1PERF-MEM(1) perf Manual PERF-MEM(1)
2
3
4
6 perf-mem - Profile memory accesses
7
9 perf mem [<options>] (record [<command>] | report)
10
12 "perf mem record" runs a command and gathers memory operation data from
13 it, into perf.data. Perf record options are accepted and are passed
14 through.
15
16 "perf mem report" displays the result. It invokes perf report with the
17 right set of options to display a memory access profile. By default,
18 loads and stores are sampled. Use the -t option to limit to loads or
19 stores.
20
21 Note that on Intel systems the memory latency reported is the
22 use-latency, not the pure load (or store latency). Use latency includes
23 any pipeline queueing delays in addition to the memory subsystem
24 latency.
25
27 <command>...
28 Any command you can specify in a shell.
29
30 -i, --input=<file>
31 Input file name.
32
33 -f, --force
34 Don’t do ownership validation
35
36 -t, --type=<type>
37 Select the memory operation type: load or store (default:
38 load,store)
39
40 -D, --dump-raw-samples
41 Dump the raw decoded samples on the screen in a format that is easy
42 to parse with one sample per line.
43
44 -x, --field-separator=<separator>
45 Specify the field separator used when dump raw samples (-D option).
46 By default, The separator is the space character.
47
48 -C, --cpu=<cpu>
49 Monitor only on the list of CPUs provided. Multiple CPUs can be
50 provided as a comma-separated list with no space: 0,1. Ranges of
51 CPUs are specified with -: 0-2. Default is to monitor all CPUS.
52
53 -U, --hide-unresolved
54 Only display entries resolved to a symbol.
55
56 -p, --phys-data
57 Record/Report sample physical addresses
58
59 --data-page-size
60 Record/Report sample data address page size
61
63 -e, --event <event>
64 Event selector. Use perf mem record -e list to list available
65 events.
66
67 -K, --all-kernel
68 Configure all used events to run in kernel space.
69
70 -U, --all-user
71 Configure all used events to run in user space.
72
73 -v, --verbose
74 Be more verbose (show counter open errors, etc)
75
76 --ldlat <n>
77 Specify desired latency for loads event. Supported on Intel and
78 Arm64 processors only. Ignored on other archs.
79
80 In addition, for report all perf report options are valid, and for
81 record all perf record options.
82
84 perf-record(1), perf-report(1)
85
86
87
88perf 01/12/2023 PERF-MEM(1)