summaryrefslogtreecommitdiff
path: root/fs/ext4/mballoc.c
diff options
context:
space:
mode:
Diffstat (limited to 'fs/ext4/mballoc.c')
-rw-r--r--fs/ext4/mballoc.c61
1 files changed, 44 insertions, 17 deletions
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 252c168454c7..9e06334771a3 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -695,13 +695,10 @@ static int __mb_check_buddy(struct ext4_buddy *e4b, char *file,
for (i = 0; i < max; i++) {
if (mb_test_bit(i, buddy)) {
- /* only single bit in buddy2 may be 1 */
+ /* only single bit in buddy2 may be 0 */
if (!mb_test_bit(i << 1, buddy2)) {
MB_CHECK_ASSERT(
mb_test_bit((i<<1)+1, buddy2));
- } else if (!mb_test_bit((i << 1) + 1, buddy2)) {
- MB_CHECK_ASSERT(
- mb_test_bit(i << 1, buddy2));
}
continue;
}
@@ -2919,7 +2916,7 @@ const struct seq_operations ext4_mb_seq_groups_ops = {
int ext4_seq_mb_stats_show(struct seq_file *seq, void *offset)
{
- struct super_block *sb = (struct super_block *)seq->private;
+ struct super_block *sb = seq->private;
struct ext4_sb_info *sbi = EXT4_SB(sb);
seq_puts(seq, "mballoc:\n");
@@ -3498,7 +3495,7 @@ int ext4_mb_init(struct super_block *sb)
spin_lock_init(&lg->lg_prealloc_lock);
}
- if (blk_queue_nonrot(bdev_get_queue(sb->s_bdev)))
+ if (bdev_nonrot(sb->s_bdev))
sbi->s_mb_max_linear_groups = 0;
else
sbi->s_mb_max_linear_groups = MB_DEFAULT_LINEAR_LIMIT;
@@ -3629,7 +3626,7 @@ static inline int ext4_issue_discard(struct super_block *sb,
return __blkdev_issue_discard(sb->s_bdev,
(sector_t)discard_block << (sb->s_blocksize_bits - 9),
(sector_t)count << (sb->s_blocksize_bits - 9),
- GFP_NOFS, 0, biop);
+ GFP_NOFS, biop);
} else
return sb_issue_discard(sb, discard_block, count, GFP_NOFS, 0);
}
@@ -4107,6 +4104,15 @@ ext4_mb_normalize_request(struct ext4_allocation_context *ac,
size = size >> bsbits;
start = start_off >> bsbits;
+ /*
+ * For tiny groups (smaller than 8MB) the chosen allocation
+ * alignment may be larger than group size. Make sure the
+ * alignment does not move allocation to a different group which
+ * makes mballoc fail assertions later.
+ */
+ start = max(start, rounddown(ac->ac_o_ex.fe_logical,
+ (ext4_lblk_t)EXT4_BLOCKS_PER_GROUP(ac->ac_sb)));
+
/* don't cover already allocated blocks in selected range */
if (ar->pleft && start <= ar->lleft) {
size -= ar->lleft + 1 - start;
@@ -4179,7 +4185,22 @@ ext4_mb_normalize_request(struct ext4_allocation_context *ac,
}
rcu_read_unlock();
- if (start + size <= ac->ac_o_ex.fe_logical &&
+ /*
+ * In this function "start" and "size" are normalized for better
+ * alignment and length such that we could preallocate more blocks.
+ * This normalization is done such that original request of
+ * ac->ac_o_ex.fe_logical & fe_len should always lie within "start" and
+ * "size" boundaries.
+ * (Note fe_len can be relaxed since FS block allocation API does not
+ * provide gurantee on number of contiguous blocks allocation since that
+ * depends upon free space left, etc).
+ * In case of inode pa, later we use the allocated blocks
+ * [pa_start + fe_logical - pa_lstart, fe_len/size] from the preallocated
+ * range of goal/best blocks [start, size] to put it at the
+ * ac_o_ex.fe_logical extent of this inode.
+ * (See ext4_mb_use_inode_pa() for more details)
+ */
+ if (start + size <= ac->ac_o_ex.fe_logical ||
start > ac->ac_o_ex.fe_logical) {
ext4_msg(ac->ac_sb, KERN_ERR,
"start %lu, size %lu, fe_logical %lu",
@@ -6398,6 +6419,7 @@ __releases(ext4_group_lock_ptr(sb, e4b->bd_group))
* @start: first group block to examine
* @max: last group block to examine
* @minblocks: minimum extent block count
+ * @set_trimmed: set the trimmed flag if at least one block is trimmed
*
* ext4_trim_all_free walks through group's block bitmap searching for free
* extents. When the free extent is found, mark it as used in group buddy
@@ -6407,7 +6429,7 @@ __releases(ext4_group_lock_ptr(sb, e4b->bd_group))
static ext4_grpblk_t
ext4_trim_all_free(struct super_block *sb, ext4_group_t group,
ext4_grpblk_t start, ext4_grpblk_t max,
- ext4_grpblk_t minblocks)
+ ext4_grpblk_t minblocks, bool set_trimmed)
{
struct ext4_buddy e4b;
int ret;
@@ -6426,7 +6448,7 @@ ext4_trim_all_free(struct super_block *sb, ext4_group_t group,
if (!EXT4_MB_GRP_WAS_TRIMMED(e4b.bd_info) ||
minblocks < EXT4_SB(sb)->s_last_trim_minblks) {
ret = ext4_try_to_trim_range(sb, &e4b, start, max, minblocks);
- if (ret >= 0)
+ if (ret >= 0 && set_trimmed)
EXT4_MB_GRP_SET_TRIMMED(e4b.bd_info);
} else {
ret = 0;
@@ -6455,7 +6477,7 @@ ext4_trim_all_free(struct super_block *sb, ext4_group_t group,
*/
int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range)
{
- struct request_queue *q = bdev_get_queue(sb->s_bdev);
+ unsigned int discard_granularity = bdev_discard_granularity(sb->s_bdev);
struct ext4_group_info *grp;
ext4_group_t group, first_group, last_group;
ext4_grpblk_t cnt = 0, first_cluster, last_cluster;
@@ -6463,6 +6485,7 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range)
ext4_fsblk_t first_data_blk =
le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block);
ext4_fsblk_t max_blks = ext4_blocks_count(EXT4_SB(sb)->s_es);
+ bool whole_group, eof = false;
int ret = 0;
start = range->start >> sb->s_blocksize_bits;
@@ -6475,14 +6498,16 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range)
range->len < sb->s_blocksize)
return -EINVAL;
/* No point to try to trim less than discard granularity */
- if (range->minlen < q->limits.discard_granularity) {
+ if (range->minlen < discard_granularity) {
minlen = EXT4_NUM_B2C(EXT4_SB(sb),
- q->limits.discard_granularity >> sb->s_blocksize_bits);
+ discard_granularity >> sb->s_blocksize_bits);
if (minlen > EXT4_CLUSTERS_PER_GROUP(sb))
goto out;
}
- if (end >= max_blks)
+ if (end >= max_blks - 1) {
end = max_blks - 1;
+ eof = true;
+ }
if (end <= first_data_blk)
goto out;
if (start < first_data_blk)
@@ -6496,6 +6521,7 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range)
/* end now represents the last cluster to discard in this group */
end = EXT4_CLUSTERS_PER_GROUP(sb) - 1;
+ whole_group = true;
for (group = first_group; group <= last_group; group++) {
grp = ext4_get_group_info(sb, group);
@@ -6512,12 +6538,13 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range)
* change it for the last group, note that last_cluster is
* already computed earlier by ext4_get_group_no_and_offset()
*/
- if (group == last_group)
+ if (group == last_group) {
end = last_cluster;
-
+ whole_group = eof ? true : end == EXT4_CLUSTERS_PER_GROUP(sb) - 1;
+ }
if (grp->bb_free >= minlen) {
cnt = ext4_trim_all_free(sb, group, first_cluster,
- end, minlen);
+ end, minlen, whole_group);
if (cnt < 0) {
ret = cnt;
break;
generated by cgit v1.2.3 (git 2.46.0) at 2025-08-21 09:29:58 +0000