(&to->di_uuid, &from->di_uuid);
+ }
+}
+
+static bool
+xfs_dinode_verify(
+ struct xfs_mount *mp,
+ struct xfs_inode *ip,
+ struct xfs_dinode *dip)
+{
+ if (dip->di_magic != cpu_to_be16(XFS_DINODE_MAGIC))
+ return false;
+
+ /* only version 3 or greater inodes are extensively verified here */
+ if (dip->di_version < 3)
+ return true;
+
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return false;
+ if (!xfs_verify_cksum((char *)dip, mp->m_sb.sb_inodesize,
+ offsetof(struct xfs_dinode, di_crc)))
+ return false;
+ if (be64_to_cpu(dip->di_ino) != ip->i_ino)
+ return false;
+ if (!uuid_equal(&dip->di_uuid, &mp->m_sb.sb_uuid))
+ return false;
+ return true;
+}
+
+void
+xfs_dinode_calc_crc(
+ struct xfs_mount *mp,
+ struct xfs_dinode *dip)
+{
+ __uint32_t crc;
+
+ if (dip->di_version < 3)
+ return;
+
+ ASSERT(xfs_sb_version_hascrc(&mp->m_sb));
+ crc = xfs_start_cksum((char *)dip, mp->m_sb.sb_inodesize,
+ offsetof(struct xfs_dinode, di_crc));
+ dip->di_crc = xfs_end_cksum(crc);
}
/*
@@ -638,17 +701,13 @@ xfs_iread(
if (error)
return error;
- /*
- * If we got something that isn't an inode it means someone
- * (nfs or dmi) has a stale handle.
- */
- if (dip->di_magic != cpu_to_be16(XFS_DINODE_MAGIC)) {
-#ifdef DEBUG
- xfs_alert(mp,
- "%s: dip->di_magic (0x%x) != XFS_DINODE_MAGIC (0x%x)",
- __func__, be16_to_cpu(dip->di_magic), XFS_DINODE_MAGIC);
-#endif /* DEBUG */
- error = XFS_ERROR(EINVAL);
+ /* even unallocated inodes are verified */
+ if (!xfs_dinode_verify(mp, ip, dip)) {
+ xfs_alert(mp, "%s: validation failed for inode %lld failed",
+ __func__, ip->i_ino);
+
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, dip);
+ error = XFS_ERROR(EFSCORRUPTED);
goto out_brelse;
}
@@ -670,10 +729,20 @@ xfs_iread(
goto out_brelse;
}
} else {
+ /*
+ * Partial initialisation of the in-core inode. Just the bits
+ * that xfs_ialloc won't overwrite or relies on being correct.
+ */
ip->i_d.di_magic = be16_to_cpu(dip->di_magic);
ip->i_d.di_version = dip->di_version;
ip->i_d.di_gen = be32_to_cpu(dip->di_gen);
ip->i_d.di_flushiter = be16_to_cpu(dip->di_flushiter);
+
+ if (dip->di_version == 3) {
+ ip->i_d.di_ino = be64_to_cpu(dip->di_ino);
+ uuid_copy(&ip->i_d.di_uuid, &dip->di_uuid);
+ }
+
/*
* Make sure to pull in the mode here as well in
* case the inode is released without being used.
diff --git a/logprint/log_misc.c b/logprint/log_misc.c
index 94fcf5b..ffc6916 100644
--- a/logprint/log_misc.c
+++ b/logprint/log_misc.c
@@ -653,7 +653,7 @@ xlog_print_trans_inode(xfs_caddr_t *ptr,
mode = dino.di_mode & S_IFMT;
size = (int)dino.di_size;
xlog_print_trans_inode_core(&dino);
- *ptr += sizeof(xfs_icdinode_t);
+ *ptr += xfs_icdinode_size(&dino);
if (*i == num_ops-1 && f->ilf_size == 3) {
return 1;
diff --git a/logprint/log_print_all.c b/logprint/log_print_all.c
index 44fb4c4..5b2d942 100644
--- a/logprint/log_print_all.c
+++ b/logprint/log_print_all.c
@@ -276,7 +276,8 @@ xlog_recover_print_inode(
f->ilf_dsize);
/* core inode comes 2nd */
- ASSERT(item->ri_buf[1].i_len == sizeof(xfs_icdinode_t));
+ ASSERT(item->ri_buf[1].i_len == xfs_icdinode_size((xfs_icdinode_t *)
+ item->ri_buf[1].i_addr));
xlog_recover_print_inode_core((xfs_icdinode_t *)
item->ri_buf[1].i_addr);
diff --git a/repair/phase6.c b/repair/phase6.c
index 00f70a2..136bb4f 100644
--- a/repair/phase6.c
+++ b/repair/phase6.c
@@ -429,6 +429,8 @@ mk_rbmino(xfs_mount_t *mp)
xfs_bmap_free_t flist;
xfs_dfiloff_t bno;
xfs_bmbt_irec_t map[XFS_BMAP_MAX_NMAP];
+ int vers;
+ int times;
/*
* first set up inode
@@ -445,16 +447,31 @@ mk_rbmino(xfs_mount_t *mp)
error);
}
- memset(&ip->i_d, 0, sizeof(xfs_icdinode_t));
+ vers = xfs_sb_version_hascrc(&mp->m_sb) ? 3 : 1;
+ ip->i_d.di_version = vers;
+ memset(&ip->i_d, 0, xfs_icdinode_size(&ip->i_d));
ip->i_d.di_magic = XFS_DINODE_MAGIC;
ip->i_d.di_mode = S_IFREG;
- ip->i_d.di_version = 1;
+ ip->i_d.di_version = vers;
ip->i_d.di_format = XFS_DINODE_FMT_EXTENTS;
ip->i_d.di_aformat = XFS_DINODE_FMT_EXTENTS;
ip->i_d.di_nlink = 1; /* account for sb ptr */
+ times = XFS_ICHGTIME_CHG | XFS_ICHGTIME_MOD;
+ if (ip->i_d.di_version == 3) {
+ ip->i_d.di_crc = 0;
+ ip->i_d.di_changecount = 1;
+ ip->i_d.di_lsn = 0;
+ ip->i_d.di_flags2 = 0;
+ ip->i_d.di_ino = mp->m_sb.sb_rbmino;
+ memset(&(ip->i_d.di_pad2[0]), 0, sizeof(ip->i_d.di_pad2));
+ platform_uuid_copy(&ip->i_d.di_uuid, &mp->m_sb.sb_uuid);
+ times |= XFS_ICHGTIME_CREATE;
+ }
+ libxfs_trans_ichgtime(tp, ip, times);
+
/*
* now the ifork
*/
@@ -661,6 +678,8 @@ mk_rsumino(xfs_mount_t *mp)
xfs_bmap_free_t flist;
xfs_dfiloff_t bno;
xfs_bmbt_irec_t map[XFS_BMAP_MAX_NMAP];
+ int vers;
+ int times;
/*
* first set up inode
@@ -678,16 +697,31 @@ mk_rsumino(xfs_mount_t *mp)
error);
}
- memset(&ip->i_d, 0, sizeof(xfs_icdinode_t));
+ vers = xfs_sb_version_hascrc(&mp->m_sb) ? 3 : 1;
+ ip->i_d.di_version = vers;
+ memset(&ip->i_d, 0, xfs_icdinode_size(&ip->i_d));
ip->i_d.di_magic = XFS_DINODE_MAGIC;
ip->i_d.di_mode = S_IFREG;
- ip->i_d.di_version = 1;
+ ip->i_d.di_version = vers;
ip->i_d.di_format = XFS_DINODE_FMT_EXTENTS;
ip->i_d.di_aformat = XFS_DINODE_FMT_EXTENTS;
ip->i_d.di_nlink = 1; /* account for sb ptr */
+ times = XFS_ICHGTIME_CHG | XFS_ICHGTIME_MOD;
+ if (ip->i_d.di_version == 3) {
+ ip->i_d.di_crc = 0;
+ ip->i_d.di_changecount = 1;
+ ip->i_d.di_lsn = 0;
+ ip->i_d.di_flags2 = 0;
+ ip->i_d.di_ino = mp->m_sb.sb_rsumino;
+ memset(&(ip->i_d.di_pad2[0]), 0, sizeof(ip->i_d.di_pad2));
+ platform_uuid_copy(&ip->i_d.di_uuid, &mp->m_sb.sb_uuid);
+ times |= XFS_ICHGTIME_CREATE;
+ }
+ libxfs_trans_ichgtime(tp, ip, times);
+
/*
* now the ifork
*/
@@ -760,6 +794,8 @@ mk_root_dir(xfs_mount_t *mp)
int error;
const mode_t mode = 0755;
ino_tree_node_t *irec;
+ int vers;
+ int times;
ASSERT(xfs_sb_version_hasdirv2(&mp->m_sb));
@@ -778,16 +814,31 @@ mk_root_dir(xfs_mount_t *mp)
/*
* take care of the core -- initialization from xfs_ialloc()
*/
- memset(&ip->i_d, 0, sizeof(xfs_icdinode_t));
+ vers = xfs_sb_version_hascrc(&mp->m_sb) ? 3 : 1;
+ ip->i_d.di_version = vers;
+ memset(&ip->i_d, 0, xfs_icdinode_size(&ip->i_d));
ip->i_d.di_magic = XFS_DINODE_MAGIC;
ip->i_d.di_mode = (__uint16_t) mode|S_IFDIR;
- ip->i_d.di_version = 1;
+ ip->i_d.di_version = vers;
ip->i_d.di_format = XFS_DINODE_FMT_EXTENTS;
ip->i_d.di_aformat = XFS_DINODE_FMT_EXTENTS;
ip->i_d.di_nlink = 1; /* account for . */
+ times = XFS_ICHGTIME_CHG | XFS_ICHGTIME_MOD;
+ if (ip->i_d.di_version == 3) {
+ ip->i_d.di_crc = 0;
+ ip->i_d.di_changecount = 1;
+ ip->i_d.di_lsn = 0;
+ ip->i_d.di_flags2 = 0;
+ ip->i_d.di_ino = mp->m_sb.sb_rootino;
+ memset(&(ip->i_d.di_pad2[0]), 0, sizeof(ip->i_d.di_pad2));
+ platform_uuid_copy(&ip->i_d.di_uuid, &mp->m_sb.sb_uuid);
+ times |= XFS_ICHGTIME_CREATE;
+ }
+ libxfs_trans_ichgtime(tp, ip, times);
+
libxfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
/*
xfsprogs-crc-patchset-v2/remote-symlink-crc 0000600 0001750 0001750 00000017013 12132011546 017710 0 ustar dave dave libxfs: add support for crc headers on remote symlinks
From: Dave Chinner <dchinner@xxxxxxxxxx>
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
include/Makefile | 4 +-
include/libxfs.h | 1 +
include/xfs_buf_item.h | 4 +-
include/xfs_symlink.h | 43 ++++++++++++++
libxfs/Makefile | 2 +-
libxfs/xfs_symlink.c | 154 ++++++++++++++++++++++++++++++++++++++++++++++++
6 files changed, 204 insertions(+), 4 deletions(-)
diff --git a/include/Makefile b/include/Makefile
index 61b30a8..b1c685d 100644
--- a/include/Makefile
+++ b/include/Makefile
@@ -28,8 +28,8 @@ QAHFILES = libxfs.h libxlog.h \
xfs_dir_sf.h xfs_extfree_item.h xfs_ialloc.h xfs_ialloc_btree.h \
xfs_inode.h xfs_inode_item.h xfs_inum.h \
xfs_log.h xfs_log_priv.h xfs_log_recover.h xfs_metadump.h \
- xfs_mount.h xfs_quota.h xfs_rtalloc.h xfs_sb.h xfs_trace.h \
- xfs_trans.h xfs_trans_space.h xfs_dfrag.h
+ xfs_mount.h xfs_quota.h xfs_rtalloc.h xfs_sb.h xfs_symlink.h \
+ xfs_trace.h xfs_trans.h xfs_trans_space.h xfs_dfrag.h
HFILES = handle.h jdm.h xqm.h xfs.h xfs_fs.h xfs_types.h
HFILES += $(PKG_PLATFORM).h
diff --git a/include/libxfs.h b/include/libxfs.h
index a4564fd..41cb585 100644
--- a/include/libxfs.h
+++ b/include/libxfs.h
@@ -56,6 +56,7 @@
#include <xfs/xfs_btree_trace.h>
#include <xfs/xfs_bmap.h>
#include <xfs/xfs_trace.h>
+#include <xfs/xfs_symlink.h>
#ifndef ARRAY_SIZE
#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
diff --git a/include/xfs_buf_item.h b/include/xfs_buf_item.h
index abae8c8..09cab4e 100644
--- a/include/xfs_buf_item.h
+++ b/include/xfs_buf_item.h
@@ -49,6 +49,7 @@ extern kmem_zone_t *xfs_buf_item_zone;
#define XFS_BLF_AGFL_BUF (1<<7)
#define XFS_BLF_AGI_BUF (1<<8)
#define XFS_BLF_DINO_BUF (1<<9)
+#define XFS_BLF_SYMLINK_BUF (1<<10)
#define XFS_BLF_TYPE_MASK \
(XFS_BLF_UDQUOT_BUF | \
@@ -58,7 +59,8 @@ extern kmem_zone_t *xfs_buf_item_zone;
XFS_BLF_AGF_BUF | \
XFS_BLF_AGFL_BUF | \
XFS_BLF_AGI_BUF | \
- XFS_BLF_DINO_BUF)
+ XFS_BLF_DINO_BUF | \
+ XFS_BLF_SYMLINK_BUF)
#define XFS_BLF_CHUNK 128
#define XFS_BLF_SHIFT 7
diff --git a/include/xfs_symlink.h b/include/xfs_symlink.h
new file mode 100644
index 0000000..bb21e6a
--- /dev/null
+++ b/include/xfs_symlink.h
@@ -0,0 +1,43 @@
+/*
+ * Copyright (c) 2012 Red Hat, Inc. All rights reserved.
+ */
+#ifndef __XFS_SYMLINK_H
+#define __XFS_SYMLINK_H 1
+
+#define XFS_SYMLINK_MAGIC 0x58534c4d /* XSLM */
+
+struct xfs_dsymlink_hdr {
+ __be32 sl_magic;
+ __be32 sl_offset;
+ __be32 sl_bytes;
+ __be32 sl_crc;
+ uuid_t sl_uuid;
+ __be64 sl_owner;
+ __be64 sl_blkno;
+ __be64 sl_lsn;
+};
+
+/*
+ * The maximum pathlen is 1024 bytes. Since the minimum file system
+ * blocksize is 512 bytes, we can get a max of 3 extents back from
+ * bmapi when crc headers are taken into account.
+ */
+#define XFS_SYMLINK_MAPS 3
+
+#define XFS_SYMLINK_BUF_SPACE(mp, bufsize) \
+ ((bufsize) - (xfs_sb_version_hascrc(&(mp)->m_sb) ? \
+ sizeof(struct xfs_dsymlink_hdr) : 0))
+
+int xfs_symlink_blocks(struct xfs_mount *mp, int pathlen);
+
+extern const struct xfs_buf_ops xfs_symlink_buf_ops;
+
+#ifdef __KERNEL__
+
+int xfs_symlink(struct xfs_inode *dp, struct xfs_name *link_name,
+ const char *target_path, umode_t mode, struct xfs_inode **ipp);
+int xfs_readlink(struct xfs_inode *ip, char *link);
+int xfs_inactive_symlink_rmt(struct xfs_inode *ip, struct xfs_trans **tpp);
+
+#endif /* __KERNEL__ */
+#endif /* __XFS_SYMLINK_H */
diff --git a/libxfs/Makefile b/libxfs/Makefile
index 28f71c8..75f365c 100644
--- a/libxfs/Makefile
+++ b/libxfs/Makefile
@@ -17,7 +17,7 @@ CFILES = cache.c init.c kmem.c logitem.c radix-tree.c rdwr.c trans.c util.c \
xfs_dir2.c xfs_dir2_leaf.c xfs_attr_leaf.c xfs_dir2_block.c \
xfs_dir2_node.c xfs_dir2_data.c xfs_dir2_sf.c xfs_bmap.c \
xfs_mount.c xfs_rtalloc.c xfs_trans.c xfs_attr.c \
- crc32.c
+ crc32.c xfs_symlink.c
CFILES += $(PKG_PLATFORM).c
PCFILES = darwin.c freebsd.c irix.c linux.c
diff --git a/libxfs/xfs_symlink.c b/libxfs/xfs_symlink.c
new file mode 100644
index 0000000..e018abc
--- /dev/null
+++ b/libxfs/xfs_symlink.c
@@ -0,0 +1,154 @@
+/*
+ * Copyright 2013 Red Hat, Inc.
+ * All rights reserved.
+ */
+
+#include "xfs.h"
+
+/*
+ * Each contiguous block has a header, so it is not just a simple pathlen
+ * to FSB conversion.
+ */
+int
+xfs_symlink_blocks(
+ struct xfs_mount *mp,
+ int pathlen)
+{
+ int fsblocks = 0;
+ int len = pathlen;
+
+ do {
+ fsblocks++;
+ len -= XFS_SYMLINK_BUF_SPACE(mp, mp->m_sb.sb_blocksize);
+ } while (len > 0);
+
+ ASSERT(fsblocks <= XFS_SYMLINK_MAPS);
+ return fsblocks;
+}
+
+/*
+ * XXX: this need to be used by mkfs/proto.c to create symlinks.
+ */
+static int
+xfs_symlink_hdr_set(
+ struct xfs_mount *mp,
+ xfs_ino_t ino,
+ uint32_t offset,
+ uint32_t size,
+ struct xfs_buf *bp)
+{
+ struct xfs_dsymlink_hdr *dsl = bp->b_addr;
+
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return 0;
+
+ dsl->sl_magic = cpu_to_be32(XFS_SYMLINK_MAGIC);
+ dsl->sl_offset = cpu_to_be32(offset);
+ dsl->sl_bytes = cpu_to_be32(size);
+ uuid_copy(&dsl->sl_uuid, &mp->m_sb.sb_uuid);
+ dsl->sl_owner = cpu_to_be64(ino);
+ dsl->sl_blkno = cpu_to_be64(bp->b_bn);
+ bp->b_ops = &xfs_symlink_buf_ops;
+
+ return sizeof(struct xfs_dsymlink_hdr);
+}
+
+/*
+ * Checking of the symlink header is split into two parts. the verifier does
+ * CRC, location and bounds checking, the unpacking function checks the path
+ * parameters and owner.
+ */
+bool
+xfs_symlink_hdr_ok(
+ struct xfs_mount *mp,
+ xfs_ino_t ino,
+ uint32_t offset,
+ uint32_t size,
+ struct xfs_buf *bp)
+{
+ struct xfs_dsymlink_hdr *dsl = bp->b_addr;
+
+ if (offset != be32_to_cpu(dsl->sl_offset))
+ return false;
+ if (size != be32_to_cpu(dsl->sl_bytes))
+ return false;
+ if (ino != be64_to_cpu(dsl->sl_owner))
+ return false;
+
+ /* ok */
+ return true;
+
+}
+
+static bool
+xfs_symlink_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_dsymlink_hdr *dsl = bp->b_addr;
+
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return false;
+ if (dsl->sl_magic != cpu_to_be32(XFS_SYMLINK_MAGIC))
+ return false;
+ if (!uuid_equal(&dsl->sl_uuid, &mp->m_sb.sb_uuid))
+ return false;
+ if (bp->b_bn != be64_to_cpu(dsl->sl_blkno))
+ return false;
+ if (be32_to_cpu(dsl->sl_offset) +
+ be32_to_cpu(dsl->sl_bytes) >= MAXPATHLEN)
+ return false;
+ if (dsl->sl_owner == 0)
+ return false;
+
+ return true;
+}
+
+static void
+xfs_symlink_read_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+
+ /* no verification of non-crc buffers */
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return;
+
+ if (!xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
+ offsetof(struct xfs_dsymlink_hdr, sl_crc)) ||
+ !xfs_symlink_verify(bp)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+}
+
+static void
+xfs_symlink_write_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_buf_log_item *bip = bp->b_fspriv;
+
+ /* no verification of non-crc buffers */
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return;
+
+ if (!xfs_symlink_verify(bp)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ return;
+ }
+
+ if (bip) {
+ struct xfs_dsymlink_hdr *dsl = bp->b_addr;
+ dsl->sl_lsn = cpu_to_be64(bip->bli_item.li_lsn);
+ }
+ xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length),
+ offsetof(struct xfs_dsymlink_hdr, sl_crc));
+}
+
+const struct xfs_buf_ops xfs_symlink_buf_ops = {
+ .verify_read = xfs_symlink_read_verify,
+ .verify_write = xfs_symlink_write_verify,
+};
+
xfsprogs-crc-patchset-v2/xfs-add-crc-to-dir2-block-2 0000600 0001750 0001750 00000100777 12132011546 020776 0 ustar dave dave xfs: add CRC checks to block format directory blocks
From: Dave Chinner <dchinner@xxxxxxxxxx>
Now that directory buffers are made from a single struct xfs_buf, we
can add CRC calculation and checking callbacks. While there, add all
the fields to the on disk structures for future functionality such
as d_type support, uuids, block numbers, owner inode, etc.
To distinguish between the different on disk formats, change the
magic numbers for the new format directory blocks.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
include/xfs_dir2_format.h | 155 +++++++++++++++++++++++++++++++++++++++++--
libxfs/xfs_dir2_block.c | 126 +++++++++++++++++++++++++----------
libxfs/xfs_dir2_data.c | 160 ++++++++++++++++++++++++++++-----------------
libxfs/xfs_dir2_leaf.c | 6 +-
libxfs/xfs_dir2_node.c | 2 +-
libxfs/xfs_dir2_priv.h | 4 +-
libxfs/xfs_dir2_sf.c | 2 +-
7 files changed, 346 insertions(+), 109 deletions(-)
diff --git a/include/xfs_dir2_format.h b/include/xfs_dir2_format.h
index 0727098..7dc26b0 100644
--- a/include/xfs_dir2_format.h
+++ b/include/xfs_dir2_format.h
@@ -1,5 +1,6 @@
/*
* Copyright (c) 2000-2001,2005 Silicon Graphics, Inc.
+ * Copyright (c) 2013 Red Hat, Inc.
* All Rights Reserved.
*
* This program is free software; you can redistribute it and/or
@@ -36,6 +37,37 @@
#define XFS_DIR2_FREE_MAGIC 0x58443246 /* XD2F: free index blocks */
/*
+ * Directory Version 3 With CRCs.
+ *
+ * The tree formats are the same as for version 2 directories. The difference
+ * is in the block header and dirent formats. In many cases the v3 structures
+ * use v2 definitions as they are no different and this makes code sharing much
+ * easier.
+ *
+ * Also, the xfs_dir3_*() functions handle both v2 and v3 formats - if the
+ * format is v2 then they switch to the existing v2 code, or the format is v3
+ * they implement the v3 functionality. This means the existing dir2 is a mix of
+ * xfs_dir2/xfs_dir3 calls and functions. The xfs_dir3 functions are called
+ * where there is a difference in the formats, otherwise the code is unchanged.
+ *
+ * Where it is possible, the code decides what to do based on the magic numbers
+ * in the blocks rather than feature bits in the superblock. This means the code
+ * is as independent of the external XFS code as possible as doesn't require
+ * passing struct xfs_mount pointers into places where it isn't really
+ * necessary.
+ *
+ * Version 3 includes:
+ *
+ * - a larger block header for CRC and identification purposes and so the
+ * offsets of all the structures inside the blocks are different.
+ *
+ * - new magic numbers to be able to detect the v2/v3 types on the fly.
+ */
+
+#define XFS_DIR3_BLOCK_MAGIC 0x58444233 /* XDB3: single block dirs */
+#define XFS_DIR3_DATA_MAGIC 0x58444433 /* XDD3: multiblock dirs */
+
+/*
* Byte offset in data block and shortform entry.
*/
typedef __uint16_t xfs_dir2_data_off_t;
@@ -117,19 +149,19 @@ static inline int xfs_dir2_sf_hdr_size(int i8count)
(sizeof(xfs_dir2_ino8_t) - sizeof(xfs_dir2_ino4_t));
}
-static inline xfs_dir2_data_aoff_t
+ static inline xfs_dir2_data_aoff_t
xfs_dir2_sf_get_offset(xfs_dir2_sf_entry_t *sfep)
{
return get_unaligned_be16(&sfep->offset.i);
}
-static inline void
+ static inline void
xfs_dir2_sf_put_offset(xfs_dir2_sf_entry_t *sfep, xfs_dir2_data_aoff_t off)
{
put_unaligned_be16(off, &sfep->offset.i);
}
-static inline int
+ static inline int
xfs_dir2_sf_entsize(struct xfs_dir2_sf_hdr *hdr, int len)
{
return sizeof(struct xfs_dir2_sf_entry) + /* namelen + offset */
@@ -139,14 +171,14 @@ xfs_dir2_sf_entsize(struct xfs_dir2_sf_hdr *hdr, int len)
sizeof(xfs_dir2_ino4_t));
}
-static inline struct xfs_dir2_sf_entry *
+ static inline struct xfs_dir2_sf_entry *
xfs_dir2_sf_firstentry(struct xfs_dir2_sf_hdr *hdr)
{
return (struct xfs_dir2_sf_entry *)
((char *)hdr + xfs_dir2_sf_hdr_size(hdr->i8count));
}
-static inline struct xfs_dir2_sf_entry *
+ static inline struct xfs_dir2_sf_entry *
xfs_dir2_sf_nextentry(struct xfs_dir2_sf_hdr *hdr,
struct xfs_dir2_sf_entry *sfep)
{
@@ -221,11 +253,43 @@ typedef struct xfs_dir2_data_free {
*/
typedef struct xfs_dir2_data_hdr {
__be32 magic; /* XFS_DIR2_DATA_MAGIC or */
- /* XFS_DIR2_BLOCK_MAGIC */
+ /* XFS_DIR2_BLOCK_MAGIC */
xfs_dir2_data_free_t bestfree[XFS_DIR2_DATA_FD_COUNT];
} xfs_dir2_data_hdr_t;
/*
+ * define a structure for all the verification fields we are adding to the
+ * directory block structures. This will be used in several structures.
+ * The magic number must be the first entry to align with all the dir2
+ * structures so we determine how to decode them just by the magic number.
+ */
+struct xfs_dir3_blk_hdr {
+ __be32 magic; /* magic number */
+ __be32 crc; /* CRC of block */
+ __be64 blkno; /* first block of the buffer */
+ __be64 lsn; /* sequence number of last write */
+ uuid_t uuid; /* filesystem we belong to */
+ __be64 owner; /* inode that owns the block */
+};
+
+struct xfs_dir3_data_hdr {
+ struct xfs_dir3_blk_hdr hdr;
+ xfs_dir2_data_free_t best_free[XFS_DIR2_DATA_FD_COUNT];
+};
+
+#define XFS_DIR3_DATA_CRC_OFF offsetof(struct xfs_dir3_data_hdr, hdr.crc)
+
+ static inline struct xfs_dir2_data_free *
+xfs_dir3_data_bestfree_p(struct xfs_dir2_data_hdr *hdr)
+{
+ if (hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC)) {
+ struct xfs_dir3_data_hdr *hdr3 = (struct xfs_dir3_data_hdr *)hdr;
+ return hdr3->best_free;
+ }
+ return hdr->bestfree;
+}
+
+/*
* Active entry in a data block.
*
* Aligned to 8 bytes. After the variable length name field there is a
@@ -280,6 +344,85 @@ xfs_dir2_data_unused_tag_p(struct xfs_dir2_data_unused *dup)
be16_to_cpu(dup->length) - sizeof(__be16));
}
+static inline struct xfs_dir2_data_unused *
+xfs_dir3_data_unused_p(struct xfs_dir2_data_hdr *hdr)
+{
+ if (hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC)) {
+ return (struct xfs_dir2_data_unused *)
+ ((char *)hdr + sizeof(struct xfs_dir3_data_hdr));
+ }
+ return (struct xfs_dir2_data_unused *)
+ ((char *)hdr + sizeof(struct xfs_dir2_data_hdr));
+}
+
+static inline size_t
+xfs_dir3_data_hdr_size(bool dir3)
+{
+ if (dir3)
+ return sizeof(struct xfs_dir3_data_hdr);
+ return sizeof(struct xfs_dir2_data_hdr);
+}
+
+static inline size_t
+xfs_dir3_data_entry_offset(struct xfs_dir2_data_hdr *hdr)
+{
+ bool dir3 = hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC);
+ return xfs_dir3_data_hdr_size(dir3);
+}
+
+static inline struct xfs_dir2_data_entry *
+xfs_dir3_data_entry_p(struct xfs_dir2_data_hdr *hdr)
+{
+ return (struct xfs_dir2_data_entry *)
+ ((char *)hdr + xfs_dir3_data_entry_offset(hdr));
+}
+
+/*
+ * Offsets of . and .. in data space (always block 0)
+ */
+static inline xfs_dir2_data_aoff_t
+xfs_dir3_data_dot_offset(struct xfs_dir2_data_hdr *hdr)
+{
+ return xfs_dir3_data_entry_offset(hdr);
+}
+
+static inline xfs_dir2_data_aoff_t
+xfs_dir3_data_dotdot_offset(struct xfs_dir2_data_hdr *hdr)
+{
+ return xfs_dir3_data_dot_offset(hdr) + xfs_dir2_data_entsize(1);
+}
+
+static inline xfs_dir2_data_aoff_t
+xfs_dir3_data_first_offset(struct xfs_dir2_data_hdr *hdr)
+{
+ return xfs_dir3_data_dotdot_offset(hdr) + xfs_dir2_data_entsize(2);
+}
+
+/*
+ * location of . and .. in data space (always block 0)
+ */
+static inline struct xfs_dir2_data_entry *
+xfs_dir3_data_dot_entry_p(struct xfs_dir2_data_hdr *hdr)
+{
+ return (struct xfs_dir2_data_entry *)
+ ((char *)hdr + xfs_dir3_data_dot_offset(hdr));
+}
+
+static inline struct xfs_dir2_data_entry *
+xfs_dir3_data_dotdot_entry_p(struct xfs_dir2_data_hdr *hdr)
+{
+ return (struct xfs_dir2_data_entry *)
+ ((char *)hdr + xfs_dir3_data_dotdot_offset(hdr));
+}
+
+static inline struct xfs_dir2_data_entry *
+xfs_dir3_data_first_entry_p(struct xfs_dir2_data_hdr *hdr)
+{
+ return (struct xfs_dir2_data_entry *)
+ ((char *)hdr + xfs_dir3_data_first_offset(hdr));
+}
+
/*
* Leaf block structures.
*
diff --git a/libxfs/xfs_dir2_block.c b/libxfs/xfs_dir2_block.c
index 2a99dea..c79199a 100644
--- a/libxfs/xfs_dir2_block.c
+++ b/libxfs/xfs_dir2_block.c
@@ -1,5 +1,6 @@
/*
* Copyright (c) 2000-2003,2005 Silicon Graphics, Inc.
+ * Copyright (c) 2013 Red Hat, Inc.
* All Rights Reserved.
*
* This program is free software; you can redistribute it and/or
@@ -40,44 +41,74 @@ xfs_dir_startup(void)
xfs_dir_hash_dotdot = xfs_da_hashname((unsigned char *)"..", 2);
}
-static void
-xfs_dir2_block_verify(
+static bool
+xfs_dir3_block_verify(
struct xfs_buf *bp)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
- struct xfs_dir2_data_hdr *hdr = bp->b_addr;
- int block_ok = 0;
-
- block_ok = hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC);
- block_ok = block_ok && __xfs_dir2_data_check(NULL, bp) == 0;
-
- if (!block_ok) {
- XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, hdr);
- xfs_buf_ioerror(bp, EFSCORRUPTED);
+ struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr;
+
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ if (hdr3->magic != cpu_to_be32(XFS_DIR3_BLOCK_MAGIC))
+ return false;
+ if (!uuid_equal(&hdr3->uuid, &mp->m_sb.sb_uuid))
+ return false;
+ if (be64_to_cpu(hdr3->blkno) != bp->b_bn)
+ return false;
+ } else {
+ if (hdr3->magic != cpu_to_be32(XFS_DIR2_BLOCK_MAGIC))
+ return false;
}
+ if (__xfs_dir2_data_check(NULL, bp))
+ return false;
+ return true;
}
static void
-xfs_dir2_block_read_verify(
+xfs_dir3_block_read_verify(
struct xfs_buf *bp)
{
- xfs_dir2_block_verify(bp);
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+
+ if ((xfs_sb_version_hascrc(&mp->m_sb) &&
+ !xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
+ XFS_DIR3_DATA_CRC_OFF)) ||
+ !xfs_dir3_block_verify(bp)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
}
static void
-xfs_dir2_block_write_verify(
+xfs_dir3_block_write_verify(
struct xfs_buf *bp)
{
- xfs_dir2_block_verify(bp);
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_buf_log_item *bip = bp->b_fspriv;
+ struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr;
+
+ if (!xfs_dir3_block_verify(bp)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ return;
+ }
+
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return;
+
+ if (bip)
+ hdr3->lsn = cpu_to_be64(bip->bli_item.li_lsn);
+
+ xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), XFS_DIR3_DATA_CRC_OFF);
}
-const struct xfs_buf_ops xfs_dir2_block_buf_ops = {
- .verify_read = xfs_dir2_block_read_verify,
- .verify_write = xfs_dir2_block_write_verify,
+const struct xfs_buf_ops xfs_dir3_block_buf_ops = {
+ .verify_read = xfs_dir3_block_read_verify,
+ .verify_write = xfs_dir3_block_write_verify,
};
static int
-xfs_dir2_block_read(
+xfs_dir3_block_read(
struct xfs_trans *tp,
struct xfs_inode *dp,
struct xfs_buf **bpp)
@@ -85,7 +116,29 @@ xfs_dir2_block_read(
struct xfs_mount *mp = dp->i_mount;
return xfs_da_read_buf(tp, dp, mp->m_dirdatablk, -1, bpp,
- XFS_DATA_FORK, &xfs_dir2_block_buf_ops);
+ XFS_DATA_FORK, &xfs_dir3_block_buf_ops);
+}
+
+static void
+xfs_dir3_block_init(
+ struct xfs_mount *mp,
+ struct xfs_buf *bp,
+ struct xfs_inode *dp)
+{
+ struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr;
+
+ bp->b_ops = &xfs_dir3_block_buf_ops;
+
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ memset(hdr3, 0, sizeof(*hdr3));
+ hdr3->magic = cpu_to_be32(XFS_DIR3_BLOCK_MAGIC);
+ hdr3->blkno = cpu_to_be64(bp->b_bn);
+ hdr3->owner = cpu_to_be64(dp->i_ino);
+ uuid_copy(&hdr3->uuid, &mp->m_sb.sb_uuid);
+ return;
+
+ }
+ hdr3->magic = cpu_to_be32(XFS_DIR2_BLOCK_MAGIC);
}
static void
@@ -105,7 +158,7 @@ xfs_dir2_block_need_space(
struct xfs_dir2_data_unused *enddup = NULL;
*compact = 0;
- bf = hdr->bestfree;
+ bf = xfs_dir3_data_bestfree_p(hdr);
/*
* If there are stale entries we'll use one for the leaf.
@@ -287,7 +340,7 @@ xfs_dir2_block_addname(
mp = dp->i_mount;
/* Read the (one and only) directory block into bp. */
- error = xfs_dir2_block_read(tp, dp, &bp);
+ error = xfs_dir3_block_read(tp, dp, &bp);
if (error)
return error;
@@ -597,7 +650,7 @@ xfs_dir2_block_lookup_int(
tp = args->trans;
mp = dp->i_mount;
- error = xfs_dir2_block_read(tp, dp, &bp);
+ error = xfs_dir3_block_read(tp, dp, &bp);
if (error)
return error;
@@ -860,9 +913,12 @@ xfs_dir2_leaf_to_block(
* These will show up in the leaf bests table.
*/
while (dp->i_d.di_size > mp->m_dirblksize) {
+ int hdrsz;
+
+ hdrsz = xfs_dir3_data_hdr_size(xfs_sb_version_hascrc(&mp->m_sb));
bestsp = xfs_dir2_leaf_bests_p(ltp);
if (be16_to_cpu(bestsp[be32_to_cpu(ltp->bestcount) - 1]) ==
- mp->m_dirblksize - (uint)sizeof(*hdr)) {
+ mp->m_dirblksize - hdrsz) {
if ((error =
xfs_dir2_leaf_trim_data(args, lbp,
(xfs_dir2_db_t)(be32_to_cpu(ltp->bestcount) - 1))))
@@ -900,8 +956,8 @@ xfs_dir2_leaf_to_block(
/*
* Start converting it to block form.
*/
- dbp->b_ops = &xfs_dir2_block_buf_ops;
- hdr->magic = cpu_to_be32(XFS_DIR2_BLOCK_MAGIC);
+ xfs_dir3_block_init(mp, dbp, dp);
+
needlog = 1;
needscan = 0;
/*
@@ -1023,16 +1079,16 @@ xfs_dir2_sf_to_block(
return error;
}
/*
- * Initialize the data block.
+ * Initialize the data block, then convert it to block format.
*/
- error = xfs_dir2_data_init(args, blkno, &bp);
+ error = xfs_dir3_data_init(args, blkno, &bp);
if (error) {
kmem_free(sfp);
return error;
}
- bp->b_ops = &xfs_dir2_block_buf_ops;
+ xfs_dir3_block_init(mp, bp, dp);
hdr = bp->b_addr;
- hdr->magic = cpu_to_be32(XFS_DIR2_BLOCK_MAGIC);
+
/*
* Compute size of block "tail" area.
*/
@@ -1042,7 +1098,7 @@ xfs_dir2_sf_to_block(
* The whole thing is initialized to free by the init routine.
* Say we're using the leaf and tail area.
*/
- dup = (xfs_dir2_data_unused_t *)(hdr + 1);
+ dup = xfs_dir3_data_unused_p(hdr);
needlog = needscan = 0;
xfs_dir2_data_use_free(tp, bp, dup, mp->m_dirblksize - i, i, &needlog,
&needscan);
@@ -1064,8 +1120,7 @@ xfs_dir2_sf_to_block(
/*
* Create entry for .
*/
- dep = (xfs_dir2_data_entry_t *)
- ((char *)hdr + XFS_DIR2_DATA_DOT_OFFSET);
+ dep = xfs_dir3_data_dot_entry_p(hdr);
dep->inumber = cpu_to_be64(dp->i_ino);
dep->namelen = 1;
dep->name[0] = '.';
@@ -1078,8 +1133,7 @@ xfs_dir2_sf_to_block(
/*
* Create entry for ..
*/
- dep = (xfs_dir2_data_entry_t *)
- ((char *)hdr + XFS_DIR2_DATA_DOTDOT_OFFSET);
+ dep = xfs_dir3_data_dotdot_entry_p(hdr);
dep->inumber = cpu_to_be64(xfs_dir2_sf_get_parent_ino(sfp));
dep->namelen = 2;
dep->name[0] = dep->name[1] = '.';
@@ -1089,7 +1143,7 @@ xfs_dir2_sf_to_block(
blp[1].hashval = cpu_to_be32(xfs_dir_hash_dotdot);
blp[1].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp,
(char *)dep - (char *)hdr));
- offset = XFS_DIR2_DATA_FIRST_OFFSET;
+ offset = xfs_dir3_data_first_offset(hdr);
/*
* Loop over existing entries, stuff them in.
*/
diff --git a/libxfs/xfs_dir2_data.c b/libxfs/xfs_dir2_data.c
index eb86739..66aab07 100644
--- a/libxfs/xfs_dir2_data.c
+++ b/libxfs/xfs_dir2_data.c
@@ -1,5 +1,6 @@
/*
* Copyright (c) 2000-2002,2005 Silicon Graphics, Inc.
+ * Copyright (c) 2013 Red Hat, Inc.
* All Rights Reserved.
*
* This program is free software; you can redistribute it and/or
@@ -49,11 +50,12 @@ __xfs_dir2_data_check(
mp = bp->b_target->bt_mount;
hdr = bp->b_addr;
- bf = hdr->bestfree;
- p = (char *)(hdr + 1);
+ bf = xfs_dir3_data_bestfree_p(hdr);
+ p = (char *)xfs_dir3_data_entry_p(hdr);
switch (be32_to_cpu(hdr->magic)) {
case XFS_DIR2_BLOCK_MAGIC:
+ case XFS_DIR3_BLOCK_MAGIC:
btp = xfs_dir2_block_tail_p(mp, hdr);
lep = xfs_dir2_block_leaf_p(btp);
endp = (char *)lep;
@@ -132,7 +134,8 @@ __xfs_dir2_data_check(
(char *)dep - (char *)hdr);
count++;
lastfree = 0;
- if (hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC)) {
+ if (hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC)) {
addr = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk,
(xfs_dir2_data_aoff_t)
((char *)dep - (char *)hdr));
@@ -152,7 +155,8 @@ __xfs_dir2_data_check(
* Need to have seen all the entries and all the bestfree slots.
*/
XFS_WANT_CORRUPTED_RETURN(freeseen == 7);
- if (hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC)) {
+ if (hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC)) {
for (i = stale = 0; i < be32_to_cpu(btp->count); i++) {
if (lep[i].address ==
cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
@@ -200,7 +204,8 @@ xfs_dir2_data_reada_verify(
switch (be32_to_cpu(hdr->magic)) {
case XFS_DIR2_BLOCK_MAGIC:
- bp->b_ops = &xfs_dir2_block_buf_ops;
+ case XFS_DIR3_BLOCK_MAGIC:
+ bp->b_ops = &xfs_dir3_block_buf_ops;
bp->b_ops->verify_read(bp);
return;
case XFS_DIR2_DATA_MAGIC:
@@ -272,12 +277,15 @@ xfs_dir2_data_freefind(
{
xfs_dir2_data_free_t *dfp; /* bestfree entry */
xfs_dir2_data_aoff_t off; /* offset value needed */
+ struct xfs_dir2_data_free *bf;
#if defined(DEBUG) && defined(__KERNEL__)
int matched; /* matched the value */
int seenzero; /* saw a 0 bestfree entry */
#endif
off = (xfs_dir2_data_aoff_t)((char *)dup - (char *)hdr);
+ bf = xfs_dir3_data_bestfree_p(hdr);
+
#if defined(DEBUG) && defined(__KERNEL__)
/*
* Validate some consistency in the bestfree table.
@@ -285,9 +293,10 @@ xfs_dir2_data_freefind(
* one we're looking for it has to be exact.
*/
ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
- hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
- for (dfp = &hdr->bestfree[0], seenzero = matched = 0;
- dfp < &hdr->bestfree[XFS_DIR2_DATA_FD_COUNT];
+ hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
+ for (dfp = &bf[0], seenzero = matched = 0;
+ dfp < &bf[XFS_DIR2_DATA_FD_COUNT];
dfp++) {
if (!dfp->offset) {
ASSERT(!dfp->length);
@@ -303,7 +312,7 @@ xfs_dir2_data_freefind(
else
ASSERT(be16_to_cpu(dfp->offset) + be16_to_cpu(dfp->length) <= off);
ASSERT(matched || be16_to_cpu(dfp->length) >= be16_to_cpu(dup->length));
- if (dfp > &hdr->bestfree[0])
+ if (dfp > &bf[0])
ASSERT(be16_to_cpu(dfp[-1].length) >= be16_to_cpu(dfp[0].length));
}
#endif
@@ -312,14 +321,12 @@ xfs_dir2_data_freefind(
* it can't be there since they're sorted.
*/
if (be16_to_cpu(dup->length) <
- be16_to_cpu(hdr->bestfree[XFS_DIR2_DATA_FD_COUNT - 1].length))
+ be16_to_cpu(bf[XFS_DIR2_DATA_FD_COUNT - 1].length))
return NULL;
/*
* Look at the three bestfree entries for our guy.
*/
- for (dfp = &hdr->bestfree[0];
- dfp < &hdr->bestfree[XFS_DIR2_DATA_FD_COUNT];
- dfp++) {
+ for (dfp = &bf[0]; dfp < &bf[XFS_DIR2_DATA_FD_COUNT]; dfp++) {
if (!dfp->offset)
return NULL;
if (be16_to_cpu(dfp->offset) == off)
@@ -343,11 +350,12 @@ xfs_dir2_data_freeinsert(
xfs_dir2_data_free_t *dfp; /* bestfree table pointer */
xfs_dir2_data_free_t new; /* new bestfree entry */
-#ifdef __KERNEL__
ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
- hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
-#endif
- dfp = hdr->bestfree;
+ hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
+
+ dfp = xfs_dir3_data_bestfree_p(hdr);
new.length = dup->length;
new.offset = cpu_to_be16((char *)dup - (char *)hdr);
@@ -384,32 +392,36 @@ xfs_dir2_data_freeremove(
xfs_dir2_data_free_t *dfp, /* bestfree entry pointer */
int *loghead) /* out: log data header */
{
-#ifdef __KERNEL__
+ struct xfs_dir2_data_free *bf;
+
ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
- hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
-#endif
+ hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
+
/*
* It's the first entry, slide the next 2 up.
*/
- if (dfp == &hdr->bestfree[0]) {
- hdr->bestfree[0] = hdr->bestfree[1];
- hdr->bestfree[1] = hdr->bestfree[2];
+ bf = xfs_dir3_data_bestfree_p(hdr);
+ if (dfp == &bf[0]) {
+ bf[0] = bf[1];
+ bf[1] = bf[2];
}
/*
* It's the second entry, slide the 3rd entry up.
*/
- else if (dfp == &hdr->bestfree[1])
- hdr->bestfree[1] = hdr->bestfree[2];
+ else if (dfp == &bf[1])
+ bf[1] = bf[2];
/*
* Must be the last entry.
*/
else
- ASSERT(dfp == &hdr->bestfree[2]);
+ ASSERT(dfp == &bf[2]);
/*
* Clear the 3rd entry, must be zero now.
*/
- hdr->bestfree[2].length = 0;
- hdr->bestfree[2].offset = 0;
+ bf[2].length = 0;
+ bf[2].offset = 0;
*loghead = 1;
}
@@ -425,23 +437,26 @@ xfs_dir2_data_freescan(
xfs_dir2_block_tail_t *btp; /* block tail */
xfs_dir2_data_entry_t *dep; /* active data entry */
xfs_dir2_data_unused_t *dup; /* unused data entry */
+ struct xfs_dir2_data_free *bf;
char *endp; /* end of block's data */
char *p; /* current entry pointer */
-#ifdef __KERNEL__
ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
- hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
-#endif
+ hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
+
/*
* Start by clearing the table.
*/
- memset(hdr->bestfree, 0, sizeof(hdr->bestfree));
+ bf = xfs_dir3_data_bestfree_p(hdr);
+ memset(bf, 0, sizeof(*bf) * XFS_DIR2_DATA_FD_COUNT);
*loghead = 1;
/*
* Set up pointers.
*/
- p = (char *)(hdr + 1);
- if (hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC)) {
+ p = (char *)xfs_dir3_data_entry_p(hdr);
+ if (hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC)) {
btp = xfs_dir2_block_tail_p(mp, hdr);
endp = (char *)xfs_dir2_block_leaf_p(btp);
} else
@@ -477,7 +492,7 @@ xfs_dir2_data_freescan(
* Give back the buffer for the created block.
*/
int /* error */
-xfs_dir2_data_init(
+xfs_dir3_data_init(
xfs_da_args_t *args, /* directory operation args */
xfs_dir2_db_t blkno, /* logical dir block number */
struct xfs_buf **bpp) /* output block buffer */
@@ -486,6 +501,7 @@ xfs_dir2_data_init(
xfs_dir2_data_hdr_t *hdr; /* data block header */
xfs_inode_t *dp; /* incore directory inode */
xfs_dir2_data_unused_t *dup; /* unused entry pointer */
+ struct xfs_dir2_data_free *bf;
int error; /* error return value */
int i; /* bestfree index */
xfs_mount_t *mp; /* filesystem mount point */
@@ -508,21 +524,34 @@ xfs_dir2_data_init(
* Initialize the header.
*/
hdr = bp->b_addr;
- hdr->magic = cpu_to_be32(XFS_DIR2_DATA_MAGIC);
- hdr->bestfree[0].offset = cpu_to_be16(sizeof(*hdr));
+
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr;
+
+ memset(hdr3, 0, sizeof(*hdr3));
+ hdr3->magic = cpu_to_be32(XFS_DIR3_DATA_MAGIC);
+ hdr3->blkno = cpu_to_be64(bp->b_bn);
+ hdr3->owner = cpu_to_be64(dp->i_ino);
+ uuid_copy(&hdr3->uuid, &mp->m_sb.sb_uuid);
+
+ } else
+ hdr->magic = cpu_to_be32(XFS_DIR2_DATA_MAGIC);
+
+ bf = xfs_dir3_data_bestfree_p(hdr);
+ bf[0].offset = cpu_to_be16(xfs_dir3_data_entry_offset(hdr));
for (i = 1; i < XFS_DIR2_DATA_FD_COUNT; i++) {
- hdr->bestfree[i].length = 0;
- hdr->bestfree[i].offset = 0;
+ bf[i].length = 0;
+ bf[i].offset = 0;
}
/*
* Set up an unused entry for the block's body.
*/
- dup = (xfs_dir2_data_unused_t *)(hdr + 1);
+ dup = xfs_dir3_data_unused_p(hdr);
dup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
- t = mp->m_dirblksize - (uint)sizeof(*hdr);
- hdr->bestfree[0].length = cpu_to_be16(t);
+ t = mp->m_dirblksize - (uint)xfs_dir3_data_entry_offset(hdr);
+ bf[0].length = cpu_to_be16(t);
dup->length = cpu_to_be16(t);
*xfs_dir2_data_unused_tag_p(dup) = cpu_to_be16((char *)dup - (char *)hdr);
/*
@@ -546,7 +575,8 @@ xfs_dir2_data_log_entry(
xfs_dir2_data_hdr_t *hdr = bp->b_addr;
ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
- hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
+ hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
xfs_trans_log_buf(tp, bp, (uint)((char *)dep - (char *)hdr),
(uint)((char *)(xfs_dir2_data_entry_tag_p(dep) + 1) -
@@ -564,9 +594,10 @@ xfs_dir2_data_log_header(
xfs_dir2_data_hdr_t *hdr = bp->b_addr;
ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
- hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
+ hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
- xfs_trans_log_buf(tp, bp, 0, sizeof(*hdr) - 1);
+ xfs_trans_log_buf(tp, bp, 0, xfs_dir3_data_entry_offset(hdr) - 1);
}
/*
@@ -581,7 +612,8 @@ xfs_dir2_data_log_unused(
xfs_dir2_data_hdr_t *hdr = bp->b_addr;
ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
- hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
+ hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
/*
* Log the first part of the unused entry.
@@ -619,6 +651,7 @@ xfs_dir2_data_make_free(
xfs_dir2_data_unused_t *newdup; /* new unused entry */
xfs_dir2_data_unused_t *postdup; /* unused entry after us */
xfs_dir2_data_unused_t *prevdup; /* unused entry before us */
+ struct xfs_dir2_data_free *bf;
mp = tp->t_mountp;
hdr = bp->b_addr;
@@ -631,7 +664,8 @@ xfs_dir2_data_make_free(
else {
xfs_dir2_block_tail_t *btp; /* block tail */
- ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
+ ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
btp = xfs_dir2_block_tail_p(mp, hdr);
endptr = (char *)xfs_dir2_block_leaf_p(btp);
}
@@ -639,7 +673,7 @@ xfs_dir2_data_make_free(
* If this isn't the start of the block, then back up to
* the previous entry and see if it's free.
*/
- if (offset > sizeof(*hdr)) {
+ if (offset > xfs_dir3_data_entry_offset(hdr)) {
__be16 *tagp; /* tag just before us */
tagp = (__be16 *)((char *)hdr + offset) - 1;
@@ -665,6 +699,7 @@ xfs_dir2_data_make_free(
* Previous and following entries are both free,
* merge everything into a single free entry.
*/
+ bf = xfs_dir3_data_bestfree_p(hdr);
if (prevdup && postdup) {
xfs_dir2_data_free_t *dfp2; /* another bestfree pointer */
@@ -679,7 +714,7 @@ xfs_dir2_data_make_free(
* since the third bestfree is there, there might be more
* entries.
*/
- needscan = (hdr->bestfree[2].length != 0);
+ needscan = (bf[2].length != 0);
/*
* Fix up the new big freespace.
*/
@@ -695,10 +730,10 @@ xfs_dir2_data_make_free(
* Remove entry 1 first then entry 0.
*/
ASSERT(dfp && dfp2);
- if (dfp == &hdr->bestfree[1]) {
- dfp = &hdr->bestfree[0];
+ if (dfp == &bf[1]) {
+ dfp = &bf[0];
ASSERT(dfp2 == dfp);
- dfp2 = &hdr->bestfree[1];
+ dfp2 = &bf[1];
}
xfs_dir2_data_freeremove(hdr, dfp2, needlogp);
xfs_dir2_data_freeremove(hdr, dfp, needlogp);
@@ -706,7 +741,7 @@ xfs_dir2_data_make_free(
* Now insert the new entry.
*/
dfp = xfs_dir2_data_freeinsert(hdr, prevdup, needlogp);
- ASSERT(dfp == &hdr->bestfree[0]);
+ ASSERT(dfp == &bf[0]);
ASSERT(dfp->length == prevdup->length);
ASSERT(!dfp[1].length);
ASSERT(!dfp[2].length);
@@ -735,7 +770,7 @@ xfs_dir2_data_make_free(
*/
else {
needscan = be16_to_cpu(prevdup->length) >
- be16_to_cpu(hdr->bestfree[2].length);
+ be16_to_cpu(bf[2].length);
}
}
/*
@@ -763,7 +798,7 @@ xfs_dir2_data_make_free(
*/
else {
needscan = be16_to_cpu(newdup->length) >
- be16_to_cpu(hdr->bestfree[2].length);
+ be16_to_cpu(bf[2].length);
}
}
/*
@@ -802,10 +837,12 @@ xfs_dir2_data_use_free(
xfs_dir2_data_unused_t *newdup; /* new unused entry */
xfs_dir2_data_unused_t *newdup2; /* another new unused entry */
int oldlen; /* old unused entry's length */
+ struct xfs_dir2_data_free *bf;
hdr = bp->b_addr;
ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
- hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC));
+ hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
ASSERT(be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG);
ASSERT(offset >= (char *)dup - (char *)hdr);
ASSERT(offset + len <= (char *)dup + be16_to_cpu(dup->length) - (char *)hdr);
@@ -815,7 +852,8 @@ xfs_dir2_data_use_free(
*/
dfp = xfs_dir2_data_freefind(hdr, dup);
oldlen = be16_to_cpu(dup->length);
- ASSERT(dfp || oldlen <= be16_to_cpu(hdr->bestfree[2].length));
+ bf = xfs_dir3_data_bestfree_p(hdr);
+ ASSERT(dfp || oldlen <= be16_to_cpu(bf[2].length));
/*
* Check for alignment with front and back of the entry.
*/
@@ -829,7 +867,7 @@ xfs_dir2_data_use_free(
*/
if (matchfront && matchback) {
if (dfp) {
- needscan = (hdr->bestfree[2].offset != 0);
+ needscan = (bf[2].offset != 0);
if (!needscan)
xfs_dir2_data_freeremove(hdr, dfp, needlogp);
}
@@ -859,7 +897,7 @@ xfs_dir2_data_use_free(
* that means we don't know if there was a better
* choice for the last slot, or not. Rescan.
*/
- needscan = dfp == &hdr->bestfree[2];
+ needscan = dfp == &bf[2];
}
}
/*
@@ -886,7 +924,7 @@ xfs_dir2_data_use_free(
* that means we don't know if there was a better
* choice for the last slot, or not. Rescan.
*/
- needscan = dfp == &hdr->bestfree[2];
+ needscan = dfp == &bf[2];
}
}
/*
@@ -914,7 +952,7 @@ xfs_dir2_data_use_free(
* the 2 new will work.
*/
if (dfp) {
- needscan = (hdr->bestfree[2].length != 0);
+ needscan = (bf[2].length != 0);
if (!needscan) {
xfs_dir2_data_freeremove(hdr, dfp, needlogp);
xfs_dir2_data_freeinsert(hdr, newdup, needlogp);
diff --git a/libxfs/xfs_dir2_leaf.c b/libxfs/xfs_dir2_leaf.c
index d303813..d83fce4 100644
--- a/libxfs/xfs_dir2_leaf.c
+++ b/libxfs/xfs_dir2_leaf.c
@@ -133,6 +133,7 @@ xfs_dir2_block_to_leaf(
int needlog; /* need to log block header */
int needscan; /* need to rescan bestfree */
xfs_trans_t *tp; /* transaction pointer */
+ struct xfs_dir2_data_free *bf;
trace_xfs_dir2_block_to_leaf(args);
@@ -161,6 +162,7 @@ xfs_dir2_block_to_leaf(
xfs_dir2_data_check(dp, dbp);
btp = xfs_dir2_block_tail_p(mp, hdr);
blp = xfs_dir2_block_leaf_p(btp);
+ bf = xfs_dir3_data_bestfree_p(hdr);
/*
* Set the counts in the leaf header.
*/
@@ -196,7 +198,7 @@ xfs_dir2_block_to_leaf(
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
ltp->bestcount = cpu_to_be32(1);
bestsp = xfs_dir2_leaf_bests_p(ltp);
- bestsp[0] = hdr->bestfree[0].length;
+ bestsp[0] = bf[0].length;
/*
* Log the data header and leaf bests table.
*/
@@ -528,7 +530,7 @@ xfs_dir2_leaf_addname(
/*
* Initialize the block.
*/
- if ((error = xfs_dir2_data_init(args, use_block, &dbp))) {
+ if ((error = xfs_dir3_data_init(args, use_block, &dbp))) {
xfs_trans_brelse(tp, lbp);
return error;
}
diff --git a/libxfs/xfs_dir2_node.c b/libxfs/xfs_dir2_node.c
index 649f677..e7820b2 100644
--- a/libxfs/xfs_dir2_node.c
+++ b/libxfs/xfs_dir2_node.c
@@ -1573,7 +1573,7 @@ xfs_dir2_node_addname_int(
if (unlikely((error = xfs_dir2_grow_inode(args,
XFS_DIR2_DATA_SPACE,
&dbno)) ||
- (error = xfs_dir2_data_init(args, dbno, &dbp))))
+ (error = xfs_dir3_data_init(args, dbno, &dbp))))
return error;
/*
diff --git a/libxfs/xfs_dir2_priv.h b/libxfs/xfs_dir2_priv.h
index 7da79f6..e6f2e0a 100644
--- a/libxfs/xfs_dir2_priv.h
+++ b/libxfs/xfs_dir2_priv.h
@@ -30,7 +30,7 @@ extern int xfs_dir_cilookup_result(struct xfs_da_args *args,
const unsigned char *name, int len);
/* xfs_dir2_block.c */
-extern const struct xfs_buf_ops xfs_dir2_block_buf_ops;
+extern const struct xfs_buf_ops xfs_dir3_block_buf_ops;
extern int xfs_dir2_block_addname(struct xfs_da_args *args);
extern int xfs_dir2_block_getdents(struct xfs_inode *dp, void *dirent,
@@ -61,7 +61,7 @@ xfs_dir2_data_freeinsert(struct xfs_dir2_data_hdr *hdr,
struct xfs_dir2_data_unused *dup, int *loghead);
extern void xfs_dir2_data_freescan(struct xfs_mount *mp,
struct xfs_dir2_data_hdr *hdr, int *loghead);
-extern int xfs_dir2_data_init(struct xfs_da_args *args, xfs_dir2_db_t blkno,
+extern int xfs_dir3_data_init(struct xfs_da_args *args, xfs_dir2_db_t blkno,
struct xfs_buf **bpp);
extern void xfs_dir2_data_log_entry(struct xfs_trans *tp, struct xfs_buf *bp,
struct xfs_dir2_data_entry *dep);
diff --git a/libxfs/xfs_dir2_sf.c b/libxfs/xfs_dir2_sf.c
index a96be76..6848d05 100644
--- a/libxfs/xfs_dir2_sf.c
+++ b/libxfs/xfs_dir2_sf.c
@@ -262,7 +262,7 @@ xfs_dir2_block_to_sf(
* Set up to loop over the block's entries.
*/
btp = xfs_dir2_block_tail_p(mp, hdr);
- ptr = (char *)(hdr + 1);
+ ptr = (char *)xfs_dir3_data_entry_p(hdr);
endptr = (char *)xfs_dir2_block_leaf_p(btp);
sfep = xfs_dir2_sf_firstentry(sfp);
/*
xfsprogs-crc-patchset-v2/xfs-add-crc-to-dir2-free-2 0000600 0001750 0001750 00000071727 12132011546 020627 0 ustar dave dave xfs: add CRC checking to dir2 free blocks
From: Dave Chinner <dchinner@xxxxxxxxxx>
This addition follows the same pattern as the dir2 block CRCs, but
with a few differences. The main difference is that the free block
header is different between the v2 and v3 formats, so an "in-core"
free block header has been added and _todisk/_from_disk functions
used to abstract the differences in structure format from the code.
This is similar to the on-disk superblock versus the in-core
superblock setup. The in-core strucutre is populated when the buffer
is read from disk, all the in memory checks and modifications are
done on the in-core version of the structure which is written back
to the buffer before the buffer is logged.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
db/check.c | 2 +-
include/xfs_dir2_format.h | 55 +++++-
libxfs/xfs_dir2_leaf.c | 15 +-
libxfs/xfs_dir2_node.c | 474 ++++++++++++++++++++++++++++++---------------
repair/phase6.c | 2 +-
5 files changed, 384 insertions(+), 164 deletions(-)
diff --git a/db/check.c b/db/check.c
index 1dfef74..368d2b7 100644
--- a/db/check.c
+++ b/db/check.c
@@ -3169,7 +3169,7 @@ process_leaf_node_dir_v2_free(
error++;
return;
}
- maxent = xfs_dir2_free_max_bests(mp);
+ maxent = xfs_dir3_free_max_bests(mp);
if (be32_to_cpu(free->hdr.firstdb) != xfs_dir2_da_to_db(mp,
dabno - mp->m_dirfreeblk) * maxent) {
if (!sflag || v)
diff --git a/include/xfs_dir2_format.h b/include/xfs_dir2_format.h
index 7dc26b0..bec058f 100644
--- a/include/xfs_dir2_format.h
+++ b/include/xfs_dir2_format.h
@@ -66,6 +66,7 @@
#define XFS_DIR3_BLOCK_MAGIC 0x58444233 /* XDB3: single block dirs */
#define XFS_DIR3_DATA_MAGIC 0x58444433 /* XDD3: multiblock dirs */
+#define XFS_DIR3_FREE_MAGIC 0x58444633 /* XDF3: free index blocks */
/*
* Byte offset in data block and shortform entry.
@@ -663,19 +664,65 @@ typedef struct xfs_dir2_free {
/* unused entries are -1 */
} xfs_dir2_free_t;
-static inline int xfs_dir2_free_max_bests(struct xfs_mount *mp)
+struct xfs_dir3_free_hdr {
+ struct xfs_dir3_blk_hdr hdr;
+ __be32 firstdb; /* db of first entry */
+ __be32 nvalid; /* count of valid entries */
+ __be32 nused; /* count of used entries */
+};
+
+struct xfs_dir3_free {
+ struct xfs_dir3_free_hdr hdr;
+ __be16 bests[]; /* best free counts */
+ /* unused entries are -1 */
+};
+
+#define XFS_DIR3_FREE_CRC_OFF offsetof(struct xfs_dir3_free, hdr.hdr.crc)
+
+/*
+ * In core version of the free block header, abstracted away from on-disk format
+ * differences. Use this in the code, and convert to/from the disk version using
+ * xfs_dir3_free_hdr_from_disk/xfs_dir3_free_hdr_to_disk.
+ */
+struct xfs_dir3_icfree_hdr {
+ __uint32_t magic;
+ __uint32_t firstdb;
+ __uint32_t nvalid;
+ __uint32_t nused;
+
+};
+
+void xfs_dir3_free_hdr_from_disk(struct xfs_dir3_icfree_hdr *to,
+ struct xfs_dir2_free *from);
+
+static inline int
+xfs_dir3_free_hdr_size(struct xfs_mount *mp)
{
- return (mp->m_dirblksize - sizeof(struct xfs_dir2_free_hdr)) /
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ return sizeof(struct xfs_dir3_free_hdr);
+ return sizeof(struct xfs_dir2_free_hdr);
+}
+
+static inline int
+xfs_dir3_free_max_bests(struct xfs_mount *mp)
+{
+ return (mp->m_dirblksize - xfs_dir3_free_hdr_size(mp)) /
sizeof(xfs_dir2_data_off_t);
}
+static inline __be16 *
+xfs_dir3_free_bests_p(struct xfs_mount *mp, struct xfs_dir2_free *free)
+{
+ return (__be16 *)((char *)free + xfs_dir3_free_hdr_size(mp));
+}
+
/*
* Convert data space db to the corresponding free db.
*/
static inline xfs_dir2_db_t
xfs_dir2_db_to_fdb(struct xfs_mount *mp, xfs_dir2_db_t db)
{
- return XFS_DIR2_FREE_FIRSTDB(mp) + db / xfs_dir2_free_max_bests(mp);
+ return XFS_DIR2_FREE_FIRSTDB(mp) + db / xfs_dir3_free_max_bests(mp);
}
/*
@@ -684,7 +731,7 @@ xfs_dir2_db_to_fdb(struct xfs_mount *mp, xfs_dir2_db_t db)
static inline int
xfs_dir2_db_to_fdindex(struct xfs_mount *mp, xfs_dir2_db_t db)
{
- return db % xfs_dir2_free_max_bests(mp);
+ return db % xfs_dir3_free_max_bests(mp);
}
/*
diff --git a/libxfs/xfs_dir2_leaf.c b/libxfs/xfs_dir2_leaf.c
index d83fce4..a1df347 100644
--- a/libxfs/xfs_dir2_leaf.c
+++ b/libxfs/xfs_dir2_leaf.c
@@ -1477,6 +1477,7 @@ xfs_dir2_node_to_leaf(
xfs_mount_t *mp; /* filesystem mount point */
int rval; /* successful free trim? */
xfs_trans_t *tp; /* transaction pointer */
+ struct xfs_dir3_icfree_hdr freehdr;
/*
* There's more than a leaf level in the btree, so there must
@@ -1534,15 +1535,15 @@ xfs_dir2_node_to_leaf(
if (error)
return error;
free = fbp->b_addr;
- ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
- ASSERT(!free->hdr.firstdb);
+ xfs_dir3_free_hdr_from_disk(&freehdr, free);
+
+ ASSERT(!freehdr.firstdb);
/*
* Now see if the leafn and free data will fit in a leaf1.
* If not, release the buffer and give up.
*/
- if (xfs_dir2_leaf_size(&leaf->hdr, be32_to_cpu(free->hdr.nvalid)) >
- mp->m_dirblksize) {
+ if (xfs_dir2_leaf_size(&leaf->hdr, freehdr.nvalid) > mp->m_dirblksize) {
xfs_trans_brelse(tp, fbp);
return 0;
}
@@ -1563,12 +1564,12 @@ xfs_dir2_node_to_leaf(
* Set up the leaf tail from the freespace block.
*/
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
- ltp->bestcount = free->hdr.nvalid;
+ ltp->bestcount = cpu_to_be32(freehdr.nvalid);
/*
* Set up the leaf bests table.
*/
- memcpy(xfs_dir2_leaf_bests_p(ltp), free->bests,
- be32_to_cpu(ltp->bestcount) * sizeof(xfs_dir2_data_off_t));
+ memcpy(xfs_dir2_leaf_bests_p(ltp), xfs_dir3_free_bests_p(mp, free),
+ freehdr.nvalid * sizeof(xfs_dir2_data_off_t));
xfs_dir2_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
xfs_dir2_leaf_log_tail(tp, lbp);
xfs_dir2_leaf_check(dp, lbp);
diff --git a/libxfs/xfs_dir2_node.c b/libxfs/xfs_dir2_node.c
index e7820b2..e1d1f22 100644
--- a/libxfs/xfs_dir2_node.c
+++ b/libxfs/xfs_dir2_node.c
@@ -1,5 +1,6 @@
/*
* Copyright (c) 2000-2005 Silicon Graphics, Inc.
+ * Copyright (c) 2013 Red Hat, Inc.
* All Rights Reserved.
*
* This program is free software; you can redistribute it and/or
@@ -40,44 +41,78 @@ static int xfs_dir2_leafn_remove(xfs_da_args_t *args, struct xfs_buf *bp,
static int xfs_dir2_node_addname_int(xfs_da_args_t *args,
xfs_da_state_blk_t *fblk);
-static void
-xfs_dir2_free_verify(
+static bool
+xfs_dir3_free_verify(
struct xfs_buf *bp)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
struct xfs_dir2_free_hdr *hdr = bp->b_addr;
- int block_ok = 0;
- block_ok = hdr->magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC);
- if (!block_ok) {
- XFS_CORRUPTION_ERROR("xfs_dir2_free_verify magic",
- XFS_ERRLEVEL_LOW, mp, hdr);
- xfs_buf_ioerror(bp, EFSCORRUPTED);
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr;
+
+ if (hdr3->magic != cpu_to_be32(XFS_DIR3_FREE_MAGIC))
+ return false;
+ if (!uuid_equal(&hdr3->uuid, &mp->m_sb.sb_uuid))
+ return false;
+ if (be64_to_cpu(hdr3->blkno) != bp->b_bn)
+ return false;
+ } else {
+ if (hdr->magic != cpu_to_be32(XFS_DIR2_FREE_MAGIC))
+ return false;
}
+
+ /* XXX: should bounds check the xfs_dir3_icfree_hdr here */
+
+ return true;
}
static void
-xfs_dir2_free_read_verify(
+xfs_dir3_free_read_verify(
struct xfs_buf *bp)
{
- xfs_dir2_free_verify(bp);
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+
+ if ((xfs_sb_version_hascrc(&mp->m_sb) &&
+ !xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
+ XFS_DIR3_FREE_CRC_OFF)) ||
+ !xfs_dir3_free_verify(bp)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
}
static void
-xfs_dir2_free_write_verify(
+xfs_dir3_free_write_verify(
struct xfs_buf *bp)
{
- xfs_dir2_free_verify(bp);
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_buf_log_item *bip = bp->b_fspriv;
+ struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr;
+
+ if (!xfs_dir3_free_verify(bp)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ return;
+ }
+
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return;
+
+ if (bip)
+ hdr3->lsn = cpu_to_be64(bip->bli_item.li_lsn);
+
+ xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), XFS_DIR3_FREE_CRC_OFF);
}
-static const struct xfs_buf_ops xfs_dir2_free_buf_ops = {
- .verify_read = xfs_dir2_free_read_verify,
- .verify_write = xfs_dir2_free_write_verify,
+static const struct xfs_buf_ops xfs_dir3_free_buf_ops = {
+ .verify_read = xfs_dir3_free_read_verify,
+ .verify_write = xfs_dir3_free_write_verify,
};
static int
-__xfs_dir2_free_read(
+__xfs_dir3_free_read(
struct xfs_trans *tp,
struct xfs_inode *dp,
xfs_dablk_t fbno,
@@ -85,7 +120,7 @@ __xfs_dir2_free_read(
struct xfs_buf **bpp)
{
return xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
- XFS_DATA_FORK, &xfs_dir2_free_buf_ops);
+ XFS_DATA_FORK, &xfs_dir3_free_buf_ops);
}
int
@@ -95,7 +130,7 @@ xfs_dir2_free_read(
xfs_dablk_t fbno,
struct xfs_buf **bpp)
{
- return __xfs_dir2_free_read(tp, dp, fbno, -1, bpp);
+ return __xfs_dir3_free_read(tp, dp, fbno, -1, bpp);
}
static int
@@ -105,7 +140,95 @@ xfs_dir2_free_try_read(
xfs_dablk_t fbno,
struct xfs_buf **bpp)
{
- return __xfs_dir2_free_read(tp, dp, fbno, -2, bpp);
+ return __xfs_dir3_free_read(tp, dp, fbno, -2, bpp);
+}
+
+
+void
+xfs_dir3_free_hdr_from_disk(
+ struct xfs_dir3_icfree_hdr *to,
+ struct xfs_dir2_free *from)
+{
+ if (from->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC)) {
+ to->magic = be32_to_cpu(from->hdr.magic);
+ to->firstdb = be32_to_cpu(from->hdr.firstdb);
+ to->nvalid = be32_to_cpu(from->hdr.nvalid);
+ to->nused = be32_to_cpu(from->hdr.nused);
+ } else {
+ struct xfs_dir3_free_hdr *hdr3 = (struct xfs_dir3_free_hdr *)from;
+
+ to->magic = be32_to_cpu(hdr3->hdr.magic);
+ to->firstdb = be32_to_cpu(hdr3->firstdb);
+ to->nvalid = be32_to_cpu(hdr3->nvalid);
+ to->nused = be32_to_cpu(hdr3->nused);
+ }
+
+ ASSERT(to->magic == XFS_DIR2_FREE_MAGIC ||
+ to->magic == XFS_DIR3_FREE_MAGIC);
+}
+
+static void
+xfs_dir3_free_hdr_to_disk(
+ struct xfs_dir2_free *to,
+ struct xfs_dir3_icfree_hdr *from)
+{
+ ASSERT(from->magic == XFS_DIR2_FREE_MAGIC ||
+ from->magic == XFS_DIR3_FREE_MAGIC);
+
+ if (from->magic == XFS_DIR2_FREE_MAGIC) {
+ to->hdr.magic = cpu_to_be32(from->magic);
+ to->hdr.firstdb = cpu_to_be32(from->firstdb);
+ to->hdr.nvalid = cpu_to_be32(from->nvalid);
+ to->hdr.nused = cpu_to_be32(from->nused);
+ } else {
+ struct xfs_dir3_free_hdr *hdr3 = (struct xfs_dir3_free_hdr *)to;
+
+ hdr3->hdr.magic = cpu_to_be32(from->magic);
+ hdr3->firstdb = cpu_to_be32(from->firstdb);
+ hdr3->nvalid = cpu_to_be32(from->nvalid);
+ hdr3->nused = cpu_to_be32(from->nused);
+ }
+}
+
+static int
+xfs_dir3_free_get_buf(
+ struct xfs_trans *tp,
+ struct xfs_inode *dp,
+ xfs_dir2_db_t fbno,
+ struct xfs_buf **bpp)
+{
+ struct xfs_mount *mp = dp->i_mount;
+ struct xfs_buf *bp;
+ int error;
+ struct xfs_dir3_icfree_hdr hdr;
+
+ error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(mp, fbno),
+ -1, &bp, XFS_DATA_FORK);
+ if (error)
+ return error;
+
+ bp->b_ops = &xfs_dir3_free_buf_ops;;
+
+ /*
+ * Initialize the new block to be empty, and remember
+ * its first slot as our empty slot.
+ */
+ hdr.magic = XFS_DIR2_FREE_MAGIC;
+ hdr.firstdb = 0;
+ hdr.nused = 0;
+ hdr.nvalid = 0;
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ struct xfs_dir3_free_hdr *hdr3 = bp->b_addr;
+
+ hdr.magic = XFS_DIR3_FREE_MAGIC;
+ hdr3->hdr.blkno = cpu_to_be64(bp->b_bn);
+ hdr3->hdr.owner = cpu_to_be64(dp->i_ino);
+ uuid_copy(&hdr3->hdr.uuid, &mp->m_sb.sb_uuid);
+
+ }
+ xfs_dir3_free_hdr_to_disk(bp->b_addr, &hdr);
+ *bpp = bp;
+ return 0;
}
/*
@@ -119,13 +242,16 @@ xfs_dir2_free_log_bests(
int last) /* last entry to log */
{
xfs_dir2_free_t *free; /* freespace structure */
+ __be16 *bests;
free = bp->b_addr;
- ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
+ bests = xfs_dir3_free_bests_p(tp->t_mountp, free);
+ ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC) ||
+ free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC));
xfs_trans_log_buf(tp, bp,
- (uint)((char *)&free->bests[first] - (char *)free),
- (uint)((char *)&free->bests[last] - (char *)free +
- sizeof(free->bests[0]) - 1));
+ (uint)((char *)&bests[first] - (char *)free),
+ (uint)((char *)&bests[last] - (char *)free +
+ sizeof(bests[0]) - 1));
}
/*
@@ -139,9 +265,9 @@ xfs_dir2_free_log_header(
xfs_dir2_free_t *free; /* freespace structure */
free = bp->b_addr;
- ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
- xfs_trans_log_buf(tp, bp, (uint)((char *)&free->hdr - (char *)free),
- (uint)(sizeof(xfs_dir2_free_hdr_t) - 1));
+ ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC) ||
+ free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC));
+ xfs_trans_log_buf(tp, bp, 0, xfs_dir3_free_hdr_size(tp->t_mountp) - 1);
}
/*
@@ -168,6 +294,7 @@ xfs_dir2_leaf_to_node(
xfs_dir2_data_off_t off; /* freespace entry value */
__be16 *to; /* pointer to freespace entry */
xfs_trans_t *tp; /* transaction pointer */
+ struct xfs_dir3_icfree_hdr freehdr;
trace_xfs_dir2_leaf_to_node(args);
@@ -184,43 +311,43 @@ xfs_dir2_leaf_to_node(
/*
* Get the buffer for the new freespace block.
*/
- error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(mp, fdb), -1, &fbp,
- XFS_DATA_FORK);
+ error = xfs_dir3_free_get_buf(tp, dp, fdb, &fbp);
if (error)
return error;
- fbp->b_ops = &xfs_dir2_free_buf_ops;
free = fbp->b_addr;
+ xfs_dir3_free_hdr_from_disk(&freehdr, free);
leaf = lbp->b_addr;
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
- /*
- * Initialize the freespace block header.
- */
- free->hdr.magic = cpu_to_be32(XFS_DIR2_FREE_MAGIC);
- free->hdr.firstdb = 0;
- ASSERT(be32_to_cpu(ltp->bestcount) <= (uint)dp->i_d.di_size / mp->m_dirblksize);
- free->hdr.nvalid = ltp->bestcount;
+ ASSERT(be32_to_cpu(ltp->bestcount) <=
+ (uint)dp->i_d.di_size / mp->m_dirblksize);
+
/*
* Copy freespace entries from the leaf block to the new block.
* Count active entries.
*/
- for (i = n = 0, from = xfs_dir2_leaf_bests_p(ltp), to = free->bests;
- i < be32_to_cpu(ltp->bestcount); i++, from++, to++) {
+ from = xfs_dir2_leaf_bests_p(ltp);
+ to = xfs_dir3_free_bests_p(mp, free);
+ for (i = n = 0; i < be32_to_cpu(ltp->bestcount); i++, from++, to++) {
if ((off = be16_to_cpu(*from)) != NULLDATAOFF)
n++;
*to = cpu_to_be16(off);
}
- free->hdr.nused = cpu_to_be32(n);
-
- lbp->b_ops = &xfs_dir2_leafn_buf_ops;
- leaf->hdr.info.magic = cpu_to_be16(XFS_DIR2_LEAFN_MAGIC);
/*
- * Log everything.
+ * Now initialize the freespace block header.
*/
- xfs_dir2_leaf_log_header(tp, lbp);
+ freehdr.nused = n;
+ freehdr.nvalid = be32_to_cpu(ltp->bestcount);
+
+ xfs_dir3_free_hdr_to_disk(fbp->b_addr, &freehdr);
+ xfs_dir2_free_log_bests(tp, fbp, 0, freehdr.nvalid - 1);
xfs_dir2_free_log_header(tp, fbp);
- xfs_dir2_free_log_bests(tp, fbp, 0, be32_to_cpu(free->hdr.nvalid) - 1);
+
+ /* convert the leaf to a leafnode */
+ leaf->hdr.info.magic = cpu_to_be16(XFS_DIR2_LEAFN_MAGIC);
+ lbp->b_ops = &xfs_dir2_leafn_buf_ops;
+ xfs_dir2_leaf_log_header(tp, lbp);
xfs_dir2_leafn_check(dp, lbp);
return 0;
}
@@ -339,6 +466,23 @@ xfs_dir2_leafn_check(
}
ASSERT(be16_to_cpu(leaf->hdr.stale) == stale);
}
+
+static void
+xfs_dir2_free_hdr_check(
+ struct xfs_mount *mp,
+ struct xfs_buf *bp,
+ xfs_dir2_db_t db)
+{
+ struct xfs_dir3_icfree_hdr hdr;
+
+ xfs_dir3_free_hdr_from_disk(&hdr, bp->b_addr);
+
+ ASSERT((hdr.firstdb % xfs_dir3_free_max_bests(mp)) == 0);
+ ASSERT(hdr.firstdb <= db);
+ ASSERT(db < hdr.firstdb + hdr.nvalid);
+}
+#else
+#define xfs_dir2_free_hdr_check(mp, dp, db)
#endif /* DEBUG */
/*
@@ -409,7 +553,8 @@ xfs_dir2_leafn_lookup_for_addname(
curbp = state->extrablk.bp;
curfdb = state->extrablk.blkno;
free = curbp->b_addr;
- ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
+ ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC) ||
+ free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC));
}
length = xfs_dir2_data_entsize(args->namelen);
/*
@@ -436,6 +581,8 @@ xfs_dir2_leafn_lookup_for_addname(
* in hand, take a look at it.
*/
if (newdb != curdb) {
+ __be16 *bests;
+
curdb = newdb;
/*
* Convert the data block to the free block
@@ -458,13 +605,8 @@ xfs_dir2_leafn_lookup_for_addname(
if (error)
return error;
free = curbp->b_addr;
- ASSERT(be32_to_cpu(free->hdr.magic) ==
- XFS_DIR2_FREE_MAGIC);
- ASSERT((be32_to_cpu(free->hdr.firstdb) %
- xfs_dir2_free_max_bests(mp)) == 0);
- ASSERT(be32_to_cpu(free->hdr.firstdb) <= curdb);
- ASSERT(curdb < be32_to_cpu(free->hdr.firstdb) +
- be32_to_cpu(free->hdr.nvalid));
+
+ xfs_dir2_free_hdr_check(mp, curbp, curdb);
}
/*
* Get the index for our entry.
@@ -473,8 +615,8 @@ xfs_dir2_leafn_lookup_for_addname(
/*
* If it has room, return it.
*/
- if (unlikely(free->bests[fi] ==
- cpu_to_be16(NULLDATAOFF))) {
+ bests = xfs_dir3_free_bests_p(mp, free);
+ if (unlikely(bests[fi] == cpu_to_be16(NULLDATAOFF))) {
XFS_ERROR_REPORT("xfs_dir2_leafn_lookup_int",
XFS_ERRLEVEL_LOW, mp);
if (curfdb != newfdb)
@@ -482,7 +624,7 @@ xfs_dir2_leafn_lookup_for_addname(
return XFS_ERROR(EFSCORRUPTED);
}
curfdb = newfdb;
- if (be16_to_cpu(free->bests[fi]) >= length)
+ if (be16_to_cpu(bests[fi]) >= length)
goto out;
}
}
@@ -496,6 +638,12 @@ out:
state->extrablk.bp = curbp;
state->extrablk.index = fi;
state->extrablk.blkno = curfdb;
+
+ /*
+ * Important: this magic number is not in the buffer - it's for
+ * buffer type information and therefore only the free/data type
+ * matters here, not whether CRCs are enabled or not.
+ */
state->extrablk.magic = XFS_DIR2_FREE_MAGIC;
} else {
state->extravalid = 0;
@@ -883,7 +1031,7 @@ xfs_dir2_leafn_rebalance(
}
static int
-xfs_dir2_data_block_free(
+xfs_dir3_data_block_free(
xfs_da_args_t *args,
struct xfs_dir2_data_hdr *hdr,
struct xfs_dir2_free *free,
@@ -894,59 +1042,68 @@ xfs_dir2_data_block_free(
{
struct xfs_trans *tp = args->trans;
int logfree = 0;
+ __be16 *bests;
+ struct xfs_dir3_icfree_hdr freehdr;
- if (!hdr) {
- /* One less used entry in the free table. */
- be32_add_cpu(&free->hdr.nused, -1);
- xfs_dir2_free_log_header(tp, fbp);
- /*
- * If this was the last entry in the table, we can trim the
- * table size back. There might be other entries at the end
- * referring to non-existent data blocks, get those too.
- */
- if (findex == be32_to_cpu(free->hdr.nvalid) - 1) {
- int i; /* free entry index */
+ xfs_dir3_free_hdr_from_disk(&freehdr, free);
- for (i = findex - 1; i >= 0; i--) {
- if (free->bests[i] != cpu_to_be16(NULLDATAOFF))
- break;
- }
- free->hdr.nvalid = cpu_to_be32(i + 1);
- logfree = 0;
- } else {
- /* Not the last entry, just punch it out. */
- free->bests[findex] = cpu_to_be16(NULLDATAOFF);
- logfree = 1;
- }
+ bests = xfs_dir3_free_bests_p(tp->t_mountp, free);
+ if (hdr) {
/*
- * If there are no useful entries left in the block,
- * get rid of the block if we can.
+ * Data block is not empty, just set the free entry to the new
+ * value.
*/
- if (!free->hdr.nused) {
- int error;
+ bests[findex] = cpu_to_be16(longest);
+ xfs_dir2_free_log_bests(tp, fbp, findex, findex);
+ return 0;
+ }
- error = xfs_dir2_shrink_inode(args, fdb, fbp);
- if (error == 0) {
- fbp = NULL;
- logfree = 0;
- } else if (error != ENOSPC || args->total != 0)
- return error;
- /*
- * It's possible to get ENOSPC if there is no
- * space reservation. In this case some one
- * else will eventually get rid of this block.
- */
+ /*
+ * One less used entry in the free table. Unused is not converted
+ * because we only need to know if it zero
+ */
+ freehdr.nused--;
+
+ if (findex == freehdr.nvalid - 1) {
+ int i; /* free entry index */
+
+ for (i = findex - 1; i >= 0; i--) {
+ if (bests[i] != cpu_to_be16(NULLDATAOFF))
+ break;
}
+ freehdr.nvalid = i + 1;
+ logfree = 0;
} else {
+ /* Not the last entry, just punch it out. */
+ bests[findex] = cpu_to_be16(NULLDATAOFF);
+ logfree = 1;
+ }
+
+ xfs_dir3_free_hdr_to_disk(free, &freehdr);
+ xfs_dir2_free_log_header(tp, fbp);
+
+ /*
+ * If there are no useful entries left in the block, get rid of the
+ * block if we can.
+ */
+ if (!freehdr.nused) {
+ int error;
+
+ error = xfs_dir2_shrink_inode(args, fdb, fbp);
+ if (error == 0) {
+ fbp = NULL;
+ logfree = 0;
+ } else if (error != ENOSPC || args->total != 0)
+ return error;
/*
- * Data block is not empty, just set the free entry to the new
- * value.
+ * It's possible to get ENOSPC if there is no
+ * space reservation. In this case some one
+ * else will eventually get rid of this block.
*/
- free->bests[findex] = cpu_to_be16(longest);
- logfree = 1;
}
+
/* Log the free entry that changed, unless we got rid of it. */
if (logfree)
xfs_dir2_free_log_bests(tp, fbp, findex, findex);
@@ -1047,10 +1204,15 @@ xfs_dir2_leafn_remove(
if (error)
return error;
free = fbp->b_addr;
- ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
- ASSERT(be32_to_cpu(free->hdr.firstdb) ==
- xfs_dir2_free_max_bests(mp) *
- (fdb - XFS_DIR2_FREE_FIRSTDB(mp)));
+#ifdef DEBUG
+ {
+ struct xfs_dir3_icfree_hdr freehdr;
+ xfs_dir3_free_hdr_from_disk(&freehdr, free);
+ ASSERT(freehdr.firstdb ==
+ xfs_dir3_free_max_bests(mp) *
+ (fdb - XFS_DIR2_FREE_FIRSTDB(mp)));
+ }
+#endif
/*
* Calculate which entry we need to fix.
*/
@@ -1081,7 +1243,7 @@ xfs_dir2_leafn_remove(
* If we got rid of the data block, we can eliminate that entry
* in the free block.
*/
- error = xfs_dir2_data_block_free(args, hdr, free,
+ error = xfs_dir3_data_block_free(args, hdr, free,
fdb, findex, fbp, longest);
if (error)
return error;
@@ -1432,6 +1594,8 @@ xfs_dir2_node_addname_int(
int needscan; /* need to rescan data frees */
__be16 *tagp; /* data entry tag pointer */
xfs_trans_t *tp; /* transaction pointer */
+ __be16 *bests;
+ struct xfs_dir3_icfree_hdr freehdr;
dp = args->dp;
mp = dp->i_mount;
@@ -1449,36 +1613,37 @@ xfs_dir2_node_addname_int(
*/
ifbno = fblk->blkno;
free = fbp->b_addr;
- ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
findex = fblk->index;
+ bests = xfs_dir3_free_bests_p(mp, free);
+ xfs_dir3_free_hdr_from_disk(&freehdr, free);
+
/*
* This means the free entry showed that the data block had
* space for our entry, so we remembered it.
* Use that data block.
*/
if (findex >= 0) {
- ASSERT(findex < be32_to_cpu(free->hdr.nvalid));
- ASSERT(be16_to_cpu(free->bests[findex]) != NULLDATAOFF);
- ASSERT(be16_to_cpu(free->bests[findex]) >= length);
- dbno = be32_to_cpu(free->hdr.firstdb) + findex;
- }
- /*
- * The data block looked at didn't have enough room.
- * We'll start at the beginning of the freespace entries.
- */
- else {
+ ASSERT(findex < freehdr.nvalid);
+ ASSERT(be16_to_cpu(bests[findex]) != NULLDATAOFF);
+ ASSERT(be16_to_cpu(bests[findex]) >= length);
+ dbno = freehdr.firstdb + findex;
+ } else {
+ /*
+ * The data block looked at didn't have enough room.
+ * We'll start at the beginning of the freespace entries.
+ */
dbno = -1;
findex = 0;
}
- }
- /*
- * Didn't come in with a freespace block, so don't have a data block.
- */
- else {
+ } else {
+ /*
+ * Didn't come in with a freespace block, so no data block.
+ */
ifbno = dbno = -1;
fbp = NULL;
findex = 0;
}
+
/*
* If we don't have a data block yet, we're going to scan the
* freespace blocks looking for one. Figure out what the
@@ -1532,20 +1697,26 @@ xfs_dir2_node_addname_int(
if (!fbp)
continue;
free = fbp->b_addr;
- ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
findex = 0;
}
/*
* Look at the current free entry. Is it good enough?
+ *
+ * The bests initialisation should be wher eteh bufer is read in
+ * the above branch. But gcc is too stupid to realise that bests
+ * iand the freehdr are actually initialised if they are placed
+ * there, so we have to do it here to avoid warnings. Blech.
*/
- if (be16_to_cpu(free->bests[findex]) != NULLDATAOFF &&
- be16_to_cpu(free->bests[findex]) >= length)
- dbno = be32_to_cpu(free->hdr.firstdb) + findex;
+ bests = xfs_dir3_free_bests_p(mp, free);
+ xfs_dir3_free_hdr_from_disk(&freehdr, free);
+ if (be16_to_cpu(bests[findex]) != NULLDATAOFF &&
+ be16_to_cpu(bests[findex]) >= length)
+ dbno = freehdr.firstdb + findex;
else {
/*
* Are we done with the freeblock?
*/
- if (++findex == be32_to_cpu(free->hdr.nvalid)) {
+ if (++findex == freehdr.nvalid) {
/*
* Drop the block.
*/
@@ -1599,11 +1770,11 @@ xfs_dir2_node_addname_int(
* If there wasn't a freespace block, the read will
* return a NULL fbp. Allocate and initialize a new one.
*/
- if( fbp == NULL ) {
- if ((error = xfs_dir2_grow_inode(args, XFS_DIR2_FREE_SPACE,
- &fbno))) {
+ if(!fbp) {
+ error = xfs_dir2_grow_inode(args, XFS_DIR2_FREE_SPACE,
+ &fbno);
+ if (error)
return error;
- }
if (unlikely(xfs_dir2_db_to_fdb(mp, dbno) != fbno)) {
xfs_alert(mp,
@@ -1631,27 +1802,24 @@ xfs_dir2_node_addname_int(
/*
* Get a buffer for the new block.
*/
- error = xfs_da_get_buf(tp, dp,
- xfs_dir2_db_to_da(mp, fbno),
- -1, &fbp, XFS_DATA_FORK);
+ error = xfs_dir3_free_get_buf(tp, dp, fbno, &fbp);
if (error)
return error;
- fbp->b_ops = &xfs_dir2_free_buf_ops;
+ free = fbp->b_addr;
+ bests = xfs_dir3_free_bests_p(mp, free);
+ xfs_dir3_free_hdr_from_disk(&freehdr, free);
/*
- * Initialize the new block to be empty, and remember
- * its first slot as our empty slot.
+ * Remember the first slot as our empty slot.
*/
- free = fbp->b_addr;
- free->hdr.magic = cpu_to_be32(XFS_DIR2_FREE_MAGIC);
- free->hdr.firstdb = cpu_to_be32(
- (fbno - XFS_DIR2_FREE_FIRSTDB(mp)) *
- xfs_dir2_free_max_bests(mp));
+ freehdr.firstdb = (fbno - XFS_DIR2_FREE_FIRSTDB(mp)) *
+ xfs_dir3_free_max_bests(mp);
free->hdr.nvalid = 0;
free->hdr.nused = 0;
} else {
free = fbp->b_addr;
- ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
+ bests = xfs_dir3_free_bests_p(mp, free);
+ xfs_dir3_free_hdr_from_disk(&freehdr, free);
}
/*
@@ -1662,20 +1830,21 @@ xfs_dir2_node_addname_int(
* If it's after the end of the current entries in the
* freespace block, extend that table.
*/
- if (findex >= be32_to_cpu(free->hdr.nvalid)) {
- ASSERT(findex < xfs_dir2_free_max_bests(mp));
- free->hdr.nvalid = cpu_to_be32(findex + 1);
+ if (findex >= freehdr.nvalid) {
+ ASSERT(findex < xfs_dir3_free_max_bests(mp));
+ freehdr.nvalid = findex + 1;
/*
* Tag new entry so nused will go up.
*/
- free->bests[findex] = cpu_to_be16(NULLDATAOFF);
+ bests[findex] = cpu_to_be16(NULLDATAOFF);
}
/*
* If this entry was for an empty data block
* (this should always be true) then update the header.
*/
- if (free->bests[findex] == cpu_to_be16(NULLDATAOFF)) {
- be32_add_cpu(&free->hdr.nused, 1);
+ if (bests[findex] == cpu_to_be16(NULLDATAOFF)) {
+ freehdr.nused++;
+ xfs_dir3_free_hdr_to_disk(fbp->b_addr, &freehdr);
xfs_dir2_free_log_header(tp, fbp);
}
/*
@@ -1684,7 +1853,7 @@ xfs_dir2_node_addname_int(
* change again.
*/
hdr = dbp->b_addr;
- free->bests[findex] = hdr->bestfree[0].length;
+ bests[findex] = hdr->bestfree[0].length;
logfree = 1;
}
/*
@@ -1743,8 +1912,9 @@ xfs_dir2_node_addname_int(
/*
* If the freespace entry is now wrong, update it.
*/
- if (be16_to_cpu(free->bests[findex]) != be16_to_cpu(hdr->bestfree[0].length)) {
- free->bests[findex] = hdr->bestfree[0].length;
+ bests = xfs_dir3_free_bests_p(mp, free); /* gcc is so stupid */
+ if (be16_to_cpu(bests[findex]) != be16_to_cpu(hdr->bestfree[0].length)) {
+ bests[findex] = hdr->bestfree[0].length;
logfree = 1;
}
/*
@@ -1980,6 +2150,7 @@ xfs_dir2_node_trim_free(
xfs_dir2_free_t *free; /* freespace structure */
xfs_mount_t *mp; /* filesystem mount point */
xfs_trans_t *tp; /* transaction pointer */
+ struct xfs_dir3_icfree_hdr freehdr;
dp = args->dp;
mp = dp->i_mount;
@@ -1997,11 +2168,12 @@ xfs_dir2_node_trim_free(
if (!bp)
return 0;
free = bp->b_addr;
- ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC));
+ xfs_dir3_free_hdr_from_disk(&freehdr, free);
+
/*
* If there are used entries, there's nothing to do.
*/
- if (be32_to_cpu(free->hdr.nused) > 0) {
+ if (freehdr.nused > 0) {
xfs_trans_brelse(tp, bp);
*rvalp = 0;
return 0;
diff --git a/repair/phase6.c b/repair/phase6.c
index 136bb4f..80701a4 100644
--- a/repair/phase6.c
+++ b/repair/phase6.c
@@ -2461,7 +2461,7 @@ longform_dir2_check_node(
if (be32_to_cpu(free->hdr.magic) != XFS_DIR2_FREE_MAGIC ||
be32_to_cpu(free->hdr.firstdb) !=
(fdb - XFS_DIR2_FREE_FIRSTDB(mp)) *
- xfs_dir2_free_max_bests(mp) ||
+ xfs_dir3_free_max_bests(mp) ||
be32_to_cpu(free->hdr.nvalid) <
be32_to_cpu(free->hdr.nused)) {
do_warn(
xfsprogs-crc-patchset-v2/xfs-add-crc-to-dir2-data-2 0000600 0001750 0001750 00000060310 12132011546 020601 0 ustar dave dave xfs: add CRC checking to dir2 data blocks
From: Dave Chinner <dchinner@xxxxxxxxxx>
This addition follows the same pattern as the dir2 block CRCs.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
include/xfs_dir2_format.h | 21 +++++-----
libxfs/xfs_dir2_block.c | 20 ++++-----
libxfs/xfs_dir2_data.c | 98 +++++++++++++++++++++++++++++++--------------
libxfs/xfs_dir2_leaf.c | 59 ++++++++++++++++-----------
libxfs/xfs_dir2_node.c | 39 ++++++++++--------
libxfs/xfs_dir2_priv.h | 12 +++---
6 files changed, 152 insertions(+), 97 deletions(-)
diff --git a/include/xfs_dir2_format.h b/include/xfs_dir2_format.h
index bec058f..dfc8ccf 100644
--- a/include/xfs_dir2_format.h
+++ b/include/xfs_dir2_format.h
@@ -283,7 +283,8 @@ struct xfs_dir3_data_hdr {
static inline struct xfs_dir2_data_free *
xfs_dir3_data_bestfree_p(struct xfs_dir2_data_hdr *hdr)
{
- if (hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC)) {
+ if (hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC)) {
struct xfs_dir3_data_hdr *hdr3 = (struct xfs_dir3_data_hdr *)hdr;
return hdr3->best_free;
}
@@ -345,17 +346,6 @@ xfs_dir2_data_unused_tag_p(struct xfs_dir2_data_unused *dup)
be16_to_cpu(dup->length) - sizeof(__be16));
}
-static inline struct xfs_dir2_data_unused *
-xfs_dir3_data_unused_p(struct xfs_dir2_data_hdr *hdr)
-{
- if (hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC)) {
- return (struct xfs_dir2_data_unused *)
- ((char *)hdr + sizeof(struct xfs_dir3_data_hdr));
- }
- return (struct xfs_dir2_data_unused *)
- ((char *)hdr + sizeof(struct xfs_dir2_data_hdr));
-}
-
static inline size_t
xfs_dir3_data_hdr_size(bool dir3)
{
@@ -379,6 +369,13 @@ xfs_dir3_data_entry_p(struct xfs_dir2_data_hdr *hdr)
((char *)hdr + xfs_dir3_data_entry_offset(hdr));
}
+static inline struct xfs_dir2_data_unused *
+xfs_dir3_data_unused_p(struct xfs_dir2_data_hdr *hdr)
+{
+ return (struct xfs_dir2_data_unused *)
+ ((char *)hdr + xfs_dir3_data_entry_offset(hdr));
+}
+
/*
* Offsets of . and .. in data space (always block 0)
*/
diff --git a/libxfs/xfs_dir2_block.c b/libxfs/xfs_dir2_block.c
index 9978a02..e1e1161 100644
--- a/libxfs/xfs_dir2_block.c
+++ b/libxfs/xfs_dir2_block.c
@@ -59,7 +59,7 @@ xfs_dir3_block_verify(
if (hdr3->magic != be32_to_cpu(XFS_DIR2_BLOCK_MAGIC))
return false;
}
- if (__xfs_dir2_data_check(NULL, bp))
+ if (__xfs_dir3_data_check(NULL, bp))
return false;
return true;
}
@@ -535,7 +535,7 @@ xfs_dir2_block_addname(
xfs_dir2_data_log_header(tp, bp);
xfs_dir2_block_log_tail(tp, bp);
xfs_dir2_data_log_entry(tp, bp, dep);
- xfs_dir2_data_check(dp, bp);
+ xfs_dir3_data_check(dp, bp);
return 0;
}
@@ -604,7 +604,7 @@ xfs_dir2_block_lookup(
dp = args->dp;
mp = dp->i_mount;
hdr = bp->b_addr;
- xfs_dir2_data_check(dp, bp);
+ xfs_dir3_data_check(dp, bp);
btp = xfs_dir2_block_tail_p(mp, hdr);
blp = xfs_dir2_block_leaf_p(btp);
/*
@@ -655,7 +655,7 @@ xfs_dir2_block_lookup_int(
return error;
hdr = bp->b_addr;
- xfs_dir2_data_check(dp, bp);
+ xfs_dir3_data_check(dp, bp);
btp = xfs_dir2_block_tail_p(mp, hdr);
blp = xfs_dir2_block_leaf_p(btp);
/*
@@ -792,7 +792,7 @@ xfs_dir2_block_removename(
xfs_dir2_data_freescan(mp, hdr, &needlog);
if (needlog)
xfs_dir2_data_log_header(tp, bp);
- xfs_dir2_data_check(dp, bp);
+ xfs_dir3_data_check(dp, bp);
/*
* See if the size as a shortform is good enough.
*/
@@ -849,7 +849,7 @@ xfs_dir2_block_replace(
*/
dep->inumber = cpu_to_be64(args->inumber);
xfs_dir2_data_log_entry(args->trans, bp, dep);
- xfs_dir2_data_check(dp, bp);
+ xfs_dir3_data_check(dp, bp);
return 0;
}
@@ -930,12 +930,14 @@ xfs_dir2_leaf_to_block(
* Read the data block if we don't already have it, give up if it fails.
*/
if (!dbp) {
- error = xfs_dir2_data_read(tp, dp, mp->m_dirdatablk, -1, &dbp);
+ error = xfs_dir3_data_read(tp, dp, mp->m_dirdatablk, -1, &dbp);
if (error)
return error;
}
hdr = dbp->b_addr;
- ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC));
+ ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC));
+
/*
* Size of the "leaf" area in the block.
*/
@@ -1213,6 +1215,6 @@ xfs_dir2_sf_to_block(
ASSERT(needscan == 0);
xfs_dir2_block_log_leaf(tp, bp, 0, be32_to_cpu(btp->count) - 1);
xfs_dir2_block_log_tail(tp, bp);
- xfs_dir2_data_check(dp, bp);
+ xfs_dir3_data_check(dp, bp);
return 0;
}
diff --git a/libxfs/xfs_dir2_data.c b/libxfs/xfs_dir2_data.c
index 66aab07..69841df 100644
--- a/libxfs/xfs_dir2_data.c
+++ b/libxfs/xfs_dir2_data.c
@@ -25,7 +25,7 @@
* Return 0 is the buffer is good, otherwise an error.
*/
int
-__xfs_dir2_data_check(
+__xfs_dir3_data_check(
struct xfs_inode *dp, /* incore inode pointer */
struct xfs_buf *bp) /* data block's buffer */
{
@@ -61,6 +61,7 @@ __xfs_dir2_data_check(
endp = (char *)lep;
break;
case XFS_DIR2_DATA_MAGIC:
+ case XFS_DIR3_DATA_MAGIC:
endp = (char *)hdr + mp->m_dirblksize;
break;
default:
@@ -173,21 +174,27 @@ __xfs_dir2_data_check(
return 0;
}
-static void
-xfs_dir2_data_verify(
+static bool
+xfs_dir3_data_verify(
struct xfs_buf *bp)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
- struct xfs_dir2_data_hdr *hdr = bp->b_addr;
- int block_ok = 0;
+ struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr;
- block_ok = hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC);
- block_ok = block_ok && __xfs_dir2_data_check(NULL, bp) == 0;
-
- if (!block_ok) {
- XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, hdr);
- xfs_buf_ioerror(bp, EFSCORRUPTED);
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ if (hdr3->magic != cpu_to_be32(XFS_DIR3_DATA_MAGIC))
+ return false;
+ if (!uuid_equal(&hdr3->uuid, &mp->m_sb.sb_uuid))
+ return false;
+ if (be64_to_cpu(hdr3->blkno) != bp->b_bn)
+ return false;
+ } else {
+ if (hdr3->magic != cpu_to_be32(XFS_DIR2_DATA_MAGIC))
+ return false;
}
+ if (__xfs_dir3_data_check(NULL, bp))
+ return false;
+ return true;
}
/*
@@ -196,7 +203,7 @@ xfs_dir2_data_verify(
* format buffer or a data format buffer on readahead.
*/
static void
-xfs_dir2_data_reada_verify(
+xfs_dir3_data_reada_verify(
struct xfs_buf *bp)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
@@ -209,7 +216,8 @@ xfs_dir2_data_reada_verify(
bp->b_ops->verify_read(bp);
return;
case XFS_DIR2_DATA_MAGIC:
- xfs_dir2_data_verify(bp);
+ case XFS_DIR3_DATA_MAGIC:
+ xfs_dir3_data_verify(bp);
return;
default:
XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, hdr);
@@ -219,32 +227,56 @@ xfs_dir2_data_reada_verify(
}
static void
-xfs_dir2_data_read_verify(
+xfs_dir3_data_read_verify(
struct xfs_buf *bp)
{
- xfs_dir2_data_verify(bp);
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+
+ if ((xfs_sb_version_hascrc(&mp->m_sb) &&
+ !xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
+ XFS_DIR3_DATA_CRC_OFF)) ||
+ !xfs_dir3_data_verify(bp)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
}
static void
-xfs_dir2_data_write_verify(
+xfs_dir3_data_write_verify(
struct xfs_buf *bp)
{
- xfs_dir2_data_verify(bp);
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_buf_log_item *bip = bp->b_fspriv;
+ struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr;
+
+ if (!xfs_dir3_data_verify(bp)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ return;
+ }
+
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return;
+
+ if (bip)
+ hdr3->lsn = cpu_to_be64(bip->bli_item.li_lsn);
+
+ xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), XFS_DIR3_DATA_CRC_OFF);
}
-const struct xfs_buf_ops xfs_dir2_data_buf_ops = {
- .verify_read = xfs_dir2_data_read_verify,
- .verify_write = xfs_dir2_data_write_verify,
+const struct xfs_buf_ops xfs_dir3_data_buf_ops = {
+ .verify_read = xfs_dir3_data_read_verify,
+ .verify_write = xfs_dir3_data_write_verify,
};
-static const struct xfs_buf_ops xfs_dir2_data_reada_buf_ops = {
- .verify_read = xfs_dir2_data_reada_verify,
- .verify_write = xfs_dir2_data_write_verify,
+static const struct xfs_buf_ops xfs_dir3_data_reada_buf_ops = {
+ .verify_read = xfs_dir3_data_reada_verify,
+ .verify_write = xfs_dir3_data_write_verify,
};
int
-xfs_dir2_data_read(
+xfs_dir3_data_read(
struct xfs_trans *tp,
struct xfs_inode *dp,
xfs_dablk_t bno,
@@ -252,18 +284,18 @@ xfs_dir2_data_read(
struct xfs_buf **bpp)
{
return xfs_da_read_buf(tp, dp, bno, mapped_bno, bpp,
- XFS_DATA_FORK, &xfs_dir2_data_buf_ops);
+ XFS_DATA_FORK, &xfs_dir3_data_buf_ops);
}
int
-xfs_dir2_data_readahead(
+xfs_dir3_data_readahead(
struct xfs_trans *tp,
struct xfs_inode *dp,
xfs_dablk_t bno,
xfs_daddr_t mapped_bno)
{
return xfs_da_reada_buf(tp, dp, bno, mapped_bno,
- XFS_DATA_FORK, &xfs_dir2_data_reada_buf_ops);
+ XFS_DATA_FORK, &xfs_dir3_data_reada_buf_ops);
}
/*
@@ -293,6 +325,7 @@ xfs_dir2_data_freefind(
* one we're looking for it has to be exact.
*/
ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
for (dfp = &bf[0], seenzero = matched = 0;
@@ -442,6 +475,7 @@ xfs_dir2_data_freescan(
char *p; /* current entry pointer */
ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
@@ -518,13 +552,12 @@ xfs_dir3_data_init(
XFS_DATA_FORK);
if (error)
return error;
- bp->b_ops = &xfs_dir2_data_buf_ops;
+ bp->b_ops = &xfs_dir3_data_buf_ops;
/*
* Initialize the header.
*/
hdr = bp->b_addr;
-
if (xfs_sb_version_hascrc(&mp->m_sb)) {
struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr;
@@ -575,6 +608,7 @@ xfs_dir2_data_log_entry(
xfs_dir2_data_hdr_t *hdr = bp->b_addr;
ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
@@ -594,6 +628,7 @@ xfs_dir2_data_log_header(
xfs_dir2_data_hdr_t *hdr = bp->b_addr;
ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
@@ -612,6 +647,7 @@ xfs_dir2_data_log_unused(
xfs_dir2_data_hdr_t *hdr = bp->b_addr;
ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
@@ -659,7 +695,8 @@ xfs_dir2_data_make_free(
/*
* Figure out where the end of the data area is.
*/
- if (hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC))
+ if (hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC))
endptr = (char *)hdr + mp->m_dirblksize;
else {
xfs_dir2_block_tail_t *btp; /* block tail */
@@ -841,6 +878,7 @@ xfs_dir2_data_use_free(
hdr = bp->b_addr;
ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
ASSERT(be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG);
diff --git a/libxfs/xfs_dir2_leaf.c b/libxfs/xfs_dir2_leaf.c
index a1df347..0f848b4 100644
--- a/libxfs/xfs_dir2_leaf.c
+++ b/libxfs/xfs_dir2_leaf.c
@@ -133,7 +133,7 @@ xfs_dir2_block_to_leaf(
int needlog; /* need to log block header */
int needscan; /* need to rescan bestfree */
xfs_trans_t *tp; /* transaction pointer */
- struct xfs_dir2_data_free *bf;
+ struct xfs_dir2_data_free *bf;
trace_xfs_dir2_block_to_leaf(args);
@@ -159,7 +159,7 @@ xfs_dir2_block_to_leaf(
ASSERT(lbp != NULL);
leaf = lbp->b_addr;
hdr = dbp->b_addr;
- xfs_dir2_data_check(dp, dbp);
+ xfs_dir3_data_check(dp, dbp);
btp = xfs_dir2_block_tail_p(mp, hdr);
blp = xfs_dir2_block_leaf_p(btp);
bf = xfs_dir3_data_bestfree_p(hdr);
@@ -188,8 +188,12 @@ xfs_dir2_block_to_leaf(
/*
* Fix up the block header, make it a data block.
*/
- dbp->b_ops = &xfs_dir2_data_buf_ops;
- hdr->magic = cpu_to_be32(XFS_DIR2_DATA_MAGIC);
+ dbp->b_ops = &xfs_dir3_data_buf_ops;
+ if (hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC))
+ hdr->magic = cpu_to_be32(XFS_DIR2_DATA_MAGIC);
+ else
+ hdr->magic = cpu_to_be32(XFS_DIR3_DATA_MAGIC);
+
if (needscan)
xfs_dir2_data_freescan(mp, hdr, &needlog);
/*
@@ -205,7 +209,7 @@ xfs_dir2_block_to_leaf(
if (needlog)
xfs_dir2_data_log_header(tp, dbp);
xfs_dir2_leaf_check(dp, lbp);
- xfs_dir2_data_check(dp, dbp);
+ xfs_dir3_data_check(dp, dbp);
xfs_dir2_leaf_log_bests(tp, lbp, 0, 0);
return 0;
}
@@ -369,6 +373,7 @@ xfs_dir2_leaf_addname(
__be16 *tagp; /* end of data entry */
xfs_trans_t *tp; /* transaction pointer */
xfs_dir2_db_t use_block; /* data block number */
+ struct xfs_dir2_data_free *bf; /* bestfree table */
trace_xfs_dir2_leaf_addname(args);
@@ -552,14 +557,15 @@ xfs_dir2_leaf_addname(
else
xfs_dir2_leaf_log_bests(tp, lbp, use_block, use_block);
hdr = dbp->b_addr;
- bestsp[use_block] = hdr->bestfree[0].length;
+ bf = xfs_dir3_data_bestfree_p(hdr);
+ bestsp[use_block] = bf[0].length;
grown = 1;
} else {
/*
* Already had space in some data block.
* Just read that one in.
*/
- error = xfs_dir2_data_read(tp, dp,
+ error = xfs_dir3_data_read(tp, dp,
xfs_dir2_db_to_da(mp, use_block),
-1, &dbp);
if (error) {
@@ -567,13 +573,14 @@ xfs_dir2_leaf_addname(
return error;
}
hdr = dbp->b_addr;
+ bf = xfs_dir3_data_bestfree_p(hdr);
grown = 0;
}
/*
* Point to the biggest freespace in our data block.
*/
dup = (xfs_dir2_data_unused_t *)
- ((char *)hdr + be16_to_cpu(hdr->bestfree[0].offset));
+ ((char *)hdr + be16_to_cpu(bf[0].offset));
ASSERT(be16_to_cpu(dup->length) >= length);
needscan = needlog = 0;
/*
@@ -606,8 +613,8 @@ xfs_dir2_leaf_addname(
* If the bests table needs to be changed, do it.
* Log the change unless we've already done that.
*/
- if (be16_to_cpu(bestsp[use_block]) != be16_to_cpu(hdr->bestfree[0].length)) {
- bestsp[use_block] = hdr->bestfree[0].length;
+ if (be16_to_cpu(bestsp[use_block]) != be16_to_cpu(bf[0].length)) {
+ bestsp[use_block] = bf[0].length;
if (!grown)
xfs_dir2_leaf_log_bests(tp, lbp, use_block, use_block);
}
@@ -627,7 +634,7 @@ xfs_dir2_leaf_addname(
xfs_dir2_leaf_log_header(tp, lbp);
xfs_dir2_leaf_log_ents(tp, lbp, lfloglow, lfloghigh);
xfs_dir2_leaf_check(dp, lbp);
- xfs_dir2_data_check(dp, dbp);
+ xfs_dir3_data_check(dp, dbp);
return 0;
}
@@ -1077,7 +1084,7 @@ xfs_dir2_leaf_lookup_int(
if (newdb != curdb) {
if (dbp)
xfs_trans_brelse(tp, dbp);
- error = xfs_dir2_data_read(tp, dp,
+ error = xfs_dir3_data_read(tp, dp,
xfs_dir2_db_to_da(mp, newdb),
-1, &dbp);
if (error) {
@@ -1118,7 +1125,7 @@ xfs_dir2_leaf_lookup_int(
ASSERT(cidb != -1);
if (cidb != curdb) {
xfs_trans_brelse(tp, dbp);
- error = xfs_dir2_data_read(tp, dp,
+ error = xfs_dir3_data_read(tp, dp,
xfs_dir2_db_to_da(mp, cidb),
-1, &dbp);
if (error) {
@@ -1164,6 +1171,7 @@ xfs_dir2_leaf_removename(
int needscan; /* need to rescan data frees */
xfs_dir2_data_off_t oldbest; /* old value of best free */
xfs_trans_t *tp; /* transaction pointer */
+ struct xfs_dir2_data_free *bf; /* bestfree table */
trace_xfs_dir2_leaf_removename(args);
@@ -1178,7 +1186,8 @@ xfs_dir2_leaf_removename(
mp = dp->i_mount;
leaf = lbp->b_addr;
hdr = dbp->b_addr;
- xfs_dir2_data_check(dp, dbp);
+ bf = xfs_dir3_data_bestfree_p(hdr);
+ xfs_dir3_data_check(dp, dbp);
/*
* Point to the leaf entry, use that to point to the data entry.
*/
@@ -1187,7 +1196,7 @@ xfs_dir2_leaf_removename(
dep = (xfs_dir2_data_entry_t *)
((char *)hdr + xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address)));
needscan = needlog = 0;
- oldbest = be16_to_cpu(hdr->bestfree[0].length);
+ oldbest = be16_to_cpu(bf[0].length);
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
bestsp = xfs_dir2_leaf_bests_p(ltp);
ASSERT(be16_to_cpu(bestsp[db]) == oldbest);
@@ -1216,16 +1225,16 @@ xfs_dir2_leaf_removename(
* If the longest freespace in the data block has changed,
* put the new value in the bests table and log that.
*/
- if (be16_to_cpu(hdr->bestfree[0].length) != oldbest) {
- bestsp[db] = hdr->bestfree[0].length;
+ if (be16_to_cpu(bf[0].length) != oldbest) {
+ bestsp[db] = bf[0].length;
xfs_dir2_leaf_log_bests(tp, lbp, db, db);
}
- xfs_dir2_data_check(dp, dbp);
+ xfs_dir3_data_check(dp, dbp);
/*
* If the data block is now empty then get rid of the data block.
*/
- if (be16_to_cpu(hdr->bestfree[0].length) ==
- mp->m_dirblksize - (uint)sizeof(*hdr)) {
+ if (be16_to_cpu(bf[0].length) ==
+ mp->m_dirblksize - xfs_dir3_data_entry_offset(hdr)) {
ASSERT(db != mp->m_dirdatablk);
if ((error = xfs_dir2_shrink_inode(args, db, dbp))) {
/*
@@ -1405,7 +1414,7 @@ xfs_dir2_leaf_trim_data(
/*
* Read the offending data block. We need its buffer.
*/
- error = xfs_dir2_data_read(tp, dp, xfs_dir2_db_to_da(mp, db), -1, &dbp);
+ error = xfs_dir3_data_read(tp, dp, xfs_dir2_db_to_da(mp, db), -1, &dbp);
if (error)
return error;
@@ -1415,10 +1424,12 @@ xfs_dir2_leaf_trim_data(
#ifdef DEBUG
{
struct xfs_dir2_data_hdr *hdr = dbp->b_addr;
+ struct xfs_dir2_data_free *bf = xfs_dir3_data_bestfree_p(hdr);
- ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC));
- ASSERT(be16_to_cpu(hdr->bestfree[0].length) ==
- mp->m_dirblksize - (uint)sizeof(*hdr));
+ ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC));
+ ASSERT(be16_to_cpu(bf[0].length) ==
+ mp->m_dirblksize - xfs_dir3_data_entry_offset(hdr));
ASSERT(db == be32_to_cpu(ltp->bestcount) - 1);
}
#endif
diff --git a/libxfs/xfs_dir2_node.c b/libxfs/xfs_dir2_node.c
index 0edaddc..9dda757 100644
--- a/libxfs/xfs_dir2_node.c
+++ b/libxfs/xfs_dir2_node.c
@@ -737,13 +737,13 @@ xfs_dir2_leafn_lookup_for_entry(
ASSERT(state->extravalid);
curbp = state->extrablk.bp;
} else {
- error = xfs_dir2_data_read(tp, dp,
+ error = xfs_dir3_data_read(tp, dp,
xfs_dir2_db_to_da(mp, newdb),
-1, &curbp);
if (error)
return error;
}
- xfs_dir2_data_check(dp, curbp);
+ xfs_dir3_data_check(dp, curbp);
curdb = newdb;
}
/*
@@ -771,7 +771,7 @@ xfs_dir2_leafn_lookup_for_entry(
state->extrablk.index = (int)((char *)dep -
(char *)curbp->b_addr);
state->extrablk.magic = XFS_DIR2_DATA_MAGIC;
- curbp->b_ops = &xfs_dir2_data_buf_ops;
+ curbp->b_ops = &xfs_dir3_data_buf_ops;
if (cmp == XFS_CMP_EXACT)
return XFS_ERROR(EEXIST);
}
@@ -786,7 +786,7 @@ xfs_dir2_leafn_lookup_for_entry(
state->extrablk.index = -1;
state->extrablk.blkno = curdb;
state->extrablk.magic = XFS_DIR2_DATA_MAGIC;
- curbp->b_ops = &xfs_dir2_data_buf_ops;
+ curbp->b_ops = &xfs_dir3_data_buf_ops;
} else {
/* If the curbp is not the CI match block, drop it */
if (state->extrablk.bp != curbp)
@@ -1136,6 +1136,7 @@ xfs_dir2_leafn_remove(
int needlog; /* need to log data header */
int needscan; /* need to rescan data frees */
xfs_trans_t *tp; /* transaction pointer */
+ struct xfs_dir2_data_free *bf; /* bestfree table */
trace_xfs_dir2_leafn_remove(args, index);
@@ -1170,7 +1171,8 @@ xfs_dir2_leafn_remove(
dbp = dblk->bp;
hdr = dbp->b_addr;
dep = (xfs_dir2_data_entry_t *)((char *)hdr + off);
- longest = be16_to_cpu(hdr->bestfree[0].length);
+ bf = xfs_dir3_data_bestfree_p(hdr);
+ longest = be16_to_cpu(bf[0].length);
needlog = needscan = 0;
xfs_dir2_data_make_free(tp, dbp, off,
xfs_dir2_data_entsize(dep->namelen), &needlog, &needscan);
@@ -1182,12 +1184,12 @@ xfs_dir2_leafn_remove(
xfs_dir2_data_freescan(mp, hdr, &needlog);
if (needlog)
xfs_dir2_data_log_header(tp, dbp);
- xfs_dir2_data_check(dp, dbp);
+ xfs_dir3_data_check(dp, dbp);
/*
* If the longest data block freespace changes, need to update
* the corresponding freeblock entry.
*/
- if (longest < be16_to_cpu(hdr->bestfree[0].length)) {
+ if (longest < be16_to_cpu(bf[0].length)) {
int error; /* error return value */
struct xfs_buf *fbp; /* freeblock buffer */
xfs_dir2_db_t fdb; /* freeblock block number */
@@ -1217,12 +1219,13 @@ xfs_dir2_leafn_remove(
* Calculate which entry we need to fix.
*/
findex = xfs_dir2_db_to_fdindex(mp, db);
- longest = be16_to_cpu(hdr->bestfree[0].length);
+ longest = be16_to_cpu(bf[0].length);
/*
* If the data block is now empty we can get rid of it
* (usually).
*/
- if (longest == mp->m_dirblksize - (uint)sizeof(*hdr)) {
+ if (longest == mp->m_dirblksize -
+ xfs_dir3_data_entry_offset(hdr)) {
/*
* Try to punch out the data block.
*/
@@ -1596,6 +1599,7 @@ xfs_dir2_node_addname_int(
xfs_trans_t *tp; /* transaction pointer */
__be16 *bests;
struct xfs_dir3_icfree_hdr freehdr;
+ struct xfs_dir2_data_free *bf;
dp = args->dp;
mp = dp->i_mount;
@@ -1853,7 +1857,8 @@ xfs_dir2_node_addname_int(
* change again.
*/
hdr = dbp->b_addr;
- bests[findex] = hdr->bestfree[0].length;
+ bf = xfs_dir3_data_bestfree_p(hdr);
+ bests[findex] = bf[0].length;
logfree = 1;
}
/*
@@ -1869,19 +1874,20 @@ xfs_dir2_node_addname_int(
/*
* Read the data block in.
*/
- error = xfs_dir2_data_read(tp, dp, xfs_dir2_db_to_da(mp, dbno),
+ error = xfs_dir3_data_read(tp, dp, xfs_dir2_db_to_da(mp, dbno),
-1, &dbp);
if (error)
return error;
hdr = dbp->b_addr;
+ bf = xfs_dir3_data_bestfree_p(hdr);
logfree = 0;
}
- ASSERT(be16_to_cpu(hdr->bestfree[0].length) >= length);
+ ASSERT(be16_to_cpu(bf[0].length) >= length);
/*
* Point to the existing unused space.
*/
dup = (xfs_dir2_data_unused_t *)
- ((char *)hdr + be16_to_cpu(hdr->bestfree[0].offset));
+ ((char *)hdr + be16_to_cpu(bf[0].offset));
needscan = needlog = 0;
/*
* Mark the first part of the unused space, inuse for us.
@@ -1913,8 +1919,8 @@ xfs_dir2_node_addname_int(
* If the freespace entry is now wrong, update it.
*/
bests = xfs_dir3_free_bests_p(mp, free); /* gcc is so stupid */
- if (be16_to_cpu(bests[findex]) != be16_to_cpu(hdr->bestfree[0].length)) {
- bests[findex] = hdr->bestfree[0].length;
+ if (be16_to_cpu(bests[findex]) != be16_to_cpu(bf[0].length)) {
+ bests[findex] = bf[0].length;
logfree = 1;
}
/*
@@ -2104,7 +2110,8 @@ xfs_dir2_node_replace(
* Point to the data entry.
*/
hdr = state->extrablk.bp->b_addr;
- ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC));
+ ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
+ hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC));
dep = (xfs_dir2_data_entry_t *)
((char *)hdr +
xfs_dir2_dataptr_to_off(state->mp, be32_to_cpu(lep->address)));
diff --git a/libxfs/xfs_dir2_priv.h b/libxfs/xfs_dir2_priv.h
index e6f2e0a..910e644 100644
--- a/libxfs/xfs_dir2_priv.h
+++ b/libxfs/xfs_dir2_priv.h
@@ -43,17 +43,17 @@ extern int xfs_dir2_leaf_to_block(struct xfs_da_args *args,
/* xfs_dir2_data.c */
#ifdef DEBUG
-#define xfs_dir2_data_check(dp,bp) __xfs_dir2_data_check(dp, bp);
+#define xfs_dir3_data_check(dp,bp) __xfs_dir3_data_check(dp, bp);
#else
-#define xfs_dir2_data_check(dp,bp)
+#define xfs_dir3_data_check(dp,bp)
#endif
-extern const struct xfs_buf_ops xfs_dir2_data_buf_ops;
+extern const struct xfs_buf_ops xfs_dir3_data_buf_ops;
-extern int __xfs_dir2_data_check(struct xfs_inode *dp, struct xfs_buf *bp);
-extern int xfs_dir2_data_read(struct xfs_trans *tp, struct xfs_inode *dp,
+extern int __xfs_dir3_data_check(struct xfs_inode *dp, struct xfs_buf *bp);
+extern int xfs_dir3_data_read(struct xfs_trans *tp, struct xfs_inode *dp,
xfs_dablk_t bno, xfs_daddr_t mapped_bno, struct xfs_buf **bpp);
-extern int xfs_dir2_data_readahead(struct xfs_trans *tp, struct xfs_inode *dp,
+extern int xfs_dir3_data_readahead(struct xfs_trans *tp, struct xfs_inode *dp,
xfs_dablk_t bno, xfs_daddr_t mapped_bno);
extern struct xfs_dir2_data_free *
xfsprogs-crc-patchset-v2/xfs-add-crc-to-dir2-leaf-3 0000600 0001750 0001750 00000262660 12132011546 020614 0 ustar dave dave xfs: add CRC checking to dir2 leaf blocks
From: Dave Chinner <dchinner@xxxxxxxxxx>
This addition follows the same pattern as the dir2 block CRCs.
Seeing as both LEAF1 and LEAFN types need to changed at the same
time, this is a pretty large amount of change. leaf block headers
need to be abstracted away from the on-disk structures (struct
xfs_dir3_icleaf_hdr), as do the base leaf entry locations.
This header abstract allows the in-core header and leaf entry
location to be passed around instead of the leaf block itself. This
saves a lot of converting individual variables from on-disk format
to host format where they are used, so there's a good chance that
the compiler will be able to produce much more optimal code as it's
not having to byteswap variables all over the place.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
db/check.c | 2 +-
db/dir2.c | 2 +-
include/xfs_da_btree.h | 23 ++
include/xfs_dir2_format.h | 60 +++-
libxfs/xfs_da_btree.c | 45 ++-
libxfs/xfs_dir2_block.c | 19 +-
libxfs/xfs_dir2_leaf.c | 786 ++++++++++++++++++++++++++++-----------------
libxfs/xfs_dir2_node.c | 475 +++++++++++++++------------
libxfs/xfs_dir2_priv.h | 32 +-
repair/dir2.c | 12 +-
repair/phase6.c | 14 +-
11 files changed, 938 insertions(+), 532 deletions(-)
diff --git a/db/check.c b/db/check.c
index 368d2b7..73d9da2 100644
--- a/db/check.c
+++ b/db/check.c
@@ -3304,7 +3304,7 @@ process_leaf_node_dir_v2_int(
error++;
return;
}
- lep = leaf->ents;
+ lep = xfs_dir3_leaf_ents_p(leaf);
for (i = stale = 0; i < be16_to_cpu(leaf->hdr.count); i++) {
if (be32_to_cpu(lep[i].address) == XFS_DIR2_NULL_DATAPTR)
stale++;
diff --git a/db/dir2.c b/db/dir2.c
index be58639..abdbfee 100644
--- a/db/dir2.c
+++ b/db/dir2.c
@@ -81,7 +81,7 @@ const field_t dir2_flds[] = {
FLD_COUNT, TYP_NONE },
{ "lbests", FLDT_DIR2_DATA_OFF, dir2_leaf_bests_offset,
dir2_leaf_bests_count, FLD_ARRAY|FLD_OFFSET|FLD_COUNT, TYP_NONE },
- { "lents", FLDT_DIR2_LEAF_ENTRY, OI(LOFF(ents)), dir2_leaf_ents_count,
+ { "lents", FLDT_DIR2_LEAF_ENTRY, OI(LOFF(__ents)), dir2_leaf_ents_count,
FLD_ARRAY|FLD_COUNT, TYP_NONE },
{ "ltail", FLDT_DIR2_LEAF_TAIL, dir2_leaf_tail_offset,
dir2_leaf_tail_count, FLD_OFFSET|FLD_COUNT, TYP_NONE },
diff --git a/include/xfs_da_btree.h b/include/xfs_da_btree.h
index ee5170c..0854b95 100644
--- a/include/xfs_da_btree.h
+++ b/include/xfs_da_btree.h
@@ -47,6 +47,29 @@ typedef struct xfs_da_blkinfo {
} xfs_da_blkinfo_t;
/*
+ * CRC enabled directory structure types
+ *
+ * The headers change size for the additional verification information, but
+ * otherwise the tree layouts and contents are unchanged.
+ */
+#define XFS_DIR3_LEAF1_MAGIC 0x3df1 /* magic number: v2 dirlf single blks */
+#define XFS_DIR3_LEAFN_MAGIC 0x3dff /* magic number: v2 dirlf multi blks */
+
+struct xfs_da3_blkinfo {
+ /*
+ * the node link manipulation code relies on the fact that the first
+ * element of this structure is the struct xfs_da_blkinfo so it can
+ * ignore the differences in the rest of the structures.
+ */
+ struct xfs_da_blkinfo hdr;
+ __be32 crc; /* CRC of block */
+ __be64 blkno; /* first block of the buffer */
+ __be64 lsn; /* sequence number of last write */
+ uuid_t uuid; /* filesystem we belong to */
+ __be64 owner; /* inode that owns the block */
+};
+
+/*
* This is the structure of the root and intermediate nodes in the Btree.
* The leaf nodes are defined above.
*
diff --git a/include/xfs_dir2_format.h b/include/xfs_dir2_format.h
index dfc8ccf..8bb4efe 100644
--- a/include/xfs_dir2_format.h
+++ b/include/xfs_dir2_format.h
@@ -470,6 +470,21 @@ typedef struct xfs_dir2_leaf_hdr {
__be16 stale; /* count of stale entries */
} xfs_dir2_leaf_hdr_t;
+struct xfs_dir3_leaf_hdr {
+ struct xfs_da3_blkinfo info; /* header for da routines */
+ __be16 count; /* count of entries */
+ __be16 stale; /* count of stale entries */
+ __be32 pad;
+};
+
+struct xfs_dir3_icleaf_hdr {
+ __uint32_t forw;
+ __uint32_t back;
+ __uint16_t magic;
+ __uint16_t count;
+ __uint16_t stale;
+};
+
/*
* Leaf block entry.
*/
@@ -489,23 +504,50 @@ typedef struct xfs_dir2_leaf_tail {
* Leaf block.
*/
typedef struct xfs_dir2_leaf {
- xfs_dir2_leaf_hdr_t hdr; /* leaf header */
- xfs_dir2_leaf_entry_t ents[]; /* entries */
+ xfs_dir2_leaf_hdr_t hdr; /* leaf header */
+ xfs_dir2_leaf_entry_t __ents[]; /* entries */
} xfs_dir2_leaf_t;
-/*
- * DB blocks here are logical directory block numbers, not filesystem blocks.
- */
+struct xfs_dir3_leaf {
+ struct xfs_dir3_leaf_hdr hdr; /* leaf header */
+ struct xfs_dir2_leaf_entry __ents[]; /* entries */
+};
+
+#define XFS_DIR3_LEAF_CRC_OFF offsetof(struct xfs_dir3_leaf_hdr, info.crc)
+
+static inline int
+xfs_dir3_leaf_hdr_size(struct xfs_dir2_leaf *lp)
+{
+ if (lp->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAF1_MAGIC) ||
+ lp->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC))
+ return sizeof(struct xfs_dir3_leaf_hdr);
+ return sizeof(struct xfs_dir2_leaf_hdr);
+}
-static inline int xfs_dir2_max_leaf_ents(struct xfs_mount *mp)
+static inline int
+xfs_dir3_max_leaf_ents(struct xfs_mount *mp, struct xfs_dir2_leaf *lp)
{
- return (mp->m_dirblksize - (uint)sizeof(struct xfs_dir2_leaf_hdr)) /
+ return (mp->m_dirblksize - xfs_dir3_leaf_hdr_size(lp)) /
(uint)sizeof(struct xfs_dir2_leaf_entry);
}
/*
* Get address of the bestcount field in the single-leaf block.
*/
+static inline struct xfs_dir2_leaf_entry *
+xfs_dir3_leaf_ents_p(struct xfs_dir2_leaf *lp)
+{
+ if (lp->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAF1_MAGIC) ||
+ lp->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC)) {
+ struct xfs_dir3_leaf *lp3 = (struct xfs_dir3_leaf *)lp;
+ return lp3->__ents;
+ }
+ return lp->__ents;
+}
+
+/*
+ * Get address of the bestcount field in the single-leaf block.
+ */
static inline struct xfs_dir2_leaf_tail *
xfs_dir2_leaf_tail_p(struct xfs_mount *mp, struct xfs_dir2_leaf *lp)
{
@@ -524,6 +566,10 @@ xfs_dir2_leaf_bests_p(struct xfs_dir2_leaf_tail *ltp)
}
/*
+ * DB blocks here are logical directory block numbers, not filesystem blocks.
+ */
+
+/*
* Convert dataptr to byte in file space
*/
static inline xfs_dir2_off_t
diff --git a/libxfs/xfs_da_btree.c b/libxfs/xfs_da_btree.c
index a31d353..63cd299 100644
--- a/libxfs/xfs_da_btree.c
+++ b/libxfs/xfs_da_btree.c
@@ -118,7 +118,8 @@ xfs_da_node_read_verify(
bp->b_ops->verify_read(bp);
return;
case XFS_DIR2_LEAFN_MAGIC:
- bp->b_ops = &xfs_dir2_leafn_buf_ops;
+ case XFS_DIR3_LEAFN_MAGIC:
+ bp->b_ops = &xfs_dir3_leafn_buf_ops;
bp->b_ops->verify_read(bp);
return;
default:
@@ -375,11 +376,18 @@ xfs_da_root_split(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
size = (int)((char *)&oldroot->btree[be16_to_cpu(oldroot->hdr.count)] -
(char *)oldroot);
} else {
- ASSERT(oldroot->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
+ struct xfs_dir3_icleaf_hdr leafhdr;
+ struct xfs_dir2_leaf_entry *ents;
+
leaf = (xfs_dir2_leaf_t *)oldroot;
- size = (int)((char *)&leaf->ents[be16_to_cpu(leaf->hdr.count)] -
- (char *)leaf);
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+ ents = xfs_dir3_leaf_ents_p(leaf);
+
+ ASSERT(leafhdr.magic == XFS_DIR2_LEAFN_MAGIC ||
+ leafhdr.magic == XFS_DIR3_LEAFN_MAGIC);
+ size = (int)((char *)&ents[leafhdr.count] - (char *)leaf);
}
+ /* XXX: can't just copy CRC headers from one block to another */
memcpy(node, oldroot, size);
xfs_trans_log_buf(tp, bp, 0, size - 1);
@@ -403,7 +411,8 @@ xfs_da_root_split(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
node->hdr.count = cpu_to_be16(2);
#ifdef DEBUG
- if (oldroot->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC)) {
+ if (oldroot->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
+ oldroot->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC)) {
ASSERT(blk1->blkno >= mp->m_dirleafblk &&
blk1->blkno < mp->m_dirfreeblk);
ASSERT(blk2->blkno >= mp->m_dirleafblk &&
@@ -761,6 +770,7 @@ xfs_da_blkinfo_onlychild_validate(struct xfs_da_blkinfo *blkinfo, __u16 level)
if (level == 1) {
ASSERT(magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
+ magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC) ||
magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
} else
ASSERT(magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
@@ -1544,6 +1554,7 @@ xfs_da_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path,
info = blk->bp->b_addr;
ASSERT(info->magic == cpu_to_be16(XFS_DA_NODE_MAGIC) ||
info->magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
+ info->magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC) ||
info->magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
blk->magic = be16_to_cpu(info->magic);
if (blk->magic == XFS_DA_NODE_MAGIC) {
@@ -1563,12 +1574,13 @@ xfs_da_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path,
NULL);
break;
case XFS_DIR2_LEAFN_MAGIC:
+ case XFS_DIR3_LEAFN_MAGIC:
+ blk->magic = XFS_DIR2_LEAFN_MAGIC;
blk->hashval = xfs_dir2_leafn_lasthash(blk->bp,
NULL);
break;
default:
- ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC ||
- blk->magic == XFS_DIR2_LEAFN_MAGIC);
+ ASSERT(0);
break;
}
}
@@ -1812,10 +1824,16 @@ xfs_da_swap_lastblock(
/*
* Get values from the moved block.
*/
- if (dead_info->magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC)) {
+ if (dead_info->magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
+ dead_info->magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC)) {
+ struct xfs_dir3_icleaf_hdr leafhdr;
+ struct xfs_dir2_leaf_entry *ents;
+
dead_leaf2 = (xfs_dir2_leaf_t *)dead_info;
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, dead_leaf2);
+ ents = xfs_dir3_leaf_ents_p(dead_leaf2);
dead_level = 0;
- dead_hash = be32_to_cpu(dead_leaf2->ents[be16_to_cpu(dead_leaf2->hdr.count) - 1].hashval);
+ dead_hash = be32_to_cpu(ents[leafhdr.count - 1].hashval);
} else {
ASSERT(dead_info->magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
dead_node = (xfs_da_intnode_t *)dead_info;
@@ -2260,10 +2278,17 @@ xfs_da_read_buf(
XFS_TEST_ERROR((magic != XFS_DA_NODE_MAGIC) &&
(magic != XFS_ATTR_LEAF_MAGIC) &&
(magic != XFS_DIR2_LEAF1_MAGIC) &&
+ (magic != XFS_DIR3_LEAF1_MAGIC) &&
(magic != XFS_DIR2_LEAFN_MAGIC) &&
+ (magic != XFS_DIR3_LEAFN_MAGIC) &&
(magic1 != XFS_DIR2_BLOCK_MAGIC) &&
+ (magic1 != XFS_DIR3_BLOCK_MAGIC) &&
(magic1 != XFS_DIR2_DATA_MAGIC) &&
- (free->hdr.magic != cpu_to_be32(XFS_DIR2_FREE_MAGIC)),
+ (magic1 != XFS_DIR3_DATA_MAGIC) &&
+ (free->hdr.magic !=
+ cpu_to_be32(XFS_DIR2_FREE_MAGIC)) &&
+ (free->hdr.magic !=
+ cpu_to_be32(XFS_DIR3_FREE_MAGIC)),
mp, XFS_ERRTAG_DA_READ_BUF,
XFS_RANDOM_DA_READ_BUF))) {
trace_xfs_da_btree_corrupt(bp, _RET_IP_);
diff --git a/libxfs/xfs_dir2_block.c b/libxfs/xfs_dir2_block.c
index 18eabd1..b98b749 100644
--- a/libxfs/xfs_dir2_block.c
+++ b/libxfs/xfs_dir2_block.c
@@ -897,6 +897,8 @@ xfs_dir2_leaf_to_block(
__be16 *tagp; /* end of entry (tag) */
int to; /* block/leaf to index */
xfs_trans_t *tp; /* transaction pointer */
+ struct xfs_dir2_leaf_entry *ents;
+ struct xfs_dir3_icleaf_hdr leafhdr;
trace_xfs_dir2_leaf_to_block(args);
@@ -904,8 +906,12 @@ xfs_dir2_leaf_to_block(
tp = args->trans;
mp = dp->i_mount;
leaf = lbp->b_addr;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC));
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+ ents = xfs_dir3_leaf_ents_p(leaf);
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
+
+ ASSERT(leafhdr.magic == XFS_DIR2_LEAF1_MAGIC ||
+ leafhdr.magic == XFS_DIR3_LEAF1_MAGIC);
/*
* If there are data blocks other than the first one, take this
* opportunity to remove trailing empty data blocks that may have
@@ -942,7 +948,7 @@ xfs_dir2_leaf_to_block(
* Size of the "leaf" area in the block.
*/
size = (uint)sizeof(xfs_dir2_block_tail_t) +
- (uint)sizeof(*lep) * (be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale));
+ (uint)sizeof(*lep) * (leafhdr.count - leafhdr.stale);
/*
* Look at the last data entry.
*/
@@ -971,18 +977,17 @@ xfs_dir2_leaf_to_block(
* Initialize the block tail.
*/
btp = xfs_dir2_block_tail_p(mp, hdr);
- btp->count = cpu_to_be32(be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale));
+ btp->count = cpu_to_be32(leafhdr.count - leafhdr.stale);
btp->stale = 0;
xfs_dir2_block_log_tail(tp, dbp);
/*
* Initialize the block leaf area. We compact out stale entries.
*/
lep = xfs_dir2_block_leaf_p(btp);
- for (from = to = 0; from < be16_to_cpu(leaf->hdr.count); from++) {
- if (leaf->ents[from].address ==
- cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
+ for (from = to = 0; from < leafhdr.count; from++) {
+ if (ents[from].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
continue;
- lep[to++] = leaf->ents[from];
+ lep[to++] = ents[from];
}
ASSERT(to == be32_to_cpu(btp->count));
xfs_dir2_block_log_leaf(tp, dbp, 0, be32_to_cpu(btp->count) - 1);
diff --git a/libxfs/xfs_dir2_leaf.c b/libxfs/xfs_dir2_leaf.c
index 0f848b4..f00b23c 100644
--- a/libxfs/xfs_dir2_leaf.c
+++ b/libxfs/xfs_dir2_leaf.c
@@ -1,5 +1,6 @@
/*
* Copyright (c) 2000-2003,2005 Silicon Graphics, Inc.
+ * Copyright (c) 2013 Red Hat, Inc.
* All Rights Reserved.
*
* This program is free software; you can redistribute it and/or
@@ -21,73 +22,257 @@
/*
* Local function declarations.
*/
-#ifdef DEBUG
-static void xfs_dir2_leaf_check(struct xfs_inode *dp, struct xfs_buf *bp);
-#else
-#define xfs_dir2_leaf_check(dp, bp)
-#endif
static int xfs_dir2_leaf_lookup_int(xfs_da_args_t *args, struct xfs_buf **lbpp,
int *indexp, struct xfs_buf **dbpp);
-static void xfs_dir2_leaf_log_bests(struct xfs_trans *tp, struct xfs_buf *bp,
+static void xfs_dir3_leaf_log_bests(struct xfs_trans *tp, struct xfs_buf *bp,
int first, int last);
-static void xfs_dir2_leaf_log_tail(struct xfs_trans *tp, struct xfs_buf *bp);
+static void xfs_dir3_leaf_log_tail(struct xfs_trans *tp, struct xfs_buf *bp);
-static void
-xfs_dir2_leaf_verify(
+/*
+ * Check the internal consistency of a leaf1 block.
+ * Pop an assert if something is wrong.
+ */
+#ifdef DEBUG
+#define xfs_dir3_leaf_check(mp, bp) \
+do { \
+ if (!xfs_dir3_leaf1_check((mp), (bp))) \
+ ASSERT(0); \
+} while (0);
+
+STATIC bool
+xfs_dir3_leaf1_check(
+ struct xfs_mount *mp,
+ struct xfs_buf *bp)
+{
+ struct xfs_dir2_leaf *leaf = bp->b_addr;
+ struct xfs_dir3_icleaf_hdr leafhdr;
+
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+
+ if (leafhdr.magic == XFS_DIR3_LEAF1_MAGIC) {
+ struct xfs_dir3_leaf_hdr *leaf3 = bp->b_addr;
+ if (be64_to_cpu(leaf3->info.blkno) != bp->b_bn)
+ return false;
+ } else if (leafhdr.magic != XFS_DIR2_LEAF1_MAGIC)
+ return false;
+
+ return xfs_dir3_leaf_check_int(mp, &leafhdr, leaf);
+}
+#else
+#define xfs_dir3_leaf_check(mp, bp)
+#endif
+
+void
+xfs_dir3_leaf_hdr_from_disk(
+ struct xfs_dir3_icleaf_hdr *to,
+ struct xfs_dir2_leaf *from)
+{
+ if (from->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC) ||
+ from->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC)) {
+ to->forw = be32_to_cpu(from->hdr.info.forw);
+ to->back = be32_to_cpu(from->hdr.info.back);
+ to->magic = be16_to_cpu(from->hdr.info.magic);
+ to->count = be16_to_cpu(from->hdr.count);
+ to->stale = be16_to_cpu(from->hdr.stale);
+ } else {
+ struct xfs_dir3_leaf_hdr *hdr3 = (struct xfs_dir3_leaf_hdr *)from;
+
+ to->forw = be32_to_cpu(hdr3->info.hdr.forw);
+ to->back = be32_to_cpu(hdr3->info.hdr.back);
+ to->magic = be16_to_cpu(hdr3->info.hdr.magic);
+ to->count = be16_to_cpu(hdr3->count);
+ to->stale = be16_to_cpu(hdr3->stale);
+ }
+
+ ASSERT(to->magic == XFS_DIR2_LEAF1_MAGIC ||
+ to->magic == XFS_DIR3_LEAF1_MAGIC ||
+ to->magic == XFS_DIR2_LEAFN_MAGIC ||
+ to->magic == XFS_DIR3_LEAFN_MAGIC);
+}
+
+void
+xfs_dir3_leaf_hdr_to_disk(
+ struct xfs_dir2_leaf *to,
+ struct xfs_dir3_icleaf_hdr *from)
+{
+ ASSERT(from->magic == XFS_DIR2_LEAF1_MAGIC ||
+ from->magic == XFS_DIR3_LEAF1_MAGIC ||
+ from->magic == XFS_DIR2_LEAFN_MAGIC ||
+ from->magic == XFS_DIR3_LEAFN_MAGIC);
+
+ if (from->magic == XFS_DIR2_LEAF1_MAGIC ||
+ from->magic == XFS_DIR2_LEAFN_MAGIC) {
+ to->hdr.info.forw = cpu_to_be32(from->forw);
+ to->hdr.info.back = cpu_to_be32(from->back);
+ to->hdr.info.magic = cpu_to_be16(from->magic);
+ to->hdr.count = cpu_to_be16(from->count);
+ to->hdr.stale = cpu_to_be16(from->stale);
+ } else {
+ struct xfs_dir3_leaf_hdr *hdr3 = (struct xfs_dir3_leaf_hdr *)to;
+
+ hdr3->info.hdr.forw = cpu_to_be32(from->forw);
+ hdr3->info.hdr.back = cpu_to_be32(from->back);
+ hdr3->info.hdr.magic = cpu_to_be16(from->magic);
+ hdr3->count = cpu_to_be16(from->count);
+ hdr3->stale = cpu_to_be16(from->stale);
+ }
+}
+
+bool
+xfs_dir3_leaf_check_int(
+ struct xfs_mount *mp,
+ struct xfs_dir3_icleaf_hdr *hdr,
+ struct xfs_dir2_leaf *leaf)
+{
+ struct xfs_dir2_leaf_entry *ents;
+ xfs_dir2_leaf_tail_t *ltp;
+ int stale;
+ int i;
+
+ ents = xfs_dir3_leaf_ents_p(leaf);
+ ltp = xfs_dir2_leaf_tail_p(mp, leaf);
+
+ /*
+ * XXX (dgc): This value is not restrictive enough.
+ * Should factor in the size of the bests table as well.
+ * We can deduce a value for that from di_size.
+ */
+ if (hdr->count > xfs_dir3_max_leaf_ents(mp, leaf))
+ return false;
+
+ /* Leaves and bests don't overlap in leaf format. */
+ if ((hdr->magic == XFS_DIR2_LEAF1_MAGIC ||
+ hdr->magic == XFS_DIR3_LEAF1_MAGIC) &&
+ (char *)&ents[hdr->count] > (char *)xfs_dir2_leaf_bests_p(ltp))
+ return false;
+
+ /* Check hash value order, count stale entries. */
+ for (i = stale = 0; i < hdr->count; i++) {
+ if (i + 1 < hdr->count) {
+ if (be32_to_cpu(ents[i].hashval) >
+ be32_to_cpu(ents[i + 1].hashval))
+ return false;
+ }
+ if (ents[i].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
+ stale++;
+ }
+ if (hdr->stale != stale)
+ return false;
+ return true;
+}
+
+static bool
+xfs_dir3_leaf_verify(
struct xfs_buf *bp,
- __be16 magic)
+ __uint16_t magic)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
- struct xfs_dir2_leaf_hdr *hdr = bp->b_addr;
- int block_ok = 0;
+ struct xfs_dir2_leaf *leaf = bp->b_addr;
+ struct xfs_dir3_icleaf_hdr leafhdr;
+
+ ASSERT(magic == XFS_DIR2_LEAF1_MAGIC || magic == XFS_DIR2_LEAFN_MAGIC);
+
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ struct xfs_dir3_leaf_hdr *leaf3 = bp->b_addr;
- block_ok = hdr->info.magic == magic;
- if (!block_ok) {
- XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, hdr);
+ if ((magic == XFS_DIR2_LEAF1_MAGIC &&
+ leafhdr.magic != XFS_DIR3_LEAF1_MAGIC) ||
+ (magic == XFS_DIR2_LEAFN_MAGIC &&
+ leafhdr.magic != XFS_DIR3_LEAFN_MAGIC))
+ return false;
+
+ if (!uuid_equal(&leaf3->info.uuid, &mp->m_sb.sb_uuid))
+ return false;
+ if (be64_to_cpu(leaf3->info.blkno) != bp->b_bn)
+ return false;
+ } else {
+ if (leafhdr.magic != magic)
+ return false;
+ }
+ return xfs_dir3_leaf_check_int(mp, &leafhdr, leaf);
+}
+
+static void
+__read_verify(
+ struct xfs_buf *bp,
+ __uint16_t magic)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+
+ if ((xfs_sb_version_hascrc(&mp->m_sb) &&
+ !xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
+ XFS_DIR3_LEAF_CRC_OFF)) ||
+ !xfs_dir3_leaf_verify(bp, magic)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
xfs_buf_ioerror(bp, EFSCORRUPTED);
}
}
static void
-xfs_dir2_leaf1_read_verify(
+__write_verify(
+ struct xfs_buf *bp,
+ __uint16_t magic)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_buf_log_item *bip = bp->b_fspriv;
+ struct xfs_dir3_leaf_hdr *hdr3 = bp->b_addr;
+
+ if (!xfs_dir3_leaf_verify(bp, magic)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ return;
+ }
+
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return;
+
+ if (bip)
+ hdr3->info.lsn = cpu_to_be64(bip->bli_item.li_lsn);
+
+ xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), XFS_DIR3_LEAF_CRC_OFF);
+}
+
+static void
+xfs_dir3_leaf1_read_verify(
struct xfs_buf *bp)
{
- xfs_dir2_leaf_verify(bp, cpu_to_be16(XFS_DIR2_LEAF1_MAGIC));
+ __read_verify(bp, XFS_DIR2_LEAF1_MAGIC);
}
static void
-xfs_dir2_leaf1_write_verify(
+xfs_dir3_leaf1_write_verify(
struct xfs_buf *bp)
{
- xfs_dir2_leaf_verify(bp, cpu_to_be16(XFS_DIR2_LEAF1_MAGIC));
+ __write_verify(bp, XFS_DIR2_LEAF1_MAGIC);
}
void
-xfs_dir2_leafn_read_verify(
+xfs_dir3_leafn_read_verify(
struct xfs_buf *bp)
{
- xfs_dir2_leaf_verify(bp, cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
+ __read_verify(bp, XFS_DIR2_LEAFN_MAGIC);
}
void
-xfs_dir2_leafn_write_verify(
+xfs_dir3_leafn_write_verify(
struct xfs_buf *bp)
{
- xfs_dir2_leaf_verify(bp, cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
+ __write_verify(bp, XFS_DIR2_LEAFN_MAGIC);
}
-static const struct xfs_buf_ops xfs_dir2_leaf1_buf_ops = {
- .verify_read = xfs_dir2_leaf1_read_verify,
- .verify_write = xfs_dir2_leaf1_write_verify,
+const struct xfs_buf_ops xfs_dir3_leaf1_buf_ops = {
+ .verify_read = xfs_dir3_leaf1_read_verify,
+ .verify_write = xfs_dir3_leaf1_write_verify,
};
-const struct xfs_buf_ops xfs_dir2_leafn_buf_ops = {
- .verify_read = xfs_dir2_leafn_read_verify,
- .verify_write = xfs_dir2_leafn_write_verify,
+const struct xfs_buf_ops xfs_dir3_leafn_buf_ops = {
+ .verify_read = xfs_dir3_leafn_read_verify,
+ .verify_write = xfs_dir3_leafn_write_verify,
};
static int
-xfs_dir2_leaf_read(
+xfs_dir3_leaf_read(
struct xfs_trans *tp,
struct xfs_inode *dp,
xfs_dablk_t fbno,
@@ -95,11 +280,11 @@ xfs_dir2_leaf_read(
struct xfs_buf **bpp)
{
return xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
- XFS_DATA_FORK, &xfs_dir2_leaf1_buf_ops);
+ XFS_DATA_FORK, &xfs_dir3_leaf1_buf_ops);
}
int
-xfs_dir2_leafn_read(
+xfs_dir3_leafn_read(
struct xfs_trans *tp,
struct xfs_inode *dp,
xfs_dablk_t fbno,
@@ -107,7 +292,81 @@ xfs_dir2_leafn_read(
struct xfs_buf **bpp)
{
return xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
- XFS_DATA_FORK, &xfs_dir2_leafn_buf_ops);
+ XFS_DATA_FORK, &xfs_dir3_leafn_buf_ops);
+}
+
+/*
+ * Initialize a new leaf block, leaf1 or leafn magic accepted.
+ */
+static void
+xfs_dir3_leaf_init(
+ struct xfs_mount *mp,
+ struct xfs_buf *bp,
+ xfs_ino_t owner,
+ __uint16_t type)
+{
+ struct xfs_dir2_leaf *leaf = bp->b_addr;
+
+ ASSERT(type == XFS_DIR2_LEAF1_MAGIC || type == XFS_DIR2_LEAFN_MAGIC);
+
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ struct xfs_dir3_leaf_hdr *leaf3 = bp->b_addr;
+
+ memset(leaf3, 0, sizeof(*leaf3));
+
+ leaf3->info.hdr.magic = (type == XFS_DIR2_LEAF1_MAGIC)
+ ? cpu_to_be16(XFS_DIR3_LEAF1_MAGIC)
+ : cpu_to_be16(XFS_DIR3_LEAFN_MAGIC);
+ leaf3->info.blkno = cpu_to_be64(bp->b_bn);
+ leaf3->info.owner = cpu_to_be64(owner);
+ uuid_copy(&leaf3->info.uuid, &mp->m_sb.sb_uuid);
+ } else {
+ memset(leaf, 0, sizeof(*leaf));
+ leaf->hdr.info.magic = cpu_to_be16(type);
+ }
+
+ /*
+ * If it's a leaf-format directory initialize the tail.
+ * Caller is responsible for initialising the bests table.
+ */
+ if (type == XFS_DIR2_LEAF1_MAGIC) {
+ struct xfs_dir2_leaf_tail *ltp;
+
+ ltp = xfs_dir2_leaf_tail_p(mp, leaf);
+ ltp->bestcount = 0;
+ bp->b_ops = &xfs_dir3_leaf1_buf_ops;
+ } else
+ bp->b_ops = &xfs_dir3_leafn_buf_ops;
+}
+
+int
+xfs_dir3_leaf_get_buf(
+ xfs_da_args_t *args,
+ xfs_dir2_db_t bno,
+ struct xfs_buf **bpp,
+ __uint16_t magic)
+{
+ struct xfs_inode *dp = args->dp;
+ struct xfs_trans *tp = args->trans;
+ struct xfs_mount *mp = dp->i_mount;
+ struct xfs_buf *bp;
+ int error;
+
+ ASSERT(magic == XFS_DIR2_LEAF1_MAGIC || magic == XFS_DIR2_LEAFN_MAGIC);
+ ASSERT(bno >= XFS_DIR2_LEAF_FIRSTDB(mp) &&
+ bno < XFS_DIR2_FREE_FIRSTDB(mp));
+
+ error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(mp, bno), -1, &bp,
+ XFS_DATA_FORK);
+ if (error)
+ return error;
+
+ xfs_dir3_leaf_init(mp, bp, dp->i_ino, magic);
+ xfs_dir3_leaf_log_header(tp, bp);
+ if (magic == XFS_DIR2_LEAF1_MAGIC)
+ xfs_dir3_leaf_log_tail(tp, bp);
+ *bpp = bp;
+ return 0;
}
/*
@@ -134,6 +393,8 @@ xfs_dir2_block_to_leaf(
int needscan; /* need to rescan bestfree */
xfs_trans_t *tp; /* transaction pointer */
struct xfs_dir2_data_free *bf;
+ struct xfs_dir2_leaf_entry *ents;
+ struct xfs_dir3_icleaf_hdr leafhdr;
trace_xfs_dir2_block_to_leaf(args);
@@ -153,27 +414,33 @@ xfs_dir2_block_to_leaf(
/*
* Initialize the leaf block, get a buffer for it.
*/
- if ((error = xfs_dir2_leaf_init(args, ldb, &lbp, XFS_DIR2_LEAF1_MAGIC))) {
+ error = xfs_dir3_leaf_get_buf(args, ldb, &lbp, XFS_DIR2_LEAF1_MAGIC);
+ if (error)
return error;
- }
- ASSERT(lbp != NULL);
+
leaf = lbp->b_addr;
hdr = dbp->b_addr;
xfs_dir3_data_check(dp, dbp);
btp = xfs_dir2_block_tail_p(mp, hdr);
blp = xfs_dir2_block_leaf_p(btp);
bf = xfs_dir3_data_bestfree_p(hdr);
+ ents = xfs_dir3_leaf_ents_p(leaf);
+
/*
* Set the counts in the leaf header.
*/
- leaf->hdr.count = cpu_to_be16(be32_to_cpu(btp->count));
- leaf->hdr.stale = cpu_to_be16(be32_to_cpu(btp->stale));
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+ leafhdr.count = be32_to_cpu(btp->count);
+ leafhdr.stale = be32_to_cpu(btp->stale);
+ xfs_dir3_leaf_hdr_to_disk(leaf, &leafhdr);
+ xfs_dir3_leaf_log_header(tp, lbp);
+
/*
* Could compact these but I think we always do the conversion
* after squeezing out stale entries.
*/
- memcpy(leaf->ents, blp, be32_to_cpu(btp->count) * sizeof(xfs_dir2_leaf_entry_t));
- xfs_dir2_leaf_log_ents(tp, lbp, 0, be16_to_cpu(leaf->hdr.count) - 1);
+ memcpy(ents, blp, be32_to_cpu(btp->count) * sizeof(xfs_dir2_leaf_entry_t));
+ xfs_dir3_leaf_log_ents(tp, lbp, 0, leafhdr.count - 1);
needscan = 0;
needlog = 1;
/*
@@ -208,15 +475,16 @@ xfs_dir2_block_to_leaf(
*/
if (needlog)
xfs_dir2_data_log_header(tp, dbp);
- xfs_dir2_leaf_check(dp, lbp);
+ xfs_dir3_leaf_check(mp, lbp);
xfs_dir3_data_check(dp, dbp);
- xfs_dir2_leaf_log_bests(tp, lbp, 0, 0);
+ xfs_dir3_leaf_log_bests(tp, lbp, 0, 0);
return 0;
}
STATIC void
-xfs_dir2_leaf_find_stale(
- struct xfs_dir2_leaf *leaf,
+xfs_dir3_leaf_find_stale(
+ struct xfs_dir3_icleaf_hdr *leafhdr,
+ struct xfs_dir2_leaf_entry *ents,
int index,
int *lowstale,
int *highstale)
@@ -225,7 +493,7 @@ xfs_dir2_leaf_find_stale(
* Find the first stale entry before our index, if any.
*/
for (*lowstale = index - 1; *lowstale >= 0; --*lowstale) {
- if (leaf->ents[*lowstale].address ==
+ if (ents[*lowstale].address ==
cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
break;
}
@@ -235,10 +503,8 @@ xfs_dir2_leaf_find_stale(
* Stop if the result would require moving more entries than using
* lowstale.
*/
- for (*highstale = index;
- *highstale < be16_to_cpu(leaf->hdr.count);
- ++*highstale) {
- if (leaf->ents[*highstale].address ==
+ for (*highstale = index; *highstale < leafhdr->count; ++*highstale) {
+ if (ents[*highstale].address ==
cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
break;
if (*lowstale >= 0 && index - *lowstale <= *highstale - index)
@@ -247,8 +513,9 @@ xfs_dir2_leaf_find_stale(
}
struct xfs_dir2_leaf_entry *
-xfs_dir2_leaf_find_entry(
- xfs_dir2_leaf_t *leaf, /* leaf structure */
+xfs_dir3_leaf_find_entry(
+ struct xfs_dir3_icleaf_hdr *leafhdr,
+ struct xfs_dir2_leaf_entry *ents,
int index, /* leaf table position */
int compact, /* need to compact leaves */
int lowstale, /* index of prev stale leaf */
@@ -256,7 +523,7 @@ xfs_dir2_leaf_find_entry(
int *lfloglow, /* low leaf logging index */
int *lfloghigh) /* high leaf logging index */
{
- if (!leaf->hdr.stale) {
+ if (!leafhdr->stale) {
xfs_dir2_leaf_entry_t *lep; /* leaf entry table pointer */
/*
@@ -264,18 +531,16 @@ xfs_dir2_leaf_find_entry(
*
* If there are no stale entries, just insert a hole at index.
*/
- lep = &leaf->ents[index];
- if (index < be16_to_cpu(leaf->hdr.count))
+ lep = &ents[index];
+ if (index < leafhdr->count)
memmove(lep + 1, lep,
- (be16_to_cpu(leaf->hdr.count) - index) *
- sizeof(*lep));
+ (leafhdr->count - index) * sizeof(*lep));
/*
* Record low and high logging indices for the leaf.
*/
*lfloglow = index;
- *lfloghigh = be16_to_cpu(leaf->hdr.count);
- be16_add_cpu(&leaf->hdr.count, 1);
+ *lfloghigh = leafhdr->count++;
return lep;
}
@@ -289,16 +554,17 @@ xfs_dir2_leaf_find_entry(
* entries before and after our insertion point.
*/
if (compact == 0)
- xfs_dir2_leaf_find_stale(leaf, index, &lowstale, &highstale);
+ xfs_dir3_leaf_find_stale(leafhdr, ents, index,
+ &lowstale, &highstale);
/*
* If the low one is better, use it.
*/
if (lowstale >= 0 &&
- (highstale == be16_to_cpu(leaf->hdr.count) ||
+ (highstale == leafhdr->count ||
index - lowstale - 1 < highstale - index)) {
ASSERT(index - lowstale - 1 >= 0);
- ASSERT(leaf->ents[lowstale].address ==
+ ASSERT(ents[lowstale].address ==
cpu_to_be32(XFS_DIR2_NULL_DATAPTR));
/*
@@ -306,37 +572,34 @@ xfs_dir2_leaf_find_entry(
* for the new entry.
*/
if (index - lowstale - 1 > 0) {
- memmove(&leaf->ents[lowstale],
- &leaf->ents[lowstale + 1],
+ memmove(&ents[lowstale], &ents[lowstale + 1],
(index - lowstale - 1) *
- sizeof(xfs_dir2_leaf_entry_t));
+ sizeof(xfs_dir2_leaf_entry_t));
}
*lfloglow = MIN(lowstale, *lfloglow);
*lfloghigh = MAX(index - 1, *lfloghigh);
- be16_add_cpu(&leaf->hdr.stale, -1);
- return &leaf->ents[index - 1];
+ leafhdr->stale--;
+ return &ents[index - 1];
}
/*
* The high one is better, so use that one.
*/
ASSERT(highstale - index >= 0);
- ASSERT(leaf->ents[highstale].address ==
- cpu_to_be32(XFS_DIR2_NULL_DATAPTR));
+ ASSERT(ents[highstale].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR));
/*
* Copy entries down to cover the stale entry and make room for the
* new entry.
*/
if (highstale - index > 0) {
- memmove(&leaf->ents[index + 1],
- &leaf->ents[index],
+ memmove(&ents[index + 1], &ents[index],
(highstale - index) * sizeof(xfs_dir2_leaf_entry_t));
}
*lfloglow = MIN(index, *lfloglow);
*lfloghigh = MAX(highstale, *lfloghigh);
- be16_add_cpu(&leaf->hdr.stale, -1);
- return &leaf->ents[index];
+ leafhdr->stale--;
+ return &ents[index];
}
/*
@@ -374,6 +637,8 @@ xfs_dir2_leaf_addname(
xfs_trans_t *tp; /* transaction pointer */
xfs_dir2_db_t use_block; /* data block number */
struct xfs_dir2_data_free *bf; /* bestfree table */
+ struct xfs_dir2_leaf_entry *ents;
+ struct xfs_dir3_icleaf_hdr leafhdr;
trace_xfs_dir2_leaf_addname(args);
@@ -381,7 +646,7 @@ xfs_dir2_leaf_addname(
tp = args->trans;
mp = dp->i_mount;
- error = xfs_dir2_leaf_read(tp, dp, mp->m_dirleafblk, -1, &lbp);
+ error = xfs_dir3_leaf_read(tp, dp, mp->m_dirleafblk, -1, &lbp);
if (error)
return error;
@@ -394,16 +659,19 @@ xfs_dir2_leaf_addname(
index = xfs_dir2_leaf_search_hash(args, lbp);
leaf = lbp->b_addr;
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
+ ents = xfs_dir3_leaf_ents_p(leaf);
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
bestsp = xfs_dir2_leaf_bests_p(ltp);
length = xfs_dir2_data_entsize(args->namelen);
+
/*
* See if there are any entries with the same hash value
* and space in their block for the new entry.
* This is good because it puts multiple same-hash value entries
* in a data block, improving the lookup of those entries.
*/
- for (use_block = -1, lep = &leaf->ents[index];
- index < be16_to_cpu(leaf->hdr.count) && be32_to_cpu(lep->hashval) == args->hashval;
+ for (use_block = -1, lep = &ents[index];
+ index < leafhdr.count && be32_to_cpu(lep->hashval) == args->hashval;
index++, lep++) {
if (be32_to_cpu(lep->address) == XFS_DIR2_NULL_DATAPTR)
continue;
@@ -436,7 +704,7 @@ xfs_dir2_leaf_addname(
* How many bytes do we need in the leaf block?
*/
needbytes = 0;
- if (!leaf->hdr.stale)
+ if (!leafhdr.stale)
needbytes += sizeof(xfs_dir2_leaf_entry_t);
if (use_block == -1)
needbytes += sizeof(xfs_dir2_data_off_t);
@@ -451,16 +719,15 @@ xfs_dir2_leaf_addname(
* If we don't have enough free bytes but we can make enough
* by compacting out stale entries, we'll do that.
*/
- if ((char *)bestsp - (char *)&leaf->ents[be16_to_cpu(leaf->hdr.count)] <
- needbytes && be16_to_cpu(leaf->hdr.stale) > 1) {
+ if ((char *)bestsp - (char *)&ents[leafhdr.count] < needbytes &&
+ leafhdr.stale > 1)
compact = 1;
- }
+
/*
* Otherwise if we don't have enough free bytes we need to
* convert to node form.
*/
- else if ((char *)bestsp - (char *)&leaf->ents[be16_to_cpu(
- leaf->hdr.count)] < needbytes) {
+ else if ((char *)bestsp - (char *)&ents[leafhdr.count] < needbytes) {
/*
* Just checking or no space reservation, give up.
*/
@@ -508,15 +775,15 @@ xfs_dir2_leaf_addname(
* point later.
*/
if (compact) {
- xfs_dir2_leaf_compact_x1(lbp, &index, &lowstale, &highstale,
- &lfloglow, &lfloghigh);
+ xfs_dir3_leaf_compact_x1(&leafhdr, ents, &index, &lowstale,
+ &highstale, &lfloglow, &lfloghigh);
}
/*
* There are stale entries, so we'll need log-low and log-high
* impossibly bad values later.
*/
- else if (be16_to_cpu(leaf->hdr.stale)) {
- lfloglow = be16_to_cpu(leaf->hdr.count);
+ else if (leafhdr.stale) {
+ lfloglow = leafhdr.count;
lfloghigh = -1;
}
/*
@@ -548,14 +815,14 @@ xfs_dir2_leaf_addname(
memmove(&bestsp[0], &bestsp[1],
be32_to_cpu(ltp->bestcount) * sizeof(bestsp[0]));
be32_add_cpu(<p->bestcount, 1);
- xfs_dir2_leaf_log_tail(tp, lbp);
- xfs_dir2_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
+ xfs_dir3_leaf_log_tail(tp, lbp);
+ xfs_dir3_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
}
/*
* If we're filling in a previously empty block just log it.
*/
else
- xfs_dir2_leaf_log_bests(tp, lbp, use_block, use_block);
+ xfs_dir3_leaf_log_bests(tp, lbp, use_block, use_block);
hdr = dbp->b_addr;
bf = xfs_dir3_data_bestfree_p(hdr);
bestsp[use_block] = bf[0].length;
@@ -616,10 +883,10 @@ xfs_dir2_leaf_addname(
if (be16_to_cpu(bestsp[use_block]) != be16_to_cpu(bf[0].length)) {
bestsp[use_block] = bf[0].length;
if (!grown)
- xfs_dir2_leaf_log_bests(tp, lbp, use_block, use_block);
+ xfs_dir3_leaf_log_bests(tp, lbp, use_block, use_block);
}
- lep = xfs_dir2_leaf_find_entry(leaf, index, compact, lowstale,
+ lep = xfs_dir3_leaf_find_entry(&leafhdr, ents, index, compact, lowstale,
highstale, &lfloglow, &lfloghigh);
/*
@@ -631,82 +898,40 @@ xfs_dir2_leaf_addname(
/*
* Log the leaf fields and give up the buffers.
*/
- xfs_dir2_leaf_log_header(tp, lbp);
- xfs_dir2_leaf_log_ents(tp, lbp, lfloglow, lfloghigh);
- xfs_dir2_leaf_check(dp, lbp);
+ xfs_dir3_leaf_hdr_to_disk(leaf, &leafhdr);
+ xfs_dir3_leaf_log_header(tp, lbp);
+ xfs_dir3_leaf_log_ents(tp, lbp, lfloglow, lfloghigh);
+ xfs_dir3_leaf_check(mp, lbp);
xfs_dir3_data_check(dp, dbp);
return 0;
}
-#ifdef DEBUG
-/*
- * Check the internal consistency of a leaf1 block.
- * Pop an assert if something is wrong.
- */
-STATIC void
-xfs_dir2_leaf_check(
- struct xfs_inode *dp, /* incore directory inode */
- struct xfs_buf *bp) /* leaf's buffer */
-{
- int i; /* leaf index */
- xfs_dir2_leaf_t *leaf; /* leaf structure */
- xfs_dir2_leaf_tail_t *ltp; /* leaf tail pointer */
- xfs_mount_t *mp; /* filesystem mount point */
- int stale; /* count of stale leaves */
-
- leaf = bp->b_addr;
- mp = dp->i_mount;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC));
- /*
- * This value is not restrictive enough.
- * Should factor in the size of the bests table as well.
- * We can deduce a value for that from di_size.
- */
- ASSERT(be16_to_cpu(leaf->hdr.count) <= xfs_dir2_max_leaf_ents(mp));
- ltp = xfs_dir2_leaf_tail_p(mp, leaf);
- /*
- * Leaves and bests don't overlap.
- */
- ASSERT((char *)&leaf->ents[be16_to_cpu(leaf->hdr.count)] <=
- (char *)xfs_dir2_leaf_bests_p(ltp));
- /*
- * Check hash value order, count stale entries.
- */
- for (i = stale = 0; i < be16_to_cpu(leaf->hdr.count); i++) {
- if (i + 1 < be16_to_cpu(leaf->hdr.count))
- ASSERT(be32_to_cpu(leaf->ents[i].hashval) <=
- be32_to_cpu(leaf->ents[i + 1].hashval));
- if (leaf->ents[i].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
- stale++;
- }
- ASSERT(be16_to_cpu(leaf->hdr.stale) == stale);
-}
-#endif /* DEBUG */
-
/*
* Compact out any stale entries in the leaf.
* Log the header and changed leaf entries, if any.
*/
void
-xfs_dir2_leaf_compact(
+xfs_dir3_leaf_compact(
xfs_da_args_t *args, /* operation arguments */
+ struct xfs_dir3_icleaf_hdr *leafhdr,
struct xfs_buf *bp) /* leaf buffer */
{
int from; /* source leaf index */
xfs_dir2_leaf_t *leaf; /* leaf structure */
int loglow; /* first leaf entry to log */
int to; /* target leaf index */
+ struct xfs_dir2_leaf_entry *ents;
leaf = bp->b_addr;
- if (!leaf->hdr.stale) {
+ if (!leafhdr->stale)
return;
- }
+
/*
* Compress out the stale entries in place.
*/
- for (from = to = 0, loglow = -1; from < be16_to_cpu(leaf->hdr.count); from++) {
- if (leaf->ents[from].address ==
- cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
+ ents = xfs_dir3_leaf_ents_p(leaf);
+ for (from = to = 0, loglow = -1; from < leafhdr->count; from++) {
+ if (ents[from].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
continue;
/*
* Only actually copy the entries that are different.
@@ -714,19 +939,21 @@ xfs_dir2_leaf_compact(
if (from > to) {
if (loglow == -1)
loglow = to;
- leaf->ents[to] = leaf->ents[from];
+ ents[to] = ents[from];
}
to++;
}
/*
* Update and log the header, log the leaf entries.
*/
- ASSERT(be16_to_cpu(leaf->hdr.stale) == from - to);
- be16_add_cpu(&leaf->hdr.count, -(be16_to_cpu(leaf->hdr.stale)));
- leaf->hdr.stale = 0;
- xfs_dir2_leaf_log_header(args->trans, bp);
+ ASSERT(leafhdr->stale == from - to);
+ leafhdr->count -= leafhdr->stale;
+ leafhdr->stale = 0;
+
+ xfs_dir3_leaf_hdr_to_disk(leaf, leafhdr);
+ xfs_dir3_leaf_log_header(args->trans, bp);
if (loglow != -1)
- xfs_dir2_leaf_log_ents(args->trans, bp, loglow, to - 1);
+ xfs_dir3_leaf_log_ents(args->trans, bp, loglow, to - 1);
}
/*
@@ -738,8 +965,9 @@ xfs_dir2_leaf_compact(
* and leaf logging indices.
*/
void
-xfs_dir2_leaf_compact_x1(
- struct xfs_buf *bp, /* leaf buffer */
+xfs_dir3_leaf_compact_x1(
+ struct xfs_dir3_icleaf_hdr *leafhdr,
+ struct xfs_dir2_leaf_entry *ents,
int *indexp, /* insertion index */
int *lowstalep, /* out: stale entry before us */
int *highstalep, /* out: stale entry after us */
@@ -750,22 +978,20 @@ xfs_dir2_leaf_compact_x1(
int highstale; /* stale entry at/after index */
int index; /* insertion index */
int keepstale; /* source index of kept stale */
- xfs_dir2_leaf_t *leaf; /* leaf structure */
int lowstale; /* stale entry before index */
int newindex=0; /* new insertion index */
int to; /* destination copy index */
- leaf = bp->b_addr;
- ASSERT(be16_to_cpu(leaf->hdr.stale) > 1);
+ ASSERT(leafhdr->stale > 1);
index = *indexp;
- xfs_dir2_leaf_find_stale(leaf, index, &lowstale, &highstale);
+ xfs_dir3_leaf_find_stale(leafhdr, ents, index, &lowstale, &highstale);
/*
* Pick the better of lowstale and highstale.
*/
if (lowstale >= 0 &&
- (highstale == be16_to_cpu(leaf->hdr.count) ||
+ (highstale == leafhdr->count ||
index - lowstale <= highstale - index))
keepstale = lowstale;
else
@@ -774,15 +1000,14 @@ xfs_dir2_leaf_compact_x1(
* Copy the entries in place, removing all the stale entries
* except keepstale.
*/
- for (from = to = 0; from < be16_to_cpu(leaf->hdr.count); from++) {
+ for (from = to = 0; from < leafhdr->count; from++) {
/*
* Notice the new value of index.
*/
if (index == from)
newindex = to;
if (from != keepstale &&
- leaf->ents[from].address ==
- cpu_to_be32(XFS_DIR2_NULL_DATAPTR)) {
+ ents[from].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR)) {
if (from == to)
*lowlogp = to;
continue;
@@ -796,7 +1021,7 @@ xfs_dir2_leaf_compact_x1(
* Copy only the entries that have moved.
*/
if (from > to)
- leaf->ents[to] = leaf->ents[from];
+ ents[to] = ents[from];
to++;
}
ASSERT(from > to);
@@ -810,8 +1035,8 @@ xfs_dir2_leaf_compact_x1(
/*
* Adjust the leaf header values.
*/
- be16_add_cpu(&leaf->hdr.count, -(from - to));
- leaf->hdr.stale = cpu_to_be16(1);
+ leafhdr->count -= from - to;
+ leafhdr->stale = 1;
/*
* Remember the low/high stale value only in the "right"
* direction.
@@ -819,75 +1044,18 @@ xfs_dir2_leaf_compact_x1(
if (lowstale >= newindex)
lowstale = -1;
else
- highstale = be16_to_cpu(leaf->hdr.count);
- *highlogp = be16_to_cpu(leaf->hdr.count) - 1;
+ highstale = leafhdr->count;
+ *highlogp = leafhdr->count - 1;
*lowstalep = lowstale;
*highstalep = highstale;
}
-/*
- * Initialize a new leaf block, leaf1 or leafn magic accepted.
- */
-int
-xfs_dir2_leaf_init(
- xfs_da_args_t *args, /* operation arguments */
- xfs_dir2_db_t bno, /* directory block number */
- struct xfs_buf **bpp, /* out: leaf buffer */
- int magic) /* magic number for block */
-{
- struct xfs_buf *bp; /* leaf buffer */
- xfs_inode_t *dp; /* incore directory inode */
- int error; /* error return code */
- xfs_dir2_leaf_t *leaf; /* leaf structure */
- xfs_dir2_leaf_tail_t *ltp; /* leaf tail structure */
- xfs_mount_t *mp; /* filesystem mount point */
- xfs_trans_t *tp; /* transaction pointer */
-
- dp = args->dp;
- ASSERT(dp != NULL);
- tp = args->trans;
- mp = dp->i_mount;
- ASSERT(bno >= XFS_DIR2_LEAF_FIRSTDB(mp) &&
- bno < XFS_DIR2_FREE_FIRSTDB(mp));
- /*
- * Get the buffer for the block.
- */
- error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(mp, bno), -1, &bp,
- XFS_DATA_FORK);
- if (error)
- return error;
-
- /*
- * Initialize the header.
- */
- leaf = bp->b_addr;
- leaf->hdr.info.magic = cpu_to_be16(magic);
- leaf->hdr.info.forw = 0;
- leaf->hdr.info.back = 0;
- leaf->hdr.count = 0;
- leaf->hdr.stale = 0;
- xfs_dir2_leaf_log_header(tp, bp);
- /*
- * If it's a leaf-format directory initialize the tail.
- * In this case our caller has the real bests table to copy into
- * the block.
- */
- if (magic == XFS_DIR2_LEAF1_MAGIC) {
- bp->b_ops = &xfs_dir2_leaf1_buf_ops;
- ltp = xfs_dir2_leaf_tail_p(mp, leaf);
- ltp->bestcount = 0;
- xfs_dir2_leaf_log_tail(tp, bp);
- } else
- bp->b_ops = &xfs_dir2_leafn_buf_ops;
- *bpp = bp;
- return 0;
-}
/*
* Log the bests entries indicated from a leaf1 block.
*/
static void
-xfs_dir2_leaf_log_bests(
+xfs_dir3_leaf_log_bests(
xfs_trans_t *tp, /* transaction pointer */
struct xfs_buf *bp, /* leaf buffer */
int first, /* first entry to log */
@@ -895,11 +1063,12 @@ xfs_dir2_leaf_log_bests(
{
__be16 *firstb; /* pointer to first entry */
__be16 *lastb; /* pointer to last entry */
- xfs_dir2_leaf_t *leaf; /* leaf structure */
+ struct xfs_dir2_leaf *leaf = bp->b_addr;
xfs_dir2_leaf_tail_t *ltp; /* leaf tail structure */
- leaf = bp->b_addr;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC));
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC) ||
+ leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAF1_MAGIC));
+
ltp = xfs_dir2_leaf_tail_p(tp->t_mountp, leaf);
firstb = xfs_dir2_leaf_bests_p(ltp) + first;
lastb = xfs_dir2_leaf_bests_p(ltp) + last;
@@ -911,7 +1080,7 @@ xfs_dir2_leaf_log_bests(
* Log the leaf entries indicated from a leaf1 or leafn block.
*/
void
-xfs_dir2_leaf_log_ents(
+xfs_dir3_leaf_log_ents(
xfs_trans_t *tp, /* transaction pointer */
struct xfs_buf *bp, /* leaf buffer */
int first, /* first entry to log */
@@ -919,13 +1088,17 @@ xfs_dir2_leaf_log_ents(
{
xfs_dir2_leaf_entry_t *firstlep; /* pointer to first entry */
xfs_dir2_leaf_entry_t *lastlep; /* pointer to last entry */
- xfs_dir2_leaf_t *leaf; /* leaf structure */
+ struct xfs_dir2_leaf *leaf = bp->b_addr;
+ struct xfs_dir2_leaf_entry *ents;
- leaf = bp->b_addr;
ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC) ||
- leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
- firstlep = &leaf->ents[first];
- lastlep = &leaf->ents[last];
+ leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAF1_MAGIC) ||
+ leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
+ leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC));
+
+ ents = xfs_dir3_leaf_ents_p(leaf);
+ firstlep = &ents[first];
+ lastlep = &ents[last];
xfs_trans_log_buf(tp, bp, (uint)((char *)firstlep - (char *)leaf),
(uint)((char *)lastlep - (char *)leaf + sizeof(*lastlep) - 1));
}
@@ -934,34 +1107,38 @@ xfs_dir2_leaf_log_ents(
* Log the header of the leaf1 or leafn block.
*/
void
-xfs_dir2_leaf_log_header(
+xfs_dir3_leaf_log_header(
struct xfs_trans *tp,
struct xfs_buf *bp)
{
- xfs_dir2_leaf_t *leaf; /* leaf structure */
+ struct xfs_dir2_leaf *leaf = bp->b_addr;
- leaf = bp->b_addr;
ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC) ||
- leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
+ leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAF1_MAGIC) ||
+ leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
+ leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC));
+
xfs_trans_log_buf(tp, bp, (uint)((char *)&leaf->hdr - (char *)leaf),
- (uint)(sizeof(leaf->hdr) - 1));
+ xfs_dir3_leaf_hdr_size(leaf) - 1);
}
/*
* Log the tail of the leaf1 block.
*/
STATIC void
-xfs_dir2_leaf_log_tail(
+xfs_dir3_leaf_log_tail(
struct xfs_trans *tp,
struct xfs_buf *bp)
{
- xfs_dir2_leaf_t *leaf; /* leaf structure */
+ struct xfs_dir2_leaf *leaf = bp->b_addr;
xfs_dir2_leaf_tail_t *ltp; /* leaf tail structure */
- xfs_mount_t *mp; /* filesystem mount point */
+ struct xfs_mount *mp = tp->t_mountp;
+
+ ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC) ||
+ leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAF1_MAGIC) ||
+ leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
+ leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC));
- mp = tp->t_mountp;
- leaf = bp->b_addr;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC));
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
xfs_trans_log_buf(tp, bp, (uint)((char *)ltp - (char *)leaf),
(uint)(mp->m_dirblksize - 1));
@@ -985,6 +1162,7 @@ xfs_dir2_leaf_lookup(
xfs_dir2_leaf_t *leaf; /* leaf structure */
xfs_dir2_leaf_entry_t *lep; /* leaf entry */
xfs_trans_t *tp; /* transaction pointer */
+ struct xfs_dir2_leaf_entry *ents;
trace_xfs_dir2_leaf_lookup(args);
@@ -996,12 +1174,14 @@ xfs_dir2_leaf_lookup(
}
tp = args->trans;
dp = args->dp;
- xfs_dir2_leaf_check(dp, lbp);
+ xfs_dir3_leaf_check(dp->i_mount, lbp);
leaf = lbp->b_addr;
+ ents = xfs_dir3_leaf_ents_p(leaf);
/*
* Get to the leaf entry and contained data entry address.
*/
- lep = &leaf->ents[index];
+ lep = &ents[index];
+
/*
* Point to the data entry.
*/
@@ -1045,18 +1225,23 @@ xfs_dir2_leaf_lookup_int(
xfs_trans_t *tp; /* transaction pointer */
xfs_dir2_db_t cidb = -1; /* case match data block no. */
enum xfs_dacmp cmp; /* name compare result */
+ struct xfs_dir2_leaf_entry *ents;
+ struct xfs_dir3_icleaf_hdr leafhdr;
dp = args->dp;
tp = args->trans;
mp = dp->i_mount;
- error = xfs_dir2_leaf_read(tp, dp, mp->m_dirleafblk, -1, &lbp);
+ error = xfs_dir3_leaf_read(tp, dp, mp->m_dirleafblk, -1, &lbp);
if (error)
return error;
*lbpp = lbp;
leaf = lbp->b_addr;
- xfs_dir2_leaf_check(dp, lbp);
+ xfs_dir3_leaf_check(mp, lbp);
+ ents = xfs_dir3_leaf_ents_p(leaf);
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+
/*
* Look for the first leaf entry with our hash value.
*/
@@ -1065,9 +1250,9 @@ xfs_dir2_leaf_lookup_int(
* Loop over all the entries with the right hash value
* looking to match the name.
*/
- for (lep = &leaf->ents[index]; index < be16_to_cpu(leaf->hdr.count) &&
- be32_to_cpu(lep->hashval) == args->hashval;
- lep++, index++) {
+ for (lep = &ents[index];
+ index < leafhdr.count && be32_to_cpu(lep->hashval) == args->hashval;
+ lep++, index++) {
/*
* Skip over stale leaf entries.
*/
@@ -1172,6 +1357,8 @@ xfs_dir2_leaf_removename(
xfs_dir2_data_off_t oldbest; /* old value of best free */
xfs_trans_t *tp; /* transaction pointer */
struct xfs_dir2_data_free *bf; /* bestfree table */
+ struct xfs_dir2_leaf_entry *ents;
+ struct xfs_dir3_icleaf_hdr leafhdr;
trace_xfs_dir2_leaf_removename(args);
@@ -1186,12 +1373,14 @@ xfs_dir2_leaf_removename(
mp = dp->i_mount;
leaf = lbp->b_addr;
hdr = dbp->b_addr;
- bf = xfs_dir3_data_bestfree_p(hdr);
xfs_dir3_data_check(dp, dbp);
+ bf = xfs_dir3_data_bestfree_p(hdr);
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+ ents = xfs_dir3_leaf_ents_p(leaf);
/*
* Point to the leaf entry, use that to point to the data entry.
*/
- lep = &leaf->ents[index];
+ lep = &ents[index];
db = xfs_dir2_dataptr_to_db(mp, be32_to_cpu(lep->address));
dep = (xfs_dir2_data_entry_t *)
((char *)hdr + xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address)));
@@ -1209,10 +1398,13 @@ xfs_dir2_leaf_removename(
/*
* We just mark the leaf entry stale by putting a null in it.
*/
- be16_add_cpu(&leaf->hdr.stale, 1);
- xfs_dir2_leaf_log_header(tp, lbp);
+ leafhdr.stale++;
+ xfs_dir3_leaf_hdr_to_disk(leaf, &leafhdr);
+ xfs_dir3_leaf_log_header(tp, lbp);
+
lep->address = cpu_to_be32(XFS_DIR2_NULL_DATAPTR);
- xfs_dir2_leaf_log_ents(tp, lbp, index, index);
+ xfs_dir3_leaf_log_ents(tp, lbp, index, index);
+
/*
* Scan the freespace in the data block again if necessary,
* log the data block header if necessary.
@@ -1227,7 +1419,7 @@ xfs_dir2_leaf_removename(
*/
if (be16_to_cpu(bf[0].length) != oldbest) {
bestsp[db] = bf[0].length;
- xfs_dir2_leaf_log_bests(tp, lbp, db, db);
+ xfs_dir3_leaf_log_bests(tp, lbp, db, db);
}
xfs_dir3_data_check(dp, dbp);
/*
@@ -1245,7 +1437,7 @@ xfs_dir2_leaf_removename(
*/
if (error == ENOSPC && args->total == 0)
error = 0;
- xfs_dir2_leaf_check(dp, lbp);
+ xfs_dir3_leaf_check(mp, lbp);
return error;
}
dbp = NULL;
@@ -1268,8 +1460,8 @@ xfs_dir2_leaf_removename(
memmove(&bestsp[db - i], bestsp,
(be32_to_cpu(ltp->bestcount) - (db - i)) * sizeof(*bestsp));
be32_add_cpu(<p->bestcount, -(db - i));
- xfs_dir2_leaf_log_tail(tp, lbp);
- xfs_dir2_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
+ xfs_dir3_leaf_log_tail(tp, lbp);
+ xfs_dir3_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
} else
bestsp[db] = cpu_to_be16(NULLDATAOFF);
}
@@ -1279,7 +1471,7 @@ xfs_dir2_leaf_removename(
else if (db != mp->m_dirdatablk)
dbp = NULL;
- xfs_dir2_leaf_check(dp, lbp);
+ xfs_dir3_leaf_check(mp, lbp);
/*
* See if we can convert to block form.
*/
@@ -1302,6 +1494,7 @@ xfs_dir2_leaf_replace(
xfs_dir2_leaf_t *leaf; /* leaf structure */
xfs_dir2_leaf_entry_t *lep; /* leaf entry */
xfs_trans_t *tp; /* transaction pointer */
+ struct xfs_dir2_leaf_entry *ents;
trace_xfs_dir2_leaf_replace(args);
@@ -1313,10 +1506,11 @@ xfs_dir2_leaf_replace(
}
dp = args->dp;
leaf = lbp->b_addr;
+ ents = xfs_dir3_leaf_ents_p(leaf);
/*
* Point to the leaf entry, get data address from it.
*/
- lep = &leaf->ents[index];
+ lep = &ents[index];
/*
* Point to the data entry.
*/
@@ -1330,7 +1524,7 @@ xfs_dir2_leaf_replace(
dep->inumber = cpu_to_be64(args->inumber);
tp = args->trans;
xfs_dir2_data_log_entry(tp, dbp, dep);
- xfs_dir2_leaf_check(dp, lbp);
+ xfs_dir3_leaf_check(dp->i_mount, lbp);
xfs_trans_brelse(tp, lbp);
return 0;
}
@@ -1352,17 +1546,22 @@ xfs_dir2_leaf_search_hash(
xfs_dir2_leaf_t *leaf; /* leaf structure */
xfs_dir2_leaf_entry_t *lep; /* leaf entry */
int mid=0; /* current leaf index */
+ struct xfs_dir2_leaf_entry *ents;
+ struct xfs_dir3_icleaf_hdr leafhdr;
leaf = lbp->b_addr;
+ ents = xfs_dir3_leaf_ents_p(leaf);
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+
#ifndef __KERNEL__
- if (!leaf->hdr.count)
+ if (!leafhdr.count)
return 0;
#endif
/*
* Note, the table cannot be empty, so we have to go through the loop.
* Binary search the leaf entries looking for our hash value.
*/
- for (lep = leaf->ents, low = 0, high = be16_to_cpu(leaf->hdr.count) - 1,
+ for (lep = ents, low = 0, high = leafhdr.count - 1,
hashwant = args->hashval;
low <= high; ) {
mid = (low + high) >> 1;
@@ -1448,23 +1647,29 @@ xfs_dir2_leaf_trim_data(
bestsp = xfs_dir2_leaf_bests_p(ltp);
be32_add_cpu(<p->bestcount, -1);
memmove(&bestsp[1], &bestsp[0], be32_to_cpu(ltp->bestcount) * sizeof(*bestsp));
- xfs_dir2_leaf_log_tail(tp, lbp);
- xfs_dir2_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
+ xfs_dir3_leaf_log_tail(tp, lbp);
+ xfs_dir3_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
return 0;
}
static inline size_t
-xfs_dir2_leaf_size(
- struct xfs_dir2_leaf_hdr *hdr,
+xfs_dir3_leaf_size(
+ struct xfs_dir3_icleaf_hdr *hdr,
int counts)
{
- int entries;
+ int entries;
+ int hdrsize;
+
+ entries = hdr->count - hdr->stale;
+ if (hdr->magic == XFS_DIR2_LEAF1_MAGIC ||
+ hdr->magic == XFS_DIR2_LEAFN_MAGIC)
+ hdrsize = sizeof(struct xfs_dir2_leaf_hdr);
+ else
+ hdrsize = sizeof(struct xfs_dir3_leaf_hdr);
- entries = be16_to_cpu(hdr->count) - be16_to_cpu(hdr->stale);
- return sizeof(xfs_dir2_leaf_hdr_t) +
- entries * sizeof(xfs_dir2_leaf_entry_t) +
- counts * sizeof(xfs_dir2_data_off_t) +
- sizeof(xfs_dir2_leaf_tail_t);
+ return hdrsize + entries * sizeof(xfs_dir2_leaf_entry_t)
+ + counts * sizeof(xfs_dir2_data_off_t)
+ + sizeof(xfs_dir2_leaf_tail_t);
}
/*
@@ -1488,6 +1693,7 @@ xfs_dir2_node_to_leaf(
xfs_mount_t *mp; /* filesystem mount point */
int rval; /* successful free trim? */
xfs_trans_t *tp; /* transaction pointer */
+ struct xfs_dir3_icleaf_hdr leafhdr;
struct xfs_dir3_icfree_hdr freehdr;
/*
@@ -1538,7 +1744,11 @@ xfs_dir2_node_to_leaf(
return 0;
lbp = state->path.blk[0].bp;
leaf = lbp->b_addr;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+
+ ASSERT(leafhdr.magic == XFS_DIR2_LEAFN_MAGIC ||
+ leafhdr.magic == XFS_DIR3_LEAFN_MAGIC);
+
/*
* Read the freespace block.
*/
@@ -1554,36 +1764,40 @@ xfs_dir2_node_to_leaf(
* Now see if the leafn and free data will fit in a leaf1.
* If not, release the buffer and give up.
*/
- if (xfs_dir2_leaf_size(&leaf->hdr, freehdr.nvalid) > mp->m_dirblksize) {
+ if (xfs_dir3_leaf_size(&leafhdr, freehdr.nvalid) > mp->m_dirblksize) {
xfs_trans_brelse(tp, fbp);
return 0;
}
/*
* If the leaf has any stale entries in it, compress them out.
- * The compact routine will log the header.
*/
- if (be16_to_cpu(leaf->hdr.stale))
- xfs_dir2_leaf_compact(args, lbp);
- else
- xfs_dir2_leaf_log_header(tp, lbp);
+ if (leafhdr.stale)
+ xfs_dir3_leaf_compact(args, &leafhdr, lbp);
- lbp->b_ops = &xfs_dir2_leaf1_buf_ops;
- leaf->hdr.info.magic = cpu_to_be16(XFS_DIR2_LEAF1_MAGIC);
+ lbp->b_ops = &xfs_dir3_leaf1_buf_ops;
+ leafhdr.magic = (leafhdr.magic == XFS_DIR2_LEAFN_MAGIC)
+ ? XFS_DIR2_LEAF1_MAGIC
+ : XFS_DIR3_LEAF1_MAGIC;
/*
* Set up the leaf tail from the freespace block.
*/
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
ltp->bestcount = cpu_to_be32(freehdr.nvalid);
+
/*
* Set up the leaf bests table.
*/
memcpy(xfs_dir2_leaf_bests_p(ltp), xfs_dir3_free_bests_p(mp, free),
freehdr.nvalid * sizeof(xfs_dir2_data_off_t));
- xfs_dir2_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
- xfs_dir2_leaf_log_tail(tp, lbp);
- xfs_dir2_leaf_check(dp, lbp);
+
+ xfs_dir3_leaf_hdr_to_disk(leaf, &leafhdr);
+ xfs_dir3_leaf_log_header(tp, lbp);
+ xfs_dir3_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
+ xfs_dir3_leaf_log_tail(tp, lbp);
+ xfs_dir3_leaf_check(mp, lbp);
+
/*
* Get rid of the freespace block.
*/
diff --git a/libxfs/xfs_dir2_node.c b/libxfs/xfs_dir2_node.c
index f87a245..9b93816 100644
--- a/libxfs/xfs_dir2_node.c
+++ b/libxfs/xfs_dir2_node.c
@@ -24,14 +24,6 @@
*/
static int xfs_dir2_leafn_add(struct xfs_buf *bp, xfs_da_args_t *args,
int index);
-#ifdef DEBUG
-static void xfs_dir2_leafn_check(struct xfs_inode *dp, struct xfs_buf *bp);
-#else
-#define xfs_dir2_leafn_check(dp, bp)
-#endif
-static void xfs_dir2_leafn_moveents(xfs_da_args_t *args, struct xfs_buf *bp_s,
- int start_s, struct xfs_buf *bp_d,
- int start_d, int count);
static void xfs_dir2_leafn_rebalance(xfs_da_state_t *state,
xfs_da_state_blk_t *blk1,
xfs_da_state_blk_t *blk2);
@@ -41,6 +33,39 @@ static int xfs_dir2_leafn_remove(xfs_da_args_t *args, struct xfs_buf *bp,
static int xfs_dir2_node_addname_int(xfs_da_args_t *args,
xfs_da_state_blk_t *fblk);
+/*
+ * Check internal consistency of a leafn block.
+ */
+#ifdef DEBUG
+#define xfs_dir3_leaf_check(mp, bp) \
+do { \
+ if (!xfs_dir3_leafn_check((mp), (bp))) \
+ ASSERT(0); \
+} while (0);
+
+static bool
+xfs_dir3_leafn_check(
+ struct xfs_mount *mp,
+ struct xfs_buf *bp)
+{
+ struct xfs_dir2_leaf *leaf = bp->b_addr;
+ struct xfs_dir3_icleaf_hdr leafhdr;
+
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+
+ if (leafhdr.magic == XFS_DIR3_LEAFN_MAGIC) {
+ struct xfs_dir3_leaf_hdr *leaf3 = bp->b_addr;
+ if (be64_to_cpu(leaf3->info.blkno) != bp->b_bn)
+ return false;
+ } else if (leafhdr.magic != XFS_DIR2_LEAFN_MAGIC)
+ return false;
+
+ return xfs_dir3_leaf_check_int(mp, &leafhdr, leaf);
+}
+#else
+#define xfs_dir3_leaf_check(mp, bp)
+#endif
+
static bool
xfs_dir3_free_verify(
struct xfs_buf *bp)
@@ -344,11 +369,19 @@ xfs_dir2_leaf_to_node(
xfs_dir2_free_log_bests(tp, fbp, 0, freehdr.nvalid - 1);
xfs_dir2_free_log_header(tp, fbp);
- /* convert the leaf to a leafnode */
- leaf->hdr.info.magic = cpu_to_be16(XFS_DIR2_LEAFN_MAGIC);
- lbp->b_ops = &xfs_dir2_leafn_buf_ops;
- xfs_dir2_leaf_log_header(tp, lbp);
- xfs_dir2_leafn_check(dp, lbp);
+ /*
+ * Converting the leaf to a leafnode is just a matter of changing the
+ * magic number and the ops. Do the change directly to the buffer as
+ * it's less work (and less code) than decoding the header to host
+ * format and back again.
+ */
+ if (leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC))
+ leaf->hdr.info.magic = cpu_to_be16(XFS_DIR2_LEAFN_MAGIC);
+ else
+ leaf->hdr.info.magic = cpu_to_be16(XFS_DIR3_LEAFN_MAGIC);
+ lbp->b_ops = &xfs_dir3_leafn_buf_ops;
+ xfs_dir3_leaf_log_header(tp, lbp);
+ xfs_dir3_leaf_check(mp, lbp);
return 0;
}
@@ -372,6 +405,8 @@ xfs_dir2_leafn_add(
int lowstale; /* previous stale entry */
xfs_mount_t *mp; /* filesystem mount point */
xfs_trans_t *tp; /* transaction pointer */
+ struct xfs_dir3_icleaf_hdr leafhdr;
+ struct xfs_dir2_leaf_entry *ents;
trace_xfs_dir2_leafn_add(args, index);
@@ -379,6 +414,8 @@ xfs_dir2_leafn_add(
mp = dp->i_mount;
tp = args->trans;
leaf = bp->b_addr;
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+ ents = xfs_dir3_leaf_ents_p(leaf);
/*
* Quick check just to make sure we are not going to index
@@ -394,15 +431,15 @@ xfs_dir2_leafn_add(
* a compact.
*/
- if (be16_to_cpu(leaf->hdr.count) == xfs_dir2_max_leaf_ents(mp)) {
- if (!leaf->hdr.stale)
+ if (leafhdr.count == xfs_dir3_max_leaf_ents(mp, leaf)) {
+ if (!leafhdr.stale)
return XFS_ERROR(ENOSPC);
- compact = be16_to_cpu(leaf->hdr.stale) > 1;
+ compact = leafhdr.stale > 1;
} else
compact = 0;
- ASSERT(index == 0 || be32_to_cpu(leaf->ents[index - 1].hashval) <= args->hashval);
- ASSERT(index == be16_to_cpu(leaf->hdr.count) ||
- be32_to_cpu(leaf->ents[index].hashval) >= args->hashval);
+ ASSERT(index == 0 || be32_to_cpu(ents[index - 1].hashval) <= args->hashval);
+ ASSERT(index == leafhdr.count ||
+ be32_to_cpu(ents[index].hashval) >= args->hashval);
if (args->op_flags & XFS_DA_OP_JUSTCHECK)
return 0;
@@ -411,62 +448,35 @@ xfs_dir2_leafn_add(
* Compact out all but one stale leaf entry. Leaves behind
* the entry closest to index.
*/
- if (compact) {
- xfs_dir2_leaf_compact_x1(bp, &index, &lowstale, &highstale,
- &lfloglow, &lfloghigh);
- }
- /*
- * Set impossible logging indices for this case.
- */
- else if (leaf->hdr.stale) {
- lfloglow = be16_to_cpu(leaf->hdr.count);
+ if (compact)
+ xfs_dir3_leaf_compact_x1(&leafhdr, ents, &index, &lowstale,
+ &highstale, &lfloglow, &lfloghigh);
+ else if (leafhdr.stale) {
+ /*
+ * Set impossible logging indices for this case.
+ */
+ lfloglow = leafhdr.count;
lfloghigh = -1;
}
/*
* Insert the new entry, log everything.
*/
- lep = xfs_dir2_leaf_find_entry(leaf, index, compact, lowstale,
+ lep = xfs_dir3_leaf_find_entry(&leafhdr, ents, index, compact, lowstale,
highstale, &lfloglow, &lfloghigh);
lep->hashval = cpu_to_be32(args->hashval);
lep->address = cpu_to_be32(xfs_dir2_db_off_to_dataptr(mp,
args->blkno, args->index));
- xfs_dir2_leaf_log_header(tp, bp);
- xfs_dir2_leaf_log_ents(tp, bp, lfloglow, lfloghigh);
- xfs_dir2_leafn_check(dp, bp);
+
+ xfs_dir3_leaf_hdr_to_disk(leaf, &leafhdr);
+ xfs_dir3_leaf_log_header(tp, bp);
+ xfs_dir3_leaf_log_ents(tp, bp, lfloglow, lfloghigh);
+ xfs_dir3_leaf_check(mp, bp);
return 0;
}
#ifdef DEBUG
-/*
- * Check internal consistency of a leafn block.
- */
-void
-xfs_dir2_leafn_check(
- struct xfs_inode *dp,
- struct xfs_buf *bp)
-{
- int i; /* leaf index */
- xfs_dir2_leaf_t *leaf; /* leaf structure */
- xfs_mount_t *mp; /* filesystem mount point */
- int stale; /* count of stale leaves */
-
- leaf = bp->b_addr;
- mp = dp->i_mount;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
- ASSERT(be16_to_cpu(leaf->hdr.count) <= xfs_dir2_max_leaf_ents(mp));
- for (i = stale = 0; i < be16_to_cpu(leaf->hdr.count); i++) {
- if (i + 1 < be16_to_cpu(leaf->hdr.count)) {
- ASSERT(be32_to_cpu(leaf->ents[i].hashval) <=
- be32_to_cpu(leaf->ents[i + 1].hashval));
- }
- if (leaf->ents[i].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
- stale++;
- }
- ASSERT(be16_to_cpu(leaf->hdr.stale) == stale);
-}
-
static void
xfs_dir2_free_hdr_check(
struct xfs_mount *mp,
@@ -494,15 +504,22 @@ xfs_dir2_leafn_lasthash(
struct xfs_buf *bp, /* leaf buffer */
int *count) /* count of entries in leaf */
{
- xfs_dir2_leaf_t *leaf; /* leaf structure */
+ struct xfs_dir2_leaf *leaf = bp->b_addr;
+ struct xfs_dir2_leaf_entry *ents;
+ struct xfs_dir3_icleaf_hdr leafhdr;
+
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+
+ ASSERT(leafhdr.magic == XFS_DIR2_LEAFN_MAGIC ||
+ leafhdr.magic == XFS_DIR3_LEAFN_MAGIC);
- leaf = bp->b_addr;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
if (count)
- *count = be16_to_cpu(leaf->hdr.count);
- if (!leaf->hdr.count)
+ *count = leafhdr.count;
+ if (!leafhdr.count)
return 0;
- return be32_to_cpu(leaf->ents[be16_to_cpu(leaf->hdr.count) - 1].hashval);
+
+ ents = xfs_dir3_leaf_ents_p(leaf);
+ return be32_to_cpu(ents[leafhdr.count - 1].hashval);
}
/*
@@ -531,16 +548,19 @@ xfs_dir2_leafn_lookup_for_addname(
xfs_dir2_db_t newdb; /* new data block number */
xfs_dir2_db_t newfdb; /* new free block number */
xfs_trans_t *tp; /* transaction pointer */
+ struct xfs_dir2_leaf_entry *ents;
+ struct xfs_dir3_icleaf_hdr leafhdr;
dp = args->dp;
tp = args->trans;
mp = dp->i_mount;
leaf = bp->b_addr;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
-#ifdef __KERNEL__
- ASSERT(be16_to_cpu(leaf->hdr.count) > 0);
-#endif
- xfs_dir2_leafn_check(dp, bp);
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+ ents = xfs_dir3_leaf_ents_p(leaf);
+
+ xfs_dir3_leaf_check(mp, bp);
+ ASSERT(leafhdr.count > 0);
+
/*
* Look up the hash value in the leaf entries.
*/
@@ -560,9 +580,9 @@ xfs_dir2_leafn_lookup_for_addname(
/*
* Loop over leaf entries with the right hash value.
*/
- for (lep = &leaf->ents[index]; index < be16_to_cpu(leaf->hdr.count) &&
- be32_to_cpu(lep->hashval) == args->hashval;
- lep++, index++) {
+ for (lep = &ents[index];
+ index < leafhdr.count && be32_to_cpu(lep->hashval) == args->hashval;
+ lep++, index++) {
/*
* Skip stale leaf entries.
*/
@@ -678,16 +698,19 @@ xfs_dir2_leafn_lookup_for_entry(
xfs_dir2_db_t newdb; /* new data block number */
xfs_trans_t *tp; /* transaction pointer */
enum xfs_dacmp cmp; /* comparison result */
+ struct xfs_dir2_leaf_entry *ents;
+ struct xfs_dir3_icleaf_hdr leafhdr;
dp = args->dp;
tp = args->trans;
mp = dp->i_mount;
leaf = bp->b_addr;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
-#ifdef __KERNEL__
- ASSERT(be16_to_cpu(leaf->hdr.count) > 0);
-#endif
- xfs_dir2_leafn_check(dp, bp);
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+ ents = xfs_dir3_leaf_ents_p(leaf);
+
+ xfs_dir3_leaf_check(mp, bp);
+ ASSERT(leafhdr.count > 0);
+
/*
* Look up the hash value in the leaf entries.
*/
@@ -702,9 +725,9 @@ xfs_dir2_leafn_lookup_for_entry(
/*
* Loop over leaf entries with the right hash value.
*/
- for (lep = &leaf->ents[index]; index < be16_to_cpu(leaf->hdr.count) &&
- be32_to_cpu(lep->hashval) == args->hashval;
- lep++, index++) {
+ for (lep = &ents[index];
+ index < leafhdr.count && be32_to_cpu(lep->hashval) == args->hashval;
+ lep++, index++) {
/*
* Skip stale leaf entries.
*/
@@ -776,8 +799,7 @@ xfs_dir2_leafn_lookup_for_entry(
return XFS_ERROR(EEXIST);
}
}
- ASSERT(index == be16_to_cpu(leaf->hdr.count) ||
- (args->op_flags & XFS_DA_OP_OKNOENT));
+ ASSERT(index == leafhdr.count || (args->op_flags & XFS_DA_OP_OKNOENT));
if (curbp) {
if (args->cmpresult == XFS_CMP_DIFFERENT) {
/* Giving back last used data block. */
@@ -822,52 +844,50 @@ xfs_dir2_leafn_lookup_int(
* Log entries and headers. Stale entries are preserved.
*/
static void
-xfs_dir2_leafn_moveents(
- xfs_da_args_t *args, /* operation arguments */
- struct xfs_buf *bp_s, /* source leaf buffer */
- int start_s, /* source leaf index */
- struct xfs_buf *bp_d, /* destination leaf buffer */
- int start_d, /* destination leaf index */
- int count) /* count of leaves to copy */
+xfs_dir3_leafn_moveents(
+ xfs_da_args_t *args, /* operation arguments */
+ struct xfs_buf *bp_s, /* source */
+ struct xfs_dir3_icleaf_hdr *shdr,
+ struct xfs_dir2_leaf_entry *sents,
+ int start_s,/* source leaf index */
+ struct xfs_buf *bp_d, /* destination */
+ struct xfs_dir3_icleaf_hdr *dhdr,
+ struct xfs_dir2_leaf_entry *dents,
+ int start_d,/* destination leaf index */
+ int count) /* count of leaves to copy */
{
- xfs_dir2_leaf_t *leaf_d; /* destination leaf structure */
- xfs_dir2_leaf_t *leaf_s; /* source leaf structure */
- int stale; /* count stale leaves copied */
- xfs_trans_t *tp; /* transaction pointer */
+ struct xfs_trans *tp = args->trans;
+ int stale; /* count stale leaves copied */
trace_xfs_dir2_leafn_moveents(args, start_s, start_d, count);
/*
* Silently return if nothing to do.
*/
- if (count == 0) {
+ if (count == 0)
return;
- }
- tp = args->trans;
- leaf_s = bp_s->b_addr;
- leaf_d = bp_d->b_addr;
+
/*
* If the destination index is not the end of the current
* destination leaf entries, open up a hole in the destination
* to hold the new entries.
*/
- if (start_d < be16_to_cpu(leaf_d->hdr.count)) {
- memmove(&leaf_d->ents[start_d + count], &leaf_d->ents[start_d],
- (be16_to_cpu(leaf_d->hdr.count) - start_d) *
- sizeof(xfs_dir2_leaf_entry_t));
- xfs_dir2_leaf_log_ents(tp, bp_d, start_d + count,
- count + be16_to_cpu(leaf_d->hdr.count) - 1);
+ if (start_d < dhdr->count) {
+ memmove(&dents[start_d + count], &dents[start_d],
+ (dhdr->count - start_d) * sizeof(xfs_dir2_leaf_entry_t));
+ xfs_dir3_leaf_log_ents(tp, bp_d, start_d + count,
+ count + dhdr->count - 1);
}
/*
* If the source has stale leaves, count the ones in the copy range
* so we can update the header correctly.
*/
- if (leaf_s->hdr.stale) {
+ if (shdr->stale) {
int i; /* temp leaf index */
for (i = start_s, stale = 0; i < start_s + count; i++) {
- if (leaf_s->ents[i].address ==
- cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
+ if (sents[i].address ==
+ cpu_to_be32(XFS_DIR2_NULL_DATAPTR))
stale++;
}
} else
@@ -875,29 +895,27 @@ xfs_dir2_leafn_moveents(
/*
* Copy the leaf entries from source to destination.
*/
- memcpy(&leaf_d->ents[start_d], &leaf_s->ents[start_s],
+ memcpy(&dents[start_d], &sents[start_s],
count * sizeof(xfs_dir2_leaf_entry_t));
- xfs_dir2_leaf_log_ents(tp, bp_d, start_d, start_d + count - 1);
+ xfs_dir3_leaf_log_ents(tp, bp_d, start_d, start_d + count - 1);
+
/*
* If there are source entries after the ones we copied,
* delete the ones we copied by sliding the next ones down.
*/
- if (start_s + count < be16_to_cpu(leaf_s->hdr.count)) {
- memmove(&leaf_s->ents[start_s], &leaf_s->ents[start_s + count],
+ if (start_s + count < shdr->count) {
+ memmove(&sents[start_s], &sents[start_s + count],
count * sizeof(xfs_dir2_leaf_entry_t));
- xfs_dir2_leaf_log_ents(tp, bp_s, start_s, start_s + count - 1);
+ xfs_dir3_leaf_log_ents(tp, bp_s, start_s, start_s + count - 1);
}
+
/*
* Update the headers and log them.
*/
- be16_add_cpu(&leaf_s->hdr.count, -(count));
- be16_add_cpu(&leaf_s->hdr.stale, -(stale));
- be16_add_cpu(&leaf_d->hdr.count, count);
- be16_add_cpu(&leaf_d->hdr.stale, stale);
- xfs_dir2_leaf_log_header(tp, bp_s);
- xfs_dir2_leaf_log_header(tp, bp_d);
- xfs_dir2_leafn_check(args->dp, bp_s);
- xfs_dir2_leafn_check(args->dp, bp_d);
+ shdr->count -= count;
+ shdr->stale -= stale;
+ dhdr->count += count;
+ dhdr->stale += stale;
}
/*
@@ -906,21 +924,25 @@ xfs_dir2_leafn_moveents(
*/
int /* sort order */
xfs_dir2_leafn_order(
- struct xfs_buf *leaf1_bp, /* leaf1 buffer */
- struct xfs_buf *leaf2_bp) /* leaf2 buffer */
+ struct xfs_buf *leaf1_bp, /* leaf1 buffer */
+ struct xfs_buf *leaf2_bp) /* leaf2 buffer */
{
- xfs_dir2_leaf_t *leaf1; /* leaf1 structure */
- xfs_dir2_leaf_t *leaf2; /* leaf2 structure */
-
- leaf1 = leaf1_bp->b_addr;
- leaf2 = leaf2_bp->b_addr;
- ASSERT(leaf1->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
- ASSERT(leaf2->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
- if (be16_to_cpu(leaf1->hdr.count) > 0 &&
- be16_to_cpu(leaf2->hdr.count) > 0 &&
- (be32_to_cpu(leaf2->ents[0].hashval) < be32_to_cpu(leaf1->ents[0].hashval) ||
- be32_to_cpu(leaf2->ents[be16_to_cpu(leaf2->hdr.count) - 1].hashval) <
- be32_to_cpu(leaf1->ents[be16_to_cpu(leaf1->hdr.count) - 1].hashval)))
+ struct xfs_dir2_leaf *leaf1 = leaf1_bp->b_addr;
+ struct xfs_dir2_leaf *leaf2 = leaf2_bp->b_addr;
+ struct xfs_dir2_leaf_entry *ents1;
+ struct xfs_dir2_leaf_entry *ents2;
+ struct xfs_dir3_icleaf_hdr hdr1;
+ struct xfs_dir3_icleaf_hdr hdr2;
+
+ xfs_dir3_leaf_hdr_from_disk(&hdr1, leaf1);
+ xfs_dir3_leaf_hdr_from_disk(&hdr2, leaf2);
+ ents1 = xfs_dir3_leaf_ents_p(leaf1);
+ ents2 = xfs_dir3_leaf_ents_p(leaf2);
+
+ if (hdr1.count > 0 && hdr2.count > 0 &&
+ (be32_to_cpu(ents2[0].hashval) < be32_to_cpu(ents1[0].hashval) ||
+ be32_to_cpu(ents2[hdr2.count - 1].hashval) <
+ be32_to_cpu(ents1[hdr1.count - 1].hashval)))
return 1;
return 0;
}
@@ -949,6 +971,10 @@ xfs_dir2_leafn_rebalance(
#endif
int oldsum; /* old total leaf count */
int swap; /* swapped leaf blocks */
+ struct xfs_dir2_leaf_entry *ents1;
+ struct xfs_dir2_leaf_entry *ents2;
+ struct xfs_dir3_icleaf_hdr hdr1;
+ struct xfs_dir3_icleaf_hdr hdr2;
args = state->args;
/*
@@ -963,11 +989,17 @@ xfs_dir2_leafn_rebalance(
}
leaf1 = blk1->bp->b_addr;
leaf2 = blk2->bp->b_addr;
- oldsum = be16_to_cpu(leaf1->hdr.count) + be16_to_cpu(leaf2->hdr.count);
+ xfs_dir3_leaf_hdr_from_disk(&hdr1, leaf1);
+ xfs_dir3_leaf_hdr_from_disk(&hdr2, leaf2);
+ ents1 = xfs_dir3_leaf_ents_p(leaf1);
+ ents2 = xfs_dir3_leaf_ents_p(leaf2);
+
+ oldsum = hdr1.count + hdr2.count;
#ifdef DEBUG
- oldstale = be16_to_cpu(leaf1->hdr.stale) + be16_to_cpu(leaf2->hdr.stale);
+ oldstale = hdr1.stale + hdr2.stale;
#endif
mid = oldsum >> 1;
+
/*
* If the old leaf count was odd then the new one will be even,
* so we need to divide the new count evenly.
@@ -975,10 +1007,10 @@ xfs_dir2_leafn_rebalance(
if (oldsum & 1) {
xfs_dahash_t midhash; /* middle entry hash value */
- if (mid >= be16_to_cpu(leaf1->hdr.count))
- midhash = be32_to_cpu(leaf2->ents[mid - be16_to_cpu(leaf1->hdr.count)].hashval);
+ if (mid >= hdr1.count)
+ midhash = be32_to_cpu(ents2[mid - hdr1.count].hashval);
else
- midhash = be32_to_cpu(leaf1->ents[mid].hashval);
+ midhash = be32_to_cpu(ents1[mid].hashval);
isleft = args->hashval <= midhash;
}
/*
@@ -992,30 +1024,42 @@ xfs_dir2_leafn_rebalance(
* Calculate moved entry count. Positive means left-to-right,
* negative means right-to-left. Then move the entries.
*/
- count = be16_to_cpu(leaf1->hdr.count) - mid + (isleft == 0);
+ count = hdr1.count - mid + (isleft == 0);
if (count > 0)
- xfs_dir2_leafn_moveents(args, blk1->bp,
- be16_to_cpu(leaf1->hdr.count) - count, blk2->bp, 0, count);
+ xfs_dir3_leafn_moveents(args, blk1->bp, &hdr1, ents1,
+ hdr1.count - count, blk2->bp,
+ &hdr2, ents2, 0, count);
else if (count < 0)
- xfs_dir2_leafn_moveents(args, blk2->bp, 0, blk1->bp,
- be16_to_cpu(leaf1->hdr.count), count);
- ASSERT(be16_to_cpu(leaf1->hdr.count) + be16_to_cpu(leaf2->hdr.count) == oldsum);
- ASSERT(be16_to_cpu(leaf1->hdr.stale) + be16_to_cpu(leaf2->hdr.stale) == oldstale);
+ xfs_dir3_leafn_moveents(args, blk2->bp, &hdr2, ents2, 0,
+ blk1->bp, &hdr1, ents1,
+ hdr1.count, count);
+
+ ASSERT(hdr1.count + hdr2.count == oldsum);
+ ASSERT(hdr1.stale + hdr2.stale == oldstale);
+
+ /* log the changes made when moving the entries */
+ xfs_dir3_leaf_hdr_to_disk(leaf1, &hdr1);
+ xfs_dir3_leaf_hdr_to_disk(leaf2, &hdr2);
+ xfs_dir3_leaf_log_header(args->trans, blk1->bp);
+ xfs_dir3_leaf_log_header(args->trans, blk2->bp);
+
+ xfs_dir3_leaf_check(args->dp->i_mount, blk1->bp);
+ xfs_dir3_leaf_check(args->dp->i_mount, blk2->bp);
+
/*
* Mark whether we're inserting into the old or new leaf.
*/
- if (be16_to_cpu(leaf1->hdr.count) < be16_to_cpu(leaf2->hdr.count))
+ if (hdr1.count < hdr2.count)
state->inleaf = swap;
- else if (be16_to_cpu(leaf1->hdr.count) > be16_to_cpu(leaf2->hdr.count))
+ else if (hdr1.count > hdr2.count)
state->inleaf = !swap;
else
- state->inleaf =
- swap ^ (blk1->index <= be16_to_cpu(leaf1->hdr.count));
+ state->inleaf = swap ^ (blk1->index <= hdr1.count);
/*
* Adjust the expected index for insertion.
*/
if (!state->inleaf)
- blk2->index = blk1->index - be16_to_cpu(leaf1->hdr.count);
+ blk2->index = blk1->index - hdr1.count;
/*
* Finally sanity check just to make sure we are not returning a
@@ -1137,6 +1181,8 @@ xfs_dir2_leafn_remove(
int needscan; /* need to rescan data frees */
xfs_trans_t *tp; /* transaction pointer */
struct xfs_dir2_data_free *bf; /* bestfree table */
+ struct xfs_dir3_icleaf_hdr leafhdr;
+ struct xfs_dir2_leaf_entry *ents;
trace_xfs_dir2_leafn_remove(args, index);
@@ -1144,11 +1190,14 @@ xfs_dir2_leafn_remove(
tp = args->trans;
mp = dp->i_mount;
leaf = bp->b_addr;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+ ents = xfs_dir3_leaf_ents_p(leaf);
+
/*
* Point to the entry we're removing.
*/
- lep = &leaf->ents[index];
+ lep = &ents[index];
+
/*
* Extract the data block and offset from the entry.
*/
@@ -1156,14 +1205,18 @@ xfs_dir2_leafn_remove(
ASSERT(dblk->blkno == db);
off = xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address));
ASSERT(dblk->index == off);
+
/*
* Kill the leaf entry by marking it stale.
* Log the leaf block changes.
*/
- be16_add_cpu(&leaf->hdr.stale, 1);
- xfs_dir2_leaf_log_header(tp, bp);
+ leafhdr.stale++;
+ xfs_dir3_leaf_hdr_to_disk(leaf, &leafhdr);
+ xfs_dir3_leaf_log_header(tp, bp);
+
lep->address = cpu_to_be32(XFS_DIR2_NULL_DATAPTR);
- xfs_dir2_leaf_log_ents(tp, bp, index, index);
+ xfs_dir3_leaf_log_ents(tp, bp, index, index);
+
/*
* Make the data entry free. Keep track of the longest freespace
* in the data block in case it changes.
@@ -1252,15 +1305,13 @@ xfs_dir2_leafn_remove(
return error;
}
- xfs_dir2_leafn_check(dp, bp);
+ xfs_dir3_leaf_check(mp, bp);
/*
* Return indication of whether this leaf block is empty enough
* to justify trying to join it with a neighbor.
*/
- *rval =
- ((uint)sizeof(leaf->hdr) +
- (uint)sizeof(leaf->ents[0]) *
- (be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale))) <
+ *rval = (xfs_dir3_leaf_hdr_size(leaf) +
+ (uint)sizeof(ents[0]) * (leafhdr.count - leafhdr.stale)) <
mp->m_dir_magicpct;
return 0;
}
@@ -1293,11 +1344,11 @@ xfs_dir2_leafn_split(
/*
* Initialize the new leaf block.
*/
- error = xfs_dir2_leaf_init(args, xfs_dir2_da_to_db(mp, blkno),
- &newblk->bp, XFS_DIR2_LEAFN_MAGIC);
- if (error) {
+ error = xfs_dir3_leaf_get_buf(args, xfs_dir2_da_to_db(mp, blkno),
+ &newblk->bp, XFS_DIR2_LEAFN_MAGIC);
+ if (error)
return error;
- }
+
newblk->blkno = blkno;
newblk->magic = XFS_DIR2_LEAFN_MAGIC;
/*
@@ -1321,8 +1372,8 @@ xfs_dir2_leafn_split(
*/
oldblk->hashval = xfs_dir2_leafn_lasthash(oldblk->bp, NULL);
newblk->hashval = xfs_dir2_leafn_lasthash(newblk->bp, NULL);
- xfs_dir2_leafn_check(args->dp, oldblk->bp);
- xfs_dir2_leafn_check(args->dp, newblk->bp);
+ xfs_dir3_leaf_check(mp, oldblk->bp);
+ xfs_dir3_leaf_check(mp, newblk->bp);
return error;
}
@@ -1348,9 +1399,10 @@ xfs_dir2_leafn_toosmall(
int error; /* error return value */
int forward; /* sibling block direction */
int i; /* sibling counter */
- xfs_da_blkinfo_t *info; /* leaf block header */
xfs_dir2_leaf_t *leaf; /* leaf structure */
int rval; /* result from path_shift */
+ struct xfs_dir3_icleaf_hdr leafhdr;
+ struct xfs_dir2_leaf_entry *ents;
/*
* Check for the degenerate case of the block being over 50% full.
@@ -1358,11 +1410,13 @@ xfs_dir2_leafn_toosmall(
* to coalesce with a sibling.
*/
blk = &state->path.blk[state->path.active - 1];
- info = blk->bp->b_addr;
- ASSERT(info->magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
- leaf = (xfs_dir2_leaf_t *)info;
- count = be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale);
- bytes = (uint)sizeof(leaf->hdr) + count * (uint)sizeof(leaf->ents[0]);
+ leaf = blk->bp->b_addr;
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
+ ents = xfs_dir3_leaf_ents_p(leaf);
+ xfs_dir3_leaf_check(mp, blk->bp);
+
+ count = leafhdr.count - leafhdr.stale;
+ bytes = xfs_dir3_leaf_hdr_size(leaf) + count * sizeof(ents[0]);
if (bytes > (state->blocksize >> 1)) {
/*
* Blk over 50%, don't try to join.
@@ -1381,7 +1435,7 @@ xfs_dir2_leafn_toosmall(
* Make altpath point to the block we want to keep and
* path point to the block we want to drop (this one).
*/
- forward = (info->forw != 0);
+ forward = (leafhdr.forw != 0);
memcpy(&state->altpath, &state->path, sizeof(state->path));
error = xfs_da_path_shift(state, &state->altpath, forward, 0,
&rval);
@@ -1397,15 +1451,17 @@ xfs_dir2_leafn_toosmall(
* We prefer coalescing with the lower numbered sibling so as
* to shrink a directory over time.
*/
- forward = be32_to_cpu(info->forw) < be32_to_cpu(info->back);
+ forward = leafhdr.forw < leafhdr.back;
for (i = 0, bp = NULL; i < 2; forward = !forward, i++) {
- blkno = forward ? be32_to_cpu(info->forw) : be32_to_cpu(info->back);
+ struct xfs_dir3_icleaf_hdr hdr2;
+
+ blkno = forward ? leafhdr.forw : leafhdr.back;
if (blkno == 0)
continue;
/*
* Read the sibling leaf block.
*/
- error = xfs_dir2_leafn_read(state->args->trans, state->args->dp,
+ error = xfs_dir3_leafn_read(state->args->trans, state->args->dp,
blkno, -1, &bp);
if (error)
return error;
@@ -1413,13 +1469,15 @@ xfs_dir2_leafn_toosmall(
/*
* Count bytes in the two blocks combined.
*/
- leaf = (xfs_dir2_leaf_t *)info;
- count = be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale);
+ count = leafhdr.count - leafhdr.stale;
bytes = state->blocksize - (state->blocksize >> 2);
+
leaf = bp->b_addr;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
- count += be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale);
- bytes -= count * (uint)sizeof(leaf->ents[0]);
+ xfs_dir3_leaf_hdr_from_disk(&hdr2, leaf);
+ ents = xfs_dir3_leaf_ents_p(leaf);
+ count += hdr2.count - hdr2.stale;
+ bytes -= count * sizeof(ents[0]);
+
/*
* Fits with at least 25% to spare.
*/
@@ -1466,34 +1524,53 @@ xfs_dir2_leafn_unbalance(
xfs_da_args_t *args; /* operation arguments */
xfs_dir2_leaf_t *drop_leaf; /* dead leaf structure */
xfs_dir2_leaf_t *save_leaf; /* surviving leaf structure */
+ struct xfs_dir3_icleaf_hdr savehdr;
+ struct xfs_dir3_icleaf_hdr drophdr;
+ struct xfs_dir2_leaf_entry *sents;
+ struct xfs_dir2_leaf_entry *dents;
args = state->args;
ASSERT(drop_blk->magic == XFS_DIR2_LEAFN_MAGIC);
ASSERT(save_blk->magic == XFS_DIR2_LEAFN_MAGIC);
drop_leaf = drop_blk->bp->b_addr;
save_leaf = save_blk->bp->b_addr;
- ASSERT(drop_leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
- ASSERT(save_leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC));
+
+ xfs_dir3_leaf_hdr_from_disk(&savehdr, save_leaf);
+ xfs_dir3_leaf_hdr_from_disk(&drophdr, drop_leaf);
+ sents = xfs_dir3_leaf_ents_p(save_leaf);
+ dents = xfs_dir3_leaf_ents_p(drop_leaf);
+
/*
* If there are any stale leaf entries, take this opportunity
* to purge them.
*/
- if (drop_leaf->hdr.stale)
- xfs_dir2_leaf_compact(args, drop_blk->bp);
- if (save_leaf->hdr.stale)
- xfs_dir2_leaf_compact(args, save_blk->bp);
+ if (drophdr.stale)
+ xfs_dir3_leaf_compact(args, &drophdr, drop_blk->bp);
+ if (savehdr.stale)
+ xfs_dir3_leaf_compact(args, &savehdr, save_blk->bp);
+
/*
* Move the entries from drop to the appropriate end of save.
*/
- drop_blk->hashval = be32_to_cpu(drop_leaf->ents[be16_to_cpu(drop_leaf->hdr.count) - 1].hashval);
+ drop_blk->hashval = be32_to_cpu(dents[drophdr.count - 1].hashval);
if (xfs_dir2_leafn_order(save_blk->bp, drop_blk->bp))
- xfs_dir2_leafn_moveents(args, drop_blk->bp, 0, save_blk->bp, 0,
- be16_to_cpu(drop_leaf->hdr.count));
+ xfs_dir3_leafn_moveents(args, drop_blk->bp, &drophdr, dents, 0,
+ save_blk->bp, &savehdr, sents, 0,
+ drophdr.count);
else
- xfs_dir2_leafn_moveents(args, drop_blk->bp, 0, save_blk->bp,
- be16_to_cpu(save_leaf->hdr.count), be16_to_cpu(drop_leaf->hdr.count));
- save_blk->hashval = be32_to_cpu(save_leaf->ents[be16_to_cpu(save_leaf->hdr.count) - 1].hashval);
- xfs_dir2_leafn_check(args->dp, save_blk->bp);
+ xfs_dir3_leafn_moveents(args, drop_blk->bp, &drophdr, dents, 0,
+ save_blk->bp, &savehdr, sents,
+ savehdr.count, drophdr.count);
+ save_blk->hashval = be32_to_cpu(sents[savehdr.count - 1].hashval);
+
+ /* log the changes made when moving the entries */
+ xfs_dir3_leaf_hdr_to_disk(save_leaf, &savehdr);
+ xfs_dir3_leaf_hdr_to_disk(drop_leaf, &drophdr);
+ xfs_dir3_leaf_log_header(args->trans, save_blk->bp);
+ xfs_dir3_leaf_log_header(args->trans, drop_blk->bp);
+
+ xfs_dir3_leaf_check(args->dp->i_mount, save_blk->bp);
+ xfs_dir3_leaf_check(args->dp->i_mount, drop_blk->bp);
}
/*
@@ -2098,13 +2175,15 @@ xfs_dir2_node_replace(
* and locked it. But paranoia is good.
*/
if (rval == EEXIST) {
+ struct xfs_dir2_leaf_entry *ents;
/*
* Find the leaf entry.
*/
blk = &state->path.blk[state->path.active - 1];
ASSERT(blk->magic == XFS_DIR2_LEAFN_MAGIC);
leaf = blk->bp->b_addr;
- lep = &leaf->ents[blk->index];
+ ents = xfs_dir3_leaf_ents_p(leaf);
+ lep = &ents[blk->index];
ASSERT(state->extravalid);
/*
* Point to the data entry.
diff --git a/libxfs/xfs_dir2_priv.h b/libxfs/xfs_dir2_priv.h
index 910e644..932565d 100644
--- a/libxfs/xfs_dir2_priv.h
+++ b/libxfs/xfs_dir2_priv.h
@@ -77,24 +77,25 @@ extern void xfs_dir2_data_use_free(struct xfs_trans *tp, struct xfs_buf *bp,
xfs_dir2_data_aoff_t len, int *needlogp, int *needscanp);
/* xfs_dir2_leaf.c */
-extern const struct xfs_buf_ops xfs_dir2_leafn_buf_ops;
+extern const struct xfs_buf_ops xfs_dir3_leafn_buf_ops;
-extern int xfs_dir2_leafn_read(struct xfs_trans *tp, struct xfs_inode *dp,
+extern int xfs_dir3_leafn_read(struct xfs_trans *tp, struct xfs_inode *dp,
xfs_dablk_t fbno, xfs_daddr_t mappedbno, struct xfs_buf **bpp);
extern int xfs_dir2_block_to_leaf(struct xfs_da_args *args,
struct xfs_buf *dbp);
extern int xfs_dir2_leaf_addname(struct xfs_da_args *args);
-extern void xfs_dir2_leaf_compact(struct xfs_da_args *args,
- struct xfs_buf *bp);
-extern void xfs_dir2_leaf_compact_x1(struct xfs_buf *bp, int *indexp,
+extern void xfs_dir3_leaf_compact(struct xfs_da_args *args,
+ struct xfs_dir3_icleaf_hdr *leafhdr, struct xfs_buf *bp);
+extern void xfs_dir3_leaf_compact_x1(struct xfs_dir3_icleaf_hdr *leafhdr,
+ struct xfs_dir2_leaf_entry *ents, int *indexp,
int *lowstalep, int *highstalep, int *lowlogp, int *highlogp);
extern int xfs_dir2_leaf_getdents(struct xfs_inode *dp, void *dirent,
size_t bufsize, xfs_off_t *offset, filldir_t filldir);
-extern int xfs_dir2_leaf_init(struct xfs_da_args *args, xfs_dir2_db_t bno,
- struct xfs_buf **bpp, int magic);
-extern void xfs_dir2_leaf_log_ents(struct xfs_trans *tp, struct xfs_buf *bp,
+extern int xfs_dir3_leaf_get_buf(struct xfs_da_args *args, xfs_dir2_db_t bno,
+ struct xfs_buf **bpp, __uint16_t magic);
+extern void xfs_dir3_leaf_log_ents(struct xfs_trans *tp, struct xfs_buf *bp,
int first, int last);
-extern void xfs_dir2_leaf_log_header(struct xfs_trans *tp,
+extern void xfs_dir3_leaf_log_header(struct xfs_trans *tp,
struct xfs_buf *bp);
extern int xfs_dir2_leaf_lookup(struct xfs_da_args *args);
extern int xfs_dir2_leaf_removename(struct xfs_da_args *args);
@@ -104,11 +105,18 @@ extern int xfs_dir2_leaf_search_hash(struct xfs_da_args *args,
extern int xfs_dir2_leaf_trim_data(struct xfs_da_args *args,
struct xfs_buf *lbp, xfs_dir2_db_t db);
extern struct xfs_dir2_leaf_entry *
-xfs_dir2_leaf_find_entry(struct xfs_dir2_leaf *leaf, int index, int compact,
- int lowstale, int highstale,
- int *lfloglow, int *lfloghigh);
+xfs_dir3_leaf_find_entry(struct xfs_dir3_icleaf_hdr *leafhdr,
+ struct xfs_dir2_leaf_entry *ents, int index, int compact,
+ int lowstale, int highstale, int *lfloglow, int *lfloghigh);
extern int xfs_dir2_node_to_leaf(struct xfs_da_state *state);
+extern void xfs_dir3_leaf_hdr_from_disk(struct xfs_dir3_icleaf_hdr *to,
+ struct xfs_dir2_leaf *from);
+extern void xfs_dir3_leaf_hdr_to_disk(struct xfs_dir2_leaf *to,
+ struct xfs_dir3_icleaf_hdr *from);
+extern bool xfs_dir3_leaf_check_int(struct xfs_mount *mp,
+ struct xfs_dir3_icleaf_hdr *hdr, struct xfs_dir2_leaf *leaf);
+
/* xfs_dir2_node.c */
extern int xfs_dir2_leaf_to_node(struct xfs_da_args *args,
struct xfs_buf *lbp);
diff --git a/repair/dir2.c b/repair/dir2.c
index c455c83..a48971f 100644
--- a/repair/dir2.c
+++ b/repair/dir2.c
@@ -1607,24 +1607,26 @@ process_leaf_block_dir2(
{
int i;
int stale;
+ struct xfs_dir2_leaf_entry *ents;
+
+ ents = xfs_dir3_leaf_ents_p(leaf);
for (i = stale = 0; i < be16_to_cpu(leaf->hdr.count); i++) {
- if ((char *)&leaf->ents[i] >= (char *)leaf + mp->m_dirblksize) {
+ if ((char *)&ents[i] >= (char *)leaf + mp->m_dirblksize) {
do_warn(
_("bad entry count in block %u of directory inode %" PRIu64 "\n"),
da_bno, ino);
return 1;
}
- if (be32_to_cpu(leaf->ents[i].address) == XFS_DIR2_NULL_DATAPTR)
+ if (be32_to_cpu(ents[i].address) == XFS_DIR2_NULL_DATAPTR)
stale++;
- else if (be32_to_cpu(leaf->ents[i].hashval) < last_hashval) {
+ else if (be32_to_cpu(ents[i].hashval) < last_hashval) {
do_warn(
_("bad hash ordering in block %u of directory inode %" PRIu64 "\n"),
da_bno, ino);
return 1;
}
- *next_hashval = last_hashval =
- be32_to_cpu(leaf->ents[i].hashval);
+ *next_hashval = last_hashval = be32_to_cpu(ents[i].hashval);
}
if (stale != be16_to_cpu(leaf->hdr.stale)) {
do_warn(
diff --git a/repair/phase6.c b/repair/phase6.c
index 80701a4..243b808 100644
--- a/repair/phase6.c
+++ b/repair/phase6.c
@@ -2327,6 +2327,7 @@ longform_dir2_check_leaf(
xfs_dir2_leaf_t *leaf;
xfs_dir2_leaf_tail_t *ltp;
int seeval;
+ struct xfs_dir2_leaf_entry *ents;
da_bno = mp->m_dirleafblk;
if (libxfs_da_read_buf(NULL, ip, da_bno, -1, &bp, XFS_DATA_FORK, NULL)) {
@@ -2336,6 +2337,7 @@ longform_dir2_check_leaf(
/* NOTREACHED */
}
leaf = bp->b_addr;
+ ents = xfs_dir3_leaf_ents_p(leaf);
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
bestsp = xfs_dir2_leaf_bests_p(ltp);
if (be16_to_cpu(leaf->hdr.info.magic) != XFS_DIR2_LEAF1_MAGIC ||
@@ -2344,8 +2346,8 @@ longform_dir2_check_leaf(
be16_to_cpu(leaf->hdr.count) <
be16_to_cpu(leaf->hdr.stale) ||
be16_to_cpu(leaf->hdr.count) >
- xfs_dir2_max_leaf_ents(mp) ||
- (char *)&leaf->ents[be16_to_cpu(
+ xfs_dir3_max_leaf_ents(mp, leaf) ||
+ (char *)&ents[be16_to_cpu(
leaf->hdr.count)] > (char *)bestsp) {
do_warn(
_("leaf block %u for directory inode %" PRIu64 " bad header\n"),
@@ -2353,7 +2355,7 @@ longform_dir2_check_leaf(
libxfs_putbuf(bp);
return 1;
}
- seeval = dir_hash_see_all(hashtab, leaf->ents,
+ seeval = dir_hash_see_all(hashtab, ents,
be16_to_cpu(leaf->hdr.count),
be16_to_cpu(leaf->hdr.stale));
if (dir_hash_check(hashtab, ip, seeval)) {
@@ -2396,6 +2398,7 @@ longform_dir2_check_node(
xfs_fileoff_t next_da_bno;
int seeval = 0;
int used;
+ struct xfs_dir2_leaf_entry *ents;
for (da_bno = mp->m_dirleafblk, next_da_bno = 0;
next_da_bno != NULLFILEOFF && da_bno < mp->m_dirfreeblk;
@@ -2411,6 +2414,7 @@ longform_dir2_check_node(
return 1;
}
leaf = bp->b_addr;
+ ents = xfs_dir3_leaf_ents_p(leaf);
if (be16_to_cpu(leaf->hdr.info.magic) != XFS_DIR2_LEAFN_MAGIC) {
if (be16_to_cpu(leaf->hdr.info.magic) ==
XFS_DA_NODE_MAGIC) {
@@ -2424,7 +2428,7 @@ longform_dir2_check_node(
libxfs_putbuf(bp);
return 1;
}
- if (be16_to_cpu(leaf->hdr.count) > xfs_dir2_max_leaf_ents(mp) ||
+ if (be16_to_cpu(leaf->hdr.count) > xfs_dir3_max_leaf_ents(mp, leaf) ||
be16_to_cpu(leaf->hdr.count) <
be16_to_cpu(leaf->hdr.stale)) {
do_warn(
@@ -2433,7 +2437,7 @@ longform_dir2_check_node(
libxfs_putbuf(bp);
return 1;
}
- seeval = dir_hash_see_all(hashtab, leaf->ents,
+ seeval = dir_hash_see_all(hashtab, ents,
be16_to_cpu(leaf->hdr.count),
be16_to_cpu(leaf->hdr.stale));
libxfs_putbuf(bp);
xfsprogs-crc-patchset-v2/dir2-sf 0000600 0001750 0001750 00000012163 12132011546 015433 0 ustar dave dave xfs: shortform directory offsets change for dir3 format
From: Dave Chinner <dchinner@xxxxxxxxxx>
Because the header size for the CRC enabled directory blocks is
larger, the offset of the first entry into a directory block is
different to the dir2 format. The shortform directory stores the
dirent's offset so that it doesn't change when moving from shortform
to block form and back again, and hence it needs to take into
account the different header sizes to maintain the correct offsets.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
db/check.c | 2 +-
include/xfs_dir2_format.h | 25 ++++++++++++++-----------
libxfs/xfs_dir2_sf.c | 6 +++---
repair/dir2.c | 7 ++++---
4 files changed, 22 insertions(+), 18 deletions(-)
diff --git a/db/check.c b/db/check.c
index 73d9da2..69cfbdb 100644
--- a/db/check.c
+++ b/db/check.c
@@ -3647,7 +3647,7 @@ process_sf_dir_v2(
dbprintf(_("dir %lld entry . %lld\n"), id->ino, id->ino);
(*dot)++;
sfe = xfs_dir2_sf_firstentry(&sf->hdr);
- offset = XFS_DIR2_DATA_FIRST_OFFSET;
+ offset = XFS_DIR3_DATA_FIRST_OFFSET(mp);
for (i = sf->hdr.count - 1, i8 = 0; i >= 0; i--) {
if ((__psint_t)sfe + xfs_dir2_sf_entsize(&sf->hdr,sfe->namelen) -
(__psint_t)sf > be64_to_cpu(dip->di_size)) {
diff --git a/include/xfs_dir2_format.h b/include/xfs_dir2_format.h
index 8bb4efe..8d33f03 100644
--- a/include/xfs_dir2_format.h
+++ b/include/xfs_dir2_format.h
@@ -228,16 +228,6 @@ xfs_dir2_sf_nextentry(struct xfs_dir2_sf_hdr *hdr,
xfs_dir2_byte_to_db(mp, XFS_DIR2_DATA_OFFSET)
/*
- * Offsets of . and .. in data space (always block 0)
- */
-#define XFS_DIR2_DATA_DOT_OFFSET \
- ((xfs_dir2_data_aoff_t)sizeof(struct xfs_dir2_data_hdr))
-#define XFS_DIR2_DATA_DOTDOT_OFFSET \
- (XFS_DIR2_DATA_DOT_OFFSET + xfs_dir2_data_entsize(1))
-#define XFS_DIR2_DATA_FIRST_OFFSET \
- (XFS_DIR2_DATA_DOTDOT_OFFSET + xfs_dir2_data_entsize(2))
-
-/*
* Describe a free area in the data block.
*
* The freespace will be formatted as a xfs_dir2_data_unused_t.
@@ -378,7 +368,20 @@ xfs_dir3_data_unused_p(struct xfs_dir2_data_hdr *hdr)
/*
* Offsets of . and .. in data space (always block 0)
- */
+ *
+ * The macros are used for shortform directories as they have no headers to read
+ * the magic number out of. Shortform directories need to know the size of the
+ * data block header because the sfe embeds the block offset of the entry into
+ * it so that it doesn't change when format conversion occurs. Bad Things Happen
+ * if we don't follow this rule.
+ */
+#define XFS_DIR3_DATA_DOT_OFFSET(mp) \
+ xfs_dir3_data_hdr_size(xfs_sb_version_hascrc(&(mp)->m_sb))
+#define XFS_DIR3_DATA_DOTDOT_OFFSET(mp) \
+ (XFS_DIR3_DATA_DOT_OFFSET(mp) + xfs_dir2_data_entsize(1))
+#define XFS_DIR3_DATA_FIRST_OFFSET(mp) \
+ (XFS_DIR3_DATA_DOTDOT_OFFSET(mp) + xfs_dir2_data_entsize(2))
+
static inline xfs_dir2_data_aoff_t
xfs_dir3_data_dot_offset(struct xfs_dir2_data_hdr *hdr)
{
diff --git a/libxfs/xfs_dir2_sf.c b/libxfs/xfs_dir2_sf.c
index 6848d05..cb23368 100644
--- a/libxfs/xfs_dir2_sf.c
+++ b/libxfs/xfs_dir2_sf.c
@@ -519,7 +519,7 @@ xfs_dir2_sf_addname_hard(
* to insert the new entry.
* If it's going to end up at the end then oldsfep will point there.
*/
- for (offset = XFS_DIR2_DATA_FIRST_OFFSET,
+ for (offset = XFS_DIR3_DATA_FIRST_OFFSET(dp->i_mount),
oldsfep = xfs_dir2_sf_firstentry(oldsfp),
add_datasize = xfs_dir2_data_entsize(args->namelen),
eof = (char *)oldsfep == &buf[old_isize];
@@ -601,7 +601,7 @@ xfs_dir2_sf_addname_pick(
sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
size = xfs_dir2_data_entsize(args->namelen);
- offset = XFS_DIR2_DATA_FIRST_OFFSET;
+ offset = XFS_DIR3_DATA_FIRST_OFFSET(mp);
sfep = xfs_dir2_sf_firstentry(sfp);
holefit = 0;
/*
@@ -672,7 +672,7 @@ xfs_dir2_sf_check(
dp = args->dp;
sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
- offset = XFS_DIR2_DATA_FIRST_OFFSET;
+ offset = XFS_DIR3_DATA_FIRST_OFFSET(dp->i_mount);
ino = xfs_dir2_sf_get_parent_ino(sfp);
i8count = ino > XFS_DIR2_MAX_SHORT_INUM;
diff --git a/repair/dir2.c b/repair/dir2.c
index 6b7c8d7..de5e737 100644
--- a/repair/dir2.c
+++ b/repair/dir2.c
@@ -662,6 +662,7 @@ process_sf_dir2_fixi8(
*/
static void
process_sf_dir2_fixoff(
+ xfs_mount_t *mp,
xfs_dinode_t *dip)
{
int i;
@@ -671,7 +672,7 @@ process_sf_dir2_fixoff(
sfp = (xfs_dir2_sf_t *)XFS_DFORK_DPTR(dip);
sfep = xfs_dir2_sf_firstentry(&sfp->hdr);
- offset = XFS_DIR2_DATA_FIRST_OFFSET;
+ offset = XFS_DIR3_DATA_FIRST_OFFSET(mp);
for (i = 0; i < sfp->hdr.count; i++) {
xfs_dir2_sf_put_offset(sfep, offset);
@@ -725,7 +726,7 @@ process_sf_dir2(
max_size = XFS_DFORK_DSIZE(dip, mp);
num_entries = sfp->hdr.count;
ino_dir_size = be64_to_cpu(dip->di_size);
- offset = XFS_DIR2_DATA_FIRST_OFFSET;
+ offset = XFS_DIR3_DATA_FIRST_OFFSET(mp);
bad_offset = *repair = 0;
ASSERT(ino_dir_size <= max_size);
@@ -1082,7 +1083,7 @@ _("would have corrected entry offsets in directory %" PRIu64 "\n"),
do_warn(
_("corrected entry offsets in directory %" PRIu64 "\n"),
ino);
- process_sf_dir2_fixoff(dip);
+ process_sf_dir2_fixoff(mp, dip);
*dino_dirty = 1;
*repair = 1;
}
xfsprogs-crc-patchset-v2/xfs-add-crc-to-da-node 0000600 0001750 0001750 00000332557 12132011546 020221 0 ustar dave dave xfs: add CRCs to dir2/da node blocks
From: Dave Chinner <dchinner@xxxxxxxxxx>
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
db/attr.c | 4 +-
db/check.c | 8 +-
db/dir.c | 4 +-
db/dir2.c | 2 +-
include/xfs_da_btree.h | 106 +++-
libxfs/xfs_attr.c | 24 +-
libxfs/xfs_attr_leaf.c | 17 +-
libxfs/xfs_da_btree.c | 1393 +++++++++++++++++++++++++++++-------------------
libxfs/xfs_dir2_node.c | 26 +-
repair/dir.c | 87 +--
repair/dir2.c | 96 ++--
repair/phase6.c | 6 +-
12 files changed, 1070 insertions(+), 703 deletions(-)
diff --git a/db/attr.c b/db/attr.c
index 74bf411..a5087b8 100644
--- a/db/attr.c
+++ b/db/attr.c
@@ -54,7 +54,7 @@ const field_t attr_flds[] = {
FLD_COUNT, TYP_NONE },
{ "entries", FLDT_ATTR_LEAF_ENTRY, OI(LOFF(entries)),
attr_leaf_entries_count, FLD_ARRAY|FLD_COUNT, TYP_NONE },
- { "btree", FLDT_ATTR_NODE_ENTRY, OI(NOFF(btree)), attr_node_btree_count,
+ { "btree", FLDT_ATTR_NODE_ENTRY, OI(NOFF(__btree)), attr_node_btree_count,
FLD_ARRAY|FLD_COUNT, TYP_NONE },
{ "nvlist", FLDT_ATTR_LEAF_NAME, attr_leaf_nvlist_offset,
attr_leaf_nvlist_count, FLD_ARRAY|FLD_OFFSET|FLD_COUNT, TYP_NONE },
@@ -144,7 +144,7 @@ const field_t attr_node_entry_flds[] = {
const field_t attr_node_hdr_flds[] = {
{ "info", FLDT_ATTR_BLKINFO, OI(HOFF(info)), C1, 0, TYP_NONE },
{ "count", FLDT_UINT16D, OI(HOFF(count)), C1, 0, TYP_NONE },
- { "level", FLDT_UINT16D, OI(HOFF(level)), C1, 0, TYP_NONE },
+ { "level", FLDT_UINT16D, OI(HOFF(__level)), C1, 0, TYP_NONE },
{ NULL }
};
diff --git a/db/check.c b/db/check.c
index 69cfbdb..efbd0c9 100644
--- a/db/check.c
+++ b/db/check.c
@@ -3236,6 +3236,7 @@ process_leaf_node_dir_v2_int(
xfs_dir2_leaf_tail_t *ltp;
xfs_da_intnode_t *node;
int stale;
+ struct xfs_da3_icnode_hdr nodehdr;
leaf = iocur_top->data;
switch (be16_to_cpu(leaf->hdr.info.magic)) {
@@ -3284,13 +3285,12 @@ process_leaf_node_dir_v2_int(
break;
case XFS_DA_NODE_MAGIC:
node = iocur_top->data;
- if (be16_to_cpu(node->hdr.level) < 1 ||
- be16_to_cpu(node->hdr.level) >
- XFS_DA_NODE_MAXDEPTH) {
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
+ if (nodehdr.level < 1 || nodehdr.level > XFS_DA_NODE_MAXDEPTH) {
if (!sflag || v)
dbprintf(_("bad node block level %d for dir ino "
"%lld block %d\n"),
- be16_to_cpu(node->hdr.level), id->ino,
+ nodehdr.level, id->ino,
dabno);
error++;
}
diff --git a/db/dir.c b/db/dir.c
index 4b7e9dc..70ecc46 100644
--- a/db/dir.c
+++ b/db/dir.c
@@ -48,7 +48,7 @@ const field_t dir_flds[] = {
FLD_COUNT, TYP_NONE },
{ "entries", FLDT_DIR_LEAF_ENTRY, OI(LOFF(entries)),
dir_leaf_entries_count, FLD_ARRAY|FLD_COUNT, TYP_NONE },
- { "btree", FLDT_DIR_NODE_ENTRY, OI(NOFF(btree)),
+ { "btree", FLDT_DIR_NODE_ENTRY, OI(NOFF(__btree)),
dir_node_btree_count, FLD_ARRAY|FLD_COUNT, TYP_NONE },
{ "namelist", FLDT_DIR_LEAF_NAME, dir_leaf_namelist_offset,
dir_leaf_namelist_count, FLD_ARRAY|FLD_OFFSET|FLD_COUNT, TYP_NONE },
@@ -112,7 +112,7 @@ const field_t dir_node_entry_flds[] = {
const field_t dir_node_hdr_flds[] = {
{ "info", FLDT_DIR_BLKINFO, OI(HOFF(info)), C1, 0, TYP_NONE },
{ "count", FLDT_UINT16D, OI(HOFF(count)), C1, 0, TYP_NONE },
- { "level", FLDT_UINT16D, OI(HOFF(level)), C1, 0, TYP_NONE },
+ { "level", FLDT_UINT16D, OI(HOFF(__level)), C1, 0, TYP_NONE },
{ NULL }
};
diff --git a/db/dir2.c b/db/dir2.c
index abdbfee..07809b0 100644
--- a/db/dir2.c
+++ b/db/dir2.c
@@ -87,7 +87,7 @@ const field_t dir2_flds[] = {
dir2_leaf_tail_count, FLD_OFFSET|FLD_COUNT, TYP_NONE },
{ "nhdr", FLDT_DIR_NODE_HDR, OI(NOFF(hdr)), dir2_node_hdr_count,
FLD_COUNT, TYP_NONE },
- { "nbtree", FLDT_DIR_NODE_ENTRY, OI(NOFF(btree)), dir2_node_btree_count,
+ { "nbtree", FLDT_DIR_NODE_ENTRY, OI(NOFF(__btree)), dir2_node_btree_count,
FLD_ARRAY|FLD_COUNT, TYP_NONE },
{ "fhdr", FLDT_DIR2_FREE_HDR, OI(FOFF(hdr)), dir2_free_hdr_count,
FLD_COUNT, TYP_NONE },
diff --git a/include/xfs_da_btree.h b/include/xfs_da_btree.h
index 0854b95..6bedb3c 100644
--- a/include/xfs_da_btree.h
+++ b/include/xfs_da_btree.h
@@ -1,5 +1,6 @@
/*
* Copyright (c) 2000,2002,2005 Silicon Graphics, Inc.
+ * Copyright (c) 2013 Red Hat, Inc.
* All Rights Reserved.
*
* This program is free software; you can redistribute it and/or
@@ -20,7 +21,6 @@
struct xfs_bmap_free;
struct xfs_inode;
-struct xfs_mount;
struct xfs_trans;
struct zone;
@@ -50,8 +50,11 @@ typedef struct xfs_da_blkinfo {
* CRC enabled directory structure types
*
* The headers change size for the additional verification information, but
- * otherwise the tree layouts and contents are unchanged.
+ * otherwise the tree layouts and contents are unchanged. Hence the da btree
+ * code can use the struct xfs_da_blkinfo for manipulating the tree links and
+ * magic numbers without modification for both v2 and v3 nodes.
*/
+#define XFS_DA3_NODE_MAGIC 0x3ebe /* magic number: non-leaf blocks */
#define XFS_DIR3_LEAF1_MAGIC 0x3df1 /* magic number: v2 dirlf single blks */
#define XFS_DIR3_LEAFN_MAGIC 0x3dff /* magic number: v2 dirlf multi blks */
@@ -80,19 +83,76 @@ struct xfs_da3_blkinfo {
*/
#define XFS_DA_NODE_MAXDEPTH 5 /* max depth of Btree */
+typedef struct xfs_da_node_hdr {
+ struct xfs_da_blkinfo info; /* block type, links, etc. */
+ __be16 count; /* count of active entries */
+ __be16 __level; /* level above leaves (leaf == 0) */
+} xfs_da_node_hdr_t;
+
+struct xfs_da3_node_hdr {
+ struct xfs_da3_blkinfo info; /* block type, links, etc. */
+ __be16 count; /* count of active entries */
+ __be16 __level; /* level above leaves (leaf == 0) */
+ __be32 __pad32;
+};
+
+#define XFS_DA3_NODE_CRC_OFF (offsetof(struct xfs_da3_node_hdr, info.crc))
+
+typedef struct xfs_da_node_entry {
+ __be32 hashval; /* hash value for this descendant */
+ __be32 before; /* Btree block before this key */
+} xfs_da_node_entry_t;
+
typedef struct xfs_da_intnode {
- struct xfs_da_node_hdr { /* constant-structure header block */
- xfs_da_blkinfo_t info; /* block type, links, etc. */
- __be16 count; /* count of active entries */
- __be16 level; /* level above leaves (leaf == 0) */
- } hdr;
- struct xfs_da_node_entry {
- __be32 hashval; /* hash value for this descendant */
- __be32 before; /* Btree block before this key */
- } btree[1]; /* variable sized array of keys */
+ struct xfs_da_node_hdr hdr;
+ struct xfs_da_node_entry __btree[];
} xfs_da_intnode_t;
-typedef struct xfs_da_node_hdr xfs_da_node_hdr_t;
-typedef struct xfs_da_node_entry xfs_da_node_entry_t;
+
+struct xfs_da3_intnode {
+ struct xfs_da3_node_hdr hdr;
+ struct xfs_da_node_entry __btree[];
+};
+
+/*
+ * In-core version of the node header to abstract the differences in the v2 and
+ * v3 disk format of the headers. Callers need to convert to/from disk format as
+ * appropriate.
+ */
+struct xfs_da3_icnode_hdr {
+ __uint32_t forw;
+ __uint32_t back;
+ __uint16_t magic;
+ __uint16_t count;
+ __uint16_t level;
+};
+
+extern void xfs_da3_node_hdr_from_disk(struct xfs_da3_icnode_hdr *to,
+ struct xfs_da_intnode *from);
+extern void xfs_da3_node_hdr_to_disk(struct xfs_da_intnode *to,
+ struct xfs_da3_icnode_hdr *from);
+
+static inline int
+xfs_da3_node_hdr_size(struct xfs_da_intnode *dap)
+{
+ if (dap->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC))
+ return sizeof(struct xfs_da3_node_hdr);
+ return sizeof(struct xfs_da_node_hdr);
+}
+
+static inline struct xfs_da_node_entry *
+xfs_da3_node_tree_p(struct xfs_da_intnode *dap)
+{
+ if (dap->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC)) {
+ struct xfs_da3_intnode *dap3 = (struct xfs_da3_intnode *)dap;
+ return dap3->__btree;
+ }
+ return dap->__btree;
+}
+
+extern void xfs_da3_intnode_from_disk(struct xfs_da3_icnode_hdr *to,
+ struct xfs_da_intnode *from);
+extern void xfs_da3_intnode_to_disk(struct xfs_da_intnode *to,
+ struct xfs_da3_icnode_hdr *from);
#define XFS_LBSIZE(mp) (mp)->m_sb.sb_blocksize
@@ -214,29 +274,29 @@ struct xfs_nameops {
/*
* Routines used for growing the Btree.
*/
-int xfs_da_node_create(xfs_da_args_t *args, xfs_dablk_t blkno, int level,
- struct xfs_buf **bpp, int whichfork);
-int xfs_da_split(xfs_da_state_t *state);
+int xfs_da3_node_create(struct xfs_da_args *args, xfs_dablk_t blkno,
+ int level, struct xfs_buf **bpp, int whichfork);
+int xfs_da3_split(xfs_da_state_t *state);
/*
* Routines used for shrinking the Btree.
*/
-int xfs_da_join(xfs_da_state_t *state);
-void xfs_da_fixhashpath(xfs_da_state_t *state,
- xfs_da_state_path_t *path_to_to_fix);
+int xfs_da3_join(xfs_da_state_t *state);
+void xfs_da3_fixhashpath(struct xfs_da_state *state,
+ struct xfs_da_state_path *path_to_to_fix);
/*
* Routines used for finding things in the Btree.
*/
-int xfs_da_node_lookup_int(xfs_da_state_t *state, int *result);
-int xfs_da_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path,
+int xfs_da3_node_lookup_int(xfs_da_state_t *state, int *result);
+int xfs_da3_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path,
int forward, int release, int *result);
/*
* Utility routines.
*/
-int xfs_da_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
+int xfs_da3_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
xfs_da_state_blk_t *new_blk);
-int xfs_da_node_read(struct xfs_trans *tp, struct xfs_inode *dp,
+int xfs_da3_node_read(struct xfs_trans *tp, struct xfs_inode *dp,
xfs_dablk_t bno, xfs_daddr_t mappedbno,
struct xfs_buf **bpp, int which_fork);
diff --git a/libxfs/xfs_attr.c b/libxfs/xfs_attr.c
index 2adf92b..bb2ccf2 100644
--- a/libxfs/xfs_attr.c
+++ b/libxfs/xfs_attr.c
@@ -967,7 +967,7 @@ restart:
* Search to see if name already exists, and get back a pointer
* to where it should go.
*/
- error = xfs_da_node_lookup_int(state, &retval);
+ error = xfs_da3_node_lookup_int(state, &retval);
if (error)
goto out;
blk = &state->path.blk[ state->path.active-1 ];
@@ -1038,7 +1038,7 @@ restart:
* in the index2/blkno2/rmtblkno2/rmtblkcnt2 fields.
*/
xfs_bmap_init(args->flist, args->firstblock);
- error = xfs_da_split(state);
+ error = xfs_da3_split(state);
if (!error) {
error = xfs_bmap_finish(&args->trans, args->flist,
&committed);
@@ -1060,7 +1060,7 @@ restart:
/*
* Addition succeeded, update Btree hashvals.
*/
- xfs_da_fixhashpath(state, &state->path);
+ xfs_da3_fixhashpath(state, &state->path);
}
/*
@@ -1131,7 +1131,7 @@ restart:
state->blocksize = state->mp->m_sb.sb_blocksize;
state->node_ents = state->mp->m_attr_node_ents;
state->inleaf = 0;
- error = xfs_da_node_lookup_int(state, &retval);
+ error = xfs_da3_node_lookup_int(state, &retval);
if (error)
goto out;
@@ -1141,14 +1141,14 @@ restart:
blk = &state->path.blk[ state->path.active-1 ];
ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC);
error = xfs_attr_leaf_remove(blk->bp, args);
- xfs_da_fixhashpath(state, &state->path);
+ xfs_da3_fixhashpath(state, &state->path);
/*
* Check to see if the tree needs to be collapsed.
*/
if (retval && (state->path.active > 1)) {
xfs_bmap_init(args->flist, args->firstblock);
- error = xfs_da_join(state);
+ error = xfs_da3_join(state);
if (!error) {
error = xfs_bmap_finish(&args->trans,
args->flist,
@@ -1226,7 +1226,7 @@ xfs_attr_node_removename(xfs_da_args_t *args)
/*
* Search to see if name exists, and get back a pointer to it.
*/
- error = xfs_da_node_lookup_int(state, &retval);
+ error = xfs_da3_node_lookup_int(state, &retval);
if (error || (retval != EEXIST)) {
if (error == 0)
error = retval;
@@ -1277,14 +1277,14 @@ xfs_attr_node_removename(xfs_da_args_t *args)
blk = &state->path.blk[ state->path.active-1 ];
ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC);
retval = xfs_attr_leaf_remove(blk->bp, args);
- xfs_da_fixhashpath(state, &state->path);
+ xfs_da3_fixhashpath(state, &state->path);
/*
* Check to see if the tree needs to be collapsed.
*/
if (retval && (state->path.active > 1)) {
xfs_bmap_init(args->flist, args->firstblock);
- error = xfs_da_join(state);
+ error = xfs_da3_join(state);
if (!error) {
error = xfs_bmap_finish(&args->trans, args->flist,
&committed);
@@ -1430,7 +1430,7 @@ xfs_attr_refillstate(xfs_da_state_t *state)
ASSERT((path->active >= 0) && (path->active < XFS_DA_NODE_MAXDEPTH));
for (blk = path->blk, level = 0; level < path->active; blk++, level++) {
if (blk->disk_blkno) {
- error = xfs_da_node_read(state->args->trans,
+ error = xfs_da3_node_read(state->args->trans,
state->args->dp,
blk->blkno, blk->disk_blkno,
&blk->bp, XFS_ATTR_FORK);
@@ -1449,7 +1449,7 @@ xfs_attr_refillstate(xfs_da_state_t *state)
ASSERT((path->active >= 0) && (path->active < XFS_DA_NODE_MAXDEPTH));
for (blk = path->blk, level = 0; level < path->active; blk++, level++) {
if (blk->disk_blkno) {
- error = xfs_da_node_read(state->args->trans,
+ error = xfs_da3_node_read(state->args->trans,
state->args->dp,
blk->blkno, blk->disk_blkno,
&blk->bp, XFS_ATTR_FORK);
@@ -1489,7 +1489,7 @@ xfs_attr_node_get(xfs_da_args_t *args)
/*
* Search to see if name exists, and get back a pointer to it.
*/
- error = xfs_da_node_lookup_int(state, &retval);
+ error = xfs_da3_node_lookup_int(state, &retval);
if (error) {
retval = error;
} else if (retval == EEXIST) {
diff --git a/libxfs/xfs_attr_leaf.c b/libxfs/xfs_attr_leaf.c
index 85cb31d..cb37198 100644
--- a/libxfs/xfs_attr_leaf.c
+++ b/libxfs/xfs_attr_leaf.c
@@ -703,6 +703,7 @@ xfs_attr_leaf_to_node(xfs_da_args_t *args)
struct xfs_buf *bp1, *bp2;
xfs_dablk_t blkno;
int error;
+ struct xfs_da_node_entry *btree;
trace_xfs_attr_leaf_to_node(args);
@@ -728,16 +729,16 @@ xfs_attr_leaf_to_node(xfs_da_args_t *args)
/*
* Set up the new root node.
*/
- error = xfs_da_node_create(args, 0, 1, &bp1, XFS_ATTR_FORK);
+ error = xfs_da3_node_create(args, 0, 1, &bp1, XFS_ATTR_FORK);
if (error)
goto out;
node = bp1->b_addr;
leaf = bp2->b_addr;
ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
/* both on-disk, don't endian-flip twice */
- node->btree[0].hashval =
- leaf->entries[be16_to_cpu(leaf->hdr.count)-1 ].hashval;
- node->btree[0].before = cpu_to_be32(blkno);
+ btree = xfs_da3_node_tree_p(node);
+ btree[0].hashval = leaf->entries[be16_to_cpu(leaf->hdr.count)-1 ].hashval;
+ btree[0].before = cpu_to_be32(blkno);
node->hdr.count = cpu_to_be16(1);
xfs_trans_log_buf(args->trans, bp1, 0, XFS_LBSIZE(dp->i_mount) - 1);
error = 0;
@@ -825,7 +826,7 @@ xfs_attr_leaf_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
* NOTE: rebalance() currently depends on the 2nd block being empty.
*/
xfs_attr_leaf_rebalance(state, oldblk, newblk);
- error = xfs_da_blk_link(state, oldblk, newblk);
+ error = xfs_da3_blk_link(state, oldblk, newblk);
if (error)
return(error);
@@ -1453,7 +1454,7 @@ xfs_attr_leaf_toosmall(xfs_da_state_t *state, int *action)
*/
forward = (info->forw != 0);
memcpy(&state->altpath, &state->path, sizeof(state->path));
- error = xfs_da_path_shift(state, &state->altpath, forward,
+ error = xfs_da3_path_shift(state, &state->altpath, forward,
0, &retval);
if (error)
return(error);
@@ -1510,10 +1511,10 @@ xfs_attr_leaf_toosmall(xfs_da_state_t *state, int *action)
*/
memcpy(&state->altpath, &state->path, sizeof(state->path));
if (blkno < blk->blkno) {
- error = xfs_da_path_shift(state, &state->altpath, forward,
+ error = xfs_da3_path_shift(state, &state->altpath, forward,
0, &retval);
} else {
- error = xfs_da_path_shift(state, &state->path, forward,
+ error = xfs_da3_path_shift(state, &state->path, forward,
0, &retval);
}
if (error)
diff --git a/libxfs/xfs_da_btree.c b/libxfs/xfs_da_btree.c
index 63cd299..3176626 100644
--- a/libxfs/xfs_da_btree.c
+++ b/libxfs/xfs_da_btree.c
@@ -1,5 +1,6 @@
/*
* Copyright (c) 2000-2005 Silicon Graphics, Inc.
+ * Copyright (c) 2013 Red Hat, Inc.
* All Rights Reserved.
*
* This program is free software; you can redistribute it and/or
@@ -31,69 +32,195 @@
/*
* Routines used for growing the Btree.
*/
-STATIC int xfs_da_root_split(xfs_da_state_t *state,
+STATIC int xfs_da3_root_split(xfs_da_state_t *state,
xfs_da_state_blk_t *existing_root,
xfs_da_state_blk_t *new_child);
-STATIC int xfs_da_node_split(xfs_da_state_t *state,
+STATIC int xfs_da3_node_split(xfs_da_state_t *state,
xfs_da_state_blk_t *existing_blk,
xfs_da_state_blk_t *split_blk,
xfs_da_state_blk_t *blk_to_add,
int treelevel,
int *result);
-STATIC void xfs_da_node_rebalance(xfs_da_state_t *state,
+STATIC void xfs_da3_node_rebalance(xfs_da_state_t *state,
xfs_da_state_blk_t *node_blk_1,
xfs_da_state_blk_t *node_blk_2);
-STATIC void xfs_da_node_add(xfs_da_state_t *state,
+STATIC void xfs_da3_node_add(xfs_da_state_t *state,
xfs_da_state_blk_t *old_node_blk,
xfs_da_state_blk_t *new_node_blk);
/*
* Routines used for shrinking the Btree.
*/
-STATIC int xfs_da_root_join(xfs_da_state_t *state,
+STATIC int xfs_da3_root_join(xfs_da_state_t *state,
xfs_da_state_blk_t *root_blk);
-STATIC int xfs_da_node_toosmall(xfs_da_state_t *state, int *retval);
-STATIC void xfs_da_node_remove(xfs_da_state_t *state,
+STATIC int xfs_da3_node_toosmall(xfs_da_state_t *state, int *retval);
+STATIC void xfs_da3_node_remove(xfs_da_state_t *state,
xfs_da_state_blk_t *drop_blk);
-STATIC void xfs_da_node_unbalance(xfs_da_state_t *state,
+STATIC void xfs_da3_node_unbalance(xfs_da_state_t *state,
xfs_da_state_blk_t *src_node_blk,
xfs_da_state_blk_t *dst_node_blk);
/*
* Utility routines.
*/
-STATIC uint xfs_da_node_lasthash(struct xfs_buf *bp, int *count);
-STATIC int xfs_da_node_order(struct xfs_buf *node1_bp,
- struct xfs_buf *node2_bp);
-STATIC int xfs_da_blk_unlink(xfs_da_state_t *state,
+STATIC int xfs_da3_blk_unlink(xfs_da_state_t *state,
xfs_da_state_blk_t *drop_blk,
xfs_da_state_blk_t *save_blk);
-STATIC void xfs_da_state_kill_altpath(xfs_da_state_t *state);
-static void
-xfs_da_node_verify(
+
+kmem_zone_t *xfs_da_state_zone; /* anchor for state struct zone */
+
+/*
+ * Allocate a dir-state structure.
+ * We don't put them on the stack since they're large.
+ */
+xfs_da_state_t *
+xfs_da_state_alloc(void)
+{
+ return kmem_zone_zalloc(xfs_da_state_zone, KM_NOFS);
+}
+
+/*
+ * Kill the altpath contents of a da-state structure.
+ */
+STATIC void
+xfs_da_state_kill_altpath(xfs_da_state_t *state)
+{
+ int i;
+
+ for (i = 0; i < state->altpath.active; i++)
+ state->altpath.blk[i].bp = NULL;
+ state->altpath.active = 0;
+}
+
+/*
+ * Free a da-state structure.
+ */
+void
+xfs_da_state_free(xfs_da_state_t *state)
+{
+ xfs_da_state_kill_altpath(state);
+#ifdef DEBUG
+ memset((char *)state, 0, sizeof(*state));
+#endif /* DEBUG */
+ kmem_zone_free(xfs_da_state_zone, state);
+}
+
+void
+xfs_da3_node_hdr_from_disk(
+ struct xfs_da3_icnode_hdr *to,
+ struct xfs_da_intnode *from)
+{
+ ASSERT(from->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC) ||
+ from->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC));
+
+ if (from->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC)) {
+ struct xfs_da3_node_hdr *hdr3 = (struct xfs_da3_node_hdr *)from;
+
+ to->forw = be32_to_cpu(hdr3->info.hdr.forw);
+ to->back = be32_to_cpu(hdr3->info.hdr.back);
+ to->magic = be16_to_cpu(hdr3->info.hdr.magic);
+ to->count = be16_to_cpu(hdr3->count);
+ to->level = be16_to_cpu(hdr3->__level);
+ return;
+ }
+ to->forw = be32_to_cpu(from->hdr.info.forw);
+ to->back = be32_to_cpu(from->hdr.info.back);
+ to->magic = be16_to_cpu(from->hdr.info.magic);
+ to->count = be16_to_cpu(from->hdr.count);
+ to->level = be16_to_cpu(from->hdr.__level);
+}
+
+void
+xfs_da3_node_hdr_to_disk(
+ struct xfs_da_intnode *to,
+ struct xfs_da3_icnode_hdr *from)
+{
+ ASSERT(from->magic == XFS_DA_NODE_MAGIC ||
+ from->magic == XFS_DA3_NODE_MAGIC);
+
+ if (from->magic == XFS_DA3_NODE_MAGIC) {
+ struct xfs_da3_node_hdr *hdr3 = (struct xfs_da3_node_hdr *)to;
+
+ hdr3->info.hdr.forw = cpu_to_be32(from->forw);
+ hdr3->info.hdr.back = cpu_to_be32(from->back);
+ hdr3->info.hdr.magic = cpu_to_be16(from->magic);
+ hdr3->count = cpu_to_be16(from->count);
+ hdr3->__level = cpu_to_be16(from->level);
+ return;
+ }
+ to->hdr.info.forw = cpu_to_be32(from->forw);
+ to->hdr.info.back = cpu_to_be32(from->back);
+ to->hdr.info.magic = cpu_to_be16(from->magic);
+ to->hdr.count = cpu_to_be16(from->count);
+ to->hdr.__level = cpu_to_be16(from->level);
+}
+
+static bool
+xfs_da3_node_verify(
struct xfs_buf *bp)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
- struct xfs_da_node_hdr *hdr = bp->b_addr;
- int block_ok = 0;
-
- block_ok = hdr->info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC);
- block_ok = block_ok &&
- be16_to_cpu(hdr->level) > 0 &&
- be16_to_cpu(hdr->count) > 0 ;
- if (!block_ok) {
- XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, hdr);
- xfs_buf_ioerror(bp, EFSCORRUPTED);
+ struct xfs_da_intnode *hdr = bp->b_addr;
+ struct xfs_da3_icnode_hdr ichdr;
+
+ xfs_da3_node_hdr_from_disk(&ichdr, hdr);
+
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ struct xfs_da3_node_hdr *hdr3 = bp->b_addr;
+
+ if (ichdr.magic != XFS_DA3_NODE_MAGIC)
+ return false;
+
+ if (!uuid_equal(&hdr3->info.uuid, &mp->m_sb.sb_uuid))
+ return false;
+ if (be64_to_cpu(hdr3->info.blkno) != bp->b_bn)
+ return false;
+ } else {
+ if (ichdr.magic != XFS_DA_NODE_MAGIC)
+ return false;
}
+ if (ichdr.level == 0)
+ return false;
+ if (ichdr.level > XFS_DA_NODE_MAXDEPTH)
+ return false;
+ if (ichdr.count == 0)
+ return false;
+
+ /*
+ * we don't know if the node is for and attribute or directory tree,
+ * so only fail if the count is outside both bounds
+ */
+ if (ichdr.count > mp->m_dir_node_ents &&
+ ichdr.count > mp->m_attr_node_ents)
+ return false;
+
+ /* XXX: hash order check? */
+ return true;
}
static void
-xfs_da_node_write_verify(
+xfs_da3_node_write_verify(
struct xfs_buf *bp)
{
- xfs_da_node_verify(bp);
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_buf_log_item *bip = bp->b_fspriv;
+ struct xfs_da3_node_hdr *hdr3 = bp->b_addr;
+
+ if (!xfs_da3_node_verify(bp)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ return;
+ }
+
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return;
+
+ if (bip)
+ hdr3->info.lsn = cpu_to_be64(bip->bli_item.li_lsn);
+
+ xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), XFS_DA3_NODE_CRC_OFF);
}
/*
@@ -103,16 +230,22 @@ xfs_da_node_write_verify(
* format of the block being read.
*/
static void
-xfs_da_node_read_verify(
+xfs_da3_node_read_verify(
struct xfs_buf *bp)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
struct xfs_da_blkinfo *info = bp->b_addr;
switch (be16_to_cpu(info->magic)) {
+ case XFS_DA3_NODE_MAGIC:
+ if (!xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
+ XFS_DA3_NODE_CRC_OFF))
+ break;
+ /* fall through */
case XFS_DA_NODE_MAGIC:
- xfs_da_node_verify(bp);
- break;
+ if (!xfs_da3_node_verify(bp))
+ break;
+ return;
case XFS_ATTR_LEAF_MAGIC:
bp->b_ops = &xfs_attr_leaf_buf_ops;
bp->b_ops->verify_read(bp);
@@ -123,21 +256,22 @@ xfs_da_node_read_verify(
bp->b_ops->verify_read(bp);
return;
default:
- XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
- mp, info);
- xfs_buf_ioerror(bp, EFSCORRUPTED);
break;
}
+
+ /* corrupt block */
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
}
-const struct xfs_buf_ops xfs_da_node_buf_ops = {
- .verify_read = xfs_da_node_read_verify,
- .verify_write = xfs_da_node_write_verify,
+const struct xfs_buf_ops xfs_da3_node_buf_ops = {
+ .verify_read = xfs_da3_node_read_verify,
+ .verify_write = xfs_da3_node_write_verify,
};
int
-xfs_da_node_read(
+xfs_da3_node_read(
struct xfs_trans *tp,
struct xfs_inode *dp,
xfs_dablk_t bno,
@@ -146,7 +280,7 @@ xfs_da_node_read(
int which_fork)
{
return xfs_da_read_buf(tp, dp, bno, mappedbno, bpp,
- which_fork, &xfs_da_node_buf_ops);
+ which_fork, &xfs_da3_node_buf_ops);
}
/*========================================================================
@@ -157,33 +291,45 @@ xfs_da_node_read(
* Create the initial contents of an intermediate node.
*/
int
-xfs_da_node_create(xfs_da_args_t *args, xfs_dablk_t blkno, int level,
- struct xfs_buf **bpp, int whichfork)
+xfs_da3_node_create(
+ struct xfs_da_args *args,
+ xfs_dablk_t blkno,
+ int level,
+ struct xfs_buf **bpp,
+ int whichfork)
{
- xfs_da_intnode_t *node;
- struct xfs_buf *bp;
- int error;
- xfs_trans_t *tp;
+ struct xfs_da_intnode *node;
+ struct xfs_trans *tp = args->trans;
+ struct xfs_mount *mp = tp->t_mountp;
+ struct xfs_da3_icnode_hdr ichdr = {0};
+ struct xfs_buf *bp;
+ int error;
trace_xfs_da_node_create(args);
+ ASSERT(level <= XFS_DA_NODE_MAXDEPTH);
- tp = args->trans;
error = xfs_da_get_buf(tp, args->dp, blkno, -1, &bp, whichfork);
if (error)
return(error);
- ASSERT(bp != NULL);
node = bp->b_addr;
- node->hdr.info.forw = 0;
- node->hdr.info.back = 0;
- node->hdr.info.magic = cpu_to_be16(XFS_DA_NODE_MAGIC);
- node->hdr.info.pad = 0;
- node->hdr.count = 0;
- node->hdr.level = cpu_to_be16(level);
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ struct xfs_da3_node_hdr *hdr3 = bp->b_addr;
+
+ ichdr.magic = XFS_DA3_NODE_MAGIC;
+ hdr3->info.blkno = cpu_to_be64(bp->b_bn);
+ hdr3->info.owner = cpu_to_be64(args->dp->i_ino);
+ uuid_copy(&hdr3->info.uuid, &mp->m_sb.sb_uuid);
+ } else {
+ ichdr.magic = XFS_DA_NODE_MAGIC;
+ }
+ ichdr.level = level;
+
+ xfs_da3_node_hdr_to_disk(node, &ichdr);
xfs_trans_log_buf(tp, bp,
- XFS_DA_LOGRANGE(node, &node->hdr, sizeof(node->hdr)));
+ XFS_DA_LOGRANGE(node, &node->hdr, xfs_da3_node_hdr_size(node)));
- bp->b_ops = &xfs_da_node_buf_ops;
+ bp->b_ops = &xfs_da3_node_buf_ops;
*bpp = bp;
return(0);
}
@@ -193,12 +339,18 @@ xfs_da_node_create(xfs_da_args_t *args, xfs_dablk_t blkno, int level,
* intermediate nodes, rebalance, etc.
*/
int /* error */
-xfs_da_split(xfs_da_state_t *state)
+xfs_da3_split(
+ struct xfs_da_state *state)
{
- xfs_da_state_blk_t *oldblk, *newblk, *addblk;
- xfs_da_intnode_t *node;
- struct xfs_buf *bp;
- int max, action, error, i;
+ struct xfs_da_state_blk *oldblk;
+ struct xfs_da_state_blk *newblk;
+ struct xfs_da_state_blk *addblk;
+ struct xfs_da_intnode *node;
+ struct xfs_buf *bp;
+ int max;
+ int action;
+ int error;
+ int i;
trace_xfs_da_split(state->args);
@@ -260,7 +412,7 @@ xfs_da_split(xfs_da_state_t *state)
addblk = newblk;
break;
case XFS_DA_NODE_MAGIC:
- error = xfs_da_node_split(state, oldblk, newblk, addblk,
+ error = xfs_da3_node_split(state, oldblk, newblk, addblk,
max - i, &action);
addblk->bp = NULL;
if (error)
@@ -278,7 +430,7 @@ xfs_da_split(xfs_da_state_t *state)
/*
* Update the btree to show the new hashval for this child.
*/
- xfs_da_fixhashpath(state, &state->path);
+ xfs_da3_fixhashpath(state, &state->path);
}
if (!addblk)
return(0);
@@ -288,7 +440,7 @@ xfs_da_split(xfs_da_state_t *state)
*/
ASSERT(state->path.active == 0);
oldblk = &state->path.blk[0];
- error = xfs_da_root_split(state, oldblk, addblk);
+ error = xfs_da3_root_split(state, oldblk, addblk);
if (error) {
addblk->bp = NULL;
return(error); /* GROT: dir is inconsistent */
@@ -299,8 +451,10 @@ xfs_da_split(xfs_da_state_t *state)
* just got bumped because of the addition of a new root node.
* There might be three blocks involved if a double split occurred,
* and the original block 0 could be at any position in the list.
+ *
+ * Note: the info structures being modified here for both v2 and v3 da
+ * headers, so we can do this linkage just using the v2 structures.
*/
-
node = oldblk->bp->b_addr;
if (node->hdr.info.forw) {
if (be32_to_cpu(node->hdr.info.forw) == addblk->blkno) {
@@ -339,18 +493,25 @@ xfs_da_split(xfs_da_state_t *state)
* the EOF, extending the inode in process.
*/
STATIC int /* error */
-xfs_da_root_split(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
- xfs_da_state_blk_t *blk2)
+xfs_da3_root_split(
+ struct xfs_da_state *state,
+ struct xfs_da_state_blk *blk1,
+ struct xfs_da_state_blk *blk2)
{
- xfs_da_intnode_t *node, *oldroot;
- xfs_da_args_t *args;
- xfs_dablk_t blkno;
- struct xfs_buf *bp;
- int error, size;
- xfs_inode_t *dp;
- xfs_trans_t *tp;
- xfs_mount_t *mp;
- xfs_dir2_leaf_t *leaf;
+ struct xfs_da_intnode *node;
+ struct xfs_da_intnode *oldroot;
+ struct xfs_da_node_entry *btree;
+ struct xfs_da3_icnode_hdr nodehdr;
+ struct xfs_da_args *args;
+ struct xfs_buf *bp;
+ struct xfs_inode *dp;
+ struct xfs_trans *tp;
+ struct xfs_mount *mp;
+ struct xfs_dir2_leaf *leaf;
+ xfs_dablk_t blkno;
+ int level;
+ int error;
+ int size;
trace_xfs_da_root_split(state->args);
@@ -359,22 +520,26 @@ xfs_da_root_split(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
* to a free space somewhere.
*/
args = state->args;
- ASSERT(args != NULL);
error = xfs_da_grow_inode(args, &blkno);
if (error)
- return(error);
+ return error;
+
dp = args->dp;
tp = args->trans;
mp = state->mp;
error = xfs_da_get_buf(tp, dp, blkno, -1, &bp, args->whichfork);
if (error)
- return(error);
- ASSERT(bp != NULL);
+ return error;
node = bp->b_addr;
oldroot = blk1->bp->b_addr;
- if (oldroot->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC)) {
- size = (int)((char *)&oldroot->btree[be16_to_cpu(oldroot->hdr.count)] -
- (char *)oldroot);
+ if (oldroot->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC) ||
+ oldroot->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC)) {
+ struct xfs_da3_icnode_hdr nodehdr;
+
+ xfs_da3_node_hdr_from_disk(&nodehdr, oldroot);
+ btree = xfs_da3_node_tree_p(oldroot);
+ size = (int)((char *)&btree[nodehdr.count] - (char *)oldroot);
+ level = nodehdr.level;
} else {
struct xfs_dir3_icleaf_hdr leafhdr;
struct xfs_dir2_leaf_entry *ents;
@@ -386,9 +551,22 @@ xfs_da_root_split(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
ASSERT(leafhdr.magic == XFS_DIR2_LEAFN_MAGIC ||
leafhdr.magic == XFS_DIR3_LEAFN_MAGIC);
size = (int)((char *)&ents[leafhdr.count] - (char *)leaf);
+ level = 0;
}
- /* XXX: can't just copy CRC headers from one block to another */
+
+ /*
+ * we can copy most of the information in the node from one block to
+ * another, but for CRC enabled headers we have to make sure that the
+ * block specific identifiers are kept intact. We update the buffer
+ * directly for this.
+ */
memcpy(node, oldroot, size);
+ if (oldroot->hdr.info.magic == cpu_to_be16(XFS_DA3_NODE_MAGIC) ||
+ oldroot->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC)) {
+ struct xfs_da3_intnode *node3 = (struct xfs_da3_intnode *)node;
+
+ node3->hdr.info.blkno = cpu_to_be64(bp->b_bn);
+ }
xfs_trans_log_buf(tp, bp, 0, size - 1);
bp->b_ops = blk1->bp->b_ops;
@@ -398,17 +576,21 @@ xfs_da_root_split(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
/*
* Set up the new root node.
*/
- error = xfs_da_node_create(args,
+ error = xfs_da3_node_create(args,
(args->whichfork == XFS_DATA_FORK) ? mp->m_dirleafblk : 0,
- be16_to_cpu(node->hdr.level) + 1, &bp, args->whichfork);
+ level + 1, &bp, args->whichfork);
if (error)
- return(error);
+ return error;
+
node = bp->b_addr;
- node->btree[0].hashval = cpu_to_be32(blk1->hashval);
- node->btree[0].before = cpu_to_be32(blk1->blkno);
- node->btree[1].hashval = cpu_to_be32(blk2->hashval);
- node->btree[1].before = cpu_to_be32(blk2->blkno);
- node->hdr.count = cpu_to_be16(2);
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
+ btree = xfs_da3_node_tree_p(node);
+ btree[0].hashval = cpu_to_be32(blk1->hashval);
+ btree[0].before = cpu_to_be32(blk1->blkno);
+ btree[1].hashval = cpu_to_be32(blk2->hashval);
+ btree[1].before = cpu_to_be32(blk2->blkno);
+ nodehdr.count = 2;
+ xfs_da3_node_hdr_to_disk(node, &nodehdr);
#ifdef DEBUG
if (oldroot->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
@@ -422,30 +604,34 @@ xfs_da_root_split(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
/* Header is already logged by xfs_da_node_create */
xfs_trans_log_buf(tp, bp,
- XFS_DA_LOGRANGE(node, node->btree,
- sizeof(xfs_da_node_entry_t) * 2));
+ XFS_DA_LOGRANGE(node, btree, sizeof(xfs_da_node_entry_t) * 2));
- return(0);
+ return 0;
}
/*
* Split the node, rebalance, then add the new entry.
*/
STATIC int /* error */
-xfs_da_node_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
- xfs_da_state_blk_t *newblk,
- xfs_da_state_blk_t *addblk,
- int treelevel, int *result)
+xfs_da3_node_split(
+ struct xfs_da_state *state,
+ struct xfs_da_state_blk *oldblk,
+ struct xfs_da_state_blk *newblk,
+ struct xfs_da_state_blk *addblk,
+ int treelevel,
+ int *result)
{
- xfs_da_intnode_t *node;
- xfs_dablk_t blkno;
- int newcount, error;
- int useextra;
+ struct xfs_da_intnode *node;
+ struct xfs_da3_icnode_hdr nodehdr;
+ xfs_dablk_t blkno;
+ int newcount;
+ int error;
+ int useextra;
trace_xfs_da_node_split(state->args);
node = oldblk->bp->b_addr;
- ASSERT(node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
/*
* With V2 dirs the extra block is data or freespace.
@@ -455,7 +641,7 @@ xfs_da_node_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
/*
* Do we have to split the node?
*/
- if ((be16_to_cpu(node->hdr.count) + newcount) > state->node_ents) {
+ if (nodehdr.count + newcount > state->node_ents) {
/*
* Allocate a new node, add to the doubly linked chain of
* nodes, then move some of our excess entries into it.
@@ -464,14 +650,14 @@ xfs_da_node_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
if (error)
return(error); /* GROT: dir is inconsistent */
- error = xfs_da_node_create(state->args, blkno, treelevel,
+ error = xfs_da3_node_create(state->args, blkno, treelevel,
&newblk->bp, state->args->whichfork);
if (error)
return(error); /* GROT: dir is inconsistent */
newblk->blkno = blkno;
newblk->magic = XFS_DA_NODE_MAGIC;
- xfs_da_node_rebalance(state, oldblk, newblk);
- error = xfs_da_blk_link(state, oldblk, newblk);
+ xfs_da3_node_rebalance(state, oldblk, newblk);
+ error = xfs_da3_blk_link(state, oldblk, newblk);
if (error)
return(error);
*result = 1;
@@ -483,7 +669,7 @@ xfs_da_node_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
* Insert the new entry(s) into the correct block
* (updating last hashval in the process).
*
- * xfs_da_node_add() inserts BEFORE the given index,
+ * xfs_da3_node_add() inserts BEFORE the given index,
* and as a result of using node_lookup_int() we always
* point to a valid entry (not after one), but a split
* operation always results in a new block whose hashvals
@@ -492,22 +678,23 @@ xfs_da_node_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
* If we had double-split op below us, then add the extra block too.
*/
node = oldblk->bp->b_addr;
- if (oldblk->index <= be16_to_cpu(node->hdr.count)) {
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
+ if (oldblk->index <= nodehdr.count) {
oldblk->index++;
- xfs_da_node_add(state, oldblk, addblk);
+ xfs_da3_node_add(state, oldblk, addblk);
if (useextra) {
if (state->extraafter)
oldblk->index++;
- xfs_da_node_add(state, oldblk, &state->extrablk);
+ xfs_da3_node_add(state, oldblk, &state->extrablk);
state->extravalid = 0;
}
} else {
newblk->index++;
- xfs_da_node_add(state, newblk, addblk);
+ xfs_da3_node_add(state, newblk, addblk);
if (useextra) {
if (state->extraafter)
newblk->index++;
- xfs_da_node_add(state, newblk, &state->extrablk);
+ xfs_da3_node_add(state, newblk, &state->extrablk);
state->extravalid = 0;
}
}
@@ -522,33 +709,53 @@ xfs_da_node_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
* NOTE: if blk2 is empty, then it will get the upper half of blk1.
*/
STATIC void
-xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
- xfs_da_state_blk_t *blk2)
+xfs_da3_node_rebalance(
+ struct xfs_da_state *state,
+ struct xfs_da_state_blk *blk1,
+ struct xfs_da_state_blk *blk2)
{
- xfs_da_intnode_t *node1, *node2, *tmpnode;
- xfs_da_node_entry_t *btree_s, *btree_d;
- int count, tmp;
- xfs_trans_t *tp;
+ struct xfs_da_intnode *node1;
+ struct xfs_da_intnode *node2;
+ struct xfs_da_intnode *tmpnode;
+ struct xfs_da_node_entry *btree1;
+ struct xfs_da_node_entry *btree2;
+ struct xfs_da_node_entry *btree_s;
+ struct xfs_da_node_entry *btree_d;
+ struct xfs_da3_icnode_hdr nodehdr1;
+ struct xfs_da3_icnode_hdr nodehdr2;
+ struct xfs_trans *tp;
+ int count;
+ int tmp;
+ int swap = 0;
trace_xfs_da_node_rebalance(state->args);
node1 = blk1->bp->b_addr;
node2 = blk2->bp->b_addr;
+ xfs_da3_node_hdr_from_disk(&nodehdr1, node1);
+ xfs_da3_node_hdr_from_disk(&nodehdr2, node2);
+ btree1 = xfs_da3_node_tree_p(node1);
+ btree2 = xfs_da3_node_tree_p(node2);
+
/*
* Figure out how many entries need to move, and in which direction.
* Swap the nodes around if that makes it simpler.
*/
- if ((be16_to_cpu(node1->hdr.count) > 0) && (be16_to_cpu(node2->hdr.count) > 0) &&
- ((be32_to_cpu(node2->btree[0].hashval) < be32_to_cpu(node1->btree[0].hashval)) ||
- (be32_to_cpu(node2->btree[be16_to_cpu(node2->hdr.count)-1].hashval) <
- be32_to_cpu(node1->btree[be16_to_cpu(node1->hdr.count)-1].hashval)))) {
+ if (nodehdr1.count > 0 && nodehdr2.count > 0 &&
+ ((be32_to_cpu(btree2[0].hashval) < be32_to_cpu(btree1[0].hashval)) ||
+ (be32_to_cpu(btree2[nodehdr2.count - 1].hashval) <
+ be32_to_cpu(btree1[nodehdr1.count - 1].hashval)))) {
tmpnode = node1;
node1 = node2;
node2 = tmpnode;
+ xfs_da3_node_hdr_from_disk(&nodehdr1, node1);
+ xfs_da3_node_hdr_from_disk(&nodehdr2, node2);
+ btree1 = xfs_da3_node_tree_p(node1);
+ btree2 = xfs_da3_node_tree_p(node2);
+ swap = 1;
}
- ASSERT(node1->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
- ASSERT(node2->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
- count = (be16_to_cpu(node1->hdr.count) - be16_to_cpu(node2->hdr.count)) / 2;
+
+ count = (nodehdr1.count - nodehdr2.count) / 2;
if (count == 0)
return;
tp = state->args->trans;
@@ -559,10 +766,11 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
/*
* Move elements in node2 up to make a hole.
*/
- if ((tmp = be16_to_cpu(node2->hdr.count)) > 0) {
+ tmp = nodehdr2.count;
+ if (tmp > 0) {
tmp *= (uint)sizeof(xfs_da_node_entry_t);
- btree_s = &node2->btree[0];
- btree_d = &node2->btree[count];
+ btree_s = &btree2[0];
+ btree_d = &btree2[count];
memmove(btree_d, btree_s, tmp);
}
@@ -570,12 +778,12 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
* Move the req'd B-tree elements from high in node1 to
* low in node2.
*/
- be16_add_cpu(&node2->hdr.count, count);
+ nodehdr2.count += count;
tmp = count * (uint)sizeof(xfs_da_node_entry_t);
- btree_s = &node1->btree[be16_to_cpu(node1->hdr.count) - count];
- btree_d = &node2->btree[0];
+ btree_s = &btree1[nodehdr1.count- count];
+ btree_d = &btree2[0];
memcpy(btree_d, btree_s, tmp);
- be16_add_cpu(&node1->hdr.count, -count);
+ nodehdr1.count -= count;
} else {
/*
* Move the req'd B-tree elements from low in node2 to
@@ -583,49 +791,60 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
*/
count = -count;
tmp = count * (uint)sizeof(xfs_da_node_entry_t);
- btree_s = &node2->btree[0];
- btree_d = &node1->btree[be16_to_cpu(node1->hdr.count)];
+ btree_s = &btree2[0];
+ btree_d = &btree1[nodehdr1.count];
memcpy(btree_d, btree_s, tmp);
- be16_add_cpu(&node1->hdr.count, count);
+ nodehdr1.count += count;
+
xfs_trans_log_buf(tp, blk1->bp,
XFS_DA_LOGRANGE(node1, btree_d, tmp));
/*
* Move elements in node2 down to fill the hole.
*/
- tmp = be16_to_cpu(node2->hdr.count) - count;
+ tmp = nodehdr2.count - count;
tmp *= (uint)sizeof(xfs_da_node_entry_t);
- btree_s = &node2->btree[count];
- btree_d = &node2->btree[0];
+ btree_s = &btree2[count];
+ btree_d = &btree2[0];
memmove(btree_d, btree_s, tmp);
- be16_add_cpu(&node2->hdr.count, -count);
+ nodehdr2.count -= count;
}
/*
* Log header of node 1 and all current bits of node 2.
*/
+ xfs_da3_node_hdr_to_disk(node1, &nodehdr1);
xfs_trans_log_buf(tp, blk1->bp,
- XFS_DA_LOGRANGE(node1, &node1->hdr, sizeof(node1->hdr)));
+ XFS_DA_LOGRANGE(node1, &node1->hdr,
+ xfs_da3_node_hdr_size(node1)));
+
+ xfs_da3_node_hdr_to_disk(node2, &nodehdr2);
xfs_trans_log_buf(tp, blk2->bp,
XFS_DA_LOGRANGE(node2, &node2->hdr,
- sizeof(node2->hdr) +
- sizeof(node2->btree[0]) * be16_to_cpu(node2->hdr.count)));
+ xfs_da3_node_hdr_size(node2) +
+ (sizeof(btree2[0]) * nodehdr2.count)));
/*
* Record the last hashval from each block for upward propagation.
* (note: don't use the swapped node pointers)
*/
- node1 = blk1->bp->b_addr;
- node2 = blk2->bp->b_addr;
- blk1->hashval = be32_to_cpu(node1->btree[be16_to_cpu(node1->hdr.count)-1].hashval);
- blk2->hashval = be32_to_cpu(node2->btree[be16_to_cpu(node2->hdr.count)-1].hashval);
+ if (swap) {
+ node1 = blk1->bp->b_addr;
+ node2 = blk2->bp->b_addr;
+ xfs_da3_node_hdr_from_disk(&nodehdr1, node1);
+ xfs_da3_node_hdr_from_disk(&nodehdr2, node2);
+ btree1 = xfs_da3_node_tree_p(node1);
+ btree2 = xfs_da3_node_tree_p(node2);
+ }
+ blk1->hashval = be32_to_cpu(btree1[nodehdr1.count - 1].hashval);
+ blk2->hashval = be32_to_cpu(btree2[nodehdr2.count - 1].hashval);
/*
* Adjust the expected index for insertion.
*/
- if (blk1->index >= be16_to_cpu(node1->hdr.count)) {
- blk2->index = blk1->index - be16_to_cpu(node1->hdr.count);
- blk1->index = be16_to_cpu(node1->hdr.count) + 1; /* make it invalid */
+ if (blk1->index >= nodehdr1.count) {
+ blk2->index = blk1->index - nodehdr1.count;
+ blk1->index = nodehdr1.count + 1; /* make it invalid */
}
}
@@ -633,18 +852,23 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
* Add a new entry to an intermediate node.
*/
STATIC void
-xfs_da_node_add(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
- xfs_da_state_blk_t *newblk)
+xfs_da3_node_add(
+ struct xfs_da_state *state,
+ struct xfs_da_state_blk *oldblk,
+ struct xfs_da_state_blk *newblk)
{
- xfs_da_intnode_t *node;
- xfs_da_node_entry_t *btree;
- int tmp;
+ struct xfs_da_intnode *node;
+ struct xfs_da3_icnode_hdr nodehdr;
+ struct xfs_da_node_entry *btree;
+ int tmp;
trace_xfs_da_node_add(state->args);
node = oldblk->bp->b_addr;
- ASSERT(node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
- ASSERT((oldblk->index >= 0) && (oldblk->index <= be16_to_cpu(node->hdr.count)));
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
+ btree = xfs_da3_node_tree_p(node);
+
+ ASSERT(oldblk->index >= 0 && oldblk->index <= nodehdr.count);
ASSERT(newblk->blkno != 0);
if (state->args->whichfork == XFS_DATA_FORK)
ASSERT(newblk->blkno >= state->mp->m_dirleafblk &&
@@ -654,23 +878,25 @@ xfs_da_node_add(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
* We may need to make some room before we insert the new node.
*/
tmp = 0;
- btree = &node->btree[ oldblk->index ];
- if (oldblk->index < be16_to_cpu(node->hdr.count)) {
- tmp = (be16_to_cpu(node->hdr.count) - oldblk->index) * (uint)sizeof(*btree);
- memmove(btree + 1, btree, tmp);
+ if (oldblk->index < nodehdr.count) {
+ tmp = (nodehdr.count - oldblk->index) * (uint)sizeof(*btree);
+ memmove(&btree[oldblk->index + 1], &btree[oldblk->index], tmp);
}
- btree->hashval = cpu_to_be32(newblk->hashval);
- btree->before = cpu_to_be32(newblk->blkno);
+ btree[oldblk->index].hashval = cpu_to_be32(newblk->hashval);
+ btree[oldblk->index].before = cpu_to_be32(newblk->blkno);
xfs_trans_log_buf(state->args->trans, oldblk->bp,
- XFS_DA_LOGRANGE(node, btree, tmp + sizeof(*btree)));
- be16_add_cpu(&node->hdr.count, 1);
+ XFS_DA_LOGRANGE(node, &btree[oldblk->index],
+ tmp + sizeof(*btree)));
+
+ nodehdr.count += 1;
+ xfs_da3_node_hdr_to_disk(node, &nodehdr);
xfs_trans_log_buf(state->args->trans, oldblk->bp,
- XFS_DA_LOGRANGE(node, &node->hdr, sizeof(node->hdr)));
+ XFS_DA_LOGRANGE(node, &node->hdr, xfs_da3_node_hdr_size(node)));
/*
* Copy the last hash value from the oldblk to propagate upwards.
*/
- oldblk->hashval = be32_to_cpu(node->btree[be16_to_cpu(node->hdr.count)-1 ].hashval);
+ oldblk->hashval = be32_to_cpu(btree[nodehdr.count - 1].hashval);
}
/*========================================================================
@@ -682,14 +908,16 @@ xfs_da_node_add(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
* possibly deallocating that block, etc...
*/
int
-xfs_da_join(xfs_da_state_t *state)
+xfs_da3_join(
+ struct xfs_da_state *state)
{
- xfs_da_state_blk_t *drop_blk, *save_blk;
- int action, error;
+ struct xfs_da_state_blk *drop_blk;
+ struct xfs_da_state_blk *save_blk;
+ int action = 0;
+ int error;
trace_xfs_da_join(state->args);
- action = 0;
drop_blk = &state->path.blk[ state->path.active-1 ];
save_blk = &state->altpath.blk[ state->path.active-1 ];
ASSERT(state->path.blk[0].magic == XFS_DA_NODE_MAGIC);
@@ -730,18 +958,18 @@ xfs_da_join(xfs_da_state_t *state)
* Remove the offending node, fixup hashvals,
* check for a toosmall neighbor.
*/
- xfs_da_node_remove(state, drop_blk);
- xfs_da_fixhashpath(state, &state->path);
- error = xfs_da_node_toosmall(state, &action);
+ xfs_da3_node_remove(state, drop_blk);
+ xfs_da3_fixhashpath(state, &state->path);
+ error = xfs_da3_node_toosmall(state, &action);
if (error)
return(error);
if (action == 0)
return 0;
- xfs_da_node_unbalance(state, drop_blk, save_blk);
+ xfs_da3_node_unbalance(state, drop_blk, save_blk);
break;
}
- xfs_da_fixhashpath(state, &state->altpath);
- error = xfs_da_blk_unlink(state, drop_blk, save_blk);
+ xfs_da3_fixhashpath(state, &state->altpath);
+ error = xfs_da3_blk_unlink(state, drop_blk, save_blk);
xfs_da_state_kill_altpath(state);
if (error)
return(error);
@@ -756,9 +984,9 @@ xfs_da_join(xfs_da_state_t *state)
* we only have one entry in the root, make the child block
* the new root.
*/
- xfs_da_node_remove(state, drop_blk);
- xfs_da_fixhashpath(state, &state->path);
- error = xfs_da_root_join(state, &state->path.blk[0]);
+ xfs_da3_node_remove(state, drop_blk);
+ xfs_da3_fixhashpath(state, &state->path);
+ error = xfs_da3_root_join(state, &state->path.blk[0]);
return(error);
}
@@ -772,8 +1000,10 @@ xfs_da_blkinfo_onlychild_validate(struct xfs_da_blkinfo *blkinfo, __u16 level)
ASSERT(magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC) ||
magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
- } else
- ASSERT(magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
+ } else {
+ ASSERT(magic == cpu_to_be16(XFS_DA_NODE_MAGIC) ||
+ magic == cpu_to_be16(XFS_DA3_NODE_MAGIC));
+ }
ASSERT(!blkinfo->forw);
ASSERT(!blkinfo->back);
}
@@ -786,52 +1016,60 @@ xfs_da_blkinfo_onlychild_validate(struct xfs_da_blkinfo *blkinfo, __u16 level)
* the old root to block 0 as the new root node.
*/
STATIC int
-xfs_da_root_join(xfs_da_state_t *state, xfs_da_state_blk_t *root_blk)
+xfs_da3_root_join(
+ struct xfs_da_state *state,
+ struct xfs_da_state_blk *root_blk)
{
- xfs_da_intnode_t *oldroot;
- xfs_da_args_t *args;
- xfs_dablk_t child;
- struct xfs_buf *bp;
- int error;
+ struct xfs_da_intnode *oldroot;
+ struct xfs_da_args *args;
+ xfs_dablk_t child;
+ struct xfs_buf *bp;
+ struct xfs_da3_icnode_hdr oldroothdr;
+ struct xfs_da_node_entry *btree;
+ int error;
trace_xfs_da_root_join(state->args);
- args = state->args;
- ASSERT(args != NULL);
ASSERT(root_blk->magic == XFS_DA_NODE_MAGIC);
+
+ args = state->args;
oldroot = root_blk->bp->b_addr;
- ASSERT(oldroot->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
- ASSERT(!oldroot->hdr.info.forw);
- ASSERT(!oldroot->hdr.info.back);
+ xfs_da3_node_hdr_from_disk(&oldroothdr, oldroot);
+ ASSERT(oldroothdr.forw == 0);
+ ASSERT(oldroothdr.back == 0);
/*
* If the root has more than one child, then don't do anything.
*/
- if (be16_to_cpu(oldroot->hdr.count) > 1)
- return(0);
+ if (oldroothdr.count > 1)
+ return 0;
/*
* Read in the (only) child block, then copy those bytes into
* the root block's buffer and free the original child block.
*/
- child = be32_to_cpu(oldroot->btree[0].before);
+ btree = xfs_da3_node_tree_p(oldroot);
+ child = be32_to_cpu(btree[0].before);
ASSERT(child != 0);
- error = xfs_da_node_read(args->trans, args->dp, child, -1, &bp,
+ error = xfs_da3_node_read(args->trans, args->dp, child, -1, &bp,
args->whichfork);
if (error)
- return(error);
- ASSERT(bp != NULL);
- xfs_da_blkinfo_onlychild_validate(bp->b_addr,
- be16_to_cpu(oldroot->hdr.level));
+ return error;
+ xfs_da_blkinfo_onlychild_validate(bp->b_addr, oldroothdr.level);
/*
* This could be copying a leaf back into the root block in the case of
* there only being a single leaf block left in the tree. Hence we have
* to update the b_ops pointer as well to match the buffer type change
- * that could occur.
+ * that could occur. For dir3 blocks we also need to update the block
+ * number in the buffer header.
*/
memcpy(root_blk->bp->b_addr, bp->b_addr, state->blocksize);
root_blk->bp->b_ops = bp->b_ops;
+ if (oldroothdr.magic == XFS_DA3_NODE_MAGIC) {
+ struct xfs_da3_blkinfo *da3 = root_blk->bp->b_addr;
+ da3->blkno = cpu_to_be64(root_blk->bp->b_bn);
+ }
xfs_trans_log_buf(args->trans, root_blk->bp, 0, state->blocksize - 1);
error = xfs_da_shrink_inode(args, child, bp);
return(error);
@@ -847,14 +1085,21 @@ xfs_da_root_join(xfs_da_state_t *state, xfs_da_state_blk_t *root_blk)
* If nothing can be done, return 0.
*/
STATIC int
-xfs_da_node_toosmall(xfs_da_state_t *state, int *action)
+xfs_da3_node_toosmall(
+ struct xfs_da_state *state,
+ int *action)
{
- xfs_da_intnode_t *node;
- xfs_da_state_blk_t *blk;
- xfs_da_blkinfo_t *info;
- int count, forward, error, retval, i;
- xfs_dablk_t blkno;
- struct xfs_buf *bp;
+ struct xfs_da_intnode *node;
+ struct xfs_da_state_blk *blk;
+ struct xfs_da_blkinfo *info;
+ xfs_dablk_t blkno;
+ struct xfs_buf *bp;
+ struct xfs_da3_icnode_hdr nodehdr;
+ int count;
+ int forward;
+ int error;
+ int retval;
+ int i;
trace_xfs_da_node_toosmall(state->args);
@@ -865,10 +1110,9 @@ xfs_da_node_toosmall(xfs_da_state_t *state, int *action)
*/
blk = &state->path.blk[ state->path.active-1 ];
info = blk->bp->b_addr;
- ASSERT(info->magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
node = (xfs_da_intnode_t *)info;
- count = be16_to_cpu(node->hdr.count);
- if (count > (state->node_ents >> 1)) {
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
+ if (nodehdr.count > (state->node_ents >> 1)) {
*action = 0; /* blk over 50%, don't try to join */
return(0); /* blk over 50%, don't try to join */
}
@@ -879,14 +1123,14 @@ xfs_da_node_toosmall(xfs_da_state_t *state, int *action)
* coalesce it with a sibling block. We choose (arbitrarily)
* to merge with the forward block unless it is NULL.
*/
- if (count == 0) {
+ if (nodehdr.count == 0) {
/*
* Make altpath point to the block we want to keep and
* path point to the block we want to drop (this one).
*/
forward = (info->forw != 0);
memcpy(&state->altpath, &state->path, sizeof(state->path));
- error = xfs_da_path_shift(state, &state->altpath, forward,
+ error = xfs_da3_path_shift(state, &state->altpath, forward,
0, &retval);
if (error)
return(error);
@@ -905,35 +1149,34 @@ xfs_da_node_toosmall(xfs_da_state_t *state, int *action)
* We prefer coalescing with the lower numbered sibling so as
* to shrink a directory over time.
*/
+ count = state->node_ents;
+ count -= state->node_ents >> 2;
+ count -= nodehdr.count;
+
/* start with smaller blk num */
- forward = (be32_to_cpu(info->forw) < be32_to_cpu(info->back));
+ forward = nodehdr.forw < nodehdr.back;
for (i = 0; i < 2; forward = !forward, i++) {
if (forward)
- blkno = be32_to_cpu(info->forw);
+ blkno = nodehdr.forw;
else
- blkno = be32_to_cpu(info->back);
+ blkno = nodehdr.back;
if (blkno == 0)
continue;
- error = xfs_da_node_read(state->args->trans, state->args->dp,
+ error = xfs_da3_node_read(state->args->trans, state->args->dp,
blkno, -1, &bp, state->args->whichfork);
if (error)
return(error);
- ASSERT(bp != NULL);
- node = (xfs_da_intnode_t *)info;
- count = state->node_ents;
- count -= state->node_ents >> 2;
- count -= be16_to_cpu(node->hdr.count);
node = bp->b_addr;
- ASSERT(node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
- count -= be16_to_cpu(node->hdr.count);
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
xfs_trans_brelse(state->args->trans, bp);
- if (count >= 0)
+
+ if (count - nodehdr.count >= 0)
break; /* fits with at least 25% to spare */
}
if (i >= 2) {
*action = 0;
- return(0);
+ return 0;
}
/*
@@ -942,28 +1185,42 @@ xfs_da_node_toosmall(xfs_da_state_t *state, int *action)
*/
memcpy(&state->altpath, &state->path, sizeof(state->path));
if (blkno < blk->blkno) {
- error = xfs_da_path_shift(state, &state->altpath, forward,
+ error = xfs_da3_path_shift(state, &state->altpath, forward,
0, &retval);
- if (error) {
- return(error);
- }
- if (retval) {
- *action = 0;
- return(0);
- }
} else {
- error = xfs_da_path_shift(state, &state->path, forward,
+ error = xfs_da3_path_shift(state, &state->path, forward,
0, &retval);
- if (error) {
- return(error);
- }
- if (retval) {
- *action = 0;
- return(0);
- }
+ }
+ if (error)
+ return error;
+ if (retval) {
+ *action = 0;
+ return 0;
}
*action = 1;
- return(0);
+ return 0;
+}
+
+/*
+ * Pick up the last hashvalue from an intermediate node.
+ */
+STATIC uint
+xfs_da3_node_lasthash(
+ struct xfs_buf *bp,
+ int *count)
+{
+ struct xfs_da_intnode *node;
+ struct xfs_da_node_entry *btree;
+ struct xfs_da3_icnode_hdr nodehdr;
+
+ node = bp->b_addr;
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
+ if (count)
+ *count = nodehdr.count;
+ if (!nodehdr.count)
+ return 0;
+ btree = xfs_da3_node_tree_p(node);
+ return be32_to_cpu(btree[nodehdr.count - 1].hashval);
}
/*
@@ -971,13 +1228,16 @@ xfs_da_node_toosmall(xfs_da_state_t *state, int *action)
* when we stop making changes, return.
*/
void
-xfs_da_fixhashpath(xfs_da_state_t *state, xfs_da_state_path_t *path)
+xfs_da3_fixhashpath(
+ struct xfs_da_state *state,
+ struct xfs_da_state_path *path)
{
- xfs_da_state_blk_t *blk;
- xfs_da_intnode_t *node;
- xfs_da_node_entry_t *btree;
- xfs_dahash_t lasthash=0;
- int level, count;
+ struct xfs_da_state_blk *blk;
+ struct xfs_da_intnode *node;
+ struct xfs_da_node_entry *btree;
+ xfs_dahash_t lasthash=0;
+ int level;
+ int count;
trace_xfs_da_fixhashpath(state->args);
@@ -995,23 +1255,26 @@ xfs_da_fixhashpath(xfs_da_state_t *state, xfs_da_state_path_t *path)
return;
break;
case XFS_DA_NODE_MAGIC:
- lasthash = xfs_da_node_lasthash(blk->bp, &count);
+ lasthash = xfs_da3_node_lasthash(blk->bp, &count);
if (count == 0)
return;
break;
}
for (blk--, level--; level >= 0; blk--, level--) {
+ struct xfs_da3_icnode_hdr nodehdr;
+
node = blk->bp->b_addr;
- ASSERT(node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
- btree = &node->btree[ blk->index ];
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
+ btree = xfs_da3_node_tree_p(node);
if (be32_to_cpu(btree->hashval) == lasthash)
break;
blk->hashval = lasthash;
- btree->hashval = cpu_to_be32(lasthash);
+ btree[blk->index].hashval = cpu_to_be32(lasthash);
xfs_trans_log_buf(state->args->trans, blk->bp,
- XFS_DA_LOGRANGE(node, btree, sizeof(*btree)));
+ XFS_DA_LOGRANGE(node, &btree[blk->index],
+ sizeof(*btree)));
- lasthash = be32_to_cpu(node->btree[be16_to_cpu(node->hdr.count)-1].hashval);
+ lasthash = be32_to_cpu(btree[nodehdr.count - 1].hashval);
}
}
@@ -1019,104 +1282,119 @@ xfs_da_fixhashpath(xfs_da_state_t *state, xfs_da_state_path_t *path)
* Remove an entry from an intermediate node.
*/
STATIC void
-xfs_da_node_remove(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk)
+xfs_da3_node_remove(
+ struct xfs_da_state *state,
+ struct xfs_da_state_blk *drop_blk)
{
- xfs_da_intnode_t *node;
- xfs_da_node_entry_t *btree;
- int tmp;
+ struct xfs_da_intnode *node;
+ struct xfs_da3_icnode_hdr nodehdr;
+ struct xfs_da_node_entry *btree;
+ int index;
+ int tmp;
trace_xfs_da_node_remove(state->args);
node = drop_blk->bp->b_addr;
- ASSERT(drop_blk->index < be16_to_cpu(node->hdr.count));
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
+ ASSERT(drop_blk->index < nodehdr.count);
ASSERT(drop_blk->index >= 0);
/*
* Copy over the offending entry, or just zero it out.
*/
- btree = &node->btree[drop_blk->index];
- if (drop_blk->index < (be16_to_cpu(node->hdr.count)-1)) {
- tmp = be16_to_cpu(node->hdr.count) - drop_blk->index - 1;
+ index = drop_blk->index;
+ btree = xfs_da3_node_tree_p(node);
+ if (index < nodehdr.count - 1) {
+ tmp = nodehdr.count - index - 1;
tmp *= (uint)sizeof(xfs_da_node_entry_t);
- memmove(btree, btree + 1, tmp);
+ memmove(&btree[index], &btree[index + 1], tmp);
xfs_trans_log_buf(state->args->trans, drop_blk->bp,
- XFS_DA_LOGRANGE(node, btree, tmp));
- btree = &node->btree[be16_to_cpu(node->hdr.count)-1];
+ XFS_DA_LOGRANGE(node, &btree[index], tmp));
+ index = nodehdr.count - 1;
}
- memset((char *)btree, 0, sizeof(xfs_da_node_entry_t));
+ memset(&btree[index], 0, sizeof(xfs_da_node_entry_t));
xfs_trans_log_buf(state->args->trans, drop_blk->bp,
- XFS_DA_LOGRANGE(node, btree, sizeof(*btree)));
- be16_add_cpu(&node->hdr.count, -1);
+ XFS_DA_LOGRANGE(node, &btree[index], sizeof(btree[index])));
+ nodehdr.count -= 1;
+ xfs_da3_node_hdr_to_disk(node, &nodehdr);
xfs_trans_log_buf(state->args->trans, drop_blk->bp,
- XFS_DA_LOGRANGE(node, &node->hdr, sizeof(node->hdr)));
+ XFS_DA_LOGRANGE(node, &node->hdr, xfs_da3_node_hdr_size(node)));
/*
* Copy the last hash value from the block to propagate upwards.
*/
- btree--;
- drop_blk->hashval = be32_to_cpu(btree->hashval);
+ drop_blk->hashval = be32_to_cpu(btree[index - 1].hashval);
}
/*
- * Unbalance the btree elements between two intermediate nodes,
+ * Unbalance the elements between two intermediate nodes,
* move all Btree elements from one node into another.
*/
STATIC void
-xfs_da_node_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
- xfs_da_state_blk_t *save_blk)
+xfs_da3_node_unbalance(
+ struct xfs_da_state *state,
+ struct xfs_da_state_blk *drop_blk,
+ struct xfs_da_state_blk *save_blk)
{
- xfs_da_intnode_t *drop_node, *save_node;
- xfs_da_node_entry_t *btree;
- int tmp;
- xfs_trans_t *tp;
+ struct xfs_da_intnode *drop_node;
+ struct xfs_da_intnode *save_node;
+ struct xfs_da_node_entry *dbtree;
+ struct xfs_da_node_entry *sbtree;
+ struct xfs_da3_icnode_hdr dhdr;
+ struct xfs_da3_icnode_hdr shdr;
+ struct xfs_trans *tp;
+ int sindex;
+ int tmp;
trace_xfs_da_node_unbalance(state->args);
drop_node = drop_blk->bp->b_addr;
save_node = save_blk->bp->b_addr;
- ASSERT(drop_node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
- ASSERT(save_node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
+ xfs_da3_node_hdr_from_disk(&dhdr, drop_node);
+ xfs_da3_node_hdr_from_disk(&shdr, save_node);
+ dbtree = xfs_da3_node_tree_p(drop_node);
+ sbtree = xfs_da3_node_tree_p(save_node);
tp = state->args->trans;
/*
* If the dying block has lower hashvals, then move all the
* elements in the remaining block up to make a hole.
*/
- if ((be32_to_cpu(drop_node->btree[0].hashval) < be32_to_cpu(save_node->btree[ 0 ].hashval)) ||
- (be32_to_cpu(drop_node->btree[be16_to_cpu(drop_node->hdr.count)-1].hashval) <
- be32_to_cpu(save_node->btree[be16_to_cpu(save_node->hdr.count)-1].hashval)))
- {
- btree = &save_node->btree[be16_to_cpu(drop_node->hdr.count)];
- tmp = be16_to_cpu(save_node->hdr.count) * (uint)sizeof(xfs_da_node_entry_t);
- memmove(btree, &save_node->btree[0], tmp);
- btree = &save_node->btree[0];
+ if ((be32_to_cpu(dbtree[0].hashval) < be32_to_cpu(sbtree[ 0 ].hashval)) ||
+ (be32_to_cpu(dbtree[dhdr.count - 1].hashval) <
+ be32_to_cpu(sbtree[shdr.count - 1].hashval))) {
+ /* XXX: check this - is memmove dst correct? */
+ tmp = shdr.count * (uint)sizeof(xfs_da_node_entry_t);
+ memmove(&sbtree[dhdr.count], &sbtree[0], tmp);
+
+ sindex = 0;
xfs_trans_log_buf(tp, save_blk->bp,
- XFS_DA_LOGRANGE(save_node, btree,
- (be16_to_cpu(save_node->hdr.count) + be16_to_cpu(drop_node->hdr.count)) *
- sizeof(xfs_da_node_entry_t)));
+ XFS_DA_LOGRANGE(save_node, &sbtree[0],
+ (shdr.count + dhdr.count) *
+ sizeof(xfs_da_node_entry_t)));
} else {
- btree = &save_node->btree[be16_to_cpu(save_node->hdr.count)];
+ sindex = shdr.count;
xfs_trans_log_buf(tp, save_blk->bp,
- XFS_DA_LOGRANGE(save_node, btree,
- be16_to_cpu(drop_node->hdr.count) *
- sizeof(xfs_da_node_entry_t)));
+ XFS_DA_LOGRANGE(save_node, &sbtree[sindex],
+ dhdr.count * sizeof(xfs_da_node_entry_t)));
}
/*
* Move all the B-tree elements from drop_blk to save_blk.
*/
- tmp = be16_to_cpu(drop_node->hdr.count) * (uint)sizeof(xfs_da_node_entry_t);
- memcpy(btree, &drop_node->btree[0], tmp);
- be16_add_cpu(&save_node->hdr.count, be16_to_cpu(drop_node->hdr.count));
+ tmp = dhdr.count * (uint)sizeof(xfs_da_node_entry_t);
+ memcpy(&sbtree[sindex], &dbtree[0], tmp);
+ shdr.count += dhdr.count;
+ xfs_da3_node_hdr_to_disk(save_node, &shdr);
xfs_trans_log_buf(tp, save_blk->bp,
XFS_DA_LOGRANGE(save_node, &save_node->hdr,
- sizeof(save_node->hdr)));
+ xfs_da3_node_hdr_size(save_node)));
/*
* Save the last hashval in the remaining block for upward propagation.
*/
- save_blk->hashval = be32_to_cpu(save_node->btree[be16_to_cpu(save_node->hdr.count)-1].hashval);
+ save_blk->hashval = be32_to_cpu(sbtree[shdr.count - 1].hashval);
}
/*========================================================================
@@ -1135,16 +1413,24 @@ xfs_da_node_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
* pruned depth-first tree search.
*/
int /* error */
-xfs_da_node_lookup_int(xfs_da_state_t *state, int *result)
+xfs_da3_node_lookup_int(
+ struct xfs_da_state *state,
+ int *result)
{
- xfs_da_state_blk_t *blk;
- xfs_da_blkinfo_t *curr;
- xfs_da_intnode_t *node;
- xfs_da_node_entry_t *btree;
- xfs_dablk_t blkno;
- int probe, span, max, error, retval;
- xfs_dahash_t hashval, btreehashval;
- xfs_da_args_t *args;
+ struct xfs_da_state_blk *blk;
+ struct xfs_da_blkinfo *curr;
+ struct xfs_da_intnode *node;
+ struct xfs_da_node_entry *btree;
+ struct xfs_da3_icnode_hdr nodehdr;
+ struct xfs_da_args *args;
+ xfs_dablk_t blkno;
+ xfs_dahash_t hashval;
+ xfs_dahash_t btreehashval;
+ int probe;
+ int span;
+ int max;
+ int error;
+ int retval;
args = state->args;
@@ -1160,7 +1446,7 @@ xfs_da_node_lookup_int(xfs_da_state_t *state, int *result)
* Read the next node down in the tree.
*/
blk->blkno = blkno;
- error = xfs_da_node_read(args->trans, args->dp, blkno,
+ error = xfs_da3_node_read(args->trans, args->dp, blkno,
-1, &blk->bp, args->whichfork);
if (error) {
blk->blkno = 0;
@@ -1169,66 +1455,73 @@ xfs_da_node_lookup_int(xfs_da_state_t *state, int *result)
}
curr = blk->bp->b_addr;
blk->magic = be16_to_cpu(curr->magic);
- ASSERT(blk->magic == XFS_DA_NODE_MAGIC ||
- blk->magic == XFS_DIR2_LEAFN_MAGIC ||
- blk->magic == XFS_ATTR_LEAF_MAGIC);
+
+ if (blk->magic == XFS_ATTR_LEAF_MAGIC) {
+ blk->hashval = xfs_attr_leaf_lasthash(blk->bp, NULL);
+ break;
+ }
+
+ if (blk->magic == XFS_DIR2_LEAFN_MAGIC ||
+ blk->magic == XFS_DIR3_LEAFN_MAGIC) {
+ blk->magic = XFS_DIR2_LEAFN_MAGIC;
+ blk->hashval = xfs_dir2_leafn_lasthash(blk->bp, NULL);
+ break;
+ }
+
+ blk->magic = XFS_DA_NODE_MAGIC;
+
/*
* Search an intermediate node for a match.
*/
- if (blk->magic == XFS_DA_NODE_MAGIC) {
- node = blk->bp->b_addr;
- max = be16_to_cpu(node->hdr.count);
- blk->hashval = be32_to_cpu(node->btree[max-1].hashval);
+ node = blk->bp->b_addr;
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
+ btree = xfs_da3_node_tree_p(node);
- /*
- * Binary search. (note: small blocks will skip loop)
- */
- probe = span = max / 2;
- hashval = args->hashval;
- for (btree = &node->btree[probe]; span > 4;
- btree = &node->btree[probe]) {
- span /= 2;
- btreehashval = be32_to_cpu(btree->hashval);
- if (btreehashval < hashval)
- probe += span;
- else if (btreehashval > hashval)
- probe -= span;
- else
- break;
- }
- ASSERT((probe >= 0) && (probe < max));
- ASSERT((span <= 4) || (be32_to_cpu(btree->hashval) == hashval));
+ max = nodehdr.count;
+ blk->hashval = be32_to_cpu(btree[max - 1].hashval);
- /*
- * Since we may have duplicate hashval's, find the first
- * matching hashval in the node.
- */
- while ((probe > 0) && (be32_to_cpu(btree->hashval) >= hashval)) {
- btree--;
- probe--;
- }
- while ((probe < max) && (be32_to_cpu(btree->hashval) < hashval)) {
- btree++;
- probe++;
- }
+ /*
+ * Binary search. (note: small blocks will skip loop)
+ */
+ probe = span = max / 2;
+ hashval = args->hashval;
+ while (span > 4) {
+ span /= 2;
+ btreehashval = be32_to_cpu(btree[probe].hashval);
+ if (btreehashval < hashval)
+ probe += span;
+ else if (btreehashval > hashval)
+ probe -= span;
+ else
+ break;
+ }
+ ASSERT((probe >= 0) && (probe < max));
+ ASSERT((span <= 4) ||
+ (be32_to_cpu(btree[probe].hashval) == hashval));
- /*
- * Pick the right block to descend on.
- */
- if (probe == max) {
- blk->index = max-1;
- blkno = be32_to_cpu(node->btree[max-1].before);
- } else {
- blk->index = probe;
- blkno = be32_to_cpu(btree->before);
- }
- } else if (blk->magic == XFS_ATTR_LEAF_MAGIC) {
- blk->hashval = xfs_attr_leaf_lasthash(blk->bp, NULL);
- break;
- } else if (blk->magic == XFS_DIR2_LEAFN_MAGIC) {
- blk->hashval = xfs_dir2_leafn_lasthash(blk->bp, NULL);
- break;
+ /*
+ * Since we may have duplicate hashval's, find the first
+ * matching hashval in the node.
+ */
+ while (probe > 0 &&
+ be32_to_cpu(btree[probe].hashval) >= hashval) {
+ probe--;
+ }
+ while (probe < max &&
+ be32_to_cpu(btree[probe].hashval) < hashval) {
+ probe++;
+ }
+
+ /*
+ * Pick the right block to descend on.
+ */
+ if (probe == max) {
+ blk->index = max - 1;
+ blkno = be32_to_cpu(btree[max - 1].before);
+ } else {
+ blk->index = probe;
+ blkno = be32_to_cpu(btree[probe].before);
}
}
@@ -1252,7 +1545,7 @@ xfs_da_node_lookup_int(xfs_da_state_t *state, int *result)
}
if (((retval == ENOENT) || (retval == ENOATTR)) &&
(blk->hashval == args->hashval)) {
- error = xfs_da_path_shift(state, &state->path, 1, 1,
+ error = xfs_da3_path_shift(state, &state->path, 1, 1,
&retval);
if (error)
return(error);
@@ -1274,16 +1567,52 @@ xfs_da_node_lookup_int(xfs_da_state_t *state, int *result)
*========================================================================*/
/*
+ * Compare two intermediate nodes for "order".
+ */
+STATIC int
+xfs_da3_node_order(
+ struct xfs_buf *node1_bp,
+ struct xfs_buf *node2_bp)
+{
+ struct xfs_da_intnode *node1;
+ struct xfs_da_intnode *node2;
+ struct xfs_da_node_entry *btree1;
+ struct xfs_da_node_entry *btree2;
+ struct xfs_da3_icnode_hdr node1hdr;
+ struct xfs_da3_icnode_hdr node2hdr;
+
+ node1 = node1_bp->b_addr;
+ node2 = node2_bp->b_addr;
+ xfs_da3_node_hdr_from_disk(&node1hdr, node1);
+ xfs_da3_node_hdr_from_disk(&node2hdr, node2);
+ btree1 = xfs_da3_node_tree_p(node1);
+ btree2 = xfs_da3_node_tree_p(node2);
+
+ if (node1hdr.count > 0 && node2hdr.count > 0 &&
+ ((be32_to_cpu(btree2[0].hashval) < be32_to_cpu(btree1[0].hashval)) ||
+ (be32_to_cpu(btree2[node2hdr.count - 1].hashval) <
+ be32_to_cpu(btree1[node1hdr.count - 1].hashval)))) {
+ return 1;
+ }
+ return 0;
+}
+
+/*
* Link a new block into a doubly linked list of blocks (of whatever type).
*/
int /* error */
-xfs_da_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
- xfs_da_state_blk_t *new_blk)
+xfs_da3_blk_link(
+ struct xfs_da_state *state,
+ struct xfs_da_state_blk *old_blk,
+ struct xfs_da_state_blk *new_blk)
{
- xfs_da_blkinfo_t *old_info, *new_info, *tmp_info;
- xfs_da_args_t *args;
- int before=0, error;
- struct xfs_buf *bp;
+ struct xfs_da_blkinfo *old_info;
+ struct xfs_da_blkinfo *new_info;
+ struct xfs_da_blkinfo *tmp_info;
+ struct xfs_da_args *args;
+ struct xfs_buf *bp;
+ int before = 0;
+ int error;
/*
* Set up environment.
@@ -1295,9 +1624,6 @@ xfs_da_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
ASSERT(old_blk->magic == XFS_DA_NODE_MAGIC ||
old_blk->magic == XFS_DIR2_LEAFN_MAGIC ||
old_blk->magic == XFS_ATTR_LEAF_MAGIC);
- ASSERT(old_blk->magic == be16_to_cpu(old_info->magic));
- ASSERT(new_blk->magic == be16_to_cpu(new_info->magic));
- ASSERT(old_blk->magic == new_blk->magic);
switch (old_blk->magic) {
case XFS_ATTR_LEAF_MAGIC:
@@ -1307,7 +1633,7 @@ xfs_da_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
before = xfs_dir2_leafn_order(old_blk->bp, new_blk->bp);
break;
case XFS_DA_NODE_MAGIC:
- before = xfs_da_node_order(old_blk->bp, new_blk->bp);
+ before = xfs_da3_node_order(old_blk->bp, new_blk->bp);
break;
}
@@ -1322,14 +1648,14 @@ xfs_da_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
new_info->forw = cpu_to_be32(old_blk->blkno);
new_info->back = old_info->back;
if (old_info->back) {
- error = xfs_da_node_read(args->trans, args->dp,
+ error = xfs_da3_node_read(args->trans, args->dp,
be32_to_cpu(old_info->back),
-1, &bp, args->whichfork);
if (error)
return(error);
ASSERT(bp != NULL);
tmp_info = bp->b_addr;
- ASSERT(be16_to_cpu(tmp_info->magic) == be16_to_cpu(old_info->magic));
+ ASSERT(tmp_info->magic == old_info->magic);
ASSERT(be32_to_cpu(tmp_info->forw) == old_blk->blkno);
tmp_info->forw = cpu_to_be32(new_blk->blkno);
xfs_trans_log_buf(args->trans, bp, 0, sizeof(*tmp_info)-1);
@@ -1343,7 +1669,7 @@ xfs_da_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
new_info->forw = old_info->forw;
new_info->back = cpu_to_be32(old_blk->blkno);
if (old_info->forw) {
- error = xfs_da_node_read(args->trans, args->dp,
+ error = xfs_da3_node_read(args->trans, args->dp,
be32_to_cpu(old_info->forw),
-1, &bp, args->whichfork);
if (error)
@@ -1364,59 +1690,20 @@ xfs_da_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
}
/*
- * Compare two intermediate nodes for "order".
- */
-STATIC int
-xfs_da_node_order(
- struct xfs_buf *node1_bp,
- struct xfs_buf *node2_bp)
-{
- xfs_da_intnode_t *node1, *node2;
-
- node1 = node1_bp->b_addr;
- node2 = node2_bp->b_addr;
- ASSERT(node1->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC) &&
- node2->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
- if ((be16_to_cpu(node1->hdr.count) > 0) && (be16_to_cpu(node2->hdr.count) > 0) &&
- ((be32_to_cpu(node2->btree[0].hashval) <
- be32_to_cpu(node1->btree[0].hashval)) ||
- (be32_to_cpu(node2->btree[be16_to_cpu(node2->hdr.count)-1].hashval) <
- be32_to_cpu(node1->btree[be16_to_cpu(node1->hdr.count)-1].hashval)))) {
- return(1);
- }
- return(0);
-}
-
-/*
- * Pick up the last hashvalue from an intermediate node.
- */
-STATIC uint
-xfs_da_node_lasthash(
- struct xfs_buf *bp,
- int *count)
-{
- xfs_da_intnode_t *node;
-
- node = bp->b_addr;
- ASSERT(node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
- if (count)
- *count = be16_to_cpu(node->hdr.count);
- if (!node->hdr.count)
- return(0);
- return be32_to_cpu(node->btree[be16_to_cpu(node->hdr.count)-1].hashval);
-}
-
-/*
* Unlink a block from a doubly linked list of blocks.
*/
STATIC int /* error */
-xfs_da_blk_unlink(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
- xfs_da_state_blk_t *save_blk)
+xfs_da3_blk_unlink(
+ struct xfs_da_state *state,
+ struct xfs_da_state_blk *drop_blk,
+ struct xfs_da_state_blk *save_blk)
{
- xfs_da_blkinfo_t *drop_info, *save_info, *tmp_info;
- xfs_da_args_t *args;
- struct xfs_buf *bp;
- int error;
+ struct xfs_da_blkinfo *drop_info;
+ struct xfs_da_blkinfo *save_info;
+ struct xfs_da_blkinfo *tmp_info;
+ struct xfs_da_args *args;
+ struct xfs_buf *bp;
+ int error;
/*
* Set up environment.
@@ -1428,8 +1715,6 @@ xfs_da_blk_unlink(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
ASSERT(save_blk->magic == XFS_DA_NODE_MAGIC ||
save_blk->magic == XFS_DIR2_LEAFN_MAGIC ||
save_blk->magic == XFS_ATTR_LEAF_MAGIC);
- ASSERT(save_blk->magic == be16_to_cpu(save_info->magic));
- ASSERT(drop_blk->magic == be16_to_cpu(drop_info->magic));
ASSERT(save_blk->magic == drop_blk->magic);
ASSERT((be32_to_cpu(save_info->forw) == drop_blk->blkno) ||
(be32_to_cpu(save_info->back) == drop_blk->blkno));
@@ -1443,7 +1728,7 @@ xfs_da_blk_unlink(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
trace_xfs_da_unlink_back(args);
save_info->back = drop_info->back;
if (drop_info->back) {
- error = xfs_da_node_read(args->trans, args->dp,
+ error = xfs_da3_node_read(args->trans, args->dp,
be32_to_cpu(drop_info->back),
-1, &bp, args->whichfork);
if (error)
@@ -1460,7 +1745,7 @@ xfs_da_blk_unlink(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
trace_xfs_da_unlink_forward(args);
save_info->forw = drop_info->forw;
if (drop_info->forw) {
- error = xfs_da_node_read(args->trans, args->dp,
+ error = xfs_da3_node_read(args->trans, args->dp,
be32_to_cpu(drop_info->forw),
-1, &bp, args->whichfork);
if (error)
@@ -1488,15 +1773,22 @@ xfs_da_blk_unlink(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
* the new bottom and the root.
*/
int /* error */
-xfs_da_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path,
- int forward, int release, int *result)
+xfs_da3_path_shift(
+ struct xfs_da_state *state,
+ struct xfs_da_state_path *path,
+ int forward,
+ int release,
+ int *result)
{
- xfs_da_state_blk_t *blk;
- xfs_da_blkinfo_t *info;
- xfs_da_intnode_t *node;
- xfs_da_args_t *args;
- xfs_dablk_t blkno=0;
- int level, error;
+ struct xfs_da_state_blk *blk;
+ struct xfs_da_blkinfo *info;
+ struct xfs_da_intnode *node;
+ struct xfs_da_args *args;
+ struct xfs_da_node_entry *btree;
+ struct xfs_da3_icnode_hdr nodehdr;
+ xfs_dablk_t blkno = 0;
+ int level;
+ int error;
trace_xfs_da_path_shift(state->args);
@@ -1511,16 +1803,17 @@ xfs_da_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path,
ASSERT((path->active > 0) && (path->active < XFS_DA_NODE_MAXDEPTH));
level = (path->active-1) - 1; /* skip bottom layer in path */
for (blk = &path->blk[level]; level >= 0; blk--, level--) {
- ASSERT(blk->bp != NULL);
node = blk->bp->b_addr;
- ASSERT(node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
- if (forward && (blk->index < be16_to_cpu(node->hdr.count)-1)) {
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
+ btree = xfs_da3_node_tree_p(node);
+
+ if (forward && (blk->index < nodehdr.count - 1)) {
blk->index++;
- blkno = be32_to_cpu(node->btree[blk->index].before);
+ blkno = be32_to_cpu(btree[blk->index].before);
break;
} else if (!forward && (blk->index > 0)) {
blk->index--;
- blkno = be32_to_cpu(node->btree[blk->index].before);
+ blkno = be32_to_cpu(btree[blk->index].before);
break;
}
}
@@ -1546,47 +1839,58 @@ xfs_da_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path,
* Read the next child block.
*/
blk->blkno = blkno;
- error = xfs_da_node_read(args->trans, args->dp, blkno, -1,
+ error = xfs_da3_node_read(args->trans, args->dp, blkno, -1,
&blk->bp, args->whichfork);
if (error)
return(error);
- ASSERT(blk->bp != NULL);
info = blk->bp->b_addr;
ASSERT(info->magic == cpu_to_be16(XFS_DA_NODE_MAGIC) ||
+ info->magic == cpu_to_be16(XFS_DA3_NODE_MAGIC) ||
info->magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
info->magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC) ||
info->magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
- blk->magic = be16_to_cpu(info->magic);
- if (blk->magic == XFS_DA_NODE_MAGIC) {
+
+
+ /*
+ * Note: we flatten the magic number to a single type so we
+ * don't have to compare against crc/non-crc types elsewhere.
+ */
+ switch (be16_to_cpu(info->magic)) {
+ case XFS_DA_NODE_MAGIC:
+ case XFS_DA3_NODE_MAGIC:
+ blk->magic = XFS_DA_NODE_MAGIC;
node = (xfs_da_intnode_t *)info;
- blk->hashval = be32_to_cpu(node->btree[be16_to_cpu(node->hdr.count)-1].hashval);
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
+ btree = xfs_da3_node_tree_p(node);
+ blk->hashval = be32_to_cpu(btree[nodehdr.count - 1].hashval);
if (forward)
blk->index = 0;
else
- blk->index = be16_to_cpu(node->hdr.count)-1;
- blkno = be32_to_cpu(node->btree[blk->index].before);
- } else {
+ blk->index = nodehdr.count - 1;
+ blkno = be32_to_cpu(btree[blk->index].before);
+ break;
+ case XFS_ATTR_LEAF_MAGIC:
+ blk->magic = XFS_ATTR_LEAF_MAGIC;
ASSERT(level == path->active-1);
blk->index = 0;
- switch(blk->magic) {
- case XFS_ATTR_LEAF_MAGIC:
- blk->hashval = xfs_attr_leaf_lasthash(blk->bp,
- NULL);
- break;
- case XFS_DIR2_LEAFN_MAGIC:
- case XFS_DIR3_LEAFN_MAGIC:
- blk->magic = XFS_DIR2_LEAFN_MAGIC;
- blk->hashval = xfs_dir2_leafn_lasthash(blk->bp,
- NULL);
- break;
- default:
- ASSERT(0);
- break;
- }
+ blk->hashval = xfs_attr_leaf_lasthash(blk->bp,
+ NULL);
+ break;
+ case XFS_DIR2_LEAFN_MAGIC:
+ case XFS_DIR3_LEAFN_MAGIC:
+ blk->magic = XFS_DIR2_LEAFN_MAGIC;
+ ASSERT(level == path->active-1);
+ blk->index = 0;
+ blk->hashval = xfs_dir2_leafn_lasthash(blk->bp,
+ NULL);
+ break;
+ default:
+ ASSERT(0);
+ break;
}
}
*result = 0;
- return(0);
+ return 0;
}
@@ -1773,22 +2077,36 @@ xfs_da_grow_inode(
* a bmap btree split to do that.
*/
STATIC int
-xfs_da_swap_lastblock(
- xfs_da_args_t *args,
- xfs_dablk_t *dead_blknop,
- struct xfs_buf **dead_bufp)
+xfs_da3_swap_lastblock(
+ struct xfs_da_args *args,
+ xfs_dablk_t *dead_blknop,
+ struct xfs_buf **dead_bufp)
{
- xfs_dablk_t dead_blkno, last_blkno, sib_blkno, par_blkno;
- struct xfs_buf *dead_buf, *last_buf, *sib_buf, *par_buf;
- xfs_fileoff_t lastoff;
- xfs_inode_t *ip;
- xfs_trans_t *tp;
- xfs_mount_t *mp;
- int error, w, entno, level, dead_level;
- xfs_da_blkinfo_t *dead_info, *sib_info;
- xfs_da_intnode_t *par_node, *dead_node;
- xfs_dir2_leaf_t *dead_leaf2;
- xfs_dahash_t dead_hash;
+ struct xfs_da_blkinfo *dead_info;
+ struct xfs_da_blkinfo *sib_info;
+ struct xfs_da_intnode *par_node;
+ struct xfs_da_intnode *dead_node;
+ struct xfs_dir2_leaf *dead_leaf2;
+ struct xfs_da_node_entry *btree;
+ struct xfs_da3_icnode_hdr par_hdr;
+ struct xfs_inode *ip;
+ struct xfs_trans *tp;
+ struct xfs_mount *mp;
+ struct xfs_buf *dead_buf;
+ struct xfs_buf *last_buf;
+ struct xfs_buf *sib_buf;
+ struct xfs_buf *par_buf;
+ xfs_dahash_t dead_hash;
+ xfs_fileoff_t lastoff;
+ xfs_dablk_t dead_blkno;
+ xfs_dablk_t last_blkno;
+ xfs_dablk_t sib_blkno;
+ xfs_dablk_t par_blkno;
+ int error;
+ int w;
+ int entno;
+ int level;
+ int dead_level;
trace_xfs_da_swap_lastblock(args);
@@ -1812,7 +2130,7 @@ xfs_da_swap_lastblock(
* Read the last block in the btree space.
*/
last_blkno = (xfs_dablk_t)lastoff - mp->m_dirblkfsbs;
- error = xfs_da_node_read(tp, ip, last_blkno, -1, &last_buf, w);
+ error = xfs_da3_node_read(tp, ip, last_blkno, -1, &last_buf, w);
if (error)
return error;
/*
@@ -1835,17 +2153,22 @@ xfs_da_swap_lastblock(
dead_level = 0;
dead_hash = be32_to_cpu(ents[leafhdr.count - 1].hashval);
} else {
- ASSERT(dead_info->magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
+ struct xfs_da3_icnode_hdr deadhdr;
+
+ ASSERT(dead_info->magic == cpu_to_be16(XFS_DA_NODE_MAGIC) ||
+ dead_info->magic == cpu_to_be16(XFS_DA3_NODE_MAGIC));
dead_node = (xfs_da_intnode_t *)dead_info;
- dead_level = be16_to_cpu(dead_node->hdr.level);
- dead_hash = be32_to_cpu(dead_node->btree[be16_to_cpu(dead_node->hdr.count) - 1].hashval);
+ xfs_da3_node_hdr_from_disk(&deadhdr, dead_node);
+ btree = xfs_da3_node_tree_p(dead_node);
+ dead_level = deadhdr.level;
+ dead_hash = be32_to_cpu(btree[deadhdr.count - 1].hashval);
}
sib_buf = par_buf = NULL;
/*
* If the moved block has a left sibling, fix up the pointers.
*/
if ((sib_blkno = be32_to_cpu(dead_info->back))) {
- error = xfs_da_node_read(tp, ip, sib_blkno, -1, &sib_buf, w);
+ error = xfs_da3_node_read(tp, ip, sib_blkno, -1, &sib_buf, w);
if (error)
goto done;
sib_info = sib_buf->b_addr;
@@ -1867,7 +2190,7 @@ xfs_da_swap_lastblock(
* If the moved block has a right sibling, fix up the pointers.
*/
if ((sib_blkno = be32_to_cpu(dead_info->forw))) {
- error = xfs_da_node_read(tp, ip, sib_blkno, -1, &sib_buf, w);
+ error = xfs_da3_node_read(tp, ip, sib_blkno, -1, &sib_buf, w);
if (error)
goto done;
sib_info = sib_buf->b_addr;
@@ -1891,31 +2214,31 @@ xfs_da_swap_lastblock(
* Walk down the tree looking for the parent of the moved block.
*/
for (;;) {
- error = xfs_da_node_read(tp, ip, par_blkno, -1, &par_buf, w);
+ error = xfs_da3_node_read(tp, ip, par_blkno, -1, &par_buf, w);
if (error)
goto done;
par_node = par_buf->b_addr;
- if (unlikely(par_node->hdr.info.magic !=
- cpu_to_be16(XFS_DA_NODE_MAGIC) ||
- (level >= 0 && level != be16_to_cpu(par_node->hdr.level) + 1))) {
+ xfs_da3_node_hdr_from_disk(&par_hdr, par_node);
+ if (level >= 0 && level != par_hdr.level + 1) {
XFS_ERROR_REPORT("xfs_da_swap_lastblock(4)",
XFS_ERRLEVEL_LOW, mp);
error = XFS_ERROR(EFSCORRUPTED);
goto done;
}
- level = be16_to_cpu(par_node->hdr.level);
+ level = par_hdr.level;
+ btree = xfs_da3_node_tree_p(par_node);
for (entno = 0;
- entno < be16_to_cpu(par_node->hdr.count) &&
- be32_to_cpu(par_node->btree[entno].hashval) < dead_hash;
+ entno < par_hdr.count &&
+ be32_to_cpu(btree[entno].hashval) < dead_hash;
entno++)
continue;
- if (unlikely(entno == be16_to_cpu(par_node->hdr.count))) {
+ if (entno == par_hdr.count) {
XFS_ERROR_REPORT("xfs_da_swap_lastblock(5)",
XFS_ERRLEVEL_LOW, mp);
error = XFS_ERROR(EFSCORRUPTED);
goto done;
}
- par_blkno = be32_to_cpu(par_node->btree[entno].before);
+ par_blkno = be32_to_cpu(btree[entno].before);
if (level == dead_level + 1)
break;
xfs_trans_brelse(tp, par_buf);
@@ -1927,13 +2250,13 @@ xfs_da_swap_lastblock(
*/
for (;;) {
for (;
- entno < be16_to_cpu(par_node->hdr.count) &&
- be32_to_cpu(par_node->btree[entno].before) != last_blkno;
+ entno < par_hdr.count &&
+ be32_to_cpu(btree[entno].before) != last_blkno;
entno++)
continue;
- if (entno < be16_to_cpu(par_node->hdr.count))
+ if (entno < par_hdr.count)
break;
- par_blkno = be32_to_cpu(par_node->hdr.info.forw);
+ par_blkno = par_hdr.forw;
xfs_trans_brelse(tp, par_buf);
par_buf = NULL;
if (unlikely(par_blkno == 0)) {
@@ -1942,27 +2265,27 @@ xfs_da_swap_lastblock(
error = XFS_ERROR(EFSCORRUPTED);
goto done;
}
- error = xfs_da_node_read(tp, ip, par_blkno, -1, &par_buf, w);
+ error = xfs_da3_node_read(tp, ip, par_blkno, -1, &par_buf, w);
if (error)
goto done;
par_node = par_buf->b_addr;
- if (unlikely(
- be16_to_cpu(par_node->hdr.level) != level ||
- par_node->hdr.info.magic != cpu_to_be16(XFS_DA_NODE_MAGIC))) {
+ xfs_da3_node_hdr_from_disk(&par_hdr, par_node);
+ if (par_hdr.level != level) {
XFS_ERROR_REPORT("xfs_da_swap_lastblock(7)",
XFS_ERRLEVEL_LOW, mp);
error = XFS_ERROR(EFSCORRUPTED);
goto done;
}
+ btree = xfs_da3_node_tree_p(par_node);
entno = 0;
}
/*
* Update the parent entry pointing to the moved block.
*/
- par_node->btree[entno].before = cpu_to_be32(dead_blkno);
+ btree[entno].before = cpu_to_be32(dead_blkno);
xfs_trans_log_buf(tp, par_buf,
- XFS_DA_LOGRANGE(par_node, &par_node->btree[entno].before,
- sizeof(par_node->btree[entno].before)));
+ XFS_DA_LOGRANGE(par_node, &btree[entno].before,
+ sizeof(btree[entno].before)));
*dead_blknop = last_blkno;
*dead_bufp = last_buf;
return 0;
@@ -2004,14 +2327,15 @@ xfs_da_shrink_inode(
* Remove extents. If we get ENOSPC for a dir we have to move
* the last block to the place we want to kill.
*/
- if ((error = xfs_bunmapi(tp, dp, dead_blkno, count,
- xfs_bmapi_aflag(w)|XFS_BMAPI_METADATA,
- 0, args->firstblock, args->flist,
- &done)) == ENOSPC) {
+ error = xfs_bunmapi(tp, dp, dead_blkno, count,
+ xfs_bmapi_aflag(w)|XFS_BMAPI_METADATA,
+ 0, args->firstblock, args->flist, &done);
+ if (error == ENOSPC) {
if (w != XFS_DATA_FORK)
break;
- if ((error = xfs_da_swap_lastblock(args, &dead_blkno,
- &dead_buf)))
+ error = xfs_da3_swap_lastblock(args, &dead_blkno,
+ &dead_buf);
+ if (error)
break;
} else {
break;
@@ -2276,6 +2600,7 @@ xfs_da_read_buf(
magic1 = be32_to_cpu(hdr->magic);
if (unlikely(
XFS_TEST_ERROR((magic != XFS_DA_NODE_MAGIC) &&
+ (magic != XFS_DA3_NODE_MAGIC) &&
(magic != XFS_ATTR_LEAF_MAGIC) &&
(magic != XFS_DIR2_LEAF1_MAGIC) &&
(magic != XFS_DIR3_LEAF1_MAGIC) &&
@@ -2346,41 +2671,3 @@ out_free:
return -1;
return mappedbno;
}
-
-kmem_zone_t *xfs_da_state_zone; /* anchor for state struct zone */
-
-/*
- * Allocate a dir-state structure.
- * We don't put them on the stack since they're large.
- */
-xfs_da_state_t *
-xfs_da_state_alloc(void)
-{
- return kmem_zone_zalloc(xfs_da_state_zone, KM_NOFS);
-}
-
-/*
- * Kill the altpath contents of a da-state structure.
- */
-STATIC void
-xfs_da_state_kill_altpath(xfs_da_state_t *state)
-{
- int i;
-
- for (i = 0; i < state->altpath.active; i++)
- state->altpath.blk[i].bp = NULL;
- state->altpath.active = 0;
-}
-
-/*
- * Free a da-state structure.
- */
-void
-xfs_da_state_free(xfs_da_state_t *state)
-{
- xfs_da_state_kill_altpath(state);
-#ifdef DEBUG
- memset((char *)state, 0, sizeof(*state));
-#endif /* DEBUG */
- kmem_zone_free(xfs_da_state_zone, state);
-}
diff --git a/libxfs/xfs_dir2_node.c b/libxfs/xfs_dir2_node.c
index edf6e03..c1c50ad 100644
--- a/libxfs/xfs_dir2_node.c
+++ b/libxfs/xfs_dir2_node.c
@@ -1348,7 +1348,7 @@ xfs_dir2_leafn_split(
* block into the leaves.
*/
xfs_dir2_leafn_rebalance(state, oldblk, newblk);
- error = xfs_da_blk_link(state, oldblk, newblk);
+ error = xfs_da3_blk_link(state, oldblk, newblk);
if (error) {
return error;
}
@@ -1430,7 +1430,7 @@ xfs_dir2_leafn_toosmall(
*/
forward = (leafhdr.forw != 0);
memcpy(&state->altpath, &state->path, sizeof(state->path));
- error = xfs_da_path_shift(state, &state->altpath, forward, 0,
+ error = xfs_da3_path_shift(state, &state->altpath, forward, 0,
&rval);
if (error)
return error;
@@ -1492,10 +1492,10 @@ xfs_dir2_leafn_toosmall(
*/
memcpy(&state->altpath, &state->path, sizeof(state->path));
if (blkno < blk->blkno)
- error = xfs_da_path_shift(state, &state->altpath, forward, 0,
+ error = xfs_da3_path_shift(state, &state->altpath, forward, 0,
&rval);
else
- error = xfs_da_path_shift(state, &state->path, forward, 0,
+ error = xfs_da3_path_shift(state, &state->path, forward, 0,
&rval);
if (error) {
return error;
@@ -1591,7 +1591,7 @@ xfs_dir2_node_addname(
* Look up the name. We're not supposed to find it, but
* this gives us the insertion point.
*/
- error = xfs_da_node_lookup_int(state, &rval);
+ error = xfs_da3_node_lookup_int(state, &rval);
if (error)
rval = error;
if (rval != ENOENT) {
@@ -1617,7 +1617,7 @@ xfs_dir2_node_addname(
* It worked, fix the hash values up the btree.
*/
if (!(args->op_flags & XFS_DA_OP_JUSTCHECK))
- xfs_da_fixhashpath(state, &state->path);
+ xfs_da3_fixhashpath(state, &state->path);
} else {
/*
* It didn't work, we need to split the leaf block.
@@ -1629,7 +1629,7 @@ xfs_dir2_node_addname(
/*
* Split the leaf block and insert the new entry.
*/
- rval = xfs_da_split(state);
+ rval = xfs_da3_split(state);
}
done:
xfs_da_state_free(state);
@@ -2007,7 +2007,7 @@ xfs_dir2_node_addname_int(
/*
* Lookup an entry in a node-format directory.
- * All the real work happens in xfs_da_node_lookup_int.
+ * All the real work happens in xfs_da3_node_lookup_int.
* The only real output is the inode number of the entry.
*/
int /* error */
@@ -2032,7 +2032,7 @@ xfs_dir2_node_lookup(
/*
* Fill in the path to the entry in the cursor.
*/
- error = xfs_da_node_lookup_int(state, &rval);
+ error = xfs_da3_node_lookup_int(state, &rval);
if (error)
rval = error;
else if (rval == ENOENT && args->cmpresult == XFS_CMP_CASE) {
@@ -2087,7 +2087,7 @@ xfs_dir2_node_removename(
/*
* Look up the entry we're deleting, set up the cursor.
*/
- error = xfs_da_node_lookup_int(state, &rval);
+ error = xfs_da3_node_lookup_int(state, &rval);
if (error)
rval = error;
/*
@@ -2111,12 +2111,12 @@ xfs_dir2_node_removename(
/*
* Fix the hash values up the btree.
*/
- xfs_da_fixhashpath(state, &state->path);
+ xfs_da3_fixhashpath(state, &state->path);
/*
* If we need to join leaf blocks, do it.
*/
if (rval && state->path.active > 1)
- error = xfs_da_join(state);
+ error = xfs_da3_join(state);
/*
* If no errors so far, try conversion to leaf format.
*/
@@ -2158,7 +2158,7 @@ xfs_dir2_node_replace(
/*
* Lookup the entry to change in the btree.
*/
- error = xfs_da_node_lookup_int(state, &rval);
+ error = xfs_da3_node_lookup_int(state, &rval);
if (error) {
rval = error;
}
diff --git a/repair/dir.c b/repair/dir.c
index daca17d..cb04267 100644
--- a/repair/dir.c
+++ b/repair/dir.c
@@ -740,6 +740,8 @@ traverse_int_dablock(xfs_mount_t *mp,
xfs_da_intnode_t *node;
xfs_dfsbno_t fsbno;
xfs_buf_t *bp;
+ struct xfs_da_node_entry *btree;
+ struct xfs_da3_icnode_hdr nodehdr;
/*
* traverse down left-side of tree until we hit the
@@ -775,20 +777,21 @@ traverse_int_dablock(xfs_mount_t *mp,
}
node = (xfs_da_intnode_t *)XFS_BUF_PTR(bp);
+ btree = xfs_da3_node_tree_p(node);
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
- if (be16_to_cpu(node->hdr.info.magic) != XFS_DA_NODE_MAGIC) {
+ if (nodehdr.magic != XFS_DA_NODE_MAGIC) {
do_warn(_("bad dir/attr magic number in inode %" PRIu64 ", "
"file bno = %u, fsbno = %" PRIu64 "\n"),
da_cursor->ino, bno, fsbno);
libxfs_putbuf(bp);
goto error_out;
}
- if (be16_to_cpu(node->hdr.count) >
- mp->m_dir_node_ents) {
+ if (nodehdr.count > mp->m_dir_node_ents) {
do_warn(_("bad record count in inode %" PRIu64 ", "
"count = %d, max = %d\n"),
da_cursor->ino,
- be16_to_cpu(node->hdr.count),
+ nodehdr.count,
mp->m_dir_node_ents);
libxfs_putbuf(bp);
goto error_out;
@@ -798,9 +801,9 @@ traverse_int_dablock(xfs_mount_t *mp,
* maintain level counter
*/
if (i == -1)
- i = da_cursor->active = be16_to_cpu(node->hdr.level);
+ i = da_cursor->active = nodehdr.level;
else {
- if (be16_to_cpu(node->hdr.level) == i - 1) {
+ if (nodehdr.level == i - 1) {
i--;
} else {
if (whichfork == XFS_DATA_FORK)
@@ -816,8 +819,7 @@ traverse_int_dablock(xfs_mount_t *mp,
}
}
- da_cursor->level[i].hashval = be32_to_cpu(
- node->btree[0].hashval);
+ da_cursor->level[i].hashval = be32_to_cpu(btree[0].hashval);
da_cursor->level[i].bp = bp;
da_cursor->level[i].bno = bno;
da_cursor->level[i].index = 0;
@@ -828,7 +830,7 @@ traverse_int_dablock(xfs_mount_t *mp,
/*
* set up new bno for next level down
*/
- bno = be32_to_cpu(node->btree[0].before);
+ bno = be32_to_cpu(btree[0].before);
} while (node != NULL && i > 1);
/*
@@ -912,6 +914,8 @@ verify_final_da_path(xfs_mount_t *mp,
int bad = 0;
int entry;
int this_level = p_level + 1;
+ struct xfs_da_node_entry *btree;
+ struct xfs_da3_icnode_hdr nodehdr;
#ifdef XR_DIR_TRACE
fprintf(stderr, "in verify_final_da_path, this_level = %d\n",
@@ -923,32 +927,34 @@ verify_final_da_path(xfs_mount_t *mp,
*/
entry = cursor->level[this_level].index;
node = (xfs_da_intnode_t *)XFS_BUF_PTR(cursor->level[this_level].bp);
+ btree = xfs_da3_node_tree_p(node);
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
/*
* check internal block consistency on this level -- ensure
* that all entries are used, encountered and expected hashvals
* match, etc.
*/
- if (entry != be16_to_cpu(node->hdr.count) - 1) {
+ if (entry != nodehdr.count - 1) {
do_warn(_("directory/attribute block used/count "
"inconsistency - %d/%hu\n"),
- entry, be16_to_cpu(node->hdr.count));
+ entry, nodehdr.count);
bad++;
}
/*
* hash values monotonically increasing ???
*/
if (cursor->level[this_level].hashval >=
- be32_to_cpu(node->btree[entry].hashval)) {
+ be32_to_cpu(btree[entry].hashval)) {
do_warn(_("directory/attribute block hashvalue inconsistency, "
"expected > %u / saw %u\n"),
cursor->level[this_level].hashval,
- be32_to_cpu(node->btree[entry].hashval));
+ be32_to_cpu(btree[entry].hashval));
bad++;
}
- if (be32_to_cpu(node->hdr.info.forw) != 0) {
+ if (nodehdr.forw != 0) {
do_warn(_("bad directory/attribute forward block pointer, "
"expected 0, saw %u\n"),
- be32_to_cpu(node->hdr.info.forw));
+ nodehdr.forw);
bad++;
}
if (bad) {
@@ -966,12 +972,11 @@ verify_final_da_path(xfs_mount_t *mp,
/*
* ok, now check descendant block number against this level
*/
- if (cursor->level[p_level].bno != be32_to_cpu(
- node->btree[entry].before)) {
+ if (cursor->level[p_level].bno != be32_to_cpu( btree[entry].before)) {
#ifdef XR_DIR_TRACE
fprintf(stderr, "bad directory btree pointer, child bno should "
"be %d, block bno is %d, hashval is %u\n",
- be16_to_cpu(node->btree[entry].before),
+ be16_to_cpu(btree[entry].before),
cursor->level[p_level].bno,
cursor->level[p_level].hashval);
fprintf(stderr, "verify_final_da_path returns 1 (bad) #1a\n");
@@ -979,14 +984,13 @@ verify_final_da_path(xfs_mount_t *mp,
return(1);
}
- if (cursor->level[p_level].hashval != be32_to_cpu(
- node->btree[entry].hashval)) {
+ if (cursor->level[p_level].hashval != be32_to_cpu(btree[entry].hashval)) {
if (!no_modify) {
do_warn(_("correcting bad hashval in non-leaf "
"dir/attr block\n\tin (level %d) in "
"inode %" PRIu64 ".\n"),
this_level, cursor->ino);
- node->btree[entry].hashval = cpu_to_be32(
+ btree[entry].hashval = cpu_to_be32(
cursor->level[p_level].hashval);
cursor->level[this_level].dirty++;
} else {
@@ -1001,7 +1005,7 @@ verify_final_da_path(xfs_mount_t *mp,
* Note: squirrel hashval away _before_ releasing the
* buffer, preventing a use-after-free problem.
*/
- hashval = be32_to_cpu(node->btree[entry].hashval);
+ hashval = be32_to_cpu(btree[entry].hashval);
/*
* release/write buffer
@@ -1085,6 +1089,8 @@ verify_da_path(xfs_mount_t *mp,
int bad;
int entry;
int this_level = p_level + 1;
+ struct xfs_da_node_entry *btree;
+ struct xfs_da3_icnode_hdr nodehdr;
/*
* index is currently set to point to the entry that
@@ -1092,20 +1098,22 @@ verify_da_path(xfs_mount_t *mp,
*/
entry = cursor->level[this_level].index;
node = (xfs_da_intnode_t *)XFS_BUF_PTR(cursor->level[this_level].bp);
+ btree = xfs_da3_node_tree_p(node);
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
/*
* if this block is out of entries, validate this
* block and move on to the next block.
* and update cursor value for said level
*/
- if (entry >= be16_to_cpu(node->hdr.count)) {
+ if (entry >= nodehdr.count) {
/*
* update the hash value for this level before
* validating it. bno value should be ok since
* it was set when the block was first read in.
*/
cursor->level[this_level].hashval =
- be32_to_cpu(node->btree[entry - 1].hashval);
+ be32_to_cpu(btree[entry - 1].hashval);
/*
* keep track of greatest block # -- that gets
@@ -1123,7 +1131,7 @@ verify_da_path(xfs_mount_t *mp,
/*
* ok, now get the next buffer and check sibling pointers
*/
- dabno = be32_to_cpu(node->hdr.info.forw);
+ dabno = nodehdr.forw;
ASSERT(dabno != 0);
fsbno = blkmap_get(cursor->blkmap, dabno);
@@ -1144,36 +1152,38 @@ verify_da_path(xfs_mount_t *mp,
}
newnode = (xfs_da_intnode_t *)XFS_BUF_PTR(bp);
+ btree = xfs_da3_node_tree_p(node);
+ xfs_da3_node_hdr_from_disk(&nodehdr, newnode);
+
/*
* verify magic number and back pointer, sanity-check
* entry count, verify level
*/
bad = 0;
- if (XFS_DA_NODE_MAGIC != be16_to_cpu(newnode->hdr.info.magic)) {
+ if (XFS_DA_NODE_MAGIC != nodehdr.magic) {
do_warn(
_("bad magic number %x in block %u (%" PRIu64 ") for directory inode %" PRIu64 "\n"),
- be16_to_cpu(newnode->hdr.info.magic),
+ nodehdr.magic,
dabno, fsbno, cursor->ino);
bad++;
}
- if (be32_to_cpu(newnode->hdr.info.back) !=
- cursor->level[this_level].bno) {
+ if (nodehdr.back != cursor->level[this_level].bno) {
do_warn(
_("bad back pointer in block %u (%"PRIu64 ") for directory inode %" PRIu64 "\n"),
dabno, fsbno, cursor->ino);
bad++;
}
- if (be16_to_cpu(newnode->hdr.count) > mp->m_dir_node_ents) {
+ if (nodehdr.count > mp->m_dir_node_ents) {
do_warn(
_("entry count %d too large in block %u (%" PRIu64 ") for directory inode %" PRIu64 "\n"),
- be16_to_cpu(newnode->hdr.count),
+ nodehdr.count,
dabno, fsbno, cursor->ino);
bad++;
}
- if (be16_to_cpu(newnode->hdr.level) != this_level) {
+ if (nodehdr.level != this_level) {
do_warn(
_("bad level %d in block %u (%" PRIu64 ") for directory inode %" PRIu64 "\n"),
- be16_to_cpu(newnode->hdr.level),
+ nodehdr.level,
dabno, fsbno, cursor->ino);
bad++;
}
@@ -1199,7 +1209,7 @@ verify_da_path(xfs_mount_t *mp,
cursor->level[this_level].dirty = 0;
cursor->level[this_level].bno = dabno;
cursor->level[this_level].hashval =
- be32_to_cpu(newnode->btree[0].hashval);
+ be32_to_cpu(btree[0].hashval);
#ifdef XR_DIR_TRACE
cursor->level[this_level].n = newnode;
#endif
@@ -1210,12 +1220,11 @@ verify_da_path(xfs_mount_t *mp,
/*
* ditto for block numbers
*/
- if (cursor->level[p_level].bno !=
- be32_to_cpu(node->btree[entry].before)) {
+ if (cursor->level[p_level].bno != be32_to_cpu(btree[entry].before)) {
#ifdef XR_DIR_TRACE
fprintf(stderr, "bad directory btree pointer, child bno "
"should be %d, block bno is %d, hashval is %u\n",
- be32_to_cpu(node->btree[entry].before),
+ be32_to_cpu(btree[entry].before),
cursor->level[p_level].bno,
cursor->level[p_level].hashval);
fprintf(stderr, "verify_da_path returns 1 (bad) #1a\n");
@@ -1227,13 +1236,13 @@ verify_da_path(xfs_mount_t *mp,
* block against the hashval in the current entry
*/
if (cursor->level[p_level].hashval !=
- be32_to_cpu(node->btree[entry].hashval)) {
+ be32_to_cpu(btree[entry].hashval)) {
if (!no_modify) {
do_warn(_("correcting bad hashval in interior "
"dir/attr block\n\tin (level %d) in "
"inode %" PRIu64 ".\n"),
this_level, cursor->ino);
- node->btree[entry].hashval = cpu_to_be32(
+ btree[entry].hashval = cpu_to_be32(
cursor->level[p_level].hashval);
cursor->level[this_level].dirty++;
} else {
diff --git a/repair/dir2.c b/repair/dir2.c
index de5e737..b774897 100644
--- a/repair/dir2.c
+++ b/repair/dir2.c
@@ -127,9 +127,10 @@ traverse_int_dir2block(xfs_mount_t *mp,
struct xfs_buf *bp;
int i;
int nex;
- xfs_da_blkinfo_t *info;
xfs_da_intnode_t *node;
bmap_ext_t lbmp;
+ struct xfs_da_node_entry *btree;
+ struct xfs_da3_icnode_hdr nodehdr;
/*
* traverse down left-side of tree until we hit the
@@ -138,7 +139,7 @@ traverse_int_dir2block(xfs_mount_t *mp,
*/
bno = mp->m_dirleafblk;
i = -1;
- info = NULL;
+ node = NULL;
da_cursor->active = 0;
do {
@@ -161,9 +162,10 @@ _("can't read block %u for directory inode %" PRIu64 "\n"),
goto error_out;
}
- info = bp->b_addr;
+ node = bp->b_addr;
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
- if (be16_to_cpu(info->magic) == XFS_DIR2_LEAFN_MAGIC) {
+ if (nodehdr.magic == XFS_DIR2_LEAFN_MAGIC) {
if ( i != -1 ) {
do_warn(
_("found non-root LEAFN node in inode %" PRIu64 " bno = %u\n"),
@@ -172,20 +174,21 @@ _("found non-root LEAFN node in inode %" PRIu64 " bno = %u\n"),
*rbno = 0;
libxfs_putbuf(bp);
return(1);
- } else if (be16_to_cpu(info->magic) != XFS_DA_NODE_MAGIC) {
+ } else if (nodehdr.magic != XFS_DA_NODE_MAGIC) {
libxfs_putbuf(bp);
do_warn(
_("bad dir magic number 0x%x in inode %" PRIu64 " bno = %u\n"),
- be16_to_cpu(info->magic),
+ nodehdr.magic,
da_cursor->ino, bno);
goto error_out;
}
- node = (xfs_da_intnode_t*)info;
- if (be16_to_cpu(node->hdr.count) > mp->m_dir_node_ents) {
+ btree = xfs_da3_node_tree_p(node);
+ if (nodehdr.count > mp->m_dir_node_ents) {
libxfs_putbuf(bp);
do_warn(
-_("bad record count in inode %" PRIu64 ", count = %d, max = %d\n"), da_cursor->ino,
- be16_to_cpu(node->hdr.count),
+_("bad record count in inode %" PRIu64 ", count = %d, max = %d\n"),
+ da_cursor->ino,
+ nodehdr.count,
mp->m_dir_node_ents);
goto error_out;
}
@@ -193,7 +196,7 @@ _("bad record count in inode %" PRIu64 ", count = %d, max = %d\n"), da_cursor->i
* maintain level counter
*/
if (i == -1) {
- i = da_cursor->active = be16_to_cpu(node->hdr.level);
+ i = da_cursor->active = nodehdr.level;
if (i >= XFS_DA_NODE_MAXDEPTH) {
do_warn(
_("bad header depth for directory inode %" PRIu64 "\n"),
@@ -203,7 +206,7 @@ _("bad header depth for directory inode %" PRIu64 "\n"),
goto error_out;
}
} else {
- if (be16_to_cpu(node->hdr.level) == i - 1) {
+ if (nodehdr.level == i - 1) {
i--;
} else {
do_warn(
@@ -214,8 +217,7 @@ _("bad directory btree for directory inode %" PRIu64 "\n"),
}
}
- da_cursor->level[i].hashval =
- be32_to_cpu(node->btree[0].hashval);
+ da_cursor->level[i].hashval = be32_to_cpu(btree[0].hashval);
da_cursor->level[i].bp = bp;
da_cursor->level[i].bno = bno;
da_cursor->level[i].index = 0;
@@ -223,8 +225,8 @@ _("bad directory btree for directory inode %" PRIu64 "\n"),
/*
* set up new bno for next level down
*/
- bno = be32_to_cpu(node->btree[0].before);
- } while (info != NULL && i > 1);
+ bno = be32_to_cpu(btree[0].before);
+ } while (node != NULL && i > 1);
/*
* now return block number and get out
@@ -306,6 +308,8 @@ verify_final_dir2_path(xfs_mount_t *mp,
int bad = 0;
int entry;
int this_level = p_level + 1;
+ struct xfs_da_node_entry *btree;
+ struct xfs_da3_icnode_hdr nodehdr;
/*
* the index should point to the next "unprocessed" entry
@@ -313,32 +317,34 @@ verify_final_dir2_path(xfs_mount_t *mp,
*/
entry = cursor->level[this_level].index;
node = (xfs_da_intnode_t *)(cursor->level[this_level].bp->b_addr);
+ btree = xfs_da3_node_tree_p(node);
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
/*
* check internal block consistency on this level -- ensure
* that all entries are used, encountered and expected hashvals
* match, etc.
*/
- if (entry != be16_to_cpu(node->hdr.count) - 1) {
+ if (entry != nodehdr.count - 1) {
do_warn(
_("directory block used/count inconsistency - %d / %hu\n"),
- entry, be16_to_cpu(node->hdr.count));
+ entry, nodehdr.count);
bad++;
}
/*
* hash values monotonically increasing ???
*/
if (cursor->level[this_level].hashval >=
- be32_to_cpu(node->btree[entry].hashval)) {
+ be32_to_cpu(btree[entry].hashval)) {
do_warn(_("directory/attribute block hashvalue inconsistency, "
"expected > %u / saw %u\n"),
cursor->level[this_level].hashval,
- be32_to_cpu(node->btree[entry].hashval));
+ be32_to_cpu(btree[entry].hashval));
bad++;
}
- if (be32_to_cpu(node->hdr.info.forw) != 0) {
+ if (nodehdr.forw != 0) {
do_warn(_("bad directory/attribute forward block pointer, "
"expected 0, saw %u\n"),
- be32_to_cpu(node->hdr.info.forw));
+ nodehdr.forw);
bad++;
}
if (bad) {
@@ -355,18 +361,17 @@ verify_final_dir2_path(xfs_mount_t *mp,
/*
* ok, now check descendant block number against this level
*/
- if (cursor->level[p_level].bno !=
- be32_to_cpu(node->btree[entry].before))
+ if (cursor->level[p_level].bno != be32_to_cpu(btree[entry].before))
return(1);
if (cursor->level[p_level].hashval !=
- be32_to_cpu(node->btree[entry].hashval)) {
+ be32_to_cpu(btree[entry].hashval)) {
if (!no_modify) {
do_warn(
_("correcting bad hashval in non-leaf dir block\n"
"\tin (level %d) in inode %" PRIu64 ".\n"),
this_level, cursor->ino);
- node->btree[entry].hashval = cpu_to_be32(
+ btree[entry].hashval = cpu_to_be32(
cursor->level[p_level].hashval);
cursor->level[this_level].dirty++;
} else {
@@ -399,8 +404,7 @@ _("would correct bad hashval in non-leaf dir block\n"
* set hashvalue to correctl reflect the now-validated
* last entry in this block and continue upwards validation
*/
- cursor->level[this_level].hashval =
- be32_to_cpu(node->btree[entry].hashval);
+ cursor->level[this_level].hashval = be32_to_cpu(btree[entry].hashval);
return(verify_final_dir2_path(mp, cursor, this_level));
}
@@ -459,6 +463,8 @@ verify_dir2_path(xfs_mount_t *mp,
bmap_ext_t *bmp;
int nex;
bmap_ext_t lbmp;
+ struct xfs_da_node_entry *btree;
+ struct xfs_da3_icnode_hdr nodehdr;
/*
* index is currently set to point to the entry that
@@ -466,20 +472,22 @@ verify_dir2_path(xfs_mount_t *mp,
*/
entry = cursor->level[this_level].index;
node = cursor->level[this_level].bp->b_addr;
+ btree = xfs_da3_node_tree_p(node);
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
/*
* if this block is out of entries, validate this
* block and move on to the next block.
* and update cursor value for said level
*/
- if (entry >= be16_to_cpu(node->hdr.count)) {
+ if (entry >= nodehdr.count) {
/*
* update the hash value for this level before
* validating it. bno value should be ok since
* it was set when the block was first read in.
*/
cursor->level[this_level].hashval =
- be32_to_cpu(node->btree[entry - 1].hashval);
+ be32_to_cpu(btree[entry - 1].hashval);
/*
* keep track of greatest block # -- that gets
@@ -497,7 +505,7 @@ verify_dir2_path(xfs_mount_t *mp,
/*
* ok, now get the next buffer and check sibling pointers
*/
- dabno = be32_to_cpu(node->hdr.info.forw);
+ dabno = nodehdr.forw;
ASSERT(dabno != 0);
nex = blkmap_getn(cursor->blkmap, dabno, mp->m_dirblkfsbs,
&bmp, &lbmp);
@@ -520,36 +528,37 @@ _("can't read block %u for directory inode %" PRIu64 "\n"),
}
newnode = bp->b_addr;
+ btree = xfs_da3_node_tree_p(newnode);
+ xfs_da3_node_hdr_from_disk(&nodehdr, node);
/*
* verify magic number and back pointer, sanity-check
* entry count, verify level
*/
bad = 0;
- if (XFS_DA_NODE_MAGIC != be16_to_cpu(newnode->hdr.info.magic)) {
+ if (XFS_DA_NODE_MAGIC != nodehdr.magic) {
do_warn(
_("bad magic number %x in block %u for directory inode %" PRIu64 "\n"),
- be16_to_cpu(newnode->hdr.info.magic),
+ nodehdr.magic,
dabno, cursor->ino);
bad++;
}
- if (be32_to_cpu(newnode->hdr.info.back) !=
- cursor->level[this_level].bno) {
+ if (nodehdr.back != cursor->level[this_level].bno) {
do_warn(
_("bad back pointer in block %u for directory inode %" PRIu64 "\n"),
dabno, cursor->ino);
bad++;
}
- if (be16_to_cpu(newnode->hdr.count) > mp->m_dir_node_ents) {
+ if (nodehdr.count > mp->m_dir_node_ents) {
do_warn(
_("entry count %d too large in block %u for directory inode %" PRIu64 "\n"),
- be16_to_cpu(newnode->hdr.count),
+ nodehdr.count,
dabno, cursor->ino);
bad++;
}
- if (be16_to_cpu(newnode->hdr.level) != this_level) {
+ if (nodehdr.level != this_level) {
do_warn(
_("bad level %d in block %u for directory inode %" PRIu64 "\n"),
- be16_to_cpu(newnode->hdr.level),
+ nodehdr.level,
dabno, cursor->ino);
bad++;
}
@@ -572,7 +581,7 @@ _("bad level %d in block %u for directory inode %" PRIu64 "\n"),
cursor->level[this_level].dirty = 0;
cursor->level[this_level].bno = dabno;
cursor->level[this_level].hashval =
- be32_to_cpu(newnode->btree[0].hashval);
+ be32_to_cpu(btree[0].hashval);
node = newnode;
entry = cursor->level[this_level].index = 0;
@@ -580,21 +589,20 @@ _("bad level %d in block %u for directory inode %" PRIu64 "\n"),
/*
* ditto for block numbers
*/
- if (cursor->level[p_level].bno !=
- be32_to_cpu(node->btree[entry].before))
+ if (cursor->level[p_level].bno != be32_to_cpu(btree[entry].before))
return(1);
/*
* ok, now validate last hashvalue in the descendant
* block against the hashval in the current entry
*/
if (cursor->level[p_level].hashval !=
- be32_to_cpu(node->btree[entry].hashval)) {
+ be32_to_cpu(btree[entry].hashval)) {
if (!no_modify) {
do_warn(
_("correcting bad hashval in interior dir block\n"
"\tin (level %d) in inode %" PRIu64 ".\n"),
this_level, cursor->ino);
- node->btree[entry].hashval = cpu_to_be32(
+ btree[entry].hashval = cpu_to_be32(
cursor->level[p_level].hashval);
cursor->level[this_level].dirty++;
} else {
diff --git a/repair/phase6.c b/repair/phase6.c
index e3d0934..090e400 100644
--- a/repair/phase6.c
+++ b/repair/phase6.c
@@ -1199,6 +1199,7 @@ map_first_dblock_fsbno(xfs_mount_t *mp,
int i;
int error;
char *ftype;
+ struct xfs_da_node_entry *btree;
/*
* traverse down left-side of tree until we hit the
@@ -1265,6 +1266,7 @@ _("can't map block %d in %s inode %" PRIu64 ", xfs_bmapi returns %d, nmap = %d\n
}
node = (xfs_da_intnode_t *)XFS_BUF_PTR(bp);
+ btree = xfs_da3_node_tree_p(node);
if (be16_to_cpu(node->hdr.info.magic) != XFS_DA_NODE_MAGIC) {
libxfs_putbuf(bp);
@@ -1275,9 +1277,9 @@ _("bad dir/attr magic number in inode %" PRIu64 ", file bno = %u, fsbno = %" PRI
}
if (i == -1)
- i = be16_to_cpu(node->hdr.level);
+ i = be16_to_cpu(node->hdr.__level);
- da_bno = be32_to_cpu(node->btree[0].before);
+ da_bno = be32_to_cpu(btree[0].before);
libxfs_putbuf(bp);
bp = NULL;
xfsprogs-crc-patchset-v2/xfs-add-crc-to-attr-leaf 0000600 0001750 0001750 00000330714 12132011546 020562 0 ustar dave dave xfs: add CRCs to attr leaf blocks
From: Dave Chinner <dchinner@xxxxxxxxxx>
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
db/attr.c | 16 +-
db/dir.c | 4 +-
db/dir2.c | 2 +-
db/metadump.c | 4 +-
include/xfs_attr_leaf.h | 122 +++-
include/xfs_da_btree.h | 5 +-
libxfs/xfs_attr.c | 66 +-
libxfs/xfs_attr_leaf.c | 1529 +++++++++++++++++++++++++++--------------------
libxfs/xfs_da_btree.c | 34 +-
repair/attr_repair.c | 4 +-
10 files changed, 1034 insertions(+), 752 deletions(-)
diff --git a/db/attr.c b/db/attr.c
index a5087b8..05049ba 100644
--- a/db/attr.c
+++ b/db/attr.c
@@ -143,7 +143,7 @@ const field_t attr_node_entry_flds[] = {
#define HOFF(f) bitize(offsetof(xfs_da_node_hdr_t, f))
const field_t attr_node_hdr_flds[] = {
{ "info", FLDT_ATTR_BLKINFO, OI(HOFF(info)), C1, 0, TYP_NONE },
- { "count", FLDT_UINT16D, OI(HOFF(count)), C1, 0, TYP_NONE },
+ { "count", FLDT_UINT16D, OI(HOFF(__count)), C1, 0, TYP_NONE },
{ "level", FLDT_UINT16D, OI(HOFF(__level)), C1, 0, TYP_NONE },
{ NULL }
};
@@ -219,7 +219,7 @@ attr_leaf_name_local_name_count(
e = &block->entries[i];
if (be16_to_cpu(e->nameidx) == off) {
if (e->flags & XFS_ATTR_LOCAL) {
- l = xfs_attr_leaf_name_local(block, i);
+ l = xfs_attr3_leaf_name_local(block, i);
return l->namelen;
} else
return 0;
@@ -248,7 +248,7 @@ attr_leaf_name_local_value_count(
e = &block->entries[i];
if (be16_to_cpu(e->nameidx) == off) {
if (e->flags & XFS_ATTR_LOCAL) {
- l = xfs_attr_leaf_name_local(block, i);
+ l = xfs_attr3_leaf_name_local(block, i);
return be16_to_cpu(l->valuelen);
} else
return 0;
@@ -285,7 +285,7 @@ attr_leaf_name_local_value_offset(
if (i >= be16_to_cpu(block->hdr.count))
return 0;
- l = xfs_attr_leaf_name_local(block, i);
+ l = xfs_attr3_leaf_name_local(block, i);
vp = (char *)&l->nameval[l->namelen];
return (int)bitize(vp - (char *)l);
}
@@ -333,7 +333,7 @@ attr_leaf_name_remote_name_count(
e = &block->entries[i];
if (be16_to_cpu(e->nameidx) == off) {
if (!(e->flags & XFS_ATTR_LOCAL)) {
- r = xfs_attr_leaf_name_remote(block, i);
+ r = xfs_attr3_leaf_name_remote(block, i);
return r->namelen;
} else
return 0;
@@ -360,11 +360,11 @@ attr_leaf_name_size(
return 0;
e = &block->entries[idx];
if (e->flags & XFS_ATTR_LOCAL) {
- l = xfs_attr_leaf_name_local(block, idx);
+ l = xfs_attr3_leaf_name_local(block, idx);
return (int)bitize(xfs_attr_leaf_entsize_local(l->namelen,
be16_to_cpu(l->valuelen)));
} else {
- r = xfs_attr_leaf_name_remote(block, idx);
+ r = xfs_attr3_leaf_name_remote(block, idx);
return (int)bitize(xfs_attr_leaf_entsize_remote(r->namelen));
}
}
@@ -412,7 +412,7 @@ attr_node_btree_count(
block = obj;
if (be16_to_cpu(block->hdr.info.magic) != XFS_DA_NODE_MAGIC)
return 0;
- return be16_to_cpu(block->hdr.count);
+ return be16_to_cpu(block->hdr.__count);
}
/*ARGSUSED*/
diff --git a/db/dir.c b/db/dir.c
index 70ecc46..a9b1654 100644
--- a/db/dir.c
+++ b/db/dir.c
@@ -111,7 +111,7 @@ const field_t dir_node_entry_flds[] = {
#define HOFF(f) bitize(offsetof(xfs_da_node_hdr_t, f))
const field_t dir_node_hdr_flds[] = {
{ "info", FLDT_DIR_BLKINFO, OI(HOFF(info)), C1, 0, TYP_NONE },
- { "count", FLDT_UINT16D, OI(HOFF(count)), C1, 0, TYP_NONE },
+ { "count", FLDT_UINT16D, OI(HOFF(__count)), C1, 0, TYP_NONE },
{ "level", FLDT_UINT16D, OI(HOFF(__level)), C1, 0, TYP_NONE },
{ NULL }
};
@@ -228,7 +228,7 @@ dir_node_btree_count(
block = obj;
if (be16_to_cpu(block->hdr.info.magic) != XFS_DA_NODE_MAGIC)
return 0;
- return be16_to_cpu(block->hdr.count);
+ return be16_to_cpu(block->hdr.__count);
}
/*ARGSUSED*/
diff --git a/db/dir2.c b/db/dir2.c
index 07809b0..2cbc22a 100644
--- a/db/dir2.c
+++ b/db/dir2.c
@@ -684,7 +684,7 @@ dir2_node_btree_count(
node = obj;
if (be16_to_cpu(node->hdr.info.magic) != XFS_DA_NODE_MAGIC)
return 0;
- return be16_to_cpu(node->hdr.count);
+ return be16_to_cpu(node->hdr.__count);
}
/*ARGSUSED*/
diff --git a/db/metadump.c b/db/metadump.c
index 0635e7b..44e7162 100644
--- a/db/metadump.c
+++ b/db/metadump.c
@@ -1282,7 +1282,7 @@ obfuscate_attr_blocks(
break;
}
if (entry->flags & XFS_ATTR_LOCAL) {
- local = xfs_attr_leaf_name_local(leaf, i);
+ local = xfs_attr3_leaf_name_local(leaf, i);
if (local->namelen == 0) {
if (show_warnings)
print_warning("zero length for "
@@ -1295,7 +1295,7 @@ obfuscate_attr_blocks(
memset(&local->nameval[local->namelen], 0,
be16_to_cpu(local->valuelen));
} else {
- remote = xfs_attr_leaf_name_remote(leaf, i);
+ remote = xfs_attr3_leaf_name_remote(leaf, i);
if (remote->namelen == 0 ||
remote->valueblk == 0) {
if (show_warnings)
diff --git a/include/xfs_attr_leaf.h b/include/xfs_attr_leaf.h
index 77de139..f9d7846 100644
--- a/include/xfs_attr_leaf.h
+++ b/include/xfs_attr_leaf.h
@@ -1,5 +1,6 @@
/*
* Copyright (c) 2000,2002-2003,2005 Silicon Graphics, Inc.
+ * Copyright (c) 2013 Red Hat, Inc.
* All Rights Reserved.
*
* This program is free software; you can redistribute it and/or
@@ -89,7 +90,7 @@ typedef struct xfs_attr_leaf_hdr { /* constant-structure header block */
typedef struct xfs_attr_leaf_entry { /* sorted on key, not name */
__be32 hashval; /* hash value of name */
- __be16 nameidx; /* index into buffer of name/value */
+ __be16 nameidx; /* index into buffer of name/value */
__u8 flags; /* LOCAL/ROOT/SECURE/INCOMPLETE flag */
__u8 pad2; /* unused pad byte */
} xfs_attr_leaf_entry_t;
@@ -115,6 +116,54 @@ typedef struct xfs_attr_leafblock {
} xfs_attr_leafblock_t;
/*
+ * CRC enabled leaf structures. Called "version 3" structures to match the
+ * version number of the directory and dablk structures for this feature, and
+ * attr2 is already taken by the variable inode attribute fork size feature.
+ */
+struct xfs_attr3_leaf_hdr {
+ struct xfs_da3_blkinfo info;
+ __be16 count;
+ __be16 usedbytes;
+ __be16 firstused;
+ __u8 holes;
+ __u8 pad1;
+ struct xfs_attr_leaf_map freemap[XFS_ATTR_LEAF_MAPSIZE];
+};
+
+#define XFS_ATTR3_LEAF_CRC_OFF (offsetof(struct xfs_attr3_leaf_hdr, info.crc))
+
+struct xfs_attr3_leafblock {
+ struct xfs_attr3_leaf_hdr hdr;
+ struct xfs_attr_leaf_entry entries[1];
+
+ /*
+ * The rest of the block contains the following structures after the
+ * leaf entries, growing from the bottom up. The variables are never
+ * referenced, the locations accessed purely from helper functions.
+ *
+ * struct xfs_attr_leaf_name_local
+ * struct xfs_attr_leaf_name_remote
+ */
+};
+
+/*
+ * incore, neutral version of the attribute leaf header
+ */
+struct xfs_attr3_icleaf_hdr {
+ __uint32_t forw;
+ __uint32_t back;
+ __uint16_t magic;
+ __uint16_t count;
+ __uint16_t usedbytes;
+ __uint16_t firstused;
+ __u8 holes;
+ struct {
+ __uint16_t base;
+ __uint16_t size;
+ } freemap[XFS_ATTR_LEAF_MAPSIZE];
+};
+
+/*
* Flags used in the leaf_entry[i].flags field.
* NOTE: the INCOMPLETE bit must not collide with the flags bits specified
* on the system call, they are "or"ed together for various operations.
@@ -147,26 +196,43 @@ typedef struct xfs_attr_leafblock {
*/
#define XFS_ATTR_LEAF_NAME_ALIGN ((uint)sizeof(xfs_dablk_t))
+static inline int
+xfs_attr3_leaf_hdr_size(struct xfs_attr_leafblock *leafp)
+{
+ if (leafp->hdr.info.magic == cpu_to_be16(XFS_ATTR3_LEAF_MAGIC))
+ return sizeof(struct xfs_attr3_leaf_hdr);
+ return sizeof(struct xfs_attr_leaf_hdr);
+}
+
+static inline struct xfs_attr_leaf_entry *
+xfs_attr3_leaf_entryp(xfs_attr_leafblock_t *leafp)
+{
+ if (leafp->hdr.info.magic == cpu_to_be16(XFS_ATTR3_LEAF_MAGIC))
+ return &((struct xfs_attr3_leafblock *)leafp)->entries[0];
+ return &leafp->entries[0];
+}
+
/*
* Cast typed pointers for "local" and "remote" name/value structs.
*/
-static inline xfs_attr_leaf_name_remote_t *
-xfs_attr_leaf_name_remote(xfs_attr_leafblock_t *leafp, int idx)
+static inline char *
+xfs_attr3_leaf_name(xfs_attr_leafblock_t *leafp, int idx)
{
- return (xfs_attr_leaf_name_remote_t *)
- &((char *)leafp)[be16_to_cpu(leafp->entries[idx].nameidx)];
+ struct xfs_attr_leaf_entry *entries = xfs_attr3_leaf_entryp(leafp);
+
+ return &((char *)leafp)[be16_to_cpu(entries[idx].nameidx)];
}
-static inline xfs_attr_leaf_name_local_t *
-xfs_attr_leaf_name_local(xfs_attr_leafblock_t *leafp, int idx)
+static inline xfs_attr_leaf_name_remote_t *
+xfs_attr3_leaf_name_remote(xfs_attr_leafblock_t *leafp, int idx)
{
- return (xfs_attr_leaf_name_local_t *)
- &((char *)leafp)[be16_to_cpu(leafp->entries[idx].nameidx)];
+ return (xfs_attr_leaf_name_remote_t *)xfs_attr3_leaf_name(leafp, idx);
}
-static inline char *xfs_attr_leaf_name(xfs_attr_leafblock_t *leafp, int idx)
+static inline xfs_attr_leaf_name_local_t *
+xfs_attr3_leaf_name_local(xfs_attr_leafblock_t *leafp, int idx)
{
- return &((char *)leafp)[be16_to_cpu(leafp->entries[idx].nameidx)];
+ return (xfs_attr_leaf_name_local_t *)xfs_attr3_leaf_name(leafp, idx);
}
/*
@@ -221,37 +287,37 @@ int xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes);
/*
* Internal routines when attribute fork size == XFS_LBSIZE(mp).
*/
-int xfs_attr_leaf_to_node(struct xfs_da_args *args);
-int xfs_attr_leaf_to_shortform(struct xfs_buf *bp,
+int xfs_attr3_leaf_to_node(struct xfs_da_args *args);
+int xfs_attr3_leaf_to_shortform(struct xfs_buf *bp,
struct xfs_da_args *args, int forkoff);
-int xfs_attr_leaf_clearflag(struct xfs_da_args *args);
-int xfs_attr_leaf_setflag(struct xfs_da_args *args);
-int xfs_attr_leaf_flipflags(xfs_da_args_t *args);
+int xfs_attr3_leaf_clearflag(struct xfs_da_args *args);
+int xfs_attr3_leaf_setflag(struct xfs_da_args *args);
+int xfs_attr3_leaf_flipflags(struct xfs_da_args *args);
/*
* Routines used for growing the Btree.
*/
-int xfs_attr_leaf_split(struct xfs_da_state *state,
+int xfs_attr3_leaf_split(struct xfs_da_state *state,
struct xfs_da_state_blk *oldblk,
struct xfs_da_state_blk *newblk);
-int xfs_attr_leaf_lookup_int(struct xfs_buf *leaf,
+int xfs_attr3_leaf_lookup_int(struct xfs_buf *leaf,
struct xfs_da_args *args);
-int xfs_attr_leaf_getvalue(struct xfs_buf *bp, struct xfs_da_args *args);
-int xfs_attr_leaf_add(struct xfs_buf *leaf_buffer,
+int xfs_attr3_leaf_getvalue(struct xfs_buf *bp, struct xfs_da_args *args);
+int xfs_attr3_leaf_add(struct xfs_buf *leaf_buffer,
struct xfs_da_args *args);
-int xfs_attr_leaf_remove(struct xfs_buf *leaf_buffer,
+int xfs_attr3_leaf_remove(struct xfs_buf *leaf_buffer,
struct xfs_da_args *args);
-int xfs_attr_leaf_list_int(struct xfs_buf *bp,
+int xfs_attr3_leaf_list_int(struct xfs_buf *bp,
struct xfs_attr_list_context *context);
/*
* Routines used for shrinking the Btree.
*/
-int xfs_attr_leaf_toosmall(struct xfs_da_state *state, int *retval);
-void xfs_attr_leaf_unbalance(struct xfs_da_state *state,
+int xfs_attr3_leaf_toosmall(struct xfs_da_state *state, int *retval);
+void xfs_attr3_leaf_unbalance(struct xfs_da_state *state,
struct xfs_da_state_blk *drop_blk,
struct xfs_da_state_blk *save_blk);
-int xfs_attr_root_inactive(struct xfs_trans **trans, struct xfs_inode *dp);
+int xfs_attr3_root_inactive(struct xfs_trans **trans, struct xfs_inode *dp);
/*
* Utility routines.
@@ -261,10 +327,12 @@ int xfs_attr_leaf_order(struct xfs_buf *leaf1_bp,
struct xfs_buf *leaf2_bp);
int xfs_attr_leaf_newentsize(int namelen, int valuelen, int blocksize,
int *local);
-int xfs_attr_leaf_read(struct xfs_trans *tp, struct xfs_inode *dp,
+int xfs_attr3_leaf_read(struct xfs_trans *tp, struct xfs_inode *dp,
xfs_dablk_t bno, xfs_daddr_t mappedbno,
struct xfs_buf **bpp);
+void xfs_attr3_leaf_hdr_from_disk(struct xfs_attr3_icleaf_hdr *to,
+ struct xfs_attr_leafblock *from);
-extern const struct xfs_buf_ops xfs_attr_leaf_buf_ops;
+extern const struct xfs_buf_ops xfs_attr3_leaf_buf_ops;
#endif /* __XFS_ATTR_LEAF_H__ */
diff --git a/include/xfs_da_btree.h b/include/xfs_da_btree.h
index 6bedb3c..0e8182c 100644
--- a/include/xfs_da_btree.h
+++ b/include/xfs_da_btree.h
@@ -55,6 +55,7 @@ typedef struct xfs_da_blkinfo {
* magic numbers without modification for both v2 and v3 nodes.
*/
#define XFS_DA3_NODE_MAGIC 0x3ebe /* magic number: non-leaf blocks */
+#define XFS_ATTR3_LEAF_MAGIC 0x3bee /* magic number: attribute leaf blks */
#define XFS_DIR3_LEAF1_MAGIC 0x3df1 /* magic number: v2 dirlf single blks */
#define XFS_DIR3_LEAFN_MAGIC 0x3dff /* magic number: v2 dirlf multi blks */
@@ -85,13 +86,13 @@ struct xfs_da3_blkinfo {
typedef struct xfs_da_node_hdr {
struct xfs_da_blkinfo info; /* block type, links, etc. */
- __be16 count; /* count of active entries */
+ __be16 __count; /* count of active entries */
__be16 __level; /* level above leaves (leaf == 0) */
} xfs_da_node_hdr_t;
struct xfs_da3_node_hdr {
struct xfs_da3_blkinfo info; /* block type, links, etc. */
- __be16 count; /* count of active entries */
+ __be16 __count; /* count of active entries */
__be16 __level; /* level above leaves (leaf == 0) */
__be32 __pad32;
};
diff --git a/libxfs/xfs_attr.c b/libxfs/xfs_attr.c
index bb2ccf2..4429cb7 100644
--- a/libxfs/xfs_attr.c
+++ b/libxfs/xfs_attr.c
@@ -659,7 +659,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
*/
dp = args->dp;
args->blkno = 0;
- error = xfs_attr_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+ error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
if (error)
return error;
@@ -667,14 +667,14 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
* Look up the given attribute in the leaf block. Figure out if
* the given flags produce an error or call for an atomic rename.
*/
- retval = xfs_attr_leaf_lookup_int(bp, args);
+ retval = xfs_attr3_leaf_lookup_int(bp, args);
if ((args->flags & ATTR_REPLACE) && (retval == ENOATTR)) {
xfs_trans_brelse(args->trans, bp);
- return(retval);
+ return retval;
} else if (retval == EEXIST) {
if (args->flags & ATTR_CREATE) { /* pure create op */
xfs_trans_brelse(args->trans, bp);
- return(retval);
+ return retval;
}
trace_xfs_attr_leaf_replace(args);
@@ -690,7 +690,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
* Add the attribute to the leaf block, transitioning to a Btree
* if required.
*/
- retval = xfs_attr_leaf_add(bp, args);
+ retval = xfs_attr3_leaf_add(bp, args);
if (retval == ENOSPC) {
/*
* Promote the attribute list to the Btree format, then
@@ -698,7 +698,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
* can manage its own transactions.
*/
xfs_bmap_init(args->flist, args->firstblock);
- error = xfs_attr_leaf_to_node(args);
+ error = xfs_attr3_leaf_to_node(args);
if (!error) {
error = xfs_bmap_finish(&args->trans, args->flist,
&committed);
@@ -763,7 +763,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
* In a separate transaction, set the incomplete flag on the
* "old" attr and clear the incomplete flag on the "new" attr.
*/
- error = xfs_attr_leaf_flipflags(args);
+ error = xfs_attr3_leaf_flipflags(args);
if (error)
return(error);
@@ -785,19 +785,19 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
* Read in the block containing the "old" attr, then
* remove the "old" attr from that block (neat, huh!)
*/
- error = xfs_attr_leaf_read(args->trans, args->dp, args->blkno,
+ error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno,
-1, &bp);
if (error)
return error;
- xfs_attr_leaf_remove(bp, args);
+ xfs_attr3_leaf_remove(bp, args);
/*
* If the result is small enough, shrink it all into the inode.
*/
if ((forkoff = xfs_attr_shortform_allfit(bp, dp))) {
xfs_bmap_init(args->flist, args->firstblock);
- error = xfs_attr_leaf_to_shortform(bp, args, forkoff);
+ error = xfs_attr3_leaf_to_shortform(bp, args, forkoff);
/* bp is gone due to xfs_da_shrink_inode */
if (!error) {
error = xfs_bmap_finish(&args->trans,
@@ -829,9 +829,9 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
/*
* Added a "remote" value, just clear the incomplete flag.
*/
- error = xfs_attr_leaf_clearflag(args);
+ error = xfs_attr3_leaf_clearflag(args);
}
- return(error);
+ return error;
}
/*
@@ -854,24 +854,24 @@ xfs_attr_leaf_removename(xfs_da_args_t *args)
*/
dp = args->dp;
args->blkno = 0;
- error = xfs_attr_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+ error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
if (error)
return error;
- error = xfs_attr_leaf_lookup_int(bp, args);
+ error = xfs_attr3_leaf_lookup_int(bp, args);
if (error == ENOATTR) {
xfs_trans_brelse(args->trans, bp);
return(error);
}
- xfs_attr_leaf_remove(bp, args);
+ xfs_attr3_leaf_remove(bp, args);
/*
* If the result is small enough, shrink it all into the inode.
*/
if ((forkoff = xfs_attr_shortform_allfit(bp, dp))) {
xfs_bmap_init(args->flist, args->firstblock);
- error = xfs_attr_leaf_to_shortform(bp, args, forkoff);
+ error = xfs_attr3_leaf_to_shortform(bp, args, forkoff);
/* bp is gone due to xfs_da_shrink_inode */
if (!error) {
error = xfs_bmap_finish(&args->trans, args->flist,
@@ -881,7 +881,7 @@ xfs_attr_leaf_removename(xfs_da_args_t *args)
ASSERT(committed);
args->trans = NULL;
xfs_bmap_cancel(args->flist);
- return(error);
+ return error;
}
/*
@@ -891,7 +891,7 @@ xfs_attr_leaf_removename(xfs_da_args_t *args)
if (committed)
xfs_trans_ijoin(args->trans, dp, 0);
}
- return(0);
+ return 0;
}
/*
@@ -909,21 +909,21 @@ xfs_attr_leaf_get(xfs_da_args_t *args)
trace_xfs_attr_leaf_get(args);
args->blkno = 0;
- error = xfs_attr_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+ error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
if (error)
return error;
- error = xfs_attr_leaf_lookup_int(bp, args);
+ error = xfs_attr3_leaf_lookup_int(bp, args);
if (error != EEXIST) {
xfs_trans_brelse(args->trans, bp);
- return(error);
+ return error;
}
- error = xfs_attr_leaf_getvalue(bp, args);
+ error = xfs_attr3_leaf_getvalue(bp, args);
xfs_trans_brelse(args->trans, bp);
if (!error && (args->rmtblkno > 0) && !(args->flags & ATTR_KERNOVAL)) {
error = xfs_attr_rmtval_get(args);
}
- return(error);
+ return error;
}
/*========================================================================
@@ -989,7 +989,7 @@ restart:
args->rmtblkcnt = 0;
}
- retval = xfs_attr_leaf_add(blk->bp, state->args);
+ retval = xfs_attr3_leaf_add(blk->bp, state->args);
if (retval == ENOSPC) {
if (state->path.active == 1) {
/*
@@ -999,7 +999,7 @@ restart:
*/
xfs_da_state_free(state);
xfs_bmap_init(args->flist, args->firstblock);
- error = xfs_attr_leaf_to_node(args);
+ error = xfs_attr3_leaf_to_node(args);
if (!error) {
error = xfs_bmap_finish(&args->trans,
args->flist,
@@ -1101,7 +1101,7 @@ restart:
* In a separate transaction, set the incomplete flag on the
* "old" attr and clear the incomplete flag on the "new" attr.
*/
- error = xfs_attr_leaf_flipflags(args);
+ error = xfs_attr3_leaf_flipflags(args);
if (error)
goto out;
@@ -1140,7 +1140,7 @@ restart:
*/
blk = &state->path.blk[ state->path.active-1 ];
ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC);
- error = xfs_attr_leaf_remove(blk->bp, args);
+ error = xfs_attr3_leaf_remove(blk->bp, args);
xfs_da3_fixhashpath(state, &state->path);
/*
@@ -1181,7 +1181,7 @@ restart:
/*
* Added a "remote" value, just clear the incomplete flag.
*/
- error = xfs_attr_leaf_clearflag(args);
+ error = xfs_attr3_leaf_clearflag(args);
if (error)
goto out;
}
@@ -1255,7 +1255,7 @@ xfs_attr_node_removename(xfs_da_args_t *args)
* Mark the attribute as INCOMPLETE, then bunmapi() the
* remote value.
*/
- error = xfs_attr_leaf_setflag(args);
+ error = xfs_attr3_leaf_setflag(args);
if (error)
goto out;
error = xfs_attr_rmtval_remove(args);
@@ -1276,7 +1276,7 @@ xfs_attr_node_removename(xfs_da_args_t *args)
*/
blk = &state->path.blk[ state->path.active-1 ];
ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC);
- retval = xfs_attr_leaf_remove(blk->bp, args);
+ retval = xfs_attr3_leaf_remove(blk->bp, args);
xfs_da3_fixhashpath(state, &state->path);
/*
@@ -1322,13 +1322,13 @@ xfs_attr_node_removename(xfs_da_args_t *args)
ASSERT(state->path.blk[0].bp);
state->path.blk[0].bp = NULL;
- error = xfs_attr_leaf_read(args->trans, args->dp, 0, -1, &bp);
+ error = xfs_attr3_leaf_read(args->trans, args->dp, 0, -1, &bp);
if (error)
goto out;
if ((forkoff = xfs_attr_shortform_allfit(bp, dp))) {
xfs_bmap_init(args->flist, args->firstblock);
- error = xfs_attr_leaf_to_shortform(bp, args, forkoff);
+ error = xfs_attr3_leaf_to_shortform(bp, args, forkoff);
/* bp is gone due to xfs_da_shrink_inode */
if (!error) {
error = xfs_bmap_finish(&args->trans,
@@ -1500,7 +1500,7 @@ xfs_attr_node_get(xfs_da_args_t *args)
/*
* Get the value, local or "remote"
*/
- retval = xfs_attr_leaf_getvalue(blk->bp, args);
+ retval = xfs_attr3_leaf_getvalue(blk->bp, args);
if (!retval && (args->rmtblkno > 0)
&& !(args->flags & ATTR_KERNOVAL)) {
retval = xfs_attr_rmtval_get(args);
diff --git a/libxfs/xfs_attr_leaf.c b/libxfs/xfs_attr_leaf.c
index cb37198..9de2244 100644
--- a/libxfs/xfs_attr_leaf.c
+++ b/libxfs/xfs_attr_leaf.c
@@ -1,5 +1,6 @@
/*
* Copyright (c) 2000-2005 Silicon Graphics, Inc.
+ * Copyright (c) 2013 Red Hat, Inc.
* All Rights Reserved.
*
* This program is free software; you can redistribute it and/or
@@ -31,68 +32,204 @@
/*
* Routines used for growing the Btree.
*/
-STATIC int xfs_attr_leaf_create(xfs_da_args_t *args, xfs_dablk_t which_block,
- struct xfs_buf **bpp);
-STATIC int xfs_attr_leaf_add_work(struct xfs_buf *leaf_buffer,
- xfs_da_args_t *args, int freemap_index);
-STATIC void xfs_attr_leaf_compact(struct xfs_da_args *args,
- struct xfs_buf *leaf_buffer);
-STATIC void xfs_attr_leaf_rebalance(xfs_da_state_t *state,
+STATIC int xfs_attr3_leaf_create(struct xfs_da_args *args,
+ xfs_dablk_t which_block, struct xfs_buf **bpp);
+STATIC int xfs_attr3_leaf_add_work(struct xfs_buf *leaf_buffer,
+ struct xfs_attr3_icleaf_hdr *ichdr,
+ struct xfs_da_args *args, int freemap_index);
+STATIC void xfs_attr3_leaf_compact(struct xfs_da_args *args,
+ struct xfs_attr3_icleaf_hdr *ichdr,
+ struct xfs_buf *leaf_buffer);
+STATIC void xfs_attr3_leaf_rebalance(xfs_da_state_t *state,
xfs_da_state_blk_t *blk1,
xfs_da_state_blk_t *blk2);
-STATIC int xfs_attr_leaf_figure_balance(xfs_da_state_t *state,
- xfs_da_state_blk_t *leaf_blk_1,
- xfs_da_state_blk_t *leaf_blk_2,
- int *number_entries_in_blk1,
- int *number_usedbytes_in_blk1);
+STATIC int xfs_attr3_leaf_figure_balance(xfs_da_state_t *state,
+ xfs_da_state_blk_t *leaf_blk_1,
+ struct xfs_attr3_icleaf_hdr *ichdr1,
+ xfs_da_state_blk_t *leaf_blk_2,
+ struct xfs_attr3_icleaf_hdr *ichdr2,
+ int *number_entries_in_blk1,
+ int *number_usedbytes_in_blk1);
/*
* Utility routines.
*/
-STATIC void xfs_attr_leaf_moveents(xfs_attr_leafblock_t *src_leaf,
- int src_start,
- xfs_attr_leafblock_t *dst_leaf,
- int dst_start, int move_count,
- xfs_mount_t *mp);
+STATIC void xfs_attr3_leaf_moveents(struct xfs_attr_leafblock *src_leaf,
+ struct xfs_attr3_icleaf_hdr *src_ichdr, int src_start,
+ struct xfs_attr_leafblock *dst_leaf,
+ struct xfs_attr3_icleaf_hdr *dst_ichdr, int dst_start,
+ int move_count, struct xfs_mount *mp);
STATIC int xfs_attr_leaf_entsize(xfs_attr_leafblock_t *leaf, int index);
-static void
-xfs_attr_leaf_verify(
+void
+xfs_attr3_leaf_hdr_from_disk(
+ struct xfs_attr3_icleaf_hdr *to,
+ struct xfs_attr_leafblock *from)
+{
+ int i;
+
+ ASSERT(from->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC) ||
+ from->hdr.info.magic == cpu_to_be16(XFS_ATTR3_LEAF_MAGIC));
+
+ if (from->hdr.info.magic == cpu_to_be16(XFS_ATTR3_LEAF_MAGIC)) {
+ struct xfs_attr3_leaf_hdr *hdr3 = (struct xfs_attr3_leaf_hdr *)from;
+
+ to->forw = be32_to_cpu(hdr3->info.hdr.forw);
+ to->back = be32_to_cpu(hdr3->info.hdr.back);
+ to->magic = be16_to_cpu(hdr3->info.hdr.magic);
+ to->count = be16_to_cpu(hdr3->count);
+ to->usedbytes = be16_to_cpu(hdr3->usedbytes);
+ to->firstused = be16_to_cpu(hdr3->firstused);
+ to->holes = hdr3->holes;
+
+ for (i = 0; i < XFS_ATTR_LEAF_MAPSIZE; i++) {
+ to->freemap[i].base = be16_to_cpu(hdr3->freemap[i].base);
+ to->freemap[i].size = be16_to_cpu(hdr3->freemap[i].size);
+ }
+ return;
+ }
+ to->forw = be32_to_cpu(from->hdr.info.forw);
+ to->back = be32_to_cpu(from->hdr.info.back);
+ to->magic = be16_to_cpu(from->hdr.info.magic);
+ to->count = be16_to_cpu(from->hdr.count);
+ to->usedbytes = be16_to_cpu(from->hdr.usedbytes);
+ to->firstused = be16_to_cpu(from->hdr.firstused);
+ to->holes = from->hdr.holes;
+
+ for (i = 0; i < XFS_ATTR_LEAF_MAPSIZE; i++) {
+ to->freemap[i].base = be16_to_cpu(from->hdr.freemap[i].base);
+ to->freemap[i].size = be16_to_cpu(from->hdr.freemap[i].size);
+ }
+}
+
+void
+xfs_attr3_leaf_hdr_to_disk(
+ struct xfs_attr_leafblock *to,
+ struct xfs_attr3_icleaf_hdr *from)
+{
+ int i;
+
+ ASSERT(from->magic == XFS_ATTR_LEAF_MAGIC ||
+ from->magic == XFS_ATTR3_LEAF_MAGIC);
+
+ if (from->magic == XFS_ATTR3_LEAF_MAGIC) {
+ struct xfs_attr3_leaf_hdr *hdr3 = (struct xfs_attr3_leaf_hdr *)to;
+
+ hdr3->info.hdr.forw = cpu_to_be32(from->forw);
+ hdr3->info.hdr.back = cpu_to_be32(from->back);
+ hdr3->info.hdr.magic = cpu_to_be16(from->magic);
+ hdr3->count = cpu_to_be16(from->count);
+ hdr3->usedbytes = cpu_to_be16(from->usedbytes);
+ hdr3->firstused = cpu_to_be16(from->firstused);
+ hdr3->holes = from->holes;
+ hdr3->pad1 = 0;
+
+ for (i = 0; i < XFS_ATTR_LEAF_MAPSIZE; i++) {
+ hdr3->freemap[i].base = cpu_to_be16(from->freemap[i].base);
+ hdr3->freemap[i].size = cpu_to_be16(from->freemap[i].size);
+ }
+ return;
+ }
+ to->hdr.info.forw = cpu_to_be32(from->forw);
+ to->hdr.info.back = cpu_to_be32(from->back);
+ to->hdr.info.magic = cpu_to_be16(from->magic);
+ to->hdr.count = cpu_to_be16(from->count);
+ to->hdr.usedbytes = cpu_to_be16(from->usedbytes);
+ to->hdr.firstused = cpu_to_be16(from->firstused);
+ to->hdr.holes = from->holes;
+ to->hdr.pad1 = 0;
+
+ for (i = 0; i < XFS_ATTR_LEAF_MAPSIZE; i++) {
+ to->hdr.freemap[i].base = cpu_to_be16(from->freemap[i].base);
+ to->hdr.freemap[i].size = cpu_to_be16(from->freemap[i].size);
+ }
+}
+
+static bool
+xfs_attr3_leaf_verify(
struct xfs_buf *bp)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
- struct xfs_attr_leaf_hdr *hdr = bp->b_addr;
- int block_ok = 0;
+ struct xfs_attr_leafblock *leaf = bp->b_addr;
+ struct xfs_attr3_icleaf_hdr ichdr;
- block_ok = hdr->info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC);
- if (!block_ok) {
- XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, hdr);
- xfs_buf_ioerror(bp, EFSCORRUPTED);
+ xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf);
+
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ struct xfs_da3_node_hdr *hdr3 = bp->b_addr;
+
+ if (ichdr.magic != XFS_ATTR3_LEAF_MAGIC)
+ return false;
+
+ if (!uuid_equal(&hdr3->info.uuid, &mp->m_sb.sb_uuid))
+ return false;
+ if (be64_to_cpu(hdr3->info.blkno) != bp->b_bn)
+ return false;
+ } else {
+ if (ichdr.magic != XFS_ATTR_LEAF_MAGIC)
+ return false;
}
+ if (ichdr.count == 0)
+ return false;
+
+ /* XXX: need to range check rest of attr header values */
+ /* XXX: hash order check? */
+
+ return true;
}
static void
-xfs_attr_leaf_read_verify(
+xfs_attr3_leaf_write_verify(
struct xfs_buf *bp)
{
- xfs_attr_leaf_verify(bp);
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_buf_log_item *bip = bp->b_fspriv;
+ struct xfs_attr3_leaf_hdr *hdr3 = bp->b_addr;
+
+ if (!xfs_attr3_leaf_verify(bp)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ return;
+ }
+
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return;
+
+ if (bip)
+ hdr3->info.lsn = cpu_to_be64(bip->bli_item.li_lsn);
+
+ xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), XFS_ATTR3_LEAF_CRC_OFF);
}
+/*
+ * leaf/node format detection on trees is sketchy, so a node read can be done on
+ * leaf level blocks when detection identifies the tree as a node format tree
+ * incorrectly. In this case, we need to swap the verifier to match the correct
+ * format of the block being read.
+ */
static void
-xfs_attr_leaf_write_verify(
- struct xfs_buf *bp)
+xfs_attr3_leaf_read_verify(
+ struct xfs_buf *bp)
{
- xfs_attr_leaf_verify(bp);
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+
+ if ((xfs_sb_version_hascrc(&mp->m_sb) &&
+ !xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
+ XFS_ATTR3_LEAF_CRC_OFF)) ||
+ !xfs_attr3_leaf_verify(bp)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
}
-const struct xfs_buf_ops xfs_attr_leaf_buf_ops = {
- .verify_read = xfs_attr_leaf_read_verify,
- .verify_write = xfs_attr_leaf_write_verify,
+const struct xfs_buf_ops xfs_attr3_leaf_buf_ops = {
+ .verify_read = xfs_attr3_leaf_read_verify,
+ .verify_write = xfs_attr3_leaf_write_verify,
};
int
-xfs_attr_leaf_read(
+xfs_attr3_leaf_read(
struct xfs_trans *tp,
struct xfs_inode *dp,
xfs_dablk_t bno,
@@ -100,7 +237,7 @@ xfs_attr_leaf_read(
struct xfs_buf **bpp)
{
return xfs_da_read_buf(tp, dp, bno, mappedbno, bpp,
- XFS_ATTR_FORK, &xfs_attr_leaf_buf_ops);
+ XFS_ATTR_FORK, &xfs_attr3_leaf_buf_ops);
}
/*========================================================================
@@ -528,7 +665,7 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
}
ASSERT(blkno == 0);
- error = xfs_attr_leaf_create(args, blkno, &bp);
+ error = xfs_attr3_leaf_create(args, blkno, &bp);
if (error) {
error = xfs_da_shrink_inode(args, 0, bp);
bp = NULL;
@@ -557,9 +694,9 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
nargs.hashval = xfs_da_hashname(sfe->nameval,
sfe->namelen);
nargs.flags = XFS_ATTR_NSP_ONDISK_TO_ARGS(sfe->flags);
- error = xfs_attr_leaf_lookup_int(bp, &nargs); /* set a->index */
+ error = xfs_attr3_leaf_lookup_int(bp, &nargs); /* set a->index */
ASSERT(error == ENOATTR);
- error = xfs_attr_leaf_add(bp, &nargs);
+ error = xfs_attr3_leaf_add(bp, &nargs);
ASSERT(error != ENOSPC);
if (error)
goto out;
@@ -596,7 +733,7 @@ xfs_attr_shortform_allfit(
continue; /* don't copy partial entries */
if (!(entry->flags & XFS_ATTR_LOCAL))
return(0);
- name_loc = xfs_attr_leaf_name_local(leaf, i);
+ name_loc = xfs_attr3_leaf_name_local(leaf, i);
if (name_loc->namelen >= XFS_ATTR_SF_ENTSIZE_MAX)
return(0);
if (be16_to_cpu(name_loc->valuelen) >= XFS_ATTR_SF_ENTSIZE_MAX)
@@ -616,29 +753,34 @@ xfs_attr_shortform_allfit(
* Convert a leaf attribute list to shortform attribute list
*/
int
-xfs_attr_leaf_to_shortform(
- struct xfs_buf *bp,
- xfs_da_args_t *args,
- int forkoff)
+xfs_attr3_leaf_to_shortform(
+ struct xfs_buf *bp,
+ struct xfs_da_args *args,
+ int forkoff)
{
- xfs_attr_leafblock_t *leaf;
- xfs_attr_leaf_entry_t *entry;
- xfs_attr_leaf_name_local_t *name_loc;
- xfs_da_args_t nargs;
- xfs_inode_t *dp;
- char *tmpbuffer;
- int error, i;
+ struct xfs_attr_leafblock *leaf;
+ struct xfs_attr3_icleaf_hdr ichdr;
+ struct xfs_attr_leaf_entry *entry;
+ struct xfs_attr_leaf_name_local *name_loc;
+ struct xfs_da_args nargs;
+ struct xfs_inode *dp = args->dp;
+ char *tmpbuffer;
+ int error;
+ int i;
trace_xfs_attr_leaf_to_sf(args);
- dp = args->dp;
tmpbuffer = kmem_alloc(XFS_LBSIZE(dp->i_mount), KM_SLEEP);
- ASSERT(tmpbuffer != NULL);
+ if (!tmpbuffer)
+ return ENOMEM;
- ASSERT(bp != NULL);
memcpy(tmpbuffer, bp->b_addr, XFS_LBSIZE(dp->i_mount));
+
leaf = (xfs_attr_leafblock_t *)tmpbuffer;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
+ xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf);
+ entry = xfs_attr3_leaf_entryp(leaf);
+
+ /* XXX (dgc): buffer is about to be marked stale - why zero it? */
memset(bp->b_addr, 0, XFS_LBSIZE(dp->i_mount));
/*
@@ -668,14 +810,14 @@ xfs_attr_leaf_to_shortform(
nargs.whichfork = XFS_ATTR_FORK;
nargs.trans = args->trans;
nargs.op_flags = XFS_DA_OP_OKNOENT;
- entry = &leaf->entries[0];
- for (i = 0; i < be16_to_cpu(leaf->hdr.count); entry++, i++) {
+
+ for (i = 0; i < ichdr.count; entry++, i++) {
if (entry->flags & XFS_ATTR_INCOMPLETE)
continue; /* don't copy partial entries */
if (!entry->nameidx)
continue;
ASSERT(entry->flags & XFS_ATTR_LOCAL);
- name_loc = xfs_attr_leaf_name_local(leaf, i);
+ name_loc = xfs_attr3_leaf_name_local(leaf, i);
nargs.name = name_loc->nameval;
nargs.namelen = name_loc->namelen;
nargs.value = &name_loc->nameval[nargs.namelen];
@@ -688,43 +830,50 @@ xfs_attr_leaf_to_shortform(
out:
kmem_free(tmpbuffer);
- return(error);
+ return error;
}
/*
* Convert from using a single leaf to a root node and a leaf.
*/
int
-xfs_attr_leaf_to_node(xfs_da_args_t *args)
+xfs_attr3_leaf_to_node(
+ struct xfs_da_args *args)
{
- xfs_attr_leafblock_t *leaf;
- xfs_da_intnode_t *node;
- xfs_inode_t *dp;
- struct xfs_buf *bp1, *bp2;
- xfs_dablk_t blkno;
- int error;
+ struct xfs_attr_leafblock *leaf;
+ struct xfs_attr3_icleaf_hdr icleafhdr;
+ struct xfs_attr_leaf_entry *entries;
struct xfs_da_node_entry *btree;
+ struct xfs_da3_icnode_hdr icnodehdr;
+ struct xfs_da_intnode *node;
+ struct xfs_inode *dp = args->dp;
+ struct xfs_mount *mp = dp->i_mount;
+ struct xfs_buf *bp1 = NULL;
+ struct xfs_buf *bp2 = NULL;
+ xfs_dablk_t blkno;
+ int error;
trace_xfs_attr_leaf_to_node(args);
- dp = args->dp;
- bp1 = bp2 = NULL;
error = xfs_da_grow_inode(args, &blkno);
if (error)
goto out;
- error = xfs_attr_leaf_read(args->trans, args->dp, 0, -1, &bp1);
+ error = xfs_attr3_leaf_read(args->trans, dp, 0, -1, &bp1);
if (error)
goto out;
- bp2 = NULL;
- error = xfs_da_get_buf(args->trans, args->dp, blkno, -1, &bp2,
- XFS_ATTR_FORK);
+ error = xfs_da_get_buf(args->trans, dp, blkno, -1, &bp2, XFS_ATTR_FORK);
if (error)
goto out;
+
+ /* copy leaf to new buffer, update identifiers */
bp2->b_ops = bp1->b_ops;
- memcpy(bp2->b_addr, bp1->b_addr, XFS_LBSIZE(dp->i_mount));
- bp1 = NULL;
- xfs_trans_log_buf(args->trans, bp2, 0, XFS_LBSIZE(dp->i_mount) - 1);
+ memcpy(bp2->b_addr, bp1->b_addr, XFS_LBSIZE(mp));
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ struct xfs_da3_blkinfo *hdr3 = bp2->b_addr;
+ hdr3->blkno = cpu_to_be64(bp2->b_bn);
+ }
+ xfs_trans_log_buf(args->trans, bp2, 0, XFS_LBSIZE(mp) - 1);
/*
* Set up the new root node.
@@ -733,17 +882,22 @@ xfs_attr_leaf_to_node(xfs_da_args_t *args)
if (error)
goto out;
node = bp1->b_addr;
+ xfs_da3_node_hdr_from_disk(&icnodehdr, node);
+ btree = xfs_da3_node_tree_p(node);
+
leaf = bp2->b_addr;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
+ xfs_attr3_leaf_hdr_from_disk(&icleafhdr, leaf);
+ entries = xfs_attr3_leaf_entryp(leaf);
+
/* both on-disk, don't endian-flip twice */
- btree = xfs_da3_node_tree_p(node);
- btree[0].hashval = leaf->entries[be16_to_cpu(leaf->hdr.count)-1 ].hashval;
+ btree[0].hashval = entries[icleafhdr.count - 1].hashval;
btree[0].before = cpu_to_be32(blkno);
- node->hdr.count = cpu_to_be16(1);
- xfs_trans_log_buf(args->trans, bp1, 0, XFS_LBSIZE(dp->i_mount) - 1);
+ icnodehdr.count = 1;
+ xfs_da3_node_hdr_to_disk(node, &icnodehdr);
+ xfs_trans_log_buf(args->trans, bp1, 0, XFS_LBSIZE(mp) - 1);
error = 0;
out:
- return(error);
+ return error;
}
@@ -756,52 +910,62 @@ out:
* or a leaf in a node attribute list.
*/
STATIC int
-xfs_attr_leaf_create(
- xfs_da_args_t *args,
- xfs_dablk_t blkno,
- struct xfs_buf **bpp)
+xfs_attr3_leaf_create(
+ struct xfs_da_args *args,
+ xfs_dablk_t blkno,
+ struct xfs_buf **bpp)
{
- xfs_attr_leafblock_t *leaf;
- xfs_attr_leaf_hdr_t *hdr;
- xfs_inode_t *dp;
- struct xfs_buf *bp;
- int error;
+ struct xfs_attr_leafblock *leaf;
+ struct xfs_attr3_icleaf_hdr ichdr;
+ struct xfs_inode *dp = args->dp;
+ struct xfs_mount *mp = dp->i_mount;
+ struct xfs_buf *bp;
+ int error;
trace_xfs_attr_leaf_create(args);
- dp = args->dp;
- ASSERT(dp != NULL);
error = xfs_da_get_buf(args->trans, args->dp, blkno, -1, &bp,
XFS_ATTR_FORK);
if (error)
- return(error);
- bp->b_ops = &xfs_attr_leaf_buf_ops;
+ return error;
+ bp->b_ops = &xfs_attr3_leaf_buf_ops;
leaf = bp->b_addr;
- memset((char *)leaf, 0, XFS_LBSIZE(dp->i_mount));
- hdr = &leaf->hdr;
- hdr->info.magic = cpu_to_be16(XFS_ATTR_LEAF_MAGIC);
- hdr->firstused = cpu_to_be16(XFS_LBSIZE(dp->i_mount));
- if (!hdr->firstused) {
- hdr->firstused = cpu_to_be16(
- XFS_LBSIZE(dp->i_mount) - XFS_ATTR_LEAF_NAME_ALIGN);
- }
+ memset(leaf, 0, XFS_LBSIZE(mp));
+
+ memset(&ichdr, 0, sizeof(ichdr));
+ ichdr.firstused = XFS_LBSIZE(mp);
+
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ struct xfs_da3_blkinfo *hdr3 = bp->b_addr;
- hdr->freemap[0].base = cpu_to_be16(sizeof(xfs_attr_leaf_hdr_t));
- hdr->freemap[0].size = cpu_to_be16(be16_to_cpu(hdr->firstused) -
- sizeof(xfs_attr_leaf_hdr_t));
+ ichdr.magic = XFS_ATTR3_LEAF_MAGIC;
+
+ hdr3->blkno = cpu_to_be64(bp->b_bn);
+ hdr3->owner = cpu_to_be64(dp->i_ino);
+ uuid_copy(&hdr3->uuid, &mp->m_sb.sb_uuid);
+
+ ichdr.freemap[0].base = sizeof(struct xfs_attr3_leaf_hdr);
+ } else {
+ ichdr.magic = XFS_ATTR_LEAF_MAGIC;
+ ichdr.freemap[0].base = sizeof(struct xfs_attr_leaf_hdr);
+ }
+ ichdr.freemap[0].size = ichdr.firstused - ichdr.freemap[0].base;
- xfs_trans_log_buf(args->trans, bp, 0, XFS_LBSIZE(dp->i_mount) - 1);
+ xfs_attr3_leaf_hdr_to_disk(leaf, &ichdr);
+ xfs_trans_log_buf(args->trans, bp, 0, XFS_LBSIZE(mp) - 1);
*bpp = bp;
- return(0);
+ return 0;
}
/*
* Split the leaf node, rebalance, then add the new entry.
*/
int
-xfs_attr_leaf_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
- xfs_da_state_blk_t *newblk)
+xfs_attr3_leaf_split(
+ struct xfs_da_state *state,
+ struct xfs_da_state_blk *oldblk,
+ struct xfs_da_state_blk *newblk)
{
xfs_dablk_t blkno;
int error;
@@ -815,7 +979,7 @@ xfs_attr_leaf_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
error = xfs_da_grow_inode(state->args, &blkno);
if (error)
return(error);
- error = xfs_attr_leaf_create(state->args, blkno, &newblk->bp);
+ error = xfs_attr3_leaf_create(state->args, blkno, &newblk->bp);
if (error)
return(error);
newblk->blkno = blkno;
@@ -825,7 +989,7 @@ xfs_attr_leaf_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
* Rebalance the entries across the two leaves.
* NOTE: rebalance() currently depends on the 2nd block being empty.
*/
- xfs_attr_leaf_rebalance(state, oldblk, newblk);
+ xfs_attr3_leaf_rebalance(state, oldblk, newblk);
error = xfs_da3_blk_link(state, oldblk, newblk);
if (error)
return(error);
@@ -839,10 +1003,10 @@ xfs_attr_leaf_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
*/
if (state->inleaf) {
trace_xfs_attr_leaf_add_old(state->args);
- error = xfs_attr_leaf_add(oldblk->bp, state->args);
+ error = xfs_attr3_leaf_add(oldblk->bp, state->args);
} else {
trace_xfs_attr_leaf_add_new(state->args);
- error = xfs_attr_leaf_add(newblk->bp, state->args);
+ error = xfs_attr3_leaf_add(newblk->bp, state->args);
}
/*
@@ -857,22 +1021,23 @@ xfs_attr_leaf_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
* Add a name to the leaf attribute list structure.
*/
int
-xfs_attr_leaf_add(
+xfs_attr3_leaf_add(
struct xfs_buf *bp,
struct xfs_da_args *args)
{
- xfs_attr_leafblock_t *leaf;
- xfs_attr_leaf_hdr_t *hdr;
- xfs_attr_leaf_map_t *map;
- int tablesize, entsize, sum, tmp, i;
+ struct xfs_attr_leafblock *leaf;
+ struct xfs_attr3_icleaf_hdr ichdr;
+ int tablesize;
+ int entsize;
+ int sum;
+ int tmp;
+ int i;
trace_xfs_attr_leaf_add(args);
leaf = bp->b_addr;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
- ASSERT((args->index >= 0)
- && (args->index <= be16_to_cpu(leaf->hdr.count)));
- hdr = &leaf->hdr;
+ xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf);
+ ASSERT(args->index >= 0 && args->index <= ichdr.count);
entsize = xfs_attr_leaf_newentsize(args->namelen, args->valuelen,
args->trans->t_mountp->m_sb.sb_blocksize, NULL);
@@ -880,25 +1045,23 @@ xfs_attr_leaf_add(
* Search through freemap for first-fit on new name length.
* (may need to figure in size of entry struct too)
*/
- tablesize = (be16_to_cpu(hdr->count) + 1)
- * sizeof(xfs_attr_leaf_entry_t)
- + sizeof(xfs_attr_leaf_hdr_t);
- map = &hdr->freemap[XFS_ATTR_LEAF_MAPSIZE-1];
- for (sum = 0, i = XFS_ATTR_LEAF_MAPSIZE-1; i >= 0; map--, i--) {
- if (tablesize > be16_to_cpu(hdr->firstused)) {
- sum += be16_to_cpu(map->size);
+ tablesize = (ichdr.count + 1) * sizeof(xfs_attr_leaf_entry_t)
+ + xfs_attr3_leaf_hdr_size(leaf);
+ for (sum = 0, i = XFS_ATTR_LEAF_MAPSIZE - 1; i >= 0; i--) {
+ if (tablesize > ichdr.firstused) {
+ sum += ichdr.freemap[i].size;
continue;
}
- if (!map->size)
+ if (!ichdr.freemap[i].size)
continue; /* no space in this map */
tmp = entsize;
- if (be16_to_cpu(map->base) < be16_to_cpu(hdr->firstused))
+ if (ichdr.freemap[i].base < ichdr.firstused)
tmp += sizeof(xfs_attr_leaf_entry_t);
- if (be16_to_cpu(map->size) >= tmp) {
- tmp = xfs_attr_leaf_add_work(bp, args, i);
- return(tmp);
+ if (ichdr.freemap[i].size >= tmp) {
+ tmp = xfs_attr3_leaf_add_work(bp, &ichdr, args, i);
+ goto out_log_hdr;
}
- sum += be16_to_cpu(map->size);
+ sum += ichdr.freemap[i].size;
}
/*
@@ -906,82 +1069,90 @@ xfs_attr_leaf_add(
* and we don't have enough freespace, then compaction will do us
* no good and we should just give up.
*/
- if (!hdr->holes && (sum < entsize))
- return(XFS_ERROR(ENOSPC));
+ if (!ichdr.holes && sum < entsize)
+ return XFS_ERROR(ENOSPC);
/*
* Compact the entries to coalesce free space.
* This may change the hdr->count via dropping INCOMPLETE entries.
*/
- xfs_attr_leaf_compact(args, bp);
+ xfs_attr3_leaf_compact(args, &ichdr, bp);
/*
* After compaction, the block is guaranteed to have only one
* free region, in freemap[0]. If it is not big enough, give up.
*/
- if (be16_to_cpu(hdr->freemap[0].size)
- < (entsize + sizeof(xfs_attr_leaf_entry_t)))
- return(XFS_ERROR(ENOSPC));
+ if (ichdr.freemap[0].size < (entsize + sizeof(xfs_attr_leaf_entry_t))) {
+ tmp = ENOSPC;
+ goto out_log_hdr;
+ }
+
+ tmp = xfs_attr3_leaf_add_work(bp, &ichdr, args, 0);
- return(xfs_attr_leaf_add_work(bp, args, 0));
+out_log_hdr:
+ xfs_attr3_leaf_hdr_to_disk(leaf, &ichdr);
+ xfs_trans_log_buf(args->trans, bp,
+ XFS_DA_LOGRANGE(leaf, &leaf->hdr,
+ xfs_attr3_leaf_hdr_size(leaf)));
+ return tmp;
}
/*
* Add a name to a leaf attribute list structure.
*/
STATIC int
-xfs_attr_leaf_add_work(
- struct xfs_buf *bp,
- xfs_da_args_t *args,
- int mapindex)
+xfs_attr3_leaf_add_work(
+ struct xfs_buf *bp,
+ struct xfs_attr3_icleaf_hdr *ichdr,
+ struct xfs_da_args *args,
+ int mapindex)
{
- xfs_attr_leafblock_t *leaf;
- xfs_attr_leaf_hdr_t *hdr;
- xfs_attr_leaf_entry_t *entry;
- xfs_attr_leaf_name_local_t *name_loc;
- xfs_attr_leaf_name_remote_t *name_rmt;
- xfs_attr_leaf_map_t *map;
- xfs_mount_t *mp;
- int tmp, i;
+ struct xfs_attr_leafblock *leaf;
+ struct xfs_attr_leaf_entry *entry;
+ struct xfs_attr_leaf_name_local *name_loc;
+ struct xfs_attr_leaf_name_remote *name_rmt;
+ struct xfs_attr_leaf_map *map;
+ struct xfs_mount *mp;
+ int tmp;
+ int i;
trace_xfs_attr_leaf_add_work(args);
leaf = bp->b_addr;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
- hdr = &leaf->hdr;
- ASSERT((mapindex >= 0) && (mapindex < XFS_ATTR_LEAF_MAPSIZE));
- ASSERT((args->index >= 0) && (args->index <= be16_to_cpu(hdr->count)));
+ ASSERT(mapindex >= 0 && mapindex < XFS_ATTR_LEAF_MAPSIZE);
+ ASSERT(args->index >= 0 && args->index <= ichdr->count);
/*
* Force open some space in the entry array and fill it in.
*/
- entry = &leaf->entries[args->index];
- if (args->index < be16_to_cpu(hdr->count)) {
- tmp = be16_to_cpu(hdr->count) - args->index;
+ entry = &xfs_attr3_leaf_entryp(leaf)[args->index];
+ if (args->index < ichdr->count) {
+ tmp = ichdr->count - args->index;
tmp *= sizeof(xfs_attr_leaf_entry_t);
- memmove((char *)(entry+1), (char *)entry, tmp);
+ memmove(entry + 1, entry, tmp);
xfs_trans_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, entry, tmp + sizeof(*entry)));
}
- be16_add_cpu(&hdr->count, 1);
+ ichdr->count++;
/*
* Allocate space for the new string (at the end of the run).
*/
- map = &hdr->freemap[mapindex];
mp = args->trans->t_mountp;
- ASSERT(be16_to_cpu(map->base) < XFS_LBSIZE(mp));
- ASSERT((be16_to_cpu(map->base) & 0x3) == 0);
- ASSERT(be16_to_cpu(map->size) >=
+ ASSERT(ichdr->freemap[mapindex].base < XFS_LBSIZE(mp));
+ ASSERT((ichdr->freemap[mapindex].base & 0x3) == 0);
+ ASSERT(ichdr->freemap[mapindex].size >=
xfs_attr_leaf_newentsize(args->namelen, args->valuelen,
mp->m_sb.sb_blocksize, NULL));
- ASSERT(be16_to_cpu(map->size) < XFS_LBSIZE(mp));
- ASSERT((be16_to_cpu(map->size) & 0x3) == 0);
- be16_add_cpu(&map->size,
- -xfs_attr_leaf_newentsize(args->namelen, args->valuelen,
- mp->m_sb.sb_blocksize, &tmp));
- entry->nameidx = cpu_to_be16(be16_to_cpu(map->base) +
- be16_to_cpu(map->size));
+ ASSERT(ichdr->freemap[mapindex].size < XFS_LBSIZE(mp));
+ ASSERT((ichdr->freemap[mapindex].size & 0x3) == 0);
+
+ ichdr->freemap[mapindex].size -=
+ xfs_attr_leaf_newentsize(args->namelen, args->valuelen,
+ mp->m_sb.sb_blocksize, &tmp);
+
+ entry->nameidx = cpu_to_be16(ichdr->freemap[mapindex].base +
+ ichdr->freemap[mapindex].size);
entry->hashval = cpu_to_be32(args->hashval);
entry->flags = tmp ? XFS_ATTR_LOCAL : 0;
entry->flags |= XFS_ATTR_NSP_ARGS_TO_ONDISK(args->flags);
@@ -996,7 +1167,7 @@ xfs_attr_leaf_add_work(
XFS_DA_LOGRANGE(leaf, entry, sizeof(*entry)));
ASSERT((args->index == 0) ||
(be32_to_cpu(entry->hashval) >= be32_to_cpu((entry-1)->hashval)));
- ASSERT((args->index == be16_to_cpu(hdr->count)-1) ||
+ ASSERT((args->index == ichdr->count - 1) ||
(be32_to_cpu(entry->hashval) <= be32_to_cpu((entry+1)->hashval)));
/*
@@ -1007,14 +1178,14 @@ xfs_attr_leaf_add_work(
* as part of this transaction (a split operation for example).
*/
if (entry->flags & XFS_ATTR_LOCAL) {
- name_loc = xfs_attr_leaf_name_local(leaf, args->index);
+ name_loc = xfs_attr3_leaf_name_local(leaf, args->index);
name_loc->namelen = args->namelen;
name_loc->valuelen = cpu_to_be16(args->valuelen);
memcpy((char *)name_loc->nameval, args->name, args->namelen);
memcpy((char *)&name_loc->nameval[args->namelen], args->value,
be16_to_cpu(name_loc->valuelen));
} else {
- name_rmt = xfs_attr_leaf_name_remote(leaf, args->index);
+ name_rmt = xfs_attr3_leaf_name_remote(leaf, args->index);
name_rmt->namelen = args->namelen;
memcpy((char *)name_rmt->name, args->name, args->namelen);
entry->flags |= XFS_ATTR_INCOMPLETE;
@@ -1025,44 +1196,41 @@ xfs_attr_leaf_add_work(
args->rmtblkcnt = XFS_B_TO_FSB(mp, args->valuelen);
}
xfs_trans_log_buf(args->trans, bp,
- XFS_DA_LOGRANGE(leaf, xfs_attr_leaf_name(leaf, args->index),
+ XFS_DA_LOGRANGE(leaf, xfs_attr3_leaf_name(leaf, args->index),
xfs_attr_leaf_entsize(leaf, args->index)));
/*
* Update the control info for this leaf node
*/
- if (be16_to_cpu(entry->nameidx) < be16_to_cpu(hdr->firstused)) {
- /* both on-disk, don't endian-flip twice */
- hdr->firstused = entry->nameidx;
- }
- ASSERT(be16_to_cpu(hdr->firstused) >=
- ((be16_to_cpu(hdr->count) * sizeof(*entry)) + sizeof(*hdr)));
- tmp = (be16_to_cpu(hdr->count)-1) * sizeof(xfs_attr_leaf_entry_t)
- + sizeof(xfs_attr_leaf_hdr_t);
- map = &hdr->freemap[0];
+ if (be16_to_cpu(entry->nameidx) < ichdr->firstused)
+ ichdr->firstused = be16_to_cpu(entry->nameidx);
+
+ ASSERT(ichdr->firstused >= ichdr->count * sizeof(xfs_attr_leaf_entry_t)
+ + xfs_attr3_leaf_hdr_size(leaf));
+ tmp = (ichdr->count - 1) * sizeof(xfs_attr_leaf_entry_t)
+ + xfs_attr3_leaf_hdr_size(leaf);
+
for (i = 0; i < XFS_ATTR_LEAF_MAPSIZE; map++, i++) {
- if (be16_to_cpu(map->base) == tmp) {
- be16_add_cpu(&map->base, sizeof(xfs_attr_leaf_entry_t));
- be16_add_cpu(&map->size,
- -((int)sizeof(xfs_attr_leaf_entry_t)));
+ if (ichdr->freemap[i].base == tmp) {
+ ichdr->freemap[i].base += sizeof(xfs_attr_leaf_entry_t);
+ ichdr->freemap[i].size -= sizeof(xfs_attr_leaf_entry_t);
}
}
- be16_add_cpu(&hdr->usedbytes, xfs_attr_leaf_entsize(leaf, args->index));
- xfs_trans_log_buf(args->trans, bp,
- XFS_DA_LOGRANGE(leaf, hdr, sizeof(*hdr)));
- return(0);
+ ichdr->usedbytes += xfs_attr_leaf_entsize(leaf, args->index);
+ return 0;
}
/*
* Garbage collect a leaf attribute list block by copying it to a new buffer.
*/
STATIC void
-xfs_attr_leaf_compact(
+xfs_attr3_leaf_compact(
struct xfs_da_args *args,
+ struct xfs_attr3_icleaf_hdr *ichdr_d,
struct xfs_buf *bp)
{
xfs_attr_leafblock_t *leaf_s, *leaf_d;
- xfs_attr_leaf_hdr_t *hdr_s, *hdr_d;
+ struct xfs_attr3_icleaf_hdr ichdr_s;
struct xfs_trans *trans = args->trans;
struct xfs_mount *mp = trans->t_mountp;
char *tmpbuffer;
@@ -1079,34 +1247,69 @@ xfs_attr_leaf_compact(
*/
leaf_s = (xfs_attr_leafblock_t *)tmpbuffer;
leaf_d = bp->b_addr;
- hdr_s = &leaf_s->hdr;
- hdr_d = &leaf_d->hdr;
- hdr_d->info = hdr_s->info; /* struct copy */
- hdr_d->firstused = cpu_to_be16(XFS_LBSIZE(mp));
- /* handle truncation gracefully */
- if (!hdr_d->firstused) {
- hdr_d->firstused = cpu_to_be16(
- XFS_LBSIZE(mp) - XFS_ATTR_LEAF_NAME_ALIGN);
- }
- hdr_d->usedbytes = 0;
- hdr_d->count = 0;
- hdr_d->holes = 0;
- hdr_d->freemap[0].base = cpu_to_be16(sizeof(xfs_attr_leaf_hdr_t));
- hdr_d->freemap[0].size = cpu_to_be16(be16_to_cpu(hdr_d->firstused) -
- sizeof(xfs_attr_leaf_hdr_t));
+ ichdr_s = *ichdr_d; /* struct copy */
+ ichdr_d->firstused = XFS_LBSIZE(mp);
+ ichdr_d->usedbytes = 0;
+ ichdr_d->count = 0;
+ ichdr_d->holes = 0;
+ ichdr_d->freemap[0].base = xfs_attr3_leaf_hdr_size(leaf_s);
+ ichdr_d->freemap[0].size = ichdr_d->firstused - ichdr_d->freemap[0].base;
/*
* Copy all entry's in the same (sorted) order,
* but allocate name/value pairs packed and in sequence.
*/
- xfs_attr_leaf_moveents(leaf_s, 0, leaf_d, 0,
- be16_to_cpu(hdr_s->count), mp);
+ xfs_attr3_leaf_moveents(leaf_s, &ichdr_s, 0, leaf_d, ichdr_d, 0,
+ ichdr_s.count, mp);
+ /*
+ * this logs the entire buffer, but the caller must write the header
+ * back to the buffer when it is finished modifying it.
+ */
xfs_trans_log_buf(trans, bp, 0, XFS_LBSIZE(mp) - 1);
kmem_free(tmpbuffer);
}
/*
+ * Compare two leaf blocks "order".
+ * Return 0 unless leaf2 should go before leaf1.
+ */
+static int
+xfs_attr3_leaf_order(
+ struct xfs_buf *leaf1_bp,
+ struct xfs_attr3_icleaf_hdr *leaf1hdr,
+ struct xfs_buf *leaf2_bp,
+ struct xfs_attr3_icleaf_hdr *leaf2hdr)
+{
+ struct xfs_attr_leaf_entry *entries1;
+ struct xfs_attr_leaf_entry *entries2;
+
+ entries1 = xfs_attr3_leaf_entryp(leaf1_bp->b_addr);
+ entries2 = xfs_attr3_leaf_entryp(leaf2_bp->b_addr);
+ if (leaf1hdr->count > 0 && leaf2hdr->count > 0 &&
+ ((be32_to_cpu(entries2[0].hashval) <
+ be32_to_cpu(entries1[0].hashval)) ||
+ (be32_to_cpu(entries2[leaf2hdr->count - 1].hashval) <
+ be32_to_cpu(entries1[leaf1hdr->count - 1].hashval)))) {
+ return 1;
+ }
+ return 0;
+}
+
+int
+xfs_attr_leaf_order(
+ struct xfs_buf *leaf1_bp,
+ struct xfs_buf *leaf2_bp)
+{
+ struct xfs_attr3_icleaf_hdr ichdr1;
+ struct xfs_attr3_icleaf_hdr ichdr2;
+
+ xfs_attr3_leaf_hdr_from_disk(&ichdr1, leaf1_bp->b_addr);
+ xfs_attr3_leaf_hdr_from_disk(&ichdr2, leaf2_bp->b_addr);
+ return xfs_attr3_leaf_order(leaf1_bp, &ichdr1, leaf2_bp, &ichdr2);
+}
+
+/*
* Redistribute the attribute list entries between two leaf nodes,
* taking into account the size of the new entry.
*
@@ -1119,14 +1322,23 @@ xfs_attr_leaf_compact(
* the "new" and "old" values can end up in different blocks.
*/
STATIC void
-xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
- xfs_da_state_blk_t *blk2)
+xfs_attr3_leaf_rebalance(
+ struct xfs_da_state *state,
+ struct xfs_da_state_blk *blk1,
+ struct xfs_da_state_blk *blk2)
{
- xfs_da_args_t *args;
- xfs_da_state_blk_t *tmp_blk;
- xfs_attr_leafblock_t *leaf1, *leaf2;
- xfs_attr_leaf_hdr_t *hdr1, *hdr2;
- int count, totallen, max, space, swap;
+ struct xfs_da_args *args;
+ struct xfs_attr_leafblock *leaf1;
+ struct xfs_attr_leafblock *leaf2;
+ struct xfs_attr3_icleaf_hdr ichdr1;
+ struct xfs_attr3_icleaf_hdr ichdr2;
+ struct xfs_attr_leaf_entry *entries1;
+ struct xfs_attr_leaf_entry *entries2;
+ int count;
+ int totallen;
+ int max;
+ int space;
+ int swap;
/*
* Set up environment.
@@ -1135,9 +1347,9 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
ASSERT(blk2->magic == XFS_ATTR_LEAF_MAGIC);
leaf1 = blk1->bp->b_addr;
leaf2 = blk2->bp->b_addr;
- ASSERT(leaf1->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
- ASSERT(leaf2->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
- ASSERT(leaf2->hdr.count == 0);
+ xfs_attr3_leaf_hdr_from_disk(&ichdr1, leaf1);
+ xfs_attr3_leaf_hdr_from_disk(&ichdr2, leaf2);
+ ASSERT(ichdr2.count == 0);
args = state->args;
trace_xfs_attr_leaf_rebalance(args);
@@ -1149,16 +1361,23 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
* second block, this code should never set "swap".
*/
swap = 0;
- if (xfs_attr_leaf_order(blk1->bp, blk2->bp)) {
+ if (xfs_attr3_leaf_order(blk1->bp, &ichdr1, blk2->bp, &ichdr2)) {
+ struct xfs_da_state_blk *tmp_blk;
+ struct xfs_attr3_icleaf_hdr tmp_ichdr;
+
tmp_blk = blk1;
blk1 = blk2;
blk2 = tmp_blk;
+
+ /* struct copies to swap them rather than reconverting */
+ tmp_ichdr = ichdr1;
+ ichdr1 = ichdr2;
+ ichdr2 = tmp_ichdr;
+
leaf1 = blk1->bp->b_addr;
leaf2 = blk2->bp->b_addr;
swap = 1;
}
- hdr1 = &leaf1->hdr;
- hdr2 = &leaf2->hdr;
/*
* Examine entries until we reduce the absolute difference in
@@ -1168,41 +1387,39 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
* "inleaf" is true if the new entry should be inserted into blk1.
* If "swap" is also true, then reverse the sense of "inleaf".
*/
- state->inleaf = xfs_attr_leaf_figure_balance(state, blk1, blk2,
- &count, &totallen);
+ state->inleaf = xfs_attr3_leaf_figure_balance(state, blk1, &ichdr1,
+ blk2, &ichdr2,
+ &count, &totallen);
if (swap)
state->inleaf = !state->inleaf;
/*
* Move any entries required from leaf to leaf:
*/
- if (count < be16_to_cpu(hdr1->count)) {
+ if (count < ichdr1.count) {
/*
* Figure the total bytes to be added to the destination leaf.
*/
/* number entries being moved */
- count = be16_to_cpu(hdr1->count) - count;
- space = be16_to_cpu(hdr1->usedbytes) - totallen;
+ count = ichdr1.count - count;
+ space = ichdr1.usedbytes - totallen;
space += count * sizeof(xfs_attr_leaf_entry_t);
/*
* leaf2 is the destination, compact it if it looks tight.
*/
- max = be16_to_cpu(hdr2->firstused)
- - sizeof(xfs_attr_leaf_hdr_t);
- max -= be16_to_cpu(hdr2->count) * sizeof(xfs_attr_leaf_entry_t);
+ max = ichdr2.firstused - xfs_attr3_leaf_hdr_size(leaf1);
+ max -= ichdr2.count * sizeof(xfs_attr_leaf_entry_t);
if (space > max)
- xfs_attr_leaf_compact(args, blk2->bp);
+ xfs_attr3_leaf_compact(args, &ichdr2, blk2->bp);
/*
* Move high entries from leaf1 to low end of leaf2.
*/
- xfs_attr_leaf_moveents(leaf1, be16_to_cpu(hdr1->count) - count,
- leaf2, 0, count, state->mp);
+ xfs_attr3_leaf_moveents(leaf1, &ichdr1, ichdr1.count - count,
+ leaf2, &ichdr2, 0, count, state->mp);
- xfs_trans_log_buf(args->trans, blk1->bp, 0, state->blocksize-1);
- xfs_trans_log_buf(args->trans, blk2->bp, 0, state->blocksize-1);
- } else if (count > be16_to_cpu(hdr1->count)) {
+ } else if (count > ichdr1.count) {
/*
* I assert that since all callers pass in an empty
* second buffer, this code should never execute.
@@ -1213,36 +1430,37 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
* Figure the total bytes to be added to the destination leaf.
*/
/* number entries being moved */
- count -= be16_to_cpu(hdr1->count);
- space = totallen - be16_to_cpu(hdr1->usedbytes);
+ count -= ichdr1.count;
+ space = totallen - ichdr1.usedbytes;
space += count * sizeof(xfs_attr_leaf_entry_t);
/*
* leaf1 is the destination, compact it if it looks tight.
*/
- max = be16_to_cpu(hdr1->firstused)
- - sizeof(xfs_attr_leaf_hdr_t);
- max -= be16_to_cpu(hdr1->count) * sizeof(xfs_attr_leaf_entry_t);
+ max = ichdr1.firstused - xfs_attr3_leaf_hdr_size(leaf1);
+ max -= ichdr1.count * sizeof(xfs_attr_leaf_entry_t);
if (space > max)
- xfs_attr_leaf_compact(args, blk1->bp);
+ xfs_attr3_leaf_compact(args, &ichdr1, blk1->bp);
/*
* Move low entries from leaf2 to high end of leaf1.
*/
- xfs_attr_leaf_moveents(leaf2, 0, leaf1,
- be16_to_cpu(hdr1->count), count, state->mp);
-
- xfs_trans_log_buf(args->trans, blk1->bp, 0, state->blocksize-1);
- xfs_trans_log_buf(args->trans, blk2->bp, 0, state->blocksize-1);
+ xfs_attr3_leaf_moveents(leaf2, &ichdr2, 0, leaf1, &ichdr1,
+ ichdr1.count, count, state->mp);
}
+ xfs_attr3_leaf_hdr_to_disk(leaf1, &ichdr1);
+ xfs_attr3_leaf_hdr_to_disk(leaf2, &ichdr2);
+ xfs_trans_log_buf(args->trans, blk1->bp, 0, state->blocksize-1);
+ xfs_trans_log_buf(args->trans, blk2->bp, 0, state->blocksize-1);
+
/*
* Copy out last hashval in each block for B-tree code.
*/
- blk1->hashval = be32_to_cpu(
- leaf1->entries[be16_to_cpu(leaf1->hdr.count)-1].hashval);
- blk2->hashval = be32_to_cpu(
- leaf2->entries[be16_to_cpu(leaf2->hdr.count)-1].hashval);
+ entries1 = xfs_attr3_leaf_entryp(leaf1);
+ entries2 = xfs_attr3_leaf_entryp(leaf2);
+ blk1->hashval = be32_to_cpu(entries1[ichdr1.count - 1].hashval);
+ blk2->hashval = be32_to_cpu(entries2[ichdr2.count - 1].hashval);
/*
* Adjust the expected index for insertion.
@@ -1256,12 +1474,12 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
* inserting. The index/blkno fields refer to the "old" entry,
* while the index2/blkno2 fields refer to the "new" entry.
*/
- if (blk1->index > be16_to_cpu(leaf1->hdr.count)) {
+ if (blk1->index > ichdr1.count) {
ASSERT(state->inleaf == 0);
- blk2->index = blk1->index - be16_to_cpu(leaf1->hdr.count);
+ blk2->index = blk1->index - ichdr1.count;
args->index = args->index2 = blk2->index;
args->blkno = args->blkno2 = blk2->blkno;
- } else if (blk1->index == be16_to_cpu(leaf1->hdr.count)) {
+ } else if (blk1->index == ichdr1.count) {
if (state->inleaf) {
args->index = blk1->index;
args->blkno = blk1->blkno;
@@ -1273,8 +1491,7 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
* is already stored in blkno2/index2, so don't
* overwrite it overwise we corrupt the tree.
*/
- blk2->index = blk1->index
- - be16_to_cpu(leaf1->hdr.count);
+ blk2->index = blk1->index - ichdr1.count;
args->index = blk2->index;
args->blkno = blk2->blkno;
if (!state->extravalid) {
@@ -1302,42 +1519,40 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
* GROT: Do a double-split for this case?
*/
STATIC int
-xfs_attr_leaf_figure_balance(xfs_da_state_t *state,
- xfs_da_state_blk_t *blk1,
- xfs_da_state_blk_t *blk2,
- int *countarg, int *usedbytesarg)
+xfs_attr3_leaf_figure_balance(
+ struct xfs_da_state *state,
+ struct xfs_da_state_blk *blk1,
+ struct xfs_attr3_icleaf_hdr *ichdr1,
+ struct xfs_da_state_blk *blk2,
+ struct xfs_attr3_icleaf_hdr *ichdr2,
+ int *countarg,
+ int *usedbytesarg)
{
- xfs_attr_leafblock_t *leaf1, *leaf2;
- xfs_attr_leaf_hdr_t *hdr1, *hdr2;
- xfs_attr_leaf_entry_t *entry;
- int count, max, index, totallen, half;
- int lastdelta, foundit, tmp;
-
- /*
- * Set up environment.
- */
- leaf1 = blk1->bp->b_addr;
- leaf2 = blk2->bp->b_addr;
- hdr1 = &leaf1->hdr;
- hdr2 = &leaf2->hdr;
- foundit = 0;
- totallen = 0;
+ struct xfs_attr_leafblock *leaf1 = blk1->bp->b_addr;
+ struct xfs_attr_leafblock *leaf2 = blk2->bp->b_addr;
+ struct xfs_attr_leaf_entry *entry;
+ int count;
+ int max;
+ int index;
+ int totallen = 0;
+ int half;
+ int lastdelta;
+ int foundit = 0;
+ int tmp;
/*
* Examine entries until we reduce the absolute difference in
* byte usage between the two blocks to a minimum.
*/
- max = be16_to_cpu(hdr1->count) + be16_to_cpu(hdr2->count);
- half = (max+1) * sizeof(*entry);
- half += be16_to_cpu(hdr1->usedbytes) +
- be16_to_cpu(hdr2->usedbytes) +
- xfs_attr_leaf_newentsize(
- state->args->namelen,
- state->args->valuelen,
- state->blocksize, NULL);
+ max = ichdr1->count + ichdr2->count;
+ half = (max + 1) * sizeof(*entry);
+ half += ichdr1->usedbytes + ichdr2->usedbytes +
+ xfs_attr_leaf_newentsize(state->args->namelen,
+ state->args->valuelen,
+ state->blocksize, NULL);
half /= 2;
lastdelta = state->blocksize;
- entry = &leaf1->entries[0];
+ entry = xfs_attr3_leaf_entryp(leaf1);
for (count = index = 0; count < max; entry++, index++, count++) {
#define XFS_ATTR_ABS(A) (((A) < 0) ? -(A) : (A))
@@ -1360,9 +1575,9 @@ xfs_attr_leaf_figure_balance(xfs_da_state_t *state,
/*
* Wrap around into the second block if necessary.
*/
- if (count == be16_to_cpu(hdr1->count)) {
+ if (count == ichdr1->count) {
leaf1 = leaf2;
- entry = &leaf1->entries[0];
+ entry = xfs_attr3_leaf_entryp(leaf1);
index = 0;
}
@@ -1393,7 +1608,7 @@ xfs_attr_leaf_figure_balance(xfs_da_state_t *state,
*countarg = count;
*usedbytesarg = totallen;
- return(foundit);
+ return foundit;
}
/*========================================================================
@@ -1412,14 +1627,20 @@ xfs_attr_leaf_figure_balance(xfs_da_state_t *state,
* GROT: allow for INCOMPLETE entries in calculation.
*/
int
-xfs_attr_leaf_toosmall(xfs_da_state_t *state, int *action)
+xfs_attr3_leaf_toosmall(
+ struct xfs_da_state *state,
+ int *action)
{
- xfs_attr_leafblock_t *leaf;
- xfs_da_state_blk_t *blk;
- xfs_da_blkinfo_t *info;
- int count, bytes, forward, error, retval, i;
- xfs_dablk_t blkno;
- struct xfs_buf *bp;
+ struct xfs_attr_leafblock *leaf;
+ struct xfs_da_state_blk *blk;
+ struct xfs_attr3_icleaf_hdr ichdr;
+ struct xfs_buf *bp;
+ xfs_dablk_t blkno;
+ int bytes;
+ int forward;
+ int error;
+ int retval;
+ int i;
trace_xfs_attr_leaf_toosmall(state->args);
@@ -1429,13 +1650,11 @@ xfs_attr_leaf_toosmall(xfs_da_state_t *state, int *action)
* to coalesce with a sibling.
*/
blk = &state->path.blk[ state->path.active-1 ];
- info = blk->bp->b_addr;
- ASSERT(info->magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
- leaf = (xfs_attr_leafblock_t *)info;
- count = be16_to_cpu(leaf->hdr.count);
- bytes = sizeof(xfs_attr_leaf_hdr_t) +
- count * sizeof(xfs_attr_leaf_entry_t) +
- be16_to_cpu(leaf->hdr.usedbytes);
+ leaf = blk->bp->b_addr;
+ xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf);
+ bytes = xfs_attr3_leaf_hdr_size(leaf) +
+ ichdr.count * sizeof(xfs_attr_leaf_entry_t) +
+ ichdr.usedbytes;
if (bytes > (state->blocksize >> 1)) {
*action = 0; /* blk over 50%, don't try to join */
return(0);
@@ -1447,12 +1666,12 @@ xfs_attr_leaf_toosmall(xfs_da_state_t *state, int *action)
* coalesce it with a sibling block. We choose (arbitrarily)
* to merge with the forward block unless it is NULL.
*/
- if (count == 0) {
+ if (ichdr.count == 0) {
/*
* Make altpath point to the block we want to keep and
* path point to the block we want to drop (this one).
*/
- forward = (info->forw != 0);
+ forward = (ichdr.forw != 0);
memcpy(&state->altpath, &state->path, sizeof(state->path));
error = xfs_da3_path_shift(state, &state->altpath, forward,
0, &retval);
@@ -1463,7 +1682,7 @@ xfs_attr_leaf_toosmall(xfs_da_state_t *state, int *action)
} else {
*action = 2;
}
- return(0);
+ return 0;
}
/*
@@ -1474,28 +1693,28 @@ xfs_attr_leaf_toosmall(xfs_da_state_t *state, int *action)
* to shrink an attribute list over time.
*/
/* start with smaller blk num */
- forward = (be32_to_cpu(info->forw) < be32_to_cpu(info->back));
+ forward = ichdr.forw < ichdr.back;
for (i = 0; i < 2; forward = !forward, i++) {
+ struct xfs_attr3_icleaf_hdr ichdr2;
if (forward)
- blkno = be32_to_cpu(info->forw);
+ blkno = ichdr.forw;
else
- blkno = be32_to_cpu(info->back);
+ blkno = ichdr.back;
if (blkno == 0)
continue;
- error = xfs_attr_leaf_read(state->args->trans, state->args->dp,
+ error = xfs_attr3_leaf_read(state->args->trans, state->args->dp,
blkno, -1, &bp);
if (error)
return(error);
- leaf = (xfs_attr_leafblock_t *)info;
- count = be16_to_cpu(leaf->hdr.count);
- bytes = state->blocksize - (state->blocksize>>2);
- bytes -= be16_to_cpu(leaf->hdr.usedbytes);
- leaf = bp->b_addr;
- count += be16_to_cpu(leaf->hdr.count);
- bytes -= be16_to_cpu(leaf->hdr.usedbytes);
- bytes -= count * sizeof(xfs_attr_leaf_entry_t);
- bytes -= sizeof(xfs_attr_leaf_hdr_t);
+ xfs_attr3_leaf_hdr_from_disk(&ichdr2, bp->b_addr);
+
+ bytes = state->blocksize - (state->blocksize >> 2) -
+ ichdr.usedbytes - ichdr2.usedbytes -
+ ((ichdr.count + ichdr2.count) *
+ sizeof(xfs_attr_leaf_entry_t)) -
+ xfs_attr3_leaf_hdr_size(leaf);
+
xfs_trans_brelse(state->args->trans, bp);
if (bytes >= 0)
break; /* fits with at least 25% to spare */
@@ -1534,32 +1753,35 @@ xfs_attr_leaf_toosmall(xfs_da_state_t *state, int *action)
* If two leaves are 37% full, when combined they will leave 25% free.
*/
int
-xfs_attr_leaf_remove(
- struct xfs_buf *bp,
- xfs_da_args_t *args)
+xfs_attr3_leaf_remove(
+ struct xfs_buf *bp,
+ struct xfs_da_args *args)
{
- xfs_attr_leafblock_t *leaf;
- xfs_attr_leaf_hdr_t *hdr;
- xfs_attr_leaf_map_t *map;
- xfs_attr_leaf_entry_t *entry;
- int before, after, smallest, entsize;
- int tablesize, tmp, i;
- xfs_mount_t *mp;
+ struct xfs_attr_leafblock *leaf;
+ struct xfs_attr3_icleaf_hdr ichdr;
+ struct xfs_attr_leaf_entry *entry;
+ struct xfs_mount *mp = args->trans->t_mountp;
+ int before;
+ int after;
+ int smallest;
+ int entsize;
+ int tablesize;
+ int tmp;
+ int i;
trace_xfs_attr_leaf_remove(args);
leaf = bp->b_addr;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
- hdr = &leaf->hdr;
- mp = args->trans->t_mountp;
- ASSERT((be16_to_cpu(hdr->count) > 0)
- && (be16_to_cpu(hdr->count) < (XFS_LBSIZE(mp)/8)));
- ASSERT((args->index >= 0)
- && (args->index < be16_to_cpu(hdr->count)));
- ASSERT(be16_to_cpu(hdr->firstused) >=
- ((be16_to_cpu(hdr->count) * sizeof(*entry)) + sizeof(*hdr)));
- entry = &leaf->entries[args->index];
- ASSERT(be16_to_cpu(entry->nameidx) >= be16_to_cpu(hdr->firstused));
+ xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf);
+
+ ASSERT(ichdr.count > 0 && ichdr.count < XFS_LBSIZE(mp) / 8);
+ ASSERT(args->index >= 0 && args->index < ichdr.count);
+ ASSERT(ichdr.firstused >= ichdr.count * sizeof(*entry) +
+ xfs_attr3_leaf_hdr_size(leaf));
+
+ entry = &xfs_attr3_leaf_entryp(leaf)[args->index];
+
+ ASSERT(be16_to_cpu(entry->nameidx) >= ichdr.firstused);
ASSERT(be16_to_cpu(entry->nameidx) < XFS_LBSIZE(mp));
/*
@@ -1568,30 +1790,28 @@ xfs_attr_leaf_remove(
* find smallest free region in case we need to replace it,
* adjust any map that borders the entry table,
*/
- tablesize = be16_to_cpu(hdr->count) * sizeof(xfs_attr_leaf_entry_t)
- + sizeof(xfs_attr_leaf_hdr_t);
- map = &hdr->freemap[0];
- tmp = be16_to_cpu(map->size);
+ tablesize = ichdr.count * sizeof(xfs_attr_leaf_entry_t)
+ + xfs_attr3_leaf_hdr_size(leaf);
+ tmp = ichdr.freemap[0].size;
before = after = -1;
smallest = XFS_ATTR_LEAF_MAPSIZE - 1;
entsize = xfs_attr_leaf_entsize(leaf, args->index);
- for (i = 0; i < XFS_ATTR_LEAF_MAPSIZE; map++, i++) {
- ASSERT(be16_to_cpu(map->base) < XFS_LBSIZE(mp));
- ASSERT(be16_to_cpu(map->size) < XFS_LBSIZE(mp));
- if (be16_to_cpu(map->base) == tablesize) {
- be16_add_cpu(&map->base,
- -((int)sizeof(xfs_attr_leaf_entry_t)));
- be16_add_cpu(&map->size, sizeof(xfs_attr_leaf_entry_t));
+ for (i = 0; i < XFS_ATTR_LEAF_MAPSIZE; i++) {
+ ASSERT(ichdr.freemap[i].base < XFS_LBSIZE(mp));
+ ASSERT(ichdr.freemap[i].size < XFS_LBSIZE(mp));
+ if (ichdr.freemap[i].base == tablesize) {
+ ichdr.freemap[i].base -= sizeof(xfs_attr_leaf_entry_t);
+ ichdr.freemap[i].size += sizeof(xfs_attr_leaf_entry_t);
}
- if ((be16_to_cpu(map->base) + be16_to_cpu(map->size))
- == be16_to_cpu(entry->nameidx)) {
+ if (ichdr.freemap[i].base + ichdr.freemap[i].size ==
+ be16_to_cpu(entry->nameidx)) {
before = i;
- } else if (be16_to_cpu(map->base)
- == (be16_to_cpu(entry->nameidx) + entsize)) {
+ } else if (ichdr.freemap[i].base ==
+ (be16_to_cpu(entry->nameidx) + entsize)) {
after = i;
- } else if (be16_to_cpu(map->size) < tmp) {
- tmp = be16_to_cpu(map->size);
+ } else if (ichdr.freemap[i].size < tmp) {
+ tmp = ichdr.freemap[i].size;
smallest = i;
}
}
@@ -1602,36 +1822,30 @@ xfs_attr_leaf_remove(
*/
if ((before >= 0) || (after >= 0)) {
if ((before >= 0) && (after >= 0)) {
- map = &hdr->freemap[before];
- be16_add_cpu(&map->size, entsize);
- be16_add_cpu(&map->size,
- be16_to_cpu(hdr->freemap[after].size));
- hdr->freemap[after].base = 0;
- hdr->freemap[after].size = 0;
+ ichdr.freemap[before].size += entsize;
+ ichdr.freemap[before].size += ichdr.freemap[after].size;
+ ichdr.freemap[after].base = 0;
+ ichdr.freemap[after].size = 0;
} else if (before >= 0) {
- map = &hdr->freemap[before];
- be16_add_cpu(&map->size, entsize);
+ ichdr.freemap[before].size += entsize;
} else {
- map = &hdr->freemap[after];
- /* both on-disk, don't endian flip twice */
- map->base = entry->nameidx;
- be16_add_cpu(&map->size, entsize);
+ ichdr.freemap[after].base = be16_to_cpu(entry->nameidx);
+ ichdr.freemap[after].size += entsize;
}
} else {
/*
* Replace smallest region (if it is smaller than free'd entry)
*/
- map = &hdr->freemap[smallest];
- if (be16_to_cpu(map->size) < entsize) {
- map->base = cpu_to_be16(be16_to_cpu(entry->nameidx));
- map->size = cpu_to_be16(entsize);
+ if (ichdr.freemap[smallest].size < entsize) {
+ ichdr.freemap[smallest].base = be16_to_cpu(entry->nameidx);
+ ichdr.freemap[smallest].size = entsize;
}
}
/*
* Did we remove the first entry?
*/
- if (be16_to_cpu(entry->nameidx) == be16_to_cpu(hdr->firstused))
+ if (be16_to_cpu(entry->nameidx) == ichdr.firstused)
smallest = 1;
else
smallest = 0;
@@ -1639,20 +1853,20 @@ xfs_attr_leaf_remove(
/*
* Compress the remaining entries and zero out the removed stuff.
*/
- memset(xfs_attr_leaf_name(leaf, args->index), 0, entsize);
- be16_add_cpu(&hdr->usedbytes, -entsize);
+ memset(xfs_attr3_leaf_name(leaf, args->index), 0, entsize);
+ ichdr.usedbytes -= entsize;
xfs_trans_log_buf(args->trans, bp,
- XFS_DA_LOGRANGE(leaf, xfs_attr_leaf_name(leaf, args->index),
+ XFS_DA_LOGRANGE(leaf, xfs_attr3_leaf_name(leaf, args->index),
entsize));
- tmp = (be16_to_cpu(hdr->count) - args->index)
- * sizeof(xfs_attr_leaf_entry_t);
- memmove((char *)entry, (char *)(entry+1), tmp);
- be16_add_cpu(&hdr->count, -1);
+ tmp = (ichdr.count - args->index) * sizeof(xfs_attr_leaf_entry_t);
+ memmove(entry, entry + 1, tmp);
+ ichdr.count--;
xfs_trans_log_buf(args->trans, bp,
- XFS_DA_LOGRANGE(leaf, entry, tmp + sizeof(*entry)));
- entry = &leaf->entries[be16_to_cpu(hdr->count)];
- memset((char *)entry, 0, sizeof(xfs_attr_leaf_entry_t));
+ XFS_DA_LOGRANGE(leaf, entry, tmp + sizeof(xfs_attr_leaf_entry_t)));
+
+ entry = &xfs_attr3_leaf_entryp(leaf)[ichdr.count];
+ memset(entry, 0, sizeof(xfs_attr_leaf_entry_t));
/*
* If we removed the first entry, re-find the first used byte
@@ -1662,130 +1876,130 @@ xfs_attr_leaf_remove(
*/
if (smallest) {
tmp = XFS_LBSIZE(mp);
- entry = &leaf->entries[0];
- for (i = be16_to_cpu(hdr->count)-1; i >= 0; entry++, i--) {
- ASSERT(be16_to_cpu(entry->nameidx) >=
- be16_to_cpu(hdr->firstused));
+ entry = xfs_attr3_leaf_entryp(leaf);
+ for (i = ichdr.count - 1; i >= 0; entry++, i--) {
+ ASSERT(be16_to_cpu(entry->nameidx) >= ichdr.firstused);
ASSERT(be16_to_cpu(entry->nameidx) < XFS_LBSIZE(mp));
if (be16_to_cpu(entry->nameidx) < tmp)
tmp = be16_to_cpu(entry->nameidx);
}
- hdr->firstused = cpu_to_be16(tmp);
- if (!hdr->firstused) {
- hdr->firstused = cpu_to_be16(
- tmp - XFS_ATTR_LEAF_NAME_ALIGN);
- }
+ ichdr.firstused = tmp;
+ if (!ichdr.firstused)
+ ichdr.firstused = tmp - XFS_ATTR_LEAF_NAME_ALIGN;
} else {
- hdr->holes = 1; /* mark as needing compaction */
+ ichdr.holes = 1; /* mark as needing compaction */
}
+ xfs_attr3_leaf_hdr_to_disk(leaf, &ichdr);
xfs_trans_log_buf(args->trans, bp,
- XFS_DA_LOGRANGE(leaf, hdr, sizeof(*hdr)));
+ XFS_DA_LOGRANGE(leaf, &leaf->hdr,
+ xfs_attr3_leaf_hdr_size(leaf)));
/*
* Check if leaf is less than 50% full, caller may want to
* "join" the leaf with a sibling if so.
*/
- tmp = sizeof(xfs_attr_leaf_hdr_t);
- tmp += be16_to_cpu(leaf->hdr.count) * sizeof(xfs_attr_leaf_entry_t);
- tmp += be16_to_cpu(leaf->hdr.usedbytes);
- return(tmp < mp->m_attr_magicpct); /* leaf is < 37% full */
+ tmp = ichdr.usedbytes + xfs_attr3_leaf_hdr_size(leaf) +
+ ichdr.count * sizeof(xfs_attr_leaf_entry_t);
+
+ return tmp < mp->m_attr_magicpct; /* leaf is < 37% full */
}
/*
* Move all the attribute list entries from drop_leaf into save_leaf.
*/
void
-xfs_attr_leaf_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
- xfs_da_state_blk_t *save_blk)
+xfs_attr3_leaf_unbalance(
+ struct xfs_da_state *state,
+ struct xfs_da_state_blk *drop_blk,
+ struct xfs_da_state_blk *save_blk)
{
- xfs_attr_leafblock_t *drop_leaf, *save_leaf, *tmp_leaf;
- xfs_attr_leaf_hdr_t *drop_hdr, *save_hdr, *tmp_hdr;
- xfs_mount_t *mp;
- char *tmpbuffer;
+ struct xfs_attr_leafblock *drop_leaf = drop_blk->bp->b_addr;
+ struct xfs_attr_leafblock *save_leaf = save_blk->bp->b_addr;
+ struct xfs_attr3_icleaf_hdr drophdr;
+ struct xfs_attr3_icleaf_hdr savehdr;
+ struct xfs_attr_leaf_entry *entry;
+ struct xfs_mount *mp = state->mp;
trace_xfs_attr_leaf_unbalance(state->args);
- /*
- * Set up environment.
- */
- mp = state->mp;
- ASSERT(drop_blk->magic == XFS_ATTR_LEAF_MAGIC);
- ASSERT(save_blk->magic == XFS_ATTR_LEAF_MAGIC);
drop_leaf = drop_blk->bp->b_addr;
save_leaf = save_blk->bp->b_addr;
- ASSERT(drop_leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
- ASSERT(save_leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
- drop_hdr = &drop_leaf->hdr;
- save_hdr = &save_leaf->hdr;
+ xfs_attr3_leaf_hdr_from_disk(&drophdr, drop_leaf);
+ xfs_attr3_leaf_hdr_from_disk(&savehdr, save_leaf);
+ entry = xfs_attr3_leaf_entryp(drop_leaf);
/*
* Save last hashval from dying block for later Btree fixup.
*/
- drop_blk->hashval = be32_to_cpu(
- drop_leaf->entries[be16_to_cpu(drop_leaf->hdr.count)-1].hashval);
+ drop_blk->hashval = be32_to_cpu(entry[drophdr.count - 1].hashval);
/*
* Check if we need a temp buffer, or can we do it in place.
* Note that we don't check "leaf" for holes because we will
* always be dropping it, toosmall() decided that for us already.
*/
- if (save_hdr->holes == 0) {
+ if (savehdr.holes == 0) {
/*
* dest leaf has no holes, so we add there. May need
* to make some room in the entry array.
*/
- if (xfs_attr_leaf_order(save_blk->bp, drop_blk->bp)) {
- xfs_attr_leaf_moveents(drop_leaf, 0, save_leaf, 0,
- be16_to_cpu(drop_hdr->count), mp);
+ if (xfs_attr3_leaf_order(save_blk->bp, &savehdr,
+ drop_blk->bp, &drophdr)) {
+ xfs_attr3_leaf_moveents(drop_leaf, &drophdr, 0,
+ save_leaf, &savehdr, 0,
+ drophdr.count, mp);
} else {
- xfs_attr_leaf_moveents(drop_leaf, 0, save_leaf,
- be16_to_cpu(save_hdr->count),
- be16_to_cpu(drop_hdr->count), mp);
+ xfs_attr3_leaf_moveents(drop_leaf, &drophdr, 0,
+ save_leaf, &savehdr,
+ savehdr.count, drophdr.count, mp);
}
} else {
/*
* Destination has holes, so we make a temporary copy
* of the leaf and add them both to that.
*/
- tmpbuffer = kmem_alloc(state->blocksize, KM_SLEEP);
- ASSERT(tmpbuffer != NULL);
- memset(tmpbuffer, 0, state->blocksize);
- tmp_leaf = (xfs_attr_leafblock_t *)tmpbuffer;
- tmp_hdr = &tmp_leaf->hdr;
- tmp_hdr->info = save_hdr->info; /* struct copy */
- tmp_hdr->count = 0;
- tmp_hdr->firstused = cpu_to_be16(state->blocksize);
- if (!tmp_hdr->firstused) {
- tmp_hdr->firstused = cpu_to_be16(
- state->blocksize - XFS_ATTR_LEAF_NAME_ALIGN);
- }
- tmp_hdr->usedbytes = 0;
- if (xfs_attr_leaf_order(save_blk->bp, drop_blk->bp)) {
- xfs_attr_leaf_moveents(drop_leaf, 0, tmp_leaf, 0,
- be16_to_cpu(drop_hdr->count), mp);
- xfs_attr_leaf_moveents(save_leaf, 0, tmp_leaf,
- be16_to_cpu(tmp_leaf->hdr.count),
- be16_to_cpu(save_hdr->count), mp);
+ struct xfs_attr_leafblock *tmp_leaf;
+ struct xfs_attr3_icleaf_hdr tmphdr;
+
+ tmp_leaf = kmem_alloc(state->blocksize, KM_SLEEP);
+ memset(tmp_leaf, 0, state->blocksize);
+ memset(&tmphdr, 0, sizeof(tmphdr));
+
+ tmphdr.magic = savehdr.magic;
+ tmphdr.forw = savehdr.forw;
+ tmphdr.back = savehdr.back;
+ tmphdr.firstused = state->blocksize;
+ if (xfs_attr3_leaf_order(save_blk->bp, &savehdr,
+ drop_blk->bp, &drophdr)) {
+ xfs_attr3_leaf_moveents(drop_leaf, &drophdr, 0,
+ tmp_leaf, &tmphdr, 0,
+ drophdr.count, mp);
+ xfs_attr3_leaf_moveents(save_leaf, &savehdr, 0,
+ tmp_leaf, &tmphdr, tmphdr.count,
+ savehdr.count, mp);
} else {
- xfs_attr_leaf_moveents(save_leaf, 0, tmp_leaf, 0,
- be16_to_cpu(save_hdr->count), mp);
- xfs_attr_leaf_moveents(drop_leaf, 0, tmp_leaf,
- be16_to_cpu(tmp_leaf->hdr.count),
- be16_to_cpu(drop_hdr->count), mp);
+ xfs_attr3_leaf_moveents(save_leaf, &savehdr, 0,
+ tmp_leaf, &tmphdr, 0,
+ savehdr.count, mp);
+ xfs_attr3_leaf_moveents(drop_leaf, &drophdr, 0,
+ tmp_leaf, &tmphdr, tmphdr.count,
+ drophdr.count, mp);
}
- memcpy((char *)save_leaf, (char *)tmp_leaf, state->blocksize);
- kmem_free(tmpbuffer);
+ memcpy(save_leaf, tmp_leaf, state->blocksize);
+ savehdr = tmphdr; /* struct copy */
+ kmem_free(tmp_leaf);
}
+ xfs_attr3_leaf_hdr_to_disk(save_leaf, &savehdr);
xfs_trans_log_buf(state->args->trans, save_blk->bp, 0,
state->blocksize - 1);
/*
* Copy out last hashval in each block for B-tree code.
*/
- save_blk->hashval = be32_to_cpu(
- save_leaf->entries[be16_to_cpu(save_leaf->hdr.count)-1].hashval);
+ entry = xfs_attr3_leaf_entryp(save_leaf);
+ save_blk->hashval = be32_to_cpu(entry[savehdr.count - 1].hashval);
}
/*========================================================================
@@ -1806,31 +2020,33 @@ xfs_attr_leaf_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
* Don't change the args->value unless we find the attribute.
*/
int
-xfs_attr_leaf_lookup_int(
- struct xfs_buf *bp,
- xfs_da_args_t *args)
+xfs_attr3_leaf_lookup_int(
+ struct xfs_buf *bp,
+ struct xfs_da_args *args)
{
- xfs_attr_leafblock_t *leaf;
- xfs_attr_leaf_entry_t *entry;
- xfs_attr_leaf_name_local_t *name_loc;
- xfs_attr_leaf_name_remote_t *name_rmt;
- int probe, span;
- xfs_dahash_t hashval;
+ struct xfs_attr_leafblock *leaf;
+ struct xfs_attr3_icleaf_hdr ichdr;
+ struct xfs_attr_leaf_entry *entry;
+ struct xfs_attr_leaf_entry *entries;
+ struct xfs_attr_leaf_name_local *name_loc;
+ struct xfs_attr_leaf_name_remote *name_rmt;
+ xfs_dahash_t hashval;
+ int probe;
+ int span;
trace_xfs_attr_leaf_lookup(args);
leaf = bp->b_addr;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
- ASSERT(be16_to_cpu(leaf->hdr.count)
- < (XFS_LBSIZE(args->dp->i_mount)/8));
+ xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf);
+ entries = xfs_attr3_leaf_entryp(leaf);
+ ASSERT(ichdr.count < XFS_LBSIZE(args->dp->i_mount) / 8);
/*
* Binary search. (note: small blocks will skip this loop)
*/
hashval = args->hashval;
- probe = span = be16_to_cpu(leaf->hdr.count) / 2;
- for (entry = &leaf->entries[probe]; span > 4;
- entry = &leaf->entries[probe]) {
+ probe = span = ichdr.count / 2;
+ for (entry = &entries[probe]; span > 4; entry = &entries[probe]) {
span /= 2;
if (be32_to_cpu(entry->hashval) < hashval)
probe += span;
@@ -1839,35 +2055,31 @@ xfs_attr_leaf_lookup_int(
else
break;
}
- ASSERT((probe >= 0) &&
- (!leaf->hdr.count
- || (probe < be16_to_cpu(leaf->hdr.count))));
- ASSERT((span <= 4) || (be32_to_cpu(entry->hashval) == hashval));
+ ASSERT(probe >= 0 && (!ichdr.count || probe < ichdr.count));
+ ASSERT(span <= 4 || be32_to_cpu(entry->hashval) == hashval);
/*
* Since we may have duplicate hashval's, find the first matching
* hashval in the leaf.
*/
- while ((probe > 0) && (be32_to_cpu(entry->hashval) >= hashval)) {
+ while (probe > 0 && be32_to_cpu(entry->hashval) >= hashval) {
entry--;
probe--;
}
- while ((probe < be16_to_cpu(leaf->hdr.count)) &&
- (be32_to_cpu(entry->hashval) < hashval)) {
+ while (probe < ichdr.count &&
+ be32_to_cpu(entry->hashval) < hashval) {
entry++;
probe++;
}
- if ((probe == be16_to_cpu(leaf->hdr.count)) ||
- (be32_to_cpu(entry->hashval) != hashval)) {
+ if (probe == ichdr.count || be32_to_cpu(entry->hashval) != hashval) {
args->index = probe;
- return(XFS_ERROR(ENOATTR));
+ return XFS_ERROR(ENOATTR);
}
/*
* Duplicate keys may be present, so search all of them for a match.
*/
- for ( ; (probe < be16_to_cpu(leaf->hdr.count)) &&
- (be32_to_cpu(entry->hashval) == hashval);
+ for (; probe < ichdr.count && (be32_to_cpu(entry->hashval) == hashval);
entry++, probe++) {
/*
* GROT: Add code to remove incomplete entries.
@@ -1881,21 +2093,22 @@ xfs_attr_leaf_lookup_int(
continue;
}
if (entry->flags & XFS_ATTR_LOCAL) {
- name_loc = xfs_attr_leaf_name_local(leaf, probe);
+ name_loc = xfs_attr3_leaf_name_local(leaf, probe);
if (name_loc->namelen != args->namelen)
continue;
- if (memcmp(args->name, (char *)name_loc->nameval, args->namelen) != 0)
+ if (memcmp(args->name, name_loc->nameval,
+ args->namelen) != 0)
continue;
if (!xfs_attr_namesp_match(args->flags, entry->flags))
continue;
args->index = probe;
- return(XFS_ERROR(EEXIST));
+ return XFS_ERROR(EEXIST);
} else {
- name_rmt = xfs_attr_leaf_name_remote(leaf, probe);
+ name_rmt = xfs_attr3_leaf_name_remote(leaf, probe);
if (name_rmt->namelen != args->namelen)
continue;
- if (memcmp(args->name, (char *)name_rmt->name,
- args->namelen) != 0)
+ if (memcmp(args->name, name_rmt->name,
+ args->namelen) != 0)
continue;
if (!xfs_attr_namesp_match(args->flags, entry->flags))
continue;
@@ -1903,11 +2116,11 @@ xfs_attr_leaf_lookup_int(
args->rmtblkno = be32_to_cpu(name_rmt->valueblk);
args->rmtblkcnt = XFS_B_TO_FSB(args->dp->i_mount,
be32_to_cpu(name_rmt->valuelen));
- return(XFS_ERROR(EEXIST));
+ return XFS_ERROR(EEXIST);
}
}
args->index = probe;
- return(XFS_ERROR(ENOATTR));
+ return XFS_ERROR(ENOATTR);
}
/*
@@ -1915,40 +2128,40 @@ xfs_attr_leaf_lookup_int(
* list structure.
*/
int
-xfs_attr_leaf_getvalue(
- struct xfs_buf *bp,
- xfs_da_args_t *args)
+xfs_attr3_leaf_getvalue(
+ struct xfs_buf *bp,
+ struct xfs_da_args *args)
{
- int valuelen;
- xfs_attr_leafblock_t *leaf;
- xfs_attr_leaf_entry_t *entry;
- xfs_attr_leaf_name_local_t *name_loc;
- xfs_attr_leaf_name_remote_t *name_rmt;
+ struct xfs_attr_leafblock *leaf;
+ struct xfs_attr3_icleaf_hdr ichdr;
+ struct xfs_attr_leaf_entry *entry;
+ struct xfs_attr_leaf_name_local *name_loc;
+ struct xfs_attr_leaf_name_remote *name_rmt;
+ int valuelen;
leaf = bp->b_addr;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
- ASSERT(be16_to_cpu(leaf->hdr.count)
- < (XFS_LBSIZE(args->dp->i_mount)/8));
- ASSERT(args->index < be16_to_cpu(leaf->hdr.count));
+ xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf);
+ ASSERT(ichdr.count < XFS_LBSIZE(args->dp->i_mount) / 8);
+ ASSERT(args->index < ichdr.count);
- entry = &leaf->entries[args->index];
+ entry = &xfs_attr3_leaf_entryp(leaf)[args->index];
if (entry->flags & XFS_ATTR_LOCAL) {
- name_loc = xfs_attr_leaf_name_local(leaf, args->index);
+ name_loc = xfs_attr3_leaf_name_local(leaf, args->index);
ASSERT(name_loc->namelen == args->namelen);
ASSERT(memcmp(args->name, name_loc->nameval, args->namelen) == 0);
valuelen = be16_to_cpu(name_loc->valuelen);
if (args->flags & ATTR_KERNOVAL) {
args->valuelen = valuelen;
- return(0);
+ return 0;
}
if (args->valuelen < valuelen) {
args->valuelen = valuelen;
- return(XFS_ERROR(ERANGE));
+ return XFS_ERROR(ERANGE);
}
args->valuelen = valuelen;
memcpy(args->value, &name_loc->nameval[args->namelen], valuelen);
} else {
- name_rmt = xfs_attr_leaf_name_remote(leaf, args->index);
+ name_rmt = xfs_attr3_leaf_name_remote(leaf, args->index);
ASSERT(name_rmt->namelen == args->namelen);
ASSERT(memcmp(args->name, name_rmt->name, args->namelen) == 0);
valuelen = be32_to_cpu(name_rmt->valuelen);
@@ -1956,15 +2169,15 @@ xfs_attr_leaf_getvalue(
args->rmtblkcnt = XFS_B_TO_FSB(args->dp->i_mount, valuelen);
if (args->flags & ATTR_KERNOVAL) {
args->valuelen = valuelen;
- return(0);
+ return 0;
}
if (args->valuelen < valuelen) {
args->valuelen = valuelen;
- return(XFS_ERROR(ERANGE));
+ return XFS_ERROR(ERANGE);
}
args->valuelen = valuelen;
}
- return(0);
+ return 0;
}
/*========================================================================
@@ -1977,13 +2190,21 @@ xfs_attr_leaf_getvalue(
*/
/*ARGSUSED*/
STATIC void
-xfs_attr_leaf_moveents(xfs_attr_leafblock_t *leaf_s, int start_s,
- xfs_attr_leafblock_t *leaf_d, int start_d,
- int count, xfs_mount_t *mp)
+xfs_attr3_leaf_moveents(
+ struct xfs_attr_leafblock *leaf_s,
+ struct xfs_attr3_icleaf_hdr *ichdr_s,
+ int start_s,
+ struct xfs_attr_leafblock *leaf_d,
+ struct xfs_attr3_icleaf_hdr *ichdr_d,
+ int start_d,
+ int count,
+ struct xfs_mount *mp)
{
- xfs_attr_leaf_hdr_t *hdr_s, *hdr_d;
- xfs_attr_leaf_entry_t *entry_s, *entry_d;
- int desti, tmp, i;
+ struct xfs_attr_leaf_entry *entry_s;
+ struct xfs_attr_leaf_entry *entry_d;
+ int desti;
+ int tmp;
+ int i;
/*
* Check for nothing to do.
@@ -1994,45 +2215,41 @@ xfs_attr_leaf_moveents(xfs_attr_leafblock_t *leaf_s, int start_s,
/*
* Set up environment.
*/
- ASSERT(leaf_s->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
- ASSERT(leaf_d->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
- hdr_s = &leaf_s->hdr;
- hdr_d = &leaf_d->hdr;
- ASSERT((be16_to_cpu(hdr_s->count) > 0) &&
- (be16_to_cpu(hdr_s->count) < (XFS_LBSIZE(mp)/8)));
- ASSERT(be16_to_cpu(hdr_s->firstused) >=
- ((be16_to_cpu(hdr_s->count)
- * sizeof(*entry_s))+sizeof(*hdr_s)));
- ASSERT(be16_to_cpu(hdr_d->count) < (XFS_LBSIZE(mp)/8));
- ASSERT(be16_to_cpu(hdr_d->firstused) >=
- ((be16_to_cpu(hdr_d->count)
- * sizeof(*entry_d))+sizeof(*hdr_d)));
-
- ASSERT(start_s < be16_to_cpu(hdr_s->count));
- ASSERT(start_d <= be16_to_cpu(hdr_d->count));
- ASSERT(count <= be16_to_cpu(hdr_s->count));
+ ASSERT(ichdr_s->magic == XFS_ATTR_LEAF_MAGIC ||
+ ichdr_s->magic == XFS_ATTR3_LEAF_MAGIC);
+ ASSERT(ichdr_s->magic == ichdr_d->magic);
+ ASSERT(ichdr_s->count > 0 && ichdr_s->count < XFS_LBSIZE(mp) / 8);
+ ASSERT(ichdr_s->firstused >= (ichdr_s->count * sizeof(*entry_s))
+ + xfs_attr3_leaf_hdr_size(leaf_s));
+ ASSERT(ichdr_d->count < XFS_LBSIZE(mp) / 8);
+ ASSERT(ichdr_d->firstused >= (ichdr_d->count * sizeof(*entry_d))
+ + xfs_attr3_leaf_hdr_size(leaf_d));
+
+ ASSERT(start_s < ichdr_s->count);
+ ASSERT(start_d <= ichdr_d->count);
+ ASSERT(count <= ichdr_s->count);
+
/*
* Move the entries in the destination leaf up to make a hole?
*/
- if (start_d < be16_to_cpu(hdr_d->count)) {
- tmp = be16_to_cpu(hdr_d->count) - start_d;
+ if (start_d < ichdr_d->count) {
+ tmp = ichdr_d->count - start_d;
tmp *= sizeof(xfs_attr_leaf_entry_t);
- entry_s = &leaf_d->entries[start_d];
- entry_d = &leaf_d->entries[start_d + count];
- memmove((char *)entry_d, (char *)entry_s, tmp);
+ entry_s = &xfs_attr3_leaf_entryp(leaf_d)[start_d];
+ entry_d = &xfs_attr3_leaf_entryp(leaf_d)[start_d + count];
+ memmove(entry_d, entry_s, tmp);
}
/*
* Copy all entry's in the same (sorted) order,
* but allocate attribute info packed and in sequence.
*/
- entry_s = &leaf_s->entries[start_s];
- entry_d = &leaf_d->entries[start_d];
+ entry_s = &xfs_attr3_leaf_entryp(leaf_s)[start_s];
+ entry_d = &xfs_attr3_leaf_entryp(leaf_d)[start_d];
desti = start_d;
for (i = 0; i < count; entry_s++, entry_d++, desti++, i++) {
- ASSERT(be16_to_cpu(entry_s->nameidx)
- >= be16_to_cpu(hdr_s->firstused));
+ ASSERT(be16_to_cpu(entry_s->nameidx) >= ichdr_s->firstused);
tmp = xfs_attr_leaf_entsize(leaf_s, start_s + i);
#ifdef GROT
/*
@@ -2041,36 +2258,34 @@ xfs_attr_leaf_moveents(xfs_attr_leafblock_t *leaf_s, int start_s,
* off for 6.2, should be revisited later.
*/
if (entry_s->flags & XFS_ATTR_INCOMPLETE) { /* skip partials? */
- memset(xfs_attr_leaf_name(leaf_s, start_s + i), 0, tmp);
- be16_add_cpu(&hdr_s->usedbytes, -tmp);
- be16_add_cpu(&hdr_s->count, -1);
+ memset(xfs_attr3_leaf_name(leaf_s, start_s + i), 0, tmp);
+ ichdr_s->usedbytes -= tmp;
+ ichdr_s->count -= 1;
entry_d--; /* to compensate for ++ in loop hdr */
desti--;
if ((start_s + i) < offset)
result++; /* insertion index adjustment */
} else {
#endif /* GROT */
- be16_add_cpu(&hdr_d->firstused, -tmp);
+ ichdr_d->firstused -= tmp;
/* both on-disk, don't endian flip twice */
entry_d->hashval = entry_s->hashval;
- /* both on-disk, don't endian flip twice */
- entry_d->nameidx = hdr_d->firstused;
+ entry_d->nameidx = cpu_to_be16(ichdr_d->firstused);
entry_d->flags = entry_s->flags;
ASSERT(be16_to_cpu(entry_d->nameidx) + tmp
<= XFS_LBSIZE(mp));
- memmove(xfs_attr_leaf_name(leaf_d, desti),
- xfs_attr_leaf_name(leaf_s, start_s + i), tmp);
+ memmove(xfs_attr3_leaf_name(leaf_d, desti),
+ xfs_attr3_leaf_name(leaf_s, start_s + i), tmp);
ASSERT(be16_to_cpu(entry_s->nameidx) + tmp
<= XFS_LBSIZE(mp));
- memset(xfs_attr_leaf_name(leaf_s, start_s + i), 0, tmp);
- be16_add_cpu(&hdr_s->usedbytes, -tmp);
- be16_add_cpu(&hdr_d->usedbytes, tmp);
- be16_add_cpu(&hdr_s->count, -1);
- be16_add_cpu(&hdr_d->count, 1);
- tmp = be16_to_cpu(hdr_d->count)
- * sizeof(xfs_attr_leaf_entry_t)
- + sizeof(xfs_attr_leaf_hdr_t);
- ASSERT(be16_to_cpu(hdr_d->firstused) >= tmp);
+ memset(xfs_attr3_leaf_name(leaf_s, start_s + i), 0, tmp);
+ ichdr_s->usedbytes -= tmp;
+ ichdr_d->usedbytes += tmp;
+ ichdr_s->count -= 1;
+ ichdr_d->count += 1;
+ tmp = ichdr_d->count * sizeof(xfs_attr_leaf_entry_t)
+ + xfs_attr3_leaf_hdr_size(leaf_d);
+ ASSERT(ichdr_d->firstused >= tmp);
#ifdef GROT
}
#endif /* GROT */
@@ -2079,71 +2294,40 @@ xfs_attr_leaf_moveents(xfs_attr_leafblock_t *leaf_s, int start_s,
/*
* Zero out the entries we just copied.
*/
- if (start_s == be16_to_cpu(hdr_s->count)) {
+ if (start_s == ichdr_s->count) {
tmp = count * sizeof(xfs_attr_leaf_entry_t);
- entry_s = &leaf_s->entries[start_s];
+ entry_s = &xfs_attr3_leaf_entryp(leaf_s)[start_s];
ASSERT(((char *)entry_s + tmp) <=
((char *)leaf_s + XFS_LBSIZE(mp)));
- memset((char *)entry_s, 0, tmp);
+ memset(entry_s, 0, tmp);
} else {
/*
* Move the remaining entries down to fill the hole,
* then zero the entries at the top.
*/
- tmp = be16_to_cpu(hdr_s->count) - count;
- tmp *= sizeof(xfs_attr_leaf_entry_t);
- entry_s = &leaf_s->entries[start_s + count];
- entry_d = &leaf_s->entries[start_s];
- memmove((char *)entry_d, (char *)entry_s, tmp);
+ tmp = (ichdr_s->count - count) - sizeof(xfs_attr_leaf_entry_t);
+ entry_s = &xfs_attr3_leaf_entryp(leaf_s)[start_s + count];
+ entry_d = &xfs_attr3_leaf_entryp(leaf_s)[start_s];
+ memmove(entry_d, entry_s, tmp);
tmp = count * sizeof(xfs_attr_leaf_entry_t);
- entry_s = &leaf_s->entries[be16_to_cpu(hdr_s->count)];
+ entry_s = &xfs_attr3_leaf_entryp(leaf_s)[ichdr_s->count];
ASSERT(((char *)entry_s + tmp) <=
((char *)leaf_s + XFS_LBSIZE(mp)));
- memset((char *)entry_s, 0, tmp);
+ memset(entry_s, 0, tmp);
}
/*
* Fill in the freemap information
*/
- hdr_d->freemap[0].base = cpu_to_be16(sizeof(xfs_attr_leaf_hdr_t));
- be16_add_cpu(&hdr_d->freemap[0].base, be16_to_cpu(hdr_d->count) *
- sizeof(xfs_attr_leaf_entry_t));
- hdr_d->freemap[0].size = cpu_to_be16(be16_to_cpu(hdr_d->firstused)
- - be16_to_cpu(hdr_d->freemap[0].base));
- hdr_d->freemap[1].base = 0;
- hdr_d->freemap[2].base = 0;
- hdr_d->freemap[1].size = 0;
- hdr_d->freemap[2].size = 0;
- hdr_s->holes = 1; /* leaf may not be compact */
-}
-
-/*
- * Compare two leaf blocks "order".
- * Return 0 unless leaf2 should go before leaf1.
- */
-int
-xfs_attr_leaf_order(
- struct xfs_buf *leaf1_bp,
- struct xfs_buf *leaf2_bp)
-{
- xfs_attr_leafblock_t *leaf1, *leaf2;
-
- leaf1 = leaf1_bp->b_addr;
- leaf2 = leaf2_bp->b_addr;
- ASSERT((leaf1->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC)) &&
- (leaf2->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC)));
- if ((be16_to_cpu(leaf1->hdr.count) > 0) &&
- (be16_to_cpu(leaf2->hdr.count) > 0) &&
- ((be32_to_cpu(leaf2->entries[0].hashval) <
- be32_to_cpu(leaf1->entries[0].hashval)) ||
- (be32_to_cpu(leaf2->entries[
- be16_to_cpu(leaf2->hdr.count)-1].hashval) <
- be32_to_cpu(leaf1->entries[
- be16_to_cpu(leaf1->hdr.count)-1].hashval)))) {
- return(1);
- }
- return(0);
+ ichdr_d->freemap[0].base = xfs_attr3_leaf_hdr_size(leaf_d);
+ ichdr_d->freemap[0].base += ichdr_d->count * sizeof(xfs_attr_leaf_entry_t);
+ ichdr_d->freemap[0].size = ichdr_d->firstused - ichdr_d->freemap[0].base;
+ ichdr_d->freemap[1].base = 0;
+ ichdr_d->freemap[2].base = 0;
+ ichdr_d->freemap[1].size = 0;
+ ichdr_d->freemap[2].size = 0;
+ ichdr_s->holes = 1; /* leaf may not be compact */
}
/*
@@ -2154,15 +2338,16 @@ xfs_attr_leaf_lasthash(
struct xfs_buf *bp,
int *count)
{
- xfs_attr_leafblock_t *leaf;
+ struct xfs_attr3_icleaf_hdr ichdr;
+ struct xfs_attr_leaf_entry *entries;
- leaf = bp->b_addr;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
+ xfs_attr3_leaf_hdr_from_disk(&ichdr, bp->b_addr);
+ entries = xfs_attr3_leaf_entryp(bp->b_addr);
if (count)
- *count = be16_to_cpu(leaf->hdr.count);
- if (!leaf->hdr.count)
- return(0);
- return be32_to_cpu(leaf->entries[be16_to_cpu(leaf->hdr.count)-1].hashval);
+ *count = ichdr.count;
+ if (!ichdr.count)
+ return 0;
+ return be32_to_cpu(entries[ichdr.count - 1].hashval);
}
/*
@@ -2172,20 +2357,21 @@ xfs_attr_leaf_lasthash(
STATIC int
xfs_attr_leaf_entsize(xfs_attr_leafblock_t *leaf, int index)
{
+ struct xfs_attr_leaf_entry *entries;
xfs_attr_leaf_name_local_t *name_loc;
xfs_attr_leaf_name_remote_t *name_rmt;
int size;
- ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
- if (leaf->entries[index].flags & XFS_ATTR_LOCAL) {
- name_loc = xfs_attr_leaf_name_local(leaf, index);
+ entries = xfs_attr3_leaf_entryp(leaf);
+ if (entries[index].flags & XFS_ATTR_LOCAL) {
+ name_loc = xfs_attr3_leaf_name_local(leaf, index);
size = xfs_attr_leaf_entsize_local(name_loc->namelen,
be16_to_cpu(name_loc->valuelen));
} else {
- name_rmt = xfs_attr_leaf_name_remote(leaf, index);
+ name_rmt = xfs_attr3_leaf_name_remote(leaf, index);
size = xfs_attr_leaf_entsize_remote(name_rmt->namelen);
}
- return(size);
+ return size;
}
/*
@@ -2210,7 +2396,7 @@ xfs_attr_leaf_newentsize(int namelen, int valuelen, int blocksize, int *local)
*local = 0;
}
}
- return(size);
+ return size;
}
/*========================================================================
@@ -2221,14 +2407,16 @@ xfs_attr_leaf_newentsize(int namelen, int valuelen, int blocksize, int *local)
* Clear the INCOMPLETE flag on an entry in a leaf block.
*/
int
-xfs_attr_leaf_clearflag(xfs_da_args_t *args)
+xfs_attr3_leaf_clearflag(
+ struct xfs_da_args *args)
{
- xfs_attr_leafblock_t *leaf;
- xfs_attr_leaf_entry_t *entry;
- xfs_attr_leaf_name_remote_t *name_rmt;
- struct xfs_buf *bp;
- int error;
+ struct xfs_attr_leafblock *leaf;
+ struct xfs_attr_leaf_entry *entry;
+ struct xfs_attr_leaf_name_remote *name_rmt;
+ struct xfs_buf *bp;
+ int error;
#ifdef DEBUG
+ struct xfs_attr3_icleaf_hdr ichdr;
xfs_attr_leaf_name_local_t *name_loc;
int namelen;
char *name;
@@ -2238,23 +2426,25 @@ xfs_attr_leaf_clearflag(xfs_da_args_t *args)
/*
* Set up the operation.
*/
- error = xfs_attr_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+ error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
if (error)
return(error);
leaf = bp->b_addr;
- ASSERT(args->index < be16_to_cpu(leaf->hdr.count));
- ASSERT(args->index >= 0);
- entry = &leaf->entries[ args->index ];
+ entry = &xfs_attr3_leaf_entryp(leaf)[args->index];
ASSERT(entry->flags & XFS_ATTR_INCOMPLETE);
#ifdef DEBUG
+ xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf);
+ ASSERT(args->index < ichdr.count);
+ ASSERT(args->index >= 0);
+
if (entry->flags & XFS_ATTR_LOCAL) {
- name_loc = xfs_attr_leaf_name_local(leaf, args->index);
+ name_loc = xfs_attr3_leaf_name_local(leaf, args->index);
namelen = name_loc->namelen;
name = (char *)name_loc->nameval;
} else {
- name_rmt = xfs_attr_leaf_name_remote(leaf, args->index);
+ name_rmt = xfs_attr3_leaf_name_remote(leaf, args->index);
namelen = name_rmt->namelen;
name = (char *)name_rmt->name;
}
@@ -2269,7 +2459,7 @@ xfs_attr_leaf_clearflag(xfs_da_args_t *args)
if (args->rmtblkno) {
ASSERT((entry->flags & XFS_ATTR_LOCAL) == 0);
- name_rmt = xfs_attr_leaf_name_remote(leaf, args->index);
+ name_rmt = xfs_attr3_leaf_name_remote(leaf, args->index);
name_rmt->valueblk = cpu_to_be32(args->rmtblkno);
name_rmt->valuelen = cpu_to_be32(args->valuelen);
xfs_trans_log_buf(args->trans, bp,
@@ -2286,34 +2476,41 @@ xfs_attr_leaf_clearflag(xfs_da_args_t *args)
* Set the INCOMPLETE flag on an entry in a leaf block.
*/
int
-xfs_attr_leaf_setflag(xfs_da_args_t *args)
+xfs_attr3_leaf_setflag(
+ struct xfs_da_args *args)
{
- xfs_attr_leafblock_t *leaf;
- xfs_attr_leaf_entry_t *entry;
- xfs_attr_leaf_name_remote_t *name_rmt;
- struct xfs_buf *bp;
+ struct xfs_attr_leafblock *leaf;
+ struct xfs_attr_leaf_entry *entry;
+ struct xfs_attr_leaf_name_remote *name_rmt;
+ struct xfs_buf *bp;
int error;
+#ifdef DEBUG
+ struct xfs_attr3_icleaf_hdr ichdr;
+#endif
trace_xfs_attr_leaf_setflag(args);
/*
* Set up the operation.
*/
- error = xfs_attr_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
+ error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
if (error)
return(error);
leaf = bp->b_addr;
- ASSERT(args->index < be16_to_cpu(leaf->hdr.count));
+#ifdef DEBUG
+ xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf);
+ ASSERT(args->index < ichdr.count);
ASSERT(args->index >= 0);
- entry = &leaf->entries[ args->index ];
+#endif
+ entry = &xfs_attr3_leaf_entryp(leaf)[args->index];
ASSERT((entry->flags & XFS_ATTR_INCOMPLETE) == 0);
entry->flags |= XFS_ATTR_INCOMPLETE;
xfs_trans_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, entry, sizeof(*entry)));
if ((entry->flags & XFS_ATTR_LOCAL) == 0) {
- name_rmt = xfs_attr_leaf_name_remote(leaf, args->index);
+ name_rmt = xfs_attr3_leaf_name_remote(leaf, args->index);
name_rmt->valueblk = 0;
name_rmt->valuelen = 0;
xfs_trans_log_buf(args->trans, bp,
@@ -2334,14 +2531,20 @@ xfs_attr_leaf_setflag(xfs_da_args_t *args)
* Note that they could be in different blocks, or in the same block.
*/
int
-xfs_attr_leaf_flipflags(xfs_da_args_t *args)
+xfs_attr3_leaf_flipflags(
+ struct xfs_da_args *args)
{
- xfs_attr_leafblock_t *leaf1, *leaf2;
- xfs_attr_leaf_entry_t *entry1, *entry2;
- xfs_attr_leaf_name_remote_t *name_rmt;
- struct xfs_buf *bp1, *bp2;
+ struct xfs_attr_leafblock *leaf1;
+ struct xfs_attr_leafblock *leaf2;
+ struct xfs_attr_leaf_entry *entry1;
+ struct xfs_attr_leaf_entry *entry2;
+ struct xfs_attr_leaf_name_remote *name_rmt;
+ struct xfs_buf *bp1;
+ struct xfs_buf *bp2;
int error;
#ifdef DEBUG
+ struct xfs_attr3_icleaf_hdr ichdr1;
+ struct xfs_attr3_icleaf_hdr ichdr2;
xfs_attr_leaf_name_local_t *name_loc;
int namelen1, namelen2;
char *name1, *name2;
@@ -2352,7 +2555,7 @@ xfs_attr_leaf_flipflags(xfs_da_args_t *args)
/*
* Read the block containing the "old" attr
*/
- error = xfs_attr_leaf_read(args->trans, args->dp, args->blkno, -1, &bp1);
+ error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp1);
if (error)
return error;
@@ -2360,7 +2563,7 @@ xfs_attr_leaf_flipflags(xfs_da_args_t *args)
* Read the block containing the "new" attr, if it is different
*/
if (args->blkno2 != args->blkno) {
- error = xfs_attr_leaf_read(args->trans, args->dp, args->blkno2,
+ error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno2,
-1, &bp2);
if (error)
return error;
@@ -2369,31 +2572,35 @@ xfs_attr_leaf_flipflags(xfs_da_args_t *args)
}
leaf1 = bp1->b_addr;
- ASSERT(args->index < be16_to_cpu(leaf1->hdr.count));
- ASSERT(args->index >= 0);
- entry1 = &leaf1->entries[ args->index ];
+ entry1 = &xfs_attr3_leaf_entryp(leaf1)[args->index];
leaf2 = bp2->b_addr;
- ASSERT(args->index2 < be16_to_cpu(leaf2->hdr.count));
- ASSERT(args->index2 >= 0);
- entry2 = &leaf2->entries[ args->index2 ];
+ entry2 = &xfs_attr3_leaf_entryp(leaf2)[args->index2];
#ifdef DEBUG
+ xfs_attr3_leaf_hdr_from_disk(&ichdr1, leaf1);
+ ASSERT(args->index < ichdr1.count);
+ ASSERT(args->index >= 0);
+
+ xfs_attr3_leaf_hdr_from_disk(&ichdr2, leaf2);
+ ASSERT(args->index2 < ichdr2.count);
+ ASSERT(args->index2 >= 0);
+
if (entry1->flags & XFS_ATTR_LOCAL) {
- name_loc = xfs_attr_leaf_name_local(leaf1, args->index);
+ name_loc = xfs_attr3_leaf_name_local(leaf1, args->index);
namelen1 = name_loc->namelen;
name1 = (char *)name_loc->nameval;
} else {
- name_rmt = xfs_attr_leaf_name_remote(leaf1, args->index);
+ name_rmt = xfs_attr3_leaf_name_remote(leaf1, args->index);
namelen1 = name_rmt->namelen;
name1 = (char *)name_rmt->name;
}
if (entry2->flags & XFS_ATTR_LOCAL) {
- name_loc = xfs_attr_leaf_name_local(leaf2, args->index2);
+ name_loc = xfs_attr3_leaf_name_local(leaf2, args->index2);
namelen2 = name_loc->namelen;
name2 = (char *)name_loc->nameval;
} else {
- name_rmt = xfs_attr_leaf_name_remote(leaf2, args->index2);
+ name_rmt = xfs_attr3_leaf_name_remote(leaf2, args->index2);
namelen2 = name_rmt->namelen;
name2 = (char *)name_rmt->name;
}
@@ -2410,7 +2617,7 @@ xfs_attr_leaf_flipflags(xfs_da_args_t *args)
XFS_DA_LOGRANGE(leaf1, entry1, sizeof(*entry1)));
if (args->rmtblkno) {
ASSERT((entry1->flags & XFS_ATTR_LOCAL) == 0);
- name_rmt = xfs_attr_leaf_name_remote(leaf1, args->index);
+ name_rmt = xfs_attr3_leaf_name_remote(leaf1, args->index);
name_rmt->valueblk = cpu_to_be32(args->rmtblkno);
name_rmt->valuelen = cpu_to_be32(args->valuelen);
xfs_trans_log_buf(args->trans, bp1,
@@ -2421,7 +2628,7 @@ xfs_attr_leaf_flipflags(xfs_da_args_t *args)
xfs_trans_log_buf(args->trans, bp2,
XFS_DA_LOGRANGE(leaf2, entry2, sizeof(*entry2)));
if ((entry2->flags & XFS_ATTR_LOCAL) == 0) {
- name_rmt = xfs_attr_leaf_name_remote(leaf2, args->index2);
+ name_rmt = xfs_attr3_leaf_name_remote(leaf2, args->index2);
name_rmt->valueblk = 0;
name_rmt->valuelen = 0;
xfs_trans_log_buf(args->trans, bp2,
@@ -2433,5 +2640,5 @@ xfs_attr_leaf_flipflags(xfs_da_args_t *args)
*/
error = xfs_trans_roll(&args->trans, args->dp);
- return(error);
+ return error;
}
diff --git a/libxfs/xfs_da_btree.c b/libxfs/xfs_da_btree.c
index 3176626..ea16bee 100644
--- a/libxfs/xfs_da_btree.c
+++ b/libxfs/xfs_da_btree.c
@@ -120,14 +120,14 @@ xfs_da3_node_hdr_from_disk(
to->forw = be32_to_cpu(hdr3->info.hdr.forw);
to->back = be32_to_cpu(hdr3->info.hdr.back);
to->magic = be16_to_cpu(hdr3->info.hdr.magic);
- to->count = be16_to_cpu(hdr3->count);
+ to->count = be16_to_cpu(hdr3->__count);
to->level = be16_to_cpu(hdr3->__level);
return;
}
to->forw = be32_to_cpu(from->hdr.info.forw);
to->back = be32_to_cpu(from->hdr.info.back);
to->magic = be16_to_cpu(from->hdr.info.magic);
- to->count = be16_to_cpu(from->hdr.count);
+ to->count = be16_to_cpu(from->hdr.__count);
to->level = be16_to_cpu(from->hdr.__level);
}
@@ -145,14 +145,14 @@ xfs_da3_node_hdr_to_disk(
hdr3->info.hdr.forw = cpu_to_be32(from->forw);
hdr3->info.hdr.back = cpu_to_be32(from->back);
hdr3->info.hdr.magic = cpu_to_be16(from->magic);
- hdr3->count = cpu_to_be16(from->count);
+ hdr3->__count = cpu_to_be16(from->count);
hdr3->__level = cpu_to_be16(from->level);
return;
}
to->hdr.info.forw = cpu_to_be32(from->forw);
to->hdr.info.back = cpu_to_be32(from->back);
to->hdr.info.magic = cpu_to_be16(from->magic);
- to->hdr.count = cpu_to_be16(from->count);
+ to->hdr.__count = cpu_to_be16(from->count);
to->hdr.__level = cpu_to_be16(from->level);
}
@@ -247,7 +247,7 @@ xfs_da3_node_read_verify(
break;
return;
case XFS_ATTR_LEAF_MAGIC:
- bp->b_ops = &xfs_attr_leaf_buf_ops;
+ bp->b_ops = &xfs_attr3_leaf_buf_ops;
bp->b_ops->verify_read(bp);
return;
case XFS_DIR2_LEAFN_MAGIC:
@@ -378,7 +378,7 @@ xfs_da3_split(
*/
switch (oldblk->magic) {
case XFS_ATTR_LEAF_MAGIC:
- error = xfs_attr_leaf_split(state, oldblk, newblk);
+ error = xfs_attr3_leaf_split(state, oldblk, newblk);
if ((error != 0) && (error != ENOSPC)) {
return(error); /* GROT: attr is inconsistent */
}
@@ -393,12 +393,12 @@ xfs_da3_split(
if (state->inleaf) {
state->extraafter = 0; /* before newblk */
trace_xfs_attr_leaf_split_before(state->args);
- error = xfs_attr_leaf_split(state, oldblk,
+ error = xfs_attr3_leaf_split(state, oldblk,
&state->extrablk);
} else {
state->extraafter = 1; /* after newblk */
trace_xfs_attr_leaf_split_after(state->args);
- error = xfs_attr_leaf_split(state, newblk,
+ error = xfs_attr3_leaf_split(state, newblk,
&state->extrablk);
}
if (error)
@@ -938,12 +938,12 @@ xfs_da3_join(
*/
switch (drop_blk->magic) {
case XFS_ATTR_LEAF_MAGIC:
- error = xfs_attr_leaf_toosmall(state, &action);
+ error = xfs_attr3_leaf_toosmall(state, &action);
if (error)
return(error);
if (action == 0)
return(0);
- xfs_attr_leaf_unbalance(state, drop_blk, save_blk);
+ xfs_attr3_leaf_unbalance(state, drop_blk, save_blk);
break;
case XFS_DIR2_LEAFN_MAGIC:
error = xfs_dir2_leafn_toosmall(state, &action);
@@ -999,7 +999,8 @@ xfs_da_blkinfo_onlychild_validate(struct xfs_da_blkinfo *blkinfo, __u16 level)
if (level == 1) {
ASSERT(magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC) ||
- magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
+ magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC) ||
+ magic == cpu_to_be16(XFS_ATTR3_LEAF_MAGIC));
} else {
ASSERT(magic == cpu_to_be16(XFS_DA_NODE_MAGIC) ||
magic == cpu_to_be16(XFS_DA3_NODE_MAGIC));
@@ -1456,7 +1457,9 @@ xfs_da3_node_lookup_int(
curr = blk->bp->b_addr;
blk->magic = be16_to_cpu(curr->magic);
- if (blk->magic == XFS_ATTR_LEAF_MAGIC) {
+ if (blk->magic == XFS_ATTR_LEAF_MAGIC ||
+ blk->magic == XFS_ATTR3_LEAF_MAGIC) {
+ blk->magic = XFS_ATTR_LEAF_MAGIC;
blk->hashval = xfs_attr_leaf_lasthash(blk->bp, NULL);
break;
}
@@ -1536,7 +1539,7 @@ xfs_da3_node_lookup_int(
retval = xfs_dir2_leafn_lookup_int(blk->bp, args,
&blk->index, state);
} else if (blk->magic == XFS_ATTR_LEAF_MAGIC) {
- retval = xfs_attr_leaf_lookup_int(blk->bp, args);
+ retval = xfs_attr3_leaf_lookup_int(blk->bp, args);
blk->index = args->index;
args->blkno = blk->blkno;
} else {
@@ -1848,7 +1851,8 @@ xfs_da3_path_shift(
info->magic == cpu_to_be16(XFS_DA3_NODE_MAGIC) ||
info->magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
info->magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC) ||
- info->magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
+ info->magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC) ||
+ info->magic == cpu_to_be16(XFS_ATTR3_LEAF_MAGIC));
/*
@@ -1870,6 +1874,7 @@ xfs_da3_path_shift(
blkno = be32_to_cpu(btree[blk->index].before);
break;
case XFS_ATTR_LEAF_MAGIC:
+ case XFS_ATTR3_LEAF_MAGIC:
blk->magic = XFS_ATTR_LEAF_MAGIC;
ASSERT(level == path->active-1);
blk->index = 0;
@@ -2602,6 +2607,7 @@ xfs_da_read_buf(
XFS_TEST_ERROR((magic != XFS_DA_NODE_MAGIC) &&
(magic != XFS_DA3_NODE_MAGIC) &&
(magic != XFS_ATTR_LEAF_MAGIC) &&
+ (magic != XFS_ATTR3_LEAF_MAGIC) &&
(magic != XFS_DIR2_LEAF1_MAGIC) &&
(magic != XFS_DIR3_LEAF1_MAGIC) &&
(magic != XFS_DIR2_LEAFN_MAGIC) &&
diff --git a/repair/attr_repair.c b/repair/attr_repair.c
index ec7f4a3..033b7fe 100644
--- a/repair/attr_repair.c
+++ b/repair/attr_repair.c
@@ -381,7 +381,7 @@ process_leaf_attr_local(
{
xfs_attr_leaf_name_local_t *local;
- local = xfs_attr_leaf_name_local(leaf, i);
+ local = xfs_attr3_leaf_name_local(leaf, i);
if (local->namelen == 0 || namecheck((char *)&local->nameval[0],
local->namelen)) {
do_warn(
@@ -435,7 +435,7 @@ process_leaf_attr_remote(
xfs_attr_leaf_name_remote_t *remotep;
char* value;
- remotep = xfs_attr_leaf_name_remote(leaf, i);
+ remotep = xfs_attr3_leaf_name_remote(leaf, i);
if (remotep->namelen == 0 || namecheck((char *)&remotep->name[0],
remotep->namelen) ||
xfsprogs-crc-patchset-v2/xfs-attr-rmt-move 0000600 0001750 0001750 00000047642 12132011546 017513 0 ustar dave dave xfs: split remote attribute code out
From: Dave Chinner <dchinner@xxxxxxxxxx>
Adding CRC support to remote attributes adds a significant amount of
remote attribute specific code. Split the existing remote attribute
code out into it's own file so that all the relevant remote
attribute code is in a single, easy to find place.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
include/libxfs.h | 1 +
include/{xfs.h => xfs_attr_remote.h} | 23 +--
libxfs/Makefile | 2 +-
libxfs/xfs.h | 9 +-
libxfs/xfs_attr.c | 296 --------------------------------
libxfs/xfs_attr_remote.c | 306 ++++++++++++++++++++++++++++++++++
6 files changed, 322 insertions(+), 315 deletions(-)
diff --git a/include/libxfs.h b/include/libxfs.h
index 41cb585..972d850 100644
--- a/include/libxfs.h
+++ b/include/libxfs.h
@@ -589,6 +589,7 @@ extern unsigned long libxfs_physmem(void); /* in kilobytes */
#include <xfs/xfs_rtalloc.h>
#include <xfs/xfs_attr_leaf.h>
+#include <xfs/xfs_attr_remote.h>
#include <xfs/xfs_quota.h>
#include <xfs/xfs_trans_space.h>
#include <xfs/xfs_log.h>
diff --git a/include/xfs.h b/include/xfs_attr_remote.h
similarity index 69%
copy from include/xfs.h
copy to include/xfs_attr_remote.h
index e32c4a8..b4be90e 100644
--- a/include/xfs.h
+++ b/include/xfs_attr_remote.h
@@ -1,5 +1,5 @@
/*
- * Copyright (c) 2005 Silicon Graphics, Inc. All Rights Reserved.
+ * Copyright (c) 2013 Red Hat, Inc.
*
* This program is free software; you can redistribute it and/or modify it
* under the terms of version 2.1 of the GNU Lesser General Public License
@@ -20,21 +20,12 @@
* License along with this program; if not, write the Free Software
* Foundation, Inc., 59 Temple Place - Suite 330, Boston MA 02111-1307,
* USA.
- *
- * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy,
- * Mountain View, CA 94043, or:
- *
- * http://www.sgi.com
- *
- * For further information regarding this notice, see:
- *
- * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/
*/
-#ifndef __XFS_H__
-#define __XFS_H__
+#ifndef __XFS_ATTR_REMOTE_H__
+#define __XFS_ATTR_REMOTE_H__
-#include <xfs/platform_defs.h>
-#include <xfs/xfs_types.h>
-#include <xfs/xfs_fs.h>
+int xfs_attr_rmtval_get(struct xfs_da_args *args);
+int xfs_attr_rmtval_set(struct xfs_da_args *args);
+int xfs_attr_rmtval_remove(struct xfs_da_args *args);
-#endif /* __XFS_H__ */
+#endif /* __XFS_ATTR_REMOTE_H__ */
diff --git a/libxfs/Makefile b/libxfs/Makefile
index 75f365c..d0b483d 100644
--- a/libxfs/Makefile
+++ b/libxfs/Makefile
@@ -16,7 +16,7 @@ CFILES = cache.c init.c kmem.c logitem.c radix-tree.c rdwr.c trans.c util.c \
xfs_ialloc_btree.c xfs_bmap_btree.c xfs_da_btree.c \
xfs_dir2.c xfs_dir2_leaf.c xfs_attr_leaf.c xfs_dir2_block.c \
xfs_dir2_node.c xfs_dir2_data.c xfs_dir2_sf.c xfs_bmap.c \
- xfs_mount.c xfs_rtalloc.c xfs_trans.c xfs_attr.c \
+ xfs_mount.c xfs_rtalloc.c xfs_trans.c xfs_attr.c xfs_attr_remote.c \
crc32.c xfs_symlink.c
CFILES += $(PKG_PLATFORM).c
diff --git a/libxfs/xfs.h b/libxfs/xfs.h
index b3b45bb..c69dc4a 100644
--- a/libxfs/xfs.h
+++ b/libxfs/xfs.h
@@ -180,14 +180,19 @@ roundup_pow_of_two(uint v)
#define XFS_BUF_SET_VTYPE_REF(a,b,c) ((void) 0)
#define XFS_BUF_SET_BDSTRAT_FUNC(a,b) ((void) 0)
-#define xfs_incore(bt,blkno,len,lockit) 0
+/* avoid gcc warning */
+#define xfs_incore(bt,blkno,len,lockit) ({ \
+ typeof(blkno) __foo = (blkno); \
+ (blkno) = __foo; \
+ NULL; \
+})
#define xfs_buf_relse(bp) libxfs_putbuf(bp)
#define xfs_read_buf(mp,devp,blkno,len,f,bpp) \
(*(bpp) = libxfs_readbuf((devp), \
(blkno), (len), 1), 0)
#define xfs_buf_get(devp,blkno,len,f) \
(libxfs_getbuf((devp), (blkno), (len)))
-#define xfs_bwrite(mp,bp) libxfs_writebuf((bp), 0)
+#define xfs_bwrite(bp) libxfs_writebuf((bp), 0)
#define XBRW_READ LIBXFS_BREAD
#define XBRW_WRITE LIBXFS_BWRITE
diff --git a/libxfs/xfs_attr.c b/libxfs/xfs_attr.c
index 4429cb7..cfc2f4b 100644
--- a/libxfs/xfs_attr.c
+++ b/libxfs/xfs_attr.c
@@ -49,13 +49,6 @@ STATIC int xfs_attr_node_removename(xfs_da_args_t *args);
STATIC int xfs_attr_fillstate(xfs_da_state_t *state);
STATIC int xfs_attr_refillstate(xfs_da_state_t *state);
-/*
- * Routines to manipulate out-of-line attribute values.
- */
-STATIC int xfs_attr_rmtval_set(xfs_da_args_t *args);
-STATIC int xfs_attr_rmtval_remove(xfs_da_args_t *args);
-
-#define ATTR_RMTVALUE_MAPSIZE 1 /* # of map entries at once */
STATIC int
xfs_attr_name_to_xname(
@@ -1518,292 +1511,3 @@ xfs_attr_node_get(xfs_da_args_t *args)
xfs_da_state_free(state);
return(retval);
}
-
-/*========================================================================
- * External routines for manipulating out-of-line attribute values.
- *========================================================================*/
-
-/*
- * Read the value associated with an attribute from the out-of-line buffer
- * that we stored it in.
- */
-int
-xfs_attr_rmtval_get(xfs_da_args_t *args)
-{
- xfs_bmbt_irec_t map[ATTR_RMTVALUE_MAPSIZE];
- xfs_mount_t *mp;
- xfs_daddr_t dblkno;
- void *dst;
- xfs_buf_t *bp;
- int nmap, error, tmp, valuelen, blkcnt, i;
- xfs_dablk_t lblkno;
-
- trace_xfs_attr_rmtval_get(args);
-
- ASSERT(!(args->flags & ATTR_KERNOVAL));
-
- mp = args->dp->i_mount;
- dst = args->value;
- valuelen = args->valuelen;
- lblkno = args->rmtblkno;
- while (valuelen > 0) {
- nmap = ATTR_RMTVALUE_MAPSIZE;
- error = xfs_bmapi_read(args->dp, (xfs_fileoff_t)lblkno,
- args->rmtblkcnt, map, &nmap,
- XFS_BMAPI_ATTRFORK);
- if (error)
- return(error);
- ASSERT(nmap >= 1);
-
- for (i = 0; (i < nmap) && (valuelen > 0); i++) {
- ASSERT((map[i].br_startblock != DELAYSTARTBLOCK) &&
- (map[i].br_startblock != HOLESTARTBLOCK));
- dblkno = XFS_FSB_TO_DADDR(mp, map[i].br_startblock);
- blkcnt = XFS_FSB_TO_BB(mp, map[i].br_blockcount);
- error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp,
- dblkno, blkcnt, 0, &bp, NULL);
- if (error)
- return(error);
-
- tmp = min_t(int, valuelen, BBTOB(bp->b_length));
- xfs_buf_iomove(bp, 0, tmp, dst, XBRW_READ);
- xfs_buf_relse(bp);
- dst += tmp;
- valuelen -= tmp;
-
- lblkno += map[i].br_blockcount;
- }
- }
- ASSERT(valuelen == 0);
- return(0);
-}
-
-/*
- * Write the value associated with an attribute into the out-of-line buffer
- * that we have defined for it.
- */
-STATIC int
-xfs_attr_rmtval_set(xfs_da_args_t *args)
-{
- xfs_mount_t *mp;
- xfs_fileoff_t lfileoff;
- xfs_inode_t *dp;
- xfs_bmbt_irec_t map;
- xfs_daddr_t dblkno;
- void *src;
- xfs_buf_t *bp;
- xfs_dablk_t lblkno;
- int blkcnt, valuelen, nmap, error, tmp, committed;
-
- trace_xfs_attr_rmtval_set(args);
-
- dp = args->dp;
- mp = dp->i_mount;
- src = args->value;
-
- /*
- * Find a "hole" in the attribute address space large enough for
- * us to drop the new attribute's value into.
- */
- blkcnt = XFS_B_TO_FSB(mp, args->valuelen);
- lfileoff = 0;
- error = xfs_bmap_first_unused(args->trans, args->dp, blkcnt, &lfileoff,
- XFS_ATTR_FORK);
- if (error) {
- return(error);
- }
- args->rmtblkno = lblkno = (xfs_dablk_t)lfileoff;
- args->rmtblkcnt = blkcnt;
-
- /*
- * Roll through the "value", allocating blocks on disk as required.
- */
- while (blkcnt > 0) {
- /*
- * Allocate a single extent, up to the size of the value.
- */
- xfs_bmap_init(args->flist, args->firstblock);
- nmap = 1;
- error = xfs_bmapi_write(args->trans, dp, (xfs_fileoff_t)lblkno,
- blkcnt,
- XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,
- args->firstblock, args->total, &map, &nmap,
- args->flist);
- if (!error) {
- error = xfs_bmap_finish(&args->trans, args->flist,
- &committed);
- }
- if (error) {
- ASSERT(committed);
- args->trans = NULL;
- xfs_bmap_cancel(args->flist);
- return(error);
- }
-
- /*
- * bmap_finish() may have committed the last trans and started
- * a new one. We need the inode to be in all transactions.
- */
- if (committed)
- xfs_trans_ijoin(args->trans, dp, 0);
-
- ASSERT(nmap == 1);
- ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
- (map.br_startblock != HOLESTARTBLOCK));
- lblkno += map.br_blockcount;
- blkcnt -= map.br_blockcount;
-
- /*
- * Start the next trans in the chain.
- */
- error = xfs_trans_roll(&args->trans, dp);
- if (error)
- return (error);
- }
-
- /*
- * Roll through the "value", copying the attribute value to the
- * already-allocated blocks. Blocks are written synchronously
- * so that we can know they are all on disk before we turn off
- * the INCOMPLETE flag.
- */
- lblkno = args->rmtblkno;
- valuelen = args->valuelen;
- while (valuelen > 0) {
- int buflen;
-
- /*
- * Try to remember where we decided to put the value.
- */
- xfs_bmap_init(args->flist, args->firstblock);
- nmap = 1;
- error = xfs_bmapi_read(dp, (xfs_fileoff_t)lblkno,
- args->rmtblkcnt, &map, &nmap,
- XFS_BMAPI_ATTRFORK);
- if (error)
- return(error);
- ASSERT(nmap == 1);
- ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
- (map.br_startblock != HOLESTARTBLOCK));
-
- dblkno = XFS_FSB_TO_DADDR(mp, map.br_startblock),
- blkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount);
-
- bp = xfs_buf_get(mp->m_ddev_targp, dblkno, blkcnt, 0);
- if (!bp)
- return ENOMEM;
-
- buflen = BBTOB(bp->b_length);
- tmp = min_t(int, valuelen, buflen);
- xfs_buf_iomove(bp, 0, tmp, src, XBRW_WRITE);
- if (tmp < buflen)
- xfs_buf_zero(bp, tmp, buflen - tmp);
-
- error = xfs_bwrite(mp, bp); /* GROT: NOTE: synchronous write */
- xfs_buf_relse(bp);
- if (error)
- return error;
- src += tmp;
- valuelen -= tmp;
-
- lblkno += map.br_blockcount;
- }
- ASSERT(valuelen == 0);
- return(0);
-}
-
-/*
- * Remove the value associated with an attribute by deleting the
- * out-of-line buffer that it is stored on.
- */
-STATIC int
-xfs_attr_rmtval_remove(xfs_da_args_t *args)
-{
- xfs_mount_t *mp;
- xfs_bmbt_irec_t map;
- xfs_buf_t *bp;
- xfs_daddr_t dblkno;
- xfs_dablk_t lblkno;
- int valuelen, blkcnt, nmap, error, done, committed;
-
- trace_xfs_attr_rmtval_remove(args);
-
- mp = args->dp->i_mount;
-
- /*
- * Roll through the "value", invalidating the attribute value's
- * blocks.
- */
- lblkno = args->rmtblkno;
- valuelen = args->rmtblkcnt;
- while (valuelen > 0) {
- /*
- * Try to remember where we decided to put the value.
- */
- nmap = 1;
- error = xfs_bmapi_read(args->dp, (xfs_fileoff_t)lblkno,
- args->rmtblkcnt, &map, &nmap,
- XFS_BMAPI_ATTRFORK);
- if (error)
- return(error);
- ASSERT(nmap == 1);
- ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
- (map.br_startblock != HOLESTARTBLOCK));
-
- dblkno = XFS_FSB_TO_DADDR(mp, map.br_startblock),
- blkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount);
-
- /*
- * If the "remote" value is in the cache, remove it.
- */
- bp = xfs_incore(mp->m_ddev_targp, dblkno, blkcnt, XBF_TRYLOCK);
- if (bp) {
- xfs_buf_stale(bp);
- xfs_buf_relse(bp);
- bp = NULL;
- }
-
- valuelen -= map.br_blockcount;
-
- lblkno += map.br_blockcount;
- }
-
- /*
- * Keep de-allocating extents until the remote-value region is gone.
- */
- lblkno = args->rmtblkno;
- blkcnt = args->rmtblkcnt;
- done = 0;
- while (!done) {
- xfs_bmap_init(args->flist, args->firstblock);
- error = xfs_bunmapi(args->trans, args->dp, lblkno, blkcnt,
- XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,
- 1, args->firstblock, args->flist,
- &done);
- if (!error) {
- error = xfs_bmap_finish(&args->trans, args->flist,
- &committed);
- }
- if (error) {
- ASSERT(committed);
- args->trans = NULL;
- xfs_bmap_cancel(args->flist);
- return(error);
- }
-
- /*
- * bmap_finish() may have committed the last trans and started
- * a new one. We need the inode to be in all transactions.
- */
- if (committed)
- xfs_trans_ijoin(args->trans, args->dp, 0);
-
- /*
- * Close out trans and start the next one in the chain.
- */
- error = xfs_trans_roll(&args->trans, args->dp);
- if (error)
- return (error);
- }
- return(0);
-}
diff --git a/libxfs/xfs_attr_remote.c b/libxfs/xfs_attr_remote.c
new file mode 100644
index 0000000..36f8b5d
--- /dev/null
+++ b/libxfs/xfs_attr_remote.c
@@ -0,0 +1,306 @@
+/*
+ * Copyright (c) 2000-2005 Silicon Graphics, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+#include <xfs.h>
+
+#define ATTR_RMTVALUE_MAPSIZE 1 /* # of map entries at once */
+
+/*
+ * Read the value associated with an attribute from the out-of-line buffer
+ * that we stored it in.
+ */
+int
+xfs_attr_rmtval_get(xfs_da_args_t *args)
+{
+ xfs_bmbt_irec_t map[ATTR_RMTVALUE_MAPSIZE];
+ xfs_mount_t *mp;
+ xfs_daddr_t dblkno;
+ void *dst;
+ xfs_buf_t *bp;
+ int nmap, error, tmp, valuelen, blkcnt, i;
+ xfs_dablk_t lblkno;
+
+ trace_xfs_attr_rmtval_get(args);
+
+ ASSERT(!(args->flags & ATTR_KERNOVAL));
+
+ mp = args->dp->i_mount;
+ dst = args->value;
+ valuelen = args->valuelen;
+ lblkno = args->rmtblkno;
+ while (valuelen > 0) {
+ nmap = ATTR_RMTVALUE_MAPSIZE;
+ error = xfs_bmapi_read(args->dp, (xfs_fileoff_t)lblkno,
+ args->rmtblkcnt, map, &nmap,
+ XFS_BMAPI_ATTRFORK);
+ if (error)
+ return(error);
+ ASSERT(nmap >= 1);
+
+ for (i = 0; (i < nmap) && (valuelen > 0); i++) {
+ ASSERT((map[i].br_startblock != DELAYSTARTBLOCK) &&
+ (map[i].br_startblock != HOLESTARTBLOCK));
+ dblkno = XFS_FSB_TO_DADDR(mp, map[i].br_startblock);
+ blkcnt = XFS_FSB_TO_BB(mp, map[i].br_blockcount);
+ error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp,
+ dblkno, blkcnt, 0, &bp, NULL);
+ if (error)
+ return(error);
+
+ tmp = min_t(int, valuelen, BBTOB(bp->b_length));
+ xfs_buf_iomove(bp, 0, tmp, dst, XBRW_READ);
+ xfs_buf_relse(bp);
+ dst += tmp;
+ valuelen -= tmp;
+
+ lblkno += map[i].br_blockcount;
+ }
+ }
+ ASSERT(valuelen == 0);
+ return(0);
+}
+
+/*
+ * Write the value associated with an attribute into the out-of-line buffer
+ * that we have defined for it.
+ */
+int
+xfs_attr_rmtval_set(xfs_da_args_t *args)
+{
+ xfs_mount_t *mp;
+ xfs_fileoff_t lfileoff;
+ xfs_inode_t *dp;
+ xfs_bmbt_irec_t map;
+ xfs_daddr_t dblkno;
+ void *src;
+ xfs_buf_t *bp;
+ xfs_dablk_t lblkno;
+ int blkcnt, valuelen, nmap, error, tmp, committed;
+
+ trace_xfs_attr_rmtval_set(args);
+
+ dp = args->dp;
+ mp = dp->i_mount;
+ src = args->value;
+
+ /*
+ * Find a "hole" in the attribute address space large enough for
+ * us to drop the new attribute's value into.
+ */
+ blkcnt = XFS_B_TO_FSB(mp, args->valuelen);
+ lfileoff = 0;
+ error = xfs_bmap_first_unused(args->trans, args->dp, blkcnt, &lfileoff,
+ XFS_ATTR_FORK);
+ if (error) {
+ return(error);
+ }
+ args->rmtblkno = lblkno = (xfs_dablk_t)lfileoff;
+ args->rmtblkcnt = blkcnt;
+
+ /*
+ * Roll through the "value", allocating blocks on disk as required.
+ */
+ while (blkcnt > 0) {
+ /*
+ * Allocate a single extent, up to the size of the value.
+ */
+ xfs_bmap_init(args->flist, args->firstblock);
+ nmap = 1;
+ error = xfs_bmapi_write(args->trans, dp, (xfs_fileoff_t)lblkno,
+ blkcnt,
+ XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,
+ args->firstblock, args->total, &map, &nmap,
+ args->flist);
+ if (!error) {
+ error = xfs_bmap_finish(&args->trans, args->flist,
+ &committed);
+ }
+ if (error) {
+ ASSERT(committed);
+ args->trans = NULL;
+ xfs_bmap_cancel(args->flist);
+ return(error);
+ }
+
+ /*
+ * bmap_finish() may have committed the last trans and started
+ * a new one. We need the inode to be in all transactions.
+ */
+ if (committed)
+ xfs_trans_ijoin(args->trans, dp, 0);
+
+ ASSERT(nmap == 1);
+ ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
+ (map.br_startblock != HOLESTARTBLOCK));
+ lblkno += map.br_blockcount;
+ blkcnt -= map.br_blockcount;
+
+ /*
+ * Start the next trans in the chain.
+ */
+ error = xfs_trans_roll(&args->trans, dp);
+ if (error)
+ return (error);
+ }
+
+ /*
+ * Roll through the "value", copying the attribute value to the
+ * already-allocated blocks. Blocks are written synchronously
+ * so that we can know they are all on disk before we turn off
+ * the INCOMPLETE flag.
+ */
+ lblkno = args->rmtblkno;
+ valuelen = args->valuelen;
+ while (valuelen > 0) {
+ int buflen;
+
+ /*
+ * Try to remember where we decided to put the value.
+ */
+ xfs_bmap_init(args->flist, args->firstblock);
+ nmap = 1;
+ error = xfs_bmapi_read(dp, (xfs_fileoff_t)lblkno,
+ args->rmtblkcnt, &map, &nmap,
+ XFS_BMAPI_ATTRFORK);
+ if (error)
+ return(error);
+ ASSERT(nmap == 1);
+ ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
+ (map.br_startblock != HOLESTARTBLOCK));
+
+ dblkno = XFS_FSB_TO_DADDR(mp, map.br_startblock),
+ blkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount);
+
+ bp = xfs_buf_get(mp->m_ddev_targp, dblkno, blkcnt, 0);
+ if (!bp)
+ return ENOMEM;
+
+ buflen = BBTOB(bp->b_length);
+ tmp = min_t(int, valuelen, buflen);
+ xfs_buf_iomove(bp, 0, tmp, src, XBRW_WRITE);
+ if (tmp < buflen)
+ xfs_buf_zero(bp, tmp, buflen - tmp);
+
+ error = xfs_bwrite(bp); /* GROT: NOTE: synchronous write */
+ xfs_buf_relse(bp);
+ if (error)
+ return error;
+ src += tmp;
+ valuelen -= tmp;
+
+ lblkno += map.br_blockcount;
+ }
+ ASSERT(valuelen == 0);
+ return(0);
+}
+
+/*
+ * Remove the value associated with an attribute by deleting the
+ * out-of-line buffer that it is stored on.
+ */
+int
+xfs_attr_rmtval_remove(xfs_da_args_t *args)
+{
+ xfs_mount_t *mp;
+ xfs_bmbt_irec_t map;
+ xfs_buf_t *bp;
+ xfs_daddr_t dblkno;
+ xfs_dablk_t lblkno;
+ int valuelen, blkcnt, nmap, error, done, committed;
+
+ trace_xfs_attr_rmtval_remove(args);
+
+ mp = args->dp->i_mount;
+
+ /*
+ * Roll through the "value", invalidating the attribute value's
+ * blocks.
+ */
+ lblkno = args->rmtblkno;
+ valuelen = args->rmtblkcnt;
+ while (valuelen > 0) {
+ /*
+ * Try to remember where we decided to put the value.
+ */
+ nmap = 1;
+ error = xfs_bmapi_read(args->dp, (xfs_fileoff_t)lblkno,
+ args->rmtblkcnt, &map, &nmap,
+ XFS_BMAPI_ATTRFORK);
+ if (error)
+ return(error);
+ ASSERT(nmap == 1);
+ ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
+ (map.br_startblock != HOLESTARTBLOCK));
+
+ dblkno = XFS_FSB_TO_DADDR(mp, map.br_startblock),
+ blkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount);
+
+ /*
+ * If the "remote" value is in the cache, remove it.
+ */
+ bp = xfs_incore(mp->m_ddev_targp, dblkno, blkcnt, XBF_TRYLOCK);
+ if (bp) {
+ xfs_buf_stale(bp);
+ xfs_buf_relse(bp);
+ bp = NULL;
+ }
+
+ valuelen -= map.br_blockcount;
+
+ lblkno += map.br_blockcount;
+ }
+
+ /*
+ * Keep de-allocating extents until the remote-value region is gone.
+ */
+ lblkno = args->rmtblkno;
+ blkcnt = args->rmtblkcnt;
+ done = 0;
+ while (!done) {
+ xfs_bmap_init(args->flist, args->firstblock);
+ error = xfs_bunmapi(args->trans, args->dp, lblkno, blkcnt,
+ XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA,
+ 1, args->firstblock, args->flist,
+ &done);
+ if (!error) {
+ error = xfs_bmap_finish(&args->trans, args->flist,
+ &committed);
+ }
+ if (error) {
+ ASSERT(committed);
+ args->trans = NULL;
+ xfs_bmap_cancel(args->flist);
+ return(error);
+ }
+
+ /*
+ * bmap_finish() may have committed the last trans and started
+ * a new one. We need the inode to be in all transactions.
+ */
+ if (committed)
+ xfs_trans_ijoin(args->trans, args->dp, 0);
+
+ /*
+ * Close out trans and start the next one in the chain.
+ */
+ error = xfs_trans_roll(&args->trans, args->dp);
+ if (error)
+ return (error);
+ }
+ return(0);
+}
+
xfsprogs-crc-patchset-v2/xfs-add-crc-to-remote-attr 0000600 0001750 0001750 00000034450 12132011546 021144 0 ustar dave dave xfs: add CRC protection to remote attributes
From: Dave Chinner <dchinner@xxxxxxxxxx>
There are two ways of doing this - the first is to add a CRC to the
remote attribute entry in the attribute block. The second is to
treat them similar to the remote symlink, where each fragment has
it's own header and identifies fragment location in the attribute.
The problem with the CRC in the remote attr entry is that we cannot
identify the owner of the metadata from the metadata blocks
themselves, or where the blocks fit into the remote attribute. The
down side to this approach is that we never know when the attribute
has been read from disk or not and so we have to verify it every
time it is read, and we must calculate it during the create
transaction and log it. We do not log CRCs for any other metadata,
and so this creates a unique set of coherency problems that, in
general, are best avoided.
Adding an identifying header to each allocated block allows us to
identify each fragment and where in the attribute it is located. It
enables us to rebuild the remote attribute from just the raw blocks
containing the attribute. It also provides us to do per-block CRCs
verification at IO time rather than during the transaction context
that creates it or every time it is read into a user buffer. Hence
it avoids all the problems that an external, logged CRC has, and
provides all the benefits of self identifying metadata.
The only complexity is that we have to add a header per fragment,
and we don't know how many fragments will be needed prior to
allocations. If we take the symlink example, the header is 56 bytes
and hence for a 4k block size filesystem, in the worst case 16
headers requires 1 extra block for the 64k attribute data. For 512
byte filesystems the worst case is an extra block for every 9
fragments (i.e. 16 extra blocks in the worse case). This will be
very rare and so it's not really a major concern.
Because allocation is done in two steps - the first finds a hole
large enough in the attribute file, the second does the allocation -
we only need to find a hole big enough for a worst case allocation.
We only need to allocate enough extra blocks for number of headers
required by the fragments, and we can calculate that as we go....
Hence it really only makes sense to use the same model as for
symlinks - it doesn't add that much complexity, does not require an
attribute tree format change, and does not require logging
calculated CRC values.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
include/xfs_attr_remote.h | 19 +++
libxfs/xfs_attr_remote.c | 321 ++++++++++++++++++++++++++++++++++++++-------
2 files changed, 290 insertions(+), 50 deletions(-)
diff --git a/include/xfs_attr_remote.h b/include/xfs_attr_remote.h
index b4be90e..9e71edf 100644
--- a/include/xfs_attr_remote.h
+++ b/include/xfs_attr_remote.h
@@ -24,6 +24,25 @@
#ifndef __XFS_ATTR_REMOTE_H__
#define __XFS_ATTR_REMOTE_H__
+#define XFS_ATTR3_RMT_MAGIC 0x5841524d /* XARM */
+
+struct xfs_attr3_rmt_hdr {
+ __be32 rm_magic;
+ __be32 rm_offset;
+ __be32 rm_bytes;
+ __be32 rm_crc;
+ uuid_t rm_uuid;
+ __be64 rm_owner;
+ __be64 rm_blkno;
+ __be64 rm_lsn;
+};
+
+#define XFS_ATTR3_RMT_CRC_OFF offsetof(struct xfs_attr3_rmt_hdr, rm_crc)
+
+#define XFS_ATTR3_RMT_BUF_SPACE(mp, bufsize) \
+ ((bufsize) - (xfs_sb_version_hascrc(&(mp)->m_sb) ? \
+ sizeof(struct xfs_attr3_rmt_hdr) : 0))
+
int xfs_attr_rmtval_get(struct xfs_da_args *args);
int xfs_attr_rmtval_set(struct xfs_da_args *args);
int xfs_attr_rmtval_remove(struct xfs_da_args *args);
diff --git a/libxfs/xfs_attr_remote.c b/libxfs/xfs_attr_remote.c
index 36f8b5d..fa112ad 100644
--- a/libxfs/xfs_attr_remote.c
+++ b/libxfs/xfs_attr_remote.c
@@ -1,5 +1,6 @@
/*
* Copyright (c) 2000-2005 Silicon Graphics, Inc.
+ * Copyright (c) 2013 Red Hat, Inc.
* All Rights Reserved.
*
* This program is free software; you can redistribute it and/or
@@ -20,58 +21,226 @@
#define ATTR_RMTVALUE_MAPSIZE 1 /* # of map entries at once */
/*
+ * Each contiguous block has a header, so it is not just a simple attribute
+ * length to FSB conversion.
+ */
+static int
+xfs_attr3_rmt_blocks(
+ struct xfs_mount *mp,
+ int attrlen)
+{
+ int fsblocks = 0;
+ int len = attrlen;
+
+ do {
+ fsblocks++;
+ len -= XFS_ATTR3_RMT_BUF_SPACE(mp, mp->m_sb.sb_blocksize);
+ } while (len > 0);
+
+ return fsblocks;
+}
+
+static bool
+xfs_attr3_rmt_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_attr3_rmt_hdr *rmt = bp->b_addr;
+
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return false;
+ if (rmt->rm_magic != cpu_to_be32(XFS_ATTR3_RMT_MAGIC))
+ return false;
+ if (!uuid_equal(&rmt->rm_uuid, &mp->m_sb.sb_uuid))
+ return false;
+ if (bp->b_bn != be64_to_cpu(rmt->rm_blkno))
+ return false;
+ if (be32_to_cpu(rmt->rm_offset) +
+ be32_to_cpu(rmt->rm_bytes) >= MAXPATHLEN)
+ return false;
+ if (rmt->rm_owner == 0)
+ return false;
+
+ return true;
+}
+
+static void
+xfs_attr3_rmt_read_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+
+ /* no verification of non-crc buffers */
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return;
+
+ if (!xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
+ XFS_ATTR3_RMT_CRC_OFF) ||
+ !xfs_attr3_rmt_verify(bp)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+}
+
+static void
+xfs_attr3_rmt_write_verify(
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_buf_log_item *bip = bp->b_fspriv;
+
+ /* no verification of non-crc buffers */
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return;
+
+ if (!xfs_attr3_rmt_verify(bp)) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ return;
+ }
+
+ if (bip) {
+ struct xfs_attr3_rmt_hdr *rmt = bp->b_addr;
+ rmt->rm_lsn = cpu_to_be64(bip->bli_item.li_lsn);
+ }
+ xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length),
+ XFS_ATTR3_RMT_CRC_OFF);
+}
+
+const struct xfs_buf_ops xfs_attr3_rmt_buf_ops = {
+ .verify_read = xfs_attr3_rmt_read_verify,
+ .verify_write = xfs_attr3_rmt_write_verify,
+};
+
+static int
+xfs_attr3_rmt_hdr_set(
+ struct xfs_mount *mp,
+ xfs_ino_t ino,
+ uint32_t offset,
+ uint32_t size,
+ struct xfs_buf *bp)
+{
+ struct xfs_attr3_rmt_hdr *rmt = bp->b_addr;
+
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return 0;
+
+ rmt->rm_magic = cpu_to_be32(XFS_ATTR3_RMT_MAGIC);
+ rmt->rm_offset = cpu_to_be32(offset);
+ rmt->rm_bytes = cpu_to_be32(size);
+ uuid_copy(&rmt->rm_uuid, &mp->m_sb.sb_uuid);
+ rmt->rm_owner = cpu_to_be64(ino);
+ rmt->rm_blkno = cpu_to_be64(bp->b_bn);
+ bp->b_ops = &xfs_attr3_rmt_buf_ops;
+
+ return sizeof(struct xfs_attr3_rmt_hdr);
+}
+
+/*
+ * Checking of the remote attribute header is split into two parts. the verifier
+ * does CRC, location and bounds checking, the unpacking function checks the
+ * attribute parameters and owner.
+ */
+static bool
+xfs_attr3_rmt_hdr_ok(
+ struct xfs_mount *mp,
+ xfs_ino_t ino,
+ uint32_t offset,
+ uint32_t size,
+ struct xfs_buf *bp)
+{
+ struct xfs_attr3_rmt_hdr *rmt = bp->b_addr;
+
+ if (offset != be32_to_cpu(rmt->rm_offset))
+ return false;
+ if (size != be32_to_cpu(rmt->rm_bytes))
+ return false;
+ if (ino != be64_to_cpu(rmt->rm_owner))
+ return false;
+
+ /* ok */
+ return true;
+
+}
+
+/*
* Read the value associated with an attribute from the out-of-line buffer
* that we stored it in.
*/
int
-xfs_attr_rmtval_get(xfs_da_args_t *args)
+xfs_attr_rmtval_get(
+ struct xfs_da_args *args)
{
- xfs_bmbt_irec_t map[ATTR_RMTVALUE_MAPSIZE];
- xfs_mount_t *mp;
- xfs_daddr_t dblkno;
- void *dst;
- xfs_buf_t *bp;
- int nmap, error, tmp, valuelen, blkcnt, i;
- xfs_dablk_t lblkno;
+ struct xfs_bmbt_irec map[ATTR_RMTVALUE_MAPSIZE];
+ struct xfs_mount *mp = args->dp->i_mount;
+ struct xfs_buf *bp;
+ xfs_daddr_t dblkno;
+ xfs_dablk_t lblkno = args->rmtblkno;
+ void *dst = args->value;
+ int valuelen = args->valuelen;
+ int nmap;
+ int error;
+ int blkcnt;
+ int i;
+ int offset = 0;
trace_xfs_attr_rmtval_get(args);
ASSERT(!(args->flags & ATTR_KERNOVAL));
- mp = args->dp->i_mount;
- dst = args->value;
- valuelen = args->valuelen;
- lblkno = args->rmtblkno;
while (valuelen > 0) {
nmap = ATTR_RMTVALUE_MAPSIZE;
error = xfs_bmapi_read(args->dp, (xfs_fileoff_t)lblkno,
args->rmtblkcnt, map, &nmap,
XFS_BMAPI_ATTRFORK);
if (error)
- return(error);
+ return error;
ASSERT(nmap >= 1);
for (i = 0; (i < nmap) && (valuelen > 0); i++) {
+ int byte_cnt;
+ char *src;
+
ASSERT((map[i].br_startblock != DELAYSTARTBLOCK) &&
(map[i].br_startblock != HOLESTARTBLOCK));
dblkno = XFS_FSB_TO_DADDR(mp, map[i].br_startblock);
blkcnt = XFS_FSB_TO_BB(mp, map[i].br_blockcount);
error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp,
- dblkno, blkcnt, 0, &bp, NULL);
+ dblkno, blkcnt, 0, &bp,
+ &xfs_attr3_rmt_buf_ops);
if (error)
- return(error);
+ return error;
+
+ byte_cnt = min_t(int, valuelen, BBTOB(bp->b_length));
+ byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, byte_cnt);
- tmp = min_t(int, valuelen, BBTOB(bp->b_length));
- xfs_buf_iomove(bp, 0, tmp, dst, XBRW_READ);
+ src = bp->b_addr;
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ if (!xfs_attr3_rmt_hdr_ok(mp, args->dp->i_ino,
+ offset, byte_cnt, bp)) {
+ xfs_alert(mp,
+"remote attribute header does not match required off/len/owner (0x%x/Ox%x,0x%llx)",
+ offset, byte_cnt, args->dp->i_ino);
+ xfs_buf_relse(bp);
+ return EFSCORRUPTED;
+
+ }
+
+ src += sizeof(struct xfs_attr3_rmt_hdr);
+ }
+
+ memcpy(dst, src, byte_cnt);
xfs_buf_relse(bp);
- dst += tmp;
- valuelen -= tmp;
+
+ offset += byte_cnt;
+ dst += byte_cnt;
+ valuelen -= byte_cnt;
lblkno += map[i].br_blockcount;
}
}
ASSERT(valuelen == 0);
- return(0);
+ return 0;
}
/*
@@ -79,35 +248,49 @@ xfs_attr_rmtval_get(xfs_da_args_t *args)
* that we have defined for it.
*/
int
-xfs_attr_rmtval_set(xfs_da_args_t *args)
+xfs_attr_rmtval_set(
+ struct xfs_da_args *args)
{
- xfs_mount_t *mp;
- xfs_fileoff_t lfileoff;
- xfs_inode_t *dp;
- xfs_bmbt_irec_t map;
- xfs_daddr_t dblkno;
- void *src;
- xfs_buf_t *bp;
- xfs_dablk_t lblkno;
- int blkcnt, valuelen, nmap, error, tmp, committed;
+ struct xfs_inode *dp = args->dp;
+ struct xfs_mount *mp = dp->i_mount;
+ struct xfs_bmbt_irec map;
+ struct xfs_buf *bp;
+ xfs_daddr_t dblkno;
+ xfs_dablk_t lblkno;
+ xfs_fileoff_t lfileoff = 0;
+ void *src = args->value;
+ int blkcnt;
+ int valuelen;
+ int nmap;
+ int error;
+ int hdrcnt = 0;
+ bool crcs = xfs_sb_version_hascrc(&mp->m_sb);
+ int offset = 0;
trace_xfs_attr_rmtval_set(args);
- dp = args->dp;
- mp = dp->i_mount;
- src = args->value;
-
/*
* Find a "hole" in the attribute address space large enough for
- * us to drop the new attribute's value into.
+ * us to drop the new attribute's value into. Because CRC enable
+ * attributes have headers, we can't just do a straight byte to FSB
+ * conversion. We calculate the worst case block count in this case
+ * and we may not need that many, so we have to handle this when
+ * allocating the blocks below.
*/
- blkcnt = XFS_B_TO_FSB(mp, args->valuelen);
- lfileoff = 0;
+ if (!crcs)
+ blkcnt = XFS_B_TO_FSB(mp, args->valuelen);
+ else
+ blkcnt = xfs_attr3_rmt_blocks(mp, args->valuelen);
+
error = xfs_bmap_first_unused(args->trans, args->dp, blkcnt, &lfileoff,
XFS_ATTR_FORK);
- if (error) {
- return(error);
- }
+ if (error)
+ return error;
+
+ /* Start with the attribute data. We'll allocate the rest afterwards. */
+ if (crcs)
+ blkcnt = XFS_B_TO_FSB(mp, args->valuelen);
+
args->rmtblkno = lblkno = (xfs_dablk_t)lfileoff;
args->rmtblkcnt = blkcnt;
@@ -115,6 +298,8 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
* Roll through the "value", allocating blocks on disk as required.
*/
while (blkcnt > 0) {
+ int committed;
+
/*
* Allocate a single extent, up to the size of the value.
*/
@@ -148,6 +333,27 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
(map.br_startblock != HOLESTARTBLOCK));
lblkno += map.br_blockcount;
blkcnt -= map.br_blockcount;
+ hdrcnt++;
+
+ /*
+ * If we have enough blocks for the attribute data, calculate
+ * how many extra blocks we need for headers. We might run
+ * through this multiple times in the case that the additional
+ * headers in the blocks needed for the data fragments spills
+ * into requiring more blocks. e.g. for 512 byte blocks, we'll
+ * spill for another block every 9 headers we require in this
+ * loop.
+ */
+
+ if (crcs && blkcnt == 0) {
+ int total_len;
+
+ total_len = args->valuelen +
+ hdrcnt * sizeof(struct xfs_attr3_rmt_hdr);
+ blkcnt = XFS_B_TO_FSB(mp, total_len);
+ blkcnt -= args->rmtblkcnt;
+ args->rmtblkcnt += blkcnt;
+ }
/*
* Start the next trans in the chain.
@@ -166,7 +372,8 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
lblkno = args->rmtblkno;
valuelen = args->valuelen;
while (valuelen > 0) {
- int buflen;
+ int byte_cnt;
+ char *buf;
/*
* Try to remember where we decided to put the value.
@@ -188,24 +395,38 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
bp = xfs_buf_get(mp->m_ddev_targp, dblkno, blkcnt, 0);
if (!bp)
return ENOMEM;
+ bp->b_ops = &xfs_attr3_rmt_buf_ops;
+
+ byte_cnt = BBTOB(bp->b_length);
+ byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, byte_cnt);
+ if (valuelen < byte_cnt) {
+ byte_cnt = valuelen;
+ }
+
+ buf = bp->b_addr;
+ buf += xfs_attr3_rmt_hdr_set(mp, dp->i_ino, offset,
+ byte_cnt, bp);
+ memcpy(buf, src, byte_cnt);
- buflen = BBTOB(bp->b_length);
- tmp = min_t(int, valuelen, buflen);
- xfs_buf_iomove(bp, 0, tmp, src, XBRW_WRITE);
- if (tmp < buflen)
- xfs_buf_zero(bp, tmp, buflen - tmp);
+ if (byte_cnt < BBTOB(bp->b_length))
+ xfs_buf_zero(bp, byte_cnt,
+ BBTOB(bp->b_length) - byte_cnt);
error = xfs_bwrite(bp); /* GROT: NOTE: synchronous write */
xfs_buf_relse(bp);
if (error)
return error;
- src += tmp;
- valuelen -= tmp;
+
+ src += byte_cnt;
+ valuelen -= byte_cnt;
+ offset += byte_cnt;
+ hdrcnt--;
lblkno += map.br_blockcount;
}
ASSERT(valuelen == 0);
- return(0);
+ ASSERT(hdrcnt == 0);
+ return 0;
}
/*
@@ -284,7 +505,7 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args)
ASSERT(committed);
args->trans = NULL;
xfs_bmap_cancel(args->flist);
- return(error);
+ return error;
}
/*
xfsprogs-crc-patchset-v2/xfs-da-blf-types 0000600 0001750 0001750 00000036121 12132011547 017253 0 ustar dave dave xfs: add buffer types to directory and attribute buffers
From: Dave Chinner <dchinner@xxxxxxxxxx>
Add buffer types to the buffer log items so that log recovery can
validate the buffers and calculate CRCs correctly after the buffers
are recovered.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
include/xfs_attr_remote.h | 2 ++
include/xfs_buf_item.h | 18 +++++++++++++++++-
include/xfs_da_btree.h | 2 ++
include/xfs_trans.h | 2 ++
libxfs/xfs.h | 1 +
libxfs/xfs_attr_leaf.c | 9 ++++++++-
libxfs/xfs_da_btree.c | 46 ++++++++++++++++++++++++++++++++++++++++++---
libxfs/xfs_dir2_block.c | 12 +++++++++---
libxfs/xfs_dir2_data.c | 8 +++++++-
libxfs/xfs_dir2_leaf.c | 24 +++++++++++++++++++----
libxfs/xfs_dir2_node.c | 17 ++++++++++++++---
libxfs/xfs_dir2_priv.h | 2 ++
12 files changed, 127 insertions(+), 16 deletions(-)
diff --git a/include/xfs_attr_remote.h b/include/xfs_attr_remote.h
index 9e71edf..28f6f10 100644
--- a/include/xfs_attr_remote.h
+++ b/include/xfs_attr_remote.h
@@ -43,6 +43,8 @@ struct xfs_attr3_rmt_hdr {
((bufsize) - (xfs_sb_version_hascrc(&(mp)->m_sb) ? \
sizeof(struct xfs_attr3_rmt_hdr) : 0))
+extern const struct xfs_buf_ops xfs_attr3_rmt_buf_ops;
+
int xfs_attr_rmtval_get(struct xfs_da_args *args);
int xfs_attr_rmtval_set(struct xfs_da_args *args);
int xfs_attr_rmtval_remove(struct xfs_da_args *args);
diff --git a/include/xfs_buf_item.h b/include/xfs_buf_item.h
index 09cab4e..640adcf 100644
--- a/include/xfs_buf_item.h
+++ b/include/xfs_buf_item.h
@@ -50,6 +50,14 @@ extern kmem_zone_t *xfs_buf_item_zone;
#define XFS_BLF_AGI_BUF (1<<8)
#define XFS_BLF_DINO_BUF (1<<9)
#define XFS_BLF_SYMLINK_BUF (1<<10)
+#define XFS_BLF_DIR_BLOCK_BUF (1<<11)
+#define XFS_BLF_DIR_DATA_BUF (1<<12)
+#define XFS_BLF_DIR_FREE_BUF (1<<13)
+#define XFS_BLF_DIR_LEAF1_BUF (1<<14)
+#define XFS_BLF_DIR_LEAFN_BUF (1<<15)
+#define XFS_BLF_DA_NODE_BUF (1<<16)
+#define XFS_BLF_ATTR_LEAF_BUF (1<<17)
+#define XFS_BLF_ATTR_RMT_BUF (1<<18)
#define XFS_BLF_TYPE_MASK \
(XFS_BLF_UDQUOT_BUF | \
@@ -60,7 +68,15 @@ extern kmem_zone_t *xfs_buf_item_zone;
XFS_BLF_AGFL_BUF | \
XFS_BLF_AGI_BUF | \
XFS_BLF_DINO_BUF | \
- XFS_BLF_SYMLINK_BUF)
+ XFS_BLF_SYMLINK_BUF | \
+ XFS_BLF_DIR_BLOCK_BUF | \
+ XFS_BLF_DIR_DATA_BUF | \
+ XFS_BLF_DIR_FREE_BUF | \
+ XFS_BLF_DIR_LEAF1_BUF | \
+ XFS_BLF_DIR_LEAFN_BUF | \
+ XFS_BLF_DA_NODE_BUF | \
+ XFS_BLF_ATTR_LEAF_BUF | \
+ XFS_BLF_ATTR_RMT_BUF)
#define XFS_BLF_CHUNK 128
#define XFS_BLF_SHIFT 7
diff --git a/include/xfs_da_btree.h b/include/xfs_da_btree.h
index 0e8182c..6fb3371 100644
--- a/include/xfs_da_btree.h
+++ b/include/xfs_da_btree.h
@@ -301,6 +301,8 @@ int xfs_da3_node_read(struct xfs_trans *tp, struct xfs_inode *dp,
xfs_dablk_t bno, xfs_daddr_t mappedbno,
struct xfs_buf **bpp, int which_fork);
+extern const struct xfs_buf_ops xfs_da3_node_buf_ops;
+
/*
* Utility routines.
*/
diff --git a/include/xfs_trans.h b/include/xfs_trans.h
index a9bd826..9e145e9 100644
--- a/include/xfs_trans.h
+++ b/include/xfs_trans.h
@@ -502,6 +502,8 @@ void xfs_trans_dquot_buf(xfs_trans_t *, struct xfs_buf *, uint);
void xfs_trans_inode_alloc_buf(xfs_trans_t *, struct xfs_buf *);
void xfs_trans_buf_set_type(struct xfs_trans *, struct xfs_buf *,
uint);
+void xfs_trans_buf_copy_type(struct xfs_buf *dst_bp,
+ struct xfs_buf *src_bp);
void xfs_trans_ichgtime(struct xfs_trans *, struct xfs_inode *, int);
void xfs_trans_ijoin(struct xfs_trans *, struct xfs_inode *, uint);
void xfs_trans_log_buf(xfs_trans_t *, struct xfs_buf *, uint, uint);
diff --git a/libxfs/xfs.h b/libxfs/xfs.h
index c69dc4a..6bec18e 100644
--- a/libxfs/xfs.h
+++ b/libxfs/xfs.h
@@ -255,6 +255,7 @@ roundup_pow_of_two(uint v)
#define xfs_trans_agflist_delta(tp, d)
#define xfs_trans_agbtree_delta(tp, d)
#define xfs_trans_buf_set_type(tp, bp, t)
+#define xfs_trans_buf_copy_type(dbp, sbp)
#define xfs_buf_readahead(a,b,c,ops) ((void) 0) /* no readahead */
#define xfs_buf_readahead_map(a,b,c,ops) ((void) 0) /* no readahead */
diff --git a/libxfs/xfs_attr_leaf.c b/libxfs/xfs_attr_leaf.c
index 9de2244..7724781 100644
--- a/libxfs/xfs_attr_leaf.c
+++ b/libxfs/xfs_attr_leaf.c
@@ -236,8 +236,13 @@ xfs_attr3_leaf_read(
xfs_daddr_t mappedbno,
struct xfs_buf **bpp)
{
- return xfs_da_read_buf(tp, dp, bno, mappedbno, bpp,
+ int err;
+
+ err = xfs_da_read_buf(tp, dp, bno, mappedbno, bpp,
XFS_ATTR_FORK, &xfs_attr3_leaf_buf_ops);
+ if (!err && tp)
+ xfs_trans_buf_set_type(tp, *bpp, XFS_BLF_ATTR_LEAF_BUF);
+ return err;
}
/*========================================================================
@@ -867,6 +872,7 @@ xfs_attr3_leaf_to_node(
goto out;
/* copy leaf to new buffer, update identifiers */
+ xfs_trans_buf_set_type(args->trans, bp2, XFS_BLF_ATTR_LEAF_BUF);
bp2->b_ops = bp1->b_ops;
memcpy(bp2->b_addr, bp1->b_addr, XFS_LBSIZE(mp));
if (xfs_sb_version_hascrc(&mp->m_sb)) {
@@ -929,6 +935,7 @@ xfs_attr3_leaf_create(
if (error)
return error;
bp->b_ops = &xfs_attr3_leaf_buf_ops;
+ xfs_trans_buf_set_type(args->trans, bp, XFS_BLF_ATTR_LEAF_BUF);
leaf = bp->b_addr;
memset(leaf, 0, XFS_LBSIZE(mp));
diff --git a/libxfs/xfs_da_btree.c b/libxfs/xfs_da_btree.c
index ea16bee..565010d 100644
--- a/libxfs/xfs_da_btree.c
+++ b/libxfs/xfs_da_btree.c
@@ -269,7 +269,6 @@ const struct xfs_buf_ops xfs_da3_node_buf_ops = {
.verify_write = xfs_da3_node_write_verify,
};
-
int
xfs_da3_node_read(
struct xfs_trans *tp,
@@ -279,8 +278,35 @@ xfs_da3_node_read(
struct xfs_buf **bpp,
int which_fork)
{
- return xfs_da_read_buf(tp, dp, bno, mappedbno, bpp,
+ int err;
+
+ err = xfs_da_read_buf(tp, dp, bno, mappedbno, bpp,
which_fork, &xfs_da3_node_buf_ops);
+ if (!err && tp) {
+ struct xfs_da_blkinfo *info = (*bpp)->b_addr;
+ int type;
+
+ switch (be16_to_cpu(info->magic)) {
+ case XFS_DA3_NODE_MAGIC:
+ case XFS_DA_NODE_MAGIC:
+ type = XFS_BLF_DA_NODE_BUF;
+ break;
+ case XFS_ATTR_LEAF_MAGIC:
+ case XFS_ATTR3_LEAF_MAGIC:
+ type = XFS_BLF_ATTR_LEAF_BUF;
+ break;
+ case XFS_DIR2_LEAFN_MAGIC:
+ case XFS_DIR3_LEAFN_MAGIC:
+ type = XFS_BLF_DIR_LEAFN_BUF;
+ break;
+ default:
+ type = 0;
+ ASSERT(0);
+ break;
+ }
+ xfs_trans_buf_set_type(tp, *bpp, type);
+ }
+ return err;
}
/*========================================================================
@@ -311,6 +337,8 @@ xfs_da3_node_create(
error = xfs_da_get_buf(tp, args->dp, blkno, -1, &bp, whichfork);
if (error)
return(error);
+ bp->b_ops = &xfs_da3_node_buf_ops;
+ xfs_trans_buf_set_type(tp, bp, XFS_BLF_DA_NODE_BUF);
node = bp->b_addr;
if (xfs_sb_version_hascrc(&mp->m_sb)) {
@@ -329,7 +357,6 @@ xfs_da3_node_create(
xfs_trans_log_buf(tp, bp,
XFS_DA_LOGRANGE(node, &node->hdr, xfs_da3_node_hdr_size(node)));
- bp->b_ops = &xfs_da3_node_buf_ops;
*bpp = bp;
return(0);
}
@@ -540,6 +567,12 @@ xfs_da3_root_split(
btree = xfs_da3_node_tree_p(oldroot);
size = (int)((char *)&btree[nodehdr.count] - (char *)oldroot);
level = nodehdr.level;
+
+ /*
+ * we are about to copy oldroot to bp, so set up the type
+ * of bp while we know exactly what it will be.
+ */
+ xfs_trans_buf_set_type(tp, bp, XFS_BLF_DA_NODE_BUF);
} else {
struct xfs_dir3_icleaf_hdr leafhdr;
struct xfs_dir2_leaf_entry *ents;
@@ -552,6 +585,12 @@ xfs_da3_root_split(
leafhdr.magic == XFS_DIR3_LEAFN_MAGIC);
size = (int)((char *)&ents[leafhdr.count] - (char *)leaf);
level = 0;
+
+ /*
+ * we are about to copy oldroot to bp, so set up the type
+ * of bp while we know exactly what it will be.
+ */
+ xfs_trans_buf_set_type(tp, bp, XFS_BLF_DIR_LEAFN_BUF);
}
/*
@@ -1067,6 +1106,7 @@ xfs_da3_root_join(
*/
memcpy(root_blk->bp->b_addr, bp->b_addr, state->blocksize);
root_blk->bp->b_ops = bp->b_ops;
+ xfs_trans_buf_copy_type(root_blk->bp, bp);
if (oldroothdr.magic == XFS_DA3_NODE_MAGIC) {
struct xfs_da3_blkinfo *da3 = root_blk->bp->b_addr;
da3->blkno = cpu_to_be64(root_blk->bp->b_bn);
diff --git a/libxfs/xfs_dir2_block.c b/libxfs/xfs_dir2_block.c
index b98b749..574e414 100644
--- a/libxfs/xfs_dir2_block.c
+++ b/libxfs/xfs_dir2_block.c
@@ -114,20 +114,26 @@ xfs_dir3_block_read(
struct xfs_buf **bpp)
{
struct xfs_mount *mp = dp->i_mount;
+ int err;
- return xfs_da_read_buf(tp, dp, mp->m_dirdatablk, -1, bpp,
+ err = xfs_da_read_buf(tp, dp, mp->m_dirdatablk, -1, bpp,
XFS_DATA_FORK, &xfs_dir3_block_buf_ops);
+ if (!err && tp)
+ xfs_trans_buf_set_type(tp, *bpp, XFS_BLF_DIR_BLOCK_BUF);
+ return err;
}
static void
xfs_dir3_block_init(
struct xfs_mount *mp,
+ struct xfs_trans *tp,
struct xfs_buf *bp,
struct xfs_inode *dp)
{
struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr;
bp->b_ops = &xfs_dir3_block_buf_ops;
+ xfs_trans_buf_set_type(tp, bp, XFS_BLF_DIR_BLOCK_BUF);
if (xfs_sb_version_hascrc(&mp->m_sb)) {
memset(hdr3, 0, sizeof(*hdr3));
@@ -964,7 +970,7 @@ xfs_dir2_leaf_to_block(
/*
* Start converting it to block form.
*/
- xfs_dir3_block_init(mp, dbp, dp);
+ xfs_dir3_block_init(mp, tp, dbp, dp);
needlog = 1;
needscan = 0;
@@ -1093,7 +1099,7 @@ xfs_dir2_sf_to_block(
kmem_free(sfp);
return error;
}
- xfs_dir3_block_init(mp, bp, dp);
+ xfs_dir3_block_init(mp, tp, bp, dp);
hdr = bp->b_addr;
/*
diff --git a/libxfs/xfs_dir2_data.c b/libxfs/xfs_dir2_data.c
index 69841df..9752ae3 100644
--- a/libxfs/xfs_dir2_data.c
+++ b/libxfs/xfs_dir2_data.c
@@ -283,8 +283,13 @@ xfs_dir3_data_read(
xfs_daddr_t mapped_bno,
struct xfs_buf **bpp)
{
- return xfs_da_read_buf(tp, dp, bno, mapped_bno, bpp,
+ int err;
+
+ err = xfs_da_read_buf(tp, dp, bno, mapped_bno, bpp,
XFS_DATA_FORK, &xfs_dir3_data_buf_ops);
+ if (!err && tp)
+ xfs_trans_buf_set_type(tp, *bpp, XFS_BLF_DIR_DATA_BUF);
+ return err;
}
int
@@ -553,6 +558,7 @@ xfs_dir3_data_init(
if (error)
return error;
bp->b_ops = &xfs_dir3_data_buf_ops;
+ xfs_trans_buf_set_type(tp, bp, XFS_BLF_DIR_DATA_BUF);
/*
* Initialize the header.
diff --git a/libxfs/xfs_dir2_leaf.c b/libxfs/xfs_dir2_leaf.c
index f00b23c..3d1ec23 100644
--- a/libxfs/xfs_dir2_leaf.c
+++ b/libxfs/xfs_dir2_leaf.c
@@ -279,8 +279,13 @@ xfs_dir3_leaf_read(
xfs_daddr_t mappedbno,
struct xfs_buf **bpp)
{
- return xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
+ int err;
+
+ err = xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
XFS_DATA_FORK, &xfs_dir3_leaf1_buf_ops);
+ if (!err && tp)
+ xfs_trans_buf_set_type(tp, *bpp, XFS_BLF_DIR_LEAF1_BUF);
+ return err;
}
int
@@ -291,8 +296,13 @@ xfs_dir3_leafn_read(
xfs_daddr_t mappedbno,
struct xfs_buf **bpp)
{
- return xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
+ int err;
+
+ err = xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
XFS_DATA_FORK, &xfs_dir3_leafn_buf_ops);
+ if (!err && tp)
+ xfs_trans_buf_set_type(tp, *bpp, XFS_BLF_DIR_LEAFN_BUF);
+ return err;
}
/*
@@ -301,6 +311,7 @@ xfs_dir3_leafn_read(
static void
xfs_dir3_leaf_init(
struct xfs_mount *mp,
+ struct xfs_trans *tp,
struct xfs_buf *bp,
xfs_ino_t owner,
__uint16_t type)
@@ -335,8 +346,11 @@ xfs_dir3_leaf_init(
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
ltp->bestcount = 0;
bp->b_ops = &xfs_dir3_leaf1_buf_ops;
- } else
+ xfs_trans_buf_set_type(tp, bp, XFS_BLF_DIR_LEAF1_BUF);
+ } else {
bp->b_ops = &xfs_dir3_leafn_buf_ops;
+ xfs_trans_buf_set_type(tp, bp, XFS_BLF_DIR_LEAFN_BUF);
+ }
}
int
@@ -361,7 +375,7 @@ xfs_dir3_leaf_get_buf(
if (error)
return error;
- xfs_dir3_leaf_init(mp, bp, dp->i_ino, magic);
+ xfs_dir3_leaf_init(mp, tp, bp, dp->i_ino, magic);
xfs_dir3_leaf_log_header(tp, bp);
if (magic == XFS_DIR2_LEAF1_MAGIC)
xfs_dir3_leaf_log_tail(tp, bp);
@@ -456,6 +470,7 @@ xfs_dir2_block_to_leaf(
* Fix up the block header, make it a data block.
*/
dbp->b_ops = &xfs_dir3_data_buf_ops;
+ xfs_trans_buf_set_type(tp, dbp, XFS_BLF_DIR_DATA_BUF);
if (hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC))
hdr->magic = cpu_to_be32(XFS_DIR2_DATA_MAGIC);
else
@@ -1776,6 +1791,7 @@ xfs_dir2_node_to_leaf(
xfs_dir3_leaf_compact(args, &leafhdr, lbp);
lbp->b_ops = &xfs_dir3_leaf1_buf_ops;
+ xfs_trans_buf_set_type(tp, lbp, XFS_BLF_DIR_LEAF1_BUF);
leafhdr.magic = (leafhdr.magic == XFS_DIR2_LEAFN_MAGIC)
? XFS_DIR2_LEAF1_MAGIC
: XFS_DIR3_LEAF1_MAGIC;
diff --git a/libxfs/xfs_dir2_node.c b/libxfs/xfs_dir2_node.c
index 9e75553..a88049b 100644
--- a/libxfs/xfs_dir2_node.c
+++ b/libxfs/xfs_dir2_node.c
@@ -130,7 +130,7 @@ xfs_dir3_free_write_verify(
xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), XFS_DIR3_FREE_CRC_OFF);
}
-static const struct xfs_buf_ops xfs_dir3_free_buf_ops = {
+const struct xfs_buf_ops xfs_dir3_free_buf_ops = {
.verify_read = xfs_dir3_free_read_verify,
.verify_write = xfs_dir3_free_write_verify,
};
@@ -144,8 +144,15 @@ __xfs_dir3_free_read(
xfs_daddr_t mappedbno,
struct xfs_buf **bpp)
{
- return xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
+ int err;
+
+ err = xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
XFS_DATA_FORK, &xfs_dir3_free_buf_ops);
+
+ /* try read returns without an error or *bpp if it lands in a hole */
+ if (!err && tp && *bpp)
+ xfs_trans_buf_set_type(tp, *bpp, XFS_BLF_DIR_FREE_BUF);
+ return err;
}
int
@@ -232,7 +239,8 @@ xfs_dir3_free_get_buf(
if (error)
return error;
- bp->b_ops = &xfs_dir3_free_buf_ops;;
+ xfs_trans_buf_set_type(tp, bp, XFS_BLF_DIR_FREE_BUF);
+ bp->b_ops = &xfs_dir3_free_buf_ops;
/*
* Initialize the new block to be empty, and remember
@@ -380,6 +388,7 @@ xfs_dir2_leaf_to_node(
else
leaf->hdr.info.magic = cpu_to_be16(XFS_DIR3_LEAFN_MAGIC);
lbp->b_ops = &xfs_dir3_leafn_buf_ops;
+ xfs_trans_buf_set_type(tp, lbp, XFS_BLF_DIR_LEAFN_BUF);
xfs_dir3_leaf_log_header(tp, lbp);
xfs_dir3_leaf_check(mp, lbp);
return 0;
@@ -795,6 +804,7 @@ xfs_dir2_leafn_lookup_for_entry(
(char *)curbp->b_addr);
state->extrablk.magic = XFS_DIR2_DATA_MAGIC;
curbp->b_ops = &xfs_dir3_data_buf_ops;
+ xfs_trans_buf_set_type(tp, curbp, XFS_BLF_DIR_DATA_BUF);
if (cmp == XFS_CMP_EXACT)
return XFS_ERROR(EEXIST);
}
@@ -809,6 +819,7 @@ xfs_dir2_leafn_lookup_for_entry(
state->extrablk.blkno = curdb;
state->extrablk.magic = XFS_DIR2_DATA_MAGIC;
curbp->b_ops = &xfs_dir3_data_buf_ops;
+ xfs_trans_buf_set_type(tp, curbp, XFS_BLF_DIR_DATA_BUF);
} else {
/* If the curbp is not the CI match block, drop it */
if (state->extrablk.bp != curbp)
diff --git a/libxfs/xfs_dir2_priv.h b/libxfs/xfs_dir2_priv.h
index 932565d..7cf573c 100644
--- a/libxfs/xfs_dir2_priv.h
+++ b/libxfs/xfs_dir2_priv.h
@@ -49,6 +49,7 @@ extern int xfs_dir2_leaf_to_block(struct xfs_da_args *args,
#endif
extern const struct xfs_buf_ops xfs_dir3_data_buf_ops;
+extern const struct xfs_buf_ops xfs_dir3_free_buf_ops;
extern int __xfs_dir3_data_check(struct xfs_inode *dp, struct xfs_buf *bp);
extern int xfs_dir3_data_read(struct xfs_trans *tp, struct xfs_inode *dp,
@@ -77,6 +78,7 @@ extern void xfs_dir2_data_use_free(struct xfs_trans *tp, struct xfs_buf *bp,
xfs_dir2_data_aoff_t len, int *needlogp, int *needscanp);
/* xfs_dir2_leaf.c */
+extern const struct xfs_buf_ops xfs_dir3_leaf1_buf_ops;
extern const struct xfs_buf_ops xfs_dir3_leafn_buf_ops;
extern int xfs_dir3_leafn_read(struct xfs_trans *tp, struct xfs_inode *dp,
xfsprogs-crc-patchset-v2/xfs-trans-type-blf-change 0000600 0001750 0001750 00000041300 12132011547 021051 0 ustar dave dave xfs: buffer type overruns blf_flags field
From: Dave Chinner <dchinner@xxxxxxxxxx>
The buffer type passed to log recvoery in the buffer log item
overruns the blf_flags field. I had assumed that flags field was a
32 bit value, and it turns out it is a unisgned short. Therefore
having 19 flags doesn't really work.
Convert the buffer type field to numeric value, and use the top 5
bits of the flags field for it. We currently have 17 types of
buffers, so using 5 bits gives us plenty of room for expansion in
future....
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
include/xfs_buf_item.h | 92 +++++++++++++++++++++++++++--------------------
include/xfs_trans.h | 4 ---
libxfs/trans.c | 2 +-
libxfs/xfs_alloc.c | 4 +--
libxfs/xfs_attr_leaf.c | 6 ++--
libxfs/xfs_bmap.c | 2 +-
libxfs/xfs_btree.c | 8 ++---
libxfs/xfs_da_btree.c | 12 +++----
libxfs/xfs_dir2_block.c | 4 +--
libxfs/xfs_dir2_data.c | 4 +--
libxfs/xfs_dir2_leaf.c | 12 +++----
libxfs/xfs_dir2_node.c | 10 +++---
libxfs/xfs_ialloc.c | 2 +-
13 files changed, 86 insertions(+), 76 deletions(-)
diff --git a/include/xfs_buf_item.h b/include/xfs_buf_item.h
index 640adcf..2573d2a 100644
--- a/include/xfs_buf_item.h
+++ b/include/xfs_buf_item.h
@@ -39,45 +39,6 @@ extern kmem_zone_t *xfs_buf_item_zone;
#define XFS_BLF_PDQUOT_BUF (1<<3)
#define XFS_BLF_GDQUOT_BUF (1<<4)
-/*
- * all buffers now need flags to tell recovery where the magic number
- * is so that it can verify and calculate the CRCs on the buffer correctly
- * once the changes have been replayed into the buffer.
- */
-#define XFS_BLF_BTREE_BUF (1<<5)
-#define XFS_BLF_AGF_BUF (1<<6)
-#define XFS_BLF_AGFL_BUF (1<<7)
-#define XFS_BLF_AGI_BUF (1<<8)
-#define XFS_BLF_DINO_BUF (1<<9)
-#define XFS_BLF_SYMLINK_BUF (1<<10)
-#define XFS_BLF_DIR_BLOCK_BUF (1<<11)
-#define XFS_BLF_DIR_DATA_BUF (1<<12)
-#define XFS_BLF_DIR_FREE_BUF (1<<13)
-#define XFS_BLF_DIR_LEAF1_BUF (1<<14)
-#define XFS_BLF_DIR_LEAFN_BUF (1<<15)
-#define XFS_BLF_DA_NODE_BUF (1<<16)
-#define XFS_BLF_ATTR_LEAF_BUF (1<<17)
-#define XFS_BLF_ATTR_RMT_BUF (1<<18)
-
-#define XFS_BLF_TYPE_MASK \
- (XFS_BLF_UDQUOT_BUF | \
- XFS_BLF_PDQUOT_BUF | \
- XFS_BLF_GDQUOT_BUF | \
- XFS_BLF_BTREE_BUF | \
- XFS_BLF_AGF_BUF | \
- XFS_BLF_AGFL_BUF | \
- XFS_BLF_AGI_BUF | \
- XFS_BLF_DINO_BUF | \
- XFS_BLF_SYMLINK_BUF | \
- XFS_BLF_DIR_BLOCK_BUF | \
- XFS_BLF_DIR_DATA_BUF | \
- XFS_BLF_DIR_FREE_BUF | \
- XFS_BLF_DIR_LEAF1_BUF | \
- XFS_BLF_DIR_LEAFN_BUF | \
- XFS_BLF_DA_NODE_BUF | \
- XFS_BLF_ATTR_LEAF_BUF | \
- XFS_BLF_ATTR_RMT_BUF)
-
#define XFS_BLF_CHUNK 128
#define XFS_BLF_SHIFT 7
#define BIT_TO_WORD_SHIFT 5
@@ -101,6 +62,55 @@ typedef struct xfs_buf_log_format {
} xfs_buf_log_format_t;
/*
+ * All buffers now need to tell recovery where the magic number
+ * is so that it can verify and calculate the CRCs on the buffer correctly
+ * once the changes have been replayed into the buffer.
+ *
+ * The type value is held in the upper 5 bits of the blf_flags field, which is
+ * an unsigned 16 bit field. Hence we need to shift it 11 bits up and down.
+ */
+#define XFS_BLFT_BITS 5
+#define XFS_BLFT_SHIFT 11
+#define XFS_BLFT_MASK (((1 << XFS_BLFT_BITS) - 1) << XFS_BLFT_SHIFT)
+
+enum xfs_blft {
+ XFS_BLFT_UNKNOWN_BUF = 0,
+ XFS_BLFT_UDQUOT_BUF,
+ XFS_BLFT_PDQUOT_BUF,
+ XFS_BLFT_GDQUOT_BUF,
+ XFS_BLFT_BTREE_BUF,
+ XFS_BLFT_AGF_BUF,
+ XFS_BLFT_AGFL_BUF,
+ XFS_BLFT_AGI_BUF,
+ XFS_BLFT_DINO_BUF,
+ XFS_BLFT_SYMLINK_BUF,
+ XFS_BLFT_DIR_BLOCK_BUF,
+ XFS_BLFT_DIR_DATA_BUF,
+ XFS_BLFT_DIR_FREE_BUF,
+ XFS_BLFT_DIR_LEAF1_BUF,
+ XFS_BLFT_DIR_LEAFN_BUF,
+ XFS_BLFT_DA_NODE_BUF,
+ XFS_BLFT_ATTR_LEAF_BUF,
+ XFS_BLFT_ATTR_RMT_BUF,
+ XFS_BLFT_SB_BUF,
+ XFS_BLFT_MAX_BUF = (1 << XFS_BLFT_BITS),
+};
+
+static inline void
+xfs_blft_to_flags(struct xfs_buf_log_format *blf, enum xfs_blft type)
+{
+ ASSERT(type > XFS_BLFT_UNKNOWN_BUF && type < XFS_BLFT_MAX_BUF);
+ blf->blf_flags &= ~XFS_BLFT_MASK;
+ blf->blf_flags |= ((type << XFS_BLFT_SHIFT) & XFS_BLFT_MASK);
+}
+
+static inline __uint16_t
+xfs_blft_from_flags(struct xfs_buf_log_format *blf)
+{
+ return (blf->blf_flags & XFS_BLFT_MASK) >> XFS_BLFT_SHIFT;
+}
+
+/*
* buf log item flags
*/
#define XFS_BLI_HOLD 0x01
@@ -153,6 +163,10 @@ void xfs_buf_attach_iodone(struct xfs_buf *,
void xfs_buf_iodone_callbacks(struct xfs_buf *);
void xfs_buf_iodone(struct xfs_buf *, struct xfs_log_item *);
+void xfs_trans_buf_set_type(struct xfs_trans *, struct xfs_buf *,
+ enum xfs_blft);
+void xfs_trans_buf_copy_type(struct xfs_buf *dst_bp, struct xfs_buf *src_bp);
+
#endif /* __KERNEL__ */
#endif /* __XFS_BUF_ITEM_H__ */
diff --git a/include/xfs_trans.h b/include/xfs_trans.h
index 9e145e9..acf1381 100644
--- a/include/xfs_trans.h
+++ b/include/xfs_trans.h
@@ -500,10 +500,6 @@ void xfs_trans_inode_buf(xfs_trans_t *, struct xfs_buf *);
void xfs_trans_stale_inode_buf(xfs_trans_t *, struct xfs_buf *);
void xfs_trans_dquot_buf(xfs_trans_t *, struct xfs_buf *, uint);
void xfs_trans_inode_alloc_buf(xfs_trans_t *, struct xfs_buf *);
-void xfs_trans_buf_set_type(struct xfs_trans *, struct xfs_buf *,
- uint);
-void xfs_trans_buf_copy_type(struct xfs_buf *dst_bp,
- struct xfs_buf *src_bp);
void xfs_trans_ichgtime(struct xfs_trans *, struct xfs_inode *, int);
void xfs_trans_ijoin(struct xfs_trans *, struct xfs_inode *, uint);
void xfs_trans_log_buf(xfs_trans_t *, struct xfs_buf *, uint, uint);
diff --git a/libxfs/trans.c b/libxfs/trans.c
index 619aad1..831e42a 100644
--- a/libxfs/trans.c
+++ b/libxfs/trans.c
@@ -218,7 +218,7 @@ libxfs_trans_inode_alloc_buf(
ASSERT(XFS_BUF_FSPRIVATE(bp, void *) != NULL);
bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *);
bip->bli_flags |= XFS_BLI_INODE_ALLOC_BUF;
- xfs_trans_buf_set_type(tp, bp, XFS_BLF_DINO_BUF);
+ xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DINO_BUF);
}
/*
diff --git a/libxfs/xfs_alloc.c b/libxfs/xfs_alloc.c
index 30fc5f4..1041f8f 100644
--- a/libxfs/xfs_alloc.c
+++ b/libxfs/xfs_alloc.c
@@ -2079,7 +2079,7 @@ xfs_alloc_log_agf(
trace_xfs_agf(tp->t_mountp, XFS_BUF_TO_AGF(bp), fields, _RET_IP_);
- xfs_trans_buf_set_type(tp, bp, XFS_BLF_AGF_BUF);
+ xfs_trans_buf_set_type(tp, bp, XFS_BLFT_AGF_BUF);
xfs_btree_offsets(fields, offsets, XFS_AGF_NUM_BITS, &first, &last);
xfs_trans_log_buf(tp, bp, (uint)first, (uint)last);
@@ -2159,7 +2159,7 @@ xfs_alloc_put_freelist(
xfs_alloc_log_agf(tp, agbp, logflags);
- xfs_trans_buf_set_type(tp, agflbp, XFS_BLF_AGFL_BUF);
+ xfs_trans_buf_set_type(tp, agflbp, XFS_BLFT_AGFL_BUF);
xfs_trans_log_buf(tp, agflbp, startoff,
startoff + sizeof(xfs_agblock_t) - 1);
return 0;
diff --git a/libxfs/xfs_attr_leaf.c b/libxfs/xfs_attr_leaf.c
index 7724781..b28266a 100644
--- a/libxfs/xfs_attr_leaf.c
+++ b/libxfs/xfs_attr_leaf.c
@@ -241,7 +241,7 @@ xfs_attr3_leaf_read(
err = xfs_da_read_buf(tp, dp, bno, mappedbno, bpp,
XFS_ATTR_FORK, &xfs_attr3_leaf_buf_ops);
if (!err && tp)
- xfs_trans_buf_set_type(tp, *bpp, XFS_BLF_ATTR_LEAF_BUF);
+ xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_ATTR_LEAF_BUF);
return err;
}
@@ -872,7 +872,7 @@ xfs_attr3_leaf_to_node(
goto out;
/* copy leaf to new buffer, update identifiers */
- xfs_trans_buf_set_type(args->trans, bp2, XFS_BLF_ATTR_LEAF_BUF);
+ xfs_trans_buf_set_type(args->trans, bp2, XFS_BLFT_ATTR_LEAF_BUF);
bp2->b_ops = bp1->b_ops;
memcpy(bp2->b_addr, bp1->b_addr, XFS_LBSIZE(mp));
if (xfs_sb_version_hascrc(&mp->m_sb)) {
@@ -935,7 +935,7 @@ xfs_attr3_leaf_create(
if (error)
return error;
bp->b_ops = &xfs_attr3_leaf_buf_ops;
- xfs_trans_buf_set_type(args->trans, bp, XFS_BLF_ATTR_LEAF_BUF);
+ xfs_trans_buf_set_type(args->trans, bp, XFS_BLFT_ATTR_LEAF_BUF);
leaf = bp->b_addr;
memset(leaf, 0, XFS_LBSIZE(mp));
diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index 5e736a5..6664265 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -1217,7 +1217,7 @@ xfs_bmap_local_to_extents_init_fn(
{
bp->b_ops = &xfs_bmbt_buf_ops;
memcpy(bp->b_addr, ifp->if_u1.if_data, ifp->if_bytes);
- xfs_trans_buf_set_type(tp, bp, XFS_BLF_BTREE_BUF);
+ xfs_trans_buf_set_type(tp, bp, XFS_BLFT_BTREE_BUF);
}
STATIC void
diff --git a/libxfs/xfs_btree.c b/libxfs/xfs_btree.c
index a7c19e9..a613294 100644
--- a/libxfs/xfs_btree.c
+++ b/libxfs/xfs_btree.c
@@ -1227,7 +1227,7 @@ xfs_btree_log_keys(
XFS_BTREE_TRACE_ARGBII(cur, bp, first, last);
if (bp) {
- xfs_trans_buf_set_type(cur->bc_tp, bp, XFS_BLF_BTREE_BUF);
+ xfs_trans_buf_set_type(cur->bc_tp, bp, XFS_BLFT_BTREE_BUF);
xfs_trans_log_buf(cur->bc_tp, bp,
xfs_btree_key_offset(cur, first),
xfs_btree_key_offset(cur, last + 1) - 1);
@@ -1252,7 +1252,7 @@ xfs_btree_log_recs(
XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY);
XFS_BTREE_TRACE_ARGBII(cur, bp, first, last);
- xfs_trans_buf_set_type(cur->bc_tp, bp, XFS_BLF_BTREE_BUF);
+ xfs_trans_buf_set_type(cur->bc_tp, bp, XFS_BLFT_BTREE_BUF);
xfs_trans_log_buf(cur->bc_tp, bp,
xfs_btree_rec_offset(cur, first),
xfs_btree_rec_offset(cur, last + 1) - 1);
@@ -1277,7 +1277,7 @@ xfs_btree_log_ptrs(
struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
int level = xfs_btree_get_level(block);
- xfs_trans_buf_set_type(cur->bc_tp, bp, XFS_BLF_BTREE_BUF);
+ xfs_trans_buf_set_type(cur->bc_tp, bp, XFS_BLFT_BTREE_BUF);
xfs_trans_log_buf(cur->bc_tp, bp,
xfs_btree_ptr_offset(cur, first, level),
xfs_btree_ptr_offset(cur, last + 1, level) - 1);
@@ -1352,7 +1352,7 @@ xfs_btree_log_block(
(cur->bc_flags & XFS_BTREE_LONG_PTRS) ?
loffsets : soffsets,
nbits, &first, &last);
- xfs_trans_buf_set_type(cur->bc_tp, bp, XFS_BLF_BTREE_BUF);
+ xfs_trans_buf_set_type(cur->bc_tp, bp, XFS_BLFT_BTREE_BUF);
xfs_trans_log_buf(cur->bc_tp, bp, first, last);
} else {
xfs_trans_log_inode(cur->bc_tp, cur->bc_private.b.ip,
diff --git a/libxfs/xfs_da_btree.c b/libxfs/xfs_da_btree.c
index 565010d..0a96f20 100644
--- a/libxfs/xfs_da_btree.c
+++ b/libxfs/xfs_da_btree.c
@@ -289,15 +289,15 @@ xfs_da3_node_read(
switch (be16_to_cpu(info->magic)) {
case XFS_DA3_NODE_MAGIC:
case XFS_DA_NODE_MAGIC:
- type = XFS_BLF_DA_NODE_BUF;
+ type = XFS_BLFT_DA_NODE_BUF;
break;
case XFS_ATTR_LEAF_MAGIC:
case XFS_ATTR3_LEAF_MAGIC:
- type = XFS_BLF_ATTR_LEAF_BUF;
+ type = XFS_BLFT_ATTR_LEAF_BUF;
break;
case XFS_DIR2_LEAFN_MAGIC:
case XFS_DIR3_LEAFN_MAGIC:
- type = XFS_BLF_DIR_LEAFN_BUF;
+ type = XFS_BLFT_DIR_LEAFN_BUF;
break;
default:
type = 0;
@@ -338,7 +338,7 @@ xfs_da3_node_create(
if (error)
return(error);
bp->b_ops = &xfs_da3_node_buf_ops;
- xfs_trans_buf_set_type(tp, bp, XFS_BLF_DA_NODE_BUF);
+ xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DA_NODE_BUF);
node = bp->b_addr;
if (xfs_sb_version_hascrc(&mp->m_sb)) {
@@ -572,7 +572,7 @@ xfs_da3_root_split(
* we are about to copy oldroot to bp, so set up the type
* of bp while we know exactly what it will be.
*/
- xfs_trans_buf_set_type(tp, bp, XFS_BLF_DA_NODE_BUF);
+ xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DA_NODE_BUF);
} else {
struct xfs_dir3_icleaf_hdr leafhdr;
struct xfs_dir2_leaf_entry *ents;
@@ -590,7 +590,7 @@ xfs_da3_root_split(
* we are about to copy oldroot to bp, so set up the type
* of bp while we know exactly what it will be.
*/
- xfs_trans_buf_set_type(tp, bp, XFS_BLF_DIR_LEAFN_BUF);
+ xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DIR_LEAFN_BUF);
}
/*
diff --git a/libxfs/xfs_dir2_block.c b/libxfs/xfs_dir2_block.c
index 574e414..dc69394 100644
--- a/libxfs/xfs_dir2_block.c
+++ b/libxfs/xfs_dir2_block.c
@@ -119,7 +119,7 @@ xfs_dir3_block_read(
err = xfs_da_read_buf(tp, dp, mp->m_dirdatablk, -1, bpp,
XFS_DATA_FORK, &xfs_dir3_block_buf_ops);
if (!err && tp)
- xfs_trans_buf_set_type(tp, *bpp, XFS_BLF_DIR_BLOCK_BUF);
+ xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_BLOCK_BUF);
return err;
}
@@ -133,7 +133,7 @@ xfs_dir3_block_init(
struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr;
bp->b_ops = &xfs_dir3_block_buf_ops;
- xfs_trans_buf_set_type(tp, bp, XFS_BLF_DIR_BLOCK_BUF);
+ xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DIR_BLOCK_BUF);
if (xfs_sb_version_hascrc(&mp->m_sb)) {
memset(hdr3, 0, sizeof(*hdr3));
diff --git a/libxfs/xfs_dir2_data.c b/libxfs/xfs_dir2_data.c
index 9752ae3..155352c 100644
--- a/libxfs/xfs_dir2_data.c
+++ b/libxfs/xfs_dir2_data.c
@@ -288,7 +288,7 @@ xfs_dir3_data_read(
err = xfs_da_read_buf(tp, dp, bno, mapped_bno, bpp,
XFS_DATA_FORK, &xfs_dir3_data_buf_ops);
if (!err && tp)
- xfs_trans_buf_set_type(tp, *bpp, XFS_BLF_DIR_DATA_BUF);
+ xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_DATA_BUF);
return err;
}
@@ -558,7 +558,7 @@ xfs_dir3_data_init(
if (error)
return error;
bp->b_ops = &xfs_dir3_data_buf_ops;
- xfs_trans_buf_set_type(tp, bp, XFS_BLF_DIR_DATA_BUF);
+ xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DIR_DATA_BUF);
/*
* Initialize the header.
diff --git a/libxfs/xfs_dir2_leaf.c b/libxfs/xfs_dir2_leaf.c
index d58c25b..12fb61d 100644
--- a/libxfs/xfs_dir2_leaf.c
+++ b/libxfs/xfs_dir2_leaf.c
@@ -280,7 +280,7 @@ xfs_dir3_leaf_read(
err = xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
XFS_DATA_FORK, &xfs_dir3_leaf1_buf_ops);
if (!err && tp)
- xfs_trans_buf_set_type(tp, *bpp, XFS_BLF_DIR_LEAF1_BUF);
+ xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_LEAF1_BUF);
return err;
}
@@ -297,7 +297,7 @@ xfs_dir3_leafn_read(
err = xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp,
XFS_DATA_FORK, &xfs_dir3_leafn_buf_ops);
if (!err && tp)
- xfs_trans_buf_set_type(tp, *bpp, XFS_BLF_DIR_LEAFN_BUF);
+ xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_LEAFN_BUF);
return err;
}
@@ -342,10 +342,10 @@ xfs_dir3_leaf_init(
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
ltp->bestcount = 0;
bp->b_ops = &xfs_dir3_leaf1_buf_ops;
- xfs_trans_buf_set_type(tp, bp, XFS_BLF_DIR_LEAF1_BUF);
+ xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DIR_LEAF1_BUF);
} else {
bp->b_ops = &xfs_dir3_leafn_buf_ops;
- xfs_trans_buf_set_type(tp, bp, XFS_BLF_DIR_LEAFN_BUF);
+ xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DIR_LEAFN_BUF);
}
}
@@ -467,7 +467,7 @@ xfs_dir2_block_to_leaf(
* Fix up the block header, make it a data block.
*/
dbp->b_ops = &xfs_dir3_data_buf_ops;
- xfs_trans_buf_set_type(tp, dbp, XFS_BLF_DIR_DATA_BUF);
+ xfs_trans_buf_set_type(tp, dbp, XFS_BLFT_DIR_DATA_BUF);
if (hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC))
hdr->magic = cpu_to_be32(XFS_DIR2_DATA_MAGIC);
else
@@ -1788,7 +1788,7 @@ xfs_dir2_node_to_leaf(
xfs_dir3_leaf_compact(args, &leafhdr, lbp);
lbp->b_ops = &xfs_dir3_leaf1_buf_ops;
- xfs_trans_buf_set_type(tp, lbp, XFS_BLF_DIR_LEAF1_BUF);
+ xfs_trans_buf_set_type(tp, lbp, XFS_BLFT_DIR_LEAF1_BUF);
leafhdr.magic = (leafhdr.magic == XFS_DIR2_LEAFN_MAGIC)
? XFS_DIR2_LEAF1_MAGIC
: XFS_DIR3_LEAFN_MAGIC;
diff --git a/libxfs/xfs_dir2_node.c b/libxfs/xfs_dir2_node.c
index 707aae8..eb430f9 100644
--- a/libxfs/xfs_dir2_node.c
+++ b/libxfs/xfs_dir2_node.c
@@ -147,7 +147,7 @@ __xfs_dir3_free_read(
/* try read returns without an error or *bpp if it lands in a hole */
if (!err && tp && *bpp)
- xfs_trans_buf_set_type(tp, *bpp, XFS_BLF_DIR_FREE_BUF);
+ xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_FREE_BUF);
return err;
}
@@ -235,7 +235,7 @@ xfs_dir3_free_get_buf(
if (error)
return error;
- xfs_trans_buf_set_type(tp, bp, XFS_BLF_DIR_FREE_BUF);
+ xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DIR_FREE_BUF);
bp->b_ops = &xfs_dir3_free_buf_ops;
/*
@@ -384,7 +384,7 @@ xfs_dir2_leaf_to_node(
else
leaf->hdr.info.magic = cpu_to_be16(XFS_DIR3_LEAFN_MAGIC);
lbp->b_ops = &xfs_dir3_leafn_buf_ops;
- xfs_trans_buf_set_type(tp, lbp, XFS_BLF_DIR_LEAFN_BUF);
+ xfs_trans_buf_set_type(tp, lbp, XFS_BLFT_DIR_LEAFN_BUF);
xfs_dir3_leaf_log_header(tp, lbp);
xfs_dir3_leaf_check(mp, lbp);
return 0;
@@ -796,7 +796,7 @@ xfs_dir2_leafn_lookup_for_entry(
(char *)curbp->b_addr);
state->extrablk.magic = XFS_DIR2_DATA_MAGIC;
curbp->b_ops = &xfs_dir3_data_buf_ops;
- xfs_trans_buf_set_type(tp, curbp, XFS_BLF_DIR_DATA_BUF);
+ xfs_trans_buf_set_type(tp, curbp, XFS_BLFT_DIR_DATA_BUF);
if (cmp == XFS_CMP_EXACT)
return XFS_ERROR(EEXIST);
}
@@ -811,7 +811,7 @@ xfs_dir2_leafn_lookup_for_entry(
state->extrablk.blkno = curdb;
state->extrablk.magic = XFS_DIR2_DATA_MAGIC;
curbp->b_ops = &xfs_dir3_data_buf_ops;
- xfs_trans_buf_set_type(tp, curbp, XFS_BLF_DIR_DATA_BUF);
+ xfs_trans_buf_set_type(tp, curbp, XFS_BLFT_DIR_DATA_BUF);
} else {
/* If the curbp is not the CI match block, drop it */
if (state->extrablk.bp != curbp)
diff --git a/libxfs/xfs_ialloc.c b/libxfs/xfs_ialloc.c
index 008c4ea..bea5254 100644
--- a/libxfs/xfs_ialloc.c
+++ b/libxfs/xfs_ialloc.c
@@ -1288,7 +1288,7 @@ xfs_ialloc_log_agi(
/*
* Log the allocation group inode header buffer.
*/
- xfs_trans_buf_set_type(tp, bp, XFS_BLF_AGI_BUF);
+ xfs_trans_buf_set_type(tp, bp, XFS_BLFT_AGI_BUF);
xfs_trans_log_buf(tp, bp, first, last);
}
xfsprogs-crc-patchset-v2/xfs__add_crc_checks_to_the_superblock-2 0000600 0001750 0001750 00000037000 12132011547 024033 0 ustar dave dave xfs: add CRC checks to the superblock
From: Dave Chinner <dchinner@xxxxxxxxxx>
With the addition of CRCs, there is such a wide and varied change to
the on disk format that it makes sense to bump the superblock
version number rather than try to use feature bits for all the new
functionality.
This commit introduces all the new superblock fields needed for all
the new functionality: feature masks similar to ext4, separate
project quota inodes, a LSN field for recovery and the CRC field.
This commit does not bump the superblock version number, however.
That will be done as a separate commit at the end of the series
after all the new functionality is present so we switch it all on in
one commit. This means that we can slowly introduce the changes
without them being active and hence maintain bisectability of the
tree.
This patch is based on a patch originally written by myself back
from SGI days, which was subsequently modified by Christoph Hellwig.
There is relatively little of that patch remaining, but the history
of the patch still should be acknowledged here.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
include/xfs_mount.h | 1 +
include/xfs_sb.h | 100 ++++++++++++++++++++++++++++++++++++---------------
libxfs/xfs_mount.c | 92 +++++++++++++++++++++++++++++++++++++++++------
3 files changed, 153 insertions(+), 40 deletions(-)
diff --git a/include/xfs_mount.h b/include/xfs_mount.h
index 28bbf46..68c02a9 100644
--- a/include/xfs_mount.h
+++ b/include/xfs_mount.h
@@ -391,6 +391,7 @@ struct xfs_perag *xfs_perag_get_tag(struct xfs_mount *mp, xfs_agnumber_t agno,
int tag);
void xfs_perag_put(struct xfs_perag *pag);
+extern void xfs_sb_calc_crc(struct xfs_buf *);
extern void xfs_mod_sb(struct xfs_trans *, __int64_t);
extern int xfs_initialize_perag(struct xfs_mount *, xfs_agnumber_t,
xfs_agnumber_t *);
diff --git a/include/xfs_sb.h b/include/xfs_sb.h
index 6a7f8b0..d6709db 100644
--- a/include/xfs_sb.h
+++ b/include/xfs_sb.h
@@ -32,6 +32,7 @@ struct xfs_mount;
#define XFS_SB_VERSION_2 2 /* 6.2 - attributes */
#define XFS_SB_VERSION_3 3 /* 6.2 - new inode version */
#define XFS_SB_VERSION_4 4 /* 6.2+ - bitmask version */
+#define XFS_SB_VERSION_5 5 /* CRC enabled filesystem */
#define XFS_SB_VERSION_NUMBITS 0x000f
#define XFS_SB_VERSION_ALLFBITS 0xfff0
#define XFS_SB_VERSION_SASHFBITS 0xf000
@@ -161,6 +162,18 @@ typedef struct xfs_sb {
*/
__uint32_t sb_bad_features2;
+ /* version 5 superblock fields start here */
+
+ /* feature masks */
+ __uint32_t sb_features_compat;
+ __uint32_t sb_features_ro_compat;
+ __uint32_t sb_features_incompat;
+
+ __uint32_t sb_crc; /* superblock crc */
+
+ xfs_ino_t sb_pquotino; /* project quota inode */
+ xfs_lsn_t sb_lsn; /* last write sequence */
+
/* must be padded to 64 bit alignment */
} xfs_sb_t;
@@ -229,7 +242,19 @@ typedef struct xfs_dsb {
* for features2 bits. Easiest just to mark it bad and not use
* it for anything else.
*/
- __be32 sb_bad_features2;
+ __be32 sb_bad_features2;
+
+ /* version 5 superblock fields start here */
+
+ /* feature masks */
+ __be32 sb_features_compat;
+ __be32 sb_features_ro_compat;
+ __be32 sb_features_incompat;
+
+ __le32 sb_crc; /* superblock crc */
+
+ __be64 sb_pquotino; /* project quota inode */
+ __be64 sb_lsn; /* last write sequence */
/* must be padded to 64 bit alignment */
} xfs_dsb_t;
@@ -250,7 +275,9 @@ typedef enum {
XFS_SBS_GQUOTINO, XFS_SBS_QFLAGS, XFS_SBS_FLAGS, XFS_SBS_SHARED_VN,
XFS_SBS_INOALIGNMT, XFS_SBS_UNIT, XFS_SBS_WIDTH, XFS_SBS_DIRBLKLOG,
XFS_SBS_LOGSECTLOG, XFS_SBS_LOGSECTSIZE, XFS_SBS_LOGSUNIT,
- XFS_SBS_FEATURES2, XFS_SBS_BAD_FEATURES2,
+ XFS_SBS_FEATURES2, XFS_SBS_BAD_FEATURES2, XFS_SBS_FEATURES_COMPAT,
+ XFS_SBS_FEATURES_RO_COMPAT, XFS_SBS_FEATURES_INCOMPAT, XFS_SBS_CRC,
+ XFS_SBS_PQUOTINO, XFS_SBS_LSN,
XFS_SBS_FIELDCOUNT
} xfs_sb_field_t;
@@ -276,6 +303,11 @@ typedef enum {
#define XFS_SB_FDBLOCKS XFS_SB_MVAL(FDBLOCKS)
#define XFS_SB_FEATURES2 XFS_SB_MVAL(FEATURES2)
#define XFS_SB_BAD_FEATURES2 XFS_SB_MVAL(BAD_FEATURES2)
+#define XFS_SB_FEATURES_COMPAT XFS_SB_MVAL(FEATURES_COMPAT)
+#define XFS_SB_FEATURES_RO_COMPAT XFS_SB_MVAL(FEATURES_RO_COMPAT)
+#define XFS_SB_FEATURES_INCOMPAT XFS_SB_MVAL(FEATURES_INCOMPAT)
+#define XFS_SB_CRC XFS_SB_MVAL(CRC)
+#define XFS_SB_PQUOTINO XFS_SB_MVAL(PQUOTINO)
#define XFS_SB_NUM_BITS ((int)XFS_SBS_FIELDCOUNT)
#define XFS_SB_ALL_BITS ((1LL << XFS_SB_NUM_BITS) - 1)
#define XFS_SB_MOD_BITS \
@@ -283,7 +315,8 @@ typedef enum {
XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | XFS_SB_GQUOTINO | \
XFS_SB_QFLAGS | XFS_SB_SHARED_VN | XFS_SB_UNIT | XFS_SB_WIDTH | \
XFS_SB_ICOUNT | XFS_SB_IFREE | XFS_SB_FDBLOCKS | XFS_SB_FEATURES2 | \
- XFS_SB_BAD_FEATURES2)
+ XFS_SB_BAD_FEATURES2 | XFS_SB_FEATURES_COMPAT | \
+ XFS_SB_FEATURES_RO_COMPAT | XFS_SB_FEATURES_INCOMPAT | XFS_SB_PQUOTINO)
/*
@@ -325,6 +358,8 @@ static inline int xfs_sb_good_version(xfs_sb_t *sbp)
return 1;
}
+ if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5)
+ return 1;
return 0;
}
@@ -365,7 +400,7 @@ static inline int xfs_sb_version_hasattr(xfs_sb_t *sbp)
{
return sbp->sb_versionnum == XFS_SB_VERSION_2 ||
sbp->sb_versionnum == XFS_SB_VERSION_3 ||
- (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
+ (XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4 &&
(sbp->sb_versionnum & XFS_SB_VERSION_ATTRBIT));
}
@@ -373,7 +408,7 @@ static inline void xfs_sb_version_addattr(xfs_sb_t *sbp)
{
if (sbp->sb_versionnum == XFS_SB_VERSION_1)
sbp->sb_versionnum = XFS_SB_VERSION_2;
- else if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4)
+ else if (XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4)
sbp->sb_versionnum |= XFS_SB_VERSION_ATTRBIT;
else
sbp->sb_versionnum = XFS_SB_VERSION_4 | XFS_SB_VERSION_ATTRBIT;
@@ -382,7 +417,7 @@ static inline void xfs_sb_version_addattr(xfs_sb_t *sbp)
static inline int xfs_sb_version_hasnlink(xfs_sb_t *sbp)
{
return sbp->sb_versionnum == XFS_SB_VERSION_3 ||
- (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
+ (XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4 &&
(sbp->sb_versionnum & XFS_SB_VERSION_NLINKBIT));
}
@@ -396,13 +431,13 @@ static inline void xfs_sb_version_addnlink(xfs_sb_t *sbp)
static inline int xfs_sb_version_hasquota(xfs_sb_t *sbp)
{
- return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
+ return XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4 &&
(sbp->sb_versionnum & XFS_SB_VERSION_QUOTABIT);
}
static inline void xfs_sb_version_addquota(xfs_sb_t *sbp)
{
- if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4)
+ if (XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4)
sbp->sb_versionnum |= XFS_SB_VERSION_QUOTABIT;
else
sbp->sb_versionnum = xfs_sb_version_tonew(sbp->sb_versionnum) |
@@ -411,13 +446,14 @@ static inline void xfs_sb_version_addquota(xfs_sb_t *sbp)
static inline int xfs_sb_version_hasalign(xfs_sb_t *sbp)
{
- return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
- (sbp->sb_versionnum & XFS_SB_VERSION_ALIGNBIT);
+ return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) ||
+ (XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4 &&
+ (sbp->sb_versionnum & XFS_SB_VERSION_ALIGNBIT));
}
static inline int xfs_sb_version_hasdalign(xfs_sb_t *sbp)
{
- return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
+ return XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4 &&
(sbp->sb_versionnum & XFS_SB_VERSION_DALIGNBIT);
}
@@ -429,38 +465,42 @@ static inline int xfs_sb_version_hasshared(xfs_sb_t *sbp)
static inline int xfs_sb_version_hasdirv2(xfs_sb_t *sbp)
{
- return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
- (sbp->sb_versionnum & XFS_SB_VERSION_DIRV2BIT);
+ return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) ||
+ (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
+ (sbp->sb_versionnum & XFS_SB_VERSION_DIRV2BIT));
}
static inline int xfs_sb_version_haslogv2(xfs_sb_t *sbp)
{
- return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
- (sbp->sb_versionnum & XFS_SB_VERSION_LOGV2BIT);
+ return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) ||
+ (XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4 &&
+ (sbp->sb_versionnum & XFS_SB_VERSION_LOGV2BIT));
}
static inline int xfs_sb_version_hasextflgbit(xfs_sb_t *sbp)
{
- return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
- (sbp->sb_versionnum & XFS_SB_VERSION_EXTFLGBIT);
+ return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) ||
+ (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
+ (sbp->sb_versionnum & XFS_SB_VERSION_EXTFLGBIT));
}
static inline int xfs_sb_version_hassector(xfs_sb_t *sbp)
{
- return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
+ return XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4 &&
(sbp->sb_versionnum & XFS_SB_VERSION_SECTORBIT);
}
static inline int xfs_sb_version_hasasciici(xfs_sb_t *sbp)
{
- return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
+ return XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4 &&
(sbp->sb_versionnum & XFS_SB_VERSION_BORGBIT);
}
static inline int xfs_sb_version_hasmorebits(xfs_sb_t *sbp)
{
- return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
- (sbp->sb_versionnum & XFS_SB_VERSION_MOREBITSBIT);
+ return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) ||
+ (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
+ (sbp->sb_versionnum & XFS_SB_VERSION_MOREBITSBIT));
}
/*
@@ -475,14 +515,16 @@ static inline int xfs_sb_version_hasmorebits(xfs_sb_t *sbp)
static inline int xfs_sb_version_haslazysbcount(xfs_sb_t *sbp)
{
- return xfs_sb_version_hasmorebits(sbp) &&
- (sbp->sb_features2 & XFS_SB_VERSION2_LAZYSBCOUNTBIT);
+ return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) ||
+ (xfs_sb_version_hasmorebits(sbp) &&
+ (sbp->sb_features2 & XFS_SB_VERSION2_LAZYSBCOUNTBIT));
}
static inline int xfs_sb_version_hasattr2(xfs_sb_t *sbp)
{
- return xfs_sb_version_hasmorebits(sbp) &&
- (sbp->sb_features2 & XFS_SB_VERSION2_ATTR2BIT);
+ return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) ||
+ (xfs_sb_version_hasmorebits(sbp) &&
+ (sbp->sb_features2 & XFS_SB_VERSION2_ATTR2BIT));
}
static inline void xfs_sb_version_addattr2(xfs_sb_t *sbp)
@@ -500,8 +542,9 @@ static inline void xfs_sb_version_removeattr2(xfs_sb_t *sbp)
static inline int xfs_sb_version_hasprojid32bit(xfs_sb_t *sbp)
{
- return xfs_sb_version_hasmorebits(sbp) &&
- (sbp->sb_features2 & XFS_SB_VERSION2_PROJID32BIT);
+ return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) ||
+ (xfs_sb_version_hasmorebits(sbp) &&
+ (sbp->sb_features2 & XFS_SB_VERSION2_PROJID32BIT));
}
static inline void xfs_sb_version_addprojid32bit(xfs_sb_t *sbp)
@@ -513,8 +556,7 @@ static inline void xfs_sb_version_addprojid32bit(xfs_sb_t *sbp)
static inline int xfs_sb_version_hascrc(xfs_sb_t *sbp)
{
- return (xfs_sb_version_hasmorebits(sbp) &&
- (sbp->sb_features2 & XFS_SB_VERSION2_CRCBIT));
+ return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5;
}
/*
diff --git a/libxfs/xfs_mount.c b/libxfs/xfs_mount.c
index 7ab3519..07b892b 100644
--- a/libxfs/xfs_mount.c
+++ b/libxfs/xfs_mount.c
@@ -70,6 +70,12 @@ static const struct {
{ offsetof(xfs_sb_t, sb_logsunit), 0 },
{ offsetof(xfs_sb_t, sb_features2), 0 },
{ offsetof(xfs_sb_t, sb_bad_features2), 0 },
+ { offsetof(xfs_sb_t, sb_features_compat), 0 },
+ { offsetof(xfs_sb_t, sb_features_ro_compat), 0 },
+ { offsetof(xfs_sb_t, sb_features_incompat), 0 },
+ { offsetof(xfs_sb_t, sb_crc), 0 },
+ { offsetof(xfs_sb_t, sb_pquotino), 0 },
+ { offsetof(xfs_sb_t, sb_lsn), 0 },
{ sizeof(xfs_sb_t), 0 }
};
@@ -127,11 +133,23 @@ xfs_mount_validate_sb(
return XFS_ERROR(EWRONGFS);
}
+
if (!xfs_sb_good_version(sbp)) {
xfs_warn(mp, "bad version");
return XFS_ERROR(EWRONGFS);
}
+ /*
+ * Do not allow Version 5 superblocks to mount right now, even though
+ * support is in place. We need to implement the proper feature masks
+ * first.
+ */
+ if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) {
+ xfs_alert(mp,
+ "Version 5 superblock detected. Experimental support not yet enabled!");
+ return XFS_ERROR(EINVAL);
+ }
+
if (unlikely(
sbp->sb_logstart == 0 && mp->m_logdev == mp->m_dev)) {
xfs_warn(mp,
@@ -264,6 +282,11 @@ xfs_sb_from_disk(
to->sb_logsunit = be32_to_cpu(from->sb_logsunit);
to->sb_features2 = be32_to_cpu(from->sb_features2);
to->sb_bad_features2 = be32_to_cpu(from->sb_bad_features2);
+ to->sb_features_compat = be32_to_cpu(from->sb_features_compat);
+ to->sb_features_ro_compat = be32_to_cpu(from->sb_features_ro_compat);
+ to->sb_features_incompat = be32_to_cpu(from->sb_features_incompat);
+ to->sb_pquotino = be64_to_cpu(from->sb_pquotino);
+ to->sb_lsn = be64_to_cpu(from->sb_lsn);
}
/*
@@ -319,13 +342,12 @@ xfs_sb_to_disk(
}
}
-static void
+static int
xfs_sb_verify(
struct xfs_buf *bp)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
struct xfs_sb sb;
- int error;
xfs_sb_from_disk(&sb, XFS_BUF_TO_SBP(bp));
@@ -333,16 +355,46 @@ xfs_sb_verify(
* Only check the in progress field for the primary superblock as
* mkfs.xfs doesn't clear it from secondary superblocks.
*/
- error = xfs_mount_validate_sb(mp, &sb, bp->b_bn == XFS_SB_DADDR);
- if (error)
- xfs_buf_ioerror(bp, error);
+ return xfs_mount_validate_sb(mp, &sb, bp->b_bn == XFS_SB_DADDR);
}
+/*
+ * If the superblock has the CRC feature bit set or the CRC field is non-null,
+ * check that the CRC is valid. We check the CRC field is non-null because a
+ * single bit error could clear the feature bit and unused parts of the
+ * superblock are supposed to be zero. Hence a non-null crc field indicates that
+ * we've potentially lost a feature bit and we should check it anyway.
+ */
static void
xfs_sb_read_verify(
struct xfs_buf *bp)
{
- xfs_sb_verify(bp);
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_dsb *dsb = XFS_BUF_TO_SBP(bp);
+ int error;
+
+ /*
+ * open code the version check to avoid needing to convert the entire
+ * superblock from disk order just to check the version number
+ */
+ if (dsb->sb_magicnum == cpu_to_be32(XFS_SB_MAGIC) &&
+ (((be16_to_cpu(dsb->sb_versionnum) & XFS_SB_VERSION_NUMBITS) ==
+ XFS_SB_VERSION_5) ||
+ dsb->sb_crc != 0)) {
+
+ if (!xfs_verify_cksum(bp->b_addr, be16_to_cpu(dsb->sb_sectsize),
+ offsetof(struct xfs_sb, sb_crc))) {
+ error = EFSCORRUPTED;
+ goto out_error;
+ }
+ }
+ error = xfs_sb_verify(bp);
+
+out_error:
+ if (error) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, error);
+ }
}
/*
@@ -355,11 +407,10 @@ static void
xfs_sb_quiet_read_verify(
struct xfs_buf *bp)
{
- struct xfs_sb sb;
+ struct xfs_dsb *dsb = XFS_BUF_TO_SBP(bp);
- xfs_sb_from_disk(&sb, XFS_BUF_TO_SBP(bp));
- if (sb.sb_magicnum == XFS_SB_MAGIC) {
+ if (dsb->sb_magicnum == cpu_to_be32(XFS_SB_MAGIC)) {
/* XFS filesystem, verify noisily! */
xfs_sb_read_verify(bp);
return;
@@ -370,9 +421,27 @@ xfs_sb_quiet_read_verify(
static void
xfs_sb_write_verify(
- struct xfs_buf *bp)
+ struct xfs_buf *bp)
{
- xfs_sb_verify(bp);
+ struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_buf_log_item *bip = bp->b_fspriv;
+ int error;
+
+ error = xfs_sb_verify(bp);
+ if (error) {
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
+ xfs_buf_ioerror(bp, error);
+ return;
+ }
+
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return;
+
+ if (bip)
+ XFS_BUF_TO_SBP(bp)->sb_lsn = cpu_to_be64(bip->bli_item.li_lsn);
+
+ xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length),
+ offsetof(struct xfs_sb, sb_crc));
}
const struct xfs_buf_ops xfs_sb_buf_ops = {
@@ -525,5 +594,6 @@ xfs_mod_sb(xfs_trans_t *tp, __int64_t fields)
ASSERT((1LL << f) & XFS_SB_MOD_BITS);
first = xfs_sb_info[f].offset;
+ xfs_trans_buf_set_type(tp, bp, XFS_BLFT_SB_BUF);
xfs_trans_log_buf(tp, bp, first, last);
}
xfsprogs-crc-patchset-v2/xfs-sb-v5-feature-masks 0000600 0001750 0001750 00000020573 12132011547 020471 0 ustar dave dave xfs: implement extended feature masks
From: Dave Chinner <dchinner@xxxxxxxxxx>
The version 5 superblock has extended feature masks for compatible,
incompatible and read-only compatible feature sets. Implement the
masking and mount-time checking for these feature masks.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
include/xfs_sb.h | 70 ++++++++++++++++++++++++++++++++++++++++++--
libxfs/xfs_mount.c | 41 ++++++++++++++++++++++----
logprint/log_print_trans.c | 18 ++++++++++++
3 files changed, 122 insertions(+), 7 deletions(-)
diff --git a/include/xfs_sb.h b/include/xfs_sb.h
index d6709db..51db6f2 100644
--- a/include/xfs_sb.h
+++ b/include/xfs_sb.h
@@ -168,8 +168,10 @@ typedef struct xfs_sb {
__uint32_t sb_features_compat;
__uint32_t sb_features_ro_compat;
__uint32_t sb_features_incompat;
+ __uint32_t sb_features_log_incompat;
__uint32_t sb_crc; /* superblock crc */
+ __uint32_t sb_pad;
xfs_ino_t sb_pquotino; /* project quota inode */
xfs_lsn_t sb_lsn; /* last write sequence */
@@ -250,8 +252,10 @@ typedef struct xfs_dsb {
__be32 sb_features_compat;
__be32 sb_features_ro_compat;
__be32 sb_features_incompat;
+ __be32 sb_features_log_incompat;
__le32 sb_crc; /* superblock crc */
+ __be32 sb_pad;
__be64 sb_pquotino; /* project quota inode */
__be64 sb_lsn; /* last write sequence */
@@ -276,7 +280,8 @@ typedef enum {
XFS_SBS_INOALIGNMT, XFS_SBS_UNIT, XFS_SBS_WIDTH, XFS_SBS_DIRBLKLOG,
XFS_SBS_LOGSECTLOG, XFS_SBS_LOGSECTSIZE, XFS_SBS_LOGSUNIT,
XFS_SBS_FEATURES2, XFS_SBS_BAD_FEATURES2, XFS_SBS_FEATURES_COMPAT,
- XFS_SBS_FEATURES_RO_COMPAT, XFS_SBS_FEATURES_INCOMPAT, XFS_SBS_CRC,
+ XFS_SBS_FEATURES_RO_COMPAT, XFS_SBS_FEATURES_INCOMPAT,
+ XFS_SBS_FEATURES_LOG_INCOMPAT, XFS_SBS_CRC, XFS_SBS_PAD,
XFS_SBS_PQUOTINO, XFS_SBS_LSN,
XFS_SBS_FIELDCOUNT
} xfs_sb_field_t;
@@ -306,6 +311,7 @@ typedef enum {
#define XFS_SB_FEATURES_COMPAT XFS_SB_MVAL(FEATURES_COMPAT)
#define XFS_SB_FEATURES_RO_COMPAT XFS_SB_MVAL(FEATURES_RO_COMPAT)
#define XFS_SB_FEATURES_INCOMPAT XFS_SB_MVAL(FEATURES_INCOMPAT)
+#define XFS_SB_FEATURES_LOG_INCOMPAT XFS_SB_MVAL(FEATURES_LOG_INCOMPAT)
#define XFS_SB_CRC XFS_SB_MVAL(CRC)
#define XFS_SB_PQUOTINO XFS_SB_MVAL(PQUOTINO)
#define XFS_SB_NUM_BITS ((int)XFS_SBS_FIELDCOUNT)
@@ -316,7 +322,8 @@ typedef enum {
XFS_SB_QFLAGS | XFS_SB_SHARED_VN | XFS_SB_UNIT | XFS_SB_WIDTH | \
XFS_SB_ICOUNT | XFS_SB_IFREE | XFS_SB_FDBLOCKS | XFS_SB_FEATURES2 | \
XFS_SB_BAD_FEATURES2 | XFS_SB_FEATURES_COMPAT | \
- XFS_SB_FEATURES_RO_COMPAT | XFS_SB_FEATURES_INCOMPAT | XFS_SB_PQUOTINO)
+ XFS_SB_FEATURES_RO_COMPAT | XFS_SB_FEATURES_INCOMPAT | \
+ XFS_SB_FEATURES_LOG_INCOMPAT | XFS_SB_PQUOTINO)
/*
@@ -559,6 +566,65 @@ static inline int xfs_sb_version_hascrc(xfs_sb_t *sbp)
return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5;
}
+
+/*
+ * Extended v5 superblock feature masks. These are to be used for new v5
+ * superblock features only.
+ *
+ * Compat features are new features that old kernels will not notice or affect
+ * and so can mount read-write without issues.
+ *
+ * RO-Compat (read only) are features that old kernels can read but will break
+ * if they write. Hence only read-only mounts of such filesystems are allowed on
+ * kernels that don't support the feature bit.
+ *
+ * InCompat features are features which old kernels will not understand and so
+ * must not mount.
+ *
+ * Log-InCompat features are for changes to log formats or new transactions that
+ * can't be replayed on older kernels. The fields are set when the filesystem is
+ * mounted, and a clean unmount clears the fields.
+ */
+#define XFS_SB_FEAT_COMPAT_ALL 0
+#define XFS_SB_FEAT_COMPAT_UNKNOWN ~XFS_SB_FEAT_COMPAT_ALL
+static inline bool
+xfs_sb_has_compat_feature(
+ struct xfs_sb *sbp,
+ __uint32_t feature)
+{
+ return (sbp->sb_features_compat & feature) != 0;
+}
+
+#define XFS_SB_FEAT_RO_COMPAT_ALL 0
+#define XFS_SB_FEAT_RO_COMPAT_UNKNOWN ~XFS_SB_FEAT_RO_COMPAT_ALL
+static inline bool
+xfs_sb_has_ro_compat_feature(
+ struct xfs_sb *sbp,
+ __uint32_t feature)
+{
+ return (sbp->sb_features_ro_compat & feature) != 0;
+}
+
+#define XFS_SB_FEAT_INCOMPAT_ALL 0
+#define XFS_SB_FEAT_INCOMPAT_UNKNOWN ~XFS_SB_FEAT_INCOMPAT_ALL
+static inline bool
+xfs_sb_has_incompat_feature(
+ struct xfs_sb *sbp,
+ __uint32_t feature)
+{
+ return (sbp->sb_features_incompat & feature) != 0;
+}
+
+#define XFS_SB_FEAT_INCOMPAT_LOG_ALL 0
+#define XFS_SB_FEAT_INCOMPAT_LOG_UNKNOWN ~XFS_SB_FEAT_INCOMPAT_LOG_ALL
+static inline bool
+xfs_sb_has_incompat_log_feature(
+ struct xfs_sb *sbp,
+ __uint32_t feature)
+{
+ return (sbp->sb_features_log_incompat & feature) != 0;
+}
+
/*
* end of superblock version macros
*/
diff --git a/libxfs/xfs_mount.c b/libxfs/xfs_mount.c
index 07b892b..f7daa2f 100644
--- a/libxfs/xfs_mount.c
+++ b/libxfs/xfs_mount.c
@@ -73,7 +73,9 @@ static const struct {
{ offsetof(xfs_sb_t, sb_features_compat), 0 },
{ offsetof(xfs_sb_t, sb_features_ro_compat), 0 },
{ offsetof(xfs_sb_t, sb_features_incompat), 0 },
+ { offsetof(xfs_sb_t, sb_features_log_incompat), 0 },
{ offsetof(xfs_sb_t, sb_crc), 0 },
+ { offsetof(xfs_sb_t, sb_pad), 0 },
{ offsetof(xfs_sb_t, sb_pquotino), 0 },
{ offsetof(xfs_sb_t, sb_lsn), 0 },
{ sizeof(xfs_sb_t), 0 }
@@ -140,14 +142,40 @@ xfs_mount_validate_sb(
}
/*
- * Do not allow Version 5 superblocks to mount right now, even though
- * support is in place. We need to implement the proper feature masks
- * first.
+ * Version 5 superblock feature mask validation. Reject combinations the
+ * kernel cannot support up front before checking anything else.
*/
if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) {
xfs_alert(mp,
- "Version 5 superblock detected. Experimental support not yet enabled!");
- return XFS_ERROR(EINVAL);
+"Version 5 superblock detected. xfsprogs has EXPERIMENTAL support enabled!\n"
+"Use of these features is at your own risk!");
+
+ if (xfs_sb_has_compat_feature(sbp,
+ XFS_SB_FEAT_COMPAT_UNKNOWN)) {
+ xfs_warn(mp,
+"Superblock has unknown compatible features (0x%x) enabled.\n"
+"Using a more recent xfsprogs is recommended.",
+ (sbp->sb_features_compat &
+ XFS_SB_FEAT_COMPAT_UNKNOWN));
+ }
+
+ if (xfs_sb_has_ro_compat_feature(sbp,
+ XFS_SB_FEAT_RO_COMPAT_UNKNOWN)) {
+ xfs_warn(mp,
+"Superblock has unknown read-only compatible features (0x%x) enabled.\n"
+"Using a more recent xfsprogs is recommended.",
+ (sbp->sb_features_ro_compat &
+ XFS_SB_FEAT_RO_COMPAT_UNKNOWN));
+ }
+ if (xfs_sb_has_incompat_feature(sbp,
+ XFS_SB_FEAT_INCOMPAT_UNKNOWN)) {
+ xfs_warn(mp,
+"Superblock has unknown incompatible features (0x%x) enabled.\n"
+"Filesystem can not be safely operated on by this xfsprogs installation",
+ (sbp->sb_features_incompat &
+ XFS_SB_FEAT_INCOMPAT_UNKNOWN));
+ return XFS_ERROR(EINVAL);
+ }
}
if (unlikely(
@@ -285,6 +313,9 @@ xfs_sb_from_disk(
to->sb_features_compat = be32_to_cpu(from->sb_features_compat);
to->sb_features_ro_compat = be32_to_cpu(from->sb_features_ro_compat);
to->sb_features_incompat = be32_to_cpu(from->sb_features_incompat);
+ to->sb_features_log_incompat =
+ be32_to_cpu(from->sb_features_log_incompat);
+ to->sb_pad = 0;
to->sb_pquotino = be64_to_cpu(from->sb_pquotino);
to->sb_lsn = be64_to_cpu(from->sb_lsn);
}
diff --git a/logprint/log_print_trans.c b/logprint/log_print_trans.c
index 86e1c42..2dd3a10 100644
--- a/logprint/log_print_trans.c
+++ b/logprint/log_print_trans.c
@@ -68,6 +68,24 @@ xfs_log_print_trans(
if (head_blk == tail_blk)
return;
+
+ /*
+ * Version 5 superblock log feature mask validation. We know the
+ * log is dirty so check if there are any unknown log features
+ * in what we need to recover. If there are unknown features
+ * (e.g. unsupported transactions) then warn about it.
+ */
+ if (XFS_SB_VERSION_NUM(&log->l_mp->m_sb) == XFS_SB_VERSION_5 &&
+ xfs_sb_has_incompat_log_feature(&log->l_mp->m_sb,
+ XFS_SB_FEAT_INCOMPAT_LOG_UNKNOWN)) {
+ printf(_(
+"Superblock has unknown incompatible log features (0x%x) enabled.\n"
+"Output may be incomplete or inaccurate. It is recommended that you\n"
+"upgrade your xfsprogs installation to match the filesystem features.\n"),
+ (log->l_mp->m_sb.sb_features_log_incompat &
+ XFS_SB_FEAT_INCOMPAT_LOG_UNKNOWN));
+ }
+
if ((error = xlog_do_recovery_pass(log, head_blk, tail_blk, XLOG_RECOVER_PASS1))) {
fprintf(stderr, _("%s: failed in xfs_do_recovery_pass, error: %d\n"),
progname, error);
xfsprogs-crc-patchset-v2/xfsprogs-mkfs-crc-support 0000600 0001750 0001750 00000225673 12132011547 021272 0 ustar dave dave xfsprogs: introduce CRC support into mkfs.xfs
From: Dave Chinner <dchinner@xxxxxxxxxx>
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
copy/xfs_copy.c | 5 +-
db/sb.c | 6 +-
include/libxfs.h | 85 ++++++++++++++----------
include/libxlog.h | 2 +-
libxfs/init.c | 20 +++---
libxfs/logitem.c | 4 +-
libxfs/rdwr.c | 143 ++++++++++++++++++++++++++-------------
libxfs/trans.c | 8 +--
libxfs/xfs.h | 15 +++--
libxfs/xfs_alloc.c | 10 +--
libxfs/xfs_alloc_btree.c | 6 +-
libxfs/xfs_attr_leaf.c | 6 +-
libxfs/xfs_attr_remote.c | 6 +-
libxfs/xfs_bmap_btree.c | 6 +-
libxfs/xfs_btree.c | 8 +--
libxfs/xfs_da_btree.c | 6 +-
libxfs/xfs_dir2_block.c | 6 +-
libxfs/xfs_dir2_data.c | 10 +--
libxfs/xfs_dir2_leaf.c | 6 +-
libxfs/xfs_dir2_node.c | 6 +-
libxfs/xfs_ialloc.c | 6 +-
libxfs/xfs_ialloc_btree.c | 6 +-
libxfs/xfs_inode.c | 4 +-
libxfs/xfs_mount.c | 18 +++--
libxfs/xfs_symlink.c | 8 +--
logprint/logprint.c | 3 +-
mkfs/maxtrres.c | 4 +-
mkfs/proto.c | 4 +-
mkfs/xfs_mkfs.c | 162 +++++++++++++++++++++++++++++++++++----------
mkfs/xfs_mkfs.h | 12 ++--
repair/attr_repair.c | 6 +-
repair/dino_chunks.c | 8 ++-
repair/dinode.c | 10 +--
repair/dir.c | 8 +--
repair/dir2.c | 2 +-
repair/phase2.c | 5 +-
repair/phase3.c | 2 +-
repair/phase6.c | 8 +--
repair/prefetch.c | 4 +-
repair/rt.c | 4 +-
repair/scan.c | 12 ++--
repair/xfs_repair.c | 7 +-
42 files changed, 423 insertions(+), 244 deletions(-)
diff --git a/copy/xfs_copy.c b/copy/xfs_copy.c
index c01e0b9..7dc7d9a 100644
--- a/copy/xfs_copy.c
+++ b/copy/xfs_copy.c
@@ -674,8 +674,11 @@ main(int argc, char **argv)
/* prepare the mount structure */
- sbp = libxfs_readbuf(xargs.ddev, XFS_SB_DADDR, 1, 0);
memset(&mbuf, 0, sizeof(xfs_mount_t));
+ mbuf.m_ddev_targ.dev = xargs.ddev;
+ mbuf.m_ddev_targ.bt_mount = &mbuf;
+ sbp = libxfs_readbuf(mbuf.m_ddev_targ, XFS_SB_DADDR, 1, 0,
+ &xfs_sb_buf_ops);
sb = &mbuf.m_sb;
libxfs_sb_from_disk(sb, XFS_BUF_TO_SBP(sbp));
diff --git a/db/sb.c b/db/sb.c
index d83db9c..f99210c 100644
--- a/db/sb.c
+++ b/db/sb.c
@@ -235,7 +235,8 @@ sb_logcheck(void)
x.logdev = x.ddev;
x.logBBsize = XFS_FSB_TO_BB(mp, mp->m_sb.sb_logblocks);
x.logBBstart = XFS_FSB_TO_DADDR(mp, mp->m_sb.sb_logstart);
- log.l_dev = (mp->m_sb.sb_logstart == 0) ? x.logdev : x.ddev;
+ log.l_dev.dev = (mp->m_sb.sb_logstart == 0) ? x.logdev : x.ddev;
+ log.l_dev.bt_mount = mp;
log.l_logsize = BBTOB(log.l_logBBsize);
log.l_logBBsize = x.logBBsize;
log.l_logBBstart = x.logBBstart;
@@ -266,8 +267,7 @@ sb_logzero(uuid_t *uuidp)
dbprintf(_("Clearing log and setting UUID\n"));
- if (libxfs_log_clear(
- (mp->m_sb.sb_logstart == 0) ? x.logdev : x.ddev,
+ if (libxfs_log_clear(mp->m_logdev_targ,
XFS_FSB_TO_DADDR(mp, mp->m_sb.sb_logstart),
(xfs_extlen_t)XFS_FSB_TO_BB(mp, mp->m_sb.sb_logblocks),
uuidp,
diff --git a/include/libxfs.h b/include/libxfs.h
index 972d850..8ee4eaa 100644
--- a/include/libxfs.h
+++ b/include/libxfs.h
@@ -116,6 +116,16 @@ typedef struct {
#define LIBXFS_EXCLUSIVELY 0x0010 /* disallow other accesses (O_EXCL) */
#define LIBXFS_DIRECT 0x0020 /* can use direct I/O, not buffered */
+/*
+ * IO verifier callbacks need the xfs_mount pointer, so we have to behave
+ * somewhat like the kernel now for userspace IO in terms of having buftarg
+ * based devices...
+ */
+struct xfs_buftarg {
+ struct xfs_mount *bt_mount;
+ dev_t dev;
+};
+
extern char *progname;
extern int libxfs_init (libxfs_init_t *);
extern void libxfs_destroy (void);
@@ -130,11 +140,12 @@ extern void platform_findsizes(char *path, int fd, long long *sz, int *bsz);
/* check or write log footer: specify device, log size in blocks & uuid */
typedef xfs_caddr_t (libxfs_get_block_t)(xfs_caddr_t, int, void *);
-extern int libxfs_log_clear (dev_t, xfs_daddr_t, uint, uuid_t *,
+extern int libxfs_log_clear (struct xfs_buftarg, xfs_daddr_t, uint, uuid_t *,
int, int, int);
extern int libxfs_log_header (xfs_caddr_t, uuid_t *, int, int, int,
libxfs_get_block_t *, void *);
+
/*
* Define a user-level mount structure with all we need
* in order to make use of the numerous XFS_* macros.
@@ -151,9 +162,12 @@ typedef struct xfs_mount {
struct xfs_inode *m_rbmip; /* pointer to bitmap inode */
struct xfs_inode *m_rsumip; /* pointer to summary inode */
struct xfs_inode *m_rootip; /* pointer to root directory */
- dev_t m_dev;
- dev_t m_logdev;
- dev_t m_rtdev;
+ struct xfs_buftarg m_ddev_targ;
+ struct xfs_buftarg m_logdev_targ;
+ struct xfs_buftarg m_rtdev_targ;
+#define m_dev m_ddev_targ
+#define m_logdev m_logdev_targ
+#define m_rtdev m_rtdev_targ
__uint8_t m_dircook_elog; /* log d-cookie entry bits */
__uint8_t m_blkbit_log; /* blocklog + NBBY */
__uint8_t m_blkbb_log; /* blocklog - BBSHIFT */
@@ -218,11 +232,6 @@ extern void libxfs_rtmount_destroy (xfs_mount_t *);
/*
* Simple I/O interface
*/
-typedef struct xfs_buftarg {
- struct xfs_mount *bt_mount;
- dev_t dev;
-} xfs_buftarg_t;
-
#define XB_PAGES 2
struct xfs_buf_map {
@@ -244,7 +253,8 @@ typedef struct xfs_buf {
xfs_daddr_t b_bn;
unsigned b_bcount;
unsigned int b_length;
- dev_t b_dev;
+ struct xfs_buftarg b_target;
+#define b_dev b_target.dev
pthread_mutex_t b_lock;
pthread_t b_holder;
unsigned int b_recur;
@@ -254,7 +264,6 @@ typedef struct xfs_buf {
void *b_addr;
int b_error;
const struct xfs_buf_ops *b_ops;
- struct xfs_buftarg *b_target;
struct xfs_perag *b_pag;
struct xfs_buf_map *b_map;
int b_nmaps;
@@ -315,12 +324,12 @@ extern struct cache_operations libxfs_bcache_operations;
#ifdef XFS_BUF_TRACING
-#define libxfs_readbuf(dev, daddr, len, flags) \
+#define libxfs_readbuf(dev, daddr, len, flags, ops) \
libxfs_trace_readbuf(__FUNCTION__, __FILE__, __LINE__, \
- (dev), (daddr), (len), (flags))
-#define libxfs_readbuf_map(dev, map, nmaps, flags) \
+ (dev), (daddr), (len), (flags), (ops))
+#define libxfs_readbuf_map(dev, map, nmaps, flags, ops) \
libxfs_trace_readbuf_map(__FUNCTION__, __FILE__, __LINE__, \
- (dev), (map), (nmaps), (flags))
+ (dev), (map), (nmaps), (flags), (ops))
#define libxfs_writebuf(buf, flags) \
libxfs_trace_writebuf(__FUNCTION__, __FILE__, __LINE__, \
(buf), (flags))
@@ -337,28 +346,34 @@ extern struct cache_operations libxfs_bcache_operations;
libxfs_trace_putbuf(__FUNCTION__, __FILE__, __LINE__, (buf))
extern xfs_buf_t *libxfs_trace_readbuf(const char *, const char *, int,
- dev_t, xfs_daddr_t, int, int);
+ struct xfs_buftarg, xfs_daddr_t, int, int,
+ const struct xfs_buf_ops *);
extern xfs_buf_t *libxfs_trace_readbuf_map(const char *, const char *, int,
- dev_t, struct xfs_buf_map *, int, int);
+ struct xfs_buftarg, struct xfs_buf_map *, int, int,
+ const struct xfs_buf_ops *);
extern int libxfs_trace_writebuf(const char *, const char *, int,
xfs_buf_t *, int);
extern xfs_buf_t *libxfs_trace_getbuf(const char *, const char *, int,
- dev_t, xfs_daddr_t, int);
+ struct xfs_buftarg, xfs_daddr_t, int);
extern xfs_buf_t *libxfs_trace_getbuf_map(const char *, const char *, int,
- dev_t, struct xfs_buf_map *, int);
+ struct xfs_buftarg, struct xfs_buf_map *, int);
extern xfs_buf_t *libxfs_trace_getbuf_flags(const char *, const char *, int,
- dev_t, xfs_daddr_t, int, unsigned int);
+ struct xfs_buftarg, xfs_daddr_t, int, unsigned int);
extern void libxfs_trace_putbuf (const char *, const char *, int,
xfs_buf_t *);
#else
-extern xfs_buf_t *libxfs_readbuf(dev_t, xfs_daddr_t, int, int);
-extern xfs_buf_t *libxfs_readbuf_map(dev_t, struct xfs_buf_map *, int, int);
+extern xfs_buf_t *libxfs_readbuf(struct xfs_buftarg, xfs_daddr_t, int, int,
+ const struct xfs_buf_ops *);
+extern xfs_buf_t *libxfs_readbuf_map(struct xfs_buftarg, struct xfs_buf_map *,
+ int, int, const struct xfs_buf_ops *);
extern int libxfs_writebuf(xfs_buf_t *, int);
-extern xfs_buf_t *libxfs_getbuf(dev_t, xfs_daddr_t, int);
-extern xfs_buf_t *libxfs_getbuf_map(dev_t, struct xfs_buf_map *, int);
-extern xfs_buf_t *libxfs_getbuf_flags(dev_t, xfs_daddr_t, int, unsigned int);
+extern xfs_buf_t *libxfs_getbuf(struct xfs_buftarg, xfs_daddr_t, int);
+extern xfs_buf_t *libxfs_getbuf_map(struct xfs_buftarg,
+ struct xfs_buf_map *, int);
+extern xfs_buf_t *libxfs_getbuf_flags(struct xfs_buftarg, xfs_daddr_t,
+ int, unsigned int);
extern void libxfs_putbuf (xfs_buf_t *);
#endif
@@ -371,11 +386,11 @@ extern int libxfs_bcache_overflowed(void);
extern int libxfs_bcache_usage(void);
/* Buffer (Raw) Interfaces */
-extern xfs_buf_t *libxfs_getbufr(dev_t, xfs_daddr_t, int);
+extern xfs_buf_t *libxfs_getbufr(struct xfs_buftarg, xfs_daddr_t, int);
extern void libxfs_putbufr(xfs_buf_t *);
extern int libxfs_writebuf_int(xfs_buf_t *, int);
-extern int libxfs_readbufr(dev_t, xfs_daddr_t, xfs_buf_t *, int, int);
+extern int libxfs_readbufr(struct xfs_buftarg, xfs_daddr_t, xfs_buf_t *, int, int);
extern int libxfs_bhash_size;
extern int libxfs_ihash_size;
@@ -461,14 +476,15 @@ extern int libxfs_trans_read_buf (xfs_mount_t *, xfs_trans_t *, dev_t,
xfs_daddr_t, int, uint, struct xfs_buf **);
*/
-struct xfs_buf *libxfs_trans_get_buf_map(struct xfs_trans *tp, dev_t dev,
- struct xfs_buf_map *map, int nmaps,
- uint flags);
+struct xfs_buf *libxfs_trans_get_buf_map(struct xfs_trans *tp,
+ struct xfs_buftarg dev,
+ struct xfs_buf_map *map, int nmaps,
+ uint flags);
static inline struct xfs_buf *
libxfs_trans_get_buf(
struct xfs_trans *tp,
- dev_t dev,
+ struct xfs_buftarg dev,
xfs_daddr_t blkno,
int numblks,
uint flags)
@@ -478,7 +494,8 @@ libxfs_trans_get_buf(
}
int libxfs_trans_read_buf_map(struct xfs_mount *mp,
- struct xfs_trans *tp, dev_t dev,
+ struct xfs_trans *tp,
+ struct xfs_buftarg dev,
struct xfs_buf_map *map, int nmaps,
uint flags, struct xfs_buf **bpp,
const struct xfs_buf_ops *ops);
@@ -487,7 +504,7 @@ static inline int
libxfs_trans_read_buf(
struct xfs_mount *mp,
struct xfs_trans *tp,
- dev_t dev,
+ struct xfs_buftarg dev,
xfs_daddr_t blkno,
int numblks,
uint flags,
@@ -507,7 +524,7 @@ typedef struct xfs_inode {
xfs_mount_t *i_mount; /* fs mount struct ptr */
xfs_ino_t i_ino; /* inode number (agno/agino) */
struct xfs_imap i_imap; /* location for xfs_imap() */
- dev_t i_dev; /* dev for this inode */
+ struct xfs_buftarg i_dev; /* dev for this inode */
xfs_ifork_t *i_afp; /* attribute fork pointer */
xfs_ifork_t i_df; /* data fork */
xfs_trans_t *i_transp; /* ptr to owning transaction */
diff --git a/include/libxlog.h b/include/libxlog.h
index 36ede59..37cbc68 100644
--- a/include/libxlog.h
+++ b/include/libxlog.h
@@ -28,7 +28,7 @@ struct xlog {
xfs_lsn_t l_tail_lsn; /* lsn of 1st LR w/ unflush buffers */
xfs_lsn_t l_last_sync_lsn;/* lsn of last LR on disk */
xfs_mount_t *l_mp; /* mount point */
- dev_t l_dev; /* dev_t of log */
+ struct xfs_buftarg l_dev; /* dev_t of log */
xfs_daddr_t l_logBBstart; /* start block of log */
int l_logsize; /* size of log in bytes */
int l_logBBsize; /* size of log in 512 byte chunks */
diff --git a/libxfs/init.c b/libxfs/init.c
index 1b11fb0..55a17d2 100644
--- a/libxfs/init.c
+++ b/libxfs/init.c
@@ -457,7 +457,7 @@ rtmount_init(
sbp = &mp->m_sb;
if (sbp->sb_rblocks == 0)
return 0;
- if (mp->m_rtdev == 0 && !(flags & LIBXFS_MOUNT_DEBUGGER)) {
+ if (mp->m_rtdev.dev == 0 && !(flags & LIBXFS_MOUNT_DEBUGGER)) {
fprintf(stderr, _("%s: filesystem has a realtime subvolume\n"),
progname);
return -1;
@@ -486,7 +486,7 @@ rtmount_init(
return -1;
}
bp = libxfs_readbuf(mp->m_rtdev,
- d - XFS_FSB_TO_BB(mp, 1), XFS_FSB_TO_BB(mp, 1), 0);
+ d - XFS_FSB_TO_BB(mp, 1), XFS_FSB_TO_BB(mp, 1), 0, NULL);
if (bp == NULL) {
fprintf(stderr, _("%s: realtime size check failed\n"),
progname);
@@ -634,9 +634,13 @@ libxfs_mount(
xfs_sb_t *sbp;
int error;
- mp->m_dev = dev;
- mp->m_rtdev = rtdev;
- mp->m_logdev = logdev;
+ mp->m_ddev_targ.dev = dev;
+ mp->m_ddev_targ.bt_mount = mp;
+ mp->m_logdev_targ.dev = logdev;
+ mp->m_logdev_targ.bt_mount = mp;
+ mp->m_rtdev_targ.dev = rtdev;
+ mp->m_rtdev_targ.bt_mount = mp;
+
mp->m_flags = (LIBXFS_MOUNT_32BITINODES|LIBXFS_MOUNT_32BITINOOPT);
mp->m_sb = *sb;
INIT_RADIX_TREE(&mp->m_perag_tree, GFP_KERNEL);
@@ -720,7 +724,7 @@ libxfs_mount(
bp = libxfs_readbuf(mp->m_dev,
d - XFS_FSS_TO_BB(mp, 1), XFS_FSS_TO_BB(mp, 1),
- !(flags & LIBXFS_MOUNT_DEBUGGER));
+ !(flags & LIBXFS_MOUNT_DEBUGGER), NULL);
if (!bp) {
fprintf(stderr, _("%s: data size check failed\n"), progname);
if (!(flags & LIBXFS_MOUNT_DEBUGGER))
@@ -728,13 +732,13 @@ libxfs_mount(
} else
libxfs_putbuf(bp);
- if (mp->m_logdev && mp->m_logdev != mp->m_dev) {
+ if (mp->m_logdev.dev && mp->m_logdev.dev != mp->m_dev.dev) {
d = (xfs_daddr_t) XFS_FSB_TO_BB(mp, mp->m_sb.sb_logblocks);
if ( (XFS_BB_TO_FSB(mp, d) != mp->m_sb.sb_logblocks) ||
(!(bp = libxfs_readbuf(mp->m_logdev,
d - XFS_FSB_TO_BB(mp, 1),
XFS_FSB_TO_BB(mp, 1),
- !(flags & LIBXFS_MOUNT_DEBUGGER)))) ) {
+ !(flags & LIBXFS_MOUNT_DEBUGGER), NULL))) ) {
fprintf(stderr, _("%s: log size checks failed\n"),
progname);
if (!(flags & LIBXFS_MOUNT_DEBUGGER))
diff --git a/libxfs/logitem.c b/libxfs/logitem.c
index 84e4c14..f298aa0 100644
--- a/libxfs/logitem.c
+++ b/libxfs/logitem.c
@@ -32,7 +32,7 @@ kmem_zone_t *xfs_ili_zone; /* inode log item zone */
xfs_buf_t *
xfs_trans_buf_item_match(
xfs_trans_t *tp,
- dev_t dev,
+ struct xfs_buftarg dev,
struct xfs_buf_map *map,
int nmaps)
{
@@ -47,7 +47,7 @@ xfs_trans_buf_item_match(
list_for_each_entry(lidp, &tp->t_items, lid_trans) {
blip = (struct xfs_buf_log_item *)lidp->lid_item;
if (blip->bli_item.li_type == XFS_LI_BUF &&
- blip->bli_buf->b_dev == dev &&
+ blip->bli_buf->b_target.dev == dev.dev &&
XFS_BUF_ADDR(blip->bli_buf) == map[0].bm_bn &&
blip->bli_buf->b_bcount == BBTOB(len)) {
ASSERT(blip->bli_buf->b_map_count == nmaps);
diff --git a/libxfs/rdwr.c b/libxfs/rdwr.c
index e9cc7b1..db4ed6e 100644
--- a/libxfs/rdwr.c
+++ b/libxfs/rdwr.c
@@ -102,7 +102,7 @@ static xfs_caddr_t next(xfs_caddr_t ptr, int offset, void *private)
int
libxfs_log_clear(
- dev_t device,
+ struct xfs_buftarg device,
xfs_daddr_t start,
uint length,
uuid_t *fs_uuid,
@@ -113,11 +113,11 @@ libxfs_log_clear(
xfs_buf_t *bp;
int len;
- if (!device || !fs_uuid)
+ if (!device.dev || !fs_uuid)
return -EINVAL;
/* first zero the log */
- libxfs_device_zero(device, start, length);
+ libxfs_device_zero(device.dev, start, length);
/* then write a log record header */
len = ((version == 2) && sunit) ? BTOBB(sunit) : 2;
@@ -200,12 +200,15 @@ libxfs_log_header(
#undef libxfs_getbuf_flags
#undef libxfs_putbuf
-xfs_buf_t *libxfs_readbuf(dev_t, xfs_daddr_t, int, int);
-xfs_buf_t *libxfs_readbuf_map(dev_t, struct xfs_buf_map *, int, int);
+xfs_buf_t *libxfs_readbuf(struct xfs_buftarg, xfs_daddr_t, int, int,
+ const struct xfs_buf_map *);
+xfs_buf_t *libxfs_readbuf_map(struct xfs_buftarg, struct xfs_buf_map *,
+ int, int, const struct xfs_buf_map *);
int libxfs_writebuf(xfs_buf_t *, int);
-xfs_buf_t *libxfs_getbuf(dev_t, xfs_daddr_t, int);
-xfs_buf_t *libxfs_getbuf_map(dev_t, struct xfs_buf_map *, int);
-xfs_buf_t *libxfs_getbuf_flags(dev_t, xfs_daddr_t, int, unsigned int);
+xfs_buf_t *libxfs_getbuf(struct xfs_buftarg, xfs_daddr_t, int);
+xfs_buf_t *libxfs_getbuf_map(struct xfs_buftarg, struct xfs_buf_map *, int);
+xfs_buf_t *libxfs_getbuf_flags(struct xfs_buftarg, xfs_daddr_t, int,
+ unsigned int);
void libxfs_putbuf (xfs_buf_t *);
#define __add_trace(bp, func, file, line) \
@@ -219,18 +222,20 @@ do { \
xfs_buf_t *
libxfs_trace_readbuf(const char *func, const char *file, int line,
- dev_t dev, xfs_daddr_t blkno, int len, int flags)
+ struct xfs_buftarg dev, xfs_daddr_t blkno, int len, int flags,
+ const struct xfs_buf_ops *ops)
{
- xfs_buf_t *bp = libxfs_readbuf(dev, blkno, len, flags);
+ xfs_buf_t *bp = libxfs_readbuf(dev, blkno, len, flags, ops);
__add_trace(bp, func, file, line);
return bp;
}
xfs_buf_t *
libxfs_trace_readbuf_map(const char *func, const char *file, int line,
- dev_t dev, struct xfs_buf_map *map, int nmaps, int flags)
+ struct xfs_buftarg dev, struct xfs_buf_map *map, int nmaps, int flags,
+ const struct xfs_buf_ops *ops)
{
- xfs_buf_t *bp = libxfs_readbuf_map(dev, map, nmaps, flags);
+ xfs_buf_t *bp = libxfs_readbuf_map(dev, map, nmaps, flags, ops);
__add_trace(bp, func, file, line);
return bp;
}
@@ -244,7 +249,7 @@ libxfs_trace_writebuf(const char *func, const char *file, int line, xfs_buf_t *b
xfs_buf_t *
libxfs_trace_getbuf(const char *func, const char *file, int line,
- dev_t device, xfs_daddr_t blkno, int len)
+ struct xfs_buftarg device, xfs_daddr_t blkno, int len)
{
xfs_buf_t *bp = libxfs_getbuf(device, blkno, len);
__add_trace(bp, func, file, line);
@@ -253,7 +258,7 @@ libxfs_trace_getbuf(const char *func, const char *file, int line,
xfs_buf_t *
libxfs_trace_getbuf_map(const char *func, const char *file, int line,
- dev_t device, struct xfs_buf_map *map, int nmaps)
+ struct xfs_buftarg device, struct xfs_buf_map *map, int nmaps)
{
xfs_buf_t *bp = libxfs_getbuf_map(device, map, nmaps);
__add_trace(bp, func, file, line);
@@ -262,7 +267,7 @@ libxfs_trace_getbuf_map(const char *func, const char *file, int line,
xfs_buf_t *
libxfs_trace_getbuf_flags(const char *func, const char *file, int line,
- dev_t device, xfs_daddr_t blkno, int len, unsigned int flags)
+ struct xfs_buftarg device, xfs_daddr_t blkno, int len, unsigned int flags)
{
xfs_buf_t *bp = libxfs_getbuf_flags(device, blkno, len, flags);
__add_trace(bp, func, file, line);
@@ -284,7 +289,7 @@ xfs_buf_t *
libxfs_getsb(xfs_mount_t *mp, int flags)
{
return libxfs_readbuf(mp->m_dev, XFS_SB_DADDR,
- XFS_FSS_TO_BB(mp, 1), flags);
+ XFS_FSS_TO_BB(mp, 1), flags, &xfs_sb_buf_ops);
}
kmem_zone_t *xfs_buf_zone;
@@ -302,7 +307,7 @@ static struct cache_mru xfs_buf_freelist =
* buffer initialisation instead of a contiguous buffer.
*/
struct xfs_bufkey {
- dev_t device;
+ struct xfs_buftarg device;
xfs_daddr_t blkno;
unsigned int bblen;
struct xfs_buf_map *map;
@@ -322,7 +327,7 @@ libxfs_bcompare(struct cache_node *node, cache_key_t key)
struct xfs_bufkey *bkey = (struct xfs_bufkey *)key;
#ifdef IO_BCOMPARE_CHECK
- if (bp->b_dev == bkey->device &&
+ if (bp->b_target.dev == bkey->device.dev &&
bp->b_bn == bkey->blkno &&
bp->b_bcount != BBTOB(bkey->bblen))
fprintf(stderr, "%lx: Badness in key lookup (length)\n"
@@ -332,7 +337,7 @@ libxfs_bcompare(struct cache_node *node, cache_key_t key)
(unsigned long long)bkey->blkno, BBTOB(bkey->bblen));
#endif
- return (bp->b_dev == bkey->device &&
+ return (bp->b_target.dev == bkey->device.dev &&
bp->b_bn == bkey->blkno &&
bp->b_bcount == BBTOB(bkey->bblen));
}
@@ -346,13 +351,14 @@ libxfs_bprint(xfs_buf_t *bp)
}
static void
-__initbuf(xfs_buf_t *bp, dev_t device, xfs_daddr_t bno, unsigned int bytes)
+__initbuf(xfs_buf_t *bp, struct xfs_buftarg device, xfs_daddr_t bno,
+ unsigned int bytes)
{
bp->b_flags = 0;
bp->b_bn = bno;
bp->b_bcount = bytes;
bp->b_length = BTOBB(bytes);
- bp->b_dev = device;
+ bp->b_target = device;
bp->b_error = 0;
if (!bp->b_addr)
bp->b_addr = memalign(libxfs_device_alignment(), bytes);
@@ -372,13 +378,15 @@ __initbuf(xfs_buf_t *bp, dev_t device, xfs_daddr_t bno, unsigned int bytes)
}
static void
-libxfs_initbuf(xfs_buf_t *bp, dev_t device, xfs_daddr_t bno, unsigned int bytes)
+libxfs_initbuf(xfs_buf_t *bp, struct xfs_buftarg device, xfs_daddr_t bno,
+ unsigned int bytes)
{
__initbuf(bp, device, bno, bytes);
}
static void
-libxfs_initbuf_map(xfs_buf_t *bp, dev_t device, struct xfs_buf_map *map, int nmaps)
+libxfs_initbuf_map(xfs_buf_t *bp, struct xfs_buftarg device,
+ struct xfs_buf_map *map, int nmaps)
{
unsigned int bytes = 0;
int i;
@@ -441,7 +449,7 @@ __libxfs_getbufr(int blen)
}
xfs_buf_t *
-libxfs_getbufr(dev_t device, xfs_daddr_t blkno, int bblen)
+libxfs_getbufr(struct xfs_buftarg device, xfs_daddr_t blkno, int bblen)
{
xfs_buf_t *bp;
int blen = BBTOB(bblen);
@@ -459,7 +467,7 @@ libxfs_getbufr(dev_t device, xfs_daddr_t blkno, int bblen)
}
xfs_buf_t *
-libxfs_getbufr_map(dev_t device, xfs_daddr_t blkno, int bblen,
+libxfs_getbufr_map(struct xfs_buftarg device, xfs_daddr_t blkno, int bblen,
struct xfs_buf_map *map, int nmaps)
{
xfs_buf_t *bp;
@@ -552,9 +560,10 @@ out_put:
}
struct xfs_buf *
-libxfs_getbuf_flags(dev_t device, xfs_daddr_t blkno, int len, unsigned int flags)
+libxfs_getbuf_flags(struct xfs_buftarg device, xfs_daddr_t blkno, int len,
+ unsigned int flags)
{
- struct xfs_bufkey key = {0};
+ struct xfs_bufkey key = {{0}};
key.device = device;
key.blkno = blkno;
@@ -564,15 +573,15 @@ libxfs_getbuf_flags(dev_t device, xfs_daddr_t blkno, int len, unsigned int flags
}
struct xfs_buf *
-libxfs_getbuf(dev_t device, xfs_daddr_t blkno, int len)
+libxfs_getbuf(struct xfs_buftarg device, xfs_daddr_t blkno, int len)
{
return libxfs_getbuf_flags(device, blkno, len, 0);
}
struct xfs_buf *
-libxfs_getbuf_map(dev_t device, struct xfs_buf_map *map, int nmaps)
+libxfs_getbuf_map(struct xfs_buftarg device, struct xfs_buf_map *map, int nmaps)
{
- struct xfs_bufkey key = {0};
+ struct xfs_bufkey key = {{0}};
int i;
key.device = device;
@@ -610,9 +619,9 @@ libxfs_putbuf(xfs_buf_t *bp)
void
libxfs_purgebuf(xfs_buf_t *bp)
{
- struct xfs_bufkey key = {0};
+ struct xfs_bufkey key = {{0}};
- key.device = bp->b_dev;
+ key.device = bp->b_target;
key.blkno = bp->b_bn;
key.bblen = bp->b_bcount >> BBSHIFT;
@@ -658,9 +667,10 @@ __read_buf(int fd, void *buf, int len, off64_t offset, int flags)
}
int
-libxfs_readbufr(dev_t dev, xfs_daddr_t blkno, xfs_buf_t *bp, int len, int flags)
+libxfs_readbufr(struct xfs_buftarg dev, xfs_daddr_t blkno, xfs_buf_t *bp,
+ int len, int flags)
{
- int fd = libxfs_device_to_fd(dev);
+ int fd = libxfs_device_to_fd(dev.dev);
int bytes = BBTOB(len);
int error;
@@ -668,7 +678,7 @@ libxfs_readbufr(dev_t dev, xfs_daddr_t blkno, xfs_buf_t *bp, int len, int flags)
error = __read_buf(fd, bp->b_addr, bytes, LIBXFS_BBTOOFF64(blkno), flags);
if (!error &&
- bp->b_dev == dev &&
+ bp->b_target.dev == dev.dev &&
bp->b_bn == blkno &&
bp->b_bcount == bytes)
bp->b_flags |= LIBXFS_B_UPTODATE;
@@ -681,22 +691,37 @@ libxfs_readbufr(dev_t dev, xfs_daddr_t blkno, xfs_buf_t *bp, int len, int flags)
}
xfs_buf_t *
-libxfs_readbuf(dev_t dev, xfs_daddr_t blkno, int len, int flags)
+libxfs_readbuf(struct xfs_buftarg dev, xfs_daddr_t blkno, int len, int flags,
+ const struct xfs_buf_ops *ops)
{
xfs_buf_t *bp;
int error;
bp = libxfs_getbuf(dev, blkno, len);
- if (bp && !(bp->b_flags & (LIBXFS_B_UPTODATE|LIBXFS_B_DIRTY))) {
- error = libxfs_readbufr(dev, blkno, bp, len, flags);
- if (error)
- bp->b_error = error;
- }
+ if (!bp)
+ return NULL;
+ if ((bp->b_flags & (LIBXFS_B_UPTODATE|LIBXFS_B_DIRTY)))
+ return bp;
+
+ /*
+ * only set the ops on a cache miss (i.e. first physical read) as the
+ * verifier may change the ops to match the typ eof buffer it contains.
+ * A cache hit might reset the verifier to the original type if we set
+ * it again, but it won't get called again and set to match the buffer
+ * contents. *cough* xfs_da_node_buf_ops *cough*.
+ */
+ bp->b_ops = ops;
+ error = libxfs_readbufr(dev, blkno, bp, len, flags);
+ if (error)
+ bp->b_error = error;
+ else if (bp->b_ops)
+ bp->b_ops->verify_read(bp);
return bp;
}
struct xfs_buf *
-libxfs_readbuf_map(dev_t dev, struct xfs_buf_map *map, int nmaps, int flags)
+libxfs_readbuf_map(struct xfs_buftarg dev, struct xfs_buf_map *map, int nmaps,
+ int flags, const struct xfs_buf_ops *ops)
{
xfs_buf_t *bp;
int error = 0;
@@ -705,15 +730,19 @@ libxfs_readbuf_map(dev_t dev, struct xfs_buf_map *map, int nmaps, int flags)
char *buf;
if (nmaps == 1)
- return libxfs_readbuf(dev, map[0].bm_bn, map[0].bm_len, flags);
+ return libxfs_readbuf(dev, map[0].bm_bn, map[0].bm_len,
+ flags, ops);
bp = libxfs_getbuf_map(dev, map, nmaps);
- if (!bp || (bp->b_flags & (LIBXFS_B_UPTODATE|LIBXFS_B_DIRTY)))
+ if (!bp)
+ return NULL;
+ bp->b_ops = ops;
+ if ((bp->b_flags & (LIBXFS_B_UPTODATE|LIBXFS_B_DIRTY)))
return bp;
ASSERT(bp->b_nmaps = nmaps);
- fd = libxfs_device_to_fd(dev);
+ fd = libxfs_device_to_fd(dev.dev);
buf = bp->b_addr;
for (i = 0; i < bp->b_nmaps; i++) {
off64_t offset = LIBXFS_BBTOOFF64(bp->b_map[i].bm_bn);
@@ -731,8 +760,11 @@ libxfs_readbuf_map(dev_t dev, struct xfs_buf_map *map, int nmaps, int flags)
offset += len;
}
- if (!error)
+ if (!error) {
bp->b_flags |= LIBXFS_B_UPTODATE;
+ if (bp->b_ops)
+ bp->b_ops->verify_read(bp);
+ }
#ifdef IO_DEBUG
printf("%lx: %s: read %lu bytes, error %d, blkno=%llu(%llu), %p\n",
pthread_self(), __FUNCTION__, buf - (char *)bp->b_addr, error,
@@ -767,9 +799,26 @@ __write_buf(int fd, void *buf, int len, off64_t offset, int flags)
int
libxfs_writebufr(xfs_buf_t *bp)
{
- int fd = libxfs_device_to_fd(bp->b_dev);
+ int fd = libxfs_device_to_fd(bp->b_target.dev);
int error = 0;
+ /*
+ * we never write buffers that are marked stale. This indicates they
+ * contain data that has been invalidated, and even if the buffer is
+ * dirty it must *never* be written. Verifiers are wonderful for finding
+ * bugs like this. Make sure the error is obvious as to the cause.
+ */
+ if (bp->b_flags & LIBXFS_B_STALE) {
+ bp->b_error = ESTALE;
+ return bp->b_error;
+ }
+
+ if (bp->b_ops) {
+ bp->b_ops->verify_write(bp);
+ if (bp->b_error)
+ return bp->b_error;
+ }
+
if (!(bp->b_flags & LIBXFS_B_DISCONTIG)) {
error = __write_buf(fd, bp->b_addr, bp->b_bcount,
LIBXFS_BBTOOFF64(bp->b_bn), bp->b_flags);
diff --git a/libxfs/trans.c b/libxfs/trans.c
index 831e42a..7cef147 100644
--- a/libxfs/trans.c
+++ b/libxfs/trans.c
@@ -386,7 +386,7 @@ libxfs_trans_bhold(
xfs_buf_t *
libxfs_trans_get_buf_map(
xfs_trans_t *tp,
- dev_t dev,
+ struct xfs_buftarg dev,
struct xfs_buf_map *map,
int nmaps,
uint f)
@@ -465,7 +465,7 @@ int
libxfs_trans_read_buf_map(
xfs_mount_t *mp,
xfs_trans_t *tp,
- dev_t dev,
+ struct xfs_buftarg dev,
struct xfs_buf_map *map,
int nmaps,
uint flags,
@@ -479,7 +479,7 @@ libxfs_trans_read_buf_map(
*bpp = NULL;
if (tp == NULL) {
- bp = libxfs_readbuf_map(dev, map, nmaps, flags);
+ bp = libxfs_readbuf_map(dev, map, nmaps, flags, ops);
if (!bp) {
return (flags & XBF_TRYLOCK) ?
EAGAIN : XFS_ERROR(ENOMEM);
@@ -498,7 +498,7 @@ libxfs_trans_read_buf_map(
goto done;
}
- bp = libxfs_readbuf_map(dev, map, nmaps, flags);
+ bp = libxfs_readbuf_map(dev, map, nmaps, flags, ops);
if (!bp) {
return (flags & XBF_TRYLOCK) ?
EAGAIN : XFS_ERROR(ENOMEM);
diff --git a/libxfs/xfs.h b/libxfs/xfs.h
index 6bec18e..4bad148 100644
--- a/libxfs/xfs.h
+++ b/libxfs/xfs.h
@@ -55,9 +55,16 @@ typedef __uint32_t inst_t; /* an instruction */
#define EWRONGFS EINVAL
#endif
+#if 0
#define m_ddev_targp m_dev
#define m_logdev_targp m_logdev
#define m_rtdev_targp m_rtdev
+#endif
+
+#define m_ddev_targp m_ddev_targ
+#define m_logdev_targp m_logdev_targ
+#define m_rtdev_targp m_rtdev_targ
+
#define xfs_error_level 0
#define STATIC static
@@ -187,11 +194,7 @@ roundup_pow_of_two(uint v)
NULL; \
})
#define xfs_buf_relse(bp) libxfs_putbuf(bp)
-#define xfs_read_buf(mp,devp,blkno,len,f,bpp) \
- (*(bpp) = libxfs_readbuf((devp), \
- (blkno), (len), 1), 0)
-#define xfs_buf_get(devp,blkno,len,f) \
- (libxfs_getbuf((devp), (blkno), (len)))
+#define xfs_buf_get(devp,blkno,len,f) (libxfs_getbuf((devp), (blkno), (len)))
#define xfs_bwrite(bp) libxfs_writebuf((bp), 0)
#define XBRW_READ LIBXFS_BREAD
@@ -372,7 +375,7 @@ void xfs_buf_item_init (xfs_buf_t *, xfs_mount_t *);
void xfs_buf_item_log (xfs_buf_log_item_t *, uint, uint);
/* xfs_trans_buf.c */
-xfs_buf_t *xfs_trans_buf_item_match(xfs_trans_t *, dev_t,
+xfs_buf_t *xfs_trans_buf_item_match(xfs_trans_t *, struct xfs_buftarg,
struct xfs_buf_map *, int);
/* local source files */
diff --git a/libxfs/xfs_alloc.c b/libxfs/xfs_alloc.c
index 1041f8f..1429c9b 100644
--- a/libxfs/xfs_alloc.c
+++ b/libxfs/xfs_alloc.c
@@ -414,7 +414,7 @@ static bool
xfs_agfl_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_agfl *agfl = XFS_BUF_TO_AGFL(bp);
int i;
@@ -443,7 +443,7 @@ static void
xfs_agfl_read_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
int agfl_ok = 1;
/*
@@ -470,7 +470,7 @@ static void
xfs_agfl_write_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_buf_log_item *bip = bp->b_fspriv;
/* no verification of non-crc AGFLs */
@@ -2205,7 +2205,7 @@ static void
xfs_agf_read_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
int agf_ok = 1;
if (xfs_sb_version_hascrc(&mp->m_sb))
@@ -2225,7 +2225,7 @@ static void
xfs_agf_write_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_buf_log_item *bip = bp->b_fspriv;
if (!xfs_agf_verify(mp, bp)) {
diff --git a/libxfs/xfs_alloc_btree.c b/libxfs/xfs_alloc_btree.c
index 949c96d..a845a15 100644
--- a/libxfs/xfs_alloc_btree.c
+++ b/libxfs/xfs_alloc_btree.c
@@ -258,7 +258,7 @@ static bool
xfs_allocbt_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
struct xfs_perag *pag = bp->b_pag;
unsigned int level;
@@ -337,7 +337,7 @@ xfs_allocbt_read_verify(
xfs_allocbt_verify(bp))) {
trace_xfs_btree_corrupt(bp, _RET_IP_);
XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
- bp->b_target->bt_mount, bp->b_addr);
+ bp->b_target.bt_mount, bp->b_addr);
xfs_buf_ioerror(bp, EFSCORRUPTED);
}
}
@@ -349,7 +349,7 @@ xfs_allocbt_write_verify(
if (!xfs_allocbt_verify(bp)) {
trace_xfs_btree_corrupt(bp, _RET_IP_);
XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
- bp->b_target->bt_mount, bp->b_addr);
+ bp->b_target.bt_mount, bp->b_addr);
xfs_buf_ioerror(bp, EFSCORRUPTED);
}
xfs_btree_sblock_calc_crc(bp);
diff --git a/libxfs/xfs_attr_leaf.c b/libxfs/xfs_attr_leaf.c
index b28266a..6d7ac3b 100644
--- a/libxfs/xfs_attr_leaf.c
+++ b/libxfs/xfs_attr_leaf.c
@@ -150,7 +150,7 @@ static bool
xfs_attr3_leaf_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_attr_leafblock *leaf = bp->b_addr;
struct xfs_attr3_icleaf_hdr ichdr;
@@ -183,7 +183,7 @@ static void
xfs_attr3_leaf_write_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_buf_log_item *bip = bp->b_fspriv;
struct xfs_attr3_leaf_hdr *hdr3 = bp->b_addr;
@@ -212,7 +212,7 @@ static void
xfs_attr3_leaf_read_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
if ((xfs_sb_version_hascrc(&mp->m_sb) &&
!xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
diff --git a/libxfs/xfs_attr_remote.c b/libxfs/xfs_attr_remote.c
index fa112ad..68956ac 100644
--- a/libxfs/xfs_attr_remote.c
+++ b/libxfs/xfs_attr_remote.c
@@ -44,7 +44,7 @@ static bool
xfs_attr3_rmt_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_attr3_rmt_hdr *rmt = bp->b_addr;
if (!xfs_sb_version_hascrc(&mp->m_sb))
@@ -68,7 +68,7 @@ static void
xfs_attr3_rmt_read_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
/* no verification of non-crc buffers */
if (!xfs_sb_version_hascrc(&mp->m_sb))
@@ -86,7 +86,7 @@ static void
xfs_attr3_rmt_write_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_buf_log_item *bip = bp->b_fspriv;
/* no verification of non-crc buffers */
diff --git a/libxfs/xfs_bmap_btree.c b/libxfs/xfs_bmap_btree.c
index 473db4a..761ef29 100644
--- a/libxfs/xfs_bmap_btree.c
+++ b/libxfs/xfs_bmap_btree.c
@@ -704,7 +704,7 @@ static bool
xfs_bmbt_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
unsigned int level;
@@ -764,7 +764,7 @@ xfs_bmbt_read_verify(
xfs_bmbt_verify(bp))) {
trace_xfs_btree_corrupt(bp, _RET_IP_);
XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
- bp->b_target->bt_mount, bp->b_addr);
+ bp->b_target.bt_mount, bp->b_addr);
xfs_buf_ioerror(bp, EFSCORRUPTED);
}
@@ -778,7 +778,7 @@ xfs_bmbt_write_verify(
xfs_warn(bp->b_target->bt_mount, "bmbt daddr 0x%llx failed", bp->b_bn);
trace_xfs_btree_corrupt(bp, _RET_IP_);
XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
- bp->b_target->bt_mount, bp->b_addr);
+ bp->b_target.bt_mount, bp->b_addr);
xfs_buf_ioerror(bp, EFSCORRUPTED);
return;
}
diff --git a/libxfs/xfs_btree.c b/libxfs/xfs_btree.c
index a613294..60a7d61 100644
--- a/libxfs/xfs_btree.c
+++ b/libxfs/xfs_btree.c
@@ -214,7 +214,7 @@ xfs_btree_lblock_calc_crc(
struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
struct xfs_buf_log_item *bip = bp->b_fspriv;
- if (!xfs_sb_version_hascrc(&bp->b_target->bt_mount->m_sb))
+ if (!xfs_sb_version_hascrc(&bp->b_target.bt_mount->m_sb))
return;
if (bip)
block->bb_u.l.bb_lsn = cpu_to_be64(bip->bli_item.li_lsn);
@@ -226,7 +226,7 @@ bool
xfs_btree_lblock_verify_crc(
struct xfs_buf *bp)
{
- if (xfs_sb_version_hascrc(&bp->b_target->bt_mount->m_sb))
+ if (xfs_sb_version_hascrc(&bp->b_target.bt_mount->m_sb))
return xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
XFS_BTREE_LBLOCK_CRC_OFF);
return true;
@@ -247,7 +247,7 @@ xfs_btree_sblock_calc_crc(
struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
struct xfs_buf_log_item *bip = bp->b_fspriv;
- if (!xfs_sb_version_hascrc(&bp->b_target->bt_mount->m_sb))
+ if (!xfs_sb_version_hascrc(&bp->b_target.bt_mount->m_sb))
return;
if (bip)
block->bb_u.s.bb_lsn = cpu_to_be64(bip->bli_item.li_lsn);
@@ -259,7 +259,7 @@ bool
xfs_btree_sblock_verify_crc(
struct xfs_buf *bp)
{
- if (xfs_sb_version_hascrc(&bp->b_target->bt_mount->m_sb))
+ if (xfs_sb_version_hascrc(&bp->b_target.bt_mount->m_sb))
return xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
XFS_BTREE_SBLOCK_CRC_OFF);
return true;
diff --git a/libxfs/xfs_da_btree.c b/libxfs/xfs_da_btree.c
index 0a96f20..fb2f95a 100644
--- a/libxfs/xfs_da_btree.c
+++ b/libxfs/xfs_da_btree.c
@@ -160,7 +160,7 @@ static bool
xfs_da3_node_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_da_intnode *hdr = bp->b_addr;
struct xfs_da3_icnode_hdr ichdr;
@@ -204,7 +204,7 @@ static void
xfs_da3_node_write_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_buf_log_item *bip = bp->b_fspriv;
struct xfs_da3_node_hdr *hdr3 = bp->b_addr;
@@ -233,7 +233,7 @@ static void
xfs_da3_node_read_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_da_blkinfo *info = bp->b_addr;
switch (be16_to_cpu(info->magic)) {
diff --git a/libxfs/xfs_dir2_block.c b/libxfs/xfs_dir2_block.c
index dc69394..004e240 100644
--- a/libxfs/xfs_dir2_block.c
+++ b/libxfs/xfs_dir2_block.c
@@ -45,7 +45,7 @@ static bool
xfs_dir3_block_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr;
if (xfs_sb_version_hascrc(&mp->m_sb)) {
@@ -68,7 +68,7 @@ static void
xfs_dir3_block_read_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
if ((xfs_sb_version_hascrc(&mp->m_sb) &&
!xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
@@ -83,7 +83,7 @@ static void
xfs_dir3_block_write_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_buf_log_item *bip = bp->b_fspriv;
struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr;
diff --git a/libxfs/xfs_dir2_data.c b/libxfs/xfs_dir2_data.c
index 155352c..8182f60 100644
--- a/libxfs/xfs_dir2_data.c
+++ b/libxfs/xfs_dir2_data.c
@@ -48,7 +48,7 @@ __xfs_dir3_data_check(
int stale; /* count of stale leaves */
struct xfs_name name;
- mp = bp->b_target->bt_mount;
+ mp = bp->b_target.bt_mount;
hdr = bp->b_addr;
bf = xfs_dir3_data_bestfree_p(hdr);
p = (char *)xfs_dir3_data_entry_p(hdr);
@@ -178,7 +178,7 @@ static bool
xfs_dir3_data_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr;
if (xfs_sb_version_hascrc(&mp->m_sb)) {
@@ -206,7 +206,7 @@ static void
xfs_dir3_data_reada_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_dir2_data_hdr *hdr = bp->b_addr;
switch (be32_to_cpu(hdr->magic)) {
@@ -230,7 +230,7 @@ static void
xfs_dir3_data_read_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
if ((xfs_sb_version_hascrc(&mp->m_sb) &&
!xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
@@ -245,7 +245,7 @@ static void
xfs_dir3_data_write_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_buf_log_item *bip = bp->b_fspriv;
struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr;
diff --git a/libxfs/xfs_dir2_leaf.c b/libxfs/xfs_dir2_leaf.c
index 12fb61d..1862e0c 100644
--- a/libxfs/xfs_dir2_leaf.c
+++ b/libxfs/xfs_dir2_leaf.c
@@ -162,7 +162,7 @@ xfs_dir3_leaf_verify(
struct xfs_buf *bp,
__uint8_t magic)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_dir2_leaf *leaf = bp->b_addr;
struct xfs_dir3_icleaf_hdr leafhdr;
@@ -194,7 +194,7 @@ __read_verify(
struct xfs_buf *bp,
__uint16_t magic)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
if ((xfs_sb_version_hascrc(&mp->m_sb) &&
!xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
@@ -210,7 +210,7 @@ __write_verify(
struct xfs_buf *bp,
__uint16_t magic)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_buf_log_item *bip = bp->b_fspriv;
struct xfs_dir3_leaf_hdr *hdr3 = bp->b_addr;
diff --git a/libxfs/xfs_dir2_node.c b/libxfs/xfs_dir2_node.c
index eb430f9..36747b1 100644
--- a/libxfs/xfs_dir2_node.c
+++ b/libxfs/xfs_dir2_node.c
@@ -66,7 +66,7 @@ static bool
xfs_dir3_free_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_dir2_free_hdr *hdr = bp->b_addr;
if (xfs_sb_version_hascrc(&mp->m_sb)) {
@@ -92,7 +92,7 @@ static void
xfs_dir3_free_read_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
if ((xfs_sb_version_hascrc(&mp->m_sb) &&
!xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
@@ -107,7 +107,7 @@ static void
xfs_dir3_free_write_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_buf_log_item *bip = bp->b_fspriv;
struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr;
diff --git a/libxfs/xfs_ialloc.c b/libxfs/xfs_ialloc.c
index bea5254..8b0fa0e 100644
--- a/libxfs/xfs_ialloc.c
+++ b/libxfs/xfs_ialloc.c
@@ -1310,7 +1310,7 @@ static bool
xfs_agi_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_agi *agi = XFS_BUF_TO_AGI(bp);
if (xfs_sb_version_hascrc(&mp->m_sb) &&
@@ -1341,7 +1341,7 @@ static void
xfs_agi_read_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
int agi_ok = 1;
if (xfs_sb_version_hascrc(&mp->m_sb))
@@ -1360,7 +1360,7 @@ static void
xfs_agi_write_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_buf_log_item *bip = bp->b_fspriv;
if (!xfs_agi_verify(bp)) {
diff --git a/libxfs/xfs_ialloc_btree.c b/libxfs/xfs_ialloc_btree.c
index ee036bf..d7c865f 100644
--- a/libxfs/xfs_ialloc_btree.c
+++ b/libxfs/xfs_ialloc_btree.c
@@ -167,7 +167,7 @@ static int
xfs_inobt_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
struct xfs_perag *pag = bp->b_pag;
unsigned int level;
@@ -222,7 +222,7 @@ xfs_inobt_read_verify(
xfs_inobt_verify(bp))) {
trace_xfs_btree_corrupt(bp, _RET_IP_);
XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
- bp->b_target->bt_mount, bp->b_addr);
+ bp->b_target.bt_mount, bp->b_addr);
xfs_buf_ioerror(bp, EFSCORRUPTED);
}
}
@@ -234,7 +234,7 @@ xfs_inobt_write_verify(
if (!xfs_inobt_verify(bp)) {
trace_xfs_btree_corrupt(bp, _RET_IP_);
XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
- bp->b_target->bt_mount, bp->b_addr);
+ bp->b_target.bt_mount, bp->b_addr);
xfs_buf_ioerror(bp, EFSCORRUPTED);
}
xfs_btree_sblock_calc_crc(bp);
diff --git a/libxfs/xfs_inode.c b/libxfs/xfs_inode.c
index d6513b9..5ca8f32 100644
--- a/libxfs/xfs_inode.c
+++ b/libxfs/xfs_inode.c
@@ -92,7 +92,7 @@ static void
xfs_inode_buf_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
int i;
int ni;
@@ -237,7 +237,7 @@ xfs_iformat(
}
if (unlikely((ip->i_d.di_flags & XFS_DIFLAG_REALTIME) &&
- !ip->i_mount->m_rtdev)) {
+ !ip->i_mount->m_rtdev.dev)) {
xfs_warn(ip->i_mount,
"corrupt dinode %Lu, has realtime flag set.",
ip->i_ino);
diff --git a/libxfs/xfs_mount.c b/libxfs/xfs_mount.c
index f7daa2f..029f891 100644
--- a/libxfs/xfs_mount.c
+++ b/libxfs/xfs_mount.c
@@ -115,6 +115,9 @@ xfs_perag_put(struct xfs_perag *pag)
/*
* Check the validity of the SB found.
+ *
+ * We don't check the in-progress flag here because it is set by userspace
+ * utilities to indicate operations are in progress....
*/
STATIC int
xfs_mount_validate_sb(
@@ -179,7 +182,7 @@ xfs_mount_validate_sb(
}
if (unlikely(
- sbp->sb_logstart == 0 && mp->m_logdev == mp->m_dev)) {
+ sbp->sb_logstart == 0 && mp->m_logdev.dev == mp->m_dev.dev)) {
xfs_warn(mp,
"filesystem is marked as having an external log; "
"specify logdev on the mount command line.");
@@ -187,7 +190,7 @@ xfs_mount_validate_sb(
}
if (unlikely(
- sbp->sb_logstart != 0 && mp->m_logdev != mp->m_dev)) {
+ sbp->sb_logstart != 0 && mp->m_logdev.dev != mp->m_dev.dev)) {
xfs_warn(mp,
"filesystem is marked as having an internal log; "
"do not specify logdev on the mount command line.");
@@ -243,11 +246,6 @@ xfs_mount_validate_sb(
}
- if (check_inprogress && sbp->sb_inprogress) {
- xfs_warn(mp, "Offline file system operation in progress!");
- return XFS_ERROR(EFSCORRUPTED);
- }
-
/*
* Version 1 directory format has never worked on Linux.
*/
@@ -377,7 +375,7 @@ static int
xfs_sb_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_sb sb;
xfs_sb_from_disk(&sb, XFS_BUF_TO_SBP(bp));
@@ -400,7 +398,7 @@ static void
xfs_sb_read_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_dsb *dsb = XFS_BUF_TO_SBP(bp);
int error;
@@ -454,7 +452,7 @@ static void
xfs_sb_write_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_buf_log_item *bip = bp->b_fspriv;
int error;
diff --git a/libxfs/xfs_symlink.c b/libxfs/xfs_symlink.c
index e018abc..6c42de7 100644
--- a/libxfs/xfs_symlink.c
+++ b/libxfs/xfs_symlink.c
@@ -29,7 +29,7 @@ xfs_symlink_blocks(
/*
* XXX: this need to be used by mkfs/proto.c to create symlinks.
*/
-static int
+int
xfs_symlink_hdr_set(
struct xfs_mount *mp,
xfs_ino_t ino,
@@ -84,7 +84,7 @@ static bool
xfs_symlink_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_dsymlink_hdr *dsl = bp->b_addr;
if (!xfs_sb_version_hascrc(&mp->m_sb))
@@ -108,7 +108,7 @@ static void
xfs_symlink_read_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
/* no verification of non-crc buffers */
if (!xfs_sb_version_hascrc(&mp->m_sb))
@@ -126,7 +126,7 @@ static void
xfs_symlink_write_verify(
struct xfs_buf *bp)
{
- struct xfs_mount *mp = bp->b_target->bt_mount;
+ struct xfs_mount *mp = bp->b_target.bt_mount;
struct xfs_buf_log_item *bip = bp->b_fspriv;
/* no verification of non-crc buffers */
diff --git a/logprint/logprint.c b/logprint/logprint.c
index 2a01780..5452e87 100644
--- a/logprint/logprint.c
+++ b/logprint/logprint.c
@@ -230,7 +230,8 @@ main(int argc, char **argv)
ASSERT(x.logBBsize <= INT_MAX);
- log.l_dev = x.logdev;
+ log.l_dev.dev = x.logdev;
+ log.l_dev.bt_mount = &mount;
log.l_logsize = BBTOB(x.logBBsize);
log.l_logBBstart = x.logBBstart;
log.l_logBBsize = x.logBBsize;
diff --git a/mkfs/maxtrres.c b/mkfs/maxtrres.c
index f12cc70..d571d77 100644
--- a/mkfs/maxtrres.c
+++ b/mkfs/maxtrres.c
@@ -67,6 +67,7 @@ max_trans_res_by_mount(
int
max_trans_res(
+ int crcs_enabled,
int dirversion,
int sectorlog,
int blocklog,
@@ -90,7 +91,8 @@ max_trans_res(
sbp->sb_inodesize = 1 << inodelog;
sbp->sb_inopblock = 1 << (blocklog - inodelog);
sbp->sb_dirblklog = dirblocklog - blocklog;
- sbp->sb_versionnum = XFS_SB_VERSION_4 |
+ sbp->sb_versionnum =
+ (crcs_enabled ? XFS_SB_VERSION_5 : XFS_SB_VERSION_4) |
(dirversion == 2 ? XFS_SB_VERSION_DIRV2BIT : 0);
libxfs_mount(&mount, sbp, 0,0,0,0);
diff --git a/mkfs/proto.c b/mkfs/proto.c
index 56eed31..4254f8a 100644
--- a/mkfs/proto.c
+++ b/mkfs/proto.c
@@ -676,7 +676,7 @@ rtinit(
error);
}
for (i = 0, ep = map; i < nmap; i++, ep++) {
- libxfs_device_zero(mp->m_dev,
+ libxfs_device_zero(mp->m_ddev_targ.dev,
XFS_FSB_TO_DADDR(mp, ep->br_startblock),
XFS_FSB_TO_BB(mp, ep->br_blockcount));
bno += ep->br_blockcount;
@@ -713,7 +713,7 @@ rtinit(
error);
}
for (i = 0, ep = map; i < nmap; i++, ep++) {
- libxfs_device_zero(mp->m_dev,
+ libxfs_device_zero(mp->m_ddev_targ.dev,
XFS_FSB_TO_DADDR(mp, ep->br_startblock),
XFS_FSB_TO_BB(mp, ep->br_blockcount));
bno += ep->br_blockcount;
diff --git a/mkfs/xfs_mkfs.c b/mkfs/xfs_mkfs.c
index c84448e..bf23de1 100644
--- a/mkfs/xfs_mkfs.c
+++ b/mkfs/xfs_mkfs.c
@@ -178,6 +178,12 @@ char *sopts[] = {
NULL
};
+char *mopts[] = {
+#define M_CRC 0
+ "crc",
+ NULL
+};
+
#define TERABYTES(count, blog) ((__uint64_t)(count) << (40 - (blog)))
#define GIGABYTES(count, blog) ((__uint64_t)(count) << (30 - (blog)))
#define MEGABYTES(count, blog) ((__uint64_t)(count) << (20 - (blog)))
@@ -952,6 +958,7 @@ main(
libxfs_init_t xi;
struct fs_topology ft;
int lazy_sb_counters;
+ int crcs_enabled;
progname = basename(argv[0]);
setlocale(LC_ALL, "");
@@ -983,13 +990,14 @@ main(
force_overwrite = 0;
worst_freelist = 0;
lazy_sb_counters = 1;
+ crcs_enabled = 0;
memset(&fsx, 0, sizeof(fsx));
memset(&xi, 0, sizeof(xi));
xi.isdirect = LIBXFS_DIRECT;
xi.isreadonly = LIBXFS_EXCLUSIVELY;
- while ((c = getopt(argc, argv, "b:d:i:l:L:n:KNp:qr:s:CfV")) != EOF) {
+ while ((c = getopt(argc, argv, "b:d:i:l:L:m:n:KNp:qr:s:CfV")) != EOF) {
switch (c) {
case 'C':
case 'f':
@@ -1455,6 +1463,25 @@ main(
illegal(optarg, "L");
label = optarg;
break;
+ case 'm':
+ p = optarg;
+ while (*p != '\0') {
+ char *value;
+
+ switch (getsubopt(&p, (constpp)mopts, &value)) {
+ case M_CRC:
+ if (!value || *value == '\0')
+ reqval('m', mopts, M_CRC);
+ c = atoi(value);
+ if (c < 0 || c > 1)
+ illegal(value, "m crc");
+ crcs_enabled = c;
+ break;
+ default:
+ unknown('m', value);
+ }
+ }
+ break;
case 'n':
p = optarg;
while (*p != '\0') {
@@ -1774,9 +1801,17 @@ _("block size %d cannot be smaller than logical sector size %d\n"),
inodelog = blocklog - libxfs_highbit32(inopblock);
isize = 1 << inodelog;
} else if (!ilflag && !isflag) {
- inodelog = XFS_DINODE_DFL_LOG;
+ inodelog = crcs_enabled ? XFS_DINODE_DFL_CRC_LOG
+ : XFS_DINODE_DFL_LOG;
isize = 1 << inodelog;
}
+ if (crcs_enabled && inodelog < XFS_DINODE_DFL_CRC_LOG) {
+ fprintf(stderr,
+ _("Minimum inode size for CRCs is %d bytes\n"),
+ 1 << XFS_DINODE_DFL_CRC_LOG);
+ usage();
+ }
+
if (xi.lisfile && (!logsize || !xi.logname)) {
fprintf(stderr,
_("if -l file then -l name and -l size are required\n"));
@@ -2025,7 +2060,7 @@ reported by the device (%u).\n"),
sectorsize, xi.rtbsize);
}
- max_tr_res = max_trans_res(dirversion,
+ max_tr_res = max_trans_res(crcs_enabled, dirversion,
sectorlog, blocklog, inodelog, dirblocklog);
ASSERT(max_tr_res);
min_logblocks = max_tr_res * XFS_MIN_LOG_FACTOR;
@@ -2295,7 +2330,7 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
*/
if (!logsize) {
logblocks = MIN(logblocks,
- agsize - XFS_PREALLOC_BLOCKS(mp));
+ XFS_ALLOC_AG_MAX_USABLE(mp));
}
if (logblocks > agsize - XFS_PREALLOC_BLOCKS(mp)) {
fprintf(stderr,
@@ -2338,6 +2373,7 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
printf(_(
"meta-data=%-22s isize=%-6d agcount=%lld, agsize=%lld blks\n"
" =%-22s sectsz=%-5u attr=%u, projid32bit=%u\n"
+ " =%-22s crc=%-5u\n"
"data =%-22s bsize=%-6u blocks=%llu, imaxpct=%u\n"
" =%-22s sunit=%-6u swidth=%u blks\n"
"naming =version %-14u bsize=%-6u ascii-ci=%d\n"
@@ -2346,6 +2382,7 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
"realtime =%-22s extsz=%-6d blocks=%lld, rtextents=%lld\n"),
dfile, isize, (long long)agcount, (long long)agsize,
"", sectorsize, attrversion, projid32bit,
+ "", crcs_enabled,
"", blocksize, (long long)dblocks, imaxpct,
"", dsunit, dswidth,
dirversion, dirblocksize, nci,
@@ -2411,9 +2448,10 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
sbp->sb_logsectlog = 0;
sbp->sb_logsectsize = 0;
}
- sbp->sb_features2 = XFS_SB_VERSION2_MKFS(lazy_sb_counters,
+ sbp->sb_features2 = XFS_SB_VERSION2_MKFS(crcs_enabled, lazy_sb_counters,
attrversion == 2, projid32bit == 1, 0);
- sbp->sb_versionnum = XFS_SB_VERSION_MKFS(iaflag, dsunit != 0,
+ sbp->sb_versionnum = XFS_SB_VERSION_MKFS(crcs_enabled, iaflag,
+ dsunit != 0,
logversion == 2, attrversion == 1,
(sectorsize != BBSIZE ||
lsectorsize != BBSIZE),
@@ -2435,13 +2473,16 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
* swap (somewhere around the page size), jfs (32k),
* ext[2,3] and reiserfs (64k) - and hopefully all else.
*/
- buf = libxfs_getbuf(xi.ddev, 0, BTOBB(WHACK_SIZE));
+ mp->m_ddev_targ.dev = xi.ddev;
+ mp->m_ddev_targ.bt_mount = mp;
+ buf = libxfs_getbuf(mp->m_ddev_targ, 0, BTOBB(WHACK_SIZE));
memset(XFS_BUF_PTR(buf), 0, WHACK_SIZE);
libxfs_writebuf(buf, LIBXFS_EXIT_ON_FAILURE);
libxfs_purgebuf(buf);
/* OK, now write the superblock */
- buf = libxfs_getbuf(xi.ddev, XFS_SB_DADDR, XFS_FSS_TO_BB(mp, 1));
+ buf = libxfs_getbuf(mp->m_ddev_targ, XFS_SB_DADDR, XFS_FSS_TO_BB(mp, 1));
+ buf->b_ops = &xfs_sb_buf_ops;
memset(XFS_BUF_PTR(buf), 0, sectorsize);
libxfs_sb_to_disk((void *)XFS_BUF_PTR(buf), sbp, XFS_SB_ALL_BITS);
libxfs_writebuf(buf, LIBXFS_EXIT_ON_FAILURE);
@@ -2460,10 +2501,11 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
/*
* Zero out the end of the device, to obliterate any
* old MD RAID (or other) metadata at the end of the device.
- * (MD sb is ~64k from the end, take out a wider swath to be sure)
+ * (MD sb is ~64k from the end, take out a wider swath to be sure)
*/
if (!xi.disfile) {
- buf = libxfs_getbuf(xi.ddev, (xi.dsize - BTOBB(WHACK_SIZE)),
+ buf = libxfs_getbuf(mp->m_ddev_targ,
+ (xi.dsize - BTOBB(WHACK_SIZE)),
BTOBB(WHACK_SIZE));
memset(XFS_BUF_PTR(buf), 0, WHACK_SIZE);
libxfs_writebuf(buf, LIBXFS_EXIT_ON_FAILURE);
@@ -2475,10 +2517,13 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
*/
if (loginternal)
xi.logdev = xi.ddev;
- if (xi.logdev)
- libxfs_log_clear(xi.logdev, XFS_FSB_TO_DADDR(mp, logstart),
+ if (xi.logdev) {
+ mp->m_logdev_targ.dev = xi.ddev;
+ libxfs_log_clear(mp->m_logdev_targ,
+ XFS_FSB_TO_DADDR(mp, logstart),
(xfs_extlen_t)XFS_FSB_TO_BB(mp, logblocks),
&sbp->sb_uuid, logversion, lsunit, XLOG_FMT);
+ }
mp = libxfs_mount(mp, sbp, xi.ddev, xi.logdev, xi.rtdev, 1);
if (mp == NULL) {
@@ -2487,13 +2532,22 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
exit(1);
}
+ /*
+ * XXX: this code is effectively shared with the kernel growfs code.
+ * These initialisations should be pulled into libxfs to keep the
+ * kernel/userspace header initialisation code the same.
+ */
for (agno = 0; agno < agcount; agno++) {
+ struct xfs_agfl *agfl;
+ int bucket;
+
/*
* Superblock.
*/
- buf = libxfs_getbuf(xi.ddev,
+ buf = libxfs_getbuf(mp->m_ddev_targ,
XFS_AG_DADDR(mp, agno, XFS_SB_DADDR),
XFS_FSS_TO_BB(mp, 1));
+ buf->b_ops = &xfs_sb_buf_ops;
memset(XFS_BUF_PTR(buf), 0, sectorsize);
libxfs_sb_to_disk((void *)XFS_BUF_PTR(buf), sbp, XFS_SB_ALL_BITS);
libxfs_writebuf(buf, LIBXFS_EXIT_ON_FAILURE);
@@ -2501,9 +2555,10 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
/*
* AG header block: freespace
*/
- buf = libxfs_getbuf(mp->m_dev,
+ buf = libxfs_getbuf(mp->m_ddev_targ,
XFS_AG_DADDR(mp, agno, XFS_AGF_DADDR(mp)),
XFS_FSS_TO_BB(mp, 1));
+ buf->b_ops = &xfs_agf_buf_ops;
agf = XFS_BUF_TO_AGF(buf);
memset(agf, 0, sectorsize);
if (agno == agcount - 1)
@@ -2522,6 +2577,9 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
nbmblocks = (xfs_extlen_t)(agsize - XFS_PREALLOC_BLOCKS(mp));
agf->agf_freeblks = cpu_to_be32(nbmblocks);
agf->agf_longest = cpu_to_be32(nbmblocks);
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ platform_uuid_copy(&agf->agf_uuid, &mp->m_sb.sb_uuid);
+
if (loginternal && agno == logagno) {
be32_add_cpu(&agf->agf_freeblks, -logblocks);
agf->agf_longest = cpu_to_be32(agsize -
@@ -2532,12 +2590,33 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
libxfs_writebuf(buf, LIBXFS_EXIT_ON_FAILURE);
/*
+ * AG freelist header block
+ */
+ buf = libxfs_getbuf(mp->m_ddev_targ,
+ XFS_AG_DADDR(mp, agno, XFS_AGFL_DADDR(mp)),
+ XFS_FSS_TO_BB(mp, 1));
+ buf->b_ops = &xfs_agfl_buf_ops;
+ agfl = XFS_BUF_TO_AGFL(buf);
+ /* setting to 0xff results in initialisation to NULLAGBLOCK */
+ memset(agfl, 0xff, sectorsize);
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ agfl->agfl_magicnum = cpu_to_be32(XFS_AGFL_MAGIC);
+ agfl->agfl_seqno = cpu_to_be32(agno);
+ platform_uuid_copy(&agfl->agfl_uuid, &mp->m_sb.sb_uuid);
+ for (bucket = 0; bucket < XFS_AGFL_SIZE(mp); bucket++)
+ agfl->agfl_bno[bucket] = cpu_to_be32(NULLAGBLOCK);
+ }
+
+ libxfs_writebuf(buf, LIBXFS_EXIT_ON_FAILURE);
+
+ /*
* AG header block: inodes
*/
- buf = libxfs_getbuf(mp->m_dev,
+ buf = libxfs_getbuf(mp->m_ddev_targ,
XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp)),
XFS_FSS_TO_BB(mp, 1));
agi = XFS_BUF_TO_AGI(buf);
+ buf->b_ops = &xfs_agi_buf_ops;
memset(agi, 0, sectorsize);
agi->agi_magicnum = cpu_to_be32(XFS_AGI_MAGIC);
agi->agi_versionnum = cpu_to_be32(XFS_AGI_VERSION);
@@ -2549,6 +2628,8 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
agi->agi_freecount = 0;
agi->agi_newino = cpu_to_be32(NULLAGINO);
agi->agi_dirino = cpu_to_be32(NULLAGINO);
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ platform_uuid_copy(&agi->agi_uuid, &mp->m_sb.sb_uuid);
for (c = 0; c < XFS_AGI_UNLINKED_BUCKETS; c++)
agi->agi_unlinked[c] = cpu_to_be32(NULLAGINO);
libxfs_writebuf(buf, LIBXFS_EXIT_ON_FAILURE);
@@ -2556,16 +2637,19 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
/*
* BNO btree root block
*/
- buf = libxfs_getbuf(mp->m_dev,
+ buf = libxfs_getbuf(mp->m_ddev_targ,
XFS_AGB_TO_DADDR(mp, agno, XFS_BNO_BLOCK(mp)),
bsize);
+ buf->b_ops = &xfs_allocbt_buf_ops;
block = XFS_BUF_TO_BLOCK(buf);
memset(block, 0, blocksize);
- block->bb_magic = cpu_to_be32(XFS_ABTB_MAGIC);
- block->bb_level = 0;
- block->bb_numrecs = cpu_to_be16(1);
- block->bb_u.s.bb_leftsib = cpu_to_be32(NULLAGBLOCK);
- block->bb_u.s.bb_rightsib = cpu_to_be32(NULLAGBLOCK);
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ xfs_btree_init_block(mp, buf, XFS_ABTB_CRC_MAGIC, 0, 1,
+ agno, XFS_BTREE_CRC_BLOCKS);
+ else
+ xfs_btree_init_block(mp, buf, XFS_ABTB_MAGIC, 0, 1,
+ agno, 0);
+
arec = XFS_ALLOC_REC_ADDR(mp, block, 1);
arec->ar_startblock = cpu_to_be32(XFS_PREALLOC_BLOCKS(mp));
if (loginternal && agno == logagno) {
@@ -2599,16 +2683,19 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
/*
* CNT btree root block
*/
- buf = libxfs_getbuf(mp->m_dev,
+ buf = libxfs_getbuf(mp->m_ddev_targ,
XFS_AGB_TO_DADDR(mp, agno, XFS_CNT_BLOCK(mp)),
bsize);
+ buf->b_ops = &xfs_allocbt_buf_ops;
block = XFS_BUF_TO_BLOCK(buf);
memset(block, 0, blocksize);
- block->bb_magic = cpu_to_be32(XFS_ABTC_MAGIC);
- block->bb_level = 0;
- block->bb_numrecs = cpu_to_be16(1);
- block->bb_u.s.bb_leftsib = cpu_to_be32(NULLAGBLOCK);
- block->bb_u.s.bb_rightsib = cpu_to_be32(NULLAGBLOCK);
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ xfs_btree_init_block(mp, buf, XFS_ABTC_CRC_MAGIC, 0, 1,
+ agno, XFS_BTREE_CRC_BLOCKS);
+ else
+ xfs_btree_init_block(mp, buf, XFS_ABTC_MAGIC, 0, 1,
+ agno, 0);
+
arec = XFS_ALLOC_REC_ADDR(mp, block, 1);
arec->ar_startblock = cpu_to_be32(XFS_PREALLOC_BLOCKS(mp));
if (loginternal && agno == logagno) {
@@ -2632,16 +2719,18 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
/*
* INO btree root block
*/
- buf = libxfs_getbuf(mp->m_dev,
+ buf = libxfs_getbuf(mp->m_ddev_targ,
XFS_AGB_TO_DADDR(mp, agno, XFS_IBT_BLOCK(mp)),
bsize);
+ buf->b_ops = &xfs_inobt_buf_ops;
block = XFS_BUF_TO_BLOCK(buf);
memset(block, 0, blocksize);
- block->bb_magic = cpu_to_be32(XFS_IBT_MAGIC);
- block->bb_level = 0;
- block->bb_numrecs = 0;
- block->bb_u.s.bb_leftsib = cpu_to_be32(NULLAGBLOCK);
- block->bb_u.s.bb_rightsib = cpu_to_be32(NULLAGBLOCK);
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ xfs_btree_init_block(mp, buf, XFS_IBT_CRC_MAGIC, 0, 0,
+ agno, XFS_BTREE_CRC_BLOCKS);
+ else
+ xfs_btree_init_block(mp, buf, XFS_IBT_MAGIC, 0, 0,
+ agno, 0);
libxfs_writebuf(buf, LIBXFS_EXIT_ON_FAILURE);
}
@@ -2656,7 +2745,7 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
/*
* Make sure we can write the last block in the realtime area.
*/
- if (mp->m_rtdev && rtblocks > 0) {
+ if (mp->m_rtdev.dev && rtblocks > 0) {
buf = libxfs_getbuf(mp->m_rtdev,
XFS_FSB_TO_BB(mp, rtblocks - 1LL), bsize);
memset(XFS_BUF_PTR(buf), 0, blocksize);
@@ -2710,7 +2799,7 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
XFS_AGB_TO_DADDR(mp, mp->m_sb.sb_agcount-1,
XFS_SB_DADDR),
XFS_FSS_TO_BB(mp, 1),
- LIBXFS_EXIT_ON_FAILURE);
+ LIBXFS_EXIT_ON_FAILURE, &xfs_sb_buf_ops);
XFS_BUF_TO_SBP(buf)->sb_rootino = cpu_to_be64(
mp->m_sb.sb_rootino);
libxfs_writebuf(buf, LIBXFS_EXIT_ON_FAILURE);
@@ -2722,7 +2811,7 @@ an AG size that is one stripe unit smaller, for example %llu.\n"),
XFS_AGB_TO_DADDR(mp, (mp->m_sb.sb_agcount-1)/2,
XFS_SB_DADDR),
XFS_FSS_TO_BB(mp, 1),
- LIBXFS_EXIT_ON_FAILURE);
+ LIBXFS_EXIT_ON_FAILURE, &xfs_sb_buf_ops);
XFS_BUF_TO_SBP(buf)->sb_rootino = cpu_to_be64(
mp->m_sb.sb_rootino);
libxfs_writebuf(buf, LIBXFS_EXIT_ON_FAILURE);
@@ -2878,6 +2967,7 @@ usage( void )
{
fprintf(stderr, _("Usage: %s\n\
/* blocksize */ [-b log=n|size=num]\n\
+/* metadata */ [-m crc=[0|1]\n\
/* data subvol */ [-d agcount=n,agsize=n,file,name=xxx,size=num,\n\
(sunit=value,swidth=value|su=num,sw=num),\n\
sectlog=n|sectsize=num\n\
diff --git a/mkfs/xfs_mkfs.h b/mkfs/xfs_mkfs.h
index f25a7f3..d10e444 100644
--- a/mkfs/xfs_mkfs.h
+++ b/mkfs/xfs_mkfs.h
@@ -23,9 +23,9 @@
XFS_SB_VERSION_EXTFLGBIT | \
XFS_SB_VERSION_DIRV2BIT)
-#define XFS_SB_VERSION_MKFS(ia,dia,log2,attr1,sflag,ci,more) (\
- ((ia)||(dia)||(log2)||(attr1)||(sflag)||(ci)||(more)) ? \
- ( XFS_SB_VERSION_4 | \
+#define XFS_SB_VERSION_MKFS(crc,ia,dia,log2,attr1,sflag,ci,more) (\
+ ((crc)||(ia)||(dia)||(log2)||(attr1)||(sflag)||(ci)||(more)) ? \
+ (((crc) ? XFS_SB_VERSION_5 : XFS_SB_VERSION_4) | \
((ia) ? XFS_SB_VERSION_ALIGNBIT : 0) | \
((dia) ? XFS_SB_VERSION_DALIGNBIT : 0) | \
((log2) ? XFS_SB_VERSION_LOGV2BIT : 0) | \
@@ -36,15 +36,17 @@
XFS_DFL_SB_VERSION_BITS | \
0 ) : XFS_SB_VERSION_1 )
-#define XFS_SB_VERSION2_MKFS(lazycount, attr2, projid32bit, parent) (\
+#define XFS_SB_VERSION2_MKFS(crc, lazycount, attr2, projid32bit, parent) (\
((lazycount) ? XFS_SB_VERSION2_LAZYSBCOUNTBIT : 0) | \
((attr2) ? XFS_SB_VERSION2_ATTR2BIT : 0) | \
((projid32bit) ? XFS_SB_VERSION2_PROJID32BIT : 0) | \
((parent) ? XFS_SB_VERSION2_PARENTBIT : 0) | \
+ ((crc) ? XFS_SB_VERSION2_CRCBIT : 0) | \
0 )
#define XFS_DFL_BLOCKSIZE_LOG 12 /* 4096 byte blocks */
#define XFS_DINODE_DFL_LOG 8 /* 256 byte inodes */
+#define XFS_DINODE_DFL_CRC_LOG 9 /* 512 byte inodes for CRCs */
#define XFS_MIN_DATA_BLOCKS 100
#define XFS_MIN_INODE_PERBLOCK 2 /* min inodes per block */
#define XFS_DFL_IMAXIMUM_PCT 25 /* max % of space for inodes */
@@ -79,7 +81,7 @@ extern void parse_proto (xfs_mount_t *mp, struct fsxattr *fsx, char **pp);
extern void res_failed (int err);
/* maxtrres.c */
-extern int max_trans_res (int dirversion,
+extern int max_trans_res (int crcs_enabled, int dirversion,
int sectorlog, int blocklog, int inodelog, int dirblocklog);
#endif /* __XFS_MKFS_H__ */
diff --git a/repair/attr_repair.c b/repair/attr_repair.c
index 033b7fe..70a2041 100644
--- a/repair/attr_repair.c
+++ b/repair/attr_repair.c
@@ -344,7 +344,7 @@ rmtval_get(xfs_mount_t *mp, xfs_ino_t ino, blkmap_t *blkmap,
break;
}
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, bno),
- XFS_FSB_TO_BB(mp, 1), 0);
+ XFS_FSB_TO_BB(mp, 1), 0, NULL);
if (!bp) {
do_warn(
_("can't read remote block for attributes of inode %" PRIu64 "\n"), ino);
@@ -673,7 +673,7 @@ process_leaf_attr_level(xfs_mount_t *mp,
}
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, dev_bno),
- XFS_FSB_TO_BB(mp, 1), 0);
+ XFS_FSB_TO_BB(mp, 1), 0, NULL);
if (!bp) {
do_warn(
_("can't read file block %u (fsbno %" PRIu64 ") for attribute fork of inode %" PRIu64 "\n"),
@@ -855,7 +855,7 @@ process_longform_attr(
}
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, bno),
- XFS_FSB_TO_BB(mp, 1), 0);
+ XFS_FSB_TO_BB(mp, 1), 0, NULL);
if (!bp) {
do_warn(
_("can't read block 0 of inode %" PRIu64 " attribute fork\n"),
diff --git a/repair/dino_chunks.c b/repair/dino_chunks.c
index 4d14c57..3e7e225 100644
--- a/repair/dino_chunks.c
+++ b/repair/dino_chunks.c
@@ -53,7 +53,7 @@ check_aginode_block(xfs_mount_t *mp,
* so no one else will overlap them.
*/
bp = libxfs_readbuf(mp->m_dev, XFS_AGB_TO_DADDR(mp, agno, agbno),
- XFS_FSB_TO_BB(mp, 1), 0);
+ XFS_FSB_TO_BB(mp, 1), 0, NULL);
if (!bp) {
do_warn(_("cannot read agbno (%u/%u), disk block %" PRId64 "\n"),
agno, agbno, XFS_AGB_TO_DADDR(mp, agno, agbno));
@@ -66,6 +66,8 @@ check_aginode_block(xfs_mount_t *mp,
XFS_OFFBNO_TO_AGINO(mp, agbno, i)))
cnt++;
}
+ if (cnt)
+ bp->b_ops = &xfs_inode_buf_ops;
libxfs_putbuf(bp);
return(cnt);
@@ -626,7 +628,8 @@ process_inode_chunk(
bplist[bp_index] = libxfs_readbuf(mp->m_dev,
XFS_AGB_TO_DADDR(mp, agno, agbno),
- XFS_FSB_TO_BB(mp, blks_per_cluster), 0);
+ XFS_FSB_TO_BB(mp, blks_per_cluster), 0,
+ NULL);
if (!bplist[bp_index]) {
do_warn(_("cannot read inode %" PRIu64 ", disk block %" PRId64 ", cnt %d\n"),
XFS_AGINO_TO_INO(mp, agno, first_irec->ino_startnum),
@@ -640,6 +643,7 @@ process_inode_chunk(
return(1);
}
agbno += blks_per_cluster;
+ bplist[bp_index]->b_ops = &xfs_inode_buf_ops;
pftrace("readbuf %p (%llu, %d) in AG %d", bplist[bp_index],
(long long)XFS_BUF_ADDR(bplist[bp_index]),
diff --git a/repair/dinode.c b/repair/dinode.c
index f564216..7a75dc8 100644
--- a/repair/dinode.c
+++ b/repair/dinode.c
@@ -837,7 +837,8 @@ get_agino_buf(xfs_mount_t *mp,
size = XFS_FSB_TO_BB(mp, MAX(1, XFS_INODES_PER_CHUNK/inodes_per_block));
bp = libxfs_readbuf(mp->m_dev, XFS_AGB_TO_DADDR(mp, agno,
- XFS_AGINO_TO_AGBNO(mp, irec->ino_startnum)), size, 0);
+ XFS_AGINO_TO_AGBNO(mp, irec->ino_startnum)), size, 0,
+ &xfs_inode_buf_ops);
if (!bp) {
do_warn(_("cannot read inode (%u/%u), disk block %" PRIu64 "\n"),
agno, irec->ino_startnum,
@@ -948,7 +949,7 @@ getfunc_btree(xfs_mount_t *mp,
ASSERT(verify_dfsbno(mp, fsbno));
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, fsbno),
- XFS_FSB_TO_BB(mp, 1), 0);
+ XFS_FSB_TO_BB(mp, 1), 0, NULL);
if (!bp) {
do_error(_("cannot read bmap block %" PRIu64 "\n"), fsbno);
return(NULLDFSBNO);
@@ -1005,7 +1006,7 @@ _("- # of bmap records in inode %" PRIu64 " less than minimum (%u, min - %u), pr
*/
libxfs_putbuf(bp);
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, fsbno),
- XFS_FSB_TO_BB(mp, 1), 0);
+ XFS_FSB_TO_BB(mp, 1), 0, NULL);
if (!bp) {
do_error(_("cannot read bmap block %" PRIu64 "\n"),
fsbno);
@@ -1502,7 +1503,8 @@ process_symlink(
if (fsbno != NULLDFSBNO)
bp = libxfs_readbuf(mp->m_dev,
XFS_FSB_TO_DADDR(mp, fsbno),
- XFS_FSB_TO_BB(mp, 1), 0);
+ XFS_FSB_TO_BB(mp, 1), 0,
+ &xfs_symlink_buf_ops);
if (!bp || fsbno == NULLDFSBNO) {
do_warn(
_("cannot read inode %" PRIu64 ", file block %d, disk block %" PRIu64 "\n"),
diff --git a/repair/dir.c b/repair/dir.c
index cb04267..925ba5d 100644
--- a/repair/dir.c
+++ b/repair/dir.c
@@ -763,7 +763,7 @@ traverse_int_dablock(xfs_mount_t *mp,
goto error_out;
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, fsbno),
- XFS_FSB_TO_BB(mp, 1), 0);
+ XFS_FSB_TO_BB(mp, 1), 0, NULL);
if (!bp) {
if (whichfork == XFS_DATA_FORK)
do_warn(
@@ -1143,7 +1143,7 @@ verify_da_path(xfs_mount_t *mp,
}
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, fsbno),
- XFS_FSB_TO_BB(mp, 1), 0);
+ XFS_FSB_TO_BB(mp, 1), 0, NULL);
if (!bp) {
do_warn(
_("can't read block %u (%" PRIu64 ") for directory inode %" PRIu64 "\n"),
@@ -2260,7 +2260,7 @@ process_leaf_dir_level(xfs_mount_t *mp,
bd_addr = (xfs_daddr_t)XFS_FSB_TO_DADDR(mp, dev_bno);
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, dev_bno),
- XFS_FSB_TO_BB(mp, 1), 0);
+ XFS_FSB_TO_BB(mp, 1), 0, NULL);
if (!bp) {
do_warn(
_("can't read file block %u (fsbno %" PRIu64 ", daddr %" PRId64 ") "
@@ -2521,7 +2521,7 @@ process_leaf_dir(
return(1);
}
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, bno),
- XFS_FSB_TO_BB(mp, 1), 0);
+ XFS_FSB_TO_BB(mp, 1), 0, NULL);
if (!bp) {
do_warn(_("can't read block 0 for directory inode %" PRIu64 "\n"),
ino);
diff --git a/repair/dir2.c b/repair/dir2.c
index b774897..274273d 100644
--- a/repair/dir2.c
+++ b/repair/dir2.c
@@ -105,7 +105,7 @@ da_read_buf(
map[i].bm_bn = XFS_FSB_TO_DADDR(mp, bmp[i].startblock);
map[i].bm_len = XFS_FSB_TO_BB(mp, bmp[i].blockcount);
}
- bp = libxfs_readbuf_map(mp->m_dev, map, nex, 0);
+ bp = libxfs_readbuf_map(mp->m_dev, map, nex, 0, NULL);
if (map != map_array)
free(map);
return bp;
diff --git a/repair/phase2.c b/repair/phase2.c
index 23b457a..5a4cf10 100644
--- a/repair/phase2.c
+++ b/repair/phase2.c
@@ -48,7 +48,8 @@ zero_log(xfs_mount_t *mp)
x.logBBsize = XFS_FSB_TO_BB(mp, mp->m_sb.sb_logblocks);
x.logBBstart = XFS_FSB_TO_DADDR(mp, mp->m_sb.sb_logstart);
- log.l_dev = logdev;
+ log.l_dev.dev = logdev;
+ log.l_dev.bt_mount = mp;
log.l_logsize = BBTOB(x.logBBsize);
log.l_logBBsize = x.logBBsize;
log.l_logBBstart = x.logBBstart;
@@ -92,7 +93,7 @@ zero_log(xfs_mount_t *mp)
}
}
- libxfs_log_clear(logdev,
+ libxfs_log_clear(log.l_dev,
XFS_FSB_TO_DADDR(mp, mp->m_sb.sb_logstart),
(xfs_extlen_t)XFS_FSB_TO_BB(mp, mp->m_sb.sb_logblocks),
&mp->m_sb.sb_uuid,
diff --git a/repair/phase3.c b/repair/phase3.c
index 80c66b5..3e43938 100644
--- a/repair/phase3.c
+++ b/repair/phase3.c
@@ -40,7 +40,7 @@ process_agi_unlinked(
bp = libxfs_readbuf(mp->m_dev,
XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp)),
- mp->m_sb.sb_sectsize/BBSIZE, 0);
+ mp->m_sb.sb_sectsize/BBSIZE, 0, &xfs_agi_buf_ops);
if (!bp)
do_error(_("cannot read agi block %" PRId64 " for ag %u\n"),
XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp)), agno);
diff --git a/repair/phase6.c b/repair/phase6.c
index 090e400..8d4d941 100644
--- a/repair/phase6.c
+++ b/repair/phase6.c
@@ -512,7 +512,7 @@ mk_rbmino(xfs_mount_t *mp)
error);
}
for (i = 0, ep = map; i < nmap; i++, ep++) {
- libxfs_device_zero(mp->m_dev,
+ libxfs_device_zero(mp->m_ddev_targ.dev,
XFS_FSB_TO_DADDR(mp, ep->br_startblock),
XFS_FSB_TO_BB(mp, ep->br_blockcount));
bno += ep->br_blockcount;
@@ -767,7 +767,7 @@ mk_rsumino(xfs_mount_t *mp)
error);
}
for (i = 0, ep = map; i < nmap; i++, ep++) {
- libxfs_device_zero(mp->m_dev,
+ libxfs_device_zero(mp->m_ddev_targ.dev,
XFS_FSB_TO_DADDR(mp, ep->br_startblock),
XFS_FSB_TO_BB(mp, ep->br_blockcount));
bno += ep->br_blockcount;
@@ -1256,7 +1256,7 @@ _("can't map block %d in %s inode %" PRIu64 ", xfs_bmapi returns %d, nmap = %d\n
*/
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, fsbno),
- XFS_FSB_TO_BB(mp, 1), 0);
+ XFS_FSB_TO_BB(mp, 1), 0, NULL);
if (!bp) {
do_warn(
@@ -1623,7 +1623,7 @@ longform_dir_entry_check(xfs_mount_t *mp,
skipit = 0;
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, fsbno),
- XFS_FSB_TO_BB(mp, 1), 0);
+ XFS_FSB_TO_BB(mp, 1), 0, NULL);
if (!bp) {
do_error(
diff --git a/repair/prefetch.c b/repair/prefetch.c
index e0eff96..f985e38 100644
--- a/repair/prefetch.c
+++ b/repair/prefetch.c
@@ -222,7 +222,7 @@ pf_scan_lbtree(
int rc;
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, dbno),
- XFS_FSB_TO_BB(mp, 1), 0);
+ XFS_FSB_TO_BB(mp, 1), 0, NULL);
if (!bp)
return 0;
@@ -721,7 +721,7 @@ init_prefetch(
xfs_mount_t *pmp)
{
mp = pmp;
- mp_fd = libxfs_device_to_fd(mp->m_dev);
+ mp_fd = libxfs_device_to_fd(mp->m_ddev_targ.dev);
pf_max_bytes = sysconf(_SC_PAGE_SIZE) << 7;
pf_max_bbs = pf_max_bytes >> BBSHIFT;
pf_max_fsbs = pf_max_bytes >> mp->m_sb.sb_blocklog;
diff --git a/repair/rt.c b/repair/rt.c
index d6ecd56..042ff46 100644
--- a/repair/rt.c
+++ b/repair/rt.c
@@ -206,7 +206,7 @@ process_rtbitmap(xfs_mount_t *mp,
continue;
}
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, bno),
- XFS_FSB_TO_BB(mp, 1));
+ XFS_FSB_TO_BB(mp, 1), NULL);
if (!bp) {
do_warn(_("can't read block %d for rtbitmap inode\n"),
bmbno);
@@ -268,7 +268,7 @@ process_rtsummary(xfs_mount_t *mp,
continue;
}
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, bno),
- XFS_FSB_TO_BB(mp, 1));
+ XFS_FSB_TO_BB(mp, 1), NULL);
if (!bp) {
do_warn(_("can't read block %d for rtsummary inode\n"),
sumbno);
diff --git a/repair/scan.c b/repair/scan.c
index 76bb7f1..82a0fed 100644
--- a/repair/scan.c
+++ b/repair/scan.c
@@ -85,7 +85,7 @@ scan_sbtree(
xfs_buf_t *bp;
bp = libxfs_readbuf(mp->m_dev, XFS_AGB_TO_DADDR(mp, agno, root),
- XFS_FSB_TO_BB(mp, 1), 0);
+ XFS_FSB_TO_BB(mp, 1), 0, NULL);
if (!bp) {
do_error(_("can't read btree block %d/%d\n"), agno, root);
return;
@@ -130,7 +130,7 @@ scan_lbtree(
int dirty = 0;
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, root),
- XFS_FSB_TO_BB(mp, 1), 0);
+ XFS_FSB_TO_BB(mp, 1), 0, NULL);
if (!bp) {
do_error(_("can't read btree block %d/%d\n"),
XFS_FSB_TO_AGNO(mp, root),
@@ -1060,7 +1060,7 @@ scan_freelist(
agflbuf = libxfs_readbuf(mp->m_dev,
XFS_AG_DADDR(mp, agno, XFS_AGFL_DADDR(mp)),
- XFS_FSS_TO_BB(mp, 1), 0);
+ XFS_FSS_TO_BB(mp, 1), 0, &xfs_agfl_buf_ops);
if (!agflbuf) {
do_abort(_("can't read agfl block for ag %d\n"), agno);
return;
@@ -1207,7 +1207,7 @@ scan_ag(
int status;
sbbuf = libxfs_readbuf(mp->m_dev, XFS_AG_DADDR(mp, agno, XFS_SB_DADDR),
- XFS_FSS_TO_BB(mp, 1), 0);
+ XFS_FSS_TO_BB(mp, 1), 0, &xfs_sb_buf_ops);
if (!sbbuf) {
do_error(_("can't get root superblock for ag %d\n"), agno);
return;
@@ -1223,7 +1223,7 @@ scan_ag(
agfbuf = libxfs_readbuf(mp->m_dev,
XFS_AG_DADDR(mp, agno, XFS_AGF_DADDR(mp)),
- XFS_FSS_TO_BB(mp, 1), 0);
+ XFS_FSS_TO_BB(mp, 1), 0, &xfs_agf_buf_ops);
if (!agfbuf) {
do_error(_("can't read agf block for ag %d\n"), agno);
libxfs_putbuf(sbbuf);
@@ -1234,7 +1234,7 @@ scan_ag(
agibuf = libxfs_readbuf(mp->m_dev,
XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp)),
- XFS_FSS_TO_BB(mp, 1), 0);
+ XFS_FSS_TO_BB(mp, 1), 0, &xfs_agi_buf_ops);
if (!agibuf) {
do_error(_("can't read agi block for ag %d\n"), agno);
libxfs_putbuf(agfbuf);
diff --git a/repair/xfs_repair.c b/repair/xfs_repair.c
index 67a7446..167b1f6 100644
--- a/repair/xfs_repair.c
+++ b/repair/xfs_repair.c
@@ -558,9 +558,12 @@ main(int argc, char **argv)
}
/* prepare the mount structure */
- sbp = libxfs_readbuf(x.ddev, XFS_SB_DADDR,
- 1 << (XFS_MAX_SECTORSIZE_LOG - BBSHIFT), 0);
memset(&xfs_m, 0, sizeof(xfs_mount_t));
+ xfs_m.m_ddev_targ.dev = x.ddev;
+ xfs_m.m_ddev_targ.bt_mount = &xfs_m;
+ sbp = libxfs_readbuf(xfs_m.m_ddev_targ, XFS_SB_DADDR,
+ 1 << (XFS_MAX_SECTORSIZE_LOG - BBSHIFT), 0,
+ &xfs_sb_buf_ops);
libxfs_sb_from_disk(&xfs_m.m_sb, XFS_BUF_TO_SBP(sbp));
/*
xfsprogs-crc-patchset-v2/xfsprogs-repair-crc-support 0000600 0001750 0001750 00000130076 12132011547 021604 0 ustar dave dave xfsprogs: add crc format support to repair
From: Dave Chinner <dchinner@xxxxxxxxxx>
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
db/sb.c | 1 +
include/libxfs.h | 5 ++
include/xfs_alloc_btree.h | 2 +-
include/xfs_bmap_btree.h | 2 +-
include/xfs_btree.h | 5 +-
include/xfs_ialloc_btree.h | 2 +-
include/xfs_symlink.h | 2 +
libxfs/rdwr.c | 20 ++++-
libxfs/xfs.h | 12 ++-
libxfs/xfs_alloc.c | 7 +-
libxfs/xfs_btree.c | 20 +++--
libxfs/xfs_ialloc.c | 2 -
libxfs/xfs_mount.c | 16 ----
libxlog/xfs_log_recover.c | 4 +-
logprint/logprint.c | 3 +
repair/agheader.c | 36 ++++++++-
repair/dino_chunks.c | 7 +-
repair/dinode.c | 190 ++++++++++++++++++++++++++------------------
repair/phase2.c | 1 +
repair/phase5.c | 157 +++++++++++++++++++++++++++---------
repair/prefetch.c | 7 +-
repair/scan.c | 158 +++++++++++++++++++-----------------
repair/scan.h | 12 ++-
repair/versions.c | 2 +-
repair/xfs_repair.c | 2 +-
25 files changed, 435 insertions(+), 240 deletions(-)
diff --git a/db/sb.c b/db/sb.c
index f99210c..af86722 100644
--- a/db/sb.c
+++ b/db/sb.c
@@ -240,6 +240,7 @@ sb_logcheck(void)
log.l_logsize = BBTOB(log.l_logBBsize);
log.l_logBBsize = x.logBBsize;
log.l_logBBstart = x.logBBstart;
+ log.l_sectBBsize = x.lbsize;
log.l_mp = mp;
if (xlog_find_tail(&log, &head_blk, &tail_blk)) {
diff --git a/include/libxfs.h b/include/libxfs.h
index 8ee4eaa..0f870f6 100644
--- a/include/libxfs.h
+++ b/include/libxfs.h
@@ -679,6 +679,7 @@ void xfs_bmbt_disk_get_all(xfs_bmbt_rec_t *r, xfs_bmbt_irec_t *s);
#define libxfs_dinode_to_disk xfs_dinode_to_disk
void xfs_dinode_from_disk(struct xfs_icdinode *,
struct xfs_dinode *);
+#define libxfs_dinode_calc_crc xfs_dinode_calc_crc
#define libxfs_idata_realloc xfs_idata_realloc
#define libxfs_idestroy_fork xfs_idestroy_fork
@@ -687,6 +688,10 @@ void xfs_dinode_from_disk(struct xfs_icdinode *,
#define libxfs_sb_from_disk xfs_sb_from_disk
#define libxfs_sb_to_disk xfs_sb_to_disk
+/* xfs_symlink.h */
+#define libxfs_symlink_blocks xfs_symlink_blocks
+#define libxfs_symlink_hdr_ok xfs_symlink_hdr_ok
+
/* xfs_rtalloc.c */
int libxfs_rtfree_extent(struct xfs_trans *, xfs_rtblock_t, xfs_extlen_t);
diff --git a/include/xfs_alloc_btree.h b/include/xfs_alloc_btree.h
index 70c3ea0..e160339 100644
--- a/include/xfs_alloc_btree.h
+++ b/include/xfs_alloc_btree.h
@@ -64,7 +64,7 @@ typedef __be32 xfs_alloc_ptr_t;
*/
#define XFS_ALLOC_BLOCK_LEN(mp) \
(xfs_sb_version_hascrc(&((mp)->m_sb)) ? \
- XFS_BTREE_SBLOCK_LEN + XFS_BTREE_CRCBLOCK_ADD : \
+ XFS_BTREE_SBLOCK_CRC_LEN : \
XFS_BTREE_SBLOCK_LEN)
/*
diff --git a/include/xfs_bmap_btree.h b/include/xfs_bmap_btree.h
index 8a28b89..20d66b0 100644
--- a/include/xfs_bmap_btree.h
+++ b/include/xfs_bmap_btree.h
@@ -140,7 +140,7 @@ typedef __be64 xfs_bmbt_ptr_t, xfs_bmdr_ptr_t;
*/
#define XFS_BMBT_BLOCK_LEN(mp) \
(xfs_sb_version_hascrc(&((mp)->m_sb)) ? \
- XFS_BTREE_LBLOCK_LEN + XFS_BTREE_CRCBLOCK_ADD : \
+ XFS_BTREE_LBLOCK_CRC_LEN : \
XFS_BTREE_LBLOCK_LEN)
#define XFS_BMBT_REC_ADDR(mp, block, index) \
diff --git a/include/xfs_btree.h b/include/xfs_btree.h
index 02f89d8..c0acbbf 100644
--- a/include/xfs_btree.h
+++ b/include/xfs_btree.h
@@ -83,7 +83,10 @@ struct xfs_btree_block {
#define XFS_BTREE_SBLOCK_LEN 16 /* size of a short form block */
#define XFS_BTREE_LBLOCK_LEN 24 /* size of a long form block */
-#define XFS_BTREE_CRCBLOCK_ADD 32 /* size of blkno + crc + uuid */
+
+/* sizes of CRC enabled btree blocks */
+#define XFS_BTREE_SBLOCK_CRC_LEN (XFS_BTREE_SBLOCK_LEN + 40)
+#define XFS_BTREE_LBLOCK_CRC_LEN (XFS_BTREE_LBLOCK_LEN + 48)
#define XFS_BTREE_SBLOCK_CRC_OFF \
offsetof(struct xfs_btree_block, bb_u.s.bb_crc)
diff --git a/include/xfs_ialloc_btree.h b/include/xfs_ialloc_btree.h
index a1bfa7a..7f5ae6b 100644
--- a/include/xfs_ialloc_btree.h
+++ b/include/xfs_ialloc_btree.h
@@ -80,7 +80,7 @@ typedef __be32 xfs_inobt_ptr_t;
*/
#define XFS_INOBT_BLOCK_LEN(mp) \
(xfs_sb_version_hascrc(&((mp)->m_sb)) ? \
- XFS_BTREE_SBLOCK_LEN + XFS_BTREE_CRCBLOCK_ADD : \
+ XFS_BTREE_SBLOCK_CRC_LEN : \
XFS_BTREE_SBLOCK_LEN)
/*
diff --git a/include/xfs_symlink.h b/include/xfs_symlink.h
index bb21e6a..55f3f2d 100644
--- a/include/xfs_symlink.h
+++ b/include/xfs_symlink.h
@@ -29,6 +29,8 @@ struct xfs_dsymlink_hdr {
sizeof(struct xfs_dsymlink_hdr) : 0))
int xfs_symlink_blocks(struct xfs_mount *mp, int pathlen);
+bool xfs_symlink_hdr_ok(struct xfs_mount *mp, xfs_ino_t ino, uint32_t offset,
+ uint32_t size, struct xfs_buf *bp);
extern const struct xfs_buf_ops xfs_symlink_buf_ops;
diff --git a/libxfs/rdwr.c b/libxfs/rdwr.c
index db4ed6e..1b56e57 100644
--- a/libxfs/rdwr.c
+++ b/libxfs/rdwr.c
@@ -444,6 +444,7 @@ __libxfs_getbufr(int blen)
} else
bp = kmem_zone_zalloc(xfs_buf_zone, 0);
pthread_mutex_unlock(&xfs_buf_freelist.cm_mutex);
+ bp->b_ops = NULL;
return bp;
}
@@ -736,6 +737,7 @@ libxfs_readbuf_map(struct xfs_buftarg dev, struct xfs_buf_map *map, int nmaps,
bp = libxfs_getbuf_map(dev, map, nmaps);
if (!bp)
return NULL;
+ bp->b_error = 0;
bp->b_ops = ops;
if ((bp->b_flags & (LIBXFS_B_UPTODATE|LIBXFS_B_DIRTY)))
return bp;
@@ -813,10 +815,20 @@ libxfs_writebufr(xfs_buf_t *bp)
return bp->b_error;
}
+ /*
+ * clear any pre-existing error status on the buffer. This can occur if
+ * the buffer is corrupt on disk and the repair process doesn't clear
+ * the error before fixing and writing it back.
+ */
+ bp->b_error = 0;
if (bp->b_ops) {
bp->b_ops->verify_write(bp);
- if (bp->b_error)
+ if (bp->b_error) {
+ fprintf(stderr,
+ _("%s: write verifer failed on bno 0x%llx/0x%x\n"),
+ __func__, (long long)bp->b_bn, bp->b_bcount);
return bp->b_error;
+ }
}
if (!(bp->b_flags & LIBXFS_B_DISCONTIG)) {
@@ -863,6 +875,12 @@ libxfs_writebuf_int(xfs_buf_t *bp, int flags)
int
libxfs_writebuf(xfs_buf_t *bp, int flags)
{
+#ifdef IO_DEBUG
+ printf("%lx: %s: dirty blkno=%llu(%llu)\n",
+ pthread_self(), __FUNCTION__,
+ (long long)LIBXFS_BBTOOFF64(bp->b_bn),
+ (long long)bp->b_bn);
+#endif
bp->b_flags |= (LIBXFS_B_DIRTY | flags);
libxfs_putbuf(bp);
return 0;
diff --git a/libxfs/xfs.h b/libxfs/xfs.h
index 4bad148..7051ec5 100644
--- a/libxfs/xfs.h
+++ b/libxfs/xfs.h
@@ -79,8 +79,16 @@ typedef __uint32_t inst_t; /* an instruction */
#define IHOLD(ip) ((void) 0)
/* stop unused var warnings by assigning mp to itself */
-#define XFS_CORRUPTION_ERROR(e,l,mp,m) do { (mp) = (mp); } while (0)
-#define XFS_ERROR_REPORT(e,l,mp) do { (mp) = (mp); } while (0)
+#define XFS_CORRUPTION_ERROR(e,l,mp,m) do { \
+ (mp) = (mp); \
+ cmn_err(CE_ALERT, "%s: XFS_CORRUPTION_ERROR", (e)); \
+} while (0)
+
+#define XFS_ERROR_REPORT(e,l,mp) do { \
+ (mp) = (mp); \
+ cmn_err(CE_ALERT, "%s: XFS_ERROR_REPORT", (e)); \
+} while (0)
+
#define XFS_QM_DQATTACH(mp,ip,flags) 0
#define XFS_ERROR(e) (e)
#define XFS_ERRLEVEL_LOW 1
diff --git a/libxfs/xfs_alloc.c b/libxfs/xfs_alloc.c
index 1429c9b..54f1b35 100644
--- a/libxfs/xfs_alloc.c
+++ b/libxfs/xfs_alloc.c
@@ -2173,8 +2173,13 @@ xfs_agf_verify(
struct xfs_agf *agf = XFS_BUF_TO_AGF(bp);
if (xfs_sb_version_hascrc(&mp->m_sb) &&
- !uuid_equal(&agf->agf_uuid, &mp->m_sb.sb_uuid))
+ !uuid_equal(&agf->agf_uuid, &mp->m_sb.sb_uuid)) {
+ char uu[64], uu2[64];
+ platform_uuid_unparse(&agf->agf_uuid, uu);
+ platform_uuid_unparse(&mp->m_sb.sb_uuid, uu2);
+
return false;
+ }
if (!(agf->agf_magicnum == cpu_to_be32(XFS_AGF_MAGIC) &&
XFS_AGF_GOOD_VERSION(be32_to_cpu(agf->agf_versionnum)) &&
diff --git a/libxfs/xfs_btree.c b/libxfs/xfs_btree.c
index 60a7d61..3814b53 100644
--- a/libxfs/xfs_btree.c
+++ b/libxfs/xfs_btree.c
@@ -391,17 +391,15 @@ xfs_btree_dup_cursor(
*/
static inline size_t xfs_btree_block_len(struct xfs_btree_cur *cur)
{
- size_t len;
-
- if (cur->bc_flags & XFS_BTREE_LONG_PTRS)
- len = XFS_BTREE_LBLOCK_LEN;
- else
- len = XFS_BTREE_SBLOCK_LEN;
+ if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
+ if (cur->bc_flags & XFS_BTREE_CRC_BLOCKS)
+ return XFS_BTREE_LBLOCK_CRC_LEN;
+ return XFS_BTREE_LBLOCK_LEN;
+ }
if (cur->bc_flags & XFS_BTREE_CRC_BLOCKS)
- len += XFS_BTREE_CRCBLOCK_ADD;
-
- return len;
+ return XFS_BTREE_SBLOCK_CRC_LEN;
+ return XFS_BTREE_SBLOCK_LEN;
}
/*
@@ -1311,7 +1309,7 @@ xfs_btree_log_block(
offsetof(struct xfs_btree_block, bb_u.s.bb_uuid),
offsetof(struct xfs_btree_block, bb_u.s.bb_owner),
offsetof(struct xfs_btree_block, bb_u.s.bb_crc),
- XFS_BTREE_SBLOCK_LEN + XFS_BTREE_CRCBLOCK_ADD
+ XFS_BTREE_SBLOCK_CRC_LEN
};
static const short loffsets[] = { /* table of offsets (long) */
offsetof(struct xfs_btree_block, bb_magic),
@@ -1325,7 +1323,7 @@ xfs_btree_log_block(
offsetof(struct xfs_btree_block, bb_u.l.bb_owner),
offsetof(struct xfs_btree_block, bb_u.l.bb_crc),
offsetof(struct xfs_btree_block, bb_u.l.bb_pad),
- XFS_BTREE_LBLOCK_LEN + XFS_BTREE_CRCBLOCK_ADD
+ XFS_BTREE_LBLOCK_CRC_LEN
};
XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY);
diff --git a/libxfs/xfs_ialloc.c b/libxfs/xfs_ialloc.c
index 8b0fa0e..282d938 100644
--- a/libxfs/xfs_ialloc.c
+++ b/libxfs/xfs_ialloc.c
@@ -182,8 +182,6 @@ xfs_ialloc_inode_init(
ino = XFS_AGINO_TO_INO(mp, agno,
XFS_OFFBNO_TO_AGINO(mp, agbno, 0));
} else if (xfs_sb_version_hasnlink(&mp->m_sb))
- version = 3;
- else if (xfs_sb_version_hasnlink(&mp->m_sb))
version = 2;
else
version = 1;
diff --git a/libxfs/xfs_mount.c b/libxfs/xfs_mount.c
index 029f891..774fd2c 100644
--- a/libxfs/xfs_mount.c
+++ b/libxfs/xfs_mount.c
@@ -181,22 +181,6 @@ xfs_mount_validate_sb(
}
}
- if (unlikely(
- sbp->sb_logstart == 0 && mp->m_logdev.dev == mp->m_dev.dev)) {
- xfs_warn(mp,
- "filesystem is marked as having an external log; "
- "specify logdev on the mount command line.");
- return XFS_ERROR(EINVAL);
- }
-
- if (unlikely(
- sbp->sb_logstart != 0 && mp->m_logdev.dev != mp->m_dev.dev)) {
- xfs_warn(mp,
- "filesystem is marked as having an internal log; "
- "do not specify logdev on the mount command line.");
- return XFS_ERROR(EINVAL);
- }
-
/*
* More sanity checking. Most of these were stolen directly from
* xfs_repair.
diff --git a/libxlog/xfs_log_recover.c b/libxlog/xfs_log_recover.c
index ad53e86..c66f9a1 100644
--- a/libxlog/xfs_log_recover.c
+++ b/libxlog/xfs_log_recover.c
@@ -19,9 +19,7 @@
#include <xfs/libxlog.h>
#define xfs_readonly_buftarg(buftarg) (0)
-
-/* avoid set-but-unused var warning. gcc is not very bright. */
-#define xlog_clear_stale_blocks(log, taillsn) ((taillsn) = (taillsn))
+#define xlog_clear_stale_blocks(log, taillsn) (0)
/*
diff --git a/logprint/logprint.c b/logprint/logprint.c
index 5452e87..cf24b6e 100644
--- a/logprint/logprint.c
+++ b/logprint/logprint.c
@@ -93,6 +93,7 @@ logstat(xfs_mount_t *mp)
x.logBBsize = XFS_FSB_TO_BB(mp, sb->sb_logblocks);
x.logBBstart = XFS_FSB_TO_DADDR(mp, sb->sb_logstart);
+ x.lbsize = 1 << mp->m_sb.sb_logsectlog;
if (!x.logname && sb->sb_logstart == 0) {
fprintf(stderr, _(" external log device not specified\n\n"));
usage();
@@ -104,6 +105,7 @@ logstat(xfs_mount_t *mp)
stat(x.dname, &s);
x.logBBsize = s.st_size >> 9;
x.logBBstart = 0;
+ x.lbsize = 1;
}
@@ -235,6 +237,7 @@ main(int argc, char **argv)
log.l_logsize = BBTOB(x.logBBsize);
log.l_logBBstart = x.logBBstart;
log.l_logBBsize = x.logBBsize;
+ log.l_sectBBsize = x.lbsize;
log.l_mp = &mount;
switch (print_operation) {
diff --git a/repair/agheader.c b/repair/agheader.c
index 769022d..bc8b1bf 100644
--- a/repair/agheader.c
+++ b/repair/agheader.c
@@ -22,6 +22,11 @@
#include "protos.h"
#include "err_protos.h"
+/*
+ * XXX (dgc): WTF is the point of all the check and repair here when phase 5
+ * recreates the AGF/AGI/AGFL completely from scratch?
+ */
+
static int
verify_set_agf(xfs_mount_t *mp, xfs_agf_t *agf, xfs_agnumber_t i)
{
@@ -104,7 +109,20 @@ verify_set_agf(xfs_mount_t *mp, xfs_agf_t *agf, xfs_agnumber_t i)
/* don't check freespace btrees -- will be checked by caller */
- return(retval);
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return retval;
+
+ if (platform_uuid_compare(&agf->agf_uuid, &mp->m_sb.sb_uuid)) {
+ char uu[64];
+
+ retval = XR_AG_AGF;
+ platform_uuid_unparse(&agf->agf_uuid, uu);
+ do_warn(_("bad uuid %s for agf %d\n"), uu, i);
+
+ if (!no_modify)
+ platform_uuid_copy(&agf->agf_uuid, &mp->m_sb.sb_uuid);
+ }
+ return retval;
}
static int
@@ -169,7 +187,21 @@ verify_set_agi(xfs_mount_t *mp, xfs_agi_t *agi, xfs_agnumber_t agno)
/* don't check inode btree -- will be checked by caller */
- return(retval);
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return retval;
+
+ if (platform_uuid_compare(&agi->agi_uuid, &mp->m_sb.sb_uuid)) {
+ char uu[64];
+
+ retval = XR_AG_AGI;
+ platform_uuid_unparse(&agi->agi_uuid, uu);
+ do_warn(_("bad uuid %s for agi %d\n"), uu, agno);
+
+ if (!no_modify)
+ platform_uuid_copy(&agi->agi_uuid, &mp->m_sb.sb_uuid);
+ }
+
+ return retval;
}
/*
diff --git a/repair/dino_chunks.c b/repair/dino_chunks.c
index 3e7e225..280423b 100644
--- a/repair/dino_chunks.c
+++ b/repair/dino_chunks.c
@@ -629,7 +629,7 @@ process_inode_chunk(
bplist[bp_index] = libxfs_readbuf(mp->m_dev,
XFS_AGB_TO_DADDR(mp, agno, agbno),
XFS_FSB_TO_BB(mp, blks_per_cluster), 0,
- NULL);
+ &xfs_inode_buf_ops);
if (!bplist[bp_index]) {
do_warn(_("cannot read inode %" PRIu64 ", disk block %" PRId64 ", cnt %d\n"),
XFS_AGINO_TO_INO(mp, agno, first_irec->ino_startnum),
@@ -776,8 +776,11 @@ process_inode_chunk(
extra_attr_check, &isa_dir, &parent);
ASSERT(is_used != 3);
- if (ino_dirty)
+ if (ino_dirty) {
dirty = 1;
+ libxfs_dinode_calc_crc(mp, dino);
+ }
+
/*
* XXX - if we want to try and keep
* track of whether we need to bang on
diff --git a/repair/dinode.c b/repair/dinode.c
index 7a75dc8..82bea55 100644
--- a/repair/dinode.c
+++ b/repair/dinode.c
@@ -86,139 +86,127 @@ _("would have cleared inode %" PRIu64 " attributes\n"), ino_num);
}
static int
-clear_dinode_core(xfs_dinode_t *dinoc, xfs_ino_t ino_num)
+clear_dinode_core(struct xfs_mount *mp, xfs_dinode_t *dinoc, xfs_ino_t ino_num)
{
int dirty = 0;
+ int i;
- if (be16_to_cpu(dinoc->di_magic) != XFS_DINODE_MAGIC) {
- dirty = 1;
-
- if (no_modify)
- return(1);
+#define __dirty_no_modify_ret(dirty) \
+ ({ (dirty) = 1; if (no_modify) return 1; })
+ if (be16_to_cpu(dinoc->di_magic) != XFS_DINODE_MAGIC) {
+ __dirty_no_modify_ret(dirty);
dinoc->di_magic = cpu_to_be16(XFS_DINODE_MAGIC);
}
if (!XFS_DINODE_GOOD_VERSION(dinoc->di_version) ||
(!fs_inode_nlink && dinoc->di_version > 1)) {
- dirty = 1;
-
- if (no_modify)
- return(1);
-
- dinoc->di_version = (fs_inode_nlink) ? 2 : 1;
+ __dirty_no_modify_ret(dirty);
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ dinoc->di_version = 3;
+ else
+ dinoc->di_version = (fs_inode_nlink) ? 2 : 1;
}
if (be16_to_cpu(dinoc->di_mode) != 0) {
- dirty = 1;
-
- if (no_modify)
- return(1);
-
+ __dirty_no_modify_ret(dirty);
dinoc->di_mode = 0;
}
if (be16_to_cpu(dinoc->di_flags) != 0) {
- dirty = 1;
-
- if (no_modify)
- return(1);
-
+ __dirty_no_modify_ret(dirty);
dinoc->di_flags = 0;
}
if (be32_to_cpu(dinoc->di_dmevmask) != 0) {
- dirty = 1;
-
- if (no_modify)
- return(1);
-
+ __dirty_no_modify_ret(dirty);
dinoc->di_dmevmask = 0;
}
if (dinoc->di_forkoff != 0) {
- dirty = 1;
-
- if (no_modify)
- return(1);
-
+ __dirty_no_modify_ret(dirty);
dinoc->di_forkoff = 0;
}
if (dinoc->di_format != XFS_DINODE_FMT_EXTENTS) {
- dirty = 1;
-
- if (no_modify)
- return(1);
-
+ __dirty_no_modify_ret(dirty);
dinoc->di_format = XFS_DINODE_FMT_EXTENTS;
}
if (dinoc->di_aformat != XFS_DINODE_FMT_EXTENTS) {
- dirty = 1;
-
- if (no_modify)
- return(1);
-
+ __dirty_no_modify_ret(dirty);
dinoc->di_aformat = XFS_DINODE_FMT_EXTENTS;
}
if (be64_to_cpu(dinoc->di_size) != 0) {
- dirty = 1;
-
- if (no_modify)
- return(1);
-
+ __dirty_no_modify_ret(dirty);
dinoc->di_size = 0;
}
if (be64_to_cpu(dinoc->di_nblocks) != 0) {
- dirty = 1;
-
- if (no_modify)
- return(1);
-
+ __dirty_no_modify_ret(dirty);
dinoc->di_nblocks = 0;
}
if (be16_to_cpu(dinoc->di_onlink) != 0) {
- dirty = 1;
-
- if (no_modify)
- return(1);
-
+ __dirty_no_modify_ret(dirty);
dinoc->di_onlink = 0;
}
if (be32_to_cpu(dinoc->di_nextents) != 0) {
- dirty = 1;
-
- if (no_modify)
- return(1);
-
+ __dirty_no_modify_ret(dirty);
dinoc->di_nextents = 0;
}
if (be16_to_cpu(dinoc->di_anextents) != 0) {
- dirty = 1;
-
- if (no_modify)
- return(1);
-
+ __dirty_no_modify_ret(dirty);
dinoc->di_anextents = 0;
}
if (dinoc->di_version > 1 &&
be32_to_cpu(dinoc->di_nlink) != 0) {
- dirty = 1;
+ __dirty_no_modify_ret(dirty);
+ dinoc->di_nlink = 0;
+ }
- if (no_modify)
- return(1);
+ /* we are done for version 1/2 inodes */
+ if (dinoc->di_version < 3)
+ return dirty;
- dinoc->di_nlink = 0;
+ if (be64_to_cpu(dinoc->di_ino) != ino_num) {
+ __dirty_no_modify_ret(dirty);
+ dinoc->di_ino = cpu_to_be64(ino_num);
}
- return(dirty);
+ if (platform_uuid_compare(&dinoc->di_uuid, &mp->m_sb.sb_uuid)) {
+ __dirty_no_modify_ret(dirty);
+ platform_uuid_copy(&dinoc->di_uuid, &mp->m_sb.sb_uuid);
+ }
+
+ for (i = 0; i < 16; i++) {
+ if (dinoc->di_pad[i] != 0) {
+ __dirty_no_modify_ret(dirty);
+ memset(dinoc->di_pad, 0, 16);
+ break;
+ }
+ }
+
+ if (be64_to_cpu(dinoc->di_flags2) != 0) {
+ __dirty_no_modify_ret(dirty);
+ dinoc->di_flags2 = 0;
+ }
+
+ if (be64_to_cpu(dinoc->di_lsn) != 0) {
+ __dirty_no_modify_ret(dirty);
+ dinoc->di_lsn = 0;
+ }
+
+ if (be64_to_cpu(dinoc->di_changecount) != 0) {
+ __dirty_no_modify_ret(dirty);
+ dinoc->di_changecount = 0;
+ }
+
+ return dirty;
}
static int
@@ -244,7 +232,7 @@ clear_dinode(xfs_mount_t *mp, xfs_dinode_t *dino, xfs_ino_t ino_num)
{
int dirty;
- dirty = clear_dinode_core(dino, ino_num);
+ dirty = clear_dinode_core(mp, dino, ino_num);
dirty += clear_dinode_unlinked(mp, dino);
/* and clear the forks */
@@ -1127,6 +1115,7 @@ process_btinode(
int level;
int numrecs;
bmap_cursor_t cursor;
+ __uint64_t magic;
dib = (xfs_bmdr_block_t *)XFS_DFORK_PTR(dip, whichfork);
lino = XFS_AGINO_TO_INO(mp, agno, ino);
@@ -1138,6 +1127,9 @@ process_btinode(
else
forkname = _("attr");
+ magic = xfs_sb_version_hascrc(&mp->m_sb) ? XFS_BMAP_CRC_MAGIC
+ : XFS_BMAP_MAGIC;
+
level = be16_to_cpu(dib->bb_level);
numrecs = be16_to_cpu(dib->bb_numrecs);
@@ -1191,9 +1183,9 @@ _("bad numrecs 0 in inode %" PRIu64 " bmap btree root block\n"),
return(1);
}
- if (scan_lbtree(be64_to_cpu(pp[i]), level, scanfunc_bmap, type,
+ if (scan_lbtree(be64_to_cpu(pp[i]), level, scan_bmapbt, type,
whichfork, lino, tot, nex, blkmapp, &cursor,
- 1, check_dups))
+ 1, check_dups, magic, &xfs_bmbt_buf_ops))
return(1);
/*
* fix key (offset) mismatches between the keys in root
@@ -1512,9 +1504,21 @@ _("cannot read inode %" PRIu64 ", file block %d, disk block %" PRIu64 "\n"),
return(1);
}
+
buf_data = (char *)XFS_BUF_PTR(bp);
- size = MIN(be64_to_cpu(dino->di_size) - amountdone,
- XFS_FSB_TO_BB(mp, 1) * BBSIZE);
+ size = MIN(be64_to_cpu(dino->di_size) - amountdone,
+ XFS_SYMLINK_BUF_SPACE(mp,
+ mp->m_sb.sb_blocksize));
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ if (!libxfs_symlink_hdr_ok(mp, lino, amountdone,
+ size, bp)) {
+ do_warn(
+_("bad symlink header ino %" PRIu64 ", file block %d, disk block %" PRIu64 "\n"),
+ lino, i, fsbno);
+ return(1);
+ }
+ buf_data += sizeof(struct xfs_dsymlink_hdr);
+ }
memmove(cptr, buf_data, size);
cptr += size;
amountdone += size;
@@ -2464,7 +2468,8 @@ process_dinode_int(xfs_mount_t *mp,
}
if (!XFS_DINODE_GOOD_VERSION(dino->di_version) ||
- (!fs_inode_nlink && dino->di_version > 1)) {
+ (!fs_inode_nlink && dino->di_version > 1) ||
+ (xfs_sb_version_hascrc(&mp->m_sb) && dino->di_version < 3) ) {
retval = 1;
if (!uncertain)
do_warn(_("bad version number 0x%x on inode %" PRIu64 "%c"),
@@ -2473,7 +2478,9 @@ process_dinode_int(xfs_mount_t *mp,
if (!verify_mode) {
if (!no_modify) {
do_warn(_(" resetting version number\n"));
- dino->di_version = (fs_inode_nlink) ? 2 : 1;
+ dino->di_version =
+ xfs_sb_version_hascrc(&mp->m_sb) ? 3 :
+ (fs_inode_nlink) ? 2 : 1;
*dirty = 1;
} else
do_warn(_(" would reset version number\n"));
@@ -2481,6 +2488,31 @@ process_dinode_int(xfs_mount_t *mp,
}
/*
+ * We don't bother checking the CRC here - we cannot guarantee that when
+ * we are called here that the inode has not already been modified in
+ * memory and hence invalidated the CRC.
+ */
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ if (be64_to_cpu(dino->di_ino) != lino) {
+ if (!uncertain)
+ do_warn(
+_("inode identifier %llu mismatch on inode %" PRIu64 "\n"),
+ be64_to_cpu(dino->di_ino), lino);
+ if (verify_mode)
+ return 1;
+ goto clear_bad_out;
+ }
+ if (platform_uuid_compare(&dino->di_uuid, &mp->m_sb.sb_uuid)) {
+ if (!uncertain)
+ do_warn(
+ _("UUID mismatch on inode %" PRIu64 "\n"), lino);
+ if (verify_mode)
+ return 1;
+ goto clear_bad_out;
+ }
+ }
+
+ /*
* blow out of here if the inode size is < 0
*/
if ((xfs_fsize_t)be64_to_cpu(dino->di_size) < 0) {
diff --git a/repair/phase2.c b/repair/phase2.c
index 5a4cf10..0481f51 100644
--- a/repair/phase2.c
+++ b/repair/phase2.c
@@ -64,6 +64,7 @@ zero_log(xfs_mount_t *mp)
ASSERT(mp->m_sb.sb_logsectlog >= BBSHIFT);
}
log.l_sectbb_mask = (1 << log.l_sectbb_log) - 1;
+ log.l_sectBBsize = 1 << mp->m_sb.sb_logsectlog;
if ((error = xlog_find_tail(&log, &head_blk, &tail_blk))) {
do_warn(_("zero_log: cannot find log head/tail "
diff --git a/repair/phase5.c b/repair/phase5.c
index 1f71cac..ce5352e 100644
--- a/repair/phase5.c
+++ b/repair/phase5.c
@@ -602,6 +602,12 @@ prop_freespace_cursor(xfs_mount_t *mp, xfs_agnumber_t agno,
xfs_alloc_ptr_t *bt_ptr;
xfs_agblock_t agbno;
bt_stat_level_t *lptr;
+ __uint32_t crc_magic;
+
+ if (magic == XFS_ABTB_MAGIC)
+ crc_magic = XFS_ABTB_CRC_MAGIC;
+ else
+ crc_magic = XFS_ABTC_CRC_MAGIC;
level++;
@@ -650,14 +656,17 @@ prop_freespace_cursor(xfs_mount_t *mp, xfs_agnumber_t agno,
/*
* initialize block header
*/
+ lptr->buf_p->b_ops = &xfs_allocbt_buf_ops;
bt_hdr = XFS_BUF_TO_BLOCK(lptr->buf_p);
memset(bt_hdr, 0, mp->m_sb.sb_blocksize);
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ xfs_btree_init_block(mp, lptr->buf_p, crc_magic, level,
+ 0, agno, XFS_BTREE_CRC_BLOCKS);
+ else
+ xfs_btree_init_block(mp, lptr->buf_p, magic, level,
+ 0, agno, 0);
- bt_hdr->bb_magic = cpu_to_be32(magic);
- bt_hdr->bb_level = cpu_to_be16(level);
bt_hdr->bb_u.s.bb_leftsib = cpu_to_be32(lptr->prev_agbno);
- bt_hdr->bb_u.s.bb_rightsib = cpu_to_be32(NULLAGBLOCK);
- bt_hdr->bb_numrecs = 0;
/*
* propagate extent record for first extent in new block up
@@ -699,6 +708,7 @@ build_freespace_tree(xfs_mount_t *mp, xfs_agnumber_t agno,
extent_tree_node_t *ext_ptr;
bt_stat_level_t *lptr;
xfs_extlen_t freeblks;
+ __uint32_t crc_magic;
#ifdef XR_BLD_FREE_TRACE
fprintf(stderr, "in build_freespace_tree, agno = %d\n", agno);
@@ -707,6 +717,10 @@ build_freespace_tree(xfs_mount_t *mp, xfs_agnumber_t agno,
freeblks = 0;
ASSERT(level > 0);
+ if (magic == XFS_ABTB_MAGIC)
+ crc_magic = XFS_ABTB_CRC_MAGIC;
+ else
+ crc_magic = XFS_ABTC_CRC_MAGIC;
/*
* initialize the first block on each btree level
@@ -728,14 +742,15 @@ build_freespace_tree(xfs_mount_t *mp, xfs_agnumber_t agno,
/*
* initialize block header
*/
+ lptr->buf_p->b_ops = &xfs_allocbt_buf_ops;
bt_hdr = XFS_BUF_TO_BLOCK(lptr->buf_p);
memset(bt_hdr, 0, mp->m_sb.sb_blocksize);
-
- bt_hdr->bb_magic = cpu_to_be32(magic);
- bt_hdr->bb_level = cpu_to_be16(i);
- bt_hdr->bb_u.s.bb_leftsib = cpu_to_be32(NULLAGBLOCK);
- bt_hdr->bb_u.s.bb_rightsib = cpu_to_be32(NULLAGBLOCK);
- bt_hdr->bb_numrecs = 0;
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ xfs_btree_init_block(mp, lptr->buf_p, crc_magic, i,
+ 0, agno, XFS_BTREE_CRC_BLOCKS);
+ else
+ xfs_btree_init_block(mp, lptr->buf_p, magic, i,
+ 0, agno, 0);
}
/*
* run along leaf, setting up records. as we have to switch
@@ -759,13 +774,17 @@ build_freespace_tree(xfs_mount_t *mp, xfs_agnumber_t agno,
/*
* block initialization, lay in block header
*/
+ lptr->buf_p->b_ops = &xfs_allocbt_buf_ops;
bt_hdr = XFS_BUF_TO_BLOCK(lptr->buf_p);
memset(bt_hdr, 0, mp->m_sb.sb_blocksize);
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ xfs_btree_init_block(mp, lptr->buf_p, crc_magic, 0,
+ 0, agno, XFS_BTREE_CRC_BLOCKS);
+ else
+ xfs_btree_init_block(mp, lptr->buf_p, magic, 0,
+ 0, agno, 0);
- bt_hdr->bb_magic = cpu_to_be32(magic);
- bt_hdr->bb_level = 0;
bt_hdr->bb_u.s.bb_leftsib = cpu_to_be32(lptr->prev_agbno);
- bt_hdr->bb_u.s.bb_rightsib = cpu_to_be32(NULLAGBLOCK);
bt_hdr->bb_numrecs = cpu_to_be16(lptr->num_recs_pb +
(lptr->modulo > 0));
#ifdef XR_BLD_FREE_TRACE
@@ -996,14 +1015,19 @@ prop_ino_cursor(xfs_mount_t *mp, xfs_agnumber_t agno, bt_status_t *btree_curs,
/*
* initialize block header
*/
+ lptr->buf_p->b_ops = &xfs_inobt_buf_ops;
bt_hdr = XFS_BUF_TO_BLOCK(lptr->buf_p);
memset(bt_hdr, 0, mp->m_sb.sb_blocksize);
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ xfs_btree_init_block(mp, lptr->buf_p, XFS_IBT_CRC_MAGIC,
+ level, 0, agno,
+ XFS_BTREE_CRC_BLOCKS);
+ else
+ xfs_btree_init_block(mp, lptr->buf_p, XFS_IBT_MAGIC,
+ level, 0, agno, 0);
- bt_hdr->bb_magic = cpu_to_be32(XFS_IBT_MAGIC);
- bt_hdr->bb_level = cpu_to_be16(level);
bt_hdr->bb_u.s.bb_leftsib = cpu_to_be32(lptr->prev_agbno);
- bt_hdr->bb_u.s.bb_rightsib = cpu_to_be32(NULLAGBLOCK);
- bt_hdr->bb_numrecs = 0;
+
/*
* propagate extent record for first extent in new block up
*/
@@ -1024,6 +1048,9 @@ prop_ino_cursor(xfs_mount_t *mp, xfs_agnumber_t agno, bt_status_t *btree_curs,
*bt_ptr = cpu_to_be32(btree_curs->level[level-1].agbno);
}
+/*
+ * XXX: yet more code that can be shared with mkfs, growfs.
+ */
static void
build_agi(xfs_mount_t *mp, xfs_agnumber_t agno,
bt_status_t *btree_curs, xfs_agino_t first_agino,
@@ -1036,6 +1063,7 @@ build_agi(xfs_mount_t *mp, xfs_agnumber_t agno,
agi_buf = libxfs_getbuf(mp->m_dev,
XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp)),
mp->m_sb.sb_sectsize/BBSIZE);
+ agi_buf->b_ops = &xfs_agi_buf_ops;
agi = XFS_BUF_TO_AGI(agi_buf);
memset(agi, 0, mp->m_sb.sb_sectsize);
@@ -1057,6 +1085,9 @@ build_agi(xfs_mount_t *mp, xfs_agnumber_t agno,
for (i = 0; i < XFS_AGI_UNLINKED_BUCKETS; i++)
agi->agi_unlinked[i] = cpu_to_be32(NULLAGINO);
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ platform_uuid_copy(&agi->agi_uuid, &mp->m_sb.sb_uuid);
+
libxfs_writebuf(agi_buf, 0);
}
@@ -1099,15 +1130,19 @@ build_ino_tree(xfs_mount_t *mp, xfs_agnumber_t agno,
/*
* initialize block header
*/
+
+ lptr->buf_p->b_ops = &xfs_inobt_buf_ops;
bt_hdr = XFS_BUF_TO_BLOCK(lptr->buf_p);
memset(bt_hdr, 0, mp->m_sb.sb_blocksize);
-
- bt_hdr->bb_magic = cpu_to_be32(XFS_IBT_MAGIC);
- bt_hdr->bb_level = cpu_to_be16(i);
- bt_hdr->bb_u.s.bb_leftsib = cpu_to_be32(NULLAGBLOCK);
- bt_hdr->bb_u.s.bb_rightsib = cpu_to_be32(NULLAGBLOCK);
- bt_hdr->bb_numrecs = 0;
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ xfs_btree_init_block(mp, lptr->buf_p, XFS_IBT_CRC_MAGIC,
+ i, 0, agno,
+ XFS_BTREE_CRC_BLOCKS);
+ else
+ xfs_btree_init_block(mp, lptr->buf_p, XFS_IBT_MAGIC,
+ i, 0, agno, 0);
}
+
/*
* run along leaf, setting up records. as we have to switch
* blocks, call the prop_ino_cursor routine to set up the new
@@ -1127,13 +1162,18 @@ build_ino_tree(xfs_mount_t *mp, xfs_agnumber_t agno,
/*
* block initialization, lay in block header
*/
+ lptr->buf_p->b_ops = &xfs_inobt_buf_ops;
bt_hdr = XFS_BUF_TO_BLOCK(lptr->buf_p);
memset(bt_hdr, 0, mp->m_sb.sb_blocksize);
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ xfs_btree_init_block(mp, lptr->buf_p, XFS_IBT_CRC_MAGIC,
+ 0, 0, agno,
+ XFS_BTREE_CRC_BLOCKS);
+ else
+ xfs_btree_init_block(mp, lptr->buf_p, XFS_IBT_MAGIC,
+ 0, 0, agno, 0);
- bt_hdr->bb_magic = cpu_to_be32(XFS_IBT_MAGIC);
- bt_hdr->bb_level = 0;
bt_hdr->bb_u.s.bb_leftsib = cpu_to_be32(lptr->prev_agbno);
- bt_hdr->bb_u.s.bb_rightsib = cpu_to_be32(NULLAGBLOCK);
bt_hdr->bb_numrecs = cpu_to_be16(lptr->num_recs_pb +
(lptr->modulo > 0));
@@ -1192,7 +1232,9 @@ build_ino_tree(xfs_mount_t *mp, xfs_agnumber_t agno,
/*
* build both the agf and the agfl for an agno given both
- * btree cursors
+ * btree cursors.
+ *
+ * XXX: yet more common code that can be shared with mkfs/growfs.
*/
static void
build_agf_agfl(xfs_mount_t *mp,
@@ -1208,10 +1250,12 @@ build_agf_agfl(xfs_mount_t *mp,
int j;
xfs_agfl_t *agfl;
xfs_agf_t *agf;
+ __be32 *freelist;
agf_buf = libxfs_getbuf(mp->m_dev,
XFS_AG_DADDR(mp, agno, XFS_AGF_DADDR(mp)),
mp->m_sb.sb_sectsize/BBSIZE);
+ agf_buf->b_ops = &xfs_agf_buf_ops;
agf = XFS_BUF_TO_AGF(agf_buf);
memset(agf, 0, mp->m_sb.sb_sectsize);
@@ -1265,31 +1309,45 @@ build_agf_agfl(xfs_mount_t *mp,
XFS_BTNUM_CNT);
#endif
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ platform_uuid_copy(&agf->agf_uuid, &mp->m_sb.sb_uuid);
+
+ /* initialise the AGFL, then fill it if there are blocks left over. */
+ agfl_buf = libxfs_getbuf(mp->m_dev,
+ XFS_AG_DADDR(mp, agno, XFS_AGFL_DADDR(mp)),
+ mp->m_sb.sb_sectsize/BBSIZE);
+ agfl_buf->b_ops = &xfs_agfl_buf_ops;
+ agfl = XFS_BUF_TO_AGFL(agfl_buf);
+
+ /* setting to 0xff results in initialisation to NULLAGBLOCK */
+ memset(agfl, 0xff, mp->m_sb.sb_sectsize);
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ agfl->agfl_magicnum = cpu_to_be32(XFS_AGFL_MAGIC);
+ agfl->agfl_seqno = cpu_to_be32(agno);
+ platform_uuid_copy(&agfl->agfl_uuid, &mp->m_sb.sb_uuid);
+ for (i = 0; i < XFS_AGFL_SIZE(mp); i++)
+ agfl->agfl_bno[i] = cpu_to_be32(NULLAGBLOCK);
+ }
+ freelist = XFS_BUF_TO_AGFL_BNO(mp, agfl_buf);
+
/*
* do we have left-over blocks in the btree cursors that should
* be used to fill the AGFL?
*/
if (bno_bt->num_free_blocks > 0 || bcnt_bt->num_free_blocks > 0) {
+
/*
- * yes - grab the AGFL buffer
- */
- agfl_buf = libxfs_getbuf(mp->m_dev,
- XFS_AG_DADDR(mp, agno, XFS_AGFL_DADDR(mp)),
- mp->m_sb.sb_sectsize/BBSIZE);
- agfl = XFS_BUF_TO_AGFL(agfl_buf);
- memset(agfl, 0, mp->m_sb.sb_sectsize);
- /*
- * ok, now grab as many blocks as we can
+ * yes, now grab as many blocks as we can
*/
i = j = 0;
while (bno_bt->num_free_blocks > 0 && i < XFS_AGFL_SIZE(mp)) {
- agfl->agfl_bno[i] = cpu_to_be32(
+ freelist[i] = cpu_to_be32(
get_next_blockaddr(agno, 0, bno_bt));
i++;
}
while (bcnt_bt->num_free_blocks > 0 && i < XFS_AGFL_SIZE(mp)) {
- agfl->agfl_bno[i] = cpu_to_be32(
+ freelist[i] = cpu_to_be32(
get_next_blockaddr(agno, 0, bcnt_bt));
i++;
}
@@ -1324,13 +1382,14 @@ build_agf_agfl(xfs_mount_t *mp,
fprintf(stderr, "writing agfl for ag %u\n", agno);
#endif
- libxfs_writebuf(agfl_buf, 0);
} else {
agf->agf_flfirst = 0;
agf->agf_fllast = cpu_to_be32(XFS_AGFL_SIZE(mp) - 1);
agf->agf_flcount = 0;
}
+ libxfs_writebuf(agfl_buf, 0);
+
ext_ptr = findbiggest_bcnt_extent(agno);
agf->agf_longest = cpu_to_be32((ext_ptr != NULL) ?
ext_ptr->ex_blockcount : 0);
@@ -1340,6 +1399,26 @@ build_agf_agfl(xfs_mount_t *mp,
libxfs_writebuf(agf_buf, 0);
+ /*
+ * now fix up the free list appropriately
+ * XXX: code lifted from mkfs, shoul dbe shared.
+ */
+ {
+ xfs_alloc_arg_t args;
+ xfs_trans_t *tp;
+
+ memset(&args, 0, sizeof(args));
+ args.tp = tp = libxfs_trans_alloc(mp, 0);
+ args.mp = mp;
+ args.agno = agno;
+ args.alignment = 1;
+ args.pag = xfs_perag_get(mp,agno);
+ libxfs_trans_reserve(tp, XFS_MIN_FREELIST(agf, mp), 0, 0, 0, 0);
+ libxfs_alloc_fix_freelist(&args, 0);
+ xfs_perag_put(args.pag);
+ libxfs_trans_commit(tp, 0);
+ }
+
#ifdef XR_BLD_FREE_TRACE
fprintf(stderr, "wrote agf for ag %u, error = %d\n", agno, error);
#endif
diff --git a/repair/prefetch.c b/repair/prefetch.c
index f985e38..a1b2454 100644
--- a/repair/prefetch.c
+++ b/repair/prefetch.c
@@ -222,7 +222,7 @@ pf_scan_lbtree(
int rc;
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, dbno),
- XFS_FSB_TO_BB(mp, 1), 0, NULL);
+ XFS_FSB_TO_BB(mp, 1), 0, &xfs_bmbt_buf_ops);
if (!bp)
return 0;
@@ -338,6 +338,11 @@ pf_read_inode_dirs(
int hasdir = 0;
int isadir;
+ bp->b_ops = &xfs_inode_buf_ops;
+ bp->b_ops->verify_read(bp);
+ if (bp->b_error)
+ return;
+
for (icnt = 0; icnt < (XFS_BUF_COUNT(bp) >> mp->m_sb.sb_inodelog); icnt++) {
dino = xfs_make_iptr(mp, bp, icnt);
diff --git a/repair/scan.c b/repair/scan.c
index 82a0fed..6c72d64 100644
--- a/repair/scan.c
+++ b/repair/scan.c
@@ -48,17 +48,6 @@ struct aghdr_cnts {
__uint64_t ifreecount;
};
-static void
-scanfunc_allocbt(
- struct xfs_btree_block *block,
- int level,
- xfs_agblock_t bno,
- xfs_agnumber_t agno,
- int suspect,
- int isroot,
- __uint32_t magic,
- struct aghdr_cnts *agcnts);
-
void
set_mp(xfs_mount_t *mpp)
{
@@ -78,20 +67,23 @@ scan_sbtree(
xfs_agnumber_t agno,
int suspect,
int isroot,
+ __uint32_t magic,
void *priv),
int isroot,
- void *priv)
+ __uint32_t magic,
+ void *priv,
+ const struct xfs_buf_ops *ops)
{
xfs_buf_t *bp;
bp = libxfs_readbuf(mp->m_dev, XFS_AGB_TO_DADDR(mp, agno, root),
- XFS_FSB_TO_BB(mp, 1), 0, NULL);
+ XFS_FSB_TO_BB(mp, 1), 0, ops);
if (!bp) {
do_error(_("can't read btree block %d/%d\n"), agno, root);
return;
}
(*func)(XFS_BUF_TO_BLOCK(bp), nlevels - 1, root, agno, suspect,
- isroot, priv);
+ isroot, magic, priv);
libxfs_putbuf(bp);
}
@@ -114,7 +106,8 @@ scan_lbtree(
bmap_cursor_t *bm_cursor,
int isroot,
int check_dups,
- int *dirty),
+ int *dirty,
+ __uint64_t magic),
int type,
int whichfork,
xfs_ino_t ino,
@@ -123,14 +116,16 @@ scan_lbtree(
blkmap_t **blkmapp,
bmap_cursor_t *bm_cursor,
int isroot,
- int check_dups)
+ int check_dups,
+ __uint64_t magic,
+ const struct xfs_buf_ops *ops)
{
xfs_buf_t *bp;
int err;
int dirty = 0;
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, root),
- XFS_FSB_TO_BB(mp, 1), 0, NULL);
+ XFS_FSB_TO_BB(mp, 1), 0, ops);
if (!bp) {
do_error(_("can't read btree block %d/%d\n"),
XFS_FSB_TO_AGNO(mp, root),
@@ -139,7 +134,8 @@ scan_lbtree(
}
err = (*func)(XFS_BUF_TO_BLOCK(bp), nlevels - 1,
type, whichfork, root, ino, tot, nex, blkmapp,
- bm_cursor, isroot, check_dups, &dirty);
+ bm_cursor, isroot, check_dups, &dirty,
+ magic);
ASSERT(dirty == 0 || (dirty && !no_modify));
@@ -152,7 +148,7 @@ scan_lbtree(
}
int
-scanfunc_bmap(
+scan_bmapbt(
struct xfs_btree_block *block,
int level,
int type,
@@ -165,7 +161,8 @@ scanfunc_bmap(
bmap_cursor_t *bm_cursor,
int isroot,
int check_dups,
- int *dirty)
+ int *dirty,
+ __uint64_t magic)
{
int i;
int err;
@@ -192,7 +189,7 @@ scanfunc_bmap(
* another inode are claiming the same block but that's
* highly unlikely.
*/
- if (be32_to_cpu(block->bb_magic) != XFS_BMAP_MAGIC) {
+ if (be32_to_cpu(block->bb_magic) != magic) {
do_warn(
_("bad magic # %#x in inode %" PRIu64 " (%s fork) bmbt block %" PRIu64 "\n"),
be32_to_cpu(block->bb_magic), ino, forkname, bno);
@@ -206,6 +203,16 @@ _("expected level %d got %d in inode %" PRIu64 ", (%s fork) bmbt block %" PRIu64
return(1);
}
+ if (magic == XFS_BMAP_CRC_MAGIC) {
+ /* verify owner */
+ if (be64_to_cpu(block->bb_u.l.bb_owner) != ino) {
+ do_warn(
+_("expected owner inode %" PRIu64 ", got %llu, bmbt block %" PRIu64 "\n"),
+ ino, be64_to_cpu(block->bb_u.l.bb_owner), bno);
+ return(1);
+ }
+ }
+
if (check_dups == 0) {
/*
* check sibling pointers. if bad we have a conflict
@@ -408,9 +415,10 @@ _("bad bmap btree ptr 0x%llx in ino %" PRIu64 "\n"),
return(1);
}
- err = scan_lbtree(be64_to_cpu(pp[i]), level, scanfunc_bmap,
+ err = scan_lbtree(be64_to_cpu(pp[i]), level, scan_bmapbt,
type, whichfork, ino, tot, nex, blkmapp,
- bm_cursor, 0, check_dups);
+ bm_cursor, 0, check_dups, magic,
+ &xfs_bmbt_buf_ops);
if (err)
return(1);
@@ -481,35 +489,7 @@ _("bad fwd (right) sibling pointer (saw %" PRIu64 " should be NULLDFSBNO)\n"
}
static void
-scanfunc_bno(
- struct xfs_btree_block *block,
- int level,
- xfs_agblock_t bno,
- xfs_agnumber_t agno,
- int suspect,
- int isroot,
- void *agcnts)
-{
- return scanfunc_allocbt(block, level, bno, agno,
- suspect, isroot, XFS_ABTB_MAGIC, agcnts);
-}
-
-static void
-scanfunc_cnt(
- struct xfs_btree_block *block,
- int level,
- xfs_agblock_t bno,
- xfs_agnumber_t agno,
- int suspect,
- int isroot,
- void *agcnts)
-{
- return scanfunc_allocbt(block, level, bno, agno,
- suspect, isroot, XFS_ABTC_MAGIC, agcnts);
-}
-
-static void
-scanfunc_allocbt(
+scan_allocbt(
struct xfs_btree_block *block,
int level,
xfs_agblock_t bno,
@@ -517,8 +497,9 @@ scanfunc_allocbt(
int suspect,
int isroot,
__uint32_t magic,
- struct aghdr_cnts *agcnts)
+ void *priv)
{
+ struct aghdr_cnts *agcnts = priv;
const char *name;
int i;
xfs_alloc_ptr_t *pp;
@@ -529,9 +510,19 @@ scanfunc_allocbt(
xfs_extlen_t lastcount = 0;
xfs_agblock_t lastblock = 0;
- assert(magic == XFS_ABTB_MAGIC || magic == XFS_ABTC_MAGIC);
-
- name = (magic == XFS_ABTB_MAGIC) ? "bno" : "cnt";
+ switch (magic) {
+ case XFS_ABTB_CRC_MAGIC:
+ case XFS_ABTB_MAGIC:
+ name = "bno";
+ break;
+ case XFS_ABTC_CRC_MAGIC:
+ case XFS_ABTC_MAGIC:
+ name = "cnt";
+ break;
+ default:
+ assert(0);
+ break;
+ }
if (be32_to_cpu(block->bb_magic) != magic) {
do_warn(_("bad magic # %#x in bt%s block %d/%d\n"),
@@ -615,7 +606,8 @@ _("%s freespace btree block claimed (state %d), agno %d, bno %d, suspect %d\n"),
continue;
}
- if (magic == XFS_ABTB_MAGIC) {
+ if (magic == XFS_ABTB_MAGIC ||
+ magic == XFS_ABTB_CRC_MAGIC) {
if (b <= lastblock) {
do_warn(_(
"out-of-order bno btree record %d (%u %u) block %u/%u\n"),
@@ -648,7 +640,8 @@ _("%s freespace btree block claimed (state %d), agno %d, bno %d, suspect %d\n"),
* no warning messages -- we'll catch
* FREE1 blocks later
*/
- if (magic == XFS_ABTC_MAGIC) {
+ if (magic == XFS_ABTC_MAGIC ||
+ magic == XFS_ABTC_CRC_MAGIC) {
set_bmap_ext(agno, b, blen,
XR_E_FREE);
break;
@@ -709,10 +702,20 @@ _("%s freespace btree block claimed (state %d), agno %d, bno %d, suspect %d\n"),
* as possible.
*/
if (bno != 0 && verify_agbno(mp, agno, bno)) {
- scan_sbtree(bno, level, agno, suspect,
- (magic == XFS_ABTB_MAGIC) ?
- scanfunc_bno : scanfunc_cnt, 0,
- (void *)agcnts);
+ switch (magic) {
+ case XFS_ABTB_CRC_MAGIC:
+ case XFS_ABTB_MAGIC:
+ scan_sbtree(bno, level, agno, suspect,
+ scan_allocbt, 0, magic, priv,
+ &xfs_allocbt_buf_ops);
+ break;
+ case XFS_ABTC_CRC_MAGIC:
+ case XFS_ABTC_MAGIC:
+ scan_sbtree(bno, level, agno, suspect,
+ scan_allocbt, 0, magic, priv,
+ &xfs_allocbt_buf_ops);
+ break;
+ }
}
}
}
@@ -896,13 +899,14 @@ _("inode rec for ino %" PRIu64 " (%d/%d) overlaps existing rec (start %d/%d)\n")
* that we aren't sure about go into the uncertain list.
*/
static void
-scanfunc_ino(
+scan_inobt(
struct xfs_btree_block *block,
int level,
xfs_agblock_t bno,
xfs_agnumber_t agno,
int suspect,
int isroot,
+ __uint32_t magic,
void *priv)
{
struct aghdr_cnts *agcnts = priv;
@@ -915,7 +919,7 @@ scanfunc_ino(
hdr_errors = 0;
- if (be32_to_cpu(block->bb_magic) != XFS_IBT_MAGIC) {
+ if (be32_to_cpu(block->bb_magic) != magic) {
do_warn(_("bad magic # %#x in inobt block %d/%d\n"),
be32_to_cpu(block->bb_magic), agno, bno);
hdr_errors++;
@@ -1032,7 +1036,8 @@ _("inode btree block claimed (state %d), agno %d, bno %d, suspect %d\n"),
if (be32_to_cpu(pp[i]) != 0 && verify_agbno(mp, agno,
be32_to_cpu(pp[i])))
scan_sbtree(be32_to_cpu(pp[i]), level, agno,
- suspect, scanfunc_ino, 0, priv);
+ suspect, scan_inobt, 0, magic, priv,
+ &xfs_inobt_buf_ops);
}
}
@@ -1041,12 +1046,12 @@ scan_freelist(
xfs_agf_t *agf,
struct aghdr_cnts *agcnts)
{
- xfs_agfl_t *agfl;
xfs_buf_t *agflbuf;
xfs_agnumber_t agno;
xfs_agblock_t bno;
int count;
int i;
+ __be32 *freelist;
agno = be32_to_cpu(agf->agf_seqno);
@@ -1065,7 +1070,7 @@ scan_freelist(
do_abort(_("can't read agfl block for ag %d\n"), agno);
return;
}
- agfl = XFS_BUF_TO_AGFL(agflbuf);
+ freelist = XFS_BUF_TO_AGFL_BNO(mp, agflbuf);
i = be32_to_cpu(agf->agf_flfirst);
if (no_modify) {
@@ -1080,7 +1085,7 @@ scan_freelist(
count = 0;
for (;;) {
- bno = be32_to_cpu(agfl->agfl_bno[i]);
+ bno = be32_to_cpu(freelist[i]);
if (verify_agbno(mp, agno, bno))
set_bmap(agno, bno, XR_E_FREE);
else
@@ -1109,11 +1114,15 @@ validate_agf(
struct aghdr_cnts *agcnts)
{
xfs_agblock_t bno;
+ __uint32_t magic;
bno = be32_to_cpu(agf->agf_roots[XFS_BTNUM_BNO]);
if (bno != 0 && verify_agbno(mp, agno, bno)) {
+ magic = xfs_sb_version_hascrc(&mp->m_sb) ? XFS_ABTB_CRC_MAGIC
+ : XFS_ABTB_MAGIC;
scan_sbtree(bno, be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]),
- agno, 0, scanfunc_bno, 1, agcnts);
+ agno, 0, scan_allocbt, 1, magic, agcnts,
+ &xfs_allocbt_buf_ops);
} else {
do_warn(_("bad agbno %u for btbno root, agno %d\n"),
bno, agno);
@@ -1121,8 +1130,11 @@ validate_agf(
bno = be32_to_cpu(agf->agf_roots[XFS_BTNUM_CNT]);
if (bno != 0 && verify_agbno(mp, agno, bno)) {
+ magic = xfs_sb_version_hascrc(&mp->m_sb) ? XFS_ABTC_CRC_MAGIC
+ : XFS_ABTC_MAGIC;
scan_sbtree(bno, be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]),
- agno, 0, scanfunc_cnt, 1, agcnts);
+ agno, 0, scan_allocbt, 1, magic, agcnts,
+ &xfs_allocbt_buf_ops);
} else {
do_warn(_("bad agbno %u for btbcnt root, agno %d\n"),
bno, agno);
@@ -1153,11 +1165,15 @@ validate_agi(
{
xfs_agblock_t bno;
int i;
+ __uint32_t magic;
bno = be32_to_cpu(agi->agi_root);
if (bno != 0 && verify_agbno(mp, agno, bno)) {
+ magic = xfs_sb_version_hascrc(&mp->m_sb) ? XFS_IBT_CRC_MAGIC
+ : XFS_IBT_MAGIC;
scan_sbtree(bno, be32_to_cpu(agi->agi_level),
- agno, 0, scanfunc_ino, 1, agcnts);
+ agno, 0, scan_inobt, 1, magic, agcnts,
+ &xfs_inobt_buf_ops);
} else {
do_warn(_("bad agbno %u for inobt root, agno %d\n"),
be32_to_cpu(agi->agi_root), agno);
diff --git a/repair/scan.h b/repair/scan.h
index 9f945cf..92593e9 100644
--- a/repair/scan.h
+++ b/repair/scan.h
@@ -35,7 +35,8 @@ int scan_lbtree(
bmap_cursor_t *bm_cursor,
int isroot,
int check_dups,
- int *dirty),
+ int *dirty,
+ __uint64_t magic),
int type,
int whichfork,
xfs_ino_t ino,
@@ -44,9 +45,11 @@ int scan_lbtree(
struct blkmap **blkmapp,
bmap_cursor_t *bm_cursor,
int isroot,
- int check_dups);
+ int check_dups,
+ __uint64_t magic,
+ const struct xfs_buf_ops *ops);
-int scanfunc_bmap(
+int scan_bmapbt(
struct xfs_btree_block *block,
int level,
int type,
@@ -59,7 +62,8 @@ int scanfunc_bmap(
bmap_cursor_t *bm_cursor,
int isroot,
int check_dups,
- int *dirty);
+ int *dirty,
+ __uint64_t magic);
void
scan_ags(
diff --git a/repair/versions.c b/repair/versions.c
index 957766a..c11a728 100644
--- a/repair/versions.c
+++ b/repair/versions.c
@@ -165,7 +165,7 @@ _("This filesystem contains features not understood by this program.\n"));
return(1);
}
- if (XFS_SB_VERSION_NUM(sb) == XFS_SB_VERSION_4) {
+ if (XFS_SB_VERSION_NUM(sb) >= XFS_SB_VERSION_4) {
if (!fs_sb_feature_bits_allowed) {
if (!no_modify) {
do_warn(
diff --git a/repair/xfs_repair.c b/repair/xfs_repair.c
index 167b1f6..8eefc48 100644
--- a/repair/xfs_repair.c
+++ b/repair/xfs_repair.c
@@ -612,7 +612,7 @@ main(int argc, char **argv)
glob_agcount = mp->m_sb.sb_agcount;
chunks_pblock = mp->m_sb.sb_inopblock / XFS_INODES_PER_CHUNK;
- max_symlink_blocks = howmany(MAXPATHLEN - 1, mp->m_sb.sb_blocksize);
+ max_symlink_blocks = libxfs_symlink_blocks(mp, MAXPATHLEN);
inodes_per_cluster = MAX(mp->m_sb.sb_inopblock,
XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_inodelog);
xfsprogs-crc-patchset-v2/xfsprogs-db-crc-support 0000600 0001750 0001750 00000106751 12132011547 020712 0 ustar dave dave xfsprogs: add crc format support to db
From: Dave Chinner <dchinner@xxxxxxxxxx>
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
db/agf.c | 3 +
db/agfl.c | 16 +++++
db/agfl.h | 2 +
db/agi.c | 3 +
db/btblock.c | 145 ++++++++++++++++++++++++++++++++++++++++++++++
db/btblock.h | 10 ++++
db/check.c | 20 ++++++-
db/field.c | 16 +++++
db/field.h | 8 +++
db/freesp.c | 11 +++-
db/init.c | 4 ++
db/inode.c | 27 +++++++++
db/inode.h | 3 +
db/sb.c | 28 ++++++++-
db/type.c | 35 ++++++++++-
db/type.h | 3 +-
include/xfs_dir2.h | 8 +++
include/xfs_inode.h | 4 +-
libxfs/init.c | 3 +-
logprint/log_misc.c | 2 +-
logprint/log_print_all.c | 4 +-
repair/dir2.c | 15 ++---
repair/phase6.c | 20 +++----
23 files changed, 359 insertions(+), 31 deletions(-)
diff --git a/db/agf.c b/db/agf.c
index 668637a..389cb43 100644
--- a/db/agf.c
+++ b/db/agf.c
@@ -69,6 +69,9 @@ const field_t agf_flds[] = {
{ "freeblks", FLDT_EXTLEN, OI(OFF(freeblks)), C1, 0, TYP_NONE },
{ "longest", FLDT_EXTLEN, OI(OFF(longest)), C1, 0, TYP_NONE },
{ "btreeblks", FLDT_UINT32D, OI(OFF(btreeblks)), C1, 0, TYP_NONE },
+ { "uuid", FLDT_UUID, OI(OFF(uuid)), C1, 0, TYP_NONE },
+ { "lsn", FLDT_UINT64X, OI(OFF(lsn)), C1, 0, TYP_NONE },
+ { "crc", FLDT_UINT32X, OI(OFF(crc)), C1, 0, TYP_NONE },
{ NULL }
};
diff --git a/db/agfl.c b/db/agfl.c
index 72dca23..e2340e6 100644
--- a/db/agfl.c
+++ b/db/agfl.c
@@ -41,8 +41,24 @@ const field_t agfl_hfld[] = { {
{ NULL }
};
+const field_t agfl_crc_hfld[] = { {
+ "", FLDT_AGFL_CRC, OI(0), C1, 0, TYP_NONE, },
+ { NULL }
+};
+
#define OFF(f) bitize(offsetof(xfs_agfl_t, agfl_ ## f))
const field_t agfl_flds[] = {
+ { "bno", FLDT_AGBLOCKNZ, OI(OFF(magicnum)), agfl_bno_size,
+ FLD_ARRAY|FLD_COUNT, TYP_DATA },
+ { NULL }
+};
+
+const field_t agfl_crc_flds[] = {
+ { "magicnum", FLDT_UINT32X, OI(OFF(magicnum)), C1, 0, TYP_NONE },
+ { "seqno", FLDT_AGNUMBER, OI(OFF(seqno)), C1, 0, TYP_NONE },
+ { "uuid", FLDT_UUID, OI(OFF(uuid)), C1, 0, TYP_NONE },
+ { "lsn", FLDT_UINT64X, OI(OFF(lsn)), C1, 0, TYP_NONE },
+ { "crc", FLDT_UINT32X, OI(OFF(crc)), C1, 0, TYP_NONE },
{ "bno", FLDT_AGBLOCKNZ, OI(OFF(bno)), agfl_bno_size,
FLD_ARRAY|FLD_COUNT, TYP_DATA },
{ NULL }
diff --git a/db/agfl.h b/db/agfl.h
index 7b7631b..177ad41 100644
--- a/db/agfl.h
+++ b/db/agfl.h
@@ -18,6 +18,8 @@
extern const struct field agfl_flds[];
extern const struct field agfl_hfld[];
+extern const struct field agfl_crc_flds[];
+extern const struct field agfl_crc_hfld[];
extern void agfl_init(void);
extern int agfl_size(void *obj, int startoff, int idx);
diff --git a/db/agi.c b/db/agi.c
index 02d5d30..6b2e889 100644
--- a/db/agi.c
+++ b/db/agi.c
@@ -54,6 +54,9 @@ const field_t agi_flds[] = {
{ "dirino", FLDT_AGINO, OI(OFF(dirino)), C1, 0, TYP_INODE },
{ "unlinked", FLDT_AGINONN, OI(OFF(unlinked)),
CI(XFS_AGI_UNLINKED_BUCKETS), FLD_ARRAY, TYP_NONE },
+ { "uuid", FLDT_UUID, OI(OFF(uuid)), C1, 0, TYP_NONE },
+ { "lsn", FLDT_UINT64X, OI(OFF(lsn)), C1, 0, TYP_NONE },
+ { "crc", FLDT_UINT32X, OI(OFF(crc)), C1, 0, TYP_NONE },
{ NULL }
};
diff --git a/db/btblock.c b/db/btblock.c
index 2c199b2..37b9903 100644
--- a/db/btblock.c
+++ b/db/btblock.c
@@ -60,6 +60,31 @@ struct xfs_db_btree {
sizeof(xfs_inobt_rec_t),
sizeof(__be32),
},
+ [/*0x424d415*/8] = { /* BMAP_CRC */
+ XFS_BTREE_LBLOCK_CRC_LEN,
+ sizeof(xfs_bmbt_key_t),
+ sizeof(xfs_bmbt_rec_t),
+ sizeof(__be64),
+ },
+ [/*0x4142544*/0xa] = { /* ABTB_CRC */
+ XFS_BTREE_SBLOCK_CRC_LEN,
+ sizeof(xfs_alloc_key_t),
+ sizeof(xfs_alloc_rec_t),
+ sizeof(__be32),
+ },
+ [/*0x414254*/0xb] = { /* ABTC_CRC */
+ XFS_BTREE_SBLOCK_CRC_LEN,
+ sizeof(xfs_alloc_key_t),
+ sizeof(xfs_alloc_rec_t),
+ sizeof(__be32),
+ },
+ [/*0x4941425*/0xc] = { /* IABT_CRC */
+ XFS_BTREE_SBLOCK_CRC_LEN,
+ sizeof(xfs_inobt_key_t),
+ sizeof(xfs_inobt_rec_t),
+ sizeof(__be32),
+ },
+
};
/*
@@ -208,6 +233,15 @@ const field_t bmapbtd_hfld[] = {
{ NULL }
};
+const field_t bmapbta_crc_hfld[] = {
+ { "", FLDT_BMAPBTA_CRC, OI(0), C1, 0, TYP_NONE },
+ { NULL }
+};
+const field_t bmapbtd_crc_hfld[] = {
+ { "", FLDT_BMAPBTD_CRC, OI(0), C1, 0, TYP_NONE },
+ { NULL }
+};
+
#define OFF(f) bitize(offsetof(struct xfs_btree_block, bb_ ## f))
const field_t bmapbta_flds[] = {
{ "magic", FLDT_UINT32X, OI(OFF(magic)), C1, 0, TYP_NONE },
@@ -237,6 +271,45 @@ const field_t bmapbtd_flds[] = {
FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_BMAPBTD },
{ NULL }
};
+/* crc enabled versions */
+const field_t bmapbta_crc_flds[] = {
+ { "magic", FLDT_UINT32X, OI(OFF(magic)), C1, 0, TYP_NONE },
+ { "level", FLDT_UINT16D, OI(OFF(level)), C1, 0, TYP_NONE },
+ { "numrecs", FLDT_UINT16D, OI(OFF(numrecs)), C1, 0, TYP_NONE },
+ { "leftsib", FLDT_DFSBNO, OI(OFF(u.l.bb_leftsib)), C1, 0, TYP_BMAPBTA },
+ { "rightsib", FLDT_DFSBNO, OI(OFF(u.l.bb_rightsib)), C1, 0, TYP_BMAPBTA },
+ { "bno", FLDT_DFSBNO, OI(OFF(u.l.bb_blkno)), C1, 0, TYP_BMAPBTD },
+ { "lsn", FLDT_UINT64X, OI(OFF(u.l.bb_lsn)), C1, 0, TYP_NONE },
+ { "uuid", FLDT_UUID, OI(OFF(u.l.bb_uuid)), C1, 0, TYP_NONE },
+ { "owner", FLDT_INO, OI(OFF(u.l.bb_owner)), C1, 0, TYP_NONE },
+ { "crc", FLDT_UINT32X, OI(OFF(u.l.bb_crc)), C1, 0, TYP_NONE },
+ { "recs", FLDT_BMAPBTAREC, btblock_rec_offset, btblock_rec_count,
+ FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_NONE },
+ { "keys", FLDT_BMAPBTAKEY, btblock_key_offset, btblock_key_count,
+ FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_NONE },
+ { "ptrs", FLDT_BMAPBTAPTR, btblock_ptr_offset, btblock_key_count,
+ FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_BMAPBTA },
+ { NULL }
+};
+const field_t bmapbtd_crc_flds[] = {
+ { "magic", FLDT_UINT32X, OI(OFF(magic)), C1, 0, TYP_NONE },
+ { "level", FLDT_UINT16D, OI(OFF(level)), C1, 0, TYP_NONE },
+ { "numrecs", FLDT_UINT16D, OI(OFF(numrecs)), C1, 0, TYP_NONE },
+ { "leftsib", FLDT_DFSBNO, OI(OFF(u.l.bb_leftsib)), C1, 0, TYP_BMAPBTD },
+ { "rightsib", FLDT_DFSBNO, OI(OFF(u.l.bb_rightsib)), C1, 0, TYP_BMAPBTD },
+ { "bno", FLDT_DFSBNO, OI(OFF(u.l.bb_blkno)), C1, 0, TYP_BMAPBTD },
+ { "lsn", FLDT_UINT64X, OI(OFF(u.l.bb_lsn)), C1, 0, TYP_NONE },
+ { "uuid", FLDT_UUID, OI(OFF(u.l.bb_uuid)), C1, 0, TYP_NONE },
+ { "owner", FLDT_INO, OI(OFF(u.l.bb_owner)), C1, 0, TYP_NONE },
+ { "crc", FLDT_UINT32X, OI(OFF(u.l.bb_crc)), C1, 0, TYP_NONE },
+ { "recs", FLDT_BMAPBTDREC, btblock_rec_offset, btblock_rec_count,
+ FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_NONE },
+ { "keys", FLDT_BMAPBTDKEY, btblock_key_offset, btblock_key_count,
+ FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_NONE },
+ { "ptrs", FLDT_BMAPBTDPTR, btblock_ptr_offset, btblock_key_count,
+ FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_BMAPBTD },
+ { NULL }
+};
#undef OFF
#define KOFF(f) bitize(offsetof(xfs_bmbt_key_t, br_ ## f))
@@ -289,6 +362,11 @@ const field_t inobt_hfld[] = {
{ NULL }
};
+const field_t inobt_crc_hfld[] = {
+ { "", FLDT_INOBT_CRC, OI(0), C1, 0, TYP_NONE },
+ { NULL }
+};
+
#define OFF(f) bitize(offsetof(struct xfs_btree_block, bb_ ## f))
const field_t inobt_flds[] = {
{ "magic", FLDT_UINT32X, OI(OFF(magic)), C1, 0, TYP_NONE },
@@ -304,6 +382,25 @@ const field_t inobt_flds[] = {
FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_INOBT },
{ NULL }
};
+const field_t inobt_crc_flds[] = {
+ { "magic", FLDT_UINT32X, OI(OFF(magic)), C1, 0, TYP_NONE },
+ { "level", FLDT_UINT16D, OI(OFF(level)), C1, 0, TYP_NONE },
+ { "numrecs", FLDT_UINT16D, OI(OFF(numrecs)), C1, 0, TYP_NONE },
+ { "leftsib", FLDT_AGBLOCK, OI(OFF(u.s.bb_leftsib)), C1, 0, TYP_INOBT },
+ { "rightsib", FLDT_AGBLOCK, OI(OFF(u.s.bb_rightsib)), C1, 0, TYP_INOBT },
+ { "bno", FLDT_DFSBNO, OI(OFF(u.s.bb_blkno)), C1, 0, TYP_INOBT },
+ { "lsn", FLDT_UINT64X, OI(OFF(u.s.bb_lsn)), C1, 0, TYP_NONE },
+ { "uuid", FLDT_UUID, OI(OFF(u.s.bb_uuid)), C1, 0, TYP_NONE },
+ { "owner", FLDT_AGNUMBER, OI(OFF(u.s.bb_owner)), C1, 0, TYP_NONE },
+ { "crc", FLDT_UINT32X, OI(OFF(u.s.bb_crc)), C1, 0, TYP_NONE },
+ { "recs", FLDT_INOBTREC, btblock_rec_offset, btblock_rec_count,
+ FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_NONE },
+ { "keys", FLDT_INOBTKEY, btblock_key_offset, btblock_key_count,
+ FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_NONE },
+ { "ptrs", FLDT_INOBTPTR, btblock_ptr_offset, btblock_key_count,
+ FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_INOBT },
+ { NULL }
+};
#undef OFF
#define KOFF(f) bitize(offsetof(xfs_inobt_key_t, ir_ ## f))
@@ -331,6 +428,11 @@ const field_t bnobt_hfld[] = {
{ NULL }
};
+const field_t bnobt_crc_hfld[] = {
+ { "", FLDT_BNOBT_CRC, OI(0), C1, 0, TYP_NONE },
+ { NULL }
+};
+
#define OFF(f) bitize(offsetof(struct xfs_btree_block, bb_ ## f))
const field_t bnobt_flds[] = {
{ "magic", FLDT_UINT32X, OI(OFF(magic)), C1, 0, TYP_NONE },
@@ -346,6 +448,25 @@ const field_t bnobt_flds[] = {
FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_BNOBT },
{ NULL }
};
+const field_t bnobt_crc_flds[] = {
+ { "magic", FLDT_UINT32X, OI(OFF(magic)), C1, 0, TYP_NONE },
+ { "level", FLDT_UINT16D, OI(OFF(level)), C1, 0, TYP_NONE },
+ { "numrecs", FLDT_UINT16D, OI(OFF(numrecs)), C1, 0, TYP_NONE },
+ { "leftsib", FLDT_AGBLOCK, OI(OFF(u.s.bb_leftsib)), C1, 0, TYP_BNOBT },
+ { "rightsib", FLDT_AGBLOCK, OI(OFF(u.s.bb_rightsib)), C1, 0, TYP_BNOBT },
+ { "bno", FLDT_DFSBNO, OI(OFF(u.s.bb_blkno)), C1, 0, TYP_BNOBT },
+ { "lsn", FLDT_UINT64X, OI(OFF(u.s.bb_lsn)), C1, 0, TYP_NONE },
+ { "uuid", FLDT_UUID, OI(OFF(u.s.bb_uuid)), C1, 0, TYP_NONE },
+ { "owner", FLDT_AGNUMBER, OI(OFF(u.s.bb_owner)), C1, 0, TYP_NONE },
+ { "crc", FLDT_UINT32X, OI(OFF(u.s.bb_crc)), C1, 0, TYP_NONE },
+ { "recs", FLDT_BNOBTREC, btblock_rec_offset, btblock_rec_count,
+ FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_NONE },
+ { "keys", FLDT_BNOBTKEY, btblock_key_offset, btblock_key_count,
+ FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_NONE },
+ { "ptrs", FLDT_BNOBTPTR, btblock_ptr_offset, btblock_key_count,
+ FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_BNOBT },
+ { NULL }
+};
#undef OFF
#define KOFF(f) bitize(offsetof(xfs_alloc_key_t, ar_ ## f))
@@ -369,6 +490,11 @@ const field_t cntbt_hfld[] = {
{ NULL }
};
+const field_t cntbt_crc_hfld[] = {
+ { "", FLDT_CNTBT_CRC, OI(0), C1, 0, TYP_NONE },
+ { NULL }
+};
+
#define OFF(f) bitize(offsetof(struct xfs_btree_block, bb_ ## f))
const field_t cntbt_flds[] = {
{ "magic", FLDT_UINT32X, OI(OFF(magic)), C1, 0, TYP_NONE },
@@ -384,6 +510,25 @@ const field_t cntbt_flds[] = {
FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_CNTBT },
{ NULL }
};
+const field_t cntbt_crc_flds[] = {
+ { "magic", FLDT_UINT32X, OI(OFF(magic)), C1, 0, TYP_NONE },
+ { "level", FLDT_UINT16D, OI(OFF(level)), C1, 0, TYP_NONE },
+ { "numrecs", FLDT_UINT16D, OI(OFF(numrecs)), C1, 0, TYP_NONE },
+ { "leftsib", FLDT_AGBLOCK, OI(OFF(u.s.bb_leftsib)), C1, 0, TYP_CNTBT },
+ { "rightsib", FLDT_AGBLOCK, OI(OFF(u.s.bb_rightsib)), C1, 0, TYP_CNTBT },
+ { "bno", FLDT_DFSBNO, OI(OFF(u.s.bb_blkno)), C1, 0, TYP_CNTBT },
+ { "lsn", FLDT_UINT64X, OI(OFF(u.s.bb_lsn)), C1, 0, TYP_NONE },
+ { "uuid", FLDT_UUID, OI(OFF(u.s.bb_uuid)), C1, 0, TYP_NONE },
+ { "owner", FLDT_AGNUMBER, OI(OFF(u.s.bb_owner)), C1, 0, TYP_NONE },
+ { "crc", FLDT_UINT32X, OI(OFF(u.s.bb_crc)), C1, 0, TYP_NONE },
+ { "recs", FLDT_CNTBTREC, btblock_rec_offset, btblock_rec_count,
+ FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_NONE },
+ { "keys", FLDT_CNTBTKEY, btblock_key_offset, btblock_key_count,
+ FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_NONE },
+ { "ptrs", FLDT_CNTBTPTR, btblock_ptr_offset, btblock_key_count,
+ FLD_ARRAY|FLD_ABASE1|FLD_COUNT|FLD_OFFSET, TYP_CNTBT },
+ { NULL }
+};
#undef OFF
#define KOFF(f) bitize(offsetof(xfs_alloc_key_t, ar_ ## f))
diff --git a/db/btblock.h b/db/btblock.h
index 0631e66..daee060 100644
--- a/db/btblock.h
+++ b/db/btblock.h
@@ -18,26 +18,36 @@
extern const struct field bmapbta_flds[];
extern const struct field bmapbta_hfld[];
+extern const struct field bmapbta_crc_flds[];
+extern const struct field bmapbta_crc_hfld[];
extern const struct field bmapbta_key_flds[];
extern const struct field bmapbta_rec_flds[];
extern const struct field bmapbtd_flds[];
extern const struct field bmapbtd_hfld[];
+extern const struct field bmapbtd_crc_flds[];
+extern const struct field bmapbtd_crc_hfld[];
extern const struct field bmapbtd_key_flds[];
extern const struct field bmapbtd_rec_flds[];
extern const struct field inobt_flds[];
extern const struct field inobt_hfld[];
+extern const struct field inobt_crc_flds[];
+extern const struct field inobt_crc_hfld[];
extern const struct field inobt_key_flds[];
extern const struct field inobt_rec_flds[];
extern const struct field bnobt_flds[];
extern const struct field bnobt_hfld[];
+extern const struct field bnobt_crc_flds[];
+extern const struct field bnobt_crc_hfld[];
extern const struct field bnobt_key_flds[];
extern const struct field bnobt_rec_flds[];
extern const struct field cntbt_flds[];
extern const struct field cntbt_hfld[];
+extern const struct field cntbt_crc_flds[];
+extern const struct field cntbt_crc_hfld[];
extern const struct field cntbt_key_flds[];
extern const struct field cntbt_rec_flds[];
diff --git a/db/check.c b/db/check.c
index efbd0c9..41c150b 100644
--- a/db/check.c
+++ b/db/check.c
@@ -800,6 +800,19 @@ blockget_f(
dbprintf(_("already have block usage information\n"));
return 0;
}
+
+ /*
+ * XXX: check does not support CRC enabled filesystems. Return success
+ * silently without doing anything here initially so that xfstests can
+ * run without modification on metadata enabled filesystems.
+ *
+ * XXX: ultimately we need to dump an error message here that xfstests
+ * filters out, or we need to actually do the work to make check support
+ * crc enabled filesystems.
+ */
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ return 0;
+
if (!init(argc, argv)) {
if (serious_error)
exitcode = 3;
@@ -1774,6 +1787,7 @@ init(
serious_error = 1;
return 0;
}
+
if (!sb_logcheck())
return 0;
rt = mp->m_sb.sb_rextents != 0;
@@ -4096,6 +4110,7 @@ scan_freelist(
xfs_agblock_t bno;
uint count;
int i;
+ __be32 *freelist;
if (XFS_SB_BLOCK(mp) != XFS_AGFL_BLOCK(mp) &&
XFS_AGF_BLOCK(mp) != XFS_AGFL_BLOCK(mp) &&
@@ -4125,9 +4140,12 @@ scan_freelist(
return;
}
+ /* open coded XFS_BUF_TO_AGFL_BNO */
+ freelist = xfs_sb_version_hascrc(&((mp)->m_sb)) ? &agfl->agfl_bno[0]
+ : (__be32 *)agfl;
count = 0;
for (;;) {
- bno = be32_to_cpu(agfl->agfl_bno[i]);
+ bno = be32_to_cpu(freelist[i]);
set_dbmap(seqno, bno, 1, DBM_FREELIST, seqno,
XFS_AGFL_BLOCK(mp));
count++;
diff --git a/db/field.c b/db/field.c
index 6903898..032b74e 100644
--- a/db/field.c
+++ b/db/field.c
@@ -48,6 +48,8 @@ const ftattr_t ftattrtab[] = {
agf_flds },
{ FLDT_AGFL, "agfl", NULL, (char *)agfl_flds, agfl_size, FTARG_SIZE,
NULL, agfl_flds },
+ { FLDT_AGFL_CRC, "agfl", NULL, (char *)agfl_crc_flds, agfl_size,
+ FTARG_SIZE, NULL, agfl_crc_flds },
{ FLDT_AGI, "agi", NULL, (char *)agi_flds, agi_size, FTARG_SIZE, NULL,
agi_flds },
{ FLDT_AGINO, "agino", fp_num, "%u", SI(bitsz(xfs_agino_t)),
@@ -86,6 +88,8 @@ const ftattr_t ftattrtab[] = {
attrshort_size, FTARG_SIZE, NULL, attr_shortform_flds },
{ FLDT_BMAPBTA, "bmapbta", NULL, (char *)bmapbta_flds, btblock_size,
FTARG_SIZE, NULL, bmapbta_flds },
+ { FLDT_BMAPBTA_CRC, "bmapbta", NULL, (char *)bmapbta_crc_flds,
+ btblock_size, FTARG_SIZE, NULL, bmapbta_crc_flds },
{ FLDT_BMAPBTAKEY, "bmapbtakey", fp_sarray, (char *)bmapbta_key_flds,
SI(bitsz(xfs_bmbt_key_t)), 0, NULL, bmapbta_key_flds },
{ FLDT_BMAPBTAPTR, "bmapbtaptr", fp_num, "%llu",
@@ -94,6 +98,8 @@ const ftattr_t ftattrtab[] = {
SI(bitsz(xfs_bmbt_rec_t)), 0, NULL, bmapbta_rec_flds },
{ FLDT_BMAPBTD, "bmapbtd", NULL, (char *)bmapbtd_flds, btblock_size,
FTARG_SIZE, NULL, bmapbtd_flds },
+ { FLDT_BMAPBTD_CRC, "bmapbtd", NULL, (char *)bmapbtd_crc_flds,
+ btblock_size, FTARG_SIZE, NULL, bmapbtd_crc_flds },
{ FLDT_BMAPBTDKEY, "bmapbtdkey", fp_sarray, (char *)bmapbtd_key_flds,
SI(bitsz(xfs_bmbt_key_t)), 0, NULL, bmapbtd_key_flds },
{ FLDT_BMAPBTDPTR, "bmapbtdptr", fp_num, "%llu",
@@ -114,6 +120,8 @@ const ftattr_t ftattrtab[] = {
SI(bitsz(xfs_bmdr_ptr_t)), 0, fa_dfsbno, NULL },
{ FLDT_BNOBT, "bnobt", NULL, (char *)bnobt_flds, btblock_size, FTARG_SIZE,
NULL, bnobt_flds },
+ { FLDT_BNOBT_CRC, "bnobt", NULL, (char *)bnobt_crc_flds, btblock_size,
+ FTARG_SIZE, NULL, bnobt_crc_flds },
{ FLDT_BNOBTKEY, "bnobtkey", fp_sarray, (char *)bnobt_key_flds,
SI(bitsz(xfs_alloc_key_t)), 0, NULL, bnobt_key_flds },
{ FLDT_BNOBTPTR, "bnobtptr", fp_num, "%u", SI(bitsz(xfs_alloc_ptr_t)),
@@ -135,6 +143,8 @@ const ftattr_t ftattrtab[] = {
{ FLDT_CHARS, "chars", fp_num, "%c", SI(bitsz(char)), 0, NULL, NULL },
{ FLDT_CNTBT, "cntbt", NULL, (char *)cntbt_flds, btblock_size, FTARG_SIZE,
NULL, cntbt_flds },
+ { FLDT_CNTBT_CRC, "cntbt", NULL, (char *)cntbt_crc_flds, btblock_size,
+ FTARG_SIZE, NULL, cntbt_crc_flds },
{ FLDT_CNTBTKEY, "cntbtkey", fp_sarray, (char *)cntbt_key_flds,
SI(bitsz(xfs_alloc_key_t)), 0, NULL, cntbt_key_flds },
{ FLDT_CNTBTPTR, "cntbtptr", fp_num, "%u", SI(bitsz(xfs_alloc_ptr_t)),
@@ -156,6 +166,8 @@ const ftattr_t ftattrtab[] = {
SI(bitsz(__int8_t)), 0, NULL, NULL },
{ FLDT_DINODE_U, "dinode_u", NULL, (char *)inode_u_flds, inode_u_size,
FTARG_SIZE|FTARG_OKEMPTY, NULL, inode_u_flds },
+ { FLDT_DINODE_V3, "dinode_v3", NULL, (char *)inode_v3_flds,
+ SI(bitsz(xfs_dinode_t)), 0, NULL, inode_v3_flds },
{ FLDT_DIR, "dir", NULL, (char *)dir_flds, dir_size, FTARG_SIZE, NULL,
dir_flds },
{ FLDT_DIR2, "dir2", NULL, (char *)dir2_flds, dir2_size, FTARG_SIZE,
@@ -246,6 +258,8 @@ const ftattr_t ftattrtab[] = {
fa_ino, NULL },
{ FLDT_INOBT, "inobt", NULL, (char *)inobt_flds, btblock_size,
FTARG_SIZE, NULL, inobt_flds },
+ { FLDT_INOBT_CRC, "inobt", NULL, (char *)inobt_crc_flds, btblock_size,
+ FTARG_SIZE, NULL, inobt_crc_flds },
{ FLDT_INOBTKEY, "inobtkey", fp_sarray, (char *)inobt_key_flds,
SI(bitsz(xfs_inobt_key_t)), 0, NULL, inobt_key_flds },
{ FLDT_INOBTPTR, "inobtptr", fp_num, "%u", SI(bitsz(xfs_inobt_ptr_t)),
@@ -254,6 +268,8 @@ const ftattr_t ftattrtab[] = {
SI(bitsz(xfs_inobt_rec_t)), 0, NULL, inobt_rec_flds },
{ FLDT_INODE, "inode", NULL, (char *)inode_flds, inode_size, FTARG_SIZE,
NULL, inode_flds },
+ { FLDT_INODE_CRC, "inode", NULL, (char *)inode_crc_flds, inode_size,
+ FTARG_SIZE, NULL, inode_crc_flds },
{ FLDT_INOFREE, "inofree", fp_num, "%#llx", SI(bitsz(xfs_inofree_t)), 0,
NULL, NULL },
{ FLDT_INT16D, "int16d", fp_num, "%d", SI(bitsz(__int16_t)),
diff --git a/db/field.h b/db/field.h
index 6962d69..f503c42 100644
--- a/db/field.h
+++ b/db/field.h
@@ -22,6 +22,7 @@ typedef enum fldt {
FLDT_AGBLOCKNZ,
FLDT_AGF,
FLDT_AGFL,
+ FLDT_AGFL_CRC,
FLDT_AGI,
FLDT_AGINO,
FLDT_AGINONN,
@@ -39,10 +40,12 @@ typedef enum fldt {
FLDT_ATTRBLOCK,
FLDT_ATTRSHORT,
FLDT_BMAPBTA,
+ FLDT_BMAPBTA_CRC,
FLDT_BMAPBTAKEY,
FLDT_BMAPBTAPTR,
FLDT_BMAPBTAREC,
FLDT_BMAPBTD,
+ FLDT_BMAPBTD_CRC,
FLDT_BMAPBTDKEY,
FLDT_BMAPBTDPTR,
FLDT_BMAPBTDREC,
@@ -53,6 +56,7 @@ typedef enum fldt {
FLDT_BMROOTDKEY,
FLDT_BMROOTDPTR,
FLDT_BNOBT,
+ FLDT_BNOBT_CRC,
FLDT_BNOBTKEY,
FLDT_BNOBTPTR,
FLDT_BNOBTREC,
@@ -64,6 +68,7 @@ typedef enum fldt {
FLDT_CHARNS,
FLDT_CHARS,
FLDT_CNTBT,
+ FLDT_CNTBT_CRC,
FLDT_CNTBTKEY,
FLDT_CNTBTPTR,
FLDT_CNTBTREC,
@@ -75,6 +80,7 @@ typedef enum fldt {
FLDT_DINODE_CORE,
FLDT_DINODE_FMT,
FLDT_DINODE_U,
+ FLDT_DINODE_V3,
FLDT_DIR,
FLDT_DIR2,
FLDT_DIR2_BLOCK_TAIL,
@@ -116,10 +122,12 @@ typedef enum fldt {
FLDT_FSIZE,
FLDT_INO,
FLDT_INOBT,
+ FLDT_INOBT_CRC,
FLDT_INOBTKEY,
FLDT_INOBTPTR,
FLDT_INOBTREC,
FLDT_INODE,
+ FLDT_INODE_CRC,
FLDT_INOFREE,
FLDT_INT16D,
FLDT_INT32D,
diff --git a/db/freesp.c b/db/freesp.c
index 472b1f7..9813137 100644
--- a/db/freesp.c
+++ b/db/freesp.c
@@ -231,6 +231,7 @@ scan_freelist(
xfs_agfl_t *agfl;
xfs_agblock_t bno;
int i;
+ __be32 *agfl_bno;
if (be32_to_cpu(agf->agf_flcount) == 0)
return;
@@ -249,8 +250,10 @@ scan_freelist(
return;
}
+ agfl_bno = xfs_sb_version_hascrc(&mp->m_sb) ? &agfl->agfl_bno[0]
+ : iocur_top->data;
for (;;) {
- bno = be32_to_cpu(agfl->agfl_bno[i]);
+ bno = be32_to_cpu(agfl_bno[i]);
addtohist(seqno, bno, 1);
if (i == be32_to_cpu(agf->agf_fllast))
break;
@@ -296,7 +299,8 @@ scanfunc_bno(
xfs_alloc_ptr_t *pp;
xfs_alloc_rec_t *rp;
- if (be32_to_cpu(block->bb_magic) != XFS_ABTB_MAGIC)
+ if (!(be32_to_cpu(block->bb_magic) == XFS_ABTB_MAGIC ||
+ be32_to_cpu(block->bb_magic) == XFS_ABTB_CRC_MAGIC))
return;
if (level == 0) {
@@ -323,7 +327,8 @@ scanfunc_cnt(
xfs_alloc_ptr_t *pp;
xfs_alloc_rec_t *rp;
- if (be32_to_cpu(block->bb_magic) != XFS_ABTC_MAGIC)
+ if (!(be32_to_cpu(block->bb_magic) == XFS_ABTC_MAGIC ||
+ be32_to_cpu(block->bb_magic) == XFS_ABTC_CRC_MAGIC))
return;
if (level == 0) {
diff --git a/db/init.c b/db/init.c
index 2a31cb8..141b484 100644
--- a/db/init.c
+++ b/db/init.c
@@ -26,6 +26,7 @@
#include "sig.h"
#include "output.h"
#include "malloc.h"
+#include "type.h"
static char **cmdline;
static int ncmdline;
@@ -145,6 +146,9 @@ init(
}
blkbb = 1 << mp->m_blkbb_log;
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ type_set_tab_crc();
+
push_cur();
init_commands();
init_sig();
diff --git a/db/inode.c b/db/inode.c
index 036717f..6596e4f 100644
--- a/db/inode.c
+++ b/db/inode.c
@@ -58,6 +58,10 @@ const field_t inode_hfld[] = {
{ "", FLDT_INODE, OI(0), C1, 0, TYP_NONE },
{ NULL }
};
+const field_t inode_crc_hfld[] = {
+ { "", FLDT_INODE_CRC, OI(0), C1, 0, TYP_NONE },
+ { NULL }
+};
/* XXX: fix this up! */
#define OFF(f) bitize(offsetof(xfs_dinode_t, di_ ## f))
@@ -70,6 +74,17 @@ const field_t inode_flds[] = {
FLD_COUNT|FLD_OFFSET, TYP_NONE },
{ NULL }
};
+const field_t inode_crc_flds[] = {
+ { "core", FLDT_DINODE_CORE, OI(OFF(magic)), C1, 0, TYP_NONE },
+ { "next_unlinked", FLDT_AGINO, OI(OFF(next_unlinked)), C1, 0,
+ TYP_INODE },
+ { "v3", FLDT_DINODE_V3, OI(OFF(magic)), C1, 0, TYP_NONE },
+ { "u", FLDT_DINODE_U, inode_u_offset, C1, FLD_OFFSET, TYP_NONE },
+ { "a", FLDT_DINODE_A, inode_a_offset, inode_a_count,
+ FLD_COUNT|FLD_OFFSET, TYP_NONE },
+ { NULL }
+};
+
#define COFF(f) bitize(offsetof(xfs_dinode_t, di_ ## f))
const field_t inode_core_flds[] = {
@@ -152,6 +167,18 @@ const field_t inode_core_flds[] = {
{ NULL }
};
+const field_t inode_v3_flds[] = {
+ { "crc", FLDT_UINT32X, OI(COFF(crc)), C1, 0, TYP_NONE },
+ { "change_count", FLDT_UINT64D, OI(COFF(changecount)), C1, 0, TYP_NONE },
+ { "lsn", FLDT_UINT64X, OI(COFF(lsn)), C1, 0, TYP_NONE },
+ { "flags2", FLDT_UINT64X, OI(COFF(flags2)), C1, 0, TYP_NONE },
+ { "crtime", FLDT_TIMESTAMP, OI(COFF(crtime)), C1, 0, TYP_NONE },
+ { "inumber", FLDT_INO, OI(COFF(ino)), C1, 0, TYP_NONE },
+ { "uuid", FLDT_UUID, OI(COFF(uuid)), C1, 0, TYP_NONE },
+ { NULL }
+};
+
+
#define TOFF(f) bitize(offsetof(xfs_timestamp_t, t_ ## f))
const field_t timestamp_flds[] = {
{ "sec", FLDT_TIME, OI(TOFF(sec)), C1, 0, TYP_NONE },
diff --git a/db/inode.h b/db/inode.h
index 6c1ac5c..1624f1d 100644
--- a/db/inode.h
+++ b/db/inode.h
@@ -18,8 +18,11 @@
extern const struct field inode_a_flds[];
extern const struct field inode_core_flds[];
+extern const struct field inode_v3_flds[];
extern const struct field inode_flds[];
+extern const struct field inode_crc_flds[];
extern const struct field inode_hfld[];
+extern const struct field inode_crc_hfld[];
extern const struct field inode_u_flds[];
extern const struct field timestamp_flds[];
diff --git a/db/sb.c b/db/sb.c
index af86722..c52d545 100644
--- a/db/sb.c
+++ b/db/sb.c
@@ -109,6 +109,14 @@ const field_t sb_flds[] = {
{ "logsunit", FLDT_UINT32D, OI(OFF(logsunit)), C1, 0, TYP_NONE },
{ "features2", FLDT_UINT32X, OI(OFF(features2)), C1, 0, TYP_NONE },
{ "bad_features2", FLDT_UINT32X, OI(OFF(bad_features2)), C1, 0, TYP_NONE },
+
+ { "features_compat", FLDT_UINT32X, OI(OFF(features_compat)), C1, 0, TYP_NONE },
+ { "features_ro_compat", FLDT_UINT32X, OI(OFF(features_ro_compat)), C1, 0, TYP_NONE },
+ { "features_incompat", FLDT_UINT32X, OI(OFF(features_incompat)), C1, 0, TYP_NONE },
+
+ { "crc", FLDT_UINT32X, OI(OFF(crc)), C1, 0, TYP_NONE },
+ { "pquotino", FLDT_INO, OI(OFF(pquotino)), C1, 0, TYP_INODE },
+ { "lsn", FLDT_UINT64X, OI(OFF(lsn)), C1, 0, TYP_NONE },
{ NULL }
};
@@ -240,7 +248,7 @@ sb_logcheck(void)
log.l_logsize = BBTOB(log.l_logBBsize);
log.l_logBBsize = x.logBBsize;
log.l_logBBstart = x.logBBstart;
- log.l_sectBBsize = x.lbsize;
+ log.l_sectBBsize = 1 << mp->m_sb.sb_logsectlog;
log.l_mp = mp;
if (xlog_find_tail(&log, &head_blk, &tail_blk)) {
@@ -595,6 +603,8 @@ version_string(
strcpy(s, "V3");
else if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4)
strcpy(s, "V4");
+ else if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5)
+ strcpy(s, "V5");
if (xfs_sb_version_hasattr(sbp))
strcat(s, ",ATTR");
@@ -626,6 +636,8 @@ version_string(
strcat(s, ",LAZYSBCOUNT");
if (xfs_sb_version_hasprojid32bit(sbp))
strcat(s, ",PROJID32BIT");
+ if (xfs_sb_version_hascrc(sbp))
+ strcat(s, ",CRC");
return s;
}
@@ -666,6 +678,10 @@ version_f(
version = mp->m_sb.sb_versionnum |
XFS_SB_VERSION_EXTFLGBIT;
break;
+ case XFS_SB_VERSION_5:
+ dbprintf(
+ _("unwritten extents always enabled for v5 superblocks.\n"));
+ break;
}
} else if (!strcasecmp(argv[1], "log2")) {
switch (XFS_SB_VERSION_NUM(&mp->m_sb)) {
@@ -686,8 +702,18 @@ version_f(
version = mp->m_sb.sb_versionnum |
XFS_SB_VERSION_LOGV2BIT;
break;
+ case XFS_SB_VERSION_5:
+ dbprintf(
+ _("Version 2 logs always enabled for v5 superblocks.\n"));
+ break;
}
+ } else if (XFS_SB_VERSION_NUM(&mp->m_sb) == XFS_SB_VERSION_5) {
+ dbprintf(
+ _("%s: Cannot change %s on v5 superblocks.\n"),
+ progname, argv[1]);
+ return 0;
} else if (!strcasecmp(argv[1], "attr1")) {
+
if (xfs_sb_version_hasattr2(&mp->m_sb)) {
if (!(mp->m_sb.sb_features2 &=
~XFS_SB_VERSION2_ATTR2BIT))
diff --git a/db/type.c b/db/type.c
index e5fd9eb..ad67543 100644
--- a/db/type.c
+++ b/db/type.c
@@ -50,7 +50,7 @@ static const cmdinfo_t type_cmd =
{ "type", NULL, type_f, 0, 1, 1, N_("[newtype]"),
N_("set/show current data type"), NULL };
-const typ_t typtab[] = {
+static const typ_t __typtab[] = {
{ TYP_AGF, "agf", handle_struct, agf_hfld },
{ TYP_AGFL, "agfl", handle_struct, agfl_hfld },
{ TYP_AGI, "agi", handle_struct, agi_hfld },
@@ -75,6 +75,39 @@ const typ_t typtab[] = {
{ TYP_NONE, NULL }
};
+static const typ_t __typtab_crc[] = {
+ { TYP_AGF, "agf", handle_struct, agf_hfld },
+ { TYP_AGFL, "agfl", handle_struct, agfl_crc_hfld },
+ { TYP_AGI, "agi", handle_struct, agi_hfld },
+ { TYP_ATTR, "attr", handle_struct, attr_hfld },
+ { TYP_BMAPBTA, "bmapbta", handle_struct, bmapbta_crc_hfld },
+ { TYP_BMAPBTD, "bmapbtd", handle_struct, bmapbtd_crc_hfld },
+ { TYP_BNOBT, "bnobt", handle_struct, bnobt_crc_hfld },
+ { TYP_CNTBT, "cntbt", handle_struct, cntbt_crc_hfld },
+ { TYP_DATA, "data", handle_block, NULL },
+ { TYP_DIR, "dir", handle_struct, dir_hfld },
+ { TYP_DIR2, "dir2", handle_struct, dir2_hfld },
+ { TYP_DQBLK, "dqblk", handle_struct, dqblk_hfld },
+ { TYP_INOBT, "inobt", handle_struct, inobt_crc_hfld },
+ { TYP_INODATA, "inodata", NULL, NULL },
+ { TYP_INODE, "inode", handle_struct, inode_crc_hfld },
+ { TYP_LOG, "log", NULL, NULL },
+ { TYP_RTBITMAP, "rtbitmap", NULL, NULL },
+ { TYP_RTSUMMARY, "rtsummary", NULL, NULL },
+ { TYP_SB, "sb", handle_struct, sb_hfld },
+ { TYP_SYMLINK, "symlink", handle_string, NULL },
+ { TYP_TEXT, "text", handle_text, NULL },
+ { TYP_NONE, NULL }
+};
+
+const typ_t *typtab = __typtab;
+
+void
+type_set_tab_crc(void)
+{
+ typtab = __typtab_crc;
+}
+
static const typ_t *
findtyp(
char *name)
diff --git a/db/type.h b/db/type.h
index 8dd7925..ae92581 100644
--- a/db/type.h
+++ b/db/type.h
@@ -43,9 +43,10 @@ typedef struct typ
pfunc_t pfunc;
const struct field *fields;
} typ_t;
-extern const typ_t typtab[], *cur_typ;
+extern const typ_t *typtab, *cur_typ;
extern void type_init(void);
+extern void type_set_tab_crc(void);
extern void handle_block(int action, const struct field *fields, int argc,
char **argv);
extern void handle_string(int action, const struct field *fields, int argc,
diff --git a/include/xfs_dir2.h b/include/xfs_dir2.h
index 2dd0cc0..2ed4d82 100644
--- a/include/xfs_dir2.h
+++ b/include/xfs_dir2.h
@@ -100,4 +100,12 @@ extern void xfs_dir2_data_use_free(struct xfs_trans *tp, struct xfs_buf *bp,
extern struct xfs_dir2_data_free *xfs_dir2_data_freefind(
struct xfs_dir2_data_hdr *hdr, struct xfs_dir2_data_unused *dup);
+extern const struct xfs_buf_ops xfs_da3_node_buf_ops;
+extern const struct xfs_buf_ops xfs_dir3_block_buf_ops;
+extern const struct xfs_buf_ops xfs_dir3_leafn_buf_ops;
+extern const struct xfs_buf_ops xfs_dir3_leaf1_buf_ops;
+extern const struct xfs_buf_ops xfs_dir3_free_buf_ops;
+extern const struct xfs_buf_ops xfs_dir3_data_buf_ops;
+
+
#endif /* __XFS_DIR2_H__ */
diff --git a/include/xfs_inode.h b/include/xfs_inode.h
index cc14743..fdca7f7 100644
--- a/include/xfs_inode.h
+++ b/include/xfs_inode.h
@@ -169,9 +169,9 @@ typedef struct xfs_icdinode {
/* structure must be padded to 64 bit alignment */
} xfs_icdinode_t;
-static inline uint xfs_icdinode_size(struct xfs_icdinode *dicp)
+static inline uint xfs_icdinode_size(int version)
{
- if (dicp->di_version == 3)
+ if (version == 3)
return sizeof(struct xfs_icdinode);
return offsetof(struct xfs_icdinode, di_next_unlinked);
}
diff --git a/libxfs/init.c b/libxfs/init.c
index 55a17d2..372dd6b 100644
--- a/libxfs/init.c
+++ b/libxfs/init.c
@@ -746,7 +746,8 @@ libxfs_mount(
}
if (bp)
libxfs_putbuf(bp);
- }
+ } else
+ mp->m_logdev.dev = mp->m_dev.dev;
/* Initialize realtime fields in the mount structure */
if (rtmount_init(mp, flags)) {
diff --git a/logprint/log_misc.c b/logprint/log_misc.c
index ffc6916..a9c3710 100644
--- a/logprint/log_misc.c
+++ b/logprint/log_misc.c
@@ -653,7 +653,7 @@ xlog_print_trans_inode(xfs_caddr_t *ptr,
mode = dino.di_mode & S_IFMT;
size = (int)dino.di_size;
xlog_print_trans_inode_core(&dino);
- *ptr += xfs_icdinode_size(&dino);
+ *ptr += xfs_icdinode_size(dino.di_version);
if (*i == num_ops-1 && f->ilf_size == 3) {
return 1;
diff --git a/logprint/log_print_all.c b/logprint/log_print_all.c
index 70b0905..4626186 100644
--- a/logprint/log_print_all.c
+++ b/logprint/log_print_all.c
@@ -295,8 +295,8 @@ xlog_recover_print_inode(
f->ilf_dsize);
/* core inode comes 2nd */
- ASSERT(item->ri_buf[1].i_len == xfs_icdinode_size((xfs_icdinode_t *)
- item->ri_buf[1].i_addr));
+ ASSERT(item->ri_buf[1].i_len == xfs_icdinode_size(1) ||
+ item->ri_buf[1].i_len == xfs_icdinode_size(3));
xlog_recover_print_inode_core((xfs_icdinode_t *)
item->ri_buf[1].i_addr);
diff --git a/repair/dir2.c b/repair/dir2.c
index 274273d..02c3bf3 100644
--- a/repair/dir2.c
+++ b/repair/dir2.c
@@ -83,7 +83,8 @@ static struct xfs_buf *
da_read_buf(
xfs_mount_t *mp,
int nex,
- bmap_ext_t *bmp)
+ bmap_ext_t *bmp,
+ const struct xfs_buf_ops *ops)
{
#define MAP_ARRAY_SZ 4
struct xfs_buf_map map_array[MAP_ARRAY_SZ];
@@ -105,7 +106,7 @@ da_read_buf(
map[i].bm_bn = XFS_FSB_TO_DADDR(mp, bmp[i].startblock);
map[i].bm_len = XFS_FSB_TO_BB(mp, bmp[i].blockcount);
}
- bp = libxfs_readbuf_map(mp->m_dev, map, nex, 0, NULL);
+ bp = libxfs_readbuf_map(mp->m_dev, map, nex, 0, ops);
if (map != map_array)
free(map);
return bp;
@@ -152,7 +153,7 @@ traverse_int_dir2block(xfs_mount_t *mp,
if (nex == 0)
goto error_out;
- bp = da_read_buf(mp, nex, bmp);
+ bp = da_read_buf(mp, nex, bmp, &xfs_da3_node_buf_ops);
if (bmp != &lbmp)
free(bmp);
if (bp == NULL) {
@@ -516,7 +517,7 @@ _("can't get map info for block %u of directory inode %" PRIu64 "\n"),
return(1);
}
- bp = da_read_buf(mp, nex, bmp);
+ bp = da_read_buf(mp, nex, bmp, &xfs_da3_node_buf_ops);
if (bmp != &lbmp)
free(bmp);
@@ -1561,7 +1562,7 @@ _("block %u for directory inode %" PRIu64 " is missing\n"),
mp->m_dirdatablk, ino);
return 1;
}
- bp = da_read_buf(mp, nex, bmp);
+ bp = da_read_buf(mp, nex, bmp, &xfs_dir3_block_buf_ops);
if (bmp != &lbmp)
free(bmp);
if (bp == NULL) {
@@ -1692,7 +1693,7 @@ _("can't map block %u for directory inode %" PRIu64 "\n"),
da_bno, ino);
goto error_out;
}
- bp = da_read_buf(mp, nex, bmp);
+ bp = da_read_buf(mp, nex, bmp, &xfs_dir3_leafn_buf_ops);
if (bmp != &lbmp)
free(bmp);
bmp = NULL;
@@ -1878,7 +1879,7 @@ _("block %" PRIu64 " for directory inode %" PRIu64 " is missing\n"),
dbno, ino);
continue;
}
- bp = da_read_buf(mp, nex, bmp);
+ bp = da_read_buf(mp, nex, bmp, &xfs_dir3_data_buf_ops);
if (bmp != &lbmp)
free(bmp);
if (bp == NULL) {
diff --git a/repair/phase6.c b/repair/phase6.c
index 8d4d941..48596e8 100644
--- a/repair/phase6.c
+++ b/repair/phase6.c
@@ -448,8 +448,7 @@ mk_rbmino(xfs_mount_t *mp)
}
vers = xfs_sb_version_hascrc(&mp->m_sb) ? 3 : 1;
- ip->i_d.di_version = vers;
- memset(&ip->i_d, 0, xfs_icdinode_size(&ip->i_d));
+ memset(&ip->i_d, 0, xfs_icdinode_size(vers));
ip->i_d.di_magic = XFS_DINODE_MAGIC;
ip->i_d.di_mode = S_IFREG;
@@ -698,8 +697,7 @@ mk_rsumino(xfs_mount_t *mp)
}
vers = xfs_sb_version_hascrc(&mp->m_sb) ? 3 : 1;
- ip->i_d.di_version = vers;
- memset(&ip->i_d, 0, xfs_icdinode_size(&ip->i_d));
+ memset(&ip->i_d, 0, xfs_icdinode_size(vers));
ip->i_d.di_magic = XFS_DINODE_MAGIC;
ip->i_d.di_mode = S_IFREG;
@@ -815,8 +813,7 @@ mk_root_dir(xfs_mount_t *mp)
* take care of the core -- initialization from xfs_ialloc()
*/
vers = xfs_sb_version_hascrc(&mp->m_sb) ? 3 : 1;
- ip->i_d.di_version = vers;
- memset(&ip->i_d, 0, xfs_icdinode_size(&ip->i_d));
+ memset(&ip->i_d, 0, xfs_icdinode_size(vers));
ip->i_d.di_magic = XFS_DINODE_MAGIC;
ip->i_d.di_mode = (__uint16_t) mode|S_IFDIR;
@@ -1256,7 +1253,7 @@ _("can't map block %d in %s inode %" PRIu64 ", xfs_bmapi returns %d, nmap = %d\n
*/
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, fsbno),
- XFS_FSB_TO_BB(mp, 1), 0, NULL);
+ XFS_FSB_TO_BB(mp, 1), 0, &xfs_da3_node_buf_ops);
if (!bp) {
do_warn(
@@ -2332,7 +2329,8 @@ longform_dir2_check_leaf(
struct xfs_dir2_leaf_entry *ents;
da_bno = mp->m_dirleafblk;
- if (libxfs_da_read_buf(NULL, ip, da_bno, -1, &bp, XFS_DATA_FORK, NULL)) {
+ if (libxfs_da_read_buf(NULL, ip, da_bno, -1, &bp, XFS_DATA_FORK,
+ &xfs_dir3_leaf1_buf_ops)) {
do_error(
_("can't read block %u for directory inode %" PRIu64 "\n"),
da_bno, ip->i_ino);
@@ -2409,7 +2407,7 @@ longform_dir2_check_node(
if (bmap_next_offset(NULL, ip, &next_da_bno, XFS_DATA_FORK))
break;
if (libxfs_da_read_buf(NULL, ip, da_bno, -1, &bp,
- XFS_DATA_FORK, NULL)) {
+ XFS_DATA_FORK, &xfs_dir3_leafn_buf_ops)) {
do_warn(
_("can't read leaf block %u for directory inode %" PRIu64 "\n"),
da_bno, ip->i_ino);
@@ -2457,7 +2455,7 @@ longform_dir2_check_node(
if (bmap_next_offset(NULL, ip, &next_da_bno, XFS_DATA_FORK))
break;
if (libxfs_da_read_buf(NULL, ip, da_bno, -1, &bp,
- XFS_DATA_FORK, NULL)) {
+ XFS_DATA_FORK, &xfs_dir3_free_buf_ops)) {
do_warn(
_("can't read freespace block %u for directory inode %" PRIu64 "\n"),
da_bno, ip->i_ino);
@@ -2579,7 +2577,7 @@ longform_dir2_entry_check(xfs_mount_t *mp,
num_bps * sizeof(struct xfs_buf*));
}
if (libxfs_da_read_buf(NULL, ip, da_bno, -1, &bplist[db],
- XFS_DATA_FORK, NULL)) {
+ XFS_DATA_FORK, &xfs_dir3_data_buf_ops)) {
do_warn(
_("can't read data block %u for directory inode %" PRIu64 "\n"),
da_bno, ino);
xfsprogs-crc-patchset-v2/xfsprogs-repair-dir-crc-support 0000600 0001750 0001750 00000052626 12132011547 022364 0 ustar dave dave xfs_repair: update for dir/attr crc format changes.
From: Dave Chinner <dchinner@xxxxxxxxxx>
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
include/xfs_attr_leaf.h | 2 +
include/xfs_dir2_format.h | 3 ++
libxfs/xfs_dir2_priv.h | 2 -
repair/attr_repair.c | 53 ++++++++++++++-----------
repair/dir2.c | 43 +++++++++++---------
repair/dir2.h | 6 +--
repair/phase6.c | 95 ++++++++++++++++++++++++++-------------------
7 files changed, 118 insertions(+), 86 deletions(-)
diff --git a/include/xfs_attr_leaf.h b/include/xfs_attr_leaf.h
index f9d7846..b3e93bb 100644
--- a/include/xfs_attr_leaf.h
+++ b/include/xfs_attr_leaf.h
@@ -332,6 +332,8 @@ int xfs_attr3_leaf_read(struct xfs_trans *tp, struct xfs_inode *dp,
struct xfs_buf **bpp);
void xfs_attr3_leaf_hdr_from_disk(struct xfs_attr3_icleaf_hdr *to,
struct xfs_attr_leafblock *from);
+void xfs_attr3_leaf_hdr_to_disk(struct xfs_attr_leafblock *to,
+ struct xfs_attr3_icleaf_hdr *from);
extern const struct xfs_buf_ops xfs_attr3_leaf_buf_ops;
diff --git a/include/xfs_dir2_format.h b/include/xfs_dir2_format.h
index 8d33f03..7137ba0 100644
--- a/include/xfs_dir2_format.h
+++ b/include/xfs_dir2_format.h
@@ -518,6 +518,9 @@ struct xfs_dir3_leaf {
#define XFS_DIR3_LEAF_CRC_OFF offsetof(struct xfs_dir3_leaf_hdr, info.crc)
+extern void xfs_dir3_leaf_hdr_from_disk(struct xfs_dir3_icleaf_hdr *to,
+ struct xfs_dir2_leaf *from);
+
static inline int
xfs_dir3_leaf_hdr_size(struct xfs_dir2_leaf *lp)
{
diff --git a/libxfs/xfs_dir2_priv.h b/libxfs/xfs_dir2_priv.h
index 7cf573c..a5f1e97 100644
--- a/libxfs/xfs_dir2_priv.h
+++ b/libxfs/xfs_dir2_priv.h
@@ -112,8 +112,6 @@ xfs_dir3_leaf_find_entry(struct xfs_dir3_icleaf_hdr *leafhdr,
int lowstale, int highstale, int *lfloglow, int *lfloghigh);
extern int xfs_dir2_node_to_leaf(struct xfs_da_state *state);
-extern void xfs_dir3_leaf_hdr_from_disk(struct xfs_dir3_icleaf_hdr *to,
- struct xfs_dir2_leaf *from);
extern void xfs_dir3_leaf_hdr_to_disk(struct xfs_dir2_leaf *to,
struct xfs_dir3_icleaf_hdr *from);
extern bool xfs_dir3_leaf_check_int(struct xfs_mount *mp,
diff --git a/repair/attr_repair.c b/repair/attr_repair.c
index 70a2041..3a48548 100644
--- a/repair/attr_repair.c
+++ b/repair/attr_repair.c
@@ -497,15 +497,17 @@ process_leaf_attr_block(
xfs_attr_leaf_entry_t *entry;
int i, start, stop, clearit, usedbs, firstb, thissize;
da_freemap_t *attr_freemap;
+ struct xfs_attr3_icleaf_hdr leafhdr;
+ xfs_attr3_leaf_hdr_from_disk(&leafhdr, leaf);
clearit = usedbs = 0;
*repair = 0;
firstb = mp->m_sb.sb_blocksize;
- stop = sizeof(xfs_attr_leaf_hdr_t);
+ stop = xfs_attr3_leaf_hdr_size(leaf);
/* does the count look sorta valid? */
- if (be16_to_cpu(leaf->hdr.count) * sizeof(xfs_attr_leaf_entry_t)
- + sizeof(xfs_attr_leaf_hdr_t) > XFS_LBSIZE(mp)) {
+ if (leafhdr.count * sizeof(xfs_attr_leaf_entry_t)
+ + xfs_attr3_leaf_hdr_size(leaf) > XFS_LBSIZE(mp)) {
do_warn(
_("bad attribute count %d in attr block %u, inode %" PRIu64 "\n"),
be16_to_cpu(leaf->hdr.count), da_bno, ino);
@@ -516,8 +518,8 @@ process_leaf_attr_block(
(void) set_da_freemap(mp, attr_freemap, 0, stop);
/* go thru each entry checking for problems */
- for (i = 0, entry = &leaf->entries[0];
- i < be16_to_cpu(leaf->hdr.count); i++, entry++) {
+ for (i = 0, entry = xfs_attr3_leaf_entryp(leaf);
+ i < leafhdr.count; i++, entry++) {
/* check if index is within some boundary. */
if (be16_to_cpu(entry->nameidx) > XFS_LBSIZE(mp)) {
@@ -538,7 +540,7 @@ process_leaf_attr_block(
}
/* mark the entry used */
- start = (__psint_t)&leaf->entries[i] - (__psint_t)leaf;
+ start = (__psint_t)&entry[i] - (__psint_t)leaf;
stop = start + sizeof(xfs_attr_leaf_entry_t);
if (set_da_freemap(mp, attr_freemap, start, stop)) {
do_warn(
@@ -584,40 +586,40 @@ process_leaf_attr_block(
* since the block will get compacted anyhow by the kernel.
*/
- if ((leaf->hdr.holes == 0 &&
- firstb != be16_to_cpu(leaf->hdr.firstused)) ||
- be16_to_cpu(leaf->hdr.firstused) > firstb) {
+ if ((leafhdr.holes == 0 &&
+ firstb != leafhdr.firstused) ||
+ leafhdr.firstused > firstb) {
if (!no_modify) {
do_warn(
_("- resetting first used heap value from %d to %d in "
"block %u of attribute fork of inode %" PRIu64 "\n"),
- be16_to_cpu(leaf->hdr.firstused),
+ leafhdr.firstused,
firstb, da_bno, ino);
- leaf->hdr.firstused = cpu_to_be16(firstb);
+ leafhdr.firstused = firstb;
*repair = 1;
} else {
do_warn(
_("- would reset first used value from %d to %d in "
"block %u of attribute fork of inode %" PRIu64 "\n"),
- be16_to_cpu(leaf->hdr.firstused),
+ leafhdr.firstused,
firstb, da_bno, ino);
}
}
- if (usedbs != be16_to_cpu(leaf->hdr.usedbytes)) {
+ if (usedbs != leafhdr.usedbytes) {
if (!no_modify) {
do_warn(
_("- resetting usedbytes cnt from %d to %d in "
"block %u of attribute fork of inode %" PRIu64 "\n"),
- be16_to_cpu(leaf->hdr.usedbytes),
+ leafhdr.usedbytes,
usedbs, da_bno, ino);
- leaf->hdr.usedbytes = cpu_to_be16(usedbs);
+ leafhdr.usedbytes = usedbs;
*repair = 1;
} else {
do_warn(
_("- would reset usedbytes cnt from %d to %d in "
"block %u of attribute fork of %" PRIu64 "\n"),
- be16_to_cpu(leaf->hdr.usedbytes),
+ leafhdr.usedbytes,
usedbs, da_bno, ino);
}
}
@@ -629,6 +631,8 @@ process_leaf_attr_block(
* we can add it then.
*/
}
+ if (*repair)
+ xfs_attr3_leaf_hdr_to_disk(leaf, &leafhdr);
free(attr_freemap);
return (clearit); /* and repair */
@@ -651,6 +655,7 @@ process_leaf_attr_level(xfs_mount_t *mp,
xfs_dablk_t prev_bno;
xfs_dahash_t current_hashval = 0;
xfs_dahash_t greatest_hashval;
+ struct xfs_attr3_icleaf_hdr leafhdr;
da_bno = da_cursor->level[0].bno;
ino = da_cursor->ino;
@@ -681,13 +686,15 @@ process_leaf_attr_level(xfs_mount_t *mp,
goto error_out;
}
- leaf = (xfs_attr_leafblock_t *)XFS_BUF_PTR(bp);
+ leaf = bp->b_addr;
+ xfs_attr3_leaf_hdr_from_disk(&leafhdr, leaf);
/* check magic number for leaf directory btree block */
- if (be16_to_cpu(leaf->hdr.info.magic) != XFS_ATTR_LEAF_MAGIC) {
+ if (!(leafhdr.magic == XFS_ATTR_LEAF_MAGIC ||
+ leafhdr.magic == XFS_ATTR3_LEAF_MAGIC)) {
do_warn(
_("bad attribute leaf magic %#x for inode %" PRIu64 "\n"),
- leaf->hdr.info.magic, ino);
+ leafhdr.magic, ino);
libxfs_putbuf(bp);
goto error_out;
}
@@ -712,10 +719,10 @@ process_leaf_attr_level(xfs_mount_t *mp,
da_cursor->level[0].hashval = greatest_hashval;
da_cursor->level[0].bp = bp;
da_cursor->level[0].bno = da_bno;
- da_cursor->level[0].index = be16_to_cpu(leaf->hdr.count);
+ da_cursor->level[0].index = leafhdr.count;
da_cursor->level[0].dirty = repair;
- if (be32_to_cpu(leaf->hdr.info.back) != prev_bno) {
+ if (leafhdr.back != prev_bno) {
do_warn(
_("bad sibling back pointer for block %u in attribute fork for inode %" PRIu64 "\n"),
da_bno, ino);
@@ -724,7 +731,7 @@ process_leaf_attr_level(xfs_mount_t *mp,
}
prev_bno = da_bno;
- da_bno = be32_to_cpu(leaf->hdr.info.forw);
+ da_bno = leafhdr.forw;
if (da_bno != 0 && verify_da_path(mp, da_cursor, 0)) {
libxfs_putbuf(bp);
@@ -891,6 +898,7 @@ process_longform_attr(
*/
switch (be16_to_cpu(leaf->hdr.info.magic)) {
case XFS_ATTR_LEAF_MAGIC: /* leaf-form attribute */
+ case XFS_ATTR3_LEAF_MAGIC:
if (process_leaf_attr_block(mp, leaf, 0, ino, blkmap,
0, &next_hashval, repair)) {
/* the block is bad. lose the attribute fork. */
@@ -901,6 +909,7 @@ process_longform_attr(
break;
case XFS_DA_NODE_MAGIC: /* btree-form attribute */
+ case XFS_DA3_NODE_MAGIC:
/* must do this now, to release block 0 before the traversal */
if (repairlinks) {
*repair = 1;
diff --git a/repair/dir2.c b/repair/dir2.c
index ec9c032..1edcd02 100644
--- a/repair/dir2.c
+++ b/repair/dir2.c
@@ -166,7 +166,8 @@ _("can't read block %u for directory inode %" PRIu64 "\n"),
node = bp->b_addr;
xfs_da3_node_hdr_from_disk(&nodehdr, node);
- if (nodehdr.magic == XFS_DIR2_LEAFN_MAGIC) {
+ if (nodehdr.magic == XFS_DIR2_LEAFN_MAGIC ||
+ nodehdr.magic == XFS_DIR3_LEAFN_MAGIC) {
if ( i != -1 ) {
do_warn(
_("found non-root LEAFN node in inode %" PRIu64 " bno = %u\n"),
@@ -175,7 +176,8 @@ _("found non-root LEAFN node in inode %" PRIu64 " bno = %u\n"),
*rbno = 0;
libxfs_putbuf(bp);
return(1);
- } else if (nodehdr.magic != XFS_DA_NODE_MAGIC) {
+ } else if (!(nodehdr.magic == XFS_DA_NODE_MAGIC ||
+ nodehdr.magic != XFS_DA3_NODE_MAGIC)) {
libxfs_putbuf(bp);
do_warn(
_("bad dir magic number 0x%x in inode %" PRIu64 " bno = %u\n"),
@@ -536,7 +538,8 @@ _("can't read block %u for directory inode %" PRIu64 "\n"),
* entry count, verify level
*/
bad = 0;
- if (XFS_DA_NODE_MAGIC != nodehdr.magic) {
+ if (!(nodehdr.magic == XFS_DA_NODE_MAGIC ||
+ nodehdr.magic == XFS_DA3_NODE_MAGIC)) {
do_warn(
_("bad magic number %x in block %u for directory inode %" PRIu64 "\n"),
nodehdr.magic,
@@ -1199,8 +1202,8 @@ process_dir2_data(
xfs_ino_t ent_ino;
d = bp->b_addr;
- bf = d->hdr.bestfree;
- ptr = (char *)d->u;
+ bf = xfs_dir3_data_bestfree_p(&d->hdr);
+ ptr = (char *)xfs_dir3_data_entry_p(&d->hdr);
badbest = lastfree = freeseen = 0;
if (be16_to_cpu(bf[0].length) == 0) {
badbest |= be16_to_cpu(bf[0].offset) != 0;
@@ -1266,7 +1269,7 @@ process_dir2_data(
do_warn(_("\twould junk block\n"));
return 1;
}
- ptr = (char *)d->u;
+ ptr = (char *)xfs_dir3_data_entry_p(&d->hdr);
/*
* Process the entries now.
*/
@@ -1575,7 +1578,8 @@ _("can't read block %u for directory inode %" PRIu64 "\n"),
* Verify the block
*/
block = bp->b_addr;
- if (be32_to_cpu(block->hdr.magic) != XFS_DIR2_BLOCK_MAGIC)
+ if (!(be32_to_cpu(block->hdr.magic) == XFS_DIR2_BLOCK_MAGIC ||
+ be32_to_cpu(block->hdr.magic) == XFS_DIR3_BLOCK_MAGIC))
do_warn(
_("bad directory block magic # %#x in block %u for directory inode %" PRIu64 "\n"),
be32_to_cpu(block->hdr.magic), mp->m_dirdatablk, ino);
@@ -1618,10 +1622,12 @@ process_leaf_block_dir2(
int i;
int stale;
struct xfs_dir2_leaf_entry *ents;
+ struct xfs_dir3_icleaf_hdr leafhdr;
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
ents = xfs_dir3_leaf_ents_p(leaf);
- for (i = stale = 0; i < be16_to_cpu(leaf->hdr.count); i++) {
+ for (i = stale = 0; i < leafhdr.count; i++) {
if ((char *)&ents[i] >= (char *)leaf + mp->m_dirblksize) {
do_warn(
_("bad entry count in block %u of directory inode %" PRIu64 "\n"),
@@ -1638,7 +1644,7 @@ _("bad hash ordering in block %u of directory inode %" PRIu64 "\n"),
}
*next_hashval = last_hashval = be32_to_cpu(ents[i].hashval);
}
- if (stale != be16_to_cpu(leaf->hdr.stale)) {
+ if (stale != leafhdr.stale) {
do_warn(
_("bad stale count in block %u of directory inode %" PRIu64 "\n"),
da_bno, ino);
@@ -1667,6 +1673,7 @@ process_leaf_level_dir2(
int nex;
xfs_dablk_t prev_bno;
bmap_ext_t lbmp;
+ struct xfs_dir3_icleaf_hdr leafhdr;
da_bno = da_cursor->level[0].bno;
ino = da_cursor->ino;
@@ -1703,15 +1710,15 @@ _("can't read file block %u for directory inode %" PRIu64 "\n"),
goto error_out;
}
leaf = bp->b_addr;
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
/*
* Check magic number for leaf directory btree block.
*/
- if (be16_to_cpu(leaf->hdr.info.magic) !=
- XFS_DIR2_LEAFN_MAGIC) {
+ if (!(leafhdr.magic == XFS_DIR2_LEAFN_MAGIC) ||
+ leafhdr.magic == XFS_DIR3_LEAFN_MAGIC) {
do_warn(
_("bad directory leaf magic # %#x for directory inode %" PRIu64 " block %u\n"),
- be16_to_cpu(leaf->hdr.info.magic),
- ino, da_bno);
+ leafhdr.magic, ino, da_bno);
libxfs_putbuf(bp);
goto error_out;
}
@@ -1733,11 +1740,10 @@ _("bad directory leaf magic # %#x for directory inode %" PRIu64 " block %u\n"),
da_cursor->level[0].hashval = greatest_hashval;
da_cursor->level[0].bp = bp;
da_cursor->level[0].bno = da_bno;
- da_cursor->level[0].index =
- be16_to_cpu(leaf->hdr.count);
+ da_cursor->level[0].index = leafhdr.count;
da_cursor->level[0].dirty = buf_dirty;
- if (be32_to_cpu(leaf->hdr.info.back) != prev_bno) {
+ if (leafhdr.back != prev_bno) {
do_warn(
_("bad sibling back pointer for block %u in directory inode %" PRIu64 "\n"),
da_bno, ino);
@@ -1745,7 +1751,7 @@ _("bad sibling back pointer for block %u in directory inode %" PRIu64 "\n"),
goto error_out;
}
prev_bno = da_bno;
- da_bno = be32_to_cpu(leaf->hdr.info.forw);
+ da_bno = leafhdr.forw;
if (da_bno != 0) {
if (verify_dir2_path(mp, da_cursor, 0)) {
libxfs_putbuf(bp);
@@ -1888,7 +1894,8 @@ _("can't read block %" PRIu64 " for directory inode %" PRIu64 "\n"),
continue;
}
data = bp->b_addr;
- if (be32_to_cpu(data->hdr.magic) != XFS_DIR2_DATA_MAGIC)
+ if (!(be32_to_cpu(data->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
+ be32_to_cpu(data->hdr.magic) == XFS_DIR3_DATA_MAGIC))
do_warn(
_("bad directory block magic # %#x in block %" PRIu64 " for directory inode %" PRIu64 "\n"),
be32_to_cpu(data->hdr.magic), dbno, ino);
diff --git a/repair/dir2.h b/repair/dir2.h
index 63e35d4..6a85015 100644
--- a/repair/dir2.h
+++ b/repair/dir2.h
@@ -33,13 +33,13 @@ typedef union {
typedef struct xfs_dir2_data {
xfs_dir2_data_hdr_t hdr; /* magic XFS_DIR2_DATA_MAGIC */
- xfs_dir2_data_union_t u[1];
+ xfs_dir2_data_union_t __u[1];
} xfs_dir2_data_t;
typedef struct xfs_dir2_block {
xfs_dir2_data_hdr_t hdr; /* magic XFS_DIR2_BLOCK_MAGIC */
- xfs_dir2_data_union_t u[1];
- xfs_dir2_leaf_entry_t leaf[1];
+ xfs_dir2_data_union_t __u[1];
+ xfs_dir2_leaf_entry_t __leaf[1];
xfs_dir2_block_tail_t tail;
} xfs_dir2_block_t;
diff --git a/repair/phase6.c b/repair/phase6.c
index 64f7a49..8900deb 100644
--- a/repair/phase6.c
+++ b/repair/phase6.c
@@ -1337,6 +1337,8 @@ entry_junked(
/*
* process a leaf block, also checks for .. entry
* and corrects it to match what we think .. should be
+ *
+ * This is a v1 directory function.
*/
static void
lf_block_dir_entry_check(xfs_mount_t *mp,
@@ -1581,6 +1583,8 @@ lf_block_dir_entry_check(xfs_mount_t *mp,
* succeeds or dies, inode never gets dirtied since all changes
* happen in file blocks. the inode size and other core info
* is already correct, it's just the leaf entries that get altered.
+ *
+ * This is a function that verifies a v1 directory structure.
*/
static void
longform_dir_entry_check(xfs_mount_t *mp,
@@ -1921,7 +1925,7 @@ longform_dir2_entry_check_data(
bp = *bpp;
d = bp->b_addr;
- ptr = (char *)d->u;
+ ptr = (char *)xfs_dir3_data_entry_p(&d->hdr);
nbad = 0;
needscan = needlog = 0;
junkit = 0;
@@ -1932,10 +1936,16 @@ longform_dir2_entry_check_data(
endptr = (char *)blp;
if (endptr > (char *)btp)
endptr = (char *)btp;
- wantmagic = XFS_DIR2_BLOCK_MAGIC;
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ wantmagic = XFS_DIR3_BLOCK_MAGIC;
+ else
+ wantmagic = XFS_DIR2_BLOCK_MAGIC;
} else {
endptr = (char *)d + mp->m_dirblksize;
- wantmagic = XFS_DIR2_DATA_MAGIC;
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ wantmagic = XFS_DIR3_DATA_MAGIC;
+ else
+ wantmagic = XFS_DIR2_DATA_MAGIC;
}
db = xfs_dir2_da_to_db(mp, da_bno);
@@ -1976,8 +1986,8 @@ longform_dir2_entry_check_data(
break;
/* check for block with no data entries */
- if ((ptr == (char *)d->u) && (ptr +
- be16_to_cpu(dup->length) >= endptr)) {
+ if ((ptr == (char *)xfs_dir3_data_entry_p(&d->hdr)) &&
+ (ptr + be16_to_cpu(dup->length) >= endptr)) {
junkit = 1;
*num_illegal += 1;
break;
@@ -2048,7 +2058,7 @@ longform_dir2_entry_check_data(
do_warn(_("would fix magic # to %#x\n"), wantmagic);
}
lastfree = 0;
- ptr = (char *)d->u;
+ ptr = (char *)xfs_dir3_data_entry_p(&d->hdr);
/*
* look at each entry. reference inode pointed to by each
* entry in the incore inode tree.
@@ -2218,7 +2228,8 @@ longform_dir2_entry_check_data(
if (ip->i_ino == inum) {
ASSERT(dep->name[0] == '.' && dep->namelen == 1);
add_inode_ref(current_irec, current_ino_offset);
- if (da_bno != 0 || dep != (xfs_dir2_data_entry_t *)d->u) {
+ if (da_bno != 0 ||
+ dep != xfs_dir3_data_entry_p(&d->hdr)) {
/* "." should be the first entry */
nbad++;
if (entry_junked(
@@ -2327,6 +2338,7 @@ longform_dir2_check_leaf(
xfs_dir2_leaf_tail_t *ltp;
int seeval;
struct xfs_dir2_leaf_entry *ents;
+ struct xfs_dir3_icleaf_hdr leafhdr;
da_bno = mp->m_dirleafblk;
if (libxfs_da_read_buf(NULL, ip, da_bno, -1, &bp, XFS_DATA_FORK,
@@ -2337,27 +2349,24 @@ longform_dir2_check_leaf(
/* NOTREACHED */
}
leaf = bp->b_addr;
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
ents = xfs_dir3_leaf_ents_p(leaf);
ltp = xfs_dir2_leaf_tail_p(mp, leaf);
bestsp = xfs_dir2_leaf_bests_p(ltp);
- if (be16_to_cpu(leaf->hdr.info.magic) != XFS_DIR2_LEAF1_MAGIC ||
- be32_to_cpu(leaf->hdr.info.forw) ||
- be32_to_cpu(leaf->hdr.info.back) ||
- be16_to_cpu(leaf->hdr.count) <
- be16_to_cpu(leaf->hdr.stale) ||
- be16_to_cpu(leaf->hdr.count) >
+ if (!(leafhdr.magic == XFS_DIR2_LEAF1_MAGIC ||
+ leafhdr.magic == XFS_DIR3_LEAF1_MAGIC) ||
+ leafhdr.forw || leafhdr.back ||
+ leafhdr.count < leaf->hdr.stale ||
+ leafhdr.count >
xfs_dir3_max_leaf_ents(mp, leaf) ||
- (char *)&ents[be16_to_cpu(
- leaf->hdr.count)] > (char *)bestsp) {
+ (char *)&ents[leafhdr.count] > (char *)bestsp) {
do_warn(
_("leaf block %u for directory inode %" PRIu64 " bad header\n"),
da_bno, ip->i_ino);
libxfs_putbuf(bp);
return 1;
}
- seeval = dir_hash_see_all(hashtab, ents,
- be16_to_cpu(leaf->hdr.count),
- be16_to_cpu(leaf->hdr.stale));
+ seeval = dir_hash_see_all(hashtab, ents, leafhdr.count, leafhdr.stale);
if (dir_hash_check(hashtab, ip, seeval)) {
libxfs_putbuf(bp);
return 1;
@@ -2399,6 +2408,9 @@ longform_dir2_check_node(
int seeval = 0;
int used;
struct xfs_dir2_leaf_entry *ents;
+ struct xfs_dir3_icleaf_hdr leafhdr;
+ struct xfs_dir3_icfree_hdr freehdr;
+ __be16 *bests;
for (da_bno = mp->m_dirleafblk, next_da_bno = 0;
next_da_bno != NULLFILEOFF && da_bno < mp->m_dirfreeblk;
@@ -2414,23 +2426,23 @@ longform_dir2_check_node(
return 1;
}
leaf = bp->b_addr;
+ xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf);
ents = xfs_dir3_leaf_ents_p(leaf);
- if (be16_to_cpu(leaf->hdr.info.magic) != XFS_DIR2_LEAFN_MAGIC) {
- if (be16_to_cpu(leaf->hdr.info.magic) ==
- XFS_DA_NODE_MAGIC) {
+ if (!(leafhdr.magic == XFS_DIR2_LEAFN_MAGIC ||
+ leafhdr.magic == XFS_DIR3_LEAFN_MAGIC)) {
+ if (leafhdr.magic == XFS_DA_NODE_MAGIC ||
+ leafhdr.magic == XFS_DA3_NODE_MAGIC) {
libxfs_putbuf(bp);
continue;
}
do_warn(
_("unknown magic number %#x for block %u in directory inode %" PRIu64 "\n"),
- be16_to_cpu(leaf->hdr.info.magic),
- da_bno, ip->i_ino);
+ leafhdr.magic, da_bno, ip->i_ino);
libxfs_putbuf(bp);
return 1;
}
- if (be16_to_cpu(leaf->hdr.count) > xfs_dir3_max_leaf_ents(mp, leaf) ||
- be16_to_cpu(leaf->hdr.count) <
- be16_to_cpu(leaf->hdr.stale)) {
+ if (leafhdr.count > xfs_dir3_max_leaf_ents(mp, leaf) ||
+ leafhdr.count < leafhdr.stale) {
do_warn(
_("leaf block %u for directory inode %" PRIu64 " bad header\n"),
da_bno, ip->i_ino);
@@ -2438,8 +2450,7 @@ longform_dir2_check_node(
return 1;
}
seeval = dir_hash_see_all(hashtab, ents,
- be16_to_cpu(leaf->hdr.count),
- be16_to_cpu(leaf->hdr.stale));
+ leafhdr.count, leafhdr.stale);
libxfs_putbuf(bp);
if (seeval != DIR_HASH_CK_OK)
return 1;
@@ -2461,35 +2472,35 @@ longform_dir2_check_node(
return 1;
}
free = bp->b_addr;
+ xfs_dir3_free_hdr_from_disk(&freehdr, free);
+ bests = xfs_dir3_free_bests_p(mp, free);
fdb = xfs_dir2_da_to_db(mp, da_bno);
- if (be32_to_cpu(free->hdr.magic) != XFS_DIR2_FREE_MAGIC ||
- be32_to_cpu(free->hdr.firstdb) !=
+ if (!(freehdr.magic == XFS_DIR2_FREE_MAGIC ||
+ freehdr.magic == XFS_DIR3_FREE_MAGIC) ||
+ freehdr.firstdb !=
(fdb - XFS_DIR2_FREE_FIRSTDB(mp)) *
xfs_dir3_free_max_bests(mp) ||
- be32_to_cpu(free->hdr.nvalid) <
- be32_to_cpu(free->hdr.nused)) {
+ freehdr.nvalid < freehdr.nused) {
do_warn(
_("free block %u for directory inode %" PRIu64 " bad header\n"),
da_bno, ip->i_ino);
libxfs_putbuf(bp);
return 1;
}
- for (i = used = 0; i < be32_to_cpu(free->hdr.nvalid); i++) {
- if (i + be32_to_cpu(free->hdr.firstdb) >=
- freetab->nents ||
- freetab->ents[i + be32_to_cpu(
- free->hdr.firstdb)].v !=
- be16_to_cpu(free->bests[i])) {
+ for (i = used = 0; i < freehdr.nvalid; i++) {
+ if (i + freehdr.firstdb >= freetab->nents ||
+ freetab->ents[i + freehdr.firstdb].v !=
+ be16_to_cpu(bests[i])) {
do_warn(
_("free block %u entry %i for directory ino %" PRIu64 " bad\n"),
da_bno, i, ip->i_ino);
libxfs_putbuf(bp);
return 1;
}
- used += be16_to_cpu(free->bests[i]) != NULLDATAOFF;
- freetab->ents[i + be32_to_cpu(free->hdr.firstdb)].s = 1;
+ used += be16_to_cpu(bests[i]) != NULLDATAOFF;
+ freetab->ents[i + freehdr.firstdb].s = 1;
}
- if (used != be32_to_cpu(free->hdr.nused)) {
+ if (used != freehdr.nused) {
do_warn(
_("free block %u for directory inode %" PRIu64 " bad nused\n"),
da_bno, ip->i_ino);
@@ -2632,6 +2643,8 @@ longform_dir2_entry_check(xfs_mount_t *mp,
/*
* shortform directory processing routines -- entry verification and
* bad entry deletion (pruning).
+ *
+ * This is a v1 directory structure function.
*/
static void
shortform_dir_entry_check(xfs_mount_t *mp,
xfsprogs-crc-patchset-v2/debug 0000600 0001750 0001750 00000000005 12132011547 015244 0 ustar dave dave ---
xfsprogs-crc-patchset-v2/bpm-progs-deb 0000600 0001750 0001750 00000007771 12132011547 016635 0 ustar dave dave xfsprogs: update 'make deb' to use tarball
From: Ben Myers <bpm@xxxxxxx>
This patch changes the build process so that 'make deb' uses the same
process of creating a source tree as the release script.
* Add a list of files which go in the release tarball in .gitcensus
This is needed so that you can create a tarball in a bare release
tree, when .git is not available.
* Modify the SRCTAR target to include files from .git_census and use tar
instead of git archive.
* Modify the SRCTARINC files to include .git_census, and include
.git_census in the 'make realclean' target.
* remove the 'make source-link' target.
Signed-off-by: Ben Myers <bpm@xxxxxxx>
---
v2: don't muck with signing the deb in this patch, rename .git_census to
.gitcensus, add .gitcensus to .gitignore.
.gitignore | 1 +
Makefile | 24 ++++++++++++------------
include/buildrules | 11 -----------
3 files changed, 13 insertions(+), 23 deletions(-)
Index: b/.gitignore
===================================================================
--- a/.gitignore 2013-02-14 10:40:25.861946781 -0600
+++ b/.gitignore 2013-02-14 10:40:39.352005790 -0600
@@ -5,6 +5,7 @@
# build system
.census
+.gitcensus
/include/platform_defs.h
/include/builddefs
/install-sh
Index: b/Makefile
===================================================================
--- a/Makefile 2013-02-14 09:43:02.411946730 -0600
+++ b/Makefile 2013-02-14 10:41:15.151360977 -0600
@@ -30,8 +30,7 @@ SRCTAR = $(PKG_NAME)-$(PKG_VERSION).tar.
CONFIGURE = aclocal.m4 configure config.guess config.sub install-sh ltmain.sh
LSRCFILES = configure.ac release.sh README VERSION $(CONFIGURE)
SRCTARINC = m4/libtool.m4 m4/lt~obsolete.m4 m4/ltoptions.m4 m4/ltsugar.m4 \
- m4/ltversion.m4 po/xfsprogs.pot $(CONFIGURE)
-
+ m4/ltversion.m4 po/xfsprogs.pot .gitcensus $(CONFIGURE)
LDIRT = config.log .ltdep .dep config.status config.cache confdefs.h \
conftest* built .census install.* install-dev.* *.gz \
autom4te.cache/* libtool include/builddefs include/platform_defs.h
@@ -120,7 +119,7 @@ distclean: clean
$(Q)rm -f $(LDIRT)
realclean: distclean
- $(Q)rm -f $(CONFIGURE)
+ $(Q)rm -f $(CONFIGURE) .gitcensus
#
# All this gunk is to allow for a make dist on an unconfigured tree
@@ -137,18 +136,19 @@ ifeq ($(HAVE_BUILDDEFS), no)
$(Q)$(MAKE) $(MAKEOPTS) -C . $@
else
$(Q)$(MAKE) $(MAKEOPTS) $(SRCDIR)
- $(Q)$(MAKE) $(MAKEOPTS) -C po
- $(Q)$(MAKE) $(MAKEOPTS) source-link
$(Q)cd $(SRCDIR) && dpkg-buildpackage
endif
-$(SRCDIR) : $(_FORCE)
+$(SRCDIR) : $(_FORCE) $(SRCTAR)
rm -fr $@
- mkdir -p $@
+ $(Q)$(TAR) -zxvf $(SRCTAR)
-$(SRCTAR) : default
- $(Q)git archive --prefix=$(SRCDIR)/ --format=tar v$(PKG_VERSION) > $(SRCDIR).tar
- $(Q)$(TAR) --transform "s,^,$(SRCDIR)/," -rf $(SRCDIR).tar \
- $(SRCTARINC)
- $(Q)$(ZIP) $(SRCDIR).tar
+$(SRCTAR) : default $(SRCTARINC) .gitcensus
+ $(Q)$(TAR) --transform "s,^,$(SRCDIR)/," -zcf $(SRCDIR).tar.gz \
+ `cat .gitcensus` $(SRCTARINC)
echo Wrote: $@
+
+.gitcensus: $(_FORCE)
+ $(Q)if test -d .git; then \
+ git ls-files > .gitcensus && echo "new .gitcensus"; \
+ fi
Index: b/include/buildrules
===================================================================
--- a/include/buildrules 2013-02-14 09:43:02.421945975 -0600
+++ b/include/buildrules 2013-02-14 09:44:13.291947757 -0600
@@ -23,17 +23,6 @@ $(SUBDIRS):
$(Q)$(MAKE) $(MAKEOPTS) -q -C $@ || $(MAKE) $(MAKEOPTS) -C $@
endif
-source-link:
- @test -z "$$DIR" && DIR="."; \
- for f in `echo $(SRCFILES) $(SUBDIRS) $(POTHEAD)`; do \
- if test -d $$f ; then \
- mkdir $(TOPDIR)/$(PKG_NAME)-$(PKG_VERSION)/$$DIR/$$f || exit $$?; \
- $(MAKEF) DIR=$$DIR/$$f -C $$f $@ || exit $$?; \
- else \
- ln $$f $(TOPDIR)/$(PKG_NAME)-$(PKG_VERSION)/$$DIR/$$f || exit $$?; \
- fi; \
- done
-
#
# Standard targets
#
_______________________________________________
xfs mailing list
xfs@xxxxxxxxxxx
http://oss.sgi.com/mailman/listinfo/xfs
--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
xfsprogs-crc-patchset-v2/xfsprogs-freespace 0000600 0001750 0001750 00000024531 12132011547 017776 0 ustar dave dave xfsprogs: Space management tool
From: Dave Chinner <dchinner@xxxxxxxxxx>
xfs_spaceman is intended as a diagnostic and control tool for space
management operations within XFS. Operations like examining free
space, managing allocation policies, issuing block discards on free
space, etc.
The tool is modelled on the xfs_io interface, allowing both
interactive and command line control of the tool, enabling it to be
used in scripts and automated management tools.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
Makefile | 3 +-
spaceman/Makefile | 34 ++++++++++++
spaceman/file.c | 149 +++++++++++++++++++++++++++++++++++++++++++++++++++++
spaceman/init.c | 119 ++++++++++++++++++++++++++++++++++++++++++
spaceman/init.h | 24 +++++++++
spaceman/space.h | 36 +++++++++++++
6 files changed, 364 insertions(+), 1 deletion(-)
diff --git a/Makefile b/Makefile
index c40fb2c..a81b8b2 100644
--- a/Makefile
+++ b/Makefile
@@ -41,7 +41,7 @@ endif
LIB_SUBDIRS = libxfs libxlog libxcmd libhandle libdisk
TOOL_SUBDIRS = copy db estimate fsck fsr growfs io logprint mkfs quota \
- mdrestore repair rtcp m4 man doc po debian
+ mdrestore repair rtcp m4 man doc po debian spaceman
SUBDIRS = include $(LIB_SUBDIRS) $(TOOL_SUBDIRS)
@@ -62,6 +62,7 @@ io: libxcmd libhandle
mkfs: libxfs
quota: libxcmd
repair: libxfs libxlog
+space: libxcmd
ifneq ($(ENABLE_BLKID), yes)
mkfs: libdisk
diff --git a/spaceman/Makefile b/spaceman/Makefile
new file mode 100644
index 0000000..ff8d23e
--- /dev/null
+++ b/spaceman/Makefile
@@ -0,0 +1,34 @@
+#
+# Copyright (c) 2012 Red Hat, Inc. All Rights Reserved.
+#
+
+TOPDIR = ..
+include $(TOPDIR)/include/builddefs
+
+LTCOMMAND = xfs_spaceman
+HFILES = init.h space.h
+CFILES = init.c \
+ file.c
+
+LLDLIBS = $(LIBXCMD)
+LTDEPENDENCIES = $(LIBXCMD)
+LLDFLAGS = -static
+
+ifeq ($(ENABLE_READLINE),yes)
+LLDLIBS += $(LIBREADLINE) $(LIBTERMCAP)
+endif
+
+ifeq ($(ENABLE_EDITLINE),yes)
+LLDLIBS += $(LIBEDITLINE) $(LIBTERMCAP)
+endif
+
+default: depend $(LTCOMMAND)
+
+include $(BUILDRULES)
+
+install: default
+ $(INSTALL) -m 755 -d $(PKG_SBIN_DIR)
+ $(LTINSTALL) -m 755 $(LTCOMMAND) $(PKG_SBIN_DIR)
+install-dev:
+
+-include .dep
diff --git a/spaceman/file.c b/spaceman/file.c
new file mode 100644
index 0000000..ea4ab0c
--- /dev/null
+++ b/spaceman/file.c
@@ -0,0 +1,149 @@
+/*
+ * Copyright (c) 2004-2005 Silicon Graphics, Inc.
+ * Copyright (c) 2012 Red Hat, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include <xfs/xfs.h>
+#include <xfs/command.h>
+#include <xfs/input.h>
+#include <sys/mman.h>
+#include "init.h"
+#include "space.h"
+
+static cmdinfo_t print_cmd;
+
+fileio_t *filetable;
+int filecount;
+fileio_t *file;
+
+static void
+print_fileio(
+ fileio_t *file,
+ int index,
+ int braces)
+{
+ printf(_("%c%03d%c %-14s (%s,%s,%s%s%s)\n"),
+ braces? '[' : ' ', index, braces? ']' : ' ', file->name,
+ file->flags & O_SYNC ? _("sync") : _("non-sync"),
+ file->flags & O_DIRECT ? _("direct") : _("non-direct"),
+ file->flags & O_RDONLY ? _("read-only") : _("read-write"),
+ file->flags & O_APPEND ? _(",append-only") : "",
+ file->flags & O_NONBLOCK ? _(",non-block") : "");
+}
+
+int
+filelist_f(void)
+{
+ int i;
+
+ for (i = 0; i < filecount; i++)
+ print_fileio(&filetable[i], i, &filetable[i] == file);
+ return 0;
+}
+
+static int
+print_f(
+ int argc,
+ char **argv)
+{
+ filelist_f();
+ return 0;
+}
+
+int
+openfile(
+ char *path,
+ xfs_fsop_geom_t *geom,
+ int flags,
+ mode_t mode)
+{
+ int fd;
+
+ fd = open(path, flags, mode);
+ if (fd < 0) {
+ if ((errno == EISDIR) && (flags & O_RDWR)) {
+ /* make it as if we asked for O_RDONLY & try again */
+ flags &= ~O_RDWR;
+ flags |= O_RDONLY;
+ fd = open(path, flags, mode);
+ if (fd < 0) {
+ perror(path);
+ return -1;
+ }
+ } else {
+ perror(path);
+ return -1;
+ }
+ }
+
+ if (xfsctl(path, fd, XFS_IOC_FSGEOMETRY, geom) < 0) {
+ perror("XFS_IOC_FSGEOMETRY");
+ close(fd);
+ return -1;
+ }
+ return fd;
+}
+
+int
+addfile(
+ char *name,
+ int fd,
+ xfs_fsop_geom_t *geometry,
+ int flags)
+{
+ char *filename;
+
+ filename = strdup(name);
+ if (!filename) {
+ perror("strdup");
+ close(fd);
+ return -1;
+ }
+
+ /* Extend the table of currently open files */
+ filetable = (fileio_t *)realloc(filetable, /* growing */
+ ++filecount * sizeof(fileio_t));
+ if (!filetable) {
+ perror("realloc");
+ filecount = 0;
+ free(filename);
+ close(fd);
+ return -1;
+ }
+
+ /* Finally, make this the new active open file */
+ file = &filetable[filecount - 1];
+ file->fd = fd;
+ file->flags = flags;
+ file->name = filename;
+ file->geom = *geometry;
+ return 0;
+}
+
+void
+file_init(void)
+{
+ print_cmd.name = "print";
+ print_cmd.altname = "p";
+ print_cmd.cfunc = print_f;
+ print_cmd.argmin = 0;
+ print_cmd.argmax = 0;
+ print_cmd.flags = CMD_FLAG_GLOBAL;
+ print_cmd.oneline = _("list current open files");
+
+ add_command(&print_cmd);
+}
diff --git a/spaceman/init.c b/spaceman/init.c
new file mode 100644
index 0000000..98c79e5
--- /dev/null
+++ b/spaceman/init.c
@@ -0,0 +1,119 @@
+/*
+ * Copyright (c) 2012 Red Hat, Inc
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include <xfs/xfs.h>
+#include <xfs/command.h>
+#include <xfs/input.h>
+#include "init.h"
+#include "space.h"
+
+char *progname;
+int exitcode;
+
+void
+usage(void)
+{
+ fprintf(stderr,
+ _("Usage: %s [-c cmd] file\n"),
+ progname);
+ exit(1);
+}
+
+static void
+init_commands(void)
+{
+ file_init();
+ help_init();
+ quit_init();
+}
+
+static int
+init_args_command(
+ int index)
+{
+ if (index >= filecount)
+ return 0;
+ file = &filetable[index++];
+ return index;
+}
+
+static int
+init_check_command(
+ const cmdinfo_t *ct)
+{
+ if (!(ct->flags & CMD_FLAG_GLOBAL))
+ return 0;
+ return 1;
+}
+
+void
+init(
+ int argc,
+ char **argv)
+{
+ int c, flags = 0;
+ mode_t mode = 0600;
+ xfs_fsop_geom_t geometry = { 0 };
+
+ progname = basename(argv[0]);
+ setlocale(LC_ALL, "");
+ bindtextdomain(PACKAGE, LOCALEDIR);
+ textdomain(PACKAGE);
+
+ while ((c = getopt(argc, argv, "c:V")) != EOF) {
+ switch (c) {
+ case 'c':
+ add_user_command(optarg);
+ break;
+ case 'V':
+ printf(_("%s version %s\n"), progname, VERSION);
+ exit(0);
+ default:
+ usage();
+ }
+ }
+
+ if (optind == argc)
+ usage();
+
+ while (optind < argc) {
+ if ((c = openfile(argv[optind], &geometry, flags, mode)) < 0)
+ exit(1);
+ if (!platform_test_xfs_fd(c)) {
+ printf(_("Not an XFS filesystem!\n"));
+ exit(1);
+ }
+ if (addfile(argv[optind], c, &geometry, flags) < 0)
+ exit(1);
+ optind++;
+ }
+
+ init_commands();
+ add_args_command(init_args_command);
+ add_check_command(init_check_command);
+}
+
+int
+main(
+ int argc,
+ char **argv)
+{
+ init(argc, argv);
+ command_loop();
+ return exitcode;
+}
diff --git a/spaceman/init.h b/spaceman/init.h
new file mode 100644
index 0000000..ecd0b5d
--- /dev/null
+++ b/spaceman/init.h
@@ -0,0 +1,24 @@
+/*
+ * Copyright (c) 2012 Red Hat, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+extern char *progname;
+extern int exitcode;
+
+#define min(a,b) (((a)<(b))?(a):(b))
+#define max(a,b) (((a)>(b))?(a):(b))
+
diff --git a/spaceman/space.h b/spaceman/space.h
new file mode 100644
index 0000000..6e1bc52
--- /dev/null
+++ b/spaceman/space.h
@@ -0,0 +1,36 @@
+/*
+ * Copyright (c) 2012 Red Hat, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+typedef struct fileio {
+ int fd; /* open file descriptor */
+ int flags; /* flags describing file state */
+ char *name; /* file name at time of open */
+ xfs_fsop_geom_t geom; /* XFS filesystem geometry */
+} fileio_t;
+
+extern fileio_t *filetable; /* open file table */
+extern int filecount; /* number of open files */
+extern fileio_t *file; /* active file in file table */
+extern int filelist_f(void);
+
+extern int openfile(char *, xfs_fsop_geom_t *, int, mode_t);
+extern int addfile(char *, int , xfs_fsop_geom_t *, int);
+
+extern void file_init(void);
+extern void help_init(void);
+extern void quit_init(void);
xfsprogs-crc-patchset-v2/spaceman-trim 0000600 0001750 0001750 00000011556 12132011547 016733 0 ustar dave dave spaceman: add FITRIM support
From: Dave Chinner <dchinner@xxxxxxxxxx>
Add support for discarding free space extents via the FITRIM
command. Make it easy to discard a single range, an entire AG or all
the freespace in the filesystem.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
spaceman/Makefile | 2 +-
spaceman/init.c | 1 +
spaceman/space.h | 1 +
spaceman/trim.c | 140 +++++++++++++++++++++++++++++++++++++++++++++++++++++
4 files changed, 143 insertions(+), 1 deletion(-)
diff --git a/spaceman/Makefile b/spaceman/Makefile
index ff8d23e..9fb9142 100644
--- a/spaceman/Makefile
+++ b/spaceman/Makefile
@@ -8,7 +8,7 @@ include $(TOPDIR)/include/builddefs
LTCOMMAND = xfs_spaceman
HFILES = init.h space.h
CFILES = init.c \
- file.c
+ file.c trim.c
LLDLIBS = $(LIBXCMD)
LTDEPENDENCIES = $(LIBXCMD)
diff --git a/spaceman/init.c b/spaceman/init.c
index f3d5397..307d91f 100644
--- a/spaceman/init.c
+++ b/spaceman/init.c
@@ -40,6 +40,7 @@ init_commands(void)
file_init();
help_init();
quit_init();
+ trim_init();
}
static int
diff --git a/spaceman/space.h b/spaceman/space.h
index 6e1bc52..7b4f034 100644
--- a/spaceman/space.h
+++ b/spaceman/space.h
@@ -34,3 +34,4 @@ extern int addfile(char *, int , xfs_fsop_geom_t *, int);
extern void file_init(void);
extern void help_init(void);
extern void quit_init(void);
+extern void trim_init(void);
diff --git a/spaceman/trim.c b/spaceman/trim.c
new file mode 100644
index 0000000..11385ff
--- /dev/null
+++ b/spaceman/trim.c
@@ -0,0 +1,140 @@
+/*
+ * Copyright (c) 2012 Red Hat, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include <xfs/xfs.h>
+#include <xfs/xfs_types.h>
+#include <xfs/command.h>
+#include <linux/fs.h>
+#include "init.h"
+#include "space.h"
+#include "input.h"
+
+#ifndef FITRIM
+#define FITRIM _IOWR('X', 121, struct fstrim_range) /* Trim */
+
+struct fstrim_range {
+ __u64 start;
+ __u64 len;
+ __u64 minlen;
+};
+#endif
+
+static cmdinfo_t trim_cmd;
+
+/*
+ * Report on trimace usage in xfs filesystem.
+ */
+static int
+trim_f(
+ int argc,
+ char **argv)
+{
+ struct fstrim_range trim = {0};
+ xfs_agnumber_t agno = 0;
+ off64_t offset = 0;
+ ssize_t length = 0;
+ ssize_t minlen = 0;
+ int aflag = 0;
+ int fflag = 0;
+ int ret;
+ int c;
+
+ while ((c = getopt(argc, argv, "a:fm:")) != EOF) {
+ switch (c) {
+ case 'a':
+ if (fflag)
+ return command_usage(&trim_cmd);
+ aflag = 1;
+ agno = atoi(optarg);
+ break;
+ case 'f':
+ if (aflag)
+ return command_usage(&trim_cmd);
+ fflag = 1;
+ break;
+ case 'm':
+ minlen = cvtnum(file->geom.blocksize,
+ file->geom.sectsize, argv[optind]);
+ break;
+ default:
+ return command_usage(&trim_cmd);
+ }
+ }
+
+ if (optind != argc - 2 && !(aflag || fflag))
+ return command_usage(&trim_cmd);
+ if (optind != argc) {
+ offset = cvtnum(file->geom.blocksize, file->geom.sectsize,
+ argv[optind]);
+ length = cvtnum(file->geom.blocksize, file->geom.sectsize,
+ argv[optind + 1]);
+ } else if (agno) {
+ offset = agno * file->geom.agblocks * file->geom.blocksize;
+ length = file->geom.agblocks * file->geom.blocksize;
+ } else {
+ offset = 0;
+ length = file->geom.datablocks * file->geom.blocksize;
+ }
+
+ trim.start = offset;
+ trim.len = length;
+ trim.minlen = minlen;
+
+ ret = ioctl(file->fd, FITRIM, (unsigned long)&trim);
+ if (ret < 0) {
+ fprintf(stderr, "%s: ioctl(FITRIM) [\"%s\"]: "
+ "%s\n", progname, file->name, strerror(errno));
+ exitcode = 1;
+ return 0;
+ }
+ return 0;
+}
+
+static void
+trim_help(void)
+{
+ printf(_(
+"\n"
+"Discard filesystem free space\n"
+"\n"
+"Options: [-m minlen] [-f]|[-a agno]|[offset length]\n"
+"\n"
+" -m minlen -- skip freespace extents smaller than minlen\n"
+" -f -- trim all the freespace in the entire filesystem\n"
+" -a agno -- trim all the freespace in the given AG agno\n"
+" offset length -- trim the freespace in the range {offset, length}\n"
+"\n"));
+
+}
+
+void
+trim_init(void)
+{
+ trim_cmd.name = "trim";
+ trim_cmd.altname = "tr";
+ trim_cmd.cfunc = trim_f;
+ trim_cmd.argmin = 1;
+ trim_cmd.argmax = 4;
+ trim_cmd.args = "[-m minlen] [-f]|[-a agno]|[offset length]\n";
+ trim_cmd.flags = CMD_FLAG_GLOBAL;
+ trim_cmd.oneline = _("Discard filesystem free space");
+ trim_cmd.help = trim_help;
+
+ add_command(&trim_cmd);
+}
+
xfsprogs-crc-patchset-v2/spaceman-prealloc-flush 0000600 0001750 0001750 00000012100 12132011547 020662 0 ustar dave dave spaceman: add new speculative prealloc control
From: Dave Chinner <dchinner@xxxxxxxxxx>
Add an control interface for purging speculative
preallocation via the new ioctls.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
spaceman/Makefile | 2 +-
spaceman/init.c | 1 +
spaceman/prealloc.c | 136 +++++++++++++++++++++++++++++++++++++++++++++++++++
spaceman/space.h | 1 +
4 files changed, 139 insertions(+), 1 deletion(-)
diff --git a/spaceman/Makefile b/spaceman/Makefile
index 9fb9142..b1f1136 100644
--- a/spaceman/Makefile
+++ b/spaceman/Makefile
@@ -8,7 +8,7 @@ include $(TOPDIR)/include/builddefs
LTCOMMAND = xfs_spaceman
HFILES = init.h space.h
CFILES = init.c \
- file.c trim.c
+ file.c prealloc.c trim.c
LLDLIBS = $(LIBXCMD)
LTDEPENDENCIES = $(LIBXCMD)
diff --git a/spaceman/init.c b/spaceman/init.c
index 314b5aa..e16ad83 100644
--- a/spaceman/init.c
+++ b/spaceman/init.c
@@ -39,6 +39,7 @@ init_commands(void)
{
file_init();
help_init();
+ prealloc_init();
quit_init();
trim_init();
}
diff --git a/spaceman/prealloc.c b/spaceman/prealloc.c
new file mode 100644
index 0000000..bef392a
--- /dev/null
+++ b/spaceman/prealloc.c
@@ -0,0 +1,136 @@
+/*
+ * Copyright (c) 2012 Red Hat, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include <xfs/xfs.h>
+#include <xfs/xfs_types.h>
+#include <xfs/command.h>
+#include "input.h"
+#include "init.h"
+#include "space.h"
+
+#ifndef XFS_IOC_FREE_EOFBLOCKS
+#define XFS_IOC_FREE_EOFBLOCKS _IOR ('X', 58, struct xfs_eofblocks)
+
+#define XFS_EOFBLOCKS_VERSION 1
+struct xfs_eofblocks {
+ __u32 eof_version;
+ __u32 eof_flags;
+ uid_t eof_uid;
+ gid_t eof_gid;
+ prid_t eof_prid;
+ __u32 pad32;
+ __u64 eof_min_file_size;
+ __u64 pad64[12];
+};
+
+/* eof_flags values */
+#define XFS_EOF_FLAGS_SYNC (1 << 0) /* sync/wait mode scan */
+#define XFS_EOF_FLAGS_UID (1 << 1) /* filter by uid */
+#define XFS_EOF_FLAGS_GID (1 << 2) /* filter by gid */
+#define XFS_EOF_FLAGS_PRID (1 << 3) /* filter by project id */
+#define XFS_EOF_FLAGS_MINFILESIZE (1 << 4) /* filter by min file size */
+
+#endif
+
+static cmdinfo_t prealloc_cmd;
+
+/*
+ * Control preallocation amounts.
+ */
+static int
+prealloc_f(
+ int argc,
+ char **argv)
+{
+ struct xfs_eofblocks eofb = {0};
+ int c;
+
+ eofb.eof_version = XFS_EOFBLOCKS_VERSION;
+
+ while ((c = getopt(argc, argv, "g:m:p:su:")) != EOF) {
+ switch (c) {
+ case 'g':
+ eofb.eof_flags |= XFS_EOF_FLAGS_GID;
+ eofb.eof_gid = atoi(optarg);
+ break;
+ case 'u':
+ eofb.eof_flags |= XFS_EOF_FLAGS_UID;
+ eofb.eof_uid = atoi(optarg);
+ break;
+ case 'p':
+ eofb.eof_flags |= XFS_EOF_FLAGS_PRID;
+ eofb.eof_prid = atoi(optarg);
+ break;
+ case 's':
+ eofb.eof_flags |= XFS_EOF_FLAGS_SYNC;
+ break;
+ case 'm':
+ eofb.eof_flags |= XFS_EOF_FLAGS_MINFILESIZE;
+ eofb.eof_min_file_size = cvtnum(file->geom.blocksize,
+ file->geom.sectsize,
+ optarg);
+ break;
+ case '?':
+ default:
+ return command_usage(&prealloc_cmd);
+ }
+ }
+ if (optind != argc)
+ return command_usage(&prealloc_cmd);
+
+ if (xfsctl(file->name, file->fd, XFS_IOC_FREE_EOFBLOCKS, &eofb) < 0) {
+ fprintf(stderr, _("%s: XFS_IOC_FREE_EOFBLOCKS on %s: %s\n"),
+ progname, file->name, strerror(errno));
+ }
+ return 0;
+}
+
+static void
+prealloc_help(void)
+{
+ printf(_(
+"\n"
+"Control speculative preallocation\n"
+"\n"
+"Options: [-s] [-ugp id] [-m minlen]\n"
+"\n"
+" -s -- synchronous flush - wait for flush to complete\n"
+" -u uid -- remove prealloc on files matching user <uid>\n"
+" -g gid -- remove prealloc on files matching group <gid>\n"
+" -p prid -- remove prealloc on files matching project <prid>\n"
+" -m minlen -- only consider files larger than <minlen>\n"
+"\n"));
+
+}
+
+void
+prealloc_init(void)
+{
+ prealloc_cmd.name = "prealloc";
+ prealloc_cmd.altname = "prealloc";
+ prealloc_cmd.cfunc = prealloc_f;
+ prealloc_cmd.argmin = 1;
+ prealloc_cmd.argmax = -1;
+ prealloc_cmd.args = "[-s] [-ugp id] [-m minlen]\n";
+ prealloc_cmd.flags = CMD_FLAG_GLOBAL;
+ prealloc_cmd.oneline = _("Control specualtive preallocation");
+ prealloc_cmd.help = prealloc_help;
+
+ add_command(&prealloc_cmd);
+}
+
diff --git a/spaceman/space.h b/spaceman/space.h
index 7b4f034..0ae3116 100644
--- a/spaceman/space.h
+++ b/spaceman/space.h
@@ -33,5 +33,6 @@ extern int addfile(char *, int , xfs_fsop_geom_t *, int);
extern void file_init(void);
extern void help_init(void);
+extern void prealloc_init(void);
extern void quit_init(void);
extern void trim_init(void);
xfsprogs-crc-patchset-v2/spaceman-ag-control 0000600 0001750 0001750 00000015212 12132011547 020016 0 ustar dave dave spaceman: AG state control
From: Dave Chinner <dchinner@xxxxxxxxxx>
Add support for a new allocation group state control ioctl. This
allows control of various AG parameters, such as whether inode
allocation is allowed in the AG, metadata preference, whether new
allocations are allowed, etc. This requires a new ioctl.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
spaceman/Makefile | 2 +-
spaceman/ag.c | 222 +++++++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 223 insertions(+), 1 deletion(-)
diff --git a/spaceman/Makefile b/spaceman/Makefile
index b1f1136..08709b3 100644
--- a/spaceman/Makefile
+++ b/spaceman/Makefile
@@ -8,7 +8,7 @@ include $(TOPDIR)/include/builddefs
LTCOMMAND = xfs_spaceman
HFILES = init.h space.h
CFILES = init.c \
- file.c prealloc.c trim.c
+ ag.c file.c prealloc.c trim.c
LLDLIBS = $(LIBXCMD)
LTDEPENDENCIES = $(LIBXCMD)
diff --git a/spaceman/ag.c b/spaceman/ag.c
new file mode 100644
index 0000000..a32d5da
--- /dev/null
+++ b/spaceman/ag.c
@@ -0,0 +1,222 @@
+/*
+ * Copyright (c) 2012 Red Hat, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include <xfs/xfs.h>
+#include <xfs/xfs_types.h>
+#include <xfs/command.h>
+#include <linux/dqblk_xfs.h>
+#include "input.h"
+#include "init.h"
+#include "space.h"
+
+#ifndef XFS_IOC_AGCONTROL
+#define XFS_IOC_AGCONTROL _IOWR ('X', 60, struct xfs_agcontrol)
+
+#define XFS_AGCONTROL_VERSION 1
+struct xfs_agcontrol {
+ __u32 version;
+ __u32 flags;
+ __u32 agno;
+ __u32 state;
+ __u64 pad[8];
+};
+
+/* control flags */
+#define XFS_AGCONTROL_GETAGFSTATE (1 << 0) /* get AGF state */
+#define XFS_AGCONTROL_SETAGFSTATE (1 << 1) /* set AGF state */
+#define XFS_AGCONTROL_GETAGISTATE (1 << 2) /* get AGI state */
+#define XFS_AGCONTROL_SETAGISTATE (1 << 3) /* set AGI state */
+
+/* state flags */
+
+/*
+ * inode and allocation states are split. AGF and AGI online state will move in
+ * sync as it is really a whole AG state. No allocation flags imply no new
+ * allocations, but inodes and extents can be removed. Readonly means no
+ * modification (alloc or free) is allowed. This is to allow different
+ * operations to be performed. e.g. emptying an AG in preparation for a shrink
+ * require NOALLOC state, but an AG that has a corrupted freespace btree might
+ * be switched to READONLY until the freespace tree is rebuilt. An AGF/AGI in
+ * this corrupt/ro state will set the relevant corruption flag in the state
+ * field....
+ */
+#define XFS_AGFSTATE_ONLINE (1 << 0) /* AGF online */
+#define XFS_AGFSTATE_NOALLOC (1 << 1) /* No new allocation */
+#define XFS_AGFSTATE_READONLY (1 << 2) /* AGF is immutable */
+#define XFS_AGFSTATE_METADATA (1 << 3) /* metadata preferred */
+#define XFS_AGFSTATE_CORRUPT_BNO (1 << 4) /* bno freespace corrupt */
+#define XFS_AGFSTATE_CORRUPT_CNT (1 << 5) /* cnt freespace corrupt */
+#define XFS_AGFSTATE_CORRUPT_AGFL (1 << 6) /* AGFL freespace corrupt */
+
+#define XFS_AGISTATE_ONLINE (1 << 0) /* AGI online */
+#define XFS_AGISTATE_NOALLOC (1 << 1) /* No new allocation */
+#define XFS_AGISTATE_READONLY (1 << 2) /* AGI is immutable */
+#define XFS_AGISTATE_CORRUPT_TREE (1 << 2) /* AGI btree corrupt */
+
+#endif
+
+static cmdinfo_t agfctl_cmd;
+static cmdinfo_t agictl_cmd;
+
+static int
+agfctl_f(
+ int argc,
+ char **argv)
+{
+ struct xfs_agcontrol agctl = {0};
+ xfs_agnumber_t agno;
+ int gflag = 0;
+ int c;
+
+ while ((c = getopt(argc, argv, "gs")) != EOF) {
+ switch (c) {
+ case 'g':
+ gflag = 1;
+ break;
+ default:
+ return command_usage(&agfctl_cmd);
+ }
+ }
+ if (optind != argc - 1)
+ return command_usage(&agfctl_cmd);
+
+ agno = atoi(argv[optind]);
+ if (agno >= file->geom.agcount) {
+ fprintf(stderr, _("%s: agno %d out of range (max %d)\n"),
+ progname, agno, file->geom.agcount);
+ exitcode = 1;
+ return 0;
+ }
+
+ agctl.version = XFS_AGCONTROL_VERSION;
+ agctl.agno = agno;
+ if (gflag)
+ agctl.flags = XFS_AGCONTROL_GETAGFSTATE;
+
+ if (xfsctl(file->name, file->fd, XFS_IOC_AGCONTROL, &agctl) < 0) {
+ fprintf(stderr, _("%s: XFS_IOC_AGCONTROL on %s: %s\n"),
+ progname, file->name, strerror(errno));
+ }
+ return 0;
+}
+
+static void
+agfctl_help(void)
+{
+ printf(_(
+"\n"
+"AGF state control\n"
+"\n"
+"Options: [-g] agno\n"
+"\n"
+" -g -- get state\n"
+" agno -- AG to operate on\n"
+"\n"));
+
+}
+
+void
+agfctl_init(void)
+{
+ agfctl_cmd.name = "agfctl";
+ agfctl_cmd.altname = "agfctl";
+ agfctl_cmd.cfunc = agfctl_f;
+ agfctl_cmd.argmin = 2;
+ agfctl_cmd.argmax = -1;
+ agfctl_cmd.args = "agno\n";
+ agfctl_cmd.flags = CMD_FLAG_GLOBAL;
+ agfctl_cmd.oneline = _("AGF state control");
+ agfctl_cmd.help = agfctl_help;
+
+ add_command(&agfctl_cmd);
+}
+
+static int
+agictl_f(
+ int argc,
+ char **argv)
+{
+ struct xfs_agcontrol agctl = {0};
+ xfs_agnumber_t agno;
+ int gflag = 0;
+ int c;
+
+ while ((c = getopt(argc, argv, "gs")) != EOF) {
+ switch (c) {
+ case 'g':
+ gflag = 1;
+ break;
+ default:
+ return command_usage(&agictl_cmd);
+ }
+ }
+ if (optind != argc - 1)
+ return command_usage(&agictl_cmd);
+
+ agno = atoi(argv[optind]);
+ if (agno >= file->geom.agcount) {
+ fprintf(stderr, _("%s: agno %d out of range (max %d)\n"),
+ progname, agno, file->geom.agcount);
+ exitcode = 1;
+ return 0;
+ }
+
+ agctl.version = XFS_AGCONTROL_VERSION;
+ agctl.agno = agno;
+ if (gflag)
+ agctl.flags = XFS_AGCONTROL_GETAGISTATE;
+
+ if (xfsctl(file->name, file->fd, XFS_IOC_AGCONTROL, &agctl) < 0) {
+ fprintf(stderr, _("%s: XFS_IOC_AGCONTROL on %s: %s\n"),
+ progname, file->name, strerror(errno));
+ exitcode = 1;
+ return 0;
+ }
+ return 0;
+}
+
+static void
+agictl_help(void)
+{
+ printf(_(
+"\n"
+"AGI state control\n"
+"\n"
+"Options: [-g] agno\n"
+"\n"
+" -g -- get state\n"
+" agno -- AG to operate on\n"
+"\n"));
+
+}
+
+void
+agictl_init(void)
+{
+ agictl_cmd.name = "agictl";
+ agictl_cmd.altname = "agictl";
+ agictl_cmd.cfunc = agictl_f;
+ agictl_cmd.argmin = 2;
+ agictl_cmd.argmax = -1;
+ agictl_cmd.args = "agno\n";
+ agictl_cmd.flags = CMD_FLAG_GLOBAL;
+ agictl_cmd.oneline = _("AGI state control");
+ agictl_cmd.help = agictl_help;
+
+ add_command(&agictl_cmd);
+}
xfsprogs-crc-patchset-v2/xfsprogs-freespace-2 0000600 0001750 0001750 00000024635 12132011547 020142 0 ustar dave dave xfsprogs: Space management tool
From: Dave Chinner <dchinner@xxxxxxxxxx>
Add freespace mapping tool modelled on the xfs_db freesp command.
The advantage of this command over xfs_db is that it can be done
online and is coherent with concurrent modifications to the
filesystem.
This requires kernel extensions to the fiemap command to map free
space indexes.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
---
spaceman/Makefile | 2 +-
spaceman/freesp.c | 395 +++++++++++++++++++++++++++++++++++++++++++++++++++++
spaceman/init.c | 1 +
spaceman/space.h | 1 +
4 files changed, 398 insertions(+), 1 deletion(-)
diff --git a/spaceman/Makefile b/spaceman/Makefile
index b1f1136..4743ad8 100644
--- a/spaceman/Makefile
+++ b/spaceman/Makefile
@@ -8,7 +8,7 @@ include $(TOPDIR)/include/builddefs
LTCOMMAND = xfs_spaceman
HFILES = init.h space.h
CFILES = init.c \
- file.c prealloc.c trim.c
+ file.c freesp.c prealloc.c trim.c
LLDLIBS = $(LIBXCMD)
LTDEPENDENCIES = $(LIBXCMD)
diff --git a/spaceman/freesp.c b/spaceman/freesp.c
new file mode 100644
index 0000000..dc1d7a5
--- /dev/null
+++ b/spaceman/freesp.c
@@ -0,0 +1,395 @@
+/*
+ * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc.
+ * Copyright (c) 2012 Red Hat, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include <xfs/xfs.h>
+#include <xfs/xfs_types.h>
+#include <xfs/command.h>
+#include <linux/fs.h>
+#include <linux/fiemap.h>
+#include "init.h"
+#include "space.h"
+
+#ifndef FIEMAPFS_FLAG_FREESP
+#define FIEMAPFS_FLAG_FREESP 0x80000000
+#define FIEMAPFS_FLAG_FREESP_SIZE 0x40000000
+#define FIEMAPFS_FLAG_FREESP_SIZE_HINT 0x20000000
+
+#define FS_IOC_FIEMAPFS _IOWR('f', 12, struct fiemap)
+#endif
+
+typedef struct histent
+{
+ int low;
+ int high;
+ long long count;
+ long long blocks;
+} histent_t;
+
+static int agcount;
+static xfs_agnumber_t *aglist;
+static int countflag;
+static int dumpflag;
+static int equalsize;
+static histent_t *hist;
+static int histcount;
+static int multsize;
+static int seen1;
+static int summaryflag;
+static long long totblocks;
+static long long totexts;
+
+static cmdinfo_t freesp_cmd;
+
+static void
+addhistent(
+ int h)
+{
+ hist = realloc(hist, (histcount + 1) * sizeof(*hist));
+ if (h == 0)
+ h = 1;
+ hist[histcount].low = h;
+ hist[histcount].count = hist[histcount].blocks = 0;
+ histcount++;
+ if (h == 1)
+ seen1 = 1;
+}
+
+static void
+addtohist(
+ xfs_agnumber_t agno,
+ xfs_agblock_t agbno,
+ off64_t len)
+{
+ int i;
+
+ if (dumpflag)
+ printf("%8d %8d %8Zu\n", agno, agbno, len);
+ totexts++;
+ totblocks += len;
+ for (i = 0; i < histcount; i++) {
+ if (hist[i].high >= len) {
+ hist[i].count++;
+ hist[i].blocks += len;
+ break;
+ }
+ }
+}
+
+static int
+hcmp(
+ const void *a,
+ const void *b)
+{
+ return ((histent_t *)a)->low - ((histent_t *)b)->low;
+}
+
+static void
+histinit(
+ int maxlen)
+{
+ int i;
+
+ if (equalsize) {
+ for (i = 1; i < maxlen; i += equalsize)
+ addhistent(i);
+ } else if (multsize) {
+ for (i = 1; i < maxlen; i *= multsize)
+ addhistent(i);
+ } else {
+ if (!seen1)
+ addhistent(1);
+ qsort(hist, histcount, sizeof(*hist), hcmp);
+ }
+ for (i = 0; i < histcount; i++) {
+ if (i < histcount - 1)
+ hist[i].high = hist[i + 1].low - 1;
+ else
+ hist[i].high = maxlen;
+ }
+}
+
+static void
+printhist(void)
+{
+ int i;
+
+ printf("%7s %7s %7s %7s %6s\n",
+ _("from"), _("to"), _("extents"), _("blocks"), _("pct"));
+ for (i = 0; i < histcount; i++) {
+ if (hist[i].count)
+ printf("%7d %7d %7lld %7lld %6.2f\n", hist[i].low,
+ hist[i].high, hist[i].count, hist[i].blocks,
+ hist[i].blocks * 100.0 / totblocks);
+ }
+}
+
+static int
+inaglist(
+ xfs_agnumber_t agno)
+{
+ int i;
+
+ if (agcount == 0)
+ return 1;
+ for (i = 0; i < agcount; i++)
+ if (aglist[i] == agno)
+ return 1;
+ return 0;
+}
+
+#define NR_EXTENTS 128
+
+static void
+scan_ag(
+ xfs_agnumber_t agno)
+{
+ struct fiemap *fiemap;
+ off64_t blocksize = file->geom.blocksize;
+ uint64_t last_logical = agno * file->geom.agblocks * blocksize;
+ uint64_t length = file->geom.agblocks * blocksize;
+ off64_t fsbperag;
+ int fiemap_flags;
+ int last = 0;
+ int map_size;
+
+
+ last_logical = (off64_t)file->geom.agblocks * blocksize * agno;
+ length = (off64_t)file->geom.agblocks * blocksize;
+ fsbperag = (off64_t)file->geom.agblocks * blocksize;
+
+ map_size = sizeof(struct fiemap) +
+ sizeof(struct fiemap_extent) * NR_EXTENTS;
+ fiemap = malloc(map_size);
+ if (!fiemap) {
+ fprintf(stderr, _("%s: fiemap malloc failed.\n"), progname);
+ exitcode = 1;
+ return;
+ }
+ if (countflag)
+ fiemap_flags = FIEMAPFS_FLAG_FREESPACE_SIZE;
+ else
+ fiemap_flags = FIEMAPFS_FLAG_FREESPACE;
+
+ while (!last) {
+ struct fiemap_extent *extent;
+ xfs_agblock_t agbno;
+ int ret;
+ int i;
+
+ memset(fiemap, 0, map_size);
+ fiemap->fm_flags = fiemap_flags;
+ fiemap_flags &= ~FIEMAPFS_FLAGS_FREESP_CONTINUE;
+
+ fiemap->fm_start = last_logical;
+ fiemap->fm_length = length;
+ fiemap->fm_extent_count = NR_EXTENTS;
+
+ ret = ioctl(file->fd, FS_IOC_FIEMAPFS, (unsigned long)fiemap);
+ if (ret < 0) {
+ fprintf(stderr, "%s: ioctl(FS_IOC_FIEMAPFS) [\"%s\"]: "
+ "%s\n", progname, file->name, strerror(errno));
+ free(fiemap);
+ exitcode = 1;
+ return;
+ }
+
+ /* No more extents to map, exit */
+ if (!fiemap->fm_mapped_extents)
+ break;
+
+ for (i = 0; i < fiemap->fm_mapped_extents; i++) {
+ off64_t aglen;
+
+ extent = &fiemap->fm_extents[i];
+
+
+ agbno = (extent->fe_physical - (fsbperag * agno)) /
+ blocksize;
+ aglen = extent->fe_length / blocksize;
+
+ addtohist(agno, agbno, aglen);
+
+
+ if (extent->fe_flags & FIEMAP_EXTENT_LAST) {
+ last = 1;
+ break;
+ }
+ }
+
+ if (fiemap_flags == FIEMAPFS_FLAG_FREESPACE) {
+ /* move our range past over what we just searched */
+ last_logical = max(last_logical,
+ extent->fe_logical + extent->fe_length);
+ } else {
+ /*
+ * we want to start the search from the current
+ * extent, but size ordered free space can be found
+ * anywhere in the range we asked for so we cannot move
+ * last_logical around. This means we need to give the
+ * search the last extent we've found back to the kernel
+ * for it to start it's search again. Move
+ * it to extent zero, and flag it as a continued call.
+ */
+ memcpy(&fiemap->fm_extents[0], extent,
+ sizeof(fiemap->fm_extents[0]));
+ fiemap_flags |= FIEMAPFS_FLAGS_FREESP_SIZE_HINT;
+ }
+
+ }
+}
+static void
+aglistadd(
+ char *a)
+{
+ aglist = realloc(aglist, (agcount + 1) * sizeof(*aglist));
+ aglist[agcount] = (xfs_agnumber_t)atoi(a);
+ agcount++;
+}
+
+static int
+init(
+ int argc,
+ char **argv)
+{
+ int c;
+ int speced = 0;
+
+ agcount = countflag = dumpflag = equalsize = multsize = optind = 0;
+ histcount = seen1 = summaryflag = 0;
+ totblocks = totexts = 0;
+ aglist = NULL;
+ hist = NULL;
+ while ((c = getopt(argc, argv, "a:bcde:h:m:s")) != EOF) {
+ switch (c) {
+ case 'a':
+ aglistadd(optarg);
+ break;
+ case 'b':
+ if (speced)
+ return 0;
+ multsize = 2;
+ speced = 1;
+ break;
+ case 'c':
+ countflag = 1;
+ break;
+ case 'd':
+ dumpflag = 1;
+ break;
+ case 'e':
+ if (speced)
+ return 0;
+ equalsize = atoi(optarg);
+ speced = 1;
+ break;
+ case 'h':
+ if (speced && !histcount)
+ return 0;
+ addhistent(atoi(optarg));
+ speced = 1;
+ break;
+ case 'm':
+ if (speced)
+ return 0;
+ multsize = atoi(optarg);
+ speced = 1;
+ break;
+ case 's':
+ summaryflag = 1;
+ break;
+ case '?':
+ return 0;
+ }
+ }
+ if (optind != argc)
+ return 0;
+ if (!speced)
+ multsize = 2;
+ histinit(file->geom.agblocks);
+ return 1;
+}
+
+/*
+ * Report on freespace usage in xfs filesystem.
+ */
+static int
+freesp_f(
+ int argc,
+ char **argv)
+{
+ xfs_agnumber_t agno;
+
+ if (!init(argc, argv))
+ return 0;
+ for (agno = 0; agno < file->geom.agcount; agno++) {
+ if (inaglist(agno))
+ scan_ag(agno);
+ }
+ if (histcount)
+ printhist();
+ if (summaryflag) {
+ printf(_("total free extents %lld\n"), totexts);
+ printf(_("total free blocks %lld\n"), totblocks);
+ printf(_("average free extent size %g\n"),
+ (double)totblocks / (double)totexts);
+ }
+ if (aglist)
+ free(aglist);
+ if (hist)
+ free(hist);
+ return 0;
+}
+
+static void
+freesp_help(void)
+{
+ printf(_(
+"\n"
+"Examine filesystem free space\n"
+"\n"
+"Options: [-bcds] [-a agno] [-e bsize] [-h h1]... [-m bmult]\n"
+"\n"
+" -b -- binary histogram bin size\n"
+" -c -- scan the by-count (size) ordered freespace tree\n"
+" -d -- debug output\n"
+" -s -- emit freespace summary information\n"
+" -a agno -- scan only the given AG agno\n"
+" -e bsize -- use fixed histogram bin size of bsize\n"
+" -h h1 -- use custom histogram bin size of h1. Multiple specifications allowed.\n"
+" -m bmult -- use histogram bin size multiplier of bmult\n"
+"\n"));
+
+}
+
+void
+freesp_init(void)
+{
+ freesp_cmd.name = "freesp";
+ freesp_cmd.altname = "fsp";
+ freesp_cmd.cfunc = freesp_f;
+ freesp_cmd.argmin = 0;
+ freesp_cmd.argmax = -1;
+ freesp_cmd.args = "[-bcds] [-a agno] [-e bsize] [-h h1]... [-m bmult]\n";
+ freesp_cmd.flags = CMD_FLAG_GLOBAL;
+ freesp_cmd.oneline = _("Examine filesystem free space");
+ freesp_cmd.help = freesp_help;
+
+ add_command(&freesp_cmd);
+}
+
diff --git a/spaceman/init.c b/spaceman/init.c
index e16ad83..f8a3cb3 100644
--- a/spaceman/init.c
+++ b/spaceman/init.c
@@ -38,6 +38,7 @@ static void
init_commands(void)
{
file_init();
+ freesp_init();
help_init();
prealloc_init();
quit_init();
diff --git a/spaceman/space.h b/spaceman/space.h
index 0ae3116..a1381a6 100644
--- a/spaceman/space.h
+++ b/spaceman/space.h
@@ -32,6 +32,7 @@ extern int openfile(char *, xfs_fsop_geom_t *, int, mode_t);
extern int addfile(char *, int , xfs_fsop_geom_t *, int);
extern void file_init(void);
+extern void freesp_init(void);
extern void help_init(void);
extern void prealloc_init(void);
extern void quit_init(void);
_______________________________________________
xfs mailing list
xfs@xxxxxxxxxxx
http://oss.sgi.com/mailman/listinfo/xfs
