On Tue, 2 Jul 2002, Keith Owens wrote:
> For netfilter, the use count reflects the number of packets being
> processed. Complex and potentially high overhead.
<blink>
_Why_???
If some rule in your chains needs a module foo, you probably want foo
to stay around until that rule is removed. Even if there's no traffic
whatsoever.
So what's the problem with making use count equal to number of rules
referencing the module? You need exclusion between chain changes and
chain traversing anyway...
> All of this requires that the module information be passed in multiple
> structures and assumes that all code is careful about reference
> counting the code it is about to execute. There has to be a better
> way!
You know, I'd rather trust core code to do something right than expect
all drivers to follow any non-trivial rules (and "you should not block
in <areas>" _is_ non-trivial enough).
No comments on the network devices - I'll need to read through the code
to answer that one...
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Jul 07 2002 - 22:00:10 EST