Lock hierarchies to avoid deadlocks
In a previous post I explained how mutexes work and the problem of race conditions.
In this post I introduce another common problem with mutexes, which is deadlocks. Deadlocks are situations when a program can’t make any progress because it is waiting for a mutex that will never be available. This might sound stupid, but it’s something that actually happens very often.
A naive example could be this:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
#include <thread>
#include <mutex>
std::mutex mutexA;
std::mutex mutexB;
void doSomething() {
std::lock_guard<std::mutex> gA(mutexA);
std::this_thread::sleep_for(std::chrono::seconds(1));
std::lock_guard<std::mutex> gB(mutexB);
}
void doSomethingElse() {
std::lock_guard<std::mutex> gB(mutexB);
std::this_thread::sleep_for(std::chrono::seconds(1));
std::lock_guard<std::mutex> gA(mutexA);
}
int main() {
std::thread t1(doSomething);
std::thread t2(doSomethingElse);
t1.join();
t2.join();
}