libvpx: Pull from upstream

source/libvpx/vp9/common/vp9_idct.c

 /*
  *  Copyright (c) 2010 The WebM project authors. All Rights Reserved.
  *
  *  Use of this source code is governed by a BSD-style license
  *  that can be found in the LICENSE file in the root of the source
  *  tree. An additional intellectual property rights grant can be found
  *  in the file PATENTS.  All contributing project authors may
  *  be found in the AUTHORS file in the root of the source tree.
  */
 
@@ skipping 207 matching lines @@
   for (i = 0; i < 8; ++i) {
     for (j = 0; j < 8; ++j)
       temp_in[j] = out[j * 8 + i];
     idct8_1d(temp_in, temp_out);
     for (j = 0; j < 8; ++j)
       dest[j * dest_stride + i] = clip_pixel(ROUND_POWER_OF_TWO(temp_out[j], 5)
                                   + dest[j * dest_stride + i]);
   }
 }
 
+void vp9_short_idct8x8_1_add_c(int16_t *input, uint8_t *dest, int dest_stride) {
+  int i, j;
+  int a1;
+  int16_t out = dct_const_round_shift(input[0] * cospi_16_64);
+  out = dct_const_round_shift(out * cospi_16_64);
+  a1 = ROUND_POWER_OF_TWO(out, 5);
+  for (j = 0; j < 8; ++j) {
+    for (i = 0; i < 8; ++i)
+      dest[i] = clip_pixel(dest[i] + a1);
+    dest += dest_stride;
+  }
+}
+
 static void iadst4_1d(int16_t *input, int16_t *output) {
   int s0, s1, s2, s3, s4, s5, s6, s7;
 
   int x0 = input[0];
   int x1 = input[1];
   int x2 = input[2];
   int x3 = input[3];
 
   if (!(x0 | x1 | x2 | x3)) {
     output[0] = output[1] = output[2] = output[3] = 0;
@@ skipping 188 matching lines @@
   for (i = 0; i < 8; ++i) {
     for (j = 0; j < 8; ++j)
       temp_in[j] = out[j * 8 + i];
     idct8_1d(temp_in, temp_out);
     for (j = 0; j < 8; ++j)
       dest[j * dest_stride + i] = clip_pixel(ROUND_POWER_OF_TWO(temp_out[j], 5)
                                   + dest[j * dest_stride + i]);
   }
 }
 
-void vp9_short_idct1_8x8_c(int16_t *input, int16_t *output) {
-  int16_t out = dct_const_round_shift(input[0] * cospi_16_64);
-  out = dct_const_round_shift(out * cospi_16_64);
-  output[0] = ROUND_POWER_OF_TWO(out, 5);
-}
-
 static void idct16_1d(int16_t *input, int16_t *output) {
   int16_t step1[16], step2[16];
   int temp1, temp2;
 
   // stage 1
   step1[0] = input[0/2];
   step1[1] = input[16/2];
   step1[2] = input[8/2];
   step1[3] = input[24/2];
   step1[4] = input[4/2];
@@ skipping 398 matching lines @@
   for (i = 0; i < 16; ++i) {
     for (j = 0; j < 16; ++j)
       temp_in[j] = out[j*16 + i];
     idct16_1d(temp_in, temp_out);
     for (j = 0; j < 16; ++j)
       dest[j * dest_stride + i] = clip_pixel(ROUND_POWER_OF_TWO(temp_out[j], 6)
                                   + dest[j * dest_stride + i]);
   }
 }
 
-void vp9_short_idct1_16x16_c(int16_t *input, int16_t *output) {
+void vp9_short_idct16x16_1_add_c(int16_t *input, uint8_t *dest,
+                                 int dest_stride) {
+  int i, j;
+  int a1;
   int16_t out = dct_const_round_shift(input[0] * cospi_16_64);
   out = dct_const_round_shift(out * cospi_16_64);
-  output[0] = ROUND_POWER_OF_TWO(out, 6);
+  a1 = ROUND_POWER_OF_TWO(out, 6);
+  for (j = 0; j < 16; ++j) {
+    for (i = 0; i < 16; ++i)
+      dest[i] = clip_pixel(dest[i] + a1);
+    dest += dest_stride;
+  }
 }
 
 static void idct32_1d(int16_t *input, int16_t *output) {
   int16_t step1[32], step2[32];
   int temp1, temp2;
 
   // stage 1
   step1[0] = input[0];
   step1[1] = input[16];
   step1[2] = input[8];
@@ skipping 378 matching lines @@
       dest[j * dest_stride + i] = clip_pixel(ROUND_POWER_OF_TWO(temp_out[j], 6)
                                   + dest[j * dest_stride + i]);
   }
 }
 
 void vp9_short_idct1_32x32_c(int16_t *input, int16_t *output) {
   int16_t out = dct_const_round_shift(input[0] * cospi_16_64);
   out = dct_const_round_shift(out * cospi_16_64);
   output[0] = ROUND_POWER_OF_TWO(out, 6);
 }
-
-void vp9_short_idct10_32x32_add_c(int16_t *input, uint8_t *dest,
-                                  int dest_stride) {
-  int16_t out[32 * 32] = { 0 };
-  int16_t *outptr = out;
-  int i, j;
-  int16_t temp_in[32], temp_out[32];
-
-  // First transform rows. Since all non-zero dct coefficients are in
-  // upper-left 4x4 area, we only need to calculate first 4 rows here.
-  for (i = 0; i < 4; ++i) {
-    idct32_1d(input, outptr);
-    input += 32;
-    outptr += 32;
-  }
-
-  // Columns
-  for (i = 0; i < 32; ++i) {
-    for (j = 0; j < 32; ++j)
-      temp_in[j] = out[j * 32 + i];
-    idct32_1d(temp_in, temp_out);
-    for (j = 0; j < 32; ++j)
-      dest[j * dest_stride + i] = clip_pixel(ROUND_POWER_OF_TWO(temp_out[j], 6)
-                                  + dest[j * dest_stride + i]);
-  }
-}