Re: XL Fortran on PowerPC- or x86-arch

In article <1hb9re1.1vppg61l9iy0dN%nospam@xxxxxxxxxxxxx>,
Richard E Maine <nospam@xxxxxxxxxxxxx> wrote:

You might have run into a distinction between
whether or not the compiler short-circuits expression evaluation.

A programming error of expecting short-circuiting is hard to detect at
compile-time. We recently switched our default to short-circuiting
because we were getting bug reports saying "this works on Intel, why
doesn't it work with Pathscale", and we determined that it didn't hurt
performance significantly in our performance test suite. This is how
de-facto standards arise...

-- greg
(employed by, not speaking for, PathScale)
.