1 perf-bench(1) 1 perf-bench(1)
2 ============= 2 =============
3 3
4 NAME 4 NAME
5 ---- 5 ----
6 perf-bench - General framework for benchmark s 6 perf-bench - General framework for benchmark suites
7 7
8 SYNOPSIS 8 SYNOPSIS
9 -------- 9 --------
10 [verse] 10 [verse]
11 'perf bench' [<common options>] <subsystem> <s 11 'perf bench' [<common options>] <subsystem> <suite> [<options>]
12 12
13 DESCRIPTION 13 DESCRIPTION
14 ----------- 14 -----------
15 This 'perf bench' command is a general framewo 15 This 'perf bench' command is a general framework for benchmark suites.
16 16
17 COMMON OPTIONS 17 COMMON OPTIONS
18 -------------- 18 --------------
19 -r:: 19 -r::
20 --repeat=:: 20 --repeat=::
21 Specify number of times to repeat the run (def !! 21 Specify amount of times to repeat the run (default 10).
22 22
23 -f:: 23 -f::
24 --format=:: 24 --format=::
25 Specify format style. 25 Specify format style.
26 Current available format styles are: 26 Current available format styles are:
27 27
28 'default':: 28 'default'::
29 Default style. This is mainly for human readin 29 Default style. This is mainly for human reading.
30 --------------------- 30 ---------------------
31 % perf bench sched pipe # 31 % perf bench sched pipe # with no style specified
32 (executing 1000000 pipe operations between two 32 (executing 1000000 pipe operations between two tasks)
33 Total time:5.855 sec 33 Total time:5.855 sec
34 5.855061 usecs/op 34 5.855061 usecs/op
35 170792 ops/sec 35 170792 ops/sec
36 --------------------- 36 ---------------------
37 37
38 'simple':: 38 'simple'::
39 This simple style is friendly for automated 39 This simple style is friendly for automated
40 processing by scripts. 40 processing by scripts.
41 --------------------- 41 ---------------------
42 % perf bench --format=simple sched pipe # 42 % perf bench --format=simple sched pipe # specified simple
43 5.988 43 5.988
44 --------------------- 44 ---------------------
45 45
46 SUBSYSTEM 46 SUBSYSTEM
47 --------- 47 ---------
48 48
49 'sched':: 49 'sched'::
50 Scheduler and IPC mechanisms. 50 Scheduler and IPC mechanisms.
51 51
52 'syscall':: <<
53 System call performance (throughput). <<
54 <<
55 'mem':: 52 'mem'::
56 Memory access performance. 53 Memory access performance.
57 54
58 'numa':: 55 'numa'::
59 NUMA scheduling and MM benchmarks. 56 NUMA scheduling and MM benchmarks.
60 57
61 'futex':: 58 'futex'::
62 Futex stressing benchmarks. 59 Futex stressing benchmarks.
63 60
64 'epoll':: 61 'epoll'::
65 Eventpoll (epoll) stressing benchmarks 62 Eventpoll (epoll) stressing benchmarks.
66 63
67 'internals':: 64 'internals'::
68 Benchmark internal perf functionality. 65 Benchmark internal perf functionality.
69 66
70 'uprobe':: <<
71 Benchmark overhead of uprobe + BPF. <<
72 <<
73 'all':: 67 'all'::
74 All benchmark subsystems. 68 All benchmark subsystems.
75 69
76 SUITES FOR 'sched' 70 SUITES FOR 'sched'
77 ~~~~~~~~~~~~~~~~~~ 71 ~~~~~~~~~~~~~~~~~~
78 *messaging*:: 72 *messaging*::
79 Suite for evaluating performance of scheduler 73 Suite for evaluating performance of scheduler and IPC mechanisms.
80 Based on hackbench by Rusty Russell. 74 Based on hackbench by Rusty Russell.
81 75
82 Options of *messaging* 76 Options of *messaging*
83 ^^^^^^^^^^^^^^^^^^^^^^ 77 ^^^^^^^^^^^^^^^^^^^^^^
84 -p:: 78 -p::
85 --pipe:: 79 --pipe::
86 Use pipe() instead of socketpair() 80 Use pipe() instead of socketpair()
87 81
88 -t:: 82 -t::
89 --thread:: 83 --thread::
90 Be multi thread instead of multi process 84 Be multi thread instead of multi process
91 85
92 -g:: 86 -g::
93 --group=:: 87 --group=::
94 Specify number of groups 88 Specify number of groups
95 89
96 -l:: 90 -l::
97 --nr_loops=:: 91 --nr_loops=::
98 Specify number of loops 92 Specify number of loops
99 93
100 Example of *messaging* 94 Example of *messaging*
101 ^^^^^^^^^^^^^^^^^^^^^^ 95 ^^^^^^^^^^^^^^^^^^^^^^
102 96
103 --------------------- 97 ---------------------
104 % perf bench sched messaging # 98 % perf bench sched messaging # run with default
105 options (20 sender and receiver processes per 99 options (20 sender and receiver processes per group)
106 (10 groups == 400 processes run) 100 (10 groups == 400 processes run)
107 101
108 Total time:0.308 sec 102 Total time:0.308 sec
109 103
110 % perf bench sched messaging -t -g 20 # 104 % perf bench sched messaging -t -g 20 # be multi-thread, with 20 groups
111 (20 sender and receiver threads per group) 105 (20 sender and receiver threads per group)
112 (20 groups == 800 threads run) 106 (20 groups == 800 threads run)
113 107
114 Total time:0.582 sec 108 Total time:0.582 sec
115 --------------------- 109 ---------------------
116 110
117 *pipe*:: 111 *pipe*::
118 Suite for pipe() system call. 112 Suite for pipe() system call.
119 Based on pipe-test-1m.c by Ingo Molnar. 113 Based on pipe-test-1m.c by Ingo Molnar.
120 114
121 Options of *pipe* 115 Options of *pipe*
122 ^^^^^^^^^^^^^^^^^ 116 ^^^^^^^^^^^^^^^^^
123 -l:: 117 -l::
124 --loop=:: 118 --loop=::
125 Specify number of loops. 119 Specify number of loops.
126 120
127 -G:: <<
128 --cgroups=:: <<
129 Names of cgroups for sender and receiver, sepa <<
130 This is useful to check cgroup context switchi <<
131 Note that perf doesn't create nor delete the c <<
132 make sure that the cgroups exist and are acces <<
133 <<
134 <<
135 Example of *pipe* 121 Example of *pipe*
136 ^^^^^^^^^^^^^^^^^ 122 ^^^^^^^^^^^^^^^^^
137 123
138 --------------------- 124 ---------------------
139 % perf bench sched pipe 125 % perf bench sched pipe
140 (executing 1000000 pipe operations between two 126 (executing 1000000 pipe operations between two tasks)
141 127
142 Total time:8.091 sec 128 Total time:8.091 sec
143 8.091833 usecs/op 129 8.091833 usecs/op
144 123581 ops/sec 130 123581 ops/sec
145 131
146 % perf bench sched pipe -l 1000 # 132 % perf bench sched pipe -l 1000 # loop 1000
147 (executing 1000 pipe operations between two ta 133 (executing 1000 pipe operations between two tasks)
148 134
149 Total time:0.016 sec 135 Total time:0.016 sec
150 16.948000 usecs/op 136 16.948000 usecs/op
151 59004 ops/sec 137 59004 ops/sec
152 <<
153 % perf bench sched pipe -G AAA,BBB <<
154 (executing 1000000 pipe operations between cgr <<
155 # Running 'sched/pipe' benchmark: <<
156 # Executed 1000000 pipe operations between two <<
157 <<
158 Total time: 6.886 [sec] <<
159 <<
160 6.886208 usecs/op <<
161 145217 ops/sec <<
162 <<
163 --------------------- 138 ---------------------
164 <<
165 SUITES FOR 'syscall' <<
166 ~~~~~~~~~~~~~~~~~~ <<
167 *basic*:: <<
168 Suite for evaluating performance of core syste <<
169 This uses a single thread simply doing getppid <<
170 cached by glibc. <<
171 <<
172 139
173 SUITES FOR 'mem' 140 SUITES FOR 'mem'
174 ~~~~~~~~~~~~~~~~ 141 ~~~~~~~~~~~~~~~~
175 *memcpy*:: 142 *memcpy*::
176 Suite for evaluating performance of simple mem 143 Suite for evaluating performance of simple memory copy in various ways.
177 144
178 Options of *memcpy* 145 Options of *memcpy*
179 ^^^^^^^^^^^^^^^^^^^ 146 ^^^^^^^^^^^^^^^^^^^
180 -l:: 147 -l::
181 --size:: 148 --size::
182 Specify size of memory to copy (default: 1MB). 149 Specify size of memory to copy (default: 1MB).
183 Available units are B, KB, MB, GB and TB (case 150 Available units are B, KB, MB, GB and TB (case insensitive).
184 151
185 -f:: 152 -f::
186 --function:: 153 --function::
187 Specify function to copy (default: default). 154 Specify function to copy (default: default).
188 Available functions are depend on the architec 155 Available functions are depend on the architecture.
189 On x86-64, x86-64-unrolled, x86-64-movsq and x 156 On x86-64, x86-64-unrolled, x86-64-movsq and x86-64-movsb are supported.
190 157
191 -l:: 158 -l::
192 --nr_loops:: 159 --nr_loops::
193 Repeat memcpy invocation this number of times. 160 Repeat memcpy invocation this number of times.
194 161
195 -c:: 162 -c::
196 --cycles:: 163 --cycles::
197 Use perf's cpu-cycles event instead of gettime 164 Use perf's cpu-cycles event instead of gettimeofday syscall.
198 165
199 *memset*:: 166 *memset*::
200 Suite for evaluating performance of simple mem 167 Suite for evaluating performance of simple memory set in various ways.
201 168
202 Options of *memset* 169 Options of *memset*
203 ^^^^^^^^^^^^^^^^^^^ 170 ^^^^^^^^^^^^^^^^^^^
204 -l:: 171 -l::
205 --size:: 172 --size::
206 Specify size of memory to set (default: 1MB). 173 Specify size of memory to set (default: 1MB).
207 Available units are B, KB, MB, GB and TB (case 174 Available units are B, KB, MB, GB and TB (case insensitive).
208 175
209 -f:: 176 -f::
210 --function:: 177 --function::
211 Specify function to set (default: default). 178 Specify function to set (default: default).
212 Available functions are depend on the architec 179 Available functions are depend on the architecture.
213 On x86-64, x86-64-unrolled, x86-64-stosq and x 180 On x86-64, x86-64-unrolled, x86-64-stosq and x86-64-stosb are supported.
214 181
215 -l:: 182 -l::
216 --nr_loops:: 183 --nr_loops::
217 Repeat memset invocation this number of times. 184 Repeat memset invocation this number of times.
218 185
219 -c:: 186 -c::
220 --cycles:: 187 --cycles::
221 Use perf's cpu-cycles event instead of gettime 188 Use perf's cpu-cycles event instead of gettimeofday syscall.
222 189
223 SUITES FOR 'numa' 190 SUITES FOR 'numa'
224 ~~~~~~~~~~~~~~~~~ 191 ~~~~~~~~~~~~~~~~~
225 *mem*:: 192 *mem*::
226 Suite for evaluating NUMA workloads. 193 Suite for evaluating NUMA workloads.
227 194
228 SUITES FOR 'futex' 195 SUITES FOR 'futex'
229 ~~~~~~~~~~~~~~~~~~ 196 ~~~~~~~~~~~~~~~~~~
230 *hash*:: 197 *hash*::
231 Suite for evaluating hash tables. 198 Suite for evaluating hash tables.
232 199
233 *wake*:: 200 *wake*::
234 Suite for evaluating wake calls. 201 Suite for evaluating wake calls.
235 202
236 *wake-parallel*:: 203 *wake-parallel*::
237 Suite for evaluating parallel wake calls. 204 Suite for evaluating parallel wake calls.
238 205
239 *requeue*:: 206 *requeue*::
240 Suite for evaluating requeue calls. 207 Suite for evaluating requeue calls.
241 208
242 *lock-pi*:: 209 *lock-pi*::
243 Suite for evaluating futex lock_pi calls. 210 Suite for evaluating futex lock_pi calls.
244 211
245 SUITES FOR 'epoll' 212 SUITES FOR 'epoll'
246 ~~~~~~~~~~~~~~~~~~ 213 ~~~~~~~~~~~~~~~~~~
247 *wait*:: 214 *wait*::
248 Suite for evaluating concurrent epoll_wait cal 215 Suite for evaluating concurrent epoll_wait calls.
249 216
250 *ctl*:: 217 *ctl*::
251 Suite for evaluating multiple epoll_ctl calls. 218 Suite for evaluating multiple epoll_ctl calls.
252 219
253 SUITES FOR 'internals' 220 SUITES FOR 'internals'
254 ~~~~~~~~~~~~~~~~~~~~~~ 221 ~~~~~~~~~~~~~~~~~~~~~~
255 *synthesize*:: 222 *synthesize*::
256 Suite for evaluating perf's event synthesis pe 223 Suite for evaluating perf's event synthesis performance.
257 224
258 SEE ALSO 225 SEE ALSO
259 -------- 226 --------
260 linkperf:perf[1] 227 linkperf:perf[1]
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.