From: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
To: Ian Campbell <Ian.Campbell@citrix.com>
Cc: "Peter X. Gao" <peterxianggao@gmail.com>,
Xen-devel@lists.xenproject.org,
Fabio Fantoni <fabio.fantoni@m2r.biz>,
Wei Liu <wei.liu2@citrix.com>,
xen-users@lists.xen.org
Subject: Re: Virtio on Xen
Date: Wed, 5 Mar 2014 13:02:15 -0500 [thread overview]
Message-ID: <20140305180215.GB9058@phenom.dumpdata.com> (raw)
In-Reply-To: <1392123700.26657.112.camel@kazak.uk.xensource.com>
On Tue, Feb 11, 2014 at 01:01:40PM +0000, Ian Campbell wrote:
> On Mon, 2014-02-10 at 10:07 -0800, Peter X. Gao wrote:
> > Thanks for your reply. I am now using virtio-net and it seems working.
> > However, Intel DPDK also requires hugepage. When a DPDK application is
> > initiating hugepage, I got the following error. Do I need to config
> > something in Xen to support hugepage?
>
> I'm not sure about the status of superpage support in mainline kernels
> for PV Xen guests. IIRC there was a requirement to add a Xen command
> line flag to enable it at the level.
>
> Or you could just use an HVM guest, since no special support is needed
> for hugepages there.
>
> But maybe I'm confused because I think your use of virtio-net would
> necessarily require that you be using an HVM not PV guest.
>
> But then looking at your logs I see Xen PV block and net but no sign of
> virtio -- so I suspect you are actually doing PV and not using
> virtio-net at all.
DPDK 1.6 is out - and you can do Xen. You need to use HVM guests and
a special module in dom0 to setup 2MB contingous pages that
is shared with the guest.
The protocol that DPDK uses is VirtIO.
See:
http://dpdk.org/browse/dpdk/commit/?id=47bd46112b710dc59b1becfb67e18da319c5debe
http://dpdk.org/browse/dpdk/commit/?id=148f963fb5323c1c6b6d5cea95084deb25cc73f8
>
> Ian.
>
>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel
next prev parent reply other threads:[~2014-03-05 18:02 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-02-07 21:19 Virtio on Xen Peter X. Gao
2014-02-10 10:04 ` Ian Campbell
2014-02-10 10:42 ` Wei Liu
2014-02-10 11:19 ` Fabio Fantoni
2014-02-10 18:07 ` Peter X. Gao
2014-02-11 13:01 ` Ian Campbell
2014-03-05 18:02 ` Konrad Rzeszutek Wilk [this message]
2014-03-05 22:33 ` Peter X. Gao
2014-02-11 13:06 ` Application using hugepages crashes PV guest Ian Campbell
-- strict thread matches above, loose matches on Subject: below --
2014-08-11 13:54 Virtio on Xen Divya Durgadas
2014-08-11 14:55 ` Wei Liu
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=20140305180215.GB9058@phenom.dumpdata.com \
--to=konrad.wilk@oracle.com \
--cc=Ian.Campbell@citrix.com \
--cc=Xen-devel@lists.xenproject.org \
--cc=fabio.fantoni@m2r.biz \
--cc=peterxianggao@gmail.com \
--cc=wei.liu2@citrix.com \
--cc=xen-users@lists.xen.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.