Re: BUG: KCSAN: data-race in do_page_fault / spectre_v4_enable_task_mitigation

[Marco Elver <elver@google.com>]

On Thu, 15 Dec 2022 at 08:32, Naresh Kamboju <naresh.kamboju@linaro.org> wrote:
>
> [Please ignore if it is already reported, and not an expert of KCSAN]
>
> On Linux next-20221215 tag arm64 allmodconfig boot failed due to following
> data-race reported by KCSAN.
>
> Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
>
> [    0.000000][    T0] Booting Linux on physical CPU 0x0000000000 [0x410fd034]
> [    0.000000][    T0] Linux version 6.1.0-next-20221214
> (tuxmake@tuxmake) (aarch64-linux-gnu-gcc (Debian 12.2.0-9) 12.2.0, GNU
> ld (GNU Binutils for Debian) 2.39) #2 SMP PREEMPT_DYNAMIC @1671022464
> [    0.000000][    T0] random: crng init done
> [    0.000000][    T0] Machine model: linux,dummy-virt
> ...
> [ 1067.461794][  T132] BUG: KCSAN: data-race in do_page_fault /
> spectre_v4_enable_task_mitigation
> [ 1067.467529][  T132]
> [ 1067.469146][  T132] write to 0xffff80000f00bfb8 of 8 bytes by task
> 93 on cpu 0:
> [ 1067.473790][  T132]  spectre_v4_enable_task_mitigation+0x2f8/0x340
> [ 1067.477964][  T132]  __switch_to+0xc4/0x200

Please provide line numbers with all reports - you can use the script
scripts/decode_stacktrace.sh (requires the vmlinux you found this
with) to do so.

It would be good to do this immediately, because having anyone else do
so is nearly impossible - and without line numbers this report will
very likely be ignored.

Thanks,
-- Marco

> [ 1067.480877][  T132]  __schedule+0x5ec/0x6c0
> [ 1067.483764][  T132]  schedule+0x6c/0x100
> [ 1067.486526][  T132]  worker_thread+0x7d8/0x8c0
> [ 1067.489581][  T132]  kthread+0x1b8/0x200
> [ 1067.492483][  T132]  ret_from_fork+0x10/0x20
> [ 1067.495450][  T132]
> [ 1067.497034][  T132] read to 0xffff80000f00bfb8 of 8 bytes by task
> 132 on cpu 0:
> [ 1067.501684][  T132]  do_page_fault+0x568/0xa40
> [ 1067.504938][  T132]  do_mem_abort+0x7c/0x180
> [ 1067.508051][  T132]  el0_da+0x64/0x100
> [ 1067.510712][  T132]  el0t_64_sync_handler+0x90/0x180
> [ 1067.514191][  T132]  el0t_64_sync+0x1a4/0x1a8
> [ 1067.517200][  T132]
> [ 1067.518758][  T132] 1 lock held by (udevadm)/132:
> [ 1067.521883][  T132]  #0: ffff00000b802c28
> (&mm->mmap_lock){++++}-{3:3}, at: do_page_fault+0x480/0xa40
> [ 1067.528399][  T132] irq event stamp: 1461
> [ 1067.531041][  T132] hardirqs last  enabled at (1460):
> [<ffff80000af83e40>] preempt_schedule_irq+0x40/0x100
> [ 1067.537176][  T132] hardirqs last disabled at (1461):
> [<ffff80000af82c84>] __schedule+0x84/0x6c0
> [ 1067.542788][  T132] softirqs last  enabled at (1423):
> [<ffff800008020688>] fpsimd_restore_current_state+0x148/0x1c0
> [ 1067.549480][  T132] softirqs last disabled at (1421):
> [<ffff8000080205fc>] fpsimd_restore_current_state+0xbc/0x1c0
> [ 1067.556127][  T132]
> [ 1067.557687][  T132] value changed: 0x0000000060000000 -> 0x0000000060001000
> [ 1067.562039][  T132]
> [ 1067.563631][  T132] Reported by Kernel Concurrency Sanitizer on:
> [ 1067.567480][  T132] CPU: 0 PID: 132 Comm: (udevadm) Tainted: G
>           T  6.1.0-next-20221214 #2
> 4185b46758ba972fed408118afddb8c426bff43a
> [ 1067.575669][  T132] Hardware name: linux,dummy-virt (DT)
>
>
> metadata:
>   repo: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/?h=next-20221214
>   config: allmodconfig
>   arch: arm64
>   Build details:
> https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20221214/
>
> --
> Linaro LKFT
> https://lkft.linaro.org