Return to BSD News archive
Xref: sserve comp.os.386bsd.questions:2508 comp.os.386bsd.bugs:758 Path: sserve!newshost.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!saimiri.primate.wisc.edu!zaphod.mps.ohio-state.edu!howland.reston.ans.net!usenet.ins.cwru.edu!lerc.nasa.gov!eagle!mikef From: mikef@sarah.lerc.nasa.gov (Mike J. Fuller) Newsgroups: comp.os.386bsd.questions,comp.os.386bsd.bugs Subject: Re: [NetBSD] 3Com ethernet problems, misc bugs, and debugging kernels Date: 17 May 1993 19:51:23 GMT Organization: Computational Materials Laboratory, NASA Lewis Research Center Lines: 69 Distribution: world Message-ID: <1t8qbr$k4a@eagle.lerc.nasa.gov> References: <1t69ec$bau@eagle.lerc.nasa.gov> Reply-To: mikef@sarah.lerc.nasa.gov NNTP-Posting-Host: sarah.lerc.nasa.gov In-reply-to: mikef@sarah.lerc.nasa.gov's message of 16 May 1993 20:50:20 GMT X-Newsreader: GNU Emacs 18.58.2, GNUS 3.14.1, NNTP 3.10 >>>>> On 16 May 1993 20:50:20 GMT, I said: Me> Both of those times, it failed to do a crash dump and reboot. Does anybody Me> know what the addresses and codes correspond to? Well, thanks to the FAQ (which, of course, showed up on our news server a few hours after my posting :-), trap type 12 = stack fault. Still doesn't tell me what the addresses and codes are, though... I did get one response about the problem itself: >>>>> On Mon, 17 May 93 10:01:18 -0700, Mark Ganter >>>>> <ganter@u.washington.edu> said: Mark> Mike, I had the same problem with the Trap 12 crashes with the kernel Mark> before May 6th. I was in the process of installing NetBSD and would Mark> crash during ftping. I grabbed the "new" kc and install disk and Mark> things are running much better. I did hear about the new install floppies, but I don't see how they could help; if the problem is in the kernel (which it appears to be), how will doing a new install change anything? After all, the kernel is built from the ksrc distribution, which hasn't been updated. I suppose it could help if the boot program has changed and the problem had to do with how things get initialized. A message from the boot program which I had noticed before, but hadn't really thought much about, reads something like this: 386bsd BOOT @0x90000 639/7168K [20/9/92] ... Warning: Base memory 639K, assuming 640K. Could this be the source of all my problems? Me> Lastly, does anybody know how to build a debugging kernel? I got one response on that: >>>>> On Mon, 17 May 93 16:19:42 +0200, chmr@edvz.tu-graz.ac.at (Christoph >>>>> Robitschko) said: Christoph> You probably want to make a kernel with the kernel debugger; For Christoph> this, you just have to include "pseudo-device ddb" and "options Christoph> DDB" in your config file. If you do this, you should also Christoph> (previously) get DDB.tar.z from ftp.cs.few.eur.nl in /pub, these Christoph> are patches for the kernel debugger to consume less space in the Christoph> kernel and to print correct stack traces. Christoph> If you really want to compile the kernel with full debugging code Christoph> -- I've been told it's over 4 MB in size... You probably got the Christoph> "out of memory" message because the compiler was using more Christoph> memory than the limit was set; Use "limit stacksize unlimited" Christoph> and "limit datasize unlimited" (in csh). I haven't tried making a kernel with the debugger yet, but changing the limits does indeed take care of the "out of memory" error when compiling with "-g". I should have known! As Homer would say, ``D'oh!''. Many thanks to Mark and Christoph for taking the time to respond, and if anyone has any more suggestions about my "panic problem", let me know and I will continue to summarize as events unfold. Mike Fuller /-----------------------------------------------------------------------------\ | Mike J. Fuller | Internet: mikef@sarah.lerc.nasa.gov | "I hate | |----------------| mikef@zippysun.math.uakron.edu | quotations." | |/\/\/\/\/\/\/\/\| Bitnet: r2mjf@akronvm | -- R.W. Emerson | \-----------------------------------------------------------------------------/