6 perf-bench - General framework for benchmark suites
11 'perf bench' [<common options>] <subsystem> <suite> [<options>]
15 This 'perf bench' command is a general framework for benchmark suites.
22 Current available format styles are:
25 Default style. This is mainly for human reading.
27 % perf bench sched pipe # with no style specified
28 (executing 1000000 pipe operations between two tasks)
35 This simple style is friendly for automated
36 processing by scripts.
38 % perf bench --format=simple sched pipe # specified simple
46 Scheduler and IPC mechanisms.
49 Memory access performance.
52 NUMA scheduling and MM benchmarks.
55 Futex stressing benchmarks.
58 All benchmark subsystems.
63 Suite for evaluating performance of scheduler and IPC mechanisms.
64 Based on hackbench by Rusty Russell.
66 Options of *messaging*
67 ^^^^^^^^^^^^^^^^^^^^^^
70 Use pipe() instead of socketpair()
74 Be multi thread instead of multi process
78 Specify number of groups
82 Specify number of loops
84 Example of *messaging*
85 ^^^^^^^^^^^^^^^^^^^^^^
88 % perf bench sched messaging # run with default
89 options (20 sender and receiver processes per group)
90 (10 groups == 400 processes run)
94 % perf bench sched messaging -t -g 20 # be multi-thread, with 20 groups
95 (20 sender and receiver threads per group)
96 (20 groups == 800 threads run)
102 Suite for pipe() system call.
103 Based on pipe-test-1m.c by Ingo Molnar.
109 Specify number of loops.
114 ---------------------
115 % perf bench sched pipe
116 (executing 1000000 pipe operations between two tasks)
122 % perf bench sched pipe -l 1000 # loop 1000
123 (executing 1000 pipe operations between two tasks)
128 ---------------------
133 Suite for evaluating performance of simple memory copy in various ways.
139 Specify length of memory to copy (default: 1MB).
140 Available units are B, KB, MB, GB and TB (case insensitive).
144 Specify routine to copy (default: default).
145 Available routines are depend on the architecture.
146 On x86-64, x86-64-unrolled, x86-64-movsq and x86-64-movsb are supported.
150 Repeat memcpy invocation this number of times.
154 Use perf's cpu-cycles event instead of gettimeofday syscall.
158 Show only the result with page faults before memcpy.
162 Show only the result without page faults before memcpy.
165 Suite for evaluating performance of simple memory set in various ways.
171 Specify length of memory to set (default: 1MB).
172 Available units are B, KB, MB, GB and TB (case insensitive).
176 Specify routine to set (default: default).
177 Available routines are depend on the architecture.
178 On x86-64, x86-64-unrolled, x86-64-stosq and x86-64-stosb are supported.
182 Repeat memset invocation this number of times.
186 Use perf's cpu-cycles event instead of gettimeofday syscall.
190 Show only the result with page faults before memset.
194 Show only the result without page faults before memset.
199 Suite for evaluating NUMA workloads.
204 Suite for evaluating hash tables.
207 Suite for evaluating wake calls.
210 Suite for evaluating requeue calls.