From patchwork Thu Nov 2 17:04:31 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Davidlohr Bueso X-Patchwork-Id: 10039021 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id AC63360291 for ; Thu, 2 Nov 2017 17:05:50 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 9800629178 for ; Thu, 2 Nov 2017 17:05:50 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 8C7A52918C; Thu, 2 Nov 2017 17:05:50 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-6.9 required=2.0 tests=BAYES_00,RCVD_IN_DNSWL_HI autolearn=unavailable version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 01CE129178 for ; Thu, 2 Nov 2017 17:05:50 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754004AbdKBRFg convert rfc822-to-8bit (ORCPT ); Thu, 2 Nov 2017 13:05:36 -0400 Received: from mx2.suse.de ([195.135.220.15]:56592 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752042AbdKBRFg (ORCPT ); Thu, 2 Nov 2017 13:05:36 -0400 X-Virus-Scanned: by amavisd-new at test-mx.suse.de Received: from relay2.suse.de (charybdis-ext.suse.de [195.135.220.254]) by mx2.suse.de (Postfix) with ESMTP id 43212AC7D; Thu, 2 Nov 2017 17:05:34 +0000 (UTC) Date: Thu, 2 Nov 2017 10:04:31 -0700 From: Davidlohr Bueso To: Waiman Long Cc: Alexander Viro , Jan Kara , Jeff Layton , "J. Bruce Fields" , Tejun Heo , Christoph Lameter , linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, Ingo Molnar , Peter Zijlstra , Andi Kleen , Dave Chinner , Boqun Feng Subject: Re: [PATCH v8 1/6] lib/dlock-list: Distributed and lock-protected lists Message-ID: <20171102170431.oq3i5mxtjcg53uot@linux-n805> References: <1509475860-16139-1-git-send-email-longman@redhat.com> <1509475860-16139-2-git-send-email-longman@redhat.com> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <1509475860-16139-2-git-send-email-longman@redhat.com> User-Agent: NeoMutt/20170421 (1.8.2) Sender: linux-fsdevel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-fsdevel@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP On Tue, 31 Oct 2017, Waiman Long wrote: >+/** >+ * dlock_lists_empty - Check if all the dlock lists are empty >+ * @dlist: Pointer to the dlock_list_heads structure >+ * Return: true if list is empty, false otherwise. >+ * >+ * This can be a pretty expensive function call. If this function is required >+ * in a performance critical path, we may have to maintain a global count >+ * of the list entries in the global dlock_list_heads structure instead. >+ */ I vote for doing this in the original version. How about the following? >+bool dlock_lists_empty(struct dlock_list_heads *dlist) >+{ >+ int idx; >+ >+ for (idx = 0; idx < nr_cpu_ids; idx++) >+ if (!list_empty(&dlist->heads[idx].list)) >+ return false; >+ return true; >+} >+EXPORT_SYMBOL(dlock_lists_empty); ----------8<----------------------------------------------- From: Davidlohr Bueso Subject: [PATCH] lib/dlock-list: Scale dlock_lists_empty() Instead of the current O(N) implementation; at the cost of adding an atomic counter. We also need to add a heads pointer to the node structure such that we can unaccount a thread doing list_del(). Signed-off-by: Davidlohr Bueso --- include/linux/dlock-list.h | 2 ++ lib/dlock-list.c | 40 ++++++++++++++++++++++++++++------------ 2 files changed, 30 insertions(+), 12 deletions(-) diff --git a/include/linux/dlock-list.h b/include/linux/dlock-list.h index c00c7f92ada4..dd73d5787885 100644 --- a/include/linux/dlock-list.h +++ b/include/linux/dlock-list.h @@ -36,6 +36,7 @@ struct dlock_list_head { struct dlock_list_heads { struct dlock_list_head *heads; + atomic_t waiters; }; /* @@ -44,6 +45,7 @@ struct dlock_list_heads { struct dlock_list_node { struct list_head list; struct dlock_list_head *head; + struct dlock_list_heads *heads; }; /* diff --git a/lib/dlock-list.c b/lib/dlock-list.c index a4ddecc01b12..bd11fc0da254 100644 --- a/lib/dlock-list.c +++ b/lib/dlock-list.c @@ -124,6 +124,8 @@ int __alloc_dlock_list_heads(struct dlock_list_heads *dlist, head->lock = __SPIN_LOCK_UNLOCKED(&head->lock); lockdep_set_class(&head->lock, key); } + + atomic_set(&dlist->waiters, 0); return 0; } EXPORT_SYMBOL(__alloc_dlock_list_heads); @@ -139,29 +141,23 @@ void free_dlock_list_heads(struct dlock_list_heads *dlist) { kfree(dlist->heads); dlist->heads = NULL; + atomic_set(&dlist->waiters, 0); } EXPORT_SYMBOL(free_dlock_list_heads); /** * dlock_lists_empty - Check if all the dlock lists are empty * @dlist: Pointer to the dlock_list_heads structure - * Return: true if list is empty, false otherwise. * - * This can be a pretty expensive function call. If this function is required - * in a performance critical path, we may have to maintain a global count - * of the list entries in the global dlock_list_heads structure instead. + * Return: true if all dlock lists are empty, false otherwise. */ bool dlock_lists_empty(struct dlock_list_heads *dlist) { - int idx; - /* Shouldn't be called before nr_dlock_lists is initialized */ WARN_ON_ONCE(!nr_dlock_lists); - for (idx = 0; idx < nr_dlock_lists; idx++) - if (!list_empty(&dlist->heads[idx].list)) - return false; - return true; + smp_mb__before_atomic(); + return !atomic_read(&dlist->waiters); } EXPORT_SYMBOL(dlock_lists_empty); @@ -179,10 +175,30 @@ void dlock_lists_add(struct dlock_list_node *node, struct dlock_list_head *head = &dlist->heads[this_cpu_read(cpu2idx)]; /* + * Serialize dlist->waiters such that a 0->1 transition is not missed, + * by another thread checking if any of the dlock lists are used. + * + * CPU0 CPU1 + * dlock_list_add() dlock_lists_empty() + * [S] atomic_inc(waiters); + * smp_mb__after_atomic(); + * smp_mb__before_atomic(); + * [L] atomic_read(waiters) + * list_add() + * + * Bump the waiters counter _before_ taking the head->lock such that we + * don't miss a thread adding itself to a list while spinning for the + * lock. + */ + atomic_inc(&dlist->waiters); + smp_mb__after_atomic(); + + /* * There is no need to disable preemption */ spin_lock(&head->lock); node->head = head; + node->heads = dlist; list_add(&node->list, &head->list); spin_unlock(&head->lock); } @@ -199,8 +215,7 @@ EXPORT_SYMBOL(dlock_lists_add); * a bug. */ void dlock_lists_del(struct dlock_list_node *node) -{ - struct dlock_list_head *head; +{ struct dlock_list_head *head; bool retry; do { @@ -214,6 +229,7 @@ void dlock_lists_del(struct dlock_list_node *node) list_del_init(&node->list); node->head = NULL; retry = false; + atomic_dec(&node->heads->waiters); } else { /* * The lock has somehow changed. Retry again if it is