AuroraMiddleware/skia2
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

remove unnecessary hsw opts

Browse Source 
These are copies of the ssse3 versions of these routines.
Instead of copying those routines, just use them.

Add inline to _portable methods to stifle warnings that
they are unused when compiling for HSW.

Change-Id: I655a844473bb3336dd2d1a80c1390259261b6243
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/294295
Reviewed-by: Shuangshuang Zhou <shuangshuang.zhou@intel.com>
Commit-Queue: Mike Klein <mtklein@google.com>
This commit is contained in:
Mike Klein 2020-06-04 11:15:10 -05:00 committed by Skia Commit-Bot
parent 7910a8f761
commit 88c6d72b42
2 changed files with 3 additions and 46 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
src/opts/SkOpts_hsw.cpp
View File

@ -28,8 +28,6 @@ namespace SkOpts {
RGBA_to_BGRA = SK_OPTS_NS::RGBA_to_BGRA;
RGBA_to_rgbA = SK_OPTS_NS::RGBA_to_rgbA;
RGBA_to_bgrA = SK_OPTS_NS::RGBA_to_bgrA;
RGB_to_RGB1 = SK_OPTS_NS::RGB_to_RGB1;
RGB_to_BGR1 = SK_OPTS_NS::RGB_to_BGR1;
gray_to_RGB1 = SK_OPTS_NS::gray_to_RGB1;
grayA_to_RGBA = SK_OPTS_NS::grayA_to_RGBA;
grayA_to_rgbA = SK_OPTS_NS::grayA_to_rgbA;

47
src/opts/SkSwizzler_opts.h
View File

@ -65,7 +65,7 @@ static void RGBA_to_BGRA_portable(uint32_t* dst, const uint32_t* src, int count)
}
}
static void RGB_to_RGB1_portable(uint32_t dst[], const uint8_t* src, int count) {
static inline void RGB_to_RGB1_portable(uint32_t dst[], const uint8_t* src, int count) {
for (int i = 0; i < count; i++) {
uint8_t r = src[0],
g = src[1],
@ -78,7 +78,7 @@ static void RGB_to_RGB1_portable(uint32_t dst[], const uint8_t* src, int count)
}
}
static void RGB_to_BGR1_portable(uint32_t dst[], const uint8_t* src, int count) {
static inline void RGB_to_BGR1_portable(uint32_t dst[], const uint8_t* src, int count) {
for (int i = 0; i < count; i++) {
uint8_t r = src[0],
g = src[1],
@ -580,48 +580,7 @@ static void premul_should_swapRB(bool kSwapRB, uint32_t* dst, const uint32_t* sr
RGBA_to_BGRA_portable(dst, src, count);
}
// Use SSSE3 impl as AVX2 impl regresses performance on some platforms.
static void insert_alpha_should_swaprb(bool kSwapRB,
uint32_t dst[], const uint8_t* src, int count) {
const __m128i alphaMask = _mm_set1_epi32(0xFF000000);
__m128i expand;
const uint8_t X = 0xFF; // Used a placeholder. The value of X is irrelevant.
if (kSwapRB) {
expand = _mm_setr_epi8(2,1,0,X, 5,4,3,X, 8,7,6,X, 11,10,9,X);
} else {
expand = _mm_setr_epi8(0,1,2,X, 3,4,5,X, 6,7,8,X, 9,10,11,X);
}
while (count >= 6) {
// Load a vector. While this actually contains 5 pixels plus an
// extra component, we will discard all but the first four pixels on
// this iteration.
__m128i rgb = _mm_loadu_si128((const __m128i*) src);
// Expand the first four pixels to RGBX and then mask to RGB(FF).
__m128i rgba = _mm_or_si128(_mm_shuffle_epi8(rgb, expand), alphaMask);
// Store 4 pixels.
_mm_storeu_si128((__m128i*) dst, rgba);
src += 4*3;
dst += 4;
count -= 4;
}
// Call portable code to finish up the tail of [0,4) pixels.
auto proc = kSwapRB ? RGB_to_BGR1_portable : RGB_to_RGB1_portable;
proc(dst, src, count);
}
/*not static*/ inline void RGB_to_RGB1(uint32_t dst[], const uint8_t* src, int count) {
insert_alpha_should_swaprb(false, dst, src, count);
}
/*not static*/ inline void RGB_to_BGR1(uint32_t dst[], const uint8_t* src, int count) {
insert_alpha_should_swaprb(true, dst, src, count);
}
// We saw no benefit from AVX2 over SSSE3 for RGB_to_RGB1 / RGB_to_BGR1.
/*not static*/ inline void gray_to_RGB1(uint32_t dst[], const uint8_t* src, int count) {
const __m256i alphas = _mm256_set1_epi8((uint8_t) 0xFF);