Re: optimization/9357: [3.2/3.3/3.4 regression] SegFault with -fssa -funroll-loops -fprofile-arcs

Re: optimization/9357: [3.2/3.3/3.4 regression] SegFault with -fssa -funroll-loops -fprofile-arcs

[Dara Hazeghi]

The following reply was made to PR optimization/9357; it has been noted by GNATS.

From: Dara Hazeghi <dhazeghi@yahoo.com>
To: gcc-gnats@gcc.gnu.org
Cc:  
Subject: Re: optimization/9357: [3.2/3.3/3.4 regression] SegFault with -fssa  -funroll-loops -fprofile-arcs
Date: Fri, 9 May 2003 22:38:52 -0700

 http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit- 
 trail&database=gcc&pr=9357
 
 Hello,
 
 I can reproduce this bug on 3.2.3 and 3.3 branch but on mainline  
 (20030509) it compiles fine.
 
 Dara
 

Re: optimization/9357: [3.2/3.3/3.4 regression] SegFault with -fssa -funroll-loops -fprofile-arcs

[Andreas Jaeger]

The following reply was made to PR optimization/9357; it has been noted by GNATS.

From: Andreas Jaeger <aj@suse.de>
To: janis187@us.ibm.com
Cc: gcc-gnats@gcc.gnu.org, gcc-bugs@gcc.gnu.org, nobody@gcc.gnu.org,
	gcc-prs@gcc.gnu.org, maddinp@yahoo.de
Subject: Re: optimization/9357: [3.2/3.3/3.4 regression] SegFault with -fssa  
 -funroll-loops -fprofile-arcs
Date: Mon, 24 Feb 2003 17:46:31 +0100

 Janis Johnson <janis187@us.ibm.com> writes:
 
 > The regression showed up with this patch:
 >
 > Sat Oct 27 12:12:09 CEST 2001  Jan Hubicka  <jh@suse.cz>
 >
 >         * cfgcleanup.c (merge_blocks_move_predecessor_nojumps,
 >         merge_blocks_move_successors_nojumps): Set BB_UPDATE_LIFE.
 >         (merge_blocks): Properly udpate life after making edge nonfallthru.
 >         * cfgrtl.c (merge_blocks_nomove): Update global_live_at_end.
 >
 > http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=9357
 
 This might be the case but -fssa is currently broken AFAIK and not
 supported really, so if it cannot get reproduced without -fssa, I
 really susspect a SSA issue,
 
 Andreas
 -- 
  Andreas Jaeger
   SuSE Labs aj@suse.de
    private aj@arthur.inka.de
     http://www.suse.de/~aj

Re: optimization/9357: [3.2/3.3/3.4 regression] SegFault with -fssa -funroll-loops -fprofile-arcs

[Janis Johnson]

The following reply was made to PR optimization/9357; it has been noted by GNATS.

From: Janis Johnson <janis187@us.ibm.com>
To: gcc-gnats@gcc.gnu.org, gcc-bugs@gcc.gnu.org, nobody@gcc.gnu.org,
   gcc-prs@gcc.gnu.org, maddinp@yahoo.de
Cc:  
Subject: Re: optimization/9357: [3.2/3.3/3.4 regression] SegFault with -fssa  
 -funroll-loops -fprofile-arcs
Date: Mon, 24 Feb 2003 08:30:12 -0800

 The regression showed up with this patch:
 
 Sat Oct 27 12:12:09 CEST 2001  Jan Hubicka  <jh@suse.cz>
 
         * cfgcleanup.c (merge_blocks_move_predecessor_nojumps,
         merge_blocks_move_successors_nojumps): Set BB_UPDATE_LIFE.
         (merge_blocks): Properly udpate life after making edge nonfallthru.
         * cfgrtl.c (merge_blocks_nomove): Update global_live_at_end.
 
 http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=9357
 
 
 

Re: optimization/9357: [3.2/3.3/3.4 regression] SegFault with -fssa -funroll-loops -fprofile-arcs

[bangerth]

Old Synopsis: gcc-3.2.1 optimization-crash
New Synopsis: [3.2/3.3/3.4 regression] SegFault with -fssa  -funroll-loops -fprofile-arcs

State-Changed-From-To: open->analyzed
State-Changed-By: bangerth
State-Changed-When: Fri Jan 17 12:42:37 2003
State-Changed-Why:
    Confirmed. Here's a smaller testcase:
    ---------------------
    void foo() {
        typedef void (*fun_ptr)();
        fun_ptr *array;
        int i;
        bar(array);
        for (i=0; i<=0x100; i++)
            array[i] = &foo;
    }
    ----------------------------
    This crashes every compiler since 3.2:
    
    g/a> /home/bangerth/bin/gcc-3.4-pre/bin/gcc -O -c c64mem.c -fssa  -funroll-loops -fprofile-arcs
    c64mem.c: In function `foo':
    c64mem.c:8: internal compiler error: Speicherzugriffsfehler
    Please submit a full bug report,
    with preprocessed source if appropriate.
    See <URL:http://www.gnu.org/software/gcc/bugs.html> for instructions.
    
    
    
    It used to work with 3.0.4, so it's a regression.
    
    W.

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=9357