On 06/02/15 1:02 PM, Christophe Gisquet wrote:
> Sure. Hopefully, the difference in number of gpr-passed args between
> UNIX64 and WIN64 can be handled.
I'll send a patch in a moment. Thanks.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://f
Hi,
2015-02-06 16:41 GMT+01:00 James Almer :
> Wouldn't it be better to just use the same code as UNIX64 instead? Now that
> we're going to load
> the address of the table to a reg, there's not point in having a whole
> separate init path for
> WIN64.
> One for X86_64 and one for X86_32 (Where a
On 06/02/15 9:49 AM, Christophe Gisquet wrote:
> diff --git a/libavcodec/x86/hevc_sao.asm b/libavcodec/x86/hevc_sao.asm
> index 5136121..8619716 100644
> --- a/libavcodec/x86/hevc_sao.asm
> +++ b/libavcodec/x86/hevc_sao.asm
> @@ -296,14 +296,16 @@ HEVC_SAO_BAND_FILTER_16 12, 64, 2
> %if WIN64
> c
On 06/02/15 7:07 AM, Hendrik Leppkes wrote:
> For some reason, the non-WIN64 version does just this, but the WIN64
> version does not.
> Any particular reason for this difference?
We support PIC on WIN64 but not with others. Or at least x86inc.asm enables it
that way.
According to msvc documenta
Hi,
2015-02-06 11:07 GMT+01:00 Hendrik Leppkes :
> I looked into the MSVC 64bit failure from this patch, and from what I
> can tell doing this doesn't work:
> movsx a_strideq, byte [pb_eo+eoq*4+1]
>
> I'm not entirely sure on the specifics why it breaks however..
> But all I could find sugges
On Thu, Feb 5, 2015 at 7:07 PM, James Almer wrote:
>
> On 05/02/15 12:49 PM, Christophe Gisquet wrote:
> > Hi,
> >
> > 2015-02-05 5:18 GMT+01:00 James Almer :
> >> Original x86 intrinsics code and initial yasm port by Pierre-Edouard
> >> Lepere.
> >> Refactoring and optimizations by James Almer.
On 05/02/15 12:49 PM, Christophe Gisquet wrote:
> Hi,
>
> 2015-02-05 5:18 GMT+01:00 James Almer :
>> Original x86 intrinsics code and initial yasm port by Pierre-Edouard Lepere.
>> Refactoring and optimizations by James Almer.
>
> No further comment from me.
Pushed, thanks.
Hi,
2015-02-05 5:18 GMT+01:00 James Almer :
> Original x86 intrinsics code and initial yasm port by Pierre-Edouard Lepere.
> Refactoring and optimizations by James Almer.
No further comment from me.
--
Christophe
___
ffmpeg-devel mailing list
ffmpeg-d
Original x86 intrinsics code and initial yasm port by Pierre-Edouard Lepere.
Refactoring and optimizations by James Almer.
Benchmarks of BQTerrace_1920x1080_60_qp22.bin with an Intel Core i5-4200U
Width 32
158583 decicycles in edge, sao_edge_filter_8 runs, 0 skips
5205 decicycles in ff_hevc_sao_e