From: Andi Kleen <ak@suse.de>
To: Jeff Garzik <jeff@garzik.org>
Cc: Virtualization Mailing List <virtualization@lists.osdl.org>,
"H. Peter Anvin" <hpa@zytor.com>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
mathiasen@gmail.com, virtualization@lists.linux-foundation.org
Subject: Re: A set of "standard" virtual devices?
Date: Mon, 2 Apr 2007 23:36:42 +0200 [thread overview]
Message-ID: <200704022336.43136.ak@suse.de> (raw)
In-Reply-To: <4611768D.1080801@garzik.org>
On Monday 02 April 2007 23:33:01 Jeff Garzik wrote:
> Andi Kleen wrote:
> >> How would that work in the case where virtualized guests don't have a
> >> visible PCI bus, and the virtual environment doesn't pretend to emulate
> >> a PCI bus?
> >
> > If they emulated one with the appropiate device
> > then distribution driver auto probing would just work transparently for them.
>
> Yes, but, ideally with paravirtualization you should be able to avoid
> the overhead of emulating many major classes of device (storage,
> network, RNG, etc.) by developing a low-overhead passthrough interface
> that does not involve PCI at all.
The implementation wouldn't need to use PCI at all. There wouldn't
even need to be PCI like registers internally. Just a pci device
with an ID somewhere in sysfs. PCI with unique IDs
is just a convenient and well established key into the driver module
collection. Once you have the right driver it can do what it wants.
-Andi
WARNING: multiple messages have this Message-ID (diff)
From: Andi Kleen <ak@suse.de>
To: Jeff Garzik <jeff@garzik.org>
Cc: virtualization@lists.linux-foundation.org,
Jeremy Fitzhardinge <jeremy@goop.org>,
"H. Peter Anvin" <hpa@zytor.com>,
Virtualization Mailing List <virtualization@lists.osdl.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
mathiasen@gmail.com
Subject: Re: A set of "standard" virtual devices?
Date: Mon, 2 Apr 2007 23:36:42 +0200 [thread overview]
Message-ID: <200704022336.43136.ak@suse.de> (raw)
In-Reply-To: <4611768D.1080801@garzik.org>
On Monday 02 April 2007 23:33:01 Jeff Garzik wrote:
> Andi Kleen wrote:
> >> How would that work in the case where virtualized guests don't have a
> >> visible PCI bus, and the virtual environment doesn't pretend to emulate
> >> a PCI bus?
> >
> > If they emulated one with the appropiate device
> > then distribution driver auto probing would just work transparently for them.
>
> Yes, but, ideally with paravirtualization you should be able to avoid
> the overhead of emulating many major classes of device (storage,
> network, RNG, etc.) by developing a low-overhead passthrough interface
> that does not involve PCI at all.
The implementation wouldn't need to use PCI at all. There wouldn't
even need to be PCI like registers internally. Just a pci device
with an ID somewhere in sysfs. PCI with unique IDs
is just a convenient and well established key into the driver module
collection. Once you have the right driver it can do what it wants.
-Andi
next prev parent reply other threads:[~2007-04-02 21:36 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-04-02 20:18 A set of "standard" virtual devices? H. Peter Anvin
2007-04-02 20:31 ` Roland Dreier
2007-04-02 20:33 ` H. Peter Anvin
2007-04-02 21:49 ` Roland Dreier
2007-04-02 20:56 ` Jeremy Fitzhardinge
2007-04-02 21:12 ` Andi Kleen
2007-04-02 21:33 ` Jeff Garzik
2007-04-02 21:33 ` Jeff Garzik
2007-04-02 21:36 ` Andi Kleen [this message]
2007-04-02 21:36 ` Andi Kleen
2007-04-02 21:42 ` Jeremy Fitzhardinge
2007-04-02 21:42 ` Jeremy Fitzhardinge
2007-04-02 21:53 ` Anthony Liguori
2007-04-02 21:53 ` Anthony Liguori
2007-04-02 22:04 ` Jeremy Fitzhardinge
2007-04-02 22:10 ` H. Peter Anvin
2007-04-02 22:10 ` H. Peter Anvin
2007-04-02 22:25 ` Jeff Garzik
2007-04-02 22:25 ` Jeff Garzik
2007-04-02 22:30 ` H. Peter Anvin
2007-04-02 22:30 ` H. Peter Anvin
2007-04-03 9:41 ` Arnd Bergmann
2007-04-03 10:41 ` Cornelia Huck
2007-04-03 12:15 ` Arnd Bergmann
2007-04-03 13:39 ` Cornelia Huck
2007-04-03 14:03 ` Arnd Bergmann
2007-04-03 16:07 ` Cornelia Huck
2007-04-03 8:29 ` Christian Borntraeger
2007-04-03 8:30 ` Andi Kleen
2007-04-03 9:17 ` Cornelia Huck
2007-04-03 9:26 ` Andi Kleen
2007-04-03 10:51 ` Cornelia Huck
2007-04-03 15:00 ` Adrian Bunk
2007-04-03 17:50 ` Arnd Bergmann
2007-04-03 17:50 ` Arnd Bergmann
2007-04-03 19:07 ` Jeremy Fitzhardinge
2007-04-03 19:42 ` Arnd Bergmann
2007-04-03 19:55 ` Jeremy Fitzhardinge
2007-04-03 20:03 ` H. Peter Anvin
2007-04-03 20:03 ` H. Peter Anvin
2007-04-03 21:00 ` Jeremy Fitzhardinge
2007-04-03 21:00 ` Jeremy Fitzhardinge
2007-04-03 21:45 ` H. Peter Anvin
2007-04-03 21:51 ` Arnd Bergmann
2007-04-03 22:10 ` H. Peter Anvin
2007-04-03 22:10 ` H. Peter Anvin
2007-04-03 22:49 ` Arnd Bergmann
2007-04-04 0:52 ` H. Peter Anvin
2007-04-04 13:11 ` Arnd Bergmann
2007-04-04 15:50 ` H. Peter Anvin
2007-04-03 20:50 ` Arnd Bergmann
2007-04-03 20:50 ` Arnd Bergmann
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=200704022336.43136.ak@suse.de \
--to=ak@suse.de \
--cc=hpa@zytor.com \
--cc=jeff@garzik.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mathiasen@gmail.com \
--cc=virtualization@lists.linux-foundation.org \
--cc=virtualization@lists.osdl.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.