On Mon, May 11, 2009 at 04:40:37PM -0400, Tom spot Callaway wrote: > On 05/11/2009 03:55 PM, Adam Williamson wrote: > > On Mon, 2009-05-11 at 14:51 -0400, Josh Boyer wrote: > > > >> I would really like to see a proliferation of secondary arches in > >> Fedora, but I don't think 'workstation' is a viable usage model for > >> them to get started. Most will have to focus on the type of hardware > >> that actually sells for that arch, and yes I realize that can be at > >> odds with some of the directions Fedora is going. > > > > Semi-sidetrack here: Ubuntu has a secondary 'architecture' called LPIA > > which is, in fact, just an alternative set of compiler optimizations > > which they claim results in better code for 'netbook > > systems' (presumably meaning 'Atom CPUs'). Would that be something we > > could look into, or not interesting? > > Sure, I see no reason why someone couldn't take that on as a "secondary > arch". It would need a new arch identifier target for rpm to prevent > namespace collisions. > To be fair, the Ubuntu 'lpia' distribution actually performs /worse/ than normal compiled x86 code. (Since the GCC version they used didn't support scheduling for Atom. notting, others, and I have all done benchmarks when the i586 switch took place to see this.) Profiling to see if the atom target actually helps at all in gcc-4.4 thanks to Jakub backporting from GCC trunk would be a worthwhile endeavour though. regards, Kyle -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
