Per powertop, it does, though kstat cpu_info has only shown it running at the 
higher speed.  Watching powertop, it does seem to switch between pstates often.

Sent from my iPhone

On Mar 29, 2009, at 9:28 AM, John Martin <John.M.Martin at Sun.COM> wrote:

Jason Rhodes wrote:
any other OSes. Maybe Jason
can have a try some linux liveCD like Ubuntu to see
if P-state works. I just want to
make solaris pm competitive to the other OSes. I will
not persist my opinion if there
is a better reason to disable C-state PDC bits by
default.
   

For what it's worth, I fired up an Ubuntu 8.10 livecd that I had, and it 
supported the two CPU speeds on my E7400 just fine.  I could even manually 
select either speed manually with the Gnome power applet.  Cute.

I also upgraded to snv_110, and it appears to work as promised.   kstat 
cpu_info lists two supported_frequencies_Hz.  What's funny, however, is that it 
made no difference in power usage.  I have the computer hooked up to a 
Kill-a-watt, and power consumption at idle is about 110W, both in build 105 and 
110, and in Ubuntu as well.

                       OpenSolaris PowerTOP version 1.1

Cn                      Avg     residency       P-states (frequencies)
C0 (cpu running)                (15.4%)         1596 Mhz        0.0%
C1                      1.4ms   (84.6%)         2793 Mhz        100.0%

Wakeups-from-idle per second: 586.4     interval: 5.0s
no ACPI power usage estimate available

Top causes for wakeups:
25.4% (148.9)                  sched :  <cross calls>
25.2% (147.5)               <kernel> :  genunix`cv_wakeup
17.1% (100.0)               <kernel> :  genunix`realitexpire
17.1% (100.0)               <kernel> :  genunix`clock
16.3% ( 95.6)           VBoxHeadless :  <cross calls>
3.9% ( 22.8)               <kernel> :  uhci`uhci_handle_root_hub_status_change
1.7% ( 10.0)               <kernel> :  ata`ghd_timeout
1.3% (  7.8)               <kernel> :  ehci`ehci_handle_root_hub_status_change
1.0% (  6.0)               <kernel> :  uhci`uhci_cmd_timeout_hdlr
0.8% (  4.8)            <interrupt> :  e1000g#2
 
Does your system ever move to the lower P-state (1596MHz)?
On my Tecra M10 and hand built i7 920, b110 + daily.0327 bfu
it quickly moves the the lowest P-state.

We've had long discussions on other aliases on whether
Penryn/Wolfdale P-state changes amount to any real reduction
in system power.



      

Reply via email to