On Wed, Jan 01, 2003 at 03:00:09PM +0100, Andy Polyakov wrote: > > > I mean I see no reason why > > > we should feel discouraged to recognize that it's sparc64. What does > > > uname -m return on your machine? But most of all I wonder do we really > > > have to have separate lines for NetBSD, FreeBSD and OpenBSD? Why can't > > > we unify those lines to asteriskBSD-platform? In which case sparc64 > > > recognition becomes a must as other flavors do support non-UltraSPARCs. > > > > Well, the information is currently obtained via a sysctl, which > > returns (on sparc) > > > > enigma# sysctl hw.model > > hw.model: Sun Microsystems UltraSparc-II > > Does it mean that 'uname -m' doesn't work? Never did? Never will? Going > to seize any time soon? But even if sysctl is the only option, we still > can map *UltraSparc* to sparc64 and I think we should favor it as it's > denoting ABI and is therefore less confusing.
uname -m returns "sparc64". I assume that the use of the hw.model sysctl is so different x86 CPUs can be distinguished for purposes of asm support (they all return "i386" from uname -m). > As for asteriskBSD. Can you test if './Configure OpenBSD-sparc64' works > on your FreeBSD machine? It does at this point in time (I copied the OpenBSD-sparc64 target to FreeBSD-sparc64), but of course there are no guarantees they will not diverge at some point. Kris
msg14715/pgp00000.pgp
Description: PGP signature
