Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 14 Jan 2003 12:37:02 -0700
From:      "Justin T. Gibbs" <gibbs@scsiguy.com>
To:        Joerg Wunsch <joerg_wunsch@uriah.heep.sax.de>
Cc:        freebsd-scsi@freebsd.org
Subject:   Re: quirk candidate: SEAGATE ST52160N
Message-ID:  <579060000.1042573021@aslan.btc.adaptec.com>
In-Reply-To: <20030114201301.A19952@uriah.heep.sax.de>
References:  <20030113220055.D50550@uriah.heep.sax.de> <147830000.1042501904@aslan.btc.adaptec.com> <20030114074853.A71485@uriah.heep.sax.de> <774150000.1042549034@aslan.scsiguy.com> <20030114143505.A12331@uriah.heep.sax.de> <808290000.1042552117@aslan.scsiguy.com> <20030114201301.A19952@uriah.heep.sax.de>

next in thread | previous in thread | raw e-mail | index | archive | help
> As Justin T. Gibbs wrote:
> 
>> > Isn't byte 0 the `35' for the command?
> 
>> Byte 0 of the sense data.
> 
> Ah, now i see.
> 
>> The strange thing is that the bogus
>> error code varied.  I wonder if we are somehow corrupting the
>> sense data.
> 
> Hmm, i've got remote GDB on that machine.  Here's the contents of
> ccb.ccb_h and ccb.csio after the command:

It's a bit hard to parse in the format that gdb provides, but it
looks like the sense data is utterly bogus in this case.

> If you've got a few more hints where to trace something,
> that should be easy to do.  The machine is a scratch box
> only which i can reboot at will.

What you should probably do is trace the sense buffer both before
and after the sense is retrieved.  Perhaps fill the target sense
buffer with a known pattern, log the transfer residual, and see if
the number of bytes overwritten is as you would expect.  That should
show that the drive is really returning bogus data.  It's either that
or stick on a SCSI bus analyzer if you happen to have access to one.

I don't recall from your earlier mail.  Is this an ahc controller?  If
so, some code in ahc_handle_seqint's BAD_STATUS handler and the SCB_SENSE
case in ahc_done() should provide the logging you need.

--
Justin


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-scsi" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?579060000.1042573021>