This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug fortran/64980] [5.0 Regression] ICE in trans-expr.c


https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64980

Mikael Morin <mikael at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mikael at gcc dot gnu.org

--- Comment #4 from Mikael Morin <mikael at gcc dot gnu.org> ---
This cures it:

Index: trans-expr.c
===================================================================
--- trans-expr.c    (rÃvision 220514)
+++ trans-expr.c    (copie de travail)
@@ -4537,6 +4537,7 @@ gfc_conv_procedure_call (gfc_se * se, gfc_symbol *
                      fsym->attr.optional
                      && e->expr_type == EXPR_VARIABLE);
             }
+#if 0
           else if (e->ts.type == BT_CLASS && fsym
                && fsym->ts.type == BT_CLASS
                && !CLASS_DATA (fsym)->as
@@ -4587,6 +4588,7 @@ gfc_conv_procedure_call (gfc_se * se, gfc_symbol *
             }
             }
           else
+#endif
             gfc_conv_expr_reference (&parmse, e);

           /* Catch base objects that are not variables.  */

So yes, r218584 is the cause.
The problem comes from the interface mapping machinery, which locally changes
THIS from class MULTI_TRAPEZIUM_T to MULTI_TRAPEZIUM_NODE_CLASS_T in the call
get_d_value_array(node).
The code from r218584 creates a temporary using the declared type
MULTI_TRAPEZIUM_T, which doesn't match what interface mapping expects.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]