Merge tag 'v4.0-rc1' into perf/core, to refresh the tree

Signed-off-by: Ingo Molnar <mingo@kernel.org>

114 files changed, 2832 insertions, 1577 deletions

diff --git a/kernel/Kconfig.locks b/kernel/Kconfig.locks
index 76768ee812b2..08561f1acd13 100644
--- a/kernel/Kconfig.locks
+++ b/kernel/Kconfig.locks
@@ -231,6 +231,10 @@ config RWSEM_SPIN_ON_OWNER
        def_bool y
        depends on SMP && RWSEM_XCHGADD_ALGORITHM && ARCH_SUPPORTS_ATOMIC_RMW
 
+config LOCK_SPIN_ON_OWNER
+       def_bool y
+       depends on MUTEX_SPIN_ON_OWNER || RWSEM_SPIN_ON_OWNER
+
 config ARCH_USE_QUEUE_RWLOCK
 	bool
 
diff --git a/kernel/Makefile b/kernel/Makefile
index a59481a3fa6c..1408b3353a3c 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -13,8 +13,8 @@ obj-y     = fork.o exec_domain.o panic.o \
 
 ifdef CONFIG_FUNCTION_TRACER
 # Do not trace debug files and internal ftrace files
-CFLAGS_REMOVE_cgroup-debug.o = -pg
-CFLAGS_REMOVE_irq_work.o = -pg
+CFLAGS_REMOVE_cgroup-debug.o = $(CC_FLAGS_FTRACE)
+CFLAGS_REMOVE_irq_work.o = $(CC_FLAGS_FTRACE)
 endif
 
 # cond_syscall is currently not LTO compatible
@@ -26,6 +26,7 @@ obj-y += power/
 obj-y += printk/
 obj-y += irq/
 obj-y += rcu/
+obj-y += livepatch/
 
 obj-$(CONFIG_CHECKPOINT_RESTORE) += kcmp.o
 obj-$(CONFIG_FREEZER) += freezer.o
@@ -142,7 +143,7 @@ endif
 kernel/system_certificates.o: $(obj)/x509_certificate_list
 
 quiet_cmd_x509certs  = CERTS   $@
-      cmd_x509certs  = cat $(X509_CERTIFICATES) /dev/null >$@ $(foreach X509,$(X509_CERTIFICATES),; echo "  - Including cert $(X509)")
+      cmd_x509certs  = cat $(X509_CERTIFICATES) /dev/null >$@ $(foreach X509,$(X509_CERTIFICATES),; $(kecho) "  - Including cert $(X509)")
 
 targets += $(obj)/x509_certificate_list
 $(obj)/x509_certificate_list: $(X509_CERTIFICATES) $(obj)/.x509.list
diff --git a/kernel/acct.c b/kernel/acct.c
index 33738ef972f3..e6c10d1a4058 100644
--- a/kernel/acct.c
+++ b/kernel/acct.c
@@ -76,10 +76,11 @@ int acct_parm[3] = {4, 2, 30};
 /*
  * External references and all of the globals.
  */
-static void do_acct_process(struct bsd_acct_struct *acct);
 
 struct bsd_acct_struct {
 	struct fs_pin		pin;
+	atomic_long_t		count;
+	struct rcu_head		rcu;
 	struct mutex		lock;
 	int			active;
 	unsigned long		needcheck;
@@ -89,6 +90,8 @@ struct bsd_acct_struct {
 	struct completion	done;
 };
 
+static void do_acct_process(struct bsd_acct_struct *acct);
+
 /*
  * Check the amount of free space and suspend/resume accordingly.
  */
@@ -124,32 +127,56 @@ out:
 	return acct->active;
 }
 
+static void acct_put(struct bsd_acct_struct *p)
+{
+	if (atomic_long_dec_and_test(&p->count))
+		kfree_rcu(p, rcu);
+}
+
+static inline struct bsd_acct_struct *to_acct(struct fs_pin *p)
+{
+	return p ? container_of(p, struct bsd_acct_struct, pin) : NULL;
+}
+
 static struct bsd_acct_struct *acct_get(struct pid_namespace *ns)
 {
 	struct bsd_acct_struct *res;
 again:
 	smp_rmb();
 	rcu_read_lock();
-	res = ACCESS_ONCE(ns->bacct);
+	res = to_acct(ACCESS_ONCE(ns->bacct));
 	if (!res) {
 		rcu_read_unlock();
 		return NULL;
 	}
-	if (!atomic_long_inc_not_zero(&res->pin.count)) {
+	if (!atomic_long_inc_not_zero(&res->count)) {
 		rcu_read_unlock();
 		cpu_relax();
 		goto again;
 	}
 	rcu_read_unlock();
 	mutex_lock(&res->lock);
-	if (!res->ns) {
+	if (res != to_acct(ACCESS_ONCE(ns->bacct))) {
 		mutex_unlock(&res->lock);
-		pin_put(&res->pin);
+		acct_put(res);
 		goto again;
 	}
 	return res;
 }
 
+static void acct_pin_kill(struct fs_pin *pin)
+{
+	struct bsd_acct_struct *acct = to_acct(pin);
+	mutex_lock(&acct->lock);
+	do_acct_process(acct);
+	schedule_work(&acct->work);
+	wait_for_completion(&acct->done);
+	cmpxchg(&acct->ns->bacct, pin, NULL);
+	mutex_unlock(&acct->lock);
+	pin_remove(pin);
+	acct_put(acct);
+}
+
 static void close_work(struct work_struct *work)
 {
 	struct bsd_acct_struct *acct = container_of(work, struct bsd_acct_struct, work);
@@ -160,44 +187,13 @@ static void close_work(struct work_struct *work)
 	complete(&acct->done);
 }
 
-static void acct_kill(struct bsd_acct_struct *acct,
-		      struct bsd_acct_struct *new)
-{
-	if (acct) {
-		struct pid_namespace *ns = acct->ns;
-		do_acct_process(acct);
-		INIT_WORK(&acct->work, close_work);
-		init_completion(&acct->done);
-		schedule_work(&acct->work);
-		wait_for_completion(&acct->done);
-		pin_remove(&acct->pin);
-		ns->bacct = new;
-		acct->ns = NULL;
-		atomic_long_dec(&acct->pin.count);
-		mutex_unlock(&acct->lock);
-		pin_put(&acct->pin);
-	}
-}
-
-static void acct_pin_kill(struct fs_pin *pin)
-{
-	struct bsd_acct_struct *acct;
-	acct = container_of(pin, struct bsd_acct_struct, pin);
-	mutex_lock(&acct->lock);
-	if (!acct->ns) {
-		mutex_unlock(&acct->lock);
-		pin_put(pin);
-		acct = NULL;
-	}
-	acct_kill(acct, NULL);
-}
-
 static int acct_on(struct filename *pathname)
 {
 	struct file *file;
 	struct vfsmount *mnt, *internal;
 	struct pid_namespace *ns = task_active_pid_ns(current);
-	struct bsd_acct_struct *acct, *old;
+	struct bsd_acct_struct *acct;
+	struct fs_pin *old;
 	int err;
 
 	acct = kzalloc(sizeof(struct bsd_acct_struct), GFP_KERNEL);
@@ -238,21 +234,21 @@ static int acct_on(struct filename *pathname)
 	mnt = file->f_path.mnt;
 	file->f_path.mnt = internal;
 
-	atomic_long_set(&acct->pin.count, 1);
-	acct->pin.kill = acct_pin_kill;
+	atomic_long_set(&acct->count, 1);
+	init_fs_pin(&acct->pin, acct_pin_kill);
 	acct->file = file;
 	acct->needcheck = jiffies;
 	acct->ns = ns;
 	mutex_init(&acct->lock);
+	INIT_WORK(&acct->work, close_work);
+	init_completion(&acct->done);
 	mutex_lock_nested(&acct->lock, 1);	/* nobody has seen it yet */
 	pin_insert(&acct->pin, mnt);
 
-	old = acct_get(ns);
-	if (old)
-		acct_kill(old, acct);
-	else
-		ns->bacct = acct;
+	rcu_read_lock();
+	old = xchg(&ns->bacct, &acct->pin);
 	mutex_unlock(&acct->lock);
+	pin_kill(old);
 	mnt_drop_write(mnt);
 	mntput(mnt);
 	return 0;
@@ -288,7 +284,8 @@ SYSCALL_DEFINE1(acct, const char __user *, name)
 		mutex_unlock(&acct_on_mutex);
 		putname(tmp);
 	} else {
-		acct_kill(acct_get(task_active_pid_ns(current)), NULL);
+		rcu_read_lock();
+		pin_kill(task_active_pid_ns(current)->bacct);
 	}
 
 	return error;
@@ -296,7 +293,8 @@ SYSCALL_DEFINE1(acct, const char __user *, name)
 
 void acct_exit_ns(struct pid_namespace *ns)
 {
-	acct_kill(acct_get(ns), NULL);
+	rcu_read_lock();
+	pin_kill(ns->bacct);
 }
 
 /*
@@ -576,7 +574,7 @@ static void slow_acct_process(struct pid_namespace *ns)
 		if (acct) {
 			do_acct_process(acct);
 			mutex_unlock(&acct->lock);
-			pin_put(&acct->pin);
+			acct_put(acct);
 		}
 	}
 }
diff --git a/kernel/audit.h b/kernel/audit.h
index 3cdffad5a1d9..1caa0d345d90 100644
--- a/kernel/audit.h
+++ b/kernel/audit.h
@@ -24,12 +24,6 @@
 #include <linux/skbuff.h>
 #include <uapi/linux/mqueue.h>
 
-/* 0 = no checking
-   1 = put_count checking
-   2 = verbose put_count checking
-*/
-#define AUDIT_DEBUG 0
-
 /* AUDIT_NAMES is the number of slots we reserve in the audit_context
  * for saving names from getname().  If we get more names we will allocate
  * a name dynamically and also add those to the list anchored by names_list. */
@@ -74,9 +68,8 @@ struct audit_cap_data {
 	};
 };
 
-/* When fs/namei.c:getname() is called, we store the pointer in name and