Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Wed, 11 Apr 2018 10:07:26 +0530
From:   Viresh Kumar <viresh.kumar@linaro.org>
To:     Patrick Bellasi <patrick.bellasi@arm.com>
Cc:     linux-kernel@vger.kernel.org, linux-pm@vger.kernel.org,
        Ingo Molnar <mingo@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        "Rafael J . Wysocki" <rafael.j.wysocki@intel.com>,
        Joel Fernandes <joelaf@google.com>,
        Steve Muckle <smuckle@google.com>,
        Juri Lelli <juri.lelli@redhat.com>,
        Dietmar Eggemann <dietmar.eggemann@arm.com>
Subject: Re: [PATCH v2] cpufreq/schedutil: Cleanup, document and fix iowait
 boost
Message-ID: <20180411043726.GJ7671@vireshk-i7>
References: <20180410155931.31973-1-patrick.bellasi@arm.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20180410155931.31973-1-patrick.bellasi@arm.com>
User-Agent: Mutt/1.5.24 (2015-08-30)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 10-04-18, 16:59, Patrick Bellasi wrote:
> The iowait boosting code has been recently updated to add a progressive
> boosting behavior which allows to be less aggressive in boosting tasks
> doing only sporadic IO operations, thus being more energy efficient for
> example on mobile platforms.
> 
> The current code is now however a bit convoluted. Some functionalities
> (e.g. iowait boost reset) are replicated in different paths and their
> documentation is slightly misaligned.
> 
> Moreover, from a functional stadpoint, the iowait boosting is also not
> always reset in systems where cpufreq policies are not shared, each CPU
> has his own policy. Indeed, when a CPU is idle for a long time we keep
> doubling the boosting instead of resetting it to the minimum frequency,
> as expected by the TICK_NSEC logic, whenever a task wakes up from IO.
> 
> Let's cleanup the code by consolidating all the IO wait boosting related
> functionality inside the already existing functions and better define
> their role:
> 
> - sugov_set_iowait_boost: is now in charge only to set/increase the IO
>      wait boost, every time a task wakes up from an IO wait.
> 
> - sugov_iowait_boost: is now in charge to reset/reduce the IO wait
>      boost, every time a sugov update is triggered, as well as
>      to (eventually) enforce the currently required IO boost value.
> 
> This is possible since these two functions are already used one after
> the other, both in single and shared frequency domains, following the
> same template:
> 
>    /* Configure IO boost, if required */
>    sugov_set_iowait_boost()
> 
>    /* Return here if freq change is in progress or throttled */
> 
>    /* Collect and aggregate utilization information */
>    sugov_get_util()
>    sugov_aggregate_util()
> 
>    /* Add IO boost if currently enabled */
>    sugov_iowait_boost()
> 
> As a extra bonus, let's also add the documentation for these two
> functions and better align the in-code documentation.
> 
> Signed-off-by: Patrick Bellasi <patrick.bellasi@arm.com>
> Reported-by: Viresh Kumar <viresh.kumar@linaro.org>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: Peter Zijlstra <peterz@infradead.org>
> Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> Cc: Viresh Kumar <viresh.kumar@linaro.org>
> Cc: Joel Fernandes <joelaf@google.com>
> Cc: Steve Muckle <smuckle@google.com>
> Cc: Juri Lelli <juri.lelli@redhat.com>
> Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
> Cc: linux-kernel@vger.kernel.org
> Cc: linux-pm@vger.kernel.org
> 
> ---
> Changes in v2:
> - Fix return in sugov_iowait_boost()'s reset code (Viresh)
> - Add iowait boost reset for sugov_update_single() (Viresh)
> - Title changed to reflact the fix from previous point
> 
> Based on today's tip/sched/core:
>    b720342 sched/core: Update preempt_notifier_key to modern API
> ---
>  kernel/sched/cpufreq_schedutil.c | 120 ++++++++++++++++++++++++++-------------
>  1 file changed, 81 insertions(+), 39 deletions(-)
> 
> diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
> index 2b124811947d..2a2ae3a0e41f 100644
> --- a/kernel/sched/cpufreq_schedutil.c
> +++ b/kernel/sched/cpufreq_schedutil.c
> @@ -51,7 +51,7 @@ struct sugov_cpu {
>  	bool			iowait_boost_pending;
>  	unsigned int		iowait_boost;
>  	unsigned int		iowait_boost_max;
> -	u64 last_update;
> +	u64			last_update;
>  
>  	/* The fields below are only needed when sharing a policy: */
>  	unsigned long		util_cfs;
> @@ -201,43 +201,97 @@ static unsigned long sugov_aggregate_util(struct sugov_cpu *sg_cpu)
>  	return min(util, sg_cpu->max);
>  }
>  
> -static void sugov_set_iowait_boost(struct sugov_cpu *sg_cpu, u64 time, unsigned int flags)
> +/**
> + * sugov_set_iowait_boost updates the IO boost at each wakeup from IO.
> + * @sg_cpu: the sugov data for the CPU to boost
> + * @time:   the update time from the caller
> + * @flags:  SCHED_CPUFREQ_IOWAIT if the task is waking up after an IO wait
> + *
> + * Each time a task wakes up after an IO operation, the CPU utilization can be
> + * boosted to a certain utilization which is doubled at each wakeup
> + * from IO, starting from the utilization of the minimum OPP to that of the
> + * maximum one.
> + */
> +static void sugov_set_iowait_boost(struct sugov_cpu *sg_cpu, u64 time,
> +				   unsigned int flags)
>  {
> -	if (flags & SCHED_CPUFREQ_IOWAIT) {
> -		if (sg_cpu->iowait_boost_pending)
> -			return;
> -
> -		sg_cpu->iowait_boost_pending = true;
> +	bool iowait = flags & SCHED_CPUFREQ_IOWAIT;
>  
> -		if (sg_cpu->iowait_boost) {
> -			sg_cpu->iowait_boost <<= 1;
> -			if (sg_cpu->iowait_boost > sg_cpu->iowait_boost_max)
> -				sg_cpu->iowait_boost = sg_cpu->iowait_boost_max;
> -		} else {
> -			sg_cpu->iowait_boost = sg_cpu->sg_policy->policy->min;
> -		}
> -	} else if (sg_cpu->iowait_boost) {
> +	/* Reset boost if the CPU appears to have been idle enough */
> +	if (sg_cpu->iowait_boost) {
>  		s64 delta_ns = time - sg_cpu->last_update;
>  
> -		/* Clear iowait_boost if the CPU apprears to have been idle. */
>  		if (delta_ns > TICK_NSEC) {
> -			sg_cpu->iowait_boost = 0;
> -			sg_cpu->iowait_boost_pending = false;
> +			sg_cpu->iowait_boost = iowait
> +				? sg_cpu->sg_policy->policy->min : 0;
> +			sg_cpu->iowait_boost_pending = iowait;
> +			return;
>  		}
>  	}
> +
> +	/* Boost only tasks waking up after IO */
> +	if (!iowait)
> +		return;
> +
> +	/* Ensure IO boost doubles only one time at each frequency increase */
> +	if (sg_cpu->iowait_boost_pending)
> +		return;
> +	sg_cpu->iowait_boost_pending = true;
> +
> +	/* Double the IO boost at each frequency increase */
> +	if (sg_cpu->iowait_boost) {
> +		sg_cpu->iowait_boost <<= 1;
> +		if (sg_cpu->iowait_boost > sg_cpu->iowait_boost_max)
> +			sg_cpu->iowait_boost = sg_cpu->iowait_boost_max;
> +		return;
> +	}
> +
> +	/* At first wakeup after IO, start with minimum boost */
> +	sg_cpu->iowait_boost = sg_cpu->sg_policy->policy->min;
>  }

The above part should be a different patch with this:

Fixes: a5a0809bc58e ("cpufreq: schedutil: Make iowait boost more energy efficient")

> -static void sugov_iowait_boost(struct sugov_cpu *sg_cpu, unsigned long *util,
> -			       unsigned long *max)
> +/**
> + * sugov_iowait_boost boosts a CPU after a wakeup from IO.
> + * @sg_cpu: the sugov data for the cpu to boost
> + * @time:   the update time from the caller
> + * @util:   the utilization to (eventually) boost
> + * @max:    the maximum value the utilization can be boosted to
> + *
> + * A CPU running a task which woken up after an IO operation can have its
> + * utilization boosted to speed up the completion of those IO operations.
> + * The IO boost value is increased each time a task wakes up from IO, in
> + * sugov_set_iowait_boost(), and it's instead decreased by this function,
> + * each time an increase has not been requested (!iowait_boost_pending).
> + *
> + * A CPU which also appears to have been idle for at least one tick has also
> + * its IO boost utilization reset.
> + *
> + * This mechanism is designed to boost high frequently IO waiting tasks, while
> + * being more conservative on tasks which does sporadic IO operations.
> + */
> +static void sugov_iowait_boost(struct sugov_cpu *sg_cpu, u64 time,
> +			       unsigned long *util, unsigned long *max)
>  {
>  	unsigned int boost_util, boost_max;
> +	s64 delta_ns;
>  
> +	/* No IOWait boost active */
>  	if (!sg_cpu->iowait_boost)
>  		return;
>  
> +	/* Clear boost if the CPU appears to have been idle enough */
> +	delta_ns = time - sg_cpu->last_update;
> +	if (delta_ns > TICK_NSEC) {
> +		sg_cpu->iowait_boost = 0;
> +		sg_cpu->iowait_boost_pending = false;
> +		return;
> +	}
> +
> +	/* An IO waiting task has just woken up, use the boost value */
>  	if (sg_cpu->iowait_boost_pending) {
>  		sg_cpu->iowait_boost_pending = false;
>  	} else {
> +		/* Reduce the boost value otherwise */
>  		sg_cpu->iowait_boost >>= 1;
>  		if (sg_cpu->iowait_boost < sg_cpu->sg_policy->policy->min) {
>  			sg_cpu->iowait_boost = 0;
> @@ -248,6 +302,10 @@ static void sugov_iowait_boost(struct sugov_cpu *sg_cpu, unsigned long *util,
>  	boost_util = sg_cpu->iowait_boost;
>  	boost_max = sg_cpu->iowait_boost_max;
>  
> +	/*
> +	 * A CPU is boosted only if its current utilization is smaller then
> +	 * the current IO boost level.
> +	 */
>  	if (*util * boost_max < *max * boost_util) {
>  		*util = boost_util;
>  		*max = boost_max;
> @@ -299,7 +357,7 @@ static void sugov_update_single(struct update_util_data *hook, u64 time,
>  	sugov_get_util(sg_cpu);
>  	max = sg_cpu->max;
>  	util = sugov_aggregate_util(sg_cpu);
> -	sugov_iowait_boost(sg_cpu, &util, &max);
> +	sugov_iowait_boost(sg_cpu, time, &util, &max);
>  	next_f = get_next_freq(sg_policy, util, max);
>  	/*
>  	 * Do not reduce the frequency if the CPU has not been idle
> @@ -325,28 +383,12 @@ static unsigned int sugov_next_freq_shared(struct sugov_cpu *sg_cpu, u64 time)
>  	for_each_cpu(j, policy->cpus) {
>  		struct sugov_cpu *j_sg_cpu = &per_cpu(sugov_cpu, j);
>  		unsigned long j_util, j_max;
> -		s64 delta_ns;
>  
>  		sugov_get_util(j_sg_cpu);
> -
> -		/*
> -		 * If the CFS CPU utilization was last updated before the
> -		 * previous frequency update and the time elapsed between the
> -		 * last update of the CPU utilization and the last frequency
> -		 * update is long enough, reset iowait_boost and util_cfs, as
> -		 * they are now probably stale. However, still consider the
> -		 * CPU contribution if it has some DEADLINE utilization
> -		 * (util_dl).
> -		 */
> -		delta_ns = time - j_sg_cpu->last_update;
> -		if (delta_ns > TICK_NSEC) {
> -			j_sg_cpu->iowait_boost = 0;
> -			j_sg_cpu->iowait_boost_pending = false;
> -		}
> -
>  		j_max = j_sg_cpu->max;
>  		j_util = sugov_aggregate_util(j_sg_cpu);
> -		sugov_iowait_boost(j_sg_cpu, &j_util, &j_max);
> +		sugov_iowait_boost(j_sg_cpu, time, &j_util, &j_max);
> +
>  		if (j_util * max > j_max * util) {
>  			util = j_util;
>  			max = j_max;

And the rest is just code rearrangement. And as Peter said, we better
have a routine to clear boost values on delta > TICK_NSEC.

Diff LGTM otherwise. Thanks.

-- 
viresh