Assembly line - start up

Slides:



Advertisements
Similar presentations
1 ECE369 ECE369 Pipelining. 2 ECE369 “toupper” :converts any lowercase characters (with ASCII codes between 97 and 122) in the null-terminated argument.
Advertisements

Morgan Kaufmann Publishers The Processor
Datorteknik DelayedLoad bild 1 Delayed Load. Datorteknik DelayedLoad bild 2 All problems solved? NO, what will happen if lw $6 $0($1) add $4 $6.
Datorteknik Pipelining bild 1 Acceleration How to improve speed? At what costs?
Datorteknik Pipeline1 bild 1 No Assembly line. Datorteknik Pipeline1 bild 2 Assembly line - start up ChassisAxelsMotorSeatsBody Start up waste.
Pipelining Hwanmo Sung CS147 Presentation Professor Sin-Min Lee.
Lecture: Pipelining Basics
1 A few words about the quiz Closed book, but you may bring in a page of handwritten notes. –You need to know what the “core” MIPS instructions do. –I.
Chapter 6. (1)Long cycle time (2)Idle transistors.
1 Lecture 17: Basic Pipelining Today’s topics:  5-stage pipeline  Hazards and instruction scheduling Mid-term exam stats:  Highest: 90, Mean: 58.
1 Recap (Pipelining). 2 What is Pipelining? A way of speeding up execution of tasks Key idea : overlap execution of multiple taks.
Computer ArchitectureFall 2007 © October 31, CS-447– Computer Architecture M,W 10-11:20am Lecture 17 Review.
Lecture: Pipelining Basics
Datorteknik PerformanceAnalyse bild 1 Performance –what is it: measures of performance The CPU Performance Equation: –Execution time as the measure –what.
Lecture 24: CPU Design Today’s topic –Multi-Cycle ALU –Introduction to Pipelining 1.
INSTRUCTION PIPELINE. Introduction An instruction pipeline is a technique used in the design of computers and other digital electronic devices to increase.
Automobile Manufacturing 1. Build frame. 60 min. 2. Add engine. 50 min. 3. Build body. 80 min. 4. Paint. 40 min. 5. Finish.45 min. 275 min. Latency: Time.
Ch6a- 2 EE/CS/CPE Computer Organization  Seattle Pacific University Automobile Manufacturing 1. Build frame. 60 min. 2. Add engine. 50 min. 3.
1 Designing a Pipelined Processor In this Chapter, we will study 1. Pipelined datapath 2. Pipelined control 3. Data Hazards 4. Forwarding 5. Branch Hazards.
1 COMP541 Pipelined MIPS Montek Singh Mar 30, 2010.
CECS 440 Pipelining.1(c) 2014 – R. W. Allison [slides adapted from D. Patterson slides with additional credits to M.J. Irwin]
1 COMP541 Pipelined MIPS Montek Singh Apr 9, 2012.
Branch Prediction CS 3220 Fall 2014 Hadi Esmaeilzadeh Georgia Institute of Technology Some slides adopted from Prof. Milos Prvulovic.
Datapath and Control AddressInstruction Memory Write Data Reg Addr Register File ALU Data Memory Address Write Data Read Data PC Read Data Read Data.
Pipelining 7/12/2013.
Speed up on cycle time Stalls – Optimizing compilers for pipelining
Lecture 18: Pipelining I.
Example Addiu $t1 $r0 3 Addiu $t1 $t1 5 Addiu $t1 $t1 7.
CDA3101 Recitation Section 8
Performance Performance The CPU Performance Equation:
ECE/CS 552: Pipelining © Prof. Mikko Lipasti
CSCI206 - Computer Organization & Programming
Lecture 16: Basic Pipelining
Lecture: Pipelining Basics
Performance of Single-cycle Design
Morgan Kaufmann Publishers The Processor
CSCI206 - Computer Organization & Programming
ECE232: Hardware Organization and Design
Lecture: Pipelining Basics
Instructional Parallelism
Chapter 4 The Processor Part 2
CSCI206 - Computer Organization & Programming
SOLUTIONS CHAPTER 4.
Lecture 16: Basic Pipelining
ELEC / Computer Architecture and Design Spring Pipelining (Chapter 6)
Guerilla Section #4 10/05 SDS & MIPS Datapath
Lecture 5: Pipelining Basics
CSCI206 - Computer Organization & Programming
Data Hazards Data Hazard
Pipelining Basic concept of assembly line
The Processor Lecture 3.6: Control Hazards
Systems I Pipelining I Topics Pipelining principles Pipeline overheads
Pipeline Principle A non-pipelined system of combination circuits (A, B, C) that computation requires total of 300 picoseconds. Comb. logic.
Basic Building Blocks Multiplexer Demultiplexer Adder +
Pipelining: Basic Concepts
ECE 352 Digital System Fundamentals
The branch instruction
Pipelining: critical path, pipeline hazards Prof. Eric Rotenberg
Reading and writing to data memory
Delayed Load What will happen if lw $6 $0($1) add $4 $6 $1
Basic Building Blocks Multiplexer Demultiplexer Adder +
Pipelining.
CSC3050 – Computer Architecture
Pipelining Basic concept of assembly line
Pipelining Appendix A and Chapter 3.
Acceleration How to improve speed? At what costs?.
Introduction to Computer Organization and Architecture
A relevant question Assuming you’ve got: One washer (takes 30 minutes)
Lecture: Pipelining Basics
Instructor: Michael Greenbaum
Presentation transcript:

Assembly line - start up 1. 2. 3. 4. 5. Chassis Axels Motor Seats Body Start up waste t

Assembly line - stop 1. 2. 3. 4. 5. Chassis Axels Motor Seats Body 1. 2. 3. 4. 5. Chassis Axels Motor Seats Body Nothing to do

Assembly line At start: final stations idle At stop: start stations idle 5 “stages” for each car Each car still takes 5 stages but... ... we produce one car each step

Assembly line  Time-set car Cycles instruction Without pipelining: = 5 5 cycles 1 instr. With pipelining = 1 5 cycles 5 instr.

Pipelining T = Nq * CPI * Tc We can bring this But what down to 1 determines this? The slowest pipeline stage “Rate determining step”

How we draw this? “At time t, pipeline step n works at the green car”

Pipeline is most efficient... ...when the work is equally shared “critical path” delay same for each stage or as close as possible

How do we break up a long critical path? Insert flip - flops!

But - “no free lunch” Delay: 30 ns 10 ns Latency: 1 cycle 3 cycles Also: The flip-flops have a cost!

Pipeline problem: Start-up waste Unexpected problem in a stage (stall) No more orders (flush)

Zero ext. = = Branch logic A ALU 4 B + = = 31 + Sgn/Ze extend