Re: [FFmpeg-devel] [PATCH] swscale/aarch64: Fix rgb24toyv12 only works with aligned width

On Wed, 18 Sep 2024, Zhao Zhili wrote: From: Zhao Zhili Since c0666d8b, rgb24toyv12 is broken for width non-aligned to 16. Add a simple wrapper to handle the non-aligned part. Signed-off-by: Zhao Zhili Co-authored-by: johzzy --- libswscale/aarch64/rgb2rgb.c | 23 ++- test

[FFmpeg-devel] [PATCH 5/5] checkasm: Print the SVE vector length at startup

--- tests/checkasm/checkasm.c | 12 +++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/tests/checkasm/checkasm.c b/tests/checkasm/checkasm.c index c932e028a5..c9d2b5faf1 100644 --- a/tests/checkasm/checkasm.c +++ b/tests/checkasm/checkasm.c @@ -94,6 +94,10 @@ #define isatt

[FFmpeg-devel] [PATCH 4/5] aarch64: Print the SVE vector length in libavutil/tests/cpu.c

This makes this aspect more visible in test logs. --- libavutil/aarch64/Makefile | 2 ++ libavutil/aarch64/cpu.h | 4 libavutil/aarch64/cpu_sve.S | 29 + libavutil/tests/cpu.c | 8 4 files changed, 43 insertions(+) create mode 100644 libavut

[FFmpeg-devel] [PATCH 3/5] aarch64: Add CPU feature flags for SVE and SVE2

Add code for detecting the feature on Linux and Windows. --- libavutil/aarch64/cpu.c | 20 libavutil/aarch64/cpu.h | 2 ++ libavutil/cpu.c | 2 ++ libavutil/cpu.h | 2 ++ libavutil/tests/cpu.c | 2 ++ tests/checkasm/checkasm.c | 2 ++ 6 files c

[FFmpeg-devel] [PATCH 2/5] configure: Add detection of assembler support for SVE/SVE2

It turns out that recent versions of MS armasm64 does support some SVE instructions, but not all of them. Test for one of the instructions that it currently doesn't support. --- Just as disclaimer, I'm not currently actively planning on writing SVE/SVE2 optimizations. However, related projects su

[FFmpeg-devel] [PATCH 1/5] aarch64: Detect I8MM on Windows via SVE-I8MM

There's no direct processor feature constant for I8MM alone, but there is a flag for SVE-I8MM (added in WinSDK 10.0.26100 and recent versions of mingw-w64). If SVE-I8MM is available, we can assume that I8MM is available. While HW supporting these features isn't yet commonly running Windows, this a

Re: [FFmpeg-devel] [PATCH v3 2/2] configure: correctly set sanitizer toolchain compilers

On Thu, 12 Sep 2024, Marvin Scholz wrote: Previously only the C compiler was set, which would lead to confusing situations where even though clang-asan was selected, it would still use g++ for C++ code, failing because configure does not support mixing compilers in this way (which is a separate

Re: [FFmpeg-devel] [PATCH v2 00/14] aarch64/vvc: Add SIMD

On Thu, 12 Sep 2024, Zhao Zhili wrote: From: Zhao Zhili Patches 1~9 has been updated according to Martin's review. Patches 10~14 are new. I have created a PR on github: https://github.com/quink-black/FFmpeg/pull/2 Thanks for testing it through that set of tests! No further comments from m

Re: [FFmpeg-devel] Procedure to enable the Windows on ARM64 FFMPEG Libraries

On Thu, 12 Sep 2024, Niranjan Kshatriya (QUIC) wrote: and compiled locally ( on ARM using WSL ) for Arm with tool chain : GitHub - Windows-on-ARM-Experiments/mingw-woarm64-build: Workflows and build scripts for Windows on Arm64 GNU cross-compiler for `aarch64-w64-mingw32` target.

Re: [FFmpeg-devel] [PATCH] tests/checkasm/sw_rgb: don't write random data past the end of the buffer

On Thu, 12 Sep 2024, Ramiro Polla wrote: On Thu, Sep 12, 2024 at 8:44 AM James Almer wrote: Should fix fate-checkasm-sw_rgb under gcc-ubsan. Signed-off-by: James Almer --- tests/checkasm/sw_rgb.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/checkasm/sw_rgb.c b/

Re: [FFmpeg-devel] [PATCH v2] configure: fix symbol prefix detection

On Wed, 11 Sep 2024, Marvin Scholz wrote: The symbol prefix check would incorrectly detect a bogus prefix in circumstances where sanitizers instrument the build, like when configuring with the clang-asan toolchain where it would detect the prefix as __odr_asan_gen_, which is obviously wrong.

Re: [FFmpeg-devel] [PATCH 3/3] aarch64/vvc: Add put_qpel_hv

On Wed, 11 Sep 2024, Zhao Zhili wrote: From: Zhao Zhili With Apple M1 (no i8mm): put_luma_hv_8_4x4_c: 2.2 ( 1.00x) put_luma_hv_8_4x4_neon: 0.8 ( 3.00x) put_luma_hv_8_8x8_c: 7.0 ( 1.00x) pu

Re: [FFmpeg-devel] [PATCH 2/3] aarch64/vvc: Add put_qpel_vx

On Wed, 11 Sep 2024, Zhao Zhili wrote: From: Zhao Zhili put_luma_v_8_4x4_c: 1.0 ( 1.00x) put_luma_v_8_4x4_neon: 0.0 ( 0.00x) put_luma_v_8_8x8_c: 3.5 ( 1.00x) put_luma_v_8_8x8_neon:

Re: [FFmpeg-devel] [PATCH 1/3] aarch64/h26x: Remove duplicate b.eq instruction

On Wed, 11 Sep 2024, Zhao Zhili wrote: From: Zhao Zhili b.eq is added by calc_all after each calc. --- libavcodec/aarch64/h26x/qpel_neon.S | 1 - 1 file changed, 1 deletion(-) diff --git a/libavcodec/aarch64/h26x/qpel_neon.S b/libavcodec/aarch64/h26x/qpel_neon.S index 8a372a76be..7868811b3b 1

Re: [FFmpeg-devel] [PATCH 6/6] avcodec/hevc: ff_hevc_(qpel/epel)_filters are signed type

On Sun, 8 Sep 2024, Zhao Zhili wrote: From: Zhao Zhili --- libavcodec/hevc/dsp_template.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) The rest of these patches seem fine (or where I don't have much of an opinion). // Martin ___ ffm

Re: [FFmpeg-devel] [PATCH 4/6] aarch64/vvc: Add put_pel/put_pel_uni/put_pel_uni_w

On Sun, 8 Sep 2024, Zhao Zhili wrote: diff --git a/libavcodec/aarch64/h26x/dsp.h b/libavcodec/aarch64/h26x/dsp.h index f72746ce03..076d01b477 100644 --- a/libavcodec/aarch64/h26x/dsp.h +++ b/libavcodec/aarch64/h26x/dsp.h @@ -248,4 +248,26 @@ NEON8_FNPROTO_PARTIAL_4(qpel, (int16_t *dst, const uin

Re: [FFmpeg-devel] aarch64: Implement support for elf_aux_info(3) on FreeBSD and OpenBSD

On Mon, 9 Sep 2024, Brad Smith wrote: aarch64: Implement support for elf_aux_info(3) on FreeBSD and OpenBSD FreeBSD 12.0+, OpenBSD -current and what will be OpenBSD 7.6 support elf_aux_info(3). Signed-off-by: Brad Smith --- configure | 2 ++ libavutil/aarch64/cpu.c | 2 +- libavut

Re: [FFmpeg-devel] [PATCH] avutil/cpu_internal: Provide ff_getauxval() wrapper for getauxvaul()

On Sat, 24 Aug 2024, Brad Smith wrote: avutil/cpu_internal: Provide ff_getauxval() wrapper for getauxvaul() Initially used for getauxval() but will be used to add support for other API. Signed-off-by: Brad Smith --- libavutil/aarch64/cpu.c | 4 ++-- libavutil/arm/cpu.c | 2 +- libavut

Re: [FFmpeg-devel] [PATCH] avcodec/avcodec: remove usage of __typeof__()

On Sun, 8 Sep 2024, James Almer wrote: It's non-standard C. Signed-off-by: James Almer --- libavcodec/avcodec.c | 14 +++--- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/libavcodec/avcodec.c b/libavcodec/avcodec.c index cb89236549..78153d12f1 100644 --- a/libavcodec/avc

Re: [FFmpeg-devel] [PATCH] avcodec/avcodec: remove usage of __typeof__()

On Sun, 8 Sep 2024, James Almer wrote: It's non-standard C. Signed-off-by: James Almer --- libavcodec/avcodec.c | 14 +++--- 1 file changed, 7 insertions(+), 7 deletions(-) LGTM // Martin ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.o

Re: [FFmpeg-devel] [PATCH] swscale/aarch64/rgb2rgb: add deinterleaveBytes neon implementation

On Fri, 30 Aug 2024, Ramiro Polla wrote: A55 A76 deinterleave_bytes_c: 70342.0 34497.5 deinterleave_bytes_neon: 21594.5 ( 3.26x) 5535.2 ( 6.23x) deinterleave_bytes_aligned_c: 71340.8 34651.2 deinter

Re: [FFmpeg-devel] [PATCH 2/2] aarch64/vvc: Bind h26x/sao filter implementation to vvc

On Wed, 28 Aug 2024, Zhao Zhili wrote: From: Zhao Zhili --- libavcodec/aarch64/h26x/dsp.h | 6 +++- libavcodec/aarch64/h26x/sao_neon.S| 44 +-- libavcodec/aarch64/hevcdsp_init_aarch64.c | 2 +- libavcodec/aarch64/vvc/Makefile | 5 +-- libavcodec

Re: [FFmpeg-devel] [PATCH 4/4] swscale/aarch64/rgb2rgb: add neon implementation for rgb24toyv12

On Thu, 29 Aug 2024, Ramiro Polla wrote: On Wed, Aug 28, 2024 at 11:23 PM Martin Storsjö wrote: On Wed, 28 Aug 2024, Ramiro Polla wrote: +2: +// load first line +ld3 {v16.8b, v17.8b, v18.8b}, [x0], #24 +ld3 {v19.8b, v20.8b, v21.8b}, [x0], #24

Re: [FFmpeg-devel] [PATCH 4/4] swscale/aarch64/rgb2rgb: add neon implementation for rgb24toyv12

On Wed, 28 Aug 2024, Ramiro Polla wrote: A55 A76 rgb24toyv12_16_200_c: 36658.8 17319.2 rgb24toyv12_16_200_neon: 12765.8 ( 2.87x) 6036.0 ( 2.87x) rgb24toyv12_128_60_c: 83329.5 39901.2 rgb24toyv12_128_60_neon: 28059.8 ( 2.9

Re: [FFmpeg-devel] [PATCH 1/4] checkasm/sw_rgb: add rgb24toyv12 tests

On Wed, 28 Aug 2024, Ramiro Polla wrote: NOTE: currently the tests for rgb24toyv12 fail for x86 since the c and mmxext implementations differ (the mmxext version averages four rgb pixels before performing the chroma calculations). --- tests/checkasm/sw_rgb.c | 89 ++

Re: [FFmpeg-devel] [PATCH] avutil/aarch64: add AV_COPY128 and AV_ZERO128 macros

On Thu, 22 Aug 2024, Ramiro Polla wrote: --- libavutil/aarch64/intreadwrite.h | 42 libavutil/intreadwrite.h | 4 ++- 2 files changed, 45 insertions(+), 1 deletion(-) create mode 100644 libavutil/aarch64/intreadwrite.h diff --git a/libavutil/aarch64/intre

Re: [FFmpeg-devel] [PATCH v2 3/7] avcodec/aarch64/mpegvideoencdsp: add neon implementations for pix_sum and pix_norm1

On Wed, 21 Aug 2024, Ramiro Polla wrote: BTW, this instruction is kinda exotic and the docs aren't super clear, so it'd be good to test manually that it really does what we want, for negative numbers and numbers close to the ends of the value range; I didn't do that manually yet. I prefer just

Re: [FFmpeg-devel] [PATCH v2 3/7] avcodec/aarch64/mpegvideoencdsp: add neon implementations for pix_sum and pix_norm1

On Wed, 21 Aug 2024, Ramiro Polla wrote: A55 A76 pix_norm1_c: 484.3 235.2 pix_norm1_neon: 193.8 ( 2.50x) 44.7 ( 5.26x) pix_sum_c: 302.8 243.7 pix_sum_neon: 81.6 ( 3.71x) 26.0 ( 9.37x) --- libavcodec/aarch64/Makefile

Re: [FFmpeg-devel] [PATCH] libswscale: arm: Don't assume aligned output in yuv2rgb functions

On Mon, 19 Aug 2024, Martin Storsjö wrote: This fixes failures in recently added checkasm tests. While the buffers in most cases are aligned, libswscale in general can't assume the output to be aligned. --- This fixes FATE failures like in http://fate.ffmpeg.org/report.cgi?time=202408190

Re: [FFmpeg-devel] Does rtspenc actually support AVFMT_GLOBALHEADER?

On Mon, 19 Aug 2024, John Cox wrote: Does rtspenc actually support AVFMT_GLOBALHEADER? It is specified in the FFOutputFormat flags but I can't see anywhere in the code where extradata is referenced like it is in other output formats which support that flag. I ask because I have an encoder that

Re: [FFmpeg-devel] [PATCH 2/7] avcodec/aarch64/mpegvideoencdsp: add neon implementations for pix_sum and pix_norm1

On Mon, 19 Aug 2024, Ramiro Polla wrote: If the stride is a negative number, the first sxtw does the right thing, but the "lsl w1, w1, #1" will zero out the upper half of the register. I'll start adding negative stride tests to checkasm to spot these bugs. That's probably useful. The other a

Re: [FFmpeg-devel] [PATCH 2/7] avcodec/aarch64/mpegvideoencdsp: add neon implementations for pix_sum and pix_norm1

On Sun, 18 Aug 2024, Ramiro Polla wrote: I had tested the real world case on the A76, but not on the A53. I spent a couple of hours with perf trying to find the source of the discrepancy but I couldn't find anything conclusive. I need to learn more about how to test cache misses. Nah, I guess

Re: [FFmpeg-devel] [PATCH 1/7] checkasm/mpegvideoencdsp: add pix_sum and pix_norm1

On Sun, 18 Aug 2024, Ramiro Polla wrote: --- tests/checkasm/Makefile | 1 + tests/checkasm/checkasm.c| 3 ++ tests/checkasm/checkasm.h| 1 + tests/checkasm/mpegvideoencdsp.c | 77 4 files changed, 82 insertions(+) create mode 100644 tests/

[FFmpeg-devel] [PATCH] libswscale: arm: Don't assume aligned output in yuv2rgb functions

This fixes failures in recently added checkasm tests. While the buffers in most cases are aligned, libswscale in general can't assume the output to be aligned. --- This fixes FATE failures like in http://fate.ffmpeg.org/report.cgi?time=20240819041749&slot=armv7-linux-gcc-9. --- libswscale/arm/yuv

Re: [FFmpeg-devel] [PATCH 3/7] avcodec/aarch64/mpegvideoencdsp: add dotprod implementation for pix_norm1

On Sun, 18 Aug 2024, Ramiro Polla wrote: A76 pix_norm1_c:231.5 pix_norm1_neon: 44.2 ( 5.24x) pix_norm1_dotprod: 20.7 (11.18x) --- libavcodec/aarch64/mpegvideoencdsp_init.c | 10 libavcodec/aarch64/mpegvideoencdsp_neon.S | 28 +++ 2 f

Re: [FFmpeg-devel] [PATCH 2/7] avcodec/aarch64/mpegvideoencdsp: add neon implementations for pix_sum and pix_norm1

On Sun, 18 Aug 2024, Ramiro Polla wrote: A53 A76 pix_norm1_c: 519.2 231.5 pix_norm1_neon: 195.0 ( 2.66x) 44.2 ( 5.24x) pix_sum_c: 344.5 242.2 pix_sum_neon:119.0 ( 2.89x) 41.7 ( 5.81x) --- Hmm, those speedups on the A53 look q

Re: [FFmpeg-devel] [TC invoked] [PATCH 2/4] lavc/mpegvideo: use H263DSP dequant function

On Sat, 6 Jul 2024, Rémi Denis-Courmont wrote: Le lauantaina 6. heinäkuuta 2024, 19.20.33 EEST Andreas Rheinhardt a écrit : Rémi Denis-Courmont: > Le lauantaina 6. heinäkuuta 2024, 18.23.00 EEST Andreas Rheinhardt a écrit : >> >> This adds an indirection. I have asked you to actually benchma

Re: [FFmpeg-devel] [PATCH] avcodec/aarch64/me_cmp: add dotprod implementations of sse16 and vsse_intra16

On Thu, 15 Aug 2024, Ramiro Polla wrote: checkasm --bench for Raspberry Pi 5 Model B Rev 1.0: sse_0_c: 241.5 sse_0_neon: 37.2 sse_0_dotprod: 22.2 vsse_4_c: 148.7 vsse_4_neon: 31.0 vsse_4_dotprod: 15.7 --- libavcodec/aarch64/me_cmp_init_aarch64.c | 14 +++ libavcodec/aarch64/me_cmp_neon.S

Re: [FFmpeg-devel] [PATCH 4/4] swscale/aarch64: add nv24/nv42 to yuv420p unscaled converter

On Thu, 15 Aug 2024, Ramiro Polla wrote: Thank you for the review. New patch attached. Thanks - this looks very straightforward and nice now! Just one minor nit below: +add x4, x4, x5, sxtw// src1 += srcPadding +add x9, x9, x5, sxtw

Re: [FFmpeg-devel] [PATCH] libavcodec/arm/mlpdsp_armv5te: fix label format to work with binutils 2.43

On Thu, 15 Aug 2024, Sebastian Ramacher wrote: On 2024-08-13 23:24:06 +0300, Martin Storsjö wrote: On Fri, 9 Aug 2024, Ross Burton wrote: > binutils 2.43 has stricter validation for labels[1] and results in errors > when building ffmpeg for armv5: > > src/libavcodec/arm/mlpdsp_a

Re: [FFmpeg-devel] [PATCH 4/4] swscale/aarch64: add nv24/nv42 to yuv420p unscaled converter

On Fri, 9 Aug 2024, Ramiro Polla wrote: checkasm --bench for Raspberry Pi 5 Model B Rev 1.0: nv24_yuv420p_128_c: 423.0 nv24_yuv420p_128_neon: 115.7 nv24_yuv420p_1920_c: 5939.5 nv24_yuv420p_1920_neon: 1339.7 nv42_yuv420p_128_c: 423.2 nv42_yuv420p_128_neon: 115.7 nv42_yuv420p_1920_c: 5907.5 nv42_y

Re: [FFmpeg-devel] [PATCH v2 5/5] swscale/aarch64/yuv2rgb: add neon yuv42{0, 2}p -> gbrp unscaled colorspace converters

On Tue, 6 Aug 2024, Ramiro Polla wrote: checkasm --bench on a Raspberry Pi 5 Model B Rev 1.0: yuv420p_gbrp_128_c: 1243.0 yuv420p_gbrp_128_neon: 453.5 yuv420p_gbrp_1920_c: 18165.5 yuv420p_gbrp_1920_neon: 6700.0 yuv422p_gbrp_128_c: 1463.5 yuv422p_gbrp_128_neon: 471.5 yuv422p_gbrp_1920_c: 21343.7 y

Re: [FFmpeg-devel] [PATCH] libavcodec/arm/mlpdsp_armv5te: fix label format to work with binutils 2.43

On Fri, 9 Aug 2024, Ross Burton wrote: binutils 2.43 has stricter validation for labels[1] and results in errors when building ffmpeg for armv5: src/libavcodec/arm/mlpdsp_armv5te.S:232: Error: junk at end of line, first unrecognized character is `0' Remove the leading zero in the "01" label t

Re: [FFmpeg-devel] [FFmpeg-cvslog] avfilter/video: don't zero allocated buffers if memory poisoning is used

On Tue, 13 Aug 2024, James Almer wrote: ffmpeg | branch: master | James Almer | Sat Aug 10 21:31:16 2024 -0300| [41307ff3e9384c51d646bff7e3dcf0d554098a8f] | committer: James Almer avfilter/video: don't zero allocated buffers if memory poisoning is used Same as in avcodec/get_buffer.c Should

Re: [FFmpeg-devel] [GASPP PATCH] Omit the "-c" argument from the preprocessing command

On Thu, 25 Jul 2024, Martin Storsjö wrote: A command like "cc -c -E" is tautological; the -c is ignored, when we explicitly specify that we want to preprocess only. Since https://github.com/llvm/llvm-project/commit/6461e537815f7fa68cef06842505353cf5600e9c and https://github.com

Re: [FFmpeg-devel] aarch64: Implement support for elf_aux_info(3) on FreeBSD and OpenBSD

On Sun, 28 Jul 2024, Rémi Denis-Courmont wrote: Le 28 juillet 2024 07:37:51 GMT+03:00, Brad Smith a écrit : On 2024-07-26 7:56 a.m., Rémi Denis-Courmont wrote: Le 26 juillet 2024 13:58:34 GMT+03:00, Brad Smith a écrit : aarch64: Implement support for elf_aux_info(3) on FreeBSD and OpenBS

[FFmpeg-devel] [GASPP PATCH] Omit the "-c" argument from the preprocessing command

A command like "cc -c -E" is tautological; the -c is ignored, when we explicitly specify that we want to preprocess only. Since https://github.com/llvm/llvm-project/commit/6461e537815f7fa68cef06842505353cf5600e9c and https://github.com/llvm/llvm-project/pull/98607, Clang now warns about the unused

Re: [FFmpeg-devel] [PATCH] checkasm: Increase the tolerance for ac3_sum_square_butterfly_float

On Tue, 23 Jul 2024, Michael Niedermayer wrote: On Wed, Jul 24, 2024 at 12:01:32AM +0300, Martin Storsjö via ffmpeg-devel wrote: Increase the tolerance from 10 ulp to 11 ulp. This fixes occasional errors for some inputs; the errors could be reproduced on aarch64/neon builds, with "che

[FFmpeg-devel] [PATCH] checkasm: Increase the tolerance for ac3_sum_square_butterfly_float

Increase the tolerance from 10 ulp to 11 ulp. This fixes occasional errors for some inputs; the errors could be reproduced on aarch64/neon builds, with "checkasm --test=ac3dsp 3446175925". --- tests/checkasm/ac3dsp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/checkas

Re: [FFmpeg-devel] Add Mediacodec audio decoders support

On Wed, 10 Jul 2024, Zhao Zhili wrote: On Jun 12, 2024, at 21:42, Matthieu Bouron wrote: Hello, This patchset adds Mediacodec audio decoders support. Currently, only AAC, AMR, MP3, FLAC, VORBIS and OPUS are supported. This is mainly useful to avoid shipping Android builds of FFmpeg that ar

[FFmpeg-devel] [PATCH 2/2] aarch64: vvc: Consistently use # for immediate constants

--- libavcodec/aarch64/vvc/alf.S | 12 ++-- 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/libavcodec/aarch64/vvc/alf.S b/libavcodec/aarch64/vvc/alf.S index 828031cb90..eec193302a 100644 --- a/libavcodec/aarch64/vvc/alf.S +++ b/libavcodec/aarch64/vvc/alf.S @@ -95,7 +95,7 @@

[FFmpeg-devel] [PATCH 1/2] aarch64: vvc: Fix compilation of alf.S with MSVC 2022 17.7 and older

Use the "ldur" instruction explicitly, instead of having the assembler implicitly convert "ldr" instructions to "ldur". This fixes build errors like these: libavcodec\aarch64\vvc\alf.o.asm(1023) : error A2518: operand 2: Memory offset must be aligned ldr q22, [x3, #24] libavc

Re: [FFmpeg-devel] [FFmpeg-cvslog] avcodec/vvc: Add aarch64 neon optimization for ALF

On Mon, 22 Jul 2024, Nuo Mi wrote: On Mon, Jul 22, 2024 at 9:15 PM Martin Storsjö wrote: On Mon, 22 Jul 2024, Zhao Zhili wrote: > ffmpeg | branch: master | Zhao Zhili | Tue Jul 16 00:19:15 2024 +0800| [2d4ef304c9e13f5e8abe37c20ddd0f17102c6393] | committer: Nuo

Re: [FFmpeg-devel] [FFmpeg-cvslog] avcodec/vvc: Add aarch64 neon optimization for ALF

On Mon, 22 Jul 2024, Zhao Zhili wrote: ffmpeg | branch: master | Zhao Zhili | Tue Jul 16 00:19:15 2024 +0800| [2d4ef304c9e13f5e8abe37c20ddd0f17102c6393] | committer: Nuo Mi avcodec/vvc: Add aarch64 neon optimization for ALF http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=2d4ef304c

Re: [FFmpeg-devel] Add Mediacodec audio decoders support

On Wed, 17 Jul 2024, Anton Khirnov wrote: Quoting Cosmin Stejerean via ffmpeg-devel (2024-07-16 22:14:19) > On Jul 16, 2024, at 8:24 PM, Rémi Denis-Courmont wrote: > > Le tiistaina 16. heinäkuuta 2024, 18.48.06 EEST Cosmin Stejerean via ffmpeg- > devel a écrit : >> To add another data poin

Re: [FFmpeg-devel] [PATCH] x86/intreadwrite: add missing casts to pointer arguments

On Thu, 11 Jul 2024, James Almer wrote: Should make strict compilers happy. Also, make AV_COPY128 use integer operations while at it. Signed-off-by: James Almer --- libavutil/x86/intreadwrite.h | 15 --- 1 file changed, 4 insertions(+), 11 deletions(-) diff --git a/libavutil/x86/in

Re: [FFmpeg-devel] [FFmpeg-cvslog] x86/intreadwrite: use intrinsics instead of inline asm for AV_COPY128

On Thu, 11 Jul 2024, James Almer wrote: On 7/11/2024 10:54 AM, Martin Storsjö wrote: On Thu, 11 Jul 2024, Martin Storsjö wrote: On Thu, 11 Jul 2024, James Almer wrote: On 7/11/2024 10:08 AM, Martin Storsjö wrote: On Wed, 10 Jul 2024, James Almer wrote: ffmpeg | branch: master | James

Re: [FFmpeg-devel] [FFmpeg-cvslog] x86/intreadwrite: use intrinsics instead of inline asm for AV_COPY128

On Thu, 11 Jul 2024, Martin Storsjö wrote: On Thu, 11 Jul 2024, James Almer wrote: On 7/11/2024 10:08 AM, Martin Storsjö wrote: On Wed, 10 Jul 2024, James Almer wrote: ffmpeg | branch: master | James Almer | Wed Jul 10 13:00:20 2024 -0300| [bd1bcb07e0f29c135103a402d71b343a09ad1690

Re: [FFmpeg-devel] [FFmpeg-cvslog] x86/intreadwrite: use intrinsics instead of inline asm for AV_COPY128

On Thu, 11 Jul 2024, James Almer wrote: On 7/11/2024 10:08 AM, Martin Storsjö wrote: On Wed, 10 Jul 2024, James Almer wrote: ffmpeg | branch: master | James Almer | Wed Jul 10 13:00:20 2024 -0300| [bd1bcb07e0f29c135103a402d71b343a09ad1690] | committer: James Almer x86/intreadwrite: use

Re: [FFmpeg-devel] [FFmpeg-cvslog] x86/intreadwrite: use intrinsics instead of inline asm for AV_COPY128

On Wed, 10 Jul 2024, James Almer wrote: ffmpeg | branch: master | James Almer | Wed Jul 10 13:00:20 2024 -0300| [bd1bcb07e0f29c135103a402d71b343a09ad1690] | committer: James Almer x86/intreadwrite: use intrinsics instead of inline asm for AV_COPY128 This has the benefit of removing any SSE -

Re: [FFmpeg-devel] [PATCH] tests/fate/filter-audio: convert atempo test to oneoff

On Fri, 5 Jul 2024, James Almer wrote: On 7/5/2024 2:38 AM, Anton Khirnov wrote: Quoting James Almer (2024-07-04 22:45:28) On 7/4/2024 4:04 PM, Anton Khirnov wrote: Filter output is not bitexact. --- Reference file at https://up.khirnov.net/7r.pcm, please put it in filter-reference/atempo.pcm

Re: [FFmpeg-devel] [PATCH v2 3/5] hlsenc: Remove bogus check for if (vs->start_pos) for appending segments

On Wed, 3 Jul 2024, Steven Liu wrote: Martin Storsjö 于2024年7月3日周三 04:46写道： On Wed, 26 Jun 2024, Martin Storsjö wrote: > Previously, vs->start_pos was never 0 here, unless using the > -hls_segment_size option, which wasn't allowed for SEGMENT_TYPE_FMP4. > Therefore, this

Re: [FFmpeg-devel] [PATCH] tests/fate/filter-audio: convert atempo test to oneoff

On Thu, 4 Jul 2024, Anton Khirnov wrote: Filter output is not bitexact. --- Reference file at https://up.khirnov.net/7r.pcm, please put it in filter-reference/atempo.pcm --- tests/fate/filter-audio.mak | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/fate/filter-audio.

Re: [FFmpeg-devel] [FFmpeg-cvslog] tests/fate/filter-audio.mak: add test for atempo audio filter

On Wed, 3 Jul 2024, Rajiv Harlalka wrote: ffmpeg | branch: master | Rajiv Harlalka | Thu Mar 21 13:53:29 2024 +0530| [fc446eea05b9bc7de9a3c6b56dae8255bb5c5b5d] | committer: Anton Khirnov tests/fate/filter-audio.mak: add test for atempo audio filter Signed-off-by: Rajiv Harlalka Signed-off-

[FFmpeg-devel] [PATCH] tests: Add a missing dependency for the filter-atempo test

--- tests/fate/filter-audio.mak | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/fate/filter-audio.mak b/tests/fate/filter-audio.mak index 8120620afe..cf6065b93f 100644 --- a/tests/fate/filter-audio.mak +++ b/tests/fate/filter-audio.mak @@ -413,6 +413,7 @@ fate-filter-hdcd-s32p: CMP = one

Re: [FFmpeg-devel] [GASPP PATCH 2/2] Handle local labels in expressions with .xword/.dword/.quad

On Mon, 24 Jun 2024, Martin Storsjö wrote: --- This might be needed in dav1d in the future. --- gas-preprocessor.pl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/gas-preprocessor.pl b/gas-preprocessor.pl index 20b927f..19b0131 100755 --- a/gas-preprocessor.pl +++ b/gas

Re: [FFmpeg-devel] [PATCH v2 3/5] hlsenc: Remove bogus check for if (vs->start_pos) for appending segments

On Wed, 26 Jun 2024, Martin Storsjö wrote: Previously, vs->start_pos was never 0 here, unless using the -hls_segment_size option, which wasn't allowed for SEGMENT_TYPE_FMP4. Therefore, this if statement was practically always taken anyway. Remove this bogus if statement, to allow cha

[FFmpeg-devel] [PATCH v2 5/5] hlsenc: Calculate the average and actual maximum bitrate of segments

Previously, the bitrate advertised in the master playlist would only be based on the nominal values in either AVCodecParameters bit_rate, or via AVCPBProperties max_bitrate. On top of this, a fudge factor of 10% is added, to account for container overhead. Neither of these bitrates may be known, a

[FFmpeg-devel] [PATCH v2 4/5] hlsenc: When not using HLS_SINGLE_FILE, set vs->size to range_length

This matches what is done in the corresponding case for HLS_SINGLE_FILE. Normally, vs->size is already initialized correctly - but when writing the initial segment, with mp4 files, vs->size has been set to the size of the init segment, while range_length contains the real size of the first segment

[FFmpeg-devel] [PATCH v2 3/5] hlsenc: Remove bogus check for if (vs->start_pos) for appending segments

Previously, vs->start_pos was never 0 here, unless using the -hls_segment_size option, which wasn't allowed for SEGMENT_TYPE_FMP4. Therefore, this if statement was practically always taken anyway. Remove this bogus if statement, to allow changing vs->start_pos to reflect the right value when not u

[FFmpeg-devel] [PATCH v2 2/5] hlsenc: Fix setting vs->start_pos when not using HLS_SINGLE_FILE or hls_segment_size

When not using HLS_SINGLE_FILE or hls_segment_size, we're writing each segment into a separate file. In that case, the file start pos for each segment will be zero. This matches the case in (hls->max_seg_size > 0) above, where we decide to switch to a new file. This fixes the calculation of "vs->

[FFmpeg-devel] [PATCH v2 1/5] hlsenc: Fix the return value accumulation in append_single_file

Both the read_byte variable (which is accumulated into append_single_file) and the return value are int64_t; give the ret variable the right corresponding type too. --- libavformat/hlsenc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libavformat/hlsenc.c b/libavformat/hlsen

Re: [FFmpeg-devel] [PATCH 2/4] hlsenc: Fix setting vs->start_pos when not using HLS_SINGLE_FILE or hls_segment_size

On Mon, 24 Jun 2024, Steven Liu wrote: On Jun 24, 2024, at 16:49, Martin Storsjö wrote: When not using HLS_SINGLE_FILE or hls_segment_size, we're writing each segment into a separate file. In that case, the file start pos for each segment will be zero. This matches the case in

[FFmpeg-devel] [PATCH] avcodec/vvcdec: Fix compiling with MSVC 2022 17.8 and older

Versions of MSVC older than 17.9 error out here with the following error: src/libavcodec/vvc/filter.c(815): error C2059: syntax error: '}' src/libavcodec/vvc/filter.c(832): error C2065: 'all_zero_bs': undeclared identifier src/libavcodec/vvc/filter.c(836): error C2065: 'all_zero_bs':

[FFmpeg-devel] [GASPP PATCH 2/2] Handle local labels in expressions with .xword/.dword/.quad

--- This might be needed in dav1d in the future. --- gas-preprocessor.pl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/gas-preprocessor.pl b/gas-preprocessor.pl index 20b927f..19b0131 100755 --- a/gas-preprocessor.pl +++ b/gas-preprocessor.pl @@ -958,7 +958,7 @@ sub handle_se

[FFmpeg-devel] [GASPP PATCH 1/2] Translate .xword and .dword to .quad

--- gas-preprocessor.pl | 2 ++ 1 file changed, 2 insertions(+) diff --git a/gas-preprocessor.pl b/gas-preprocessor.pl index b0c343e..20b927f 100755 --- a/gas-preprocessor.pl +++ b/gas-preprocessor.pl @@ -1169,6 +1169,8 @@ sub handle_serialized_line { $line =~ s/\.syntax/$comm$&/x if $a

Re: [FFmpeg-devel] [PATCH v4 3/3] swscale/aarch64: Add argb/abgr to yuv

On Mon, 24 Jun 2024, Zhao Zhili wrote: From: Zhao Zhili Test on Apple M1 with kperf: : -O3 : -O3 -fno-vectorize abgr_to_uv_8_c : 19.4 : 26.1 abgr_to_uv_8_neon : 29.9 : 51.1 abgr_to_uv_128_c

Re: [FFmpeg-devel] [PATCH v3 2/3] swscale/aarch64: Add bgra/rgba to yuv

On Mon, 24 Jun 2024, Zhao Zhili wrote: From: Zhao Zhili Test on Apple M1 with kperf : -O3 : -O3 -fno-vectorize bgra_to_uv_8_c : 13.4 : 27.5 bgra_to_uv_8_neon : 37.4 : 41.7 bgra_to_uv_128_c

Re: [FFmpeg-devel] [PATCH v3 1/3] swscale/aarch64: Add bgr24 to yuv

On Mon, 24 Jun 2024, Zhao Zhili wrote: From: Zhao Zhili Test on Apple M1 with kperf : -O3 : -O3 -fno-vectorize bgr24_to_uv_8_c : 28.5 : 52.5 bgr24_to_uv_8_neon : 54.5 : 59.7 bgr24_to_uv_128_c

[FFmpeg-devel] [PATCH 4/4] hlsenc: Calculate the average and actual maximum bitrate of segments

Previously, the bitrate advertised in the master playlist would only be based on the nominal values in either AVCodecParameters bit_rate, or via AVCPBProperties max_bitrate. On top of this, a fudge factor of 10% is added, to account for container overhead. Neither of these bitrates may be known, a

[FFmpeg-devel] [PATCH 3/4] hlsenc: When not using HLS_SINGLE_FILE, set vs->size to range_length

This matches what is done in the corresponding case for HLS_SINGLE_FILE. Normally, vs->size is already initialized correctly - but when writing the initial segment, with mp4 files, vs->size has been set to the size of the init segment, while range_length contains the real size of the first segment

[FFmpeg-devel] [PATCH 2/4] hlsenc: Fix setting vs->start_pos when not using HLS_SINGLE_FILE or hls_segment_size

When not using HLS_SINGLE_FILE or hls_segment_size, we're writing each segment into a separate file. In that case, the file start pos for each segment will be zero. This matches the case in (hls->max_seg_size > 0) above, where we decide to switch to a new file. This fixes the calculation of "vs->

[FFmpeg-devel] [PATCH 1/4] hlsenc: Fix the return value accumulation in append_single_file

Both the read_byte variable (which is accumulated into append_single_file) and the return value are int64_t; give the ret variable the right corresponding type too. --- libavformat/hlsenc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libavformat/hlsenc.c b/libavformat/hlsen

Re: [FFmpeg-devel] [PATCH v3] movenc: Add an option for resilient, hybrid fragmented/non-fragmented muxing

On Thu, 20 Jun 2024, Dennis Sädtler wrote: On 2024-06-20 15:47, Timo Rothenpieler wrote: On 20/06/2024 15:46, Martin Storsjö wrote: On Wed, 19 Jun 2024, Martin Storsjö wrote: This allows ending up with a normal, non-fragmented file when the file is finished, while keeping the file readable

Re: [FFmpeg-devel] [PATCH] aarch64: Add OpenBSD runtime detection of dotprod and i8mm using sysctl

On Sat, 22 Jun 2024, Brad Smith wrote: [PATCH] aarch64: Add OpenBSD runtime detection of dotprod and i8mm using sysctl Signed-off-by: Brad Smith --- libavutil/aarch64/cpu.c | 35 +++ 1 file changed, 35 insertions(+) diff --git a/libavutil/aarch64/cpu.c b/libavut

Re: [FFmpeg-devel] [PATCH v3] movenc: Add an option for resilient, hybrid fragmented/non-fragmented muxing

On Wed, 19 Jun 2024, Martin Storsjö wrote: This allows ending up with a normal, non-fragmented file when the file is finished, while keeping the file readable if writing is aborted abruptly at any point. (Normally when writing a mov/mp4 file, the unfinished file is completely useless unless it

Re: [FFmpeg-devel] [PATCH 2/2] swscale/aarch64: Add bgra/rgba to yuv

On Thu, 20 Jun 2024, Zhao Zhili wrote: On Jun 19, 2024, at 20:05, Rémi Denis-Courmont wrote: Le 19 juin 2024 11:24:28 GMT+02:00, Zhao Zhili mailto:quinkbl...@foxmail.com>> a écrit : On Jun 19, 2024, at 15:07, Rémi Denis-Courmont wrote: Le 15 juin 2024 11:57:18 GMT+02:00, Zhao Zhili a é

[FFmpeg-devel] [PATCH v3] movenc: Add an option for resilient, hybrid fragmented/non-fragmented muxing

This allows ending up with a normal, non-fragmented file when the file is finished, while keeping the file readable if writing is aborted abruptly at any point. (Normally when writing a mov/mp4 file, the unfinished file is completely useless unless it is finished properly.) This results in a file

Re: [FFmpeg-devel] [PATCH v2] movenc: Add an option for hiding fragments at the end

On Mon, 17 Jun 2024, Gyan Doshi via ffmpeg-devel wrote: On 2024-06-17 04:08 pm, Martin Storsjö wrote: On Sat, 15 Jun 2024, Gyan Doshi wrote: On 2024-06-15 03:54 am, Dennis Sädtler via ffmpeg-devel wrote: On 2024-06-14 13:23, Gyan Doshi wrote: On 2024-06-14 04:35 pm, Timo Rothenpieler

Re: [FFmpeg-devel] [PATCH] swscale/aarch64: Add argb/abgr to yuv

On Sun, 16 Jun 2024, Zhao Zhili wrote: From: Zhao Zhili Test on Apple M1 with kperf: abgr_to_uv_8_c: 19.4 abgr_to_uv_8_neon: 29.9 abgr_to_uv_128_c: 146.4 abgr_to_uv_128_neon: 85.1 abgr_to_uv_1080_c: 1162.6 abgr_to_uv_1080_neon: 819.6 abgr_to_uv_1920_c: 2063.6 abgr_to_uv_1920_neon: 1435.1 abgr

Re: [FFmpeg-devel] [PATCH 1/2] swscale/aarch64: Add bgr24 to yuv

On Sat, 15 Jun 2024, Zhao Zhili wrote: From: Zhao Zhili Test on Apple M1 with kperf bgr24_to_uv_8_c: 41.5 bgr24_to_uv_8_neon: 41.8 bgr24_to_uv_128_c: 133.5 bgr24_to_uv_128_neon: 94.3 bgr24_to_uv_1080_c: 960.5 bgr24_to_uv_1080_neon: 751.0 bgr24_to_uv_1920_c: 1695.3 bgr24_to_uv_1920_neon: 1357.

Re: [FFmpeg-devel] [PATCH 2/2] swscale/aarch64: Add bgra/rgba to yuv

On Sat, 15 Jun 2024, Zhao Zhili wrote: From: Zhao Zhili Test on Apple M1 with kperf bgra_to_uv_8_c: 13.4 bgra_to_uv_8_neon: 37.4 bgra_to_uv_128_c: 155.9 bgra_to_uv_128_neon: 91.7 bgra_to_uv_1080_c: 1173.2 bgra_to_uv_1080_neon: 822.7 bgra_to_uv_1920_c: 2078.2 bgra_to_uv_1920_neon: 1437.7 bgra_

Re: [FFmpeg-devel] [PATCH v2 2/2] avutil/macos_kperf: Fix assert which makes kperf failed to run

On Tue, 18 Jun 2024, Zhao Zhili wrote: From: Zhao Zhili On m1, kpc_get_counter_count(KPC_MASK) return 8 in my test. The exact value doesn't matter in our case, as long as we have a sufficiently large array Signed-off-by: Zhao Zhili --- libavutil/macos_kperf.c | 16 ++-- 1 file cha

Re: [FFmpeg-devel] [PATCH 1/2] avutil/timer: define macos kperf as AV_READ_TIME

On Mon, 17 Jun 2024, Zhao Zhili wrote: On Jun 17, 2024, at 19:15, Martin Storsjö wrote: On Wed, 12 Jun 2024, Zhao Zhili wrote: From: Zhao Zhili Firstly, make ff_kperf_cycles as an implementation of AV_READ_TIME avoids code duplication. Secondly, fix compilation error since 6a18c0bc87e

Re: [FFmpeg-devel] [PATCH 1/2] avutil/timer: define macos kperf as AV_READ_TIME

On Mon, 17 Jun 2024, Martin Storsjö wrote: On Wed, 12 Jun 2024, Zhao Zhili wrote: From: Zhao Zhili Firstly, make ff_kperf_cycles as an implementation of AV_READ_TIME avoids code duplication. Secondly, fix compilation error since 6a18c0bc87e when macos-kperf is enabled. mach_time.h is

Re: [FFmpeg-devel] [PATCH 1/2] avutil/timer: define macos kperf as AV_READ_TIME

On Wed, 12 Jun 2024, Zhao Zhili wrote: From: Zhao Zhili Firstly, make ff_kperf_cycles as an implementation of AV_READ_TIME avoids code duplication. Secondly, fix compilation error since 6a18c0bc87e when macos-kperf is enabled. mach_time.h is included only when CONFIG_MACOS_KPERF is 0. The err

Re: [FFmpeg-devel] [PATCH 2/2] avutil/macos_kperf: Fix assert which makes kperf failed to run

On Wed, 12 Jun 2024, Zhao Zhili wrote: From: Zhao Zhili On m1, kpc_get_counter_count(KPC_MASK) return 8. The exact value doesn't matter in our case. This is somewhat unexpected, I had expected that this API was originally tested on an m1. I guess it might depend on what OS version you're us

Re: [FFmpeg-devel] [PATCH v2] movenc: Add an option for hiding fragments at the end

On Sat, 15 Jun 2024, Gyan Doshi wrote: On 2024-06-15 03:54 am, Dennis Sädtler via ffmpeg-devel wrote: On 2024-06-14 13:23, Gyan Doshi wrote: On 2024-06-14 04:35 pm, Timo Rothenpieler wrote: On 14/06/2024 12:44, Martin Storsjö wrote: On Fri, 14 Jun 2024, Gyan Doshi wrote: On 2024-06-14 02

[FFmpeg-devel] [PATCH v2] aarch64: Use cntvct_el0 as timer register on Android and macOS

The default timer register pmccntr_el0 usually requires enabling access with e.g. a kernel module. On macOS, using cntvct_el0 gives measurements with the same magnitude as mach_absolute_time (which is used currently), but possibly with a little less overhead/noise. --- cntvct_el0 should have less

Re: [FFmpeg-devel] [PATCH v2] movenc: Add an option for hiding fragments at the end

On Fri, 14 Jun 2024, Timo Rothenpieler wrote: On 14/06/2024 12:44, Martin Storsjö wrote: On Fri, 14 Jun 2024, Gyan Doshi wrote: On 2024-06-14 02:18 am, Martin Storsjö wrote: On Thu, 13 Jun 2024, Gyan Doshi wrote: On 2024-06-13 06:20 pm, Martin Storsjö wrote: I'd otherwise want to

  1   2   3   4   5   6   7   8   9   10   >