On 06/22/10 - 03:33:43PM, Eric Blake wrote:
On 06/15/2010 06:35 AM, Chris Lalancette wrote:
> On 06/15/10 - 01:13:00PM, Daniel P. Berrange wrote:
>> You're accessing 'driver' here without first locking it
>
> D'oh, of course. Thanks for that. Updated patch below.
>
>
>>From 56b1d0463790b52e23975968c54d4f9f9a3c5fbd Mon Sep 17 00:00:00 2001
> From: Chris Lalancette <clalance(a)redhat.com>
> Date: Mon, 14 Jun 2010 17:12:35 -0400
> Subject: [PATCH v2 1/2] Check for active PCI devices when doing nodedevice
operations.
>
> In the current libvirt PCI code, there is no checking whether
> a PCI device is in use by a guest when doing node device
> detach or reattach. This causes problems when a device is
> assigned to a guest, and the administrator starts issuing
> nodedevice commands. Make it so that we check the list
> of active devices when trying to detach/reattach, and only
> allow the operation if the device is not assigned to a guest.
>
> v2: Add locking to qemudNodeDeviceDettach and qemudNodeDeviceReAttach
>
> Signed-off-by: Chris Lalancette <clalance(a)redhat.com>
> ---
> - pciDettachDevice(dev) < 0)
> + pciDettachDevice(dev, driver->activePciHostdevs) < 0)
ACK.
Thanks, I pushed this now.
--
Chris Lalancette