>
> the code moving should be in a seperate patch idealy, no big issue here as
> this
> is not a very complex patch
>
> otherwise patches should be ok
>
> also tested on linux x86-32/64, mingw32/64, linux qemu mips
> thanks
>
>
>
patch split (code move/asm)
and pushed
Thanks for review and
On Tue, Mar 20, 2018 at 09:06:53PM +0100, Martin Vignali wrote:
> Hello,
>
> New patchs in attach without AVX2 version
[...]
> rgb2rgb.c | 23 ---
> rgb2rgb.h |6 +++---
> rgb2rgb_template.c | 21 +
> x86/rgb2rgb.c |6
On 3/20/18, Martin Vignali wrote:
> Hello,
>
> New patchs in attach without AVX2 version
>
> Martin
>
probably ok. michealni?
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
Hello,
New patchs in attach without AVX2 version
Martin
0001-swscale-rgb-add-X86-SIMD-SSSE3-for.patch
Description: Binary data
0002-swscale-rgb-add-X86-SIMD-SSSE3-for.patch
Description: Binary data
0003-checkasm-swscale-add-test-for-rgb-shuffle_bytes-func.patch
Description: Binary data
2018-03-18 19:04 GMT+01:00, Paul B Mahol :
> On 3/18/18, Carl Eugen Hoyos wrote:
>> 2018-03-18 18:20 GMT+01:00, Paul B Mahol :
>>> On 3/18/18, Carl Eugen Hoyos wrote:
2018-03-18 17:46 GMT+01:00, Martin Vignali
On 3/18/18, Carl Eugen Hoyos wrote:
> 2018-03-18 18:20 GMT+01:00, Paul B Mahol :
>> On 3/18/18, Carl Eugen Hoyos wrote:
>>> 2018-03-18 17:46 GMT+01:00, Martin Vignali :
2018-03-18 17:37 GMT+01:00 Paul B
2018-03-18 18:20 GMT+01:00, Paul B Mahol :
> On 3/18/18, Carl Eugen Hoyos wrote:
>> 2018-03-18 17:46 GMT+01:00, Martin Vignali :
>>> 2018-03-18 17:37 GMT+01:00 Paul B Mahol :
>>>
On 3/18/18, Nicolas George
Paul B Mahol (2018-03-18):
> You are almost always wrong.
And you, just now, were rude.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
On 3/18/18, Carl Eugen Hoyos wrote:
> 2018-03-18 17:46 GMT+01:00, Martin Vignali :
>> 2018-03-18 17:37 GMT+01:00 Paul B Mahol :
>>
>>> On 3/18/18, Nicolas George wrote:
>>> > Martin Vignali (2018-03-18):
>>> >> I
2018-03-18 17:46 GMT+01:00, Martin Vignali :
> 2018-03-18 17:37 GMT+01:00 Paul B Mahol :
>
>> On 3/18/18, Nicolas George wrote:
>> > Martin Vignali (2018-03-18):
>> >> I run the test again with a bigger width (512 instead of 128)
>> >>
On 3/18/2018 1:28 PM, Nicolas George wrote:
> Martin Vignali (2018-03-18):
>> I run the test again with a bigger width (512 instead of 128)
>> This is my result :
>> shuffle_bytes_0321_c: 128.6
>> shuffle_bytes_0321_ssse3: 41.6
>> shuffle_bytes_0321_avx2: 23.4
>
> IIUC, these benchmarks are
2018-03-18 17:37 GMT+01:00 Paul B Mahol :
> On 3/18/18, Nicolas George wrote:
> > Martin Vignali (2018-03-18):
> >> I run the test again with a bigger width (512 instead of 128)
> >> This is my result :
> >> shuffle_bytes_0321_c: 128.6
> >>
On 3/18/18, Nicolas George wrote:
> Martin Vignali (2018-03-18):
>> I run the test again with a bigger width (512 instead of 128)
>> This is my result :
>> shuffle_bytes_0321_c: 128.6
>> shuffle_bytes_0321_ssse3: 41.6
>> shuffle_bytes_0321_avx2: 23.4
>
> IIUC, these benchmarks
Martin Vignali (2018-03-18):
> I run the test again with a bigger width (512 instead of 128)
> This is my result :
> shuffle_bytes_0321_c: 128.6
> shuffle_bytes_0321_ssse3: 41.6
> shuffle_bytes_0321_avx2: 23.4
IIUC, these benchmarks are expressed in CPU cycles. But what James says
is that it can
On 3/18/2018 1:23 PM, Martin Vignali wrote:
> 2018-03-18 16:49 GMT+01:00 James Almer :
>
>> On 3/18/2018 12:08 PM, Martin Vignali wrote:
>>> 2018-03-03 18:20 GMT+01:00 Martin Vignali :
>>>
Hello,
Patch in attach add SIMD for the 5
2018-03-18 16:49 GMT+01:00 James Almer :
> On 3/18/2018 12:08 PM, Martin Vignali wrote:
> > 2018-03-03 18:20 GMT+01:00 Martin Vignali :
> >
> >> Hello,
> >>
> >> Patch in attach add SIMD for the 5 shuffle_bytes func for rgb2rgb
> >> The new SIMD are
On 3/18/2018 12:08 PM, Martin Vignali wrote:
> 2018-03-03 18:20 GMT+01:00 Martin Vignali :
>
>> Hello,
>>
>> Patch in attach add SIMD for the 5 shuffle_bytes func for rgb2rgb
>> The new SIMD are write using external ASM.
>>
>> Also add checkasm test for theses func
>>
2018-03-18 16:08 GMT+01:00, Martin Vignali :
> shuffle_bytes_0321_c: 51.4
> shuffle_bytes_0321_ssse3: 18.7
> shuffle_bytes_0321_avx2: 12.7
> shuffle_bytes_1230_c: 126.9
> shuffle_bytes_1230_ssse3: 16.7
> shuffle_bytes_1230_avx2: 12.9
> shuffle_bytes_2103_c: 52.4
>
2018-03-03 18:20 GMT+01:00 Martin Vignali :
> Hello,
>
> Patch in attach add SIMD for the 5 shuffle_bytes func for rgb2rgb
> The new SIMD are write using external ASM.
>
> Also add checkasm test for theses func
> Restricted to x86_64, because the scalar part doesn't
Hello,
Patch in attach add SIMD for the 5 shuffle_bytes func for rgb2rgb
The new SIMD are write using external ASM.
Also add checkasm test for theses func
Restricted to x86_64, because the scalar part doesn't compile on x86_32
I consider for the scalar part that the src_size value is a multiple
20 matches
Mail list logo