freebsd-dev/sys/ia64/include/vmparam.h
Alan Cox 7bfda801a8 Change the management of cached pages (PQ_CACHE) in two fundamental
ways:

(1) Cached pages are no longer kept in the object's resident page
splay tree and memq.  Instead, they are kept in a separate per-object
splay tree of cached pages.  However, access to this new per-object
splay tree is synchronized by the _free_ page queues lock, not to be
confused with the heavily contended page queues lock.  Consequently, a
cached page can be reclaimed by vm_page_alloc(9) without acquiring the
object's lock or the page queues lock.

This solves a problem independently reported by tegge@ and Isilon.
Specifically, they observed the page daemon consuming a great deal of
CPU time because of pages bouncing back and forth between the cache
queue (PQ_CACHE) and the inactive queue (PQ_INACTIVE).  The source of
this problem turned out to be a deadlock avoidance strategy employed
when selecting a cached page to reclaim in vm_page_select_cache().
However, the root cause was really that reclaiming a cached page
required the acquisition of an object lock while the page queues lock
was already held.  Thus, this change addresses the problem at its
root, by eliminating the need to acquire the object's lock.

Moreover, keeping cached pages in the object's primary splay tree and
memq was, in effect, optimizing for the uncommon case.  Cached pages
are reclaimed far, far more often than they are reactivated.  Instead,
this change makes reclamation cheaper, especially in terms of
synchronization overhead, and reactivation more expensive, because
reactivated pages will have to be reentered into the object's primary
splay tree and memq.

(2) Cached pages are now stored alongside free pages in the physical
memory allocator's buddy queues, increasing the likelihood that large
allocations of contiguous physical memory (i.e., superpages) will
succeed.

Finally, as a result of this change long-standing restrictions on when
and where a cached page can be reclaimed and returned by
vm_page_alloc(9) are eliminated.  Specifically, calls to
vm_page_alloc(9) specifying VM_ALLOC_INTERRUPT can now reclaim and
return a formerly cached page.  Consequently, a call to malloc(9)
specifying M_NOWAIT is less likely to fail.

Discussed with: many over the course of the summer, including jeff@,
   Justin Husted @ Isilon, peter@, tegge@
Tested by: an earlier version by kris@
Approved by: re (kensmith)
2007-09-25 06:25:06 +00:00

204 lines
6.4 KiB
C

/* $FreeBSD$ */
/* From: NetBSD: vmparam.h,v 1.6 1997/09/23 23:23:23 mjacob Exp */
#ifndef _MACHINE_VMPARAM_H
#define _MACHINE_VMPARAM_H
/*-
* Copyright (c) 1988 University of Utah.
* Copyright (c) 1992, 1993
* The Regents of the University of California. All rights reserved.
*
* This code is derived from software contributed to Berkeley by
* the Systems Programming Group of the University of Utah Computer
* Science Department and Ralph Campbell.
*
* 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.
*
* from: Utah $Hdr: vmparam.h 1.16 91/01/18$
*
* @(#)vmparam.h 8.2 (Berkeley) 4/22/94
*/
/*
* Machine dependent constants for ia64.
*/
/*
* USRSTACK is the top (end) of the user stack. Immediately above the user
* stack resides the syscall gateway page.
*/
#define USRSTACK VM_MAX_ADDRESS
/*
* Virtual memory related constants, all in bytes
*/
#ifndef MAXTSIZ
#define MAXTSIZ (1<<30) /* max text size (1G) */
#endif
#ifndef DFLDSIZ
#define DFLDSIZ (1<<27) /* initial data size (128M) */
#endif
#ifndef MAXDSIZ
#define MAXDSIZ (1<<30) /* max data size (1G) */
#endif
#ifndef DFLSSIZ
#define DFLSSIZ (1<<21) /* initial stack size (2M) */
#endif
#ifndef MAXSSIZ
#define MAXSSIZ (1<<28) /* max stack size (256M) */
#endif
#ifndef SGROWSIZ
#define SGROWSIZ (128UL*1024) /* amount to grow stack */
#endif
/*
* Boundary at which to place first MAPMEM segment if not explicitly
* specified. Should be a power of two. This allows some slop for
* the data segment to grow underneath the first mapped segment.
*/
#define MMSEG 0x200000
/*
* The size of the clock loop.
*/
#define LOOPPAGES (maxfree - firstfree)
/*
* The time for a process to be blocked before being very swappable.
* This is a number of seconds which the system takes as being a non-trivial
* amount of real time. You probably shouldn't change this;
* it is used in subtle ways (fractions and multiples of it are, that is, like
* half of a ``long time'', almost a long time, etc.)
* It is related to human patience and other factors which don't really
* change over time.
*/
#define MAXSLP 20
/*
* A swapped in process is given a small amount of core without being bothered
* by the page replacement algorithm. Basically this says that if you are
* swapped in you deserve some resources. We protect the last SAFERSS
* pages against paging and will just swap you out rather than paging you.
* Note that each process has at least UPAGES pages which are not
* paged anyways, in addition to SAFERSS.
*/
#define SAFERSS 10 /* nominal ``small'' resident set size
protected against replacement */
/*
* We need region 7 virtual addresses for pagetables.
*/
#define UMA_MD_SMALL_ALLOC
/*
* The physical address space is sparsely populated.
*/
#define VM_PHYSSEG_SPARSE
/*
* The number of PHYSSEG entries is equal to the number of phys_avail
* entries.
*/
#define VM_PHYSSEG_MAX 49
/*
* Create three free page pools: VM_FREEPOOL_DEFAULT is the default pool
* from which physical pages are allocated and VM_FREEPOOL_DIRECT is
* the pool from which physical pages for small UMA objects are
* allocated.
*/
#define VM_NFREEPOOL 3
#define VM_FREEPOOL_CACHE 2
#define VM_FREEPOOL_DEFAULT 0
#define VM_FREEPOOL_DIRECT 1
/*
* Create one free page list.
*/
#define VM_NFREELIST 1
#define VM_FREELIST_DEFAULT 0
/*
* An allocation size of 256MB is supported in order to optimize the
* use of the identity mappings in region 7 by UMA.
*/
#define VM_NFREEORDER 16
/*
* Manipulating region bits of an address.
*/
#define IA64_RR_BASE(n) (((u_int64_t) (n)) << 61)
#define IA64_RR_MASK(x) ((x) & ((1L << 61) - 1))
#define IA64_PHYS_TO_RR6(x) ((x) | IA64_RR_BASE(6))
#define IA64_PHYS_TO_RR7(x) ((x) | IA64_RR_BASE(7))
/*
* Page size of the identity mappings in region 7.
*/
#ifndef LOG2_ID_PAGE_SIZE
#define LOG2_ID_PAGE_SIZE 28 /* 256M */
#endif
#define IA64_ID_PAGE_SHIFT (LOG2_ID_PAGE_SIZE)
#define IA64_ID_PAGE_SIZE (1<<(LOG2_ID_PAGE_SIZE))
#define IA64_ID_PAGE_MASK (IA64_ID_PAGE_SIZE-1)
#define IA64_BACKINGSTORE IA64_RR_BASE(4)
/*
* Mach derived constants
*/
/* user/kernel map constants */
#define VM_MIN_ADDRESS 0
#define VM_MAX_ADDRESS IA64_RR_BASE(5)
#define VM_GATEWAY_SIZE PAGE_SIZE
#define VM_MAXUSER_ADDRESS (VM_MAX_ADDRESS + VM_GATEWAY_SIZE)
#define VM_MIN_KERNEL_ADDRESS VM_MAXUSER_ADDRESS
#define VM_MAX_KERNEL_ADDRESS (IA64_RR_BASE(6) - 1)
#define KERNBASE VM_MAX_ADDRESS
/* virtual sizes (bytes) for various kernel submaps */
#ifndef VM_KMEM_SIZE
#define VM_KMEM_SIZE (12 * 1024 * 1024)
#endif
/*
* How many physical pages per KVA page allocated.
* min(max(max(VM_KMEM_SIZE, Physical memory/VM_KMEM_SIZE_SCALE),
* VM_KMEM_SIZE_MIN), VM_KMEM_SIZE_MAX)
* is the total KVA space allocated for kmem_map.
*/
#ifndef VM_KMEM_SIZE_SCALE
#define VM_KMEM_SIZE_SCALE (4) /* XXX 8192 byte pages */
#endif
/* initial pagein size of beginning of executable file */
#ifndef VM_INITIAL_PAGEIN
#define VM_INITIAL_PAGEIN 16
#endif
#endif /* !_MACHINE_VMPARAM_H */