From owner-cvs-all@FreeBSD.ORG Thu Jan 15 03:40:45 2004 Return-Path: Delivered-To: cvs-all@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E32D816A4CE; Thu, 15 Jan 2004 03:40:45 -0800 (PST) Received: from smtp.des.no (flood.des.no [217.116.83.31]) by mx1.FreeBSD.org (Postfix) with ESMTP id 103A243D49; Thu, 15 Jan 2004 03:40:44 -0800 (PST) (envelope-from des@des.no) Received: by smtp.des.no (Pony Express, from userid 666) id 186A6530A; Thu, 15 Jan 2004 12:40:43 +0100 (CET) Received: from dwp.des.no (des.no [80.203.228.37]) by smtp.des.no (Pony Express) with ESMTP id 6E92A5309; Thu, 15 Jan 2004 12:40:34 +0100 (CET) Received: by dwp.des.no (Postfix, from userid 2602) id 0CAC033C6A; Thu, 15 Jan 2004 12:40:34 +0100 (CET) To: Tim Robbins References: <200401151015.i0FAF49u009868@repoman.freebsd.org> <20040115111219.GA20914@cat.robbins.dropbear.id.au> From: des@des.no (Dag-Erling =?iso-8859-1?q?Sm=F8rgrav?=) Date: Thu, 15 Jan 2004 12:40:34 +0100 In-Reply-To: <20040115111219.GA20914@cat.robbins.dropbear.id.au> (Tim Robbins's message of "Thu, 15 Jan 2004 22:12:19 +1100") Message-ID: User-Agent: Gnus/5.090024 (Oort Gnus v0.24) Emacs/21.3 (berkeley-unix) MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable X-Spam-Checker-Version: SpamAssassin 2.61 (1.212.2.1-2003-12-09-exp) on flood.des.no X-Spam-Level: ss X-Spam-Status: No, hits=2.6 required=5.0 tests=RCVD_IN_DYNABLOCK, RCVD_IN_SORBS autolearn=no version=2.61 cc: cvs-src@FreeBSD.org cc: src-committers@FreeBSD.org cc: cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/kern init_main.c kern_descrip.c sys_pipe.c uipc_syscalls.c uipc_usrreq.c vfs_syscalls.c src/sys/sys filedesc.h X-BeenThere: cvs-all@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: CVS commit messages for the entire tree List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 15 Jan 2004 11:40:46 -0000 Tim Robbins writes: > Thanks for committing this. I'm interested, though, what was the reason y= ou > changed it to use a single bitmap instead of two levels? The increased complexity didn't actually translate into a performance improvement. There was no reason to use two levels instead of one. The theoretical advantage is logarithmic rather than linear early growth, but in practice it wasn't noticeable. The cost of the logic required to maintain and use the high-level bitmap outweighs the gain (which is to save a handful of comparisons against zero in the low-level bitmap). DES --=20 Dag-Erling Sm=F8rgrav - des@des.no