4c8d244fae
NuttX kernel should not use the syscall functions, especially after enabling CONFIG_SCHED_INSTRUMENTATION_SYSCALL, all system functions will be traced to backend, which will impact system performance. Signed-off-by: chao an <anchao@xiaomi.com>
230 lines
7.6 KiB
C
230 lines
7.6 KiB
C
/****************************************************************************
|
|
* sched/pthread/pthread_mutexunlock.c
|
|
*
|
|
* Licensed to the Apache Software Foundation (ASF) under one or more
|
|
* contributor license agreements. See the NOTICE file distributed with
|
|
* this work for additional information regarding copyright ownership. The
|
|
* ASF licenses this file to you under the Apache License, Version 2.0 (the
|
|
* "License"); you may not use this file except in compliance with the
|
|
* License. You may obtain a copy of the License at
|
|
*
|
|
* http://www.apache.org/licenses/LICENSE-2.0
|
|
*
|
|
* Unless required by applicable law or agreed to in writing, software
|
|
* distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
|
|
* WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
|
|
* License for the specific language governing permissions and limitations
|
|
* under the License.
|
|
*
|
|
****************************************************************************/
|
|
|
|
/****************************************************************************
|
|
* Included Files
|
|
****************************************************************************/
|
|
|
|
#include <nuttx/config.h>
|
|
|
|
#include <unistd.h>
|
|
#include <pthread.h>
|
|
#include <sched.h>
|
|
#include <assert.h>
|
|
#include <errno.h>
|
|
#include <debug.h>
|
|
|
|
#include "pthread/pthread.h"
|
|
|
|
/****************************************************************************
|
|
* Private Functions
|
|
****************************************************************************/
|
|
|
|
/****************************************************************************
|
|
* Name: pthread_mutex_islocked
|
|
*
|
|
* Description:
|
|
* Return true is the mutex is locked.
|
|
*
|
|
* Input Parameters:
|
|
* None
|
|
*
|
|
* Returned Value:
|
|
* Returns true if the mutex is locked
|
|
*
|
|
****************************************************************************/
|
|
|
|
static inline bool pthread_mutex_islocked(FAR struct pthread_mutex_s *mutex)
|
|
{
|
|
int semcount = mutex->sem.semcount;
|
|
|
|
/* The underlying semaphore should have a count less than 2:
|
|
*
|
|
* 1 == mutex is unlocked.
|
|
* 0 == mutex is locked with no waiters
|
|
* -n == mutex is locked with 'n' waiters.
|
|
*/
|
|
|
|
DEBUGASSERT(semcount < 2);
|
|
return semcount < 1;
|
|
}
|
|
|
|
/****************************************************************************
|
|
* Public Functions
|
|
****************************************************************************/
|
|
|
|
/****************************************************************************
|
|
* Name: pthread_mutex_unlock
|
|
*
|
|
* Description:
|
|
* The pthread_mutex_unlock() function releases the mutex object referenced
|
|
* by mutex. The manner in which a mutex is released is dependent upon the
|
|
* mutex's type attribute. If there are threads blocked on the mutex object
|
|
* referenced by mutex when pthread_mutex_unlock() is called, resulting in
|
|
* the mutex becoming available, the scheduling policy is used to determine
|
|
* which thread shall acquire the mutex. (In the case of
|
|
* PTHREAD_MUTEX_RECURSIVE mutexes, the mutex becomes available when the
|
|
* count reaches zero and the calling thread no longer has any locks on
|
|
* this mutex).
|
|
*
|
|
* If a signal is delivered to a thread waiting for a mutex, upon return
|
|
* from the signal handler the thread resumes waiting for the mutex as if
|
|
* it was not interrupted.
|
|
*
|
|
* Input Parameters:
|
|
* None
|
|
*
|
|
* Returned Value:
|
|
* None
|
|
*
|
|
* Assumptions:
|
|
*
|
|
****************************************************************************/
|
|
|
|
int pthread_mutex_unlock(FAR pthread_mutex_t *mutex)
|
|
{
|
|
int ret = EPERM;
|
|
|
|
sinfo("mutex=%p\n", mutex);
|
|
DEBUGASSERT(mutex != NULL);
|
|
if (mutex == NULL)
|
|
{
|
|
return EINVAL;
|
|
}
|
|
|
|
/* Make sure the semaphore is stable while we make the following checks.
|
|
* This all needs to be one atomic action.
|
|
*/
|
|
|
|
sched_lock();
|
|
|
|
/* The unlock operation is only performed if the mutex is actually locked.
|
|
* EPERM *must* be returned if the mutex type is PTHREAD_MUTEX_ERRORCHECK
|
|
* or PTHREAD_MUTEX_RECURSIVE, or the mutex is a robust mutex, and the
|
|
* current thread does not own the mutex. Behavior is undefined for the
|
|
* remaining case.
|
|
*/
|
|
|
|
if (pthread_mutex_islocked(mutex))
|
|
{
|
|
#if !defined(CONFIG_PTHREAD_MUTEX_UNSAFE) || defined(CONFIG_PTHREAD_MUTEX_TYPES)
|
|
/* Does the calling thread own the semaphore? If no, should we return
|
|
* an error?
|
|
*
|
|
* Error checking is always performed for ERRORCHECK and RECURSIVE
|
|
* mutex types. Error checking is only performed for NORMAL (or
|
|
* DEFAULT) mutex type if the NORMAL mutex is robust. That is either:
|
|
*
|
|
* 1. CONFIG_PTHREAD_MUTEX_ROBUST is defined, or
|
|
* 2. CONFIG_PTHREAD_MUTEX_BOTH is defined and the robust flag is set
|
|
*/
|
|
|
|
#if defined(CONFIG_PTHREAD_MUTEX_ROBUST)
|
|
/* Not that error checking is always performed if the configuration has
|
|
* CONFIG_PTHREAD_MUTEX_ROBUST defined. Just check if the calling
|
|
* thread owns the semaphore.
|
|
*/
|
|
|
|
if (mutex->pid != nxsched_gettid())
|
|
|
|
#elif defined(CONFIG_PTHREAD_MUTEX_UNSAFE) && defined(CONFIG_PTHREAD_MUTEX_TYPES)
|
|
/* If mutex types are not supported, then all mutexes are NORMAL (or
|
|
* DEFAULT). Error checking should never be performed for the
|
|
* non-robust NORMAL mutex type.
|
|
*/
|
|
|
|
if (mutex->type != PTHREAD_MUTEX_NORMAL &&
|
|
mutex->pid != nxsched_gettid())
|
|
|
|
#else /* CONFIG_PTHREAD_MUTEX_BOTH */
|
|
/* Skip the error check if this is a non-robust NORMAL mutex */
|
|
|
|
bool errcheck = ((mutex->flags & _PTHREAD_MFLAGS_ROBUST) != 0);
|
|
#ifdef CONFIG_PTHREAD_MUTEX_TYPES
|
|
errcheck |= (mutex->type != PTHREAD_MUTEX_NORMAL);
|
|
#endif
|
|
|
|
/* Does the calling thread own the semaphore? If not should we report
|
|
* the EPERM error?
|
|
*/
|
|
|
|
if (errcheck && mutex->pid != nxsched_gettid())
|
|
#endif
|
|
{
|
|
/* No... return an EPERM error.
|
|
*
|
|
* Per POSIX: "EPERM should be returned if the mutex type is
|
|
* PTHREAD_MUTEX_ERRORCHECK or PTHREAD_MUTEX_RECURSIVE, or the
|
|
* mutex is a robust mutex, and the current thread does not own
|
|
* the mutex."
|
|
*
|
|
* For the case of the non-robust PTHREAD_MUTEX_NORMAL mutex,
|
|
* the behavior is undefined.
|
|
*/
|
|
|
|
serr("ERROR: Holder=%d returning EPERM\n", mutex->pid);
|
|
ret = EPERM;
|
|
}
|
|
else
|
|
#endif /* !CONFIG_PTHREAD_MUTEX_UNSAFE || CONFIG_PTHREAD_MUTEX_TYPES */
|
|
|
|
#ifdef CONFIG_PTHREAD_MUTEX_TYPES
|
|
/* Yes, the caller owns the semaphore.. Is this a recursive mutex? */
|
|
|
|
if (mutex->type == PTHREAD_MUTEX_RECURSIVE && mutex->nlocks > 1)
|
|
{
|
|
/* This is a recursive mutex and we there are multiple locks held.
|
|
* Retain the mutex lock, just decrement the count of locks held,
|
|
* and return success.
|
|
*/
|
|
|
|
mutex->nlocks--;
|
|
ret = OK;
|
|
}
|
|
else
|
|
|
|
#endif /* CONFIG_PTHREAD_MUTEX_TYPES */
|
|
|
|
/* This is either a non-recursive mutex or is the outermost unlock of
|
|
* a recursive mutex.
|
|
*
|
|
* In the case where the calling thread is NOT the holder of the
|
|
* thread, the behavior is undefined per POSIX.
|
|
* Here we do the same as GLIBC:
|
|
* We allow the other thread to release the mutex even though it does
|
|
* not own it.
|
|
*/
|
|
|
|
{
|
|
/* Nullify the pid and lock count then post the semaphore */
|
|
|
|
mutex->pid = INVALID_PROCESS_ID;
|
|
#ifdef CONFIG_PTHREAD_MUTEX_TYPES
|
|
mutex->nlocks = 0;
|
|
#endif
|
|
ret = pthread_mutex_give(mutex);
|
|
}
|
|
}
|
|
|
|
sched_unlock();
|
|
sinfo("Returning %d\n", ret);
|
|
return ret;
|
|
}
|