19.md


1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67

[\<- State machine concepts](18.md)

---

# Multiple control outputs

## Background on register swap problem

### Datapath and control

- We've learned about muxes and registers as datapath elements
- Mux selects and load enables are control signals that determine how datapath is used
- If these control signals need to be sequenced across multiple cycles, a state machine is the typical way to do this

### Example 2: Register Swap

- Swap the contents of R1 and R2
- Can't be done in a single cycle
- Use R3 as temporary holding register

![diagram](19.1.png)

### Another view of the datapath

- To make the control signals more clear

![diagram](19.2.png)

---

## State diagram

### Sequence

- Since there's a single bus, can only copy/move one value at a time
	- Copy the value in R2 into R3
		- Select R2 to drive the bus and assert R3's enable
	- Copy the value in R1 into R2
		- Select R1 to drive the bus and assert R2's enable
	- Copy the value in R3 into R1
		- Select R3 to drive the bus and assert R1's enable
- Each of these steps is a different state in a state machine

![diagram](19.3.png)

---

## State table and implementation

### State Table & Assignment

![diagram](19.4.png)

### Visually deriving equations

- Y2 = (stateB) + (stateC)
	- `= !y2*y2 + y2*!y1 = y2^y1`
- Y1 = w\*(stateA) + (stateC)
	- `= w*!y2*!y1 + y2*!y1`

### K-maps for Next State

![diagram](19.5.png)

### Implementation

![diagram](19.6.png)