On Sat, Feb 21, 2009 at 08:54:58AM -0800, Jesse Keating wrote: > If your package was built in x86_64 but with access to the 32bit > content, what is the resulting rpm, is it x86_64, or ix86? If it's It contains ix86 code, so that would be better. At the moment my mock build makes a .x86_64.rpm file: http://homes.merjis.com/~rich/mingw/fedora-rawhide/x86_64/RPMS/darwinx-odcctools-590.36-0.20060413.6.fc11.x86_64.rpm > ix86, the "best" thing would likely to exclusivearch it to ix86, and > file a ticket for a multilib hardcoding so that it'll be copied into the > x86_64 repo, like wine is. I also hate these hacks, as I'd rather see > effort put into making the code work, if the software itself is > generally useful. If its not generally useful, then does it really > belong in Fedora, let alone hacked around to force it to be multilib ? Well, generally useful code may still have a strange internal implementation ... Look at emacs ... Rich. -- Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones Read my OCaml programming blog: http://camltastic.blogspot.com/ Fedora now supports 68 OCaml packages (the OPEN alternative to F#) http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list