linux-kernel - Re: WARNING at switch_mm_irqs_off, followed by frozen machine

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] [day] [month] [year] [list]

Date:   Sat, 20 Jun 2020 03:59:43 +0300
From:   Ilkka Prusi <ilkka.prusi@...inet.fi>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     linux-kernel@...r.kernel.org, riel@...riel.com, luto@...nel.org,
        x86@...nel.org
Subject: Re: WARNING at switch_mm_irqs_off, followed by frozen machine

Hi,

On 18.6.2020 13.21, Peter Zijlstra wrote:
> On Wed, Jun 17, 2020 at 02:32:39AM +0300, Ilkka Prusi wrote:
>> Hi,
>>
>> Yesterday my computer with kernel version 5.7.2 was frozen badly enough that
>> hard reset was necessary (did not react to SysRq keys). Upon checking logs I
>> found following warning and information from the time just before resetting
>> it.
>>
>> Computer has AMD Ryzen 7 2700, Asus B450 motherboard.
>>
>> 5.8-rc1 encountered BUG() and did not boot (iommu and smp_processor_id()
>> called from wrong context, I'll see if I can catch log somehow).
> Since you're building your own kernels, can you make sure to always
> have:
>
>   - CONFIG_DEBUG_BUGVERBOSE=y
>   - CONFIG_DEBUG_INFO=y
>
> and then when reporting, run the thing through:
>
>    ./scripts/decode_stacktrace.sh

Ok, thanks for letting me know, I'll try that.


> And then pattern match that against my local defconfig build of
> arch/x86/mm/tlb.o and pray my compiler did anyting like your
> (unspecified) compiler.

Forgot to mention that, currently it is:

gcc (Debian 9.3.0-13) 9.3.0

I'll try your suggestions, sorry for not replying sooner.

--

  - Ilkka Prusi