Le decadi 30 messidor, an CCXXIII, Paul B Mahol a écrit :
> Subject: can amerge use .needs_fifo?
.needs_fifo is evil, it breaks normal scheduling and thus can cause
unlimited memory consumption even in normal filter graphs.
Why do you want to know? Is there a particular case that is not working f
Le decadi 30 messidor, an CCXXIII, compn a écrit :
> I also trust JB.
>
> But I remember the libav fork and some vlc devels/admins during that
> time.
>
> http://ffmpeg.org/pipermail/ffmpeg-devel/2011-January/106458.html
>
> i think vlc would be a good host as long as JB is there.
Sorry for not
Dana 19. 7. 2015. 12:28 osoba "Nicolas George" napisala
je:
>
> Le decadi 30 messidor, an CCXXIII, Paul B Mahol a écrit :
> > Subject: can amerge use .needs_fifo?
>
> .needs_fifo is evil, it breaks normal scheduling and thus can cause
> unlimited memory consumption even in normal filter graphs.
>
On Thu, Jul 16, 2015 at 09:01:53PM +0100, Derek Buitenhuis wrote:
> On Thu, Jul 16, 2015 at 7:43 PM, Clément Bœsch wrote:
> >> +Reverses a clip. Requires memory to buffer the entire clip, so trimming
> >> is suggested.
> >
> > We use infinitive form, so "Reverse".
>
> Done locally.
>
> > You mi
On Sun, Jul 19, 2015 at 1:13 PM, Clément Bœsch wrote:
>> Is there any reasonable way to determine when to print such a warning? Seems
>> silly to warn over e.g. 40 frames.
>
> I meant in the documentation. I don't know for the code.
OK.
>> How does this look:
>>
>> -vf trim=end=10,reverse
>>
Chris Spencer gmail.com> writes:
> -On Windows, you need to run the IDL files through command{widl}.
> +On Windows, you need to run the IDL files through command{midl}.
Is this intended?
> -DeckLink is very picky about the formats it supports. Pixel format is
> +DeckLink is very picky abou
Hi Carl,
On 19 July 2015 at 13:53, Carl Eugen Hoyos wrote:
>
> Chris Spencer gmail.com> writes:
>
> > -On Windows, you need to run the IDL files through command{widl}.
> > +On Windows, you need to run the IDL files through command{midl}.
>
> Is this intended?
Yes, widl doesn't exist. The to
On 19 July 2015 at 14:08, Chris Spencer wrote:
> The Blackmagic documentation is kind of strange on this one actually.
> For bmdFormat8BitBGRA it says "alpha channel is valid". For
> bmdFormat8BitBGRA it says "alpha channel may be valid". Not sure how
> best to deal with that to be honest.
Sorry,
Hi Donny,
On Sat, Jul 18, 2015 at 9:48 PM, Donny Yang wrote:
> +uint32_t x, y;
> +uint32_t leftmost_x = input->width;
> +uint32_t rightmost_x = 0;
> +uint32_t topmost_y = input->height;
> +uint32_t bottommost_y = 0;
> +const uint8_t *input_data = input->data[0];
> +ui
Chris Spencer gmail.com> writes:
> > > -On Windows, you need to run the IDL files
> > > through command{widl}.
> > > +On Windows, you need to run the IDL files
> > > through command{midl}.
> >
> > Is this intended?
>
> Yes, widl doesn't exist. The tool is called midl.
Then please send a s
On 19 July 2015 at 14:30, Carl Eugen Hoyos wrote:
> Chris Spencer gmail.com> writes:
>
>> > > -On Windows, you need to run the IDL files
>> > > through command{widl}.
>> > > +On Windows, you need to run the IDL files
>> > > through command{midl}.
>> >
>> > Is this intended?
>>
>> Yes, widl do
Chris Spencer gmail.com> writes:
> Ok so with the HDMI input connected to my graphics card
> the alpha channel is zero with both ARGB and BGRA
Then please (only) use 0RGB and BGR0.
> I don't really know what the codec_tag does; I just
> copied the values from raw.c. What do I need to do
> wi
Chris Spencer gmail.com> writes:
> + This option is deprecated, use
> + option{pixel_format=yuv422p10} instead.
I just remembered this mail:
http://thread.gmane.org/gmane.comp.video.ffmpeg.devel/190657
v4l2 has an option input_format that accepts both
(for decklink) "v210" and "uyvy422" as st
On 19 July 2015 at 15:13, Carl Eugen Hoyos wrote:
> Chris Spencer gmail.com> writes:
>
>> + This option is deprecated, use
>> + option{pixel_format=yuv422p10} instead.
>
> I just remembered this mail:
> http://thread.gmane.org/gmane.comp.video.ffmpeg.devel/190657
>
> v4l2 has an option input_for
Hi,
I'm having trouble accessing a client's private data in my sample
application. After forking I have an AVIOContext client, but how do I
access its HTTPContext's AVOption fields? For example, how do I get
the method or how do I set the http_code?
Regards,
Stephan
__
On Sun, Jul 19, 2015 at 12:59:34PM +0200, Nicolas George wrote:
> Le decadi 30 messidor, an CCXXIII, compn a écrit :
> > I also trust JB.
> >
> > But I remember the libav fork and some vlc devels/admins during that
> > time.
> >
> > http://ffmpeg.org/pipermail/ffmpeg-devel/2011-January/106458.htm
On Sun, Jul 19, 2015 at 05:43:41PM +0200, Michael Niedermayer wrote:
[...]
> we could even move trac to the 3.7ghz i7 hetzner box. if we wanted
sorry mixed them up, hetzner is a E3-1246 v3 @ 3.50GHz
the OVH one would be E5-1620 v2 3.7ghz
not that it matters
[...]
--
Michael GnuPG fingerprin
---
This is an attempt to reproduce the exact same behaviour as the
selective color tool in photoshop (and maybe other adobe tool). Since
I didn't find time to write the documentation yet, the filter can be
summarized with something like: it allows you to adjust cyan, magenta,
yellow on only a cert
Signed-off-by: Paul B Mahol
---
Now with example.
---
doc/filters.texi | 57 +++
libavfilter/Makefile | 1 +
libavfilter/af_sidechaincompress.c | 328 +
libavfilter/allfilters.c | 1 +
4 files changed, 387 inse
On Sun, Jul 19, 2015 at 03:08:33AM +, Donny Yang wrote:
> Signed-off-by: Donny Yang
> ---
> libavcodec/pngdec.c | 10 +-
> 1 file changed, 5 insertions(+), 5 deletions(-)
this breaks playback of clock.png
[...]
--
Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0
On Sat, Jul 18, 2015 at 01:59:09PM +0300, Vesselin Bontchev wrote:
> Thanks for all the feedback.
>
> Vesselin
> doc/general.texi |2
> doc/muxers.texi|7 ++
> libavformat/isom.h |8 +++
> libavformat/mov.c | 128
> +
> 4
On Sat, Jul 18, 2015 at 08:20:24PM +0100, Chris Spencer wrote:
> GetBufferedAudioSampleFrameCount() used to take an unsigned long, but this
> changed at some point and the latest version of the DeckLink SDK (10.4.1)
> takes an unsigned int.
>
> Signed-off-by: Chris Spencer
> ---
> libavdevice/
Signed-off-by: Donny Yang
---
libavcodec/pngdec.c | 11 ++-
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/libavcodec/pngdec.c b/libavcodec/pngdec.c
index eae778b..cb1cebb 100644
--- a/libavcodec/pngdec.c
+++ b/libavcodec/pngdec.c
@@ -856,13 +856,13 @@ static int decode_fct
On Sat, Jul 18, 2015 at 08:20:25PM +0100, Chris Spencer wrote:
> Signed-off-by: Chris Spencer
> ---
> configure | 2 ++
> 1 file changed, 2 insertions(+)
applied
thanks
[...]
--
Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB
I am the wisest man alive, for I know one
On 19 July 2015 at 19:31, Michael Niedermayer wrote:
> does this not need some version check so its correct for older SDK too
> ?
Probably a good idea, yes. I've just checked and it changed in 10.4 so
I'll put that in.
Chris
___
ffmpeg-devel mailing li
On Sun, Jul 19, 2015 at 06:34:06PM +, Donny Yang wrote:
> Signed-off-by: Donny Yang
> ---
> libavcodec/pngdec.c | 11 ++-
> 1 file changed, 6 insertions(+), 5 deletions(-)
works with the files i tested
ill leave review and apply to paul
thx
[...]
--
Michael GnuPG fingerprint: 9
Le primidi 1er thermidor, an CCXXIII, Stephan Holljes a écrit :
> I'm having trouble accessing a client's private data in my sample
> application. After forking I have an AVIOContext client, but how do I
> access its HTTPContext's AVOption fields? For example, how do I get
> the method or how do I
In version 10.4 of the DeckLink SDK, GetBufferedAudioSampleFrameCount() was
changed to take an unsigned int instead of an unsigned long.
Signed-off-by: Chris Spencer
---
libavdevice/decklink_common.h | 6 ++
1 file changed, 6 insertions(+)
diff --git a/libavdevice/decklink_common.h b/libav
Signed-off-by: Paul B Mahol
---
libavfilter/af_amerge.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavfilter/af_amerge.c b/libavfilter/af_amerge.c
index 0a0a79f..62a11f7 100644
--- a/libavfilter/af_amerge.c
+++ b/libavfilter/af_amerge.c
@@ -32,7 +32,7 @@
#include "buff
Le primidi 1er thermidor, an CCXXIII, Paul B Mahol a écrit :
> Signed-off-by: Paul B Mahol
> ---
> libavfilter/af_amerge.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
Ok.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
__
On 7/19/15, Donny Yang wrote:
> ---
> libavcodec/pngdec.c | 47 +++
> 1 file changed, 27 insertions(+), 20 deletions(-)
>
Do you have samples this fix or way how to generate such samples?
___
ffmpeg-devel mai
On 7/19/15, Donny Yang wrote:
> Signed-off-by: Donny Yang
> ---
> libavcodec/pngdec.c | 10 +-
> 1 file changed, 5 insertions(+), 5 deletions(-)
>
> diff --git a/libavcodec/pngdec.c b/libavcodec/pngdec.c
> index eae778b..7a5c464 100644
> --- a/libavcodec/pngdec.c
> +++ b/libavcodec/pngde
On 19/07/15 6:08 PM, Vesselin Bontchev wrote:
> diff --git a/libavutil/tea.h b/libavutil/tea.h
> new file mode 100644
> index 000..1db6987
> --- /dev/null
> +++ b/libavutil/tea.h
> @@ -0,0 +1,66 @@
> +/*
> + * A 32-bit implementation of the TEA algorithm
> + * Copyright (c) 2015 Vesselin Bontch
Hi,
I need support for TEA (Tiny Encryption Algorithm) in FFmpeg for my upcoming
patch (which makes Audible AA files playable).
Thanks,
VesselinFrom 491afe746e3a1f723798224ee56fd57a028ed4da Mon Sep 17 00:00:00 2001
From: Vesselin Bontchev
Date: Sun, 19 Jul 2015 22:25:53 +0200
Subject: [PATCH] A
Thanks for all the feedback.
VesselinFrom 6e427af7f9450856c5d96734647e760d1d0f7ce2 Mon Sep 17 00:00:00 2001
From: Vesselin Bontchev
Date: Sun, 19 Jul 2015 22:25:53 +0200
Subject: [PATCH] Add support for TEA (Tiny Encryption Algorithm)
---
libavutil/Makefile |3 +
libavutil/tea.c
On Mon, Jul 20, 2015 at 02:16:28AM +0300, Vesselin Bontchev wrote:
> Thanks for all the feedback.
>
> Vesselin
> libavutil/Makefile |3
> libavutil/tea.c | 213
> +++
> libavutil/tea.h | 71 +++
> tests/fate
From 9eac81d24472916636f2b0ad21cf8560f0acf20a Mon Sep 17 00:00:00 2001
From: ZhouXiaoyong
Date: Mon, 20 Jul 2015 11:07:05 +0800
Subject: [PATCH 2/2] avcodec: loongson relocate constants of idctdsp and
h264pred
Signed-off-by: ZhouXiaoyong
---
libavcodec/mips/constants.c| 5
libavcode
From 9c95155e90ff5d083e56cdd2565792c9e314302a Mon Sep 17 00:00:00 2001
From: ZhouXiaoyong
Date: Mon, 20 Jul 2015 10:58:30 +0800
Subject: [PATCH 1/2] avcodec: loongson constants redefined with macros
Signed-off-by: ZhouXiaoyong
---
libavcodec/mips/constants.c | 52 ++
Signed-off-by: James Almer
---
libavfilter/x86/vf_ssim.asm | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libavfilter/x86/vf_ssim.asm b/libavfilter/x86/vf_ssim.asm
index be6a4d3..6661987 100644
--- a/libavfilter/x86/vf_ssim.asm
+++ b/libavfilter/x86/vf_ssim.asm
@@ -97,8 +
~20% faster than ssse3. Also enabled for x86_32
Signed-off-by: James Almer
---
libavfilter/x86/vf_ssim.asm| 62 --
libavfilter/x86/vf_ssim_init.c | 5
2 files changed, 64 insertions(+), 3 deletions(-)
diff --git a/libavfilter/x86/vf_ssim.asm b/l
sorry, the last patch aborded, please review this one.
because it could avoid to use load when use immediate value ff_pb_80 in
idctdsp_mmi.c.
---
From 40399677fd67087db950c7f0f8ca382e5bc2cfd2 Mon Sep 17 00:00:00 2001
From: ZhouXiaoyong
Date: Mon, 20 Jul 2015 11:07:05 +0800
Subject: [PATCH 2/2] a
41 matches
Mail list logo
