Orange Pi5 kernel

^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  1) For a higher level overview, try: perf report --sort comm,dso
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  2) Sample related events with: perf record -e '{cycles,instructions}:S'
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  3) Compare performance results with: perf diff [<old file> <new file>]
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  4) Boolean options have negative forms, e.g.: perf report --no-children
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  5) Customize output of perf script with: perf script -F event,ip,sym
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  6) Generate a script for your data: perf script -g <lang>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  7) Save output of perf stat using: perf stat record <target workload>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  8) Create an archive with symtabs to analyse on other machine: perf archive
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  9) Search options using a keyword: perf report -h <keyword>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) Use parent filter to see specific call path: perf report -p <regex>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) List events using substring match: perf list <keyword>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) To see list of saved events and attributes: perf evlist -v
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) Use --symfs <dir> if your symbol files are in non-standard locations
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) To see callchains in a more compact form: perf report -g folded
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) Show individual samples with: perf script
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) Limit to show entries above 5% only: perf report --percent-limit 5
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) Profiling branch (mis)predictions with: perf record -b / perf report
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) To show assembler sample contexts use perf record -b / perf script -F +brstackinsn --xed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) Treat branches as callchains: perf report --branch-history
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) To count events in every 1000 msec: perf stat -I 1000
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) Print event counts in CSV format with: perf stat -x,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) If you have debuginfo enabled, try: perf report -s sym,srcline
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) For memory address profiling, try: perf mem record / perf mem report
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) For tracepoint events, try: perf report -s trace_fields
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) To record callchains for each sample: perf record -g
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) To record every process run by a user: perf record -u <user>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) Skip collecting build-id when recording: perf record -B
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) To change sampling frequency to 100 Hz: perf record -F 100
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) See assembly instructions with percentage: perf annotate <symbol>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) If you prefer Intel style assembly, try: perf annotate -M intel
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) For hierarchical output, try: perf report --hierarchy
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) Order by the overhead of source file name and line number: perf report -s srcline
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) System-wide collection from all CPUs: perf record -a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) Show current config key-value pairs: perf config --list
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) Show user configuration overrides: perf config --user --list
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) To add Node.js USDT(User-Level Statically Defined Tracing): perf buildid-cache --add `which node`
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) To report cacheline events from previous recording: perf c2c report
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) To browse sample contexts use perf report --sample 10 and select in context menu
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) To separate samples by time use perf report --sort time,overhead,sym
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) To set sample time separation other than 100ms with --sort time use --time-quantum
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) Add -I to perf record to sample register values, which will be visible in perf report sample context.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) To show IPC for sampling periods use perf record -e '{cycles,instructions}:S' and then browse context
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) To show context switches in perf report sample context add --switch-events to perf record.