(j3.2006) binding labels and global identifiers

[Malcolm Cohen]

I wrote:
> one might ask why malloc
> (and sin and cos and ...) are even in the user namespace at all, but they were
> originally in that namespace

Van Snyder claimed:
>They weren't in the namespace of the Univac 1100 compilers (well, Univac
>didn't have malloc).

Univac didn't have a C compiler then.

If they did have a C compiler at some point, then sin and cos et al were in the 
namespace.

>So "they were originally in the namespace" seems to be a feeble reason
>for trying to cope with Stu's mistake.

Nope.  (not feeble, not a reason, not coping, not Stu, not a mistake).  It is 
just a fact.

>If processors feel the need to produce a linker name that is different
>from the Fortran name (in the no-BIND(C) case), they should use a name
>that cannot be a valid C name

Right, so your assertion is that we are not permitted to use C in our 
implementation.

This is ridiculous and annoying.

The standard has no business poking its nose into implementation technologies.

Cheers,
-- 
................................Malcolm Cohen, Nihon NAG, Tokyo.