Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932676AbdDQXxv (ORCPT ); Mon, 17 Apr 2017 19:53:51 -0400 Received: from mx0a-001b2d01.pphosted.com ([148.163.156.1]:41302 "EHLO mx0a-001b2d01.pphosted.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755526AbdDQXpj (ORCPT ); Mon, 17 Apr 2017 19:45:39 -0400 From: "Paul E. McKenney" To: linux-kernel@vger.kernel.org Cc: mingo@kernel.org, jiangshanlai@gmail.com, dipankar@in.ibm.com, akpm@linux-foundation.org, mathieu.desnoyers@efficios.com, josh@joshtriplett.org, tglx@linutronix.de, peterz@infradead.org, rostedt@goodmis.org, dhowells@redhat.com, edumazet@google.com, fweisbec@gmail.com, oleg@redhat.com, bobby.prani@gmail.com, "Paul E. McKenney" Subject: [PATCH v2 tip/core/rcu 11/39] rcu: Place guard on rcu_all_qs() and rcu_note_context_switch() actions Date: Mon, 17 Apr 2017 16:44:58 -0700 X-Mailer: git-send-email 2.5.2 In-Reply-To: <20170417234452.GB19013@linux.vnet.ibm.com> References: <20170417234452.GB19013@linux.vnet.ibm.com> X-TM-AS-GCONF: 00 x-cbid: 17041723-0036-0000-0000-000001E2D52A X-IBM-SpamModules-Scores: X-IBM-SpamModules-Versions: BY=3.00006935; HX=3.00000240; KW=3.00000007; PH=3.00000004; SC=3.00000208; SDB=6.00848869; UDB=6.00419100; IPR=6.00627510; BA=6.00005292; NDR=6.00000001; ZLA=6.00000005; ZF=6.00000009; ZB=6.00000000; ZP=6.00000000; ZH=6.00000000; ZU=6.00000002; MB=3.00015073; XFM=3.00000013; UTC=2017-04-17 23:45:37 X-IBM-AV-DETECTION: SAVI=unused REMOTE=unused XFE=unused x-cbparentid: 17041723-0037-0000-0000-00003FC8C481 Message-Id: <1492472726-3841-11-git-send-email-paulmck@linux.vnet.ibm.com> X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:,, definitions=2017-04-17_20:,, signatures=0 X-Proofpoint-Spam-Details: rule=outbound_notspam policy=outbound score=0 spamscore=0 suspectscore=1 malwarescore=0 phishscore=0 adultscore=0 bulkscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.0.1-1702020001 definitions=main-1704170205 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 8381 Lines: 201 The rcu_all_qs() and rcu_note_context_switch() do a series of checks, taking various actions to supply RCU with quiescent states, depending on the outcomes of the various checks. This is a bit much for scheduling fastpaths, so this commit creates a separate ->rcu_urgent_qs field in the rcu_dynticks structure that acts as a global guard for these checks. Thus, in the common case, rcu_all_qs() and rcu_note_context_switch() check the ->rcu_urgent_qs field, find it false, and simply return. Signed-off-by: Paul E. McKenney Cc: Peter Zijlstra --- .../Design/Data-Structures/Data-Structures.html | 11 ++++++- kernel/rcu/tree.c | 38 ++++++++++++++-------- kernel/rcu/tree.h | 3 +- kernel/rcu/tree_exp.h | 2 ++ kernel/rcu/tree_plugin.h | 8 +++-- 5 files changed, 44 insertions(+), 18 deletions(-) diff --git a/Documentation/RCU/Design/Data-Structures/Data-Structures.html b/Documentation/RCU/Design/Data-Structures/Data-Structures.html index e4bf20a68fa3..4dec89097559 100644 --- a/Documentation/RCU/Design/Data-Structures/Data-Structures.html +++ b/Documentation/RCU/Design/Data-Structures/Data-Structures.html @@ -1106,6 +1106,7 @@ Its fields are as follows: 3 atomic_t dynticks; 4 bool rcu_need_heavy_qs; 5 unsigned long rcu_qs_ctr; + 6 bool rcu_urgent_qs;

The ->dynticks_nesting field counts the @@ -1131,12 +1132,20 @@ it is willing to call for heavy-weight dyntick-counter operations. This flag is checked by RCU's context-switch and cond_resched() code, which provide a momentary idle sojourn in response. -

Finally the ->rcu_qs_ctr field is used to record +

The ->rcu_qs_ctr field is used to record quiescent states from cond_resched(). Because cond_resched() can execute quite frequently, this must be quite lightweight, as in a non-atomic increment of this per-CPU field. +

Finally, the ->rcu_urgent_qs field is used to record +the fact that the RCU core code would really like to see a quiescent +state from the corresponding CPU, with the various other fields indicating +just how badly RCU wants this quiescent state. +This flag is checked by RCU's context-switch and cond_resched() +code, which, if nothing else, non-atomically increment ->rcu_qs_ctr +in response. + diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 3c62ea06edb3..aa81cc6c365e 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -464,8 +464,14 @@ void rcu_note_context_switch(void) trace_rcu_utilization(TPS("Start context switch")); rcu_sched_qs(); rcu_preempt_note_context_switch(); + /* Load rcu_urgent_qs before other flags. */ + if (!smp_load_acquire(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs))) + goto out; + this_cpu_write(rcu_dynticks.rcu_urgent_qs, false); if (unlikely(raw_cpu_read(rcu_dynticks.rcu_need_heavy_qs))) rcu_momentary_dyntick_idle(); + this_cpu_inc(rcu_dynticks.rcu_qs_ctr); +out: trace_rcu_utilization(TPS("End context switch")); barrier(); /* Avoid RCU read-side critical sections leaking up. */ } @@ -488,29 +494,26 @@ void rcu_all_qs(void) { unsigned long flags; + if (!raw_cpu_read(rcu_dynticks.rcu_urgent_qs)) + return; + preempt_disable(); + /* Load rcu_urgent_qs before other flags. */ + if (!smp_load_acquire(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs))) { + preempt_enable(); + return; + } + this_cpu_write(rcu_dynticks.rcu_urgent_qs, false); barrier(); /* Avoid RCU read-side critical sections leaking down. */ if (unlikely(raw_cpu_read(rcu_dynticks.rcu_need_heavy_qs))) { local_irq_save(flags); rcu_momentary_dyntick_idle(); local_irq_restore(flags); } - if (unlikely(raw_cpu_read(rcu_sched_data.cpu_no_qs.b.exp))) { - /* - * Yes, we just checked a per-CPU variable with preemption - * enabled, so we might be migrated to some other CPU at - * this point. That is OK because in that case, the - * migration will supply the needed quiescent state. - * We might end up needlessly disabling preemption and - * invoking rcu_sched_qs() on the destination CPU, but - * the probability and cost are both quite low, so this - * should not be a problem in practice. - */ - preempt_disable(); + if (unlikely(raw_cpu_read(rcu_sched_data.cpu_no_qs.b.exp))) rcu_sched_qs(); - preempt_enable(); - } this_cpu_inc(rcu_dynticks.rcu_qs_ctr); barrier(); /* Avoid RCU read-side critical sections leaking up. */ + preempt_enable(); } EXPORT_SYMBOL_GPL(rcu_all_qs); @@ -1246,6 +1249,7 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp, { unsigned long jtsq; bool *rnhqp; + bool *ruqp; unsigned long rjtsc; struct rcu_node *rnp; @@ -1281,11 +1285,15 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp, * might not be the case for nohz_full CPUs looping in the kernel. */ rnp = rdp->mynode; + ruqp = per_cpu_ptr(&rcu_dynticks.rcu_urgent_qs, rdp->cpu); if (time_after(jiffies, rdp->rsp->gp_start + jtsq) && READ_ONCE(rdp->rcu_qs_ctr_snap) != per_cpu(rcu_dynticks.rcu_qs_ctr, rdp->cpu) && READ_ONCE(rdp->gpnum) == rnp->gpnum && !rdp->gpwrap) { trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, TPS("rqc")); return 1; + } else { + /* Load rcu_qs_ctr before store to rcu_urgent_qs. */ + smp_store_release(ruqp, true); } /* Check for the CPU being offline. */ @@ -1321,6 +1329,8 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp, (time_after(jiffies, rdp->rsp->gp_start + jtsq) || time_after(jiffies, rdp->rsp->jiffies_resched))) { WRITE_ONCE(*rnhqp, true); + /* Store rcu_need_heavy_qs before rcu_urgent_qs. */ + smp_store_release(ruqp, true); rdp->rsp->jiffies_resched += 5; /* Re-enable beating. */ } diff --git a/kernel/rcu/tree.h b/kernel/rcu/tree.h index b212cd0f22c7..d2f276fc2edc 100644 --- a/kernel/rcu/tree.h +++ b/kernel/rcu/tree.h @@ -113,8 +113,9 @@ struct rcu_dynticks { /* Process level is worth LLONG_MAX/2. */ int dynticks_nmi_nesting; /* Track NMI nesting level. */ atomic_t dynticks; /* Even value for idle, else odd. */ - bool rcu_need_heavy_qs; /* GP old, need heavy quiescent state. */ + bool rcu_need_heavy_qs; /* GP old, need heavy quiescent state. */ unsigned long rcu_qs_ctr; /* Light universal quiescent state ctr. */ + bool rcu_urgent_qs; /* GP old need light quiescent state. */ #ifdef CONFIG_NO_HZ_FULL_SYSIDLE long long dynticks_idle_nesting; /* irq/process nesting level from idle. */ diff --git a/kernel/rcu/tree_exp.h b/kernel/rcu/tree_exp.h index a7b639ccd46e..a1f52bbe9db6 100644 --- a/kernel/rcu/tree_exp.h +++ b/kernel/rcu/tree_exp.h @@ -331,6 +331,8 @@ static void sync_sched_exp_handler(void *data) return; } __this_cpu_write(rcu_sched_data.cpu_no_qs.b.exp, true); + /* Store .exp before .rcu_urgent_qs. */ + smp_store_release(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs), true); resched_cpu(smp_processor_id()); } diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h index 0a62a8f1caac..621296a6694b 100644 --- a/kernel/rcu/tree_plugin.h +++ b/kernel/rcu/tree_plugin.h @@ -1860,7 +1860,9 @@ static void __call_rcu_nocb_enqueue(struct rcu_data *rdp, trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("WakeEmpty")); } else { - rdp->nocb_defer_wakeup = RCU_NOGP_WAKE; + WRITE_ONCE(rdp->nocb_defer_wakeup, RCU_NOGP_WAKE); + /* Store ->nocb_defer_wakeup before ->rcu_urgent_qs. */ + smp_store_release(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs), true); trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("WakeEmptyIsDeferred")); } @@ -1872,7 +1874,9 @@ static void __call_rcu_nocb_enqueue(struct rcu_data *rdp, trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("WakeOvf")); } else { - rdp->nocb_defer_wakeup = RCU_NOGP_WAKE_FORCE; + WRITE_ONCE(rdp->nocb_defer_wakeup, RCU_NOGP_WAKE_FORCE); + /* Store ->nocb_defer_wakeup before ->rcu_urgent_qs. */ + smp_store_release(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs), true); trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("WakeOvfIsDeferred")); } -- 2.5.2
 
Quick Quiz: