On 2012年06月05日 22:18, Alan Stern wrote:
On Tue, 5 Jun 2012, Lan Tianyu wrote:
The patch does not implement anything at all for "auto". It should do
_something_.
Yeah, if we don't power off port without device when echo "auto"> portX/control,
this patch does nothing about "auto". Original plan is to deal with port without
devices firstly. Do you mean I should add the process of port with device in this
patch?
At least make a start. For example, have "auto" turn off the power if
no device is attached to the port.
In your reply for v2, hint that user space can power off port without
device, so I removed related code. Did I misunderstand?
control has three options. "auto", "on" and "off"
"auto" - if port without device, power off the port.
(This patch only cover ports without device)
Is this option really needed? Can't userspace check first to see
whether there is a device, and then write "off" if there isn't?
For ports with device, there is a proposal that if the device was not
in the suspend at that pointer, the power would remain on but it would
power off when it was suspended. If the the device was in the suspend,
"auto" means the device could be put into much lower state so the device
need to resume and suspend again.
That makes more sense. Without this, however, there is no need for
"auto".
If "auto" does nothing, there's no reason to put it in this patch at
all. It could be added in a later patch, along with its
implementation.
OK. I prefer to remove "auto" option in this patch. We can further discuss
the function of "auto".
Alan Stern
--
Best Regards
Tianyu Lan
linux kernel enabling team
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
