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 1499 matching lines @@
   }
 }
 
 void vp9_highbd_idct4(const tran_low_t *input, tran_low_t *output, int bd) {
   tran_low_t step[4];
   tran_high_t temp1, temp2;
   (void) bd;
   // stage 1
   temp1 = (input[0] + input[2]) * cospi_16_64;
   temp2 = (input[0] - input[2]) * cospi_16_64;
-  step[0] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step[1] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step[0] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step[1] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = input[1] * cospi_24_64 - input[3] * cospi_8_64;
   temp2 = input[1] * cospi_8_64 + input[3] * cospi_24_64;
-  step[2] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step[3] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step[2] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step[3] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   // stage 2
   output[0] = WRAPLOW(step[0] + step[3], bd);
   output[1] = WRAPLOW(step[1] + step[2], bd);
   output[2] = WRAPLOW(step[1] - step[2], bd);
   output[3] = WRAPLOW(step[0] - step[3], bd);
 }
 
 void vp9_highbd_idct4x4_16_add_c(const tran_low_t *input, uint8_t *dest8,
                                  int stride, int bd) {
@@ skipping 19 matching lines @@
       dest[j * stride + i] = highbd_clip_pixel_add(
           dest[j * stride + i], ROUND_POWER_OF_TWO(temp_out[j], 4), bd);
     }
   }
 }
 
 void vp9_highbd_idct4x4_1_add_c(const tran_low_t *input, uint8_t *dest8,
                                 int dest_stride, int bd) {
   int i;
   tran_high_t a1;
-  tran_low_t out = WRAPLOW(dct_const_round_shift(input[0] * cospi_16_64), bd);
+  tran_low_t out = WRAPLOW(
+      highbd_dct_const_round_shift(input[0] * cospi_16_64, bd), bd);
   uint16_t *dest = CONVERT_TO_SHORTPTR(dest8);
 
-  out = WRAPLOW(dct_const_round_shift(out * cospi_16_64), bd);
+  out = WRAPLOW(highbd_dct_const_round_shift(out * cospi_16_64, bd), bd);
   a1 = ROUND_POWER_OF_TWO(out, 4);
 
   for (i = 0; i < 4; i++) {
     dest[0] = highbd_clip_pixel_add(dest[0], a1, bd);
     dest[1] = highbd_clip_pixel_add(dest[1], a1, bd);
     dest[2] = highbd_clip_pixel_add(dest[2], a1, bd);
     dest[3] = highbd_clip_pixel_add(dest[3], a1, bd);
     dest += dest_stride;
   }
 }
 
 void vp9_highbd_idct8(const tran_low_t *input, tran_low_t *output, int bd) {
   tran_low_t step1[8], step2[8];
   tran_high_t temp1, temp2;
   // stage 1
   step1[0] = input[0];
   step1[2] = input[4];
   step1[1] = input[2];
   step1[3] = input[6];
   temp1 = input[1] * cospi_28_64 - input[7] * cospi_4_64;
   temp2 = input[1] * cospi_4_64 + input[7] * cospi_28_64;
-  step1[4] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[7] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[4] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[7] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = input[5] * cospi_12_64 - input[3] * cospi_20_64;
   temp2 = input[5] * cospi_20_64 + input[3] * cospi_12_64;
-  step1[5] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[6] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[5] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[6] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   // stage 2 & stage 3 - even half
   vp9_highbd_idct4(step1, step1, bd);
 
   // stage 2 - odd half
   step2[4] = WRAPLOW(step1[4] + step1[5], bd);
   step2[5] = WRAPLOW(step1[4] - step1[5], bd);
   step2[6] = WRAPLOW(-step1[6] + step1[7], bd);
   step2[7] = WRAPLOW(step1[6] + step1[7], bd);
 
   // stage 3 - odd half
   step1[4] = step2[4];
   temp1 = (step2[6] - step2[5]) * cospi_16_64;
   temp2 = (step2[5] + step2[6]) * cospi_16_64;
-  step1[5] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[6] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[5] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[6] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   step1[7] = step2[7];
 
   // stage 4
   output[0] = WRAPLOW(step1[0] + step1[7], bd);
   output[1] = WRAPLOW(step1[1] + step1[6], bd);
   output[2] = WRAPLOW(step1[2] + step1[5], bd);
   output[3] = WRAPLOW(step1[3] + step1[4], bd);
   output[4] = WRAPLOW(step1[3] - step1[4], bd);
   output[5] = WRAPLOW(step1[2] - step1[5], bd);
   output[6] = WRAPLOW(step1[1] - step1[6], bd);
@@ skipping 24 matching lines @@
       dest[j * stride + i] = highbd_clip_pixel_add(
           dest[j * stride + i], ROUND_POWER_OF_TWO(temp_out[j], 5), bd);
     }
   }
 }
 
 void vp9_highbd_idct8x8_1_add_c(const tran_low_t *input, uint8_t *dest8,
                                 int stride, int bd) {
   int i, j;
   tran_high_t a1;
-  tran_low_t out = WRAPLOW(dct_const_round_shift(input[0] * cospi_16_64), bd);
+  tran_low_t out = WRAPLOW(
+      highbd_dct_const_round_shift(input[0] * cospi_16_64, bd), bd);
   uint16_t *dest = CONVERT_TO_SHORTPTR(dest8);
-  out = WRAPLOW(dct_const_round_shift(out * cospi_16_64), bd);
+  out = WRAPLOW(highbd_dct_const_round_shift(out * cospi_16_64, bd), bd);
   a1 = ROUND_POWER_OF_TWO(out, 5);
   for (j = 0; j < 8; ++j) {
     for (i = 0; i < 8; ++i)
       dest[i] = highbd_clip_pixel_add(dest[i], a1, bd);
     dest += stride;
   }
 }
 
 static void highbd_iadst4(const tran_low_t *input, tran_low_t *output, int bd) {
   tran_high_t s0, s1, s2, s3, s4, s5, s6, s7;
@@ skipping 20 matching lines @@
 
   s0 = s0 + s3 + s5;
   s1 = s1 - s4 - s6;
   s3 = s2;
   s2 = sinpi_3_9 * s7;
 
   // 1-D transform scaling factor is sqrt(2).
   // The overall dynamic range is 14b (input) + 14b (multiplication scaling)
   // + 1b (addition) = 29b.
   // Hence the output bit depth is 15b.
-  output[0] = WRAPLOW(dct_const_round_shift(s0 + s3), bd);
-  output[1] = WRAPLOW(dct_const_round_shift(s1 + s3), bd);
-  output[2] = WRAPLOW(dct_const_round_shift(s2), bd);
-  output[3] = WRAPLOW(dct_const_round_shift(s0 + s1 - s3), bd);
+  output[0] = WRAPLOW(highbd_dct_const_round_shift(s0 + s3, bd), bd);
+  output[1] = WRAPLOW(highbd_dct_const_round_shift(s1 + s3, bd), bd);
+  output[2] = WRAPLOW(highbd_dct_const_round_shift(s2, bd), bd);
+  output[3] = WRAPLOW(highbd_dct_const_round_shift(s0 + s1 - s3, bd), bd);
 }
 
 void vp9_highbd_iht4x4_16_add_c(const tran_low_t *input, uint8_t *dest8,
                                 int stride, int tx_type, int bd) {
   const highbd_transform_2d IHT_4[] = {
     { vp9_highbd_idct4, vp9_highbd_idct4  },    // DCT_DCT  = 0
     { highbd_iadst4, vp9_highbd_idct4 },    // ADST_DCT = 1
     { vp9_highbd_idct4, highbd_iadst4 },    // DCT_ADST = 2
     { highbd_iadst4, highbd_iadst4 }    // ADST_ADST = 3
   };
@@ skipping 44 matching lines @@
   // stage 1
   s0 = cospi_2_64  * x0 + cospi_30_64 * x1;
   s1 = cospi_30_64 * x0 - cospi_2_64  * x1;
   s2 = cospi_10_64 * x2 + cospi_22_64 * x3;
   s3 = cospi_22_64 * x2 - cospi_10_64 * x3;
   s4 = cospi_18_64 * x4 + cospi_14_64 * x5;
   s5 = cospi_14_64 * x4 - cospi_18_64 * x5;
   s6 = cospi_26_64 * x6 + cospi_6_64  * x7;
   s7 = cospi_6_64  * x6 - cospi_26_64 * x7;
 
-  x0 = WRAPLOW(dct_const_round_shift(s0 + s4), bd);
-  x1 = WRAPLOW(dct_const_round_shift(s1 + s5), bd);
-  x2 = WRAPLOW(dct_const_round_shift(s2 + s6), bd);
-  x3 = WRAPLOW(dct_const_round_shift(s3 + s7), bd);
-  x4 = WRAPLOW(dct_const_round_shift(s0 - s4), bd);
-  x5 = WRAPLOW(dct_const_round_shift(s1 - s5), bd);
-  x6 = WRAPLOW(dct_const_round_shift(s2 - s6), bd);
-  x7 = WRAPLOW(dct_const_round_shift(s3 - s7), bd);
+  x0 = WRAPLOW(highbd_dct_const_round_shift(s0 + s4, bd), bd);
+  x1 = WRAPLOW(highbd_dct_const_round_shift(s1 + s5, bd), bd);
+  x2 = WRAPLOW(highbd_dct_const_round_shift(s2 + s6, bd), bd);
+  x3 = WRAPLOW(highbd_dct_const_round_shift(s3 + s7, bd), bd);
+  x4 = WRAPLOW(highbd_dct_const_round_shift(s0 - s4, bd), bd);
+  x5 = WRAPLOW(highbd_dct_const_round_shift(s1 - s5, bd), bd);
+  x6 = WRAPLOW(highbd_dct_const_round_shift(s2 - s6, bd), bd);
+  x7 = WRAPLOW(highbd_dct_const_round_shift(s3 - s7, bd), bd);
 
   // stage 2
   s0 = x0;
   s1 = x1;
   s2 = x2;
   s3 = x3;
   s4 =  cospi_8_64  * x4 + cospi_24_64 * x5;
   s5 =  cospi_24_64 * x4 - cospi_8_64  * x5;
   s6 = -cospi_24_64 * x6 + cospi_8_64  * x7;
   s7 =  cospi_8_64  * x6 + cospi_24_64 * x7;
 
   x0 = WRAPLOW(s0 + s2, bd);
   x1 = WRAPLOW(s1 + s3, bd);
   x2 = WRAPLOW(s0 - s2, bd);
   x3 = WRAPLOW(s1 - s3, bd);
-  x4 = WRAPLOW(dct_const_round_shift(s4 + s6), bd);
-  x5 = WRAPLOW(dct_const_round_shift(s5 + s7), bd);
-  x6 = WRAPLOW(dct_const_round_shift(s4 - s6), bd);
-  x7 = WRAPLOW(dct_const_round_shift(s5 - s7), bd);
+  x4 = WRAPLOW(highbd_dct_const_round_shift(s4 + s6, bd), bd);
+  x5 = WRAPLOW(highbd_dct_const_round_shift(s5 + s7, bd), bd);
+  x6 = WRAPLOW(highbd_dct_const_round_shift(s4 - s6, bd), bd);
+  x7 = WRAPLOW(highbd_dct_const_round_shift(s5 - s7, bd), bd);
 
   // stage 3
   s2 = cospi_16_64 * (x2 + x3);
   s3 = cospi_16_64 * (x2 - x3);
   s6 = cospi_16_64 * (x6 + x7);
   s7 = cospi_16_64 * (x6 - x7);
 
-  x2 = WRAPLOW(dct_const_round_shift(s2), bd);
-  x3 = WRAPLOW(dct_const_round_shift(s3), bd);
-  x6 = WRAPLOW(dct_const_round_shift(s6), bd);
-  x7 = WRAPLOW(dct_const_round_shift(s7), bd);
+  x2 = WRAPLOW(highbd_dct_const_round_shift(s2, bd), bd);
+  x3 = WRAPLOW(highbd_dct_const_round_shift(s3, bd), bd);
+  x6 = WRAPLOW(highbd_dct_const_round_shift(s6, bd), bd);
+  x7 = WRAPLOW(highbd_dct_const_round_shift(s7, bd), bd);
 
   output[0] = WRAPLOW(x0, bd);
   output[1] = WRAPLOW(-x4, bd);
   output[2] = WRAPLOW(x6, bd);
   output[3] = WRAPLOW(-x2, bd);
   output[4] = WRAPLOW(x3, bd);
   output[5] = WRAPLOW(-x7, bd);
   output[6] = WRAPLOW(x5, bd);
   output[7] = WRAPLOW(-x1, bd);
 }
@@ skipping 88 matching lines @@
   step2[1] = step1[1];
   step2[2] = step1[2];
   step2[3] = step1[3];
   step2[4] = step1[4];
   step2[5] = step1[5];
   step2[6] = step1[6];
   step2[7] = step1[7];
 
   temp1 = step1[8] * cospi_30_64 - step1[15] * cospi_2_64;
   temp2 = step1[8] * cospi_2_64 + step1[15] * cospi_30_64;
-  step2[8] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[15] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[8] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[15] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   temp1 = step1[9] * cospi_14_64 - step1[14] * cospi_18_64;
   temp2 = step1[9] * cospi_18_64 + step1[14] * cospi_14_64;
-  step2[9] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[14] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[9] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[14] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   temp1 = step1[10] * cospi_22_64 - step1[13] * cospi_10_64;
   temp2 = step1[10] * cospi_10_64 + step1[13] * cospi_22_64;
-  step2[10] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[13] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[10] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[13] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   temp1 = step1[11] * cospi_6_64 - step1[12] * cospi_26_64;
   temp2 = step1[11] * cospi_26_64 + step1[12] * cospi_6_64;
-  step2[11] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[12] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[11] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[12] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   // stage 3
   step1[0] = step2[0];
   step1[1] = step2[1];
   step1[2] = step2[2];
   step1[3] = step2[3];
 
   temp1 = step2[4] * cospi_28_64 - step2[7] * cospi_4_64;
   temp2 = step2[4] * cospi_4_64 + step2[7] * cospi_28_64;
-  step1[4] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[7] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[4] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[7] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = step2[5] * cospi_12_64 - step2[6] * cospi_20_64;
   temp2 = step2[5] * cospi_20_64 + step2[6] * cospi_12_64;
-  step1[5] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[6] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[5] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[6] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   step1[8] = WRAPLOW(step2[8] + step2[9], bd);
   step1[9] = WRAPLOW(step2[8] - step2[9], bd);
   step1[10] = WRAPLOW(-step2[10] + step2[11], bd);
   step1[11] = WRAPLOW(step2[10] + step2[11], bd);
   step1[12] = WRAPLOW(step2[12] + step2[13], bd);
   step1[13] = WRAPLOW(step2[12] - step2[13], bd);
   step1[14] = WRAPLOW(-step2[14] + step2[15], bd);
   step1[15] = WRAPLOW(step2[14] + step2[15], bd);
 
   // stage 4
   temp1 = (step1[0] + step1[1]) * cospi_16_64;
   temp2 = (step1[0] - step1[1]) * cospi_16_64;
-  step2[0] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[1] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[0] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[1] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = step1[2] * cospi_24_64 - step1[3] * cospi_8_64;
   temp2 = step1[2] * cospi_8_64 + step1[3] * cospi_24_64;
-  step2[2] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[3] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[2] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[3] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   step2[4] = WRAPLOW(step1[4] + step1[5], bd);
   step2[5] = WRAPLOW(step1[4] - step1[5], bd);
   step2[6] = WRAPLOW(-step1[6] + step1[7], bd);
   step2[7] = WRAPLOW(step1[6] + step1[7], bd);
 
   step2[8] = step1[8];
   step2[15] = step1[15];
   temp1 = -step1[9] * cospi_8_64 + step1[14] * cospi_24_64;
   temp2 = step1[9] * cospi_24_64 + step1[14] * cospi_8_64;
-  step2[9] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[14] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[9] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[14] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = -step1[10] * cospi_24_64 - step1[13] * cospi_8_64;
   temp2 = -step1[10] * cospi_8_64 + step1[13] * cospi_24_64;
-  step2[10] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[13] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[10] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[13] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   step2[11] = step1[11];
   step2[12] = step1[12];
 
   // stage 5
   step1[0] = WRAPLOW(step2[0] + step2[3], bd);
   step1[1] = WRAPLOW(step2[1] + step2[2], bd);
   step1[2] = WRAPLOW(step2[1] - step2[2], bd);
   step1[3] = WRAPLOW(step2[0] - step2[3], bd);
   step1[4] = step2[4];
   temp1 = (step2[6] - step2[5]) * cospi_16_64;
   temp2 = (step2[5] + step2[6]) * cospi_16_64;
-  step1[5] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[6] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[5] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[6] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   step1[7] = step2[7];
 
   step1[8] = WRAPLOW(step2[8] + step2[11], bd);
   step1[9] = WRAPLOW(step2[9] + step2[10], bd);
   step1[10] = WRAPLOW(step2[9] - step2[10], bd);
   step1[11] = WRAPLOW(step2[8] - step2[11], bd);
   step1[12] = WRAPLOW(-step2[12] + step2[15], bd);
   step1[13] = WRAPLOW(-step2[13] + step2[14], bd);
   step1[14] = WRAPLOW(step2[13] + step2[14], bd);
   step1[15] = WRAPLOW(step2[12] + step2[15], bd);
 
   // stage 6
   step2[0] = WRAPLOW(step1[0] + step1[7], bd);
   step2[1] = WRAPLOW(step1[1] + step1[6], bd);
   step2[2] = WRAPLOW(step1[2] + step1[5], bd);
   step2[3] = WRAPLOW(step1[3] + step1[4], bd);
   step2[4] = WRAPLOW(step1[3] - step1[4], bd);
   step2[5] = WRAPLOW(step1[2] - step1[5], bd);
   step2[6] = WRAPLOW(step1[1] - step1[6], bd);
   step2[7] = WRAPLOW(step1[0] - step1[7], bd);
   step2[8] = step1[8];
   step2[9] = step1[9];
   temp1 = (-step1[10] + step1[13]) * cospi_16_64;
   temp2 = (step1[10] + step1[13]) * cospi_16_64;
-  step2[10] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[13] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[10] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[13] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = (-step1[11] + step1[12]) * cospi_16_64;
   temp2 = (step1[11] + step1[12]) * cospi_16_64;
-  step2[11] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[12] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[11] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[12] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   step2[14] = step1[14];
   step2[15] = step1[15];
 
   // stage 7
   output[0] = WRAPLOW(step2[0] + step2[15], bd);
   output[1] = WRAPLOW(step2[1] + step2[14], bd);
   output[2] = WRAPLOW(step2[2] + step2[13], bd);
   output[3] = WRAPLOW(step2[3] + step2[12], bd);
   output[4] = WRAPLOW(step2[4] + step2[11], bd);
   output[5] = WRAPLOW(step2[5] + step2[10], bd);
@@ skipping 76 matching lines @@
   s7 = x6 * cospi_19_64 - x7 * cospi_13_64;
   s8 = x8 * cospi_17_64 + x9 * cospi_15_64;
   s9 = x8 * cospi_15_64 - x9 * cospi_17_64;
   s10 = x10 * cospi_21_64 + x11 * cospi_11_64;
   s11 = x10 * cospi_11_64 - x11 * cospi_21_64;
   s12 = x12 * cospi_25_64 + x13 * cospi_7_64;
   s13 = x12 * cospi_7_64  - x13 * cospi_25_64;
   s14 = x14 * cospi_29_64 + x15 * cospi_3_64;
   s15 = x14 * cospi_3_64  - x15 * cospi_29_64;
 
-  x0 = WRAPLOW(dct_const_round_shift(s0 + s8), bd);
-  x1 = WRAPLOW(dct_const_round_shift(s1 + s9), bd);
-  x2 = WRAPLOW(dct_const_round_shift(s2 + s10), bd);
-  x3 = WRAPLOW(dct_const_round_shift(s3 + s11), bd);
-  x4 = WRAPLOW(dct_const_round_shift(s4 + s12), bd);
-  x5 = WRAPLOW(dct_const_round_shift(s5 + s13), bd);
-  x6 = WRAPLOW(dct_const_round_shift(s6 + s14), bd);
-  x7 = WRAPLOW(dct_const_round_shift(s7 + s15), bd);
-  x8  = WRAPLOW(dct_const_round_shift(s0 - s8), bd);
-  x9  = WRAPLOW(dct_const_round_shift(s1 - s9), bd);
-  x10 = WRAPLOW(dct_const_round_shift(s2 - s10), bd);
-  x11 = WRAPLOW(dct_const_round_shift(s3 - s11), bd);
-  x12 = WRAPLOW(dct_const_round_shift(s4 - s12), bd);
-  x13 = WRAPLOW(dct_const_round_shift(s5 - s13), bd);
-  x14 = WRAPLOW(dct_const_round_shift(s6 - s14), bd);
-  x15 = WRAPLOW(dct_const_round_shift(s7 - s15), bd);
+  x0 = WRAPLOW(highbd_dct_const_round_shift(s0 + s8, bd), bd);
+  x1 = WRAPLOW(highbd_dct_const_round_shift(s1 + s9, bd), bd);
+  x2 = WRAPLOW(highbd_dct_const_round_shift(s2 + s10, bd), bd);
+  x3 = WRAPLOW(highbd_dct_const_round_shift(s3 + s11, bd), bd);
+  x4 = WRAPLOW(highbd_dct_const_round_shift(s4 + s12, bd), bd);
+  x5 = WRAPLOW(highbd_dct_const_round_shift(s5 + s13, bd), bd);
+  x6 = WRAPLOW(highbd_dct_const_round_shift(s6 + s14, bd), bd);
+  x7 = WRAPLOW(highbd_dct_const_round_shift(s7 + s15, bd), bd);
+  x8  = WRAPLOW(highbd_dct_const_round_shift(s0 - s8, bd), bd);
+  x9  = WRAPLOW(highbd_dct_const_round_shift(s1 - s9, bd), bd);
+  x10 = WRAPLOW(highbd_dct_const_round_shift(s2 - s10, bd), bd);
+  x11 = WRAPLOW(highbd_dct_const_round_shift(s3 - s11, bd), bd);
+  x12 = WRAPLOW(highbd_dct_const_round_shift(s4 - s12, bd), bd);
+  x13 = WRAPLOW(highbd_dct_const_round_shift(s5 - s13, bd), bd);
+  x14 = WRAPLOW(highbd_dct_const_round_shift(s6 - s14, bd), bd);
+  x15 = WRAPLOW(highbd_dct_const_round_shift(s7 - s15, bd), bd);
 
   // stage 2
   s0 = x0;
   s1 = x1;
   s2 = x2;
   s3 = x3;
   s4 = x4;
   s5 = x5;
   s6 = x6;
   s7 = x7;
   s8 = x8 * cospi_4_64 + x9 * cospi_28_64;
   s9 = x8 * cospi_28_64 - x9 * cospi_4_64;
   s10 = x10 * cospi_20_64 + x11 * cospi_12_64;
   s11 = x10 * cospi_12_64 - x11 * cospi_20_64;
   s12 = -x12 * cospi_28_64 + x13 * cospi_4_64;
   s13 = x12 * cospi_4_64 + x13 * cospi_28_64;
   s14 = -x14 * cospi_12_64 + x15 * cospi_20_64;
   s15 = x14 * cospi_20_64 + x15 * cospi_12_64;
 
   x0 = WRAPLOW(s0 + s4, bd);
   x1 = WRAPLOW(s1 + s5, bd);
   x2 = WRAPLOW(s2 + s6, bd);
   x3 = WRAPLOW(s3 + s7, bd);
   x4 = WRAPLOW(s0 - s4, bd);
   x5 = WRAPLOW(s1 - s5, bd);
   x6 = WRAPLOW(s2 - s6, bd);
   x7 = WRAPLOW(s3 - s7, bd);
-  x8 = WRAPLOW(dct_const_round_shift(s8 + s12), bd);
-  x9 = WRAPLOW(dct_const_round_shift(s9 + s13), bd);
-  x10 = WRAPLOW(dct_const_round_shift(s10 + s14), bd);
-  x11 = WRAPLOW(dct_const_round_shift(s11 + s15), bd);
-  x12 = WRAPLOW(dct_const_round_shift(s8 - s12), bd);
-  x13 = WRAPLOW(dct_const_round_shift(s9 - s13), bd);
-  x14 = WRAPLOW(dct_const_round_shift(s10 - s14), bd);
-  x15 = WRAPLOW(dct_const_round_shift(s11 - s15), bd);
+  x8 = WRAPLOW(highbd_dct_const_round_shift(s8 + s12, bd), bd);
+  x9 = WRAPLOW(highbd_dct_const_round_shift(s9 + s13, bd), bd);
+  x10 = WRAPLOW(highbd_dct_const_round_shift(s10 + s14, bd), bd);
+  x11 = WRAPLOW(highbd_dct_const_round_shift(s11 + s15, bd), bd);
+  x12 = WRAPLOW(highbd_dct_const_round_shift(s8 - s12, bd), bd);
+  x13 = WRAPLOW(highbd_dct_const_round_shift(s9 - s13, bd), bd);
+  x14 = WRAPLOW(highbd_dct_const_round_shift(s10 - s14, bd), bd);
+  x15 = WRAPLOW(highbd_dct_const_round_shift(s11 - s15, bd), bd);
 
   // stage 3
   s0 = x0;
   s1 = x1;
   s2 = x2;
   s3 = x3;
   s4 = x4 * cospi_8_64 + x5 * cospi_24_64;
   s5 = x4 * cospi_24_64 - x5 * cospi_8_64;
   s6 = -x6 * cospi_24_64 + x7 * cospi_8_64;
   s7 = x6 * cospi_8_64 + x7 * cospi_24_64;
   s8 = x8;
   s9 = x9;
   s10 = x10;
   s11 = x11;
   s12 = x12 * cospi_8_64 + x13 * cospi_24_64;
   s13 = x12 * cospi_24_64 - x13 * cospi_8_64;
   s14 = -x14 * cospi_24_64 + x15 * cospi_8_64;
   s15 = x14 * cospi_8_64 + x15 * cospi_24_64;
 
   x0 = WRAPLOW(s0 + s2, bd);
   x1 = WRAPLOW(s1 + s3, bd);
   x2 = WRAPLOW(s0 - s2, bd);
   x3 = WRAPLOW(s1 - s3, bd);
-  x4 = WRAPLOW(dct_const_round_shift(s4 + s6), bd);
-  x5 = WRAPLOW(dct_const_round_shift(s5 + s7), bd);
-  x6 = WRAPLOW(dct_const_round_shift(s4 - s6), bd);
-  x7 = WRAPLOW(dct_const_round_shift(s5 - s7), bd);
+  x4 = WRAPLOW(highbd_dct_const_round_shift(s4 + s6, bd), bd);
+  x5 = WRAPLOW(highbd_dct_const_round_shift(s5 + s7, bd), bd);
+  x6 = WRAPLOW(highbd_dct_const_round_shift(s4 - s6, bd), bd);
+  x7 = WRAPLOW(highbd_dct_const_round_shift(s5 - s7, bd), bd);
   x8 = WRAPLOW(s8 + s10, bd);
   x9 = WRAPLOW(s9 + s11, bd);
   x10 = WRAPLOW(s8 - s10, bd);
   x11 = WRAPLOW(s9 - s11, bd);
-  x12 = WRAPLOW(dct_const_round_shift(s12 + s14), bd);
-  x13 = WRAPLOW(dct_const_round_shift(s13 + s15), bd);
-  x14 = WRAPLOW(dct_const_round_shift(s12 - s14), bd);
-  x15 = WRAPLOW(dct_const_round_shift(s13 - s15), bd);
+  x12 = WRAPLOW(highbd_dct_const_round_shift(s12 + s14, bd), bd);
+  x13 = WRAPLOW(highbd_dct_const_round_shift(s13 + s15, bd), bd);
+  x14 = WRAPLOW(highbd_dct_const_round_shift(s12 - s14, bd), bd);
+  x15 = WRAPLOW(highbd_dct_const_round_shift(s13 - s15, bd), bd);
 
   // stage 4
   s2 = (- cospi_16_64) * (x2 + x3);
   s3 = cospi_16_64 * (x2 - x3);
   s6 = cospi_16_64 * (x6 + x7);
   s7 = cospi_16_64 * (-x6 + x7);
   s10 = cospi_16_64 * (x10 + x11);
   s11 = cospi_16_64 * (-x10 + x11);
   s14 = (- cospi_16_64) * (x14 + x15);
   s15 = cospi_16_64 * (x14 - x15);
 
-  x2 = WRAPLOW(dct_const_round_shift(s2), bd);
-  x3 = WRAPLOW(dct_const_round_shift(s3), bd);
-  x6 = WRAPLOW(dct_const_round_shift(s6), bd);
-  x7 = WRAPLOW(dct_const_round_shift(s7), bd);
-  x10 = WRAPLOW(dct_const_round_shift(s10), bd);
-  x11 = WRAPLOW(dct_const_round_shift(s11), bd);
-  x14 = WRAPLOW(dct_const_round_shift(s14), bd);
-  x15 = WRAPLOW(dct_const_round_shift(s15), bd);
+  x2 = WRAPLOW(highbd_dct_const_round_shift(s2, bd), bd);
+  x3 = WRAPLOW(highbd_dct_const_round_shift(s3, bd), bd);
+  x6 = WRAPLOW(highbd_dct_const_round_shift(s6, bd), bd);
+  x7 = WRAPLOW(highbd_dct_const_round_shift(s7, bd), bd);
+  x10 = WRAPLOW(highbd_dct_const_round_shift(s10, bd), bd);
+  x11 = WRAPLOW(highbd_dct_const_round_shift(s11, bd), bd);
+  x14 = WRAPLOW(highbd_dct_const_round_shift(s14, bd), bd);
+  x15 = WRAPLOW(highbd_dct_const_round_shift(s15, bd), bd);
 
   output[0] = WRAPLOW(x0, bd);
   output[1] = WRAPLOW(-x8, bd);
   output[2] = WRAPLOW(x12, bd);
   output[3] = WRAPLOW(-x4, bd);
   output[4] = WRAPLOW(x6, bd);
   output[5] = WRAPLOW(x14, bd);
   output[6] = WRAPLOW(x10, bd);
   output[7] = WRAPLOW(x2, bd);
   output[8] = WRAPLOW(x3, bd);
@@ skipping 66 matching lines @@
       dest[j * stride + i] = highbd_clip_pixel_add(
           dest[j * stride + i], ROUND_POWER_OF_TWO(temp_out[j], 6), bd);
     }
   }
 }
 
 void vp9_highbd_idct16x16_1_add_c(const tran_low_t *input, uint8_t *dest8,
                                   int stride, int bd) {
   int i, j;
   tran_high_t a1;
-  tran_low_t out = WRAPLOW(dct_const_round_shift(input[0] * cospi_16_64), bd);
+  tran_low_t out = WRAPLOW(
+      highbd_dct_const_round_shift(input[0] * cospi_16_64, bd), bd);
   uint16_t *dest = CONVERT_TO_SHORTPTR(dest8);
 
-  out = WRAPLOW(dct_const_round_shift(out * cospi_16_64), bd);
+  out = WRAPLOW(highbd_dct_const_round_shift(out * cospi_16_64, bd), bd);
   a1 = ROUND_POWER_OF_TWO(out, 6);
   for (j = 0; j < 16; ++j) {
     for (i = 0; i < 16; ++i)
       dest[i] = highbd_clip_pixel_add(dest[i], a1, bd);
     dest += stride;
   }
 }
 
 static void highbd_idct32(const tran_low_t *input, tran_low_t *output, int bd) {
   tran_low_t step1[32], step2[32];
@@ skipping 13 matching lines @@
   step1[9] = input[18];
   step1[10] = input[10];
   step1[11] = input[26];
   step1[12] = input[6];
   step1[13] = input[22];
   step1[14] = input[14];
   step1[15] = input[30];
 
   temp1 = input[1] * cospi_31_64 - input[31] * cospi_1_64;
   temp2 = input[1] * cospi_1_64 + input[31] * cospi_31_64;
-  step1[16] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[31] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[16] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[31] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   temp1 = input[17] * cospi_15_64 - input[15] * cospi_17_64;
   temp2 = input[17] * cospi_17_64 + input[15] * cospi_15_64;
-  step1[17] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[30] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[17] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[30] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   temp1 = input[9] * cospi_23_64 - input[23] * cospi_9_64;
   temp2 = input[9] * cospi_9_64 + input[23] * cospi_23_64;
-  step1[18] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[29] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[18] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[29] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   temp1 = input[25] * cospi_7_64 - input[7] * cospi_25_64;
   temp2 = input[25] * cospi_25_64 + input[7] * cospi_7_64;
-  step1[19] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[28] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[19] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[28] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   temp1 = input[5] * cospi_27_64 - input[27] * cospi_5_64;
   temp2 = input[5] * cospi_5_64 + input[27] * cospi_27_64;
-  step1[20] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[27] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[20] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[27] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   temp1 = input[21] * cospi_11_64 - input[11] * cospi_21_64;
   temp2 = input[21] * cospi_21_64 + input[11] * cospi_11_64;
-  step1[21] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[26] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[21] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[26] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   temp1 = input[13] * cospi_19_64 - input[19] * cospi_13_64;
   temp2 = input[13] * cospi_13_64 + input[19] * cospi_19_64;
-  step1[22] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[25] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[22] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[25] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   temp1 = input[29] * cospi_3_64 - input[3] * cospi_29_64;
   temp2 = input[29] * cospi_29_64 + input[3] * cospi_3_64;
-  step1[23] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[24] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[23] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[24] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   // stage 2
   step2[0] = step1[0];
   step2[1] = step1[1];
   step2[2] = step1[2];
   step2[3] = step1[3];
   step2[4] = step1[4];
   step2[5] = step1[5];
   step2[6] = step1[6];
   step2[7] = step1[7];
 
   temp1 = step1[8] * cospi_30_64 - step1[15] * cospi_2_64;
   temp2 = step1[8] * cospi_2_64 + step1[15] * cospi_30_64;
-  step2[8] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[15] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[8] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[15] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   temp1 = step1[9] * cospi_14_64 - step1[14] * cospi_18_64;
   temp2 = step1[9] * cospi_18_64 + step1[14] * cospi_14_64;
-  step2[9] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[14] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[9] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[14] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   temp1 = step1[10] * cospi_22_64 - step1[13] * cospi_10_64;
   temp2 = step1[10] * cospi_10_64 + step1[13] * cospi_22_64;
-  step2[10] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[13] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[10] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[13] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   temp1 = step1[11] * cospi_6_64 - step1[12] * cospi_26_64;
   temp2 = step1[11] * cospi_26_64 + step1[12] * cospi_6_64;
-  step2[11] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[12] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[11] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[12] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   step2[16] = WRAPLOW(step1[16] + step1[17], bd);
   step2[17] = WRAPLOW(step1[16] - step1[17], bd);
   step2[18] = WRAPLOW(-step1[18] + step1[19], bd);
   step2[19] = WRAPLOW(step1[18] + step1[19], bd);
   step2[20] = WRAPLOW(step1[20] + step1[21], bd);
   step2[21] = WRAPLOW(step1[20] - step1[21], bd);
   step2[22] = WRAPLOW(-step1[22] + step1[23], bd);
   step2[23] = WRAPLOW(step1[22] + step1[23], bd);
   step2[24] = WRAPLOW(step1[24] + step1[25], bd);
   step2[25] = WRAPLOW(step1[24] - step1[25], bd);
   step2[26] = WRAPLOW(-step1[26] + step1[27], bd);
   step2[27] = WRAPLOW(step1[26] + step1[27], bd);
   step2[28] = WRAPLOW(step1[28] + step1[29], bd);
   step2[29] = WRAPLOW(step1[28] - step1[29], bd);
   step2[30] = WRAPLOW(-step1[30] + step1[31], bd);
   step2[31] = WRAPLOW(step1[30] + step1[31], bd);
 
   // stage 3
   step1[0] = step2[0];
   step1[1] = step2[1];
   step1[2] = step2[2];
   step1[3] = step2[3];
 
   temp1 = step2[4] * cospi_28_64 - step2[7] * cospi_4_64;
   temp2 = step2[4] * cospi_4_64 + step2[7] * cospi_28_64;
-  step1[4] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[7] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[4] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[7] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = step2[5] * cospi_12_64 - step2[6] * cospi_20_64;
   temp2 = step2[5] * cospi_20_64 + step2[6] * cospi_12_64;
-  step1[5] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[6] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[5] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[6] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
 
   step1[8] = WRAPLOW(step2[8] + step2[9], bd);
   step1[9] = WRAPLOW(step2[8] - step2[9], bd);
   step1[10] = WRAPLOW(-step2[10] + step2[11], bd);
   step1[11] = WRAPLOW(step2[10] + step2[11], bd);
   step1[12] = WRAPLOW(step2[12] + step2[13], bd);
   step1[13] = WRAPLOW(step2[12] - step2[13], bd);
   step1[14] = WRAPLOW(-step2[14] + step2[15], bd);
   step1[15] = WRAPLOW(step2[14] + step2[15], bd);
 
   step1[16] = step2[16];
   step1[31] = step2[31];
   temp1 = -step2[17] * cospi_4_64 + step2[30] * cospi_28_64;
   temp2 = step2[17] * cospi_28_64 + step2[30] * cospi_4_64;
-  step1[17] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[30] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[17] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[30] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = -step2[18] * cospi_28_64 - step2[29] * cospi_4_64;
   temp2 = -step2[18] * cospi_4_64 + step2[29] * cospi_28_64;
-  step1[18] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[29] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[18] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[29] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   step1[19] = step2[19];
   step1[20] = step2[20];
   temp1 = -step2[21] * cospi_20_64 + step2[26] * cospi_12_64;
   temp2 = step2[21] * cospi_12_64 + step2[26] * cospi_20_64;
-  step1[21] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[26] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[21] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[26] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = -step2[22] * cospi_12_64 - step2[25] * cospi_20_64;
   temp2 = -step2[22] * cospi_20_64 + step2[25] * cospi_12_64;
-  step1[22] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[25] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[22] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[25] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   step1[23] = step2[23];
   step1[24] = step2[24];
   step1[27] = step2[27];
   step1[28] = step2[28];
 
   // stage 4
   temp1 = (step1[0] + step1[1]) * cospi_16_64;
   temp2 = (step1[0] - step1[1]) * cospi_16_64;
-  step2[0] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[1] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[0] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[1] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = step1[2] * cospi_24_64 - step1[3] * cospi_8_64;
   temp2 = step1[2] * cospi_8_64 + step1[3] * cospi_24_64;
-  step2[2] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[3] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[2] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[3] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   step2[4] = WRAPLOW(step1[4] + step1[5], bd);
   step2[5] = WRAPLOW(step1[4] - step1[5], bd);
   step2[6] = WRAPLOW(-step1[6] + step1[7], bd);
   step2[7] = WRAPLOW(step1[6] + step1[7], bd);
 
   step2[8] = step1[8];
   step2[15] = step1[15];
   temp1 = -step1[9] * cospi_8_64 + step1[14] * cospi_24_64;
   temp2 = step1[9] * cospi_24_64 + step1[14] * cospi_8_64;
-  step2[9] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[14] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[9] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[14] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = -step1[10] * cospi_24_64 - step1[13] * cospi_8_64;
   temp2 = -step1[10] * cospi_8_64 + step1[13] * cospi_24_64;
-  step2[10] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[13] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[10] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[13] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   step2[11] = step1[11];
   step2[12] = step1[12];
 
   step2[16] = WRAPLOW(step1[16] + step1[19], bd);
   step2[17] = WRAPLOW(step1[17] + step1[18], bd);
   step2[18] = WRAPLOW(step1[17] - step1[18], bd);
   step2[19] = WRAPLOW(step1[16] - step1[19], bd);
   step2[20] = WRAPLOW(-step1[20] + step1[23], bd);
   step2[21] = WRAPLOW(-step1[21] + step1[22], bd);
   step2[22] = WRAPLOW(step1[21] + step1[22], bd);
   step2[23] = WRAPLOW(step1[20] + step1[23], bd);
 
   step2[24] = WRAPLOW(step1[24] + step1[27], bd);
   step2[25] = WRAPLOW(step1[25] + step1[26], bd);
   step2[26] = WRAPLOW(step1[25] - step1[26], bd);
   step2[27] = WRAPLOW(step1[24] - step1[27], bd);
   step2[28] = WRAPLOW(-step1[28] + step1[31], bd);
   step2[29] = WRAPLOW(-step1[29] + step1[30], bd);
   step2[30] = WRAPLOW(step1[29] + step1[30], bd);
   step2[31] = WRAPLOW(step1[28] + step1[31], bd);
 
   // stage 5
   step1[0] = WRAPLOW(step2[0] + step2[3], bd);
   step1[1] = WRAPLOW(step2[1] + step2[2], bd);
   step1[2] = WRAPLOW(step2[1] - step2[2], bd);
   step1[3] = WRAPLOW(step2[0] - step2[3], bd);
   step1[4] = step2[4];
   temp1 = (step2[6] - step2[5]) * cospi_16_64;
   temp2 = (step2[5] + step2[6]) * cospi_16_64;
-  step1[5] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[6] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[5] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[6] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   step1[7] = step2[7];
 
   step1[8] = WRAPLOW(step2[8] + step2[11], bd);
   step1[9] = WRAPLOW(step2[9] + step2[10], bd);
   step1[10] = WRAPLOW(step2[9] - step2[10], bd);
   step1[11] = WRAPLOW(step2[8] - step2[11], bd);
   step1[12] = WRAPLOW(-step2[12] + step2[15], bd);
   step1[13] = WRAPLOW(-step2[13] + step2[14], bd);
   step1[14] = WRAPLOW(step2[13] + step2[14], bd);
   step1[15] = WRAPLOW(step2[12] + step2[15], bd);
 
   step1[16] = step2[16];
   step1[17] = step2[17];
   temp1 = -step2[18] * cospi_8_64 + step2[29] * cospi_24_64;
   temp2 = step2[18] * cospi_24_64 + step2[29] * cospi_8_64;
-  step1[18] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[29] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[18] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[29] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = -step2[19] * cospi_8_64 + step2[28] * cospi_24_64;
   temp2 = step2[19] * cospi_24_64 + step2[28] * cospi_8_64;
-  step1[19] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[28] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[19] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[28] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = -step2[20] * cospi_24_64 - step2[27] * cospi_8_64;
   temp2 = -step2[20] * cospi_8_64 + step2[27] * cospi_24_64;
-  step1[20] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[27] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[20] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[27] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = -step2[21] * cospi_24_64 - step2[26] * cospi_8_64;
   temp2 = -step2[21] * cospi_8_64 + step2[26] * cospi_24_64;
-  step1[21] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[26] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[21] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[26] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   step1[22] = step2[22];
   step1[23] = step2[23];
   step1[24] = step2[24];
   step1[25] = step2[25];
   step1[30] = step2[30];
   step1[31] = step2[31];
 
   // stage 6
   step2[0] = WRAPLOW(step1[0] + step1[7], bd);
   step2[1] = WRAPLOW(step1[1] + step1[6], bd);
   step2[2] = WRAPLOW(step1[2] + step1[5], bd);
   step2[3] = WRAPLOW(step1[3] + step1[4], bd);
   step2[4] = WRAPLOW(step1[3] - step1[4], bd);
   step2[5] = WRAPLOW(step1[2] - step1[5], bd);
   step2[6] = WRAPLOW(step1[1] - step1[6], bd);
   step2[7] = WRAPLOW(step1[0] - step1[7], bd);
   step2[8] = step1[8];
   step2[9] = step1[9];
   temp1 = (-step1[10] + step1[13]) * cospi_16_64;
   temp2 = (step1[10] + step1[13]) * cospi_16_64;
-  step2[10] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[13] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[10] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[13] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = (-step1[11] + step1[12]) * cospi_16_64;
   temp2 = (step1[11] + step1[12]) * cospi_16_64;
-  step2[11] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step2[12] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step2[11] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step2[12] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   step2[14] = step1[14];
   step2[15] = step1[15];
 
   step2[16] = WRAPLOW(step1[16] + step1[23], bd);
   step2[17] = WRAPLOW(step1[17] + step1[22], bd);
   step2[18] = WRAPLOW(step1[18] + step1[21], bd);
   step2[19] = WRAPLOW(step1[19] + step1[20], bd);
   step2[20] = WRAPLOW(step1[19] - step1[20], bd);
   step2[21] = WRAPLOW(step1[18] - step1[21], bd);
   step2[22] = WRAPLOW(step1[17] - step1[22], bd);
@@ skipping 25 matching lines @@
   step1[13] = WRAPLOW(step2[2] - step2[13], bd);
   step1[14] = WRAPLOW(step2[1] - step2[14], bd);
   step1[15] = WRAPLOW(step2[0] - step2[15], bd);
 
   step1[16] = step2[16];
   step1[17] = step2[17];
   step1[18] = step2[18];
   step1[19] = step2[19];
   temp1 = (-step2[20] + step2[27]) * cospi_16_64;
   temp2 = (step2[20] + step2[27]) * cospi_16_64;
-  step1[20] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[27] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[20] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[27] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = (-step2[21] + step2[26]) * cospi_16_64;
   temp2 = (step2[21] + step2[26]) * cospi_16_64;
-  step1[21] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[26] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[21] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[26] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = (-step2[22] + step2[25]) * cospi_16_64;
   temp2 = (step2[22] + step2[25]) * cospi_16_64;
-  step1[22] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[25] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[22] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[25] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   temp1 = (-step2[23] + step2[24]) * cospi_16_64;
   temp2 = (step2[23] + step2[24]) * cospi_16_64;
-  step1[23] = WRAPLOW(dct_const_round_shift(temp1), bd);
-  step1[24] = WRAPLOW(dct_const_round_shift(temp2), bd);
+  step1[23] = WRAPLOW(highbd_dct_const_round_shift(temp1, bd), bd);
+  step1[24] = WRAPLOW(highbd_dct_const_round_shift(temp2, bd), bd);
   step1[28] = step2[28];
   step1[29] = step2[29];
   step1[30] = step2[30];
   step1[31] = step2[31];
 
   // final stage
   output[0] = WRAPLOW(step1[0] + step1[31], bd);
   output[1] = WRAPLOW(step1[1] + step1[30], bd);
   output[2] = WRAPLOW(step1[2] + step1[29], bd);
   output[3] = WRAPLOW(step1[3] + step1[28], bd);
@@ skipping 93 matching lines @@
     }
   }
 }
 
 void vp9_highbd_idct32x32_1_add_c(const tran_low_t *input, uint8_t *dest8,
                                   int stride, int bd) {
   int i, j;
   int a1;
   uint16_t *dest = CONVERT_TO_SHORTPTR(dest8);
 
-  tran_low_t out = WRAPLOW(dct_const_round_shift(input[0] * cospi_16_64), bd);
-  out = WRAPLOW(dct_const_round_shift(out * cospi_16_64), bd);
+  tran_low_t out = WRAPLOW(
+      highbd_dct_const_round_shift(input[0] * cospi_16_64, bd), bd);
+  out = WRAPLOW(highbd_dct_const_round_shift(out * cospi_16_64, bd), bd);
   a1 = ROUND_POWER_OF_TWO(out, 6);
 
   for (j = 0; j < 32; ++j) {
     for (i = 0; i < 32; ++i)
       dest[i] = highbd_clip_pixel_add(dest[i], a1, bd);
     dest += stride;
   }
 }
 
 // idct
@@ skipping 79 matching lines @@
 
 void vp9_highbd_iht16x16_add(TX_TYPE tx_type, const tran_low_t *input,
                            uint8_t *dest, int stride, int eob, int bd) {
   if (tx_type == DCT_DCT) {
     vp9_highbd_idct16x16_add(input, dest, stride, eob, bd);
   } else {
     vp9_highbd_iht16x16_256_add(input, dest, stride, tx_type, bd);
   }
 }
 #endif  // CONFIG_VP9_HIGHBITDEPTH