Messages in this thread | | | Date | Sat, 21 May 2016 00:01:00 -0400 | From | Waiman Long <> | Subject | Re: sem_lock() vs qspinlocks |
| |
On 05/20/2016 08:59 PM, Davidlohr Bueso wrote: > On Fri, 20 May 2016, Peter Zijlstra wrote: > >> On Fri, May 20, 2016 at 04:47:43PM -0400, Waiman Long wrote: >> >>> >Similarly, and I know you hate it, but afaict, then semantically >>> >queued_spin_is_contended() ought to be: >>> > >>> >- return atomic_read(&lock->val) & ~_Q_LOCKED_MASK; >>> >+ return atomic_read(&lock->val); >>> > >> >>> Looking for contended lock, you need to consider the lock waiters >>> also. So >>> looking at the whole word is right. >> >> No, you _only_ need to look at the lock waiters. > > Is there anyway to do this in a single atomic_read? My thought is that > otherwise > we could further expand the race window of when the lock is and isn't > contended (as returned to by the user). Ie avoiding crap like: > > atomic_read(&lock->val) && atomic_read(&lock->val) != _Q_LOCKED_VAL > > In any case, falsely returning for the 'locked, uncontended' case, vs > completely > ignoring waiters is probably the lesser evil :). > > Thanks, > Davidlohr
The existing code is doing that, but I would argue that including the locked, but uncontended case isn't a bad idea.
Cheers, Longman
| |