Hi all, here is a patch which does several things:
1) It fixes the original problem in the PR (cf. comments 0 and 6) by
adding code which checks if a generic interface has a specific
procedure of the same name (see resolve_procedure_interface).
2) It fixes other problems found along the way (cf. comment 7), which
are all due to the fact that the checks for interface declarations in
PROCEDURE statements came too early (checks for generics, statement
functions and intrinsics are moved from "match_procedure_interface" to
"resolve_procedure_interface"). For the intrinsics we do the same
trick as for PR51081, by setting the flavor at parsing stage and later
setting the INTRINSIC attribute at resolution stage.
3) It does minor cleanup related to "gfc_is_intrinsic", by moving some
checks into the routine, which were typically done before calling the
routine in a several places. (This is also related to the recent patch
for PR51081.) Note that 'use_assoc' is not sufficient to identify a
routine as non-intrinsic (apparently this was wrongly assumed in some
cases).
The patch was regtested successfully on x86_64-unknown-linux-gnu. Ok for trunk?
Cheers,
Janus
2012-07-31 Janus Weil <ja...@gcc.gnu.org>
PR fortran/42418
* decl.c (match_procedure_interface): Move some checks to
'resolve_procedure_interface'. Set flavor if appropriate.
* expr.c (gfc_check_pointer_assign): Cleanup of 'gfc_is_intrinsic'.
* intrinsic.c (gfc_is_intrinsic): Additional checks for attributes which
identify a procedure as being non-intrinsic.
* resolve.c (resolve_procedure_interface): Checks moved here from
'match_procedure_interface'. Minor cleanup.
(resolve_formal_arglist,resolve_symbol): Cleanup of
'resolve_procedure_interface'
(resolve_actual_arglist,is_external_proc): Cleanup of
'gfc_is_intrinsic'.
2012-07-31 Janus Weil <ja...@gcc.gnu.org>
PR fortran/42418
* gfortran.dg/proc_decl_29.f90: New.
pr42418_v4.diff
Description: Binary data
proc_decl_29.f90
Description: Binary data
