Re: [PATCH v10 10/19] qspinlock, x86: Allow unfair spinlock in a virtual guest

[Peter Zijlstra <peterz@xxxxxxxxxxxxx>]

On Wed, May 07, 2014 at 11:01:38AM -0400, Waiman Long wrote:


No, we want the unfair thing for VIRT, not PARAVIRT.


> diff --git a/kernel/locking/qspinlock.c b/kernel/locking/qspinlock.c
> index 9e7659e..10e87e1 100644
> --- a/kernel/locking/qspinlock.c
> +++ b/kernel/locking/qspinlock.c
> @@ -227,6 +227,14 @@ static __always_inline int get_qlock(struct qspinlock *lock)
>  {
>  	struct __qspinlock *l = (void *)lock;
>  
> +#ifdef CONFIG_PARAVIRT_UNFAIR_LOCKS
> +	if (static_key_false(&paravirt_unfairlocks_enabled))
> +		/*
> +		 * Need to use atomic operation to get the lock when
> +		 * lock stealing can happen.
> +		 */
> +		return cmpxchg(&l->locked, 0, _Q_LOCKED_VAL) == 0;

That's missing {}.

> +#endif


>  	barrier();
>  	ACCESS_ONCE(l->locked) = _Q_LOCKED_VAL;
>  	barrier();


But no, what you want is:

static __always_inline bool virt_lock(struct qspinlock *lock)
{
#ifdef CONFIG_VIRT_MUCK
	if (static_key_false(&virt_unfairlocks_enabled)) {
		while (!queue_spin_trylock(lock))
			cpu_relax();

		return true;
	}
#else
	return false;
}


void queue_spin_lock_slowpath(struct qspinlock *lock, u32 val)
{
	if (virt_lock(lock))
		return;

	...
}
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html