On Fri, 02 Oct 2015 09:53:03 -0400 Wayne Sallee <wa...@waynesallee.com> wrote:
OK, the backtrace shows that the actual illegal instruction happened earlier - there are several error handlers that come after the problem began and so we have to dissemble earlier. From the backtrace, it looks to me that this is the exact place the failure began: #8 0x00007fe021043dde in __gmpn_mul_1 () from /usr/lib/libgmp.so.10 Which is in libgmp as William Harrington suspected. mpn_mul_1 is known to use CPU-optimized assembly: http://lists.gforge.inria.fr/pipermail/ecm-discuss/2007-November/003969.html There they mention a --enable-fat gmp configure option that allows gmp to autodetect the CPU at run time. So that's one option. Let's see what the offending instruction is before passing judgment on what to do: Ok, try this chant: gdb -c core /usr/libexec/gcc/x86_64-unknown-linux-gnu/4.9.2/cc1plus (gdb) disassemble 0x00007fe021043dde,+8 Cheers, Mike -- http://lists.linuxfromscratch.org/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
