shithub: libvpx

Info  •  Files  •  Log  •  Branches

Use correct buffer size in vp8 subpixel variance

In vp8_sub_pixel_variance8x8_neon the temp2 buffer is only initialized
to kHeight8 * kWidth8. However, in the case that xoffset != 0 and
yoffset == 0, var_filter_block2d_bil_w8 is called with output_width
kHeight8PlusOne.

Thanks to cmugurel for diagnosing and yulius for the patch.

Change-Id: Ib71ffd96ffad963c92b8b7ca23f303942785b8e0
https://code.google.com/p/webrtc/issues/detail?id=4190

--- a/vp8/common/arm/neon/vp8_subpixelvariance_neon.c

+++ b/vp8/common/arm/neon/vp8_subpixelvariance_neon.c

@@ -1003,7 +1003,7 @@

         const unsigned char *dst,

         int dst_stride,

         unsigned int *sse) {

-  DECLARE_ALIGNED_ARRAY(kAlign16, uint8_t, temp2, kHeight8 * kWidth8);

+  DECLARE_ALIGNED_ARRAY(kAlign16, uint8_t, temp2, kHeight8PlusOne * kWidth8);

   DECLARE_ALIGNED_ARRAY(kAlign16, uint8_t, fdata3, kHeight8PlusOne * kWidth8);

   if (xoffset == 0) {

     var_filter_block2d_bil_w8(src, temp2, src_stride, kWidth8, kHeight8,

@@ -1021,4 +1021,3 @@

   }

   return variance8x8_neon(temp2, kWidth8, dst, dst_stride, sse);

 }

-