-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 08 Mar 2003 15:17:08 +0100, Josip Gracin wrote:
> > Can you recompile the kernel with Red Hat's config? > > Nope. .oO > > What steps are necessary to reproduce it? > > cd /usr/src/linux-2.4 > make mrproper > cp configs/kernel-2.4.18-athlon.config .config > make oldconfig > make dep > make bzImage > /* each file gets the following warning: > 'cc1: warning: -malign-functions is obsolete, use -falign-functions' */ > > and finally > > gcc -D__KERNEL__ -I/usr/src/linux-2.4.18-24.8.0/include -Wall > -Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common > -fomit-frame-pointer -pipe -mpreferred-stack-boundary=2 -march=i686 > -malign-functions=4 -nostdinc -I > /usr/lib/gcc-lib/i386-redhat-linux/3.2/include -DKBUILD_BASENAME=md > -DEXPORT_SYMTAB -c md.c > cc1: warning: -malign-functions is obsolete, use -falign-functions > md.c: In function `do_md_run': > md.c:1745: internal error: Segmentation fault > Please submit a full bug report, > with preprocessed source if appropriate. > > I can repeat the whole procedure (starting with make mrproper) and it > breaks at the exact same place. However, if I use different configs, it > breaks at different places, but for the same config always at the same > place. Just to make one thing sure, if you don't restart at the beginning, but re-run "make bzImage" several times, does it always fail at the same line same file? > > Most of the time, segmentation faults during kernel compilation are > > due to unstable hardware. Running "make bzImage" (for instance) > > again compiles a bit more, but then segfaults again. > > I understand that, but in my case the error is always at the same place. > I try to compile while running different other applications but nothing > changes. There're two things left you could try. First is to compile the kernel after entering "export LC_ALL=C". Second is to verify integrity of all your installed packages as "root": # rpm -qa | xargs -n 1 -t rpm -V &> rpm-Va.txt # less rpm-Va.txt If that doesn't report any damaged core components, check out https://bugzilla.redhat.com. There are several bug reports on segfaults during kernel compilation, but most of them are flagged "NEEDINFO" and are unanswered by the original reporters. I would also assume that the reporters see hardware instability. However, if you have a clearly reproducable test-case, it would be worthing submitting detailed info into Red Hat's bugzilla system. Be told that Jakub Jelinek from Red Hat or another developer is likely to ask for the compiler output and file created when adding "-save-temps -v" to the compiler command-line. > > ?? I don't see what that bug report has to do with GCC. > > Nothing, except that it makes me more nervous since I neither have a working > GCC nor a working libc. What else is there in a UNIX system? Well, what else doesn't compile for you? - -- -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.7 (GNU/Linux) iD8DBQE+ainw0iMVcrivHFQRAmJGAJwJSc+7I95DIpL4DzcKcrRG1X8oBgCggPQ9 uCA6x7Ujm2h/zks67k6LDOk= =/R7M -----END PGP SIGNATURE----- -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
