BÁO cáo bài tập lớn hệ điều HÀNH đề tài PThread and multicore programming

Trường Đại học Bách Khoa Thành phố Hồ Chí Minh Phần 2Phần code • Thread=8 Hình 2.1.5: Real time of dotpodmutex.c with thread=8... Comparison chart between dotprodserial.c and dotprodmute

Trang 1

BÁO CÁO BÀI TẬP LỚN

HỆ ĐIỀU HÀNH

Đề tài : PThread and Multicore programming

GVHD: GV: Nguyễn Quang Hùng

Thực hiện: Nguyễn Lê Minh Bảo

Hà Việt Đức -Đoàn Thảo Nhi - 2014010 Trịnh Tiến Đạt - 2012959

Từ Hoàng Phiếm - 2014112

Trang 2

Mục lục

1 Phần Lý thuyết . 3

2 Phần code . 4

2.1 Mutexes . 4

2.1.1 Review, compile and run the dotprodserial.c program 4

2.1.2 Review, compile and run the dotprodmutex.c program 4

2.1.3 Execute the dotprodmutex.c program with thread are arranged 7

2.2 Condition Variables . 11

2.2.1 Review, compile and run the condvar.c program 11

2.2.2 Review, compile, run the bug4.c program and fix the problem 13

2.3 Figure out and fix the global sum . 14

2.4 Compile and run arrayloops.c . 19

Trang 4

Phần code Phần 2

Mutexes

2.1

2.1.1 Review, compile and run the dotprodserial.c program

The dotprodserial.c program used to calculate the indirection of two vectors with

length =100000 Two vectors is calculated to be shown struct DOTDATA This

pro-gram does by running sequentially on a thread

Output : Result When dotprodserial.c run

Hình 2.1.1: Real time of dotpodserial.c.

2.1.2 Review, compile and run the dotprodmutex.c program

The dotprodmutex.c program used to calculate the direction of two vectors with length=100000 But here is the difference:

• In main() function has initialize with the number of thread is 4 to run parallel processes

• In dotprod() function use pthreadmutexlock và pthreadmutexunlock is one of the main means to synchronize flows and to protect shared data when multiple writing happens

Output : Result when the dotprodmutex.c run

Trang 5

Hình 2.1.2: Real time of dotpodmutex.c

Output : The dotprodmutex.c run with thread = 2, 4, 8, 16, 32 và sum=100000

• Thread=2

Hình 2.1.3: Real time of dotpodmutex.c with thread=2.

• Thread=4

Trang 6

Trường Đại học Bách Khoa Thành phố Hồ Chí Minh Phần 2

Phần code

• Thread=8

• Thread=16

Trang 7

• Thread=32

Comparison chart between dotprodserial.c and dotprodmutex.c

2.1.3 Execute the dotprodmutex.c program with thread are arranged

Explain the code :

• Wanting the thread to be printed in the same self, we need to let pthreadjoin within for the time of creating pthreadcreate Because the thread is created it will run and wait for pthreadjoin to pay the result, then create a new thread new should order Print it will be true

• As for pthreadjoin at another for another round, it only works to wait all the thread returns the result before the main () stops

• If there is no pthreadjoin, when the main () is finished, no matter whether other threads are finished or not, the program is terminated

Code after fixing :

Trang 8

Phần code

#include <pthread.h>

#include <stdio.h>

#include <stdlib.h>

typedef struct

{

double *a;

double *b;

double sum;

int veclen;

} DOTDATA;

/* Define globally accessible variables and a mutex */

#define NUMTHRDS 4

#define VECLEN 100000

DOTDATA dotstr;

pthread_t callThd[NUMTHRDS];

pthread_mutex_t mutexsum;

void *dotprod(void *arg)

{

/* Define and use local variables for convenience */

int i, start, end, len ;

long offset;

double mysum, *x, *y;

offset = (long)arg;

len = dotstr.veclen;

start = offset*len;

end = start + len;

x = dotstr.a;

y = dotstr.b;

/*

Trang 9

Perform the dot product and assign result

to the appropriate variable in the structure

*/

mysum = 0;

for (i=start; i<end ; i++)

{

mysum += (x[i] * y[i]);

}

/*

Lock a mutex prior to updating the value in the shared

structure, and unlock it upon updating

*/

pthread_mutex_lock (&mutexsum);

dotstr.sum += mysum;

printf("Thread %ld did %d to %d: mysum=%f global

sum=%f\n",offset,start,end,mysum,dotstr.sum);

pthread_mutex_unlock (&mutexsum);

pthread_exit((void*) 0);

}

int main (int argc, char *argv[])

{

long i;

double *a, *b;

void *status;

pthread_attr_t attr;

/* Assign storage and initialize values */

a = (double*) malloc (NUMTHRDS*VECLEN*sizeof(double));

b = (double*) malloc (NUMTHRDS*VECLEN*sizeof(double));

for (i=0; i<VECLEN*NUMTHRDS; i++) {

a[i]=1;

b[i]=a[i];

}

Trang 10

Phần code

dotstr.veclen = VECLEN;

dotstr.a = a;

dotstr.b = b;

dotstr.sum=0;

pthread_mutex_init(&mutexsum, NULL);

/* Create threads to perform the dotproduct */

pthread_attr_init(&attr);

pthread_attr_setdetachstate(&attr, PTHREAD_CREATE_JOINABLE);

for(i=0;i<NUMTHRDS;i++)

{

/* Each thread works on a different set of data

* The offset is specified by 'i' The size of

* the data for each thread is indicated by VECLEN

*/

pthread_create(&callThd[i], &attr, dotprod, (void *)i); /* Wait on the other threads- Fix*/

pthread_join(callThd[i], &status);

}

pthread_attr_destroy(&attr);

/* After joining, print out the results and cleanup */

printf ("Sum = %f \n", dotstr.sum);

free (a);

free (b);

pthread_mutex_destroy(&mutexsum);

pthread_exit(NULL);

}

Trang 11

Condition Variables

2.2

2.2.1 Review, compile and run the condvar.c program

Bài code mẫu condvar.c sử dụng Pthread contidion variables Ở dưới hàm main(), chúng ta sẽ tạo ra 3 thread Hai trong số chúng sẽ tăng biến count, trong khi đó, thread còn lại sẽ xem giá trị của count Khi biến count tăng lên một giới hạn cho trước, waiting thread được báo hiệu bởi một trong 2 thread tăng biến count Waiting thread sẽ "awakens" và sau đó sửa đổi count Chương trình sẽ tiếp tục cho đến khi các thread đạt tới giá trị TCOUNT Cuối cùng, chương trình sẽ in ra giá trị cuối cùng của biến count.

OUTPUT:

Trang 12

Phần code

inc_count(): thread 3, count = 1, unlocking mutex

Starting watch_count(): thread 1

watch_count(): thread 1 Count= 2 Going into wait

inc_count(): thread 2, count = 12 Threshold reached Just sent signal

watch_count(): thread 1 Condition signal received Count= 12

watch_count(): thread 1 Updating the value of count

watch_count(): thread 1 count now = 137

watch_count(): thread 1 Unlocking mutex

Main(): Waited and joined with 3 threads Final value of count = 145 Done

Trang 13

2.2.2 Review, compile, run the bug4.c program and fix the problem

1 Run the bug4.c program:

Hình 2.2.1: The result after running the bug4.c program.

2 Run the bug4.c program:

Trang 14

Phần code

Figure out and fix the global sum

2.3

Result bug6.c

#First

thread: 0 starting start=0 end=99999

thread: 2 done Global sum now is=103536

Final Global Sum=562298

#Second

Trang 15

Giải thích: Do biến sum là biến global scope và các threads truy cập vào cùng 1

địa chỉ biến sum nên sẽ xảy ra tình trạng race condition Giải pháp được đưa ra là sử dụng mutex clock

Trang 16

Phần code

#include <pthread.h>

#include <stdio.h>

#include <stdlib.h>

/* Define global data where everyone can see them */

#define NUMTHRDS 8

#define VECLEN 100000

pthread_mutex_t mutexsum;

int *a, *b;

long sum = 0.0;

void *dotprod(void *arg) {

/* Each thread works on a different set of data

* The offset is specified by the arg parameter The size of

* the data for each thread is indicated by VECLEN

*/

int i, start, end, offset, len;

long tid;

long local_sum = 0;

tid = (long)arg;

offset = tid;

len = VECLEN;

start = offset*len;

end = start + len;

/* Perform my section of the dot product */

printf("thread: %ld starting start=%d

end=%d\n",tid,start,end-1);

for (i=start ; i < end ; i++) local_sum += (a[i] *

b[i]);

printf("thread: %ld done Global sum now

is=%li\n",tid,sum);

Trang 17

int main (int argc, char *argv[]) {

long i;

void *status;

pthread_t threads[NUMTHRDS];

pthread_attr_t attr;

/* Assign storage and initialize values */

a = (int*) malloc (NUMTHRDS*VECLEN*sizeof(int));

b = (int*) malloc (NUMTHRDS*VECLEN*sizeof(int));

for (i=0 ; i < VECLEN*NUMTHRDS ; i++) a[i] = b[i] = 1;

/* Initialize mutex variable */

pthread_mutex_init(&mutexsum, NULL);

/* Create threads as joinable, each of which will

execute the dot product

* routine Their offset into the global vectors is specified by passing

* the "i" argument in pthread_create()

*/

pthread_attr_init(&attr);

pthread_attr_setdetachstate(&attr,

PTHREAD_CREATE_JOINABLE);

for (i=0 ; i < NUMTHRDS ; i++)

pthread_create(&threads[i], &attr, dotprod, (void

*)i);

pthread_attr_destroy(&attr);

/* Wait on the other threads for final result */

for (i=0 ; i < NUMTHRDS ; i++)

pthread_join(threads[i], &status);

/* After joining, print out the results and cleanup */

printf ("Final Global Sum=%li\n",sum);

free (a);

free (b);

pthread_mutex_destroy(&mutexsum);

pthread_exit(NULL);

}

Trang 18

Phần code

Result bug6fix.c

#First

Final Global Sum=800000

#Second

Trang 19

Compile and run arrayloops.c

2.4

Arrayloops.c là 1 ví dụ khác của việc dùng mutex để cập nhật biến toàn cục sum

Results:

Thread 0 doing iterations 0 to 249999

Done Sum= 4.999995e+11

Check Sum= 4.999995e+11

Định dạng
Số trang	19
Dung lượng	1,66 MB