Why isn't the conclusion to just use the mutex at least on Linux? It seems to have both consistently low latency and great throughput. What else can you get from the spinlock?
Sigh. From the article itself:
> I hope to have convinced you to avoid spinlocks, even for very short sections. Spinlocks are only a benefit if you really really care about the fast path and are certain that the lock will essentially always be free to take. On my computer the fast path on a spinlock is roughly 7ns and on a mutex it’s roughly 14ns, so both of those numbers are very fast. But when spinlocks go bad, the scheduler can be really confused, at least on Linux.
That said, his conclusion is in fact just as you stated, if you read in between the lines. (This paper is really about the scheduler, so it's not exactly that conclusion.)