Return to BSD News archive
Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.uwa.edu.au!DIALix!not-for-mail From: julian@perth.DIALix.oz.au (Julian Elischer) Newsgroups: comp.os.386bsd.questions Subject: Re: BT946C problem with NetBSD 1.0 Date: 23 Dec 1994 02:59:18 +0800 Organization: DIALix Services, Perth, Australia. Lines: 24 Sender: julian@perth.DIALix.oz.au Message-ID: <3dcia6$nff$1@perth.DIALix.oz.au> References: <3d5gg7$3b6@titania.pps.pgh.pa.us> NNTP-Posting-Host: perth.dialix.oz.au NNTP-Posting-User: julian X-Newsreader: NN version 6.5.0 #59 (NOV) ksulliva@oberon.pps.pgh.pa.us (Kevin Sullivan) writes: >Dec 19 02:01:08 sparta /netbsd: bt0: mbi at 0xf870ca34 should be found, stat=81. >.resync At TFS.COM we've been experimenting with this error... it seems to us that the best thing to do is comment out the printf, because the condition is almost unavoidable occasionally with busy multiple scsi devices and under OSF1 (yes the same drivers) it actually made things worse becasue while it was scrolling up the error messages, it missed further SCSI interrupts. :( We'll be doing some more testing on it but it seems that this 'error' is benign, though there may be a better way of handling the devices that reduce it's frequency.. it happens when an interrupt service call services two requests, but the second interrupt has already been posted, so when it is serviced, it doesn't find anything to do.. and prints that (slightly misleading) message. julian