Re: c/4596: GCC 3.0.1 optimization causes segmentation violation in executable

Re: c/4596: GCC 3.0.1 optimization causes segmentation violation in executable

[rth]

The following reply was made to PR c/4596; it has been noted by GNATS.

From: rth@gcc.gnu.org
To: Matt.Rhodes@synergex.com, gcc-bugs@gcc.gnu.org, gcc-gnats@gcc.gnu.org,
  gcc-prs@gcc.gnu.org, nobody@gcc.gnu.org, rth@gcc.gnu.org
Cc:  
Subject: Re: c/4596: GCC 3.0.1 optimization causes segmentation violation in executable
Date: 20 Nov 2001 04:32:34 -0000

 Synopsis: GCC 3.0.1 optimization causes segmentation violation in executable
 
 Responsible-Changed-From-To: unassigned->rth
 Responsible-Changed-By: rth
 Responsible-Changed-When: Mon Nov 19 20:32:33 2001
 Responsible-Changed-Why:
     Mine.
 State-Changed-From-To: open->closed
 State-Changed-By: rth
 State-Changed-When: Mon Nov 19 20:32:33 2001
 State-Changed-Why:
     In the future, provide _all_ of the information requested in
       http://gcc.gnu.org/bugs.html
     as you've not provided enough information.
     
     That said, the existance of -fomit-frame-pointer in your
     command line suggests that perhaps 
       http://gcc.gnu.org/ml/gcc-patches/2001-11/msg00698.html
     might solve your problem.  If not, submit a proper report.
 
 http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=4596&database=gcc

Re: c/4596: GCC 3.0.1 optimization causes segmentation violation in executable

[rth]

Synopsis: GCC 3.0.1 optimization causes segmentation violation in executable

Responsible-Changed-From-To: unassigned->rth
Responsible-Changed-By: rth
Responsible-Changed-When: Mon Nov 19 20:32:33 2001
Responsible-Changed-Why:
    Mine.
State-Changed-From-To: open->closed
State-Changed-By: rth
State-Changed-When: Mon Nov 19 20:32:33 2001
State-Changed-Why:
    In the future, provide _all_ of the information requested in
      http://gcc.gnu.org/bugs.html
    as you've not provided enough information.
    
    That said, the existance of -fomit-frame-pointer in your
    command line suggests that perhaps 
      http://gcc.gnu.org/ml/gcc-patches/2001-11/msg00698.html
    might solve your problem.  If not, submit a proper report.

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=4596&database=gcc

c/4596: GCC 3.0.1 optimization causes segmentation violation in executable

[Matt.Rhodes]

>Number:         4596
>Category:       c
>Synopsis:       GCC 3.0.1 optimization causes segmentation violation in executable
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          wrong-code
>Submitter-Id:   net
>Arrival-Date:   Wed Oct 17 13:56:00 PDT 2001
>Closed-Date:
>Last-Modified:
>Originator:     Matt Rhodes
>Release:        gcc version 3.0.1
>Organization:
>Environment:
Red Hat Linux 7.1 running on a Pentium 120.
>Description:
When the lcmdln.c module is compiled with -O3 optimization, the executable (dblibr) fails with a segmentation violation when it is run.  However, when lcmdln.c is compiled with
-O3 in addition to -fschedule-insns explicitly specified, the excutable runs without the segmentation violation. Also, it executes without error if lcmdln.c is only compiled with -O1 optimization.
>How-To-Repeat:
$ gcc -O3 -finline-functions -funroll-loops
-fomit-frame-pointer -fexpensive-optimizations
-fschedule-insns2 -fstrength-reduce -fno-strict-aliasing
-trigraphs -march=pentium -fPIC -o lcmdln.o -c lcmdln.c
$ gcc -o dblibr calloc.o lcreat.o lmain.o lmisc.o lcmdln.o lerrint.o nordisam.a duisam.a dutil.a stblib.a -lbsd -lm
$ dblibr -c bldut.olb bldut1.dbo
>Fix:
Add -fschedule-insns to compiler options when building lcmdln.c module OR change -O3 compiler option to -O1.
>Release-Note:
>Audit-Trail:
>Unformatted: