On Mon, 14 Jan 2013 05:32:39 +0100, Luca Barbato wrote:
> By default libav sets it to 3/4 while x264 sets it to 9/10.
>
> CC: libav-sta...@libav.org
> ---
> libavcodec/libx264.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/libavcodec/libx264.c b/libavcodec/libx264
On Mon, 14 Jan 2013 23:49:36 +0100, Diego Biurrun wrote:
> On Mon, Jan 14, 2013 at 10:28:57PM +0100, Alexandra Khirnova wrote:
> > @@ -484,10 +499,54 @@ static int dvdsub_decode(AVCodecContext *avctx,
> > +
> > +if (ctx->has_palette) {
> > +int i;
> > +av_log(avctx, AV_LOG_DEB
On Sun, 13 Jan 2013, Diego Biurrun wrote:
> ---
>
> Now using local variables for HIDDEN and FUNCTION_PREFIX, as suggested
> by Loren.
>
> libavutil/x86/x86inc.asm | 40
> libavutil/x86/x86util.asm |1 +
> 2 files changed, 29 insertions(+), 12 delet
On 01/14/2013 04:28 PM, Diego Biurrun wrote:
> ---
>
> Added comments about macro arguments as requested by Justin.
>
> libavutil/x86/x86util.asm | 11 +--
> 1 files changed, 5 insertions(+), 6 deletions(-)
>
> diff --git a/libavutil/x86/x86util.asm b/libavutil/x86/x86util.asm
> index
On 01/14/2013 04:29 PM, Diego Biurrun wrote:
> ---
>
> Added macro argument comments as requested by Justin.
>
> libavutil/x86/x86util.asm | 14 ++
> 1 files changed, 6 insertions(+), 8 deletions(-)
>
> diff --git a/libavutil/x86/x86util.asm b/libavutil/x86/x86util.asm
> index 474
On 15/01/13 00:39, Diego Biurrun wrote:
> ---
>
> This is likely incorrect, but I'm sending it nonetheless to troll
> somebody into coming up with a better fix. Right now this is
> triggering build failures on some FATE configs due to not always
> returning a value from a non-void function.
>
A
On Tue, Jan 15, 2013 at 01:08:17AM +0100, Diego Biurrun wrote:
> These flags are as linker-specific as other LDFLAGS and thus
> need to be translated to the correct linker syntax.
> ---
>
> Tested to work and more correct than previous attempts to fix the issue.
Disregard, I need to sleep instead
---
configure |3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/configure b/configure
index e422b60..1ea812e 100755
--- a/configure
+++ b/configure
@@ -3596,6 +3596,7 @@ check_cflags -Wstrict-prototypes
enabled extra_warnings && check_cflags -Winline
# add some linker
These flags are as linker-specific as other LDFLAGS and thus
need to be translated to the correct linker syntax.
---
Tested to work and more correct than previous attempts to fix the issue.
configure |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/configure b/configure
i
On Mon, Jan 14, 2013 at 11:38:00PM +0100, Diego Biurrun wrote:
> From: Sean McGovern
>
> These flags are as linker-specific as other LDFLAGS and thus need
> to be translated to the correct syntax for each linker.
>
> Signed-off-by: Diego Biurrun
> ---
>
> Simplified version that avoids the ext
On 14/01/13 19:12, Martin Storsjö wrote:
> This is mostly useful for encryption together with the RTP muxer,
> but could also be set up as IO towards the peer with the SDP
> demuxer with custom IO.
> ---
> libavformat/Makefile |1 +
> libavformat/allformats.c |1 +
> libavformat/srtppr
On 15/01/13 00:39, Diego Biurrun wrote:
> ---
>
> This is likely incorrect, but I'm sending it nonetheless to troll
> somebody into coming up with a better fix. Right now this is
> triggering build failures on some FATE configs due to not always
> returning a value from a non-void function.
>
>
---
This is likely incorrect, but I'm sending it nonetheless to troll
somebody into coming up with a better fix. Right now this is
triggering build failures on some FATE configs due to not always
returning a value from a non-void function.
libavformat/nutdec.c |2 +-
1 files changed, 1 inse
On Mon, Jan 14, 2013 at 10:28:57PM +0100, Alexandra Khirnova wrote:
> From: Joakim Plate
>
> This is how Matroska store palette information for vobsubs
storeS
> --- a/libavcodec/dvdsubdec.c
> +++ b/libavcodec/dvdsubdec.c
> @@ -25,6 +25,11 @@
>
> +typedef struct DVDSubContext {
> + uint32_t p
From: Sean McGovern
These flags are as linker-specific as other LDFLAGS and thus need
to be translated to the correct syntax for each linker.
Signed-off-by: Diego Biurrun
---
Simplified version that avoids the extra echo call.
configure |2 +-
1 files changed, 1 insertions(+), 1 deletion
On Mon, Jan 14, 2013 at 11:51:01PM +0200, Martin Storsjö wrote:
> On Mon, 14 Jan 2013, Diego Biurrun wrote:
> >On Mon, Jan 14, 2013 at 08:12:44PM +0200, Martin Storsjö wrote:
> >>This is mostly useful for encryption together with the RTP muxer,
> >>but could also be set up as IO towards the peer wi
On Mon, 14 Jan 2013, Diego Biurrun wrote:
On Mon, Jan 14, 2013 at 08:12:44PM +0200, Martin Storsjö wrote:
This is mostly useful for encryption together with the RTP muxer,
but could also be set up as IO towards the peer with the SDP
demuxer with custom IO.
---
libavformat/Makefile |1 +
On Mon, 14 Jan 2013, Diego Biurrun wrote:
On Mon, Jan 14, 2013 at 09:45:10PM +0200, Martin Storsjö wrote:
--- /dev/null
+++ b/libavutil/hmac.h
@@ -0,0 +1,92 @@
+
+/**
+ * Allocate an AVHMAC context.
+ * @param type The hash function used for the HMAC.
+ */
+AVHMAC *av_hmac_alloc(enum AVHMACType
On Mon, 14 Jan 2013, Diego Biurrun wrote:
On Mon, Jan 14, 2013 at 08:12:42PM +0200, Martin Storsjö wrote:
--- /dev/null
+++ b/libavformat/srtp.c
@@ -0,0 +1,291 @@
+
+void ff_srtp_free(struct SRTPContext *s)
+{
+av_freep(&s->aes);
+if (s->hmac)
+av_hmac_free(s->hmac);
+s->hm
On Mon, Jan 14, 2013 at 09:45:10PM +0200, Martin Storsjö wrote:
> --- /dev/null
> +++ b/libavutil/hmac.h
> @@ -0,0 +1,92 @@
> +
> +/**
> + * Allocate an AVHMAC context.
> + * @param type The hash function used for the HMAC.
> + */
> +AVHMAC *av_hmac_alloc(enum AVHMACType type);
> +/**
> + * Free an
On Mon, Jan 14, 2013 at 08:12:44PM +0200, Martin Storsjö wrote:
> This is mostly useful for encryption together with the RTP muxer,
> but could also be set up as IO towards the peer with the SDP
> demuxer with custom IO.
> ---
> libavformat/Makefile |1 +
> libavformat/allformats.c |1
On Mon, 14 Jan 2013, Diego Biurrun wrote:
On Mon, Jan 14, 2013 at 08:12:41PM +0200, Martin Storsjö wrote:
This supports HMAC-MD5 and HMAC-SHA1 for now, other hashes are
simple to add.
---
.. nits as agreed to on IRC ..
--- a/doc/APIchanges
+++ b/doc/APIchanges
@@ -13,6 +13,9 @@ libavutil:
On Mon, Jan 14, 2013 at 08:12:42PM +0200, Martin Storsjö wrote:
>
> --- /dev/null
> +++ b/libavformat/srtp.c
> @@ -0,0 +1,291 @@
> +
> +void ff_srtp_free(struct SRTPContext *s)
> +{
> +av_freep(&s->aes);
> +if (s->hmac)
> +av_hmac_free(s->hmac);
> +s->hmac = NULL;
> +}
This do
On Mon, Jan 14, 2013 at 08:12:41PM +0200, Martin Storsjö wrote:
> This supports HMAC-MD5 and HMAC-SHA1 for now, other hashes are
> simple to add.
> ---
.. nits as agreed to on IRC ..
> --- a/doc/APIchanges
> +++ b/doc/APIchanges
> @@ -13,6 +13,9 @@ libavutil: 2012-10-22
>
> API changes, mo
---
Added macro argument comments as requested by Justin.
libavutil/x86/x86util.asm | 14 ++
1 files changed, 6 insertions(+), 8 deletions(-)
diff --git a/libavutil/x86/x86util.asm b/libavutil/x86/x86util.asm
index 4740b7f..9236991 100644
--- a/libavutil/x86/x86util.asm
+++ b/liba
From: Joakim Plate
This is how Matroska store palette information for vobsubs
Signed-off-by: Alexandra Khirnova
---
libavcodec/dvdsubdec.c | 69
1 file changed, 64 insertions(+), 5 deletions(-)
diff --git a/libavcodec/dvdsubdec.c b/libavcodec
---
Added comments about macro arguments as requested by Justin.
libavutil/x86/x86util.asm | 11 +--
1 files changed, 5 insertions(+), 6 deletions(-)
diff --git a/libavutil/x86/x86util.asm b/libavutil/x86/x86util.asm
index 9300faf..4740b7f 100644
--- a/libavutil/x86/x86util.asm
+++ b/
On Mon, 14 Jan 2013, Diego Biurrun wrote:
These warnings have no false positives and point to serious bugs.
---
Squashed both patches into one and dropped the clang options that have
no effect.
configure |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/configure b/confi
These warnings have no false positives and point to serious bugs.
---
Squashed both patches into one and dropped the clang options that have
no effect.
configure |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/configure b/configure
index 3d5fe7c..751d22e 100755
--- a/co
On 14/01/2013 1:28 PM, Derek Buitenhuis wrote:
>> Something along the lines of
>>
>> #if defined(_MSC_VER)
>> #define FMT %Iu
>> #else
>> #define FMT %zu
>> #endif
>>
>> and replacing %zu by FMT below seems simpler. Or is there something
>> I am missing?
>
> I wasn't able to get that to
On 14/01/13 21:47, Diego Biurrun wrote:
> The stupid thing is that -Werror=return-type has no effect with gcc,
> but works as advertised with clang. I'm inclined to think that adding
> all options for both compilers would give us maximum coverage, at least
> on FATE.
Has half effect on gcc, it is
Le lundi 14 janvier 2013 22:40:45, Diego Biurrun a écrit :
> > Umm, no. Sure, we don't have these annotations everywhere, and we
> > don't warn about it by default, but removing isn't necessary either.
>
> Why do you want to keep them in just a few select places? We probably
> have dozens or hund
On Mon, 14 Jan 2013, Diego Biurrun wrote:
On Mon, Jan 14, 2013 at 10:36:08PM +0200, Martin Storsjö wrote:
On Mon, 14 Jan 2013, Diego Biurrun wrote:
We do not enable warnings about unused function parameters, so
there is little point in trying to silence those warnings.
Besides, some fringe com
On Mon, Jan 14, 2013 at 10:31:27PM +0200, Martin Storsjö wrote:
> On Mon, 14 Jan 2013, Diego Biurrun wrote:
> >--- a/configure
> >+++ b/configure
> >@@ -3693,6 +3693,9 @@ elif enabled clang; then
> >check_cflags -Werror=implicit-function-declaration
> >check_cflags -Werror=missing-prototype
On Mon, Jan 14, 2013 at 10:36:08PM +0200, Martin Storsjö wrote:
> On Mon, 14 Jan 2013, Diego Biurrun wrote:
> >We do not enable warnings about unused function parameters, so
> >there is little point in trying to silence those warnings.
> >Besides, some fringe compilers choke on these attributes.
>
On Mon, 14 Jan 2013, Diego Biurrun wrote:
We do not enable warnings about unused function parameters, so
there is little point in trying to silence those warnings.
Besides, some fringe compilers choke on these attributes.
---
This is a more complete version of Rémi's patch.
libavcodec/dxva2_h2
We do not enable warnings about unused function parameters, so
there is little point in trying to silence those warnings.
Besides, some fringe compilers choke on these attributes.
---
This is a more complete version of Rémi's patch.
libavcodec/dxva2_h264.c |3 +--
libavcodec/dxva2_mpeg2.
On Mon, 14 Jan 2013, Diego Biurrun wrote:
---
configure |3 +++
1 file changed, 3 insertions(+)
diff --git a/configure b/configure
index 60d8fff..6306094 100755
--- a/configure
+++ b/configure
@@ -3693,6 +3693,9 @@ elif enabled clang; then
check_cflags -Qunused-arguments
check_cflags
On 14/01/13 20:45, Martin Storsjö wrote:
> This supports HMAC-MD5 and HMAC-SHA1 for now, other hashes are
> simple to add.
> ---
> Added a typedef for the context struct and added doxygen docs for
> the new public API, as requested by Anton, using proper AVERROR
> error codes as requested by Luca.
On 14/01/13 20:43, Diego Biurrun wrote:
> ---
> configure |3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/configure b/configure
> index 60d8fff..6306094 100755
> --- a/configure
> +++ b/configure
> @@ -3693,6 +3693,9 @@ elif enabled clang; then
> check_cflags -Qunused-argument
On 14/01/13 20:43, Diego Biurrun wrote:
> These warnings have no false positives and point to serious bugs.
> ---
> configure |1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/configure b/configure
> index 554d578..60d8fff 100755
> --- a/configure
> +++ b/configure
> @@ -3684,6 +3684,7
On 14/01/13 20:43, Martin Storsjö wrote:
> On Mon, 14 Jan 2013, Luca Barbato wrote:
>
>> Catch some more mistakes that result in crashes on SunCC.
>
> Build failures, not crashes, afaik (that is, the same as on clang).
>
Thanks for catching, amended.
___
On Monday, January 14, 2013, Martin Storsjö wrote:
> On Mon, 14 Jan 2013, Sean McGovern wrote:
>
>> Several of the options were incorrect for suncc.
>> Let's just shove the GCC options through
>> the pre-existing compiler flags filter.
>> ---
>> configure | 2 +-
>> 1 file changed, 1 insertion(+),
---
libavcodec/vdpau_internal.h |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libavcodec/vdpau_internal.h b/libavcodec/vdpau_internal.h
index 87c94c8..c740d94 100644
--- a/libavcodec/vdpau_internal.h
+++ b/libavcodec/vdpau_internal.h
@@ -34,8 +34,8 @@ static inline uin
On Mon, Jan 14, 2013 at 07:58:05PM +0200, Martin Storsjö wrote:
> On Mon, 14 Jan 2013, Sean McGovern wrote:
> >--- a/configure
> >+++ b/configure
> >@@ -2239,7 +2239,8 @@ suncc_flags(){
> >-fomit-frame-pointer) echo -xregs=frameptr;;
> >-fPIC)echo -KPIC -
This supports HMAC-MD5 and HMAC-SHA1 for now, other hashes are
simple to add.
---
Added a typedef for the context struct and added doxygen docs for
the new public API, as requested by Anton, using proper AVERROR
error codes as requested by Luca.
doc/APIchanges |3 ++
libavutil/Makefile
On Mon, 14 Jan 2013, Luca Barbato wrote:
Catch some more mistakes that result in crashes on SunCC.
Build failures, not crashes, afaik (that is, the same as on clang).
// Martin
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.o
These warnings have no false positives and point to serious bugs.
---
configure |1 +
1 file changed, 1 insertion(+)
diff --git a/configure b/configure
index 554d578..60d8fff 100755
--- a/configure
+++ b/configure
@@ -3684,6 +3684,7 @@ elif enabled gcc; then
check_cflags -Werror=implicit
---
configure |3 +++
1 file changed, 3 insertions(+)
diff --git a/configure b/configure
index 60d8fff..6306094 100755
--- a/configure
+++ b/configure
@@ -3693,6 +3693,9 @@ elif enabled clang; then
check_cflags -Qunused-arguments
check_cflags -Werror=implicit-function-declaration
On Mon, 14 Jan 2013 13:44:09 -0500, Justin Ruggles
wrote:
> ---
> libavfilter/vf_fps.c |1 +
> 1 files changed, 1 insertions(+), 0 deletions(-)
>
> diff --git a/libavfilter/vf_fps.c b/libavfilter/vf_fps.c
> index dc20114..8fd51bd 100644
> --- a/libavfilter/vf_fps.c
> +++ b/libavfilter/vf_f
Catch some more mistakes that result in crashes on SunCC.
Other compilers treat non-void return from void functions as error
by default.
---
configure | 1 +
1 file changed, 1 insertion(+)
diff --git a/configure b/configure
index 554d578..08d631d 100755
--- a/configure
+++ b/configure
@@ -3685,6
Hi,
On Mon, Jan 14, 2013 at 11:21 AM, Anton Khirnov wrote:
> It's been returning an error value since
> bad446e251405dc250c3cbee199072e083a1e4b9
>
> Also check for the errors it returns.
> ---
> libavcodec/h264.c |3 ++-
> libavcodec/h264.h |2 +-
> libavcodec/h264_refs.c | 1
On Mon, Jan 14, 2013 at 08:21:20PM +0100, Anton Khirnov wrote:
> It's been returning an error value since
> bad446e251405dc250c3cbee199072e083a1e4b9
>
> Also check for the errors it returns.
> ---
> libavcodec/h264.c |3 ++-
> libavcodec/h264.h |2 +-
> libavcodec/h264_refs.c |
It's been returning an error value since
bad446e251405dc250c3cbee199072e083a1e4b9
Also check for the errors it returns.
---
libavcodec/h264.c |3 ++-
libavcodec/h264.h |2 +-
libavcodec/h264_refs.c | 10 ++
3 files changed, 9 insertions(+), 6 deletions(-)
diff --git a
On 14/01/13 20:02, Martin Storsjö wrote:
> On Mon, 14 Jan 2013, Luca Barbato wrote:
>
>> On 14/01/13 19:12, Martin Storsjö wrote:
>>> This supports the AES_CM_128_HMAC_SHA1_80 and
>>> AES_CM_128_HMAC_SHA1_32 cipher suites (from RFC 4568) at the
>>> moment. The main missing features are replay prot
On Mon, 14 Jan 2013, Luca Barbato wrote:
On 14/01/13 19:12, Martin Storsjö wrote:
This supports the AES_CM_128_HMAC_SHA1_80 and
AES_CM_128_HMAC_SHA1_32 cipher suites (from RFC 4568) at the
moment. The main missing features are replay protection (which can be
added later without changing the int
On Fri, 11 Jan 2013, Luca Barbato wrote:
On 10/01/13 20:29, Jan Gerber wrote:
forwarding patch from ffmpeg-devel to fix theora video quality settings
from Maximilian Seesslen.
Seems ok, queued as well.
Pushed.
// Martin
___
libav-devel mailing
On Fri, 11 Jan 2013, Luca Barbato wrote:
On 10/01/13 20:34, Jan Gerber wrote:
another patch that should be applied to libav.
Seems ok, queued.
Didn't seem to be pushed yet, so pushed now.
// Martin
___
libav-devel mailing list
libav-devel@libav.
On 14/01/13 19:12, Martin Storsjö wrote:
> This only takes care of decrypting incoming packets; the outgoing
> RTCP packets are not encrypted. This is enough for some use cases,
> and signalling crypto keys for use with outgoing RTCP packets
> doesn't fit as simply into the API. If the SDP demuxer
On 14/01/2013 19:33, Diego Biurrun wrote:
> Changing these names could be considered an API/ABI break.
Aren't ff_* symbols considered internals?
--
Diego Elio Pettenò — Flameeyes
flamee...@flameeyes.eu — http://blog.flameeyes.eu/
___
libav-devel mailin
On 14/01/13 19:12, Martin Storsjö wrote:
> Also print port numbers for this protocol.
> ---
> libavformat/sdp.c | 15 ++-
> 1 file changed, 14 insertions(+), 1 deletion(-)
>
Ok, remind me to draft something for the documentation if you do not
have time.
lu
___
---
libavfilter/vf_fps.c |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/libavfilter/vf_fps.c b/libavfilter/vf_fps.c
index dc20114..8fd51bd 100644
--- a/libavfilter/vf_fps.c
+++ b/libavfilter/vf_fps.c
@@ -104,6 +104,7 @@ static av_cold void uninit(AVFilterContext *ctx)
{
On Mon, Jan 14, 2013 at 05:18:04PM +0100, Luca Barbato wrote:
>
> --- a/libavcodec/libvpxdec.c
> +++ b/libavcodec/libvpxdec.c
> @@ -117,7 +112,13 @@ static av_cold int vp8_free(AVCodecContext *avctx)
>
> -AVCodec ff_libvpx_decoder = {
> +AVCodec ff_libvpx_vp8_decoder = {
> .name =
On 14/01/13 19:12, Martin Storsjö wrote:
> This supports the AES_CM_128_HMAC_SHA1_80 and
> AES_CM_128_HMAC_SHA1_32 cipher suites (from RFC 4568) at the
> moment. The main missing features are replay protection (which can be
> added later without changing the internal API), and the F8 and null
> cip
On Mon, 14 Jan 2013, Luca Barbato wrote:
On 14/01/13 19:12, Martin Storsjö wrote:
+ if (outlen < c->hashlen)
+ return -1;
May I have an AVERROR for it? (the rest looks fine.)
Changed locally into AVERROR(EINVAL).
// Martin___
libav-devel mailing
On 14/01/2013 8:37 AM, Diego Biurrun wrote:
> Something along the lines of
>
> #if defined(_MSC_VER)
> #define FMT %Iu
> #else
> #define FMT %zu
> #endif
>
> and replacing %zu by FMT below seems simpler. Or is there something
> I am missing?
I wasn't able to get that to expand properl
On 14/01/13 19:12, Martin Storsjö wrote:
> + if (outlen < c->hashlen)
> + return -1;
May I have an AVERROR for it? (the rest looks fine.)
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
On Mon, 14 Jan 2013, Martin Storsjö wrote:
This is mostly useful for encryption together with the RTP muxer,
but could also be set up as IO towards the peer with the SDP
demuxer with custom IO.
---
libavformat/Makefile |1 +
libavformat/allformats.c |1 +
libavformat/srtpproto.c | 14
Also print port numbers for this protocol.
---
libavformat/sdp.c | 15 ++-
1 file changed, 14 insertions(+), 1 deletion(-)
diff --git a/libavformat/sdp.c b/libavformat/sdp.c
index 57044d4..186f83b 100644
--- a/libavformat/sdp.c
+++ b/libavformat/sdp.c
@@ -128,7 +128,7 @@ static int
This is mostly useful for encryption together with the RTP muxer,
but could also be set up as IO towards the peer with the SDP
demuxer with custom IO.
---
libavformat/Makefile |1 +
libavformat/allformats.c |1 +
libavformat/srtpproto.c | 144 +
This only takes care of decrypting incoming packets; the outgoing
RTCP packets are not encrypted. This is enough for some use cases,
and signalling crypto keys for use with outgoing RTCP packets
doesn't fit as simply into the API. If the SDP demuxer is hooked
up with custom IO, the return packets c
This supports the AES_CM_128_HMAC_SHA1_80 and
AES_CM_128_HMAC_SHA1_32 cipher suites (from RFC 4568) at the
moment. The main missing features are replay protection (which can be
added later without changing the internal API), and the F8 and null
ciphers.
---
libavformat/srtp.c | 291 ++
This supports HMAC-MD5 and HMAC-SHA1 for now, other hashes are
simple to add.
---
doc/APIchanges |3 ++
libavutil/Makefile |2 +
libavutil/hmac.c| 135 +++
libavutil/hmac.h| 44 +
libavutil/version.h |2 +-
On Mon, 14 Jan 2013, Sean McGovern wrote:
Several of the options were incorrect for suncc.
Let's just shove the GCC options through
the pre-existing compiler flags filter.
---
configure | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configure b/configure
index e9cdfab..da5634
On Mon, 14 Jan 2013, Sean McGovern wrote:
-mimpure-text is for GCC only, and suncc's equivalent of
-shared is -G.
---
configure | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/configure b/configure
index a2bf6e1..e9cdfab 100755
--- a/configure
+++ b/configure
@@ -2239,7 +223
Several of the options were incorrect for suncc.
Let's just shove the GCC options through
the pre-existing compiler flags filter.
---
configure | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configure b/configure
index e9cdfab..da56349 100755
--- a/configure
+++ b/configure
@@
-mimpure-text is for GCC only, and suncc's equivalent of
-shared is -G.
---
configure | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/configure b/configure
index a2bf6e1..e9cdfab 100755
--- a/configure
+++ b/configure
@@ -2239,7 +2239,8 @@ suncc_flags(){
-fomit-f
On 14/01/13 05:32, Luca Barbato wrote:
> By default libav sets it to 3/4 while x264 sets it to 9/10.
>
> CC: libav-sta...@libav.org
> ---
> libavcodec/libx264.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/libavcodec/libx264.c b/libavcodec/libx264.c
> index f536f7f.
---
Tested on a variety of configs, but that pesky emms bug prevents full testing
of x86_32 --disable-asm. So, more testing would be appreciated (and on MSVC).
---
libavcodec/x86/dsputilenc.asm | 152 +
libavcodec/x86/dsputilenc_mmx.c | 201
On Wed, Aug 01, 2012 at 12:18:02AM +0200, Diego Biurrun wrote:
> ---
> libavcodec/x86/ac3dsp.asm |3 ---
> libavcodec/x86/dsputilenc_yasm.asm |2 --
> libavutil/x86/x86util.asm | 29 +
> 3 files changed, 13 insertions(+), 21 deletions(-)
OKe
Luca Barbato writes:
>
> On 12/01/13 19:32, Luca Barbato wrote:
> > From: Joakim Plate
> >
>
Hehe. Always fun to see an old patch comming to life again ;)
/Joakim
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mail
On Sat, 12 Jan 2013 04:31:21 +0100, Luca Barbato wrote:
> rc_buffer_size is not set before.
>
> Solve the initial the rate control underflow issue reported in
> bug 222.
>
> CC: libav-sta...@libav.org
> ---
> avconv_opt.c | 2 --
> libavcodec/utils.c | 3 +++
> 2 files changed, 3 inserti
On 14/01/13 17:18, Luca Barbato wrote:
> Again, this time without blind and untested rebases...
>
Ok'd on IRC by Ronald.
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
Support older libvpx versions.
---
configure | 16 ++--
libavcodec/Makefile| 6 --
libavcodec/allcodecs.c | 4 ++--
libavcodec/libvpxdec.c | 17 +++--
libavcodec/libvpxenc.c | 34 ++
5 files changed, 45 insertions(+), 3
This feature is experimental use at your risk
---
libavcodec/allcodecs.c | 1 +
libavcodec/codec_desc.c | 7 +++
libavcodec/libvpxdec.c | 26 --
libavcodec/libvpxenc.c | 38 --
4 files changed, 68 insertions(+), 4 deletions(-)
d
---
libavformat/nut.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/libavformat/nut.c b/libavformat/nut.c
index 196e04e..9267226 100644
--- a/libavformat/nut.c
+++ b/libavformat/nut.c
@@ -39,6 +39,7 @@ const AVCodecTag ff_nut_data_tags[] = {
};
const AVCodecTag ff_nut_video_tags[] = {
+
From: Tom Finegan
---
libavcodec/avcodec.h | 1 +
libavformat/matroska.c | 1 +
2 files changed, 2 insertions(+)
diff --git a/libavcodec/avcodec.h b/libavcodec/avcodec.h
index e57170b..09b9b42 100644
--- a/libavcodec/avcodec.h
+++ b/libavcodec/avcodec.h
@@ -267,6 +267,7 @@ enum AVCodecID {
Again, this time without blind and untested rebases...
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
On 14/01/13 17:00, Martin Storsjö wrote:
> This makes the behaviour defined when they wrap around. The value
> assigned to expected_prior was a uint32_t already.
> ---
> libavformat/rtpdec.h |6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
Ok.
_
This makes the behaviour defined when they wrap around. The value
assigned to expected_prior was a uint32_t already.
---
libavformat/rtpdec.h |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/libavformat/rtpdec.h b/libavformat/rtpdec.h
index 1cae1e4..a0ee655 100644
--- a
On 14/01/13 16:44, Diego Biurrun wrote:
> ---
> libavcodec/Makefile |1 -
> 1 files changed, 0 insertions(+), 1 deletions(-)
>
Ok.
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
---
libavcodec/Makefile |1 -
1 files changed, 0 insertions(+), 1 deletions(-)
diff --git a/libavcodec/Makefile b/libavcodec/Makefile
index 1001c2d..2330f85 100644
--- a/libavcodec/Makefile
+++ b/libavcodec/Makefile
@@ -385,7 +385,6 @@ OBJS-$(CONFIG_VC1_DXVA2_HWACCEL) += dxva2_vc1.o
OB
On 01/14/2013 05:01 AM, Martin Storsjö wrote:
> From: Michael Niedermayer
>
> ---
> libavcodec/aacenc.c |2 +-
> libavcodec/libfaac.c |2 +-
> libavcodec/libfdk-aacenc.c |2 +-
> libavcodec/libmp3lame.c |2 +-
> libavcodec/libopencore-amr.c |2 +-
> l
On 14/01/13 16:25, Ronald S. Bultje wrote:
> Commented out in original code also, I'm trying to not make unrelated
> functional modifications.
>
The patch looks fine and you tested it a lot from what I can see, commit
anytime.
___
libav-devel mailing l
Hi,
On Mon, Jan 14, 2013 at 6:11 AM, Luca Barbato wrote:
> On 14/01/13 06:46, Ronald S. Bultje wrote:
>
>> libavcodec/h264.c | 12 -
>> libavcodec/h264.h | 5 +-
>> libavcodec/h264_refs.c | 143
>> ++---
>> 3 files changed, 114 insert
On 14/01/13 06:46, Ronald S. Bultje wrote:
> libavcodec/h264.c | 12 -
> libavcodec/h264.h | 5 +-
> libavcodec/h264_refs.c | 143
> ++---
> 3 files changed, 114 insertions(+), 46 deletions(-)
>
> +#if 0
> +if (mm
On Sat, Jan 12, 2013 at 07:57:09PM -0500, Derek Buitenhuis wrote:
>
> --- a/libavcodec/tableprint.h
> +++ b/libavcodec/tableprint.h
> @@ -71,20 +71,32 @@ void write_uint32_t_2d_array(const void *, int, int);
>
> -#define WRITE_ARRAY(prefix, type, name) \
> +#if defined(_MSC_VER)
On 14/01/13 14:29, Martin Storsjö wrote:
> On Mon, 14 Jan 2013, Luca Barbato wrote:
>
>> On 14/01/13 10:43, Martin Storsjö wrote:
>>> Without this, we'd signal a huge loss rate (due to wraparound) if
>>> we had received one packet more than expected (that is, one seq
>>> number sent twice). The co
On Mon, 14 Jan 2013, Luca Barbato wrote:
On 14/01/13 10:43, Martin Storsjö wrote:
Without this, we'd signal a huge loss rate (due to wraparound) if
we had received one packet more than expected (that is, one seq
number sent twice). The code has a check for lost_interval <= 0,
but that doesn't d
On 14/01/13 10:43, Martin Storsjö wrote:
> Without this, we'd signal a huge loss rate (due to wraparound) if
> we had received one packet more than expected (that is, one seq
> number sent twice). The code has a check for lost_interval <= 0,
> but that doesn't do what was intended as long as the va
1 - 100 of 115 matches
Mail list logo
