From owner-freebsd-current Thu Dec 17 16:10:14 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id QAA10441 for freebsd-current-outgoing; Thu, 17 Dec 1998 16:10:14 -0800 (PST) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from alpo.whistle.com (alpo.whistle.com [207.76.204.38]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id QAA10415 for ; Thu, 17 Dec 1998 16:10:04 -0800 (PST) (envelope-from julian@whistle.com) Received: (from daemon@localhost) by alpo.whistle.com (8.8.5/8.8.5) id QAA13024; Thu, 17 Dec 1998 16:02:24 -0800 (PST) Received: from current1.whistle.com(207.76.205.22) via SMTP by alpo.whistle.com, id smtpda13022; Fri Dec 18 00:02:21 1998 Message-ID: <36799B88.2781E494@whistle.com> Date: Thu, 17 Dec 1998 16:02:16 -0800 From: Julian Elischer Organization: Whistle Communications X-Mailer: Mozilla 3.0Gold (X11; I; FreeBSD 2.2.7-RELEASE i386) MIME-Version: 1.0 To: Brian Feldman CC: "current@FreeBSD.ORG" Subject: Re: Linux Threads patches available References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Brian Feldman wrote: > For one, I object. Keep it as a patchset, or make a vendor > branch, but the overhead and memory use of these modifications > needs to be considered carefully! I know I did much of the > LinuxThreads work, and would love to see > it going into the kernel, but the ramifications need to be > examined. I've been looking closely at the linux_clone() and signal changes as modified by Richard. I really see no change that I consider harmful. I won't check this in without a couple of core sponsors but I think that the patches as they stand are really close enough to production-quality to be usable. A slight change might be to make the process signal structure a zone allocated item rather than a malloc'd structure, however I don't think that that should be a real problem. The sructure is 328 (?) bytes long, but it already exists. The fields are simply being shifted from one place to another. The biggest thing is the (struct sigacts) which moves from the u area to kvm. Since the U area is fixed in size (always has a page available) We do lose some KVM. However with the amount of work going into Linux Threads we really will gain by piggybacking on this, and we will need these changes anyway to be able to develope and POSIX complient code. People have been very silent on thi which is amazing considering what a step forwards this is... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message