pcm@iwarpj.intel.com (Phil C. Miller) (07/22/89)
In article <2916@ast.cs.vu.nl> ast@cs.vu.nl (Andy Tanenbaum) writes: >In article <4660@crash.cts.com> jca@pnet01.cts.com (John C. Archambeau) writes: >>Your insight along with anyone else's will be appreciated as always... > >I regard the 286 as a transient. It and its weird little segments will >eventually pass from view. I don't want to do any work on the compiler >or system to cater to it. Too much trouble. Two points: 1. AT class machines using the 286 are in EXTREMELY widespread use, both in the University community and in the hacker community. Being one of the people in the latter group, I understand what you're saying: the 286 is not everyone's favorite architecture. On the other hand, not everyone can afford to go out and buy a 386 box. I think that the 286 will be with us for a long time. 2. You probably don't need to work on a 286 compiler. Several people in the Minix community, including me, have offered in the past to contribute their time to produce a compiler which understands the 286. The offer is still open. I'm sure that enough interested parties could be found to port the libraries, commands, etc. >Andy Tanenbaum (ast@cs.vu.nl) Phil Miller {...}!tektronix!ogccse!pcm
henry@utzoo.uucp (Henry Spencer) (07/23/89)
In article <4686@omepd.UUCP> pcm@iwarpj.UUCP (Phil C. Miller) writes: > ... I think that the 286 will be with us for a long time. So will the pdp8. (Yes, there are still pdp8s aplenty in the world.) But nobody writes software for them any more. >2. You probably don't need to work on a 286 compiler. Several people in the > Minix community, including me, have offered in the past to contribute > their time to produce a compiler which understands the 286. The offer is > still open. I'm sure that enough interested parties could be found to > port the libraries, commands, etc. The trouble is that "port" in this context does not mean "clean up", it means "change" (some would say "break"). The problem with the 286 is not that its segments are ugly -- almost everything about the x86 family is ugly -- but that programs cannot just rely on the compiler to sort it all out. Unless you accept massive inefficiency, EVERY PROGRAM that wants to use them has to be munged to know about them and manage them. This is an abomination, and Andy is doing the right thing in refusing to have anything to do with it. -- 1961-1969: 8 years of Apollo. | Henry Spencer at U of Toronto Zoology 1969-1989: 20 years of nothing.| uunet!attcan!utzoo!henry henry@zoo.toronto.edu