From owner-freebsd-bugs Wed Mar 31 22:30:21 1999 Delivered-To: freebsd-bugs@freebsd.org Received: from freefall.freebsd.org (freefall.FreeBSD.ORG [204.216.27.21]) by hub.freebsd.org (Postfix) with ESMTP id 28F65155BC for ; Wed, 31 Mar 1999 22:30:19 -0800 (PST) (envelope-from gnats@FreeBSD.org) Received: (from gnats@localhost) by freefall.freebsd.org (8.9.2/8.9.2) id WAA37710; Wed, 31 Mar 1999 22:30:02 -0800 (PST) (envelope-from gnats@FreeBSD.org) Date: Wed, 31 Mar 1999 22:30:02 -0800 (PST) Message-Id: <199904010630.WAA37710@freefall.freebsd.org> To: freebsd-bugs@FreeBSD.org Cc: From: "Justin T. Gibbs" Subject: Re: kern/10866: ahc2740 panic Reply-To: "Justin T. Gibbs" Sender: owner-freebsd-bugs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org The following reply was made to PR kern/10866; it has been noted by GNATS. From: "Justin T. Gibbs" To: tjohnson@wallace.dlib.vt.edu Cc: FreeBSD-gnats-submit@FreeBSD.ORG Subject: Re: kern/10866: ahc2740 panic Date: Wed, 31 Mar 1999 23:16:13 -0700 >>Description: > >The system panics a short time after boot. > >The debugger reports: (retyped) >panic ahc0 brkadrint, Illegal Sequencer Address referenced at seqaddr 0x1e >mp_lock=01000001 cpuid=1 lapic.id=01000000 >stopped at Debugger+0x37: movl $0,in_Debugger > >the controler is: >Mar 27 11:25:11 teryx /kernel: Probing for devices on the EISA bus >Mar 27 11:25:11 teryx /kernel: ahc0: at 0x1c0 >0->0x1cff irq 14 on eisa0 slot 1 >Mar 27 11:25:11 teryx /kernel: ahc0: aic7770 <= Rev C, Twin Chann >Mar 27 11:25:11 teryx /kernel: el, A SCSI Id=7, B SCSI Id=7, primary A, 4/255 >SCBs I have the same controller and cannot reproduce the problem. The error code that you are encountering indicates that the sequencer jumped to an illegal address, but the instruction that does this is executed for every single initiated command, so if it were a code bug, it would show up as soon as we tried talking to a device. My guess is that the sequencer instruction ram on your controller is experiencing intermittent corruption that causes this jump instruction to 'mutate' after the program has been loaded. The aic7770, unlike its later PCI brethren, does not have parity protected instruction ram, so we won't get a more direct indication of this kind of failure. Perhaps you have a cooling problem or the controller is in the process of failing. -- Justin To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-bugs" in the body of the message