395635 - Full render in JavaScript Progressive Raytracer is 20 times slower than Opera 9.5

Status: RESOLVED DUPLICATE of bug 424715

(Core :: Layout, defect)

Description

[Adam Hauner]

On nontroppo.org there are several performance tests for comparing Opera 9.5 (Kestrel) with other browsers. http://nontroppo.org/timer/kestrel_tests/

One of tests is JavaScript Progressive Raytracer.
http://nontroppo.org/timer/progressive_raytracer.html (attached as testcase)

It's full render takes cca 20 times more time in Firefox3.0a8 than in first alpha of Opera 9.5. Results from several runs of test:
Kestrel on XP:         28.8 s
Firefox 3.0a7 on XP:  613.5 s

On my notebook with actual nightly Fx:
Opera 9.50 Alpha (build: 9500)
40.953 seconds

Firefox/3.0a8pre (rv:1.9a8pre Gecko/2007090905)
801.217 seconds

Comment 1

[Adam Hauner]

Attachment: JS Progressive Raytracer (testcase)

Comment 2

[Olli Pettay [:smaug][[e-mail]]]

Attachment: jprof

Flat Profile

Total hit count: 13621
Count %Total  Function Name
3146   23.1     nsLineBox::LastChild() const
3085   22.6     nsLayoutUtils::GetLastSibling(nsIFrame*)
3031   22.3     nsLineBox::RFindLineContaining(nsIFrame*, nsLineList_iterator const&, nsLineList_iterator&, int*)
3021   22.2     nsFrameList::AppendFrames(nsIFrame*, nsIFrame*)
...

This might be dup of some other bug.

Comment 3

[Adam Hauner]

nsLineBox::LastChild() is mentioned by BZ in bug 40988 comment #62 and also in bug 237735 comment #4, but there isn't probably any bug for it.

nsLayoutUtils::GetLastSibling is probably bug 233463 - adding it to dependencies

nsLineBox::RFindLineContaining is mentioned by BZ in bug 304598 comment #19

Comment 4

[Damian Shaw [Quan]]

I know this bug is the older bug with more details. But the other bug is confirmed, has more CC, votes and seems to have more developer attention. If you strongly disagree with my choice I will be happy to reverse the duplicate.

Feel free to copy over as much information as you want.