On 8/22/2018 8:44 PM, Mark Thompson wrote: > Including a unit test. > --- > libavcodec/Makefile | 3 +- > libavcodec/h264_levels.c | 130 +++++++++++++++++++++++ > libavcodec/h264_levels.h | 53 ++++++++++ > libavcodec/tests/.gitignore | 1 + > libavcodec/tests/h264_levels.c | 183 +++++++++++++++++++++++++++++++++ > tests/fate/libavcodec.mak | 5 + > 6 files changed, 374 insertions(+), 1 deletion(-) > create mode 100644 libavcodec/h264_levels.c > create mode 100644 libavcodec/h264_levels.h > create mode 100644 libavcodec/tests/h264_levels.c > > diff --git a/libavcodec/Makefile b/libavcodec/Makefile > index cbbfc9af2e..d07a9073af 100644 > --- a/libavcodec/Makefile > +++ b/libavcodec/Makefile > @@ -355,7 +355,7 @@ OBJS-$(CONFIG_H264_OMX_ENCODER) += omx.o > OBJS-$(CONFIG_H264_QSV_DECODER) += qsvdec_h2645.o > OBJS-$(CONFIG_H264_QSV_ENCODER) += qsvenc_h264.o > OBJS-$(CONFIG_H264_RKMPP_DECODER) += rkmppdec.o > -OBJS-$(CONFIG_H264_VAAPI_ENCODER) += vaapi_encode_h264.o > +OBJS-$(CONFIG_H264_VAAPI_ENCODER) += h264_levels.o vaapi_encode_h264.o > OBJS-$(CONFIG_H264_VIDEOTOOLBOX_ENCODER) += videotoolboxenc.o > OBJS-$(CONFIG_H264_V4L2M2M_DECODER) += v4l2_m2m_dec.o > OBJS-$(CONFIG_H264_V4L2M2M_ENCODER) += v4l2_m2m_enc.o > @@ -1134,6 +1134,7 @@ TESTPROGS-$(CONFIG_IDCTDSP) += dct > TESTPROGS-$(CONFIG_IIRFILTER) += iirfilter > TESTPROGS-$(HAVE_MMX) += motion > TESTPROGS-$(CONFIG_MPEGVIDEO) += mpeg12framerate > +TESTPROGS-$(CONFIG_H264_VAAPI_ENCODER) += h264_levels
Needing h264_vaapi_encoder to test this is not ideal. You'd be limiting the amount of fate clients testing an internal module to those with an unrelated external encoder enabled. Maybe just add h264_levels.o to h264parse, and check for that instead? Alternatively, since you need to add that object to h264_metadata to fix the build failure Michael reported for patch 28, you could check for that bsf instead. _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
