![]() | |||||||||||
Messages in this thread |
On 8/7/06, Pavel Machek <pavel@suse.cz> wrote: > > When the current hdaps driver is queried for recent keyboard/mouse activity > > (which is provided by the hardware for use in disk parking daemons), it > > simply returns the last readout. However, every hardware query resets the > > activity flag in the hardware, and this is triggered by (almost) any > > hdaps sysfs attribute read, so the flag could be reset before it is > > observed and is thus nearly useless. > > > > This patch makes the activities flags persist for 0.1sec, by remembering > > when was the last time we saw them set. This gives apps like the hdaps > > daemon enough time to poll the flag. > And should not hdapsd get it from input interface? At this point the disk may parked and its queue frozen, and may depend on the keyboard/mouse activity to be released. So hdapsd mustn't do anything that involves potential swapping or userspace with unlocked memory, and the input infrastructure is too large and flexible for such a guarantee. I guess this is why IBM provided this low-level hook at the embedded controller level. Shem - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ | ||||||||||
| Last update: 2006-08-07 18:21 [from the cache] ©2003-2008 | |||||||||||