From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:41871)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <paolo.bonzini@gmail.com>) id 1VFHiK-0006TW-Dx
	for qemu-devel@nongnu.org; Fri, 30 Aug 2013 01:57:52 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <paolo.bonzini@gmail.com>) id 1VFHiB-0004ku-L7
	for qemu-devel@nongnu.org; Fri, 30 Aug 2013 01:57:44 -0400
Received: from mail-ee0-x230.google.com ([2a00:1450:4013:c00::230]:54896)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <paolo.bonzini@gmail.com>) id 1VFHiB-0004ko-FI
	for qemu-devel@nongnu.org; Fri, 30 Aug 2013 01:57:35 -0400
Received: by mail-ee0-f48.google.com with SMTP id l10so666943eei.7
	for <qemu-devel@nongnu.org>; Thu, 29 Aug 2013 22:57:34 -0700 (PDT)
Sender: Paolo Bonzini <paolo.bonzini@gmail.com>
Message-ID: <52203448.1080101@redhat.com>
Date: Fri, 30 Aug 2013 07:57:28 +0200
From: Paolo Bonzini <pbonzini@redhat.com>
MIME-Version: 1.0
References: <4C818617-784B-4685-B98C-44E9AFFBD8B8@bluezbox.com>
In-Reply-To: <4C818617-784B-4685-B98C-44E9AFFBD8B8@bluezbox.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] PCNet-PCI I/O ops byte order
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Oleksandr Tymoshenko <gonzo@bluezbox.com>
Cc: qemu-devel@nongnu.org

Il 29/08/2013 22:56, Oleksandr Tymoshenko ha scritto:
> Hello,
> 
> I'm working on QEMU/mips support for FreeBSD. qemu-system-mipsel works
> just fine but I ran into a problem with qemu-system-mips. There are two 
> devices on PCI bus in MALTA machine emulation. ATA IDE controller and
> PCNet NIC:  hw/ide/pci.c and hw/net/pcnet-pci.c respectively. Problem is
> I/O ops byte order for these two devices are defined inconsistently.
> It's  DEVICE_LITTLE_ENDIAN for bmdma_addr_ioport_ops and 
> DEVICE_NATIVE_ENDIAN for pcnet_io_ops. And since byte swapping
> in my case performed by bus driver I can't get consistent behaviour
> for these devices on big-endian system.
> 
> I don't have real hardware to run my code on but shouldn't all devices
> on PCI bus treat words as little endian?
> 
> Thank you

Yes, you are right.  Aurelien Jarno just posted a patch to use
DEVICE_LITTLE_ENDIAN for pcnet_io_ops.

Paolo