On Wed, 2012-09-19 at 13:27 +0100, James Bottomley wrote: > I think we could do this with a couple of flags sitting inside struct > device itself: one for pm state and capabilities defined at a generic > level and one for device specific pm state. The latter would be for > things like the door lock information which is very specific to CDs > (although not specific to SCSI CDs). Alternatively, even if we can't > use these capabilities at the generic pm level, we still need an > internal state set of flags because power state stuff traverses the > stack and struct device is the only universal object in that stack. FWIW the SATA Device Sleep support, which is currently only used (in a patch which is pending) automatically as an extension of the AHCI link power management, can also be triggered manually by a GPIO line; I've got a platform where it's done with an ACPI call. I suspect we'll see that kind of manual setup on a few embedded platforms too. If we're working on hooking this up through the device layers, perhaps that's worth taking into consideration? -- David Woodhouse Open Source Technology Centre David.Woodhouse@xxxxxxxxx Intel Corporation
Attachment:
smime.p7s
Description: S/MIME cryptographic signature
