Re: A different performance measure

Hi,

First, I would like to congratulate Dave, Andy, and all others that
have worked to reduce context switch times. This work should
a very positive impact on overall system performance.

[Dave Hudson <dave@humbug.demon.co.uk> writes]
> Running with my DX/4-100 test system and a kernel built under gcc 1.42, this
> is now showing 15 us context switches :-) (measured over 1M switch ops).
> The same system with a DX/2-66 (not the same system I usually mention for a
> DX/2) shows 21 us. FWIW with this DX/2 system I get a perf1 figure of 75
> us.

perf1 attempts to measure the time to do a context switch + the time to
send a message that doesn't contain any data. Now, the difference between
the 21 us number and the 75 us number is awfully big. Is the 54 us
difference totally due to the messaging overhead? Or perhaps the
perf1/sched_yield() measuring something different that the old perf1?

                                        Mike Larson
Received on Fri Jan 20 09:48:13 1995