[x265] [PATCH UPDATE] asm: cvt32to16_shr_sse2

[Steve Borho]

On Thu, Oct 17, 2013 at 10:53 AM, Min Chen <chenm003 at 163.com> wrote:

> # HG changeset patch
> # User Min Chen <chenm003 at 163.com>
> # Date 1382025002 -28800
> # Node ID ecd0378bf284026478454495b32afed1fddffd26
> # Parent  03370e851f3b754d4a8531e8ade92fca43de5cc4
> asm: cvt32to16_shr_sse2
>

awesome... except this patch is missing pixel-util.asm.  If you send it to
me via e-mail or Skype I can fold it into this patch.

Also; patches which add assembly functions should remove the intrinsic
function.


> diff -r 03370e851f3b -r ecd0378bf284 source/common/CMakeLists.txt
> --- a/source/common/CMakeLists.txt      Thu Oct 17 23:49:38 2013 +0800
> +++ b/source/common/CMakeLists.txt      Thu Oct 17 23:50:02 2013 +0800
> @@ -153,7 +153,7 @@
>          add_definitions(-DHAVE_ALIGNED_STACK=0)
>      endif()
>
> -    set(ASMS pixel-a.asm const-a.asm cpu-a.asm sad-a.asm mc-a.asm
> mc-a2.asm ipfilter8.asm)
> +    set(ASMS pixel-a.asm const-a.asm cpu-a.asm sad-a.asm mc-a.asm
> mc-a2.asm ipfilter8.asm pixel-util.asm)
>      if (X64)
>          add_definitions(-DARCH_X86_64=1)
>      else()
> diff -r 03370e851f3b -r ecd0378bf284 source/common/x86/asm-primitives.cpp
> --- a/source/common/x86/asm-primitives.cpp      Thu Oct 17 23:49:38 2013
> +0800
> +++ b/source/common/x86/asm-primitives.cpp      Thu Oct 17 23:50:02 2013
> +0800
> @@ -312,6 +312,8 @@
>          p.sa8d[BLOCK_8x8]   = x265_pixel_sa8d_8x8_sse2;
>          p.sa8d[BLOCK_16x16] = x265_pixel_sa8d_16x16_sse2;
>          SA8D_INTER_FROM_BLOCK(sse2);
> +
> +        p.cvt32to16_shr = x265_cvt32to16_shr_sse2;
>      }
>      if (cpuMask & X265_CPU_SSSE3)
>      {
> diff -r 03370e851f3b -r ecd0378bf284 source/common/x86/pixel.h
> --- a/source/common/x86/pixel.h Thu Oct 17 23:49:38 2013 +0800
> +++ b/source/common/x86/pixel.h Thu Oct 17 23:50:02 2013 +0800
> @@ -201,6 +201,8 @@
>  uint64_t x265_pixel_sa8d_satd_16x16_xop(pixel *pix1, intptr_t stride1,
> pixel *pix2, intptr_t stride2);
>  uint64_t x265_pixel_sa8d_satd_16x16_avx2(pixel *pix1, intptr_t stride1,
> pixel *pix2, intptr_t stride2);
>
> +void x265_cvt32to16_shr_sse2(short *dst, int *src, intptr_t, int, int);
> +
>  #define DECL_SSD(width, suffix) \
>      int x265_pixel_ssd_ ## width ## x64_ ## suffix(pixel *, intptr_t,
> pixel *, intptr_t); \
>      int x265_pixel_ssd_ ## width ## x48_ ## suffix(pixel *, intptr_t,
> pixel *, intptr_t); \
>
> _______________________________________________
> x265-devel mailing list
> x265-devel at videolan.org
> https://mailman.videolan.org/listinfo/x265-devel
>



-- 
Steve Borho
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20131017/c16233e7/attachment.html>