Far be it from to put down performance work, but I think you are missing
the point. When I put that test into lmbench, I put in fork/exit
vs for/exec("/tmp/foo")/exit vs fork/exec("sh -c /tmp/foo")/exit.
The point was /not/ that fork/exit are too slow but that execing /bin/sh
-c /tmp/foo was very slow. The first can be fairly easily speeded up,
but so what? It doesn't happen in any performance critical path that
I've seen. The latter is much harder to speed up, but it's where the
real rewards are.
Short summary: don't worry about shaving 10 usecs off fork/exit when you
can shave 10,000 usecs off /bin/sh -c.
-
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/