*BSD News Article 18449


Return to BSD News archive

Newsgroups: comp.os.386bsd.questions
Path: sserve!newshost.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!haven.umd.edu!uunet!world!hd
From: hd@world.std.com (HD Associates)
Subject: Re: Microcode bug in AHA1542C?
Message-ID: <CA7Fon.FAp@world.std.com>
Organization: The World Public Access UNIX, Brookline, MA
References: <211cnr$gpp@cleese.apana.org.au> <219qmi$2ih@wzv.win.tue.nl>
Date: Thu, 15 Jul 1993 11:53:10 GMT
Lines: 49

In article <219qmi$2ih@wzv.win.tue.nl> guido@gvr.win.tue.nl (Guido van Rooij) writes:
>newton@cleese.apana.org.au (Mark Newton) writes:
>
>
>>I've recently bought an Adaptec 1542C so that I'd have something to plug
>>my 170Mb SCSI disk into.  Things are not going smoothly, however.
>
>>Writing to the SCSI disk causes hundreds of errors (including "partially
>>allocated inode", "incorrect block count", "unknown file type", and
>>many more) next time fsck runs over its filesystems.
>
>>Rodney Grimes mentioned a microcode bug in some 1542C's that causes data
>>corruption under UNIX and suggested that I call Adaptec tech support about
...
>
>The rumour was that teh 1542C's contained a micro-code bug. But as
>far as I know, it is due to the fact that the (early) 1542C's are
>extermely sensitive to wrong cabling. I don't know the details
>but I think you should check your cabling. The newer 1542C's shouldn't
>have this problem anymore. (you can always ask o comp.periphs.scsi)
>

I believe this is right.  I was one of the people having all kinds of problems
with the 1542C - completely failing for me until I got a new SCSI-II
cable from Black Box for my external peripherals, and then working with
some external peripherals, not working with others, working only with the
termination purposely screwed up and so on.  I have several old devices
around I use for testing and it was a real problem.

I returned the board due to a target mode firmware bug and got it back
with new firmware and Adaptec's modification to the REQ/ACK signals.
Everything now works perfectly.

First be sure you have SCSI-II cables.  There is some discussion of
this over in comp.periphs.scsi.  I would buy them from a good mail
order outfit that knows cables and not a corner computer store.

If you still have problems, tell your distributor that you are having
problems due to the REQ/ACK signals and you know that Adaptec has an
engineering fix for it and you'd like to exchange your board for one
that has the fix.  See if that flies.

Adaptec is probably right about their board being OK and the cabling /
peripherals in the wrong, but the board was a real pain in the neck
until I got the modification.
-- 
Peter Dufault               Real Time Machine Control and Simulation
HD Associates               Voice: 508 433 6936
hd@world.std.com            Fax:   508 433 5267