mutex

A "safe" mutex lock

Safe means that there's no deadlock.

Example

var mu sync.Mutex

mu.Lock()

More Code....

// deadlock! I don't know why Go runtime will not panic.
mu.Lock()

However, the project can prevent that.

var mu Mutex

mu.Lock()

More Code....

// safe to lock again. Actually it doesn't lock.
mu.Lock()

But this will cause the chaos of lock state.

So another function will be useful.

IsLocked()

var mu Mutex

mu.Lock()

More Code....

if !mu.Locked() {
    mu.Lock()
}

What about unlock?

To avoid the chaos of lock state, there's a TryUnlock()

If you can't ensure the lock is locked or not.

Use TryUnlock()

It's safe to call concurrently(Of course!).

Recursive lock

Recursive lock acts like C++ STL Recursive lock.

It can lock multiple times, but if you need unlock, you need to unlock it multiple times. Or it will cause deadlock.

var rm Recursive
for i := 0; i < 5; i++ {
    rm.Lock()
}

More Code ...

for i := 0; i < 5; i++ {
    rm.Unlock()
}