[FFmpeg-devel] [PATCH] checkasm: add support for vvc alf
vvc_alf_filter_chroma_4x4_10_c: 661.9 vvc_alf_filter_chroma_4x4_10_avx2: 138.4 vvc_alf_filter_chroma_4x8_10_c: 1287.4 vvc_alf_filter_chroma_4x8_10_avx2: 258.9 vvc_alf_filter_chroma_4x12_10_c: 2017.1 vvc_alf_filter_chroma_4x12_10_avx2: 375.9 vvc_alf_filter_chroma_4x16_10_c: 2604.6 vvc_alf_filter_chroma_4x16_10_avx2: 489.9 vvc_alf_filter_chroma_4x20_10_c: 3074.6 vvc_alf_filter_chroma_4x20_10_avx2: 626.1 vvc_alf_filter_chroma_4x24_10_c: 3674.1 vvc_alf_filter_chroma_4x24_10_avx2: 728.1 vvc_alf_filter_chroma_4x28_10_c: 4427.6 vvc_alf_filter_chroma_4x28_10_avx2: 876.6 vvc_alf_filter_chroma_8x4_10_c: 1269.1 vvc_alf_filter_chroma_8x4_10_avx2: 293.6 vvc_alf_filter_chroma_8x8_10_c: 2511.6 vvc_alf_filter_chroma_8x8_10_avx2: 497.4 vvc_alf_filter_chroma_8x12_10_c: 3645.9 vvc_alf_filter_chroma_8x12_10_avx2: 735.6 vvc_alf_filter_chroma_8x16_10_c: 8241.1 vvc_alf_filter_chroma_8x16_10_avx2: 972.4 vvc_alf_filter_chroma_8x20_10_c: 6074.1 vvc_alf_filter_chroma_8x20_10_avx2: 1269.6 vvc_alf_filter_chroma_8x24_10_c: 7268.6 vvc_alf_filter_chroma_8x24_10_avx2: 1447.4 vvc_alf_filter_chroma_8x28_10_c: 8707.4 vvc_alf_filter_chroma_8x28_10_avx2: 1727.9 vvc_alf_filter_chroma_12x4_10_c: 1831.9 vvc_alf_filter_chroma_12x4_10_avx2: 386.4 vvc_alf_filter_chroma_12x8_10_c: 3847.6 vvc_alf_filter_chroma_12x8_10_avx2: 738.6 vvc_alf_filter_chroma_12x12_10_c: 5460.6 vvc_alf_filter_chroma_12x12_10_avx2: 1098.9 vvc_alf_filter_chroma_12x16_10_c: 7642.6 vvc_alf_filter_chroma_12x16_10_avx2: 1445.4 vvc_alf_filter_chroma_12x20_10_c: 9056.6 vvc_alf_filter_chroma_12x20_10_avx2: 1898.9 vvc_alf_filter_chroma_12x24_10_c: 10862.4 vvc_alf_filter_chroma_12x24_10_avx2: 3073.1 vvc_alf_filter_chroma_12x28_10_c: 13005.6 vvc_alf_filter_chroma_12x28_10_avx2: 2587.9 vvc_alf_filter_chroma_16x4_10_c: 4406.1 vvc_alf_filter_chroma_16x4_10_avx2: 249.9 vvc_alf_filter_chroma_16x8_10_c: 4846.9 vvc_alf_filter_chroma_16x8_10_avx2: 478.6 vvc_alf_filter_chroma_16x12_10_c: 7247.1 vvc_alf_filter_chroma_16x12_10_avx2: 704.4 vvc_alf_filter_chroma_16x16_10_c: 10163.1 vvc_alf_filter_chroma_16x16_10_avx2: 934.4 vvc_alf_filter_chroma_16x20_10_c: 12041.1 vvc_alf_filter_chroma_16x20_10_avx2: 1196.1 vvc_alf_filter_chroma_16x24_10_c: 14448.9 vvc_alf_filter_chroma_16x24_10_avx2: 1390.9 vvc_alf_filter_chroma_16x28_10_c: 16846.9 vvc_alf_filter_chroma_16x28_10_avx2: 1664.9 vvc_alf_filter_chroma_20x4_10_c: 3036.6 vvc_alf_filter_chroma_20x4_10_avx2: 373.6 vvc_alf_filter_chroma_20x8_10_c: 6190.4 vvc_alf_filter_chroma_20x8_10_avx2: 715.6 vvc_alf_filter_chroma_20x12_10_c: 9278.9 vvc_alf_filter_chroma_20x12_10_avx2: 1096.4 vvc_alf_filter_chroma_20x16_10_c: 21400.9 vvc_alf_filter_chroma_20x16_10_avx2: 1451.9 vvc_alf_filter_chroma_20x20_10_c: 16755.6 vvc_alf_filter_chroma_20x20_10_avx2: 1862.9 vvc_alf_filter_chroma_20x24_10_c: 18029.9 vvc_alf_filter_chroma_20x24_10_avx2: 2110.9 vvc_alf_filter_chroma_20x28_10_c: 21003.6 vvc_alf_filter_chroma_20x28_10_avx2: 2585.4 vvc_alf_filter_chroma_24x4_10_c: 6182.9 vvc_alf_filter_chroma_24x4_10_avx2: 501.6 vvc_alf_filter_chroma_24x8_10_c: 7429.9 vvc_alf_filter_chroma_24x8_10_avx2: 964.1 vvc_alf_filter_chroma_24x12_10_c: 11429.9 vvc_alf_filter_chroma_24x12_10_avx2: 1436.9 vvc_alf_filter_chroma_24x16_10_c: 14822.4 vvc_alf_filter_chroma_24x16_10_avx2: 1983.6 vvc_alf_filter_chroma_24x20_10_c: 18020.4 vvc_alf_filter_chroma_24x20_10_avx2: 2418.9 vvc_alf_filter_chroma_24x24_10_c: 22183.1 vvc_alf_filter_chroma_24x24_10_avx2: 2896.6 vvc_alf_filter_chroma_24x28_10_c: 25198.4 vvc_alf_filter_chroma_24x28_10_avx2: 3298.6 vvc_alf_filter_chroma_28x4_10_c: 4363.6 vvc_alf_filter_chroma_28x4_10_avx2: 623.1 vvc_alf_filter_chroma_28x8_10_c: 8466.9 vvc_alf_filter_chroma_28x8_10_avx2: 1242.4 vvc_alf_filter_chroma_28x12_10_c: 13418.9 vvc_alf_filter_chroma_28x12_10_avx2: 1786.4 vvc_alf_filter_chroma_28x16_10_c: 16942.9 vvc_alf_filter_chroma_28x16_10_avx2: 2396.9 vvc_alf_filter_chroma_28x20_10_c: 21186.6 vvc_alf_filter_chroma_28x20_10_avx2: 2968.1 vvc_alf_filter_chroma_28x24_10_c: 25374.9 vvc_alf_filter_chroma_28x24_10_avx2: 11726.6 vvc_alf_filter_chroma_28x28_10_c: 29567.4 vvc_alf_filter_chroma_28x28_10_avx2: 4132.9 vvc_alf_filter_luma_4x4_10_c: 1032.6 vvc_alf_filter_luma_4x4_10_avx2: 239.6 vvc_alf_filter_luma_4x8_10_c: 2833.4 vvc_alf_filter_luma_4x8_10_avx2: 459.1 vvc_alf_filter_luma_4x12_10_c: 3114.9 vvc_alf_filter_luma_4x12_10_avx2: 670.1 vvc_alf_filter_luma_4x16_10_c: 4366.6 vvc_alf_filter_luma_4x16_10_avx2: 891.1 vvc_alf_filter_luma_4x20_10_c: 5139.6 vvc_alf_filter_luma_4x20_10_avx2: 1132.1 vvc_alf_filter_luma_4x24_10_c: 6161.9 vvc_alf_filter_luma_4x24_10_avx2: 1358.1 vvc_alf_filter_luma_4x28_10_c: 7178.6 vvc_alf_filter_luma_4x28_10_avx2: 1582.6 vvc_alf_filter_luma_8x4_10_c: 2122.9 vvc_alf_filter_luma_8x4_10_avx2: 489.4 vvc_alf_filter_luma_8x8_10_c: 4190.1 vvc_alf_filter_luma_8x8_10_avx2: 897.1 vvc_alf_filter_luma_8x12_10_c: 6112.9 vvc_alf_filter_luma_8x12_10_avx2: 1327.6 vvc_alf_filter_luma_8x16_10_c: 8631.4 vvc_alf_filter_luma_8x16_10_avx2: 1765.4 vvc_alf_filter_luma_8x20_10_c:
Re: [FFmpeg-devel] [PATCH 1/2] vvcdec: alf, add avx2 for luma and chroma filter
On Sun, Feb 26, 2023 at 1:48 PM Nuo Mi wrote: > got 11%~26% performance for 1080P and 4k video > > clipbefore after delta > RitualDance_1920x1080_60_10_420_32_LD.2635 4322.8% > RitualDance_1920x1080_60_10_420_37_RA.266 43 4811.6% > Tango2_3840x2160_60_10_420_27_LD.2667.9 1026.5% > --- > libavcodec/vvcdsp.c | 3 + > libavcodec/x86/Makefile | 2 + > libavcodec/x86/vvc_alf.asm | 301 +++ > libavcodec/x86/vvcdsp.h | 44 + > libavcodec/x86/vvcdsp_init.c | 81 ++ > 5 files changed, 431 insertions(+) > create mode 100644 libavcodec/x86/vvc_alf.asm > create mode 100644 libavcodec/x86/vvcdsp.h > create mode 100644 libavcodec/x86/vvcdsp_init.c > > diff --git a/libavcodec/vvcdsp.c b/libavcodec/vvcdsp.c > index 801bd0189d..399631503f 100644 > --- a/libavcodec/vvcdsp.c > +++ b/libavcodec/vvcdsp.c > @@ -313,4 +313,7 @@ void ff_vvc_dsp_init(VVCDSPContext *vvcdsp, int > bit_depth) > VVC_DSP(8); > break; > } > +#if ARCH_X86 > +ff_vvc_dsp_init_x86(vvcdsp, bit_depth); > +#endif > } > diff --git a/libavcodec/x86/Makefile b/libavcodec/x86/Makefile > index 118daca333..23b2fb42bb 100644 > --- a/libavcodec/x86/Makefile > +++ b/libavcodec/x86/Makefile > @@ -82,6 +82,7 @@ OBJS-$(CONFIG_VP9_DECODER) += > x86/vp9dsp_init.o\ >x86/vp9dsp_init_12bpp.o \ >x86/vp9dsp_init_16bpp.o > OBJS-$(CONFIG_WEBP_DECODER)+= x86/vp8dsp_init.o > +OBJS-$(CONFIG_VVC_DECODER) += x86/vvcdsp_init.o > > > # GCC inline assembly optimizations > @@ -202,4 +203,5 @@ X86ASM-OBJS-$(CONFIG_VP9_DECODER) += > x86/vp9intrapred.o\ >x86/vp9lpf_16bpp.o\ >x86/vp9mc.o \ >x86/vp9mc_16bpp.o > +X86ASM-OBJS-$(CONFIG_VVC_DECODER) += x86/vvc_alf.o > X86ASM-OBJS-$(CONFIG_WEBP_DECODER) += x86/vp8dsp.o > diff --git a/libavcodec/x86/vvc_alf.asm b/libavcodec/x86/vvc_alf.asm > new file mode 100644 > index 00..c3e4074be7 > --- /dev/null > +++ b/libavcodec/x86/vvc_alf.asm > @@ -0,0 +1,301 @@ > > +;** > +;* VVC Adaptive Loop Filter SIMD optimizations > +;* > +;* Copyright (c) 2023 Nuo Mi > +;* > +;* This file is part of FFmpeg. > +;* > +;* FFmpeg is free software; you can redistribute it and/or > +;* modify it under the terms of the GNU Lesser General Public > +;* License as published by the Free Software Foundation; either > +;* version 2.1 of the License, or (at your option) any later version. > +;* > +;* FFmpeg is distributed in the hope that it will be useful, > +;* but WITHOUT ANY WARRANTY; without even the implied warranty of > +;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU > +;* Lesser General Public License for more details. > +;* > +;* You should have received a copy of the GNU Lesser General Public > +;* License along with FFmpeg; if not, write to the Free Software > +;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA > 02110-1301 USA > > +;** > + > +%include "libavutil/x86/x86util.asm" > + > +SECTION_RODATA > + > +%macro PARAM_SHUFFE 1 > +%assign i (%1 * 2) > +%assign j ((i + 1) << 8) + (i) > +param_shuffe_%+%1: > +%rep 2 > +times 4 dw j > +times 4 dw (j + 0x0808) > +%endrep > +%endmacro > + > +PARAM_SHUFFE 0 > +PARAM_SHUFFE 1 > +PARAM_SHUFFE 2 > +PARAM_SHUFFE 3 > + > +dw_64: dd 64 > + > +SECTION .text > + > +%if HAVE_AVX2_EXTERNAL > + > +;%1-%3 out > +;%4 clip or filter > +%macro LOAD_LUMA_PARAMS_W16 4 > +%ifidn clip, %4 > +movum%1, [%4q + 0 * 32] > +movum%2, [%4q + 1 * 32] > +movum%3, [%4q + 2 * 32] > +%elifidn filter, %4 > +movuxm%1, [%4q + 0 * 16] > +movuxm%2, [%4q + 1 * 16] > +movuxm%3, [%4q + 2 * 16] > +pmovsxbwm%1, xm%1 > +pmovsxbwm%2, xm%2 > +pmovsxbwm%3, xm%3 > +%else > +%error "need filter or clip for the fourth param" > +%endif > +%endmacro > + > +%macro LOAD_LUMA_PARAMS_W16 6 > +LOAD_LUMA_PARAMS_W16%1, %2, %3, %4 > +;m%1 = 03 02 01 00 > +;m%2 = 07 06 05 04 > +;m%3 = 11 10 09 08 > + > +vshufpd m%5, m%1, m%2, 0b0011 ;06 02 05 01 > +vshufpd m%6, m%3, m%5, 0b1001 ;06 10 01 09 > + > +vshufpd m%1, m%1, m%6, 0b1100 ;06 03 09 00 > +vshufpd m%2, m%2, m%6, 0b0110 ;10 07 01 04 > +vshufpd
[FFmpeg-devel] [PATCH 1/2] vvcdec: alf, add avx2 for luma and chroma filter
got 11%~26% performance for 1080P and 4k video clipbefore after delta RitualDance_1920x1080_60_10_420_32_LD.2635 4322.8% RitualDance_1920x1080_60_10_420_37_RA.266 43 4811.6% Tango2_3840x2160_60_10_420_27_LD.2667.9 1026.5% --- libavcodec/vvcdsp.c | 3 + libavcodec/x86/Makefile | 2 + libavcodec/x86/vvc_alf.asm | 301 +++ libavcodec/x86/vvcdsp.h | 44 + libavcodec/x86/vvcdsp_init.c | 81 ++ 5 files changed, 431 insertions(+) create mode 100644 libavcodec/x86/vvc_alf.asm create mode 100644 libavcodec/x86/vvcdsp.h create mode 100644 libavcodec/x86/vvcdsp_init.c diff --git a/libavcodec/vvcdsp.c b/libavcodec/vvcdsp.c index 801bd0189d..399631503f 100644 --- a/libavcodec/vvcdsp.c +++ b/libavcodec/vvcdsp.c @@ -313,4 +313,7 @@ void ff_vvc_dsp_init(VVCDSPContext *vvcdsp, int bit_depth) VVC_DSP(8); break; } +#if ARCH_X86 +ff_vvc_dsp_init_x86(vvcdsp, bit_depth); +#endif } diff --git a/libavcodec/x86/Makefile b/libavcodec/x86/Makefile index 118daca333..23b2fb42bb 100644 --- a/libavcodec/x86/Makefile +++ b/libavcodec/x86/Makefile @@ -82,6 +82,7 @@ OBJS-$(CONFIG_VP9_DECODER) += x86/vp9dsp_init.o \ x86/vp9dsp_init_12bpp.o \ x86/vp9dsp_init_16bpp.o OBJS-$(CONFIG_WEBP_DECODER)+= x86/vp8dsp_init.o +OBJS-$(CONFIG_VVC_DECODER) += x86/vvcdsp_init.o # GCC inline assembly optimizations @@ -202,4 +203,5 @@ X86ASM-OBJS-$(CONFIG_VP9_DECODER) += x86/vp9intrapred.o\ x86/vp9lpf_16bpp.o\ x86/vp9mc.o \ x86/vp9mc_16bpp.o +X86ASM-OBJS-$(CONFIG_VVC_DECODER) += x86/vvc_alf.o X86ASM-OBJS-$(CONFIG_WEBP_DECODER) += x86/vp8dsp.o diff --git a/libavcodec/x86/vvc_alf.asm b/libavcodec/x86/vvc_alf.asm new file mode 100644 index 00..c3e4074be7 --- /dev/null +++ b/libavcodec/x86/vvc_alf.asm @@ -0,0 +1,301 @@ +;** +;* VVC Adaptive Loop Filter SIMD optimizations +;* +;* Copyright (c) 2023 Nuo Mi +;* +;* This file is part of FFmpeg. +;* +;* FFmpeg is free software; you can redistribute it and/or +;* modify it under the terms of the GNU Lesser General Public +;* License as published by the Free Software Foundation; either +;* version 2.1 of the License, or (at your option) any later version. +;* +;* FFmpeg is distributed in the hope that it will be useful, +;* but WITHOUT ANY WARRANTY; without even the implied warranty of +;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU +;* Lesser General Public License for more details. +;* +;* You should have received a copy of the GNU Lesser General Public +;* License along with FFmpeg; if not, write to the Free Software +;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA +;** + +%include "libavutil/x86/x86util.asm" + +SECTION_RODATA + +%macro PARAM_SHUFFE 1 +%assign i (%1 * 2) +%assign j ((i + 1) << 8) + (i) +param_shuffe_%+%1: +%rep 2 +times 4 dw j +times 4 dw (j + 0x0808) +%endrep +%endmacro + +PARAM_SHUFFE 0 +PARAM_SHUFFE 1 +PARAM_SHUFFE 2 +PARAM_SHUFFE 3 + +dw_64: dd 64 + +SECTION .text + +%if HAVE_AVX2_EXTERNAL + +;%1-%3 out +;%4 clip or filter +%macro LOAD_LUMA_PARAMS_W16 4 +%ifidn clip, %4 +movum%1, [%4q + 0 * 32] +movum%2, [%4q + 1 * 32] +movum%3, [%4q + 2 * 32] +%elifidn filter, %4 +movuxm%1, [%4q + 0 * 16] +movuxm%2, [%4q + 1 * 16] +movuxm%3, [%4q + 2 * 16] +pmovsxbwm%1, xm%1 +pmovsxbwm%2, xm%2 +pmovsxbwm%3, xm%3 +%else +%error "need filter or clip for the fourth param" +%endif +%endmacro + +%macro LOAD_LUMA_PARAMS_W16 6 +LOAD_LUMA_PARAMS_W16%1, %2, %3, %4 +;m%1 = 03 02 01 00 +;m%2 = 07 06 05 04 +;m%3 = 11 10 09 08 + +vshufpd m%5, m%1, m%2, 0b0011 ;06 02 05 01 +vshufpd m%6, m%3, m%5, 0b1001 ;06 10 01 09 + +vshufpd m%1, m%1, m%6, 0b1100 ;06 03 09 00 +vshufpd m%2, m%2, m%6, 0b0110 ;10 07 01 04 +vshufpd m%3, m%3, m%5, 0b0110 ;02 11 05 08 + +vpermpd m%1, m%1, 0b01_11_10_00 ;09 06 03 00 +vshufpd m%2, m%2, m%2, 0b1001 ;10 07 04 01 +vpermpd m%3, m%3, 0b10_00_01_11 ;11 08 05 02 +%endmacro + +%macro LOAD_LUMA_PARAMS_W4 6 +%ifidn clip,
Re: [FFmpeg-devel] [PATCH 1/2] lavfi/coreimage: suppress OpenGL deprecation warning
On Sun, Feb 26, 2023 at 1:48 AM Thilo Borgmann wrote: > > Am 25.02.23 um 16:26 schrieb Jun Zhao: > > From: Jun Zhao > > > > suppress OpenGL deprecation warning. > > > > Signed-off-by: Jun Zhao > > --- > > libavfilter/vf_coreimage.m | 3 ++- > > 1 file changed, 2 insertions(+), 1 deletion(-) > > > > diff --git a/libavfilter/vf_coreimage.m b/libavfilter/vf_coreimage.m > > index b1959861de..17b2cde6fe 100644 > > --- a/libavfilter/vf_coreimage.m > > +++ b/libavfilter/vf_coreimage.m > > @@ -22,7 +22,8 @@ > >* @file > >* Video processing based on Apple's CoreImage API > >*/ > > - > > +#define GL_SILENCE_DEPRECATION > > +#define CI_SILENCE_GL_DEPRECATION > > #import > > #import > > Why do you want to silence a framework warning? > Ok, will drop the fix ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-devel] videolan ffmpeg git issue [please dont push before reading]
On Sat, Feb 25, 2023 at 12:37:32PM +0530, Gyan Doshi wrote: > > > On 2023-02-25 06:44 am, Michael Niedermayer wrote: > > Hi all > > > > it seems videolan had a double RAID failure and was restored from backup > > that caused ffmpeg developer git for master and release/5.1 to rewind by > > a few commits. > > git pull gives 'WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED!' > > Is this expected? i guess so, i guess this was recreated when it was restored dig git.videolan.org. SSHFP should match it thx [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB Why not whip the teacher when the pupil misbehaves? -- Diogenes of Sinope signature.asc Description: PGP signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-devel] videolan ffmpeg git issue [please dont push before reading]
On Sat, Feb 25, 2023 at 11:45:51AM +0100, Michael Niedermayer wrote: > On Sat, Feb 25, 2023 at 02:14:44AM +0100, Michael Niedermayer wrote: > > Hi all > > > > it seems videolan had a double RAID failure and was restored from backup > > that caused ffmpeg developer git for master and release/5.1 to rewind by > > a few commits. > > It would be slightly messy if different commits are pushed on that rewond > > HEADs- So if you push something (when its possible again) please make sure > > that teh commits match the previous (pre restore) ones > > > > Its not the end of the world if others are pushed, we would then need > > a merge commit to reconcile these 2 histories > > Some update > > videolan took this opertunity to switch the repositories from EOL gitosis > to gitolite. This added some hickups > first being some nonworkingness with multiple admins in gitolite. Which > i belive was due to single keys being used with multiple users, i fixed > that for a 2nd key i added so it seems thats the issue > > and a 2nd issue that causes branches & heads to disappear from public view > when one pushed to them i suspect thats a permission / user/group issue > but i cannot verify or fix this. > ATM thresh is sleeping and i suspect these issues will be fixed once he > awakes The permissions have been fixed by thresh. Ive added the missing commits back, and pushing seems to work again. Ive also changed our github mirror config so it cannot receed as that happened temporarly as one of 2 sync scripts aparently force pushed teh receeded HEADs There are still some issues with git hooks. So pushing may not be protected against trailing whitespace or such things. Also the ML and IRC may not receive log mails. Thresh plans / is working on the hooks this evening. They are expected to be workinhg againg in the european morning thx [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB The educated differ from the uneducated as much as the living from the dead. -- Aristotle signature.asc Description: PGP signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-devel] [PATCH 2/2] fftools/ffmpeg_filter: initialize the 'o' to silence the warning
Am 25.02.23 um 16:26 schrieb Jun Zhao: From: Jun Zhao silence the warning: variable 'o' is used uninitialized whenever '&&' condition is false Signed-off-by: Jun Zhao --- fftools/ffmpeg_filter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) LGTM. -Thilo ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-devel] [PATCH 1/2] lavfi/coreimage: suppress OpenGL deprecation warning
Am 25.02.23 um 16:26 schrieb Jun Zhao: From: Jun Zhao suppress OpenGL deprecation warning. Signed-off-by: Jun Zhao --- libavfilter/vf_coreimage.m | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/libavfilter/vf_coreimage.m b/libavfilter/vf_coreimage.m index b1959861de..17b2cde6fe 100644 --- a/libavfilter/vf_coreimage.m +++ b/libavfilter/vf_coreimage.m @@ -22,7 +22,8 @@ * @file * Video processing based on Apple's CoreImage API */ - +#define GL_SILENCE_DEPRECATION +#define CI_SILENCE_GL_DEPRECATION #import #import Why do you want to silence a framework warning? -Thilo ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-devel] [PATCH 1/2] doc/examples/muxing: Remove unnecessary ret
ignore the patch 1-2, send with the wrong branch On Sat, Feb 25, 2023 at 11:25 PM Jun Zhao wrote: > > From: Jun Zhao > > Remove unnecessary ret and make the code more compact > > Signed-off-by: Jun Zhao > --- > doc/examples/muxing.c | 5 + > 1 file changed, 1 insertion(+), 4 deletions(-) > > diff --git a/doc/examples/muxing.c b/doc/examples/muxing.c > index 3acb778322..cd997d5431 100644 > --- a/doc/examples/muxing.c > +++ b/doc/examples/muxing.c > @@ -219,8 +219,6 @@ static AVFrame *alloc_audio_frame(enum AVSampleFormat > sample_fmt, >int sample_rate, int nb_samples) > { > AVFrame *frame = av_frame_alloc(); > -int ret; > - > if (!frame) { > fprintf(stderr, "Error allocating an audio frame\n"); > exit(1); > @@ -232,8 +230,7 @@ static AVFrame *alloc_audio_frame(enum AVSampleFormat > sample_fmt, > frame->nb_samples = nb_samples; > > if (nb_samples) { > -ret = av_frame_get_buffer(frame, 0); > -if (ret < 0) { > +if (av_frame_get_buffer(frame, 0) < 0) { > fprintf(stderr, "Error allocating an audio buffer\n"); > exit(1); > } > -- > 2.25.1 > > ___ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe". -- === Jun zhao/赵军 +++ ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
[FFmpeg-devel] [PATCH 2/2] fftools/ffmpeg_filter: initialize the 'o' to silence the warning
From: Jun Zhao silence the warning: variable 'o' is used uninitialized whenever '&&' condition is false Signed-off-by: Jun Zhao --- fftools/ffmpeg_filter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fftools/ffmpeg_filter.c b/fftools/ffmpeg_filter.c index 1f5bbf6c4d..3504a3cc0a 100644 --- a/fftools/ffmpeg_filter.c +++ b/fftools/ffmpeg_filter.c @@ -362,7 +362,7 @@ fail: static int filter_opt_apply(AVFilterContext *f, const char *key, const char *val) { -const AVOption *o; +const AVOption *o = NULL; int ret; ret = av_opt_set(f, key, val, AV_OPT_SEARCH_CHILDREN); -- 2.25.1 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
[FFmpeg-devel] [PATCH 1/2] lavfi/coreimage: suppress OpenGL deprecation warning
From: Jun Zhao suppress OpenGL deprecation warning. Signed-off-by: Jun Zhao --- libavfilter/vf_coreimage.m | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/libavfilter/vf_coreimage.m b/libavfilter/vf_coreimage.m index b1959861de..17b2cde6fe 100644 --- a/libavfilter/vf_coreimage.m +++ b/libavfilter/vf_coreimage.m @@ -22,7 +22,8 @@ * @file * Video processing based on Apple's CoreImage API */ - +#define GL_SILENCE_DEPRECATION +#define CI_SILENCE_GL_DEPRECATION #import #import -- 2.25.1 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
[FFmpeg-devel] [PATCH 2/2] doc/protocols: update rtsp options
From: Jun Zhao Split the rtsp options to muxer/demuxer, and update the options. Signed-off-by: Jun Zhao --- doc/protocols.texi | 64 ++ 1 file changed, 64 insertions(+) diff --git a/doc/protocols.texi b/doc/protocols.texi index 0df38d790c..5e9198e67c 100644 --- a/doc/protocols.texi +++ b/doc/protocols.texi @@ -1178,6 +1178,59 @@ Options can be set on the @command{ffmpeg}/@command{ffplay} command line, or set in code via @code{AVOption}s or in @code{avformat_open_input}. +@subsection Muxer +The following options are supported. + +@table @option +@item rtsp_transport +Set RTSP transport protocols. + +It accepts the following values: +@table @samp +@item udp +Use UDP as lower transport protocol. + +@item tcp +Use TCP (interleaving within the RTSP control channel) as lower +transport protocol. +@end table + +Default value is @samp{0}. + +@item rtsp_flags +Set RTSP flags. + +The following values are accepted: +@table @samp +@item latm +Use MP4A-LATM packetization instead of MPEG4-GENERIC for AAC. +@item rfc2190 +Use RFC 2190 packetization instead of RFC 4629 for H.263. +@item skip_rtcp +Don't send RTCP sender reports. +@item h264_mode0 +Use mode 0 for H.264 in RTP. +@item send_bye +Send RTCP BYE packets when finishing. +@end table + +Default value is @samp{0}. + + +@item min_port +Set minimum local UDP port. Default value is 5000. + +@item max_port +Set maximum local UDP port. Default value is 65000. + +@item buffer_size +Set the maximum socket buffer size in bytes. + +@item pkt_size +Set max send packet size (in bytes). Default value is 1472. +@end table + +@subsection Demuxer The following options are supported. @table @option @@ -1203,6 +1256,10 @@ Use UDP multicast as lower transport protocol. @item http Use HTTP tunneling as lower transport protocol, which is useful for passing proxies. + +@item https +Use HTTPs tunneling as lower transport protocol, which is useful for +passing proxies and widely used for security consideration. @end table Multiple lower transport protocols may be specified, in that case they are @@ -1220,6 +1277,9 @@ Accept packets only from negotiated peer address and port. Act as a server, listening for an incoming connection. @item prefer_tcp Try TCP for RTP transport first, if TCP is available as RTSP RTP transport. +@item satip_raw +Export raw MPEG-TS stream instead of demuxing. The flag will simply write out +the raw stream, with the original PAT/PMT/PIDs intact. @end table Default value is @samp{none}. @@ -1232,6 +1292,7 @@ The following flags are accepted: @item video @item audio @item data +@item subtitle @end table By default it accepts all media types. @@ -1256,6 +1317,9 @@ Set socket TCP I/O timeout in microseconds. @item user_agent Override User-Agent header. If not specified, it defaults to the libavformat identifier string. + +@item buffer_size +Set the maximum socket buffer size in bytes. @end table When receiving data over UDP, the demuxer tries to reorder received packets -- 2.25.1 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
[FFmpeg-devel] [PATCH 1/2] doc/examples/muxing: Remove unnecessary ret
From: Jun Zhao Remove unnecessary ret and make the code more compact Signed-off-by: Jun Zhao --- doc/examples/muxing.c | 5 + 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/doc/examples/muxing.c b/doc/examples/muxing.c index 3acb778322..cd997d5431 100644 --- a/doc/examples/muxing.c +++ b/doc/examples/muxing.c @@ -219,8 +219,6 @@ static AVFrame *alloc_audio_frame(enum AVSampleFormat sample_fmt, int sample_rate, int nb_samples) { AVFrame *frame = av_frame_alloc(); -int ret; - if (!frame) { fprintf(stderr, "Error allocating an audio frame\n"); exit(1); @@ -232,8 +230,7 @@ static AVFrame *alloc_audio_frame(enum AVSampleFormat sample_fmt, frame->nb_samples = nb_samples; if (nb_samples) { -ret = av_frame_get_buffer(frame, 0); -if (ret < 0) { +if (av_frame_get_buffer(frame, 0) < 0) { fprintf(stderr, "Error allocating an audio buffer\n"); exit(1); } -- 2.25.1 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-devel] FFmpeg 6.0
On Fri, Feb 24, 2023 at 09:31:33AM +0530, Gyan Doshi wrote: > > > On 2023-02-21 09:28 pm, Michael Niedermayer wrote: > > On Tue, Feb 21, 2023 at 02:20:26PM +0100, Jean-Baptiste Kempf wrote: > > > > > > On Tue, 21 Feb 2023, at 12:38, Michael Niedermayer wrote: > > > > On Fri, Feb 10, 2023 at 06:47:03PM +0100, Michael Niedermayer wrote: > > > > > Hi all > > > > > > > > > > i plan to branch off release/6.0 from master in the next days > > > > > If theres something blocking and i should wait, please reply here > > > > > > > > > > 6.0 release will be maybe 1 week after the branch point > > > > > once it has branched all important fixes should be backported of > > > > > course > > > > What name shall 6.0 bear ? > > > Why don’t you select 3 or 4 names from the list > > > And we do a proper vote > > how do you select 3 or 4 names while maintaining it as a > > proper vote where people can add name suggestions ? > > > > > > > on vote.ffmpeg.org? > > Is naming the only hold-up for this release? There are some bugfixes which need to be backported which is blocked by the videolan git server issue atm thx [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB If the United States is serious about tackling the national security threats related to an insecure 5G network, it needs to rethink the extent to which it values corporate profits and government espionage over security.-Bruce Schneier signature.asc Description: PGP signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
Re: [FFmpeg-devel] videolan ffmpeg git issue [please dont push before reading]
On Sat, Feb 25, 2023 at 02:14:44AM +0100, Michael Niedermayer wrote: > Hi all > > it seems videolan had a double RAID failure and was restored from backup > that caused ffmpeg developer git for master and release/5.1 to rewind by > a few commits. > It would be slightly messy if different commits are pushed on that rewond > HEADs- So if you push something (when its possible again) please make sure > that teh commits match the previous (pre restore) ones > > Its not the end of the world if others are pushed, we would then need > a merge commit to reconcile these 2 histories Some update videolan took this opertunity to switch the repositories from EOL gitosis to gitolite. This added some hickups first being some nonworkingness with multiple admins in gitolite. Which i belive was due to single keys being used with multiple users, i fixed that for a 2nd key i added so it seems thats the issue and a 2nd issue that causes branches & heads to disappear from public view when one pushed to them i suspect thats a permission / user/group issue but i cannot verify or fix this. ATM thresh is sleeping and i suspect these issues will be fixed once he awakes thx [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB If you fake or manipulate statistics in a paper in physics you will never get a job again. If you fake or manipulate statistics in a paper in medicin you will get a job for life at the pharma industry. signature.asc Description: PGP signature ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".