On Tue, 31 Aug 2010, Oliver Neukum wrote: > Am Dienstag, 31. August 2010, 16:43:38 schrieb Alan Stern: > > On Tue, 31 Aug 2010, Oliver Neukum wrote: > > > > > Am Dienstag, 31. August 2010, 16:13:26 schrieb Alan Stern: > > > > Worse. What happens when the system sleeps and the device does not > > > loose power? > > > > You mean the device remains at full power during system sleep? How > > could that happen? > > Not full power, any power at all. There are three cases > > a) loss of power -> reset_resume() > b) device not buggy -> resume() - no further trouble > c) device buggy -> spurious change event > > We run into (c) if eg. the reader shares power with a wakeup source. Ah yes. In fact, that must be a problem even now. At least adding autosuspend support won't make it any worse than it already is. > > For those buggy readers, it would be better to have something > > which could be implemented as a sysfs attribute in usb-storage. > > Why usb-storage? Any drive for real rotational media has the same > requirement. > Waiting for minutes for a suspend without medium is ridiculous, > as is spinning down a disk after only two seconds. Suggestions for a sysfs interface? Should the sd driver add its own "suspend_delay_no_medium" attribute? Alan Stern -- 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
