[2412] in linux-scsi channel archive

home help back first fref pref prev next nref lref last post

Re: ADAPTEC AHA2940AU with Linux

daemon@ATHENA.MIT.EDU (Dave Wreski)
Fri Sep 5 05:56:41 1997

Date: 	Thu, 4 Sep 1997 17:01:15 -0400 (EDT)
From: Dave Wreski <dave@nic.com>
To: Rob Block <rblock@Eng.Sun.COM>
cc: Russell Price <rjprice@eagle.ais.net>, Doug Ledford <dledford@dialnet.net>,
        linux-scsi@vger.rutgers.edu
In-Reply-To: <199709040052.RAA17898@century.eng.sun.com>



On Wed, 3 Sep 1997, Rob Block wrote:

> Actually, I upgraded to 2.0.30 (and even installed the pre-7 patch for
> 2.0.31).  Now the card is recognized, though I'm having some other
> problems... I have a UMAX scanner, and if the scanner cable is attached to
> the external port (even with the scanner turned off), the kernel panics
> during scsi initialization, with a message like:
> 	"Error: something reset the scsi bus"
> If I disconnect the cable and reboot the system comes up okay.

Guys,

If you really need to get this card working right away, or would just like
to help find the bugs, you could try the following new patch just
released:

ftp.dialnet.net/pub/linux/aic7xxx/aic7xxx-abort-Sep04.patch.gz

There are also other files, including a README, that you should check out.
I've cc'd Doug, as I'm sure he'd like to hear your success or failure
stories..

Incidentally, I know that the adaptec has very touchy termination and
resistence requirements, and requires very good scsi cables for proper
operation.

Dave


> 
> (Next time I'm at my PC I'll write down the exact panic message I got and
> repost it.)
> 
> I tried changing my lilo.conf parameter from
> 	"append aic7xxx=ultra"
> to
> 	"append aic7xxx=no_reset,ultra"
> (on a suggestion I read) but that made no difference.
> 
> This kind of defeats the purpose of my getting the card to begin with,
> so I can use the scanner with Linux.  If I reconnect the scanner after
> bootup and run the umax driver utility, it doesn't find the scanner.
> (That might be perfectly normal, I'm new at using SCSI...)
> I looked at the SCSISelect boot utility that comes up before the OS, but
> didn't see anything useful to try to tweak.  At least some progress has
> been made - my card is recognized by Linux.  Now if only I can get the
> scanner to work with it...
> 
> If anyone has used this type of card with a scanner or has any clue what
> might be going wrong, I'd appreciate the help.
> 
> 
> Thanks,
> Rob
> 
> 
> Russell Price writes:
>  > In article <340BC1E1.C7A8CBE2@sun.com> you wrote:
>  > 
>  > : Can you tell me what changes you made to the kernel to get your card to work?
>  > 
>  > :  Probing PCI hardware.
>  > :  Warning : Unknown PCI device (9004:6178).
>  > : 
>  > : Looking at the driver files aic7xxx.c and pci.h, I noticed:
>  > : 
>  > : static const char *board_names[] = {
>  > : ...
>  > :  "AHA-2940 Ultra",  /* AIC_7881 */
>  > : ...
>  > : }
>  > : 
>  > : and
>  > : 
>  > : #define PCI_DEVICE_ID_ADAPTEC_7881      0x8178
>  > : 
>  > : which would lead me to believe the correct device address for the card is 8178, not 6178.
>  > : 
>  > 
>  > Try the 2.0.30 kernel.  The pci.h in 2.0.30 has the following:
>  > 
>  > 	#define PCI_DEVICE_ID_ADAPTEC_7861	0x6178
>  > 
>  > My 2940AU works just fine with it. Here's what my /proc/pci says:
>  > 
>  >   Bus  0, device   9, function  0:
>  >     SCSI storage controller: Adaptec AIC-7861 (rev 1).
>  >       Medium devsel.  Fast back-to-back capable.  IRQ 12.  Master Capable.  Latency=32.  Min Gnt=4.Max Lat=4.
>  >       I/O at 0x6100.
>  >       Non-prefetchable 32 bit memory at 0xe4000000.
>  > 
>  > 
>  > If you can't/won't go to 2.0.30, you can get away with changing the
>  > 0x8178 to 0x6178 -- I did that when I was running 2.0.18.
>  > 
>  > 
>  > -- 
>  > Russ Price                         "Doesn't 'Microsoft' mean 'small and limp?'"
>  > rjXprice@ais.net
>  > 
>  > (DROP THE X TO REACH ME...)
>  > 
> 


home help back first fref pref prev next nref lref last post