Signed-off-by: James Almer jamr...@gmail.com
---
configure | 5 +
libavutil/x86/cpu.h | 3 +++
2 files changed, 8 insertions(+)
diff --git a/configure b/configure
index 72cf831..96cbe5d 100755
--- a/configure
+++ b/configure
@@ -270,6 +270,7 @@ Optimization options (experts only
Based on x264 code
Signed-off-by: James Almer jamr...@gmail.com
---
libavutil/cpu.c | 6 ++
libavutil/cpu.h | 2 ++
libavutil/x86/cpu.c | 16 +++-
3 files changed, 19 insertions(+), 5 deletions(-)
diff --git a/libavutil/cpu.c b/libavutil/cpu.c
index 972e4eb..d651eb2
Based on x264 code
Signed-off-by: James Almer jamr...@gmail.com
---
configure | 5 +
libavutil/cpu.c | 3 +++
libavutil/cpu.h | 1 +
libavutil/x86/cpu.c | 5 -
libavutil/x86/cpu.h | 3 +++
5 files changed, 16 insertions(+), 1 deletion(-)
diff --git a/configure b
These are the missing instruction sets introduced with Haswell/Piledriver CPUs.
Last two patches are based on x264 detection code.
James Almer (3):
x86: add missing XOP checks and macros
x86: add detection for FMA3 instruction set
x86: add detection for Bit Manipulation Instruction sets
On 22/02/14 1:29 PM, Luca Barbato wrote:
On 22/02/14 06:53, James Almer wrote:
These are the missing instruction sets introduced with Haswell/Piledriver
CPUs.
Last two patches are based on x264 detection code.
The set doesn't look bad at all but I'm wondering about it's usage:
James
On 22/02/14 11:57 PM, Luca Barbato wrote:
On 23/02/14 00:52, Dave Yeo wrote:
HAVE_LIST has ARCH_EXT_LIST
ARCH_EXT_LIST has ARCH_EXT_LIST_X86
I'm wondering why it is broken for you since it should not.
lu
https://fate.libav.org/x86.os2.444/2014002516
Probably related.
On 23/02/14 11:20 AM, Janne Grunau wrote:
Do you plan to write assembly using any of these instructions? Having
the tests while not using the instructions is just an exercise in
completeness.
Janne
No, not for BMI1/2. I saw the TODO line as i was adding FMA3 so i
thought i might as well
Build only on x86_32 targets.
Signed-off-by: James Almer jamr...@gmail.com
---
libavcodec/x86/dcadsp.asm| 55 +---
libavcodec/x86/dcadsp_init.c | 44 +--
2 files changed, 69 insertions(+), 30 deletions(-)
diff --git
registers, which would benefit AMD users
unlike these AVX/FMA3 ymm ones. Thoughts?
James Almer (3):
x86/synth_filter: add synth_filter_fma3
x86/synth_filter: add synth_filter_sse
x86/synth_filter: add synth_filter_avx
libavcodec/x86/dcadsp.asm| 109
Sandy Bridge Win64:
180 cycles in ff_synth_filter_inner_sse2
150 cycles in ff_synth_filter_inner_avx
Also switch to a three operand format for some instructions to avoid
assembly errors with Yasm 1.1.0 or older.
Signed-off-by: James Almer jamr...@gmail.com
---
libavcodec/x86/dcadsp.asm| 76
Signed-off-by: James Almer jamr...@gmail.com
---
libavcodec/x86/dcadsp.asm| 28 +++-
libavcodec/x86/dcadsp_init.c | 4
2 files changed, 15 insertions(+), 17 deletions(-)
diff --git a/libavcodec/x86/dcadsp.asm b/libavcodec/x86/dcadsp.asm
index 0d7c86e..e1842ef
On 04/03/14 3:48 PM, Diego Biurrun wrote:
Some optimized functions reference optimized symbols, so the functions
must be explicitly disabled when those symbols are unavailable.
---
libavcodec/x86/dcadsp_init.c |4
1 file changed, 4 insertions(+)
diff --git
On 06/03/14 5:47 PM, Luca Barbato wrote:
On 06/03/14 21:34, Reinhard Tartler wrote:
Do we want this in release/10?
Yes.
The current stable version (x265 0.8) has X265_BUILD == 7. This change
would make libav 10 only support the development branch, and most users
and even distros usually
On 10/03/14 11:37 AM, Diego Biurrun wrote:
---
libavformat/apetag.c |6 --
libavformat/asfdec.c |8 +---
libavformat/avidec.c |4 ++--
libavformat/bink.c | 10 +++---
libavformat/cafdec.c |5 -
libavformat/crcenc.c
On 10/03/14 11:55 PM, Tim Walker wrote:
On 11 Mar 2014, at 03:38, James Almer jamr...@gmail.com wrote:
On 10/03/14 11:37 AM, Diego Biurrun wrote:
---
libavformat/apetag.c |6 --
libavformat/asfdec.c |8 +---
libavformat/avidec.c |4
On 13/08/13 11:49 PM, Luca Barbato wrote:
---
libavcodec/avcodec.h | 5 +
libavcodec/avpacket.c | 56
++
libavcodec/internal.h | 36
libavformat/avformat.h | 6 --
4 files changed, 97
On 14/03/14 7:56 AM, Christophe Gisquet wrote:
Hi,
2014-03-04 3:25 GMT+01:00 James Almer jamr...@gmail.com:
-INIT_XMM sse2
+%macro SETZERO 1
+%if cpuflag(sse2)
+pxor %1, %1
+%else
+xorps %1, %1, %1
+%endif
+%endmacro
+
+%macro SHUF 2
+%if cpuflag(sse2
On 14/03/14 8:02 AM, Christophe Gisquet wrote:
Hi,
2014-03-04 3:25 GMT+01:00 James Almer jamr...@gmail.com:
snip
Don't know fma3 but this is straightforward replacement of mul+add by
a mac instruction. If the avx code is ok, I don't see how this
wouldn't.
I just noticed i can replace
On 19/03/14 2:24 PM, Ben Avison wrote:
diff --git a/libavcodec/mlpdsp.h b/libavcodec/mlpdsp.h
index bd864d9..7b7640e 100644
--- a/libavcodec/mlpdsp.h
+++ b/libavcodec/mlpdsp.h
@@ -23,6 +23,7 @@
#define AVCODEC_MLPDSP_H
#include stdint.h
+#include mlp.h
void
Signed-off-by: James Almer jamr...@gmail.com
---
libavcodec/x86/dcadsp.asm| 22 ++
libavcodec/x86/dcadsp_init.c | 6 ++
2 files changed, 28 insertions(+)
diff --git a/libavcodec/x86/dcadsp.asm b/libavcodec/x86/dcadsp.asm
index 662cb96..59d96bf 100644
--- a/libavcodec
Sandy Bridge Win64:
180 cycles in ff_synth_filter_inner_sse2
150 cycles in ff_synth_filter_inner_avx
Also switch some instructions to a three operand format to avoid
assembly errors with Yasm 1.1.0 or older.
Signed-off-by: James Almer jamr...@gmail.com
---
libavcodec/x86/dcadsp.asm| 85
Build only on x86_32 targets.
Signed-off-by: James Almer jamr...@gmail.com
---
libavcodec/x86/dcadsp.asm| 55 +---
libavcodec/x86/dcadsp_init.c | 45 ++--
2 files changed, 70 insertions(+), 30 deletions(-)
diff --git
ymm ones. Thoughts?
James Almer (3):
x86/synth_filter: add synth_filter_sse
x86/synth_filter: add synth_filter_avx
x86/synth_filter: add synth_filter_fma3
libavcodec/x86/dcadsp.asm| 138 ---
libavcodec/x86/dcadsp_init.c | 55 +++--
2
On 24/03/14 11:28 AM, Diego Biurrun wrote:
On Thu, Mar 20, 2014 at 03:37:56PM -0300, James Almer wrote:
--- a/libavcodec/x86/dcadsp.asm
+++ b/libavcodec/x86/dcadsp.asm
@@ -230,6 +230,14 @@ DCA_LFE_FIR 1
mova m12, [ptr1 + j + mmsize]
%endif
+%if cpuflag(fma3)
+fmaddps
On 24/03/14 3:07 PM, Diego Biurrun wrote:
On Mon, Mar 24, 2014 at 02:59:08PM -0300, James Almer wrote:
On 24/03/14 11:28 AM, Diego Biurrun wrote:
On Thu, Mar 20, 2014 at 03:37:56PM -0300, James Almer wrote:
--- a/libavcodec/x86/dcadsp.asm
+++ b/libavcodec/x86/dcadsp.asm
@@ -230,6 +230,14
On 28/03/14 4:15 PM, Jason Garrett-Glaser wrote:
On Thu, Mar 20, 2014 at 11:37 AM, James Almer jamr...@gmail.com wrote:
Build only on x86_32 targets.
Signed-off-by: James Almer jamr...@gmail.com
---
libavcodec/x86/dcadsp.asm| 55
On 28/03/14 4:16 PM, Jason Garrett-Glaser wrote:
On Thu, Mar 20, 2014 at 11:37 AM, James Almer jamr...@gmail.com wrote:
Sandy Bridge Win64:
180 cycles in ff_synth_filter_inner_sse2
150 cycles in ff_synth_filter_inner_avx
Also switch some instructions to a three operand format to avoid
100755 tools/gen-rc
diff --git a/tools/gen-rc b/tools/gen-rc
new file mode 100755
index 000..269f2f6
--- /dev/null
+++ b/tools/gen-rc
@@ -0,0 +1,122 @@
+#!/bin/sh
+#
+# Copyright (c) 2012 James Almer
+# Copyright (c) 2013 Tiancheng Timothy Gu
+#
+# This file is part of Libav
This fixes compilation failures with --disable-fma3
Signed-off-by: James Almer jamr...@gmail.com
---
See https://fate.libav.org/x86_32-linux-suncc-nosse/20140405142549
libavcodec/x86/dcadsp.asm| 2 --
libavcodec/x86/dcadsp_init.c | 2 --
2 files changed, 4 deletions(-)
diff --git
On 05/04/14 5:20 PM, Diego Biurrun wrote:
On Sat, Apr 05, 2014 at 02:00:53PM -0300, James Almer wrote:
This fixes compilation failures with --disable-fma3
Signed-off-by: James Almer jamr...@gmail.com
---
See https://fate.libav.org/x86_32-linux-suncc-nosse/20140405142549
libavcodec/x86
On 31/05/14 4:23 AM, Anton Khirnov wrote:
From: Tudor Suciu tudor.su...@gmail.com
Signed-off-by: Anton Khirnov an...@khirnov.net
---
The webm official page does not mention that, but Google seems to claim those
are now offically supported. Nice to see they are keeping all the matroska
On 04/06/14 3:33 PM, Andrew Stone wrote:
By comparing versions of dictionaries, it's possible to detect if metadata
has changed.
---
libavutil/dict.c | 8
libavutil/dict.h | 10 ++
2 files changed, 18 insertions(+)
Maybe revision is a better name to avoid confusion with
On 24/06/14 11:26 AM, Luca Barbato wrote:
From: Pierre Edouard Lepere pierre-edouard.lep...@insa-rennes.fr
Signed-off-by: Luca Barbato lu_z...@gentoo.org
---
libavcodec/hevc.c | 19 -
libavcodec/hevcdsp.c | 5 ++
libavcodec/hevcdsp.h | 4 +-
On 24/06/14 11:26 AM, Luca Barbato wrote:
From: Pierre Edouard Lepere pierre-edouard.lep...@insa-rennes.fr
The functions only support x86_64.
Fixes from Hendrik Leppkes and James Almer
Signed-off-by: Luca Barbato lu_z...@gentoo.org
---
libavcodec/hevcdsp.c |6
On 25/06/14 3:44 PM, Luca Barbato wrote:
On 25/06/14 20:33, James Almer wrote:
On 24/06/14 11:26 AM, Luca Barbato wrote:
From: Pierre Edouard Lepere pierre-edouard.lep...@insa-rennes.fr
The functions only support x86_64.
Fixes from Hendrik Leppkes and James Almer
Signed-off-by: Luca
---
This applies cleanly after PATCH 14/15, and of course requires relevant changes
to
hevc_init.c
I think i got every function right, but in any case fixing any of them is a
single
line change.
In the end, out of 190 functions, only 44 were SSE4.
libavcodec/x86/hevc_mc.asm | 363
On 07/07/14 4:08 PM, Martin Storsjö wrote:
This silences a warning with gcc.
---
In my defense, clang didn't show this warning.
---
libavformat/mov.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavformat/mov.c b/libavformat/mov.c
index 4a2d265..853c3e5 100644
On 21/07/14 6:21 PM, Diego Biurrun wrote:
---
I'm slightly unsure about this one. MMX does not require 16-byte aligned
buffers, nor does PowerPC IIRC, but SSE and AltiVec do, so I believe my
solution is closer to the original intention. Please do correct me if I
am wrong...
configure
On 12/08/14 1:54 PM, Anton Khirnov wrote:
---
doc/APIchanges | 3 +++
libavutil/mem.c | 20
libavutil/mem.h | 10 ++
libavutil/version.h | 2 +-
4 files changed, 34 insertions(+), 1 deletion(-)
Shouldn't this be in avstring.h/c?
On 15/08/14 5:13 PM, Diego Biurrun wrote:
This avoids a number of redefinition warnings.
---
libavcodec/dxva2_internal.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/libavcodec/dxva2_internal.h b/libavcodec/dxva2_internal.h
index f35a076..d50c0ff 100644
---
On 31/08/14 4:24 PM, Luca Barbato wrote:
---
libavcodec/avcodec.h | 5 +
libavcodec/avpacket.c | 56
++
libavcodec/internal.h | 36
libavformat/avformat.h | 6 --
4 files changed, 97
This makes the addition of arch optimized functions easier.
Signed-off-by: James Almer jamr...@gmail.com
---
libavcodec/Makefile | 2 +-
libavcodec/jpeg2000dec.c | 72 +--
libavcodec/jpeg2000dsp.c | 98
This makes the addition of arch optimized functions easier.
Signed-off-by: James Almer jamr...@gmail.com
---
Now with proper names for the dsp functions, Irreversible MCT (ICT) and
Reversible MCT (RCT) respectively.
No other changes.
libavcodec/Makefile | 2 +-
libavcodec/jpeg2000dec.c
This silences warnings about passing arguments from incompatible pointer type
when targeting Windows Vista or newer.
Signed-off-by: James Almer jamr...@gmail.com
---
Only tested with Mingw-w64 v3.
Can someone test MSVC or ICL? I know the former defines the struct, so the
configure check should
Wrap the function calls in a similar fashion to how it's being done
with the critical section API.
Signed-off-by: James Almer jamr...@gmail.com
---
compat/w32pthreads.h | 64 +++-
1 file changed, 38 insertions(+), 26 deletions(-)
diff --git
This silences warnings about passing arguments from incompatible pointer type
when targeting Windows Vista or newer.
Signed-off-by: James Almer jamr...@gmail.com
---
Sent the wrong version earlier, my bad.
The request for testing i made before still stands.
compat/w32pthreads.h | 23
Wrap the function calls in a similar fashion to how it's being done
with the critical section API.
Signed-off-by: James Almer jamr...@gmail.com
---
compat/w32pthreads.h | 63 ++--
1 file changed, 37 insertions(+), 26 deletions(-)
diff --git
On 09/10/14 5:03 AM, Martin Storsjö wrote:
On Wed, 8 Oct 2014, James Almer wrote:
Wrap the function calls in a similar fashion to how it's being done
with the critical section API.
Signed-off-by: James Almer jamr...@gmail.com
---
compat/w32pthreads.h | 63
Wrap the function calls in a similar fashion to how it's being done
with the critical section API.
Signed-off-by: James Almer jamr...@gmail.com
---
compat/w32pthreads.h | 60 +---
1 file changed, 38 insertions(+), 22 deletions(-)
diff --git
This silences warnings about passing arguments from incompatible pointer type
when targeting Windows Vista or newer.
Signed-off-by: James Almer jamr...@gmail.com
---
compat/w32pthreads.h | 23 +--
configure| 2 ++
2 files changed, 15 insertions(+), 10 deletions
On 29/10/14 12:46 PM, Anton Khirnov wrote:
It is required by (at least) the ogg demuxer.
Mark the current semi-public apriv API for removal.
---
doc/APIchanges | 3 +++
libavcodec/Makefile| 12 ++
libavcodec/version.h | 4 ++--
On 24/11/14 8:05 PM, Vittorio Giovara wrote:
On Mon, Nov 24, 2014 at 5:58 PM, Henrik Gramner hen...@gramner.com wrote:
+mova m2, [r2+r1]
+mova m3, [r2+r1+mmsize]
+pxor m2, m6
+pxor m3, m6
pxor m2, m6, [r2+r1]
pxor m3, m6, [r2+r1+mmsize]
Avoids two moves in AVX, otherwise
On 27/11/14 2:43 PM, Reinhard Tartler wrote:
Hi,
it seems that tests/lena.pnm is not really redistributable. I'm
proposing to replace it with an image I've taken this summer. The
patch itself is too large to post it here, which is why I've uploaded
it to
On 27/11/14 3:03 PM, Reinhard Tartler wrote:
On Thu, Nov 27, 2014 at 12:59 PM, James Almer jamr...@gmail.com wrote:
On 27/11/14 2:43 PM, Reinhard Tartler wrote:
Hi,
it seems that tests/lena.pnm is not really redistributable. I'm
proposing to replace it with an image I've taken this summer
On 12/12/14 5:57 PM, Vittorio Giovara wrote:
From: James Almer jamr...@gmail.com
Signed-off-by: James Almer jamr...@gmail.com
Signed-off-by: Michael Niedermayer michae...@gmx.at
Signed-off-by: Vittorio Giovara vittorio.giov...@gmail.com
---
Changelog | 1 +
libavformat
On 16/12/14 11:58 AM, Vittorio Giovara wrote:
From: James Almer jamr...@gmail.com
Signed-off-by: Vittorio Giovara vittorio.giov...@gmail.com
---
Dropped the sign-offs since the file was modified.
Addressed Anton's and James' comment.
Vittorio
Changelog | 1
On 16/12/14 3:05 PM, Vittorio Giovara wrote:
On Tue, Dec 16, 2014 at 6:10 PM, James Almer jamr...@gmail.com wrote:
On 16/12/14 11:58 AM, Vittorio Giovara wrote:
From: James Almer jamr...@gmail.com
Signed-off-by: Vittorio Giovara vittorio.giov...@gmail.com
---
Dropped the sign-offs since
On 16/03/15 5:00 AM, Niels Möller wrote:
James Almer jamr...@gmail.com writes:
Valgrind is complaining about this code (Conditional jump or move
depends on uninitialised
value error), as seen here
https://fate.libav.org/x86_64-linux-gcc-valgrind/20150316044429
Zero initializing
On 13/03/15 12:24 PM, Luca Barbato wrote:
On 13/03/15 16:17, Diego Biurrun wrote:
From: Niels Möller ni...@lysator.liu.se
---
Changes since last round:
- XLL disabled by default.
- Return error on too many downmix coefficients
This has survived Oracle, so it's good to go IMO and will
On 12/03/15 1:14 PM, Luca Barbato wrote:
On 12/03/15 13:54, Derek Buitenhuis wrote:
On 3/11/2015 7:07 PM, Luca Barbato wrote:
On top of it, I wasn't aware it exists
The feature has been complete and in FFmpeg master for 1.5+ years...
I do not think pretend FFmpeg doesn't exist and never
On 28/03/15 2:52 PM, Justin Ruggles wrote:
On 03/28/2015 01:42 PM, Himangi Saraogi wrote:
---
libavcodec/tiffenc.c | 11 ++-
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/libavcodec/tiffenc.c b/libavcodec/tiffenc.c
index 169360f..46e4207 100644
---
On 22/03/15 12:49 PM, Vittorio Giovara wrote:
+// AAN IDCT
If this isn't already in the tree somewhere and it's generic enough that it can
be reused,
then it should be shared like faanidct and added to idctdsp.
And if it's HQ/HQA specific, it still could be split into a new hqdsp context
for
On 23/03/15 10:23 AM, Luca Barbato wrote:
On 23/03/15 12:45, Hendrik Leppkes wrote:
---
configure | 4 +
libavcodec/Makefile| 1 +
libavcodec/allcodecs.c | 1 +
libavcodec/libdcadec.c | 197
+
4 files changed, 203
On 02/05/15 5:22 AM, Luca Barbato wrote:
On 02/05/15 06:23, Anton Khirnov wrote:
Quoting Vittorio Giovara (2015-05-02 01:17:08)
The enum is used by lavc, lavf and lavfi, and it is referenced by lavu,
so it sementically belongs to lavu more than any other.
This change allows to drop an
id should be an integer, not a string. It is also optional, so use
contentType instead which is the proper attribute for these values.
This fixes an MPD validation error.
Signed-off-by: James Almer jamr...@gmail.com
---
libavformat/dashenc.c | 4 ++--
1 file changed, 2 insertions(+), 2
On 12/05/15 6:00 PM, Luca Barbato wrote:
On 11/05/15 17:25, Derek Buitenhuis wrote:
From: Gopu Govindaswamy g...@multicorewareinc.com
The x265pic.bitDepth is set on encode_frame while I assume that this
information should be used at init now.
I'm not sure how recent is this api version,
Signed-off-by: James Almer jamr...@gmail.com
---
libavutil/cpu.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/libavutil/cpu.c b/libavutil/cpu.c
index 354d21e..4e8ef61 100644
--- a/libavutil/cpu.c
+++ b/libavutil/cpu.c
@@ -90,8 +90,7 @@ int av_parse_cpu_flags(const char
On 19/05/15 7:49 AM, Vittorio Giovara wrote:
---
Another set of eyes for the pixel format mapping would be welcome.
Vittorio
libavcodec/libvpx.c| 26 ++
libavcodec/libvpx.h| 2 ++
libavcodec/libvpxdec.c | 4 ++--
3 files changed, 30 insertions(+), 2
The first check is done without the AVIOContext, so alloc it only if said check
succeeds
Signed-off-by: James Almer jamr...@gmail.com
---
libavformat/mpjpegdec.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/libavformat/mpjpegdec.c b/libavformat/mpjpegdec.c
index
On 02/06/15 8:09 AM, Vittorio Giovara wrote:
+/* Alpha compression function */
+static void compress_alpha(uint8_t *dst, ptrdiff_t stride, const uint8_t
*block)
+{
+int i, j;
+int dist, bias, dist4, dist2, bits, mask;
+int mn, mx;
+
+/* Find min/max color */
+mn = mx
On 11/06/15 11:56 AM, Luca Barbato wrote:
@@ -321,8 +321,12 @@ static av_cold int vpx_init(AVCodecContext *avctx,
/* 0-3: For non-zero values the encoder increasingly optimizes for
reduced
complexity playback on low powered devices at the expense of encode
quality. */
-
On 11/06/15 11:56 AM, Luca Barbato wrote:
---
I tied the supported formats to the ABI version.
configure | 12 ++--
libavcodec/libvpx.c| 8 ++--
libavcodec/libvpxenc.c | 6 +-
3 files changed, 17 insertions(+), 9 deletions(-)
diff --git a/configure
On 03/06/15 11:29 AM, Luca Barbato wrote:
On 31/05/15 17:01, Luca Barbato wrote:
On 27/05/15 22:25, James Almer wrote:
Yes, that plus a considerable amount of ifdeffery in the code. It
will be ugly, but i also think it's worth keeping compatibility with
at least 1.3.0
1.3.0 explodes
This was a regression introduced with d8ffb2055f0e0fcb5d025bab72eb19c2a886c125.
Signed-off-by: James Almer jamr...@gmail.com
---
configure | 2 ++
1 file changed, 2 insertions(+)
diff --git a/configure b/configure
index cdc5a8d..a29cd38 100755
--- a/configure
+++ b/configure
@@ -2216,6 +2216,8
On 27/05/15 1:15 PM, Vittorio Giovara wrote:
Bump the minimum libvpx version to 1.4.0 so that all pixel
formats are present. Add new VP9 profiles.
Signed-off-by: Vittorio Giovara vittorio.giov...@gmail.com
---
Modified as requested.
Vittorio
configure | 23
On 27/05/15 3:24 PM, James Almer wrote:
On 27/05/15 1:15 PM, Vittorio Giovara wrote:
Bump the minimum libvpx version to 1.4.0 so that all pixel
formats are present. Add new VP9 profiles.
Signed-off-by: Vittorio Giovara vittorio.giov...@gmail.com
---
Modified as requested.
Vittorio
On 27/05/15 5:04 PM, Martin Storsjö wrote:
On Wed, 27 May 2015, Vittorio Giovara wrote:
Bump the minimum libvpx version to 1.4.0 so that all pixel
formats are present. Add new VP9 profiles.
Sorry to be a bit late to the party, but how bad would it be to keep compat
with older versions?
On 26/05/15 2:29 PM, James Almer wrote:
Signed-off-by: James Almer jamr...@gmail.com
---
No changes from last revision.
doc/APIchanges | 3 +++
libavutil/cpu.c | 3 +++
libavutil/cpu.h | 1 +
libavutil/version.h | 4 ++--
libavutil/x86/cpu.c | 17 ++---
5
On 28/05/15 9:11 AM, Vittorio Giovara wrote:
On Wed, May 27, 2015 at 7:24 PM, James Almer jamr...@gmail.com wrote:
As Luca said, require_pkg_config. If you use require, vpx = 1.4.0 is just
used as a name
to report a failure. It does not check for that version.
I swear I couldn't get
This was a regression introduced with d8ffb2055f0e0fcb5d025bab72eb19c2a886c125.
Signed-off-by: James Almer jamr...@gmail.com
---
configure | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/configure b/configure
index 30d6f18..2458adb 100755
--- a/configure
+++ b/configure
On 02/06/15 4:27 AM, Martin Storsjö wrote:
On Mon, 1 Jun 2015, James Almer wrote:
On 01/06/15 7:54 AM, Martin Storsjö wrote:
This fixes dxva2 detection (i.e. correctly realizes that it isn't
available) for WinRT, where dxva2api.h does exist, but these definitions
are omitted (when targeting
On 25/05/15 2:31 AM, Steve Lhomme wrote:
On Sun, May 24, 2015 at 1:13 PM, Luca Barbato lu_z...@gentoo.org wrote:
On 24/05/15 07:59, Steve Lhomme wrote:
Any update on this patch ?
If it works for you I'll merge it Monday. I do not have mean to test it
directly I guess.
Yes, it works.
On 25/05/15 11:49 AM, Steve Lhomme wrote:
---
configure | 5 -
1 file changed, 5 deletions(-)
diff --git a/configure b/configure
index 18280b9..a9ecad1 100755
--- a/configure
+++ b/configure
@@ -1555,7 +1555,6 @@ HAVE_LIST=
atomics_native
dos_paths
d3d11_cobj
-
On 25/05/15 8:24 PM, James Almer wrote:
On 25/05/15 11:49 AM, Steve Lhomme wrote:
---
configure | 5 -
1 file changed, 5 deletions(-)
diff --git a/configure b/configure
index 18280b9..a9ecad1 100755
--- a/configure
+++ b/configure
@@ -1555,7 +1555,6 @@ HAVE_LIST
Signed-off-by: James Almer jamr...@gmail.com
---
Updated with a new libavutil version after the d3d11 patch.
doc/APIchanges | 3 +++
libavutil/cpu.c | 3 +++
libavutil/cpu.h | 1 +
libavutil/version.h | 4 ++--
libavutil/x86/cpu.c | 17 ++---
5 files changed, 23
Signed-off-by: James Almer jamr...@gmail.com
---
The FMA4 functions from libavresample's audio_mix need to be handled
differently. Disabling them if avxslow is true is pointless since no
CPU out there currently has FMA4 and a fast float execution unit. So
I'm thinking about duplicating them
Signed-off-by: James Almer jamr...@gmail.com
---
doc/APIchanges | 3 +++
libavutil/cpu.c | 3 +++
libavutil/cpu.h | 1 +
libavutil/version.h | 4 ++--
libavutil/x86/cpu.c | 17 ++---
5 files changed, 23 insertions(+), 5 deletions(-)
diff --git a/doc/APIchanges b/doc
Signed-off-by: James Almer jamr...@gmail.com
---
libavcodec/x86/dcadsp_init.c | 4 ++--
libavcodec/x86/dct_init.c | 2 +-
libavcodec/x86/fft_init.c | 2 +-
libavfilter/x86/af_volume_init.c | 2 +-
libavresample/x86/audio_convert_init.c | 4
Signed-off-by: James Almer jamr...@gmail.com
---
No changes from last revision.
doc/APIchanges | 3 +++
libavutil/cpu.c | 3 +++
libavutil/cpu.h | 1 +
libavutil/version.h | 4 ++--
libavutil/x86/cpu.c | 17 ++---
5 files changed, 23 insertions(+), 5 deletions
Signed-off-by: James Almer jamr...@gmail.com
---
libavutil/cpu_internal.h | 12
libavutil/x86/cpu.h | 18 ++
2 files changed, 30 insertions(+)
diff --git a/libavutil/cpu_internal.h b/libavutil/cpu_internal.h
index 3bfe8a8..2e9b44b 100644
--- a/libavutil
On 19/08/15 4:43 PM, Anton Khirnov wrote:
---
libavcodec/hevc.c | 6 +-
libavcodec/hevc.h | 2 +-
libavcodec/hevcdsp.c | 24 +-
libavcodec/hevcdsp.h | 5 +-
libavcodec/hevcdsp_template.c | 8 +-
libavcodec/x86/Makefile | 3 +-
On 19/08/15 8:23 PM, Ronald S. Bultje wrote:
Hi,
On Wed, Aug 19, 2015 at 6:34 PM, James Almer jamr...@gmail.com wrote:
On 19/08/15 4:43 PM, Anton Khirnov wrote:
---
libavcodec/hevc.c | 6 +-
libavcodec/hevc.h | 2 +-
libavcodec/hevcdsp.c | 24
On 21/08/15 8:43 PM, Luca Barbato wrote:
Broken in f9ab4fe1f7c1e9d410ca5ee2c9ff8d2892aad068
---
Sorry.
libavcodec/h264_sei.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/h264_sei.c b/libavcodec/h264_sei.c
index 361d4de..ddf1b6f 100644
---
On 21/08/15 4:19 AM, Anton Khirnov wrote:
+
+add dstq, dststrideq
+add srcq, srcstrideq
+
+%assign i (i + 1)
+%endrep
+
+dec heightq
This and every other case should be heightd. There's no guarantee the high bits
will be zero
on every x86_64 target.
This is the
Module: libav
Branch: master
Commit: f9ab4fe1f7c1e9d410ca5ee2c9ff8d2892aad068
Author:John Högberg john.hogberg at ericsson.com
Committer: Luca Barbato lu_zero at gentoo.org
Date: Fri Aug 7 19:30:38 2015 +
h264: Discard currently unsupported registered sei
Signed-off-by:
On 21/08/15 4:19 AM, Anton Khirnov wrote:
+%macro PUT_WEIGHTED_PRED 3
+%if %1
+cglobal hevc_put_weighted_pred_avg_ %+ %2 %+ _ %+ %3, 11, 11, 8, denom,
weight0, weight1, offset0, offset1, dst, dststride, src0, src1, srcstride,
height
+%else
+cglobal hevc_put_weighted_pred_ %+ %2 %+ _ %+
On 22/08/15 1:16 PM, Anton Khirnov wrote:
+%macro QPEL_8 2
+%if %2
+%define postfixv
+%define mvfrac myq
Same here and below the else, rename this to mvfracq and add a mvfracd.
+%define pixstride srcstrideq
+%define pixstride3 sstride3q
+%define src_m3
On 23/08/15 3:27 PM, Anton Khirnov wrote:
Quoting James Almer (2015-08-22 23:58:41)
On 22/08/15 1:16 PM, Anton Khirnov wrote:
+%macro QPEL_8 2
+%if %2
+%define postfixv
+%define mvfrac myq
Same here and below the else, rename this to mvfracq and add a mvfracd
On 20/08/15 3:35 AM, Anton Khirnov wrote:
Quoting James Almer (2015-08-20 00:34:58)
On 19/08/15 4:43 PM, Anton Khirnov wrote:
---
libavcodec/hevc.c | 6 +-
libavcodec/hevc.h | 2 +-
libavcodec/hevcdsp.c | 24 +-
libavcodec/hevcdsp.h | 5
Signed-off-by: James Almer jamr...@gmail.com
---
libavformat/rtmpproto.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavformat/rtmpproto.c b/libavformat/rtmpproto.c
index 1db7495..ec4b0e7 100644
--- a/libavformat/rtmpproto.c
+++ b/libavformat/rtmpproto.c
@@ -971,7 +971,7
1 - 100 of 362 matches
Mail list logo
