Merge branch 'perf/jump-labels' into perf/core

Merge reason: After much naming discussion, there seems to be consensus
              now - queue it up for v3.4.

Signed-off-by: Ingo Molnar <mingo@elte.hu>

7 files changed, 119 insertions, 57 deletions

diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h
index f7c69580fea7..2172da2d9bb4 100644
--- a/include/linux/jump_label.h
+++ b/include/linux/jump_label.h
@@ -9,15 +9,15 @@
  *
  * Jump labels provide an interface to generate dynamic branches using
  * self-modifying code. Assuming toolchain and architecture support the result
- * of a "if (static_branch(&key))" statement is a unconditional branch (which
+ * of a "if (static_key_false(&key))" statement is a unconditional branch (which
  * defaults to false - and the true block is placed out of line).
  *
- * However at runtime we can change the 'static' branch target using
- * jump_label_{inc,dec}(). These function as a 'reference' count on the key
+ * However at runtime we can change the branch target using
+ * static_key_slow_{inc,dec}(). These function as a 'reference' count on the key
  * object and for as long as there are references all branches referring to
  * that particular key will point to the (out of line) true block.
  *
- * Since this relies on modifying code the jump_label_{inc,dec}() functions
+ * Since this relies on modifying code the static_key_slow_{inc,dec}() functions
  * must be considered absolute slow paths (machine wide synchronization etc.).
  * OTOH, since the affected branches are unconditional their runtime overhead
  * will be absolutely minimal, esp. in the default (off) case where the total
@@ -26,12 +26,26 @@
  *
  * When the control is directly exposed to userspace it is prudent to delay the
  * decrement to avoid high frequency code modifications which can (and do)
- * cause significant performance degradation. Struct jump_label_key_deferred and
- * jump_label_dec_deferred() provide for this.
+ * cause significant performance degradation. Struct static_key_deferred and
+ * static_key_slow_dec_deferred() provide for this.
  *
  * Lacking toolchain and or architecture support, it falls back to a simple
  * conditional branch.
- */
+ *
+ * struct static_key my_key = STATIC_KEY_INIT_TRUE;
+ *
+ *   if (static_key_true(&my_key)) {
+ *   }
+ *
+ * will result in the true case being in-line and starts the key with a single
+ * reference. Mixing static_key_true() and static_key_false() on the same key is not
+ * allowed.
+ *
+ * Not initializing the key (static data is initialized to 0s anyway) is the
+ * same as using STATIC_KEY_INIT_FALSE and static_key_false() is
+ * equivalent with static_branch().
+ *
+*/
 
 #include <linux/types.h>
 #include <linux/compiler.h>
@@ -39,16 +53,17 @@
 
 #if defined(CC_HAVE_ASM_GOTO) && defined(CONFIG_JUMP_LABEL)
 
-struct jump_label_key {
+struct static_key {
 	atomic_t enabled;
+/* Set lsb bit to 1 if branch is default true, 0 ot */
 	struct jump_entry *entries;
 #ifdef CONFIG_MODULES
-	struct jump_label_mod *next;
+	struct static_key_mod *next;
 #endif
 };
 
-struct jump_label_key_deferred {
-	struct jump_label_key key;
+struct static_key_deferred {
+	struct static_key key;
 	unsigned long timeout;
 	struct delayed_work work;
 };
@@ -66,13 +81,34 @@ struct module;
 
 #ifdef HAVE_JUMP_LABEL
 
-#ifdef CONFIG_MODULES
-#define JUMP_LABEL_INIT {ATOMIC_INIT(0), NULL, NULL}
-#else
-#define JUMP_LABEL_INIT {ATOMIC_INIT(0), NULL}
-#endif
+#define JUMP_LABEL_TRUE_BRANCH 1UL
+
+static
+inline struct jump_entry *jump_label_get_entries(struct static_key *key)
+{
+	return (struct jump_entry *)((unsigned long)key->entries
+						& ~JUMP_LABEL_TRUE_BRANCH);
+}
+
+static inline bool jump_label_get_branch_default(struct static_key *key)
+{
+	if ((unsigned long)key->entries & JUMP_LABEL_TRUE_BRANCH)
+		return true;
+	return false;
+}
+
+static __always_inline bool static_key_false(struct static_key *key)
+{
+	return arch_static_branch(key);
+}
 
-static __always_inline bool static_branch(struct jump_label_key *key)
+static __always_inline bool static_key_true(struct static_key *key)
+{
+	return !static_key_false(key);
+}
+
+/* Deprecated. Please use 'static_key_false() instead. */
+static __always_inline bool static_branch(struct static_key *key)
 {
 	return arch_static_branch(key);
 }
@@ -88,21 +124,24 @@ extern void arch_jump_label_transform(struct jump_entry *entry,
 extern void arch_jump_label_transform_static(struct jump_entry *entry,
 					     enum jump_label_type type);
 extern int jump_label_text_reserved(void *start, void *end);
-extern void jump_label_inc(struct jump_label_key *key);
-extern void jump_label_dec(struct jump_label_key *key);
-extern void jump_label_dec_deferred(struct jump_label_key_deferred *key);
-extern bool jump_label_enabled(struct jump_label_key *key);
+extern void static_key_slow_inc(struct static_key *key);
+extern void static_key_slow_dec(struct static_key *key);
+extern void static_key_slow_dec_deferred(struct static_key_deferred *key);
+extern bool static_key_enabled(struct static_key *key);
 extern void jump_label_apply_nops(struct module *mod);
-extern void jump_label_rate_limit(struct jump_label_key_deferred *key,
-		unsigned long rl);
+extern void
+jump_label_rate_limit(struct static_key_deferred *key, unsigned long rl);
+
+#define STATIC_KEY_INIT_TRUE ((struct static_key) \
+	{ .enabled = ATOMIC_INIT(1), .entries = (void *)1 })
+#define STATIC_KEY_INIT_FALSE ((struct static_key) \
+	{ .enabled = ATOMIC_INIT(0), .entries = (void *)0 })
 
 #else  /* !HAVE_JUMP_LABEL */
 
 #include <linux/atomic.h>
 
-#define JUMP_LABEL_INIT {ATOMIC_INIT(0)}
-
-struct jump_label_key {
+struct static_key {
 	atomic_t enabled;
 };
 
@@ -110,30 +149,45 @@ static __always_inline void jump_label_init(void)
 {
 }
 
-struct jump_label_key_deferred {
-	struct jump_label_key  key;
+struct static_key_deferred {
+	struct static_key  key;
 };
 
-static __always_inline bool static_branch(struct jump_label_key *key)
+static __always_inline bool static_key_false(struct static_key *key)
+{
+	if (unlikely(atomic_read(&key->enabled)) > 0)
+		return true;
+	return false;
+}
+
+static __always_inline bool static_key_true(struct static_key *key)
 {
-	if (unlikely(atomic_read(&key->enabled)))
+	if (likely(atomic_read(&key->enabled)) > 0)
 		return true;
 	return false;
 }
 
-static inline void jump_label_inc(struct jump_label_key *key)
+/* Deprecated. Please use 'static_key_false() instead. */
+static __always_inline bool static_branch(struct static_key *key)
+{
+	if (unlikely(atomic_read(&key->enabled)) > 0)
+		return true;
+	return false;
+}
+
+static inline void static_key_slow_inc(struct static_key *key)
 {
 	atomic_inc(&key->enabled);
 }
 
-static inline void jump_label_dec(struct jump_label_key *key)
+static inline void static_key_slow_dec(struct static_key *key)
 {
 	atomic_dec(&key->enabled);
 }
 
-static inline void jump_label_dec_deferred(struct jump_label_key_deferred *key)
+static inline void static_key_slow_dec_deferred(struct static_key_deferred *key)
 {
-	jump_label_dec(&key->key);
+	static_key_slow_dec(&key->key);
 }
 
 static inline int jump_label_text_reserved(void *start, void *end)
@@ -144,9 +198,9 @@ static inline int jump_label_text_reserved(void *start, void *end)
 static inline void jump_label_lock(void) {}
 static inline void jump_label_unlock(void) {}
 
-static inline bool jump_label_enabled(struct jump_label_key *key)
+static inline bool static_key_enabled(struct static_key *key)
 {
-	return !!atomic_read(&key->enabled);
+	return (atomic_read(&key->enabled) > 0);
 }
 
 static inline int jump_label_apply_nops(struct module *mod)
@@ -154,13 +208,20 @@ static inline int jump_label_apply_nops(struct module *mod)
 	return 0;
 }
 
-static inline void jump_label_rate_limit(struct jump_label_key_deferred *key,
+static inline void
+jump_label_rate_limit(struct static_key_deferred *key,
 		unsigned long rl)
 {
 }
+
+#define STATIC_KEY_INIT_TRUE ((struct static_key) \
+		{ .enabled = ATOMIC_INIT(1) })
+#define STATIC_KEY_INIT_FALSE ((struct static_key) \
+		{ .enabled = ATOMIC_INIT(0) })
+
 #endif	/* HAVE_JUMP_LABEL */
 
-#define jump_label_key_enabled	((struct jump_label_key){ .enabled = ATOMIC_INIT(1), })
-#define jump_label_key_disabled	((struct jump_label_key){ .enabled = ATOMIC_INIT(0), })
+#define STATIC_KEY_INIT STATIC_KEY_INIT_FALSE
+#define jump_label_enabled static_key_enabled
 
 #endif	/* _LINUX_JUMP_LABEL_H */
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 0eac07c95255..7dfaae7846ab 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -214,8 +214,8 @@ enum {
 #include <linux/skbuff.h>
 
 #ifdef CONFIG_RPS
-#include <linux/jump_label.h>
-extern struct jump_label_key rps_needed;
+#include <linux/static_key.h>
+extern struct static_key rps_needed;
 #endif
 
 struct neighbour;
diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index b809265607d0..29734be334c1 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -163,13 +163,13 @@ extern struct ctl_path nf_net_ipv4_netfilter_sysctl_path[];
 extern struct list_head nf_hooks[NFPROTO_NUMPROTO][NF_MAX_HOOKS];
 
 #if defined(CONFIG_JUMP_LABEL)
-#include <linux/jump_label.h>
-extern struct jump_label_key nf_hooks_needed[NFPROTO_NUMPROTO][NF_MAX_HOOKS];
+#include <linux/static_key.h>
+extern struct static_key nf_hooks_needed[NFPROTO_NUMPROTO][NF_MAX_HOOKS];
 static inline bool nf_hooks_active(u_int8_t pf, unsigned int hook)
 {
 	if (__builtin_constant_p(pf) &&
 	    __builtin_constant_p(hook))
-		return static_branch(&nf_hooks_needed[pf][hook]);
+		return static_key_false(&nf_hooks_needed[pf][hook]);
 
 	return !list_empty(&nf_hooks[pf][hook]);
 }
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 92a056f6d18d..64426b71381f 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -514,7 +514,7 @@ struct perf_guest_info_callbacks {
 #include <linux/ftrace.h>
 #include <linux/cpu.h>
 #include <linux/irq_work.h>
-#include <linux/jump_label.h>
+#include <linux/static_key.h>
 #include <linux/atomic.h>
 #include <asm/local.h>
 
@@ -1041,7 +1041,7 @@ static inline int is_software_event(struct perf_event *event)
 	return event->pmu->task_ctx_nr == perf_sw_context;
 }
 
-extern struct jump_label_key perf_swevent_enabled[PERF_COUNT_SW_MAX];
+extern struct static_key perf_swevent_enabled[PERF_COUNT_SW_MAX];
 
 extern void __perf_sw_event(u32, u64, struct pt_regs *, u64);
 
@@ -1069,7 +1069,7 @@ perf_sw_event(u32 event_id, u64 nr, struct pt_regs *regs, u64 addr)
 {
 	struct pt_regs hot_regs;
 
-	if (static_branch(&perf_swevent_enabled[event_id])) {
+	if (static_key_false(&perf_swevent_enabled[event_id])) {
 		if (!regs) {
 			perf_fetch_caller_regs(&hot_regs);
 			regs = &hot_regs;
@@ -1078,12 +1078,12 @@ perf_sw_event(u32 event_id, u64 nr, struct pt_regs *regs, u64 addr)
 	}
 }
 
-extern struct jump_label_key_deferred perf_sched_events;
+extern struct static_key_deferred perf_sched_events;
 
 static inline void perf_event_task_sched_in(struct task_struct *prev,
 					    struct task_struct *task)
 {
-	if (static_branch(&perf_sched_events.key))
+	if (static_key_false(&perf_sched_events.key))
 		__perf_event_task_sched_in(prev, task);
 }
 
@@ -1092,7 +1092,7 @@ static inline void perf_event_task_sched_out(struct task_struct *prev,
 {
 	perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, NULL, 0);
 
-	if (static_branch(&perf_sched_events.key))
+	if (static_key_false(&perf_sched_events.key))
 		__perf_event_task_sched_out(prev, next);
 }
 
diff --git a/include/linux/static_key.h b/include/linux/static_key.h
new file mode 100644
index 000000000000..27bd3f8a0857
--- /dev/null
+++ b/include/linux/static_key.h
@@ -0,0 +1 @@
+#include <linux/jump_label.h>
diff --git a/include/linux/tracepoint.h b/include/linux/tracepoint.h
index fc36da97ff7e..bd96ecd0e05c 100644
--- a/include/linux/tracepoint.h
+++ b/include/linux/tracepoint.h
@@ -17,7 +17,7 @@
 #include <linux/errno.h>
 #include <linux/types.h>
 #include <linux/rcupdate.h>
-#include <linux/jump_label.h>
+#include <linux/static_key.h>
 
 struct module;
 struct tracepoint;
@@ -29,7 +29,7 @@ struct tracepoint_func {
 
 struct tracepoint {
 	const char *name;		/* Tracepoint name */
-	struct jump_label_key key;
+	struct static_key key;
 	void (*regfunc)(void);
 	void (*unregfunc)(void);
 	struct tracepoint_func __rcu *funcs;
@@ -145,7 +145,7 @@ static inline void tracepoint_synchronize_unregister(void)
 	extern struct tracepoint __tracepoint_##name;			\
 	static inline void trace_##name(proto)				\
 	{								\
-		if (static_branch(&__tracepoint_##name.key))		\
+		if (static_key_false(&__tracepoint_##name.key))		\
 			__DO_TRACE(&__tracepoint_##name,		\
 				TP_PROTO(data_proto),			\
 				TP_ARGS(data_args),			\
@@ -188,7 +188,7 @@ static inline void tracepoint_synchronize_unregister(void)
 	__attribute__((section("__tracepoints_strings"))) = #name;	 \
 	struct tracepoint __tracepoint_##name				 \
 	__attribute__((section("__tracepoints"))) =			 \
-		{ __tpstrtab_##name, JUMP_LABEL_INIT, reg, unreg, NULL };\
+		{ __tpstrtab_##name, STATIC_KEY_INIT_FALSE, reg, unreg, NULL };\
 	static struct tracepoint * const __tracepoint_ptr_##name __used	 \
 	__attribute__((section("__tracepoints_ptrs"))) =		 \
 		&__tracepoint_##name;
diff --git a/include/net/sock.h b/include/net/sock.h
index 91c1c8baf020..dcde2d9268cd 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -55,7 +55,7 @@
 #include <linux/uaccess.h>
 #include <linux/memcontrol.h>
 #include <linux/res_counter.h>
-#include <linux/jump_label.h>
+#include <linux/static_key.h>
 
 #include <linux/filter.h>
 #include <linux/rculist_nulls.h>
@@ -924,13 +924,13 @@ inline void sk_refcnt_debug_release(const struct sock *sk)
 #endif /* SOCK_REFCNT_DEBUG */
 
 #if defined(CONFIG_CGROUP_MEM_RES_CTLR_KMEM) && defined(CONFIG_NET)
-extern struct jump_label_key memcg_socket_limit_enabled;
+extern struct static_key memcg_socket_limit_enabled;
 static inline struct cg_proto *parent_cg_proto(struct proto *proto,
 					       struct cg_proto *cg_proto)
 {
 	return proto->proto_cgroup(parent_mem_cgroup(cg_proto->memcg));
 }
-#define mem_cgroup_sockets_enabled static_branch(&memcg_socket_limit_enabled)
+#define mem_cgroup_sockets_enabled static_key_false(&memcg_socket_limit_enabled)
 #else
 #define mem_cgroup_sockets_enabled 0
 static inline struct cg_proto *parent_cg_proto(struct proto *proto,