From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753469AbXDMNC0 (ORCPT ); Fri, 13 Apr 2007 09:02:26 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753499AbXDMNC0 (ORCPT ); Fri, 13 Apr 2007 09:02:26 -0400 Received: from mail.screens.ru ([213.234.233.54]:33975 "EHLO mail.screens.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753469AbXDMNCO (ORCPT ); Fri, 13 Apr 2007 09:02:14 -0400 Date: Fri, 13 Apr 2007 17:02:01 +0400 From: Oleg Nesterov To: Andrew Morton Cc: Davide Libenzi , "Eric W. Biederman" , Ingo Molnar , Linus Torvalds , "Rafael J. Wysocki" , Roland McGrath , Rusty Russell , linux-kernel@vger.kernel.org Subject: [PATCH 2/3] make kthread_create() more scalable Message-ID: <20070413130201.GA170@tv-sign.ru> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.11 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org If kernel_thread(kthread) succeeds, kthread() can not fail on its path to complete(&create->started) + schedule(). After that it can't be woken because nobody can see the new task yet. This means: - we don't need tasklist_lock for find_task_by_pid(). - create_kthread() doesn't need to wait for create->started. Instead, kthread_create() first waits for create->created to get the result of kernel_thread(), then waits for create->started to synchronize with kthread(). Signed-off-by: Oleg Nesterov --- 2.6.21-rc5/kernel/kthread.c~1_CREATE 2007-04-13 14:39:21.000000000 +0400 +++ 2.6.21-rc5/kernel/kthread.c 2007-04-13 14:52:44.000000000 +0400 @@ -24,11 +24,11 @@ struct kthread_create_info /* Information passed to kthread() from kthreadd. */ int (*threadfn)(void *data); void *data; + struct completion created; struct completion started; /* Result passed back to kthread_create() from kthreadd. */ - struct task_struct *result; - struct completion done; + pid_t result; struct list_head list; }; @@ -91,15 +91,9 @@ static void create_kthread(struct kthrea /* We want our own signal handler (we take no signals by default). */ pid = kernel_thread(kthread, create, CLONE_FS | CLONE_FILES | SIGCHLD); - if (pid < 0) { - create->result = ERR_PTR(pid); - } else { - wait_for_completion(&create->started); - read_lock(&tasklist_lock); - create->result = find_task_by_pid(pid); - read_unlock(&tasklist_lock); - } - complete(&create->done); + create->result = pid; + + complete(&create->created); } /** @@ -127,27 +121,31 @@ struct task_struct *kthread_create(int ( ...) { struct kthread_create_info create; + struct task_struct *ret; + va_list args; create.threadfn = threadfn; create.data = data; + init_completion(&create.created); init_completion(&create.started); - init_completion(&create.done); spin_lock(&kthread_create_lock); list_add_tail(&create.list, &kthread_create_list); - wake_up_process(kthreadd_task); spin_unlock(&kthread_create_lock); + wake_up_process(kthreadd_task); - wait_for_completion(&create.done); + wait_for_completion(&create.created); + if (create.result < 0) + return ERR_PTR(create.result); - if (!IS_ERR(create.result)) { - va_list args; - va_start(args, namefmt); - vsnprintf(create.result->comm, sizeof(create.result->comm), - namefmt, args); - va_end(args); - } - return create.result; + wait_for_completion(&create.started); + ret = find_task_by_pid(create.result); + + va_start(args, namefmt); + vsnprintf(ret->comm, sizeof(ret->comm), namefmt, args); + va_end(args); + + return ret; } EXPORT_SYMBOL(kthread_create);