On Sunday, September 23, 2012, Lan Tianyu wrote: > 于 2012/9/23 3:54, Rafael J. Wysocki 写道: > > On Friday, September 21, 2012, Peter Stuge wrote: > >> Sarah Sharp wrote: > >>>>> If userspace really wants the port off (e.g. to disconnect and > >>>>> reconnect a misbehaving device), then it can set the sysfs file > >>>>> to off. > >>>> > >>>> And unless all ganged ports are also off it will fail. Userspace will > >>>> want to know about that, and why, along with a reference that can be > >>>> matched against the list of ganged sets of ports. > >>> > >>> Ah, right. Yes, we should probably add some dmesg info lines about > >>> why we can't turn off a port right now. > >> > >> dmesg isn't very programmable. Maybe ioctl after all? Trust me that I > >> don't have a preference, it's just that I don't know how sysfs could > >> report back the relevant failure info in a race-free way. > > > > Well, we actually need to handle power domains appropriately. > > > > Unfortunately, ACPI doesn't support the power domain concept directly and > > tries to kind of hide them behind power resources, but let's face it, they > > are what they are. > > > > Some work in that direction has been done in the ARM space, where we have > > much more direct access to hardware, and I suppose it may be extended to > > things like "ganged sets of ports" (which actually are power domains). > > > Do you mean arm has used generic pm domain but ACPI has not yet? Yes. > We should associate ACPI power resource with generic pm domain. Yes, we probably should. > Has generic pm domain been exposed to usr space? (e.g via sysfs) No, it hasn't. Mostly because it is not clear how the interface should look like. Thanks, Rafael -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
