Re: [patch] lockless, scalable get_pid(), for_each_process() elimination, 2.5.35-BK

From: William Lee Irwin III (wli@holomorphy.com)
Date: Tue Sep 17 2002 - 21:22:42 EST


From: William Lee Irwin III <wli@holomorphy.com>
> The issues addressed here are extremely important for the workloads I
> must support.

On Tue, Sep 17, 2002 at 06:27:22PM -0700, David S. Miller wrote:
> Have you published test tools that emulate your workload?
> These would be very useful, probably to find problems even
> outside the scope of pid lookups.

By and large I've been using existing tools, for instance, multiple
tiobench instances to obtain a large task count. The test cases I've
had to construct by hand are generally meant to trigger pagetable OOM,
which I've been assigned to do something about. In general, I attempt
to simulate the VM and I/O characteristics of databases. Sometimes I
have to get a bit inventive, e.g. recent 2*dbench 512 on tmpfs test.

Thanks,
Bill
-
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 : Mon Sep 23 2002 - 22:00:21 EST