Chrome Version: chromium-bin-4.0.295.0_p35884 and earlier was the same with firefox-3.5 (not firefox-3.0 and not firefox-bin) What steps will reproduce the problem? 1. Just start chromium or chromium-bin 2. Wait for about 5 seconds What is the expected result? Should just keep running What happens instead? The application runs for about 5 seconds and then gets killed by SIGABRT Please provide any additional information below. Attach a screenshot if possible. Gentoo (updated regularyly) Vanilla 2.6.32.2 (and several earlier versions) nVidia Corporation G72 [GeForce 7500 LE] (rev a1) nvidia-drivers-190.53-r1 (and a couple of earlier versions) Dualhead with Xinerama xorg-server-1.6.5-r1 (was the same on 1.5) I've attached the strace output, gdb backtrace and emerge --info output at http://code.google.com/p/chromium/issues/detail?id=32081 Reproducible: Always Steps to Reproduce: 1. Start 2. Wait Actual Results: Crash Expected Results: Run I've already posted this bug at http://code.google.com/p/chromium/issues/detail?id=32081 and were the bug was classified as a "NSS Gentoo porting bug"
emerge --info needs to be attached to bug report here. You will also need to provide nss/nspr version being used. I have not been able to duplicate this so please be complete.
[ebuild R ] dev-libs/nspr-4.8 USE="debug -ipv6" 0 kB [ebuild R ] dev-libs/nss-3.12.5 USE="-utils" 0 kB
http://chromium.googlecode.com/issues/attachment?aid=-6287086432198115199&name=emerge-info.txt http://chromium.googlecode.com/issues/attachment?aid=-4633507144528995837&name=strace-chromium-bin.txt http://chromium.googlecode.com/issues/attachment?aid=-6352066652704513804&name=gdb-chromium.txt and last but not least http://code.google.com/p/chromium/issues/detail?id=32081
If you would please rebuild xulrunner/firefox both and retest. I have a feeling your hitting an alighnment bug in gcc.
The bug doesn't show up with freshly recompiled xulrunner-1.9.2-r1 and firefox-3.6-r1 any more. chromium and chromium-bin are still getting Jory A. Pratt: thanks for the tip I've followed your suggestion and recompiled xulrunner and firefox. Firefox works fine now. I've tried the same earlier and it didn't do the trick. I've tried to switch to xulrunner-bin eather with no result. So now it might be just the update, which made the things change. (Firefox stopped to work with the update from 3.0 to 3.5; firefox-bin kept working all the time). Any suggestions to solve the chromium problem?
If you would disable all optimizations other then -O2 for your cflags and recompile chromium, if this resolves your problem, then we definitely have hit the alignment bug in 4.3.x gcc. I am gonna temp, assign this to mozilla team, if problem is resolved we will re-assign to gcc team.
Jory A. Pratt: 1. firefox works fine for now (see comment 5) so I can't reproduce the bug with it; 2. to get the gdb backtrace I had to recompile chromium, nss and nspr disabling all the optimizations, which I did: http://chromium.googlecode.com/issues/attachment?aid=-6352066652704513804&name=gdb-chromium.txt 3. I've already posted this bug on the chromium bug tracker: http://code.google.com/p/chromium/issues/detail?id=32081 - the last proposition ther was, I should debug myself
LD_LIBRARY_PATH="/usr/lib/nspr:/usr/lib/nss:/usr/lib:/lib" chromium and let me know what you get, if it fails also please run in gdb with same library path.
yes, thank you with LD_LIBRARY_PATH set correctly chromium and chromium-bin work fine. who is responsible for this variable being not set properly? I'll post this result to chromium bug tracker God save the Queen
(In reply to comment #9) > yes, thank you > > with LD_LIBRARY_PATH set correctly chromium and chromium-bin work fine. > who is responsible for this variable being not set properly? > > I'll post this result to chromium bug tracker > > God save the Queen > I am responsible for it, but it is handled via env.d entries, If you would run env-update && source /etc/profile and retest would be appreciated. The 08{nss,nspr} in /etc/env.d should update ld.so.conf properly to find the libs.
(In reply to comment #10) > I am responsible for it, but it is handled via env.d entries, If you would run > env-update && source /etc/profile and retest would be appreciated. The > 08{nss,nspr} in /etc/env.d should update ld.so.conf properly to find the libs. > it did work $ cat /etc/env.d/08nspr LDPATH=/usr/lib/nspr $ cat /etc/env.d/08nss LDPATH=/usr/lib/nss thanks again
As reporter has reported issue is all resolved I am closing.
