Re: [PATCH v4 RFC 2/6] x86/MSI-X: track host and guest mask‑all requests separately

[Andrew Cooper <andrew.cooper3@citrix.com>]

On 25/06/15 09:01, Jan Beulich wrote:
>>>> On 24.06.15 at 19:15, <andrew.cooper3@citrix.com> wrote:
>> On 22/06/15 15:47, Jan Beulich wrote:
>>> Host uses of the bits will be added subsequently, and must not be
>>> overridden by guests (including Dom0, namely when acting on behalf of
>>> a guest).
>>>
>>> Signed-off-by: Jan Beulich <jbeulich@suse.com>
>>>
>>> --- a/xen/arch/x86/msi.c
>>> +++ b/xen/arch/x86/msi.c
>>> @@ -846,6 +846,12 @@ static int msix_capability_init(struct p
>>>  
>>>      if ( !msix->used_entries )
>>>      {
>>> +        msix->host_maskall = 0;
>>> +        if ( !msix->guest_maskall )
>>> +            control &= ~PCI_MSIX_FLAGS_MASKALL;
>>> +        else
>>> +            control |= PCI_MSIX_FLAGS_MASKALL;
>> Is the guest (or hardware) in a position to influence guest_maskall at
>> this point?  I am not sure that it is.
> Of course - via the cfg write intercept (i.e. the hunk immediately
> following this one).

So it is.  I had got the call chronology confused.

Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>  (although
perhaps better to wait until we understand Sanders MSI masking problem).

~Andrew