From: Simon Horman <horms@verge.net.au>
To: "Han, Weidong" <weidong.han@intel.com>
Cc: "'xen-devel@lists.xensource.com'" <xen-devel@lists.xensource.com>
Subject: Re: Does it support to specify virtual function in guest for assigned device
Date: Wed, 9 Sep 2009 08:46:29 +1000 [thread overview]
Message-ID: <20090908224629.GA4075@verge.net.au> (raw)
In-Reply-To: <715D42877B251141A38726ABF5CABF2C054B5B4C35@pdsmsx503.ccr.corp.intel.com>
On Tue, Sep 08, 2009 at 04:10:57PM +0800, Han, Weidong wrote:
> Hi Simon,
>
> You did lots of work on multiple functions passthrough. We can specify
> virtual slot for assigned device like pci=[ '01:00.0@5' ]. Does it
> support to specify virtual function in guest for assigned device? For
> example, I want to specify virtual bdf 00:05.1 for device 01:00.1.
Hi Weidong,
Yes, as of changeset "xend: pass-through: Extend multi-function mapping"
(20012:3242351f9c67) is is possible to specify the virtual function.
However, this is only possible for multi-function pass-through.
When only a single function is passed through the virtual function
will always be zero, because PCI devices must always have function 0
present.
So the closest that you could do when passing through 01:00.1 is to get
virtual function 00:05.0 by specifying '01:00.1@5'.
If you passed through another function of 01:00 at the same time as 01:00.1,
say 01:00.0, then you could explicitly choose to have 01:00.1
assigned to 00:05.1 and 01:00.0 assigned to 00:05.0 using '00:05.0=0,1=1@5'.
Actually that would be the default as it is just an identity mapping,
so the same result would be achieved by '00:05.0,1@5'. A more interesting
case would be to reverse the mapping so 00:05.0 -> 01:00.1 and
00:05.1 -> 00:05.0, which could be achieved using '00:05.0=1,1=0@5'.
next prev parent reply other threads:[~2009-09-08 22:46 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-09-08 8:10 Does it support to specify virtual function in guest for assigned device Han, Weidong
2009-09-08 22:46 ` Simon Horman [this message]
2009-09-08 22:52 ` Han, Weidong
2009-09-08 23:17 ` Simon Horman
2009-09-09 1:37 ` Han, Weidong
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=20090908224629.GA4075@verge.net.au \
--to=horms@verge.net.au \
--cc=weidong.han@intel.com \
--cc=xen-devel@lists.xensource.com \
/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.