[Bug c++/25861] New: tree check fail at c-common.c:2430

[Bug c++/25861] New: tree check fail at c-common.c:2430

[dcb314 at hotmail dot com]

I just tried to compile package gnugk-2.2.3-2 from Suse Linux with a recent
GNU C compiler version 4.2 snapshot 20060114. 

The compiler snapshot said

/home/dcb/gnu/42-20060114/results/bin/g++ -g -O3 -Wall -fmessage-length=0
-DHAS_RADIUS=1 -DHAS_MSG_NOSIGNAL=1  -D'MANUFACTURER=GNU'
-D'PROGRAMMNAME=Gatekeeper' -DPTRACING  -D_REENTRANT -Wall  -O2 -g
-fmessage-length=0 -D_FORTIFY_SOURCE=2 -fexceptions -DP_64BIT
-I/usr/share/pwlib//include -DPTRACING
-I/usr/include/openh323 -DMAJOR_VERSION=2 -DMINOR_VERSION=2 -DBUILD_NUMBER=3
-Os   -felide-constructors -x c++ -c main.cxx -o obj_linux_x86_64_r/main.o
/usr/include/ptlib/notifier.h: In constructor
'PNotifierFunction::PNotifierFunction(void*)':
/usr/include/ptlib/notifier.h:42: internal compiler error: tree check: expected
tree that contains 'decl with visibility' structure, have 'parm_decl'  in
c_common_truthvalue_conversion, at c-common.c:2430
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.
make: *** [obj_linux_x86_64_r/main.o] Error 1

Preprocessed source code attached.


-- 
           Summary: tree check fail at c-common.c:2430
           Product: gcc
           Version: 4.2.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: dcb314 at hotmail dot com
  GCC host triplet: x86_64-suse-linux


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25861

[Bug c++/25861] tree check fail at c-common.c:2430

[dcb314 at hotmail dot com]

------- Comment #1 from dcb314 at hotmail dot com  2006-01-19 18:48 -------
Created an attachment (id=10676)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=10676&action=view)
C++ source code


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25861

[Bug c++/25861] tree check fail at c-common.c:2430

[pinskia at gcc dot gnu dot org]

------- Comment #2 from pinskia at gcc dot gnu dot org  2006-01-19 18:53 -------
Reducing.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25861

[Bug c++/25861] tree check fail at c-common.c:2430

[pinskia at gcc dot gnu dot org]

------- Comment #3 from pinskia at gcc dot gnu dot org  2006-01-19 18:54 -------
This has been faling since at least 20051219. 


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25861

[Bug c++/25861] tree check fail at c-common.c:2430

[dcb314 at hotmail dot com]

------- Comment #4 from dcb314 at hotmail dot com  2006-01-19 19:17 -------
(In reply to comment #3)
> This has been faling since at least 20051219. 

I rather suspect there would be a useful job for someone
to take each weekly snapshot of gcc 4.2 and make
sure it compiles some recentish distribution of Linux
from one of the big players like Suse or Redhat.

It would be an easy for anyone with access to a compile
farm.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25861

[Bug c++/25861] tree check fail at c-common.c:2430

[pinskia at gcc dot gnu dot org]

------- Comment #5 from pinskia at gcc dot gnu dot org  2006-01-19 20:16 -------
Semi reduced:
void f(void);
class PNotifierFunction 
{
  public:
    PNotifierFunction(
      void * obj
    ) { 
    void * object;
      object = ((&(obj)&&(obj)!=__null)?(obj): (f(),(obj)));
     }
};


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25861

[Bug c/25861] [4.2 Regression] tree check fail at c-common.c:2430

[pinskia at gcc dot gnu dot org]

------- Comment #6 from pinskia at gcc dot gnu dot org  2006-01-19 20:18 -------
Reduced C testcase:
int f(void *a)
{
  return !(&a);
}


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
          Component|c++                         |c
     Ever Confirmed|0                           |1
   GCC host triplet|x86_64-suse-linux           |
           Keywords|                            |ice-on-valid-code
   Last reconfirmed|0000-00-00 00:00:00         |2006-01-19 20:18:57
               date|                            |
            Summary|tree check fail at c-       |[4.2 Regression] tree check
                   |common.c:2430               |fail at c-common.c:2430
   Target Milestone|---                         |4.2.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25861

[Bug c/25861] [4.2 Regression] tree check fail at c-common.c:2430

[pinskia at gcc dot gnu dot org]

------- Comment #7 from pinskia at gcc dot gnu dot org  2006-01-19 20:20 -------
Caused by:
2005-12-14  Ben Elliston  <bje@au.ibm.com>

        * c-common.c (c_common_truthvalue_conversion): Generalise warning
        for addresses converted to booleans; not just function addresses.
        * c-typeck.c (build_binary_op): Warn for address comparisons which
        can never be NULL (eg. func == NULL or &var == NULL).
        * common.opt (Walways-true): New option.
        * c-opts.c (c_common_handle_option): Set it with -Wall.
        * doc/invoke.texi: Document it.


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |bje at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25861

[Bug c/25861] [4.2 Regression] tree check fail at c-common.c:2430

[pinskia at gcc dot gnu dot org]

------- Comment #8 from pinskia at gcc dot gnu dot org  2006-01-23 19:17 -------
This is easy to fix.


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |pinskia at gcc dot gnu dot
                   |dot org                     |org
             Status|NEW                         |ASSIGNED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25861

[Bug c/25861] [4.2 Regression] tree check fail at c-common.c:2430

[pinskia at gcc dot gnu dot org]

------- Comment #9 from pinskia at gcc dot gnu dot org  2006-01-24 16:16 -------
The full patch (with some cleanup) is:
Index: c-common.c
===================================================================
--- c-common.c  (revision 110173)
+++ c-common.c  (working copy)
@@ -2456,25 +2456,26 @@ c_common_truthvalue_conversion (tree exp

     case ADDR_EXPR:
       {
-       if (DECL_P (TREE_OPERAND (expr, 0))
-           && !DECL_WEAK (TREE_OPERAND (expr, 0)))
+       tree inner = TREE_OPERAND (expr, 0);
+       if (DECL_P (inner)
+           && (TREE_CODE (inner) == PARM_DECL || !DECL_WEAK (inner)))
          {
            /* Common Ada/Pascal programmer's mistake.  We always warn
               about this since it is so bad.  */
            warning (OPT_Walways_true, "the address of %qD, will always
evaluate as %<true%>",
-                    TREE_OPERAND (expr, 0));
+                    inner);
            return truthvalue_true_node;
          }

        /* If we are taking the address of an external decl, it might be
           zero if it is weak, so we cannot optimize.  */
-       if (DECL_P (TREE_OPERAND (expr, 0))
-           && DECL_EXTERNAL (TREE_OPERAND (expr, 0)))
+       if (DECL_P (inner)
+           && DECL_EXTERNAL (inner))
          break;

-       if (TREE_SIDE_EFFECTS (TREE_OPERAND (expr, 0)))
+       if (TREE_SIDE_EFFECTS (inner))
          return build2 (COMPOUND_EXPR, truthvalue_type_node,
-                        TREE_OPERAND (expr, 0), truthvalue_true_node);
+                        inner, truthvalue_true_node);
        else
          return truthvalue_true_node;
       }


I am testing this right now.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25861

[Bug c/25861] [4.2 Regression] tree check fail at c-common.c:2430

[pinskia at gcc dot gnu dot org]

------- Comment #10 from pinskia at gcc dot gnu dot org  2006-01-25 20:40 -------
Patch posted:
http://gcc.gnu.org/ml/gcc-patches/2006-01/msg01762.html


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                URL|                            |http://gcc.gnu.org/ml/gcc-
                   |                            |patches/2006-
                   |                            |01/msg01762.html
           Keywords|                            |patch


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25861

[Bug c/25861] [4.2 Regression] tree check fail at c-common.c:2430

[pinskia at gcc dot gnu dot org]

------- Comment #11 from pinskia at gcc dot gnu dot org  2006-01-26 14:59 -------
Subject: Bug 25861

Author: pinskia
Date: Thu Jan 26 14:59:26 2006
New Revision: 110256

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=110256
Log:
2006-01-26  Andrew Pinski  <pinskia@physics.uc.edu>

        PR C/25861
        * c-common.c (c_common_truthvalue_conversion) <case ADDR_EXPR>:
        Use a new variable, inner.  PARM_DECLs are always non-weak.

2006-01-26  Andrew Pinski  <pinskia@physics.uc.edu>

        PR C/25861
        * gcc.c-torture/compile/pr25861.c: New test.


Added:
    trunk/gcc/testsuite/gcc.c-torture/compile/pr25861.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/c-common.c
    trunk/gcc/testsuite/ChangeLog


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25861

[Bug c/25861] [4.2 Regression] tree check fail at c-common.c:2430

[pinskia at gcc dot gnu dot org]

------- Comment #12 from pinskia at gcc dot gnu dot org  2006-01-26 14:59 -------
Fixed.


-- 

pinskia at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25861