Re: [PATCH] libata: Add ALPM power state accounting to the AHCI driver

[Tejun Heo <tj@kernel.org>]

Arjan van de Ven wrote:
> On Sun, 15 Nov 2009 03:27:33 -0500
> Jeff Garzik <jgarzik@pobox.com> wrote:
> 
>> What little there is of driver-specific behavior can be handled from 
>> existing callbacks (->enable_pm) or by creating a driver-specific 
>> function that calls a generic function (eg. ahci_alpm_set_accounting 
>> could call ata_alpm_set_accounting, before twiddling AHCI's
>> PORT_IRQ_MASK).
> 
> the whole concept of needing that accounting flag is AHCI specific;
> if ever any of the other chip drivers goes to do ALPM, it'll be using
> explicit software control, which doesn't need this kind of enable flag;
> The only reason there is an enable flag is that the hw based accounting
> is resulting in extra interrupts that you don't want except when you
> want the accounting (read: powertop is running)

There's DIPM.

-- 
tejun