linux-kernel - Re: [PATCH] soundwire: stream: only change state if needed

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]

Message-ID: <7a362c6a-364e-499a-e841-0a919f48bf84@linux.intel.com>
Date:   Tue, 17 Mar 2020 10:31:29 -0500
From:   Pierre-Louis Bossart <pierre-louis.bossart@...ux.intel.com>
To:     Srinivas Kandagatla <srinivas.kandagatla@...aro.org>,
        alsa-devel@...a-project.org
Cc:     linux-kernel@...r.kernel.org, tiwai@...e.de, broonie@...nel.org,
        vkoul@...nel.org, gregkh@...uxfoundation.org, jank@...ence.com,
        slawomir.blauciak@...el.com,
        Bard liao <yung-chuan.liao@...ux.intel.com>,
        Rander Wang <rander.wang@...ux.intel.com>,
        Ranjani Sridharan <ranjani.sridharan@...ux.intel.com>,
        Hui Wang <hui.wang@...onical.com>,
        Sanyog Kale <sanyog.r.kale@...el.com>
Subject: Re: [PATCH] soundwire: stream: only change state if needed


>>>> The change below would be an error case for Intel, so it's probably 
>>>> better if we go with your suggestion. You have a very specific state 
>>>> handling due to your power amps and it's probably better to keep it 
>>>> platform-specific.
>>>
>>> Just trying to understand, why would it be error for Intel case?
>>>
>>> IMO, If stream state is SDW_STREAM_ENABLED that also implicit that 
>>> its prepared too. Similar thing with SDW_STREAM_DEPREPARED.
>>> Isn't it?
>>
>> the stream state is a scalar value, not a mask. The state machine only 
>> allows transition from CONFIGURED TO PREPARED or from DEPREPARED TO 
>> PREPARED, or DISABLED to PREPARED.
>> There is no allowed transition from ENABLED TO PREPARED, you have to 
>> go through the DISABLED state and make sure a bank switch occurred, 
>> and re-do a bank switch to prepare again.
> I agree with you if are on single dai case. Am happy to move to stream 
> handling to machine driver for now.
> 
> But this also means that in cases like multi-codec its not recommended 
> to call sdw_prepare and sdw_enable in a single function from codec.
> Which might be worth documenting.

Well, the bigger question is why use sdw_stream functions at the codec 
level in the first place? All other codec drivers seem to have no issue 
leaving the dais owned by the master control stream state changes.

I am not saying I object or it's bad, just that there were significant 
changes in usages of the 'stream' concept since it was introduced, as 
well as threads in MIPI circles to clarify the prepare/enable 
dependencies, so it'd be useful to have a complete picture of what 
different platforms need/want.