| Message ID | CAJZRZVkuSRo=bBMprdUhzc=3oCPpn84HMSVmNhzPTrZ5JDKNpg@mail.gmail.com |
|---|---|
| State | New |
| Headers | show |
| Series | [FFmpeg-devel,v2] avcodec/cuvidddec: Guess pixel format based on probed bit depth and chroma sampling | expand |
| Context | Check | Description |
|---|---|---|
| yinshiyou/commit_msg_loongarch64 | warning | The first line of the commit message must start with a context terminated by a colon and a space, for example "lavu/opt: " or "doc: ". |
| andriy/commit_msg_x86 | warning | The first line of the commit message must start with a context terminated by a colon and a space, for example "lavu/opt: " or "doc: ". |
| andriy/make_x86 | success | Make finished |
| andriy/make_fate_x86 | success | Make fate finished |
On 02/08/2024 15:31, Roman Arzumanyan wrote: > Hello world, > > Please find v2 of the previous patch attached. > Both bit depth and chroma sampling (420 and 444 as supported by Nvdec) are > now taken into account when selecting sw_pix_fmt. It's actually AV_PIX_FMT_YUV444P16 for both 10 and 12 bit, due to how the bits are aligned opposite to what ffmpeg expects in AV_PIX_FMT_YUV444P10. No need to resend for that, will ammend locally.
Thanks a lot Timo ! Just for my own education: in my application I follow the usual sequence of API calls to init hw device context and open the codec. What's the right place to probe the input? I was under the impression that it's done within avformat_find_stream_info(). пт, 2 авг. 2024 г. в 16:57, Timo Rothenpieler <timo@rothenpieler.org>: > On 02/08/2024 15:31, Roman Arzumanyan wrote: > > Hello world, > > > > Please find v2 of the previous patch attached. > > Both bit depth and chroma sampling (420 and 444 as supported by Nvdec) > are > > now taken into account when selecting sw_pix_fmt. > > It's actually AV_PIX_FMT_YUV444P16 for both 10 and 12 bit, due to how > the bits are aligned opposite to what ffmpeg expects in > AV_PIX_FMT_YUV444P10. > > No need to resend for that, will ammend locally. > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". >
On 02.08.2024 16:05, Roman Arzumanyan wrote: > Thanks a lot Timo ! > > Just for my own education: in my application I follow the usual sequence of > API calls to init hw device context and open the codec. What's the right > place to probe the input? I was under the impression that it's done within > avformat_find_stream_info(). ffmpeg.c calls avcodec_parameters_to_context() on the codecpars got from the demuxer/probing phase. I'm not actually sure where the probing itself happens.
From a0c0c8497e75987ae771a466c9f0fce5c3ef106c Mon Sep 17 00:00:00 2001 From: Roman Arzumanyan <r.arzumanyan@visionlabs.ai> Date: Thu, 1 Aug 2024 16:35:22 +0300 Subject: [PATCH] Guess pixel format based on bit depth and chroma sampling: --- libavcodec/cuviddec.c | 25 ++++++++++++++++++++++--- 1 file changed, 22 insertions(+), 3 deletions(-) diff --git a/libavcodec/cuviddec.c b/libavcodec/cuviddec.c index f88ad75e88..a7185fecbf 100644 --- a/libavcodec/cuviddec.c +++ b/libavcodec/cuviddec.c @@ -834,22 +834,41 @@ static av_cold int cuvid_decode_init(AVCodecContext *avctx) int ret = 0; enum AVPixelFormat pix_fmts[3] = { AV_PIX_FMT_CUDA, - AV_PIX_FMT_NV12, + AV_PIX_FMT_NONE, AV_PIX_FMT_NONE }; int probed_width = avctx->coded_width ? avctx->coded_width : 1280; int probed_height = avctx->coded_height ? avctx->coded_height : 720; - int probed_bit_depth = 8; + int probed_bit_depth = 8, is_yuv444 = 0; const AVPixFmtDescriptor *probe_desc = av_pix_fmt_desc_get(avctx->pix_fmt); if (probe_desc && probe_desc->nb_components) probed_bit_depth = probe_desc->comp[0].depth; + if (probe_desc && !probe_desc->log2_chroma_w && !probe_desc->log2_chroma_h) + is_yuv444 = 1; + + // Pick pixel format based on bit depth and chroma sampling. + // Only 420 and 444 sampling are supported by HW so far, no need to check for 422. + switch (probed_bit_depth) { + case 8: + pix_fmts[1] = is_yuv444 ? AV_PIX_FMT_YUV444P : AV_PIX_FMT_NV12; + break; + case 10: + pix_fmts[1] = is_yuv444 ? AV_PIX_FMT_YUV444P10 : AV_PIX_FMT_P010; + break; + case 12: + pix_fmts[1] = is_yuv444 ? AV_PIX_FMT_YUV444P16 : AV_PIX_FMT_P016; + break; + default: + break; + } + ctx->pkt = avctx->internal->in_pkt; // Accelerated transcoding scenarios with 'ffmpeg' require that the // pix_fmt be set to AV_PIX_FMT_CUDA early. The sw_pix_fmt, and the // pix_fmt for non-accelerated transcoding, do not need to be correct - // but need to be set to something. We arbitrarily pick NV12. + // but need to be set to something. ret = ff_get_format(avctx, pix_fmts); if (ret < 0) { av_log(avctx, AV_LOG_ERROR, "ff_get_format failed: %d\n", ret); -- 2.34.1