On Sun, Feb 28, 2016 at 4:52 PM, Muhammad Faiz wrote:
> optimize draw_bar_yuv (slightly faster)
> optimize draw_axis (about 2x faster)
>
> Signed-off-by: Muhammad Faiz
> ---
> libavfilter/avf_showcqt.c | 212
> +-
> 1 file changed, 115 insertions(+),
On 02/29/2016 07:39 AM, Mats Peterson wrote:
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Once again, this patch set is needed for stream copy of raw palettized
data, since the packets will
From: Clément Bœsch
---
Changes since latest version:
- remove unused 32-bit path
- make 16-bit path more accurate by mirroring the MMX code (still not bitexact)
- the code as originally trying to process 2 lines at a time to save chroma pre
mult computations and avoid re-reading the whole line
On Thu, Feb 25, 2016 at 02:56:52PM +0100, Clément Bœsch wrote:
> On Thu, Feb 25, 2016 at 02:47:45PM +0100, Hendrik Leppkes wrote:
> > On Thu, Feb 25, 2016 at 2:35 PM, Clément Bœsch wrote:
> > > From: Clément Bœsch
> > >
> > > This option is typically useful when cross-compiling dynamic libraries
On Mon, Feb 29, 2016 at 01:45:28AM +0100, Carl Eugen Hoyos wrote:
> Hi!
>
> I believe attached patch fixes a possible issue when auto-detecting jpeg.
>
> Please review, Carl Eugen
> img2dec.c |1 +
> 1 file changed, 1 insertion(+)
> db2e658f6b6761e038746e51569a9faf5516e0ff patchsoslength.d
Hi!
Attached patch fixes ticket #5271 for me.
Please comment, Carl Eugen
diff --git a/libavformat/mov.c b/libavformat/mov.c
index 043f4a9..888b2ad 100644
--- a/libavformat/mov.c
+++ b/libavformat/mov.c
@@ -1461,6 +1461,11 @@ static int mov_read_ares(MOVContext *c, AVIOContext *pb,
MOVAtom atom)
Michael Niedermayer niedermayer.cc> writes:
> > img2dec.c |1 +
> > 1 file changed, 1 insertion(+)
> > db2e658f6b6761e038746e51569a9faf5516e0ff patchsoslength.diff
> > diff --git a/libavformat/img2dec.c b/libavformat/img2dec.c
>
> LGTM
Patch applied.
Thank you, Carl Eugen
__
On Sat, Feb 27, 2016 at 09:56:01PM +0200, Andrew Shulgin wrote:
> Some Hikvison RTSP cameras send wrong sprop-parameter-sets in SDP - it's
> missing PPS.
>
> Example of correct sprop-parameter-sets:
> sprop-parameter-sets=Z0IAH5W4PASaEAAAcIAAFfkAQA==,aM48gA==
> Example of wrong sprop-parameter-set
Derek Buitenhuis gmail.com> writes:
> On 2/26/2016 3:05 PM, Carl Eugen Hoyos wrote:
> > Or in other words: Except for jpeg, there is no
> > probe function that returns 0 although the decoder
> > can decode the image.
>
> I'll take your word for it on this part.
Please don't: Instead double-ch
Correct SPS and PPS are in the stream.
Maybe it's reasonale to use existing SPS. But i'm not sure how to implement
that.
Yeah really, setting extradata_size to 0 should not be skipped.
On Mon, Feb 29, 2016 at 1:01 PM, Michael Niedermayer wrote:
> On Sat, Feb 27, 2016 at 09:56:01PM +0200, Andrew
From: Clément Bœsch
---
TODO: doc, bump, Changelog
example: -vf bench=start,gradfun,format=rgba,hqx,bench=stop
---
libavfilter/Makefile | 2 +
libavfilter/allfilters.c | 2 +
libavfilter/f_bench.c| 132 +++
3 files changed, 136 insertions(
Hi!
Attached patch adds auto-detection for pcx images.
Please comment, Carl Eugen
diff --git a/libavformat/Makefile b/libavformat/Makefile
index 782908e..edfa695 100644
--- a/libavformat/Makefile
+++ b/libavformat/Makefile
@@ -218,6 +218,7 @@ OBJS-$(CONFIG_IMAGE_EXR_PIPE_DEMUXER) += img2dec.o
Corrected the patch a bit, now setting extradata_size to 0
On Mon, Feb 29, 2016 at 1:13 PM, Andrew Shulgin
wrote:
> Correct SPS and PPS are in the stream.
> Maybe it's reasonale to use existing SPS. But i'm not sure how to
> implement that.
> Yeah really, setting extradata_size to 0 should not b
Carl Eugen Hoyos ag.or.at> writes:
> +|| b[1] > 6
Locally fixed to "> 5"
Carl Eugen
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
On Mon, Feb 29, 2016 at 12:01:35PM +, Carl Eugen Hoyos wrote:
> Carl Eugen Hoyos ag.or.at> writes:
>
> > +|| b[1] > 6
>
> Locally fixed to "> 5"
patch probably ok
thx
[...]
--
Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB
He who knows, does not speak. H
On Mon, Feb 29, 2016 at 11:52:24AM +0100, Carl Eugen Hoyos wrote:
> Hi!
>
> Attached patch fixes ticket #5271 for me.
>
> Please comment, Carl Eugen
> mov.c |5 +
> 1 file changed, 5 insertions(+)
> ef08b944e3cb77bd7311187ecbfbdae719147d92 patchaviddv.diff
> diff --git a/libavformat/mo
On Mon, Feb 29, 2016 at 01:59:22PM +0200, Andrew Shulgin wrote:
> Corrected the patch a bit, now setting extradata_size to 0
>
> On Mon, Feb 29, 2016 at 1:13 PM, Andrew Shulgin
> wrote:
>
> > Correct SPS and PPS are in the stream.
> > Maybe it's reasonale to use existing SPS. But i'm not sure ho
Hi,
On Sun, Feb 28, 2016 at 8:26 PM, Timothy Gu wrote:
> ---
> libavcodec/x86/Makefile | 3 +-
> libavcodec/x86/vc1dsp.asm| 585
> ---
> libavcodec/x86/vc1dsp_loopfilter.asm | 317 +++
> libavcodec/x86/vc1dsp_mc.asm
Also i can send you an email with public RTSP URL privately if you ask.
On Mon, Feb 29, 2016 at 3:10 PM, Andrew Shulgin
wrote:
> Extracting SPS without the missig PPS results in unplayable FLV.
>
> ffmpeg -thread_queue_size 512 -rtsp_transport tcp -i rtsp://
> admin:12345@192.88.99.1/h264/ch1/ma
On Fri, Feb 19, 2016 at 17:51:00 +0100, Nicolas George wrote:
> Changing the name of existing muxers is annoying, but for new features I
> would like much better if we did not wave around the name of a hash function
> that has known vulnerabilities since 20 years: "streamhash" would be an
> obvious
Extracting SPS without the missig PPS results in unplayable FLV.
ffmpeg -thread_queue_size 512 -rtsp_transport tcp -i rtsp://
admin:12345@192.88.99.1/h264/ch1/main/av_stream -ar 44100 -ac 2 -acodec
pcm_s16le -f s16le -ac 2 -i /dev/zero -c:v copy -acodec aac -strict
experimental -ab 128k -f flv wro
Small fix.
--
Mats Peterson
http://matsp888.no-ip.org/~mats/
>From ca4010bd52169e4a10513981fc876334eb2b4e5e Mon Sep 17 00:00:00 2001
From: Mats Peterson
Date: Mon, 29 Feb 2016 14:10:54 +0100
Subject: [PATCH 2/4 v2] lavf/movenc: Add support for palette side data packets
---
libavformat/movenc.c
New patch set.
--
Mats Peterson
http://matsp888.no-ip.org/~mats/
>From 46b854f4b872d67b3aed118c929250f983dbab7b Mon Sep 17 00:00:00 2001
From: Mats Peterson
Date: Mon, 29 Feb 2016 14:23:50 +0100
Subject: [PATCH v2 1/4] lavf/avienc: Add support for palette side data packets
---
libavformat/avie
--
Mats Peterson
http://matsp888.no-ip.org/~mats/
>From f659982a4799aa3bdde3b3498fd04ba236174b7d Mon Sep 17 00:00:00 2001
From: Mats Peterson
Date: Mon, 29 Feb 2016 14:24:03 +0100
Subject: [PATCH v2 2/4] lavf/movenc: Add support for palette side data packets
---
libavformat/movenc.c | 22 +++
--
Mats Peterson
http://matsp888.no-ip.org/~mats/
>From 750d1d19dbb234c5d2d07dd8235afa48d3f1ba1f Mon Sep 17 00:00:00 2001
From: Mats Peterson
Date: Mon, 29 Feb 2016 14:24:31 +0100
Subject: [PATCH v2 3/4] lavf/rawutils: New function ff_get_raw_palette()
---
libavformat/rawutils.c | 16 +++
--
Mats Peterson
http://matsp888.no-ip.org/~mats/
>From 56ed6a5c0ac819e0cbb941034868490271f61801 Mon Sep 17 00:00:00 2001
From: Mats Peterson
Date: Mon, 29 Feb 2016 14:24:53 +0100
Subject: [PATCH v2 4/4] lavf/internal.h: Add declaration for ff_get_raw_palette()
---
libavformat/internal.h |
Le primidi 11 ventôse, an CCXXIV, Moritz Barsnick a écrit :
> I would like to propose a (quite trivial) patch renaming first the
> sources, then the muxers, and of course aliasing the existing muxer
> names.
No objection from me. If you manage to make it so that the default is MD5
when the name is
On Mon, Feb 29, 2016 at 2:41 PM, Nicolas George wrote:
>
>> I don't quite understand: Is there something that needs to be fixed in
>> the existing md5enc code?
>
> Apparently not, which means I am misremembering something.
>
You just can't reuse the classes for mutliple components, options are fi
Fixes ticket 4313.
---
ffmpeg.c| 1 +
libavcodec/Makefile | 1 +
libavcodec/allcodecs.c | 1 +
libavcodec/vp9_superframe_bsf.c | 189
libavformat/ivfenc.c| 13 +++
libavformat/matroskaenc.c
If pkt->size == 0 && pkt->side_data_elems == 0 after bsf->filter()
returns, the packet is considered dropped.
---
ffmpeg.c| 2 ++
libavformat/mux.c | 24 +---
libavformat/utils.c | 5 +
3 files changed, 20 insertions(+), 11 deletions(-)
diff --git a/ffmpeg.
Michael Niedermayer niedermayer.cc> writes:
> On Mon, Feb 29, 2016 at 12:01:35PM +, Carl Eugen Hoyos wrote:
> > Carl Eugen Hoyos ag.or.at> writes:
> >
> > > +|| b[1] > 6
> >
> > Locally fixed to "> 5"
>
> patch probably ok
Pushed.
Thank you, Carl Eugen
_
Le primidi 11 ventôse, an CCXXIV, Ronald S. Bultje a écrit :
> If pkt->size == 0 && pkt->side_data_elems == 0 after bsf->filter()
> returns, the packet is considered dropped.
I think the doxy for av_apply_bitstream_filters() needs to be updated
accordingly.
Regards,
--
Nicolas George
signat
Autodetected by default. Encode using -codec:v vtenc.
Signed-off-by: Rick Kern
---
MAINTAINERS|1 +
configure | 19 +
libavcodec/Makefile|1 +
libavcodec/allcodecs.c |1 +
libavcodec/vtenc.c | 1320
5
Mats Peterson skrev: (29 februari 2016
14:27:13 CET)
>New patch set.
It seems that you, Michael, don't care much about adding support for palette
side data packets, while Reimar advocates for it. Is the situation correctly
understood?
--
Mats Peterson
http://matsp888.no-ip.org/~mats/
On Mon, Feb 29, 2016 at 11:14:17PM +0800, Rick Kern wrote:
[...]
> +if (!profile) {
> +//VideoToolbox auto-selects profile and level.
> +return true;
> +} else if (!av_strcasecmp("baseline", profile)) {
> +if (!strcmp("auto", level)) *profile_level_val =
> kVTP
Fixes ticket 4313.
---
ffmpeg.c| 1 +
libavcodec/Makefile | 1 +
libavcodec/allcodecs.c | 1 +
libavcodec/vp9_superframe_bsf.c | 189
libavformat/ivfenc.c| 13 +++
libavformat/matroskaenc.c
If pkt->size == 0 && pkt->side_data_elems == 0 after bsf->filter()
returns, the packet is considered dropped.
---
ffmpeg.c | 2 ++
libavcodec/avcodec.h | 3 ++-
libavformat/avformat.h | 4 +++-
libavformat/mux.c | 24 +---
libavformat/utils.c| 5 +++
On Mon, 29 Feb 2016 23:14:17 +0800
Rick Kern wrote:
> Autodetected by default. Encode using -codec:v vtenc.
>
> Signed-off-by: Rick Kern
> ---
OK, autodetection seems to work properly on non-OSX too. I'll apply in
18 hours or so.
___
ffmpeg-devel mai
> On Feb 29, 2016, at 11:19 PM, Clément Bœsch wrote:
>
> On Mon, Feb 29, 2016 at 11:14:17PM +0800, Rick Kern wrote:
> [...]
>> +if (!profile) {
>> +//VideoToolbox auto-selects profile and level.
>> +return true;
>> +} else if (!av_strcasecmp("baseline", profile)) {
>> +
Please ignore this patch - the options code need to be updated.
> On Feb 29, 2016, at 11:58 PM, wm4 wrote:
>
> On Mon, 29 Feb 2016 23:14:17 +0800
> Rick Kern wrote:
>
>> Autodetected by default. Encode using -codec:v vtenc.
>>
>> Signed-off-by: Rick Kern
>> ---
>
> OK, autodetection seems t
Le primidi 11 ventôse, an CCXXIV, Richard Kern a écrit :
> No, doesn’t look like it. The VideoToolbox constants aren’t 1-to-1 with the
> command line options.
They do not need to be, you can use intermediate integer constants just as
well as you currently use string constants, with the extra bene
On Sun, Feb 28, 2016 at 11:08:59PM +0100, Marton Balint wrote:
>
> On Sun, 28 Feb 2016, Michael Niedermayer wrote:
>
> >On Sat, Feb 20, 2016 at 02:11:49AM +0100, Marton Balint wrote:
> >>This also fixes reading gapless metadata when the entries do not start with
> >>the
> >>mean atom. Such sampl
On Mon, Feb 29, 2016 at 4:57 AM Ronald S. Bultje wrote:
> This is kind of hard to review, but I'm going to assume that there's no
> actual code changes, in which case this LGTM.
>
No there isn't. Applied, thanks.
Timothy
>
___
ffmpeg-devel mailing li
Autodetected by default. Encode using -codec:v vtenc.
Signed-off-by: Rick Kern
---
MAINTAINERS|1 +
configure | 19 +
libavcodec/Makefile|1 +
libavcodec/allcodecs.c |1 +
libavcodec/vtenc.c | 1339
5
On Tue, 1 Mar 2016 01:30:58 +0800
Rick Kern wrote:
> Autodetected by default. Encode using -codec:v vtenc.
>
> Signed-off-by: Rick Kern
> ---
> MAINTAINERS|1 +
> configure | 19 +
> libavcodec/Makefile|1 +
> libavcodec/allcodecs.c |1 +
> libavcode
On 2/29/16, Clement Boesch wrote:
> From: Clement Boesch
>
> ---
> TODO: doc, bump, Changelog
>
> example: -vf bench=start,gradfun,format=rgba,hqx,bench=stop
Nice, LGTM with documentation changes.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
h
On Tue, Mar 01, 2016 at 01:30:58AM +0800, Rick Kern wrote:
> Autodetected by default. Encode using -codec:v vtenc.
>
> Signed-off-by: Rick Kern
[...]
> +/*
> + * contiguous_buf_size is 0 if not contiguous, and the size of the buffer
> + * containing all planes if so.
> + */
> +static int get_cv_
On Mon, Feb 29, 2016 at 06:57:15PM +0100, wm4 wrote:
[...]
> > +static bool get_vt_profile_level(AVCodecContext *avctx,
> > + CFStringRef*profile_level_val)
> > +{
> > +VTEncContext *vtctx = avctx->priv_data;
> > +int64_t profile = vtctx->profile;
> > +
>
Le primidi 11 ventôse, an CCXXIV, Clement Boesch a écrit :
> > > +case H264_PROF_BASELINE:
> > > +switch (vtctx->level) {
> > > +case 0: *profile_level_val =
> > > kVTProfileLevel_H264_Baseline_AutoLevel; break;
On Mon, Feb 29, 2016 at 08:14:10PM +0100, Nicolas George wrote:
> Le primidi 11 ventôse, an CCXXIV, Clement Boesch a écrit :
> > > > +case H264_PROF_BASELINE:
> > > > +switch (vtctx->level) {
> > > > +case 0: *profile_level_val =
> > > > kVTProfileLevel_H264_Ba
On Mon, 29 Feb 2016 20:07:15 +0100
Clément Bœsch wrote:
> On Mon, Feb 29, 2016 at 06:57:15PM +0100, wm4 wrote:
> [...]
> > > +static bool get_vt_profile_level(AVCodecContext *avctx,
> > > + CFStringRef*profile_level_val)
> > > +{
> > > +VTEncContext *vtctx
Hi James,
On Thu, Feb 25, 2016 at 6:50 PM, James Almer wrote:
>
> Did you try running the FATE suite using threadsanitizer, or just decoded
> random
> videos? I remember like half the tests were failing because of data races and
> most of them pointed to code changed by your patches. It would be
On Fri, Feb 26, 2016 at 5:04 PM, Ronald S. Bultje wrote:
>
> To be a little bit more explicit, it should be relatively easy to
> accomplish this by changing the position of qscale in the relevant struct,
> and only copy that half of the struct (where entries are obviously grouped)
> where we actua
On Mon, Feb 29, 2016 at 11:47:14AM -0800, Wan-Teh Chang wrote:
> On Fri, Feb 26, 2016 at 5:04 PM, Ronald S. Bultje wrote:
> >
> > To be a little bit more explicit, it should be relatively easy to
> > accomplish this by changing the position of qscale in the relevant struct,
> > and only copy that
Hi,
On Mon, Feb 29, 2016 at 2:47 PM, Wan-Teh Chang wrote:
> On Fri, Feb 26, 2016 at 5:04 PM, Ronald S. Bultje
> wrote:
> >
> > To be a little bit more explicit, it should be relatively easy to
> > accomplish this by changing the position of qscale in the relevant
> struct,
> > and only copy tha
Signed-off-by: Michael Niedermayer
---
tests/fate/seek.mak |2 ++
tests/ref/seek/cache-pipe | 49 +
2 files changed, 51 insertions(+)
create mode 100644 tests/ref/seek/cache-pipe
diff --git a/tests/fate/seek.mak b/tests/fate/seek.mak
index
Signed-off-by: Michael Niedermayer
---
libavformat/seek-test.c |2 ++
1 file changed, 2 insertions(+)
diff --git a/libavformat/seek-test.c b/libavformat/seek-test.c
index bfd06db..904b435 100644
--- a/libavformat/seek-test.c
+++ b/libavformat/seek-test.c
@@ -80,6 +80,8 @@ int main(int argc,
On Mon, Feb 29, 2016 at 04:12:56PM +0100, Mats Peterson wrote:
> Mats Peterson skrev: (29 februari 2016
> 14:27:13 CET)
> >New patch set.
>
> It seems that you, Michael, don't care much about adding support for palette
> side data packets, while Reimar advocates for it. Is the situation correct
Is it possible for AVCodecContext 's reordered_opaque to be exactly the
same for 3 or 4 consecutive frames when packet's pts is fed in it ? Is it
because of B frames? I am seeing this, in this case how does one re-orders
the pics to render?
Ratin
___
f
On Sun, Feb 28, 2016 at 12:50:24PM +0100, Michael Niedermayer wrote:
> Signed-off-by: Michael Niedermayer
> ---
> tests/fate/libavcodec.mak |5 +
> 1 file changed, 5 insertions(+)
patchset applied
[...]
--
Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB
it is n
On 02/29/2016 09:42 PM, Michael Niedermayer wrote:
On Mon, Feb 29, 2016 at 04:12:56PM +0100, Mats Peterson wrote:
Mats Peterson skrev: (29 februari 2016
14:27:13 CET)
New patch set.
It seems that you, Michael, don't care much about adding support for palette
side data packets, while Reimar
This bug was found by Dmitry Vyukov. If two threads may call
ff_thread_report_progress at the same time, progress[field] may
decrease. For example, suppose progress[field] is 10 and two threads
call ff_thread_report_progress to update progress[field] to 11 and
12, respectively. If the second thread
Hello !
This patch enables background sending of UDP packets with specified delay.
When sending packets without a delay some devices with small RX buffer
( MAG200 STB, for example) will drop tail packets in burst causing
decoding errors.
It needs to specify "fifo_size" with "packet_gap" .
The ou
On Mon, Feb 29, 2016 at 03:10:41PM +0200, Andrew Shulgin wrote:
> Extracting SPS without the missig PPS results in unplayable FLV.
>
> ffmpeg -thread_queue_size 512 -rtsp_transport tcp -i rtsp://
> admin:12345@192.88.99.1/h264/ch1/main/av_stream -ar 44100 -ac 2 -acodec
> pcm_s16le -f s16le -ac 2 -
Yep, you're right.
On Tue, Mar 1, 2016 at 2:09 AM, Michael Niedermayer
wrote:
> On Mon, Feb 29, 2016 at 03:10:41PM +0200, Andrew Shulgin wrote:
> > Extracting SPS without the missig PPS results in unplayable FLV.
> >
> > ffmpeg -thread_queue_size 512 -rtsp_transport tcp -i rtsp://
> > admin:1234
On Thu, Feb 25, 2016 at 1:11 PM, Michael Niedermayer
wrote:
> On Wed, Feb 24, 2016 at 09:20:11PM -0500, Ganesh Ajjanagadde wrote:
>> Add some more verbose info regarding why the imprecise and slow floor(x+0.5)
>> hack
>> is used; helpful for future maintenance.
>>
>> Signed-off-by: Ganesh Ajjanag
On 1/25/16, Hendrik Leppkes wrote:
> On Mon, Jan 25, 2016 at 1:13 PM, Michael Niedermayer
> wrote:
>> On Mon, Jan 25, 2016 at 10:24:36AM +, Carl Eugen Hoyos wrote:
>>> Kieran Kunhya kunhya.com> writes:
>>>
>>> > The internal encoder is superior to libvo-aacenc.
>>>
>>> I thought this was the
On Tue, Mar 01, 2016 at 02:11:28AM +0200, Andrew Shulgin wrote:
> Yep, you're right.
ok, applied
thanks
>
> On Tue, Mar 1, 2016 at 2:09 AM, Michael Niedermayer
> wrote:
>
> > On Mon, Feb 29, 2016 at 03:10:41PM +0200, Andrew Shulgin wrote:
> > > Extracting SPS without the missig PPS results i
Thanks :)
On Tue, Mar 1, 2016 at 2:43 AM, Michael Niedermayer
wrote:
> On Tue, Mar 01, 2016 at 02:11:28AM +0200, Andrew Shulgin wrote:
> > Yep, you're right.
>
> ok, applied
>
> thanks
>
>
> >
> > On Tue, Mar 1, 2016 at 2:09 AM, Michael Niedermayer
>
> > wrote:
> >
> > > On Mon, Feb 29, 2016 at
Adding early support for a subset of the proposed colour elements
according to the latest version of spec:
https://mailarchive.ietf.org/arch/search/?email_list=cellar&gbt=1&index=hIKLhMdgTMTEwUTeA4ct38h0tmE
I've left out elements for pix_fmt related things as there still
seems to be some discussio
Signed-off-by: Neil Birkbeck
---
libavformat/dump.c | 21 +
1 file changed, 21 insertions(+)
diff --git a/libavformat/dump.c b/libavformat/dump.c
index 56c285d..9e7c12b 100644
--- a/libavformat/dump.c
+++ b/libavformat/dump.c
@@ -26,6 +26,7 @@
#include "libavutil/display.h"
On Mon, Feb 22, 2016 at 11:34 PM, Andrey Utkin
wrote:
> Hi!
> I am aware of news that AAC encoder got stable status recently.
>
> But you could find this interesting. We've got an ffmpeg build from
> October 2014, and it performs three times faster on AAC encoding than
> recent 3.0 release. There
Correct the order of the load/store operation and the memory barrier in
avpriv_atomic_int_get and avpriv_atomic_int_set.
Use the atomic get and set functions in ff_thread_report_progress and
ff_thread_await_progress.
---
libavcodec/pthread_frame.c | 11 +++
libavutil/atomic.c | 48
---
tests/fate/filter-audio.mak |7 +
tests/ref/fate/filter-hls2 | 768 +++
2 files changed, 775 insertions(+)
create mode 100644 tests/ref/fate/filter-hls1
create mode 100644 tests/ref/fate/filter-hls2
diff --git a/tests/fate/filter-audio.mak b/tes
Hi,
On Mon, Feb 29, 2016 at 5:41 PM, Wan-Teh Chang wrote:
> This bug was found by Dmitry Vyukov. If two threads may call
> ff_thread_report_progress at the same time, progress[field] may
> decrease. For example, suppose progress[field] is 10 and two threads
> call ff_thread_report_progress to up
Hi,
On Mon, Feb 29, 2016 at 9:30 PM, Ganesh Ajjanagadde
wrote:
> Ideally, FATE should have some basic plotting/performance
> infrastructure, e.g a client can submit perf figures so that evolution
> over time can be viewed. No idea why this can't be done.
It can - it just hasn't been done yet :
Hi,
On Mon, Feb 29, 2016 at 9:35 PM, Wan-Teh Chang wrote:
> Correct the order of the load/store operation and the memory barrier in
> avpriv_atomic_int_get and avpriv_atomic_int_set.
>
This sounds useful. Is there some documentation on which one should be used
under what conditions? I.e. when d
On Mon, Feb 29, 2016 at 10:02:46PM -0500, Ronald S. Bultje wrote:
> Hi,
>
> On Mon, Feb 29, 2016 at 9:30 PM, Ganesh Ajjanagadde
> wrote:
>
> > Ideally, FATE should have some basic plotting/performance
> > infrastructure, e.g a client can submit perf figures so that evolution
> > over time can be
On 2/29/2016 11:47 PM, Michael Niedermayer wrote:
> ---
> tests/fate/filter-audio.mak |7 +
> tests/ref/fate/filter-hls2 | 768
> +++
> 2 files changed, 775 insertions(+)
> create mode 100644 tests/ref/fate/filter-hls1
> create mode 100644 tests/ref
It makes no sense whatsoever to do this at each function call; we
already have a table for this.
Yields a 2x improvement in find_min_book (x86-64, Haswell+GCC):
ffmpeg -i sin.flac -acodec aac -y sin.aac
find_min_book
old
605 decicycles in find_min_book, 8388453 runs,155 skips.9x
606 de
On Tue, Mar 01, 2016 at 12:17:33AM -0300, James Almer wrote:
> On 2/29/2016 11:47 PM, Michael Niedermayer wrote:
> > ---
> > tests/fate/filter-audio.mak |7 +
> > tests/ref/fate/filter-hls2 | 768
> > +++
> > 2 files changed, 775 insertions(+)
> > cr
On Mon, Feb 29, 2016 at 10:07 PM, Michael Niedermayer
wrote:
> On Mon, Feb 29, 2016 at 10:02:46PM -0500, Ronald S. Bultje wrote:
>> Hi,
>>
>> On Mon, Feb 29, 2016 at 9:30 PM, Ganesh Ajjanagadde
>> wrote:
>>
>> > Ideally, FATE should have some basic plotting/performance
>> > infrastructure, e.g a
On 3/1/2016 12:24 AM, Michael Niedermayer wrote:
> On Tue, Mar 01, 2016 at 12:17:33AM -0300, James Almer wrote:
>> On 2/29/2016 11:47 PM, Michael Niedermayer wrote:
>>> +fate-filter-hls2: tests/data/hls-list.m3u8
>>
>> Why not just add fate-filter-hls1 as dep here, and get rid of the m3u8
>> targe
On Tue, Mar 01, 2016 at 12:32:27AM -0300, James Almer wrote:
> On 3/1/2016 12:24 AM, Michael Niedermayer wrote:
> > On Tue, Mar 01, 2016 at 12:17:33AM -0300, James Almer wrote:
> >> On 2/29/2016 11:47 PM, Michael Niedermayer wrote:
> >>> +fate-filter-hls2: tests/data/hls-list.m3u8
> >>
> >> Why not
Hi mentors
Forwarding as it says "please forward this e-mail to any mentors from your org
new to Outreachy this round and encourage them to participate."
- Forwarded message from Marina Zhurakhinskaya -
Date: Mon, 29 Feb 2016 22:47:14 -0500 (EST)
From: Marina Zhurakhinskaya
To: outre
---
libavformat/sdp.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavformat/sdp.c b/libavformat/sdp.c
index 2ab37a8..368402b 100644
--- a/libavformat/sdp.c
+++ b/libavformat/sdp.c
@@ -706,7 +706,7 @@ static char *sdp_write_media_attributes(char *buff, int
size, AVCodecCo
Replaced color_range with av_frame_get_color_range().
Rick Kern (1):
VideoToolbox H.264 Encoder
MAINTAINERS|1 +
configure | 19 +
libavcodec/Makefile|1 +
libavcodec/allcodecs.c |1 +
libavcodec/vtenc.c | 1339
Autodetected by default. Encode using -codec:v vtenc.
Signed-off-by: Rick Kern
---
MAINTAINERS|1 +
configure | 19 +
libavcodec/Makefile|1 +
libavcodec/allcodecs.c |1 +
libavcodec/vtenc.c | 1339
5
On 02/29/2016 02:27 PM, Mats Peterson wrote:
New patch set.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Sorry, Michael, but this won't be the last version. There are other
codecs than raw
This patch set adds support for non-raw codecs that use a palette.
Try to do stream copy to and from avi with the files below:
QuickTime Animation (RLE):
https://drive.google.com/open?id=0B3_pEBoLs0faREo1SlRydmV1LU0
QuickTime Graphics (SMC):
https://drive.google.com/open?id=0B3_pEBoLs0faODd5RVB
--
Mats Peterson
http://matsp888.no-ip.org/~mats/
>From 74f44252ec10720b6d7813f7fe0bc166c871ecbd Mon Sep 17 00:00:00 2001
From: Mats Peterson
Date: Tue, 1 Mar 2016 08:06:38 +0100
Subject: [PATCH v3 2/5] lavf/movenc: Add support for palette side data packets
---
libavformat/movenc.c | 45
--
Mats Peterson
http://matsp888.no-ip.org/~mats/
>From 701ba074d4d5c0d0fe0b1b814db41df940a2c7d7 Mon Sep 17 00:00:00 2001
From: Mats Peterson
Date: Tue, 1 Mar 2016 08:07:29 +0100
Subject: [PATCH v3 3/5] lavf/riffenc: Handle non-raw codecs with a palette
---
libavformat/riffenc.c | 20 +++
--
Mats Peterson
http://matsp888.no-ip.org/~mats/
>From 9a816b6d3d592684e08eac3120aaf94146b179ce Mon Sep 17 00:00:00 2001
From: Mats Peterson
Date: Tue, 1 Mar 2016 08:07:56 +0100
Subject: [PATCH v3 4/5] lavf/rawutils: New function ff_get_raw_palette()
---
libavformat/rawutils.c | 16
--
Mats Peterson
http://matsp888.no-ip.org/~mats/
>From a8a3cef658f4d9c3e4f9a5f9839cfc8761379d83 Mon Sep 17 00:00:00 2001
From: Mats Peterson
Date: Tue, 1 Mar 2016 08:08:19 +0100
Subject: [PATCH v3 5/5] lavf/internal.h: Add declaration for ff_get_raw_palette()
---
libavformat/internal.h | 12
94 matches
Mail list logo
