From owner-freebsd-current@FreeBSD.ORG Tue Jul 20 14:33:46 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 3B4F316A4CF; Tue, 20 Jul 2004 14:33:46 +0000 (GMT) Received: from storm.FreeBSD.org.uk (storm.FreeBSD.org.uk [194.242.157.42]) by mx1.FreeBSD.org (Postfix) with ESMTP id B1AEE43D48; Tue, 20 Jul 2004 14:33:45 +0000 (GMT) (envelope-from mark@grondar.org) Received: from storm.FreeBSD.org.uk (Ugrondar@localhost [127.0.0.1]) i6KEXisd021058; Tue, 20 Jul 2004 15:33:44 +0100 (BST) (envelope-from mark@grondar.org) Received: (from Ugrondar@localhost)i6KEXiO8021057; Tue, 20 Jul 2004 15:33:44 +0100 (BST) (envelope-from mark@grondar.org) X-Authentication-Warning: storm.FreeBSD.org.uk: Ugrondar set sender to mark@grondar.org using -f Received: from grondar.org (localhost [127.0.0.1])i6KEJxJX079041; Tue, 20 Jul 2004 15:19:59 +0100 (BST) (envelope-from mark@grondar.org) Message-Id: <200407201419.i6KEJxJX079041@grimreaper.grondar.org> To: Robert Watson From: Mark Murray In-Reply-To: Your message of "Tue, 20 Jul 2004 09:32:08 EDT." Date: Tue, 20 Jul 2004 15:19:59 +0100 Sender: mark@grondar.org cc: jesk cc: freebsd-current@FreeBSD.ORG Subject: Re: I/O or Threading Suffer X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 20 Jul 2004 14:33:46 -0000 Robert Watson writes: > On Tue, 20 Jul 2004, Scott Long wrote: > > > If you are dd'ing from /dev/random, then you are depleting the entropy > > pool. Anything else that tries to get random numbers is going to block > > in strange ways. Trying just dd'ing from /dev/zero and see if that > > makes a difference. > > With Yarrow, /dev/random should just keep on chugging, so this is unlikely > to be the source. However, lots of /dev/random I/O may take a while to > yield if it's working hard in kernel, so if that is combined with a > scheduling nit of some sort, we could be looking at a starvation issue. When you read /dev/random, you only exercise the output generator, and this is a pretty simple piece of code. Yarrow's kernel thread is used ONLY for the entropy harvesting, and is only indirectly coupled to the output generator. M -- Mark Murray iumop ap!sdn w,I idlaH