Penguin

Differences between current version and previous revision of mlock(2).

Other diffs: Previous Major Revision, Previous Author, or view the Annotated Edit History

Newer page: version 2 Last edited on Monday, March 10, 2003 8:07:10 pm by PerryLorier
Older page: version 1 Last edited on Tuesday, June 4, 2002 12:23:43 am by perry Revert
@@ -1,104 +1,36 @@
-MLOCK  
-!!!MLOCK  
-NAME  
-SYNOPSIS  
-DESCRIPTION  
-RETURN VALUE  
-ERRORS  
-CONFORMING TO  
-SEE ALSO  
-----  
 !!NAME 
-  
-  
 mlock - disable paging for some parts of memory 
 !!SYNOPSIS 
+ #include <sys/mman.h>  
  
+ __int mlock(const void *__''addr''__, size_t __''len''__);__  
  
-__#include  
-__''addr''__, size_t__ ''len''__);  
-__  
 !!DESCRIPTION 
+mlock(2) disables paging for the memory in the range starting at ''addr'' with length ''len'' bytes. All pages which contain a part of the specified memory range are  
+guaranteed be resident in RAM when the mlock(2) system call returns successfully and they are guaranteed to stay in RAM until the pages are unlocked by munlock(2) or  
+munlockall(2), or until the process terminates or starts another program with exec(2). Child processes do not inherit page locks across a fork(2).  
  
+Memory locking has two main applications: real-time algorithms and high-security data processing. Real-time applications require deterministic timing, and, like  
+scheduling, paging is one major cause of unexpected program execution delays. Real-time applications will usually also switch to a real-time scheduler with  
+__sched_setscheduler__. Cryptographic security software often handles critical bytes like passwords or secret keys as data structures. As a result of paging, these secrets could be transfered onto a persistent swap store medium, where they might be accessible to the enemy long after the security software has erased the secrets in RAM and terminated.  
  
-__mlock__ disables paging for the memory in the range  
-starting at ''addr'' with length ''len'' bytes . All  
- pages which contain a part of the specified memory range are  
-guaranteed be resident in RAM when the __mlock__ system  
-call returns successfully and they are guaranteed to stay in  
- RAM until the pages are unlocked by __munlock__ or  
-__munlockall__, or until the process terminates or starts  
-another program with __exec__. Child processes do not  
-inherit page locks across a __fork__
+Memory locks do not stack, i .e., pages which have been locked several times by calls to mlock(2) or mlockall(2) will be unlocked by a single call to munlock(2) for the corresponding range or by munlockall(2). Pages which are mapped to several locations or by several processes stay locked into RAM as long as they are locked at least at one location or by at least one process. 
  
+On POSIX systems on which mlock and munlock are available, ___POSIX_MEMLOCK_RANGE__ is defined in <unistd.h> and the value PAGESIZE from <limits.h> indicates the number of bytes per page.  
  
-Memory locking has two main applications: real-time  
-algorithms and high-security data processing. Real-time  
-applications require deterministic timing, and, like  
-scheduling, paging is one major cause of unexpected program  
-execution delays. Real-time applications will usually also  
-switch to a real-time scheduler with  
-__sched_setscheduler__. Cryptographic security software  
-often handles critical bytes like passwords or secret keys  
-as data structures. As a result of paging, these secrets  
-could be transfered onto a persistent swap store medium,  
-where they might be accessible to the enemy long after the  
-security software has erased the secrets in RAM and  
-terminated.  
  
-  
-Memory locks do not stack, i.e., pages which have been  
-locked several times by calls to __mlock__ or  
-__mlockall__ will be unlocked by a single call to  
-__munlock__ for the corresponding range or by  
-__munlockall__. Pages which are mapped to several  
-locations or by several processes stay locked into RAM as  
-long as they are locked at least at one location or by at  
-least one process.  
-  
-  
-On POSIX systems on which __mlock__ and __munlock__  
-are available, ___POSIX_MEMLOCK_RANGE__ is defined in  
-__PAGESIZE__ from  
-__  
 !!RETURN VALUE 
+On success, mlock(2) returns zero. On error, -1 is returned, ''errno'' is set appropriately, and no changes are made to any locks in the address space of the  
+process.  
  
-  
-On success, __mlock__ returns zero. On error, -1 is  
-returned, ''errno'' is set appropriately, and no changes  
-are made to any locks in the address space of the  
-process.  
 !!ERRORS 
+;[ENOMEM]: Some of the specified address range does not correspond to mapped pages in the address space of the process or the process tried to exceed the maximum number of allowed locked pages.  
+;[EPERM]: The calling process does not have appropriate privileges. Only root processes are allowed to lock pages.  
+;[EINVAL]: ''len'' was not a positive number.  
  
-  
-__ENOMEM__  
-  
-  
-Some of the specified address range does not correspond to  
-mapped pages in the address space of the process or the  
-process tried to exceed the maximum number of allowed locked  
-pages.  
-  
-  
-__EPERM__  
-  
-  
-The calling process does not have appropriate privileges.  
-Only root processes are allowed to lock pages.  
-  
-  
-__EINVAL__  
-  
-  
-''len'' was not a positive number.  
 !!CONFORMING TO 
  
+POSIX.1b, SVr4. SVr4 documents an additional [EAGAIN] error code.  
  
-POSIX.1b, SVr4. SVr4 documents an additional EAGAIN error  
-code.  
 !!SEE ALSO 
-  
-  
- munlock(2), mlockall(2),  
- munlockall(2)  
-----  
+munlock(2), mlockall(2), munlockall(2) 
This page is a man page (or other imported legacy content). We are unable to automatically determine the license status of this page.