Hacker News Reader

Besides other benefits mentioned, a spinlock is a user-space construct; you don't incur the overhead switching to kernel mode at the loss of the benefits and guarantees those primitives provide. (I'm simplifying some.)

View on HN

I think spinlock is used when the critical section is relatively smaller where the threads are likely to perform small amount of work after acquiring locks. Thus locks are not expected to be held for longer duration. The other threads contending for the lock prefer to just spin and poll (and of course burn CPU) with the hope that lock will soon be relinquished by the owner thread. This way spinning threads avoid the overhead of context switching as well.

Goes without saying, spinlocks are not useful on uniprocessor machines or single core machines.

On Win32, this has been abstracted into the critical section API, which lives entirely in user-space and is a lightweight mutex. Many of the functionality implemented in the article are already part of the API (spin-aquire, etc.).

Though from what I've read, on Linux, pthreads implements its mutexes in user-space (futex) as well and are fairly cheap to use.

Yes, the difference is that spin locks never enter the kernel, while a good mutex only does on contention. In practice most mutexes use an hibrid approach.

Reply on news.ycombinator.com