linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: JaeJoon Jung <rgbi3307@gmail.com>
To: SeongJae Park <sj@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	damon@lists.linux.dev,  linux-kernel@vger.kernel.org,
	linux-mm@kvack.org
Subject: Re: [RFC PATCH v3 07/37] mm/damon/core: apply access reports to high level snapshot
Date: Fri, 12 Dec 2025 22:20:04 +0900	[thread overview]
Message-ID: <CAHOvCC4WQPo5HrCJ+958t1ab2NDzbg493XYarqACARTpYskVzQ@mail.gmail.com> (raw)
In-Reply-To: <20251208062943.68824-8-sj@kernel.org>

On Mon, 8 Dec 2025 at 15:35, SeongJae Park <sj@kernel.org> wrote:
>
> Now any DAMON API callers can report their observed access information.
> The DAMON core layer is just ignoring those, though.  Update the core to
> use the reported information at building the high level access pattern
> snapshot.

It seems inefficient to repeatedly access the damon_access_reports[1000] array
using a for loop in the kdamond_check_reported_accesses() function.
It is inefficient to for loop through the entire
damon_access_reports[1000] array.
When CONFIG_HZ and jiffies are increased as follows and
damond sample_interval is 5000us (5ms), the time flow diagram is as follows.

CONFIG_HZ 1000, jiffies == 1ms
damond sample_interval == 5000us (5ms)

reports_len(==): [0 ... 5]
                                                     [*]
        0       1       2       3        4       5     6    7    8
9     997  998  999
        [====|====|====|====|====]-----|----|----|----|  ....  |------|-------|
jiffies++    1        2       3       4        5     0    0    0   0
    0      0       0
damond_fn(sample interval)      -5[0<]

reports_len(==): [997 ... 2]
                                  [*]
        0           1            2    3    4    5    6    7    8   9
    997    998   999
        [======|======]----|----|----|-----|----|----|----|  ....  [=====|=====]
jiffies++   1001       1002   3   4   5    6     7    8    9       997
  998    999
damond_fn(sample interval)
                            -5[997<]

It seems that only the section corresponding to the sample interval ([==|==])
can be cycled as follows.  And, how about enjoying damon_access_reports[1000]
as damon_access_reports[500]?  Even if it reduce the 1000ms to 500ms
array space, it seems that it can sufficiently report and process within
the sample interval of 5ms.

 static unsigned int kdamond_check_reported_accesses(struct damon_ctx *ctx)
 {
-       int i;
+       int i = damon_access_reports_len;
+       unsigned int nr = 0;
        struct damon_access_report *report;
        struct damon_target *t;

@@ -2904,16 +2905,18 @@ static unsigned int
kdamond_check_reported_accesses(struct damon_ctx *ctx)
                return 0;

        mutex_lock(&damon_access_reports_lock);
-       for (i = 0; i < damon_access_reports_len; i++) {
-               report = &damon_access_reports[i];
-               if (time_before(report->report_jiffies,
-                                       jiffies -
-                                       usecs_to_jiffies(
-                                               ctx->attrs.sample_interval)))
-                       continue;
+       report = &damon_access_reports[i];
+       while (time_after(report->report_jiffies,
+               jiffies - usecs_to_jiffies(ctx->attrs.sample_interval))) {
                damon_for_each_target(t, ctx)
                        kdamond_apply_access_report(report, t, ctx);
+               if (++nr >= DAMON_ACCESS_REPORTS_CAP)
+                       break;
+
+               i = (i == 0) ? (DAMON_ACCESS_REPORTS_CAP - 1) : (i - 1);
+               report = &damon_access_reports[i];
        }
+
        mutex_unlock(&damon_access_reports_lock);
        /* For nr_accesses_bp, absence of access should also be reported. */
        return kdamond_apply_zero_access_report(ctx);
}

Thanks,
JaeJoon

>
> Signed-off-by: SeongJae Park <sj@kernel.org>
> ---
>  include/linux/damon.h |  1 +
>  mm/damon/core.c       | 68 ++++++++++++++++++++++++++++++++++++++++++-
>  2 files changed, 68 insertions(+), 1 deletion(-)
>
> diff --git a/include/linux/damon.h b/include/linux/damon.h
> index b8ebb2aa02c8..b04c2e36833a 100644
> --- a/include/linux/damon.h
> +++ b/include/linux/damon.h
> @@ -83,6 +83,7 @@ struct damon_region {
>         unsigned int age;
>  /* private: Internal value for age calculation. */
>         unsigned int last_nr_accesses;
> +       bool access_reported;
>  };
>
>  /**
> diff --git a/mm/damon/core.c b/mm/damon/core.c
> index 296117d5e7f7..a14754a47c7f 100644
> --- a/mm/damon/core.c
> +++ b/mm/damon/core.c
> @@ -137,6 +137,7 @@ struct damon_region *damon_new_region(unsigned long start, unsigned long end)
>
>         region->age = 0;
>         region->last_nr_accesses = 0;
> +       region->access_reported = false;
>
>         return region;
>  }
> @@ -2745,6 +2746,68 @@ static void kdamond_init_ctx(struct damon_ctx *ctx)
>         }
>  }
>
> +static void kdamond_apply_access_report(struct damon_access_report *report,
> +               struct damon_target *t, struct damon_ctx *ctx)
> +{
> +       struct damon_region *r;
> +
> +       /* todo: make search faster, e.g., binary search? */
> +       damon_for_each_region(r, t) {
> +               if (report->addr < r->ar.start)
> +                       continue;
> +               if (r->ar.end < report->addr + report->size)
> +                       continue;
> +               if (!r->access_reported)
> +                       damon_update_region_access_rate(r, true, &ctx->attrs);
> +               r->access_reported = true;
> +       }
> +}
> +
> +static unsigned int kdamond_apply_zero_access_report(struct damon_ctx *ctx)
> +{
> +       struct damon_target *t;
> +       struct damon_region *r;
> +       unsigned int max_nr_accesses = 0;
> +
> +       damon_for_each_target(t, ctx) {
> +               damon_for_each_region(r, t) {
> +                       if (r->access_reported)
> +                               r->access_reported = false;
> +                       else
> +                               damon_update_region_access_rate(r, false,
> +                                               &ctx->attrs);
> +                       max_nr_accesses = max(max_nr_accesses, r->nr_accesses);
> +               }
> +       }
> +       return max_nr_accesses;
> +}
> +
> +static unsigned int kdamond_check_reported_accesses(struct damon_ctx *ctx)
> +{
> +       int i;
> +       struct damon_access_report *report;
> +       struct damon_target *t;
> +
> +       /* currently damon_access_report supports only physical address */
> +       if (damon_target_has_pid(ctx))
> +               return 0;
> +
> +       mutex_lock(&damon_access_reports_lock);
> +       for (i = 0; i < damon_access_reports_len; i++) {
> +               report = &damon_access_reports[i];
> +               if (time_before(report->report_jiffies,
> +                                       jiffies -
> +                                       usecs_to_jiffies(
> +                                               ctx->attrs.sample_interval)))
> +                       continue;
> +               damon_for_each_target(t, ctx)
> +                       kdamond_apply_access_report(report, t, ctx);
> +       }
> +       mutex_unlock(&damon_access_reports_lock);
> +       /* For nr_accesses_bp, absence of access should also be reported. */
> +       return kdamond_apply_zero_access_report(ctx);
> +}
> +
>  /*
>   * The monitoring daemon that runs as a kernel thread
>   */
> @@ -2790,7 +2853,10 @@ static int kdamond_fn(void *data)
>                 kdamond_usleep(sample_interval);
>                 ctx->passed_sample_intervals++;
>
> -               if (ctx->ops.check_accesses)
> +               /* todo: make these non-exclusive */
> +               if (ctx->sample_control.primitives_enabled.page_fault)
> +                       max_nr_accesses = kdamond_check_reported_accesses(ctx);
> +               else if (ctx->ops.check_accesses)
>                         max_nr_accesses = ctx->ops.check_accesses(ctx);
>
>                 if (ctx->passed_sample_intervals >= next_aggregation_sis)
> --
> 2.47.3
>


  reply	other threads:[~2025-12-12 13:20 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-12-08  6:29 [RFC PATCH v3 00/37] mm/damon: introduce per-CPUs/threads/write/read monitoring SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 01/37] mm/damon/core: implement damon_report_access() SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 02/37] mm/damon: define struct damon_sample_control SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 03/37] mm/damon/core: commit damon_sample_control SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 04/37] mm/damon/core: implement damon_report_page_fault() SeongJae Park
2025-12-12 12:46   ` JaeJoon Jung
2025-12-12 22:47     ` SeongJae Park
2025-12-13  0:31       ` JaeJoon Jung
2025-12-13  0:56         ` SeongJae Park
2025-12-13  1:37           ` JaeJoon Jung
2025-12-08  6:29 ` [RFC PATCH v3 05/37] mm/{mprotect,memory}: (no upstream-aimed hack) implement MM_CP_DAMON SeongJae Park
2025-12-08 11:19   ` David Hildenbrand (Red Hat)
2025-12-09  4:56     ` SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 06/37] mm/damon/paddr: support page fault access check primitive SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 07/37] mm/damon/core: apply access reports to high level snapshot SeongJae Park
2025-12-12 13:20   ` JaeJoon Jung [this message]
2025-12-12 23:11     ` SeongJae Park
2025-12-13  1:10       ` JaeJoon Jung
2025-12-13  3:21         ` SeongJae Park
2025-12-13  4:09           ` JaeJoon Jung
2025-12-13  5:53             ` SeongJae Park
2025-12-13  9:17               ` SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 08/37] mm/damon/sysfs: implement monitoring_attrs/sample/ dir SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 09/37] mm/damon/sysfs: implement sample/primitives/ dir SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 10/37] mm/damon/sysfs: connect primitives directory with core SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 11/37] Docs/mm/damon/design: document page fault sampling primitive SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 12/37] Docs/admin-guide/mm/damon/usage: document sample primitives dir SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 13/37] mm/damon: extend damon_access_report for origin CPU reporting SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 14/37] mm/damon/core: report access origin cpu of page faults SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 15/37] mm/damon: implement sample filter data structure for cpus-only monitoring SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 16/37] mm/damon/core: implement damon_sample_filter manipulations SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 17/37] mm/damon/core: commit damon_sample_filters SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 18/37] mm/damon/core: apply sample filter to access reports SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 19/37] mm/damon/sysfs: implement sample/filters/ directory SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 20/37] mm/damon/sysfs: implement sample filter directory SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 21/37] mm/damon/sysfs: implement type, matching, allow files under sample filter dir SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 22/37] mm/damon/sysfs: implement cpumask file " SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 23/37] mm/damon/sysfs: connect sample filters with core layer SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 24/37] Docs/mm/damon/design: document sample filters SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 25/37] Docs/admin-guide/mm/damon/usage: document sample filters dir SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 26/37] mm/damon: extend damon_access_report for access-origin thread info SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 27/37] mm/damon/core: report access-generated thread id of the fault event SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 28/37] mm/damon: extend damon_sample_filter for threads SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 29/37] mm/damon/core: support threads type sample filter SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 30/37] mm/damon/sysfs: support thread based access sample filtering SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 31/37] Docs/mm/damon/design: document threads type sample filter SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 32/37] Docs/admin-guide/mm/damon/usage: document tids_arr file SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 33/37] mm/damon: support reporting write access SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 34/37] mm/damon/core: report whether the page fault was for writing SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 35/37] mm/damon/core: support write access sample filter SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 36/37] mm/damon/sysfs: support write-type " SeongJae Park
2025-12-08  6:29 ` [RFC PATCH v3 37/37] Docs/mm/damon/design: document write access sample filter type SeongJae Park

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAHOvCC4WQPo5HrCJ+958t1ab2NDzbg493XYarqACARTpYskVzQ@mail.gmail.com \
    --to=rgbi3307@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=damon@lists.linux.dev \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=sj@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox