Shall we clear the cpumap in XEN_DOMCTL_setvcpuaffinity

Shall we clear the cpumap in XEN_DOMCTL_setvcpuaffinity

[Jiang, Yunhong]

When browse code, I noticed if the op->u.vcpuaffinity.cpumap is NULL, a random value of new_affinity will be passed to vcpu_set_affinity(), as in followed code.

I'm not sure if this is expected result. IMO, we should return -EINVAL if no cpu specified, as the return value at vcpu_set_affinity() for empty cpumap.

Did I missed anything?

    case XEN_DOMCTL_getvcpuaffinity:
    {
        domid_t dom = op->domain;
        struct domain *d = rcu_lock_domain_by_id(dom);
        struct vcpu *v;
        cpumask_t new_affinity;
......
        if ( op->cmd == XEN_DOMCTL_setvcpuaffinity )
        {
            xenctl_cpumap_to_cpumask(		>>>>> the new_affinity is not initialized for if cpumap==NULL
                &new_affinity, &op->u.vcpuaffinity.cpumap);
            ret = vcpu_set_affinity(v, &new_affinity);
        }

Thanks
--jyh

Re: Shall we clear the cpumap in XEN_DOMCTL_setvcpuaffinity

[Keir Fraser]

On 12/05/2010 08:22, "Jiang, Yunhong" <yunhong.jiang@intel.com> wrote:

> When browse code, I noticed if the op->u.vcpuaffinity.cpumap is NULL, a random
> value of new_affinity will be passed to vcpu_set_affinity(), as in followed
> code.
> 
> I'm not sure if this is expected result. IMO, we should return -EINVAL if no
> cpu specified, as the return value at vcpu_set_affinity() for empty cpumap.

Xen-unstable:21350.

 Thanks,
 Keir

> Did I missed anything?
> 
>     case XEN_DOMCTL_getvcpuaffinity:
>     {
>         domid_t dom = op->domain;
>         struct domain *d = rcu_lock_domain_by_id(dom);
>         struct vcpu *v;
>         cpumask_t new_affinity;
> ......
>         if ( op->cmd == XEN_DOMCTL_setvcpuaffinity )
>         {
>             xenctl_cpumap_to_cpumask(  >>>>> the new_affinity is not
> initialized for if cpumap==NULL
>                 &new_affinity, &op->u.vcpuaffinity.cpumap);
>             ret = vcpu_set_affinity(v, &new_affinity);
>         }
> 
> Thanks
> --jyh

RE: Shall we clear the cpumap in XEN_DOMCTL_setvcpuaffinity

[Jiang, Yunhong]

Thanks, although "no changes found" after hg pull :-)

--jyh

>-----Original Message-----
>From: Keir Fraser [mailto:keir.fraser@eu.citrix.com]
>Sent: Wednesday, May 12, 2010 3:43 PM
>To: Jiang, Yunhong
>Cc: xen-devel@lists.xensource.com
>Subject: Re: Shall we clear the cpumap in XEN_DOMCTL_setvcpuaffinity
>
>On 12/05/2010 08:22, "Jiang, Yunhong" <yunhong.jiang@intel.com> wrote:
>
>> When browse code, I noticed if the op->u.vcpuaffinity.cpumap is NULL, a random
>> value of new_affinity will be passed to vcpu_set_affinity(), as in followed
>> code.
>>
>> I'm not sure if this is expected result. IMO, we should return -EINVAL if no
>> cpu specified, as the return value at vcpu_set_affinity() for empty cpumap.
>
>Xen-unstable:21350.
>
> Thanks,
> Keir
>
>> Did I missed anything?
>>
>>     case XEN_DOMCTL_getvcpuaffinity:
>>     {
>>         domid_t dom = op->domain;
>>         struct domain *d = rcu_lock_domain_by_id(dom);
>>         struct vcpu *v;
>>         cpumask_t new_affinity;
>> ......
>>         if ( op->cmd == XEN_DOMCTL_setvcpuaffinity )
>>         {
>>             xenctl_cpumap_to_cpumask(  >>>>> the new_affinity is not
>> initialized for if cpumap==NULL
>>                 &new_affinity, &op->u.vcpuaffinity.cpumap);
>>             ret = vcpu_set_affinity(v, &new_affinity);
>>         }
>>
>> Thanks
>> --jyh
>

Re: Shall we clear the cpumap in XEN_DOMCTL_setvcpuaffinity

[Keir Fraser]

It's in the staging tree
http://xenbits.xensource.com/staging/xen-unstable.hg

 -- Keir

On 12/05/2010 08:57, "Jiang, Yunhong" <yunhong.jiang@intel.com> wrote:

> Thanks, although "no changes found" after hg pull :-)
> 
> --jyh
> 
>> -----Original Message-----
>> From: Keir Fraser [mailto:keir.fraser@eu.citrix.com]
>> Sent: Wednesday, May 12, 2010 3:43 PM
>> To: Jiang, Yunhong
>> Cc: xen-devel@lists.xensource.com
>> Subject: Re: Shall we clear the cpumap in XEN_DOMCTL_setvcpuaffinity
>> 
>> On 12/05/2010 08:22, "Jiang, Yunhong" <yunhong.jiang@intel.com> wrote:
>> 
>>> When browse code, I noticed if the op->u.vcpuaffinity.cpumap is NULL, a
>>> random
>>> value of new_affinity will be passed to vcpu_set_affinity(), as in followed
>>> code.
>>> 
>>> I'm not sure if this is expected result. IMO, we should return -EINVAL if no
>>> cpu specified, as the return value at vcpu_set_affinity() for empty cpumap.
>> 
>> Xen-unstable:21350.
>> 
>> Thanks,
>> Keir
>> 
>>> Did I missed anything?
>>> 
>>>     case XEN_DOMCTL_getvcpuaffinity:
>>>     {
>>>         domid_t dom = op->domain;
>>>         struct domain *d = rcu_lock_domain_by_id(dom);
>>>         struct vcpu *v;
>>>         cpumask_t new_affinity;
>>> ......
>>>         if ( op->cmd == XEN_DOMCTL_setvcpuaffinity )
>>>         {
>>>             xenctl_cpumap_to_cpumask(  >>>>> the new_affinity is not
>>> initialized for if cpumap==NULL
>>>                 &new_affinity, &op->u.vcpuaffinity.cpumap);
>>>             ret = vcpu_set_affinity(v, &new_affinity);
>>>         }
>>> 
>>> Thanks
>>> --jyh
>> 
>