From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751083AbWCSPFs (ORCPT ); Sun, 19 Mar 2006 10:05:48 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751504AbWCSPFr (ORCPT ); Sun, 19 Mar 2006 10:05:47 -0500 Received: from omx2-ext.sgi.com ([192.48.171.19]:21391 "EHLO omx2.sgi.com") by vger.kernel.org with ESMTP id S1751083AbWCSPFr (ORCPT ); Sun, 19 Mar 2006 10:05:47 -0500 From: Paul Jackson To: Andrew Morton Cc: "Eric W. Biederman" , Simon.Derr@bull.net, Paul Jackson , linux-kernel@vger.kernel.org Date: Sun, 19 Mar 2006 07:05:38 -0800 Message-Id: <20060319150538.16855.53670.sendpatchset@jackhammer.engr.sgi.com> Subject: [PATCH] Cpuset: remove unnecessary NULL check comment fix Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org From: Paul Jackson Make the comments referring to the (ab)use of the top_cpuset during a tasks exit more explicit - hopefully clearer. This is now called "the_top_cpuset_hack", instead of the "Hack." Signed-off-by: Paul Jackson --- Andrew - this goes right after "cpuset-remove-unnecessary-NULL-check". kernel/cpuset.c | 9 +++++---- 1 files changed, 5 insertions(+), 4 deletions(-) --- 2.6.16-rc6-mm2.orig/kernel/cpuset.c 2006-03-19 06:45:00.600257470 -0800 +++ 2.6.16-rc6-mm2/kernel/cpuset.c 2006-03-19 06:49:27.495730165 -0800 @@ -2023,7 +2023,7 @@ void cpuset_fork(struct task_struct *chi * because tsk is already marked PF_EXITING, so attach_task() won't * mess with it, or task is a failed fork, never visible to attach_task. * - * Hack: + * the_top_cpuset_hack: * * Set the exiting tasks cpuset to the root cpuset (top_cpuset). * @@ -2062,7 +2062,7 @@ void cpuset_exit(struct task_struct *tsk struct cpuset *cs; cs = tsk->cpuset; - tsk->cpuset = &top_cpuset; /* Hack - see comment above */ + tsk->cpuset = &top_cpuset; /* the_top_cpuset_hack - see above */ if (notify_on_release(cs)) { char *pathbuf = NULL; @@ -2375,8 +2375,9 @@ void __cpuset_memory_pressure_bump(void) * - No need to task_lock(tsk) on this tsk->cpuset reference, as it * doesn't really matter if tsk->cpuset changes after we read it, * and we take manage_mutex, keeping attach_task() from changing it - * anyway. No need to check that tsk->cpuset != NULL, thanks to the - * cpuset_exit() Hack. + * anyway. No need to check that tsk->cpuset != NULL, thanks to + * the_top_cpuset_hack in cpuset_exit(), which sets an exiting tasks + * cpuset to top_cpuset. */ static int proc_cpuset_show(struct seq_file *m, void *v) -- I won't rest till it's the best ... Programmer, Linux Scalability Paul Jackson 1.650.933.1373