1 .\" Copyright (C) 2002 Robert Love
2 .\" and Copyright (C) 2006 Michael Kerrisk
4 .\" %%%LICENSE_START(GPLv2+_DOC_FULL)
5 .\" This is free documentation; you can redistribute it and/or
6 .\" modify it under the terms of the GNU General Public License as
7 .\" published by the Free Software Foundation; either version 2 of
8 .\" the License, or (at your option) any later version.
10 .\" The GNU General Public License's references to "object code"
11 .\" and "executables" are to be interpreted as the output of any
12 .\" document formatting or typesetting system, including
13 .\" intermediate and printed output.
15 .\" This manual is distributed in the hope that it will be useful,
16 .\" but WITHOUT ANY WARRANTY; without even the implied warranty of
17 .\" MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18 .\" GNU General Public License for more details.
20 .\" You should have received a copy of the GNU General Public
21 .\" License along with this manual; if not, see
22 .\" <http://www.gnu.org/licenses/>.
25 .\" 2002-11-19 Robert Love <rml@tech9.net> - initial version
26 .\" 2004-04-20 mtk - fixed description of return value
27 .\" 2004-04-22 aeb - added glibc prototype history
28 .\" 2005-05-03 mtk - noted that sched_setaffinity may cause thread
29 .\" migration and that CPU affinity is a per-thread attribute.
30 .\" 2006-02-03 mtk -- Major rewrite
31 .\" 2008-11-12, mtk, removed CPU_*() macro descriptions to a
32 .\" separate CPU_SET(3) page.
34 .TH SCHED_SETAFFINITY 2 2014-05-10 "Linux" "Linux Programmer's Manual"
36 sched_setaffinity, sched_getaffinity \- \
37 set and get a thread's CPU affinity mask
40 .BR "#define _GNU_SOURCE" " /* See feature_test_macros(7) */"
43 .BI "int sched_setaffinity(pid_t " pid ", size_t " cpusetsize ,
44 .BI " const cpu_set_t *" mask );
46 .BI "int sched_getaffinity(pid_t " pid ", size_t " cpusetsize ,
47 .BI " cpu_set_t *" mask );
50 A thread's CPU affinity mask determines the set of CPUs on which
51 it is eligible to run.
52 On a multiprocessor system, setting the CPU affinity mask
53 can be used to obtain performance benefits.
55 by dedicating one CPU to a particular thread
56 (i.e., setting the affinity mask of that thread to specify a single CPU,
57 and setting the affinity mask of all other threads to exclude that CPU),
58 it is possible to ensure maximum execution speed for that thread.
59 Restricting a thread to run on a single CPU also avoids
60 the performance cost caused by the cache invalidation that occurs
61 when a thread ceases to execute on one CPU and then
62 recommences execution on a different CPU.
64 A CPU affinity mask is represented by the
66 structure, a "CPU set", pointed to by
68 A set of macros for manipulating CPU sets is described in
71 .BR sched_setaffinity ()
72 sets the CPU affinity mask of the thread whose ID is
74 to the value specified by
78 is zero, then the calling thread is used.
81 is the length (in bytes) of the data pointed to by
83 Normally this argument would be specified as
84 .IR "sizeof(cpu_set_t)" .
86 If the thread specified by
88 is not currently running on one of the CPUs specified in
90 then that thread is migrated to one of the CPUs specified in
93 .BR sched_getaffinity ()
94 writes the affinity mask of the thread whose ID is
98 structure pointed to by
102 argument specifies the size (in bytes) of
106 is zero, then the mask of the calling thread is returned.
109 .BR sched_setaffinity ()
111 .BR sched_getaffinity ()
113 On error, \-1 is returned, and
115 is set appropriately.
119 A supplied memory address was invalid.
122 The affinity bit mask
124 contains no processors that are currently physically on the system
125 and permitted to the thread according to any restrictions that
126 may be imposed by the "cpuset" mechanism described in
130 .RB ( sched_getaffinity ()
131 and, in kernels before 2.6.9,
132 .BR sched_setaffinity ())
134 is smaller than the size of the affinity mask used by the kernel.
137 .RB ( sched_setaffinity ())
138 The calling thread does not have appropriate privileges.
139 The caller needs an effective user ID equal to the real user ID
140 or effective user ID of the thread identified by
142 or it must possess the
147 The thread whose ID is \fIpid\fP could not be found.
149 The CPU affinity system calls were introduced in Linux kernel 2.5.8.
150 The system call wrappers were introduced in glibc 2.3.
151 Initially, the glibc interfaces included a
157 argument was removed, but was then restored in glibc 2.3.4, with type
160 These system calls are Linux-specific.
163 .BR sched_setaffinity (),
164 the set of CPUs on which the thread will actually run is
165 the intersection of the set specified in the
167 argument and the set of CPUs actually present on the system.
168 The system may further restrict the set of CPUs on which the thread
169 runs if the "cpuset" mechanism described in
172 These restrictions on the actual set of CPUs on which the thread
173 will run are silently imposed by the kernel.
176 has a description of the Linux scheduling scheme.
178 The affinity mask is a per-thread attribute that can be
179 adjusted independently for each of the threads in a thread group.
180 The value returned from a call to
182 can be passed in the argument
186 as 0 will set the attribute for the calling thread,
187 and passing the value returned from a call to
189 will set the attribute for the main thread of the thread group.
190 (If you are using the POSIX threads API, then use
191 .BR pthread_setaffinity_np (3)
193 .BR sched_setaffinity ().)
197 inherits its parent's CPU affinity mask.
198 The affinity mask is preserved across an
201 This manual page describes the glibc interface for the CPU affinity calls.
202 The actual system call interface is slightly different, with the
205 .IR "unsigned long\ *" ,
206 reflecting the fact that the underlying implementation of CPU
207 sets is a simple bit mask.
209 .BR sched_getaffinity ()
210 system call returns the size (in bytes) of the
212 data type that is used internally by the kernel to
213 represent the CPU set bit mask.
223 .BR sched_get_priority_max (2),
224 .BR sched_get_priority_min (2),
225 .BR sched_getscheduler (2),
226 .BR sched_setscheduler (2),
229 .BR pthread_setaffinity_np (3),
230 .BR sched_getcpu (3),
231 .BR capabilities (7),
235 This page is part of release 3.67 of the Linux
238 A description of the project,
239 information about reporting bugs,
240 and the latest version of this page,
242 \%http://www.kernel.org/doc/man\-pages/.