linux-kernel - Re: snd_hda_intel/sst-acpi sound breakage on suspend/resume since 5.6-rc1

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]

Date:   Thu, 19 Mar 2020 14:17:35 +0100
From:   Cezary Rojewski <cezary.rojewski@...el.com>
To:     Dominik Brodowski <linux@...inikbrodowski.net>,
        kuninori.morimoto.gx@...esas.com
Cc:     Pierre-Louis Bossart <pierre-louis.bossart@...ux.intel.com>,
        Keyon Jie <yang.jie@...ux.intel.com>,
        alsa-devel@...a-project.org, curtis@...ainey.com,
        linux-kernel@...r.kernel.org, tiwai@...e.com,
        liam.r.girdwood@...ux.intel.com, broonie@...nel.org
Subject: Re: snd_hda_intel/sst-acpi sound breakage on suspend/resume since
 5.6-rc1

On 2020-03-19 14:00, Dominik Brodowski wrote:
> On Wed, Mar 18, 2020 at 11:20:55PM +0100, Cezary Rojewski wrote:

>>
>> Thanks for quick reply. Revert of said commit fixes stream==NULL issue for
>> me. See if there were any changes in dmesg.
>> Will ask technicians to assist me on site tomorrow.
> 
> Have some good news now, namely that a bisect is complete: That pointed to
> 1272063a7ee4 ("ASoC: soc-core: care .ignore_suspend for Component suspend");
> therefore I've added Kuninori Morimoto to this e-mail thread.
> 
> Additionally, I have tested mainline (v5.6-rc6+ as of 5076190daded) with
> *both* 64df6afa0dab (which you suggested yesterday) and 1272063a7ee4
> reverted. And that works like a charm as well.
> 
> Hope this helps!
> 
> Thanks,
> 	Dominik
> 

To make everyone not miss a bit - I believe we had 2 issues here, even 
though that one may seem harmless from user perspective:

 From IPC logs indeed it looks like a redundant (additional) stream 
initialization has occurred - said redundant stream is destroyed right 
after it has been created, and only to be recreated yet again.. Can 
share the logs if required.

While hw_params() handled doubled init nicely, _reset and _free
did not (during on pcm_close()) -> secondary invokes attempted to RESET
and FREE stream despite it being destroyed long ago. With revert of
patch I had mentioned, no lines:

!!!	haswell-pcm-audio haswell-pcm-audio: warning: stream is NULL, no
stream to reset, ignore it.
!!!	haswell-pcm-audio haswell-pcm-audio: warning: stream is NULL, no
stream to free, ignore it.

should appear.

I'll focus now on the commits you found offending during your bisect. 
Thank you Dominik!

Czarek