[FFmpeg-devel] [PATCH] lavc/cuvid: fail early if GPU can't handle video parameters

[Pavel Koshevoy]

On 1/22/17 11:13 AM, Philip Langdale wrote:
> On Sat, 21 Jan 2017 10:27:30 -0700
> pkoshevoy at gmail.com wrote:
>
>> From: Pavel Koshevoy <pkoshevoy at gmail.com>
>>
>> Evidently CUVID doesn't support decoding 422 or 444 chroma formats,
>> and only a limited set of resolutions per codec are supported.
>>
>> Unfortunately CUVID silently drops packets for video of unsupported
>> resolution.  However, it will error-out at cuvidCreateDecoder call
>> if the indicated video resolution is not supported.
>>
>> Given that stream resolution and pixel format are typically known as
>> a result of probing it is better to use this information during
>> avcodec_open2 call to fail immediately, rather than proceeding to
>> decode and never receiving any frames from the decoder nor receiving
>> any indication of decode failure.
> So I'm confused. I agree that these errors do not emerge early enough
> to induce failure at decoder init time, but they should all trigger
> errors at first-frame decode time. That's certainly true for the
> chroma format error, because I added that logic.
>
> Assuming that is true, you should just be dealing with first-frame
> failure and then falling back to SW decoding, or whatever policy
> you want in your app (I assume 8K SW decoding won't go well...).
>

Yeah, but I don't even get a 1st frame failure with YUV420P 8K from 
h264_cuvid -- all I ever get is a 0 from avcodec_send_packet and 
AVERROR(EAGAIN) from avcodec_receive_frame.  This is what I've observed 
with GeForce GT 730 and GeForce GTX 1060 on openSUSE 42.2 and Ubuntu 
16.04 with latest nvidia drivers.

I will try to make my application more robust too -- I will have to 
buffer all the packets sent to the decoder until the 1st frame either 
succeeds or fails, so I can re-send the same packets to another decoder 
in case of 1st frame failure.

   Pavel.