At 1:12 PM -0800 3/16/04, Wes Peters wrote: >On Sunday 14 March 2004 05:34 pm, Stephen McKay wrote: > > > If the intention is to use the sparc conversion is as the > > template for architectures I care about then now is the > > first time I can contribute to improving the process. > >sparc64 is, afaik, the last 64 bit architecture to change >over to 64BTT. PowerPC is still 32-bTT, and I think that's a 64-bit architecture. But the people working on that have talked about switching to 64-bTT before PowerPC becomes a tier-1 platform. >The 32 bit architectures don't matter because the likelihood >of them surviving until when 32BTT becomes a problem is nil. >(That's probably true of the sparc64 architecture too, but >keeping sparc64 in line with the rest of the 64-bitters is >a good idea.) I should also note that a big reason I wanted sparc64 to be 64-bTT is that it's a big-endian architecture, so bugs in the handling of time_t will show up differently on it than on ia64 or amd64. By moving sparc64 to 64-bTT right now, we will notice bugs right now that we might not notice on the other platforms for quite some time -- even though the bug would exist on all 64-bTT platforms. [I may have mentioned that before. I forget if I did] -- Garance Alistair Drosehn = gad_at_gilead.netel.rpi.edu Senior Systems Programmer or gad_at_freebsd.org Rensselaer Polytechnic Institute or drosih_at_rpi.eduReceived on Wed Mar 17 2004 - 12:18:27 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:47 UTC