From 88b91c7ca49bc8600cf1106eb891d08c1965b9ce Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 26 Oct 2009 10:24:31 -0700 Subject: rcu: Simplify creating of lockdep class for root rcu_node Use lockdep_set_class() to simplify the code and to avoid any additional overhead in the !LOCKDEP case. Also move the definition of rcu_root_class into kernel/rcutree.c, as suggested by Lai Jiangshan. Signed-off-by: Peter Zijlstra Signed-off-by: Paul E. McKenney Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <1256577871443-git-send-email-> Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'kernel/rcutree.c') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index ddbf111e9e18..055f1a941a9e 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -51,6 +51,8 @@ /* Data structures. */ +static struct lock_class_key rcu_root_class; + #define RCU_STATE_INITIALIZER(name) { \ .level = { &name.node[0] }, \ .levelcnt = { \ @@ -1666,8 +1668,7 @@ static void __init rcu_init_one(struct rcu_state *rsp) cpustride *= rsp->levelspread[i]; rnp = rsp->level[i]; for (j = 0; j < rsp->levelcnt[i]; j++, rnp++) { - if (rnp != rcu_get_root(rsp)) - spin_lock_init(&rnp->lock); + spin_lock_init(&rnp->lock); rnp->gpnum = 0; rnp->qsmask = 0; rnp->qsmaskinit = 0; @@ -1690,7 +1691,7 @@ static void __init rcu_init_one(struct rcu_state *rsp) INIT_LIST_HEAD(&rnp->blocked_tasks[1]); } } - spin_lock_init(&rcu_get_root(rsp)->lock); + lockdep_set_class(&rcu_get_root(rsp)->lock, &rcu_root_class); } /* -- cgit v1.2.2 From 281d150c5f8892f158747594ab49ce2823fd8b8c Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 2 Nov 2009 13:52:27 -0800 Subject: rcu: Prepare for synchronization fixes: clean up for non-NO_HZ handling of ->completed counter Impose a clear locking design on non-NO_HZ handling of the ->completed counter. This increases the distance between the RCU and the CPU-hotplug mechanisms. Signed-off-by: Paul E. McKenney Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com Cc: # .32.x LKML-Reference: <12571987491353-git-send-email-> Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 67 +++++++++++++++++++++++++------------------------------- 1 file changed, 30 insertions(+), 37 deletions(-) (limited to 'kernel/rcutree.c') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 69f4efec3449..26249abf24dc 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -178,8 +178,28 @@ static struct rcu_node *rcu_get_root(struct rcu_state *rsp) return &rsp->node[0]; } +/* + * Record the specified "completed" value, which is later used to validate + * dynticks counter manipulations and CPU-offline checks. Specify + * "rsp->completed - 1" to unconditionally invalidate any future dynticks + * manipulations and CPU-offline checks. Such invalidation is useful at + * the beginning of a grace period. + */ +static void dyntick_record_completed(struct rcu_state *rsp, long comp) +{ + rsp->dynticks_completed = comp; +} + #ifdef CONFIG_SMP +/* + * Recall the previously recorded value of the completion for dynticks. + */ +static long dyntick_recall_completed(struct rcu_state *rsp) +{ + return rsp->dynticks_completed; +} + /* * If the specified CPU is offline, tell the caller that it is in * a quiescent state. Otherwise, whack it with a reschedule IPI. @@ -337,27 +357,8 @@ void rcu_irq_exit(void) set_need_resched(); } -/* - * Record the specified "completed" value, which is later used to validate - * dynticks counter manipulations. Specify "rsp->completed - 1" to - * unconditionally invalidate any future dynticks manipulations (which is - * useful at the beginning of a grace period). - */ -static void dyntick_record_completed(struct rcu_state *rsp, long comp) -{ - rsp->dynticks_completed = comp; -} - #ifdef CONFIG_SMP -/* - * Recall the previously recorded value of the completion for dynticks. - */ -static long dyntick_recall_completed(struct rcu_state *rsp) -{ - return rsp->dynticks_completed; -} - /* * Snapshot the specified CPU's dynticks counter so that we can later * credit them with an implicit quiescent state. Return 1 if this CPU @@ -421,24 +422,8 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp) #else /* #ifdef CONFIG_NO_HZ */ -static void dyntick_record_completed(struct rcu_state *rsp, long comp) -{ -} - #ifdef CONFIG_SMP -/* - * If there are no dynticks, then the only way that a CPU can passively - * be in a quiescent state is to be offline. Unlike dynticks idle, which - * is a point in time during the prior (already finished) grace period, - * an offline CPU is always in a quiescent state, and thus can be - * unconditionally applied. So just return the current value of completed. - */ -static long dyntick_recall_completed(struct rcu_state *rsp) -{ - return rsp->completed; -} - static int dyntick_save_progress_counter(struct rcu_data *rdp) { return 0; @@ -1146,6 +1131,7 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed) long lastcomp; struct rcu_node *rnp = rcu_get_root(rsp); u8 signaled; + u8 forcenow; if (!rcu_gp_in_progress(rsp)) return; /* No grace period in progress, nothing to force. */ @@ -1182,16 +1168,23 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed) if (rcu_process_dyntick(rsp, lastcomp, dyntick_save_progress_counter)) goto unlock_ret; + /* fall into next case. */ + + case RCU_SAVE_COMPLETED: /* Update state, record completion counter. */ + forcenow = 0; spin_lock(&rnp->lock); if (lastcomp == rsp->completed && - rsp->signaled == RCU_SAVE_DYNTICK) { + rsp->signaled == signaled) { rsp->signaled = RCU_FORCE_QS; dyntick_record_completed(rsp, lastcomp); + forcenow = signaled == RCU_SAVE_COMPLETED; } spin_unlock(&rnp->lock); - break; + if (!forcenow) + break; + /* fall into next case. */ case RCU_FORCE_QS: -- cgit v1.2.2 From d09b62dfa336447c52a5ec9bb88adbc479b0f3b8 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 2 Nov 2009 13:52:28 -0800 Subject: rcu: Fix synchronization for rcu_process_gp_end() uses of ->completed counter Impose a clear locking design on the rcu_process_gp_end() function's use of the ->completed counter. This is done by creating a ->completed field in the rcu_node structure, which can safely be accessed under the protection of that structure's lock. Performance and scalability are maintained by using a form of double-checked locking, so that rcu_process_gp_end() only acquires the leaf rcu_node structure's ->lock if a grace period has recently ended. This fix reduces rcutorture failure rate by at least two orders of magnitude under heavy stress with force_quiescent_state() being invoked artificially often. Without this fix, unsynchronized access to the ->completed field can cause rcu_process_gp_end() to advance callbacks whose grace period has not yet expired. (Bad idea!) Signed-off-by: Paul E. McKenney Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com Cc: # .32.x LKML-Reference: <12571987494069-git-send-email-> Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 128 ++++++++++++++++++++++++++++++++++--------------------- 1 file changed, 80 insertions(+), 48 deletions(-) (limited to 'kernel/rcutree.c') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 26249abf24dc..9e068d112153 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -569,6 +569,76 @@ check_for_new_grace_period(struct rcu_state *rsp, struct rcu_data *rdp) return ret; } +/* + * Advance this CPU's callbacks, but only if the current grace period + * has ended. This may be called only from the CPU to whom the rdp + * belongs. In addition, the corresponding leaf rcu_node structure's + * ->lock must be held by the caller, with irqs disabled. + */ +static void +__rcu_process_gp_end(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp) +{ + /* Did another grace period end? */ + if (rdp->completed != rnp->completed) { + + /* Advance callbacks. No harm if list empty. */ + rdp->nxttail[RCU_DONE_TAIL] = rdp->nxttail[RCU_WAIT_TAIL]; + rdp->nxttail[RCU_WAIT_TAIL] = rdp->nxttail[RCU_NEXT_READY_TAIL]; + rdp->nxttail[RCU_NEXT_READY_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; + + /* Remember that we saw this grace-period completion. */ + rdp->completed = rnp->completed; + } +} + +/* + * Advance this CPU's callbacks, but only if the current grace period + * has ended. This may be called only from the CPU to whom the rdp + * belongs. + */ +static void +rcu_process_gp_end(struct rcu_state *rsp, struct rcu_data *rdp) +{ + unsigned long flags; + struct rcu_node *rnp; + + local_irq_save(flags); + rnp = rdp->mynode; + if (rdp->completed == ACCESS_ONCE(rnp->completed) || /* outside lock. */ + !spin_trylock(&rnp->lock)) { /* irqs already off, retry later. */ + local_irq_restore(flags); + return; + } + __rcu_process_gp_end(rsp, rnp, rdp); + spin_unlock_irqrestore(&rnp->lock, flags); +} + +/* + * Do per-CPU grace-period initialization for running CPU. The caller + * must hold the lock of the leaf rcu_node structure corresponding to + * this CPU. + */ +static void +rcu_start_gp_per_cpu(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp) +{ + /* Prior grace period ended, so advance callbacks for current CPU. */ + __rcu_process_gp_end(rsp, rnp, rdp); + + /* + * Because this CPU just now started the new grace period, we know + * that all of its callbacks will be covered by this upcoming grace + * period, even the ones that were registered arbitrarily recently. + * Therefore, advance all outstanding callbacks to RCU_WAIT_TAIL. + * + * Other CPUs cannot be sure exactly when the grace period started. + * Therefore, their recently registered callbacks must pass through + * an additional RCU_NEXT_READY stage, so that they will be handled + * by the next RCU grace period. + */ + rdp->nxttail[RCU_NEXT_READY_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; + rdp->nxttail[RCU_WAIT_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; +} + /* * Start a new RCU grace period if warranted, re-initializing the hierarchy * in preparation for detecting the next grace period. The caller must hold @@ -596,26 +666,14 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags) dyntick_record_completed(rsp, rsp->completed - 1); note_new_gpnum(rsp, rdp); - /* - * Because this CPU just now started the new grace period, we know - * that all of its callbacks will be covered by this upcoming grace - * period, even the ones that were registered arbitrarily recently. - * Therefore, advance all outstanding callbacks to RCU_WAIT_TAIL. - * - * Other CPUs cannot be sure exactly when the grace period started. - * Therefore, their recently registered callbacks must pass through - * an additional RCU_NEXT_READY stage, so that they will be handled - * by the next RCU grace period. - */ - rdp->nxttail[RCU_NEXT_READY_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; - rdp->nxttail[RCU_WAIT_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; - /* Special-case the common single-level case. */ if (NUM_RCU_NODES == 1) { rcu_preempt_check_blocked_tasks(rnp); rnp->qsmask = rnp->qsmaskinit; rnp->gpnum = rsp->gpnum; + rnp->completed = rsp->completed; rsp->signaled = RCU_SIGNAL_INIT; /* force_quiescent_state OK. */ + rcu_start_gp_per_cpu(rsp, rnp, rdp); spin_unlock_irqrestore(&rnp->lock, flags); return; } @@ -648,6 +706,9 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags) rcu_preempt_check_blocked_tasks(rnp); rnp->qsmask = rnp->qsmaskinit; rnp->gpnum = rsp->gpnum; + rnp->completed = rsp->completed; + if (rnp == rdp->mynode) + rcu_start_gp_per_cpu(rsp, rnp, rdp); spin_unlock(&rnp->lock); /* irqs remain disabled. */ } @@ -658,34 +719,6 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags) spin_unlock_irqrestore(&rsp->onofflock, flags); } -/* - * Advance this CPU's callbacks, but only if the current grace period - * has ended. This may be called only from the CPU to whom the rdp - * belongs. - */ -static void -rcu_process_gp_end(struct rcu_state *rsp, struct rcu_data *rdp) -{ - long completed_snap; - unsigned long flags; - - local_irq_save(flags); - completed_snap = ACCESS_ONCE(rsp->completed); /* outside of lock. */ - - /* Did another grace period end? */ - if (rdp->completed != completed_snap) { - - /* Advance callbacks. No harm if list empty. */ - rdp->nxttail[RCU_DONE_TAIL] = rdp->nxttail[RCU_WAIT_TAIL]; - rdp->nxttail[RCU_WAIT_TAIL] = rdp->nxttail[RCU_NEXT_READY_TAIL]; - rdp->nxttail[RCU_NEXT_READY_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; - - /* Remember that we saw this grace-period completion. */ - rdp->completed = completed_snap; - } - local_irq_restore(flags); -} - /* * Clean up after the prior grace period and let rcu_start_gp() start up * the next grace period if one is needed. Note that the caller must @@ -697,7 +730,6 @@ static void cpu_quiet_msk_finish(struct rcu_state *rsp, unsigned long flags) WARN_ON_ONCE(!rcu_gp_in_progress(rsp)); rsp->completed = rsp->gpnum; rsp->signaled = RCU_GP_IDLE; - rcu_process_gp_end(rsp, rsp->rda[smp_processor_id()]); rcu_start_gp(rsp, flags); /* releases root node's rnp->lock. */ } @@ -1539,21 +1571,16 @@ static void __cpuinit rcu_init_percpu_data(int cpu, struct rcu_state *rsp, int preemptable) { unsigned long flags; - long lastcomp; unsigned long mask; struct rcu_data *rdp = rsp->rda[cpu]; struct rcu_node *rnp = rcu_get_root(rsp); /* Set up local state, ensuring consistent view of global state. */ spin_lock_irqsave(&rnp->lock, flags); - lastcomp = rsp->completed; - rdp->completed = lastcomp; - rdp->gpnum = lastcomp; rdp->passed_quiesc = 0; /* We could be racing with new GP, */ rdp->qs_pending = 1; /* so set up to respond to current GP. */ rdp->beenonline = 1; /* We have now been online. */ rdp->preemptable = preemptable; - rdp->passed_quiesc_completed = lastcomp - 1; rdp->qlen_last_fqs_check = 0; rdp->n_force_qs_snap = rsp->n_force_qs; rdp->blimit = blimit; @@ -1575,6 +1602,11 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp, int preemptable) spin_lock(&rnp->lock); /* irqs already disabled. */ rnp->qsmaskinit |= mask; mask = rnp->grpmask; + if (rnp == rdp->mynode) { + rdp->gpnum = rnp->completed; /* if GP in progress... */ + rdp->completed = rnp->completed; + rdp->passed_quiesc_completed = rnp->completed - 1; + } spin_unlock(&rnp->lock); /* irqs already disabled. */ rnp = rnp->parent; } while (rnp != NULL && !(rnp->qsmaskinit & mask)); -- cgit v1.2.2 From 9160306e6f5b68bb64630c9031c517ca1cf463db Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 2 Nov 2009 13:52:29 -0800 Subject: rcu: Fix note_new_gpnum() uses of ->gpnum Impose a clear locking design on the note_new_gpnum() function's use of the ->gpnum counter. This is done by updating rdp->gpnum only from the corresponding leaf rcu_node structure's rnp->gpnum field, and even then only under the protection of that same rcu_node structure's ->lock field. Performance and scalability are maintained using a form of double-checked locking, and excessive spinning is avoided by use of the spin_trylock() function. The use of spin_trylock() is safe due to the fact that CPUs who fail to acquire this lock will try again later. The hierarchical nature of the rcu_node data structure limits contention (which could be limited further if need be using the RCU_FANOUT kernel parameter). Without this patch, obscure but quite possible races could result in a quiescent state that occurred during one grace period to be accounted to the following grace period, causing this following grace period to end prematurely. Not good! Signed-off-by: Paul E. McKenney Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com Cc: # .32.x LKML-Reference: <12571987492350-git-send-email-> Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 32 +++++++++++++++++++++++++++----- 1 file changed, 27 insertions(+), 5 deletions(-) (limited to 'kernel/rcutree.c') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 9e068d112153..ec007dd22632 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -540,13 +540,33 @@ static void check_cpu_stall(struct rcu_state *rsp, struct rcu_data *rdp) /* * Update CPU-local rcu_data state to record the newly noticed grace period. * This is used both when we started the grace period and when we notice - * that someone else started the grace period. + * that someone else started the grace period. The caller must hold the + * ->lock of the leaf rcu_node structure corresponding to the current CPU, + * and must have irqs disabled. */ +static void __note_new_gpnum(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp) +{ + if (rdp->gpnum != rnp->gpnum) { + rdp->qs_pending = 1; + rdp->passed_quiesc = 0; + rdp->gpnum = rnp->gpnum; + } +} + static void note_new_gpnum(struct rcu_state *rsp, struct rcu_data *rdp) { - rdp->qs_pending = 1; - rdp->passed_quiesc = 0; - rdp->gpnum = rsp->gpnum; + unsigned long flags; + struct rcu_node *rnp; + + local_irq_save(flags); + rnp = rdp->mynode; + if (rdp->gpnum == ACCESS_ONCE(rnp->gpnum) || /* outside lock. */ + !spin_trylock(&rnp->lock)) { /* irqs already off, retry later. */ + local_irq_restore(flags); + return; + } + __note_new_gpnum(rsp, rnp, rdp); + spin_unlock_irqrestore(&rnp->lock, flags); } /* @@ -637,6 +657,9 @@ rcu_start_gp_per_cpu(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_dat */ rdp->nxttail[RCU_NEXT_READY_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; rdp->nxttail[RCU_WAIT_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; + + /* Set state so that this CPU will detect the next quiescent state. */ + __note_new_gpnum(rsp, rnp, rdp); } /* @@ -664,7 +687,6 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags) rsp->jiffies_force_qs = jiffies + RCU_JIFFIES_TILL_FORCE_QS; record_gp_stall_check_time(rsp); dyntick_record_completed(rsp, rsp->completed - 1); - note_new_gpnum(rsp, rdp); /* Special-case the common single-level case. */ if (NUM_RCU_NODES == 1) { -- cgit v1.2.2 From 4bcfe055030d9e953945def3864f7e6997b27782 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 10 Nov 2009 13:37:21 -0800 Subject: rcu: Rename dynticks_completed to completed_fqs This field is used whether or not CONFIG_NO_HZ is set, so the old name of ->dynticks_completed is quite misleading. Change to ->completed_fqs, given that it the value that force_quiescent_state() is trying to drive the ->completed field away from. Signed-off-by: Paul E. McKenney Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <12578890423298-git-send-email-> Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/rcutree.c') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index ec007dd22632..26fc7807761d 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -187,7 +187,7 @@ static struct rcu_node *rcu_get_root(struct rcu_state *rsp) */ static void dyntick_record_completed(struct rcu_state *rsp, long comp) { - rsp->dynticks_completed = comp; + rsp->completed_fqs = comp; } #ifdef CONFIG_SMP @@ -197,7 +197,7 @@ static void dyntick_record_completed(struct rcu_state *rsp, long comp) */ static long dyntick_recall_completed(struct rcu_state *rsp) { - return rsp->dynticks_completed; + return rsp->completed_fqs; } /* -- cgit v1.2.2 From c64ac3ce06558e534aec62b1fadeb0a3f111dac1 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 10 Nov 2009 13:37:22 -0800 Subject: rcu: Simplify association of quiescent states with grace periods The rdp->passed_quiesc_completed fields are used to properly associate the recorded quiescent state with a grace period. It is OK to wrongly associate a given quiescent state with a preceding grace period, but it is fatal to associate a given quiescent state with a grace period that begins after the quiescent state occurred. Grace periods are numbered, and the following fields track them: o ->gpnum is the number of the grace period currently in progress, or the number of the last grace period to complete if no grace period is currently in progress. o ->completed is the number of the last grace period to have completed. These two fields are equal if there is no grace period in progress, otherwise ->gpnum is one greater than ->completed. But the rdp->passed_quiesc_completed field compared against ->completed, and if equal, the quiescent state is presumed to count against the current grace period. The earlier code copied rdp->completed to rdp->passed_quiesc_completed, which has been made to work, but is error-prone. In contrast, copying one less than rdp->gpnum is guaranteed safe, because rdp->gpnum is not incremented until after the start of the corresponding grace period. At the end of the grace period, when ->completed has incremented, then any quiescent periods recorded previously will be discarded. Signed-off-by: Paul E. McKenney Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <12578890421011-git-send-email-> Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/rcutree.c') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 26fc7807761d..d8024192c73b 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -100,7 +100,7 @@ void rcu_sched_qs(int cpu) struct rcu_data *rdp; rdp = &per_cpu(rcu_sched_data, cpu); - rdp->passed_quiesc_completed = rdp->completed; + rdp->passed_quiesc_completed = rdp->gpnum - 1; barrier(); rdp->passed_quiesc = 1; rcu_preempt_note_context_switch(cpu); @@ -111,7 +111,7 @@ void rcu_bh_qs(int cpu) struct rcu_data *rdp; rdp = &per_cpu(rcu_bh_data, cpu); - rdp->passed_quiesc_completed = rdp->completed; + rdp->passed_quiesc_completed = rdp->gpnum - 1; barrier(); rdp->passed_quiesc = 1; } -- cgit v1.2.2 From b32e9eb6ad29572b4451847d0e8227c9be2b6d69 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 12 Nov 2009 22:35:03 -0800 Subject: rcu: Accelerate callback processing on CPUs not detecting GP end An earlier fix for a race resulted in a situation where the CPUs other than the CPU that detected the end of the grace period would not process their callbacks until the next grace period started. This means that these other CPUs would unnecessarily demand that an extra grace period be started. This patch eliminates this extra grace period and speeds callback processing by propagating rsp->completed to the rcu_node structures in the case where the CPU detecting the end of the grace period sees no reason to start a new grace period. Signed-off-by: Paul E. McKenney Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <1258094104417-git-send-email-> Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 18 +++++++++++++++++- 1 file changed, 17 insertions(+), 1 deletion(-) (limited to 'kernel/rcutree.c') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index d8024192c73b..b4efb9e36680 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -676,7 +676,23 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags) struct rcu_node *rnp = rcu_get_root(rsp); if (!cpu_needs_another_gp(rsp, rdp)) { - spin_unlock_irqrestore(&rnp->lock, flags); + if (rnp->completed == rsp->completed) { + spin_unlock_irqrestore(&rnp->lock, flags); + return; + } + spin_unlock(&rnp->lock); /* irqs remain disabled. */ + + /* + * Propagate new ->completed value to rcu_node structures + * so that other CPUs don't have to wait until the start + * of the next grace period to process their callbacks. + */ + rcu_for_each_node_breadth_first(rsp, rnp) { + spin_lock(&rnp->lock); /* irqs already disabled. */ + rnp->completed = rsp->completed; + spin_unlock(&rnp->lock); /* irqs remain disabled. */ + } + local_irq_restore(flags); return; } -- cgit v1.2.2 From 8e9aa8f067d2dcd9457980ced618e1cffbcfba46 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 12 Nov 2009 22:35:04 -0800 Subject: rcu: Simplify association of forced quiescent states with grace periods The force_quiescent_state() function also took a snapshot of the ->completed field, which was as obnoxious as it was in rcu_sched_qs() and friends. So snapshot ->gpnum-1. Also, since the dyntick_record_completed() and dyntick_recall_completed() functions are now simple assignments that are independent of CONFIG_NO_HZ, and since their names are now misleading, get rid of them. Signed-off-by: Paul E. McKenney Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <12580941042308-git-send-email-> Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 27 +++------------------------ 1 file changed, 3 insertions(+), 24 deletions(-) (limited to 'kernel/rcutree.c') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index b4efb9e36680..3df04381ea3e 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -178,28 +178,8 @@ static struct rcu_node *rcu_get_root(struct rcu_state *rsp) return &rsp->node[0]; } -/* - * Record the specified "completed" value, which is later used to validate - * dynticks counter manipulations and CPU-offline checks. Specify - * "rsp->completed - 1" to unconditionally invalidate any future dynticks - * manipulations and CPU-offline checks. Such invalidation is useful at - * the beginning of a grace period. - */ -static void dyntick_record_completed(struct rcu_state *rsp, long comp) -{ - rsp->completed_fqs = comp; -} - #ifdef CONFIG_SMP -/* - * Recall the previously recorded value of the completion for dynticks. - */ -static long dyntick_recall_completed(struct rcu_state *rsp) -{ - return rsp->completed_fqs; -} - /* * If the specified CPU is offline, tell the caller that it is in * a quiescent state. Otherwise, whack it with a reschedule IPI. @@ -702,7 +682,6 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags) rsp->signaled = RCU_GP_INIT; /* Hold off force_quiescent_state. */ rsp->jiffies_force_qs = jiffies + RCU_JIFFIES_TILL_FORCE_QS; record_gp_stall_check_time(rsp); - dyntick_record_completed(rsp, rsp->completed - 1); /* Special-case the common single-level case. */ if (NUM_RCU_NODES == 1) { @@ -1214,7 +1193,7 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed) goto unlock_ret; /* no emergency and done recently. */ rsp->n_force_qs++; spin_lock(&rnp->lock); - lastcomp = rsp->completed; + lastcomp = rsp->gpnum - 1; signaled = rsp->signaled; rsp->jiffies_force_qs = jiffies + RCU_JIFFIES_TILL_FORCE_QS; if (lastcomp == rsp->gpnum) { @@ -1248,7 +1227,7 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed) if (lastcomp == rsp->completed && rsp->signaled == signaled) { rsp->signaled = RCU_FORCE_QS; - dyntick_record_completed(rsp, lastcomp); + rsp->completed_fqs = lastcomp; forcenow = signaled == RCU_SAVE_COMPLETED; } spin_unlock(&rnp->lock); @@ -1259,7 +1238,7 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed) case RCU_FORCE_QS: /* Check dyntick-idle state, send IPI to laggarts. */ - if (rcu_process_dyntick(rsp, dyntick_recall_completed(rsp), + if (rcu_process_dyntick(rsp, rsp->completed_fqs, rcu_implicit_dynticks_qs)) goto unlock_ret; -- cgit v1.2.2 From 560d4bc0df9a5e63b980432282d8c2bd3559ec74 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 13 Nov 2009 19:51:38 -0800 Subject: rcu: Further cleanups of use of lastcomp Now that a copy of the rsp->completed flag is available in all rcu_node structures, make full use of it. It is still legitimate to access rsp->completed while holding the root rcu_node structure's lock, however. Also, tighten up force_quiescent_state()'s checks for end of current grace period. Signed-off-by: Paul E. McKenney Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <1258170699933-git-send-email-> Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) (limited to 'kernel/rcutree.c') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 3df04381ea3e..24bbf2ce0605 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -817,7 +817,7 @@ cpu_quiet(int cpu, struct rcu_state *rsp, struct rcu_data *rdp, long lastcomp) rnp = rdp->mynode; spin_lock_irqsave(&rnp->lock, flags); - if (lastcomp != ACCESS_ONCE(rsp->completed)) { + if (lastcomp != rnp->completed) { /* * Someone beat us to it for this grace period, so leave. @@ -935,7 +935,6 @@ static void rcu_adopt_orphan_cbs(struct rcu_state *rsp) static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp) { unsigned long flags; - long lastcomp; unsigned long mask; struct rcu_data *rdp = rsp->rda[cpu]; struct rcu_node *rnp; @@ -971,7 +970,6 @@ static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp) spin_unlock(&rnp->lock); /* irqs remain disabled. */ rnp = rnp->parent; } while (rnp != NULL); - lastcomp = rsp->completed; spin_unlock_irqrestore(&rsp->onofflock, flags); @@ -1145,7 +1143,7 @@ static int rcu_process_dyntick(struct rcu_state *rsp, long lastcomp, rcu_for_each_leaf_node(rsp, rnp) { mask = 0; spin_lock_irqsave(&rnp->lock, flags); - if (rsp->completed != lastcomp) { + if (rnp->completed != lastcomp) { spin_unlock_irqrestore(&rnp->lock, flags); return 1; } @@ -1159,7 +1157,7 @@ static int rcu_process_dyntick(struct rcu_state *rsp, long lastcomp, if ((rnp->qsmask & bit) != 0 && f(rsp->rda[cpu])) mask |= bit; } - if (mask != 0 && rsp->completed == lastcomp) { + if (mask != 0 && rnp->completed == lastcomp) { /* cpu_quiet_msk() releases rnp->lock. */ cpu_quiet_msk(mask, rsp, rnp, flags); @@ -1196,7 +1194,7 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed) lastcomp = rsp->gpnum - 1; signaled = rsp->signaled; rsp->jiffies_force_qs = jiffies + RCU_JIFFIES_TILL_FORCE_QS; - if (lastcomp == rsp->gpnum) { + if(!rcu_gp_in_progress(rsp)) { rsp->n_force_qs_ngp++; spin_unlock(&rnp->lock); goto unlock_ret; /* no GP in progress, time updated. */ @@ -1224,7 +1222,8 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed) /* Update state, record completion counter. */ forcenow = 0; spin_lock(&rnp->lock); - if (lastcomp == rsp->completed && + if (lastcomp + 1 == rsp->gpnum && + lastcomp == rsp->completed && rsp->signaled == signaled) { rsp->signaled = RCU_FORCE_QS; rsp->completed_fqs = lastcomp; -- cgit v1.2.2 From 2f51f9884f6a36b0fe9636d5a1937e5cbd25723b Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 13 Nov 2009 19:51:39 -0800 Subject: rcu: Eliminate __rcu_pending() false positives Now that there are both ->gpnum and ->completed fields in the rcu_node structure, __rcu_pending() should check rdp->gpnum and rdp->completed against rnp->gpnum and rdp->completed, respectively, instead of the prior comparison against the rcu_state fields rsp->gpnum and rsp->completed. Given the old comparison, __rcu_pending() could return 1, resulting in a needless raise_softirq(RCU_SOFTIRQ). This useless work would happen if RCU responded to a scheduling-clock interrupt after the rcu_state fields had been updated, but before the rcu_node fields had been updated. Changing the comparison from the rcu_state fields to the rcu_node fields prevents this useless work from happening. Signed-off-by: Paul E. McKenney Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <12581706991966-git-send-email-> Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'kernel/rcutree.c') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 24bbf2ce0605..9b36d6d7fb97 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -1403,6 +1403,8 @@ EXPORT_SYMBOL_GPL(call_rcu_bh); */ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp) { + struct rcu_node *rnp = rdp->mynode; + rdp->n_rcu_pending++; /* Check for CPU stalls, if enabled. */ @@ -1427,13 +1429,13 @@ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp) } /* Has another RCU grace period completed? */ - if (ACCESS_ONCE(rsp->completed) != rdp->completed) { /* outside lock */ + if (ACCESS_ONCE(rnp->completed) != rdp->completed) { /* outside lock */ rdp->n_rp_gp_completed++; return 1; } /* Has a new RCU grace period started? */ - if (ACCESS_ONCE(rsp->gpnum) != rdp->gpnum) { /* outside lock */ + if (ACCESS_ONCE(rnp->gpnum) != rdp->gpnum) { /* outside lock */ rdp->n_rp_gp_started++; return 1; } -- cgit v1.2.2 From b668c9cf3e58739dac54a1d6f42f2b4bdd980b3e Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sun, 22 Nov 2009 08:53:48 -0800 Subject: rcu: Fix grace-period-stall bug on large systems with CPU hotplug When the last CPU of a given leaf rcu_node structure goes offline, all of the tasks queued on that leaf rcu_node structure (due to having blocked in their current RCU read-side critical sections) are requeued onto the root rcu_node structure. This requeuing is carried out by rcu_preempt_offline_tasks(). However, it is possible that these queued tasks are the only thing preventing the leaf rcu_node structure from reporting a quiescent state up the rcu_node hierarchy. Unfortunately, the old code would fail to do this reporting, resulting in a grace-period stall given the following sequence of events: 1. Kernel built for more than 32 CPUs on 32-bit systems or for more than 64 CPUs on 64-bit systems, so that there is more than one rcu_node structure. (Or CONFIG_RCU_FANOUT is artificially set to a number smaller than CONFIG_NR_CPUS.) 2. The kernel is built with CONFIG_TREE_PREEMPT_RCU. 3. A task running on a CPU associated with a given leaf rcu_node structure blocks while in an RCU read-side critical section -and- that CPU has not yet passed through a quiescent state for the current RCU grace period. This will cause the task to be queued on the leaf rcu_node's blocked_tasks[] array, in particular, on the element of this array corresponding to the current grace period. 4. Each of the remaining CPUs corresponding to this same leaf rcu_node structure pass through a quiescent state. However, the task is still in its RCU read-side critical section, so these quiescent states cannot be reported further up the rcu_node hierarchy. Nevertheless, all bits in the leaf rcu_node structure's ->qsmask field are now zero. 5. Each of the remaining CPUs go offline. (The events in step #4 and #5 can happen in any order as long as each CPU passes through a quiescent state before going offline.) 6. When the last CPU goes offline, __rcu_offline_cpu() will invoke rcu_preempt_offline_tasks(), which will move the task to the root rcu_node structure, but without reporting a quiescent state up the rcu_node hierarchy (and this failure to report a quiescent state is the bug). But because this leaf rcu_node structure's ->qsmask field is already zero and its ->block_tasks[] entries are all empty, force_quiescent_state() will skip this rcu_node structure. Therefore, grace periods are now hung. This patch abstracts some code out of rcu_read_unlock_special(), calling the result task_quiet() by analogy with cpu_quiet(), and invokes task_quiet() from both rcu_read_lock_special() and __rcu_offline_cpu(). Invoking task_quiet() from __rcu_offline_cpu() reports the quiescent state up the rcu_node hierarchy, fixing the bug. This ends up requiring a separate lock_class_key per level of the rcu_node hierarchy, which this patch also provides. Signed-off-by: Paul E. McKenney Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <12589088301770-git-send-email-> Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 40 +++++++++++++++++++++------------------- 1 file changed, 21 insertions(+), 19 deletions(-) (limited to 'kernel/rcutree.c') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 9b36d6d7fb97..b79bfcd28e95 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -51,7 +51,7 @@ /* Data structures. */ -static struct lock_class_key rcu_root_class; +static struct lock_class_key rcu_node_class[NUM_RCU_LVLS]; #define RCU_STATE_INITIALIZER(name) { \ .level = { &name.node[0] }, \ @@ -936,6 +936,7 @@ static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp) { unsigned long flags; unsigned long mask; + int need_quiet = 0; struct rcu_data *rdp = rsp->rda[cpu]; struct rcu_node *rnp; @@ -949,29 +950,30 @@ static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp) spin_lock(&rnp->lock); /* irqs already disabled. */ rnp->qsmaskinit &= ~mask; if (rnp->qsmaskinit != 0) { - spin_unlock(&rnp->lock); /* irqs remain disabled. */ + if (rnp != rdp->mynode) + spin_unlock(&rnp->lock); /* irqs remain disabled. */ break; } - - /* - * If there was a task blocking the current grace period, - * and if all CPUs have checked in, we need to propagate - * the quiescent state up the rcu_node hierarchy. But that - * is inconvenient at the moment due to deadlock issues if - * this should end the current grace period. So set the - * offlined CPU's bit in ->qsmask in order to force the - * next force_quiescent_state() invocation to clean up this - * mess in a deadlock-free manner. - */ - if (rcu_preempt_offline_tasks(rsp, rnp, rdp) && !rnp->qsmask) - rnp->qsmask |= mask; - + if (rnp == rdp->mynode) + need_quiet = rcu_preempt_offline_tasks(rsp, rnp, rdp); + else + spin_unlock(&rnp->lock); /* irqs remain disabled. */ mask = rnp->grpmask; - spin_unlock(&rnp->lock); /* irqs remain disabled. */ rnp = rnp->parent; } while (rnp != NULL); - spin_unlock_irqrestore(&rsp->onofflock, flags); + /* + * We still hold the leaf rcu_node structure lock here, and + * irqs are still disabled. The reason for this subterfuge is + * because invoking task_quiet() with ->onofflock held leads + * to deadlock. + */ + spin_unlock(&rsp->onofflock); /* irqs remain disabled. */ + rnp = rdp->mynode; + if (need_quiet) + task_quiet(rnp, flags); + else + spin_unlock_irqrestore(&rnp->lock, flags); rcu_adopt_orphan_cbs(rsp); } @@ -1731,6 +1733,7 @@ static void __init rcu_init_one(struct rcu_state *rsp) rnp = rsp->level[i]; for (j = 0; j < rsp->levelcnt[i]; j++, rnp++) { spin_lock_init(&rnp->lock); + lockdep_set_class(&rnp->lock, &rcu_node_class[i]); rnp->gpnum = 0; rnp->qsmask = 0; rnp->qsmaskinit = 0; @@ -1753,7 +1756,6 @@ static void __init rcu_init_one(struct rcu_state *rsp) INIT_LIST_HEAD(&rnp->blocked_tasks[1]); } } - lockdep_set_class(&rcu_get_root(rsp)->lock, &rcu_root_class); } /* -- cgit v1.2.2 From 9f680ab41485edfdc96331b70afa7513aa0a7720 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sun, 22 Nov 2009 08:53:49 -0800 Subject: rcu: Eliminate unneeded function wrapping The functions rcu_init() is a wrapper for __rcu_init(), and also sets up the CPU-hotplug notifier for rcu_barrier_cpu_hotplug(). But TINY_RCU doesn't need CPU-hotplug notification, and the rcu_barrier_cpu_hotplug() is a simple wrapper for rcu_cpu_notify(). So push rcu_init() out to kernel/rcutree.c and kernel/rcutiny.c and get rid of the wrapper function rcu_barrier_cpu_hotplug(). Signed-off-by: Paul E. McKenney Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <12589088302320-git-send-email-> Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 17 ++++++++++++++--- 1 file changed, 14 insertions(+), 3 deletions(-) (limited to 'kernel/rcutree.c') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index b79bfcd28e95..e3d3bbddbcd5 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -1644,8 +1644,8 @@ static void __cpuinit rcu_online_cpu(int cpu) /* * Handle CPU online/offline notification events. */ -int __cpuinit rcu_cpu_notify(struct notifier_block *self, - unsigned long action, void *hcpu) +static int __cpuinit rcu_cpu_notify(struct notifier_block *self, + unsigned long action, void *hcpu) { long cpu = (long)hcpu; @@ -1781,8 +1781,10 @@ do { \ } \ } while (0) -void __init __rcu_init(void) +void __init rcu_init(void) { + int i; + rcu_bootup_announce(); #ifdef CONFIG_RCU_CPU_STALL_DETECTOR printk(KERN_INFO "RCU-based detection of stalled CPUs is enabled.\n"); @@ -1791,6 +1793,15 @@ void __init __rcu_init(void) RCU_INIT_FLAVOR(&rcu_bh_state, rcu_bh_data); __rcu_init_preempt(); open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); + + /* + * We don't need protection against CPU-hotplug here because + * this is called early in boot, before either interrupts + * or the scheduler are operational. + */ + cpu_notifier(rcu_cpu_notify, 0); + for_each_online_cpu(i) + rcu_cpu_notify(NULL, CPU_UP_PREPARE, (void *)(long)i); } #include "rcutree_plugin.h" -- cgit v1.2.2 From 6ebb237bece23275d1da149b61a342f0d4d06a08 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sun, 22 Nov 2009 08:53:50 -0800 Subject: rcu: Re-arrange code to reduce #ifdef pain Remove #ifdefs from kernel/rcupdate.c and include/linux/rcupdate.h by moving code to include/linux/rcutiny.h, include/linux/rcutree.h, and kernel/rcutree.c. Also remove some definitions that are no longer used. Signed-off-by: Paul E. McKenney Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <1258908830885-git-send-email-> Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 80 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 80 insertions(+) (limited to 'kernel/rcutree.c') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index e3d3bbddbcd5..4ca7e0292fd8 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -46,6 +46,7 @@ #include #include #include +#include #include "rcutree.h" @@ -79,6 +80,8 @@ DEFINE_PER_CPU(struct rcu_data, rcu_sched_data); struct rcu_state rcu_bh_state = RCU_STATE_INITIALIZER(rcu_bh_state); DEFINE_PER_CPU(struct rcu_data, rcu_bh_data); +static int rcu_scheduler_active __read_mostly; + /* * Return true if an RCU grace period is in progress. The ACCESS_ONCE()s @@ -1396,6 +1399,68 @@ void call_rcu_bh(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) } EXPORT_SYMBOL_GPL(call_rcu_bh); +/** + * synchronize_sched - wait until an rcu-sched grace period has elapsed. + * + * Control will return to the caller some time after a full rcu-sched + * grace period has elapsed, in other words after all currently executing + * rcu-sched read-side critical sections have completed. These read-side + * critical sections are delimited by rcu_read_lock_sched() and + * rcu_read_unlock_sched(), and may be nested. Note that preempt_disable(), + * local_irq_disable(), and so on may be used in place of + * rcu_read_lock_sched(). + * + * This means that all preempt_disable code sequences, including NMI and + * hardware-interrupt handlers, in progress on entry will have completed + * before this primitive returns. However, this does not guarantee that + * softirq handlers will have completed, since in some kernels, these + * handlers can run in process context, and can block. + * + * This primitive provides the guarantees made by the (now removed) + * synchronize_kernel() API. In contrast, synchronize_rcu() only + * guarantees that rcu_read_lock() sections will have completed. + * In "classic RCU", these two guarantees happen to be one and + * the same, but can differ in realtime RCU implementations. + */ +void synchronize_sched(void) +{ + struct rcu_synchronize rcu; + + if (rcu_blocking_is_gp()) + return; + + init_completion(&rcu.completion); + /* Will wake me after RCU finished. */ + call_rcu_sched(&rcu.head, wakeme_after_rcu); + /* Wait for it. */ + wait_for_completion(&rcu.completion); +} +EXPORT_SYMBOL_GPL(synchronize_sched); + +/** + * synchronize_rcu_bh - wait until an rcu_bh grace period has elapsed. + * + * Control will return to the caller some time after a full rcu_bh grace + * period has elapsed, in other words after all currently executing rcu_bh + * read-side critical sections have completed. RCU read-side critical + * sections are delimited by rcu_read_lock_bh() and rcu_read_unlock_bh(), + * and may be nested. + */ +void synchronize_rcu_bh(void) +{ + struct rcu_synchronize rcu; + + if (rcu_blocking_is_gp()) + return; + + init_completion(&rcu.completion); + /* Will wake me after RCU finished. */ + call_rcu_bh(&rcu.head, wakeme_after_rcu); + /* Wait for it. */ + wait_for_completion(&rcu.completion); +} +EXPORT_SYMBOL_GPL(synchronize_rcu_bh); + /* * Check to see if there is any immediate RCU-related work to be done * by the current CPU, for the specified type of RCU, returning 1 if so. @@ -1480,6 +1545,21 @@ int rcu_needs_cpu(int cpu) rcu_preempt_needs_cpu(cpu); } +/* + * This function is invoked towards the end of the scheduler's initialization + * process. Before this is called, the idle task might contain + * RCU read-side critical sections (during which time, this idle + * task is booting the system). After this function is called, the + * idle tasks are prohibited from containing RCU read-side critical + * sections. + */ +void rcu_scheduler_starting(void) +{ + WARN_ON(num_online_cpus() != 1); + WARN_ON(nr_context_switches() > 0); + rcu_scheduler_active = 1; +} + static DEFINE_PER_CPU(struct rcu_head, rcu_barrier_head) = {NULL}; static atomic_t rcu_barrier_cpu_count; static DEFINE_MUTEX(rcu_barrier_mutex); -- cgit v1.2.2 From d3f6bad3911736e44ba11f3f3f6ac4e8c837fdfc Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 2 Dec 2009 12:10:13 -0800 Subject: rcu: Rename "quiet" functions The number of "quiet" functions has grown recently, and the names are no longer very descriptive. The point of all of these functions is to do some portion of the task of reporting a quiescent state, so rename them accordingly: o cpu_quiet() becomes rcu_report_qs_rdp(), which reports a quiescent state to the per-CPU rcu_data structure. If this turns out to be a new quiescent state for this grace period, then rcu_report_qs_rnp() will be invoked to propagate the quiescent state up the rcu_node hierarchy. o cpu_quiet_msk() becomes rcu_report_qs_rnp(), which reports a quiescent state for a given CPU (or possibly a set of CPUs) up the rcu_node hierarchy. o cpu_quiet_msk_finish() becomes rcu_report_qs_rsp(), which reports a full set of quiescent states to the global rcu_state structure. o task_quiet() becomes rcu_report_unblock_qs_rnp(), which reports a quiescent state due to a task exiting an RCU read-side critical section that had previously blocked in that same critical section. As indicated by the new name, this type of quiescent state is reported up the rcu_node hierarchy (using rcu_report_qs_rnp() to do so). Signed-off-by: Paul E. McKenney Acked-by: Josh Triplett Acked-by: Lai Jiangshan Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <12597846163698-git-send-email-> Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 66 +++++++++++++++++++++++++++++++------------------------- 1 file changed, 37 insertions(+), 29 deletions(-) (limited to 'kernel/rcutree.c') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 4ca7e0292fd8..a9f51031d3e8 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -740,11 +740,13 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags) } /* - * Clean up after the prior grace period and let rcu_start_gp() start up - * the next grace period if one is needed. Note that the caller must - * hold rnp->lock, as required by rcu_start_gp(), which will release it. + * Report a full set of quiescent states to the specified rcu_state + * data structure. This involves cleaning up after the prior grace + * period and letting rcu_start_gp() start up the next grace period + * if one is needed. Note that the caller must hold rnp->lock, as + * required by rcu_start_gp(), which will release it. */ -static void cpu_quiet_msk_finish(struct rcu_state *rsp, unsigned long flags) +static void rcu_report_qs_rsp(struct rcu_state *rsp, unsigned long flags) __releases(rcu_get_root(rsp)->lock) { WARN_ON_ONCE(!rcu_gp_in_progress(rsp)); @@ -754,15 +756,16 @@ static void cpu_quiet_msk_finish(struct rcu_state *rsp, unsigned long flags) } /* - * Similar to cpu_quiet(), for which it is a helper function. Allows - * a group of CPUs to be quieted at one go, though all the CPUs in the - * group must be represented by the same leaf rcu_node structure. - * That structure's lock must be held upon entry, and it is released - * before return. + * Similar to rcu_report_qs_rdp(), for which it is a helper function. + * Allows quiescent states for a group of CPUs to be reported at one go + * to the specified rcu_node structure, though all the CPUs in the group + * must be represented by the same rcu_node structure (which need not be + * a leaf rcu_node structure, though it often will be). That structure's + * lock must be held upon entry, and it is released before return. */ static void -cpu_quiet_msk(unsigned long mask, struct rcu_state *rsp, struct rcu_node *rnp, - unsigned long flags) +rcu_report_qs_rnp(unsigned long mask, struct rcu_state *rsp, + struct rcu_node *rnp, unsigned long flags) __releases(rnp->lock) { struct rcu_node *rnp_c; @@ -798,21 +801,23 @@ cpu_quiet_msk(unsigned long mask, struct rcu_state *rsp, struct rcu_node *rnp, /* * Get here if we are the last CPU to pass through a quiescent - * state for this grace period. Invoke cpu_quiet_msk_finish() + * state for this grace period. Invoke rcu_report_qs_rsp() * to clean up and start the next grace period if one is needed. */ - cpu_quiet_msk_finish(rsp, flags); /* releases rnp->lock. */ + rcu_report_qs_rsp(rsp, flags); /* releases rnp->lock. */ } /* - * Record a quiescent state for the specified CPU, which must either be - * the current CPU. The lastcomp argument is used to make sure we are - * still in the grace period of interest. We don't want to end the current - * grace period based on quiescent states detected in an earlier grace - * period! + * Record a quiescent state for the specified CPU to that CPU's rcu_data + * structure. This must be either called from the specified CPU, or + * called when the specified CPU is known to be offline (and when it is + * also known that no other CPU is concurrently trying to help the offline + * CPU). The lastcomp argument is used to make sure we are still in the + * grace period of interest. We don't want to end the current grace period + * based on quiescent states detected in an earlier grace period! */ static void -cpu_quiet(int cpu, struct rcu_state *rsp, struct rcu_data *rdp, long lastcomp) +rcu_report_qs_rdp(int cpu, struct rcu_state *rsp, struct rcu_data *rdp, long lastcomp) { unsigned long flags; unsigned long mask; @@ -827,8 +832,8 @@ cpu_quiet(int cpu, struct rcu_state *rsp, struct rcu_data *rdp, long lastcomp) * The race with GP start is resolved by the fact that we * hold the leaf rcu_node lock, so that the per-CPU bits * cannot yet be initialized -- so we would simply find our - * CPU's bit already cleared in cpu_quiet_msk() if this race - * occurred. + * CPU's bit already cleared in rcu_report_qs_rnp() if this + * race occurred. */ rdp->passed_quiesc = 0; /* try again later! */ spin_unlock_irqrestore(&rnp->lock, flags); @@ -846,7 +851,7 @@ cpu_quiet(int cpu, struct rcu_state *rsp, struct rcu_data *rdp, long lastcomp) */ rdp->nxttail[RCU_NEXT_READY_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; - cpu_quiet_msk(mask, rsp, rnp, flags); /* releases rnp->lock */ + rcu_report_qs_rnp(mask, rsp, rnp, flags); /* rlses rnp->lock */ } } @@ -877,8 +882,11 @@ rcu_check_quiescent_state(struct rcu_state *rsp, struct rcu_data *rdp) if (!rdp->passed_quiesc) return; - /* Tell RCU we are done (but cpu_quiet() will be the judge of that). */ - cpu_quiet(rdp->cpu, rsp, rdp, rdp->passed_quiesc_completed); + /* + * Tell RCU we are done (but rcu_report_qs_rdp() will be the + * judge of that). + */ + rcu_report_qs_rdp(rdp->cpu, rsp, rdp, rdp->passed_quiesc_completed); } #ifdef CONFIG_HOTPLUG_CPU @@ -968,13 +976,13 @@ static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp) /* * We still hold the leaf rcu_node structure lock here, and * irqs are still disabled. The reason for this subterfuge is - * because invoking task_quiet() with ->onofflock held leads - * to deadlock. + * because invoking rcu_report_unblock_qs_rnp() with ->onofflock + * held leads to deadlock. */ spin_unlock(&rsp->onofflock); /* irqs remain disabled. */ rnp = rdp->mynode; if (need_quiet) - task_quiet(rnp, flags); + rcu_report_unblock_qs_rnp(rnp, flags); else spin_unlock_irqrestore(&rnp->lock, flags); @@ -1164,8 +1172,8 @@ static int rcu_process_dyntick(struct rcu_state *rsp, long lastcomp, } if (mask != 0 && rnp->completed == lastcomp) { - /* cpu_quiet_msk() releases rnp->lock. */ - cpu_quiet_msk(mask, rsp, rnp, flags); + /* rcu_report_qs_rnp() releases rnp->lock. */ + rcu_report_qs_rnp(mask, rsp, rnp, flags); continue; } spin_unlock_irqrestore(&rnp->lock, flags); -- cgit v1.2.2 From cf244dc01bf68e1ad338b82447f8686d24ea4435 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 2 Dec 2009 12:10:14 -0800 Subject: rcu: Enable fourth level of TREE_RCU hierarchy Enable a fourth level of rcu_node hierarchy for TREE_RCU and TREE_PREEMPT_RCU. This is for stress-testing and experiemental purposes only, although in theory this would enable 16,777,216 CPUs on 64-bit systems, though only 1,048,576 CPUs on 32-bit systems. Normal experimental use of this fourth level will normally set CONFIG_RCU_FANOUT=2, requiring a 16-CPU system, though the more adventurous (and more fortunate) experimenters may wish to chose CONFIG_RCU_FANOUT=3 for 81-CPU systems or even CONFIG_RCU_FANOUT=4 for 256-CPU systems. Signed-off-by: Paul E. McKenney Acked-by: Josh Triplett Acked-by: Lai Jiangshan Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <12597846161257-git-send-email-> Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'kernel/rcutree.c') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index a9f51031d3e8..d47e03e5792a 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -60,7 +60,8 @@ static struct lock_class_key rcu_node_class[NUM_RCU_LVLS]; NUM_RCU_LVL_0, /* root of hierarchy. */ \ NUM_RCU_LVL_1, \ NUM_RCU_LVL_2, \ - NUM_RCU_LVL_3, /* == MAX_RCU_LVLS */ \ + NUM_RCU_LVL_3, \ + NUM_RCU_LVL_4, /* == MAX_RCU_LVLS */ \ }, \ .signaled = RCU_GP_IDLE, \ .gpnum = -300, \ @@ -1877,6 +1878,9 @@ void __init rcu_init(void) #ifdef CONFIG_RCU_CPU_STALL_DETECTOR printk(KERN_INFO "RCU-based detection of stalled CPUs is enabled.\n"); #endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ +#if NUM_RCU_LVL_4 != 0 + printk(KERN_INFO "Experimental four-level hierarchy is enabled.\n"); +#endif /* #if NUM_RCU_LVL_4 != 0 */ RCU_INIT_FLAVOR(&rcu_sched_state, rcu_sched_data); RCU_INIT_FLAVOR(&rcu_bh_state, rcu_bh_data); __rcu_init_preempt(); -- cgit v1.2.2 From d9a3da0699b24a589b27a61e1a5b5bd30d9db669 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 2 Dec 2009 12:10:15 -0800 Subject: rcu: Add expedited grace-period support for preemptible RCU Implement an synchronize_rcu_expedited() for preemptible RCU that actually is expedited. This uses synchronize_sched_expedited() to force all threads currently running in a preemptible-RCU read-side critical section onto the appropriate ->blocked_tasks[] list, then takes a snapshot of all of these lists and waits for them to drain. Signed-off-by: Paul E. McKenney Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <1259784616158-git-send-email-> Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'kernel/rcutree.c') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index d47e03e5792a..53ae9598f798 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -948,7 +948,7 @@ static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp) { unsigned long flags; unsigned long mask; - int need_quiet = 0; + int need_report = 0; struct rcu_data *rdp = rsp->rda[cpu]; struct rcu_node *rnp; @@ -967,7 +967,7 @@ static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp) break; } if (rnp == rdp->mynode) - need_quiet = rcu_preempt_offline_tasks(rsp, rnp, rdp); + need_report = rcu_preempt_offline_tasks(rsp, rnp, rdp); else spin_unlock(&rnp->lock); /* irqs remain disabled. */ mask = rnp->grpmask; @@ -982,10 +982,12 @@ static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp) */ spin_unlock(&rsp->onofflock); /* irqs remain disabled. */ rnp = rdp->mynode; - if (need_quiet) + if (need_report & RCU_OFL_TASKS_NORM_GP) rcu_report_unblock_qs_rnp(rnp, flags); else spin_unlock_irqrestore(&rnp->lock, flags); + if (need_report & RCU_OFL_TASKS_EXP_GP) + rcu_report_exp_rnp(rsp, rnp); rcu_adopt_orphan_cbs(rsp); } @@ -1843,6 +1845,8 @@ static void __init rcu_init_one(struct rcu_state *rsp) rnp->level = i; INIT_LIST_HEAD(&rnp->blocked_tasks[0]); INIT_LIST_HEAD(&rnp->blocked_tasks[1]); + INIT_LIST_HEAD(&rnp->blocked_tasks[2]); + INIT_LIST_HEAD(&rnp->blocked_tasks[3]); } } } -- cgit v1.2.2