27.9.1. Microbenchmarks
It eventually has to come to that, hasn’t it?
-
userland/gcc/busy_loop.c described at C busy loop
Of course, there is a continuum between what is a "microbenchmark" and a "macrobechmark".
One would hope that every microbenchmark exercises a concentrated subset of part of an important macro benchmark, otherwise what’s the point, right?
Also for parametrized "macro benchmark", you can always in theory reduce the problem size to be so small that it might be more appropriate to call it a micro benchmark.
So our working definition will be more of the type: "does it solve an understandable useful high level problem from start to end?".
If the answer is yes, then we call it a macro benchmark, otherwise micro.
Bibliography: