From: "Or Sagi" <ors-DHAoeUZcRjAAvxtiuMwx3w@public.gmane.org>
To: <linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>
Cc: kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
Subject: [RFC] Deferred interrupt handling.
Date: Wed, 18 Jul 2007 12:53:19 +0300 [thread overview]
Message-ID: <005901c7c921$79c006c0$6d401440$@com> (raw)
Gentlepeople,
We're currently working on PCI device pass-through support for KVM. In this
model all physical hardware access to specific devices will be performed by
the VM, and not by the host.
In particular, this requires interrupt handling to be done by the guest --
The host shouldn't load the corresponding device driver or otherwise access
the device. Since the host kernel is not aware of the device semantics it
cannot acknowledge the interrupt at the device level.
As far as the host kernel is concerned the VM is a user level process. We
require the ability to forward interrupt handling to such entities. The
current kernel interrupt handling path doesn't allow deferring interrupt
handling _and_ acknowledgement.
We believe that this requires changing the current interrupt path in several
fashions:
0. Adding an IRQ_DEFERRED mechanism to the interrupt handling path. ISRs
returning IRQ_DEFERRED will keep the interrupt masked until future
acknowledge.
1. Deferred acknowledge mechanism which would acknowledge the APIC and
unmask
the interrupt.
An alternative idea was suggested at
http://lists.xensource.com/archives/html/xen-devel/2007-05/msg01148.html ---
which is based on reversing the polarity of the received interrupt. Once
the guest acknowledges the interrupt at the device level, the polarity is
reversed again by the host.
This may not be the best choice performance wise -- twice the interrupt
count.
Another issue we are aware of is the effect deferred handling will have on
other devices sharing the interrupt. This seems to be a problem regardless
of the solution.
We wouldn't like to assume VT-d availability right now.
Any ideas ? Thoughts ?
-- Ors
-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
next reply other threads:[~2007-07-18 9:53 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-07-18 9:53 Or Sagi [this message]
2007-07-18 11:30 ` [RFC] Deferred interrupt handling Alan Cox
[not found] ` <20070718123040.615795b7-v58gJUvfdfWUJIigds3554dd74u8MsAO@public.gmane.org>
2007-07-18 16:36 ` Dor Laor
[not found] ` <64F9B87B6B770947A9F8391472E032160CC16C92-yEcIvxbTEBqsx+V+t5oei8rau4O3wl8o3fe8/T/H7NteoWH0uzbU5w@public.gmane.org>
2007-07-18 16:47 ` Alan Cox
[not found] ` <20070718174736.0baf6d0e-v58gJUvfdfWUJIigds3554dd74u8MsAO@public.gmane.org>
2007-07-18 16:46 ` Avi Kivity
[not found] ` <469E43D5.70802-atKUWr5tajBWk0Htik3J/w@public.gmane.org>
2007-07-18 16:57 ` Alan Cox
[not found] ` <20070718175708.6968923e-v58gJUvfdfWUJIigds3554dd74u8MsAO@public.gmane.org>
2007-07-18 16:57 ` Avi Kivity
[not found] ` <469E4679.205-atKUWr5tajBWk0Htik3J/w@public.gmane.org>
2007-07-18 17:14 ` Alan Cox
2007-07-18 17:09 ` Dor Laor
[not found] ` <64F9B87B6B770947A9F8391472E032160CC16D06-yEcIvxbTEBqsx+V+t5oei8rau4O3wl8o3fe8/T/H7NteoWH0uzbU5w@public.gmane.org>
2007-07-18 17:33 ` Alan Cox
[not found] ` <20070718183358.0a0886aa-v58gJUvfdfWUJIigds3554dd74u8MsAO@public.gmane.org>
2007-07-18 21:29 ` Dor Laor
[not found] ` <64F9B87B6B770947A9F8391472E032160CC16F3C-yEcIvxbTEBqsx+V+t5oei8rau4O3wl8o3fe8/T/H7NteoWH0uzbU5w@public.gmane.org>
2007-07-18 21:46 ` Alan Cox
[not found] ` <20070718224649.3e49a978-v58gJUvfdfWUJIigds3554dd74u8MsAO@public.gmane.org>
2007-07-25 14:18 ` Pavel Machek
2007-07-18 22:23 ` Or Sagi
2007-07-18 19:14 ` Lennart Sorensen
[not found] ` <20070718191433.GB24545-1wCw9BSqJbv44Nm34jS7GywD8/FfD2ys@public.gmane.org>
2007-07-19 9:23 ` Avi Kivity
[not found] ` <469F2D8C.7020805-atKUWr5tajBWk0Htik3J/w@public.gmane.org>
2007-07-19 13:17 ` Lennart Sorensen
[not found] ` <20070719131718.GC5623-1wCw9BSqJbv44Nm34jS7GywD8/FfD2ys@public.gmane.org>
2007-07-19 13:38 ` Avi Kivity
[not found] ` <469F6941.9010901-atKUWr5tajBWk0Htik3J/w@public.gmane.org>
2007-07-19 16:23 ` Lennart Sorensen
2007-07-18 15:57 ` Lennart Sorensen
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='005901c7c921$79c006c0$6d401440$@com' \
--to=ors-dhaoeuzcrjaavxtiumwx3w@public.gmane.org \
--cc=kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.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