Messages in this thread | | | Date | Sat, 4 Sep 2021 12:14:29 +0200 | From | Peter Zijlstra <> | Subject | Re: [RFC] locking: rwbase: Take care of ordering guarantee for fastpath reader |
| |
On Fri, Sep 03, 2021 at 10:50:58PM +0800, Boqun Feng wrote: > > * ordering, either is strong enough to provide ACQUIRE order > > * for the above load of @readers. > > */ > > rwbase_set_and_save_current_state(state); > > raw_spin_lock_irqsave(&rtm->wait_lock, flags); > > > > while (readers) { > > ... > > readers = atomic_read(&rwb->readers); > > The above should be _acquire(), right? Pairs with the last reader > exiting the critical section and dec ->readers to 0. If so, it > undermines the necessity of the restructure?
This is the one that's followed by set_current_state(), no?
| |