doc/en_US.ISO8859-1/books/developers-handbook/vm/chapter.sgml
Peter Pentchev 1f359cbbab /tmp/msg
2003-01-12 18:20:55 +00:00

260 lines
13 KiB
Text

<!--
The FreeBSD Documentation Project
$FreeBSD$
-->
<chapter id="vm">
<chapterinfo>
<authorgroup>
<author>
<firstname>Matthew</firstname>
<surname>Dillon</surname>
<contrib>Contributed by </contrib>
</author>
</authorgroup>
<!-- 6 Feb 1999 -->
</chapterinfo>
<title>Virtual Memory System</title>
<sect1 id="vm-physmem">
<title>Management of physical
memory&mdash;<literal>vm_page_t</literal></title>
<para>Physical memory is managed on a page-by-page basis through the
<literal>vm_page_t</literal> structure. Pages of physical memory are
categorized through the placement of their respective
<literal>vm_page_t</literal> structures on one of several paging
queues.</para>
<para>A page can be in a wired, active, inactive, cache, or free state.
Except for the wired state, the page is typically placed in a doubly
link list queue representing the state that it is in. Wired pages
are not placed on any queue.</para>
<para>FreeBSD implements a more involved paging queue for cached and
free pages in order to implement page coloring. Each of these states
involves multiple queues arranged according to the size of the
processor's L1 and L2 caches. When a new page needs to be allocated,
FreeBSD attempts to obtain one that is reasonably well aligned from
the point of view of the L1 and L2 caches relative to the VM object
the page is being allocated for.</para>
<para>Additionally, a page may be held with a reference count or locked
with a busy count. The VM system also implements an <quote>ultimate
locked</quote> state for a page using the PG_BUSY bit in the page's
flags.</para>
<para>In general terms, each of the paging queues operates in a LRU
fashion. A page is typically placed in a wired or active state
initially. When wired, the page is usually associated with a page
table somewhere. The VM system ages the page by scanning pages in a
more active paging queue (LRU) in order to move them to a less-active
paging queue. Pages that get moved into the cache are still
associated with a VM object but are candidates for immediate reuse.
Pages in the free queue are truly free. FreeBSD attempts to minimize
the number of pages in the free queue, but a certain minimum number of
truly free pages must be maintained in order to accommodate page
allocation at interrupt time.</para>
<para>If a process attempts to access a page that does not exist in its
page table but does exist in one of the paging queues (such as the
inactive or cache queues), a relatively inexpensive page reactivation
fault occurs which causes the page to be reactivated. If the page
does not exist in system memory at all, the process must block while
the page is brought in from disk.</para>
<para>FreeBSD dynamically tunes its paging queues and attempts to
maintain reasonable ratios of pages in the various queues as well as
attempts to maintain a reasonable breakdown of clean vs. dirty pages.
The amount of rebalancing that occurs depends on the system's memory
load. This rebalancing is implemented by the pageout daemon and
involves laundering dirty pages (syncing them with their backing
store), noticing when pages are activity referenced (resetting their
position in the LRU queues or moving them between queues), migrating
pages between queues when the queues are out of balance, and so forth.
FreeBSD's VM system is willing to take a reasonable number of
reactivation page faults to determine how active or how idle a page
actually is. This leads to better decisions being made as to when to
launder or swap-out a page.</para>
</sect1>
<sect1 id="vm-cache">
<title>The unified buffer
cache&mdash;<literal>vm_object_t</literal></title>
<para>FreeBSD implements the idea of a generic <quote>VM object</quote>.
VM objects can be associated with backing store of various
types&mdash;unbacked, swap-backed, physical device-backed, or
file-backed storage. Since the filesystem uses the same VM objects to
manage in-core data relating to files, the result is a unified buffer
cache.</para>
<para>VM objects can be <emphasis>shadowed</emphasis>. That is, they
can be stacked on top of each other. For example, you might have a
swap-backed VM object stacked on top of a file-backed VM object in
order to implement a MAP_PRIVATE mmap()ing. This stacking is also
used to implement various sharing properties, including
copy-on-write, for forked address spaces.</para>
<para>It should be noted that a <literal>vm_page_t</literal> can only be
associated with one VM object at a time. The VM object shadowing
implements the perceived sharing of the same page across multiple
instances.</para>
</sect1>
<sect1 id="vm-fileio">
<title>Filesystem I/O&mdash;<literal>struct buf</literal></title>
<para>vnode-backed VM objects, such as file-backed objects, generally
need to maintain their own clean/dirty info independent from the VM
system's idea of clean/dirty. For example, when the VM system decides
to synchronize a physical page to its backing store, the VM system
needs to mark the page clean before the page is actually written to
its backing store. Additionally, filesystems need to be able to map
portions of a file or file metadata into KVM in order to operate on
it.</para>
<para>The entities used to manage this are known as filesystem buffers,
<literal>struct buf</literal>'s, or
<literal>bp</literal>'s. When a filesystem needs to operate on a
portion of a VM object, it typically maps part of the object into a
struct buf and the maps the pages in the struct buf into KVM. In the
same manner, disk I/O is typically issued by mapping portions of
objects into buffer structures and then issuing the I/O on the buffer
structures. The underlying vm_page_t's are typically busied for the
duration of the I/O. Filesystem buffers also have their own notion of
being busy, which is useful to filesystem driver code which would
rather operate on filesystem buffers instead of hard VM pages.</para>
<para>FreeBSD reserves a limited amount of KVM to hold mappings from
struct bufs, but it should be made clear that this KVM is used solely
to hold mappings and does not limit the ability to cache data.
Physical data caching is strictly a function of
<literal>vm_page_t</literal>'s, not filesystem buffers. However,
since filesystem buffers are used to placehold I/O, they do inherently
limit the amount of concurrent I/O possible. However, as there are usually a
few thousand filesystem buffers available, this is not usually a
problem.</para>
</sect1>
<sect1 id="vm-pagetables">
<title>Mapping Page Tables&mdash;<literal>vm_map_t, vm_entry_t</literal></title>
<para>FreeBSD separates the physical page table topology from the VM
system. All hard per-process page tables can be reconstructed on the
fly and are usually considered throwaway. Special page tables such as
those managing KVM are typically permanently preallocated. These page
tables are not throwaway.</para>
<para>FreeBSD associates portions of vm_objects with address ranges in
virtual memory through <literal>vm_map_t</literal> and
<literal>vm_entry_t</literal> structures. Page tables are directly
synthesized from the
<literal>vm_map_t</literal>/<literal>vm_entry_t</literal>/
<literal>vm_object_t</literal> hierarchy. Recall that I mentioned
that physical pages are only directly associated with a
<literal>vm_object</literal>; that is not quite true.
<literal>vm_page_t</literal>'s are also linked into page tables that
they are actively associated with. One <literal>vm_page_t</literal>
can be linked into several <emphasis>pmaps</emphasis>, as page tables
are called. However, the hierarchical association holds, so all
references to the same page in the same object reference the same
<literal>vm_page_t</literal> and thus give us buffer cache unification
across the board.</para>
</sect1>
<sect1 id="vm-kvm">
<title>KVM Memory Mapping</title>
<para>FreeBSD uses KVM to hold various kernel structures. The single
largest entity held in KVM is the filesystem buffer cache. That is,
mappings relating to <literal>struct buf</literal> entities.</para>
<para>Unlike Linux, FreeBSD does <emphasis>not</emphasis> map all of physical memory into
KVM. This means that FreeBSD can handle memory configurations up to
4G on 32 bit platforms. In fact, if the mmu were capable of it,
FreeBSD could theoretically handle memory configurations up to 8TB on
a 32 bit platform. However, since most 32 bit platforms are only
capable of mapping 4GB of ram, this is a moot point.</para>
<para>KVM is managed through several mechanisms. The main mechanism
used to manage KVM is the <emphasis>zone allocator</emphasis>. The
zone allocator takes a chunk of KVM and splits it up into
constant-sized blocks of memory in order to allocate a specific type
of structure. You can use <command>vmstat -m</command> to get an
overview of current KVM utilization broken down by zone.</para>
</sect1>
<sect1 id="vm-tuning">
<title>Tuning the FreeBSD VM system</title>
<para>A concerted effort has been made to make the FreeBSD kernel
dynamically tune itself. Typically you do not need to mess with
anything beyond the <option>maxusers</option> and
<option>NMBCLUSTERS</option> kernel config options. That is, kernel
compilation options specified in (typically)
<filename>/usr/src/sys/i386/conf/<replaceable>CONFIG_FILE</replaceable></filename>.
A description of all available kernel configuration options can be
found in <filename>/usr/src/sys/i386/conf/LINT</filename>.</para>
<para>In a large system configuration you may wish to increase
<option>maxusers</option>. Values typically range from 10 to 128.
Note that raising <option>maxusers</option> too high can cause the
system to overflow available KVM resulting in unpredictable operation.
It is better to leave <option>maxusers</option> at some reasonable number and add other
options, such as <option>NMBCLUSTERS</option>, to increase specific
resources.</para>
<para>If your system is going to use the network heavily, you may want
to increase <option>NMBCLUSTERS</option>. Typical values range from
1024 to 4096.</para>
<para>The <literal>NBUF</literal> parameter is also traditionally used
to scale the system. This parameter determines the amount of KVA the
system can use to map filesystem buffers for I/O. Note that this
parameter has nothing whatsoever to do with the unified buffer cache!
This parameter is dynamically tuned in 3.0-CURRENT and later kernels
and should generally not be adjusted manually. We recommend that you
<emphasis>not</emphasis> try to specify an <literal>NBUF</literal>
parameter. Let the system pick it. Too small a value can result in
extremely inefficient filesystem operation while too large a value can
starve the page queues by causing too many pages to become wired
down.</para>
<para>By default, FreeBSD kernels are not optimized. You can set
debugging and optimization flags with the
<literal>makeoptions</literal> directive in the kernel configuration.
Note that you should not use <option>-g</option> unless you can
accommodate the large (typically 7 MB+) kernels that result.</para>
<programlisting>makeoptions DEBUG="-g"
makeoptions COPTFLAGS="-O -pipe"</programlisting>
<para>Sysctl provides a way to tune kernel parameters at run-time. You
typically do not need to mess with any of the sysctl variables,
especially the VM related ones.</para>
<para>Run time VM and system tuning is relatively straightforward.
First, use Soft Updates on your UFS/FFS filesystems whenever possible.
<filename>/usr/src/sys/ufs/ffs/README.softupdates</filename> contains
instructions (and restrictions) on how to configure it.</para>
<para>Second, configure sufficient swap. You should have a swap
partition configured on each physical disk, up to four, even on your
<quote>work</quote> disks. You should have at least 2x the swap space
as you have main memory, and possibly even more if you do not have a
lot of memory. You should also size your swap partition based on the
maximum memory configuration you ever intend to put on the machine so
you do not have to repartition your disks later on. If you want to be
able to accommodate a crash dump, your first swap partition must be at
least as large as main memory and <filename>/var/crash</filename> must
have sufficient free space to hold the dump.</para>
<para>NFS-based swap is perfectly acceptable on 4.X or later systems,
but you must be aware that the NFS server will take the brunt of the
paging load.</para>
</sect1>
</chapter>