[Webkit-unassigned] [Bug 37292] http://trac.webkit.org/changeset/57215 caused perf/memory regressions

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=37292





--- Comment #5 from Ojan Vafai <ojan at chromium.org>  2010-04-09 16:05:13 PST ---
OK. I've created a reduced perf test that I'll attach shortly. The page cycler
runs through 30 pages and does 10 cycles. I took the tests where
median_after_patch / median_before_patch > 1.1. In other words, the pages that
regressed. In the test case, I load each one in an iframe 100 times.

I ran each case 5 times in Safari. Numbers are in milliseconds.

Before r57215: 106454, 105744, 102916, 101746, 103130
After r57215: 116417, 115711, 116228, 122100, 115701
After r57215 with canUseGlyphCache changes commented out: 105953, 106326,
105762, 105035, 105254

So, clearly the canUseGlyphCache changes account for all of the perf
regression.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.