From mboxrd@z Thu Jan 1 00:00:00 1970 From: Yang Hongyang Subject: Re: [PATCH] remove DMA_nBIT_MASK macro Date: Thu, 07 May 2009 16:29:21 +0800 Message-ID: <4A029BE1.4090309@cn.fujitsu.com> References: <20090507141114M.fujita.tomonori@lab.ntt.co.jp> <20090506223452.65e545d8.akpm@linux-foundation.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: FUJITA Tomonori , linux-kernel@vger.kernel.org, netdev@vger.kernel.org, linux-arm-kernel@lists.arm.linux.org.uk To: Andrew Morton Return-path: Received: from cn.fujitsu.com ([222.73.24.84]:57093 "EHLO song.cn.fujitsu.com" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1756106AbZEGI2Z (ORCPT ); Thu, 7 May 2009 04:28:25 -0400 In-Reply-To: <20090506223452.65e545d8.akpm@linux-foundation.org> Sender: netdev-owner@vger.kernel.org List-ID: Andrew Morton wrote: > On Thu, 7 May 2009 14:14:16 +0900 FUJITA Tomonori wrote: > >> We replaced all DMA_nBIT_MASK macros with DMA_BIT_MASK(n) but why do >> we still keep DMA_nBIT_MASK macros in include/linux/dma-mapping.h? >> >> As long as these macros exist, people use them. The current git has >> two users and linux-next have other users. >> >> Is it better to remove DMA_nBIT_MASK macros completely now? > > Yes, the plan is to remove them. > > Doing so will break lots and lots of out-of-tree drivers, causing > people some grief. Is there any way in which we can cause their use to > cause __deprecated warnings for a couple of months, to give people a > chance to migrate? > > Shall we use something like below to warn people? #define DMA_64BIT_MASK DMA_BIT_MASK(64);(__deprecated warnings:use DMA_BIT_MASK(64) instead) ? -- Regards Yang Hongyang