Re: Fixed point support for GCC cross compiler

shafi wrote:
> Someone told me that one can add support in two ways.
> 1. Include the support through the compiler (change the internals).
> 2. Outside the compiler, through a library. (library support)


My original comments assumed you didn't want to redefine floats or
doubles. If you're willing to redefine floats by linking to a new
fixed-point library that might work.


You could give people a choice of linking to your fixed point lib, or
to a standard floating point lib. You'd have to go with a fixed number
of decimal places though, or else the front-end of the compiler would
have to be changed to support a new syntax.


I can't give you the details on how to do it because I'm not an expert
on the GNU tools. I think there's a forum specifically for gcc people
and that might be a good place to ask this question. Look around at the
main GNU site for a link to their forum.


Eric

Return to the comp.compilers page.
Search the comp.compilers archives again.