4 ms·
We’d considered using mflat, but we’re not that performance constrained (and prefer the slightly smaller binary size with register windows enabled). I may do s
by jakeinspace 3y ago
We’d considered using mflat, but we’re not that performance constrained (and prefer the slightly smaller binary size with register windows enabled). I may do some profiling of the under flow/overflow interrupts though since you’ve now got me second guessing myself.
- okl 3y agoRegisters asr22/23 contain a cycle counter that you can use to time stuff. If it's not present, there's a register in the DSU that counts cycles but that requires an access via the AHB bus. You can measure a lot of things with those cycle counters, like context switch and interrupt handling times, memcpy vs naive for-loop, linear vs. binary search on small arrays... I'd expect a few microseconds per overflow at most but it depends a lot on the characteristics of the system. Of course, if the application is not sensitive to a few microseconds here and a few microseconds there that optimization might not be worth it.