1OSLAT(8)                    System Manager's Manual                   OSLAT(8)
2
3
4

NAME

6       oslat - OS Latency Detector
7

SYNOPSIS

9       oslat [ -shvz ] [ -b bucket-size ] [ -B bias ] [ -c cpu-list ] [ -C
10             cpu-main-thread ] [ -f rt-prio ] [ -m workload-mem ] [-t runtime
11             ] [ -T trace-threshold ] [ -w workload ]
12

DESCRIPTION

14       oslat is an open source userspace polling mode stress program to detect
15       OS level latency.  The program runs a busy loop with no or various
16       workloads, collecting TSC information and measuring the time frequently
17       during the process.
18

OPTIONS

20       -b, --bucket-size=N
21              Specify the number of the buckets (4-1024).
22
23       -B, --bias=USEC
24              Add a bias to all the buckets using the estimated mininum.
25
26       -c, --cpu-list=CPULIST
27              Specify CPUs to run on.  For example, '1,3,5,7-15'.
28
29       -C, --cpu-main-thread=CORE
30              Specify which CPU the main thread runs on.  Default is cpu0.
31
32       -f, --rtprio=PRIORITY
33              Using specific SCHED_FIFO priority (1-99).  Otherwise use the
34              default priority, normally it will be SCHED_OTHER.
35
36       -m, --workload-mem=SIZE
37              Size of the memory to use for the workload (e.g., 4K, 1M).
38              Total memory usage will be this value multiplies 2*N, because
39              there will be src/dst buffers for each thread, and N is the
40              number of processors for testing.
41
42       -t, --runtime=SEC
43              Specify test duration, e.g., 60, 20m, 2H (m/M: minutes, h/H:
44              hours, d/D: days).  By default the unit is s/second.
45
46       -T, --trace-threshold=THRESHOLD
47              Stop the test when threshold triggered (in USEC).  At the
48              meantime, print a marker in ftrace and stop ftrace too.
49
50       -w, --workload=WORKLOAD
51              Specify a kind of workload, default is no workload.  Options:
52              "no", "memmove".
53
54       -s, --single-preheat
55              Use a single thread when measuring latency at preheat stage
56              NOTE: please make sure the CPU frequency on all testing cores
57              are locked before using this parmater.  If you don't know how to
58              lock the freq then please don't use this parameter.
59
60       -h, --help
61              Show the help message.
62
63       -v, --version
64              Show the version of the program.
65
66       -z, --zero-omit
67              Don't display buckets in the output histogram if all zeros.
68

AUTHOR

70       oslat was written by Peter Xu <peterx@redhat.com>.
71
72
73
74                                August 17, 2020                       OSLAT(8)
Impressum