Re: [PATCH v2 0/9] Remove spin_unlock_wait()

From: Manfred Spraul
Date: Fri Jul 07 2017 - 13:48:19 EST

Hi Ingo,

On 07/07/2017 10:31 AM, Ingo Molnar wrote:

There's another, probably just as significant advantage: queued_spin_unlock_wait()
is 'read-only', while spin_lock()+spin_unlock() dirties the lock cache line. On
any bigger system this should make a very measurable difference - if
spin_unlock_wait() is ever used in a performance critical code path.
At least for ipc/sem:
Dirtying the cacheline (in the slow path) allows to remove a smp_mb() in the hot path.
So for sem_lock(), I either need a primitive that dirties the cacheline or sem_lock() must continue to use spin_lock()/spin_unlock().