Re: [PATCH] locking/pvqspinlock: restore/set vcpu_hashed state after failing adaptive locking spinning

From: xinhui
Date: Thu Jul 14 2016 - 04:22:52 EST




On 2016å07æ14æ 15:53, Wanpeng Li wrote:
From: Wanpeng Li <wanpeng.li@xxxxxxxxxxx>

When the lock holder vCPU is racing with the queue head:

CPU 0 (lock holder) CPU1 (queue head)
=================== =================
spin_lock(); spin_lock();
pv_kick_node(); pv_wait_head_or_lock();
if (!lp) {
lp = pv_hash(lock, pn);
xchg(&l->locked, _Q_SLOW_VAL);
}
WRITE_ONCE(pn->state, vcpu_halted);
cmpxchg(&pn->state,
vcpu_halted, vcpu_hashed);
WRITE_ONCE(l->locked, _Q_SLOW_VAL);
(void)pv_hash(lock, pn);

In this case, lock holder inserts the pv_node of queue head into the
hash table and set _Q_SLOW_VAL unnecessary. This patch avoids it by
restoring/setting vcpu_halted state after failing adaptive locking
spinning.


good catching. this is hash entry leak.

After apply Waiman's pathset "locking/pvqspinlock: Fix missed PV wakeup & support PPC", this is not the big problem.
But we can still save another cmpxchg.

So
Reviewed-by: Pan Xinhui <xinhui.pan@xxxxxxxxxxxxxxxxxx>

Cc: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxxxxx>
Cc: Waiman Long <Waiman.Long@xxxxxxx>
Cc: Davidlohr Bueso <dave@xxxxxxxxxxxx>
Signed-off-by: Wanpeng Li <wanpeng.li@xxxxxxxxxxx>
---
kernel/locking/qspinlock_paravirt.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/locking/qspinlock_paravirt.h b/kernel/locking/qspinlock_paravirt.h
index 21ede57..ac7d20b 100644
--- a/kernel/locking/qspinlock_paravirt.h
+++ b/kernel/locking/qspinlock_paravirt.h
@@ -450,7 +450,7 @@ pv_wait_head_or_lock(struct qspinlock *lock, struct mcs_spinlock *node)
goto gotlock;
}
}
- WRITE_ONCE(pn->state, vcpu_halted);
+ WRITE_ONCE(pn->state, vcpu_hashed);
qstat_inc(qstat_pv_wait_head, true);
qstat_inc(qstat_pv_wait_again, waitcnt);
pv_wait(&l->locked, _Q_SLOW_VAL);