Re: [libav-devel] [PATCH 20/20] shorten: decode all channels in the same call to avcodec_decode_audio3().

On Thu, Oct 13, 2011 at 05:45:57PM -0400, Justin Ruggles wrote: > This allows for outputting the final frame. Otherwise the decoding will stop > after the first channel of the last frame when flushing the decoder with > empty packets. > > FATE reference updated accordingly. > --- > libavcodec/sho

Re: [libav-devel] [PATCH 19/20] shorten: enable CODEC_CAP_DELAY to flush any remaining frames in the buffer.

On Thu, Oct 13, 2011 at 05:45:56PM -0400, Justin Ruggles wrote: > Also need to stop decoding when the FN_QUIT command is found so that a > trailing seek table isn't decoded as a normal frame. > > Updated FATE reference. More valid audio is now decoded. > --- > libavcodec/shorten.c

Re: [libav-devel] [PATCH 18/20] shorten: do not use block size to determine whether to read the header.

On Thu, Oct 13, 2011 at 05:45:55PM -0400, Justin Ruggles wrote: > --- > libavcodec/shorten.c |6 -- > 1 files changed, 4 insertions(+), 2 deletions(-) > > diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c > index 7fdd86c..357512f 100644 > --- a/libavcodec/shorten.c > +++ b/libavcod

Re: [libav-devel] [PATCH 17/20] shorten: do not return after reading header

On Thu, Oct 13, 2011 at 05:45:54PM -0400, Justin Ruggles wrote: > --- > libavcodec/shorten.c |3 +-- > 1 files changed, 1 insertions(+), 2 deletions(-) > > diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c > index 8031758..7fdd86c 100644 > --- a/libavcodec/shorten.c > +++ b/libavcodec/

Re: [libav-devel] [PATCH 16/20] shorten: check output buffer size before decoding

On Thu, Oct 13, 2011 at 05:45:53PM -0400, Justin Ruggles wrote: > --- > libavcodec/shorten.c |8 +++- > 1 files changed, 7 insertions(+), 1 deletions(-) > > diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c > index ec50fc1..8031758 100644 > --- a/libavcodec/shorten.c > +++ b/libavc

Re: [libav-devel] [PATCH 15/20] shorten: use av_clip_int16() for output sample clipping

On Thu, Oct 13, 2011 at 05:45:52PM -0400, Justin Ruggles wrote: > --- > libavcodec/shorten.c |2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) > > diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c > index 5ffd634..ec50fc1 100644 > --- a/libavcodec/shorten.c > +++ b/libavcodec/s

Re: [libav-devel] [PATCH 14/20] shorten: use bytestream functions to decode the embedded WAVE header

On Thu, Oct 13, 2011 at 05:45:51PM -0400, Justin Ruggles wrote: > --- > libavcodec/shorten.c | 43 +-- > 1 files changed, 17 insertions(+), 26 deletions(-) looks OK ___ libav-devel mailing list libav-devel@libav

Re: [libav-devel] [PATCH 12/20] shorten: merge decoding of FN_DIFF* subblocks into decode_subframe_lpc()

On Thu, Oct 13, 2011 at 05:45:49PM -0400, Justin Ruggles wrote: > --- > libavcodec/shorten.c | 83 +++-- > 1 files changed, 39 insertions(+), 44 deletions(-) probably OK ___ libav-devel mailing list libav-de

Re: [libav-devel] [PATCH 13/20] shorten: add some comments

On Thu, Oct 13, 2011 at 05:45:50PM -0400, Justin Ruggles wrote: > --- > libavcodec/shorten.c | 18 ++ > 1 files changed, 18 insertions(+), 0 deletions(-) OK ___ libav-devel mailing list libav-devel@libav.org https://lists.libav.org/mai

Re: [libav-devel] [PATCH 11/20] cosmetics: remove some needless commented-out stuff

On Thu, Oct 13, 2011 at 05:45:48PM -0400, Justin Ruggles wrote: > --- > libavcodec/shorten.c |2 -- > 1 files changed, 0 insertions(+), 2 deletions(-) > > diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c > index 3f06e57..aeb07d7 100644 > --- a/libavcodec/shorten.c > +++ b/libavcodec/s

Re: [libav-devel] [PATCH 10/20] shorten: validate block size

On Thu, Oct 13, 2011 at 05:45:47PM -0400, Justin Ruggles wrote: > --- > libavcodec/shorten.c | 17 +++-- > 1 files changed, 15 insertions(+), 2 deletions(-) > > diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c > index 2502587..3f06e57 100644 > --- a/libavcodec/shorten.c > ++

Re: [libav-devel] [PATCH 09/20] shorten: move declaration of 'ret' to top of shorten_decode_frame()

On Thu, Oct 13, 2011 at 05:45:46PM -0400, Justin Ruggles wrote: > --- > libavcodec/shorten.c |4 ++-- > 1 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c > index 638d886..2502587 100644 > --- a/libavcodec/shorten.c > +++ b/libavcodec

Re: [libav-devel] [PATCH 08/20] shorten: pass on error value from allocate_buffers() instead of returning -1

On Thu, Oct 13, 2011 at 05:45:45PM -0400, Justin Ruggles wrote: > --- > libavcodec/shorten.c |6 +++--- > 1 files changed, 3 insertions(+), 3 deletions(-) > > diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c > index ad274b4..638d886 100644 > --- a/libavcodec/shorten.c > +++ b/libavcod

Re: [libav-devel] [PATCH 07/20] shorten: check for realloc failure

On Thu, Oct 13, 2011 at 05:45:44PM -0400, Justin Ruggles wrote: > --- > libavcodec/shorten.c | 20 +--- > 1 files changed, 17 insertions(+), 3 deletions(-) LGTM ___ libav-devel mailing list libav-devel@libav.org https://lists.libav.org

Re: [libav-devel] [PATCH 06/20] shorten: move decoding of prediction order and applying of global offset to decode_subframe_lpc().

On Thu, Oct 13, 2011 at 05:45:43PM -0400, Justin Ruggles wrote: > --- > libavcodec/shorten.c | 44 +++- > 1 files changed, 27 insertions(+), 17 deletions(-) looks OK ___ libav-devel mailing list libav-devel@liba

Re: [libav-devel] [PATCH 05/20] shorten: only calculate output size when returning decoded samples, otherwise just set data_size to zero.

On Thu, Oct 13, 2011 at 05:45:42PM -0400, Justin Ruggles wrote: > --- > libavcodec/shorten.c | 10 ++ > 1 files changed, 6 insertions(+), 4 deletions(-) LGTM ___ libav-devel mailing list libav-devel@libav.org https://lists.libav.org/mailman/li

Re: [libav-devel] [PATCH 04/20] cosmetics: reindent

On Thu, Oct 13, 2011 at 05:45:41PM -0400, Justin Ruggles wrote: > --- > libavcodec/shorten.c | 162 > +- > 1 files changed, 81 insertions(+), 81 deletions(-) obviously OK ___ libav-devel mailing list lib

Re: [libav-devel] [PATCH 03/20] shorten: separate processing of audio commands from non-audio commands

On Thu, Oct 13, 2011 at 05:45:40PM -0400, Justin Ruggles wrote: > --- > libavcodec/shorten.c | 73 ++--- > 1 files changed, 39 insertions(+), 34 deletions(-) looks reasonable ___ libav-devel mailing list lib

Re: [libav-devel] [PATCH 02/20] shorten: skip some fields in the WAV header embedded in the shorten header.

On Thu, Oct 13, 2011 at 05:45:39PM -0400, Justin Ruggles wrote: > fixes incorrect bitrate reporting and potential misreporting of the number of > channels. > --- > libavcodec/shorten.c |7 --- > 1 files changed, 4 insertions(+), 3 deletions(-) > > diff --git a/libavcodec/shorten.c b/libav

Re: [libav-devel] [PATCH 01/20] shorten: split reading of file header into a separate functions

On Thu, Oct 13, 2011 at 05:45:38PM -0400, Justin Ruggles wrote: > --- > libavcodec/shorten.c | 131 > +++--- > 1 files changed, 70 insertions(+), 61 deletions(-) LGTM ___ libav-devel mailing list libav-devel

Re: [libav-devel] [PATCH] prores: add '-pix_fmt yuv422p10le' to FATE tests to keep CRCs endian independent

Hi, On Thu, Oct 13, 2011 at 3:37 PM, Diego Biurrun wrote: > --- >  tests/fate/prores.mak |   10 +- >  1 files changed, 5 insertions(+), 5 deletions(-) > > diff --git a/tests/fate/prores.mak b/tests/fate/prores.mak > index 7be756d..1392f14 100644 > --- a/tests/fate/prores.mak > +++ b/tests

[libav-devel] [PATCH] prores: add '-pix_fmt yuv422p10le' to FATE tests to keep CRCs endian independent

--- tests/fate/prores.mak | 10 +- 1 files changed, 5 insertions(+), 5 deletions(-) diff --git a/tests/fate/prores.mak b/tests/fate/prores.mak index 7be756d..1392f14 100644 --- a/tests/fate/prores.mak +++ b/tests/fate/prores.mak @@ -7,9 +7,9 @@ FATE_PRORES = fate-prores-422

Re: [libav-devel] [libav-commits] prores: add FATE tests

Hi, On Thu, Oct 13, 2011 at 2:45 PM, Diego Biurrun wrote: > +fate-prores-422:       CMD = framecrc -vsync 0 -i > $(SAMPLES)/prores/Sequence_1-Apple_ProRes_422.mov > +fate-prores-422_hq:    CMD = framecrc -vsync 0 -i > $(SAMPLES)/prores/Sequence_1-Apple_ProRes_422_HQ.mov > +fate-prores-422_lt:  

[libav-devel] [PATCH 5/5] adpcmdec: calculate actual number of output samples for each decoder.

This also allows for removing some of the buf_size checks and using the sample count for some of the decoding loops. --- libavcodec/adpcm.c | 351 ++-- 1 files changed, 231 insertions(+), 120 deletions(-) diff --git a/libavcodec/adpcm.c b/libavcode

[libav-devel] [PATCH] mov: Prevent illegal writes when chapter titles are very short.

--- libavformat/mov.c | 23 +++ 1 files changed, 15 insertions(+), 8 deletions(-) diff --git a/libavformat/mov.c b/libavformat/mov.c index 0e2ad1f..06057a2 100644 --- a/libavformat/mov.c +++ b/libavformat/mov.c @@ -2369,14 +2369,21 @@ static void mov_read_chapters(AVFormatCo

[libav-devel] [PATCH 20/20] shorten: decode all channels in the same call to avcodec_decode_audio3().

This allows for outputting the final frame. Otherwise the decoding will stop after the first channel of the last frame when flushing the decoder with empty packets. FATE reference updated accordingly. --- libavcodec/shorten.c | 28 tests/ref/fate/los

[libav-devel] [PATCH 18/20] shorten: do not use block size to determine whether to read the header.

--- libavcodec/shorten.c |6 -- 1 files changed, 4 insertions(+), 2 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index 7fdd86c..357512f 100644 --- a/libavcodec/shorten.c +++ b/libavcodec/shorten.c @@ -102,6 +102,7 @@ typedef struct ShortenContext { int blocks

[libav-devel] [PATCH 19/20] shorten: enable CODEC_CAP_DELAY to flush any remaining frames in the buffer.

Also need to stop decoding when the FN_QUIT command is found so that a trailing seek table isn't decoded as a normal frame. Updated FATE reference. More valid audio is now decoded. --- libavcodec/shorten.c | 17 ++--- tests/ref/fate/lossless-shortenaudio |2 +- 2

[libav-devel] [PATCH 17/20] shorten: do not return after reading header

--- libavcodec/shorten.c |3 +-- 1 files changed, 1 insertions(+), 2 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index 8031758..7fdd86c 100644 --- a/libavcodec/shorten.c +++ b/libavcodec/shorten.c @@ -442,9 +442,8 @@ static int shorten_decode_frame(AVCodecContext *av

[libav-devel] [PATCH 16/20] shorten: check output buffer size before decoding

--- libavcodec/shorten.c |8 +++- 1 files changed, 7 insertions(+), 1 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index ec50fc1..8031758 100644 --- a/libavcodec/shorten.c +++ b/libavcodec/shorten.c @@ -550,9 +550,15 @@ static int shorten_decode_frame(AVCodecConte

[libav-devel] [PATCH 15/20] shorten: use av_clip_int16() for output sample clipping

--- libavcodec/shorten.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index 5ffd634..ec50fc1 100644 --- a/libavcodec/shorten.c +++ b/libavcodec/shorten.c @@ -254,7 +254,7 @@ static int16_t * interleave_buffer(int16_t *samples

[libav-devel] [PATCH 14/20] shorten: use bytestream functions to decode the embedded WAVE header

--- libavcodec/shorten.c | 43 +-- 1 files changed, 17 insertions(+), 26 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index a06eb3c..5ffd634 100644 --- a/libavcodec/shorten.c +++ b/libavcodec/shorten.c @@ -28,6 +28,7 @@ #includ

[libav-devel] [PATCH 13/20] shorten: add some comments

--- libavcodec/shorten.c | 18 ++ 1 files changed, 18 insertions(+), 0 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index f6f920c..a06eb3c 100644 --- a/libavcodec/shorten.c +++ b/libavcodec/shorten.c @@ -409,6 +409,7 @@ static int shorten_decode_frame(AV

[libav-devel] [PATCH 12/20] shorten: merge decoding of FN_DIFF* subblocks into decode_subframe_lpc()

--- libavcodec/shorten.c | 83 +++-- 1 files changed, 39 insertions(+), 44 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index aeb07d7..f6f920c 100644 --- a/libavcodec/shorten.c +++ b/libavcodec/shorten.c @@ -267,37 +267,54 @@

[libav-devel] [PATCH 11/20] cosmetics: remove some needless commented-out stuff

--- libavcodec/shorten.c |2 -- 1 files changed, 0 insertions(+), 2 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index 3f06e57..aeb07d7 100644 --- a/libavcodec/shorten.c +++ b/libavcodec/shorten.c @@ -409,7 +409,6 @@ static int shorten_decode_frame(AVCodecContext *avc

[libav-devel] [PATCH 10/20] shorten: validate block size

--- libavcodec/shorten.c | 17 +++-- 1 files changed, 15 insertions(+), 2 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index 2502587..3f06e57 100644 --- a/libavcodec/shorten.c +++ b/libavcodec/shorten.c @@ -330,8 +330,16 @@ static int read_header(ShortenCont

[libav-devel] [PATCH 09/20] shorten: move declaration of 'ret' to top of shorten_decode_frame()

--- libavcodec/shorten.c |4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index 638d886..2502587 100644 --- a/libavcodec/shorten.c +++ b/libavcodec/shorten.c @@ -382,6 +382,8 @@ static int shorten_decode_frame(AVCodecContext *a

[libav-devel] [PATCH 08/20] shorten: pass on error value from allocate_buffers() instead of returning -1

--- libavcodec/shorten.c |6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index ad274b4..638d886 100644 --- a/libavcodec/shorten.c +++ b/libavcodec/shorten.c @@ -306,7 +306,7 @@ static int decode_subframe_lpc(ShortenContext *

[libav-devel] [PATCH 06/20] shorten: move decoding of prediction order and applying of global offset to decode_subframe_lpc().

--- libavcodec/shorten.c | 44 +++- 1 files changed, 27 insertions(+), 17 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index a4313ae..52bf2b2 100644 --- a/libavcodec/shorten.c +++ b/libavcodec/shorten.c @@ -260,20 +260,41 @@ stati

[libav-devel] [PATCH 07/20] shorten: check for realloc failure

--- libavcodec/shorten.c | 20 +--- 1 files changed, 17 insertions(+), 3 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index 52bf2b2..ad274b4 100644 --- a/libavcodec/shorten.c +++ b/libavcodec/shorten.c @@ -116,6 +116,7 @@ static int allocate_buffers(Shor

[libav-devel] [PATCH 05/20] shorten: only calculate output size when returning decoded samples, otherwise just set data_size to zero.

--- libavcodec/shorten.c | 10 ++ 1 files changed, 6 insertions(+), 4 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index 62cead4..a4313ae 100644 --- a/libavcodec/shorten.c +++ b/libavcodec/shorten.c @@ -385,6 +385,7 @@ static int shorten_decode_frame(AVCodecCont

[libav-devel] [PATCH 04/20] cosmetics: reindent

--- libavcodec/shorten.c | 162 +- 1 files changed, 81 insertions(+), 81 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index 2a9b9ab..62cead4 100644 --- a/libavcodec/shorten.c +++ b/libavcodec/shorten.c @@ -427,91 +427,91 @@

[libav-devel] [PATCH 03/20] shorten: separate processing of audio commands from non-audio commands

--- libavcodec/shorten.c | 73 ++--- 1 files changed, 39 insertions(+), 34 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index af798da..2a9b9ab 100644 --- a/libavcodec/shorten.c +++ b/libavcodec/shorten.c @@ -69,6 +69,9 @@ #de

[libav-devel] [PATCH 02/20] shorten: skip some fields in the WAV header embedded in the shorten header.

fixes incorrect bitrate reporting and potential misreporting of the number of channels. --- libavcodec/shorten.c |7 --- 1 files changed, 4 insertions(+), 3 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index af12e08..af798da 100644 --- a/libavcodec/shorten.c +++ b

[libav-devel] [PATCH 01/20] shorten: split reading of file header into a separate functions

--- libavcodec/shorten.c | 131 +++--- 1 files changed, 70 insertions(+), 61 deletions(-) diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c index dd1bef8..af12e08 100644 --- a/libavcodec/shorten.c +++ b/libavcodec/shorten.c @@ -273,6 +273,73 @@

[libav-devel] [PATCH 00/20] shorten clean-up

Justin Ruggles (20): shorten: split reading of file header into a separate functions shorten: skip some fields in the WAV header embedded in the shorten header. shorten: separate processing of audio commands from non-audio commands cosmetics: reindent shorten: only calculate outpu

Re: [libav-devel] [PATCH] w32threads: support for frame multithreading

On Thu, Oct 13, 2011 at 11:27:00PM +0200, Janne Grunau wrote: > > --- /dev/null > +++ b/libavcodec/w32pthreads.h > @@ -0,0 +1,208 @@ > + > +#ifndef AVCODEC_W32PTHREAD_H > +#define AVCODEC_W32PTHREAD_H This is not quite the filename (missing 's'). > +static int pthread_create(pthread_t *thread, c

Re: [libav-devel] [PATCH] w32threads: support for frame multithreading

Hi, On Thu, Oct 13, 2011 at 11:27 PM, Janne Grunau wrote: > From: Steven Walters > > Replace our incomplete w32threads implementation with x264's pthreads > w32threads wrapper. > Relicensed to LGPL with kind permission by Pegasys Inc. > > Signed-off-by: Janne Grunau > --- >  libavcodec/Makefile

[libav-devel] [PATCH] w32threads: support for frame multithreading

From: Steven Walters Replace our incomplete w32threads implementation with x264's pthreads w32threads wrapper. Relicensed to LGPL with kind permission by Pegasys Inc. Signed-off-by: Janne Grunau --- libavcodec/Makefile |2 +- libavcodec/h264.c|2 +- libavcodec/h264_direct.

[libav-devel] Replace incomplete w32threads implementation

Hej, our w32threads thread helpers lack support for frame multithreading and nobody seems interested in implemeting it. Unfortunately win-pthreads causes trouble for VLC, JB can fill the details. It was suggested, that libav could use x264's w32threads pthreads wrapper. Pegasys Inc. and Steven gav

Re: [libav-devel] [PATCH 4/6] fmtconvert: check compile-time x86 instruction set flags

On 10/13/2011 04:38 PM, Ronald S. Bultje wrote: > Hi, > > On Wed, Oct 12, 2011 at 9:24 PM, Justin Ruggles > wrote: >> -if(mm_flags & AV_CPU_FLAG_SSE2){ >> +if (HAVE_SSE && mm_flags & AV_CPU_FLAG_SSE2) { > > (Not specific review of this patch, but ...) why don't we have a > HAVE_

Re: [libav-devel] [PATCH 2/6] fmtconvert: use mova instead of defining a new macro

On 10/13/2011 04:43 PM, Ronald S. Bultje wrote: > Hi, > > On Mon, Oct 10, 2011 at 6:48 AM, Justin Ruggles > wrote: >> @@ -214,12 +208,12 @@ cglobal float_interleave2_%1, 3,4,%2, dst, src, len, >> src1 >> %endmacro >> >> INIT_MMX >> -%define MOVPS movq >> %define PUNPCKLDQ punpckldq >> %

Re: [libav-devel] [PATCH 5/5] adpcmdec: calculate actual number of output samples for each decoder.

On Wed, 12 Oct 2011 19:04:12 -0400, Justin Ruggles wrote: > This also allows for removing some of the buf_size checks and using the > sample count for some of the decoding loops. > --- > libavcodec/adpcm.c | 351 > ++-- > 1 files changed, 231 in

Re: [libav-devel] [PATCH 2/6] fmtconvert: use mova instead of defining a new macro

Hi, On Mon, Oct 10, 2011 at 6:48 AM, Justin Ruggles wrote: > @@ -214,12 +208,12 @@ cglobal float_interleave2_%1, 3,4,%2, dst, src, len, > src1 >  %endmacro > >  INIT_MMX > -%define MOVPS     movq >  %define PUNPCKLDQ punpckldq >  %define PUNPCKHDQ punpckhdq >  FLOAT_INTERLEAVE2 mmx, 0 >  INIT_XM

Re: [libav-devel] [PATCH 6/6] fmtconvert: port float_to_int16_interleave() 2-channel x86 inline asm to yasm

Hi, On Wed, Oct 12, 2011 at 9:25 PM, Justin Ruggles wrote: > --- >  libavcodec/x86/fmtconvert.asm   |   52 + >  libavcodec/x86/fmtconvert_mmx.c |   69 +- >  2 files changed, 61 insertions(+), 60 deletions(-) I think all of Loren/Jas

Re: [libav-devel] [PATCH 4/6] fmtconvert: check compile-time x86 instruction set flags

Hi, On Wed, Oct 12, 2011 at 9:24 PM, Justin Ruggles wrote: > -        if(mm_flags & AV_CPU_FLAG_SSE2){ > +        if (HAVE_SSE && mm_flags & AV_CPU_FLAG_SSE2) { (Not specific review of this patch, but ...) why don't we have a HAVE_SSE2 macro? Ronald _

Re: [libav-devel] [PATCH 3/6] fmtconvert: port float_to_int16() x86 inline asm to yasm

Hi, 2011/10/10 Måns Rullgård : > Justin Ruggles writes: > >> --- >>  libavcodec/x86/fmtconvert.asm   |   42 ++ >>  libavcodec/x86/fmtconvert_mmx.c |   89 >> ++- >>  2 files changed, 55 insertions(+), 76 deletions(-) > > \o/ > > (This is not a p

Re: [libav-devel] [PATCH 5/6] fmtconvert: port int32_to_float_fmul_scalar() x86 inline asm to yasm

Hi, On Mon, Oct 10, 2011 at 6:48 AM, Justin Ruggles wrote: > --- >  libavcodec/x86/dsputil_yasm.asm |    8 - >  libavcodec/x86/fmtconvert.asm   |   46 ++ >  libavcodec/x86/fmtconvert_mmx.c |   59 -- >  libavutil/x86/x86util.asm  

Re: [libav-devel] [PATCH 3/5] avformat: add AVInputFormat flag AVFMT_NO_BYTE_SEEK.

On Thu, Oct 13, 2011 at 03:04:32PM -0400, Justin Ruggles wrote: > This will prevent seeking by bytes for formats that cannot handle this. > --- > doc/APIchanges |3 +++ > libavformat/avformat.h |3 ++- > libavformat/utils.c|2 ++ > libavformat/version.h |2 +- > 4 file

Re: [libav-devel] [PATCH 5/7] truespeech: simplify truespeech_read_frame() by using get_bits()

On Thu, Oct 13, 2011 at 03:08:08PM -0400, Justin Ruggles wrote: > --- > libavcodec/truespeech.c | 139 > ++- > 1 files changed, 52 insertions(+), 87 deletions(-) OK ___ libav-devel mailing list libav-devel@l

Re: [libav-devel] [PATCH 4/7] truespeech: decode directly to output buffer instead of a temp buffer

On Thu, Oct 13, 2011 at 03:08:07PM -0400, Justin Ruggles wrote: > --- > libavcodec/truespeech.c | 16 ++-- > 1 files changed, 6 insertions(+), 10 deletions(-) LGTM ___ libav-devel mailing list libav-devel@libav.org https://lists.libav.org/

Re: [libav-devel] [PATCH 2/7] truespeech: check for large enough output buffer rather than truncating output

On Thu, Oct 13, 2011 at 03:08:05PM -0400, Justin Ruggles wrote: > --- > libavcodec/truespeech.c | 16 > 1 files changed, 12 insertions(+), 4 deletions(-) looks reasonable ___ libav-devel mailing list libav-devel@libav.org https://list

Re: [libav-devel] [PATCH 3/7] truespeech: check to make sure channels == 1

On Thu, Oct 13, 2011 at 03:08:06PM -0400, Justin Ruggles wrote: > --- > libavcodec/truespeech.c |5 + > 1 files changed, 5 insertions(+), 0 deletions(-) > > diff --git a/libavcodec/truespeech.c b/libavcodec/truespeech.c > index 6f663f3..555ec87 100644 > --- a/libavcodec/truespeech.c > +++

Re: [libav-devel] [PATCH 1/7] truespeech: remove unneeded zero-size packet check.

On Thu, Oct 13, 2011 at 03:08:04PM -0400, Justin Ruggles wrote: > This is already checked in avcodec_decode_audio3() > --- > libavcodec/truespeech.c |3 --- > 1 files changed, 0 insertions(+), 3 deletions(-) > > diff --git a/libavcodec/truespeech.c b/libavcodec/truespeech.c > index 6b6d131..0

Re: [libav-devel] PPC warning patrol & PIC

On Thursday, October 13, 2011, Måns Rullgård wrote: >> Could/should I run a PPC valgrind instance instead, or some >> combination of this with what already exists? > > Valgrind would probably be more useful than the warning patrol. OK, I will go ahead and replace it soon. -- Sean McG. __

Re: [libav-devel] PPC warning patrol & PIC

"Sean McGovern" writes: > Hi folks, > > Is there any interest in me continuing to run these configurations? > Does anyone find them useful? The PIC one is certainly useful. We had a bug hiding there, as you probably recall. > Could/should I run a PPC valgrind instance instead, or some > combin

Re: [libav-devel] [PATCH 2/2] id3v2: reduce the scope of some non-globally-used symbols/structures

On 10/10/2011 11:27 AM, Diego Biurrun wrote: > --- > libavformat/id3v2.c | 22 ++ > libavformat/id3v2.h | 10 -- > 2 files changed, 14 insertions(+), 18 deletions(-) patch looks ok. -Justin ___ libav-devel mailing list

Re: [libav-devel] PPC warning patrol & PIC

Hi, On Thu, Oct 13, 2011 at 9:25 PM, Sean McGovern wrote: > Could/should I run a PPC valgrind instance instead, or some combination of > this with what already exists? PPC valgrind would be incredibly useful. Ronald ___ libav-devel mailing list libav

Re: [libav-devel] [PATCH 7/7] truespeech: use sizeof() instead of hardcoded sizes

On Thu, Oct 13, 2011 at 03:08:10PM -0400, Justin Ruggles wrote: > --- > libavcodec/truespeech.c |6 +++--- > 1 files changed, 3 insertions(+), 3 deletions(-) LGTM Diego ___ libav-devel mailing list libav-devel@libav.org https://lists.libav.org/mail

[libav-devel] PPC warning patrol & PIC

Hi folks, Is there any interest in me continuing to run these configurations? Does anyone find them useful? Could/should I run a PPC valgrind instance instead, or some combination of this with what already exists? -- Sean McG. ___ libav-devel mailin

Re: [libav-devel] [PATCH 1/2] id3v2: cosmetics: move some declarations before the places they are used

On 10/10/2011 11:27 AM, Diego Biurrun wrote: > --- > libavformat/id3v2.c | 130 +- > 1 files changed, 65 insertions(+), 65 deletions(-) fine. -Justin ___ libav-devel mailing list libav-devel@libav.org

Re: [libav-devel] [PATCH 6/7] truespeech: remove unneeded variable, 'consumed'

On Thu, Oct 13, 2011 at 03:08:09PM -0400, Justin Ruggles wrote: > increment the 'buf' pointer instead, and consume the whole packet. Looks sane. Diego ___ libav-devel mailing list libav-devel@libav.org https://lists.libav.org/mailman/listinfo/libav-deve

[libav-devel] [PATCH 7/7] truespeech: use sizeof() instead of hardcoded sizes

--- libavcodec/truespeech.c |6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/libavcodec/truespeech.c b/libavcodec/truespeech.c index 72d828a..7747fca 100644 --- a/libavcodec/truespeech.c +++ b/libavcodec/truespeech.c @@ -127,7 +127,7 @@ static void truespeech_correlate

[libav-devel] [PATCH 6/7] truespeech: remove unneeded variable, 'consumed'

increment the 'buf' pointer instead, and consume the whole packet. --- libavcodec/truespeech.c |7 +++ 1 files changed, 3 insertions(+), 4 deletions(-) diff --git a/libavcodec/truespeech.c b/libavcodec/truespeech.c index c48a9c1..72d828a 100644 --- a/libavcodec/truespeech.c +++ b/libavcod

[libav-devel] [PATCH 1/7] truespeech: remove unneeded zero-size packet check.

This is already checked in avcodec_decode_audio3() --- libavcodec/truespeech.c |3 --- 1 files changed, 0 insertions(+), 3 deletions(-) diff --git a/libavcodec/truespeech.c b/libavcodec/truespeech.c index 6b6d131..0ab2cab 100644 --- a/libavcodec/truespeech.c +++ b/libavcodec/truespeech.c @@ -

[libav-devel] [PATCH 5/7] truespeech: simplify truespeech_read_frame() by using get_bits()

--- libavcodec/truespeech.c | 139 ++- 1 files changed, 52 insertions(+), 87 deletions(-) diff --git a/libavcodec/truespeech.c b/libavcodec/truespeech.c index eec77ea..c48a9c1 100644 --- a/libavcodec/truespeech.c +++ b/libavcodec/truespeech.c @@ -21,6

[libav-devel] [PATCH 4/7] truespeech: decode directly to output buffer instead of a temp buffer

--- libavcodec/truespeech.c | 16 ++-- 1 files changed, 6 insertions(+), 10 deletions(-) diff --git a/libavcodec/truespeech.c b/libavcodec/truespeech.c index 555ec87..eec77ea 100644 --- a/libavcodec/truespeech.c +++ b/libavcodec/truespeech.c @@ -346,7 +346,6 @@ static int truespeech

[libav-devel] [PATCH 3/7] truespeech: check to make sure channels == 1

--- libavcodec/truespeech.c |5 + 1 files changed, 5 insertions(+), 0 deletions(-) diff --git a/libavcodec/truespeech.c b/libavcodec/truespeech.c index 6f663f3..555ec87 100644 --- a/libavcodec/truespeech.c +++ b/libavcodec/truespeech.c @@ -56,6 +56,11 @@ static av_cold int truespeech_deco

[libav-devel] [PATCH 2/7] truespeech: check for large enough output buffer rather than truncating output

--- libavcodec/truespeech.c | 16 1 files changed, 12 insertions(+), 4 deletions(-) diff --git a/libavcodec/truespeech.c b/libavcodec/truespeech.c index 0ab2cab..6f663f3 100644 --- a/libavcodec/truespeech.c +++ b/libavcodec/truespeech.c @@ -342,14 +342,22 @@ static int truespee

[libav-devel] [PATCH 0/7] truespeech clean-up

Justin Ruggles (7): truespeech: remove unneeded zero-size packet check. truespeech: check for large enough output buffer rather than truncating output truespeech: check to make sure channels == 1 truespeech: decode directly to output buffer instead of a temp buffer truespeech: simplif

[libav-devel] [PATCH 3/5] avformat: add AVInputFormat flag AVFMT_NO_BYTE_SEEK.

This will prevent seeking by bytes for formats that cannot handle this. --- doc/APIchanges |3 +++ libavformat/avformat.h |3 ++- libavformat/utils.c|2 ++ libavformat/version.h |2 +- 4 files changed, 8 insertions(+), 2 deletions(-) diff --git a/doc/APIchanges b/doc/

Re: [libav-devel] [PATCH 1/4] mlpdec: validate that the reported channel count matches the actual output channel count

On Thu, Oct 13, 2011 at 02:53:09PM -0400, Justin Ruggles wrote: > --- > libavcodec/mlpdec.c |7 ++- > 1 files changed, 6 insertions(+), 1 deletions(-) > > diff --git a/libavcodec/mlpdec.c b/libavcodec/mlpdec.c > index c90285a..15b8de4 100644 > --- a/libavcodec/mlpdec.c > +++ b/libavcodec/

[libav-devel] [PATCH 1/4] mlpdec: validate that the reported channel count matches the actual output channel count

--- libavcodec/mlpdec.c |7 ++- 1 files changed, 6 insertions(+), 1 deletions(-) diff --git a/libavcodec/mlpdec.c b/libavcodec/mlpdec.c index c90285a..15b8de4 100644 --- a/libavcodec/mlpdec.c +++ b/libavcodec/mlpdec.c @@ -909,7 +909,12 @@ static int output_data_internal(MLPDecodeContext *

Re: [libav-devel] [PATCH v2 4/5] rtsp: Remove the separate filter_source variable

On Thu, 13 Oct 2011 14:20:30 +0300, Martin Storsjö wrote: > Read it as a flag from the flags field instead. > --- > libavformat/rtsp.c |6 ++ > libavformat/rtsp.h |4 > 2 files changed, 2 insertions(+), 8 deletions(-) > > diff --git a/libavformat/rtsp.c b/libavformat/rtsp.c > i

Re: [libav-devel] [PATCH v2 2/5] rtsp: Simplify AVOption definitions

On Thu, 13 Oct 2011 14:20:28 +0300, Martin Storsjö wrote: > Use defines for shortening common parts, omit the .dbl named > initializer (since it's the first element in the union). > --- > libavformat/rtsp.c |4 +++- > 1 files changed, 3 insertions(+), 1 deletions(-) > Ok. -- Anton Khirno

Re: [libav-devel] [PATCH v2 1/5] rtsp: Merge the AVOption lists

On Thu, 13 Oct 2011 14:20:27 +0300, Martin Storsjö wrote: > This eases adding options that are common for both. The > AV_OPT_FLAG_EN/DECODING_PARAM still indicates whether they belong > to the muxer or demuxer. > --- > libavformat/rtsp.c|7 +++ > libavformat/rtsp.h|3 +++ > l

Re: [libav-devel] [PATCH v2 3/5] rtsp: Accept options via private avoptions instead of URL options

On 10/13/11 3:51 PM, Martin Storsjö wrote: On Thu, 13 Oct 2011, Luca Barbato wrote: On 10/13/11 2:02 PM, Martin Storsjö wrote: Additionally, that still has the issue of parsing options from URLs and redirects - if I give the input url rtsp+tcp://foo/bar, which redirects me to rtsp://other/serv

Re: [libav-devel] [PATCH] pcm: reduce pointer type casting

On Thu, Oct 13, 2011 at 03:09:51PM +0100, Mans Rullgard wrote: > Making 'samples' a pointer to uint8_t simplifies the DECODE > macro and reduces the amount of type casting overall. > > This also fixes some signed overflows on left shift. > > Signed-off-by: Mans Rullgard > --- > libavcodec/pcm.c

Re: [libav-devel] [PATCH] vc1: explicitly zero interlaced mode coding variables for progressive mode

Hi, On Thu, Oct 13, 2011 at 6:39 AM, shahriman AMS wrote: > - Original Message - >> From: Kostya Shishkov >> To: libav development >> Cc: >> Sent: Thursday, October 13, 2011 2:12 PM >> Subject: [libav-devel] [PATCH] vc1: explicitly zero interlaced mode coding >> variables for progressi

[libav-devel] [PATCH] pcm: reduce pointer type casting

Making 'samples' a pointer to uint8_t simplifies the DECODE macro and reduces the amount of type casting overall. This also fixes some signed overflows on left shift. Signed-off-by: Mans Rullgard --- libavcodec/pcm.c | 77 ++ 1 files changed

Re: [libav-devel] [PATCH 1/1] Always generate a proper WAVEFORMATEX structure in ff_put_wav_header

On 10/13/2011 11:34 AM, John Brooks wrote: > > The cbSize field should be included in all cases, even with PCM where > its value is ignored. > > Fixes encoding PCM audio in Matroska for some players which insist on > a full WAVEFORMATEX structure for A_MS/ACM audio. Looks ok. Is there regtests

Re: [libav-devel] [PATCH v2 3/5] rtsp: Accept options via private avoptions instead of URL options

On Thu, 13 Oct 2011, Luca Barbato wrote: On 10/13/11 2:02 PM, Martin Storsjö wrote: Additionally, that still has the issue of parsing options from URLs and redirects - if I give the input url rtsp+tcp://foo/bar, which redirects me to rtsp://other/server, I'll reparse that url, forgetting about

Re: [libav-devel] [PATCH] vc1: explicitly zero interlaced mode coding variables for progressive mode

- Original Message - > From: Kostya Shishkov > To: libav development > Cc: > Sent: Thursday, October 13, 2011 2:12 PM > Subject: [libav-devel] [PATCH] vc1: explicitly zero interlaced mode coding > variables for progressive mode > > Both v->fcm and v->field_mode are used in common code,

Re: [libav-devel] [PATCH v2 3/5] rtsp: Accept options via private avoptions instead of URL options

On 10/13/11 2:02 PM, Martin Storsjö wrote: Additionally, that still has the issue of parsing options from URLs and redirects - if I give the input url rtsp+tcp://foo/bar, which redirects me to rtsp://other/server, I'll reparse that url, forgetting about the options passed in the previous one. Sur

Re: [libav-devel] [PATCH v2 3/5] rtsp: Accept options via private avoptions instead of URL options

On Thu, 13 Oct 2011, Luca Barbato wrote: On 10/13/11 1:20 PM, Martin Storsjö wrote: +{ "rtsp_transport", "RTSP transport protocols", OFFSET(lower_transport_mask), AV_OPT_TYPE_FLAGS, {0}, 0, INT_MAX, DEC|ENC, "rtsp_transport" }, \ +{ "udp", "UDP", 0, AV_OPT_TYPE_CONST, {1<< RTSP_LOWER_

Re: [libav-devel] [PATCH v2 3/5] rtsp: Accept options via private avoptions instead of URL options

On 10/13/11 1:20 PM, Martin Storsjö wrote: +{ "rtsp_transport", "RTSP transport protocols", OFFSET(lower_transport_mask), AV_OPT_TYPE_FLAGS, {0}, 0, INT_MAX, DEC|ENC, "rtsp_transport" }, \ +{ "udp", "UDP", 0, AV_OPT_TYPE_CONST, {1<< RTSP_LOWER_TRANSPORT_UDP}, 0, 0, DEC|ENC, "rtsp_trans

[libav-devel] [PATCH v2 5/5] doc: Update the documentation on setting options for RTSP

--- doc/protocols.texi | 23 +++ 1 files changed, 15 insertions(+), 8 deletions(-) diff --git a/doc/protocols.texi b/doc/protocols.texi index cc35982..8aee5f6 100644 --- a/doc/protocols.texi +++ b/doc/protocols.texi @@ -246,12 +246,15 @@ supporting it (currently Darwin Strea

[libav-devel] [PATCH v2 3/5] rtsp: Accept options via private avoptions instead of URL options

Eventually, the old way of passing options by adding stuff to the URL can be dropped. This avoids having to tamper with the user-specified URL to pass options on the transport mode. This also works better with redirects, since the options don't need to be parsed out from the URL. --- libavformat/

[libav-devel] [PATCH v2 4/5] rtsp: Remove the separate filter_source variable

Read it as a flag from the flags field instead. --- libavformat/rtsp.c |6 ++ libavformat/rtsp.h |4 2 files changed, 2 insertions(+), 8 deletions(-) diff --git a/libavformat/rtsp.c b/libavformat/rtsp.c index 23c7b24..52a657b 100644 --- a/libavformat/rtsp.c +++ b/libavformat/rtsp

[libav-devel] [PATCH v2 2/5] rtsp: Simplify AVOption definitions

Use defines for shortening common parts, omit the .dbl named initializer (since it's the first element in the union). --- libavformat/rtsp.c |4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/libavformat/rtsp.c b/libavformat/rtsp.c index 49c148c..e4c4c98 100644 --- a/libav

  1   2   >