freebsd-dev/sys/vm/vm_meter.c
Alan Cox ebcddc7217 Introduce a new page queue, PQ_LAUNDRY, for storing unreferenced, dirty
pages, specificially, dirty pages that have passed once through the inactive
queue.  A new, dedicated thread is responsible for both deciding when to
launder pages and actually laundering them.  The new policy uses the
relative sizes of the inactive and laundry queues to determine whether to
launder pages at a given point in time.  In general, this leads to more
intelligent swapping behavior, since the laundry thread will avoid pageouts
when the marginal benefit of doing so is low.  Previously, without a
dedicated queue for dirty pages, the page daemon didn't have the information
to determine whether pageout provides any benefit to the system.  Thus, the
previous policy often resulted in small but steadily increasing amounts of
swap usage when the system is under memory pressure, even when the inactive
queue consisted mostly of clean pages.  This change addresses that issue,
and also paves the way for some future virtual memory system improvements by
removing the last source of object-cached clean pages, i.e., PG_CACHE pages.

The new laundry thread sleeps while waiting for a request from the page
daemon thread(s).  A request is raised by setting the variable
vm_laundry_request and waking the laundry thread.  We request launderings
for two reasons: to try and balance the inactive and laundry queue sizes
("background laundering"), and to quickly make up for a shortage of free
pages and clean inactive pages ("shortfall laundering").  When background
laundering is requested, the laundry thread computes the number of page
daemon wakeups that have taken place since the last laundering.  If this
number is large enough relative to the ratio of the laundry and (global)
inactive queue sizes, we will launder vm_background_launder_target pages at
vm_background_launder_rate KB/s.  Otherwise, the laundry thread goes back
to sleep without doing any work.  When scanning the laundry queue during
background laundering, reactivated pages are counted towards the laundry
thread's target.

In contrast, shortfall laundering is requested when an inactive queue scan
fails to meet its target.  In this case, the laundry thread attempts to
launder enough pages to meet v_free_target within 0.5s, which is the
inactive queue scan period.

A laundry request can be latched while another is currently being
serviced.  In particular, a shortfall request will immediately preempt a
background laundering.

This change also redefines the meaning of vm_cnt.v_reactivated and removes
the functions vm_page_cache() and vm_page_try_to_cache().  The new meaning
of vm_cnt.v_reactivated now better reflects its name.  It represents the
number of inactive or laundry pages that are returned to the active queue
on account of a reference.

In collaboration with:	markj
Reviewed by:	kib
Tested by:	pho
Sponsored by:	Dell EMC Isilon
Differential Revision:	https://reviews.freebsd.org/D8302
2016-11-09 18:48:37 +00:00

319 lines
10 KiB
C

/*-
* Copyright (c) 1982, 1986, 1989, 1993
* The Regents of the University of California. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 4. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* @(#)vm_meter.c 8.4 (Berkeley) 1/4/94
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/mutex.h>
#include <sys/proc.h>
#include <sys/resource.h>
#include <sys/rwlock.h>
#include <sys/sx.h>
#include <sys/vmmeter.h>
#include <sys/smp.h>
#include <vm/vm.h>
#include <vm/vm_page.h>
#include <vm/vm_extern.h>
#include <vm/vm_param.h>
#include <vm/pmap.h>
#include <vm/vm_map.h>
#include <vm/vm_object.h>
#include <sys/sysctl.h>
struct vmmeter vm_cnt;
SYSCTL_UINT(_vm, VM_V_FREE_MIN, v_free_min,
CTLFLAG_RW, &vm_cnt.v_free_min, 0, "Minimum low-free-pages threshold");
SYSCTL_UINT(_vm, VM_V_FREE_TARGET, v_free_target,
CTLFLAG_RW, &vm_cnt.v_free_target, 0, "Desired free pages");
SYSCTL_UINT(_vm, VM_V_FREE_RESERVED, v_free_reserved,
CTLFLAG_RW, &vm_cnt.v_free_reserved, 0, "Pages reserved for deadlock");
SYSCTL_UINT(_vm, VM_V_INACTIVE_TARGET, v_inactive_target,
CTLFLAG_RW, &vm_cnt.v_inactive_target, 0, "Pages desired inactive");
SYSCTL_UINT(_vm, VM_V_PAGEOUT_FREE_MIN, v_pageout_free_min,
CTLFLAG_RW, &vm_cnt.v_pageout_free_min, 0, "Min pages reserved for kernel");
SYSCTL_UINT(_vm, OID_AUTO, v_free_severe,
CTLFLAG_RW, &vm_cnt.v_free_severe, 0, "Severe page depletion point");
static int
sysctl_vm_loadavg(SYSCTL_HANDLER_ARGS)
{
#ifdef SCTL_MASK32
u_int32_t la[4];
if (req->flags & SCTL_MASK32) {
la[0] = averunnable.ldavg[0];
la[1] = averunnable.ldavg[1];
la[2] = averunnable.ldavg[2];
la[3] = averunnable.fscale;
return SYSCTL_OUT(req, la, sizeof(la));
} else
#endif
return SYSCTL_OUT(req, &averunnable, sizeof(averunnable));
}
SYSCTL_PROC(_vm, VM_LOADAVG, loadavg, CTLTYPE_STRUCT | CTLFLAG_RD |
CTLFLAG_MPSAFE, NULL, 0, sysctl_vm_loadavg, "S,loadavg",
"Machine loadaverage history");
/*
* This function aims to determine if the object is mapped,
* specifically, if it is referenced by a vm_map_entry. Because
* objects occasionally acquire transient references that do not
* represent a mapping, the method used here is inexact. However, it
* has very low overhead and is good enough for the advisory
* vm.vmtotal sysctl.
*/
static bool
is_object_active(vm_object_t obj)
{
return (obj->ref_count > obj->shadow_count);
}
static int
vmtotal(SYSCTL_HANDLER_ARGS)
{
struct vmtotal total;
vm_object_t object;
struct proc *p;
struct thread *td;
bzero(&total, sizeof(total));
/*
* Calculate process statistics.
*/
sx_slock(&allproc_lock);
FOREACH_PROC_IN_SYSTEM(p) {
if (p->p_flag & P_SYSTEM)
continue;
PROC_LOCK(p);
switch (p->p_state) {
case PRS_NEW:
PROC_UNLOCK(p);
continue;
break;
default:
FOREACH_THREAD_IN_PROC(p, td) {
thread_lock(td);
switch (td->td_state) {
case TDS_INHIBITED:
if (TD_IS_SWAPPED(td))
total.t_sw++;
else if (TD_IS_SLEEPING(td)) {
if (td->td_priority <= PZERO)
total.t_dw++;
else
total.t_sl++;
if (td->td_wchan ==
&vm_cnt.v_free_count)
total.t_pw++;
}
break;
case TDS_CAN_RUN:
total.t_sw++;
break;
case TDS_RUNQ:
case TDS_RUNNING:
total.t_rq++;
thread_unlock(td);
continue;
default:
break;
}
thread_unlock(td);
}
}
PROC_UNLOCK(p);
}
sx_sunlock(&allproc_lock);
/*
* Calculate object memory usage statistics.
*/
mtx_lock(&vm_object_list_mtx);
TAILQ_FOREACH(object, &vm_object_list, object_list) {
/*
* Perform unsynchronized reads on the object. In
* this case, the lack of synchronization should not
* impair the accuracy of the reported statistics.
*/
if ((object->flags & OBJ_FICTITIOUS) != 0) {
/*
* Devices, like /dev/mem, will badly skew our totals.
*/
continue;
}
if (object->ref_count == 0) {
/*
* Also skip unreferenced objects, including
* vnodes representing mounted file systems.
*/
continue;
}
if (object->ref_count == 1 &&
(object->flags & OBJ_NOSPLIT) != 0) {
/*
* Also skip otherwise unreferenced swap
* objects backing tmpfs vnodes, and POSIX or
* SysV shared memory.
*/
continue;
}
total.t_vm += object->size;
total.t_rm += object->resident_page_count;
if (is_object_active(object)) {
total.t_avm += object->size;
total.t_arm += object->resident_page_count;
}
if (object->shadow_count > 1) {
/* shared object */
total.t_vmshr += object->size;
total.t_rmshr += object->resident_page_count;
if (is_object_active(object)) {
total.t_avmshr += object->size;
total.t_armshr += object->resident_page_count;
}
}
}
mtx_unlock(&vm_object_list_mtx);
total.t_free = vm_cnt.v_free_count + vm_cnt.v_cache_count;
return (sysctl_handle_opaque(oidp, &total, sizeof(total), req));
}
/*
* vm_meter_cnt() - accumulate statistics from all cpus and the global cnt
* structure.
*
* The vmmeter structure is now per-cpu as well as global. Those
* statistics which can be kept on a per-cpu basis (to avoid cache
* stalls between cpus) can be moved to the per-cpu vmmeter. Remaining
* statistics, such as v_free_reserved, are left in the global
* structure.
*/
u_int
vm_meter_cnt(size_t offset)
{
struct pcpu *pcpu;
u_int count;
int i;
count = *(u_int *)((char *)&vm_cnt + offset);
CPU_FOREACH(i) {
pcpu = pcpu_find(i);
count += *(u_int *)((char *)&pcpu->pc_cnt + offset);
}
return (count);
}
static int
cnt_sysctl(SYSCTL_HANDLER_ARGS)
{
u_int count;
count = vm_meter_cnt((char *)arg1 - (char *)&vm_cnt);
return (SYSCTL_OUT(req, &count, sizeof(count)));
}
SYSCTL_PROC(_vm, VM_TOTAL, vmtotal, CTLTYPE_OPAQUE|CTLFLAG_RD|CTLFLAG_MPSAFE,
0, sizeof(struct vmtotal), vmtotal, "S,vmtotal",
"System virtual memory statistics");
SYSCTL_NODE(_vm, OID_AUTO, stats, CTLFLAG_RW, 0, "VM meter stats");
static SYSCTL_NODE(_vm_stats, OID_AUTO, sys, CTLFLAG_RW, 0,
"VM meter sys stats");
static SYSCTL_NODE(_vm_stats, OID_AUTO, vm, CTLFLAG_RW, 0,
"VM meter vm stats");
SYSCTL_NODE(_vm_stats, OID_AUTO, misc, CTLFLAG_RW, 0, "VM meter misc stats");
#define VM_STATS(parent, var, descr) \
SYSCTL_PROC(parent, OID_AUTO, var, \
CTLTYPE_UINT | CTLFLAG_RD | CTLFLAG_MPSAFE, &vm_cnt.var, 0, \
cnt_sysctl, "IU", descr)
#define VM_STATS_VM(var, descr) VM_STATS(_vm_stats_vm, var, descr)
#define VM_STATS_SYS(var, descr) VM_STATS(_vm_stats_sys, var, descr)
VM_STATS_SYS(v_swtch, "Context switches");
VM_STATS_SYS(v_trap, "Traps");
VM_STATS_SYS(v_syscall, "System calls");
VM_STATS_SYS(v_intr, "Device interrupts");
VM_STATS_SYS(v_soft, "Software interrupts");
VM_STATS_VM(v_vm_faults, "Address memory faults");
VM_STATS_VM(v_io_faults, "Page faults requiring I/O");
VM_STATS_VM(v_cow_faults, "Copy-on-write faults");
VM_STATS_VM(v_cow_optim, "Optimized COW faults");
VM_STATS_VM(v_zfod, "Pages zero-filled on demand");
VM_STATS_VM(v_ozfod, "Optimized zero fill pages");
VM_STATS_VM(v_swapin, "Swap pager pageins");
VM_STATS_VM(v_swapout, "Swap pager pageouts");
VM_STATS_VM(v_swappgsin, "Swap pages swapped in");
VM_STATS_VM(v_swappgsout, "Swap pages swapped out");
VM_STATS_VM(v_vnodein, "Vnode pager pageins");
VM_STATS_VM(v_vnodeout, "Vnode pager pageouts");
VM_STATS_VM(v_vnodepgsin, "Vnode pages paged in");
VM_STATS_VM(v_vnodepgsout, "Vnode pages paged out");
VM_STATS_VM(v_intrans, "In transit page faults");
VM_STATS_VM(v_reactivated, "Pages reactivated by pagedaemon");
VM_STATS_VM(v_pdwakeups, "Pagedaemon wakeups");
VM_STATS_VM(v_pdpages, "Pages analyzed by pagedaemon");
VM_STATS_VM(v_pdshortfalls, "Page reclamation shortfalls");
VM_STATS_VM(v_tcached, "Total pages cached");
VM_STATS_VM(v_dfree, "Pages freed by pagedaemon");
VM_STATS_VM(v_pfree, "Pages freed by exiting processes");
VM_STATS_VM(v_tfree, "Total pages freed");
VM_STATS_VM(v_page_size, "Page size in bytes");
VM_STATS_VM(v_page_count, "Total number of pages in system");
VM_STATS_VM(v_free_reserved, "Pages reserved for deadlock");
VM_STATS_VM(v_free_target, "Pages desired free");
VM_STATS_VM(v_free_min, "Minimum low-free-pages threshold");
VM_STATS_VM(v_free_count, "Free pages");
VM_STATS_VM(v_wire_count, "Wired pages");
VM_STATS_VM(v_active_count, "Active pages");
VM_STATS_VM(v_inactive_target, "Desired inactive pages");
VM_STATS_VM(v_inactive_count, "Inactive pages");
VM_STATS_VM(v_laundry_count, "Pages eligible for laundering");
VM_STATS_VM(v_cache_count, "Pages on cache queue");
VM_STATS_VM(v_pageout_free_min, "Min pages reserved for kernel");
VM_STATS_VM(v_interrupt_free_min, "Reserved pages for interrupt code");
VM_STATS_VM(v_forks, "Number of fork() calls");
VM_STATS_VM(v_vforks, "Number of vfork() calls");
VM_STATS_VM(v_rforks, "Number of rfork() calls");
VM_STATS_VM(v_kthreads, "Number of fork() calls by kernel");
VM_STATS_VM(v_forkpages, "VM pages affected by fork()");
VM_STATS_VM(v_vforkpages, "VM pages affected by vfork()");
VM_STATS_VM(v_rforkpages, "VM pages affected by rfork()");
VM_STATS_VM(v_kthreadpages, "VM pages affected by fork() by kernel");