From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:51156)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <lersek@redhat.com>) id 1Um9Og-00048W-3o
	for qemu-devel@nongnu.org; Mon, 10 Jun 2013 17:13:04 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <lersek@redhat.com>) id 1Um9Ob-0007NS-Hq
	for qemu-devel@nongnu.org; Mon, 10 Jun 2013 17:13:01 -0400
Received: from mx1.redhat.com ([209.132.183.28]:51177)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <lersek@redhat.com>) id 1Um9Ob-0007NJ-AE
	for qemu-devel@nongnu.org; Mon, 10 Jun 2013 17:12:57 -0400
Message-ID: <51B641C8.2050501@redhat.com>
Date: Mon, 10 Jun 2013 23:14:48 +0200
From: Laszlo Ersek <lersek@redhat.com>
MIME-Version: 1.0
References: <1370882347-31129-1-git-send-email-mst@redhat.com>
	<87r4g9rdwe.fsf@codemonkey.ws> <51B62C48.5060303@redhat.com>
	<87ehc920ym.fsf@codemonkey.ws> <51B63A74.1030905@redhat.com>
In-Reply-To: <51B63A74.1030905@redhat.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] [PATCH] qemu: piix: PCI bridge ACPI hotplug support
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: Anthony Liguori <anthony@codemonkey.ws>
Cc: Peter Maydell <peter.maydell@linaro.org>, "Michael S. Tsirkin" <mst@redhat.com>, David Woodhouse <dwmw2@infradead.org>, seabios@seabios.org, qemu-devel@nongnu.org, Kevin O'Connor <kevin@koconnor.net>, Gerd Hoffmann <kraxel@redhat.com>, Jordan Justen <jljusten@gmail.com>

Sorry about replying to myself...

On 06/10/13 22:43, Laszlo Ersek wrote:

> It's not a design purity argument, just what's cheaper.

You probably consider SMBIOS/ACPI table generation "guest code" that has
no place in the machine emulator. You're likely willing to expose the
needed bits on a case-by-case basis, in their barest representation.

I must consider the entire stack, from the libvirt top, down to the
guest kernel or root shell, and see where it is cheapest to effect
changes. The higher I can hoist something the better.

Modulo, it depends on what element of the stack people think of as
central / indispensable -- that element should support the feature directly.

(Returning to my earlier example, it would be theoretically viable *I
guess* to equip libvirt to pre-format a Type 3 SMBIOS blob in a file,
and pass it with "-smbios file=...", but many people don't use libvirt.
It also might not apply to other emulators supported by libvirt.)

Thanks,
Laszlo