Mistake on this page?
Report an issue in GitHub or email us

CriticalSectionLock

The CriticalSectionLock class provides a mechanism to access a resource without interruption. With the CriticalSectionLock::enable API, you can enter critical section with interrupts disabled. The CriticalSectionLock::disable() API is the exit from critical section, and the last exit call restores the state of interrupts.

CriticalSectionLock class is based on RAII approach. In other words, the constructor acquires the lock, and the destructor destroys it automatically when it is out of scope. We do not recommend you use CriticalSectionLock as global or a member of a class because you will enter critical section on object creation, and all interrupts will be disabled.

Mbed OS supports nesting of critical section, and the destructor enables interrupts only when you exit from the last nested critical section.

Note: You must not use time-consuming operations, standard library and RTOS functions inside critical section.

CriticalSectionLock class reference

Static Public Member Functions
static void enable ()
 Mark the start of a critical section. More...
static void disable ()
 Mark the end of a critical section. More...

CriticalSectionLock example

Here is an example that demonstrates a race condition issue and how CriticalSectionLock helps resolves it.

/*
 * Copyright (c) 2017 - 2020 Arm Limited and affiliates.
 * SPDX-License-Identifier: Apache-2.0
 */

#include "mbed.h"

#define  USE_CRITICAL_SECTION_LOCK      1   // Set 0 to see race condition
// Note: Might require few runs to see race condition

#define THREAD_CNT  8

int32_t value = 100000;
volatile int32_t counter = 0;

void increment(void)
{
    for (int i = 0; i < value; i++) {
#if (USE_CRITICAL_SECTION_LOCK == 1)
        CriticalSectionLock  lock;
#endif
        counter += 1;
    }
}

int get_count(void)
{
    if (counter == (value * THREAD_CNT)) {
        printf("No Race condition\n");
    } else {
        printf("Race condition\n");
    }
    return counter;
}

int main()
{
    Thread counter_thread[THREAD_CNT];

    for (int i = 0; i < THREAD_CNT; i++) {
        counter_thread[i].start(callback(increment));
    }

    // Wait for the threads to finish
    for (int i = 0; i < THREAD_CNT; i++) {
        counter_thread[i].join();
    }
    printf("Counter = %d\n", get_count());
}

Important Information for this Arm website

This site uses cookies to store information on your computer. By continuing to use our site, you consent to our cookies. If you are not happy with the use of these cookies, please review our Cookie Policy to learn how they can be disabled. By disabling cookies, some features of the site will not work.