Re: [PATCH for-4.5] reset PCI devices on force removal even when QEMU returns error

[Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>]

On Fri, Dec 12, 2014 at 04:13:52PM +0100, Sander Eikelenboom wrote:
> Hi Konrad,
> 
> This doesn't seem to be applied yet, nor does it seem to have a release-(N)ACK 
> from you ?

Hm, Stefano:

- Is this a regression?

- What are the risks of this not going in? I presume that it just means
  we haven't reset it in sysfs. But the xc_deassign_device operation
  if not done will not affect the hypervisor - which will move the
  device to dom0 upon guest teardown.

> 
> --
> Sander
> 
> 
> 
> Friday, November 28, 2014, 5:53:09 PM, you wrote:
> 
> > On do_pci_remove when QEMU returns error, we just bail out early without
> > resetting the device. On domain shutdown we are racing with QEMU exiting
> > and most often QEMU closes the QMP connection before executing the
> > requested command.
> 
> > In these cases if force=1, it makes sense to go ahead with rest of the
> > PCI device removal, that includes resetting the device and calling
> > xc_deassign_device. Otherwise we risk not resetting the device properly
> > on domain shutdown.
> 
> > Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
> 
> > diff --git a/tools/libxl/libxl_pci.c b/tools/libxl/libxl_pci.c
> > index 316643c..0ac0b93 100644
> > --- a/tools/libxl/libxl_pci.c
> > +++ b/tools/libxl/libxl_pci.c
> > @@ -1243,7 +1245,7 @@ static int do_pci_remove(libxl__gc *gc, uint32_t domid,
> >              rc = ERROR_INVAL;
> >              goto out_fail;
> >          }
> > -        if (rc) {
> > +        if (rc && !force) {
> >              rc = ERROR_FAIL;
> >              goto out_fail;
> >          }
> 
>