From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([140.186.70.92]:55329)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <anthony@codemonkey.ws>) id 1Qg5Vi-0001x7-6m
	for qemu-devel@nongnu.org; Sun, 10 Jul 2011 21:42:11 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <anthony@codemonkey.ws>) id 1Qg5Vh-0005J7-7Y
	for qemu-devel@nongnu.org; Sun, 10 Jul 2011 21:42:10 -0400
Received: from mail-yx0-f173.google.com ([209.85.213.173]:41020)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <anthony@codemonkey.ws>) id 1Qg5Vh-0005J1-57
	for qemu-devel@nongnu.org; Sun, 10 Jul 2011 21:42:09 -0400
Received: by yxt3 with SMTP id 3so1650711yxt.4
	for <qemu-devel@nongnu.org>; Sun, 10 Jul 2011 18:42:07 -0700 (PDT)
Message-ID: <4E1A54ED.7040606@codemonkey.ws>
Date: Sun, 10 Jul 2011 20:42:05 -0500
From: Anthony Liguori <anthony@codemonkey.ws>
MIME-Version: 1.0
References: <1310321709-30770-1-git-send-email-avi@redhat.com>	<1310321709-30770-32-git-send-email-avi@redhat.com>
	<alpine.LNX.2.00.1107110030120.9345@linmac>
In-Reply-To: <alpine.LNX.2.00.1107110030120.9345@linmac>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] [RFC v3 31/56] ac97: convert to memory API
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: malc <av1474@comtv.ru>
Cc: Avi Kivity <avi@redhat.com>, kvm@vger.kernel.org, qemu-devel@nongnu.org

On 07/10/2011 03:33 PM, malc wrote:
> On Sun, 10 Jul 2011, Avi Kivity wrote:
>
>> fixes BAR sizing as well.
>
> I find this patch disgusting, the read and write handlers in particular.

Shouldn't it be possible to do something like:

typedef struct OldMemoryRegionOps {
     MemoryRegionOps parent_ops;
     CPUReadMemoryFunc *readfn[3];
     CPUWriteMemoryFunc *writefn[3];
     void *opaque;
} OldMemoryRegionOps;

That should allow old-style implementations to be converted without 
introducing trampoline functions everywhere.

Regards,

Anthony Liguori

>
> [..snip..]
>