Re: c++/9193: Testsuite error on mips-irix6 with g++.dg/abi/empty6.C (missing -Wabi warning)

Re: c++/9193: Testsuite error on mips-irix6 with g++.dg/abi/empty6.C (missing -Wabi warning)

[Mark Mitchell]

The following reply was made to PR c++/9193; it has been noted by GNATS.

From: Mark Mitchell <mark@codesourcery.com>
To: "Kaveh R. Ghazi" <ghazi@caip.rutgers.edu>,
   "gcc-gnats@gcc.gnu.org" <gcc-gnats@gcc.gnu.org>
Cc: "gcc-bugs@gcc.gnu.org" <gcc-bugs@gcc.gnu.org>
Subject: Re: c++/9193: Testsuite error on mips-irix6 with
 g++.dg/abi/empty6.C (missing -Wabi warning)
Date: Wed, 08 Jan 2003 09:47:23 -0800

 --On Wednesday, January 08, 2003 12:37:50 PM -0500 "Kaveh R. Ghazi" 
 <ghazi@caip.rutgers.edu> wrote:
 
 >  > From: Mark Mitchell <mark@codesourcery.com>
 >  >
 >  > > and indeed no -Wabi warning is issued.  The failure occurs only when
 > I  > > run the testsuite with -mabi=64, it succeeds without that flag as
 > noted  > > in the test report above.
 >  >
 >  > That's correct behavior; it means that there's no change under that
 > ABI.
 >
 > For my own education, would you please help me understand why the
 > irix6 -mabi=n32 vs -mabi=64 flag affects this, but other 64 bit
 > platforms, or sparc -m64, don't see a change?
 
 Essentially, there's no logical answer.  We were assuming that
 DECL_FIELD_OFFSET meant something it only means sometimes.  On platforms
 where it meant what we thought it meant, we don't need to warn; elsewhere,
 we do.
 
 -- 
 Mark Mitchell                mark@codesourcery.com
 CodeSourcery, LLC            http://www.codesourcery.com

Re: c++/9193: Testsuite error on mips-irix6 with g++.dg/abi/empty6.C (missing -Wabi warning)

[ghazi]

Synopsis: Testsuite error on mips-irix6 with g++.dg/abi/empty6.C (missing -Wabi warning)

State-Changed-From-To: open->closed
State-Changed-By: ghazi
State-Changed-When: Mon Mar  3 15:55:18 2003
State-Changed-Why:
    This problem seems to have stopped occuring around Feb 27.  I'd feel more comfortable closing it if I was sure it's fixed rather than "hidden".  The regression was last seen here:
    http://gcc.gnu.org/ml/gcc-testresults/2003-01/msg01250.html
    and is gone here:
    http://gcc.gnu.org/ml/gcc-testresults/2003-02/msg01447.html
    

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=9193

Re: c++/9193: Testsuite error on mips-irix6 with g++.dg/abi/empty6.C (missing -Wabi warning)

[Kaveh R. Ghazi]

The following reply was made to PR c++/9193; it has been noted by GNATS.

From: "Kaveh R. Ghazi" <ghazi@caip.rutgers.edu>
To: gcc-gnats@gcc.gnu.org, mark@codesourcery.com
Cc: gcc-bugs@gcc.gnu.org
Subject: Re: c++/9193: Testsuite error on mips-irix6 with g++.dg/abi/empty6.C (missing -Wabi warning)
Date: Wed, 8 Jan 2003 12:37:50 -0500 (EST)

  > From: Mark Mitchell <mark@codesourcery.com>
  > 
  > > and indeed no -Wabi warning is issued.  The failure occurs only when I
  > > run the testsuite with -mabi=64, it succeeds without that flag as noted
  > > in the test report above.
  > 
  > That's correct behavior; it means that there's no change under that ABI.
 
 For my own education, would you please help me understand why the
 irix6 -mabi=n32 vs -mabi=64 flag affects this, but other 64 bit
 platforms, or sparc -m64, don't see a change?
 
 
  > 
  > So, would you do the magic to make the test not run in that situation?
 
 Sure, I'll give it a try.  Thanks for analyzing this one.
 
 		--Kaveh
 --
 Kaveh R. Ghazi			ghazi@caip.rutgers.edu

Re: c++/9193: Testsuite error on mips-irix6 with g++.dg/abi/empty6.C (missing -Wabi warning)

[Mark Mitchell]

The following reply was made to PR c++/9193; it has been noted by GNATS.

From: Mark Mitchell <mark@codesourcery.com>
To: "ghazi@caip.rutgers.edu" <ghazi@caip.rutgers.edu>,
   "gcc-gnats@gcc.gnu.org" <gcc-gnats@gcc.gnu.org>
Cc:  
Subject: Re: c++/9193: Testsuite error on mips-irix6 with
 g++.dg/abi/empty6.C (missing -Wabi warning)
Date: Mon, 06 Jan 2003 17:38:07 -0800

 --On Monday, January 06, 2003 05:14:13 PM +0000 "ghazi@caip.rutgers.edu" 
 <ghazi@caip.rutgers.edu> wrote:
 
 > and indeed no -Wabi warning is issued.  The failure occurs only when I
 > run the testsuite with -mabi=64, it succeeds without that flag as noted
 > in the test report above.
 
 That's correct behavior; it means that there's no change under that ABI.
 
 So, would you do the magic to make the test not run in that situation?
 
 (I'm happy to do it, but I can't test it, so you have a better chance of
 getting it right.)
 
 Thanks!
 
 -- 
 Mark Mitchell                mark@codesourcery.com
 CodeSourcery, LLC            http://www.codesourcery.com

c++/9193: Testsuite error on mips-irix6 with g++.dg/abi/empty6.C (missing -Wabi warning)

[ghazi]

>Number:         9193
>Category:       c++
>Synopsis:       Testsuite error on mips-irix6 with g++.dg/abi/empty6.C (missing -Wabi warning)
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Mon Jan 06 09:16:01 PST 2003
>Closed-Date:
>Last-Modified:
>Originator:     Kaveh Ghazi
>Release:        gcc version 3.4 20030105 (experimental)
>Organization:
>Environment:
mips-sgi-irix6.5
>Description:
I'm getting a testsuite failure for g++.dg/abi/empty6.C on mips-sgi-irix6.5 as seen here:  http://gcc.gnu.org/ml/gcc-testresults/2003-01/msg00233.html

The error is not new, it appears to have started from the very day the testcase was checked in: http://gcc.gnu.org/ml/gcc-testresults/2002-09/msg00909.html

The logfile says:
FAIL: g++.dg/abi/empty6.C  (test for warnings, line 6)

and indeed no -Wabi warning is issued.  The failure occurs only when I run the testsuite with -mabi=64, it succeeds without that flag as noted in the test report above.

>How-To-Repeat:
Target mips-sgi-irix6.5 and compile empty6.ii with:

cc1plus -fpreprocessed empty6.ii -quiet -dumpbase empty6.C -mabi=64 -auxbase-strip empty6.C -Wabi -version -fmessage-length=0 -o empty6.C

and note that no -Wabi warning is issued.

# 1 "/caip/u58/ghazi/gcc-testing/egcc-CVS20030105/gcc/testsuite/g++.dg/abi/empty6.C"
# 1 "<built-in>"
# 1 "<command line>"
# 1 "/caip/u58/ghazi/gcc-testing/egcc-CVS20030105/gcc/testsuite/g++.dg/abi/empty6.C"


struct A {};

struct B {
  A a;
  virtual void f () {}
};


>Fix:

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