Re: [libav-devel] [PATCH 05/15] lavr: x86: optimized 2-channel fltp to s16 conversion

Hi, On Sat, Jul 14, 2012 at 9:29 PM, Justin Ruggles wrote: > --- > libavresample/x86/audio_convert.asm| 37 > > libavresample/x86/audio_convert_init.c |9 +++ > 2 files changed, 46 insertions(+), 0 deletions(-) > > diff --git a/libavresample/x86/au

Re: [libav-devel] [PATCH 04/15] lavr: x86: optimized 6-channel s16p to flt conversion

Hi, On Sat, Jul 21, 2012 at 12:12 PM, Justin Ruggles wrote: > +%if cpuflag(ssse3) > +pshufb m3, m0, unpack_odd ; m3 = 12, 13, 14, 15 > +pshufb m0, unpack_even ; m0 = 0, 1, 2, 3 > +pshufb m4, m1, unpack_odd ; m4 = 16, 17, 18,

Re: [libav-devel] [PATCH 03/15] lavr: x86: optimized 2-channel s16p to flt conversion

Hi, On Sat, Jul 14, 2012 at 9:29 PM, Justin Ruggles wrote: > --- > libavresample/x86/audio_convert.asm| 49 > > libavresample/x86/audio_convert_init.c |9 ++ > 2 files changed, 58 insertions(+), 0 deletions(-) LGTM. Ronald ___

Re: [libav-devel] [PATCH 02/15] lavr: x86: optimized 6-channel s16p to s16 conversion

Hi, On Sat, Jul 14, 2012 at 9:29 PM, Justin Ruggles wrote: > --- > libavresample/x86/audio_convert.asm| 62 > > libavresample/x86/audio_convert_init.c |9 + > 2 files changed, 71 insertions(+), 0 deletions(-) > > diff --git a/libavresample/x86/audi

Re: [libav-devel] [PATCH 01/15] lavr: x86: optimized 2-channel s16p to s16 conversion

Hi, On Sat, Jul 14, 2012 at 9:29 PM, Justin Ruggles wrote: > --- > libavresample/x86/audio_convert.asm| 36 > > libavresample/x86/audio_convert_init.c | 13 +++ > 2 files changed, 49 insertions(+), 0 deletions(-) > > diff --git a/libavresample/x8

Re: [libav-devel] [PATCH] lavr: x86: improve non-SSE4 version of S16_TO_S32_SX macro

Hi, On Sat, Jul 14, 2012 at 10:33 AM, Justin Ruggles wrote: > On 06/26/2012 04:55 PM, Justin Ruggles wrote: >> Removes a false dependency on existing contents of the 2nd dst register, >> giving better performance for OOE. >> --- >> libavresample/x86/util.asm |3 ++- >> 1 files changed, 2 ins

Re: [libav-devel] [libav-commits] x86: fft: convert sse inline asm to yasm

Hi, On Tue, Jul 24, 2012 at 3:05 PM, Jason Garrett-Glaser wrote: > On Tue, Jul 24, 2012 at 9:02 AM, John Stebbins > wrote: >> On 07/24/2012 05:53 PM, Jason Garrett-Glaser wrote: >>> >>> On Tue, Jul 24, 2012 at 8:34 AM, Måns Rullgård wrote: Jason Garrett-Glaser writes: > On

[libav-devel] [PATCH] Replace x*155/100 by x*101581>>16.

From: "Ronald S. Bultje" Idea stolen from webp (by Pascal Massimino) - because it's Cool. --- libavcodec/vp8.c | 12 ++-- 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/libavcodec/vp8.c b/libavcodec/vp8.c index d0e2a0c..e4cfbcb 100644 --- a/libavco

Re: [libav-devel] [PATCH] H264: Remove 3dnow qpel code.

Hi, On Tue, Jul 24, 2012 at 7:25 AM, Luca Barbato wrote: > On 7/24/12 4:45 AM, Ronald S. Bultje wrote: >> >> Hi, >> >> On Mon, Jul 23, 2012 at 7:45 PM, Ronald S. Bultje >> wrote: >>> >>> Hi, >>> >>> On Mon, Jul 23, 2012 at 5:

Re: [libav-devel] [PATCH] H264: Remove 3dnow qpel code.

Hi, On Mon, Jul 23, 2012 at 7:45 PM, Ronald S. Bultje wrote: > Hi, > > On Mon, Jul 23, 2012 at 5:37 PM, Daniel Kang wrote: >> On Mon, Jul 23, 2012 at 5:21 PM, Diego Biurrun wrote: >>> >>> On Mon, Jul 23, 2012 at 05:12:23PM -0700, Daniel Kang wrote: >>>

Re: [libav-devel] [PATCH] H264: Remove 3dnow qpel code.

Hi, On Mon, Jul 23, 2012 at 5:37 PM, Daniel Kang wrote: > On Mon, Jul 23, 2012 at 5:21 PM, Diego Biurrun wrote: >> >> On Mon, Jul 23, 2012 at 05:12:23PM -0700, Daniel Kang wrote: >> > From: Daniel Kang >> > >> > The only CPUs that have 3dnow and don't have mmxext are 12 years old. >> > --- >> >

Re: [libav-devel] [PATCH] h264: refactor NAL decode loop

Hi, On Mon, Jul 23, 2012 at 7:29 AM, Luca Barbato wrote: > From: "Ronald S. Bultje" > > Write out the NAL decoding loops in full so that they are easier > to parse for a preprocessor without it having to be aware of macros > or other such things in C code. > >

Re: [libav-devel] [PATCH] h264: refactor NAL decode loop.

Hi, On Mon, Jul 23, 2012 at 7:14 AM, Kostya Shishkov wrote: > On Mon, Jul 23, 2012 at 07:11:49AM -0700, Ronald S. Bultje wrote: >> Hi, >> >> On Mon, Jul 23, 2012 at 2:05 AM, Diego Biurrun wrote: >> > On Sun, Jul 22, 2012 at 08:46:10PM -0700, Ronald S. Bultje wrote:

Re: [libav-devel] [PATCH] h264: convert loop filter strength dsp function to yasm.

Hi, On Sun, Jul 22, 2012 at 1:16 PM, Ronald S. Bultje wrote: > From: "Ronald S. Bultje" > > This completes the conversion of h264dsp to yasm; note that h264 also > uses some dsputil functions, most notably qpel. Performance-wise, the > yasm-version is ~10 cycles faster

Re: [libav-devel] [PATCH 2/2] vp3: don't use calls to inline asm in yasm code.

Hi, On Sun, Jul 22, 2012 at 2:38 PM, Ronald S. Bultje wrote: > From: "Ronald S. Bultje" > > Mixing yasm and inline asm is a bad idea, since if either yasm or inline > asm is not supported by your toolchain, all of the asm stops working. > Thus, better to use eith

Re: [libav-devel] [PATCH] h264: refactor NAL decode loop.

Hi, On Mon, Jul 23, 2012 at 2:05 AM, Diego Biurrun wrote: > On Sun, Jul 22, 2012 at 08:46:10PM -0700, Ronald S. Bultje wrote: >> From: "Ronald S. Bultje" >> >> Write out the NAL decoding loops in full so that they are easier to >> parse for a preprocessor wi

[libav-devel] [PATCH] h264: refactor NAL decode loop.

From: "Ronald S. Bultje" Write out the NAL decoding loops in full so that they are easier to parse for a preprocessor without it having to be aware of macros or other such things in C code. This also makes the code more readable. --- libavcodec/h26

Re: [libav-devel] [PATCH 1/2] build: support non-standard replacements for -E flag

Hi, On Sun, Jul 22, 2012 at 5:35 PM, Måns Rullgård wrote: > Diego Biurrun writes: > >> On Mon, Jul 23, 2012 at 01:16:07AM +0100, Måns Rullgård wrote: >>> Diego Biurrun writes: >>> >>> > On Mon, Jul 23, 2012 at 12:16:41AM +0100, Mans Rullgard wrote: >>> >> This allows using non-standard flags fo

Re: [libav-devel] [PATCH] swscale: Mark all init functions as av_cold

Hi, On Sun, Jul 22, 2012 at 3:30 PM, Diego Biurrun wrote: > --- > libswscale/output.c | 15 --- > libswscale/ppc/swscale_altivec.c |3 ++- > libswscale/ppc/yuv2rgb_altivec.c | 11 +++ > libswscale/rgb2rgb.c |3 ++- > libswscale/sparc/yuv2r

[libav-devel] [PATCH] swscale: place inline assembly bilinear scaler under HAVE_INLINE_ASM.

From: "Ronald S. Bultje" --- libswscale/utils.c | 10 +- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/libswscale/utils.c b/libswscale/utils.c index d8fee58..a6b5a18 100644 --- a/libswscale/utils.c +++ b/libswscale/utils.c @@ -576,7 +576,7 @@ fail:

[libav-devel] [PATCH] x86/dsputil: put inline asm under HAVE_INLINE_ASM.

From: "Ronald S. Bultje" This allows compiling with compilers that don't support gcc-style inline assembly. --- libavcodec/dct-test.c|2 +- libavcodec/x86/dsputil_mmx.c | 69 -- libavcodec/x86/h264_qpel_mmx.c |4 ++-

[libav-devel] [PATCH] x86/dsputil: put inline asm under HAVE_INLINE_ASM.

From: "Ronald S. Bultje" This allows compiling with compilers that don't support gcc-style inline assembly. --- libavcodec/dct-test.c|2 +- libavcodec/x86/dsputil_mmx.c | 69 -- libavcodec/x86/h264_qpel_mmx.c |4 ++-

[libav-devel] [PATCH 2/2] vp3: don't use calls to inline asm in yasm code.

From: "Ronald S. Bultje" Mixing yasm and inline asm is a bad idea, since if either yasm or inline asm is not supported by your toolchain, all of the asm stops working. Thus, better to use either one or the other alone. --- libavcodec/x86/vp3dsp.

[libav-devel] [PATCH 1/2] x86/dsputil: put inline asm under HAVE_INLINE_ASM.

From: "Ronald S. Bultje" This allows compiling with compilers that don't support gcc-style inline assembly. --- libavcodec/x86/dsputil_mmx.c | 69 -- libavcodec/x86/h264_qpel_mmx.c |4 ++- libavcodec/x86/idct_mmx.c|4

[libav-devel] [PATCH] swscale: add fast bilinear scaler under HAVE_INLINE_ASM.

From: "Ronald S. Bultje" --- libswscale/utils.c | 10 +- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/libswscale/utils.c b/libswscale/utils.c index d8fee58..a6b5a18 100644 --- a/libswscale/utils.c +++ b/libswscale/utils.c @@ -576,7 +576,7 @@ fail:

[libav-devel] [PATCH] swscale: add missing HAVE_INLINE_ASM check.

From: "Ronald S. Bultje" The function called in this block is under HAVE_INLINE_ASM itself also. --- libswscale/swscale.c |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libswscale/swscale.c b/libswscale/swscale.c index 7ae5af3..5cfa7f2 100644 --- a/libswscale

[libav-devel] [PATCH] h264: convert loop filter strength dsp function to yasm.

From: "Ronald S. Bultje" This completes the conversion of h264dsp to yasm; note that h264 also uses some dsputil functions, most notably qpel. Performance-wise, the yasm-version is ~10 cycles faster (182->172) on x86-64, and ~8 cycles faster (201->193) on x86-32. ---

[libav-devel] [PATCH] h264: convert loop filter strength dsp function to yasm.

From: "Ronald S. Bultje" This completes the conversion of h264dsp to yasm; note that h264 also uses some dsputil functions, most notably qpel. Performance-wise, the yasm-version is ~10 cycles faster (182->172) on x86-64, and ~8 cycles faster (201->193) on x86-32. ---

Re: [libav-devel] [PATCH] h264: refactor NAL decoding loop.

Hi, On Sun, Jul 22, 2012 at 8:17 AM, Måns Rullgård wrote: > "Ronald S. Bultje" writes: > >>> This manner of splitting things is incredibly weird-looking. Instead of >>> trying to unify these rather different fragments, turning the second >>> ha

Re: [libav-devel] [PATCH] h264: refactor NAL decoding loop.

Hi, On Sat, Jul 21, 2012 at 5:19 PM, Måns Rullgård wrote: > "Ronald S. Bultje" writes: > >> From: "Ronald S. Bultje" >> >> This removes some code duplication between the 3 different versions, >> and aligns brackets in such a way that it is now

Re: [libav-devel] [PATCH] lavfi: put inline assembly under HAVE_INLINE_ASM.

Hi, On Sat, Jul 21, 2012 at 5:03 PM, Ronald S. Bultje wrote: > From: "Ronald S. Bultje" > > This allows compiling this code using compilers that do not understand > gcc-style inline assembly. > --- > libavfilter/x86/gradfun.c |6 ++ > libavfilter/x86/yad

[libav-devel] [PATCH] yadif: use emms_c() instead of inline assembly for emms invocations.

From: "Ronald S. Bultje" --- libavfilter/vf_yadif.c |5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/libavfilter/vf_yadif.c b/libavfilter/vf_yadif.c index 40cfdf5..02feeb2 100644 --- a/libavfilter/vf_yadif.c +++ b/libavfilter/vf_yadif.c @@ -135,9 +135,8 @@ s

[libav-devel] [PATCH] lavfi: put inline assembly under HAVE_INLINE_ASM.

From: "Ronald S. Bultje" This allows compiling this code using compilers that do not understand gcc-style inline assembly. --- libavfilter/x86/gradfun.c |6 ++ libavfilter/x86/yadif.c |6 ++ 2 files changed, 12 insertions(+) diff --git a/libavfilter/x86/g

[libav-devel] [PATCH] h264: refactor NAL decoding loop.

From: "Ronald S. Bultje" This removes some code duplication between the 3 different versions, and aligns brackets in such a way that it is now possible to pull this code through a naive pre-processor that doesn't necessarily have to be aware of compiler-macros. --- libavcode

[libav-devel] [PATCH] lavf: use conditional notation for default codec in muxer declarations.

From: "Ronald S. Bultje" This removes the use of macro nesting in these code constructs, which makes it easier to parse in pre-processors. --- libavformat/asfenc.c | 12 ++-- libavformat/avienc.c |6 +- libavformat/flvenc.c |6 +- l

[libav-devel] [PATCH] ac3: don't use different names for option tables in the template file.

From: "Ronald S. Bultje" The variables which are declared in the teplate file are static and therefore there is no symbol clash. --- libavcodec/ac3enc_fixed.c |2 +- libavcodec/ac3enc_opts_template.c |6 -- libavcodec/eac3enc.c |2 +- 3 files

Re: [libav-devel] [PATCH 3/3] dsputil: x86: add SHUFFLE_MASK_W macro

Hi, On Fri, Jul 20, 2012 at 3:38 PM, Justin Ruggles wrote: > From: Jason Garrett-Glaser > > Simplifies pshufb masks that operate on words. > --- > Ported from x264. LGPL relicense ok'd by Jason on IRC. > > libavcodec/x86/dsputil_yasm.asm |2 +- > libavutil/x86/x86util.asm | 12 +

[libav-devel] [PATCH] adpcm: don't duplicate identical AVSampleFmt array for each encoder.

From: "Ronald S. Bultje" --- libavcodec/adpcmenc.c |6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/libavcodec/adpcmenc.c b/libavcodec/adpcmenc.c index 73c5403..fa40c22 100644 --- a/libavcodec/adpcmenc.c +++ b/libavcodec/adpcmenc.c @@ -716,6 +716,

[libav-devel] [PATCH] ac3: define option table name before including option table template.

From: "Ronald S. Bultje" --- libavcodec/ac3enc_fixed.c |1 + libavcodec/ac3enc_float.c |1 + libavcodec/ac3enc_opts_template.c |8 +--- libavcodec/eac3enc.c |1 + 4 files changed, 4 insertions(+), 7 deletions(-) diff --git a/

[libav-devel] [PATCH] lavf: use conditional notation for default codec in muxer declarations.

From: "Ronald S. Bultje" --- libavformat/asfenc.c | 12 ++-- libavformat/avienc.c |6 +- libavformat/flvenc.c |6 +- libavformat/matroskaenc.c | 21 ++--- libavformat/movenc.c | 21 ++--- libavforma

[libav-devel] [PATCH] rtsp: remove terminal comma in FF_RTP_FLAG_OPTS macro.

From: "Ronald S. Bultje" --- libavformat/movenc.c |2 +- libavformat/rtpenc.c |2 +- libavformat/rtpenc.h |2 +- libavformat/rtsp.c |2 +- 4 files changed, 4 insertions(+), 4 deletions(-) diff --git a/libavformat/movenc.c b/libavformat/movenc.c index 350ac95..2a5f

Re: [libav-devel] [PATCH] libavcodec: implement LOCAL_ALIGNED() without variadic macro arguments.

Hi, On Sun, Jul 15, 2012 at 3:31 PM, Måns Rullgård wrote: > "Ronald S. Bultje" writes: > >> Hi, >> >> On Sun, Jul 8, 2012 at 1:34 PM, Luca Barbato wrote: >>> On 07/06/2012 07:55 PM, Ronald S. Bultje wrote: >>>> From: "Ronald S. Bu

Re: [libav-devel] [PATCH] dsputil: drop unnecessary empty parameter

Hi, On Sun, Jul 15, 2012 at 12:26 PM, Luca Barbato wrote: > From: "Ronald S. Bultje" > > --- > libavcodec/dsputil.h |6 +++--- > 1 files changed, 3 insertions(+), 3 deletions(-) (Copying directly from pastebin, so sorry for the line wraps) I was thinkin

[libav-devel] [PATCH] swscale: rename x86/rgb2rgb.c to rgb2rgb_mmx.c.

From: "Ronald S. Bultje" This is consistent with names of other files that contain x86 SIMD optimizations (MMX and up). --- libswscale/x86/Makefile |2 +- libswscale/x86/{rgb2rgb.c => rgb2rgb_mmx.c} |0 2 files changed, 1 insertion(+), 1 deleti

Re: [libav-devel] [PATCH] swscale: x86: Conditionally compile inline Assembly optimizations

Hi, On Sun, Jul 15, 2012 at 5:26 AM, Diego Biurrun wrote: > Skip the optimizations if the compiler does not support inline Assembly. > --- > libswscale/rgb2rgb.c|2 +- > libswscale/x86/Makefile |5 +++-- > libswscale/yuv2rgb.c|2 +- > 3 files changed, 5 insertions(+), 4 delet

[libav-devel] [PATCH] swscale: use HAVE_INLINE_ASM in direct calls to inline assembly.

From: "Ronald S. Bultje" --- libswscale/swscale.c |4 ++-- libswscale/utils.c | 10 +- 2 files changed, 7 insertions(+), 7 deletions(-) diff --git a/libswscale/swscale.c b/libswscale/swscale.c index 7ae5af3..0f8ef2b 100644 --- a/libswscale/swscale.c +++ b/libswscale

[libav-devel] [PATCH 4/4] swscale: rename x86/rgb2rgb.c to rgb2rgb_mmx.c.

From: "Ronald S. Bultje" This is consistent with names of other files that contain x86 SIMD optimizations (MMX and up). --- libswscale/rgb2rgb.c |2 +- libswscale/x86/Makefile |2 +- libswscale/x86/rgb2rgb.c | 144 --

[libav-devel] [PATCH 3/4] swscale: use HAVE_INLINE_ASM in x86/rgb2rgb.c.

From: "Ronald S. Bultje" --- libswscale/x86/rgb2rgb.c |6 ++ 1 file changed, 6 insertions(+) diff --git a/libswscale/x86/rgb2rgb.c b/libswscale/x86/rgb2rgb.c index 282618c..353b0aa 100644 --- a/libswscale/x86/rgb2rgb.c +++ b/libswscale/x86/rgb2rgb.c @@ -33,6 +33,8 @@

[libav-devel] [PATCH 2/4] swscale: use HAVE_INLINE_ASM in swscale_mmx.c.

From: "Ronald S. Bultje" --- libswscale/x86/swscale_mmx.c |6 ++ 1 file changed, 6 insertions(+) diff --git a/libswscale/x86/swscale_mmx.c b/libswscale/x86/swscale_mmx.c index 37d786a..b57b7a4 100644 --- a/libswscale/x86/swscale_mmx.c +++ b/libswscale/x86/swscale_mmx.c @@ -

[libav-devel] [PATCH 1/4] swscale: use HAVE_INLINE_ASM in yuv2rgb_mmx.c.

From: "Ronald S. Bultje" --- libswscale/x86/yuv2rgb_mmx.c |6 ++ 1 file changed, 6 insertions(+) diff --git a/libswscale/x86/yuv2rgb_mmx.c b/libswscale/x86/yuv2rgb_mmx.c index 0eaea77..f534e0e 100644 --- a/libswscale/x86/yuv2rgb_mmx.c +++ b/libswscale/x86/yuv2rgb_mmx.c @@ -

Re: [libav-devel] [PATCH] libavcodec: implement LOCAL_ALIGNED() without variadic macro arguments.

Hi, On Sun, Jul 8, 2012 at 1:34 PM, Luca Barbato wrote: > On 07/06/2012 07:55 PM, Ronald S. Bultje wrote: >> From: "Ronald S. Bultje" >> >> MSVC has incomplete variadic macro argument handling, meaning it does >> support it to some extend, but it will basic

Re: [libav-devel] [PATCH] snow: refactor code to work around a compiler bug in MSVC.

Hi, On Sun, Jul 8, 2012 at 3:43 PM, Benjamin Larsson wrote: > On 07/07/2012 06:06 PM, Ronald S. Bultje wrote: >> >> From: "Ronald S. Bultje" >> >> This fixes the compiler error "cannot convert from 'BlockNode' to >> 'int16_t'&

Re: [libav-devel] [PATCH 1/2] vp8: Include the thread headers before using the pthread types

Hi, On Sat, Jul 14, 2012 at 2:19 PM, Martin Storsjö wrote: > This was unnoticed on linux, since stdlib.h apparently includes > files declaring the pthread_mutex_t and pthread_cond_t types. > --- > libavcodec/vp8.c |6 -- > libavcodec/vp8.h |5 + > 2 files changed, 5 insertions(+)

Re: [libav-devel] [PATCH] vc1dec: Do not use random pred_flag if motion vector data is skipped

Hi, On Sat, Jul 14, 2012 at 6:37 PM, Mashiat Sarker Shakkhar wrote: > This fixes SA10143.vc1 from test-suite. Also partially fixes MC-VC1.ts > from videolan streams archive. > --- > libavcodec/vc1dec.c |2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) > > diff --git a/libavcodec/vc1d

Re: [libav-devel] [PATCH 1/4] fft-test: don't use getopt().

Hi, On Mon, Jul 9, 2012 at 12:44 PM, Anton Khirnov wrote: > > After some discussion on IRC I think I'll drop this set and we'll go > back to the original Ronald's getopt() replacement, just moved to some > less controversial place like compat/. > > I'll send a new patch tomorrow if nobody beats m

Re: [libav-devel] [PATCH] configure: check for math functions without linking to -lm also.

Hi, On Sun, Jul 8, 2012 at 12:48 PM, Måns Rullgård wrote: > "Ronald S. Bultje" writes: > >> Hi, >> >> On Sat, Jul 7, 2012 at 3:57 PM, Derek Buitenhuis >> wrote: >>> On 07/07/2012 6:09 PM, Diego Biurrun wrote: >>>> And this is require

Re: [libav-devel] [PATCH] snow: refactor code to work around a compiler bug in MSVC.

Hi, On Sun, Jul 8, 2012 at 12:46 PM, Måns Rullgård wrote: > "Ronald S. Bultje" writes: > >> From: "Ronald S. Bultje" >> >> This fixes the compiler error "cannot convert from 'BlockNode' to >> 'int16_t'". > >

Re: [libav-devel] [PATCH] dct/fft-test: use a replacement getopt() if the system has none present.

Hi, On Sun, Jul 8, 2012 at 12:36 PM, Måns Rullgård wrote: > "Ronald S. Bultje" writes: >> On Wed, Jul 4, 2012 at 2:09 PM, Måns Rullgård wrote: >>> "Ronald S. Bultje" writes: >>>> On Tue, Jul 3, 2012 at 8:14 PM, Ronald S. Bultje >>&g

Re: [libav-devel] [PATCH] configure: check for math functions without linking to -lm also.

Hi, On Sat, Jul 7, 2012 at 3:57 PM, Derek Buitenhuis wrote: > On 07/07/2012 6:09 PM, Diego Biurrun wrote: >> And this is required - why? > > Libm / math is not A Thing in MSVC and isn't required. > > It complains about -lm, but ignores it. > > I think this is mostly to shut up a warning. Correct

Re: [libav-devel] [PATCH] x86/timer: implement an intrinsic-based version for rdtsc (AV_READ_TIME).

Hi, On Sat, Jul 7, 2012 at 1:49 PM, Måns Rullgård wrote: > "Ronald S. Bultje" wrote: >>On Fri, Jul 6, 2012 at 10:30 AM, Luca Barbato >>wrote: >>> On 07/06/2012 07:22 PM, Ronald S. Bultje wrote: >>>> From: "Ronald S. Bultje" >>>

Re: [libav-devel] [PATCH 4/4] graph2dot: don't use getopt().

Hi, On Sat, Jul 7, 2012 at 1:47 PM, Måns Rullgård wrote: > "Ronald S. Bultje" wrote: >>On Sat, Jul 7, 2012 at 11:03 AM, Anton Khirnov >>wrote: >>> -while ((c = getopt(argc, argv, "hi:o:")) != -1) { >>> -switch (c) { &

Re: [libav-devel] [PATCH] configure: support linkers that expect non-standard library pre/suffixes.

Hi, On Sat, Jul 7, 2012 at 2:03 PM, Ronald S. Bultje wrote: > From: "Ronald S. Bultje" > > This is similar in style to how it currently supports some exotic > compilers (e.g. see CC_O). > --- > configure | 13 ++--- > 1 file changed, 10 insertions(+)

Re: [libav-devel] [PATCH] cmutils: include shellapi.h on Win32 (for CommandLineToArgvW).

Hi, On Sat, Jul 7, 2012 at 2:06 PM, Martin Storsjö wrote: > On Sat, 7 Jul 2012, Ronald S. Bultje wrote: > >> From: "Ronald S. Bultje" >> >> --- >> cmdutils.c |1 + >> 1 file changed, 1 insertion(+) >> >> diff --git a/cmdutil

[libav-devel] [PATCH] configure: support linkers that expect non-standard library pre/suffixes.

From: "Ronald S. Bultje" This is similar in style to how it currently supports some exotic compilers (e.g. see CC_O). --- configure | 13 ++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/configure b/configure index ebc6310..c8ac2b2 100755 --- a/confi

[libav-devel] [PATCH] configure: check for math functions without linking to -lm also.

From: "Ronald S. Bultje" --- configure |4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/configure b/configure index da60e8e..ebc6310 100755 --- a/configure +++ b/configure @@ -2943,7 +2943,9 @@ for thread in $THREADS_LIST; do fi done -check_lib math

[libav-devel] [PATCH] cmutils: include shellapi.h on Win32 (for CommandLineToArgvW).

From: "Ronald S. Bultje" --- cmdutils.c |1 + 1 file changed, 1 insertion(+) diff --git a/cmdutils.c b/cmdutils.c index 6d13bd6..11a5f03 100644 --- a/cmdutils.c +++ b/cmdutils.c @@ -162,6 +162,7 @@ static const OptionDef *find_option(const OptionDef *po, const char *name) #

Re: [libav-devel] [PATCH] x86/timer: implement an intrinsic-based version for rdtsc (AV_READ_TIME).

Hi, On Fri, Jul 6, 2012 at 10:30 AM, Luca Barbato wrote: > On 07/06/2012 07:22 PM, Ronald S. Bultje wrote: >> From: "Ronald S. Bultje" > > Seems fine. Pushed. Ronald ___ libav-devel mailing list libav-devel@libav.org https:

Re: [libav-devel] [PATCH 4/4] graph2dot: don't use getopt().

Hi, On Sat, Jul 7, 2012 at 11:03 AM, Anton Khirnov wrote: > -while ((c = getopt(argc, argv, "hi:o:")) != -1) { > -switch (c) { > -case 'h': > +while (optindex < argc) { > +char *opt = argv[optindex++]; > + > +if (!opt[0]) { Can that trigger? I don't think

Re: [libav-devel] [PATCH] mxfdec: replace x>>av_log2(sizeof(..)) by x/sizeof(..).

Hi, On Fri, Jul 6, 2012 at 11:10 AM, Luca Barbato wrote: > On 07/06/2012 07:31 PM, Ronald S. Bultje wrote: >> Hi, >> >> On Fri, Jul 6, 2012 at 10:29 AM, Luca Barbato wrote: >>> On 07/06/2012 07:12 PM, Ronald S. Bultje wrote: >>>> Hi, >>>>

[libav-devel] [PATCH] snow: refactor code to work around a compiler bug in MSVC.

From: "Ronald S. Bultje" This fixes the compiler error "cannot convert from 'BlockNode' to 'int16_t'". --- libavcodec/snowenc.c |7 ++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/libavcodec/snowenc.c b/libavcodec/snowenc.c index

Re: [libav-devel] [PATCH] dct/fft-test: use a replacement getopt() if the system has none present.

Hi, On Sat, Jul 7, 2012 at 5:26 AM, Diego Biurrun wrote: > On Fri, Jul 06, 2012 at 09:59:50PM -0700, Ronald S. Bultje wrote: >> On Fri, Jul 6, 2012 at 6:53 PM, Diego Biurrun wrote: >> > On Fri, Jul 06, 2012 at 10:33:19AM -0700, Ronald S. Bultje wrote: >> >> On Fri,

[libav-devel] [PATCH] x86/cpu: implement get/set_eflags using intrinsics.

From: "Ronald S. Bultje" --- configure |3 +++ libavutil/x86/cpu.c | 14 ++ 2 files changed, 17 insertions(+) diff --git a/configure b/configure index 598b33c..bb4d55f 100755 --- a/configure +++ b/configure @@ -1126,6 +1126,7 @@ HAVE_LIST="

Re: [libav-devel] [PATCH] dct/fft-test: use a replacement getopt() if the system has none present.

Hi, On Fri, Jul 6, 2012 at 6:53 PM, Diego Biurrun wrote: > On Fri, Jul 06, 2012 at 10:33:19AM -0700, Ronald S. Bultje wrote: >> On Fri, Jul 6, 2012 at 10:27 AM, Luca Barbato wrote: >> > On 07/06/2012 07:13 PM, Ronald S. Bultje wrote: >> >> On Wed, Jul 4, 2012 at

Re: [libav-devel] [PATCH 2/3] x86/cpu: implement support for xgetbv through intrinsics.

Hi, On Fri, Jul 6, 2012 at 10:03 AM, Ronald S. Bultje wrote: > check_code cc mmintrin.h "_mm_empty()" && enable mm_empty > +check_code cc immintrin.h "xgetbv(0)" && enable xgetbv Missing _, fixed locally... Ronald

[libav-devel] [PATCH] libavcodec: implement LOCAL_ALIGNED() without variadic macro arguments.

From: "Ronald S. Bultje" MSVC has incomplete variadic macro argument handling, meaning it does support it to some extend, but it will basically handle all variadic arguments as a single argument, including the comma that separates them, thus making it useless for anything else than as

Re: [libav-devel] [PATCH] dct/fft-test: use a replacement getopt() if the system has none present.

Hi, On Fri, Jul 6, 2012 at 10:27 AM, Luca Barbato wrote: > On 07/06/2012 07:13 PM, Ronald S. Bultje wrote: >> On Wed, Jul 4, 2012 at 2:09 PM, Måns Rullgård wrote: >>> "Ronald S. Bultje" writes: >>>> On Tue, Jul 3, 2012 at 8:14 PM, Ronald S. Bultje &

Re: [libav-devel] [PATCH] mxfdec: replace x>>av_log2(sizeof(..)) by x/sizeof(..).

Hi, On Fri, Jul 6, 2012 at 10:29 AM, Luca Barbato wrote: > On 07/06/2012 07:12 PM, Ronald S. Bultje wrote: >> Hi, >> >> On Thu, Jul 5, 2012 at 4:17 PM, Ronald S. Bultje wrote: >>> On Wed, Jul 4, 2012 at 8:14 AM, Ronald S. Bultje wrote: >>>> From: &q

[libav-devel] [PATCH] x86/timer: implement an intrinsic-based version for rdtsc (AV_READ_TIME).

From: "Ronald S. Bultje" --- configure |2 ++ libavutil/x86/timer.h |8 2 files changed, 10 insertions(+) diff --git a/configure b/configure index 07dbb46..61b773f 100755 --- a/configure +++ b/configure @@ -1125,6 +1125,7 @@ HAVE_LIST="

[libav-devel] [PATCH] x86/cpu: implement get/set_eflags using MSVC-style inline assembly.

From: "Ronald S. Bultje" --- configure |3 +++ libavutil/x86/cpu.c | 21 + 2 files changed, 24 insertions(+) diff --git a/configure b/configure index 72c4943..07dbb46 100755 --- a/configure +++ b/configure @@ -1120,6 +1120,7 @@ HAVE_LIST=&quo

Re: [libav-devel] [PATCH] dct/fft-test: use a replacement getopt() if the system has none present.

Hi, On Wed, Jul 4, 2012 at 2:09 PM, Måns Rullgård wrote: > "Ronald S. Bultje" writes: >> On Tue, Jul 3, 2012 at 8:14 PM, Ronald S. Bultje wrote: >>> From: "Ronald S. Bultje" >>> >>> This allows compiling and running these tests on syst

Re: [libav-devel] [PATCH] mxfdec: replace x>>av_log2(sizeof(..)) by x/sizeof(..).

Hi, On Thu, Jul 5, 2012 at 4:17 PM, Ronald S. Bultje wrote: > On Wed, Jul 4, 2012 at 8:14 AM, Ronald S. Bultje wrote: >> From: "Ronald S. Bultje" >> >> Also replace x>>av_log2(sizeof(..)) + 1 by x/sizeof(..). The +1 is >> probably meant to emul

[libav-devel] [PATCH 3/3] x86/cpu: implement support for cpuid through intrinsics.

From: "Ronald S. Bultje" --- configure |2 ++ libavutil/x86/cpu.c | 14 ++ 2 files changed, 16 insertions(+) diff --git a/configure b/configure index 70b3f56..72c4943 100755 --- a/configure +++ b/configure @@ -1061,6 +1061,7 @@ HAVE_LIST=&q

[libav-devel] [PATCH 2/3] x86/cpu: implement support for xgetbv through intrinsics.

From: "Ronald S. Bultje" --- configure |2 ++ libavutil/x86/cpu.c | 11 +++ 2 files changed, 13 insertions(+) diff --git a/configure b/configure index 3d02e75..70b3f56 100755 --- a/configure +++ b/configure @@ -1169,6 +1169,7 @@ HAVE_LIST="

[libav-devel] [PATCH 1/3] lavu: use intrinsic for emms on systems lacking inline asm support.

From: "Ronald S. Bultje" --- configure|3 +++ libavutil/internal.h |7 +-- 2 files changed, 8 insertions(+), 2 deletions(-) diff --git a/configure b/configure index 4c1cdc3..3d02e75 100755 --- a/configure +++ b/configure @@ -1117,6 +1117,7 @@

Re: [libav-devel] [PATCH] mxfdec: replace x>>av_log2(sizeof(..)) by x/sizeof(..).

Hi, On Wed, Jul 4, 2012 at 8:14 AM, Ronald S. Bultje wrote: > From: "Ronald S. Bultje" > > Also replace x>>av_log2(sizeof(..)) + 1 by x/sizeof(..). The +1 is > probably meant to emulate av_log2_ceil(sizeof(..)) in cases where ".." > is not a power of

Re: [libav-devel] [PATCH] lavu: use intrinsic for emms on systems lacking inline asm support.

On Wed, Jul 4, 2012 at 2:04 PM, Ronald S. Bultje wrote: > Hi, > > On Wed, Jul 4, 2012 at 12:36 PM, Måns Rullgård wrote: >> "Ronald S. Bultje" writes: >> >>> Hi, >>> >>> On Wed, Jul 4, 2012 at 10:15 AM, Måns Rullgård wrote: >>&

Re: [libav-devel] [PATCH] dct/fft-test: use a replacement getopt() if the system has none present.

Hi, On Tue, Jul 3, 2012 at 8:14 PM, Ronald S. Bultje wrote: > From: "Ronald S. Bultje" > > This allows compiling and running these tests on systems lacking a built- > in version of getopt(), such as MSVC. > --- > configure |2 ++ >

Re: [libav-devel] [PATCH] lavu: use intrinsic for emms on systems lacking inline asm support.

Hi, On Wed, Jul 4, 2012 at 12:36 PM, Måns Rullgård wrote: > "Ronald S. Bultje" writes: > >> Hi, >> >> On Wed, Jul 4, 2012 at 10:15 AM, Måns Rullgård wrote: >>> "Ronald S. Bultje" writes: >>> >>>>

Re: [libav-devel] [PATCH] vp8: Add ifdef guards around the sse2 loopfilter in the sse2slow branch too

Hi, On Wed, Jul 4, 2012 at 12:28 PM, Martin Storsjö wrote: > This was missed in the the previous commit in 70a1c800. > --- > libavcodec/x86/vp8dsp-init.c |2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/libavcodec/x86/vp8dsp-init.c b/libavcodec/x86/vp8dsp-init.c > index a0e8f9b..589

Re: [libav-devel] [PATCH] mxfdec: replace x>>av_log2(sizeof(..)) by x/sizeof(..).

Hi, On Wed, Jul 4, 2012 at 5:08 AM, Tomas Härdin wrote: > On Wed, 2012-07-04 at 12:14 +0100, Måns Rullgård wrote: >> Kostya Shishkov writes: >> >> > On Wed, Jul 04, 2012 at 11:09:57AM +0100, Måns Rullgård wrote: >> >> "Ronald S. Bultje"

Re: [libav-devel] [PATCH] x86/cpu: implement support for cpuid through intrinsics.

Hi, On Wed, Jul 4, 2012 at 10:35 AM, Måns Rullgård wrote: > "Ronald S. Bultje" writes: > >> From: "Ronald S. Bultje" >> >> --- >> configure |9 + >> libavutil/x86/cpu.c | 14 ++ >> 2 files chan

Re: [libav-devel] [PATCH] x86/cpu: implement support for xgetbv through intrinsics.

Hi, On Wed, Jul 4, 2012 at 10:31 AM, Måns Rullgård wrote: > "Ronald S. Bultje" writes: > >> From: "Ronald S. Bultje" >> >> --- >> configure |9 + >> libavutil/x86/cpu.c | 11 +++ >> 2 files chan

Re: [libav-devel] [PATCH] lavu: use intrinsic for emms on systems lacking inline asm support.

Hi, On Wed, Jul 4, 2012 at 10:15 AM, Måns Rullgård wrote: > "Ronald S. Bultje" writes: > >> From: "Ronald S. Bultje" >> >> --- >> configure|9 + >> libavutil/internal.h |6 +- >> 2 files changed,

[libav-devel] [PATCH] lavu: use intrinsic for emms on systems lacking inline asm support.

From: "Ronald S. Bultje" --- configure|9 + libavutil/internal.h |7 +-- 2 files changed, 14 insertions(+), 2 deletions(-) diff --git a/configure b/configure index 351d8a0..4fc20a2 100755 --- a/configure +++ b/configure @@ -1118,7 +1118,9 @@

[libav-devel] [PATCH] x86/cpu: implement support for cpuid through intrinsics.

From: "Ronald S. Bultje" --- configure |9 + libavutil/x86/cpu.c | 14 ++ 2 files changed, 23 insertions(+) diff --git a/configure b/configure index 35d4406..976b59d 100755 --- a/configure +++ b/configure @@ -1062,6 +1062,7 @@ HAVE_LIST=&q

[libav-devel] [PATCH] x86/cpu: implement support for xgetbv through intrinsics.

From: "Ronald S. Bultje" --- configure |9 + libavutil/x86/cpu.c | 11 +++ 2 files changed, 20 insertions(+) diff --git a/configure b/configure index 4fc20a2..35d4406 100755 --- a/configure +++ b/configure @@ -1092,6 +1092,7 @@ HAVE_LIST="

[libav-devel] [PATCH] lavu: use intrinsic for emms on systems lacking inline asm support.

From: "Ronald S. Bultje" --- configure|9 + libavutil/internal.h |6 +- 2 files changed, 14 insertions(+), 1 deletion(-) diff --git a/configure b/configure index 351d8a0..4fc20a2 100755 --- a/configure +++ b/configure @@ -1118,7 +1118,9 @@

Re: [libav-devel] [PATCH 2/2] vp8: loopfilter >=sse2 functions need aligned stack on x86-32.

Hi, On Wed, Jul 4, 2012 at 8:01 AM, Måns Rullgård wrote: > "Ronald S. Bultje" writes: > >> Hi, >> >> On Tue, Jul 3, 2012 at 9:11 PM, Ronald S. Bultje wrote: >>> +#if ARCH_X86_64 || HAVE_ALIGNED_STACK >> >> The ARCH_X86_64 here is not becau

[libav-devel] [PATCH] mxfdec: replace x>>av_log2(sizeof(..)) by x/sizeof(..).

From: "Ronald S. Bultje" Also replace x>>av_log2(sizeof(..)) + 1 by x/sizeof(..). The +1 is probably meant to emulate av_log2_ceil(sizeof(..)) in cases where ".." is not a power of two. --- libavformat/mxfdec.c |8 1 file changed, 4 insertions(+), 4

Re: [libav-devel] [PATCH 2/2] vp8: loopfilter >=sse2 functions need aligned stack on x86-32.

Hi, On Tue, Jul 3, 2012 at 9:11 PM, Ronald S. Bultje wrote: > +#if ARCH_X86_64 || HAVE_ALIGNED_STACK The ARCH_X86_64 here is not because that one has aligned stack (it does on win64 also; see e.g. [1]). Aligned or unaligned stack on win64 (or any 64bit platform for that matter) is

<    2   3   4   5   6   7   8   9   10   11   >