V4: Fix rang check error in assert base on Mark's review
V3: Clean the code logic base on Michael's review.
V2: Add set_ue_golomb_long() to support 32bits UE golomb and update the unit
test.
From fa3f59e5fcb2cddcc44b0e895bfa02caa491fee5 Mon Sep 17 00:00:00 2001
From: Jun Zhao
Date: Fri, 2 Jun 201
On 2017/6/8 18:02, Mark Thompson wrote:
> On 08/06/17 01:29, Jun Zhao wrote:
>> V3: Clean the code logic base on Michael's review.
>> V2: Add Add set_ue_golomb_long() to support 32bits UE golomb and update the
>> unit test.
>>
>> From 4de3e0c30af7bc1901562000eda018a6d6849292 Mon Sep 17 00:00:00
Fixes OOM
Fixes: 2131/clusterfuzz-testcase-minimized-4718045157130240
Found-by: continuous fuzzing process
https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer
---
libavcodec/shorten.c | 4
1 file changed, 4 insertions(+)
diff --git a/libavcode
On Thu, Jun 08, 2017 at 06:35:07PM -0400, Ronald S. Bultje wrote:
> Hi,
>
> On Thu, Jun 8, 2017 at 6:10 PM, Michael Niedermayer
> wrote:
>
> > Signed value in
> > Unsigned
> > INTeger type
>
> [..]
> > Both SUINT and unsigned should produce identical binaries
>
> This seems to go against the r
---
libavfilter/vf_colorspace.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/libavfilter/vf_colorspace.c b/libavfilter/vf_colorspace.c
index d1e9a1c609..71ea08a20f 100644
--- a/libavfilter/vf_colorspace.c
+++ b/libavfilter/vf_colorspace.c
@@ -1090,6 +1090,7 @@ static const AVOption colorspa
From: Anton Mitrofanov
The use of rsp was pretty much hardcoded there and probably didn't work
otherwise with stack_size > 0.
---
libavutil/x86/x86inc.asm | 19 ++-
1 file changed, 10 insertions(+), 9 deletions(-)
diff --git a/libavutil/x86/x86inc.asm b/libavutil/x86/x86inc.asm
From: Henrik Gramner
We overload the `call` instruction with a macro, but it would misbehave when
the macro argument wasn't a valid identifier. Fix it by explicitly checking
if the argument is an identifier.
---
libavutil/x86/x86inc.asm | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
From: Henrik Gramner
Due to a peculiarity in the ModR/M addressing encoding, the r12 and r13
registers sometimes requires an additional byte when used as a base register.
r14 and r15 doesn't have that issue, so prefer using them.
---
libavutil/x86/x86inc.asm | 16
1 file change
From: Henrik Gramner
There's no point in emitting a rep prefix before ret on modern CPUs.
---
libavutil/x86/x86inc.asm | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavutil/x86/x86inc.asm b/libavutil/x86/x86inc.asm
index f2a6a3f1db..44069741cc 100644
--- a/libavutil/x86/x
From: Henrik Gramner
Simplifies writing assembly code that depends on available instructions.
LZCNT implies SSE2
BMI1 implies AVX+LZCNT
AVX2 implies BMI2
---
This is the patch I was talking about. Where should I put the aesni define?
x264 doesn't have it but I will try to get it upstreamed.
l
Incorporate some of the recent changes committed to x264. This is an initial
set with no controversial changes: no nasm requirement, no avx512.
I do want your comments on where I should put the aesni define in the last
patch. I will make a note on that one too. I will attempt to upstream that
d
On request by Rostislav Pehlivanov (atomnuker), I've been working on
SSE/AVX accelerated version of pvq_search().
The attached patch is my work so far.
At the moment, for me, at the default bitrate
the function is 2.5 times faster than the current C version.
(My cpu is Intel Westmere class.)
The
Hi,
On Thu, Jun 8, 2017 at 6:10 PM, Michael Niedermayer
wrote:
> Signed value in
> Unsigned
> INTeger type
[..]
> Both SUINT and unsigned should produce identical binaries
This seems to go against the rule that code should be as simple as possible.
Unsigned is simpler than SUINT if the outcom
On Sat, May 27, 2017 at 12:24:16PM +0200, wm4 wrote:
> On Sat, 27 May 2017 03:56:42 +0200
> Michael Niedermayer wrote:
>
> > On Fri, May 26, 2017 at 07:06:44PM -0400, Ronald S. Bultje wrote:
> > > Hi,
> > >
> > > On Fri, May 26, 2017 at 6:34 PM, Michael Niedermayer
> > > > > > wrote:
> > >
This makes the code easier to understand and allows future changes simplifying
some of the
scanf() away
Signed-off-by: Michael Niedermayer
---
libavcodec/htmlsubtitles.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/htmlsubtitles.c b/libavcodec/htmlsubtitles.c
i
Fixes Timeout
Fixes: 2127/clusterfuzz-testcase-minimized-6595787859427328
Signed-off-by: Michael Niedermayer
---
libavcodec/htmlsubtitles.c | 12 ++--
1 file changed, 10 insertions(+), 2 deletions(-)
diff --git a/libavcodec/htmlsubtitles.c b/libavcodec/htmlsubtitles.c
index 16295daa0c..
Hello,
You can create a FipsContext struct to store decoding data
instead of using Fips Header struct
(you can take a look to other image decoder)
Also you should need to split the patch for libavcodec part and libavformat
part
Nit : The coding style for if/else is
if {
} else {
}
Not
if {
On 6/8/2017 4:56 PM, Moritz Barsnick wrote:
> On Thu, Jun 08, 2017 at 22:34:15 +0530, Paras Chadha wrote:
>> +t64 = (((uint64_t) ptr8[0]) << 56) | (((uint64_t) ptr8[1]) <<
>> 48) | (((uint64_t) ptr8[2]) << 40) | (((uint64_t) ptr8[3]) << 32) | (ptr8[4]
>> << 24) | (ptr8[5] << 16) | (pt
On Thu, Jun 08, 2017 at 22:34:15 +0530, Paras Chadha wrote:
> +t64 = (((uint64_t) ptr8[0]) << 56) | (((uint64_t) ptr8[1]) <<
> 48) | (((uint64_t) ptr8[2]) << 40) | (((uint64_t) ptr8[3]) << 32) | (ptr8[4]
> << 24) | (ptr8[5] << 16) | (ptr8[6] << 8) | ptr8[7];
I think you can (or shoul
On Thu, 8 Jun 2017 12:28:41 -0700
Sasi Inguva wrote:
> Ping!
>
If nobody has another comment, I guess I can push it tomorrow or next
week.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Ping!
On Jun 6, 2017 11:17 AM, "Sasi Inguva" wrote:
> Got it. Added to the description.
>
> On Tue, Jun 6, 2017 at 6:51 AM, wm4 wrote:
>
>> On Sat, 3 Jun 2017 19:31:37 -0700
>> Sasi Inguva wrote:
>>
>> > > -// Offset the DTS by ctts[0] to make the PTS of the first frame 0
>> > > -if (c
On Wed, Jun 07, 2017 at 10:56:29PM +0200, Matthieu Bouron wrote:
> On Tue, Jun 06, 2017 at 09:39:25AM -0700, Aman Gupta wrote:
> > On Mon, Jun 5, 2017 at 1:22 PM, Aman Gupta wrote:
> >
> > > From: Aman Gupta
> > >
> > > Android TV and FireOS hardware supports mpeg2 hardware decoding via
> > > Me
On Thu, Jun 08, 2017 at 01:07:40PM +0100, Mark Thompson wrote:
> On 08/06/17 12:35, Michael Niedermayer wrote:
> > On Thu, Jun 08, 2017 at 11:02:30AM +0100, Mark Thompson wrote:
> >> On 08/06/17 01:29, Jun Zhao wrote:
> >>> V3: Clean the code logic base on Michael's review.
> >>> V2: Add Add set_ue
Hi,
On Thu, Jun 1, 2017 at 5:55 PM, Michael Niedermayer
wrote:
> On Thu, Jun 01, 2017 at 07:48:38AM -0400, Ronald S. Bultje wrote:
> > Hi,
> >
> > On Thu, Jun 1, 2017 at 2:27 AM, raymond zheng <
> raymondzheng1...@gmail.com>
> > wrote:
> >
> > > This patch is only for solving the bug introduced
It supports all 2-d images alongwith bzero, bscale and blank keywords.
RGBA images are supported as NAXIS3 = 3 or 4 i.e. Planes in RGBA order. Also
CTYPE = 'RGB ' should be present.
It currently does not support XTENSION keyword.
Signed-off-by: Paras Chadha
---
Changelog| 1 +
vp9_diag_downright_16x16_12bpp_c: 149.0
vp9_diag_downright_16x16_12bpp_sse2: 67.8
vp9_diag_downright_16x16_12bpp_ssse3: 45.6
vp9_diag_downright_16x16_12bpp_avx: 36.6
vp9_diag_downright_16x16_12bpp_avx2: 25.5
~30% faster than avx
Signed-off-by: Ilia Valiakhmetov
---
libavcodec/x86/vp9dsp_init_16
On 08/06/17 12:35, Michael Niedermayer wrote:
> On Thu, Jun 08, 2017 at 11:02:30AM +0100, Mark Thompson wrote:
>> On 08/06/17 01:29, Jun Zhao wrote:
>>> V3: Clean the code logic base on Michael's review.
>>> V2: Add Add set_ue_golomb_long() to support 32bits UE golomb and update the
>>> unit test.
On 6/8/17, wm4 wrote:
> On Thu, 8 Jun 2017 13:24:39 +0200
> Paul B Mahol wrote:
>
>> On 6/8/17, wm4 wrote:
>> > On Thu, 8 Jun 2017 11:36:02 +0200
>> > Paul B Mahol wrote:
>> >
>> >> On 6/8/17, wm4 wrote:
>> >> > On Thu, 8 Jun 2017 08:13:48 +0200
>> >> > Paul B Mahol wrote:
>> >> >
>> >> >> S
On Thu, 8 Jun 2017 13:24:39 +0200
Paul B Mahol wrote:
> On 6/8/17, wm4 wrote:
> > On Thu, 8 Jun 2017 11:36:02 +0200
> > Paul B Mahol wrote:
> >
> >> On 6/8/17, wm4 wrote:
> >> > On Thu, 8 Jun 2017 08:13:48 +0200
> >> > Paul B Mahol wrote:
> >> >
> >> >> Signed-off-by: Paul B Mahol
> >
On Thu, Jun 08, 2017 at 11:02:30AM +0100, Mark Thompson wrote:
> On 08/06/17 01:29, Jun Zhao wrote:
> > V3: Clean the code logic base on Michael's review.
> > V2: Add Add set_ue_golomb_long() to support 32bits UE golomb and update the
> > unit test.
> >
> > From 4de3e0c30af7bc1901562000eda018a6d6
On 6/8/17, wm4 wrote:
> On Thu, 8 Jun 2017 11:36:02 +0200
> Paul B Mahol wrote:
>
>> On 6/8/17, wm4 wrote:
>> > On Thu, 8 Jun 2017 08:13:48 +0200
>> > Paul B Mahol wrote:
>> >
>> >> Signed-off-by: Paul B Mahol
>> >> ---
>> >
>> > Feeding the coefficients as additional audio inputs, and using
Hi,
On Wed, Jun 7, 2017 at 11:41 PM, Vittorio Giovara <
vittorio.giov...@gmail.com> wrote:
> Signed-off-by: Vittorio Giovara
> ---
> Refreshing an old patch from last November...
>
> After this, vf_colorspace supports *all* the primaries recognized
> by libavutil.
LGTM.
Ronald
___
On 08/06/17 01:29, Jun Zhao wrote:
> V3: Clean the code logic base on Michael's review.
> V2: Add Add set_ue_golomb_long() to support 32bits UE golomb and update the
> unit test.
>
> From 4de3e0c30af7bc1901562000eda018a6d6849292 Mon Sep 17 00:00:00 2001
> From: Jun Zhao
> Date: Fri, 2 Jun 2017 1
On Thu, 8 Jun 2017 11:36:02 +0200
Paul B Mahol wrote:
> On 6/8/17, wm4 wrote:
> > On Thu, 8 Jun 2017 08:13:48 +0200
> > Paul B Mahol wrote:
> >
> >> Signed-off-by: Paul B Mahol
> >> ---
> >
> > Feeding the coefficients as additional audio inputs, and using amovie
> > to load them, seems l
On 6/8/17, wm4 wrote:
> On Thu, 8 Jun 2017 08:13:48 +0200
> Paul B Mahol wrote:
>
>> Signed-off-by: Paul B Mahol
>> ---
>
> Feeding the coefficients as additional audio inputs, and using amovie
> to load them, seems like an extremely awkward choice. It would make
> sense to me if there actually
On Thu, 8 Jun 2017 08:13:48 +0200
Paul B Mahol wrote:
> Signed-off-by: Paul B Mahol
> ---
Feeding the coefficients as additional audio inputs, and using amovie
to load them, seems like an extremely awkward choice. It would make
sense to me if there actually were a set of input coefficients per
On Wed, 7 Jun 2017 21:23:12 +0200 (CEST)
Marton Balint wrote:
> On Wed, 7 Jun 2017, wm4 wrote:
>
> > On Wed, 31 May 2017 18:26:01 +0200
> > wm4 wrote:
> >
> >> Fixes detection of some TV sample as 24.5 FPS. With the patch applied,
> >> it's detected as 25 FPS.
> >>
> >> This is enabled for m
On Wed, Jun 7, 2017, at 11:40 AM, Gyan wrote:
> On Sat, Jun 3, 2017 at 3:24 PM, Gyan wrote:
>
> > The support forum at
> >
> > http://ffmpeg.gusari.org/index.php
> >
> > which is linked to at
> >
> > http://www.ffmpeg.org/contact.html#Forums
> >
> > now sports a banner announcing its likely closur
38 matches
Mail list logo
