Date: Fri, 11 Oct 2002 09:53:04 -0700 From: "Kirby Kuehl (kkuehl)" <kkuehl@cisco.com> To: "'Kevin Oberman'" <oberman@es.net> Cc: <freebsd-stable@freebsd.org> Subject: RE: Freebsd 4.7-Stable kernel panic. Message-ID: <000701c27146$ae19efe0$dc2746ab@amer.cisco.com> In-Reply-To: <20021011161159.B80CB5D04@ptavv.es.net>
next in thread | previous in thread | raw e-mail | index | archive | help
I do follow this routine: make buildworld make buildkernel KERNCONF=MYKERNEL make installkernel KERNCONF=MYKERNEL make installworld mergemaster I just turned off the MATH_EMULATE today when I was turning on the debugging options. I also turned off apm today in an effor to eliminate things I didn't need hoping to find the source of the panic. Kirby Kuehl -----Original Message----- From: Kevin Oberman [mailto:oberman@es.net] Sent: Friday, October 11, 2002 9:12 AM To: Kirby Kuehl Cc: freebsd-stable@freebsd.org Subject: Re: Freebsd 4.7-Stable kernel panic. > Date: Fri, 11 Oct 2002 08:16:50 -0700 (PDT) > From: Kirby Kuehl <kkuehl@cisco.com> > Sender: owner-freebsd-stable@FreeBSD.ORG > > I do not believe this panic is hardware related as I am currently > composing this letter on the same system (I had a backup of my working > kernel) > Current Working Kernel: > FreeBSD kkuehl-fbsd.cisco.com 4.7-RC FreeBSD 4.7-RC #5: Fri Oct 4 > 09:51:47 PDT 2002 > root@kkuehl-fbsd.cisco.com:/usr/obj/usr/src/sys/ELITE i386 > > This IBM T30 Thinkpad has been running freebsd-stable happily for > approximately two months with the same kernel config. (I have attached > a diff between my kernel config and GENERIC) > > After cvsuping on 10/11/2002, building and installing world and > building and installing kernel and mergemastering I get an immediate > kernel panic. > > I then enabled the following in hopes of captureing a backtrace: > > Line from /etc/fstab showing swap partition: > /dev/ad0s1b none swap sw 0 0 > > Line from /etc/rc.conf enabling dumpdev: dumpdev="/dev/ad0s1b" > > Relevant lines from /usr/src/sys/i386/conf/MYKERNEL: > options DDB, DDB_UNATTENDED > makeoptions DEBUG=-g > > Unfortunately, the kernel panics/reboots (a second time) before it > copies the core from swap to /var/crash. > > Am I missing a step in obtaining a core / dump? Is there another > approach? I realize that this report is unhelpful without a backtrace. I can only speculate, but you did not mention doing "make installworld" between installing the new kernel and running mergemaster. It sounds like you installed the new world BEFORE building the kernel. This can be VERY dangerous! You now have an updated world that is out of sync with the old (working) kernel leaving you with no options except a complete re-build of the system. Sounds like you were lucky, though, and your old kernel still works. Please follow the handbook recommendation on rebuilding a system. Looking at your kernel, most things look OK. > -options MATH_EMULATE #Support for x87 emulation This is something I have not tried. I have always suspected that I could delete this, but have never done so. Possible issue. > # SCSI peripherals > -device scbus # SCSI bus (required) > -device da # Direct Access (disks) If you are really sure that you want will not plug any storage device into the USB, this is fine. > -device psm0 at atkbdc? irq 12 > +device psm0 at atkbdc? flags 0x20 irq 12 I found that I needed flags 0x2000 to keep the mouse on my T30 working. > # Power management support (see LINT for more options) > -device apm0 at nexus? disable flags 0x20 # Advanced Power Management > +#device apm0 at nexus? disable flags 0x20 # Advanced Power Management Seems odd to not use APM on a laptop, but it should not cause a crash. > -device wi You are disabling your wireless card if you have one. (You may want it that way.) Then again, noting your e-mail address, you might prefer a different wireless card. > -device umass # Disks/Mass storage - Requires scbus and da > +#device umass # Disks/Mass storage - Requires scbus and da This fits with the removed scbus and da above. Once again, you are forcing a kernel rebuild if you ever want to use one of these devices. That eliminates one of the USB's nice features. So I don't see anything that SHOULD cause a failure, but I am a bit suspicious of your elimination of option MATH_EMULATE. Guess I will have to upgrade my T30 and see what happens. Coincidentally, I last updated that system on Oct. 5, just hours after you did yours. I may be in the same hole you are in about an hour. R. Kevin Oberman, Network Engineer Energy Sciences Network (ESnet) Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab) E-mail: oberman@es.net Phone: +1 510 486-8634 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?000701c27146$ae19efe0$dc2746ab>