Docsity
Docsity
Prepare for your exams
Prepare for your exams

Study with the several resources on Docsity

Earn points to download
Earn points to download

Earn points by helping other students or get them with a premium plan

Guidelines and tips
Guidelines and tips
Sell on Docsity
Sell on Docsity
Docsity AI
Docsity AI
Log inSign up

Prepare for your exams

Study with the several resources on Docsity

Find documents

Prepare for your exams with the study notes shared by other students like you on Docsity

Search for your university

Find the specific documents for your university's exams

Docsity AINEW

Summarize your documents, ask them questions, convert them into quizzes and concept maps

Explore questions

Clear up your doubts by reading the answers to questions asked by your fellow students

Earn points to download

Earn points by helping other students or get them with a premium plan

Share documents
download point icon20 Points

For each uploaded document

Answer questions
download point icon5 Points

For each given answer (max 1 per day)

All the ways to get free points
Get points immediately

Choose a premium plan with all the points you need

Study Opportunities

Choose your next study program

Get in touch with the best universities in the world. Search through thousands of universities and official partners

Community

Ask the community

Ask the community for help and clear up your study doubts

Free resources

Our save-the-student-ebooks!

Download our free guides on studying techniques, anxiety management strategies, and thesis advice from Docsity tutors

Branch Predictor Simulator Project for ECE 463/521 - Spring 2006, Study Guides, Projects, Research of Electrical and Electronics Engineering

North Carolina State University (NCSU)Electrical and Electronics Engineering

A project for ece 463 and 521 students where they will construct a branch predictor simulator and design branch predictors using local, global, and hybrid predictors. Students are encouraged to work in pairs and use c/c++ or java for implementation. The project includes designing, coding, and reporting on the results.

Typology: Study Guides, Projects, Research

Pre 2010

Uploaded on 03/18/2009

koofers-user-35m
koofers-user-35m 🇺🇸

4

(1)

10 documents

1 / 6

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
–1–
ECE 463/521: Spring 2006
Project #2: Branch Prediction
Design due: Friday, March 31, 2006, 11 PM
Code due: Monday, April 3, 2006, 11 PM
Report due: Friday, April 7, 2006, 11 PM
Project rules
1. All students are encouraged to work in teams of two, using pair programming. Pair
programming means programming where two people sit at the same workstation, writing
the code collaboratively.
2. ECE 521 students will have additional parts to do; therefore both members of a pair
should be from the same class (463 or 521).
3. You may not work with the same partner on more than one project this semester.
4. You must register your partnership by posting on the
Pair-Programming Partners
message board, under the topic Project 2.”
5. Sharing of code between teams will be considered cheating, and will be penalized in
accordance with the Academic Integrity policy.
6. It is acceptable for you to compare your results with other groups to help debug your
program. It is not acceptable to collaborate on the final experiments.
7. You must do all your work in the C/C++ or Java languages. C++ (or Java) is
encouraged because it enables straightforward code-reuse and division of labor.
8. Homework will be submitted over the Wolfware Submit system and run in the
Eos/Unity environment.
Project description
In this project you will construct a branch predictor simulator and use it to design branch
predictors.
Specification of the simulator
1. (
All students
) Model a local branch predictor with parameters (
m
,
n
)
m
= the number of low-order PC bits used to index into the local branch-
history table.
n
= number of bits in the local branch-history register.
Refer to the diagram of the local branch predictor at the top of the next page.
pf3
pf4
pf5

Partial preview of the text

Download Branch Predictor Simulator Project for ECE 463/521 - Spring 2006 and more Study Guides, Projects, Research Electrical and Electronics Engineering in PDF only on Docsity!

ECE 463/521: Spring 2006

Project #2: Branch Prediction

Design due: Friday, March 31, 2006, 11 PM

Code due: Monday, April 3, 2006, 11 PM

Report due: Friday, April 7, 2006, 11 PM

Project rules

  1. All students are encouraged to work in teams of two, using pair programming. Pair programming means programming where two people sit at the same workstation, writing the code collaboratively.
  2. ECE 521 students will have additional parts to do; therefore both members of a pair should be from the same class (463 or 521).
  3. You may not work with the same partner on more than one project this semester.
  4. You must register your partnership by posting on the Pair-Programming Partners message board, under the topic Project 2.”
  5. Sharing of code between teams wi accordance with the Academic Integrity policy.ll be considered cheating, and will be penalized in
  6. It is acceptable for you to compare your results with other groups to help debug your program. It is not acceptable to collaborate on the final experiments.
  7. You must do all your work in the C/C++ or Java languages. C++ (or Java) is encouraged because it enables straightforward code-reuse and division of labor.
  8. Homework will be submitted over the Wolfware Submit system and run in the Eos/Unity environment.

Project description

In this project you will construct a branch predictor simulator and use it to design branch predictors.

Specification of the simulator

  1. ( All students) Model a local branch predictor with parameters ( m, n) m = the number of low-order PC bits used to index into the local branch- history table. n = number of bits in the local branch-history register. Refer to the diagram of the local branch predictor at the top of the next page.

History regs. Counters Actual branch outcomes Prediction

Program Counter

m+1:

  • The table of local branch history registers is indexed by m low-order bits of the address of the branch instruction (i.e., PC value when the branch is fetched, not the branch target). These m bits are the least-significant bits of the address, no including the two least-significant bits of the PC, since these are always zero (that is, use bits m+1 through 2 of the PC).
  • The value in a branch-history register is used to index into an array of counters. Thus, if n is the number of bits in a branch-history register, there will be 2^ n counters.
  • The branch history registers are all initialized to zero at the beginning of the simulation
  • Every time a branch is predicted, the branch outcome is shifted into the most- significant bit position of the local branch-history register containing the history of that branch. ( Note: In real hardware, we would not be able to shift the branch outcome in until the EX stage of the pipeline. However, this is just a simulation, and we read the branch outcome from the trace file, so we can use it right away.)
  • There are no tags and no tag-checking (no miss signals for the predictor).
  • Each of the 2^ n^ counters contains 2 bits. (a) The counter is incremented when the branch is taken and decremented when it is not taken. (b) The counter saturates at 0 and 3 (c) The counter is updated after the prediction is made. (d) If the counter value is greater than or equal to two, the branch is predicted taken otherwise it is predicted not taken. (e) All counters should be set to 2 when the simulation begins.
  1. ( All students) Model a global branch predictor with parameter n. n = number of bits in the global branch history register.

PC

mmmm+1 : 2

(a) The counter saturates at 0 and 3 (the counter will never go below zero or above three). (b) All choice counters are initialized to 1 at the beginning of the simulation (c) The index for the chooser table is bit k+1 to bit 2 of the address of the branch (the value of the PC when the branch is encountered). (d) Only the predictor that was selected (local or global) is updated with the branch behavior. (e) The global history register will be updated even if the local branch predictor is selected. You should be able to select any of the predictors (local, global, or hybrid) based on a command-line parameter. The parameters of the simulation are: { m, n 1 } for local branch predictor. { n 2 } for global branch predictor. { { m, n 1 }, { n 2 }, k} for the hybrid predictor.

Input

The simulator reads a trace file in the following format: 0x〈hex branch PC〉 t/n 0x〈hex branch PC〉 t/n … where 〈hex branch PC〉 holds the hexadecimal form of address of the branch instruction in memory. “t” indicates that the branch is actually taken. “n” indicates that the branch is not taken. Example 1100023b t 0a050201 n

Output

The simulation output should include:

  1. The number of branches encountered.
  2. The number of branch mispredictions (predicted taken when not taken, predicted not taken when taken).
  3. Misprediction rate (#mispredictions / #branches).

Program interface requirements

  1. Your code must be able to run in the Unix environment on the Eos/Unity system.
  2. A makefile must be provided.
  3. A “make” will create a executable file--“brn_predictor”.
  4. The program must be executable from the command line, using the following format.

local: brn_predictor [tracefile name] [m1] [n1] global: brn_predictor [tracefile name] [n2] hybrid: brn_predictor [tracefile name] [m1] [n1] [n2] [k]

Example: For local branch predictor, eos% brn_predictor gcc_trace.txt 12 8 The first parameter is the number of low-order PC bits ( m = 12). The second parameter is the number of bits in history register ( n = 8).

Experiments

For the traces given in the course locker (directory /afs/eos/courses/ece/ece521/common/www/homework/projects/2/traces), use your simulator as follows: (1) ( All students ) Do a preliminary design-space search using the following configurations. Local Branch Predictor : (5 ≤ m ≤ 15, 2 ≤ n ≤ m), where n is even.

  • Produce one graph for each benchmark. y-axis: Branch-misprediction rate x-axis: m
  • Draw conclusions and discuss the results. Compare and contrast with (2). (2) ( All students) Do a preliminary design-space search using the following configurations: Global Branch Predictor: (5 ≤ n ≤ 15)
  • Produce one graph for each benchmark. y-axis: Branch-misprediction rate x-axis: n There should be one curve with data points for each value of n.
  • Draw conclusions and discuss the result. Compare and contrast with (a). (3) ( All students ) Do an intelligent search of the design space to minimize misprediction rate and minimize predictor cost in bits given a maximum budget of (a) 16 kilobytes of storage, and (b) 32 kilobytes of storage. (That is, do one analysis for 16 KB and another for 32 KB.) What to include in the report: For each of the benchmarks individually, report the best two predictors based on the constraints. Justify your choice with graphs and/or tables and discussion.