On Sun, 26 Aug 2012, Luca Barbato wrote:
On 8/26/12 5:35 PM, Samuel Pitoiset wrote:
---
This patch fixes the following crash:
Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_INVALID_ADDRESS at address: 0x0018
[Switching to process 16366 thread 0x651b]
On Fri, 17 Aug 2012, Martin Storsjö wrote:
The condition for calling the rtpdec cleanup was broken in
df8cf076c86.
---
libavformat/rtsp.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavformat/rtsp.c b/libavformat/rtsp.c
index ce18610..0390862 100644
--- a/libavformat
This C99 feature is generally not used in the libav codebase,
since it breaks building with some fringe compilers (GCC 2.95,
MSVC).
---
libavfilter/vf_hqdn3d.c |3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/libavfilter/vf_hqdn3d.c b/libavfilter/vf_hqdn3d.c
index 535657a..
This also makes sure the function pointers are stored via the same
union member as they are accessed via.
---
avprobe.c | 27 ++-
1 file changed, 14 insertions(+), 13 deletions(-)
diff --git a/avprobe.c b/avprobe.c
index 00e0ad0..3059f5e 100644
--- a/avprobe.c
+++ b/avpr
Compilation seems to fail on some GCC 3.4 instances due to this
duplicate declaration.
---
libavfilter/vf_hqdn3d.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavfilter/vf_hqdn3d.c b/libavfilter/vf_hqdn3d.c
index 5bb896c..71b670d 100644
--- a/libavfilter/vf_hqdn3d.c
++
This fixes crashes introduced by 2e8f3cbcda5, the codec can be null
when called from parsers.
---
This fixes the issue pointed out by Uoti in
http://lists.libav.org/pipermail/libav-devel/2012-August/033836.html
and that Samuel tried to fix differently in
http://patches.libav.org/patch/26549/.
lib
This is required for playback with the Stagefright RTSP framework
on Android.
---
libavformat/sdp.c | 12
1 file changed, 12 insertions(+)
diff --git a/libavformat/sdp.c b/libavformat/sdp.c
index 2304a68..b89f4b7 100644
--- a/libavformat/sdp.c
+++ b/libavformat/sdp.c
@@ -155,8 +155
On Tue, 28 Aug 2012, Diego Biurrun wrote:
On Tue, Aug 28, 2012 at 01:03:29PM +0100, Mans Rullgard wrote:
--- a/configure
+++ b/configure
@@ -2291,6 +2293,11 @@ add_ldflags $_flags $_ldflags
+probe_cc hostld "$host_ld"
+host_ldflags_filter=$_flags_filter
+add_host_ldflags $_flags $_ldflags
+HOS
On Tue, 28 Aug 2012, Mans Rullgard wrote:
This adds new HOSTLD and related settings for host linker allowing
it to be different from HOSTCC.
Signed-off-by: Mans Rullgard
---
Makefile | 2 +-
common.mak | 2 +-
configure | 15 ---
3 files changed, 14 insertions(+), 5 deletions(-)
On Sat, 25 Aug 2012, Samuel Pitoiset wrote:
---
You can test it by using the following link:
rtsp://elmer.tapir.caltech.edu/Lecture6_tape1_web.mov
However, the server seems a little weird because sometimes
it fails the connection and lot of packets are missed...
I tested that link with QuickTim
On Tue, 28 Aug 2012, Mans Rullgard wrote:
This enables replacing the -l and -L flags used to specify the
just-built libraries when linking the tools and shared libs with
non-standard syntaxes. System library flags are already handled
by the filtering mechanism in configure.
Signed-off-by: Mans
This fixes building with DEBUG defined after the function was made
static and the prototype removed in d7f9786cbc.
---
libavcodec/audio_frame_queue.c | 32
1 file changed, 16 insertions(+), 16 deletions(-)
diff --git a/libavcodec/audio_frame_queue.c b/libavcodec
On Tue, 28 Aug 2012, Diego Elio Pettenò wrote:
On 28/08/2012 04:59, Martin Storsjö wrote:
const char *pset_string = "; sprop-parameter-sets=";
+const char *profile_string = "; profile-level-id=";
Any particular reasons why these are not `static const char name[
---
libavformat/sdp.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libavformat/sdp.c b/libavformat/sdp.c
index 7413386..9bb4815 100644
--- a/libavformat/sdp.c
+++ b/libavformat/sdp.c
@@ -154,8 +154,8 @@ static char *extradata2psets(AVCodecContext *c)
{
char *pse
On Tue, 28 Aug 2012, Mans Rullgard wrote:
This enables replacing the -l and -L flags used to specify the
just-built libraries when linking the tools and shared libs with
non-standard syntaxes. System library flags are already handled
by the filtering mechanism in configure.
Signed-off-by: Mans
The Windows SDK in MSVC doesn't have mkdir, only _mkdir, and
MSDN says one should include direct.h to use it.
---
tools/ismindex.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/tools/ismindex.c b/tools/ismindex.c
index f6b0c6c..bf8c69d 100644
--- a/tools/ismindex.c
+++
---
tools/qt-faststart.c |3 +++
1 file changed, 3 insertions(+)
diff --git a/tools/qt-faststart.c b/tools/qt-faststart.c
index 8d2196c..f33d6fa 100644
--- a/tools/qt-faststart.c
+++ b/tools/qt-faststart.c
@@ -32,6 +32,9 @@
#ifdef __MINGW32__
#define fseeko(x, y, z) fseeko64(x, y, z)
#defi
---
libavcodec/motion-test.c |2 --
libswscale/colorspace-test.c |1 -
2 files changed, 3 deletions(-)
diff --git a/libavcodec/motion-test.c b/libavcodec/motion-test.c
index 4c80368..bc4f0da 100644
--- a/libavcodec/motion-test.c
+++ b/libavcodec/motion-test.c
@@ -26,8 +26,6 @@
#inclu
---
tools/cws2fws.c |6 ++
tools/pktdumper.c |6 ++
2 files changed, 12 insertions(+)
diff --git a/tools/cws2fws.c b/tools/cws2fws.c
index 68f7953..74588c1 100644
--- a/tools/cws2fws.c
+++ b/tools/cws2fws.c
@@ -6,11 +6,17 @@
* This utility converts compressed Macromedia Flash
The MSVC headers don't define PATH_MAX by default, unless _POSIX_
is defined before including limits.h. Defining _POSIX_ makes the
open/write/close function declarations go away though.
---
tools/pktdumper.c |4
1 file changed, 4 insertions(+)
diff --git a/tools/pktdumper.c b/tools/pktdu
---
tools/graph2dot.c |7 +++
1 file changed, 7 insertions(+)
diff --git a/tools/graph2dot.c b/tools/graph2dot.c
index 51a1ab1..e7f487d 100644
--- a/tools/graph2dot.c
+++ b/tools/graph2dot.c
@@ -18,7 +18,10 @@
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 US
On Tue, 28 Aug 2012, Måns Rullgård wrote:
Martin Storsjö writes:
The Windows SDK in MSVC doesn't have mkdir, only _mkdir, and
MSDN says one should include direct.h to use it.
---
tools/ismindex.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/tools/isminde
On Tue, 28 Aug 2012, Måns Rullgård wrote:
Martin Storsjö writes:
The MSVC headers don't define PATH_MAX by default, unless _POSIX_
is defined before including limits.h. Defining _POSIX_ makes the
open/write/close function declarations go away though.
Why does this file not use stdio?
Also change the snprintf size to use the full buffer, since
snprintf always null-terminates the buffer.
---
tools/pktdumper.c |8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/tools/pktdumper.c b/tools/pktdumper.c
index 9243c93..fffeeeb 100644
--- a/tools/pktdumper.c
+
PATH_MAX is not necessarily available on all systems, e.g. it's
normally not available on MSVC, and is not guaranteed to defined
on a POSIX system either.
---
tools/pktdumper.c |6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/tools/pktdumper.c b/tools/pktdumper.c
index
The buffer is filled with a strncpy(foo, sizeof(buffer) - 1). If
the size is insufficient, the last byte of the buffer needs to be
zero initialized.
---
tools/pktdumper.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/pktdumper.c b/tools/pktdumper.c
index fffeeeb..4f1
This takes care of null-terminating the buffer if it is too small,
which wasn't handled properly before.
---
tools/pktdumper.c |5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/tools/pktdumper.c b/tools/pktdumper.c
index fffeeeb..f21a8e0 100644
--- a/tools/pktdumper.c
+++
On Thu, 30 Aug 2012, Måns Rullgård wrote:
Martin Storsjö writes:
This takes care of null-terminating the buffer if it is too small,
which wasn't handled properly before.
---
tools/pktdumper.c |5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/tools/pktdumpe
On Thu, 30 Aug 2012, Martin Storsjö wrote:
On Thu, 30 Aug 2012, Måns Rullgård wrote:
Martin Storsjö writes:
This takes care of null-terminating the buffer if it is too small,
which wasn't handled properly before.
---
tools/pktdumper.c |5 +++--
1 file changed, 3 insertions(
---
libavfilter/vf_delogo.c| 14 +++---
libavfilter/vf_lut.c | 22 +++---
libavfilter/vsrc_testsrc.c | 12 ++--
3 files changed, 24 insertions(+), 24 deletions(-)
diff --git a/libavfilter/vf_delogo.c b/libavfilter/vf_delogo.c
index 5e2cf48..2b4705b
---
libavformat/crypto.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavformat/crypto.c b/libavformat/crypto.c
index 7095360..93c9f23 100644
--- a/libavformat/crypto.c
+++ b/libavformat/crypto.c
@@ -48,7 +48,7 @@ typedef struct {
#define D AV_OPT_FLAG_DECODING_PARAM
On Mon, 3 Sep 2012, Diego Biurrun wrote:
---
No comment on the rest of the patch itself, but the comment is slightly
incorrect - if you say "simplify", one would assume that it doesn't change
behaviour, while afaik this is the fix for e.g. some of the fate failures
(where HAVE_FMA4 == 1 but
---
libavcodec/g722dec.c|2 +-
libavcodec/libvpxenc.c |2 +-
libavcodec/mpegvideo.h |2 +-
libavcodec/options_table.h | 14 +++---
libavdevice/libcdio.c |2 +-
libavfilter/vf_drawtext.c |2 +-
libavformat/movenc.c|2 +-
libavform
---
libavcodec/ac3dec.c |2 +-
libavcodec/libx264.c |8
libavfilter/af_amix.c|2 +-
libavfilter/af_asyncts.c |2 +-
libavresample/options.c |8
5 files changed, 11 insertions(+), 11 deletions(-)
diff --git a/libavcodec/ac3dec.c b/libavcodec/ac3d
---
libavfilter/buffersrc.c|2 +-
libavfilter/vsrc_testsrc.c |2 +-
libavutil/opt.c|2 +-
3 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/libavfilter/buffersrc.c b/libavfilter/buffersrc.c
index 62b9e99..8e344cb 100644
--- a/libavfilter/buffersrc.c
+++ b/l
---
libavutil/opt.h |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavutil/opt.h b/libavutil/opt.h
index 8f800fc..775267f 100644
--- a/libavutil/opt.h
+++ b/libavutil/opt.h
@@ -261,10 +261,10 @@ typedef struct AVOption {
* the default value for scalar options
On Tue, 4 Sep 2012, Luca Barbato wrote:
On 9/3/12 8:23 PM, Martin Storsjö wrote:
---
libavcodec/g722dec.c|2 +-
libavcodec/libvpxenc.c |2 +-
libavcodec/mpegvideo.h |2 +-
libavcodec/options_table.h | 14 +++---
libavdevice/libcdio.c |2
From: Michael Niedermayer
Previously, the value given to put_bits was 10 bits long for positive
predictors, even though 9 bits were to be written. The extra bit could
in some cases overwrite existing bits in the bitstream writer cache.
This fixes a failed assert in put_bits.h, when running a ver
Previously, the put_bits call writing the value wrote a value
larger than the number of bits specified, failing asserts
in debug mode. There was no actual bitstream writer corruption,
since the overwritten bit already always was set to 1.
---
libavcodec/proresenc.c |3 +--
1 file changed, 1 in
The ParserContext class doesn't have an AVClass, required for
using it as a logging class.
---
libavcodec/parser.c |8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/libavcodec/parser.c b/libavcodec/parser.c
index 29904f9..dd2d836 100644
--- a/libavcodec/parser.c
+++ b/
The data in coded_frame isn't allocated using get_buffer, but
is copied from the input frame to the encoder, so we should
not try to free it ourselves.
This fixes an assert failure when running in debug mode.
---
libavcodec/proresenc.c |3 ---
1 file changed, 3 deletions(-)
diff --git a/liba
On Tue, 4 Sep 2012, Derek Buitenhuis wrote:
On 04/09/2012 7:59 AM, Martin Storsjö wrote:
The data in coded_frame isn't allocated using get_buffer, but
is copied from the input frame to the encoder, so we should
not try to free it ourselves.
This fixes an assert failure when running in
On Tue, 4 Sep 2012, Derek Buitenhuis wrote:
On 04/09/2012 7:48 AM, Martin Storsjö wrote:
The ParserContext class doesn't have an AVClass, required for
using it as a logging class.
---
libavcodec/parser.c |8
1 file changed, 4 insertions(+), 4 deletions(-)
Do we not
On Tue, 4 Sep 2012, Kostya Shishkov wrote:
On Tue, Sep 04, 2012 at 02:21:25PM +0300, Martin Storsjö wrote:
From: Michael Niedermayer
Previously, the value given to put_bits was 10 bits long for positive
predictors, even though 9 bits were to be written. The extra bit could
in some cases
On Tue, 4 Sep 2012, Kostya Shishkov wrote:
On Tue, Sep 04, 2012 at 02:36:34PM +0300, Martin Storsjö wrote:
Previously, the put_bits call writing the value wrote a value
larger than the number of bits specified, failing asserts
in debug mode. There was no actual bitstream writer corruption
From: Michael Niedermayer
It seems the condition was flipped from what was intended.
---
libavcodec/elbg.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/elbg.c b/libavcodec/elbg.c
index 31195f3..0aa8e16 100644
--- a/libavcodec/elbg.c
+++ b/libavcodec/elbg.c
@@
From: Michael Niedermayer
It now just checks uninitialized and unused data.
---
libavcodec/mpegvideo_enc.c |1 -
1 file changed, 1 deletion(-)
diff --git a/libavcodec/mpegvideo_enc.c b/libavcodec/mpegvideo_enc.c
index aeafb4d..77b0ab5 100644
--- a/libavcodec/mpegvideo_enc.c
+++ b/libavcodec
From: Michael Niedermayer
It is possible in various error pathes as well as gap handling
that this has already been allocated. Its not clear why that
would be a problem with the current code, thus disable the
assert to avoid common assert failure when asserts are enabled.
---
libavcodec/h264.c |
From: Michael Niedermayer
This assert is no longer true since h264 error concealment needs
last_picture_ptr to be set.
---
libavcodec/mpegvideo.c |3 ---
1 file changed, 3 deletions(-)
diff --git a/libavcodec/mpegvideo.c b/libavcodec/mpegvideo.c
index e9aff3b..718df8b 100644
--- a/libavcode
From: Michael Niedermayer
This fixes assert failures when running in debug mode.
---
libavcodec/svq1enc.c |1 +
1 file changed, 1 insertion(+)
diff --git a/libavcodec/svq1enc.c b/libavcodec/svq1enc.c
index 3d90558..d1a55bd 100644
--- a/libavcodec/svq1enc.c
+++ b/libavcodec/svq1enc.c
@@ -486
On Tue, 4 Sep 2012, Benjamin Larsson wrote:
On 09/04/2012 03:20 PM, Martin Storsjö wrote:
From: Michael Niedermayer
It is possible in various error pathes as well as gap handling
that this has already been allocated. Its not clear why that
would be a problem with the current code, thus
From: Michael Niedermayer
It is possible in various error paths as well as gap handling
that this has already been allocated. It is not clear why that
would be a problem with the current code, thus disable the
assert to avoid a common assert failure when asserts are enabled.
---
libavcodec/h264.
From: Mans Rullgard
---
configure | 23 ---
1 file changed, 20 insertions(+), 3 deletions(-)
diff --git a/configure b/configure
index c6c5c19..c2dc898 100755
--- a/configure
+++ b/configure
@@ -2709,9 +2709,6 @@ case $target_os in
ranlib=:
enable dos_path
From: Luca Barbato
A second expansion forces the preprocessor to tokenize properly.
---
libavcodec/dsputil.h | 10 +++---
1 file changed, 7 insertions(+), 3 deletions(-)
diff --git a/libavcodec/dsputil.h b/libavcodec/dsputil.h
index e47cd2c..bb70428 100644
--- a/libavcodec/dsputil.h
+++ b
From: "Ronald S. Bultje"
---
compat/msvcrt/snprintf.c | 63 +++
compat/strtod.c | 94 ++
configure|7
libavutil/Makefile |3 ++
4 files changed, 167 insertions(+)
create mode
From: "Ronald S. Bultje"
Use with the following configure parameters:
--cc='c99wrap cl' --ld='c99wrap link'
---
These 4 patches (plus http://patches.libav.org/patch/26881/, which
isn't strictly necessary but does help) are the only things that
remain for using MSVC, using the new converter.
---
On Wed, 5 Sep 2012, Ronald S. Bultje wrote:
Hi,
On Mon, Sep 3, 2012 at 11:23 AM, Martin Storsjö wrote:
---
libavcodec/g722dec.c|2 +-
libavcodec/libvpxenc.c |2 +-
libavcodec/mpegvideo.h |2 +-
libavcodec/options_table.h | 14 +++---
libavdevice
On Wed, 5 Sep 2012, Diego Biurrun wrote:
On Tue, Sep 04, 2012 at 11:47:23PM +0300, Martin Storsjö wrote:
From: "Ronald S. Bultje"
Use with the following configure parameters:
--cc='c99wrap cl' --ld='c99wrap link'
This should be documented somewhere at some poi
On Wed, 5 Sep 2012, Diego Biurrun wrote:
---
With this patch applied the fate-vsynth1-mpeg2-ivlc-qprd passes even in
debug mode, it was segfaulting before, and all of FATE passes with
-DDEBUG in CFLAGS.
The only thing broken about the av_dlog calls is the log context (which
doesn't have an AV
On Tue, 4 Sep 2012, Martin Storsjö wrote:
From: Michael Niedermayer
It is possible in various error paths as well as gap handling
that this has already been allocated. It is not clear why that
would be a problem with the current code, thus disable the
assert to avoid a common assert failure
From: Michael Niedermayer
---
libavutil/opt.h |8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/libavutil/opt.h b/libavutil/opt.h
index 775267f..ca13682 100644
--- a/libavutil/opt.h
+++ b/libavutil/opt.h
@@ -64,7 +64,7 @@
*
* static const AVOption options[] = {
---
tests/fate.sh |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/fate.sh b/tests/fate.sh
index e04c871..c267182 100755
--- a/tests/fate.sh
+++ b/tests/fate.sh
@@ -107,6 +107,6 @@ mkdir -p ${build}
configure >configure.log 2>&1 || fail $? "error configuring"
compil
This disables the following warnings:
C4100: unreferenced formal parameter
(1035 occurrances)
C4214: nonstandard extension used : bit field types other than int
(609 occurances)
C4996: 'avpriv_snprintf': This function or variable may be unsafe.
Consider using _snprintf_s instead. To disabl
---
tests/fate.sh |1 +
1 file changed, 1 insertion(+)
diff --git a/tests/fate.sh b/tests/fate.sh
index b95b222..411fa28 100755
--- a/tests/fate.sh
+++ b/tests/fate.sh
@@ -49,6 +49,7 @@ configure()(
${cpu:+--cpu="$cpu"}\
${cross_pr
On Wed, 5 Sep 2012, Samuel Pitoiset wrote:
---
This patch fixes the following crash:
Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_INVALID_ADDRESS at address: 0x0018
[Switching to process 17337 thread 0x661b]
avio_rb32 (s=0x0) at aviobuf.c:552
552 uns
On Wed, 5 Sep 2012, Samuel Pitoiset wrote:
+/* JFIF header */
+put_marker(&pbc, APP0);
+put_bits(&pbc, 16, 16);
+ff_put_string(&pbc, "JFIF", 1);
This uses libavcodec private symbols that can't be accessed from libavformat
like this (try building with --enable-shared on linux,
On Wed, 5 Sep 2012, Diego Biurrun wrote:
On Wed, Sep 05, 2012 at 08:43:46PM +0300, Martin Storsjö wrote:
--- a/tests/fate.sh
+++ b/tests/fate.sh
@@ -107,6 +107,6 @@ mkdir -p ${build}
configure >configure.log 2>&1 || fail $? "error configuring"
compile >compile.log
The default ones work, but outputs the preprocessed file on stdout
(into config.log).
---
This requires an updated c99wrap (pushed to git master right now)
that handles the -Fi parameter properly.
---
configure |1 +
1 file changed, 1 insertion(+)
diff --git a/configure b/configure
index 25a5
On Thu, 6 Sep 2012, Samuel Pitoiset wrote:
---
This patch fixes the following crash:
Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_INVALID_ADDRESS at address: 0x0018
[Switching to process 17337 thread 0x661b]
avio_rb32 (s=0x0) at aviobuf.c:552
552 uns
On Thu, 6 Sep 2012, Samuel Pitoiset wrote:
---
Changelog| 1 +
libavformat/Makefile | 1 +
libavformat/rtpdec.c | 1 +
libavformat/rtpdec_formats.h | 1 +
libavformat/rtpdec_jpeg.c| 330 +++
libavformat/version.h
On Thu, 6 Sep 2012, Samuel Pitoiset wrote:
On Thu, Sep 6, 2012 at 3:00 PM, Martin Storsjö wrote:
On Thu, 6 Sep 2012, Samuel Pitoiset wrote:
---
Changelog| 1 +
libavformat/Makefile | 1 +
libavformat/rtpdec.c | 1 +
libavformat/rtpdec_formats.h | 1
On Thu, 6 Sep 2012, Samuel Pitoiset wrote:
These symbols will be used from the RTP/JPEG depacketizer.
---
libavcodec/mjpeg.c | 63 -
libavcodec/mjpeg.h | 76 ++---
libavcodec/mjpeg2jpeg_bsf.c | 12 +++
---
libavformat/os_support.h |7 +++
1 file changed, 7 insertions(+)
diff --git a/libavformat/os_support.h b/libavformat/os_support.h
index 8e2eb83..533b06b 100644
--- a/libavformat/os_support.h
+++ b/libavformat/os_support.h
@@ -38,6 +38,13 @@
# define fstat(f,s) _fstati64((f), (s))
#
This allows opening an existing file for updating.
Is combo used by anything already, which this breaks?
---
libavformat/file.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavformat/file.c b/libavformat/file.c
index 0e3577d..c42dfe9 100644
--- a/libavformat/file.c
+++
This muxer splits the output from the ismv muxer into individual
files, in realtime.
The same can also be done by the standalone tool ismindex, but this
muxer is needed for doing it in realtime (especially for live
streams that need extra handling for updating the lookahead fields
in the fragment
On Thu, 6 Sep 2012, Måns Rullgård wrote:
Martin Storsjö writes:
This allows opening an existing file for updating.
Is combo used by anything already, which this breaks?
---
libavformat/file.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavformat/file.c b
On Thu, 6 Sep 2012, Måns Rullgård wrote:
Martin Storsjö writes:
---
libavformat/os_support.h |7 +++
1 file changed, 7 insertions(+)
diff --git a/libavformat/os_support.h b/libavformat/os_support.h
index 8e2eb83..533b06b 100644
--- a/libavformat/os_support.h
+++ b/libavformat
On Thu, 6 Sep 2012, Martin Storsjö wrote:
On Thu, 6 Sep 2012, Samuel Pitoiset wrote:
---
This patch fixes the following crash:
Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_INVALID_ADDRESS at address: 0x0018
[Switching to process 17337 thread
For demuxers/decoders that reset the channel layout regularly
(e.g. rtpdec_qt) to a default value of 0, we avoid reinitializing
the resampling needlessly, since the call to
guess_input_channel_layout within the resample_changed block would
end up with the same channel layout anyway.
Does this make
The default channel layout is guessed at startup, but this is
required if the demuxer/decoder resets the channel layout back
to 0/unknown regularly.
---
avplay.c |2 ++
1 file changed, 2 insertions(+)
diff --git a/avplay.c b/avplay.c
index 0c642e0..6f8f2cb 100644
--- a/avplay.c
+++ b/avplay.c
---
libavformat/mov_chan.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavformat/mov_chan.c b/libavformat/mov_chan.c
index 800bb10..b1fa13f 100644
--- a/libavformat/mov_chan.c
+++ b/libavformat/mov_chan.c
@@ -580,7 +580,7 @@ int ff_mov_read_chan(AVFormatContext *s, AVI
If regularly parsing new chan atoms (as in rtpdec_qt), but the
chan atoms don't actually contain any channel layout, don't reset
the value that the caller has filled in (by guessing).
---
This is an alternative solution to the patches for avconv/avplay.
---
libavformat/mov_chan.c |7 ---
1
On Fri, 7 Sep 2012, Måns Rullgård wrote:
Martin Storsjö writes:
On Thu, 6 Sep 2012, Måns Rullgård wrote:
Martin Storsjö writes:
---
libavformat/os_support.h |7 +++
1 file changed, 7 insertions(+)
diff --git a/libavformat/os_support.h b/libavformat/os_support.h
index 8e2eb83
On Fri, 7 Sep 2012, Måns Rullgård wrote:
Martin Storsjö writes:
On Thu, 6 Sep 2012, Måns Rullgård wrote:
Martin Storsjö writes:
This allows opening an existing file for updating.
Is combo used by anything already, which this breaks?
---
libavformat/file.c |2 +-
1 file changed, 1
On Thu, 6 Sep 2012, Samuel Pitoiset wrote:
These symbols will be used from the RTP/JPEG depacketizer.
---
libavcodec/mjpeg.c | 14 +++---
libavcodec/mjpeg.h | 14 +++---
libavcodec/mjpeg2jpeg_bsf.c | 12 ++--
libavcodec/mjpegdec.c | 24 ---
On Thu, 6 Sep 2012, Samuel Pitoiset wrote:
---
TODO:
1) Do not hardcoded video pixel dimensions (any ideas ?)
s1->streams[0]->codec->width/height
2) QuickTimePlayer doesn't play the video
---
Changelog | 1 +
libavformat/Makefile | 1 +
libavformat/rtpenc.c | 6 +++
On Fri, 7 Sep 2012, Diego Biurrun wrote:
On Fri, Sep 07, 2012 at 10:59:26AM +0300, Martin Storsjö wrote:
--- a/libavformat/mov_chan.c
+++ b/libavformat/mov_chan.c
@@ -579,9 +579,10 @@ int ff_mov_read_chan(AVFormatContext *s, AVIOContext *pb,
AVStream *st,
-if (layout_tag == 0
On Fri, 7 Sep 2012, Samuel Pitoiset wrote:
---
Fix build with --disable-everything --enable-demuxer=rtsp
---
Changelog| 1 +
configure| 2 +-
libavformat/Makefile | 1 +
libavformat/rtpdec.c | 1 +
libavformat/rtpdec_formats.h | 1 +
l
On Mon, 3 Sep 2012, Diego Biurrun wrote:
---
...l_mmx_avg_template.c => dsputil_avg_template.c} |0
libavcodec/x86/dsputil_mmx.c |8
...l_mmx_qns_template.c => dsputil_qns_template.c} |0
...l_mmx_rnd_template.c => dsputil_rnd_template.c} |0
libavcodec
On Fri, 7 Sep 2012, Justin Ruggles wrote:
On 09/07/2012 03:57 AM, Martin Storsjö wrote:
For demuxers/decoders that reset the channel layout regularly
(e.g. rtpdec_qt) to a default value of 0, we avoid reinitializing
the resampling needlessly, since the call to
guess_input_channel_layout within
On Sat, 8 Sep 2012, Måns Rullgård wrote:
Hendrik Leppkes writes:
On Sat, Sep 8, 2012 at 12:09 PM, Ronald S. Bultje wrote:
This sounds like the wrong fix tbh. The first chunk looks OK but
wouldn't other systems, eg old GCC versions, be similarly affected?
Shouldn't the true fix live elsewhe
On Sat, 8 Sep 2012, Måns Rullgård wrote:
Martin Storsjö writes:
On Sat, 8 Sep 2012, Måns Rullgård wrote:
Hendrik Leppkes writes:
On Sat, Sep 8, 2012 at 12:09 PM, Ronald S. Bultje wrote:
This sounds like the wrong fix tbh. The first chunk looks OK but
wouldn't other systems, e
On Sat, 8 Sep 2012, Måns Rullgård wrote:
Hendrik Leppkes writes:
On Sat, Sep 8, 2012 at 1:40 PM, Martin Storsjö wrote:
On Sat, 8 Sep 2012, Måns Rullgård wrote:
Hendrik Leppkes writes:
On Sat, Sep 8, 2012 at 12:09 PM, Ronald S. Bultje
wrote:
This sounds like the wrong fix tbh
On Fri, 7 Sep 2012, Mans Rullgard wrote:
Signed-off-by: Mans Rullgard
---
configure | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configure b/configure
index 25a56e7..61d8012 100755
--- a/configure
+++ b/configure
@@ -2323,7 +2323,7 @@ probe_cc(){
_flags='-nologo'
This reverts parts of e0c6cce4472. There is external mmx asm that
requires this alignment.
This fixes crashes when using swscale in builds with external mmx,
without inline assembly.
---
libswscale/utils.c |6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/libswscale/uti
This allows using it from libavformat as well. This will be used
by the RTP/JPEG depacketizer.
---
libavcodec/bitstream.c |2 +-
libavcodec/mjpegenc.c |6 +++---
libavcodec/mpeg4videoenc.c |2 +-
libavcodec/put_bits.h |4 ++--
4 files changed, 7 insertions(+), 7 delet
On Fri, 7 Sep 2012, Samuel Pitoiset wrote:
---
Changelog| 1 +
libavcodec/Makefile | 1 +
libavformat/Makefile | 1 +
libavformat/rtpdec.c | 1 +
libavformat/rtpdec_formats.h | 1 +
libavformat/rtpdec_jpeg.c| 331 +
On Fri, 7 Sep 2012, Martin Storsjö wrote:
On Fri, 7 Sep 2012, Måns Rullgård wrote:
Martin Storsjö writes:
On Thu, 6 Sep 2012, Måns Rullgård wrote:
Martin Storsjö writes:
---
libavformat/os_support.h |7 +++
1 file changed, 7 insertions(+)
diff --git a/libavformat
On Sun, 9 Sep 2012, Martin Storsjö wrote:
This allows using it from libavformat as well. This will be used
by the RTP/JPEG depacketizer.
---
libavcodec/bitstream.c |2 +-
libavcodec/mjpegenc.c |6 +++---
libavcodec/mpeg4videoenc.c |2 +-
libavcodec/put_bits.h |4
Previously the file descriptor was stored in the priv_data pointer.
---
libavformat/file.c | 33 +
1 file changed, 21 insertions(+), 12 deletions(-)
diff --git a/libavformat/file.c b/libavformat/file.c
index 0e3577d..3cfd28c 100644
--- a/libavformat/file.c
+++ b/
1 - 100 of 1817 matches
Mail list logo
