Re: About Power Manager

Typically OEMIdle cuts the cpu clock only, leaving the peripheral clocks
unaffected. Depends on the capability of the cpu. But you're correct, you
can't be adjusting the clock in a way that negatively affects other devices.
Power states that affect other devices would be managed by the higher level
PM driver, which should have knowledge of whether that device can be safely
disabled or not.

--
Dean Ramsier - eMVP
BSQUARE Corporation


"Stanley" <Stanley@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote in message
news:4B7AB72A-D6AB-4B04-9329-C93D06C8744E@xxxxxxxxxxxxxxxx
Excuse me, there is another question when I implement the OEMIdle and the
PM
functions device driver.
If I move the CPU into idle mode, I will reduce the CPU clock for power s
aving. Is it OK for reducing CPU clock in the OEMIdle?
However, the device sharing with the same clock source with CPU will also
be
affected.
Ex: USB must work in 48Mhz. The CPU clock can't lower than it.
I am afaid that the CPU can't get smaller power in this condition.
Please give me some advices.

Best regards,
"Stanley" wrote:

Thanks you all,
I think I've understood the difference.


"Paul Monson" wrote:

PM manages the power state of peripheral devices. Many peripheral
devices are relatively slow to start and stop. As a result the power
state is adjusted less frequently(when compared to OEM idle), usually
after several seconds or minutes have passed.

OEM idle usually only affects the CPU power states and happens as
frequently as possible.

Paul Monson
Intrinsyc


Dean Ramsier wrote:
The same as it was before. The kernel constantly transitions in and
out of
Idle (meaning OEMIdle) whenever a thread blocks and nothing else is
scheduled to run. This happens at the millisecond level and is
transparent
to the Power Manager, which works with system wide events on a much
larger
time scale.

Maybe the key to understanding this is to recognize that the kernel
Idle
refers to idling the processor at the thread level (microscopic) and
the PM
refers to system level events (macroscopic). The two execute in
parallel,
and the PM has no idea if the kernel has placed the processor in a
low power
mode via OEMIdle.



.

Relevant Pages

  • Re: About Power Manager
    ... me what is the condition that the system will enter the IDLE mode? ... Depends on the capability of the cpu. ... If I move the CPU into idle mode, I will reduce the CPU clock for power ... Is it OK for reducing CPU clock in the OEMIdle? ...
    (microsoft.public.windowsce.platbuilder)
  • Re: ASUS Crosshair help
    ... start cranking the CPU clock from 200 to something higher. ... To experiment in Windows, you are in luck. ... The upper slider should control the CPU clock. ... Orthos will run two copies of the Prime95 Torture Test. ...
    (alt.comp.periphs.mainboard.asus)
  • Re: About Power Manager
    ... Excuse me, When I test the OEMIdle, I put a debug message in it. ... me what is the condition that the system will enter the IDLE mode? ... Depends on the capability of the cpu. ... Power states that affect other devices would be managed by the higher level ...
    (microsoft.public.windowsce.platbuilder)
  • Re: Stolen and degraded time and schedulers
    ... I presume you mean monotonic real time.) ... but I don't think that's terribly useful. ... work_time is probably most naturally measured in cpu clock cycles rather ... In the same way that clock modulation gates the cpu clock, ...
    (Linux-Kernel)
  • Re: timing a method
    ... That's stated to be a count of CPU clock cycles, ... >>be independent of CPU clock rate and drift, ... and Dr John, this is all said and good. ... So no reasonably accurate is going to be reasonably ...
    (alt.comp.lang.borland-delphi)