usr/src/lib/libc/sys/mlock.2

.\" Copyright (c) 1993 The Regents of the University of California.
.\" All rights reserved.
.\"
.\" %sccs.include.redist.roff%
.\"
.\"     @(#)mlock.2     8.1 (Berkeley) %G%
.\"
.Dd June 2, 1993
.Dt MLOCK 2
.Os
.Sh NAME
.Nm mlock ,
.Nm munlock
.Nd lock (unlock) physical pages in memory
.Sh SYNOPSIS
.Fd #include <sys/types.h>
.Fd #include <sys/mman.h>
.Ft int
.Fn mlock "caddr_t addr" "size_t len"
.Ft int
.Fn munlock "caddr_t addr" "size_t len"
.Sh DESCRIPTION
The
.Nm mlock
system call
locks into memory the physical pages associated with the virtual address
range starting at
.Fa addr
for
.Fa len
bytes.
The
.Nm munlock
call unlocks pages previously locked by one or more
.Nm mlock
calls.
For both, the
.Fa addr
parameter should be aligned to a multiple of the page size.
If the
.Fa len
parameter is not a multiple of the page size, it will be rounded up
to be so.
The entire range must be allocated.
.Pp
After an
.Nm mlock
call, the indicated pages will cause neither a non-resident page
or address-translation fault until they are unlocked.
They may still cause protection-violation faults or TLB-miss faults on
architectures with software-managed TLBs.
The physical pages remain in memory until all locked mappings for the pages
are removed.
Multiple processes may have the same physical pages locked via their own
virtual address mappings.
A single process may likewise have pages multiply-locked via different virtual
mappings of the same pages or via nested
.Nm mlock
calls on the same address range.
Unlocking is performed explicitly by
.Nm munlock
or implicitly by a call to
.Nm munmap
which deallocates the unmapped address range.
Locked mappings are not inherited by the child process after a
.Xr fork 2 .
.Pp
Since physical memory is a potentially scarce resource, processes are
limited in how much they can lock down.
A single process can
.Nm mlock
the minimum of
a system-wide ``wired pages'' limit and
the per-process
.Li RLIMIT_MEMLOCK
resource limit.
.Sh RETURN VALUES
A return value of 0 indicates that the call
succeeded and all pages in the range have either been locked or unlocked.
A return value of -1 indicates an error occurred and the locked
status of all pages in the range remains unchanged.
In this case, the global location
.Va errno
is set to indicate the error.
.Sh ERRORS
.Fn Mlock
will fail if:
.Bl -tag -width Er
.It Bq Er EINVAL
The address given is not page aligned or the length is negative.
.It Bq Er EAGAIN
Locking the indicated range would exceed either the system or per-process
limit for locked memory.
.It Bq Er ENOMEM
Some portion of the indicated address range is not allocated.
There was an error faulting/mapping a page.
.El
.Fn Munlock
will fail if:
.Bl -tag -width Er
.It Bq Er EINVAL
The address given is not page aligned or the length is negative.
.It Bq Er ENOMEM
Some portion of the indicated address range is not allocated.
Some portion of the indicated address range is not locked.
.El
.Sh "SEE ALSO"
.Xr fork 2 ,
.Xr mmap 2 ,
.Xr munmap 2 ,
.Xr setrlimit 2 ,
.Xr getpagesize 3
.Sh BUGS
Unlike The Sun implementation, multiple
.Nm mlock
calls on the same address range require the corresponding number of
.Nm munlock
calls to actually unlock the pages, i.e.
.Nm mlock
nests.
This should be considered a consequence of the implementation
and not a feature.
.Pp
The per-process resource limit is a limit on the amount of virtual
memory locked, while the system-wide limit is for the number of locked
physical pages.
Hence a process with two distinct locked mappings of the same physical page
counts as 2 pages against the per-process limit and as only a single page
in the system limit.
.Sh HISTORY
The
.Fn mlock
and
.Fn munlock
calls are
.Ud .