On 2008/11/12 21:43 +0100, Jan Nieuwenhuizen wrote: > John Mandereau: > > CPP='x86_64-linux-gcc -E -x c-header' > > /home/lilydev/git/newgub/target/linux-64/build/glibc-core-2.3/elf/ld-linux-x86-64.so.2 > > --library-path > > /home/lilydev/git/newgub/target/linux-64/build/glibc-core-2.3:/home/lilydev/git/newgub/target/linux-64/build/glibc-core-2.3/math:/home/lilydev/git/newgub/target/linux-64/build/glibc-core-2.3/elf:/home/lilydev/git/newgub/target/linux-64/build/glibc-core-2.3/dlfcn:/home/lilydev/git/newgub/target/linux-64/build/glibc-core-2.3/nss:/home/lilydev/git/newgub/target/linux-64/build/glibc-core-2.3/nis:/home/lilydev/git/newgub/target/linux-64/build/glibc-core-2.3/rt:/home/lilydev/git/newgub/target/linux-64/build/glibc-core-2.3/resolv:/home/lilydev/git/newgub/target/linux-64/build/glibc-core-2.3/crypt:/home/lilydev/git/newgub/target/linux-64/build/glibc-core-2.3/linuxthreads > > > > /home/lilydev/git/newgub/target/linux-64/build/glibc-core-2.3/sunrpc/rpcgen > > -Y ../scripts -c rpcsvc/bootparam_prot.x -o > > /home/lilydev/git/newgub/target/linux-64/build/glibc-core-2.3/sunrpc/xbootparam_prot.T > > make[4]: *** > > [/home/lilydev/git/newgub/target/linux-64/build/glibc-core-2.3/sunrpc/xbootparam_prot.stmp] > > Floating point exception > > Ugh. Is this reproducible? What if you run the command manually? Make sure > to set > the path, ie, run > > PATH=/home/lilydev/git/newgub/target/linux-64/root/usr/cross/bin:$PATH > CPP='x86.... rpcgen -Y ...
If I run the ld-linux-x86-64.so.2 command above manually with the PATH you suggested, it exits successfully. I tried to rebuild glibc-core with "rm ...status; make lilypond" twice, and I even did "rm -rf target/linux-64; make lilypond", but it didn't help, I got the same crash again. Unlike first time I got this crash, I ran the compilations on an idle system with Firefox and my email-client opened. I'll try to call make commands manually later... if I manage to reproduce GUB in the shell. > The only halfway relevant thing I find is > > http://bugs.gentoo.org/show_bug.cgi?id=45035 I guess this is not really related. > Not nice. Do you have enough ram/swapspace? How's your hardware? 4 GB RAM, Intel Core 2 Duo at 3,17 GHz, Fedora 9 64-bit, Linux 2.6.26.6. I've never encountered random freezes or crashes, so I guess my RAM is OK. I don't know any easy way to detect FPU/CPU hardware problems, but I've also done some basic signal processing computing and didn't get any strange result. > What if you add > > def force_sequential_build (self): > return True > > to the relevant classes in glibc.py [and or] glibc-core.py? It doesn't help, a similar crash happens, with exactly the same floating point exception error message. This crash seems not to be related with the CPU load. Best, John _______________________________________________ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
