Re: Ada's Slide To Oblivion ...

[Ian Wild <ian@cfmu.eurocontrol.be>]

Nick Roberts wrote:
>  
> Allow me to try to clarify. The C language requires (in practice if not
> strictly in theory) that all pointers fit into one machine word. On 32-bit
> architectures, this almost invariably forces the use of a 'flat' address
> space (just an offset, with no segment number or equivalent).


C, in practice as well as theory, has two flavours of pointer,
and there's no operation that will convert between code
pointers and data pointers.  You simply *can't* read or write
via code space pointers, nor can you call a data pointer.  (In
fact, it's not unusual for the two ranges of pointers, when viewed
as integers, to overlap, since "somewhere near the bottom" is
a pretty good place for both to start.)


> Which means
> that, for many architectures, the operating system cannot use segmentation
> (or other memory divisions) to detect a call or jump into read-write memory.

There's nothing in C that prevents split I/D spaces being
used if they're available, as witnessed by the fact that C
(and, indeed, Unix) ran on PDP-11s of yore.  Of course, if the
processor /can't/ separate instructions from data, it can't do
it for *any* language, so it'd be unfair to criticise C alone
for this.