Merge branch 'linux-linaro-lsk-v4.4-android' of git://git.linaro.org/kernel/linux-linaro-stable.git

* linux-linaro-lsk-v4.4-android: (1212 commits)
  ANDROID: sdcardfs: Change current->fs under lock
  ANDROID: sdcardfs: Don't use OVERRIDE_CRED macro
  ANDROID: restrict store of prefer_idle as boolean
  BACKPORT: arm/syscalls: Optimize address limit check
  UPSTREAM: syscalls: Use CHECK_DATA_CORRUPTION for addr_limit_user_check
  BACKPORT: arm64/syscalls: Check address limit on user-mode return
  BACKPORT: x86/syscalls: Check address limit on user-mode return
  BACKPORT: lkdtm: add bad USER_DS test
  UPSTREAM: bug: switch data corruption check to __must_check
  BACKPORT: lkdtm: Add tests for struct list corruption
  UPSTREAM: bug: Provide toggle for BUG on data corruption
  UPSTREAM: list: Split list_del() debug checking into separate function
  UPSTREAM: rculist: Consolidate DEBUG_LIST for list_add_rcu()
  BACKPORT: list: Split list_add() debug checking into separate function
  FROMLIST: ANDROID: binder: Add BINDER_GET_NODE_INFO_FOR_REF ioctl.
  BACKPORT: arm64/vdso: Fix nsec handling for CLOCK_MONOTONIC_RAW
  ANDROID: arm64: mm: fix 4.4.154 merge
  BACKPORT: zsmalloc: introduce zs_huge_class_size()
  BACKPORT: zram: drop max_zpage_size and use zs_huge_class_size()
  ANDROID: tracing: fix race condition reading saved tgids
  ...

Change-Id: I9f23db35eb926b6fa0d7af7dbbb55c9a37d536fc

15 files changed, 258 insertions, 120 deletions

diff --git a/kernel/audit_watch.c b/kernel/audit_watch.c
index a162661c9d60..f45a9a5d3e47 100644
--- a/kernel/audit_watch.c
+++ b/kernel/audit_watch.c
@@ -419,6 +419,13 @@ int audit_add_watch(struct audit_krule *krule, struct list_head **list)
 	struct path parent_path;
 	int h, ret = 0;
 
+	/*
+	 * When we will be calling audit_add_to_parent, krule->watch might have
+	 * been updated and watch might have been freed.
+	 * So we need to keep a reference of watch.
+	 */
+	audit_get_watch(watch);
+
 	mutex_unlock(&audit_filter_mutex);
 
 	/* Avoid calling path_lookup under audit_filter_mutex. */
@@ -427,8 +434,10 @@ int audit_add_watch(struct audit_krule *krule, struct list_head **list)
 	/* caller expects mutex locked */
 	mutex_lock(&audit_filter_mutex);
 
-	if (ret)
+	if (ret) {
+		audit_put_watch(watch);
 		return ret;
+	}
 
 	/* either find an old parent or attach a new one */
 	parent = audit_find_parent(d_backing_inode(parent_path.dentry));
@@ -446,6 +455,7 @@ int audit_add_watch(struct audit_krule *krule, struct list_head **list)
 	*list = &audit_inode_hash[h];
 error:
 	path_put(&parent_path);
+	audit_put_watch(watch);
 	return ret;
 }
 
diff --git a/kernel/fork.c b/kernel/fork.c
index 80445ca0420b..5a90b0f6d668 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1136,7 +1136,9 @@ static int copy_sighand(unsigned long clone_flags, struct task_struct *tsk)
 		return -ENOMEM;
 
 	atomic_set(&sig->count, 1);
+	spin_lock_irq(&current->sighand->siglock);
 	memcpy(sig->action, current->sighand->action, sizeof(sig->action));
+	spin_unlock_irq(&current->sighand->siglock);
 	return 0;
 }
 
@@ -1364,6 +1366,18 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 
 	cpufreq_task_times_init(p);
 
+	/*
+	 * This _must_ happen before we call free_task(), i.e. before we jump
+	 * to any of the bad_fork_* labels. This is to avoid freeing
+	 * p->set_child_tid which is (ab)used as a kthread's data pointer for
+	 * kernel threads (PF_KTHREAD).
+	 */
+	p->set_child_tid = (clone_flags & CLONE_CHILD_SETTID) ? child_tidptr : NULL;
+	/*
+	 * Clear TID on mm_release()?
+	 */
+	p->clear_child_tid = (clone_flags & CLONE_CHILD_CLEARTID) ? child_tidptr : NULL;
+
 	ftrace_graph_init_task(p);
 
 	rt_mutex_init_task(p);
@@ -1525,11 +1539,6 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 		}
 	}
 
-	p->set_child_tid = (clone_flags & CLONE_CHILD_SETTID) ? child_tidptr : NULL;
-	/*
-	 * Clear TID on mm_release()?
-	 */
-	p->clear_child_tid = (clone_flags & CLONE_CHILD_CLEARTID) ? child_tidptr : NULL;
 #ifdef CONFIG_BLOCK
 	p->plug = NULL;
 #endif
diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index 0dfa25276f39..6a0cd0d585d2 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -373,7 +373,6 @@ void handle_nested_irq(unsigned int irq)
 	raw_spin_lock_irq(&desc->lock);
 
 	desc->istate &= ~(IRQS_REPLAY | IRQS_WAITING);
-	kstat_incr_irqs_this_cpu(desc);
 
 	action = desc->action;
 	if (unlikely(!action || irqd_irq_disabled(&desc->irq_data))) {
@@ -381,6 +380,7 @@ void handle_nested_irq(unsigned int irq)
 		goto out_unlock;
 	}
 
+	kstat_incr_irqs_this_cpu(desc);
 	irqd_set(&desc->irq_data, IRQD_IRQ_INPROGRESS);
 	raw_spin_unlock_irq(&desc->lock);
 
@@ -447,13 +447,13 @@ void handle_simple_irq(struct irq_desc *desc)
 		goto out_unlock;
 
 	desc->istate &= ~(IRQS_REPLAY | IRQS_WAITING);
-	kstat_incr_irqs_this_cpu(desc);
 
 	if (unlikely(!desc->action || irqd_irq_disabled(&desc->irq_data))) {
 		desc->istate |= IRQS_PENDING;
 		goto out_unlock;
 	}
 
+	kstat_incr_irqs_this_cpu(desc);
 	handle_irq_event(desc);
 
 out_unlock:
@@ -497,7 +497,6 @@ void handle_level_irq(struct irq_desc *desc)
 		goto out_unlock;
 
 	desc->istate &= ~(IRQS_REPLAY | IRQS_WAITING);
-	kstat_incr_irqs_this_cpu(desc);
 
 	/*
 	 * If its disabled or no action available
@@ -508,6 +507,7 @@ void handle_level_irq(struct irq_desc *desc)
 		goto out_unlock;
 	}
 
+	kstat_incr_irqs_this_cpu(desc);
 	handle_irq_event(desc);
 
 	cond_unmask_irq(desc);
@@ -567,7 +567,6 @@ void handle_fasteoi_irq(struct irq_desc *desc)
 		goto out;
 
 	desc->istate &= ~(IRQS_REPLAY | IRQS_WAITING);
-	kstat_incr_irqs_this_cpu(desc);
 
 	/*
 	 * If its disabled or no action available
@@ -579,6 +578,7 @@ void handle_fasteoi_irq(struct irq_desc *desc)
 		goto out;
 	}
 
+	kstat_incr_irqs_this_cpu(desc);
 	if (desc->istate & IRQS_ONESHOT)
 		mask_irq(desc);
 
diff --git a/kernel/kthread.c b/kernel/kthread.c
index d9b0be5c6a5f..f529aafc1180 100644
--- a/kernel/kthread.c
+++ b/kernel/kthread.c
@@ -313,10 +313,16 @@ struct task_struct *kthread_create_on_node(int (*threadfn)(void *data),
 	task = create->result;
 	if (!IS_ERR(task)) {
 		static const struct sched_param param = { .sched_priority = 0 };
+		char name[TASK_COMM_LEN];
 		va_list args;
 
 		va_start(args, namefmt);
-		vsnprintf(task->comm, sizeof(task->comm), namefmt, args);
+		/*
+		 * task is already visible to other tasks, so updating
+		 * COMM must be protected.
+		 */
+		vsnprintf(name, sizeof(name), namefmt, args);
+		set_task_comm(task, name);
 		va_end(args);
 		/*
 		 * root may have changed our (kthreadd's) priority or CPU mask.
diff --git a/kernel/locking/osq_lock.c b/kernel/locking/osq_lock.c
index 05a37857ab55..8d7047ecef4e 100644
--- a/kernel/locking/osq_lock.c
+++ b/kernel/locking/osq_lock.c
@@ -104,6 +104,19 @@ bool osq_lock(struct optimistic_spin_queue *lock)
 
 	prev = decode_cpu(old);
 	node->prev = prev;
+
+	/*
+	 * osq_lock()			unqueue
+	 *
+	 * node->prev = prev		osq_wait_next()
+	 * WMB				MB
+	 * prev->next = node		next->prev = prev // unqueue-C
+	 *
+	 * Here 'node->prev' and 'next->prev' are the same variable and we need
+	 * to ensure these stores happen in-order to avoid corrupting the list.
+	 */
+	smp_wmb();
+
 	WRITE_ONCE(prev->next, node);
 
 	/*
diff --git a/kernel/locking/rwsem-xadd.c b/kernel/locking/rwsem-xadd.c
index a4d4de05b2d1..1be33caf157d 100644
--- a/kernel/locking/rwsem-xadd.c
+++ b/kernel/locking/rwsem-xadd.c
@@ -511,6 +511,33 @@ struct rw_semaphore *rwsem_wake(struct rw_semaphore *sem)
 	unsigned long flags;
 
 	/*
+	* __rwsem_down_write_failed_common(sem)
+	*   rwsem_optimistic_spin(sem)
+	*     osq_unlock(sem->osq)
+	*   ...
+	*   atomic_long_add_return(&sem->count)
+	*
+	*      - VS -
+	*
+	*              __up_write()
+	*                if (atomic_long_sub_return_release(&sem->count) < 0)
+	*                  rwsem_wake(sem)
+	*                    osq_is_locked(&sem->osq)
+	*
+	* And __up_write() must observe !osq_is_locked() when it observes the
+	* atomic_long_add_return() in order to not miss a wakeup.
+	*
+	* This boils down to:
+	*
+	* [S.rel] X = 1                [RmW] r0 = (Y += 0)
+	*         MB                         RMB
+	* [RmW]   Y += 1               [L]   r1 = X
+	*
+	* exists (r0=1 /\ r1=0)
+	*/
+	smp_rmb();
+
+	/*
 	 * If a spinner is present, it is not necessary to do the wakeup.
 	 * Try to do wakeup only if the trylock succeeds to minimize
 	 * spinlock contention which may introduce too much delay in the
diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig
index 2098f7d5fd9e..e746f7d1dc16 100644
--- a/kernel/power/Kconfig
+++ b/kernel/power/Kconfig
@@ -119,6 +119,7 @@ config PM_SLEEP
 	def_bool y
 	depends on SUSPEND || HIBERNATE_CALLBACKS
 	select PM
+	select SRCU
 
 config PM_SLEEP_SMP
 	def_bool y
diff --git a/kernel/sched/tune.c b/kernel/sched/tune.c
index d444fc1a4d58..5e47c29b44f6 100644
--- a/kernel/sched/tune.c
+++ b/kernel/sched/tune.c
@@ -569,7 +569,7 @@ prefer_idle_write(struct cgroup_subsys_state *css, struct cftype *cft,
 	    u64 prefer_idle)
 {
 	struct schedtune *st = css_st(css);
-	st->prefer_idle = prefer_idle;
+	st->prefer_idle = !!prefer_idle;
 
 	return 0;
 }
diff --git a/kernel/sys.c b/kernel/sys.c
index 0df4753d4969..ede0c1f4b860 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -1144,18 +1144,19 @@ static int override_release(char __user *release, size_t len)
 
 SYSCALL_DEFINE1(newuname, struct new_utsname __user *, name)
 {
-	int errno = 0;
+	struct new_utsname tmp;
 
 	down_read(&uts_sem);
-	if (copy_to_user(name, utsname(), sizeof *name))
-		errno = -EFAULT;
+	memcpy(&tmp, utsname(), sizeof(tmp));
 	up_read(&uts_sem);
+	if (copy_to_user(name, &tmp, sizeof(tmp)))
+		return -EFAULT;
 
-	if (!errno && override_release(name->release, sizeof(name->release)))
-		errno = -EFAULT;
-	if (!errno && override_architecture(name))
-		errno = -EFAULT;
-	return errno;
+	if (override_release(name->release, sizeof(name->release)))
+		return -EFAULT;
+	if (override_architecture(name))
+		return -EFAULT;
+	return 0;
 }
 
 #ifdef __ARCH_WANT_SYS_OLD_UNAME
@@ -1164,55 +1165,46 @@ SYSCALL_DEFINE1(newuname, struct new_utsname __user *, name)
  */
 SYSCALL_DEFINE1(uname, struct old_utsname __user *, name)
 {
-	int error = 0;
+	struct old_utsname tmp;
 
 	if (!name)
 		return -EFAULT;
 
 	down_read(&uts_sem);
-	if (copy_to_user(name, utsname(), sizeof(*name)))
-		error = -EFAULT;
+	memcpy(&tmp, utsname(), sizeof(tmp));
 	up_read(&uts_sem);
+	if (copy_to_user(name, &tmp, sizeof(tmp)))
+		return -EFAULT;
 
-	if (!error && override_release(name->release, sizeof(name->release)))
-		error = -EFAULT;
-	if (!error && override_architecture(name))
-		error = -EFAULT;
-	return error;
+	if (override_release(name->release, sizeof(name->release)))
+		return -EFAULT;
+	if (override_architecture(name))
+		return -EFAULT;
+	return 0;
 }
 
 SYSCALL_DEFINE1(olduname, struct oldold_utsname __user *, name)
 {
-	int error;
+	struct oldold_utsname tmp = {};
 
 	if (!name)
 		return -EFAULT;
-	if (!access_ok(VERIFY_WRITE, name, sizeof(struct oldold_utsname)))
-		return -EFAULT;
 
 	down_read(&uts_sem);
-	error = __copy_to_user(&name->sysname, &utsname()->sysname,
-			       __OLD_UTS_LEN);
-	error |= __put_user(0, name->sysname + __OLD_UTS_LEN);
-	error |= __copy_to_user(&name->nodename, &utsname()->nodename,
-				__OLD_UTS_LEN);
-	error |= __put_user(0, name->nodename + __OLD_UTS_LEN);
-	error |= __copy_to_user(&name->release, &utsname()->release,
-				__OLD_UTS_LEN);
-	error |= __put_user(0, name->release + __OLD_UTS_LEN);
-	error |= __copy_to_user(&name->version, &utsname()->version,
-				__OLD_UTS_LEN);
-	error |= __put_user(0, name->version + __OLD_UTS_LEN);
-	error |= __copy_to_user(&name->machine, &utsname()->machine,
-				__OLD_UTS_LEN);
-	error |= __put_user(0, name->machine + __OLD_UTS_LEN);
+	memcpy(&tmp.sysname, &utsname()->sysname, __OLD_UTS_LEN);
+	memcpy(&tmp.nodename, &utsname()->nodename, __OLD_UTS_LEN);
+	memcpy(&tmp.release, &utsname()->release, __OLD_UTS_LEN);
+	memcpy(&tmp.version, &utsname()->version, __OLD_UTS_LEN);
+	memcpy(&tmp.machine, &utsname()->machine, __OLD_UTS_LEN);
 	up_read(&uts_sem);
+	if (copy_to_user(name, &tmp, sizeof(tmp)))
+		return -EFAULT;
 
-	if (!error && override_architecture(name))
-		error = -EFAULT;
-	if (!error && override_release(name->release, sizeof(name->release)))
-		error = -EFAULT;
-	return error ? -EFAULT : 0;
+	if (override_architecture(name))
+		return -EFAULT;
+	if (override_release(name->release, sizeof(name->release)))
+		return -EFAULT;
+	return 0;
 }
 #endif
 
@@ -1226,17 +1218,18 @@ SYSCALL_DEFINE2(sethostname, char __user *, name, int, len)
 
 	if (len < 0 || len > __NEW_UTS_LEN)
 		return -EINVAL;
-	down_write(&uts_sem);
 	errno = -EFAULT;
 	if (!copy_from_user(tmp, name, len)) {
-		struct new_utsname *u = utsname();
+		struct new_utsname *u;
 
+		down_write(&uts_sem);
+		u = utsname();
 		memcpy(u->nodename, tmp, len);
 		memset(u->nodename + len, 0, sizeof(u->nodename) - len);
 		errno = 0;
 		uts_proc_notify(UTS_PROC_HOSTNAME);
+		up_write(&uts_sem);
 	}
-	up_write(&uts_sem);
 	return errno;
 }
 
@@ -1244,8 +1237,9 @@ SYSCALL_DEFINE2(sethostname, char __user *, name, int, len)
 
 SYSCALL_DEFINE2(gethostname, char __user *, name, int, len)
 {
-	int i, errno;
+	int i;
 	struct new_utsname *u;
+	char tmp[__NEW_UTS_LEN + 1];
 
 	if (len < 0)
 		return -EINVAL;
@@ -1254,11 +1248,11 @@ SYSCALL_DEFINE2(gethostname, char __user *, name, int, len)
 	i = 1 + strlen(u->nodename);
 	if (i > len)
 		i = len;
-	errno = 0;
-	if (copy_to_user(name, u->nodename, i))
-		errno = -EFAULT;
+	memcpy(tmp, u->nodename, i);
 	up_read(&uts_sem);
-	return errno;
+	if (copy_to_user(name, tmp, i))
+		return -EFAULT;
+	return 0;
 }
 
 #endif
@@ -1277,17 +1271,18 @@ SYSCALL_DEFINE2(setdomainname, char __user *, name, int, len)
 	if (len < 0 || len > __NEW_UTS_LEN)
 		return -EINVAL;
 
-	down_write(&uts_sem);
 	errno = -EFAULT;
 	if (!copy_from_user(tmp, name, len)) {
-		struct new_utsname *u = utsname();
+		struct new_utsname *u;
 
+		down_write(&uts_sem);
+		u = utsname();
 		memcpy(u->domainname, tmp, len);
 		memset(u->domainname + len, 0, sizeof(u->domainname) - len);
 		errno = 0;
 		uts_proc_notify(UTS_PROC_DOMAINNAME);
+		up_write(&uts_sem);
 	}
-	up_write(&uts_sem);
 	return errno;
 }
 
diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c
index 7ab5eafea8b2..210b8e726a97 100644
--- a/kernel/trace/blktrace.c
+++ b/kernel/trace/blktrace.c
@@ -1716,6 +1716,10 @@ static ssize_t sysfs_blk_trace_attr_store(struct device *dev,
 	mutex_lock(&bdev->bd_mutex);
 
 	if (attr == &dev_attr_enable) {
+		if (!!value == !!q->blk_trace) {
+			ret = 0;
+			goto out_unlock_bdev;
+		}
 		if (value)
 			ret = blk_trace_setup_queue(q, bdev);
 		else
diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index fdaa88f38aec..74b20e3ab8c6 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -1513,6 +1513,8 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned long nr_pages)
 	tmp_iter_page = first_page;
 
 	do {
+		cond_resched();
+
 		to_remove_page = tmp_iter_page;
 		rb_inc_page(cpu_buffer, &tmp_iter_page);
 
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index efdddec2af6e..2bee793749d8 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -1359,11 +1359,11 @@ void tracing_reset_all_online_cpus(void)
 
 #define SAVED_CMDLINES_DEFAULT 128
 #define NO_CMDLINE_MAP UINT_MAX
-static unsigned saved_tgids[SAVED_CMDLINES_DEFAULT];
 static arch_spinlock_t trace_cmdline_lock = __ARCH_SPIN_LOCK_UNLOCKED;
 struct saved_cmdlines_buffer {
 	unsigned map_pid_to_cmdline[PID_MAX_DEFAULT+1];
 	unsigned *map_cmdline_to_pid;
+	unsigned *map_cmdline_to_tgid;
 	unsigned cmdline_num;
 	int cmdline_idx;
 	char *saved_cmdlines;
@@ -1397,12 +1397,23 @@ static int allocate_cmdlines_buffer(unsigned int val,
 		return -ENOMEM;
 	}
 
+	s->map_cmdline_to_tgid = kmalloc_array(val,
+					       sizeof(*s->map_cmdline_to_tgid),
+					       GFP_KERNEL);
+	if (!s->map_cmdline_to_tgid) {
+		kfree(s->map_cmdline_to_pid);
+		kfree(s->saved_cmdlines);
+		return -ENOMEM;
+	}
+
 	s->cmdline_idx = 0;
 	s->cmdline_num = val;
 	memset(&s->map_pid_to_cmdline, NO_CMDLINE_MAP,
 	       sizeof(s->map_pid_to_cmdline));
 	memset(s->map_cmdline_to_pid, NO_CMDLINE_MAP,
 	       val * sizeof(*s->map_cmdline_to_pid));
+	memset(s->map_cmdline_to_tgid, NO_CMDLINE_MAP,
+	       val * sizeof(*s->map_cmdline_to_tgid));
 
 	return 0;
 }
@@ -1568,14 +1579,17 @@ static int trace_save_cmdline(struct task_struct *tsk)
 	if (!tsk->pid || unlikely(tsk->pid > PID_MAX_DEFAULT))
 		return 0;
 
+	preempt_disable();
 	/*
 	 * It's not the end of the world if we don't get
 	 * the lock, but we also don't want to spin
 	 * nor do we want to disable interrupts,
 	 * so if we miss here, then better luck next time.
 	 */
-	if (!arch_spin_trylock(&trace_cmdline_lock))
+	if (!arch_spin_trylock(&trace_cmdline_lock)) {
+		preempt_enable();
 		return 0;
+	}
 
 	idx = savedcmd->map_pid_to_cmdline[tsk->pid];
 	if (idx == NO_CMDLINE_MAP) {
@@ -1598,8 +1612,9 @@ static int trace_save_cmdline(struct task_struct *tsk)
 	}
 
 	set_cmdline(idx, tsk->comm);
-	saved_tgids[idx] = tsk->tgid;
+	savedcmd->map_cmdline_to_tgid[idx] = tsk->tgid;
 	arch_spin_unlock(&trace_cmdline_lock);
+	preempt_enable();
 
 	return 1;
 }
@@ -1641,19 +1656,29 @@ void trace_find_cmdline(int pid, char comm[])
 	preempt_enable();
 }
 
-int trace_find_tgid(int pid)
+static int __find_tgid_locked(int pid)
 {
 	unsigned map;
 	int tgid;
 
-	preempt_disable();
-	arch_spin_lock(&trace_cmdline_lock);
 	map = savedcmd->map_pid_to_cmdline[pid];
 	if (map != NO_CMDLINE_MAP)
-		tgid = saved_tgids[map];
+		tgid = savedcmd->map_cmdline_to_tgid[map];
 	else
 		tgid = -1;
 
+	return tgid;
+}
+
+int trace_find_tgid(int pid)
+{
+	int tgid;
+
+	preempt_disable();
+	arch_spin_lock(&trace_cmdline_lock);
+
+	tgid = __find_tgid_locked(pid);
+
 	arch_spin_unlock(&trace_cmdline_lock);
 	preempt_enable();
 
@@ -3970,10 +3995,15 @@ tracing_saved_cmdlines_size_read(struct file *filp, char __user *ubuf,
 {
 	char buf[64];
 	int r;
+	unsigned int n;
 
+	preempt_disable();
 	arch_spin_lock(&trace_cmdline_lock);
-	r = scnprintf(buf, sizeof(buf), "%u\n", savedcmd->cmdline_num);
+	n = savedcmd->cmdline_num;
 	arch_spin_unlock(&trace_cmdline_lock);
+	preempt_enable();
+
+	r = scnprintf(buf, sizeof(buf), "%u\n", n);
 
 	return simple_read_from_buffer(ubuf, cnt, ppos, buf, r);
 }
@@ -3982,6 +4012,7 @@ static void free_saved_cmdlines_buffer(struct saved_cmdlines_buffer *s)
 {
 	kfree(s->saved_cmdlines);
 	kfree(s->map_cmdline_to_pid);
+	kfree(s->map_cmdline_to_tgid);
 	kfree(s);
 }
 
@@ -3998,10 +4029,12 @@ static int tracing_resize_saved_cmdlines(unsigned int val)
 		return -ENOMEM;
 	}
 
+	preempt_disable();
 	arch_spin_lock(&trace_cmdline_lock);
 	savedcmd_temp = savedcmd;
 	savedcmd = s;
 	arch_spin_unlock(&trace_cmdline_lock);
+	preempt_enable();
 	free_saved_cmdlines_buffer(savedcmd_temp);
 
 	return 0;
@@ -4220,33 +4253,61 @@ tracing_saved_tgids_read(struct file *file, char __user *ubuf,
 	char *file_buf;
 	char *buf;
 	int len = 0;
-	int pid;
 	int i;
+	int *pids;
+	int n = 0;
 
-	file_buf = kmalloc(SAVED_CMDLINES_DEFAULT*(16+1+16), GFP_KERNEL);
-	if (!file_buf)
-		return -ENOMEM;
+	preempt_disable();
+	arch_spin_lock(&trace_cmdline_lock);
 
-	buf = file_buf;
+	pids = kmalloc_array(savedcmd->cmdline_num, 2*sizeof(int), GFP_KERNEL);
+	if (!pids) {
+		arch_spin_unlock(&trace_cmdline_lock);
+		preempt_enable();
+		return -ENOMEM;
+	}
 
-	for (i = 0; i < SAVED_CMDLINES_DEFAULT; i++) {
-		int tgid;
-		int r;
+	for (i = 0; i < savedcmd->cmdline_num; i++) {
+		int pid;
 
 		pid = savedcmd->map_cmdline_to_pid[i];
 		if (pid == -1 || pid == NO_CMDLINE_MAP)
 			continue;
 
-		tgid = trace_find_tgid(pid);
-		r = sprintf(buf, "%d %d\n", pid, tgid);
+		pids[n] = pid;
+		pids[n+1] = __find_tgid_locked(pid);
+		n += 2;
+	}
+	arch_spin_unlock(&trace_cmdline_lock);
+	preempt_enable();
+
+	if (n == 0) {
+		kfree(pids);
+		return 0;
+	}
+
+	/* enough to hold max pair of pids + space, lr and nul */
+	len = n * 12;
+	file_buf = kmalloc(len, GFP_KERNEL);
+	if (!file_buf) {
+		kfree(pids);
+		return -ENOMEM;
+	}
+
+	buf = file_buf;
+	for (i = 0; i < n && len > 0; i += 2) {
+		int r;
+
+		r = snprintf(buf, len, "%d %d\n", pids[i], pids[i+1]);
 		buf += r;
-		len += r;
+		len -= r;
 	}
 
 	len = simple_read_from_buffer(ubuf, cnt, ppos,
-				      file_buf, len);
+				      file_buf, buf - file_buf);
 
 	kfree(file_buf);
+	kfree(pids);
 
 	return len;
 }
@@ -6587,7 +6648,9 @@ rb_simple_write(struct file *filp, const char __user *ubuf,
 
 	if (buffer) {
 		mutex_lock(&trace_types_lock);
-		if (val) {
+		if (!!val == tracer_tracing_is_on(tr)) {
+			val = 0; /* do nothing */
+		} else if (val) {
 			tracer_tracing_on(tr);
 			if (tr->current_trace->start)
 				tr->current_trace->start(tr);
diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c
index 68bb89ad9d28..1dc887bab085 100644
--- a/kernel/trace/trace_uprobe.c
+++ b/kernel/trace/trace_uprobe.c
@@ -969,7 +969,7 @@ probe_event_disable(struct trace_uprobe *tu, struct trace_event_file *file)
 
 		list_del_rcu(&link->list);
 		/* synchronize with u{,ret}probe_trace_func */
-		synchronize_sched();
+		synchronize_rcu();
 		kfree(link);
 
 		if (!list_empty(&tu->tp.files))
diff --git a/kernel/user_namespace.c b/kernel/user_namespace.c
index 88fefa68c516..a965df4b54f5 100644
--- a/kernel/user_namespace.c
+++ b/kernel/user_namespace.c
@@ -602,9 +602,26 @@ static ssize_t map_write(struct file *file, const char __user *buf,
 	struct uid_gid_map new_map;
 	unsigned idx;
 	struct uid_gid_extent *extent = NULL;
-	unsigned long page = 0;
+	unsigned long page;
 	char *kbuf, *pos, *next_line;
-	ssize_t ret = -EINVAL;
+	ssize_t ret;
+
+	/* Only allow < page size writes at the beginning of the file */
+	if ((*ppos != 0) || (count >= PAGE_SIZE))
+		return -EINVAL;
+
+	/* Get a buffer */
+	page = __get_free_page(GFP_TEMPORARY);
+	kbuf = (char *) page;
+	if (!page)
+		return -ENOMEM;
+
+	/* Slurp in the user data */
+	if (copy_from_user(kbuf, buf, count)) {
+		free_page(page);
+		return -EFAULT;
+	}
+	kbuf[count] = '\0';
 
 	/*
 	 * The userns_state_mutex serializes all writes to any given map.
@@ -638,24 +655,6 @@ static ssize_t map_write(struct file *file, const char __user *buf,
 	if (cap_valid(cap_setid) && !file_ns_capable(file, ns, CAP_SYS_ADMIN))
 		goto out;
 
-	/* Get a buffer */
-	ret = -ENOMEM;
-	page = __get_free_page(GFP_TEMPORARY);
-	kbuf = (char *) page;
-	if (!page)
-		goto out;
-
-	/* Only allow < page size writes at the beginning of the file */
-	ret = -EINVAL;
-	if ((*ppos != 0) || (count >= PAGE_SIZE))
-		goto out;
-
-	/* Slurp in the user data */
-	ret = -EFAULT;
-	if (copy_from_user(kbuf, buf, count))
-		goto out;
-	kbuf[count] = '\0';
-
 	/* Parse the user data */
 	ret = -EINVAL;
 	pos = kbuf;
diff --git a/kernel/utsname_sysctl.c b/kernel/utsname_sysctl.c
index c8eac43267e9..d2b3b2973456 100644
--- a/kernel/utsname_sysctl.c
+++ b/kernel/utsname_sysctl.c
@@ -17,7 +17,7 @@
 
 #ifdef CONFIG_PROC_SYSCTL
 
-static void *get_uts(struct ctl_table *table, int write)
+static void *get_uts(struct ctl_table *table)
 {
 	char *which = table->data;
 	struct uts_namespace *uts_ns;
@@ -25,21 +25,9 @@ static void *get_uts(struct ctl_table *table, int write)
 	uts_ns = current->nsproxy->uts_ns;
 	which = (which - (char *)&init_uts_ns) + (char *)uts_ns;
 
-	if (!write)
-		down_read(&uts_sem);
-	else
-		down_write(&uts_sem);
 	return which;
 }
 
-static void put_uts(struct ctl_table *table, int write, void *which)
-{
-	if (!write)
-		up_read(&uts_sem);
-	else
-		up_write(&uts_sem);
-}
-
 /*
  *	Special case of dostring for the UTS structure. This has locks
  *	to observe. Should this be in kernel/sys.c ????
@@ -49,13 +37,34 @@ static int proc_do_uts_string(struct ctl_table *table, int write,
 {
 	struct ctl_table uts_table;
 	int r;
+	char tmp_data[__NEW_UTS_LEN + 1];
+
 	memcpy(&uts_table, table, sizeof(uts_table));
-	uts_table.data = get_uts(table, write);
+	uts_table.data = tmp_data;
+
+	/*
+	 * Buffer the value in tmp_data so that proc_dostring() can be called
+	 * without holding any locks.
+	 * We also need to read the original value in the write==1 case to
+	 * support partial writes.
+	 */
+	down_read(&uts_sem);
+	memcpy(tmp_data, get_uts(table), sizeof(tmp_data));
+	up_read(&uts_sem);
 	r = proc_dostring(&uts_table, write, buffer, lenp, ppos);
-	put_uts(table, write, uts_table.data);
 
-	if (write)
+	if (write) {
+		/*
+		 * Write back the new value.
+		 * Note that, since we dropped uts_sem, the result can
+		 * theoretically be incorrect if there are two parallel writes
+		 * at non-zero offsets to the same sysctl.
+		 */
+		down_write(&uts_sem);
+		memcpy(get_uts(table), tmp_data, sizeof(tmp_data));
+		up_write(&uts_sem);
 		proc_sys_poll_notify(table->poll);
+	}
 
 	return r;
 }