From owner-freebsd-arch Wed Dec 13 6:36: 0 2000 From owner-freebsd-arch@FreeBSD.ORG Wed Dec 13 06:35:58 2000 Return-Path: Delivered-To: freebsd-arch@freebsd.org Received: from mail-relay.eunet.no (mail-relay.eunet.no [193.71.71.242]) by hub.freebsd.org (Postfix) with ESMTP id D05ED37B400 for ; Wed, 13 Dec 2000 06:35:57 -0800 (PST) Received: from login-1.eunet.no (login-1.eunet.no [193.75.110.2]) by mail-relay.eunet.no (8.9.3/8.9.3/GN) with ESMTP id PAA88042; Wed, 13 Dec 2000 15:35:53 +0100 (CET) (envelope-from mbendiks@eunet.no) Received: from localhost (mbendiks@localhost) by login-1.eunet.no (8.9.3/8.8.8) with ESMTP id PAA37292; Wed, 13 Dec 2000 15:35:53 +0100 (CET) (envelope-from mbendiks@eunet.no) X-Authentication-Warning: login-1.eunet.no: mbendiks owned process doing -bs Date: Wed, 13 Dec 2000 15:35:53 +0100 (CET) From: Marius Bendiksen To: Seigo Tanimura Cc: dillon@earth.backplane.com, bright@wintelcom.net, arch@FreeBSD.ORG Subject: Re: Even 1GB KVA is not enough, but we have no more space In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > Sorry, please forget this matter. The culprit truned out to be a race > condition of testing-and-setting ffs_inode_hash_lock in > ffs_vget(). Mutexifing the test-and-sleep-or-set and wakeup-reset > operations was enough to fix the problem. Now my kernel allocates up > to 316K vnodes on make buildworld and release, so malloc(9) pool of > 200MB is sufficient. (316K vnodes occupy about 90-100MB in kmem_map) For older systems, I used to replace the wakeup with wakeup_one. I guess that would reduce lock contention too. I sent a patch on that topic a while ago to one of the lists when someone was seeing problems with ffs_vget deadlocks. Marius To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message