Re: [libav-devel] [PATCH] configure: add libavresample to rpath

On 25/04/12 22:11, Sean McGovern wrote: > --- > configure |2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) > > diff --git a/configure b/configure > index 3632bda..1dc2e3e 100755 > --- a/configure > +++ b/configure > @@ -3084,7 +3084,7 @@ enabled extra_warnings && check_cflags -Winlin

[libav-devel] [PATCH] configure: add libavresample to rpath

--- configure |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/configure b/configure index 3632bda..1dc2e3e 100755 --- a/configure +++ b/configure @@ -3084,7 +3084,7 @@ enabled extra_warnings && check_cflags -Winline # add some linker flags check_ldflags -Wl,--warn-com

Re: [libav-devel] [PATCH 3/3] lavr: add x86-optimized function for mixing 2 to 1 s16p with q6 coeffs

On Wed, Apr 25, 2012 at 5:46 PM, Jason Garrett-Glaser wrote: > On Wed, Apr 25, 2012 at 4:44 PM, Justin Ruggles > wrote: >> --- >>  libavresample/x86/audio_mix.asm    |   43 >> >>  libavresample/x86/audio_mix_init.c |    5 >>  2 files changed, 48 insertio

Re: [libav-devel] [PATCH 3/3] lavr: add x86-optimized function for mixing 2 to 1 s16p with q6 coeffs

On Wed, Apr 25, 2012 at 4:44 PM, Justin Ruggles wrote: > --- >  libavresample/x86/audio_mix.asm    |   43 > >  libavresample/x86/audio_mix_init.c |    5 >  2 files changed, 48 insertions(+), 0 deletions(-) > > diff --git a/libavresample/x86/audio_mix.asm

Re: [libav-devel] [PATCH 1/3] lavr: add C functions for mixing 2 to 1 channels with s16p format

2012/4/25 Måns Rullgård : > Alex Converse writes: > >>> +    while (len > 4) { >>> +        *dst++ = (*src0++ * m0 + *src1++ * m1) >> 6; >>> +        *dst++ = (*src0++ * m0 + *src1++ * m1) >> 6; >>> +        *dst++ = (*src0++ * m0 + *src1++ * m1) >> 6; >>> +        *dst++ = (*src0++ * m0 + *src1++

Re: [libav-devel] [PATCH] test programs: don't ignore the return value of fwrite()

On Wed, Apr 25, 2012 at 07:03:47PM -0400, Sean McGovern wrote: > > --- a/tests/rotozoom.c > +++ b/tests/rotozoom.c > @@ -23,10 +23,21 @@ > > +static void check_result(size_t size) > +{ > +if(errno) > +{ For future reference: This is not K&R style, check out the relevant section in the d

Re: [libav-devel] [PATCH 1/3] lavr: add C functions for mixing 2 to 1 channels with s16p format

Alex Converse writes: >> +    while (len > 4) { >> +        *dst++ = (*src0++ * m0 + *src1++ * m1) >> 6; >> +        *dst++ = (*src0++ * m0 + *src1++ * m1) >> 6; >> +        *dst++ = (*src0++ * m0 + *src1++ * m1) >> 6; >> +        *dst++ = (*src0++ * m0 + *src1++ * m1) >> 6; > > Isn't this prone

Re: [libav-devel] [PATCH 1/3] lavr: add C functions for mixing 2 to 1 channels with s16p format

On Wed, Apr 25, 2012 at 4:44 PM, Justin Ruggles wrote: > > --- >  libavresample/audio_mix.c |   50 > + >  1 files changed, 50 insertions(+), 0 deletions(-) > > diff --git a/libavresample/audio_mix.c b/libavresample/audio_mix.c > index d5f2efb..8115d52 10

Re: [libav-devel] [PATCH] mov: Parse EC3SpecificBox (dec3 atom).

2012年4月26日8:47 Justin Ruggles : > On 04/25/2012 07:38 PM, Yusuke Nakamura wrote: > > > 2012年4月26日8:16 Justin Ruggles > >: > > > > On 04/25/2012 06:31 PM, Yusuke Nakamura wrote: > > > > > +st->codec->channels = ((int[]){2,1,2,3,3,4,4,5})[acmod] + > lfeo

Re: [libav-devel] [PATCH] mov: Parse EC3SpecificBox (dec3 atom).

On 04/25/2012 07:38 PM, Yusuke Nakamura wrote: > 2012年4月26日8:16 Justin Ruggles >: > > On 04/25/2012 06:31 PM, Yusuke Nakamura wrote: > > > +st->codec->channels = ((int[]){2,1,2,3,3,4,4,5})[acmod] + lfeon; > > +st->codec->channel_layout = avpr

[libav-devel] [PATCH 2/3] lavr: add x86-optimized functions for mixing 2 to 1 s16p with float coeffs

--- libavresample/x86/audio_mix.asm| 61 libavresample/x86/audio_mix_init.c | 13 2 files changed, 74 insertions(+), 0 deletions(-) diff --git a/libavresample/x86/audio_mix.asm b/libavresample/x86/audio_mix.asm index ef30f02..4f79f83 100644 --

[libav-devel] [PATCH 3/3] lavr: add x86-optimized function for mixing 2 to 1 s16p with q6 coeffs

--- libavresample/x86/audio_mix.asm| 43 libavresample/x86/audio_mix_init.c |5 2 files changed, 48 insertions(+), 0 deletions(-) diff --git a/libavresample/x86/audio_mix.asm b/libavresample/x86/audio_mix.asm index 4f79f83..2f65912 100644 --- a/

[libav-devel] [PATCH 1/3] lavr: add C functions for mixing 2 to 1 channels with s16p format

--- libavresample/audio_mix.c | 50 + 1 files changed, 50 insertions(+), 0 deletions(-) diff --git a/libavresample/audio_mix.c b/libavresample/audio_mix.c index d5f2efb..8115d52 100644 --- a/libavresample/audio_mix.c +++ b/libavresample/audio_mix.c @@

[libav-devel] [PATCH 0/3] optimized 2-to-1 channel mixing

This adds some channel-specific C and x86 asm functions for mixing 2 channels to 1 channel. Here are some benchmarks: all tests are with x86_64 G = generic C function C = channel-specific C function mix 2-to-1 s16p/flt --- Atom: G - 84612 C - 40639 SSE2 - 6223 Athlon64:

Re: [libav-devel] [PATCH] test programs: don't ignore the return value of fwrite()

Sean McGovern writes: > --- > tests/rotozoom.c | 17 ++--- > tests/videogen.c | 17 ++--- > 2 files changed, 28 insertions(+), 6 deletions(-) > > diff --git a/tests/rotozoom.c b/tests/rotozoom.c > index 9ce45cd..eea67c3 100644 > --- a/tests/rotozoom.c > +++ b/tests/ro

Re: [libav-devel] [PATCH] mov: Parse EC3SpecificBox (dec3 atom).

2012年4月26日8:16 Justin Ruggles : > On 04/25/2012 06:31 PM, Yusuke Nakamura wrote: > > > +st->codec->channels = ((int[]){2,1,2,3,3,4,4,5})[acmod] + lfeon; > > +st->codec->channel_layout = avpriv_ac3_channel_layout_tab[acmod]; > > +if (lfeon) > > +st->codec->channel_layout |= AV_C

Re: [libav-devel] [PATCH] mov: Parse EC3SpecificBox (dec3 atom).

On 04/25/2012 06:31 PM, Yusuke Nakamura wrote: > +st->codec->channels = ((int[]){2,1,2,3,3,4,4,5})[acmod] + lfeon; > +st->codec->channel_layout = avpriv_ac3_channel_layout_tab[acmod]; > +if (lfeon) > +st->codec->channel_layout |= AV_CH_LOW_FREQUENCY; st->codec->channels = av_

[libav-devel] [PATCH] test programs: don't ignore the return value of fwrite()

--- tests/rotozoom.c | 17 ++--- tests/videogen.c | 17 ++--- 2 files changed, 28 insertions(+), 6 deletions(-) diff --git a/tests/rotozoom.c b/tests/rotozoom.c index 9ce45cd..eea67c3 100644 --- a/tests/rotozoom.c +++ b/tests/rotozoom.c @@ -23,10 +23,21 @@ #include

[libav-devel] [PATCH] mov: Parse EC3SpecificBox (dec3 atom).

Skip to parse fields for additional independent substreams and its associated dependent substreams since libavcodec's eac3 decoder doesn't support them yet. 0001-mov-Parse-EC3SpecificBox-dec3-atom.patch Description: Binary data ___ libav-devel mailing l

Re: [libav-devel] [PATCH] libaacplus: introduce encoder support

On 25/04/12 15:08, Harfe Leier wrote: > The patch#6 (for reference codec's sbr_main.c) from libaacplus 2.0.2 > supports max bitrate up to 72000bps. > Nice to know, let me update my libaacplus to this version =) lu -- Luca Barbato Gentoo/linux http://dev.gentoo.org/~lu_zero __

Re: [libav-devel] [PATCH] vp8: armv6: fix non-armv6t2 build

On 25/04/12 15:09, Mans Rullgard wrote: > The assembler may fail to place literal pools close enough to > instructions referencing them. An explicit .ltorg directive > fixes this. > > Signed-off-by: Mans Rullgard > --- > libavcodec/arm/vp8dsp_armv6.S |2 ++ > 1 file changed, 2 insertions(+)

[libav-devel] [PATCH] vp8: armv6: fix non-armv6t2 build

The assembler may fail to place literal pools close enough to instructions referencing them. An explicit .ltorg directive fixes this. Signed-off-by: Mans Rullgard --- libavcodec/arm/vp8dsp_armv6.S |2 ++ 1 file changed, 2 insertions(+) diff --git a/libavcodec/arm/vp8dsp_armv6.S b/libavcode

Re: [libav-devel] [PATCH] libaacplus: introduce encoder support

The patch#6 (for reference codec's sbr_main.c) from libaacplus 2.0.2 supports max bitrate up to 72000bps. On 23 April 2012 19:14, Luca Barbato wrote: > On 23/04/12 10:13, Kieran Kunhya wrote: >>> +    if (avctx->bit_rate > 64*1024) { >>> +        av_log(avctx, AV_LOG_WARNING, >>> +              

Re: [libav-devel] [PATCH] build: icc: silence some warnings

On 25/04/12 13:15, Mans Rullgard wrote: > This disables the warning "external declaration in primary source file" > which is issued when a prototype for an extern function is found in a > .c file rather than a header file. We have such prototypes for asm > functions where a separate header file wo

Re: [libav-devel] [PATCH] avplay: use libavresample for sample format conversion and channel mixing

On 04/25/2012 04:49 PM, Justin Ruggles wrote: > SDL only supports s16p sample format and a limited number of channel layouts. > Some versions of SDL on some systems support 4-channel and 6-channel output, > but it's safer overall to downmix any layout with more than 2 channels to > stereo. oops,

[libav-devel] [PATCH] avplay: use libavresample for sample format conversion and channel mixing

SDL only supports s16p sample format and a limited number of channel layouts. Some versions of SDL on some systems support 4-channel and 6-channel output, but it's safer overall to downmix any layout with more than 2 channels to stereo. --- avplay.c | 124 +

Re: [libav-devel] [PATCH 1/3] avutil: add av_parse_cpu_flags() function

Hi, On Wed, Apr 25, 2012 at 1:08 PM, Luca Barbato wrote: > The patchset seems fine. +1, and is possibly useful in other test applications also. Ronald ___ libav-devel mailing list libav-devel@libav.org https://lists.libav.org/mailman/listinfo/libav-de

Re: [libav-devel] [PATCH 2/2] vp8: armv6 optimisations

Hi, On Wed, Apr 25, 2012 at 10:27 AM, Luca Barbato wrote: > On 25/04/12 05:12, Mans Rullgard wrote: >> Based on patch by Ronald S. Bultje , >> partially ported from libvpx. >> >> Signed-off-by: Mans Rullgard >> --- >>  libavcodec/arm/vp8dsp_armv6.S      | 1573 >> +++

[libav-devel] [PATCH] build: icc: silence some warnings

This disables the warning "external declaration in primary source file" which is issued when a prototype for an extern function is found in a .c file rather than a header file. We have such prototypes for asm functions where a separate header file would be pointless. Signed-off-by: Mans Rullgard

Re: [libav-devel] [PATCH 1/3] avutil: add av_parse_cpu_flags() function

The patchset seems fine. -- Luca Barbato Gentoo/linux http://dev.gentoo.org/~lu_zero ___ libav-devel mailing list libav-devel@libav.org https://lists.libav.org/mailman/listinfo/libav-devel

[libav-devel] [PATCH 1/3] avutil: add av_parse_cpu_flags() function

This moves the cpu flag parsing code from avconv to avutil so it can be accessed elsewhere. Signed-off-by: Mans Rullgard --- avconv.c | 59 - doc/APIchanges |3 +++ libavutil/avutil.h |2 +- libavutil/cpu.c| 62 +++

[libav-devel] [PATCH 3/3] fft-test: add option to set cpuflag mask

This can be useful for testing. Signed-off-by: Mans Rullgard --- libavcodec/fft-test.c | 10 +- tests/fate/fft.mak| 26 ++ 2 files changed, 23 insertions(+), 13 deletions(-) diff --git a/libavcodec/fft-test.c b/libavcodec/fft-test.c index f890de6..70c0655

[libav-devel] [PATCH 2/3] cpu: recognise only cpu flag names pertinent to the architecture

Signed-off-by: Mans Rullgard --- libavutil/cpu.c |6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/libavutil/cpu.c b/libavutil/cpu.c index d11ff40..6532744 100644 --- a/libavutil/cpu.c +++ b/libavutil/cpu.c @@ -63,7 +63,9 @@ int av_parse_cpu_flags(const char *s) #defin

Re: [libav-devel] [PATCH 1/2] mov: support eac3 audio

On 24/04/2012 9:27 PM, Justin Ruggles wrote: > This one looks ok. Pushed with updated comment, as discussed on IRC. - Derek ___ libav-devel mailing list libav-devel@libav.org https://lists.libav.org/mailman/listinfo/libav-devel

Re: [libav-devel] [PATCH] configure: add POWER7 support

Sean McGovern writes: > On Wednesday, April 25, 2012, Måns Rullgård wrote: >> Sean McGovern writes: >> Could you try disabling -mvsx for the DCT code like this: $ rm libavcodec/jfdct*.o $ make ECFLAGS=-mno-vsx Then run fate again and see if it makes a difference.

Re: [libav-devel] [PATCH] configure: add POWER7 support

On Wednesday, April 25, 2012, Måns Rullgård wrote: > Sean McGovern writes: > >>> Could you try disabling -mvsx for the DCT code like this: >>> >>> $ rm libavcodec/jfdct*.o >>> $ make ECFLAGS=-mno-vsx >>> >>> Then run fate again and see if it makes a difference. If it doesn't, do >>> the same thi

Re: [libav-devel] [PATCH 2/4] avcodec: refactor avcodec_encode_audio2() to merge common branches

On 04/20/2012 04:45 PM, Alex Converse wrote: > On Wed, Mar 21, 2012 at 12:55 PM, Justin Ruggles > wrote: >> >> --- >> libavcodec/utils.c |9 - >> 1 files changed, 4 insertions(+), 5 deletions(-) >> >> diff --git a/libavcodec/utils.c b/libavcodec/utils.c >> index f66295a..6494499 1006

Re: [libav-devel] [PATCH] avf: fix faulty check in has_duration

Hi, On Tue, Apr 24, 2012 at 10:14 PM, Luca Barbato wrote: > An invalid duration is AV_NOPTS_VALUE not 0. > --- >  libavformat/utils.c |    2 +- >  1 files changed, 1 insertions(+), 1 deletions(-) > > diff --git a/libavformat/utils.c b/libavformat/utils.c > index bcdba8c..b167e96 100644 > --- a/li

Re: [libav-devel] [PATCH 2/3] aacps: align some arrays

On 25/04/12 06:09, Måns Rullgård wrote: > Mans Rullgard writes: > >> Signed-off-by: Mans Rullgard >> --- >> libavcodec/aacps.c | 19 +++ >> libavcodec/aacps.h | 20 ++-- >> libavcodec/aacps_tablegen.c | 20 ++-- >> libavcod

Re: [libav-devel] [PATCH 2/2] vp8: armv6 optimisations

On 25/04/12 05:12, Mans Rullgard wrote: > Based on patch by Ronald S. Bultje , > partially ported from libvpx. > > Signed-off-by: Mans Rullgard > --- > libavcodec/arm/vp8dsp_armv6.S | 1573 > > libavcodec/arm/vp8dsp_init_armv6.c | 86 ++- > 2 files ch

Re: [libav-devel] [PATCH 1/2] vp8: arm: separate ARMv6 functions from NEON

On 25/04/12 05:12, Mans Rullgard wrote: > This is a preparation for complete ARMv6 optimisations. > > Signed-off-by: Mans Rullgard > --- Looks fine. -- Luca Barbato Gentoo/linux http://dev.gentoo.org/~lu_zero ___ libav-devel mailing list libav-dev

Re: [libav-devel] [PATCH 1/2] Add libavresample

Hi, On Wed, Apr 25, 2012 at 8:31 AM, Justin Ruggles wrote: > On 04/25/2012 05:36 AM, Diego Biurrun wrote: > >> Nice to see this committed. :) >> >> On Fri, Apr 20, 2012 at 12:02:56PM -0400, Justin Ruggles wrote: >>> >>>  libavresample/avresample-test.c        |  340 ++ >>> ---

Re: [libav-devel] [PATCH 1/2] Add libavresample

On 04/25/2012 05:36 AM, Diego Biurrun wrote: > Nice to see this committed. :) > > On Fri, Apr 20, 2012 at 12:02:56PM -0400, Justin Ruggles wrote: >> >> libavresample/avresample-test.c| 340 ++ >> --- /dev/null >> +++ b/libavresample/Makefile >> @@ -0,0 +1,14 @@ >> + >

Re: [libav-devel] [PATCH 1/2] avserver: check return value of ftruncate()

On Wed, Apr 25, 2012 at 02:08:53PM +0100, Mans Rullgard wrote: > Signed-off-by: Mans Rullgard > --- > avserver.c |5 - > 1 file changed, 4 insertions(+), 1 deletion(-) LGTM Diego ___ libav-devel mailing list libav-devel@libav.org https://lists

Re: [libav-devel] [PATCH] configure: add POWER7 support

Sean McGovern writes: >> Could you try disabling -mvsx for the DCT code like this: >> >> $ rm libavcodec/jfdct*.o >> $ make ECFLAGS=-mno-vsx >> >> Then run fate again and see if it makes a difference. If it doesn't, do >> the same thing with dsputil.o and mpegvideo*.o. > > Same result with all o

Re: [libav-devel] [PATCH] configure: add POWER7 support

On Wednesday, April 25, 2012, Måns Rullgård wrote: > Sean McGovern writes: > >> 2012/4/24 Måns Rullgård : >>> Sean McGovern writes: >>> 2012/4/24 Måns Rullgård : > Sean McGovern writes: > Comparing the output of dct-test between vsx and non-vsx gives identical output. >>>

Re: [libav-devel] [PATCH 3/3] aacps: NEON optimisations

Mans Rullgard writes: > Signed-off-by: Mans Rullgard > --- > libavcodec/aacpsdsp.c |3 + > libavcodec/aacpsdsp.h |1 + > libavcodec/arm/Makefile|6 +- > libavcodec/arm/aacpsdsp_init_arm.c | 53 +++ > libavcodec/arm/aacpsdsp_neon.S | 2

Re: [libav-devel] [PATCH 2/3] aacps: align some arrays

Mans Rullgard writes: > Signed-off-by: Mans Rullgard > --- > libavcodec/aacps.c | 19 +++ > libavcodec/aacps.h | 20 ++-- > libavcodec/aacps_tablegen.c | 20 ++-- > libavcodec/aacps_tablegen.h | 15 --- > liba

[libav-devel] [PATCH 2/2] avserver: remove daemon mode

This code spews a multitude of warnings with glibc (unchecked return values), some of them possibly warranted. Furthermore, the deamonisation is not suitable for use with typical startup scripts as it does not provide the PID of the daemon in any way. Users wishing to run avserver as a daemon can

[libav-devel] [PATCH 1/2] avserver: check return value of ftruncate()

Signed-off-by: Mans Rullgard --- avserver.c |5 - 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/avserver.c b/avserver.c index 2b16932..5777a08 100644 --- a/avserver.c +++ b/avserver.c @@ -2572,8 +2572,11 @@ static int http_start_receive_data(HTTPContext *c) if (c->stre

[libav-devel] [PATCH 1/2] vp8: arm: separate ARMv6 functions from NEON

This is a preparation for complete ARMv6 optimisations. Signed-off-by: Mans Rullgard --- libavcodec/arm/Makefile|7 +- libavcodec/arm/vp8dsp.h| 78 +++ libavcodec/arm/vp8dsp_armv6.S | 53 + libavcodec/arm/vp8dsp_init_arm.c | 144

[libav-devel] [PATCH 0/2] vp8: armv6 optimisations

Addressed Diego's nit-picking. Fixed copyright and licence text. Mans Rullgard (2): vp8: arm: separate ARMv6 functions from NEON vp8: armv6 optimisations libavcodec/arm/Makefile|7 +- libavcodec/arm/vp8dsp.h| 78 ++ libavcodec/arm/vp8dsp_armv6.S | 1626 +++

Re: [libav-devel] [PATCH] configure: add POWER7 support

Sean McGovern writes: > 2012/4/24 Måns Rullgård : >> Sean McGovern writes: >> >>> 2012/4/24 Måns Rullgård : Sean McGovern writes: >>> Comparing the output of dct-test between vsx and non-vsx gives >>> identical output. >> >> Can you post the .err file from a failing test, doesn't matt

Re: [libav-devel] [PATCH 1/2] Add libavresample

Nice to see this committed. :) On Fri, Apr 20, 2012 at 12:02:56PM -0400, Justin Ruggles wrote: > > libavresample/avresample-test.c| 340 ++ > --- /dev/null > +++ b/libavresample/Makefile > @@ -0,0 +1,14 @@ > + > +TESTPROGS = avresample > --- /dev/null > +++ b/libavres