搭建环境
这里编写的是用户层面的 util, 直接调用内核态的系统调用 sleep 即可.
#include "kernel/types.h"
#include "kernel/stat.h"
#include "user/user.h"
int
main(int argc, char* argv[])
{
if(argc < 2){
fprintf(2, "Usage: sleep number of ticks...\n");
exit(1);
}
sleep(atoi(argv[1]));
exit(0);
}使用两个管道进行父子进程通信, 需要注意的是如果管道的写端没有close, 那么管道中数据为空时对管道的读取将会阻塞. 因此对于不需要的管道描述符, 要尽可能早的关闭.
#include "kernel/types.h"
#include "kernel/stat.h"
#include "user/user.h"
int
main(int argc, char* argv[])
{
int fd[] = {3, 4};
pipe(fd);
write(fd[1], "ping", 4);
int pid = fork();
int status;
if (pid == 0) {
// child
int pid = getpid();
char word1[10];
read(fd[0], word1, 4);
printf("%d: received %s\n", pid, word1);
write(fd[1], "pong", 4);
exit(0);
} else {
// parent
int pid = getpid();
wait(&status);
char word2[10];
if (status == 0) {
read(fd[0], word2, 4);
printf("%d: received %s\n", pid, word2);
} else {
exit(-1);
}
}
exit(0);
}primes 要求用并行的方式实现埃式筛. 可以先理解一下 C++ 版本的埃式筛实现思路 ，再对照着实现.
Bell Labs and CSP Threads 中指出了并行实现埃式筛的伪代码:
p = get a number from left neighbor
print p
loop:
n = get a number from left neighbor
if (p does not divide n)
send n to right neighbor
一个生成过程可以将数字2、3、4、...、1000输入到管道的左端: 管道中的第一个过程消除2的倍数, 第二个过程消除3的倍数, 第三个过程消除5的倍数, 依此类推. 如下图所示:
#include "kernel/types.h"
#include "user/user.h"
#include "kernel/stat.h"
void sive(int pL[]) {
// pL is left neighbor
int pR[2];
pipe(pR);
close(pL[1]);
int prime, n, pid;
if (read(pL[0], &prime, sizeof(int)) == 0) {
close(pL[0]);
exit(0);
}
printf("prime %d\n", prime);
pid = fork();
if (pid > 0) {
close(pR[0]);
while (read(pL[0], &n, sizeof(int))) {
if (n % prime) write(pR[1], &n, sizeof(int));
}
close(pL[0]);
close(pR[1]);
wait((int*)0);
exit(0);
} else if (pid == 0) {
sive(pR);
close(pL[0]);
}
}
int main(int argc, char* argv[]) {
int pf[2];
pipe(pf);
int prime, n, pid;
pid = fork();
prime = 2;
printf("prime: %d\n", prime);
if (pid > 0) {
close(pf[0]);
for (n = 2; n <= 35; n ++ ) {
if (n % prime)
write(pf[1], &n, sizeof(int)); // 未筛掉的数由创建的子进程继续筛
}
close(pf[1]);
wait((int*)0); // 父进程要等待子进程结束
} else if (pid == 0) {
sive(pf);
}
exit(0);
}仿照 user/ls.c 来编写如何读取文件夹或者文件.
#include "kernel/types.h"
#include "kernel/stat.h"
#include "user/user.h"
#include "kernel/fs.h"
void my_print(char* str){ // 辅助函数1
printf("%s\n", str);
}
char*
fmtname(char* path){ // 辅助函数2 - 规范化文件名
static char buf[DIRSIZ + 1];
char* ptr;
// Find first character after last slash
for(ptr=path+strlen(path); ptr >= path && *ptr != '/'; ptr--)
;
ptr++;
// Return blank-padded name.
if(strlen(ptr) >= DIRSIZ)
return ptr;
memmove(buf, ptr, strlen(ptr));
memset(buf+strlen(ptr), ' ', DIRSIZ-strlen(ptr));
return buf;
}
char* strcat(char* dest, const char* src){ // 辅助函数3 - 拼接字符串
char* ptr = dest;
while(*ptr != '\0') ptr++;
while(*src != '\0'){
*ptr = *src;
ptr++;
src++;
}
*ptr = '\0';
return dest;
}
void removeEndSpace(char* str){ // 辅助函数4 - 去除末尾空格
int cnt = 0;
for (int i = 0; str[i]; i ++ ){
if (str[i] != ' ') str[cnt ++ ] = str[i];
}
str[cnt] = '\0';
}
void find(char* path, char* file){
char buf[512], *p;
int fd;
struct dirent de;
struct stat st;
if((fd = open(path, 0)) < 0){
fprintf(2, "find: cannot open %s\n", path);
return;
}
if(fstat(fd, &st) < 0){
fprintf(2, "find: cannot stat %s\n", path);
close(fd);
return;
}
// st.type 一定是文件夹
if(st.type == T_FILE){
fprintf(2, "find: <path> <file>");
close(fd);
return;
}
if(strlen(path) + 1 + DIRSIZ + 1 > sizeof(buf)){
printf("find: path too long\n");
close(fd);
return;
}
strcpy(buf, path);
p = buf + strlen(buf);
*p++ = '/';
while(read(fd, &de, sizeof(de)) == sizeof(de)){
if(de.inum == 0) continue;
memmove(p, de.name, DIRSIZ);
p[DIRSIZ] = 0;
char* filename = fmtname(buf);
removeEndSpace(filename);
if(stat(buf, &st) < 0){
printf("find: cannot stat %s\n", buf);
continue;
}
if(st.type == T_FILE){ // 是文件，判断是否是dest_file
if(strcmp(filename, file) == 0){ // 是
// printf("current file(dir) is %s %d %d\n", filename, strcmp(filename, "."), strlen(filename));
printf("%s\n", buf);
// close(fd);
// exit(0); // success
}else { // 不是，继续寻找
continue;
}
}else if(st.type == T_DIR){
if (*filename == '.') continue;
find(buf, file);
}
}
close(fd);
}
int
main(int argc, char* argv[]){
if(argc < 2){
fprintf(2, "find: find search_path dest_file...\n");
exit(1);
}
find(argv[1], argv[2]);
// TODO
// in xargs, there is a bug in find.c: when type "sh < xargstest.sh", the output is
// $ $ $ $ $ $ hello
// hello
// hello
// $ $
exit(0);
}见代码注释.
#include "kernel/types.h"
#include "kernel/stat.h"
#include "user/user.h"
#include "kernel/param.h"
#include <stddef.h>
int main(int argc, char* argv[]) {
char stdin_buf[512];
read(0, stdin_buf, sizeof stdin_buf);
// debug: output the argv
// for (int i = 0; i < argc; i ++ ) printf("argv[%d] is %s, ", i, argv[i]);
// printf("\n");
// debug: output the buf
// printf("the content of buf is %s\n", stdin_buf);
char* arguments[MAXARG];
int count = 0;
char word[512];
for (int i = 0, j = 0; stdin_buf[i] != '\0'; i ++ , j ++ ) {
if (stdin_buf[i] == ' ' || stdin_buf[i] == '\n') {
word[j] = '\0';
arguments[count] = malloc(strlen(word) + 1);
strcpy(arguments[count], word);
j = -1;
count ++ ;
} else {
word[j] = stdin_buf[i];
}
}
// debug: output the arguments
// for (int i = 0; i < count; i ++ ) printf("arguments[%d] is %s, length is %d\n", i, arguments[i], strlen(arguments[i]));
// get the latter length of the argv
int len = 0;
for (int i = 1; argv[i] != NULL; i ++ ) len ++ ;
// debug: test the output len
// printf("%d\n", len);
// get the command
char* command = "";
strcpy(command, argv[1]);
// debug: test output command
// printf("command is %s\n", command);
// combine the former argv and the latter argv
char* new_arg[count + len];
for (int i = 0; i < len; i ++ ) {
new_arg[i] = malloc(strlen(argv[i + 1] + 1));
if (new_arg[i] == NULL) {
printf("failed to allocate memory\n");
exit(1);
}
strcpy(new_arg[i], argv[i + 1]);
}
for (int i = 0; i < count; i ++ ) {
new_arg[len + i] = malloc(strlen(arguments[i] + 1));
if (new_arg[len + i] == NULL) {
printf("failed to allocate memory\n");
exit(1);
}
strcpy(new_arg[len + i], arguments[i]);
}
new_arg[len + count] = 0; // set end flag 0 for the new_arg
// debug: test output new_arg
// for (int i = 0; new_arg[i] != NULL; i ++ ) printf("new_arg[%d] is %s\n", i, new_arg[i]);
// free memory
for (int i = 0; i < count; i ++ ) free(arguments[i]);
for (int i = 0; i < count + len; i ++ ) free(new_arg[i]);
// exec
int pid, status;
pid = fork();
if (pid == 0) { // child
exec(command, new_arg);
printf("exec failed\n");
exit(1);
} else {
// printf("parent waiting\n");
wait(&status);
// printf("the child exited with status %d\n", status);
}
// TODO
// I don't know why when I type the "echo 1 2 | xargs echo bye", the stdout isn't correct
exit(0);
}