run pthreads from pthread - c

Question

I have a file which contains a list of filenames in which I wanna search for a word and replace it
I modified the code a little just to show only relevant parts here
The problem is that if I have only one file in that list, it won't process it with multi threads because the threads are working only if I have multiple files
So I want to keep the current threads configuration but I wanna add some threads at the processing part
I have this code:
struct words_list {
char word[20];
struct words_list * next;
};
FILE * INFILE;
int num_thread = 10;
// Mutex variables
pthread_mutex_t input_queue;
pthread_mutex_t word_list;
int main(int argc,char **argv)
{
//some code is missing
if((INFILE = fopen(myfile,"r")) == NULL) {
fprintf(stderr,"Can't open input file\n");
exit(0);
}
for(i = 0 ; i < number_thread; i++)
{
if(pthread_create(&thread_id[i],NULL,&search,NULL) != 0)
{
i--;
fprintf(stderr,RED "\nError in creating thread\n" NONE);
}
}
for(i = 0 ; i < number_thread; i++)
if(pthread_join(thread_id[i],NULL) != 0)
{
fprintf(stderr,RED "\nError in joining thread\n" NONE);
}
fflush(INFILE);
fclose(INFILE);
}
void * search(void * data)
{
char file[20];
while (!feof(INFILE))
{
if (fgets(file,sizeof(file),INFILE) != NULL)
{
if (strlen(file) < 8)
break;
if (file[strlen (file) - 1] == '\n')
file[strlen (file) - 1] = '\0';
}
process(file);
}
return NULL;
}
void process(char *filename)
{
char buff[512];
char word[20];
struct words_list * curr_word = first_word;
if(verbose != 0)
fprintf(stderr,"Processing: %s\n",filename);
while(curr_word != NULL)
{
//some code missing
pthread_mutex_lock(&word_list);
strncpy(word,curr_word->word,sizeof(word) - 1);
pthread_mutex_unlock(&word_list);
**//replace_word must run with multiple threads**
ret = replace_word(word,buff,sizeof(buff));
//end of threads part
//code missing
}
}
How can I add other pthreads at the bold part so it can process with multiple threads each file?

Answer 1

Instead of assigning one thread per file. Why shouldn't you allocate a thread to process portion of each file. As shown below, I allocated a thread for every 50 bytes of data in a file. In this technique, even if you have only one file in the list, you can still use multiple threads to parse it. Hope it helps.
#include "stdio.h"
#include "pthread.h"
#include "sys/stat.h"
#include "string.h"
#include "fcntl.h"
#define TOTAL_NUMBER_THREADS 100
struct words_list {
char word[20];
struct words_list * next;
};
struct file_segment {
char filename[50];
size_t foffset;
size_t size;
}fs[TOTAL_NUMBER_THREADS];
FILE * INFILE;
int num_thread=0;
// Mutex variables
pthread_mutex_t input_queue;
pthread_mutex_t word_list;
pthread_t thread_id[TOTAL_NUMBER_THREADS];
void *process( void *arg);
void segment_file(char *filename)
{
int fd;
int offset=0;
struct stat statbuf;
size_t size;
fd = open(filename, O_RDONLY);
if(fd < 0)
{
perror("fopen");
return;
}
fstat(fd, &statbuf);
size=statbuf.st_size;
while((offset < size) && (num_thread <= 100))
{
strncpy(fs[num_thread].filename, filename, sizeof(fs[num_thread].filename));
fs[num_thread].foffset=offset;
fs[num_thread].size=(size>50)?50:size;
offset+=fs[num_thread].size;
if(pthread_create(&thread_id[num_thread],NULL,&process,&fs[num_thread]) != 0)
{
fprintf(stderr,"\nError in creating thread\n");
}
num_thread++;
}
return;
}
void *process( void *arg)
{
char buf[50];
struct file_segment *fs;
char word[20];
//struct words_list * curr_word = first_word;
fs = (struct file_segment *) arg;
FILE *fp;
fp=fopen(fs->filename, "r");
fseek(fp, fs->foffset, SEEK_SET);
fread(buf,1,fs->size,fp);
while(curr_word != NULL)
{
//some code missing
pthread_mutex_lock(&word_list);
strncpy(word,curr_word->word,sizeof(word) - 1);
pthread_mutex_unlock(&word_list);
**//replace_word must run with multiple threads**
ret = replace_word(word,buff,sizeof(buff));
//end of threads part
//code missing
}
//printf("Filename: %s\n Info: %s\n", fs->filename, buf);
printf("%s", buf);
return;
}
int main(int argc,char **argv)
{
//some code is missing
char file[50];
int i;
if((INFILE = fopen("list.txt","r")) == NULL) {
fprintf(stderr,"Can't open input file\n");
return 0; }
while (!feof(INFILE))
{
if (fgets(file,sizeof(file),INFILE) != NULL)
{
if (strlen(file) < 8)
break;
if (file[strlen (file) - 1] == '\n')
file[strlen (file) - 1] = '\0';
}
segment_file(file);
}
for(i = 0 ; i < num_thread; i++)
if(pthread_join(thread_id[i],NULL) != 0)
{
fprintf(stderr,"\nError in joining thread\n");
}
fflush(INFILE);
fclose(INFILE);
}