Timeline for Number of branches reported by perf stat not equal to real number of test command
Current License: CC BY-SA 4.0
Post Revisions
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| 2 days ago | comment | added | ilnurKh | yes, some other actions can explain constant diff, but not the multiplication | |
| Apr 28 at 5:54 | comment | added | singhatulks | There is lot more that happened in the process than just the loop in your code. First ld had to load your binary and the shared libraries. Then you are also calling shared lib functions. So it makes sense that it's much higher than the number of iterations in your code. P.S. As an answer in related question suggested by SO says, use perf_event_open. | |
| Apr 26 at 15:13 | history | edited | RKou | CC BY-SA 4.0 |
edited body
|
| Apr 25 at 18:48 | comment | added | o11c |
I've found perf to be one of the buggiest parts of the kernel - it's extremely hardware-dependent, and I suspect something about it's not getting saved/restored properly when rescheduling happens (which is pretty frequent). It's quite possible that the recorded stats are actually a mix of your process and some other process (including the idle process) or kernel interrupts or ...
|
|
| Apr 25 at 17:14 | comment | added | Barmar | I suspect this has something to do with hardware branch prediction. | |
| S Apr 25 at 13:04 | review | First questions | |||
| Apr 25 at 15:06 | |||||
| S Apr 25 at 13:04 | history | asked | ilnurKh | CC BY-SA 4.0 |