FAILED: patch "[PATCH] f2fs: fix to migrate all curseg types during" failed to apply to 5.10-stable tree

FAILED: patch "[PATCH] f2fs: fix to migrate all curseg types during" failed to apply to 5.10-stable tree

[gregkh]

The patch below does not apply to the 5.10-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-5.10.y
git checkout FETCH_HEAD
git cherry-pick -x 8ec06f50ddd8d201bd7e55b896ae28ed9d4cb7d1
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<stable@vger.kernel.org>' --in-reply-to '2026090947-bankable-stroller-237b@gregkh' --subject-prefix 'PATCH 5.10.y' 'HEAD^..'

Possible dependencies:



thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From 8ec06f50ddd8d201bd7e55b896ae28ed9d4cb7d1 Mon Sep 17 00:00:00 2001
From: Daeho Jeong <daehojeong@google.com>
Date: Wed, 19 Aug 2026 11:06:35 -0700
Subject: [PATCH] f2fs: fix to migrate all curseg types during
 free_segment_range

In free_segment_range(), the curseg evacuation loop only iterates up to
NR_CURSEG_PERSIST_TYPE (0..5), missing non-persistent in-memory curseg
types such as CURSEG_COLD_DATA_PINNED and CURSEG_ALL_DATA_ATGC.

Even though these in-memory curseg types are not saved in the on-disk
checkpoint header, they still occupy active physical segments at runtime.
If an active in-memory curseg happens to be allocated within the segment
range being truncated during filesystem shrink, failing to evacuate it
will cause subsequent writes to the curseg attempting out-of-bounds I/O
on the truncated storage range.

Fix this by expanding the curseg evacuation loop upper bound to
NR_CURSEG_TYPE to ensure all active curseg types are safely migrated
out of the target range.

Fixes: d0b9e42ab615 ("f2fs: introduce inmem curseg")
Cc: stable@vger.kernel.org
Signed-off-by: Daeho Jeong <daehojeong@google.com>
Signed-off-by: Sunmin Jeong <s_min.jeong@samsung.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>

diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index 3215630c353d..5917ee917d87 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -2240,7 +2240,7 @@ static int free_segment_range(struct f2fs_sb_info *sbi,
 	f2fs_reset_gc_victim_resource(sbi, start, end);
 
 	/* Move out cursegs from the target range */
-	for (type = CURSEG_HOT_DATA; type < NR_CURSEG_PERSIST_TYPE; type++) {
+	for (type = CURSEG_HOT_DATA; type < NR_CURSEG_TYPE; type++) {
 		err = f2fs_allocate_segment_for_resize(sbi, type, start, end);
 		if (err)
 			goto out;

[PATCH 5.10.y 1/3] f2fs: delete f2fs_get_new_segment() declaration

[Sasha Levin]

From: Chao Yu <chao@kernel.org>

[ Upstream commit e24e8333d0f3110473c6e6a20afd51911984e901 ]

Commit 093749e296e2 ("f2fs: support age threshold based garbage
collection") added this declaration, but w/ definition, delete
it.

Signed-off-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Stable-dep-of: 8ec06f50ddd8 ("f2fs: fix to migrate all curseg types during free_segment_range")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/f2fs/f2fs.h | 2 --
 1 file changed, 2 deletions(-)

diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 948e3512f60ff..c99436e829af1 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -3365,8 +3365,6 @@ bool f2fs_segment_has_free_slot(struct f2fs_sb_info *sbi, int segno);
 void f2fs_init_inmem_curseg(struct f2fs_sb_info *sbi);
 void f2fs_save_inmem_curseg(struct f2fs_sb_info *sbi);
 void f2fs_restore_inmem_curseg(struct f2fs_sb_info *sbi);
-void f2fs_get_new_segment(struct f2fs_sb_info *sbi,
-			unsigned int *newseg, bool new_sec, int dir);
 void f2fs_allocate_segment_for_resize(struct f2fs_sb_info *sbi, int type,
 					unsigned int start, unsigned int end);
 void f2fs_allocate_new_section(struct f2fs_sb_info *sbi, int type, bool force);
-- 
2.53.0

[PATCH 5.10.y 2/3] f2fs: stop checkpoint when get a out-of-bounds segment

[Sasha Levin]

From: Zhiguo Niu <zhiguo.niu@unisoc.com>

[ Upstream commit f9e28904e6442019043a8e94ec6747a064d06003 ]

There is low probability that an out-of-bounds segment will be got
on a small-capacity device. In order to prevent subsequent write requests
allocating block address from this invalid segment, which may cause
unexpected issue, stop checkpoint should be performed.

Also introduce a new stop cp reason: STOP_CP_REASON_NO_SEGMENT.

Note, f2fs_stop_checkpoint(, false) is complex and it may sleep, so we should
move it outside segmap_lock spinlock coverage in get_new_segment().

Signed-off-by: Zhiguo Niu <zhiguo.niu@unisoc.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>

[5.15 dependency adaptation]
Keep the existing left/right allocation policy and stop checkpointing after
releasing segmap_lock when either free-section search is exhausted. Use the
existing two-argument f2fs_stop_checkpoint() API; omit the stop-reason enum
hunk because this branch has no checkpoint-reason recording infrastructure.

Prepare the existing resize loop for 8ec06f50ddd8 without adding functions:
make f2fs_allocate_segment_for_resize() return -EIO when checkpointing has
failed, and have free_segment_range() unwind on that error. The older
allocation helpers do not return errors, so use the existing checkpoint
error flag rather than importing their newer interfaces. Leave the loop's
NR_CURSEG_PERSIST_TYPE bound for the target commit to change.

[ sashal: Reduced backport -- upstream f9e28904e6442 touches 2 file(s), this
  backport carries 3. Not backported here:
  include/linux/f2fs_fs.h
  This note is generated from the file lists only; see the resolution record
  for the reasoning. ]

Stable-dep-of: 8ec06f50ddd8 ("f2fs: fix to migrate all curseg types during free_segment_range")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/f2fs/f2fs.h    |  2 +-
 fs/f2fs/gc.c      |  7 +++++--
 fs/f2fs/segment.c | 20 +++++++++++++++++---
 3 files changed, 23 insertions(+), 6 deletions(-)

diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index c99436e829af1..b9fb11708392e 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -3365,7 +3365,7 @@ bool f2fs_segment_has_free_slot(struct f2fs_sb_info *sbi, int segno);
 void f2fs_init_inmem_curseg(struct f2fs_sb_info *sbi);
 void f2fs_save_inmem_curseg(struct f2fs_sb_info *sbi);
 void f2fs_restore_inmem_curseg(struct f2fs_sb_info *sbi);
-void f2fs_allocate_segment_for_resize(struct f2fs_sb_info *sbi, int type,
+int f2fs_allocate_segment_for_resize(struct f2fs_sb_info *sbi, int type,
 					unsigned int start, unsigned int end);
 void f2fs_allocate_new_section(struct f2fs_sb_info *sbi, int type, bool force);
 void f2fs_allocate_new_segments(struct f2fs_sb_info *sbi);
diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index 379839f5b4b74..33a2079b77c2b 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -1877,8 +1877,11 @@ static int free_segment_range(struct f2fs_sb_info *sbi,
 	mutex_unlock(&DIRTY_I(sbi)->seglist_lock);
 
 	/* Move out cursegs from the target range */
-	for (type = CURSEG_HOT_DATA; type < NR_CURSEG_PERSIST_TYPE; type++)
-		f2fs_allocate_segment_for_resize(sbi, type, start, end);
+	for (type = CURSEG_HOT_DATA; type < NR_CURSEG_PERSIST_TYPE; type++) {
+		err = f2fs_allocate_segment_for_resize(sbi, type, start, end);
+		if (err)
+			goto out;
+	}
 
 	/* do GC to move out valid blocks in the range */
 	for (segno = start; segno <= end; segno += sbi->segs_per_sec) {
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 1313caa60a4d2..f0c96826ee3e0 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -2479,6 +2479,7 @@ static void get_new_segment(struct f2fs_sb_info *sbi,
 	bool init = true;
 	int go_left = 0;
 	int i;
+	int ret = 0;
 
 	spin_lock(&free_i->segmap_lock);
 
@@ -2494,7 +2495,10 @@ static void get_new_segment(struct f2fs_sb_info *sbi,
 		if (dir == ALLOC_RIGHT) {
 			secno = find_next_zero_bit(free_i->free_secmap,
 							MAIN_SECS(sbi), 0);
-			f2fs_bug_on(sbi, secno >= MAIN_SECS(sbi));
+			if (secno >= MAIN_SECS(sbi)) {
+				ret = -ENOSPC;
+				goto out_unlock;
+			}
 		} else {
 			go_left = 1;
 			left_start = hint - 1;
@@ -2510,7 +2514,10 @@ static void get_new_segment(struct f2fs_sb_info *sbi,
 		}
 		left_start = find_next_zero_bit(free_i->free_secmap,
 							MAIN_SECS(sbi), 0);
-		f2fs_bug_on(sbi, left_start >= MAIN_SECS(sbi));
+		if (left_start >= MAIN_SECS(sbi)) {
+			ret = -ENOSPC;
+			goto out_unlock;
+		}
 		break;
 	}
 	secno = left_start;
@@ -2551,7 +2558,13 @@ static void get_new_segment(struct f2fs_sb_info *sbi,
 	f2fs_bug_on(sbi, test_bit(segno, free_i->free_segmap));
 	__set_inuse(sbi, segno);
 	*newseg = segno;
+out_unlock:
 	spin_unlock(&free_i->segmap_lock);
+
+	if (ret) {
+		f2fs_stop_checkpoint(sbi, false);
+		f2fs_bug_on(sbi, 1);
+	}
 }
 
 static void reset_curseg(struct f2fs_sb_info *sbi, int type, int modified)
@@ -2898,7 +2911,7 @@ static bool need_new_seg(struct f2fs_sb_info *sbi, int type)
 	return false;
 }
 
-void f2fs_allocate_segment_for_resize(struct f2fs_sb_info *sbi, int type,
+int f2fs_allocate_segment_for_resize(struct f2fs_sb_info *sbi, int type,
 					unsigned int start, unsigned int end)
 {
 	struct curseg_info *curseg = CURSEG_I(sbi, type);
@@ -2929,6 +2942,7 @@ void f2fs_allocate_segment_for_resize(struct f2fs_sb_info *sbi, int type,
 
 	mutex_unlock(&curseg->curseg_mutex);
 	up_read(&SM_I(sbi)->curseg_lock);
+	return f2fs_cp_error(sbi) ? -EIO : 0;
 }
 
 static void __allocate_new_segment(struct f2fs_sb_info *sbi, int type,
-- 
2.53.0

[PATCH 5.10.y 3/3] f2fs: fix to migrate all curseg types during free_segment_range

[Sasha Levin]

From: Daeho Jeong <daehojeong@google.com>

[ Upstream commit 8ec06f50ddd8d201bd7e55b896ae28ed9d4cb7d1 ]

In free_segment_range(), the curseg evacuation loop only iterates up to
NR_CURSEG_PERSIST_TYPE (0..5), missing non-persistent in-memory curseg
types such as CURSEG_COLD_DATA_PINNED and CURSEG_ALL_DATA_ATGC.

Even though these in-memory curseg types are not saved in the on-disk
checkpoint header, they still occupy active physical segments at runtime.
If an active in-memory curseg happens to be allocated within the segment
range being truncated during filesystem shrink, failing to evacuate it
will cause subsequent writes to the curseg attempting out-of-bounds I/O
on the truncated storage range.

Fix this by expanding the curseg evacuation loop upper bound to
NR_CURSEG_TYPE to ensure all active curseg types are safely migrated
out of the target range.

Fixes: d0b9e42ab615 ("f2fs: introduce inmem curseg")
Cc: stable@vger.kernel.org
Signed-off-by: Daeho Jeong <daehojeong@google.com>
Signed-off-by: Sunmin Jeong <s_min.jeong@samsung.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/f2fs/gc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index 33a2079b77c2b..d090ddc4be755 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -1877,7 +1877,7 @@ static int free_segment_range(struct f2fs_sb_info *sbi,
 	mutex_unlock(&DIRTY_I(sbi)->seglist_lock);
 
 	/* Move out cursegs from the target range */
-	for (type = CURSEG_HOT_DATA; type < NR_CURSEG_PERSIST_TYPE; type++) {
+	for (type = CURSEG_HOT_DATA; type < NR_CURSEG_TYPE; type++) {
 		err = f2fs_allocate_segment_for_resize(sbi, type, start, end);
 		if (err)
 			goto out;
-- 
2.53.0