[FFmpeg-devel] [PATCH] avcodec/nvenc: remove useless context pop

Signed-off-by: leozhang --- libavcodec/nvenc.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/libavcodec/nvenc.c b/libavcodec/nvenc.c index ae25afd7f8..fdf06d6102 100644 --- a/libavcodec/nvenc.c +++ b/libavcodec/nvenc.c @@ -607,9 +607,6 @@ static av_cold int nvenc_check_device

[FFmpeg-devel] [PATCH] avcodec/nvenc: remove useless context pop

Signed-off-by: leozhang --- libavcodec/nvenc.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/libavcodec/nvenc.c b/libavcodec/nvenc.c index ae25afd7f8..fdf06d6102 100644 --- a/libavcodec/nvenc.c +++ b/libavcodec/nvenc.c @@ -607,9 +607,6 @@ static av_cold int nvenc_check_device

[FFmpeg-devel] [PATCH v2] avcodec/cuviddec: fix CUDA_ERROR_INVALID_CONTEXT error found by cuda-memcheck tool

Test command like below: cuda-memcheck ./ffmpeg -hwaccel cuvid -c:v h264_cuvid -i input_file -c:v h264_nvenc -f null - Signed-off-by: leozhang --- libavcodec/cuviddec.c | 7 +++ 1 file changed, 7 insertions(+) diff --git a/libavcodec/cuviddec.c b/libavcodec/cuviddec.c index 5e698d4cd0

Re: [FFmpeg-devel] [PATCH] avcodec/cuviddec: fix CUDA_ERROR_INVALID_CONTEXT error found by cuda-memcheck tool

Timo Rothenpieler 于2020年11月20日周五 上午12:29写道： > > On 18.11.2020 09:24, leozhang wrote: > > Test command like below: > > cuda-memcheck ./ffmpeg -hwaccel cuvid -c:v h264_cuvid -i input_file -c:v > > h264_nvenc -f null - > > > > Signed-off-by: leozhang >

[FFmpeg-devel] [PATCH] avcodec/cuviddec: fix CUDA_ERROR_INVALID_CONTEXT error found by cuda-memcheck tool

Test command like below: cuda-memcheck ./ffmpeg -hwaccel cuvid -c:v h264_cuvid -i input_file -c:v h264_nvenc -f null - Signed-off-by: leozhang --- libavcodec/cuviddec.c | 15 +++ 1 file changed, 15 insertions(+) diff --git a/libavcodec/cuviddec.c b/libavcodec/cuviddec.c index

[FFmpeg-devel] [PATCH v2] nv-codec-headers: add cuMemcpy

Signed-off-by: leozhang --- include/ffnvcodec/dynlink_cuda.h | 2 ++ include/ffnvcodec/dynlink_loader.h | 4 2 files changed, 6 insertions(+) diff --git a/include/ffnvcodec/dynlink_cuda.h b/include/ffnvcodec/dynlink_cuda.h index 117fdc5..baafa49 100644 --- a/include/ffnvcodec

Re: [FFmpeg-devel] [PATCH] nv-codec-headers: add cuMemcpy

Timo Rothenpieler 于2020年11月13日周五 下午8:20写道： > > Please also add cuMemcpyAsync while at it. Will add it. > > What for and where is this needed? cuMemcpy is used to copy 1D array between host and device. For example, the gauss blur filter weights can be calculated by CPU, then memcpy to device memory

[FFmpeg-devel] [PATCH] nv-codec-headers: add cuMemcpy

Signed-off-by: leozhang --- include/ffnvcodec/dynlink_cuda.h | 1 + include/ffnvcodec/dynlink_loader.h | 2 ++ 2 files changed, 3 insertions(+) diff --git a/include/ffnvcodec/dynlink_cuda.h b/include/ffnvcodec/dynlink_cuda.h index 117fdc5..1686894 100644 --- a/include/ffnvcodec/dynlink_cuda.h

[FFmpeg-devel] [RFC] support half-precision float in *_cuda filters

Half precision float is supported for those devices sm capacity >=5.3. Theoretically, half float can achieve 2x performance speed up. So is there any plan to support half float in FFmpeg? Or any problem so that half float can't be met up? BR

[FFmpeg-devel] [PATCH] avfilter/vf_scale_cuda: unload cuModule when uninit

Signed-off-by: leozhang --- libavfilter/vf_scale_cuda.c | 8 1 file changed, 8 insertions(+) diff --git a/libavfilter/vf_scale_cuda.c b/libavfilter/vf_scale_cuda.c index 1ffb73f831..46d69b00ee 100644 --- a/libavfilter/vf_scale_cuda.c +++ b/libavfilter/vf_scale_cuda.c @@ -119,7 +119,15

Re: [FFmpeg-devel] [PATCH] avutil/video_enc_params: fix code comment

zhilizhao 于2020年8月13日周四 下午12:55写道： > > > > > On Aug 13, 2020, at 11:57 AM, leozhang wrote: > > > > Signed-off-by: leozhang > > --- > > libavutil/video_enc_params.h | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > &