[Tinyos-host-mote-wg] RE: [Tinyos-2.0wg] Interrupt interface

[Phil Buonadonna]

Maybe nesc could add pragmas to do compile time checking of constant
function arguments!  

ha ha ha ha ha

I'm soooooooo funny.

pb
 

> -----Original Message-----
> From: Philip Levis [mailto:pal at cs.stanford.edu] 
> Sent: Thursday, October 20, 2005 2:01 PM
> To: Phil Buonadonna
> Cc: Tinyos-2.0wg at Mail.Millennium.Berkeley.EDU
> Subject: RE: [Tinyos-2.0wg] Interrupt interface
> 
> On Thu, 2005-10-20 at 13:38, Phil Buonadonna wrote:
> 
> > Not too much to understand, I think. Except that there 
> are/will be devices
> > which can be set to interrupt on both edges, not just one.  
> If we stick to
> > the present interface, we consciously block this 
> capability. I think that's
> > just fine, but we should be aware of it.
> 
> This is a tough one.
> 
> On one hand, having two interfaces means, well, having two interfaces.
> On the other, the Interrupt interface is an HIL interface (e.g., the
> MSP430Interrupt interface has more functionality). Allowing additional
> parameters that could be invalid (e.g., making it take an
> interrupt_edge_t of HIGH_TO_LOW LOW_TO_HIGH or CHANGE) goes against
> static binding, in that the error is at runtime.
> 
> Phil
> 
> 
> 
> 

_______________________________________________
Tinyos-2.0wg mailing list
Tinyos-2.0wg at Mail.Millennium.Berkeley.EDU
https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-2.0wg