Re: [Xen-devel] [PATCH 6/18 V2]: PVH xen: Introduce PVH guest type

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 16.03.13 at 01:32, Mukesh Rathor <mukesh.rathor@xxxxxxxxxx> wrote:
> @@ -277,8 +280,8 @@ struct domain
>      struct rangeset *iomem_caps;
>      struct rangeset *irq_caps;
>  
> -    /* Is this an HVM guest? */
> -    bool_t           is_hvm;
> +    /* !is_pvh && !is_hvm ==> PV, else PVH or HVM */
> +    enum {hvm_guest=1, pvh_guest} guest_type;

Even if not used explicitly anywhere right now, please make the PV
guest case explicit here too (at once avoiding the =1).

Quite likely we'll want to also have the enum have a tag, so we
can eventually pass value from this enumeration to functions.

I'm also mildly puzzled by you using "_guest" suffixes here rather
than, as usual, a common prefix.

And of course, please properly format this.

> @@ -718,10 +725,14 @@ void watchdog_domain_destroy(struct domain *d);
>  
>  #define VM_ASSIST(_d,_t) (test_bit((_t), &(_d)->vm_assist))
>  
> -#define is_hvm_domain(d) ((d)->is_hvm)
> +#define is_hvm_domain(d) ((d)->guest_type == hvm_guest)
>  #define is_hvm_vcpu(v)   (is_hvm_domain(v->domain))
> +#define is_pvh_domain(d) ((d)->guest_type == pvh_guest)
> +#define is_pvh_vcpu(v)   (is_pvh_domain(v->domain))
>  #define is_pinned_vcpu(v) ((v)->domain->is_pinned || \
>                             cpumask_weight((v)->cpu_affinity) == 1)
> +#define is_hvm_or_pvh_domain(d) (is_hvm_domain(d) || is_pvh_domain(d))
> +#define is_hvm_or_pvh_vcpu(v) (is_hvm_or_pvh_domain(v->domain))

These surely can have better names, if they're needed at all:
Wouldn't !is_pv_domain() do what you need?

Jan


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