1GETRUSAGE(2) Linux Programmer's Manual GETRUSAGE(2)
2
3
4
6 getrusage - get resource usage
7
9 #include <sys/time.h>
10 #include <sys/resource.h>
11
12 int getrusage(int who, struct rusage *usage);
13
15 getrusage() returns resource usage measures for who, which can be one
16 of the following:
17
18 RUSAGE_SELF
19 Return resource usage statistics for the calling process, which
20 is the sum of resources used by all threads in the process.
21
22 RUSAGE_CHILDREN
23 Return resource usage statistics for all children of the calling
24 process that have terminated and been waited for. These statis‐
25 tics will include the resources used by grandchildren, and fur‐
26 ther removed descendants, if all of the intervening descendants
27 waited on their terminated children.
28
29 RUSAGE_THREAD (since Linux 2.6.26)
30 Return resource usage statistics for the calling thread.
31
32 The resource usages are returned in the structure pointed to by usage,
33 which has the following form:
34
35 struct rusage {
36 struct timeval ru_utime; /* user CPU time used */
37 struct timeval ru_stime; /* system CPU time used */
38 long ru_maxrss; /* maximum resident set size */
39 long ru_ixrss; /* integral shared memory size */
40 long ru_idrss; /* integral unshared data size */
41 long ru_isrss; /* integral unshared stack size */
42 long ru_minflt; /* page reclaims (soft page faults) */
43 long ru_majflt; /* page faults (hard page faults) */
44 long ru_nswap; /* swaps */
45 long ru_inblock; /* block input operations */
46 long ru_oublock; /* block output operations */
47 long ru_msgsnd; /* IPC messages sent */
48 long ru_msgrcv; /* IPC messages received */
49 long ru_nsignals; /* signals received */
50 long ru_nvcsw; /* voluntary context switches */
51 long ru_nivcsw; /* involuntary context switches */
52 };
53
54 Not all fields are completed; unmaintained fields are set to zero by
55 the kernel. (The unmaintained fields are provided for compatibility
56 with other systems, and because they may one day be supported on
57 Linux.) The fields are interpreted as follows:
58
59 ru_utime
60 This is the total amount of time spent executing in user mode,
61 expressed in a timeval structure (seconds plus microseconds).
62
63 ru_stime
64 This is the total amount of time spent executing in kernel mode,
65 expressed in a timeval structure (seconds plus microseconds).
66
67 ru_maxrss (since Linux 2.6.32)
68 This is the maximum resident set size used (in kilobytes). For
69 RUSAGE_CHILDREN, this is the resident set size of the largest
70 child, not the maximum resident set size of the process tree.
71
72 ru_ixrss (unmaintained)
73 This field is currently unused on Linux.
74
75 ru_idrss (unmaintained)
76 This field is currently unused on Linux.
77
78 ru_isrss (unmaintained)
79 This field is currently unused on Linux.
80
81 ru_minflt
82 The number of page faults serviced without any I/O activity;
83 here I/O activity is avoided by “reclaiming” a page frame from
84 the list of pages awaiting reallocation.
85
86 ru_majflt
87 The number of page faults serviced that required I/O activity.
88
89 ru_nswap (unmaintained)
90 This field is currently unused on Linux.
91
92 ru_inblock (since Linux 2.6.22)
93 The number of times the file system had to perform input.
94
95 ru_oublock (since Linux 2.6.22)
96 The number of times the file system had to perform output.
97
98 ru_msgsnd (unmaintained)
99 This field is currently unused on Linux.
100
101 ru_msgrcv (unmaintained)
102 This field is currently unused on Linux.
103
104 ru_nsignals (unmaintained)
105 This field is currently unused on Linux.
106
107 ru_nvcsw (since Linux 2.6)
108 The number of times a context switch resulted due to a process
109 voluntarily giving up the processor before its time slice was
110 completed (usually to await availability of a resource).
111
112 ru_nivcsw (since Linux 2.6)
113 The number of times a context switch resulted due to a higher
114 priority process becoming runnable or because the current
115 process exceeded its time slice.
116
118 On success, zero is returned. On error, -1 is returned, and errno is
119 set appropriately.
120
122 EFAULT usage points outside the accessible address space.
123
124 EINVAL who is invalid.
125
127 SVr4, 4.3BSD. POSIX.1-2001 specifies getrusage(), but specifies only
128 the fields ru_utime and ru_stime.
129
130 RUSAGE_THREAD is Linux-specific.
131
133 Resource usage metrics are preserved across an execve(2).
134
135 Including <sys/time.h> is not required these days, but increases porta‐
136 bility. (Indeed, struct timeval is defined in <sys/time.h>.)
137
138 In Linux kernel versions before 2.6.9, if the disposition of SIGCHLD is
139 set to SIG_IGN then the resource usages of child processes are automat‐
140 ically included in the value returned by RUSAGE_CHILDREN, although
141 POSIX.1-2001 explicitly prohibits this. This nonconformance is recti‐
142 fied in Linux 2.6.9 and later.
143
144 The structure definition shown at the start of this page was taken from
145 4.3BSD Reno.
146
147 Ancient systems provided a vtimes() function with a similar purpose to
148 getrusage(). For backward compatibility, glibc also provides vtimes().
149 All new applications should be written using getrusage().
150
151 See also the description of /proc/PID/stat in proc(5).
152
154 clock_gettime(2), getrlimit(2), times(2), wait(2), wait4(2), clock(3)
155
157 This page is part of release 3.53 of the Linux man-pages project. A
158 description of the project, and information about reporting bugs, can
159 be found at http://www.kernel.org/doc/man-pages/.
160
161
162
163Linux 2010-09-26 GETRUSAGE(2)