This is the mail archive of the gcc-patches@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]

[gfotran, committed] Fix PR 20467


I committed this patch (+ the testcase from the PR) to the mainline under the
obviously correct rule after bubblestrapping and testing.  The frontend
allowed dummy arguments to be defined as statement functions, which caused a
segfault during resolution.

I will commit this to 4.0 once testing finishes.

- Tobi

Index: ChangeLog
===================================================================
RCS file: /cvs/gcc/gcc/gcc/fortran/ChangeLog,v
retrieving revision 1.352
diff -u -p -r1.352 ChangeLog
--- ChangeLog   14 Mar 2005 15:23:38 -0000      1.352
+++ ChangeLog   14 Mar 2005 19:49:58 -0000
@@ -1,3 +1,9 @@
+2005-03-14  Tobias Schl"uter  <tobias.schlueter@physik.uni-muenchen.de>
+
+       PR fortran/20467
+       * symbol.c (check_conflict): A dummy argument can't be a statement
+       function.
+
 2005-03-14  Zdenek Dvorak  <dvorakz@suse.cz>

        * fortran/trans-intrinsic.c (gfc_conv_intrinsic_ishft): Convert
Index: symbol.c
===================================================================
RCS file: /cvs/gcc/gcc/gcc/fortran/symbol.c,v
retrieving revision 1.26
diff -u -p -r1.26 symbol.c
--- symbol.c    24 Feb 2005 18:26:27 -0000      1.26
+++ symbol.c    14 Mar 2005 19:49:58 -0000
@@ -369,6 +369,7 @@ check_conflict (symbol_attribute * attr,
        {
        case PROC_ST_FUNCTION:
          conf2 (in_common);
+         conf2 (dummy);
          break;

        case PROC_MODULE:


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