316623 - Stuck Process running RandomStyles hc_nodtdstaff.xml

Status: RESOLVED WORKSFORME

(Core :: Web Painting, defect)

Description

[Bob Clary [:bc] (inactive)]

Automated RandomStyles testing on WiNXP with today's FF trunk:

http://test.bclary.com/tests/w3.org/2001/DOM-Test-Suite/build/ecmascript/level1/core/files/hc_nodtdstaff.xml
seed=140;skip=255;changesPerInterval=144;interval=246

Eventually the browser stops repainting the window. CPU is at 0%, closing the browser leaves a process running, then attempting to open a new instance of the browser starts a new process even if MOZ_NO_REMOTE is not set.

Guessing on Layout: View Rendering

Comment 1

[Bob Clary [:bc] (inactive)]

ditto: 
http://test.bclary.com/tests/w3.org/2001/DOM-Test-Suite/build/ecmascript/level1/core/files/hc_staff.xml
seed=140;skip=255;changesPerInterval=144;interval=246

Comment 2

[Bob Clary [:bc] (inactive)]

Attachment: list of xml urls and parameters

I am getting similar behavior with the other xml files in the DOM Test Suite with Random Styles. I'll attach the list here along with the Random Styles parameters for testing purposes. This list will become part of the automated fuzzbusting tests.

Comment 3

[Jesse Ruderman]

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9a1) Gecko/20051203 Firefox/1.6a1

I can't reproduce using the URLs in comment 0 and comment 1.  I let the status bar counter go pretty high (20000 and 10000).

Do you still see this bug, and if so, how far does the status bar counter get on each combination of URL and Random Styles parameters?

Comment 4

[Bob Clary [:bc] (inactive)]

(In reply to comment #3)

> 
> Do you still see this bug, and if so, how far does the status bar counter get
> on each combination of URL and Random Styles parameters?
> 

Not on trunk, but still on 1.8 branch. It gets to 15375. We can mark it works for me if you think that is best.

Comment 5

[Daniel Veditz [:dveditz]]

I actually crash with a null dereference at
 nsGridRowLayout::GetGrid() Line 141

Before the crash it stopped repainting and did appear "stuck", but crashed when I attempted to resize the window.

Comment 6

[Daniel Veditz [:dveditz]]

No sign of a fix, not realistic for 1.8.0.1

Comment 7

[Daniel Veditz [:dveditz]]

Not making 1.8.0.2 either :-(

Comment 8

[Josh Aas]

Seems likely that this got fixed or was mitigated since anyone last looked at it. Bob - can you still reproduce? Can we open this up and mark it WFM? I wasn't able to see any issues with Firefox 10 on Win 7 in a VM.

Comment 9

[Bob Clary [:bc] (inactive)]

I can't reproduce since I don't quite have the same setup as back then. I don't think it will be worth trying to run these exact tests now since Jesse has been running random styles for 7 years and has filed tons of bugs already. -> WFM.