ShuriZma
/
dolphin
mirror of https://github.com/dolphin-emu/dolphin.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Last commit only requires SSSE3, not SSE4. 

git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@6769 8ced0084-cf51-0410-be5f-012b33b47a6e
This commit is contained in:
xsacha 2011-01-07 13:54:18 +00:00
parent 53474403e2
commit a6acc99a89
1 changed files with 98 additions and 94 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
Source/Core/VideoCommon/Src/TextureDecoder.cpp
View File

@ -1068,8 +1068,10 @@ PC_TexFormat TexDecoder_Decode_RGBA(u32 * dst, const u8 * src, int width, int he
for (int y = 0; y < height; y += 4)
for (int x = 0; x < width; x += 8)
{
#if _M_SSE >= 0x401
// SSE4 intrinsics: About 5-10% faster than SSE2 version
#if _M_SSE >= 0x301
if (cpu_info.bSSSE3)
{
// SSSE3 intrinsics: About 5-10% faster than SSE2 version
for (int iy = 0; iy < 4; ++iy, src+=8)
{
const __m128i mask3210 = _mm_set_epi8(3, 3, 3, 3, 2, 2, 2, 2, 1, 1, 1, 1, 0, 0, 0, 0);
@ -1086,7 +1088,9 @@ PC_TexFormat TexDecoder_Decode_RGBA(u32 * dst, const u8 * src, int width, int he
_mm_storeu_si128(quaddst, rgba0);
_mm_storeu_si128(quaddst+1, rgba1);
}
#else
} else
#endif
{
// JSD optimized with SSE2 intrinsics.
// Produces an ~86% speed increase over reference C implementation.
// Each loop iteration processes 4 rows from 4 64-bit reads.
@ -1165,7 +1169,7 @@ PC_TexFormat TexDecoder_Decode_RGBA(u32 * dst, const u8 * src, int width, int he
_mm_storeu_si128(quaddst+1, rgba7);
src += 8;
#endif
}
}
#if 0
// Reference C implementation