Re: [patch 1/2] Touchscreen support for sharp sl-5500

[Russell King <rmk+lkml@arm.linux.org.uk>]

On Mon, Jul 25, 2005 at 11:47:25AM -0500, Dmitry Torokhov wrote:
> On 7/25/05, Russell King <rmk+lkml@arm.linux.org.uk> wrote:
> > If you look at _my_ version, you'll notice that it doesn't use the
> > class interface stuff.  A previous version of it did, and this seems
> > to be what the collie stuff is based upon.
> 
> I was only commenting on something that was posted on LKML for
> inclusion into input subtree that I am interested in. I don't track
> ARM development that closely. Where can we see your version, please?

See earlier in this thread, 24th July.

> > What I suggest is that the collie folk need to update their driver
> > to my version so that we don't have two different forks of the same
> > driver in existance.  Then we can start discussing whether things
> > should be using kthreads or not.
> 
> Do you have any reason why, generally speaking, threads should not be
> used? They seem to clean up code in drivers quite a bit.

It depends what the reasoning is behind them.  The touchscreen driver
is threaded because it wants to collect touschreen samples independently
of the availability of a user thread.  Moreover, obtaining ADC samples
needs a sleeping context since it may take a while to complete.

However, putting all UCB interrupts into a thread does not make sense
to me - if we allow UCB interrupts to sleep, it allows one UCB interrupt
to be processed at the exclusion of the others.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:  2.6 Serial core