From owner-freebsd-current Mon Dec 8 14:54:35 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id OAA23899 for current-outgoing; Mon, 8 Dec 1997 14:54:35 -0800 (PST) (envelope-from owner-freebsd-current) Received: from sci.brooklyn.cuny.edu (dayton@louis.sci.brooklyn.cuny.edu [146.245.1.7]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id OAA23877 for ; Mon, 8 Dec 1997 14:54:23 -0800 (PST) (envelope-from dayton@sci.brooklyn.cuny.edu) Received: (from dayton@localhost) by sci.brooklyn.cuny.edu (8.8.7/8.8.4) id RAA26506; Mon, 8 Dec 1997 17:53:14 -0500 (EST) Date: Mon, 8 Dec 1997 17:53:14 -0500 (EST) Message-Id: <199712082253.RAA26506@sci.brooklyn.cuny.edu> From: Dayton Clark To: current@freebsd.org Subject: -current, goliath, smp -- problems Sender: owner-freebsd-current@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Folks, I tired of people saying how good stable -current is ;-). I've had a system w/4 PPro 200 on a goliath motherboard for about a year now. It's had some hardware problems, but I believe those are fixed now. It runs multiprocessor Solaris and NT without a hitch but what I really want to run is FreeBSD SMP (this message is appropriate for current also). Alas, the fates are against me. Lack of time on my part is the main reason I haven't made more progress, but I've had enough time recently to get *some* evidence about what's going wrong. I will pick away at the problem, as time permits, and I hope you all have some insights on how I might proceed, or (hope, hope) perhaps someone knows just what the problem is. The system: Motherboard AMI Goliath EISA & PCI busses Chipset 450GX Processor 4X PPRO 200 Cache 256K Memory 64M Disk Adaptec 2940w 4G, 8x cdrom Video ATI Mach 32 OS FreeBSD-current SNAP-971203 (also some Sept snap). The problems: There are three problems I'm having the first two are sporadic and typcially disappear on when repeated. They also occur only with the SMP kernel (the generic smp kernel except NCPU=4), I believe. I suspect that the first two are related. o Commands (e.g., ls, mv) die with error code = 0. A message is printed on stderr, perhaps by the shell, stating that the process died with error code 0. Repeat it and it usually works. o When I try a large make (make world or making emacs) programs will disappear, i.e. make all of a sudden can't find cc or rm (after many successful uses of the command). Repeat the make and it will continue. o Emacs doesn't work at all and I can't do much without emacs 8-). The emacs from the packages distribution hangs. I brought over emacs 20.2 it dereferences a null pointer in malloc_internal(). I compiled 19.34 in /usr/ports and it gets a segment violation in malloc_internal() but at a different location. This happens with both the UP and SMP kernels. Any help would be appreciated. thanks dayton Dayton Clark CIS Department dayton@brooklyn.cuny.edu Brooklyn College/CUNY 1-718-951-4811 Brooklyn, New York 11210 1-718-951-4842 (fax)