Closed
Bug 222061
Opened 22 years ago
Closed 22 years ago
Setting style.display on div with iframe to 'none' crashes browser
Categories
(Core Graveyard :: GFX: Gtk, defect)
Tracking
(Not tracked)
RESOLVED
DUPLICATE
of bug 205492
People
(Reporter: cxreg, Assigned: blizzard)
References
()
Details
(Keywords: crash)
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20031010 Galeon/1.3.9
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20031010 Galeon/1.3.9
In http://genericorp.net/~count/mozcrash.html (minimum case, derived from
espn.com) mozilla crashes immediately upon loading the page. Crash will not
occur if all of the following are not met:
must have iframe
must have some displayed text
assign 'none' to style.display DOM element of div
Reproducible: Always
Steps to Reproduce:
1. load example page
2. watch mozilla crash
3. rinse, repeat
Actual Results:
segfault
Expected Results:
unsure
Comment 1•22 years ago
|
||
WFM, 2003-10-13-05 trunk Linux
![]() |
||
Comment 2•22 years ago
|
||
Crashes for me on
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.5) Gecko/20031003 Firebird/0.7
![]() |
||
Comment 4•22 years ago
|
||
Dave, Nathan, could you please test a current build and check whether that
crashes for you too?
![]() |
||
Comment 5•22 years ago
|
||
I just tried with the 20031025 nightly, and it no longer crashes - however, I do
get the following printed to stderr when it loads the page:
Gtk-WARNING **: invalid cast from (NULL) pointer to `GtkObject'
![]() |
||
Comment 6•22 years ago
|
||
Chris, what's up with that warning? I've seen that too, recently...
![]() |
||
Comment 7•22 years ago
|
||
I was a bit curious, so I built my own nightly using the same options that the
Debian mozilla packages use, and it *did* still crash. Offhand my guess would
be it's triggered by something in either GTK2 or XFT (My Firebird build uses
both of those, and exhibits the crash as well). I'm building again, sans
"--disable-debug --enable-strip-libs" to see if I can get any useful debugging
information from the crash.
./configure --prefix=/home/kraken/mozilla --enable-svg --without-system-mng
--without-system-png --disable-pedantic
--with-default-mozilla-five-home=/home/kraken/mozilla --disable-debug
--disable-tests --disable-short-wchar --enable-xprint --enable-strip-libs
--enable-crypto --enable-mathml --enable-oji --enable-extensions=all
--enable-ldap --with-system-zlib --enable-freetype2
--enable-default-toolkit=gtk2 --enable-toolkit=gtk2 --enable-xft
![]() |
||
Comment 9•22 years ago
|
||
I recompiled for debugging, and ran
./mozilla -g http://genericorp.net/~count/mozcrash.html
When it crashed, I did a thread apply all bt full. The complete output of the
session is at http://drunkmonkey.org/mozcrash.txt
![]() |
||
Updated•22 years ago
|
Status: UNCONFIRMED → NEW
Ever confirmed: true
Comment 10•22 years ago
|
||
I can reproduce the crash too.
This is the reason:
(gdb) print mDrawingarea
$1 = (MozDrawingarea *) 0x0
![]() |
Assignee | |
Comment 11•22 years ago
|
||
Same as bug 222061? I have a workaround patch in my moz rpms for Red Hat that
just print out an NS_WARNING and return failure.
![]() |
Assignee | |
Comment 12•22 years ago
|
||
-> dup
*** This bug has been marked as a duplicate of 205492 ***
Status: NEW → RESOLVED
Closed: 22 years ago
Resolution: --- → DUPLICATE
Updated•17 years ago
|
Product: Core → Core Graveyard
You need to log in
before you can comment on or make changes to this bug.
Description
•