Re: guaranteed no-op function pointer?

["bluejack" <bluujack@xxxxxxxxx>]

FYI: Doing some testing on a variety of hardware suggests that calling
a function takes about 3 times as long as testing a variable, which
means that in the common scenario, testing the variable and then not
calling is far better than calling a function that does nothing,
although the less happy case of testing the variable and calling the
function is, accordingly more expensive.