On 8/21/20, Paul B Mahol wrote:
> Overall speed changes for 1920x1080, yuv422p10le, 60fps from: 0.19x to
> 0.343x
>
> Signed-off-by: Paul B Mahol
Will apply soon.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo
On Fri, Aug 21, 2020 at 15:10:09 +0200, Moritz Barsnick wrote:
> Let me add my promised benchmarks. I have nothing modern here (these
> CPUs are 10, 18 and 6 years old), but I can confirm the impressive
> improvements:
>
> Intel Atom D525 Intel Pentium 4 Intel Haswell
On Fri, Aug 21, 2020 at 13:13:29 +0200, Paul B Mahol wrote:
> Overall speed changes for 1920x1080, yuv422p10le, 60fps from: 0.19x to 0.343x
Let me add my promised benchmarks. I have nothing modern here (these
CPUs are 10, 18 and 6 years old), but I can confirm the impressive
improvements:
Overall speed changes for 1920x1080, yuv422p10le, 60fps from: 0.19x to 0.343x
Signed-off-by: Paul B Mahol
---
libavcodec/Makefile | 2 +-
libavcodec/cfhd.c | 337 --
libavcodec/cfhd.h | 3 +
libavcodec/cfhddsp.c | 118 +++
libavc
On Thu, Aug 20, 2020 at 12:53:05 +0200, Moritz Barsnick wrote:
> Alas, with the patchset, the following command quickly terminates with
> Illegal instruction in ff_cfhd_horiz_filter_clip10_sse2 ():
The same on this CPU, obviously, because it also doesn't support
SSE4.1:
[barsnick@paradise ffmpeg
On 8/20/2020 7:53 AM, Moritz Barsnick wrote:
> On Sun, Aug 16, 2020 at 18:25:12 +0200, Paul B Mahol wrote:
>> On 8/16/20, Paul B Mahol wrote:
>>> Please help porting this to linux and 64bit calling convention.
>>
>> New patch attached.
>>
>> This one does not allocate stack on x32.
>
> I wanted t
On Sun, Aug 16, 2020 at 18:25:12 +0200, Paul B Mahol wrote:
> On 8/16/20, Paul B Mahol wrote:
> > Please help porting this to linux and 64bit calling convention.
>
> New patch attached.
>
> This one does not allocate stack on x32.
I wanted to benchmark on several machines (newest I have is a Hasw
On 8/17/2020 10:50 AM, Paul B Mahol wrote:
> On 8/16/20, Paul B Mahol wrote:
>> On 8/16/20, Paul B Mahol wrote:
>>> Hi,
>>>
>>> patch attached.
>>>
>>> Please help porting this to linux and 64bit calling convention.
>>>
>>
>> New patch attached.
>>
>> This one does not allocate stack on x32.
>>
>
On 8/16/20, Paul B Mahol wrote:
> On 8/16/20, Paul B Mahol wrote:
>> Hi,
>>
>> patch attached.
>>
>> Please help porting this to linux and 64bit calling convention.
>>
>
> New patch attached.
>
> This one does not allocate stack on x32.
>
This is latest patch for cfhd that should be reviewed.
Pl
On 8/17/20, Anton Khirnov wrote:
> Quoting Paul B Mahol (2020-08-14 14:24:25)
> >From 874fd9e604a6dcd55cca77c7256a633e5739da77 Mon Sep 17 00:00:00 2001
>>From: Paul B Mahol
>>Date: Sun, 9 Aug 2020 17:47:34 +0200
>>Subject: [PATCH] avcodec/cfhd: add x86 SIMD
>>
>>---
>> libavcodec/Makefile
Quoting Paul B Mahol (2020-08-14 14:24:25)
>From 874fd9e604a6dcd55cca77c7256a633e5739da77 Mon Sep 17 00:00:00 2001
>From: Paul B Mahol
>Date: Sun, 9 Aug 2020 17:47:34 +0200
>Subject: [PATCH] avcodec/cfhd: add x86 SIMD
>
>---
> libavcodec/Makefile | 2 +-
> libavcodec/cfhd.c
Hi Paul,
sorry, forgot to reply to this...
On 2020-08-15 12:55 +0200, Paul B Mahol wrote:
> On 8/15/20, Alexander Strasser wrote:
> > On 2020-08-14 20:22 +0100, Derek Buitenhuis wrote:
> >> On 14/08/2020 20:13, Paul B Mahol wrote:
> >> > What specific insults in this thread?
> >>
> >> You wrote
On 8/16/20, Paul B Mahol wrote:
> Hi,
>
> patch attached.
>
> Please help porting this to linux and 64bit calling convention.
>
New patch attached.
This one does not allocate stack on x32.
0001-avcodec-cfhd-add-x86-SIMD.patch
Description: Binary data
___
On 8/16/20, Michael Niedermayer wrote:
> On Sun, Aug 16, 2020 at 08:32:20AM +0200, Paul B Mahol wrote:
>> On 8/16/20, Michael Niedermayer wrote:
>> > On Sat, Aug 15, 2020 at 07:29:40PM +0200, Paul B Mahol wrote:
>> >> On 8/15/20, Michael Niedermayer wrote:
>> >> > On Fri, Aug 14, 2020 at 02:24:2
On Sun, Aug 16, 2020 at 08:32:20AM +0200, Paul B Mahol wrote:
> On 8/16/20, Michael Niedermayer wrote:
> > On Sat, Aug 15, 2020 at 07:29:40PM +0200, Paul B Mahol wrote:
> >> On 8/15/20, Michael Niedermayer wrote:
> >> > On Fri, Aug 14, 2020 at 02:24:25PM +0200, Paul B Mahol wrote:
> >> >> On 8/14
On 8/16/2020 12:01 PM, Paul B Mahol wrote:
> On 8/16/20, James Almer wrote:
>> On 8/16/2020 11:09 AM, Paul B Mahol wrote:
>>> On 8/16/20, Paul B Mahol wrote:
Hi,
patch attached.
Please help porting this to linux and 64bit calling convention.
>>>
>>> New patch attache
On 8/16/20, James Almer wrote:
> On 8/16/2020 11:09 AM, Paul B Mahol wrote:
>> On 8/16/20, Paul B Mahol wrote:
>>> Hi,
>>>
>>> patch attached.
>>>
>>> Please help porting this to linux and 64bit calling convention.
>>>
>>
>> New patch attached, could build on x64, please report any build failure.
On 8/16/2020 11:09 AM, Paul B Mahol wrote:
> On 8/16/20, Paul B Mahol wrote:
>> Hi,
>>
>> patch attached.
>>
>> Please help porting this to linux and 64bit calling convention.
>>
>
> New patch attached, could build on x64, please report any build failure.
[...]
> diff --git a/libavcodec/x86/cfh
On 8/16/20, Paul B Mahol wrote:
> Hi,
>
> patch attached.
>
> Please help porting this to linux and 64bit calling convention.
>
New patch attached, could build on x64, please report any build failure.
0001-avcodec-cfhd-add-x86-SIMD.patch
Description: Binary data
Hi,
patch attached.
Please help porting this to linux and 64bit calling convention.
0001-avcodec-cfhd-add-x86-SIMD.patch
Description: Binary data
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
T
On 8/16/20, Michael Niedermayer wrote:
> On Sat, Aug 15, 2020 at 07:29:40PM +0200, Paul B Mahol wrote:
>> On 8/15/20, Michael Niedermayer wrote:
>> > On Fri, Aug 14, 2020 at 02:24:25PM +0200, Paul B Mahol wrote:
>> >> On 8/14/20, Paul B Mahol wrote:
>> >> > On 8/13/20, Paul B Mahol wrote:
>> >>
On Sat, Aug 15, 2020 at 07:29:40PM +0200, Paul B Mahol wrote:
> On 8/15/20, Michael Niedermayer wrote:
> > On Fri, Aug 14, 2020 at 02:24:25PM +0200, Paul B Mahol wrote:
> >> On 8/14/20, Paul B Mahol wrote:
> >> > On 8/13/20, Paul B Mahol wrote:
> >> >> Hi,
> >> >>
> >> >> patch attached.
> >> >>
On 8/15/20, Michael Niedermayer wrote:
> On Fri, Aug 14, 2020 at 02:24:25PM +0200, Paul B Mahol wrote:
>> On 8/14/20, Paul B Mahol wrote:
>> > On 8/13/20, Paul B Mahol wrote:
>> >> Hi,
>> >>
>> >> patch attached.
>> >>
>> >> Please review and/or benchmark, especially .asm file.
>> >>
>> >
>> > U
On Fri, Aug 14, 2020 at 02:24:25PM +0200, Paul B Mahol wrote:
> On 8/14/20, Paul B Mahol wrote:
> > On 8/13/20, Paul B Mahol wrote:
> >> Hi,
> >>
> >> patch attached.
> >>
> >> Please review and/or benchmark, especially .asm file.
> >>
> >
> > Updated patch attached.
> >
>
> Sorry, wrong patch.
On 8/15/20, Alexander Strasser wrote:
> Hi Derek,
> hi all!
>
> On 2020-08-14 20:22 +0100, Derek Buitenhuis wrote:
>> On 14/08/2020 20:13, Paul B Mahol wrote:
>> >> Resending because I accidentally replied to James instead of the list.
>> >> Woops.
>> >>
>> >> I guess it was not clear to me this i
Hi Derek,
hi all!
On 2020-08-14 20:22 +0100, Derek Buitenhuis wrote:
> On 14/08/2020 20:13, Paul B Mahol wrote:
> >> Resending because I accidentally replied to James instead of the list.
> >> Woops.
> >>
> >> I guess it was not clear to me this is not the initial thread, since it is
> >> not
> >>
On 14/08/2020 20:13, Paul B Mahol wrote:
>> Resending because I accidentally replied to James instead of the list.
>> Woops.
>>
>> I guess it was not clear to me this is not the initial thread, since it is
>> not
>> a v2 patch, and no other thread is titled this, or seems to include SIMD?
>> Perhap
On 8/14/20, Derek Buitenhuis wrote:
> On 14/08/2020 19:01, James Almer wrote:
>> On the very first email from this thread, he said "Please review and/or
>> benchmark, especially .asm file". He did not state his benchmarks were
>> irrelevant at first, but he did ask others for theirs.
>
> Resending
On 14/08/2020 19:01, James Almer wrote:
> On the very first email from this thread, he said "Please review and/or
> benchmark, especially .asm file". He did not state his benchmarks were
> irrelevant at first, but he did ask others for theirs.
Resending because I accidentally replied to James inst
On 8/14/2020 2:44 PM, Derek Buitenhuis wrote:
> On 14/08/2020 18:36, Paul B Mahol wrote:
>> You are not being helpful at all.
>
> It's called having basic standards for commit messages.
>
>> I clearly asked for testers to give their benchmarks as mine are
>> little relevant.
>
> You did not. Fee
On 14/08/2020 18:36, Paul B Mahol wrote:
> You are not being helpful at all.
It's called having basic standards for commit messages.
> I clearly asked for testers to give their benchmarks as mine are
> little relevant.
You did not. Feel free to point out where you did, because I cannnot
find it.
On 8/14/20, Derek Buitenhuis wrote:
> On 14/08/2020 16:37, Paul B Mahol wrote:
>>> I would expect any SIMD patch to include benchmarks showing it
>>> is actually faster.
>> It is faster, but I have only Celeron CPU to test.
>
> I can't tell if you're trolling or not, but that doesn't
> change the
On 8/14/20, Derek Buitenhuis wrote:
> On 14/08/2020 16:37, Paul B Mahol wrote:
>>> I would expect any SIMD patch to include benchmarks showing it
>>> is actually faster.
>> It is faster, but I have only Celeron CPU to test.
>
> I can't tell if you're trolling or not, but that doesn't
> change the
On 14/08/2020 16:37, Paul B Mahol wrote:
>> I would expect any SIMD patch to include benchmarks showing it
>> is actually faster.
> It is faster, but I have only Celeron CPU to test.
I can't tell if you're trolling or not, but that doesn't
change the standards...
- Derek
_
On 8/14/20, Derek Buitenhuis wrote:
>>> Updated patch attached.
>>>
>>
>> Sorry, wrong patch.
>
> I would expect any SIMD patch to include benchmarks showing it
> is actually faster.
It is faster, but I have only Celeron CPU to test.
>
> - Derek
> ___
Aug 13, 2020, 18:23 by one...@gmail.com:
> Hi,
>
> patch attached.
>
> Please review and/or benchmark, especially .asm file.
>
I took a look. Its just the horizontal pass of an inverse 2-6 idwt with
clipping.
The code is so simple I wasn't able to find any obvious ways to improve it,
except perh
>> Updated patch attached.
>>
>
> Sorry, wrong patch.
I would expect any SIMD patch to include benchmarks showing it
is actually faster.
- Derek
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To
On 8/13/20, Paul B Mahol wrote:
> Hi,
>
> patch attached.
>
> Please review and/or benchmark, especially .asm file.
>
Updated patch attached.
0001-avcodec-cfhd-add-x86-SIMD.patch
Description: Binary data
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmp
On 8/14/20, Paul B Mahol wrote:
> On 8/13/20, Paul B Mahol wrote:
>> Hi,
>>
>> patch attached.
>>
>> Please review and/or benchmark, especially .asm file.
>>
>
> Updated patch attached.
>
Sorry, wrong patch.
New patch attached.
0001-avcodec-cfhd-add-x86-SIMD.patch
Description: Binary data
___
Heya,
Curiosity: Since it seems to be SSSE3, does it work on both 32bits and 64bits
CPU?
Best,
On Thu, 13 Aug 2020, at 18:23, Paul B Mahol wrote:
> Hi,
>
> patch attached.
>
> Please review and/or benchmark, especially .asm file.
>
> ___
> ffmpeg-d
Hi,
patch attached.
Please review and/or benchmark, especially .asm file.
0001-avcodec-cfhd-add-x86-SIMD.patch
Description: Binary data
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscr
41 matches
Mail list logo
