Re: [libvirt] virNodeDevice APIs and VFIO

2013-04-24 Thread Laine Stump
On 04/23/2013 12:34 PM, Laine Stump wrote: On 04/23/2013 11:46 AM, Daniel P. Berrange wrote: On Tue, Apr 23, 2013 at 10:52:09AM -0400, Laine Stump wrote: Yesterday for the first time I consciously noticed the virNodeDeviceDettach and virNodeDeviceReAttach APIs, and found that they are

Re: [libvirt] virNodeDevice APIs and VFIO

2013-04-24 Thread Daniel P. Berrange
On Wed, Apr 24, 2013 at 09:56:34AM -0400, Laine Stump wrote: On 04/23/2013 12:34 PM, Laine Stump wrote: On 04/23/2013 11:46 AM, Daniel P. Berrange wrote: On Tue, Apr 23, 2013 at 10:52:09AM -0400, Laine Stump wrote: Yesterday for the first time I consciously noticed the

Re: [libvirt] virNodeDevice APIs and VFIO

2013-04-24 Thread Laine Stump
On 04/24/2013 10:01 AM, Daniel P. Berrange wrote: On Wed, Apr 24, 2013 at 09:56:34AM -0400, Laine Stump wrote: On 04/23/2013 12:34 PM, Laine Stump wrote: On 04/23/2013 11:46 AM, Daniel P. Berrange wrote: On Tue, Apr 23, 2013 at 10:52:09AM -0400, Laine Stump wrote: Yesterday for the first time

[libvirt] virNodeDevice APIs and VFIO

2013-04-23 Thread Laine Stump
Yesterday for the first time I consciously noticed the virNodeDeviceDettach and virNodeDeviceReAttach APIs, and found that they are hardcoded to bind to/unbind from the pci-stub driver for qemu, and the pciback driver for Xen. If we want these APIs to be useful for VFIO, they will need to bind to

Re: [libvirt] virNodeDevice APIs and VFIO

2013-04-23 Thread Eric Blake
On 04/23/2013 08:52 AM, Laine Stump wrote: Yesterday for the first time I consciously noticed the virNodeDeviceDettach and virNodeDeviceReAttach APIs, and found that they are hardcoded to bind to/unbind from the pci-stub driver for qemu, and the pciback driver for Xen. If we want these APIs to

Re: [libvirt] virNodeDevice APIs and VFIO

2013-04-23 Thread Daniel P. Berrange
On Tue, Apr 23, 2013 at 10:52:09AM -0400, Laine Stump wrote: Yesterday for the first time I consciously noticed the virNodeDeviceDettach and virNodeDeviceReAttach APIs, and found that they are hardcoded to bind to/unbind from the pci-stub driver for qemu, and the pciback driver for Xen. If we

Re: [libvirt] virNodeDevice APIs and VFIO

2013-04-23 Thread Laine Stump
On 04/23/2013 11:46 AM, Daniel P. Berrange wrote: On Tue, Apr 23, 2013 at 10:52:09AM -0400, Laine Stump wrote: Yesterday for the first time I consciously noticed the virNodeDeviceDettach and virNodeDeviceReAttach APIs, and found that they are hardcoded to bind to/unbind from the pci-stub