Bug 58143 - PerformanceTiming returns inconsistent values when timing is null.
Summary: PerformanceTiming returns inconsistent values when timing is null.
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: Page Loading (show other bugs)
Version: 528+ (Nightly build)
Hardware: PC OS X 10.5
: P2 Normal
Assignee: Nobody
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2011-04-08 08:54 PDT by Vsevolod Vlasov
Modified: 2011-04-11 13:46 PDT (History)
7 users (show)

See Also:


Attachments
Patch (3.89 KB, patch)
2011-04-08 14:53 PDT, Vsevolod Vlasov
no flags Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Vsevolod Vlasov 2011-04-08 08:54:22 PDT
PerformanceTiming returns inconsistent values when timing is null.
Comment 1 Vsevolod Vlasov 2011-04-08 14:53:28 PDT
Created attachment 88876 [details]
Patch
Comment 2 Eric Seidel (no email) 2011-04-10 15:42:48 PDT
Comment on attachment 88876 [details]
Patch

How do we test this?
Comment 3 Vsevolod Vlasov 2011-04-10 17:20:02 PDT
This is tested in fast/dom/webtiming*.html layout tests.
After landing of http://codereview.chromium.org/6823006/ these tests would fail, unless this patch is also landed.
Comment 4 Eric Seidel (no email) 2011-04-10 17:35:12 PDT
(In reply to comment #3)
> This is tested in fast/dom/webtiming*.html layout tests.
> After landing of http://codereview.chromium.org/6823006/ these tests would fail, unless this patch is also landed.

Ideally the ChagneLog would contain this information. :)
Comment 5 Pavel Feldman 2011-04-10 23:04:33 PDT
Comment on attachment 88876 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=88876&action=review

> Source/WebCore/page/PerformanceTiming.cpp:264
> +        return requestStart();

You probably should do the same for the secureConnection timers above and for the responseEnd below.
Comment 6 Vsevolod Vlasov 2011-04-11 10:33:59 PDT
Comment on attachment 88876 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=88876&action=review

>> Source/WebCore/page/PerformanceTiming.cpp:264
>> +        return requestStart();
> 
> You probably should do the same for the secureConnection timers above and for the responseEnd below.

secureConnection: "If this attribute is available but HTTPS is not used, this attribute must return zero.", so zero is correct.
respondEnd is filled from documentLoadTiming, not resourceLoadTiming, so its value is correct as well.
Comment 7 Tony Gentilcore 2011-04-11 10:46:23 PDT
Thanks for taking care of this.
Comment 8 WebKit Commit Bot 2011-04-11 13:43:15 PDT
The commit-queue encountered the following flaky tests while processing attachment 88876 [details]:

fast/workers/storage/use-same-database-in-page-and-workers.html bug 50995 (author: dumi@chromium.org)
The commit-queue is continuing to process your patch.
Comment 9 WebKit Commit Bot 2011-04-11 13:45:55 PDT
Comment on attachment 88876 [details]
Patch

Clearing flags on attachment: 88876

Committed r83491: <http://trac.webkit.org/changeset/83491>
Comment 10 WebKit Commit Bot 2011-04-11 13:46:00 PDT
All reviewed patches have been landed.  Closing bug.