From owner-freebsd-hackers Wed Jul 10 00:47:00 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id AAA18767 for hackers-outgoing; Wed, 10 Jul 1996 00:47:00 -0700 (PDT) Received: from critter.tfs.com ([140.145.230.252]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id AAA18762; Wed, 10 Jul 1996 00:46:57 -0700 (PDT) Received: from critter.tfs.com (localhost [127.0.0.1]) by critter.tfs.com (8.7.5/8.7.3) with ESMTP id JAA00363; Wed, 10 Jul 1996 09:46:25 +0200 (MET DST) To: Nate Williams cc: "Jordan K. Hubbard" , hackers@freebsd.org Subject: Re: [Fwd: Parallel laplink abuse leads to death of kernel secondary timer] In-reply-to: Your message of "Mon, 08 Jul 1996 23:29:51 MDT." <199607090529.XAA16499@rocky.mt.sri.com> Date: Wed, 10 Jul 1996 09:46:24 +0200 Message-ID: <361.836984784@critter.tfs.com> From: Poul-Henning Kamp Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Well, lets keep in mind that PLIP is a disgusting thing in the first place, but as we all know, in certain circumstances that may not matter. I agree that it should work intelligently and not do odd things to your system in any set of circumstances, but I don't think that I will be able to find the time needed for the correct change to this, until this problems stands in my way 3AM some morning or something similar. Sorry. Poul-Henning PS: Bruce has pointed out that tweaking which spl-level we use could be used to do it, (try splvm() and see if it helps :-) In message <199607090529.XAA16499@rocky.mt.sri.com>, Nate Williams writes: >> >> Yow, this one's pretty cool! :-) I guess we always knew that PLIP was a >> high-overhead proposition, but it's interesting to see that it only >> croaks on the Pentium. > >FWIW, I responded to this on Usenet and basically blamed his hardware. >I've used PLIP to mount NFS disks and done build worlds on the two >laptops I have, one a 486/75, and the other a Pentium/75. The 'servers' >have been my 486/66 at home and my P-100 ASUS box at work, and I've >never seen any problems with the timers dying. NFS mounting the disks >over PLIP was a *really* good way of generating an incredibly high >interrupt load on my 486/66, although the Pentium didn't seem to mind it >as much. Both laptops seemed to not notice it much since I suspect they >were CPU and/or I/O bound most of the time. > > >Nate -- Poul-Henning Kamp | phk@FreeBSD.ORG FreeBSD Core-team. http://www.freebsd.org/~phk | phk@login.dknet.dk Private mailbox. whois: [PHK] | phk@ref.tfs.com TRW Financial Systems, Inc. Future will arrive by its own means, progress not so.