[FFmpeg-devel] [PATCH v4] mdct15: add assembly optimizations for the 15-point FFT

c:1802 decicycles in fft15,16774635 runs, 2581 skips avx: 865 decicycles in fft15,16776378 runs,838 skips Signed-off-by: Rostislav Pehlivanov --- libavcodec/mdct15.c | 182 +-- libavcodec/mdct15.h | 26

[FFmpeg-devel] [PATCH] configure: use pkg-config for libgme, if available

The pkg-config file is relatively new (2013), so some distros might not have it yet. And the -lstdc++ being required for the static lib is only present since the last release in December 2016. --- configure | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/configure

Re: [FFmpeg-devel] [PATCH v3] mdct15: add assembly optimizations for the 15-point FFT

On Fri, Jun 23, 2017 at 12:44 AM, Rostislav Pehlivanov wrote: > +%macro FFT5 3 ; %1 - in_offset, %2 - dst1 (64bit used), %3 - dst2 > +movddup xm0, [inq + 0*16 + 0 + %1] ; in[ 0].re, in[ 0].im, in[ 0].re, > in[ 0].im > +movsd xm1, [inq + 1*16 + 8 + %1] ; in[

[FFmpeg-devel] Fwd: [PATCH] configure: use or require pkg-config for libvorbis

I wasn't sure if adding use_pkg_config or completely changing to pkg-config-only would be more appropriate so I submit two patches doing one or the other. Cleanest would be changing to require pkg-config, but I leave the decision to the committer, if either are accepted.

[FFmpeg-devel] [PATCH] configure: require pkg-config for libgme

--- configure | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/configure b/configure index dd9608540e..08bd6c0ae8 100755 --- a/configure +++ b/configure @@ -5826,7 +5826,7 @@ enabled fontconfig&& enable libfontconfig enabled libfontconfig && require_pkg_config

Re: [FFmpeg-devel] [PATCH V5 1/4] lavc/golomb: add value range comment for set_ue_golomb().

On 2017/6/19 8:52, Jun Zhao wrote: > On 2017/6/14 11:00, Jun Zhao wrote: >> V5: Split the patch. >> V4: Fix rang check error in assert base on Mark's review >> V3: Clean the code logic base on Michael's review. >> V2: Add set_ue_golomb_long() to support 32bits UE golomb and update the unit >>

[FFmpeg-devel] [PATCH v3] mdct15: add assembly optimizations for the 15-point FFT

c:1802 decicycles in fft15,16774635 runs, 2581 skips fma3: 935 decicycles in fft15,16775893 runs, 1323 skips Signed-off-by: Rostislav Pehlivanov --- libavcodec/mdct15.c | 182 +-- libavcodec/mdct15.h | 26

Re: [FFmpeg-devel] [PATCH v2] mdct15: add assembly optimizations for the 15-point FFT

On 6/22/2017 5:40 PM, Rostislav Pehlivanov wrote: > c:1802 decicycles in fft15,16774635 runs, 2581 skips > fma3: 935 decicycles in fft15,16775893 runs, 1323 skips > --- > libavcodec/mdct15.c | 186 > ++- > libavcodec/mdct15.h |

[FFmpeg-devel] [PATCHv3 3/4] avformat: make flush_packets a tri-state and set it to -1 (auto) by default

If flushing is not disabled, then mux.c will signal the end of the packets with an AVIO_DATA_MARKER_FLUSH_POINT, and aviobuf will be able to decide to flush or not based on the preferred minimum packet size set by the used protocol. Signed-off-by: Marton Balint ---

Re: [FFmpeg-devel] [PATCH] vf_spp: only assign function pointers if permutation matches expectations.

On Wed, Jun 21, 2017 at 09:03:11PM -0400, Ronald S. Bultje wrote: > Hi, > > On Wed, Jun 21, 2017 at 8:42 PM, Michael Niedermayer > wrote: > > > On Wed, Jun 21, 2017 at 09:46:58AM -0400, Ronald S. Bultje wrote: > > > --- > > > libavfilter/x86/vf_spp.c | 15

[FFmpeg-devel] [PATCH v2] mdct15: add assembly optimizations for the 15-point FFT

c:1802 decicycles in fft15,16774635 runs, 2581 skips fma3: 935 decicycles in fft15,16775893 runs, 1323 skips --- libavcodec/mdct15.c | 186 ++- libavcodec/mdct15.h | 26 +++--- libavcodec/x86/Makefile | 2 +

Re: [FFmpeg-devel] [PATCH] avformat/oggenc: check for stream private data in ogg_free()

On 6/22/2017 4:06 PM, Paul B Mahol wrote: > On 6/22/17, James Almer wrote: >> Fixes a NULL pointer derefence when ogg_init() returns a failure and >> a stream's private data was not yet allocated. >> >> This is a regression since 3c5a53cdfa099bba8bd951f95b85727b4b3b5d68 >> >>

Re: [FFmpeg-devel] [PATCH] avformat/oggenc: check for stream private data in ogg_free()

On 6/22/17, James Almer wrote: > Fixes a NULL pointer derefence when ogg_init() returns a failure and > a stream's private data was not yet allocated. > > This is a regression since 3c5a53cdfa099bba8bd951f95b85727b4b3b5d68 > > Signed-off-by: James Almer >

[FFmpeg-devel] [PATCH] avformat/oggenc: check for stream private data in ogg_free()

Fixes a NULL pointer derefence when ogg_init() returns a failure and a stream's private data was not yet allocated. This is a regression since 3c5a53cdfa099bba8bd951f95b85727b4b3b5d68 Signed-off-by: James Almer --- libavformat/oggenc.c | 2 ++ 1 file changed, 2 insertions(+)

Re: [FFmpeg-devel] [PATCH 1/2] checkasm: add sbrdsp tests

On 6/22/2017 9:56 AM, Matthieu Bouron wrote: > --- > > The following patchset applies on top of Clément's aacpsdsp patchset. > > --- > tests/checkasm/Makefile | 3 +- > tests/checkasm/checkasm.c | 1 + > tests/checkasm/checkasm.h | 1 + > tests/checkasm/sbrdsp.c | 297 >

[FFmpeg-devel] [PATCH 2/2] lavc/aarch64: add sbrdsp neon implementation

tests/checkasm/checkasm --bench --test=sbrdsp (git)-[aarch64a checkasm: using

[FFmpeg-devel] [PATCH 1/2] checkasm: add sbrdsp tests

--- The following patchset applies on top of Clément's aacpsdsp patchset. --- tests/checkasm/Makefile | 3 +- tests/checkasm/checkasm.c | 1 + tests/checkasm/checkasm.h | 1 + tests/checkasm/sbrdsp.c | 297 ++ 4 files changed, 301

[FFmpeg-devel] [PATCH 5/5] lavc/aacpsdsp: use ptrdiff_t for stride in hybrid_analysis

From: Clément Bœsch --- libavcodec/aacpsdsp.h | 4 +++- libavcodec/aacpsdsp_template.c | 2 +- libavcodec/aarch64/aacpsdsp_init_aarch64.c | 2 +- libavcodec/aarch64/aacpsdsp_neon.S | 1 - libavcodec/arm/aacpsdsp_init_arm.c | 2

[FFmpeg-devel] [PATCH 4/5] lavc/aarch64: add a few SIMD functions for AAC PS

From: Clément Bœsch ☭ tests/checkasm/checkasm --bench --test=aacpsdsp checkasm: using random seed 3318985180 MMX implied by specified flags MMX implied by specified flags NEON: - aacpsdsp.add_squares[OK] - aacpsdsp.mul_pair_single[OK] - aacpsdsp.hybrid_analysis

[FFmpeg-devel] [PATCH 2/5] checkasm: add AAC PS tests

From: Clément Bœsch This includes various fixes and improvements from James Almer. Signed-off-by: James Almer --- tests/checkasm/Makefile | 1 + tests/checkasm/aacpsdsp.c | 161 ++ tests/checkasm/checkasm.c

[FFmpeg-devel] [PATCH 1/5] lavc/arm: fix lack of precision in ff_ps_stereo_interpolate_neon

From: Clément Bœsch The code originally pre-multiply by 2 the steps, causing the running sum of the h factors to drift away due to the lack of precision. It quickly causes an inaccuracy > 0.01. I tried diverse approaches such as multiply by 2.0 (instead of adding the value

[FFmpeg-devel] [PATCH 3/5] lavc/aacpsdsp: fix a few spaces (cosmetics)

From: Clément Bœsch --- libavcodec/aacpsdsp_template.c | 9 + 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/libavcodec/aacpsdsp_template.c b/libavcodec/aacpsdsp_template.c index 3049ce8b79..4f193604bc 100644 --- a/libavcodec/aacpsdsp_template.c +++

Re: [FFmpeg-devel] [PATCH] checkasm/aarch64: fix tests returning a float

On Wed, Jun 21, 2017 at 10:19:33PM +0200, Matthieu Bouron wrote: > On Wed, Jun 21, 2017 at 04:57:53PM -0300, James Almer wrote: > > On 6/19/2017 6:08 AM, Matthieu Bouron wrote: > > > Avoids overriding v0 (which containins the result of the tested > > > function) in checkasm_call_checked. > > > >

Re: [FFmpeg-devel] [PATCH] Fix static linking openssl library

(This really belongs to ffmpeg-user.) On Thu, Jun 22, 2017 at 08:14:56 +0300, Alexandr Topilski wrote: > use_pkg_config openssl openssl/ssl.h SSL_library_init > check_pkg_config openssl openssl/ssl.h SSL_library_init > pkg-config --exists --print-errors openssl > check_func_headers openssl/ssl.h