292541 - Firefox/Xorg crash when site loads

Status: RESOLVED WONTFIX

(Firefox :: General, defect)

Description

[Garrett Cooper]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.7) Gecko/20050414 Firefox/1.0.3
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.7) Gecko/20050426 Firefox/1.0.3

When I go to the website above, almost every single time I go there my system
hardlocks and I am unable to do anything but login via SSH from other other
machine. I attempt to kill Firefox, X, or anything to bring my PC back to life
but unfortunately it won't NOHUP for me.



Reproducible: Always

Steps to Reproduce:
1. Go to website.
Actual Results:  
Hardlock!

Expected Results:  
Loaded the page.

I have debug information possibly since I compiled firefox with the debug flag,
so if you can help me find out where that is I can help you. The interesting
thing is that it may be a Javascript error like a buffer overrun or something
that is causing Firefox to crash, but this does not occur on my Windows version
of firefox, so it is something related to how my Linux version was built or
something else I overlooked. Since this is something that may be a security
issue, I will mark it as a security problem.

Additional info about machine/relevant programs:

gman@sprsd gman $ emerge --info
Portage 2.0.51.19 (default-linux/x86/2005.0, gcc-3.4.3-20050110,
glibc-2.3.4.20050125-r1, 2.6.11-gentoo-r6 i686)
=================================================================
System uname: 2.6.11-gentoo-r6 i686 Intel(R) Pentium(R) 4 CPU 2.40GHz
Gentoo Base System version 1.4.16
Python:              dev-lang/python-2.3.5 [2.3.5 (#1, Apr 28 2005, 12:21:49)]
distcc 2.16 i686-pc-linux-gnu (protocols 1 and 2) (default port 3632) [disabled]
dev-lang/python:     2.3.5
sys-apps/sandbox:    [Not Present]
sys-devel/autoconf:  2.59-r6, 2.13
sys-devel/automake:  1.7.9-r1, 1.8.5-r3, 1.5, 1.4_p6, 1.6.3, 1.9.4
sys-devel/binutils:  2.15.92.0.2-r7
sys-devel/libtool:   1.5.16
virtual/os-headers:  2.6.8.1-r2
ACCEPT_KEYWORDS="x86"
AUTOCLEAN="yes"
CFLAGS="-O3 -march=pentium4 -fomit-frame-pointer -mmmx -msse -msse2
-mfpmath=sse,387"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3.3/env /usr/kde/3.3/share
/config /usr/kde/3.3/shutdown /usr/kde/3.4/env /usr/kde/3.4/share/config /usr/kd
e/3.4/shutdown /usr/kde/3/share/config /usr/lib/X11/xkb /usr/share/config /var/q
mail/control"
CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d"
CXXFLAGS="-O3 -march=pentium4 -fomit-frame-pointer -mmmx -msse -msse2 -mfpmath=s
se,387"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoaddcvs autoconfig ccache distlocks keepwork sandbox sfperms strict
"
GENTOO_MIRRORS="http://the-outsider.no-ip.com/ http://distfiles.gentoo.org/"
MAKEOPTS="-j2"
PKGDIR="/usr/portage/packages"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/cstm/portage"
SYNC="rsync://the-outsider.no-ip.com/gentoo-portage/"
USE="x86 X alsa apm arts avi berkdb bitmap-fonts cdr cjk crypt cups divx4linux
dvd dvdr emboss fam flac foomaticdb fortran gdbm gif gpm gtk gtk2 imagemagick
imlib immqt-bc java jpeg ldap libg++ libwww live mad mmx mmx2 motif mozilla mp3
mpeg ncurses nls nptl nptlonly nvidia ogg oggvorbis opengl oss pam pdflib perl
png python qt quicktime readline samba sdl spell sse sse2 ssl svga tcltk tcpd
tiff truetype truetype-fonts type1-fonts unicode vorbis xml2 xmms xprint xv xvid
zlib
video_cards_nvidia userland_GNU kernel_linux libc_glibc"
Unset:  ASFLAGS, CBUILD, CTARGET, LANG, LC_ALL, LDFLAGS, LINGUAS

gman@sprsd gman $ emerge -pv mozilla-firefox freetype =x11-libs/gtk+-1.2.10-r11 gtk+

These are the packages that I would merge, in order:

Calculating dependencies ...done!
[ebuild   R   ] www-client/mozilla-firefox-1.0.3  +debug -gnome +java +ldap
-mozdevelop -moznoxft -mozsvg -mozxmlterm -xinerama +xprint 32,015 kB
[ebuild   R   ] media-libs/freetype-2.1.9-r1  -bindist -debug -doc +zlib 969 kB
[ebuild   R   ] x11-libs/gtk+-1.2.10-r11  -debug +nls 2,880 kB
[ebuild   R   ] x11-libs/gtk+-2.6.4-r1  -debug -doc +jpeg -static +tiff 10,985 kB

Comment 1

[Daniel Veditz [:dveditz]]

wfm on linux and windows... any more details on how to reproduce this? The link
above is a blog, does it depend on particular content that may have scrolled off?

Theres an embedded mp3, which for me triggers the missing plugin infobar. Do you
have a player set up for that, and maybe that's what's causing the problem?

Comment 2

[Garrett Cooper]

I am using mplayerplug-in 2.80 that was built with the gecko-sdk 1.7.5, but the
issue isn't the audio I believe since the audio plugin seems to work perfectly
fine on other blogs. Maybe it's something to do with the embed tag or something?

Comment 3

[Garrett Cooper]

Got the same behavior on another blog site: http://xanga.com/ellis219. I will
try disabling the whiteheart theme but first I will check with Windows based
Firefox to see what the similarities are between the 2 sites.

Comment 4

[Garrett Cooper]

Apparently whatever affects the compiled source affects the binary as well. I
tried out the binary and it affects it just as much-if not worse
surprisingly-than the source compiled version.

Overall, I'm wondering then if it was some level of bad parsing in the HTML. I
did see a suspicious section
(<xmp><script>document.write('<!---->')</script><!----></xmp>) in the blog sites
that gave me issues, and I know that <xmp> is an outdated HTML tag, so maybe it
Firefox misinterprets the commenting produced by the Javascript and crashes.

Note: I thought that going to those websites crashed Xorg with the source
compiled browser, but in fact if I press CRTL-ALT-BACKSPACE, the key command
kills Xorg and the displayed graphics remain on the screen.

Once again, this issue does not affect Windows based binaries of Firefox, so the
issue at hand may be a combination of something of an unchecked error with
Freetype, GTK, or Xorg.

Comment 5

[Garrett Cooper]

Also, sorry for the extra comments, but even when I do have all of the web
features turned off, Firefox crashes. That was the reasoning behind my
assumption that the issue is a parsing error or something similar.

Comment 6

[Garrett Cooper]

Ok, found out some more info...

I can load up the page with firefox 1.0.4 on FC2 perfectly fine.

Commenting out the CSS in the header it seems as if the problem goes away...
(?!?!). Thus, I'm going to do some more inspection on the CSS.

Comment 7

[Garrett Cooper]

Bingo. Found the problem part in the CSS that's messing with my Firefox
(specificially the commented out portion of the code).


BODY { FONT-SIZE: x-small; BACKGROUND:
/*url(http://students.washington.edu/audits/tinkywinkysogay.bmp) fixed no-repeat
center center;*/ COLOR: #000000; FONT-FAMILY: Comic Sans MS; TEXT-ALIGN: center;
TEXT-DECORATION: none
}


So apparently the image or the image invoking CSS is halting the browser. I
wonder then... does parsing the entire HTML occur prior to turning off web
features. That would make sense why my browser crashed when the 'web features'
were off.

Comment 8

[Garrett Cooper]

Ok, apparently the issue doesn't affect my Celeron box interestingly enough...

gman@D-128-208-47-172 gman $ emerge info && emerge -pv mozilla-firefox xorg-x11
freetype
Portage 2.0.51.19 (default-linux/x86/2005.0, gcc-3.4.3-20050110, glibc-2.3.5-r0,
2.6.11-gentoo-r4 i686)
=================================================================
System uname: 2.6.11-gentoo-r4 i686 Intel(R) Celeron(TM) CPU                1200MHz
Gentoo Base System version 1.4.16
Python:              dev-lang/python-2.3.5 [2.3.5 (#1, Apr 30 2005, 05:31:29)]
distcc 2.16 i686-pc-linux-gnu (protocols 1 and 2) (default port 3632) [enabled]
dev-lang/python:     2.3.5
sys-apps/sandbox:    [Not Present]
sys-devel/autoconf:  2.59-r6, 2.13
sys-devel/automake:  1.7.9-r1, 1.8.5-r3, 1.5, 1.4_p6, 1.6.3, 1.9.5
sys-devel/binutils:  2.15.92.0.2-r7
sys-devel/libtool:   1.5.16
virtual/os-headers:  2.6.8.1-r2
ACCEPT_KEYWORDS="x86"
AUTOCLEAN="yes"
CFLAGS="-O3 -pipe -fomit-frame-pointer -march=pentium3 -mmmx -msse -mfpmath=sse,387"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3/share/config
/usr/lib/X11/xkb /usr/share/config /var/bind /var/qmail/control"
CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d"
CXXFLAGS="-O3 -pipe -fomit-frame-pointer -march=pentium3 -mmmx -msse
-mfpmath=sse,387"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoaddcvs autoconfig ccache distcc distlocks sandbox sfperms strict"
GENTOO_MIRRORS="http://distfiles.gentoo.org/"
MAKEOPTS="-j3"
PKGDIR="/usr/portage/packages"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/usr/local/zugaina-portage /cstm/portage"
SYNC="rsync://rsync.gentoo.org/gentoo-portage"
USE="x86 X alsa berkdb bitmap-fonts crypt cups emboss encode foomaticdb fortran
gd gdbm gif gnutls gpm gtk gtk2 imap jpeg ldap libg++ libwww maildir mmx motif
mp3 mysql ncurses nls oss pam perl png python readline samba sdl spell sse ssl
svga tcpd tiff truetype truetype-fonts type1-fonts xml2 zlib userland_GNU
kernel_linux elibc_glibc"
Unset:  ASFLAGS, CBUILD, CTARGET, LANG, LC_ALL, LDFLAGS, LINGUAS


These are the packages that I would merge, in order:

Calculating dependencies ...done!
[ebuild   R   ] www-client/mozilla-firefox-1.0.4  -debug -gnome -java +ldap
-mozdevelop -moznoxft -mozsvg -mozxmlterm -xinerama -xprint 0 kB 
[ebuild   R   ] x11-base/xorg-x11-6.8.2-r1  -3dfx -3dnow +bitmap-fonts -cjk
-debug -dlloader -dmx -doc -font-server -hardened -insecure-drivers -ipv6
-minimal +mmx +nls -opengl +pam -sdk +sse -static +truetype-fonts +type1-fonts
(-uclibc) -xprint -xv 0 kB
[ebuild   R   ] media-libs/freetype-2.1.9-r1  -bindist -debug -doc +zlib 0 kB

Some of the major hardware differences:
P4 box has an GeForce4 Ti4200 with OpenGL enabled via the nvidia driver where
the Celeron box has an SiS630 chipset.

The sse2 optimization flags as well as the nvidia driver/OpenGL flags may have
something to do with this issue, but I cannot be sure...

Whatever is occurring though is possibly not directly due to Firefox, but rather
is something that is triggered by Firefox in Xorg.

Comment 9

[Garrett Cooper]

Well, I figured out what the issue is. It's some sort of rendering issue with
XFCE 4.2. Needless to say after I switched to Fluxbox for a little while to test
things, I had no further issues with the blog site. I'm going to close this bug
with Mozilla since the issue instead lies with XFCE.