From owner-freebsd-hackers@FreeBSD.ORG Wed Aug 6 10:05:32 2003 Return-Path: Delivered-To: freebsd-hackers@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4806F37B404 for ; Wed, 6 Aug 2003 10:05:32 -0700 (PDT) Received: from HAL9000.homeunix.com (ip114.bella-vista.sfo.interquest.net [66.199.86.114]) by mx1.FreeBSD.org (Postfix) with ESMTP id 359CC43FAF for ; Wed, 6 Aug 2003 10:05:31 -0700 (PDT) (envelope-from das@freebsd.org) Received: from HAL9000.homeunix.com (localhost [127.0.0.1]) by HAL9000.homeunix.com (8.12.9/8.12.9) with ESMTP id h76H5IjX007219; Wed, 6 Aug 2003 10:05:19 -0700 (PDT) (envelope-from das@freebsd.org) Received: (from das@localhost) by HAL9000.homeunix.com (8.12.9/8.12.9/Submit) id h76H5HtJ007218; Wed, 6 Aug 2003 10:05:17 -0700 (PDT) (envelope-from das@freebsd.org) Date: Wed, 6 Aug 2003 10:05:17 -0700 From: David Schultz To: Jan Grant Message-ID: <20030806170517.GA6961@HAL9000.homeunix.com> Mail-Followup-To: Jan Grant , Roderick van Domburg , freebsd-hackers References: <20030806044351.GA3881@HAL9000.homeunix.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: cc: freebsd-hackers cc: Roderick van Domburg Subject: Re: Tuning HZ for semi-realtime applications X-BeenThere: freebsd-hackers@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Technical Discussions relating to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 06 Aug 2003 17:05:32 -0000 On Wed, Aug 06, 2003, Jan Grant wrote: > On Tue, 5 Aug 2003, David Schultz wrote: > > > 100 Hz works just fine for > > interactive jobs; humans can't tell the difference.[1] > > They can if they're using X :-) I gave Denim* a trial recently; it was > unusable at 100Hz and fine at 1000. Yes, polling a tablet device is an interesting case where it matters. If you reread my last message carefully, you'll noticed that I suggested bumping up the default HZ a bit, since the overhead appears to be well into the noise even at 500 Hz. An alternative would be to implement a high-resolution timer facility similar to the Solaris cyclics implementation.