On Sat, 18 Jul 1998 13:55:10 +0200, Andrew Derrick Balsa
<andrebalsa@altern.org> said:
> The "Memory Rusting Effect" however is something completely
> different. People have been detecting memory fragmentation problems in
> the 2.1.x kernels for some time now. Those problems don't exist on
> 2.0.x kernels.
That's because the caches which they are concerned with don't exist on
2.0. That's not an excuse for not getting the cache management right,
of course.
> However, many developers believe that adding more layers of code to
> tackle special cases will help solve the problem. I believe this is
> the wrong approach.
Absolutely. For what it's worth, 2.1.110 is radically simpler in most
cases regarding vm/cache management, and performs enormously better on
low memory. :)
> Adding code to tackle the special cases increases code complexity,
> gains nothing in terms of performance for the general case, and
> ultimately leads to statements of the kind: "If you only have 8Mb of
> RAM on your Linux box, better stick to 2.0.x kernels". :-(
Your statement is in itself a generalisation which omits some important
special cases and which is, therefor, inherently flawed. For what it's
worth, 2.1.110 is so much simpler that for some very important special
cases it simply ceases to work. Just a little special case code deals
with that nicely and results in what appears to be a very stable
system. The trick is to make sure that the special case code does not
disturb the normal running of the system. What went wrong with 2.1 was
that people were trying to fix some special cases by applying all sorts
of rules which _seriously_ affected the normal cases; most of that no
longer applies in 2.1.110.
--Stephen
-
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.altern.org/andrebalsa/doc/lkml-faq.html