Research on device power management

Research on device power management

[Holger Macht]

[-- Attachment #1: Type: text/plain, Size: 264 bytes --]

Hi,

Due to the recent discussions about device power management (runtime PM),
one could find the following research helpful. It contains some device
tests/measurements for four different laptops.

    http://powersave.sourceforge.net/pm-report/

Regards,
	Holger

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

RE: Research on device power management

[Gross, Mark]

[-- Attachment #1: Type: text/plain, Size: 927 bytes --]

Very cool.

Can I get some data on your amp meter?  I'm thinking about building a
home brew thing with a current sense resister and a PIC 877.

Also, I would be interested in USB mouse and wireless mouse effects on
the system.  I know low speed usb devices result in a PCI interrupt
every msec, so it would be a nice data point to have (battery time costs
of using a USB mouse)

--mgross

>-----Original Message-----
>From: linux-pm-bounces@lists.osdl.org
[mailto:linux-pm-bounces@lists.osdl.org]
>On Behalf Of Holger Macht
>Sent: Friday, October 28, 2005 5:05 AM
>To: linux-pm@lists.osdl.org
>Subject: [linux-pm] Research on device power management
>
>Hi,
>
>Due to the recent discussions about device power management (runtime
PM),
>one could find the following research helpful. It contains some device
>tests/measurements for four different laptops.
>
>    http://powersave.sourceforge.net/pm-report/
>
>Regards,
>	Holger


[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

RE: Research on device power management

[Alan Stern]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 433 bytes --]

On Fri, 28 Oct 2005, Gross, Mark wrote:

> Also, I would be interested in USB mouse and wireless mouse effects on
> the system.  I know low speed usb devices result in a PCI interrupt
> every msec,

No they don't.  Interrupts occur only when data is transferred.

On the other hand, connecting USB devices (of any speed) causes ongoing
DMA activity, which can prevent CPUs from entering their most power-saving 
states.

Alan Stern


[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

RE: Research on device power management

[Gross, Mark]

[-- Attachment #1: Type: text/plain, Size: 866 bytes --]

Your right.  I miss spoke.  The USB bus is polled for low / speed
devices such as mice and keyboards at 1000Hz, but this is done by the
host controller or root hub.

--mgross

>-----Original Message-----
>From: Alan Stern [mailto:stern@rowland.harvard.edu]
>Sent: Friday, October 28, 2005 9:36 AM
>To: Gross, Mark
>Cc: Holger Macht; linux-pm@lists.osdl.org
>Subject: RE: [linux-pm] Research on device power management
>
>On Fri, 28 Oct 2005, Gross, Mark wrote:
>
>> Also, I would be interested in USB mouse and wireless mouse effects
on
>> the system.  I know low speed usb devices result in a PCI interrupt
>> every msec,
>
>No they don't.  Interrupts occur only when data is transferred.
>
>On the other hand, connecting USB devices (of any speed) causes ongoing
>DMA activity, which can prevent CPUs from entering their most
power-saving
>states.
>
>Alan Stern



[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

Re: Research on device power management

[Holger Macht]

[-- Attachment #1: Type: text/plain, Size: 634 bytes --]

On Fri 28. Oct 08:59:44, Gross, Mark wrote:
> Very cool.
> 
> Can I get some data on your amp meter?  I'm thinking about building a
> home brew thing with a current sense resister and a PIC 877.

It is a "Voltcraft M-4660A" with rs232 interface. Anything more specific I
can tell you?

> 
> Also, I would be interested in USB mouse and wireless mouse effects on
> the system.  I know low speed usb devices result in a PCI interrupt
> every msec, so it would be a nice data point to have (battery time costs
> of using a USB mouse)

Added --> http://powersave.sourceforge.net/pm-report/#nx5000usbmouse

> 
> --mgross

Regards,
	Holger

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]