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 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 ~10.42x/~5.23x (32-bit/64-bit) for horizontal ratios <= 2, ~9.49x/~4.64x for ratios within (2, 4], ~6.43x/~3.18x for ratios within (4, 8], and ~5.42x/~2.50x for ratios > 8 when not memory-bound on Haswell as compared with the current SSE2 implementation.
Showing the changes to codec/processing/src/downsample/downsample.cpp (1 of 5 files). Show all
Unified · Split
| @@ -109,6 +109,9 @@ void CDownsampling::InitDownsampleFuncs (SDownsampleFuncs& sDownsampleFunc, int · 108 unmodified lines | ||
| 109 | 109 | sDownsampleFunc.pfQuarterDownsampler = DyadicBilinearQuarterDownsampler_sse4; |
| 110 | 110 | sDownsampleFunc.pfGeneralRatioChroma = GeneralBilinearAccurateDownsamplerWrap_sse41; |
| 111 | 111 | } |
| 112 | + if (iCpuFlag & WELS_CPU_AVX2) { | |
| 113 | + sDownsampleFunc.pfGeneralRatioLuma = GeneralBilinearFastDownsamplerWrap_avx2; | |
| 114 | + } | |
| 112 | 115 | #endif//X86_ASM |
| 113 | 116 | |
| 114 | 117 | #if defined(HAVE_NEON) |