On Mon, Sep 13, 1999 at 09:57:26AM +0930, Brett Lymn wrote:
> According to Ignatios Souvatzis:
> >
> >No.
> >
> 
> Thanks for all the definite answers on this one :-)  I was reasonably
> certain it was wrong...
> 
> >
> >Somebody should fix it.
> >
> 
> I will fix it - it's biting me, I have the hardware to test it so I
> don't think it unreasonable that I get this fixed.  The only question
> I have is if tsleep cannot be called, what can I use?  The particular
> function is looping (possibly infinitely - aaargh) on a register
> status change and is using tsleep to throttle the poll interval.

Hm.

In the interupt? 

That is: you get an interupt for event X, and have to do some processing that
involves waiting for event Y, and can NOT leave interupt context (that is,
wake up the upper half of the driver, and let THAT one wait for the hardware?)

	-is

-- 
 * Progress (n.): The process through which Usenet has evolved from
   smart people in front of dumb terminals to dumb people in front of
   smart terminals.  -- obs@burnout.demon.co.uk (obscurity)