From nobody@FreeBSD.ORG Wed May 12 23:53:40 1999
Return-Path: <nobody@FreeBSD.ORG>
Received: by hub.freebsd.org (Postfix, from userid 32767)
	id B884B153BE; Wed, 12 May 1999 23:53:40 -0700 (PDT)
Message-Id: <19990513065340.B884B153BE@hub.freebsd.org>
Date: Wed, 12 May 1999 23:53:40 -0700 (PDT)
From: dnsadmin@aeonflux.ent
Sender: nobody@FreeBSD.ORG
To: freebsd-gnats-submit@freebsd.org
Subject: Adaptec 2940 UW SCSI Controller BIOS 1.34.3 wont work under 3.1R
X-Send-Pr-Version: www-1.0

>Number:         11681
>Category:       i386
>Synopsis:       Adaptec 2940 UW SCSI Controller BIOS 1.34.3 wont work under 3.1R
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    gibbs
>State:          closed
>Quarter:        
>Keywords:       
>Date-Required:  
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Thu May 13 00:00:01 PDT 1999
>Closed-Date:    Sat Feb 24 11:42:28 PST 2001
>Last-Modified:  Sat Feb 24 11:43:38 PST 2001
>Originator:     James F. Porter
>Release:        3.1R
>Organization:
Aeon Flux Enterprises, Inc.
>Environment:
FreeBSD www2.aeonflux.net 3.1-RELEASE FreeBSD 3.1-RELEASE #3: Fri Apr 30 16:38:58 EDT 1999     root@www2.aeonflux.net:/usr/src/sys/c
ompile/WWW2  i386
>Description:
We run a variety of PC's with the AHA 2940UW controller. ALL systems
work perfectly as long as we don't use a BIOS revision higher than
1.26.  Bios revisions above that including the current 1.34.3 latest
AHA 2940UW shipping bios keep generating SCSI BUS Resets. And the
system never loads even if you are NOT booting from the SCSI device.
The O/S cycles thru ah0 0:0:0:0 thru 0:0:15:0 and then starts over.
It doesn't even matter whether a drive is connected or not.  This is
a major problem for us (we're an small ISP) as we can no longer obtain
cards with the 1.26 or earlier bios and Adaptec as specifically said
that the 1.34.3 should be used because of all the fixes in it.
>How-To-Repeat:
simply put ANY Adaptec 2940UW PCI SCSI controller in a PC
with a bios revision higher than 1.26 and the problem automatically
repeats.
>Fix:
none.  we need one desperately as this is holding up our entire
upgrade program.

>Release-Note:
>Audit-Trail:

From: "Justin T. Gibbs" <gibbs@plutotech.com>
To: dnsadmin@aeonflux.ent
Cc: freebsd-gnats-submit@FreeBSD.ORG
Subject: Re: i386/11681: Adaptec 2940 UW SCSI Controller BIOS 1.34.3 wont work under 3.1R 
Date: Thu, 13 May 1999 07:13:34 -0600

 >We run a variety of PC's with the AHA 2940UW controller. ALL systems
 >work perfectly as long as we don't use a BIOS revision higher than
 >1.26.
 
 You'll have to be a bit more specific as I have several 2940UW controllers
 with BIOS revisions in the 30's and they all work fine.  My guess is that
 you are encountering an interration between a specific rev of the aic7880
 and the chipset on your motherboard.  A dmesg from a boot -v (or possible a
 boot -v sent to a serial console so you can capture it all) will give me
 a clue about your problem, but I may only be able to help you out if you
 can send me a bare bones system (MB, controller, case, powersupply) for a
 day or two so I can analyze the problem with a PCI bus analyzer.
 
 --
 Justin
 

From: DNS Administrator <dnsadmin@aeonflux.net>
To: freebsd-gnats-submit@freebsd.org
Cc:  
Subject: Re: i386/11681: Adaptec 2940 UW SCSI Controller BIOS 1.34.3
  wont work under 3.1R
Date: Thu, 13 May 1999 13:53:45 -0400

 This is right off the Adaptec main chip: AHA-2940UW,  945300-01 E, 9746
 This is off the BIOS prom.  589217, r E, 9100, @ 1996, v1.32
 
 The card works just fine in any Novel or Windows system.  We upgraded the
 bios to 1.34.3 in an attempt to make it work.  The system NEVER reaches a
 boot prompt it simply hangs retesting the SCSI bus with/without a drive
 attached.  
 If you can give me a method to get a dmesg without reaching a prompt.
 
 
 >From: "Justin T. Gibbs" <gibbs@plutotech.com>
 >    To: dnsadmin@aeonflux.ent
 >    Cc: freebsd-gnats-submit@FreeBSD.ORG
 >    Subject: Re: i386/11681: Adaptec 2940 UW SCSI Controller BIOS 1.34.3
 wont work under 3.1R 
 >    Date: Thu, 13 May 1999 07:13:34 -0600
 >
 >     >We run a variety of PC's with the AHA 2940UW controller. ALL systems
 >     >work perfectly as long as we don't use a BIOS revision higher than
 >     >1.26.
 >     
 >     You'll have to be a bit more specific as I have several 2940UW
 controllers
 >     with BIOS revisions in the 30's and they all work fine.  My guess is
 that
 >     you are encountering an interration between a specific rev of the
 aic7880
 >     and the chipset on your motherboard.  A dmesg from a boot -v (or
 possible a
 >     boot -v sent to a serial console so you can capture it all) will give me
 >     a clue about your problem, but I may only be able to help you out if you
 >     can send me a bare bones system (MB, controller, case, powersupply)
 for a
 >     day or two so I can analyze the problem with a PCI bus analyzer.
 >     
 >     --
 >     Justin
 

From: Cy Schubert <cschuber@uumail.gov.bc.ca>
To: freebsd-gnats-submit@freebsd.org
Cc: dnsadmin@aeonflux.net, "Justin T. Gibbs" <gibbs@plutotech.com>
Subject: Re: i386/11681: Adaptec 2940 UW SCSI Controller BIOS 
 1.34.3
Date: Thu, 13 May 1999 14:59:56 -0700

 I had the same problem, as follows on 3.1R:
 
 1.  Upgraded my system BIOS to fix a minor Y2K bug.
 
 2.  Installed a 2940UW (replacing a 1542CF) with BIOS 1.34.x.
 
 3.  System would not boot with the timeouts ah0 0:0:0:0 thru 
 0:0:15:0
     as described in the PR.
 
 4.  Backed out of the system BIOS upgrade performed in step 1, 
 restoring
     the original system BIOS that came with the moterboard.  This 
 fixed
     the problem.
 
 
 Regards,                       Phone:  (250)387-8437
 Cy Schubert                      Fax:  (250)387-5766
 Open Systems Group          Internet:  Cy.Schubert@uumail.gov.bc.ca
 ITSD                                   Cy.Schubert@gems8.gov.bc.ca
 Province of BC            
                       "e**(i*pi)+1=0"
 
 
 
 

From: "Justin T. Gibbs" <gibbs@plutotech.com>
To: Cy Schubert - ITSD Open Systems Group <Cy.Schubert@uumail.gov.bc.ca>
Cc: freebsd-gnats-submit@freebsd.org, dnsadmin@aeonflux.net,
	"Justin T. Gibbs" <gibbs@plutotech.com>
Subject: Re: i386/11681: Adaptec 2940 UW SCSI Controller BIOS 1.34.3 
Date: Thu, 13 May 1999 16:16:34 -0600

 >4.  Backed out of the system BIOS upgrade performed in step 1, 
 >restoring
 >    the original system BIOS that came with the moterboard.  This 
 >fixed
 >    the problem.
 
 Hmm.  Then this may be as simple as setting the bus master enable bit
 in the driver.  I already do this in -current and will be bringing
 that change over shortly.
 
 --
 Justin
 
 
 

From: Cy Schubert <cschuber@uumail.gov.bc.ca>
To: "Justin T. Gibbs" <gibbs@plutotech.com>
Cc: Cy Schubert - ITSD Open Systems Group <Cy.Schubert@uumail.gov.bc.ca>,
	freebsd-gnats-submit@freebsd.org, dnsadmin@aeonflux.net
Subject: Re: i386/11681: Adaptec 2940 UW SCSI Controller BIOS 
 1.34.3
Date: Thu, 13 May 1999 15:54:19 -0700

 I have a QDI Explorer I.  QDI has a Y2K patch for the Explorer II.  
 Their tech told me (also verified by their Web Site) that there is 
 no Y2K BIOS for the Explorer I but assured me that Explorer II Y2K 
 BIOS will work with Explorer I motherboards.  Two weeks later I 
 purchased a 2940UW and FreeBSD wouldn't boot.  I restored the old 
 BIOS and FreeBSD was able to boot.  I suspect that there may be 
 some timing problems on the bus or as you suggest a bus mastering 
 problem.
 
 
 Regards,                       Phone:  (250)387-8437
 Cy Schubert                      Fax:  (250)387-5766
 Open Systems Group          Internet:  Cy.Schubert@uumail.gov.bc.ca
 ITSD                                   Cy.Schubert@gems8.gov.bc.ca
 Province of BC            
                       "e**(i*pi)+1=0"
 
 In message <199905132226.QAA25150@pluto.plutotech.com>, "Justin T. 
 Gibbs" write
 s:
 > >4.  Backed out of the system BIOS upgrade performed in step 1, 
 > >restoring
 > >    the original system BIOS that came with the moterboard.  This 
 > >fixed
 > >    the problem.
 > 
 > Hmm.  Then this may be as simple as setting the bus master enable bit
 > in the driver.  I already do this in -current and will be bringing
 > that change over shortly.
 > 
 > --
 > Justin
 > 
 > 
 
 
 

From: DNS Administrator <dnsadmin@aeonflux.net>
To: Cy Schubert - ITSD Open Systems Group <Cy.Schubert@uumail.gov.bc.ca>,
	freebsd-gnats-submit@freebsd.org
Cc: "Justin T. Gibbs" <gibbs@plutotech.com>
Subject: Re: i386/11681: Adaptec 2940 UW SCSI Controller BIOS  1.34.3
Date: Fri, 14 May 1999 06:20:51 -0400

 >3.  System would not boot with the timeouts ah0 0:0:0:0 thru 
 >0:0:15:0
 >    as described in the PR.
 
 Well I'm humbled.  Or irritated as it may be.  The 2940UW bios 1.34.3 works
 now on my current configuration.  I guess we'll have to stop bying the <
 $150 motherboards and stick with only ASUS boards.  The only modification I
 had to make was to turn off in the bios > 1 gig support for DOS.
 Everything works fine now.  It is puzzling though why the 1.25 bios cards
 work on the cheaper boards.  Not once have we ever had problems with them
 
 Thanks to all who contributed.
 
 James.
 

From: "Justin T. Gibbs" <gibbs@plutotech.com>
To: DNS Administrator <dnsadmin@aeonflux.net>
Cc: Cy Schubert - ITSD Open Systems Group <Cy.Schubert@uumail.gov.bc.ca>,
	freebsd-gnats-submit@freebsd.org,
	"Justin T. Gibbs" <gibbs@plutotech.com>
Subject: Re: i386/11681: Adaptec 2940 UW SCSI Controller BIOS 1.34.3 
Date: Fri, 14 May 1999 08:24:55 -0600

 >>3.  System would not boot with the timeouts ah0 0:0:0:0 thru 
 >>0:0:15:0
 >>    as described in the PR.
 >
 >Well I'm humbled.  Or irritated as it may be.  The 2940UW bios 1.34.3 works
 >now on my current configuration.  I guess we'll have to stop bying the <
 >$150 motherboards and stick with only ASUS boards.  The only modification I
 >had to make was to turn off in the bios > 1 gig support for DOS.
 >Everything works fine now.  It is puzzling though why the 1.25 bios cards
 >work on the cheaper boards.  Not once have we ever had problems with them
 
 It's not necessarily puzzling at all.  This is probably an interraction
 between the revision of the aic7880 on the newer SCSI controllers and
 the particular PCI chipset used on the cheaper motherboards.  As I
 mentioned before, if you send me a system for a day or two, I can
 probably make it work.  This would benefit anyone using motherboards
 with that chipset...
 
 --
 Justin
 

From: DNS Administrator <dnsadmin@aeonflux.net>
To: "Justin T. Gibbs" <gibbs@plutotech.com>
Cc: Cy Schubert - ITSD Open Systems Group <Cy.Schubert@uumail.gov.bc.ca>,
	freebsd-gnats-submit@freebsd.org,
	"Justin T. Gibbs" <gibbs@plutotech.com>
Subject: Re: i386/11681: Adaptec 2940 UW SCSI Controller BIOS 1.34.3 
Date: Fri, 14 May 1999 12:12:29 -0400

 >It's not necessarily puzzling at all.  This is probably an interraction
 >between the revision of the aic7880 on the newer SCSI controllers and
 >the particular PCI chipset used on the cheaper motherboards.  As I
 >mentioned before, if you send me a system for a day or two, I can
 >probably make it work.  This would benefit anyone using motherboards
 >with that chipset...
 
 I may be able to do that in a week or two.  The system we had to get going
 under 3.1 was a critical webserver. I was up all night getting back on the
 air.  Give me a address.
 
Responsible-Changed-From-To: freebsd-bugs->gibbs 
Responsible-Changed-By: steve 
Responsible-Changed-When: Sun May 16 17:30:01 PDT 1999 
Responsible-Changed-Why:  
Looks like Mr. Gibbs is taking care of this one. 
State-Changed-From-To: open->closed 
State-Changed-By: gibbs 
State-Changed-When: Sat Feb 24 11:42:28 PST 2001 
State-Changed-Why:  
The bug was likely caused by the PCI 2.1 retry bug of the 7880 Rev B. 
A workaround for this bug, that has a negligible impact on performance, 
has been in the driver for some time. 

http://www.freebsd.org/cgi/query-pr.cgi?pr=11681 
>Unformatted:
