[Webkit-unassigned] [Bug 55011] Fragment parsing does not need to use HTMLSourceTracker

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=55011





--- Comment #8 from Andy Estes <aestes at apple.com>  2011-02-22 21:34:39 PST ---
(In reply to comment #1)
> I'm not sure I believe it, but this patch seems to make our tiny-innerHTML benchmark twice as fast:
> 
> Before patch:
> avg 5586.1
> median 5594
> stdev 41.295157101045135
> min 5425
> max 5633
> 
> After Patch:
> avg 2603.9
> median 2609.5
> stdev 32.500615378789355
> min 2475
> max 2649
> 
> I suspect (if those numbers are to be believed!) that's all due to avoiding bug 55005.

I'm not able to reproduce as much of a speedup as you're seeing. Here are the numbers I'm getting on tiny-innerHTML:

Before:
avg 2990.85
median 2967
stdev 68.55382921471272
min 2907
max 3198

After:
avg 2562
median 2569
stdev 33.406586176980134
min 2420
max 2582

Is it possible that you were testing with debug builds? That would explain why your before numbers are so high since malloc() is much more expensive in debug.

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