Commit 11ca81f8 authored by Yunqing Wang's avatar Yunqing Wang
Browse files

Add vp9_idct4_1d_sse2

Added SSE2 idct4_1d which is called by vp9_short_iht4x4. Also,
modified the parameter type passed to vp9_short_iht functions to
make it work with rtcd prototype.

Change-Id: I81ba7cb4db6738f1923383b52a06deb760923ffe
parent 2e055322
......@@ -26,6 +26,7 @@
#include <math.h>
#include "./vpx_config.h"
#include "./vp9_rtcd.h"
#include "vp9/common/vp9_systemdependent.h"
#include "vp9/common/vp9_blockd.h"
#include "vp9/common/vp9_common.h"
......@@ -109,7 +110,7 @@ void vp9_dc_only_inv_walsh_add_c(int input_dc, uint8_t *pred_ptr,
}
}
static void idct4_1d(int16_t *input, int16_t *output) {
void vp9_idct4_1d_c(int16_t *input, int16_t *output) {
int16_t step[4];
int temp1, temp2;
// stage 1
......@@ -140,7 +141,7 @@ void vp9_short_idct4x4llm_c(int16_t *input, int16_t *output, int pitch) {
for (i = 0; i < 4; ++i) {
for (j = 0; j < 4; ++j)
temp_in[j] = input[j];
idct4_1d(temp_in, outptr);
vp9_idct4_1d(temp_in, outptr);
input += 4;
outptr += 4;
}
......@@ -149,7 +150,7 @@ void vp9_short_idct4x4llm_c(int16_t *input, int16_t *output, int pitch) {
for (i = 0; i < 4; ++i) {
for (j = 0; j < 4; ++j)
temp_in[j] = out[j * 4 + i];
idct4_1d(temp_in, temp_out);
vp9_idct4_1d(temp_in, temp_out);
for (j = 0; j < 4; ++j)
output[j * half_pitch + i] = ROUND_POWER_OF_TWO(temp_out[j], 4);
}
......@@ -205,7 +206,7 @@ static void idct8_1d(int16_t *input, int16_t *output) {
step1[6] = dct_const_round_shift(temp2);
// stage 2 & stage 3 - even half
idct4_1d(step1, step1);
vp9_idct4_1d(step1, step1);
// stage 2 - odd half
step2[4] = step1[4] + step1[5];
......@@ -298,24 +299,23 @@ static void iadst4_1d(int16_t *input, int16_t *output) {
output[3] = dct_const_round_shift(s3);
}
static const transform_2d IHT_4[] = {
{ idct4_1d, idct4_1d }, // DCT_DCT = 0
{ iadst4_1d, idct4_1d }, // ADST_DCT = 1
{ idct4_1d, iadst4_1d }, // DCT_ADST = 2
{ iadst4_1d, iadst4_1d } // ADST_ADST = 3
};
void vp9_short_iht4x4_c(int16_t *input, int16_t *output,
int pitch, TX_TYPE tx_type) {
int pitch, int tx_type) {
const transform_2d IHT_4[] = {
{ vp9_idct4_1d, vp9_idct4_1d }, // DCT_DCT = 0
{ iadst4_1d, vp9_idct4_1d }, // ADST_DCT = 1
{ vp9_idct4_1d, iadst4_1d }, // DCT_ADST = 2
{ iadst4_1d, iadst4_1d } // ADST_ADST = 3
};
int i, j;
int16_t out[4 * 4];
int16_t *outptr = out;
int16_t temp_in[4], temp_out[4];
const transform_2d ht = IHT_4[tx_type];
// inverse transform row vectors
for (i = 0; i < 4; ++i) {
ht.rows(input, outptr);
IHT_4[tx_type].rows(input, outptr);
input += 4;
outptr += 4;
}
......@@ -324,7 +324,7 @@ void vp9_short_iht4x4_c(int16_t *input, int16_t *output,
for (i = 0; i < 4; ++i) {
for (j = 0; j < 4; ++j)
temp_in[j] = out[j * 4 + i];
ht.cols(temp_in, temp_out);
IHT_4[tx_type].cols(temp_in, temp_out);
for (j = 0; j < 4; ++j)
output[j * pitch + i] = ROUND_POWER_OF_TWO(temp_out[j], 4);
}
......@@ -415,7 +415,7 @@ static const transform_2d IHT_8[] = {
};
void vp9_short_iht8x8_c(int16_t *input, int16_t *output,
int pitch, TX_TYPE tx_type) {
int pitch, int tx_type) {
int i, j;
int16_t out[8 * 8];
int16_t *outptr = out;
......@@ -838,7 +838,7 @@ static const transform_2d IHT_16[] = {
};
void vp9_short_iht16x16_c(int16_t *input, int16_t *output,
int pitch, TX_TYPE tx_type) {
int pitch, int tx_type) {
int i, j;
int16_t out[16 * 16];
int16_t *outptr = out;
......
......@@ -322,6 +322,9 @@ specialize vp9_short_iht4x4
prototype void vp9_short_iht16x16 "int16_t *input, int16_t *output, int pitch, int tx_type"
specialize vp9_short_iht16x16
prototype void vp9_idct4_1d "int16_t *input, int16_t *output"
specialize vp9_idct4_1d sse2
# dct and add
prototype void vp9_dc_only_idct_add "int input_dc, uint8_t *pred_ptr, uint8_t *dst_ptr, int pitch, int stride"
......
......@@ -77,10 +77,10 @@ void vp9_dc_only_idct_add_sse2(int input_dc, uint8_t *pred_ptr,
void vp9_short_idct4x4llm_sse2(int16_t *input, int16_t *output, int pitch) {
const __m128i zero = _mm_setzero_si128();
const __m128i eight = _mm_set1_epi16(8);
const __m128i cst = _mm_setr_epi16((short)cospi_16_64, (short)cospi_16_64,
(short)cospi_16_64, (short)-cospi_16_64,
(short)cospi_24_64, (short)-cospi_8_64,
(short)cospi_8_64, (short)cospi_24_64);
const __m128i cst = _mm_setr_epi16((int16_t)cospi_16_64, (int16_t)cospi_16_64,
(int16_t)cospi_16_64, (int16_t)-cospi_16_64,
(int16_t)cospi_24_64, (int16_t)-cospi_8_64,
(int16_t)cospi_8_64, (int16_t)cospi_24_64);
const __m128i rounding = _mm_set1_epi32(DCT_CONST_ROUNDING);
const int half_pitch = pitch >> 1;
__m128i input0, input1, input2, input3;
......@@ -198,4 +198,40 @@ void vp9_short_idct4x4llm_sse2(int16_t *input, int16_t *output, int pitch) {
input3 = _mm_srli_si128(input3, 8);
_mm_storel_epi64((__m128i *)(output + 2 * half_pitch), input3);
}
void vp9_idct4_1d_sse2(int16_t *input, int16_t *output) {
const __m128i zero = _mm_setzero_si128();
const __m128i c1 = _mm_setr_epi16((int16_t)cospi_16_64, (int16_t)cospi_16_64,
(int16_t)cospi_16_64, (int16_t)-cospi_16_64,
(int16_t)cospi_24_64, (int16_t)-cospi_8_64,
(int16_t)cospi_8_64, (int16_t)cospi_24_64);
const __m128i c2 = _mm_setr_epi16(1, 1, 1, 1, 1, -1, 1, -1);
const __m128i rounding = _mm_set1_epi32(DCT_CONST_ROUNDING);
__m128i in, temp;
// Load input data.
in = _mm_loadl_epi64((__m128i *)input);
// Construct i3, i1, i3, i1, i2, i0, i2, i0
in = _mm_shufflelo_epi16(in, 0xd8);
in = _mm_unpacklo_epi32(in, in);
// Stage 1
in = _mm_madd_epi16(in, c1);
in = _mm_add_epi32(in, rounding);
in = _mm_srai_epi32(in, DCT_CONST_BITS);
in = _mm_packs_epi32(in, zero);
// Stage 2
temp = _mm_shufflelo_epi16(in, 0x9c);
in = _mm_shufflelo_epi16(in, 0xc9);
in = _mm_unpacklo_epi64(temp, in);
in = _mm_madd_epi16(in, c2);
in = _mm_packs_epi32(in, zero);
// Store results
_mm_storel_epi64((__m128i *)output, in);
}
#endif
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment