From: Marcelo Tosatti <mtosatti@redhat.com>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org,
Minchan Kim <minchan@kernel.org>,
Matthew Wilcox <willy@infradead.org>,
Mel Gorman <mgorman@techsingularity.net>,
Nicolas Saenz Julienne <nsaenzju@redhat.com>,
Juri Lelli <juri.lelli@redhat.com>,
Thomas Gleixner <tglx@linutronix.de>,
Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
"Paul E. McKenney" <paulmck@kernel.org>
Subject: [patch v5] mm: lru_cache_disable: replace work queue synchronization with synchronize_rcu
Date: Thu, 10 Mar 2022 10:22:12 -0300 [thread overview]
Message-ID: <Yin7hDxdt0s/x+fp@fuller.cnet> (raw)
In-Reply-To: <20220304163554.8872fe5d5a9d634f7a2884f5@linux-foundation.org>
On systems that run FIFO:1 applications that busy loop,
any SCHED_OTHER task that attempts to execute
on such a CPU (such as work threads) will not
be scheduled, which leads to system hangs.
Commit d479960e44f27e0e52ba31b21740b703c538027c ("mm: disable LRU
pagevec during the migration temporarily") relies on
queueing work items on all online CPUs to ensure visibility
of lru_disable_count.
To fix this, replace the usage of work items with synchronize_rcu,
which provides the same guarantees.
Readers of lru_disable_count are protected by either disabling
preemption or rcu_read_lock:
preempt_disable, local_irq_disable [bh_lru_lock()]
rcu_read_lock [rt_spin_lock CONFIG_PREEMPT_RT]
preempt_disable [local_lock !CONFIG_PREEMPT_RT]
Since v5.1 kernel, synchronize_rcu() is guaranteed to wait on
preempt_disable() regions of code. So any CPU which sees
lru_disable_count = 0 will have exited the critical
section when synchronize_rcu() returns.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Reviewed-by: Nicolas Saenz Julienne <nsaenzju@redhat.com>
Acked-by: Minchan Kim <minchan@kernel.org>
---
v5: changelog improvements (Andrew Morton)
v4: improve comment clarity, mention synchronize_rcu guarantees
on v5.1 (Andrew Morton /
Paul E. McKenney)
v3: update stale comment (Nicolas Saenz Julienne)
v2: rt_spin_lock calls rcu_read_lock, no need
to add it before local_lock on swap.c (Nicolas Saenz Julienne)
diff --git a/mm/swap.c b/mm/swap.c
index bcf3ac288b56..b5ee163daa66 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -831,8 +831,7 @@ inline void __lru_add_drain_all(bool force_all_cpus)
for_each_online_cpu(cpu) {
struct work_struct *work = &per_cpu(lru_add_drain_work, cpu);
- if (force_all_cpus ||
- pagevec_count(&per_cpu(lru_pvecs.lru_add, cpu)) ||
+ if (pagevec_count(&per_cpu(lru_pvecs.lru_add, cpu)) ||
data_race(pagevec_count(&per_cpu(lru_rotate.pvec, cpu))) ||
pagevec_count(&per_cpu(lru_pvecs.lru_deactivate_file, cpu)) ||
pagevec_count(&per_cpu(lru_pvecs.lru_deactivate, cpu)) ||
@@ -876,15 +875,21 @@ atomic_t lru_disable_count = ATOMIC_INIT(0);
void lru_cache_disable(void)
{
atomic_inc(&lru_disable_count);
-#ifdef CONFIG_SMP
/*
- * lru_add_drain_all in the force mode will schedule draining on
- * all online CPUs so any calls of lru_cache_disabled wrapped by
- * local_lock or preemption disabled would be ordered by that.
- * The atomic operation doesn't need to have stronger ordering
- * requirements because that is enforced by the scheduling
- * guarantees.
+ * Readers of lru_disable_count are protected by either disabling
+ * preemption or rcu_read_lock:
+ *
+ * preempt_disable, local_irq_disable [bh_lru_lock()]
+ * rcu_read_lock [rt_spin_lock CONFIG_PREEMPT_RT]
+ * preempt_disable [local_lock !CONFIG_PREEMPT_RT]
+ *
+ * Since v5.1 kernel, synchronize_rcu() is guaranteed to wait on
+ * preempt_disable() regions of code. So any CPU which sees
+ * lru_disable_count = 0 will have exited the critical
+ * section when synchronize_rcu() returns.
*/
+ synchronize_rcu();
+#ifdef CONFIG_SMP
__lru_add_drain_all(true);
#else
lru_add_and_bh_lrus_drain();
next prev parent reply other threads:[~2022-03-10 15:43 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-02-22 16:01 [patch v2] mm: lru_cache_disable: replace work queue synchronization with synchronize_rcu Marcelo Tosatti
2022-02-22 16:07 ` [patch v3] " Marcelo Tosatti
2022-02-22 16:25 ` Nicolas Saenz Julienne
2022-03-04 1:03 ` Andrew Morton
2022-03-04 1:49 ` Paul E. McKenney
2022-03-04 15:08 ` Marcelo Tosatti
2022-03-04 16:02 ` Paul E. McKenney
2022-03-04 15:11 ` Marcelo Tosatti
2022-03-04 16:29 ` [patch v4] " Marcelo Tosatti
2022-03-05 0:35 ` Andrew Morton
2022-03-07 18:52 ` Marcelo Tosatti
2022-03-10 13:22 ` Marcelo Tosatti [this message]
2022-03-11 2:23 ` [patch v5] " Andrew Morton
2022-03-11 8:35 ` Sebastian Andrzej Siewior
2022-03-12 0:40 ` Andrew Morton
2022-03-12 20:39 ` Marcelo Tosatti
2022-03-13 9:23 ` Hillf Danton
2022-03-31 13:52 ` Borislav Petkov
2022-04-28 18:00 ` Marcelo Tosatti
2022-05-28 21:18 ` Andrew Morton
2022-05-28 22:54 ` Michael Larabel
2022-05-29 0:48 ` Michael Larabel
2022-06-19 12:14 ` Thorsten Leemhuis
2022-06-22 0:15 ` Andrew Morton
2022-03-05 4:33 ` [patch v4] " Paul E. McKenney
2022-03-08 17:41 ` Minchan Kim
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=Yin7hDxdt0s/x+fp@fuller.cnet \
--to=mtosatti@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=bigeasy@linutronix.de \
--cc=juri.lelli@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@techsingularity.net \
--cc=minchan@kernel.org \
--cc=nsaenzju@redhat.com \
--cc=paulmck@kernel.org \
--cc=tglx@linutronix.de \
--cc=willy@infradead.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;
as well as URLs for NNTP newsgroup(s).