Tim Waugh (tim@cyberelk.demon.co.uk)
Tue, 1 Dec 1998 23:45:25 +0000 (GMT)
On Wed, 2 Dec 1998, David Campbell wrote:
> >From a driver porting view, having blocking calls as default is nice
> since then you don't need any major modifications.
Yes. (Aside from porting UDI drivers in the future..) The mail I sent
out just now with the 1284.3 interface snippet in had functions that you
could just put NULL as the async parameter, and they would be blocking.
> For non-blocking calls (please give an example where would you use them?)
(Shrug) I couldn't think of any. (Other than a UDI driver, where
nothing's _allowed_ to block.)
> A simple macro would fail for SMP machines as the situation could
> arise that the test says it will not block, the other processor grabs
> the resource, the blocking function is called and blocks... (or am I
> being paranoid again)
I think you're being paranoid again.. ;-)
Tim.
*/
-- To unsubscribe, send mail to: linux-parport-request@torque.net --
-- with the single word "unsubscribe" in the body of the message. --
This archive was generated by hypermail 2.0b3 on Wed 30 Dec 1998 - 10:18:49 EST