RE: [Xen-devel] [PATCH] Add VMDq support to ixgbe

["Fischer, Anna" <anna.fischer@xxxxxx>]

> Subject: [Xen-devel] [PATCH] Add VMDq support to ixgbe
> 
> This patch adds experimental VMDq support (AKA Netchannel2 vmq) to the
> ixgbe driver.  This applies to the Netchannel2 tree, and should NOT be
> applied to the "normal" development tree.
> 
> To enable VMDq functionality, load the driver with the command-line
> parameter VMDQ=<num queues>, as in:
> 
> $ modprobe ixgbe VMDQ=8
> 
> You can then set up PV domains to use the device by modifying your VM
> configuration file from
>      vif = [ '<whatever>' ]
> to
>      vif2 = [ 'pdev=<netdev>' ]
> where <netdev> is the interface name for your 82598 board, e.g peth0 in
> dom0.
> 
> The Netchannel2 code is VERY experimental at this stage and should not
> be used in production environments.  This patch is intended to support
> further development and testing efforts.

I have ported the netchannel2 frontend driver to run as PV-on-HVM network 
driver in an HVM guest. I am running only a single guest at the moment. I use a 
bridged setup, so the backend drivers plug into a bridge in dom0. 

My guest needs multiple VIFs. This works fine as long as I never have multiple 
VIFs plugged into the same bridge. I have tried with multiple queues from 
different NIC ports, and with multiple queues from the same NIC port. As long 
as I have a separate bridge per VIF, then my guest networking is up and running 
nicely. 

As soon as I have two or more VIFs plugged into the same bridge that connects 
to the physical network (using standard Xen setup scripts), then the system 
freezes if there is network traffic. It does not freeze on the first packet I 
believe, but very shortly after network communication takes place. I would 
attach logs but as the system freezes there are no error messages or anything 
that could indicate what the problem is. Is there a reason why netchannel2 VMDQ 
support does not work when multiple queues (or backend devices) share the same 
bridge?

Thanks,
Anna


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel