Re: [FFmpeg-devel] avutil/x86util : add macro for 128 bits constant load

2017-12-02 Thread Martin Vignali 
2017-12-02 13:13 GMT+01:00 Henrik Gramner :

> On Fri, Dec 1, 2017 at 9:03 PM, Martin Vignali 
> wrote:
> > If no one have objections, i will push these patch tomorrow.
> >
> > Martin
>
> Follow James' suggestion to use >16 instead of ==32, otherwise OK.
>

Pushed, with mmsize > 16 in the macro patch

Thanks

Martin
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] avutil/x86util : add macro for 128 bits constant load

2017-12-02 Thread Henrik Gramner 
On Fri, Dec 1, 2017 at 9:03 PM, Martin Vignali  wrote:
> If no one have objections, i will push these patch tomorrow.
>
> Martin

Follow James' suggestion to use >16 instead of ==32, otherwise OK.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] avutil/x86util : add macro for 128 bits constant load

2017-12-01 Thread Martin Vignali 
2017-11-28 21:04 GMT+01:00 Henrik Gramner :

> On Mon, Nov 27, 2017 at 11:37 PM, James Almer  wrote:
> > On 11/27/2017 7:33 PM, James Darnley wrote:
> >> If the condition was made "mmsize > 16" would this work correctly for
> >> zmm registers?  (Assume I finally push my AVX-512 patches).
> >
> > No, there's no EVEX variant of vbroadcasti128. For that you would need
> > to use vbroadcasti32x4 or vbroadcasti64x2.
>
> x86inc handles that.
>

If no one have objections, i will push these patch tomorrow.

Martin
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] avutil/x86util : add macro for 128 bits constant load

2017-11-28 Thread Henrik Gramner 
On Mon, Nov 27, 2017 at 11:37 PM, James Almer  wrote:
> On 11/27/2017 7:33 PM, James Darnley wrote:
>> If the condition was made "mmsize > 16" would this work correctly for
>> zmm registers?  (Assume I finally push my AVX-512 patches).
>
> No, there's no EVEX variant of vbroadcasti128. For that you would need
> to use vbroadcasti32x4 or vbroadcasti64x2.

x86inc handles that.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] avutil/x86util : add macro for 128 bits constant load

2017-11-27 Thread James Almer 
On 11/27/2017 7:33 PM, James Darnley wrote:
> On 2017-11-27 20:19, Martin Vignali wrote:
>> +%macro VBROADCASTI128 2 ; dst xmm/ymm, src : 128bits val
>> +%if mmsize == 32
>> +vbroadcasti128 %1, %2
>> +%else
>> +mova   %1, %2
>> +%endif
>> +%endmacro
> 
> If the condition was made "mmsize > 16" would this work correctly for
> zmm registers?  (Assume I finally push my AVX-512 patches).

No, there's no EVEX variant of vbroadcasti128. For that you would need
to use vbroadcasti32x4 or vbroadcasti64x2.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] avutil/x86util : add macro for 128 bits constant load

2017-11-27 Thread James Darnley 
On 2017-11-27 20:19, Martin Vignali wrote:
> +%macro VBROADCASTI128 2 ; dst xmm/ymm, src : 128bits val
> +%if mmsize == 32
> +vbroadcasti128 %1, %2
> +%else
> +mova   %1, %2
> +%endif
> +%endmacro

If the condition was made "mmsize > 16" would this work correctly for
zmm registers?  (Assume I finally push my AVX-512 patches).

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel