[143/227] mm: lru_cache_disable: replace work queue synchronization with synchronize_rcu

Commit Message

Andrew Morton March 22, 2022, 9:45 p.m. UTC

From: Marcelo Tosatti <[e-mail]>
Subject: mm: lru_cache_disable: replace work queue synchronization with synchronize_rcu

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 d479960e44f27e0e5 ("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.

Link: https://lkml.kernel.org/r/Yin7hDxdt0s/[e-mail]
Signed-off-by: Marcelo Tosatti <[e-mail]>
Reviewed-by: Nicolas Saenz Julienne <[e-mail]>
Acked-by: Minchan Kim <[e-mail]>
Cc: Matthew Wilcox <[e-mail]>
Cc: Mel Gorman <[e-mail]>
Cc: Juri Lelli <[e-mail]>
Cc: Thomas Gleixner <[e-mail]>
Cc: Sebastian Andrzej Siewior <[e-mail]>
Cc: Paul E. McKenney <[e-mail]>
Signed-off-by: Andrew Morton <[e-mail]>
---

 mm/swap.c |   23 ++++++++++++++---------
 1 file changed, 14 insertions(+), 9 deletions(-)

Patch

--- a/mm/swap.c~mm-lru_cache_disable-replace-work-queue-synchronization-with-synchronize_rcu
+++ a/mm/swap.c
@@ -831,8 +831,7 @@  inline void __lru_add_drain_all(bool for
 	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
 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();