--- Comment #7 from Steve Kargl <sgk at troutmask dot apl.washington.edu> ---
On Wed, Mar 14, 2018 at 02:41:19AM +0000, kargl at gcc dot gnu.org wrote:
> > Yep, gfortran is missing a simplification. When
> > simplification of size(a) occurs, the int(1) is
> > still an expression with EXPR_FUNCTION.
> Testing a patch.
Bummer. The patch causes 2 regressions.
FAIL: gfortran.dg/pr84734.f90 -O (test for errors, line 3)
FAIL: gfortran.dg/typebound_proc_32.f90 -O (test for excess errors)
The former should fail, but doesn't. The latter
should pass, but fails.