From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932100AbYBZUDS (ORCPT ); Tue, 26 Feb 2008 15:03:18 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752598AbYBZUDF (ORCPT ); Tue, 26 Feb 2008 15:03:05 -0500 Received: from pne-smtpout3-sn2.hy.skanova.net ([81.228.8.111]:54373 "EHLO pne-smtpout3-sn2.hy.skanova.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751682AbYBZUDE (ORCPT ); Tue, 26 Feb 2008 15:03:04 -0500 Date: Tue, 26 Feb 2008 22:02:17 +0200 From: Pekka Paalanen To: Andrew Morton Cc: linux-kernel@vger.kernel.org, Ingo Molnar , Christoph Hellwig , Arjan van de Ven , Pavel Roskin , Steven Rostedt , Peter Zijlstra , Andy Whitcroft Subject: Re: [RFC] mmiotrace full patch, preview 1 Message-ID: <20080226220217.747d47a5@daedalus.pq.iki.fi> In-Reply-To: <20080225144922.ae716e7d.akpm@linux-foundation.org> References: <20080224190323.77adf683@daedalus.pq.iki.fi> <20080225144922.ae716e7d.akpm@linux-foundation.org> X-Mailer: Claws Mail 3.0.2 (GTK+ 2.12.5; x86_64-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 25 Feb 2008 14:49:22 -0800 Andrew Morton wrote: > Please feed the diff through scritps/checkpatch.pl and consider addressing > the things which it finds. I checked that, but I didn't think any of them were worth fixing. And since this is a work in progress and a in a state which I do not want committed yet, I did not sign this patch. mmio-mod.c is still a mess. > WARNING: Use of volatile is usually wrong: see Documentation/volatile-considered-harmful.txt > #1097: FILE: arch/x86/mm/mmio-mod.c:437: > +void iounmap_trace(volatile void __iomem *addr) I believe the 'volatile' belongs here, it is also in the declaration of iounmap() in arch/x86/mm/ioremap.c. > WARNING: externs should be avoided in .c files > #1766: FILE: arch/x86/mm/testmmiotrace.c:7: > +extern void __iomem *ioremap_nocache_trace(unsigned long offset, > > WARNING: Use of volatile is usually wrong: see Documentation/volatile-considered-harmful.txt > #1768: FILE: arch/x86/mm/testmmiotrace.c:9: > +extern void iounmap_trace(volatile void __iomem *addr); > > WARNING: externs should be avoided in .c files > #1768: FILE: arch/x86/mm/testmmiotrace.c:9: > +extern void iounmap_trace(volatile void __iomem *addr); These are all in testmmiotrace.c which calls the traced ioremap functions directly. These direct calls will go away and it will call the normal ioremap functions. > WARNING: Use of volatile is usually wrong: see Documentation/volatile-considered-harmful.txt > #1790: FILE: arch/x86/mm/testmmiotrace.c:31: > + volatile unsigned int v; Since testmmiotrace does not use v for anything other than return value of ioread8/16/32(), I wanted to prevent the compiler from optimizing it away. Now thinking it again, ioread*() must guarantee that the read is always executed. I'll remove v altogether. Patch for the other issues you mentioned is brewing. Thanks. -- Pekka Paalanen http://www.iki.fi/pq/