top命o是Linux下常用的性能分析工具Q能够实时显C系l中各个q程的资源占用状况，cM于Windows的Q务管理器。下面详l介l它的用方法?
| top - 01:06:48 up 1:22, 1 user, load average: 0.06,
0.60, 0.48 Tasks: 29 total, 1 running, 28 sleeping, 0 stopped, 0 zombie Cpu(s): 0.3% us, 1.0% sy, 0.0% ni, 98.7% id, 0.0% wa, 0.0% hi, 0.0% si Mem: 191272k total, 173656k used, 17616k free, 22052k buffers Swap: 192772k total, 0k used, 192772k free, 123988k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 1379 root 16 0 7976 2456 1980 S 0.7 1.3 0:11.03 sshd 14704 root 16 0 2128 980 796 R 0.7 0.5 0:02.72 top 1 root 16 0 1992 632 544 S 0.0 0.3 0:00.90 init 2 root 34 19 0 0 0 S 0.0 0.0 0:00.00 ksoftirqd/0 3 root RT 0 0 0 0 S 0.0 0.0 0:00.00 watchdog/0 |
l计信息?/h4>
前五行是pȝ整体的统计信息。第一行是d队列信息Q同 uptime 命o的执行结果。其内容如下Q?/p>
| 01:06:48 | 当前旉 |
| up 1:22 | pȝq行旉Q格式ؓ??/td> |
| 1 user | 当前d用户?/td> |
| load average: 0.06, 0.60, 0.48 | pȝ负蝲Q即d队列的^均长度?br> 三个数值分别ؓ 1分钟?分钟?5分钟前到现在的^均倹{?/td> |
W二、三行ؓq程和CPU的信息。当有多个CPUӞq些内容可能会超q两行。内容如下：
| Tasks: 29 total | q程L |
| 1 running | 正在q行的进E数 |
| 28 sleeping | 睡眠的进E数 |
| 0 stopped | 停止的进E数 |
| 0 zombie | 僵尸q程?/td> |
| Cpu(s): 0.3% us | 用户I间占用CPU癑ֈ?/td> |
| 1.0% sy | 内核I间占用CPU癑ֈ?/td> |
| 0.0% ni | 用户q程I间内改变过优先U的q程占用CPU癑ֈ?/td> |
| 98.7% id | I闲CPU癑ֈ?/td> |
| 0.0% wa | {待输入输出的CPU旉癑ֈ?/td> |
| 0.0% hi | |
| 0.0% si |
最后两行ؓ内存信息。内容如下：
| Mem: 191272k total | 物理内存总量 |
| 173656k used | 使用的物理内存总量 |
| 17616k free | I闲内存总量 |
| 22052k buffers | 用作内核~存的内存量 |
| Swap: 192772k total | 交换区总量 |
| 0k used | 使用的交换区总量 |
| 192772k free | I闲交换区总量 |
| 123988k cached | ~冲的交换区总量?br>
内存中的内容被换出到交换区，而后又被换入到内存，但用过的交换区未被覆盖， 该数值即?strong>内容已存在于内存?/strong>的交换区的大?br> 相应的内存再ơ被换出时可不必再对交换区写入?/td> |
q程信息?/h4>
l计信息区域的下ҎCZ各个q程的详l信息。首先来认识一下各列的含义?/p>
| 序号 | 列名 | 含义 |
| a | PID | q程id |
| b | PPID | 父进Eid |
| c | RUSER | Real user name |
| d | UID | q程所有者的用户id |
| e | USER | q程所有者的用户?/td> |
| f | GROUP | q程所有者的l名 |
| g | TTY | 启动q程的终端名。不是从l端启动的进E则昄?? |
| h | PR | 优先U?/td> |
| i | NI | nice倹{负DC高优先U，正DCZ优先U?/td> |
| j | P | 最后用的CPUQ仅在多CPU环境下有意义 |
| k | %CPU | 上次更新到现在的CPU旉占用癑ֈ?/td> |
| l | TIME | q程使用的CPU旉总计Q单位秒 |
| m | TIME+ | q程使用的CPU旉总计Q单?/100U?/td> |
| n | %MEM | q程使用?strong>物理内存癑ֈ?/td> |
| o | VIRT | q程使用的虚拟内存总量Q单位kb。VIRT=SWAP+RES |
| p | SWAP | q程使用的虚拟内存中Q被换出的大，单位kb?/td> |
| q | RES | q程使用的、未被换出的物理内存大小Q单位kb。RES=CODE+DATA |
| r | CODE | 可执行代码占用的物理内存大小Q单位kb |
| s | DATA | 可执行代码以外的部分(数据D??占用?strong>物理? 存大，单位kb |
| t | SHR | ׃n内存大小Q单位kb |
| u | nFLT | 面错误ơ数 |
| v | nDRT | 最后一ơ写入到现在Q被修改q的面数?/td> |
| w | S | q程状态?br>
D=不可中断的睡眠状?br>
R=q行 S=睡眠 T=跟踪/停止 Z=僵尸q程 |
| x | COMMAND | 命o?命o?/td> |
| y | WCHAN | 若该q程在睡眠，则显C睡眠中的系l函数名 |
| z | Flags | d标志Q参?sched.h |
默认情况下仅昄比较重要?PID、USER、PR、NI、VIRT、RES、SHR、S?CPU?MEM、TIME+、COMMAND 列。可以通过下面的快捷键来更ҎC内宏V?/p>
更改昄内容
通过 f 键可以选择昄的内宏V按 f 键之后会昄列的列表Q按 a-z 卛_昄或隐藏对应的列，最后按回R键确定?/p>
?o 键可以改变列的显C顺序。按写?a-z 可以相应的列向右移动，而大写的 A-Z 可以相应的列向左移动。最后按回R键确定?/p>
按大写的 F ?O 键，然后?a-z 可以进E按照相应的列进行排序。而大写的 R 键可以将当前的排序倒{?/p>
命o使用
1Q?工具Q命令）名称
top
2Q工P命oQ作?br>
昄pȝ当前的进E和其他状况Q?
top是一个动态显CE?卛_以通过用户按键来不断刷新当前状?如果在前台执行该命o,它将独占前台,直到用户l止该程序ؓ?
比较准确的说,top命o提供了实时的对系l处理器的状态监?它将昄pȝ中CPU最“敏感”的Q务列?该命令可以按CPU使用.内存使用和执行时?
对Q务进行排序；而且该命令的很多Ҏ都可以通过交互式命令或者在个h定制文g中进行设?
3Q环境设|?br>
在Linux下用?br>
4Q用方?br>
4Q?使用格式
top [-] [d] [p] [q] [c] [C] [S] [s] [n]
4Q?参数说明
d 指定每两ơ屏q信息刷C间的旉间隔。当然用户可以用s交互命o来改变之?br>
p 通过指定监控q程ID来仅仅监控某个进E的状态?br>
q该选项top没有M延迟的进行刷新。如果调用程序有用户权限Q那么top以
可能高的优先q行?br>
S 指定累计模式
s 使top命o在安全模式中q行。这去除交互命令所带来的潜在危险?br>
i 使top不显CZQ何闲|或者僵死进E?br>
c 昄整个命o行而不只是昄命o?br>
4.3其他
下面介绍在top命o执行q程中可以用的一些交互命令。从使用角度来看Q熟l的掌握q些命o比掌握选项q重要一些。这些命令都是单字母的，如果在命
令行选项中用了s选项Q则可能其中一些命令会被屏蔽掉?br>
Ctrl+L 擦除q且重写屏幕?br>
h或? 昄帮助画面Q给Z些简短的命oȝ说明?br>
k
l止一个进E。系l将提示用户输入需要终止的q程PIDQ以及需要发送给该进E什么样的信受一般的l止q程可以使用15信号Q如果不能正常结束那׃?
信号9强制l束该进E。默认值是信号15。在安全模式中此命o被屏蔽?br>
i 忽略闲置和僵死进E。这是一个开兛_命o?br>
q 退出程序?br>
r 重新安排一个进E的优先U别。系l提C用戯入需要改变的q程PID以及需要设|的
q程优先U倹{输入一个正值将使优先降低Q反之则可以使该q程拥有更高的优先权。默认值是10?br>
S 切换到篏计模式?br>
s 改变两次h之间的gq时间。系l将提示用户输入新的旉Q单位ؓs。如果有数Q?
换成m s。输?值则pȝ不断刷斎ͼ默认值是5
s。需要注意的是如果设|太的旉Q很可能会引起不断刷斎ͼ从而根本来不及看清昄的情况，而且pȝ负蝲也会大大增加?br>
f或者F 从当前显CZd或者删除项目?br>
o或者O 改变昄目的顺序?br>
l 切换昄q_负蝲和启动时间信息?br>
m 切换昄内存信息?br>
t 切换昄q程和CPU状态信息?br>
c 切换昄命o名称和完整命令行?br>
M Ҏȝ内存大小q行排序?br>
P ҎCPU使用癑ֈ比大进行排序?br>
T Ҏ旉/累计旉q行排序?br>
W 当前设|写入~/.toprc文g中。这是写top配置文g的推荐方法?/span>
]]>
#include <sys/types.h>
#include <errno.h>
#include <fcntl.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#define FILE_PATH "LockFile.txt"
#define MAXLINE 4096 /* max text line length */
#define FILE_MODE (S_IRUSR | S_IWUSR | S_IRGRP | S_IROTH)
int LockSet(int fd, int Type)
{
struct flock lock;
lock.l_type = Type; /* F_RDLCK, F_WRLCK, F_UNLCK */
lock.l_start = 0; /* byte offset, relative to l_whence */
lock.l_whence = SEEK_END; /* SEEK_SET, SEEK_CUR, SEEK_END */
lock.l_len = 0; /* #bytes (0 means to EOF) */
return fcntl( fd, F_SETLK, &lock);
}
int main(int argc, char **argv)
{
char line[MAXLINE] = {0};
int fd = open(FILE_PATH, O_RDWR | O_CREAT| O_APPEND , FILE_MODE);
if ( LockSet(fd, F_WRLCK) < 0 )
{
if (errno == EACCES || errno == EAGAIN)
{
printf("This program has running!\n");
close(fd);
}
else
{
printf("call fcntl function error.%s\n", strerror(errno) );
}
exit(1);
}
printf("lock file [%s] successful
\n", FILE_PATH);
snprintf(line, sizeof(line), "%ld\n", (long) getpid());
lseek(fd, 0, SEEK_END);
write(fd, line, strlen(line));
getchar();
if ( LockSet(fd, F_UNLCK) < 0 )
{
printf("call fcntl function error.%s\n", strerror(errno) );
}
close(fd);
return 0;
}
http://www.shnenglu.com/Files/bujiwu/RecordLocking.rar#include <errno.h>
#include <fcntl.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#define FILE_PATH "LockFile.txt"
#define MAXLINE 4096 /* max text line length */
#define FILE_MODE (S_IRUSR | S_IWUSR | S_IRGRP | S_IROTH)
int LockSet(int fd, int Type)
{
struct flock lock;
lock.l_type = Type; /* F_RDLCK, F_WRLCK, F_UNLCK */
lock.l_start = 0; /* byte offset, relative to l_whence */
lock.l_whence = SEEK_END; /* SEEK_SET, SEEK_CUR, SEEK_END */
lock.l_len = 0; /* #bytes (0 means to EOF) */
return fcntl( fd, F_SETLK, &lock);
}
int main(int argc, char **argv)
{
char line[MAXLINE] = {0};
int fd = open(FILE_PATH, O_RDWR | O_CREAT| O_APPEND , FILE_MODE);
if ( LockSet(fd, F_WRLCK) < 0 )
{
if (errno == EACCES || errno == EAGAIN)
{
printf("This program has running!\n");
close(fd);
}
else
{
printf("call fcntl function error.%s\n", strerror(errno) );
}
exit(1);
}
printf("lock file [%s] successful
\n", FILE_PATH);snprintf(line, sizeof(line), "%ld\n", (long) getpid());
lseek(fd, 0, SEEK_END);
write(fd, line, strlen(line));
getchar();
if ( LockSet(fd, F_UNLCK) < 0 )
{
printf("call fcntl function error.%s\n", strerror(errno) );
}
close(fd);
return 0;
}
]]>
sem_destroy()
sem_post()
sem_wait()
//g++ -o semaphore semaphore.cpp -lpthread
#include <semaphore.h>
#include <errno.h>
#include <unistd.h>
#include <pthread.h>
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include "ThreadMutex.h"
#define MAXSIZE 10
sem_t g_sem_max;
sem_t g_sem_zero;
CThreadMutex g_mutex;
int g_count = 0;
int sem_wait_i(sem_t* psem)
{
int rv = 0;
while ( 0 != ( rv = sem_wait(psem) ) )
{
if ( errno == EINTR )
{
continue;
}
else
{
printf("sem_wait %s\n", strerror(errno) );
break;
}
}
return rv;
}
void* producer_thread(void* Parameter)
{
while( true )
{
sleep(2);
if( sem_wait_i(&g_sem_max) != 0 )
{
break;
}
g_mutex.Lock();
g_count++;
printf("Producer=%d\n", g_count);
sem_post(&g_sem_zero);
g_mutex.UnLock();
}
return NULL;
}
void* consumer_thread(void* Parameter)
{
while( true )
{
sleep(3);
if( sem_wait_i(&g_sem_zero) != 0 )
{
break;
}
g_mutex.Lock();
g_count--;
printf("Consumer=%d\n", g_count);
sem_post(&g_sem_max);
g_mutex.UnLock();
}
return NULL;
}
int main(int argc, char* argv[])
{
if ( sem_init(&g_sem_zero, 0, 0) != 0 )
{
printf("sem_init %s\n", strerror(errno) );
return -1;
}
if ( sem_init(&g_sem_max, 0, MAXSIZE) != 0 )
{
printf("sem_init %s\n", strerror(errno) );
return -1;
}
pthread_t pro_thread;
pthread_t sum_thread;
if ( pthread_create( &sum_thread, NULL, consumer_thread, NULL) != 0 )
{
printf("pthread_create%s\n", strerror(errno) );
return -1;
}
if ( pthread_create( &pro_thread, NULL, producer_thread, NULL) != 0 )
{
printf("pthread_create%s\n", strerror(errno) );
return -1;
}
if ( pthread_join(pro_thread, NULL) != 0 )
{
printf("pthread_join %s\n", strerror(errno) );
return -1;
}
if ( pthread_join(sum_thread, NULL) != 0 )
{
printf("pthread_join %s\n", strerror(errno) );
return -1;
}
sem_destroy(&g_sem_zero);
sem_destroy(&g_sem_max);
return 0;
}
http://www.shnenglu.com/Files/bujiwu/semaphore.rar#include <semaphore.h>
#include <errno.h>
#include <unistd.h>
#include <pthread.h>
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include "ThreadMutex.h"
#define MAXSIZE 10
sem_t g_sem_max;
sem_t g_sem_zero;
CThreadMutex g_mutex;
int g_count = 0;
int sem_wait_i(sem_t* psem)
{
int rv = 0;
while ( 0 != ( rv = sem_wait(psem) ) )
{
if ( errno == EINTR )
{
continue;
}
else
{
printf("sem_wait %s\n", strerror(errno) );
break;
}
}
return rv;
}
void* producer_thread(void* Parameter)
{
while( true )
{
sleep(2);
if( sem_wait_i(&g_sem_max) != 0 )
{
break;
}
g_mutex.Lock();
g_count++;
printf("Producer=%d\n", g_count);
sem_post(&g_sem_zero);
g_mutex.UnLock();
}
return NULL;
}
void* consumer_thread(void* Parameter)
{
while( true )
{
sleep(3);
if( sem_wait_i(&g_sem_zero) != 0 )
{
break;
}
g_mutex.Lock();
g_count--;
printf("Consumer=%d\n", g_count);
sem_post(&g_sem_max);
g_mutex.UnLock();
}
return NULL;
}
int main(int argc, char* argv[])
{
if ( sem_init(&g_sem_zero, 0, 0) != 0 )
{
printf("sem_init %s\n", strerror(errno) );
return -1;
}
if ( sem_init(&g_sem_max, 0, MAXSIZE) != 0 )
{
printf("sem_init %s\n", strerror(errno) );
return -1;
}
pthread_t pro_thread;
pthread_t sum_thread;
if ( pthread_create( &sum_thread, NULL, consumer_thread, NULL) != 0 )
{
printf("pthread_create%s\n", strerror(errno) );
return -1;
}
if ( pthread_create( &pro_thread, NULL, producer_thread, NULL) != 0 )
{
printf("pthread_create%s\n", strerror(errno) );
return -1;
}
if ( pthread_join(pro_thread, NULL) != 0 )
{
printf("pthread_join %s\n", strerror(errno) );
return -1;
}
if ( pthread_join(sum_thread, NULL) != 0 )
{
printf("pthread_join %s\n", strerror(errno) );
return -1;
}
sem_destroy(&g_sem_zero);
sem_destroy(&g_sem_max);
return 0;
}
]]>
struct itimerval *ovalue);
DESCRIPTION
The system provides each process with three interval timers, each decrementing in a distinct time domain. When any timer expires, a
signal is sent to the process, and the timer (potentially) restarts.
ITIMER_REAL decrements in real time, and delivers SIGALRM upon expiration.
ITIMER_VIRTUAL decrements only when the process is executing, and delivers SIGVTALRM upon expiration.
ITIMER_PROF decrements both when the process executes and when the system is executing on behalf of the process. Coupled with
ITIMER_VIRTUAL, this timer is usually used to profile the time spent by the application in user and kernel space. SIG-
PROF is delivered upon expiration.
Timer values are defined by the following structures:
struct itimerval {
struct timeval it_interval; /* next value */
struct timeval it_value; /* current value */
};
struct timeval {
long tv_sec; /* seconds */
long tv_usec; /* microseconds */
};
#include <stdio.h>
#include <unistd.h>
#include <signal.h>
#include <string.h>
#include <sys/time.h>
#include <errno.h>
void PrintMsg(int Num)
{
printf("%s\n", "Hello World");
return;
}
int main(int argc, char* argv[])
{
signal(SIGALRM, PrintMsg);
struct itimerval tick;
tick.it_value.tv_sec = 10; //十秒钟后启动定时器
tick.it_value.tv_usec = 0;
tick.it_interval.tv_sec =1; //定时器启动后Q每?U将执行相应的函?/span>
tick.it_interval.tv_usec = 0;
//setitimer触发SIGALRM信号
int ret = setitimer(ITIMER_REAL, &tick, NULL);
if ( ret != 0)
{
printf("Set timer error. %s \n", strerror(errno) );
return -1;
}
printf("Wait!\n");
getchar();
return 0;
}
#include <unistd.h>
#include <signal.h>
#include <string.h>
#include <sys/time.h>
#include <errno.h>
void PrintMsg(int Num)
{
printf("%s\n", "Hello World");
return;
}
int main(int argc, char* argv[])
{
signal(SIGALRM, PrintMsg);
struct itimerval tick;
tick.it_value.tv_sec = 10; //十秒钟后启动定时器
tick.it_value.tv_usec = 0;
tick.it_interval.tv_sec =1; //定时器启动后Q每?U将执行相应的函?/span>
tick.it_interval.tv_usec = 0;
//setitimer触发SIGALRM信号
int ret = setitimer(ITIMER_REAL, &tick, NULL);
if ( ret != 0)
{
printf("Set timer error. %s \n", strerror(errno) );
return -1;
}
printf("Wait!\n");
getchar();
return 0;
}
]]>
//Mutex.cpp
#include <pthread.h>
#include <iostream>
using namespace std;
class ThreadMutex
{
public:
ThreadMutex()
{
pthread_mutex_init(&mtx, NULL);
}
~ThreadMutex()
{
pthread_mutex_destroy( &mtx );
}
inline void Lock()
{
pthread_mutex_lock( &mtx );
}
inline void UnLock()
{
pthread_mutex_unlock( &mtx );
}
private:
pthread_mutex_t mtx;
};
//以下为测试用?br>ThreadMutex g_Mutex;
void *PrintMsg(void *lpPara)
{
char *msg = (char *)lpPara;
g_Mutex.Lock();
for(int i=0; i< 5; i++ )
{
cout << msg << endl;
sleep( 1 );
}
g_Mutex.UnLock();
return NULL;
}
int main()
{
pthread_t t1,t2;
//创徏两个工作U程Q第1个线E打?0?Q第2个线E打?0??/span>
pthread_create( &t1, NULL, &PrintMsg, (void *)"First print thread" );
pthread_create( &t2, NULL, &PrintMsg, (void *)"Second print thread" );
//{待U程l束
pthread_join( t1, NULL);
pthread_join( t2, NULL);
return 0;
}
#include <pthread.h>
#include <iostream>
using namespace std;
class ThreadMutex
{
public:
ThreadMutex()
{
pthread_mutex_init(&mtx, NULL);
}
~ThreadMutex()
{
pthread_mutex_destroy( &mtx );
}
inline void Lock()
{
pthread_mutex_lock( &mtx );
}
inline void UnLock()
{
pthread_mutex_unlock( &mtx );
}
private:
pthread_mutex_t mtx;
};
//以下为测试用?br>ThreadMutex g_Mutex;
void *PrintMsg(void *lpPara)
{
char *msg = (char *)lpPara;
g_Mutex.Lock();
for(int i=0; i< 5; i++ )
{
cout << msg << endl;
sleep( 1 );
}
g_Mutex.UnLock();
return NULL;
}
int main()
{
pthread_t t1,t2;
//创徏两个工作U程Q第1个线E打?0?Q第2个线E打?0??/span>
pthread_create( &t1, NULL, &PrintMsg, (void *)"First print thread" );
pthread_create( &t2, NULL, &PrintMsg, (void *)"Second print thread" );
//{待U程l束
pthread_join( t1, NULL);
pthread_join( t2, NULL);
return 0;
}
通过g++ -o Mutex Mutex.cpp -lpthread~译?br>
]]>
#include <sys/ipc.h>
#include <sys/shm.h>
#include <unistd.h>
#include <errno.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <errno.h>
#include <sys/types.h>
using namespace std;
int main(int argc, char* argv[])
{
key_t key = ftok("/etc/passwd", 'a');
if( -1 == key )
{
fprintf(stderr, "Creat Key Error:%s\n", strerror(errno));
return -1;
}
/*
if your IPC are used by processes related bya fork()
(parent-children-grandchildren) use IPC_PRIVATE as the key
*/
int shm_id=shmget( key, 1024, IPC_CREAT |0660);
if( shm_id == -1 )
{
if ( shmctl(shm_id, IPC_RMID, NULL) == -1 )
{
fprintf(stderr, "shmctl remove shmid error:%s\n", strerror(errno ));
return -1;
}
shm_id=shmget( key, 1024, IPC_CREAT |0660);
if( shm_id == -1 )
{
fprintf(stderr, "shmget create error:%s\n", strerror(errno) );
return -1;
}
}
char* pMap = (char *)shmat(shm_id, NULL, 0);
if( (int)pMap == -1)
{
fprintf(stderr, "shmat create error:%s\n", strerror(errno) );
shmctl(shm_id, IPC_RMID, NULL);
return -1;
}
strcpy( pMap, "Bujiwu Swallow");
getchar();
shmdt(pMap);
shmctl(shm_id, IPC_RMID, NULL);
return 0;
}
#include <sys/shm.h>
#include <unistd.h>
#include <errno.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <errno.h>
#include <sys/types.h>
using namespace std;
int main(int argc, char* argv[])
{
key_t key = ftok("/etc/passwd", 'a');
if( -1 == key )
{
fprintf(stderr, "Creat Key Error:%s\n", strerror(errno));
return -1;
}
/*
if your IPC are used by processes related bya fork()
(parent-children-grandchildren
) use IPC_PRIVATE as the key*/
int shm_id=shmget( key, 1024, IPC_CREAT |0660);
if( shm_id == -1 )
{
if ( shmctl(shm_id, IPC_RMID, NULL) == -1 )
{
fprintf(stderr, "shmctl remove shmid error:%s\n", strerror(errno ));
return -1;
}
shm_id=shmget( key, 1024, IPC_CREAT |0660);
if( shm_id == -1 )
{
fprintf(stderr, "shmget create error:%s\n", strerror(errno) );
return -1;
}
}
char* pMap = (char *)shmat(shm_id, NULL, 0);
if( (int)pMap == -1)
{
fprintf(stderr, "shmat create error:%s\n", strerror(errno) );
shmctl(shm_id, IPC_RMID, NULL);
return -1;
}
strcpy( pMap, "Bujiwu Swallow");
getchar();
shmdt(pMap);
shmctl(shm_id, IPC_RMID, NULL);
return 0;
}
#include <sys/ipc.h>
#include <sys/shm.h>
#include <unistd.h>
#include <errno.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <errno.h>
#include <sys/types.h>
using namespace std;
int main(int argc, char* argv[])
{
key_t key = ftok("/etc/passwd", 'a');
if( -1 == key )
{
fprintf(stderr, "Creat Key Error:%s\n", strerror(errno));
return -1;
}
/*
if your IPC are used by processes related bya fork()
(parent-children-grandchildren) use IPC_PRIVATE as the key
*/
int shm_id=shmget( key, 1024, IPC_CREAT|0660);
if( shm_id == -1 )
{
if ( shmctl(shm_id, IPC_RMID, NULL) == -1 )
{
fprintf(stderr, "shmctl remove shmid error:%s\n", strerror(errno ));
return -1;
}
shm_id=shmget( key, 1024, IPC_CREAT |0660);
if( shm_id == -1 )
{
fprintf(stderr, "shmget create error:%s\n", strerror(errno) );
return -1;
}
}
char* pMap = (char *)shmat(shm_id, NULL, 0);
if( (int)pMap == -1)
{
fprintf(stderr, "shmat create error:%s\n", strerror(errno) );
shmctl(shm_id, IPC_RMID, NULL);
return -1;
}
printf("Shared memory::%s\n", pMap);
shmdt(pMap);
shmctl(shm_id, IPC_RMID, NULL);
return 0;
}
#include <sys/shm.h>
#include <unistd.h>
#include <errno.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <errno.h>
#include <sys/types.h>
using namespace std;
int main(int argc, char* argv[])
{
key_t key = ftok("/etc/passwd", 'a');
if( -1 == key )
{
fprintf(stderr, "Creat Key Error:%s\n", strerror(errno));
return -1;
}
/*
if your IPC are used by processes related bya fork()
(parent-children-grandchildren
) use IPC_PRIVATE as the key*/
int shm_id=shmget( key, 1024, IPC_CREAT|0660);
if( shm_id == -1 )
{
if ( shmctl(shm_id, IPC_RMID, NULL) == -1 )
{
fprintf(stderr, "shmctl remove shmid error:%s\n", strerror(errno ));
return -1;
}
shm_id=shmget( key, 1024, IPC_CREAT |0660);
if( shm_id == -1 )
{
fprintf(stderr, "shmget create error:%s\n", strerror(errno) );
return -1;
}
}
char* pMap = (char *)shmat(shm_id, NULL, 0);
if( (int)pMap == -1)
{
fprintf(stderr, "shmat create error:%s\n", strerror(errno) );
shmctl(shm_id, IPC_RMID, NULL);
return -1;
}
printf("Shared memory::%s\n", pMap);
shmdt(pMap);
shmctl(shm_id, IPC_RMID, NULL);
return 0;
}
http://www.shnenglu.com/Files/bujiwu/ShareMemory.rar
]]>
]]>
1、什么是Core Dump?
Core的意思是内存, Dump的意思是扔出? 堆出?
开发和使用UnixE序? 有时E序莫名其妙的down? 却没有Q何的提示(有时候会提示core dumped). q时候可以查看一下有没有形如core.q程L文g生成, q个文g便是操作pȝ把程序down掉时的内存内Ҏ出来生成? 它可以做试程序的参?
core dump又叫核心转储, 当程序运行过E中发生异常, E序异常退出时, 由操作系l把E序当前的内存状况存储在一个core文g? 叫core dump.
2、如何用core文g?
gdb -c core文g路径 [应用E序的\径]
q去后输入where回R, 可以显C程序在哪一行当掉的, 在哪个函C.
3、ؓ什么没有core文g生成?
有时候程序down? 但是core文g却没有生? core文g的生成跟你当前系l的环境讄有关p? 可以用下面的语句讄一? 然后再运行程序便成生成core文g.
ulimit -c unlimited
core文g生成的位|一般于q行E序的\径相? 文g名一般ؓcore.q程?br>
4、用gdb查看core文g:
下面我们可以在发生运行时信号引v的错误时发生core dump?
发生core dump之后, 用gdbq行查看core文g的内? 以定位文件中引发core dump的行.
gdb [exec file] [core file]
?
gdb ./test test.core
在进入gdb? 用bt命o查看backtrace以检查发生程序运行到哪里, 来定位core dump的文?>?
]]>
{
pid_t pid;
pid_t sid;
// fork to leave parent
pid = fork ();
if (pid < 0)
{
std::cout << "failed to fork!" << std::endl;
exit (EXIT_FAILURE);
}
else if (pid > 0)
{
//parent process go to exit
exit (EXIT_SUCCESS);
}
//child process
sid = setsid ();
if (sid < 0)
{
std::cout << "Failed to set session id." << std::endl;
exit (EXIT_FAILURE);
}
return;
}
]]>
2 #include <sys/types.h>
3 #include <stdio.h>
4 #include <string.h>
5 #include <errno.h>
6
7 int main(int argc ,char* argv[])
8 {
9 pid_t my_pid = getpid();
10 pid_t parent_pid = getppid();
11
12 printf("ProcessID =%ld ParentID =%ld\n", my_pid, parent_pid);
13
14 pid_t child = fork();
15
16 printf("-------------------------------------\n");
17 printf("Fork Process ID = %ld \n", child );
18
19 switch( child )
20 {
21 case -1:
22 printf("Create my child process error by fork func, errno=%ld\n", strerror(errno) );
23 break;
24
25 case 0:
26 printf("This is child process! Process ID = %ld\n", getpid() );
27 break;
28
29 default:
30 printf("This is not child process! Process ID= %ld Child Process ID=%ld\n", getpid(), child);
31
32 }
33
34 printf("who am I? Process ID=%ld Parent Process ID= %ld\n", getpid(), getppid());
35
36 return 0;
37 }
38
http://www.shnenglu.com/Files/bujiwu/fork.rar
]]>
#在PATH中找到可执行文gE序的\径?br>export PATH =$PATH:$HOME/bin
#gcc扑ֈ头文件的路径
C_INCLUDE_PATH=/usr/include/libxml2:/MyLib
export C_INCLUDE_PATH
#g++扑ֈ头文件的路径
CPLUS_INCLUDE_PATH=$CPLUS_INCLUDE_PATH:/usr/include/libxml2:/MyLib
export CPLUS_INCLUDE_PATH
#扑ֈ动态链接库的\?br>LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/MyLib
export LD_LIBRARY_PATH
#扑ֈ静态库的\?br>LIBRARY_PATH=$LIBRARY_PATH:/MyLib
export LIBRARY_PATH
]]>
2 #include <pthread.h> /* pthread functions and data structures */
3 #include <unistd.h> /* head file for sleep function */
4
5 /* global mutex for our program. assignment initializes it */
6 pthread_mutex_t a_mutex = PTHREAD_MUTEX_INITIALIZER;
7
8 /* function to be executed by the new thread */
9 void* do_loop(void* data)
10 {
11 int me = *((int*)data); /* thread identifying number */
12 int rc; /* contain mutex lock/unlock results */
13
14 /* lock the mutex, to assure exclusive */
15 rc = pthread_mutex_lock(&a_mutex);
16
17 for (int i=0; i< 4; i++)
18 {
19 printf("'%d' - Got '%d'\n", me, i);
20 sleep(1);
21 }
22
23 /* unlock mutex */
24 rc = pthread_mutex_unlock(&a_mutex);
25
26 /* terminate the thread */
27 pthread_exit(NULL);
28 }
29
30 /* like any CPlusPlus program, program's execution begins in main */
31 /*g++ -o Thread Thread.cpp -lpthread*/
32 int main(int argc, char* argv[])
33 {
34 int thr_id; /* thread ID for the newly created thread */
35 pthread_t p_thread1; /* thread's structure */
36 pthread_t p_thread2; /* thread's structure */
37 int a = 1; /* thread 1 identifying number */
38 int b = 2; /* thread 2 identifying number */
39
40 /* create a new thread that will execute 'do_loop()' */
41 thr_id = pthread_create(&p_thread1, NULL, do_loop, (void*)&a);
42
43 /* create a new thread */
44 thr_id = pthread_create(&p_thread2, NULL, do_loop, (void*)&b);
45
46 /*wait until sub thread exit*/
47 pthread_join(p_thread1, NULL);
48 pthread_join(p_thread2, NULL);
49
50 return 0;
51 }
52
~译Q?nbsp;gcc -o pthread pthread.c -lpthread
]]>
打解?.tar.gz)
tar -cvzf firefox-3.0.3.tar.gz firefox #firefox文gҎ包ؓfirefox-3.0.3.tar.gz
tar -xvzf firefox-3.0.3.tar.gz
打解?.tar.bz2)
tar -cvjf firefox-3.0.3.tar.bz2 firefox/ #firefox文gҎ包ؓfirefox-3.0.3.tar.bz2
tar -xvjf firefox-3.0.3.tar.bz2
参数说明Q?br>c 创徏新的档案文g。如果用h备䆾一个目录或是一些文Ӟp选择q个选项?br>x 从档案文件中释放文g?br>v 详细报告tar处理的文件信息。如无此选项Qtar不报告文件信息?br>z 用gzip来压~?解压~文Ӟ加上该选项后可以将档案文gq行压羃Q但q原时也一定要使用该选项q行解压~?
f 使用档案文g或设备，q个选项通常是必选的?br>
]]>
不需要手工添加各个文件的依赖关系Q用vpath自动搜烦文gQ?br> 试工程Q?br> http://www.shnenglu.com/Files/bujiwu/Linux.rar
]]>