Re: [RFCv4,19/21] media: vim2m: add request support

[Dmitry Osipenko]

On 12.03.2018 15:32, Alexandre Courbot wrote:
> On Mon, Mar 12, 2018 at 5:15 PM, Tomasz Figa <tfiga@xxxxxxxxxxxx> wrote:
>> Hi Paul, Dmitry,
>>
>> On Mon, Mar 12, 2018 at 5:10 PM, Paul Kocialkowski
>> <paul.kocialkowski@xxxxxxxxxxx> wrote:
>>> Hi,
>>>
>>> On Sun, 2018-03-11 at 22:42 +0300, Dmitry Osipenko wrote:
>>>> Hello,
>>>>
>>>> On 07.03.2018 19:37, Paul Kocialkowski wrote:
>>>>> Hi,
>>>>>
>>>>> First off, I'd like to take the occasion to say thank-you for your
>>>>> work.
>>>>> This is a major piece of plumbing that is required for me to add
>>>>> support
>>>>> for the Allwinner CedarX VPU hardware in upstream Linux. Other
>>>>> drivers,
>>>>> such as tegra-vde (that was recently merged in staging) are also
>>>>> badly
>>>>> in need of this API.
>>>>
>>>> Certainly it would be good to have a common UAPI. Yet I haven't got my
>>>> hands on
>>>> trying to implement the V4L interface for the tegra-vde driver, but
>>>> I've taken a
>>>> look at Cedrus driver and for now I've one question:
>>>>
>>>> Would it be possible (or maybe already is) to have a single IOCTL that
>>>> takes input/output buffers with codec parameters, processes the
>>>> request(s) and returns to userspace when everything is done? Having 5
>>>> context switches for a single frame decode (like Cedrus VAAPI driver
>>>> does) looks like a bit of overhead.
>>>
>>> The V4L2 interface exposes ioctls for differents actions and I don't
>>> think there's a combined ioctl for this. The request API was introduced
>>> precisely because we need to have consistency between the various ioctls
>>> needed for each frame. Maybe one single (atomic) ioctl would have worked
>>> too, but that's apparently not how the V4L2 API was designed.
>>>
>>> I don't think there is any particular overhead caused by having n ioctls
>>> instead of a single one. At least that would be very surprising IMHO.
>>
>> Well, there is small syscall overhead, which normally shouldn't be
>> very painful, although with all the speculative execution hardening,
>> can't be sure of anything anymore. :)
>>
>> Hans and Alex can correct me if I'm wrong, but I believe there is a
>> more atomic-like API being planned, which would only need one IOCTL to
>> do everything. However, that would be a more serious change to the
>> V4L2 interfaces, so should be decoupled from Request API itself.
> 
> Indeed, we discussed the possibility to setup and submit requests in
> one syscall, similarly (at least in spirit) to the DRM atomic API.
> 
> This has only been discussed though, and as a feature to consider
> *after* the request API is merged for codecs (as the more complex
> camera use-cases would benefit more from it). As Tomasz mentioned, the
> overhead of ioctls is somehow negligible compared to the workload of
> the encoding/decoding itself, although I suppose it can still add up.
> The main advantage I can see for this is a simpler and less
> error-prone setup of requests for user-space.

Indeed, atomic API should be nicer from a userspace perspective and a bit more
optimal for at least some of workloads. Would be good to have it.