From owner-freebsd-current Wed Apr 26 06:50:05 1995 Return-Path: current-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id GAA24751 for current-outgoing; Wed, 26 Apr 1995 06:50:05 -0700 Received: from hda.com (hda.com [199.232.40.182]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id GAA24745 for ; Wed, 26 Apr 1995 06:50:03 -0700 Received: (dufault@localhost) by hda.com (8.6.9/8.3) id JAA27705; Wed, 26 Apr 1995 09:47:55 -0400 From: Peter Dufault Message-Id: <199504261347.JAA27705@hda.com> Subject: Re: panic: pmap_zero_page: CMAP busy To: bde@zeta.org.au (Bruce Evans) Date: Wed, 26 Apr 1995 09:47:54 -0400 (EDT) Cc: current@freefall.cdrom.com In-Reply-To: <199504260757.RAA10941@godzilla.zeta.org.au> from "Bruce Evans" at Apr 26, 95 05:57:58 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1433 Sender: current-owner@FreeBSD.org Precedence: bulk Bruce Evans writes: > > >The panic is caused by the difference between these two lines in > >the config: > > >>From GENERIC (works): > >> config kernel root on wd0 swap on wd0 and wd1 and wd2 and vn0 dumps on wd0 > > >>From DESMOND (panics): > >> config kernel root on wd0 swap on wd0 dumps on wd0 > > Maybe this has something to do with the bogus test for the end of > swdevt[] in i386/autoconf.c. The array is terminated by device NODEV > but device 0 is tested for (the test for > 0 degenerates to == 0 > because dev_t is unsigned). Apparently there are usually 0's sufficiently > close to the end of swdevt[] for there to be no problems. This is not reproducible. The scarey thing is it persisted through two full kernel builds, including the directory removal. The building was done on my production system so it isn't anything quirky about the new system. It went away when I reverted to GENERIC and stayed away as I slowly moved it back to the original reduced config. I stupidly didn't save the original config, but I'm 90% sure I now have the exact same reduced configuration I originally did and I'm not getting any problems. Now if I can figure out why it can't find the BOCA BB1008 multiport serial card at 0x100 - 0x138 IRQ12 ... Peter -- Peter Dufault Real Time Machine Control and Simulation HD Associates, Inc. Voice: 508 433 6936 dufault@hda.com Fax: 508 433 5267