Bus Protocols and Interfacing (adopted Steven and Marios’s slides) Bus basics I/O transactions MPC823 bus Reference: Chapter 13 of “White Book”

Slides:



Advertisements
Similar presentations
Bus arbitration Processor and DMA controllers both need to initiate data transfers on the bus and access main memory. The device that is allowed to initiate.
Advertisements

Homework Reading Machine Projects Labs
IT253: Computer Organization
Computer Architecture
1  1998 Morgan Kaufmann Publishers Interfacing Processors and Peripherals.
Datorteknik BusInterfacing bild 1 Bus Interfacing Processor-Memory Bus –High speed memory bus Backplane Bus –Processor-Interface bus –This is what we usually.
9/20/6Lecture 3 - Instruction Set - Al1 The Hardware Interface.
Chapter 2 Microprocessor Bus Transfers. Big- and Little-Endian Ordering Bit-endian processor architecture –High-order-byte-first (H-O-B-F) map the highest-order.
Embedded Systems Design: A Unified Hardware/Software Introduction, (c) 2000 Vahid/Givargis 1 A simple bus bus structure ProcessorMemory rd'/wr enable addr[0-11]
EECS 470 Busses in the real world Lecture 22 – Fall 2013.
Interfacing Processors and Peripherals Andreas Klappenecker CPSC321 Computer Architecture.
DIRECT MEMORY ACCESS CS 147 Thursday July 5,2001 SEEMA RAI.
Midterm Tuesday October 23 Covers Chapters 3 through 6 - Buses, Clocks, Timing, Edge Triggering, Level Triggering - Cache Memory Systems - Internal Memory.
TECH CH03 System Buses Computer Components Computer Function
680XX Hardware Interface Outline Goal Reading
1 COMP541 Interrupts, DMA, Serial I/O Montek Singh April 24, 2007.
1 EECS 373 Design of Microprocessor-Based Systems Prabal Dutta University of Michigan Lecture 4: Memory-Mapped I/O, Bus Architectures January 20, 2015.
9/20/6Lecture 3 - Instruction Set - Al Hardware interface (part 2)
CS-334: Computer Architecture
Computer Architecture Lecture 08 Fasih ur Rehman.
Digital System Bus A bus in a digital system is a collection of (usually unbroken) signal lines that carry module-to-module communications. The signals.
CPU BASICS, THE BUS, CLOCKS, I/O SUBSYSTEM Philip Chan.
NS Training Hardware.
CHAPTER 3 TOP LEVEL VIEW OF COMPUTER FUNCTION AND INTERCONNECTION
I/O Example: Disk Drives To access data: — seek: position head over the proper track (8 to 20 ms. avg.) — rotational latency: wait for desired sector (.5.
Top Level View of Computer Function and Interconnection.
DEVICES AND COMMUNICATION BUSES FOR DEVICES NETWORK
Computer Architecture System Interface Units Iolanthe II approaches Coromandel Harbour.
Interrupts, Buses Chapter 6.2.5, Introduction to Interrupts Interrupts are a mechanism by which other modules (e.g. I/O) may interrupt normal.
Direct Memory Access (DMA) Microprocessors I -1. Topics to be discussed  Basic DMA Concept Basic DMA Concept  DMA pins and timing DMA pins and timing.
MBG 1 CIS501, Fall 99 Lecture 18: Input/Output (I/O): Buses and Peripherals Michael B. Greenwald Computer Architecture CIS 501 Fall 1999.
EEE440 Computer Architecture
ECEG-3202 Computer Architecture and Organization Chapter 3 Top Level View of Computer Function and Interconnection.
Modes of transfer in computer
1 EECS 373 Design of Microprocessor-Based Systems Mark Brehob University of Michigan Lecture 4: Bit of ssembly, Memory-mapped I/O, APB January 21, 2014.
Computer Hardware A computer is made of internal components Central Processor Unit Internal External and external components.
12/16/  List the elements of 8255A Programmable Peripheral Interface (PPI)  Explain its various operating modes  Develop a simple program to.
Spring 2009W. Rhett DavisNC State UniversityECE 406Slide 1 ECE 406 – Design of Complex Digital Systems Lecture 16: Introduction to Buses and Interfaces.
Dr Mohamed Menacer College of Computer Science and Engineering, Taibah University CE-321: Computer.
Introduction to Microprocessors - chapter3 1 Chapter 3 The 8085 Microprocessor Architecture.
IT3002 Computer Architecture
CS 478: Microcontroller Systems University of Wisconsin-Eau Claire Dan Ernst Bus Protocols and Interfacing Bus basics I/O transactions MPC555 bus Reference:
MACHINE CYCLE AND T-STATE
Fall EE 333 Lillevik 333f06-l22 University of Portland School of Engineering Computer Organization Lecture 22 Project 6 Hard disk drive Bus arbitration.
Chapter 3 System Buses.  Hardwired systems are inflexible  General purpose hardware can do different tasks, given correct control signals  Instead.
8085 INTERNAL ARCHITECTURE.  Upon completing this topic, you should be able to: State all the register available in the 8085 microprocessor and explain.
1 EECS 373 Design of Microprocessor-Based Systems Prabal Dutta University of Michigan Lecture 4: Review, Simulation, ABI, and Memory-Mapped I/O September.
Amdahl’s Law & I/O Control Method 1. Amdahl’s Law The overall performance of a system is a result of the interaction of all of its components. System.
EECS 373 – lecture 4 Buses, serial communication and digital design.
Networked Embedded Systems Pengyu Zhang & Sachin Katti EE107 Spring 2016 Lecture 11 Direct Memory Access.
Interconnection Structures
Department of Computer Science and Engineering
Bus Interfacing Processor-Memory Bus Backplane Bus I/O Bus
Chapter 6 Input/Output Organization
16.317: Microprocessor System Design I
Slides developed in part by Mark Brehob & Prabal Dutta
EE 107 Fall 2017 Lecture 7 Serial Buses – I2C Direct Memory Access
EECS 373 Design of Microprocessor-Based Systems Mark Brehob
Md. Mojahidul Islam Lecturer Dept. of Computer Science & Engineering
Md. Mojahidul Islam Lecturer Dept. of Computer Science & Engineering
Suppose that a certain program takes 200 seconds of elapsed time to execute. Out of these 200 seconds, 180 seconds is the CPU time and the rest is I/O.
William Stallings Computer Organization and Architecture 7th Edition
Presentation transcript:

Bus Protocols and Interfacing (adopted Steven and Marios’s slides) Bus basics I/O transactions MPC823 bus Reference: Chapter 13 of “White Book”

Basic example Discuss a basic bus protocol –Asynchronous (no clock) –Initiator and Target –REQ#, ACK#, Data[7:0], ADS[7:0], CMD CMD=0 is read, CMD=1 is write. REQ# low means initiator is requesting something. ACK# low means target has done its job.

A read transaction Say initiator wants to read location 0x24 –Initiator sets ADS=0x24, CMD=0. –Initiator then sets REQ# to low. (why do we need a delay? How much of a delay?) –Target sees read request. –Target drives data onto data bus. –Target then sets ACK# to low. –Initiator grabs the data from the data bus. –Initiator sets REQ# to high, stops driving ADS and CMD –Target stops driving data, sets ACK# to low terminating the transaction

Read transaction ADS[7:0] CMD Data[7:0] REQ# ACK# ?? 0x24 ?? 0x55 A B C D E FG

A write transcation (write 0xF4 to location 0x31) –Initiator sets ADS=0x31, CMD=1, Data=0xF4 –Initiator then sets REQ# to low. –Target sees write request. –Target reads data from data bus. (Just has to store in a register, need not write all the way to memory!) –Target then sets ACK# to low. –Initiator sets REQ# to high & stops driving other lines. –Target sets ACK# to high terminating the transaction

The push-button (if ADS=0x04 write 0 or 1 depending on button) ADS[0] ADS[1] ADS[2] ADS[3] ADS[4] ADS[5] ADS[6] ADS[7] REQ# Button (0 or 1) 0 Data[0] Data[7].. Delay ACK# What about CMD?

The LED (1 bit reg written by LSB of address 0x05) ADS[2] ADS[0] ADS[1] ADS[3] ADS[4] ADS[5] ADS[6] ADS[7] REQ# Flip-flop which controls LED clock D DATA[2] DATA[0] DATA[1] DATA[3] DATA[4] DATA[5] DATA[6] DATA[7] Delay ACK#

MPC823 Bus The basic function of the MPC823 bus is similar, though slightly more complicated. (Chapter 13 of “White Book”) Timing is controlled by a global clock; all signals are in reference to the rising edge of this clock. 32-bit data bus D[0:31] 26-bit address bus A[6:31] A[0:5] not sent off chip On-chip peripherals still see 32 address bits Basic control lines: RD/#WR #TS (transfer start)—like #REQ but only asserted on first clock cycle of transaction #TA (transfer acknowledge)—like #ACK

MPC823 Read Master drives address, RD/#WR, and #TS to initiate a read transaction. Address and RD/#WR guaranteed valid at same rising clock edge that TS’ is asserted. Master deasserts #TS after one cycle, but keeps driving address and RD/#WR until it sees #TA. Slaves look at address and RD/#WR when #TS asserted. One of them will drive data and assert #TA. Master smaples data on same rising clock edge when #TA is asserted. Minimum transactions takes two clock cycles. Transactions can take longer; slow slaves add wait states by not asserting #TA.

MPC823 Read CLK A[6:31] D[0:31] RD/#WR #TS #TA ?? A1 D1 A2 D2 A1 A2

MPC823 Write Same start as for a read, except for polarity of RD/WR#. Master drives data by 2 nd cycle. Slaves look at address and RD/#WR when #TS is asserted. One of them will read data and assert #TA. As with reads, minimum transaction length is two cycles. Slaves can take longer, however, by not asserting #TA. Master keeps driving address, RD/#WR, and data until it sees #TA asserted.

MPC823 Write CLK A[6:31] D[0:31] RD/#WR #TS #TA ?? A1 D1 A2 D2 A1 A2

TSI[0:1] – Specifies the size of the data to be transferred. DP[0:3] – Data parity #Burst – Indicates a burst transaction #BDIP – Burst Data In Progress (more burst stuff) #BI – Burst Inhibit. Indicates that the slave doesn’t support burst transactions Many More… Of course, things aren’t that simple

Example of complexity: The Burst Mechanism (13.4.4) Burst transfers are used to move 16 bytes at a time #BURST must be asserted by master #BI must not be asserted by slave Must be a 16-byte aligned access Supports critical word first.

Arbitration (page 13-28) Requesting deviceArbiter Request the bus Assert #BR ACK bus mastership Wait for #BB to be deasserted Assert #BB Negate #BR Perform data transfer Release bus mastership Negate #BB Grant bus arbitration Assert #BG Terminate Arbitration Negate #BG

Transfer Alignment 0x MPC823 external bus supports natural address alignment Byte access: Any address alignment Half-word access: Address bit 31 equal to 0 Word access: Address bits 31 and 30 equal to 0

Dealing with Smaller Accesses: Reads 0x Assume that the word value 0x is stored at 0x1000 and that r4 contains 0x1000. What happens on the following transfers? D[0:7] D[8:15] D[16:23] D[24:31] lbz r3, 0 (r4) lbz r3, 1 (r4) lbz r3, 2 (r4) lbz r3, 3 (r4) lhz r3, 0 (r4) lhz r3, 2 (r4)

Dealing with Smaller Accesses: Writes How about the following transfers? D[0:7] D[8:15] D[16:23] D[24:31] stb r2, 0 (r4) stb r2, 1 (r4) stb r2, 2 (r4) stb r2, 3 (r4) sth r2, 0 (r4) sth r2, 1 (r4) 0x

On a write access, which two factors determine which bits in a 32-bit word are updated? On most wide buses, the master drives byte enable lines instead of less significant address bits Moto (16 bits): LDS’, UDS’ (no address LSB) 32-bit buses: Replace low 2 address bits with 4 byte enables MPC823 does not: Full byte address provided Size (byte, halfword, word) encoded on two control lines TSIZ[0-1] Dealing with Smaller Accesses

Unaligned Accesses 0x x100_ Consider two adjacent 32-bit memory locations and assume that r4 = 0x1000. What happens when the CPU executes the following instructions? lwz r3, 2 (r4) lwz r3, 1 (r4)

Problems with Unaligned Accesses What are some problems with unaligned accesses?

Basic bus issues What are the basic wires for specifying the transaction and moving the data –What are the types of transactions? How are they specified? –How is length of data transfer specified? Who can delay (insert wait states?) How is arbitration done? Out-of-order transfers allowed? –Any restrictions? Error reporting? Weirdness? –Alignment for example.

Transaction types Usually read/write with a length –But in a given domain, other info might be important. Data vs. Code access. I/O vs. memory access Hints to target device –Length might be arbitrary.

Delaying Who can delay and how –Usually a target (slave) can delay –Sometimes initiator (master) can delay –Sometimes initiator can drop the transaction –Sometimes the target has options on how to delay.

Arbitration Fairness –Even sharing, priority sharing, weighted sharing Mechanism –Centralized arbiter –Distributed arbiter –Combination Duration –Until done –Until someone else requests –Until certain time passes. –Combination

Out-of-order Does the bus allow transactions to complete out-of-order? –If so, can increase bandwidth (why?) –If so, might have to worry about ordering issues Memory consistency models not a topic for this class (take EECS 570!) but basics are pretty easy to grasp

Out-of-order: Ordering problem Processor 1 write I=1 write J=1 Processor 2 Write J=2 Write I=2 If both programs are executed in order, is there any setting of J and I which is impossible?