diff options
| author |  tomhudson@google.com <tomhudson@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81> | 2012-01-11 19:29:08 +0000 |
|---|---|---|
| committer | tomhudson@google.com <tomhudson@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81> | 2012-01-11 19:29:08 +0000 |
| commit | 054ff1efa4f9187ce7fd20aaf3aed7cecf14e12b (patch) | |
| tree | 135488684848de773a9887dc535ef35460290a5d /src/effects | |
| parent | 47628314a1511bd6ab996abf126fc90b57be6065 (diff) | |
Fix the string we're checking for in the #if defined(), so that loops actually do get unrolled on Windows (15% perf win claimed many months ago).
http://codereview.appspot.com/5528078
git-svn-id: http://skia.googlecode.com/svn/trunk@3019 2bbb7eff-a529-9590-31e7-b0007b416f81
Diffstat (limited to 'src/effects')
| -rw-r--r-- | src/effects/SkBlurMask.cpp | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/src/effects/SkBlurMask.cpp b/src/effects/SkBlurMask.cpp index 79fbcf7efd..95a1e6b2e5 100644 --- a/src/effects/SkBlurMask.cpp +++ b/src/effects/SkBlurMask.cpp @@ -16,7 +16,7 @@ // breakeven on Mac, and ~15% slowdown on Linux. // Reading a word at a time when bulding the sum buffer seems to give // us no appreciable speedup on Windows or Mac, and 2% slowdown on Linux. -#if defined(BUILD_FOR_WIN_32) +#if defined(SK_BUILD_FOR_WIN32) #define UNROLL_KERNEL_LOOP 1 #endif |