On Thu, May 25, 2017 at 02:58:53PM -0700, Paul E. McKenney wrote:
> There is material describing the ordering guarantees provided by
> spin_unlock_wait(), but it is not necessarily easy to find. This commit
> therefore adds a docbook header comment to this function informally
> describing its semantics.
>
> Signed-off-by: Paul E. McKenney <[email protected]>
> Acked-by: Peter Zijlstra <[email protected]>
> ---
> include/linux/spinlock.h | 20 ++++++++++++++++++++
> 1 file changed, 20 insertions(+)
>
> diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h
> index 59248dcc6ef3..d9510e8522d4 100644
> --- a/include/linux/spinlock.h
> +++ b/include/linux/spinlock.h
> @@ -369,6 +369,26 @@ static __always_inline int spin_trylock_irq(spinlock_t *lock)
> raw_spin_trylock_irqsave(spinlock_check(lock), flags); \
> })
>
> +/**
> + * spin_unlock_wait - Interpose between successive critical sections
> + * @lock: the spinlock whose critical sections are to be interposed.
> + *
> + * Semantically this is equivalent to a spin_lock() immediately
> + * followed by a spin_unlock(). However, most architectures have
> + * more efficient implementations in which the spin_unlock_wait()
> + * cannot block concurrent lock acquisition, and in some cases
> + * where spin_unlock_wait() does not write to the lock variable.
> + * Nevertheless, spin_unlock_wait() can have high overhead, so if
> + * you feel the need to use it, please check to see if there is
> + * a better way to get your job done.
> + *
> + * The ordering guarantees provided by spin_unlock_wait() are:
> + *
> + * 1. All accesses preceding the spin_unlock_wait() happen before
> + * any accesses in later critical sections for this same lock.
> + * 2. All accesses following the spin_unlock_wait() happen after
> + * any accesses in earlier critical sections for this same lock.
> + */
[From a discussion with Paul, Alan]
I understand that some implementation would need to "be strengthened" to
meet this "spin_lock(); spin_unlock()" semantics; please compare with
726328d92a42b6d4b76078e2659f43067f82c4e8
("locking/spinlock, arch: Update and fix spin_unlock_wait() implementations")
Should we "relax" this description? Should we integrate it with changes
to the implementation(s)? [...] What do you think?
Andrea
> static __always_inline void spin_unlock_wait(spinlock_t *lock)
> {
> raw_spin_unlock_wait(&lock->rlock);
> --
> 2.5.2
>