Merge commit 'fixes.2015.02.23a' into core/rcu

 Conflicts:
	kernel/rcu/tree.c
Signed-off-by: Ingo Molnar <mingo@kernel.org>

include/linux/compiler.h

@@ -20,12 +20,14 @@
 # define __pmem		__attribute__((noderef, address_space(5)))
 #ifdef CONFIG_SPARSE_RCU_POINTER
 # define __rcu		__attribute__((noderef, address_space(4)))
-#else
+#else /* CONFIG_SPARSE_RCU_POINTER */
 # define __rcu
-#endif
+#endif /* CONFIG_SPARSE_RCU_POINTER */
+# define __private	__attribute__((noderef))
 extern void __chk_user_ptr(const volatile void __user *);
 extern void __chk_io_ptr(const volatile void __iomem *);
-#else
+# define ACCESS_PRIVATE(p, member) (*((typeof((p)->member) __force *) &(p)->member))
+#else /* __CHECKER__ */
 # define __user
 # define __kernel
 # define __safe
@@ -44,7 +46,9 @@ extern void __chk_io_ptr(const volatile void __iomem *);
 # define __percpu
 # define __rcu
 # define __pmem
-#endif
+# define __private
+# define ACCESS_PRIVATE(p, member) ((p)->member)
+#endif /* __CHECKER__ */
 
 /* Indirect macros required for expanded argument pasting, eg. __LINE__. */
 #define ___PASTE(a,b) a##b

include/linux/irq.h

@@ -137,7 +137,7 @@ struct irq_domain;
  * @msi_desc:		MSI descriptor
  */
 struct irq_common_data {
-	unsigned int		state_use_accessors;
+	unsigned int		__private state_use_accessors;
 #ifdef CONFIG_NUMA
 	unsigned int		node;
 #endif
@@ -208,7 +208,7 @@ enum {
 	IRQD_FORWARDED_TO_VCPU		= (1 << 20),
 };
 
-#define __irqd_to_state(d)		((d)->common->state_use_accessors)
+#define __irqd_to_state(d) ACCESS_PRIVATE((d)->common, state_use_accessors)
 
 static inline bool irqd_is_setaffinity_pending(struct irq_data *d)
 {
@@ -299,6 +299,8 @@ static inline void irqd_clr_forwarded_to_vcpu(struct irq_data *d)
 	__irqd_to_state(d) &= ~IRQD_FORWARDED_TO_VCPU;
 }
 
+#undef __irqd_to_state
+
 static inline irq_hw_number_t irqd_to_hwirq(struct irq_data *d)
 {
 	return d->hwirq;

include/linux/rcupdate.h

@@ -360,8 +360,6 @@ void rcu_user_exit(void);
 #else
 static inline void rcu_user_enter(void) { }
 static inline void rcu_user_exit(void) { }
-static inline void rcu_user_hooks_switch(struct task_struct *prev,
-					 struct task_struct *next) { }
 #endif /* CONFIG_NO_HZ_FULL */
 
 #ifdef CONFIG_RCU_NOCB_CPU

include/linux/srcu.h

@@ -99,8 +99,23 @@ void process_srcu(struct work_struct *work);
 	}
 
 /*
- * define and init a srcu struct at build time.
- * dont't call init_srcu_struct() nor cleanup_srcu_struct() on it.
+ * Define and initialize a srcu struct at build time.
+ * Do -not- call init_srcu_struct() nor cleanup_srcu_struct() on it.
+ *
+ * Note that although DEFINE_STATIC_SRCU() hides the name from other
+ * files, the per-CPU variable rules nevertheless require that the
+ * chosen name be globally unique.  These rules also prohibit use of
+ * DEFINE_STATIC_SRCU() within a function.  If these rules are too
+ * restrictive, declare the srcu_struct manually.  For example, in
+ * each file:
+ *
+ *	static struct srcu_struct my_srcu;
+ *
+ * Then, before the first use of each my_srcu, manually initialize it:
+ *
+ *	init_srcu_struct(&my_srcu);
+ *
+ * See include/linux/percpu-defs.h for the rules on per-CPU variables.
  */
 #define __DEFINE_SRCU(name, is_static)					\
 	static DEFINE_PER_CPU(struct srcu_struct_array, name##_srcu_array);\

kernel/irq/internals.h

@@ -160,6 +160,8 @@ irq_put_desc_unlock(struct irq_desc *desc, unsigned long flags)
 	__irq_put_desc_unlock(desc, flags, false);
 }
 
+#define __irqd_to_state(d) ACCESS_PRIVATE((d)->common, state_use_accessors)
+
 /*
  * Manipulation functions for irq_data.state
  */
@@ -188,6 +190,8 @@ static inline bool irqd_has_set(struct irq_data *d, unsigned int mask)
 	return __irqd_to_state(d) & mask;
 }
 
+#undef __irqd_to_state
+
 static inline void kstat_incr_irqs_this_cpu(struct irq_desc *desc)
 {
 	__this_cpu_inc(*desc->kstat_irqs);

kernel/rcu/rcutorture.c

@@ -932,12 +932,14 @@ rcu_torture_writer(void *arg)
 	int nsynctypes = 0;
 
 	VERBOSE_TOROUT_STRING("rcu_torture_writer task started");
-	pr_alert("%s" TORTURE_FLAG
-		 " Grace periods expedited from boot/sysfs for %s,\n",
-		 torture_type, cur_ops->name);
-	pr_alert("%s" TORTURE_FLAG
-		 " Testing of dynamic grace-period expediting diabled.\n",
-		 torture_type);
+	if (!can_expedite) {
+		pr_alert("%s" TORTURE_FLAG
+			 " Grace periods expedited from boot/sysfs for %s,\n",
+			 torture_type, cur_ops->name);
+		pr_alert("%s" TORTURE_FLAG
+			 " Disabled dynamic grace-period expediting.\n",
+			 torture_type);
+	}
 
 	/* Initialize synctype[] array.  If none set, take default. */
 	if (!gp_cond1 && !gp_exp1 && !gp_normal1 && !gp_sync1)

kernel/rcu/tiny_plugin.h

@@ -23,7 +23,7 @@
  */
 
 #include <linux/kthread.h>
-#include <linux/module.h>
+#include <linux/init.h>
 #include <linux/debugfs.h>
 #include <linux/seq_file.h>
 
@@ -122,18 +122,7 @@ static int __init rcutiny_trace_init(void)
 	debugfs_remove_recursive(rcudir);
 	return 1;
 }
-
-static void __exit rcutiny_trace_cleanup(void)
-{
-	debugfs_remove_recursive(rcudir);
-}
-
-module_init(rcutiny_trace_init);
-module_exit(rcutiny_trace_cleanup);
-
-MODULE_AUTHOR("Paul E. McKenney");
-MODULE_DESCRIPTION("Read-Copy Update tracing for tiny implementation");
-MODULE_LICENSE("GPL");
+device_initcall(rcutiny_trace_init);
 
 static void check_cpu_stall(struct rcu_ctrlblk *rcp)
 {

kernel/rcu/tree.c

@@ -108,7 +108,6 @@ RCU_STATE_INITIALIZER(rcu_sched, 's', call_rcu_sched);
 RCU_STATE_INITIALIZER(rcu_bh, 'b', call_rcu_bh);
 
 static struct rcu_state *const rcu_state_p;
-static struct rcu_data __percpu *const rcu_data_p;
 LIST_HEAD(rcu_struct_flavors);
 
 /* Dump rcu_node combining tree at boot to verify correct setup. */
@@ -1083,13 +1082,12 @@ static int dyntick_save_progress_counter(struct rcu_data *rdp,
 	rcu_sysidle_check_cpu(rdp, isidle, maxj);
 	if ((rdp->dynticks_snap & 0x1) == 0) {
 		trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, TPS("dti"));
-		return 1;
-	} else {
 		if (ULONG_CMP_LT(READ_ONCE(rdp->gpnum) + ULONG_MAX / 4,
 				 rdp->mynode->gpnum))
 			WRITE_ONCE(rdp->gpwrap, true);
-		return 0;
+		return 1;
 	}
+	return 0;
 }
 
 /*
@@ -1173,15 +1171,16 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp,
 			smp_mb(); /* ->cond_resched_completed before *rcrmp. */
 			WRITE_ONCE(*rcrmp,
 				   READ_ONCE(*rcrmp) + rdp->rsp->flavor_mask);
-			resched_cpu(rdp->cpu);  /* Force CPU into scheduler. */
-			rdp->rsp->jiffies_resched += 5; /* Enable beating. */
-		} else if (ULONG_CMP_GE(jiffies, rdp->rsp->jiffies_resched)) {
-			/* Time to beat on that CPU again! */
-			resched_cpu(rdp->cpu);  /* Force CPU into scheduler. */
-			rdp->rsp->jiffies_resched += 5; /* Re-enable beating. */
 		}
+		rdp->rsp->jiffies_resched += 5; /* Re-enable beating. */
 	}
 
+	/* And if it has been a really long time, kick the CPU as well. */
+	if (ULONG_CMP_GE(jiffies,
+			 rdp->rsp->gp_start + 2 * jiffies_till_sched_qs) ||
+	    ULONG_CMP_GE(jiffies, rdp->rsp->gp_start + jiffies_till_sched_qs))
+		resched_cpu(rdp->cpu);  /* Force CPU into scheduler. */
+
 	return 0;
 }
 
@@ -1246,7 +1245,7 @@ static void rcu_dump_cpu_stacks(struct rcu_state *rsp)
 				if (rnp->qsmask & (1UL << cpu))
 					dump_cpu_task(rnp->grplo + cpu);
 		}
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 	}
 }
 
@@ -1266,12 +1265,12 @@ static void print_other_cpu_stall(struct rcu_state *rsp, unsigned long gpnum)
 	raw_spin_lock_irqsave_rcu_node(rnp, flags);
 	delta = jiffies - READ_ONCE(rsp->jiffies_stall);
 	if (delta < RCU_STALL_RAT_DELAY || !rcu_gp_in_progress(rsp)) {
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 		return;
 	}
 	WRITE_ONCE(rsp->jiffies_stall,
 		   jiffies + 3 * rcu_jiffies_till_stall_check() + 3);
-	raw_spin_unlock_irqrestore(&rnp->lock, flags);
+	raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 
 	/*
 	 * OK, time to rat on our buddy...
@@ -1292,7 +1291,7 @@ static void print_other_cpu_stall(struct rcu_state *rsp, unsigned long gpnum)
 					ndetected++;
 				}
 		}
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 	}
 
 	print_cpu_stall_info_end();
@@ -1357,7 +1356,7 @@ static void print_cpu_stall(struct rcu_state *rsp)
 	if (ULONG_CMP_GE(jiffies, READ_ONCE(rsp->jiffies_stall)))
 		WRITE_ONCE(rsp->jiffies_stall,
 			   jiffies + 3 * rcu_jiffies_till_stall_check() + 3);
-	raw_spin_unlock_irqrestore(&rnp->lock, flags);
+	raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 
 	/*
 	 * Attempt to revive the RCU machinery by forcing a context switch.
@@ -1595,7 +1594,7 @@ rcu_start_future_gp(struct rcu_node *rnp, struct rcu_data *rdp,
 	}
 unlock_out:
 	if (rnp != rnp_root)
-		raw_spin_unlock(&rnp_root->lock);
+		raw_spin_unlock_rcu_node(rnp_root);
 out:
 	if (c_out != NULL)
 		*c_out = c;
@@ -1814,7 +1813,7 @@ static void note_gp_changes(struct rcu_state *rsp, struct rcu_data *rdp)
 		return;
 	}
 	needwake = __note_gp_changes(rsp, rnp, rdp);
-	raw_spin_unlock_irqrestore(&rnp->lock, flags);
+	raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 	if (needwake)
 		rcu_gp_kthread_wake(rsp);
 }
@@ -1839,7 +1838,7 @@ static bool rcu_gp_init(struct rcu_state *rsp)
 	raw_spin_lock_irq_rcu_node(rnp);
 	if (!READ_ONCE(rsp->gp_flags)) {
 		/* Spurious wakeup, tell caller to go back to sleep.  */
-		raw_spin_unlock_irq(&rnp->lock);
+		raw_spin_unlock_irq_rcu_node(rnp);
 		return false;
 	}
 	WRITE_ONCE(rsp->gp_flags, 0); /* Clear all flags: New grace period. */
@@ -1849,7 +1848,7 @@ static bool rcu_gp_init(struct rcu_state *rsp)
 		 * Grace period already in progress, don't start another.
 		 * Not supposed to be able to happen.
 		 */
-		raw_spin_unlock_irq(&rnp->lock);
+		raw_spin_unlock_irq_rcu_node(rnp);
 		return false;
 	}
 
@@ -1858,7 +1857,7 @@ static bool rcu_gp_init(struct rcu_state *rsp)
 	/* Record GP times before starting GP, hence smp_store_release(). */
 	smp_store_release(&rsp->gpnum, rsp->gpnum + 1);
 	trace_rcu_grace_period(rsp->name, rsp->gpnum, TPS("start"));
-	raw_spin_unlock_irq(&rnp->lock);
+	raw_spin_unlock_irq_rcu_node(rnp);
 
 	/*
 	 * Apply per-leaf buffered online and offline operations to the
@@ -1872,7 +1871,7 @@ static bool rcu_gp_init(struct rcu_state *rsp)
 		if (rnp->qsmaskinit == rnp->qsmaskinitnext &&
 		    !rnp->wait_blkd_tasks) {
 			/* Nothing to do on this leaf rcu_node structure. */
-			raw_spin_unlock_irq(&rnp->lock);
+			raw_spin_unlock_irq_rcu_node(rnp);
 			continue;
 		}
 
@@ -1906,7 +1905,7 @@ static bool rcu_gp_init(struct rcu_state *rsp)
 			rcu_cleanup_dead_rnp(rnp);
 		}
 
-		raw_spin_unlock_irq(&rnp->lock);
+		raw_spin_unlock_irq_rcu_node(rnp);
 	}
 
 	/*
@@ -1937,7 +1936,7 @@ static bool rcu_gp_init(struct rcu_state *rsp)
 		trace_rcu_grace_period_init(rsp->name, rnp->gpnum,
 					    rnp->level, rnp->grplo,
 					    rnp->grphi, rnp->qsmask);
-		raw_spin_unlock_irq(&rnp->lock);
+		raw_spin_unlock_irq_rcu_node(rnp);
 		cond_resched_rcu_qs();
 		WRITE_ONCE(rsp->gp_activity, jiffies);
 	}
@@ -1995,7 +1994,7 @@ static void rcu_gp_fqs(struct rcu_state *rsp, bool first_time)
 		raw_spin_lock_irq_rcu_node(rnp);
 		WRITE_ONCE(rsp->gp_flags,
 			   READ_ONCE(rsp->gp_flags) & ~RCU_GP_FLAG_FQS);
-		raw_spin_unlock_irq(&rnp->lock);
+		raw_spin_unlock_irq_rcu_node(rnp);
 	}
 }
 
@@ -2025,7 +2024,7 @@ static void rcu_gp_cleanup(struct rcu_state *rsp)
 	 * safe for us to drop the lock in order to mark the grace
 	 * period as completed in all of the rcu_node structures.
 	 */
-	raw_spin_unlock_irq(&rnp->lock);
+	raw_spin_unlock_irq_rcu_node(rnp);
 
 	/*
 	 * Propagate new ->completed value to rcu_node structures so
@@ -2047,7 +2046,7 @@ static void rcu_gp_cleanup(struct rcu_state *rsp)
 		/* smp_mb() provided by prior unlock-lock pair. */
 		nocb += rcu_future_gp_cleanup(rsp, rnp);
 		sq = rcu_nocb_gp_get(rnp);
-		raw_spin_unlock_irq(&rnp->lock);
+		raw_spin_unlock_irq_rcu_node(rnp);
 		rcu_nocb_gp_cleanup(sq);
 		cond_resched_rcu_qs();
 		WRITE_ONCE(rsp->gp_activity, jiffies);
@@ -2070,7 +2069,7 @@ static void rcu_gp_cleanup(struct rcu_state *rsp)
 				       READ_ONCE(rsp->gpnum),
 				       TPS("newreq"));
 	}
-	raw_spin_unlock_irq(&rnp->lock);
+	raw_spin_unlock_irq_rcu_node(rnp);
 }
 
 /*
@@ -2236,18 +2235,20 @@ static bool rcu_start_gp(struct rcu_state *rsp)
 }
 
 /*
- * 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, which
- * is released before return.
+ * Report a full set of quiescent states to the specified rcu_state data
+ * structure.  Invoke rcu_gp_kthread_wake() to awaken the grace-period
+ * kthread if another grace period is required.  Whether we wake
+ * the grace-period kthread or it awakens itself for the next round
+ * of quiescent-state forcing, that kthread will clean up after the
+ * just-completed grace period.  Note that the caller must hold rnp->lock,
+ * which is released before return.
  */
 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));
 	WRITE_ONCE(rsp->gp_flags, READ_ONCE(rsp->gp_flags) | RCU_GP_FLAG_FQS);
-	raw_spin_unlock_irqrestore(&rcu_get_root(rsp)->lock, flags);
+	raw_spin_unlock_irqrestore_rcu_node(rcu_get_root(rsp), flags);
 	swake_up(&rsp->gp_wq);  /* Memory barrier implied by swake_up() path. */
 }
 
@@ -2277,7 +2278,7 @@ rcu_report_qs_rnp(unsigned long mask, struct rcu_state *rsp,
 			 * Our bit has already been cleared, or the
 			 * relevant grace period is already over, so done.
 			 */
-			raw_spin_unlock_irqrestore(&rnp->lock, flags);
+			raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 			return;
 		}
 		WARN_ON_ONCE(oldmask); /* Any child must be all zeroed! */
@@ -2289,7 +2290,7 @@ rcu_report_qs_rnp(unsigned long mask, struct rcu_state *rsp,
 		if (rnp->qsmask != 0 || rcu_preempt_blocked_readers_cgp(rnp)) {
 
 			/* Other bits still set at this level, so done. */
-			raw_spin_unlock_irqrestore(&rnp->lock, flags);
+			raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 			return;
 		}
 		mask = rnp->grpmask;
@@ -2299,7 +2300,7 @@ rcu_report_qs_rnp(unsigned long mask, struct rcu_state *rsp,
 
 			break;
 		}
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 		rnp_c = rnp;
 		rnp = rnp->parent;
 		raw_spin_lock_irqsave_rcu_node(rnp, flags);
@@ -2331,7 +2332,7 @@ static void rcu_report_unblock_qs_rnp(struct rcu_state *rsp,
 
 	if (rcu_state_p == &rcu_sched_state || rsp != rcu_state_p ||
 	    rnp->qsmask != 0 || rcu_preempt_blocked_readers_cgp(rnp)) {
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 		return;  /* Still need more quiescent states! */
 	}
 
@@ -2348,19 +2349,14 @@ static void rcu_report_unblock_qs_rnp(struct rcu_state *rsp,
 	/* Report up the rest of the hierarchy, tracking current ->gpnum. */
 	gps = rnp->gpnum;
 	mask = rnp->grpmask;
-	raw_spin_unlock(&rnp->lock);	/* irqs remain disabled. */
+	raw_spin_unlock_rcu_node(rnp);	/* irqs remain disabled. */
 	raw_spin_lock_rcu_node(rnp_p);	/* irqs already disabled. */
 	rcu_report_qs_rnp(mask, rsp, rnp_p, gps, flags);
 }
 
 /*
  * 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!
+ * structure.  This must be called from the specified CPU.
  */
 static void
 rcu_report_qs_rdp(int cpu, struct rcu_state *rsp, struct rcu_data *rdp)
@@ -2385,14 +2381,14 @@ rcu_report_qs_rdp(int cpu, struct rcu_state *rsp, struct rcu_data *rdp)
 		 */
 		rdp->cpu_no_qs.b.norm = true;	/* need qs for new gp. */
 		rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_qs_ctr);
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 		return;
 	}
 	mask = rdp->grpmask;
 	if ((rnp->qsmask & mask) == 0) {
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 	} else {
-		rdp->core_needs_qs = 0;
+		rdp->core_needs_qs = false;
 
 		/*
 		 * This GP can't end until cpu checks in, so all of our
@@ -2601,10 +2597,11 @@ static void rcu_cleanup_dead_rnp(struct rcu_node *rnp_leaf)
 		rnp->qsmaskinit &= ~mask;
 		rnp->qsmask &= ~mask;
 		if (rnp->qsmaskinit) {
-			raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */
+			raw_spin_unlock_rcu_node(rnp);
+			/* irqs remain disabled. */
 			return;
 		}
-		raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */
+		raw_spin_unlock_rcu_node(rnp); /* irqs remain disabled. */
 	}
 }
 
@@ -2627,7 +2624,7 @@ static void rcu_cleanup_dying_idle_cpu(int cpu, struct rcu_state *rsp)
 	mask = rdp->grpmask;
 	raw_spin_lock_irqsave_rcu_node(rnp, flags); /* Enforce GP memory-order guarantee. */
 	rnp->qsmaskinitnext &= ~mask;
-	raw_spin_unlock_irqrestore(&rnp->lock, flags);
+	raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 }
 
 /*
@@ -2861,7 +2858,7 @@ static void force_qs_rnp(struct rcu_state *rsp,
 			rcu_report_qs_rnp(mask, rsp, rnp, rnp->gpnum, flags);
 		} else {
 			/* Nothing to do here, so just drop the lock. */
-			raw_spin_unlock_irqrestore(&rnp->lock, flags);
+			raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 		}
 	}
 }
@@ -2897,11 +2894,11 @@ static void force_quiescent_state(struct rcu_state *rsp)
 	raw_spin_unlock(&rnp_old->fqslock);
 	if (READ_ONCE(rsp->gp_flags) & RCU_GP_FLAG_FQS) {
 		rsp->n_force_qs_lh++;
-		raw_spin_unlock_irqrestore(&rnp_old->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp_old, flags);
 		return;  /* Someone beat us to it. */
 	}
 	WRITE_ONCE(rsp->gp_flags, READ_ONCE(rsp->gp_flags) | RCU_GP_FLAG_FQS);
-	raw_spin_unlock_irqrestore(&rnp_old->lock, flags);
+	raw_spin_unlock_irqrestore_rcu_node(rnp_old, flags);
 	swake_up(&rsp->gp_wq); /* Memory barrier implied by swake_up() path. */
 }
 
@@ -2927,7 +2924,7 @@ __rcu_process_callbacks(struct rcu_state *rsp)
 	if (cpu_needs_another_gp(rsp, rdp)) {
 		raw_spin_lock_rcu_node(rcu_get_root(rsp)); /* irqs disabled. */
 		needwake = rcu_start_gp(rsp);
-		raw_spin_unlock_irqrestore(&rcu_get_root(rsp)->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rcu_get_root(rsp), flags);
 		if (needwake)
 			rcu_gp_kthread_wake(rsp);
 	} else {
@@ -3018,7 +3015,7 @@ static void __call_rcu_core(struct rcu_state *rsp, struct rcu_data *rdp,
 
 			raw_spin_lock_rcu_node(rnp_root);
 			needwake = rcu_start_gp(rsp);
-			raw_spin_unlock(&rnp_root->lock);
+			raw_spin_unlock_rcu_node(rnp_root);
 			if (needwake)
 				rcu_gp_kthread_wake(rsp);
 		} else {
@@ -3438,14 +3435,14 @@ static void sync_exp_reset_tree_hotplug(struct rcu_state *rsp)
 	rcu_for_each_leaf_node(rsp, rnp) {
 		raw_spin_lock_irqsave_rcu_node(rnp, flags);
 		if (rnp->expmaskinit == rnp->expmaskinitnext) {
-			raw_spin_unlock_irqrestore(&rnp->lock, flags);
+			raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 			continue;  /* No new CPUs, nothing to do. */
 		}
 
 		/* Update this node's mask, track old value for propagation. */
 		oldmask = rnp->expmaskinit;
 		rnp->expmaskinit = rnp->expmaskinitnext;
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 
 		/* If was already nonzero, nothing to propagate. */
 		if (oldmask)
@@ -3460,7 +3457,7 @@ static void sync_exp_reset_tree_hotplug(struct rcu_state *rsp)
 			if (rnp_up->expmaskinit)
 				done = true;
 			rnp_up->expmaskinit |= mask;
-			raw_spin_unlock_irqrestore(&rnp_up->lock, flags);
+			raw_spin_unlock_irqrestore_rcu_node(rnp_up, flags);
 			if (done)
 				break;
 			mask = rnp_up->grpmask;
@@ -3483,7 +3480,7 @@ static void __maybe_unused sync_exp_reset_tree(struct rcu_state *rsp)
 		raw_spin_lock_irqsave_rcu_node(rnp, flags);
 		WARN_ON_ONCE(rnp->expmask);
 		rnp->expmask = rnp->expmaskinit;
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 	}
 }
 
@@ -3524,11 +3521,11 @@ static void __rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp,
 			if (!rnp->expmask)
 				rcu_initiate_boost(rnp, flags);
 			else
-				raw_spin_unlock_irqrestore(&rnp->lock, flags);
+				raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 			break;
 		}
 		if (rnp->parent == NULL) {
-			raw_spin_unlock_irqrestore(&rnp->lock, flags);
+			raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 			if (wake) {
 				smp_mb(); /* EGP done before wake_up(). */
 				swake_up(&rsp->expedited_wq);
@@ -3536,7 +3533,7 @@ static void __rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp,
 			break;
 		}
 		mask = rnp->grpmask;
-		raw_spin_unlock(&rnp->lock); /* irqs remain disabled */
+		raw_spin_unlock_rcu_node(rnp); /* irqs remain disabled */
 		rnp = rnp->parent;
 		raw_spin_lock_rcu_node(rnp); /* irqs already disabled */
 		WARN_ON_ONCE(!(rnp->expmask & mask));
@@ -3571,7 +3568,7 @@ static void rcu_report_exp_cpu_mult(struct rcu_state *rsp, struct rcu_node *rnp,
 
 	raw_spin_lock_irqsave_rcu_node(rnp, flags);
 	if (!(rnp->expmask & mask)) {
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 		return;
 	}
 	rnp->expmask &= ~mask;
@@ -3732,7 +3729,7 @@ static void sync_rcu_exp_select_cpus(struct rcu_state *rsp,
 		 */
 		if (rcu_preempt_has_tasks(rnp))
 			rnp->exp_tasks = rnp->blkd_tasks.next;
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 
 		/* IPI the remaining CPUs for expedited quiescent state. */
 		mask = 1;
@@ -3749,7 +3746,7 @@ static void sync_rcu_exp_select_cpus(struct rcu_state *rsp,
 			raw_spin_lock_irqsave_rcu_node(rnp, flags);
 			if (cpu_online(cpu) &&
 			    (rnp->expmask & mask)) {
-				raw_spin_unlock_irqrestore(&rnp->lock, flags);
+				raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 				schedule_timeout_uninterruptible(1);
 				if (cpu_online(cpu) &&
 				    (rnp->expmask & mask))
@@ -3758,7 +3755,7 @@ static void sync_rcu_exp_select_cpus(struct rcu_state *rsp,
 			}
 			if (!(rnp->expmask & mask))
 				mask_ofl_ipi &= ~mask;
-			raw_spin_unlock_irqrestore(&rnp->lock, flags);
+			raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 		}
 		/* Report quiescent states for those that went offline. */
 		mask_ofl_test |= mask_ofl_ipi;
@@ -4165,7 +4162,7 @@ static void rcu_init_new_rnp(struct rcu_node *rnp_leaf)
 			return;
 		raw_spin_lock_rcu_node(rnp); /* Interrupts already disabled. */
 		rnp->qsmaskinit |= mask;
-		raw_spin_unlock(&rnp->lock); /* Interrupts remain disabled. */
+		raw_spin_unlock_rcu_node(rnp); /* Interrupts remain disabled. */
 	}
 }
 
@@ -4189,7 +4186,7 @@ rcu_boot_init_percpu_data(int cpu, struct rcu_state *rsp)
 	rdp->rsp = rsp;
 	mutex_init(&rdp->exp_funnel_mutex);
 	rcu_boot_init_nocb_percpu_data(rdp);
-	raw_spin_unlock_irqrestore(&rnp->lock, flags);
+	raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 }
 
 /*
@@ -4217,7 +4214,7 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp)
 	rcu_sysidle_init_percpu_data(rdp->dynticks);
 	atomic_set(&rdp->dynticks->dynticks,
 		   (atomic_read(&rdp->dynticks->dynticks) & ~0x1) + 1);
-	raw_spin_unlock(&rnp->lock);		/* irqs remain disabled. */
+	raw_spin_unlock_rcu_node(rnp);		/* irqs remain disabled. */
 
 	/*
 	 * Add CPU to leaf rcu_node pending-online bitmask.  Any needed
@@ -4238,7 +4235,7 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp)
 	rdp->rcu_qs_ctr_snap = per_cpu(rcu_qs_ctr, cpu);
 	rdp->core_needs_qs = false;
 	trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpuonl"));
-	raw_spin_unlock_irqrestore(&rnp->lock, flags);
+	raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 }
 
 static void rcu_prepare_cpu(int cpu)
@@ -4360,7 +4357,7 @@ static int __init rcu_spawn_gp_kthread(void)
 			sp.sched_priority = kthread_prio;
 			sched_setscheduler_nocheck(t, SCHED_FIFO, &sp);
 		}
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 		wake_up_process(t);
 	}
 	rcu_spawn_nocb_kthreads();
@@ -4451,8 +4448,8 @@ static void __init rcu_init_one(struct rcu_state *rsp)
 		cpustride *= levelspread[i];
 		rnp = rsp->level[i];
 		for (j = 0; j < levelcnt[i]; j++, rnp++) {
-			raw_spin_lock_init(&rnp->lock);
-			lockdep_set_class_and_name(&rnp->lock,
+			raw_spin_lock_init(&ACCESS_PRIVATE(rnp, lock));
+			lockdep_set_class_and_name(&ACCESS_PRIVATE(rnp, lock),
 						   &rcu_node_class[i], buf[i]);
 			raw_spin_lock_init(&rnp->fqslock);
 			lockdep_set_class_and_name(&rnp->fqslock,

kernel/rcu/tree.h

@@ -150,8 +150,9 @@ struct rcu_dynticks {
  * Definition for node within the RCU grace-period-detection hierarchy.
  */
 struct rcu_node {
-	raw_spinlock_t lock;	/* Root rcu_node's lock protects some */
-				/*  rcu_state fields as well as following. */
+	raw_spinlock_t __private lock;	/* Root rcu_node's lock protects */
+					/*  some rcu_state fields as well as */
+					/*  following. */
 	unsigned long gpnum;	/* Current grace period for this node. */
 				/*  This will either be equal to or one */
 				/*  behind the root rcu_node's gpnum. */
@@ -682,7 +683,7 @@ static inline void rcu_nocb_q_lengths(struct rcu_data *rdp, long *ql, long *qll)
 #endif /* #else #ifdef CONFIG_PPC */
 
 /*
- * Wrappers for the rcu_node::lock acquire.
+ * Wrappers for the rcu_node::lock acquire and release.
  *
  * Because the rcu_nodes form a tree, the tree traversal locking will observe
  * different lock values, this in turn means that an UNLOCK of one level
@@ -691,29 +692,48 @@ static inline void rcu_nocb_q_lengths(struct rcu_data *rdp, long *ql, long *qll)
  *
  * In order to restore full ordering between tree levels, augment the regular
  * lock acquire functions with smp_mb__after_unlock_lock().
+ *
+ * As ->lock of struct rcu_node is a __private field, therefore one should use
+ * these wrappers rather than directly call raw_spin_{lock,unlock}* on ->lock.
  */
 static inline void raw_spin_lock_rcu_node(struct rcu_node *rnp)
 {
-	raw_spin_lock(&rnp->lock);
+	raw_spin_lock(&ACCESS_PRIVATE(rnp, lock));
 	smp_mb__after_unlock_lock();
 }
 
+static inline void raw_spin_unlock_rcu_node(struct rcu_node *rnp)
+{
+	raw_spin_unlock(&ACCESS_PRIVATE(rnp, lock));
+}
+
 static inline void raw_spin_lock_irq_rcu_node(struct rcu_node *rnp)
 {
-	raw_spin_lock_irq(&rnp->lock);
+	raw_spin_lock_irq(&ACCESS_PRIVATE(rnp, lock));
 	smp_mb__after_unlock_lock();
 }
 
-#define raw_spin_lock_irqsave_rcu_node(rnp, flags)	\
-do {							\
-	typecheck(unsigned long, flags);		\
-	raw_spin_lock_irqsave(&(rnp)->lock, flags);	\
-	smp_mb__after_unlock_lock();			\
+static inline void raw_spin_unlock_irq_rcu_node(struct rcu_node *rnp)
+{
+	raw_spin_unlock_irq(&ACCESS_PRIVATE(rnp, lock));
+}
+
+#define raw_spin_lock_irqsave_rcu_node(rnp, flags)			\
+do {									\
+	typecheck(unsigned long, flags);				\
+	raw_spin_lock_irqsave(&ACCESS_PRIVATE(rnp, lock), flags);	\
+	smp_mb__after_unlock_lock();					\
+} while (0)
+
+#define raw_spin_unlock_irqrestore_rcu_node(rnp, flags)			\
+do {									\
+	typecheck(unsigned long, flags);				\
+	raw_spin_unlock_irqrestore(&ACCESS_PRIVATE(rnp, lock), flags);	\
 } while (0)
 
 static inline bool raw_spin_trylock_rcu_node(struct rcu_node *rnp)
 {
-	bool locked = raw_spin_trylock(&rnp->lock);
+	bool locked = raw_spin_trylock(&ACCESS_PRIVATE(rnp, lock));
 
 	if (locked)
 		smp_mb__after_unlock_lock();

kernel/rcu/tree_plugin.h

@@ -235,7 +235,7 @@ static void rcu_preempt_ctxt_queue(struct rcu_node *rnp, struct rcu_data *rdp)
 		rnp->gp_tasks = &t->rcu_node_entry;
 	if (!rnp->exp_tasks && (blkd_state & RCU_EXP_BLKD))
 		rnp->exp_tasks = &t->rcu_node_entry;
-	raw_spin_unlock(&rnp->lock); /* rrupts remain disabled. */
+	raw_spin_unlock_rcu_node(rnp); /* interrupts remain disabled. */
 
 	/*
 	 * Report the quiescent state for the expedited GP.  This expedited
@@ -489,7 +489,7 @@ void rcu_read_unlock_special(struct task_struct *t)
 							 !!rnp->gp_tasks);
 			rcu_report_unblock_qs_rnp(rcu_state_p, rnp, flags);
 		} else {
-			raw_spin_unlock_irqrestore(&rnp->lock, flags);
+			raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 		}
 
 		/* Unboost if we were boosted. */
@@ -518,14 +518,14 @@ static void rcu_print_detail_task_stall_rnp(struct rcu_node *rnp)
 
 	raw_spin_lock_irqsave_rcu_node(rnp, flags);
 	if (!rcu_preempt_blocked_readers_cgp(rnp)) {
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 		return;
 	}
 	t = list_entry(rnp->gp_tasks->prev,
 		       struct task_struct, rcu_node_entry);
 	list_for_each_entry_continue(t, &rnp->blkd_tasks, rcu_node_entry)
 		sched_show_task(t);
-	raw_spin_unlock_irqrestore(&rnp->lock, flags);
+	raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 }
 
 /*
@@ -807,7 +807,6 @@ void exit_rcu(void)
 #else /* #ifdef CONFIG_PREEMPT_RCU */
 
 static struct rcu_state *const rcu_state_p = &rcu_sched_state;
-static struct rcu_data __percpu *const rcu_data_p = &rcu_sched_data;
 
 /*
  * Tell them what RCU they are running.
@@ -991,7 +990,7 @@ static int rcu_boost(struct rcu_node *rnp)
 	 * might exit their RCU read-side critical sections on their own.
 	 */
 	if (rnp->exp_tasks == NULL && rnp->boost_tasks == NULL) {
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 		return 0;
 	}
 
@@ -1028,7 +1027,7 @@ static int rcu_boost(struct rcu_node *rnp)
 	 */
 	t = container_of(tb, struct task_struct, rcu_node_entry);
 	rt_mutex_init_proxy_locked(&rnp->boost_mtx, t);
-	raw_spin_unlock_irqrestore(&rnp->lock, flags);
+	raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 	/* Lock only for side effect: boosts task t's priority. */
 	rt_mutex_lock(&rnp->boost_mtx);
 	rt_mutex_unlock(&rnp->boost_mtx);  /* Then keep lockdep happy. */
@@ -1088,7 +1087,7 @@ static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags)
 
 	if (!rcu_preempt_blocked_readers_cgp(rnp) && rnp->exp_tasks == NULL) {
 		rnp->n_balk_exp_gp_tasks++;
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 		return;
 	}
 	if (rnp->exp_tasks != NULL ||
@@ -1098,13 +1097,13 @@ static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags)
 	     ULONG_CMP_GE(jiffies, rnp->boost_time))) {
 		if (rnp->exp_tasks == NULL)
 			rnp->boost_tasks = rnp->gp_tasks;
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 		t = rnp->boost_kthread_task;
 		if (t)
 			rcu_wake_cond(t, rnp->boost_kthread_status);
 	} else {
 		rcu_initiate_boost_trace(rnp);
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 	}
 }
 
@@ -1172,7 +1171,7 @@ static int rcu_spawn_one_boost_kthread(struct rcu_state *rsp,
 		return PTR_ERR(t);
 	raw_spin_lock_irqsave_rcu_node(rnp, flags);
 	rnp->boost_kthread_task = t;
-	raw_spin_unlock_irqrestore(&rnp->lock, flags);
+	raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 	sp.sched_priority = kthread_prio;
 	sched_setscheduler_nocheck(t, SCHED_FIFO, &sp);
 	wake_up_process(t); /* get to TASK_INTERRUPTIBLE quickly. */
@@ -1308,7 +1307,7 @@ static void rcu_prepare_kthreads(int cpu)
 static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags)
 	__releases(rnp->lock)
 {
-	raw_spin_unlock_irqrestore(&rnp->lock, flags);
+	raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 }
 
 static void invoke_rcu_callbacks_kthread(void)
@@ -1559,7 +1558,7 @@ static void rcu_prepare_for_idle(void)
 		rnp = rdp->mynode;
 		raw_spin_lock_rcu_node(rnp); /* irqs already disabled. */
 		needwake = rcu_accelerate_cbs(rsp, rnp, rdp);
-		raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */
+		raw_spin_unlock_rcu_node(rnp); /* irqs remain disabled. */
 		if (needwake)
 			rcu_gp_kthread_wake(rsp);
 	}
@@ -2064,7 +2063,7 @@ static void rcu_nocb_wait_gp(struct rcu_data *rdp)
 
 	raw_spin_lock_irqsave_rcu_node(rnp, flags);
 	needwake = rcu_start_future_gp(rnp, rdp, &c);
-	raw_spin_unlock_irqrestore(&rnp->lock, flags);
+	raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 	if (needwake)
 		rcu_gp_kthread_wake(rdp->rsp);
 

scripts/checkpatch.pl

@@ -269,7 +269,8 @@ our $Sparse	= qr{
 			__init_refok|
 			__kprobes|
 			__ref|
-			__rcu
+			__rcu|
+			__private
 		}x;
 our $InitAttributePrefix = qr{__(?:mem|cpu|dev|net_|)};
 our $InitAttributeData = qr{$InitAttributePrefix(?:initdata\b)};