From patchwork Fri May 14 02:26:56 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Yunsheng Lin X-Patchwork-Id: 12256977 X-Patchwork-Delegate: kuba@kernel.org 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=-16.8 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER,INCLUDES_PATCH, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_GIT 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 DF4BCC43462 for ; Fri, 14 May 2021 02:27:06 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id C4A646143C for ; Fri, 14 May 2021 02:27:06 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231368AbhENC2P (ORCPT ); Thu, 13 May 2021 22:28:15 -0400 Received: from szxga04-in.huawei.com ([45.249.212.190]:3666 "EHLO szxga04-in.huawei.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230129AbhENC2N (ORCPT ); Thu, 13 May 2021 22:28:13 -0400 Received: from DGGEMS414-HUB.china.huawei.com (unknown [172.30.72.60]) by szxga04-in.huawei.com (SkyGuard) with ESMTP id 4FhC5b0mHCz1BMNk; Fri, 14 May 2021 10:24:19 +0800 (CST) Received: from localhost.localdomain (10.69.192.56) by DGGEMS414-HUB.china.huawei.com (10.3.19.214) with Microsoft SMTP Server id 14.3.498.0; Fri, 14 May 2021 10:26:57 +0800 From: Yunsheng Lin To: , CC: , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , Subject: [PATCH net v8 1/3] net: sched: fix packet stuck problem for lockless qdisc Date: Fri, 14 May 2021 10:26:56 +0800 Message-ID: <1620959218-17250-2-git-send-email-linyunsheng@huawei.com> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1620959218-17250-1-git-send-email-linyunsheng@huawei.com> References: <1620959218-17250-1-git-send-email-linyunsheng@huawei.com> MIME-Version: 1.0 X-Originating-IP: [10.69.192.56] X-CFilter-Loop: Reflected Precedence: bulk List-ID: X-Mailing-List: bpf@vger.kernel.org X-Patchwork-Delegate: kuba@kernel.org Lockless qdisc has below concurrent problem: cpu0 cpu1 . . q->enqueue . . . qdisc_run_begin() . . . dequeue_skb() . . . sch_direct_xmit() . . . . q->enqueue . qdisc_run_begin() . return and do nothing . . qdisc_run_end() . cpu1 enqueue a skb without calling __qdisc_run() because cpu0 has not released the lock yet and spin_trylock() return false for cpu1 in qdisc_run_begin(), and cpu0 do not see the skb enqueued by cpu1 when calling dequeue_skb() because cpu1 may enqueue the skb after cpu0 calling dequeue_skb() and before cpu0 calling qdisc_run_end(). Lockless qdisc has below another concurrent problem when tx_action is involved: cpu0(serving tx_action) cpu1 cpu2 . . . . q->enqueue . . qdisc_run_begin() . . dequeue_skb() . . . q->enqueue . . . . sch_direct_xmit() . . . qdisc_run_begin() . . return and do nothing . . . clear __QDISC_STATE_SCHED . . qdisc_run_begin() . . return and do nothing . . . . . . qdisc_run_end() . This patch fixes the above data race by: 1. If the first spin_trylock() return false and STATE_MISSED is not set, set STATE_MISSED and retry another spin_trylock() in case other CPU may not see STATE_MISSED after it releases the lock. 2. reschedule if STATE_MISSED is set after the lock is released at the end of qdisc_run_end(). For tx_action case, STATE_MISSED is also set when cpu1 is at the end if qdisc_run_end(), so tx_action will be rescheduled again to dequeue the skb enqueued by cpu2. Clear STATE_MISSED before retrying a dequeuing when dequeuing returns NULL in order to reduce the overhead of the second spin_trylock() and __netif_schedule() calling. Also clear the STATE_MISSED before calling __netif_schedule() at the end of qdisc_run_end() to avoid doing another round of dequeuing in the pfifo_fast_dequeue(). The performance impact of this patch, tested using pktgen and dummy netdev with pfifo_fast qdisc attached: threads without+this_patch with+this_patch delta 1 2.61Mpps 2.60Mpps -0.3% 2 3.97Mpps 3.82Mpps -3.7% 4 5.62Mpps 5.59Mpps -0.5% 8 2.78Mpps 2.77Mpps -0.3% 16 2.22Mpps 2.22Mpps -0.0% Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Acked-by: Jakub Kicinski Tested-by: Juergen Gross Signed-off-by: Yunsheng Lin --- V7: Clear STATE_MISSED before calling __netif_schedule() as suggested by Jakub. V6: Check MISSED after the first trylock, and remove the automic test and set for performance sake as suggested by Jakub. V4: Change STATE_NEED_RESCHEDULE to STATE_MISSED mirroring NAPI's NAPIF_STATE_MISSED, and add Juergen's "Tested-by" tag for there is only renaming and typo fixing between V4 and V3. V3: Fix a compile error and a few comment typo, remove the __QDISC_STATE_DEACTIVATED checking, and update the performance data. V2: Avoid the overhead of fixing the data race as much as possible. --- include/net/sch_generic.h | 35 ++++++++++++++++++++++++++++++++++- net/sched/sch_generic.c | 19 +++++++++++++++++++ 2 files changed, 53 insertions(+), 1 deletion(-) diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index f7a6e14..1e62551 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -36,6 +36,7 @@ struct qdisc_rate_table { enum qdisc_state_t { __QDISC_STATE_SCHED, __QDISC_STATE_DEACTIVATED, + __QDISC_STATE_MISSED, }; struct qdisc_size_table { @@ -159,8 +160,33 @@ static inline bool qdisc_is_empty(const struct Qdisc *qdisc) static inline bool qdisc_run_begin(struct Qdisc *qdisc) { if (qdisc->flags & TCQ_F_NOLOCK) { + if (spin_trylock(&qdisc->seqlock)) + goto nolock_empty; + + /* If the MISSED flag is set, it means other thread has + * set the MISSED flag before second spin_trylock(), so + * we can return false here to avoid multi cpus doing + * the set_bit() and second spin_trylock() concurrently. + */ + if (test_bit(__QDISC_STATE_MISSED, &qdisc->state)) + return false; + + /* Set the MISSED flag before the second spin_trylock(), + * if the second spin_trylock() return false, it means + * other cpu holding the lock will do dequeuing for us + * or it will see the MISSED flag set after releasing + * lock and reschedule the net_tx_action() to do the + * dequeuing. + */ + set_bit(__QDISC_STATE_MISSED, &qdisc->state); + + /* Retry again in case other CPU may not see the new flag + * after it releases the lock at the end of qdisc_run_end(). + */ if (!spin_trylock(&qdisc->seqlock)) return false; + +nolock_empty: WRITE_ONCE(qdisc->empty, false); } else if (qdisc_is_running(qdisc)) { return false; @@ -176,8 +202,15 @@ static inline bool qdisc_run_begin(struct Qdisc *qdisc) static inline void qdisc_run_end(struct Qdisc *qdisc) { write_seqcount_end(&qdisc->running); - if (qdisc->flags & TCQ_F_NOLOCK) + if (qdisc->flags & TCQ_F_NOLOCK) { spin_unlock(&qdisc->seqlock); + + if (unlikely(test_bit(__QDISC_STATE_MISSED, + &qdisc->state))) { + clear_bit(__QDISC_STATE_MISSED, &qdisc->state); + __netif_schedule(qdisc); + } + } } static inline bool qdisc_may_bulk(const struct Qdisc *qdisc) diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 44991ea..795d986 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -640,8 +640,10 @@ static struct sk_buff *pfifo_fast_dequeue(struct Qdisc *qdisc) { struct pfifo_fast_priv *priv = qdisc_priv(qdisc); struct sk_buff *skb = NULL; + bool need_retry = true; int band; +retry: for (band = 0; band < PFIFO_FAST_BANDS && !skb; band++) { struct skb_array *q = band2list(priv, band); @@ -652,6 +654,23 @@ static struct sk_buff *pfifo_fast_dequeue(struct Qdisc *qdisc) } if (likely(skb)) { qdisc_update_stats_at_dequeue(qdisc, skb); + } else if (need_retry && + test_bit(__QDISC_STATE_MISSED, &qdisc->state)) { + /* Delay clearing the STATE_MISSED here to reduce + * the overhead of the second spin_trylock() in + * qdisc_run_begin() and __netif_schedule() calling + * in qdisc_run_end(). + */ + clear_bit(__QDISC_STATE_MISSED, &qdisc->state); + + /* Make sure dequeuing happens after clearing + * STATE_MISSED. + */ + smp_mb__after_atomic(); + + need_retry = false; + + goto retry; } else { WRITE_ONCE(qdisc->empty, true); } From patchwork Fri May 14 02:26:57 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Yunsheng Lin X-Patchwork-Id: 12256973 X-Patchwork-Delegate: kuba@kernel.org 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=-16.8 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER,INCLUDES_PATCH, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_GIT 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 B94D8C433ED for ; Fri, 14 May 2021 02:27:05 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 96A2061285 for ; Fri, 14 May 2021 02:27:05 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231254AbhENC2N (ORCPT ); Thu, 13 May 2021 22:28:13 -0400 Received: from szxga04-in.huawei.com ([45.249.212.190]:3664 "EHLO szxga04-in.huawei.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229956AbhENC2N (ORCPT ); Thu, 13 May 2021 22:28:13 -0400 Received: from DGGEMS414-HUB.china.huawei.com (unknown [172.30.72.60]) by szxga04-in.huawei.com (SkyGuard) with ESMTP id 4FhC5Z5fnBz1BMNW; Fri, 14 May 2021 10:24:18 +0800 (CST) Received: from localhost.localdomain (10.69.192.56) by DGGEMS414-HUB.china.huawei.com (10.3.19.214) with Microsoft SMTP Server id 14.3.498.0; Fri, 14 May 2021 10:26:57 +0800 From: Yunsheng Lin To: , CC: , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , Subject: [PATCH net v8 2/3] net: sched: fix tx action rescheduling issue during deactivation Date: Fri, 14 May 2021 10:26:57 +0800 Message-ID: <1620959218-17250-3-git-send-email-linyunsheng@huawei.com> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1620959218-17250-1-git-send-email-linyunsheng@huawei.com> References: <1620959218-17250-1-git-send-email-linyunsheng@huawei.com> MIME-Version: 1.0 X-Originating-IP: [10.69.192.56] X-CFilter-Loop: Reflected Precedence: bulk List-ID: X-Mailing-List: bpf@vger.kernel.org X-Patchwork-Delegate: kuba@kernel.org Currently qdisc_run() checks the STATE_DEACTIVATED of lockless qdisc before calling __qdisc_run(), which ultimately clear the STATE_MISSED when all the skb is dequeued. If STATE_DEACTIVATED is set before clearing STATE_MISSED, there may be rescheduling of net_tx_action() at the end of qdisc_run_end(), see below: CPU0(net_tx_atcion) CPU1(__dev_xmit_skb) CPU2(dev_deactivate) . . . . set STATE_MISSED . . __netif_schedule() . . . set STATE_DEACTIVATED . . qdisc_reset() . . . .<--------------- . synchronize_net() clear __QDISC_STATE_SCHED | . . . | . . . | . some_qdisc_is_busy() . | . return *false* . | . . test STATE_DEACTIVATED | . . __qdisc_run() *not* called | . . . | . . test STATE_MISS | . . __netif_schedule()--------| . . . . . . . . __qdisc_run() is not called by net_tx_atcion() in CPU0 because CPU2 has set STATE_DEACTIVATED flag during dev_deactivate(), and STATE_MISSED is only cleared in __qdisc_run(), __netif_schedule is called at the end of qdisc_run_end(), causing tx action rescheduling problem. qdisc_run() called by net_tx_action() runs in the softirq context, which should has the same semantic as the qdisc_run() called by __dev_xmit_skb() protected by rcu_read_lock_bh(). And there is a synchronize_net() between STATE_DEACTIVATED flag being set and qdisc_reset()/some_qdisc_is_busy in dev_deactivate(), we can safely bail out for the deactived lockless qdisc in net_tx_action(), and qdisc_reset() will reset all skb not dequeued yet. So add the rcu_read_lock() explicitly to protect the qdisc_run() and do the STATE_DEACTIVATED checking in net_tx_action() before calling qdisc_run_begin(). Another option is to do the checking in the qdisc_run_end(), but it will add unnecessary overhead for non-tx_action case, because __dev_queue_xmit() will not see qdisc with STATE_DEACTIVATED after synchronize_net(), the qdisc with STATE_DEACTIVATED can only be seen by net_tx_action() because of __netif_schedule(). The STATE_DEACTIVATED checking in qdisc_run() is to avoid race between net_tx_action() and qdisc_reset(), see: commit d518d2ed8640 ("net/sched: fix race between deactivation and dequeue for NOLOCK qdisc"). As the bailout added above for deactived lockless qdisc in net_tx_action() provides better protection for the race without calling qdisc_run() at all, so remove the STATE_DEACTIVATED checking in qdisc_run(). After qdisc_reset(), there is no skb in qdisc to be dequeued, so clear the STATE_MISSED in dev_reset_queue() too. Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Acked-by: Jakub Kicinski Signed-off-by: Yunsheng Lin V7: Clearing STATE_MISSED before calling __netif_schedule() has avoid the endless rescheduling problem, but there may still be a unnecessary rescheduling, so adjust the commit log. --- include/net/pkt_sched.h | 7 +------ net/core/dev.c | 26 ++++++++++++++++++++++---- net/sched/sch_generic.c | 4 +++- 3 files changed, 26 insertions(+), 11 deletions(-) diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h index f5c1bee..6d7b12c 100644 --- a/include/net/pkt_sched.h +++ b/include/net/pkt_sched.h @@ -128,12 +128,7 @@ void __qdisc_run(struct Qdisc *q); static inline void qdisc_run(struct Qdisc *q) { if (qdisc_run_begin(q)) { - /* NOLOCK qdisc must check 'state' under the qdisc seqlock - * to avoid racing with dev_qdisc_reset() - */ - if (!(q->flags & TCQ_F_NOLOCK) || - likely(!test_bit(__QDISC_STATE_DEACTIVATED, &q->state))) - __qdisc_run(q); + __qdisc_run(q); qdisc_run_end(q); } } diff --git a/net/core/dev.c b/net/core/dev.c index 222b1d3..d596cd7 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5025,25 +5025,43 @@ static __latent_entropy void net_tx_action(struct softirq_action *h) sd->output_queue_tailp = &sd->output_queue; local_irq_enable(); + rcu_read_lock(); + while (head) { struct Qdisc *q = head; spinlock_t *root_lock = NULL; head = head->next_sched; - if (!(q->flags & TCQ_F_NOLOCK)) { - root_lock = qdisc_lock(q); - spin_lock(root_lock); - } /* We need to make sure head->next_sched is read * before clearing __QDISC_STATE_SCHED */ smp_mb__before_atomic(); + + if (!(q->flags & TCQ_F_NOLOCK)) { + root_lock = qdisc_lock(q); + spin_lock(root_lock); + } else if (unlikely(test_bit(__QDISC_STATE_DEACTIVATED, + &q->state))) { + /* There is a synchronize_net() between + * STATE_DEACTIVATED flag being set and + * qdisc_reset()/some_qdisc_is_busy() in + * dev_deactivate(), so we can safely bail out + * early here to avoid data race between + * qdisc_deactivate() and some_qdisc_is_busy() + * for lockless qdisc. + */ + clear_bit(__QDISC_STATE_SCHED, &q->state); + continue; + } + clear_bit(__QDISC_STATE_SCHED, &q->state); qdisc_run(q); if (root_lock) spin_unlock(root_lock); } + + rcu_read_unlock(); } xfrm_dev_backlog(sd); diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 795d986..d86c4cc 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -1177,8 +1177,10 @@ static void dev_reset_queue(struct net_device *dev, qdisc_reset(qdisc); spin_unlock_bh(qdisc_lock(qdisc)); - if (nolock) + if (nolock) { + clear_bit(__QDISC_STATE_MISSED, &qdisc->state); spin_unlock_bh(&qdisc->seqlock); + } } static bool some_qdisc_is_busy(struct net_device *dev) From patchwork Fri May 14 02:26:58 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Yunsheng Lin X-Patchwork-Id: 12256975 X-Patchwork-Delegate: kuba@kernel.org 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=-16.8 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER,INCLUDES_PATCH, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_GIT 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 1329BC43461 for ; Fri, 14 May 2021 02:27:06 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id DDC9F61421 for ; Fri, 14 May 2021 02:27:05 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231348AbhENC2O (ORCPT ); Thu, 13 May 2021 22:28:14 -0400 Received: from szxga04-in.huawei.com ([45.249.212.190]:3665 "EHLO szxga04-in.huawei.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229548AbhENC2N (ORCPT ); Thu, 13 May 2021 22:28:13 -0400 Received: from DGGEMS414-HUB.china.huawei.com (unknown [172.30.72.60]) by szxga04-in.huawei.com (SkyGuard) with ESMTP id 4FhC5Z6RzSz1BMNd; Fri, 14 May 2021 10:24:18 +0800 (CST) Received: from localhost.localdomain (10.69.192.56) by DGGEMS414-HUB.china.huawei.com (10.3.19.214) with Microsoft SMTP Server id 14.3.498.0; Fri, 14 May 2021 10:26:58 +0800 From: Yunsheng Lin To: , CC: , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , Subject: [PATCH net v8 3/3] net: sched: fix tx action reschedule issue with stopped queue Date: Fri, 14 May 2021 10:26:58 +0800 Message-ID: <1620959218-17250-4-git-send-email-linyunsheng@huawei.com> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1620959218-17250-1-git-send-email-linyunsheng@huawei.com> References: <1620959218-17250-1-git-send-email-linyunsheng@huawei.com> MIME-Version: 1.0 X-Originating-IP: [10.69.192.56] X-CFilter-Loop: Reflected Precedence: bulk List-ID: X-Mailing-List: bpf@vger.kernel.org X-Patchwork-Delegate: kuba@kernel.org The netdev qeueue might be stopped when byte queue limit has reached or tx hw ring is full, net_tx_action() may still be rescheduled endlessly if STATE_MISSED is set, which consumes a lot of cpu without dequeuing and transmiting any skb because the netdev queue is stopped, see qdisc_run_end(). This patch fixes it by checking the netdev queue state before calling qdisc_run() and clearing STATE_MISSED if netdev queue is stopped during qdisc_run(), the net_tx_action() is recheduled again when netdev qeueue is restarted, see netif_tx_wake_queue(). As there is time window betewwn netif_xmit_frozen_or_stopped() checking and STATE_MISSED clearing, between which STATE_MISSED may set by net_tx_action() scheduled by netif_tx_wake_queue(), so set the STATE_MISSED again if netdev queue is restarted. Fixes: 6b3ba9146fe6 ("net: sched: allow qdiscs to handle locking") Reported-by: Michal Kubecek Acked-by: Jakub Kicinski Signed-off-by: Yunsheng Lin --- V8: Change qdisc_maybe_stop_tx() to qdisc_maybe_clear_missed() as suggested by Jakub. V7: Fix the netif_tx_wake_queue() data race noted by Jakub. V6: Drop NET_XMIT_DROP checking for it is not really relevant to this patch, and it may cause performance performance regression with multi pktgen threads on dummy netdev with pfifo_fast qdisc case. --- net/core/dev.c | 3 ++- net/sched/sch_generic.c | 27 ++++++++++++++++++++++++++- 2 files changed, 28 insertions(+), 2 deletions(-) diff --git a/net/core/dev.c b/net/core/dev.c index d596cd7..ef8cf76 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3853,7 +3853,8 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, if (q->flags & TCQ_F_NOLOCK) { rc = q->enqueue(skb, q, &to_free) & NET_XMIT_MASK; - qdisc_run(q); + if (likely(!netif_xmit_frozen_or_stopped(txq))) + qdisc_run(q); if (unlikely(to_free)) kfree_skb_list(to_free); diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index d86c4cc..fc8b56b 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -35,6 +35,25 @@ const struct Qdisc_ops *default_qdisc_ops = &pfifo_fast_ops; EXPORT_SYMBOL(default_qdisc_ops); +static void qdisc_maybe_clear_missed(struct Qdisc *q, + const struct netdev_queue *txq) +{ + clear_bit(__QDISC_STATE_MISSED, &q->state); + + /* Make sure the below netif_xmit_frozen_or_stopped() + * checking happens after clearing STATE_MISSED. + */ + smp_mb__after_atomic(); + + /* Checking netif_xmit_frozen_or_stopped() again to + * make sure STATE_MISSED is set if the STATE_MISSED + * set by netif_tx_wake_queue()'s rescheduling of + * net_tx_action() is cleared by the above clear_bit(). + */ + if (!netif_xmit_frozen_or_stopped(txq)) + set_bit(__QDISC_STATE_MISSED, &q->state); +} + /* Main transmission queue. */ /* Modifications to data participating in scheduling must be protected with @@ -74,6 +93,7 @@ static inline struct sk_buff *__skb_dequeue_bad_txq(struct Qdisc *q) } } else { skb = SKB_XOFF_MAGIC; + qdisc_maybe_clear_missed(q, txq); } } @@ -242,6 +262,7 @@ static struct sk_buff *dequeue_skb(struct Qdisc *q, bool *validate, } } else { skb = NULL; + qdisc_maybe_clear_missed(q, txq); } if (lock) spin_unlock(lock); @@ -251,8 +272,10 @@ static struct sk_buff *dequeue_skb(struct Qdisc *q, bool *validate, *validate = true; if ((q->flags & TCQ_F_ONETXQUEUE) && - netif_xmit_frozen_or_stopped(txq)) + netif_xmit_frozen_or_stopped(txq)) { + qdisc_maybe_clear_missed(q, txq); return skb; + } skb = qdisc_dequeue_skb_bad_txq(q); if (unlikely(skb)) { @@ -311,6 +334,8 @@ bool sch_direct_xmit(struct sk_buff *skb, struct Qdisc *q, HARD_TX_LOCK(dev, txq, smp_processor_id()); if (!netif_xmit_frozen_or_stopped(txq)) skb = dev_hard_start_xmit(skb, dev, txq, &ret); + else + qdisc_maybe_clear_missed(q, txq); HARD_TX_UNLOCK(dev, txq); } else {