Cache Lab: Understanding Cache Memories Solution

This lab is managed by Rohan Sadale (sadal005@umn.edu) and Nathan  Bittner (bittn037@umn.edu)

 

 

 

 

1    Logistics

 

 

This is an individual project.  You must run this lab on CSE lab machines.  (Though this project can be completed on a personal linux machine, your final submission must be tested on a CSE lab machine. We will grade your project on a CSE lab machine.)

You will not receive any credits if we are not able to run your code.

 

 

2    Overview

 

 

This lab will help you understand the impact that cache memories can have on the performance of your C

programs.

 

The lab consists of two parts. In the first part you will write a small C program (about 200-300 lines) that simulates the behavior of a cache memory. In the second part, you will optimize a small matrix transpose function, with the goal of minimizing the number of cache misses.

 

 

3    Downloading the assignment

 

 

Download cachelab-handout.tar from Moodle

 

 

Start by copying cachelab-handout.tar to a protected Linux directory in which you plan to do your work. Then give the command

 

linux tar  xvf cachelab-handout.tar

 

 

 

 

 

 

This will create a directory called cachelab-handout that contains a number of files.  You will be modifying two files: csim.c and trans.c. To compile these files, type:

 

linux make   clean linux  make

 

WARNING: Do not let the Windows WinZip program open up your .tar file (many Web browsers are set to do this automatically). Instead, save the file to your Linux directory and use the Linux tar program to extract the files. In general, for this class you should NEVER use any platform other than Linux to modify your files. Doing so can cause loss of data (and important work!).

 

 

4    Description

 

 

The lab has two parts. In Part A you will implement a cache simulator. In Part B you will write a matrix transpose function that is optimized for cache performance.

 

 

4.1   Reference Trace Files

 

The traces subdirectory of the handout directory contains a collection of reference trace files that we will use to evaluate the correctness of the cache simulator you write in Part A. The trace files are generated by a Linux program called valgrind. For example, typing

 

linux valgrind --log-fd=1 --tool=lackey -v --trace-mem=yes ls -l

 

on the command line runs the executable program “ls  -l”, captures a trace of each of its memory accesses in the order they occur, and prints them on stdout.

Valgrind memory traces have the following form:

 

I 0400d7d4,8

M   0421c7f0,4

L  04f6b868,8

S  7ff0005c8,8

 

Each line denotes one or two memory accesses. The format of each line is

 

[space]operation address,size

 

The operation field denotes the type of memory access: “I” denotes an instruction load, “L” a data load, “S” a data store, and “M” a data modify (i.e., a data load followed by a data store). There is never a space before each “I”. There is always a space before each “M”, “L”, and “S”. The address field specifies a 64-bit hexadecimal memory address. The size field specifies the number of bytes accessed by the operation.

 

 

 

 

 

4.2   Part  A: Writing a Cache Simulator

 

In Part A you will write a cache simulator in csim.c that takes a valgrind memory trace as input, simulates the hit/miss behavior of a cache memory on this trace, and outputs the total number of hits, misses, and evictions.

We have provided you with the binary executable of a reference cache simulator, called csim-MRU-ref, that simulates the behavior of a cache with arbitrary size and associativity on a valgrind trace file. It uses the MRU (most-recently used) replacement policy when choosing which cache line to evict.

The reference simulator takes the following command-line arguments:

 

Usage: ./csim-MRU-ref [-hv]  -s  <s   -E <E  -b <b  -t <tracefile

 

 

• -h: Optional help flag that prints usage info

 

• -v: Optional verbose flag that displays trace info

 

• -s <s:  Number of set index bits (S = 2s is the number of sets)

 

• -E <E: Associativity (number of lines per set)

 

• -b <b: Number of block bits (B = 2b is the block size)

 

• -t <tracefile: Name of the valgrind trace to replay

 

 

linux ./csim-MRU-ref -s  4  -E 1  -b 4  -t traces/yi.trace hits:4  misses:5 evictions:3

 

 

The same example in verbose mode:

 

linux ./csim-MRU-ref -v -s  4  -E 1  -b 4  -t traces/yi.trace

L  10,1  miss

M   20,1  miss hit

L  22,1 hit

S  18,1 hit

L  110,1 miss eviction

L  210,1 miss eviction

M   12,1 miss eviction  hit hits:4  misses:5 evictions:3

 

Your job for Part A is to fill in the csim.c file so that it takes the same command line arguments and produces the identical output as the reference simulator. Notice that this file is almost completely empty. You’ll need to write it from scratch.

 

 

 

 

 

Programming Rules for Part  A

 

• Include your name and loginID in the header comment for csim.c.

 

• Your csim.c file must compile without warnings in order to receive credit.

 

• Your simulator must work correctly for arbitrary s, E,  and b.  This means that you will need to allocate storage for your simulator’s data structures using the malloc function. Type “man malloc” for information about this function.

 

• For this lab, we are interested only in data cache performance, so your simulator should ignore all instruction cache accesses (lines starting with “I”). Recall that valgrind always puts “I” in the first column (with no preceding space), and “M”, “L”, and “S” in the second column (with a preceding space). This may help you parse the trace.

 

• To receive credit for Part A, you must call the function printSummary, with the total number of hits, misses, and evictions, at the end of your main function:

 

printSummary(hit_count, miss_count, eviction_count);

 

• For this this lab, you should assume that memory accesses are aligned properly, such that a single memory access never crosses block boundaries.  By making this assumption, you can ignore the request sizes in the valgrind traces.

 

 

4.3   Part  B: Optimizing Matrix Transpose

 

In Part B you will write a transpose function in trans.c that causes as few cache misses as possible.

 



ji
 
Let A denote a matrix, and Aij denote the component on the ith row and jth column. The transpose of A, denoted AT , is a matrix such that Aij = AT .

 

To help you get started, we have given you an example transpose function in trans.c that computes the transpose of N × M matrix A and stores the results in M × N matrix B:

 

char trans_desc[]  =  "Simple row-wise scan  transpose";

void trans(int  M,  int N,   int A[N][M], int  B[M][N])

 

The example transpose function is correct, but it is inefficient because the access pattern results in relatively many cache misses.

Your job in Part B is to write a similar function, called transpose_submit, that minimizes the number of cache misses across different sized matrices:

 

char transpose_submit_desc[] =  "Transpose submission";

void transpose_submit(int  M,  int N,   int A[N][M], int  B[M][N]);

 

Do not change the description string (“Transpose  submission”) for your transpose_submit function.  The autograder searches for this string to determine which transpose function to evaluate for credit.

 

 

 

 

 

Programming Rules for Part  B

 

• Include your name and loginID in the header comment for trans.c.

 

• Your code in trans.c must compile without warnings to receive credit.

 

• You are allowed to define at most 12 local variables of type int per transpose function.1

 

• You are not allowed to side-step the previous rule by using any variables of type long or by using any bit tricks to store more than one value to a single variable.

 

• Your transpose function may not use recursion.

 

• If you choose to use helper functions, you may not have more than 12 local variables on the stack at a time between your helper functions and your top level transpose function. For example, if your transpose declares 8 variables, and then you call a function which uses 4 variables, which calls another function which uses 2, you will have 14 variables on the stack, and you will be in violation of the rule.

 

• Your transpose function may not modify array A. You may, however, do whatever you want with the contents of array B.

 

• You are NOT allowed to define any arrays in your code or to use any variant of malloc.

 

 

5    Evaluation

 

 

This section describes how your work will be evaluated. The full score for this lab is 60 points:

 

• Part A: 27 Points

 

• Part B: 26 Points

 

• Style: 7 Points

 

 

5.1   Evaluation for Part  A

 

For Part A, we will run your cache simulator using different cache parameters and traces. There are eight test cases, each worth 3 points, except for the last case, which is worth 6 points:

 

linux ./csim -s 1  -E 1  -b 1  -t traces/yi2.trace linux ./csim -s 4  -E 2  -b 4  -t traces/yi.trace linux ./csim -s 2  -E 1  -b 4  -t traces/dave.trace linux ./csim -s 2  -E 1  -b 3  -t traces/trans.trace linux ./csim -s 2  -E 2  -b 3  -t traces/trans.trace

 

1 The reason for this restriction is that our testing code is not able to count references to the stack. We want you to limit your references to the stack and focus on the access patterns of the source and destination arrays.

 

 

 

 

 

 

linux ./csim -s 2  -E 4  -b 3  -t traces/trans.trace linux ./csim -s 5  -E 1  -b 5  -t traces/trans.trace linux ./csim -s 5  -E 1  -b 5  -t traces/long.trace

 

You can use the reference simulator csim-MRU-ref to obtain the correct answer for each of these test cases. During debugging, use the -v option for a detailed record of each hit and miss.

For each test case, outputting the correct number of cache hits, misses and evictions will give you full credit for that test case.  Each of your reported number of hits, misses and evictions is worth 1/3 of the credit for that test case. That is, if a particular test case is worth 3 points, and your simulator outputs the correct number of hits and misses, but reports the wrong number of evictions, then you will earn 2 points.

 

 

5.2   Evaluation for Part  B

 

For Part B, we will evaluate the correctness and performance of your transpose_submit function on three different-sized output matrices:

 

• 32 × 32 (M = 32, N = 32)

 

• 64 × 64 (M = 64, N = 64)

 

• 61 × 67 (M = 61, N = 67)

 

 

5.2.1   Performance (26 pts)

 

For each matrix size, the performance of your transpose_submit function is evaluated by using valgrind to extract the address trace for your function, and then using the reference simulator to replay this trace on a cache with parameters (s = 5, E = 1, b = 5).

Your performance score for each matrix size scales linearly with the number of misses, m, up to some threshold:

 

 

• 32 × 32: 8 points if m < 300, 0 points if m 600

 

• 64 × 64: 8 points if m < 1, 600, 0 points if m 2, 000

 

• 61 × 67: 10 points if m < 2, 000, 0 points if m 3, 000

 

 

Your code must be correct to receive any performance points for a particular size. Your code only needs to be correct for these three cases and you can optimize it specifically for these three cases. In particular, it is perfectly OK for your function to explicitly check for the input sizes and implement separate code optimized for each case.

 

 

 

 

 

5.3   Evaluation for Style

 

There are 7 points for coding style. These will be assigned manually by the course staff.

 

The course staff will inspect your code in Part B for illegal arrays and excessive local variables. Good code comments are essential in Part-B to explain the design of your transpose submit function.

 

 

6    Working on the Lab

 

 

6.1   Working  on Part  A

 

We have provided you with an autograding program, called test-csim, that tests the correctness of your cache simulator on the reference traces. Be sure to compile your simulator before running the test:

linux  make

linux ./test-csim


 

 

 

Your  simulator        Reference simulator

Points  (s,E,b)  Hits     Misses Evicts  Hits     Misses Evicts 

3          (1,1,1)  9          8          6          9          8          6          traces/yi2.trace

3          (4,2,4)  5          4          1          5          4          1          traces/yi.trace

3          (2,1,4)  2          3          1          2          3          1          traces/dave.trace

3          (2,1,3)  167      71        67        167      71        67        traces/trans.trace

3          (2,2,3)  173      65        57        173      65        57        traces/trans.trace

3          (2,4,3)  197      41        25        197      41        25        traces/trans.trace

3          (5,1,5)  231      7          0          231      7          0          traces/trans.trace

6          (5,1,5)  265189            21775  21743  265189            21775  21743  traces/long.trace

27                                                                                           

TEST_CSIM_RESULTS=27

 

 

 

For each test, it shows the number of points you earned, the cache parameters, the input trace file, and a comparison of the results from your simulator and the reference simulator.

Here are some hints and suggestions for working on Part A:

 

• Do your initial debugging on the small traces, such as traces/dave.trace.

 

• The reference simulator takes an optional -v argument that enables verbose output, displaying the hits, misses, and evictions that occur as a result of each memory access.  You are not required to implement this feature in your csim.c code, but we strongly recommend that you do so.  It will help you debug by allowing you to directly compare the behavior of your simulator with the reference simulator on the reference trace files.

 

• We recommend that you use the getopt function to parse your command line arguments. You’ll need the following header files:

 

 

 

#include <getopt.h

#include <stdlib.h

#include <unistd.h

 

See “man  3  getopt” for details.

 

• Each data load (L) or store (S) operation can cause at most one cache miss. The data modify operation (M) is treated as a load followed by a store to the same address. Thus, an M operation can result in two cache hits, or a miss and a hit plus a possible eviction.

 

• If you would like to use C0-style contracts from 15-122, you can include contracts.h, which we have provided in the handout directory for your convenience.

 

 

6.2   Working  on Part  B

 

We have provided you with an autograding program, called test-trans.c, that tests the correctness and performance of each of the transpose functions that you have registered with the autograder.

You can register up to 100 versions of the transpose function in your trans.c file. Each transpose version has the following form:

 

/*  Header  comment */

char trans_simple_desc[] =  "A  simple transpose";

void trans_simple(int  M,  int N,   int A[N][M], int  B[M][N])

{

/*  your transpose  code here */

}

 

Register a particular transpose function with the autograder by making a call of the form:

 

registerTransFunction(trans_simple,  trans_simple_desc);

 

in the registerFunctions routine in trans.c.  At runtime, the autograder will evaluate each reg- istered transpose function and print the results.  Of course, one of the registered functions must be the transpose_submit function that you are submitting for credit:

 

registerTransFunction(transpose_submit,  transpose_submit_desc);

 

See the default trans.c function for an example of how this works.

 

The autograder takes the matrix size as input. It uses valgrind to generate a trace of each registered trans- pose function. It then evaluates each trace by running the reference simulator on a cache with parameters (s = 5, E = 1, b = 5).

For example, to test your registered transpose functions on a 32 × 32 matrix, rebuild test-trans, and then run it with the appropriate values for M and N :

 

linux make

linux ./test-trans -M  32   -N  32

Step 1: Evaluating registered  transpose funcs for correctness:

func 0  (Transpose submission): correctness:  1

func 1  (Simple row-wise scan transpose): correctness: 1 func 2  (column-wise scan transpose): correctness: 1

func 3  (using a zig-zag access pattern):  correctness: 1

 

Step 2:  Generating memory   traces for registered transpose funcs.

 

Step 3:  Evaluating performance of registered  transpose funcs (s=5, E=1, b=5)

func 0  (Transpose submission): hits:1766,  misses:287, evictions:255

func 1  (Simple row-wise scan transpose): hits:870, misses:1183, evictions:1151 func 2  (column-wise scan transpose): hits:870, misses:1183, evictions:1151

func 3  (using a zig-zag access pattern):  hits:1076, misses:977, evictions:945

 

Summary   for official submission (func 0):  correctness=1 misses=287

 

In this example, we have registered four different transpose functions in trans.c.  The test-trans program tests each of the registered functions, displays the results for each, and extracts the results for the official submission.

Here are some hints and suggestions for working on Part B.

 

• The test-trans program saves the trace for function i in file trace.fi.2   These trace files are invaluable debugging tools that can help you understand exactly where the hits and misses for each transpose function are coming from. To debug a particular function, simply run its trace through the reference simulator with the verbose option:

 

linux ./csim-ref  -v -s 5  -E 1  -b 5  -t trace.f0

S  68312c,1 miss L  683140,8 miss L  683124,4 hit

L  683120,4 hit

L  603124,4 miss eviction

S  6431a0,4 miss

...

 

• Since your transpose function is being evaluated on a direct-mapped cache, conflict misses are a potential problem.  Think about the potential for conflict misses in your code, especially along the diagonal. Try to think of access patterns that will decrease the number of these conflict misses.

 

• Blocking is a useful technique for reducing cache misses. See

 

 

6.3   Putting  it all Together

 

We have provided you with a driver program, called ./driver.py, that performs a complete evaluation of your simulator and transpose code.  This is the same program your instructor uses to evaluate your handins. The driver uses test-csim to evaluate your simulator, and it uses test-trans to evaluate your submitted transpose function on the three matrix sizes. Then it prints a summary of your results and the points you have earned.

To run the driver, type:

 

linux ./driver.py

 

 

7    Handing in Your Work

 

 

Each time you type make  in the cachelab-handout directory, the Makefile creates a tarball, called

userid-handin.tar, that contains your current csim.c and trans.c files.

 

 

Upload your userid-handin.tar file on Moodle. userid is usually your x500 id. IMPORTANT:

• Do not create the handin tarball on a Windows or Mac machine,and do not handin files in any other

archive format, such as .zip, .gzip, or .tgz files.

 

• Please don’t create userid-handin.tar manually.  If your code compiles without errors, the make command should generate the necessary submission file. We expect submission in the format mentioned above. There will be a penalty if the submission is not in required format.