From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43)
	id 1MxHuH-0006sR-GZ
	for qemu-devel@nongnu.org; Mon, 12 Oct 2009 06:13:33 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43)
	id 1MxHuC-0006pJ-UY
	for qemu-devel@nongnu.org; Mon, 12 Oct 2009 06:13:32 -0400
Received: from [199.232.76.173] (port=34949 helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1MxHuC-0006pA-4f
	for qemu-devel@nongnu.org; Mon, 12 Oct 2009 06:13:28 -0400
Received: from mx1.redhat.com ([209.132.183.28]:49468)
	by monty-python.gnu.org with esmtp (Exim 4.60)
	(envelope-from <kraxel@redhat.com>) id 1MxHuB-0006Rb-Oq
	for qemu-devel@nongnu.org; Mon, 12 Oct 2009 06:13:28 -0400
Message-ID: <4AD30136.3000506@redhat.com>
Date: Mon, 12 Oct 2009 12:13:10 +0200
From: Gerd Hoffmann <kraxel@redhat.com>
MIME-Version: 1.0
Subject: Re: [Qemu-devel] [RFC PATCH v2 5/5] ne2k_isa: add property for option
	rom loading.
References: <1254918996-26050-1-git-send-email-kraxel@redhat.com>
	<1254918996-26050-6-git-send-email-kraxel@redhat.com>
	<4ACC92C8.8060700@codemonkey.ws> <4ACC95D3.6070305@redhat.com>
	<4ACC9765.3030300@codemonkey.ws> <4ACC9EF9.1010201@redhat.com>
	<4ACCA2FC.6010002@codemonkey.ws> <4ACCA55A.1020602@redhat.com>
	<4ACCDF44.9010704@codemonkey.ws>
In-Reply-To: <4ACCDF44.9010704@codemonkey.ws>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
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: Anthony Liguori <anthony@codemonkey.ws>
Cc: qemu-devel@nongnu.org

>>>> It's pc-specific though, so when we go the route of loading roms
>>>> unconditionally we need to wrap that into a machine-specific helper
>>>> function so it happes on TARGET_I386 only.
>>>
>>> No, it's not pc-specific. An e1000 card on a PPC still has an x86 option
>>> rom. Whether it gets loaded and how it gets loaded depends on the
>>> target, but not the existence of the rom on the device.
>>
>> Yep, the *loading* is what I was referring to (see $subject) ...
>
> Well, I guess I'm confused about where we stand.

loading the rom is x86 specific ...

> Are you suggesting that we drop the pxe property and load the roms
> unconditionally?

Yes, I think we should do that (on x86), given BEV works nicely.
The e1000 rom needs fixing first of course ;)

> The tricky thing here is that we only want to load a
> particular rom once. There's no need to load the rtl8139 multiple times
> for multiple nics.

As hw/loader.c keeps track of the roms this should be easy to do.

The rom_add_option() macro in hw/loader.h should become a function which 
loads the rom on x86 and does nothing on other archs.  Then the nic 
drivers can simply call rom_add_option("pxe-${driver}.bin") unconditionally.

cheers,
   Gerd