On Thursday 31 July 2014 at 14:19:43, Ivan Shapovalov wrote:
> [...]
>
> Ivan Shapovalov (6):
> reiser4: fix reiser4_post_{commit,write_back}_hook() and their invocations.
> reiser4: make space_allocator's check_blocks() reusable.
> reiser4: add an implementation of "block lists", splitted off the discard code.
> reiser4: blocknr_list: use kmem_cache instead of kmalloc for allocating entries.
> reiser4: blocknr_set: use kmem_cache instead of kmalloc for allocating entries.
> reiser4: discard support: initial implementation using blocknr_list, without extent padding.
>
> [...]
This is a diff between previous two patchsets and this one, for ease of reviewing.
diff --git a/fs/reiser4/block_alloc.c b/fs/reiser4/block_alloc.c
index 98080a1..324b11c 100644
--- a/fs/reiser4/block_alloc.c
+++ b/fs/reiser4/block_alloc.c
@@ -9,6 +9,7 @@ reiser4/README */
#include "block_alloc.h"
#include "tree.h"
#include "super.h"
+#include "discard.h"
#include <linux/types.h> /* for __u?? */
#include <linux/fs.h> /* for struct super_block */
@@ -1144,7 +1145,7 @@ void reiser4_post_write_back_hook(void)
} while (ret == -E_REPEAT);
if (ret) {
- warning("intelfx-8", "discard atom failed (%ld)", ret);
+ warning("intelfx-8", "discard atom failed (%d)", ret);
}
atom = get_current_atom_locked();
diff --git a/fs/reiser4/discard.c b/fs/reiser4/discard.c
index 8442619..7a07afc 100644
--- a/fs/reiser4/discard.c
+++ b/fs/reiser4/discard.c
@@ -53,23 +53,21 @@
* the atom's delete set becomes "the discard set" -- list of blocks that have
* to be considered for discarding.
*
- * On atom commit we will generate a minimal superset of the discard set,
- * comprised of whole erase units.
- *
* Discarding is performed before completing deferred deallocations, hence all
* extents in the discard set are still marked as allocated and cannot contain
* any data. Thus we can avoid any checks for blocks directly present in the
* discard set.
*
- * However, we pad each extent from both sides to erase unit boundaries, and
- * these paddings still have to be checked if they fall outside of initial
- * extent (may not happen if block size > erase unit size).
+ * For now, we don't perform "padding" of extents to erase unit boundaries.
+ * This means if extents are not aligned with the device's erase unit lattice,
+ * the partial erase units at head and tail of extents are truncated by kernel
+ * (in blkdev_issue_discard()).
*
* So, at commit time the following actions take place:
* - delete sets are merged to form the discard set;
* - elements of the discard set are sorted;
* - the discard set is iterated, joining any adjacent extents;
- * - <TODO>
+ * - for each extent, a single call to blkdev_issue_discard() is done.
*/
#include "discard.h"
@@ -98,92 +96,20 @@ static int discard_extent(txn_atom *atom UNUSED_ARG,
{
struct super_block *sb = reiser4_get_current_sb();
struct block_device *bdev = sb->s_bdev;
- struct queue_limits *limits = &bdev_get_queue(bdev)->limits;
- sector_t extent_start_sec, extent_end_sec,
- unit_sec, request_start_sec = 0, request_len_sec = 0;
- reiser4_block_nr unit_start_blk, unit_len_blk;
- int ret, erase_unit_counter = 0;
+ sector_t extent_start_sec, extent_len_sec;
const int sec_per_blk = sb->s_blocksize >> 9;
- /* from blkdev_issue_discard():
- * Zero-sector (unknown) and one-sector granularities are the same. */
- const int granularity = max(limits->discard_granularity >> 9, 1U);
- const int alignment = (bdev_discard_alignment(bdev) >> 9) % granularity;
-
/* we assume block = N * sector */
assert("intelfx-7", sec_per_blk > 0);
/* convert extent to sectors */
extent_start_sec = *start * sec_per_blk;
- extent_end_sec = (*start + *len) * sec_per_blk;
+ extent_len_sec = *len * sec_per_blk;
- /* round down extent start sector to an erase unit boundary */
- unit_sec = extent_start_sec;
- if (granularity > 1) {
- sector_t tmp = extent_start_sec - alignment;
- unit_sec -= sector_div(tmp, granularity);
- }
-
- /* iterate over erase units in the extent */
- do {
- /* considering erase unit:
- * [unit_sec; unit_sec + granularity) */
-
- /* calculate block range for erase unit:
- * [unit_start_blk; unit_start_blk+unit_len_blk) */
- unit_start_blk = unit_sec;
- do_div(unit_start_blk, sec_per_blk);
-
- if (granularity > 1) {
- unit_len_blk = unit_sec + granularity - 1;
- do_div(unit_len_blk, sec_per_blk);
- ++unit_len_blk;
-
- assert("intelfx-22", unit_len_blk > unit_start_blk);
-
- unit_len_blk -= unit_start_blk;
- } else {
- unit_len_blk = 1;
- }
-
- if (reiser4_check_blocks(&unit_start_blk, &unit_len_blk, 0)) {
- /* OK. Add this unit to the accumulator.
- * We accumulate discard units to call blkdev_issue_discard()
- * not too frequently. */
-
- if (request_len_sec > 0) {
- request_len_sec += granularity;
- } else {
- request_start_sec = unit_sec;
- request_len_sec = granularity;
- }
- } else {
- /* This unit can't be discarded. Discard what's been accumulated
- * so far. */
- if (request_len_sec > 0) {
- ret = __discard_extent(bdev, request_start_sec, request_len_sec);
- if (ret != 0) {
- return ret;
- }
- request_len_sec = 0;
- }
- }
-
- unit_sec += granularity;
- ++erase_unit_counter;
- } while (unit_sec < extent_end_sec);
-
- /* Discard the last accumulated request. */
- if (request_len_sec > 0) {
- ret = __discard_extent(bdev, request_start_sec, request_len_sec);
- if (ret != 0) {
- return ret;
- }
- }
-
- return 0;
+ /* discard the extent, don't pad it to erase unit boundaries for now */
+ return __discard_extent(bdev, extent_start_sec, extent_len_sec);
}
int discard_atom(txn_atom *atom, struct list_head *processed_set)
@@ -201,6 +127,7 @@ int discard_atom(txn_atom *atom, struct list_head *processed_set)
if (list_empty(&atom->discard.delete_set)) {
/* Nothing left to discard. */
+ spin_unlock_atom(atom);
return 0;
}
diff --git a/fs/reiser4/plugin/space/bitmap.c b/fs/reiser4/plugin/space/bitmap.c
index 03bc5e7..3da3f6b 100644
--- a/fs/reiser4/plugin/space/bitmap.c
+++ b/fs/reiser4/plugin/space/bitmap.c
@@ -1259,8 +1259,7 @@ int reiser4_check_blocks_bitmap(const reiser4_block_nr * start,
reiser4_block_nr end;
bmap_nr_t bmap, end_bmap;
- bmap_off_t offset;
- bmap_off_t end_offset;
+ bmap_off_t offset, end_offset;
const bmap_off_t max_offset = bmap_bit_count(super->s_blocksize);
assert("intelfx-9", start != NULL);
@@ -1270,9 +1269,8 @@ int reiser4_check_blocks_bitmap(const reiser4_block_nr * start,
check_block_range(start, len);
end = *start + *len - 1;
} else {
- /* end is used as temporary len here */
- end = 1;
- check_block_range(start, &end);
+ /* on next line, end is used as temporary len for check_block_range() */
+ end = 1; check_block_range(start, &end);
end = *start;
}
diff --git a/fs/reiser4/txnmgr.h b/fs/reiser4/txnmgr.h
index 05990d8..72b84a2 100644
--- a/fs/reiser4/txnmgr.h
+++ b/fs/reiser4/txnmgr.h
@@ -256,8 +256,10 @@ struct txn_atom {
} nodiscard;
struct {
- /* The atom's delete set. It collects block numbers which were
- deallocated with BA_DEFER, i. e. of ordinary nodes. */
+ /* The atom's delete set. It collects all blocks that have been
+ deallocated (both immediate and deferred) during the transaction.
+ These blocks are considered for discarding at commit time.
+ For details see discard.c */
struct list_head delete_set;
} discard;
};
Thanks,
--
Ivan Shapovalov / intelfx /
Attachment:
signature.asc
Description: This is a digitally signed message part.
