linux-kernel - Re: [syzbot] [xfs?] KASAN: slab-use-after-free Read in xfs_buf

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CANp29Y6PLo-Lr81SWp4qK9avLKpTGhXMAUDrZe3OYYgfWuaKVw@mail.gmail.com>
Date: Mon, 19 Jan 2026 09:53:18 +0100
From: Aleksandr Nogikh <nogikh@...gle.com>
To: Christoph Hellwig <hch@...radead.org>
Cc: syzbot <syzbot+0391d34e801643e2809b@...kaller.appspotmail.com>, 
	linux-kernel@...r.kernel.org, linux-xfs@...r.kernel.org, 
	syzkaller-bugs@...glegroups.com, Dmitry Vyukov <dvyukov@...gle.com>
Subject: Re: [syzbot] [xfs?] KASAN: slab-use-after-free Read in xfs_buf_rele (4)

On Mon, Jan 19, 2026 at 9:37 AM Christoph Hellwig <hch@...radead.org> wrote:
>
> So I'm not sure what this test does that it always triggers the lockdep
> keys, but that makes it impossible to validate the original xfs report.
>
> Is there a way to force running syzbot reproducers without lockdep?

Not directly, but you could explicitly modify lockdep's Kconfig in
your test patch to disable lockdep entirely.

>
> Note that I've also had it running locally for quite a while, an even
> with lockdep enabled I'm somehow not hitting the lockdep splat.
> Although that is using my normal debug config and not the provided
> one.

Hmm, yes, that sounds weird.

I wonder if it's because we run the reproducers in threaded mode when
handling #syz test commands on the syzbot side, which leads to even
more syscalls being executed in parallel. Or the system just got lucky
once when it was generating the reproducer - overall, "BUG:
MAX_LOCKDEP_KEYS too low!" [1] seems to be a popular sink for
different reproducers on our side :(

[1] https://syzkaller.appspot.com/bug?extid=a70a6358abd2c3f9550f

>
> On Mon, Jan 19, 2026 at 12:34:03AM -0800, syzbot wrote:
> > Hello,
> >
> > syzbot has tested the proposed patch but the reproducer is still triggering an issue:
> > BUG: MAX_LOCKDEP_KEYS too low!
> >
> > BUG: MAX_LOCKDEP_KEYS too low!
> > turning off the locking correctness validator.
> > CPU: 1 UID: 0 PID: 7123 Comm: syz-executor Not tainted syzkaller #0 PREEMPT
> > Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/03/2025
> > Call trace:
> >  show_stack+0x2c/0x3c arch/arm64/kernel/stacktrace.c:499 (C)
> >  __dump_stack+0x30/0x40 lib/dump_stack.c:94
> >  dump_stack_lvl+0xd8/0x12c lib/dump_stack.c:120
> >  dump_stack+0x1c/0x28 lib/dump_stack.c:129
> >  register_lock_class+0x310/0x348 kernel/locking/lockdep.c:1332
> >  __lock_acquire+0xbc/0x30a4 kernel/locking/lockdep.c:5112
> >  lock_acquire+0x140/0x2e0 kernel/locking/lockdep.c:5868
> >  touch_wq_lockdep_map+0xa8/0x164 kernel/workqueue.c:3940
> >  __flush_workqueue+0xfc/0x109c kernel/workqueue.c:3982
> >  drain_workqueue+0xa4/0x310 kernel/workqueue.c:4146
> >  destroy_workqueue+0xb4/0xd90 kernel/workqueue.c:5903
> >  xfs_destroy_mount_workqueues+0xac/0xdc fs/xfs/xfs_super.c:649
> >  xfs_fs_put_super+0x128/0x144 fs/xfs/xfs_super.c:1262
> >  generic_shutdown_super+0x12c/0x2b8 fs/super.c:643
> >  kill_block_super+0x44/0x90 fs/super.c:1722
> >  xfs_kill_sb+0x20/0x58 fs/xfs/xfs_super.c:2297
> >  deactivate_locked_super+0xc4/0x12c fs/super.c:474
> >  deactivate_super+0xe0/0x100 fs/super.c:507
> >  cleanup_mnt+0x31c/0x3ac fs/namespace.c:1318
> >  __cleanup_mnt+0x20/0x30 fs/namespace.c:1325
> >  task_work_run+0x1dc/0x260 kernel/task_work.c:233
> >  resume_user_mode_work include/linux/resume_user_mode.h:50 [inline]
> >  __exit_to_user_mode_loop kernel/entry/common.c:44 [inline]
> >  exit_to_user_mode_loop+0x10c/0x18c kernel/entry/common.c:75
> >  __exit_to_user_mode_prepare include/linux/irq-entry-common.h:226 [inline]
> >  exit_to_user_mode_prepare_legacy include/linux/irq-entry-common.h:242 [inline]
> >  arm64_exit_to_user_mode arch/arm64/kernel/entry-common.c:81 [inline]
> >  el0_svc+0x17c/0x26c arch/arm64/kernel/entry-common.c:725
> >  el0t_64_sync_handler+0x84/0x12c arch/arm64/kernel/entry-common.c:743
> >  el0t_64_sync+0x198/0x19c arch/arm64/kernel/entry.S:596
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> > XFS (loop0): Unmounting Filesystem c496e05e-540d-4c72-b591-04d79d8b4eeb
> >
> >
> > Tested on:
> >
> > commit:         3e548540 increase LOCKDEP_CHAINS_BITS
> > git tree:       git://git.infradead.org/users/hch/xfs.git xfs-buf-hash
> > console output: https://syzkaller.appspot.com/x/log.txt?x=101b0d22580000
> > kernel config:  https://syzkaller.appspot.com/x/.config?x=6c6138f827b10ea4
> > dashboard link: https://syzkaller.appspot.com/bug?extid=0391d34e801643e2809b
> > compiler:       Debian clang version 20.1.8 (++20250708063551+0c9f909b7976-1~exp1~20250708183702.136), Debian LLD 20.1.8
> > userspace arch: arm64
> >
> > Note: no patches were applied.
> ---end quoted text---
>