|   | 
      | 
  
  
      | 
      | 
  
 
     | 
    | 
  
  
     | 
    | 
  
  
    |   | 
      | 
  
  
    | 
         
xen-devel
Re: [Xen-devel] Xen power management default governor
 
On 11/01/10 23:36, Kaushik Barde wrote:
 
I believe, userspace governor gives better application control for instance,
controlling P-states through existing userspace frequency scaling daemons.
Just easier to set one's own power management policy.
-Kaushik
    
 
 Yes, but that's true, but that doesn't explain why userspace is the 
default, especially as there doesn't seem to be any userspace governor 
in the distribution and anyway, the userspace default was abandoned in 
the mainline Linux kernel because ondemand works rather better.
 On most machines, "userspace" with no running governor daemon is the 
same as "performance".  However, on some BIOS revisions of some 
hardware, "userspace" is the same as "powersave" and this basically 
makes it look as though the system is running rather slower than you 
would expect.
 So while userspace might be good for people who have clear cut ideas of 
what they think the policy for scaling should be, it's not good for 
everyone else and, in fact, it might lead people to believe that 
performance is actually rather poor.
 
-----Original Message-----
From: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
[mailto:xen-devel-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf Of John Haxby
Sent: Monday, January 11, 2010 4:51 AM
To: xen-devel@xxxxxxxxxxxxxxxxxxx
Subject: [Xen-devel] Xen power management default governor
Can anyone tell me why the default governor for power management is
"userspace"?   If there any reason why it shouldn't be "ondemand"?
And is there a userspace governor daemon that I've overlooked?
jch
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
    
 
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
 
 |   
 
 | 
    | 
  
  
    |   | 
    |