From mboxrd@z Thu Jan  1 00:00:00 1970
From: Avi Kivity <avi@redhat.com>
Subject: Re: [Qemu-devel] [PATCH 0/2] Fix wide ioport access cracking
Date: Thu, 11 Aug 2011 19:22:54 +0300
Message-ID: <4E4401DE.9090504@redhat.com>
References: <1313048426-17273-1-git-send-email-avi@redhat.com> <alpine.LFD.2.02.1108111021530.30725@bbs.intern> <4E43927B.90606@redhat.com> <4E4392F7.8020002@redhat.com> <alpine.LFD.2.02.1108111040590.4090@bbs.intern> <4E43A482.5040401@redhat.com> <alpine.LFD.2.02.1108111807260.27291@bbs.intern> <4E440169.8030403@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Cc: qemu-devel@nongnu.org, kvm@vger.kernel.org
To: Gerhard Wiesinger <lists@wiesinger.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx1.redhat.com ([209.132.183.28]:50236 "EHLO mx1.redhat.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1751212Ab1HKQW6 (ORCPT <rfc822;kvm@vger.kernel.org>);
	Thu, 11 Aug 2011 12:22:58 -0400
In-Reply-To: <4E440169.8030403@redhat.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

On 08/11/2011 07:20 PM, Avi Kivity wrote:
> On 08/11/2011 07:08 PM, Gerhard Wiesinger wrote:
>>
>> (gdb) frame 4
>> #4  0x000000000041eb9b in pci_update_mappings (d=0x1a90bc0)
>>     at /root/download/qemu/git/qemu-kvm-test/hw/pci.c:1134
>> 1134                memory_region_del_subregion(r->address_space, 
>> r->memory);
>> (gdb) print i
>> $1 = <value optimized out>
>> (gdb) print *r
>> $2 = {addr = 22058952032257, size = 32, filtered_size = 
>> 171717340864446496,
>>   type = 1 '\001', memory = 0x1a90000, address_space = 0x200019282f0}
>> (gdb) print d->io_regions[0]
>> $3 = {addr = 22058952032257, size = 32, filtered_size = 
>> 171717340864446496,
>>   type = 1 '\001', memory = 0x1a90000, address_space = 0x200019282f0}
>
> Yikes, this looks like corruption, it the leading 0x2000 in 
> address_space is out of place.
>
> Can you step through lsi pci bar registration and place a data 
> breakpoint on address_space, and see where it gets this value?
>
> 'addr' looks bad too.
>

Or maybe it's just -O2 screwing up debug information.  Please change 
./configure to set -O1 and redo.

Please print *r.memory as well.


-- 
error compiling committee.c: too many arguments to function