282851 - firefox stops responding when rendering massive table

Status: RESOLVED DUPLICATE of bug 267525

(Core :: Layout: Tables, defect)

Description

[J.]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.8b2) Gecko/20050219 Firefox/1.0+
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.8b2) Gecko/20050219 Firefox/1.0+

this page is in french:

www.playagain.net make firefox crashes in some of it's sections.

thie bug exists since 1.0 as far i know. Using this nighty build (19-02-05) gets
the same.





Reproducible: Always

Steps to Reproduce:
1.go to www.playagain.net
2.in the left panel, (under jeux) select "M.A.M.E"
3.in the menu will appear choose MAME 0.92
4. The page will no tload. firefox will hang or if you are fast enough, you can
try to press many times the "stop" button and maybe firefox will crash or
restore, giving the page uncomplete.
Actual Results:  
firefox hangs or crashes

Expected Results:  
view the page normally

Comment 1

[J.]

Damn... your search engine have problems???

search playagain

got Zarro Boogs found. 

create a bug report

search playagain .. again

got 2 results...

doh! :-)

J.

Comment 2

[Al Behnke]

This might be very slow processing rather than totally hungup.  I have
simplified the web page and placed a copy at 
http://mysite.verizon.net/abehnke/mame8.html . The problem seems to something
about the STYLE on the table of 'border-collapse : collapse' with td 
rowspan="5509".  If I remove either the style or the rowspan then processing
speeds up many times.  The mame8.html page takes less than 3 seconds to load
using MS IE and about 60 seconds on Foxfire 1.0.  This is on Win XP system.

Comment 3

[Steve England [:stevee]]

Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.8b2) Gecko/20050330
Firefox/1.0+
Although now for step 3 you have to "choose MAME 0.94", this page takes an
incredibley long time to load up, and makes firefox give the impression that it
has hung. The page did eventually load for me, but it was quite a wait, and the
throbber stopped its animation whilst the page was thinking.

Comment 4

[Steve England [:stevee]]

-> NEW
Edited summary
Testcase (not really, but an URL that causes the problem)
Request blocking

Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.8b2) Gecko/20050403
Firefox/1.0+

1. Go to http://mysite.verizon.net/abehnke/mame8.html
2. Observe firefox stops responding. You can't move between tabs, resize the
window, etc. The throbber in the tabbar stops animating too. Eventually firefox
finishes thinking about the page and you can browse again normal. It can take up
to one minute for your browser to recover.

It seems to be an enourmous table.

Comment 5

[Steve England [:stevee]]

*** Bug 282403 has been marked as a duplicate of this bug. ***

Comment 6

[Steve England [:stevee]]

Maybe bug 203358 and bug 290866 are related.

Comment 7

[Steve England [:stevee]]

*** Bug 291589 has been marked as a duplicate of this bug. ***

Comment 8

[Bernd]

>Testcase (not really, but an URL that causes the problem)
Then don't mark that it has testcase but create a testcase. 
Thats the usual deficit of all the dupes of bug 54542. If you want this fixed
you don't scream in mozillazine but  create a minimized testcase that shows the
issue, demonstrates that the issue is different from all the other bugs (e.g.
not a dupe). 
Second you provide a jprof
(http://lxr.mozilla.org/mozilla/source/tools/jprof/README.html) or try to get 
CodeAnalyst on Windows running to see whats going on. In short after this step
one can pinpoint which code takes the time.
Third you iterate over your testcase to make it smaller and find what should be
removed from the testcase to make it render fast.
If you don't plan to spend on this
https://bugzilla.mozilla.org/show_bug.cgi?id=54542#c140 shows the earliest
horizon, but it goes definetely after http://tinyurl.com/3hw8y.

So in short: this will not make it for 1.8b2. And if I don't get help also not
for 1.9.

Comment 9

[Steve England [:stevee]]

*** Bug 294249 has been marked as a duplicate of this bug. ***

Comment 10

[Steve England [:stevee]]

*** Bug 260283 has been marked as a duplicate of this bug. ***

Comment 11

[Ryuichi KUBUKI]

*** This bug has been marked as a duplicate of 267525 ***