From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-patches-return-289139-listarch-gcc-patches=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 27316 invoked by alias); 11 Apr 2011 11:49:23 -0000
Received: (qmail 27303 invoked by uid 22791); 11 Apr 2011 11:49:23 -0000
X-SWARE-Spam-Status: No, hits=-3.1 required=5.0	tests=AWL,BAYES_00,TW_BF,T_RP_MATCHES_RCVD
X-Spam-Check-By: sourceware.org
Received: from cantor2.suse.de (HELO mx2.suse.de) (195.135.220.15)    by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Mon, 11 Apr 2011 11:49:18 +0000
Received: from relay2.suse.de (charybdis-ext.suse.de [195.135.221.2])	by mx2.suse.de (Postfix) with ESMTP id 20666867E2;	Mon, 11 Apr 2011 13:49:17 +0200 (CEST)
Date: Mon, 11 Apr 2011 11:49:00 -0000
From: Michael Matz <matz@suse.de>
To: Dominique Dhumieres <dominiq@lps.ens.fr>
Cc: paul.richard.thomas@gmail.com, gcc-patches@gcc.gnu.org,	fortran@gcc.gnu.org
Subject: Re: Implement stack arrays even for unknown sizes
In-Reply-To: <20110410132925.5B1493BE18@mailhost.lps.ens.fr>
Message-ID: <Pine.LNX.4.64.1104111346210.1989@wotan.suse.de>
References: <20110409100809.BF9CD3BE18@mailhost.lps.ens.fr> <BANLkTikwa8es+FxG2ThwWMo4MFGWVeddvg@mail.gmail.com> <20110410132925.5B1493BE18@mailhost.lps.ens.fr>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-IsSubscribed: yes
Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-patches.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-patches/>
List-Post: <mailto:gcc-patches@gcc.gnu.org>
List-Help: <mailto:gcc-patches-help@gcc.gnu.org>
Sender: gcc-patches-owner@gcc.gnu.org
X-SW-Source: 2011-04/txt/msg00743.txt.bz2

Hi,

On Sun, 10 Apr 2011, Dominique Dhumieres wrote:

> > I find that both nf.f90 and capacita.f90 segfault in runtime for any stack size.
> 
> On x86_64-apple-darwin10, nf.f90 "works". However if I run it through
> valgrind I get
> 
> ==64815== Memcheck, a memory error detector
> ==64815== Copyright (C) 2002-2010, and GNU GPL'd, by Julian Seward et al.
> ==64815== Using Valgrind-3.6.1 and LibVEX; rerun with -h for copyright info
> ==64815== Command: a.out --max-stackframe=2118496
> ==64815==
> ==64815== Warning: set address range perms: large range [0x7ffe6c000000, 0x7fff5bc01000) (defined)
> ==64815== Warning: client switching stacks?  SP change: 0x7fff5bffe410 --> 0x7fff5be0cef0
> ==64815==          to suppress, use: --max-stackframe=2037024 or greater

See?  That's whay I meant with having to use a large ulimit for stack 
size.  Usually stack overflows symptom is a simple segfault.  What ulimit 
-s have you used for your capacita tests?

> The segfault for capacita.f90 occurs in the subroutine fourir at the line
> 
>             write(unit=*, fmt=*) "error in fourier: n=", ntot
> 
> AFAICT the problem occurs in the loop
> 
>       do m=1,ntot/4-1
>         E(m) = exp(m*h)
>       end do
> 
> If I print ntot, loc(ntot) before it I get
> 
>         2048      140734799794712
> 
> After the loop loc(ntot) is
> 
>               9205357642636066816
> 
> and any attemp to print its values yields a segfault.

I'll poke at polyhedron somewhat.


Ciao,
Michael.