From owner-freebsd-current Sun Aug 26 14:39:27 2001 Delivered-To: freebsd-current@freebsd.org Received: from InterJet.elischer.org (c421509-a.pinol1.sfba.home.com [24.7.86.9]) by hub.freebsd.org (Postfix) with ESMTP id 864B537B405 for ; Sun, 26 Aug 2001 14:39:23 -0700 (PDT) (envelope-from julian@elischer.org) Received: from elischer.org (InterJet.elischer.org [192.168.1.1]) by InterJet.elischer.org (8.9.1a/8.9.1) with ESMTP id OAA69919 for ; Sun, 26 Aug 2001 14:51:37 -0700 (PDT) Message-ID: <3B896A82.DAAB4110@elischer.org> Date: Sun, 26 Aug 2001 14:30:42 -0700 From: Julian Elischer X-Mailer: Mozilla 4.7 [en] (X11; U; FreeBSD 5.0-CURRENT i386) X-Accept-Language: en, hu MIME-Version: 1.0 To: current@freebsd.org Subject: KSE kernel comparissons Content-Type: text/plain; charset=iso-8859-2 Content-Transfer-Encoding: 7bit Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Comparative times for 'make buildworld' for unmodified and KSE (milestone-2) kernels unmodified -current 2138.464u 3358.378s 1:37:39.77 93.8% 842+1080k 45105+176988io 3208pf+0w modified KSE kernel 2143.716u 3363.311s 1:37:50.33 93.8% 841+1081k 45435+176988io 3214pf+0w I'm very glad to see that the overhead added was not too great. (well within the margin of error I'm sure) Same system, same source tree, just different kernel. /usr/obj was deleted before the previous reboot in both cases. Stdout not redirected, running via ssh from another machine. No soft-updates. ref4# size /k* (GENERIC config) text data bss dec hex filename 3180804 275436 350612 3806852 3a1684 /kernel.normal 3188036 275436 350836 3814308 3a33a4 /kernel.kse My guess is that the size increase in the text area is due to the extra code here and there to take the extra dereference from (p) to (td->td_proc) and the places where there is both a td variabel and a p variable. (with extra code to initialise them) Possibly an extra 1K for code to initialise the more complicated structures too. More actual code will be needed to get away from 1:1, so this is just a baseline. the next steps are for us a s a groupt to decide if this is really the way we want to go, and if so, whether we want to commit these changes to make them available for the world to work on as a base for real threading support. The alternative is to do linux-type threading with processes (peter wemm has been investigating a variant on this scheme). This is probably a no-turning-back commit. It's presently checked in on the FreeBSD p4 tree based on freefall, so it's safe, but we need to make a decision on where it goes next. -- +------------------------------------+ ______ _ __ | __--_|\ Julian Elischer | \ U \/ / hard at work in | / \ julian@elischer.org +------>x USA \ a very strange | ( OZ ) \___ ___ | country ! +- X_.---._/ presently in San Francisco \_/ \\ v To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message