1perf-lock(1) 2============ 3 4NAME 5---- 6perf-lock - Analyze lock events 7 8SYNOPSIS 9-------- 10[verse] 11'perf lock' {record|report|script|info|contention} 12 13DESCRIPTION 14----------- 15You can analyze various lock behaviours 16and statistics with this 'perf lock' command. 17 18 'perf lock record <command>' records lock events 19 between start and end <command>. And this command 20 produces the file "perf.data" which contains tracing 21 results of lock events. 22 23 'perf lock report' reports statistical data. 24 25 'perf lock script' shows raw lock events. 26 27 'perf lock info' shows metadata like threads or addresses 28 of lock instances. 29 30 'perf lock contention' shows contention statistics. 31 32COMMON OPTIONS 33-------------- 34 35-i:: 36--input=<file>:: 37 Input file name. (default: perf.data unless stdin is a fifo) 38 39-v:: 40--verbose:: 41 Be more verbose (show symbol address, etc). 42 43-q:: 44--quiet:: 45 Do not show any message. (Suppress -v) 46 47-D:: 48--dump-raw-trace:: 49 Dump raw trace in ASCII. 50 51-f:: 52--force:: 53 Don't complain, do it. 54 55--vmlinux=<file>:: 56 vmlinux pathname 57 58--kallsyms=<file>:: 59 kallsyms pathname 60 61 62REPORT OPTIONS 63-------------- 64 65-k:: 66--key=<value>:: 67 Sorting key. Possible values: acquired (default), contended, 68 avg_wait, wait_total, wait_max, wait_min. 69 70-F:: 71--field=<value>:: 72 Output fields. By default it shows all the fields but users can 73 customize that using this. Possible values: acquired, contended, 74 avg_wait, wait_total, wait_max, wait_min. 75 76-c:: 77--combine-locks:: 78 Merge lock instances in the same class (based on name). 79 80-t:: 81--threads:: 82 The -t option is to show per-thread lock stat like below: 83 84 $ perf lock report -t -F acquired,contended,avg_wait 85 86 Name acquired contended avg wait (ns) 87 88 perf 240569 9 5784 89 swapper 106610 19 543 90 :15789 17370 2 14538 91 ContainerMgr 8981 6 874 92 sleep 5275 1 11281 93 ContainerThread 4416 4 944 94 RootPressureThr 3215 5 1215 95 rcu_preempt 2954 0 0 96 ContainerMgr 2560 0 0 97 unnamed 1873 0 0 98 EventManager_De 1845 1 636 99 futex-default-S 1609 0 0 100 101-E:: 102--entries=<value>:: 103 Display this many entries. 104 105 106INFO OPTIONS 107------------ 108 109-t:: 110--threads:: 111 dump thread list in perf.data 112 113-m:: 114--map:: 115 dump map of lock instances (address:name table) 116 117 118CONTENTION OPTIONS 119-------------- 120 121-k:: 122--key=<value>:: 123 Sorting key. Possible values: contended, wait_total (default), 124 wait_max, wait_min, avg_wait. 125 126-F:: 127--field=<value>:: 128 Output fields. By default it shows all but the wait_min fields 129 and users can customize that using this. Possible values: 130 contended, wait_total, wait_max, wait_min, avg_wait. 131 132-t:: 133--threads:: 134 Show per-thread lock contention stat 135 136-b:: 137--use-bpf:: 138 Use BPF program to collect lock contention stats instead of 139 using the input data. 140 141-a:: 142--all-cpus:: 143 System-wide collection from all CPUs. 144 145-C:: 146--cpu:: 147 Collect samples only on the list of CPUs provided. Multiple CPUs can be 148 provided as a comma-separated list with no space: 0,1. Ranges of CPUs 149 are specified with -: 0-2. Default is to monitor all CPUs. 150 151-p:: 152--pid=:: 153 Record events on existing process ID (comma separated list). 154 155--tid=:: 156 Record events on existing thread ID (comma separated list). 157 158--map-nr-entries:: 159 Maximum number of BPF map entries (default: 10240). 160 161--max-stack:: 162 Maximum stack depth when collecting lock contention (default: 8). 163 164--stack-skip 165 Number of stack depth to skip when finding a lock caller (default: 3). 166 167-E:: 168--entries=<value>:: 169 Display this many entries. 170 171 172SEE ALSO 173-------- 174linkperf:perf[1] 175