Timeline for Simple C++ event loop - static vs dynamic dispatch performance
Current License: CC BY-SA 4.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 24 at 8:57 | vote | accept | Jnuk | ||
| Apr 24 at 6:21 | answer | added | G. Sliepen | timeline score: 2 | |
| Apr 24 at 5:50 | history | edited | G. Sliepen |
edited tags
|
|
| Apr 24 at 1:23 | comment | added | Rish |
Did you benchmark with optimizations enabled? Compiling with -O3, I find the run time more or less the same for both (which is kinda expected).
|
|
| Apr 23 at 23:24 | comment | added | J_H | This submission is about performance, yet it does not include any per-line CPU profiling measurements. It would benefit from some godbolt.org links that highlight what's different in the generated dispatch instructions. // Each example has an Event2 "distractor" thread going on, fine, counting down to negative infinity. It's unclear how much work it does. In particular, if there is a "task scheduling fairness" difference between the examples, we might be measuring that lambdas always do half as many decrements, or that they randomly did half as many on a particular run. | |
| Apr 23 at 22:22 | history | edited | toolic | CC BY-SA 4.0 |
deleted 11 characters in body
|
| S Apr 23 at 22:19 | review | First questions | |||
| Apr 23 at 22:22 | |||||
| S Apr 23 at 22:19 | history | asked | Jnuk | CC BY-SA 4.0 |