i did see that option but when i selected it i got no output; maybe i have to look for the output somewhere else?
the thing i was surprised by was not really that i could beat it by an instruction or two by rewriting the c code, but that i could reduce the number of instructions in the inner loop by 37% with no apparent tradeoffs by hand-applying an optimization that the compiler already does apply, itself, on arm
Seems it doesn't like -S being present (and it makes sense as you don't want assembly, you want a disassembled object file). In general there's no need for -S in compiler explorer.
the thing i was surprised by was not really that i could beat it by an instruction or two by rewriting the c code, but that i could reduce the number of instructions in the inner loop by 37% with no apparent tradeoffs by hand-applying an optimization that the compiler already does apply, itself, on arm