2 files changed, 205 insertions, 83 deletions

diff --git a/kernel/events/core.c b/kernel/events/core.c
index 953c1434837..17b3c6cf160 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -175,8 +175,8 @@ int sysctl_perf_event_sample_rate __read_mostly	= DEFAULT_MAX_SAMPLE_RATE;
 static int max_samples_per_tick __read_mostly	= DIV_ROUND_UP(DEFAULT_MAX_SAMPLE_RATE, HZ);
 static int perf_sample_period_ns __read_mostly	= DEFAULT_SAMPLE_PERIOD_NS;
 
-static atomic_t perf_sample_allowed_ns __read_mostly =
-	ATOMIC_INIT( DEFAULT_SAMPLE_PERIOD_NS * DEFAULT_CPU_TIME_MAX_PERCENT / 100);
+static int perf_sample_allowed_ns __read_mostly =
+	DEFAULT_SAMPLE_PERIOD_NS * DEFAULT_CPU_TIME_MAX_PERCENT / 100;
 
 void update_perf_cpu_limits(void)
 {
@@ -184,7 +184,7 @@ void update_perf_cpu_limits(void)
 
 	tmp *= sysctl_perf_cpu_time_max_percent;
 	do_div(tmp, 100);
-	atomic_set(&perf_sample_allowed_ns, tmp);
+	ACCESS_ONCE(perf_sample_allowed_ns) = tmp;
 }
 
 static int perf_rotate_context(struct perf_cpu_context *cpuctx);
@@ -193,7 +193,7 @@ int perf_proc_update_handler(struct ctl_table *table, int write,
 		void __user *buffer, size_t *lenp,
 		loff_t *ppos)
 {
-	int ret = proc_dointvec(table, write, buffer, lenp, ppos);
+	int ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
 
 	if (ret || !write)
 		return ret;
@@ -228,14 +228,15 @@ int perf_cpu_time_max_percent_handler(struct ctl_table *table, int write,
  * we detect that events are taking too long.
  */
 #define NR_ACCUMULATED_SAMPLES 128
-DEFINE_PER_CPU(u64, running_sample_length);
+static DEFINE_PER_CPU(u64, running_sample_length);
 
 void perf_sample_event_took(u64 sample_len_ns)
 {
 	u64 avg_local_sample_len;
 	u64 local_samples_len;
+	u64 allowed_ns = ACCESS_ONCE(perf_sample_allowed_ns);
 
-	if (atomic_read(&perf_sample_allowed_ns) == 0)
+	if (allowed_ns == 0)
 		return;
 
 	/* decay the counter by 1 average sample */
@@ -251,7 +252,7 @@ void perf_sample_event_took(u64 sample_len_ns)
 	 */
 	avg_local_sample_len = local_samples_len/NR_ACCUMULATED_SAMPLES;
 
-	if (avg_local_sample_len <= atomic_read(&perf_sample_allowed_ns))
+	if (avg_local_sample_len <= allowed_ns)
 		return;
 
 	if (max_samples_per_tick <= 1)
@@ -262,10 +263,9 @@ void perf_sample_event_took(u64 sample_len_ns)
 	perf_sample_period_ns = NSEC_PER_SEC / sysctl_perf_event_sample_rate;
 
 	printk_ratelimited(KERN_WARNING
-			"perf samples too long (%lld > %d), lowering "
+			"perf samples too long (%lld > %lld), lowering "
 			"kernel.perf_event_max_sample_rate to %d\n",
-			avg_local_sample_len,
-			atomic_read(&perf_sample_allowed_ns),
+			avg_local_sample_len, allowed_ns,
 			sysctl_perf_event_sample_rate);
 
 	update_perf_cpu_limits();
@@ -899,6 +899,7 @@ static void unclone_ctx(struct perf_event_context *ctx)
 		put_ctx(ctx->parent_ctx);
 		ctx->parent_ctx = NULL;
 	}
+	ctx->generation++;
 }
 
 static u32 perf_event_pid(struct perf_event *event, struct task_struct *p)
@@ -1136,6 +1137,8 @@ list_add_event(struct perf_event *event, struct perf_event_context *ctx)
 	ctx->nr_events++;
 	if (event->attr.inherit_stat)
 		ctx->nr_stat++;
+
+	ctx->generation++;
 }
 
 /*
@@ -1201,6 +1204,9 @@ static void perf_event__header_size(struct perf_event *event)
 	if (sample_type & PERF_SAMPLE_DATA_SRC)
 		size += sizeof(data->data_src.val);
 
+	if (sample_type & PERF_SAMPLE_TRANSACTION)
+		size += sizeof(data->txn);
+
 	event->header_size = size;
 }
 
@@ -1310,6 +1316,8 @@ list_del_event(struct perf_event *event, struct perf_event_context *ctx)
 	 */
 	if (event->state > PERF_EVENT_STATE_OFF)
 		event->state = PERF_EVENT_STATE_OFF;
+
+	ctx->generation++;
 }
 
 static void perf_group_detach(struct perf_event *event)
@@ -2146,22 +2154,38 @@ static void ctx_sched_out(struct perf_event_context *ctx,
 }
 
 /*
- * Test whether two contexts are equivalent, i.e. whether they
- * have both been cloned from the same version of the same context
- * and they both have the same number of enabled events.
- * If the number of enabled events is the same, then the set
- * of enabled events should be the same, because these are both
- * inherited contexts, therefore we can't access individual events
- * in them directly with an fd; we can only enable/disable all
- * events via prctl, or enable/disable all events in a family
- * via ioctl, which will have the same effect on both contexts.
+ * Test whether two contexts are equivalent, i.e. whether they have both been
+ * cloned from the same version of the same context.
+ *
+ * Equivalence is measured using a generation number in the context that is
+ * incremented on each modification to it; see unclone_ctx(), list_add_event()
+ * and list_del_event().
  */
 static int context_equiv(struct perf_event_context *ctx1,
 			 struct perf_event_context *ctx2)
 {
-	return ctx1->parent_ctx && ctx1->parent_ctx == ctx2->parent_ctx
-		&& ctx1->parent_gen == ctx2->parent_gen
-		&& !ctx1->pin_count && !ctx2->pin_count;
+	/* Pinning disables the swap optimization */
+	if (ctx1->pin_count || ctx2->pin_count)
+		return 0;
+
+	/* If ctx1 is the parent of ctx2 */
+	if (ctx1 == ctx2->parent_ctx && ctx1->generation == ctx2->parent_gen)
+		return 1;
+
+	/* If ctx2 is the parent of ctx1 */
+	if (ctx1->parent_ctx == ctx2 && ctx1->parent_gen == ctx2->generation)
+		return 1;
+
+	/*
+	 * If ctx1 and ctx2 have the same parent; we flatten the parent
+	 * hierarchy, see perf_event_init_context().
+	 */
+	if (ctx1->parent_ctx && ctx1->parent_ctx == ctx2->parent_ctx &&
+			ctx1->parent_gen == ctx2->parent_gen)
+		return 1;
+
+	/* Unmatched */
+	return 0;
 }
 
 static void __perf_event_sync_stat(struct perf_event *event,
@@ -2244,7 +2268,7 @@ static void perf_event_context_sched_out(struct task_struct *task, int ctxn,
 {
 	struct perf_event_context *ctx = task->perf_event_ctxp[ctxn];
 	struct perf_event_context *next_ctx;
-	struct perf_event_context *parent;
+	struct perf_event_context *parent, *next_parent;
 	struct perf_cpu_context *cpuctx;
 	int do_switch = 1;
 
@@ -2256,10 +2280,18 @@ static void perf_event_context_sched_out(struct task_struct *task, int ctxn,
 		return;
 
 	rcu_read_lock();
-	parent = rcu_dereference(ctx->parent_ctx);
 	next_ctx = next->perf_event_ctxp[ctxn];
-	if (parent && next_ctx &&
-	    rcu_dereference(next_ctx->parent_ctx) == parent) {
+	if (!next_ctx)
+		goto unlock;
+
+	parent = rcu_dereference(ctx->parent_ctx);
+	next_parent = rcu_dereference(next_ctx->parent_ctx);
+
+	/* If neither context have a parent context; they cannot be clones. */
+	if (!parent && !next_parent)
+		goto unlock;
+
+	if (next_parent == ctx || next_ctx == parent || next_parent == parent) {
 		/*
 		 * Looks like the two contexts are clones, so we might be
 		 * able to optimize the context switch.  We lock both
@@ -2287,6 +2319,7 @@ static void perf_event_context_sched_out(struct task_struct *task, int ctxn,
 		raw_spin_unlock(&next_ctx->lock);
 		raw_spin_unlock(&ctx->lock);
 	}
+unlock:
 	rcu_read_unlock();
 
 	if (do_switch) {
@@ -4572,6 +4605,9 @@ void perf_output_sample(struct perf_output_handle *handle,
 	if (sample_type & PERF_SAMPLE_DATA_SRC)
 		perf_output_put(handle, data->data_src.val);
 
+	if (sample_type & PERF_SAMPLE_TRANSACTION)
+		perf_output_put(handle, data->txn);
+
 	if (!event->attr.watermark) {
 		int wakeup_events = event->attr.wakeup_events;
 
@@ -5100,24 +5136,23 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event)
 	unsigned int size;
 	char tmp[16];
 	char *buf = NULL;
-	const char *name;
-
-	memset(tmp, 0, sizeof(tmp));
+	char *name;
 
 	if (file) {
 		struct inode *inode;
 		dev_t dev;
-		/*
-		 * d_path works from the end of the rb backwards, so we
-		 * need to add enough zero bytes after the string to handle
-		 * the 64bit alignment we do later.
-		 */
-		buf = kzalloc(PATH_MAX + sizeof(u64), GFP_KERNEL);
+
+		buf = kmalloc(PATH_MAX, GFP_KERNEL);
 		if (!buf) {
 			name = strncpy(tmp, "//enomem", sizeof(tmp));
 			goto got_name;
 		}
-		name = d_path(&file->f_path, buf, PATH_MAX);
+		/*
+		 * d_path() works from the end of the rb backwards, so we
+		 * need to add enough zero bytes after the string to handle
+		 * the 64bit alignment we do later.
+		 */
+		name = d_path(&file->f_path, buf, PATH_MAX - sizeof(u64));
 		if (IS_ERR(name)) {
 			name = strncpy(tmp, "//toolong", sizeof(tmp));
 			goto got_name;
@@ -5130,21 +5165,19 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event)
 		min = MINOR(dev);
 
 	} else {
-		if (arch_vma_name(mmap_event->vma)) {
-			name = strncpy(tmp, arch_vma_name(mmap_event->vma),
-				       sizeof(tmp) - 1);
+		name = (char *)arch_vma_name(vma);
+		if (name) {
+			name = strncpy(tmp, name, sizeof(tmp) - 1);
 			tmp[sizeof(tmp) - 1] = '\0';
 			goto got_name;
 		}
 
-		if (!vma->vm_mm) {
-			name = strncpy(tmp, "[vdso]", sizeof(tmp));
-			goto got_name;
-		} else if (vma->vm_start <= vma->vm_mm->start_brk &&
+		if (vma->vm_start <= vma->vm_mm->start_brk &&
 				vma->vm_end >= vma->vm_mm->brk) {
 			name = strncpy(tmp, "[heap]", sizeof(tmp));
 			goto got_name;
-		} else if (vma->vm_start <= vma->vm_mm->start_stack &&
+		}
+		if (vma->vm_start <= vma->vm_mm->start_stack &&
 				vma->vm_end >= vma->vm_mm->start_stack) {
 			name = strncpy(tmp, "[stack]", sizeof(tmp));
 			goto got_name;
@@ -5155,7 +5188,14 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event)
 	}
 
 got_name:
-	size = ALIGN(strlen(name)+1, sizeof(u64));
+	/*
+	 * Since our buffer works in 8 byte units we need to align our string
+	 * size to a multiple of 8. However, we must guarantee the tail end is
+	 * zero'd out to avoid leaking random bits to userspace.
+	 */
+	size = strlen(name)+1;
+	while (!IS_ALIGNED(size, sizeof(u64)))
+		name[size++] = '\0';
 
 	mmap_event->file_name = name;
 	mmap_event->file_size = size;
@@ -7126,7 +7166,6 @@ SYSCALL_DEFINE5(perf_event_open,
 	}
 
 	perf_install_in_context(ctx, event, event->cpu);
-	++ctx->generation;
 	perf_unpin_context(ctx);
 	mutex_unlock(&ctx->mutex);
 
@@ -7209,7 +7248,6 @@ perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu,
 	WARN_ON_ONCE(ctx->parent_ctx);
 	mutex_lock(&ctx->mutex);
 	perf_install_in_context(ctx, event, cpu);
-	++ctx->generation;
 	perf_unpin_context(ctx);
 	mutex_unlock(&ctx->mutex);
 
diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c
index ad8e1bdca70..ae9e1d2ef25 100644
--- a/kernel/events/uprobes.c
+++ b/kernel/events/uprobes.c
@@ -35,6 +35,7 @@
 #include <linux/kdebug.h>	/* notifier mechanism */
 #include "../../mm/internal.h"	/* munlock_vma_page */
 #include <linux/percpu-rwsem.h>
+#include <linux/task_work.h>
 
 #include <linux/uprobes.h>
 
@@ -1096,21 +1097,22 @@ void uprobe_munmap(struct vm_area_struct *vma, unsigned long start, unsigned lon
 }
 
 /* Slot allocation for XOL */
-static int xol_add_vma(struct xol_area *area)
+static int xol_add_vma(struct mm_struct *mm, struct xol_area *area)
 {
-	struct mm_struct *mm = current->mm;
 	int ret = -EALREADY;
 
 	down_write(&mm->mmap_sem);
 	if (mm->uprobes_state.xol_area)
 		goto fail;
 
-	ret = -ENOMEM;
-	/* Try to map as high as possible, this is only a hint. */
-	area->vaddr = get_unmapped_area(NULL, TASK_SIZE - PAGE_SIZE, PAGE_SIZE, 0, 0);
-	if (area->vaddr & ~PAGE_MASK) {
-		ret = area->vaddr;
-		goto fail;
+	if (!area->vaddr) {
+		/* Try to map as high as possible, this is only a hint. */
+		area->vaddr = get_unmapped_area(NULL, TASK_SIZE - PAGE_SIZE,
+						PAGE_SIZE, 0, 0);
+		if (area->vaddr & ~PAGE_MASK) {
+			ret = area->vaddr;
+			goto fail;
+		}
 	}
 
 	ret = install_special_mapping(mm, area->vaddr, PAGE_SIZE,
@@ -1120,30 +1122,19 @@ static int xol_add_vma(struct xol_area *area)
 
 	smp_wmb();	/* pairs with get_xol_area() */
 	mm->uprobes_state.xol_area = area;
-	ret = 0;
  fail:
 	up_write(&mm->mmap_sem);
 
 	return ret;
 }
 
-/*
- * get_xol_area - Allocate process's xol_area if necessary.
- * This area will be used for storing instructions for execution out of line.
- *
- * Returns the allocated area or NULL.
- */
-static struct xol_area *get_xol_area(void)
+static struct xol_area *__create_xol_area(unsigned long vaddr)
 {
 	struct mm_struct *mm = current->mm;
-	struct xol_area *area;
 	uprobe_opcode_t insn = UPROBE_SWBP_INSN;
+	struct xol_area *area;
 
-	area = mm->uprobes_state.xol_area;
-	if (area)
-		goto ret;
-
-	area = kzalloc(sizeof(*area), GFP_KERNEL);
+	area = kmalloc(sizeof(*area), GFP_KERNEL);
 	if (unlikely(!area))
 		goto out;
 
@@ -1155,13 +1146,14 @@ static struct xol_area *get_xol_area(void)
 	if (!area->page)
 		goto free_bitmap;
 
-	/* allocate first slot of task's xol_area for the return probes */
+	area->vaddr = vaddr;
+	init_waitqueue_head(&area->wq);
+	/* Reserve the 1st slot for get_trampoline_vaddr() */
 	set_bit(0, area->bitmap);
-	copy_to_page(area->page, 0, &insn, UPROBE_SWBP_INSN_SIZE);
 	atomic_set(&area->slot_count, 1);
-	init_waitqueue_head(&area->wq);
+	copy_to_page(area->page, 0, &insn, UPROBE_SWBP_INSN_SIZE);
 
-	if (!xol_add_vma(area))
+	if (!xol_add_vma(mm, area))
 		return area;
 
 	__free_page(area->page);
@@ -1170,9 +1162,25 @@ static struct xol_area *get_xol_area(void)
  free_area:
 	kfree(area);
  out:
+	return NULL;
+}
+
+/*
+ * get_xol_area - Allocate process's xol_area if necessary.
+ * This area will be used for storing instructions for execution out of line.
+ *
+ * Returns the allocated area or NULL.
+ */
+static struct xol_area *get_xol_area(void)
+{
+	struct mm_struct *mm = current->mm;
+	struct xol_area *area;
+
+	if (!mm->uprobes_state.xol_area)
+		__create_xol_area(0);
+
 	area = mm->uprobes_state.xol_area;
- ret:
-	smp_read_barrier_depends();     /* pairs with wmb in xol_add_vma() */
+	smp_read_barrier_depends();	/* pairs with wmb in xol_add_vma() */
 	return area;
 }
 
@@ -1345,14 +1353,6 @@ void uprobe_free_utask(struct task_struct *t)
 }
 
 /*
- * Called in context of a new clone/fork from copy_process.
- */
-void uprobe_copy_process(struct task_struct *t)
-{
-	t->utask = NULL;
-}
-
-/*
  * Allocate a uprobe_task object for the task if if necessary.
  * Called when the thread hits a breakpoint.
  *
@@ -1367,6 +1367,90 @@ static struct uprobe_task *get_utask(void)
 	return current->utask;
 }
 
+static int dup_utask(struct task_struct *t, struct uprobe_task *o_utask)
+{
+	struct uprobe_task *n_utask;
+	struct return_instance **p, *o, *n;
+
+	n_utask = kzalloc(sizeof(struct uprobe_task), GFP_KERNEL);
+	if (!n_utask)
+		return -ENOMEM;
+	t->utask = n_utask;
+
+	p = &n_utask->return_instances;
+	for (o = o_utask->return_instances; o; o = o->next) {
+		n = kmalloc(sizeof(struct return_instance), GFP_KERNEL);
+		if (!n)
+			return -ENOMEM;
+
+		*n = *o;
+		atomic_inc(&n->uprobe->ref);
+		n->next = NULL;
+
+		*p = n;
+		p = &n->next;
+		n_utask->depth++;
+	}
+
+	return 0;
+}
+
+static void uprobe_warn(struct task_struct *t, const char *msg)
+{
+	pr_warn("uprobe: %s:%d failed to %s\n",
+			current->comm, current->pid, msg);
+}
+
+static void dup_xol_work(struct callback_head *work)
+{
+	kfree(work);
+
+	if (current->flags & PF_EXITING)
+		return;
+
+	if (!__create_xol_area(current->utask->vaddr))
+		uprobe_warn(current, "dup xol area");
+}
+
+/*
+ * Called in context of a new clone/fork from copy_process.
+ */
+void uprobe_copy_process(struct task_struct *t, unsigned long flags)
+{
+	struct uprobe_task *utask = current->utask;
+	struct mm_struct *mm = current->mm;
+	struct callback_head *work;
+	struct xol_area *area;
+
+	t->utask = NULL;
+
+	if (!utask || !utask->return_instances)
+		return;
+
+	if (mm == t->mm && !(flags & CLONE_VFORK))
+		return;
+
+	if (dup_utask(t, utask))
+		return uprobe_warn(t, "dup ret instances");
+
+	/* The task can fork() after dup_xol_work() fails */
+	area = mm->uprobes_state.xol_area;
+	if (!area)
+		return uprobe_warn(t, "dup xol area");
+
+	if (mm == t->mm)
+		return;
+
+	/* TODO: move it into the union in uprobe_task */
+	work = kmalloc(sizeof(*work), GFP_KERNEL);
+	if (!work)
+		return uprobe_warn(t, "dup xol area");
+
+	utask->vaddr = area->vaddr;
+	init_task_work(work, dup_xol_work);
+	task_work_add(t, work, true);
+}
+
 /*
  * Current area->vaddr notion assume the trampoline address is always
  * equal area->vaddr.