From: Theodore Ts'o Subject: Re: better patch for linux/bitops.h Date: Wed, 4 May 2016 23:50:28 -0400 Message-ID: <20160505035028.GD10776@thunk.org> References: <20160504190723.GD3901@thunk.org> <572A6CDD.10503@av8n.com> <572A6F1C.2080708@av8n.com> <28624BFC-7C63-4F38-9F67-7CBFB0C6499B@zytor.com> <0015E1DE-DFF9-4CCE-805E-7AC286021BED@zytor.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: "H. Peter Anvin" , John Denker , linux-kernel@vger.kernel.org, Stephan Mueller , Herbert Xu , Andi Kleen , Sandy Harris , cryptography@lakedaemon.net, linux-crypto@vger.kernel.org To: Jeffrey Walton Return-path: Received: from imap.thunk.org ([74.207.234.97]:48374 "EHLO imap.thunk.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755160AbcEEDuj (ORCPT ); Wed, 4 May 2016 23:50:39 -0400 Content-Disposition: inline In-Reply-To: Sender: linux-crypto-owner@vger.kernel.org List-ID: Instead of arguing over who's "sane" or "insane", can we come up with a agreed upon set of tests, and a set of compiler and compiler versions for which these tests must achieve at least *working* code? Bonus points if they achieve optimal code, but what's important is that for a wide range of GCC versions (from ancient RHEL distributions to bleeding edge gcc 5.x compilers) this *must* work. >From my perspective, clang and ICC producing corret code is a very nice to have, but most shops I know of don't yet assume that clang produces code that is trustworthy for production systems, although it *is* great for as far as generating compiler warnings to find potential bugs. But instead of arguing over what works and doesn't, let's just create the the test set and just try it on a wide range of compilers and architectures, hmmm? - Ted