Re: [Xen-devel] [PATCH 1/6] xen: Add support for hiding and unhiding pcie passthrough devices

[Wei Liu <wei.liu2@xxxxxxxxxx>]

On Thu, Jul 06, 2017 at 02:45:18AM -0600, Jan Beulich wrote:
> >>> On 05.07.17 at 21:38, <venu.busireddy@xxxxxxxxxx> wrote:
> > On 2017-07-04 09:46:58 -0600, Jan Beulich wrote:
> >> >>> On 27.06.17 at 19:14, <venu.busireddy@xxxxxxxxxx> wrote:
> >> 
> >> First of all, please Cc all maintainers of code you modify.
> > 
> > I was using the names spit out by the scripts/get_maintainer.pl script
> > for the patch file. I didn't know that the script had a "-f" option, and
> > without it, the script spits out only two names, which I included. I now
> > have Cc'ed all the names that the "-f" option produced. Interestingly,
> > Daniel's name is not in the "-f" output, and hence, I am still confused
> > what the correct list is!
> 
> I can't talk about the script, except that it is known to have
> limitations. Generally, changes to the public interface should be
> Cc-ed to all REST maintainers.
> 
> >> > Add support for hiding and unhiding (by introducing two new hypercall
> >> > subops) pci devices that trigger AER fatal errors while assigned to
> >> > guests in passthrough mode. Hiding of the device is done by assigning
> >> > it to dom_xen dummy domain.
> >> 
> >> Would you mind explaining why simply de-assigning the device
> >> (with an existing operation) isn't suitable here? (This explanation
> >> would presumably belong either in the description here or in the
> >> cover letter.)
> > 
> > My initial thinking (for the first revision) was that the guest and
> > the device together are party to the evil things, and hence the guest
> > should be killed. But I agree that unassigning the device should be
> > sufficient. Once the device is removed, the guest can't do much that
> > any other guest can't. Therefore, I will change this patchset to simply
> > unassign the device from the guest.
> > 
> > Is that acceptable?
> 
> I think so, but I may be missing parts of your reasoning as to why
> hiding the device may be a good thing.

My thought exactly.

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