[Bug fortran/41121] New: compile-time error when building BLAS with -fimplicit-none

[Bug fortran/41121] New: compile-time error when building BLAS with -fimplicit-none

[barron dot bichon at swri dot org]

With version r150889, compiling several BLAS and LAPACK source files produce
errors with the '-fimplicit-none' flag. For instance, when building
http://www.netlib.org/blas/dgbmv.f I get the following error message:

> gfortran -c -fimplicit-none dgbmv.f
dgbmv.f:297.72:

      END
                                                                        1
Error: Symbol 'a1' at (1) has no IMPLICIT type
dgbmv.f:297.72:

      END
                                                                        1
Error: Symbol 'a2' at (1) has no IMPLICIT type
dgbmv.f:143.19:

      INTRINSIC MAX,MIN
                   1
Error: Function 'max' at (1) has no IMPLICIT type
dgbmv.f:143.23:

      INTRINSIC MAX,MIN
                       1
Error: Function 'min' at (1) has no IMPLICIT type

When '-fimplicit-none' is removed, the file compiles without error. Tested on
OS X and Linux.

(My first bug report, so apologies if I'm missing something.)


-- 
           Summary: compile-time error when building BLAS with -fimplicit-
                    none
           Product: gcc
           Version: 4.5.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: fortran
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: barron dot bichon at swri dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41121

[Bug fortran/41121] [4.5 Regression] compile-time error when building BLAS with -fimplicit-none

[jv244 at cam dot ac dot uk]

------- Comment #1 from jv244 at cam dot ac dot uk  2009-08-19 15:26 -------
a reduced testcase, but not quite covering the original

SUBROUTINE S1
 INTRINSIC MIN
 INTEGER :: I,J
 write(6,*) MIN(I,J)

END SUBROUTINE


-- 

jv244 at cam dot ac dot uk changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
      Known to fail|                            |4.5.0
      Known to work|                            |4.4.2
   Last reconfirmed|0000-00-00 00:00:00         |2009-08-19 15:26:19
               date|                            |
            Summary|compile-time error when     |[4.5 Regression] compile-
                   |building BLAS with -        |time error when building
                   |fimplicit-none              |BLAS with -fimplicit-none


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41121

[Bug fortran/41121] [4.5 Regression] compile-time error when building BLAS with -fimplicit-none

[burnus at gcc dot gnu dot org]

-- 

burnus at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.5.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41121

[Bug fortran/41121] [4.5 Regression] compile-time error when building BLAS with -fimplicit-none

[janus at gcc dot gnu dot org]

------- Comment #2 from janus at gcc dot gnu dot org  2009-08-20 07:45 -------
Curiously, this does not happen when using the IMPLICIT NONE statement, instead
of -fimplicit-none:

  IMPLICIT NONE
  INTRINSIC MIN
  INTEGER :: I,J
  print *,MIN(I,J)
END


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41121

[Bug fortran/41121] [4.5 Regression] compile-time error when building BLAS with -fimplicit-none

[janus at gcc dot gnu dot org]

------- Comment #3 from janus at gcc dot gnu dot org  2009-08-20 08:04 -------
Here's the fix:

Index: gcc/fortran/resolve.c
===================================================================
--- gcc/fortran/resolve.c       (revision 150933)
+++ gcc/fortran/resolve.c       (working copy)
@@ -10280,7 +10280,7 @@ resolve_symbol (gfc_symbol *sym)

   /* Resolve formal namespaces.  */
   if (sym->formal_ns && sym->formal_ns != gfc_current_ns
-      && !sym->attr.contained)
+      && !sym->attr.contained && !sym->attr.intrinsic)
     gfc_resolve (sym->formal_ns);

   /* Make sure the formal namespace is present.  */

Will commit as obvious after performing a regtest.


-- 

janus at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |janus at gcc dot gnu dot org
                   |dot org                     |
             Status|NEW                         |ASSIGNED
   Last reconfirmed|2009-08-19 15:26:19         |2009-08-20 08:04:37
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41121

[Bug fortran/41121] [4.5 Regression] compile-time error when building BLAS with -fimplicit-none

[jv244 at cam dot ac dot uk]

------- Comment #4 from jv244 at cam dot ac dot uk  2009-08-20 08:25 -------
(In reply to comment #3)
> Here's the fix:

does it fully fix the original, i.e. a1 and a2 shouldn't be warned for either.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41121

[Bug fortran/41121] [4.5 Regression] compile-time error when building BLAS with -fimplicit-none

[janus at gcc dot gnu dot org]

------- Comment #5 from janus at gcc dot gnu dot org  2009-08-20 08:59 -------
(In reply to comment #4)
> does it fully fix the original, i.e. a1 and a2 shouldn't be warned for either.

It does.

Btw these 'a1' and 'a2' don't appear anywhere in dgbmv.f. I think they are the
formal arguements of MIN or MAX.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41121

[Bug fortran/41121] [4.5 Regression] compile-time error when building BLAS with -fimplicit-none

[janus at gcc dot gnu dot org]

------- Comment #6 from janus at gcc dot gnu dot org  2009-08-20 09:33 -------
Subject: Bug 41121

Author: janus
Date: Thu Aug 20 09:33:01 2009
New Revision: 150957

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=150957
Log:
2009-08-20  Janus Weil  <janus@gcc.gnu.org>

        PR fortran/41121
        * resolve.c (resolve_symbol): Don't resolve formal_ns of intrinsic
        procedures.

2009-08-20  Janus Weil  <janus@gcc.gnu.org>

        PR fortran/41121
        * gfortran.dg/intrinsic_5.f90: New.


Added:
    trunk/gcc/testsuite/gfortran.dg/intrinsic_5.f90
Modified:
    trunk/gcc/fortran/ChangeLog
    trunk/gcc/fortran/resolve.c
    trunk/gcc/testsuite/ChangeLog


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41121

[Bug fortran/41121] [4.5 Regression] compile-time error when building BLAS with -fimplicit-none

[janus at gcc dot gnu dot org]

------- Comment #7 from janus at gcc dot gnu dot org  2009-08-20 09:36 -------
Fixed with r150957. Closing.


-- 

janus at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41121

[Bug fortran/41121] [4.5 Regression] compile-time error when building BLAS with -fimplicit-none

[hjl at gcc dot gnu dot org]

------- Comment #8 from hjl at gcc dot gnu dot org  2009-08-30 02:07 -------
Subject: Bug 41121

Author: hjl
Date: Sun Aug 30 02:06:32 2009
New Revision: 151218

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=151218
Log:
2009-08-29  H.J. Lu  <hongjiu.lu@intel.com>

        Backport from mainline:
        2009-08-26  H.J. Lu  <hongjiu.lu@intel.com>

        PR fortran/41162
        * gfortran.dg/pr41162.f: New.

        2009-08-26  Richard Guenther  <rguenther@suse.de>

        PR middle-end/41163
        * gcc.c-torture/compile/pr41163.c: New testcase.

        2009-08-25  Janus Weil  <janus@gcc.gnu.org>

        PR fortran/41139
        * gfortran.dg/proc_ptr_25.f90: New.
        * gfortran.dg/proc_ptr_comp_18.f90: New.
        * gfortran.dg/proc_ptr_comp_19.f90: New.

        2009-08-20  Michael Matz  <matz@suse.de>

        PR fortran/41126
        * gfortran.dg/pr41126.f90: New test.

        2009-08-20  Janus Weil  <janus@gcc.gnu.org>

        PR fortran/41121
        * gfortran.dg/intrinsic_5.f90: New.

        2009-08-19  Jason Merrill  <jason@redhat.com>

        PR c++/41120
        * g++.dg/other/gc4.C: New.

Added:
    branches/gcc-4_4-branch/gcc/testsuite/g++.dg/other/gc4.C
      - copied unchanged from r151217, trunk/gcc/testsuite/g++.dg/other/gc4.C
    branches/gcc-4_4-branch/gcc/testsuite/gcc.c-torture/compile/pr41163.c
      - copied unchanged from r151217,
trunk/gcc/testsuite/gcc.c-torture/compile/pr41163.c
    branches/gcc-4_4-branch/gcc/testsuite/gfortran.dg/intrinsic_5.f90
      - copied unchanged from r151217,
trunk/gcc/testsuite/gfortran.dg/intrinsic_5.f90
    branches/gcc-4_4-branch/gcc/testsuite/gfortran.dg/pr41126.f90
      - copied unchanged from r151217,
trunk/gcc/testsuite/gfortran.dg/pr41126.f90
    branches/gcc-4_4-branch/gcc/testsuite/gfortran.dg/pr41162.f
      - copied unchanged from r151216,
trunk/gcc/testsuite/gfortran.dg/pr41162.f
    branches/gcc-4_4-branch/gcc/testsuite/gfortran.dg/proc_ptr_25.f90
      - copied unchanged from r151217,
trunk/gcc/testsuite/gfortran.dg/proc_ptr_25.f90
    branches/gcc-4_4-branch/gcc/testsuite/gfortran.dg/proc_ptr_comp_18.f90
      - copied unchanged from r151217,
trunk/gcc/testsuite/gfortran.dg/proc_ptr_comp_18.f90
    branches/gcc-4_4-branch/gcc/testsuite/gfortran.dg/proc_ptr_comp_19.f90
      - copied unchanged from r151217,
trunk/gcc/testsuite/gfortran.dg/proc_ptr_comp_19.f90
Modified:
    branches/gcc-4_4-branch/gcc/testsuite/ChangeLog


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41121