From owner-freebsd-current Thu Aug 26 13:15: 1 1999 Delivered-To: freebsd-current@freebsd.org Received: from apollo.backplane.com (apollo.backplane.com [209.157.86.2]) by hub.freebsd.org (Postfix) with ESMTP id 6E2D514C38 for ; Thu, 26 Aug 1999 13:14:54 -0700 (PDT) (envelope-from dillon@apollo.backplane.com) Received: (from dillon@localhost) by apollo.backplane.com (8.9.3/8.9.1) id NAA24637; Thu, 26 Aug 1999 13:13:20 -0700 (PDT) (envelope-from dillon) Date: Thu, 26 Aug 1999 13:13:20 -0700 (PDT) From: Matthew Dillon Message-Id: <199908262013.NAA24637@apollo.backplane.com> To: "John W. DeBoskey" Cc: phk@critter.freebsd.dk (Poul-Henning Kamp), julian@whistle.com, jwd@unx.sas.com, freebsd-current@FreeBSD.ORG Subject: Re: -current kernel problems (spec_getpages & vm_fault) References: <199908261958.PAA46391@bb01f39.unx.sas.com> Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG That fixes a problem with ccd, but not the one causing John's failures. You will note that with John's failure's the I/O is properly page-aligned. The fix to ccd deals with a misalignment problem. But I believe I have found the problem... it is a bug in vm/swap_pager.c that occurs when you have more then one swap partition. The code is improperly crossing a physical disk device boundry and vm/vm_swap.c is returning a failure because of that. For example, if you attempt to read 8K from a swap-backed VN device and 4K of it is in one swap partition and the other 4K is in another, vm/swap_pager.c is not properly breaking up the transaction. I'll have a fix in a moment. -Matt :which would be this commit... : :phk 1999/08/26 07:46:11 PDT : : Modified files: : sys/dev/ccd ccd.c : Log: : Initialize the dev->si_bsize fields. : : Submitted by: tegge : Reviewed by: phk : : Revision Changes Path : 1.53 +5 -1 src/sys/dev/ccd/ccd.c : : : I need to figure out why my 11:30am EST cvsup didn't pick :this file up. It's 44 minutes infront of cvsup... : :Thanks! :john To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message