http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50684
--- Comment #10 from janus at gcc dot gnu.org 2011-10-13 12:27:58 UTC --- (In reply to comment #9) > I think the comments of the following program are correct; however, the > compilers I tested either rejected all - or they compiled it without warning. > Thus, I might have misunderstood something. I would agree with your comments. > On the other hand, if I change > "call move_alloc" to "deallocate", gfortran follows my invalid/OK pattern. With move_alloc, gfortran-4.5 rejects only (1), gfortran-4.6 rejects all of them. > I have not checked whether it works with your patch, but I think it still > rejects (3). It does. In that sense I agree that the patch is not completely correct. > call move_alloc (tx, ta) ! Invalid (1) There 'tx' here should be an 'x', I guess. > call move_alloc (x%a, a) ! Invalid (2) This one corresponds to the original test case.
