418388 - Selecting and navigating product criteria (very Ajax heavy) produces huge CPU load

Status: RESOLVED INCOMPLETE

(Core :: General, defect)

Description

[gummi]

User-Agent:       Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.8.0.7) Gecko/20060909 Firefox/1.5.0.7
Build Identifier: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.8.0.7) Gecko/20060909 Firefox/1.5.0.7

At the site, click on the top navigation bar "Composants et Connectique". Choose, for example, "Disque Dur Interne". Then refine the criteria using the options on the navigation bar on the left hand side. Going through these options gives a huge spike in the CPU load and makes Camino unusable. I suspect it has something to do with the DHTML.

Reproducible: Always

Steps to Reproduce:
1.At the site, click on the top navigation bar "Composants et Connectique"
2.Choose, for example, "Disque Dur Interne".
3.Then refine the criteria using the options on the navigation bar on the left hand side.
Actual Results:  
Unusable application.

Expected Results:  
Allow one to use to application.

Comment 1

[Chris Lawson (gone)]

Can you get a sample and attach it here, please? Directions at

http://caminobrowser.org/documentation/bugzilla/#hang

What version of Camino are you reporting this against? Please use the version in question when filing bugs in the future, as Bugzilla automatically captures the user-agent string of the browser you file the bug with.

cl

Comment 2

[gummi]

Attachment: Sample output generated by Activity Monitor for Camino whilst visiting navigating surcouf.com

Camino version: Version 1.6b2 (1.8.1.12pre 2008011814). But this problem can be recreated in the official build 1.5.5.

Comment 3

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

I see this, and it's much better on the trunk, though it's still pretty laggy in Camino and Minefield.  I see a bunch of DOM, selection, and (!) mozInlineSpellcheck stuff in my sampling, though.

I suppose we should send this to Core somewhere and let them see what they can do to improve pref?

Comment 4

[Stuart Morgan]

I'm downgrading this, since it doesn't hang on trunk. We should attach a trunk sample here, and send it to core.

Comment 5

[Chris Lawson (gone)]

gummi, can you please capture a sample once again, this time using a nightly build (from http://caminobrowser.org/contribute/ )?

Comment 6

[gummi]

Attachment: surcouf.com hang-CPU spike -- Second sample using latest nightly build - 260408

Comment 7

[gummi]

Attachment: surcouf.com hang-CPU spike -- Third sample using latest nightly build - 260408

In the last two samples I was performing different tasks on the website that choked the latest nightly build of Camino.

Comment 8

[Chris Lawson (gone)]

Looking through Core bugs and having taken my own sample, I wonder if maybe this is a dupe of bug 77945.

Comment 9

[Wayne Mery (:wsmwk)]

I don't see a spike. Still a valid testcase?

Comment 10

[Tyler Downer [He/Him]]

No reply, INCO. Please reopen if you still see this bug in Firefox 3.5 or later in Firefox safe mode (http://support.mozilla.com/en-US/kb/Safe+Mode) or a new Firefox profile (http://support.mozilla.com/en-US/kb/Profiles) with updated plugins. A new testcase that shows this would be helpful if it still happens.