6.6-stable review patch. If anyone has any objections, please let me know.
------------------
From: Naohiro Aota <naohiro.aota@xxxxxxx>
commit 8cd44dd1d17a23d5cc8c443c659ca57aa76e2fa5 upstream.
When btrfs makes a block group read-only, it adds all free regions in the
block group to space_info->bytes_readonly. That free space excludes
reserved and pinned regions. OTOH, when btrfs makes the block group
read-write again, it moves all the unused regions into the block group's
zone_unusable. That unused region includes reserved and pinned regions.
As a result, it counts too much zone_unusable bytes.
Fortunately (or unfortunately), having erroneous zone_unusable does not
affect the calculation of space_info->bytes_readonly, because free
space (num_bytes in btrfs_dec_block_group_ro) calculation is done based on
the erroneous zone_unusable and it reduces the num_bytes just to cancel the
error.
This behavior can be easily discovered by adding a WARN_ON to check e.g,
"bg->pinned > 0" in btrfs_dec_block_group_ro(), and running fstests test
case like btrfs/282.
Fix it by properly considering pinned and reserved in
btrfs_dec_block_group_ro(). Also, add a WARN_ON and introduce
btrfs_space_info_update_bytes_zone_unusable() to catch a similar mistake.
Fixes: 169e0da91a21 ("btrfs: zoned: track unusable bytes for zones")
CC: stable@xxxxxxxxxxxxxxx # 5.15+
Signed-off-by: Naohiro Aota <naohiro.aota@xxxxxxx>
Reviewed-by: Josef Bacik <josef@xxxxxxxxxxxxxx>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@xxxxxxx>
Signed-off-by: David Sterba <dsterba@xxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
fs/btrfs/block-group.c | 13 ++++++++-----
fs/btrfs/extent-tree.c | 3 ++-
fs/btrfs/free-space-cache.c | 4 +++-
fs/btrfs/space-info.c | 2 +-
fs/btrfs/space-info.h | 1 +
include/trace/events/btrfs.h | 8 ++++++++
6 files changed, 23 insertions(+), 8 deletions(-)
--- a/fs/btrfs/block-group.c
+++ b/fs/btrfs/block-group.c
@@ -1214,8 +1214,8 @@ int btrfs_remove_block_group(struct btrf
block_group->space_info->total_bytes -= block_group->length;
block_group->space_info->bytes_readonly -=
(block_group->length - block_group->zone_unusable);
- block_group->space_info->bytes_zone_unusable -=
- block_group->zone_unusable;
+ btrfs_space_info_update_bytes_zone_unusable(fs_info, block_group->space_info,
+ -block_group->zone_unusable);
block_group->space_info->disk_total -= block_group->length * factor;
spin_unlock(&block_group->space_info->lock);
@@ -1399,7 +1399,8 @@ static int inc_block_group_ro(struct btr
if (btrfs_is_zoned(cache->fs_info)) {
/* Migrate zone_unusable bytes to readonly */
sinfo->bytes_readonly += cache->zone_unusable;
- sinfo->bytes_zone_unusable -= cache->zone_unusable;
+ btrfs_space_info_update_bytes_zone_unusable(cache->fs_info, sinfo,
+ -cache->zone_unusable);
cache->zone_unusable = 0;
}
cache->ro++;
@@ -3023,9 +3024,11 @@ void btrfs_dec_block_group_ro(struct btr
if (btrfs_is_zoned(cache->fs_info)) {
/* Migrate zone_unusable bytes back */
cache->zone_unusable =
- (cache->alloc_offset - cache->used) +
+ (cache->alloc_offset - cache->used - cache->pinned -
+ cache->reserved) +
(cache->length - cache->zone_capacity);
- sinfo->bytes_zone_unusable += cache->zone_unusable;
+ btrfs_space_info_update_bytes_zone_unusable(cache->fs_info, sinfo,
+ cache->zone_unusable);
sinfo->bytes_readonly -= cache->zone_unusable;
}
num_bytes = cache->length - cache->reserved -
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -2749,7 +2749,8 @@ static int unpin_extent_range(struct btr
readonly = true;
} else if (btrfs_is_zoned(fs_info)) {
/* Need reset before reusing in a zoned block group */
- space_info->bytes_zone_unusable += len;
+ btrfs_space_info_update_bytes_zone_unusable(fs_info, space_info,
+ len);
readonly = true;
}
spin_unlock(&cache->lock);
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -2721,8 +2721,10 @@ static int __btrfs_add_free_space_zoned(
* If the block group is read-only, we should account freed space into
* bytes_readonly.
*/
- if (!block_group->ro)
+ if (!block_group->ro) {
block_group->zone_unusable += to_unusable;
+ WARN_ON(block_group->zone_unusable > block_group->length);
+ }
spin_unlock(&ctl->tree_lock);
if (!used) {
spin_lock(&block_group->lock);
--- a/fs/btrfs/space-info.c
+++ b/fs/btrfs/space-info.c
@@ -312,7 +312,7 @@ void btrfs_add_bg_to_space_info(struct b
found->bytes_used += block_group->used;
found->disk_used += block_group->used * factor;
found->bytes_readonly += block_group->bytes_super;
- found->bytes_zone_unusable += block_group->zone_unusable;
+ btrfs_space_info_update_bytes_zone_unusable(info, found, block_group->zone_unusable);
if (block_group->length > 0)
found->full = 0;
btrfs_try_granting_tickets(info, found);
--- a/fs/btrfs/space-info.h
+++ b/fs/btrfs/space-info.h
@@ -197,6 +197,7 @@ btrfs_space_info_update_##name(struct bt
DECLARE_SPACE_INFO_UPDATE(bytes_may_use, "space_info");
DECLARE_SPACE_INFO_UPDATE(bytes_pinned, "pinned");
+DECLARE_SPACE_INFO_UPDATE(bytes_zone_unusable, "zone_unusable");
int btrfs_init_space_info(struct btrfs_fs_info *fs_info);
void btrfs_add_bg_to_space_info(struct btrfs_fs_info *info,
--- a/include/trace/events/btrfs.h
+++ b/include/trace/events/btrfs.h
@@ -2430,6 +2430,14 @@ DEFINE_EVENT(btrfs__space_info_update, u
TP_ARGS(fs_info, sinfo, old, diff)
);
+DEFINE_EVENT(btrfs__space_info_update, update_bytes_zone_unusable,
+
+ TP_PROTO(const struct btrfs_fs_info *fs_info,
+ const struct btrfs_space_info *sinfo, u64 old, s64 diff),
+
+ TP_ARGS(fs_info, sinfo, old, diff)
+);
+
DECLARE_EVENT_CLASS(btrfs_raid56_bio,
TP_PROTO(const struct btrfs_raid_bio *rbio,
