Hi, all,
Our team plans to contribute SIMD optimization on PowerPC for HEVC Encoder
or Decoder in FFmpeg. Before it, we'd like to know if there has been related
works (completed or in progress) and if this optimization is in demand.
Thanks!
Zhenan Lin
Institute of Computer Science and Tech
MSVC/ICL does not have unistd.h as standard and instead declares read/close
functions in io.h. This is currently handled in some files but not in a
couple. Adding an include for io.h where needed adds the correct
declarations.
0001-msvc-fix-implicitly-declared-read-close.patch
Description: Binary
On Fri, Aug 01, 2014 at 08:37:27PM -0400, compn wrote:
> patch from https://trac.ffmpeg.org/ticket/3824
>
> -compn
> diff --git a/libavformat/http.c b/libavformat/http.c
> index 33585b0..3dffaee 100644
> --- a/libavformat/http.c
> +++ b/libavformat/http.c
> @@ -229,6 +229,7 @@ redo:
> me
patch from https://trac.ffmpeg.org/ticket/3824
-compndiff --git a/libavformat/http.c b/libavformat/http.c
index 33585b0..3dffaee 100644
--- a/libavformat/http.c
+++ b/libavformat/http.c
@@ -229,6 +229,7 @@ redo:
memset(&s->auth_state, 0, sizeof(s->auth_state));
attempts =
Disfavor video streams with unknown resolution and no packets
Fixes seeking in audio-only-speex.flv
Signed-off-by: Michael Niedermayer
---
libavformat/utils.c | 20 ++--
1 file changed, 14 insertions(+), 6 deletions(-)
diff --git a/libavformat/utils.c b/libavformat/utils.c
ind
On Fri, Aug 01, 2014 at 03:53:12PM +0200, Nedeljko Babic wrote:
> From: Djordje Pesut
>
> Move the existing code to a new template file.
If code was only moved (I have not checked), I don't think
adding your copyright is appropriate.
___
ffmpeg-devel m
On Fri, Aug 01, 2014 at 03:53:08PM +0200, Nedeljko Babic wrote:
> +#if !defined(_AAC_FLOAT_EMU_)
> +#define _AAC_FLOAT_EMU_
That is not out usual style
1) We use #ifndef
2) Identifiers starting with _ + upper case letter are reserved by POSIX
and never should be used
3) The naming convention is li
Von meinem iPhone gesendet
> Am 01.08.2014 um 14:45 schrieb Carl Eugen Hoyos :
>
> Daniel Oberhoff gmail.com> writes:
>
>> And how about that test/data/fate/filter-pixfmts-rectification.rep
>> file?
>
> Imo, please send a patch without fate test first,
> this can still be done.
>
Kinda a
On Fri, Aug 01, 2014 at 03:31:30PM +0200, Benoit Fouet wrote:
> ---
> libavcodec/h264_mp4toannexb_bsf.c | 16
> 1 file changed, 8 insertions(+), 8 deletions(-)
applied
thanks
[...]
--
Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB
The greatest way to
Additional information related to my first message.
Building with gcc produces no trace of avfoundation support in full help.
Building with clang shows...
AVFoundation input device AVOptions:
-list_devices .D.. list available devices (from 0 to 1)
(default 0)
true
Summary: No clang or gcc invocation incantation seems to produce
avfoundation indev support.
Need someone smarter than me to suggest how I can debug this, to
produce a Portfile fix for MacPorts.
Thanks.
___
Yesterday I fired up ffmpeg to capture a MacBook Pro iSight Webcam stream.
Sounds e
Andreas Cadhalpun wrote:
Hi all,
some of you may have noticed a weird ffmpeg package in the NEW queue[1].
Let me explain:
... ... ...
and make kittens cry...
... ... ...
Primary reason for Debian kittens crying right now... is ffmpeg being missing...
From: Djordje Pesut
Add fixed poind code.
Signed-off-by: Nedeljko Babic
---
libavcodec/aac.h | 51 +---
libavcodec/aac_defines.h | 77 ++
libavcodec/aacsbr.c | 3 +
libavcodec/aacsbr.h | 13 +-
libavcodec/aacsbr_fixed.c| 555 ++
From: Djordje Pesut
Add fixed point implementation.
Signed-off-by: Nedeljko Babic
---
libavcodec/aac_defines.h | 36 +
libavcodec/aacps.c | 350 ++---
libavcodec/aacps.h | 28 ++--
libavcodec/aacps_fixed.c | 25 +++
From: Jovan Zelincevic
Build system modified.
Signed-off-by: Nedeljko Babic
---
libavcodec/Makefile | 19 +++
1 file changed, 11 insertions(+), 8 deletions(-)
diff --git a/libavcodec/Makefile b/libavcodec/Makefile
index 5931554..2f68cf5 100644
--- a/libavcodec/Makefile
+++ b/l
From: Djordje Pesut
Move the existing code to a new template file.
Signed-off-by: Nedeljko Babic
---
libavcodec/aacsbr.c | 1418 --
libavcodec/aacsbr.h | 45 ++
libavcodec/aacsbr_template.c | 1406 +
From: Jovan Zelincevic
Add fixed point implementation of functions for generating tables.
Signed-off-by: Nedeljko Babic
---
libavcodec/aacps_fixed_tablegen.c | 94 +
libavcodec/aacps_fixed_tablegen.h | 429 ++
2 files changed, 523 insertions(+)
cre
From: Djordje Pesut
Move existing code to a new file.
Signed-off-by: Nedeljko Babic
---
libavcodec/aacps.c | 157 -
libavcodec/aacps_float.c | 177 +++
2 files changed, 177 insertions(+), 157 deletions(-)
From: Djordje Pesut
Add fixed point implementation
Signed-off-by: Nedeljko Babic
---
libavcodec/aac.h | 102 --
libavcodec/aacdec.c | 5 +
libavcodec/aacdec_fixed.c| 458 ++
libavcodec/aacdec_template.c | 461 ++
From: Jovan Zelincevic
Build system modified.
Signed-off-by: Nedeljko Babic
---
libavcodec/Makefile | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/libavcodec/Makefile b/libavcodec/Makefile
index 3ce317d..5931554 100644
--- a/libavcodec/Makefile
+++ b/libavcodec/Make
From: Jovan Zelincevic
Signed-off-by: Nedeljko Babic
---
Changelog | 1 +
doc/general.texi| 2 +-
doc/mips.txt| 6 ++
libavutil/version.h | 2 +-
4 files changed, 9 insertions(+), 2 deletions(-)
diff --git a/Changelog b/Changelog
index 067f72a..c786d19 100644
--- a/Ch
From: Jovan Zelincevic
Converte tables to fixed point.
Signed-off-by: Nedeljko Babic
---
libavcodec/aacsbrdata.h | 688
1 file changed, 344 insertions(+), 344 deletions(-)
diff --git a/libavcodec/aacsbrdata.h b/libavcodec/aacsbrdata.h
index 125
From: Jovan Zelincevic
Add fixed point implementation of functions for generating tables
Signed-off-by: Nedeljko Babic
---
libavcodec/aacdectab.h | 34 +--
libavcodec/aactab.c| 486 +
libavcodec/aactab.h|
From: Djordje Pesut
Add float emulation
Signed-off-by: Nedeljko Babic
---
libavcodec/aac_float_emu.h | 698 +
libavcodec/float_emu.h | 400 ++
libavcodec/float_emu_tab.c | 296 +++
3 files changed, 1394 ins
From: Jovan Zelincevic
Build system modified
Signed-off-by: Nedeljko Babic
---
configure | 1 +
libavcodec/Makefile | 16 +---
libavcodec/aacdec.c | 1 -
libavcodec/aacdec_fixed.c | 1 -
libavcodec/allcodecs.c| 1 +
5 files changed, 15 insertions
Few months ago I sent implementation of fixed point AAC decoder for review.
Since it was sent in parallel with implementation of fixed point AC3 decoder,
we decided to first finish the work on AC3 decoder and then to continue with
AAC using experience gained from submission process for AC3.
W
On 31 July 2014 19:13, Michael Niedermayer wrote:
> On Mon, Jul 28, 2014 at 02:41:54PM +0100, Kieran Kunhya wrote:
>> ---
>> libavcodec/opus.c| 11 +---
>> libavcodec/opus.h|9 +++
>> libavcodec/opus_parser.c | 138
>> +-
>> liba
---
libavcodec/h264_mp4toannexb_bsf.c | 16
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/libavcodec/h264_mp4toannexb_bsf.c
b/libavcodec/h264_mp4toannexb_bsf.c
index e0c1385..739ff95 100644
--- a/libavcodec/h264_mp4toannexb_bsf.c
+++ b/libavcodec/h264_mp4toannexb_
Daniel Oberhoff gmail.com> writes:
> And how about that test/data/fate/filter-pixfmts-rectification.rep
> file?
Imo, please send a patch without fate test first,
this can still be done.
Did you already explain if / how this filter is
different from / better than the vignette filter?
Carl Eu
Am 01.08.2014 um 13:07 schrieb Carl Eugen Hoyos :
> Daniel Oberhoff gmail.com> writes:
>
>> all work now, except vuy444p
>
> It's yuv444p
oops :).
And how about that test/data/fate/filter-pixfmts-rectification.rep file? Should
I commit that?
Best
__
Daniel Oberhoff gmail.com> writes:
> all work now, except vuy444p
It's yuv444p
Carl Eugen
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Am 01.08.2014 um 12:22 schrieb Clément Bœsch :
> On Fri, Aug 01, 2014 at 12:13:22PM +0200, Daniel Oberhoff wrote:
>>
>> Am 29.07.2014 um 09:54 schrieb Carl Eugen Hoyos :
>>
>>> Daniel Oberhoff gmail.com> writes:
>>>
OBJS-$(CONFIG_ZOOMPAN_FILTER)+= vf_zoompan.o
+OBJS-
On Thu, Jul 31, 2014 at 02:32:02PM +0200, Benoit Fouet wrote:
> source index, as well as dest one, is unconditionnaly set afterwards,
> before being effectively used.
> ---
> libavcodec/h264.c | 1 -
> 1 file changed, 1 deletion(-)
>
> diff --git a/libavcodec/h264.c b/libavcodec/h264.c
> index 10
On Fri, Aug 01, 2014 at 01:54:14AM +0200, Michael Niedermayer wrote:
> On Thu, Jul 31, 2014 at 03:40:51PM +0200, Benoit Fouet wrote:
> > In order not to break a sequence like "SPS IDR SPS IDR", the boolean
> > telling that the SPS/PPS has been seen should always be set.
> > ---
> > libavcodec/h264
On Fri, Aug 01, 2014 at 12:13:22PM +0200, Daniel Oberhoff wrote:
>
> Am 29.07.2014 um 09:54 schrieb Carl Eugen Hoyos :
>
> > Daniel Oberhoff gmail.com> writes:
> >
> >> OBJS-$(CONFIG_ZOOMPAN_FILTER)+= vf_zoompan.o
> >> +OBJS-$(CONFIG_RECTIFICATION_FILTER) += vf_rectific
On Fri, Aug 1, 2014 at 12:19 PM, Clément Bœsch wrote:
> On Fri, Aug 01, 2014 at 12:10:50PM +0200, Daniel Oberhoff wrote:
> >
> > Am 29.07.2014 um 09:43 schrieb Clément Bœsch :
> >
> > > On Tue, Jul 29, 2014 at 08:04:50AM +0200, Daniel Oberhoff wrote:
> > >> Updated patch:
> >
> > …
> >
> > >> +st
On Fri, Aug 01, 2014 at 12:10:50PM +0200, Daniel Oberhoff wrote:
>
> Am 29.07.2014 um 09:43 schrieb Clément Bœsch :
>
> > On Tue, Jul 29, 2014 at 08:04:50AM +0200, Daniel Oberhoff wrote:
> >> Updated patch:
>
> …
>
> >> +static av_cold int init(AVFilterContext *ctx)
> >> +{
> >> +return 0;
Am 29.07.2014 um 09:54 schrieb Carl Eugen Hoyos :
> Daniel Oberhoff gmail.com> writes:
>
>> OBJS-$(CONFIG_ZOOMPAN_FILTER)+= vf_zoompan.o
>> +OBJS-$(CONFIG_RECTIFICATION_FILTER) += vf_rectification.o
>
>> REGISTER_FILTER(ZOOMPAN,zoompan,vf);
>> +
Am 29.07.2014 um 09:43 schrieb Clément Bœsch :
> On Tue, Jul 29, 2014 at 08:04:50AM +0200, Daniel Oberhoff wrote:
>> Updated patch:
…
>> +static av_cold int init(AVFilterContext *ctx)
>> +{
>> +return 0;
>> +}
>> +
>> +static av_cold void uninit(AVFilterContext *ctx)
>> +{
>> +}
>> +
>
> I
On Fri, Aug 01, 2014 at 01:54:57AM +0200, Carl Eugen Hoyos wrote:
> Hi!
>
> When transcoding an input file with subtitles to mkv, ffmpeg by default tries
> to encode the subtitles even if the input contains bitmap subtitles.
> Attached patch should fix this issue reported in ticket #3819 (and bef
40 matches
Mail list logo
