Xen 
 
Home Products Support Community News
 
   
 

xen-devel

[Xen-devel] Re: [PATCH] Fix xm pci-attach/detach for inactive managed do

To: Masaki Kanno <kanno.masaki@xxxxxxxxxxxxxx>
Subject: [Xen-devel] Re: [PATCH] Fix xm pci-attach/detach for inactive managed domains
From: Simon Horman <horms@xxxxxxxxxxxx>
Date: Fri, 15 May 2009 10:58:45 +1000
Cc: xen-devel@xxxxxxxxxxxxxxxxxxx
Delivery-date: Thu, 14 May 2009 17:59:16 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <20090515003015.GA23201@xxxxxxxxxxxx>
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
References: <5CC9D493E098B9kanno.masaki@xxxxxxxxxxxxxx> <20090515003015.GA23201@xxxxxxxxxxxx>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Mutt/1.5.18 (2008-05-17)
On Fri, May 15, 2009 at 10:30:15AM +1000, Simon Horman wrote:
> On Thu, May 14, 2009 at 09:59:55PM +0900, Masaki Kanno wrote:
> Content-Description: Mail message body
> > Hi,
> > 
> > Currently, xm pci-attach and xm pci-detach don't work for inactive 
> > managed domains.  This patch solves the problems.
> > 
> > # xm new vm1
> > Using config file "/etc/xen/vm1".
> > # xm pci-list vm1
> > VSlt domain   bus   slot   func
> > 0x6  0x0      0x01  0x00   0x0
> > 0x7  0x0      0x01  0x00   0x1
> > # xm pci-attach vm1 00:1a.1 8
> > Error: 'vslot'
> > Usage: xm pci-attach [-o|--options=<opt>] <Domain> <domain:bus:slot.func> 
> > [virtual slot]
> > 
> > Insert a new pass-through pci device.
> > # xm pci-detach vm1 01:00.1
> > Error: 'vslot'
> > Usage: xm pci-detach <Domain> <domain:bus:slot.func>
> > 
> > Remove a domain's pass-through pci device.
> 
> Good morning Kanno-san,
> 
> unfortunately this change does not seem to be correct to me.
> 
> In the case of pci-attach if I follow the folliwing steps:
> 
> $ xm new hvm.conf
> $ xm pci-attach hvm 00:1a.0
> $ xm start hvm
> [ wait a bit ]
> $ xm pci-list hvm
> 
> Then pci-list never indicates that 00:1a.0 has been assigned a vslot.
> 
> In the case of pci-detach, I am wondering if it makes any
> sense to detach a function from a paused domain as the guest os
> really needs to release the device for pci-detach to complete successfully.
> 
> 
> On an semi-related note, I am seeing an error when trying to connect
> to the console when I start a domain. I will investigate further.
> 
> $ xm new hvm.conf
> $ xm start -c hvm
> /usr/lib64/xen/bin/xenconsole: invalid option -- 1
> Invalid number of arguments
> Try `/usr/lib64/xen/bin/xenconsole --help' for more information.

Sorry, this was not quite correct. The console seems to work
but runs into trouble if the domain is restarted after
pci-attach was performed while the guest was running.

The following is without any patches applied.

$ xm new hvm.conf
$ xm start -c hvm
[ wait a bit, guest boots]
^]
[ back to host console ]
$ xm pci-attach hvm 00:1a.0
$ xm pci-list hgm
[ looks ok ]
$ xm destroy debian
$ xm start -c debian
/usr/lib64/xen/bin/xenconsole: invalid option -- 1
Invalid number of arguments
Try `/usr/lib64/xen/bin/xenconsole --help' for more information.
$ xm list
ame                                        ID   Mem VCPUs      State  Time(s)
Domain-0                                     0  1024     2     r-----  413.1
hvm                                              128     1               0.0
$ xm start -c debian
xenconsole: Could not read tty from store: No such file or directory
$ xm list
xm list
Name                                        ID   Mem VCPUs      State Time(s)
Domain-0                                     0  1024     2     r-----  413.9
debian                                      17   128     1     ----c-    0.0

*sigh*, it feels like there are many bugs in there.

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