From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id BBB15C77B75 for ; Fri, 21 Apr 2023 07:44:27 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229559AbjDUHo1 (ORCPT ); Fri, 21 Apr 2023 03:44:27 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:52810 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229642AbjDUHo0 (ORCPT ); Fri, 21 Apr 2023 03:44:26 -0400 Received: from elvis.franken.de (elvis.franken.de [193.175.24.41]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id E0EB010D7 for ; Fri, 21 Apr 2023 00:44:24 -0700 (PDT) Received: from uucp (helo=alpha) by elvis.franken.de with local-bsmtp (Exim 3.36 #1) id 1pplRQ-0004xg-00; Fri, 21 Apr 2023 09:44:24 +0200 Received: by alpha.franken.de (Postfix, from userid 1000) id 1F8ECC0122; Fri, 21 Apr 2023 09:41:35 +0200 (CEST) Date: Fri, 21 Apr 2023 09:41:35 +0200 From: Thomas Bogendoerfer To: Jiaxun Yang Cc: Nick Desaulniers , "linux-mips@vger.kernel.org" , llvm@lists.linux.dev, Nathan Chancellor Subject: Re: [PATCH 1/8] MIPS: Replace assembly isa level directives with macros Message-ID: <20230421074134.GA6209@alpha.franken.de> References: <20230409104309.13887-1-jiaxun.yang@flygoat.com> <20230409104309.13887-2-jiaxun.yang@flygoat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.10.1 (2018-07-13) Precedence: bulk List-ID: X-Mailing-List: linux-mips@vger.kernel.org On Thu, Apr 20, 2023 at 08:29:03PM +0100, Jiaxun Yang wrote: > Yes, GAS and LLVM sometimes have different opinions on what a instruction > feature should belong to. Personally I think there is no right or wrong in most case. > > So generally when we try to use some inline assembly features that toolchain > may consider belongs to higher ISA level we will use `.set mips64r2` directives. > > Having this patch just unified the defined arch across the tree, so it happens to fix > some cases where `.set` was given a improper option. I'd prefer, if we don't magically fix something by doing this massive replacement. So first bug fixing then cleanup. And what I don't like is the name of the #defines (I know it's not your choice, ), they don't tell me anything and it's still not clear which one should be used in which case. I see one use case, which is enabling 64bit instruction inside a 32bit kernel. What are the others ? Do we really need all of them ? For example the change in arch/mips/mm/cex-oct.S, this is for a octeon kernel, which only supports and works with 64bit kernels... Thomas. -- Crap can work. Given enough thrust pigs will fly, but it's not necessarily a good idea. [ RFC1925, 2.3 ]