From owner-freebsd-current Wed Feb 11 12:06:13 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA08411 for current-outgoing; Wed, 11 Feb 1998 12:06:13 -0800 (PST) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from smtp02.primenet.com (smtp02.primenet.com [206.165.6.132]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA08394 for ; Wed, 11 Feb 1998 12:06:10 -0800 (PST) (envelope-from tlambert@usr07.primenet.com) Received: (from daemon@localhost) by smtp02.primenet.com (8.8.8/8.8.8) id NAA07251; Wed, 11 Feb 1998 13:06:06 -0700 (MST) Received: from usr07.primenet.com(206.165.6.207) via SMTP by smtp02.primenet.com, id smtpd007228; Wed Feb 11 13:06:00 1998 Received: (from tlambert@localhost) by usr07.primenet.com (8.8.5/8.8.5) id NAA10426; Wed, 11 Feb 1998 13:05:57 -0700 (MST) From: Terry Lambert Message-Id: <199802112005.NAA10426@usr07.primenet.com> Subject: Re: Hollywood (Re: PATCH.M ) To: hasty@rah.star-gate.com (Amancio Hasty) Date: Wed, 11 Feb 1998 20:05:57 +0000 (GMT) Cc: tlambert@primenet.com, current@FreeBSD.ORG In-Reply-To: <199802111308.FAA02308@rah.star-gate.com> from "Amancio Hasty" at Feb 11, 98 05:08:15 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > Hi Terry, > > Where is code ? If you want me to test or check out the patches just > let me know. It was posted to -current under the subject "PATCH.C". BTW: Doug Rabson has reviewed these patches in my last monolithic attempt containing the locking patches, the EXCLUDE flag patches, the nameifree patches, and a number of other namei cleanups. The patches I sent him applied cleanly -- most likely because he did not wait for weeks of kernel changes before applying them, unlike some people. His single complaint, which I was unable to resolve to his satisfaction, was that there was a pathname buffer memory leak in the NFS server rename code. The NFS server code must allocate and free pathname buffers, just as the system call code must allocate and free pathname buffers: the patch enforced a more orthogonal VFS interface, and both are VFS consumers. In order to attempt to resolve his problem, I sent him a kernel memory leak test framework. In it, you can snapshot any set of named VM variables (anything you can get back from vmstat), and exervice any combination of VOP code paths. My sample code included with this framework tests only the namei() freeing VOP's. It localized the problem, but he was unable to put any time into fixing it. These new patches do not include the monolithic changes that caused him problems: only the things he was prepared to commit inre: NFS locking. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe current" in the body of the message