368908 - frame in ati support page hangs browser

Status: RESOLVED INCOMPLETE

(Firefox :: General, defect)

Description

[towsonu2003]

User-Agent:       Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.1.1) Gecko/20061208 Firefox/2.0.0.1
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.1.1) Gecko/20061208 Firefox/2.0.0.1

Simply go to https://support.ati.com/ics/support/default.asp?deptID=894 and it will hang for 10-15 seconds...

Reproducible: Always




using Ubuntu 6.06 but Firefox 2.0.0.1
Tried with clean profile and nightly...

How do you debug hangs???

Comment 1

[Ria Klaassen (not reading all bugmail)]

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9a2pre) Gecko/20070131 Minefield/3.0a2pre
Has indeed a high CPU usage peak but only for a few seconds here. 

Comment 3

[Bob Clary [:bc] (inactive)]

I can't reproduce a hang with bonecho or minefield debug builds from today on fc6.

If the browser is really stuck somewhere, you can try to attach to it in a debugger and see if you have reproducible stacks or if you can determine the code that is causing the problem. A debug build would be best for that though.

Looks to me that the issue is the folder script used in the nav frame. <https://support.ati.com/ics/support/inc/folder.js>.  Probably a dupe of a known perf bug...

Comment 4

[towsonu2003]

> see if you have reproducible stacks

I doubt I can ever find the code that's causing this... but do you have a pointer on how to do this? I can copy-paste stuff ;) thanks.

Comment 5

[towsonu2003]

that's all I can do: this frame hangs my browser, *everytime*: https://support.ati.com/ics/support/KBFolder.asp?questionID=21578

I tried to isolate stuff, but once I saved the frame/page locally to my hard disk, I couldn't make it hang no more because I didn't know what I was seeing and where to change so the saved page could still work as if it was online... 

Comment 6

[Boris Zbarsky [:bzbarsky]]

Profile shows nothing obvious other than lots of style resolution, lots of image loading, a _huge_ amount of setting of the "display" property, and a lot of eval() calls in JS.  Oh, and lots of document.write.

Offhand, the "display" thing is the most likely issue, and sounds like just bad design on the part of the sight.  But it's hard to tell with all this mess.

Comment 7

[Martijn Wargers (dead)]

Attachment: unminimised testcase

This returns appr. 2100ms for me with the latest 1.8.1 branch build.
With a current trunk build, it returns appr. 1100ms for me.
When removing the image cruft, the testcase is appr. 30% faster for me.

Comment 8

[towsonu2003]

(In reply to comment #7)
> Created an attachment (id=255225) [details]
> unminimised testcase
> 
> This returns appr. 2100ms for me with the latest 1.8.1 branch build.
> With a current trunk build, it returns appr. 1100ms for me.
> When removing the image cruft, the testcase is appr. 30% faster for me. 
> 

I got a pop up saying "11402" with my current Fx 2.0.0.1. I take that's 11402ms? It hanged the browser for a while, although a little less than the hang time I experience at the frame link I sent... Do you need me to try it out with nightly builds or something? 

Comment 9

[towsonu2003]

using nightly build, attached "unminimized testcase" hangs a while (a bit less). pop up says "4764". 

ati website ( http://support.ati.com ) that has the frame hangs about three times longer I believe..

nightly build: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9a2pre) Gecko/2007020611 Minefield/3.0a2pre

Comment 10

[towsonu2003]

(In reply to comment #9)
> using nightly build, attached "unminimized testcase" hangs a while (a bit
> less). pop up says "4764". 
> 

sorry for spamming: I meant "using nightly build with new profile". thanks for the help

Comment 11

[towsonu2003]

(In reply to comment #7)
> Created an attachment (id=255225) [details]
> unminimised testcase
> 

ok this time I tried it on an up-to-date Dapper vmplayer (have vmware-tools installed and running, if that makes any difference) with new profile and nightly firefox build, the popup said "6459" -it hanged a little longer than it did in my previous comment-...

I went to support.ati.com and it hanged as much as it did without vmware on nightly build (mentioned in above comment). 

this build was: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9a3pre) Gecko/20070215 Minefield/3.0a3pre

Comment 12

[Joe Smith]

WFM  -  Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.0.2pre) Gecko/2008072309 Minefield/3.0.2pre ID:2008072309 on Ubuntu 8.10 Alpha 2

Visited support.ati.com with no issues.

When I went to https://support.ati.com/ics/support/default.asp?deptID=894 received this error in console:

CSS Error (https://support.ati.com/ics/support/KBFolder.asp :0.30): Error in parsing value for property 'position'.  Declaration dropped.

Just no, this error also popped up:
WARNING: An event was posted to a thread that will never run it (rejected): file nsThread.cpp, line 359
WARNING: leaking reference to nsTimerImpl: file nsTimerImpl.cpp, line 464

Other than that, it loaded a bit slow, but that could just be due to the network.

Comment 13

[José Jeria]

WFM: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.0.1) Gecko/2008070206 Firefox/3.0.1

Comment 14

[Tyler Downer [He/Him]]

This bug was reported on Firefox 2.x or older, which is no longer supported and will not be receiving any more updates. I strongly suggest that you update to Firefox 3.6.3 or later, update your plugins (flash, adobe, etc.), and retest in a new profile. If you still see the issue with the updated Firefox, please post here. Otherwise, please close as RESOLVED > WORKSFORME
http://www.mozilla.com
http://support.mozilla.com/kb/Managing+profiles
http://support.mozilla.com/kb/Safe+mode

Comment 15

[Tyler Downer [He/Him]]

No reply, INCOMPLETE. Please retest with Firefox 3.6.x or later and a new profile (http://support.mozilla.com/kb/Managing+profiles). If you continue to see this issue with the newest firefox and a new profile, then please comment on this bug.