This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: GNU Fortran90?
- To: gcc at gcc dot gnu dot org
- Subject: Re: GNU Fortran90?
- From: Andrew Vaught <andy at xena dot eas dot asu dot edu>
- Date: Sat, 25 Nov 2000 14:02:50 -0700 (MST)
>I feel we have to tread carefully here ... Do these people assume that
>- because Fortran 9x supports array syntax - they can automagically use
>the AltiVec instructions of the G4 ? 1)
>
>Note that Andy Vaught decided months ago to decompose all array
>arithmetic into its consituent loops in g95 because that makes code
>generation much easier.
>
>This means that to be able to use AltiVec, we have to build a
>vectorising pass into the middle-end of GCC (and then g77 code would
>profit as well).
This isn't quite correct. Vector expressions are stored as such in
g95's intermediate structures. If there is a mechanism in the GBE for
implementing a vector expression then we can use it. If not, we have to
generate the tree structures with loops that are equivalent to the vector
expression. Another possibility is generating tree structures that take
advantage of multiple processors via a standard thread interface.
Andy
----------------- XOLD(K,IC,I)=
Andy Vaught .... DO ITERS=1, 10 XOLD(K,IC,I)
andy@xena.eas.asu.edu | | /CALLMSOLVE(A,B,X,I,ITERS,TOL)+(RANNYU(0)
Arizona State University ======|WRITE(6,'(I5,2X,F12.6)')ITERS,TOL -HALF)
Tempe, Arizona USA OOOOOO \ENDDORETURN PARAMETER(ZERO=1.D0)*TENTH*DELTA