Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!munnari.OZ.AU!news.ecn.uoknor.edu!feed1.news.erols.com!howland.erols.net!news.mathworks.com!fu-berlin.de!news-ber1.dfn.de!news-fra1.dfn.de!news-koe1.dfn.de!RRZ.Uni-Koeln.DE!usenet From: se@x14.mi.uni-koeln.de (Stefan Esser) Newsgroups: comp.unix.bsd.freebsd.misc Subject: Re: Problem with NCR 53c810 PCI SCSI Date: 22 Dec 1996 01:46:41 GMT Organization: Regional Computing Center, University of Cologne Lines: 83 Message-ID: <59i3u1$lpo@news.rrz.uni-koeln.de> References: <54g4g0$t45@nntp.hut.fi> <54o0ps$72n@nntp.hut.fi> NNTP-Posting-Host: annexr2-38.slip.uni-koeln.de Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Newsreader: knews 0.9.7 To: kja2@cc.hut.fi (Kari T Jokiranta) Xref: euryale.cc.adfa.oz.au comp.unix.bsd.freebsd.misc:32845 In article <54o0ps$72n@nntp.hut.fi>, > What wonders me is that BIOS always finds the disks, as does the > dos drivers that came with the PCI card. > I also tried Linux 2.0 bootdisk with NCR support several times, > and it booted OK every time. Hi! Sorry for the late reply, but I don't read USENET news more often than every other month. If you have any problem with the NCR driver under FreeBSD (or NetBSD or the BSD NCR driver under Linux) you should send mail to the FreeBSD mail lists (scsi@freebsd.org) or to my mail address directly <se@freebsd.org>. > I noticed that if I let the disks spin for 10-15 minutes before > I boot FreeBSD, the problem disappears. Seems like the NCR driver > doesn't like to coldboot my computer? This makes it even more likely, that there is some kind of hardware problem ... The BSD NCR driver puts a much higher load on the hardware than any other NCR SCSI driver I know of, since it tries to have the NCR work as an independent processor. All other NCR drivers I know about are written in such a way, that only very short code sequences are executed by the NCR, and then the host CPU decides what to do next. The NCR driver does also enable FAST SCSI and tags by default, and this may lead to problems with marginal hardware. > I include here some of the error messages as well as the messages when > it boots normally, so someone who's interested can take a look at them.. > > I booted FreeBSD with option -v, so the listings are long. > > -- dmesg when fail ---------------------------------------- > -- (I wonder where the beginning of the messages has disappeared. > -- Is there a limit in /dev/klog?). Sure, it must be covered by kernel RAM ... The default buffer size is 4KB. > (ncr0:5:0): COMMAND FAILED (9 ff) @f0ab394c. > ncr0: aborting job ... > ncr0:0: ERROR (90:0) (0-21-b8) (0/13) @ (418:43000060). > script cmd = 878b0000 > reg: da 00 00 13 47 00 05 1f 31 00 00 21 80 01 00 02. > ncr0: restart (fatal error). Your boot message log below doesn't show any SCSI device with ID 5. Seems there is something severely wrong ! > (ncr0:6:0): COMMAND FAILED (9 ff) @f0ab394c. > ncr0: aborting job ... > ncr0:0: ERROR (90:0) (0-21-b8) (0/13) @ (418:43000060). > script cmd = 878b0000 > reg: da 00 00 13 47 00 06 1f 31 00 00 21 80 01 00 02. > ncr0: restart (fatal error). > -- dmesg when success ------------------------------------ > > FreeBSD 2.1.5-RELEASE #0: Tue Oct 22 01:06:32 EET DST 1996 > chip0 <SiS 85c496> rev 49 on pci0:5 > ncr0 <ncr 53c810 scsi> rev 17 int a irq 11 on pci0:12 > (ncr0:0:0): "IBM WDS-3200 !J S550" type 0 fixed SCSI 2 > (ncr0:1:0): "IBM MXT-540SL !N JA6I" type 0 fixed SCSI 2 > (ncr0:2:0): "MICROP 1578-15 19U0" type 0 fixed SCSI 1 There is nothing found at SCSI ID 5 and 6 ... If your problem still exists, then do please send mail and give some more details: - How often does the boot fail ? - Does the system always work correctly, if the boot succeeded ? - Are all the devices internal ? - How long is the SCSI bus cable ? - These are old drives, which draw a high current at 12V. Is your power supply strong enough to start them all at the same time ? Regards, STefan