Re: [PATCH v4] loop: Fix NULL pointer dereference in lo_rw_aio()

[Tetsuo Handa]

On 2026/07/13 12:04, Hillf Danton wrote:
> On Tue, 9 Jun 2026 18:50:13 +0100 Al Viro wrote:
>> On Sun, Jun 07, 2026 at 07:54:58PM +0900, Tetsuo Handa wrote:
>>> syzbot is reporting NULL pointer dereference in lo_rw_aio() [1][2].
>>> An analysis by the Gemini AI collaborator [3] considers that this problem
>>> is caused by a timing shift primarily exposed by commit 65565ca5f99b
>>> ("block: unify the synchronous bi_end_io callbacks"), along with helper
>>> refactorings like commit 92c3737a2473 ("block: add a bio_submit_or_kill
>>> helper").
>>>
>>> But due to difficulty of reproducing this race, discussion about what is
>>> happening and how to fix this problem is stalling. Also, we haven't
>>> identified how many filesystems are subjected to this problem.
>>>
>>> Therefore, this patch introduces a grace period for flushing pending I/O
>>> requests (which should be a good thing from the perspective of defensive
>>> programming) so that we won't hit NULL pointer dereference problem, and
>>> also emits BUG: message in order to help filesystem developers identify
>>> the caller of an I/O request that failed to wait for completion so that
>>> filesystem developers can fix such caller to wait for completion.
>>>
>>> Note that emitting BUG: message is enabled only if CONFIG_KCOV=y, for
>>> this check is a waste of computation resources for almost all users.
>>
>> Still breaks xfs/259, same as the version in next-20260605...
>>
> And syzbot report [1]
> 
> [1] https://lore.kernel.org/lkml/6a54399f.278cdcd2.e885.0023.GAE@xxxxxxxxxx/

No. This report is from upstream tree where "loop: Fix NULL pointer dereference
in lo_rw_aio()" is not yet applied.

[  102.784602][ T5324] loop0: detected capacity change from 0 to 2048
[  102.837593][ T5324] Alternate GPT is invalid, using primary GPT.
[  102.841605][ T5324]  loop0: p2 p3 p7
[  102.939203][ T5285] I/O error, dev loop0, sector 408 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 2
[  102.947057][ T5324] loop0: detected capacity change from 0 to 512
[  102.951560][ T5287] I/O error, dev loop0, sector 208 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 2
[  102.964277][ T5287] I/O error, dev loop0, sector 208 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 2
[  102.969633][ T5285] I/O error, dev loop0, sector 408 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 2
[  102.979918][ T5285] Buffer I/O error on dev loop0p7, logical block 8, async page read
[  102.984620][ T5287] Buffer I/O error on dev loop0p3, logical block 8, async page read
[  103.009916][ T5287] udevd[5287]: inotify_add_watch(7, /dev/loop0p3, 10) failed: No such file or directory
[  103.014532][  T176] ==================================================================
[  103.018401][  T176] BUG: KASAN: slab-use-after-free in update_io_ticks+0xa9/0x280
[  103.022211][  T176] Read of size 8 at addr ffff88803236b9a8 by task kworker/u4:6/176
[  103.025664][  T176] 
[  103.026746][  T176] CPU: 0 UID: 0 PID: 176 Comm: kworker/u4:6 Not tainted syzkaller #0 PREEMPT(full) 
[  103.026761][  T176] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04/01/2014
[  103.026768][  T176] Workqueue: loop0 loop_rootcg_workfn

Al, can you reproduce your xfs/259 breakage and share kernel messages with
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?id=d908729e7467462bce1db3bc1bffec2e2d1f6c81 ?

I'm floating this patch in linux-next tree, but syzbot hasn't succeeded to hit
"BUG: %s/%u is doing I/O request on loop%d in Lo_rundown state.\n" message.
I think we need to send to upstream tree so that syzbot can hit this message within a few days.
To do so, we need to agree that your xfs/259 breakage should be addressed on the xfstests side.