Jean Tourrilhes <[EMAIL PROTECTED]> writes:
> Dag, I think this is not correct. First you can have sleep
> below 10ms, and second you didn't read carefully my sentence above.
Sure I did, but the docs I had for the ACT dongle said I should wait a few
ms, not 20 us. If it had said 20 us then I would certainly have used
udelay!! Until somebody sends me a doc which sais something else, the
driver will not be changed. I used _much_ resources the last time I got the
driver working, because it was so hard to find people using this
dongle. I'm not going to do anything until somebody that are using this
dongle 1) changes the driver 2) test that it works OK with a couple of
machines, and 3) sends me a patch the I can include ;-)
The suggested maximum timeout for udelay is btw. 1000 us. If I sent a patch
to Alan using udelays longer than 1000, he would store it in /dev/null (I
have tried!)
But really, if it's no harm sleeping longer than the few ms you want, then
I would choose kernel timers!! Modern machines can do a lot of work in a
few ms you know ;-) I really hate the xir2ps_cs driver that locks the
machine for a second when you insert the card. At least I'm not going to
produce such code.
-- 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
