EECS 373 Design of Microprocessor-Based Systems Mark Brehob University of Michigan

1

EECS 373Design of Microprocessor-Based Systems

Mark BrehobUniversity of Michigan

Lecture 7: Finish interrupts, group talks, start on serial busesFeb 4th 2014

Announcements

• Recall we have a Monday night group formation meeting from 6:30-8:00pm – Everyone needs to be there.– In 1500 EECS.

• Exam will be on Tuesday 2/25 in class.

• HW3 will be posted later this week.

• Thursday’s lecture will be ~30 minutes of class material and 50 minutes of project discussion.

2

High-level review of interrupts

• Why do we need them? Why are the alternatives unacceptable? – Convince me!

• What sources of interrupts are there?– Hardware and software!

• What makes them difficult to deal with?– Interrupt controllers are complex: there is a lot to

do!• Enable/disable, prioritize, allow premption

(nested interrupts), etc.– Software issues are non-trivial

• Can’t trash work of task you interrupted• Need to be able to restore state• Shared data issues are a real pain

3

4

SmartFusion interrupt sources

5

And the interrupt vectors (in startup_a2fxxxm3.s found in CMSIS, startup_gcc)g_pfnVectors:

.word _estack .word Reset_Handler .word NMI_Handler .word HardFault_Handler .word MemManage_Handler .word BusFault_Handler .word UsageFault_Handler .word 0 .word 0 .word 0 .word 0 .word SVC_Handler .word DebugMon_Handler .word 0 .word PendSV_Handler .word SysTick_Handler .word WdogWakeup_IRQHandler .word BrownOut_1_5V_IRQHandler .word BrownOut_3_3V_IRQHandler.............. (they continue)

6

Interrupt handlers

7

Pending interrupts

8

The normal case. Once Interrupt request is seen, processor puts it in “pending” state even if hardware drops the request. IPS is cleared by the hardware once we jump to the ISR.

This figure and those following are from The Definitive Guide to the ARM Cortex-M3, Section 7.4

9

In this case, the processor never took the interrupt because we cleared the IPS by hand (via a memory-mapped I/O register)

10

11

Answer

12

Interrupt pulses before entering ISR

13

Answer

14

15

16

17

18

19

20

21

22

Masking

23

24

The xPSR register layout

25

ARM interrupt summary

1. We’ve got a bunch of memory-mapped registers that control things (NVIC)– Enable/disable individual interrupts– Set/clear pending– Interrupt priority and preemption

2. We’ve got to understand how the hardware interrupt lines interact with the NVIC

3. And how we figure out where to set the PC to point to for a given interrupt source.

26

1. NVIC registers (example)

27

1. More registers (example)

28

1. Yet another part of the NVIC registers!

29

2. How external lines interact with the NVIC

30

The normal case. Once Interrupt request is seen, processor puts it in “pending” state even if hardware drops the request. IPS is cleared by the hardware once we jump to the ISR.

This figure and those following are from The Definitive Guide to the ARM Cortex-M3, Section 7.4

3. How the hardware figures out what to set the PC to

g_pfnVectors: .word _estack .word Reset_Handler .word NMI_Handler .word HardFault_Handler .word MemManage_Handler .word BusFault_Handler .word UsageFault_Handler .word 0 .word 0 .word 0 .word 0 .word SVC_Handler .word DebugMon_Handler .word 0 .word PendSV_Handler .word SysTick_Handler .word WdogWakeup_IRQHandler .word BrownOut_1_5V_IRQHandler .word BrownOut_3_3V_IRQHandler.............. (they continue)

31

So let’s say a GPIO pin goes high - When will we get an interrupt? - What happens if the interrupt is allowed to proceed?

32

What happens when we return from an ISR?

33

Other stuff: The xPSR register layout

34

Example of Complexity: The Reset Interrupt

1) No power2) System is held in RESET as long as VCC15 < 0.8V

a) In reset: registers forced to defaultb) RC-Osc begins to oscillatec) MSS_CCC drives RC-Osc/4 into FCLKd) PORESET_N is held low

3) Once VCC15GOOD, PORESET_N goes higha) MSS reads from eNVM address 0x0 and 0x4

35

Group talks in EECS 373

36

Talk about talks

• Groups of 2-3 folks– Not your lab partner

• 12 minutes for the talk, ~3 minutes for questions

• Four parts– Meet with me about 2-3 weeks ahead of time to

discuss topic idea.– Give first practice talk about 1-2 weeks before

scheduled date (10%)– Second practice talk a few days before scheduled

date (10%)– Give talk in class (80%)

37

Topic talk (continued)

• Each talk must include– An explanation of how the topic relates to

embedded systems– An understanding of high-level issues including

tradeoffs• Need to produce at least one original graph

explaining tradeoffs.– Some detailed explanation of a relevant part of

the topic– Where others can go to learn more information

• We’ll take 10 minutes at the end of class to form groups of 2-3.– By 9pm groups should fill out form at http://

tinyurl.com/373TTW14 If you don’t have a group by that time, send me an e-mail and I’ll form them.

38

Start Serial Buses

Some generic information

(UART, I2C, and SPI next Tuesday)

39

Fun with buses

• A multidrop bus (MDB) is a computer bus in which all components are connected to the same set of electrical wires. (from Wikipedia)– In the general case, a bus may have more than

one device capable of driving it.• That is, it may be a “multi-master” bas as

discussed earlier.

How can we handle multiple (potential) bus drivers? (1/3)

• Tri-state devices, just haveone device drive at a time. Everyone can read though– Pros:

• Very common, fairly fast, pin-efficient.

– Cons:• Tri-state devices can be slow.

– Especially drive-to-tristate?• Need to be sure two folks not driving at the same

time– Let out the magic smoke.

– Most common solution (at least historically)• Ethernet, PCI, etc.

How can we handle multiple (potential) bus drivers? (2/3)

• MUX– Just have each device generate its data, and

have a MUX select.• That’s a LOT of pins.

– Consider a 32-bit bus with 6 potential drivers.

» Draw the figure.» How many pins needed for the MUX?

– Not generally realistic for an “on-PCB” design as we’ll need an extra device (or a lot of pins on one device)• But reasonable on-chip

– In fact AHB, APB do this.

How can we handle multiple (potential) bus drivers? (3/3)

• “pull-up” aka “open collector” aka “wired AND”– Wire is pulled high by

a resistor– If any device pulls the

wire low, it goes low.• Pros:

– If two devices both drive the bus, it still works!

• Cons:– Rise-time is very slow. – Constant power drain.

• Pros:– If two devices both

drive the bus, it still works!

• Cons:– Rise-time is very

slow. – Constant power

drain.

• Sees use in I2C, CAN.

External memory attaches to the processor via the external memory controller and bus

44

Atmel SAM3U