I disagree. While most non-algorithmic optimisations probably aren't
worth the effort, there are a few tricks of the trade that are very easy
and really make the performance difference. (With present technology).
One of these is dividing a function into a common case which runs fast,
and a general case which runs slower. Adding the keyword "inline" when
the common case code is tiny is just a final tweak. Really, the
compiler could inline those tiny functions itself; the problem is that
the compiler doesn't know that the general case will only be called
rarely. (Profile driven compilation may fix this, but the technology's
not readily available yet).
-- Jamie
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/