getrusage(2)
πΈ getrusage β the resource usage receipt bestie π
ok so real talk: this is getrusage(2), a system call (section 2, the cool section), and it's literally the function that hands you a lil receipt of how much CPU, memory, and I/O your program has been eating. pull up a chair π«Ά
β¨ NAME β¨
getrusage β get resource usage. that's it, that's the whole vibe. it tells you what your process has been consuming. π
π LIBRARY π
Standard C library (libc, link it with -lc) β so like, the normal one, nothing exotic, no cap.
π SYNOPSIS π
#include <sys/resource.h>
int getrusage(int who, struct rusage *usage);
π DESCRIPTION π
so getrusage() returns resource usage measures for who, and bestie, who can be one of these three icons:
RUSAGE_SELFβ returns the resource usage stats for the calling process, which is the sum of resources used by all the threads in that process. yes, all of them. the whole squad's tab. πRUSAGE_CHILDRENβ returns the resource usage stats for all children of the calling process that have terminated and been waited for. and it gets messier: these stats also include the resources used by grandchildren, and even further-removed descendants, if all of the intervening descendants waited on their terminated children. family tree accounting, fr fr. π³RUSAGE_THREAD(since Linux 2.6.26) β returns the resource usage stats for the calling thread only. just you, main character energy. BUT β οΈ you must define the_GNU_SOURCEfeature test macro (before including any header file, do not play with me on this) in order to obtain the definition of this constant from<sys/resource.h>.
π¦ the struct, look at her π¦
the resource usages come back in the structure pointed to by usage, which has this exact form:
struct rusage {
struct timeval ru_utime; /* user CPU time used */
struct timeval ru_stime; /* system CPU time used */
long ru_maxrss; /* maximum resident set size */
long ru_ixrss; /* integral shared memory size */
long ru_idrss; /* integral unshared data size */
long ru_isrss; /* integral unshared stack size */
long ru_minflt; /* page reclaims (soft page faults) */
long ru_majflt; /* page faults (hard page faults) */
long ru_nswap; /* swaps */
long ru_inblock; /* block input operations */
long ru_oublock; /* block output operations */
long ru_msgsnd; /* IPC messages sent */
long ru_msgrcv; /* IPC messages received */
long ru_nsignals; /* signals received */
long ru_nvcsw; /* voluntary context switches */
long ru_nivcsw; /* involuntary context switches */
};
β οΈ important lore: not all fields are completed. unmaintained fields are set to zero by the kernel. (the unmaintained fields are there for compatibility with other systems, and because they may one day be supported on Linux β hope springs eternal π₯².)
π field-by-field, let's get into it π
ru_utimeβ the total amount of time spent executing in user mode, expressed in atimevalstructure (seconds plus microseconds).ru_stimeβ the total amount of time spent executing in kernel mode, expressed in atimevalstructure (seconds plus microseconds).ru_maxrss(since Linux 2.6.32) β the maximum resident set size used, in KiB. and heads up: forRUSAGE_CHILDREN, this is the resident set size of the largest child, not the maximum resident set size of the whole process tree. don't let it gaslight you. πru_ixrss(unmaintained) β this field is currently unused on Linux.ru_idrss(unmaintained) β this field is currently unused on Linux.ru_isrss(unmaintained) β this field is currently unused on Linux.ru_minfltβ the number of page faults serviced without any I/O activity; here, I/O activity is avoided by βreclaimingβ a page frame from the list of pages awaiting reallocation. recycling era β»οΈru_majfltβ the number of page faults serviced that required I/O activity. the expensive ones. πru_nswap(unmaintained) β this field is currently unused on Linux.ru_inblock(since Linux 2.6.22) β the number of times the filesystem had to perform input.ru_oublock(since Linux 2.6.22) β the number of times the filesystem had to perform output.ru_msgsnd(unmaintained) β this field is currently unused on Linux.ru_msgrcv(unmaintained) β this field is currently unused on Linux.ru_nsignals(unmaintained) β this field is currently unused on Linux.ru_nvcsw(since Linux 2.6) β the number of times a context switch resulted due to a process voluntarily giving up the processor before its time slice was completed (usually to await availability of a resource). polite queen, yielded the CPU herself πru_nivcsw(since Linux 2.6) β the number of times a context switch resulted due to a higher priority process becoming runnable or because the current process exceeded its time slice. jumped, basically. π
β RETURN VALUE β
On success, zero is returned. On error, -1 is returned, and errno is set to indicate the error. classic. π
π ERRORS π
EFAULTβ usage points outside the accessible address space. you gave it a garbage pointer bestie. πEINVALβ who is invalid. pick one of the three, i beg.
π§ ATTRIBUTES π§
for an explanation of the terms used in this section, see attributes(7).
- Interface:
getrusage() - Attribute: Thread safety
- Value:
MT-Safeβ
so yeah, it's MT-Safe, run it from your threads, it's not gonna snitch. π«Ά
π STANDARDS π
POSIX.1-2024. periodt.
POSIX.1 specifies getrusage(), but it only specifies the fields ru_utime and ru_stime. everything else? bonus content. β¨
RUSAGE_THREAD is Linux-specific. very linux of her. π
π HISTORY π
4.3BSD, SVr4, SUSv1, POSIX.1-2001 XSI. ancient icon behavior.
Before Linux 2.6.9, if the disposition of SIGCHLD is set to SIG_IGN, then the resource usages of child processes are automatically included in the value returned by RUSAGE_CHILDREN β even though POSIX.1-2001 explicitly prohibits this. messy. π¬ this nonconformance is rectified in Linux 2.6.9 and later.
The structure definition shown at the start of this page was taken from 4.3BSD Reno. vintage.
Ancient systems provided a vtimes() function with a similar purpose to getrusage(). for backward compatibility, glibc (up until Linux 2.32) also provides vtimes(). all new applications should be written using getrusage(). (Since Linux 2.33, glibc no longer provides an vtimes() implementation.) so like, RIP vtimes(), you will not be missed boo π
π NOTES π
Resource usage metrics are preserved across execve(2). so you can exec away and your stats still come with you. slay. β¨
π₯ SEE ALSO π₯
clock_gettime(2), getrlimit(2), times(2), wait(2), wait4(2), clock(3), proc_pid_stat(5), proc_pid_io(5)
go read those too, i know you want to, iykyk ππ
and that's the whole page, bestie β this was the getrusage(2) man page from Linux man-pages 6.20, dated 2026-02-08. go check your own CPU usage, you're probably the main character of the process table. mwah π«Άπβ¨
π¦ where this page comes from
- arch rolling: man-pages 6.20-1 Β· BSD-2-Clause AND BSD-3-Clause AND BSD-4.3TAHOE AND BSD-4-Clause-UC AND GPL-1.0-or-later AND GPL-2.0-only AND GPL-2.0-or-later AND GPL-3.0-or-later AND LGPL-3.0-linking-exception AND LGPL-3.0-or-later AND LicenseRef-Public-Domain AND LicenseRef-UltraPermissive AND Linux-man-pages-1-para AND Linux-man-pages-copyleft AND Linux-man-pages-copyleft-2-para AND Linux-man-pages-copyleft-var AND MIT AND Spencer-94
- nixpkgs unstable: man-pages 6.20
- π the page's license: Linux-man-pages-copyleft
- Copyright 1992, Drew Eckhardt <drew@cs.colorado.edu>
- Copyright 2002, Michael Kerrisk <mtk.manpages@gmail.com>
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash on : a language model's rewrite of the original, under its terms