Re: [PATCH] locking/rtmutex: remove unused cmpxchg_relaxed

[Davidlohr Bueso]

Cc'ing tglx.

On Tue, 21 Jan 2020, Alex Shi wrote:

> No one use this macro after it was introduced. Better to remove it?

You also need to remove it for the CONFIG_DEBUG_RT_MUTEXES=y case.

> Signed-off-by: Alex Shi <alex.shi@xxxxxxxxxxxxxxxxx>
> Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> Cc: Davidlohr Bueso <dave@xxxxxxxxxxxx>
> Cc: Ingo Molnar <mingo@xxxxxxxxxx>
> Cc: Will Deacon <will@xxxxxxxxxx>
> Cc: linux-kernel@xxxxxxxxxxxxxxx
> ---
> kernel/locking/rtmutex.c | 1 -
> 1 file changed, 1 deletion(-)
>
> diff --git a/kernel/locking/rtmutex.c b/kernel/locking/rtmutex.c
> index 851bbb10819d..a849964a348d 100644
> --- a/kernel/locking/rtmutex.c
> +++ b/kernel/locking/rtmutex.c
> @@ -141,7 +141,6 @@ static void fixup_rt_mutex_waiters(struct rt_mutex *lock)
>  * set up.
>  */
> #ifndef CONFIG_DEBUG_RT_MUTEXES
> -# define rt_mutex_cmpxchg_relaxed(l,c,n) (cmpxchg_relaxed(&l->owner, c, n) == c)
> # define rt_mutex_cmpxchg_acquire(l,c,n) (cmpxchg_acquire(&l->owner, c, n) == c)
> # define rt_mutex_cmpxchg_release(l,c,n) (cmpxchg_release(&l->owner, c, n) == c)

Hmm unrelated, but do we want CCAS for rtmutex fastpath? Ie:

    (l->owner == c && cmpxchg_acquire(&l->owner, c, n) == c)

That would optimize for the contended case and avoid the cmpxchg - it would
also help if we ever do the top-waiter spin thing.

Thanks,
Davidlohr