1ALLOC_HUGEPAGES(2)         Linux Programmer's Manual        ALLOC_HUGEPAGES(2)
2
3
4

NAME

6       alloc_hugepages, free_hugepages - allocate or free huge pages
7

SYNOPSIS

9       void *alloc_hugepages(int key, void *addr, size_t len,
10                             int prot, int flag);
11
12       int free_hugepages(void *addr);
13

DESCRIPTION

15       The system calls alloc_hugepages() and free_hugepages() were introduced
16       in Linux 2.5.36 and removed again in 2.5.54.  They existed only on i386
17       and  ia64  (when  built with CONFIG_HUGETLB_PAGE).  In Linux 2.4.20 the
18       syscall numbers exist, but the calls fail with the error ENOSYS.
19
20       On i386 the memory management hardware knows about  ordinary  pages  (4
21       KiB)  and  huge  pages  (2  or 4 MiB).  Similarly ia64 knows about huge
22       pages of several sizes.  These system calls serve  to  map  huge  pages
23       into the process's memory or to free them again.  Huge pages are locked
24       into memory, and are not swapped.
25
26       The key argument is an identifier.  When zero the  pages  are  private,
27       and not inherited by children.  When positive the pages are shared with
28       other applications using the same key,  and  inherited  by  child  pro‐
29       cesses.
30
31       The  addr argument of free_hugepages() tells which page is being freed:
32       it was the return value of a call to alloc_hugepages().  (The memory is
33       first  actually freed when all users have released it.)  The addr argu‐
34       ment of alloc_hugepages() is a hint, that the kernel  may  or  may  not
35       follow.  Addresses must be properly aligned.
36
37       The  len  argument is the length of the required segment.  It must be a
38       multiple of the huge page size.
39
40       The prot argument specifies the memory protection of the  segment.   It
41       is one of PROT_READ, PROT_WRITE, PROT_EXEC.
42
43       The flag argument is ignored, unless key is positive.  In that case, if
44       flag is IPC_CREAT, then a new huge page segment is  created  when  none
45       with  the  given  key existed.  If this flag is not set, then ENOENT is
46       returned when no segment with the given key exists.
47

RETURN VALUE

49       On success, alloc_hugepages() returns the  allocated  virtual  address,
50       and free_hugepages() returns zero.  On error, -1 is returned, and errno
51       is set appropriately.
52

ERRORS

54       ENOSYS The system call is not supported on this kernel.
55

FILES

57       /proc/sys/vm/nr_hugepages Number of configured hugetlb pages.  This can
58       be read and written.
59
60       /proc/meminfo  Gives info on the number of configured hugetlb pages and
61       on their size in the three variables  HugePages_Total,  HugePages_Free,
62       Hugepagesize.
63

CONFORMING TO

65       These  calls  are specific to Linux on Intel processors, and should not
66       be used in programs intended to be portable.
67

NOTES

69       These system calls are gone; they existed only in Linux 2.5.36  through
70       to  2.5.54.  Now the hugetlbfs file system can be used instead.  Memory
71       backed by huge pages (if the CPU supports them) is  obtained  by  using
72       mmap(2) to map files in this virtual file system.
73
74       The  maximal number of huge pages can be specified using the hugepages=
75       boot parameter.
76
77

COLOPHON

79       This page is part of release 3.25 of the Linux  man-pages  project.   A
80       description  of  the project, and information about reporting bugs, can
81       be found at http://www.kernel.org/doc/man-pages/.
82
83
84
85Linux                             2007-05-31                ALLOC_HUGEPAGES(2)
Impressum