ref: 4418b790a79342a0b89cac0c83e07bc4d8729262
parent: 82fe8c9f367ebbe72c5bd8199c1971649f952afa
author: Christian Duvivier <cduvivier@google.com>
date: Fri Mar 15 11:50:55 EDT 2013
Faster vp9_short_fdct16x16. Scalar path is about 1.5x faster (3.1% overall encoder speedup). SSE2 path is about 7.2x faster (7.8% overall encoder speedup). Change-Id: I06da5ad0cdae2488431eabf002b0d898d66d8289
--- a/vp9/common/vp9_rtcd_defs.sh
+++ b/vp9/common/vp9_rtcd_defs.sh
@@ -598,7 +598,7 @@
specialize vp9_short_fdct32x32
prototype void vp9_short_fdct16x16 "int16_t *InputData, int16_t *OutputData, int pitch"
-specialize vp9_short_fdct16x16
+specialize vp9_short_fdct16x16 sse2
prototype void vp9_short_walsh4x4_x8 "int16_t *InputData, int16_t *OutputData, int pitch"
specialize vp9_short_walsh4x4_x8
--- a/vp9/encoder/vp9_dct.c
+++ b/vp9/encoder/vp9_dct.c
@@ -269,6 +269,185 @@
}
}
+void vp9_short_fdct16x16_c(int16_t *input, int16_t *output, int pitch) {+ // The 2D transform is done with two passes which are actually pretty
+ // similar. In the first one, we transform the columns and transpose
+ // the results. In the second one, we transform the rows. To achieve that,
+ // as the first pass results are transposed, we tranpose the columns (that
+ // is the transposed rows) and transpose the results (so that it goes back
+ // in normal/row positions).
+ const int stride = pitch >> 1;
+ int pass;
+ // We need an intermediate buffer between passes.
+ int16_t intermediate[256];
+ int16_t *in = input;
+ int16_t *out = intermediate;
+ // Do the two transform/transpose passes
+ for (pass = 0; pass < 2; ++pass) {+ /*canbe16*/ int step1[8];
+ /*canbe16*/ int step2[8];
+ /*canbe16*/ int step3[8];
+ /*canbe16*/ int input[8];
+ /*needs32*/ int temp1, temp2;
+ int i;
+ for (i = 0; i < 16; i++) {+ if (0 == pass) {+ // Calculate input for the first 8 results.
+ input[0] = (in[0 * stride] + in[15 * stride]) << 2;
+ input[1] = (in[1 * stride] + in[14 * stride]) << 2;
+ input[2] = (in[2 * stride] + in[13 * stride]) << 2;
+ input[3] = (in[3 * stride] + in[12 * stride]) << 2;
+ input[4] = (in[4 * stride] + in[11 * stride]) << 2;
+ input[5] = (in[5 * stride] + in[10 * stride]) << 2;
+ input[6] = (in[6 * stride] + in[ 9 * stride]) << 2;
+ input[7] = (in[7 * stride] + in[ 8 * stride]) << 2;
+ // Calculate input for the next 8 results.
+ step1[0] = (in[7 * stride] - in[ 8 * stride]) << 2;
+ step1[1] = (in[6 * stride] - in[ 9 * stride]) << 2;
+ step1[2] = (in[5 * stride] - in[10 * stride]) << 2;
+ step1[3] = (in[4 * stride] - in[11 * stride]) << 2;
+ step1[4] = (in[3 * stride] - in[12 * stride]) << 2;
+ step1[5] = (in[2 * stride] - in[13 * stride]) << 2;
+ step1[6] = (in[1 * stride] - in[14 * stride]) << 2;
+ step1[7] = (in[0 * stride] - in[15 * stride]) << 2;
+ } else {+ // Calculate input for the first 8 results.
+ input[0] = ((in[0 * 16] + 1) >> 2) + ((in[15 * 16] + 1) >> 2);
+ input[1] = ((in[1 * 16] + 1) >> 2) + ((in[14 * 16] + 1) >> 2);
+ input[2] = ((in[2 * 16] + 1) >> 2) + ((in[13 * 16] + 1) >> 2);
+ input[3] = ((in[3 * 16] + 1) >> 2) + ((in[12 * 16] + 1) >> 2);
+ input[4] = ((in[4 * 16] + 1) >> 2) + ((in[11 * 16] + 1) >> 2);
+ input[5] = ((in[5 * 16] + 1) >> 2) + ((in[10 * 16] + 1) >> 2);
+ input[6] = ((in[6 * 16] + 1) >> 2) + ((in[ 9 * 16] + 1) >> 2);
+ input[7] = ((in[7 * 16] + 1) >> 2) + ((in[ 8 * 16] + 1) >> 2);
+ // Calculate input for the next 8 results.
+ step1[0] = ((in[7 * 16] + 1) >> 2) - ((in[ 8 * 16] + 1) >> 2);
+ step1[1] = ((in[6 * 16] + 1) >> 2) - ((in[ 9 * 16] + 1) >> 2);
+ step1[2] = ((in[5 * 16] + 1) >> 2) - ((in[10 * 16] + 1) >> 2);
+ step1[3] = ((in[4 * 16] + 1) >> 2) - ((in[11 * 16] + 1) >> 2);
+ step1[4] = ((in[3 * 16] + 1) >> 2) - ((in[12 * 16] + 1) >> 2);
+ step1[5] = ((in[2 * 16] + 1) >> 2) - ((in[13 * 16] + 1) >> 2);
+ step1[6] = ((in[1 * 16] + 1) >> 2) - ((in[14 * 16] + 1) >> 2);
+ step1[7] = ((in[0 * 16] + 1) >> 2) - ((in[15 * 16] + 1) >> 2);
+ }
+ // Work on the first eight values; fdct8_1d(input, even_results);
+ {+ /*canbe16*/ int s0, s1, s2, s3, s4, s5, s6, s7;
+ /*needs32*/ int t0, t1, t2, t3;
+ /*canbe16*/ int x0, x1, x2, x3;
+
+ // stage 1
+ s0 = input[0] + input[7];
+ s1 = input[1] + input[6];
+ s2 = input[2] + input[5];
+ s3 = input[3] + input[4];
+ s4 = input[3] - input[4];
+ s5 = input[2] - input[5];
+ s6 = input[1] - input[6];
+ s7 = input[0] - input[7];
+
+ // fdct4_1d(step, step);
+ x0 = s0 + s3;
+ x1 = s1 + s2;
+ x2 = s1 - s2;
+ x3 = s0 - s3;
+ t0 = (x0 + x1) * cospi_16_64;
+ t1 = (x0 - x1) * cospi_16_64;
+ t2 = x3 * cospi_8_64 + x2 * cospi_24_64;
+ t3 = x3 * cospi_24_64 - x2 * cospi_8_64;
+ out[0] = dct_const_round_shift(t0);
+ out[4] = dct_const_round_shift(t2);
+ out[8] = dct_const_round_shift(t1);
+ out[12] = dct_const_round_shift(t3);
+
+ // Stage 2
+ t0 = (s6 - s5) * cospi_16_64;
+ t1 = (s6 + s5) * cospi_16_64;
+ t2 = dct_const_round_shift(t0);
+ t3 = dct_const_round_shift(t1);
+
+ // Stage 3
+ x0 = s4 + t2;
+ x1 = s4 - t2;
+ x2 = s7 - t3;
+ x3 = s7 + t3;
+
+ // Stage 4
+ t0 = x0 * cospi_28_64 + x3 * cospi_4_64;
+ t1 = x1 * cospi_12_64 + x2 * cospi_20_64;
+ t2 = x2 * cospi_12_64 + x1 * -cospi_20_64;
+ t3 = x3 * cospi_28_64 + x0 * -cospi_4_64;
+ out[2] = dct_const_round_shift(t0);
+ out[6] = dct_const_round_shift(t2);
+ out[10] = dct_const_round_shift(t1);
+ out[14] = dct_const_round_shift(t3);
+ }
+ // Work on the next eight values; step1 -> odd_results
+ {+ // step 2
+ temp1 = (step1[5] - step1[2]) * cospi_16_64;
+ temp2 = (step1[4] - step1[3]) * cospi_16_64;
+ step2[2] = dct_const_round_shift(temp1);
+ step2[3] = dct_const_round_shift(temp2);
+ temp1 = (step1[4] + step1[3]) * cospi_16_64;
+ temp2 = (step1[5] + step1[2]) * cospi_16_64;
+ step2[4] = dct_const_round_shift(temp1);
+ step2[5] = dct_const_round_shift(temp2);
+ // step 3
+ step3[0] = step1[0] + step2[3];
+ step3[1] = step1[1] + step2[2];
+ step3[2] = step1[1] - step2[2];
+ step3[3] = step1[0] - step2[3];
+ step3[4] = step1[7] - step2[4];
+ step3[5] = step1[6] - step2[5];
+ step3[6] = step1[6] + step2[5];
+ step3[7] = step1[7] + step2[4];
+ // step 4
+ temp1 = step3[1] * -cospi_8_64 + step3[6] * cospi_24_64;
+ temp2 = step3[2] * -cospi_24_64 - step3[5] * cospi_8_64;
+ step2[1] = dct_const_round_shift(temp1);
+ step2[2] = dct_const_round_shift(temp2);
+ temp1 = step3[2] * -cospi_8_64 + step3[5] * cospi_24_64;
+ temp2 = step3[1] * cospi_24_64 + step3[6] * cospi_8_64;
+ step2[5] = dct_const_round_shift(temp1);
+ step2[6] = dct_const_round_shift(temp2);
+ // step 5
+ step1[0] = step3[0] + step2[1];
+ step1[1] = step3[0] - step2[1];
+ step1[2] = step3[3] - step2[2];
+ step1[3] = step3[3] + step2[2];
+ step1[4] = step3[4] + step2[5];
+ step1[5] = step3[4] - step2[5];
+ step1[6] = step3[7] - step2[6];
+ step1[7] = step3[7] + step2[6];
+ // step 6
+ temp1 = step1[0] * cospi_30_64 + step1[7] * cospi_2_64;
+ temp2 = step1[1] * cospi_14_64 + step1[6] * cospi_18_64;
+ out[1] = dct_const_round_shift(temp1);
+ out[9] = dct_const_round_shift(temp2);
+ temp1 = step1[2] * cospi_22_64 + step1[5] * cospi_10_64;
+ temp2 = step1[3] * cospi_6_64 + step1[4] * cospi_26_64;
+ out[5] = dct_const_round_shift(temp1);
+ out[13] = dct_const_round_shift(temp2);
+ temp1 = step1[3] * -cospi_26_64 + step1[4] * cospi_6_64;
+ temp2 = step1[2] * -cospi_10_64 + step1[5] * cospi_22_64;
+ out[3] = dct_const_round_shift(temp1);
+ out[11] = dct_const_round_shift(temp2);
+ temp1 = step1[1] * -cospi_18_64 + step1[6] * cospi_14_64;
+ temp2 = step1[0] * -cospi_2_64 + step1[7] * cospi_30_64;
+ out[7] = dct_const_round_shift(temp1);
+ out[15] = dct_const_round_shift(temp2);
+ }
+ // Do next column (which is a transposed row in second/horizontal pass)
+ in++;
+ out += 16;
+ }
+ // Setup in/out for next pass.
+ in = intermediate;
+ out = output;
+ }
+}
+
static void fadst8_1d(int16_t *input, int16_t *output) {int s0, s1, s2, s3, s4, s5, s6, s7;
@@ -421,132 +600,145 @@
// Rewrote to use same algorithm as others.
-static void fdct16_1d(int16_t input[16], int16_t output[16]) {- int16_t step[16];
- int temp1, temp2;
+static void fdct16_1d(int16_t in[16], int16_t out[16]) {+ /*canbe16*/ int step1[8];
+ /*canbe16*/ int step2[8];
+ /*canbe16*/ int step3[8];
+ /*canbe16*/ int input[8];
+ /*needs32*/ int temp1, temp2;
// step 1
- step[ 0] = input[0] + input[15];
- step[ 1] = input[1] + input[14];
- step[ 2] = input[2] + input[13];
- step[ 3] = input[3] + input[12];
- step[ 4] = input[4] + input[11];
- step[ 5] = input[5] + input[10];
- step[ 6] = input[6] + input[ 9];
- step[ 7] = input[7] + input[ 8];
- step[ 8] = input[7] - input[ 8];
- step[ 9] = input[6] - input[ 9];
- step[10] = input[5] - input[10];
- step[11] = input[4] - input[11];
- step[12] = input[3] - input[12];
- step[13] = input[2] - input[13];
- step[14] = input[1] - input[14];
- step[15] = input[0] - input[15];
+ input[0] = in[0] + in[15];
+ input[1] = in[1] + in[14];
+ input[2] = in[2] + in[13];
+ input[3] = in[3] + in[12];
+ input[4] = in[4] + in[11];
+ input[5] = in[5] + in[10];
+ input[6] = in[6] + in[ 9];
+ input[7] = in[7] + in[ 8];
- fdct8_1d(step, step);
+ step1[0] = in[7] - in[ 8];
+ step1[1] = in[6] - in[ 9];
+ step1[2] = in[5] - in[10];
+ step1[3] = in[4] - in[11];
+ step1[4] = in[3] - in[12];
+ step1[5] = in[2] - in[13];
+ step1[6] = in[1] - in[14];
+ step1[7] = in[0] - in[15];
+ // fdct8_1d(step, step);
+ {+ /*canbe16*/ int s0, s1, s2, s3, s4, s5, s6, s7;
+ /*needs32*/ int t0, t1, t2, t3;
+ /*canbe16*/ int x0, x1, x2, x3;
+
+ // stage 1
+ s0 = input[0] + input[7];
+ s1 = input[1] + input[6];
+ s2 = input[2] + input[5];
+ s3 = input[3] + input[4];
+ s4 = input[3] - input[4];
+ s5 = input[2] - input[5];
+ s6 = input[1] - input[6];
+ s7 = input[0] - input[7];
+
+ // fdct4_1d(step, step);
+ x0 = s0 + s3;
+ x1 = s1 + s2;
+ x2 = s1 - s2;
+ x3 = s0 - s3;
+ t0 = (x0 + x1) * cospi_16_64;
+ t1 = (x0 - x1) * cospi_16_64;
+ t2 = x3 * cospi_8_64 + x2 * cospi_24_64;
+ t3 = x3 * cospi_24_64 - x2 * cospi_8_64;
+ out[0] = dct_const_round_shift(t0);
+ out[4] = dct_const_round_shift(t2);
+ out[8] = dct_const_round_shift(t1);
+ out[12] = dct_const_round_shift(t3);
+
+ // Stage 2
+ t0 = (s6 - s5) * cospi_16_64;
+ t1 = (s6 + s5) * cospi_16_64;
+ t2 = dct_const_round_shift(t0);
+ t3 = dct_const_round_shift(t1);
+
+ // Stage 3
+ x0 = s4 + t2;
+ x1 = s4 - t2;
+ x2 = s7 - t3;
+ x3 = s7 + t3;
+
+ // Stage 4
+ t0 = x0 * cospi_28_64 + x3 * cospi_4_64;
+ t1 = x1 * cospi_12_64 + x2 * cospi_20_64;
+ t2 = x2 * cospi_12_64 + x1 * -cospi_20_64;
+ t3 = x3 * cospi_28_64 + x0 * -cospi_4_64;
+ out[2] = dct_const_round_shift(t0);
+ out[6] = dct_const_round_shift(t2);
+ out[10] = dct_const_round_shift(t1);
+ out[14] = dct_const_round_shift(t3);
+ }
+
// step 2
- output[8] = step[8];
- output[9] = step[9];
- temp1 = (-step[10] + step[13]) * cospi_16_64;
- temp2 = (-step[11] + step[12]) * cospi_16_64;
- output[10] = dct_const_round_shift(temp1);
- output[11] = dct_const_round_shift(temp2);
- temp1 = (step[11] + step[12]) * cospi_16_64;
- temp2 = (step[10] + step[13]) * cospi_16_64;
- output[12] = dct_const_round_shift(temp1);
- output[13] = dct_const_round_shift(temp2);
- output[14] = step[14];
- output[15] = step[15];
+ temp1 = (step1[5] - step1[2]) * cospi_16_64;
+ temp2 = (step1[4] - step1[3]) * cospi_16_64;
+ step2[2] = dct_const_round_shift(temp1);
+ step2[3] = dct_const_round_shift(temp2);
+ temp1 = (step1[4] + step1[3]) * cospi_16_64;
+ temp2 = (step1[5] + step1[2]) * cospi_16_64;
+ step2[4] = dct_const_round_shift(temp1);
+ step2[5] = dct_const_round_shift(temp2);
// step 3
- step[ 8] = output[8] + output[11];
- step[ 9] = output[9] + output[10];
- step[ 10] = output[9] - output[10];
- step[ 11] = output[8] - output[11];
- step[ 12] = -output[12] + output[15];
- step[ 13] = -output[13] + output[14];
- step[ 14] = output[13] + output[14];
- step[ 15] = output[12] + output[15];
+ step3[0] = step1[0] + step2[3];
+ step3[1] = step1[1] + step2[2];
+ step3[2] = step1[1] - step2[2];
+ step3[3] = step1[0] - step2[3];
+ step3[4] = step1[7] - step2[4];
+ step3[5] = step1[6] - step2[5];
+ step3[6] = step1[6] + step2[5];
+ step3[7] = step1[7] + step2[4];
// step 4
- output[8] = step[8];
- temp1 = -step[9] * cospi_8_64 + step[14] * cospi_24_64;
- temp2 = -step[10] * cospi_24_64 - step[13] * cospi_8_64;
- output[9] = dct_const_round_shift(temp1);
- output[10] = dct_const_round_shift(temp2);
- output[11] = step[11];
- output[12] = step[12];
- temp1 = -step[10] * cospi_8_64 + step[13] * cospi_24_64;
- temp2 = step[9] * cospi_24_64 + step[14] * cospi_8_64;
- output[13] = dct_const_round_shift(temp1);
- output[14] = dct_const_round_shift(temp2);
- output[15] = step[15];
+ temp1 = step3[1] * -cospi_8_64 + step3[6] * cospi_24_64;
+ temp2 = step3[2] * -cospi_24_64 - step3[5] * cospi_8_64;
+ step2[1] = dct_const_round_shift(temp1);
+ step2[2] = dct_const_round_shift(temp2);
+ temp1 = step3[2] * -cospi_8_64 + step3[5] * cospi_24_64;
+ temp2 = step3[1] * cospi_24_64 + step3[6] * cospi_8_64;
+ step2[5] = dct_const_round_shift(temp1);
+ step2[6] = dct_const_round_shift(temp2);
// step 5
- step[8] = output[8] + output[9];
- step[9] = output[8] - output[9];
- step[10] = -output[10] + output[11];
- step[11] = output[10] + output[11];
- step[12] = output[12] + output[13];
- step[13] = output[12] - output[13];
- step[14] = -output[14] + output[15];
- step[15] = output[14] + output[15];
+ step1[0] = step3[0] + step2[1];
+ step1[1] = step3[0] - step2[1];
+ step1[2] = step3[3] - step2[2];
+ step1[3] = step3[3] + step2[2];
+ step1[4] = step3[4] + step2[5];
+ step1[5] = step3[4] - step2[5];
+ step1[6] = step3[7] - step2[6];
+ step1[7] = step3[7] + step2[6];
// step 6
- output[0] = step[0];
- output[8] = step[4];
- output[4] = step[2];
- output[12] = step[6];
- output[2] = step[1];
- output[10] = step[5];
- output[6] = step[3];
- output[14] = step[7];
+ temp1 = step1[0] * cospi_30_64 + step1[7] * cospi_2_64;
+ temp2 = step1[1] * cospi_14_64 + step1[6] * cospi_18_64;
+ out[1] = dct_const_round_shift(temp1);
+ out[9] = dct_const_round_shift(temp2);
- temp1 = step[8] * cospi_30_64 + step[15] * cospi_2_64;
- temp2 = step[9] * cospi_14_64 + step[14] * cospi_18_64;
- output[1] = dct_const_round_shift(temp1);
- output[9] = dct_const_round_shift(temp2);
+ temp1 = step1[2] * cospi_22_64 + step1[5] * cospi_10_64;
+ temp2 = step1[3] * cospi_6_64 + step1[4] * cospi_26_64;
+ out[5] = dct_const_round_shift(temp1);
+ out[13] = dct_const_round_shift(temp2);
- temp1 = step[10] * cospi_22_64 + step[13] * cospi_10_64;
- temp2 = step[11] * cospi_6_64 + step[12] * cospi_26_64;
- output[5] = dct_const_round_shift(temp1);
- output[13] = dct_const_round_shift(temp2);
+ temp1 = step1[3] * -cospi_26_64 + step1[4] * cospi_6_64;
+ temp2 = step1[2] * -cospi_10_64 + step1[5] * cospi_22_64;
+ out[3] = dct_const_round_shift(temp1);
+ out[11] = dct_const_round_shift(temp2);
- temp1 = -step[11] * cospi_26_64 + step[12] * cospi_6_64;
- temp2 = -step[10] * cospi_10_64 + step[13] * cospi_22_64;
- output[3] = dct_const_round_shift(temp1);
- output[11] = dct_const_round_shift(temp2);
-
- temp1 = -step[9] * cospi_18_64 + step[14] * cospi_14_64;
- temp2 = -step[8] * cospi_2_64 + step[15] * cospi_30_64;
- output[7] = dct_const_round_shift(temp1);
- output[15] = dct_const_round_shift(temp2);
-}
-
-void vp9_short_fdct16x16_c(int16_t *input, int16_t *out, int pitch) {- int shortpitch = pitch >> 1;
- int i, j;
- int16_t output[256];
- int16_t temp_in[16], temp_out[16];
-
- // First transform columns
- for (i = 0; i < 16; i++) {- for (j = 0; j < 16; j++)
- temp_in[j] = input[j * shortpitch + i] << 2;
- fdct16_1d(temp_in, temp_out);
- for (j = 0; j < 16; j++)
- output[j * 16 + i] = (temp_out[j] + 1) >> 2;
- }
-
- // Then transform rows
- for (i = 0; i < 16; ++i) {- for (j = 0; j < 16; ++j)
- temp_in[j] = output[j + i * 16];
- fdct16_1d(temp_in, temp_out);
- for (j = 0; j < 16; ++j)
- out[j + i * 16] = temp_out[j];
- }
+ temp1 = step1[1] * -cospi_18_64 + step1[6] * cospi_14_64;
+ temp2 = step1[0] * -cospi_2_64 + step1[7] * cospi_30_64;
+ out[7] = dct_const_round_shift(temp1);
+ out[15] = dct_const_round_shift(temp2);
}
void fadst16_1d(int16_t *input, int16_t *output) {--- a/vp9/encoder/x86/vp9_dct_sse2_intrinsics.c
+++ b/vp9/encoder/x86/vp9_dct_sse2_intrinsics.c
@@ -270,3 +270,629 @@
_mm_storeu_si128 ((__m128i *)(output + 7 * 8), in7);
}
}
+
+void vp9_short_fdct16x16_sse2(int16_t *input, int16_t *output, int pitch) {+ // The 2D transform is done with two passes which are actually pretty
+ // similar. In the first one, we transform the columns and transpose
+ // the results. In the second one, we transform the rows. To achieve that,
+ // as the first pass results are transposed, we tranpose the columns (that
+ // is the transposed rows) and transpose the results (so that it goes back
+ // in normal/row positions).
+ const int stride = pitch >> 1;
+ int pass;
+ // We need an intermediate buffer between passes.
+ int16_t intermediate[256];
+ int16_t *in = input;
+ int16_t *out = intermediate;
+ // Constants
+ // When we use them, in one case, they are all the same. In all others
+ // it's a pair of them that we need to repeat four times. This is done
+ // by constructing the 32 bit constant corresponding to that pair.
+ const __m128i k__cospi_p16_p16 = _mm_set1_epi16(cospi_16_64);
+ const __m128i k__cospi_p16_m16 = pair_set_epi16(cospi_16_64, -cospi_16_64);
+ const __m128i k__cospi_p24_p08 = pair_set_epi16(cospi_24_64, cospi_8_64);
+ const __m128i k__cospi_m24_m08 = pair_set_epi16(-cospi_24_64, -cospi_8_64);
+ const __m128i k__cospi_m08_p24 = pair_set_epi16(-cospi_8_64, cospi_24_64);
+ const __m128i k__cospi_p28_p04 = pair_set_epi16(cospi_28_64, cospi_4_64);
+ const __m128i k__cospi_m04_p28 = pair_set_epi16(-cospi_4_64, cospi_28_64);
+ const __m128i k__cospi_p12_p20 = pair_set_epi16(cospi_12_64, cospi_20_64);
+ const __m128i k__cospi_m20_p12 = pair_set_epi16(-cospi_20_64, cospi_12_64);
+ const __m128i k__cospi_p30_p02 = pair_set_epi16(cospi_30_64, cospi_2_64);
+ const __m128i k__cospi_p14_p18 = pair_set_epi16(cospi_14_64, cospi_18_64);
+ const __m128i k__cospi_m02_p30 = pair_set_epi16(-cospi_2_64, cospi_30_64);
+ const __m128i k__cospi_m18_p14 = pair_set_epi16(-cospi_18_64, cospi_14_64);
+ const __m128i k__cospi_p22_p10 = pair_set_epi16(cospi_22_64, cospi_10_64);
+ const __m128i k__cospi_p06_p26 = pair_set_epi16(cospi_6_64, cospi_26_64);
+ const __m128i k__cospi_m10_p22 = pair_set_epi16(-cospi_10_64, cospi_22_64);
+ const __m128i k__cospi_m26_p06 = pair_set_epi16(-cospi_26_64, cospi_6_64);
+ const __m128i k__DCT_CONST_ROUNDING = _mm_set1_epi32(DCT_CONST_ROUNDING);
+ const __m128i kOne = _mm_set1_epi16(1);
+ // Do the two transform/transpose passes
+ for (pass = 0; pass < 2; ++pass) {+ // We process eight columns (transposed rows in second pass) at a time.
+ int column_start;
+ for (column_start = 0; column_start < 16; column_start += 8) {+ __m128i in00, in01, in02, in03, in04, in05, in06, in07;
+ __m128i in08, in09, in10, in11, in12, in13, in14, in15;
+ __m128i input0, input1, input2, input3, input4, input5, input6, input7;
+ __m128i step1_0, step1_1, step1_2, step1_3;
+ __m128i step1_4, step1_5, step1_6, step1_7;
+ __m128i step2_1, step2_2, step2_3, step2_4, step2_5, step2_6;
+ __m128i step3_0, step3_1, step3_2, step3_3;
+ __m128i step3_4, step3_5, step3_6, step3_7;
+ __m128i res00, res01, res02, res03, res04, res05, res06, res07;
+ __m128i res08, res09, res10, res11, res12, res13, res14, res15;
+ // Load and pre-condition input.
+ if (0 == pass) {+ in00 = _mm_loadu_si128((const __m128i *)(in + 0 * stride));
+ in01 = _mm_loadu_si128((const __m128i *)(in + 1 * stride));
+ in02 = _mm_loadu_si128((const __m128i *)(in + 2 * stride));
+ in03 = _mm_loadu_si128((const __m128i *)(in + 3 * stride));
+ in04 = _mm_loadu_si128((const __m128i *)(in + 4 * stride));
+ in05 = _mm_loadu_si128((const __m128i *)(in + 5 * stride));
+ in06 = _mm_loadu_si128((const __m128i *)(in + 6 * stride));
+ in07 = _mm_loadu_si128((const __m128i *)(in + 7 * stride));
+ in08 = _mm_loadu_si128((const __m128i *)(in + 8 * stride));
+ in09 = _mm_loadu_si128((const __m128i *)(in + 9 * stride));
+ in10 = _mm_loadu_si128((const __m128i *)(in + 10 * stride));
+ in11 = _mm_loadu_si128((const __m128i *)(in + 11 * stride));
+ in12 = _mm_loadu_si128((const __m128i *)(in + 12 * stride));
+ in13 = _mm_loadu_si128((const __m128i *)(in + 13 * stride));
+ in14 = _mm_loadu_si128((const __m128i *)(in + 14 * stride));
+ in15 = _mm_loadu_si128((const __m128i *)(in + 15 * stride));
+ // x = x << 2
+ in00 = _mm_slli_epi16(in00, 2);
+ in01 = _mm_slli_epi16(in01, 2);
+ in02 = _mm_slli_epi16(in02, 2);
+ in03 = _mm_slli_epi16(in03, 2);
+ in04 = _mm_slli_epi16(in04, 2);
+ in05 = _mm_slli_epi16(in05, 2);
+ in06 = _mm_slli_epi16(in06, 2);
+ in07 = _mm_slli_epi16(in07, 2);
+ in08 = _mm_slli_epi16(in08, 2);
+ in09 = _mm_slli_epi16(in09, 2);
+ in10 = _mm_slli_epi16(in10, 2);
+ in11 = _mm_slli_epi16(in11, 2);
+ in12 = _mm_slli_epi16(in12, 2);
+ in13 = _mm_slli_epi16(in13, 2);
+ in14 = _mm_slli_epi16(in14, 2);
+ in15 = _mm_slli_epi16(in15, 2);
+ } else {+ in00 = _mm_loadu_si128((const __m128i *)(in + 0 * 16));
+ in01 = _mm_loadu_si128((const __m128i *)(in + 1 * 16));
+ in02 = _mm_loadu_si128((const __m128i *)(in + 2 * 16));
+ in03 = _mm_loadu_si128((const __m128i *)(in + 3 * 16));
+ in04 = _mm_loadu_si128((const __m128i *)(in + 4 * 16));
+ in05 = _mm_loadu_si128((const __m128i *)(in + 5 * 16));
+ in06 = _mm_loadu_si128((const __m128i *)(in + 6 * 16));
+ in07 = _mm_loadu_si128((const __m128i *)(in + 7 * 16));
+ in08 = _mm_loadu_si128((const __m128i *)(in + 8 * 16));
+ in09 = _mm_loadu_si128((const __m128i *)(in + 9 * 16));
+ in10 = _mm_loadu_si128((const __m128i *)(in + 10 * 16));
+ in11 = _mm_loadu_si128((const __m128i *)(in + 11 * 16));
+ in12 = _mm_loadu_si128((const __m128i *)(in + 12 * 16));
+ in13 = _mm_loadu_si128((const __m128i *)(in + 13 * 16));
+ in14 = _mm_loadu_si128((const __m128i *)(in + 14 * 16));
+ in15 = _mm_loadu_si128((const __m128i *)(in + 15 * 16));
+ // x = (x + 1) >> 2
+ in00 = _mm_add_epi16(in00, kOne);
+ in01 = _mm_add_epi16(in01, kOne);
+ in02 = _mm_add_epi16(in02, kOne);
+ in03 = _mm_add_epi16(in03, kOne);
+ in04 = _mm_add_epi16(in04, kOne);
+ in05 = _mm_add_epi16(in05, kOne);
+ in06 = _mm_add_epi16(in06, kOne);
+ in07 = _mm_add_epi16(in07, kOne);
+ in08 = _mm_add_epi16(in08, kOne);
+ in09 = _mm_add_epi16(in09, kOne);
+ in10 = _mm_add_epi16(in10, kOne);
+ in11 = _mm_add_epi16(in11, kOne);
+ in12 = _mm_add_epi16(in12, kOne);
+ in13 = _mm_add_epi16(in13, kOne);
+ in14 = _mm_add_epi16(in14, kOne);
+ in15 = _mm_add_epi16(in15, kOne);
+ in00 = _mm_srai_epi16(in00, 2);
+ in01 = _mm_srai_epi16(in01, 2);
+ in02 = _mm_srai_epi16(in02, 2);
+ in03 = _mm_srai_epi16(in03, 2);
+ in04 = _mm_srai_epi16(in04, 2);
+ in05 = _mm_srai_epi16(in05, 2);
+ in06 = _mm_srai_epi16(in06, 2);
+ in07 = _mm_srai_epi16(in07, 2);
+ in08 = _mm_srai_epi16(in08, 2);
+ in09 = _mm_srai_epi16(in09, 2);
+ in10 = _mm_srai_epi16(in10, 2);
+ in11 = _mm_srai_epi16(in11, 2);
+ in12 = _mm_srai_epi16(in12, 2);
+ in13 = _mm_srai_epi16(in13, 2);
+ in14 = _mm_srai_epi16(in14, 2);
+ in15 = _mm_srai_epi16(in15, 2);
+ }
+ in += 8;
+ // Calculate input for the first 8 results.
+ {+ input0 = _mm_add_epi16(in00, in15);
+ input1 = _mm_add_epi16(in01, in14);
+ input2 = _mm_add_epi16(in02, in13);
+ input3 = _mm_add_epi16(in03, in12);
+ input4 = _mm_add_epi16(in04, in11);
+ input5 = _mm_add_epi16(in05, in10);
+ input6 = _mm_add_epi16(in06, in09);
+ input7 = _mm_add_epi16(in07, in08);
+ }
+ // Calculate input for the next 8 results.
+ {+ step1_0 = _mm_sub_epi16(in07, in08);
+ step1_1 = _mm_sub_epi16(in06, in09);
+ step1_2 = _mm_sub_epi16(in05, in10);
+ step1_3 = _mm_sub_epi16(in04, in11);
+ step1_4 = _mm_sub_epi16(in03, in12);
+ step1_5 = _mm_sub_epi16(in02, in13);
+ step1_6 = _mm_sub_epi16(in01, in14);
+ step1_7 = _mm_sub_epi16(in00, in15);
+ }
+ // Work on the first eight values; fdct8_1d(input, even_results);
+ {+ // Add/substract
+ const __m128i q0 = _mm_add_epi16(input0, input7);
+ const __m128i q1 = _mm_add_epi16(input1, input6);
+ const __m128i q2 = _mm_add_epi16(input2, input5);
+ const __m128i q3 = _mm_add_epi16(input3, input4);
+ const __m128i q4 = _mm_sub_epi16(input3, input4);
+ const __m128i q5 = _mm_sub_epi16(input2, input5);
+ const __m128i q6 = _mm_sub_epi16(input1, input6);
+ const __m128i q7 = _mm_sub_epi16(input0, input7);
+ // Work on first four results
+ {+ // Add/substract
+ const __m128i r0 = _mm_add_epi16(q0, q3);
+ const __m128i r1 = _mm_add_epi16(q1, q2);
+ const __m128i r2 = _mm_sub_epi16(q1, q2);
+ const __m128i r3 = _mm_sub_epi16(q0, q3);
+ // Interleave to do the multiply by constants which gets us
+ // into 32 bits.
+ const __m128i t0 = _mm_unpacklo_epi16(r0, r1);
+ const __m128i t1 = _mm_unpackhi_epi16(r0, r1);
+ const __m128i t2 = _mm_unpacklo_epi16(r2, r3);
+ const __m128i t3 = _mm_unpackhi_epi16(r2, r3);
+ const __m128i u0 = _mm_madd_epi16(t0, k__cospi_p16_p16);
+ const __m128i u1 = _mm_madd_epi16(t1, k__cospi_p16_p16);
+ const __m128i u2 = _mm_madd_epi16(t0, k__cospi_p16_m16);
+ const __m128i u3 = _mm_madd_epi16(t1, k__cospi_p16_m16);
+ const __m128i u4 = _mm_madd_epi16(t2, k__cospi_p24_p08);
+ const __m128i u5 = _mm_madd_epi16(t3, k__cospi_p24_p08);
+ const __m128i u6 = _mm_madd_epi16(t2, k__cospi_m08_p24);
+ const __m128i u7 = _mm_madd_epi16(t3, k__cospi_m08_p24);
+ // dct_const_round_shift
+ const __m128i v0 = _mm_add_epi32(u0, k__DCT_CONST_ROUNDING);
+ const __m128i v1 = _mm_add_epi32(u1, k__DCT_CONST_ROUNDING);
+ const __m128i v2 = _mm_add_epi32(u2, k__DCT_CONST_ROUNDING);
+ const __m128i v3 = _mm_add_epi32(u3, k__DCT_CONST_ROUNDING);
+ const __m128i v4 = _mm_add_epi32(u4, k__DCT_CONST_ROUNDING);
+ const __m128i v5 = _mm_add_epi32(u5, k__DCT_CONST_ROUNDING);
+ const __m128i v6 = _mm_add_epi32(u6, k__DCT_CONST_ROUNDING);
+ const __m128i v7 = _mm_add_epi32(u7, k__DCT_CONST_ROUNDING);
+ const __m128i w0 = _mm_srai_epi32(v0, DCT_CONST_BITS);
+ const __m128i w1 = _mm_srai_epi32(v1, DCT_CONST_BITS);
+ const __m128i w2 = _mm_srai_epi32(v2, DCT_CONST_BITS);
+ const __m128i w3 = _mm_srai_epi32(v3, DCT_CONST_BITS);
+ const __m128i w4 = _mm_srai_epi32(v4, DCT_CONST_BITS);
+ const __m128i w5 = _mm_srai_epi32(v5, DCT_CONST_BITS);
+ const __m128i w6 = _mm_srai_epi32(v6, DCT_CONST_BITS);
+ const __m128i w7 = _mm_srai_epi32(v7, DCT_CONST_BITS);
+ // Combine
+ res00 = _mm_packs_epi32(w0, w1);
+ res08 = _mm_packs_epi32(w2, w3);
+ res04 = _mm_packs_epi32(w4, w5);
+ res12 = _mm_packs_epi32(w6, w7);
+ }
+ // Work on next four results
+ {+ // Interleave to do the multiply by constants which gets us
+ // into 32 bits.
+ const __m128i d0 = _mm_unpacklo_epi16(q6, q5);
+ const __m128i d1 = _mm_unpackhi_epi16(q6, q5);
+ const __m128i e0 = _mm_madd_epi16(d0, k__cospi_p16_m16);
+ const __m128i e1 = _mm_madd_epi16(d1, k__cospi_p16_m16);
+ const __m128i e2 = _mm_madd_epi16(d0, k__cospi_p16_p16);
+ const __m128i e3 = _mm_madd_epi16(d1, k__cospi_p16_p16);
+ // dct_const_round_shift
+ const __m128i f0 = _mm_add_epi32(e0, k__DCT_CONST_ROUNDING);
+ const __m128i f1 = _mm_add_epi32(e1, k__DCT_CONST_ROUNDING);
+ const __m128i f2 = _mm_add_epi32(e2, k__DCT_CONST_ROUNDING);
+ const __m128i f3 = _mm_add_epi32(e3, k__DCT_CONST_ROUNDING);
+ const __m128i s0 = _mm_srai_epi32(f0, DCT_CONST_BITS);
+ const __m128i s1 = _mm_srai_epi32(f1, DCT_CONST_BITS);
+ const __m128i s2 = _mm_srai_epi32(f2, DCT_CONST_BITS);
+ const __m128i s3 = _mm_srai_epi32(f3, DCT_CONST_BITS);
+ // Combine
+ const __m128i r0 = _mm_packs_epi32(s0, s1);
+ const __m128i r1 = _mm_packs_epi32(s2, s3);
+ // Add/substract
+ const __m128i x0 = _mm_add_epi16(q4, r0);
+ const __m128i x1 = _mm_sub_epi16(q4, r0);
+ const __m128i x2 = _mm_sub_epi16(q7, r1);
+ const __m128i x3 = _mm_add_epi16(q7, r1);
+ // Interleave to do the multiply by constants which gets us
+ // into 32 bits.
+ const __m128i t0 = _mm_unpacklo_epi16(x0, x3);
+ const __m128i t1 = _mm_unpackhi_epi16(x0, x3);
+ const __m128i t2 = _mm_unpacklo_epi16(x1, x2);
+ const __m128i t3 = _mm_unpackhi_epi16(x1, x2);
+ const __m128i u0 = _mm_madd_epi16(t0, k__cospi_p28_p04);
+ const __m128i u1 = _mm_madd_epi16(t1, k__cospi_p28_p04);
+ const __m128i u2 = _mm_madd_epi16(t0, k__cospi_m04_p28);
+ const __m128i u3 = _mm_madd_epi16(t1, k__cospi_m04_p28);
+ const __m128i u4 = _mm_madd_epi16(t2, k__cospi_p12_p20);
+ const __m128i u5 = _mm_madd_epi16(t3, k__cospi_p12_p20);
+ const __m128i u6 = _mm_madd_epi16(t2, k__cospi_m20_p12);
+ const __m128i u7 = _mm_madd_epi16(t3, k__cospi_m20_p12);
+ // dct_const_round_shift
+ const __m128i v0 = _mm_add_epi32(u0, k__DCT_CONST_ROUNDING);
+ const __m128i v1 = _mm_add_epi32(u1, k__DCT_CONST_ROUNDING);
+ const __m128i v2 = _mm_add_epi32(u2, k__DCT_CONST_ROUNDING);
+ const __m128i v3 = _mm_add_epi32(u3, k__DCT_CONST_ROUNDING);
+ const __m128i v4 = _mm_add_epi32(u4, k__DCT_CONST_ROUNDING);
+ const __m128i v5 = _mm_add_epi32(u5, k__DCT_CONST_ROUNDING);
+ const __m128i v6 = _mm_add_epi32(u6, k__DCT_CONST_ROUNDING);
+ const __m128i v7 = _mm_add_epi32(u7, k__DCT_CONST_ROUNDING);
+ const __m128i w0 = _mm_srai_epi32(v0, DCT_CONST_BITS);
+ const __m128i w1 = _mm_srai_epi32(v1, DCT_CONST_BITS);
+ const __m128i w2 = _mm_srai_epi32(v2, DCT_CONST_BITS);
+ const __m128i w3 = _mm_srai_epi32(v3, DCT_CONST_BITS);
+ const __m128i w4 = _mm_srai_epi32(v4, DCT_CONST_BITS);
+ const __m128i w5 = _mm_srai_epi32(v5, DCT_CONST_BITS);
+ const __m128i w6 = _mm_srai_epi32(v6, DCT_CONST_BITS);
+ const __m128i w7 = _mm_srai_epi32(v7, DCT_CONST_BITS);
+ // Combine
+ res02 = _mm_packs_epi32(w0, w1);
+ res14 = _mm_packs_epi32(w2, w3);
+ res10 = _mm_packs_epi32(w4, w5);
+ res06 = _mm_packs_epi32(w6, w7);
+ }
+ }
+ // Work on the next eight values; step1 -> odd_results
+ {+ // step 2
+ {+ const __m128i t0 = _mm_unpacklo_epi16(step1_5, step1_2);
+ const __m128i t1 = _mm_unpackhi_epi16(step1_5, step1_2);
+ const __m128i t2 = _mm_unpacklo_epi16(step1_4, step1_3);
+ const __m128i t3 = _mm_unpackhi_epi16(step1_4, step1_3);
+ const __m128i u0 = _mm_madd_epi16(t0, k__cospi_p16_m16);
+ const __m128i u1 = _mm_madd_epi16(t1, k__cospi_p16_m16);
+ const __m128i u2 = _mm_madd_epi16(t2, k__cospi_p16_m16);
+ const __m128i u3 = _mm_madd_epi16(t3, k__cospi_p16_m16);
+ // dct_const_round_shift
+ const __m128i v0 = _mm_add_epi32(u0, k__DCT_CONST_ROUNDING);
+ const __m128i v1 = _mm_add_epi32(u1, k__DCT_CONST_ROUNDING);
+ const __m128i v2 = _mm_add_epi32(u2, k__DCT_CONST_ROUNDING);
+ const __m128i v3 = _mm_add_epi32(u3, k__DCT_CONST_ROUNDING);
+ const __m128i w0 = _mm_srai_epi32(v0, DCT_CONST_BITS);
+ const __m128i w1 = _mm_srai_epi32(v1, DCT_CONST_BITS);
+ const __m128i w2 = _mm_srai_epi32(v2, DCT_CONST_BITS);
+ const __m128i w3 = _mm_srai_epi32(v3, DCT_CONST_BITS);
+ // Combine
+ step2_2 = _mm_packs_epi32(w0, w1);
+ step2_3 = _mm_packs_epi32(w2, w3);
+ }
+ {+ const __m128i t0 = _mm_unpacklo_epi16(step1_5, step1_2);
+ const __m128i t1 = _mm_unpackhi_epi16(step1_5, step1_2);
+ const __m128i t2 = _mm_unpacklo_epi16(step1_4, step1_3);
+ const __m128i t3 = _mm_unpackhi_epi16(step1_4, step1_3);
+ const __m128i u0 = _mm_madd_epi16(t0, k__cospi_p16_p16);
+ const __m128i u1 = _mm_madd_epi16(t1, k__cospi_p16_p16);
+ const __m128i u2 = _mm_madd_epi16(t2, k__cospi_p16_p16);
+ const __m128i u3 = _mm_madd_epi16(t3, k__cospi_p16_p16);
+ // dct_const_round_shift
+ const __m128i v0 = _mm_add_epi32(u0, k__DCT_CONST_ROUNDING);
+ const __m128i v1 = _mm_add_epi32(u1, k__DCT_CONST_ROUNDING);
+ const __m128i v2 = _mm_add_epi32(u2, k__DCT_CONST_ROUNDING);
+ const __m128i v3 = _mm_add_epi32(u3, k__DCT_CONST_ROUNDING);
+ const __m128i w0 = _mm_srai_epi32(v0, DCT_CONST_BITS);
+ const __m128i w1 = _mm_srai_epi32(v1, DCT_CONST_BITS);
+ const __m128i w2 = _mm_srai_epi32(v2, DCT_CONST_BITS);
+ const __m128i w3 = _mm_srai_epi32(v3, DCT_CONST_BITS);
+ // Combine
+ step2_5 = _mm_packs_epi32(w0, w1);
+ step2_4 = _mm_packs_epi32(w2, w3);
+ }
+ // step 3
+ {+ step3_0 = _mm_add_epi16(step1_0, step2_3);
+ step3_1 = _mm_add_epi16(step1_1, step2_2);
+ step3_2 = _mm_sub_epi16(step1_1, step2_2);
+ step3_3 = _mm_sub_epi16(step1_0, step2_3);
+ step3_4 = _mm_sub_epi16(step1_7, step2_4);
+ step3_5 = _mm_sub_epi16(step1_6, step2_5);
+ step3_6 = _mm_add_epi16(step1_6, step2_5);
+ step3_7 = _mm_add_epi16(step1_7, step2_4);
+ }
+ // step 4
+ {+ const __m128i t0 = _mm_unpacklo_epi16(step3_1, step3_6);
+ const __m128i t1 = _mm_unpackhi_epi16(step3_1, step3_6);
+ const __m128i t2 = _mm_unpacklo_epi16(step3_2, step3_5);
+ const __m128i t3 = _mm_unpackhi_epi16(step3_2, step3_5);
+ const __m128i u0 = _mm_madd_epi16(t0, k__cospi_m08_p24);
+ const __m128i u1 = _mm_madd_epi16(t1, k__cospi_m08_p24);
+ const __m128i u2 = _mm_madd_epi16(t2, k__cospi_m24_m08);
+ const __m128i u3 = _mm_madd_epi16(t3, k__cospi_m24_m08);
+ // dct_const_round_shift
+ const __m128i v0 = _mm_add_epi32(u0, k__DCT_CONST_ROUNDING);
+ const __m128i v1 = _mm_add_epi32(u1, k__DCT_CONST_ROUNDING);
+ const __m128i v2 = _mm_add_epi32(u2, k__DCT_CONST_ROUNDING);
+ const __m128i v3 = _mm_add_epi32(u3, k__DCT_CONST_ROUNDING);
+ const __m128i w0 = _mm_srai_epi32(v0, DCT_CONST_BITS);
+ const __m128i w1 = _mm_srai_epi32(v1, DCT_CONST_BITS);
+ const __m128i w2 = _mm_srai_epi32(v2, DCT_CONST_BITS);
+ const __m128i w3 = _mm_srai_epi32(v3, DCT_CONST_BITS);
+ // Combine
+ step2_1 = _mm_packs_epi32(w0, w1);
+ step2_2 = _mm_packs_epi32(w2, w3);
+ }
+ {+ const __m128i t0 = _mm_unpacklo_epi16(step3_1, step3_6);
+ const __m128i t1 = _mm_unpackhi_epi16(step3_1, step3_6);
+ const __m128i t2 = _mm_unpacklo_epi16(step3_2, step3_5);
+ const __m128i t3 = _mm_unpackhi_epi16(step3_2, step3_5);
+ const __m128i u0 = _mm_madd_epi16(t0, k__cospi_p24_p08);
+ const __m128i u1 = _mm_madd_epi16(t1, k__cospi_p24_p08);
+ const __m128i u2 = _mm_madd_epi16(t2, k__cospi_m08_p24);
+ const __m128i u3 = _mm_madd_epi16(t3, k__cospi_m08_p24);
+ // dct_const_round_shift
+ const __m128i v0 = _mm_add_epi32(u0, k__DCT_CONST_ROUNDING);
+ const __m128i v1 = _mm_add_epi32(u1, k__DCT_CONST_ROUNDING);
+ const __m128i v2 = _mm_add_epi32(u2, k__DCT_CONST_ROUNDING);
+ const __m128i v3 = _mm_add_epi32(u3, k__DCT_CONST_ROUNDING);
+ const __m128i w0 = _mm_srai_epi32(v0, DCT_CONST_BITS);
+ const __m128i w1 = _mm_srai_epi32(v1, DCT_CONST_BITS);
+ const __m128i w2 = _mm_srai_epi32(v2, DCT_CONST_BITS);
+ const __m128i w3 = _mm_srai_epi32(v3, DCT_CONST_BITS);
+ // Combine
+ step2_6 = _mm_packs_epi32(w0, w1);
+ step2_5 = _mm_packs_epi32(w2, w3);
+ }
+ // step 5
+ {+ step1_0 = _mm_add_epi16(step3_0, step2_1);
+ step1_1 = _mm_sub_epi16(step3_0, step2_1);
+ step1_2 = _mm_sub_epi16(step3_3, step2_2);
+ step1_3 = _mm_add_epi16(step3_3, step2_2);
+ step1_4 = _mm_add_epi16(step3_4, step2_5);
+ step1_5 = _mm_sub_epi16(step3_4, step2_5);
+ step1_6 = _mm_sub_epi16(step3_7, step2_6);
+ step1_7 = _mm_add_epi16(step3_7, step2_6);
+ }
+ // step 6
+ {+ const __m128i t0 = _mm_unpacklo_epi16(step1_0, step1_7);
+ const __m128i t1 = _mm_unpackhi_epi16(step1_0, step1_7);
+ const __m128i t2 = _mm_unpacklo_epi16(step1_1, step1_6);
+ const __m128i t3 = _mm_unpackhi_epi16(step1_1, step1_6);
+ const __m128i u0 = _mm_madd_epi16(t0, k__cospi_p30_p02);
+ const __m128i u1 = _mm_madd_epi16(t1, k__cospi_p30_p02);
+ const __m128i u2 = _mm_madd_epi16(t2, k__cospi_p14_p18);
+ const __m128i u3 = _mm_madd_epi16(t3, k__cospi_p14_p18);
+ // dct_const_round_shift
+ const __m128i v0 = _mm_add_epi32(u0, k__DCT_CONST_ROUNDING);
+ const __m128i v1 = _mm_add_epi32(u1, k__DCT_CONST_ROUNDING);
+ const __m128i v2 = _mm_add_epi32(u2, k__DCT_CONST_ROUNDING);
+ const __m128i v3 = _mm_add_epi32(u3, k__DCT_CONST_ROUNDING);
+ const __m128i w0 = _mm_srai_epi32(v0, DCT_CONST_BITS);
+ const __m128i w1 = _mm_srai_epi32(v1, DCT_CONST_BITS);
+ const __m128i w2 = _mm_srai_epi32(v2, DCT_CONST_BITS);
+ const __m128i w3 = _mm_srai_epi32(v3, DCT_CONST_BITS);
+ // Combine
+ res01 = _mm_packs_epi32(w0, w1);
+ res09 = _mm_packs_epi32(w2, w3);
+ }
+ {+ const __m128i t0 = _mm_unpacklo_epi16(step1_2, step1_5);
+ const __m128i t1 = _mm_unpackhi_epi16(step1_2, step1_5);
+ const __m128i t2 = _mm_unpacklo_epi16(step1_3, step1_4);
+ const __m128i t3 = _mm_unpackhi_epi16(step1_3, step1_4);
+ const __m128i u0 = _mm_madd_epi16(t0, k__cospi_p22_p10);
+ const __m128i u1 = _mm_madd_epi16(t1, k__cospi_p22_p10);
+ const __m128i u2 = _mm_madd_epi16(t2, k__cospi_p06_p26);
+ const __m128i u3 = _mm_madd_epi16(t3, k__cospi_p06_p26);
+ // dct_const_round_shift
+ const __m128i v0 = _mm_add_epi32(u0, k__DCT_CONST_ROUNDING);
+ const __m128i v1 = _mm_add_epi32(u1, k__DCT_CONST_ROUNDING);
+ const __m128i v2 = _mm_add_epi32(u2, k__DCT_CONST_ROUNDING);
+ const __m128i v3 = _mm_add_epi32(u3, k__DCT_CONST_ROUNDING);
+ const __m128i w0 = _mm_srai_epi32(v0, DCT_CONST_BITS);
+ const __m128i w1 = _mm_srai_epi32(v1, DCT_CONST_BITS);
+ const __m128i w2 = _mm_srai_epi32(v2, DCT_CONST_BITS);
+ const __m128i w3 = _mm_srai_epi32(v3, DCT_CONST_BITS);
+ // Combine
+ res05 = _mm_packs_epi32(w0, w1);
+ res13 = _mm_packs_epi32(w2, w3);
+ }
+ {+ const __m128i t0 = _mm_unpacklo_epi16(step1_2, step1_5);
+ const __m128i t1 = _mm_unpackhi_epi16(step1_2, step1_5);
+ const __m128i t2 = _mm_unpacklo_epi16(step1_3, step1_4);
+ const __m128i t3 = _mm_unpackhi_epi16(step1_3, step1_4);
+ const __m128i u0 = _mm_madd_epi16(t0, k__cospi_m10_p22);
+ const __m128i u1 = _mm_madd_epi16(t1, k__cospi_m10_p22);
+ const __m128i u2 = _mm_madd_epi16(t2, k__cospi_m26_p06);
+ const __m128i u3 = _mm_madd_epi16(t3, k__cospi_m26_p06);
+ // dct_const_round_shift
+ const __m128i v0 = _mm_add_epi32(u0, k__DCT_CONST_ROUNDING);
+ const __m128i v1 = _mm_add_epi32(u1, k__DCT_CONST_ROUNDING);
+ const __m128i v2 = _mm_add_epi32(u2, k__DCT_CONST_ROUNDING);
+ const __m128i v3 = _mm_add_epi32(u3, k__DCT_CONST_ROUNDING);
+ const __m128i w0 = _mm_srai_epi32(v0, DCT_CONST_BITS);
+ const __m128i w1 = _mm_srai_epi32(v1, DCT_CONST_BITS);
+ const __m128i w2 = _mm_srai_epi32(v2, DCT_CONST_BITS);
+ const __m128i w3 = _mm_srai_epi32(v3, DCT_CONST_BITS);
+ // Combine
+ res11 = _mm_packs_epi32(w0, w1);
+ res03 = _mm_packs_epi32(w2, w3);
+ }
+ {+ const __m128i t0 = _mm_unpacklo_epi16(step1_0, step1_7);
+ const __m128i t1 = _mm_unpackhi_epi16(step1_0, step1_7);
+ const __m128i t2 = _mm_unpacklo_epi16(step1_1, step1_6);
+ const __m128i t3 = _mm_unpackhi_epi16(step1_1, step1_6);
+ const __m128i u0 = _mm_madd_epi16(t0, k__cospi_m02_p30);
+ const __m128i u1 = _mm_madd_epi16(t1, k__cospi_m02_p30);
+ const __m128i u2 = _mm_madd_epi16(t2, k__cospi_m18_p14);
+ const __m128i u3 = _mm_madd_epi16(t3, k__cospi_m18_p14);
+ // dct_const_round_shift
+ const __m128i v0 = _mm_add_epi32(u0, k__DCT_CONST_ROUNDING);
+ const __m128i v1 = _mm_add_epi32(u1, k__DCT_CONST_ROUNDING);
+ const __m128i v2 = _mm_add_epi32(u2, k__DCT_CONST_ROUNDING);
+ const __m128i v3 = _mm_add_epi32(u3, k__DCT_CONST_ROUNDING);
+ const __m128i w0 = _mm_srai_epi32(v0, DCT_CONST_BITS);
+ const __m128i w1 = _mm_srai_epi32(v1, DCT_CONST_BITS);
+ const __m128i w2 = _mm_srai_epi32(v2, DCT_CONST_BITS);
+ const __m128i w3 = _mm_srai_epi32(v3, DCT_CONST_BITS);
+ // Combine
+ res15 = _mm_packs_epi32(w0, w1);
+ res07 = _mm_packs_epi32(w2, w3);
+ }
+ }
+ // Transpose the results, do it as two 8x8 transposes.
+ {+ // 00 01 02 03 04 05 06 07
+ // 10 11 12 13 14 15 16 17
+ // 20 21 22 23 24 25 26 27
+ // 30 31 32 33 34 35 36 37
+ // 40 41 42 43 44 45 46 47
+ // 50 51 52 53 54 55 56 57
+ // 60 61 62 63 64 65 66 67
+ // 70 71 72 73 74 75 76 77
+ const __m128i tr0_0 = _mm_unpacklo_epi16(res00, res01);
+ const __m128i tr0_1 = _mm_unpacklo_epi16(res02, res03);
+ const __m128i tr0_2 = _mm_unpackhi_epi16(res00, res01);
+ const __m128i tr0_3 = _mm_unpackhi_epi16(res02, res03);
+ const __m128i tr0_4 = _mm_unpacklo_epi16(res04, res05);
+ const __m128i tr0_5 = _mm_unpacklo_epi16(res06, res07);
+ const __m128i tr0_6 = _mm_unpackhi_epi16(res04, res05);
+ const __m128i tr0_7 = _mm_unpackhi_epi16(res06, res07);
+ // 00 10 01 11 02 12 03 13
+ // 20 30 21 31 22 32 23 33
+ // 04 14 05 15 06 16 07 17
+ // 24 34 25 35 26 36 27 37
+ // 40 50 41 51 42 52 43 53
+ // 60 70 61 71 62 72 63 73
+ // 54 54 55 55 56 56 57 57
+ // 64 74 65 75 66 76 67 77
+ const __m128i tr1_0 = _mm_unpacklo_epi32(tr0_0, tr0_1);
+ const __m128i tr1_1 = _mm_unpacklo_epi32(tr0_2, tr0_3);
+ const __m128i tr1_2 = _mm_unpackhi_epi32(tr0_0, tr0_1);
+ const __m128i tr1_3 = _mm_unpackhi_epi32(tr0_2, tr0_3);
+ const __m128i tr1_4 = _mm_unpacklo_epi32(tr0_4, tr0_5);
+ const __m128i tr1_5 = _mm_unpacklo_epi32(tr0_6, tr0_7);
+ const __m128i tr1_6 = _mm_unpackhi_epi32(tr0_4, tr0_5);
+ const __m128i tr1_7 = _mm_unpackhi_epi32(tr0_6, tr0_7);
+ // 00 10 20 30 01 11 21 31
+ // 40 50 60 70 41 51 61 71
+ // 02 12 22 32 03 13 23 33
+ // 42 52 62 72 43 53 63 73
+ // 04 14 24 34 05 15 21 36
+ // 44 54 64 74 45 55 61 76
+ // 06 16 26 36 07 17 27 37
+ // 46 56 66 76 47 57 67 77
+ const __m128i tr2_0 = _mm_unpacklo_epi64(tr1_0, tr1_4);
+ const __m128i tr2_1 = _mm_unpackhi_epi64(tr1_0, tr1_4);
+ const __m128i tr2_2 = _mm_unpacklo_epi64(tr1_2, tr1_6);
+ const __m128i tr2_3 = _mm_unpackhi_epi64(tr1_2, tr1_6);
+ const __m128i tr2_4 = _mm_unpacklo_epi64(tr1_1, tr1_5);
+ const __m128i tr2_5 = _mm_unpackhi_epi64(tr1_1, tr1_5);
+ const __m128i tr2_6 = _mm_unpacklo_epi64(tr1_3, tr1_7);
+ const __m128i tr2_7 = _mm_unpackhi_epi64(tr1_3, tr1_7);
+ // 00 10 20 30 40 50 60 70
+ // 01 11 21 31 41 51 61 71
+ // 02 12 22 32 42 52 62 72
+ // 03 13 23 33 43 53 63 73
+ // 04 14 24 34 44 54 64 74
+ // 05 15 25 35 45 55 65 75
+ // 06 16 26 36 46 56 66 76
+ // 07 17 27 37 47 57 67 77
+ _mm_storeu_si128 ((__m128i *)(out + 0 * 16), tr2_0);
+ _mm_storeu_si128 ((__m128i *)(out + 1 * 16), tr2_1);
+ _mm_storeu_si128 ((__m128i *)(out + 2 * 16), tr2_2);
+ _mm_storeu_si128 ((__m128i *)(out + 3 * 16), tr2_3);
+ _mm_storeu_si128 ((__m128i *)(out + 4 * 16), tr2_4);
+ _mm_storeu_si128 ((__m128i *)(out + 5 * 16), tr2_5);
+ _mm_storeu_si128 ((__m128i *)(out + 6 * 16), tr2_6);
+ _mm_storeu_si128 ((__m128i *)(out + 7 * 16), tr2_7);
+ }
+ {+ // 00 01 02 03 04 05 06 07
+ // 10 11 12 13 14 15 16 17
+ // 20 21 22 23 24 25 26 27
+ // 30 31 32 33 34 35 36 37
+ // 40 41 42 43 44 45 46 47
+ // 50 51 52 53 54 55 56 57
+ // 60 61 62 63 64 65 66 67
+ // 70 71 72 73 74 75 76 77
+ const __m128i tr0_0 = _mm_unpacklo_epi16(res08, res09);
+ const __m128i tr0_1 = _mm_unpacklo_epi16(res10, res11);
+ const __m128i tr0_2 = _mm_unpackhi_epi16(res08, res09);
+ const __m128i tr0_3 = _mm_unpackhi_epi16(res10, res11);
+ const __m128i tr0_4 = _mm_unpacklo_epi16(res12, res13);
+ const __m128i tr0_5 = _mm_unpacklo_epi16(res14, res15);
+ const __m128i tr0_6 = _mm_unpackhi_epi16(res12, res13);
+ const __m128i tr0_7 = _mm_unpackhi_epi16(res14, res15);
+ // 00 10 01 11 02 12 03 13
+ // 20 30 21 31 22 32 23 33
+ // 04 14 05 15 06 16 07 17
+ // 24 34 25 35 26 36 27 37
+ // 40 50 41 51 42 52 43 53
+ // 60 70 61 71 62 72 63 73
+ // 54 54 55 55 56 56 57 57
+ // 64 74 65 75 66 76 67 77
+ const __m128i tr1_0 = _mm_unpacklo_epi32(tr0_0, tr0_1);
+ const __m128i tr1_1 = _mm_unpacklo_epi32(tr0_2, tr0_3);
+ const __m128i tr1_2 = _mm_unpackhi_epi32(tr0_0, tr0_1);
+ const __m128i tr1_3 = _mm_unpackhi_epi32(tr0_2, tr0_3);
+ const __m128i tr1_4 = _mm_unpacklo_epi32(tr0_4, tr0_5);
+ const __m128i tr1_5 = _mm_unpacklo_epi32(tr0_6, tr0_7);
+ const __m128i tr1_6 = _mm_unpackhi_epi32(tr0_4, tr0_5);
+ const __m128i tr1_7 = _mm_unpackhi_epi32(tr0_6, tr0_7);
+ // 00 10 20 30 01 11 21 31
+ // 40 50 60 70 41 51 61 71
+ // 02 12 22 32 03 13 23 33
+ // 42 52 62 72 43 53 63 73
+ // 04 14 24 34 05 15 21 36
+ // 44 54 64 74 45 55 61 76
+ // 06 16 26 36 07 17 27 37
+ // 46 56 66 76 47 57 67 77
+ const __m128i tr2_0 = _mm_unpacklo_epi64(tr1_0, tr1_4);
+ const __m128i tr2_1 = _mm_unpackhi_epi64(tr1_0, tr1_4);
+ const __m128i tr2_2 = _mm_unpacklo_epi64(tr1_2, tr1_6);
+ const __m128i tr2_3 = _mm_unpackhi_epi64(tr1_2, tr1_6);
+ const __m128i tr2_4 = _mm_unpacklo_epi64(tr1_1, tr1_5);
+ const __m128i tr2_5 = _mm_unpackhi_epi64(tr1_1, tr1_5);
+ const __m128i tr2_6 = _mm_unpacklo_epi64(tr1_3, tr1_7);
+ const __m128i tr2_7 = _mm_unpackhi_epi64(tr1_3, tr1_7);
+ // 00 10 20 30 40 50 60 70
+ // 01 11 21 31 41 51 61 71
+ // 02 12 22 32 42 52 62 72
+ // 03 13 23 33 43 53 63 73
+ // 04 14 24 34 44 54 64 74
+ // 05 15 25 35 45 55 65 75
+ // 06 16 26 36 46 56 66 76
+ // 07 17 27 37 47 57 67 77
+ // Store results
+ _mm_storeu_si128 ((__m128i *)(out + 8 + 0 * 16), tr2_0);
+ _mm_storeu_si128 ((__m128i *)(out + 8 + 1 * 16), tr2_1);
+ _mm_storeu_si128 ((__m128i *)(out + 8 + 2 * 16), tr2_2);
+ _mm_storeu_si128 ((__m128i *)(out + 8 + 3 * 16), tr2_3);
+ _mm_storeu_si128 ((__m128i *)(out + 8 + 4 * 16), tr2_4);
+ _mm_storeu_si128 ((__m128i *)(out + 8 + 5 * 16), tr2_5);
+ _mm_storeu_si128 ((__m128i *)(out + 8 + 6 * 16), tr2_6);
+ _mm_storeu_si128 ((__m128i *)(out + 8 + 7 * 16), tr2_7);
+ }
+ out += 8*16;
+ }
+ // Setup in/out for next pass.
+ in = intermediate;
+ out = output;
+ }
+}
--
⑨