Messages in this thread | | | Date | Wed, 27 Jan 2021 23:43:52 +0100 | From | Peter Zijlstra <> | Subject | Re: [PATCH 1/2] qspinlock: Ensure writes are pushed out of core write buffer |
| |
On Wed, Jan 27, 2021 at 09:01:08PM +0100, Alexander A Sverdlin wrote: > From: Alexander Sverdlin <alexander.sverdlin@nokia.com> > > Ensure writes are pushed out of core write buffer to prevent waiting code > on another cores from spinning longer than necessary.
Our smp_wmb() as defined does not have that property. You're relying on some arch specific details which do not belong in common code.
> diff --git a/kernel/locking/mcs_spinlock.h b/kernel/locking/mcs_spinlock.h > index 5e10153..10e497a 100644 > --- a/kernel/locking/mcs_spinlock.h > +++ b/kernel/locking/mcs_spinlock.h > @@ -89,6 +89,11 @@ void mcs_spin_lock(struct mcs_spinlock **lock, struct mcs_spinlock *node) > return; > } > WRITE_ONCE(prev->next, node); > + /* > + * This is necessary to make sure that the corresponding "while" in the > + * mcs_spin_unlock() doesn't loop forever > + */
This comment is utterly inadequate, since it does not describe an explicit ordering between two (or more) stores.
> + smp_wmb(); > > /* Wait until the lock holder passes the lock down. */ > arch_mcs_spin_lock_contended(&node->locked); > diff --git a/kernel/locking/qspinlock.c b/kernel/locking/qspinlock.c > index cbff6ba..577fe01 100644 > --- a/kernel/locking/qspinlock.c > +++ b/kernel/locking/qspinlock.c > @@ -469,6 +469,12 @@ void queued_spin_lock_slowpath(struct qspinlock *lock, u32 val) > > /* Link @node into the waitqueue. */ > WRITE_ONCE(prev->next, node); > + /* > + * This is necessary to make sure that the corresponding > + * smp_cond_load_relaxed() below (running on another core) > + * doesn't spin forever. > + */ > + smp_wmb();
That's insane, cache coherency should not allow that to happen in the first place. Our smp_wmb() cannot help with that.
> pv_wait_node(node, prev); > arch_mcs_spin_lock_contended(&node->locked);
| |