From 499fe05dc030b31d15c2ad9717e6bcee839f7d12 Mon Sep 17 00:00:00 2001
From: Yaowu Xu <yaowu@google.com>
Date: Fri, 22 Feb 2013 16:18:08 -0800
Subject: [PATCH] optimize forward 16x16 DCT for accuracy

This commit added pre/post scaling for first half of fDCT16x16 to
reduce error, by simulation of 100,000 blocks for random inputs,
the average sse reduced from 2.1/block to 0.0498/block.

also enabled tests for 16x16 fDCT and iDCT

Change-Id: Id2a95f0464c6dd4118797d456237ae90274c0f02
---
 test/dct16x16_test.cc |   2 +-
 test/test.mk          |   3 +-
 vp9/encoder/vp9_dct.c | 440 +-----------------------------------------
 3 files changed, 5 insertions(+), 440 deletions(-)

diff --git a/test/dct16x16_test.cc b/test/dct16x16_test.cc
index 194d5fb74..f6d2d5994 100644
--- a/test/dct16x16_test.cc
+++ b/test/dct16x16_test.cc
@@ -289,7 +289,7 @@ TEST(VP9Idct16x16Test, AccuracyCheck) {
     }
   }
 }
-#if 0
+#if 1
 // we need enable fdct test once we re-do the 16 point fdct.
 TEST(VP9Fdct16x16Test, AccuracyCheck) {
   ACMRandom rnd(ACMRandom::DeterministicSeed());
diff --git a/test/test.mk b/test/test.mk
index 29657625c..d879f9fb2 100644
--- a/test/test.mk
+++ b/test/test.mk
@@ -72,8 +72,9 @@ endif
 
 LIBVPX_TEST_SRCS-$(CONFIG_VP9)         += convolve_test.cc
 LIBVPX_TEST_SRCS-$(CONFIG_VP9_ENCODER) += fdct4x4_test.cc
+
 LIBVPX_TEST_SRCS-$(CONFIG_VP9_ENCODER) += fdct8x8_test.cc
-#LIBVPX_TEST_SRCS-$(CONFIG_VP9_ENCODER) += dct16x16_test.cc
+LIBVPX_TEST_SRCS-$(CONFIG_VP9_ENCODER) += dct16x16_test.cc
 LIBVPX_TEST_SRCS-$(CONFIG_VP9_ENCODER) += variance_test.cc
 #LIBVPX_TEST_SRCS-$(CONFIG_VP9_ENCODER) += dct32x32_test.cc
 
diff --git a/vp9/encoder/vp9_dct.c b/vp9/encoder/vp9_dct.c
index a459e949b..6894c5fee 100644
--- a/vp9/encoder/vp9_dct.c
+++ b/vp9/encoder/vp9_dct.c
@@ -918,441 +918,7 @@ void vp9_short_walsh8x4_x8_c(short *input, short *output, int pitch) {
   vp9_short_walsh4x4_x8_c(input + 4, output + 16, pitch);
 }
 
-#define TEST_INT_16x16_DCT 1
-#if !TEST_INT_16x16_DCT
 
-static void dct16x16_1d(double input[16], double output[16]) {
-  static const double C1 = 0.995184726672197;
-  static const double C2 = 0.98078528040323;
-  static const double C3 = 0.956940335732209;
-  static const double C4 = 0.923879532511287;
-  static const double C5 = 0.881921264348355;
-  static const double C6 = 0.831469612302545;
-  static const double C7 = 0.773010453362737;
-  static const double C8 = 0.707106781186548;
-  static const double C9 = 0.634393284163646;
-  static const double C10 = 0.555570233019602;
-  static const double C11 = 0.471396736825998;
-  static const double C12 = 0.38268343236509;
-  static const double C13 = 0.290284677254462;
-  static const double C14 = 0.195090322016128;
-  static const double C15 = 0.098017140329561;
-
-  vp9_clear_system_state(); // Make it simd safe : __asm emms;
-  {
-    double step[16];
-    double intermediate[16];
-    double 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];
-
-    // step 2
-    output[0] = step[0] + step[7];
-    output[1] = step[1] + step[6];
-    output[2] = step[2] + step[5];
-    output[3] = step[3] + step[4];
-    output[4] = step[3] - step[4];
-    output[5] = step[2] - step[5];
-    output[6] = step[1] - step[6];
-    output[7] = step[0] - step[7];
-
-    temp1 = step[ 8]*C7;
-    temp2 = step[15]*C9;
-    output[ 8] = temp1 + temp2;
-
-    temp1 = step[ 9]*C11;
-    temp2 = step[14]*C5;
-    output[ 9] = temp1 - temp2;
-
-    temp1 = step[10]*C3;
-    temp2 = step[13]*C13;
-    output[10] = temp1 + temp2;
-
-    temp1 = step[11]*C15;
-    temp2 = step[12]*C1;
-    output[11] = temp1 - temp2;
-
-    temp1 = step[11]*C1;
-    temp2 = step[12]*C15;
-    output[12] = temp2 + temp1;
-
-    temp1 = step[10]*C13;
-    temp2 = step[13]*C3;
-    output[13] = temp2 - temp1;
-
-    temp1 = step[ 9]*C5;
-    temp2 = step[14]*C11;
-    output[14] = temp2 + temp1;
-
-    temp1 = step[ 8]*C9;
-    temp2 = step[15]*C7;
-    output[15] = temp2 - temp1;
-
-    // step 3
-    step[ 0] = output[0] + output[3];
-    step[ 1] = output[1] + output[2];
-    step[ 2] = output[1] - output[2];
-    step[ 3] = output[0] - output[3];
-
-    temp1 = output[4]*C14;
-    temp2 = output[7]*C2;
-    step[ 4] = temp1 + temp2;
-
-    temp1 = output[5]*C10;
-    temp2 = output[6]*C6;
-    step[ 5] = temp1 + temp2;
-
-    temp1 = output[5]*C6;
-    temp2 = output[6]*C10;
-    step[ 6] = temp2 - temp1;
-
-    temp1 = output[4]*C2;
-    temp2 = output[7]*C14;
-    step[ 7] = temp2 - temp1;
-
-    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];
-
-    // step 4
-    output[ 0] = (step[ 0] + step[ 1]);
-    output[ 8] = (step[ 0] - step[ 1]);
-
-    temp1 = step[2]*C12;
-    temp2 = step[3]*C4;
-    temp1 = temp1 + temp2;
-    output[ 4] = 2*(temp1*C8);
-
-    temp1 = step[2]*C4;
-    temp2 = step[3]*C12;
-    temp1 = temp2 - temp1;
-    output[12] = 2*(temp1*C8);
-
-    output[ 2] = 2*((step[4] + step[ 5])*C8);
-    output[14] = 2*((step[7] - step[ 6])*C8);
-
-    temp1 = step[4] - step[5];
-    temp2 = step[6] + step[7];
-    output[ 6] = (temp1 + temp2);
-    output[10] = (temp1 - temp2);
-
-    intermediate[8] = step[8] + step[14];
-    intermediate[9] = step[9] + step[15];
-
-    temp1 = intermediate[8]*C12;
-    temp2 = intermediate[9]*C4;
-    temp1 = temp1 - temp2;
-    output[3] = 2*(temp1*C8);
-
-    temp1 = intermediate[8]*C4;
-    temp2 = intermediate[9]*C12;
-    temp1 = temp2 + temp1;
-    output[13] = 2*(temp1*C8);
-
-    output[ 9] = 2*((step[10] + step[11])*C8);
-
-    intermediate[11] = step[10] - step[11];
-    intermediate[12] = step[12] + step[13];
-    intermediate[13] = step[12] - step[13];
-    intermediate[14] = step[ 8] - step[14];
-    intermediate[15] = step[ 9] - step[15];
-
-    output[15] = (intermediate[11] + intermediate[12]);
-    output[ 1] = -(intermediate[11] - intermediate[12]);
-
-    output[ 7] = 2*(intermediate[13]*C8);
-
-    temp1 = intermediate[14]*C12;
-    temp2 = intermediate[15]*C4;
-    temp1 = temp1 - temp2;
-    output[11] = -2*(temp1*C8);
-
-    temp1 = intermediate[14]*C4;
-    temp2 = intermediate[15]*C12;
-    temp1 = temp2 + temp1;
-    output[ 5] = 2*(temp1*C8);
-  }
-  vp9_clear_system_state(); // Make it simd safe : __asm emms;
-}
-
-void vp9_short_fdct16x16_c(short *input, short *out, int pitch) {
-  vp9_clear_system_state(); // Make it simd safe : __asm emms;
-  {
-    int shortpitch = pitch >> 1;
-    int i, j;
-    double output[256];
-    // First transform columns
-    for (i = 0; i < 16; i++) {
-        double temp_in[16], temp_out[16];
-        for (j = 0; j < 16; j++)
-            temp_in[j] = input[j*shortpitch + i];
-        dct16x16_1d(temp_in, temp_out);
-        for (j = 0; j < 16; j++)
-            output[j*16 + i] = temp_out[j];
-    }
-    // Then transform rows
-    for (i = 0; i < 16; ++i) {
-        double temp_in[16], temp_out[16];
-        for (j = 0; j < 16; ++j)
-            temp_in[j] = output[j + i*16];
-        dct16x16_1d(temp_in, temp_out);
-        for (j = 0; j < 16; ++j)
-            output[j + i*16] = temp_out[j];
-    }
-    // Scale by some magic number
-    for (i = 0; i < 256; i++)
-        out[i] = (short)round(output[i]/2);
-  }
-  vp9_clear_system_state(); // Make it simd safe : __asm emms;
-}
-
-#else
-
-#define NEW_FDCT16 1
-#if !NEW_FDCT16
-static const int16_t C1 = 16305;
-static const int16_t C2 = 16069;
-static const int16_t C3 = 15679;
-static const int16_t C4 = 15137;
-static const int16_t C5 = 14449;
-static const int16_t C6 = 13623;
-static const int16_t C7 = 12665;
-static const int16_t C8 = 11585;
-static const int16_t C9 = 10394;
-static const int16_t C10 = 9102;
-static const int16_t C11 = 7723;
-static const int16_t C12 = 6270;
-static const int16_t C13 = 4756;
-static const int16_t C14 = 3196;
-static const int16_t C15 = 1606;
-
-#define RIGHT_SHIFT 14
-#define ROUNDING (1 << (RIGHT_SHIFT - 1))
-
-static void dct16x16_1d(int16_t input[16], int16_t output[16],
-                        int last_shift_bits) {
-    int16_t step[16];
-    int intermediate[16];
-    int temp1, temp2;
-    int final_shift = RIGHT_SHIFT;
-    int final_rounding = ROUNDING;
-    int output_shift = 0;
-    int output_rounding = 0;
-
-    final_shift += last_shift_bits;
-    if (final_shift > 0)
-    final_rounding = 1 << (final_shift - 1);
-
-    output_shift += last_shift_bits;
-    if (output_shift > 0)
-      output_rounding = 1 << (output_shift - 1);
-
-    // 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];
-
-    // step 2
-    output[0] = step[0] + step[7];
-    output[1] = step[1] + step[6];
-    output[2] = step[2] + step[5];
-    output[3] = step[3] + step[4];
-    output[4] = step[3] - step[4];
-    output[5] = step[2] - step[5];
-    output[6] = step[1] - step[6];
-    output[7] = step[0] - step[7];
-
-    temp1 = step[ 8] * C7;
-    temp2 = step[15] * C9;
-    output[ 8] = (temp1 + temp2 + ROUNDING) >> RIGHT_SHIFT;
-
-    temp1 = step[ 9] * C11;
-    temp2 = step[14] * C5;
-    output[ 9] = (temp1 - temp2 + ROUNDING) >> RIGHT_SHIFT;
-
-    temp1 = step[10] * C3;
-    temp2 = step[13] * C13;
-    output[10] = (temp1 + temp2 + ROUNDING) >> RIGHT_SHIFT;
-
-    temp1 = step[11] * C15;
-    temp2 = step[12] * C1;
-    output[11] = (temp1 - temp2 + ROUNDING) >> RIGHT_SHIFT;
-
-    temp1 = step[11] * C1;
-    temp2 = step[12] * C15;
-    output[12] = (temp2 + temp1 + ROUNDING) >> RIGHT_SHIFT;
-
-    temp1 = step[10] * C13;
-    temp2 = step[13] * C3;
-    output[13] = (temp2 - temp1 + ROUNDING) >> RIGHT_SHIFT;
-
-    temp1 = step[ 9] * C5;
-    temp2 = step[14] * C11;
-    output[14] = (temp2 + temp1 + ROUNDING) >> RIGHT_SHIFT;
-
-    temp1 = step[ 8] * C9;
-    temp2 = step[15] * C7;
-    output[15] = (temp2 - temp1 + ROUNDING) >> RIGHT_SHIFT;
-
-    // step 3
-    step[ 0] = output[0] + output[3];
-    step[ 1] = output[1] + output[2];
-    step[ 2] = output[1] - output[2];
-    step[ 3] = output[0] - output[3];
-
-    temp1 = output[4] * C14;
-    temp2 = output[7] * C2;
-    step[ 4] = (temp1 + temp2 + ROUNDING) >> RIGHT_SHIFT;
-
-    temp1 = output[5] * C10;
-    temp2 = output[6] * C6;
-    step[ 5] = (temp1 + temp2 + ROUNDING) >> RIGHT_SHIFT;
-
-    temp1 = output[5] * C6;
-    temp2 = output[6] * C10;
-    step[ 6] = (temp2 - temp1 + ROUNDING) >> RIGHT_SHIFT;
-
-    temp1 = output[4] * C2;
-    temp2 = output[7] * C14;
-    step[ 7] = (temp2 - temp1 + ROUNDING) >> RIGHT_SHIFT;
-
-    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];
-
-    // step 4
-    output[ 0] = (step[ 0] + step[ 1] + output_rounding) >> output_shift;
-    output[ 8] = (step[ 0] - step[ 1] + output_rounding) >> output_shift;
-
-    temp1 = step[2] * C12;
-    temp2 = step[3] * C4;
-    temp1 = (temp1 + temp2 + final_rounding) >> final_shift;
-    output[ 4] = (2 * (temp1 * C8) + ROUNDING) >> RIGHT_SHIFT;
-
-    temp1 = step[2] * C4;
-    temp2 = step[3] * C12;
-    temp1 = (temp2 - temp1 + final_rounding) >> final_shift;
-    output[12] = (2 * (temp1 * C8) + ROUNDING) >> RIGHT_SHIFT;
-
-    output[ 2] = (2 * ((step[4] + step[ 5]) * C8) + final_rounding)
-        >> final_shift;
-    output[14] = (2 * ((step[7] - step[ 6]) * C8) + final_rounding)
-        >> final_shift;
-
-    temp1 = step[4] - step[5];
-    temp2 = step[6] + step[7];
-    output[ 6] = (temp1 + temp2 + output_rounding) >> output_shift;
-    output[10] = (temp1 - temp2 + output_rounding) >> output_shift;
-
-    intermediate[8] = step[8] + step[14];
-    intermediate[9] = step[9] + step[15];
-
-    temp1 = intermediate[8] * C12;
-    temp2 = intermediate[9] * C4;
-    temp1 = (temp1 - temp2 + final_rounding) >> final_shift;
-    output[3] = (2 * (temp1 * C8) + ROUNDING) >> RIGHT_SHIFT;
-
-    temp1 = intermediate[8] * C4;
-    temp2 = intermediate[9] * C12;
-    temp1 = (temp2 + temp1 + final_rounding) >> final_shift;
-    output[13] = (2 * (temp1 * C8) + ROUNDING) >> RIGHT_SHIFT;
-
-    output[ 9] = (2 * ((step[10] + step[11]) * C8) + final_rounding)
-        >> final_shift;
-
-    intermediate[11] = step[10] - step[11];
-    intermediate[12] = step[12] + step[13];
-    intermediate[13] = step[12] - step[13];
-    intermediate[14] = step[ 8] - step[14];
-    intermediate[15] = step[ 9] - step[15];
-
-    output[15] = (intermediate[11] + intermediate[12] + output_rounding)
-        >> output_shift;
-    output[ 1] = (intermediate[12] - intermediate[11] + output_rounding)
-        >> output_shift;
-
-    output[ 7] = (2 * (intermediate[13] * C8) + final_rounding) >> final_shift;
-
-    temp1 = intermediate[14] * C12;
-    temp2 = intermediate[15] * C4;
-    temp1 = (temp1 - temp2 + final_rounding) >> final_shift;
-    output[11] = (-2 * (temp1 * C8) + ROUNDING) >> RIGHT_SHIFT;
-
-    temp1 = intermediate[14] * C4;
-    temp2 = intermediate[15] * C12;
-    temp1 = (temp2 + temp1 + final_rounding) >> final_shift;
-    output[ 5] = (2 * (temp1 * C8) + ROUNDING) >> RIGHT_SHIFT;
-}
-
-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 *outptr = &output[0];
-
-    // First transform columns
-    for (i = 0; i < 16; i++) {
-        int16_t temp_in[16];
-        int16_t temp_out[16];
-        for (j = 0; j < 16; j++)
-            temp_in[j] = input[j * shortpitch + i];
-        dct16x16_1d(temp_in, temp_out, 0);
-        for (j = 0; j < 16; j++)
-            output[j * 16 + i] = temp_out[j];
-    }
-
-    // Then transform rows
-    for (i = 0; i < 16; ++i) {
-        dct16x16_1d(outptr, out, 1);
-        outptr += 16;
-        out += 16;
-    }
-}
-#undef RIGHT_SHIFT
-#undef ROUNDING
-
-#else
 // Rewrote to use same algorithm as others.
 static void fdct16_1d(int16_t input[16], int16_t output[16]) {
   int16_t step[16];
@@ -1466,10 +1032,10 @@ void vp9_short_fdct16x16_c(int16_t *input, int16_t *out, int pitch) {
   // First transform columns
   for (i = 0; i < 16; i++) {
     for (j = 0; j < 16; j++)
-      temp_in[j] = input[j * shortpitch + i];
+      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];
+      output[j * 16 + i] = (temp_out[j] + 1) >> 2;
   }
 
   // Then transform rows
@@ -1481,8 +1047,6 @@ void vp9_short_fdct16x16_c(int16_t *input, int16_t *out, int pitch) {
       out[j + i * 16] = temp_out[j];
   }
 }
-#endif
-#endif
 
 #if CONFIG_INTHT16X16
 void fadst16_1d(int16_t *input, int16_t *output) {