Only two functions that use xop multiply-accumulate instructions where the
first operand is the same as the fourth actually took advantage of the macros.
This further reduces differences with x264's x86inc.
Signed-off-by: James Almer
---
libavcodec/x86/flacdsp.asm | 9 +
libavutil/
On Mon, Aug 3, 2015 at 8:04 AM, Clément Bœsch wrote:
> naive question, what about the few SECTION_TEXT 32?
I don't know why 32-byte alignment was used for the text section in
the first place. Functions are only 16-byte aligned anyway.
___
ffmpeg-devel ma
On Sun, Aug 02, 2015 at 10:40:02PM +0200, Henrik Gramner wrote:
> The .text section is already 16-byte aligned by default on all supported
> platforms so `SECTION_TEXT` isn't any different from `SECTION .text`.
naive question, what about the few SECTION_TEXT 32?
[...]
--
Clément B.
pgpiCW7CJ2
On Sun, Aug 02, 2015 at 07:45:57PM -0400, Ronald S. Bultje wrote:
> Hi,
>
> On Sun, Aug 2, 2015 at 4:40 PM, Henrik Gramner wrote:
>
> > ---
> > libavutil/x86/x86inc.asm | 32 +---
> > 1 file changed, 21 insertions(+), 11 deletions(-)
>
>
> lgtm.
applied
Henrik, i
On Sun, Aug 02, 2015 at 07:36:51PM -0400, Ronald S. Bultje wrote:
> Hi,
>
> On Sun, Aug 2, 2015 at 4:40 PM, Henrik Gramner wrote:
>
> > The bug was fixed in 1.3.0, so only perform the workaround in earlier
> > versions.
> > ---
> > libavutil/x86/x86inc.asm | 20 +++-
> > 1 file
Hi,
On Sun, Aug 2, 2015 at 4:40 PM, Henrik Gramner wrote:
> Change ALLOC_STACK to always align the stack before allocating stack space
> for
> consistency. Previously alignment would occur either before or after
> allocating
> stack space depending on whether manual alignment was required or not
Hi,
On Sun, Aug 2, 2015 at 7:56 PM, James Almer wrote:
> On 02/08/15 5:40 PM, Henrik Gramner wrote:
> > Change ALLOC_STACK to always align the stack before allocating stack
> space for
> > consistency. Previously alignment would occur either before or after
> allocating
> > stack space depending
On 02/08/15 5:40 PM, Henrik Gramner wrote:
> Change ALLOC_STACK to always align the stack before allocating stack space for
> consistency. Previously alignment would occur either before or after
> allocating
> stack space depending on whether manual alignment was required or not.
> ---
> libavcod
Hi,
On Sun, Aug 2, 2015 at 4:40 PM, Henrik Gramner wrote:
> ---
> libavutil/x86/x86inc.asm | 32 +---
> 1 file changed, 21 insertions(+), 11 deletions(-)
lgtm.
Ronald
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
Hi,
On Sun, Aug 2, 2015 at 4:40 PM, Henrik Gramner wrote:
> The bug was fixed in 1.3.0, so only perform the workaround in earlier
> versions.
> ---
> libavutil/x86/x86inc.asm | 20 +++-
> 1 file changed, 11 insertions(+), 9 deletions(-)
>
> diff --git a/libavutil/x86/x86inc.asm
Hi,
On Sun, Aug 2, 2015 at 4:40 PM, Henrik Gramner wrote:
> The .text section is already 16-byte aligned by default on all supported
> platforms so `SECTION_TEXT` isn't any different from `SECTION .text`.-;
> Overrides the default .text section.
> -; Silences warnings when defining structures.
>
Signed-off-by: James Almer
---
tools/crypto_bench.c | 4
1 file changed, 4 insertions(+)
diff --git a/tools/crypto_bench.c b/tools/crypto_bench.c
index 5af2dc1..5ae4715 100644
--- a/tools/crypto_bench.c
+++ b/tools/crypto_bench.c
@@ -106,6 +106,7 @@ static void run_lavu_ ## suffix(uint8_t *
OpenSSL and Tomcrypt des modules are not benched together as there's
some naming collisions between the two.
Signed-off-by: James Almer
---
tools/crypto_bench.c | 58
1 file changed, 58 insertions(+)
diff --git a/tools/crypto_bench.c b/tools/
Signed-off-by: James Almer
---
tools/crypto_bench.c | 24
1 file changed, 24 insertions(+)
diff --git a/tools/crypto_bench.c b/tools/crypto_bench.c
index 87e99d0..fb749ac 100644
--- a/tools/crypto_bench.c
+++ b/tools/crypto_bench.c
@@ -80,6 +80,7 @@ struct hash_impl {
#
Signed-off-by: James Almer
---
tools/crypto_bench.c | 24
1 file changed, 24 insertions(+)
diff --git a/tools/crypto_bench.c b/tools/crypto_bench.c
index d1e7c30..87e99d0 100644
--- a/tools/crypto_bench.c
+++ b/tools/crypto_bench.c
@@ -79,6 +79,7 @@ struct hash_impl {
#
Signed-off-by: James Almer
---
tools/crypto_bench.c | 45 +
1 file changed, 45 insertions(+)
diff --git a/tools/crypto_bench.c b/tools/crypto_bench.c
index 79629bc..d1e7c30 100644
--- a/tools/crypto_bench.c
+++ b/tools/crypto_bench.c
@@ -75,6 +75,7 @@
On date Friday 2015-07-31 15:53:04 +0200, Michael Niedermayer wrote:
> Hi all
>
> Ive been in FFmepg since 14 years and been the leader since 11 years
> and i feel that iam not the best person for the leader position.
> I had hoped for a long time that the fork situation would resolve and
> both s
On Sun, Aug 2, 2015 at 11:52 PM, wm4 wrote:
> On Sun, 2 Aug 2015 23:45:36 +0200
> Hendrik Leppkes wrote:
>
>> On Sun, Aug 2, 2015 at 11:27 PM, Ivan Uskov wrote:
>> > Hello Michael,
>> >
>> > Sunday, August 2, 2015, 9:46:23 PM, you wrote:
>> >>> MN> it appears the file was not in mfx_dispatch pre
On Sun, 2 Aug 2015 23:45:36 +0200
Hendrik Leppkes wrote:
> On Sun, Aug 2, 2015 at 11:27 PM, Ivan Uskov wrote:
> > Hello Michael,
> >
> > Sunday, August 2, 2015, 9:46:23 PM, you wrote:
> >>> MN> it appears the file was not in mfx_dispatch previously
> >>> MN> so a check in confgure might be neede
On Sun, Aug 2, 2015 at 11:27 PM, Ivan Uskov wrote:
> Hello Michael,
>
> Sunday, August 2, 2015, 9:46:23 PM, you wrote:
>>> MN> it appears the file was not in mfx_dispatch previously
>>> MN> so a check in confgure might be needed
>>> As I can see here
>>> https://github.com/lu-zero/mfx_dispatch/tre
On Sun, Aug 02, 2015 at 05:35:59PM +0200, Paul B Mahol wrote:
> Signed-off-by: Paul B Mahol
> ---
> doc/filters.texi | 61
> libavfilter/Makefile | 1 +
> libavfilter/allfilters.c | 1 +
> libavfilter/vf_convolution.c | 337
>
Hello Michael,
Sunday, August 2, 2015, 9:46:23 PM, you wrote:
>> MN> it appears the file was not in mfx_dispatch previously
>> MN> so a check in confgure might be needed
>> As I can see here
>> https://github.com/lu-zero/mfx_dispatch/tree/master/mfx
>> The mfxjpeg.h was added 17 days ago and marke
---
libavutil/x86/x86inc.asm | 32 +---
1 file changed, 21 insertions(+), 11 deletions(-)
diff --git a/libavutil/x86/x86inc.asm b/libavutil/x86/x86inc.asm
index 49b4107..8dfc24f 100644
--- a/libavutil/x86/x86inc.asm
+++ b/libavutil/x86/x86inc.asm
@@ -1,7 +1,7 @@
;
The .text section is already 16-byte aligned by default on all supported
platforms so `SECTION_TEXT` isn't any different from `SECTION .text`.
---
libavcodec/x86/aacpsdsp.asm | 2 +-
libavcodec/x86/audiodsp.asm | 2 +-
libavcodec/x86/blockdsp.asm | 2 +-
libavcodec/x8
Change ALLOC_STACK to always align the stack before allocating stack space for
consistency. Previously alignment would occur either before or after allocating
stack space depending on whether manual alignment was required or not.
---
libavcodec/x86/h264_deblock.asm | 4 +--
libavutil/x86/x86inc.a
The bug was fixed in 1.3.0, so only perform the workaround in earlier versions.
---
libavutil/x86/x86inc.asm | 20 +++-
1 file changed, 11 insertions(+), 9 deletions(-)
diff --git a/libavutil/x86/x86inc.asm b/libavutil/x86/x86inc.asm
index e176715..8581551 100644
--- a/libavutil/x
This brings x86inc.asm in libavutil up to date with x86inc.asm in x264.
They're not 100% identical but the difference is tiny compared to before.
Henrik Gramner (4):
x86inc: Support arbitrary stack alignments
x86inc: Disable vpbroadcastq workaround in newer yasm versions
x86inc: Drop SECTIO
On Sun, 2 Aug 2015 20:27:17 +0200
wm4 wrote:
> On Sun, 2 Aug 2015 15:31:07 +0300
> Ivan Uskov wrote:
>
> > Hello All,
> >
> > The attached patch adds QSV-based mjpeg video decoder.
> > Please review.
> >
> >
>
> Is mjpeg decoding so important that we need QSV decoding of it?
some ip vide
Hello wm4,
Sunday, August 2, 2015, 9:38:33 PM, you wrote:
>> w> Is mjpeg decoding so important that we need QSV decoding of it?
>> Why not? It is for free.
w> Having to maintain additional code has a cost, though.
Near about zero, since qsv core common for all formats.
--
Best regards,
Ivan
On Sun, Aug 02, 2015 at 09:23:21PM +0300, Ivan Uskov wrote:
> Hello Michael,
>
> Sunday, August 2, 2015, 8:55:45 PM, you wrote:
>
> >> +#if QSV_VERSION_ATLEAST(1, 3)
> >> +#include
> >> +#endif
>
> MN> this seems not working
>
> MN> CC libavcodec/qsv.o
> MN> libavcodec/qsv.c:33:25: fatal
On Sun, 2 Aug 2015 21:36:15 +0300
Ivan Uskov wrote:
> Hello wm4,
>
> Sunday, August 2, 2015, 9:27:17 PM, you wrote:
>
> w> Is mjpeg decoding so important that we need QSV decoding of it?
> Why not? It is for free.
Having to maintain additional code has a cost, though.
_
Hello wm4,
Sunday, August 2, 2015, 9:27:17 PM, you wrote:
w> Is mjpeg decoding so important that we need QSV decoding of it?
Why not? It is for free.
--
Best regards,
Ivanmailto:ivan.us...@nablet.com
___
ffmpeg-devel mai
On Sun, 2 Aug 2015 15:31:07 +0300
Ivan Uskov wrote:
> Hello All,
>
> The attached patch adds QSV-based mjpeg video decoder.
> Please review.
>
>
Is mjpeg decoding so important that we need QSV decoding of it?
___
ffmpeg-devel mailing list
ffmpeg-d
Hello Michael,
Sunday, August 2, 2015, 8:55:45 PM, you wrote:
>> +#if QSV_VERSION_ATLEAST(1, 3)
>> +#include
>> +#endif
MN> this seems not working
MN> CC libavcodec/qsv.o
MN> libavcodec/qsv.c:33:25: fatal error: mfx/mfxjpeg.h: No such file or
directory
MN> #include
MN>
On Sun, Aug 02, 2015 at 06:57:15PM +0200, Paul B Mahol wrote:
> On 7/24/15, Michael Niedermayer wrote:
> > On Fri, Jul 24, 2015 at 07:09:16AM +, Paul B Mahol wrote:
[...]
> >> +static int filter_frame(AVFilterLink *inlink, AVFrame *inpicref)
> >> +{
> >> +AVFilterContext *ctx = inlink->dst
On Sun, Aug 02, 2015 at 03:31:07PM +0300, Ivan Uskov wrote:
> Hello All,
>
> The attached patch adds QSV-based mjpeg video decoder.
> Please review.
>
>
> --
> Best regards,
> Ivan mailto:ivan.us...@nablet.com
> Changelog |1
> configure
On 7/24/15, Michael Niedermayer wrote:
> On Fri, Jul 24, 2015 at 07:09:16AM +, Paul B Mahol wrote:
> [..]
>
>> +static av_cold void uninit(AVFilterContext *ctx)
>> +{
>> +FrameRateContext *s = ctx->priv;
>> +int i;
>> +
>> +for (i = s->frst + 1; i > s->last; i++) {
>> +if (
Signed-off-by: Paul B Mahol
---
doc/filters.texi | 61
libavfilter/Makefile | 1 +
libavfilter/allfilters.c | 1 +
libavfilter/vf_convolution.c | 337 +++
4 files changed, 400 insertions(+)
create mode 100644 libavfilt
On 8/1/15, Carl Eugen Hoyos wrote:
> Hi!
>
> Attached patch fixes ticket #4747 for me, I don't know how to
> detect that the wave atom contains no frma / alac atom...
And that is mentioned because?
> Please comment, Carl Eugen
___
ffmpeg-devel mailing
Hello All,
The attached patch adds QSV-based mjpeg video decoder.
Please review.
--
Best regards,
Ivan mailto:ivan.us...@nablet.com
0001-QSV-MJPEG-video-decoder-has-been-added.patch
Description: Binary data
___
ffmpeg-deve
Hi,
On Sun, Aug 2, 2015 at 7:10 AM, Hendrik Leppkes wrote:
> On Sun, Aug 2, 2015 at 1:07 PM, Hendrik Leppkes
> wrote:
> > It needs to point to the value from the sps rps, not the final computed
> one from the slice header.
> > ---
> > libavcodec/dxva2_hevc.c | 2 +-
> > libavcodec/hevc.h
02.08.2015, 07:44, "James Almer" :
> On 30/07/15 7:46 AM, Vesselin Bontchev wrote:
>> From 06b0c0013404a67c72ea14a3c90730c0c4bd5b9a Mon Sep 17 00:00:00 2001
>> From: Vesselin Bontchev
>> Date: Sun, 19 Jul 2015 23:16:36 +0200
>> Subject: [PATCH] Add support for Audible AA files
>>
>> + AVIOCon
On Sun, 2 Aug 2015 13:07:31 +0200
Hendrik Leppkes wrote:
> It needs to point to the value from the sps rps, not the final
> computed one from the slice header. ---
it looks ok to me, but i did not review code.
i would use the ^^ above as a commit message though.
maybe something like
"dxva2_hev
On Sun, Aug 2, 2015 at 1:07 PM, Hendrik Leppkes wrote:
> It needs to point to the value from the sps rps, not the final computed one
> from the slice header.
> ---
> libavcodec/dxva2_hevc.c | 2 +-
> libavcodec/hevc.h | 1 +
> libavcodec/hevc_ps.c| 1 +
> 3 files changed, 3 insertions(
It needs to point to the value from the sps rps, not the final computed one
from the slice header.
---
libavcodec/dxva2_hevc.c | 2 +-
libavcodec/hevc.h | 1 +
libavcodec/hevc_ps.c| 1 +
3 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/libavcodec/dxva2_hevc.c b/libavcodec/d
45 matches
Mail list logo
