Re: [RFC for-5.1 4/4] spapr: Don't allow unplug of NVLink2 devices

[David Gibson]

On Thu, Mar 26, 2020 at 01:27:40PM +0100, Greg Kurz wrote:
> On Thu, 26 Mar 2020 16:40:09 +1100
> David Gibson <address@hidden> wrote:
> 
> > Currently, we can't properly handle unplug of NVLink2 devices, because we
> > don't have code to tear down their special memory resources.  There's not
> > a lot of impetus to implement that. Since hardware NVLink2 devices can't
> > be hot unplugged, the guest side drivers don't usually support unplug
> > anyway.
> > 
> > Therefore, simply prevent unplug of NVLink2 devices.
> > 
> 
> This could maybe considered as a valid fix for 5.0 since this prevents
> guest crashes IIUC. But since this requires the two preliminary cleanup
> patches, I understand you may prefer to postpone that to 5.1.

Yeah, it's arguably a bug, but not a regression, so I'm inclined to
leave it to 5.1.

> 
> > Signed-off-by: David Gibson <address@hidden>
> > ---
> 
> Reviewed-by: Greg Kurz <address@hidden>
> 
> >  hw/ppc/spapr_pci.c | 5 +++++
> >  1 file changed, 5 insertions(+)
> > 
> > diff --git a/hw/ppc/spapr_pci.c b/hw/ppc/spapr_pci.c
> > index 55ca9dee1e..5c8262413a 100644
> > --- a/hw/ppc/spapr_pci.c
> > +++ b/hw/ppc/spapr_pci.c
> > @@ -1666,6 +1666,11 @@ static void spapr_pci_unplug_request(HotplugHandler 
> > *plug_handler,
> >              return;
> >          }
> >  
> > +        if (spapr_phb_is_nvlink_dev(pdev, phb)) {
> > +            error_setg(errp, "PCI: Cannot unplug NVLink2 devices");
> > +            return;
> > +        }
> > +
> >          /* ensure any other present functions are pending unplug */
> >          if (PCI_FUNC(pdev->devfn) == 0) {
> >              for (i = 1; i < 8; i++) {
> 

-- 
David Gibson                    | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au  | minimalist, thank you.  NOT _the_ _other_
                                | _way_ _around_!
http://www.ozlabs.org/~dgibson

signature.asc
Description: PGP signature