GNUTLS_SHUT_RDWR means GnuTLS will keep waiting for the server's
termination reply. But since we don't shutdown the TCP connection at
this point yet, GnuTLS will just keep skipping actual data from the
server, which basically is perceived as hang.
Use GNUTLS_SHUT_WR instead, which doesn't have
When the display is preempted, all vdpau objects are implicitly
destroyed by the driver. API calls will return either
VDP_STATUS_DISPLAY_PREEMPTED (if the display is still preempted), or
VDP_STATUS_INVALID_HANDLE (driver recovered from preemption). Since
libavcodec's vdpau code has no concept of
On Thu, 28 May 2015 11:30:57 +0300
Martin Storsjö mar...@martin.st wrote:
This fixes building with gcrypt-backed gnutls versions, broken
in 57cde2b180.
---
libavformat/tls_gnutls.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/libavformat/tls_gnutls.c
On Thu, 28 May 2015 11:48:49 +0300 (EEST)
Martin Storsjö mar...@martin.st wrote:
On Thu, 28 May 2015, wm4 wrote:
On Thu, 28 May 2015 11:30:57 +0300
Martin Storsjö mar...@martin.st wrote:
This fixes building with gcrypt-backed gnutls versions, broken
in 57cde2b180
This was probably broken some time ago. The breakage is now part of the
ABI. For example, we have:
AV_PIX_FMT_XYZ12BE
AV_PIX_FMT_NV16
AV_PIX_FMT_NV20LE
AV_PIX_FMT_NV20LE is wrong. It has the value 113, but as little-endian
format it should be even. This must have been quite obvious
OSX does not know MSG_NOSIGNAL, and provides its own non-standard
mechanism instead. I guess Apple hates standards.
---
Haven't really checked whether this really prevents SIGPIPE (too
bothersome to test), only that the function call succeeds. I don't
think there's a reason to check the success
On Thu, 28 May 2015 18:09:44 +0200
Luca Barbato lu_z...@gentoo.org wrote:
On 28/05/15 16:26, wm4 wrote:
This was probably broken some time ago. The breakage is now part of the
ABI. For example, we have:
Fine by me, this kind of thing is too brittle, do we have a macro
relying on it?
I
On Thu, 28 May 2015 18:37:00 +0200
Luca Barbato lu_z...@gentoo.org wrote:
On 28/05/15 18:23, wm4 wrote:
Haven't really checked whether this really prevents SIGPIPE (too
bothersome to test), only that the function call succeeds. I don't
think there's a reason to check the success normally
On Wed, 27 May 2015 16:41:10 +0100
Vittorio Giovara vittorio.giov...@gmail.com wrote:
From: wm4 nfx...@googlemail.com
Apparently it can happen that a mp3 file has junk data between id3 tag
and actual mp3 data. Skip this to avoid outputting nonsense timestamps.
(Two packets had the same
On Tue, 26 May 2015 14:24:38 +0100
Vittorio Giovara vittorio.giov...@gmail.com wrote:
---
libavformat/avio_internal.h | 7 +++
libavformat/aviobuf.c | 8
2 files changed, 15 insertions(+)
diff --git a/libavformat/avio_internal.h b/libavformat/avio_internal.h
index
There is no need to have this mess in network.c.
---
libavformat/network.c | 91 +--
libavformat/tls.h | 6
libavformat/tls_gnutls.c | 30 ++--
libavformat/tls_openssl.c | 68 +--
4 files
Move the OpenSSL and GnuTLS implementations to their own files. Other
than the connection code (including options) and some boilerplate, no
code is actually shared.
---
configure | 4 +-
libavformat/Makefile | 3 +-
libavformat/allformats.c | 3 +-
On Sun, 24 May 2015 21:21:07 +0200
Luca Barbato lu_z...@gentoo.org wrote:
On 24/05/15 17:17, Kieran Kunhya wrote:
On 24 May 2015 at 15:20, Luca Barbato lu_z...@gentoo.org wrote:
Should work on top of the hap set.
I'll try to get a not-completely-horrible encoder later.
What does
On Fri, 22 May 2015 22:25:53 +0300 (EEST)
Martin Storsjö mar...@martin.st wrote:
On Fri, 22 May 2015, wm4 wrote:
Move the OpenSSL and GnuTLS implementations to their own files. Other
than the connection code (including options) and some boilerplate, no
code is actually shared.
Well
On Fri, 22 May 2015 09:40:49 +0300
Martin Storsjö mar...@martin.st wrote:
From: James Almer jamr...@gmail.com
---
While such a configuration probably doesn't make much sense, we
shouldn't fail to compile there.
---
libavformat/tls.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
Move the OpenSSL and GnuTLS implementations to their own files. Other
than the connection code (including options) and some boilerplate, no
code is actually shared.
---
There is a minor change in behavior: now the equivalent of TLS_shutdown
is called both on normal closing, and closing on the
There is no need to have this mess in network.c.
---
libavformat/network.c | 83 +++
libavformat/tls_common.h | 6
libavformat/tls_gnutls.c | 26 +++
libavformat/tls_openssl.c | 64
4 files
On Thu, 21 May 2015 16:09:01 +0300
Martin Storsjö mar...@martin.st wrote:
---
libavformat/tls.c | 83
++-
1 file changed, 27 insertions(+), 56 deletions(-)
diff --git a/libavformat/tls.c b/libavformat/tls.c
index f768653..36ca754a
On Thu, 21 May 2015 16:08:59 +0300
Martin Storsjö mar...@martin.st wrote:
This avoids hijacking the fd, by reading using the normal
URLContext functions instead. This allowing reading data that has
been buffered in the underlying URLContext.
This avoids using the libraries own send
On Thu, 21 May 2015 16:09:00 +0300
Martin Storsjö mar...@martin.st wrote:
Since the underlying URLContext read functions are used,
they handle interruption, without having to handle it at
this level.
---
libavformat/tls.c | 35 +++
1 file changed, 11
On Thu, 21 May 2015 22:07:39 +0300 (EEST)
Martin Storsjö mar...@martin.st wrote:
On Thu, 21 May 2015, Martin Storsjö wrote:
On Thu, 21 May 2015, wm4 wrote:
On Thu, 21 May 2015 16:08:59 +0300
Martin Storsjö mar...@martin.st wrote:
This avoids hijacking the fd, by reading using
Personally, I need the decoder to back out if get_format() returns no
usable pixel format. This didn't work because the error code was not
propagated down the call chain. This in turn happened because the
variable declaration removed in this patch shadowed the variable, whose
value is returned at
On Thu, 14 May 2015 19:30:57 +0200
Luca Barbato lu_z...@gentoo.org wrote:
Just setting data to NULL and size to 0 is not safe because of side
data.
Reported-By: Bernd Pfrommer
---
libavcodec/avcodec.h | 10 ++
1 file changed, 6 insertions(+), 4 deletions(-)
diff --git
On Sun, 10 May 2015 13:46:36 +0200
Luca Barbato lu_z...@gentoo.org wrote:
On 10/05/15 12:08, wm4 wrote:
what do you think? (Personally, I think it's rather questionable. But
on the other hand it's inconvenient for users, and I'm trying to avoid
format-specific hacks in generic code
On Fri, 8 May 2015 15:43:11 +0100
Vittorio Giovara vittorio.giov...@gmail.com wrote:
From: wm4 nfx...@googlemail.com
APIC tags always have a description. Tag writers obviously leave it
empty if there is no description. In this case, libavformat would export
as title. Do not set the title
On Sat, 9 May 2015 21:32:21 +0200
Luca Barbato lu_z...@gentoo.org wrote:
8 flags are not enough.
---
I need an additional flags and 8 are already in use.
I'd land this patch as we are going to break the ABI soon anyway.
libavutil/pixdesc.h | 21 +
1 file changed,
On Sun, 10 May 2015 20:34:47 +0300
Rémi Denis-Courmont r...@remlab.net wrote:
Le dimanche 10 mai 2015, 19:15:24 wm4 a écrit :
GnuTLS uses writev() to write to the socket without giving MSG_NOSIGNAL.
Should that not be fixed upstream? Do they specifically want to support non-
socket file
On Sun, 10 May 2015 20:38:09 +0300
Rémi Denis-Courmont r...@remlab.net wrote:
Le dimanche 10 mai 2015, 19:15:25 wm4 a écrit :
A low-effort fix for fixing SIGPIPE being raised in some situations
because OpenSSL uses a naive write() to send data to the socket without
using MSG_NOSIGNAL
A low-effort fix for fixing SIGPIPE being raised in some situations
because OpenSSL uses a naive write() to send data to the socket without
using MSG_NOSIGNAL.
The idea and code for blocking and unqueuing SIGPIPE is taken from:
GnuTLS uses writev() to write to the socket without giving MSG_NOSIGNAL.
---
libavformat/tls.c | 8
1 file changed, 8 insertions(+)
diff --git a/libavformat/tls.c b/libavformat/tls.c
index d5de5ee..4a52e84 100644
--- a/libavformat/tls.c
+++ b/libavformat/tls.c
@@ -137,6 +137,13 @@
On Fri, 8 May 2015 09:02:23 +
John Högberg john.hogb...@ericsson.com wrote:
Hi again,
I've changed my approach a bit to make it more general; instead of adding a
carrier type field to AVStream, I've added support for raising any event
through omission. Formats set `base_event_flags`
On Tue, 5 May 2015 14:19:44 +0100
Vittorio Giovara vittorio.giov...@gmail.com wrote:
On Sat, May 2, 2015 at 5:50 AM, Anton Khirnov an...@khirnov.net wrote:
Quoting Vittorio Giovara (2015-05-02 01:17:17)
Add an helper function and a compatibility layer for exporting
information through
On Wed, 29 Apr 2015 15:28:50 +0100
Vittorio Giovara vittorio.giov...@gmail.com wrote:
how about avplay: Always initialize diff
Proper bikeshed:
avplay: fix use of uninitialized variable
Because someone reading the commit message has no idea what diff is.
On Wed, 15 Apr 2015 15:04:32 +0100
Vittorio Giovara vittorio.giov...@gmail.com wrote:
Fix a rather lenghty warning from clang.
---
libavcodec/vc1_pred.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/libavcodec/vc1_pred.c b/libavcodec/vc1_pred.c
index
This is optional, but ensures that linking with -Wl,--as-needed does not
drop the library containing the MMAL VC driver. The driver normally
registers itself in the library constructor, but since no symbols are
explicitly referenced, the linker could remove it with as-needed
enabled.
---
At first
---
Requested by Diego some time ago.
---
libavcodec/mmaldec.c | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/libavcodec/mmaldec.c b/libavcodec/mmaldec.c
index c59a496..1f29d2b 100644
--- a/libavcodec/mmaldec.c
+++ b/libavcodec/mmaldec.c
@@ -24,6 +24,12 @@
*
This seems to happen sometimes with old .avi files. They are just
confusing and don't help anyone and use sloppy language.
---
libavcodec/mpeg4videodec.c | 5 -
1 file changed, 5 deletions(-)
diff --git a/libavcodec/mpeg4videodec.c b/libavcodec/mpeg4videodec.c
index 0966e07..518d117 100644
On Fri, 10 Apr 2015 23:40:30 +0530
Himangi Saraogi himangi...@gmail.com wrote:
---
libavcodec/faxcompr.c | 2 +-
libavformat/ac3dec.c | 2 +-
libavformat/mp3dec.c | 2 +-
libavformat/rtmppkt.c | 3 ++-
4 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/libavcodec/faxcompr.c
On Sat, 14 Mar 2015 21:41:13 +0530
Himangi Saraogi himangi...@gmail.com wrote:
---
libavformat/ac3dec.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavformat/ac3dec.c b/libavformat/ac3dec.c
index 4ceffa5..38a2a8f 100644
--- a/libavformat/ac3dec.c
+++
On Wed, 1 Apr 2015 01:38:29 +0200
Vittorio Giovara vittorio.giov...@gmail.com wrote:
This value is to be used to reflect the real intentions of the developer
regarding a correct error reporting.
---
In the codebase there are a couple of places were this is evident, and this
error kind will
On Fri, 27 Mar 2015 23:04:08 +0100
wm4 nfx...@googlemail.com wrote:
Based on a patch by Rodger Combs.
---
Add elenril's and Martin's suggestions.
---
configure | 13 +
libavcodec/Makefile| 1 +
libavcodec/allcodecs.c | 2 +
libavcodec/mmaldec.c | 770
On Sat, 28 Mar 2015 10:04:14 -0600
John Stebbins stebb...@jetheaddev.com wrote:
On 03/21/2015 10:28 AM, John Stebbins wrote:
On 03/21/2015 09:45 AM, wm4 wrote:
On Sat, 21 Mar 2015 09:20:40 -0600
John Stebbins stebb...@jetheaddev.com wrote:
On 03/21/2015 06:30 AM, wm4 wrote:
On Fri
Based on a patch by Rodger Combs.
---
Add elenril's and Martin's suggestions.
---
configure | 13 +
libavcodec/Makefile| 1 +
libavcodec/allcodecs.c | 2 +
libavcodec/mmaldec.c | 770 +
libavutil/pixdesc.c| 4 +
On Fri, 27 Mar 2015 11:54:46 +0200 (EET)
Martin Storsjö mar...@martin.st wrote:
On Thu, 26 Mar 2015, wm4 wrote:
On Thu, 26 Mar 2015 10:18:14 +0200 (EET)
Martin Storsjö mar...@martin.st wrote:
On Tue, 24 Mar 2015, wm4 wrote:
Based on a patch by Rodger Combs.
---
Changed
On Thu, 26 Mar 2015 10:18:14 +0200 (EET)
Martin Storsjö mar...@martin.st wrote:
On Tue, 24 Mar 2015, wm4 wrote:
Based on a patch by Rodger Combs.
---
Changed the way the configure test works. Now it will first run a test
without adding the insane RPI driver include and library paths
Based on a patch by Rodger Combs.
---
Implement format changes, some other fixes.
---
configure | 11 +
libavcodec/Makefile| 1 +
libavcodec/allcodecs.c | 2 +
libavcodec/mmaldec.c | 759 +
libavutil/pixdesc.c| 4 +
Based on a patch by Rodger Combs.
---
Changed the way the configure test works. Now it will first run a test
without adding the insane RPI driver include and library paths (which
are usually missing from the compiler's standard paths).
Added a call to bcm_host_init() on init.
Fixed some minor
On Sat, 21 Mar 2015 09:20:40 -0600
John Stebbins stebb...@jetheaddev.com wrote:
On 03/21/2015 06:30 AM, wm4 wrote:
On Fri, 20 Mar 2015 16:47:57 -0600
John Stebbins stebb...@jetheaddev.com wrote:
Some subtitles (e.g. PGS) do not require a duration since they have
explicit end
On Fri, 20 Mar 2015 16:47:57 -0600
John Stebbins stebb...@jetheaddev.com wrote:
Some subtitles (e.g. PGS) do not require a duration since they have
explicit end-of-subtitle indication in the stream. This provides
a way to omit the unnecessary duration while muxing.
---
On Fri, 20 Mar 2015 22:23:36 +0200 (EET)
Martin Storsjö mar...@martin.st wrote:
On Fri, 20 Mar 2015, wm4 wrote:
From: wm4 wm4@nowhere
Based on a patch by Rodger Combs.
---
Pretty messy at times because the decoder is asynchronous and buffer
management is messy.
Don't hit me
From: wm4 wm4@nowhere
Based on a patch by Rodger Combs.
---
Pretty messy at times because the decoder is asynchronous and buffer
management is messy.
Don't hit me for the crap in require_mmal() - everyone does this on this
platform. The -isystem and -fgnu89-inline flags are to evade warnings
On Fri, 20 Mar 2015 22:52:29 +0200 (EET)
Martin Storsjö mar...@martin.st wrote:
On Fri, 20 Mar 2015, wm4 wrote:
On Fri, 20 Mar 2015 22:23:36 +0200 (EET)
Martin Storsjö mar...@martin.st wrote:
On Fri, 20 Mar 2015, wm4 wrote:
From: wm4 wm4@nowhere
Based on a patch by Rodger
On Thu, 12 Mar 2015 18:36:51 -0400
Bradley Sepos brad...@bradleysepos.com wrote:
Updated version of patch by Carl Eugen Hoyos:
https://patches.libav.org/patch/42533/
https://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=79209f5d6ca31c73260f1c5f5dbaa8395102d9f1
Works in limited testing.
On Tue, 10 Mar 2015 01:28:53 +
Vittorio Giovara vittorio.giov...@gmail.com wrote:
This field is designed for marking codec properties useful to lavc internals.
Two internal capabilities are added:
- CODEC_INT_CAP_THREAD_SAFE: codec can be opened without locks;
-
On Tue, 24 Feb 2015 15:33:53 +
Vittorio Giovara vittorio.giov...@gmail.com wrote:
Enabled only in debug mode and only when run under Valgrind.
Signed-off-by: Vittorio Giovara vittorio.giov...@gmail.com
---
This is a neat trick I found reading
On Fri, 20 Feb 2015 23:50:47 +0200 (EET)
Martin Storsjö mar...@martin.st wrote:
On Fri, 20 Feb 2015, wm4 wrote:
On Fri, 20 Feb 2015 22:24:51 +0200
Martin Storsjö mar...@martin.st wrote:
Since 4abfa387b, we insert index entries based on composition timestamp,
not
dts
On Fri, 20 Feb 2015 22:24:51 +0200
Martin Storsjö mar...@martin.st wrote:
Since 4abfa387b, we insert index entries based on composition timestamp, not
dts, and these aren't necessarily monotonous. av_add_index_entry does a sorted
insert (instead of appending entries at the end), while the mov
On Thu, 12 Feb 2015 14:31:10 +0100
Hendrik Leppkes h.lepp...@gmail.com wrote:
On Thu, Feb 12, 2015 at 10:54 AM, Anton Khirnov an...@khirnov.net wrote:
---
avconv.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/avconv.c b/avconv.c
index 8e76863..6349cae 100644
--- a/avconv.c
On Tue, 10 Feb 2015 14:56:05 -0500
Vittorio Giovara vittorio.giov...@gmail.com wrote:
On Tue, Feb 10, 2015 at 2:45 PM, Luca Barbato lu_z...@gentoo.org wrote:
On 10/02/15 20:43, Vittorio Giovara wrote:
On Tue, Feb 10, 2015 at 2:36 PM, Alex Converse alex.conve...@gmail.com
wrote:
On
On Sun, 08 Feb 2015 21:33:26 +0100
Anton Khirnov an...@khirnov.net wrote:
Quoting wm4 (2015-02-06 14:53:40)
diff --git a/libavformat/avformat.h b/libavformat/avformat.h
index df9a602..a0556e8 100644
--- a/libavformat/avformat.h
+++ b/libavformat/avformat.h
@@ -1210,58 +1210,10
On Fri, 6 Feb 2015 18:38:44 +
Vittorio Giovara vittorio.giov...@gmail.com wrote:
Related to warning: suggest braces around initialization of subobject.
---
libavcodec/dct-test.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libavcodec/dct-test.c
Currently, all pixel formats which support mor than 8 bits per
component have a big endian and little endian variant, e.g.
AV_PIX_FMT_RGBA64LE and AV_PIX_FMT_RGBA64BE. There are also
native aliases defined to the compile-time endian of the
platform, like AV_PIX_FMT_RGBA64.
I'm arguing that the BE
A avcodec_copy_context() function exists, but it also sets the codec
and wipes the complete destination context. It does codec-specific
initialization like avcodec_alloc_context3() does (like allocating the
private context, so decoder-specific options can be set).
This is very inconvenient if you
This is a proposal for an API extension.
Currently, some pixel formats support alpha, but whether the alpha
component contains something useful or just garbage is not part of the
pixel format definition. This applies at least to packed RGB formats,
where the 4th component is either alpha or
Nothing uses it, and it provides no public API.
Archeological finds:
Commit 101036adb9 added the API.
Commit a8dd8dc6e9 made mpegts.c use it.
Commit af8aae3fa3 disabled it by default in mpegts.c.
Commit ae2bb52cd2 removed all uses of this from mpegts.c.
---
libavformat/Makefile | 1 -
This is not an API change; the fields were explicitly declared private
before.
---
libavformat/asfdec.c | 10 +++
libavformat/avformat.h | 50 +---
libavformat/dv.c | 4 +--
libavformat/internal.h | 41 +++
libavformat/mtv.c |
On Fri, 6 Feb 2015 13:51:48 +
Vittorio Giovara vittorio.giov...@gmail.com wrote:
On Fri, Feb 6, 2015 at 11:32 AM, wm4 nfx...@googlemail.com wrote:
This is a proposal for an API extension.
Currently, some pixel formats support alpha, but whether the alpha
component contains something
On Fri, 16 Jan 2015 16:59:13 +0900
김건수 gunsoo83@samsung.com wrote:
Let me explain our user scenario.
Libavcodec will be distributed with our application.
When the application starts to play a video, it will check the availability
of libva.
If libva is possible to use, the
On Tue, 2 Dec 2014 16:48:25 +
Vittorio Giovara vittorio.giov...@gmail.com wrote:
From: Gonzalo Garramuno ggarr...@gmail.com
This is needed to avoid banding artifacts when gammaing the picture.
Currently, if done with a video filter, the process is done on uints
instead of full float.
---
This is a bit confusing. This change is not included in Libav 11:
2014-08-xx - xxx - lavf 56.03.0 - avformat.h
Add AVFormatContext.max_ts_probe.
while this one is:
2014-08-28 - 9301486 - lavc 56.1.0 - avcodec.h
Add AV_PKT_DATA_STEREO3D to export container-level stereo3d
---
I did not expect that it would crash. And why should it? So just make it work.
---
libavcodec/utils.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/libavcodec/utils.c b/libavcodec/utils.c
index 8cbd47b..e2d9096 100644
--- a/libavcodec/utils.c
+++ b/libavcodec/utils.c
@@ -2079,6
On Mon, 01 Dec 2014 14:09:20 +0100
Anton Khirnov an...@khirnov.net wrote:
Quoting wm4 (2014-12-01 13:53:29)
---
I did not expect that it would crash. And why should it? So just make it
work.
---
libavcodec/utils.c | 3 +++
1 file changed, 3 insertions(+)
diff --git
On Mon, 1 Dec 2014 13:56:57 +0100
Nicolas George geo...@nsup.org wrote:
Le primidi 11 frimaire, an CCXXIII, wm4 a écrit :
I did not expect that it would crash. And why should it?
Because it does not make any sense.
I used it for making the decoder release all surfaces before
uninitializing
, doing nothing is ok by definition.
Based on a patch by wm4 nfx...@googlemail.com.
---
libavutil/thread.h | 50 ++
1 file changed, 50 insertions(+)
create mode 100644 libavutil/thread.h
diff --git a/libavutil/thread.h b/libavutil/thread.h
On Fri, 14 Nov 2014 11:38:09 +0100
Anton Khirnov an...@khirnov.net wrote:
Quoting wm4 (2014-11-11 17:26:47)
The buffer pool has to atomically add and remove entries from the linked
list of available buffers. This was done by removing the entire list
with a CAS operation, working
On Fri, 14 Nov 2014 11:42:32 +0100
Anton Khirnov an...@khirnov.net wrote:
Mention explicitly that the array must be filled contiguously.
---
libavutil/frame.h | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/libavutil/frame.h b/libavutil/frame.h
index
This helps keeping the code clean if Libav is compiled for targets
without threading. Since we assume that no threads of any kind are used
in such configurations, doing nothing is ok by definition.
---
Makefile| 2 +-
compat/nopthreads.h | 53
The buffer pool has to atomically add and remove entries from the linked
list of available buffers. This was done by removing the entire list
with a CAS operation, working on it, and then setting it back again
(using a retry-loop in case another thread was doing the same thing).
This could
On Fri, 14 Nov 2014 20:44:59 +0100
Anton Khirnov an...@khirnov.net wrote:
Quoting wm4 (2014-11-14 13:10:13)
On Fri, 14 Nov 2014 11:42:32 +0100
Anton Khirnov an...@khirnov.net wrote:
Mention explicitly that the array must be filled contiguously.
---
libavutil/frame.h | 4
The buffer pool has to atomically add and remove entries from the linked
list of available buffers. This was done by removing the entire list
with a CAS operation, working on it, and then setting it back again
(using a retry-loop in case another thread was doing the same thing).
This could
:32, wm4 wrote:
FFmpeg removed this bad design of writing the timestamps into the ASS
packet, so demuxing mkv files longer than 10 hours and with ASS
subtitles would work just fine, as far as I can see. (Although I would
wonder how you'd create such a file, since all tools probably read
On Mon, 03 Nov 2014 12:32:13 +0100
Anton Khirnov an...@khirnov.net wrote:
Quoting Luca Barbato (2014-10-29 16:55:31)
On 29/10/14 16:46, Anton Khirnov wrote:
It is required by (at least) the ogg demuxer.
Mark the current semi-public apriv API for removal.
---
doc/APIchanges
On Wed, 29 Oct 2014 18:40:27 -0700
John Stebbins stebb...@jetheaddev.com wrote:
On 10/29/2014 03:30 PM, wm4 wrote:
On Wed, 29 Oct 2014 11:06:05 -0700
John Stebbins stebb...@jetheaddev.com wrote:
The matroska spec says blockduration == 0 means the frame is not a
keyframe. Since all
On Wed, 29 Oct 2014 11:06:05 -0700
John Stebbins stebb...@jetheaddev.com wrote:
The matroska spec says blockduration == 0 means the frame is not a
keyframe. Since all subtitles are keyframes, 0 blockduration should
not be written.
Fixes mkvalidator error messages for PGS subtitles.
---
On Wed, 29 Oct 2014 20:39:46 +0100
Luca Barbato lu_z...@gentoo.org wrote:
CC: libav-sta...@libav.org
---
It is surprising that ASS supports only ~10 hours movies.
libavformat/matroskaenc.c | 21 +++--
1 file changed, 19 insertions(+), 2 deletions(-)
diff --git
On Wed, 29 Oct 2014 23:40:21 +0100
Luca Barbato lu_z...@gentoo.org wrote:
On 29/10/14 23:30, wm4 wrote:
Even mkvmerge writes ASS subtitles with duration 0 with an explicit
blockduration element:
| + Block group at 6289
| + Block (track number 1, 1 frame(s), timecode 60.000s = 00:01
On Wed, 10 Sep 2014 20:38:45 +0300
Rémi Denis-Courmont r...@remlab.net wrote:
Since the VDPAU pixel format does not distinguish between different
VDPAU video surface chroma types, we need another way to pass this
datum to the application.
Until now, VDPAU in libavcodec has only supported
On Wed, 10 Sep 2014 21:48:14 +0300
Rémi Denis-Courmont r...@remlab.net wrote:
Le mercredi 10 septembre 2014, 20:39:50 wm4 a écrit :
I think it would be better in terms of backwards compatibility to have
the user call a function to signal support for formats other than
VDP_CHROMA_TYPE_420
On Sat, 06 Sep 2014 18:59:30 +0300
Rémi Denis-Courmont r...@remlab.net wrote:
Le samedi 6 septembre 2014, 17:28:59 wm4 a écrit :
Specifically as regards preemption, I think it remains doable. In
principles, this makes no real difference; you can still wrap the
callback in a similarly
On Sun, 07 Sep 2014 22:25:06 +0300
Rémi Denis-Courmont r...@remlab.net wrote:
Le dimanche 7 septembre 2014, 13:16:32 wm4 a écrit :
Your API design pretends preemption doesn't exist. You're ignoring a
major aspect of the vdpau API. I don't see how this is reasonable.
That is not true. I
On Sat, 06 Sep 2014 18:10:17 +0300
Rémi Denis-Courmont r...@remlab.net wrote:
Hello,
Le vendredi 5 septembre 2014, 18:13:49 wm4 a écrit :
+static VdpStatus vdp_enosys(/* Unspecified, not void - leave empty */)
+{
+return VDP_STATUS_NO_IMPLEMENTATION;
+}
+
+static void
On Thu, 4 Sep 2014 16:11:15 +0300
Rémi Denis-Courmont r...@remlab.net wrote:
From: Rémi Denis-Courmont re...@nvidia.com
Using the not so new init and uninit callbacks, avcodec can now take
care of creating and destroying the VDPAU decoder instance.
The application is still responsible
On Fri, 29 Aug 2014 01:00:45 +0200
Diego Biurrun di...@biurrun.de wrote:
On Sun, Aug 24, 2014 at 12:28:56AM +0200, Clément Bœsch wrote:
Kieran suggested tonight on #ffmpeg-devel to have a common mailing-list
between the two projects to start communicating again in sane terms.
The
On Sun, 24 Aug 2014 00:28:56 +0200
Clément Bœsch u...@pkh.me wrote:
Hi,
Kieran suggested tonight on #ffmpeg-devel to have a common mailing-list
between the two projects to start communicating again in sane terms.
The proposition would be a mailing-list where the 2 projects would send
the
On Sun, 24 Aug 2014 13:00:07 +
Anton Khirnov an...@khirnov.net wrote:
Hi,
I hereby present my entry to this year's ridiculously oversized patchset
competition.
The theme this time is further decoupling of lavf and lavc. Currently,
AVStream
has an embedded AVCodecContext, which is
On Sat, 23 Aug 2014 10:09:28 -0400
Reinhard Tartler siret...@tauware.de wrote:
This option facilitates testing shared libarary builds: for instance
fate builders do no longer need to set LD_LIBRARY_PATH as the binaries will
get the right search paths hardcoded into their executable file.
On Wed, 13 Aug 2014 22:35:47 +0200
Diego Biurrun di...@biurrun.de wrote:
---
Ahem ...
I'll still charge Anton a Guinness, but invite Justin to another :)
libavutil/frame.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/libavutil/frame.c b/libavutil/frame.c
index
On Tue, 12 Aug 2014 16:54:50 +
Anton Khirnov an...@khirnov.net wrote:
---
doc/APIchanges | 3 +++
libavutil/mem.c | 20
libavutil/mem.h | 10 ++
libavutil/version.h | 2 +-
4 files changed, 34 insertions(+), 1 deletion(-)
diff --git
On Sat, 9 Aug 2014 21:22:13 +0300 (EEST)
Martin Storsjö mar...@martin.st wrote:
On Thu, 31 Jul 2014, Andrew Stone wrote:
It won't hurt servers that don't care about the header,
and those that do will include it by default.
---
Changelog | 1 +
doc/protocols.texi | 2 +-
On Sat, 09 Aug 2014 10:26:16 -0400
Justin Ruggles justin.rugg...@gmail.com wrote:
On 08/02/2014 10:56 AM, Luca Barbato wrote:
---
+int avresample_config(AVAudioResampleContext *avr, AVFrame *out, AVFrame
*in,
+ AVDictionary **opts)
+{
+int ret;
+
+
601 - 700 of 810 matches
Mail list logo
