Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-16.6 required=3.0 tests=DKIMWL_WL_MED,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH, MAILING_LIST_MULTI,SIGNED_OFF_BY,SPF_PASS,USER_AGENT_GIT,USER_IN_DEF_DKIM_WL autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 28AB6C43381 for ; Thu, 28 Feb 2019 20:55:50 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id D3C432084F for ; Thu, 28 Feb 2019 20:55:49 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="ispxyN1+" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1732991AbfB1Uzt (ORCPT ); Thu, 28 Feb 2019 15:55:49 -0500 Received: from mail-it1-f202.google.com ([209.85.166.202]:33824 "EHLO mail-it1-f202.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1725918AbfB1Uzs (ORCPT ); Thu, 28 Feb 2019 15:55:48 -0500 Received: by mail-it1-f202.google.com with SMTP id x87so4806007ita.1 for ; Thu, 28 Feb 2019 12:55:47 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20161025; h=date:message-id:mime-version:subject:from:to:cc; bh=qYHPVLPTJqt2/yueEyv19gjc97Ua0g0bGwOrdfwdLSM=; b=ispxyN1+DuqWGmsrjtQdGHM47HzXNkE7NToode5CzJSQQ0kF/gb0KwQUNPLoGvFnWr 2xttMsQU0At2CgikJlLYkedDVmnPPDPL8wh+PMOlW3DVm6CoxAwg4jrEb1mgpsE5R4G1 pz+ZtUXEzQnDpMoCoZTs/RkehwXK4b25xcQ4pnMwupElduyXbWUsjrZQFKddpvcOGMxD 65UwcZvgRRQtmfllKR3pOhqQwBlAr0Z+XeVTfTpCBeZhBXAE1EzzoKeN+X2y3OAhl8Mt DT7w2Hr/i64fmGdEYXO3bN3GjhmC3lvGl602iIrJCEA1LreqBK7LUOhT+liMWn/PiRfH MjgQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:date:message-id:mime-version:subject:from:to:cc; bh=qYHPVLPTJqt2/yueEyv19gjc97Ua0g0bGwOrdfwdLSM=; b=Zr/DIjdFTu+CRVA4sVCtiG274OI0vig57bYSRaX/cOA8FQUvSirnae05W1nSJyqiPB kiLePk6Ygv3X5uokfQD11+sbbAeQzxG4o8PsnAUGjgtZtKtfYkb1io350Jex+pAiQYN3 0hq/t3DqT8L9gnoXMvOWJsy9AC8GlFJOGfq7z1WZCqgh682/9agsyblY559CbfnZ2ojM CgMq9OWUZeubI/vqQ7jOyI7ML7pL0r2geBschKuAL9w1FR5+UM9Dfrd2bmrFnObhkiY5 AmSnGglXs76xG463VZNR7JXp+ZDPHU6KscPwpTSjfm+G+T8WlLf9U1eoBHkZYy8GWIJ4 Il/w== X-Gm-Message-State: APjAAAWGS48CS81XCByflNhBrKgw6XMg34nNUNP7lKj88HjpgR/vzFRc xMw8y6cemvPzOhXe2uu2utQ7degCJ7a1tw== X-Google-Smtp-Source: APXvYqxMyfkZj52aBIz/ykbB17iV91me+v+OFfGPu/vh8FfFbiuLe9+JXq6MXAY7S/btizHF3eFipouAGq0UoA== X-Received: by 2002:a24:1a17:: with SMTP id 23mr1149255iti.9.1551387346889; Thu, 28 Feb 2019 12:55:46 -0800 (PST) Date: Thu, 28 Feb 2019 12:55:43 -0800 Message-Id: <20190228205543.172713-1-edumazet@google.com> Mime-Version: 1.0 X-Mailer: git-send-email 2.21.0.352.gf09ad66450-goog Subject: [PATCH net] net: sched: put back q.qlen into a single location From: Eric Dumazet To: "David S . Miller" Cc: netdev , Eric Dumazet , Eric Dumazet , John Fastabend , Jamal Hadi Salim , Cong Wang , Jiri Pirko Content-Type: text/plain; charset="UTF-8" Sender: netdev-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org In the series fc8b81a5981f ("Merge branch 'lockless-qdisc-series'") John made the assumption that the data path had no need to read the qdisc qlen (number of packets in the qdisc). It is true when pfifo_fast is used as the root qdisc, or as direct MQ/MQPRIO children. But pfifo_fast can be used as leaf in class full qdiscs, and existing logic needs to access the child qlen in an efficient way. HTB breaks badly, since it uses cl->leaf.q->q.qlen in : htb_activate() -> WARN_ON() htb_dequeue_tree() to decide if a class can be htb_deactivated when it has no more packets. HFSC, DRR, CBQ, QFQ have similar issues, and some calls to qdisc_tree_reduce_backlog() also read q.qlen directly. Using qdisc_qlen_sum() (which iterates over all possible cpus) in the data path is a non starter. It seems we have to put back qlen in a central location, at least for stable kernels. For all qdisc but pfifo_fast, qlen is guarded by the qdisc lock, so the existing q.qlen{++|--} are correct. For 'lockless' qdisc (pfifo_fast so far), we need to use atomic_{inc|dec}() because the spinlock might be not held (for example from pfifo_fast_enqueue() and pfifo_fast_dequeue()) This patch adds atomic_qlen (in the same location than qlen) and renames the following helpers, since we want to express they can be used without qdisc lock, and that qlen is no longer percpu. - qdisc_qstats_cpu_qlen_dec -> qdisc_qstats_atomic_qlen_dec() - qdisc_qstats_cpu_qlen_inc -> qdisc_qstats_atomic_qlen_inc() Later (net-next) we might revert this patch by tracking all these qlen uses and replace them by a more efficient method (not having to access a precise qlen, but an empty/non_empty status that might be less expensive to maintain/track). Another possibility is to have a legacy pfifo_fast version that would be used when used a a child qdisc, since the parent qdisc needs a spinlock anyway. But then, future lockless qdiscs would also have the same problem. Fixes: 7e66016f2c65 ("net: sched: helpers to sum qlen and qlen for per cpu logic") Signed-off-by: Eric Dumazet Cc: John Fastabend Cc: Jamal Hadi Salim Cc: Cong Wang Cc: Jiri Pirko --- include/net/sch_generic.h | 31 +++++++++++++------------------ net/core/gen_stats.c | 2 -- net/sched/sch_generic.c | 13 ++++++------- 3 files changed, 19 insertions(+), 27 deletions(-) diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 9481f2c142e26ee1174653d673e6134edd9851da..e7eb4aa6ccc94e24f2d08d92d9df59db903664ab 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -51,7 +51,10 @@ struct qdisc_size_table { struct qdisc_skb_head { struct sk_buff *head; struct sk_buff *tail; - __u32 qlen; + union { + u32 qlen; + atomic_t atomic_qlen; + }; spinlock_t lock; }; @@ -408,27 +411,19 @@ static inline void qdisc_cb_private_validate(const struct sk_buff *skb, int sz) BUILD_BUG_ON(sizeof(qcb->data) < sz); } -static inline int qdisc_qlen_cpu(const struct Qdisc *q) -{ - return this_cpu_ptr(q->cpu_qstats)->qlen; -} - static inline int qdisc_qlen(const struct Qdisc *q) { return q->q.qlen; } -static inline int qdisc_qlen_sum(const struct Qdisc *q) +static inline u32 qdisc_qlen_sum(const struct Qdisc *q) { - __u32 qlen = q->qstats.qlen; - int i; + u32 qlen = q->qstats.qlen; - if (q->flags & TCQ_F_NOLOCK) { - for_each_possible_cpu(i) - qlen += per_cpu_ptr(q->cpu_qstats, i)->qlen; - } else { + if (q->flags & TCQ_F_NOLOCK) + qlen += atomic_read(&q->q.atomic_qlen); + else qlen += q->q.qlen; - } return qlen; } @@ -825,14 +820,14 @@ static inline void qdisc_qstats_cpu_backlog_inc(struct Qdisc *sch, this_cpu_add(sch->cpu_qstats->backlog, qdisc_pkt_len(skb)); } -static inline void qdisc_qstats_cpu_qlen_inc(struct Qdisc *sch) +static inline void qdisc_qstats_atomic_qlen_inc(struct Qdisc *sch) { - this_cpu_inc(sch->cpu_qstats->qlen); + atomic_inc(&sch->q.atomic_qlen); } -static inline void qdisc_qstats_cpu_qlen_dec(struct Qdisc *sch) +static inline void qdisc_qstats_atomic_qlen_dec(struct Qdisc *sch) { - this_cpu_dec(sch->cpu_qstats->qlen); + atomic_dec(&sch->q.atomic_qlen); } static inline void qdisc_qstats_cpu_requeues_inc(struct Qdisc *sch) diff --git a/net/core/gen_stats.c b/net/core/gen_stats.c index 9bf1b9ad17806dfaa579317408e5c4707d014cc0..ac679f74ba4754e69c51dd9b3fb9b23611782300 100644 --- a/net/core/gen_stats.c +++ b/net/core/gen_stats.c @@ -291,7 +291,6 @@ __gnet_stats_copy_queue_cpu(struct gnet_stats_queue *qstats, for_each_possible_cpu(i) { const struct gnet_stats_queue *qcpu = per_cpu_ptr(q, i); - qstats->qlen = 0; qstats->backlog += qcpu->backlog; qstats->drops += qcpu->drops; qstats->requeues += qcpu->requeues; @@ -307,7 +306,6 @@ void __gnet_stats_copy_queue(struct gnet_stats_queue *qstats, if (cpu) { __gnet_stats_copy_queue_cpu(qstats, cpu); } else { - qstats->qlen = q->qlen; qstats->backlog = q->backlog; qstats->drops = q->drops; qstats->requeues = q->requeues; diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 968a85fe4d4a9a0f5887989789cdd97ae7033f5a..de31f2f3b9730bc2e5d35648483d43ea2d86d008 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -68,7 +68,7 @@ static inline struct sk_buff *__skb_dequeue_bad_txq(struct Qdisc *q) skb = __skb_dequeue(&q->skb_bad_txq); if (qdisc_is_percpu_stats(q)) { qdisc_qstats_cpu_backlog_dec(q, skb); - qdisc_qstats_cpu_qlen_dec(q); + qdisc_qstats_atomic_qlen_dec(q); } else { qdisc_qstats_backlog_dec(q, skb); q->q.qlen--; @@ -108,7 +108,7 @@ static inline void qdisc_enqueue_skb_bad_txq(struct Qdisc *q, if (qdisc_is_percpu_stats(q)) { qdisc_qstats_cpu_backlog_inc(q, skb); - qdisc_qstats_cpu_qlen_inc(q); + qdisc_qstats_atomic_qlen_inc(q); } else { qdisc_qstats_backlog_inc(q, skb); q->q.qlen++; @@ -147,7 +147,7 @@ static inline int dev_requeue_skb_locked(struct sk_buff *skb, struct Qdisc *q) qdisc_qstats_cpu_requeues_inc(q); qdisc_qstats_cpu_backlog_inc(q, skb); - qdisc_qstats_cpu_qlen_inc(q); + qdisc_qstats_atomic_qlen_inc(q); skb = next; } @@ -252,7 +252,7 @@ static struct sk_buff *dequeue_skb(struct Qdisc *q, bool *validate, skb = __skb_dequeue(&q->gso_skb); if (qdisc_is_percpu_stats(q)) { qdisc_qstats_cpu_backlog_dec(q, skb); - qdisc_qstats_cpu_qlen_dec(q); + qdisc_qstats_atomic_qlen_dec(q); } else { qdisc_qstats_backlog_dec(q, skb); q->q.qlen--; @@ -645,7 +645,7 @@ static int pfifo_fast_enqueue(struct sk_buff *skb, struct Qdisc *qdisc, if (unlikely(err)) return qdisc_drop_cpu(skb, qdisc, to_free); - qdisc_qstats_cpu_qlen_inc(qdisc); + qdisc_qstats_atomic_qlen_inc(qdisc); /* Note: skb can not be used after skb_array_produce(), * so we better not use qdisc_qstats_cpu_backlog_inc() */ @@ -670,7 +670,7 @@ static struct sk_buff *pfifo_fast_dequeue(struct Qdisc *qdisc) if (likely(skb)) { qdisc_qstats_cpu_backlog_dec(qdisc, skb); qdisc_bstats_cpu_update(qdisc, skb); - qdisc_qstats_cpu_qlen_dec(qdisc); + qdisc_qstats_atomic_qlen_dec(qdisc); } return skb; @@ -714,7 +714,6 @@ static void pfifo_fast_reset(struct Qdisc *qdisc) struct gnet_stats_queue *q = per_cpu_ptr(qdisc->cpu_qstats, i); q->backlog = 0; - q->qlen = 0; } } -- 2.21.0.352.gf09ad66450-goog