[comp.lang.c] 80*86 vs. C?

gwyn@brl-smoke.ARPA (Doug Gwyn ) (08/20/87)

In article <2163@xanth.UUCP> kent@xanth.UUCP (Kent Paul Dolan) writes:
>I don't think it is a useful endeavor to labor at such length to embed
>the hardware design mistakes of the past in the C standards of the
>future.

Please explain how we're doing this.  I don't recall anything
in the draft proposed American National Standard for C that is
present solely because of segmented architectures.  Certainly
not the requirement that no data object be given an address
indistinguishable from a null pointer; that constraint is
logically necessary and has been with us for a long time.

Undoubtedly there are some 80*86 compilers that purport to be
C compilers but do not correctly implement anything that could
reasonably be called "C".  I suggest not buying them (even
better, let the vendors know why you don't buy such products).

drw@cullvax.UUCP (Dale Worley) (08/21/87)

gwyn@brl-smoke.ARPA (Doug Gwyn ) writes:
> In article <2163@xanth.UUCP> kent@xanth.UUCP (Kent Paul Dolan) writes:
> >I don't think it is a useful endeavor to labor at such length to embed
> >the hardware design mistakes of the past in the C standards of the
> >future.
> Please explain how we're doing this.

I suspect that kent wants:

All addresses are pointers into an enormous, linear address space.
Thus, any two pointers can be subtracted, one can point off the
beginning or end of a malloc'ed section of memory, etc.

All addresses are byte-addressed, and are the same size and alignment
as ints.

One can dereference NULL.

etc.

These are the same complaints that are voiced here all the time,
usually in the form: "Lots of [bad] code depends on this, so it should
work."  It is best summarized by the phrase "idiots who think all the
world's a Vax".

C is (or should be) a high-level language.  Hardware architecture
shouldn't leak into the code one writes, at least when the code
conforms to an ANSI standard.

Dale
-- 
Dale Worley	Cullinet Software		ARPA: cullvax!drw@eddie.mit.edu
UUCP: ...!seismo!harvard!mit-eddie!cullvax!drw
OS/2: Yesterday's software tomorrow	    Nuclear war?  There goes my career!