From owner-freebsd-arch Mon Dec 10 15:40:28 2001 Delivered-To: freebsd-arch@freebsd.org Received: from duke.cs.duke.edu (duke.cs.duke.edu [152.3.140.1]) by hub.freebsd.org (Postfix) with ESMTP id F113E37B419; Mon, 10 Dec 2001 15:40:20 -0800 (PST) Received: from grasshopper.cs.duke.edu (grasshopper.cs.duke.edu [152.3.145.30]) by duke.cs.duke.edu (8.9.3/8.9.3) with ESMTP id SAA09397; Mon, 10 Dec 2001 18:40:15 -0500 (EST) Received: (from gallatin@localhost) by grasshopper.cs.duke.edu (8.11.3/8.9.1) id fBANdoT14965; Mon, 10 Dec 2001 18:39:50 -0500 (EST) (envelope-from gallatin@cs.duke.edu) From: Andrew Gallatin MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <15381.18374.269597.931300@grasshopper.cs.duke.edu> Date: Mon, 10 Dec 2001 18:39:50 -0500 (EST) To: Matthew Dillon Cc: John Baldwin , freebsd-arch@FreeBSD.org Subject: Re: Proposed auto-sizing patch to sysinstall (was Re: Using a la In-Reply-To: <200112102221.fBAMLD648896@apollo.backplane.com> References: <200112102221.fBAMLD648896@apollo.backplane.com> X-Mailer: VM 6.75 under 21.1 (patch 12) "Channel Islands" XEmacs Lucid Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Matthew Dillon writes: > > In regards to the "dump" being smarter... system core dump is only > a dump of physical memory. It does not dump VM Objects. Of course, > it does wind up dumping whatever is in physical memory, such as the > file cache, but it's important that it do so or we kernel hackers > would have a hard time tracking problems down. Also, you don't want I think you're one of the few people who might actually care about the contents of the buffercache. That's why it would be optional. > to make a system core dump too 'smart'... the system could be in a > very corrupted state when dumping so we can't safely traverse dozens of > system structures. True, but in nearly three years of using partial dumps, I can't remember that ever happening. And I was mucking with the VM system in my zero-copy work. Again, that's why it would be optional. Anyway, have a look at the partial dump diffs done 2 years ago by Darrell Anderson. http://www.cs.duke.edu/~anderson/freebsd/partialdump These are old, non-context diffs to 4.0 (at the time -current). There'd be some mergework to do, but I think it would be quite handy. Drew PS: We eventually abandoned partial dump in favor of netdump, which works on roughly the same principals but is quite a bit faster, but more of a local hack: http://www.cs.duke.edu/~anderson/freebsd/netdump/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message