[lkml]   [2003]   [Aug]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PM] Patrick: which part of "maintainer" and "peer review" needs explaining to you?
On Mon, Aug 25, 2003 at 06:14:15PM +0100, Russell King wrote:
> On Mon, Aug 25, 2003 at 09:57:20AM -0700, Matt Porter wrote:
> > Alternatively, you could leave the platform model as is (it's only
> > for really dumb devices).
> Thing is that we use the platform model for off chip devices as well.
> On ARM, its gets used for any device which the platform code knows
> where it is located. ie, a platform device.
> > On PPC, we have an OCP (on chip peripheral)
> > model that is mostly integrated into the device model now. OCP is
> > just another bus/driver type and so PM works in the normal fashion.
> Ah, but OCP can't be used to describe a platform dependent SMC91x
> network interface that some random designer decided to drop into
> their design - it isn't part of the SoC.

There's nothing inherent in OCP (except the name) that prevents this.
In fact, that's one of the directions we wanted to go with it because
of PPC4xx external bus controller devices that are board specific.

> > There's a driver API around it as well so we can cleanly share drivers
> > across various SoC implementations with different base address,
> > IRQ mappings, etc. It might be more useful to extende this across
> > the architectures that need it.
> Note that we've already done some public work on providing flexible
> platform device support to satisfy the needs of platform people -
> by adding the variable number of resources to the platform device.

Yes, I recall asking for more interrupt resources when I thought it
might be useful. However, the PPC4xx stuff is requiring a move
to a finer grained specification of internal bus types to properly
implement PM. It's necessary to know if a device is a child of
the PLB, OPB, EBC, etc. to know when to power down various internal
bus drivers as well. I don't see that happening with platform
devices even with the addition of PM ops. Perhaps this isn't
applicable to more than PPC, though.

> Also note that most of the x86 ISA PCMCIA devices _are_ platform
> devices today. As of this new power management model, they're
> broken due to the fact that they no longer receive power management
> events.

I see.

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
More majordomo info at
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:48    [W:0.069 / U:1.952 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site