diff options
| author | darin@chromium.org <darin@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2009-12-09 21:05:56 +0000 |
|---|---|---|
| committer | darin@chromium.org <darin@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2009-12-09 21:05:56 +0000 |
| commit | 8fbfa3dc1620eeb1da8908c359bdbbb1467c44ff (patch) | |
| tree | e993637c6728e094cdb98d7f1386b014f4292e66 /chrome/renderer/paint_aggregator.h | |
| parent | 0f4e0eb5e6f88bc639e52c63023bf2ad46a70fbd (diff) | |
| download | chromium_src-8fbfa3dc1620eeb1da8908c359bdbbb1467c44ff.zip chromium_src-8fbfa3dc1620eeb1da8908c359bdbbb1467c44ff.tar.gz chromium_src-8fbfa3dc1620eeb1da8908c359bdbbb1467c44ff.tar.bz2 | |
Limit the total number of paint rects.
Based on some testing on a single core machine, reducing this
number below 10 doesn't seem to help. On some of the page
cycler tests (intl2 in particular), I observed some excessive
paint rect counts (>50). Those outliers seem to explain the
perf regression observed for single core machines.
R=brettw
BUG=29589
TEST=none
Review URL: http://codereview.chromium.org/481002
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@34189 0039d316-1c4b-4281-b951-d872f2087c98
Diffstat (limited to 'chrome/renderer/paint_aggregator.h')
| -rw-r--r-- | chrome/renderer/paint_aggregator.h | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/chrome/renderer/paint_aggregator.h b/chrome/renderer/paint_aggregator.h index d1af49b..9a595be 100644 --- a/chrome/renderer/paint_aggregator.h +++ b/chrome/renderer/paint_aggregator.h @@ -50,6 +50,7 @@ class PaintAggregator { gfx::Rect ScrollPaintRect(const gfx::Rect& paint_rect, int dx, int dy) const; bool ShouldInvalidateScrollRect(const gfx::Rect& rect) const; void InvalidateScrollRect(); + void CombinePaintRects(); PendingUpdate update_; }; |