Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 21 Aug 1996 22:28:09 +0200 (MET DST)
From:      Stefan Esser <se@zpr.uni-koeln.de>
To:        Eric Blood <eblood@winky.reno.nv.us>
Cc:        freebsd-bugs@freebsd.org
Subject:   Re: NCR825 problem
Message-ID:  <199608212028.WAA07186@x14.mi.uni-koeln.de>
In-Reply-To: <12434.840521241@winky>
References:  <12434.840521241@winky>

next in thread | previous in thread | raw e-mail | index | archive | help
Eric Blood writes:
 > 
 > On a SuperMicro motherboard, NCR825 SCSI, 1 gig wide SCSI drive
 > and a default kernel from both the 2.1.0 CD and the 2.2-960501-SNAP
 > CD I get the following:
 > 
 > changing root device to sd0a
 > de0: enabling AUI port
 > (very long pause)
 > ncr0: aborting job ...
 > ncr0:2: ERROR (90:0) (8-0-0) (0/13) @ (a54:50000000).
 >         script cmd = 740a8700
 > 	reg:     de 00 00 13 47 00 0f 0f 35 08 81 00 90 00 0f 02.

I assume that your CD-ROM drive is on SCSI ID 2 ?

Seems that some time-out occured. The error code
indicates that the NCR chip's internal processor
was interrupted (0x90 == DMA Fifo Empty + Abort)
by the host CPU.

It seems the NCR chip was waiting for the drive 
to complete some command, but it took too long
and a time expired.

 > ncr0: restart (fatal error).
 > sd0(ncr0:0:0): COMMAND FAILED (9 ff) @f0c30a00.
 > sd0(ncr0:0:0): WIDE SCSI (16 bit) enabled.
 > sd0(ncr0:0:0): FAST SCSI-2 100ns (10 Mb/sec) offset 8.
 > sd1(ncr0:1:0): FAST SCSI-2 100ns (10 Mb/sec) offset 8.

Seems the devices at SCSI ID 0 and 1 recovered 
and are back to normal again. What is the transfer
mode of device with ID 2 ?

 > Only once did this cause problems because the install program
 > said it couldn't see the CD.  Rebooting fixed this.

Guess your CD-ROM takes too long to complete self-test.
What model is it ?

 > When I install a custom kernel with all of the unnecessary drivers
 > removed, this error doesn't occur.  Also, for either case, the
 > system will stay up for days with no problems.

Hmm, might it have been caused by a cold boot at the
time you observed the aborted command, while the later
system start-up was a warm boot with all devices spun
up and ready ?

Regards, STefan



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199608212028.WAA07186>