>>> On 15.03.18 at 12:48, <roger....@citrix.com> wrote:
> On Wed, Mar 14, 2018 at 10:51:07AM -0600, Jan Beulich wrote:
>> >>> On 14.03.18 at 15:04, <roger....@citrix.com> wrote:
>> > --- a/xen/drivers/vpci/vpci.c
>> > +++ b/xen/drivers/vpci/vpci.c
>> > @@ -47,6 +47,10 @@ void vpci_remove_device(struct pci_dev *pdev)
>> >          xfree(r);
>> >      }
>> >      spin_unlock(&pdev->vpci->lock);
>> > +#ifdef __XEN__
>> > +    /* NB: fields below are not exposed to the user-space test harness. */
>> > +    xfree(pdev->vpci->msi);
>> > +#endif
>> 
>> Would it maybe be better to add such dummy field(s), to avoid the
>> #ifdef here? Anyway, with or without that
>> Reviewed-by: Jan Beulich <jbeul...@suse.com>
> 
> For the msi structure that's doable, but for msix it's more complex
> because it includes the vpci_arch_msix_entry structure, and that would
> mean exposing more stuff to the user-space test harness.

I don't understand: These are pointers, hence it suffices to
use the structures here without actually defining their members
anywhere; you don't even need to declare them as an empty
structure. You'd just need to make sure that the test tool fills
the pointer fields with NULL (so that free()ing them is a no-op).

Jan


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Reply via email to