[Bug c++/37217] New: ICE on Valid Code

[Bug c++/37217] New: ICE on Valid Code

[mckelvey at maskull dot com]

This code has been working forever:

$ g++ -v
Using built-in specs.
Target: i686-pc-cygwin
Configured with: /cygdrive/f/Home/cvsroot/gcc/configure --verbose
--enable-threads --disable-nls --disable-win32-registry
--enable-languages=c,c++
Thread model: posix
gcc version 4.4.0 20080823 (experimental) (GCC) 

User@MCKELVEY-XP ~/PD
$ uname -a
CYGWIN_NT-5.1 MCKELVEY-XP 1.5.25(0.156/4/2) 2008-06-12 19:34 i686 Cygwin



$ g++ -c -g -fno-elide-constructors      -DUSE_MUTEX=1  -pedantic-errors
-Werror -ansi -fno-common -Wall -Wold-style-cast -Wsign-promo -Wpointer-arith
-Wundef -Wwrite-strings -Winvalid-pch -Woverloaded-virtual -Wcast-qual -Wextra
-Wredundant-decls -Wshadow -Wcast-align -Wcomment -fstrict-aliasing -Winit-self
-Wmissing-include-dirs -Wswitch-default -Wswitch-enum -Wlogical-op -Wconversion
-Wmissing-declarations -Wunused -MMD   --save-temp   -fimplicit-templates -o
Types.o Types.cc
In file included from
/usr/local/lib/gcc/i686-pc-cygwin/4.4.0/../../../../include/c++/4.4.0/bits/localefwd.h:48,
                 from
/usr/local/lib/gcc/i686-pc-cygwin/4.4.0/../../../../include/c++/4.4.0/locale:46,
                 from Types.h:37,
                 from Types.cc:29:
/usr/local/lib/gcc/i686-pc-cygwin/4.4.0/../../../../include/c++/4.4.0/i686-pc-cygwin/bits/c++locale.h:
In function 'int std::__convert_from_v(int* const&, char*, int, const char*,
...)':
/usr/local/lib/gcc/i686-pc-cygwin/4.4.0/../../../../include/c++/4.4.0/i686-pc-cygwin/bits/c++locale.h:67:
internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.



$ alias CONFIGURECVS
alias CONFIGURECVS='/cygdrive/f/Home/cvsroot/gcc/configure     --verbose
--enable-threads --disable-nls     --disable-win32-registry
--enable-languages=c,c++'

User@MCKELVEY-XP ~/PD
$ alias BUILD
alias BUILD='nice make CFLAGS='\'''\'' BOOT_CFLAGS='\'''\'' LIBCFLAGS='\''-g
-O'\''     CXXFLAGS='\''-O'\'' LIBCXXFLAGS='\''-g -O'\''     bootstrap'


-- 
           Summary: ICE on Valid Code
           Product: gcc
           Version: 4.4.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: mckelvey at maskull dot com
 GCC build triplet: i686-pc-cygwin
  GCC host triplet: i686-pc-cygwin
GCC target triplet: i686-pc-cygwin


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217

[Bug c++/37217] ICE on Valid Code

[mckelvey at maskull dot com]

------- Comment #1 from mckelvey at maskull dot com  2008-08-24 01:41 -------
Created an attachment (id=16135)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16135&action=view)
Compressed preprocessed source


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217

[Bug c++/37217] ICE on Valid Code

[pinskia at gcc dot gnu dot org]

------- Comment #2 from pinskia at gcc dot gnu dot org  2008-08-24 22:20 -------
Reducing, -Wconversion causes the ICE.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217

[Bug c++/37217] [4.4 Regression] -Wconversion causes ICE with __builtin_strcmp with one char compare

[pinskia at gcc dot gnu dot org]

------- Comment #3 from pinskia at gcc dot gnu dot org  2008-08-24 22:27 -------
Confirmed, reduced testcase:
    int   __convert_from_v( )   {
     char* __old;
     if (__builtin_strcmp(__old, "C")) {}
}


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
  GCC build triplet|i686-pc-cygwin              |
   GCC host triplet|i686-pc-cygwin              |
 GCC target triplet|i686-pc-cygwin              |
           Keywords|                            |ice-on-valid-code
   Last reconfirmed|0000-00-00 00:00:00         |2008-08-24 22:27:02
               date|                            |
            Summary|ICE on Valid Code           |[4.4 Regression] -
                   |                            |Wconversion causes ICE with
                   |                            |__builtin_strcmp with one
                   |                            |char compare
   Target Milestone|---                         |4.4.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217

[Bug c++/37217] [4.4 Regression] -Wconversion causes ICE with __builtin_strcmp with one char compare

[burnus at gcc dot gnu dot org]

------- Comment #4 from burnus at gcc dot gnu dot org  2008-08-25 08:26 -------
My PR 37230 is presumably a duplicate of this PR. There, valgrind shows for
"gcc -Wconcersion":

==26156== Invalid read of size 2
==26156==    at 0x458558: conversion_warning (c-common.c:1571)
==26156==    by 0x4593F7: convert_and_check (c-common.c:1839)
==26156==    by 0x42BF71: build_binary_op (c-typeck.c:8434)

1571 is this line:
  1566    /* If any operand is artificial, then this expression was generated
  1567       by the compiler and we do not warn.  */
  1568    for (i = 0; i < expr_num_operands; i++)
  1569      {
  1570        tree op = TREE_OPERAND (expr, i);
  1571        if (DECL_P (op) && DECL_ARTIFICIAL (op))
  1572          return;
  1573      }

Working: 2008-08-20-r139252
Failing: 2008-08-21-r139367
Failing: 20080824 (experimental) [trunk revision 139535]


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217

[Bug c++/37217] [4.4 Regression] -Wconversion causes ICE with __builtin_strcmp with one char compare

[burnus at gcc dot gnu dot org]

------- Comment #5 from burnus at gcc dot gnu dot org  2008-08-25 09:31 -------
Manuel, I think it was caused by your check in for:

2008-08-17  Manuel Lopez-Ibanez  <manu@gcc.gnu.org>

        PR c++/35602
        * c-common.c (conversion_warning): Do not want for artificial
        expressions.

It fits time wise and valgrind points at lines you added there.


-- 

burnus at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217

[Bug c++/37217] [4.4 Regression] -Wconversion causes ICE with __builtin_strcmp with one char compare

[manu at gcc dot gnu dot org]

------- Comment #6 from manu at gcc dot gnu dot org  2008-08-25 10:22 -------
(In reply to comment #5)
> Manuel, I think it was caused by your check in for:
> 
> 2008-08-17  Manuel Lopez-Ibanez  <manu@gcc.gnu.org>
> 
>         PR c++/35602
>         * c-common.c (conversion_warning): Do not want for artificial
>         expressions.
> 
> It fits time wise and valgrind points at lines you added there.

I cannot reproduce this with any of the testcases in x86_64-unknow-linux-gnu
and that is all I have access to. Please attach your testcase from PR37230 and
if it could be reduced for x86_64-unknow-linux-gnu by Andrew or someone else,
that would be extremely helpful.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217

[Bug c++/37217] [4.4 Regression] -Wconversion causes ICE with __builtin_strcmp with one char compare

[burnus at gcc dot gnu dot org]

------- Comment #7 from burnus at gcc dot gnu dot org  2008-08-25 11:37 -------
Created an attachment (id=16144)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16144&action=view)
test2.i.gz

> I cannot reproduce this with any of the testcases in x86_64-unknow-linux-gnu
> and that is all I have access to. Please attach your testcase from PR37230 and
> if it could be reduced for x86_64-unknow-linux-gnu by Andrew or someone else,
> that would be extremely helpful.

Test case from PR37230. Run as:
$ gcc -Wconversion test2.i

That gives the mentioned segfault and the valgrind error on my
x86_64-unknow-linux-gnu.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217

[Bug c++/37217] [4.4 Regression] -Wconversion causes ICE with __builtin_strcmp with one char compare

[manu at gcc dot gnu dot org]

------- Comment #8 from manu at gcc dot gnu dot org  2008-08-25 12:36 -------
The fix is trivial:

Index: gcc/c-common.c
===================================================================
--- gcc/c-common.c      (revision 139373)
+++ gcc/c-common.c      (working copy)
@@ -1566,11 +1566,11 @@ conversion_warning (tree type, tree expr
   /* If any operand is artificial, then this expression was generated
      by the compiler and we do not warn.  */
   for (i = 0; i < expr_num_operands; i++)
     {
       tree op = TREE_OPERAND (expr, i);
-      if (DECL_P (op) && DECL_ARTIFICIAL (op))
+      if (op && DECL_P (op) && DECL_ARTIFICIAL (op))
        return;
     }

   switch (TREE_CODE (expr))
     {

However, the testcase is too big for the testsuite...


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217

[Bug c++/37217] [4.4 Regression] -Wconversion causes ICE with __builtin_strcmp with one char compare

[pinskia at gcc dot gnu dot org]

------- Comment #9 from pinskia at gcc dot gnu dot org  2008-08-25 18:59 -------
(In reply to comment #8)
> However, the testcase is too big for the testsuite...
The C++ testcase I attached is small.  See comment #4.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217

[Bug c++/37217] [4.4 Regression] -Wconversion causes ICE with __builtin_strcmp with one char compare

[manu at gcc dot gnu dot org]

------- Comment #10 from manu at gcc dot gnu dot org  2008-08-25 19:07 -------
(In reply to comment #9)
> (In reply to comment #8)
> > However, the testcase is too big for the testsuite...
> The C++ testcase I attached is small.  See comment #4.

I don't get the ICE with that testcase in x86_64-unknown-linux-gnu.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217

[Bug c++/37217] [4.4 Regression] -Wconversion causes ICE with __builtin_strcmp with one char compare

[manu at gcc dot gnu dot org]

------- Comment #11 from manu at gcc dot gnu dot org  2008-08-25 21:53 -------
(In reply to comment #10)
> (In reply to comment #9)
> > (In reply to comment #8)
> > > However, the testcase is too big for the testsuite...
> > The C++ testcase I attached is small.  See comment #4.
> 
> I don't get the ICE with that testcase in x86_64-unknown-linux-gnu.
> 

Reduced testcase for x86_64-unknown-linux-gnu.

typedef struct Tcl_ResolvedVarInfo {
 char *re_guts;
} regex_t;
void TclReComp(regex_t *re)
{
 if (re->re_guts == ((void *)0)) ;
}


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217

[Bug c++/37217] [4.4 Regression] -Wconversion causes ICE with __builtin_strcmp with one char compare

[manu at gcc dot gnu dot org]

------- Comment #12 from manu at gcc dot gnu dot org  2008-08-26 07:58 -------
Patch: http://gcc.gnu.org/ml/gcc-patches/2008-08/msg01913.html

Please test that it fixes the ICE in i686-pc-cygwin. Thanks.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217

[Bug c++/37217] [4.4 Regression] -Wconversion causes ICE with __builtin_strcmp with one char compare

[burnus at gcc dot gnu dot org]

------- Comment #13 from burnus at gcc dot gnu dot org  2008-08-27 15:36 -------
*** Bug 37230 has been marked as a duplicate of this bug. ***


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217

[Bug c++/37217] [4.4 Regression] -Wconversion causes ICE with __builtin_strcmp with one char compare

[manu at gcc dot gnu dot org]

------- Comment #14 from manu at gcc dot gnu dot org  2008-08-28 00:02 -------
Subject: Bug 37217

Author: manu
Date: Thu Aug 28 00:00:49 2008
New Revision: 139682

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=139682
Log:
2008-08-27  Manuel Lopez-Ibanez  <manu@gcc.gnu.org>

        PR 37217
        * c-common.c (convesion_warning): Check for null operands.
testsuite/
        * gcc.dg/pr37217.c: New.

Added:
    trunk/gcc/testsuite/gcc.dg/pr37217.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/c-common.c
    trunk/gcc/testsuite/ChangeLog


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217

[Bug c++/37217] [4.4 Regression] -Wconversion causes ICE with __builtin_strcmp with one char compare

[manu at gcc dot gnu dot org]

------- Comment #15 from manu at gcc dot gnu dot org  2008-08-28 00:09 -------
Subject: Bug 37217

Author: manu
Revision: 139682
Modified property: svn:log

Modified: svn:log at Thu Aug 28 00:09:20 2008
------------------------------------------------------------------------------
--- svn:log (original)
+++ svn:log Thu Aug 28 00:09:20 2008
@@ -1,6 +1,6 @@
 2008-08-27  Manuel Lopez-Ibanez  <manu@gcc.gnu.org>

        PR 37217
-       * c-common.c (convesion_warning): Check for null operands.
+       * c-common.c (conversion_warning): Check for null operands.
 testsuite/
        * gcc.dg/pr37217.c: New.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217

[Bug c++/37217] [4.4 Regression] -Wconversion causes ICE with __builtin_strcmp with one char compare

[manu at gcc dot gnu dot org]

------- Comment #16 from manu at gcc dot gnu dot org  2008-08-28 00:12 -------
Fixed in GCC 4.4.


-- 

manu at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217

[Bug c++/37217] [4.4 Regression] -Wconversion causes ICE with __builtin_strcmp with one char compare

[mckelvey at maskull dot com]

------- Comment #17 from mckelvey at maskull dot com  2008-08-28 02:41 -------
(In reply to comment #12)
> Patch: http://gcc.gnu.org/ml/gcc-patches/2008-08/msg01913.html
> 
> Please test that it fixes the ICE in i686-pc-cygwin. Thanks.
> 

Yes! Thanks!


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37217