Re: [PATCH v3] cpuidle: poll_state: Add time limit to poll_idle()

[Rik van Riel <riel@surriel.com>]

[-- Attachment #1: Type: text/plain, Size: 2160 bytes --]

On Thu, 2018-03-22 at 18:09 +0100, Rafael J. Wysocki wrote:
> 
> +++ linux-pm/drivers/cpuidle/poll_state.c
> @@ -10,6 +10,7 @@
>  #include <linux/sched/idle.h>
>  
>  #define POLL_IDLE_TIME_LIMIT	(TICK_NSEC / 16)
> +#define POLL_IDLE_COUNT		1000
>  
>  static int __cpuidle poll_idle(struct cpuidle_device *dev,
>  			       struct cpuidle_driver *drv, int
> index)
> @@ -18,9 +19,14 @@ static int __cpuidle poll_idle(struct cp
>  
>  	local_irq_enable();
>  	if (!current_set_polling_and_test()) {
> +		unsigned int loop_count = 0;
> +
>  		while (!need_resched()) {
>  			cpu_relax();
> +			if (loop_count++ < POLL_IDLE_COUNT)
> +				continue;
>  
> +			loop_count = 0;
>  			if (local_clock() - time_start >
> POLL_IDLE_TIME_LIMIT)
>  				break;
>  		}

OK, I am still seeing a performance
degradation with the above, though
not throughout the entire workload.

It appears that making the idle loop
do anything besides cpu_relax() for
a significant amount of time slows
things down.

I plan to try two more things:

1) Disable polling on SMT systems, with
   the idea that putting one thread to
   sleep with monitor/mwait in C1 will
   allow the other thread to run faster.

2) Insert more cpu_relax() calls into the
   main loop, so the CPU core spends more
   of its time in cpu_relax() and less
   time doing other things:

static int __cpuidle poll_idle(struct cpuidle_device *dev,
             struct cpuidle_driver *drv, int index)
{
  u64 time_start = local_clock();

  local_irq_enable();
  if (!current_set_polling_and_test()) {
    unsigned int loop_count = 0;

    while (!need_resched()) {
      cpu_relax();
      cpu_relax();
      cpu_relax();
      cpu_relax();
      cpu_relax();
      cpu_relax();
      cpu_relax();
      cpu_relax();
      if (loop_count++ < POLL_IDLE_COUNT)
        continue;

      loop_count = 0;
      if (local_clock() - time_start > POLL_IDLE_TIME_LIMIT)
        break;
    }
  }
  current_clr_polling();
                                         
  return index;
}

I will let you know how they perform.

-- 
All Rights Reversed.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 488 bytes --]