Revision 03049269de433cb5fe2859be9ae4469ceb1163ed authored by Michal Hocko on 25 March 2016, 21:20:33 UTC, committed by Linus Torvalds on 25 March 2016, 23:37:42 UTC
wake_oom_reaper has allowed only 1 oom victim to be queued.  The main
reason for that was the simplicity as other solutions would require some
way of queuing.  The current approach is racy and that was deemed
sufficient as the oom_reaper is considered a best effort approach to
help with oom handling when the OOM victim cannot terminate in a
reasonable time.  The race could lead to missing an oom victim which can
get stuck

out_of_memory
  wake_oom_reaper
    cmpxchg // OK
    			oom_reaper
			  oom_reap_task
			    __oom_reap_task
oom_victim terminates
			      atomic_inc_not_zero // fail
out_of_memory
  wake_oom_reaper
    cmpxchg // fails
			  task_to_reap = NULL

This race requires 2 OOM invocations in a short time period which is not
very likely but certainly not impossible.  E.g.  the original victim
might have not released a lot of memory for some reason.

The situation would improve considerably if wake_oom_reaper used a more
robust queuing.  This is what this patch implements.  This means adding
oom_reaper_list list_head into task_struct (eat a hole before embeded
thread_struct for that purpose) and a oom_reaper_lock spinlock for
queuing synchronization.  wake_oom_reaper will then add the task on the
queue and oom_reaper will dequeue it.

Signed-off-by: Michal Hocko <mhocko@suse.com>
Cc: Vladimir Davydov <vdavydov@virtuozzo.com>
Cc: Andrea Argangeli <andrea@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
1 parent bc448e8
Raw File
blk-mq-cpumap.c
/*
 * CPU <-> hardware queue mapping helpers
 *
 * Copyright (C) 2013-2014 Jens Axboe
 */
#include <linux/kernel.h>
#include <linux/threads.h>
#include <linux/module.h>
#include <linux/mm.h>
#include <linux/smp.h>
#include <linux/cpu.h>

#include <linux/blk-mq.h>
#include "blk.h"
#include "blk-mq.h"

static int cpu_to_queue_index(unsigned int nr_cpus, unsigned int nr_queues,
			      const int cpu)
{
	return cpu * nr_queues / nr_cpus;
}

static int get_first_sibling(unsigned int cpu)
{
	unsigned int ret;

	ret = cpumask_first(topology_sibling_cpumask(cpu));
	if (ret < nr_cpu_ids)
		return ret;

	return cpu;
}

int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues,
			    const struct cpumask *online_mask)
{
	unsigned int i, nr_cpus, nr_uniq_cpus, queue, first_sibling;
	cpumask_var_t cpus;

	if (!alloc_cpumask_var(&cpus, GFP_ATOMIC))
		return 1;

	cpumask_clear(cpus);
	nr_cpus = nr_uniq_cpus = 0;
	for_each_cpu(i, online_mask) {
		nr_cpus++;
		first_sibling = get_first_sibling(i);
		if (!cpumask_test_cpu(first_sibling, cpus))
			nr_uniq_cpus++;
		cpumask_set_cpu(i, cpus);
	}

	queue = 0;
	for_each_possible_cpu(i) {
		if (!cpumask_test_cpu(i, online_mask)) {
			map[i] = 0;
			continue;
		}

		/*
		 * Easy case - we have equal or more hardware queues. Or
		 * there are no thread siblings to take into account. Do
		 * 1:1 if enough, or sequential mapping if less.
		 */
		if (nr_queues >= nr_cpus || nr_cpus == nr_uniq_cpus) {
			map[i] = cpu_to_queue_index(nr_cpus, nr_queues, queue);
			queue++;
			continue;
		}

		/*
		 * Less then nr_cpus queues, and we have some number of
		 * threads per cores. Map sibling threads to the same
		 * queue.
		 */
		first_sibling = get_first_sibling(i);
		if (first_sibling == i) {
			map[i] = cpu_to_queue_index(nr_uniq_cpus, nr_queues,
							queue);
			queue++;
		} else
			map[i] = map[first_sibling];
	}

	free_cpumask_var(cpus);
	return 0;
}

unsigned int *blk_mq_make_queue_map(struct blk_mq_tag_set *set)
{
	unsigned int *map;

	/* If cpus are offline, map them to first hctx */
	map = kzalloc_node(sizeof(*map) * nr_cpu_ids, GFP_KERNEL,
				set->numa_node);
	if (!map)
		return NULL;

	if (!blk_mq_update_queue_map(map, set->nr_hw_queues, cpu_online_mask))
		return map;

	kfree(map);
	return NULL;
}

/*
 * We have no quick way of doing reverse lookups. This is only used at
 * queue init time, so runtime isn't important.
 */
int blk_mq_hw_queue_to_node(unsigned int *mq_map, unsigned int index)
{
	int i;

	for_each_possible_cpu(i) {
		if (index == mq_map[i])
			return local_memory_node(cpu_to_node(i));
	}

	return NUMA_NO_NODE;
}
back to top