> From: [log in to unmask] (Malcolm Cohen)
> [log in to unmask] said:
> > Finally, I just got email from Richard Shapiro, a compiler developer
> > at SGI. In response to a request of mine, he has added a compiler feature
> > to SGI's f90 which should go at least some way toward solving the
> > problem:
> >
> > > From [log in to unmask] Fri Feb 5 10:55:07 1999
> > > You will be happy to know that in Fortran 90 v7.3 (about to go to Beta) we
> > > have added an option to force allocations larger than a certain size (both
> > > for AUTOMATIC arrays and compiler generated temps) to go on the heap
> > > instead of the stack.
> >
> > This feature originated in Cray's f90.
> Well I don't doubt that SGIf90's feature may have originated in Cray's
> f90, but some other vendors have had this for some time; as it happens
> NAGWare f90 has always allocated large dynamic objects on the heap
> precisely because the default stack size on Unix etc. is so small.
Yes, and the Absoft f90 compiler (at least the LINUX version) appears
to do this as well.
I don't know who did it first, and sorry I left the impression that
Cray did (which they might have -- I just don't know).
FWIW, the Cray/SGI implementation, AFAIK, allows the user to set
a threshold at compile time. Auto. objects larger than the threshold
will be allocated on the heap and smaller objects on the stack.
-P.
*** "Freedom's just another word for nothing left to lose." (B. Yeltsin)***
*********** Peter S. Shenkin; Schrodinger, Inc.; (201)433-2014 ************
*********** [log in to unmask]; http://www.schrodinger.com ***********
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|