Re: [Xen-devel] [PATCH 1/7] iommu: introduce the concept of BFN...

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 16.03.18 at 11:31, <Paul.Durrant@xxxxxxxxxx> wrote:
>> From: Jan Beulich [mailto:JBeulich@xxxxxxxx]
>> Sent: 15 March 2018 13:40
>> 
>> >>> On 12.02.18 at 11:47, <paul.durrant@xxxxxxxxxx> wrote:
>> > --- a/xen/include/xen/iommu.h
>> > +++ b/xen/include/xen/iommu.h
>> > @@ -23,11 +23,15 @@
>> >  #include <xen/page-defs.h>
>> >  #include <xen/spinlock.h>
>> >  #include <xen/pci.h>
>> > +#include <xen/typesafe.h>
>> >  #include <public/hvm/ioreq.h>
>> >  #include <public/domctl.h>
>> >  #include <asm/device.h>
>> >  #include <asm/iommu.h>
>> >
>> > +TYPE_SAFE(unsigned long, bfn);
>> > +#define INVALID_BFN      _bfn(~0UL)
>> 
>> Please accompany this by a grep fodder (like the others have) and
>> perhaps also PRI_bfn. And while the type definition logically belongs
>> here, you will also want to add bfn_t with a description of its
>> purpose into the comment at the top of xen/mm.h. I guess you'll
>> need to replace / amend "host" in the MFN description there at the
>> same time.
> 
> Should I move the TYPE_SAFE evaluation to xen/mm.h then? If I leave it here 
> then I'll presumably need some ifdef hackery in mm.h if you want be to define 
> bfn_t there too.

No, I don't think the type definition itself needs to go there. If
anything, add a comment there saying that bfn_t lives in iommu.h.

Jan


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