dany/libwebp
1
0
Fork 0
mirror of https://github.com/webmproject/libwebp.git synced 2025-10-31 10:25:46 +01:00
Code Issues Actions Packages Projects Releases Wiki Activity

SSE2: slightly faster FTransformWHT

Browse Source 
goes from 0.3% to 0.1% overall CPU time, but...

Change-Id: I4c9a92b1e1d6b58ed57c6b890366f1dbeaf84f84
This commit is contained in:
Pascal Massimino
2015-06-30 17:08:21 +02:00
parent 36e9c4bc50
commit 8ef9a63b45
1 changed files with 28 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files

48
src/dsp/enc_sse2.c
View File

@@ -463,28 +463,36 @@ static void FTransform2(const uint8_t* src, const uint8_t* ref, int16_t* out) {
FTransformPass2(&v01h, &v32h, out + 16); FTransformPass2(&v01h, &v32h, out + 16);
} }
static void FTransformWHT(const int16_t* in, int16_t* out) { static void FTransformWHTRow(const int16_t* const in, __m128i* const out) {
int32_t tmp[16]; const __m128i kMult1 = _mm_set_epi16(0, 0, 0, 0, 1, 1, 1, 1);
int i; const __m128i kMult2 = _mm_set_epi16(0, 0, 0, 0, -1, 1, -1, 1);
for (i = 0; i < 4; ++i, in += 64) { const __m128i src0 = _mm_loadl_epi64((__m128i*)&in[0 * 16]);
const int a0 = (in[0 * 16] + in[2 * 16]); const __m128i src1 = _mm_loadl_epi64((__m128i*)&in[1 * 16]);
const int a1 = (in[1 * 16] + in[3 * 16]); const __m128i src2 = _mm_loadl_epi64((__m128i*)&in[2 * 16]);
const int a2 = (in[1 * 16] - in[3 * 16]); const __m128i src3 = _mm_loadl_epi64((__m128i*)&in[3 * 16]);
const int a3 = (in[0 * 16] - in[2 * 16]); const __m128i A01 = _mm_unpacklo_epi16(src0, src1); // A0 A1 | ...
tmp[0 + i * 4] = a0 + a1; const __m128i A23 = _mm_unpacklo_epi16(src2, src3); // A2 A3 | ...
tmp[1 + i * 4] = a3 + a2; const __m128i B0 = _mm_adds_epi16(A01, A23); // a0 | a1 | ...
tmp[2 + i * 4] = a3 - a2; const __m128i B1 = _mm_subs_epi16(A01, A23); // a3 | a2 | ...
tmp[3 + i * 4] = a0 - a1; const __m128i C0 = _mm_unpacklo_epi32(B0, B1); // a0 | a1 | a3 | a2
const __m128i C1 = _mm_unpacklo_epi32(B1, B0); // a3 | a2 | a0 | a1
const __m128i D0 = _mm_madd_epi16(C0, kMult1); // out0, out1
const __m128i D1 = _mm_madd_epi16(C1, kMult2); // out2, out3
*out = _mm_unpacklo_epi64(D0, D1);
} }
static void FTransformWHT(const int16_t* in, int16_t* out) {
__m128i row0, row1, row2, row3;
FTransformWHTRow(in + 0 * 64, &row0);
FTransformWHTRow(in + 1 * 64, &row1);
FTransformWHTRow(in + 2 * 64, &row2);
FTransformWHTRow(in + 3 * 64, &row3);
{ {
const __m128i src0 = _mm_loadu_si128((__m128i*)&tmp[0]); const __m128i a0 = _mm_add_epi32(row0, row2);
const __m128i src1 = _mm_loadu_si128((__m128i*)&tmp[4]); const __m128i a1 = _mm_add_epi32(row1, row3);
const __m128i src2 = _mm_loadu_si128((__m128i*)&tmp[8]); const __m128i a2 = _mm_sub_epi32(row1, row3);
const __m128i src3 = _mm_loadu_si128((__m128i*)&tmp[12]); const __m128i a3 = _mm_sub_epi32(row0, row2);
const __m128i a0 = _mm_add_epi32(src0, src2);
const __m128i a1 = _mm_add_epi32(src1, src3);
const __m128i a2 = _mm_sub_epi32(src1, src3);
const __m128i a3 = _mm_sub_epi32(src0, src2);
const __m128i b0 = _mm_srai_epi32(_mm_add_epi32(a0, a1), 1); const __m128i b0 = _mm_srai_epi32(_mm_add_epi32(a0, a1), 1);
const __m128i b1 = _mm_srai_epi32(_mm_add_epi32(a3, a2), 1); const __m128i b1 = _mm_srai_epi32(_mm_add_epi32(a3, a2), 1);
const __m128i b2 = _mm_srai_epi32(_mm_sub_epi32(a3, a2), 1); const __m128i b2 = _mm_srai_epi32(_mm_sub_epi32(a3, a2), 1);