Return to BSD News archive
Path: sserve!newshost.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!pacbell.com!ihnp4.ucsd.edu!mvb.saic.com!MathWorks.Com!news.kei.com!bloom-beacon.mit.edu!ai-lab!life.ai.mit.edu!mycroft From: mycroft@duality.gnu.ai.mit.edu (Charles Hannum) Newsgroups: comp.os.386bsd.development Subject: Re: Coexisting on the PC Date: 11 Mar 1994 19:51:39 GMT Organization: MIT Artificial Intelligence Lab Lines: 33 Message-ID: <MYCROFT.94Mar11145139@duality.gnu.ai.mit.edu> References: <CLz6un.A9G@rex.uokhsc.edu> <2l3nvj$fpc@u.cc.utah.edu> <1994Mar10.090638.1364@softwords.bc.ca> <1994Mar11.075510.10894@softwords.bc.ca> NNTP-Posting-Host: duality.gnu.ai.mit.edu In-reply-to: jnemeth@cue.bc.ca's message of Fri, 11 Mar 94 07:55:10 GMT In article <1994Mar11.075510.10894@softwords.bc.ca> jnemeth@cue.bc.ca (John Nemeth) writes: All PC stuff should be removed from the platform independent parts of the kernel. I just looked for `#if.*i386' in my NetBSD-current kernel source tree. I found: kern/init_main.c:#if defined(i386) || defined(mac68k) || defined(amiga) kern/init_main.c:#if !defined(i386) && !defined(mac68k) && !defined(amiga) kern/subr_mcount.c:#if defined(i386) kern/subr_mcount.c:#if defined(i386) sys/disklabel.h:#ifdef i386 sys/disklabel.h:#ifdef i386 vm/vm_glue.c:#if defined(i386) || defined(pc532) vm/vm_glue.c:#if defined(i386) || defined(pc532) Arguably, we might want a machine/disklabel.h, but I don't think it would be worthwhile to split subr_mcount.c. The init_main.c #if's are because the timer must be set earlier on some machines so that delay() will work. The vm_glue.c thing is because the page tables are in the user's vm_map, and we must be careful not to inherit them across fork(). There were others, but they have all been removed. -- - Charles Hannum NetBSD group Working ports: i386, hp300, amiga, sparc, mac68k, pc532. In progress: pmax, sun3.