python-pillow/Pillow
1
1
Fork 0
mirror of https://github.com/python-pillow/Pillow.git synced 2025-08-20 20:24:45 +03:00
Code Issues Packages Projects Releases Wiki Activity

SIMD Filter. better loading

Browse Source
This commit is contained in:
Alexander 2017-09-13 02:26:45 +03:00 committed by Alexander Karpinsky
parent 0e17b8313b
commit 74d95f5e9d
1 changed files with 12 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
libImaging/FilterSIMD_3x3i_4u8.c
View File

@ -48,35 +48,33 @@ ImagingFilter3x3i_4u8(Imaging imOut, Imaging im, const INT16* kernel,
__m128i kernel21 = _mm_set_epi16( __m128i kernel21 = _mm_set_epi16(
kernel[8], kernel[7], kernel[8], kernel[7], kernel[8], kernel[7], kernel[8], kernel[7],
kernel[6], 0, kernel[6], 0); kernel[6], 0, kernel[6], 0);
// __m128i source;
__m128i pix00, pix10, pix20; __m128i pix00, pix10, pix20;
// __m128i pix01, pix11, pix21;
out[0] = in0[0]; out[0] = in0[0];
x = 1; x = 1;
MM_KERNEL_LOAD(0, 0);
for (; x < im->xsize-1-3; x += 4) { for (; x < im->xsize-1-3; x += 4) {
__m128i ss0 = _mm_set1_epi32(offset); __m128i ss0 = _mm_set1_epi32(offset);
__m128i ss1 = _mm_set1_epi32(offset); __m128i ss1 = _mm_set1_epi32(offset);
__m128i ss2 = _mm_set1_epi32(offset); __m128i ss2 = _mm_set1_epi32(offset);
__m128i ss3 = _mm_set1_epi32(offset); __m128i ss3 = _mm_set1_epi32(offset);
MM_KERNEL_LOAD(0, x-1);
MM_KERNEL_SUM(ss0, 0, 0, _mm_unpacklo_epi8, _mm_unpacklo_epi32); MM_KERNEL_SUM(ss0, 0, 0, _mm_unpacklo_epi8, _mm_unpacklo_epi32);
MM_KERNEL_SUM(ss0, 0, 0, _mm_unpackhi_epi8, _mm_unpackhi_epi32); MM_KERNEL_SUM(ss0, 0, 0, _mm_unpackhi_epi8, _mm_unpackhi_epi32);
ss0 = _mm_srai_epi32(ss0, PRECISION_BITS);
MM_KERNEL_SUM(ss1, 0, 1, _mm_unpacklo_epi8, _mm_unpacklo_epi32); MM_KERNEL_SUM(ss1, 0, 1, _mm_unpacklo_epi8, _mm_unpacklo_epi32);
MM_KERNEL_SUM(ss1, 0, 1, _mm_unpackhi_epi8, _mm_unpackhi_epi32); MM_KERNEL_SUM(ss1, 0, 1, _mm_unpackhi_epi8, _mm_unpackhi_epi32);
ss1 = _mm_srai_epi32(ss1, PRECISION_BITS); ss0 = _mm_packs_epi32(
ss0 = _mm_packs_epi32(ss0, ss1); _mm_srai_epi32(ss0, PRECISION_BITS),
_mm_srai_epi32(ss1, PRECISION_BITS));
MM_KERNEL_LOAD(0, x+1); MM_KERNEL_SUM(ss2, 0, 0, _mm_unpackhi_epi8, _mm_unpacklo_epi32);
MM_KERNEL_SUM(ss2, 0, 0, _mm_unpacklo_epi8, _mm_unpacklo_epi32); MM_KERNEL_SUM(ss3, 0, 1, _mm_unpackhi_epi8, _mm_unpacklo_epi32);
MM_KERNEL_SUM(ss2, 0, 0, _mm_unpackhi_epi8, _mm_unpackhi_epi32); MM_KERNEL_LOAD(0, x+3);
ss2 = _mm_srai_epi32(ss2, PRECISION_BITS); MM_KERNEL_SUM(ss2, 0, 0, _mm_unpacklo_epi8, _mm_unpackhi_epi32);
MM_KERNEL_SUM(ss3, 0, 1, _mm_unpacklo_epi8, _mm_unpacklo_epi32); MM_KERNEL_SUM(ss3, 0, 1, _mm_unpacklo_epi8, _mm_unpackhi_epi32);
MM_KERNEL_SUM(ss3, 0, 1, _mm_unpackhi_epi8, _mm_unpackhi_epi32); ss2 = _mm_packs_epi32(
ss3 = _mm_srai_epi32(ss3, PRECISION_BITS); _mm_srai_epi32(ss2, PRECISION_BITS),
ss2 = _mm_packs_epi32(ss2, ss3); _mm_srai_epi32(ss3, PRECISION_BITS));
ss0 = _mm_packus_epi16(ss0, ss2); ss0 = _mm_packus_epi16(ss0, ss2);
_mm_storeu_si128((__m128i*) &out[x], ss0); _mm_storeu_si128((__m128i*) &out[x], ss0);