RE: bogomips ???

Stephen C. Tweedie (sct@redhat.com)
Fri, 23 Apr 1999 23:16:57 +0100 (BST)


Hi,

On Fri, 23 Apr 1999 14:58:27 +0200, BROWN Nick <Nick.BROWN@coe.fr> said:

> On my Dell Pentium-II 233 MHz, I have 231.<mumble> bogomips. So why do I
> have 358.<blarg> bogomips on my Pentium MMX 180MHz, which is most definitely
> slower (if it was accidentally twice as fast, I'd have noticed when
> compiling the kernel!) ?

See, I _said_ it was a bogus measurement of performance!

It only measures the speed of a single delay loop. That can be
profoundly affected by seemingly small details of the CPU's internal
implementation. It has been known for the bogomips to almost double in
value after the insertion of a no-operation assembler instruction into
the delay loop: it just depends on how that affects the CPU pipeline.

--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.tux.org/lkml/