This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Let's talk about INIT logic. The discussion will cover three topics: the rationale, the assumptions, and the implementation. The Rationale In much the same way that a flip-flop has an asynchronous input that takes the state of that flip-flop to some known value, we might like to have an additional input that takes the circuit to a known state independently of the application of the other inputs. Inasmuch as the circuit is already asynchronous, it does little good to refer to this input as being any more asynchronous than the others. But the premise here is that its assertion can take the circuit back to its initial state regardless of the present total state of the room, that is, independently of the state and the inputs. With this in mind, we want to talk about applying an input (INIT) that will return the circuit to the initial state for as long as it is asserted. INIT is distinct from x 1 and x 2 . We presume that if the initial state is stable for some input x 1 x 2 , then once the INIT signal is de-asserted, then the initial state will cause itself to happen for the "right" input x 1 x 2 . After all, that is the definition of stable. But what if that particular input isn't being applied when the INIT is de-asserted ? The Assumptions This is easier to set up via a story. Imagine that the display room has an emergency exit that can be used to bypass both the entrance and exit turnstiles where appropriate. It's a doorway that is used in, well, emergencies. On one particularly busy day, traffic through the display room is heavy. At one particular moment when two people are in the room (and the entrance turnstile is locked as a result), a great hue and cry is raised....
View Full Document
This note was uploaded on 01/17/2011 for the course ECE 3504 at Virginia Tech.