NetBSD Problem Report #14529

Received: (qmail 10360 invoked from network); 10 Nov 2001 22:23:51 -0000
Message-Id: <20011110222350.4CC7A7E27@beowulf.gw.com>
Date: Sat, 10 Nov 2001 17:23:50 -0500 (EST)
From: christos@zoulas.com
Reply-To: christos@zoulas.com
To: gnats-bugs@gnats.netbsd.org
Subject: aic7880 gets stuck probing devices
X-Send-Pr-Version: 3.95

>Number:         14529
>Category:       kern
>Synopsis:       aic7880 gets stuck probing devices
>Confidential:   no
>Severity:       serious
>Priority:       high
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sat Nov 10 22:24:00 +0000 2001
>Closed-Date:    
>Last-Modified:  Mon Nov 12 09:04:01 +0000 2001
>Originator:     Christos Zoulas
>Release:        <NetBSD-current source date>
>Organization:
	None, but will keep trying.
>Environment:

System: NetBSD beowulf.gw.com 1.5X NetBSD 1.5X (GW-GENERIC) #36: Sat Jul 28 11:41:19 EDT 2001 kim@nix.suominen.org:/net/nix/src-3/NetBSD/cvsroot/src/sys/arch/i386/compile/GW-GENERIC i386
Architecture: i386
Machine: i386
>Description:

....
ahc0 at pci0 dev 11 function 0
ahc0: interrupting at irq 11
ahc0: aic7880 Wide Channel A, SCSI Id=7, 16/255 SCBs
scsibus0 at ahc0: 16 targets, 8 luns per target
...
scsibus0: waiting 2 seconds for devices to settle...
probe(ahc0:0:0:0): SCB 1d - timed out while idle, SEQADDR == 0x18c
SCSIRATE == 0x0
probe(ahc0:0:0:0): Queuing a BDR SCB
probe(ahc0:0:0:0): SCB 1d - timed out while idle, SEQADDR == 0x18c
SCSIRATE == 0x0
probe(ahc0:0:0:0): no longer in timeout, status = 0
ahc0: Issued Channel A Bus Reset. 1 SCBs aborted
probe(ahc0:0:1:0): SCB 1d - timed out while idle, SEQADDR == 0x18c
SCSIRATE == 0x0
probe(ahc0:0:1:0): Queuing a BDR SCB
probe(ahc0:0:1:0): SCB 1d - timed out while idle, SEQADDR == 0x18c
SCSIRATE == 0x0
probe(ahc0:0:1:0): no longer in timeout, status = 0
ahc0: Issued Channel A Bus Reset. 1 SCBs aborted
syncing disks... done
rebooting...

>How-To-Repeat:
	The only thing on the bus is a cdrw drive which works under linux
	in approximately the same hardware configuration.
>Fix:
	N/A
>Release-Note:
>Audit-Trail:

From: Gergely EGERVARY <mauzi@expertlan.hu>
To: <christos@zoulas.com>
Cc: <gnats-bugs@gnats.netbsd.org>
Subject: Re: kern/14529: aic7880 gets stuck probing devices
Date: Sun, 11 Nov 2001 19:40:22 +0100 (CET)

 have you turned off 16bit/wide initialisation in the scsiselect utility?

 also, please check my pr kern/14527 - if you could reproduce my problem...

 thanks,

 -- mauzi


From: Dave Tyson <Dave.Tyson@liverpool.ac.uk>
To: <christos@zoulas.com>
Cc: <gnats-bugs@gnats.netbsd.org>
Subject: Re: kern/14529: aic7880 gets stuck probing devices
Date: Mon, 12 Nov 2001 09:02:59 +0000 (GMT)

 On Sat, 10 Nov 2001 christos@zoulas.com wrote:

 >
 >>Number:         14529
 >>Category:       kern
 >>Synopsis:       aic7880 gets stuck probing devices
 >>Confidential:   no
 >>Severity:       serious
 >>Priority:       high
 >>Responsible:    kern-bug-people
 >>State:          open
 >>Class:          sw-bug
 >>Submitter-Id:   net
 >>Arrival-Date:   Sat Nov 10 14:24:00 PST 2001
 >>Closed-Date:
 >>Last-Modified:
 >>Originator:     Christos Zoulas
 >>Release:        <NetBSD-current source date>
 >>Organization:
 >	None, but will keep trying.
 >>Environment:
 >
 >System: NetBSD beowulf.gw.com 1.5X NetBSD 1.5X (GW-GENERIC) #36: Sat Jul 28 11:41:19 EDT 2001 kim@nix.suominen.org:/net/nix/src-3/NetBSD/cvsroot/src/sys/arch/i386/compile/GW-GENERIC i386
 >Architecture: i386
 >Machine: i386
 >>Description:
 >
 >....
 >ahc0 at pci0 dev 11 function 0
 >ahc0: interrupting at irq 11
 >ahc0: aic7880 Wide Channel A, SCSI Id=7, 16/255 SCBs
 >scsibus0 at ahc0: 16 targets, 8 luns per target
 >...
 >scsibus0: waiting 2 seconds for devices to settle...
 >probe(ahc0:0:0:0): SCB 1d - timed out while idle, SEQADDR == 0x18c
 >SCSIRATE == 0x0
 >probe(ahc0:0:0:0): Queuing a BDR SCB
 >probe(ahc0:0:0:0): SCB 1d - timed out while idle, SEQADDR == 0x18c
 >SCSIRATE == 0x0

 ...snip snip snip...

 I reported this same problem in January this year, see kern/11891.
 Unfortunately the two system's were in 24/7 service and so it was
 difficult to do any debugging.

 Having recently gotten a spare 2944 (so I could leave the production
 machines running) I carried out a few tests. Initially I thought the
 problem may lie with a duff flash revisions on the 2944, however it this
 proved to be a false lead as after flashing all the 2944's to bios 2.20.0,
 they worked perfectly under 1.5.2 on a crappy pentium 1 system, but still
 failed on the production Pentium II system. Swapping the Tyan Tiger LX
 motherboard for a BX version, but keeping everything else the same
 cured the problem.

 My guess is the the SCSI code in 1.5.x is driving the controllers harder
 than the 1.3.x and 1.4.x releases and this triggered a problem with the
 motherboard. However if the problem does not show under linux then the
 NetBSD code should be fixed. I can probably get one of the LX boards up
 and running with a celeron 400 to do some testing if needed.

 Dave

 --
 =====================================================================
 Computing Services Dept         Phone/Fax: 0151-794-3731/3759
 The University of Liverpool     Email: dtyson@liv.ac.uk
 Chadwick Tower, Peach Street    WWW:   http://www.liv.ac.uk/~dtyson
 Liverpool L69 7ZF
 =====================================================================

>Unformatted:
 	Sat Nov 10 17:21:36 EST 2001

Home	PR Database Search