[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH v7a 09/14] xen/arm: move and rename is_vcpu_running function to sched.h



On Wed, Jul 2, 2014 at 4:49 PM, Ian Campbell <Ian.Campbell@xxxxxxxxxx> wrote:
> On Tue, 2014-07-01 at 08:55 +0100, Jan Beulich wrote:
>> >>> On 01.07.14 at 08:41, <vijay.kilari@xxxxxxxxx> wrote:
>> > @@ -829,6 +829,20 @@ void watchdog_domain_destroy(struct domain *d);
>> >  #define need_iommu(d)    (0)
>> >  #endif
>> >
>> > +static inline bool_t is_vcpu_online(const struct domain *d, unsigned int 
>> > vcpuid)
>>
>> I'm sorry for paying attention to this only now, but a function named
>> this way would ordinarily have a single const struct vcpu * parameter.
>> If the decoding of the (domain, vcpuid) tuple is to remain in the
>> function I guess it ought to be renamed to e.g. is_domain_vcpu_online()
>
> Looking at the callers it doesn't seem unreasonable to me to push the
> indexing of d->vcpus into them.
>
> d is used for the d->max_vcpus check -- which the callers all have
> already or are naturally avoiding and the check for d->vcpus == NULL,
> which I suspect is out of the question for the existing callers (but
> would need checking)

Yes, check for d->max_vcpus is already taken care by for_each_set_bit
and check for d->vcpus == NULL can be done by caller and
with this, is_vcpu_online() can take vcpu is parameter

>
>> (I'd further suggest to drop the is_ prefix, but I realize others may be
>> of different opinion).
>>
>> Jan
>>
>
>

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


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.