Merge branches 'doc.2017.04.12a', 'fixes.2017.04.19a' and 'srcu.2017.04.21a' into HEAD

doc.2017.04.12a: Documentation updates
fixes.2017.04.19a: Miscellaneous fixes
srcu.2017.04.21a: Parallelize SRCU callback handling

Documentation/RCU/00-INDEX

@@ -17,7 +17,7 @@ rcu_dereference.txt
 rcubarrier.txt
 	- RCU and Unloadable Modules
 rculist_nulls.txt
-	- RCU list primitives for use with SLAB_DESTROY_BY_RCU
+	- RCU list primitives for use with SLAB_TYPESAFE_BY_RCU
 rcuref.txt
 	- Reference-count design for elements of lists/arrays protected by RCU
 rcu.txt

Documentation/RCU/Design/Data-Structures/Data-Structures.html

@@ -1185,6 +1185,9 @@ Its fields are as follows:
   1   int dynticks_nesting;
   2   int dynticks_nmi_nesting;
   3   atomic_t dynticks;
+  4   bool rcu_need_heavy_qs;
+  5   unsigned long rcu_qs_ctr;
+  6   bool rcu_urgent_qs;
 </pre>
 
 <p>The <tt>-&gt;dynticks_nesting</tt> field counts the
@@ -1198,11 +1201,32 @@ NMIs are counted by the <tt>-&gt;dynticks_nmi_nesting</tt>
 field, except that NMIs that interrupt non-dyntick-idle execution
 are not counted.
 
-</p><p>Finally, the <tt>-&gt;dynticks</tt> field counts the corresponding
+</p><p>The <tt>-&gt;dynticks</tt> field counts the corresponding
 CPU's transitions to and from dyntick-idle mode, so that this counter
 has an even value when the CPU is in dyntick-idle mode and an odd
 value otherwise.
 
+</p><p>The <tt>-&gt;rcu_need_heavy_qs</tt> field is used
+to record the fact that the RCU core code would really like to
+see a quiescent state from the corresponding CPU, so much so that
+it is willing to call for heavy-weight dyntick-counter operations.
+This flag is checked by RCU's context-switch and <tt>cond_resched()</tt>
+code, which provide a momentary idle sojourn in response.
+
+</p><p>The <tt>-&gt;rcu_qs_ctr</tt> field is used to record
+quiescent states from <tt>cond_resched()</tt>.
+Because <tt>cond_resched()</tt> can execute quite frequently, this
+must be quite lightweight, as in a non-atomic increment of this
+per-CPU field.
+
+</p><p>Finally, the <tt>-&gt;rcu_urgent_qs</tt> field is used to record
+the fact that the RCU core code would really like to see a quiescent
+state from the corresponding CPU, with the various other fields indicating
+just how badly RCU wants this quiescent state.
+This flag is checked by RCU's context-switch and <tt>cond_resched()</tt>
+code, which, if nothing else, non-atomically increment <tt>-&gt;rcu_qs_ctr</tt>
+in response.
+
 <table>
 <tr><th>&nbsp;</th></tr>
 <tr><th align="left">Quick Quiz:</th></tr>

Documentation/RCU/rculist_nulls.txt

@@ -1,5 +1,5 @@
 Using hlist_nulls to protect read-mostly linked lists and
-objects using SLAB_DESTROY_BY_RCU allocations.
+objects using SLAB_TYPESAFE_BY_RCU allocations.
 
 Please read the basics in Documentation/RCU/listRCU.txt
 
@@ -7,7 +7,7 @@ Using special makers (called 'nulls') is a convenient way
 to solve following problem :
 
 A typical RCU linked list managing objects which are
-allocated with SLAB_DESTROY_BY_RCU kmem_cache can
+allocated with SLAB_TYPESAFE_BY_RCU kmem_cache can
 use following algos :
 
 1) Lookup algo
@@ -96,7 +96,7 @@ unlock_chain(); // typically a spin_unlock()
 3) Remove algo
 --------------
 Nothing special here, we can use a standard RCU hlist deletion.
-But thanks to SLAB_DESTROY_BY_RCU, beware a deleted object can be reused
+But thanks to SLAB_TYPESAFE_BY_RCU, beware a deleted object can be reused
 very very fast (before the end of RCU grace period)
 
 if (put_last_reference_on(obj) {

Documentation/RCU/whatisRCU.txt

@@ -928,7 +928,8 @@ d.	Do you need RCU grace periods to complete even in the face
 
 e.	Is your workload too update-intensive for normal use of
 	RCU, but inappropriate for other synchronization mechanisms?
-	If so, consider SLAB_DESTROY_BY_RCU.  But please be careful!
+	If so, consider SLAB_TYPESAFE_BY_RCU (which was originally
+	named SLAB_DESTROY_BY_RCU).  But please be careful!
 
 f.	Do you need read-side critical sections that are respected
 	even though they are in the middle of the idle loop, during

arch/Kconfig

@@ -320,6 +320,9 @@ config HAVE_CMPXCHG_LOCAL
 config HAVE_CMPXCHG_DOUBLE
 	bool
 
+config ARCH_WEAK_RELEASE_ACQUIRE
+	bool
+
 config ARCH_WANT_IPC_PARSE_VERSION
 	bool
 

arch/powerpc/Kconfig

@@ -99,6 +99,7 @@ config PPC
 	select ARCH_USE_BUILTIN_BSWAP
 	select ARCH_USE_CMPXCHG_LOCKREF		if PPC64
 	select ARCH_WANT_IPC_PARSE_VERSION
+	select ARCH_WEAK_RELEASE_ACQUIRE
 	select BINFMT_ELF
 	select BUILDTIME_EXTABLE_SORT
 	select CLONE_BACKWARDS

drivers/gpu/drm/i915/i915_gem.c

@@ -4552,7 +4552,7 @@ i915_gem_load_init(struct drm_i915_private *dev_priv)
 	dev_priv->requests = KMEM_CACHE(drm_i915_gem_request,
 					SLAB_HWCACHE_ALIGN |
 					SLAB_RECLAIM_ACCOUNT |
-					SLAB_DESTROY_BY_RCU);
+					SLAB_TYPESAFE_BY_RCU);
 	if (!dev_priv->requests)
 		goto err_vmas;
 

drivers/gpu/drm/i915/i915_gem_request.h

@@ -493,7 +493,7 @@ static inline struct drm_i915_gem_request *
 __i915_gem_active_get_rcu(const struct i915_gem_active *active)
 {
 	/* Performing a lockless retrieval of the active request is super
-	 * tricky. SLAB_DESTROY_BY_RCU merely guarantees that the backing
+	 * tricky. SLAB_TYPESAFE_BY_RCU merely guarantees that the backing
 	 * slab of request objects will not be freed whilst we hold the
 	 * RCU read lock. It does not guarantee that the request itself
 	 * will not be freed and then *reused*. Viz,

drivers/staging/lustre/lustre/ldlm/ldlm_lockd.c

@@ -1071,7 +1071,7 @@ int ldlm_init(void)
 	ldlm_lock_slab = kmem_cache_create("ldlm_locks",
 					   sizeof(struct ldlm_lock), 0,
 					   SLAB_HWCACHE_ALIGN |
-					   SLAB_DESTROY_BY_RCU, NULL);
+					   SLAB_TYPESAFE_BY_RCU, NULL);
 	if (!ldlm_lock_slab) {
 		kmem_cache_destroy(ldlm_resource_slab);
 		return -ENOMEM;

fs/jbd2/journal.c

@@ -2340,7 +2340,7 @@ static int jbd2_journal_init_journal_head_cache(void)
 	jbd2_journal_head_cache = kmem_cache_create("jbd2_journal_head",
 				sizeof(struct journal_head),
 				0,		/* offset */
-				SLAB_TEMPORARY | SLAB_DESTROY_BY_RCU,
+				SLAB_TEMPORARY | SLAB_TYPESAFE_BY_RCU,
 				NULL);		/* ctor */
 	retval = 0;
 	if (!jbd2_journal_head_cache) {

fs/signalfd.c

@@ -38,7 +38,7 @@ void signalfd_cleanup(struct sighand_struct *sighand)
 	/*
 	 * The lockless check can race with remove_wait_queue() in progress,
 	 * but in this case its caller should run under rcu_read_lock() and
-	 * sighand_cachep is SLAB_DESTROY_BY_RCU, we can safely return.
+	 * sighand_cachep is SLAB_TYPESAFE_BY_RCU, we can safely return.
 	 */
 	if (likely(!waitqueue_active(wqh)))
 		return;

include/linux/dma-fence.h

@@ -229,7 +229,7 @@ static inline struct dma_fence *dma_fence_get_rcu(struct dma_fence *fence)
  *
  * Function returns NULL if no refcount could be obtained, or the fence.
  * This function handles acquiring a reference to a fence that may be
- * reallocated within the RCU grace period (such as with SLAB_DESTROY_BY_RCU),
+ * reallocated within the RCU grace period (such as with SLAB_TYPESAFE_BY_RCU),
  * so long as the caller is using RCU on the pointer to the fence.
  *
  * An alternative mechanism is to employ a seqlock to protect a bunch of
@@ -257,7 +257,7 @@ dma_fence_get_rcu_safe(struct dma_fence * __rcu *fencep)
 		 * have successfully acquire a reference to it. If it no
 		 * longer matches, we are holding a reference to some other
 		 * reallocated pointer. This is possible if the allocator
-		 * is using a freelist like SLAB_DESTROY_BY_RCU where the
+		 * is using a freelist like SLAB_TYPESAFE_BY_RCU where the
 		 * fence remains valid for the RCU grace period, but it
 		 * may be reallocated. When using such allocators, we are
 		 * responsible for ensuring the reference we get is to

include/linux/kvm_host.h

@@ -375,8 +375,6 @@ struct kvm {
 	struct mutex slots_lock;
 	struct mm_struct *mm; /* userspace tied to this vm */
 	struct kvm_memslots *memslots[KVM_ADDRESS_SPACE_NUM];
-	struct srcu_struct srcu;
-	struct srcu_struct irq_srcu;
 	struct kvm_vcpu *vcpus[KVM_MAX_VCPUS];
 
 	/*
@@ -429,6 +427,8 @@ struct kvm {
 	struct list_head devices;
 	struct dentry *debugfs_dentry;
 	struct kvm_stat_data **debugfs_stat_data;
+	struct srcu_struct srcu;
+	struct srcu_struct irq_srcu;
 };
 
 #define kvm_err(fmt, ...) \

include/linux/rcu_node_tree.h

@@ -0,0 +1,99 @@
+/*
+ * RCU node combining tree definitions.  These are used to compute
+ * global attributes while avoiding common-case global contention.  A key
+ * property that these computations rely on is a tournament-style approach
+ * where only one of the tasks contending a lower level in the tree need
+ * advance to the next higher level.  If properly configured, this allows
+ * unlimited scalability while maintaining a constant level of contention
+ * on the root node.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, you can access it online at
+ * http://www.gnu.org/licenses/gpl-2.0.html.
+ *
+ * Copyright IBM Corporation, 2017
+ *
+ * Author: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
+ */
+
+#ifndef __LINUX_RCU_NODE_TREE_H
+#define __LINUX_RCU_NODE_TREE_H
+
+/*
+ * Define shape of hierarchy based on NR_CPUS, CONFIG_RCU_FANOUT, and
+ * CONFIG_RCU_FANOUT_LEAF.
+ * In theory, it should be possible to add more levels straightforwardly.
+ * In practice, this did work well going from three levels to four.
+ * Of course, your mileage may vary.
+ */
+
+#ifdef CONFIG_RCU_FANOUT
+#define RCU_FANOUT CONFIG_RCU_FANOUT
+#else /* #ifdef CONFIG_RCU_FANOUT */
+# ifdef CONFIG_64BIT
+# define RCU_FANOUT 64
+# else
+# define RCU_FANOUT 32
+# endif
+#endif /* #else #ifdef CONFIG_RCU_FANOUT */
+
+#ifdef CONFIG_RCU_FANOUT_LEAF
+#define RCU_FANOUT_LEAF CONFIG_RCU_FANOUT_LEAF
+#else /* #ifdef CONFIG_RCU_FANOUT_LEAF */
+#define RCU_FANOUT_LEAF 16
+#endif /* #else #ifdef CONFIG_RCU_FANOUT_LEAF */
+
+#define RCU_FANOUT_1	      (RCU_FANOUT_LEAF)
+#define RCU_FANOUT_2	      (RCU_FANOUT_1 * RCU_FANOUT)
+#define RCU_FANOUT_3	      (RCU_FANOUT_2 * RCU_FANOUT)
+#define RCU_FANOUT_4	      (RCU_FANOUT_3 * RCU_FANOUT)
+
+#if NR_CPUS <= RCU_FANOUT_1
+#  define RCU_NUM_LVLS	      1
+#  define NUM_RCU_LVL_0	      1
+#  define NUM_RCU_NODES	      NUM_RCU_LVL_0
+#  define NUM_RCU_LVL_INIT    { NUM_RCU_LVL_0 }
+#  define RCU_NODE_NAME_INIT  { "rcu_node_0" }
+#  define RCU_FQS_NAME_INIT   { "rcu_node_fqs_0" }
+#elif NR_CPUS <= RCU_FANOUT_2
+#  define RCU_NUM_LVLS	      2
+#  define NUM_RCU_LVL_0	      1
+#  define NUM_RCU_LVL_1	      DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_1)
+#  define NUM_RCU_NODES	      (NUM_RCU_LVL_0 + NUM_RCU_LVL_1)
+#  define NUM_RCU_LVL_INIT    { NUM_RCU_LVL_0, NUM_RCU_LVL_1 }
+#  define RCU_NODE_NAME_INIT  { "rcu_node_0", "rcu_node_1" }
+#  define RCU_FQS_NAME_INIT   { "rcu_node_fqs_0", "rcu_node_fqs_1" }
+#elif NR_CPUS <= RCU_FANOUT_3
+#  define RCU_NUM_LVLS	      3
+#  define NUM_RCU_LVL_0	      1
+#  define NUM_RCU_LVL_1	      DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_2)
+#  define NUM_RCU_LVL_2	      DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_1)
+#  define NUM_RCU_NODES	      (NUM_RCU_LVL_0 + NUM_RCU_LVL_1 + NUM_RCU_LVL_2)
+#  define NUM_RCU_LVL_INIT    { NUM_RCU_LVL_0, NUM_RCU_LVL_1, NUM_RCU_LVL_2 }
+#  define RCU_NODE_NAME_INIT  { "rcu_node_0", "rcu_node_1", "rcu_node_2" }
+#  define RCU_FQS_NAME_INIT   { "rcu_node_fqs_0", "rcu_node_fqs_1", "rcu_node_fqs_2" }
+#elif NR_CPUS <= RCU_FANOUT_4
+#  define RCU_NUM_LVLS	      4
+#  define NUM_RCU_LVL_0	      1
+#  define NUM_RCU_LVL_1	      DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_3)
+#  define NUM_RCU_LVL_2	      DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_2)
+#  define NUM_RCU_LVL_3	      DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_1)
+#  define NUM_RCU_NODES	      (NUM_RCU_LVL_0 + NUM_RCU_LVL_1 + NUM_RCU_LVL_2 + NUM_RCU_LVL_3)
+#  define NUM_RCU_LVL_INIT    { NUM_RCU_LVL_0, NUM_RCU_LVL_1, NUM_RCU_LVL_2, NUM_RCU_LVL_3 }
+#  define RCU_NODE_NAME_INIT  { "rcu_node_0", "rcu_node_1", "rcu_node_2", "rcu_node_3" }
+#  define RCU_FQS_NAME_INIT   { "rcu_node_fqs_0", "rcu_node_fqs_1", "rcu_node_fqs_2", "rcu_node_fqs_3" }
+#else
+# error "CONFIG_RCU_FANOUT insufficient for NR_CPUS"
+#endif /* #if (NR_CPUS) <= RCU_FANOUT_1 */
+
+#endif /* __LINUX_RCU_NODE_TREE_H */

include/linux/rcu_segcblist.h

@@ -0,0 +1,712 @@
+/*
+ * RCU segmented callback lists
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, you can access it online at
+ * http://www.gnu.org/licenses/gpl-2.0.html.
+ *
+ * Copyright IBM Corporation, 2017
+ *
+ * Authors: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
+ */
+
+#ifndef __KERNEL_RCU_SEGCBLIST_H
+#define __KERNEL_RCU_SEGCBLIST_H
+
+/* Simple unsegmented callback lists. */
+struct rcu_cblist {
+	struct rcu_head *head;
+	struct rcu_head **tail;
+	long len;
+	long len_lazy;
+};
+
+#define RCU_CBLIST_INITIALIZER(n) { .head = NULL, .tail = &n.head }
+
+/* Initialize simple callback list. */
+static inline void rcu_cblist_init(struct rcu_cblist *rclp)
+{
+	rclp->head = NULL;
+	rclp->tail = &rclp->head;
+	rclp->len = 0;
+	rclp->len_lazy = 0;
+}
+
+/* Is simple callback list empty? */
+static inline bool rcu_cblist_empty(struct rcu_cblist *rclp)
+{
+	return !rclp->head;
+}
+
+/* Return number of callbacks in simple callback list. */
+static inline long rcu_cblist_n_cbs(struct rcu_cblist *rclp)
+{
+	return rclp->len;
+}
+
+/* Return number of lazy callbacks in simple callback list. */
+static inline long rcu_cblist_n_lazy_cbs(struct rcu_cblist *rclp)
+{
+	return rclp->len_lazy;
+}
+
+/*
+ * Debug function to actually count the number of callbacks.
+ * If the number exceeds the limit specified, return -1.
+ */
+static inline long rcu_cblist_count_cbs(struct rcu_cblist *rclp, long lim)
+{
+	int cnt = 0;
+	struct rcu_head **rhpp = &rclp->head;
+
+	for (;;) {
+		if (!*rhpp)
+			return cnt;
+		if (++cnt > lim)
+			return -1;
+		rhpp = &(*rhpp)->next;
+	}
+}
+
+/*
+ * Dequeue the oldest rcu_head structure from the specified callback
+ * list.  This function assumes that the callback is non-lazy, but
+ * the caller can later invoke rcu_cblist_dequeued_lazy() if it
+ * finds otherwise (and if it cares about laziness).  This allows
+ * different users to have different ways of determining laziness.
+ */
+static inline struct rcu_head *rcu_cblist_dequeue(struct rcu_cblist *rclp)
+{
+	struct rcu_head *rhp;
+
+	rhp = rclp->head;
+	if (!rhp)
+		return NULL;
+	rclp->len--;
+	rclp->head = rhp->next;
+	if (!rclp->head)
+		rclp->tail = &rclp->head;
+	return rhp;
+}
+
+/*
+ * Account for the fact that a previously dequeued callback turned out
+ * to be marked as lazy.
+ */
+static inline void rcu_cblist_dequeued_lazy(struct rcu_cblist *rclp)
+{
+	rclp->len_lazy--;
+}
+
+/*
+ * Interim function to return rcu_cblist head pointer.  Longer term, the
+ * rcu_cblist will be used more pervasively, removing the need for this
+ * function.
+ */
+static inline struct rcu_head *rcu_cblist_head(struct rcu_cblist *rclp)
+{
+	return rclp->head;
+}
+
+/*
+ * Interim function to return rcu_cblist head pointer.  Longer term, the
+ * rcu_cblist will be used more pervasively, removing the need for this
+ * function.
+ */
+static inline struct rcu_head **rcu_cblist_tail(struct rcu_cblist *rclp)
+{
+	WARN_ON_ONCE(rcu_cblist_empty(rclp));
+	return rclp->tail;
+}
+
+/* Complicated segmented callback lists.  ;-) */
+
+/*
+ * Index values for segments in rcu_segcblist structure.
+ *
+ * The segments are as follows:
+ *
+ * [head, *tails[RCU_DONE_TAIL]):
+ *	Callbacks whose grace period has elapsed, and thus can be invoked.
+ * [*tails[RCU_DONE_TAIL], *tails[RCU_WAIT_TAIL]):
+ *	Callbacks waiting for the current GP from the current CPU's viewpoint.
+ * [*tails[RCU_WAIT_TAIL], *tails[RCU_NEXT_READY_TAIL]):
+ *	Callbacks that arrived before the next GP started, again from
+ *	the current CPU's viewpoint.  These can be handled by the next GP.
+ * [*tails[RCU_NEXT_READY_TAIL], *tails[RCU_NEXT_TAIL]):
+ *	Callbacks that might have arrived after the next GP started.
+ *	There is some uncertainty as to when a given GP starts and
+ *	ends, but a CPU knows the exact times if it is the one starting
+ *	or ending the GP.  Other CPUs know that the previous GP ends
+ *	before the next one starts.
+ *
+ * Note that RCU_WAIT_TAIL cannot be empty unless RCU_NEXT_READY_TAIL is also
+ * empty.
+ *
+ * The ->gp_seq[] array contains the grace-period number at which the
+ * corresponding segment of callbacks will be ready to invoke.  A given
+ * element of this array is meaningful only when the corresponding segment
+ * is non-empty, and it is never valid for RCU_DONE_TAIL (whose callbacks
+ * are already ready to invoke) or for RCU_NEXT_TAIL (whose callbacks have
+ * not yet been assigned a grace-period number).
+ */
+#define RCU_DONE_TAIL		0	/* Also RCU_WAIT head. */
+#define RCU_WAIT_TAIL		1	/* Also RCU_NEXT_READY head. */
+#define RCU_NEXT_READY_TAIL	2	/* Also RCU_NEXT head. */
+#define RCU_NEXT_TAIL		3
+#define RCU_CBLIST_NSEGS	4
+
+struct rcu_segcblist {
+	struct rcu_head *head;
+	struct rcu_head **tails[RCU_CBLIST_NSEGS];
+	unsigned long gp_seq[RCU_CBLIST_NSEGS];
+	long len;
+	long len_lazy;
+};
+
+#define RCU_SEGCBLIST_INITIALIZER(n) \
+{ \
+	.head = NULL, \
+	.tails[RCU_DONE_TAIL] = &n.head, \
+	.tails[RCU_WAIT_TAIL] = &n.head, \
+	.tails[RCU_NEXT_READY_TAIL] = &n.head, \
+	.tails[RCU_NEXT_TAIL] = &n.head, \
+}
+
+/*
+ * Initialize an rcu_segcblist structure.
+ */
+static inline void rcu_segcblist_init(struct rcu_segcblist *rsclp)
+{
+	int i;
+
+	BUILD_BUG_ON(RCU_NEXT_TAIL + 1 != ARRAY_SIZE(rsclp->gp_seq));
+	BUILD_BUG_ON(ARRAY_SIZE(rsclp->tails) != ARRAY_SIZE(rsclp->gp_seq));
+	rsclp->head = NULL;
+	for (i = 0; i < RCU_CBLIST_NSEGS; i++)
+		rsclp->tails[i] = &rsclp->head;
+	rsclp->len = 0;
+	rsclp->len_lazy = 0;
+}
+
+/*
+ * Is the specified rcu_segcblist structure empty?
+ *
+ * But careful!  The fact that the ->head field is NULL does not
+ * necessarily imply that there are no callbacks associated with
+ * this structure.  When callbacks are being invoked, they are
+ * removed as a group.  If callback invocation must be preempted,
+ * the remaining callbacks will be added back to the list.  Either
+ * way, the counts are updated later.
+ *
+ * So it is often the case that rcu_segcblist_n_cbs() should be used
+ * instead.
+ */
+static inline bool rcu_segcblist_empty(struct rcu_segcblist *rsclp)
+{
+	return !rsclp->head;
+}
+
+/* Return number of callbacks in segmented callback list. */
+static inline long rcu_segcblist_n_cbs(struct rcu_segcblist *rsclp)
+{
+	return READ_ONCE(rsclp->len);
+}
+
+/* Return number of lazy callbacks in segmented callback list. */
+static inline long rcu_segcblist_n_lazy_cbs(struct rcu_segcblist *rsclp)
+{
+	return rsclp->len_lazy;
+}
+
+/* Return number of lazy callbacks in segmented callback list. */
+static inline long rcu_segcblist_n_nonlazy_cbs(struct rcu_segcblist *rsclp)
+{
+	return rsclp->len - rsclp->len_lazy;
+}
+
+/*
+ * Is the specified rcu_segcblist enabled, for example, not corresponding
+ * to an offline or callback-offloaded CPU?
+ */
+static inline bool rcu_segcblist_is_enabled(struct rcu_segcblist *rsclp)
+{
+	return !!rsclp->tails[RCU_NEXT_TAIL];
+}
+
+/*
+ * Disable the specified rcu_segcblist structure, so that callbacks can
+ * no longer be posted to it.  This structure must be empty.
+ */
+static inline void rcu_segcblist_disable(struct rcu_segcblist *rsclp)
+{
+	WARN_ON_ONCE(!rcu_segcblist_empty(rsclp));
+	WARN_ON_ONCE(rcu_segcblist_n_cbs(rsclp));
+	WARN_ON_ONCE(rcu_segcblist_n_lazy_cbs(rsclp));
+	rsclp->tails[RCU_NEXT_TAIL] = NULL;
+}
+
+/*
+ * Is the specified segment of the specified rcu_segcblist structure
+ * empty of callbacks?
+ */
+static inline bool rcu_segcblist_segempty(struct rcu_segcblist *rsclp, int seg)
+{
+	if (seg == RCU_DONE_TAIL)
+		return &rsclp->head == rsclp->tails[RCU_DONE_TAIL];
+	return rsclp->tails[seg - 1] == rsclp->tails[seg];
+}
+
+/*
+ * Are all segments following the specified segment of the specified
+ * rcu_segcblist structure empty of callbacks?  (The specified
+ * segment might well contain callbacks.)
+ */
+static inline bool rcu_segcblist_restempty(struct rcu_segcblist *rsclp, int seg)
+{
+	return !*rsclp->tails[seg];
+}
+
+/*
+ * Does the specified rcu_segcblist structure contain callbacks that
+ * are ready to be invoked?
+ */
+static inline bool rcu_segcblist_ready_cbs(struct rcu_segcblist *rsclp)
+{
+	return rcu_segcblist_is_enabled(rsclp) &&
+	       &rsclp->head != rsclp->tails[RCU_DONE_TAIL];
+}
+
+/*
+ * Does the specified rcu_segcblist structure contain callbacks that
+ * are still pending, that is, not yet ready to be invoked?
+ */
+static inline bool rcu_segcblist_pend_cbs(struct rcu_segcblist *rsclp)
+{
+	return rcu_segcblist_is_enabled(rsclp) &&
+	       !rcu_segcblist_restempty(rsclp, RCU_DONE_TAIL);
+}
+
+/*
+ * Dequeue and return the first ready-to-invoke callback.  If there
+ * are no ready-to-invoke callbacks, return NULL.  Disables interrupts
+ * to avoid interference.  Does not protect from interference from other
+ * CPUs or tasks.
+ */
+static inline struct rcu_head *
+rcu_segcblist_dequeue(struct rcu_segcblist *rsclp)
+{
+	unsigned long flags;
+	int i;
+	struct rcu_head *rhp;
+
+	local_irq_save(flags);
+	if (!rcu_segcblist_ready_cbs(rsclp)) {
+		local_irq_restore(flags);
+		return NULL;
+	}
+	rhp = rsclp->head;
+	BUG_ON(!rhp);
+	rsclp->head = rhp->next;
+	for (i = RCU_DONE_TAIL; i < RCU_CBLIST_NSEGS; i++) {
+		if (rsclp->tails[i] != &rhp->next)
+			break;
+		rsclp->tails[i] = &rsclp->head;
+	}
+	smp_mb(); /* Dequeue before decrement for rcu_barrier(). */
+	WRITE_ONCE(rsclp->len, rsclp->len - 1);
+	local_irq_restore(flags);
+	return rhp;
+}
+
+/*
+ * Account for the fact that a previously dequeued callback turned out
+ * to be marked as lazy.
+ */
+static inline void rcu_segcblist_dequeued_lazy(struct rcu_segcblist *rsclp)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	rsclp->len_lazy--;
+	local_irq_restore(flags);
+}
+
+/*
+ * Return a pointer to the first callback in the specified rcu_segcblist
+ * structure.  This is useful for diagnostics.
+ */
+static inline struct rcu_head *
+rcu_segcblist_first_cb(struct rcu_segcblist *rsclp)
+{
+	if (rcu_segcblist_is_enabled(rsclp))
+		return rsclp->head;
+	return NULL;
+}
+
+/*
+ * Return a pointer to the first pending callback in the specified
+ * rcu_segcblist structure.  This is useful just after posting a given
+ * callback -- if that callback is the first pending callback, then
+ * you cannot rely on someone else having already started up the required
+ * grace period.
+ */
+static inline struct rcu_head *
+rcu_segcblist_first_pend_cb(struct rcu_segcblist *rsclp)
+{
+	if (rcu_segcblist_is_enabled(rsclp))
+		return *rsclp->tails[RCU_DONE_TAIL];
+	return NULL;
+}
+
+/*
+ * Does the specified rcu_segcblist structure contain callbacks that
+ * have not yet been processed beyond having been posted, that is,
+ * does it contain callbacks in its last segment?
+ */
+static inline bool rcu_segcblist_new_cbs(struct rcu_segcblist *rsclp)
+{
+	return rcu_segcblist_is_enabled(rsclp) &&
+	       !rcu_segcblist_restempty(rsclp, RCU_NEXT_READY_TAIL);
+}
+
+/*
+ * Enqueue the specified callback onto the specified rcu_segcblist
+ * structure, updating accounting as needed.  Note that the ->len
+ * field may be accessed locklessly, hence the WRITE_ONCE().
+ * The ->len field is used by rcu_barrier() and friends to determine
+ * if it must post a callback on this structure, and it is OK
+ * for rcu_barrier() to sometimes post callbacks needlessly, but
+ * absolutely not OK for it to ever miss posting a callback.
+ */
+static inline void rcu_segcblist_enqueue(struct rcu_segcblist *rsclp,
+					 struct rcu_head *rhp, bool lazy)
+{
+	WRITE_ONCE(rsclp->len, rsclp->len + 1); /* ->len sampled locklessly. */
+	if (lazy)
+		rsclp->len_lazy++;
+	smp_mb(); /* Ensure counts are updated before callback is enqueued. */
+	rhp->next = NULL;
+	*rsclp->tails[RCU_NEXT_TAIL] = rhp;
+	rsclp->tails[RCU_NEXT_TAIL] = &rhp->next;
+}
+
+/*
+ * Entrain the specified callback onto the specified rcu_segcblist at
+ * the end of the last non-empty segment.  If the entire rcu_segcblist
+ * is empty, make no change, but return false.
+ *
+ * This is intended for use by rcu_barrier()-like primitives, -not-
+ * for normal grace-period use.  IMPORTANT:  The callback you enqueue
+ * will wait for all prior callbacks, NOT necessarily for a grace
+ * period.  You have been warned.
+ */
+static inline bool rcu_segcblist_entrain(struct rcu_segcblist *rsclp,
+					 struct rcu_head *rhp, bool lazy)
+{
+	int i;
+
+	if (rcu_segcblist_n_cbs(rsclp) == 0)
+		return false;
+	WRITE_ONCE(rsclp->len, rsclp->len + 1);
+	if (lazy)
+		rsclp->len_lazy++;
+	smp_mb(); /* Ensure counts are updated before callback is entrained. */
+	rhp->next = NULL;
+	for (i = RCU_NEXT_TAIL; i > RCU_DONE_TAIL; i--)
+		if (rsclp->tails[i] != rsclp->tails[i - 1])
+			break;
+	*rsclp->tails[i] = rhp;
+	for (; i <= RCU_NEXT_TAIL; i++)
+		rsclp->tails[i] = &rhp->next;
+	return true;
+}
+
+/*
+ * Extract only the counts from the specified rcu_segcblist structure,
+ * and place them in the specified rcu_cblist structure.  This function
+ * supports both callback orphaning and invocation, hence the separation
+ * of counts and callbacks.  (Callbacks ready for invocation must be
+ * orphaned and adopted separately from pending callbacks, but counts
+ * apply to all callbacks.  Locking must be used to make sure that
+ * both orphaned-callbacks lists are consistent.)
+ */
+static inline void rcu_segcblist_extract_count(struct rcu_segcblist *rsclp,
+					       struct rcu_cblist *rclp)
+{
+	rclp->len_lazy += rsclp->len_lazy;
+	rclp->len += rsclp->len;
+	rsclp->len_lazy = 0;
+	WRITE_ONCE(rsclp->len, 0); /* ->len sampled locklessly. */
+}
+
+/*
+ * Extract only those callbacks ready to be invoked from the specified
+ * rcu_segcblist structure and place them in the specified rcu_cblist
+ * structure.
+ */
+static inline void rcu_segcblist_extract_done_cbs(struct rcu_segcblist *rsclp,
+						  struct rcu_cblist *rclp)
+{
+	int i;
+
+	if (!rcu_segcblist_ready_cbs(rsclp))
+		return; /* Nothing to do. */
+	*rclp->tail = rsclp->head;
+	rsclp->head = *rsclp->tails[RCU_DONE_TAIL];
+	*rsclp->tails[RCU_DONE_TAIL] = NULL;
+	rclp->tail = rsclp->tails[RCU_DONE_TAIL];
+	for (i = RCU_CBLIST_NSEGS - 1; i >= RCU_DONE_TAIL; i--)
+		if (rsclp->tails[i] == rsclp->tails[RCU_DONE_TAIL])
+			rsclp->tails[i] = &rsclp->head;
+}
+
+/*
+ * Extract only those callbacks still pending (not yet ready to be
+ * invoked) from the specified rcu_segcblist structure and place them in
+ * the specified rcu_cblist structure.  Note that this loses information
+ * about any callbacks that might have been partway done waiting for
+ * their grace period.  Too bad!  They will have to start over.
+ */
+static inline void
+rcu_segcblist_extract_pend_cbs(struct rcu_segcblist *rsclp,
+			       struct rcu_cblist *rclp)
+{
+	int i;
+
+	if (!rcu_segcblist_pend_cbs(rsclp))
+		return; /* Nothing to do. */
+	*rclp->tail = *rsclp->tails[RCU_DONE_TAIL];
+	rclp->tail = rsclp->tails[RCU_NEXT_TAIL];
+	*rsclp->tails[RCU_DONE_TAIL] = NULL;
+	for (i = RCU_DONE_TAIL + 1; i < RCU_CBLIST_NSEGS; i++)
+		rsclp->tails[i] = rsclp->tails[RCU_DONE_TAIL];
+}
+
+/*
+ * Move the entire contents of the specified rcu_segcblist structure,
+ * counts, callbacks, and all, to the specified rcu_cblist structure.
+ * @@@ Why do we need this???  Moving early-boot CBs to NOCB lists?
+ * @@@ Memory barrier needed?  (Not if only used at boot time...)
+ */
+static inline void rcu_segcblist_extract_all(struct rcu_segcblist *rsclp,
+					     struct rcu_cblist *rclp)
+{
+	rcu_segcblist_extract_done_cbs(rsclp, rclp);
+	rcu_segcblist_extract_pend_cbs(rsclp, rclp);
+	rcu_segcblist_extract_count(rsclp, rclp);
+}
+
+/*
+ * Insert counts from the specified rcu_cblist structure in the
+ * specified rcu_segcblist structure.
+ */
+static inline void rcu_segcblist_insert_count(struct rcu_segcblist *rsclp,
+					      struct rcu_cblist *rclp)
+{
+	rsclp->len_lazy += rclp->len_lazy;
+	/* ->len sampled locklessly. */
+	WRITE_ONCE(rsclp->len, rsclp->len + rclp->len);
+	rclp->len_lazy = 0;
+	rclp->len = 0;
+}
+
+/*
+ * Move callbacks from the specified rcu_cblist to the beginning of the
+ * done-callbacks segment of the specified rcu_segcblist.
+ */
+static inline void rcu_segcblist_insert_done_cbs(struct rcu_segcblist *rsclp,
+						 struct rcu_cblist *rclp)
+{
+	int i;
+
+	if (!rclp->head)
+		return; /* No callbacks to move. */
+	*rclp->tail = rsclp->head;
+	rsclp->head = rclp->head;
+	for (i = RCU_DONE_TAIL; i < RCU_CBLIST_NSEGS; i++)
+		if (&rsclp->head == rsclp->tails[i])
+			rsclp->tails[i] = rclp->tail;
+		else
+			break;
+	rclp->head = NULL;
+	rclp->tail = &rclp->head;
+}
+
+/*
+ * Move callbacks from the specified rcu_cblist to the end of the
+ * new-callbacks segment of the specified rcu_segcblist.
+ */
+static inline void rcu_segcblist_insert_pend_cbs(struct rcu_segcblist *rsclp,
+						 struct rcu_cblist *rclp)
+{
+	if (!rclp->head)
+		return; /* Nothing to do. */
+	*rsclp->tails[RCU_NEXT_TAIL] = rclp->head;
+	rsclp->tails[RCU_NEXT_TAIL] = rclp->tail;
+	rclp->head = NULL;
+	rclp->tail = &rclp->head;
+}
+
+/*
+ * Advance the callbacks in the specified rcu_segcblist structure based
+ * on the current value passed in for the grace-period counter.
+ */
+static inline void rcu_segcblist_advance(struct rcu_segcblist *rsclp,
+					 unsigned long seq)
+{
+	int i, j;
+
+	WARN_ON_ONCE(!rcu_segcblist_is_enabled(rsclp));
+	if (rcu_segcblist_restempty(rsclp, RCU_DONE_TAIL))
+		return;
+
+	/*
+	 * Find all callbacks whose ->gp_seq numbers indicate that they
+	 * are ready to invoke, and put them into the RCU_DONE_TAIL segment.
+	 */
+	for (i = RCU_WAIT_TAIL; i < RCU_NEXT_TAIL; i++) {
+		if (ULONG_CMP_LT(seq, rsclp->gp_seq[i]))
+			break;
+		rsclp->tails[RCU_DONE_TAIL] = rsclp->tails[i];
+	}
+
+	/* If no callbacks moved, nothing more need be done. */
+	if (i == RCU_WAIT_TAIL)
+		return;
+
+	/* Clean up tail pointers that might have been misordered above. */
+	for (j = RCU_WAIT_TAIL; j < i; j++)
+		rsclp->tails[j] = rsclp->tails[RCU_DONE_TAIL];
+
+	/*
+	 * Callbacks moved, so clean up the misordered ->tails[] pointers
+	 * that now point into the middle of the list of ready-to-invoke
+	 * callbacks.  The overall effect is to copy down the later pointers
+	 * into the gap that was created by the now-ready segments.
+	 */
+	for (j = RCU_WAIT_TAIL; i < RCU_NEXT_TAIL; i++, j++) {
+		if (rsclp->tails[j] == rsclp->tails[RCU_NEXT_TAIL])
+			break;  /* No more callbacks. */
+		rsclp->tails[j] = rsclp->tails[i];
+		rsclp->gp_seq[j] = rsclp->gp_seq[i];
+	}
+}
+
+/*
+ * "Accelerate" callbacks based on more-accurate grace-period information.
+ * The reason for this is that RCU does not synchronize the beginnings and
+ * ends of grace periods, and that callbacks are posted locally.  This in
+ * turn means that the callbacks must be labelled conservatively early
+ * on, as getting exact information would degrade both performance and
+ * scalability.  When more accurate grace-period information becomes
+ * available, previously posted callbacks can be "accelerated", marking
+ * them to complete at the end of the earlier grace period.
+ *
+ * This function operates on an rcu_segcblist structure, and also the
+ * grace-period sequence number seq at which new callbacks would become
+ * ready to invoke.  Returns true if there are callbacks that won't be
+ * ready to invoke until seq, false otherwise.
+ */
+static inline bool rcu_segcblist_accelerate(struct rcu_segcblist *rsclp,
+					    unsigned long seq)
+{
+	int i;
+
+	WARN_ON_ONCE(!rcu_segcblist_is_enabled(rsclp));
+	if (rcu_segcblist_restempty(rsclp, RCU_DONE_TAIL))
+		return false;
+
+	/*
+	 * Find the segment preceding the oldest segment of callbacks
+	 * whose ->gp_seq[] completion is at or after that passed in via
+	 * "seq", skipping any empty segments.  This oldest segment, along
+	 * with any later segments, can be merged in with any newly arrived
+	 * callbacks in the RCU_NEXT_TAIL segment, and assigned "seq"
+	 * as their ->gp_seq[] grace-period completion sequence number.
+	 */
+	for (i = RCU_NEXT_READY_TAIL; i > RCU_DONE_TAIL; i--)
+		if (rsclp->tails[i] != rsclp->tails[i - 1] &&
+		    ULONG_CMP_LT(rsclp->gp_seq[i], seq))
+			break;
+
+	/*
+	 * If all the segments contain callbacks that correspond to
+	 * earlier grace-period sequence numbers than "seq", leave.
+	 * Assuming that the rcu_segcblist structure has enough
+	 * segments in its arrays, this can only happen if some of
+	 * the non-done segments contain callbacks that really are
+	 * ready to invoke.  This situation will get straightened
+	 * out by the next call to rcu_segcblist_advance().
+	 *
+	 * Also advance to the oldest segment of callbacks whose
+	 * ->gp_seq[] completion is at or after that passed in via "seq",
+	 * skipping any empty segments.
+	 */
+	if (++i >= RCU_NEXT_TAIL)
+		return false;
+
+	/*
+	 * Merge all later callbacks, including newly arrived callbacks,
+	 * into the segment located by the for-loop above.  Assign "seq"
+	 * as the ->gp_seq[] value in order to correctly handle the case
+	 * where there were no pending callbacks in the rcu_segcblist
+	 * structure other than in the RCU_NEXT_TAIL segment.
+	 */
+	for (; i < RCU_NEXT_TAIL; i++) {
+		rsclp->tails[i] = rsclp->tails[RCU_NEXT_TAIL];
+		rsclp->gp_seq[i] = seq;
+	}
+	return true;
+}
+
+/*
+ * Scan the specified rcu_segcblist structure for callbacks that need
+ * a grace period later than the one specified by "seq".  We don't look
+ * at the RCU_DONE_TAIL or RCU_NEXT_TAIL segments because they don't
+ * have a grace-period sequence number.
+ */
+static inline bool rcu_segcblist_future_gp_needed(struct rcu_segcblist *rsclp,
+						  unsigned long seq)
+{
+	int i;
+
+	for (i = RCU_WAIT_TAIL; i < RCU_NEXT_TAIL; i++)
+		if (rsclp->tails[i - 1] != rsclp->tails[i] &&
+		    ULONG_CMP_LT(seq, rsclp->gp_seq[i]))
+			return true;
+	return false;
+}
+
+/*
+ * Interim function to return rcu_segcblist head pointer.  Longer term, the
+ * rcu_segcblist will be used more pervasively, removing the need for this
+ * function.
+ */
+static inline struct rcu_head *rcu_segcblist_head(struct rcu_segcblist *rsclp)
+{
+	return rsclp->head;
+}
+
+/*
+ * Interim function to return rcu_segcblist head pointer.  Longer term, the
+ * rcu_segcblist will be used more pervasively, removing the need for this
+ * function.
+ */
+static inline struct rcu_head **rcu_segcblist_tail(struct rcu_segcblist *rsclp)
+{
+	WARN_ON_ONCE(rcu_segcblist_empty(rsclp));
+	return rsclp->tails[RCU_NEXT_TAIL];
+}
+
+#endif /* __KERNEL_RCU_SEGCBLIST_H */

include/linux/rculist.h

@@ -509,7 +509,8 @@ static inline void hlist_add_tail_rcu(struct hlist_node *n,
 {
 	struct hlist_node *i, *last = NULL;
 
-	for (i = hlist_first_rcu(h); i; i = hlist_next_rcu(i))
+	/* Note: write side code, so rcu accessors are not needed. */
+	for (i = h->first; i; i = i->next)
 		last = i;
 
 	if (last) {

include/linux/rcupdate.h

@@ -363,15 +363,20 @@ static inline void rcu_init_nohz(void)
 #ifdef CONFIG_TASKS_RCU
 #define TASKS_RCU(x) x
 extern struct srcu_struct tasks_rcu_exit_srcu;
-#define rcu_note_voluntary_context_switch(t) \
+#define rcu_note_voluntary_context_switch_lite(t) \
 	do { \
-		rcu_all_qs(); \
 		if (READ_ONCE((t)->rcu_tasks_holdout)) \
 			WRITE_ONCE((t)->rcu_tasks_holdout, false); \
 	} while (0)
+#define rcu_note_voluntary_context_switch(t) \
+	do { \
+		rcu_all_qs(); \
+		rcu_note_voluntary_context_switch_lite(t); \
+	} while (0)
 #else /* #ifdef CONFIG_TASKS_RCU */
 #define TASKS_RCU(x) do { } while (0)
-#define rcu_note_voluntary_context_switch(t)	rcu_all_qs()
+#define rcu_note_voluntary_context_switch_lite(t)	do { } while (0)
+#define rcu_note_voluntary_context_switch(t)		rcu_all_qs()
 #endif /* #else #ifdef CONFIG_TASKS_RCU */
 
 /**
@@ -1127,11 +1132,11 @@ do { \
  * if the UNLOCK and LOCK are executed by the same CPU or if the
  * UNLOCK and LOCK operate on the same lock variable.
  */
-#ifdef CONFIG_PPC
+#ifdef CONFIG_ARCH_WEAK_RELEASE_ACQUIRE
 #define smp_mb__after_unlock_lock()	smp_mb()  /* Full ordering for lock. */
-#else /* #ifdef CONFIG_PPC */
+#else /* #ifdef CONFIG_ARCH_WEAK_RELEASE_ACQUIRE */
 #define smp_mb__after_unlock_lock()	do { } while (0)
-#endif /* #else #ifdef CONFIG_PPC */
+#endif /* #else #ifdef CONFIG_ARCH_WEAK_RELEASE_ACQUIRE */
 
 
 #endif /* __LINUX_RCUPDATE_H */

include/linux/rcutiny.h

@@ -33,6 +33,11 @@ static inline int rcu_dynticks_snap(struct rcu_dynticks *rdtp)
 	return 0;
 }
 
+static inline bool rcu_eqs_special_set(int cpu)
+{
+	return false;  /* Never flag non-existent other CPUs! */
+}
+
 static inline unsigned long get_state_synchronize_rcu(void)
 {
 	return 0;
@@ -87,10 +92,11 @@ static inline void kfree_call_rcu(struct rcu_head *head,
 	call_rcu(head, func);
 }
 
-static inline void rcu_note_context_switch(void)
-{
-	rcu_sched_qs();
-}
+#define rcu_note_context_switch(preempt) \
+	do { \
+		rcu_sched_qs(); \
+		rcu_note_voluntary_context_switch_lite(current); \
+	} while (0)
 
 /*
  * Take advantage of the fact that there is only one CPU, which
@@ -212,14 +218,14 @@ static inline void exit_rcu(void)
 {
 }
 
-#ifdef CONFIG_DEBUG_LOCK_ALLOC
+#if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_SRCU)
 extern int rcu_scheduler_active __read_mostly;
 void rcu_scheduler_starting(void);
-#else /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
+#else /* #if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_SRCU) */
 static inline void rcu_scheduler_starting(void)
 {
 }
-#endif /* #else #ifdef CONFIG_DEBUG_LOCK_ALLOC */
+#endif /* #else #if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_SRCU) */
 
 #if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_RCU_TRACE)
 
@@ -237,6 +243,10 @@ static inline bool rcu_is_watching(void)
 
 #endif /* #else defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_RCU_TRACE) */
 
+static inline void rcu_request_urgent_qs_task(struct task_struct *t)
+{
+}
+
 static inline void rcu_all_qs(void)
 {
 	barrier(); /* Avoid RCU read-side critical sections leaking across. */

include/linux/rcutree.h

@@ -30,7 +30,7 @@
 #ifndef __LINUX_RCUTREE_H
 #define __LINUX_RCUTREE_H
 
-void rcu_note_context_switch(void);
+void rcu_note_context_switch(bool preempt);
 int rcu_needs_cpu(u64 basem, u64 *nextevt);
 void rcu_cpu_stall_reset(void);
 
@@ -41,7 +41,7 @@ void rcu_cpu_stall_reset(void);
  */
 static inline void rcu_virt_note_context_switch(int cpu)
 {
-	rcu_note_context_switch();
+	rcu_note_context_switch(false);
 }
 
 void synchronize_rcu_bh(void);
@@ -108,6 +108,7 @@ void rcu_scheduler_starting(void);
 extern int rcu_scheduler_active __read_mostly;
 
 bool rcu_is_watching(void);
+void rcu_request_urgent_qs_task(struct task_struct *t);
 
 void rcu_all_qs(void);
 

include/linux/slab.h

@@ -28,7 +28,7 @@
 #define SLAB_STORE_USER		0x00010000UL	/* DEBUG: Store the last owner for bug hunting */
 #define SLAB_PANIC		0x00040000UL	/* Panic if kmem_cache_create() fails */
 /*
- * SLAB_DESTROY_BY_RCU - **WARNING** READ THIS!
+ * SLAB_TYPESAFE_BY_RCU - **WARNING** READ THIS!
  *
  * This delays freeing the SLAB page by a grace period, it does _NOT_
  * delay object freeing. This means that if you do kmem_cache_free()
@@ -61,8 +61,10 @@
  *
  * rcu_read_lock before reading the address, then rcu_read_unlock after
  * taking the spinlock within the structure expected at that address.
+ *
+ * Note that SLAB_TYPESAFE_BY_RCU was originally named SLAB_DESTROY_BY_RCU.
  */
-#define SLAB_DESTROY_BY_RCU	0x00080000UL	/* Defer freeing slabs to RCU */
+#define SLAB_TYPESAFE_BY_RCU	0x00080000UL	/* Defer freeing slabs to RCU */
 #define SLAB_MEM_SPREAD		0x00100000UL	/* Spread some memory over cpuset */
 #define SLAB_TRACE		0x00200000UL	/* Trace allocations and frees */
 

include/linux/srcu.h

@@ -22,7 +22,7 @@
  *	   Lai Jiangshan <laijs@cn.fujitsu.com>
  *
  * For detailed explanation of Read-Copy Update mechanism see -
- * 		Documentation/RCU/ *.txt
+ *		Documentation/RCU/ *.txt
  *
  */
 
@@ -32,35 +32,9 @@
 #include <linux/mutex.h>
 #include <linux/rcupdate.h>
 #include <linux/workqueue.h>
+#include <linux/rcu_segcblist.h>
 
-struct srcu_array {
-	unsigned long lock_count[2];
-	unsigned long unlock_count[2];
-};
-
-struct rcu_batch {
-	struct rcu_head *head, **tail;
-};
-
-#define RCU_BATCH_INIT(name) { NULL, &(name.head) }
-
-struct srcu_struct {
-	unsigned long completed;
-	struct srcu_array __percpu *per_cpu_ref;
-	spinlock_t queue_lock; /* protect ->batch_queue, ->running */
-	bool running;
-	/* callbacks just queued */
-	struct rcu_batch batch_queue;
-	/* callbacks try to do the first check_zero */
-	struct rcu_batch batch_check0;
-	/* callbacks done with the first check_zero and the flip */
-	struct rcu_batch batch_check1;
-	struct rcu_batch batch_done;
-	struct delayed_work work;
-#ifdef CONFIG_DEBUG_LOCK_ALLOC
-	struct lockdep_map dep_map;
-#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
-};
+struct srcu_struct;
 
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 
@@ -82,46 +56,15 @@ int init_srcu_struct(struct srcu_struct *sp);
 #define __SRCU_DEP_MAP_INIT(srcu_name)
 #endif /* #else #ifdef CONFIG_DEBUG_LOCK_ALLOC */
 
-void process_srcu(struct work_struct *work);
-
-#define __SRCU_STRUCT_INIT(name)					\
-	{								\
-		.completed = -300,					\
-		.per_cpu_ref = &name##_srcu_array,			\
-		.queue_lock = __SPIN_LOCK_UNLOCKED(name.queue_lock),	\
-		.running = false,					\
-		.batch_queue = RCU_BATCH_INIT(name.batch_queue),	\
-		.batch_check0 = RCU_BATCH_INIT(name.batch_check0),	\
-		.batch_check1 = RCU_BATCH_INIT(name.batch_check1),	\
-		.batch_done = RCU_BATCH_INIT(name.batch_done),		\
-		.work = __DELAYED_WORK_INITIALIZER(name.work, process_srcu, 0),\
-		__SRCU_DEP_MAP_INIT(name)				\
-	}
-
-/*
- * 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_array, name##_srcu_array);\
-	is_static struct srcu_struct name = __SRCU_STRUCT_INIT(name)
-#define DEFINE_SRCU(name)		__DEFINE_SRCU(name, /* not static */)
-#define DEFINE_STATIC_SRCU(name)	__DEFINE_SRCU(name, static)
+#ifdef CONFIG_TINY_SRCU
+#include <linux/srcutiny.h>
+#elif defined(CONFIG_TREE_SRCU)
+#include <linux/srcutree.h>
+#elif defined(CONFIG_CLASSIC_SRCU)
+#include <linux/srcuclassic.h>
+#else
+#error "Unknown SRCU implementation specified to kernel configuration"
+#endif
 
 /**
  * call_srcu() - Queue a callback for invocation after an SRCU grace period
@@ -147,9 +90,6 @@ void cleanup_srcu_struct(struct srcu_struct *sp);
 int __srcu_read_lock(struct srcu_struct *sp) __acquires(sp);
 void __srcu_read_unlock(struct srcu_struct *sp, int idx) __releases(sp);
 void synchronize_srcu(struct srcu_struct *sp);
-void synchronize_srcu_expedited(struct srcu_struct *sp);
-unsigned long srcu_batches_completed(struct srcu_struct *sp);
-void srcu_barrier(struct srcu_struct *sp);
 
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 

include/linux/srcuclassic.h

@@ -0,0 +1,101 @@
+/*
+ * Sleepable Read-Copy Update mechanism for mutual exclusion,
+ *	classic v4.11 variant.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, you can access it online at
+ * http://www.gnu.org/licenses/gpl-2.0.html.
+ *
+ * Copyright (C) IBM Corporation, 2017
+ *
+ * Author: Paul McKenney <paulmck@us.ibm.com>
+ */
+
+#ifndef _LINUX_SRCU_CLASSIC_H
+#define _LINUX_SRCU_CLASSIC_H
+
+struct srcu_array {
+	unsigned long lock_count[2];
+	unsigned long unlock_count[2];
+};
+
+struct rcu_batch {
+	struct rcu_head *head, **tail;
+};
+
+#define RCU_BATCH_INIT(name) { NULL, &(name.head) }
+
+struct srcu_struct {
+	unsigned long completed;
+	struct srcu_array __percpu *per_cpu_ref;
+	spinlock_t queue_lock; /* protect ->batch_queue, ->running */
+	bool running;
+	/* callbacks just queued */
+	struct rcu_batch batch_queue;
+	/* callbacks try to do the first check_zero */
+	struct rcu_batch batch_check0;
+	/* callbacks done with the first check_zero and the flip */
+	struct rcu_batch batch_check1;
+	struct rcu_batch batch_done;
+	struct delayed_work work;
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+	struct lockdep_map dep_map;
+#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
+};
+
+void process_srcu(struct work_struct *work);
+
+#define __SRCU_STRUCT_INIT(name)					\
+	{								\
+		.completed = -300,					\
+		.per_cpu_ref = &name##_srcu_array,			\
+		.queue_lock = __SPIN_LOCK_UNLOCKED(name.queue_lock),	\
+		.running = false,					\
+		.batch_queue = RCU_BATCH_INIT(name.batch_queue),	\
+		.batch_check0 = RCU_BATCH_INIT(name.batch_check0),	\
+		.batch_check1 = RCU_BATCH_INIT(name.batch_check1),	\
+		.batch_done = RCU_BATCH_INIT(name.batch_done),		\
+		.work = __DELAYED_WORK_INITIALIZER(name.work, process_srcu, 0),\
+		__SRCU_DEP_MAP_INIT(name)				\
+	}
+
+/*
+ * 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_array, name##_srcu_array);\
+	is_static struct srcu_struct name = __SRCU_STRUCT_INIT(name)
+#define DEFINE_SRCU(name)		__DEFINE_SRCU(name, /* not static */)
+#define DEFINE_STATIC_SRCU(name)	__DEFINE_SRCU(name, static)
+
+void synchronize_srcu_expedited(struct srcu_struct *sp);
+void srcu_barrier(struct srcu_struct *sp);
+unsigned long srcu_batches_completed(struct srcu_struct *sp);
+
+#endif

include/linux/srcutiny.h

@@ -0,0 +1,81 @@
+/*
+ * Sleepable Read-Copy Update mechanism for mutual exclusion,
+ *	tiny variant.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, you can access it online at
+ * http://www.gnu.org/licenses/gpl-2.0.html.
+ *
+ * Copyright (C) IBM Corporation, 2017
+ *
+ * Author: Paul McKenney <paulmck@us.ibm.com>
+ */
+
+#ifndef _LINUX_SRCU_TINY_H
+#define _LINUX_SRCU_TINY_H
+
+#include <linux/swait.h>
+
+struct srcu_struct {
+	int srcu_lock_nesting[2];	/* srcu_read_lock() nesting depth. */
+	struct swait_queue_head srcu_wq;
+					/* Last srcu_read_unlock() wakes GP. */
+	unsigned long srcu_gp_seq;	/* GP seq # for callback tagging. */
+	struct rcu_segcblist srcu_cblist;
+					/* Pending SRCU callbacks. */
+	int srcu_idx;			/* Current reader array element. */
+	bool srcu_gp_running;		/* GP workqueue running? */
+	bool srcu_gp_waiting;		/* GP waiting for readers? */
+	struct work_struct srcu_work;	/* For driving grace periods. */
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+	struct lockdep_map dep_map;
+#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
+};
+
+void srcu_drive_gp(struct work_struct *wp);
+
+#define __SRCU_STRUCT_INIT(name)					\
+{									\
+	.srcu_wq = __SWAIT_QUEUE_HEAD_INITIALIZER(name.srcu_wq),	\
+	.srcu_cblist = RCU_SEGCBLIST_INITIALIZER(name.srcu_cblist),	\
+	.srcu_work = __WORK_INITIALIZER(name.srcu_work, srcu_drive_gp),	\
+	__SRCU_DEP_MAP_INIT(name)					\
+}
+
+/*
+ * This odd _STATIC_ arrangement is needed for API compatibility with
+ * Tree SRCU, which needs some per-CPU data.
+ */
+#define DEFINE_SRCU(name) \
+	struct srcu_struct name = __SRCU_STRUCT_INIT(name)
+#define DEFINE_STATIC_SRCU(name) \
+	static struct srcu_struct name = __SRCU_STRUCT_INIT(name)
+
+void synchronize_srcu(struct srcu_struct *sp);
+
+static inline void synchronize_srcu_expedited(struct srcu_struct *sp)
+{
+	synchronize_srcu(sp);
+}
+
+static inline void srcu_barrier(struct srcu_struct *sp)
+{
+	synchronize_srcu(sp);
+}
+
+static inline unsigned long srcu_batches_completed(struct srcu_struct *sp)
+{
+	return 0;
+}
+
+#endif

include/linux/srcutree.h

@@ -0,0 +1,139 @@
+/*
+ * Sleepable Read-Copy Update mechanism for mutual exclusion,
+ *	tree variant.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, you can access it online at
+ * http://www.gnu.org/licenses/gpl-2.0.html.
+ *
+ * Copyright (C) IBM Corporation, 2017
+ *
+ * Author: Paul McKenney <paulmck@us.ibm.com>
+ */
+
+#ifndef _LINUX_SRCU_TREE_H
+#define _LINUX_SRCU_TREE_H
+
+#include <linux/rcu_node_tree.h>
+#include <linux/completion.h>
+
+struct srcu_node;
+struct srcu_struct;
+
+/*
+ * Per-CPU structure feeding into leaf srcu_node, similar in function
+ * to rcu_node.
+ */
+struct srcu_data {
+	/* Read-side state. */
+	unsigned long srcu_lock_count[2];	/* Locks per CPU. */
+	unsigned long srcu_unlock_count[2];	/* Unlocks per CPU. */
+
+	/* Update-side state. */
+	spinlock_t lock ____cacheline_internodealigned_in_smp;
+	struct rcu_segcblist srcu_cblist;	/* List of callbacks.*/
+	unsigned long srcu_gp_seq_needed;	/* Furthest future GP needed. */
+	bool srcu_cblist_invoking;		/* Invoking these CBs? */
+	struct delayed_work work;		/* Context for CB invoking. */
+	struct rcu_head srcu_barrier_head;	/* For srcu_barrier() use. */
+	struct srcu_node *mynode;		/* Leaf srcu_node. */
+	int cpu;
+	struct srcu_struct *sp;
+};
+
+/*
+ * Node in SRCU combining tree, similar in function to rcu_data.
+ */
+struct srcu_node {
+	spinlock_t lock;
+	unsigned long srcu_have_cbs[4];		/* GP seq for children */
+						/*  having CBs, but only */
+						/*  is > ->srcu_gq_seq. */
+	struct srcu_node *srcu_parent;		/* Next up in tree. */
+	int grplo;				/* Least CPU for node. */
+	int grphi;				/* Biggest CPU for node. */
+};
+
+/*
+ * Per-SRCU-domain structure, similar in function to rcu_state.
+ */
+struct srcu_struct {
+	struct srcu_node node[NUM_RCU_NODES];	/* Combining tree. */
+	struct srcu_node *level[RCU_NUM_LVLS + 1];
+						/* First node at each level. */
+	struct mutex srcu_cb_mutex;		/* Serialize CB preparation. */
+	spinlock_t gp_lock;			/* protect ->srcu_cblist */
+	struct mutex srcu_gp_mutex;		/* Serialize GP work. */
+	unsigned int srcu_idx;			/* Current rdr array element. */
+	unsigned long srcu_gp_seq;		/* Grace-period seq #. */
+	unsigned long srcu_gp_seq_needed;	/* Latest gp_seq needed. */
+	atomic_t srcu_exp_cnt;			/* # ongoing expedited GPs. */
+	struct srcu_data __percpu *sda;		/* Per-CPU srcu_data array. */
+	unsigned long srcu_barrier_seq;		/* srcu_barrier seq #. */
+	struct mutex srcu_barrier_mutex;	/* Serialize barrier ops. */
+	struct completion srcu_barrier_completion;
+						/* Awaken barrier rq at end. */
+	atomic_t srcu_barrier_cpu_cnt;		/* # CPUs not yet posting a */
+						/*  callback for the barrier */
+						/*  operation. */
+	struct delayed_work work;
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+	struct lockdep_map dep_map;
+#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
+};
+
+/* Values for state variable (bottom bits of ->srcu_gp_seq). */
+#define SRCU_STATE_IDLE		0
+#define SRCU_STATE_SCAN1	1
+#define SRCU_STATE_SCAN2	2
+
+void process_srcu(struct work_struct *work);
+
+#define __SRCU_STRUCT_INIT(name)					\
+	{								\
+		.sda = &name##_srcu_data,				\
+		.gp_lock = __SPIN_LOCK_UNLOCKED(name.gp_lock),		\
+		.srcu_gp_seq_needed = 0 - 1,				\
+		__SRCU_DEP_MAP_INIT(name)				\
+	}
+
+/*
+ * 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_data, name##_srcu_data);\
+	is_static struct srcu_struct name = __SRCU_STRUCT_INIT(name)
+#define DEFINE_SRCU(name)		__DEFINE_SRCU(name, /* not static */)
+#define DEFINE_STATIC_SRCU(name)	__DEFINE_SRCU(name, static)
+
+void synchronize_srcu_expedited(struct srcu_struct *sp);
+void srcu_barrier(struct srcu_struct *sp);
+unsigned long srcu_batches_completed(struct srcu_struct *sp);
+
+#endif

include/linux/types.h

@@ -209,7 +209,7 @@ struct ustat {
  * naturally due ABI requirements, but some architectures (like CRIS) have
  * weird ABI and we need to ask it explicitly.
  *
- * The alignment is required to guarantee that bits 0 and 1 of @next will be
+ * The alignment is required to guarantee that bit 0 of @next will be
  * clear under normal conditions -- as long as we use call_rcu(),
  * call_rcu_bh(), call_rcu_sched(), or call_srcu() to queue callback.
  *

include/net/sock.h

@@ -993,7 +993,7 @@ struct smc_hashinfo;
 struct module;
 
 /*
- * caches using SLAB_DESTROY_BY_RCU should let .next pointer from nulls nodes
+ * caches using SLAB_TYPESAFE_BY_RCU should let .next pointer from nulls nodes
  * un-modified. Special care is taken when initializing object to zero.
  */
 static inline void sk_prot_clear_nulls(struct sock *sk, int size)

init/Kconfig

@@ -526,6 +526,35 @@ config SRCU
 	  permits arbitrary sleeping or blocking within RCU read-side critical
 	  sections.
 
+config CLASSIC_SRCU
+	bool "Use v4.11 classic SRCU implementation"
+	default n
+	depends on RCU_EXPERT && SRCU
+	help
+	  This option selects the traditional well-tested classic SRCU
+	  implementation from v4.11, as might be desired for enterprise
+	  Linux distributions.  Without this option, the shiny new
+	  Tiny SRCU and Tree SRCU implementations are used instead.
+	  At some point, it is hoped that Tiny SRCU and Tree SRCU
+	  will accumulate enough test time and confidence to allow
+	  Classic SRCU to be dropped entirely.
+
+	  Say Y if you need a rock-solid SRCU.
+
+	  Say N if you would like help test Tree SRCU.
+
+config TINY_SRCU
+	bool
+	default y if TINY_RCU && !CLASSIC_SRCU
+	help
+	  This option selects the single-CPU non-preemptible version of SRCU.
+
+config TREE_SRCU
+	bool
+	default y if !TINY_RCU && !CLASSIC_SRCU
+	help
+	  This option selects the full-fledged version of SRCU.
+
 config TASKS_RCU
 	bool
 	default n
@@ -612,11 +641,17 @@ config RCU_FANOUT_LEAF
 	  initialization.  These systems tend to run CPU-bound, and thus
 	  are not helped by synchronized interrupts, and thus tend to
 	  skew them, which reduces lock contention enough that large
-	  leaf-level fanouts work well.
+	  leaf-level fanouts work well.  That said, setting leaf-level
+	  fanout to a large number will likely cause problematic
+	  lock contention on the leaf-level rcu_node structures unless
+	  you boot with the skew_tick kernel parameter.
 
 	  Select a specific number if testing RCU itself.
 
-	  Select the maximum permissible value for large systems.
+	  Select the maximum permissible value for large systems, but
+	  please understand that you may also need to set the skew_tick
+	  kernel boot parameter to avoid contention on the rcu_node
+	  structure's locks.
 
 	  Take the default if unsure.
 

kernel/fork.c

@@ -1313,7 +1313,7 @@ void __cleanup_sighand(struct sighand_struct *sighand)
 	if (atomic_dec_and_test(&sighand->count)) {
 		signalfd_cleanup(sighand);
 		/*
-		 * sighand_cachep is SLAB_DESTROY_BY_RCU so we can free it
+		 * sighand_cachep is SLAB_TYPESAFE_BY_RCU so we can free it
 		 * without an RCU grace period, see __lock_task_sighand().
 		 */
 		kmem_cache_free(sighand_cachep, sighand);
@@ -2144,7 +2144,7 @@ void __init proc_caches_init(void)
 {
 	sighand_cachep = kmem_cache_create("sighand_cache",
 			sizeof(struct sighand_struct), 0,
-			SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_DESTROY_BY_RCU|
+			SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_TYPESAFE_BY_RCU|
 			SLAB_NOTRACK|SLAB_ACCOUNT, sighand_ctor);
 	signal_cachep = kmem_cache_create("signal_cache",
 			sizeof(struct signal_struct), 0,

kernel/locking/lockdep.c

@@ -1144,10 +1144,10 @@ print_circular_bug_header(struct lock_list *entry, unsigned int depth,
 		return 0;
 
 	printk("\n");
-	printk("======================================================\n");
-	printk("[ INFO: possible circular locking dependency detected ]\n");
+	pr_warn("======================================================\n");
+	pr_warn("WARNING: possible circular locking dependency detected\n");
 	print_kernel_ident();
-	printk("-------------------------------------------------------\n");
+	pr_warn("------------------------------------------------------\n");
 	printk("%s/%d is trying to acquire lock:\n",
 		curr->comm, task_pid_nr(curr));
 	print_lock(check_src);
@@ -1482,11 +1482,11 @@ print_bad_irq_dependency(struct task_struct *curr,
 		return 0;
 
 	printk("\n");
-	printk("======================================================\n");
-	printk("[ INFO: %s-safe -> %s-unsafe lock order detected ]\n",
+	pr_warn("=====================================================\n");
+	pr_warn("WARNING: %s-safe -> %s-unsafe lock order detected\n",
 		irqclass, irqclass);
 	print_kernel_ident();
-	printk("------------------------------------------------------\n");
+	pr_warn("-----------------------------------------------------\n");
 	printk("%s/%d [HC%u[%lu]:SC%u[%lu]:HE%u:SE%u] is trying to acquire:\n",
 		curr->comm, task_pid_nr(curr),
 		curr->hardirq_context, hardirq_count() >> HARDIRQ_SHIFT,
@@ -1711,10 +1711,10 @@ print_deadlock_bug(struct task_struct *curr, struct held_lock *prev,
 		return 0;
 
 	printk("\n");
-	printk("=============================================\n");
-	printk("[ INFO: possible recursive locking detected ]\n");
+	pr_warn("============================================\n");
+	pr_warn("WARNING: possible recursive locking detected\n");
 	print_kernel_ident();
-	printk("---------------------------------------------\n");
+	pr_warn("--------------------------------------------\n");
 	printk("%s/%d is trying to acquire lock:\n",
 		curr->comm, task_pid_nr(curr));
 	print_lock(next);
@@ -2061,10 +2061,10 @@ static void print_collision(struct task_struct *curr,
 			struct lock_chain *chain)
 {
 	printk("\n");
-	printk("======================\n");
-	printk("[chain_key collision ]\n");
+	pr_warn("============================\n");
+	pr_warn("WARNING: chain_key collision\n");
 	print_kernel_ident();
-	printk("----------------------\n");
+	pr_warn("----------------------------\n");
 	printk("%s/%d: ", current->comm, task_pid_nr(current));
 	printk("Hash chain already cached but the contents don't match!\n");
 
@@ -2360,10 +2360,10 @@ print_usage_bug(struct task_struct *curr, struct held_lock *this,
 		return 0;
 
 	printk("\n");
-	printk("=================================\n");
-	printk("[ INFO: inconsistent lock state ]\n");
+	pr_warn("================================\n");
+	pr_warn("WARNING: inconsistent lock state\n");
 	print_kernel_ident();
-	printk("---------------------------------\n");
+	pr_warn("--------------------------------\n");
 
 	printk("inconsistent {%s} -> {%s} usage.\n",
 		usage_str[prev_bit], usage_str[new_bit]);
@@ -2425,10 +2425,10 @@ print_irq_inversion_bug(struct task_struct *curr,
 		return 0;
 
 	printk("\n");
-	printk("=========================================================\n");
-	printk("[ INFO: possible irq lock inversion dependency detected ]\n");
+	pr_warn("========================================================\n");
+	pr_warn("WARNING: possible irq lock inversion dependency detected\n");
 	print_kernel_ident();
-	printk("---------------------------------------------------------\n");
+	pr_warn("--------------------------------------------------------\n");
 	printk("%s/%d just changed the state of lock:\n",
 		curr->comm, task_pid_nr(curr));
 	print_lock(this);
@@ -3170,10 +3170,10 @@ print_lock_nested_lock_not_held(struct task_struct *curr,
 		return 0;
 
 	printk("\n");
-	printk("==================================\n");
-	printk("[ BUG: Nested lock was not taken ]\n");
+	pr_warn("==================================\n");
+	pr_warn("WARNING: Nested lock was not taken\n");
 	print_kernel_ident();
-	printk("----------------------------------\n");
+	pr_warn("----------------------------------\n");
 
 	printk("%s/%d is trying to lock:\n", curr->comm, task_pid_nr(curr));
 	print_lock(hlock);
@@ -3383,10 +3383,10 @@ print_unlock_imbalance_bug(struct task_struct *curr, struct lockdep_map *lock,
 		return 0;
 
 	printk("\n");
-	printk("=====================================\n");
-	printk("[ BUG: bad unlock balance detected! ]\n");
+	pr_warn("=====================================\n");
+	pr_warn("WARNING: bad unlock balance detected!\n");
 	print_kernel_ident();
-	printk("-------------------------------------\n");
+	pr_warn("-------------------------------------\n");
 	printk("%s/%d is trying to release lock (",
 		curr->comm, task_pid_nr(curr));
 	print_lockdep_cache(lock);
@@ -3880,10 +3880,10 @@ print_lock_contention_bug(struct task_struct *curr, struct lockdep_map *lock,
 		return 0;
 
 	printk("\n");
-	printk("=================================\n");
-	printk("[ BUG: bad contention detected! ]\n");
+	pr_warn("=================================\n");
+	pr_warn("WARNING: bad contention detected!\n");
 	print_kernel_ident();
-	printk("---------------------------------\n");
+	pr_warn("---------------------------------\n");
 	printk("%s/%d is trying to contend lock (",
 		curr->comm, task_pid_nr(curr));
 	print_lockdep_cache(lock);
@@ -4244,10 +4244,10 @@ print_freed_lock_bug(struct task_struct *curr, const void *mem_from,
 		return;
 
 	printk("\n");
-	printk("=========================\n");
-	printk("[ BUG: held lock freed! ]\n");
+	pr_warn("=========================\n");
+	pr_warn("WARNING: held lock freed!\n");
 	print_kernel_ident();
-	printk("-------------------------\n");
+	pr_warn("-------------------------\n");
 	printk("%s/%d is freeing memory %p-%p, with a lock still held there!\n",
 		curr->comm, task_pid_nr(curr), mem_from, mem_to-1);
 	print_lock(hlock);
@@ -4302,11 +4302,11 @@ static void print_held_locks_bug(void)
 		return;
 
 	printk("\n");
-	printk("=====================================\n");
-	printk("[ BUG: %s/%d still has locks held! ]\n",
+	pr_warn("====================================\n");
+	pr_warn("WARNING: %s/%d still has locks held!\n",
 	       current->comm, task_pid_nr(current));
 	print_kernel_ident();
-	printk("-------------------------------------\n");
+	pr_warn("------------------------------------\n");
 	lockdep_print_held_locks(current);
 	printk("\nstack backtrace:\n");
 	dump_stack();
@@ -4371,7 +4371,7 @@ retry:
 	} while_each_thread(g, p);
 
 	printk("\n");
-	printk("=============================================\n\n");
+	pr_warn("=============================================\n\n");
 
 	if (unlock)
 		read_unlock(&tasklist_lock);
@@ -4401,10 +4401,10 @@ asmlinkage __visible void lockdep_sys_exit(void)
 		if (!debug_locks_off())
 			return;
 		printk("\n");
-		printk("================================================\n");
-		printk("[ BUG: lock held when returning to user space! ]\n");
+		pr_warn("================================================\n");
+		pr_warn("WARNING: lock held when returning to user space!\n");
 		print_kernel_ident();
-		printk("------------------------------------------------\n");
+		pr_warn("------------------------------------------------\n");
 		printk("%s/%d is leaving the kernel with locks still held!\n",
 				curr->comm, curr->pid);
 		lockdep_print_held_locks(curr);
@@ -4421,13 +4421,13 @@ void lockdep_rcu_suspicious(const char *file, const int line, const char *s)
 #endif /* #ifdef CONFIG_PROVE_RCU_REPEATEDLY */
 	/* Note: the following can be executed concurrently, so be careful. */
 	printk("\n");
-	pr_err("===============================\n");
-	pr_err("[ ERR: suspicious RCU usage.  ]\n");
+	pr_warn("=============================\n");
+	pr_warn("WARNING: suspicious RCU usage\n");
 	print_kernel_ident();
-	pr_err("-------------------------------\n");
-	pr_err("%s:%d %s!\n", file, line, s);
-	pr_err("\nother info that might help us debug this:\n\n");
-	pr_err("\n%srcu_scheduler_active = %d, debug_locks = %d\n",
+	pr_warn("-----------------------------\n");
+	printk("%s:%d %s!\n", file, line, s);
+	printk("\nother info that might help us debug this:\n\n");
+	printk("\n%srcu_scheduler_active = %d, debug_locks = %d\n",
 	       !rcu_lockdep_current_cpu_online()
 			? "RCU used illegally from offline CPU!\n"
 			: !rcu_is_watching()

kernel/locking/rtmutex-debug.c

@@ -102,10 +102,11 @@ void debug_rt_mutex_print_deadlock(struct rt_mutex_waiter *waiter)
 		return;
 	}
 
-	printk("\n============================================\n");
-	printk(  "[ BUG: circular locking deadlock detected! ]\n");
-	printk("%s\n", print_tainted());
-	printk(  "--------------------------------------------\n");
+	pr_warn("\n");
+	pr_warn("============================================\n");
+	pr_warn("WARNING: circular locking deadlock detected!\n");
+	pr_warn("%s\n", print_tainted());
+	pr_warn("--------------------------------------------\n");
 	printk("%s/%d is deadlocking current task %s/%d\n\n",
 	       task->comm, task_pid_nr(task),
 	       current->comm, task_pid_nr(current));

kernel/rcu/Makefile

@@ -3,7 +3,9 @@
 KCOV_INSTRUMENT := n
 
 obj-y += update.o sync.o
-obj-$(CONFIG_SRCU) += srcu.o
+obj-$(CONFIG_CLASSIC_SRCU) += srcu.o
+obj-$(CONFIG_TREE_SRCU) += srcutree.o
+obj-$(CONFIG_TINY_SRCU) += srcutiny.o
 obj-$(CONFIG_RCU_TORTURE_TEST) += rcutorture.o
 obj-$(CONFIG_RCU_PERF_TEST) += rcuperf.o
 obj-$(CONFIG_TREE_RCU) += tree.o

kernel/rcu/rcu.h

@@ -56,6 +56,83 @@
 #define DYNTICK_TASK_EXIT_IDLE	   (DYNTICK_TASK_NEST_VALUE + \
 				    DYNTICK_TASK_FLAG)
 
+
+/*
+ * Grace-period counter management.
+ */
+
+#define RCU_SEQ_CTR_SHIFT	2
+#define RCU_SEQ_STATE_MASK	((1 << RCU_SEQ_CTR_SHIFT) - 1)
+
+/*
+ * Return the counter portion of a sequence number previously returned
+ * by rcu_seq_snap() or rcu_seq_current().
+ */
+static inline unsigned long rcu_seq_ctr(unsigned long s)
+{
+	return s >> RCU_SEQ_CTR_SHIFT;
+}
+
+/*
+ * Return the state portion of a sequence number previously returned
+ * by rcu_seq_snap() or rcu_seq_current().
+ */
+static inline int rcu_seq_state(unsigned long s)
+{
+	return s & RCU_SEQ_STATE_MASK;
+}
+
+/*
+ * Set the state portion of the pointed-to sequence number.
+ * The caller is responsible for preventing conflicting updates.
+ */
+static inline void rcu_seq_set_state(unsigned long *sp, int newstate)
+{
+	WARN_ON_ONCE(newstate & ~RCU_SEQ_STATE_MASK);
+	WRITE_ONCE(*sp, (*sp & ~RCU_SEQ_STATE_MASK) + newstate);
+}
+
+/* Adjust sequence number for start of update-side operation. */
+static inline void rcu_seq_start(unsigned long *sp)
+{
+	WRITE_ONCE(*sp, *sp + 1);
+	smp_mb(); /* Ensure update-side operation after counter increment. */
+	WARN_ON_ONCE(rcu_seq_state(*sp) != 1);
+}
+
+/* Adjust sequence number for end of update-side operation. */
+static inline void rcu_seq_end(unsigned long *sp)
+{
+	smp_mb(); /* Ensure update-side operation before counter increment. */
+	WARN_ON_ONCE(!rcu_seq_state(*sp));
+	WRITE_ONCE(*sp, (*sp | RCU_SEQ_STATE_MASK) + 1);
+}
+
+/* Take a snapshot of the update side's sequence number. */
+static inline unsigned long rcu_seq_snap(unsigned long *sp)
+{
+	unsigned long s;
+
+	s = (READ_ONCE(*sp) + 2 * RCU_SEQ_STATE_MASK + 1) & ~RCU_SEQ_STATE_MASK;
+	smp_mb(); /* Above access must not bleed into critical section. */
+	return s;
+}
+
+/* Return the current value the update side's sequence number, no ordering. */
+static inline unsigned long rcu_seq_current(unsigned long *sp)
+{
+	return READ_ONCE(*sp);
+}
+
+/*
+ * Given a snapshot from rcu_seq_snap(), determine whether or not a
+ * full update-side operation has occurred.
+ */
+static inline bool rcu_seq_done(unsigned long *sp, unsigned long s)
+{
+	return ULONG_CMP_GE(READ_ONCE(*sp), s);
+}
+
 /*
  * debug_rcu_head_queue()/debug_rcu_head_unqueue() are used internally
  * by call_rcu() and rcu callback execution, and are therefore not part of the
@@ -109,12 +186,12 @@ static inline bool __rcu_reclaim(const char *rn, struct rcu_head *head)
 
 	rcu_lock_acquire(&rcu_callback_map);
 	if (__is_kfree_rcu_offset(offset)) {
-		RCU_TRACE(trace_rcu_invoke_kfree_callback(rn, head, offset));
+		RCU_TRACE(trace_rcu_invoke_kfree_callback(rn, head, offset);)
 		kfree((void *)head - offset);
 		rcu_lock_release(&rcu_callback_map);
 		return true;
 	} else {
-		RCU_TRACE(trace_rcu_invoke_callback(rn, head));
+		RCU_TRACE(trace_rcu_invoke_callback(rn, head);)
 		head->func(head);
 		rcu_lock_release(&rcu_callback_map);
 		return false;
@@ -144,4 +221,76 @@ void rcu_test_sync_prims(void);
  */
 extern void resched_cpu(int cpu);
 
+#if defined(SRCU) || !defined(TINY_RCU)
+
+#include <linux/rcu_node_tree.h>
+
+extern int rcu_num_lvls;
+extern int num_rcu_lvl[];
+extern int rcu_num_nodes;
+static bool rcu_fanout_exact;
+static int rcu_fanout_leaf;
+
+/*
+ * Compute the per-level fanout, either using the exact fanout specified
+ * or balancing the tree, depending on the rcu_fanout_exact boot parameter.
+ */
+static inline void rcu_init_levelspread(int *levelspread, const int *levelcnt)
+{
+	int i;
+
+	if (rcu_fanout_exact) {
+		levelspread[rcu_num_lvls - 1] = rcu_fanout_leaf;
+		for (i = rcu_num_lvls - 2; i >= 0; i--)
+			levelspread[i] = RCU_FANOUT;
+	} else {
+		int ccur;
+		int cprv;
+
+		cprv = nr_cpu_ids;
+		for (i = rcu_num_lvls - 1; i >= 0; i--) {
+			ccur = levelcnt[i];
+			levelspread[i] = (cprv + ccur - 1) / ccur;
+			cprv = ccur;
+		}
+	}
+}
+
+/*
+ * Do a full breadth-first scan of the rcu_node structures for the
+ * specified rcu_state structure.
+ */
+#define rcu_for_each_node_breadth_first(rsp, rnp) \
+	for ((rnp) = &(rsp)->node[0]; \
+	     (rnp) < &(rsp)->node[rcu_num_nodes]; (rnp)++)
+
+/*
+ * Do a breadth-first scan of the non-leaf rcu_node structures for the
+ * specified rcu_state structure.  Note that if there is a singleton
+ * rcu_node tree with but one rcu_node structure, this loop is a no-op.
+ */
+#define rcu_for_each_nonleaf_node_breadth_first(rsp, rnp) \
+	for ((rnp) = &(rsp)->node[0]; \
+	     (rnp) < (rsp)->level[rcu_num_lvls - 1]; (rnp)++)
+
+/*
+ * Scan the leaves of the rcu_node hierarchy for the specified rcu_state
+ * structure.  Note that if there is a singleton rcu_node tree with but
+ * one rcu_node structure, this loop -will- visit the rcu_node structure.
+ * It is still a leaf node, even if it is also the root node.
+ */
+#define rcu_for_each_leaf_node(rsp, rnp) \
+	for ((rnp) = (rsp)->level[rcu_num_lvls - 1]; \
+	     (rnp) < &(rsp)->node[rcu_num_nodes]; (rnp)++)
+
+/*
+ * Iterate over all possible CPUs in a leaf RCU node.
+ */
+#define for_each_leaf_node_possible_cpu(rnp, cpu) \
+	for ((cpu) = cpumask_next(rnp->grplo - 1, cpu_possible_mask); \
+	     cpu <= rnp->grphi; \
+	     cpu = cpumask_next((cpu), cpu_possible_mask))
+
+#endif /* #if defined(SRCU) || !defined(TINY_RCU) */
+
 #endif /* __LINUX_RCU_H */

kernel/rcu/rcutorture.c

@@ -559,19 +559,34 @@ static void srcu_torture_barrier(void)
 
 static void srcu_torture_stats(void)
 {
-	int cpu;
-	int idx = srcu_ctlp->completed & 0x1;
+	int __maybe_unused cpu;
+	int idx;
 
-	pr_alert("%s%s per-CPU(idx=%d):",
+#if defined(CONFIG_TREE_SRCU) || defined(CONFIG_CLASSIC_SRCU)
+#ifdef CONFIG_TREE_SRCU
+	idx = srcu_ctlp->srcu_idx & 0x1;
+#else /* #ifdef CONFIG_TREE_SRCU */
+	idx = srcu_ctlp->completed & 0x1;
+#endif /* #else #ifdef CONFIG_TREE_SRCU */
+	pr_alert("%s%s Tree SRCU per-CPU(idx=%d):",
 		 torture_type, TORTURE_FLAG, idx);
 	for_each_possible_cpu(cpu) {
 		unsigned long l0, l1;
 		unsigned long u0, u1;
 		long c0, c1;
-		struct srcu_array *counts = per_cpu_ptr(srcu_ctlp->per_cpu_ref, cpu);
+#ifdef CONFIG_TREE_SRCU
+		struct srcu_data *counts;
 
+		counts = per_cpu_ptr(srcu_ctlp->sda, cpu);
+		u0 = counts->srcu_unlock_count[!idx];
+		u1 = counts->srcu_unlock_count[idx];
+#else /* #ifdef CONFIG_TREE_SRCU */
+		struct srcu_array *counts;
+
+		counts = per_cpu_ptr(srcu_ctlp->per_cpu_ref, cpu);
 		u0 = counts->unlock_count[!idx];
 		u1 = counts->unlock_count[idx];
+#endif /* #else #ifdef CONFIG_TREE_SRCU */
 
 		/*
 		 * Make sure that a lock is always counted if the corresponding
@@ -579,14 +594,26 @@ static void srcu_torture_stats(void)
 		 */
 		smp_rmb();
 
+#ifdef CONFIG_TREE_SRCU
+		l0 = counts->srcu_lock_count[!idx];
+		l1 = counts->srcu_lock_count[idx];
+#else /* #ifdef CONFIG_TREE_SRCU */
 		l0 = counts->lock_count[!idx];
 		l1 = counts->lock_count[idx];
+#endif /* #else #ifdef CONFIG_TREE_SRCU */
 
 		c0 = l0 - u0;
 		c1 = l1 - u1;
 		pr_cont(" %d(%ld,%ld)", cpu, c0, c1);
 	}
 	pr_cont("\n");
+#elif defined(CONFIG_TINY_SRCU)
+	idx = READ_ONCE(srcu_ctlp->srcu_idx) & 0x1;
+	pr_alert("%s%s Tiny SRCU per-CPU(idx=%d): (%d,%d)\n",
+		 torture_type, TORTURE_FLAG, idx,
+		 READ_ONCE(srcu_ctlp->srcu_lock_nesting[!idx]),
+		 READ_ONCE(srcu_ctlp->srcu_lock_nesting[idx]));
+#endif
 }
 
 static void srcu_torture_synchronize_expedited(void)

kernel/rcu/srcu.c

@@ -22,7 +22,7 @@
  *	   Lai Jiangshan <laijs@cn.fujitsu.com>
  *
  * For detailed explanation of Read-Copy Update mechanism see -
- * 		Documentation/RCU/ *.txt
+ *		Documentation/RCU/ *.txt
  *
  */
 
@@ -243,8 +243,14 @@ static bool srcu_readers_active(struct srcu_struct *sp)
  * cleanup_srcu_struct - deconstruct a sleep-RCU structure
  * @sp: structure to clean up.
  *
- * Must invoke this after you are finished using a given srcu_struct that
- * was initialized via init_srcu_struct(), else you leak memory.
+ * Must invoke this only after you are finished using a given srcu_struct
+ * that was initialized via init_srcu_struct().  This code does some
+ * probabalistic checking, spotting late uses of srcu_read_lock(),
+ * synchronize_srcu(), synchronize_srcu_expedited(), and call_srcu().
+ * If any such late uses are detected, the per-CPU memory associated with
+ * the srcu_struct is simply leaked and WARN_ON() is invoked.  If the
+ * caller frees the srcu_struct itself, a use-after-free crash will likely
+ * ensue, but at least there will be a warning printed.
  */
 void cleanup_srcu_struct(struct srcu_struct *sp)
 {

kernel/rcu/srcutiny.c

@@ -0,0 +1,215 @@
+/*
+ * Sleepable Read-Copy Update mechanism for mutual exclusion,
+ *	tiny version for non-preemptible single-CPU use.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, you can access it online at
+ * http://www.gnu.org/licenses/gpl-2.0.html.
+ *
+ * Copyright (C) IBM Corporation, 2017
+ *
+ * Author: Paul McKenney <paulmck@us.ibm.com>
+ */
+
+#include <linux/export.h>
+#include <linux/mutex.h>
+#include <linux/preempt.h>
+#include <linux/rcupdate_wait.h>
+#include <linux/sched.h>
+#include <linux/delay.h>
+#include <linux/srcu.h>
+
+#include <linux/rcu_node_tree.h>
+#include "rcu.h"
+
+static int init_srcu_struct_fields(struct srcu_struct *sp)
+{
+	sp->srcu_lock_nesting[0] = 0;
+	sp->srcu_lock_nesting[1] = 0;
+	init_swait_queue_head(&sp->srcu_wq);
+	sp->srcu_gp_seq = 0;
+	rcu_segcblist_init(&sp->srcu_cblist);
+	sp->srcu_gp_running = false;
+	sp->srcu_gp_waiting = false;
+	sp->srcu_idx = 0;
+	INIT_WORK(&sp->srcu_work, srcu_drive_gp);
+	return 0;
+}
+
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+
+int __init_srcu_struct(struct srcu_struct *sp, const char *name,
+		       struct lock_class_key *key)
+{
+	/* Don't re-initialize a lock while it is held. */
+	debug_check_no_locks_freed((void *)sp, sizeof(*sp));
+	lockdep_init_map(&sp->dep_map, name, key, 0);
+	return init_srcu_struct_fields(sp);
+}
+EXPORT_SYMBOL_GPL(__init_srcu_struct);
+
+#else /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
+
+/*
+ * init_srcu_struct - initialize a sleep-RCU structure
+ * @sp: structure to initialize.
+ *
+ * Must invoke this on a given srcu_struct before passing that srcu_struct
+ * to any other function.  Each srcu_struct represents a separate domain
+ * of SRCU protection.
+ */
+int init_srcu_struct(struct srcu_struct *sp)
+{
+	return init_srcu_struct_fields(sp);
+}
+EXPORT_SYMBOL_GPL(init_srcu_struct);
+
+#endif /* #else #ifdef CONFIG_DEBUG_LOCK_ALLOC */
+
+/*
+ * cleanup_srcu_struct - deconstruct a sleep-RCU structure
+ * @sp: structure to clean up.
+ *
+ * Must invoke this after you are finished using a given srcu_struct that
+ * was initialized via init_srcu_struct(), else you leak memory.
+ */
+void cleanup_srcu_struct(struct srcu_struct *sp)
+{
+	WARN_ON(sp->srcu_lock_nesting[0] || sp->srcu_lock_nesting[1]);
+	flush_work(&sp->srcu_work);
+	WARN_ON(rcu_seq_state(sp->srcu_gp_seq));
+	WARN_ON(sp->srcu_gp_running);
+	WARN_ON(sp->srcu_gp_waiting);
+	WARN_ON(!rcu_segcblist_empty(&sp->srcu_cblist));
+}
+EXPORT_SYMBOL_GPL(cleanup_srcu_struct);
+
+/*
+ * Counts the new reader in the appropriate per-CPU element of the
+ * srcu_struct.  Must be called from process context.
+ * Returns an index that must be passed to the matching srcu_read_unlock().
+ */
+int __srcu_read_lock(struct srcu_struct *sp)
+{
+	int idx;
+
+	idx = READ_ONCE(sp->srcu_idx);
+	WRITE_ONCE(sp->srcu_lock_nesting[idx], sp->srcu_lock_nesting[idx] + 1);
+	return idx;
+}
+EXPORT_SYMBOL_GPL(__srcu_read_lock);
+
+/*
+ * Removes the count for the old reader from the appropriate element of
+ * the srcu_struct.  Must be called from process context.
+ */
+void __srcu_read_unlock(struct srcu_struct *sp, int idx)
+{
+	int newval = sp->srcu_lock_nesting[idx] - 1;
+
+	WRITE_ONCE(sp->srcu_lock_nesting[idx], newval);
+	if (!newval && READ_ONCE(sp->srcu_gp_waiting))
+		swake_up(&sp->srcu_wq);
+}
+EXPORT_SYMBOL_GPL(__srcu_read_unlock);
+
+/*
+ * Workqueue handler to drive one grace period and invoke any callbacks
+ * that become ready as a result.  Single-CPU and !PREEMPT operation
+ * means that we get away with murder on synchronization.  ;-)
+ */
+void srcu_drive_gp(struct work_struct *wp)
+{
+	int idx;
+	struct rcu_cblist ready_cbs;
+	struct srcu_struct *sp;
+	struct rcu_head *rhp;
+
+	sp = container_of(wp, struct srcu_struct, srcu_work);
+	if (sp->srcu_gp_running || rcu_segcblist_empty(&sp->srcu_cblist))
+		return; /* Already running or nothing to do. */
+
+	/* Tag recently arrived callbacks and wait for readers. */
+	WRITE_ONCE(sp->srcu_gp_running, true);
+	rcu_segcblist_accelerate(&sp->srcu_cblist,
+				 rcu_seq_snap(&sp->srcu_gp_seq));
+	rcu_seq_start(&sp->srcu_gp_seq);
+	idx = sp->srcu_idx;
+	WRITE_ONCE(sp->srcu_idx, !sp->srcu_idx);
+	WRITE_ONCE(sp->srcu_gp_waiting, true);  /* srcu_read_unlock() wakes! */
+	swait_event(sp->srcu_wq, !READ_ONCE(sp->srcu_lock_nesting[idx]));
+	WRITE_ONCE(sp->srcu_gp_waiting, false); /* srcu_read_unlock() cheap. */
+	rcu_seq_end(&sp->srcu_gp_seq);
+
+	/* Update callback list based on GP, and invoke ready callbacks. */
+	rcu_segcblist_advance(&sp->srcu_cblist,
+			      rcu_seq_current(&sp->srcu_gp_seq));
+	if (rcu_segcblist_ready_cbs(&sp->srcu_cblist)) {
+		rcu_cblist_init(&ready_cbs);
+		local_irq_disable();
+		rcu_segcblist_extract_done_cbs(&sp->srcu_cblist, &ready_cbs);
+		local_irq_enable();
+		rhp = rcu_cblist_dequeue(&ready_cbs);
+		for (; rhp != NULL; rhp = rcu_cblist_dequeue(&ready_cbs)) {
+			local_bh_disable();
+			rhp->func(rhp);
+			local_bh_enable();
+		}
+		local_irq_disable();
+		rcu_segcblist_insert_count(&sp->srcu_cblist, &ready_cbs);
+		local_irq_enable();
+	}
+	WRITE_ONCE(sp->srcu_gp_running, false);
+
+	/*
+	 * If more callbacks, reschedule ourselves.  This can race with
+	 * a call_srcu() at interrupt level, but the ->srcu_gp_running
+	 * checks will straighten that out.
+	 */
+	if (!rcu_segcblist_empty(&sp->srcu_cblist))
+		schedule_work(&sp->srcu_work);
+}
+EXPORT_SYMBOL_GPL(srcu_drive_gp);
+
+/*
+ * Enqueue an SRCU callback on the specified srcu_struct structure,
+ * initiating grace-period processing if it is not already running.
+ */
+void call_srcu(struct srcu_struct *sp, struct rcu_head *head,
+	       rcu_callback_t func)
+{
+	unsigned long flags;
+
+	head->func = func;
+	local_irq_save(flags);
+	rcu_segcblist_enqueue(&sp->srcu_cblist, head, false);
+	local_irq_restore(flags);
+	if (!READ_ONCE(sp->srcu_gp_running))
+		schedule_work(&sp->srcu_work);
+}
+EXPORT_SYMBOL_GPL(call_srcu);
+
+/*
+ * synchronize_srcu - wait for prior SRCU read-side critical-section completion
+ */
+void synchronize_srcu(struct srcu_struct *sp)
+{
+	struct rcu_synchronize rs;
+
+	init_rcu_head_on_stack(&rs.head);
+	init_completion(&rs.completion);
+	call_srcu(sp, &rs.head, wakeme_after_rcu);
+	wait_for_completion(&rs.completion);
+	destroy_rcu_head_on_stack(&rs.head);
+}
+EXPORT_SYMBOL_GPL(synchronize_srcu);

kernel/rcu/srcutree.c

@@ -0,0 +1,996 @@
+/*
+ * Sleepable Read-Copy Update mechanism for mutual exclusion.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, you can access it online at
+ * http://www.gnu.org/licenses/gpl-2.0.html.
+ *
+ * Copyright (C) IBM Corporation, 2006
+ * Copyright (C) Fujitsu, 2012
+ *
+ * Author: Paul McKenney <paulmck@us.ibm.com>
+ *	   Lai Jiangshan <laijs@cn.fujitsu.com>
+ *
+ * For detailed explanation of Read-Copy Update mechanism see -
+ *		Documentation/RCU/ *.txt
+ *
+ */
+
+#include <linux/export.h>
+#include <linux/mutex.h>
+#include <linux/percpu.h>
+#include <linux/preempt.h>
+#include <linux/rcupdate_wait.h>
+#include <linux/sched.h>
+#include <linux/smp.h>
+#include <linux/delay.h>
+#include <linux/srcu.h>
+
+#include "rcu.h"
+
+static void srcu_invoke_callbacks(struct work_struct *work);
+static void srcu_reschedule(struct srcu_struct *sp, unsigned long delay);
+
+/*
+ * Initialize SRCU combining tree.  Note that statically allocated
+ * srcu_struct structures might already have srcu_read_lock() and
+ * srcu_read_unlock() running against them.  So if the is_static parameter
+ * is set, don't initialize ->srcu_lock_count[] and ->srcu_unlock_count[].
+ */
+static void init_srcu_struct_nodes(struct srcu_struct *sp, bool is_static)
+{
+	int cpu;
+	int i;
+	int level = 0;
+	int levelspread[RCU_NUM_LVLS];
+	struct srcu_data *sdp;
+	struct srcu_node *snp;
+	struct srcu_node *snp_first;
+
+	/* Work out the overall tree geometry. */
+	sp->level[0] = &sp->node[0];
+	for (i = 1; i < rcu_num_lvls; i++)
+		sp->level[i] = sp->level[i - 1] + num_rcu_lvl[i - 1];
+	rcu_init_levelspread(levelspread, num_rcu_lvl);
+
+	/* Each pass through this loop initializes one srcu_node structure. */
+	rcu_for_each_node_breadth_first(sp, snp) {
+		spin_lock_init(&snp->lock);
+		for (i = 0; i < ARRAY_SIZE(snp->srcu_have_cbs); i++)
+			snp->srcu_have_cbs[i] = 0;
+		snp->grplo = -1;
+		snp->grphi = -1;
+		if (snp == &sp->node[0]) {
+			/* Root node, special case. */
+			snp->srcu_parent = NULL;
+			continue;
+		}
+
+		/* Non-root node. */
+		if (snp == sp->level[level + 1])
+			level++;
+		snp->srcu_parent = sp->level[level - 1] +
+				   (snp - sp->level[level]) /
+				   levelspread[level - 1];
+	}
+
+	/*
+	 * Initialize the per-CPU srcu_data array, which feeds into the
+	 * leaves of the srcu_node tree.
+	 */
+	WARN_ON_ONCE(ARRAY_SIZE(sdp->srcu_lock_count) !=
+		     ARRAY_SIZE(sdp->srcu_unlock_count));
+	level = rcu_num_lvls - 1;
+	snp_first = sp->level[level];
+	for_each_possible_cpu(cpu) {
+		sdp = per_cpu_ptr(sp->sda, cpu);
+		spin_lock_init(&sdp->lock);
+		rcu_segcblist_init(&sdp->srcu_cblist);
+		sdp->srcu_cblist_invoking = false;
+		sdp->srcu_gp_seq_needed = sp->srcu_gp_seq;
+		sdp->mynode = &snp_first[cpu / levelspread[level]];
+		for (snp = sdp->mynode; snp != NULL; snp = snp->srcu_parent) {
+			if (snp->grplo < 0)
+				snp->grplo = cpu;
+			snp->grphi = cpu;
+		}
+		sdp->cpu = cpu;
+		INIT_DELAYED_WORK(&sdp->work, srcu_invoke_callbacks);
+		sdp->sp = sp;
+		if (is_static)
+			continue;
+
+		/* Dynamically allocated, better be no srcu_read_locks()! */
+		for (i = 0; i < ARRAY_SIZE(sdp->srcu_lock_count); i++) {
+			sdp->srcu_lock_count[i] = 0;
+			sdp->srcu_unlock_count[i] = 0;
+		}
+	}
+}
+
+/*
+ * Initialize non-compile-time initialized fields, including the
+ * associated srcu_node and srcu_data structures.  The is_static
+ * parameter is passed through to init_srcu_struct_nodes(), and
+ * also tells us that ->sda has already been wired up to srcu_data.
+ */
+static int init_srcu_struct_fields(struct srcu_struct *sp, bool is_static)
+{
+	mutex_init(&sp->srcu_cb_mutex);
+	mutex_init(&sp->srcu_gp_mutex);
+	sp->srcu_idx = 0;
+	sp->srcu_gp_seq = 0;
+	atomic_set(&sp->srcu_exp_cnt, 0);
+	sp->srcu_barrier_seq = 0;
+	mutex_init(&sp->srcu_barrier_mutex);
+	atomic_set(&sp->srcu_barrier_cpu_cnt, 0);
+	INIT_DELAYED_WORK(&sp->work, process_srcu);
+	if (!is_static)
+		sp->sda = alloc_percpu(struct srcu_data);
+	init_srcu_struct_nodes(sp, is_static);
+	smp_store_release(&sp->srcu_gp_seq_needed, 0); /* Init done. */
+	return sp->sda ? 0 : -ENOMEM;
+}
+
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+
+int __init_srcu_struct(struct srcu_struct *sp, const char *name,
+		       struct lock_class_key *key)
+{
+	/* Don't re-initialize a lock while it is held. */
+	debug_check_no_locks_freed((void *)sp, sizeof(*sp));
+	lockdep_init_map(&sp->dep_map, name, key, 0);
+	spin_lock_init(&sp->gp_lock);
+	return init_srcu_struct_fields(sp, false);
+}
+EXPORT_SYMBOL_GPL(__init_srcu_struct);
+
+#else /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
+
+/**
+ * init_srcu_struct - initialize a sleep-RCU structure
+ * @sp: structure to initialize.
+ *
+ * Must invoke this on a given srcu_struct before passing that srcu_struct
+ * to any other function.  Each srcu_struct represents a separate domain
+ * of SRCU protection.
+ */
+int init_srcu_struct(struct srcu_struct *sp)
+{
+	spin_lock_init(&sp->gp_lock);
+	return init_srcu_struct_fields(sp, false);
+}
+EXPORT_SYMBOL_GPL(init_srcu_struct);
+
+#endif /* #else #ifdef CONFIG_DEBUG_LOCK_ALLOC */
+
+/*
+ * First-use initialization of statically allocated srcu_struct
+ * structure.  Wiring up the combining tree is more than can be
+ * done with compile-time initialization, so this check is added
+ * to each update-side SRCU primitive.  Use ->gp_lock, which -is-
+ * compile-time initialized, to resolve races involving multiple
+ * CPUs trying to garner first-use privileges.
+ */
+static void check_init_srcu_struct(struct srcu_struct *sp)
+{
+	unsigned long flags;
+
+	WARN_ON_ONCE(rcu_scheduler_active == RCU_SCHEDULER_INIT);
+	/* The smp_load_acquire() pairs with the smp_store_release(). */
+	if (!rcu_seq_state(smp_load_acquire(&sp->srcu_gp_seq_needed))) /*^^^*/
+		return; /* Already initialized. */
+	spin_lock_irqsave(&sp->gp_lock, flags);
+	if (!rcu_seq_state(sp->srcu_gp_seq_needed)) {
+		spin_unlock_irqrestore(&sp->gp_lock, flags);
+		return;
+	}
+	init_srcu_struct_fields(sp, true);
+	spin_unlock_irqrestore(&sp->gp_lock, flags);
+}
+
+/*
+ * Returns approximate total of the readers' ->srcu_lock_count[] values
+ * for the rank of per-CPU counters specified by idx.
+ */
+static unsigned long srcu_readers_lock_idx(struct srcu_struct *sp, int idx)
+{
+	int cpu;
+	unsigned long sum = 0;
+
+	for_each_possible_cpu(cpu) {
+		struct srcu_data *cpuc = per_cpu_ptr(sp->sda, cpu);
+
+		sum += READ_ONCE(cpuc->srcu_lock_count[idx]);
+	}
+	return sum;
+}
+
+/*
+ * Returns approximate total of the readers' ->srcu_unlock_count[] values
+ * for the rank of per-CPU counters specified by idx.
+ */
+static unsigned long srcu_readers_unlock_idx(struct srcu_struct *sp, int idx)
+{
+	int cpu;
+	unsigned long sum = 0;
+
+	for_each_possible_cpu(cpu) {
+		struct srcu_data *cpuc = per_cpu_ptr(sp->sda, cpu);
+
+		sum += READ_ONCE(cpuc->srcu_unlock_count[idx]);
+	}
+	return sum;
+}
+
+/*
+ * Return true if the number of pre-existing readers is determined to
+ * be zero.
+ */
+static bool srcu_readers_active_idx_check(struct srcu_struct *sp, int idx)
+{
+	unsigned long unlocks;
+
+	unlocks = srcu_readers_unlock_idx(sp, idx);
+
+	/*
+	 * Make sure that a lock is always counted if the corresponding
+	 * unlock is counted. Needs to be a smp_mb() as the read side may
+	 * contain a read from a variable that is written to before the
+	 * synchronize_srcu() in the write side. In this case smp_mb()s
+	 * A and B act like the store buffering pattern.
+	 *
+	 * This smp_mb() also pairs with smp_mb() C to prevent accesses
+	 * after the synchronize_srcu() from being executed before the
+	 * grace period ends.
+	 */
+	smp_mb(); /* A */
+
+	/*
+	 * If the locks are the same as the unlocks, then there must have
+	 * been no readers on this index at some time in between. This does
+	 * not mean that there are no more readers, as one could have read
+	 * the current index but not have incremented the lock counter yet.
+	 *
+	 * Possible bug: There is no guarantee that there haven't been
+	 * ULONG_MAX increments of ->srcu_lock_count[] since the unlocks were
+	 * counted, meaning that this could return true even if there are
+	 * still active readers.  Since there are no memory barriers around
+	 * srcu_flip(), the CPU is not required to increment ->srcu_idx
+	 * before running srcu_readers_unlock_idx(), which means that there
+	 * could be an arbitrarily large number of critical sections that
+	 * execute after srcu_readers_unlock_idx() but use the old value
+	 * of ->srcu_idx.
+	 */
+	return srcu_readers_lock_idx(sp, idx) == unlocks;
+}
+
+/**
+ * srcu_readers_active - returns true if there are readers. and false
+ *                       otherwise
+ * @sp: which srcu_struct to count active readers (holding srcu_read_lock).
+ *
+ * Note that this is not an atomic primitive, and can therefore suffer
+ * severe errors when invoked on an active srcu_struct.  That said, it
+ * can be useful as an error check at cleanup time.
+ */
+static bool srcu_readers_active(struct srcu_struct *sp)
+{
+	int cpu;
+	unsigned long sum = 0;
+
+	for_each_possible_cpu(cpu) {
+		struct srcu_data *cpuc = per_cpu_ptr(sp->sda, cpu);
+
+		sum += READ_ONCE(cpuc->srcu_lock_count[0]);
+		sum += READ_ONCE(cpuc->srcu_lock_count[1]);
+		sum -= READ_ONCE(cpuc->srcu_unlock_count[0]);
+		sum -= READ_ONCE(cpuc->srcu_unlock_count[1]);
+	}
+	return sum;
+}
+
+#define SRCU_INTERVAL		1
+
+/**
+ * cleanup_srcu_struct - deconstruct a sleep-RCU structure
+ * @sp: structure to clean up.
+ *
+ * Must invoke this after you are finished using a given srcu_struct that
+ * was initialized via init_srcu_struct(), else you leak memory.
+ */
+void cleanup_srcu_struct(struct srcu_struct *sp)
+{
+	int cpu;
+
+	WARN_ON_ONCE(atomic_read(&sp->srcu_exp_cnt));
+	if (WARN_ON(srcu_readers_active(sp)))
+		return; /* Leakage unless caller handles error. */
+	flush_delayed_work(&sp->work);
+	for_each_possible_cpu(cpu)
+		flush_delayed_work(&per_cpu_ptr(sp->sda, cpu)->work);
+	if (WARN_ON(rcu_seq_state(READ_ONCE(sp->srcu_gp_seq)) != SRCU_STATE_IDLE) ||
+	    WARN_ON(srcu_readers_active(sp))) {
+		pr_info("cleanup_srcu_struct: Active srcu_struct %p state: %d\n", sp, rcu_seq_state(READ_ONCE(sp->srcu_gp_seq)));
+		return; /* Caller forgot to stop doing call_srcu()? */
+	}
+	free_percpu(sp->sda);
+	sp->sda = NULL;
+}
+EXPORT_SYMBOL_GPL(cleanup_srcu_struct);
+
+/*
+ * Counts the new reader in the appropriate per-CPU element of the
+ * srcu_struct.  Must be called from process context.
+ * Returns an index that must be passed to the matching srcu_read_unlock().
+ */
+int __srcu_read_lock(struct srcu_struct *sp)
+{
+	int idx;
+
+	idx = READ_ONCE(sp->srcu_idx) & 0x1;
+	__this_cpu_inc(sp->sda->srcu_lock_count[idx]);
+	smp_mb(); /* B */  /* Avoid leaking the critical section. */
+	return idx;
+}
+EXPORT_SYMBOL_GPL(__srcu_read_lock);
+
+/*
+ * Removes the count for the old reader from the appropriate per-CPU
+ * element of the srcu_struct.  Note that this may well be a different
+ * CPU than that which was incremented by the corresponding srcu_read_lock().
+ * Must be called from process context.
+ */
+void __srcu_read_unlock(struct srcu_struct *sp, int idx)
+{
+	smp_mb(); /* C */  /* Avoid leaking the critical section. */
+	this_cpu_inc(sp->sda->srcu_unlock_count[idx]);
+}
+EXPORT_SYMBOL_GPL(__srcu_read_unlock);
+
+/*
+ * We use an adaptive strategy for synchronize_srcu() and especially for
+ * synchronize_srcu_expedited().  We spin for a fixed time period
+ * (defined below) to allow SRCU readers to exit their read-side critical
+ * sections.  If there are still some readers after a few microseconds,
+ * we repeatedly block for 1-millisecond time periods.
+ */
+#define SRCU_RETRY_CHECK_DELAY		5
+
+/*
+ * Start an SRCU grace period.
+ */
+static void srcu_gp_start(struct srcu_struct *sp)
+{
+	struct srcu_data *sdp = this_cpu_ptr(sp->sda);
+	int state;
+
+	RCU_LOCKDEP_WARN(!lockdep_is_held(&sp->gp_lock),
+			 "Invoked srcu_gp_start() without ->gp_lock!");
+	WARN_ON_ONCE(ULONG_CMP_GE(sp->srcu_gp_seq, sp->srcu_gp_seq_needed));
+	rcu_segcblist_advance(&sdp->srcu_cblist,
+			      rcu_seq_current(&sp->srcu_gp_seq));
+	(void)rcu_segcblist_accelerate(&sdp->srcu_cblist,
+				       rcu_seq_snap(&sp->srcu_gp_seq));
+	rcu_seq_start(&sp->srcu_gp_seq);
+	state = rcu_seq_state(READ_ONCE(sp->srcu_gp_seq));
+	WARN_ON_ONCE(state != SRCU_STATE_SCAN1);
+}
+
+/*
+ * Track online CPUs to guide callback workqueue placement.
+ */
+DEFINE_PER_CPU(bool, srcu_online);
+
+void srcu_online_cpu(unsigned int cpu)
+{
+	WRITE_ONCE(per_cpu(srcu_online, cpu), true);
+}
+
+void srcu_offline_cpu(unsigned int cpu)
+{
+	WRITE_ONCE(per_cpu(srcu_online, cpu), false);
+}
+
+/*
+ * Place the workqueue handler on the specified CPU if online, otherwise
+ * just run it whereever.  This is useful for placing workqueue handlers
+ * that are to invoke the specified CPU's callbacks.
+ */
+static bool srcu_queue_delayed_work_on(int cpu, struct workqueue_struct *wq,
+				       struct delayed_work *dwork,
+				       unsigned long delay)
+{
+	bool ret;
+
+	preempt_disable();
+	if (READ_ONCE(per_cpu(srcu_online, cpu)))
+		ret = queue_delayed_work_on(cpu, wq, dwork, delay);
+	else
+		ret = queue_delayed_work(wq, dwork, delay);
+	preempt_enable();
+	return ret;
+}
+
+/*
+ * Schedule callback invocation for the specified srcu_data structure,
+ * if possible, on the corresponding CPU.
+ */
+static void srcu_schedule_cbs_sdp(struct srcu_data *sdp, unsigned long delay)
+{
+	srcu_queue_delayed_work_on(sdp->cpu, system_power_efficient_wq,
+				   &sdp->work, delay);
+}
+
+/*
+ * Schedule callback invocation for all srcu_data structures associated
+ * with the specified srcu_node structure, if possible, on the corresponding
+ * CPUs.
+ */
+static void srcu_schedule_cbs_snp(struct srcu_struct *sp, struct srcu_node *snp)
+{
+	int cpu;
+
+	for (cpu = snp->grplo; cpu <= snp->grphi; cpu++)
+		srcu_schedule_cbs_sdp(per_cpu_ptr(sp->sda, cpu),
+				      atomic_read(&sp->srcu_exp_cnt) ? 0 : SRCU_INTERVAL);
+}
+
+/*
+ * Note the end of an SRCU grace period.  Initiates callback invocation
+ * and starts a new grace period if needed.
+ *
+ * The ->srcu_cb_mutex acquisition does not protect any data, but
+ * instead prevents more than one grace period from starting while we
+ * are initiating callback invocation.  This allows the ->srcu_have_cbs[]
+ * array to have a finite number of elements.
+ */
+static void srcu_gp_end(struct srcu_struct *sp)
+{
+	bool cbs;
+	unsigned long gpseq;
+	int idx;
+	int idxnext;
+	struct srcu_node *snp;
+
+	/* Prevent more than one additional grace period. */
+	mutex_lock(&sp->srcu_cb_mutex);
+
+	/* End the current grace period. */
+	spin_lock_irq(&sp->gp_lock);
+	idx = rcu_seq_state(sp->srcu_gp_seq);
+	WARN_ON_ONCE(idx != SRCU_STATE_SCAN2);
+	rcu_seq_end(&sp->srcu_gp_seq);
+	gpseq = rcu_seq_current(&sp->srcu_gp_seq);
+	spin_unlock_irq(&sp->gp_lock);
+	mutex_unlock(&sp->srcu_gp_mutex);
+	/* A new grace period can start at this point.  But only one. */
+
+	/* Initiate callback invocation as needed. */
+	idx = rcu_seq_ctr(gpseq) % ARRAY_SIZE(snp->srcu_have_cbs);
+	idxnext = (idx + 1) % ARRAY_SIZE(snp->srcu_have_cbs);
+	rcu_for_each_node_breadth_first(sp, snp) {
+		spin_lock_irq(&snp->lock);
+		cbs = false;
+		if (snp >= sp->level[rcu_num_lvls - 1])
+			cbs = snp->srcu_have_cbs[idx] == gpseq;
+		snp->srcu_have_cbs[idx] = gpseq;
+		rcu_seq_set_state(&snp->srcu_have_cbs[idx], 1);
+		spin_unlock_irq(&snp->lock);
+		if (cbs) {
+			smp_mb(); /* GP end before CB invocation. */
+			srcu_schedule_cbs_snp(sp, snp);
+		}
+	}
+
+	/* Callback initiation done, allow grace periods after next. */
+	mutex_unlock(&sp->srcu_cb_mutex);
+
+	/* Start a new grace period if needed. */
+	spin_lock_irq(&sp->gp_lock);
+	gpseq = rcu_seq_current(&sp->srcu_gp_seq);
+	if (!rcu_seq_state(gpseq) &&
+	    ULONG_CMP_LT(gpseq, sp->srcu_gp_seq_needed)) {
+		srcu_gp_start(sp);
+		spin_unlock_irq(&sp->gp_lock);
+		/* Throttle expedited grace periods: Should be rare! */
+		srcu_reschedule(sp, atomic_read(&sp->srcu_exp_cnt) &&
+				    rcu_seq_ctr(gpseq) & 0xf
+				    ? 0
+				    : SRCU_INTERVAL);
+	} else {
+		spin_unlock_irq(&sp->gp_lock);
+	}
+}
+
+/*
+ * Funnel-locking scheme to scalably mediate many concurrent grace-period
+ * requests.  The winner has to do the work of actually starting grace
+ * period s.  Losers must either ensure that their desired grace-period
+ * number is recorded on at least their leaf srcu_node structure, or they
+ * must take steps to invoke their own callbacks.
+ */
+static void srcu_funnel_gp_start(struct srcu_struct *sp,
+				 struct srcu_data *sdp,
+				 unsigned long s)
+{
+	unsigned long flags;
+	int idx = rcu_seq_ctr(s) % ARRAY_SIZE(sdp->mynode->srcu_have_cbs);
+	struct srcu_node *snp = sdp->mynode;
+	unsigned long snp_seq;
+
+	/* Each pass through the loop does one level of the srcu_node tree. */
+	for (; snp != NULL; snp = snp->srcu_parent) {
+		if (rcu_seq_done(&sp->srcu_gp_seq, s) && snp != sdp->mynode)
+			return; /* GP already done and CBs recorded. */
+		spin_lock_irqsave(&snp->lock, flags);
+		if (ULONG_CMP_GE(snp->srcu_have_cbs[idx], s)) {
+			snp_seq = snp->srcu_have_cbs[idx];
+			spin_unlock_irqrestore(&snp->lock, flags);
+			if (snp == sdp->mynode && snp_seq != s) {
+				smp_mb(); /* CBs after GP! */
+				srcu_schedule_cbs_sdp(sdp, 0);
+			}
+			return;
+		}
+		snp->srcu_have_cbs[idx] = s;
+		spin_unlock_irqrestore(&snp->lock, flags);
+	}
+
+	/* Top of tree, must ensure the grace period will be started. */
+	spin_lock_irqsave(&sp->gp_lock, flags);
+	if (ULONG_CMP_LT(sp->srcu_gp_seq_needed, s)) {
+		/*
+		 * Record need for grace period s.  Pair with load
+		 * acquire setting up for initialization.
+		 */
+		smp_store_release(&sp->srcu_gp_seq_needed, s); /*^^^*/
+	}
+
+	/* If grace period not already done and none in progress, start it. */
+	if (!rcu_seq_done(&sp->srcu_gp_seq, s) &&
+	    rcu_seq_state(sp->srcu_gp_seq) == SRCU_STATE_IDLE) {
+		WARN_ON_ONCE(ULONG_CMP_GE(sp->srcu_gp_seq, sp->srcu_gp_seq_needed));
+		srcu_gp_start(sp);
+		queue_delayed_work(system_power_efficient_wq, &sp->work,
+				   atomic_read(&sp->srcu_exp_cnt)
+				   ? 0
+				   : SRCU_INTERVAL);
+	}
+	spin_unlock_irqrestore(&sp->gp_lock, flags);
+}
+
+/*
+ * Wait until all readers counted by array index idx complete, but
+ * loop an additional time if there is an expedited grace period pending.
+ * The caller must ensure that ->srcu_idx is not changed while checking.
+ */
+static bool try_check_zero(struct srcu_struct *sp, int idx, int trycount)
+{
+	for (;;) {
+		if (srcu_readers_active_idx_check(sp, idx))
+			return true;
+		if (--trycount + !!atomic_read(&sp->srcu_exp_cnt) <= 0)
+			return false;
+		udelay(SRCU_RETRY_CHECK_DELAY);
+	}
+}
+
+/*
+ * Increment the ->srcu_idx counter so that future SRCU readers will
+ * use the other rank of the ->srcu_(un)lock_count[] arrays.  This allows
+ * us to wait for pre-existing readers in a starvation-free manner.
+ */
+static void srcu_flip(struct srcu_struct *sp)
+{
+	WRITE_ONCE(sp->srcu_idx, sp->srcu_idx + 1);
+
+	/*
+	 * Ensure that if the updater misses an __srcu_read_unlock()
+	 * increment, that task's next __srcu_read_lock() will see the
+	 * above counter update.  Note that both this memory barrier
+	 * and the one in srcu_readers_active_idx_check() provide the
+	 * guarantee for __srcu_read_lock().
+	 */
+	smp_mb(); /* D */  /* Pairs with C. */
+}
+
+/*
+ * Enqueue an SRCU callback on the srcu_data structure associated with
+ * the current CPU and the specified srcu_struct structure, initiating
+ * grace-period processing if it is not already running.
+ *
+ * Note that all CPUs must agree that the grace period extended beyond
+ * all pre-existing SRCU read-side critical section.  On systems with
+ * more than one CPU, this means that when "func()" is invoked, each CPU
+ * is guaranteed to have executed a full memory barrier since the end of
+ * its last corresponding SRCU read-side critical section whose beginning
+ * preceded the call to call_rcu().  It also means that each CPU executing
+ * an SRCU read-side critical section that continues beyond the start of
+ * "func()" must have executed a memory barrier after the call_rcu()
+ * but before the beginning of that SRCU read-side critical section.
+ * Note that these guarantees include CPUs that are offline, idle, or
+ * executing in user mode, as well as CPUs that are executing in the kernel.
+ *
+ * Furthermore, if CPU A invoked call_rcu() and CPU B invoked the
+ * resulting SRCU callback function "func()", then both CPU A and CPU
+ * B are guaranteed to execute a full memory barrier during the time
+ * interval between the call to call_rcu() and the invocation of "func()".
+ * This guarantee applies even if CPU A and CPU B are the same CPU (but
+ * again only if the system has more than one CPU).
+ *
+ * Of course, these guarantees apply only for invocations of call_srcu(),
+ * srcu_read_lock(), and srcu_read_unlock() that are all passed the same
+ * srcu_struct structure.
+ */
+void call_srcu(struct srcu_struct *sp, struct rcu_head *rhp,
+	       rcu_callback_t func)
+{
+	unsigned long flags;
+	bool needgp = false;
+	unsigned long s;
+	struct srcu_data *sdp;
+
+	check_init_srcu_struct(sp);
+	rhp->func = func;
+	local_irq_save(flags);
+	sdp = this_cpu_ptr(sp->sda);
+	spin_lock(&sdp->lock);
+	rcu_segcblist_enqueue(&sdp->srcu_cblist, rhp, false);
+	rcu_segcblist_advance(&sdp->srcu_cblist,
+			      rcu_seq_current(&sp->srcu_gp_seq));
+	s = rcu_seq_snap(&sp->srcu_gp_seq);
+	(void)rcu_segcblist_accelerate(&sdp->srcu_cblist, s);
+	if (ULONG_CMP_LT(sdp->srcu_gp_seq_needed, s)) {
+		sdp->srcu_gp_seq_needed = s;
+		needgp = true;
+	}
+	spin_unlock_irqrestore(&sdp->lock, flags);
+	if (needgp)
+		srcu_funnel_gp_start(sp, sdp, s);
+}
+EXPORT_SYMBOL_GPL(call_srcu);
+
+/*
+ * Helper function for synchronize_srcu() and synchronize_srcu_expedited().
+ */
+static void __synchronize_srcu(struct srcu_struct *sp)
+{
+	struct rcu_synchronize rcu;
+
+	RCU_LOCKDEP_WARN(lock_is_held(&sp->dep_map) ||
+			 lock_is_held(&rcu_bh_lock_map) ||
+			 lock_is_held(&rcu_lock_map) ||
+			 lock_is_held(&rcu_sched_lock_map),
+			 "Illegal synchronize_srcu() in same-type SRCU (or in RCU) read-side critical section");
+
+	if (rcu_scheduler_active == RCU_SCHEDULER_INACTIVE)
+		return;
+	might_sleep();
+	check_init_srcu_struct(sp);
+	init_completion(&rcu.completion);
+	init_rcu_head_on_stack(&rcu.head);
+	call_srcu(sp, &rcu.head, wakeme_after_rcu);
+	wait_for_completion(&rcu.completion);
+	destroy_rcu_head_on_stack(&rcu.head);
+}
+
+/**
+ * synchronize_srcu_expedited - Brute-force SRCU grace period
+ * @sp: srcu_struct with which to synchronize.
+ *
+ * Wait for an SRCU grace period to elapse, but be more aggressive about
+ * spinning rather than blocking when waiting.
+ *
+ * Note that synchronize_srcu_expedited() has the same deadlock and
+ * memory-ordering properties as does synchronize_srcu().
+ */
+void synchronize_srcu_expedited(struct srcu_struct *sp)
+{
+	bool do_norm = rcu_gp_is_normal();
+
+	check_init_srcu_struct(sp);
+	if (!do_norm) {
+		atomic_inc(&sp->srcu_exp_cnt);
+		smp_mb__after_atomic(); /* increment before GP. */
+	}
+	__synchronize_srcu(sp);
+	if (!do_norm) {
+		smp_mb__before_atomic(); /* GP before decrement. */
+		WARN_ON_ONCE(atomic_dec_return(&sp->srcu_exp_cnt) < 0);
+	}
+}
+EXPORT_SYMBOL_GPL(synchronize_srcu_expedited);
+
+/**
+ * synchronize_srcu - wait for prior SRCU read-side critical-section completion
+ * @sp: srcu_struct with which to synchronize.
+ *
+ * Wait for the count to drain to zero of both indexes. To avoid the
+ * possible starvation of synchronize_srcu(), it waits for the count of
+ * the index=((->srcu_idx & 1) ^ 1) to drain to zero at first,
+ * and then flip the srcu_idx and wait for the count of the other index.
+ *
+ * Can block; must be called from process context.
+ *
+ * Note that it is illegal to call synchronize_srcu() from the corresponding
+ * SRCU read-side critical section; doing so will result in deadlock.
+ * However, it is perfectly legal to call synchronize_srcu() on one
+ * srcu_struct from some other srcu_struct's read-side critical section,
+ * as long as the resulting graph of srcu_structs is acyclic.
+ *
+ * There are memory-ordering constraints implied by synchronize_srcu().
+ * On systems with more than one CPU, when synchronize_srcu() returns,
+ * each CPU is guaranteed to have executed a full memory barrier since
+ * the end of its last corresponding SRCU-sched read-side critical section
+ * whose beginning preceded the call to synchronize_srcu().  In addition,
+ * each CPU having an SRCU read-side critical section that extends beyond
+ * the return from synchronize_srcu() is guaranteed to have executed a
+ * full memory barrier after the beginning of synchronize_srcu() and before
+ * the beginning of that SRCU read-side critical section.  Note that these
+ * guarantees include CPUs that are offline, idle, or executing in user mode,
+ * as well as CPUs that are executing in the kernel.
+ *
+ * Furthermore, if CPU A invoked synchronize_srcu(), which returned
+ * to its caller on CPU B, then both CPU A and CPU B are guaranteed
+ * to have executed a full memory barrier during the execution of
+ * synchronize_srcu().  This guarantee applies even if CPU A and CPU B
+ * are the same CPU, but again only if the system has more than one CPU.
+ *
+ * Of course, these memory-ordering guarantees apply only when
+ * synchronize_srcu(), srcu_read_lock(), and srcu_read_unlock() are
+ * passed the same srcu_struct structure.
+ */
+void synchronize_srcu(struct srcu_struct *sp)
+{
+	if (rcu_gp_is_expedited())
+		synchronize_srcu_expedited(sp);
+	else
+		__synchronize_srcu(sp);
+}
+EXPORT_SYMBOL_GPL(synchronize_srcu);
+
+/*
+ * Callback function for srcu_barrier() use.
+ */
+static void srcu_barrier_cb(struct rcu_head *rhp)
+{
+	struct srcu_data *sdp;
+	struct srcu_struct *sp;
+
+	sdp = container_of(rhp, struct srcu_data, srcu_barrier_head);
+	sp = sdp->sp;
+	if (atomic_dec_and_test(&sp->srcu_barrier_cpu_cnt))
+		complete(&sp->srcu_barrier_completion);
+}
+
+/**
+ * srcu_barrier - Wait until all in-flight call_srcu() callbacks complete.
+ * @sp: srcu_struct on which to wait for in-flight callbacks.
+ */
+void srcu_barrier(struct srcu_struct *sp)
+{
+	int cpu;
+	struct srcu_data *sdp;
+	unsigned long s = rcu_seq_snap(&sp->srcu_barrier_seq);
+
+	check_init_srcu_struct(sp);
+	mutex_lock(&sp->srcu_barrier_mutex);
+	if (rcu_seq_done(&sp->srcu_barrier_seq, s)) {
+		smp_mb(); /* Force ordering following return. */
+		mutex_unlock(&sp->srcu_barrier_mutex);
+		return; /* Someone else did our work for us. */
+	}
+	rcu_seq_start(&sp->srcu_barrier_seq);
+	init_completion(&sp->srcu_barrier_completion);
+
+	/* Initial count prevents reaching zero until all CBs are posted. */
+	atomic_set(&sp->srcu_barrier_cpu_cnt, 1);
+
+	/*
+	 * Each pass through this loop enqueues a callback, but only
+	 * on CPUs already having callbacks enqueued.  Note that if
+	 * a CPU already has callbacks enqueue, it must have already
+	 * registered the need for a future grace period, so all we
+	 * need do is enqueue a callback that will use the same
+	 * grace period as the last callback already in the queue.
+	 */
+	for_each_possible_cpu(cpu) {
+		sdp = per_cpu_ptr(sp->sda, cpu);
+		spin_lock_irq(&sdp->lock);
+		atomic_inc(&sp->srcu_barrier_cpu_cnt);
+		sdp->srcu_barrier_head.func = srcu_barrier_cb;
+		if (!rcu_segcblist_entrain(&sdp->srcu_cblist,
+					   &sdp->srcu_barrier_head, 0))
+			atomic_dec(&sp->srcu_barrier_cpu_cnt);
+		spin_unlock_irq(&sdp->lock);
+	}
+
+	/* Remove the initial count, at which point reaching zero can happen. */
+	if (atomic_dec_and_test(&sp->srcu_barrier_cpu_cnt))
+		complete(&sp->srcu_barrier_completion);
+	wait_for_completion(&sp->srcu_barrier_completion);
+
+	rcu_seq_end(&sp->srcu_barrier_seq);
+	mutex_unlock(&sp->srcu_barrier_mutex);
+}
+EXPORT_SYMBOL_GPL(srcu_barrier);
+
+/**
+ * srcu_batches_completed - return batches completed.
+ * @sp: srcu_struct on which to report batch completion.
+ *
+ * Report the number of batches, correlated with, but not necessarily
+ * precisely the same as, the number of grace periods that have elapsed.
+ */
+unsigned long srcu_batches_completed(struct srcu_struct *sp)
+{
+	return sp->srcu_idx;
+}
+EXPORT_SYMBOL_GPL(srcu_batches_completed);
+
+/*
+ * Core SRCU state machine.  Push state bits of ->srcu_gp_seq
+ * to SRCU_STATE_SCAN2, and invoke srcu_gp_end() when scan has
+ * completed in that state.
+ */
+static void srcu_advance_state(struct srcu_struct *sp)
+{
+	int idx;
+
+	mutex_lock(&sp->srcu_gp_mutex);
+
+	/*
+	 * Because readers might be delayed for an extended period after
+	 * fetching ->srcu_idx for their index, at any point in time there
+	 * might well be readers using both idx=0 and idx=1.  We therefore
+	 * need to wait for readers to clear from both index values before
+	 * invoking a callback.
+	 *
+	 * The load-acquire ensures that we see the accesses performed
+	 * by the prior grace period.
+	 */
+	idx = rcu_seq_state(smp_load_acquire(&sp->srcu_gp_seq)); /* ^^^ */
+	if (idx == SRCU_STATE_IDLE) {
+		spin_lock_irq(&sp->gp_lock);
+		if (ULONG_CMP_GE(sp->srcu_gp_seq, sp->srcu_gp_seq_needed)) {
+			WARN_ON_ONCE(rcu_seq_state(sp->srcu_gp_seq));
+			spin_unlock_irq(&sp->gp_lock);
+			mutex_unlock(&sp->srcu_gp_mutex);
+			return;
+		}
+		idx = rcu_seq_state(READ_ONCE(sp->srcu_gp_seq));
+		if (idx == SRCU_STATE_IDLE)
+			srcu_gp_start(sp);
+		spin_unlock_irq(&sp->gp_lock);
+		if (idx != SRCU_STATE_IDLE) {
+			mutex_unlock(&sp->srcu_gp_mutex);
+			return; /* Someone else started the grace period. */
+		}
+	}
+
+	if (rcu_seq_state(READ_ONCE(sp->srcu_gp_seq)) == SRCU_STATE_SCAN1) {
+		idx = 1 ^ (sp->srcu_idx & 1);
+		if (!try_check_zero(sp, idx, 1)) {
+			mutex_unlock(&sp->srcu_gp_mutex);
+			return; /* readers present, retry later. */
+		}
+		srcu_flip(sp);
+		rcu_seq_set_state(&sp->srcu_gp_seq, SRCU_STATE_SCAN2);
+	}
+
+	if (rcu_seq_state(READ_ONCE(sp->srcu_gp_seq)) == SRCU_STATE_SCAN2) {
+
+		/*
+		 * SRCU read-side critical sections are normally short,
+		 * so check at least twice in quick succession after a flip.
+		 */
+		idx = 1 ^ (sp->srcu_idx & 1);
+		if (!try_check_zero(sp, idx, 2)) {
+			mutex_unlock(&sp->srcu_gp_mutex);
+			return; /* readers present, retry later. */
+		}
+		srcu_gp_end(sp);  /* Releases ->srcu_gp_mutex. */
+	}
+}
+
+/*
+ * Invoke a limited number of SRCU callbacks that have passed through
+ * their grace period.  If there are more to do, SRCU will reschedule
+ * the workqueue.  Note that needed memory barriers have been executed
+ * in this task's context by srcu_readers_active_idx_check().
+ */
+static void srcu_invoke_callbacks(struct work_struct *work)
+{
+	bool more;
+	struct rcu_cblist ready_cbs;
+	struct rcu_head *rhp;
+	struct srcu_data *sdp;
+	struct srcu_struct *sp;
+
+	sdp = container_of(work, struct srcu_data, work.work);
+	sp = sdp->sp;
+	rcu_cblist_init(&ready_cbs);
+	spin_lock_irq(&sdp->lock);
+	smp_mb(); /* Old grace periods before callback invocation! */
+	rcu_segcblist_advance(&sdp->srcu_cblist,
+			      rcu_seq_current(&sp->srcu_gp_seq));
+	if (sdp->srcu_cblist_invoking ||
+	    !rcu_segcblist_ready_cbs(&sdp->srcu_cblist)) {
+		spin_unlock_irq(&sdp->lock);
+		return;  /* Someone else on the job or nothing to do. */
+	}
+
+	/* We are on the job!  Extract and invoke ready callbacks. */
+	sdp->srcu_cblist_invoking = true;
+	rcu_segcblist_extract_done_cbs(&sdp->srcu_cblist, &ready_cbs);
+	spin_unlock_irq(&sdp->lock);
+	rhp = rcu_cblist_dequeue(&ready_cbs);
+	for (; rhp != NULL; rhp = rcu_cblist_dequeue(&ready_cbs)) {
+		local_bh_disable();
+		rhp->func(rhp);
+		local_bh_enable();
+	}
+
+	/*
+	 * Update counts, accelerate new callbacks, and if needed,
+	 * schedule another round of callback invocation.
+	 */
+	spin_lock_irq(&sdp->lock);
+	rcu_segcblist_insert_count(&sdp->srcu_cblist, &ready_cbs);
+	(void)rcu_segcblist_accelerate(&sdp->srcu_cblist,
+				       rcu_seq_snap(&sp->srcu_gp_seq));
+	sdp->srcu_cblist_invoking = false;
+	more = rcu_segcblist_ready_cbs(&sdp->srcu_cblist);
+	spin_unlock_irq(&sdp->lock);
+	if (more)
+		srcu_schedule_cbs_sdp(sdp, 0);
+}
+
+/*
+ * Finished one round of SRCU grace period.  Start another if there are
+ * more SRCU callbacks queued, otherwise put SRCU into not-running state.
+ */
+static void srcu_reschedule(struct srcu_struct *sp, unsigned long delay)
+{
+	bool pushgp = true;
+
+	spin_lock_irq(&sp->gp_lock);
+	if (ULONG_CMP_GE(sp->srcu_gp_seq, sp->srcu_gp_seq_needed)) {
+		if (!WARN_ON_ONCE(rcu_seq_state(sp->srcu_gp_seq))) {
+			/* All requests fulfilled, time to go idle. */
+			pushgp = false;
+		}
+	} else if (!rcu_seq_state(sp->srcu_gp_seq)) {
+		/* Outstanding request and no GP.  Start one. */
+		srcu_gp_start(sp);
+	}
+	spin_unlock_irq(&sp->gp_lock);
+
+	if (pushgp)
+		queue_delayed_work(system_power_efficient_wq, &sp->work, delay);
+}
+
+/*
+ * This is the work-queue function that handles SRCU grace periods.
+ */
+void process_srcu(struct work_struct *work)
+{
+	struct srcu_struct *sp;
+
+	sp = container_of(work, struct srcu_struct, work.work);
+
+	srcu_advance_state(sp);
+	srcu_reschedule(sp, atomic_read(&sp->srcu_exp_cnt) ? 0 : SRCU_INTERVAL);
+}
+EXPORT_SYMBOL_GPL(process_srcu);

kernel/rcu/tiny.c

@@ -79,7 +79,7 @@ EXPORT_SYMBOL(__rcu_is_watching);
  */
 static int rcu_qsctr_help(struct rcu_ctrlblk *rcp)
 {
-	RCU_TRACE(reset_cpu_stall_ticks(rcp));
+	RCU_TRACE(reset_cpu_stall_ticks(rcp);)
 	if (rcp->donetail != rcp->curtail) {
 		rcp->donetail = rcp->curtail;
 		return 1;
@@ -125,7 +125,7 @@ void rcu_bh_qs(void)
  */
 void rcu_check_callbacks(int user)
 {
-	RCU_TRACE(check_cpu_stalls());
+	RCU_TRACE(check_cpu_stalls();)
 	if (user)
 		rcu_sched_qs();
 	else if (!in_softirq())
@@ -143,7 +143,7 @@ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp)
 	const char *rn = NULL;
 	struct rcu_head *next, *list;
 	unsigned long flags;
-	RCU_TRACE(int cb_count = 0);
+	RCU_TRACE(int cb_count = 0;)
 
 	/* Move the ready-to-invoke callbacks to a local list. */
 	local_irq_save(flags);
@@ -152,7 +152,7 @@ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp)
 		local_irq_restore(flags);
 		return;
 	}
-	RCU_TRACE(trace_rcu_batch_start(rcp->name, 0, rcp->qlen, -1));
+	RCU_TRACE(trace_rcu_batch_start(rcp->name, 0, rcp->qlen, -1);)
 	list = rcp->rcucblist;
 	rcp->rcucblist = *rcp->donetail;
 	*rcp->donetail = NULL;
@@ -162,7 +162,7 @@ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp)
 	local_irq_restore(flags);
 
 	/* Invoke the callbacks on the local list. */
-	RCU_TRACE(rn = rcp->name);
+	RCU_TRACE(rn = rcp->name;)
 	while (list) {
 		next = list->next;
 		prefetch(next);
@@ -171,9 +171,9 @@ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp)
 		__rcu_reclaim(rn, list);
 		local_bh_enable();
 		list = next;
-		RCU_TRACE(cb_count++);
+		RCU_TRACE(cb_count++;)
 	}
-	RCU_TRACE(rcu_trace_sub_qlen(rcp, cb_count));
+	RCU_TRACE(rcu_trace_sub_qlen(rcp, cb_count);)
 	RCU_TRACE(trace_rcu_batch_end(rcp->name,
 				      cb_count, 0, need_resched(),
 				      is_idle_task(current),
@@ -221,7 +221,7 @@ static void __call_rcu(struct rcu_head *head,
 	local_irq_save(flags);
 	*rcp->curtail = head;
 	rcp->curtail = &head->next;
-	RCU_TRACE(rcp->qlen++);
+	RCU_TRACE(rcp->qlen++;)
 	local_irq_restore(flags);
 
 	if (unlikely(is_idle_task(current))) {
@@ -254,8 +254,8 @@ EXPORT_SYMBOL_GPL(call_rcu_bh);
 void __init rcu_init(void)
 {
 	open_softirq(RCU_SOFTIRQ, rcu_process_callbacks);
-	RCU_TRACE(reset_cpu_stall_ticks(&rcu_sched_ctrlblk));
-	RCU_TRACE(reset_cpu_stall_ticks(&rcu_bh_ctrlblk));
+	RCU_TRACE(reset_cpu_stall_ticks(&rcu_sched_ctrlblk);)
+	RCU_TRACE(reset_cpu_stall_ticks(&rcu_bh_ctrlblk);)
 
 	rcu_early_boot_tests();
 }

kernel/rcu/tiny_plugin.h

@@ -52,7 +52,7 @@ static struct rcu_ctrlblk rcu_bh_ctrlblk = {
 	RCU_TRACE(.name = "rcu_bh")
 };
 
-#ifdef CONFIG_DEBUG_LOCK_ALLOC
+#if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_SRCU)
 #include <linux/kernel_stat.h>
 
 int rcu_scheduler_active __read_mostly;
@@ -65,15 +65,16 @@ EXPORT_SYMBOL_GPL(rcu_scheduler_active);
  * to RCU_SCHEDULER_RUNNING, skipping the RCU_SCHEDULER_INIT stage.
  * The reason for this is that Tiny RCU does not need kthreads, so does
  * not have to care about the fact that the scheduler is half-initialized
- * at a certain phase of the boot process.
+ * at a certain phase of the boot process.  Unless SRCU is in the mix.
  */
 void __init rcu_scheduler_starting(void)
 {
 	WARN_ON(nr_context_switches() > 0);
-	rcu_scheduler_active = RCU_SCHEDULER_RUNNING;
+	rcu_scheduler_active = IS_ENABLED(CONFIG_SRCU)
+		? RCU_SCHEDULER_INIT : RCU_SCHEDULER_RUNNING;
 }
 
-#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
+#endif /* #if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_SRCU) */
 
 #ifdef CONFIG_RCU_TRACE
 
@@ -162,8 +163,8 @@ static void reset_cpu_stall_ticks(struct rcu_ctrlblk *rcp)
 
 static void check_cpu_stalls(void)
 {
-	RCU_TRACE(check_cpu_stall(&rcu_bh_ctrlblk));
-	RCU_TRACE(check_cpu_stall(&rcu_sched_ctrlblk));
+	RCU_TRACE(check_cpu_stall(&rcu_bh_ctrlblk);)
+	RCU_TRACE(check_cpu_stall(&rcu_sched_ctrlblk);)
 }
 
 #endif /* #ifdef CONFIG_RCU_TRACE */

kernel/rcu/tree.c

@@ -97,8 +97,8 @@ struct rcu_state sname##_state = { \
 	.gpnum = 0UL - 300UL, \
 	.completed = 0UL - 300UL, \
 	.orphan_lock = __RAW_SPIN_LOCK_UNLOCKED(&sname##_state.orphan_lock), \
-	.orphan_nxttail = &sname##_state.orphan_nxtlist, \
-	.orphan_donetail = &sname##_state.orphan_donelist, \
+	.orphan_pend = RCU_CBLIST_INITIALIZER(sname##_state.orphan_pend), \
+	.orphan_done = RCU_CBLIST_INITIALIZER(sname##_state.orphan_done), \
 	.barrier_mutex = __MUTEX_INITIALIZER(sname##_state.barrier_mutex), \
 	.name = RCU_STATE_NAME(sname), \
 	.abbr = sabbr, \
@@ -123,7 +123,7 @@ static int rcu_fanout_leaf = RCU_FANOUT_LEAF;
 module_param(rcu_fanout_leaf, int, 0444);
 int rcu_num_lvls __read_mostly = RCU_NUM_LVLS;
 /* Number of rcu_nodes at specified level. */
-static int num_rcu_lvl[] = NUM_RCU_LVL_INIT;
+int num_rcu_lvl[] = NUM_RCU_LVL_INIT;
 int rcu_num_nodes __read_mostly = NUM_RCU_NODES; /* Total # rcu_nodes in use. */
 /* panic() on RCU Stall sysctl. */
 int sysctl_panic_on_rcu_stall __read_mostly;
@@ -199,7 +199,7 @@ static const int gp_cleanup_delay;
 
 /*
  * Number of grace periods between delays, normalized by the duration of
- * the delay.  The longer the the delay, the more the grace periods between
+ * the delay.  The longer the delay, the more the grace periods between
  * each delay.  The reason for this normalization is that it means that,
  * for non-zero delays, the overall slowdown of grace periods is constant
  * regardless of the duration of the delay.  This arrangement balances
@@ -272,11 +272,19 @@ void rcu_bh_qs(void)
 	}
 }
 
-static DEFINE_PER_CPU(int, rcu_sched_qs_mask);
+/*
+ * Steal a bit from the bottom of ->dynticks for idle entry/exit
+ * control.  Initially this is for TLB flushing.
+ */
+#define RCU_DYNTICK_CTRL_MASK 0x1
+#define RCU_DYNTICK_CTRL_CTR  (RCU_DYNTICK_CTRL_MASK + 1)
+#ifndef rcu_eqs_special_exit
+#define rcu_eqs_special_exit() do { } while (0)
+#endif
 
 static DEFINE_PER_CPU(struct rcu_dynticks, rcu_dynticks) = {
 	.dynticks_nesting = DYNTICK_TASK_EXIT_IDLE,
-	.dynticks = ATOMIC_INIT(1),
+	.dynticks = ATOMIC_INIT(RCU_DYNTICK_CTRL_CTR),
 #ifdef CONFIG_NO_HZ_FULL_SYSIDLE
 	.dynticks_idle_nesting = DYNTICK_TASK_NEST_VALUE,
 	.dynticks_idle = ATOMIC_INIT(1),
@@ -290,15 +298,20 @@ static DEFINE_PER_CPU(struct rcu_dynticks, rcu_dynticks) = {
 static void rcu_dynticks_eqs_enter(void)
 {
 	struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks);
-	int special;
+	int seq;
 
 	/*
-	 * CPUs seeing atomic_inc_return() must see prior RCU read-side
+	 * CPUs seeing atomic_add_return() must see prior RCU read-side
 	 * critical sections, and we also must force ordering with the
 	 * next idle sojourn.
 	 */
-	special = atomic_inc_return(&rdtp->dynticks);
-	WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && special & 0x1);
+	seq = atomic_add_return(RCU_DYNTICK_CTRL_CTR, &rdtp->dynticks);
+	/* Better be in an extended quiescent state! */
+	WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) &&
+		     (seq & RCU_DYNTICK_CTRL_CTR));
+	/* Better not have special action (TLB flush) pending! */
+	WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) &&
+		     (seq & RCU_DYNTICK_CTRL_MASK));
 }
 
 /*
@@ -308,15 +321,22 @@ static void rcu_dynticks_eqs_enter(void)
 static void rcu_dynticks_eqs_exit(void)
 {
 	struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks);
-	int special;
+	int seq;
 
 	/*
-	 * CPUs seeing atomic_inc_return() must see prior idle sojourns,
+	 * CPUs seeing atomic_add_return() must see prior idle sojourns,
 	 * and we also must force ordering with the next RCU read-side
 	 * critical section.
 	 */
-	special = atomic_inc_return(&rdtp->dynticks);
-	WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && !(special & 0x1));
+	seq = atomic_add_return(RCU_DYNTICK_CTRL_CTR, &rdtp->dynticks);
+	WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) &&
+		     !(seq & RCU_DYNTICK_CTRL_CTR));
+	if (seq & RCU_DYNTICK_CTRL_MASK) {
+		atomic_andnot(RCU_DYNTICK_CTRL_MASK, &rdtp->dynticks);
+		smp_mb__after_atomic(); /* _exit after clearing mask. */
+		/* Prefer duplicate flushes to losing a flush. */
+		rcu_eqs_special_exit();
+	}
 }
 
 /*
@@ -333,9 +353,9 @@ static void rcu_dynticks_eqs_online(void)
 {
 	struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks);
 
-	if (atomic_read(&rdtp->dynticks) & 0x1)
+	if (atomic_read(&rdtp->dynticks) & RCU_DYNTICK_CTRL_CTR)
 		return;
-	atomic_add(0x1, &rdtp->dynticks);
+	atomic_add(RCU_DYNTICK_CTRL_CTR, &rdtp->dynticks);
 }
 
 /*
@@ -347,7 +367,7 @@ bool rcu_dynticks_curr_cpu_in_eqs(void)
 {
 	struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks);
 
-	return !(atomic_read(&rdtp->dynticks) & 0x1);
+	return !(atomic_read(&rdtp->dynticks) & RCU_DYNTICK_CTRL_CTR);
 }
 
 /*
@@ -358,7 +378,7 @@ int rcu_dynticks_snap(struct rcu_dynticks *rdtp)
 {
 	int snap = atomic_add_return(0, &rdtp->dynticks);
 
-	return snap;
+	return snap & ~RCU_DYNTICK_CTRL_MASK;
 }
 
 /*
@@ -367,7 +387,7 @@ int rcu_dynticks_snap(struct rcu_dynticks *rdtp)
  */
 static bool rcu_dynticks_in_eqs(int snap)
 {
-	return !(snap & 0x1);
+	return !(snap & RCU_DYNTICK_CTRL_CTR);
 }
 
 /*
@@ -387,14 +407,34 @@ static bool rcu_dynticks_in_eqs_since(struct rcu_dynticks *rdtp, int snap)
 static void rcu_dynticks_momentary_idle(void)
 {
 	struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks);
-	int special = atomic_add_return(2, &rdtp->dynticks);
+	int special = atomic_add_return(2 * RCU_DYNTICK_CTRL_CTR,
+					&rdtp->dynticks);
 
 	/* It is illegal to call this from idle state. */
-	WARN_ON_ONCE(!(special & 0x1));
+	WARN_ON_ONCE(!(special & RCU_DYNTICK_CTRL_CTR));
 }
 
-DEFINE_PER_CPU_SHARED_ALIGNED(unsigned long, rcu_qs_ctr);
-EXPORT_PER_CPU_SYMBOL_GPL(rcu_qs_ctr);
+/*
+ * Set the special (bottom) bit of the specified CPU so that it
+ * will take special action (such as flushing its TLB) on the
+ * next exit from an extended quiescent state.  Returns true if
+ * the bit was successfully set, or false if the CPU was not in
+ * an extended quiescent state.
+ */
+bool rcu_eqs_special_set(int cpu)
+{
+	int old;
+	int new;
+	struct rcu_dynticks *rdtp = &per_cpu(rcu_dynticks, cpu);
+
+	do {
+		old = atomic_read(&rdtp->dynticks);
+		if (old & RCU_DYNTICK_CTRL_CTR)
+			return false;
+		new = old | RCU_DYNTICK_CTRL_MASK;
+	} while (atomic_cmpxchg(&rdtp->dynticks, old, new) != old);
+	return true;
+}
 
 /*
  * Let the RCU core know that this CPU has gone through the scheduler,
@@ -403,44 +443,14 @@ EXPORT_PER_CPU_SYMBOL_GPL(rcu_qs_ctr);
  * memory barriers to let the RCU core know about it, regardless of what
  * this CPU might (or might not) do in the near future.
  *
- * We inform the RCU core by emulating a zero-duration dyntick-idle
- * period, which we in turn do by incrementing the ->dynticks counter
- * by two.
+ * We inform the RCU core by emulating a zero-duration dyntick-idle period.
  *
  * The caller must have disabled interrupts.
  */
 static void rcu_momentary_dyntick_idle(void)
 {
-	struct rcu_data *rdp;
-	int resched_mask;
-	struct rcu_state *rsp;
-
-	/*
-	 * Yes, we can lose flag-setting operations.  This is OK, because
-	 * the flag will be set again after some delay.
-	 */
-	resched_mask = raw_cpu_read(rcu_sched_qs_mask);
-	raw_cpu_write(rcu_sched_qs_mask, 0);
-
-	/* Find the flavor that needs a quiescent state. */
-	for_each_rcu_flavor(rsp) {
-		rdp = raw_cpu_ptr(rsp->rda);
-		if (!(resched_mask & rsp->flavor_mask))
-			continue;
-		smp_mb(); /* rcu_sched_qs_mask before cond_resched_completed. */
-		if (READ_ONCE(rdp->mynode->completed) !=
-		    READ_ONCE(rdp->cond_resched_completed))
-			continue;
-
-		/*
-		 * Pretend to be momentarily idle for the quiescent state.
-		 * This allows the grace-period kthread to record the
-		 * quiescent state, with no need for this CPU to do anything
-		 * further.
-		 */
-		rcu_dynticks_momentary_idle();
-		break;
-	}
+	raw_cpu_write(rcu_dynticks.rcu_need_heavy_qs, false);
+	rcu_dynticks_momentary_idle();
 }
 
 /*
@@ -448,14 +458,22 @@ static void rcu_momentary_dyntick_idle(void)
  * and requires special handling for preemptible RCU.
  * The caller must have disabled interrupts.
  */
-void rcu_note_context_switch(void)
+void rcu_note_context_switch(bool preempt)
 {
 	barrier(); /* Avoid RCU read-side critical sections leaking down. */
 	trace_rcu_utilization(TPS("Start context switch"));
 	rcu_sched_qs();
 	rcu_preempt_note_context_switch();
-	if (unlikely(raw_cpu_read(rcu_sched_qs_mask)))
+	/* Load rcu_urgent_qs before other flags. */
+	if (!smp_load_acquire(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs)))
+		goto out;
+	this_cpu_write(rcu_dynticks.rcu_urgent_qs, false);
+	if (unlikely(raw_cpu_read(rcu_dynticks.rcu_need_heavy_qs)))
 		rcu_momentary_dyntick_idle();
+	this_cpu_inc(rcu_dynticks.rcu_qs_ctr);
+	if (!preempt)
+		rcu_note_voluntary_context_switch_lite(current);
+out:
 	trace_rcu_utilization(TPS("End context switch"));
 	barrier(); /* Avoid RCU read-side critical sections leaking up. */
 }
@@ -478,29 +496,26 @@ void rcu_all_qs(void)
 {
 	unsigned long flags;
 
+	if (!raw_cpu_read(rcu_dynticks.rcu_urgent_qs))
+		return;
+	preempt_disable();
+	/* Load rcu_urgent_qs before other flags. */
+	if (!smp_load_acquire(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs))) {
+		preempt_enable();
+		return;
+	}
+	this_cpu_write(rcu_dynticks.rcu_urgent_qs, false);
 	barrier(); /* Avoid RCU read-side critical sections leaking down. */
-	if (unlikely(raw_cpu_read(rcu_sched_qs_mask))) {
+	if (unlikely(raw_cpu_read(rcu_dynticks.rcu_need_heavy_qs))) {
 		local_irq_save(flags);
 		rcu_momentary_dyntick_idle();
 		local_irq_restore(flags);
 	}
-	if (unlikely(raw_cpu_read(rcu_sched_data.cpu_no_qs.b.exp))) {
-		/*
-		 * Yes, we just checked a per-CPU variable with preemption
-		 * enabled, so we might be migrated to some other CPU at
-		 * this point.  That is OK because in that case, the
-		 * migration will supply the needed quiescent state.
-		 * We might end up needlessly disabling preemption and
-		 * invoking rcu_sched_qs() on the destination CPU, but
-		 * the probability and cost are both quite low, so this
-		 * should not be a problem in practice.
-		 */
-		preempt_disable();
+	if (unlikely(raw_cpu_read(rcu_sched_data.cpu_no_qs.b.exp)))
 		rcu_sched_qs();
-		preempt_enable();
-	}
-	this_cpu_inc(rcu_qs_ctr);
+	this_cpu_inc(rcu_dynticks.rcu_qs_ctr);
 	barrier(); /* Avoid RCU read-side critical sections leaking up. */
+	preempt_enable();
 }
 EXPORT_SYMBOL_GPL(rcu_all_qs);
 
@@ -712,16 +727,6 @@ void rcutorture_record_progress(unsigned long vernum)
 }
 EXPORT_SYMBOL_GPL(rcutorture_record_progress);
 
-/*
- * Does the CPU have callbacks ready to be invoked?
- */
-static int
-cpu_has_callbacks_ready_to_invoke(struct rcu_data *rdp)
-{
-	return &rdp->nxtlist != rdp->nxttail[RCU_DONE_TAIL] &&
-	       rdp->nxttail[RCU_NEXT_TAIL] != NULL;
-}
-
 /*
  * Return the root node of the specified rcu_state structure.
  */
@@ -752,21 +757,17 @@ static int rcu_future_needs_gp(struct rcu_state *rsp)
 static bool
 cpu_needs_another_gp(struct rcu_state *rsp, struct rcu_data *rdp)
 {
-	int i;
-
 	if (rcu_gp_in_progress(rsp))
 		return false;  /* No, a grace period is already in progress. */
 	if (rcu_future_needs_gp(rsp))
 		return true;  /* Yes, a no-CBs CPU needs one. */
-	if (!rdp->nxttail[RCU_NEXT_TAIL])
+	if (!rcu_segcblist_is_enabled(&rdp->cblist))
 		return false;  /* No, this is a no-CBs (or offline) CPU. */
-	if (*rdp->nxttail[RCU_NEXT_READY_TAIL])
+	if (!rcu_segcblist_restempty(&rdp->cblist, RCU_NEXT_READY_TAIL))
 		return true;  /* Yes, CPU has newly registered callbacks. */
-	for (i = RCU_WAIT_TAIL; i < RCU_NEXT_TAIL; i++)
-		if (rdp->nxttail[i - 1] != rdp->nxttail[i] &&
-		    ULONG_CMP_LT(READ_ONCE(rsp->completed),
-				 rdp->nxtcompleted[i]))
-			return true;  /* Yes, CBs for future grace period. */
+	if (rcu_segcblist_future_gp_needed(&rdp->cblist,
+					   READ_ONCE(rsp->completed)))
+		return true;  /* Yes, CBs for future grace period. */
 	return false; /* No grace period needed. */
 }
 
@@ -1150,6 +1151,24 @@ bool notrace rcu_is_watching(void)
 }
 EXPORT_SYMBOL_GPL(rcu_is_watching);
 
+/*
+ * If a holdout task is actually running, request an urgent quiescent
+ * state from its CPU.  This is unsynchronized, so migrations can cause
+ * the request to go to the wrong CPU.  Which is OK, all that will happen
+ * is that the CPU's next context switch will be a bit slower and next
+ * time around this task will generate another request.
+ */
+void rcu_request_urgent_qs_task(struct task_struct *t)
+{
+	int cpu;
+
+	barrier();
+	cpu = task_cpu(t);
+	if (!task_curr(t))
+		return; /* This task is not running on that CPU. */
+	smp_store_release(per_cpu_ptr(&rcu_dynticks.rcu_urgent_qs, cpu), true);
+}
+
 #if defined(CONFIG_PROVE_RCU) && defined(CONFIG_HOTPLUG_CPU)
 
 /*
@@ -1235,7 +1254,8 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp,
 				    bool *isidle, unsigned long *maxj)
 {
 	unsigned long jtsq;
-	int *rcrmp;
+	bool *rnhqp;
+	bool *ruqp;
 	unsigned long rjtsc;
 	struct rcu_node *rnp;
 
@@ -1271,11 +1291,15 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp,
 	 * might not be the case for nohz_full CPUs looping in the kernel.
 	 */
 	rnp = rdp->mynode;
+	ruqp = per_cpu_ptr(&rcu_dynticks.rcu_urgent_qs, rdp->cpu);
 	if (time_after(jiffies, rdp->rsp->gp_start + jtsq) &&
-	    READ_ONCE(rdp->rcu_qs_ctr_snap) != per_cpu(rcu_qs_ctr, rdp->cpu) &&
+	    READ_ONCE(rdp->rcu_qs_ctr_snap) != per_cpu(rcu_dynticks.rcu_qs_ctr, rdp->cpu) &&
 	    READ_ONCE(rdp->gpnum) == rnp->gpnum && !rdp->gpwrap) {
 		trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, TPS("rqc"));
 		return 1;
+	} else {
+		/* Load rcu_qs_ctr before store to rcu_urgent_qs. */
+		smp_store_release(ruqp, true);
 	}
 
 	/* Check for the CPU being offline. */
@@ -1292,7 +1316,7 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp,
 	 * in-kernel CPU-bound tasks cannot advance grace periods.
 	 * So if the grace period is old enough, make the CPU pay attention.
 	 * Note that the unsynchronized assignments to the per-CPU
-	 * rcu_sched_qs_mask variable are safe.  Yes, setting of
+	 * rcu_need_heavy_qs variable are safe.  Yes, setting of
 	 * bits can be lost, but they will be set again on the next
 	 * force-quiescent-state pass.  So lost bit sets do not result
 	 * in incorrect behavior, merely in a grace period lasting
@@ -1306,16 +1330,13 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp,
 	 * is set too high, we override with half of the RCU CPU stall
 	 * warning delay.
 	 */
-	rcrmp = &per_cpu(rcu_sched_qs_mask, rdp->cpu);
-	if (time_after(jiffies, rdp->rsp->gp_start + jtsq) ||
-	    time_after(jiffies, rdp->rsp->jiffies_resched)) {
-		if (!(READ_ONCE(*rcrmp) & rdp->rsp->flavor_mask)) {
-			WRITE_ONCE(rdp->cond_resched_completed,
-				   READ_ONCE(rdp->mynode->completed));
-			smp_mb(); /* ->cond_resched_completed before *rcrmp. */
-			WRITE_ONCE(*rcrmp,
-				   READ_ONCE(*rcrmp) + rdp->rsp->flavor_mask);
-		}
+	rnhqp = &per_cpu(rcu_dynticks.rcu_need_heavy_qs, rdp->cpu);
+	if (!READ_ONCE(*rnhqp) &&
+	    (time_after(jiffies, rdp->rsp->gp_start + jtsq) ||
+	     time_after(jiffies, rdp->rsp->jiffies_resched))) {
+		WRITE_ONCE(*rnhqp, true);
+		/* Store rcu_need_heavy_qs before rcu_urgent_qs. */
+		smp_store_release(ruqp, true);
 		rdp->rsp->jiffies_resched += 5; /* Re-enable beating. */
 	}
 
@@ -1475,7 +1496,8 @@ static void print_other_cpu_stall(struct rcu_state *rsp, unsigned long gpnum)
 
 	print_cpu_stall_info_end();
 	for_each_possible_cpu(cpu)
-		totqlen += per_cpu_ptr(rsp->rda, cpu)->qlen;
+		totqlen += rcu_segcblist_n_cbs(&per_cpu_ptr(rsp->rda,
+							    cpu)->cblist);
 	pr_cont("(detected by %d, t=%ld jiffies, g=%ld, c=%ld, q=%lu)\n",
 	       smp_processor_id(), (long)(jiffies - rsp->gp_start),
 	       (long)rsp->gpnum, (long)rsp->completed, totqlen);
@@ -1529,7 +1551,8 @@ static void print_cpu_stall(struct rcu_state *rsp)
 	print_cpu_stall_info(rsp, smp_processor_id());
 	print_cpu_stall_info_end();
 	for_each_possible_cpu(cpu)
-		totqlen += per_cpu_ptr(rsp->rda, cpu)->qlen;
+		totqlen += rcu_segcblist_n_cbs(&per_cpu_ptr(rsp->rda,
+							    cpu)->cblist);
 	pr_cont(" (t=%lu jiffies g=%ld c=%ld q=%lu)\n",
 		jiffies - rsp->gp_start,
 		(long)rsp->gpnum, (long)rsp->completed, totqlen);
@@ -1631,30 +1654,6 @@ void rcu_cpu_stall_reset(void)
 		WRITE_ONCE(rsp->jiffies_stall, jiffies + ULONG_MAX / 2);
 }
 
-/*
- * Initialize the specified rcu_data structure's default callback list
- * to empty.  The default callback list is the one that is not used by
- * no-callbacks CPUs.
- */
-static void init_default_callback_list(struct rcu_data *rdp)
-{
-	int i;
-
-	rdp->nxtlist = NULL;
-	for (i = 0; i < RCU_NEXT_SIZE; i++)
-		rdp->nxttail[i] = &rdp->nxtlist;
-}
-
-/*
- * Initialize the specified rcu_data structure's callback list to empty.
- */
-static void init_callback_list(struct rcu_data *rdp)
-{
-	if (init_nocb_callback_list(rdp))
-		return;
-	init_default_callback_list(rdp);
-}
-
 /*
  * Determine the value that ->completed will have at the end of the
  * next subsequent grace period.  This is used to tag callbacks so that
@@ -1709,7 +1708,6 @@ rcu_start_future_gp(struct rcu_node *rnp, struct rcu_data *rdp,
 		    unsigned long *c_out)
 {
 	unsigned long c;
-	int i;
 	bool ret = false;
 	struct rcu_node *rnp_root = rcu_get_root(rdp->rsp);
 
@@ -1755,13 +1753,11 @@ rcu_start_future_gp(struct rcu_node *rnp, struct rcu_data *rdp,
 	/*
 	 * Get a new grace-period number.  If there really is no grace
 	 * period in progress, it will be smaller than the one we obtained
-	 * earlier.  Adjust callbacks as needed.  Note that even no-CBs
-	 * CPUs have a ->nxtcompleted[] array, so no no-CBs checks needed.
+	 * earlier.  Adjust callbacks as needed.
 	 */
 	c = rcu_cbs_completed(rdp->rsp, rnp_root);
-	for (i = RCU_DONE_TAIL; i < RCU_NEXT_TAIL; i++)
-		if (ULONG_CMP_LT(c, rdp->nxtcompleted[i]))
-			rdp->nxtcompleted[i] = c;
+	if (!rcu_is_nocb_cpu(rdp->cpu))
+		(void)rcu_segcblist_accelerate(&rdp->cblist, c);
 
 	/*
 	 * If the needed for the required grace period is already
@@ -1793,9 +1789,7 @@ out:
 
 /*
  * Clean up any old requests for the just-ended grace period.  Also return
- * whether any additional grace periods have been requested.  Also invoke
- * rcu_nocb_gp_cleanup() in order to wake up any no-callbacks kthreads
- * waiting for this grace period to complete.
+ * whether any additional grace periods have been requested.
  */
 static int rcu_future_gp_cleanup(struct rcu_state *rsp, struct rcu_node *rnp)
 {
@@ -1841,57 +1835,27 @@ static void rcu_gp_kthread_wake(struct rcu_state *rsp)
 static bool rcu_accelerate_cbs(struct rcu_state *rsp, struct rcu_node *rnp,
 			       struct rcu_data *rdp)
 {
-	unsigned long c;
-	int i;
-	bool ret;
-
-	/* If the CPU has no callbacks, nothing to do. */
-	if (!rdp->nxttail[RCU_NEXT_TAIL] || !*rdp->nxttail[RCU_DONE_TAIL])
-		return false;
-
-	/*
-	 * Starting from the sublist containing the callbacks most
-	 * recently assigned a ->completed number and working down, find the
-	 * first sublist that is not assignable to an upcoming grace period.
-	 * Such a sublist has something in it (first two tests) and has
-	 * a ->completed number assigned that will complete sooner than
-	 * the ->completed number for newly arrived callbacks (last test).
-	 *
-	 * The key point is that any later sublist can be assigned the
-	 * same ->completed number as the newly arrived callbacks, which
-	 * means that the callbacks in any of these later sublist can be
-	 * grouped into a single sublist, whether or not they have already
-	 * been assigned a ->completed number.
-	 */
-	c = rcu_cbs_completed(rsp, rnp);
-	for (i = RCU_NEXT_TAIL - 1; i > RCU_DONE_TAIL; i--)
-		if (rdp->nxttail[i] != rdp->nxttail[i - 1] &&
-		    !ULONG_CMP_GE(rdp->nxtcompleted[i], c))
-			break;
+	bool ret = false;
 
-	/*
-	 * If there are no sublist for unassigned callbacks, leave.
-	 * At the same time, advance "i" one sublist, so that "i" will
-	 * index into the sublist where all the remaining callbacks should
-	 * be grouped into.
-	 */
-	if (++i >= RCU_NEXT_TAIL)
+	/* If no pending (not yet ready to invoke) callbacks, nothing to do. */
+	if (!rcu_segcblist_pend_cbs(&rdp->cblist))
 		return false;
 
 	/*
-	 * Assign all subsequent callbacks' ->completed number to the next
-	 * full grace period and group them all in the sublist initially
-	 * indexed by "i".
+	 * Callbacks are often registered with incomplete grace-period
+	 * information.  Something about the fact that getting exact
+	 * information requires acquiring a global lock...  RCU therefore
+	 * makes a conservative estimate of the grace period number at which
+	 * a given callback will become ready to invoke.	The following
+	 * code checks this estimate and improves it when possible, thus
+	 * accelerating callback invocation to an earlier grace-period
+	 * number.
 	 */
-	for (; i <= RCU_NEXT_TAIL; i++) {
-		rdp->nxttail[i] = rdp->nxttail[RCU_NEXT_TAIL];
-		rdp->nxtcompleted[i] = c;
-	}
-	/* Record any needed additional grace periods. */
-	ret = rcu_start_future_gp(rnp, rdp, NULL);
+	if (rcu_segcblist_accelerate(&rdp->cblist, rcu_cbs_completed(rsp, rnp)))
+		ret = rcu_start_future_gp(rnp, rdp, NULL);
 
 	/* Trace depending on how much we were able to accelerate. */
-	if (!*rdp->nxttail[RCU_WAIT_TAIL])
+	if (rcu_segcblist_restempty(&rdp->cblist, RCU_WAIT_TAIL))
 		trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("AccWaitCB"));
 	else
 		trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("AccReadyCB"));
@@ -1911,32 +1875,15 @@ static bool rcu_accelerate_cbs(struct rcu_state *rsp, struct rcu_node *rnp,
 static bool rcu_advance_cbs(struct rcu_state *rsp, struct rcu_node *rnp,
 			    struct rcu_data *rdp)
 {
-	int i, j;
-
-	/* If the CPU has no callbacks, nothing to do. */
-	if (!rdp->nxttail[RCU_NEXT_TAIL] || !*rdp->nxttail[RCU_DONE_TAIL])
+	/* If no pending (not yet ready to invoke) callbacks, nothing to do. */
+	if (!rcu_segcblist_pend_cbs(&rdp->cblist))
 		return false;
 
 	/*
 	 * Find all callbacks whose ->completed numbers indicate that they
 	 * are ready to invoke, and put them into the RCU_DONE_TAIL sublist.
 	 */
-	for (i = RCU_WAIT_TAIL; i < RCU_NEXT_TAIL; i++) {
-		if (ULONG_CMP_LT(rnp->completed, rdp->nxtcompleted[i]))
-			break;
-		rdp->nxttail[RCU_DONE_TAIL] = rdp->nxttail[i];
-	}
-	/* Clean up any sublist tail pointers that were misordered above. */
-	for (j = RCU_WAIT_TAIL; j < i; j++)
-		rdp->nxttail[j] = rdp->nxttail[RCU_DONE_TAIL];
-
-	/* Copy down callbacks to fill in empty sublists. */
-	for (j = RCU_WAIT_TAIL; i < RCU_NEXT_TAIL; i++, j++) {
-		if (rdp->nxttail[j] == rdp->nxttail[RCU_NEXT_TAIL])
-			break;
-		rdp->nxttail[j] = rdp->nxttail[i];
-		rdp->nxtcompleted[j] = rdp->nxtcompleted[i];
-	}
+	rcu_segcblist_advance(&rdp->cblist, rnp->completed);
 
 	/* Classify any remaining callbacks. */
 	return rcu_accelerate_cbs(rsp, rnp, rdp);
@@ -1981,7 +1928,7 @@ static bool __note_gp_changes(struct rcu_state *rsp, struct rcu_node *rnp,
 		trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpustart"));
 		need_gp = !!(rnp->qsmask & rdp->grpmask);
 		rdp->cpu_no_qs.b.norm = need_gp;
-		rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_qs_ctr);
+		rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_dynticks.rcu_qs_ctr);
 		rdp->core_needs_qs = need_gp;
 		zero_cpu_stall_ticks(rdp);
 		WRITE_ONCE(rdp->gpwrap, false);
@@ -2579,7 +2526,7 @@ rcu_report_qs_rdp(int cpu, struct rcu_state *rsp, struct rcu_data *rdp)
 		 * within the current grace period.
 		 */
 		rdp->cpu_no_qs.b.norm = true;	/* need qs for new gp. */
-		rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_qs_ctr);
+		rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_dynticks.rcu_qs_ctr);
 		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 		return;
 	}
@@ -2653,13 +2600,8 @@ rcu_send_cbs_to_orphanage(int cpu, struct rcu_state *rsp,
 	 * because _rcu_barrier() excludes CPU-hotplug operations, so it
 	 * cannot be running now.  Thus no memory barrier is required.
 	 */
-	if (rdp->nxtlist != NULL) {
-		rsp->qlen_lazy += rdp->qlen_lazy;
-		rsp->qlen += rdp->qlen;
-		rdp->n_cbs_orphaned += rdp->qlen;
-		rdp->qlen_lazy = 0;
-		WRITE_ONCE(rdp->qlen, 0);
-	}
+	rdp->n_cbs_orphaned += rcu_segcblist_n_cbs(&rdp->cblist);
+	rcu_segcblist_extract_count(&rdp->cblist, &rsp->orphan_done);
 
 	/*
 	 * Next, move those callbacks still needing a grace period to
@@ -2667,31 +2609,18 @@ rcu_send_cbs_to_orphanage(int cpu, struct rcu_state *rsp,
 	 * Some of the callbacks might have gone partway through a grace
 	 * period, but that is too bad.  They get to start over because we
 	 * cannot assume that grace periods are synchronized across CPUs.
-	 * We don't bother updating the ->nxttail[] array yet, instead
-	 * we just reset the whole thing later on.
 	 */
-	if (*rdp->nxttail[RCU_DONE_TAIL] != NULL) {
-		*rsp->orphan_nxttail = *rdp->nxttail[RCU_DONE_TAIL];
-		rsp->orphan_nxttail = rdp->nxttail[RCU_NEXT_TAIL];
-		*rdp->nxttail[RCU_DONE_TAIL] = NULL;
-	}
+	rcu_segcblist_extract_pend_cbs(&rdp->cblist, &rsp->orphan_pend);
 
 	/*
 	 * Then move the ready-to-invoke callbacks to the orphanage,
 	 * where some other CPU will pick them up.  These will not be
 	 * required to pass though another grace period: They are done.
 	 */
-	if (rdp->nxtlist != NULL) {
-		*rsp->orphan_donetail = rdp->nxtlist;
-		rsp->orphan_donetail = rdp->nxttail[RCU_DONE_TAIL];
-	}
+	rcu_segcblist_extract_done_cbs(&rdp->cblist, &rsp->orphan_done);
 
-	/*
-	 * Finally, initialize the rcu_data structure's list to empty and
-	 * disallow further callbacks on this CPU.
-	 */
-	init_callback_list(rdp);
-	rdp->nxttail[RCU_NEXT_TAIL] = NULL;
+	/* Finally, disallow further callbacks on this CPU.  */
+	rcu_segcblist_disable(&rdp->cblist);
 }
 
 /*
@@ -2700,7 +2629,6 @@ rcu_send_cbs_to_orphanage(int cpu, struct rcu_state *rsp,
  */
 static void rcu_adopt_orphan_cbs(struct rcu_state *rsp, unsigned long flags)
 {
-	int i;
 	struct rcu_data *rdp = raw_cpu_ptr(rsp->rda);
 
 	/* No-CBs CPUs are handled specially. */
@@ -2709,13 +2637,11 @@ static void rcu_adopt_orphan_cbs(struct rcu_state *rsp, unsigned long flags)
 		return;
 
 	/* Do the accounting first. */
-	rdp->qlen_lazy += rsp->qlen_lazy;
-	rdp->qlen += rsp->qlen;
-	rdp->n_cbs_adopted += rsp->qlen;
-	if (rsp->qlen_lazy != rsp->qlen)
+	rdp->n_cbs_adopted += rcu_cblist_n_cbs(&rsp->orphan_done);
+	if (rcu_cblist_n_lazy_cbs(&rsp->orphan_done) !=
+	    rcu_cblist_n_cbs(&rsp->orphan_done))
 		rcu_idle_count_callbacks_posted();
-	rsp->qlen_lazy = 0;
-	rsp->qlen = 0;
+	rcu_segcblist_insert_count(&rdp->cblist, &rsp->orphan_done);
 
 	/*
 	 * We do not need a memory barrier here because the only way we
@@ -2723,24 +2649,13 @@ static void rcu_adopt_orphan_cbs(struct rcu_state *rsp, unsigned long flags)
 	 * we are the task doing the rcu_barrier().
 	 */
 
-	/* First adopt the ready-to-invoke callbacks. */
-	if (rsp->orphan_donelist != NULL) {
-		*rsp->orphan_donetail = *rdp->nxttail[RCU_DONE_TAIL];
-		*rdp->nxttail[RCU_DONE_TAIL] = rsp->orphan_donelist;
-		for (i = RCU_NEXT_SIZE - 1; i >= RCU_DONE_TAIL; i--)
-			if (rdp->nxttail[i] == rdp->nxttail[RCU_DONE_TAIL])
-				rdp->nxttail[i] = rsp->orphan_donetail;
-		rsp->orphan_donelist = NULL;
-		rsp->orphan_donetail = &rsp->orphan_donelist;
-	}
-
-	/* And then adopt the callbacks that still need a grace period. */
-	if (rsp->orphan_nxtlist != NULL) {
-		*rdp->nxttail[RCU_NEXT_TAIL] = rsp->orphan_nxtlist;
-		rdp->nxttail[RCU_NEXT_TAIL] = rsp->orphan_nxttail;
-		rsp->orphan_nxtlist = NULL;
-		rsp->orphan_nxttail = &rsp->orphan_nxtlist;
-	}
+	/* First adopt the ready-to-invoke callbacks, then the done ones. */
+	rcu_segcblist_insert_done_cbs(&rdp->cblist, &rsp->orphan_done);
+	WARN_ON_ONCE(!rcu_cblist_empty(&rsp->orphan_done));
+	rcu_segcblist_insert_pend_cbs(&rdp->cblist, &rsp->orphan_pend);
+	WARN_ON_ONCE(!rcu_cblist_empty(&rsp->orphan_pend));
+	WARN_ON_ONCE(rcu_segcblist_empty(&rdp->cblist) !=
+		     !rcu_segcblist_n_cbs(&rdp->cblist));
 }
 
 /*
@@ -2748,14 +2663,14 @@ static void rcu_adopt_orphan_cbs(struct rcu_state *rsp, unsigned long flags)
  */
 static void rcu_cleanup_dying_cpu(struct rcu_state *rsp)
 {
-	RCU_TRACE(unsigned long mask);
-	RCU_TRACE(struct rcu_data *rdp = this_cpu_ptr(rsp->rda));
-	RCU_TRACE(struct rcu_node *rnp = rdp->mynode);
+	RCU_TRACE(unsigned long mask;)
+	RCU_TRACE(struct rcu_data *rdp = this_cpu_ptr(rsp->rda);)
+	RCU_TRACE(struct rcu_node *rnp = rdp->mynode;)
 
 	if (!IS_ENABLED(CONFIG_HOTPLUG_CPU))
 		return;
 
-	RCU_TRACE(mask = rdp->grpmask);
+	RCU_TRACE(mask = rdp->grpmask;)
 	trace_rcu_grace_period(rsp->name,
 			       rnp->gpnum + 1 - !!(rnp->qsmask & mask),
 			       TPS("cpuofl"));
@@ -2828,9 +2743,11 @@ static void rcu_cleanup_dead_cpu(int cpu, struct rcu_state *rsp)
 	rcu_adopt_orphan_cbs(rsp, flags);
 	raw_spin_unlock_irqrestore(&rsp->orphan_lock, flags);
 
-	WARN_ONCE(rdp->qlen != 0 || rdp->nxtlist != NULL,
-		  "rcu_cleanup_dead_cpu: Callbacks on offline CPU %d: qlen=%lu, nxtlist=%p\n",
-		  cpu, rdp->qlen, rdp->nxtlist);
+	WARN_ONCE(rcu_segcblist_n_cbs(&rdp->cblist) != 0 ||
+		  !rcu_segcblist_empty(&rdp->cblist),
+		  "rcu_cleanup_dead_cpu: Callbacks on offline CPU %d: qlen=%lu, 1stCB=%p\n",
+		  cpu, rcu_segcblist_n_cbs(&rdp->cblist),
+		  rcu_segcblist_first_cb(&rdp->cblist));
 }
 
 /*
@@ -2840,14 +2757,17 @@ static void rcu_cleanup_dead_cpu(int cpu, struct rcu_state *rsp)
 static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp)
 {
 	unsigned long flags;
-	struct rcu_head *next, *list, **tail;
-	long bl, count, count_lazy;
-	int i;
+	struct rcu_head *rhp;
+	struct rcu_cblist rcl = RCU_CBLIST_INITIALIZER(rcl);
+	long bl, count;
 
 	/* If no callbacks are ready, just return. */
-	if (!cpu_has_callbacks_ready_to_invoke(rdp)) {
-		trace_rcu_batch_start(rsp->name, rdp->qlen_lazy, rdp->qlen, 0);
-		trace_rcu_batch_end(rsp->name, 0, !!READ_ONCE(rdp->nxtlist),
+	if (!rcu_segcblist_ready_cbs(&rdp->cblist)) {
+		trace_rcu_batch_start(rsp->name,
+				      rcu_segcblist_n_lazy_cbs(&rdp->cblist),
+				      rcu_segcblist_n_cbs(&rdp->cblist), 0);
+		trace_rcu_batch_end(rsp->name, 0,
+				    !rcu_segcblist_empty(&rdp->cblist),
 				    need_resched(), is_idle_task(current),
 				    rcu_is_callbacks_kthread());
 		return;
@@ -2855,73 +2775,62 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp)
 
 	/*
 	 * Extract the list of ready callbacks, disabling to prevent
-	 * races with call_rcu() from interrupt handlers.
+	 * races with call_rcu() from interrupt handlers.  Leave the
+	 * callback counts, as rcu_barrier() needs to be conservative.
 	 */
 	local_irq_save(flags);
 	WARN_ON_ONCE(cpu_is_offline(smp_processor_id()));
 	bl = rdp->blimit;
-	trace_rcu_batch_start(rsp->name, rdp->qlen_lazy, rdp->qlen, bl);
-	list = rdp->nxtlist;
-	rdp->nxtlist = *rdp->nxttail[RCU_DONE_TAIL];
-	*rdp->nxttail[RCU_DONE_TAIL] = NULL;
-	tail = rdp->nxttail[RCU_DONE_TAIL];
-	for (i = RCU_NEXT_SIZE - 1; i >= 0; i--)
-		if (rdp->nxttail[i] == rdp->nxttail[RCU_DONE_TAIL])
-			rdp->nxttail[i] = &rdp->nxtlist;
+	trace_rcu_batch_start(rsp->name, rcu_segcblist_n_lazy_cbs(&rdp->cblist),
+			      rcu_segcblist_n_cbs(&rdp->cblist), bl);
+	rcu_segcblist_extract_done_cbs(&rdp->cblist, &rcl);
 	local_irq_restore(flags);
 
 	/* Invoke callbacks. */
-	count = count_lazy = 0;
-	while (list) {
-		next = list->next;
-		prefetch(next);
-		debug_rcu_head_unqueue(list);
-		if (__rcu_reclaim(rsp->name, list))
-			count_lazy++;
-		list = next;
-		/* Stop only if limit reached and CPU has something to do. */
-		if (++count >= bl &&
+	rhp = rcu_cblist_dequeue(&rcl);
+	for (; rhp; rhp = rcu_cblist_dequeue(&rcl)) {
+		debug_rcu_head_unqueue(rhp);
+		if (__rcu_reclaim(rsp->name, rhp))
+			rcu_cblist_dequeued_lazy(&rcl);
+		/*
+		 * Stop only if limit reached and CPU has something to do.
+		 * Note: The rcl structure counts down from zero.
+		 */
+		if (-rcu_cblist_n_cbs(&rcl) >= bl &&
 		    (need_resched() ||
 		     (!is_idle_task(current) && !rcu_is_callbacks_kthread())))
 			break;
 	}
 
 	local_irq_save(flags);
-	trace_rcu_batch_end(rsp->name, count, !!list, need_resched(),
-			    is_idle_task(current),
+	count = -rcu_cblist_n_cbs(&rcl);
+	trace_rcu_batch_end(rsp->name, count, !rcu_cblist_empty(&rcl),
+			    need_resched(), is_idle_task(current),
 			    rcu_is_callbacks_kthread());
 
-	/* Update count, and requeue any remaining callbacks. */
-	if (list != NULL) {
-		*tail = rdp->nxtlist;
-		rdp->nxtlist = list;
-		for (i = 0; i < RCU_NEXT_SIZE; i++)
-			if (&rdp->nxtlist == rdp->nxttail[i])
-				rdp->nxttail[i] = tail;
-			else
-				break;
-	}
+	/* Update counts and requeue any remaining callbacks. */
+	rcu_segcblist_insert_done_cbs(&rdp->cblist, &rcl);
 	smp_mb(); /* List handling before counting for rcu_barrier(). */
-	rdp->qlen_lazy -= count_lazy;
-	WRITE_ONCE(rdp->qlen, rdp->qlen - count);
 	rdp->n_cbs_invoked += count;
+	rcu_segcblist_insert_count(&rdp->cblist, &rcl);
 
 	/* Reinstate batch limit if we have worked down the excess. */
-	if (rdp->blimit == LONG_MAX && rdp->qlen <= qlowmark)
+	count = rcu_segcblist_n_cbs(&rdp->cblist);
+	if (rdp->blimit == LONG_MAX && count <= qlowmark)
 		rdp->blimit = blimit;
 
 	/* Reset ->qlen_last_fqs_check trigger if enough CBs have drained. */
-	if (rdp->qlen == 0 && rdp->qlen_last_fqs_check != 0) {
+	if (count == 0 && rdp->qlen_last_fqs_check != 0) {
 		rdp->qlen_last_fqs_check = 0;
 		rdp->n_force_qs_snap = rsp->n_force_qs;
-	} else if (rdp->qlen < rdp->qlen_last_fqs_check - qhimark)
-		rdp->qlen_last_fqs_check = rdp->qlen;
-	WARN_ON_ONCE((rdp->nxtlist == NULL) != (rdp->qlen == 0));
+	} else if (count < rdp->qlen_last_fqs_check - qhimark)
+		rdp->qlen_last_fqs_check = count;
+	WARN_ON_ONCE(rcu_segcblist_empty(&rdp->cblist) != (count == 0));
 
 	local_irq_restore(flags);
 
 	/* Re-invoke RCU core processing if there are callbacks remaining. */
-	if (cpu_has_callbacks_ready_to_invoke(rdp))
+	if (rcu_segcblist_ready_cbs(&rdp->cblist))
 		invoke_rcu_core();
 }
 
@@ -3087,7 +2996,7 @@ __rcu_process_callbacks(struct rcu_state *rsp)
 	bool needwake;
 	struct rcu_data *rdp = raw_cpu_ptr(rsp->rda);
 
-	WARN_ON_ONCE(rdp->beenonline == 0);
+	WARN_ON_ONCE(!rdp->beenonline);
 
 	/* Update RCU state based on any recent quiescent states. */
 	rcu_check_quiescent_state(rsp, rdp);
@@ -3105,7 +3014,7 @@ __rcu_process_callbacks(struct rcu_state *rsp)
 	}
 
 	/* If there are callbacks ready, invoke them. */
-	if (cpu_has_callbacks_ready_to_invoke(rdp))
+	if (rcu_segcblist_ready_cbs(&rdp->cblist))
 		invoke_rcu_callbacks(rsp, rdp);
 
 	/* Do any needed deferred wakeups of rcuo kthreads. */
@@ -3177,7 +3086,8 @@ static void __call_rcu_core(struct rcu_state *rsp, struct rcu_data *rdp,
 	 * invoking force_quiescent_state() if the newly enqueued callback
 	 * is the only one waiting for a grace period to complete.
 	 */
-	if (unlikely(rdp->qlen > rdp->qlen_last_fqs_check + qhimark)) {
+	if (unlikely(rcu_segcblist_n_cbs(&rdp->cblist) >
+		     rdp->qlen_last_fqs_check + qhimark)) {
 
 		/* Are we ignoring a completed grace period? */
 		note_gp_changes(rsp, rdp);
@@ -3195,10 +3105,10 @@ static void __call_rcu_core(struct rcu_state *rsp, struct rcu_data *rdp,
 			/* Give the grace period a kick. */
 			rdp->blimit = LONG_MAX;
 			if (rsp->n_force_qs == rdp->n_force_qs_snap &&
-			    *rdp->nxttail[RCU_DONE_TAIL] != head)
+			    rcu_segcblist_first_pend_cb(&rdp->cblist) != head)
 				force_quiescent_state(rsp);
 			rdp->n_force_qs_snap = rsp->n_force_qs;
-			rdp->qlen_last_fqs_check = rdp->qlen;
+			rdp->qlen_last_fqs_check = rcu_segcblist_n_cbs(&rdp->cblist);
 		}
 	}
 }
@@ -3238,7 +3148,7 @@ __call_rcu(struct rcu_head *head, rcu_callback_t func,
 	rdp = this_cpu_ptr(rsp->rda);
 
 	/* Add the callback to our list. */
-	if (unlikely(rdp->nxttail[RCU_NEXT_TAIL] == NULL) || cpu != -1) {
+	if (unlikely(!rcu_segcblist_is_enabled(&rdp->cblist)) || cpu != -1) {
 		int offline;
 
 		if (cpu != -1)
@@ -3257,23 +3167,21 @@ __call_rcu(struct rcu_head *head, rcu_callback_t func,
 		 */
 		BUG_ON(cpu != -1);
 		WARN_ON_ONCE(!rcu_is_watching());
-		if (!likely(rdp->nxtlist))
-			init_default_callback_list(rdp);
+		if (rcu_segcblist_empty(&rdp->cblist))
+			rcu_segcblist_init(&rdp->cblist);
 	}
-	WRITE_ONCE(rdp->qlen, rdp->qlen + 1);
-	if (lazy)
-		rdp->qlen_lazy++;
-	else
+	rcu_segcblist_enqueue(&rdp->cblist, head, lazy);
+	if (!lazy)
 		rcu_idle_count_callbacks_posted();
-	smp_mb();  /* Count before adding callback for rcu_barrier(). */
-	*rdp->nxttail[RCU_NEXT_TAIL] = head;
-	rdp->nxttail[RCU_NEXT_TAIL] = &head->next;
 
 	if (__is_kfree_rcu_offset((unsigned long)func))
 		trace_rcu_kfree_callback(rsp->name, head, (unsigned long)func,
-					 rdp->qlen_lazy, rdp->qlen);
+					 rcu_segcblist_n_lazy_cbs(&rdp->cblist),
+					 rcu_segcblist_n_cbs(&rdp->cblist));
 	else
-		trace_rcu_callback(rsp->name, head, rdp->qlen_lazy, rdp->qlen);
+		trace_rcu_callback(rsp->name, head,
+				   rcu_segcblist_n_lazy_cbs(&rdp->cblist),
+				   rcu_segcblist_n_cbs(&rdp->cblist));
 
 	/* Go handle any RCU core processing required. */
 	__call_rcu_core(rsp, rdp, head, flags);
@@ -3519,41 +3427,6 @@ void cond_synchronize_sched(unsigned long oldstate)
 }
 EXPORT_SYMBOL_GPL(cond_synchronize_sched);
 
-/* Adjust sequence number for start of update-side operation. */
-static void rcu_seq_start(unsigned long *sp)
-{
-	WRITE_ONCE(*sp, *sp + 1);
-	smp_mb(); /* Ensure update-side operation after counter increment. */
-	WARN_ON_ONCE(!(*sp & 0x1));
-}
-
-/* Adjust sequence number for end of update-side operation. */
-static void rcu_seq_end(unsigned long *sp)
-{
-	smp_mb(); /* Ensure update-side operation before counter increment. */
-	WRITE_ONCE(*sp, *sp + 1);
-	WARN_ON_ONCE(*sp & 0x1);
-}
-
-/* Take a snapshot of the update side's sequence number. */
-static unsigned long rcu_seq_snap(unsigned long *sp)
-{
-	unsigned long s;
-
-	s = (READ_ONCE(*sp) + 3) & ~0x1;
-	smp_mb(); /* Above access must not bleed into critical section. */
-	return s;
-}
-
-/*
- * Given a snapshot from rcu_seq_snap(), determine whether or not a
- * full update-side operation has occurred.
- */
-static bool rcu_seq_done(unsigned long *sp, unsigned long s)
-{
-	return ULONG_CMP_GE(READ_ONCE(*sp), s);
-}
-
 /*
  * 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.
@@ -3577,7 +3450,7 @@ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp)
 	/* Is the RCU core waiting for a quiescent state from this CPU? */
 	if (rcu_scheduler_fully_active &&
 	    rdp->core_needs_qs && rdp->cpu_no_qs.b.norm &&
-	    rdp->rcu_qs_ctr_snap == __this_cpu_read(rcu_qs_ctr)) {
+	    rdp->rcu_qs_ctr_snap == __this_cpu_read(rcu_dynticks.rcu_qs_ctr)) {
 		rdp->n_rp_core_needs_qs++;
 	} else if (rdp->core_needs_qs && !rdp->cpu_no_qs.b.norm) {
 		rdp->n_rp_report_qs++;
@@ -3585,7 +3458,7 @@ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp)
 	}
 
 	/* Does this CPU have callbacks ready to invoke? */
-	if (cpu_has_callbacks_ready_to_invoke(rdp)) {
+	if (rcu_segcblist_ready_cbs(&rdp->cblist)) {
 		rdp->n_rp_cb_ready++;
 		return 1;
 	}
@@ -3649,10 +3522,10 @@ static bool __maybe_unused rcu_cpu_has_callbacks(bool *all_lazy)
 
 	for_each_rcu_flavor(rsp) {
 		rdp = this_cpu_ptr(rsp->rda);
-		if (!rdp->nxtlist)
+		if (rcu_segcblist_empty(&rdp->cblist))
 			continue;
 		hc = true;
-		if (rdp->qlen != rdp->qlen_lazy || !all_lazy) {
+		if (rcu_segcblist_n_nonlazy_cbs(&rdp->cblist) || !all_lazy) {
 			al = false;
 			break;
 		}
@@ -3761,7 +3634,7 @@ static void _rcu_barrier(struct rcu_state *rsp)
 				__call_rcu(&rdp->barrier_head,
 					   rcu_barrier_callback, rsp, cpu, 0);
 			}
-		} else if (READ_ONCE(rdp->qlen)) {
+		} else if (rcu_segcblist_n_cbs(&rdp->cblist)) {
 			_rcu_barrier_trace(rsp, "OnlineQ", cpu,
 					   rsp->barrier_sequence);
 			smp_call_function_single(cpu, rcu_barrier_func, rsp, 1);
@@ -3870,8 +3743,9 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp)
 	rdp->qlen_last_fqs_check = 0;
 	rdp->n_force_qs_snap = rsp->n_force_qs;
 	rdp->blimit = blimit;
-	if (!rdp->nxtlist)
-		init_callback_list(rdp);  /* Re-enable callbacks on this CPU. */
+	if (rcu_segcblist_empty(&rdp->cblist) && /* No early-boot CBs? */
+	    !init_nocb_callback_list(rdp))
+		rcu_segcblist_init(&rdp->cblist);  /* Re-enable callbacks. */
 	rdp->dynticks->dynticks_nesting = DYNTICK_TASK_EXIT_IDLE;
 	rcu_sysidle_init_percpu_data(rdp->dynticks);
 	rcu_dynticks_eqs_online();
@@ -3890,12 +3764,16 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp)
 	rdp->gpnum = rnp->completed; /* Make CPU later note any new GP. */
 	rdp->completed = rnp->completed;
 	rdp->cpu_no_qs.b.norm = true;
-	rdp->rcu_qs_ctr_snap = per_cpu(rcu_qs_ctr, cpu);
+	rdp->rcu_qs_ctr_snap = per_cpu(rcu_dynticks.rcu_qs_ctr, cpu);
 	rdp->core_needs_qs = false;
 	trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpuonl"));
 	raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 }
 
+/*
+ * Invoked early in the CPU-online process, when pretty much all
+ * services are available.  The incoming CPU is not present.
+ */
 int rcutree_prepare_cpu(unsigned int cpu)
 {
 	struct rcu_state *rsp;
@@ -3909,6 +3787,9 @@ int rcutree_prepare_cpu(unsigned int cpu)
 	return 0;
 }
 
+/*
+ * Update RCU priority boot kthread affinity for CPU-hotplug changes.
+ */
 static void rcutree_affinity_setting(unsigned int cpu, int outgoing)
 {
 	struct rcu_data *rdp = per_cpu_ptr(rcu_state_p->rda, cpu);
@@ -3916,20 +3797,34 @@ static void rcutree_affinity_setting(unsigned int cpu, int outgoing)
 	rcu_boost_kthread_setaffinity(rdp->mynode, outgoing);
 }
 
+/*
+ * Near the end of the CPU-online process.  Pretty much all services
+ * enabled, and the CPU is now very much alive.
+ */
 int rcutree_online_cpu(unsigned int cpu)
 {
 	sync_sched_exp_online_cleanup(cpu);
 	rcutree_affinity_setting(cpu, -1);
+	if (IS_ENABLED(CONFIG_TREE_SRCU))
+		srcu_online_cpu(cpu);
 	return 0;
 }
 
+/*
+ * Near the beginning of the process.  The CPU is still very much alive
+ * with pretty much all services enabled.
+ */
 int rcutree_offline_cpu(unsigned int cpu)
 {
 	rcutree_affinity_setting(cpu, cpu);
+	if (IS_ENABLED(CONFIG_TREE_SRCU))
+		srcu_offline_cpu(cpu);
 	return 0;
 }
 
-
+/*
+ * Near the end of the offline process.  We do only tracing here.
+ */
 int rcutree_dying_cpu(unsigned int cpu)
 {
 	struct rcu_state *rsp;
@@ -3939,6 +3834,9 @@ int rcutree_dying_cpu(unsigned int cpu)
 	return 0;
 }
 
+/*
+ * The outgoing CPU is gone and we are running elsewhere.
+ */
 int rcutree_dead_cpu(unsigned int cpu)
 {
 	struct rcu_state *rsp;
@@ -3956,6 +3854,10 @@ int rcutree_dead_cpu(unsigned int cpu)
  * incoming CPUs are not allowed to use RCU read-side critical sections
  * until this function is called.  Failing to observe this restriction
  * will result in lockdep splats.
+ *
+ * Note that this function is special in that it is invoked directly
+ * from the incoming CPU rather than from the cpuhp_step mechanism.
+ * This is because this function must be invoked at a precise location.
  */
 void rcu_cpu_starting(unsigned int cpu)
 {
@@ -3978,9 +3880,6 @@ void rcu_cpu_starting(unsigned int cpu)
 
 #ifdef CONFIG_HOTPLUG_CPU
 /*
- * The CPU is exiting the idle loop into the arch_cpu_idle_dead()
- * function.  We now remove it from the rcu_node tree's ->qsmaskinit
- * bit masks.
  * The CPU is exiting the idle loop into the arch_cpu_idle_dead()
  * function.  We now remove it from the rcu_node tree's ->qsmaskinit
  * bit masks.
@@ -3999,6 +3898,14 @@ static void rcu_cleanup_dying_idle_cpu(int cpu, struct rcu_state *rsp)
 	raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 }
 
+/*
+ * The outgoing function has no further need of RCU, so remove it from
+ * the list of CPUs that RCU must track.
+ *
+ * Note that this function is special in that it is invoked directly
+ * from the outgoing CPU rather than from the cpuhp_step mechanism.
+ * This is because this function must be invoked at a precise location.
+ */
 void rcu_report_dead(unsigned int cpu)
 {
 	struct rcu_state *rsp;
@@ -4013,6 +3920,10 @@ void rcu_report_dead(unsigned int cpu)
 }
 #endif
 
+/*
+ * On non-huge systems, use expedited RCU grace periods to make suspend
+ * and hibernation run faster.
+ */
 static int rcu_pm_notify(struct notifier_block *self,
 			 unsigned long action, void *hcpu)
 {
@@ -4083,7 +3994,7 @@ early_initcall(rcu_spawn_gp_kthread);
  * task is booting the system, and such primitives are no-ops).  After this
  * function is called, any synchronous grace-period primitives are run as
  * expedited, with the requesting task driving the grace period forward.
- * A later core_initcall() rcu_exp_runtime_mode() will switch to full
+ * A later core_initcall() rcu_set_runtime_mode() will switch to full
  * runtime RCU functionality.
  */
 void rcu_scheduler_starting(void)
@@ -4095,31 +4006,6 @@ void rcu_scheduler_starting(void)
 	rcu_test_sync_prims();
 }
 
-/*
- * Compute the per-level fanout, either using the exact fanout specified
- * or balancing the tree, depending on the rcu_fanout_exact boot parameter.
- */
-static void __init rcu_init_levelspread(int *levelspread, const int *levelcnt)
-{
-	int i;
-
-	if (rcu_fanout_exact) {
-		levelspread[rcu_num_lvls - 1] = rcu_fanout_leaf;
-		for (i = rcu_num_lvls - 2; i >= 0; i--)
-			levelspread[i] = RCU_FANOUT;
-	} else {
-		int ccur;
-		int cprv;
-
-		cprv = nr_cpu_ids;
-		for (i = rcu_num_lvls - 1; i >= 0; i--) {
-			ccur = levelcnt[i];
-			levelspread[i] = (cprv + ccur - 1) / ccur;
-			cprv = ccur;
-		}
-	}
-}
-
 /*
  * Helper function for rcu_init() that initializes one rcu_state structure.
  */
@@ -4129,9 +4015,7 @@ static void __init rcu_init_one(struct rcu_state *rsp)
 	static const char * const fqs[] = RCU_FQS_NAME_INIT;
 	static struct lock_class_key rcu_node_class[RCU_NUM_LVLS];
 	static struct lock_class_key rcu_fqs_class[RCU_NUM_LVLS];
-	static u8 fl_mask = 0x1;
 
-	int levelcnt[RCU_NUM_LVLS];		/* # nodes in each level. */
 	int levelspread[RCU_NUM_LVLS];		/* kids/node in each level. */
 	int cpustride = 1;
 	int i;
@@ -4146,20 +4030,16 @@ static void __init rcu_init_one(struct rcu_state *rsp)
 
 	/* Initialize the level-tracking arrays. */
 
-	for (i = 0; i < rcu_num_lvls; i++)
-		levelcnt[i] = num_rcu_lvl[i];
 	for (i = 1; i < rcu_num_lvls; i++)
-		rsp->level[i] = rsp->level[i - 1] + levelcnt[i - 1];
-	rcu_init_levelspread(levelspread, levelcnt);
-	rsp->flavor_mask = fl_mask;
-	fl_mask <<= 1;
+		rsp->level[i] = rsp->level[i - 1] + num_rcu_lvl[i - 1];
+	rcu_init_levelspread(levelspread, num_rcu_lvl);
 
 	/* Initialize the elements themselves, starting from the leaves. */
 
 	for (i = rcu_num_lvls - 1; i >= 0; i--) {
 		cpustride *= levelspread[i];
 		rnp = rsp->level[i];
-		for (j = 0; j < levelcnt[i]; j++, rnp++) {
+		for (j = 0; j < num_rcu_lvl[i]; j++, rnp++) {
 			raw_spin_lock_init(&ACCESS_PRIVATE(rnp, lock));
 			lockdep_set_class_and_name(&ACCESS_PRIVATE(rnp, lock),
 						   &rcu_node_class[i], buf[i]);
@@ -4332,6 +4212,8 @@ void __init rcu_init(void)
 	for_each_online_cpu(cpu) {
 		rcutree_prepare_cpu(cpu);
 		rcu_cpu_starting(cpu);
+		if (IS_ENABLED(CONFIG_TREE_SRCU))
+			srcu_online_cpu(cpu);
 	}
 }
 

kernel/rcu/tree.h

@@ -30,80 +30,8 @@
 #include <linux/seqlock.h>
 #include <linux/swait.h>
 #include <linux/stop_machine.h>
-
-/*
- * Define shape of hierarchy based on NR_CPUS, CONFIG_RCU_FANOUT, and
- * CONFIG_RCU_FANOUT_LEAF.
- * In theory, it should be possible to add more levels straightforwardly.
- * In practice, this did work well going from three levels to four.
- * Of course, your mileage may vary.
- */
-
-#ifdef CONFIG_RCU_FANOUT
-#define RCU_FANOUT CONFIG_RCU_FANOUT
-#else /* #ifdef CONFIG_RCU_FANOUT */
-# ifdef CONFIG_64BIT
-# define RCU_FANOUT 64
-# else
-# define RCU_FANOUT 32
-# endif
-#endif /* #else #ifdef CONFIG_RCU_FANOUT */
-
-#ifdef CONFIG_RCU_FANOUT_LEAF
-#define RCU_FANOUT_LEAF CONFIG_RCU_FANOUT_LEAF
-#else /* #ifdef CONFIG_RCU_FANOUT_LEAF */
-# ifdef CONFIG_64BIT
-# define RCU_FANOUT_LEAF 64
-# else
-# define RCU_FANOUT_LEAF 32
-# endif
-#endif /* #else #ifdef CONFIG_RCU_FANOUT_LEAF */
-
-#define RCU_FANOUT_1	      (RCU_FANOUT_LEAF)
-#define RCU_FANOUT_2	      (RCU_FANOUT_1 * RCU_FANOUT)
-#define RCU_FANOUT_3	      (RCU_FANOUT_2 * RCU_FANOUT)
-#define RCU_FANOUT_4	      (RCU_FANOUT_3 * RCU_FANOUT)
-
-#if NR_CPUS <= RCU_FANOUT_1
-#  define RCU_NUM_LVLS	      1
-#  define NUM_RCU_LVL_0	      1
-#  define NUM_RCU_NODES	      NUM_RCU_LVL_0
-#  define NUM_RCU_LVL_INIT    { NUM_RCU_LVL_0 }
-#  define RCU_NODE_NAME_INIT  { "rcu_node_0" }
-#  define RCU_FQS_NAME_INIT   { "rcu_node_fqs_0" }
-#elif NR_CPUS <= RCU_FANOUT_2
-#  define RCU_NUM_LVLS	      2
-#  define NUM_RCU_LVL_0	      1
-#  define NUM_RCU_LVL_1	      DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_1)
-#  define NUM_RCU_NODES	      (NUM_RCU_LVL_0 + NUM_RCU_LVL_1)
-#  define NUM_RCU_LVL_INIT    { NUM_RCU_LVL_0, NUM_RCU_LVL_1 }
-#  define RCU_NODE_NAME_INIT  { "rcu_node_0", "rcu_node_1" }
-#  define RCU_FQS_NAME_INIT   { "rcu_node_fqs_0", "rcu_node_fqs_1" }
-#elif NR_CPUS <= RCU_FANOUT_3
-#  define RCU_NUM_LVLS	      3
-#  define NUM_RCU_LVL_0	      1
-#  define NUM_RCU_LVL_1	      DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_2)
-#  define NUM_RCU_LVL_2	      DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_1)
-#  define NUM_RCU_NODES	      (NUM_RCU_LVL_0 + NUM_RCU_LVL_1 + NUM_RCU_LVL_2)
-#  define NUM_RCU_LVL_INIT    { NUM_RCU_LVL_0, NUM_RCU_LVL_1, NUM_RCU_LVL_2 }
-#  define RCU_NODE_NAME_INIT  { "rcu_node_0", "rcu_node_1", "rcu_node_2" }
-#  define RCU_FQS_NAME_INIT   { "rcu_node_fqs_0", "rcu_node_fqs_1", "rcu_node_fqs_2" }
-#elif NR_CPUS <= RCU_FANOUT_4
-#  define RCU_NUM_LVLS	      4
-#  define NUM_RCU_LVL_0	      1
-#  define NUM_RCU_LVL_1	      DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_3)
-#  define NUM_RCU_LVL_2	      DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_2)
-#  define NUM_RCU_LVL_3	      DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_1)
-#  define NUM_RCU_NODES	      (NUM_RCU_LVL_0 + NUM_RCU_LVL_1 + NUM_RCU_LVL_2 + NUM_RCU_LVL_3)
-#  define NUM_RCU_LVL_INIT    { NUM_RCU_LVL_0, NUM_RCU_LVL_1, NUM_RCU_LVL_2, NUM_RCU_LVL_3 }
-#  define RCU_NODE_NAME_INIT  { "rcu_node_0", "rcu_node_1", "rcu_node_2", "rcu_node_3" }
-#  define RCU_FQS_NAME_INIT   { "rcu_node_fqs_0", "rcu_node_fqs_1", "rcu_node_fqs_2", "rcu_node_fqs_3" }
-#else
-# error "CONFIG_RCU_FANOUT insufficient for NR_CPUS"
-#endif /* #if (NR_CPUS) <= RCU_FANOUT_1 */
-
-extern int rcu_num_lvls;
-extern int rcu_num_nodes;
+#include <linux/rcu_segcblist.h>
+#include <linux/rcu_node_tree.h>
 
 /*
  * Dynticks per-CPU state.
@@ -113,6 +41,9 @@ struct rcu_dynticks {
 				    /* Process level is worth LLONG_MAX/2. */
 	int dynticks_nmi_nesting;   /* Track NMI nesting level. */
 	atomic_t dynticks;	    /* Even value for idle, else odd. */
+	bool rcu_need_heavy_qs;     /* GP old, need heavy quiescent state. */
+	unsigned long rcu_qs_ctr;   /* Light universal quiescent state ctr. */
+	bool rcu_urgent_qs;	    /* GP old need light quiescent state. */
 #ifdef CONFIG_NO_HZ_FULL_SYSIDLE
 	long long dynticks_idle_nesting;
 				    /* irq/process nesting level from idle. */
@@ -261,41 +192,6 @@ struct rcu_node {
  */
 #define leaf_node_cpu_bit(rnp, cpu) (1UL << ((cpu) - (rnp)->grplo))
 
-/*
- * Do a full breadth-first scan of the rcu_node structures for the
- * specified rcu_state structure.
- */
-#define rcu_for_each_node_breadth_first(rsp, rnp) \
-	for ((rnp) = &(rsp)->node[0]; \
-	     (rnp) < &(rsp)->node[rcu_num_nodes]; (rnp)++)
-
-/*
- * Do a breadth-first scan of the non-leaf rcu_node structures for the
- * specified rcu_state structure.  Note that if there is a singleton
- * rcu_node tree with but one rcu_node structure, this loop is a no-op.
- */
-#define rcu_for_each_nonleaf_node_breadth_first(rsp, rnp) \
-	for ((rnp) = &(rsp)->node[0]; \
-	     (rnp) < (rsp)->level[rcu_num_lvls - 1]; (rnp)++)
-
-/*
- * Scan the leaves of the rcu_node hierarchy for the specified rcu_state
- * structure.  Note that if there is a singleton rcu_node tree with but
- * one rcu_node structure, this loop -will- visit the rcu_node structure.
- * It is still a leaf node, even if it is also the root node.
- */
-#define rcu_for_each_leaf_node(rsp, rnp) \
-	for ((rnp) = (rsp)->level[rcu_num_lvls - 1]; \
-	     (rnp) < &(rsp)->node[rcu_num_nodes]; (rnp)++)
-
-/*
- * Iterate over all possible CPUs in a leaf RCU node.
- */
-#define for_each_leaf_node_possible_cpu(rnp, cpu) \
-	for ((cpu) = cpumask_next(rnp->grplo - 1, cpu_possible_mask); \
-	     cpu <= rnp->grphi; \
-	     cpu = cpumask_next((cpu), cpu_possible_mask))
-
 /*
  * Union to allow "aggregate OR" operation on the need for a quiescent
  * state by the normal and expedited grace periods.
@@ -336,34 +232,9 @@ struct rcu_data {
 					/* period it is aware of. */
 
 	/* 2) batch handling */
-	/*
-	 * If nxtlist is not NULL, it is partitioned as follows.
-	 * Any of the partitions might be empty, in which case the
-	 * pointer to that partition will be equal to the pointer for
-	 * the following partition.  When the list is empty, all of
-	 * the nxttail elements point to the ->nxtlist pointer itself,
-	 * which in that case is NULL.
-	 *
-	 * [nxtlist, *nxttail[RCU_DONE_TAIL]):
-	 *	Entries that batch # <= ->completed
-	 *	The grace period for these entries has completed, and
-	 *	the other grace-period-completed entries may be moved
-	 *	here temporarily in rcu_process_callbacks().
-	 * [*nxttail[RCU_DONE_TAIL], *nxttail[RCU_WAIT_TAIL]):
-	 *	Entries that batch # <= ->completed - 1: waiting for current GP
-	 * [*nxttail[RCU_WAIT_TAIL], *nxttail[RCU_NEXT_READY_TAIL]):
-	 *	Entries known to have arrived before current GP ended
-	 * [*nxttail[RCU_NEXT_READY_TAIL], *nxttail[RCU_NEXT_TAIL]):
-	 *	Entries that might have arrived after current GP ended
-	 *	Note that the value of *nxttail[RCU_NEXT_TAIL] will
-	 *	always be NULL, as this is the end of the list.
-	 */
-	struct rcu_head *nxtlist;
-	struct rcu_head **nxttail[RCU_NEXT_SIZE];
-	unsigned long	nxtcompleted[RCU_NEXT_SIZE];
-					/* grace periods for sublists. */
-	long		qlen_lazy;	/* # of lazy queued callbacks */
-	long		qlen;		/* # of queued callbacks, incl lazy */
+	struct rcu_segcblist cblist;	/* Segmented callback list, with */
+					/* different callbacks waiting for */
+					/* different grace periods. */
 	long		qlen_last_fqs_check;
 					/* qlen at last check for QS forcing */
 	unsigned long	n_cbs_invoked;	/* count of RCU cbs invoked. */
@@ -482,7 +353,6 @@ struct rcu_state {
 	struct rcu_node *level[RCU_NUM_LVLS + 1];
 						/* Hierarchy levels (+1 to */
 						/*  shut bogus gcc warning) */
-	u8 flavor_mask;				/* bit in flavor mask. */
 	struct rcu_data __percpu *rda;		/* pointer of percu rcu_data. */
 	call_rcu_func_t call;			/* call_rcu() flavor. */
 	int ncpus;				/* # CPUs seen so far. */
@@ -502,14 +372,11 @@ struct rcu_state {
 
 	raw_spinlock_t orphan_lock ____cacheline_internodealigned_in_smp;
 						/* Protect following fields. */
-	struct rcu_head *orphan_nxtlist;	/* Orphaned callbacks that */
+	struct rcu_cblist orphan_pend;		/* Orphaned callbacks that */
 						/*  need a grace period. */
-	struct rcu_head **orphan_nxttail;	/* Tail of above. */
-	struct rcu_head *orphan_donelist;	/* Orphaned callbacks that */
+	struct rcu_cblist orphan_done;		/* Orphaned callbacks that */
 						/*  are ready to invoke. */
-	struct rcu_head **orphan_donetail;	/* Tail of above. */
-	long qlen_lazy;				/* Number of lazy callbacks. */
-	long qlen;				/* Total number of callbacks. */
+						/* (Contains counts.) */
 	/* End of fields guarded by orphan_lock. */
 
 	struct mutex barrier_mutex;		/* Guards barrier fields. */
@@ -596,6 +463,7 @@ extern struct rcu_state rcu_preempt_state;
 #endif /* #ifdef CONFIG_PREEMPT_RCU */
 
 int rcu_dynticks_snap(struct rcu_dynticks *rdtp);
+bool rcu_eqs_special_set(int cpu);
 
 #ifdef CONFIG_RCU_BOOST
 DECLARE_PER_CPU(unsigned int, rcu_cpu_kthread_status);
@@ -673,6 +541,14 @@ static bool rcu_nohz_full_cpu(struct rcu_state *rsp);
 static void rcu_dynticks_task_enter(void);
 static void rcu_dynticks_task_exit(void);
 
+#ifdef CONFIG_SRCU
+void srcu_online_cpu(unsigned int cpu);
+void srcu_offline_cpu(unsigned int cpu);
+#else /* #ifdef CONFIG_SRCU */
+void srcu_online_cpu(unsigned int cpu) { }
+void srcu_offline_cpu(unsigned int cpu) { }
+#endif /* #else #ifdef CONFIG_SRCU */
+
 #endif /* #ifndef RCU_TREE_NONCORE */
 
 #ifdef CONFIG_RCU_TRACE

kernel/rcu/tree_exp.h

@@ -292,7 +292,7 @@ static bool exp_funnel_lock(struct rcu_state *rsp, unsigned long s)
 			trace_rcu_exp_funnel_lock(rsp->name, rnp->level,
 						  rnp->grplo, rnp->grphi,
 						  TPS("wait"));
-			wait_event(rnp->exp_wq[(s >> 1) & 0x3],
+			wait_event(rnp->exp_wq[rcu_seq_ctr(s) & 0x3],
 				   sync_exp_work_done(rsp,
 						      &rdp->exp_workdone2, s));
 			return true;
@@ -331,6 +331,8 @@ static void sync_sched_exp_handler(void *data)
 		return;
 	}
 	__this_cpu_write(rcu_sched_data.cpu_no_qs.b.exp, true);
+	/* Store .exp before .rcu_urgent_qs. */
+	smp_store_release(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs), true);
 	resched_cpu(smp_processor_id());
 }
 
@@ -531,7 +533,8 @@ static void rcu_exp_wait_wake(struct rcu_state *rsp, unsigned long s)
 				rnp->exp_seq_rq = s;
 			spin_unlock(&rnp->exp_lock);
 		}
-		wake_up_all(&rnp->exp_wq[(rsp->expedited_sequence >> 1) & 0x3]);
+		smp_mb(); /* All above changes before wakeup. */
+		wake_up_all(&rnp->exp_wq[rcu_seq_ctr(rsp->expedited_sequence) & 0x3]);
 	}
 	trace_rcu_exp_grace_period(rsp->name, s, TPS("endwake"));
 	mutex_unlock(&rsp->exp_wake_mutex);
@@ -609,9 +612,9 @@ static void _synchronize_rcu_expedited(struct rcu_state *rsp,
 	/* Wait for expedited grace period to complete. */
 	rdp = per_cpu_ptr(rsp->rda, raw_smp_processor_id());
 	rnp = rcu_get_root(rsp);
-	wait_event(rnp->exp_wq[(s >> 1) & 0x3],
-		   sync_exp_work_done(rsp,
-				      &rdp->exp_workdone0, s));
+	wait_event(rnp->exp_wq[rcu_seq_ctr(s) & 0x3],
+		   sync_exp_work_done(rsp, &rdp->exp_workdone0, s));
+	smp_mb(); /* Workqueue actions happen before return. */
 
 	/* Let the next expedited grace period start. */
 	mutex_unlock(&rsp->exp_mutex);
@@ -735,15 +738,3 @@ void synchronize_rcu_expedited(void)
 EXPORT_SYMBOL_GPL(synchronize_rcu_expedited);
 
 #endif /* #else #ifdef CONFIG_PREEMPT_RCU */
-
-/*
- * Switch to run-time mode once Tree RCU has fully initialized.
- */
-static int __init rcu_exp_runtime_mode(void)
-{
-	rcu_test_sync_prims();
-	rcu_scheduler_active = RCU_SCHEDULER_RUNNING;
-	rcu_test_sync_prims();
-	return 0;
-}
-core_initcall(rcu_exp_runtime_mode);

kernel/rcu/tree_plugin.h

@@ -1350,10 +1350,10 @@ static bool __maybe_unused rcu_try_advance_all_cbs(void)
 		 */
 		if ((rdp->completed != rnp->completed ||
 		     unlikely(READ_ONCE(rdp->gpwrap))) &&
-		    rdp->nxttail[RCU_DONE_TAIL] != rdp->nxttail[RCU_NEXT_TAIL])
+		    rcu_segcblist_pend_cbs(&rdp->cblist))
 			note_gp_changes(rsp, rdp);
 
-		if (cpu_has_callbacks_ready_to_invoke(rdp))
+		if (rcu_segcblist_ready_cbs(&rdp->cblist))
 			cbs_ready = true;
 	}
 	return cbs_ready;
@@ -1461,7 +1461,7 @@ static void rcu_prepare_for_idle(void)
 	rdtp->last_accelerate = jiffies;
 	for_each_rcu_flavor(rsp) {
 		rdp = this_cpu_ptr(rsp->rda);
-		if (!*rdp->nxttail[RCU_DONE_TAIL])
+		if (rcu_segcblist_pend_cbs(&rdp->cblist))
 			continue;
 		rnp = rdp->mynode;
 		raw_spin_lock_rcu_node(rnp); /* irqs already disabled. */
@@ -1529,7 +1529,7 @@ static void rcu_oom_notify_cpu(void *unused)
 
 	for_each_rcu_flavor(rsp) {
 		rdp = raw_cpu_ptr(rsp->rda);
-		if (rdp->qlen_lazy != 0) {
+		if (rcu_segcblist_n_lazy_cbs(&rdp->cblist)) {
 			atomic_inc(&oom_callback_count);
 			rsp->call(&rdp->oom_head, rcu_oom_callback);
 		}
@@ -1709,7 +1709,7 @@ __setup("rcu_nocbs=", rcu_nocb_setup);
 
 static int __init parse_rcu_nocb_poll(char *arg)
 {
-	rcu_nocb_poll = 1;
+	rcu_nocb_poll = true;
 	return 0;
 }
 early_param("rcu_nocb_poll", parse_rcu_nocb_poll);
@@ -1860,7 +1860,9 @@ static void __call_rcu_nocb_enqueue(struct rcu_data *rdp,
 			trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu,
 					    TPS("WakeEmpty"));
 		} else {
-			rdp->nocb_defer_wakeup = RCU_NOGP_WAKE;
+			WRITE_ONCE(rdp->nocb_defer_wakeup, RCU_NOGP_WAKE);
+			/* Store ->nocb_defer_wakeup before ->rcu_urgent_qs. */
+			smp_store_release(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs), true);
 			trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu,
 					    TPS("WakeEmptyIsDeferred"));
 		}
@@ -1872,7 +1874,9 @@ static void __call_rcu_nocb_enqueue(struct rcu_data *rdp,
 			trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu,
 					    TPS("WakeOvf"));
 		} else {
-			rdp->nocb_defer_wakeup = RCU_NOGP_WAKE_FORCE;
+			WRITE_ONCE(rdp->nocb_defer_wakeup, RCU_NOGP_WAKE_FORCE);
+			/* Store ->nocb_defer_wakeup before ->rcu_urgent_qs. */
+			smp_store_release(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs), true);
 			trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu,
 					    TPS("WakeOvfIsDeferred"));
 		}
@@ -1930,30 +1934,26 @@ static bool __maybe_unused rcu_nocb_adopt_orphan_cbs(struct rcu_state *rsp,
 						     struct rcu_data *rdp,
 						     unsigned long flags)
 {
-	long ql = rsp->qlen;
-	long qll = rsp->qlen_lazy;
+	long ql = rcu_cblist_n_cbs(&rsp->orphan_done);
+	long qll = rcu_cblist_n_lazy_cbs(&rsp->orphan_done);
 
 	/* If this is not a no-CBs CPU, tell the caller to do it the old way. */
 	if (!rcu_is_nocb_cpu(smp_processor_id()))
 		return false;
-	rsp->qlen = 0;
-	rsp->qlen_lazy = 0;
 
 	/* First, enqueue the donelist, if any.  This preserves CB ordering. */
-	if (rsp->orphan_donelist != NULL) {
-		__call_rcu_nocb_enqueue(rdp, rsp->orphan_donelist,
-					rsp->orphan_donetail, ql, qll, flags);
-		ql = qll = 0;
-		rsp->orphan_donelist = NULL;
-		rsp->orphan_donetail = &rsp->orphan_donelist;
+	if (!rcu_cblist_empty(&rsp->orphan_done)) {
+		__call_rcu_nocb_enqueue(rdp, rcu_cblist_head(&rsp->orphan_done),
+					rcu_cblist_tail(&rsp->orphan_done),
+					ql, qll, flags);
 	}
-	if (rsp->orphan_nxtlist != NULL) {
-		__call_rcu_nocb_enqueue(rdp, rsp->orphan_nxtlist,
-					rsp->orphan_nxttail, ql, qll, flags);
-		ql = qll = 0;
-		rsp->orphan_nxtlist = NULL;
-		rsp->orphan_nxttail = &rsp->orphan_nxtlist;
+	if (!rcu_cblist_empty(&rsp->orphan_pend)) {
+		__call_rcu_nocb_enqueue(rdp, rcu_cblist_head(&rsp->orphan_pend),
+					rcu_cblist_tail(&rsp->orphan_pend),
+					ql, qll, flags);
 	}
+	rcu_cblist_init(&rsp->orphan_done);
+	rcu_cblist_init(&rsp->orphan_pend);
 	return true;
 }
 
@@ -2395,16 +2395,16 @@ static bool init_nocb_callback_list(struct rcu_data *rdp)
 		return false;
 
 	/* If there are early-boot callbacks, move them to nocb lists. */
-	if (rdp->nxtlist) {
-		rdp->nocb_head = rdp->nxtlist;
-		rdp->nocb_tail = rdp->nxttail[RCU_NEXT_TAIL];
-		atomic_long_set(&rdp->nocb_q_count, rdp->qlen);
-		atomic_long_set(&rdp->nocb_q_count_lazy, rdp->qlen_lazy);
-		rdp->nxtlist = NULL;
-		rdp->qlen = 0;
-		rdp->qlen_lazy = 0;
+	if (!rcu_segcblist_empty(&rdp->cblist)) {
+		rdp->nocb_head = rcu_segcblist_head(&rdp->cblist);
+		rdp->nocb_tail = rcu_segcblist_tail(&rdp->cblist);
+		atomic_long_set(&rdp->nocb_q_count,
+				rcu_segcblist_n_cbs(&rdp->cblist));
+		atomic_long_set(&rdp->nocb_q_count_lazy,
+				rcu_segcblist_n_lazy_cbs(&rdp->cblist));
+		rcu_segcblist_init(&rdp->cblist);
 	}
-	rdp->nxttail[RCU_NEXT_TAIL] = NULL;
+	rcu_segcblist_disable(&rdp->cblist);
 	return true;
 }
 

kernel/rcu/tree_trace.c

@@ -41,11 +41,11 @@
 #include <linux/mutex.h>
 #include <linux/debugfs.h>
 #include <linux/seq_file.h>
+#include <linux/prefetch.h>
 
 #define RCU_TREE_NONCORE
 #include "tree.h"
-
-DECLARE_PER_CPU_SHARED_ALIGNED(unsigned long, rcu_qs_ctr);
+#include "rcu.h"
 
 static int r_open(struct inode *inode, struct file *file,
 					const struct seq_operations *op)
@@ -121,7 +121,7 @@ static void print_one_rcu_data(struct seq_file *m, struct rcu_data *rdp)
 		   cpu_is_offline(rdp->cpu) ? '!' : ' ',
 		   ulong2long(rdp->completed), ulong2long(rdp->gpnum),
 		   rdp->cpu_no_qs.b.norm,
-		   rdp->rcu_qs_ctr_snap == per_cpu(rcu_qs_ctr, rdp->cpu),
+		   rdp->rcu_qs_ctr_snap == per_cpu(rdp->dynticks->rcu_qs_ctr, rdp->cpu),
 		   rdp->core_needs_qs);
 	seq_printf(m, " dt=%d/%llx/%d df=%lu",
 		   rcu_dynticks_snap(rdp->dynticks),
@@ -130,17 +130,15 @@ static void print_one_rcu_data(struct seq_file *m, struct rcu_data *rdp)
 		   rdp->dynticks_fqs);
 	seq_printf(m, " of=%lu", rdp->offline_fqs);
 	rcu_nocb_q_lengths(rdp, &ql, &qll);
-	qll += rdp->qlen_lazy;
-	ql += rdp->qlen;
+	qll += rcu_segcblist_n_lazy_cbs(&rdp->cblist);
+	ql += rcu_segcblist_n_cbs(&rdp->cblist);
 	seq_printf(m, " ql=%ld/%ld qs=%c%c%c%c",
 		   qll, ql,
-		   ".N"[rdp->nxttail[RCU_NEXT_READY_TAIL] !=
-			rdp->nxttail[RCU_NEXT_TAIL]],
-		   ".R"[rdp->nxttail[RCU_WAIT_TAIL] !=
-			rdp->nxttail[RCU_NEXT_READY_TAIL]],
-		   ".W"[rdp->nxttail[RCU_DONE_TAIL] !=
-			rdp->nxttail[RCU_WAIT_TAIL]],
-		   ".D"[&rdp->nxtlist != rdp->nxttail[RCU_DONE_TAIL]]);
+		   ".N"[!rcu_segcblist_segempty(&rdp->cblist, RCU_NEXT_TAIL)],
+		   ".R"[!rcu_segcblist_segempty(&rdp->cblist,
+						RCU_NEXT_READY_TAIL)],
+		   ".W"[!rcu_segcblist_segempty(&rdp->cblist, RCU_WAIT_TAIL)],
+		   ".D"[!rcu_segcblist_segempty(&rdp->cblist, RCU_DONE_TAIL)]);
 #ifdef CONFIG_RCU_BOOST
 	seq_printf(m, " kt=%d/%c ktl=%x",
 		   per_cpu(rcu_cpu_has_work, rdp->cpu),
@@ -278,7 +276,9 @@ static void print_one_rcu_state(struct seq_file *m, struct rcu_state *rsp)
 	seq_printf(m, "nfqs=%lu/nfqsng=%lu(%lu) fqlh=%lu oqlen=%ld/%ld\n",
 		   rsp->n_force_qs, rsp->n_force_qs_ngp,
 		   rsp->n_force_qs - rsp->n_force_qs_ngp,
-		   READ_ONCE(rsp->n_force_qs_lh), rsp->qlen_lazy, rsp->qlen);
+		   READ_ONCE(rsp->n_force_qs_lh),
+		   rcu_cblist_n_lazy_cbs(&rsp->orphan_done),
+		   rcu_cblist_n_cbs(&rsp->orphan_done));
 	for (rnp = &rsp->node[0]; rnp - &rsp->node[0] < rcu_num_nodes; rnp++) {
 		if (rnp->level != level) {
 			seq_puts(m, "\n");

kernel/rcu/update.c

@@ -124,7 +124,7 @@ EXPORT_SYMBOL(rcu_read_lock_sched_held);
  * non-expedited counterparts?  Intended for use within RCU.  Note
  * that if the user specifies both rcu_expedited and rcu_normal, then
  * rcu_normal wins.  (Except during the time period during boot from
- * when the first task is spawned until the rcu_exp_runtime_mode()
+ * when the first task is spawned until the rcu_set_runtime_mode()
  * core_initcall() is invoked, at which point everything is expedited.)
  */
 bool rcu_gp_is_normal(void)
@@ -190,6 +190,39 @@ void rcu_end_inkernel_boot(void)
 
 #endif /* #ifndef CONFIG_TINY_RCU */
 
+/*
+ * Test each non-SRCU synchronous grace-period wait API.  This is
+ * useful just after a change in mode for these primitives, and
+ * during early boot.
+ */
+void rcu_test_sync_prims(void)
+{
+	if (!IS_ENABLED(CONFIG_PROVE_RCU))
+		return;
+	synchronize_rcu();
+	synchronize_rcu_bh();
+	synchronize_sched();
+	synchronize_rcu_expedited();
+	synchronize_rcu_bh_expedited();
+	synchronize_sched_expedited();
+}
+
+#if !defined(CONFIG_TINY_RCU) || defined(CONFIG_SRCU)
+
+/*
+ * Switch to run-time mode once RCU has fully initialized.
+ */
+static int __init rcu_set_runtime_mode(void)
+{
+	rcu_test_sync_prims();
+	rcu_scheduler_active = RCU_SCHEDULER_RUNNING;
+	rcu_test_sync_prims();
+	return 0;
+}
+core_initcall(rcu_set_runtime_mode);
+
+#endif /* #if !defined(CONFIG_TINY_RCU) || defined(CONFIG_SRCU) */
+
 #ifdef CONFIG_PREEMPT_RCU
 
 /*
@@ -632,6 +665,7 @@ static void check_holdout_task(struct task_struct *t,
 		put_task_struct(t);
 		return;
 	}
+	rcu_request_urgent_qs_task(t);
 	if (!needreport)
 		return;
 	if (*firstreport) {
@@ -817,23 +851,6 @@ static void rcu_spawn_tasks_kthread(void)
 
 #endif /* #ifdef CONFIG_TASKS_RCU */
 
-/*
- * Test each non-SRCU synchronous grace-period wait API.  This is
- * useful just after a change in mode for these primitives, and
- * during early boot.
- */
-void rcu_test_sync_prims(void)
-{
-	if (!IS_ENABLED(CONFIG_PROVE_RCU))
-		return;
-	synchronize_rcu();
-	synchronize_rcu_bh();
-	synchronize_sched();
-	synchronize_rcu_expedited();
-	synchronize_rcu_bh_expedited();
-	synchronize_sched_expedited();
-}
-
 #ifdef CONFIG_PROVE_RCU
 
 /*

kernel/sched/core.c

@@ -3378,7 +3378,7 @@ static void __sched notrace __schedule(bool preempt)
 		hrtick_clear(rq);
 
 	local_irq_disable();
-	rcu_note_context_switch();
+	rcu_note_context_switch(preempt);
 
 	/*
 	 * Make sure that signal_pending_state()->signal_pending() below

kernel/signal.c

@@ -1237,7 +1237,7 @@ struct sighand_struct *__lock_task_sighand(struct task_struct *tsk,
 		}
 		/*
 		 * This sighand can be already freed and even reused, but
-		 * we rely on SLAB_DESTROY_BY_RCU and sighand_ctor() which
+		 * we rely on SLAB_TYPESAFE_BY_RCU and sighand_ctor() which
 		 * initializes ->siglock: this slab can't go away, it has
 		 * the same object type, ->siglock can't be reinitialized.
 		 *

mm/kasan/kasan.c

@@ -413,7 +413,7 @@ void kasan_cache_create(struct kmem_cache *cache, size_t *size,
 	*size += sizeof(struct kasan_alloc_meta);
 
 	/* Add free meta. */
-	if (cache->flags & SLAB_DESTROY_BY_RCU || cache->ctor ||
+	if (cache->flags & SLAB_TYPESAFE_BY_RCU || cache->ctor ||
 	    cache->object_size < sizeof(struct kasan_free_meta)) {
 		cache->kasan_info.free_meta_offset = *size;
 		*size += sizeof(struct kasan_free_meta);
@@ -561,7 +561,7 @@ static void kasan_poison_slab_free(struct kmem_cache *cache, void *object)
 	unsigned long rounded_up_size = round_up(size, KASAN_SHADOW_SCALE_SIZE);
 
 	/* RCU slabs could be legally used after free within the RCU period */
-	if (unlikely(cache->flags & SLAB_DESTROY_BY_RCU))
+	if (unlikely(cache->flags & SLAB_TYPESAFE_BY_RCU))
 		return;
 
 	kasan_poison_shadow(object, rounded_up_size, KASAN_KMALLOC_FREE);
@@ -572,7 +572,7 @@ bool kasan_slab_free(struct kmem_cache *cache, void *object)
 	s8 shadow_byte;
 
 	/* RCU slabs could be legally used after free within the RCU period */
-	if (unlikely(cache->flags & SLAB_DESTROY_BY_RCU))
+	if (unlikely(cache->flags & SLAB_TYPESAFE_BY_RCU))
 		return false;
 
 	shadow_byte = READ_ONCE(*(s8 *)kasan_mem_to_shadow(object));

mm/kmemcheck.c

@@ -95,7 +95,7 @@ void kmemcheck_slab_alloc(struct kmem_cache *s, gfp_t gfpflags, void *object,
 void kmemcheck_slab_free(struct kmem_cache *s, void *object, size_t size)
 {
 	/* TODO: RCU freeing is unsupported for now; hide false positives. */
-	if (!s->ctor && !(s->flags & SLAB_DESTROY_BY_RCU))
+	if (!s->ctor && !(s->flags & SLAB_TYPESAFE_BY_RCU))
 		kmemcheck_mark_freed(object, size);
 }
 

mm/mmu_notifier.c

@@ -21,7 +21,7 @@
 #include <linux/slab.h>
 
 /* global SRCU for all MMs */
-static struct srcu_struct srcu;
+DEFINE_STATIC_SRCU(srcu);
 
 /*
  * This function allows mmu_notifier::release callback to delay a call to
@@ -252,12 +252,6 @@ static int do_mmu_notifier_register(struct mmu_notifier *mn,
 
 	BUG_ON(atomic_read(&mm->mm_users) <= 0);
 
-	/*
-	 * Verify that mmu_notifier_init() already run and the global srcu is
-	 * initialized.
-	 */
-	BUG_ON(!srcu.per_cpu_ref);
-
 	ret = -ENOMEM;
 	mmu_notifier_mm = kmalloc(sizeof(struct mmu_notifier_mm), GFP_KERNEL);
 	if (unlikely(!mmu_notifier_mm))
@@ -406,9 +400,3 @@ void mmu_notifier_unregister_no_release(struct mmu_notifier *mn,
 	mmdrop(mm);
 }
 EXPORT_SYMBOL_GPL(mmu_notifier_unregister_no_release);
-
-static int __init mmu_notifier_init(void)
-{
-	return init_srcu_struct(&srcu);
-}
-subsys_initcall(mmu_notifier_init);

mm/rmap.c

@@ -430,7 +430,7 @@ static void anon_vma_ctor(void *data)
 void __init anon_vma_init(void)
 {
 	anon_vma_cachep = kmem_cache_create("anon_vma", sizeof(struct anon_vma),
-			0, SLAB_DESTROY_BY_RCU|SLAB_PANIC|SLAB_ACCOUNT,
+			0, SLAB_TYPESAFE_BY_RCU|SLAB_PANIC|SLAB_ACCOUNT,
 			anon_vma_ctor);
 	anon_vma_chain_cachep = KMEM_CACHE(anon_vma_chain,
 			SLAB_PANIC|SLAB_ACCOUNT);
@@ -481,7 +481,7 @@ struct anon_vma *page_get_anon_vma(struct page *page)
 	 * If this page is still mapped, then its anon_vma cannot have been
 	 * freed.  But if it has been unmapped, we have no security against the
 	 * anon_vma structure being freed and reused (for another anon_vma:
-	 * SLAB_DESTROY_BY_RCU guarantees that - so the atomic_inc_not_zero()
+	 * SLAB_TYPESAFE_BY_RCU guarantees that - so the atomic_inc_not_zero()
 	 * above cannot corrupt).
 	 */
 	if (!page_mapped(page)) {

mm/slab.c

@@ -1728,7 +1728,7 @@ static void slab_destroy(struct kmem_cache *cachep, struct page *page)
 
 	freelist = page->freelist;
 	slab_destroy_debugcheck(cachep, page);
-	if (unlikely(cachep->flags & SLAB_DESTROY_BY_RCU))
+	if (unlikely(cachep->flags & SLAB_TYPESAFE_BY_RCU))
 		call_rcu(&page->rcu_head, kmem_rcu_free);
 	else
 		kmem_freepages(cachep, page);
@@ -1924,7 +1924,7 @@ static bool set_objfreelist_slab_cache(struct kmem_cache *cachep,
 
 	cachep->num = 0;
 
-	if (cachep->ctor || flags & SLAB_DESTROY_BY_RCU)
+	if (cachep->ctor || flags & SLAB_TYPESAFE_BY_RCU)
 		return false;
 
 	left = calculate_slab_order(cachep, size,
@@ -2030,7 +2030,7 @@ __kmem_cache_create (struct kmem_cache *cachep, unsigned long flags)
 	if (size < 4096 || fls(size - 1) == fls(size-1 + REDZONE_ALIGN +
 						2 * sizeof(unsigned long long)))
 		flags |= SLAB_RED_ZONE | SLAB_STORE_USER;
-	if (!(flags & SLAB_DESTROY_BY_RCU))
+	if (!(flags & SLAB_TYPESAFE_BY_RCU))
 		flags |= SLAB_POISON;
 #endif
 #endif

mm/slab.h

@@ -126,7 +126,7 @@ static inline unsigned long kmem_cache_flags(unsigned long object_size,
 
 /* Legal flag mask for kmem_cache_create(), for various configurations */
 #define SLAB_CORE_FLAGS (SLAB_HWCACHE_ALIGN | SLAB_CACHE_DMA | SLAB_PANIC | \
-			 SLAB_DESTROY_BY_RCU | SLAB_DEBUG_OBJECTS )
+			 SLAB_TYPESAFE_BY_RCU | SLAB_DEBUG_OBJECTS )
 
 #if defined(CONFIG_DEBUG_SLAB)
 #define SLAB_DEBUG_FLAGS (SLAB_RED_ZONE | SLAB_POISON | SLAB_STORE_USER)
@@ -415,7 +415,7 @@ static inline size_t slab_ksize(const struct kmem_cache *s)
 	 * back there or track user information then we can
 	 * only use the space before that information.
 	 */
-	if (s->flags & (SLAB_DESTROY_BY_RCU | SLAB_STORE_USER))
+	if (s->flags & (SLAB_TYPESAFE_BY_RCU | SLAB_STORE_USER))
 		return s->inuse;
 	/*
 	 * Else we can use all the padding etc for the allocation

mm/slab_common.c

@@ -39,7 +39,7 @@ static DECLARE_WORK(slab_caches_to_rcu_destroy_work,
  * Set of flags that will prevent slab merging
  */
 #define SLAB_NEVER_MERGE (SLAB_RED_ZONE | SLAB_POISON | SLAB_STORE_USER | \
-		SLAB_TRACE | SLAB_DESTROY_BY_RCU | SLAB_NOLEAKTRACE | \
+		SLAB_TRACE | SLAB_TYPESAFE_BY_RCU | SLAB_NOLEAKTRACE | \
 		SLAB_FAILSLAB | SLAB_KASAN)
 
 #define SLAB_MERGE_SAME (SLAB_RECLAIM_ACCOUNT | SLAB_CACHE_DMA | \
@@ -500,7 +500,7 @@ static void slab_caches_to_rcu_destroy_workfn(struct work_struct *work)
 	struct kmem_cache *s, *s2;
 
 	/*
-	 * On destruction, SLAB_DESTROY_BY_RCU kmem_caches are put on the
+	 * On destruction, SLAB_TYPESAFE_BY_RCU kmem_caches are put on the
 	 * @slab_caches_to_rcu_destroy list.  The slab pages are freed
 	 * through RCU and and the associated kmem_cache are dereferenced
 	 * while freeing the pages, so the kmem_caches should be freed only
@@ -537,7 +537,7 @@ static int shutdown_cache(struct kmem_cache *s)
 	memcg_unlink_cache(s);
 	list_del(&s->list);
 
-	if (s->flags & SLAB_DESTROY_BY_RCU) {
+	if (s->flags & SLAB_TYPESAFE_BY_RCU) {
 		list_add_tail(&s->list, &slab_caches_to_rcu_destroy);
 		schedule_work(&slab_caches_to_rcu_destroy_work);
 	} else {

mm/slob.c

@@ -126,7 +126,7 @@ static inline void clear_slob_page_free(struct page *sp)
 
 /*
  * struct slob_rcu is inserted at the tail of allocated slob blocks, which
- * were created with a SLAB_DESTROY_BY_RCU slab. slob_rcu is used to free
+ * were created with a SLAB_TYPESAFE_BY_RCU slab. slob_rcu is used to free
  * the block using call_rcu.
  */
 struct slob_rcu {
@@ -524,7 +524,7 @@ EXPORT_SYMBOL(ksize);
 
 int __kmem_cache_create(struct kmem_cache *c, unsigned long flags)
 {
-	if (flags & SLAB_DESTROY_BY_RCU) {
+	if (flags & SLAB_TYPESAFE_BY_RCU) {
 		/* leave room for rcu footer at the end of object */
 		c->size += sizeof(struct slob_rcu);
 	}
@@ -598,7 +598,7 @@ static void kmem_rcu_free(struct rcu_head *head)
 void kmem_cache_free(struct kmem_cache *c, void *b)
 {
 	kmemleak_free_recursive(b, c->flags);
-	if (unlikely(c->flags & SLAB_DESTROY_BY_RCU)) {
+	if (unlikely(c->flags & SLAB_TYPESAFE_BY_RCU)) {
 		struct slob_rcu *slob_rcu;
 		slob_rcu = b + (c->size - sizeof(struct slob_rcu));
 		slob_rcu->size = c->size;

mm/slub.c

@@ -1687,7 +1687,7 @@ static void rcu_free_slab(struct rcu_head *h)
 
 static void free_slab(struct kmem_cache *s, struct page *page)
 {
-	if (unlikely(s->flags & SLAB_DESTROY_BY_RCU)) {
+	if (unlikely(s->flags & SLAB_TYPESAFE_BY_RCU)) {
 		struct rcu_head *head;
 
 		if (need_reserve_slab_rcu) {
@@ -2963,7 +2963,7 @@ static __always_inline void slab_free(struct kmem_cache *s, struct page *page,
 	 * slab_free_freelist_hook() could have put the items into quarantine.
 	 * If so, no need to free them.
 	 */
-	if (s->flags & SLAB_KASAN && !(s->flags & SLAB_DESTROY_BY_RCU))
+	if (s->flags & SLAB_KASAN && !(s->flags & SLAB_TYPESAFE_BY_RCU))
 		return;
 	do_slab_free(s, page, head, tail, cnt, addr);
 }
@@ -3433,7 +3433,7 @@ static int calculate_sizes(struct kmem_cache *s, int forced_order)
 	 * the slab may touch the object after free or before allocation
 	 * then we should never poison the object itself.
 	 */
-	if ((flags & SLAB_POISON) && !(flags & SLAB_DESTROY_BY_RCU) &&
+	if ((flags & SLAB_POISON) && !(flags & SLAB_TYPESAFE_BY_RCU) &&
 			!s->ctor)
 		s->flags |= __OBJECT_POISON;
 	else
@@ -3455,7 +3455,7 @@ static int calculate_sizes(struct kmem_cache *s, int forced_order)
 	 */
 	s->inuse = size;
 
-	if (((flags & (SLAB_DESTROY_BY_RCU | SLAB_POISON)) ||
+	if (((flags & (SLAB_TYPESAFE_BY_RCU | SLAB_POISON)) ||
 		s->ctor)) {
 		/*
 		 * Relocate free pointer after the object if it is not
@@ -3537,7 +3537,7 @@ static int kmem_cache_open(struct kmem_cache *s, unsigned long flags)
 	s->flags = kmem_cache_flags(s->size, flags, s->name, s->ctor);
 	s->reserved = 0;
 
-	if (need_reserve_slab_rcu && (s->flags & SLAB_DESTROY_BY_RCU))
+	if (need_reserve_slab_rcu && (s->flags & SLAB_TYPESAFE_BY_RCU))
 		s->reserved = sizeof(struct rcu_head);
 
 	if (!calculate_sizes(s, -1))
@@ -5042,7 +5042,7 @@ SLAB_ATTR_RO(cache_dma);
 
 static ssize_t destroy_by_rcu_show(struct kmem_cache *s, char *buf)
 {
-	return sprintf(buf, "%d\n", !!(s->flags & SLAB_DESTROY_BY_RCU));
+	return sprintf(buf, "%d\n", !!(s->flags & SLAB_TYPESAFE_BY_RCU));
 }
 SLAB_ATTR_RO(destroy_by_rcu);
 

net/dccp/ipv4.c

@@ -950,7 +950,7 @@ static struct proto dccp_v4_prot = {
 	.orphan_count		= &dccp_orphan_count,
 	.max_header		= MAX_DCCP_HEADER,
 	.obj_size		= sizeof(struct dccp_sock),
-	.slab_flags		= SLAB_DESTROY_BY_RCU,
+	.slab_flags		= SLAB_TYPESAFE_BY_RCU,
 	.rsk_prot		= &dccp_request_sock_ops,
 	.twsk_prot		= &dccp_timewait_sock_ops,
 	.h.hashinfo		= &dccp_hashinfo,

net/dccp/ipv6.c

@@ -1012,7 +1012,7 @@ static struct proto dccp_v6_prot = {
 	.orphan_count	   = &dccp_orphan_count,
 	.max_header	   = MAX_DCCP_HEADER,
 	.obj_size	   = sizeof(struct dccp6_sock),
-	.slab_flags	   = SLAB_DESTROY_BY_RCU,
+	.slab_flags	   = SLAB_TYPESAFE_BY_RCU,
 	.rsk_prot	   = &dccp6_request_sock_ops,
 	.twsk_prot	   = &dccp6_timewait_sock_ops,
 	.h.hashinfo	   = &dccp_hashinfo,

net/ipv4/tcp_ipv4.c

@@ -2398,7 +2398,7 @@ struct proto tcp_prot = {
 	.sysctl_rmem		= sysctl_tcp_rmem,
 	.max_header		= MAX_TCP_HEADER,
 	.obj_size		= sizeof(struct tcp_sock),
-	.slab_flags		= SLAB_DESTROY_BY_RCU,
+	.slab_flags		= SLAB_TYPESAFE_BY_RCU,
 	.twsk_prot		= &tcp_timewait_sock_ops,
 	.rsk_prot		= &tcp_request_sock_ops,
 	.h.hashinfo		= &tcp_hashinfo,

net/ipv6/tcp_ipv6.c

@@ -1919,7 +1919,7 @@ struct proto tcpv6_prot = {
 	.sysctl_rmem		= sysctl_tcp_rmem,
 	.max_header		= MAX_TCP_HEADER,
 	.obj_size		= sizeof(struct tcp6_sock),
-	.slab_flags		= SLAB_DESTROY_BY_RCU,
+	.slab_flags		= SLAB_TYPESAFE_BY_RCU,
 	.twsk_prot		= &tcp6_timewait_sock_ops,
 	.rsk_prot		= &tcp6_request_sock_ops,
 	.h.hashinfo		= &tcp_hashinfo,

net/llc/af_llc.c

@@ -142,7 +142,7 @@ static struct proto llc_proto = {
 	.name	  = "LLC",
 	.owner	  = THIS_MODULE,
 	.obj_size = sizeof(struct llc_sock),
-	.slab_flags = SLAB_DESTROY_BY_RCU,
+	.slab_flags = SLAB_TYPESAFE_BY_RCU,
 };
 
 /**

net/llc/llc_conn.c

@@ -506,7 +506,7 @@ static struct sock *__llc_lookup_established(struct llc_sap *sap,
 again:
 	sk_nulls_for_each_rcu(rc, node, laddr_hb) {
 		if (llc_estab_match(sap, daddr, laddr, rc)) {
-			/* Extra checks required by SLAB_DESTROY_BY_RCU */
+			/* Extra checks required by SLAB_TYPESAFE_BY_RCU */
 			if (unlikely(!atomic_inc_not_zero(&rc->sk_refcnt)))
 				goto again;
 			if (unlikely(llc_sk(rc)->sap != sap ||
@@ -565,7 +565,7 @@ static struct sock *__llc_lookup_listener(struct llc_sap *sap,
 again:
 	sk_nulls_for_each_rcu(rc, node, laddr_hb) {
 		if (llc_listener_match(sap, laddr, rc)) {
-			/* Extra checks required by SLAB_DESTROY_BY_RCU */
+			/* Extra checks required by SLAB_TYPESAFE_BY_RCU */
 			if (unlikely(!atomic_inc_not_zero(&rc->sk_refcnt)))
 				goto again;
 			if (unlikely(llc_sk(rc)->sap != sap ||

net/llc/llc_sap.c

@@ -328,7 +328,7 @@ static struct sock *llc_lookup_dgram(struct llc_sap *sap,
 again:
 	sk_nulls_for_each_rcu(rc, node, laddr_hb) {
 		if (llc_dgram_match(sap, laddr, rc)) {
-			/* Extra checks required by SLAB_DESTROY_BY_RCU */
+			/* Extra checks required by SLAB_TYPESAFE_BY_RCU */
 			if (unlikely(!atomic_inc_not_zero(&rc->sk_refcnt)))
 				goto again;
 			if (unlikely(llc_sk(rc)->sap != sap ||

net/netfilter/nf_conntrack_core.c

@@ -914,7 +914,7 @@ static unsigned int early_drop_list(struct net *net,
 			continue;
 
 		/* kill only if still in same netns -- might have moved due to
-		 * SLAB_DESTROY_BY_RCU rules.
+		 * SLAB_TYPESAFE_BY_RCU rules.
 		 *
 		 * We steal the timer reference.  If that fails timer has
 		 * already fired or someone else deleted it. Just drop ref
@@ -1069,7 +1069,7 @@ __nf_conntrack_alloc(struct net *net,
 
 	/*
 	 * Do not use kmem_cache_zalloc(), as this cache uses
-	 * SLAB_DESTROY_BY_RCU.
+	 * SLAB_TYPESAFE_BY_RCU.
 	 */
 	ct = kmem_cache_alloc(nf_conntrack_cachep, gfp);
 	if (ct == NULL)
@@ -1114,7 +1114,7 @@ void nf_conntrack_free(struct nf_conn *ct)
 	struct net *net = nf_ct_net(ct);
 
 	/* A freed object has refcnt == 0, that's
-	 * the golden rule for SLAB_DESTROY_BY_RCU
+	 * the golden rule for SLAB_TYPESAFE_BY_RCU
 	 */
 	NF_CT_ASSERT(atomic_read(&ct->ct_general.use) == 0);
 
@@ -1878,7 +1878,7 @@ int nf_conntrack_init_start(void)
 	nf_conntrack_cachep = kmem_cache_create("nf_conntrack",
 						sizeof(struct nf_conn),
 						NFCT_INFOMASK + 1,
-						SLAB_DESTROY_BY_RCU | SLAB_HWCACHE_ALIGN, NULL);
+						SLAB_TYPESAFE_BY_RCU | SLAB_HWCACHE_ALIGN, NULL);
 	if (!nf_conntrack_cachep)
 		goto err_cachep;
 

net/smc/af_smc.c

@@ -101,7 +101,7 @@ struct proto smc_proto = {
 	.unhash		= smc_unhash_sk,
 	.obj_size	= sizeof(struct smc_sock),
 	.h.smc_hash	= &smc_v4_hashinfo,
-	.slab_flags	= SLAB_DESTROY_BY_RCU,
+	.slab_flags	= SLAB_TYPESAFE_BY_RCU,
 };
 EXPORT_SYMBOL_GPL(smc_proto);
 

tools/testing/selftests/rcutorture/bin/kvm-test-1-run.sh

@@ -170,7 +170,7 @@ qemu_append="`identify_qemu_append "$QEMU"`"
 # Pull in Kconfig-fragment boot parameters
 boot_args="`configfrag_boot_params "$boot_args" "$config_template"`"
 # Generate kernel-version-specific boot parameters
-boot_args="`per_version_boot_params "$boot_args" $builddir/.config $seconds`"
+boot_args="`per_version_boot_params "$boot_args" $resdir/.config $seconds`"
 
 if test -n "$TORTURE_BUILDONLY"
 then