From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261449AbVGCO1u (ORCPT ); Sun, 3 Jul 2005 10:27:50 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261452AbVGCO1u (ORCPT ); Sun, 3 Jul 2005 10:27:50 -0400 Received: from 167.imtp.Ilyichevsk.Odessa.UA ([195.66.192.167]:7583 "HELO port.imtp.ilyichevsk.odessa.ua") by vger.kernel.org with SMTP id S261449AbVGCO1n (ORCPT ); Sun, 3 Jul 2005 10:27:43 -0400 From: Denis Vlasenko To: Jesper Juhl Subject: Re: [PATCH] all-arch delay functions take 2 Date: Sun, 3 Jul 2005 17:27:27 +0300 User-Agent: KMail/1.5.4 Cc: linux-kernel@vger.kernel.org, Arjan van de Ven , Andrew Morton , Russell King , Vojtech Pavlik References: <200507031650.13770.vda@ilport.com.ua> <9a8748490507030714d2a1b45@mail.gmail.com> In-Reply-To: <9a8748490507030714d2a1b45@mail.gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset="koi8-r" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200507031727.27445.vda@ilport.com.ua> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sunday 03 July 2005 17:14, Jesper Juhl wrote: > On 7/3/05, Denis Vlasenko wrote: > > Hi folks, > > > > This patch makes mdelay/udelay/ndelay calls as small > > as possible (just a function call), uses macros which > > do compile-time checks on delay duration if parameter > > is a constant, otherwise check is done at run time > > to prevent udelay(-1) disasters. > > > [snip] > > > > > -#define ndelay(n) udelay((n) * 5) > > +//BOGUS! #define ndelay(n) udelay((n) * 5) > > > If it's bogus, why not just remove it instead of leaving it in as a > comment (and a C++ style comment at that) ? I wanted to draw attention of someone who knows what's going on. Can be removed altogether instead. > [snip] > > +//extern struct timer_opts* timer; > > Why add this at all if you are just going to comment it out anyway? Sorry, this is a leftover meant to be deleted. -- vda