3Program Control
3 – 19
CLKOUT
Address for
Instruction Fetch
Interrupt
Instruction
Executing
n–2 n–1 n NOP
1st instr of
serv routine
n–1 n n+1
interrupt
vector i
i+1
Figure 3.2 Interrupt Latency (Timer,
IRQIRQ
IRQIRQ
IRQ
xx
xx
x
, SPORT, HIP, & Analog Interrupts)
(Note that this latency for the timer interrupt only applies for the ADSP-2171,
ADSP-2181, and ADSP-21msp58/59 processors. See the next section for a
description of timer interrupt latency on the ADSP-2101, ADSP-2105,
ADSP-2115, ADSP-2111.)
For a pending interrupt that is masked, the latency from execution of the
instruction that unmasks the interrupt (in IMASK) to the first instruction of
the service routine is one cycle. This one-cycle latency is similar to that shown
in Figure 3.3 for the timer interrupt of the ADSP-2101/2105/2111/2115, with
the “n” instruction executing being the instruction that writes to IMASK (to
unmask the interrupt).
3.4.3.1 Timer Interrupt Latency on ADSP-2101, ADSP-2105, ADSP-2115, ADSP-2111
For the timer interrupt on these processors, the latency from when the
interrupt occurs to when the first instruction of the service routine is executed
is only one cycle. This is shown in Figure 3.3. The single cycle of latency is
needed to fetch the instruction stored at the interrupt vector location.
CLKIN
Address for
Instruction Fetch
Instruction
Executing
n+1
interrupt
vector i
i+1
n NOP
1st instr of
serv routine
Timer
Value
tcount=1 tcount=0
Figure 3.3 Timer Interrupt Latency for ADSP-2101, ADSP-2105, ADSP-2115, ADSP-2111