The mlock() function uses the mappings established for the address range [addr, addr + len) to identify pages to be locked in memory. If the page identified by a mapping changes, such as occurs when a copy of a writable MAP_PRIVATE page is made upon the first store, the lock will be transferred to the newly copied private page.

The munlock() function removes locks established with mlock().

A given page may be locked multiple times by executing an mlock() through different mappings. That is, if two different processes lock the same page, then the page will remain locked until both processes remove their locks. However, within a given mapping, page locks do not nest - multiple mlock() operations on the same address in the same process will all be removed with a single munlock(). Of course, a page locked in one process and mapped in another (or visible through a different mapping in the locking process) is still locked in memory. This fact can be used to create applications that do nothing other than lock important data in memory, thereby avoiding page I/O faults on references from other processes in the system.

If the mapping through which an mlock() has been performed is removed, an munlock() is implicitly performed. An munlock() is also performed implicitly when a page is deleted through file removal or truncation.

Locks established with mlock() are not inherited by a child process after a fork() and are not nested.

Attempts to mlock() more memory than a system-specific limit will fail.

Upon successful completion, the mlock() and munlock() functions return 0. Otherwise, no changes are made to any locks in the address space of the process, the functions return -1 and set errno to indicate the error.

The mlock() and munlock() functions will fail if:

EINVAL

The addr argument is not a multiple of the page size as returned by sysconf(3C).

ENOMEM

Addresses in the range [addr, addr + len) are invalid for the address space of a process, or specify one or more pages which are not mapped.

ENOSYS

The system does not support this memory locking interface.

EPERM

The process's effective user ID is not superuser.

The mlock() function will fail if:

EAGAIN

Some or all of the memory identified by the range [addr, addr + len) could not be locked because of insufficient system resources.

Because of the impact on system resources, the use of mlock() and munlock() is restricted to the superuser.

See attributes(5) for descriptions of the following attributes:

fork(2), memcntl(2), mmap(2), plock(3C), mlockall(3C), sysconf(3C), attributes(5), standards(5)