Re: [PATCH v2] nptl: Add backoff mechanism to spinlock loop

[Adhemerval Zanella <adhemerval.zanella@linaro.org>]

On 28/03/2022 05:47, Wangyang Guo via Libc-alpha wrote:
> When mutiple threads waiting for lock at the same time, once lock owner
> releases the lock, waiters will see lock available and all try to lock,
> which may cause an expensive CAS storm.
> 
> Binary exponential backoff with random jitter is introduced. As try-lock
> attempt increases, there is more likely that a larger number threads
> compete for adaptive mutex lock, so increase wait time in exponential.
> A random jitter is also added to avoid synchronous try-lock from other
> threads.
> 
> v2: Remove read-check before try-lock for performance.
> 
> Signed-off-by: Wangyang Guo <wangyang.guo@intel.com>
> ---
>  nptl/pthread_mutex_lock.c | 25 ++++++++++++++++---------
>  1 file changed, 16 insertions(+), 9 deletions(-)
> 
> diff --git a/nptl/pthread_mutex_lock.c b/nptl/pthread_mutex_lock.c
> index d2e652d151..7e75ec1cba 100644
> --- a/nptl/pthread_mutex_lock.c
> +++ b/nptl/pthread_mutex_lock.c
> @@ -26,6 +26,7 @@
>  #include <futex-internal.h>
>  #include <stap-probe.h>
>  #include <shlib-compat.h>
> +#include <random-bits.h>
>  
>  /* Some of the following definitions differ when pthread_mutex_cond_lock.c
>     includes this file.  */
> @@ -64,11 +65,6 @@ lll_mutex_lock_optimized (pthread_mutex_t *mutex)
>  # define PTHREAD_MUTEX_VERSIONS 1
>  #endif
>  
> -#ifndef LLL_MUTEX_READ_LOCK
> -# define LLL_MUTEX_READ_LOCK(mutex) \
> -  atomic_load_relaxed (&(mutex)->__data.__lock)
> -#endif
> -
>  static int __pthread_mutex_lock_full (pthread_mutex_t *mutex)
>       __attribute_noinline__;
>  
> @@ -138,17 +134,28 @@ PTHREAD_MUTEX_LOCK (pthread_mutex_t *mutex)
>  	  int cnt = 0;
>  	  int max_cnt = MIN (max_adaptive_count (),
>  			     mutex->__data.__spins * 2 + 10);
> +	  int spin_count, exp_backoff = 1;
> +	  unsigned int jitter = random_bits ();

This will issue a syscall for architectures that do not have clock_gettime
on vDSO, which is a performance regression.  You will need to move the
jitter setup to be arch-specific, where the generic interface setting
no random jitter.

>  	  do
>  	    {
> -	      if (cnt++ >= max_cnt)
> +	      /* In each loop, spin count is exponential backoff plus
> +	         random jitter, random range is [0, exp_backoff-1].  */
> +	      spin_count = exp_backoff + (jitter & (exp_backoff - 1));
> +	      cnt += spin_count;
> +	      if (cnt >= max_cnt)
>  		{
> +		  /* If cnt exceeds max spin count, just go to wait
> +		     queue.  */
>  		  LLL_MUTEX_LOCK (mutex);
>  		  break;
>  		}
> -	      atomic_spin_nop ();
> +	      do
> +		  atomic_spin_nop ();
> +	      while (--spin_count > 0);
> +	      /* Binary exponential backoff, prepare for next loop.  */
> +	      exp_backoff <<= 1;
>  	    }
> -	  while (LLL_MUTEX_READ_LOCK (mutex) != 0
> -		 || LLL_MUTEX_TRYLOCK (mutex) != 0);
> +	  while (LLL_MUTEX_TRYLOCK (mutex) != 0);
>  
>  	  mutex->__data.__spins += (cnt - mutex->__data.__spins) / 8;
>  	}