From: "Joel Fernandes (Google)" <joel@joelfernandes.org>
To: rcu@vger.kernel.org
Cc: linux-kernel@vger.kernel.org, rushikesh.s.kadam@intel.com,
urezki@gmail.com, neeraj.iitr10@gmail.com, frederic@kernel.org,
paulmck@kernel.org, rostedt@goodmis.org,
Vineeth Pillai <vineeth@bitbyteword.org>,
Joel Fernandes <joel@joelfernandes.org>
Subject: [PATCH v6 2/4] rcu: shrinker for lazy rcu
Date: Thu, 22 Sep 2022 22:01:02 +0000 [thread overview]
Message-ID: <20220922220104.2446868-3-joel@joelfernandes.org> (raw)
In-Reply-To: <20220922220104.2446868-1-joel@joelfernandes.org>
From: Vineeth Pillai <vineeth@bitbyteword.org>
The shrinker is used to speed up the free'ing of memory potentially held
by RCU lazy callbacks. RCU kernel module test cases show this to be
effective. Test is introduced in a later patch.
Signed-off-by: Vineeth Pillai <vineeth@bitbyteword.org>
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
---
kernel/rcu/tree_nocb.h | 52 ++++++++++++++++++++++++++++++++++++++++++
1 file changed, 52 insertions(+)
diff --git a/kernel/rcu/tree_nocb.h b/kernel/rcu/tree_nocb.h
index 661c685aba3f..1a182b9c4f6c 100644
--- a/kernel/rcu/tree_nocb.h
+++ b/kernel/rcu/tree_nocb.h
@@ -1332,6 +1332,55 @@ int rcu_nocb_cpu_offload(int cpu)
}
EXPORT_SYMBOL_GPL(rcu_nocb_cpu_offload);
+static unsigned long
+lazy_rcu_shrink_count(struct shrinker *shrink, struct shrink_control *sc)
+{
+ int cpu;
+ unsigned long count = 0;
+
+ /* Snapshot count of all CPUs */
+ for_each_possible_cpu(cpu) {
+ struct rcu_data *rdp = per_cpu_ptr(&rcu_data, cpu);
+
+ count += READ_ONCE(rdp->lazy_len);
+ }
+
+ return count ? count : SHRINK_EMPTY;
+}
+
+static unsigned long
+lazy_rcu_shrink_scan(struct shrinker *shrink, struct shrink_control *sc)
+{
+ int cpu;
+ unsigned long flags;
+ unsigned long count = 0;
+
+ /* Snapshot count of all CPUs */
+ for_each_possible_cpu(cpu) {
+ struct rcu_data *rdp = per_cpu_ptr(&rcu_data, cpu);
+ int _count = READ_ONCE(rdp->lazy_len);
+
+ if (_count == 0)
+ continue;
+ rcu_nocb_lock_irqsave(rdp, flags);
+ WRITE_ONCE(rdp->lazy_len, 0);
+ rcu_nocb_unlock_irqrestore(rdp, flags);
+ wake_nocb_gp(rdp, false);
+ sc->nr_to_scan -= _count;
+ count += _count;
+ if (sc->nr_to_scan <= 0)
+ break;
+ }
+ return count ? count : SHRINK_STOP;
+}
+
+static struct shrinker lazy_rcu_shrinker = {
+ .count_objects = lazy_rcu_shrink_count,
+ .scan_objects = lazy_rcu_shrink_scan,
+ .batch = 0,
+ .seeks = DEFAULT_SEEKS,
+};
+
void __init rcu_init_nohz(void)
{
int cpu;
@@ -1362,6 +1411,9 @@ void __init rcu_init_nohz(void)
if (!rcu_state.nocb_is_setup)
return;
+ if (register_shrinker(&lazy_rcu_shrinker, "rcu-lazy"))
+ pr_err("Failed to register lazy_rcu shrinker!\n");
+
if (!cpumask_subset(rcu_nocb_mask, cpu_possible_mask)) {
pr_info("\tNote: kernel parameter 'rcu_nocbs=', 'nohz_full', or 'isolcpus=' contains nonexistent CPUs.\n");
cpumask_and(rcu_nocb_mask, cpu_possible_mask,
--
2.37.3.998.g577e59143f-goog
next prev parent reply other threads:[~2022-09-22 22:01 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-09-22 22:01 [PATCH v6 0/4] rcu: call_rcu() power improvements Joel Fernandes (Google)
2022-09-22 22:01 ` [PATCH v6 1/4] rcu: Make call_rcu() lazy to save power Joel Fernandes (Google)
2022-09-23 21:44 ` Paul E. McKenney
2022-09-24 16:20 ` Joel Fernandes
2022-09-24 21:11 ` Paul E. McKenney
2022-09-24 22:56 ` Joel Fernandes
2022-09-25 17:31 ` Joel Fernandes
2022-09-26 17:42 ` Paul E. McKenney
2022-09-26 21:07 ` Joel Fernandes
2022-09-26 22:37 ` Paul E. McKenney
2022-09-26 23:33 ` Joel Fernandes
2022-09-26 23:53 ` Paul E. McKenney
2022-10-03 19:33 ` Joel Fernandes
2022-10-03 19:49 ` Paul E. McKenney
2022-09-24 22:46 ` Frederic Weisbecker
2022-09-24 23:28 ` Joel Fernandes
2022-09-25 1:00 ` Joel Fernandes
2022-09-25 22:00 ` Frederic Weisbecker
2022-09-26 15:04 ` Joel Fernandes
2022-09-26 17:33 ` Paul E. McKenney
2022-09-26 23:37 ` Joel Fernandes
2022-09-25 22:09 ` Frederic Weisbecker
2022-09-26 17:45 ` Paul E. McKenney
2022-09-25 8:57 ` Uladzislau Rezki
2022-09-25 17:46 ` Joel Fernandes
2022-09-26 17:48 ` Paul E. McKenney
2022-09-26 19:32 ` Uladzislau Rezki
2022-09-26 21:02 ` Joel Fernandes
2022-09-26 22:32 ` Paul E. McKenney
2022-09-26 23:47 ` Joel Fernandes
2022-09-26 23:59 ` Paul E. McKenney
2022-09-27 1:49 ` Joel Fernandes
2022-09-27 3:22 ` Paul E. McKenney
2022-09-27 13:05 ` Joel Fernandes
2022-09-27 14:14 ` Paul E. McKenney
2022-09-27 14:22 ` Joel Fernandes
2022-09-27 14:30 ` Paul E. McKenney
2022-09-27 15:25 ` Joel Fernandes
2022-09-27 15:59 ` Paul E. McKenney
[not found] ` <CAEXW_YRpAjvmBPzRA-hRQpuaDuZUzfndLb3q+e3BUyWprg5wkQ@mail.gmail.com>
2022-09-27 3:21 ` Paul E. McKenney
2022-09-26 22:27 ` Paul E. McKenney
2022-09-26 19:39 ` Uladzislau Rezki
2022-09-26 20:54 ` Joel Fernandes
2022-09-26 22:35 ` Paul E. McKenney
2022-09-26 23:44 ` Joel Fernandes
2022-09-26 23:57 ` Paul E. McKenney
2022-09-27 1:16 ` Joel Fernandes
2022-09-27 3:20 ` Paul E. McKenney
2022-09-27 14:08 ` Uladzislau Rezki
2022-09-27 14:30 ` Joel Fernandes
2022-09-27 14:59 ` Uladzislau Rezki
2022-09-27 15:13 ` Uladzislau Rezki
2022-09-27 21:31 ` Uladzislau Rezki
2022-09-27 22:05 ` Joel Fernandes
2022-09-27 22:29 ` Joel Fernandes
2022-09-30 16:11 ` Uladzislau Rezki
2022-10-04 11:35 ` Uladzislau Rezki
2022-10-04 18:06 ` Joel Fernandes
2022-09-27 15:14 ` Joel Fernandes
2022-09-22 22:01 ` Joel Fernandes (Google) [this message]
2022-09-22 22:01 ` [PATCH v6 3/4] rcuscale: Add laziness and kfree tests Joel Fernandes (Google)
2022-09-22 22:01 ` [PATCH v6 4/4] percpu-refcount: Use call_rcu_flush() for atomic switch Joel Fernandes (Google)
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=20220922220104.2446868-3-joel@joelfernandes.org \
--to=joel@joelfernandes.org \
--cc=frederic@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=neeraj.iitr10@gmail.com \
--cc=paulmck@kernel.org \
--cc=rcu@vger.kernel.org \
--cc=rostedt@goodmis.org \
--cc=rushikesh.s.kadam@intel.com \
--cc=urezki@gmail.com \
--cc=vineeth@bitbyteword.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