From owner-freebsd-hackers Thu Aug 8 18:20:50 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id SAA29201 for hackers-outgoing; Thu, 8 Aug 1996 18:20:50 -0700 (PDT) Received: from irz301.inf.tu-dresden.de (irz301.inf.tu-dresden.de [141.76.1.11]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id SAA29196; Thu, 8 Aug 1996 18:20:47 -0700 (PDT) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by irz301.inf.tu-dresden.de (8.6.12/8.6.12-s1) with ESMTP id DAA16678; Fri, 9 Aug 1996 03:20:44 +0200 Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id DAA13076; Fri, 9 Aug 1996 03:20:44 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.7.5/8.6.9) id XAA00307; Thu, 8 Aug 1996 23:13:46 +0200 (MET DST) From: J Wunsch Message-Id: <199608082113.XAA00307@uriah.heep.sax.de> Subject: Re: "Panick" - help needed... To: freebsd-hackers@freebsd.org (FreeBSD hackers) Date: Thu, 8 Aug 1996 23:13:46 +0200 (MET DST) Cc: archer@cmr.kiev.ua (Litvin Alexander B.), dyson@freebsd.org Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: from "Litvin Alexander B." at "Aug 8, 96 10:19:17 pm" X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E X-Mailer: ELM [version 2.4ME+ PL17 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Litvin Alexander B. wrote: > I tried to figure out how would FreeBSD behave under heavy load. > For this purpose I wrote the simple program. It creates 200 copies > of itself, and then each copy tries to give system some work, that > is opens a file and writes into it randomly and also allocates memory, > uses it and frees memory afterwards. Here it is: > kernel page directory invalid pdir=0x52e063 va=0xc000 > ^^^^^^^^ > actual numbers may differ > panic: invalid page directory Well, i could not reproduce this with a ``somewhat -current'' system. I ran your program for more than 20 minutes, causing the system to virtually halt every other activity. (I've accidentally switched to the X11 screen, and it took the system about 15 minutes to switch back... ;-) By using rtrprio i was able to keep a `top' process alive on the second screen, and it reported a system load of ~ 1100 after this 20 minutes torture. Of course, the disks became very noisy. :) The only abnomality was that there were some 400 leftover temp files around in my /tmp (where the program ran), and when finally rm'ing them, i've got a: panic: vm_page_free: invalid wire count (4), pindex: 0x6 In case somebody has an idea, here's the stack trace: (kgdb) backtrace [...] #9 0xf011703e in panic () #10 0xf01b2688 in vm_page_freechk_and_unqueue () #11 0xf01b26cd in vm_page_free () #12 0xf01b0049 in vm_object_terminate () #13 0xf01afeab in vm_object_deallocate () #14 0xf01b4b0a in pager_cache () #15 0xf01b5847 in vnode_pager_uncache () #16 0xf0130ecf in unlink () #17 0xf01c6367 in syscall () #18 0xf01babb5 in Xsyscall () #19 0x1266 in ?? () #20 0x107f in ?? () I could analyze the coredump further if somebody points me into a useful direction. The system is: i586/133, 32 MB RAM, NCR 53c810, 2 disks (with distributed swap space). Kernel from around July 20th. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-)