Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753214AbYKRX42 (ORCPT ); Tue, 18 Nov 2008 18:56:28 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751737AbYKRX4S (ORCPT ); Tue, 18 Nov 2008 18:56:18 -0500 Received: from one.firstfloor.org ([213.235.205.2]:50599 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751035AbYKRX4S (ORCPT ); Tue, 18 Nov 2008 18:56:18 -0500 Date: Wed, 19 Nov 2008 01:06:05 +0100 From: Andi Kleen To: Roland McGrath Cc: Andi Kleen , Ingo Molnar , Jan Beulich , heukelum@fastmail.fm, Thomas Gleixner , Alexander van Heukelum , Glauber Costa , LKML , Nick Piggin , "H. Peter Anvin" Subject: Re: [RFC,v2] x86_64: save_args out of line Message-ID: <20081119000605.GQ6703@one.firstfloor.org> References: <20081117175232.GA13766@mailshack.com> <49228648.76E4.0078.0@novell.com> <20081118111633.GA21036@mailshack.com> <4922C863.76E4.0078.0@novell.com> <20081118140349.GC23479@elte.hu> <4922E4D4.76E4.0078.0@novell.com> <20081118150024.GD30358@elte.hu> <20081118225336.60FFA1544EB@magilla.localdomain> <20081118233552.GP6703@one.firstfloor.org> <20081118234511.80C4F1544F1@magilla.localdomain> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20081118234511.80C4F1544F1@magilla.localdomain> User-Agent: Mutt/1.4.2.1i Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1520 Lines: 30 On Tue, Nov 18, 2008 at 03:45:11PM -0800, Roland McGrath wrote: > Yes, I mean a single macro that produces both the instruction and the CFI > pseudo-op to go with it. This is the essential characteristic that makes > it an improvement for maintaining the code. The main problem we have now > is that it's easy to write/modify plain assembly instructions and forget to > add or update the CFI to match. A well-considered set of macros can solve > this without making it any harder for the average assembly programmer to > understand what each line of the source means intuitively. Hmm, but if the assembler cannot auto generate it how should the assembler writer know if he should use the macro or the direct instruction without understanding CFI? Also what will the assembler reader do? Do they first have to understand CFI to understand everything? I personally would probably just resort to objdump -S in this situation. I think you're saying that for the user the macros would be just equivalent, but if that's true they could be just auto generated by the assembler. But it's obviously not, so you'll end up with the Linux magic asm dialect (and its maintenance disadvantages) and you'll still require CFI knowledge to understand/write everything anyways. -Andi -- 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/