From owner-freebsd-bugs Thu Feb 6 12:38:56 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA03629 for bugs-outgoing; Thu, 6 Feb 1997 12:38:56 -0800 (PST) Received: from pompano.pcola.gulf.net (root@pompano.pcola.gulf.net [198.69.72.14]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA03618; Thu, 6 Feb 1997 12:38:50 -0800 (PST) Received: from pompano.pcola.gulf.net (spatula@localhost.gulf.net [127.0.0.1]) by pompano.pcola.gulf.net (8.8.5/8.7.3) with SMTP id OAA22175; Thu, 6 Feb 1997 14:38:44 -0600 (CST) Date: Thu, 6 Feb 1997 14:38:43 -0600 (CST) From: Prisoner X-Sender: spatula@pompano.pcola.gulf.net To: freebsd-questions@freebsd.org cc: freebsd-bugs@freebsd.org Subject: What's going on here? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-bugs@freebsd.org X-Loop: FreeBSD.org Precedence: bulk First off, thanks to all in #freebsd who have been helping me with this page-fault thing. I have some more information that may be useful to diagnosing the problems I have been having. In a nutshell, my system encounters a lot of page faults while in kernel mode. They can be made to stop by not using swap at all, but that isn't really such a great solution. My most recent trace from the debugger reveals this: (I've reversed the order so it's chronological) Xsyscall at _Xsyscall+0x35 _syscall(27,27,5,efbf75d,efbfdb54) at _syscall+0x183 _execve(f137c800,efbfff94,efbfff84,8097060,efbfd75d) at _execve+0x1c7 _exec_aout_imgoct(efbffe98,f01bf6c18,f137c800,0,0) at _exec_aout_imgoct+0x15f _exec_new_vmspace(efbffe98,0,0,efbffe98,3e000) at _exec_new_vmspace+0x3d _pmap_remove_pages(f1425164,0,efbfe000,0,efbde000) at _pmap_remove_pages+0x7b Then the kernel catches a signal 12 (page not present) and chokes. Some interesting related information: I have observed that preceding a crash and burn, some programs begin to behave strangely. For example, while trying to make a crash happen, I was compiling a kernel with pgcc, which refused to run because of a signal 10. After crashing and rebooting, the compile went fine. Yesterday, something similar happened with BitchX, only on signal 4 (illegal instruction of all things). Netscape will frequently refuse to run because of signal 11. All of these program will run fine after rebooting, however. Other information that may be of use: I know that there are bad sectors on my swap partition, but I had the install program check and mark bad blocks. Also, this is happening with 3.0-SNAP, but also was happening with 2.1.5. This is running on a P100, intel chipset, with 32 megs of RAM and 74 megs swap. At this point, I'm trying to get my computer to crash again to see if the fault occurs in the same place consistantly. Nick -- "I don't kill, and I don't murder, or manufacture atomic weapons!" - Nerf Herder Nick Johnson, run for your lives. http://www.gulf.net/~spatula/