GPU: Add some more optimizations hints in CopyLineExpand() for SSSE3.
This commit is contained in:
rogerman
parent
75766495f9
commit
25356b70f1
|
|
@ -279,6 +279,378 @@ static FORCEINLINE void CopyLineExpand(void *__restrict dst, const void *__restr
|
|||
}
|
||||
}
|
||||
#ifdef ENABLE_SSSE3
|
||||
else if (INTEGERSCALEHINT == 5)
|
||||
{
|
||||
__m128i srcPixOut[5];
|
||||
|
||||
for (size_t srcX = 0, dstX = 0; srcX < GPU_FRAMEBUFFER_NATIVE_WIDTH / (sizeof(__m128i) / ELEMENTSIZE); srcX++, dstX+=INTEGERSCALEHINT)
|
||||
{
|
||||
const __m128i srcVec = _mm_load_si128((__m128i *)src + srcX);
|
||||
|
||||
if (ELEMENTSIZE == 1)
|
||||
{
|
||||
srcPixOut[0] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 3));
|
||||
srcPixOut[1] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 3, 3, 3, 3, 4, 4, 4, 4, 4, 5, 5, 5, 5, 5, 6, 6));
|
||||
srcPixOut[2] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 6, 6, 6, 7, 7, 7, 7, 7, 8, 8, 8, 8, 8, 9, 9, 9));
|
||||
srcPixOut[3] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 9, 9,10,10,10,10,10,11,11,11,11,11,12,12,12,12));
|
||||
srcPixOut[4] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(12,13,13,13,13,13,14,14,14,14,14,15,15,15,15,15));
|
||||
}
|
||||
else if (ELEMENTSIZE == 2)
|
||||
{
|
||||
srcPixOut[0] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 2, 3, 2, 3, 2, 3));
|
||||
srcPixOut[1] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 2, 3, 2, 3, 4, 5, 4, 5, 4, 5, 4, 5, 4, 5, 6, 7));
|
||||
srcPixOut[2] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 6, 7, 6, 7, 6, 7, 6, 7, 8, 9, 8, 9, 8, 9, 8, 9));
|
||||
srcPixOut[3] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 8, 9,10,11,10,11,10,11,10,11,10,11,12,13,12,13));
|
||||
srcPixOut[4] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(12,13,12,13,12,13,14,15,14,15,14,15,14,15,14,15));
|
||||
}
|
||||
else if (ELEMENTSIZE == 4)
|
||||
{
|
||||
srcPixOut[0] = _mm_shuffle_epi32(srcVec, 0x00);
|
||||
srcPixOut[1] = _mm_shuffle_epi32(srcVec, 0x54);
|
||||
srcPixOut[2] = _mm_shuffle_epi32(srcVec, 0xA5);
|
||||
srcPixOut[3] = _mm_shuffle_epi32(srcVec, 0xEA);
|
||||
srcPixOut[4] = _mm_shuffle_epi32(srcVec, 0xFF);
|
||||
}
|
||||
|
||||
for (size_t lx = 0; lx < (size_t)INTEGERSCALEHINT; lx++)
|
||||
{
|
||||
_mm_store_si128((__m128i *)dst + dstX + lx, srcPixOut[lx]);
|
||||
}
|
||||
|
||||
if (SCALEVERTICAL)
|
||||
{
|
||||
for (size_t ly = 1; ly < (size_t)INTEGERSCALEHINT; ly++)
|
||||
{
|
||||
for (size_t lx = 0; lx < (size_t)INTEGERSCALEHINT; lx++)
|
||||
{
|
||||
_mm_store_si128((__m128i *)dst + dstX + ((GPU_FRAMEBUFFER_NATIVE_WIDTH / (sizeof(__m128i) / ELEMENTSIZE) * INTEGERSCALEHINT) * ly) + lx, srcPixOut[lx]);
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
else if (INTEGERSCALEHINT == 6)
|
||||
{
|
||||
__m128i srcPixOut[6];
|
||||
|
||||
for (size_t srcX = 0, dstX = 0; srcX < GPU_FRAMEBUFFER_NATIVE_WIDTH / (sizeof(__m128i) / ELEMENTSIZE); srcX++, dstX+=INTEGERSCALEHINT)
|
||||
{
|
||||
const __m128i srcVec = _mm_load_si128((__m128i *)src + srcX);
|
||||
|
||||
if (ELEMENTSIZE == 1)
|
||||
{
|
||||
srcPixOut[0] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2));
|
||||
srcPixOut[1] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 2, 2, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 5, 5));
|
||||
srcPixOut[2] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 5, 5, 5, 5, 6, 6, 6, 6, 6, 6, 7, 7, 7, 7, 7, 7));
|
||||
srcPixOut[3] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 8, 8, 8, 8, 8, 8, 9, 9, 9, 9, 9, 9,10,10,10,10));
|
||||
srcPixOut[4] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(10,10,11,11,11,11,11,11,12,12,12,12,12,12,13,13));
|
||||
srcPixOut[5] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(13,13,13,13,14,14,14,14,14,14,15,15,15,15,15,15));
|
||||
}
|
||||
else if (ELEMENTSIZE == 2)
|
||||
{
|
||||
srcPixOut[0] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 2, 3, 2, 3));
|
||||
srcPixOut[1] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 2, 3, 2, 3, 2, 3, 2, 3, 4, 5, 4, 5, 4, 5, 4, 5));
|
||||
srcPixOut[2] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 4, 5, 4, 5, 6, 7, 6, 7, 6, 7, 6, 7, 6, 7, 6, 7));
|
||||
srcPixOut[3] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 8, 9, 8, 9, 8, 9, 8, 9, 8, 9, 8, 9,10,11,10,11));
|
||||
srcPixOut[4] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(10,11,10,11,10,11,10,11,12,13,12,13,12,13,12,13));
|
||||
srcPixOut[5] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(12,13,12,13,14,15,14,15,14,15,14,15,14,15,14,15));
|
||||
}
|
||||
else if (ELEMENTSIZE == 4)
|
||||
{
|
||||
srcPixOut[0] = _mm_shuffle_epi32(srcVec, 0x00);
|
||||
srcPixOut[1] = _mm_shuffle_epi32(srcVec, 0x50);
|
||||
srcPixOut[2] = _mm_shuffle_epi32(srcVec, 0x55);
|
||||
srcPixOut[3] = _mm_shuffle_epi32(srcVec, 0xAA);
|
||||
srcPixOut[4] = _mm_shuffle_epi32(srcVec, 0xFA);
|
||||
srcPixOut[5] = _mm_shuffle_epi32(srcVec, 0xFF);
|
||||
}
|
||||
|
||||
for (size_t lx = 0; lx < (size_t)INTEGERSCALEHINT; lx++)
|
||||
{
|
||||
_mm_store_si128((__m128i *)dst + dstX + lx, srcPixOut[lx]);
|
||||
}
|
||||
|
||||
if (SCALEVERTICAL)
|
||||
{
|
||||
for (size_t ly = 1; ly < (size_t)INTEGERSCALEHINT; ly++)
|
||||
{
|
||||
for (size_t lx = 0; lx < (size_t)INTEGERSCALEHINT; lx++)
|
||||
{
|
||||
_mm_store_si128((__m128i *)dst + dstX + ((GPU_FRAMEBUFFER_NATIVE_WIDTH / (sizeof(__m128i) / ELEMENTSIZE) * INTEGERSCALEHINT) * ly) + lx, srcPixOut[lx]);
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
else if (INTEGERSCALEHINT == 7)
|
||||
{
|
||||
__m128i srcPixOut[7];
|
||||
|
||||
for (size_t srcX = 0, dstX = 0; srcX < GPU_FRAMEBUFFER_NATIVE_WIDTH / (sizeof(__m128i) / ELEMENTSIZE); srcX++, dstX+=INTEGERSCALEHINT)
|
||||
{
|
||||
const __m128i srcVec = _mm_load_si128((__m128i *)src + srcX);
|
||||
|
||||
if (ELEMENTSIZE == 1)
|
||||
{
|
||||
srcPixOut[0] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 2, 2));
|
||||
srcPixOut[1] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4));
|
||||
srcPixOut[2] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 4, 4, 4, 5, 5, 5, 5, 5, 5, 5, 6, 6, 6, 6, 6, 6));
|
||||
srcPixOut[3] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 6, 7, 7, 7, 7, 7, 7, 7, 8, 8, 8, 8, 8, 8, 8, 9));
|
||||
srcPixOut[4] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 9, 9, 9, 9, 9, 9,10,10,10,10,10,10,10,11,11,11));
|
||||
srcPixOut[5] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(11,11,11,11,12,12,12,12,12,12,12,13,13,13,13,13));
|
||||
srcPixOut[6] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(13,13,14,14,14,14,14,14,14,15,15,15,15,15,15,15));
|
||||
}
|
||||
else if (ELEMENTSIZE == 2)
|
||||
{
|
||||
srcPixOut[0] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 2, 3));
|
||||
srcPixOut[1] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 4, 5, 4, 5));
|
||||
srcPixOut[2] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 4, 5, 4, 5, 4, 5, 4, 5, 4, 5, 6, 7, 6, 7, 6, 7));
|
||||
srcPixOut[3] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 6, 7, 6, 7, 6, 7, 6, 7, 8, 9, 8, 9, 8, 9, 8, 9));
|
||||
srcPixOut[4] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 8, 9, 8, 9, 8, 9,10,11,10,11,10,11,10,11,10,11));
|
||||
srcPixOut[5] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(10,11,10,11,12,13,12,13,12,13,12,13,12,13,12,13));
|
||||
srcPixOut[6] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(12,13,14,15,14,15,14,15,14,15,14,15,14,15,14,15));
|
||||
}
|
||||
else if (ELEMENTSIZE == 4)
|
||||
{
|
||||
srcPixOut[0] = _mm_shuffle_epi32(srcVec, 0x00);
|
||||
srcPixOut[1] = _mm_shuffle_epi32(srcVec, 0x40);
|
||||
srcPixOut[2] = _mm_shuffle_epi32(srcVec, 0x55);
|
||||
srcPixOut[3] = _mm_shuffle_epi32(srcVec, 0xA5);
|
||||
srcPixOut[4] = _mm_shuffle_epi32(srcVec, 0xAA);
|
||||
srcPixOut[5] = _mm_shuffle_epi32(srcVec, 0xFE);
|
||||
srcPixOut[6] = _mm_shuffle_epi32(srcVec, 0xFF);
|
||||
}
|
||||
|
||||
for (size_t lx = 0; lx < (size_t)INTEGERSCALEHINT; lx++)
|
||||
{
|
||||
_mm_store_si128((__m128i *)dst + dstX + lx, srcPixOut[lx]);
|
||||
}
|
||||
|
||||
if (SCALEVERTICAL)
|
||||
{
|
||||
for (size_t ly = 1; ly < (size_t)INTEGERSCALEHINT; ly++)
|
||||
{
|
||||
for (size_t lx = 0; lx < (size_t)INTEGERSCALEHINT; lx++)
|
||||
{
|
||||
_mm_store_si128((__m128i *)dst + dstX + ((GPU_FRAMEBUFFER_NATIVE_WIDTH / (sizeof(__m128i) / ELEMENTSIZE) * INTEGERSCALEHINT) * ly) + lx, srcPixOut[lx]);
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
else if (INTEGERSCALEHINT == 8)
|
||||
{
|
||||
__m128i srcPixOut[8];
|
||||
|
||||
for (size_t srcX = 0, dstX = 0; srcX < GPU_FRAMEBUFFER_NATIVE_WIDTH / (sizeof(__m128i) / ELEMENTSIZE); srcX++, dstX+=INTEGERSCALEHINT)
|
||||
{
|
||||
const __m128i srcVec = _mm_load_si128((__m128i *)src + srcX);
|
||||
|
||||
if (ELEMENTSIZE == 1)
|
||||
{
|
||||
srcPixOut[0] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1));
|
||||
srcPixOut[1] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3));
|
||||
srcPixOut[2] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 4, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 5, 5, 5, 5));
|
||||
srcPixOut[3] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 6, 6, 6, 6, 6, 6, 6, 6, 7, 7, 7, 7, 7, 7, 7, 7));
|
||||
srcPixOut[4] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 8, 8, 8, 8, 8, 8, 8, 8, 9, 9, 9, 9, 9, 9, 9, 9));
|
||||
srcPixOut[5] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(10,10,10,10,10,10,10,10,11,11,11,11,11,11,11,11));
|
||||
srcPixOut[6] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(12,12,12,12,12,12,12,12,13,13,13,13,13,13,13,13));
|
||||
srcPixOut[7] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(14,14,14,14,14,14,14,14,15,15,15,15,15,15,15,15));
|
||||
}
|
||||
else if (ELEMENTSIZE == 2)
|
||||
{
|
||||
srcPixOut[0] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1));
|
||||
srcPixOut[1] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3));
|
||||
srcPixOut[2] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 4, 5, 4, 5, 4, 5, 4, 5, 4, 5, 4, 5, 4, 5, 4, 5));
|
||||
srcPixOut[3] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 6, 7, 6, 7, 6, 7, 6, 7, 6, 7, 6, 7, 6, 7, 6, 7));
|
||||
srcPixOut[4] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 8, 9, 8, 9, 8, 9, 8, 9, 8, 9, 8, 9, 8, 9, 8, 9));
|
||||
srcPixOut[5] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(10,11,10,11,10,11,10,11,10,11,10,11,10,11,10,11));
|
||||
srcPixOut[6] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(12,13,12,13,12,13,12,13,12,13,12,13,12,13,12,13));
|
||||
srcPixOut[7] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(14,15,14,15,14,15,14,15,14,15,14,15,14,15,14,15));
|
||||
}
|
||||
else if (ELEMENTSIZE == 4)
|
||||
{
|
||||
srcPixOut[0] = _mm_shuffle_epi32(srcVec, 0x00);
|
||||
srcPixOut[1] = _mm_shuffle_epi32(srcVec, 0x00);
|
||||
srcPixOut[2] = _mm_shuffle_epi32(srcVec, 0x55);
|
||||
srcPixOut[3] = _mm_shuffle_epi32(srcVec, 0x55);
|
||||
srcPixOut[4] = _mm_shuffle_epi32(srcVec, 0xAA);
|
||||
srcPixOut[5] = _mm_shuffle_epi32(srcVec, 0xAA);
|
||||
srcPixOut[6] = _mm_shuffle_epi32(srcVec, 0xFF);
|
||||
srcPixOut[7] = _mm_shuffle_epi32(srcVec, 0xFF);
|
||||
}
|
||||
|
||||
for (size_t lx = 0; lx < (size_t)INTEGERSCALEHINT; lx++)
|
||||
{
|
||||
_mm_store_si128((__m128i *)dst + dstX + lx, srcPixOut[lx]);
|
||||
}
|
||||
|
||||
if (SCALEVERTICAL)
|
||||
{
|
||||
for (size_t ly = 1; ly < (size_t)INTEGERSCALEHINT; ly++)
|
||||
{
|
||||
for (size_t lx = 0; lx < (size_t)INTEGERSCALEHINT; lx++)
|
||||
{
|
||||
_mm_store_si128((__m128i *)dst + dstX + ((GPU_FRAMEBUFFER_NATIVE_WIDTH / (sizeof(__m128i) / ELEMENTSIZE) * INTEGERSCALEHINT) * ly) + lx, srcPixOut[lx]);
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
else if (INTEGERSCALEHINT == 12)
|
||||
{
|
||||
__m128i srcPixOut[12];
|
||||
|
||||
for (size_t srcX = 0, dstX = 0; srcX < GPU_FRAMEBUFFER_NATIVE_WIDTH / (sizeof(__m128i) / ELEMENTSIZE); srcX++, dstX+=INTEGERSCALEHINT)
|
||||
{
|
||||
const __m128i srcVec = _mm_load_si128((__m128i *)src + srcX);
|
||||
|
||||
if (ELEMENTSIZE == 1)
|
||||
{
|
||||
srcPixOut[ 0] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1));
|
||||
srcPixOut[ 1] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2));
|
||||
srcPixOut[ 2] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3));
|
||||
srcPixOut[ 3] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5));
|
||||
srcPixOut[ 4] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 5, 5, 5, 5, 5, 5, 5, 5, 6, 6, 6, 6, 6, 6, 6, 6));
|
||||
srcPixOut[ 5] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 6, 6, 6, 6, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7));
|
||||
srcPixOut[ 6] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 9, 9, 9, 9));
|
||||
srcPixOut[ 7] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 9, 9, 9, 9, 9, 9, 9, 9,10,10,10,10,10,10,10,10));
|
||||
srcPixOut[ 8] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(10,10,10,10,11,11,11,11,11,11,11,11,11,11,11,11));
|
||||
srcPixOut[ 9] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(12,12,12,12,12,12,12,12,12,12,12,12,13,13,13,13));
|
||||
srcPixOut[10] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(13,13,13,13,13,13,13,13,14,14,14,14,14,14,14,14));
|
||||
srcPixOut[11] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(14,14,14,14,15,15,15,15,15,15,15,15,15,15,15,15));
|
||||
}
|
||||
else if (ELEMENTSIZE == 2)
|
||||
{
|
||||
srcPixOut[ 0] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1));
|
||||
srcPixOut[ 1] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 0, 1, 0, 1, 0, 1, 0, 1, 2, 3, 2, 3, 2, 3, 2, 3));
|
||||
srcPixOut[ 2] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3));
|
||||
srcPixOut[ 3] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 4, 5, 4, 5, 4, 5, 4, 5, 4, 5, 4, 5, 4, 5, 4, 5));
|
||||
srcPixOut[ 4] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 4, 5, 4, 5, 4, 5, 4, 5, 6, 7, 6, 7, 6, 7, 6, 7));
|
||||
srcPixOut[ 5] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 6, 7, 6, 7, 6, 7, 6, 7, 6, 7, 6, 7, 6, 7, 6, 7));
|
||||
srcPixOut[ 6] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 8, 9, 8, 9, 8, 9, 8, 9, 8, 9, 8, 9, 8, 9, 8, 9));
|
||||
srcPixOut[ 7] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 8, 9, 8, 9, 8, 9, 8, 9,10,11,10,11,10,11,10,11));
|
||||
srcPixOut[ 8] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(10,11,10,11,10,11,10,11,10,11,10,11,10,11,10,11));
|
||||
srcPixOut[ 9] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(12,13,12,13,12,13,12,13,12,13,12,13,12,13,12,13));
|
||||
srcPixOut[10] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(12,13,12,13,12,13,12,13,14,15,14,15,14,15,14,15));
|
||||
srcPixOut[11] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(14,15,14,15,14,15,14,15,14,15,14,15,14,15,14,15));
|
||||
}
|
||||
else if (ELEMENTSIZE == 4)
|
||||
{
|
||||
srcPixOut[ 0] = _mm_shuffle_epi32(srcVec, 0x00);
|
||||
srcPixOut[ 1] = _mm_shuffle_epi32(srcVec, 0x00);
|
||||
srcPixOut[ 2] = _mm_shuffle_epi32(srcVec, 0x00);
|
||||
srcPixOut[ 3] = _mm_shuffle_epi32(srcVec, 0x55);
|
||||
srcPixOut[ 4] = _mm_shuffle_epi32(srcVec, 0x55);
|
||||
srcPixOut[ 5] = _mm_shuffle_epi32(srcVec, 0x55);
|
||||
srcPixOut[ 6] = _mm_shuffle_epi32(srcVec, 0xAA);
|
||||
srcPixOut[ 7] = _mm_shuffle_epi32(srcVec, 0xAA);
|
||||
srcPixOut[ 8] = _mm_shuffle_epi32(srcVec, 0xAA);
|
||||
srcPixOut[ 9] = _mm_shuffle_epi32(srcVec, 0xFF);
|
||||
srcPixOut[10] = _mm_shuffle_epi32(srcVec, 0xFF);
|
||||
srcPixOut[11] = _mm_shuffle_epi32(srcVec, 0xFF);
|
||||
}
|
||||
|
||||
for (size_t lx = 0; lx < (size_t)INTEGERSCALEHINT; lx++)
|
||||
{
|
||||
_mm_store_si128((__m128i *)dst + dstX + lx, srcPixOut[lx]);
|
||||
}
|
||||
|
||||
if (SCALEVERTICAL)
|
||||
{
|
||||
for (size_t ly = 1; ly < (size_t)INTEGERSCALEHINT; ly++)
|
||||
{
|
||||
for (size_t lx = 0; lx < (size_t)INTEGERSCALEHINT; lx++)
|
||||
{
|
||||
_mm_store_si128((__m128i *)dst + dstX + ((GPU_FRAMEBUFFER_NATIVE_WIDTH / (sizeof(__m128i) / ELEMENTSIZE) * INTEGERSCALEHINT) * ly) + lx, srcPixOut[lx]);
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
else if (INTEGERSCALEHINT == 16)
|
||||
{
|
||||
__m128i srcPixOut[16];
|
||||
|
||||
for (size_t srcX = 0, dstX = 0; srcX < GPU_FRAMEBUFFER_NATIVE_WIDTH / (sizeof(__m128i) / ELEMENTSIZE); srcX++, dstX+=INTEGERSCALEHINT)
|
||||
{
|
||||
const __m128i srcVec = _mm_load_si128((__m128i *)src + srcX);
|
||||
|
||||
if (ELEMENTSIZE == 1)
|
||||
{
|
||||
srcPixOut[ 0] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0));
|
||||
srcPixOut[ 1] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1));
|
||||
srcPixOut[ 2] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2));
|
||||
srcPixOut[ 3] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3));
|
||||
srcPixOut[ 4] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4));
|
||||
srcPixOut[ 5] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5));
|
||||
srcPixOut[ 6] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6));
|
||||
srcPixOut[ 7] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7));
|
||||
srcPixOut[ 8] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8));
|
||||
srcPixOut[ 9] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9));
|
||||
srcPixOut[10] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10));
|
||||
srcPixOut[11] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(11,11,11,11,11,11,11,11,11,11,11,11,11,11,11,11));
|
||||
srcPixOut[12] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(12,12,12,12,12,12,12,12,12,12,12,12,12,12,12,12));
|
||||
srcPixOut[13] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(13,13,13,13,13,13,13,13,13,13,13,13,13,13,13,13));
|
||||
srcPixOut[14] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(14,14,14,14,14,14,14,14,14,14,14,14,14,14,14,14));
|
||||
srcPixOut[15] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15));
|
||||
}
|
||||
else if (ELEMENTSIZE == 2)
|
||||
{
|
||||
srcPixOut[ 0] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1));
|
||||
srcPixOut[ 1] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1));
|
||||
srcPixOut[ 2] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3));
|
||||
srcPixOut[ 3] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3, 2, 3));
|
||||
srcPixOut[ 4] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 4, 5, 4, 5, 4, 5, 4, 5, 4, 5, 4, 5, 4, 5, 4, 5));
|
||||
srcPixOut[ 5] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 4, 5, 4, 5, 4, 5, 4, 5, 4, 5, 4, 5, 4, 5, 4, 5));
|
||||
srcPixOut[ 6] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 6, 7, 6, 7, 6, 7, 6, 7, 6, 7, 6, 7, 6, 7, 6, 7));
|
||||
srcPixOut[ 7] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 6, 7, 6, 7, 6, 7, 6, 7, 6, 7, 6, 7, 6, 7, 6, 7));
|
||||
srcPixOut[ 8] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 8, 9, 8, 9, 8, 9, 8, 9, 8, 9, 8, 9, 8, 9, 8, 9));
|
||||
srcPixOut[ 9] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8( 8, 9, 8, 9, 8, 9, 8, 9, 8, 9, 8, 9, 8, 9, 8, 9));
|
||||
srcPixOut[10] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(10,11,10,11,10,11,10,11,10,11,10,11,10,11,10,11));
|
||||
srcPixOut[11] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(10,11,10,11,10,11,10,11,10,11,10,11,10,11,10,11));
|
||||
srcPixOut[12] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(12,13,12,13,12,13,12,13,12,13,12,13,12,13,12,13));
|
||||
srcPixOut[13] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(12,13,12,13,12,13,12,13,12,13,12,13,12,13,12,13));
|
||||
srcPixOut[14] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(14,15,14,15,14,15,14,15,14,15,14,15,14,15,14,15));
|
||||
srcPixOut[15] = _mm_shuffle_epi8(srcVec, _mm_setr_epi8(14,15,14,15,14,15,14,15,14,15,14,15,14,15,14,15));
|
||||
}
|
||||
else if (ELEMENTSIZE == 4)
|
||||
{
|
||||
srcPixOut[ 0] = _mm_shuffle_epi32(srcVec, 0x00);
|
||||
srcPixOut[ 1] = _mm_shuffle_epi32(srcVec, 0x00);
|
||||
srcPixOut[ 2] = _mm_shuffle_epi32(srcVec, 0x00);
|
||||
srcPixOut[ 3] = _mm_shuffle_epi32(srcVec, 0x00);
|
||||
srcPixOut[ 4] = _mm_shuffle_epi32(srcVec, 0x55);
|
||||
srcPixOut[ 5] = _mm_shuffle_epi32(srcVec, 0x55);
|
||||
srcPixOut[ 6] = _mm_shuffle_epi32(srcVec, 0x55);
|
||||
srcPixOut[ 7] = _mm_shuffle_epi32(srcVec, 0x55);
|
||||
srcPixOut[ 8] = _mm_shuffle_epi32(srcVec, 0xAA);
|
||||
srcPixOut[ 9] = _mm_shuffle_epi32(srcVec, 0xAA);
|
||||
srcPixOut[10] = _mm_shuffle_epi32(srcVec, 0xAA);
|
||||
srcPixOut[11] = _mm_shuffle_epi32(srcVec, 0xAA);
|
||||
srcPixOut[12] = _mm_shuffle_epi32(srcVec, 0xFF);
|
||||
srcPixOut[13] = _mm_shuffle_epi32(srcVec, 0xFF);
|
||||
srcPixOut[14] = _mm_shuffle_epi32(srcVec, 0xFF);
|
||||
srcPixOut[15] = _mm_shuffle_epi32(srcVec, 0xFF);
|
||||
}
|
||||
|
||||
for (size_t lx = 0; lx < (size_t)INTEGERSCALEHINT; lx++)
|
||||
{
|
||||
_mm_store_si128((__m128i *)dst + dstX + lx, srcPixOut[lx]);
|
||||
}
|
||||
|
||||
if (SCALEVERTICAL)
|
||||
{
|
||||
for (size_t ly = 1; ly < (size_t)INTEGERSCALEHINT; ly++)
|
||||
{
|
||||
for (size_t lx = 0; lx < (size_t)INTEGERSCALEHINT; lx++)
|
||||
{
|
||||
_mm_store_si128((__m128i *)dst + dstX + ((GPU_FRAMEBUFFER_NATIVE_WIDTH / (sizeof(__m128i) / ELEMENTSIZE) * INTEGERSCALEHINT) * ly) + lx, srcPixOut[lx]);
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
else if (INTEGERSCALEHINT > 1)
|
||||
{
|
||||
const size_t scale = dstWidth / GPU_FRAMEBUFFER_NATIVE_WIDTH;
|
||||
|
|
|
|||
Loading…
Reference in New Issue