* [PATCH v3] fsck.f2fs: sanitize invalid segment type during block update
@ 2026-08-04 17:18 Daeho Jeong
2026-08-05 1:21 ` [f2fs-dev] " Chao Yu
0 siblings, 1 reply; 2+ messages in thread
From: Daeho Jeong @ 2026-08-04 17:18 UTC (permalink / raw)
To: linux-kernel, linux-f2fs-devel, kernel-team; +Cc: Daeho Jeong
From: Daeho Jeong <daehojeong@google.com>
If an invalid segment type (type < 0 || type >= NO_CHECK_TYPE) is read
from SIT during block update, update_block() now validates the type,
reports corruption via ASSERT_MSG(), fixes se->type directly, and falls
back to a valid active curseg type based on its SSA summary block type
when fix_on is enabled. find_next_free_block() also validates want_type to
prevent out-of-bounds accesses.
Signed-off-by: Daeho Jeong <daehojeong@google.com>
---
v3: do not update se->orig_type.
v2: used ASSERT_MSG() instead of ERR_MSG().
used the correct type info for fixed_type.
fixed se->type on top of fixing the local type only.
---
fsck/mount.c | 5 +++++
fsck/segment.c | 20 ++++++++++++++++++--
2 files changed, 23 insertions(+), 2 deletions(-)
diff --git a/fsck/mount.c b/fsck/mount.c
index 6f640a0..1804403 100644
--- a/fsck/mount.c
+++ b/fsck/mount.c
@@ -3059,6 +3059,11 @@ int find_next_free_block(struct f2fs_sb_info *sbi, u64 *to, int left,
u64 end_blkaddr = (get_sb(segment_count_main) <<
get_sb(log_blocks_per_seg)) + get_sb(main_blkaddr);
+ if (want_type < 0 || want_type >= NO_CHECK_TYPE) {
+ ASSERT_MSG("Invalid segment type %d for block allocation", want_type);
+ return -EINVAL;
+ }
+
if (c.zoned_model == F2FS_ZONED_HM && !new_sec) {
struct curseg_info *curseg = CURSEG_I(sbi, want_type);
unsigned int segs_per_zone = sbi->segs_per_sec * sbi->secs_per_zone;
diff --git a/fsck/segment.c b/fsck/segment.c
index 96de22a..9f3fa0c 100644
--- a/fsck/segment.c
+++ b/fsck/segment.c
@@ -770,15 +770,32 @@ int update_block(struct f2fs_sb_info *sbi, void *buf, u32 *blkaddr,
struct seg_entry *se;
struct f2fs_summary sum;
u64 new_blkaddr, old_blkaddr = *blkaddr, offset;
- int ret, type;
+ int ret, type, sum_type;
if (c.zoned_model != F2FS_ZONED_HM)
return dev_write_block(buf, old_blkaddr, WRITE_LIFE_NONE);
+ sum_type = get_sum_entry(sbi, old_blkaddr, &sum);
+
/* update sit bitmap & valid_blocks && se->type for old block*/
se = get_seg_entry(sbi, GET_SEGNO(sbi, old_blkaddr));
offset = OFFSET_IN_SEG(sbi, old_blkaddr);
type = se->type;
+ if (type < 0 || type >= NO_CHECK_TYPE) {
+ int fixed_type = (sum_type == SEG_TYPE_NODE ||
+ sum_type == SEG_TYPE_CUR_NODE) ?
+ CURSEG_HOT_NODE : CURSEG_HOT_DATA;
+
+ ASSERT_MSG("Invalid SIT segment type %d at blkaddr 0x%"PRIx64,
+ type, old_blkaddr);
+
+ if (!c.fix_on)
+ return -EINVAL;
+
+ FIX_MSG("Correct invalid segment type %d -> %d for blkaddr 0x%"PRIx64"\n",
+ type, fixed_type, old_blkaddr);
+ se->type = type = fixed_type;
+ }
se->valid_blocks--;
f2fs_clear_bit(offset, (char *)se->cur_valid_map);
if (need_fsync_data_record(sbi))
@@ -811,7 +828,6 @@ int update_block(struct f2fs_sb_info *sbi, void *buf, u32 *blkaddr,
f2fs_set_sit_bitmap(sbi, new_blkaddr);
/* update SSA */
- get_sum_entry(sbi, old_blkaddr, &sum);
update_sum_entry(sbi, new_blkaddr, &sum);
if (IS_DATASEG(type)) {
--
2.55.0.571.g244d577d93-goog
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [f2fs-dev] [PATCH v3] fsck.f2fs: sanitize invalid segment type during block update
2026-08-04 17:18 [PATCH v3] fsck.f2fs: sanitize invalid segment type during block update Daeho Jeong
@ 2026-08-05 1:21 ` Chao Yu
0 siblings, 0 replies; 2+ messages in thread
From: Chao Yu @ 2026-08-05 1:21 UTC (permalink / raw)
To: Daeho Jeong, linux-kernel, linux-f2fs-devel, kernel-team
Cc: chao, Daeho Jeong
On 8/5/26 01:18, Daeho Jeong wrote:
> From: Daeho Jeong <daehojeong@google.com>
>
> If an invalid segment type (type < 0 || type >= NO_CHECK_TYPE) is read
> from SIT during block update, update_block() now validates the type,
> reports corruption via ASSERT_MSG(), fixes se->type directly, and falls
> back to a valid active curseg type based on its SSA summary block type
> when fix_on is enabled. find_next_free_block() also validates want_type to
> prevent out-of-bounds accesses.
>
> Signed-off-by: Daeho Jeong <daehojeong@google.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Thanks,
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-08-05 1:21 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-08-04 17:18 [PATCH v3] fsck.f2fs: sanitize invalid segment type during block update Daeho Jeong
2026-08-05 1:21 ` [f2fs-dev] " Chao Yu
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).