Subject: Re: Faster pipes, now with uvm_loan()
To: None <tech-kern@netbsd.org, port-vax@netbsd.org>
From: Anders Magnusson <ragge@ludd.luth.se>
List: tech-kern
Date: 04/26/2001 10:17:56
>> So: no much changes in latency on vax; the delay may be caused by
>> something else (context switching maybe?)
> On a VAX? Probably that darn "calls" instruction :-)
>
It's more likely that it is the time measurement. The smallest measurable
time difference on a MicroVAX (of that type) is 10mS. Look at these
ping times, for example:
PING olivaw.ludd.luth.se (130.240.16.39): 56 data bytes
64 bytes from 130.240.16.39: icmp_seq=0 ttl=255 time=29.998 ms
64 bytes from 130.240.16.39: icmp_seq=1 ttl=255 time=19.999 ms
64 bytes from 130.240.16.39: icmp_seq=2 ttl=255 time=19.999 ms
-- Ragge