perf commands
17 commands, each with a one-line explanation. Click a command to copy it, or its title for details.
Commands 17 shown
-
perf record -g ./my_appRecord performance data with call-graph info
-
perf reportInteractive display of recorded perf data
-
perf listList all supported perf events on current system
-
perf stat ls -laShow hardware counters for a command
-
perf record -e cache-misses ./my_appRecord only cache-miss events
-
perf topShow live per-function CPU usage (like top)
-
perf annotateShow annotated assembly for hottest function
-
perf record -F 99 -ag -- sleep 30 && perf script > out.perfCollect data suitable for flamegraph generation
-
perf stat -e syscalls:sys_enter_read ./my_appCount specific syscall events
-
perf record -p 1234 sleep 5Profile an already-running process for 5 seconds
-
perf top -g --call-graph dwarfLive top hot functions with caller chains; dwarf unwind works without frame pointers
-
perf record -F 99 -a -g -- sleep 30 && perf report --stdioSample whole machine at 99Hz for 30s, then print collapsed report; 99Hz dodges lockstep aliasing
-
perf script | stackcollapse-perf.pl | flamegraph.pl > flame.svgTurn perf.data into an interactive flame graph SVG (Brendan Gregg FlameGraph tools)
-
perf stat -d -d -d -- ./appRun with three -d for L1/LLC/TLB and stalled-cycle detail; reveals memory-bound vs CPU-bound
-
perf sched record -- sleep 10 && perf sched latency --sort maxRecord context switches then show per-task wakeup latency; find runqueue starvation
-
perf trace -s -p <pid>strace-like syscall trace with -s summary; lower overhead, no ptrace stop-the-world
-
perf probe --add 'tcp_sendmsg size' && perf record -e probe:tcp_sendmsg -a -- sleep 5Add a kprobe with an argument, then record it; inspect kernel internals without rebuilding
no commands match — try a different search