On Thu, May 30, 2024 at 03:45:52PM +0200, Frederic Weisbecker wrote: > Now that the (de-)offloading process can only apply to offline CPUs, > there is no more concurrency between rcu_core and nocb kthreads. Also > the mutation now happens on empty queues. > > Therefore the state machine can be reduced to a single bit called > SEGCBLIST_OFFLOADED. Simplify the transition as follows: > > * Upon offloading: queue the rdp to be added to the rcuog list and > wait for the rcuog kthread to set the SEGCBLIST_OFFLOADED bit. Unpark > rcuo kthread. > > * Upon de-offloading: Park rcuo kthread. Queue the rdp to be removed > from the rcuog list and wait for the rcuog kthread to clear the > SEGCBLIST_OFFLOADED bit. > > Signed-off-by: Frederic Weisbecker <[email protected]> > --- [...] > diff --git a/kernel/rcu/tree_nocb.h b/kernel/rcu/tree_nocb.h > index 24daf606de0c..72a2990d2087 100644 > --- a/kernel/rcu/tree_nocb.h > +++ b/kernel/rcu/tree_nocb.h [...] > @@ -1079,29 +1080,14 @@ static int rcu_nocb_rdp_deoffload(struct rcu_data > *rdp) > * but we stick to paranoia in this rare path. > */ > rcu_nocb_lock_irqsave(rdp, flags); > - rcu_segcblist_clear_flags(&rdp->cblist, SEGCBLIST_KTHREAD_GP); > - rcu_nocb_unlock_irqrestore(rdp, flags); > + rcu_segcblist_clear_flags(&rdp->cblist, SEGCBLIST_OFFLOADED); > + raw_spin_unlock_irqrestore(&rdp->nocb_lock, flags); >
Dropping rdp->nocb_lock unconditionally means we are the holder of it, right? If so, I think we better replace the above rcu_nocb_lock_irqsave() with raw_spin_lock_irqsave(). Regards, Boqun > list_del(&rdp->nocb_entry_rdp); > } > + > mutex_unlock(&rdp_gp->nocb_gp_kthread_mutex); > > - /* > - * Lock one last time to acquire latest callback updates from kthreads > - * so we can later handle callbacks locally without locking. > - */ > - rcu_nocb_lock_irqsave(rdp, flags); > - /* > - * Theoretically we could clear SEGCBLIST_LOCKING after the nocb > - * lock is released but how about being paranoid for once? > - */ > - rcu_segcblist_clear_flags(cblist, SEGCBLIST_LOCKING); > - /* > - * Without SEGCBLIST_LOCKING, we can't use > - * rcu_nocb_unlock_irqrestore() anymore. > - */ > - raw_spin_unlock_irqrestore(&rdp->nocb_lock, flags); > - > return 0; > } > [...]
