CS252 S05. Outline. Dynamic Branch Prediction. Static Branch Prediction. Dynamic Branch Prediction. Dynamic Branch Prediction

Size: px
Start display at page:

Download "CS252 S05. Outline. Dynamic Branch Prediction. Static Branch Prediction. Dynamic Branch Prediction. Dynamic Branch Prediction"

Transcription

1 Outline CMSC Computer Systems Architecture Lecture 9 Instruction Level Parallelism (Static & Dynamic Branch ion) ILP Compiler techniques to increase ILP Loop Unrolling Static Branch ion Dynamic Branch ion Overcoming Data Hazards with Dynamic Scheduling omasulo Algorithm Conclusion CMSC - 8 (from Patterson) Static Branch ion Previously scheduled code around delayed branch o reorder code around branches Need to predict branch statically during compile Simplest scheme is to predict a branch as taken Average misprediction = untaken branch frequency = % SPEC9 More accurate scheme predicts branches using profile information collected from earlier runs, and modify prediction based on last run: Misprediction Rate % % % % % % % % compress eqntott 8% espresso gcc Integer CMSC - 8 (from Patterson) % % li doduc ear hydrod H&P Figure. % 6% 9% % mdljdp Floating Point sucor % Dynamic Branch ion Why does prediction work? Underlying algorithm has regularities Data that is being operated on has regularities Instruction sequence has redundancies that are artifacts of way that humans/compilers think about problems Is dynamic branch prediction better than static branch prediction? Seems to be here are a small number of important branches in programs that have dynamic behavior CMSC - 8 (from Patterson) Dynamic Branch ion Performance = ƒ(accuracy, cost of misprediction) Branch History able (BH): table of -bit values indexed by lower bits of PC address index Says whether or not branch taken last time No address check (may refer to wrong branch) N aken N Not aken Problem: in a loop, -bit BH will cause two mispredictions (avg is 9 loop iterations before exit): End of loop, when it exits instead of looping as before First time through loop on next time through code, when it predicts exit instead of looping Dynamic Branch ion Solution: -bit prediction scheme where predictor changes prediction only if it mispredicts twice in a row aken Not aken Red: stop, not taken Green: go, taken H&P Figure. Adds hysteresis to decision making process N N N aken N Not aken CMSC - 8 (from Patterson) CMSC - 8 (from Patterson) 6 CS S

2 BH Accuracy Correlated Branch ion Mispredict because either: Wrong guess for that branch Got branch history of wrong branch when indexing into the table 96 entry table: Misprediction Rate SPEC89 % 8% 6% % % % 8% 6% % % % eqntott 8% % espresso gcc Integer CMSC - 8 (from Patterson) % % 9% li spice doduc % H&P Figure. spice 9% 9% fpppp Floating Point matrix % % nasa7 Idea record m most recently executed branches as taken or not taken, and use that pattern to select the proper n-bit branch history table In general, (m,n) predictor means record last m branches to select between m history tables, each with n-bit counters hus, old -bit BH is a (,) predictor Global Branch History: m-bit shift register keeping /N status of last m branches. Each entry in table has m n-bit predictors Also known as -level adaptive predictor Depends on previous branches! 7 CMSC - 8 (from Patterson) if (aa == ) aa = ; if (bb == ) bb = ; if (aa!= bb) { 8 Correlating Branches Correlated Branch ion (,) predictor w/ Behavior of recent branches selects between four predictions of next branch, updating just that prediction Or, addr bits + history bits give us 6-bit index into 6 = 6 predictors, each having two bits 8 total bits. Global branch history CMSC - 8 (from Patterson) Branch address ion -bits per branch predictor 9 Possible choices Local history + branch address Global branch history + branch address Global branch history only (no branch address)» Ignores branch instruction Branch address Global branch history Index into or Local branch history or Calculations Accuracy of Different Schemes 96-entry (,) predictor (i.e., -bit BH) k x = 8k bits k = address bits How to use the same # bits w/ a (,) predictor? 8k bits w/ -bit BH means k BHs the (, ) implies an entry has four BHs k entries, i.e. a (,) predictor w/ entries Frequency of Mispredictions % 8% 6% % % % 8% 6% % % % SPEC89 96 Entries -bit BH Unlimited Entries -bit BH Entries (,) BH % 6% 6% % % % % nasa7 matrix tomcatv doducd spice fpppp gcc H&P Figure.7 6% % % expresso eqntott li,96 entries: -bits per entry Unlimited entries: -bits/entry, entries (,) CMSC - 8 (from Patterson) CMSC - 8 (from Patterson) CS S

3 ournament ors Multilevel branch predictor Use n-bit saturating counter to choose between predictors Usually choice is between global and local predictors or correct. or incorrect. CMSC - 8 (from Patterson) N-bit Saturating Counter Used to choose between predictors X & Y N-bit counter value between and n - Counter operations Increment by (up to n -)» If X is correct & Y is incorrect Decrement by (down to )» If Y is correct & X is incorrect Choose predictor X if counter > n-, Y otherwise Can be used as predictor (X = taken, Y = not taken) = taken N = not taken ournament or : DEC Alpha 6 ournament predictor using K -bit counters 8K indexed by local branch address. Chooses between: Global predictor K entries indexed by history of last branches ( = K) 8K Each entry is a standard -bit predictor Local predictor K Local history table: K -bit entries recording last branches, index by branch address he pattern of the last occurrences of that K particular branch used to index table of K entries with -bit saturating counters otal size of predictor = 8K + 8K + K + K = 9K CMSC - 8 (from Patterson) (,) or B: BNEZ // branch B: BNEZ // branch B:,N,,N, B:,,,,N or Branch ion N N N N N aken N or ion based on state of predictor N Branch ion N Not aken N (,) or (,) or w/ Saturating Counter B: BNEZ // branch B: BNEZ // branch B:,N,,N, B:,,,,N aken Not aken N N aken N Not aken N B: BNEZ // branch B: BNEZ // branch B:,N,,N, B:,,,,N or Branch ion N N N N N N N or Branch ion N N N or Branch ion N N N N N N N or Branch ion N N N CS S

4 (,) or w/ Global History + Branch B: BNEZ // branch B: BNEZ // branch B:,N,,N, B:,,,,N or Branch ion N N aken N or N P / P Last global branch Not taken / aken Choose predictor based on last global branch action Not aken Branch ion N (,) or w/ Global History + Branch B: BNEZ // branch B: BNEZ // branch B:,N,,N, B:,,,,N or / / / / / / Branch ion N N N N N N aken N or / / / / / / N P / P Last global branch Not taken / aken Branch ion N N Not aken N (,) or w/ Local History + Branch B: BNEZ // branch B: BNEZ // branch B:,N,,N, B:,,,,N or Branch ion N N aken N or N P / P Last local branch Not taken / aken Choose predictor based on last local branch action Not aken Branch ion N (,) or w/ Local History + Branch B: BNEZ // branch B: BNEZ // branch B:,N,,N, B:,,,,N or / / / / / / Branch ion N N N N N aken N or / / / / / / N P / P Last local branch Not taken / aken Branch ion N N Not aken N (,) Global or (no Branch Addr) B: BNEZ // branch N B: BNEZ // branch N aken Not aken B:,N,,N, B:,,,,N P / P / P / P History = / / / Branch actions stored in Global History Branch Branch Iter History or ion History or ion?/?/?/? Same?/?/?/??/?/?/? ors! N?/?/?/??/?/?/??/?/?/??/?/?/? N?/?/?/??/?/?/??/?/?/? N Exit History based on last global branch actions; chose predictor based on history (,) Global or (no Branch Addr) N B: BNEZ // branch B: BNEZ // branch N aken Not aken B:,N,,N, B:,,,,N P / P / P / P History = / / / Iter History Exit or /// /// /// /// /// /// Branch ion N N N N History or /// /// /// /// /// Branch ion N N N N N CS S

5 (,) Global or (no Branch Addr) N aken aken B: BNEZ // branch N N B: BNEZ // branch Not aken Not aken B:,N,,N, N B:,,,,N P / P / P / P History = / / / ournament or -bit tournament predictor Indexed by branch address Chooses between two predictors. (,) Global or. (,) or w/ Local History Iter Exit History or /// /// /// /// /// /// Branch ion N N N N N N History or /// /// /// /// /// Branch ion N N N N N N Iter Exit, N N N N, N N N Branch or N N, N N N N N, N N Branch or N Iter Exit ournament or -bit tournament predictor Indexed by branch address Chooses between two predictors. (,) Global or. (,) or w/ Local History, N N N N Branch, or N N N N N N N N N, N N N N N, N N Branch or N N N N N Comparing ors (H&P Fig..8) Advantage of tournament predictor is ability to select the right predictor for a particular branch Particularly crucial for integer benchmarks. A typical tournament predictor will select the global predictor almost % of the time for the SPEC integer benchmarks and less than % of the time for the SPEC FP benchmarks CMSC - 8 (from Patterson) 8 Pentium Misprediction Rate (per instructions, not per branch) 6% misprediction rate per branch SPECint (9% of IN instructions are branch) % misprediction rate per branch SPECfp (% of FP instructions are branch) H&P Figure.8 Branch arget Buffers (BB) Branch target calculation is costly and stalls the instruction fetch. BB stores PCs the same way as caches he PC of a branch is sent to the BB When a match is found the corresponding ed PC is returned If the branch was predicted taken, instruction fetch continues at the returned predicted PC SPECint SPECfp CMSC - 8 (from Patterson) 9 CMSC - 8 (from Patterson) CS S

6 Branch arget Buffers Dynamic Branch ion Summary H&P Figure. ion becoming important part of execution Branch History able: bits for loop accuracy Correlation: Recently executed branches correlated with next branch Either different branches (GA) Or different executions of same branches (PA) ournament predictors take insight to next level, by using multiple predictors Usually one based on global information and one based on local information, and combining them with a selector In 6, tournament predictors using K bits are in processors like the Power and Pentium Branch arget Buffer: include branch address & prediction CMSC - 8 (from Patterson) CMSC - 8 (from Patterson) CS S

Administrivia. CMSC 411 Computer Systems Architecture Lecture 14 Instruction Level Parallelism (cont.) Control Dependencies

Administrivia. CMSC 411 Computer Systems Architecture Lecture 14 Instruction Level Parallelism (cont.) Control Dependencies Administrivia CMSC 411 Computer Systems Architecture Lecture 14 Instruction Level Parallelism (cont.) HW #3, on memory hierarchy, due Tuesday Continue reading Chapter 3 of H&P Alan Sussman als@cs.umd.edu

More information

Lecture 8 Dynamic Branch Prediction, Superscalar and VLIW. Computer Architectures S

Lecture 8 Dynamic Branch Prediction, Superscalar and VLIW. Computer Architectures S Lecture 8 Dynamic Branch Prediction, Superscalar and VLIW Computer Architectures 521480S Dynamic Branch Prediction Performance = ƒ(accuracy, cost of misprediction) Branch History Table (BHT) is simplest

More information

Instruction Level Parallelism

Instruction Level Parallelism Instruction Level Parallelism The potential overlap among instruction execution is called Instruction Level Parallelism (ILP) since instructions can be executed in parallel. There are mainly two approaches

More information

Instruction Level Parallelism. Taken from

Instruction Level Parallelism. Taken from Instruction Level Parallelism Taken from http://www.cs.utsa.edu/~dj/cs3853/lecture5.ppt Outline ILP Compiler techniques to increase ILP Loop Unrolling Static Branch Prediction Dynamic Branch Prediction

More information

Some material adapted from Mohamed Younis, UMBC CMSC 611 Spr 2003 course slides Some material adapted from Hennessy & Patterson / 2003 Elsevier

Some material adapted from Mohamed Younis, UMBC CMSC 611 Spr 2003 course slides Some material adapted from Hennessy & Patterson / 2003 Elsevier Some material adapted from Mohamed Younis, UMBC CMSC 611 Spr 2003 course slides Some material adapted from Hennessy & Patterson / 2003 Elsevier Science ! CPI = (1-branch%) * non-branch CPI + branch% *

More information

Tomasulo Loop Example

Tomasulo Loop Example Tomasulo Loop Example Loop: LD F0 0 R1 MULTD F4 F0 F2 SD F4 0 R1 SUBI R1 R1 #8 BNEZ R1 Loop This time assume Multiply takes 4 clocks Assume 1st load takes 8 clocks, 2nd load takes 1 clock Clocks for SUBI,

More information

CSE4201 Instruction Level Parallelism. Branch Prediction

CSE4201 Instruction Level Parallelism. Branch Prediction CSE4201 Instruction Level Parallelism Branch Prediction Prof. Mokhtar Aboelaze York University Based on Slides by Prof. L. Bhuyan (UCR) Prof. M. Shaaban (RIT) 1 Introduction With dynamic scheduling that

More information

NOW Handout Page 1. Outline. Csci 211 Computer System Architecture. Lec 4 Instruction Level Parallelism. Instruction Level Parallelism

NOW Handout Page 1. Outline. Csci 211 Computer System Architecture. Lec 4 Instruction Level Parallelism. Instruction Level Parallelism Outline Csci 211 Computer System Architecture Lec 4 Instruction Level Parallelism Xiuzhen Cheng Department of Computer Sciences The George Washington University ILP Compiler techniques to increase ILP

More information

COSC 6385 Computer Architecture. Instruction Level Parallelism

COSC 6385 Computer Architecture. Instruction Level Parallelism COSC 6385 Computer Architecture Instruction Level Parallelism Spring 2013 Instruction Level Parallelism Pipelining allows for overlapping the execution of instructions Limitations on the (pipelined) execution

More information

As the amount of ILP to exploit grows, control dependences rapidly become the limiting factor.

As the amount of ILP to exploit grows, control dependences rapidly become the limiting factor. Hiroaki Kobayashi // As the amount of ILP to exploit grows, control dependences rapidly become the limiting factor. Branches will arrive up to n times faster in an n-issue processor, and providing an instruction

More information

Dynamic Hardware Prediction. Basic Branch Prediction Buffers. N-bit Branch Prediction Buffers

Dynamic Hardware Prediction. Basic Branch Prediction Buffers. N-bit Branch Prediction Buffers Dynamic Hardware Prediction Importance of control dependences Branches and jumps are frequent Limiting factor as ILP increases (Amdahl s law) Schemes to attack control dependences Static Basic (stall the

More information

COSC 6385 Computer Architecture Dynamic Branch Prediction

COSC 6385 Computer Architecture Dynamic Branch Prediction COSC 6385 Computer Architecture Dynamic Branch Prediction Edgar Gabriel Spring 208 Pipelining Pipelining allows for overlapping the execution of instructions Limitations on the (pipelined) execution of

More information

NOW Handout Page 1. Review from Last Time. CSE 820 Graduate Computer Architecture. Lec 7 Instruction Level Parallelism. Recall from Pipelining Review

NOW Handout Page 1. Review from Last Time. CSE 820 Graduate Computer Architecture. Lec 7 Instruction Level Parallelism. Recall from Pipelining Review Review from Last Time CSE 820 Graduate Computer Architecture Lec 7 Instruction Level Parallelism Based on slides by David Patterson 4 papers: All about where to draw line between HW and SW IBM set foundations

More information

CSE 502 Graduate Computer Architecture. Lec 8-10 Instruction Level Parallelism

CSE 502 Graduate Computer Architecture. Lec 8-10 Instruction Level Parallelism CSE 502 Graduate Computer Architecture Lec 8-10 Instruction Level Parallelism Larry Wittie Computer Science, StonyBrook University http://www.cs.sunysb.edu/~cse502 and ~lw Slides adapted from David Patterson,

More information

Instruction Fetch and Branch Prediction. CprE 581 Computer Systems Architecture Readings: Textbook (4 th ed 2.3, 2.9); (5 th ed 3.

Instruction Fetch and Branch Prediction. CprE 581 Computer Systems Architecture Readings: Textbook (4 th ed 2.3, 2.9); (5 th ed 3. Instruction Fetch and Branch Prediction CprE 581 Computer Systems Architecture Readings: Textbook (4 th ed 2.3, 2.9); (5 th ed 3.3) 1 Frontend and Backend Feedback: - Prediction correct or not, update

More information

Multicycle ALU Operations 2/28/2011. Diversified Pipelines The Path Toward Superscalar Processors. Limitations of Our Simple 5 stage Pipeline

Multicycle ALU Operations 2/28/2011. Diversified Pipelines The Path Toward Superscalar Processors. Limitations of Our Simple 5 stage Pipeline //11 Limitations of Our Simple stage Pipeline Diversified Pipelines The Path Toward Superscalar Processors HPCA, Spring 11 Assumes single cycle EX stage for all instructions This is not feasible for Complex

More information

Reduction of Control Hazards (Branch) Stalls with Dynamic Branch Prediction

Reduction of Control Hazards (Branch) Stalls with Dynamic Branch Prediction ISA Support Needed By CPU Reduction of Control Hazards (Branch) Stalls with Dynamic Branch Prediction So far we have dealt with control hazards in instruction pipelines by: 1 2 3 4 Assuming that the branch

More information

Review Tomasulo. Lecture 17: ILP and Dynamic Execution #2: Branch Prediction, Multiple Issue. Tomasulo Algorithm and Branch Prediction

Review Tomasulo. Lecture 17: ILP and Dynamic Execution #2: Branch Prediction, Multiple Issue. Tomasulo Algorithm and Branch Prediction CS252 Graduate Computer Architecture Lecture 17: ILP and Dynamic Execution #2: Branch Prediction, Multiple Issue March 23, 01 Prof. David A. Patterson Computer Science 252 Spring 01 Review Tomasulo Reservations

More information

Chapter 3 (CONT) Instructor: Josep Torrellas CS433. Copyright J. Torrellas 1999,2001,2002,2007,2013 1

Chapter 3 (CONT) Instructor: Josep Torrellas CS433. Copyright J. Torrellas 1999,2001,2002,2007,2013 1 Chapter 3 (CONT) Instructor: Josep Torrellas CS433 Copyright J. Torrellas 1999,2001,2002,2007,2013 1 Dynamic Hardware Branch Prediction Control hazards are sources of losses, especially for processors

More information

ESE 545 Computer Architecture Instruction-Level Parallelism (ILP) and Static & Dynamic Instruction Scheduling Instruction level parallelism

ESE 545 Computer Architecture Instruction-Level Parallelism (ILP) and Static & Dynamic Instruction Scheduling Instruction level parallelism Computer Architecture ESE 545 Computer Architecture Instruction-Level Parallelism (ILP) and Static & Dynamic Instruction Scheduling 1 Outline ILP Compiler techniques to increase ILP Loop Unrolling Static

More information

Dynamic Control Hazard Avoidance

Dynamic Control Hazard Avoidance Dynamic Control Hazard Avoidance Consider Effects of Increasing the ILP Control dependencies rapidly become the limiting factor they tend to not get optimized by the compiler more instructions/sec ==>

More information

CS422 Computer Architecture

CS422 Computer Architecture CS422 Computer Architecture Spring 2004 Lecture 14, 18 Feb 2004 Bhaskaran Raman Department of CSE IIT Kanpur http://web.cse.iitk.ac.in/~cs422/index.html Dealing with Control Hazards Software techniques:

More information

Page 1. Today s Big Idea. Lecture 18: Branch Prediction + analysis resources => ILP

Page 1. Today s Big Idea. Lecture 18: Branch Prediction + analysis resources => ILP CS252 Graduate Computer Architecture Lecture 18: Branch Prediction + analysis resources => ILP April 2, 2 Prof. David E. Culler Computer Science 252 Spring 2 Today s Big Idea Reactive: past actions cause

More information

Lecture 5: VLIW, Software Pipelining, and Limits to ILP. Review: Tomasulo

Lecture 5: VLIW, Software Pipelining, and Limits to ILP. Review: Tomasulo Lecture 5: VLIW, Software Pipelining, and Limits to ILP Professor David A. Patterson Computer Science 252 Spring 1998 DAP.F96 1 Review: Tomasulo Prevents Register as bottleneck Avoids WAR, WAW hazards

More information

Lecture 15: Instruc.on Level Parallelism -- Introduc.on, Compiler Techniques, and Advanced Branch Predic.on

Lecture 15: Instruc.on Level Parallelism -- Introduc.on, Compiler Techniques, and Advanced Branch Predic.on Lecture 15: Instruc.on Level Parallelism -- Introduc.on, Compiler Techniques, and Advanced Branch Predic.on CSE 564 Computer Architecture Summer 2017 Department of Computer Science and Engineering Yonghong

More information

Page 1. CISC 662 Graduate Computer Architecture. Lecture 8 - ILP 1. Pipeline CPI. Pipeline CPI (I) Pipeline CPI (II) Michela Taufer

Page 1. CISC 662 Graduate Computer Architecture. Lecture 8 - ILP 1. Pipeline CPI. Pipeline CPI (I) Pipeline CPI (II) Michela Taufer CISC 662 Graduate Computer Architecture Lecture 8 - ILP 1 Michela Taufer Pipeline CPI http://www.cis.udel.edu/~taufer/teaching/cis662f07 Powerpoint Lecture Notes from John Hennessy and David Patterson

More information

Functional Units. Registers. The Big Picture: Where are We Now? The Five Classic Components of a Computer Processor Input Control Memory

Functional Units. Registers. The Big Picture: Where are We Now? The Five Classic Components of a Computer Processor Input Control Memory The Big Picture: Where are We Now? CS152 Computer Architecture and Engineering Lecture 18 The Five Classic Components of a Computer Processor Input Control Dynamic Scheduling (Cont), Speculation, and ILP

More information

Announcements. ECE4750/CS4420 Computer Architecture L10: Branch Prediction. Edward Suh Computer Systems Laboratory

Announcements. ECE4750/CS4420 Computer Architecture L10: Branch Prediction. Edward Suh Computer Systems Laboratory ECE4750/CS4420 Computer Architecture L10: Branch Prediction Edward Suh Computer Systems Laboratory suh@csl.cornell.edu Announcements Lab2 and prelim grades Back to the regular office hours 2 1 Overview

More information

Advanced Computer Architecture

Advanced Computer Architecture Advanced Computer Architecture Instruction Level Parallelism Myung Hoon, Sunwoo School of Electrical and Computer Engineering Ajou University Outline ILP Compiler techniques to increase ILP Loop Unrolling

More information

Lecture 5: VLIW, Software Pipelining, and Limits to ILP Professor David A. Patterson Computer Science 252 Spring 1998

Lecture 5: VLIW, Software Pipelining, and Limits to ILP Professor David A. Patterson Computer Science 252 Spring 1998 Lecture 5: VLIW, Software Pipelining, and Limits to ILP Professor David A. Patterson Computer Science 252 Spring 1998 DAP.F96 1 Review: Tomasulo Prevents Register as bottleneck Avoids WAR, WAW hazards

More information

Pipelining. Ideal speedup is number of stages in the pipeline. Do we achieve this? 2. Improve performance by increasing instruction throughput ...

Pipelining. Ideal speedup is number of stages in the pipeline. Do we achieve this? 2. Improve performance by increasing instruction throughput ... CHAPTER 6 1 Pipelining Instruction class Instruction memory ister read ALU Data memory ister write Total (in ps) Load word 200 100 200 200 100 800 Store word 200 100 200 200 700 R-format 200 100 200 100

More information

EI 338: Computer Systems Engineering (Operating Systems & Computer Architecture)

EI 338: Computer Systems Engineering (Operating Systems & Computer Architecture) EI 338: Computer Systems Engineering (Operating Systems & Computer Architecture) Dept. of Computer Science & Engineering Chentao Wu wuct@cs.sjtu.edu.cn Download lectures ftp://public.sjtu.edu.cn User:

More information

Instruction-Level Parallelism and Its Exploitation (Part III) ECE 154B Dmitri Strukov

Instruction-Level Parallelism and Its Exploitation (Part III) ECE 154B Dmitri Strukov Instruction-Level Parallelism and Its Exploitation (Part III) ECE 154B Dmitri Strukov Dealing With Control Hazards Simplest solution to stall pipeline until branch is resolved and target address is calculated

More information

NOW Handout Page 1. COSC 5351 Advanced Computer Architecture

NOW Handout Page 1. COSC 5351 Advanced Computer Architecture COSC 5351 Advanced Computer Slides modified from Hennessy CS252 course slides ILP Compiler techniques to increase ILP Loop Unrolling Static Branch Prediction Dynamic Branch Prediction Overcoming Data Hazards

More information

CMSC 411 Computer Systems Architecture Lecture 13 Instruction Level Parallelism 6 (Limits to ILP & Threading)

CMSC 411 Computer Systems Architecture Lecture 13 Instruction Level Parallelism 6 (Limits to ILP & Threading) CMSC 411 Computer Systems Architecture Lecture 13 Instruction Level Parallelism 6 (Limits to ILP & Threading) Limits to ILP Conflicting studies of amount of ILP Benchmarks» vectorized Fortran FP vs. integer

More information

Old formulation of branch paths w/o prediction. bne $2,$3,foo subu $3,.. ld $2,..

Old formulation of branch paths w/o prediction. bne $2,$3,foo subu $3,.. ld $2,.. Old formulation of branch paths w/o prediction bne $2,$3,foo subu $3,.. ld $2,.. Cleaner formulation of Branching Front End Back End Instr Cache Guess PC Instr PC dequeue!= dcache arch. PC branch resolution

More information

Instruction Level Parallelism (ILP)

Instruction Level Parallelism (ILP) 1 / 26 Instruction Level Parallelism (ILP) ILP: The simultaneous execution of multiple instructions from a program. While pipelining is a form of ILP, the general application of ILP goes much further into

More information

Appendix C. Authors: John Hennessy & David Patterson. Copyright 2011, Elsevier Inc. All rights Reserved. 1

Appendix C. Authors: John Hennessy & David Patterson. Copyright 2011, Elsevier Inc. All rights Reserved. 1 Appendix C Authors: John Hennessy & David Patterson Copyright 2011, Elsevier Inc. All rights Reserved. 1 Figure C.2 The pipeline can be thought of as a series of data paths shifted in time. This shows

More information

Instruction-Level Parallelism Dynamic Branch Prediction. Reducing Branch Penalties

Instruction-Level Parallelism Dynamic Branch Prediction. Reducing Branch Penalties Instruction-Level Parallelism Dynamic Branch Prediction CS448 1 Reducing Branch Penalties Last chapter static schemes Move branch calculation earlier in pipeline Static branch prediction Always taken,

More information

Chapter-5 Memory Hierarchy Design

Chapter-5 Memory Hierarchy Design Chapter-5 Memory Hierarchy Design Unlimited amount of fast memory - Economical solution is memory hierarchy - Locality - Cost performance Principle of locality - most programs do not access all code or

More information

ECE473 Computer Architecture and Organization. Pipeline: Control Hazard

ECE473 Computer Architecture and Organization. Pipeline: Control Hazard Computer Architecture and Organization Pipeline: Control Hazard Lecturer: Prof. Yifeng Zhu Fall, 2015 Portions of these slides are derived from: Dave Patterson UCB Lec 15.1 Pipelining Outline Introduction

More information

Lecture 8: Compiling for ILP and Branch Prediction. Advanced pipelining and instruction level parallelism

Lecture 8: Compiling for ILP and Branch Prediction. Advanced pipelining and instruction level parallelism Lecture 8: Compiling for ILP and Branch Prediction Kunle Olukotun Gates 302 kunle@ogun.stanford.edu http://www-leland.stanford.edu/class/ee282h/ 1 Advanced pipelining and instruction level parallelism

More information

CS422 Computer Architecture

CS422 Computer Architecture CS422 Computer Architecture Spring 2004 Lecture 17, 26 Feb 2004 Bhaskaran Raman Department of CSE IIT Kanpur http://web.cse.iitk.ac.in/~cs422/index.html Speculation Wish to move instructions across branches

More information

Topics. Digital Systems Architecture EECE EECE Predication, Prediction, and Speculation

Topics. Digital Systems Architecture EECE EECE Predication, Prediction, and Speculation Digital Systems Architecture EECE 343-01 EECE 292-02 Predication, Prediction, and Speculation Dr. William H. Robinson February 25, 2004 http://eecs.vanderbilt.edu/courses/eece343/ Topics Aha, now I see,

More information

Virtual memory why? Virtual memory parameters Compared to first-level cache Parameter First-level cache Virtual memory. Virtual memory concepts

Virtual memory why? Virtual memory parameters Compared to first-level cache Parameter First-level cache Virtual memory. Virtual memory concepts Lecture 16 Virtual memory why? Virtual memory: Virtual memory concepts (5.10) Protection (5.11) The memory hierarchy of Alpha 21064 (5.13) Virtual address space proc 0? s space proc 1 Physical memory Virtual

More information

Multiple Instruction Issue and Hardware Based Speculation

Multiple Instruction Issue and Hardware Based Speculation Multiple Instruction Issue and Hardware Based Speculation Soner Önder Michigan Technological University, Houghton MI www.cs.mtu.edu/~soner Hardware Based Speculation Exploiting more ILP requires that we

More information

Lecture 19: Memory Hierarchy Five Ways to Reduce Miss Penalty (Second Level Cache) Admin

Lecture 19: Memory Hierarchy Five Ways to Reduce Miss Penalty (Second Level Cache) Admin Lecture 19: Memory Hierarchy Five Ways to Reduce Miss Penalty (Second Level Cache) Professor Alvin R. Lebeck Computer Science 220 Fall 1999 Exam Average 76 90-100 4 80-89 3 70-79 3 60-69 5 < 60 1 Admin

More information

Exploitation of instruction level parallelism

Exploitation of instruction level parallelism Exploitation of instruction level parallelism Computer Architecture J. Daniel García Sánchez (coordinator) David Expósito Singh Francisco Javier García Blas ARCOS Group Computer Science and Engineering

More information

Super Scalar. Kalyan Basu March 21,

Super Scalar. Kalyan Basu March 21, Super Scalar Kalyan Basu basu@cse.uta.edu March 21, 2007 1 Super scalar Pipelines A pipeline that can complete more than 1 instruction per cycle is called a super scalar pipeline. We know how to build

More information

Page # CISC 662 Graduate Computer Architecture. Lecture 8 - ILP 1. Pipeline CPI. Pipeline CPI (I) Michela Taufer

Page # CISC 662 Graduate Computer Architecture. Lecture 8 - ILP 1. Pipeline CPI. Pipeline CPI (I) Michela Taufer CISC 662 Graduate Computer Architecture Lecture 8 - ILP 1 Michela Taufer http://www.cis.udel.edu/~taufer/teaching/cis662f07 Powerpoint Lecture Notes from John Hennessy and David Patterson s: Computer Architecture,

More information

Recall from Pipelining Review. Instruction Level Parallelism and Dynamic Execution

Recall from Pipelining Review. Instruction Level Parallelism and Dynamic Execution 332 Advanced Computer Architecture Chapter 4 Instruction Level Parallelism and Dynamic Execution January 2004 Paul H J Kelly These lecture notes are partly based on the course text, Hennessy and Patterson

More information

Outline EEL 5764 Graduate Computer Architecture. Chapter 3 Limits to ILP and Simultaneous Multithreading. Overcoming Limits - What do we need??

Outline EEL 5764 Graduate Computer Architecture. Chapter 3 Limits to ILP and Simultaneous Multithreading. Overcoming Limits - What do we need?? Outline EEL 7 Graduate Computer Architecture Chapter 3 Limits to ILP and Simultaneous Multithreading! Limits to ILP! Thread Level Parallelism! Multithreading! Simultaneous Multithreading Ann Gordon-Ross

More information

CISC 662 Graduate Computer Architecture Lecture 13 - Limits of ILP

CISC 662 Graduate Computer Architecture Lecture 13 - Limits of ILP CISC 662 Graduate Computer Architecture Lecture 13 - Limits of ILP Michela Taufer http://www.cis.udel.edu/~taufer/teaching/cis662f07 Powerpoint Lecture Notes from John Hennessy and David Patterson s: Computer

More information

Control Dependence, Branch Prediction

Control Dependence, Branch Prediction Control Dependence, Branch Prediction Outline Control dependences Branch evaluation delay Branch delay slot Branch prediction Static Dynamic correlating, local, global. Control Dependences Program correctness

More information

Instruction Level Parallelism

Instruction Level Parallelism Instruction Level Parallelism Software View of Computer Architecture COMP2 Godfrey van der Linden 200-0-0 Introduction Definition of Instruction Level Parallelism(ILP) Pipelining Hazards & Solutions Dynamic

More information

Branch Prediction Chapter 3

Branch Prediction Chapter 3 1 Branch Prediction Chapter 3 2 More on Dependencies We will now look at further techniques to deal with dependencies which negatively affect ILP in program code. A dependency may be overcome in two ways:

More information

CISC 662 Graduate Computer Architecture Lecture 11 - Hardware Speculation Branch Predictions

CISC 662 Graduate Computer Architecture Lecture 11 - Hardware Speculation Branch Predictions CISC 662 Graduate Computer Architecture Lecture 11 - Hardware Speculation Branch Predictions Michela Taufer http://www.cis.udel.edu/~taufer/teaching/cis6627 Powerpoint Lecture Notes from John Hennessy

More information

Improving Cache Performance. Dr. Yitzhak Birk Electrical Engineering Department, Technion

Improving Cache Performance. Dr. Yitzhak Birk Electrical Engineering Department, Technion Improving Cache Performance Dr. Yitzhak Birk Electrical Engineering Department, Technion 1 Cache Performance CPU time = (CPU execution clock cycles + Memory stall clock cycles) x clock cycle time Memory

More information

CMSC411 Fall 2013 Midterm 2 Solutions

CMSC411 Fall 2013 Midterm 2 Solutions CMSC411 Fall 2013 Midterm 2 Solutions 1. (12 pts) Memory hierarchy a. (6 pts) Suppose we have a virtual memory of size 64 GB, or 2 36 bytes, where pages are 16 KB (2 14 bytes) each, and the machine has

More information

Metodologie di Progettazione Hardware-Software

Metodologie di Progettazione Hardware-Software Metodologie di Progettazione Hardware-Software Advanced Pipelining and Instruction-Level Paralelism Metodologie di Progettazione Hardware/Software LS Ing. Informatica 1 ILP Instruction-level Parallelism

More information

Lecture 9: Case Study MIPS R4000 and Introduction to Advanced Pipelining Professor Randy H. Katz Computer Science 252 Spring 1996

Lecture 9: Case Study MIPS R4000 and Introduction to Advanced Pipelining Professor Randy H. Katz Computer Science 252 Spring 1996 Lecture 9: Case Study MIPS R4000 and Introduction to Advanced Pipelining Professor Randy H. Katz Computer Science 252 Spring 1996 RHK.SP96 1 Review: Evaluating Branch Alternatives Two part solution: Determine

More information

Reducing Branch Costs via Branch Alignment

Reducing Branch Costs via Branch Alignment Reducing Branch Costs via Branch Alignment Brad Calder and Dirk Grunwald Department of Computer Science Campus Box 30 University of Colorado Boulder, CO 80309 030 USA fcalder,grunwaldg@cs.colorado.edu

More information

HY425 Lecture 05: Branch Prediction

HY425 Lecture 05: Branch Prediction HY425 Lecture 05: Branch Prediction Dimitrios S. Nikolopoulos University of Crete and FORTH-ICS October 19, 2011 Dimitrios S. Nikolopoulos HY425 Lecture 05: Branch Prediction 1 / 45 Exploiting ILP in hardware

More information

Static Branch Prediction

Static Branch Prediction Static Branch Prediction Branch prediction schemes can be classified into static and dynamic schemes. Static methods are usually carried out by the compiler. They are static because the prediction is already

More information

UG4 Honours project selection: Talk to Vijay or Boris if interested in computer architecture projects

UG4 Honours project selection: Talk to Vijay or Boris if interested in computer architecture projects Announcements UG4 Honours project selection: Talk to Vijay or Boris if interested in computer architecture projects Inf3 Computer Architecture - 2017-2018 1 Last time: Tomasulo s Algorithm Inf3 Computer

More information

Lecture 7 Instruction Level Parallelism (5) EEC 171 Parallel Architectures John Owens UC Davis

Lecture 7 Instruction Level Parallelism (5) EEC 171 Parallel Architectures John Owens UC Davis Lecture 7 Instruction Level Parallelism (5) EEC 171 Parallel Architectures John Owens UC Davis Credits John Owens / UC Davis 2007 2009. Thanks to many sources for slide material: Computer Organization

More information

HY425 Lecture 09: Software to exploit ILP

HY425 Lecture 09: Software to exploit ILP HY425 Lecture 09: Software to exploit ILP Dimitrios S. Nikolopoulos University of Crete and FORTH-ICS November 4, 2010 ILP techniques Hardware Dimitrios S. Nikolopoulos HY425 Lecture 09: Software to exploit

More information

CISC 662 Graduate Computer Architecture Lecture 13 - Limits of ILP

CISC 662 Graduate Computer Architecture Lecture 13 - Limits of ILP CISC 662 Graduate Computer Architecture Lecture 13 - Limits of ILP Michela Taufer http://www.cis.udel.edu/~taufer/teaching/cis662f07 Powerpoint Lecture Notes from John Hennessy and David Patterson s: Computer

More information

HY425 Lecture 09: Software to exploit ILP

HY425 Lecture 09: Software to exploit ILP HY425 Lecture 09: Software to exploit ILP Dimitrios S. Nikolopoulos University of Crete and FORTH-ICS November 4, 2010 Dimitrios S. Nikolopoulos HY425 Lecture 09: Software to exploit ILP 1 / 44 ILP techniques

More information

Review: Evaluating Branch Alternatives. Lecture 3: Introduction to Advanced Pipelining. Review: Evaluating Branch Prediction

Review: Evaluating Branch Alternatives. Lecture 3: Introduction to Advanced Pipelining. Review: Evaluating Branch Prediction Review: Evaluating Branch Alternatives Lecture 3: Introduction to Advanced Pipelining Two part solution: Determine branch taken or not sooner, AND Compute taken branch address earlier Pipeline speedup

More information

Memory Hierarchy 3 Cs and 6 Ways to Reduce Misses

Memory Hierarchy 3 Cs and 6 Ways to Reduce Misses Memory Hierarchy 3 Cs and 6 Ways to Reduce Misses Soner Onder Michigan Technological University Randy Katz & David A. Patterson University of California, Berkeley Four Questions for Memory Hierarchy Designers

More information

CMSC 611: Advanced Computer Architecture

CMSC 611: Advanced Computer Architecture CMSC 611: Advanced Computer Architecture Performance Some material adapted from Mohamed Younis, UMBC CMSC 611 Spr 2003 course slides Some material adapted from Hennessy & Patterson / 2003 Elsevier Science

More information

Improving Performance: Pipelining

Improving Performance: Pipelining Improving Performance: Pipelining Memory General registers Memory ID EXE MEM WB Instruction Fetch (includes PC increment) ID Instruction Decode + fetching values from general purpose registers EXE EXEcute

More information

CS 152 Computer Architecture and Engineering. Lecture 8 - Memory Hierarchy-III

CS 152 Computer Architecture and Engineering. Lecture 8 - Memory Hierarchy-III CS 152 Computer Architecture and Engineering Lecture 8 - Memory Hierarchy-III Krste Asanovic Electrical Engineering and Computer Sciences University of California at Berkeley http://www.eecs.berkeley.edu/~krste

More information

Classification Steady-State Cache Misses: Techniques To Improve Cache Performance:

Classification Steady-State Cache Misses: Techniques To Improve Cache Performance: #1 Lec # 9 Winter 2003 1-21-2004 Classification Steady-State Cache Misses: The Three C s of cache Misses: Compulsory Misses Capacity Misses Conflict Misses Techniques To Improve Cache Performance: Reduce

More information

CS425 Computer Systems Architecture

CS425 Computer Systems Architecture CS425 Computer Systems Architecture Fall 2018 Static Instruction Scheduling 1 Techniques to reduce stalls CPI = Ideal CPI + Structural stalls per instruction + RAW stalls per instruction + WAR stalls per

More information

Hardware Speculation Support

Hardware Speculation Support Hardware Speculation Support Conditional instructions Most common form is conditional move BNEZ R1, L ;if MOV R2, R3 ;then CMOVZ R2,R3, R1 L: ;else Other variants conditional loads and stores nullification

More information

Lecture 6 MIPS R4000 and Instruction Level Parallelism. Computer Architectures S

Lecture 6 MIPS R4000 and Instruction Level Parallelism. Computer Architectures S Lecture 6 MIPS R4000 and Instruction Level Parallelism Computer Architectures 521480S Case Study: MIPS R4000 (200 MHz, 64-bit instructions, MIPS-3 instruction set) 8 Stage Pipeline: first half of fetching

More information

Looking for Instruction Level Parallelism (ILP) Branch Prediction. Branch Prediction. Importance of Branch Prediction

Looking for Instruction Level Parallelism (ILP) Branch Prediction. Branch Prediction. Importance of Branch Prediction Looking for Instruction Level Parallelism (ILP) Branch Prediction We want to identify and exploit ILP instructions that can potentially be executed at the same time. Branches are 15-20% of instructions

More information

Branch prediction ( 3.3) Dynamic Branch Prediction

Branch prediction ( 3.3) Dynamic Branch Prediction prediction ( 3.3) Static branch prediction (built into the architecture) The default is to assume that branches are not taken May have a design which predicts that branches are taken It is reasonable to

More information

Outline EEL 5764 Graduate Computer Architecture. Chapter 2 - Instruction Level Parallelism. Recall from Pipelining Review

Outline EEL 5764 Graduate Computer Architecture. Chapter 2 - Instruction Level Parallelism. Recall from Pipelining Review Outline EEL 5764 Graduate Computer Architecture Chapter 2 - Instruction Level Parallelism Ann Gordon-Ross Electrical and Computer Engineering University of Florida ILP Compiler techniques to increase ILP

More information

CS 152 Computer Architecture and Engineering. Lecture 8 - Memory Hierarchy-III

CS 152 Computer Architecture and Engineering. Lecture 8 - Memory Hierarchy-III CS 152 Computer Architecture and Engineering Lecture 8 - Memory Hierarchy-III Krste Asanovic Electrical Engineering and Computer Sciences University of California at Berkeley http://www.eecs.berkeley.edu/~krste!

More information

Predict Not Taken. Revisiting Branch Hazard Solutions. Filling the delay slot (e.g., in the compiler) Delayed Branch

Predict Not Taken. Revisiting Branch Hazard Solutions. Filling the delay slot (e.g., in the compiler) Delayed Branch branch taken Revisiting Branch Hazard Solutions Stall Predict Not Taken Predict Taken Branch Delay Slot Branch I+1 I+2 I+3 Predict Not Taken branch not taken Branch I+1 IF (bubble) (bubble) (bubble) (bubble)

More information

Topic 14: Dealing with Branches

Topic 14: Dealing with Branches Topic 14: Dealing with Branches COS / ELE 375 Computer Architecture and Organization Princeton University Fall 2015 Prof. David August 1 FLASHBACK: Pipeline Hazards Control Hazards What is the next instruction?

More information

Topic 14: Dealing with Branches

Topic 14: Dealing with Branches Topic 14: Dealing with Branches COS / ELE 375 FLASHBACK: Pipeline Hazards Control Hazards What is the next instruction? Branch instructions take time to compute this. Stall, Predict, or Delay: Computer

More information

CISC 662 Graduate Computer Architecture Lecture 18 - Cache Performance. Why More on Memory Hierarchy?

CISC 662 Graduate Computer Architecture Lecture 18 - Cache Performance. Why More on Memory Hierarchy? CISC 662 Graduate Computer Architecture Lecture 18 - Cache Performance Michela Taufer Powerpoint Lecture Notes from John Hennessy and David Patterson s: Computer Architecture, 4th edition ---- Additional

More information

Branch statistics. 66% forward (i.e., slightly over 50% of total branches). Most often Not Taken 33% backward. Almost all Taken

Branch statistics. 66% forward (i.e., slightly over 50% of total branches). Most often Not Taken 33% backward. Almost all Taken Branch statistics Branches occur every 4-7 instructions on average in integer programs, commercial and desktop applications; somewhat less frequently in scientific ones Unconditional branches : 20% (of

More information

NOW Handout Page # Why More on Memory Hierarchy? CISC 662 Graduate Computer Architecture Lecture 18 - Cache Performance

NOW Handout Page # Why More on Memory Hierarchy? CISC 662 Graduate Computer Architecture Lecture 18 - Cache Performance CISC 66 Graduate Computer Architecture Lecture 8 - Cache Performance Michela Taufer Performance Why More on Memory Hierarchy?,,, Processor Memory Processor-Memory Performance Gap Growing Powerpoint Lecture

More information

Types of Cache Misses: The Three C s

Types of Cache Misses: The Three C s Types of Cache Misses: The Three C s 1 Compulsory: On the first access to a block; the block must be brought into the cache; also called cold start misses, or first reference misses. 2 Capacity: Occur

More information

Lecture-13 (ROB and Multi-threading) CS422-Spring

Lecture-13 (ROB and Multi-threading) CS422-Spring Lecture-13 (ROB and Multi-threading) CS422-Spring 2018 Biswa@CSE-IITK Cycle 62 (Scoreboard) vs 57 in Tomasulo Instruction status: Read Exec Write Exec Write Instruction j k Issue Oper Comp Result Issue

More information

1993. (BP-2) (BP-5, BP-10) (BP-6, BP-10) (BP-7, BP-10) YAGS (BP-10) EECC722

1993. (BP-2) (BP-5, BP-10) (BP-6, BP-10) (BP-7, BP-10) YAGS (BP-10) EECC722 Dynamic Branch Prediction Dynamic branch prediction schemes run-time behavior of branches to make predictions. Usually information about outcomes of previous occurrences of branches are used to predict

More information

Looking for Instruction Level Parallelism (ILP) Branch Prediction. Branch Prediction. Importance of Branch Prediction

Looking for Instruction Level Parallelism (ILP) Branch Prediction. Branch Prediction. Importance of Branch Prediction Looking for Instruction Level Parallelism (ILP) Branch Prediction We want to identify and exploit ILP instructions that can potentially be executed at the same time. Branches are 5-20% of instructions

More information

Page 1 ILP. ILP Basics & Branch Prediction. Smarter Schedule. Basic Block Problems. Parallelism independent enough

Page 1 ILP. ILP Basics & Branch Prediction. Smarter Schedule. Basic Block Problems. Parallelism independent enough ILP ILP Basics & Branch Prediction Today s topics: Compiler hazard mitigation loop unrolling SW pipelining Branch Prediction Parallelism independent enough e.g. avoid s» control correctly predict decision

More information

Instruction-Level Parallelism. Instruction Level Parallelism (ILP)

Instruction-Level Parallelism. Instruction Level Parallelism (ILP) Instruction-Level Parallelism CS448 1 Pipelining Instruction Level Parallelism (ILP) Limited form of ILP Overlapping instructions, these instructions can be evaluated in parallel (to some degree) Pipeline

More information

Cache Performance! ! Memory system and processor performance:! ! Improving memory hierarchy performance:! CPU time = IC x CPI x Clock time

Cache Performance! ! Memory system and processor performance:! ! Improving memory hierarchy performance:! CPU time = IC x CPI x Clock time Cache Performance!! Memory system and processor performance:! CPU time = IC x CPI x Clock time CPU performance eqn. CPI = CPI ld/st x IC ld/st IC + CPI others x IC others IC CPI ld/st = Pipeline time +

More information

COSC4201. Prof. Mokhtar Aboelaze York University

COSC4201. Prof. Mokhtar Aboelaze York University COSC4201 Chapter 3 Multi Cycle Operations Prof. Mokhtar Aboelaze York University Based on Slides by Prof. L. Bhuyan (UCR) Prof. M. Shaaban (RTI) 1 Multicycle Operations More than one function unit, each

More information

09-1 Multicycle Pipeline Operations

09-1 Multicycle Pipeline Operations 09-1 Multicycle Pipeline Operations 09-1 Material may be added to this set. Material Covered Section 3.7. Long-Latency Operations (Topics) Typical long-latency instructions: floating point Pipelined v.

More information

Getting CPI under 1: Outline

Getting CPI under 1: Outline CMSC 411 Computer Systems Architecture Lecture 12 Instruction Level Parallelism 5 (Improving CPI) Getting CPI under 1: Outline More ILP VLIW branch target buffer return address predictor superscalar more

More information

CS 152 Computer Architecture and Engineering. Lecture 8 - Memory Hierarchy-III

CS 152 Computer Architecture and Engineering. Lecture 8 - Memory Hierarchy-III CS 152 Computer Architecture and Engineering Lecture 8 - Memory Hierarchy-III Krste Asanovic Electrical Engineering and Computer Sciences University of California at Berkeley http://www.eecs.berkeley.edu/~krste

More information

BRANCH PREDICTORS. Mahdi Nazm Bojnordi. CS/ECE 6810: Computer Architecture. Assistant Professor School of Computing University of Utah

BRANCH PREDICTORS. Mahdi Nazm Bojnordi. CS/ECE 6810: Computer Architecture. Assistant Professor School of Computing University of Utah BRANCH PREDICTORS Mahdi Nazm Bojnordi Assistant Professor School of Computing University of Utah CS/ECE 6810: Computer Architecture Overview Announcements Homework 2 release: Sept. 26 th This lecture Dynamic

More information