mirror of
https://github.com/RIOT-OS/RIOT.git
synced 2024-12-29 04:50:03 +01:00
58ae12d000
Sadly GCC currently doesn't optimize as expected and the additional return value increases the ROM a bit. As a result, the application doesn't fit into the ROM of four additional boards. |
||
|---|---|---|
| .. | ||
| main.c | ||
| Makefile | ||
| Makefile.ci | ||
| README.md | ||
Benchmark for sys/atomic_utils
This application will perform 100.000 repetitions (or 1.000.000 on
Cortex-M7 and ESP32) for each atomic operation and will print the total time it
took in a table. For comparison, the speed of C11 atomics and plain volatile
accesses are also printed.
Expectations
Lower is better!
Plain volatile accesses are not atomic, and therefore should perform faster
than actual atomic operations. If atomic operations turn out to be faster
(by more than rounding errors and noise), something is odd.
The atomic_utils aim to be at least as fast as C11 atomics and often faster.
If the atomic_utils implementation performs slower than C11 atomics, you have
found potential for further optimization.