On 08/25/2011 06:55 PM, Sarah Sharp wrote:
On Thu, Aug 25, 2011 at 06:29:19PM +0200, Sebastian Andrzej Siewior wrote:
Would it make sense to make this test only when LPM is requested which
is only the case when the user requests it? One could add a message in
powertop which would warn the user that this 2.1 device could fail to
suspend.
Currently, there is no way to enable or disable USB 2.1 LPM from userspace.
Andiry's patches are enabling it automatically in the xHCI driver after
the device passes the LPM test, when it is enumerated.
I like to have such things enabled by default but it seems that not all
devices on all cores handle it well. We have one device which fails to
at this with core A but works on core B and we don't know _why_.
We don't have a knob to enable/disable usb2.1 LPM but this does not
mean that we can't add one if it makes sense.
I'm not totally against the idea that the kernel is trying to be clever
but I'm curios why it fails on some cores but then not for all devices.
And if we do know the root cause we could be better at testing and
avoid it where we can forecast the outcome.
Sarah Sharp
Sebastian
--
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
