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.