thirdparty/ppsspp
2
0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Speed up color conversion using SSE.

Browse source 
Probably not the very most optimal implementation, but faster and SSE2.
This commit is contained in:
Unknown W. Brackets 2013-10-31 00:34:18 -07:00
parent 82761de992
commit f42cd11ddb
1 changed files with 69 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

73
GPU/GLES/TextureCache.cpp
View file

@ -527,11 +527,33 @@ void TextureCache::UpdateSamplingParams(TexCacheEntry &entry, bool force) {
static void ConvertColors(void *dstBuf, const void *srcBuf, GLuint dstFmt, int numPixels) {
const u32 *src = (const u32 *)srcBuf;
u32 *dst = (u32 *)dstBuf;
// TODO: All these can be further sped up with SSE or NEON.
// TODO: NEON.
switch (dstFmt) {
case GL_UNSIGNED_SHORT_4_4_4_4:
{
for (int i = 0; i < (numPixels + 1) / 2; i++) {
#ifdef _M_SSE
const __m128i maskA = _mm_set1_epi32(0x000F000F);
const __m128i maskB = _mm_set1_epi32(0x00F000F0);
const __m128i maskG = _mm_set1_epi32(0x0F000F00);
const __m128i maskR = _mm_set1_epi32(0xF000F000);
__m128i *srcp = (__m128i *)src;
__m128i *dstp = (__m128i *)dst;
const int sseChunks = numPixels / 8;
for (int i = 0; i < sseChunks; ++i) {
__m128i c = _mm_load_si128(&srcp[i]);
__m128i v = _mm_and_si128(_mm_srli_epi32(c, 12), maskA);
v = _mm_or_si128(v, _mm_and_si128(_mm_srli_epi32(c, 4), maskB));
v = _mm_or_si128(v, _mm_and_si128(_mm_slli_epi32(c, 4), maskG));
v = _mm_or_si128(v, _mm_and_si128(_mm_slli_epi32(c, 12), maskR));
_mm_store_si128(&dstp[i], v);
}
// The remainder is done in chunks of 2, SSE was chunks of 8.
int i = sseChunks * 8 / 2;
#else
int i = 0;
#endif
for (; i < (numPixels + 1) / 2; i++) {
u32 c = src[i];
dst[i] = ((c >> 12) & 0x000F000F) |
((c >> 4) & 0x00F000F0) |
@ -540,9 +562,32 @@ static void ConvertColors(void *dstBuf, const void *srcBuf, GLuint dstFmt, int n
}
}
break;
// Final Fantasy 2 uses this heavily in animated textures.
case GL_UNSIGNED_SHORT_5_5_5_1:
{
for (int i = 0; i < (numPixels + 1) / 2; i++) {
#ifdef _M_SSE
const __m128i maskA = _mm_set1_epi32(0x00010001);
const __m128i maskB = _mm_set1_epi32(0x003E003E);
const __m128i maskG = _mm_set1_epi32(0x07C007C0);
const __m128i maskR = _mm_set1_epi32(0xF800F800);
__m128i *srcp = (__m128i *)src;
__m128i *dstp = (__m128i *)dst;
const int sseChunks = numPixels / 8;
for (int i = 0; i < sseChunks; ++i) {
__m128i c = _mm_load_si128(&srcp[i]);
__m128i v = _mm_and_si128(_mm_srli_epi32(c, 15), maskA);
v = _mm_or_si128(v, _mm_and_si128(_mm_srli_epi32(c, 9), maskB));
v = _mm_or_si128(v, _mm_and_si128(_mm_slli_epi32(c, 1), maskG));
v = _mm_or_si128(v, _mm_and_si128(_mm_slli_epi32(c, 11), maskR));
_mm_store_si128(&dstp[i], v);
}
// The remainder is done in chunks of 2, SSE was chunks of 8.
int i = sseChunks * 8 / 2;
#else
int i = 0;
#endif
for (; i < (numPixels + 1) / 2; i++) {
u32 c = src[i];
dst[i] = ((c >> 15) & 0x00010001) |
((c >> 9) & 0x003E003E) |
@ -553,7 +598,27 @@ static void ConvertColors(void *dstBuf, const void *srcBuf, GLuint dstFmt, int n
break;
case GL_UNSIGNED_SHORT_5_6_5:
{
for (int i = 0; i < (numPixels + 1) / 2; i++) {
#ifdef _M_SSE
const __m128i maskB = _mm_set1_epi32(0x001F001F);
const __m128i maskG = _mm_set1_epi32(0x07E007E0);
const __m128i maskR = _mm_set1_epi32(0xF800F800);
__m128i *srcp = (__m128i *)src;
__m128i *dstp = (__m128i *)dst;
const int sseChunks = numPixels / 8;
for (int i = 0; i < sseChunks; ++i) {
__m128i c = _mm_load_si128(&srcp[i]);
__m128i v = _mm_and_si128(_mm_srli_epi32(c, 11), maskB);
v = _mm_or_si128(v, _mm_and_si128(c, maskG));
v = _mm_or_si128(v, _mm_and_si128(_mm_slli_epi32(c, 11), maskR));
_mm_store_si128(&dstp[i], v);
}
// The remainder is done in chunks of 2, SSE was chunks of 8.
int i = sseChunks * 8 / 2;
#else
int i = 0;
#endif
for (; i < (numPixels + 1) / 2; i++) {
u32 c = src[i];
dst[i] = ((c >> 11) & 0x001F001F) |
((c >> 0) & 0x07E007E0) |