From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from thunk.org ([140.239.227.29]:42388 "EHLO thunker.thunk.org") by vger.kernel.org with ESMTP id S267747AbUHPQSN (ORCPT ); Mon, 16 Aug 2004 12:18:13 -0400 Date: Mon, 16 Aug 2004 12:17:56 -0400 From: "Theodore Ts'o" Subject: Re: Using get_cycles for add_timer_randomness Message-ID: <20040816161756.GC16521@thunk.org> References: <200308160303.17612.arnd@arndb.de> <20040810162435.GK24690@krispykreme> <20040814183623.GB5637@krispykreme> <20040815154821.2d6edb03.akpm@osdl.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20040815154821.2d6edb03.akpm@osdl.org> Sender: "Theodore Ts'o" To: Andrew Morton Cc: Anton Blanchard , arnd@arndb.de, richm@oldelvet.org.uk, linux-arch@vger.kernel.org List-ID: On Sun, Aug 15, 2004 at 03:48:21PM -0700, Andrew Morton wrote: > From: Arnd Bergmann > > I noticed that only i386 and x86-64 are currently using a high resolution > timer source when adding randomness. Since many architectures have a > working get_cycles() implementation, it seems rather straightforward to use > that. My only concern about using get_cycles is the speed question; on some architectures, could (particularly those without TSC registers or equivalent hardware support) could get_cycles() be slow enough to cause latency problems? - Ted