FAILED: patch "[PATCH] btrfs: zoned: requeue to unused block group list if zone" failed to apply to 6.12-stable tree
The patch below does not apply to the 6.12-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to stable@vger.kernel.org.
To reproduce the conflict and resubmit, you may use the following commands:
git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.12.y git checkout FETCH_HEAD git cherry-pick -x 62be7afcc13b2727bdc6a4c91aefed6b452e6ecc # <resolve conflicts, build, test, etc.> git commit -s git send-email --to 'stable@vger.kernel.org' --in-reply-to '2025081853-parrot-skeleton-78e1@gregkh' --subject-prefix 'PATCH 6.12.y' HEAD^..
Possible dependencies:
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
From 62be7afcc13b2727bdc6a4c91aefed6b452e6ecc Mon Sep 17 00:00:00 2001 From: Naohiro Aota naohiro.aota@wdc.com Date: Sun, 29 Jun 2025 23:18:29 +0900 Subject: [PATCH] btrfs: zoned: requeue to unused block group list if zone finish failed
btrfs_zone_finish() can fail for several reason. If it is -EAGAIN, we need to try it again later. So, put the block group to the retry list properly.
Failing to do so will keep the removable block group intact until remount and can causes unnecessary ENOSPC.
Fixes: 74e91b12b115 ("btrfs: zoned: zone finish unused block group") CC: stable@vger.kernel.org # 6.1+ Reviewed-by: Johannes Thumshirn johannes.thumshirn@wdc.com Signed-off-by: Naohiro Aota naohiro.aota@wdc.com Signed-off-by: David Sterba dsterba@suse.com
diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c index 3ddf9fe52b9d..47c6d040176c 100644 --- a/fs/btrfs/block-group.c +++ b/fs/btrfs/block-group.c @@ -1639,8 +1639,10 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) ret = btrfs_zone_finish(block_group); if (ret < 0) { btrfs_dec_block_group_ro(block_group); - if (ret == -EAGAIN) + if (ret == -EAGAIN) { + btrfs_link_bg_list(block_group, &retry_list); ret = 0; + } goto next; }
From: Boris Burkov boris@bur.io
[ Upstream commit 7cbce3cb4c5cfffd8b08f148e2136afc1ec1ba94 ]
All other users of the bg_list list_head increment the refcount when adding to a list and decrement it when deleting from the list. Just for the sake of uniformity and to try to avoid refcounting bugs, do it for this list as well.
This does not fix any known ref-counting bug, as the reference belongs to a single task (trans_handle is not shared and this represents trans_handle->new_bgs linkage) and will not lose its original refcount while that thread is running. And BLOCK_GROUP_FLAG_NEW protects against ref-counting errors "moving" the block group to the unused list without taking a ref.
With that said, I still believe it is simpler to just hold the extra ref count for this list user as well.
Reviewed-by: Filipe Manana fdmanana@suse.com Signed-off-by: Boris Burkov boris@bur.io Signed-off-by: David Sterba dsterba@suse.com Stable-dep-of: 62be7afcc13b ("btrfs: zoned: requeue to unused block group list if zone finish failed") Signed-off-by: Sasha Levin sashal@kernel.org --- fs/btrfs/block-group.c | 2 ++ fs/btrfs/transaction.c | 1 + 2 files changed, 3 insertions(+)
diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c index dd35e29d8082..7a18d862821b 100644 --- a/fs/btrfs/block-group.c +++ b/fs/btrfs/block-group.c @@ -2784,6 +2784,7 @@ void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans) spin_lock(&fs_info->unused_bgs_lock); list_del_init(&block_group->bg_list); clear_bit(BLOCK_GROUP_FLAG_NEW, &block_group->runtime_flags); + btrfs_put_block_group(block_group); spin_unlock(&fs_info->unused_bgs_lock);
/* @@ -2922,6 +2923,7 @@ struct btrfs_block_group *btrfs_make_block_group(struct btrfs_trans_handle *tran } #endif
+ btrfs_get_block_group(cache); list_add_tail(&cache->bg_list, &trans->new_bgs); btrfs_inc_delayed_refs_rsv_bg_inserts(fs_info);
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 24806e19c7c4..d8f7232fd9b7 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -2111,6 +2111,7 @@ static void btrfs_cleanup_pending_block_groups(struct btrfs_trans_handle *trans) */ spin_lock(&fs_info->unused_bgs_lock); list_del_init(&block_group->bg_list); + btrfs_put_block_group(block_group); spin_unlock(&fs_info->unused_bgs_lock); } }
From: Boris Burkov boris@bur.io
[ Upstream commit 0497dfba98c00edbc7af12d53c0b1138eb318bf7 ]
Similar to mark_bg_unused() and mark_bg_to_reclaim(), we have a few places that use bg_list with refcounting, mostly for retrying failures to reclaim/delete unused.
These have custom logic for handling locking and refcounting the bg_list properly, but they actually all want to do the same thing, so pull that logic out into a helper. Unfortunately, mark_bg_unused() does still need the NEW flag to avoid prematurely marking stuff unused (even if refcount is fine, we don't want to mess with bg creation), so it cannot use the new helper.
Reviewed-by: Filipe Manana fdmanana@suse.com Signed-off-by: Boris Burkov boris@bur.io Signed-off-by: David Sterba dsterba@suse.com Stable-dep-of: 62be7afcc13b ("btrfs: zoned: requeue to unused block group list if zone finish failed") Signed-off-by: Sasha Levin sashal@kernel.org --- fs/btrfs/block-group.c | 55 ++++++++++++++++++++++++------------------ 1 file changed, 31 insertions(+), 24 deletions(-)
diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c index 7a18d862821b..841954bd788f 100644 --- a/fs/btrfs/block-group.c +++ b/fs/btrfs/block-group.c @@ -1459,6 +1459,32 @@ static bool clean_pinned_extents(struct btrfs_trans_handle *trans, return ret == 0; }
+/* + * Link the block_group to a list via bg_list. + * + * @bg: The block_group to link to the list. + * @list: The list to link it to. + * + * Use this rather than list_add_tail() directly to ensure proper respect + * to locking and refcounting. + * + * Returns: true if the bg was linked with a refcount bump and false otherwise. + */ +static bool btrfs_link_bg_list(struct btrfs_block_group *bg, struct list_head *list) +{ + struct btrfs_fs_info *fs_info = bg->fs_info; + bool added = false; + + spin_lock(&fs_info->unused_bgs_lock); + if (list_empty(&bg->bg_list)) { + btrfs_get_block_group(bg); + list_add_tail(&bg->bg_list, list); + added = true; + } + spin_unlock(&fs_info->unused_bgs_lock); + return added; +} + /* * Process the unused_bgs list and remove any that don't have any allocated * space inside of them. @@ -1574,8 +1600,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) * drop under the "next" label for the * fs_info->unused_bgs list. */ - btrfs_get_block_group(block_group); - list_add_tail(&block_group->bg_list, &retry_list); + btrfs_link_bg_list(block_group, &retry_list);
trace_btrfs_skip_unused_block_group(block_group); spin_unlock(&block_group->lock); @@ -1948,20 +1973,8 @@ void btrfs_reclaim_bgs_work(struct work_struct *work) spin_unlock(&space_info->lock);
next: - if (ret && !READ_ONCE(space_info->periodic_reclaim)) { - /* Refcount held by the reclaim_bgs list after splice. */ - spin_lock(&fs_info->unused_bgs_lock); - /* - * This block group might be added to the unused list - * during the above process. Move it back to the - * reclaim list otherwise. - */ - if (list_empty(&bg->bg_list)) { - btrfs_get_block_group(bg); - list_add_tail(&bg->bg_list, &retry_list); - } - spin_unlock(&fs_info->unused_bgs_lock); - } + if (ret && !READ_ONCE(space_info->periodic_reclaim)) + btrfs_link_bg_list(bg, &retry_list); btrfs_put_block_group(bg);
mutex_unlock(&fs_info->reclaim_bgs_lock); @@ -2001,13 +2014,8 @@ void btrfs_mark_bg_to_reclaim(struct btrfs_block_group *bg) { struct btrfs_fs_info *fs_info = bg->fs_info;
- spin_lock(&fs_info->unused_bgs_lock); - if (list_empty(&bg->bg_list)) { - btrfs_get_block_group(bg); + if (btrfs_link_bg_list(bg, &fs_info->reclaim_bgs)) trace_btrfs_add_reclaim_block_group(bg); - list_add_tail(&bg->bg_list, &fs_info->reclaim_bgs); - } - spin_unlock(&fs_info->unused_bgs_lock); }
static int read_bg_from_eb(struct btrfs_fs_info *fs_info, const struct btrfs_key *key, @@ -2923,8 +2931,7 @@ struct btrfs_block_group *btrfs_make_block_group(struct btrfs_trans_handle *tran } #endif
- btrfs_get_block_group(cache); - list_add_tail(&cache->bg_list, &trans->new_bgs); + btrfs_link_bg_list(cache, &trans->new_bgs); btrfs_inc_delayed_refs_rsv_bg_inserts(fs_info);
set_avail_alloc_bits(fs_info, type);
From: Naohiro Aota naohiro.aota@wdc.com
[ Upstream commit 62be7afcc13b2727bdc6a4c91aefed6b452e6ecc ]
btrfs_zone_finish() can fail for several reason. If it is -EAGAIN, we need to try it again later. So, put the block group to the retry list properly.
Failing to do so will keep the removable block group intact until remount and can causes unnecessary ENOSPC.
Fixes: 74e91b12b115 ("btrfs: zoned: zone finish unused block group") CC: stable@vger.kernel.org # 6.1+ Reviewed-by: Johannes Thumshirn johannes.thumshirn@wdc.com Signed-off-by: Naohiro Aota naohiro.aota@wdc.com Signed-off-by: David Sterba dsterba@suse.com Signed-off-by: Sasha Levin sashal@kernel.org --- fs/btrfs/block-group.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c index 841954bd788f..13b223eef995 100644 --- a/fs/btrfs/block-group.c +++ b/fs/btrfs/block-group.c @@ -1623,8 +1623,10 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) ret = btrfs_zone_finish(block_group); if (ret < 0) { btrfs_dec_block_group_ro(block_group); - if (ret == -EAGAIN) + if (ret == -EAGAIN) { + btrfs_link_bg_list(block_group, &retry_list); ret = 0; + } goto next; }
linux-stable-mirror@lists.linaro.org
-
gregkh＠linuxfoundation.org -
Sasha Levin