Re: [patch] kernel sysfs events layer

[Jan Dittmer]

Tim Hockin wrote:
> On Wed, Sep 15, 2004 at 10:19:50AM +0200, Karol Kozimor wrote:
> 
>>On Wednesday 15 of September 2004 08:45, Jan Dittmer wrote:
>>
>>>What's wrong about fixing acpi to have something like
>>>/sys/devices/acpi/buttons/power/, that spits out the event?
>>>Just curious...
>>
>>Well, the fact that you'd have to somehow:
>>1) pass the list of all the drivers that register notify handlers to the 
>>userspace
>>2) make userspace daemons hold ~10 sysfs nodes open
> 
> 
> I *think* the suggestion was to add sysfs nodes for ACPI objects so that
> the assosicated kobject could be the "source" argument to the uevent API.
> Not that each kobject would have an event stream of it's own :)

Yepp.

> I'm not deeply into ACPI, so I'm not sure if it is possible to enumerate
> all event sources, or if GPEs would come from nowhere...

Well if you do a find /sys | grep acpi, there are basically all nodes
already there - just no values in there :-(
I think Andrew's suggestions (in this same thread) make more sense, ie.
deliver button events via the input layer...

Jan
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/