generic-library/vpx
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

variance_avx2: sync variance functions with c-code

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

Change-Id: I4850b36e18dc8b399108342be4bfe0b684aefb78
This commit is contained in:
James Zern
2016-09-19 16:17:24 -07:00
parent 0695843a21
commit 6acd061aad
1 changed files with 8 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
vpx_dsp/x86/variance_avx2.c
View File

@@ -61,7 +61,7 @@ unsigned int vpx_variance32x16_avx2(const uint8_t *src, int src_stride,
int sum;
variance_avx2(src, src_stride, ref, ref_stride, 32, 16, sse, &sum,
vpx_get32x32var_avx2, 32);
return *sse - (((int64_t)sum * sum) >> 9);
return *sse - (uint32_t)(((int64_t)sum * sum) >> 9);
}
unsigned int vpx_variance32x32_avx2(const uint8_t *src, int src_stride,
@@ -70,7 +70,7 @@ unsigned int vpx_variance32x32_avx2(const uint8_t *src, int src_stride,
int sum;
variance_avx2(src, src_stride, ref, ref_stride, 32, 32, sse, &sum,
vpx_get32x32var_avx2, 32);
return *sse - (((int64_t)sum * sum) >> 10);
return *sse - (uint32_t)(((int64_t)sum * sum) >> 10);
}
unsigned int vpx_variance64x64_avx2(const uint8_t *src, int src_stride,
@@ -79,7 +79,7 @@ unsigned int vpx_variance64x64_avx2(const uint8_t *src, int src_stride,
int sum;
variance_avx2(src, src_stride, ref, ref_stride, 64, 64, sse, &sum,
vpx_get32x32var_avx2, 32);
return *sse - (((int64_t)sum * sum) >> 12);
return *sse - (uint32_t)(((int64_t)sum * sum) >> 12);
}
unsigned int vpx_variance64x32_avx2(const uint8_t *src, int src_stride,
@@ -88,7 +88,7 @@ unsigned int vpx_variance64x32_avx2(const uint8_t *src, int src_stride,
int sum;
variance_avx2(src, src_stride, ref, ref_stride, 64, 32, sse, &sum,
vpx_get32x32var_avx2, 32);
return *sse - (((int64_t)sum * sum) >> 11);
return *sse - (uint32_t)(((int64_t)sum * sum) >> 11);
}
unsigned int vpx_sub_pixel_variance32xh_avx2(const uint8_t *src, int src_stride,
@@ -115,7 +115,7 @@ unsigned int vpx_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 vpx_sub_pixel_variance32x32_avx2(const uint8_t *src,
@@ -125,7 +125,7 @@ unsigned int vpx_sub_pixel_variance32x32_avx2(const uint8_t *src,
unsigned int *sse) {
const int se = vpx_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 vpx_sub_pixel_avg_variance64x64_avx2(
@@ -142,7 +142,7 @@ unsigned int vpx_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 vpx_sub_pixel_avg_variance32x32_avx2(
@@ -151,5 +151,5 @@ unsigned int vpx_sub_pixel_avg_variance32x32_avx2(
// Process 32 elements in parallel.
const int se = vpx_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);
}