1 files changed, 270 insertions, 56 deletions

diff --git a/kernel/events/core.c b/kernel/events/core.c
index 7fee567153f0..301079d06f24 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -36,6 +36,7 @@
 #include <linux/perf_event.h>
 #include <linux/ftrace_event.h>
 #include <linux/hw_breakpoint.h>
+#include <linux/mm_types.h>
 
 #include "internal.h"
 
@@ -371,6 +372,8 @@ void perf_cgroup_switch(struct task_struct *task, int mode)
 
 	list_for_each_entry_rcu(pmu, &pmus, entry) {
 		cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
+		if (cpuctx->unique_pmu != pmu)
+			continue; /* ensure we process each cpuctx once */
 
 		/*
 		 * perf_cgroup_events says at least one
@@ -394,9 +397,10 @@ void perf_cgroup_switch(struct task_struct *task, int mode)
 
 			if (mode & PERF_CGROUP_SWIN) {
 				WARN_ON_ONCE(cpuctx->cgrp);
-				/* set cgrp before ctxsw in to
-				 * allow event_filter_match() to not
-				 * have to pass task around
+				/*
+				 * set cgrp before ctxsw in to allow
+				 * event_filter_match() to not have to pass
+				 * task around
 				 */
 				cpuctx->cgrp = perf_cgroup_from_task(task);
 				cpu_ctx_sched_in(cpuctx, EVENT_ALL, task);
@@ -467,14 +471,13 @@ static inline int perf_cgroup_connect(int fd, struct perf_event *event,
 {
 	struct perf_cgroup *cgrp;
 	struct cgroup_subsys_state *css;
-	struct file *file;
-	int ret = 0, fput_needed;
+	struct fd f = fdget(fd);
+	int ret = 0;
 
-	file = fget_light(fd, &fput_needed);
-	if (!file)
+	if (!f.file)
 		return -EBADF;
 
-	css = cgroup_css_from_dir(file, perf_subsys_id);
+	css = cgroup_css_from_dir(f.file, perf_subsys_id);
 	if (IS_ERR(css)) {
 		ret = PTR_ERR(css);
 		goto out;
@@ -500,7 +503,7 @@ static inline int perf_cgroup_connect(int fd, struct perf_event *event,
 		ret = -EINVAL;
 	}
 out:
-	fput_light(file, fput_needed);
+	fdput(f);
 	return ret;
 }
 
@@ -3233,21 +3236,18 @@ unlock:
 
 static const struct file_operations perf_fops;
 
-static struct file *perf_fget_light(int fd, int *fput_needed)
+static inline int perf_fget_light(int fd, struct fd *p)
 {
-	struct file *file;
-
-	file = fget_light(fd, fput_needed);
-	if (!file)
-		return ERR_PTR(-EBADF);
+	struct fd f = fdget(fd);
+	if (!f.file)
+		return -EBADF;
 
-	if (file->f_op != &perf_fops) {
-		fput_light(file, *fput_needed);
-		*fput_needed = 0;
-		return ERR_PTR(-EBADF);
+	if (f.file->f_op != &perf_fops) {
+		fdput(f);
+		return -EBADF;
 	}
-
-	return file;
+	*p = f;
+	return 0;
 }
 
 static int perf_event_set_output(struct perf_event *event,
@@ -3279,22 +3279,19 @@ static long perf_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 
 	case PERF_EVENT_IOC_SET_OUTPUT:
 	{
-		struct file *output_file = NULL;
-		struct perf_event *output_event = NULL;
-		int fput_needed = 0;
 		int ret;
-
 		if (arg != -1) {
-			output_file = perf_fget_light(arg, &fput_needed);
-			if (IS_ERR(output_file))
-				return PTR_ERR(output_file);
-			output_event = output_file->private_data;
+			struct perf_event *output_event;
+			struct fd output;
+			ret = perf_fget_light(arg, &output);
+			if (ret)
+				return ret;
+			output_event = output.file->private_data;
+			ret = perf_event_set_output(event, output_event);
+			fdput(output);
+		} else {
+			ret = perf_event_set_output(event, NULL);
 		}
-
-		ret = perf_event_set_output(event, output_event);
-		if (output_event)
-			fput_light(output_file, fput_needed);
-
 		return ret;
 	}
 
@@ -3677,7 +3674,7 @@ unlock:
 		atomic_inc(&event->mmap_count);
 	mutex_unlock(&event->mmap_mutex);
 
-	vma->vm_flags |= VM_RESERVED;
+	vma->vm_flags |= VM_DONTEXPAND | VM_DONTDUMP;
 	vma->vm_ops = &perf_mmap_vmops;
 
 	return ret;
@@ -3764,6 +3761,132 @@ int perf_unregister_guest_info_callbacks(struct perf_guest_info_callbacks *cbs)
 }
 EXPORT_SYMBOL_GPL(perf_unregister_guest_info_callbacks);
 
+static void
+perf_output_sample_regs(struct perf_output_handle *handle,
+			struct pt_regs *regs, u64 mask)
+{
+	int bit;
+
+	for_each_set_bit(bit, (const unsigned long *) &mask,
+			 sizeof(mask) * BITS_PER_BYTE) {
+		u64 val;
+
+		val = perf_reg_value(regs, bit);
+		perf_output_put(handle, val);
+	}
+}
+
+static void perf_sample_regs_user(struct perf_regs_user *regs_user,
+				  struct pt_regs *regs)
+{
+	if (!user_mode(regs)) {
+		if (current->mm)
+			regs = task_pt_regs(current);
+		else
+			regs = NULL;
+	}
+
+	if (regs) {
+		regs_user->regs = regs;
+		regs_user->abi  = perf_reg_abi(current);
+	}
+}
+
+/*
+ * Get remaining task size from user stack pointer.
+ *
+ * It'd be better to take stack vma map and limit this more
+ * precisly, but there's no way to get it safely under interrupt,
+ * so using TASK_SIZE as limit.
+ */
+static u64 perf_ustack_task_size(struct pt_regs *regs)
+{
+	unsigned long addr = perf_user_stack_pointer(regs);
+
+	if (!addr || addr >= TASK_SIZE)
+		return 0;
+
+	return TASK_SIZE - addr;
+}
+
+static u16
+perf_sample_ustack_size(u16 stack_size, u16 header_size,
+			struct pt_regs *regs)
+{
+	u64 task_size;
+
+	/* No regs, no stack pointer, no dump. */
+	if (!regs)
+		return 0;
+
+	/*
+	 * Check if we fit in with the requested stack size into the:
+	 * - TASK_SIZE
+	 *   If we don't, we limit the size to the TASK_SIZE.
+	 *
+	 * - remaining sample size
+	 *   If we don't, we customize the stack size to
+	 *   fit in to the remaining sample size.
+	 */
+
+	task_size  = min((u64) USHRT_MAX, perf_ustack_task_size(regs));
+	stack_size = min(stack_size, (u16) task_size);
+
+	/* Current header size plus static size and dynamic size. */
+	header_size += 2 * sizeof(u64);
+
+	/* Do we fit in with the current stack dump size? */
+	if ((u16) (header_size + stack_size) < header_size) {
+		/*
+		 * If we overflow the maximum size for the sample,
+		 * we customize the stack dump size to fit in.
+		 */
+		stack_size = USHRT_MAX - header_size - sizeof(u64);
+		stack_size = round_up(stack_size, sizeof(u64));
+	}
+
+	return stack_size;
+}
+
+static void
+perf_output_sample_ustack(struct perf_output_handle *handle, u64 dump_size,
+			  struct pt_regs *regs)
+{
+	/* Case of a kernel thread, nothing to dump */
+	if (!regs) {
+		u64 size = 0;
+		perf_output_put(handle, size);
+	} else {
+		unsigned long sp;
+		unsigned int rem;
+		u64 dyn_size;
+
+		/*
+		 * We dump:
+		 * static size
+		 *   - the size requested by user or the best one we can fit
+		 *     in to the sample max size
+		 * data
+		 *   - user stack dump data
+		 * dynamic size
+		 *   - the actual dumped size
+		 */
+
+		/* Static size. */
+		perf_output_put(handle, dump_size);
+
+		/* Data. */
+		sp = perf_user_stack_pointer(regs);
+		rem = __output_copy_user(handle, (void *) sp, dump_size);
+		dyn_size = dump_size - rem;
+
+		perf_output_skip(handle, rem);
+
+		/* Dynamic size. */
+		perf_output_put(handle, dyn_size);
+	}
+}
+
 static void __perf_event_header__init_id(struct perf_event_header *header,
 					 struct perf_sample_data *data,
 					 struct perf_event *event)
@@ -4024,6 +4147,28 @@ void perf_output_sample(struct perf_output_handle *handle,
 			perf_output_put(handle, nr);
 		}
 	}
+
+	if (sample_type & PERF_SAMPLE_REGS_USER) {
+		u64 abi = data->regs_user.abi;
+
+		/*
+		 * If there are no regs to dump, notice it through
+		 * first u64 being zero (PERF_SAMPLE_REGS_ABI_NONE).
+		 */
+		perf_output_put(handle, abi);
+
+		if (abi) {
+			u64 mask = event->attr.sample_regs_user;
+			perf_output_sample_regs(handle,
+						data->regs_user.regs,
+						mask);
+		}
+	}
+
+	if (sample_type & PERF_SAMPLE_STACK_USER)
+		perf_output_sample_ustack(handle,
+					  data->stack_user_size,
+					  data->regs_user.regs);
 }
 
 void perf_prepare_sample(struct perf_event_header *header,
@@ -4075,6 +4220,49 @@ void perf_prepare_sample(struct perf_event_header *header,
 		}
 		header->size += size;
 	}
+
+	if (sample_type & PERF_SAMPLE_REGS_USER) {
+		/* regs dump ABI info */
+		int size = sizeof(u64);
+
+		perf_sample_regs_user(&data->regs_user, regs);
+
+		if (data->regs_user.regs) {
+			u64 mask = event->attr.sample_regs_user;
+			size += hweight64(mask) * sizeof(u64);
+		}
+
+		header->size += size;
+	}
+
+	if (sample_type & PERF_SAMPLE_STACK_USER) {
+		/*
+		 * Either we need PERF_SAMPLE_STACK_USER bit to be allways
+		 * processed as the last one or have additional check added
+		 * in case new sample type is added, because we could eat
+		 * up the rest of the sample size.
+		 */
+		struct perf_regs_user *uregs = &data->regs_user;
+		u16 stack_size = event->attr.sample_stack_user;
+		u16 size = sizeof(u64);
+
+		if (!uregs->abi)
+			perf_sample_regs_user(uregs, regs);
+
+		stack_size = perf_sample_ustack_size(stack_size, header->size,
+						     uregs->regs);
+
+		/*
+		 * If there is something to dump, add space for the dump
+		 * itself and for the field that tells the dynamic size,
+		 * which is how many have been actually dumped.
+		 */
+		if (stack_size)
+			size += sizeof(u64) + stack_size;
+
+		data->stack_user_size = stack_size;
+		header->size += size;
+	}
 }
 
 static void perf_event_output(struct perf_event *event,
@@ -4227,7 +4415,7 @@ static void perf_event_task_event(struct perf_task_event *task_event)
 	rcu_read_lock();
 	list_for_each_entry_rcu(pmu, &pmus, entry) {
 		cpuctx = get_cpu_ptr(pmu->pmu_cpu_context);
-		if (cpuctx->active_pmu != pmu)
+		if (cpuctx->unique_pmu != pmu)
 			goto next;
 		perf_event_task_ctx(&cpuctx->ctx, task_event);
 
@@ -4373,7 +4561,7 @@ static void perf_event_comm_event(struct perf_comm_event *comm_event)
 	rcu_read_lock();
 	list_for_each_entry_rcu(pmu, &pmus, entry) {
 		cpuctx = get_cpu_ptr(pmu->pmu_cpu_context);
-		if (cpuctx->active_pmu != pmu)
+		if (cpuctx->unique_pmu != pmu)
 			goto next;
 		perf_event_comm_ctx(&cpuctx->ctx, comm_event);
 
@@ -4569,7 +4757,7 @@ got_name:
 	rcu_read_lock();
 	list_for_each_entry_rcu(pmu, &pmus, entry) {
 		cpuctx = get_cpu_ptr(pmu->pmu_cpu_context);
-		if (cpuctx->active_pmu != pmu)
+		if (cpuctx->unique_pmu != pmu)
 			goto next;
 		perf_event_mmap_ctx(&cpuctx->ctx, mmap_event,
 					vma->vm_flags & VM_EXEC);
@@ -5670,8 +5858,8 @@ static void update_pmu_context(struct pmu *pmu, struct pmu *old_pmu)
 
 		cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu);
 
-		if (cpuctx->active_pmu == old_pmu)
-			cpuctx->active_pmu = pmu;
+		if (cpuctx->unique_pmu == old_pmu)
+			cpuctx->unique_pmu = pmu;
 	}
 }
 
@@ -5806,7 +5994,7 @@ skip_type:
 		cpuctx->ctx.pmu = pmu;
 		cpuctx->jiffies_interval = 1;
 		INIT_LIST_HEAD(&cpuctx->rotation_list);
-		cpuctx->active_pmu = pmu;
+		cpuctx->unique_pmu = pmu;
 	}
 
 got_cpu_context:
@@ -5967,7 +6155,7 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu,
 
 	event->parent		= parent_event;
 
-	event->ns		= get_pid_ns(current->nsproxy->pid_ns);
+	event->ns		= get_pid_ns(task_active_pid_ns(current));
 	event->id		= atomic64_inc_return(&perf_event_id);
 
 	event->state		= PERF_EVENT_STATE_INACTIVE;
@@ -6151,6 +6339,28 @@ static int perf_copy_attr(struct perf_event_attr __user *uattr,
 			attr->branch_sample_type = mask;
 		}
 	}
+
+	if (attr->sample_type & PERF_SAMPLE_REGS_USER) {
+		ret = perf_reg_validate(attr->sample_regs_user);
+		if (ret)
+			return ret;
+	}
+
+	if (attr->sample_type & PERF_SAMPLE_STACK_USER) {
+		if (!arch_perf_have_user_stack_dump())
+			return -ENOSYS;
+
+		/*
+		 * We have __u32 type for the size, but so far
+		 * we can only use __u16 as maximum due to the
+		 * __u16 sample size limit.
+		 */
+		if (attr->sample_stack_user >= USHRT_MAX)
+			ret = -EINVAL;
+		else if (!IS_ALIGNED(attr->sample_stack_user, sizeof(u64)))
+			ret = -EINVAL;
+	}
+
 out:
 	return ret;
 
@@ -6229,12 +6439,11 @@ SYSCALL_DEFINE5(perf_event_open,
 	struct perf_event_attr attr;
 	struct perf_event_context *ctx;
 	struct file *event_file = NULL;
-	struct file *group_file = NULL;
+	struct fd group = {NULL, 0};
 	struct task_struct *task = NULL;
 	struct pmu *pmu;
 	int event_fd;
 	int move_group = 0;
-	int fput_needed = 0;
 	int err;
 
 	/* for future expandability... */
@@ -6264,17 +6473,15 @@ SYSCALL_DEFINE5(perf_event_open,
 	if ((flags & PERF_FLAG_PID_CGROUP) && (pid == -1 || cpu == -1))
 		return -EINVAL;
 
-	event_fd = get_unused_fd_flags(O_RDWR);
+	event_fd = get_unused_fd();
 	if (event_fd < 0)
 		return event_fd;
 
 	if (group_fd != -1) {
-		group_file = perf_fget_light(group_fd, &fput_needed);
-		if (IS_ERR(group_file)) {
-			err = PTR_ERR(group_file);
+		err = perf_fget_light(group_fd, &group);
+		if (err)
 			goto err_fd;
-		}
-		group_leader = group_file->private_data;
+		group_leader = group.file->private_data;
 		if (flags & PERF_FLAG_FD_OUTPUT)
 			output_event = group_leader;
 		if (flags & PERF_FLAG_FD_NO_GROUP)
@@ -6450,7 +6657,7 @@ SYSCALL_DEFINE5(perf_event_open,
 	 * of the group leader will find the pointer to itself in
 	 * perf_group_detach().
 	 */
-	fput_light(group_file, fput_needed);
+	fdput(group);
 	fd_install(event_fd, event_file);
 	return event_fd;
 
@@ -6464,7 +6671,7 @@ err_task:
 	if (task)
 		put_task_struct(task);
 err_group_fd:
-	fput_light(group_file, fput_needed);
+	fdput(group);
 err_fd:
 	put_unused_fd(event_fd);
 	return err;
@@ -7227,7 +7434,7 @@ unlock:
 device_initcall(perf_event_sysfs_init);
 
 #ifdef CONFIG_CGROUP_PERF
-static struct cgroup_subsys_state *perf_cgroup_create(struct cgroup *cont)
+static struct cgroup_subsys_state *perf_cgroup_css_alloc(struct cgroup *cont)
 {
 	struct perf_cgroup *jc;
 
@@ -7244,7 +7451,7 @@ static struct cgroup_subsys_state *perf_cgroup_create(struct cgroup *cont)
 	return &jc->css;
 }
 
-static void perf_cgroup_destroy(struct cgroup *cont)
+static void perf_cgroup_css_free(struct cgroup *cont)
 {
 	struct perf_cgroup *jc;
 	jc = container_of(cgroup_subsys_state(cont, perf_subsys_id),
@@ -7285,9 +7492,16 @@ static void perf_cgroup_exit(struct cgroup *cgrp, struct cgroup *old_cgrp,
 struct cgroup_subsys perf_subsys = {
 	.name		= "perf_event",
 	.subsys_id	= perf_subsys_id,
-	.create		= perf_cgroup_create,
-	.destroy	= perf_cgroup_destroy,
+	.css_alloc	= perf_cgroup_css_alloc,
+	.css_free	= perf_cgroup_css_free,
 	.exit		= perf_cgroup_exit,
 	.attach		= perf_cgroup_attach,
+
+	/*
+	 * perf_event cgroup doesn't handle nesting correctly.
+	 * ctx->nr_cgroups adjustments should be propagated through the
+	 * cgroup hierarchy.  Fix it and remove the following.
+	 */
+	.broken_hierarchy = true,
 };
 #endif /* CONFIG_CGROUP_PERF */