Hi, Alex
On 10/13/2015 11:27 PM, Alex Williamson wrote:
On Tue, 2015-10-13 at 16:41 +0800, Cao jin wrote:
In case user regret when hot-adding multi-function, should roll back,
device_del the function added but not exposed to the guest.
As Michael suggests, this patch should come first, before
On Tue, 2015-10-13 at 16:41 +0800, Cao jin wrote:
> In case user regret when hot-adding multi-function, should roll back,
> device_del the function added but not exposed to the guest.
As Michael suggests, this patch should come first, before we actually
enable multi-function hot-add.
> Signed-off
On Tue, Oct 13, 2015 at 05:54:34PM +0800, Cao jin wrote:
> Hi Michael
>
> On 10/13/2015 04:51 PM, Michael S. Tsirkin wrote:
> >On Tue, Oct 13, 2015 at 04:41:35PM +0800, Cao jin wrote:
> >>In case user regret when hot-adding multi-function, should roll back,
> >>device_del the function added but no
Hi Michael
On 10/13/2015 04:51 PM, Michael S. Tsirkin wrote:
On Tue, Oct 13, 2015 at 04:41:35PM +0800, Cao jin wrote:
In case user regret when hot-adding multi-function, should roll back,
device_del the function added but not exposed to the guest.
Signed-off-by: Cao jin
I think this patch s
On Tue, Oct 13, 2015 at 04:41:35PM +0800, Cao jin wrote:
> In case user regret when hot-adding multi-function, should roll back,
> device_del the function added but not exposed to the guest.
>
> Signed-off-by: Cao jin
I think this patch should come first, before we enable the
functionality that
In case user regret when hot-adding multi-function, should roll back,
device_del the function added but not exposed to the guest.
Signed-off-by: Cao jin
---
hw/pci/pci_host.c | 6 +-
hw/pci/pcie.c | 22 +-
2 files changed, 22 insertions(+), 6 deletions(-)
diff --git
