[FFmpeg-devel] [PATCH] avcodec/vp9: avx2 implementation of ipred_dl_16x16_16

vp9_diag_downleft_16x16_12bpp_ssse3: 32.0 vp9_diag_downleft_16x16_12bpp_avx: 32.4 vp9_diag_downleft_16x16_12bpp_avx2: 25.5 Benchmarked with 1 runs Signed-off-by: Ilia --- libavcodec/x86/vp9dsp_init_16bpp.c| 2 ++ libavcodec/x86/vp9intrapred_16bpp.asm | 39 +++ 2 files

[FFmpeg-devel] [PATCH] libavcodec/vp9 ipred_dl_32x32_16 avx2 version

vp9_diag_downleft_32x32_8bpp_c: 580.2 vp9_diag_downleft_32x32_8bpp_sse2: 75.6 vp9_diag_downleft_32x32_8bpp_ssse3: 73.7 vp9_diag_downleft_32x32_8bpp_avx: 72.7 vp9_diag_downleft_32x32_10bpp_c: 1101.2 vp9_diag_downleft_32x32_10bpp_sse2: 145.4 vp9_diag_downleft_32x32_10bpp_ssse3: 137.5 vp9_diag_downlef

[FFmpeg-devel] [PATCH] libavcodec/vp9: ipred_dl_32x32_16 avx2 implementation

vp9_diag_downleft_32x32_8bpp_c: 580.2 vp9_diag_downleft_32x32_8bpp_sse2: 75.6 vp9_diag_downleft_32x32_8bpp_ssse3: 73.7 vp9_diag_downleft_32x32_8bpp_avx: 72.7 vp9_diag_downleft_32x32_10bpp_c: 1101.2 vp9_diag_downleft_32x32_10bpp_sse2: 145.4 vp9_diag_downleft_32x32_10bpp_ssse3: 137.5 vp9_diag_downlef

[FFmpeg-devel] [PATCH] avcodec/vp9: ipred_dr_16x16_16 avx2 implementation

vp9_diag_downright_16x16_12bpp_c: 149.0 vp9_diag_downright_16x16_12bpp_sse2: 67.8 vp9_diag_downright_16x16_12bpp_ssse3: 45.6 vp9_diag_downright_16x16_12bpp_avx: 36.6 vp9_diag_downright_16x16_12bpp_avx2: 25.5 ~30% faster than avx Signed-off-by: Ilia Valiakhmetov --- libavcodec/x86

[FFmpeg-devel] [PATCH] avcodec/vp9: ipred_dr_16x16_16 avx2 implementation

Signed-off-by: Ilia Valiakhmetov --- libavcodec/x86/vp9dsp_init_16bpp.c| 2 ++ libavcodec/x86/vp9intrapred_16bpp.asm | 56 +++ 2 files changed, 58 insertions(+) diff --git a/libavcodec/x86/vp9dsp_init_16bpp.c b/libavcodec/x86/vp9dsp_init_16bpp.c index

[FFmpeg-devel] [PATCH] avcodec/vp9: ipred_dr_16x16_16 avx2 implementation

Signed-off-by: Ilia Valiakhmetov --- libavcodec/x86/vp9dsp_init_16bpp.c| 2 ++ libavcodec/x86/vp9intrapred_16bpp.asm | 56 +++ 2 files changed, 58 insertions(+) diff --git a/libavcodec/x86/vp9dsp_init_16bpp.c b/libavcodec/x86/vp9dsp_init_16bpp.c index

[FFmpeg-devel] [PATCHv2 1/2] avcodec: add execute3() api to utilize the main function of avpriv_slicethread_create().

Signed-off-by: Ilia Valiakhmetov --- libavcodec/avcodec.h | 7 ++- libavcodec/options.c | 1 + libavcodec/pthread_slice.c | 26 -- libavcodec/utils.c | 14 ++ 4 files changed, 45 insertions(+), 3 deletions(-) diff --git a/libavcodec

[FFmpeg-devel] [PATCH 2/2] avcodec/pthread_slice: add main function support for avpriv_slicethread_create()

--- libavcodec/internal.h | 4 libavcodec/pthread_slice.c | 33 ++--- libavcodec/thread.h| 1 + libavutil/slicethread.h| 18 ++ 4 files changed, 37 insertions(+), 19 deletions(-) diff --git a/libavcodec/internal.h b/libavcodec/in

[FFmpeg-devel] [PATCHv2 2/2] avcodec/pthread_slice: add ff_slice_thread_execute_with_mainfunc()

Signed-off-by: Ilia Valiakhmetov v2: --- libavcodec/internal.h | 4 libavcodec/pthread_slice.c | 22 -- libavcodec/thread.h| 4 +++- 3 files changed, 27 insertions(+), 3 deletions(-) diff --git a/libavcodec/internal.h b/libavcodec/internal.h index

[FFmpeg-devel] [PATCH 2/2] avcodec/vp9: change avctx->execute3 in favor of ff_slice_thread_execute_with_mainfunc()

Signed-off-by: Ilia Valiakhmetov v8: --- libavcodec/vp9.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/libavcodec/vp9.c b/libavcodec/vp9.c index b780262..a71045e 100644 --- a/libavcodec/vp9.c +++ b/libavcodec/vp9.c @@ -1628,7 +1628,7 @@ FF_ENABLE_DEPRECATION_WARNINGS

[FFmpeg-devel] [PATCH] avcodec/vp9: add tile threading support

Signed-off-by: Ilia Valiakhmetov --- Changelog | 1 + 1 file changed, 1 insertion(+) diff --git a/Changelog b/Changelog index cae5254..8a4818a 100644 --- a/Changelog +++ b/Changelog @@ -43,6 +43,7 @@ version : - add --disable-autodetect build switch - drop deprecated qtkit input device (use

[FFmpeg-devel] [PATCH] Changelog: add vp9 tile threading support

Signed-off-by: Ilia Valiakhmetov --- Changelog | 1 + 1 file changed, 1 insertion(+) diff --git a/Changelog b/Changelog index 22928de..ca0758a 100644 --- a/Changelog +++ b/Changelog @@ -46,6 +46,7 @@ version : - haas audio filter - SUP/PGS subtitle muxer - convolve video filter +- VP9 tile

[FFmpeg-devel] [PATCH] avcodec/vp9: add 64-bit ipred_dr_32x32_16 avx2 implementation

vp9_diag_downright_32x32_12bpp_c: 429.7 vp9_diag_downright_32x32_12bpp_sse2: 158.9 vp9_diag_downright_32x32_12bpp_ssse3: 144.6 vp9_diag_downright_32x32_12bpp_avx: 141.0 vp9_diag_downright_32x32_12bpp_avx2: 73.8 Almost 50% faster than avx implementation --- libavcodec/x86/vp9dsp_init_16bpp.c|

[FFmpeg-devel] [PATCH] avcodec/vp9: AVX2 ipred_dl_32x32 improvement

avx Signed-off-by: Ilia Valiakhmetov --- libavcodec/x86/vp9intrapred_16bpp.asm | 47 --- 1 file changed, 33 insertions(+), 14 deletions(-) diff --git a/libavcodec/x86/vp9intrapred_16bpp.asm b/libavcodec/x86/vp9intrapred_16bpp.asm index 8d8d65e..33a8a7f 100644

[FFmpeg-devel] [PATCH] avcodec/vp9: AVX2 ipred_vl_16x16

vp9_vert_left_16x16_12bpp_c: 273.8 vp9_vert_left_16x16_12bpp_sse2: 69.4 vp9_vert_left_16x16_12bpp_ssse3: 35.3 vp9_vert_left_16x16_12bpp_avx: 34.6 vp9_vert_left_16x16_12bpp_avx2: 22.4 ~35% faster than avx Signed-off-by: Ilia Valiakhmetov --- libavcodec/x86/vp9dsp_init_16bpp.c| 2

[FFmpeg-devel] [PATCH 1/2] avcodec: add execute3() api to utilize the main function of avpriv_slicethread_create().

Signed-off-by: Ilia Valiakhmetov --- libavcodec/avcodec.h | 7 ++- libavcodec/options.c | 1 + libavcodec/pthread_slice.c | 27 +-- libavcodec/utils.c | 13 + 4 files changed, 45 insertions(+), 3 deletions(-) diff --git a/libavcodec

[FFmpeg-devel] [PATCH 0/2] Tile threading support for vp9

argument - main function for avpriv_slicethread_create(), it is used for the loopfilter. Ilia Valiakhmetov (2): avcodec: add execute3() api to utilize the main function of avpriv_slicethread_create(). avcodec/vp9: Add tile threading support libavcodec/avcodec.h | 7 +- libavcodec