hc
2024-02-20 102a0743326a03cd1a1202ceda21e175b7d3575c
kernel/tools/perf/Documentation/perf-bench.txt
....@@ -49,6 +49,9 @@
4949 'sched'::
5050 Scheduler and IPC mechanisms.
5151
52
+'syscall'::
53
+ System call performance (throughput).
54
+
5255 'mem'::
5356 Memory access performance.
5457
....@@ -57,6 +60,12 @@
5760
5861 'futex'::
5962 Futex stressing benchmarks.
63
+
64
+'epoll'::
65
+ Eventpoll (epoll) stressing benchmarks.
66
+
67
+'internals'::
68
+ Benchmark internal perf functionality.
6069
6170 'all'::
6271 All benchmark subsystems.
....@@ -131,6 +140,14 @@
131140 59004 ops/sec
132141 ---------------------
133142
143
+SUITES FOR 'syscall'
144
+~~~~~~~~~~~~~~~~~~
145
+*basic*::
146
+Suite for evaluating performance of core system call throughput (both usecs/op and ops/sec metrics).
147
+This uses a single thread simply doing getppid(2), which is a simple syscall where the result is not
148
+cached by glibc.
149
+
150
+
134151 SUITES FOR 'mem'
135152 ~~~~~~~~~~~~~~~~
136153 *memcpy*::
....@@ -203,6 +220,18 @@
203220 *lock-pi*::
204221 Suite for evaluating futex lock_pi calls.
205222
223
+SUITES FOR 'epoll'
224
+~~~~~~~~~~~~~~~~~~
225
+*wait*::
226
+Suite for evaluating concurrent epoll_wait calls.
227
+
228
+*ctl*::
229
+Suite for evaluating multiple epoll_ctl calls.
230
+
231
+SUITES FOR 'internals'
232
+~~~~~~~~~~~~~~~~~~~~~~
233
+*synthesize*::
234
+Suite for evaluating perf's event synthesis performance.
206235
207236 SEE ALSO
208237 --------