sem.c

在linux下的进程间通信的源代码

/*
 * linux/ipc/sem.c
 * Copyright (C) 1992 Krishna Balasubramanian
 * Copyright (C) 1995 Eric Schenk, Bruno Haible
 *
 * IMPLEMENTATION NOTES ON CODE REWRITE (Eric Schenk, January 1995):
 * This code underwent a massive rewrite in order to solve some problems
 * with the original code. In particular the original code failed to
 * wake up processes that were waiting for semval to go to 0 if the
 * value went to 0 and was then incremented rapidly enough. In solving
 * this problem I have also modified the implementation so that it
 * processes pending operations in a FIFO manner, thus give a guarantee
 * that processes waiting for a lock on the semaphore won't starve
 * unless another locking process fails to unlock.
 * In addition the following two changes in behavior have been introduced:
 * - The original implementation of semop returned the value
 *   last semaphore element examined on success. This does not
 *   match the manual page specifications, and effectively
 *   allows the user to read the semaphore even if they do not
 *   have read permissions. The implementation now returns 0
 *   on success as stated in the manual page.
 * - There is some confusion over whether the set of undo adjustments
 *   to be performed at exit should be done in an atomic manner.
 *   That is, if we are attempting to decrement the semval should we queue
 *   up and wait until we can do so legally?
 *   The original implementation attempted to do this.
 *   The current implementation does not do so. This is because I don't
 *   think it is the right thing (TM) to do, and because I couldn't
 *   see a clean way to get the old behavior with the new design.
 *   The POSIX standard and SVID should be consulted to determine
 *   what behavior is mandated.
 *
 * Further notes on refinement (Christoph Rohland, December 1998):
 * - The POSIX standard says, that the undo adjustments simply should
 *   redo. So the current implementation is o.K.
 * - The previous code had two flaws:
 *   1) It actively gave the semaphore to the next waiting process
 *      sleeping on the semaphore. Since this process did not have the
 *      cpu this led to many unnecessary context switches and bad
 *      performance. Now we only check which process should be able to
 *      get the semaphore and if this process wants to reduce some
 *      semaphore value we simply wake it up without doing the
 *      operation. So it has to try to get it later. Thus e.g. the
 *      running process may reacquire the semaphore during the current
 *      time slice. If it only waits for zero or increases the semaphore,
 *      we do the operation in advance and wake it up.
 *   2) It did not wake up all zero waiting processes. We try to do
 *      better but only get the semops right which only wait for zero or
 *      increase. If there are decrement operations in the operations
 *      array we do the same as before.
 *
 * /proc/sysvipc/sem support (c) 1999 Dragos Acostachioaie <[e-mail]>
 *
 * SMP-threaded, sysctl's added
 * (c) 1999 Manfred Spraul <[e-mail]>
 */

#include <linux/config.h>
#include <linux/malloc.h>
#include <linux/spinlock.h>
#include <linux/init.h>
#include <linux/proc_fs.h>
#include <asm/uaccess.h>
#include "util.h"


#define sem_lock(id)	((struct sem_array*)ipc_lock(&sem_ids,id))
#define sem_unlock(id)	ipc_unlock(&sem_ids,id)
#define sem_rmid(id)	((struct sem_array*)ipc_rmid(&sem_ids,id))
#define sem_checkid(sma, semid)	\
	ipc_checkid(&sem_ids,&sma->sem_perm,semid)
#define sem_buildid(id, seq) \
	ipc_buildid(&sem_ids, id, seq)
static struct ipc_ids sem_ids;

static int newary (key_t, int, int);
static void freeary (int id);
#ifdef CONFIG_PROC_FS
static int sysvipc_sem_read_proc(char *buffer, char **start, off_t offset, int length, int *eof, void *data);
#endif

#define SEMMSL_FAST	256 /* 512 bytes on stack */
#define SEMOPM_FAST	64  /* ~ 372 bytes on stack */

/*
 * linked list protection:
 *	sem_undo.id_next,
 *	sem_array.sem_pending{,last},
 *	sem_array.sem_undo: sem_lock() for read/write
 *	sem_undo.proc_next: only "current" is allowed to read/write that field.
 *	
 */

int sem_ctls[4] = {SEMMSL, SEMMNS, SEMOPM, SEMMNI};
#define sc_semmsl	(sem_ctls[0])
#define sc_semmns	(sem_ctls[1])
#define sc_semopm	(sem_ctls[2])
#define sc_semmni	(sem_ctls[3])

static int used_sems;

void __init sem_init (void)
{
	used_sems = 0;
	ipc_init_ids(&sem_ids,sc_semmni);

#ifdef CONFIG_PROC_FS
	create_proc_read_entry("sysvipc/sem", 0, 0, sysvipc_sem_read_proc, NULL);
#endif
}

static int newary (key_t key, int nsems, int semflg)
{
	int id;
	struct sem_array *sma;
	int size;

	if (!nsems)
		return -EINVAL;
	if (used_sems + nsems > sc_semmns)
		return -ENOSPC;

	size = sizeof (*sma) + nsems * sizeof (struct sem);
	sma = (struct sem_array *) ipc_alloc(size);
	if (!sma) {
		return -ENOMEM;
	}
	memset (sma, 0, size);
	id = ipc_addid(&sem_ids, &sma->sem_perm, sc_semmni);
	if(id == -1) {
		ipc_free(sma, size);
		return -ENOSPC;
	}
	used_sems += nsems;

	sma->sem_perm.mode = (semflg & S_IRWXUGO);
	sma->sem_perm.key = key;

	sma->sem_base = (struct sem *) &sma[1];
	/* sma->sem_pending = NULL; */
	sma->sem_pending_last = &sma->sem_pending;
	/* sma->undo = NULL; */
	sma->sem_nsems = nsems;
	sma->sem_ctime = CURRENT_TIME;
	sem_unlock(id);

	return sem_buildid(id, sma->sem_perm.seq);
}

asmlinkage long sys_semget (key_t key, int nsems, int semflg)
{
	int id, err = -EINVAL;
	struct sem_array *sma;

	if (nsems < 0 || nsems > sc_semmsl)
		return -EINVAL;
	down(&sem_ids.sem);
	
	if (key == IPC_PRIVATE) {
		err = newary(key, nsems, semflg);
	} else if ((id = ipc_findkey(&sem_ids, key)) == -1) {  /* key not used */
		if (!(semflg & IPC_CREAT))
			err = -ENOENT;
		else
			err = newary(key, nsems, semflg);
	} else if (semflg & IPC_CREAT && semflg & IPC_EXCL) {
		err = -EEXIST;
	} else {
		sma = sem_lock(id);
		if(sma==NULL)
			BUG();
		if (nsems > sma->sem_nsems)
			err = -EINVAL;
		else if (ipcperms(&sma->sem_perm, semflg))
			err = -EACCES;
		else
			err = sem_buildid(id, sma->sem_perm.seq);
		sem_unlock(id);
	}

	up(&sem_ids.sem);
	return err;
}

/* doesn't acquire the sem_lock on error! */
static int sem_revalidate(int semid, struct sem_array* sma, int nsems, short flg)
{
	struct sem_array* smanew;

	smanew = sem_lock(semid);
	if(smanew==NULL)
		return -EIDRM;
	if(smanew != sma || sem_checkid(sma,semid) || sma->sem_nsems != nsems) {
		sem_unlock(semid);
		return -EIDRM;
	}

	if (ipcperms(&sma->sem_perm, flg)) {
		sem_unlock(semid);
		return -EACCES;
	}
	return 0;
}
/* Manage the doubly linked list sma->sem_pending as a FIFO:
 * insert new queue elements at the tail sma->sem_pending_last.
 */
static inline void append_to_queue (struct sem_array * sma,
				    struct sem_queue * q)
{
	*(q->prev = sma->sem_pending_last) = q;
	*(sma->sem_pending_last = &q->next) = NULL;
}

static inline void prepend_to_queue (struct sem_array * sma,
				     struct sem_queue * q)
{
	q->next = sma->sem_pending;
	*(q->prev = &sma->sem_pending) = q;
	if (q->next)
		q->next->prev = &q->next;
	else /* sma->sem_pending_last == &sma->sem_pending */
		sma->sem_pending_last = &q->next;
}

static inline void remove_from_queue (struct sem_array * sma,
				      struct sem_queue * q)
{
	*(q->prev) = q->next;
	if (q->next)
		q->next->prev = q->prev;
	else /* sma->sem_pending_last == &q->next */
		sma->sem_pending_last = q->prev;
	q->prev = NULL; /* mark as removed */
}

/*
 * Determine whether a sequence of semaphore operations would succeed
 * all at once. Return 0 if yes, 1 if need to sleep, else return error code.
 */

static int try_atomic_semop (struct sem_array * sma, struct sembuf * sops,
			     int nsops, struct sem_undo *un, int pid,
			     int do_undo)
{
	int result, sem_op;
	struct sembuf *sop;
	struct sem * curr;

	for (sop = sops; sop < sops + nsops; sop++) {
		curr = sma->sem_base + sop->sem_num;
		sem_op = sop->sem_op;

		if (!sem_op && curr->semval)
			goto would_block;

		curr->sempid = (curr->sempid << 16) | pid;
		curr->semval += sem_op;
		if (sop->sem_flg & SEM_UNDO)
			un->semadj[sop->sem_num] -= sem_op;

		if (curr->semval < 0)
			goto would_block;
		if (curr->semval > SEMVMX)
			goto out_of_range;
	}

	if (do_undo)
	{
		sop--;
		result = 0;
		goto undo;
	}

	sma->sem_otime = CURRENT_TIME;
	return 0;

out_of_range:
	result = -ERANGE;
	goto undo;

would_block:
	if (sop->sem_flg & IPC_NOWAIT)
		result = -EAGAIN;
	else
		result = 1;

undo:
	while (sop >= sops) {
		curr = sma->sem_base + sop->sem_num;
		curr->semval -= sop->sem_op;
		curr->sempid >>= 16;

		if (sop->sem_flg & SEM_UNDO)
			un->semadj[sop->sem_num] += sop->sem_op;
		sop--;
	}

	return result;
}

/* Go through the pending queue for the indicated semaphore
 * looking for tasks that can be completed.
 */
static void update_queue (struct sem_array * sma)
{
	int error;
	struct sem_queue * q;

	for (q = sma->sem_pending; q; q = q->next) {
			
		if (q->status == 1)
			continue;	/* this one was woken up before */

		error = try_atomic_semop(sma, q->sops, q->nsops,
					 q->undo, q->pid, q->alter);

		/* Does q->sleeper still need to sleep? */
		if (error <= 0) {
				/* Found one, wake it up */
			wake_up_process(q->sleeper);
			if (error == 0 && q->alter) {
				/* if q-> alter let it self try */
				q->status = 1;
				return;
			}
			q->status = error;
			remove_from_queue(sma,q);
		}
	}
}

/* The following counts are associated to each semaphore:
 *   semncnt        number of tasks waiting on semval being nonzero
 *   semzcnt        number of tasks waiting on semval being zero
 * This model assumes that a task waits on exactly one semaphore.
 * Since semaphore operations are to be performed atomically, tasks actually
 * wait on a whole sequence of semaphores simultaneously.
 * The counts we return here are a rough approximation, but still
 * warrant that semncnt+semzcnt>0 if the task is on the pending queue.
 */
static int count_semncnt (struct sem_array * sma, ushort semnum)
{
	int semncnt;
	struct sem_queue * q;

	semncnt = 0;
	for (q = sma->sem_pending; q; q = q->next) {
		struct sembuf * sops = q->sops;
		int nsops = q->nsops;
		int i;
		for (i = 0; i < nsops; i++)
			if (sops[i].sem_num == semnum
			    && (sops[i].sem_op < 0)
			    && !(sops[i].sem_flg & IPC_NOWAIT))
				semncnt++;
	}
	return semncnt;
}
static int count_semzcnt (struct sem_array * sma, ushort semnum)
{
	int semzcnt;
	struct sem_queue * q;

	semzcnt = 0;
	for (q = sma->sem_pending; q; q = q->next) {
		struct sembuf * sops = q->sops;
		int nsops = q->nsops;
		int i;
		for (i = 0; i < nsops; i++)
			if (sops[i].sem_num == semnum
			    && (sops[i].sem_op == 0)
			    && !(sops[i].sem_flg & IPC_NOWAIT))
				semzcnt++;
	}
	return semzcnt;
}

/* Free a semaphore set. */
static void freeary (int id)
{
	struct sem_array *sma;
	struct sem_undo *un;
	struct sem_queue *q;
	int size;

	sma = sem_rmid(id);

	/* Invalidate the existing undo structures for this semaphore set.
	 * (They will be freed without any further action in sem_exit()
	 * or during the next semop.)
	 */
	for (un = sma->undo; un; un = un->id_next)
		un->semid = -1;

	/* Wake up all pending processes and let them fail with EIDRM. */
	for (q = sma->sem_pending; q; q = q->next) {
		q->status = -EIDRM;
		q->prev = NULL;
		wake_up_process(q->sleeper); /* doesn't sleep */
	}
	sem_unlock(id);

	used_sems -= sma->sem_nsems;
	size = sizeof (*sma) + sma->sem_nsems * sizeof (struct sem);
	ipc_free(sma, size);
}

static unsigned long copy_semid_to_user(void *buf, struct semid64_ds *in, int version)
{
	switch(version) {
	case IPC_64:
		return copy_to_user(buf, in, sizeof(*in));
	case IPC_OLD:
	    {
		struct semid_ds out;

		ipc64_perm_to_ipc_perm(&in->sem_perm, &out.sem_perm);

		out.sem_otime	= in->sem_otime;
		out.sem_ctime	= in->sem_ctime;
		out.sem_nsems	= in->sem_nsems;

		return copy_to_user(buf, &out, sizeof(out));
	    }
	default:
		return -EINVAL;
	}
}

int semctl_nolock(int semid, int semnum, int cmd, int version, union semun arg)
{
	int err = -EINVAL;

	switch(cmd) {
	case IPC_INFO:
	case SEM_INFO:
	{
		struct seminfo seminfo;
		int max_id;

		memset(&seminfo,0,sizeof(seminfo));
		seminfo.semmni = sc_semmni;
		seminfo.semmns = sc_semmns;
		seminfo.semmsl = sc_semmsl;
		seminfo.semopm = sc_semopm;
		seminfo.semvmx = SEMVMX;
		seminfo.semmnu = SEMMNU;
		seminfo.semmap = SEMMAP;
		seminfo.semume = SEMUME;
		down(&sem_ids.sem);
		if (cmd == SEM_INFO) {
			seminfo.semusz = sem_ids.in_use;
			seminfo.semaem = used_sems;
		} else {
			seminfo.semusz = SEMUSZ;
			seminfo.semaem = SEMAEM;
		}
		max_id = sem_ids.max_id;
		up(&sem_ids.sem);
		if (copy_to_user (arg.__buf, &seminfo, sizeof(struct seminfo))) 
			return -EFAULT;
		return (max_id < 0) ? 0: max_id;
	}
	case SEM_STAT:
	{
		struct sem_array *sma;
		struct semid64_ds tbuf;
		int id;

		if(semid > sem_ids.size)
			return -EINVAL;

		memset(&tbuf,0,sizeof(tbuf));

		sma = sem_lock(semid);
		if(sma == NULL)
			return -EINVAL;

		err = -EACCES;
		if (ipcperms (&sma->sem_perm, S_IRUGO))
			goto out_unlock;
		id = sem_buildid(semid, sma->sem_perm.seq);

		kernel_to_ipc64_perm(&sma->sem_perm, &tbuf.sem_perm);
		tbuf.sem_otime  = sma->sem_otime;
		tbuf.sem_ctime  = sma->sem_ctime;
		tbuf.sem_nsems  = sma->sem_nsems;
		sem_unlock(semid);
		if (copy_semid_to_user (arg.buf, &tbuf, version))
			return -EFAULT;
		return id;
	}
	default:
		return -EINVAL;
	}
	return err;
out_unlock:
	sem_unlock(semid);
	return err;
}

int semctl_main(int semid, int semnum, int cmd, int version, union semun arg)
{
	struct sem_array *sma;
	struct sem* curr;
	int err;
	ushort fast_sem_io[SEMMSL_FAST];
	ushort* sem_io = fast_sem_io;
	int nsems;

	sma = sem_lock(semid);
	if(sma==NULL)
		return -EINVAL;

	nsems = sma->sem_nsems;

	err=-EIDRM;
	if (sem_checkid(sma,semid))
		goto out_unlock;

	err = -EACCES;
	if (ipcperms (&sma->sem_perm, (cmd==SETVAL||cmd==SETALL)?S_IWUGO:S_IRUGO))
		goto out_unlock;

	switch (cmd) {
	case GETALL:
	{
		ushort *array = arg.array;
		int i;

		if(nsems > SEMMSL_FAST) {
			sem_unlock(semid);			
			sem_io = ipc_alloc(sizeof(ushort)*nsems);
			if(sem_io == NULL)
				return -ENOMEM;
			err = sem_revalidate(semid, sma, nsems, S_IRUGO);
			if(err)
				goto out_free;
		}

		for (i = 0; i < sma->sem_nsems; i++)
			sem_io[i] = sma->sem_base[i].semval;
		sem_unlock(semid);
		err = 0;
		if(copy_to_user(array, sem_io, nsems*sizeof(ushort)))
			err = -EFAULT;