History

[Processing/x86] Add an AVX2 implementation of GeneralBilinearAccurateDownsample

Keep track of relative pixel offsets and utilize pshufb to efficiently extract relevant pixels for horizontal scaling ratios <= 8. Because pshufb does not cross 128-bit lanes, the overhead of address calculations and loads is relatively greater as compared with an SSSE3/SSE4.1 implementation. Fall back to a generic approach for ratios > 8. The implementation assumes that data beyond the end of each line, before the next line begins, can be dirtied; which AFAICT is safe with the current usage of these routines. Speedup is ~8.52x/~6.89x (32-bit/64-bit) for horizontal ratios <= 2, ~7.81x/~6.13x for ratios within (2, 4], ~5.81x/~4.52x for ratios within (4, 8], and ~5.06x/~4.09x for ratios > 8 when not memory-bound on Haswell as compared with the current SSE2 implementation.

Sindre Aamås committed 10 years ago

Commit e4902151 parent b43e58a5 files · +616 −0Browse files

Showing the changes to codec/processing/src/downsample/downsample.h (1 of 5 files). Show all

Unified · Split

codec/processing/src/downsample/downsample.h+4−0
@@ -104,6 +104,7 @@ GeneralDownsampleFunc GeneralBilinearAccurateDownsamplerWrap_sse2; · 103 unmodified lines
104104 GeneralDownsampleFunc GeneralBilinearFastDownsamplerWrap_ssse3;
105105 GeneralDownsampleFunc GeneralBilinearAccurateDownsamplerWrap_sse41;
106106 GeneralDownsampleFunc GeneralBilinearFastDownsamplerWrap_avx2;
107+GeneralDownsampleFunc GeneralBilinearAccurateDownsamplerWrap_avx2;
107108
108109 SpecificDownsampleFunc DyadicBilinearOneThirdDownsampler_ssse3;
109110 SpecificDownsampleFunc DyadicBilinearOneThirdDownsampler_sse4;
@@ -126,6 +127,9 @@ void GeneralBilinearAccurateDownsampler_sse41 (uint8_t* pDst, int32_t iDstStride · 16 unmodified lines
126127 void GeneralBilinearFastDownsampler_avx2 (uint8_t* pDst, int32_t iDstStride, int32_t iDstWidth,
127128 int32_t iDstHeight, uint8_t* pSrc, int32_t iSrcStride, uint32_t uiScaleX,
128129 uint32_t uiScaleY);
130+void GeneralBilinearAccurateDownsampler_avx2 (uint8_t* pDst, int32_t iDstStride, int32_t iDstWidth,
131+ int32_t iDstHeight, uint8_t* pSrc, int32_t iSrcStride, uint32_t uiScaleX,
132+ uint32_t uiScaleY);
129133
130134 WELSVP_EXTERN_C_END
131135 #endif

Keyboard shortcuts

General

Show keyboard shortcuts
?
Focus the current search
/

Issues and pull requests

Open filters
f

Issues, pull requests and Inbox

Select next item
j↓
Select previous item
k↑
Open selected item
Enter
Dismiss filters or clear selection
Esc

Repository code

Go to file
t