17.05.2015 Views

TI486 Microprocessor - Al Kossow's Bitsavers

TI486 Microprocessor - Al Kossow's Bitsavers

TI486 Microprocessor - Al Kossow's Bitsavers

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Functional Timing<br />

Figure 3-11 illustrates transitioning to pipelined addressing during a burst of<br />

bus cycles. Cycle 2 makes the transition to pipelined addressing. Comparing<br />

Cycle 2 to Cycle 1 of Figure 3-10 illustrates that a transition cycle is the same<br />

whenever it occurs consisting of at least T1, T2 (NA is asserted at that time),<br />

and T2P (provided the <strong>TI486</strong>SLC/E microprocessor has an internal bus<br />

request already pending). T2P states are repeated if wait states are added to<br />

the cycle. Cycles 2, 3, and 4 in Figure 3-11 show that once address pipe lining<br />

is achieved it can be maintained with two-state bus cycles consisting only of<br />

T1P and T2P.<br />

Once a pipelined bus cycle is in progress, pipelined timing is maintained for<br />

the next cycle by asserting NA and detecting that the <strong>TI486</strong>SLC/E<br />

microprocessor enters T2P during the current bus cycle. The current bus cycle<br />

must end in state T2P for pipelining to be maintained in the next cycle. T2P is<br />

identified by the assertion of ADS. Figure 3-10 and Figure 3-11 each show<br />

pipelining ending after Cycle 4. This occurred because the <strong>TI486</strong>SLC/E CPU<br />

did not have an internal bus request prior to the acknowledgment of Cycle 4.<br />

Figure 3~ 11. Transitioning to Pipelined Address During Burst of Bus Cycles<br />

CLK2<br />

Idle<br />

I Cycle 1<br />

1 Non-Pipelined<br />

+<br />

(Write)<br />

Ti 1<br />

A23-A1,~1 I<br />

BHE, BLs<br />

M/IO,D/C<br />

T1<br />

I<br />

1<br />

.1 ..<br />

1<br />

I<br />

Cycle 2<br />

Non-Pipelined<br />

(Read)<br />

I<br />

I<br />

.~<br />

T1 T2P I<br />

wlR i~~&"--r-l------I-'<br />

Cycle 3 1<br />

Pipelined I<br />

(Write) .1 ..<br />

1 1<br />

T1P I T2P I T1P<br />

Cycle 4<br />

Pipelined<br />

(Read)<br />

Note:<br />

, 1.1<br />

•<br />

I<br />

«<br />

I<br />

:<br />

1<br />

Valid<br />

I<br />

1 1 1<br />

~i ~<br />

V~lid3 ~<br />

V~lid4 ~<br />

«<br />

2;<br />

Out 1 }-T--I- 1 1-$: In 2 < Out 3 I }-I- 1 In4--1<br />

:<br />

- 1<br />

1 I I I 1<br />

I I I I I I I I<br />

Following any idle bus state (Ti), addresses are non-pipelined bus cycles, NA is sampled only during wait states.<br />

Therefore, to begin address pipelining during a group of non-pipelined bus cycles requires a non-pipelined cycle with at<br />

least one wait state (Cycle 2 above).<br />

The complete bus state transition diagram, including operation with pipelined<br />

address is given in Figure 3-12. This is a superset of the diagram for<br />

non-pipelined address. The three additional bus states for pipelined address<br />

are shaded.<br />

3-28<br />

<strong>TI486</strong>SLCIE Bus Interface

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!