From owner-freebsd-current Sun Mar 25 10:19:17 2001 Delivered-To: freebsd-current@freebsd.org Received: from earth.backplane.com (earth-nat-cw.backplane.com [208.161.114.67]) by hub.freebsd.org (Postfix) with ESMTP id 562CD37B718; Sun, 25 Mar 2001 10:19:13 -0800 (PST) (envelope-from dillon@earth.backplane.com) Received: (from dillon@localhost) by earth.backplane.com (8.11.2/8.9.3) id f2PIIlp11246; Sun, 25 Mar 2001 10:18:47 -0800 (PST) (envelope-from dillon) Date: Sun, 25 Mar 2001 10:18:47 -0800 (PST) From: Matt Dillon Message-Id: <200103251818.f2PIIlp11246@earth.backplane.com> To: Kirk McKusick Cc: Jeroen Ruigrok/Asmodai , Peter Wemm , Paul Saab , Soeren Schmidt , current@FreeBSD.ORG Subject: Re: vm page panic References: <200103251514.HAA04116@beastie.mckusick.com> Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG : Hi guys, : : ok, sources cvsupped yesterday afternoon, just before my ffs_alloc.c : commit [which I did, obviously, add myself locally]. : : Box had been running for a while when all of a sudden it got into a : panic: : : vm_page_alloc: free/cache page 0xc0776fa4 was dirty : : a trace in ddb shows: : : allocbuf() : getblk() : ffs_balloc() : ffs_write() : vn_rdwr() : elf_coredump() : coredump() This particular panic will have a delayed effect, so the trace probably isn't the problem. Something occured some period of time prior that caused a (probably cache) page to be marked dirty. Then later when vm_page_alloc() tries to reuse the page it notices that it's dirty when it shouldn't be. I have not seen this panic literaly for years - there was a bug in John Dyson's original code that could cause it but it was fixed a very long time ago. So I'm guessing that the panic is a side effect of a newly introduced bug somewhere in current. Possibly a race of some sort with the mutex/threading/interrupt code, but I don't know. -Matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message