From: SeongJae Park <[email protected]>
Even somehow the initial monitoring target regions are well constructed
to fulfill the assumption (pages in same region have similar access
frequencies), the data access pattern can be dynamically changed. This
will result in low monitoring quality. To keep the assumption as much
as possible, DAMON adaptively merges and splits each region.
For each ``aggregation interval``, it compares the access frequencies of
adjacent regions and merges those if the frequency difference is small.
Then, after it reports and clears the aggregated access frequency of
each region, it splits each region into two or third regions of random
size, if the total number of regions after the splits wouldn't exceed
the user-specified maximum number of regions.
In this way, DAMON provides its best-effort quality and minimal overhead
while keeping the overhead bound.
Signed-off-by: SeongJae Park <[email protected]>
---
include/linux/damon.h | 7 +-
mm/damon.c | 181 +++++++++++++++++++++++++++++++++++++++++-
2 files changed, 182 insertions(+), 6 deletions(-)
diff --git a/include/linux/damon.h b/include/linux/damon.h
index 649fb8b6209f..9588bc162c3a 100644
--- a/include/linux/damon.h
+++ b/include/linux/damon.h
@@ -65,6 +65,7 @@ struct damon_task {
* @sample_interval: The time between access samplings.
* @aggr_interval: The time between monitor results aggregations.
* @min_nr_regions: The number of initial monitoring regions.
+ * @max_nr_regions: The maximum number of monitoring regions.
*
* For each @sample_interval, DAMON checks whether each region is accessed or
* not. It aggregates and keeps the access information (number of accesses to
@@ -115,6 +116,7 @@ struct damon_ctx {
unsigned long sample_interval;
unsigned long aggr_interval;
unsigned long min_nr_regions;
+ unsigned long max_nr_regions;
struct timespec64 last_aggregation;
@@ -133,8 +135,9 @@ struct damon_ctx {
};
int damon_set_pids(struct damon_ctx *ctx, int *pids, ssize_t nr_pids);
-int damon_set_attrs(struct damon_ctx *ctx, unsigned long sample_int,
- unsigned long aggr_int, unsigned long min_nr_reg);
+int damon_set_attrs(struct damon_ctx *ctx,
+ unsigned long sample_int, unsigned long aggr_int,
+ unsigned long min_nr_reg, unsigned long max_nr_reg);
int damon_start(struct damon_ctx *ctx);
int damon_stop(struct damon_ctx *ctx);
diff --git a/mm/damon.c b/mm/damon.c
index aba02c652b51..5cf39b3ad222 100644
--- a/mm/damon.c
+++ b/mm/damon.c
@@ -176,6 +176,26 @@ static unsigned int nr_damon_regions(struct damon_task *t)
return nr_regions;
}
+/* Returns the size upper limit for each monitoring region */
+static unsigned long damon_region_sz_limit(struct damon_ctx *ctx)
+{
+ struct damon_task *t;
+ struct damon_region *r;
+ unsigned long sz = 0;
+
+ damon_for_each_task(t, ctx) {
+ damon_for_each_region(r, t)
+ sz += r->ar.end - r->ar.start;
+ }
+
+ if (ctx->min_nr_regions)
+ sz /= ctx->min_nr_regions;
+ if (sz < MIN_REGION)
+ sz = MIN_REGION;
+
+ return sz;
+}
+
/*
* Functions for DAMON core logics and features
*/
@@ -226,6 +246,145 @@ static void kdamond_reset_aggregated(struct damon_ctx *c)
}
}
+#define sz_damon_region(r) (r->ar.end - r->ar.start)
+
+/*
+ * Merge two adjacent regions into one region
+ */
+static void damon_merge_two_regions(struct damon_region *l,
+ struct damon_region *r)
+{
+ l->nr_accesses = (l->nr_accesses * sz_damon_region(l) +
+ r->nr_accesses * sz_damon_region(r)) /
+ (sz_damon_region(l) + sz_damon_region(r));
+ l->ar.end = r->ar.end;
+ damon_destroy_region(r);
+}
+
+#define diff_of(a, b) (a > b ? a - b : b - a)
+
+/*
+ * Merge adjacent regions having similar access frequencies
+ *
+ * t task affected by merge operation
+ * thres '->nr_accesses' diff threshold for the merge
+ * sz_limit size upper limit of each region
+ */
+static void damon_merge_regions_of(struct damon_task *t, unsigned int thres,
+ unsigned long sz_limit)
+{
+ struct damon_region *r, *prev = NULL, *next;
+
+ damon_for_each_region_safe(r, next, t) {
+ if (prev && prev->ar.end == r->ar.start &&
+ diff_of(prev->nr_accesses, r->nr_accesses) <= thres &&
+ sz_damon_region(prev) + sz_damon_region(r) <= sz_limit)
+ damon_merge_two_regions(prev, r);
+ else
+ prev = r;
+ }
+}
+
+/*
+ * Merge adjacent regions having similar access frequencies
+ *
+ * threshold '->nr_accesses' diff threshold for the merge
+ * sz_limit size upper limit of each region
+ *
+ * This function merges monitoring target regions which are adjacent and their
+ * access frequencies are similar. This is for minimizing the monitoring
+ * overhead under the dynamically changeable access pattern. If a merge was
+ * unnecessarily made, later 'kdamond_split_regions()' will revert it.
+ */
+static void kdamond_merge_regions(struct damon_ctx *c, unsigned int threshold,
+ unsigned long sz_limit)
+{
+ struct damon_task *t;
+
+ damon_for_each_task(t, c)
+ damon_merge_regions_of(t, threshold, sz_limit);
+}
+
+/*
+ * Split a region in two
+ *
+ * r the region to be split
+ * sz_r size of the first sub-region that will be made
+ */
+static void damon_split_region_at(struct damon_ctx *ctx,
+ struct damon_region *r, unsigned long sz_r)
+{
+ struct damon_region *new;
+
+ new = damon_new_region(ctx, r->ar.start + sz_r, r->ar.end);
+ r->ar.end = new->ar.start;
+
+ damon_insert_region(new, r, damon_next_region(r));
+}
+
+/* Split every region in the given task into 'nr_subs' regions */
+static void damon_split_regions_of(struct damon_ctx *ctx,
+ struct damon_task *t, int nr_subs)
+{
+ struct damon_region *r, *next;
+ unsigned long sz_region, sz_sub = 0;
+ int i;
+
+ damon_for_each_region_safe(r, next, t) {
+ sz_region = r->ar.end - r->ar.start;
+
+ for (i = 0; i < nr_subs - 1 &&
+ sz_region > 2 * MIN_REGION; i++) {
+ /*
+ * Randomly select size of left sub-region to be at
+ * least 10 percent and at most 90% of original region
+ */
+ sz_sub = ALIGN_DOWN(damon_rand(1, 10) *
+ sz_region / 10, MIN_REGION);
+ /* Do not allow blank region */
+ if (sz_sub == 0 || sz_sub >= sz_region)
+ continue;
+
+ damon_split_region_at(ctx, r, sz_sub);
+ sz_region = sz_sub;
+ }
+ }
+}
+
+/*
+ * splits every target region into two randomly-sized regions
+ *
+ * This function splits every target region into two random-sized regions if
+ * current total number of the regions is equal or smaller than half of the
+ * user-specified maximum number of regions. This is for maximizing the
+ * monitoring accuracy under the dynamically changeable access patterns. If a
+ * split was unnecessarily made, later 'kdamond_merge_regions()' will revert
+ * it.
+ */
+static void kdamond_split_regions(struct damon_ctx *ctx)
+{
+ struct damon_task *t;
+ unsigned int nr_regions = 0;
+ static unsigned int last_nr_regions;
+ int nr_subregions = 2;
+
+ damon_for_each_task(t, ctx)
+ nr_regions += nr_damon_regions(t);
+
+ if (nr_regions > ctx->max_nr_regions / 2)
+ return;
+
+ /* If number of regions is not changed, we are maybe in corner case */
+ if (last_nr_regions == nr_regions &&
+ nr_regions < ctx->max_nr_regions / 3)
+ nr_subregions = 3;
+
+ damon_for_each_task(t, ctx)
+ damon_split_regions_of(ctx, t, nr_subregions);
+
+ last_nr_regions = nr_regions;
+}
+
/*
* Check whether current monitoring should be stopped
*
@@ -269,10 +428,14 @@ static int kdamond_fn(void *data)
struct damon_ctx *ctx = (struct damon_ctx *)data;
struct damon_task *t;
struct damon_region *r, *next;
+ unsigned int max_nr_accesses = 0;
+ unsigned long sz_limit = 0;
pr_info("kdamond (%d) starts\n", ctx->kdamond->pid);
if (ctx->init_target_regions)
ctx->init_target_regions(ctx);
+ sz_limit = damon_region_sz_limit(ctx);
+
while (!kdamond_need_stop(ctx)) {
if (ctx->prepare_access_checks)
ctx->prepare_access_checks(ctx);
@@ -282,14 +445,16 @@ static int kdamond_fn(void *data)
usleep_range(ctx->sample_interval, ctx->sample_interval + 1);
if (ctx->check_accesses)
- ctx->check_accesses(ctx);
+ max_nr_accesses = ctx->check_accesses(ctx);
if (kdamond_aggregate_interval_passed(ctx)) {
if (ctx->aggregate_cb)
ctx->aggregate_cb(ctx);
+ kdamond_merge_regions(ctx, max_nr_accesses / 10,
+ sz_limit);
kdamond_reset_aggregated(ctx);
+ kdamond_split_regions(ctx);
}
-
}
damon_for_each_task(t, ctx) {
damon_for_each_region_safe(r, next, t)
@@ -399,24 +564,32 @@ int damon_set_pids(struct damon_ctx *ctx, int *pids, ssize_t nr_pids)
* @sample_int: time interval between samplings
* @aggr_int: time interval between aggregations
* @min_nr_reg: minimal number of regions
+ * @max_nr_reg: maximum number of regions
*
* This function should not be called while the kdamond is running.
* Every time interval is in micro-seconds.
*
* Return: 0 on success, negative error code otherwise.
*/
-int damon_set_attrs(struct damon_ctx *ctx, unsigned long sample_int,
- unsigned long aggr_int, unsigned long min_nr_reg)
+int damon_set_attrs(struct damon_ctx *ctx,
+ unsigned long sample_int, unsigned long aggr_int,
+ unsigned long min_nr_reg, unsigned long max_nr_reg)
{
if (min_nr_reg < 3) {
pr_err("min_nr_regions (%lu) must be at least 3\n",
min_nr_reg);
return -EINVAL;
}
+ if (min_nr_reg > max_nr_reg) {
+ pr_err("invalid nr_regions. min (%lu) > max (%lu)\n",
+ min_nr_reg, max_nr_reg);
+ return -EINVAL;
+ }
ctx->sample_interval = sample_int;
ctx->aggr_interval = aggr_int;
ctx->min_nr_regions = min_nr_reg;
+ ctx->max_nr_regions = max_nr_reg;
return 0;
}
--
2.17.1
On 2020-06-15T18:19:17+02:00 SeongJae Park <[email protected]> wrote:
> From: SeongJae Park <[email protected]>
>
> Even somehow the initial monitoring target regions are well constructed
> to fulfill the assumption (pages in same region have similar access
> frequencies), the data access pattern can be dynamically changed. This
> will result in low monitoring quality. To keep the assumption as much
> as possible, DAMON adaptively merges and splits each region.
>
> For each ``aggregation interval``, it compares the access frequencies of
> adjacent regions and merges those if the frequency difference is small.
> Then, after it reports and clears the aggregated access frequency of
> each region, it splits each region into two or third regions of random
> size, if the total number of regions after the splits wouldn't exceed
> the user-specified maximum number of regions.
>
> In this way, DAMON provides its best-effort quality and minimal overhead
> while keeping the overhead bound.
>
> Signed-off-by: SeongJae Park <[email protected]>
> ---
Reviewed-by: Leonard Foerster <[email protected]>