Pavel Machek <[EMAIL PROTECTED]> writes:
> Hi!
>
> > PS. Looks like Pavel is going to be happy ;-)
>
> Yes! Notice that you _could_ create kernel thread and emulate needed
> parts of irmanager in it...
Yes, but I have already made this _very_ complicated mechanism for doing
delays from interrupt context ;-) The change speed functions are now coded
as small state machines, but I have managed to abstract away most of the
complicated stuff into separate functions, so the change speed functions
will be pretty simple after all (thanks to some hints from Alan Cox).
Other changes:
o IrLAP does not know anything about irda_device. Irda_device will mostly
consist of utility functions
o No change speed function for IrLAP to call. Instead each frame is tagged
with the speed it should be sent out at. Previously frames could be sent
out in the wrong speed because of the device queue, and the fact that
changing speed takes some time to do.
I'm testing this stuff right now, so I'm expecting a few crashes the next
couple of hours ;-)
-- Dag
--
/ Dag Brattli | The Linux-IrDA Project /
// University of Tromsoe, Norway | Infrared communication for Linux //
/// http://www.cs.uit.no/~dagb | http://www.cs.uit.no/linux-irda/ ///
_______________________________________________
Linux-IrDA mailing list - [EMAIL PROTECTED]
http://www4.pasta.cs.UiT.No/mailman/listinfo/linux-irda
