[Bug fortran/82250] Fortran OpenACC acc_on_device early folding

[tschwinge at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82250

--- Comment #12 from Thomas Schwinge <tschwinge at gcc dot gnu.org> ---
(In reply to Tobias Burnus from comment #11)
> > Not the whole commit, but only one small hunk of it was reverted
> 
> Which is enough to prevent the folding when the compiler has not been
> configured for offloading. If I recall correctly: before your commit, both
> the host compiler with offloading configured and a compiler not configured
> for offloading folded it to a constant. But with reverting, only a compiler
> configured for offloading folds it as it calls the folding function twice
> while without offloading configured, it is old called once.

I (still) don't follow.  Could you please provide a concrete test case and
actual vs. expected outcome.