From patchwork Wed Aug 30 02:54:44 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: NeilBrown X-Patchwork-Id: 13369922 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id B5B26C6FA8F for ; Wed, 30 Aug 2023 02:59:19 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235449AbjH3C6u (ORCPT ); Tue, 29 Aug 2023 22:58:50 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:43700 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S234124AbjH3C6V (ORCPT ); Tue, 29 Aug 2023 22:58:21 -0400 Received: from smtp-out2.suse.de (smtp-out2.suse.de [195.135.220.29]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 5B33E185 for ; Tue, 29 Aug 2023 19:58:19 -0700 (PDT) Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by smtp-out2.suse.de (Postfix) with ESMTPS id 1FAAB1F461; Wed, 30 Aug 2023 02:58:18 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1693364298; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=Y8jphQss9HeYy7fr6607tDpyIo8Hc3AHgFDPPac0GpA=; b=hqaNSsonP4tWNLuT+HHELqoxn4J7Lz8tUbH+t2UDwpiAypOEFLgorafAW0i3WUdpnx9XCx 4OIk2fjwCjiIqvXkSQ0U1lRAHWXkLD4zIcux7HTeIO/z/mx3W6SJDu/uQXq0iEb6x8U1rN aYd2lDAujBHPY8V3YzrZilRyERZs23I= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1693364298; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=Y8jphQss9HeYy7fr6607tDpyIo8Hc3AHgFDPPac0GpA=; b=6gR6xs8jAK6aX9RKGRzmjQVNsyf2C56ur+7vg+RwG/8zlMyTPaMBXuZeMHsb6q9cOEwoBZ 91U8GQaqOCA00ECQ== Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by imap2.suse-dmz.suse.de (Postfix) with ESMTPS id D565213301; Wed, 30 Aug 2023 02:58:16 +0000 (UTC) Received: from dovecot-director2.suse.de ([192.168.254.65]) by imap2.suse-dmz.suse.de with ESMTPSA id w0bXIUiw7mTZYwAAMHmgww (envelope-from ); Wed, 30 Aug 2023 02:58:16 +0000 From: NeilBrown To: Chuck Lever , Jeff Layton Cc: linux-nfs@vger.kernel.org Subject: [PATCH 01/10] SQUASH: revise comments in SUNRPC: change service idle list to be an llist Date: Wed, 30 Aug 2023 12:54:44 +1000 Message-ID: <20230830025755.21292-2-neilb@suse.de> X-Mailer: git-send-email 2.41.0 In-Reply-To: <20230830025755.21292-1-neilb@suse.de> References: <20230830025755.21292-1-neilb@suse.de> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-nfs@vger.kernel.org Revise some comments to hopefully make the more clear and less verbose. Signed-off-by: NeilBrown --- include/linux/sunrpc/svc.h | 6 +++--- net/sunrpc/svc_xprt.c | 20 +++++++------------- 2 files changed, 10 insertions(+), 16 deletions(-) diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index 5216f95411e3..ed20a2ea1f81 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -271,7 +271,7 @@ enum { * @rqstp: the thread which is now busy * * By convention a thread is busy if rq_idle.next points to rq_idle. - * This ensures it is not on the idle list. + * This will never be the case for threads on the idle list. */ static inline void svc_thread_set_busy(struct svc_rqst *rqstp) { @@ -283,9 +283,9 @@ static inline void svc_thread_set_busy(struct svc_rqst *rqstp) * @rqstp: the thread which might be busy * * By convention a thread is busy if rq_idle.next points to rq_idle. - * This ensures it is not on the idle list. + * This will never be the case for threads on the idle list. */ -static inline bool svc_thread_busy(struct svc_rqst *rqstp) +static inline bool svc_thread_busy(const struct svc_rqst *rqstp) { return rqstp->rq_idle.next == &rqstp->rq_idle; } diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 81327001e074..17c43bde35c9 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -734,22 +734,16 @@ static void svc_rqst_wait_for_work(struct svc_rqst *rqstp) llist_add(&rqstp->rq_idle, &pool->sp_idle_threads); if (unlikely(!rqst_should_sleep(rqstp))) - /* maybe there were no idle threads when some work - * became ready and so nothing was woken. We've just - * become idle so someone can to the work - maybe us. - * But we cannot reliably remove ourselves from the - * idle list - we can only remove the first task which - * might be us, and might not. - * So remove and wake it, then schedule(). If it was - * us, we won't sleep. If it is some other thread, they - * will do the work. + /* Work just became available. This thread cannot simply + * choose not to sleep as it *must* wait until removed. + * So wake the first waiter - whether it is this + * thread or some other, it will get the work done. */ svc_pool_wake_idle_thread(pool); - /* We mustn't continue while on the idle list, and we - * cannot remove outselves reliably. The only "work" - * we can do while on the idle list is to freeze. - * So loop until someone removes us + /* Since a thread cannot remove itself from an llist, + * schedule until someone else removes @rqstp from + * the idle list. */ while (!svc_thread_busy(rqstp)) { schedule(); From patchwork Wed Aug 30 02:54:45 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: NeilBrown X-Patchwork-Id: 13369921 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id EFDE0C83F12 for ; Wed, 30 Aug 2023 02:59:19 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S234124AbjH3C6u (ORCPT ); Tue, 29 Aug 2023 22:58:50 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:43712 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S239380AbjH3C60 (ORCPT ); Tue, 29 Aug 2023 22:58:26 -0400 Received: from smtp-out2.suse.de (smtp-out2.suse.de [IPv6:2001:67c:2178:6::1d]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 5FDE4185 for ; Tue, 29 Aug 2023 19:58:24 -0700 (PDT) Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by smtp-out2.suse.de (Postfix) with ESMTPS id 169FC1F461; Wed, 30 Aug 2023 02:58:23 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1693364303; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=p/rjT0WyrDuNAK7/O2cc5TCZ/nbr9qia158n5WxZS+c=; b=i+f7rPFT6IFxmpb5z0IVt0jdNUOA17QovWRjCgyzSYW2RdY9vZcdCxe7FxXpvrqpZELZg7 sugtaJQrYoUeoUgmrC98Lrk+1vD/qqpnZoZxDoJF3tPxMzjapPzSViyRCTj3kxCmxk8qkW sD1rKIMyTS66xbeQFAu7dtk7THx7gvw= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1693364303; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=p/rjT0WyrDuNAK7/O2cc5TCZ/nbr9qia158n5WxZS+c=; b=K7C1pi1NeIaSgB0WV1xUfgk9rDNSZLQgid1wUAmp8xzIRnpBt0JY/rd7LJ+wVu+zl5LGa6 zLkqkDBrGCjRwbAA== Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by imap2.suse-dmz.suse.de (Postfix) with ESMTPS id C90FF13301; Wed, 30 Aug 2023 02:58:21 +0000 (UTC) Received: from dovecot-director2.suse.de ([192.168.254.65]) by imap2.suse-dmz.suse.de with ESMTPSA id 6h+LHk2w7mTjYwAAMHmgww (envelope-from ); Wed, 30 Aug 2023 02:58:21 +0000 From: NeilBrown To: Chuck Lever , Jeff Layton Cc: linux-nfs@vger.kernel.org Subject: [PATCH 02/10] llist: add interface to check if a node is on a list. Date: Wed, 30 Aug 2023 12:54:45 +1000 Message-ID: <20230830025755.21292-3-neilb@suse.de> X-Mailer: git-send-email 2.41.0 In-Reply-To: <20230830025755.21292-1-neilb@suse.de> References: <20230830025755.21292-1-neilb@suse.de> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-nfs@vger.kernel.org With list.h lists, it is easy to test if a node is on a list, providing it was initialised and that it is removed with list_del_init(). This patch provides similar functionality for llist.h lists. init_llist_node() marks a node as being not-on-any-list be setting the ->next pointer to the node itself. llist_on_list() tests if the node is on any list. llist_del_first_init() remove the first element from a llist, and marks it as being off-list. Signed-off-by: NeilBrown --- include/linux/llist.h | 42 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 42 insertions(+) diff --git a/include/linux/llist.h b/include/linux/llist.h index 85bda2d02d65..dcb91e3bac1c 100644 --- a/include/linux/llist.h +++ b/include/linux/llist.h @@ -73,6 +73,33 @@ static inline void init_llist_head(struct llist_head *list) list->first = NULL; } +/** + * init_llist_node - initialize lock-less list node + * @node: the node to be initialised + * + * In cases where there is a need to test if a node is on + * a list or not, this initialises the node to clearly + * not be on any list. + */ +static inline void init_llist_node(struct llist_node *node) +{ + node->next = node; +} + +/** + * llist_on_list - test if a lock-list list node is on a list + * @node: the node to test + * + * When a node is on a list the ->next pointer will be NULL or + * some other node. It can never point to itself. We use that + * in init_llist_node() to record that a node is not on any list, + * and here to test whether it is on any list. + */ +static inline bool llist_on_list(const struct llist_node *node) +{ + return node->next != node; +} + /** * llist_entry - get the struct of this entry * @ptr: the &struct llist_node pointer. @@ -249,6 +276,21 @@ static inline struct llist_node *__llist_del_all(struct llist_head *head) extern struct llist_node *llist_del_first(struct llist_head *head); +/** + * llist_del_first_init - delete first entry from lock-list and mark is as being off-list + * @head: the head of lock-less list to delete from. + * + * This behave the same as llist_del_first() except that llist_init_node() is called + * on the returned node so that llist_on_list() will report false for the node. + */ +static inline struct llist_node *llist_del_first_init(struct llist_head *head) +{ + struct llist_node *n = llist_del_first(head); + + if (n) + init_llist_node(n); + return n; +} struct llist_node *llist_reverse_order(struct llist_node *head); #endif /* LLIST_H */ From patchwork Wed Aug 30 02:54:46 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: NeilBrown X-Patchwork-Id: 13369923 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 0F92FC83F16 for ; Wed, 30 Aug 2023 02:59:20 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S239380AbjH3C6u (ORCPT ); Tue, 29 Aug 2023 22:58:50 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:51492 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S241790AbjH3C6c (ORCPT ); Tue, 29 Aug 2023 22:58:32 -0400 Received: from smtp-out1.suse.de (smtp-out1.suse.de [IPv6:2001:67c:2178:6::1c]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 9B7E0193 for ; Tue, 29 Aug 2023 19:58:29 -0700 (PDT) Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by smtp-out1.suse.de (Postfix) with ESMTPS id 07FB0211E2; Wed, 30 Aug 2023 02:58:28 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1693364308; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=vYpBzJ2tgW0L8l/P6jQ0OFleSS3iL0SFlQSCIuezm6Y=; b=svCsPu52P7bVnBSL2VrTSJJ5EvJMwLXtM2X1vdBLDEm2LqvHwUFnNkqAv2RKjp8Mh040E5 U9frutU0yHUuRLymqoI26+EqeH+aq1mP10yQNNB7cD7jMB1ri3XrCDkhGgMLzPL565iCRr rm3QQTcebrjKHc2BcCtTeioNSnfOsho= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1693364308; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=vYpBzJ2tgW0L8l/P6jQ0OFleSS3iL0SFlQSCIuezm6Y=; b=wDfkscRZLdzetFV0Db9DOTL2+PbLnZMRnvk49KEmqEPnsn3/CNaCdU4LPXpLn8zW0qf2Ek yzLAbw4vGNkHKiAA== Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by imap2.suse-dmz.suse.de (Postfix) with ESMTPS id BC36C13301; Wed, 30 Aug 2023 02:58:26 +0000 (UTC) Received: from dovecot-director2.suse.de ([192.168.254.65]) by imap2.suse-dmz.suse.de with ESMTPSA id q5+CG1Kw7mTvYwAAMHmgww (envelope-from ); Wed, 30 Aug 2023 02:58:26 +0000 From: NeilBrown To: Chuck Lever , Jeff Layton Cc: linux-nfs@vger.kernel.org Subject: [PATCH 03/10] SQUASH use new llist interfaces in SUNRPC: change service idle list to be an llist Date: Wed, 30 Aug 2023 12:54:46 +1000 Message-ID: <20230830025755.21292-4-neilb@suse.de> X-Mailer: git-send-email 2.41.0 In-Reply-To: <20230830025755.21292-1-neilb@suse.de> References: <20230830025755.21292-1-neilb@suse.de> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-nfs@vger.kernel.org Use init_llist_node, llist_on_list etc for checking if a node is on a llist. Discard svc_thread_set_busy() completely and simplify svc_thread_busy() This can only be squashed if the patch to llist is moved earlier in the topic. Signed-off-by: NeilBrown --- include/linux/sunrpc/svc.h | 17 ++--------------- net/sunrpc/svc.c | 5 ++--- 2 files changed, 4 insertions(+), 18 deletions(-) diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index ed20a2ea1f81..ad4572630335 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -266,28 +266,15 @@ enum { RQ_DATA, /* request has data */ }; -/** - * svc_thread_set_busy - mark a thread as busy - * @rqstp: the thread which is now busy - * - * By convention a thread is busy if rq_idle.next points to rq_idle. - * This will never be the case for threads on the idle list. - */ -static inline void svc_thread_set_busy(struct svc_rqst *rqstp) -{ - rqstp->rq_idle.next = &rqstp->rq_idle; -} - /** * svc_thread_busy - check if a thread as busy * @rqstp: the thread which might be busy * - * By convention a thread is busy if rq_idle.next points to rq_idle. - * This will never be the case for threads on the idle list. + * A thread is only busy when it is not an the idle list. */ static inline bool svc_thread_busy(const struct svc_rqst *rqstp) { - return rqstp->rq_idle.next == &rqstp->rq_idle; + return !llist_on_list(&rqstp->rq_idle); } #define SVC_NET(rqst) (rqst->rq_xprt ? rqst->rq_xprt->xpt_net : rqst->rq_bc_net) diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index addbf28ea50a..5673f30db295 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -642,7 +642,7 @@ svc_rqst_alloc(struct svc_serv *serv, struct svc_pool *pool, int node) folio_batch_init(&rqstp->rq_fbatch); - svc_thread_set_busy(rqstp); + init_llist_node(&rqstp->rq_idle); rqstp->rq_server = serv; rqstp->rq_pool = pool; @@ -705,11 +705,10 @@ void svc_pool_wake_idle_thread(struct svc_pool *pool) rcu_read_lock(); spin_lock_bh(&pool->sp_lock); - ln = llist_del_first(&pool->sp_idle_threads); + ln = llist_del_first_init(&pool->sp_idle_threads); spin_unlock_bh(&pool->sp_lock); if (ln) { rqstp = llist_entry(ln, struct svc_rqst, rq_idle); - svc_thread_set_busy(rqstp); WRITE_ONCE(rqstp->rq_qtime, ktime_get()); wake_up_process(rqstp->rq_task); From patchwork Wed Aug 30 02:54:47 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: NeilBrown X-Patchwork-Id: 13369924 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 32582C83F1A for ; Wed, 30 Aug 2023 02:59:20 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S240275AbjH3C6v (ORCPT ); Tue, 29 Aug 2023 22:58:51 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:51528 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S241795AbjH3C6h (ORCPT ); Tue, 29 Aug 2023 22:58:37 -0400 Received: from smtp-out1.suse.de (smtp-out1.suse.de [IPv6:2001:67c:2178:6::1c]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 37ED5193 for ; Tue, 29 Aug 2023 19:58:34 -0700 (PDT) Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by smtp-out1.suse.de (Postfix) with ESMTPS id EF21F211E2; Wed, 30 Aug 2023 02:58:32 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1693364312; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=2n1lVgi8MIilrPJpoy6Sl1Kb9nOLArpLGPX+tGvwaVA=; b=gM2S22vJMMgTHXGwij/WL9up04z4il/Vj0gD/+GEH9R5X0Z+b+7Mlsm0EvV9XM0Pf5T9CK 9sipZZE1go1M69F/2D/gxbiOEdBAtvlg12cMFXmMnv/BSXDV4WNcIjw9e4YDLFRLkBmBCD dYZk6DB2SZ2w01ScWpFZdb1sSPkP0l0= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1693364312; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=2n1lVgi8MIilrPJpoy6Sl1Kb9nOLArpLGPX+tGvwaVA=; b=Upo7kO5pgngzIYHpRuxy2eYOkXH8FLroMXZJBsmrhsJ6E1maOfro9ZSj/DvUAG3AZtmltl HE9btvV3x2g66PBw== Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by imap2.suse-dmz.suse.de (Postfix) with ESMTPS id AFF2413301; Wed, 30 Aug 2023 02:58:31 +0000 (UTC) Received: from dovecot-director2.suse.de ([192.168.254.65]) by imap2.suse-dmz.suse.de with ESMTPSA id NxLCGFew7mT2YwAAMHmgww (envelope-from ); Wed, 30 Aug 2023 02:58:31 +0000 From: NeilBrown To: Chuck Lever , Jeff Layton Cc: linux-nfs@vger.kernel.org Subject: [PATCH 04/10] llist: add llist_del_first_this() Date: Wed, 30 Aug 2023 12:54:47 +1000 Message-ID: <20230830025755.21292-5-neilb@suse.de> X-Mailer: git-send-email 2.41.0 In-Reply-To: <20230830025755.21292-1-neilb@suse.de> References: <20230830025755.21292-1-neilb@suse.de> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-nfs@vger.kernel.org llist_del_first_this() deletes a specific entry from an llist, providing it is at the head of the list. Multiple threads can call this concurrently providing they each offer a different entry. This can be uses for a set of worker threads which are on the llist when they are idle. The head can always be woken, and when it is woken it can remove itself, and possibly wake the next if there is an excess of work to do. Signed-off-by: NeilBrown --- include/linux/llist.h | 4 ++++ lib/llist.c | 28 ++++++++++++++++++++++++++++ 2 files changed, 32 insertions(+) diff --git a/include/linux/llist.h b/include/linux/llist.h index dcb91e3bac1c..2c982ff7475a 100644 --- a/include/linux/llist.h +++ b/include/linux/llist.h @@ -291,6 +291,10 @@ static inline struct llist_node *llist_del_first_init(struct llist_head *head) init_llist_node(n); return n; } + +extern bool llist_del_first_this(struct llist_head *head, + struct llist_node *this); + struct llist_node *llist_reverse_order(struct llist_node *head); #endif /* LLIST_H */ diff --git a/lib/llist.c b/lib/llist.c index 6e668fa5a2c6..f21d0cfbbaaa 100644 --- a/lib/llist.c +++ b/lib/llist.c @@ -65,6 +65,34 @@ struct llist_node *llist_del_first(struct llist_head *head) } EXPORT_SYMBOL_GPL(llist_del_first); +/** + * llist_del_first_this - delete given entry of lock-less list if it is first + * @head: the head for your lock-less list + * @this: a list entry. + * + * If head of the list is given entry, delete and return %true else + * return %false. + * + * Multiple callers can safely call this concurrently with multiple + * llist_add() callers, providing all the callers offer a different @this. + */ +bool llist_del_first_this(struct llist_head *head, + struct llist_node *this) +{ + struct llist_node *entry, *next; + + /* acquire ensures orderig wrt try_cmpxchg() is llist_del_first() */ + entry = smp_load_acquire(&head->first); + do { + if (entry != this) + return false; + next = READ_ONCE(entry->next); + } while (!try_cmpxchg(&head->first, &entry, next)); + + return true; +} +EXPORT_SYMBOL_GPL(llist_del_first_this); + /** * llist_reverse_order - reverse order of a llist chain * @head: first item of the list to be reversed From patchwork Wed Aug 30 02:54:48 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: NeilBrown X-Patchwork-Id: 13369926 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 21D7AC83F18 for ; Wed, 30 Aug 2023 02:59:20 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S241795AbjH3C6w (ORCPT ); Tue, 29 Aug 2023 22:58:52 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:48058 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S241800AbjH3C6m (ORCPT ); Tue, 29 Aug 2023 22:58:42 -0400 Received: from smtp-out1.suse.de (smtp-out1.suse.de [IPv6:2001:67c:2178:6::1c]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 29F22CC9 for ; Tue, 29 Aug 2023 19:58:39 -0700 (PDT) Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by smtp-out1.suse.de (Postfix) with ESMTPS id E0A28211E2; Wed, 30 Aug 2023 02:58:37 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1693364317; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=zrpGcNIXY/c9eUBbtJHrWJaLj28atrQ28uD8aoEE9Wo=; b=FJs6yieYEbVfO5jkm5qL8FPMvqZNABcXYRzxDP1rf+9AEuNuMehF4fsCTxJe4yc7K5J6V3 k6EO8Co2QFZCBBiDxPI9ecuMpBwo/ddsKroK+EtcSCZypaRLoiBISTOrQr6M2MQEosvpu0 ABwqXMTocv44PULYPQxOtCt8+SfsYdg= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1693364317; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=zrpGcNIXY/c9eUBbtJHrWJaLj28atrQ28uD8aoEE9Wo=; b=JSOrCS/n+1mDd+gwRYoH+2N/4Lt4885Rahz7QJ+XWD7sIDscbwj+0RAelP5NAbeVwkxR86 yMEF0tgCkTbcinBw== Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by imap2.suse-dmz.suse.de (Postfix) with ESMTPS id A3C4913301; Wed, 30 Aug 2023 02:58:36 +0000 (UTC) Received: from dovecot-director2.suse.de ([192.168.254.65]) by imap2.suse-dmz.suse.de with ESMTPSA id e6zKFVyw7mQBZAAAMHmgww (envelope-from ); Wed, 30 Aug 2023 02:58:36 +0000 From: NeilBrown To: Chuck Lever , Jeff Layton Cc: linux-nfs@vger.kernel.org Subject: [PATCH 05/10] lib: add light-weight queuing mechanism. Date: Wed, 30 Aug 2023 12:54:48 +1000 Message-ID: <20230830025755.21292-6-neilb@suse.de> X-Mailer: git-send-email 2.41.0 In-Reply-To: <20230830025755.21292-1-neilb@suse.de> References: <20230830025755.21292-1-neilb@suse.de> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-nfs@vger.kernel.org lwq is a FIFO single-linked queue that only requires a spinlock for dequeueing, which happens in process context. Enqueueing is atomic with no spinlock and can happen in any context. Include a unit test for basic functionality - runs at boot time. Does not use kunit framework. Signed-off-by: NeilBrown --- include/linux/lwq.h | 120 +++++++++++++++++++++++++++++++++++ lib/Kconfig | 5 ++ lib/Makefile | 2 +- lib/lwq.c | 149 ++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 275 insertions(+), 1 deletion(-) create mode 100644 include/linux/lwq.h create mode 100644 lib/lwq.c diff --git a/include/linux/lwq.h b/include/linux/lwq.h new file mode 100644 index 000000000000..52b9c81b493a --- /dev/null +++ b/include/linux/lwq.h @@ -0,0 +1,120 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ + +#ifndef LWQ_H +#define LWQ_H +/* + * light-weight single-linked queue built from llist + * + * Entries can be enqueued from any context with no locking. + * Entries can be dequeued from process context with integrated locking. + */ +#include +#include +#include + +struct lwq_node { + struct llist_node node; +}; + +struct lwq { + spinlock_t lock; + struct llist_node *ready; /* entries to be dequeued */ + struct llist_head new; /* entries being enqueued */ +}; + +/** + * lwq_init - initialise a lwq + * @q: the lwq object + */ +static inline void lwq_init(struct lwq *q) +{ + spin_lock_init(&q->lock); + q->ready = NULL; + init_llist_head(&q->new); +} + +/** + * lwq_empty - test if lwq contains any entry + * @q: the lwq object + * + * This empty test contains an acquire barrier so that if a wakeup + * is sent when lwq_dequeue returns true, it is safe to go to sleep after + * a test on lwq_empty(). + */ +static inline bool lwq_empty(struct lwq *q) +{ + /* acquire ensures ordering wrt lwq_enqueue() */ + return smp_load_acquire(&q->ready) == NULL && llist_empty(&q->new); +} + +struct llist_node *__lwq_dequeue(struct lwq *q); +/** + * lwq_dequeue - dequeue first (oldest) entry from lwq + * @q: the queue to dequeue from + * @type: the type of object to return + * @member: them member in returned object which is an lwq_node. + * + * Remove a single object from the lwq and return it. This will take + * a spinlock and so must always be called in the same context, typcially + * process contet. + */ +#define lwq_dequeue(q, type, member) \ + ({ struct llist_node *_n = __lwq_dequeue(q); \ + _n ? container_of(_n, type, member.node) : NULL; }) + +struct llist_node *lwq_dequeue_all(struct lwq *q); + +/** + * lwq_for_each_safe - iterate over detached queue allowing deletion + * @_n: iterator variable + * @_t1: temporary struct llist_node ** + * @_t2: temporary struct llist_node * + * @_l: address of llist_node pointer from lwq_dequeue_all() + * @_member: member in _n where lwq_node is found. + * + * Iterate over members in a dequeued list. If the iterator variable + * is set to NULL, the iterator removes that entry from the queue. + */ +#define lwq_for_each_safe(_n, _t1, _t2, _l, _member) \ + for (_t1 = (_l); \ + *(_t1) ? (_n = container_of(*(_t1), typeof(*(_n)), _member.node),\ + _t2 = ((*_t1)->next), \ + true) \ + : false; \ + (_n) ? (_t1 = &(_n)->_member.node.next, 0) \ + : ((*(_t1) = (_t2)), 0)) + +/** + * lwq_enqueue - add a new item to the end of the queue + * @n - the lwq_node embedded in the item to be added + * @q - the lwq to append to. + * + * No locking is needed to append to the queue so this can + * be called from any context. + * Return %true is the list may have previously been empty. + */ +static inline bool lwq_enqueue(struct lwq_node *n, struct lwq *q) +{ + /* acquire enqures ordering wrt lwq_dequeue */ + return llist_add(&n->node, &q->new) && + smp_load_acquire(&q->ready) == NULL; +} + +/** + * lwq_enqueue_batch - add a list of new items to the end of the queue + * @n - the lwq_node embedded in the first item to be added + * @q - the lwq to append to. + * + * No locking is needed to append to the queue so this can + * be called from any context. + * Return %true is the list may have previously been empty. + */ +static inline bool lwq_enqueue_batch(struct llist_node *n, struct lwq *q) +{ + struct llist_node *e = n; + + /* acquire enqures ordering wrt lwq_dequeue */ + return llist_add_batch(llist_reverse_order(n), e, &q->new) && + smp_load_acquire(&q->ready) == NULL; +} +#endif /* LWQ_H */ diff --git a/lib/Kconfig b/lib/Kconfig index 5c2da561c516..6620bdba4f94 100644 --- a/lib/Kconfig +++ b/lib/Kconfig @@ -763,3 +763,8 @@ config ASN1_ENCODER config POLYNOMIAL tristate + +config LWQ_TEST + bool "RPC: enable boot-time test for lwq queuing" + help + Enable boot-time test of lwq functionality. diff --git a/lib/Makefile b/lib/Makefile index 1ffae65bb7ee..4b67c2d6af62 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -45,7 +45,7 @@ obj-y += lockref.o obj-y += bcd.o sort.o parser.o debug_locks.o random32.o \ bust_spinlocks.o kasprintf.o bitmap.o scatterlist.o \ list_sort.o uuid.o iov_iter.o clz_ctz.o \ - bsearch.o find_bit.o llist.o memweight.o kfifo.o \ + bsearch.o find_bit.o llist.o lwq.o memweight.o kfifo.o \ percpu-refcount.o rhashtable.o base64.o \ once.o refcount.o rcuref.o usercopy.o errseq.o bucket_locks.o \ generic-radix-tree.o diff --git a/lib/lwq.c b/lib/lwq.c new file mode 100644 index 000000000000..d6be6dda3867 --- /dev/null +++ b/lib/lwq.c @@ -0,0 +1,149 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Light weight single-linked queue. + * + * Entries are enqueued to the head of an llist, with no blocking. + * This can happen in any context. + * + * Entries are dequeued using a spinlock to protect against + * multiple access. The llist is staged in reverse order, and refreshed + * from the llist when it exhausts. + */ +#include +#include + +struct llist_node *__lwq_dequeue(struct lwq *q) +{ + struct llist_node *this; + + if (lwq_empty(q)) + return NULL; + spin_lock(&q->lock); + this = q->ready; + if (!this && !llist_empty(&q->new)) { + /* ensure queue doesn't appear transiently lwq_empty */ + smp_store_release(&q->ready, (void *)1); + this = llist_reverse_order(llist_del_all(&q->new)); + if (!this) + q->ready = NULL; + } + if (this) + q->ready = llist_next(this); + spin_unlock(&q->lock); + return this; +} + +/** + * lwq_dequeue_all - dequeue all currently enqueued objects + * @q: the queue to dequeue from + * + * Remove and return a linked list of llist_nodes of all the objects that were + * in the queue. The first on the list will be the object that was least + * recently enqueued. + */ +struct llist_node *lwq_dequeue_all(struct lwq *q) +{ + struct llist_node *r, *t, **ep; + + if (lwq_empty(q)) + return NULL; + + spin_lock(&q->lock); + r = q->ready; + q->ready = NULL; + t = llist_del_all(&q->new); + spin_unlock(&q->lock); + ep = &r; + while (*ep) + ep = &(*ep)->next; + *ep = llist_reverse_order(t); + return r; +} + +#if IS_ENABLED(CONFIG_LWQ_TEST) + +#include +#include +#include +#include +#include +struct tnode { + struct lwq_node n; + int i; + int c; +}; + +static int lwq_exercise(void *qv) +{ + struct lwq *q = qv; + int cnt; + struct tnode *t; + + for (cnt = 0; cnt < 10000; cnt++) { + wait_var_event(q, (t = lwq_dequeue(q, struct tnode, n)) != NULL); + t->c++; + if (lwq_enqueue(&t->n, q)) + wake_up_var(q); + } + while (!kthread_should_stop()) + schedule_timeout_idle(1); + return 0; +} + +static int lwq_test(void) +{ + int i; + struct lwq q; + struct llist_node *l, **t1, *t2; + struct tnode *t; + struct task_struct *threads[8]; + + printk(KERN_INFO "testing lwq....\n"); + lwq_init(&q); + printk(KERN_INFO " lwq: run some threads\n"); + for (i = 0; i < ARRAY_SIZE(threads); i++) + threads[i] = kthread_run(lwq_exercise, &q, "lwq-test-%d", i); + for (i = 0; i < 100; i++) { + t = kmalloc(sizeof(*t), GFP_KERNEL); + t->i = i; + t->c = 0; + if (lwq_enqueue(&t->n, &q)) + wake_up_var(&q); + }; + /* wait for threads to exit */ + for (i = 0; i < ARRAY_SIZE(threads); i++) + if (!IS_ERR_OR_NULL(threads[i])) + kthread_stop(threads[i]); + printk(KERN_INFO " lwq: dequeue first 50:"); + for (i = 0; i < 50 ; i++) { + if (i && (i % 10) == 0) { + printk(KERN_CONT "\n"); + printk(KERN_INFO " lwq: ... "); + } + t = lwq_dequeue(&q, struct tnode, n); + printk(KERN_CONT " %d(%d)", t->i, t->c); + kfree(t); + } + printk(KERN_CONT "\n"); + l = lwq_dequeue_all(&q); + printk(KERN_INFO " lwq: delete the multiples of 3 (test lwq_for_each_safe())\n"); + lwq_for_each_safe(t, t1, t2, &l, n) { + if ((t->i % 3) == 0) { + t->i = -1; + kfree(t); + t = NULL; + } + } + if (l) + lwq_enqueue_batch(l, &q); + printk(KERN_INFO " lwq: dequeue remaining:"); + while ((t = lwq_dequeue(&q, struct tnode, n)) != NULL) { + printk(KERN_CONT " %d", t->i); + kfree(t); + } + printk(KERN_CONT "\n"); + return 0; +} + +module_init(lwq_test); +#endif /* CONFIG_LWQ_TEST*/ From patchwork Wed Aug 30 02:54:49 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: NeilBrown X-Patchwork-Id: 13369925 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 46809C83F19 for ; Wed, 30 Aug 2023 02:59:20 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S241796AbjH3C6x (ORCPT ); Tue, 29 Aug 2023 22:58:53 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:48068 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S241803AbjH3C6q (ORCPT ); Tue, 29 Aug 2023 22:58:46 -0400 Received: from smtp-out2.suse.de (smtp-out2.suse.de [195.135.220.29]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 33C89185 for ; Tue, 29 Aug 2023 19:58:44 -0700 (PDT) Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by smtp-out2.suse.de (Postfix) with ESMTPS id EB08F1F45F; Wed, 30 Aug 2023 02:58:42 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1693364322; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=C9BZBFi3a/8MiWUA1dYlaushegUfXjp4xF87chH87zE=; b=doD1el3brmioiH6s0S+vsos0Iu+I7RyLPuzMTY46t5qwvSUTXu9eGhFthpKjX8Yw9F2ZUc bS4yvq4FZzXniBsXm68xEON+JJz+klqZSujLDpNy+6+o6rMwkF/j0UnyfvKHVJBWauWWfR ScVo6cNpKcKUS6TCoqWTD+pzhoISEI0= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1693364322; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=C9BZBFi3a/8MiWUA1dYlaushegUfXjp4xF87chH87zE=; b=nhKz3BAjoEDCoxw+oU0RI95m7kXbDOxsswyR35+f8kHQ/aEq7O4j1RuQFcy1ZwpEbllFfH D5ssVE+vrpUfmgCQ== Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by imap2.suse-dmz.suse.de (Postfix) with ESMTPS id A9DF013301; Wed, 30 Aug 2023 02:58:41 +0000 (UTC) Received: from dovecot-director2.suse.de ([192.168.254.65]) by imap2.suse-dmz.suse.de with ESMTPSA id ZDQOF2Gw7mQHZAAAMHmgww (envelope-from ); Wed, 30 Aug 2023 02:58:41 +0000 From: NeilBrown To: Chuck Lever , Jeff Layton Cc: linux-nfs@vger.kernel.org Subject: [PATCH 06/10] SUNRPC: only have one thread waking up at a time Date: Wed, 30 Aug 2023 12:54:49 +1000 Message-ID: <20230830025755.21292-7-neilb@suse.de> X-Mailer: git-send-email 2.41.0 In-Reply-To: <20230830025755.21292-1-neilb@suse.de> References: <20230830025755.21292-1-neilb@suse.de> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-nfs@vger.kernel.org Currently if several items of work become available in quick succession, that number of threads (if available) will be woken. By the time some of them wake up another thread that was already cache-warm might have come along and completed the work. Anecdotal evidence suggests as many as 15% of wakes find nothing to do once they get to the point of looking. This patch changes svc_pool_wake_idle_thread() to wake the first thread on the queue but NOT remove it. Subsequent calls will wake the same thread. Once that thread starts it will dequeue itself and after dequeueing some work to do, it will wake the next thread if there is more work ready. This results in a more orderly increase in the number of busy threads. As a bonus, this allows us to reduce locking around the idle queue. svc_pool_wake_idle_thread() no longer needs to take a lock (beyond rcu_read_lock()) as it doesn't manipulate the queue, it just looks at the first item. The thread itself can avoid locking by using the new llist_del_first_this() interface. This will safely remove the thread itself if it is the head. If it isn't the head, it will do nothing. If multiple threads call this concurrently only one will succeed. The others will do nothing, so no corruption can result. If a thread wakes up and finds that it cannot dequeue itself that means either - that it wasn't woken because it was the head of the queue. Maybe the freezer woke it. In that case it can go back to sleep (after trying to freeze of course). - some other thread found there was nothing to do very recently, and placed itself on the head of the queue in front of this thread. It must check again after placing itself there, so it can be deemed to be responsible for any pending work, and this thread can go back to sleep until woken. No code ever tests for busy threads any more. Only each thread itself cares if it is busy. So svc_thread_busy() is no longer needed. Signed-off-by: NeilBrown --- include/linux/sunrpc/svc.h | 11 ----------- net/sunrpc/svc.c | 14 ++++++-------- net/sunrpc/svc_xprt.c | 35 ++++++++++++++++++++++------------- 3 files changed, 28 insertions(+), 32 deletions(-) diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index ad4572630335..dafa362b4fdd 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -266,17 +266,6 @@ enum { RQ_DATA, /* request has data */ }; -/** - * svc_thread_busy - check if a thread as busy - * @rqstp: the thread which might be busy - * - * A thread is only busy when it is not an the idle list. - */ -static inline bool svc_thread_busy(const struct svc_rqst *rqstp) -{ - return !llist_on_list(&rqstp->rq_idle); -} - #define SVC_NET(rqst) (rqst->rq_xprt ? rqst->rq_xprt->xpt_net : rqst->rq_bc_net) /* diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 5673f30db295..3267d740235e 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -642,7 +642,6 @@ svc_rqst_alloc(struct svc_serv *serv, struct svc_pool *pool, int node) folio_batch_init(&rqstp->rq_fbatch); - init_llist_node(&rqstp->rq_idle); rqstp->rq_server = serv; rqstp->rq_pool = pool; @@ -704,17 +703,16 @@ void svc_pool_wake_idle_thread(struct svc_pool *pool) struct llist_node *ln; rcu_read_lock(); - spin_lock_bh(&pool->sp_lock); - ln = llist_del_first_init(&pool->sp_idle_threads); - spin_unlock_bh(&pool->sp_lock); + ln = READ_ONCE(pool->sp_idle_threads.first); if (ln) { rqstp = llist_entry(ln, struct svc_rqst, rq_idle); - WRITE_ONCE(rqstp->rq_qtime, ktime_get()); - wake_up_process(rqstp->rq_task); + if (!task_is_running(rqstp->rq_task)) { + wake_up_process(rqstp->rq_task); + trace_svc_wake_up(rqstp->rq_task->pid); + percpu_counter_inc(&pool->sp_threads_woken); + } rcu_read_unlock(); - percpu_counter_inc(&pool->sp_threads_woken); - trace_svc_wake_up(rqstp->rq_task->pid); return; } rcu_read_unlock(); diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 17c43bde35c9..a51570a4cbf2 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -732,20 +732,16 @@ static void svc_rqst_wait_for_work(struct svc_rqst *rqstp) if (rqst_should_sleep(rqstp)) { set_current_state(TASK_IDLE | TASK_FREEZABLE); llist_add(&rqstp->rq_idle, &pool->sp_idle_threads); + if (likely(rqst_should_sleep(rqstp))) + schedule(); - if (unlikely(!rqst_should_sleep(rqstp))) - /* Work just became available. This thread cannot simply - * choose not to sleep as it *must* wait until removed. - * So wake the first waiter - whether it is this - * thread or some other, it will get the work done. + while (!llist_del_first_this(&pool->sp_idle_threads, + &rqstp->rq_idle)) { + /* Cannot @rqstp from idle list, so some other thread + * must have queued itself after finding + * no work to do, so they have taken responsibility + * for any outstanding work. */ - svc_pool_wake_idle_thread(pool); - - /* Since a thread cannot remove itself from an llist, - * schedule until someone else removes @rqstp from - * the idle list. - */ - while (!svc_thread_busy(rqstp)) { schedule(); set_current_state(TASK_IDLE | TASK_FREEZABLE); } @@ -835,6 +831,15 @@ static void svc_handle_xprt(struct svc_rqst *rqstp, struct svc_xprt *xprt) svc_xprt_release(rqstp); } +static void wake_next(struct svc_rqst *rqstp) +{ + if (!rqst_should_sleep(rqstp)) + /* More work pending after I dequeued some, + * wake another worker + */ + svc_pool_wake_idle_thread(rqstp->rq_pool); +} + /** * svc_recv - Receive and process the next request on any transport * @rqstp: an idle RPC service thread @@ -854,13 +859,16 @@ void svc_recv(struct svc_rqst *rqstp) clear_bit(SP_TASK_PENDING, &pool->sp_flags); - if (svc_thread_should_stop(rqstp)) + if (svc_thread_should_stop(rqstp)) { + wake_next(rqstp); return; + } rqstp->rq_xprt = svc_xprt_dequeue(pool); if (rqstp->rq_xprt) { struct svc_xprt *xprt = rqstp->rq_xprt; + wake_next(rqstp); /* Normally we will wait up to 5 seconds for any required * cache information to be provided. When there are no * idle threads, we reduce the wait time. @@ -885,6 +893,7 @@ void svc_recv(struct svc_rqst *rqstp) if (req) { list_del(&req->rq_bc_list); spin_unlock_bh(&serv->sv_cb_lock); + wake_next(rqstp); svc_process_bc(req, rqstp); return; From patchwork Wed Aug 30 02:54:50 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: NeilBrown X-Patchwork-Id: 13369928 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 23591C83F12 for ; Wed, 30 Aug 2023 02:59:53 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S241790AbjH3C7V (ORCPT ); Tue, 29 Aug 2023 22:59:21 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:42376 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S241794AbjH3C6w (ORCPT ); Tue, 29 Aug 2023 22:58:52 -0400 Received: from smtp-out2.suse.de (smtp-out2.suse.de [195.135.220.29]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 326E1185 for ; Tue, 29 Aug 2023 19:58:49 -0700 (PDT) Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by smtp-out2.suse.de (Postfix) with ESMTPS id E55011F45F; Wed, 30 Aug 2023 02:58:47 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1693364327; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=yxAuqOWoWdISxUI+nr2fJZf2VC7P+7rSLtp5JL74Rns=; b=ULD0qaHKa9rkyKH+QNDoxFCH3mnjF7UJguDDdYQydAHZlr0E3nFsKapqlx/HUn1gRItr2S tvO3bcr1YxFyH6DHk1eYWjn589/aTwZbDiHx2BTzN6862Y8vFyA9RLbL2aceZXkuWmmEHe aRowEqiWUlOVi7cp0XrvyhssmYWUZZs= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1693364327; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=yxAuqOWoWdISxUI+nr2fJZf2VC7P+7rSLtp5JL74Rns=; b=2YpTOaIUCi8PHBG7fqO/dknQmw3oaiAgi4PvMkigJwZj6uZ/sDUA5QOsByA2Xc4c6R3Cn4 xU0Yd9sPxbi0H1DQ== Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by imap2.suse-dmz.suse.de (Postfix) with ESMTPS id A5DB013301; Wed, 30 Aug 2023 02:58:46 +0000 (UTC) Received: from dovecot-director2.suse.de ([192.168.254.65]) by imap2.suse-dmz.suse.de with ESMTPSA id hK42Fmaw7mQRZAAAMHmgww (envelope-from ); Wed, 30 Aug 2023 02:58:46 +0000 From: NeilBrown To: Chuck Lever , Jeff Layton Cc: linux-nfs@vger.kernel.org Subject: [PATCH 07/10] SUNRPC: use lwq for sp_sockets - renamed to sp_xprts Date: Wed, 30 Aug 2023 12:54:50 +1000 Message-ID: <20230830025755.21292-8-neilb@suse.de> X-Mailer: git-send-email 2.41.0 In-Reply-To: <20230830025755.21292-1-neilb@suse.de> References: <20230830025755.21292-1-neilb@suse.de> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-nfs@vger.kernel.org lwq avoids using back pointers in lists, and uses less locking. This introduces a new spinlock, but the other one will be removed in a future patch. For svc_clean_up_xprts(), we now dequeue the entire queue, walk it to remove and process the xprts that need cleaning up, then re-enqueue the remaining queue. Signed-off-by: NeilBrown --- include/linux/sunrpc/svc.h | 3 +- include/linux/sunrpc/svc_xprt.h | 2 +- net/sunrpc/svc.c | 2 +- net/sunrpc/svc_xprt.c | 57 ++++++++++----------------------- 4 files changed, 21 insertions(+), 43 deletions(-) diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index dafa362b4fdd..7ff9fe785e49 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -17,6 +17,7 @@ #include #include #include +#include #include #include #include @@ -34,7 +35,7 @@ struct svc_pool { unsigned int sp_id; /* pool id; also node id on NUMA */ spinlock_t sp_lock; /* protects all fields */ - struct list_head sp_sockets; /* pending sockets */ + struct lwq sp_xprts; /* pending transports */ unsigned int sp_nrthreads; /* # of threads in pool */ struct list_head sp_all_threads; /* all server threads */ struct llist_head sp_idle_threads; /* idle server threads */ diff --git a/include/linux/sunrpc/svc_xprt.h b/include/linux/sunrpc/svc_xprt.h index fa55d12dc765..8e20cd60e2e7 100644 --- a/include/linux/sunrpc/svc_xprt.h +++ b/include/linux/sunrpc/svc_xprt.h @@ -54,7 +54,7 @@ struct svc_xprt { const struct svc_xprt_ops *xpt_ops; struct kref xpt_ref; struct list_head xpt_list; - struct list_head xpt_ready; + struct lwq_node xpt_ready; unsigned long xpt_flags; struct svc_serv *xpt_server; /* service for transport */ diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 3267d740235e..63cddb8cb08d 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -508,7 +508,7 @@ __svc_create(struct svc_program *prog, unsigned int bufsize, int npools, i, serv->sv_name); pool->sp_id = i; - INIT_LIST_HEAD(&pool->sp_sockets); + lwq_init(&pool->sp_xprts); INIT_LIST_HEAD(&pool->sp_all_threads); init_llist_head(&pool->sp_idle_threads); spin_lock_init(&pool->sp_lock); diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index a51570a4cbf2..2399811582cc 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -201,7 +201,6 @@ void svc_xprt_init(struct net *net, struct svc_xprt_class *xcl, kref_init(&xprt->xpt_ref); xprt->xpt_server = serv; INIT_LIST_HEAD(&xprt->xpt_list); - INIT_LIST_HEAD(&xprt->xpt_ready); INIT_LIST_HEAD(&xprt->xpt_deferred); INIT_LIST_HEAD(&xprt->xpt_users); mutex_init(&xprt->xpt_mutex); @@ -472,9 +471,7 @@ void svc_xprt_enqueue(struct svc_xprt *xprt) pool = svc_pool_for_cpu(xprt->xpt_server); percpu_counter_inc(&pool->sp_sockets_queued); - spin_lock_bh(&pool->sp_lock); - list_add_tail(&xprt->xpt_ready, &pool->sp_sockets); - spin_unlock_bh(&pool->sp_lock); + lwq_enqueue(&xprt->xpt_ready, &pool->sp_xprts); svc_pool_wake_idle_thread(pool); } @@ -487,18 +484,9 @@ static struct svc_xprt *svc_xprt_dequeue(struct svc_pool *pool) { struct svc_xprt *xprt = NULL; - if (list_empty(&pool->sp_sockets)) - goto out; - - spin_lock_bh(&pool->sp_lock); - if (likely(!list_empty(&pool->sp_sockets))) { - xprt = list_first_entry(&pool->sp_sockets, - struct svc_xprt, xpt_ready); - list_del_init(&xprt->xpt_ready); + xprt = lwq_dequeue(&pool->sp_xprts, struct svc_xprt, xpt_ready); + if (xprt) svc_xprt_get(xprt); - } - spin_unlock_bh(&pool->sp_lock); -out: return xprt; } @@ -708,7 +696,7 @@ rqst_should_sleep(struct svc_rqst *rqstp) return false; /* was a socket queued? */ - if (!list_empty(&pool->sp_sockets)) + if (!lwq_empty(&pool->sp_xprts)) return false; /* are we shutting down? */ @@ -1047,7 +1035,6 @@ static void svc_delete_xprt(struct svc_xprt *xprt) spin_lock_bh(&serv->sv_lock); list_del_init(&xprt->xpt_list); - WARN_ON_ONCE(!list_empty(&xprt->xpt_ready)); if (test_bit(XPT_TEMP, &xprt->xpt_flags)) serv->sv_tmpcnt--; spin_unlock_bh(&serv->sv_lock); @@ -1098,36 +1085,26 @@ static int svc_close_list(struct svc_serv *serv, struct list_head *xprt_list, st return ret; } -static struct svc_xprt *svc_dequeue_net(struct svc_serv *serv, struct net *net) +static void svc_clean_up_xprts(struct svc_serv *serv, struct net *net) { - struct svc_pool *pool; struct svc_xprt *xprt; - struct svc_xprt *tmp; int i; for (i = 0; i < serv->sv_nrpools; i++) { - pool = &serv->sv_pools[i]; - - spin_lock_bh(&pool->sp_lock); - list_for_each_entry_safe(xprt, tmp, &pool->sp_sockets, xpt_ready) { - if (xprt->xpt_net != net) - continue; - list_del_init(&xprt->xpt_ready); - spin_unlock_bh(&pool->sp_lock); - return xprt; + struct svc_pool *pool = &serv->sv_pools[i]; + struct llist_node *q, **t1, *t2; + + q = lwq_dequeue_all(&pool->sp_xprts); + lwq_for_each_safe(xprt, t1, t2, &q, xpt_ready) { + if (xprt->xpt_net == net) { + set_bit(XPT_CLOSE, &xprt->xpt_flags); + svc_delete_xprt(xprt); + xprt = NULL; + } } - spin_unlock_bh(&pool->sp_lock); - } - return NULL; -} -static void svc_clean_up_xprts(struct svc_serv *serv, struct net *net) -{ - struct svc_xprt *xprt; - - while ((xprt = svc_dequeue_net(serv, net))) { - set_bit(XPT_CLOSE, &xprt->xpt_flags); - svc_delete_xprt(xprt); + if (q) + lwq_enqueue_batch(q, &pool->sp_xprts); } } From patchwork Wed Aug 30 02:54:51 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: NeilBrown X-Patchwork-Id: 13369927 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id E82DBC6FA8F for ; Wed, 30 Aug 2023 02:59:52 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S241793AbjH3C7W (ORCPT ); Tue, 29 Aug 2023 22:59:22 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:42396 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S241799AbjH3C64 (ORCPT ); Tue, 29 Aug 2023 22:58:56 -0400 Received: from smtp-out1.suse.de (smtp-out1.suse.de [IPv6:2001:67c:2178:6::1c]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 23567193 for ; Tue, 29 Aug 2023 19:58:54 -0700 (PDT) Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by smtp-out1.suse.de (Postfix) with ESMTPS id DA559211E2; Wed, 30 Aug 2023 02:58:52 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1693364332; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=put9AXxrcv+eoyJ+SDrIafBWhaHUsLY7x9zPi5RlRNA=; b=yOQCMKKUV9RyGzP5q7mqD+xggwwkj2mcvny5pcjQbDWKUvn7ZP9gB0A+beagYdoxhh6s4L rJ5mAEfRmdrmzzJG+b0ZLj0d598ZFtMGM/HZJlsKpMr8eOxYgztP8+uYrRuUrU0NvR8bXr wcao79OS8XusrLv0saCPDzoyIqZLQW0= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1693364332; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=put9AXxrcv+eoyJ+SDrIafBWhaHUsLY7x9zPi5RlRNA=; b=YEGyXhh/WsKnr6jI6KWVYtLaQXklSAkkXZwL8MIYHmFVn6TKfmzJI6ZOYJ8CTU0yEHjL6W Mj9YxtmlhUYUZ/Ag== Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by imap2.suse-dmz.suse.de (Postfix) with ESMTPS id 9B9E013301; Wed, 30 Aug 2023 02:58:51 +0000 (UTC) Received: from dovecot-director2.suse.de ([192.168.254.65]) by imap2.suse-dmz.suse.de with ESMTPSA id k3qZE2uw7mQfZAAAMHmgww (envelope-from ); Wed, 30 Aug 2023 02:58:51 +0000 From: NeilBrown To: Chuck Lever , Jeff Layton Cc: linux-nfs@vger.kernel.org Subject: [PATCH 08/10] SUNRPC: change sp_nrthreads to atomic_t Date: Wed, 30 Aug 2023 12:54:51 +1000 Message-ID: <20230830025755.21292-9-neilb@suse.de> X-Mailer: git-send-email 2.41.0 In-Reply-To: <20230830025755.21292-1-neilb@suse.de> References: <20230830025755.21292-1-neilb@suse.de> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-nfs@vger.kernel.org Using an atomic_t avoids the need to take a spinlock (which can soon be removed). Choosing a thread to kill needs to be careful as we cannot set the "die now" bit atomically with the test on the count. Instead we temporarily increase the count. Signed-off-by: NeilBrown --- fs/nfsd/nfssvc.c | 3 ++- include/linux/sunrpc/svc.h | 2 +- net/sunrpc/svc.c | 37 ++++++++++++++++++++----------------- 3 files changed, 23 insertions(+), 19 deletions(-) diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 062f51fe4dfb..5e455ced0711 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -718,7 +718,8 @@ int nfsd_get_nrthreads(int n, int *nthreads, struct net *net) if (nn->nfsd_serv != NULL) { for (i = 0; i < nn->nfsd_serv->sv_nrpools && i < n; i++) - nthreads[i] = nn->nfsd_serv->sv_pools[i].sp_nrthreads; + nthreads[i] = + atomic_read(&nn->nfsd_serv->sv_pools[i].sp_nrthreads); } return 0; diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index 7ff9fe785e49..9d0fcd6148ae 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -36,7 +36,7 @@ struct svc_pool { unsigned int sp_id; /* pool id; also node id on NUMA */ spinlock_t sp_lock; /* protects all fields */ struct lwq sp_xprts; /* pending transports */ - unsigned int sp_nrthreads; /* # of threads in pool */ + atomic_t sp_nrthreads; /* # of threads in pool */ struct list_head sp_all_threads; /* all server threads */ struct llist_head sp_idle_threads; /* idle server threads */ diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 63cddb8cb08d..9524af33ace9 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -681,8 +681,8 @@ svc_prepare_thread(struct svc_serv *serv, struct svc_pool *pool, int node) serv->sv_nrthreads += 1; spin_unlock_bh(&serv->sv_lock); + atomic_inc(&pool->sp_nrthreads); spin_lock_bh(&pool->sp_lock); - pool->sp_nrthreads++; list_add_rcu(&rqstp->rq_all, &pool->sp_all_threads); spin_unlock_bh(&pool->sp_lock); return rqstp; @@ -727,23 +727,24 @@ svc_pool_next(struct svc_serv *serv, struct svc_pool *pool, unsigned int *state) } static struct svc_pool * -svc_pool_victim(struct svc_serv *serv, struct svc_pool *pool, unsigned int *state) +svc_pool_victim(struct svc_serv *serv, struct svc_pool *target_pool, + unsigned int *state) { + struct svc_pool *pool; unsigned int i; +retry: + pool = target_pool; + if (pool != NULL) { - spin_lock_bh(&pool->sp_lock); - if (pool->sp_nrthreads) + if (atomic_inc_not_zero(&pool->sp_nrthreads)) goto found_pool; - spin_unlock_bh(&pool->sp_lock); return NULL; } else { for (i = 0; i < serv->sv_nrpools; i++) { pool = &serv->sv_pools[--(*state) % serv->sv_nrpools]; - spin_lock_bh(&pool->sp_lock); - if (pool->sp_nrthreads) + if (atomic_inc_not_zero(&pool->sp_nrthreads)) goto found_pool; - spin_unlock_bh(&pool->sp_lock); } return NULL; } @@ -751,8 +752,12 @@ svc_pool_victim(struct svc_serv *serv, struct svc_pool *pool, unsigned int *stat found_pool: set_bit(SP_VICTIM_REMAINS, &pool->sp_flags); set_bit(SP_NEED_VICTIM, &pool->sp_flags); - spin_unlock_bh(&pool->sp_lock); - return pool; + if (!atomic_dec_and_test(&pool->sp_nrthreads)) + return pool; + /* Nothing left in this pool any more */ + clear_bit(SP_NEED_VICTIM, &pool->sp_flags); + clear_bit(SP_VICTIM_REMAINS, &pool->sp_flags); + goto retry; } static int @@ -828,13 +833,10 @@ svc_stop_kthreads(struct svc_serv *serv, struct svc_pool *pool, int nrservs) int svc_set_num_threads(struct svc_serv *serv, struct svc_pool *pool, int nrservs) { - if (pool == NULL) { + if (!pool) nrservs -= serv->sv_nrthreads; - } else { - spin_lock_bh(&pool->sp_lock); - nrservs -= pool->sp_nrthreads; - spin_unlock_bh(&pool->sp_lock); - } + else + nrservs -= atomic_read(&pool->sp_nrthreads); if (nrservs > 0) return svc_start_kthreads(serv, pool, nrservs); @@ -921,10 +923,11 @@ svc_exit_thread(struct svc_rqst *rqstp) struct svc_pool *pool = rqstp->rq_pool; spin_lock_bh(&pool->sp_lock); - pool->sp_nrthreads--; list_del_rcu(&rqstp->rq_all); spin_unlock_bh(&pool->sp_lock); + atomic_dec(&pool->sp_nrthreads); + spin_lock_bh(&serv->sv_lock); serv->sv_nrthreads -= 1; spin_unlock_bh(&serv->sv_lock); From patchwork Wed Aug 30 02:54:52 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: NeilBrown X-Patchwork-Id: 13369929 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 4437DC83F16 for ; Wed, 30 Aug 2023 02:59:53 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231800AbjH3C7W (ORCPT ); Tue, 29 Aug 2023 22:59:22 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:42402 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S241800AbjH3C7B (ORCPT ); Tue, 29 Aug 2023 22:59:01 -0400 Received: from smtp-out1.suse.de (smtp-out1.suse.de [IPv6:2001:67c:2178:6::1c]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 21DE3185 for ; Tue, 29 Aug 2023 19:58:59 -0700 (PDT) Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by smtp-out1.suse.de (Postfix) with ESMTPS id BDE21211E2; Wed, 30 Aug 2023 02:58:57 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1693364337; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=7u/9LoXbnHKQTbn+5IfntA6WpU7iCnvel9gZ/zvF8aY=; b=P35rDTIIMYHLQuJ6gZGLsbhL/GkSQLpMPRNkLpyOIYVDcXsAxGyKdSxXggKI1ArM+RwvYc iDczgsIIwyoFDTFWe+t7HdEEDHWiWIl3q50hMP9LvJeJbMLVRt0ZcZJrQM5gJKIwnui403 DksC9PqcGTmPbkCdL4O/s/mNqMFAm+Q= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1693364337; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=7u/9LoXbnHKQTbn+5IfntA6WpU7iCnvel9gZ/zvF8aY=; b=bMqh0yIUP1rcyGFrtL6IxZrB9kJEb91GcFEoJTBc6f0dQg+9OHapGm0DL14lqKqNhRs/ie h2KdyOtXTpTSQMCA== Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by imap2.suse-dmz.suse.de (Postfix) with ESMTPS id 81FAD13301; Wed, 30 Aug 2023 02:58:56 +0000 (UTC) Received: from dovecot-director2.suse.de ([192.168.254.65]) by imap2.suse-dmz.suse.de with ESMTPSA id +4KKDXCw7mQnZAAAMHmgww (envelope-from ); Wed, 30 Aug 2023 02:58:56 +0000 From: NeilBrown To: Chuck Lever , Jeff Layton Cc: linux-nfs@vger.kernel.org Subject: [PATCH 09/10] SUNRPC: discard sp_lock Date: Wed, 30 Aug 2023 12:54:52 +1000 Message-ID: <20230830025755.21292-10-neilb@suse.de> X-Mailer: git-send-email 2.41.0 In-Reply-To: <20230830025755.21292-1-neilb@suse.de> References: <20230830025755.21292-1-neilb@suse.de> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-nfs@vger.kernel.org sp_lock is now only used to protect sp_all_threads. This isn't needed as sp_all_threads is only manipulated through svc_set_num_threads(), which must be locked. Read-acccess only requires rcu_read_lock(). So no more locking is needed. Signed-off-by: NeilBrown --- include/linux/sunrpc/svc.h | 1 - net/sunrpc/svc.c | 10 +++++----- 2 files changed, 5 insertions(+), 6 deletions(-) diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index 9d0fcd6148ae..8ce1392c1a35 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -34,7 +34,6 @@ */ struct svc_pool { unsigned int sp_id; /* pool id; also node id on NUMA */ - spinlock_t sp_lock; /* protects all fields */ struct lwq sp_xprts; /* pending transports */ atomic_t sp_nrthreads; /* # of threads in pool */ struct list_head sp_all_threads; /* all server threads */ diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 9524af33ace9..61ea8ce7975f 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -511,7 +511,6 @@ __svc_create(struct svc_program *prog, unsigned int bufsize, int npools, lwq_init(&pool->sp_xprts); INIT_LIST_HEAD(&pool->sp_all_threads); init_llist_head(&pool->sp_idle_threads); - spin_lock_init(&pool->sp_lock); percpu_counter_init(&pool->sp_messages_arrived, 0, GFP_KERNEL); percpu_counter_init(&pool->sp_sockets_queued, 0, GFP_KERNEL); @@ -682,9 +681,12 @@ svc_prepare_thread(struct svc_serv *serv, struct svc_pool *pool, int node) spin_unlock_bh(&serv->sv_lock); atomic_inc(&pool->sp_nrthreads); - spin_lock_bh(&pool->sp_lock); + + /* Protected by whatever lock the service uses when calling + * svc_set_num_threads() + */ list_add_rcu(&rqstp->rq_all, &pool->sp_all_threads); - spin_unlock_bh(&pool->sp_lock); + return rqstp; } @@ -922,9 +924,7 @@ svc_exit_thread(struct svc_rqst *rqstp) struct svc_serv *serv = rqstp->rq_server; struct svc_pool *pool = rqstp->rq_pool; - spin_lock_bh(&pool->sp_lock); list_del_rcu(&rqstp->rq_all); - spin_unlock_bh(&pool->sp_lock); atomic_dec(&pool->sp_nrthreads); From patchwork Wed Aug 30 02:54:53 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: NeilBrown X-Patchwork-Id: 13369930 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 33D19C83F18 for ; Wed, 30 Aug 2023 02:59:53 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S240745AbjH3C7W (ORCPT ); Tue, 29 Aug 2023 22:59:22 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:45740 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S241807AbjH3C7G (ORCPT ); Tue, 29 Aug 2023 22:59:06 -0400 Received: from smtp-out2.suse.de (smtp-out2.suse.de [195.135.220.29]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id ECDC0185 for ; Tue, 29 Aug 2023 19:59:03 -0700 (PDT) Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by smtp-out2.suse.de (Postfix) with ESMTPS id A3B5C1F45F; Wed, 30 Aug 2023 02:59:02 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1693364342; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=wMkCPWkxuyCtogIQr+wijd8g7517SUjyBJmL+548GBc=; b=l70nyySbtZIZLb9V718u56AHXcjShGliVMWoB+pvr7v8I72+uMtrphlZmTToxHr4IiyC/e AQUnq/a/bYk7eMLHkLa4qzFPvHaZDwSTKPScqatqXFpr8zUIca+49LdepNRkk814B4qos7 ZXdCFCfkmgmy2Ddnp2EnsBfpRYnix3o= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1693364342; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=wMkCPWkxuyCtogIQr+wijd8g7517SUjyBJmL+548GBc=; b=Fcp3OBM9goNgnjh9a7AInZBEuRPeh9d4KaIo7auBiZk4yazU5TtF6WL+U3YOgdX3PAMIlD Ofa99dfxDkg/agDw== Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by imap2.suse-dmz.suse.de (Postfix) with ESMTPS id 6233913301; Wed, 30 Aug 2023 02:59:01 +0000 (UTC) Received: from dovecot-director2.suse.de ([192.168.254.65]) by imap2.suse-dmz.suse.de with ESMTPSA id axaCBXWw7mQuZAAAMHmgww (envelope-from ); Wed, 30 Aug 2023 02:59:01 +0000 From: NeilBrown To: Chuck Lever , Jeff Layton Cc: linux-nfs@vger.kernel.org Subject: [PATCH 10/10] SUNRPC: change the back-channel queue to lwq Date: Wed, 30 Aug 2023 12:54:53 +1000 Message-ID: <20230830025755.21292-11-neilb@suse.de> X-Mailer: git-send-email 2.41.0 In-Reply-To: <20230830025755.21292-1-neilb@suse.de> References: <20230830025755.21292-1-neilb@suse.de> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-nfs@vger.kernel.org This removes the need to store and update back-links in the list. It also remove the need for the _bh version of spin_lock(). Signed-off-by: NeilBrown --- include/linux/sunrpc/svc.h | 3 +-- include/linux/sunrpc/xprt.h | 3 ++- net/sunrpc/backchannel_rqst.c | 5 +---- net/sunrpc/svc.c | 3 +-- net/sunrpc/svc_xprt.c | 12 +++--------- net/sunrpc/xprtrdma/backchannel.c | 4 +--- 6 files changed, 9 insertions(+), 21 deletions(-) diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index 8ce1392c1a35..c1feaf0d1542 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -90,10 +90,9 @@ struct svc_serv { int (*sv_threadfn)(void *data); #if defined(CONFIG_SUNRPC_BACKCHANNEL) - struct list_head sv_cb_list; /* queue for callback requests + struct lwq sv_cb_list; /* queue for callback requests * that arrive over the same * connection */ - spinlock_t sv_cb_lock; /* protects the svc_cb_list */ bool sv_bc_enabled; /* service uses backchannel */ #endif /* CONFIG_SUNRPC_BACKCHANNEL */ }; diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index b52411bcfe4e..5413bf694b18 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -57,6 +57,7 @@ struct xprt_class; struct seq_file; struct svc_serv; struct net; +#include /* * This describes a complete RPC request @@ -121,7 +122,7 @@ struct rpc_rqst { int rq_ntrans; #if defined(CONFIG_SUNRPC_BACKCHANNEL) - struct list_head rq_bc_list; /* Callback service list */ + struct lwq_node rq_bc_list; /* Callback service list */ unsigned long rq_bc_pa_state; /* Backchannel prealloc state */ struct list_head rq_bc_pa_list; /* Backchannel prealloc list */ #endif /* CONFIG_SUNRPC_BACKCHANEL */ diff --git a/net/sunrpc/backchannel_rqst.c b/net/sunrpc/backchannel_rqst.c index 44b7c89a635f..caa94cf57123 100644 --- a/net/sunrpc/backchannel_rqst.c +++ b/net/sunrpc/backchannel_rqst.c @@ -83,7 +83,6 @@ static struct rpc_rqst *xprt_alloc_bc_req(struct rpc_xprt *xprt) return NULL; req->rq_xprt = xprt; - INIT_LIST_HEAD(&req->rq_bc_list); /* Preallocate one XDR receive buffer */ if (xprt_alloc_xdr_buf(&req->rq_rcv_buf, gfp_flags) < 0) { @@ -367,8 +366,6 @@ void xprt_complete_bc_request(struct rpc_rqst *req, uint32_t copied) dprintk("RPC: add callback request to list\n"); xprt_get(xprt); - spin_lock(&bc_serv->sv_cb_lock); - list_add(&req->rq_bc_list, &bc_serv->sv_cb_list); - spin_unlock(&bc_serv->sv_cb_lock); + lwq_enqueue(&req->rq_bc_list, &bc_serv->sv_cb_list); svc_pool_wake_idle_thread(&bc_serv->sv_pools[0]); } diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 61ea8ce7975f..3d3aaed8311c 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -438,8 +438,7 @@ EXPORT_SYMBOL_GPL(svc_bind); static void __svc_init_bc(struct svc_serv *serv) { - INIT_LIST_HEAD(&serv->sv_cb_list); - spin_lock_init(&serv->sv_cb_lock); + lwq_init(&serv->sv_cb_list); } #else static void diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 2399811582cc..b4beb38152ab 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -705,7 +705,7 @@ rqst_should_sleep(struct svc_rqst *rqstp) #if defined(CONFIG_SUNRPC_BACKCHANNEL) if (svc_is_backchannel(rqstp)) { - if (!list_empty(&rqstp->rq_server->sv_cb_list)) + if (!lwq_empty(&rqstp->rq_server->sv_cb_list)) return false; } #endif @@ -875,18 +875,12 @@ void svc_recv(struct svc_rqst *rqstp) struct svc_serv *serv = rqstp->rq_server; struct rpc_rqst *req; - spin_lock_bh(&serv->sv_cb_lock); - req = list_first_entry_or_null(&serv->sv_cb_list, - struct rpc_rqst, rq_bc_list); + req = lwq_dequeue(&serv->sv_cb_list, + struct rpc_rqst, rq_bc_list); if (req) { - list_del(&req->rq_bc_list); - spin_unlock_bh(&serv->sv_cb_lock); wake_next(rqstp); - svc_process_bc(req, rqstp); - return; } - spin_unlock_bh(&serv->sv_cb_lock); } #endif } diff --git a/net/sunrpc/xprtrdma/backchannel.c b/net/sunrpc/xprtrdma/backchannel.c index bfc434ec52a7..8c817e755262 100644 --- a/net/sunrpc/xprtrdma/backchannel.c +++ b/net/sunrpc/xprtrdma/backchannel.c @@ -263,9 +263,7 @@ void rpcrdma_bc_receive_call(struct rpcrdma_xprt *r_xprt, /* Queue rqst for ULP's callback service */ bc_serv = xprt->bc_serv; xprt_get(xprt); - spin_lock(&bc_serv->sv_cb_lock); - list_add(&rqst->rq_bc_list, &bc_serv->sv_cb_list); - spin_unlock(&bc_serv->sv_cb_lock); + lwq_enqueue(&rqst->rq_bc_list, &bc_serv->sv_cb_list); svc_pool_wake_idle_thread(&bc_serv->sv_pools[0]);