Bug 15958 - base64 spends 1.1% of total time checking for special Infinity case
Summary: base64 spends 1.1% of total time checking for special Infinity case
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: JavaScriptCore (show other bugs)
Version: 528+ (Nightly build)
Hardware: Mac OS X 10.4
: P2 Normal
Assignee: Geoffrey Garen
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2007-11-12 12:53 PST by Eric Seidel (no email)
Modified: 2007-11-13 00:05 PST (History)
2 users (show)

See Also:

Attachments
Patch (1.39 KB, patch)
2007-11-12 23:14 PST, Geoffrey Garen 		sam: review+
Details | Formatted Diff | Diff
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Eric Seidel (no email) 2007-11-12 12:53:03 PST 
base64 spends 1.1% of total time working around toDouble/kjs_dota logic mismatch

Well, here's another softball:

at least 1.4% of *total time* in base64 is spent on this line in UString::toDouble:

if (strncmp(c, "Infinity", 8) == 0) {

We already have our own custom dota (aka kjs_dota) we just need a way to call into that logic where it doesn't ever return "Infinity" and instead does what UString::toDouble wants it to.
Comment 1 Geoffrey Garen 2007-11-12 22:25:04 PST 
I'll take this.
Comment 2 Geoffrey Garen 2007-11-12 23:14:42 PST 
Created attachment 17221 [details]
Patch
Comment 3 Sam Weinig 2007-11-12 23:48:12 PST 
Comment on attachment 17221 [details]
Patch

Looking reasonable.  r=me.
Comment 4 Eric Seidel (no email) 2007-11-12 23:48:50 PST 
Comment on attachment 17221 [details]
Patch

Oh.  I figured we'd had up kjs_dota, but this looks great too. :)
Comment 5 Geoffrey Garen 2007-11-13 00:05:19 PST 
Committed revision 27748.