diff options
| author |  tomhudson@google.com <tomhudson@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81> | 2012-06-11 12:42:24 +0000 |
|---|---|---|
| committer | tomhudson@google.com <tomhudson@google.com@2bbb7eff-a529-9590-31e7-b0007b416f81> | 2012-06-11 12:42:24 +0000 |
| commit | f8a2289667e2ad2a4776e43485ce29f32e584d0b (patch) | |
| tree | 8caee1281a58cdbd14a034213e285dffdcdbcdda /make.py | |
| parent | 23113dd4206abadd95d401821e90ceb66232e4ed (diff) | |
Manually unroll loop in convolution (blur) - most GL drivers do, but some
don't, and this leads to 20-30% benchmark speedup.
http://codereview.appspot.com/6308057/
git-svn-id: http://skia.googlecode.com/svn/trunk@4218 2bbb7eff-a529-9590-31e7-b0007b416f81
Diffstat (limited to 'make.py')
0 files changed, 0 insertions, 0 deletions