From: Anthony Liguori <aliguori@us.ibm.com>
To: Gerd Hoffmann <kraxel@redhat.com>
Cc: Andre Przywara <andre.przywara@amd.com>,
Glauber Costa <glommer@redhat.com>,
qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] Re: [PATCH v2 9/9] Add -kvm option
Date: Mon, 12 Oct 2009 09:04:36 -0500 [thread overview]
Message-ID: <4AD33774.1030001@us.ibm.com> (raw)
In-Reply-To: <4AD319F6.70506@redhat.com>
Gerd Hoffmann wrote:
> On 10/08/09 01:28, Anthony Liguori wrote:
>> Glauber Costa wrote:
>>> even if we have qdev on the irq controllers, one could still come up
>>> with
>>> situations in which we'd like to force the use of one device over
>>> another.
>>
>> Right, my assumption is that the devices will have different qdev names
>> and therefore a user can select which one gets used.
>>
>> Right now, -device is just additive. I'm not sure the best way to
>> express something like, replace this existing device with this new
>> device. Maybe some trickery with id=. Gerd, any thoughts?
>
> -M $machine gives you a barebone machine with all core devices which
> belong to it. You can't easily remove and/or replace devices.
> Especially not something central as the IRQ controller. But also no
> other core components, i.e. you wouldn't stick a piix4 ide controller
> into a Q35 machine. Just say 'no'.
This seems fundamentally flawed to me. If you cannot remove a device
from a machine using command line options, then how do we support
something like -net none? Do we make the default machine not contain a nic?
The value of a machine type to a user is that it presents a useful
machine--not a barebones machine. A user should not have to think about
which type of nic they need or whether they want to enable usb.
>
> That leaves two options:
>
> (1) create two devices, create new machines which use the kvm
> versions (aka -M pc-kvm).
> (2) make using the in-kernel kvm code a device property.
(3) Add the ability to remove device from a machine type.
--
Regards,
Anthony Liguori
next prev parent reply other threads:[~2009-10-12 14:04 UTC|newest]
Thread overview: 70+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-10-07 22:08 [Qemu-devel] [PATCH v2] Add in-kernel irqchip Glauber Costa
2009-10-07 22:08 ` [Qemu-devel] [PATCH v2 1/9] add base-addr field to io apic state Glauber Costa
2009-10-07 22:08 ` [Qemu-devel] [PATCH v2 2/9] Save missing fields in VMState Glauber Costa
2009-10-07 22:08 ` [Qemu-devel] [PATCH v2 3/9] provide in-kernel ioapic Glauber Costa
2009-10-07 22:08 ` [Qemu-devel] [PATCH v2 4/9] provide in-kernel apic Glauber Costa
2009-10-07 22:08 ` [Qemu-devel] [PATCH v2 5/9] provide apic_set_irq_delivered Glauber Costa
2009-10-07 22:08 ` [Qemu-devel] [PATCH v2 6/9] provide in-kernel i8259 chip Glauber Costa
2009-10-07 22:08 ` [Qemu-devel] [PATCH v2 7/9] initialize " Glauber Costa
2009-10-07 22:08 ` [Qemu-devel] [PATCH v2 8/9] Initialize in-kernel irqchip Glauber Costa
2009-10-07 22:08 ` [Qemu-devel] [PATCH v2 9/9] Add -kvm option Glauber Costa
2009-10-07 23:00 ` [Qemu-devel] " Anthony Liguori
2009-10-07 23:14 ` Glauber Costa
2009-10-07 23:28 ` Anthony Liguori
2009-10-12 11:58 ` Gerd Hoffmann
2009-10-12 14:04 ` Anthony Liguori [this message]
2009-10-12 15:34 ` Gerd Hoffmann
2009-10-12 16:31 ` Anthony Liguori
2009-10-13 8:05 ` Gerd Hoffmann
2009-10-13 15:33 ` Anthony Liguori
2009-10-13 19:26 ` Markus Armbruster
2009-10-13 20:43 ` Anthony Liguori
2009-10-14 8:08 ` Gerd Hoffmann
2009-10-13 15:36 ` Anthony Liguori
2009-10-13 22:57 ` Jamie Lokier
2009-10-08 14:07 ` Avi Kivity
2009-10-08 14:23 ` Anthony Liguori
2009-10-08 14:30 ` Avi Kivity
2009-10-08 14:33 ` Anthony Liguori
2009-10-08 14:41 ` Avi Kivity
2009-10-08 14:56 ` Anthony Liguori
2009-10-08 15:05 ` Avi Kivity
2009-10-08 15:14 ` Anthony Liguori
2009-10-12 11:15 ` Gerd Hoffmann
2009-10-08 14:34 ` Gleb Natapov
2009-10-08 12:02 ` [Qemu-devel] Re: [PATCH v2 7/9] initialize i8259 chip Jan Kiszka
2009-10-08 13:55 ` [Qemu-devel] Re: [PATCH v2 4/9] provide in-kernel apic Anthony Liguori
2009-10-08 14:09 ` Avi Kivity
2009-10-08 14:22 ` Glauber Costa
2009-10-09 10:06 ` Jamie Lokier
2009-10-09 14:30 ` Glauber Costa
2009-10-09 16:48 ` Jamie Lokier
2009-10-09 18:06 ` Glauber Costa
2009-10-09 19:49 ` Anthony Liguori
2009-10-11 9:10 ` Avi Kivity
2009-10-12 13:41 ` Anthony Liguori
2009-10-08 14:26 ` Anthony Liguori
2009-10-08 14:31 ` Avi Kivity
2009-10-08 14:39 ` Anthony Liguori
2009-10-08 14:46 ` Glauber Costa
2009-10-08 14:44 ` Glauber Costa
2009-10-08 11:46 ` [Qemu-devel] Re: [PATCH v2 3/9] provide in-kernel ioapic Jan Kiszka
2009-10-08 13:49 ` Anthony Liguori
2009-10-08 13:54 ` Avi Kivity
2009-10-08 15:53 ` Jan Kiszka
2009-10-08 16:07 ` Jamie Lokier
2009-10-08 16:12 ` Anthony Liguori
2009-10-08 16:17 ` Avi Kivity
2009-10-08 16:22 ` Gleb Natapov
2009-10-08 16:29 ` Avi Kivity
2009-10-08 16:34 ` Gleb Natapov
2009-10-08 16:42 ` Avi Kivity
2009-10-08 17:11 ` Gleb Natapov
2009-10-09 10:02 ` Jamie Lokier
2009-10-09 12:02 ` Gleb Natapov
2009-10-09 14:32 ` Glauber Costa
2009-10-09 16:49 ` Jamie Lokier
[not found] ` <m31vlctk9q.fsf@neno.mitica>
2009-10-09 21:34 ` Glauber Costa
2009-10-12 13:20 ` Anthony Liguori
2009-10-12 14:18 ` Jamie Lokier
2009-10-12 14:49 ` Anthony Liguori
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4AD33774.1030001@us.ibm.com \
--to=aliguori@us.ibm.com \
--cc=andre.przywara@amd.com \
--cc=glommer@redhat.com \
--cc=kraxel@redhat.com \
--cc=qemu-devel@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).