X Developers' Conference 2009 (September 28-30)

Matthew Garrett mjg59 at srcf.ucam.org
Fri Jun 26 21:58:19 PDT 2009


On Fri, Jun 26, 2009 at 09:42:02PM -0700, Alan Coopersmith wrote:
> Matthew Garrett wrote:
> > I don't think USB 3 really adds a lot here (wakeup from input devices 
> > has been part of the spec for a long time), but I'd be happy to talk 
> > about power management at the X-related platform level.
> 
> That would be good.   My comment was relating to the note in the Wikipedia
> entry that USB 3.0 adds "Link-, Device-, and Function-level power management"
> so I was imagining we could have per-device properties via Xi2 that could
> let you say "stop powering my tablet when it's idle for 5 minutes or there
> are no open apps selecting for events from it, but leave my mouse powered up
> so it can wake the system up".   Perhaps that's just silly given the power
> draw of most input devices, or I'm misunderstanding the description - that's
> why I was hoping someone who knew more could tell us if there's anything
> worthwhile to do there.

There's possibly some more fine-grained support, but even 1.1 allows for 
devices to be suspended until they generate a further wake event. I 
suspect that X's best approach to this will probably be to just keep 
devices open when needed and then close them to allow the kernel to 
suspend them, but it's worth talking about what level of functionality 
would need to be exposed.

-- 
Matthew Garrett | mjg59 at srcf.ucam.org


More information about the xorg-devel mailing list