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 for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <d5944fc0-4f8a-9881-c336-b418d76540c5@gmail.com>
Date:   Wed, 19 May 2021 16:15:43 +0300
From:   Dmitry Osipenko <digetx@...il.com>
To:     Jaroslav Kysela <perex@...ex.cz>, Mark Brown <broonie@...nel.org>
Cc:     Thierry Reding <thierry.reding@...il.com>,
        Jonathan Hunter <jonathanh@...dia.com>,
        Takashi Iwai <tiwai@...e.com>, Ion Agorria <ion@...rria.com>,
        Svyatoslav Ryhel <clamor95@...il.com>,
        Frank Rowand <frowand.list@...il.com>,
        Liam Girdwood <lgirdwood@...il.com>,
        linux-tegra@...r.kernel.org, devicetree@...r.kernel.org,
        alsa-devel@...a-project.org, linux-kernel@...r.kernel.org,
        Rob Herring <robh@...nel.org>
Subject: Re: Question about Tegra UCMs

19.05.2021 14:13, Jaroslav Kysela пишет:
> Dne 19. 05. 21 v 0:31 Dmitry Osipenko napsal(a):
>> Mark, could you please help me to understand the UCM naming scheme that ALSA uses..
>>
>> About a year ago I tried to complain to Jaroslav Kysela in a comment to the UCM change [1] that it should be breaking the naming scheme of Tegra UCMs, but haven't got a meaningful reply and moved on to other things.
>>
>> [1] https://github.com/alsa-project/alsa-ucm-conf/commit/8ff2d50745efbb6959324f672460e413f0b618b8
> 
> I'm sorry about that, but it's better to create a tracked ticket (issue or
> pull request).
> 
>> Today I noticed that the naming scheme changed again and I still don't understand what to do about it.
>>
>> I have two devices:
>>
>>  1. Acer Picasso tablet that uses "Acer Iconia Tab A500 WM8903" for the card model name.
>>
>>  2. Google Nexus 7 that uses "ASUS Google Nexus 7 ALC5642".
>>
>> Previously UCMs were picked up by pulseaudio from these paths:
>>
>>  1. /usr/share/alsa/ucm2/Acer Iconia Tab A500 WM8903/
>>  2. /usr/share/alsa/ucm2/ASUS Google Nexus 7 ALC5642/
>>
>> Now the lookup paths are changed to:
>>
>>  1. /usr/share/alsa/ucm2/Acer_Iconia_Tab/
>>  2. /usr/share/alsa/ucm2/ASUS_Google_Nex/
> 
> Yes, it's based on the driver name (which is incorrectly set /or not set/ in
> your case).
> 
> Lookup paths (with description):
> 
> https://github.com/alsa-project/alsa-ucm-conf/blob/master/ucm2/ucm.conf
> 
> The latest scheme is even different - lookups were moved to ucm2/conf.d with
> redirection to the more descriptive layered configuration tree structure, so
> the other developers can immediately identify the hardware which is
> configured. See Qualcomm examples. The long card names does not help us so much.
> 
>> Strace shows that pulseaudio searches UCMs only at these paths.
>>
>> The output of /proc/asound/cards:
>>
>>  0 [WM8903         ]: Acer_Iconia_Tab - Acer Iconia Tab A500 WM8903
>>                       Acer Iconia Tab A500 WM8903
>>
>>  0 [ALC5642        ]: ASUS_Google_Nex - ASUS Google Nexus 7 ALC5642
>>                       ASUS Google Nexus 7 ALC5642
> 
> Fields are explained in:
> 
> https://github.com/alsa-project/alsa-ucm-conf/blob/master/ucm2/README.md
> 
>> Is there anything on the kernel side that I could change to get a working naming scheme? If yes, I may try to do something about it in the v2, thanks in advance.
> 
> Try to set a meaningful driver name (usually the code handling the ASoC card
> creation). It should be very close to the kernel module name (but more user
> friendly). The current code for your hardware use the auto-generated driver
> name from the ALSA long name.

The current name isn't auto-generated, it's specified via the
nvidia,model device-tree property.

> Then try to reuse the existing configs - for example your Nexus 7 config has
> many blocks from codecs/rt5640/* .
> 
> Anyway, create a PR so we can discuss the details.

I will create the PR, thank you very much. It's important problem which
can't be postponed anymore since sound doesn't work without UCMs on
Tegra devices.

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ