Summary: | mail-client/evolution-2.10.2 fails to load components when compiled with hardened profile | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Gunther Gruber <pitfall66> |
Component: | Hardened | Assignee: | GNOME Office (OBSOLETE) <gnome-office+disabled> |
Status: | RESOLVED DUPLICATE | ||
Severity: | minor | CC: | fjsanchez, pitfall66 |
Priority: | High | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- |
Description
Gunther Gruber
2007-08-15 22:41:49 UTC
This ain't exactly a hardened bug perse. I'd say this user is hitting the same problem as described in bug #141448 Just to be sure to understand. When compiling evolution with a non hardened profile, evolution is fully working but not if you compile it with a hardened profile, right ? How do you change between hardened and non hardened btw ? Profiles like this (hardened, selinux, regular) are not meant to be changed to one another like this and could lead to multiple breakage. could you look at bug #141448 and tell us if it's similar and if the informations there helps you fix evolution ? The bug http://bugs.gentoo.org/show_bug.cgi?id=141448 is indeed the same. When I said I switched to a non hardened profile I meant that I switch the compiler. gcc-config -l shows : [1] i686-pc-linux-gnu-3.4.6 * [2] i686-pc-linux-gnu-3.4.6-hardenednopie [3] i686-pc-linux-gnu-3.4.6-hardenednopiessp [4] i686-pc-linux-gnu-3.4.6-hardenednossp [5] i686-pc-linux-gnu-3.4.6-vanilla [6] i686-pc-linux-gnu-4.1.2 where 1 is the compiler with all the hardened flags set and 5 and 6 are the compilers without any hardened flag. I think I used 6 for evolution and it worked. (But maybe it was 5). If its important I give it a retry. (hardened for gcc-4.1.2 is masked by the way) Cant http://bugs.gentoo.org/show_bug.cgi?id=141448 reproduce if the rebind fix for evolution works because "elfkickers" won't compile on my system, maybe that is another issue. I got the same warnings from the other bug and this one with the LD_BIND_NOW variable set so I'm marking this as a duplicate. Thanks for reporting *** This bug has been marked as a duplicate of bug 141448 *** FYI: Switching compilers is not an ideal solution at all. It's only one subset flag -z now that gets passed to the linker when using a hardened compiler that exposes this problem. Suggested work around for the moment would be simply to do what I did in http://bugs.gentoo.org/show_bug.cgi?id=141448#c24 *** Bug 195350 has been marked as a duplicate of this bug. *** |