Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Linux

1,684 views

Published on

Published in: Education, Technology
  • Be the first to comment

  • Be the first to like this

Linux

  1. 1. Linux<br />12 January 2010<br />Knowx innovation<br />1<br />
  2. 2. Unix<br />Developed at Bell Labs in late 1960s<br />Owned by AT&T at the time, which meant you had to buy it once AT&T figured out it was valuable<br />Written in C (a high level language)<br />Needs to be compiled to run<br />Can be easily moved to a different system by re-compiling for new hardware<br />12 January 2010<br />Knowx innovation<br />2<br />
  3. 3. Why Linux<br />Entire OS source code is free<br />Full freedom to study, modify, redistribute. No payments or restrictions.<br />Kernel and many packages follow GNU GPL, LGPL or BSD style license.<br />Powerful command line (shells)<br />Multitasking, SMP, NUMA<br />12 January 2010<br />Knowx innovation<br />3<br />
  4. 4. Protected memory<br />True multiuser capabilities<br />Filesystem choices:ext3, JFS, XFS, reiserFS<br />Highly modular & scalable: From embedded systems to IBM mainframes <br />12 January 2010<br />Knowx innovation<br />4<br />
  5. 5. distribution<br /><ul><li>Debian (related: Knoppix, (k)ubuntu, Mepis, Xandros)
  6. 6. RedHat (related: Fedora, CentOS)
  7. 7. SUSE (related: OpenSUSE)
  8. 8. Mandriva
  9. 9. Slackware
  10. 10. Embedded Linuxes</li></ul>12 January 2010<br />Knowx innovation<br />5<br />
  11. 11. GNU/Linux architecture<br />12 January 2010<br />Knowx innovation<br />6<br />
  12. 12. Linux kernel<br />Mediates access to the hardware and peripherals<br />Shells <br />Provides user access to the kernel<br />Applications <br />Provide the useful function for the operating system<br />12 January 2010<br />Knowx innovation<br />7<br />
  13. 13. Kernel architecture<br />12 January 2010<br />Knowx innovation<br />8<br />
  14. 14. Gnu compiler collection(gcc)<br />Compiler and set of utilities to build binaries from high level source code.<br />Standard for embedded systems development<br />Since its supports so many different target architectures<br />Supports a number of languages<br />C, C++, ada, java, FORTRAN, Pascal<br />12 January 2010<br />Knowx innovation<br />9<br />
  15. 15. Stage of compilation<br />12 January 2010<br />Knowx innovation<br />10<br />
  16. 16. 12 January 2010<br />Knowx innovation<br />11<br />
  17. 17. makefile<br />12 January 2010<br />Knowx innovation<br />12<br />
  18. 18. make utility uses a developer-created input file to describe the project built<br />GNU uses the name Makefile as the default name for its input file.<br />12 January 2010<br />Knowx innovation<br />13<br />
  19. 19. appexp : main.o app.o bar.o lib.o<br /> gcc –o appexp main.o app.o bar.o lib.o<br />main.o : main.c lib.h app.h<br /> gcc –c -o main.o main.c<br />app.o: app.c lib.h app.h<br /> gcc –c -o app.o app.c<br />bar.o : bar.c lib.h<br /> gcc –c -o bar.o bar.c<br />lib.o : lib.c lib.h<br /> gcc –c -o lib.o lib.c<br />Line 1: is the rule<br />The portion of the rule before the colon is called targetand after the colon is called dependencies.<br />12 January 2010<br />Knowx innovation<br />14<br />
  20. 20. File handling<br />Accomplished through the standard C library<br />We can create and manipulate ASCII text or binary files with the same API.<br />API<br />fopen, fclose, fwrite , fread, fseek, and rewind.<br />12 January 2010<br />Knowx innovation<br />15<br />
  21. 21. fopen <br />Opening a file can also be the mechanism to create a file<br />prototype:<br />FILE * fopen(const char *filename,const char *mode)<br />filename - file we wish to access or create<br />mode – mode we wish to use<br />FILE * - fopen returns the FILE pointer (FILE *) <br />12 January 2010<br />Knowx innovation<br />16<br />
  22. 22. Example<br />FILE *fin; <br />fin = fopen(&quot;inpfile.txt&quot;, &quot;r&quot;); <br />12 January 2010<br />Knowx innovation<br />17<br />
  23. 23. Program<br />#include &lt;stdio.h&gt; <br />#define MYFILE &quot;missing.txt&quot; <br />main() <br />{<br />FILE *fin; <br /> fin = fopen( MYFILE, &quot;r&quot; ); /* Try to open the file for read */ <br /> if (fin == (FILE *)0) /* Check for failure to open */<br /> {<br />printf(“error in opening file”); /* Emit an error message and exit */<br />exit(-1);<br />} <br /> fclose( fin ); /* All was well, close the file */ <br />}<br />12 January 2010<br />Knowx innovation<br />18<br />
  24. 24. fread() and fwrite()<br />Prototypes:<br />size_t fread(void *ptr, size_t size, size_t nmemb,FILE *stream);<br />size_t fwrite(const void *ptr, size_t size, size_t nmemb,FILE *stream);<br />fseek()<br />Prototype:<br />int fseek (FILE * stream, long offset, int whence);<br />Function allows to the new position given an index<br />whence argument defines <br />12 January 2010<br />Knowx innovation<br />19<br />
  25. 25. SEEK_SET - moves the file position to the position defined by offset<br />SEEK_CUR -moves the file position the number of bytes defined by offset from the current file position<br />SEEK_END - moves the file position to the number of bytes defined by offset from the end of the file<br />rewind ()<br />Resets the file read pointer back to the start of the file<br />Prototype<br />void rewind (FILE * stream);<br />12 January 2010<br />Knowx innovation<br />20<br />
  26. 26. Program<br />#include &lt;stdio.h&gt;<br />#define MAX_LINE 40 <br />#define FILENAME &quot;myfile.txt“<br />typedef struct <br />{ int id; float x_coord; float y_coord; char name[MAX_LINE+1]; <br />}<br />MY_TYPE_T; <br />MY_TYPE_T object;<br />int main()<br /> {<br />int i;<br />FILE *fin; <br />fin = fopen( FILENAME, &quot;r&quot; ); /* Open the input file */<br />12 January 2010<br />Knowx innovation<br />21<br />
  27. 27. if (fin == (FILE *)0) <br /> exit(-1);<br />fseek( fin, (2 * sizeof(MY_TYPE_T)), SEEK_SET ); /* Get the last entry */<br />fread( &object, sizeof(MY_TYPE_T), 1, fin ); <br />printf(&quot;%d %f %f %s &quot;, object.id, object.x_coord, object.y_coord, object.name );<br />rewind( fin ); /* Get the second to last entry */ <br />fseek( fin, (1 * sizeof(MY_TYPE_T)), SEEK_SET );<br />fread( &object, sizeof(MY_TYPE_T), 1, fin );<br />printf(&quot;%d %f %f %s &quot;, object.id, object.x_coord, object.y_coord, object.name ); <br />12 January 2010<br />Knowx innovation<br />22<br />
  28. 28. /* Get the first entry */<br />rewind( fin ); <br />fread( &object, sizeof(MY_TYPE_T), 1, fin ); <br />printf(&quot;%d %f %f %s &quot;, object.id, object.x_coord, object.y_coord, object.name ); fclose( fin );<br />return 0; <br />} <br />12 January 2010<br />Knowx innovation<br />23<br />
  29. 29. getpid() - to get the current process ID.<br />getppid() - to get the parent process ID.<br />getuid() - to get the user ID.<br />getgid() - to get the group ID.<br />12 January 2010<br />Knowx innovation<br />24<br />
  30. 30. Program<br />#include &lt;stdio.h&gt; <br />#include &lt;unistd.h&gt; <br />#include &lt;sys/types.h&gt; <br />int main()<br />{<br />pid_t myPid; <br />pid_t myParentPid; <br />gid_t myGid; <br />uid_t myUid; <br />myPid = getpid(); <br />myParentPid = getppid(); <br />myGid = getgid();<br />12 January 2010<br />Knowx innovation<br />25<br />
  31. 31. myUid = getuid(); <br /> printf( &quot;my process id is %d &quot;, myPid ); <br /> printf( &quot;my parent&apos;s process id is %d &quot;, myParentPid );<br />printf( &quot;my group id is %d &quot;, myGid );<br />printf( &quot;my user id is %d &quot;, myUid ); return 0; <br />} <br />Output<br />my process id is 10932.<br />my parents process id is 10795<br />My group id is 500<br />My user id is 500<br />12 January 2010<br />Knowx innovation<br />26<br />
  32. 32. Process api<br />Fork<br />Create a new child process<br />Wait<br />Suspend execution until a child processes exits<br />Signal<br />Install a new signal handler<br />Exec<br />Replace the current process image with a new process image<br />12 January 2010<br />Knowx innovation<br />27<br />
  33. 33. Fork<br />Creating new processes with in a given process.<br />return value of fork is <br />greater than zero then in the parent context. <br />Equal to zero then in the child context.<br />less than zero then error occurred.<br />12 January 2010<br />Knowx innovation<br />28<br />
  34. 34. Example<br />pid_t pid;<br />pid = fork();<br />if(pid&gt;0)<br /> {/*parent context*/}<br />else if (pid ==0)<br /> {/*child context*/}<br />else<br />{/*error occurred no child created*/}<br />12 January 2010<br />Knowx innovation<br />29<br />
  35. 35. 12 January 2010<br />Knowx innovation<br />30<br />
  36. 36. Program<br />#include &lt;sys/types.h&gt; <br />#include &lt;sys/wait.h&gt; <br />#include &lt;unistd.h&gt;<br />#include &lt;stdio.h&gt; <br />#include &lt;errno.h&gt; <br />int main() <br />{<br />pid_t ret; <br />int status, I, role = -1;<br />12 January 2010<br />Knowx innovation<br />31<br />
  37. 37. ret = fork();<br /> if (ret &gt; 0)<br /> { printf(&quot;Parent: This is the parent process (pid %d) &quot;, getpid());<br /> for (i = 0 ; i &lt; 6 ; i++) <br /> {<br /> printf(&quot;Parent: At count %d &quot;, i); sleep(1); }<br /> ret = wait( &status ); <br /> role = 0; <br /> }<br />else if (ret == 0)<br /> {<br /> printf(&quot;Child: This is the child process (pid %d) &quot;, getpid()); <br />12 January 2010<br />Knowx innovation<br />32<br />
  38. 38. for (i = 0 ; i &lt; 6 ; i++) <br /> {<br />printf(&quot;Child: At count %d &quot;, i); <br /> sleep(1); } role = 1;<br /> }<br />else <br /> { <br /> printf(&quot;Parent: Error trying to fork() (%d) &quot;, errno); <br /> }<br />printf(&quot;%s: Exiting... &quot;, ((role == 0) ? &quot;Parent&quot; : &quot;Child&quot;)); return 0; <br />} <br />12 January 2010<br />Knowx innovation<br />33<br />
  39. 39. wait<br />Suspend the calling process until a child process(created by this process) exits or until a signal is delivered.<br />Prototype<br />pid_t wait (int *status);<br />12 January 2010<br />Knowx innovation<br />34<br />
  40. 40. Signal<br />Install a signal handler for a process.<br />Prototype<br />sighandler_t signal (int signum, sighandler_t handler);<br />SIGHUP - hang up –commonly used to restart a task.<br />SIGKILL - kill signal<br />SIGINT - interrupt from the keyboard<br />SIGSTOP – stop process<br />SIGQUIT – quit signal from keyboard<br />12 January 2010<br />Knowx innovation<br />35<br />
  41. 41. Program<br />#include &lt;stdio.h&gt; <br />#include &lt;sys/types.h&gt;<br />#include &lt;sys/wait.h&gt;<br />#include &lt;unistd.h&gt; <br />#include &lt;signal.h&gt;<br />#include &lt;errno.h&gt; <br />void usr1_handler( int sig_num ) <br />{<br /> printf( &quot;Process (%d) got the SIGUSR1 &quot;, getpid() );<br /> }<br />int main() <br />{ <br />pid_t ret;<br />int status role = -1; <br />signal( SIGUSR1, usr1_handler ); <br />12 January 2010<br />Knowx innovation<br />36<br />
  42. 42. ret = fork();<br />if (ret &gt; 0)<br /> { /* Parent Context */<br />printf( &quot;Parent: This is the parent process (pid %d) &quot;, getpid() ); <br />role = 0; <br />pause(); <br />printf( &quot;Parent: Awaiting child exit &quot; );<br />ret = wait( &status );<br />}<br />12 January 2010<br />Knowx innovation<br />37<br />
  43. 43. else if (ret == 0) <br />{ /* Child Context */ <br />printf( &quot;Child: This is the child process (pid %d) &quot;, getpid() );<br />role = 1; <br />pause(); <br />} <br />else <br />{ /* Parent Context -- Error */ <br />printf( &quot;Parent: Error trying to fork() (%d) &quot;, errno ); <br />}<br />printf( &quot;%s: Exiting... &quot;, ((role == 0) ? &quot;Parent&quot; : &quot;Child&quot;) ); return 0;<br />} <br />12 January 2010<br />Knowx innovation<br />38<br />
  44. 44. exec<br />Replaces the current process image altogether.<br />Once the exec function replaces the current process, its pid is the same as the creating process<br />Permits the current process context to be replace with the program specifies as the first argument.<br />12 January 2010<br />Knowx innovation<br />39<br />
  45. 45. Example<br /> execl(“/bin/ls”,”ls”,”-la”,null);<br />This command replaces the current process with the ls (list directory)<br />Prototypes for the variants of exec<br />int execl (const char *path,const char * arg, …,..)<br />int execlp(const char *path,const char *arg,……)<br />int execv(const char *path, char *const argv[])<br />int execvp(const char *file, char *const argv[])<br />12 January 2010<br />Knowx innovation<br />40<br />
  46. 46. Program for shell interpreter<br />#include &lt;sys/types.h&gt; <br />#include &lt;sys/wait.h&gt;<br />#include &lt;unistd.h&gt;<br />#include &lt;stdio.h&gt;<br />#include &lt;stdlib.h&gt; <br />#include &lt;string.h&gt; <br />#define MAX_LINE 80 <br />int main() <br />{<br />int status; <br />pid_t childpid;<br />char cmd[MAX_LINE+1]; <br />char *sret;<br />12 January 2010<br />Knowx innovation<br />41<br />
  47. 47. while (1) <br />{<br />printf(&quot;mysh&gt;&quot;); <br />sret = fgets( cmd, sizeof(cmd), stdin ); <br />if (sret == NULL) <br />exit(-1);<br />cmd[ strlen(cmd)-1] = 0; <br />if (!strncmp(cmd, &quot;bye&quot;, 3)) <br />exit(0);<br />childpid = fork(); <br />if (childpid == 0)<br /> { execlp( cmd, cmd, 0 ); }<br />else if (childpid &gt; 0) <br /> { waitpid( childpid, &status, 0 ); } <br />printf(&quot; &quot;); <br />} <br />return 0;<br />} <br />12 January 2010<br />Knowx innovation<br />42<br />
  48. 48. Message queues<br />Messages are small collections of data (400 bytes, for example) that can be passed between cooperating programs through a message queue.<br />Messages within a queue can be of different types, and any process with proper permissions can receive the messages.<br />12 January 2010<br />Knowx innovation<br />43<br />
  49. 49. <ul><li>Creating message queue </li></ul>API<br />msgget(key_k key , int msgflg)<br />the key which signifies the name given to the queue<br />flags argument must contain the permission bits for the new queue<br />IPC_CREAT if the queue is being created<br />IPC_EXCL return an error if the message queue already exists<br />Return the msgid<br />Example <br />msgget(111 , 0666 | IPC_CREAT);<br />12 January 2010<br />Knowx innovation<br />44<br />
  50. 50. Program<br />#include &lt;stdio.h&gt; <br />#include &lt;sys/msg.h&gt;<br />#define MY_MQ_ID 111<br />int main()<br />{<br />int msgid;<br />/* Create the message queue with the id MY_MQ_ID */ <br />msgid = msgget( MY_MQ_ID, 0666 | IPC_CREAT ); <br />if (msgid &gt;= 0) <br /> { <br />printf( &quot;Created a Message Queue %d &quot;, msgid ); <br />}<br />return 0;<br />} <br />12 January 2010<br />Knowx innovation<br />45<br />
  51. 51. <ul><li>Sending the message</li></ul>API<br />int msgsnd (int msqid, const void *msgp, size_t msgsz, int msgflg); <br />Takes 4 parameters<br />msqid is the queue id of the existing queue.<br />msgp is the pointer that contains the address of the structure that holds the message and type.<br />struct message<br />{ long mtype; //The message type.priority<br />char mesg [MSGSZ];//The message is of length MSGSZ. }; <br />12 January 2010<br />Knowx innovation<br />46<br />
  52. 52. MSGSZ is the length of the message sent in bytes.<br />MSGFLG specifies the action to be taken if one or more of the following are true.<br />The number of bytes already in the queue is equal to msg_qbytes.<br />The total number of messages on all queues on the system has reached a maximum limit<br />Action to be taken<br />If (msgflg & IPC_NOWAIT) is non-zero, the message will not be sent and the calling process will return immediately<br />If (msgflg & IPC_NOWAIT) is 0, the calling process will suspend execution until one of the following occurs: <br />12 January 2010<br />Knowx innovation<br />47<br />
  53. 53. Program<br />#include &lt;sys/types.h&gt; <br />#include &lt;sys/ipc.h&gt; <br />#include &lt;sys/msg.h&gt;<br />#include &lt;stdio.h&gt; <br />#include &lt;string.h&gt; <br />#define MSGSZ 128 <br />typedef struct msgbuf /* will hold the message to be put in the queue */<br />{ long mtype; /* priority of message */<br />char mtext[MSGSZ]; /* the message that is stored */<br /> }<br />message_buf; <br />main() <br />{ int msqid; <br />int msgflg = IPC_CREAT | 0666; <br />key_t key; message_buf sbuf; size_t buf_length; key = 10;<br />printf(&quot;Calling msgget with key %#lx and flag %#o &quot;,key,msgflg);<br />12 January 2010<br />Knowx innovation<br />48<br />
  54. 54. /*A queue is created using the msget function with a key value 10 and the flag parameter being IPC_CREAT|06668 */<br /> if ((msqid = msgget(key, msgflg )) &lt; 0)<br />{ perror(&quot;msgget&quot;); <br /> exit(1); } <br />else <br />printf(&quot;msgget: msgget succeeded: msqid = %d &quot;, msqid); <br />sbuf.mtype = 1; /*setting the priority as 1 */<br />printf(&quot;msgget: msgget succeeded: msqid = %d &quot;, msqid);<br />(void) strcpy(sbuf.mtext, &quot;I am in the queue?&quot;); <br />/* copy the text &quot;I am in the queue&quot; into the array mtext which is message array */<br />printf(&quot;msgget: msgget succeeded: msqid = %d &quot;, msqid);<br />12 January 2010<br />Knowx innovation<br />49<br />
  55. 55. buf_length = strlen(sbuf.mtext) + 1 ; <br />/*sending the message with option IPC_NOWAIT */<br />if (msgsnd(msqid, &sbuf, buf_length, IPC_NOWAIT) &lt; 0)<br />{ <br />printf (&quot;%d, %d, %s, %d &quot;, msqid, sbuf.mtype, sbuf.mtext, buf_length); <br /> perror(&quot;msgsnd&quot;); <br /> exit(1); <br />} <br />else printf(&quot;Message: &quot;%s&quot; Sent &quot;, sbuf.mtext); <br />exit(0);<br /> }<br />12 January 2010<br />Knowx innovation<br />50<br />
  56. 56. <ul><li>Receive the message</li></ul>API<br />int msgrcv (int msqid, void *msgp, size_t msgsz, long msgtyp, int msgflg);<br />Arguments-<br />msqid is the queue id of the existing queue.<br />msgp is the point to a receiving buffer large enough to hold the received message<br />msgszthe maximum size of the received message<br />12 January 2010<br />Knowx innovation<br />51<br />
  57. 57. program<br />#include &lt;sys/types.h&gt; <br />#include &lt;sys/ipc.h&gt; <br />#include &lt;sys/msg.h&gt; <br />#include &lt;stdio.h&gt; <br />#define MSGSZ 128 <br />typedef struct msgbuf /* structure that will hold the message obtained from the queue*/<br />{<br /> long mtype; /*priority of message */<br /> char mtext[MSGSZ]; /*message that stored */<br />} message_buf;<br />main() <br />{<br />12 January 2010<br />Knowx innovation<br />52<br />
  58. 58. int msqid; <br />key_t key; <br />message_buf rbuf; <br />key = 10;<br />/*creating a queue with the key value 10, */<br />if ((msqid = msgget(key, 0666)) &lt; 0)<br />{ perror(&quot;msgget&quot;); <br />exit(1); } <br />/*acquires the message from the queue into rbuf*/<br />if (msgrcv(msqid, &rbuf, MSGSZ, 1, 0) &lt; 0) <br /> { perror(&quot;msgrcv&quot;);<br />exit(1); }<br />printf(&quot;%s &quot;, rbuf.mtext); <br />exit(0); <br />} <br />12 January 2010<br />Knowx innovation<br />53<br />
  59. 59. <ul><li>Configuring message queue</li></ul>API<br />msgctl(int msqid, int cmd, struct msqid_ds *buf)<br />3 Arguments <br />msqid is the queue id <br />cmd is a command constant<br />*bufis a pointer to structure.<br />Use to get the info about a message queue<br />Set the info for a message queue<br />Remove a message queue<br />12 January 2010<br />Knowx innovation<br />54<br />
  60. 60. 12 January 2010<br />Knowx innovation<br />55<br />
  61. 61. Variable that is protected.<br />It provides a means to restrict access to a resource that is shared amongst two or more processes<br />Two operations are permitted , called acquire and release.<br />2 types of semaphore.<br />Binary and Counting <br />12 January 2010<br />Knowx innovation<br />56<br />Semaphore<br />
  62. 62. The signal notations<br />P(semaphore variable) for wait<br />V(semaphore variable) for signal<br />12 January 2010<br />Knowx innovation<br />57<br />
  63. 63. A binary semaphore is a variable that can take only the values 0 and 1.<br />Represents a single resource and therefore when one process has acquired it,others are blocked until it is released.<br />P(sv) if sv greater than 0 , decrement sv.ifsvis zero suspend execution of the process<br />V(sv) if other process has been suspended waiting for sv, make it resume.if no waiting, increment sv<br />12 January 2010<br />Knowx innovation<br />58<br />Binary semaphore<br />
  64. 64. 12 January 2010<br />Knowx innovation<br />59<br />
  65. 65. Represent shared resources in quantities greater than one.<br />It could represent the entire set of buffers by setting its value to the number of buffers available.<br />Whenever the process acquires the semaphore, the value decrement .<br />When the semaphore value reaches zero, process are blocked until it becomes a non zero.<br />12 January 2010<br />Knowx innovation<br />60<br />Counting semaphore<br />
  66. 66. 12 January 2010<br />Knowx innovation<br />61<br />
  67. 67. <ul><li>Creating semaphore</li></ul>API<br />semget( key_t key, int num_sems, int sem_flags)<br />Key is the semaphore key<br />Num_sems is the Semaphore count<br />Flags <br /><ul><li>To acquire or release of semaphore</li></ul>API<br />semop(int sem_id, struct sembuf *sem_ops,size_tnum_sem_ops)<br />12 January 2010<br />Knowx innovation<br />62<br />
  68. 68. sem_id, is the semaphore identifier, as returned from semget.<br />sem_ops, is a pointer to an array of structures, each of which will have at least the following members:<br />Struct sembuf<br />{short sem_num;<br />short sem_op;<br />short sem_flg;<br />}<br />12 January 2010<br />Knowx innovation<br />63<br />
  69. 69. Program to acquire the semaphore.<br />#define MY_SEM_ID 111 <br />#define MY_SEMARRAY_ID 112 <br />#define NUM_SEMAPHORES 10 <br />#include &lt;stdio.h&gt; <br />#include &lt;sys/sem.h&gt; <br />#include &lt;stdlib.h&gt;<br />#include &quot;common.h&quot; <br />int main() <br />{ <br />int semid; <br />struct sembufsb;<br />/* Get the semaphore with the id MY_SEM_ID */<br />semid = semget( MY_SEM_ID, 1, 0 );<br />12 January 2010<br />Knowx innovation<br />64<br />
  70. 70. if (semid &gt;= 0) <br />{<br />sb.sem_num = 0; <br />sb.sem_op= -1; <br />sb.sem_flg= 0;<br />printf( &quot;semacq: Attempting to acquire semaphore %d &quot;, semid ); <br />/* Acquire the semaphore */ <br /> if (semop( semid, &sb, 1 ) == -1) <br /> {<br />printf(&quot;semacq: semop failed. &quot;); <br />exit(-1);<br />} <br /> printf( &quot;semacq: Semaphore acquired %d &quot;, semid);<br /> }<br />return 0;<br />} <br />12 January 2010<br />Knowx innovation<br />65<br />
  71. 71. Program to release the semaphore.<br />#define MY_SEM_ID 111 <br />#define MY_SEMARRAY_ID 112 <br />#define NUM_SEMAPHORES 10 <br />#include &lt;stdio.h&gt; <br />#include &lt;sys/sem.h&gt; <br />#include &lt;stdlib.h&gt;<br /> #include &quot;common.h&quot; <br />int main() <br />{ <br />int semid; <br />struct sembufsb;<br /> /* Get the semaphore with the id MY_SEM_ID */<br />semid = semget( MY_SEM_ID, 1, 0 );<br />12 January 2010<br />Knowx innovation<br />66<br />
  72. 72. if (semid &gt;= 0) <br />{<br />printf( &quot;semrel: Releasing semaphore %d &quot;, semid);<br />sb.sem_num = 0; <br />sb.sem_op= 1; <br />sb.sem_flg= 0; <br />/* Release the semaphore */ <br />if (semop( semid, &sb, 1 ) == -1) <br />{<br /> printf(&quot;semrel: semop failed. &quot;); <br />exit(-1); <br />}<br />printf( &quot;semrel: Semaphore released %d &quot;, semid ); <br />}<br />return 0;<br />} <br />12 January 2010<br />Knowx innovation<br />67<br />
  73. 73. 12 January 2010<br />Knowx innovation<br />68<br />

×