Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1423049AbWBBBhQ (ORCPT ); Wed, 1 Feb 2006 20:37:16 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1423047AbWBBBhP (ORCPT ); Wed, 1 Feb 2006 20:37:15 -0500 Received: from ns.miraclelinux.com ([219.118.163.66]:31163 "EHLO mail01.miraclelinux.com") by vger.kernel.org with ESMTP id S1423044AbWBBBhO (ORCPT ); Wed, 1 Feb 2006 20:37:14 -0500 Date: Thu, 2 Feb 2006 10:37:11 +0900 To: "Luck, Tony" Cc: linux-kernel@vger.kernel.org, Richard Henderson , Ivan Kokshaysky , linux-ia64@vger.kernel.org, linuxsh-dev@lists.sourceforge.net, Andi Kleen Subject: Re: [patch 41/44] make thread_info.flags an unsigned long Message-ID: <20060202013711.GA7143@miraclelinux.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.9i From: mita@miraclelinux.com (Akinobu Mita) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1394 Lines: 32 On Wed, Feb 01, 2006 at 10:17:43AM -0800, Luck, Tony wrote: > > --- 2.6-git.orig/include/asm-ia64/thread_info.h > +++ 2.6-git/include/asm-ia64/thread_info.h > @@ -24,7 +24,7 @@ > struct thread_info { > struct task_struct *task; /* XXX not really needed, except for dup_task_struct() */ > struct exec_domain *exec_domain;/* execution domain */ > - __u32 flags; /* thread_info flags (see TIF_*) */ > + unsigned long flags; /* thread_info flags (see TIF_*) */ > __u32 cpu; /* current CPU */ > mm_segment_t addr_limit; /* user-level address space limit */ > int preempt_count; /* 0=premptable, <0=BUG; will also serve as bh-counter */ > > This leaves a useless hole in the structure. Tell me again why > this is a good thing? This patch stops warning message introduced by the "[patch 25/44] ia64: use generic bitops" which replaces __set_bit() family to generic one. But I realize that I can't replace to generic one for ia64. Because __set_bit() family are using the pointer to __u32, but generic __set_bit() family are using the pointer to unsigned long. So the patch 25/44 is wrong and the change above is unnessesary. Thanks - 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/