On Sun, Jun 17, 2012 at 3:45 AM, Richard W.M. Jones <rjones@xxxxxxxxxx> wrote: > I would also suspect code generation or GC. Did you try asking on > upstream Coq / OCaml mailing lists? No, I haven't. But see below. > Having said that, relatively long-running programs are working OK for > me. How about on i386? I haven't had a single problem on x86_64; all of the failures I've seen have been on i386. Yesterday, I tried building again to see if anything had changed. It had. The plugin compilations were failing because, apparently, -fPIC had not been passed to the compiler when building the shared objects. I poked at it for a little bit before giving up. Today I tried building again to try to localize that problem ... and now I'm getting this: bin/coqtop.opt -boot -nois -compile theories/Init/Peano File "/home/jamesjer/rpmbuild/BUILD/coq-8.3pl4/theories/Init/Peano.v", line 78, character 28-29: Error: No interpretation for numeral 0. make[1]: *** [theories/Init/Peano.vo] Error 1 This is getting weirder and weirder. I don't even know where to start looking. I'd believe my i386 VM is damaged somehow, except I'm seeing these kinds of problems on koji builds, too. Have you tried running nontrivial OCaml programs on an i386 platform? Thanks, -- Jerry James http://www.jamezone.org/ -- devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/devel
