[patch 0/4] SATA power savings patches (ALPM)

[patch 0/4] SATA power savings patches (ALPM)

[Kristen Carlson Accardi]

Hi Jeff,
Here's the most recent patches for ALPM.  These are also located at:
http://www.kernel.org/pub/linux/kernel/people/kristen/patches/SATA/alpm

These patches implement Aggressive Link Power management for AHCI 
controllers.  This feature is described in detail in the AHCI 1.x spec.
It provides power savings of anywhere from .5-1.5 Watts depending on
the system when it is enabled.  (a graph demonstrating power savings
of this patch along with other power patches is located at 
http://www.linuxpowertop.org/results.php for your enjoyment).

This patch is changed from the previous version in that it incorporates all
the feedback I've gotten on the patches to date.  It changes the default
of the ahci driver to be max_performance rather than min_power, so userspace
will need to explicitly enable alpm via the sysfs interface when it deems
it an appropriate time to do this.  In addition, the patch was changed to
allow either HIPM or DIPM as a prerequisite for enabling ALPM, rather than 
just using HIPM.

Thanks,
Kristen
-- 

Re: [patch 0/4] SATA power savings patches (ALPM)

[Bill Davidsen]

Kristen Carlson Accardi wrote:
> Hi Jeff,
> Here's the most recent patches for ALPM.  These are also located at:
> http://www.kernel.org/pub/linux/kernel/people/kristen/patches/SATA/alpm
> 
> These patches implement Aggressive Link Power management for AHCI 
> controllers.  This feature is described in detail in the AHCI 1.x spec.
> It provides power savings of anywhere from .5-1.5 Watts depending on
> the system when it is enabled.  (a graph demonstrating power savings
> of this patch along with other power patches is located at 
> http://www.linuxpowertop.org/results.php for your enjoyment).
> 
> This patch is changed from the previous version in that it incorporates all
> the feedback I've gotten on the patches to date.  It changes the default
> of the ahci driver to be max_performance rather than min_power, so userspace
> will need to explicitly enable alpm via the sysfs interface when it deems
> it an appropriate time to do this.  In addition, the patch was changed to
> allow either HIPM or DIPM as a prerequisite for enabling ALPM, rather than 
> just using HIPM.
> 
I suspect these changes would be useful for servers as well, extending 
run time when a failover to UPS is detected.

-- 
Bill Davidsen <davidsen@tmr.com>
   "We have more to fear from the bungling of the incompetent than from
the machinations of the wicked."  - from Slashdot