Re: [Gm2] REAL precision

[Gaius Mulley]

SiTex Graphics <address@hidden> writes:

> Hi Gaius,
>
> By default GM2 translates the REAL type as a double precision number.
> Is there a way I can build the compiler with REAL=float, and
> LONGREAL=double?  Or perhaps persuade you to make REAL=float the
> default bahavior? That's been the convention for every other M2
> compiler I've used...

Hi Scott,

it probably isn't that difficult to make the changes to the compiler.
We would need a configure option to alter the default.  Nearly all the
changes occur in gm2/gccgm2.c (although there might be a few libraries
which need to pick up the change).

The reason for the current default is that gm2 supports
LONGREAL/REAL/SHORTREAL which map onto C's long double, double, float.

How important is this configuration?  I'm currently in the middle of
implementing multidimensional dynamic arrays - which I believe is the
final ISO feature that is required.  Thereafter implement five more
library ISO modules (fix loads of bugs) and I propose to release 1.0.

Is the REAL/LONGREAL float/double configuration useful enough to make
it into 1.0 or just afterwards?

regards,
Gaius