Date: Thu, 04 Dec 1997 17:45:56 -0800 From: Julian Elischer <julian@whistle.com> To: Poul-Henning Kamp <phk@freebsd.org> Cc: current@freebsd.org Subject: Re: weird current behaviour... Message-ID: <34875CD4.7566F4CF@whistle.com> References: <8106.881230997@critter.freebsd.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
All the fast ones only have interrupt context processing. the slow ones have a userland context. (the ping process) The rc456 programs are finishing up their quantum before allowing the ping to run and recieve the response. This is possibly because they have a higher priority than the ping becsause the idlprio stuff doesn't really update or change the priority.. Poul-Henning Kamp wrote: > > Network: > > machine-A en0 ------------ en0 machine-B > de0 fxp0 > | \ > | ----------------- de0 machine-C > | > hub ----- machine-D > | > | > ed0 > machine-E > > These two cases: > > 1. ping from A to E > 2. ping from A to D > > I see the round-trip time flutter from 9msec to 70 msec... > > Every other path the round-trip time is OK, including confusingly > enough: > > from E to A > from D to A > from B&C to E > from B&C to D > > What gives ? > > Is the if_de driver broken somehow ? > > Does anybody else see this ? > > Machine-A is a PR440FX motherboard with 2xP6/233 running current-SMP. > > I have tried two different de0 cards and both does this. > > -- > Poul-Henning Kamp FreeBSD coreteam member > phk@FreeBSD.ORG "Real hackers run -current on their laptop."
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?34875CD4.7566F4CF>