Marcelo Tosatti wrote:
> On Wed, 31 May 2000, Arne Thomassen wrote:
> > W Replace the big kernel lock with more fine-grained locking as far as
> > possible (completely? ;-)
>
> The correct thing to do is to fine-grain when we notice contention.
> Fine-graining "as much as possible" is a direct way to hell.
W Automatic compile-time lock correctness checker :-)
Note that excessive fine-graining is also a direct way to *slow
everything down*. Even uncontended locks have overhead, and more locks
mean more overhead. Which is why Marcelo is right from a "raw speed"
perspective. Of course it is also right from a software engineering
perspective, but a lock correctness checker would simplify that :-)
-- 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/
This archive was generated by hypermail 2b29 : Wed Jun 07 2000 - 21:00:12 EST