Hi all,
I'm working on implementation of the different FPU options for libgfortran, the runtime library for gfortran (e.g., on which floating-point exception do we want a SIGFPE to be raised, what rounding-mode does the user want, etc.).
There is nothing in standard C (or even POSIX) to do that, and we'll have to include some processor-specific code. How does the build system allow that? Can we have a system of MACRO defined depending on the target processor (something like TARGET_FPU_IS_PPC)?
PS: If I said something stupid or offended your sense of riguor, please don't be too harsh on someone new to this kind of problem.
FX
