Messages in this thread | | | Date | Fri, 27 Jan 2012 10:20:45 -0800 | From | Arun Sharma <> | Subject | Re: [PATCH 00/13] perf_events: add support for sampling taken branches (v3) |
| |
On 1/27/12 4:09 AM, Peter Zijlstra wrote: >> Here is a kernel example, where we want to sample indirect calls: >> $ perf record -a -C 1 -b ind_call -e r1c4:k sleep 10 >> $ perf report -b >> # >> # Overhead Source Symbol Target Symbol >> # ........ .......................... .......................... >> # >> 36.36% [k] __delay [k] delay_tsc >> 9.09% [k] ktime_get [k] read_tsc >> 9.09% [k] getnstimeofday [k] read_tsc >> 9.09% [k] notifier_call_chain [k] tick_notify >> 4.55% [k] cpuidle_idle_call [k] intel_idle >> 4.55% [k] cpuidle_idle_call [k] menu_reflect >> 2.27% [k] handle_irq [k] handle_edge_irq >> 2.27% [k] ack_apic_edge [k] native_apic_mem_write >> 2.27% [k] hpet_interrupt_handler [k] hrtimer_interrupt >> 2.27% [k] __run_hrtimer [k] watchdog_timer_fn >> 2.27% [k] enqueue_task [k] enqueue_task_rt >> 2.27% [k] try_to_wake_up [k] select_task_rq_rt >> 2.27% [k] do_timer [k] read_tsc >> >> Due to HW limitations, branch filtering may be approximate on >> Core, Atom processors. It is more accurate on Nehalem, Westmere >> and best on Sandy Bridge. > > Can I have you ACK on this userspace stuff (patches 11-13)?
While the current "Source -> Target" based UI works well for many cases, it'd be nice to have "-g -b any_call" to result in a callgraph like output, so for userspace programs compiled without frame pointers, we get a limited callgraph.
-Arun
| |