diff options
Diffstat (limited to 'kernel/rcu/tree.c')
| -rw-r--r-- | kernel/rcu/tree.c | 452 |
1 files changed, 339 insertions, 113 deletions
diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index d91c9156fab2..550193a9ce76 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -1,12 +1,12 @@ // SPDX-License-Identifier: GPL-2.0+ /* - * Read-Copy Update mechanism for mutual exclusion + * Read-Copy Update mechanism for mutual exclusion (tree-based version) * * Copyright IBM Corporation, 2008 * * Authors: Dipankar Sarma <dipankar@in.ibm.com> * Manfred Spraul <manfred@colorfullife.com> - * Paul E. McKenney <paulmck@linux.ibm.com> Hierarchical version + * Paul E. McKenney <paulmck@linux.ibm.com> * * Based on the original work by Paul McKenney <paulmck@linux.ibm.com> * and inputs from Rusty Russell, Andrea Arcangeli and Andi Kleen. @@ -150,6 +150,7 @@ static void rcu_boost_kthread_setaffinity(struct rcu_node *rnp, int outgoingcpu) static void invoke_rcu_core(void); static void rcu_report_exp_rdp(struct rcu_data *rdp); static void sync_sched_exp_online_cleanup(int cpu); +static void check_cb_ovld_locked(struct rcu_data *rdp, struct rcu_node *rnp); /* rcuc/rcub kthread realtime priority */ static int kthread_prio = IS_ENABLED(CONFIG_RCU_BOOST) ? 1 : 0; @@ -342,14 +343,17 @@ bool rcu_eqs_special_set(int cpu) { int old; int new; + int new_old; struct rcu_data *rdp = &per_cpu(rcu_data, cpu); + new_old = atomic_read(&rdp->dynticks); do { - old = atomic_read(&rdp->dynticks); + old = new_old; if (old & RCU_DYNTICK_CTRL_CTR) return false; new = old | RCU_DYNTICK_CTRL_MASK; - } while (atomic_cmpxchg(&rdp->dynticks, old, new) != old); + new_old = atomic_cmpxchg(&rdp->dynticks, old, new); + } while (new_old != old); return true; } @@ -410,10 +414,15 @@ static long blimit = DEFAULT_RCU_BLIMIT; static long qhimark = DEFAULT_RCU_QHIMARK; #define DEFAULT_RCU_QLOMARK 100 /* Once only this many pending, use blimit. */ static long qlowmark = DEFAULT_RCU_QLOMARK; +#define DEFAULT_RCU_QOVLD_MULT 2 +#define DEFAULT_RCU_QOVLD (DEFAULT_RCU_QOVLD_MULT * DEFAULT_RCU_QHIMARK) +static long qovld = DEFAULT_RCU_QOVLD; /* If this many pending, hammer QS. */ +static long qovld_calc = -1; /* No pre-initialization lock acquisitions! */ module_param(blimit, long, 0444); module_param(qhimark, long, 0444); module_param(qlowmark, long, 0444); +module_param(qovld, long, 0444); static ulong jiffies_till_first_fqs = ULONG_MAX; static ulong jiffies_till_next_fqs = ULONG_MAX; @@ -818,11 +827,12 @@ static __always_inline void rcu_nmi_enter_common(bool irq) incby = 1; } else if (tick_nohz_full_cpu(rdp->cpu) && rdp->dynticks_nmi_nesting == DYNTICK_IRQ_NONIDLE && - READ_ONCE(rdp->rcu_urgent_qs) && !rdp->rcu_forced_tick) { + READ_ONCE(rdp->rcu_urgent_qs) && + !READ_ONCE(rdp->rcu_forced_tick)) { raw_spin_lock_rcu_node(rdp->mynode); // Recheck under lock. if (rdp->rcu_urgent_qs && !rdp->rcu_forced_tick) { - rdp->rcu_forced_tick = true; + WRITE_ONCE(rdp->rcu_forced_tick, true); tick_dep_set_cpu(rdp->cpu, TICK_DEP_BIT_RCU); } raw_spin_unlock_rcu_node(rdp->mynode); @@ -899,7 +909,7 @@ static void rcu_disable_urgency_upon_qs(struct rcu_data *rdp) WRITE_ONCE(rdp->rcu_need_heavy_qs, false); if (tick_nohz_full_cpu(rdp->cpu) && rdp->rcu_forced_tick) { tick_dep_clear_cpu(rdp->cpu, TICK_DEP_BIT_RCU); - rdp->rcu_forced_tick = false; + WRITE_ONCE(rdp->rcu_forced_tick, false); } } @@ -1072,7 +1082,8 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp) rnhqp = &per_cpu(rcu_data.rcu_need_heavy_qs, rdp->cpu); if (!READ_ONCE(*rnhqp) && (time_after(jiffies, rcu_state.gp_start + jtsq * 2) || - time_after(jiffies, rcu_state.jiffies_resched))) { + time_after(jiffies, rcu_state.jiffies_resched) || + rcu_state.cbovld)) { WRITE_ONCE(*rnhqp, true); /* Store rcu_need_heavy_qs before rcu_urgent_qs. */ smp_store_release(ruqp, true); @@ -1089,8 +1100,8 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp) * So hit them over the head with the resched_cpu() hammer! */ if (tick_nohz_full_cpu(rdp->cpu) && - time_after(jiffies, - READ_ONCE(rdp->last_fqs_resched) + jtsq * 3)) { + (time_after(jiffies, READ_ONCE(rdp->last_fqs_resched) + jtsq * 3) || + rcu_state.cbovld)) { WRITE_ONCE(*ruqp, true); resched_cpu(rdp->cpu); WRITE_ONCE(rdp->last_fqs_resched, jiffies); @@ -1126,8 +1137,9 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp) static void trace_rcu_this_gp(struct rcu_node *rnp, struct rcu_data *rdp, unsigned long gp_seq_req, const char *s) { - trace_rcu_future_grace_period(rcu_state.name, rnp->gp_seq, gp_seq_req, - rnp->level, rnp->grplo, rnp->grphi, s); + trace_rcu_future_grace_period(rcu_state.name, READ_ONCE(rnp->gp_seq), + gp_seq_req, rnp->level, + rnp->grplo, rnp->grphi, s); } /* @@ -1174,7 +1186,7 @@ static bool rcu_start_this_gp(struct rcu_node *rnp_start, struct rcu_data *rdp, TPS("Prestarted")); goto unlock_out; } - rnp->gp_seq_needed = gp_seq_req; + WRITE_ONCE(rnp->gp_seq_needed, gp_seq_req); if (rcu_seq_state(rcu_seq_current(&rnp->gp_seq))) { /* * We just marked the leaf or internal node, and a @@ -1199,18 +1211,18 @@ static bool rcu_start_this_gp(struct rcu_node *rnp_start, struct rcu_data *rdp, } trace_rcu_this_gp(rnp, rdp, gp_seq_req, TPS("Startedroot")); WRITE_ONCE(rcu_state.gp_flags, rcu_state.gp_flags | RCU_GP_FLAG_INIT); - rcu_state.gp_req_activity = jiffies; - if (!rcu_state.gp_kthread) { + WRITE_ONCE(rcu_state.gp_req_activity, jiffies); + if (!READ_ONCE(rcu_state.gp_kthread)) { trace_rcu_this_gp(rnp, rdp, gp_seq_req, TPS("NoGPkthread")); goto unlock_out; } - trace_rcu_grace_period(rcu_state.name, READ_ONCE(rcu_state.gp_seq), TPS("newreq")); + trace_rcu_grace_period(rcu_state.name, rcu_state.gp_seq, TPS("newreq")); ret = true; /* Caller must wake GP kthread. */ unlock_out: /* Push furthest requested GP to leaf node and rcu_data structure. */ if (ULONG_CMP_LT(gp_seq_req, rnp->gp_seq_needed)) { - rnp_start->gp_seq_needed = rnp->gp_seq_needed; - rdp->gp_seq_needed = rnp->gp_seq_needed; + WRITE_ONCE(rnp_start->gp_seq_needed, rnp->gp_seq_needed); + WRITE_ONCE(rdp->gp_seq_needed, rnp->gp_seq_needed); } if (rnp != rnp_start) raw_spin_unlock_rcu_node(rnp); @@ -1235,12 +1247,13 @@ static bool rcu_future_gp_cleanup(struct rcu_node *rnp) } /* - * Awaken the grace-period kthread. Don't do a self-awaken (unless in - * an interrupt or softirq handler), and don't bother awakening when there - * is nothing for the grace-period kthread to do (as in several CPUs raced - * to awaken, and we lost), and finally don't try to awaken a kthread that - * has not yet been created. If all those checks are passed, track some - * debug information and awaken. + * Awaken the grace-period kthread. Don't do a self-awaken (unless in an + * interrupt or softirq handler, in which case we just might immediately + * sleep upon return, resulting in a grace-period hang), and don't bother + * awakening when there is nothing for the grace-period kthread to do + * (as in several CPUs raced to awaken, we lost), and finally don't try + * to awaken a kthread that has not yet been created. If all those checks + * are passed, track some debug information and awaken. * * So why do the self-wakeup when in an interrupt or softirq handler * in the grace-period kthread's context? Because the kthread might have @@ -1250,10 +1263,10 @@ static bool rcu_future_gp_cleanup(struct rcu_node *rnp) */ static void rcu_gp_kthread_wake(void) { - if ((current == rcu_state.gp_kthread && - !in_irq() && !in_serving_softirq()) || - !READ_ONCE(rcu_state.gp_flags) || - !rcu_state.gp_kthread) + struct task_struct *t = READ_ONCE(rcu_state.gp_kthread); + + if ((current == t && !in_irq() && !in_serving_softirq()) || + !READ_ONCE(rcu_state.gp_flags) || !t) return; WRITE_ONCE(rcu_state.gp_wake_time, jiffies); WRITE_ONCE(rcu_state.gp_wake_seq, READ_ONCE(rcu_state.gp_seq)); @@ -1321,7 +1334,7 @@ static void rcu_accelerate_cbs_unlocked(struct rcu_node *rnp, rcu_lockdep_assert_cblist_protected(rdp); c = rcu_seq_snap(&rcu_state.gp_seq); - if (!rdp->gpwrap && ULONG_CMP_GE(rdp->gp_seq_needed, c)) { + if (!READ_ONCE(rdp->gpwrap) && ULONG_CMP_GE(rdp->gp_seq_needed, c)) { /* Old request still live, so mark recent callbacks. */ (void)rcu_segcblist_accelerate(&rdp->cblist, c); return; @@ -1386,7 +1399,7 @@ static void __maybe_unused rcu_advance_cbs_nowake(struct rcu_node *rnp, static bool __note_gp_changes(struct rcu_node *rnp, struct rcu_data *rdp) { bool ret = false; - bool need_gp; + bool need_qs; const bool offloaded = IS_ENABLED(CONFIG_RCU_NOCB_CPU) && rcu_segcblist_is_offloaded(&rdp->cblist); @@ -1400,10 +1413,13 @@ static bool __note_gp_changes(struct rcu_node *rnp, struct rcu_data *rdp) unlikely(READ_ONCE(rdp->gpwrap))) { if (!offloaded) ret = rcu_advance_cbs(rnp, rdp); /* Advance CBs. */ + rdp->core_needs_qs = false; trace_rcu_grace_period(rcu_state.name, rdp->gp_seq, TPS("cpuend")); } else { if (!offloaded) ret = rcu_accelerate_cbs(rnp, rdp); /* Recent CBs. */ + if (rdp->core_needs_qs) + rdp->core_needs_qs = !!(rnp->qsmask & rdp->grpmask); } /* Now handle the beginnings of any new-to-this-CPU grace periods. */ @@ -1415,14 +1431,14 @@ static bool __note_gp_changes(struct rcu_node *rnp, struct rcu_data *rdp) * go looking for one. */ trace_rcu_grace_period(rcu_state.name, rnp->gp_seq, TPS("cpustart")); - need_gp = !!(rnp->qsmask & rdp->grpmask); - rdp->cpu_no_qs.b.norm = need_gp; - rdp->core_needs_qs = need_gp; + need_qs = !!(rnp->qsmask & rdp->grpmask); + rdp->cpu_no_qs.b.norm = need_qs; + rdp->core_needs_qs = need_qs; zero_cpu_stall_ticks(rdp); } rdp->gp_seq = rnp->gp_seq; /* Remember new grace-period state. */ if (ULONG_CMP_LT(rdp->gp_seq_needed, rnp->gp_seq_needed) || rdp->gpwrap) - rdp->gp_seq_needed = rnp->gp_seq_needed; + WRITE_ONCE(rdp->gp_seq_needed, rnp->gp_seq_needed); WRITE_ONCE(rdp->gpwrap, false); rcu_gpnum_ovf(rnp, rdp); return ret; @@ -1651,8 +1667,7 @@ static void rcu_gp_fqs_loop(void) WRITE_ONCE(rcu_state.jiffies_kick_kthreads, jiffies + (j ? 3 * j : 2)); } - trace_rcu_grace_period(rcu_state.name, - READ_ONCE(rcu_state.gp_seq), + trace_rcu_grace_period(rcu_state.name, rcu_state.gp_seq, TPS("fqswait")); rcu_state.gp_state = RCU_GP_WAIT_FQS; ret = swait_event_idle_timeout_exclusive( @@ -1666,13 +1681,11 @@ static void rcu_gp_fqs_loop(void) /* If time for quiescent-state forcing, do it. */ if (ULONG_CMP_GE(jiffies, rcu_state.jiffies_force_qs) || (gf & RCU_GP_FLAG_FQS)) { - trace_rcu_grace_period(rcu_state.name, - READ_ONCE(rcu_state.gp_seq), + trace_rcu_grace_period(rcu_state.name, rcu_state.gp_seq, TPS("fqsstart")); rcu_gp_fqs(first_gp_fqs); first_gp_fqs = false; - trace_rcu_grace_period(rcu_state.name, - READ_ONCE(rcu_state.gp_seq), + trace_rcu_grace_period(rcu_state.name, rcu_state.gp_seq, TPS("fqsend")); cond_resched_tasks_rcu_qs(); WRITE_ONCE(rcu_state.gp_activity, jiffies); @@ -1683,8 +1696,7 @@ static void rcu_gp_fqs_loop(void) cond_resched_tasks_rcu_qs(); WRITE_ONCE(rcu_state.gp_activity, jiffies); WARN_ON(signal_pending(current)); - trace_rcu_grace_period(rcu_state.name, - READ_ONCE(rcu_state.gp_seq), + trace_rcu_grace_period(rcu_state.name, rcu_state.gp_seq, TPS("fqswaitsig")); ret = 1; /* Keep old FQS timing. */ j = jiffies; @@ -1701,8 +1713,9 @@ static void rcu_gp_fqs_loop(void) */ static void rcu_gp_cleanup(void) { - unsigned long gp_duration; + int cpu; bool needgp = false; + unsigned long gp_duration; unsigned long new_gp_seq; bool offloaded; struct rcu_data *rdp; @@ -1748,6 +1761,12 @@ static void rcu_gp_cleanup(void) needgp = __note_gp_changes(rnp, rdp) || needgp; /* smp_mb() provided by prior unlock-lock pair. */ needgp = rcu_future_gp_cleanup(rnp) || needgp; + // Reset overload indication for CPUs no longer overloaded + if (rcu_is_leaf_node(rnp)) + for_each_leaf_node_cpu_mask(rnp, cpu, rnp->cbovldmask) { + rdp = per_cpu_ptr(&rcu_data, cpu); + check_cb_ovld_locked(rdp, rnp); + } sq = rcu_nocb_gp_get(rnp); raw_spin_unlock_irq_rcu_node(rnp); rcu_nocb_gp_cleanup(sq); @@ -1774,9 +1793,9 @@ static void rcu_gp_cleanup(void) rcu_segcblist_is_offloaded(&rdp->cblist); if ((offloaded || !rcu_accelerate_cbs(rnp, rdp)) && needgp) { WRITE_ONCE(rcu_state.gp_flags, RCU_GP_FLAG_INIT); - rcu_state.gp_req_activity = jiffies; + WRITE_ONCE(rcu_state.gp_req_activity, jiffies); trace_rcu_grace_period(rcu_state.name, - READ_ONCE(rcu_state.gp_seq), + rcu_state.gp_seq, TPS("newreq")); } else { WRITE_ONCE(rcu_state.gp_flags, @@ -1795,8 +1814,7 @@ static int __noreturn rcu_gp_kthread(void *unused) /* Handle grace-period start. */ for (;;) { - trace_rcu_grace_period(rcu_state.name, - READ_ONCE(rcu_state.gp_seq), + trace_rcu_grace_period(rcu_state.name, rcu_state.gp_seq, TPS("reqwait")); rcu_state.gp_state = RCU_GP_WAIT_GPS; swait_event_idle_exclusive(rcu_state.gp_wq, @@ -1809,8 +1827,7 @@ static int __noreturn rcu_gp_kthread(void *unused) cond_resched_tasks_rcu_qs(); WRITE_ONCE(rcu_state.gp_activity, jiffies); WARN_ON(signal_pending(current)); - trace_rcu_grace_period(rcu_state.name, - READ_ONCE(rcu_state.gp_seq), + trace_rcu_grace_period(rcu_state.name, rcu_state.gp_seq, TPS("reqwaitsig")); } @@ -1881,7 +1898,7 @@ static void rcu_report_qs_rnp(unsigned long mask, struct rcu_node *rnp, WARN_ON_ONCE(oldmask); /* Any child must be all zeroed! */ WARN_ON_ONCE(!rcu_is_leaf_node(rnp) && rcu_preempt_blocked_readers_cgp(rnp)); - rnp->qsmask &= ~mask; + WRITE_ONCE(rnp->qsmask, rnp->qsmask & ~mask); trace_rcu_quiescent_state_report(rcu_state.name, rnp->gp_seq, mask, rnp->qsmask, rnp->level, rnp->grplo, rnp->grphi, @@ -1904,7 +1921,7 @@ static void rcu_report_qs_rnp(unsigned long mask, struct rcu_node *rnp, rnp_c = rnp; rnp = rnp->parent; raw_spin_lock_irqsave_rcu_node(rnp, flags); - oldmask = rnp_c->qsmask; + oldmask = READ_ONCE(rnp_c->qsmask); } /* @@ -1987,6 +2004,8 @@ rcu_report_qs_rdp(int cpu, struct rcu_data *rdp) return; } mask = rdp->grpmask; + if (rdp->cpu == smp_processor_id()) + rdp->core_needs_qs = false; if ((rnp->qsmask & mask) == 0) { raw_spin_unlock_irqrestore_rcu_node(rnp, flags); } else { @@ -2052,7 +2071,7 @@ int rcutree_dying_cpu(unsigned int cpu) return 0; blkd = !!(rnp->qsmask & rdp->grpmask); - trace_rcu_grace_period(rcu_state.name, rnp->gp_seq, + trace_rcu_grace_period(rcu_state.name, READ_ONCE(rnp->gp_seq), blkd ? TPS("cpuofl") : TPS("cpuofl-bgp")); return 0; } @@ -2294,10 +2313,13 @@ static void force_qs_rnp(int (*f)(struct rcu_data *rdp)) struct rcu_data *rdp; struct rcu_node *rnp; + rcu_state.cbovld = rcu_state.cbovldnext; + rcu_state.cbovldnext = false; rcu_for_each_leaf_node(rnp) { cond_resched_tasks_rcu_qs(); mask = 0; raw_spin_lock_irqsave_rcu_node(rnp, flags); + rcu_state.cbovldnext |= !!rnp->cbovldmask; if (rnp->qsmask == 0) { if (!IS_ENABLED(CONFIG_PREEMPT_RCU) || rcu_preempt_blocked_readers_cgp(rnp)) { @@ -2579,11 +2601,48 @@ static void rcu_leak_callback(struct rcu_head *rhp) } /* - * Helper function for call_rcu() and friends. The cpu argument will - * normally be -1, indicating "currently running CPU". It may specify - * a CPU only if that CPU is a no-CBs CPU. Currently, only rcu_barrier() - * is expected to specify a CPU. + * Check and if necessary update the leaf rcu_node structure's + * ->cbovldmask bit corresponding to the current CPU based on that CPU's + * number of queued RCU callbacks. The caller must hold the leaf rcu_node + * structure's ->lock. */ +static void check_cb_ovld_locked(struct rcu_data *rdp, struct rcu_node *rnp) +{ + raw_lockdep_assert_held_rcu_node(rnp); + if (qovld_calc <= 0) + return; // Early boot and wildcard value set. + if (rcu_segcblist_n_cbs(&rdp->cblist) >= qovld_calc) + WRITE_ONCE(rnp->cbovldmask, rnp->cbovldmask | rdp->grpmask); + else + WRITE_ONCE(rnp->cbovldmask, rnp->cbovldmask & ~rdp->grpmask); +} + +/* + * Check and if necessary update the leaf rcu_node structure's + * ->cbovldmask bit corresponding to the current CPU based on that CPU's + * number of queued RCU callbacks. No locks need be held, but the + * caller must have disabled interrupts. + * + * Note that this function ignores the possibility that there are a lot + * of callbacks all of which have already seen the end of their respective + * grace periods. This omission is due to the need for no-CBs CPUs to + * be holding ->nocb_lock to do this check, which is too heavy for a + * common-case operation. + */ +static void check_cb_ovld(struct rcu_data *rdp) +{ + struct rcu_node *const rnp = rdp->mynode; + + if (qovld_calc <= 0 || + ((rcu_segcblist_n_cbs(&rdp->cblist) >= qovld_calc) == + !!(READ_ONCE(rnp->cbovldmask) & rdp->grpmask))) + return; // Early boot wildcard value or already set correctly. + raw_spin_lock_rcu_node(rnp); + check_cb_ovld_locked(rdp, rnp); + raw_spin_unlock_rcu_node(rnp); +} + +/* Helper function for call_rcu() and friends. */ static void __call_rcu(struct rcu_head *head, rcu_callback_t func) { @@ -2621,9 +2680,10 @@ __call_rcu(struct rcu_head *head, rcu_callback_t func) rcu_segcblist_init(&rdp->cblist); } + check_cb_ovld(rdp); if (rcu_nocb_try_bypass(rdp, head, &was_alldone, flags)) return; // Enqueued onto ->nocb_bypass, so just leave. - /* If we get here, rcu_nocb_try_bypass() acquired ->nocb_lock. */ + // If no-CBs CPU gets here, rcu_nocb_try_bypass() acquired ->nocb_lock. rcu_segcblist_enqueue(&rdp->cblist, head); if (__is_kfree_rcu_offset((unsigned long)func)) trace_rcu_kfree_callback(rcu_state.name, head, @@ -2689,22 +2749,47 @@ EXPORT_SYMBOL_GPL(call_rcu); #define KFREE_DRAIN_JIFFIES (HZ / 50) #define KFREE_N_BATCHES 2 +/* + * This macro defines how many entries the "records" array + * will contain. It is based on the fact that the size of + * kfree_rcu_bulk_data structure becomes exactly one page. + */ +#define KFREE_BULK_MAX_ENTR ((PAGE_SIZE / sizeof(void *)) - 3) + +/** + * struct kfree_rcu_bulk_data - single block to store kfree_rcu() pointers + * @nr_records: Number of active pointers in the array + * @records: Array of the kfree_rcu() pointers + * @next: Next bulk object in the block chain + * @head_free_debug: For debug, when CONFIG_DEBUG_OBJECTS_RCU_HEAD is set + */ +struct kfree_rcu_bulk_data { + unsigned long nr_records; + void *records[KFREE_BULK_MAX_ENTR]; + struct kfree_rcu_bulk_data *next; + struct rcu_head *head_free_debug; +}; + /** * struct kfree_rcu_cpu_work - single batch of kfree_rcu() requests * @rcu_work: Let queue_rcu_work() invoke workqueue handler after grace period * @head_free: List of kfree_rcu() objects waiting for a grace period + * @bhead_free: Bulk-List of kfree_rcu() objects waiting for a grace period * @krcp: Pointer to @kfree_rcu_cpu structure */ struct kfree_rcu_cpu_work { struct rcu_work rcu_work; struct rcu_head *head_free; + struct kfree_rcu_bulk_data *bhead_free; struct kfree_rcu_cpu *krcp; }; /** * struct kfree_rcu_cpu - batch up kfree_rcu() requests for RCU grace period * @head: List of kfree_rcu() objects not yet waiting for a grace period + * @bhead: Bulk-List of kfree_rcu() objects not yet waiting for a grace period + * @bcached: Keeps at most one object for later reuse when build chain blocks * @krw_arr: Array of batches of kfree_rcu() objects waiting for a grace period * @lock: Synchronize access to this structure * @monitor_work: Promote @head to @head_free after KFREE_DRAIN_JIFFIES @@ -2718,6 +2803,8 @@ struct kfree_rcu_cpu_work { */ struct kfree_rcu_cpu { struct rcu_head *head; + struct kfree_rcu_bulk_data *bhead; + struct kfree_rcu_bulk_data *bcached; struct kfree_rcu_cpu_work krw_arr[KFREE_N_BATCHES]; spinlock_t lock; struct delayed_work monitor_work; @@ -2727,14 +2814,24 @@ struct kfree_rcu_cpu { static DEFINE_PER_CPU(struct kfree_rcu_cpu, krc); +static __always_inline void +debug_rcu_head_unqueue_bulk(struct rcu_head *head) +{ +#ifdef CONFIG_DEBUG_OBJECTS_RCU_HEAD + for (; head; head = head->next) + debug_rcu_head_unqueue(head); +#endif +} + /* * This function is invoked in workqueue context after a grace period. - * It frees all the objects queued on ->head_free. + * It frees all the objects queued on ->bhead_free or ->head_free. */ static void kfree_rcu_work(struct work_struct *work) { unsigned long flags; struct rcu_head *head, *next; + struct kfree_rcu_bulk_data *bhead, *bnext; struct kfree_rcu_cpu *krcp; struct kfree_rcu_cpu_work *krwp; @@ -2744,22 +2841,44 @@ static void kfree_rcu_work(struct work_struct *work) spin_lock_irqsave(&krcp->lock, flags); head = krwp->head_free; krwp->head_free = NULL; + bhead = krwp->bhead_free; + krwp->bhead_free = NULL; spin_unlock_irqrestore(&krcp->lock, flags); - // List "head" is now private, so traverse locklessly. + /* "bhead" is now private, so traverse locklessly. */ + for (; bhead; bhead = bnext) { + bnext = bhead->next; + + debug_rcu_head_unqueue_bulk(bhead->head_free_debug); + + rcu_lock_acquire(&rcu_callback_map); + trace_rcu_invoke_kfree_bulk_callback(rcu_state.name, + bhead->nr_records, bhead->records); + + kfree_bulk(bhead->nr_records, bhead->records); + rcu_lock_release(&rcu_callback_map); + + if (cmpxchg(&krcp->bcached, NULL, bhead)) + free_page((unsigned long) bhead); + + cond_resched_tasks_rcu_qs(); + } + + /* + * Emergency case only. It can happen under low memory + * condition when an allocation gets failed, so the "bulk" + * path can not be temporary maintained. + */ for (; head; head = next) { unsigned long offset = (unsigned long)head->func; next = head->next; - // Potentially optimize with kfree_bulk in future. debug_rcu_head_unqueue(head); rcu_lock_acquire(&rcu_callback_map); trace_rcu_invoke_kfree_callback(rcu_state.name, head, offset); - if (!WARN_ON_ONCE(!__is_kfree_rcu_offset(offset))) { - /* Could be optimized with kfree_bulk() in future. */ + if (!WARN_ON_ONCE(!__is_kfree_rcu_offset(offset))) kfree((void *)head - offset); - } rcu_lock_release(&rcu_callback_map); cond_resched_tasks_rcu_qs(); @@ -2774,26 +2893,48 @@ static void kfree_rcu_work(struct work_struct *work) */ static inline bool queue_kfree_rcu_work(struct kfree_rcu_cpu *krcp) { + struct kfree_rcu_cpu_work *krwp; + bool queued = false; int i; - struct kfree_rcu_cpu_work *krwp = NULL; lockdep_assert_held(&krcp->lock); - for (i = 0; i < KFREE_N_BATCHES; i++) - if (!krcp->krw_arr[i].head_free) { - krwp = &(krcp->krw_arr[i]); - break; - } - // If a previous RCU batch is in progress, we cannot immediately - // queue another one, so return false to tell caller to retry. - if (!krwp) - return false; + for (i = 0; i < KFREE_N_BATCHES; i++) { + krwp = &(krcp->krw_arr[i]); - krwp->head_free = krcp->head; - krcp->head = NULL; - INIT_RCU_WORK(&krwp->rcu_work, kfree_rcu_work); - queue_rcu_work(system_wq, &krwp->rcu_work); - return true; + /* + * Try to detach bhead or head and attach it over any + * available corresponding free channel. It can be that + * a previous RCU batch is in progress, it means that + * immediately to queue another one is not possible so + * return false to tell caller to retry. + */ + if ((krcp->bhead && !krwp->bhead_free) || + (krcp->head && !krwp->head_free)) { + /* Channel 1. */ + if (!krwp->bhead_free) { + krwp->bhead_free = krcp->bhead; + krcp->bhead = NULL; + } + + /* Channel 2. */ + if (!krwp->head_free) { + krwp->head_free = krcp->head; + krcp->head = NULL; + } + + /* + * One work is per one batch, so there are two "free channels", + * "bhead_free" and "head_free" the batch can handle. It can be + * that the work is in the pending state when two channels have + * been detached following each other, one by one. + */ + queue_rcu_work(system_wq, &krwp->rcu_work); + queued = true; + } + } + + return queued; } static inline void kfree_rcu_drain_unlock(struct kfree_rcu_cpu *krcp, @@ -2830,19 +2971,65 @@ static void kfree_rcu_monitor(struct work_struct *work) spin_unlock_irqrestore(&krcp->lock, flags); } +static inline bool +kfree_call_rcu_add_ptr_to_bulk(struct kfree_rcu_cpu *krcp, + struct rcu_head *head, rcu_callback_t func) +{ + struct kfree_rcu_bulk_data *bnode; + + if (unlikely(!krcp->initialized)) + return false; + + lockdep_assert_held(&krcp->lock); + + /* Check if a new block is required. */ + if (!krcp->bhead || + krcp->bhead->nr_records == KFREE_BULK_MAX_ENTR) { + bnode = xchg(&krcp->bcached, NULL); + if (!bnode) { + WARN_ON_ONCE(sizeof(struct kfree_rcu_bulk_data) > PAGE_SIZE); + + bnode = (struct kfree_rcu_bulk_data *) + __get_free_page(GFP_NOWAIT | __GFP_NOWARN); + } + + /* Switch to emergency path. */ + if (unlikely(!bnode)) + return false; + + /* Initialize the new block. */ + bnode->nr_records = 0; + bnode->next = krcp->bhead; + bnode->head_free_debug = NULL; + + /* Attach it to the head. */ + krcp->bhead = bnode; + } + +#ifdef CONFIG_DEBUG_OBJECTS_RCU_HEAD + head->func = func; + head->next = krcp->bhead->head_free_debug; + krcp->bhead->head_free_debug = head; +#endif + + /* Finally insert. */ + krcp->bhead->records[krcp->bhead->nr_records++] = + (void *) head - (unsigned long) func; + + return true; +} + /* - * Queue a request for lazy invocation of kfree() after a grace period. + * Queue a request for lazy invocation of kfree_bulk()/kfree() after a grace + * period. Please note there are two paths are maintained, one is the main one + * that uses kfree_bulk() interface and second one is emergency one, that is + * used only when the main path can not be maintained temporary, due to memory + * pressure. * * Each kfree_call_rcu() request is added to a batch. The batch will be drained - * every KFREE_DRAIN_JIFFIES number of jiffies. All the objects in the batch - * will be kfree'd in workqueue context. This allows us to: - * - * 1. Batch requests together to reduce the number of grace periods during - * heavy kfree_rcu() load. - * - * 2. It makes it possible to use kfree_bulk() on a large number of - * kfree_rcu() requests thus reducing cache misses and the per-object - * overhead of kfree(). + * every KFREE_DRAIN_JIFFIES number of jiffies. All the objects in the batch will + * be free'd in workqueue context. This allows us to: batch requests together to + * reduce the number of grace periods during heavy kfree_rcu() load. */ void kfree_call_rcu(struct rcu_head *head, rcu_callback_t func) { @@ -2861,9 +3048,16 @@ void kfree_call_rcu(struct rcu_head *head, rcu_callback_t func) __func__, head); goto unlock_return; } - head->func = func; - head->next = krcp->head; - krcp->head = head; + + /* + * Under high memory pressure GFP_NOWAIT can fail, + * in that case the emergency path is maintained. + */ + if (unlikely(!kfree_call_rcu_add_ptr_to_bulk(krcp, head, func))) { + head->func = func; + head->next = krcp->head; + krcp->head = head; + } // Set timer to drain after KFREE_DRAIN_JIFFIES. if (rcu_scheduler_active == RCU_SCHEDULER_RUNNING && @@ -3075,24 +3269,32 @@ static void rcu_barrier_trace(const char *s, int cpu, unsigned long done) /* * RCU callback function for rcu_barrier(). If we are last, wake * up the task executing rcu_barrier(). + * + * Note that the value of rcu_state.barrier_sequence must be captured + * before the atomic_dec_and_test(). Otherwise, if this CPU is not last, + * other CPUs might count the value down to zero before this CPU gets + * around to invoking rcu_barrier_trace(), which might result in bogus + * data from the next instance of rcu_barrier(). */ static void rcu_barrier_callback(struct rcu_head *rhp) { + unsigned long __maybe_unused s = rcu_state.barrier_sequence; + if (atomic_dec_and_test(&rcu_state.barrier_cpu_count)) { - rcu_barrier_trace(TPS("LastCB"), -1, - rcu_state.barrier_sequence); + rcu_barrier_trace(TPS("LastCB"), -1, s); complete(&rcu_state.barrier_completion); } else { - rcu_barrier_trace(TPS("CB"), -1, rcu_state.barrier_sequence); + rcu_barrier_trace(TPS("CB"), -1, s); } } /* * Called with preemption disabled, and from cross-cpu IRQ context. */ -static void rcu_barrier_func(void *unused) +static void rcu_barrier_func(void *cpu_in) { - struct rcu_data *rdp = raw_cpu_ptr(&rcu_data); + uintptr_t cpu = (uintptr_t)cpu_in; + struct rcu_data *rdp = per_cpu_ptr(&rcu_data, cpu); rcu_barrier_trace(TPS("IRQ"), -1, rcu_state.barrier_sequence); rdp->barrier_head.func = rcu_barrier_callback; @@ -3119,7 +3321,7 @@ static void rcu_barrier_func(void *unused) */ void rcu_barrier(void) { - int cpu; + uintptr_t cpu; struct rcu_data *rdp; unsigned long s = rcu_seq_snap(&rcu_state.barrier_sequence); @@ -3142,13 +3344,14 @@ void rcu_barrier(void) rcu_barrier_trace(TPS("Inc1"), -1, rcu_state.barrier_sequence); /* - * Initialize the count to one rather than to zero in order to - * avoid a too-soon return to zero in case of a short grace period - * (or preemption of this task). Exclude CPU-hotplug operations - * to ensure that no offline CPU has callbacks queued. + * Initialize the count to two rather than to zero in order + * to avoid a too-soon return to zero in case of an immediate + * invocation of the just-enqueued callback (or preemption of + * this task). Exclude CPU-hotplug operations to ensure that no + * offline non-offloaded CPU has callbacks queued. */ init_completion(&rcu_state.barrier_completion); - atomic_set(&rcu_state.barrier_cpu_count, 1); + atomic_set(&rcu_state.barrier_cpu_count, 2); get_online_cpus(); /* @@ -3158,13 +3361,23 @@ void rcu_barrier(void) */ for_each_possible_cpu(cpu) { rdp = per_cpu_ptr(&rcu_data, cpu); - if (!cpu_online(cpu) && + if (cpu_is_offline(cpu) && !rcu_segcblist_is_offloaded(&rdp->cblist)) continue; - if (rcu_segcblist_n_cbs(&rdp->cblist)) { + if (rcu_segcblist_n_cbs(&rdp->cblist) && cpu_online(cpu)) { rcu_barrier_trace(TPS("OnlineQ"), cpu, rcu_state.barrier_sequence); - smp_call_function_single(cpu, rcu_barrier_func, NULL, 1); + smp_call_function_single(cpu, rcu_barrier_func, (void *)cpu, 1); + } else if (rcu_segcblist_n_cbs(&rdp->cblist) && + cpu_is_offline(cpu)) { + rcu_barrier_trace(TPS("OfflineNoCBQ"), cpu, + rcu_state.barrier_sequence); + local_irq_disable(); + rcu_barrier_func((void *)cpu); + local_irq_enable(); + } else if (cpu_is_offline(cpu)) { + rcu_barrier_trace(TPS("OfflineNoCBNoQ"), cpu, + rcu_state.barrier_sequence); } else { rcu_barrier_trace(TPS("OnlineNQ"), cpu, rcu_state.barrier_sequence); @@ -3176,7 +3389,7 @@ void rcu_barrier(void) * Now that we have an rcu_barrier_callback() callback on each * CPU, and thus each counted, remove the initial count. */ - if (atomic_dec_and_test(&rcu_state.barrier_cpu_count)) + if (atomic_sub_and_test(2, &rcu_state.barrier_cpu_count)) complete(&rcu_state.barrier_completion); /* Wait for all rcu_barrier_callback() callbacks to be invoked. */ @@ -3275,12 +3488,12 @@ int rcutree_prepare_cpu(unsigned int cpu) rnp = rdp->mynode; raw_spin_lock_rcu_node(rnp); /* irqs already disabled. */ rdp->beenonline = true; /* We have now been online. */ - rdp->gp_seq = rnp->gp_seq; - rdp->gp_seq_needed = rnp->gp_seq; + rdp->gp_seq = READ_ONCE(rnp->gp_seq); + rdp->gp_seq_needed = rdp->gp_seq; rdp->cpu_no_qs.b.norm = true; rdp->core_needs_qs = false; rdp->rcu_iw_pending = false; - rdp->rcu_iw_gp_seq = rnp->gp_seq - 1; + rdp->rcu_iw_gp_seq = rdp->gp_seq - 1; trace_rcu_grace_period(rcu_state.name, rdp->gp_seq, TPS("cpuonl")); raw_spin_unlock_irqrestore_rcu_node(rnp, flags); rcu_prepare_kthreads(cpu); @@ -3378,7 +3591,7 @@ void rcu_cpu_starting(unsigned int cpu) rnp = rdp->mynode; mask = rdp->grpmask; raw_spin_lock_irqsave_rcu_node(rnp, flags); - rnp->qsmaskinitnext |= mask; + WRITE_ONCE(rnp->qsmaskinitnext, rnp->qsmaskinitnext | mask); oldmask = rnp->expmaskinitnext; rnp->expmaskinitnext |= mask; oldmask ^= rnp->expmaskinitnext; @@ -3431,7 +3644,7 @@ void rcu_report_dead(unsigned int cpu) rcu_report_qs_rnp(mask, rnp, rnp->gp_seq, flags); raw_spin_lock_irqsave_rcu_node(rnp, flags); } - rnp->qsmaskinitnext &= ~mask; + WRITE_ONCE(rnp->qsmaskinitnext, rnp->qsmaskinitnext & ~mask); raw_spin_unlock_irqrestore_rcu_node(rnp, flags); raw_spin_unlock(&rcu_state.ofl_lock); @@ -3545,7 +3758,10 @@ static int __init rcu_spawn_gp_kthread(void) } rnp = rcu_get_root(); raw_spin_lock_irqsave_rcu_node(rnp, flags); - rcu_state.gp_kthread = t; + WRITE_ONCE(rcu_state.gp_activity, jiffies); + WRITE_ONCE(rcu_state.gp_req_activity, jiffies); + // Reset .gp_activity and .gp_req_activity before setting .gp_kthread. + smp_store_release(&rcu_state.gp_kthread, t); /* ^^^ */ raw_spin_unlock_irqrestore_rcu_node(rnp, flags); wake_up_process(t); rcu_spawn_nocb_kthreads(); @@ -3769,8 +3985,11 @@ static void __init kfree_rcu_batch_init(void) struct kfree_rcu_cpu *krcp = per_cpu_ptr(&krc, cpu); spin_lock_init(&krcp->lock); - for (i = 0; i < KFREE_N_BATCHES; i++) + for (i = 0; i < KFREE_N_BATCHES; i++) { + INIT_RCU_WORK(&krcp->krw_arr[i].rcu_work, kfree_rcu_work); krcp->krw_arr[i].krcp = krcp; + } + INIT_DELAYED_WORK(&krcp->monitor_work, kfree_rcu_monitor); krcp->initialized = true; } @@ -3809,6 +4028,13 @@ void __init rcu_init(void) rcu_par_gp_wq = alloc_workqueue("rcu_par_gp", WQ_MEM_RECLAIM, 0); WARN_ON(!rcu_par_gp_wq); srcu_init(); + + /* Fill in default value for rcutree.qovld boot parameter. */ + /* -After- the rcu_node ->lock fields are initialized! */ + if (qovld < 0) + qovld_calc = DEFAULT_RCU_QOVLD_MULT * qhimark; + else + qovld_calc = qovld; } #include "tree_stall.h" |