QuantLib-0.9.0-5.fc9.src.rpm Failed To Build From Source against the rawhide tree. See http://fedoraproject.org/wiki/FTBFS for more information.
Created attachment 301254 [details] root.log.bz2 root.log for x86_64
Created attachment 301255 [details] build.log.bz2 build.log for x86_64
OK, I officially have no idea why this one failed to build on your side. The build log says: g++: Internal error: Segmentation fault (program cc1plus) Please submit a full bug report. See <http://bugzilla.redhat.com/bugzilla> for instructions. make[1]: *** [riskstats.o] Error 1 Quantlib is a huge memory and resource hog to build, perhaps the system wasn't qualified to do the job? ;) This one built fine on March 24th (with gcc 4.3.0-3): http://koji.fedoraproject.org/packages/QuantLib/0.9.0/5.fc9/data/logs/x86_64/build.log The root.log doesn't show what gcc you're using, so I suppose it is possible that a gcc bug has crept in between gcc 4.3.0-3 and now. I'm throwing this one to NEEDINFO, I'd appreciate if you could give this another pass.
I filed 441090: RFE: capture core files into $resultdir because I didn't capture the segfault. The failure occurred on a build system with 4GB of RAM and 2GB of swap, 2 dual-core 3GHz processors, after 44 minutes running. I'm re-running the build.
my build failed again in exactly the same place. It's using gcc-4.3.0-6 in the chroot. Hoping it was just the system, I built it on another system, this time an 8GB 2 quad-core processors. It failed after about 10 minutes. This sure smells like a gcc bug.
Changing version to '9' as part of upcoming Fedora 9 GA. More information and reason for this action is here: http://fedoraproject.org/wiki/BugZappers/HouseKeeping
Fixed in rawhide (gcc 4.3.1-2). http://koji.fedoraproject.org/koji/buildinfo?buildID=53701