From owner-freebsd-current@FreeBSD.ORG Fri Jun 17 10:07:35 2005 Return-Path: X-Original-To: freebsd-current@FreeBSD.org Delivered-To: freebsd-current@FreeBSD.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id DED7A16A41C; Fri, 17 Jun 2005 10:07:35 +0000 (GMT) (envelope-from rwatson@FreeBSD.org) Received: from cyrus.watson.org (cyrus.watson.org [204.156.12.53]) by mx1.FreeBSD.org (Postfix) with ESMTP id B1A9543D1F; Fri, 17 Jun 2005 10:07:35 +0000 (GMT) (envelope-from rwatson@FreeBSD.org) Received: from fledge.watson.org (fledge.watson.org [204.156.12.50]) by cyrus.watson.org (Postfix) with ESMTP id 6427A46BC7; Fri, 17 Jun 2005 06:07:35 -0400 (EDT) Date: Fri, 17 Jun 2005 11:09:54 +0100 (BST) From: Robert Watson X-X-Sender: robert@fledge.watson.org To: "Stephane E. Potvin" In-Reply-To: <20050616184127.L27625@fledge.watson.org> Message-ID: <20050617110902.C56734@fledge.watson.org> References: <42B18536.3080200@videotron.ca> <20050616151502.X27625@fledge.watson.org> <42B192D2.7000505@videotron.ca> <20050616181820.E27625@fledge.watson.org> <42B1B784.8010405@videotron.ca> <20050616184127.L27625@fledge.watson.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: alc@FreeBSD.org, freebsd-current@FreeBSD.org Subject: Re: Reboot while booting with new per-CPU allocator X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Jun 2005 10:07:36 -0000 On Thu, 16 Jun 2005, Robert Watson wrote: >> Great :) It did the trick. The laptop is happily booting with the new >> allocator now. Thanks a lot to you and Alan Cox. > > Looks like what basically happened is this these kern_malloc.c changes > increase the memory burden on UMA as statistics structures for malloc > types now get allocated from UMA. It looks like, from your dmesg, you > have a fair number of modules loaded, so the storage for the statistics > comes out of the early UMA page pool, whereas before it came out of BSS. > We'll see if further tuning is required or not with large numbers of > modules. The thing that surprised me, though, is the unclean failure > mode. The other report saw a clean panic which presumably made > debugging it much easier... Interestingly, there's been a bunch of reports of this in the past few days, and there weren't immediately after the malloc commit. I wonder if some other recent change has increased the amount of UMA memory allocated early in the boot, increasing the level of reports... Robert N M Watson