Detailed counter stats perf
Run with three -d for L1/LLC/TLB and stalled-cycle detail; reveals memory-bound vs CPU-bound
perf stat -d -d -d -- ./app
more perf
all 17 commands →
Scheduler latency analysis
perf sched record -- sleep 10 && perf sched latency --sort max
Syscall trace via perf
perf trace -s -p <pid>
Dynamic probe on function
perf probe --add 'tcp_sendmsg size' && perf record -e probe:tcp_sendmsg -a -- sleep 5
Record CPU cycles
perf record -g ./my_app
Show perf report
perf report