From owner-freebsd-current Wed Mar 8 08:10:09 1995 Return-Path: current-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id IAA08507 for current-outgoing; Wed, 8 Mar 1995 08:10:09 -0800 Received: from kitten.mcs.com (Kitten.mcs.com [192.160.127.90]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id IAA08490 for ; Wed, 8 Mar 1995 08:10:07 -0800 Received: from mailbox.mcs.com (Mailbox.mcs.com [192.160.127.87]) by kitten.mcs.com (8.6.10/8.6.9) with SMTP id KAA04034; Wed, 8 Mar 1995 10:09:39 -0600 Received: by mailbox.mcs.com (/\==/\ Smail3.1.28.1 #28.5) id ; Wed, 8 Mar 95 10:15 CST Received: by mercury.mcs.com (/\==/\ Smail3.1.28.1 #28.5) id ; Wed, 8 Mar 95 10:04 CST Message-Id: From: fredriks@mcs.com (Lars Fredriksen) Subject: Re: Swapper going sour? To: phk@ref.tfs.com (Poul-Henning Kamp) Date: Wed, 8 Mar 1995 10:04:37 -0600 (CST) Cc: fredriks@mcs.com, current@FreeBSD.org In-Reply-To: <199503080712.XAA28582@ref.tfs.com> from "Poul-Henning Kamp" at Mar 7, 95 11:12:57 pm X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 1352 Sender: current-owner@FreeBSD.org Precedence: bulk Poul-Henning Kamp writes: > > > > > Hi, > > On two kernels build yesterday and today I get the following > > mesage within a couple of minutes of startup. What gives? > > > > I/O to empty block???? > find this as see what it means. Oh, I will. > > ahb_scsi_cmd0: more than 33 DMA segs > ahb/drive becomes confused, possibly bad block. This to me seems to indicate to me that someone enqueued a bad buffer. > > swap_pager: I/O error - async pageout failed; blkno 4294967295, size 67584512, error 5 > swap write fails. These numbers cannot be right. Bkno = -1 and a size of more than 6MB. > > sd1: oops not queued > ahb/scsi is confused No it just refused to queue the job since it was greater than 33 DMA segs. > > biodone: buffer already done > confuses the kernel > > swap_pager_finish: I/O error, clean of page fbf000 failed > the world ends... No it does not. The machine works fine! > Hm, my request didn't result in the response I expected. It is probably my fault for being to terse. I didn't send it out as a request that someone hold my hand and walk the code with me, but rather as a heads up in case there is actually something wrong. Lars -- ------------------------------------------------------------------- Lars Fredriksen fredriks@mcs.com (home) lars@fredriks.pr.mcs.net (home-home) fredriks@asiago.cs.wisc.edu