On 01/12/17 16:48, Devin Heitmueller wrote:
> Hello,
>
> It looks like a recent patch causes VA-API H.264 encode to stop working and
> an assertion to be thrown. I ran a git bisect and narrowed it down to the
> following commit:
>
> 32a618a948c20f18db102d0b0976790222a57105 is the first bad commit
> commit 32a618a948c20f18db102d0b0976790222a57105
> Author: Mark Thompson <s...@jkqxz.net>
> Date: Wed Oct 18 19:46:53 2017 +0100
>
> vaapi_h264: Do not use deprecated header type
>
> SEI headers should be inserted as generic raw data (the old specific
> type has been deprecated in libva2).
>
>
> When run with the above patch, I get the following output:
>
> [h264_vaapi @ 0x37d0a20] Warning: some packed headers are not supported (want
> 0xd, got 0xb).
> [h264_vaapi @ 0x37d0a20] The encode compression level option is not supported
> with this VAAPI version.
> ffmpeg: i965_drv_video.c:352: va_enc_packed_type_to_idx: Assertion `0' failed.
>
> Here’s the vainfo output which provides the version info for the driver,
> va-api version, etc. This is on a Haswell system running Centos 7.
>
> libva info: VA-API version 0.34.0
> libva info: va_getDriverName() returns 0
> libva info: Trying to open /usr/lib64/dri/i965_drv_video.so
> libva info: Found init function __vaDriverInit_0_34
> libva info: va_openDriver() returns 0
> vainfo: VA-API version: 0.34 (libva 1.2.1)
> vainfo: Driver version: Intel i965 driver - 1.2.2
Upgrading to a version less than four years old might be a plan - I admit we do
notionally support that version because of old RHEL/CentOS, but it is not well
tested (as you are finding).
>
> I’m using the following command line for testing:
>
> ./ffmpeg -y -vaapi_device /dev/dri/card0 -i /home/devin/inputfile.ts -vf
> 'format=nv12,hwupload' -c:v h264_vaapi out.mp4
>
> Any suggestions that could be offered would be greatly appreciated. Likewise
> please let me know if there is any other information I can provide that would
> assist in getting this resolved.
Try this? (Not tested, hardware which can run a version that old isn't
immediately to hand.)
--- a/libavcodec/vaapi_encode_h264.c
+++ b/libavcodec/vaapi_encode_h264.c
@@ -261,7 +261,8 @@ static int
vaapi_encode_h264_write_extra_header(AVCodecContext *avctx,
return 0;
#if !CONFIG_VAAPI_1
- } else if (priv->sei_cbr_workaround_needed) {
+ } else if (priv->sei_cbr_workaround_needed &&
+ ctx->va_packed_headers & VA_ENC_PACKED_HEADER_SLICE) {
// Insert a zero-length header using the old SEI type. This is
// required to avoid triggering broken behaviour on Intel platforms
// in CBR mode where an invalid SEI message is generated by the
(Probably wants a comment pointing at the driver code to explain why that
condition, too.)
- Mark
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
