PERF-BENCH(1) perf Manual PERF-BENCH(1) (NAME) perf-bench - (SYNOPSIS) perf bench [] [] (DESCRIPTION) perf bench () . (COMMON OPTIONS) -r, --repeat= ( ). -f, --format= . : default . . .ft C % perf bench sched pipe # with no style specified (executing 1000000 pipe operations between two tasks) Total time:5.855 sec 5.855061 usecs/op 170792 ops/sec .ft simple . .ft C % perf bench --format=simple sched pipe # specified simple 5.988 .ft sched IPC. syscall ( ) . mem . numa NUMA (MM). futex futex. epoll Eventpoll (epoll). internals perf. uprobe uprobe + BPF. all . sched (SUITES FOR sched) messaging IPC. hackbench Rusty Russell. messaging -p, --pipe pipe() socketpair() -t, --thread (multi- thread) (multi-process) -g, --group= -l, --nr_loops= () messaging .ft C % perf bench sched messaging # run with default options (20 sender and receiver processes per group) (10 groups == 400 processes run) Total time:0.308 sec % perf bench sched messaging -t -g 20 # be multi-thread, with 20 groups (20 sender and receiver threads per group) (20 groups == 800 threads run) Total time:0.582 sec .ft pipe pipe(). pipe-test-1m.c Ingo Molnar. pipe -l, --loop= (). -G, --cgroups= cgroup . (context switching) cgroup . perf cgroup cgroup . pipe .ft C % perf bench sched pipe (executing 1000000 pipe operations between two tasks) Total time:8.091 sec 8.091833 usecs/op 123581 ops/sec % perf bench sched pipe -l 1000 # loop 1000 (executing 1000 pipe operations between two tasks) Total time:0.016 sec 16.948000 usecs/op 59004 ops/sec % perf bench sched pipe -G AAA,BBB (executing 1000000 pipe operations between cgroups) # Running 'sched/pipe' benchmark: # Executed 1000000 pipe operations between two processes Total time: 6.886 [sec] 6.886208 usecs/op 145217 ops/sec .ft basic ( usecs/op ops/sec). getppid(2) glibc . mem memcpy . memcpy -s, --size (: 1MB). B KB MB GB TB ( ). -p, --page (page-size) (: 4KB). 4KB 2MB 1GB ( ). -k, --chunk (chunk-size) . (: 0 full-extent) B KB MB GB TB ( ). -f, --function (: default). . x86-64 x86-64-unrolled x86-64-movsq x86-64-movsb . -l, --nr_loops memcpy . -c, --cycles cpu-cycles perf gettimeofday. memset . memset -s, --size (: 1MB). B KB MB GB TB ( ). -p, --page (page-size) (: 4KB). 4KB 2MB 1GB ( ). -k, --chunk (chunk-size) . (: 0 full-extent) B KB MB GB TB ( ). -f, --function (: default). . x86-64 x86-64-unrolled x86-64-stosq x86-64-stosb . -l, --nr_loops memset . -c, --cycles cpu-cycles perf gettimeofday. mmap mmap(). mmap -s, --size (: 1MB). B KB MB GB TB ( ). -p, --page (page-size) (: 4KB). 4KB 2MB 1GB ( ). -r, --randomize (seed) (: 0 ). -f, --function (: all). demand populate (demand faulting) (eager mapping) . -l, --nr_loops mmap() . -c, --cycles cpu-cycles perf gettimeofday. -t, --threads= mmap/munmap. mem NUMA. futex hash (hash tables). wake wake. wake-parallel wake. requeue requeue. lock-pi lock_pi futex. epoll wait epoll_wait. ctl epoll_ctl. internals synthesize perf. (SEE ALSO) perf(1) perf 09/04/2026 PERF-BENCH(1)