variance_avx2: sync variance functions with c-code

add missing int64 -> uint32 cast; quiets -Wshorten-64-to-32 warnings

Change-Id: I4850b36e18dc8b399108342be4bfe0b684aefb78
(cherry picked from commit 6acd061aad8cf62000cc9117390d0c94581a8591)

aom_dsp/x86/variance_avx2.c

@@ -62,7 +62,7 @@ unsigned int aom_variance32x16_avx2(const uint8_t *src, int src_stride,
   int sum;
   variance_avx2(src, src_stride, ref, ref_stride, 32, 16, sse, &sum,
                 aom_get32x32var_avx2, 32);
-  return *sse - (((int64_t)sum * sum) >> 9);
+  return *sse - (uint32_t)(((int64_t)sum * sum) >> 9);
 }
 
 unsigned int aom_variance32x32_avx2(const uint8_t *src, int src_stride,
@@ -71,7 +71,7 @@ unsigned int aom_variance32x32_avx2(const uint8_t *src, int src_stride,
   int sum;
   variance_avx2(src, src_stride, ref, ref_stride, 32, 32, sse, &sum,
                 aom_get32x32var_avx2, 32);
-  return *sse - (((int64_t)sum * sum) >> 10);
+  return *sse - (uint32_t)(((int64_t)sum * sum) >> 10);
 }
 
 unsigned int aom_variance64x64_avx2(const uint8_t *src, int src_stride,
@@ -80,7 +80,7 @@ unsigned int aom_variance64x64_avx2(const uint8_t *src, int src_stride,
   int sum;
   variance_avx2(src, src_stride, ref, ref_stride, 64, 64, sse, &sum,
                 aom_get32x32var_avx2, 32);
-  return *sse - (((int64_t)sum * sum) >> 12);
+  return *sse - (uint32_t)(((int64_t)sum * sum) >> 12);
 }
 
 unsigned int aom_variance64x32_avx2(const uint8_t *src, int src_stride,
@@ -89,7 +89,7 @@ unsigned int aom_variance64x32_avx2(const uint8_t *src, int src_stride,
   int sum;
   variance_avx2(src, src_stride, ref, ref_stride, 64, 32, sse, &sum,
                 aom_get32x32var_avx2, 32);
-  return *sse - (((int64_t)sum * sum) >> 11);
+  return *sse - (uint32_t)(((int64_t)sum * sum) >> 11);
 }
 
 unsigned int aom_sub_pixel_variance32xh_avx2(const uint8_t *src, int src_stride,
@@ -116,7 +116,7 @@ unsigned int aom_sub_pixel_variance64x64_avx2(const uint8_t *src,
                                       dst + 32, dst_stride, 64, &sse2);
   const int se = se1 + se2;
   *sse = sse1 + sse2;
-  return *sse - (((int64_t)se * se) >> 12);
+  return *sse - (uint32_t)(((int64_t)se * se) >> 12);
 }
 
 unsigned int aom_sub_pixel_variance32x32_avx2(const uint8_t *src,
@@ -126,7 +126,7 @@ unsigned int aom_sub_pixel_variance32x32_avx2(const uint8_t *src,
                                               unsigned int *sse) {
   const int se = aom_sub_pixel_variance32xh_avx2(
       src, src_stride, x_offset, y_offset, dst, dst_stride, 32, sse);
-  return *sse - (((int64_t)se * se) >> 10);
+  return *sse - (uint32_t)(((int64_t)se * se) >> 10);
 }
 
 unsigned int aom_sub_pixel_avg_variance64x64_avx2(
@@ -143,7 +143,7 @@ unsigned int aom_sub_pixel_avg_variance64x64_avx2(
 
   *sse = sse1 + sse2;
 
-  return *sse - (((int64_t)se * se) >> 12);
+  return *sse - (uint32_t)(((int64_t)se * se) >> 12);
 }
 
 unsigned int aom_sub_pixel_avg_variance32x32_avx2(
@@ -152,5 +152,5 @@ unsigned int aom_sub_pixel_avg_variance32x32_avx2(
   // Process 32 elements in parallel.
   const int se = aom_sub_pixel_avg_variance32xh_avx2(
       src, src_stride, x_offset, y_offset, dst, dst_stride, sec, 32, 32, sse);
-  return *sse - (((int64_t)se * se) >> 10);
+  return *sse - (uint32_t)(((int64_t)se * se) >> 10);
 }