> 在 2016年6月26日，00:13，Peter Zijlstra <pet...@infradead.org> 写道： > > On Sat, Jun 25, 2016 at 06:09:22PM +0200, Peter Zijlstra wrote: >> That works here, but it would not work for the need_resched() in >> mutex_spin_on_owner() and mutex_optimistic_spin() which need equal >> treatment. >> >> Because those too we want to limit. >> >> The count thing, while a little more cumbersome, is more widely >> applicable than just the one OSQ case where we happen to have a cpu >> number. > > Although I suppose that mutex_spin_on_owner() (and with that the rsem > variant) could use task_cpu(lock->owner) once we've established that the > owner pointer is still valid. > > yes, What I am going to fix next is these XXX_spin_on_owner, including mutex_spin_on_owner, rwsem_spin_on_owner ….
by the way I still think mutex_unlock has a big overload too.
