From owner-freebsd-current Sun Apr 28 18:58:18 2002 Delivered-To: freebsd-current@freebsd.org Received: from apollo.backplane.com (apollo.backplane.com [216.240.41.2]) by hub.freebsd.org (Postfix) with ESMTP id 433C737B41F; Sun, 28 Apr 2002 18:58:13 -0700 (PDT) Received: (from dillon@localhost) by apollo.backplane.com (8.11.6/8.9.1) id g3T1wDh60486; Sun, 28 Apr 2002 18:58:13 -0700 (PDT) (envelope-from dillon) Date: Sun, 28 Apr 2002 18:58:13 -0700 (PDT) From: Matthew Dillon Message-Id: <200204290158.g3T1wDh60486@apollo.backplane.com> To: Robert Watson Cc: current@FreeBSD.ORG Subject: Re: Page fault in swp_pager_meta_build() References: Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG :It currently has no swap started at all, which is one reason I was rather :puzzled to see this panic: : :192.168.50.1:/cboss/devel/nfsroot/crash2.cboss.tislabs.com / nfs ro 0 0 :proc /proc procfs rw 0 0 :/dev/ad0s1e /mnt ufs rw 0 0 :... :Should it even be hitting this code if swap hasn't been enabled? I've run :into a couple of other weird bugs and wouldn't be surprised if there is a :memory allocation problem. The problem I was actually trying to reproduce :with these two crash boxes was one where the socket used by NFS get :zero'd, resulting in a null pointer dereference. The other one is in odd :panic in the mutex code during an early VFS operation. : :Robert N M Watson FreeBSD Core Team, TrustedBSD Project The case can be hit but it shouldn't have found any swap to free. Huh. Maybe there is a degenerate case in the swap code that blows up if swap is compiled in but no swap has been added. If the problem goes away when you add a tiny amount of swap that would confirm it. What is the 'swhash_mask' global contain? -Matt Matthew Dillon To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message