projects / platform / upstream / libSkiaSharp.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: f91e3d4) | patch
Add sk_calloc. Remove SkMemory_stdlib, which seems unused.
authormtklein@google.com <mtklein@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81>
Fri, 20 Sep 2013 14:31:45 +0000 (14:31 +0000)
committermtklein@google.com <mtklein@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81>
Fri, 20 Sep 2013 14:31:45 +0000 (14:31 +0000)
commit519f9677a41239808f41a7c13ef1f6e05eb1ed50
treea7a1acf5fc2c9d8c2390c48bc1a604628ff513e5tree | snapshot
parentf91e3d4f54de9976b6538decadd977b19e49eaddcommit | diff
Add sk_calloc.  Remove SkMemory_stdlib, which seems unused.

I'm seeing basically no difference between malloc + bzero and calloc on my desktop, but on a Galaxy Nexus calloc is never slower, and significantly faster once the allocation size becomes large, both for allocation and for _reading_.

BUG=skia:1662
R=reed@google.com

Review URL: https://codereview.chromium.org/24251008

git-svn-id: http://skia.googlecode.com/svn/trunk@11414 2bbb7eff-a529-9590-31e7-b0007b416f81
bench/MemoryBench.cpp diff | blob | history
gyp/tests.gyp diff | blob | history
include/core/SkTypes.h diff | blob | history
src/core/SkMemory_stdlib.cpp [deleted file] blob | history
src/ports/SkMemory_malloc.cpp diff | blob | history
src/ports/SkMemory_mozalloc.cpp diff | blob | history
tests/MemoryTest.cpp [new file with mode: 0644] blob
Domain: Dotnet / API;