Bug 28292 - Need to extend DumpRenderTree to expose number of worker threads
: Need to extend DumpRenderTree to expose number of worker threads
Status: RESOLVED FIXED
: WebKit
WebCore Misc.
: 528+ (Nightly build)
: PC Mac OS X 10.5
: P2 Normal
Assigned To:
:
:
:
:
  Show dependency treegraph
 
Reported: 2009-08-13 21:00 PST by
Modified: 2009-08-18 13:21 PST (History)


Attachments
proposed patch (43.12 KB, patch)
2009-08-13 21:45 PST, Andrew Wilson
no flags Review Patch | Details | Formatted Diff | Diff
patch with fix for gtk type error (43.13 KB, patch)
2009-08-13 21:51 PST, Andrew Wilson
eric: review+
atwilson: commit‑queue-
Review Patch | Details | Formatted Diff | Diff


Note

You need to log in before you can comment on or make changes to this bug.


Description From 2009-08-13 21:00:32 PST
There's currently no way to track if/when worker threads actually shut down from javascript layout tests.

We need to extend DumpRenderTree to expose this information so we can write layout tests to verify this functionality.
------- Comment #1 From 2009-08-13 21:45:54 PST -------
Created an attachment (id=34806) [details]
proposed patch

Turns out adding new APIs to WebKit is trickier than it appears. We need to put together a wiki page on this.
------- Comment #2 From 2009-08-13 21:51:19 PST -------
Created an attachment (id=34808) [details]
patch with fix for gtk type error
------- Comment #3 From 2009-08-17 17:37:48 PST -------
(From update of attachment 34808 [details])
Wow.  I'm just blown away by how thorough this change is.  Really, wow.  I don't think I would have had the stamina to change all 4 platforms like that.

This is technically a style violation:
 32         if (layoutTestController.workerThreadCount == 1) {
 33             log("PASS: Worker thread created");
 34         } else {
 35             log("FAIL: After thread creation: layoutTestController.workerThreadCount = " + layoutTestController.workerThreadCount);
 36         }

As far as I can tell, this all looks correct.  I'm certainly no COM expert, so there could be errors there, but looks fine to me.
------- Comment #4 From 2009-08-18 13:21:53 PST -------
Committed as r47429.