Implement the generic btree operations required to manipulate refcount
btree blocks. The implementation is similar to the bmapbt, though it
will only allocate and free blocks from the AG.
v2: Remove init_rec_from_key since we no longer need it, and add
tracepoints when refcount btree operations fail.
Since the refcount root and level fields are separate from the
existing roots and levels array, they need a separate logging flag.
Signed-off-by: Darrick J. Wong <darrick.wong@xxxxxxxxxx>
[hch: fix logging of AGF refcount btree fields]
Signed-off-by: Christoph Hellwig <hch@xxxxxx>
---
fs/xfs/Makefile | 1
fs/xfs/libxfs/xfs_alloc.c | 4 +
fs/xfs/libxfs/xfs_format.h | 5 +
fs/xfs/libxfs/xfs_refcount.c | 177 ++++++++++++++++++++++++++++++++
fs/xfs/libxfs/xfs_refcount.h | 30 +++++
fs/xfs/libxfs/xfs_refcount_btree.c | 197 ++++++++++++++++++++++++++++++++++++
6 files changed, 413 insertions(+), 1 deletion(-)
create mode 100644 fs/xfs/libxfs/xfs_refcount.c
create mode 100644 fs/xfs/libxfs/xfs_refcount.h
diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile
index 3f579af..e0c3067 100644
--- a/fs/xfs/Makefile
+++ b/fs/xfs/Makefile
@@ -55,6 +55,7 @@ xfs-y += $(addprefix libxfs/, \
xfs_ag_resv.o \
xfs_rmap.o \
xfs_rmap_btree.o \
+ xfs_refcount.o \
xfs_refcount_btree.o \
xfs_sb.o \
xfs_symlink_remote.o \
diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index 14f8a69d..be5d0df 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -2326,6 +2326,10 @@ xfs_alloc_log_agf(
offsetof(xfs_agf_t, agf_longest),
offsetof(xfs_agf_t, agf_btreeblks),
offsetof(xfs_agf_t, agf_uuid),
+ offsetof(xfs_agf_t, agf_refcount_root),
+ offsetof(xfs_agf_t, agf_refcount_level),
+ /* needed so that we don't log the whole rest of the structure: */
+ offsetof(xfs_agf_t, agf_spare64),
sizeof(xfs_agf_t)
};
diff --git a/fs/xfs/libxfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h
index 3b0cae2..45bbdad 100644
--- a/fs/xfs/libxfs/xfs_format.h
+++ b/fs/xfs/libxfs/xfs_format.h
@@ -680,7 +680,10 @@ typedef struct xfs_agf {
#define XFS_AGF_LONGEST 0x00000400
#define XFS_AGF_BTREEBLKS 0x00000800
#define XFS_AGF_UUID 0x00001000
-#define XFS_AGF_NUM_BITS 13
+#define XFS_AGF_REFCOUNT_ROOT 0x00002000
+#define XFS_AGF_REFCOUNT_LEVEL 0x00004000
+#define XFS_AGF_SPARE64 0x00008000
+#define XFS_AGF_NUM_BITS 16
#define XFS_AGF_ALL_BITS ((1 << XFS_AGF_NUM_BITS) - 1)
#define XFS_AGF_FLAGS \
diff --git a/fs/xfs/libxfs/xfs_refcount.c b/fs/xfs/libxfs/xfs_refcount.c
new file mode 100644
index 0000000..4d483b5
--- /dev/null
+++ b/fs/xfs/libxfs/xfs_refcount.c
@@ -0,0 +1,177 @@
+/*
+ * Copyright (C) 2016 Oracle. All Rights Reserved.
+ *
+ * Author: Darrick J. Wong <darrick.wong@xxxxxxxxxx>
+ *
+ * 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; either version 2
+ * of the License, or (at your option) any later version.
+ *
+ * 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"
+#include "xfs_fs.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
+#include "xfs_sb.h"
+#include "xfs_mount.h"
+#include "xfs_defer.h"
+#include "xfs_btree.h"
+#include "xfs_bmap.h"
+#include "xfs_refcount_btree.h"
+#include "xfs_alloc.h"
+#include "xfs_error.h"
+#include "xfs_trace.h"
+#include "xfs_cksum.h"
+#include "xfs_trans.h"
+#include "xfs_bit.h"
+#include "xfs_refcount.h"
+
+/*
+ * Look up the first record less than or equal to [bno, len] in the btree
+ * given by cur.
+ */
+int
+xfs_refcountbt_lookup_le(
+ struct xfs_btree_cur *cur,
+ xfs_agblock_t bno,
+ int *stat)
+{
+ trace_xfs_refcountbt_lookup(cur->bc_mp, cur->bc_private.a.agno, bno,
+ XFS_LOOKUP_LE);
+ cur->bc_rec.rc.rc_startblock = bno;
+ cur->bc_rec.rc.rc_blockcount = 0;
+ return xfs_btree_lookup(cur, XFS_LOOKUP_LE, stat);
+}
+
+/*
+ * Look up the first record greater than or equal to [bno, len] in the btree
+ * given by cur.
+ */
+int
+xfs_refcountbt_lookup_ge(
+ struct xfs_btree_cur *cur,
+ xfs_agblock_t bno,
+ int *stat)
+{
+ trace_xfs_refcountbt_lookup(cur->bc_mp, cur->bc_private.a.agno, bno,
+ XFS_LOOKUP_GE);
+ cur->bc_rec.rc.rc_startblock = bno;
+ cur->bc_rec.rc.rc_blockcount = 0;
+ return xfs_btree_lookup(cur, XFS_LOOKUP_GE, stat);
+}
+
+/*
+ * Get the data from the pointed-to record.
+ */
+int
+xfs_refcountbt_get_rec(
+ struct xfs_btree_cur *cur,
+ struct xfs_refcount_irec *irec,
+ int *stat)
+{
+ union xfs_btree_rec *rec;
+ int error;
+
+ error = xfs_btree_get_rec(cur, &rec, stat);
+ if (!error && *stat == 1) {
+ irec->rc_startblock = be32_to_cpu(rec->refc.rc_startblock);
+ irec->rc_blockcount = be32_to_cpu(rec->refc.rc_blockcount);
+ irec->rc_refcount = be32_to_cpu(rec->refc.rc_refcount);
+ trace_xfs_refcountbt_get(cur->bc_mp, cur->bc_private.a.agno,
+ irec);
+ }
+ return error;
+}
+
+/*
+ * Update the record referred to by cur to the value given
+ * by [bno, len, refcount].
+ * This either works (return 0) or gets an EFSCORRUPTED error.
+ */
+STATIC int
+xfs_refcountbt_update(
+ struct xfs_btree_cur *cur,
+ struct xfs_refcount_irec *irec)
+{
+ union xfs_btree_rec rec;
+ int error;
+
+ trace_xfs_refcountbt_update(cur->bc_mp, cur->bc_private.a.agno, irec);
+ rec.refc.rc_startblock = cpu_to_be32(irec->rc_startblock);
+ rec.refc.rc_blockcount = cpu_to_be32(irec->rc_blockcount);
+ rec.refc.rc_refcount = cpu_to_be32(irec->rc_refcount);
+ error = xfs_btree_update(cur, &rec);
+ if (error)
+ trace_xfs_refcountbt_update_error(cur->bc_mp,
+ cur->bc_private.a.agno, error, _RET_IP_);
+ return error;
+}
+
+/*
+ * Insert the record referred to by cur to the value given
+ * by [bno, len, refcount].
+ * This either works (return 0) or gets an EFSCORRUPTED error.
+ */
+STATIC int
+xfs_refcountbt_insert(
+ struct xfs_btree_cur *cur,
+ struct xfs_refcount_irec *irec,
+ int *i)
+{
+ int error;
+
+ trace_xfs_refcountbt_insert(cur->bc_mp, cur->bc_private.a.agno, irec);
+ cur->bc_rec.rc.rc_startblock = irec->rc_startblock;
+ cur->bc_rec.rc.rc_blockcount = irec->rc_blockcount;
+ cur->bc_rec.rc.rc_refcount = irec->rc_refcount;
+ error = xfs_btree_insert(cur, i);
+ XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, *i == 1, out_error);
+out_error:
+ if (error)
+ trace_xfs_refcountbt_insert_error(cur->bc_mp,
+ cur->bc_private.a.agno, error, _RET_IP_);
+ return error;
+}
+
+/*
+ * Remove the record referred to by cur, then set the pointer to the spot
+ * where the record could be re-inserted, in case we want to increment or
+ * decrement the cursor.
+ * This either works (return 0) or gets an EFSCORRUPTED error.
+ */
+STATIC int
+xfs_refcountbt_delete(
+ struct xfs_btree_cur *cur,
+ int *i)
+{
+ struct xfs_refcount_irec irec;
+ int found_rec;
+ int error;
+
+ error = xfs_refcountbt_get_rec(cur, &irec, &found_rec);
+ if (error)
+ goto out_error;
+ XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, found_rec == 1, out_error);
+ trace_xfs_refcountbt_delete(cur->bc_mp, cur->bc_private.a.agno, &irec);
+ error = xfs_btree_delete(cur, i);
+ XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, *i == 1, out_error);
+ if (error)
+ goto out_error;
+ error = xfs_refcountbt_lookup_ge(cur, irec.rc_startblock, &found_rec);
+out_error:
+ if (error)
+ trace_xfs_refcountbt_delete_error(cur->bc_mp,
+ cur->bc_private.a.agno, error, _RET_IP_);
+ return error;
+}
diff --git a/fs/xfs/libxfs/xfs_refcount.h b/fs/xfs/libxfs/xfs_refcount.h
new file mode 100644
index 0000000..8ea65c6
--- /dev/null
+++ b/fs/xfs/libxfs/xfs_refcount.h
@@ -0,0 +1,30 @@
+/*
+ * Copyright (C) 2016 Oracle. All Rights Reserved.
+ *
+ * Author: Darrick J. Wong <darrick.wong@xxxxxxxxxx>
+ *
+ * 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; either version 2
+ * of the License, or (at your option) any later version.
+ *
+ * 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_REFCOUNT_H__
+#define __XFS_REFCOUNT_H__
+
+extern int xfs_refcountbt_lookup_le(struct xfs_btree_cur *cur,
+ xfs_agblock_t bno, int *stat);
+extern int xfs_refcountbt_lookup_ge(struct xfs_btree_cur *cur,
+ xfs_agblock_t bno, int *stat);
+extern int xfs_refcountbt_get_rec(struct xfs_btree_cur *cur,
+ struct xfs_refcount_irec *irec, int *stat);
+
+#endif /* __XFS_REFCOUNT_H__ */
diff --git a/fs/xfs/libxfs/xfs_refcount_btree.c b/fs/xfs/libxfs/xfs_refcount_btree.c
index 359cf0c..7093c71 100644
--- a/fs/xfs/libxfs/xfs_refcount_btree.c
+++ b/fs/xfs/libxfs/xfs_refcount_btree.c
@@ -44,6 +44,153 @@ xfs_refcountbt_dup_cursor(
cur->bc_private.a.dfops);
}
+STATIC void
+xfs_refcountbt_set_root(
+ struct xfs_btree_cur *cur,
+ union xfs_btree_ptr *ptr,
+ int inc)
+{
+ struct xfs_buf *agbp = cur->bc_private.a.agbp;
+ struct xfs_agf *agf = XFS_BUF_TO_AGF(agbp);
+ xfs_agnumber_t seqno = be32_to_cpu(agf->agf_seqno);
+ struct xfs_perag *pag = xfs_perag_get(cur->bc_mp, seqno);
+
+ ASSERT(ptr->s != 0);
+
+ agf->agf_refcount_root = ptr->s;
+ be32_add_cpu(&agf->agf_refcount_level, inc);
+ pag->pagf_refcount_level += inc;
+ xfs_perag_put(pag);
+
+ xfs_alloc_log_agf(cur->bc_tp, agbp,
+ XFS_AGF_REFCOUNT_ROOT | XFS_AGF_REFCOUNT_LEVEL);
+}
+
+STATIC int
+xfs_refcountbt_alloc_block(
+ struct xfs_btree_cur *cur,
+ union xfs_btree_ptr *start,
+ union xfs_btree_ptr *new,
+ int *stat)
+{
+ struct xfs_alloc_arg args; /* block allocation args */
+ int error; /* error return value */
+
+ memset(&args, 0, sizeof(args));
+ args.tp = cur->bc_tp;
+ args.mp = cur->bc_mp;
+ args.type = XFS_ALLOCTYPE_NEAR_BNO;
+ args.fsbno = XFS_AGB_TO_FSB(cur->bc_mp, cur->bc_private.a.agno,
+ xfs_refc_block(args.mp));
+ args.firstblock = args.fsbno;
+ xfs_rmap_ag_owner(&args.oinfo, XFS_RMAP_OWN_REFC);
+ args.minlen = args.maxlen = args.prod = 1;
+
+ error = xfs_alloc_vextent(&args);
+ if (error)
+ goto out_error;
+ trace_xfs_refcountbt_alloc_block(cur->bc_mp, cur->bc_private.a.agno,
+ args.agbno, 1);
+ if (args.fsbno == NULLFSBLOCK) {
+ XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT);
+ *stat = 0;
+ return 0;
+ }
+ ASSERT(args.agno == cur->bc_private.a.agno);
+ ASSERT(args.len == 1);
+
+ new->s = cpu_to_be32(args.agbno);
+
+ XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT);
+ *stat = 1;
+ return 0;
+
+out_error:
+ XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR);
+ return error;
+}
+
+STATIC int
+xfs_refcountbt_free_block(
+ struct xfs_btree_cur *cur,
+ struct xfs_buf *bp)
+{
+ struct xfs_mount *mp = cur->bc_mp;
+ struct xfs_trans *tp = cur->bc_tp;
+ xfs_fsblock_t fsbno = XFS_DADDR_TO_FSB(mp, XFS_BUF_ADDR(bp));
+ struct xfs_owner_info oinfo;
+
+ trace_xfs_refcountbt_free_block(cur->bc_mp, cur->bc_private.a.agno,
+ XFS_FSB_TO_AGBNO(cur->bc_mp, fsbno), 1);
+ xfs_rmap_ag_owner(&oinfo, XFS_RMAP_OWN_REFC);
+ xfs_bmap_add_free(mp, cur->bc_private.a.dfops, fsbno, 1,
+ &oinfo);
+ xfs_trans_binval(tp, bp);
+ return 0;
+}
+
+STATIC int
+xfs_refcountbt_get_minrecs(
+ struct xfs_btree_cur *cur,
+ int level)
+{
+ return cur->bc_mp->m_refc_mnr[level != 0];
+}
+
+STATIC int
+xfs_refcountbt_get_maxrecs(
+ struct xfs_btree_cur *cur,
+ int level)
+{
+ return cur->bc_mp->m_refc_mxr[level != 0];
+}
+
+STATIC void
+xfs_refcountbt_init_key_from_rec(
+ union xfs_btree_key *key,
+ union xfs_btree_rec *rec)
+{
+ ASSERT(rec->refc.rc_startblock != 0);
+
+ key->refc.rc_startblock = rec->refc.rc_startblock;
+}
+
+STATIC void
+xfs_refcountbt_init_rec_from_cur(
+ struct xfs_btree_cur *cur,
+ union xfs_btree_rec *rec)
+{
+ ASSERT(cur->bc_rec.rc.rc_startblock != 0);
+
+ rec->refc.rc_startblock = cpu_to_be32(cur->bc_rec.rc.rc_startblock);
+ rec->refc.rc_blockcount = cpu_to_be32(cur->bc_rec.rc.rc_blockcount);
+ rec->refc.rc_refcount = cpu_to_be32(cur->bc_rec.rc.rc_refcount);
+}
+
+STATIC void
+xfs_refcountbt_init_ptr_from_cur(
+ struct xfs_btree_cur *cur,
+ union xfs_btree_ptr *ptr)
+{
+ struct xfs_agf *agf = XFS_BUF_TO_AGF(cur->bc_private.a.agbp);
+
+ ASSERT(cur->bc_private.a.agno == be32_to_cpu(agf->agf_seqno));
+ ASSERT(agf->agf_refcount_root != 0);
+
+ ptr->s = agf->agf_refcount_root;
+}
+
+STATIC __int64_t
+xfs_refcountbt_key_diff(
+ struct xfs_btree_cur *cur,
+ union xfs_btree_key *key)
+{
+ struct xfs_refcount_irec *rec = &cur->bc_rec.rc;
+ struct xfs_refcount_key *kp = &key->refc;
+
+ return (__int64_t)be32_to_cpu(kp->rc_startblock) - rec->rc_startblock;
+}
+
STATIC bool
xfs_refcountbt_verify(
struct xfs_buf *bp)
@@ -106,12 +253,62 @@ const struct xfs_buf_ops xfs_refcountbt_buf_ops = {
.verify_write = xfs_refcountbt_write_verify,
};
+#if defined(DEBUG) || defined(XFS_WARN)
+STATIC int
+xfs_refcountbt_keys_inorder(
+ struct xfs_btree_cur *cur,
+ union xfs_btree_key *k1,
+ union xfs_btree_key *k2)
+{
+ return be32_to_cpu(k1->refc.rc_startblock) <
+ be32_to_cpu(k2->refc.rc_startblock);
+}
+
+STATIC int
+xfs_refcountbt_recs_inorder(
+ struct xfs_btree_cur *cur,
+ union xfs_btree_rec *r1,
+ union xfs_btree_rec *r2)
+{
+ struct xfs_refcount_irec a, b;
+
+ int ret = be32_to_cpu(r1->refc.rc_startblock) +
+ be32_to_cpu(r1->refc.rc_blockcount) <=
+ be32_to_cpu(r2->refc.rc_startblock);
+ if (!ret) {
+ a.rc_startblock = be32_to_cpu(r1->refc.rc_startblock);
+ a.rc_blockcount = be32_to_cpu(r1->refc.rc_blockcount);
+ a.rc_refcount = be32_to_cpu(r1->refc.rc_refcount);
+ b.rc_startblock = be32_to_cpu(r2->refc.rc_startblock);
+ b.rc_blockcount = be32_to_cpu(r2->refc.rc_blockcount);
+ b.rc_refcount = be32_to_cpu(r2->refc.rc_refcount);
+ trace_xfs_refcount_rec_order_error(cur->bc_mp,
+ cur->bc_private.a.agno, &a, &b);
+ }
+
+ return ret;
+}
+#endif /* DEBUG */
+
static const struct xfs_btree_ops xfs_refcountbt_ops = {
.rec_len = sizeof(struct xfs_refcount_rec),
.key_len = sizeof(struct xfs_refcount_key),
.dup_cursor = xfs_refcountbt_dup_cursor,
+ .set_root = xfs_refcountbt_set_root,
+ .alloc_block = xfs_refcountbt_alloc_block,
+ .free_block = xfs_refcountbt_free_block,
+ .get_minrecs = xfs_refcountbt_get_minrecs,
+ .get_maxrecs = xfs_refcountbt_get_maxrecs,
+ .init_key_from_rec = xfs_refcountbt_init_key_from_rec,
+ .init_rec_from_cur = xfs_refcountbt_init_rec_from_cur,
+ .init_ptr_from_cur = xfs_refcountbt_init_ptr_from_cur,
+ .key_diff = xfs_refcountbt_key_diff,
.buf_ops = &xfs_refcountbt_buf_ops,
+#if defined(DEBUG) || defined(XFS_WARN)
+ .keys_inorder = xfs_refcountbt_keys_inorder,
+ .recs_inorder = xfs_refcountbt_recs_inorder,
+#endif
};
/*
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
