linux-kernel - Re: [PATCH v3 4/9] media: venus: vdec: add video decoder files

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:   Thu, 24 Nov 2016 15:16:16 +0200
From:   Stanimir Varbanov <stanimir.varbanov@...aro.org>
To:     nicolas@...fresne.ca,
        Stanimir Varbanov <stanimir.varbanov@...aro.org>,
        Hans Verkuil <hverkuil@...all.nl>,
        Mauro Carvalho Chehab <mchehab@...nel.org>
Cc:     Andy Gross <andy.gross@...aro.org>,
        Bjorn Andersson <bjorn.andersson@...aro.org>,
        Stephen Boyd <sboyd@...eaurora.org>,
        Srinivas Kandagatla <srinivas.kandagatla@...aro.org>,
        linux-media@...r.kernel.org, linux-kernel@...r.kernel.org,
        linux-arm-msm@...r.kernel.org
Subject: Re: [PATCH v3 4/9] media: venus: vdec: add video decoder files

Hi Nicolas,

On 11/23/2016 10:24 PM, Nicolas Dufresne wrote:
> Le lundi 21 novembre 2016 à 18:09 +0200, Stanimir Varbanov a écrit :
>>>> Meanwhile I have found bigger obstacle - I cannot run multiple
>> instances
>>>> simultaneously. By m2m design it can execute only one job (m2m
>> context)
>>>> at a time per m2m device. Can you confirm that my observation is
>> correct?
>>>  
>>> The m2m framework assumes a single HW instance, yes. Do you have
>> multiple
>>> HW decoders? I might not understand what you mean...
>>>  
>>
>> I mean that I can start and execute up to 16 decoder sessions
>> simultaneously. Its a firmware responsibility how those sessions are
>> scheduled and how the hardware is shared between them. Of course
>> depending on the resolution the firmware can refuse to start the
>> session
>> because the hardware will be overloaded and will not be able to
>> satisfy
>> the bitrate requirements.
> 
> This is similar to S5P-MFC driver, which you may have notice not use
> m2m framework.

Thanks for the note.

I have started to look into m2m because Hans asked me to reuse the ioctl
helpers that it provides.

I have no problem with usage of the m2m API if they help me to reduce
code size and doesn't impact performance.


-- 
regards,
Stan