Revision 074c238177a75f5e79af3b2cb6a84e54823ef950 authored by Mel Gorman on 25 March 2015, 22:55:42 UTC, committed by Linus Torvalds on 25 March 2015, 23:20:31 UTC
Dave Chinner reported the following on https://lkml.org/lkml/2015/3/1/226

  Across the board the 4.0-rc1 numbers are much slower, and the degradation
  is far worse when using the large memory footprint configs. Perf points
  straight at the cause - this is from 4.0-rc1 on the "-o bhash=101073" config:

   -   56.07%    56.07%  [kernel]            [k] default_send_IPI_mask_sequence_phys
      - default_send_IPI_mask_sequence_phys
         - 99.99% physflat_send_IPI_mask
            - 99.37% native_send_call_func_ipi
                 smp_call_function_many
               - native_flush_tlb_others
                  - 99.85% flush_tlb_page
                       ptep_clear_flush
                       try_to_unmap_one
                       rmap_walk
                       try_to_unmap
                       migrate_pages
                       migrate_misplaced_page
                     - handle_mm_fault
                        - 99.73% __do_page_fault
                             trace_do_page_fault
                             do_async_page_fault
                           + async_page_fault
              0.63% native_send_call_func_single_ipi
                 generic_exec_single
                 smp_call_function_single

This is showing excessive migration activity even though excessive
migrations are meant to get throttled.  Normally, the scan rate is tuned
on a per-task basis depending on the locality of faults.  However, if
migrations fail for any reason then the PTE scanner may scan faster if
the faults continue to be remote.  This means there is higher system CPU
overhead and fault trapping at exactly the time we know that migrations
cannot happen.  This patch tracks when migration failures occur and
slows the PTE scanner.

Signed-off-by: Mel Gorman <mgorman@suse.de>
Reported-by: Dave Chinner <david@fromorbit.com>
Tested-by: Dave Chinner <david@fromorbit.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Aneesh Kumar <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
1 parent b191f9b
Raw File
mq_sysctl.c
/*
 *  Copyright (C) 2007 IBM Corporation
 *
 *  Author: Cedric Le Goater <clg@fr.ibm.com>
 *
 *  This program is free software; you can redistribute it and/or
 *  modify it under the terms of the GNU General Public License as
 *  published by the Free Software Foundation, version 2 of the
 *  License.
 */

#include <linux/nsproxy.h>
#include <linux/ipc_namespace.h>
#include <linux/sysctl.h>

#ifdef CONFIG_PROC_SYSCTL
static void *get_mq(struct ctl_table *table)
{
	char *which = table->data;
	struct ipc_namespace *ipc_ns = current->nsproxy->ipc_ns;
	which = (which - (char *)&init_ipc_ns) + (char *)ipc_ns;
	return which;
}

static int proc_mq_dointvec(struct ctl_table *table, int write,
			    void __user *buffer, size_t *lenp, loff_t *ppos)
{
	struct ctl_table mq_table;
	memcpy(&mq_table, table, sizeof(mq_table));
	mq_table.data = get_mq(table);

	return proc_dointvec(&mq_table, write, buffer, lenp, ppos);
}

static int proc_mq_dointvec_minmax(struct ctl_table *table, int write,
	void __user *buffer, size_t *lenp, loff_t *ppos)
{
	struct ctl_table mq_table;
	memcpy(&mq_table, table, sizeof(mq_table));
	mq_table.data = get_mq(table);

	return proc_dointvec_minmax(&mq_table, write, buffer,
					lenp, ppos);
}
#else
#define proc_mq_dointvec NULL
#define proc_mq_dointvec_minmax NULL
#endif

static int msg_max_limit_min = MIN_MSGMAX;
static int msg_max_limit_max = HARD_MSGMAX;

static int msg_maxsize_limit_min = MIN_MSGSIZEMAX;
static int msg_maxsize_limit_max = HARD_MSGSIZEMAX;

static struct ctl_table mq_sysctls[] = {
	{
		.procname	= "queues_max",
		.data		= &init_ipc_ns.mq_queues_max,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_mq_dointvec,
	},
	{
		.procname	= "msg_max",
		.data		= &init_ipc_ns.mq_msg_max,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_mq_dointvec_minmax,
		.extra1		= &msg_max_limit_min,
		.extra2		= &msg_max_limit_max,
	},
	{
		.procname	= "msgsize_max",
		.data		= &init_ipc_ns.mq_msgsize_max,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_mq_dointvec_minmax,
		.extra1		= &msg_maxsize_limit_min,
		.extra2		= &msg_maxsize_limit_max,
	},
	{
		.procname	= "msg_default",
		.data		= &init_ipc_ns.mq_msg_default,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_mq_dointvec_minmax,
		.extra1		= &msg_max_limit_min,
		.extra2		= &msg_max_limit_max,
	},
	{
		.procname	= "msgsize_default",
		.data		= &init_ipc_ns.mq_msgsize_default,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_mq_dointvec_minmax,
		.extra1		= &msg_maxsize_limit_min,
		.extra2		= &msg_maxsize_limit_max,
	},
	{}
};

static struct ctl_table mq_sysctl_dir[] = {
	{
		.procname	= "mqueue",
		.mode		= 0555,
		.child		= mq_sysctls,
	},
	{}
};

static struct ctl_table mq_sysctl_root[] = {
	{
		.procname	= "fs",
		.mode		= 0555,
		.child		= mq_sysctl_dir,
	},
	{}
};

struct ctl_table_header *mq_register_sysctl_table(void)
{
	return register_sysctl_table(mq_sysctl_root);
}
back to top