On Thu, 2016-08-18 at 17:39 -0700, Jason Low wrote: > Imre reported an issue where threads are getting starved when trying > to acquire a mutex. Threads acquiring a mutex can get arbitrarily delayed > sleeping on a mutex because other threads can continually steal the lock > in the fastpath and/or through optimistic spinning. > > Waiman has developed patches that allow waiters to return to optimistic > spinning, thus reducing the probability that starvation occurs. However, > Imre still sees this starvation problem in the workloads when optimistic > spinning is disabled. > > This patch adds an additional boolean to the mutex that gets used in > the CONFIG_SMP && !CONFIG_MUTEX_SPIN_ON_OWNER cases. The flag signifies > whether or not other threads need to yield to a waiter and gets set > when a waiter spends too much time waiting for the mutex. The threshold > is currently set to 16 wakeups, and once the wakeup threshold is exceeded, > other threads must yield to the top waiter. The flag gets cleared > immediately after the top waiter acquires the mutex. > > This prevents waiters from getting starved without sacrificing much > much performance, as lock stealing is still allowed and only > temporarily disabled when it is detected that a waiter has been waiting > for too long.
Changes from v3 (in peterz locking/core) -> v4: 1. Fixed patch title. It should be "Prevent lock starvation when spinning is disabled" instead of "when spinning is enabled". 2. Call clear_yield_to_waiter() when a top waiter exits in the 'err' case. 3. Only clear yield_to_waiter if the thread is the top waiter and not if it is a non-top waiter that received a signal.
