getcpu(2)

1getcpu(2)                     System Calls Manual                    getcpu(2)
2
3
4

NAME

6       getcpu  -  determine  CPU  and NUMA node on which the calling thread is
7       running
8

LIBRARY

10       Standard C library (libc, -lc)
11

SYNOPSIS

13       #define _GNU_SOURCE             /* See feature_test_macros(7) */
14       #include <sched.h>
15
16       int getcpu(unsigned int *_Nullable cpu, unsigned int *_Nullable node);
17

DESCRIPTION

19       The getcpu() system call identifies the processor and node on which the
20       calling thread or process is currently running and writes them into the
21       integers pointed to by the cpu and node arguments.  The processor is  a
22       unique  small  integer  identifying  a CPU.  The node is a unique small
23       identifier identifying a NUMA node.  When either cpu or  node  is  NULL
24       nothing is written to the respective pointer.
25
26       The  information  placed in cpu is guaranteed to be current only at the
27       time of the  call:  unless  the  CPU  affinity  has  been  fixed  using
28       sched_setaffinity(2),  the  kernel  might  change  the CPU at any time.
29       (Normally this does not happen because the scheduler tries to  minimize
30       movements  between  CPUs  to keep caches hot, but it is possible.)  The
31       caller must allow for the possibility that the information returned  in
32       cpu and node is no longer current by the time the call returns.
33

RETURN VALUE

35       On  success, 0 is returned.  On error, -1 is returned, and errno is set
36       to indicate the error.
37

ERRORS

39       EFAULT Arguments point outside the calling process's address space.
40

STANDARDS

42       Linux.
43

HISTORY

45       Linux 2.6.19 (x86-64 and i386), glibc 2.29.
46
47   C library/kernel differences
48       The kernel system call has a third argument:
49
50           int getcpu(unsigned int *cpu, unsigned int *node,
51                      struct getcpu_cache *tcache);
52
53       The tcache argument is unused since Linux 2.6.24,  and  (when  invoking
54       the system call directly) should be specified as NULL, unless portabil‐
55       ity to Linux 2.6.23 or earlier is required.
56
57       In Linux 2.6.23 and earlier, if the tcache argument was non-NULL,  then
58       it  specified  a  pointer  to a caller-allocated buffer in thread-local
59       storage that was used to provide a caching mechanism for getcpu().  Use
60       of  the  cache could speed getcpu() calls, at the cost that there was a
61       very small chance that the returned information would be out  of  date.
62       The  caching  mechanism was considered to cause problems when migrating
63       threads between CPUs, and so the argument is now ignored.
64

NOTES

66       Linux makes a best effort to make this call as fast as  possible.   (On
67       some architectures, this is done via an implementation in the vdso(7).)
68       The intention of getcpu() is to allow programs  to  make  optimizations
69       with per-CPU data or for NUMA optimization.
70