Re: [PATCH 3/7] locking/rwsem: check for active writer/spinner before wakeup

2014-08-05 Thread Waiman Long
On 08/04/2014 05:20 PM, Jason Low wrote: On Sun, 2014-08-03 at 22:36 -0400, Waiman Long wrote: On a highly contended rwsem, spinlock contention due to the slow rwsem_wake() call can be a significant portion of the total CPU cycles used. With writer lock stealing and writer optimistic spinning, t

Re: [PATCH 3/7] locking/rwsem: check for active writer/spinner before wakeup

2014-08-04 Thread Jason Low
On Sun, 2014-08-03 at 22:36 -0400, Waiman Long wrote: > On a highly contended rwsem, spinlock contention due to the slow > rwsem_wake() call can be a significant portion of the total CPU cycles > used. With writer lock stealing and writer optimistic spinning, there > is also a pretty good chance th

[PATCH 3/7] locking/rwsem: check for active writer/spinner before wakeup

2014-08-03 Thread Waiman Long
On a highly contended rwsem, spinlock contention due to the slow rwsem_wake() call can be a significant portion of the total CPU cycles used. With writer lock stealing and writer optimistic spinning, there is also a pretty good chance that the lock may have been stolen before the waker wakes up the