Re: [PATCH v8 6/8] xen/arm: introduce GNTTABOP_cache_flush

[Ian Campbell <Ian.Campbell@citrix.com>]

On Tue, 2014-10-21 at 10:14 +0100, Jan Beulich wrote:
> >>> On 21.10.14 at 11:04, <Ian.Campbell@citrix.com> wrote:
> > On Mon, 2014-10-20 at 17:31 +0100, Stefano Stabellini wrote:
> >> On Mon, 20 Oct 2014, Jan Beulich wrote:
> >> > >>> On 20.10.14 at 17:24, <Ian.Campbell@citrix.com> wrote:
> >> > > On Mon, 2014-10-20 at 10:48 +0100, Stefano Stabellini wrote:
> >> > >>  
> >> > >> @@ -574,6 +576,25 @@ struct gnttab_swap_grant_ref {
> >> > >>  typedef struct gnttab_swap_grant_ref gnttab_swap_grant_ref_t;
> >> > >>  DEFINE_XEN_GUEST_HANDLE(gnttab_swap_grant_ref_t);
> >> > >>  
> >> > >> +/*
> >> > >> + * Issue one or more cache maintenance operations on a portion of a
> >> > >> + * page granted to the calling domain by a foreign domain.
> >> > >> + */
> >> > >> +struct gnttab_cache_flush {
> >> > >> +    union {
> >> > >> +        uint64_t dev_bus_addr;
> >> > >> +        grant_ref_t ref;
> >> > >> +    } a;
> >> > >> +    uint16_t offset; /* offset from start of grant */
> >> > >> +    uint16_t length; /* size within the grant */
> >> > > 
> >> > > So are these not valid when used with dev_bus_addr? I can see that being
> >> > > the case for offset, but length too?
> >> > 
> >> > Both ought to be valid, and dev_bus_addr should be page aligned.
> >> 
> >> That's right.
> > 
> > I would have expected you to correct the comment in the repost then
> > (which I notice also still says arm in the subject)
> 
> I have to admit I'm having trouble seeing which of the comments
> you think is wrong/misleading.

start of grant/size within the grant, in the context of using
dev_bus_addr, seemed strange/inconsistent.

But you made me think again and of course dev_bus_addr is required to
refer to a granted page isn't it. So I retract that comment.

Is it not worth being explicit about dev_bus_addr being page aligned and
equal to something previously returned by a map operation?

Ian.