Am 29.01.2014 09:39, schrieb Steve Beattie: > The most proper way that I can see to address this would be to rely on > the default dpkg-buildflags to get the basic level of protections. Then > to get all the protections, build depend on hardening-wrapper and export > DEB_BUILD_HARDENING=1 in debian/rules. I've attached a debdiff that I've > verified builds on all available architectures for ureadahead, since > that package is also hitting this issue.
yes that was the fix I had in mind. > The least invasive workaround would be to export MALLOC_CHECK=2 at build > time (i.e. in debian/rules), as this causes glibc to abort without > attempting to produce a backtrace when it detects internal malloc > corruption. This unfortunately still leaves configure believing that > 'gcc -static' doesn't work, but it at least causes builds not to hang. sure, that would mimic the behaviour we did see before the glibc update, but papering over the original issue. In practice this shouldn't be an issue because we only build static binaries in very few cases. -- You received this bug notification because you are a member of Ubuntu Desktop Bugs, which is subscribed to evolution-data-server in Ubuntu. https://bugs.launchpad.net/bugs/1266492 Title: ld:i386 crashes with -static -fPIE -pie To manage notifications about this bug go to: https://bugs.launchpad.net/eglibc/+bug/1266492/+subscriptions -- desktop-bugs mailing list desktop-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/desktop-bugs
