Re: [libav-devel] [PATCH] h264: avoid pointless copying of ref lists

On Thu, 17 Jan 2013 14:03:26 +0100, Anton Khirnov wrote: > ref_list is constructed from other fields per slice when needed, so do > not copy it for both frame and slice threading. > default_ref_list is constructed per frame and still needs to be copied > to per-slice contexts for slice threading,

Re: [libav-devel] [PATCH] Remove put_no_rnd_pixels_l2 function pointer for w=16 from dsputil.

On 18/01/13 07:07, Ronald S. Bultje wrote: > From: "Ronald S. Bultje" > > The function is never used. > --- > libavcodec/dsputil.c | 3 +-- > libavcodec/dsputil.h | 2 +- > libavcodec/dsputil_template.c | 4 > libavcodec/vp3.c | 2 +- > libavcodec/vp56.c

Re: [libav-devel] [PATCH 46/59] sipr: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:12:05 -0500, Justin Ruggles wrote: > --- > libavcodec/sipr.c | 15 ++- > libavcodec/sipr.h |1 - > 2 files changed, 6 insertions(+), 10 deletions(-) > > diff --git a/libavcodec/sipr.c b/libavcodec/sipr.c > index d482b0f..6db529e 100644 > --- a/libavcode

Re: [libav-devel] [PATCH 45/59] shorten: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:12:04 -0500, Justin Ruggles wrote: > --- > libavcodec/shorten.c | 14 +- > 1 files changed, 5 insertions(+), 9 deletions(-) > > diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c > index 1dc010f..48bef32 100644 > --- a/libavcodec/shorten.c > +++ b/lib

Re: [libav-devel] [PATCH 44/59] s302m: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:12:03 -0500, Justin Ruggles wrote: > --- > libavcodec/s302m.c | 34 -- > 1 files changed, 8 insertions(+), 26 deletions(-) > > diff --git a/libavcodec/s302m.c b/libavcodec/s302m.c > index 76fab59..061f015 100644 > --- a/libavcodec/s302m.

Re: [libav-devel] [PATCH 43/59] ra288: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:12:02 -0500, Justin Ruggles wrote: > --- > libavcodec/ra288.c | 14 +- > 1 files changed, 5 insertions(+), 9 deletions(-) > > diff --git a/libavcodec/ra288.c b/libavcodec/ra288.c > index 8266673..d1037b1 100644 > --- a/libavcodec/ra288.c > +++ b/libavcodec/

Re: [libav-devel] [PATCH 42/59] ra144: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:12:01 -0500, Justin Ruggles wrote: > --- > libavcodec/ra144.h|1 - > libavcodec/ra144dec.c | 13 + > 2 files changed, 5 insertions(+), 9 deletions(-) > > diff --git a/libavcodec/ra144.h b/libavcodec/ra144.h > index 73f83f0..81d6964 100644 > --- a/li

Re: [libav-devel] [PATCH 41/59] ralf: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:12:00 -0500, Justin Ruggles wrote: > --- > libavcodec/ralf.c | 19 +++ > 1 files changed, 7 insertions(+), 12 deletions(-) > > diff --git a/libavcodec/ralf.c b/libavcodec/ralf.c > index 580def3..b163a89 100644 > --- a/libavcodec/ralf.c > +++ b/libavcode

Re: [libav-devel] [PATCH 40/59] qdm2: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:11:59 -0500, Justin Ruggles wrote: > --- > libavcodec/qdm2.c | 15 +-- > 1 files changed, 5 insertions(+), 10 deletions(-) > > diff --git a/libavcodec/qdm2.c b/libavcodec/qdm2.c > index 269a051..1286cc5 100644 > --- a/libavcodec/qdm2.c > +++ b/libavcodec/qd

Re: [libav-devel] [PATCH 39/59] qcelp: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:11:58 -0500, Justin Ruggles wrote: > --- > libavcodec/qcelpdec.c | 14 +- > 1 files changed, 5 insertions(+), 9 deletions(-) > > diff --git a/libavcodec/qcelpdec.c b/libavcodec/qcelpdec.c > index b702175..48cf555 100644 > --- a/libavcodec/qcelpdec.c > +++ b

Re: [libav-devel] [PATCH 38/59] pcm-bluray: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:11:57 -0500, Justin Ruggles wrote: > --- > libavcodec/pcm-mpeg.c | 29 ++--- > 1 files changed, 6 insertions(+), 23 deletions(-) > > diff --git a/libavcodec/pcm-mpeg.c b/libavcodec/pcm-mpeg.c > index c8c683d..41bcb3e 100644 > --- a/libavcodec/pc

Re: [libav-devel] [PATCH 37/59] nellymoser: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:11:56 -0500, Justin Ruggles wrote: > --- > libavcodec/nellymoserdec.c | 14 +- > 1 files changed, 5 insertions(+), 9 deletions(-) > > diff --git a/libavcodec/nellymoserdec.c b/libavcodec/nellymoserdec.c > index b6aa6f5..4177eaf 100644 > --- a/libavcodec/nel

Re: [libav-devel] [PATCH 36/59] mpc7/8: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:11:55 -0500, Justin Ruggles wrote: > --- > libavcodec/mpc.h |1 - > libavcodec/mpc7.c | 13 + > libavcodec/mpc8.c | 13 + > 3 files changed, 10 insertions(+), 17 deletions(-) > > diff --git a/libavcodec/mpc.h b/libavcodec/mpc.h > index 86

Re: [libav-devel] [PATCH 35/59] mpegaudio: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:11:54 -0500, Justin Ruggles wrote: > --- > libavcodec/mpegaudiodec.c | 55 +--- > 1 files changed, 26 insertions(+), 29 deletions(-) > > diff --git a/libavcodec/mpegaudiodec.c b/libavcodec/mpegaudiodec.c > index fda0280..e5db694

Re: [libav-devel] [PATCH 34/59] mlp/truehd: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:11:53 -0500, Justin Ruggles wrote: > --- > libavcodec/mlpdec.c | 17 ++--- > 1 files changed, 6 insertions(+), 11 deletions(-) > > diff --git a/libavcodec/mlpdec.c b/libavcodec/mlpdec.c > index 3852f6e..63aa477 100644 > --- a/libavcodec/mlpdec.c > +++ b/lib

Re: [libav-devel] [PATCH 33/59] mace: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:11:52 -0500, Justin Ruggles wrote: > --- > libavcodec/mace.c | 16 +--- > 1 files changed, 5 insertions(+), 11 deletions(-) > > diff --git a/libavcodec/mace.c b/libavcodec/mace.c > index c78a207..368e5d7 100644 > --- a/libavcodec/mace.c > +++ b/libavcodec/m

Re: [libav-devel] [PATCH 32/59] libspeex: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:11:51 -0500, Justin Ruggles wrote: > --- > libavcodec/libspeexdec.c | 14 +- > 1 files changed, 5 insertions(+), 9 deletions(-) > > diff --git a/libavcodec/libspeexdec.c b/libavcodec/libspeexdec.c > index 8aa82ca..549ff1d 100644 > --- a/libavcodec/libspeexd

Re: [libav-devel] [PATCH 31/59] libopus: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:11:50 -0500, Justin Ruggles wrote: > --- > libavcodec/libopusdec.c | 25 - > 1 files changed, 12 insertions(+), 13 deletions(-) > > diff --git a/libavcodec/libopusdec.c b/libavcodec/libopusdec.c > index 15fa493..0bf040b 100644 > --- a/libavcodec

Re: [libav-devel] [PATCH 30/59] libopencore-amr: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:11:49 -0500, Justin Ruggles wrote: > --- > libavcodec/libopencore-amr.c | 28 ++-- > 1 files changed, 10 insertions(+), 18 deletions(-) > > diff --git a/libavcodec/libopencore-amr.c b/libavcodec/libopencore-amr.c > index 8dc2e9f..6646586 100644

Re: [libav-devel] [PATCH 29/59] libgsm: decode directly to the user-provided AVFrame

On Wed, 9 Jan 2013 12:11:48 -0500, Justin Ruggles wrote: > --- > libavcodec/libgsm.c | 14 +- > 1 files changed, 5 insertions(+), 9 deletions(-) > > diff --git a/libavcodec/libgsm.c b/libavcodec/libgsm.c > index b4d1c74..310185a 100644 > --- a/libavcodec/libgsm.c > +++ b/libavco

[libav-devel] [PATCH] Remove put_no_rnd_pixels_l2 function pointer for w=16 from dsputil.

From: "Ronald S. Bultje" The function is never used. --- libavcodec/dsputil.c | 3 +-- libavcodec/dsputil.h | 2 +- libavcodec/dsputil_template.c | 4 libavcodec/vp3.c | 2 +- libavcodec/vp56.c | 6 +++--- 5 files changed, 6 insertions(+), 11 delet

Re: [libav-devel] [PATCH] Convert vorbisdsp x86 functions from inline asm to yasm.

Hi, On Thu, Jan 17, 2013 at 3:29 PM, Vitor Sessak wrote: > On Wed, Jan 16, 2013 at 1:58 AM, Ronald S. Bultje wrote: >> +INIT_XMM sse >> +cglobal vorbis_inverse_coupling, 3, 3, 6, mag, ang, block_size >> +movsxdifnidnblock_sizeq, block_sized >> +mova m5, [pdw_8

Re: [libav-devel] [PATCH] fate: update ref after rv30_loop_filter fix

On Fri, Jan 18, 2013 at 12:26:12AM +0200, Martin Storsjö wrote: > On Thu, 17 Jan 2013, Luca Barbato wrote: > > >--- > > > >Leftover from the previous patch... > > > >tests/ref/fate/rv30 | 164 > >++-- > >1 file changed, 82 insertions(+), 82 deletions

Re: [libav-devel] [PATCH] Convert vorbisdsp x86 functions from inline asm to yasm.

On Wed, Jan 16, 2013 at 1:58 AM, Ronald S. Bultje wrote: Hi! > +INIT_XMM sse > +cglobal vorbis_inverse_coupling, 3, 3, 6, mag, ang, block_size > +movsxdifnidnblock_sizeq, block_sized > +mova m5, [pdw_8000] > +leamagq, [magq+block_sizeq*

Re: [libav-devel] [PATCH] fate: update ref after rv30_loop_filter fix

On Thu, 17 Jan 2013, Luca Barbato wrote: --- Now fate gpl passes as well. tests/ref/fate/filter-delogo | 164 +-- tests/ref/fate/rv30 | 164 +-- 2 files changed, 164 insertions(+), 164 deletions(-) Ok I gu

[libav-devel] [PATCH] fate: update ref after rv30_loop_filter fix

--- Now fate gpl passes as well. tests/ref/fate/filter-delogo | 164 +-- tests/ref/fate/rv30 | 164 +-- 2 files changed, 164 insertions(+), 164 deletions(-) diff --git a/tests/ref/fate/filter-delogo b/tests

Re: [libav-devel] [PATCH] fate: update ref after rv30_loop_filter fix

On 17/01/13 23:28, Martin Storsjö wrote: > On Thu, 17 Jan 2013, Luca Barbato wrote: > >> On 17/01/13 23:26, Martin Storsjö wrote: >>> On Thu, 17 Jan 2013, Luca Barbato wrote: >>> --- Leftover from the previous patch... tests/ref/fate/rv30 | 164 +++

Re: [libav-devel] [PATCH] fate: update ref after rv30_loop_filter fix

On Thu, 17 Jan 2013, Luca Barbato wrote: On 17/01/13 23:26, Martin Storsjö wrote: On Thu, 17 Jan 2013, Luca Barbato wrote: --- Leftover from the previous patch... tests/ref/fate/rv30 | 164 ++-- 1 file changed, 82 insertions(+), 82 deletions(-)

Re: [libav-devel] [PATCH] fate: update ref after rv30_loop_filter fix

On 17/01/13 23:26, Martin Storsjö wrote: > On Thu, 17 Jan 2013, Luca Barbato wrote: > >> --- >> >> Leftover from the previous patch... >> >> tests/ref/fate/rv30 | 164 >> ++-- >> 1 file changed, 82 insertions(+), 82 deletions(-) > > Is this enough? T

Re: [libav-devel] [PATCH] Convert vorbisdsp x86 functions from inline asm to yasm.

Hi, On Thu, Jan 17, 2013 at 12:52 PM, Luca Barbato wrote: > On 17/01/13 21:37, Justin Ruggles wrote: >> On 01/17/2013 11:25 AM, Ronald S. Bultje wrote: >>> +cglobal vorbis_inverse_coupling, 3, 3, 6, mag, ang, block_size >>> +movsxdifnidnblock_sizeq, block_sized >> >> I assume you'll remov

Re: [libav-devel] [PATCH] fate: update ref after rv30_loop_filter fix

On Thu, 17 Jan 2013, Luca Barbato wrote: --- Leftover from the previous patch... tests/ref/fate/rv30 | 164 ++-- 1 file changed, 82 insertions(+), 82 deletions(-) Is this enough? The filter-delogo test also started failing. Also, have you visua

[libav-devel] [PATCH] fate: update ref after rv30_loop_filter fix

--- Leftover from the previous patch... tests/ref/fate/rv30 | 164 ++-- 1 file changed, 82 insertions(+), 82 deletions(-) diff --git a/tests/ref/fate/rv30 b/tests/ref/fate/rv30 index 4883423..89a9d8c 100644 --- a/tests/ref/fate/rv30 +++ b/tests/re

Re: [libav-devel] [PATCH] Convert vorbisdsp x86 functions from inline asm to yasm.

On 17/01/13 21:37, Justin Ruggles wrote: > On 01/17/2013 11:25 AM, Ronald S. Bultje wrote: >> +cglobal vorbis_inverse_coupling, 3, 3, 6, mag, ang, block_size >> +movsxdifnidnblock_sizeq, block_sized > > I assume you'll remove the movsxdifnidn when changing block_size to intptr_t > > Other

Re: [libav-devel] [PATCH] lavc: Move vector_fmul_add to AVFloatDSPContext

Hi, On Jan 17, 2013 12:29 PM, "Justin Ruggles" wrote: > > On 01/17/2013 11:28 AM, Ronald S. Bultje wrote: > > Hi, > > > > On Thu, Jan 17, 2013 at 2:15 AM, Reinhard Tartler wrote: > >> On Thu, Jan 17, 2013 at 11:11 AM, Martin Storsjö wrote: > >>> On Thu, 17 Jan 2013, Reinhard Tartler wrote: > >>

Re: [libav-devel] [PATCH] lavc: Move vector_fmul_add to AVFloatDSPContext

On 17/01/13 21:17, Reinhard Tartler wrote: > On Thu, Jan 17, 2013 at 9:02 PM, Luca Barbato wrote: >> On 17/01/13 11:15, Reinhard Tartler wrote: If the members are added at the end, there shouldn't be too much of an issue though. >>> >>> If that works, then that's clearly preferable

Re: [libav-devel] [PATCH] Convert vorbisdsp x86 functions from inline asm to yasm.

On 01/17/2013 11:25 AM, Ronald S. Bultje wrote: > +cglobal vorbis_inverse_coupling, 3, 3, 6, mag, ang, block_size > +movsxdifnidnblock_sizeq, block_sized I assume you'll remove the movsxdifnidn when changing block_size to intptr_t Otherwise LGTM -Justin _

Re: [libav-devel] [RFC] Dropping 0.6

On 17/01/13 21:32, Derek Buitenhuis wrote: > Sean and Reinhard have already OK'd this idea. > > This is a formal email to ask if there are any objections. > If nobody is using it I see no point in keeping it around. ___ libav-devel mailing list libav-

[libav-devel] [RFC] Dropping 0.6

Sean and Reinhard have already OK'd this idea. This is a formal email to ask if there are any objections. - Derek ___ libav-devel mailing list libav-devel@libav.org https://lists.libav.org/mailman/listinfo/libav-devel

Re: [libav-devel] [PATCH] lavc: Move vector_fmul_add to AVFloatDSPContext

On 01/17/2013 11:28 AM, Ronald S. Bultje wrote: > Hi, > > On Thu, Jan 17, 2013 at 2:15 AM, Reinhard Tartler wrote: >> On Thu, Jan 17, 2013 at 11:11 AM, Martin Storsjö wrote: >>> On Thu, 17 Jan 2013, Reinhard Tartler wrote: >>> On Wed, Jan 16, 2013 at 7:08 PM, Ronald S. Bultje wrote: >

Re: [libav-devel] [PATCH] Change block_size type from int to intptr_t.

On 01/17/2013 11:25 AM, Ronald S. Bultje wrote: > From: "Ronald S. Bultje" > > This saves one instruction in the x86-64 assembly. > --- > libavcodec/arm/vorbisdsp_init_arm.c | 3 ++- > libavcodec/ppc/vorbisdsp_altivec.c | 2 +- > libavcodec/vorbis.h | 2 +- > libavcodec/vorbisde

Re: [libav-devel] [PATCH] Move vorbis_inverse_coupling from dsputil to vorbisdspcontext.

On 01/17/2013 11:24 AM, Ronald S. Bultje wrote: > + > +#include "libavutil/attributes.h" Why is attributes.h needed? > +#include "libavutil/cpu.h" > +#include "libavutil/arm/cpu.h" > +#include "libavcodec/vorbisdsp.h" > + > +void ff_vorbis_inverse_coupling_neon(float *mag, float *ang, int blocksi

Re: [libav-devel] [PATCH] lavc: Move vector_fmul_add to AVFloatDSPContext

On Thu, Jan 17, 2013 at 9:02 PM, Luca Barbato wrote: > On 17/01/13 11:15, Reinhard Tartler wrote: >>> If the members are added at the end, there shouldn't be too much of an issue >>> though. >> >> If that works, then that's clearly preferable to bumping major! > > Shall we stash it within the Evil

Re: [libav-devel] [RFC] [PATCH 2/2] x86: vp3dsp: Separate VP3-specific inline asm functions from dsputil

On 01/17/2013 11:18 AM, Ronald S. Bultje wrote: > Hi, > > On Thu, Jan 17, 2013 at 7:35 AM, Diego Biurrun wrote: >> On Thu, Jan 17, 2013 at 07:09:02AM -0800, Ronald S. Bultje wrote: >>> On Thu, Jan 17, 2013 at 1:45 AM, Diego Biurrun wrote: --- This is a quick idea I had to give Dan

Re: [libav-devel] [PATCH] lavc: Move vector_fmul_add to AVFloatDSPContext

On 17/01/13 11:15, Reinhard Tartler wrote: >> If the members are added at the end, there shouldn't be too much of an issue >> though. > > If that works, then that's clearly preferable to bumping major! Shall we stash it within the Evil Plan big bump? lu

Re: [libav-devel] [PATCH] float_dsp: Add #ifdef HAVE_INLINE_ASM around vector_fmul_window

Hi, On Thu, Jan 17, 2013 at 8:59 AM, Martin Storsjö wrote: > This fixes builds on 64bit MSVC. > --- > libavutil/x86/float_dsp_init.c |6 +++--- > 1 file changed, 3 insertions(+), 3 deletions(-) LGTM. Ronald ___ libav-devel mailing list libav-deve

[libav-devel] [PATCH] float_dsp: Add #ifdef HAVE_INLINE_ASM around vector_fmul_window

This fixes builds on 64bit MSVC. --- libavutil/x86/float_dsp_init.c |6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/libavutil/x86/float_dsp_init.c b/libavutil/x86/float_dsp_init.c index 56fb32b..1c678cb 100644 --- a/libavutil/x86/float_dsp_init.c +++ b/libavutil/x86/fl

Re: [libav-devel] [PATCH] lavc: Move vector_fmul_add to AVFloatDSPContext

Hi, On Thu, Jan 17, 2013 at 2:15 AM, Reinhard Tartler wrote: > On Thu, Jan 17, 2013 at 11:11 AM, Martin Storsjö wrote: >> On Thu, 17 Jan 2013, Reinhard Tartler wrote: >> >>> On Wed, Jan 16, 2013 at 7:08 PM, Ronald S. Bultje >>> wrote: Hi, On Jan 16, 2013 10:01 AM, "Justin Ru

[libav-devel] [PATCH] Change block_size type from int to intptr_t.

From: "Ronald S. Bultje" This saves one instruction in the x86-64 assembly. --- libavcodec/arm/vorbisdsp_init_arm.c | 3 ++- libavcodec/ppc/vorbisdsp_altivec.c | 2 +- libavcodec/vorbis.h | 2 +- libavcodec/vorbisdec.c | 2 +- libavcodec/vorbisdsp.h | 5

[libav-devel] [PATCH] Convert vorbisdsp x86 functions from inline asm to yasm.

From: "Ronald S. Bultje" --- libavcodec/x86/Makefile | 1 + libavcodec/x86/dsputil_mmx.c| 3 -- libavcodec/x86/dsputil_mmx.h| 2 - libavcodec/x86/vorbisdsp.asm| 84 + libavcodec/x86/vorbisdsp_init.c | 73 +++--

[libav-devel] [PATCH] Move vorbis_inverse_coupling from dsputil to vorbisdspcontext.

From: "Ronald S. Bultje" Conveniently (together with Justin's earlier patches), this makes our vorbis decoder entirely independent of dsputil. (Untested on ppc/arm.) --- libavcodec/Makefile | 2 +- libavcodec/arm/Makefile | 3 ++ libavcodec/arm/dsputil_init_neon.

Re: [libav-devel] [RFC] [PATCH 2/2] x86: vp3dsp: Separate VP3-specific inline asm functions from dsputil

Hi, On Thu, Jan 17, 2013 at 7:35 AM, Diego Biurrun wrote: > On Thu, Jan 17, 2013 at 07:09:02AM -0800, Ronald S. Bultje wrote: >> On Thu, Jan 17, 2013 at 1:45 AM, Diego Biurrun wrote: >> > --- >> > >> > This is a quick idea I had to give Daniel another small, separate piece >> > of dsputil inline

Re: [libav-devel] [RFC] [PATCH 2/2] x86: vp3dsp: Separate VP3-specific inline asm functions from dsputil

On Thu, Jan 17, 2013 at 07:09:02AM -0800, Ronald S. Bultje wrote: > On Thu, Jan 17, 2013 at 1:45 AM, Diego Biurrun wrote: > > --- > > > > This is a quick idea I had to give Daniel another small, separate piece > > of dsputil inline assembly to work on w/o having to fiddle with the C > > code. The

Re: [libav-devel] [PATCH] sensible error codes for avidec

On Thu, Jan 17, 2013 at 4:09 PM, Diego Biurrun wrote: >> >> On Thu, Jan 17, 2013 at 11:57 AM, Jindřich Makovička >> >> wrote: >> >> --- a/libavformat/avidec.c >> >> +++ b/libavformat/avidec.c >> >> @@ -1015,8 +1015,9 @@ static int avi_read_packet(AVFormatContext *s, >> >> AVPacket *pkt) >> >>

[libav-devel] [PATCH] Convert vorbisdsp x86 functions from inline asm to yasm.

From: "Ronald S. Bultje" --- libavcodec/x86/Makefile | 3 +- libavcodec/x86/dsputil_mmx.c| 3 -- libavcodec/x86/dsputil_mmx.h| 2 - libavcodec/x86/vorbisdsp.asm| 84 + libavcodec/x86/vorbisdsp_init.c | 71 +++-

Re: [libav-devel] [PATCH] sensible error codes for avidec

On Thu, Jan 17, 2013 at 01:56:17PM +0100, Jindřich Makovička wrote: > On Thu, Jan 17, 2013 at 12:37 PM, Diego Biurrun wrote: > > On Thu, Jan 17, 2013 at 12:22:49PM +0100, Jindřich Makovička wrote: > >> On Thu, Jan 17, 2013 at 11:57 AM, Jindřich Makovička > >> wrote: > >> --- a/libavformat/avidec

Re: [libav-devel] [RFC] [PATCH 2/2] x86: vp3dsp: Separate VP3-specific inline asm functions from dsputil

Hi, On Thu, Jan 17, 2013 at 1:45 AM, Diego Biurrun wrote: > --- > > This is a quick idea I had to give Daniel another small, separate piece > of dsputil inline assembly to work on w/o having to fiddle with the C > code. The vp3dsp_template ugliness will go away the moment this is > converted to

Re: [libav-devel] [PATCH] theora: Skip zero-sized headers

On 17/01/13 15:11, Martin Storsjö wrote: > This fixes a regression since d9cf5f51 with theora over RTP > (possibly with other variants of theora as well). > > In theora over RTP, the second of the 3 headers turns out to be > 0 bytes long, which prior to d9cf5f51 worked just fine. After > d9cf5f51,

Re: [libav-devel] [PATCH 5/5] avconv: Increase the SDP buffer size to fit xiph SDPs

On Thu, 17 Jan 2013, Diego Elio Pettenò wrote: On 17/01/2013 15:21, Martin Storsjö wrote: Yes, it's not all that nice, but not actually too much of an issue for the systems where people actually use the avconv front end AFAIK... Good point — I guess the only place I don't know how would re

Re: [libav-devel] [PATCH 5/5] avconv: Increase the SDP buffer size to fit xiph SDPs

On 17/01/2013 15:21, Martin Storsjö wrote: >> > > Yes, it's not all that nice, but not actually too much of an issue for > the systems where people actually use the avconv front end AFAIK... Good point — I guess the only place I don't know how would react for this would be Windows.. if you feel i

Re: [libav-devel] [PATCH 5/5] avconv: Increase the SDP buffer size to fit xiph SDPs

On Thu, 17 Jan 2013, Diego Elio Pettenò wrote: On 17/01/2013 15:12, Martin Storsjö wrote: -char sdp[2048]; +char sdp[16384]; Hrm, 16KiB on stack? (theoretically) Yes, it's not all that nice, but not actually too much of an issue for the systems where people actually use the avconv

Re: [libav-devel] [PATCH 5/5] avconv: Increase the SDP buffer size to fit xiph SDPs

On 17/01/2013 15:12, Martin Storsjö wrote: > -char sdp[2048]; > +char sdp[16384]; Hrm, 16KiB on stack? (theoretically) -- Diego Elio Pettenò — Flameeyes flamee...@flameeyes.eu — http://blog.flameeyes.eu/ ___ libav-devel mailing list libav-devel

Re: [libav-devel] [PATCH] lavc: support ZenoXVID custom tag

On Thu, 17 Jan 2013, Luca Barbato wrote: Looks like this kind of samples are produced by certain Russian equipment. --- libavcodec/h263dec.c | 2 +- libavformat/riff.c | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) Looks ok I guess, someone more familiar with riff formats perhaps can

[libav-devel] [PATCH 5/5] avconv: Increase the SDP buffer size to fit xiph SDPs

--- avconv.c |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/avconv.c b/avconv.c index dd998ab..f086cbb 100644 --- a/avconv.c +++ b/avconv.c @@ -1367,7 +1367,7 @@ static int output_packet(InputStream *ist, const AVPacket *pkt) static void print_sdp(void) { -char sdp

[libav-devel] [PATCH 4/5] rtpdec: Move setting the parsing flags to the actual depacketizers

This gets rid of almost all the codec specific details from the generic rtpdec code. --- libavformat/rtpdec.c | 12 libavformat/rtpdec_h263.c | 10 ++ libavformat/rtpdec_h263_rfc2190.c |9 + libavformat/rtpdec_h264.c |9

[libav-devel] [PATCH 3/5] rtpdec: Split handling of mpeg12 audio/video to a separate depacketizer

This also adds checking of mallocs. --- libavformat/Makefile |1 + libavformat/rtpdec.c | 40 +++ libavformat/rtpdec_formats.h |2 ++ libavformat/rtpdec_mpeg12.c | 62 ++ 4 files changed, 69 insertions(+),

[libav-devel] [PATCH 2/5] rtpdec: Split mpegts parsing to a normal depacketizer

--- libavformat/Makefile |1 + libavformat/rtpdec.c | 56 +-- libavformat/rtpdec.h |6 --- libavformat/rtpdec_formats.h |1 + libavformat/rtpdec_mpegts.c | 100 ++ libavformat/rtsp.c | 1

[libav-devel] [PATCH 1/5] rtpdec: Reorder payload handler registration alphabetically

--- libavformat/rtpdec.c | 33 +++-- 1 file changed, 15 insertions(+), 18 deletions(-) diff --git a/libavformat/rtpdec.c b/libavformat/rtpdec.c index e052ee6..28b4033 100644 --- a/libavformat/rtpdec.c +++ b/libavformat/rtpdec.c @@ -61,39 +61,36 @@ void ff_register_d

[libav-devel] [PATCH] theora: Skip zero-sized headers

This fixes a regression since d9cf5f51 with theora over RTP (possibly with other variants of theora as well). In theora over RTP, the second of the 3 headers turns out to be 0 bytes long, which prior to d9cf5f51 worked just fine. After d9cf5f51, reading from the bitstream reader fails (since the r

Re: [libav-devel] [PATCH] libcdio: support recent cdio-paranoia

On 17/01/2013 14:53, Luca Barbato wrote: > You lose the paranoia option, but the code is much simpler, I could try > baking something later if you are willing to test it a bit. You better find somebody else to test it for real as I haven't used an audio cd on linux for the past ten years or so...

Re: [libav-devel] [PATCH] libcdio: support recent cdio-paranoia

On 17/01/13 14:43, Diego Elio Pettenò wrote: > Mostly LGTM considering the note you already fixed locally. > > How feasible it is to have this work with non-CDIO Paranoia library as well? > You lose the paranoia option, but the code is much simpler, I could try baking something later if you are

Re: [libav-devel] [PATCH] libcdio: support recent cdio-paranoia

Mostly LGTM considering the note you already fixed locally. How feasible it is to have this work with non-CDIO Paranoia library as well? -- Diego Elio Pettenò — Flameeyes flamee...@flameeyes.eu — http://blog.flameeyes.eu/ ___ libav-devel mailing list l

[libav-devel] [PATCH] lavc: support ZenoXVID custom tag

Looks like this kind of samples are produced by certain Russian equipment. --- libavcodec/h263dec.c | 2 +- libavformat/riff.c | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/libavcodec/h263dec.c b/libavcodec/h263dec.c index fc5f565..7e64538 100644 --- a/libavcodec/h263dec.c

[libav-devel] [PATCH] h264: avoid pointless copying of ref lists

ref_list is constructed from other fields per slice when needed, so do not copy it for both frame and slice threading. default_ref_list is constructed per frame and still needs to be copied to per-slice contexts for slice threading, but a copy is not needed for frame threading. --- libavcodec/h264

Re: [libav-devel] [PATCH] sensible error codes for avidec

On Thu, Jan 17, 2013 at 12:37 PM, Diego Biurrun wrote: > On Thu, Jan 17, 2013 at 12:22:49PM +0100, Jindřich Makovička wrote: >> On Thu, Jan 17, 2013 at 11:57 AM, Jindřich Makovička >> wrote: >> > >> > these patches change return -1 in avidec to return AVERROR(EINVAL), >> > except one case where

Re: [libav-devel] [PATCH] sensible error codes for avidec

On Thu, Jan 17, 2013 at 12:22:49PM +0100, Jindřich Makovička wrote: > On Thu, Jan 17, 2013 at 11:57 AM, Jindřich Makovička > wrote: > > > > these patches change return -1 in avidec to return AVERROR(EINVAL), > > except one case where the return -1 corresponds to an EOF condition in > > an interle

Re: [libav-devel] [PATCH] sensible error codes for avidec

On 17/01/13 12:22, Jindřich Makovička wrote: > I forgot to amend a lost semicolon before sending, here are fixed patches. Thank you, looks fine to me. ___ libav-devel mailing list libav-devel@libav.org https://lists.libav.org/mailman/listinfo/libav-devel

Re: [libav-devel] [PATCH] sensible error codes for avidec

I forgot to amend a lost semicolon before sending, here are fixed patches. On Thu, Jan 17, 2013 at 11:57 AM, Jindřich Makovička wrote: > Hi, > > these patches change return -1 in avidec to return AVERROR(EINVAL), > except one case where the return -1 corresponds to an EOF condition in > an interl

[libav-devel] [PATCH] sensible error codes for avidec

Hi, these patches change return -1 in avidec to return AVERROR(EINVAL), except one case where the return -1 corresponds to an EOF condition in an interleaved AVI file. Regards, -- Jindřich Makovička 0001-avidec-return-EOF-when-no-more-frames-are-available.patch Description: Binary data 0002-a

Re: [libav-devel] [PATCH] libcdio: support recent cdio-paranoia

On 17/01/13 11:25, Luca Barbato wrote: > enabled libcdio && > -check_lib2 "cdio/cdda.h cdio/paranoia.h" cdio_cddap_open -lcdio_paranoia > -lcdio_cdda -lcdio > +check_lib2 "cdio/cdda.h cdio/paranoia.h" cdio_cddap_open -lcdio_paranoia > -lcdio_cdda -lcdio || check_lib2 "cdio/paranoia/cdda.

[libav-devel] [PATCH] libcdio: support recent cdio-paranoia

Upstream decided to split the paranoia interface and move the headers accordingly. --- I'd land it in release/9 as well. configure | 4 +++- libavdevice/libcdio.c | 7 +++ 2 files changed, 10 insertions(+), 1 deletion(-) diff --git a/configure b/configure index 8cc9cdc..f8244ee

Re: [libav-devel] [PATCH] lavc: Move vector_fmul_add to AVFloatDSPContext

On Thu, Jan 17, 2013 at 11:11 AM, Martin Storsjö wrote: > On Thu, 17 Jan 2013, Reinhard Tartler wrote: > >> On Wed, Jan 16, 2013 at 7:08 PM, Ronald S. Bultje >> wrote: >>> >>> Hi, >>> >>> On Jan 16, 2013 10:01 AM, "Justin Ruggles" >>> wrote: On 01/16/2013 12:46 PM, Ronald S. Bultj

Re: [libav-devel] [PATCH] lavc: Move vector_fmul_add to AVFloatDSPContext

On Thu, 17 Jan 2013, Reinhard Tartler wrote: On Wed, Jan 16, 2013 at 7:08 PM, Ronald S. Bultje wrote: Hi, On Jan 16, 2013 10:01 AM, "Justin Ruggles" wrote: On 01/16/2013 12:46 PM, Ronald S. Bultje wrote: Hi, On Jan 16, 2013 9:43 AM, "Luca Barbato" mailto:lu_z...@gentoo.org>> wrote: On

Re: [libav-devel] [PATCH] lavc: Move vector_fmul_add to AVFloatDSPContext

On Wed, Jan 16, 2013 at 7:08 PM, Ronald S. Bultje wrote: > Hi, > > On Jan 16, 2013 10:01 AM, "Justin Ruggles" wrote: >> >> On 01/16/2013 12:46 PM, Ronald S. Bultje wrote: >> > Hi, >> > >> > On Jan 16, 2013 9:43 AM, "Luca Barbato" > > > wrote: >> >> >> >> On 16/01/13 11:

[libav-devel] [RFC] [PATCH 2/2] x86: vp3dsp: Separate VP3-specific inline asm functions from dsputil

--- This is a quick idea I had to give Daniel another small, separate piece of dsputil inline assembly to work on w/o having to fiddle with the C code. The vp3dsp_template ugliness will go away the moment this is converted to yasm , which should be really soon. Passes all tests on x86_64 and x86

[libav-devel] [PATCH 1/2] x86: dsputil: Drop some unused macro definitions

--- libavcodec/x86/dsputil_mmx.c |4 1 files changed, 0 insertions(+), 4 deletions(-) diff --git a/libavcodec/x86/dsputil_mmx.c b/libavcodec/x86/dsputil_mmx.c index b16f7e4..a2f60e3 100644 --- a/libavcodec/x86/dsputil_mmx.c +++ b/libavcodec/x86/dsputil_mmx.c @@ -196,14 +196,12 @@ DECLARE