Newsgroups: comp.std.c Path: utzoo!henry From: henry@utzoo.uucp (Henry Spencer) Subject: fork/exec Message-ID: <1989Aug15.012607.4529@utzoo.uucp> Organization: U of Toronto Zoology References: <148@trigon.UUCP> <207600029@s.cs.uiuc.edu> <941@lakesys.UUCP> <2357@auspex.auspex.com> <5672@ficc.uu.net> Date: Tue, 15 Aug 89 01:26:07 GMT In article <5672@ficc.uu.net> peter@ficc.uu.net (Peter da Silva) writes: >Hmmm. Does POSIX specify that fork() is the process-creation mechanism? Yes. >... while the fork()-exec() pair is singularly elegant, it's not >implementable (without a massive number of kludges) in a wide variety of >operating systems: OS/9, VMS, RSX, AmigaOS... The technical term for this is "tough luck". POSIX is a *Unix* standard. The inability of defective operating systems to emulate it was (a) well known, and (b) not a consideration. On the hardware used by VMS and RSX, it is verifiably possible to implement fork(), since Unix runs on those machines. The same is probably true of the Amiga, given that the Amigoids assure us that the Amiga is superior to the Atari ST in every way :-), and Minix does fork() just fine on the ST. Dunno about OS/9, although given how much help Minix gets from the ST on fork() -- none -- it ought to be practical on 6809s as well, given enough memory. So doing fork() for the "operating systems" you name is just a Small Matter Of Programming for their maintainers. (That chortling sound you hear is all the Unix old-timers watching the frantic scramble for Unix compatibility by all the people who spent years sneering at Unix. Nyah nyah, we told you so! :-) :-) :-)) -- V7 /bin/mail source: 554 lines.| Henry Spencer at U of Toronto Zoology 1989 X.400 specs: 2200+ pages. | uunet!attcan!utzoo!henry henry@zoo.toronto.edu