On hold until earlier patch has been fixed.
On Wed, Nov 13, 2013 at 12:46 PM, <muru...@multicorewareinc.com> wrote: > # HG changeset patch > # User Murugan Vairavel <muru...@multicorewareinc.com> > # Date 1384326967 -19800 > # Wed Nov 13 12:46:07 2013 +0530 > # Node ID 481cdfc251de0f99ef0a3c4fd53c786b79b5f182 > # Parent 69d4c1cfc8bed7c63bfdaa1073196e0874d14ebe > pixelsub_ps, Removed unused old code > > diff -r 69d4c1cfc8be -r 481cdfc251de source/common/pixel.cpp > --- a/source/common/pixel.cpp Wed Nov 13 12:31:12 2013 +0530 > +++ b/source/common/pixel.cpp Wed Nov 13 12:46:07 2013 +0530 > @@ -971,7 +971,6 @@ > p.weightpUniPixel = weightUnidirPix; > p.weightpUni = weightUnidir; > > - p.pixelsub_ps = pixelsub_ps_c; > p.pixeladd_ss = pixeladd_ss_c; > > p.scale1D_128to64 = scale1D_128to64; > diff -r 69d4c1cfc8be -r 481cdfc251de source/common/primitives.h > --- a/source/common/primitives.h Wed Nov 13 12:31:12 2013 +0530 > +++ b/source/common/primitives.h Wed Nov 13 12:46:07 2013 +0530 > @@ -162,7 +162,6 @@ > typedef void (*blockcpy_sp_t)(int bx, int by, int16_t *dst, intptr_t > dstride, pixel *src, intptr_t sstride); // dst is aligned > typedef void (*blockcpy_ps_t)(int bx, int by, pixel *dst, intptr_t > dstride, int16_t *src, intptr_t sstride); // dst is aligned > typedef void (*blockcpy_sc_t)(int bx, int by, int16_t *dst, intptr_t > dstride, uint8_t *src, intptr_t sstride); // dst is aligned > -typedef void (*pixelsub_ps_t)(int bx, int by, int16_t *dst, intptr_t > dstride, pixel *src0, pixel *src1, intptr_t sstride0, intptr_t sstride1); > typedef void (*pixeladd_ss_t)(int bx, int by, int16_t *dst, intptr_t > dstride, int16_t *src0, int16_t *src1, intptr_t sstride0, intptr_t > sstride1); > typedef void (*pixelavg_pp_t)(pixel *dst, intptr_t dstride, pixel *src0, > intptr_t sstride0, pixel *src1, intptr_t sstride1, int weight); > typedef void (*blockfill_s_t)(int16_t *dst, intptr_t dstride, int16_t > val); > @@ -277,7 +276,6 @@ > > weightpUni_t weightpUni; > weightpUniPixel_t weightpUniPixel; > - pixelsub_ps_t pixelsub_ps; > pixeladd_ss_t pixeladd_ss; > pixelavg_pp_t pixelavg_pp[NUM_LUMA_PARTITIONS]; > > diff -r 69d4c1cfc8be -r 481cdfc251de source/common/vec/blockcopy-sse3.cpp > --- a/source/common/vec/blockcopy-sse3.cpp Wed Nov 13 12:31:12 2013 > +0530 > +++ b/source/common/vec/blockcopy-sse3.cpp Wed Nov 13 12:46:07 2013 > +0530 > @@ -170,55 +170,6 @@ > } > } > > -void pixelsub_ps(int bx, int by, int16_t *dst, intptr_t dstride, uint8_t > *src0, uint8_t *src1, intptr_t sstride0, intptr_t sstride1) > -{ > - size_t aligncheck = (size_t)dst | (size_t)src0 | bx | sstride0 | > sstride1 | dstride; > - > - if (!(aligncheck & 15)) > - { > - // fast path, multiples of 16 pixel wide blocks > - for (int y = 0; y < by; y++) > - { > - for (int x = 0; x < bx; x += 16) > - { > - __m128i word0, word1; > - __m128i word3, word4; > - __m128i mask = _mm_setzero_si128(); > - > - word0 = _mm_load_si128((__m128i const*)(src0 + x)); // > load 16 bytes from src1 > - word1 = _mm_load_si128((__m128i const*)(src1 + x)); // > load 16 bytes from src2 > - > - word3 = _mm_unpacklo_epi8(word0, mask); // interleave > with zero extensions > - word4 = _mm_unpacklo_epi8(word1, mask); > - _mm_store_si128((__m128i*)&dst[x], _mm_subs_epi16(word3, > word4)); // store block into dst > - > - word3 = _mm_unpackhi_epi8(word0, mask); // interleave > with zero extensions > - word4 = _mm_unpackhi_epi8(word1, mask); > - _mm_store_si128((__m128i*)&dst[x + 8], > _mm_subs_epi16(word3, word4)); // store block into dst > - } > - > - src0 += sstride0; > - src1 += sstride1; > - dst += dstride; > - } > - } > - else > - { > - // slow path, irregular memory alignments or sizes > - for (int y = 0; y < by; y++) > - { > - for (int x = 0; x < bx; x++) > - { > - dst[x] = (int16_t)(src0[x] - src1[x]); > - } > - > - src0 += sstride0; > - src1 += sstride1; > - dst += dstride; > - } > - } > -} > - > void pixeladd_ss(int bx, int by, int16_t *dst, intptr_t dstride, int16_t > *src0, int16_t *src1, intptr_t sstride0, intptr_t sstride1) > { > size_t aligncheck = (size_t)dst | (size_t)src0 | sstride0 | sstride1 > | dstride; > @@ -315,7 +266,6 @@ > p.blockcpy_pp = blockcopy_pp; > p.blockcpy_ps = blockcopy_ps; > p.blockcpy_sp = blockcopy_sp; > - p.pixelsub_ps = pixelsub_ps; > p.pixeladd_ss = pixeladd_ss; > #endif // if HIGH_BIT_DEPTH > } > diff -r 69d4c1cfc8be -r 481cdfc251de source/test/pixelharness.cpp > --- a/source/test/pixelharness.cpp Wed Nov 13 12:31:12 2013 +0530 > +++ b/source/test/pixelharness.cpp Wed Nov 13 12:46:07 2013 +0530 > @@ -358,29 +358,6 @@ > return true; > } > > -bool PixelHarness::check_pixelsub_sp(pixelsub_ps_t ref, pixelsub_ps_t opt) > -{ > - ALIGN_VAR_16(int16_t, ref_dest[64 * 64]); > - ALIGN_VAR_16(int16_t, opt_dest[64 * 64]); > - int bx = 64; > - int by = 64; > - int j = 0; > - for (int i = 0; i < ITERS; i++) > - { > - opt(bx, by, opt_dest, 64, pbuf2 + j, pbuf1 + j, STRIDE, STRIDE); > - ref(bx, by, ref_dest, 64, pbuf2 + j, pbuf1 + j, STRIDE, STRIDE); > - > - if (memcmp(ref_dest, opt_dest, 64 * 64 * sizeof(int16_t))) > - return false; > - > - j += INCR; > - bx = 4 * ((rand() & 15) + 1); > - by = 4 * ((rand() & 15) + 1); > - } > - > - return true; > -} > - > bool PixelHarness::check_pixeladd_ss(pixeladd_ss_t ref, pixeladd_ss_t opt) > { > ALIGN_VAR_16(int16_t, ref_dest[64 * 64]); > @@ -886,15 +863,6 @@ > } > } > > - if (opt.pixelsub_ps) > - { > - if (!check_pixelsub_sp(ref.pixelsub_ps, opt.pixelsub_ps)) > - { > - printf("Luma Substract failed!\n"); > - return false; > - } > - } > - > if (opt.pixeladd_ss) > { > if (!check_pixeladd_ss(ref.pixeladd_ss, opt.pixeladd_ss)) > @@ -1114,12 +1082,6 @@ > REPORT_SPEEDUP(opt.weightpUni, ref.weightpUni, (int16_t*)sbuf1, > pbuf1, 64, 64, 32, 32, 128, 1 << 9, 10, 100); > } > > - if (opt.pixelsub_ps) > - { > - printf("Pixel Sub"); > - REPORT_SPEEDUP(opt.pixelsub_ps, ref.pixelsub_ps, 64, 64, > (int16_t*)pbuf1, FENC_STRIDE, pbuf2, pbuf1, STRIDE, STRIDE); > - } > - > if (opt.pixeladd_ss) > { > printf("pixel_ss add"); > diff -r 69d4c1cfc8be -r 481cdfc251de source/test/pixelharness.h > --- a/source/test/pixelharness.h Wed Nov 13 12:31:12 2013 +0530 > +++ b/source/test/pixelharness.h Wed Nov 13 12:46:07 2013 +0530 > @@ -49,7 +49,6 @@ > bool check_calcrecon(calcrecon_t ref, calcrecon_t opt); > bool check_weightpUni(weightpUniPixel_t ref, weightpUniPixel_t opt); > bool check_weightpUni(weightpUni_t ref, weightpUni_t opt); > - bool check_pixelsub_sp(pixelsub_ps_t ref, pixelsub_ps_t opt); > bool check_pixeladd_ss(pixeladd_ss_t ref, pixeladd_ss_t opt); > bool check_downscale_t(downscale_t ref, downscale_t opt); > bool check_cvt32to16_shr_t(cvt32to16_shr_t ref, cvt32to16_shr_t opt); > _______________________________________________ > x265-devel mailing list > x265-devel@videolan.org > https://mailman.videolan.org/listinfo/x265-devel >
_______________________________________________ x265-devel mailing list x265-devel@videolan.org https://mailman.videolan.org/listinfo/x265-devel