From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43)
	id 1KLTXI-0004lX-1V
	for qemu-devel@nongnu.org; Tue, 22 Jul 2008 21:53:00 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43)
	id 1KLTXG-0004lK-JK
	for qemu-devel@nongnu.org; Tue, 22 Jul 2008 21:52:59 -0400
Received: from [199.232.76.173] (port=56531 helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1KLTXG-0004lH-Gn
	for qemu-devel@nongnu.org; Tue, 22 Jul 2008 21:52:58 -0400
Received: from an-out-0708.google.com ([209.85.132.240]:56114)
	by monty-python.gnu.org with esmtp (Exim 4.60)
	(envelope-from <anthony@codemonkey.ws>) id 1KLTXG-0002yM-31
	for qemu-devel@nongnu.org; Tue, 22 Jul 2008 21:52:58 -0400
Received: by an-out-0708.google.com with SMTP id d18so851243and.130
	for <qemu-devel@nongnu.org>; Tue, 22 Jul 2008 18:52:57 -0700 (PDT)
Message-ID: <48868EDB.7010809@codemonkey.ws>
Date: Tue, 22 Jul 2008 20:52:27 -0500
From: Anthony Liguori <anthony@codemonkey.ws>
MIME-Version: 1.0
Subject: Re: [Qemu-devel] [PATCH] Save 3MB ioport table memory
References: <20080721120211.GD4501@implementation.uk.xensource.com>
In-Reply-To: <20080721120211.GD4501@implementation.uk.xensource.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
Reply-To: qemu-devel@nongnu.org
List-Id: qemu-devel.nongnu.org
List-Unsubscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: qemu-devel@nongnu.org
Cc: Samuel Thibault <samuel.thibault@eu.citrix.com>

Samuel Thibault wrote:
> Hello,
>
> On 64bit machines, qemu uses 3MB of memory for the ioport tables, while
> the actual use of them is quite sparse.  The patch below saves that
> memory by leaving them zeroed and just test for that and use the default
> pointer.
>   

I really like the idea of this patch.  You have some whitespace damage 
(tabs instead of 8 spaces) and it needs a Signed-off-by: if you'd like 
me to apply it.

> Samuel
>
> Index: vl.c
> ===================================================================
> --- vl.c	(révision 4917)
> +++ vl.c	(copie de travail)
> @@ -279,17 +279,29 @@
>  static uint32_t default_ioport_readw(void *opaque, uint32_t address)
>  {
>      uint32_t data;
> -    data = ioport_read_table[0][address](ioport_opaque[address], address);
> +    IOPortReadFunc *func = ioport_read_table[0][address];
> +    if (!func)
> +	    func = default_ioport_readb;
>   

Perhaps you can introduce an accessor?  Something like:

static uint32_t ioport_read_defaults(void *opaque, int index, uint32_t 
address)
{
    static IOPortReadFunc *default_funcs[3] = {default_ioport_readb, 
default_ioport_readw, default_ioport_readl};
    IOPortReadFunc *func = ioport_read_table[index][address];
    if (!func)
        func = default_funcs[index];
    return func(opaque, address);
}

That would simplify the rest of the changes significantly.

Regards,

Anthony Liguori