My system ist P4 2,533 1024 MB Ram Use="-3dnow sse" ....-march=petium4 --- Here the interesting part of the strace output: stat64("/usr/X11R6/lib/X11/fonts/encodings/fonts.cache-1", {st_mode=S_IFREG|0644 , st_size=17, ...}) = 0 fstat64(32, {st_mode=S_IFREG|0644, st_size=17, ...}) = 0 mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x40 235000 read(32, "\"large\" 0 \".dir\"\n", 4096) = 17 read(32, "", 4096) = 0 close(32) = 0 munmap(0x40235000, 4096) = 0 open("/usr/X11R6/lib/X11/fonts/truetype/fonts.cache-1", O_RDONLY) = 32 stat64("/usr/X11R6/lib/X11/fonts/truetype", {st_mode=S_IFDIR|0755, st_size=1152, ...}) = 0 stat64("/usr/X11R6/lib/X11/fonts/truetype/fonts.cache-1", {st_mode=S_IFREG|0644, st_size=25553, ...}) = 0 close(32) = 0 stat64("/usr/X11R6/lib/X11/fonts/truetype", {st_mode=S_IFDIR|0755, st_size=1152, ...}) = 0 open("/usr/X11R6/lib/X11/fonts/Speedo/fonts.cache-1", O_RDONLY) = 32 stat64("/usr/X11R6/lib/X11/fonts/Speedo", {st_mode=S_IFDIR|0755, st_size=432, .. .}) = 0 stat64("/usr/X11R6/lib/X11/fonts/Speedo/fonts.cache-1", {st_mode=S_IFREG|0644, s t_size=0, ...}) = 0 close(32) = 0 stat64("/usr/X11R6/lib/X11/fonts/Speedo", {st_mode=S_IFDIR|0755, st_size=432, .. .}) = 0 open("/usr/X11R6/lib/X11/fonts/local/fonts.cache-1", O_RDONLY) = 32 stat64("/usr/X11R6/lib/X11/fonts/local", {st_mode=S_IFDIR|0755, st_size=144, ... }) = 0 stat64("/usr/X11R6/lib/X11/fonts/local/fonts.cache-1", {st_mode=S_IFREG|0644, st _size=0, ...}) = 0 close(32) = 0 stat64("/usr/X11R6/lib/X11/fonts/local", {st_mode=S_IFDIR|0755, st_size=144, ... }) = 0 open("/usr/X11R6/lib/X11/fonts/Type1/fonts.cache-1", O_RDONLY) = 32 stat64("/usr/X11R6/lib/X11/fonts/Type1", {st_mode=S_IFDIR|0755, st_size=2008, .. .}) = 0 stat64("/usr/X11R6/lib/X11/fonts/Type1/fonts.cache-1", {st_mode=S_IFREG|0644, st _size=14302, ...}) = 0 close(32) = 0 stat64("/usr/X11R6/lib/X11/fonts/Type1", {st_mode=S_IFDIR|0755, st_size=2008, .. .}) = 0 open("/usr/X11R6/lib/X11/fonts/75dpi/fonts.cache-1", O_RDONLY) = 32 stat64("/usr/X11R6/lib/X11/fonts/75dpi", {st_mode=S_IFDIR|0755, st_size=79368, . ..}) = 0 stat64("/usr/X11R6/lib/X11/fonts/75dpi/fonts.cache-1", {st_mode=S_IFREG|0644, st _size=0, ...}) = 0 close(32) = 0 stat64("/usr/X11R6/lib/X11/fonts/75dpi", {st_mode=S_IFDIR|0755, st_size=79368, . ..}) = 0 open("/usr/X11R6/lib/X11/fonts/util/fonts.cache-1", O_RDONLY) = 32 stat64("/usr/X11R6/lib/X11/fonts/util", {st_mode=S_IFDIR|0755, st_size=600, ...} ) = 0 stat64("/usr/X11R6/lib/X11/fonts/util/fonts.cache-1", {st_mode=S_IFREG|0644, st_ size=0, ...}) = 0 close(32) = 0 stat64("/usr/X11R6/lib/X11/fonts/util", {st_mode=S_IFDIR|0755, st_size=600, ...} ) = 0 open("/usr/X11R6/lib/X11/fonts/misc/fonts.cache-1", O_RDONLY) = 32 stat64("/usr/X11R6/lib/X11/fonts/misc", {st_mode=S_IFDIR|0755, st_size=13400, .. .}) = 0 stat64("/usr/X11R6/lib/X11/fonts/misc/fonts.cache-1", {st_mode=S_IFREG|0644, st_ size=0, ...}) = 0 close(32) = 0 stat64("/usr/X11R6/lib/X11/fonts/misc", {st_mode=S_IFDIR|0755, st_size=13400, .. .}) = 0 open("/usr/X11R6/lib/X11/fonts/TTF/fonts.cache-1", O_RDONLY) = 32 stat64("/usr/X11R6/lib/X11/fonts/TTF", {st_mode=S_IFDIR|0755, st_size=584, ...}) = 0 stat64("/usr/X11R6/lib/X11/fonts/TTF/fonts.cache-1", {st_mode=S_IFREG|0644, st_s ize=7684, ...}) = 0 close(32) = 0 stat64("/usr/X11R6/lib/X11/fonts/TTF", {st_mode=S_IFDIR|0755, st_size=584, ...}) = 0 open("/usr/X11R6/lib/X11/fonts/CID/fonts.cache-1", O_RDONLY) = 32 stat64("/usr/X11R6/lib/X11/fonts/CID", {st_mode=S_IFDIR|0755, st_size=176, ...}) = 0 stat64("/usr/X11R6/lib/X11/fonts/CID/fonts.cache-1", {st_mode=S_IFREG|0644, st_s ize=0, ...}) = 0 close(32) = 0 stat64("/usr/X11R6/lib/X11/fonts/CID", {st_mode=S_IFDIR|0755, st_size=176, ...}) = 0 open("/usr/X11R6/lib/X11/fonts/encodings/large/fonts.cache-1", O_RDONLY) = 32 stat64("/usr/X11R6/lib/X11/fonts/encodings/large", {st_mode=S_IFDIR|0755, st_siz e=504, ...}) = 0 stat64("/usr/X11R6/lib/X11/fonts/encodings/large/fonts.cache-1", {st_mode=S_IFRE G|0644, st_size=0, ...}) = 0 fstat64(32, {st_mode=S_IFREG|0644, st_size=0, ...}) = 0 mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x40 235000 read(32, "", 4096) = 0 close(32) = 0 munmap(0x40235000, 4096) = 0 writev(5, [{"\22\0\6\0\24\0 \2G\1\0\0;\1\0\0\10\0\310\0\0\0\0\0\22\0"..., 1096}, {"RENDER", 6}, {"\0\0", 2}], 3) = 1104 read(5, "\34(#\1\24\0 \2G\1\0\0h\206_\216\0\0 \2\0\0\0\0\330\365"..., 32) = 32 read(5, "\34($\1\24\0 \2\'\0\0\0h\206_\216\0\0\0\0h\206_\216\330"..., 32) = 32 read(5, "\34(%\1\24\0 \2%\0\0\0h\206_\216\0\0\0\0h\206_\216\330"..., 32) = 32 read(5, "\34(&\1\24\0 \2$\0\0\0h\206_\216\0\0\0\0h\206_\216\330"..., 32) = 32 read(5, "\34(\'\1\24\0 \2&\1\0\0h\206_\216\0\0\0\0h\206_\216\330"..., 32) = 32 read(5, "\23\0B\0010\0 \0020\0 \2\0\23\f\10\340\220t\0100\0 \2\1"..., 32) = 32 read(5, "\22\0F\0010\0 \0020\0 \2\0\23\f\10\340\220t\0100\0 \2\1"..., 32) = 32 read(5, "\1\220H\1\0\0\0\0\1\236\0\277\0\0\0\0\0\0\0\0\0\0\0\000"..., 32) = 32 write(5, "\236\0\3\0\0\0\0\0\7\0\0\0\236\1\1\0", 16) = 16 read(5, "\1\264I\1\0\0\0\0\0\0\0\0\2\0\0\0\0\0\0\0\0\0\0\0000\232"..., 32) = 32 read(5, "\1\231J\1\240\0\0\0\20\0\0\0\1\0\0\0\7\0\0\0\20\0\0\0\206"..., 32) = 32 read(5, "g\0\0\0\1\1\r\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\1\0\0\0\0\0"..., 640) = 640 --- SIGSEGV (Segmentation fault) --- unlink("/home/dg1qkt/.mozilla/default/rsxhxi6v.slt/lock") = 0 _exit(11) = ?
I have seen this before. Hard to track though ... usually: 1) after running prelink 2) after merging galeon (weird, yeah) 3) after installing glibc (prob related to prelink here). Usually a remerge fixed it for me. Now, 1) Do you use prelink ? 2) Please include the info of 'emerge info'
[General comment:] Both Mozilla and Galeon (using the stock Mozilla libs) started just hanging and burning CPU for me some time ago. I deleted config files, etc, no luck, and started using Phoenix instead. Eventually I got the mozilla CVS tree so I could build Phoenix. I just ran Galeon, it found the mozilla/dist/bin 1.3 libs, worked fine, no hang, no cpu burn.
*** Bug 13375 has been marked as a duplicate of this bug. ***
And a just remerge of mozilla (not galeon mind you) ?
I do not use prelink and it is not installed. I tried galeon because mozilla did not work, but it segfaults, too. :-( I already tried a remerge of Mozilla, but the problem is still the same. So i fetched the source from Mozilla.org and compiled it myself. It runs fine. I used ACCEPT_KEYWORDS="x86" to get a perl version which works with the current openjade version. Here the Information you requested: ----snip------------------ helena root # emerge info Portage 2.0.46-r4 (default-x86-1.4, gcc-3.2.1, glibc-2.3.1-r2) ================================================================= System uname: 2.4.20 i686 Intel(R) Pentium(R) 4 CPU 2.53GHz USE="x86 oss apm avi crypt cups encode gif jpeg libg++ libwww mikmod mmx mpeg ncurses nls pdflib png qtmt quicktime spell truetype xml2 xmms xv zlib alsa gdbm berkdb slang readline arts svga java guile X sdl gpm tcpd pam ssl perl python esd imlib oggvorbis gnome gtk qt kde motif opengl cdr -3dnow sse" ARCH="x86" COMPILER="gcc3" CHOST="i686-pc-linux-gnu" CFLAGS="-march=pentium4 -O3 -pipe" CXXFLAGS="-march=pentium4 -O3 -pipe" ACCEPT_KEYWORDS="x86" CONFIG_PROTECT="/etc /var/qmail/control /usr/share/config /usr/kde/2/share/config /usr/kde/3/share/config /usr/X11R6/lib/X11/xkb:/usr/kde/3/share/config:/usr/share/config" CONFIG_PROTECT_MASK="/etc/gconf /etc/env.d" MAKEOPTS="-j2" JDK_HOME="" JAVA_HOME="/opt/blackdown-jre-1.3.1" AUTOCLEAN="yes" SYNC="rsync://rsync.gentoo.org/gentoo-portage" GENTOO_MIRRORS="http://www.ibiblio.org/pub/Linux/distributions/gentoo" ----------snap----------------------------------------------- By the way, please ignore my bad english
I also want to report that downloading the source and building it myself with the following parameters allows mozilla 1.2.1 to work correctly (except that mozgest does not work) - ./configure --prefix=/usr --enable-xft --enable-toolkit-gtk --disable-debug --enable-optimize=-Os --enable-strip My emerge info is Portage 2.0.46-r4 (default-x86-1.4, gcc-3.2.1, glibc-2.3.1-r3) ================================================================= System uname: 2.5.54 i686 Intel(R) Xeon(TM) CPU 1.70GHz USE="x86 oss 3dnow apm avi crypt gif jpeg libg++ mikmod mmx mpeg ncurses nls pdf lib png qtmt quicktime spell xml2 xmms xv zlib gtkhtml gdbm berkdb slang readlin e arts tetex bonobo tcltk java guile X sdl gpm tcpd pam libwww ssl perl python e sd imlib gnome gtk qt motif opengl cdr -alsa -svga cups aalib truetype ldap pda kde encode gtk2 oggvorbis" ARCH="x86" COMPILER="gcc3" CHOST="i686-pc-linux-gnu" CFLAGS="-march=pentium3 -Os -pipe -fexpensive-optimizations -fpeephole -fgcse -f omit-frame-pointer -funroll-loops" CXXFLAGS="-march=pentium3 -Os -pipe -fexpensive-optimizations -fpeephole -fgcse -fomit-frame-pointer -funroll-loops" ACCEPT_KEYWORDS="x86 ~x86" CONFIG_PROTECT="/etc /var/qmail/control /usr/kde/2/share/config /usr/kde/3/share /config /usr/kde/3.1/share/config /usr/share/config" CONFIG_PROTECT_MASK="/etc/gconf /etc/env.d" MAKEOPTS="-j2" JDK_HOME="/opt/blackdown-jdk-1.4.1_beta" JAVA_HOME="/opt/blackdown-jdk-1.4.1_beta" AUTOCLEAN="yes" SYNC="rsync://rsync.gentoo.org/gentoo-portage" GENTOO_MIRRORS="http://www.ibiblio.org/pub/Linux/distributions/gentoo"
Kai, can you try with -Os or -O0 like in CFLAGS/CXXFLAGS instead of -O3 like comment #6 suggests ? Maybe also edit the ebuild, and change the '--enable-optimize="-O2" \' in the configure line to: --enable-optimize="-Os" \ or --enable-optimize="-O0" \ If you could test both, and maybe also -O2, I will appreciate. Thanks.
I tried out the change the ebuild to use -Os instead of -O2 and now mozilla works perfectly from the ebuild, including mozgest. So it seems like -O2 was the problem for atleast my seg fault. Maybe this can be a global change? Naren
Ok, great. Thanks guys :)
narendra, sorry, but could you try with -O1 ?
Ok tried the ebuild with -O1 and that works great as well. So seems that so far only -O2 is a problem. Naren
Another update. I tried -O3 and this works as well. Seems like only -O2 causes problems. Actually I am going to recompile with -O2 to verify. Naren
Ok This is bizarre. I got the mozilla emerge to run with all combination of optimizations. Here is how I did it. I built and installed the source from mozilla.org without the ebuild. Then I ran mozilla, and it ran fine and just quit it. I deleted /usr/bin/mozilla /usr/bin/mozilla-config /usr/lib/mozilla-1.2.1 and /usr/lib/mozilla. Then I build mozilla-1.2.1-r4 using the ebuild. I did not change anything in the ebuild. Then I run mozilla from the ebuild and it works great. So I do not know how this works. Any explainations for this one?
I fixed the ebuild to force -O1, so it could be why it works now when you remerge it ...
Should be fixed now. If somebody still have issues, reopen.
I have those startup crashes aswell... only on one machine, not on the other. I see no other difference between them but the processor (p4 works, athlon xp not). The latest version I could run on the athlon box was 1.1. I nearly start believing in supernatural mozillas, because when I emerged 1.2.1 first (last monday) I got a version that crashed as root, started as normal user but did not show me my start page. When I selected "about" in the help menu, it told me it was version 1.1. When I unmerge mozilla and delete the stuff in /usr/lib/mozilla and re-emerge 1.2.1, it does not even start. -mcpu=i686 -Os did not help. Does anyone have an idea what my problem is? My "emerge info", if that is helpful: Portage 2.0.46-r9 (default-x86-1.4, gcc-3.2.1, glibc-2.3.1-r2) ================================================================= System uname: 2.4.20 i686 AMD Athlon(tm) Processor GENTOO_MIRRORS="http://linux.rz.ruhr-uni-bochum.de/download/gentoo-mirror" CONFIG_PROTECT="/etc /var/qmail/control /usr/kde/2/share/config /usr/X11R6/lib/X11/xkb /usr/kde/3.1/share/config /usr/kde/3/share/config /usr/share/config" CONFIG_PROTECT_MASK="/etc/gconf /etc/env.d" PORTDIR="/usr/portage" DISTDIR="/usr/portage/distfiles" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR_OVERLAY="" USE="x86 oss apm avi crypt cups encode gif jpeg libg++ mikmod mpeg ncurses nls pdflib png quicktime spell truetype xml2 xmms xv zlib gdbm berkdb slang readline arts tetex svga java mysql X sdl gpm tcpd pam libwww ssl perl python esd imlib oggvorbis gtk qt kde motif opengl mozilla doc gnome qtmt mmx 3dnow oci8" COMPILER="gcc3" CHOST="i686-pc-linux-gnu" CFLAGS="-march=athlon-xp -O3 -pipe" CXXFLAGS="-march=athlon-xp -O3 -pipe" ACCEPT_KEYWORDS="x86" MAKEOPTS="-j2" AUTOCLEAN="yes" SYNC="rsync://linux.rz.ruhr-uni-bochum.de/gentoo-portage" FEATURES="sandbox ccache"
I'd like to report that mozilla-1.2.1-r5 segfaults on startup for me unless I move both of: /usr/X11R6/lib/X11/fonts/75dpi/helvR10.pcf.gz /usr/X11R6/lib/X11/fonts/100dpi/helvR08.pcf.gz out of the way. Seems *marginally* related to this bug so I report it here. geep root # emerge info Portage 2.0.46-r9 (default-x86-1.4, gcc-3.2.1, glibc-2.2.5-r7,2.3.1-r2) ================================================================= System uname: 2.4.19-gentoo-r10 i686 AMD Athlon(tm) MP 1800+ GENTOO_MIRRORS="http://www.ibiblio.org/pub/Linux/distributions/gentoo" CONFIG_PROTECT="/etc /var/qmail/control /usr/kde/2/share/config /usr/X11R6/lib/X11/xkb /usr/kde/3/share/config /usr/share/config" CONFIG_PROTECT_MASK="/etc/gconf /etc/env.d" PORTDIR="/usr/portage" DISTDIR="/mnt/auto/qusbox/distfiles" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR_OVERLAY="/home/jrray/portage_newebuilds" USE="x86 oss 3dnow apm avi crypt encode gif jpeg libg++ mikmod mmx mpeg ncurses pdflib png qtmt quicktime spell truetype xml2 xmms xv zlib alsa gdbm berkdb slang readline arts tetex bonobo svga tcltk java guile mysql postgres X sdl gpm tcpd pam libwww ssl perl python esd imlib oggvorbis gtk qt motif opengl mozilla cdr scanner dvd -cups -nls -gnome -kde" COMPILER="gcc3" CHOST="i686-pc-linux-gnu" CFLAGS="-march=athlon-mp -O3 -pipe -fomit-frame-pointer" CXXFLAGS="-march=athlon-mp -O3 -pipe -fomit-frame-pointer" ACCEPT_KEYWORDS="x86" MAKEOPTS="-j3" AUTOCLEAN="yes" SYNC="rsync://rsync.gentoo.org/gentoo-portage" FEATURES="sandbox ccache"
I recently received an email that would crash Mozilla until I also moved /usr/X11R6/lib/X11/fonts/misc/6x12-ISO8859-1.pcf.gz out of the way. Something's up with moz and fonts...
to all people having problems with: - kde showing white boxes instead of fixed fonts - gnome displaying many error message boxes on startup - mozilla crashing right at startup (i.e. not starting at all) i recommend the following: emerge -u /usr/portage/media-libs/freetype/freetype-2.1.3-r2.ebuild for me, this fixed all these issues at once :) consequently, i guess they all were related to problems handling compressed fonts.. ps. maybe someone should link this hint from all other current bug reports dealing with the above mentioned problems. can't hurt ;)
btw. these issues were observed with even the most recent stable versions of the mentioned apps (kde 3.1, gnome 2.2-r2, mozilla 1.2.1-r5). nb. after emerging (the so far marked 'unstable') freetype, reboot (or restart the runlevel, xfree or whatever. for me, rebooting did the trick and i did not mind as this wasn't a 300-days-uptime-yeah-i'm-so-373373-server but my every-day workstation ;)
The suggested freetype update did not help me, but deleting the two fonts as suggested in Comment #17 worked fine for me. Really really strange...
Right, I guess we could nuke those for now by default ? At least until a new freetype version are out ....
Finally got that sucker. Seems there was a ~/.fonts-cache-1 with wrong contents. I googled for it and found some hints on the garnome list - maybe someone who understands what they are talking about should check this. I deleted that file, started mozilla 1.2.4-r5 and it works - even with those fonts from comment #17 in place. Could someone who had those problems please verify this? I'm just trying it on a different machine, so stay tuned.
Latest news from the other machine: emerge mozilla -> mozilla crashes. Removed .fonts-cache-1 -> still crashed. fc-cache as root -> works fine. Would be nice if someone could verify. Maybe if jrray puts his fonts back where they were and tries?
Verified. I had the problem as described in the original report, too. None of the tips stated above helped except for running fc-cache. Works fine now. Maybe this can be included in the mozilla ebuild (at least as a postinstall message) ?
I have the same problem, however running fc-cache does /not/ fix the problem. It is a font related issue, however. I have a directory where I've copied my truetype fonts from Windows, and the problem only started since I've been using those fonts. If I move the fonts out of the way and run fc-cache, the problem disappears. So, in my case, Mozilla just isn't playing nice with some of the fonts in that directory. We can probably generalize it to say that Mozilla has trouble with some fonts, and crashes rather than handling this problem correctly.
Ok semms to me like this bugreport really is related to several font-related bugs. Running fc-cache solved it for me but I didn't change anything in any font direcory. I removed some of the fonts as suggested in the tips above (without running fc-cache, maybe it was meaningsless without that but I am not a font expert and I don't know about that) but I restored them since it didn't help. Maybe those are all different problems with the same symptom.
Well, I've got a solution for my problem: I noticed that other gtk applications weren't crashing with my Windows fonts active, they just didn't use them. So, I ran a GTK app from a terminal window, and got a ton of "Couldn't open file for font X" errors. I checked the permissions, and sure enough, no read permission. I've added read permission to all my fonts for all users, and now Mozilla doesn't crash anymore. What I think is that in Mozilla's code, instead of handling problems with a font gracefully (printing an error message to stderr, for example), it just crashes and burns. Presumeably, if you can find where that is in the code, you could probably wrap up all these problems at one. Graeme
*** Bug 18267 has been marked as a duplicate of this bug. ***
Since 1.2.1 is no longer relevant (1.4 is out, folks). This bug should be removed from bugzilla. AFAIK this is a non-issue with both the stable and unstable Mozilla builds of late.
This was actually fixed by azarah
it seems the newer moz ebuilds still set -O to 1 and refer to this bug, should they?
