]> granicus.if.org Git - libvpx/commit
projects / libvpx / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: d84bf53) | patch
Fix 32-bit unsigned integer overflow in block_variance
authorkyslov <kyslov@google.com>
Fri, 22 Feb 2019 22:15:22 +0000 (14:15 -0800)
committerkyslov <kyslov@google.com>
Fri, 22 Feb 2019 23:55:18 +0000 (15:55 -0800)
commit7e1c766158070c5abcce19d35467a65656836367
treecbc98a0ba45abe44b836bcb27847d7ac65f3bcabtree | snapshot
parentd84bf53f607982f002f1900e4c5894ba9b470770commit | diff
Fix 32-bit unsigned integer overflow in block_variance

When compiled for High Bitdepth SSE can overflow 32-bit unsigned
integer, so change it to 64 bit. Also fixing unit/int mismatch of sum

BUG=webm:1601

Change-Id: Ib576ed1d5579b0c2b4661058aa64119560b652bf
vp9/encoder/vp9_aq_variance.c diff | blob | history
Unnamed repository; edit this file 'description' to name the repository.