Re: [Xen-devel] [PATCH v8 3/3] VT-d: Fix vt-d Device-TLB flush timeout issue

["Xu, Quan" <quan.xu@xxxxxxxxx>]

On March 24, 2016 11:38pm, <dario.faggioli@xxxxxxxxxx> wrote:
> On Thu, 2016-03-24 at 13:57 +0800, Quan Xu wrote:
> > If Device-TLB flush timed out, we would hide the target ATS device and
> > crash the domain owning this ATS device. If impacted domain is
> > hardware domain, just throw out a warning.
> >
> > The hidden device should be disallowed to be further assigned to any
> > domain.
> >
> What is "should be disallowed" supposed to mean here? Isn't the situation 
> that,
> by hiding the device, which this patch is doing, we actually disallow any 
> further
> assignment?
> 
> If yes, this should rather be (something like):
> 
> "By hiding the device, we make sure it can't be assigned to any domain any
> longer."
> 

 Let me introduce what happen first.

If the ATS device flush timed out, it would be assigned to dom_xen (dummy 
domain).
Then, .e.g, the ATS device is 0000:81:00.0 in my platform. If the ATS device 
0000:81:00.0 is hidden and we assign the ATS device 0000:81:00.0 to another 
domain with xl tool,
It would fail and print out:
   """ libxl: error: libxl_pci.c:1216:libxl__device_pci_add: PCI device 
0000:81:00.0 already assigned to a different guest? """

To be honest, I'd prefer your description. do you think?

Quan

> Other than this, the patch looks good to me, but I'll re-review it when the 
> new
> version comes out (with the other patches from the preliminary series folded 
> in),
> before saying Reviewed-by.

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel