python-pillow/Pillow
1
1
Fork 0
mirror of https://github.com/python-pillow/Pillow.git synced 2025-08-21 04:34:47 +03:00
Code Issues Packages Projects Releases Wiki Activity

SIMD AlphaComposite. sse4 implementation

Browse Source
This commit is contained in:
homm 2016-06-12 19:40:56 +03:00 committed by Alexander Karpinsky
parent 2dc15b8016
commit c7ce11f0f5
1 changed files with 74 additions and 21 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

95
src/libImaging/AlphaComposite.c
View File

@ -10,6 +10,14 @@
#include "Imaging.h" #include "Imaging.h"
#include <emmintrin.h>
#include <mmintrin.h>
#include <smmintrin.h>
#if defined(__AVX2__)
#include <immintrin.h>
#endif
#define PRECISION_BITS 7 #define PRECISION_BITS 7
typedef struct { typedef struct {
@ -23,6 +31,13 @@ Imaging
ImagingAlphaComposite(Imaging imDst, Imaging imSrc) { ImagingAlphaComposite(Imaging imDst, Imaging imSrc) {
Imaging imOut; Imaging imOut;
int x, y; int x, y;
int xsize = imDst->xsize;
__m128i mm_max_alpha = _mm_set1_epi32(255);
__m128i mm_max_alpha2 = _mm_set1_epi32(255 * 255);
__m128i mm_get_lo = _mm_set_epi8(
-1,-1, 5,4, 5,4, 5,4, -1,-1, 1,0, 1,0, 1,0);
__m128i mm_get_hi = _mm_set_epi8(
-1,-1, 13,12, 13,12, 13,12, -1,-1, 9,8, 9,8, 9,8);
/* Check arguments */ /* Check arguments */
if (!imDst || !imSrc || strcmp(imDst->mode, "RGBA") || if (!imDst || !imSrc || strcmp(imDst->mode, "RGBA") ||
@ -41,43 +56,81 @@ ImagingAlphaComposite(Imaging imDst, Imaging imSrc) {
return NULL; return NULL;
} }
#define MM_SHIFTDIV255_epi16(src)\
_mm_srli_epi16(_mm_add_epi16(src, _mm_srli_epi16(src, 8)), 8)
for (y = 0; y < imDst->ysize; y++) { for (y = 0; y < imDst->ysize; y++) {
rgba8 *dst = (rgba8 *)imDst->image[y]; rgba8 *dst = (rgba8 *)imDst->image[y];
rgba8 *src = (rgba8 *)imSrc->image[y]; rgba8 *src = (rgba8 *)imSrc->image[y];
rgba8 *out = (rgba8 *)imOut->image[y]; rgba8 *out = (rgba8 *)imOut->image[y];
for (x = 0; x < imDst->xsize; x++) { x = 0;
if (src->a == 0) {
for (; x < xsize - 4; x += 4) {
__m128i mm_dst = _mm_loadu_si128((__m128i *) &dst[x]);
__m128i mm_dst_lo = _mm_unpacklo_epi8(mm_dst, _mm_setzero_si128());
__m128i mm_dst_hi = _mm_unpackhi_epi8(mm_dst, _mm_setzero_si128());
__m128i mm_src = _mm_loadu_si128((__m128i *) &src[x]);
__m128i mm_src_lo = _mm_unpacklo_epi8(mm_src, _mm_setzero_si128());
__m128i mm_src_hi = _mm_unpackhi_epi8(mm_src, _mm_setzero_si128());
__m128i mm_dst_a = _mm_srli_epi32(mm_dst, 24);
__m128i mm_src_a = _mm_srli_epi32(mm_src, 24);
// Compute coefficients
// blend = dst->a * (255 - src->a); 16 bits
__m128i mm_blend = _mm_mullo_epi32(mm_dst_a, _mm_sub_epi32(mm_max_alpha, mm_src_a));
// outa = src->a * 255 + dst->a * (255 - src->a); 16 bits
__m128i mm_outa = _mm_add_epi32(_mm_mullo_epi32(mm_src_a, mm_max_alpha), mm_blend);
__m128i mm_coef1 = _mm_mullo_epi32(mm_src_a, mm_max_alpha2);
// 8 bits
mm_coef1 = _mm_cvtps_epi32(_mm_div_ps(_mm_cvtepi32_ps(mm_coef1), _mm_cvtepi32_ps(mm_outa)));
// 8 bits
__m128i mm_coef2 = _mm_sub_epi32(mm_max_alpha, mm_coef1);
__m128i mm_out_lo = _mm_add_epi16(
_mm_mullo_epi16(mm_src_lo, _mm_shuffle_epi8(mm_coef1, mm_get_lo)),
_mm_mullo_epi16(mm_dst_lo, _mm_shuffle_epi8(mm_coef2, mm_get_lo)));
mm_out_lo = _mm_or_si128(mm_out_lo, _mm_slli_epi64(
_mm_unpacklo_epi32(mm_outa, _mm_setzero_si128()), 48));
mm_out_lo = MM_SHIFTDIV255_epi16(mm_out_lo);
__m128i mm_out_hi = _mm_add_epi16(
_mm_mullo_epi16(mm_src_hi, _mm_shuffle_epi8(mm_coef1, mm_get_hi)),
_mm_mullo_epi16(mm_dst_hi, _mm_shuffle_epi8(mm_coef2, mm_get_hi)));
mm_out_hi = _mm_or_si128(mm_out_hi, _mm_slli_epi64(
_mm_unpackhi_epi32(mm_outa, _mm_setzero_si128()), 48));
mm_out_hi = MM_SHIFTDIV255_epi16(mm_out_hi);
_mm_storeu_si128((__m128i *) &out[x],
_mm_packus_epi16(mm_out_lo, mm_out_hi));
}
for (; x < xsize; x += 1) {
if (src[x].a == 0) {
// Copy 4 bytes at once. // Copy 4 bytes at once.
*out = *dst; out[x] = dst[x];
} else { } else {
// Integer implementation with increased precision. // Integer implementation with increased precision.
// Each variable has extra meaningful bits. // Each variable has extra meaningful bits.
// Divisions are rounded. // Divisions are rounded.
UINT32 tmpr, tmpg, tmpb; UINT32 tmpr, tmpg, tmpb;
UINT32 blend = dst->a * (255 - src->a); UINT32 blend = dst[x].a * (255 - src[x].a);
UINT32 outa255 = src->a * 255 + blend; UINT32 outa255 = src[x].a * 255 + blend;
// There we use 7 bits for precision. // There we use 7 bits for precision.
// We could use more, but we go beyond 32 bits. // We could use more, but we go beyond 32 bits.
UINT32 coef1 = src->a * 255 * 255 * (1 << PRECISION_BITS) / outa255; UINT32 coef1 = src[x].a * 255 * 255 * (1<<PRECISION_BITS) / outa255;
UINT32 coef2 = 255 * (1 << PRECISION_BITS) - coef1; UINT32 coef2 = 255 * (1<<PRECISION_BITS) - coef1;
tmpr = src->r * coef1 + dst->r * coef2; tmpr = src[x].r * coef1 + dst[x].r * coef2;
tmpg = src->g * coef1 + dst->g * coef2; tmpg = src[x].g * coef1 + dst[x].g * coef2;
tmpb = src->b * coef1 + dst->b * coef2; tmpb = src[x].b * coef1 + dst[x].b * coef2;
out->r = out[x].r = SHIFTFORDIV255(tmpr + (0x80<<PRECISION_BITS)) >> PRECISION_BITS;
SHIFTFORDIV255(tmpr + (0x80 << PRECISION_BITS)) >> PRECISION_BITS; out[x].g = SHIFTFORDIV255(tmpg + (0x80<<PRECISION_BITS)) >> PRECISION_BITS;
out->g = out[x].b = SHIFTFORDIV255(tmpb + (0x80<<PRECISION_BITS)) >> PRECISION_BITS;
SHIFTFORDIV255(tmpg + (0x80 << PRECISION_BITS)) >> PRECISION_BITS; out[x].a = SHIFTFORDIV255(outa255 + 0x80);
out->b =
SHIFTFORDIV255(tmpb + (0x80 << PRECISION_BITS)) >> PRECISION_BITS;
out->a = SHIFTFORDIV255(outa255 + 0x80);
} }
dst++;
src++;
out++;
} }
} }