libvpx: Pull from upstream

source/libvpx/vp8/common/variance_c.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.
  */
 
+#include "./vp8_rtcd.h"
+#include "filter.h"
+#include "variance.h"
 
-#include "variance.h"
-#include "filter.h"
-
-
-unsigned int vp8_get_mb_ss_c
-(
-    const short *src_ptr
-)
-{
-    unsigned int i = 0, sum = 0;
-
-    do
-    {
-        sum += (src_ptr[i] * src_ptr[i]);
-        i++;
-    }
-    while (i < 256);
-
-    return sum;
+/* This is a bad idea.
+ * ctz = count trailing zeros */
+static int ctz(int a) {
+  int b = 0;
+  while (a != 1) {
+    a >>= 1;
+    b++;
+  }
+  return b;
 }
 
-
-static void variance(
+static unsigned int variance(
     const unsigned char *src_ptr,
     int  source_stride,
     const unsigned char *ref_ptr,
     int  recon_stride,
     int  w,
     int  h,
-    unsigned int *sse,
-    int *sum)
+    unsigned int *sse)
 {
     int i, j;
-    int diff;
+    int diff, sum;
 
-    *sum = 0;
+    sum = 0;
     *sse = 0;
 
     for (i = 0; i < h; i++)
     {
         for (j = 0; j < w; j++)
         {
             diff = src_ptr[j] - ref_ptr[j];
-            *sum += diff;
+            sum += diff;
             *sse += diff * diff;
         }
 
         src_ptr += source_stride;
         ref_ptr += recon_stride;
     }
+
+    return (*sse - (((unsigned int)sum * sum) >> (int)((ctz(w) + ctz(h)))));
 }
 
-
-unsigned int vp8_variance16x16_c(
-    const unsigned char *src_ptr,
-    int  source_stride,
-    const unsigned char *ref_ptr,
-    int  recon_stride,
-    unsigned int *sse)
-{
-    unsigned int var;
-    int avg;
-
-
-    variance(src_ptr, source_stride, ref_ptr, recon_stride, 16, 16, &var, &avg);
-    *sse = var;
-    return (var - (((unsigned int)avg * avg) >> 8));
-}
-
-unsigned int vp8_variance8x16_c(
-    const unsigned char *src_ptr,
-    int  source_stride,
-    const unsigned char *ref_ptr,
-    int  recon_stride,
-    unsigned int *sse)
-{
-    unsigned int var;
-    int avg;
-
-
-    variance(src_ptr, source_stride, ref_ptr, recon_stride, 8, 16, &var, &avg);
-    *sse = var;
-    return (var - (((unsigned int)avg * avg) >> 7));
-}
-
-unsigned int vp8_variance16x8_c(
-    const unsigned char *src_ptr,
-    int  source_stride,
-    const unsigned char *ref_ptr,
-    int  recon_stride,
-    unsigned int *sse)
-{
-    unsigned int var;
-    int avg;
-
-
-    variance(src_ptr, source_stride, ref_ptr, recon_stride, 16, 8, &var, &avg);
-    *sse = var;
-    return (var - (((unsigned int)avg * avg) >> 7));
-}
-
-
-unsigned int vp8_variance8x8_c(
-    const unsigned char *src_ptr,
-    int  source_stride,
-    const unsigned char *ref_ptr,
-    int  recon_stride,
-    unsigned int *sse)
-{
-    unsigned int var;
-    int avg;
-
-
-    variance(src_ptr, source_stride, ref_ptr, recon_stride, 8, 8, &var, &avg);
-    *sse = var;
-    return (var - (((unsigned int)avg * avg) >> 6));
-}
-
-unsigned int vp8_variance4x4_c(
-    const unsigned char *src_ptr,
-    int  source_stride,
-    const unsigned char *ref_ptr,
-    int  recon_stride,
-    unsigned int *sse)
-{
-    unsigned int var;
-    int avg;
-
-
-    variance(src_ptr, source_stride, ref_ptr, recon_stride, 4, 4, &var, &avg);
-    *sse = var;
-    return (var - (((unsigned int)avg * avg) >> 4));
-}
-
-
-unsigned int vp8_mse16x16_c(
-    const unsigned char *src_ptr,
-    int  source_stride,
-    const unsigned char *ref_ptr,
-    int  recon_stride,
-    unsigned int *sse)
-{
-    unsigned int var;
-    int avg;
-
-    variance(src_ptr, source_stride, ref_ptr, recon_stride, 16, 16, &var, &avg);
-    *sse = var;
-    return var;
-}
-
-
 /****************************************************************************
  *
  *  ROUTINE       : filter_block2d_bil_first_pass
  *
  *  INPUTS        : UINT8  *src_ptr          : Pointer to source block.
  *                  UINT32 src_pixels_per_line : Stride of input block.
  *                  UINT32 pixel_step        : Offset between filter input samples (see notes).
  *                  UINT32 output_height     : Input block height.
  *                  UINT32 output_width      : Input block width.
  *                  INT32  *vp8_filter          : Array of 2 bi-linear filter taps.
@@ skipping 123 matching lines @@
 
     HFilter = vp8_bilinear_filters[xoffset];
     VFilter = vp8_bilinear_filters[yoffset];
 
     /* First filter 1d Horizontal */
     var_filter_block2d_bil_first_pass(src_ptr, FData3, src_pixels_per_line, 1, 5, 4, HFilter);
 
     /* Now filter Verticaly */
     var_filter_block2d_bil_second_pass(FData3, temp2, 4,  4,  4,  4, VFilter);
 
-    return vp8_variance4x4_c(temp2, 4, dst_ptr, dst_pixels_per_line, sse);
+    return variance(temp2, 4, dst_ptr, dst_pixels_per_line, 4, 4, sse);
 }
 
 
 unsigned int vp8_sub_pixel_variance8x8_c
 (
     const unsigned char  *src_ptr,
     int  src_pixels_per_line,
     int  xoffset,
     int  yoffset,
     const unsigned char *dst_ptr,
     int dst_pixels_per_line,
     unsigned int *sse
 )
 {
     unsigned short FData3[9*8]; /* Temp data bufffer used in filtering */
     unsigned char  temp2[20*16];
     const short *HFilter, *VFilter;
 
     HFilter = vp8_bilinear_filters[xoffset];
     VFilter = vp8_bilinear_filters[yoffset];
 
     var_filter_block2d_bil_first_pass(src_ptr, FData3, src_pixels_per_line, 1, 9, 8, HFilter);
     var_filter_block2d_bil_second_pass(FData3, temp2, 8, 8, 8, 8, VFilter);
 
-    return vp8_variance8x8_c(temp2, 8, dst_ptr, dst_pixels_per_line, sse);
+    return variance(temp2, 8, dst_ptr, dst_pixels_per_line, 8, 8, sse);
 }
 
 unsigned int vp8_sub_pixel_variance16x16_c
 (
     const unsigned char  *src_ptr,
     int  src_pixels_per_line,
     int  xoffset,
     int  yoffset,
     const unsigned char *dst_ptr,
     int dst_pixels_per_line,
     unsigned int *sse
 )
 {
     unsigned short FData3[17*16];   /* Temp data bufffer used in filtering */
     unsigned char  temp2[20*16];
     const short *HFilter, *VFilter;
 
     HFilter = vp8_bilinear_filters[xoffset];
     VFilter = vp8_bilinear_filters[yoffset];
 
     var_filter_block2d_bil_first_pass(src_ptr, FData3, src_pixels_per_line, 1, 17, 16, HFilter);
     var_filter_block2d_bil_second_pass(FData3, temp2, 16, 16, 16, 16, VFilter);
 
-    return vp8_variance16x16_c(temp2, 16, dst_ptr, dst_pixels_per_line, sse);
+    return variance(temp2, 16, dst_ptr, dst_pixels_per_line, 16, 16, sse);
 }
 
 
 unsigned int vp8_variance_halfpixvar16x16_h_c(
     const unsigned char *src_ptr,
     int  source_stride,
     const unsigned char *ref_ptr,
     int  recon_stride,
     unsigned int *sse)
 {
@@ skipping 55 matching lines @@
     unsigned short FData3[16*9];    /* Temp data bufffer used in filtering */
     unsigned char  temp2[20*16];
     const short *HFilter, *VFilter;
 
     HFilter = vp8_bilinear_filters[xoffset];
     VFilter = vp8_bilinear_filters[yoffset];
 
     var_filter_block2d_bil_first_pass(src_ptr, FData3, src_pixels_per_line, 1, 9, 16, HFilter);
     var_filter_block2d_bil_second_pass(FData3, temp2, 16, 16, 8, 16, VFilter);
 
-    return vp8_variance16x8_c(temp2, 16, dst_ptr, dst_pixels_per_line, sse);
+    return variance(temp2, 16, dst_ptr, dst_pixels_per_line, 16, 8, sse);
 }
 
 unsigned int vp8_sub_pixel_variance8x16_c
 (
     const unsigned char  *src_ptr,
     int  src_pixels_per_line,
     int  xoffset,
     int  yoffset,
     const unsigned char *dst_ptr,
     int dst_pixels_per_line,
     unsigned int *sse
 )
 {
     unsigned short FData3[9*16];    /* Temp data bufffer used in filtering */
     unsigned char  temp2[20*16];
     const short *HFilter, *VFilter;
 
 
     HFilter = vp8_bilinear_filters[xoffset];
     VFilter = vp8_bilinear_filters[yoffset];
 
 
     var_filter_block2d_bil_first_pass(src_ptr, FData3, src_pixels_per_line, 1, 17, 8, HFilter);
     var_filter_block2d_bil_second_pass(FData3, temp2, 8, 8, 16, 8, VFilter);
 
-    return vp8_variance8x16_c(temp2, 8, dst_ptr, dst_pixels_per_line, sse);
+    return variance(temp2, 8, dst_ptr, dst_pixels_per_line, 8, 16, sse);
 }