From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753777Ab3KDQdA (ORCPT ); Mon, 4 Nov 2013 11:33:00 -0500 Received: from imap.thunk.org ([74.207.234.97]:54124 "EHLO imap.thunk.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753397Ab3KDQc6 (ORCPT ); Mon, 4 Nov 2013 11:32:58 -0500 Date: Mon, 4 Nov 2013 11:32:48 -0500 From: "Theodore Ts'o" To: Randy Dunlap Cc: Stephen Rothwell , linux-next@vger.kernel.org, LKML , Kees Cook , Greg Kroah-Hartman Subject: Re: [PATCH -next] random: needs linux/device.h for struct device Message-ID: <20131104163248.GC7108@thunk.org> Mail-Followup-To: Theodore Ts'o , Randy Dunlap , Stephen Rothwell , linux-next@vger.kernel.org, LKML , Kees Cook , Greg Kroah-Hartman References: <20131104175916.763e3c0cf001b9f948f2baa7@canb.auug.org.au> <5277C5F8.3080609@infradead.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <5277C5F8.3080609@infradead.org> User-Agent: Mutt/1.5.21 (2010-09-15) X-SA-Exim-Connect-IP: X-SA-Exim-Mail-From: tytso@thunk.org X-SA-Exim-Scanned: No (on imap.thunk.org); SAEximRunCond expanded to false Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Nov 04, 2013 at 08:06:16AM -0800, Randy Dunlap wrote: > on x86_64: > > drivers/char/random.c: In function 'add_device_attach_randomness': > drivers/char/random.c:837:41: error: dereferencing pointer to incomplete type > > > need to #include for struct device, I think. The patch which introduced this is going to be dropped from the random tree because I'm concerned the amount of entropy that will be added will get overstated, especially on CPU's that don't have cycle counters (ARM and MIPS in particular). I plan to rework it to be safer, although I suspect using device attach times simply won't be doable on CPU's that don't have a cycle counter register, or some other fine-grained, highly light-weight timer (I can't use the clocksource infrastructure since those time sources are not necessarily available during early boot, and there is no guarantee how expensive some of the clock sources might be; some can be quite expensive, IBM X440 servers with their "scalability cable" ). - Ted