[RFC PATCH] mm/damon/core: use number of passed access sampling as a timer

[RFC PATCH] mm/damon/core: use number of passed access sampling as a timer

[SeongJae Park]

DAMON sleeps for sampling interval after each sampling, and check if
it's time for further doing aggregation and ops updating using
ktime_get_coarse_ts64() and baseline timestamps for the two periodic
operations.  That's for making the operations occur at deterministic
timing.  However, it turned out it could still result in indeterministic
and even not-that-intuitive results.

After all, timer functions, and especially sleep functions that DAMON
uses to wait for specific timing, could contain some errors.  Those
errors are legal, so no problem.  However, depending on such legal
timing errors, the nr_accesses can be larger than aggregation interval
divided by sampling interval.  For example, with the default setting (5
ms sampling interval and 100 ms aggregation interval) we frequently show
regions having nr_accesses larger than 20.  Also, if the execution of a
DAMOS scheme takes a long time, next aggregation could happen before
enough number of samples are collected.

Since access check sampling is the smallest unit work of DAMON, using
the number of passed sampling intervals as the DAMON-internal timer can
easily avoid these problems.  That is, convert aggregation and ops
update intervals to numbers of sampling intervals that need to be passed
before those operations be executed, count the number of passed sampling
intervals, and invoke the operations as soon as the specific amount of
sampling intervals passed.  Make the change.

Signed-off-by: SeongJae Park <sj@kernel.org>
---
 include/linux/damon.h | 14 ++++++--
 mm/damon/core.c       | 84 +++++++++++++++++++------------------------
 2 files changed, 48 insertions(+), 50 deletions(-)

diff --git a/include/linux/damon.h b/include/linux/damon.h
index ab3089de1478..9a32b8fd0bd3 100644
--- a/include/linux/damon.h
+++ b/include/linux/damon.h
@@ -524,8 +524,18 @@ struct damon_ctx {
 	struct damon_attrs attrs;
 
 /* private: internal use only */
-	struct timespec64 last_aggregation;
-	struct timespec64 last_ops_update;
+	/* number of sample intervals that passed since this context started */
+	unsigned long passed_sample_intervals;
+	/*
+	 * number of sample intervals that should be passed before next
+	 * aggregation
+	 */
+	unsigned long next_aggregation_sis;
+	/*
+	 * number of sample intervals that should be passed before next ops
+	 * update
+	 */
+	unsigned long next_ops_update_sis;
 
 /* public: */
 	struct task_struct *kdamond;
diff --git a/mm/damon/core.c b/mm/damon/core.c
index 988dc39e44b1..83af336bb0e6 100644
--- a/mm/damon/core.c
+++ b/mm/damon/core.c
@@ -456,8 +456,11 @@ struct damon_ctx *damon_new_ctx(void)
 	ctx->attrs.aggr_interval = 100 * 1000;
 	ctx->attrs.ops_update_interval = 60 * 1000 * 1000;
 
-	ktime_get_coarse_ts64(&ctx->last_aggregation);
-	ctx->last_ops_update = ctx->last_aggregation;
+	ctx->passed_sample_intervals = 0;
+	ctx->next_aggregation_sis = ctx->attrs.aggr_interval /
+		ctx->attrs.sample_interval;
+	ctx->next_ops_update_sis = ctx->attrs.ops_update_interval /
+		ctx->attrs.sample_interval;
 
 	mutex_init(&ctx->kdamond_lock);
 
@@ -577,6 +580,9 @@ static void damon_update_monitoring_results(struct damon_ctx *ctx,
  */
 int damon_set_attrs(struct damon_ctx *ctx, struct damon_attrs *attrs)
 {
+	unsigned long sample_interval;
+	unsigned long remaining_interval_us;
+
 	if (attrs->min_nr_regions < 3)
 		return -EINVAL;
 	if (attrs->min_nr_regions > attrs->max_nr_regions)
@@ -584,6 +590,20 @@ int damon_set_attrs(struct damon_ctx *ctx, struct damon_attrs *attrs)
 	if (attrs->sample_interval > attrs->aggr_interval)
 		return -EINVAL;
 
+	sample_interval = attrs->sample_interval ? attrs->sample_interval : 1;
+
+	/* adjust next_aggregation_sis */
+	remaining_interval_us = ctx->attrs.sample_interval *
+		(ctx->next_aggregation_sis - ctx->passed_sample_intervals);
+	ctx->next_aggregation_sis = ctx->passed_sample_intervals +
+		remaining_interval_us / sample_interval;
+
+	/* adjust next_ops_update_sis */
+	remaining_interval_us = ctx->attrs.sample_interval *
+		(ctx->next_ops_update_sis - ctx->passed_sample_intervals);
+	ctx->next_ops_update_sis = ctx->passed_sample_intervals +
+		remaining_interval_us / sample_interval;
+
 	damon_update_monitoring_results(ctx, attrs);
 	ctx->attrs = *attrs;
 	return 0;
@@ -757,38 +777,6 @@ int damon_stop(struct damon_ctx **ctxs, int nr_ctxs)
 	return err;
 }
 
-/*
- * damon_check_reset_time_interval() - Check if a time interval is elapsed.
- * @baseline:	the time to check whether the interval has elapsed since
- * @interval:	the time interval (microseconds)
- *
- * See whether the given time interval has passed since the given baseline
- * time.  If so, it also updates the baseline to current time for next check.
- *
- * Return:	true if the time interval has passed, or false otherwise.
- */
-static bool damon_check_reset_time_interval(struct timespec64 *baseline,
-		unsigned long interval)
-{
-	struct timespec64 now;
-
-	ktime_get_coarse_ts64(&now);
-	if ((timespec64_to_ns(&now) - timespec64_to_ns(baseline)) <
-			interval * 1000)
-		return false;
-	*baseline = now;
-	return true;
-}
-
-/*
- * Check whether it is time to flush the aggregated information
- */
-static bool kdamond_aggregate_interval_passed(struct damon_ctx *ctx)
-{
-	return damon_check_reset_time_interval(&ctx->last_aggregation,
-			ctx->attrs.aggr_interval);
-}
-
 /*
  * Reset the aggregated monitoring results ('nr_accesses' of each region).
  */
@@ -1292,18 +1280,6 @@ static void kdamond_split_regions(struct damon_ctx *ctx)
 	last_nr_regions = nr_regions;
 }
 
-/*
- * Check whether it is time to check and apply the operations-related data
- * structures.
- *
- * Returns true if it is.
- */
-static bool kdamond_need_update_operations(struct damon_ctx *ctx)
-{
-	return damon_check_reset_time_interval(&ctx->last_ops_update,
-			ctx->attrs.ops_update_interval);
-}
-
 /*
  * Check whether current monitoring should be stopped
  *
@@ -1436,6 +1412,8 @@ static int kdamond_fn(void *data)
 	sz_limit = damon_region_sz_limit(ctx);
 
 	while (!kdamond_need_stop(ctx)) {
+		unsigned long sample_interval;
+
 		if (kdamond_wait_activation(ctx))
 			break;
 
@@ -1446,11 +1424,17 @@ static int kdamond_fn(void *data)
 			break;
 
 		kdamond_usleep(ctx->attrs.sample_interval);
+		ctx->passed_sample_intervals++;
 
 		if (ctx->ops.check_accesses)
 			max_nr_accesses = ctx->ops.check_accesses(ctx);
 
-		if (kdamond_aggregate_interval_passed(ctx)) {
+		sample_interval = ctx->attrs.sample_interval ?
+			ctx->attrs.sample_interval : 1;
+		if (ctx->passed_sample_intervals ==
+				ctx->next_aggregation_sis) {
+			ctx->next_aggregation_sis +=
+				ctx->attrs.aggr_interval / sample_interval;
 			kdamond_merge_regions(ctx,
 					max_nr_accesses / 10,
 					sz_limit);
@@ -1465,7 +1449,11 @@ static int kdamond_fn(void *data)
 				ctx->ops.reset_aggregated(ctx);
 		}
 
-		if (kdamond_need_update_operations(ctx)) {
+		if (ctx->passed_sample_intervals ==
+				ctx->next_ops_update_sis) {
+			ctx->next_ops_update_sis +=
+				ctx->attrs.ops_update_interval /
+				sample_interval;
 			if (ctx->ops.update)
 				ctx->ops.update(ctx);
 			sz_limit = damon_region_sz_limit(ctx);
-- 
2.25.1

Re: [RFC PATCH] mm/damon/core: use number of passed access sampling as a timer

[SeongJae Park]

On Sun, 27 Aug 2023 00:37:27 +0000 SeongJae Park <sj@kernel.org> wrote:

> DAMON sleeps for sampling interval after each sampling, and check if
> it's time for further doing aggregation and ops updating using
> ktime_get_coarse_ts64() and baseline timestamps for the two periodic
> operations.  That's for making the operations occur at deterministic
> timing.  However, it turned out it could still result in indeterministic
> and even not-that-intuitive results.
> 
> After all, timer functions, and especially sleep functions that DAMON
> uses to wait for specific timing, could contain some errors.  Those
> errors are legal, so no problem.  However, depending on such legal
> timing errors, the nr_accesses can be larger than aggregation interval
> divided by sampling interval.  For example, with the default setting (5
> ms sampling interval and 100 ms aggregation interval) we frequently show
> regions having nr_accesses larger than 20.  Also, if the execution of a
> DAMOS scheme takes a long time, next aggregation could happen before
> enough number of samples are collected.
> 
> Since access check sampling is the smallest unit work of DAMON, using
> the number of passed sampling intervals as the DAMON-internal timer can
> easily avoid these problems.  That is, convert aggregation and ops
> update intervals to numbers of sampling intervals that need to be passed
> before those operations be executed, count the number of passed sampling
> intervals, and invoke the operations as soon as the specific amount of
> sampling intervals passed.  Make the change.
> 
> Signed-off-by: SeongJae Park <sj@kernel.org>
> ---
>  include/linux/damon.h | 14 ++++++--
>  mm/damon/core.c       | 84 +++++++++++++++++++------------------------
>  2 files changed, 48 insertions(+), 50 deletions(-)
> 
> diff --git a/include/linux/damon.h b/include/linux/damon.h
> index ab3089de1478..9a32b8fd0bd3 100644
> --- a/include/linux/damon.h
> +++ b/include/linux/damon.h
> @@ -524,8 +524,18 @@ struct damon_ctx {
>  	struct damon_attrs attrs;
>  
>  /* private: internal use only */
> -	struct timespec64 last_aggregation;
> -	struct timespec64 last_ops_update;
> +	/* number of sample intervals that passed since this context started */
> +	unsigned long passed_sample_intervals;
> +	/*
> +	 * number of sample intervals that should be passed before next
> +	 * aggregation
> +	 */
> +	unsigned long next_aggregation_sis;
> +	/*
> +	 * number of sample intervals that should be passed before next ops
> +	 * update
> +	 */
> +	unsigned long next_ops_update_sis;
>  
>  /* public: */
>  	struct task_struct *kdamond;
> diff --git a/mm/damon/core.c b/mm/damon/core.c
> index 988dc39e44b1..83af336bb0e6 100644
> --- a/mm/damon/core.c
> +++ b/mm/damon/core.c
> @@ -456,8 +456,11 @@ struct damon_ctx *damon_new_ctx(void)
>  	ctx->attrs.aggr_interval = 100 * 1000;
>  	ctx->attrs.ops_update_interval = 60 * 1000 * 1000;
>  
> -	ktime_get_coarse_ts64(&ctx->last_aggregation);
> -	ctx->last_ops_update = ctx->last_aggregation;
> +	ctx->passed_sample_intervals = 0;
> +	ctx->next_aggregation_sis = ctx->attrs.aggr_interval /
> +		ctx->attrs.sample_interval;
> +	ctx->next_ops_update_sis = ctx->attrs.ops_update_interval /
> +		ctx->attrs.sample_interval;
>  
>  	mutex_init(&ctx->kdamond_lock);
>  
> @@ -577,6 +580,9 @@ static void damon_update_monitoring_results(struct damon_ctx *ctx,
>   */
>  int damon_set_attrs(struct damon_ctx *ctx, struct damon_attrs *attrs)
>  {
> +	unsigned long sample_interval;
> +	unsigned long remaining_interval_us;
> +
>  	if (attrs->min_nr_regions < 3)
>  		return -EINVAL;
>  	if (attrs->min_nr_regions > attrs->max_nr_regions)
> @@ -584,6 +590,20 @@ int damon_set_attrs(struct damon_ctx *ctx, struct damon_attrs *attrs)
>  	if (attrs->sample_interval > attrs->aggr_interval)
>  		return -EINVAL;
>  
> +	sample_interval = attrs->sample_interval ? attrs->sample_interval : 1;
> +
> +	/* adjust next_aggregation_sis */
> +	remaining_interval_us = ctx->attrs.sample_interval *
> +		(ctx->next_aggregation_sis - ctx->passed_sample_intervals);
> +	ctx->next_aggregation_sis = ctx->passed_sample_intervals +
> +		remaining_interval_us / sample_interval;
> +
> +	/* adjust next_ops_update_sis */
> +	remaining_interval_us = ctx->attrs.sample_interval *
> +		(ctx->next_ops_update_sis - ctx->passed_sample_intervals);
> +	ctx->next_ops_update_sis = ctx->passed_sample_intervals +
> +		remaining_interval_us / sample_interval;
> +
>  	damon_update_monitoring_results(ctx, attrs);
>  	ctx->attrs = *attrs;
>  	return 0;
> @@ -757,38 +777,6 @@ int damon_stop(struct damon_ctx **ctxs, int nr_ctxs)
>  	return err;
>  }

So, the new timer variables are initialized in damon_new_ctx() as the default
(5ms sampling interval, 100ms aggregation interval, and 1s ops update interval)
and adjusted in damon_set_attrs().  It means the first intervals will be the
default ones always.  I will make those initialized at the beginning of
kdamond_fn() from the next spin.

[...]
> @@ -1436,6 +1412,8 @@ static int kdamond_fn(void *data)
>  	sz_limit = damon_region_sz_limit(ctx);

As mentioned abovely, the timer variables will be initialized around here (at
the beggining of kdamond_fn(), before going into the loop), in the next spin.


Thanks,
SJ

>  
>  	while (!kdamond_need_stop(ctx)) {
> +		unsigned long sample_interval;
> +
>  		if (kdamond_wait_activation(ctx))
>  			break;
>  
> @@ -1446,11 +1424,17 @@ static int kdamond_fn(void *data)
>  			break;
>  
>  		kdamond_usleep(ctx->attrs.sample_interval);
> +		ctx->passed_sample_intervals++;
>  
>  		if (ctx->ops.check_accesses)
>  			max_nr_accesses = ctx->ops.check_accesses(ctx);
>  
> -		if (kdamond_aggregate_interval_passed(ctx)) {
> +		sample_interval = ctx->attrs.sample_interval ?
> +			ctx->attrs.sample_interval : 1;
> +		if (ctx->passed_sample_intervals ==
> +				ctx->next_aggregation_sis) {
> +			ctx->next_aggregation_sis +=
> +				ctx->attrs.aggr_interval / sample_interval;
>  			kdamond_merge_regions(ctx,
>  					max_nr_accesses / 10,
>  					sz_limit);
> @@ -1465,7 +1449,11 @@ static int kdamond_fn(void *data)
>  				ctx->ops.reset_aggregated(ctx);
>  		}
>  
> -		if (kdamond_need_update_operations(ctx)) {
> +		if (ctx->passed_sample_intervals ==
> +				ctx->next_ops_update_sis) {
> +			ctx->next_ops_update_sis +=
> +				ctx->attrs.ops_update_interval /
> +				sample_interval;
>  			if (ctx->ops.update)
>  				ctx->ops.update(ctx);
>  			sz_limit = damon_region_sz_limit(ctx);
> -- 
> 2.25.1
> 
> 

Re: [RFC PATCH] mm/damon/core: use number of passed access sampling as a timer

[SeongJae Park]

On Sun, 27 Aug 2023 00:37:27 +0000 SeongJae Park <sj@kernel.org> wrote:

> DAMON sleeps for sampling interval after each sampling, and check if
> it's time for further doing aggregation and ops updating using
> ktime_get_coarse_ts64() and baseline timestamps for the two periodic
> operations.  That's for making the operations occur at deterministic
> timing.  However, it turned out it could still result in indeterministic
> and even not-that-intuitive results.
> 
> After all, timer functions, and especially sleep functions that DAMON
> uses to wait for specific timing, could contain some errors.  Those
> errors are legal, so no problem.  However, depending on such legal
> timing errors, the nr_accesses can be larger than aggregation interval
> divided by sampling interval.  For example, with the default setting (5
> ms sampling interval and 100 ms aggregation interval) we frequently show
> regions having nr_accesses larger than 20.  Also, if the execution of a
> DAMOS scheme takes a long time, next aggregation could happen before
> enough number of samples are collected.
> 
> Since access check sampling is the smallest unit work of DAMON, using
> the number of passed sampling intervals as the DAMON-internal timer can
> easily avoid these problems.  That is, convert aggregation and ops
> update intervals to numbers of sampling intervals that need to be passed
> before those operations be executed, count the number of passed sampling
> intervals, and invoke the operations as soon as the specific amount of
> sampling intervals passed.  Make the change.
> 
> Signed-off-by: SeongJae Park <sj@kernel.org>
> ---
>  include/linux/damon.h | 14 ++++++--
>  mm/damon/core.c       | 84 +++++++++++++++++++------------------------
>  2 files changed, 48 insertions(+), 50 deletions(-)
> 
[...]
> diff --git a/mm/damon/core.c b/mm/damon/core.c
> index 988dc39e44b1..83af336bb0e6 100644
> --- a/mm/damon/core.c
> +++ b/mm/damon/core.c
[...]
> @@ -577,6 +580,9 @@ static void damon_update_monitoring_results(struct damon_ctx *ctx,
>   */
>  int damon_set_attrs(struct damon_ctx *ctx, struct damon_attrs *attrs)
>  {
> +	unsigned long sample_interval;
> +	unsigned long remaining_interval_us;
> +
>  	if (attrs->min_nr_regions < 3)
>  		return -EINVAL;
>  	if (attrs->min_nr_regions > attrs->max_nr_regions)
> @@ -584,6 +590,20 @@ int damon_set_attrs(struct damon_ctx *ctx, struct damon_attrs *attrs)
>  	if (attrs->sample_interval > attrs->aggr_interval)
>  		return -EINVAL;
>  
> +	sample_interval = attrs->sample_interval ? attrs->sample_interval : 1;
> +
> +	/* adjust next_aggregation_sis */
> +	remaining_interval_us = ctx->attrs.sample_interval *
> +		(ctx->next_aggregation_sis - ctx->passed_sample_intervals);
> +	ctx->next_aggregation_sis = ctx->passed_sample_intervals +
> +		remaining_interval_us / sample_interval;
> +
> +	/* adjust next_ops_update_sis */
> +	remaining_interval_us = ctx->attrs.sample_interval *
> +		(ctx->next_ops_update_sis - ctx->passed_sample_intervals);
> +	ctx->next_ops_update_sis = ctx->passed_sample_intervals +
> +		remaining_interval_us / sample_interval;

The above remaining interval based adjustments mean we will respect the old
intervals for one more time.  However, user wants to change it right now, so
respecting it yet makes no sense.  Also, since this function is to be called
while no aggregation is ongoing, the remaining_interval_us for
next_aggregation_sis will be always zero.  So, simply ignoring past and
resetting the timings as below may make better sense.

    ctx->next_aggregation_sis = ctx->passed_sample_intervals +
    		attrs->aggr_interval / sample_interval;

    ctx->next_ops_update_sis = ctx->passed_sample_intervals +
    		attrs->ops_update_interval / sample_interval;

I will replace the adjustments as above.


Thanks,
SJ

[...]