* [syzbot] [ext4?] BUG: unable to handle kernel paging request in jbd2_journal_file_inode
@ 2026-11-15 21:33 syzbot
2026-11-16 5:44 ` [PATCH] ext4: attach jinode before inline data conversion in setattr Qing Zhang
0 siblings, 1 reply; 4+ messages in thread
From: syzbot @ 2026-11-15 21:33 UTC (permalink / raw)
To: jack, jack, kartikey406, linux-ext4, linux-kernel, syzkaller-bugs,
tytso
Hello,
syzbot found the following issue on:
HEAD commit: 9d66f26fefac Merge tag 'vfs-7.4-rc3.fixes.2' of git://git...
git tree: upstream
console+strace: https://syzkaller.appspot.com/x/log.txt?x=15ee2a3d580000
kernel config: https://syzkaller.appspot.com/x/.config?x=d16c0eae9c338e3c
dashboard link: https://syzkaller.appspot.com/bug?extid=9c29b92a3b6500bb347d
compiler: Debian clang version 22.1.8 (1~deb13u4), Debian LLD 22.1.8
syz repro: https://syzkaller.appspot.com/x/repro.syz?x=1185c4c3580000
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=1585c4c3580000
Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/74b7d989ed20/disk-9d66f26f.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/181fa79fbd3e/vmlinux-9d66f26f.xz
kernel image: https://storage.googleapis.com/syzbot-assets/83339d886058/bzImage-9d66f26f.xz
mounted in repro: https://storage.googleapis.com/syzbot-assets/8e682084d20b/mount_0.gz
fsck result: OK (log: https://syzkaller.appspot.com/x/fsck.log?x=1685c4c3580000)
The issue was bisected to:
commit ed9356a30e59c7cc3198e7fc46cfedf3767b9b17
Author: Deepanshu Kartikey <kartikey406@gmail.com>
Date: Sat Feb 7 04:36:07 2026 +0000
ext4: convert inline data to extents when truncate exceeds inline size
bisection log: https://syzkaller.appspot.com/x/bisect.txt?x=16d2f5a1580000
final oops: https://syzkaller.appspot.com/x/report.txt?x=15d2f5a1580000
console output: https://syzkaller.appspot.com/x/log.txt?x=11d2f5a1580000
AI-generated root cause summary:
In ext4, path-based truncate(2) bypasses ext4_file_open(), leaving
EXT4_I(inode)->jinode as NULL if the file has not been opened for writing. When
truncating an inline-data file to a size larger than its inline capacity,
ext4_setattr() calls ext4_convert_inline_data() to convert the file to extent
storage without first attaching a jinode. During conversion,
ext4_convert_inline_data_nolock() allocates a block via ext4_map_blocks(), which
in data=ordered mode invokes ext4_jbd2_inode_add_write() with
EXT4_I(inode)->jinode (NULL), causing a NULL-pointer dereference when
jbd2_journal_file_inode() accesses jinode->i_flags.
full analysis: https://syzkaller.appspot.com/x/ai_root_cause?x=9c29b92a3b6500bb347d
IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+9c29b92a3b6500bb347d@syzkaller.appspotmail.com
Fixes: ed9356a30e59 ("ext4: convert inline data to extents when truncate exceeds inline size")
EXT4-fs (loop0): mounted filesystem 00000000-0000-0000-0000-000000000000 r/w with ordered data mode. Quota mode: journalled.
Oops: general protection fault, probably for non-canonical address 0xdffffc0000000005: 0000 [#1] SMP KASAN PTI
KASAN: null-ptr-deref in range [0x0000000000000028-0x000000000000002f]
CPU: 1 UID: 0 PID: 5626 Comm: syz-executor342 Not tainted syzkaller #0 PREEMPT_{RT,(full)}
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 11/06/2026
RIP: 0010:jbd2_journal_file_inode+0x10e/0x5c0 fs/jbd2/transaction.c:2717
Code: 94 ff 49 8b 06 48 89 44 24 08 48 8d b8 e8 08 00 00 48 89 7c 24 30 e8 21 08 d1 08 48 8b 04 24 48 8d 78 28 48 89 f8 48 c1 e8 03 <80> 3c 28 00 74 05 e8 87 30 94 ff 48 8b 1c 24 4c 09 7b 28 4c 8d 7b
RSP: 0018:ffffc90004c17690 EFLAGS: 00010206
RAX: 0000000000000005 RBX: 0000000000000000 RCX: ffffffff9ada4a01
RDX: 000000009c34c801 RSI: ffffffff8bed2560 RDI: 0000000000000028
RBP: dffffc0000000000 R08: ffffffff8b781fe2 R09: 0000000000000000
R10: dffffc0000000000 R11: fffffbfff21ab398 R12: 1ffff110067de680
R13: 00000000000007ff R14: ffff888033ef3400 R15: 0000000000000006
FS: 000055556ada1400(0000) GS:ffff888124811000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007ffd3ee05c74 CR3: 00000000426c8000 CR4: 00000000003526f0
Call Trace:
<TASK>
ext4_jbd2_inode_add_write fs/ext4/ext4_jbd2.h:336 [inline]
ext4_map_blocks+0x161e/0x1c60 fs/ext4/inode.c:967
ext4_convert_inline_data_nolock+0x25d/0x980 fs/ext4/inline.c:1139
ext4_convert_inline_data+0x484/0x5c0 fs/ext4/inline.c:2012
ext4_setattr+0xef2/0x1560 fs/ext4/inode.c:7104
notify_change+0xbb8/0xec0 fs/attr.c:556
do_truncate+0x1c2/0x250 fs/open.c:68
vfs_truncate+0x4ba/0x540 fs/open.c:118
ksys_truncate+0xf3/0x1c0 fs/open.c:142
__do_sys_truncate fs/open.c:154 [inline]
__se_sys_truncate fs/open.c:152 [inline]
__x64_sys_truncate+0x5b/0x70 fs/open.c:152
do_syscall_x64 arch/x86/entry/syscall_64.c:61 [inline]
do_syscall_64+0x16a/0x530 arch/x86/entry/syscall_64.c:89
entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f518900a909
Code: c0 79 93 eb d5 48 8d 7c 1d 00 eb 99 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 d8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffefc2154d8 EFLAGS: 00000246 ORIG_RAX: 000000000000004c
RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f518900a909
RDX: 00007ffefc215360 RSI: 0000000002fffffd RDI: 00002000000033c0
RBP: 0000000000000001 R08: 0000000000000000 R09: 000055556ada2560
R10: 0000000000000000 R11: 0000000000000246 R12: 00007ffefc215500
R13: 0000000000000002 R14: 431bde82d7b634db R15: 00007f518906a04b
</TASK>
Modules linked in:
---[ end trace 0000000000000000 ]---
RIP: 0010:jbd2_journal_file_inode+0x10e/0x5c0 fs/jbd2/transaction.c:2717
Code: 94 ff 49 8b 06 48 89 44 24 08 48 8d b8 e8 08 00 00 48 89 7c 24 30 e8 21 08 d1 08 48 8b 04 24 48 8d 78 28 48 89 f8 48 c1 e8 03 <80> 3c 28 00 74 05 e8 87 30 94 ff 48 8b 1c 24 4c 09 7b 28 4c 8d 7b
RSP: 0018:ffffc90004c17690 EFLAGS: 00010206
RAX: 0000000000000005 RBX: 0000000000000000 RCX: ffffffff9ada4a01
RDX: 000000009c34c801 RSI: ffffffff8bed2560 RDI: 0000000000000028
RBP: dffffc0000000000 R08: ffffffff8b781fe2 R09: 0000000000000000
R10: dffffc0000000000 R11: fffffbfff21ab398 R12: 1ffff110067de680
R13: 00000000000007ff R14: ffff888033ef3400 R15: 0000000000000006
FS: 000055556ada1400(0000) GS:ffff888124811000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007ffd3ee05c74 CR3: 00000000426c8000 CR4: 00000000003526f0
----------------
Code disassembly (best guess), 2 bytes skipped:
0: 49 8b 06 mov (%r14),%rax
3: 48 89 44 24 08 mov %rax,0x8(%rsp)
8: 48 8d b8 e8 08 00 00 lea 0x8e8(%rax),%rdi
f: 48 89 7c 24 30 mov %rdi,0x30(%rsp)
14: e8 21 08 d1 08 call 0x8d1083a
19: 48 8b 04 24 mov (%rsp),%rax
1d: 48 8d 78 28 lea 0x28(%rax),%rdi
21: 48 89 f8 mov %rdi,%rax
24: 48 c1 e8 03 shr $0x3,%rax
* 28: 80 3c 28 00 cmpb $0x0,(%rax,%rbp,1) <-- trapping instruction
2c: 74 05 je 0x33
2e: e8 87 30 94 ff call 0xff9430ba
33: 48 8b 1c 24 mov (%rsp),%rbx
37: 4c 09 7b 28 or %r15,0x28(%rbx)
3b: 4c rex.WR
3c: 8d .byte 0x8d
3d: 7b .byte 0x7b
---
This report is generated by a bot. It may contain errors.
See https://goo.gl/tpsmEJ for more information about syzbot.
syzbot engineers can be reached at syzkaller@googlegroups.com.
syzbot will keep track of this issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
For information about bisection process see: https://goo.gl/tpsmEJ#bisection
If the report is already addressed, let syzbot know by replying with:
#syz fix: exact-commit-title
If you want to overwrite report's subsystems, reply with:
#syz set subsystems: new-subsystem
(See the list of subsystem names on the web dashboard)
If the report is a duplicate of another one, reply with:
#syz dup: exact-subject-of-another-report
If you want to undo deduplication, reply with:
#syz undup
^ permalink raw reply [flat|nested] 4+ messages in thread* [PATCH] ext4: attach jinode before inline data conversion in setattr
2026-11-15 21:33 [syzbot] [ext4?] BUG: unable to handle kernel paging request in jbd2_journal_file_inode syzbot
@ 2026-11-16 5:44 ` Qing Zhang
2026-11-16 5:44 ` syzbot
2026-11-16 5:55 ` sashiko-bot
0 siblings, 2 replies; 4+ messages in thread
From: Qing Zhang @ 2026-11-16 5:44 UTC (permalink / raw)
To: tytso
Cc: adilger.kernel, libaokun, jack, ojaswin, ritesh.list, kartikey406,
yi.zhang, linux-ext4, linux-kernel, Qing Zhang, Ahmed Khan,
syzbot+9c29b92a3b6500bb347d, stable
Commit ed9356a30e59 ("ext4: convert inline data to extents when
truncate exceeds inline size") added inline data conversion to
ext4_setattr(). Pathname-based truncate bypasses ext4_file_open(),
so the inode may lack a jinode. In ordered mode, the conversion
can pass this NULL jinode to jbd2_journal_file_inode(), causing a
NULL pointer dereference.
Attach the jinode before ext4_convert_inline_data() and propagate
attachment errors. The attachment in ext4_truncate_up() occurs
after the conversion and is too late.
Fixes: ed9356a30e59 ("ext4: convert inline data to extents when truncate exceeds inline size")
Reported-by: Ahmed Khan <ahmadpsn95@gmail.com>
Link: https://lore.kernel.org/r/CAOaS1==g3mdoqmS2AjOC0_-woXJbA_X0LJQt7+M7d1=+gfyQqw@mail.gmail.com
Reported-by: syzbot+9c29b92a3b6500bb347d@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=9c29b92a3b6500bb347d
Cc: stable@vger.kernel.org
Signed-off-by: Qing Zhang <diasyzhang@tencent.com>
---
#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
fs/ext4/inode.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 10cd85d4c721..1e31933ae98c 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -7127,6 +7127,9 @@ int ext4_setattr(const struct mnt_idmap *idmap, struct dentry *dentry,
*/
if (ext4_has_inline_data(inode) &&
attr->ia_size > EXT4_I(inode)->i_inline_size) {
+ error = ext4_inode_attach_jinode(inode);
+ if (error)
+ goto err_out;
error = ext4_convert_inline_data(inode);
if (error)
goto err_out;
--
2.43.7
^ permalink raw reply related [flat|nested] 4+ messages in thread* Re: [PATCH] ext4: attach jinode before inline data conversion in setattr
2026-11-16 5:44 ` [PATCH] ext4: attach jinode before inline data conversion in setattr Qing Zhang
@ 2026-11-16 5:44 ` syzbot
2026-11-16 5:55 ` sashiko-bot
1 sibling, 0 replies; 4+ messages in thread
From: syzbot @ 2026-11-16 5:44 UTC (permalink / raw)
To: diasyzhangqing
Cc: adilger.kernel, ahmadpsn95, diasyzhang, diasyzhangqing, jack,
kartikey406, libaokun, linux-ext4, linux-kernel, ojaswin,
ritesh.list, stable, tytso, yi.zhang, syzkaller-bugs
> Commit ed9356a30e59 ("ext4: convert inline data to extents when
> truncate exceeds inline size") added inline data conversion to
> ext4_setattr(). Pathname-based truncate bypasses ext4_file_open(),
> so the inode may lack a jinode. In ordered mode, the conversion
> can pass this NULL jinode to jbd2_journal_file_inode(), causing a
> NULL pointer dereference.
>
> Attach the jinode before ext4_convert_inline_data() and propagate
> attachment errors. The attachment in ext4_truncate_up() occurs
> after the conversion and is too late.
>
> Fixes: ed9356a30e59 ("ext4: convert inline data to extents when truncate exceeds inline size")
> Reported-by: Ahmed Khan <ahmadpsn95@gmail.com>
> Link: https://lore.kernel.org/r/CAOaS1==g3mdoqmS2AjOC0_-woXJbA_X0LJQt7+M7d1=+gfyQqw@mail.gmail.com
> Reported-by: syzbot+9c29b92a3b6500bb347d@syzkaller.appspotmail.com
> Closes: https://syzkaller.appspot.com/bug?extid=9c29b92a3b6500bb347d
> Cc: stable@vger.kernel.org
> Signed-off-by: Qing Zhang <diasyzhang@tencent.com>
> ---
>
> #syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
Patch testing via #syz test is temporarily disabled.
See https://github.com/google/syzkaller/issues/8122 for details.
> fs/ext4/inode.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
> index 10cd85d4c721..1e31933ae98c 100644
> --- a/fs/ext4/inode.c
> +++ b/fs/ext4/inode.c
> @@ -7127,6 +7127,9 @@ int ext4_setattr(const struct mnt_idmap *idmap, struct dentry *dentry,
> */
> if (ext4_has_inline_data(inode) &&
> attr->ia_size > EXT4_I(inode)->i_inline_size) {
> + error = ext4_inode_attach_jinode(inode);
> + if (error)
> + goto err_out;
> error = ext4_convert_inline_data(inode);
> if (error)
> goto err_out;
> --
> 2.43.7
>
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] ext4: attach jinode before inline data conversion in setattr
2026-11-16 5:44 ` [PATCH] ext4: attach jinode before inline data conversion in setattr Qing Zhang
2026-11-16 5:44 ` syzbot
@ 2026-11-16 5:55 ` sashiko-bot
1 sibling, 0 replies; 4+ messages in thread
From: sashiko-bot @ 2026-11-16 5:55 UTC (permalink / raw)
To: Qing Zhang; +Cc: linux-ext4, tytso
> Commit ed9356a30e59 ("ext4: convert inline data to extents when
> truncate exceeds inline size") added inline data conversion to
> ext4_setattr(). Pathname-based truncate bypasses ext4_file_open(),
> so the inode may lack a jinode. In ordered mode, the conversion
> can pass this NULL jinode to jbd2_journal_file_inode(), causing a
> NULL pointer dereference.
>
> Attach the jinode before ext4_convert_inline_data() and propagate
> attachment errors. The attachment in ext4_truncate_up() occurs
> after the conversion and is too late.
>
> Fixes: ed9356a30e59 ("ext4: convert inline data to extents when truncate exceeds inline size")
> Reported-by: Ahmed Khan <ahmadpsn95@gmail.com>
> Link: https://lore.kernel.org/r/CAOaS1==g3mdoqmS2AjOC0_-woXJbA_X0LJQt7+M7d1=+gfyQqw@mail.gmail.com
> Reported-by: syzbot+9c29b92a3b6500bb347d@syzkaller.appspotmail.com
> Closes: https://syzkaller.appspot.com/bug?extid=9c29b92a3b6500bb347d
> Cc: stable@vger.kernel.org
> Signed-off-by: Qing Zhang <diasyzhang@tencent.com>
Sashiko has reviewed this patch and found no issues.
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261116054424.62067-1-diasyzhang@tencent.com?part=1
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2026-11-16 5:55 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-11-15 21:33 [syzbot] [ext4?] BUG: unable to handle kernel paging request in jbd2_journal_file_inode syzbot
2026-11-16 5:44 ` [PATCH] ext4: attach jinode before inline data conversion in setattr Qing Zhang
2026-11-16 5:44 ` syzbot
2026-11-16 5:55 ` sashiko-bot
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox