Re: [PATCH v2 02/13] rwsem: shorter spinlocked section in rwsem_down_failed_common()

[Peter Hurley <peter@hurleysoftware.com>]

On Fri, 2013-03-15 at 03:54 -0700, Michel Lespinasse wrote:
> This change reduces the size of the spinlocked and TASK_UNINTERRUPTIBLE
> sections in rwsem_down_failed_common():
> 
> - We only need the sem->wait_lock to insert ourselves on the wait_list;
>   the waiter node can be prepared outside of the wait_lock.
> 
> - The task state only needs to be set to TASK_UNINTERRUPTIBLE immediately
>   before checking if we actually need to sleep; it doesn't need to protect
>   the entire function.
> 
> Signed-off-by: Michel Lespinasse <walken@google.com>
> 
> ---
>  lib/rwsem.c | 8 +++-----
>  1 file changed, 3 insertions(+), 5 deletions(-)
> 
> diff --git a/lib/rwsem.c b/lib/rwsem.c
> index 672eb33218ac..40636454cf3c 100644
> --- a/lib/rwsem.c
> +++ b/lib/rwsem.c
> @@ -188,14 +188,12 @@ rwsem_down_failed_common(struct rw_semaphore *sem,
>  	struct task_struct *tsk = current;
>  	signed long count;
>  
> -	set_task_state(tsk, TASK_UNINTERRUPTIBLE);
> -
>  	/* set up my own style of waitqueue */
> -	raw_spin_lock_irq(&sem->wait_lock);
>  	waiter.task = tsk;
>  	waiter.type = type;
>  	get_task_struct(tsk);
>  
> +	raw_spin_lock_irq(&sem->wait_lock);
>  	if (list_empty(&sem->wait_list))
>  		adjustment += RWSEM_WAITING_BIAS;
>  	list_add_tail(&waiter.list, &sem->wait_list);
> @@ -218,7 +216,8 @@ rwsem_down_failed_common(struct rw_semaphore *sem,
>  	raw_spin_unlock_irq(&sem->wait_lock);
>  
>  	/* wait to be given the lock */
> -	for (;;) {
> +	while (true) {

I would drop this gratutious change.

> +		set_task_state(tsk, TASK_UNINTERRUPTIBLE);
>  		if (!waiter.task)
>  			break;
>  
> @@ -231,7 +230,6 @@ rwsem_down_failed_common(struct rw_semaphore *sem,
>  			}
>  		raw_spin_unlock_irq(&sem->wait_lock);
>  		schedule();
> -		set_task_state(tsk, TASK_UNINTERRUPTIBLE);
>  	}
>  
>  	tsk->state = TASK_RUNNING;