Freebsd Theme Song

Kris Kennaway kris at obsecurity.org
Sat Dec 10 13:24:28 PST 2005


On Sat, Dec 10, 2005 at 01:14:18PM -0800, Danial Thom wrote:

> Well thats just hogwash Kris. Pure bridging
> performance is a measure of the efficiency of the
> kernel to do rote tasks like respond to
> interrupts, and the latencies in performing those
> tasks. Its the best way, IMO, to exercise and
> measure the efficiency of an isolated kernel. It
> requires no userland activity, so your results
> aren't muddled by millions of system calls. Its a
> way to compare apples to apples, which is how
> good testing is done.
> 
> As long as you don't have your filesystem on a
> network, you're in good shape. But thats not even
> the point. The point is that the purpose of
> tearing apart 4.x was to go MP, and MP
> performance is dismill across the board.

This statement is simply false.  It's actually quite funny to read.

For the readers at home: Denial is once again taking his narrow view
of the world ("everything about the OS is accurately measured by how
fast the kernel routes network packets!") and extrapolating it to
infinity, then jumping up and down about it.

Kris
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-questions/attachments/20051210/5b4e3b3a/attachment.bin


More information about the freebsd-questions mailing list