Re: [FFmpeg-devel] [PATCH 3/3] avcodec/adpcm: Remove setting min_channel to value it is already set to

2024-07-27 Thread Peter Ross
On Fri, Jul 26, 2024 at 11:08:32PM +0200, Michael Niedermayer wrote: > Signed-off-by: Michael Niedermayer > --- > libavcodec/adpcm.c | 4 > 1 file changed, 4 deletions(-) > > diff --git a/libavcodec/adpcm.c b/libavcodec/adpcm.c > index 8f9a8a8c8e2..6c53d0bcf39 100644 > --- a/libavcodec/adpc

Re: [FFmpeg-devel] [PATCH 6/6] lavu/cpu: deprecate AV_CPU_FLAG_RV{F, D}

2024-07-27 Thread Rémi Denis-Courmont
Le perjantaina 26. heinäkuuta 2024, 12.16.11 EEST Andreas Rheinhardt a écrit : > Rémi Denis-Courmont: > > --- > > > > doc/APIchanges | 3 +++ > > libavutil/cpu.h | 3 +++ > > libavutil/version.h | 1 + > > 3 files changed, 7 insertions(+) > > > > diff --git a/doc/APIchanges b/doc/APIcha

Re: [FFmpeg-devel] [PATCH 6/6] lavu/cpu: deprecate AV_CPU_FLAG_RV{F, D}

2024-07-27 Thread Rémi Denis-Courmont
Le lauantaina 27. heinäkuuta 2024, 15.22.27 EEST Rémi Denis-Courmont a écrit : > Le perjantaina 26. heinäkuuta 2024, 12.16.11 EEST Andreas Rheinhardt a écrit : > > Rémi Denis-Courmont: > > > --- > > > > > > doc/APIchanges | 3 +++ > > > libavutil/cpu.h | 3 +++ > > > libavutil/version.h

[FFmpeg-devel] [PATCH 1/1] lavc/ac3dsp: fix R-V CPU requirements

2024-07-27 Thread Rémi Denis-Courmont
It probably will not matter on any real hardware, but the Zbb optimisations do not require Zba. And then, we need HAVE_RVV to build the RVV stuff. --- libavcodec/riscv/ac3dsp_init.c | 36 -- 1 file changed, 21 insertions(+), 15 deletions(-) diff --git a/libavcodec/

[FFmpeg-devel] [PATCH 1/1] lavu/riscv: count bytes rather than words for bswap32

2024-07-27 Thread Rémi Denis-Courmont
This removes the dependency on Zba at essentially zero cost. --- libavcodec/riscv/bswapdsp_init.c | 8 libavcodec/riscv/bswapdsp_rvb.S | 3 ++- libavutil/riscv/bswap_rvb.S | 10 +- libswscale/riscv/rgb2rgb.c | 2 +- libswscale/riscv/rgb2rgb_rvb.S | 3 +-- 5 files

Re: [FFmpeg-devel] [PATCH v2 1/1] lavfi/vf_gopromax_opencl: add GoPor Max 360 video filter

2024-07-27 Thread TADANO Tokumei
On 2024/07/27 13:30, TADANO Tokumei wrote: Add an OpenCL filter for filtering GoPro Max native .360 files into standard equirectangular or youtube equiangular cubemap (eac) projection. The .360 file contains separated two video streams. This filter combine two streams into single stream with s

[FFmpeg-devel] [PATCH 1/2] lavc/vc1dsp: unify R-V V DC bypass functions

2024-07-27 Thread Rémi Denis-Courmont
--- libavcodec/riscv/vc1dsp_rvv.S | 126 ++ 1 file changed, 35 insertions(+), 91 deletions(-) diff --git a/libavcodec/riscv/vc1dsp_rvv.S b/libavcodec/riscv/vc1dsp_rvv.S index 5189d5e855..548ef9d3bf 100644 --- a/libavcodec/riscv/vc1dsp_rvv.S +++ b/libavcodec/riscv/v

[FFmpeg-devel] [PATCH 2/2] lavc/vc1dsp: use saturating arithmetic for RVV inv_trans_dc

2024-07-27 Thread Rémi Denis-Courmont
T-Head C908 (cycles): vc1dsp.vc1_inv_trans_4x4_dc_c: 113.7 vc1dsp.vc1_inv_trans_4x4_dc_rvv_i32: 46.5 (before) vc1dsp.vc1_inv_trans_4x4_dc_rvv_i32: 45.5 (after) vc1dsp.vc1_inv_trans_4x8_dc_c: 230.7 vc1dsp.vc1_inv_trans_4x8_dc_rvv_i32: 65.7 (before) vc1dsp.vc1_inv_trans_4x8_dc_rvv_i32: 52.5

[FFmpeg-devel] [PATCH 1/2] configure: check if assembler supports RV zicbop

2024-07-27 Thread Rémi Denis-Courmont
zicbop is the Cache Block Operation, Prefetch extension to RVI. --- configure | 4 1 file changed, 4 insertions(+) diff --git a/configure b/configure index f6f5c29fea..40774e1a88 100755 --- a/configure +++ b/configure @@ -2218,6 +2218,7 @@ ARCH_EXT_LIST_PPC=" ARCH_EXT_LIST_RISCV=" rv

[FFmpeg-devel] [PATCH 2/2] lavc/videodsp: RISC-V zicbop prefetch

2024-07-27 Thread Rémi Denis-Courmont
There are currently no ways to run-time detect the CPU capability, so we take it for granted (in the worst case, it will execute NOPs). --- libavcodec/riscv/Makefile| 2 ++ libavcodec/riscv/videodsp.S | 34 ++ libavcodec/riscv/videodsp_init.c | 36

Re: [FFmpeg-devel] [PATCH 3/3] avcodec/adpcm: Remove setting min_channel to value it is already set to

2024-07-27 Thread Michael Niedermayer
On Sat, Jul 27, 2024 at 09:17:01PM +1000, Peter Ross wrote: > On Fri, Jul 26, 2024 at 11:08:32PM +0200, Michael Niedermayer wrote: > > Signed-off-by: Michael Niedermayer > > --- > > libavcodec/adpcm.c | 4 > > 1 file changed, 4 deletions(-) > > > > diff --git a/libavcodec/adpcm.c b/libavcod

Re: [FFmpeg-devel] [PATCH 1/3] avformat/mov: Check sample_sizes before using it

2024-07-27 Thread Michael Niedermayer
Hi On Fri, Jul 26, 2024 at 07:24:38PM -0300, James Almer wrote: [...] > > Deja vu. Didn't you send something like this before? > > > > Also, can i get the sample? As with other issues, we shouldn't reach > > No, it was me: > https://ffmpeg.org//pipermail/ffmpeg-devel/2024-June/330391.html Iam s

Re: [FFmpeg-devel] [PATCH] swscale: [loongarch] Fix checkasm-sw_yuv2rgb failure.

2024-07-27 Thread Michael Niedermayer
On Thu, Jul 25, 2024 at 06:01:52PM +0800, 陈昊 wrote: > LGTM will apply and backport (if it applies cleanly) thx [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB Avoid a single point of failure, be that a person or equipment. signature.asc Description: PGP signa

Re: [FFmpeg-devel] [PATCH 1/3] avformat/mov: Check sample_sizes before using it

2024-07-27 Thread James Almer
On 7/27/2024 7:06 PM, Michael Niedermayer wrote: Hi On Fri, Jul 26, 2024 at 07:24:38PM -0300, James Almer wrote: [...] Deja vu. Didn't you send something like this before? Also, can i get the sample? As with other issues, we shouldn't reach No, it was me: https://ffmpeg.org//pipermail/ffmpeg

[FFmpeg-devel] [PATCH 01/11] avcodec/vvcdec: thread, ensure the parse stage gets the highest priority

2024-07-27 Thread Nuo Mi
The parser stage is not parallelizable. We need to schedule it as soon as possible to create later stages, which are more parallelizable clips | before | after | delta ||---|-- RitualDance_1920x1080_

[FFmpeg-devel] [PATCH 02/11] avcodec/vvcdec: refact, combine bs tab with tu tab

2024-07-27 Thread Nuo Mi
--- libavcodec/vvc/dec.c| 23 --- libavcodec/vvc/dec.h| 2 -- libavcodec/vvc/filter.c | 4 ++-- 3 files changed, 6 insertions(+), 23 deletions(-) diff --git a/libavcodec/vvc/dec.c b/libavcodec/vvc/dec.c index d04f68e4cf..ee009d4181 100644 --- a/libavcodec/vvc/dec.c +

[FFmpeg-devel] [PATCH 03/11] avcodec/vvcdec: remove unnecessary perframe initializations

2024-07-27 Thread Nuo Mi
deblock, sao, alf skip, imtf, ipm, cqt_depth, cb_pos_x, cb_pos_y, cb_height, cp_mv, tb_pos_x0, tb_pos_y0, tb_width, tb_height --- libavcodec/vvc/dec.c | 63 1 file changed, 46 insertions(+), 17 deletions(-) diff --git a/libavcodec/vvc/dec.c b/libavcode

[FFmpeg-devel] [PATCH 05/11] avcodec/vvcdec: refact out is_available from is_a0_available

2024-07-27 Thread Nuo Mi
--- libavcodec/vvc/mvs.c | 18 -- 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/libavcodec/vvc/mvs.c b/libavcodec/vvc/mvs.c index 1788a7150b..905edf1cd0 100644 --- a/libavcodec/vvc/mvs.c +++ b/libavcodec/vvc/mvs.c @@ -545,6 +545,16 @@ typedef struct NeighbourContex

[FFmpeg-devel] [PATCH 06/11] avcodec/vvcdec: do not zero frame mvf table

2024-07-27 Thread Nuo Mi
--- libavcodec/vvc/dec.c | 11 +++ libavcodec/vvc/mvs.c | 2 +- 2 files changed, 12 insertions(+), 1 deletion(-) diff --git a/libavcodec/vvc/dec.c b/libavcodec/vvc/dec.c index 568229d2c3..be23f2bd54 100644 --- a/libavcodec/vvc/dec.c +++ b/libavcodec/vvc/dec.c @@ -159,6 +159,16 @@ static

[FFmpeg-devel] [PATCH 04/11] avcodec/vvcdec: split ctu table to zero init and no zero init parts

2024-07-27 Thread Nuo Mi
cus need to init to zero, other parts are not --- libavcodec/vvc/ctu.c | 11 ++- libavcodec/vvc/ctu.h | 3 +-- libavcodec/vvc/dec.c | 21 +++-- libavcodec/vvc/dec.h | 5 +++-- libavcodec/vvc/inter.c | 3 +-- libavcodec/vvc/intra.c | 5 ++--- 6 files changed, 20

[FFmpeg-devel] [PATCH 08/11] avcodec/vvcdec: do not zero frame cpm table

2024-07-27 Thread Nuo Mi
--- libavcodec/vvc/ctu.c | 15 --- libavcodec/vvc/dec.c | 5 ++--- 2 files changed, 10 insertions(+), 10 deletions(-) diff --git a/libavcodec/vvc/ctu.c b/libavcodec/vvc/ctu.c index d39dd579ae..579337759f 100644 --- a/libavcodec/vvc/ctu.c +++ b/libavcodec/vvc/ctu.c @@ -1080,10 +1080,1

[FFmpeg-devel] [PATCH 09/11] avcodec/vvcdec: do not zero frame msf mmi table

2024-07-27 Thread Nuo Mi
--- libavcodec/vvc/ctu.c | 8 +--- libavcodec/vvc/dec.c | 4 ++-- libavcodec/vvc/mvs.c | 1 - 3 files changed, 7 insertions(+), 6 deletions(-) diff --git a/libavcodec/vvc/ctu.c b/libavcodec/vvc/ctu.c index 579337759f..06b57215a4 100644 --- a/libavcodec/vvc/ctu.c +++ b/libavcodec/vvc/ctu.c @@

[FFmpeg-devel] [PATCH 07/11] avcodec/vvcdec: check_available, use && instead of &= for shortcut evaluation

2024-07-27 Thread Nuo Mi
--- libavcodec/vvc/mvs.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/libavcodec/vvc/mvs.c b/libavcodec/vvc/mvs.c index fe047cfd5a..e3f18f1861 100644 --- a/libavcodec/vvc/mvs.c +++ b/libavcodec/vvc/mvs.c @@ -614,9 +614,9 @@ static int check_available(Neighbour *n, const

[FFmpeg-devel] [PATCH 11/11] avcodec/vvcdec: move frame tab memset from the main thread to worker threads

2024-07-27 Thread Nuo Mi
memset tables in the main thread can become a bottleneck for the decoder. For example, if it takes 1% of the processing time for one core, the maximum achievable FPS will be 100. Move the memeset to worker threads will fix the issue. --- libavcodec/vvc/dec.c| 13 - libavcodec/vvc/thread.

[FFmpeg-devel] [PATCH 10/11] avcodec/vvcdec: do not zero frame qp table

2024-07-27 Thread Nuo Mi
For luma, qp can only change at the CU level, so the qp tab size is related to the CU. For chroma, considering the joint CbCr, the QP tab size is related to the TU. --- libavcodec/vvc/dec.c | 7 ++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/libavcodec/vvc/dec.c b/libavcodec

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

2024-07-27 Thread Brad Smith
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 OpenBSD FreeBSD 12.0+, OpenBSD -current and what will be OpenBSD 7.6 support elf_aux_info(3). Signed-off-by: Brad Smi

Re: [FFmpeg-devel] [PATCH 1/2 v2] avformat/avisynth: remove atexit() handler

2024-07-27 Thread Stephen Hutchinson
On 7/19/24 12:56 PM, Stephen Hutchinson wrote: The atexit() handler in the avisynth demuxer was added because there was a conflict in AvxSynth that arose due to their use of C++ global objects, particularly in relation to having added a logging function relying on log4cpp. This conflict was resp