---
libavcodec/Makefile | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/libavcodec/Makefile b/libavcodec/Makefile
index d586dbd..2b1e221 100644
--- a/libavcodec/Makefile
+++ b/libavcodec/Makefile
@@ -762,7 +762,8 @@ OBJS-$(CONFIG_AAC_ADTSTOASC_BSF) +=
---
This is not meant to be pushed, much less as-is.
mov.c is huge and it could probably use some splitting. What the
best way to split it might be, I have no idea.
This also fixes standalone compilation of the CAF demuxer, for
completely nonobvious reasons.
libavformat/Makefile | 3 +--
The rtmpcrypt protocol requires it.
---
libavformat/Makefile | 14 +-
libavformat/rtmpdigest.c | 67
libavformat/rtmpproto.c | 36 --
3 files changed, 74 insertions(+), 43 deletions(-)
create mode 100644
On Wed, Mar 15, 2017 at 01:40:03PM +0100, Luca Barbato wrote:
> --- a/configure
> +++ b/configure
> @@ -4836,8 +4838,8 @@ check_header soundcard.h
>
> -enabled jack_indev && check_lib jack jack/jack.h jack_client_open -ljack &&
> -check_func jack_port_get_latency_range -ljack
> +enabled
On Tue, Mar 14, 2017 at 11:42:45PM +0100, Luca Barbato wrote:
> --- a/configure
> +++ b/configure
> @@ -1636,7 +1638,6 @@ SYSTEM_FUNCS="
> inet_aton
> isatty
> -jack_port_get_latency_range
This is incorrect or incomplete. There are matching ifdefs in
libavdevice/jack.c.
Diego
On Wed, Mar 15, 2017 at 12:31:18PM +0100, Luca Barbato wrote:
> On 15/03/2017 11:36, Diego Biurrun wrote:
> > On Tue, Mar 14, 2017 at 11:42:45PM +0100, Luca Barbato wrote:
> >> ---
> >> configure | 8
> >> 1 file changed, 4 insertions(+), 4 del
On Tue, Mar 14, 2017 at 11:42:45PM +0100, Luca Barbato wrote:
> ---
> configure | 8
> 1 file changed, 4 insertions(+), 4 deletions(-)
Why?
Diego
___
libav-devel mailing list
libav-devel@libav.org
On Fri, Mar 10, 2017 at 03:17:36PM +0100, Diego Biurrun wrote:
> Also do some small cosmetic changes: Drop pointless _MMX suffix from ABSD2
> macro name, drop pointless check for MMX support, we always assume MMX is
> available in our SIMD code, fix spelling.
> ---
> libavcodec/x
On Tue, Mar 14, 2017 at 01:38:00AM +, Mark Thompson wrote:
> --- /dev/null
> +++ b/libavcodec/vp9_superframe_bsf.c
> @@ -0,0 +1,206 @@
> +/*
> + * Vp9 invisible (alt-ref) frame to superframe merge bitstream filter
nit: VP9
Diego
___
libav-devel
On Sat, Mar 11, 2017 at 11:33:31AM +0100, Luca Barbato wrote:
> On 09/06/2016 17:12, Diego Biurrun wrote:
> > From: Alexandra Hájková <alexan...@khirnov.net>
> >
> > ---
> > libavcodec/vorbis_parser.c | 32 +++---
>
On Sun, Mar 12, 2017 at 05:27:48PM +0100, Luca Barbato wrote:
> The field is expressed in 6 bits so up to 63 bits amplitudes could be
> encoded.
>
> Reported and debugged by Uoti Urpala.
You may want to add his email address here.
possibly OK
Diego
Also do some small cosmetic changes: Drop pointless _MMX suffix from ABSD2
macro name, drop pointless check for MMX support, we always assume MMX is
available in our SIMD code, fix spelling.
---
libavcodec/x86/audiodsp.asm | 5 +
libavutil/x86/x86util.asm | 52
---
libavcodec/x86/fft.asm | 94 +++---
1 file changed, 43 insertions(+), 51 deletions(-)
diff --git a/libavcodec/x86/fft.asm b/libavcodec/x86/fft.asm
index 2c74381..63e92f7 100644
--- a/libavcodec/x86/fft.asm
+++ b/libavcodec/x86/fft.asm
@@ -190,6
---
libavcodec/x86/vp8dsp.asm | 143 +-
1 file changed, 66 insertions(+), 77 deletions(-)
diff --git a/libavcodec/x86/vp8dsp.asm b/libavcodec/x86/vp8dsp.asm
index b0f6b83..4270cdd 100644
--- a/libavcodec/x86/vp8dsp.asm
+++ b/libavcodec/x86/vp8dsp.asm
---
libavcodec/x86/h264_idct.asm | 52
1 file changed, 28 insertions(+), 24 deletions(-)
diff --git a/libavcodec/x86/h264_idct.asm b/libavcodec/x86/h264_idct.asm
index eb99476..73631ff 100644
--- a/libavcodec/x86/h264_idct.asm
+++
On Tue, Mar 07, 2017 at 06:39:35AM +, Konda Raju wrote:
> From 45fe0b3a56e82ad745b4cd81f4ab206c5c2983b8 Mon Sep 17 00:00:00 2001
> From: Konda Raju
> Date: Tue, 7 Mar 2017 12:02:14 +0530
> Subject: [PATCH] add initial QP value options
>
> --- a/libavcodec/nvenc_h264.c
> +++
On Mon, Mar 06, 2017 at 08:21:19PM +0100, Luca Barbato wrote:
> Recent version of MacOS provide more posix API that the previous
> versions and recent toolchains do not fail the check even if
> they are supposed to target an older version without this flag.
configure: Add -no_weak_imports to
We should only check for clock_gettime() if _POSIX_MONOTONIC_CLOCK is
available and do a full link check for clock_gettime() in all cases.
Bug-Id: 1033
---
Bug 1033 is on OS X, I have no way to test this ..
configure | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git
NASM is more actively maintained and permits generating dependency information
as a sideeffect of assembling, thus cutting build times in half.
---
Added a note to the changelog on how to restore old behavior.
.travis.yml | 4 ++--
Changelog| 2 ++
configure|
On Mon, Mar 06, 2017 at 11:34:20AM +0100, wm4 wrote:
> --- a/libavcodec/avcodec.h
> +++ b/libavcodec/avcodec.h
> @@ -161,7 +168,22 @@
> * and will result in arbitrary behavior.
> *
> * Some codecs might require using the new API; using the old API will return
> - * an error when calling it.
On Thu, Mar 02, 2017 at 02:57:36PM +0100, Diego Biurrun wrote:
> SDL adds some "special" CFLAGS that interfere with building normal
> binaries. Capture those CFLAGS separately and avoid adding them to
> the general CFLAGS.
> ---
> configure | 8 +++-
> 1 file
On Thu, Mar 02, 2017 at 01:24:10PM +0100, Diego Biurrun wrote:
> ---
>
> Moved the command line selection before changing the default.
>
> configure | 30 ++
> 1 file changed, 22 insertions(+), 8 deletio
On Mon, Mar 06, 2017 at 10:38:12AM +0100, wm4 wrote:
> --- a/libavcodec/avcodec.h
> +++ b/libavcodec/avcodec.h
> @@ -161,7 +168,20 @@
> + * A codec is not allowed to return EAGAIN for both sending and receiving.
> This
> + * would be an invalid state, which could put the codec user into an
On Sat, Mar 04, 2017 at 11:57:52PM +, Mark Thompson wrote:
> --- a/libavfilter/Makefile
> +++ b/libavfilter/Makefile
> @@ -77,6 +77,7 @@ OBJS-$(CONFIG_NULL_FILTER) += vf_null.o
> OBJS-$(CONFIG_OPENCL_PROGRAM_FILTER) += vf_opencl_program.o opencl.o
>
On Sat, Mar 04, 2017 at 11:57:47PM +, Mark Thompson wrote:
> --- a/configure
> +++ b/configure
> @@ -1272,6 +1273,7 @@ HWACCEL_LIBRARY_LIST="
> mmal
> nvenc
> omx
> +opencl
> vaapi
> vda
> vdpau
Does opencl have a compatible license? Which one?
Diego
On Sat, Mar 04, 2017 at 11:57:41PM +, Mark Thompson wrote:
> --- a/libavfilter/vf_hwmap.c
> +++ b/libavfilter/vf_hwmap.c
> @@ -59,30 +61,61 @@ static int hwmap_config_output(AVFilterLink *outlink)
> -if (err < 0)
> +if (err < 0) {
> +av_log(avctx,
On Sat, Mar 04, 2017 at 11:57:39PM +, Mark Thompson wrote:
> Some frames contexts are not usable without additional format-specific
"frame context" or "frame's contexts"
> state in hwctx. This adds new functions map_frames_from and
> map_frames_to to set this up appropriately when deriving
On Wed, Mar 01, 2017 at 11:14:56PM +0100, Diego Biurrun wrote:
> Explicitly use "elf32" instead of "elf" and set the values in one place.
> Also set objformat to "win64" on Windows x86_64, which is required for nasm.
> ---
>
> Switched "elf&quo
On Wed, Mar 01, 2017 at 10:07:26PM +0100, Diego Biurrun wrote:
> ---
> libavcodec/x86/dct32.asm | 6 ++
> libavcodec/x86/fft.asm | 3 +--
> libavcodec/x86/imdct36.asm | 3 +--
> libavcodec/x86/rv40dsp.asm | 3 +--
> 4 files changed, 5 insertions(+), 10 deletions(
SDL adds some "special" CFLAGS that interfere with building normal
binaries. Capture those CFLAGS separately and avoid adding them to
the general CFLAGS.
---
configure | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/configure b/configure
index 4eab439..7035f11 100755
NASM is more actively maintained and permits generating dependency information
as a sideeffect of assembling, thus cutting build times in half.
---
Moved switching the default after command line assembler selection.
.travis.yml | 4 ++--
configure| 6 +++---
---
Moved the command line selection before changing the default.
configure | 30 ++
1 file changed, 22 insertions(+), 8 deletions(-)
diff --git a/configure b/configure
index a58ef1c..e2823d1 100755
--- a/configure
+++ b/configure
@@ -268,6 +268,7 @@ Toolchain
On Thu, Mar 02, 2017 at 11:02:52AM +0100, Luca Barbato wrote:
> On 02/03/2017 09:41, Diego Biurrun wrote:
> > On Thu, Mar 02, 2017 at 09:20:45AM +0100, Luca Barbato wrote:
> >> On 01/03/2017 22:07, Diego Biurrun wrote:
> >>> NASM is more actively maintained and
On Thu, Mar 02, 2017 at 09:20:45AM +0100, Luca Barbato wrote:
> On 01/03/2017 22:07, Diego Biurrun wrote:
> > NASM is more actively maintained and permits generating dependency
> > information
> > as a sideeffect of assembling, thus cutting build times in half.
>
On Wed, Mar 01, 2017 at 07:52:37PM -0500, Vittorio Giovara wrote:
> ---
> This should partially alleviate the breakage.
> Vittorio
Partially as in ..
Diego
___
libav-devel mailing list
libav-devel@libav.org
Explicitly use "elf32" instead of "elf" and set the values in one place.
Also set objformat to "win64" on Windows x86_64, which is required for nasm.
---
Switched "elf" to "elf32", fixed a test in the arm section to check for
both elf variants.
configure | 19 +--
1 file
Also set objformat to "win64" on Windows x86_64.
---
configure | 16 +++-
1 file changed, 7 insertions(+), 9 deletions(-)
diff --git a/configure b/configure
index 7e788be..b92dd5b 100755
--- a/configure
+++ b/configure
@@ -2563,7 +2563,6 @@ cc_default="gcc"
host_cc_default="gcc"
This iteration adds fallback from nasm to yasm and a few preliminaries.
[PATCH 1/4] x86: Merge align directives into SECTION_RODATA
[PATCH 2/4] build: Simplify object format handling for nasm/yasm
[PATCH 3/4] build: Prefer NASM assembler over YASM
[PATCH 4/4] build: Make x86 assembler
NASM is more actively maintained and permits generating dependency information
as a sideeffect of assembling, thus cutting build times in half.
---
.travis.yml | 4 ++--
configure| 12 ++--
doc/optimization.txt | 8
doc/platform.texi| 12 ++--
4
---
libavcodec/x86/dct32.asm | 6 ++
libavcodec/x86/fft.asm | 3 +--
libavcodec/x86/imdct36.asm | 3 +--
libavcodec/x86/rv40dsp.asm | 3 +--
4 files changed, 5 insertions(+), 10 deletions(-)
diff --git a/libavcodec/x86/dct32.asm b/libavcodec/x86/dct32.asm
index 2c4c32e..cfd5f52 100644
---
Now falls back from nasm to yasm.
configure | 26 --
1 file changed, 20 insertions(+), 6 deletions(-)
diff --git a/configure b/configure
index 8516315..a61e978 100755
--- a/configure
+++ b/configure
@@ -268,6 +268,7 @@ Toolchain options:
--objcc=OCC
libavutil uses pthreads in the buffer code (abstracted through a header).
---
configure | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configure b/configure
index cdeae9a..7e788be 100755
--- a/configure
+++ b/configure
@@ -2524,7 +2524,7 @@ avdevice_extralibs="libm_extralibs"
On Tue, Feb 28, 2017 at 09:56:21PM +0100, Diego Biurrun wrote:
> ---
>
> Simplified version, according to Vittorio's suggestion.
>
> tests/Makefile | 1 +
> tests/fate-run.sh| 4
> tests/fate/build.mak | 17 +
> 3 files changed, 22 inse
On Tue, Feb 28, 2017 at 10:12:51PM +0100, Diego Biurrun wrote:
> ---
> Makefile | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
OKed by Luca on IRC.
Diego
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.or
On Wed, Mar 01, 2017 at 08:41:43AM +0100, Luca Barbato wrote:
> On 28/02/2017 22:12, Diego Biurrun wrote:
> > ---
> > avbuild/library.mak | 2 ++
> > 1 file changed, 2 insertions(+)
> >
> > diff --git a/avbuild/library.mak b/avbuild/library.mak
> > index e5
---
Makefile | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/Makefile b/Makefile
index a3cd084..03b1a1f 100644
--- a/Makefile
+++ b/Makefile
@@ -179,6 +179,7 @@ distclean::
config:
$(SRC_PATH)/configure $(value LIBAV_CONFIGURATION)
+build: all alltools
---
avbuild/library.mak | 2 ++
1 file changed, 2 insertions(+)
diff --git a/avbuild/library.mak b/avbuild/library.mak
index e5f6d7d..0ee0be9 100644
--- a/avbuild/library.mak
+++ b/avbuild/library.mak
@@ -1,6 +1,8 @@
include $(SRC_PATH)/avbuild/common.mak
+ifeq (,$(filter
---
Simplified version, according to Vittorio's suggestion.
tests/Makefile | 1 +
tests/fate-run.sh| 4
tests/fate/build.mak | 17 +
3 files changed, 22 insertions(+)
create mode 100644 tests/fate/build.mak
diff --git a/tests/Makefile b/tests/Makefile
index
On Tue, Feb 28, 2017 at 08:06:08PM +0100, Diego Biurrun wrote:
> ---
>
> Note that this feature eliminates fallback from yasm to nasm (or
> from nasm to yasm after nasm has become the preferred default).
> Seems like an acceptable tradeoff to me.
This was not deemed to be an acce
---
Makefile | 6 +++---
configure | 8
2 files changed, 11 insertions(+), 3 deletions(-)
diff --git a/Makefile b/Makefile
index 8313ca9..f9c74c1 100644
--- a/Makefile
+++ b/Makefile
@@ -30,7 +30,7 @@ CCFLAGS = $(CPPFLAGS) $(CFLAGS)
OBJCFLAGS += $(EOBJCFLAGS)
OBJCCFLAGS =
---
Note that this feature eliminates fallback from yasm to nasm (or
from nasm to yasm after nasm has become the preferred default).
Seems like an acceptable tradeoff to me.
configure | 18 --
1 file changed, 12 insertions(+), 6 deletions(-)
diff --git a/configure b/configure
NASM is more actively maintained and permits generating dependency information
as a sideeffect of assembling, thus cutting build times in half.
---
.travis.yml | 4 ++--
configure| 14 +++---
doc/optimization.txt | 8
doc/platform.texi| 12 ++--
This iteration settles for "x86asm" as the value for names of things
related to the NASM-style x86 assembler, which seems to be the
consensus value.
I have also added a commit that allows selecting the assembler on
the configure command line, similar to what we offer for cc etc.
Note that this
Also drop stray duplicate OBJCC config.mak entry.
---
configure | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/configure b/configure
index 9f5f009..2adaa23 100755
--- a/configure
+++ b/configure
@@ -2960,7 +2960,7 @@ fi
ar_default="${cross_prefix}${ar_default}"
This fixes many warnings of the sort
warning: label alone on a line without a colon might be in error
---
libavcodec/x86/hevc_add_res.asm | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/libavcodec/x86/hevc_add_res.asm b/libavcodec/x86/hevc_add_res.asm
index
On Wed, Feb 22, 2017 at 05:55:14PM +0100, Diego Biurrun wrote:
> On Mon, Feb 20, 2017 at 11:19:38PM +0200, Martin Storsjö wrote:
> > On Fri, 17 Feb 2017, Diego Biurrun wrote:
> >
> > >None of them are specific to the YASM assembler.
> > >---
> > >
>
---
Updated to match the change to "x86-asm" in the previous patch.
Makefile | 6 +++---
configure | 8
2 files changed, 11 insertions(+), 3 deletions(-)
diff --git a/Makefile b/Makefile
index a049afd..d1b0374 100644
--- a/Makefile
+++ b/Makefile
@@ -30,7 +30,7 @@ CCFLAGS =
On Mon, Feb 27, 2017 at 03:54:11PM +0100, Diego Biurrun wrote:
> Previously, all link-time dependencies were added for all libraries,
> resulting in bogus link-time dependencies since not all dependencies
> are shared across libraries. Also, in some cases like libavutil, not
> all
Previously, all link-time dependencies were added for all libraries,
resulting in bogus link-time dependencies since not all dependencies
are shared across libraries. Also, in some cases like libavutil, not
all dependencies were taken into account, resulting in some cases of
underlinking.
To
On Mon, Feb 27, 2017 at 02:11:49PM +, Konda Raju wrote:
> From 7cb5b9d551dc0755630e8f308542d6e325169525 Mon Sep 17 00:00:00 2001
> From: Raju Konda
Konda Raju or Raju Konda? This is a tad confusing..
> Date: Wed, 22 Feb 2017 19:20:53 +0530
> Subject: [PATCH 1/2] add
On Mon, Feb 27, 2017 at 01:46:14PM +0100, Diego Biurrun wrote:
> The codec used in those files is WMV3/WMV9, not WMV2/WMV8.
> ---
>
> Better names for the individual tests.
>
> tests/fate/microsoft.mak | 12 ++--
> tests/ref/fate/{wmv
The codec used in those files is WMV3/WMV9, not WMV2/WMV8.
---
Better names for the individual tests.
tests/fate/microsoft.mak | 12 ++--
tests/ref/fate/{wmv8-drm => wmv3-drm-dec} | 0
tests/ref/fate/{wmv8-drm-nodec => wmv3-drm-nodec} | 0
3 files
On Wed, Feb 22, 2017 at 05:38:59PM +0100, Diego Biurrun wrote:
> ---
> doc/examples/Makefile | 46 +++---
> 1 file changed, 23 insertions(+), 23 deletions(-)
OKed by Luca on IRC.
Diego
___
libav-deve
On Wed, Feb 22, 2017 at 05:39:02PM +0100, Diego Biurrun wrote:
> This allows dropping /dev/null as reference value when no output is generated.
> ---
> tests/fate-run.sh | 2 +-
> tests/fate/libavcodec.mak | 4 +---
> tests/fate/libavutil.mak | 13 ++---
>
On Sat, Feb 25, 2017 at 06:46:33PM +0100, Luca Barbato wrote:
> On 25/02/2017 17:22, Diego Biurrun wrote:
> > This fixes the test with mmxext disabled.
> > ---
> > tests/fate/qt.mak | 2 +-
> > tests/ref/fate/svq3-2 | 18 +-
> > 2 files change
Previously, all link-time dependencies were added for all libraries,
resulting in bogus link-time dependencies since not all dependencies
are shared across libraries. Also, in some cases like libavutil, not
all dependencies were taken into account, resulting in some cases of
underlinking.
To
Leaving those variables in an undefined state allows them getting implicitly
enabled when they are declared as weak dependencies of other components.
In that case, the library check is not run and required linker flags are not
added, resulting in a failing build.
Fixes linking when enabling
On Fri, Feb 24, 2017 at 07:59:13PM +0100, Luca Barbato wrote:
> On 24/02/2017 19:24, Anton Khirnov wrote:
> > It's very much non-obvious what exactly does this fix and how, and why
> > this specific fix is the right way to do it.
>
> Having a mismatch between the number of channels in the stream
This fixes the test with mmxext disabled.
---
tests/fate/qt.mak | 2 +-
tests/ref/fate/svq3-2 | 18 +-
2 files changed, 10 insertions(+), 10 deletions(-)
diff --git a/tests/fate/qt.mak b/tests/fate/qt.mak
index 761db8d..b51168f 100644
--- a/tests/fate/qt.mak
+++
On Thu, Feb 23, 2017 at 10:11:04PM +0100, Janne Grunau wrote:
> On 2017-02-21 18:26:24 +0100, Diego Biurrun wrote:
> > External dependencies cannot be handled as weak dependencies since they need
> > to be explicitly enabled. If a weak dependency is set, the variable
&g
On Wed, Feb 22, 2017 at 12:53:35PM -0500, Vittorio Giovara wrote:
> --- /dev/null
> +++ b/libavcodec/pixlet.c
> @@ -0,0 +1,689 @@
> +static int read_high_coeffs(AVCodecContext *avctx, uint8_t *src, int16_t
> *dst,
> +int size, int64_t c, int a, int64_t d,
> +
On Wed, Feb 22, 2017 at 06:03:42PM +0100, Diego Biurrun wrote:
> On Wed, Feb 22, 2017 at 11:49:36AM -0500, Vittorio Giovara wrote:
> > On Wed, Feb 22, 2017 at 11:39 AM, Diego Biurrun <di...@biurrun.de> wrote:
> > > --- /dev/null
> > > +++ b/tests/fate
On Wed, Feb 22, 2017 at 11:49:36AM -0500, Vittorio Giovara wrote:
> On Wed, Feb 22, 2017 at 11:39 AM, Diego Biurrun <di...@biurrun.de> wrote:
> > --- /dev/null
> > +++ b/tests/fate/examples.mak
> > @@ -0,0 +1,32 @@
> > +FATE_EXAMPLES-$(CONFIG_DECODE_AUDIO
On Mon, Feb 20, 2017 at 11:19:38PM +0200, Martin Storsjö wrote:
> On Fri, 17 Feb 2017, Diego Biurrun wrote:
>
> >None of them are specific to the YASM assembler.
> >---
> >
> >"x86-asm" it is.
>
> Kinda ok with me, but I want to hear acks/nacks from
---
doc/encoders.texi| 2 +-
doc/faq.texi | 8
doc/filters.texi | 4 ++--
tests/fate-run.sh| 4 ++--
tests/fate/demux.mak | 34 +-
tests/fate/h264.mak | 2 +-
tests/fate/microsoft.mak | 2 +-
The codec used in those files is WMV3/WMV9, not WMV2/WMV8.
---
tests/fate/microsoft.mak | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/tests/fate/microsoft.mak b/tests/fate/microsoft.mak
index 5b32c7f..4e66307 100644
--- a/tests/fate/microsoft.mak
+++
This allows dropping /dev/null as reference value when no output is generated.
---
tests/fate-run.sh | 2 +-
tests/fate/libavcodec.mak | 4 +---
tests/fate/libavutil.mak | 13 ++---
3 files changed, 8 insertions(+), 11 deletions(-)
diff --git a/tests/fate-run.sh
---
People keep forgetting to run "make check", this integrates the build
targets into fate so they will show up as failed tests.
tests/Makefile | 1 +
tests/fate-run.sh | 4
tests/fate/build.mak| 14 ++
tests/fate/examples.mak | 32
---
doc/examples/Makefile | 46 +++---
1 file changed, 23 insertions(+), 23 deletions(-)
diff --git a/doc/examples/Makefile b/doc/examples/Makefile
index c42c68f..40c9557 100644
--- a/doc/examples/Makefile
+++ b/doc/examples/Makefile
@@ -1,31 +1,31 @@
On Sun, Feb 05, 2017 at 01:14:46PM +0100, Diego Biurrun wrote:
> ---
>
> This covers a regression I introduced a long time ago.
>
> tests/fate/qt.mak | 10 --
> tests/ref/fate/{svq3 => svq3-1} | 0
> tests/ref/fate/svq3-2 | 20 +++
From: Kieran Kunhya <kie...@kunhya.com>
Decodes YUV 4:2:2 10-bit and RGB 12-bit files.
Older files with more subbands, skips, Bayer, alpha not supported.
Further fixes and refactorings by Anton Khirnov <an...@khirnov.net>,
Diego Biurrun <di...@biurrun.de>, Vittorio Giov
---
tests/fate/video.mak | 12
tests/ref/fate/cfhd-1 | 11 +++
tests/ref/fate/cfhd-2 | 11 +++
tests/ref/fate/cfhd-3 | 11 +++
4 files changed, 45 insertions(+)
create mode 100644 tests/ref/fate/cfhd-1
create mode 100644 tests/ref/fate/cfhd-2
create mode
libavcodec/vaapi.h:58:1: warning: attribute 'deprecated' is ignored, place it
after "struct" to apply attribute to type declaration [-Wignored-attributes]
---
libavcodec/vaapi.h | 3 +--
libavcodec/xvmc.h | 2 +-
2 files changed, 2 insertions(+), 3 deletions(-)
diff --git a/libavcodec/vaapi.h
On Wed, Feb 22, 2017 at 09:56:36AM +0100, Luca Barbato wrote:
> ---
> tests/ref/seek/lavf-mkv | 26 +-
> 1 file changed, 13 insertions(+), 13 deletions(-)
Hopefully OK
Diego
___
libav-devel mailing list
libav-devel@libav.org
On Tue, Feb 21, 2017 at 07:58:04PM +0100, Luca Barbato wrote:
> On 21/02/2017 18:26, Diego Biurrun wrote:
> > all: $(AVPROGS) alltools checkheaders examples testprogs
>
> What?
That's a local change that I have to force building all targets on Oracle.
Notice that it's not pa
---
This was previously approved.
configure | 26 +-
1 file changed, 9 insertions(+), 17 deletions(-)
diff --git a/configure b/configure
index 6f1be32..ef6a8e0 100755
--- a/configure
+++ b/configure
@@ -1608,7 +1608,6 @@ SYSTEM_FUNCS="
CommandLineToArgvW
Previously, all link-time dependencies were added for all libraries,
resulting in bogus link-time dependencies since not all dependencies
are shared across libraries. Also, in some cases like libavutil, not
all dependencies were taken into account, resulting in some cases of
underlinking.
To
External dependencies cannot be handled as weak dependencies since they need
to be explicitly enabled. If a weak dependency is set, the variable
corresponding
to the weak dependency can be enabled without the rest of the build system
settings, resulting in a failing build.
This reverts commit
On Tue, Feb 21, 2017 at 02:52:01PM +0100, Luca Barbato wrote:
> On 16/02/2017 10:16, Diego Biurrun wrote:
> > This unclutters the top-level directory and groups related files together.
> > ---
> >
> > Cleaner and simpler now that the logic is separate from the examples.
On Thu, Feb 16, 2017 at 10:16:05AM +0100, Diego Biurrun wrote:
> ---
>
> Noticed one more stray EXAMPLES reference.
>
> Makefile | 2 +-
> doc/Makefile | 2 +-
> 2 files changed, 2 insertions(+), 2 deletions(-)
ping the set
Diego
__
On Tue, Feb 21, 2017 at 09:19:06AM +, Konda Raju wrote:
> From caa77fa0cecd9371fba3da72b338880d5e67fff6 Mon Sep 17 00:00:00 2001
> From: rajukonda
This looks like a Subversion user name or something, please configure
your Git with your full name.
Diego
On Mon, Feb 20, 2017 at 03:10:34AM +0100, Luca Barbato wrote:
> ---
> libavformat/rtsp.c | 50 +-
> 1 file changed, 25 insertions(+), 25 deletions(-)
rtsp: Lazily set up the pollfd array once
Diego
___
On Mon, Feb 20, 2017 at 03:10:33AM +0100, Luca Barbato wrote:
> And use av_malloc_array.
> ---
> libavformat/rtsp.c | 9 ++---
> 1 file changed, 6 insertions(+), 3 deletions(-)
Lazily allocate the pollfd array
Diego
___
libav-devel mailing list
None of them are specific to the YASM assembler.
---
"x86-asm" it is.
Makefile | 12 +++
avbuild/arch.mak | 2 +-
configure | 65 +++
libavcodec/x86/Makefile | 72
---
libavcodec/nvenc.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libavcodec/nvenc.c b/libavcodec/nvenc.c
index 90b9d1a..30da83a 100644
--- a/libavcodec/nvenc.c
+++ b/libavcodec/nvenc.c
@@ -479,7 +479,7 @@ typedef struct GUIDTuple {
#define PRESET(name, ...)
On Thu, Feb 16, 2017 at 10:57:21PM +0200, Martin Storsjö wrote:
> On Tue, 14 Feb 2017, Diego Biurrun wrote:
>
> >None of them are specific to the YASM assembler.
> >---
> >
> >Same as before, better names for the following are welcome:
> >
> >- --disab
On Thu, Feb 16, 2017 at 11:01:18PM +0200, Martin Storsjö wrote:
> On Thu, 16 Feb 2017, Diego Biurrun wrote:
> >NASM is more actively maintained and permits generating dependency
> >information
> >as a sideeffect of assembling, thus cutting build times in half.
NASM is more actively maintained and permits generating dependency information
as a sideeffect of assembling, thus cutting build times in half.
---
Updated the documentation to reflect that nasm is now preferred over yasm.
.travis.yml | 4 ++--
configure| 14 +++---
Have check_pkg_config() enable variables and set cflags and extralibs
instead of relegating that task to require_pkg_config. This simplifies
require_pkg_config(), is consistent with what other helper functions
like check_lib() do and allows getting rid of some manual variable
setting in places
This is less confusing than encountering "" in the argument list.
---
configure | 15 +++
1 file changed, 7 insertions(+), 8 deletions(-)
diff --git a/configure b/configure
index 4635b73..287b1e5 100755
--- a/configure
+++ b/configure
@@ -1139,7 +1139,6 @@ require_pkg_config(){
601 - 700 of 10585 matches
Mail list logo