I understand that the dynamic library loader can select processor-specific libraries at runtime in /usr/lib/i686 or similar. That makes sense. But:
1. Does this happen automatically, or does the executable require code to look there?
2. Can we actually compile two libraries (i586 and i686) and add them both to the rpm packages, or does that have to be done explicitly by the build process?
3. Is there some tutorial for programmers on doing this type of thing?
It just seems like such an easy solution to our mmx problem, I wonder why it's not more often employed.
Thanks, Austin
--
Austin Acton Hon.B.Sc.
Synthetic Organic Chemist, Teaching Assistant
Department of Chemistry, York University, Toronto
MandrakeClub Volunteer (www.mandrakeclub.com)
homepage: www.groundstate.ca