Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1423315AbXBHWxm (ORCPT ); Thu, 8 Feb 2007 17:53:42 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1423338AbXBHWxl (ORCPT ); Thu, 8 Feb 2007 17:53:41 -0500 Received: from scrub.xs4all.nl ([194.109.195.176]:35090 "EHLO scrub.xs4all.nl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1423315AbXBHWxl (ORCPT ); Thu, 8 Feb 2007 17:53:41 -0500 Date: Thu, 8 Feb 2007 23:53:09 +0100 (CET) From: Roman Zippel X-X-Sender: roman@scrub.home To: Linus Torvalds cc: linux-kernel@vger.kernel.org, Oleg Verych Subject: Re: The who needs reviews anyways [PATCH] In-Reply-To: Message-ID: References: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1750 Lines: 43 Hi, On Thu, 8 Feb 2007, Linus Torvalds wrote: > Historically, people used to do: > - /bin/sh was the "standard shell" (bash) > - /bin/[t]csh is what clueless weenies use for interactive work. > > (Yeah, I'm not a [t]csh fan ;) > > And you did break that. > > It's quite possible that all modern distributions will install /bin/bash > as a link to /bin/sh, but I don't see the point of that particular change. > We aren't even all that bash-centric any more. If you have a > POSIX-compatible shell in /bin/sh, it really _should_ work. It just can't > be something really broken. I don't quite understand, the Makefile doesn't care anymore about /bin/sh with this patch, the Makefile checks only for $BASH and /bin/bash (equivalent to adding "#! /bin/bash" to scripts) and if the latter fails it's possible some of our scripts will fail. We could make sure that all our scripts are POSIX clean, but is it really worth the effort? It would make casual kbuild hacking only even more difficult, as one has to check it works with the various shells. > > - proper quoting > > - proper indentation > > One thing I'm wondering about is whether we could have a "does this warn" > test. I guess you can do it with -Werror, but it might be nice to have > some tests for "does the -Wxyzzy flag warn also for proper code" Adding something like try-compile should be possible, but we should be careful with this, the more checks we add the more work is done at every make invocation. bye, Roman - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/