Re: [PATCH v13 04/10] mm: split a folio in minimum folio order chunks

["Pankaj Raghav (Samsung)" <kernel@pankajraghav.com>]

On Fri, Sep 06, 2024 at 02:52:38PM +0800, Lai, Yi wrote:
Hi Yi,

> 
> I used Syzkaller and found that there is task hang in soft_offline_page in Linux-next tree - next-20240902.

I don't know if it is related, but we had a fix for this commit for a
ltp failure due to locking issues that is there in next-20240905 but not
in next-20240902.

Fix: https://lore.kernel.org/linux-next/20240902124931.506061-2-kernel@pankajraghav.com/

Is this reproducible also on next-20240905?

> 
> After bisection and the first bad commit is:
> "
> fd031210c9ce mm: split a folio in minimum folio order chunks
> "
> 
> All detailed into can be found at:
> https://github.com/laifryiee/syzkaller_logs/tree/main/240904_155526_soft_offline_page
> Syzkaller repro code:
> https://github.com/laifryiee/syzkaller_logs/tree/main/240904_155526_soft_offline_page/repro.c
> Syzkaller repro syscall steps:
> https://github.com/laifryiee/syzkaller_logs/tree/main/240904_155526_soft_offline_page/repro.prog
> Syzkaller report:
> https://github.com/laifryiee/syzkaller_logs/tree/main/240904_155526_soft_offline_page/repro.report
> Kconfig(make olddefconfig):
> https://github.com/laifryiee/syzkaller_logs/tree/main/240904_155526_soft_offline_page/kconfig_origin
> Bisect info:
> https://github.com/laifryiee/syzkaller_logs/tree/main/240904_155526_soft_offline_page/bisect_info.log
> bzImage:
> https://github.com/laifryiee/syzkaller_logs/raw/f633dcbc3a8e4ca5f52f0110bc75ff17d9885db4/240904_155526_soft_offline_page/bzImage_ecc768a84f0b8e631986f9ade3118fa37852fef0
> Issue dmesg:
> https://github.com/laifryiee/syzkaller_logs/blob/main/240904_155526_soft_offline_page/ecc768a84f0b8e631986f9ade3118fa37852fef0_dmesg.log
> 
> "
> [  447.976688]  ? __pfx_soft_offline_page.part.0+0x10/0x10
> [  447.977255]  ? __sanitizer_cov_trace_const_cmp4+0x1a/0x20
> [  447.977858]  soft_offline_page+0x97/0xc0
> [  447.978281]  do_madvise.part.0+0x1a45/0x2a30
> [  447.978742]  ? __pfx___lock_acquire+0x10/0x10
> [  447.979227]  ? __pfx_do_madvise.part.0+0x10/0x10
> [  447.979716]  ? __this_cpu_preempt_check+0x21/0x30
> [  447.980225]  ? __this_cpu_preempt_check+0x21/0x30
> [  447.980729]  ? lock_release+0x441/0x870
> [  447.981160]  ? __this_cpu_preempt_check+0x21/0x30
> [  447.981656]  ? seqcount_lockdep_reader_access.constprop.0+0xb4/0xd0
> [  447.982321]  ? lockdep_hardirqs_on+0x89/0x110
> [  447.982771]  ? trace_hardirqs_on+0x51/0x60
> [  447.983191]  ? seqcount_lockdep_reader_access.constprop.0+0xc0/0xd0
> [  447.983819]  ? __sanitizer_cov_trace_cmp4+0x1a/0x20
> [  447.984282]  ? ktime_get_coarse_real_ts64+0xbf/0xf0
> [  447.984673]  __x64_sys_madvise+0x139/0x180
> [  447.984997]  x64_sys_call+0x19a5/0x2140
> [  447.985307]  do_syscall_64+0x6d/0x140
> [  447.985600]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
> [  447.986011] RIP: 0033:0x7f782623ee5d
> [  447.986248] RSP: 002b:00007fff9ddaffb8 EFLAGS: 00000217 ORIG_RAX: 000000000000001c
> [  447.986709] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f782623ee5d
> [  447.987147] RDX: 0000000000000065 RSI: 0000000000003000 RDI: 0000000020d51000
> [  447.987584] RBP: 00007fff9ddaffc0 R08: 00007fff9ddafff0 R09: 00007fff9ddafff0
> [  447.988022] R10: 00007fff9ddafff0 R11: 0000000000000217 R12: 00007fff9ddb0118
> [  447.988428] R13: 0000000000401716 R14: 0000000000403e08 R15: 00007f782645d000
> [  447.988799]  </TASK>
> [  447.988921]
> [  447.988921] Showing all locks held in the system:
> [  447.989237] 1 lock held by khungtaskd/33:
> [  447.989447]  #0: ffffffff8705c500 (rcu_read_lock){....}-{1:2}, at: debug_show_all_locks+0x73/0x3c0
> [  447.989947] 1 lock held by repro/628:
> [  447.990144]  #0: ffffffff87258a28 (mf_mutex){+.+.}-{3:3}, at: soft_offline_page.part.0+0xda/0xf40
> [  447.990611]
> [  447.990701] =============================================
> 
> "
> 
> I hope you find it useful.
> 
> Regards,
> Yi Lai
>