Xen devel - Feb 2008 - [PATCH] Fix xm vcpu-pin command

Hi,

When I tested xm vcpu-pin command, I encountered the following strange 
problem.  I encountered it on x86, but I didn''t encounter it on ia64. 

On x86:
 # xm vcpu-list
 Name                                ID  VCPU   CPU State   Time(s) CPU Affinity
 Domain-0                             0     0     0   -b-      31.3 any cpu
 Domain-0                             0     1     1   r--      30.1 any cpu
 # xm vcpu-pin Domain-0 0 64
 # xm vcpu-list
 Name                                ID  VCPU   CPU State   Time(s) CPU Affinity
 Domain-0                             0     0     0   -b-      31.5 0
 Domain-0                             0     1     1   r--      31.1 any cpu
 # xm vcpu-pin Domain-0 0 65
 # xm vcpu-list
 Name                                ID  VCPU   CPU State   Time(s) CPU Affinity
 Domain-0                             0     0     1   -b-      32.0 1
 Domain-0                             0     1     0   r--      31.6 any cpu

On ia64:
 # xm vcpu-list
 Name                                ID  VCPU   CPU State   Time(s) CPU Affinity
 Domain-0                             0     0     0   r--      13.2 any cpu
 Domain-0                             0     1     1   -b-      10.8 any cpu
 # xm vcpu-pin Domain-0 0 64
 Error: Cannot pin vcpu: 0 to cpu: [64] - (22, ''Invalid
argument'')
 Usage: xm vcpu-pin <Domain> <VCPU|all> <CPUs|all>
 
 Set which CPUs a VCPU can use.
 # xm vcpu-list
 Name                                ID  VCPU   CPU State   Time(s) CPU Affinity
 Domain-0                             0     0     0   r--      13.5 any cpu
 Domain-0                             0     1     1   -b-      11.0 any cpu
 # xm vcpu-pin Domain-0 0 65
 Error: Cannot pin vcpu: 0 to cpu: [65] - (22, ''Invalid
argument'')
 Usage: xm vcpu-pin <Domain> <VCPU|all> <CPUs|all>
 
 Set which CPUs a VCPU can use.
 # xm vcpu-list
 Name                                ID  VCPU   CPU State   Time(s) CPU Affinity
 Domain-0                             0     0     0   r--      13.7 any cpu
 Domain-0                             0     1     1   -b-      11.3 any cpu


I could not understand why the problem occurs.  But I would like to 
solve the problem, so I send an ad hoc patch.  If you can solve the 
problem, I hope that the problem is solved by your patch. 

Signed-off-by: Masaki Kanno <kanno.masaki@jp.fujitsu.com>

Best regards,
 Kan



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

On 19/2/08 09:45, "Masaki Kanno" <kanno.masaki@jp.fujitsu.com>
wrote:
> I could not understand why the problem occurs.  But I would like to
> solve the problem, so I send an ad hoc patch.  If you can solve the
> problem, I hope that the problem is solved by your patch.
In C, shifting by more bits than the width of the value being shifted is
undefined. This explains the difference in behaviour between ia64 and x86.
Your patch seems sensible.

 -- Keir



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Tue, 19 Feb 2008 09:50:33 +0000, Keir Fraser wrote:
>On 19/2/08 09:45, "Masaki Kanno"
<kanno.masaki@jp.fujitsu.com> wrote:
>
>> I could not understand why the problem occurs.  But I would like to
>> solve the problem, so I send an ad hoc patch.  If you can solve the
>> problem, I hope that the problem is solved by your patch.
>
>In C, shifting by more bits than the width of the value being shifted is
>undefined. This explains the difference in behaviour between ia64 and x86.
>Your patch seems sensible.
Well, then could you apply my patch?

Best regards,
 Kan



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

It''s in the queue


On 20/2/08 02:30, "Masaki Kanno" <kanno.masaki@jp.fujitsu.com>
wrote:
> Tue, 19 Feb 2008 09:50:33 +0000, Keir Fraser wrote:
> 
>> On 19/2/08 09:45, "Masaki Kanno"
<kanno.masaki@jp.fujitsu.com> wrote:
>> 
>>> I could not understand why the problem occurs.  But I would like to
>>> solve the problem, so I send an ad hoc patch.  If you can solve the
>>> problem, I hope that the problem is solved by your patch.
>> 
>> In C, shifting by more bits than the width of the value being shifted
is
>> undefined. This explains the difference in behaviour between ia64 and
x86.
>> Your patch seems sensible.
> 
> Well, then could you apply my patch?
> 
> Best regards,
>  Kan
> 
> 


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel