From owner-freebsd-questions Tue Feb 27 04:36:46 1996 Return-Path: owner-questions Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id EAA15607 for questions-outgoing; Tue, 27 Feb 1996 04:36:46 -0800 (PST) Received: from Sysiphos (Sysiphos.MI.Uni-Koeln.DE [134.95.212.10]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id EAA15602 for ; Tue, 27 Feb 1996 04:36:43 -0800 (PST) Received: by Sysiphos id AA29101 (5.67b/IDA-1.5 for questions@freebsd.org); Tue, 27 Feb 1996 13:35:53 +0100 Message-Id: <199602271235.AA29101@Sysiphos> From: se@zpr.uni-koeln.de (Stefan Esser) Date: Tue, 27 Feb 1996 13:35:52 +0100 In-Reply-To: Archie Cobbs "Re: My NCR810 dies on -stable, -current, but not 2.1.0-release" (Feb 26, 15:08) X-Mailer: Mail User's Shell (7.2.6 alpha(2) 7/9/95) To: Archie Cobbs Subject: Re: My NCR810 dies on -stable, -current, but not 2.1.0-release Cc: questions@freebsd.org Sender: owner-questions@freebsd.org Precedence: bulk On Feb 26, 15:08, Archie Cobbs wrote: } Subject: Re: My NCR810 dies on -stable, -current, but not 2.1.0-release } } > On Feb 26, 0:37, "Brian Litzinger" wrote: } > } Subject: My NCR810 dies on -stable, -current, but not 2.1.0-release } > } Earlier I reported that my NCR 810 PCI controllers was locking up } > } after the 'changing root device to sd0a'. } > } } > } I happened to wait for about 30 seconds with the system stuck and } > } got this message: } > } } > } changing root device to sd0a } > } [about 30 second delay] } > } ncr0: aborting job ... } > } ncr0:1: ERROR (90:0) (8-0-0) (0/13) @ (c84:50000000). } > } script cmd = 740a8700 } > } reg: de 00 00 13 47 00 06 1f 35 08 00 00 90 00 0f 02. } > } ncr0: restart (fatal error) } > } sd0(ncr0:0:0) COMMAND FAILED (9 ff) @f0a21800. } > } sd0(ncr0:0:0) FAST SCSI-2 100ns (10 Mb/sec) offset 8. } > } panic: cannot mount root } > } > No, but I'm surprised you see a difference } > between 2.1R and -stable ... } } In trying to do an installation of 2.1R the other day from a SCSI } CD-ROM (TEAC cd-50) to a SCSI disk, a very similar error (visible on } virtual terminal #2) prevented the installation from ever getting beyond } about 15-20% of the bin distribution (I made several attempts). Hmm, if it works at all, then it really should work all the way ... Except if there are problems with the SCSI bus cable/terminators, for example, which cause random errors. } I tried a different CD-ROM (a generic quad speed) and everything worked. } The original double speed cd-rom is a lot older and probably just } sucks in general. The door gets jammed sometimes too. I think I'm going } to throw it away. Ok. There is one difference between 2.1 and 2.0.5: Due to a oversight, CDROMs were used with synch. transfers under 2.1, while only asynch. transfers had been used for them under a GENERIC 2.0.5 kernel. } I don't have the exact screen output, but it looked very similar and I } definately remember the "ncr0: restart (fatal error)" line. The motherboard } is an ASUS SP3-G motherboard (with a rev 2 ncr chip). Sorry I don't have } more details right now, but FWIW... Well, the hardware should be supported better than anything else, since this is the development base for the PCI and NCR code :) The problem you describe can be caused by roo many terminators, for example, and it is well possible that slower transfers still work, but that FAST transfers suffer ... Regards, STefan -- Stefan Esser, Zentrum fuer Paralleles Rechnen Tel: +49 221 4706021 Universitaet zu Koeln, Weyertal 80, 50931 Koeln FAX: +49 221 4705160 ============================================================================== http://www.zpr.uni-koeln.de/~se