c++/6453: g++ regression in 3.1

c++/6453: g++ regression in 3.1

[green]

>Number:         6453
>Category:       c++
>Synopsis:       g++ regression in 3.1
>Confidential:   no
>Severity:       non-critical
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          ice-on-illegal-code
>Submitter-Id:   net
>Arrival-Date:   Thu Apr 25 08:16:00 PDT 2002
>Closed-Date:
>Last-Modified:
>Originator:     Anthony Green
>Release:        3.1 20020424 (prerelease)
>Organization:
>Environment:
System: Linux louie 2.4.7-10smp #1 SMP Thu Sep 6 17:09:31 EDT 2001 i686 unknown
Architecture: i686

	
host: i686-pc-linux-gnu
build: i686-pc-linux-gnu
target: i686-pc-linux-gnu
configured with: /louie/green/3.1/gcc/configure --prefix=/louie/green/3.1/i --enable-threads --enable-languages=c,c++,java : (reconfigured)  : (reconfigured) 
>Description:
Given the following nonsense code...

void *foo()
{
  char x[100];
  x++;
  return (void *) x;
}

Red Hat's g++ 2.96 reports:

$ g++ -c c.cc
c.cc: In function `void *foo ()':
c.cc:4: non-lvalue in increment
c.cc:3: warning: address of local variable `x' returned

Unfortunately I don't have an older FSF release handy to test on (like
2.95, or 3.0), so I don't know what they do.  However, the current 3.1
branch says:

$ g++ -c c.cc
non-lvalue

in

increment

address

of

local

variable

`
Internal compiler error: Error reporting routines re-entered.
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://www.gnu.org/software/gcc/bugs.html> for instructions.


>How-To-Repeat:
Compile the following with g++:

void *foo()
{
  char x[100];
  x++;
  return (void *) x;
}

>Fix:
	
>Release-Note:
>Audit-Trail:
>Unformatted:

Re: c++/6453: g++ regression in 3.1

[green]

Synopsis: g++ regression in 3.1

State-Changed-From-To: open->closed
State-Changed-By: green
State-Changed-When: Fri Apr 26 14:02:50 2002
State-Changed-Why:
    This was a transient problem for which I have no explanation.  update, "make clean" and rebuild a few days laters results in proper output.

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=6453

Re: c++/6453: g++ regression in 3.1

[Manfred Hollstein]

The following reply was made to PR c++/6453; it has been noted by GNATS.

From: Manfred Hollstein <manfredh@redhat.com>
To: green@redhat.com
Cc: gcc-gnats@gcc.gnu.org
Subject: Re: c++/6453: g++ regression in 3.1
Date: Thu, 25 Apr 2002 17:31:36 +0200

 green@redhat.com wrote:
 >>Number:         6453
 >>Category:       c++
 >>Synopsis:       g++ regression in 3.1
 >>Confidential:   no
 >>Severity:       non-critical
 >>Priority:       medium
 >>Responsible:    unassigned
 >>State:          open
 >>Class:          ice-on-illegal-code
 >>Submitter-Id:   net
 >>Arrival-Date:   Thu Apr 25 08:16:00 PDT 2002
 >>Closed-Date:
 >>Last-Modified:
 >>Originator:     Anthony Green
 >>Release:        3.1 20020424 (prerelease)
 >>Organization:
 >>Environment:
 > 
 > System: Linux louie 2.4.7-10smp #1 SMP Thu Sep 6 17:09:31 EDT 2001 i686 unknown
 > Architecture: i686
 > 
 > 	
 > host: i686-pc-linux-gnu
 > build: i686-pc-linux-gnu
 > target: i686-pc-linux-gnu
 > configured with: /louie/green/3.1/gcc/configure --prefix=/louie/green/3.1/i --enable-threads --enable-languages=c,c++,java : (reconfigured)  : (reconfigured) 
 
 Hmm, my snapshot from yesterday's branch in CVS (2002/04/24 10:12:11
 +0200) works without generating an ICE. I did a clean build of
 everything, though, ie. I don't have any (reconfigured) notes.
 
 My full Configure with: line reads like this:
 
 Configured with: ../gcc-3.1/configure --host=i686-redhat7-linux-gnu 
 --target=i686-redhat7-linux-gnu --build=i686-redhat7-linux-gnu 
 --srcdir=../gcc-3.1 --prefix=/opt/gnu 
 --exec-prefix=/opt/gnu/H-ia32-linux --with-gnu-as --with-gnu-ld 
 --disable-checking --enable-clocale=gnu --enable-interpreter 
 --enable-shared --enable-threads=posix --with-system-zlib 
 --with-local-prefix=/opt/gnu --enable-languages=c++,java,ada 
 --enable-nls --verbose
 
 Cheers.
 
 l8er
 manfred
 
 > 
 >>Description:
 > 
 > Given the following nonsense code...
 > 
 > void *foo()
 > {
 >   char x[100];
 >   x++;
 >   return (void *) x;
 > }
 > 
 > Red Hat's g++ 2.96 reports:
 > 
 > $ g++ -c c.cc
 > c.cc: In function `void *foo ()':
 > c.cc:4: non-lvalue in increment
 > c.cc:3: warning: address of local variable `x' returned
 > 
 > Unfortunately I don't have an older FSF release handy to test on (like
 > 2.95, or 3.0), so I don't know what they do.  However, the current 3.1
 > branch says:
 > 
 > $ g++ -c c.cc
 > non-lvalue
 > 
 > in
 > 
 > increment
 > 
 > address
 > 
 > of
 > 
 > local
 > 
 > variable
 > 
 > `
 > Internal compiler error: Error reporting routines re-entered.
 > Please submit a full bug report,
 > with preprocessed source if appropriate.
 > See <URL:http://www.gnu.org/software/gcc/bugs.html> for instructions.
 > 
 > 
 > 
 >>How-To-Repeat:
 > 
 > Compile the following with g++:
 > 
 > void *foo()
 > {
 >   char x[100];
 >   x++;
 >   return (void *) x;
 > }
 > 
 > 
 >>Fix:
 > 
 > 	
 > 
 >>Release-Note:
 >>Audit-Trail:
 >>Unformatted:
 > 
 >