1 files changed, 290 insertions, 38 deletions

diff --git a/kernel/events/core.c b/kernel/events/core.c
index 5de0b801bc7b..4ff86d57f9e5 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -1941,6 +1941,11 @@ static void perf_put_aux_event(struct perf_event *event)
 	}
 }
 
+static bool perf_need_aux_event(struct perf_event *event)
+{
+	return !!event->attr.aux_output || !!event->attr.aux_sample_size;
+}
+
 static int perf_get_aux_event(struct perf_event *event,
 			      struct perf_event *group_leader)
 {
@@ -1953,7 +1958,17 @@ static int perf_get_aux_event(struct perf_event *event,
 	if (!group_leader)
 		return 0;
 
-	if (!perf_aux_output_match(event, group_leader))
+	/*
+	 * aux_output and aux_sample_size are mutually exclusive.
+	 */
+	if (event->attr.aux_output && event->attr.aux_sample_size)
+		return 0;
+
+	if (event->attr.aux_output &&
+	    !perf_aux_output_match(event, group_leader))
+		return 0;
+
+	if (event->attr.aux_sample_size && !group_leader->pmu->snapshot_aux)
 		return 0;
 
 	if (!atomic_long_inc_not_zero(&group_leader->refcount))
@@ -2666,6 +2681,25 @@ perf_install_in_context(struct perf_event_context *ctx,
 	 */
 	smp_store_release(&event->ctx, ctx);
 
+	/*
+	 * perf_event_attr::disabled events will not run and can be initialized
+	 * without IPI. Except when this is the first event for the context, in
+	 * that case we need the magic of the IPI to set ctx->is_active.
+	 *
+	 * The IOC_ENABLE that is sure to follow the creation of a disabled
+	 * event will issue the IPI and reprogram the hardware.
+	 */
+	if (__perf_effective_state(event) == PERF_EVENT_STATE_OFF && ctx->nr_events) {
+		raw_spin_lock_irq(&ctx->lock);
+		if (ctx->task == TASK_TOMBSTONE) {
+			raw_spin_unlock_irq(&ctx->lock);
+			return;
+		}
+		add_event_to_ctx(event, ctx);
+		raw_spin_unlock_irq(&ctx->lock);
+		return;
+	}
+
 	if (!task) {
 		cpu_function_call(cpu, __perf_install_in_context, event);
 		return;
@@ -3204,10 +3238,21 @@ static void perf_event_context_sched_out(struct task_struct *task, int ctxn,
 		raw_spin_lock(&ctx->lock);
 		raw_spin_lock_nested(&next_ctx->lock, SINGLE_DEPTH_NESTING);
 		if (context_equiv(ctx, next_ctx)) {
+			struct pmu *pmu = ctx->pmu;
+
 			WRITE_ONCE(ctx->task, next);
 			WRITE_ONCE(next_ctx->task, task);
 
-			swap(ctx->task_ctx_data, next_ctx->task_ctx_data);
+			/*
+			 * PMU specific parts of task perf context can require
+			 * additional synchronization. As an example of such
+			 * synchronization see implementation details of Intel
+			 * LBR call stack data profiling;
+			 */
+			if (pmu->swap_task_ctx)
+				pmu->swap_task_ctx(ctx, next_ctx);
+			else
+				swap(ctx->task_ctx_data, next_ctx->task_ctx_data);
 
 			/*
 			 * RCU_INIT_POINTER here is safe because we've not
@@ -4229,8 +4274,9 @@ find_get_context(struct pmu *pmu, struct task_struct *task,
 
 	if (!task) {
 		/* Must be root to operate on a CPU event: */
-		if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN))
-			return ERR_PTR(-EACCES);
+		err = perf_allow_cpu(&event->attr);
+		if (err)
+			return ERR_PTR(err);
 
 		cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu);
 		ctx = &cpuctx->ctx;
@@ -4539,6 +4585,8 @@ static void _free_event(struct perf_event *event)
 
 	unaccount_event(event);
 
+	security_perf_event_free(event);
+
 	if (event->rb) {
 		/*
 		 * Can happen when we close an event with re-directed output.
@@ -4992,6 +5040,10 @@ perf_read(struct file *file, char __user *buf, size_t count, loff_t *ppos)
 	struct perf_event_context *ctx;
 	int ret;
 
+	ret = security_perf_event_read(event);
+	if (ret)
+		return ret;
+
 	ctx = perf_event_ctx_lock(event);
 	ret = __perf_read(event, buf, count);
 	perf_event_ctx_unlock(event, ctx);
@@ -5288,6 +5340,11 @@ static long perf_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	struct perf_event_context *ctx;
 	long ret;
 
+	/* Treat ioctl like writes as it is likely a mutating operation. */
+	ret = security_perf_event_write(event);
+	if (ret)
+		return ret;
+
 	ctx = perf_event_ctx_lock(event);
 	ret = _perf_ioctl(event, cmd, arg);
 	perf_event_ctx_unlock(event, ctx);
@@ -5639,10 +5696,8 @@ static void perf_mmap_close(struct vm_area_struct *vma)
 		perf_pmu_output_stop(event);
 
 		/* now it's safe to free the pages */
-		if (!rb->aux_mmap_locked)
-			atomic_long_sub(rb->aux_nr_pages, &mmap_user->locked_vm);
-		else
-			atomic64_sub(rb->aux_mmap_locked, &vma->vm_mm->pinned_vm);
+		atomic_long_sub(rb->aux_nr_pages - rb->aux_mmap_locked, &mmap_user->locked_vm);
+		atomic64_sub(rb->aux_mmap_locked, &vma->vm_mm->pinned_vm);
 
 		/* this has to be the last one */
 		rb_free_aux(rb);
@@ -5753,6 +5808,10 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma)
 	if (!(vma->vm_flags & VM_SHARED))
 		return -EINVAL;
 
+	ret = security_perf_event_read(event);
+	if (ret)
+		return ret;
+
 	vma_size = vma->vm_end - vma->vm_start;
 
 	if (vma->vm_pgoff == 0) {
@@ -5859,13 +5918,7 @@ accounting:
 
 	user_locked = atomic_long_read(&user->locked_vm) + user_extra;
 
-	if (user_locked <= user_lock_limit) {
-		/* charge all to locked_vm */
-	} else if (atomic_long_read(&user->locked_vm) >= user_lock_limit) {
-		/* charge all to pinned_vm */
-		extra = user_extra;
-		user_extra = 0;
-	} else {
+	if (user_locked > user_lock_limit) {
 		/*
 		 * charge locked_vm until it hits user_lock_limit;
 		 * charge the rest from pinned_vm
@@ -5878,7 +5931,7 @@ accounting:
 	lock_limit >>= PAGE_SHIFT;
 	locked = atomic64_read(&vma->vm_mm->pinned_vm) + extra;
 
-	if ((locked > lock_limit) && perf_paranoid_tracepoint_raw() &&
+	if ((locked > lock_limit) && perf_is_paranoid() &&
 		!capable(CAP_IPC_LOCK)) {
 		ret = -EPERM;
 		goto unlock;
@@ -6208,6 +6261,122 @@ perf_output_sample_ustack(struct perf_output_handle *handle, u64 dump_size,
 	}
 }
 
+static unsigned long perf_prepare_sample_aux(struct perf_event *event,
+					  struct perf_sample_data *data,
+					  size_t size)
+{
+	struct perf_event *sampler = event->aux_event;
+	struct ring_buffer *rb;
+
+	data->aux_size = 0;
+
+	if (!sampler)
+		goto out;
+
+	if (WARN_ON_ONCE(READ_ONCE(sampler->state) != PERF_EVENT_STATE_ACTIVE))
+		goto out;
+
+	if (WARN_ON_ONCE(READ_ONCE(sampler->oncpu) != smp_processor_id()))
+		goto out;
+
+	rb = ring_buffer_get(sampler->parent ? sampler->parent : sampler);
+	if (!rb)
+		goto out;
+
+	/*
+	 * If this is an NMI hit inside sampling code, don't take
+	 * the sample. See also perf_aux_sample_output().
+	 */
+	if (READ_ONCE(rb->aux_in_sampling)) {
+		data->aux_size = 0;
+	} else {
+		size = min_t(size_t, size, perf_aux_size(rb));
+		data->aux_size = ALIGN(size, sizeof(u64));
+	}
+	ring_buffer_put(rb);
+
+out:
+	return data->aux_size;
+}
+
+long perf_pmu_snapshot_aux(struct ring_buffer *rb,
+			   struct perf_event *event,
+			   struct perf_output_handle *handle,
+			   unsigned long size)
+{
+	unsigned long flags;
+	long ret;
+
+	/*
+	 * Normal ->start()/->stop() callbacks run in IRQ mode in scheduler
+	 * paths. If we start calling them in NMI context, they may race with
+	 * the IRQ ones, that is, for example, re-starting an event that's just
+	 * been stopped, which is why we're using a separate callback that
+	 * doesn't change the event state.
+	 *
+	 * IRQs need to be disabled to prevent IPIs from racing with us.
+	 */
+	local_irq_save(flags);
+	/*
+	 * Guard against NMI hits inside the critical section;
+	 * see also perf_prepare_sample_aux().
+	 */
+	WRITE_ONCE(rb->aux_in_sampling, 1);
+	barrier();
+
+	ret = event->pmu->snapshot_aux(event, handle, size);
+
+	barrier();
+	WRITE_ONCE(rb->aux_in_sampling, 0);
+	local_irq_restore(flags);
+
+	return ret;
+}
+
+static void perf_aux_sample_output(struct perf_event *event,
+				   struct perf_output_handle *handle,
+				   struct perf_sample_data *data)
+{
+	struct perf_event *sampler = event->aux_event;
+	unsigned long pad;
+	struct ring_buffer *rb;
+	long size;
+
+	if (WARN_ON_ONCE(!sampler || !data->aux_size))
+		return;
+
+	rb = ring_buffer_get(sampler->parent ? sampler->parent : sampler);
+	if (!rb)
+		return;
+
+	size = perf_pmu_snapshot_aux(rb, sampler, handle, data->aux_size);
+
+	/*
+	 * An error here means that perf_output_copy() failed (returned a
+	 * non-zero surplus that it didn't copy), which in its current
+	 * enlightened implementation is not possible. If that changes, we'd
+	 * like to know.
+	 */
+	if (WARN_ON_ONCE(size < 0))
+		goto out_put;
+
+	/*
+	 * The pad comes from ALIGN()ing data->aux_size up to u64 in
+	 * perf_prepare_sample_aux(), so should not be more than that.
+	 */
+	pad = data->aux_size - size;
+	if (WARN_ON_ONCE(pad >= sizeof(u64)))
+		pad = 8;
+
+	if (pad) {
+		u64 zero = 0;
+		perf_output_copy(handle, &zero, pad);
+	}
+
+out_put:
+	ring_buffer_put(rb);
+}
+
 static void __perf_event_header__init_id(struct perf_event_header *header,
 					 struct perf_sample_data *data,
 					 struct perf_event *event)
@@ -6527,6 +6696,13 @@ void perf_output_sample(struct perf_output_handle *handle,
 	if (sample_type & PERF_SAMPLE_PHYS_ADDR)
 		perf_output_put(handle, data->phys_addr);
 
+	if (sample_type & PERF_SAMPLE_AUX) {
+		perf_output_put(handle, data->aux_size);
+
+		if (data->aux_size)
+			perf_aux_sample_output(event, handle, data);
+	}
+
 	if (!event->attr.watermark) {
 		int wakeup_events = event->attr.wakeup_events;
 
@@ -6715,6 +6891,35 @@ void perf_prepare_sample(struct perf_event_header *header,
 
 	if (sample_type & PERF_SAMPLE_PHYS_ADDR)
 		data->phys_addr = perf_virt_to_phys(data->addr);
+
+	if (sample_type & PERF_SAMPLE_AUX) {
+		u64 size;
+
+		header->size += sizeof(u64); /* size */
+
+		/*
+		 * Given the 16bit nature of header::size, an AUX sample can
+		 * easily overflow it, what with all the preceding sample bits.
+		 * Make sure this doesn't happen by using up to U16_MAX bytes
+		 * per sample in total (rounded down to 8 byte boundary).
+		 */
+		size = min_t(size_t, U16_MAX - header->size,
+			     event->attr.aux_sample_size);
+		size = rounddown(size, 8);
+		size = perf_prepare_sample_aux(event, data, size);
+
+		WARN_ON_ONCE(size + header->size > U16_MAX);
+		header->size += size;
+	}
+	/*
+	 * If you're adding more sample types here, you likely need to do
+	 * something about the overflowing header::size, like repurpose the
+	 * lowest 3 bits of size, which should be always zero at the moment.
+	 * This raises a more important question, do we really need 512k sized
+	 * samples and why, so good argumentation is in order for whatever you
+	 * do here next.
+	 */
+	WARN_ON_ONCE(header->size & 7);
 }
 
 static __always_inline int
@@ -10066,7 +10271,7 @@ static struct lock_class_key cpuctx_lock;
 
 int perf_pmu_register(struct pmu *pmu, const char *name, int type)
 {
-	int cpu, ret;
+	int cpu, ret, max = PERF_TYPE_MAX;
 
 	mutex_lock(&pmus_lock);
 	ret = -ENOMEM;
@@ -10079,12 +10284,17 @@ int perf_pmu_register(struct pmu *pmu, const char *name, int type)
 		goto skip_type;
 	pmu->name = name;
 
-	if (type < 0) {
-		type = idr_alloc(&pmu_idr, pmu, PERF_TYPE_MAX, 0, GFP_KERNEL);
-		if (type < 0) {
-			ret = type;
+	if (type != PERF_TYPE_SOFTWARE) {
+		if (type >= 0)
+			max = type;
+
+		ret = idr_alloc(&pmu_idr, pmu, max, 0, GFP_KERNEL);
+		if (ret < 0)
 			goto free_pdc;
-		}
+
+		WARN_ON(type >= 0 && ret != type);
+
+		type = ret;
 	}
 	pmu->type = type;
 
@@ -10161,7 +10371,16 @@ got_cpu_context:
 	if (!pmu->event_idx)
 		pmu->event_idx = perf_event_idx_default;
 
-	list_add_rcu(&pmu->entry, &pmus);
+	/*
+	 * Ensure the TYPE_SOFTWARE PMUs are at the head of the list,
+	 * since these cannot be in the IDR. This way the linear search
+	 * is fast, provided a valid software event is provided.
+	 */
+	if (type == PERF_TYPE_SOFTWARE || !name)
+		list_add_rcu(&pmu->entry, &pmus);
+	else
+		list_add_tail_rcu(&pmu->entry, &pmus);
+
 	atomic_set(&pmu->exclusive_cnt, 0);
 	ret = 0;
 unlock:
@@ -10174,7 +10393,7 @@ free_dev:
 	put_device(pmu->dev);
 
 free_idr:
-	if (pmu->type >= PERF_TYPE_MAX)
+	if (pmu->type != PERF_TYPE_SOFTWARE)
 		idr_remove(&pmu_idr, pmu->type);
 
 free_pdc:
@@ -10196,7 +10415,7 @@ void perf_pmu_unregister(struct pmu *pmu)
 	synchronize_rcu();
 
 	free_percpu(pmu->pmu_disable_count);
-	if (pmu->type >= PERF_TYPE_MAX)
+	if (pmu->type != PERF_TYPE_SOFTWARE)
 		idr_remove(&pmu_idr, pmu->type);
 	if (pmu_bus_running) {
 		if (pmu->nr_addr_filters)
@@ -10266,9 +10485,8 @@ static int perf_try_init_event(struct pmu *pmu, struct perf_event *event)
 
 static struct pmu *perf_init_event(struct perf_event *event)
 {
+	int idx, type, ret;
 	struct pmu *pmu;
-	int idx;
-	int ret;
 
 	idx = srcu_read_lock(&pmus_srcu);
 
@@ -10280,13 +10498,28 @@ static struct pmu *perf_init_event(struct perf_event *event)
 			goto unlock;
 	}
 
+	/*
+	 * PERF_TYPE_HARDWARE and PERF_TYPE_HW_CACHE
+	 * are often aliases for PERF_TYPE_RAW.
+	 */
+	type = event->attr.type;
+	if (type == PERF_TYPE_HARDWARE || type == PERF_TYPE_HW_CACHE)
+		type = PERF_TYPE_RAW;
+
+again:
 	rcu_read_lock();
-	pmu = idr_find(&pmu_idr, event->attr.type);
+	pmu = idr_find(&pmu_idr, type);
 	rcu_read_unlock();
 	if (pmu) {
 		ret = perf_try_init_event(pmu, event);
+		if (ret == -ENOENT && event->attr.type != type) {
+			type = event->attr.type;
+			goto again;
+		}
+
 		if (ret)
 			pmu = ERR_PTR(ret);
+
 		goto unlock;
 	}
 
@@ -10618,11 +10851,20 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu,
 		}
 	}
 
+	err = security_perf_event_alloc(event);
+	if (err)
+		goto err_callchain_buffer;
+
 	/* symmetric to unaccount_event() in _free_event() */
 	account_event(event);
 
 	return event;
 
+err_callchain_buffer:
+	if (!event->parent) {
+		if (event->attr.sample_type & PERF_SAMPLE_CALLCHAIN)
+			put_callchain_buffers();
+	}
 err_addr_filters:
 	kfree(event->addr_filter_ranges);
 
@@ -10673,7 +10915,7 @@ static int perf_copy_attr(struct perf_event_attr __user *uattr,
 
 	attr->size = size;
 
-	if (attr->__reserved_1 || attr->__reserved_2)
+	if (attr->__reserved_1 || attr->__reserved_2 || attr->__reserved_3)
 		return -EINVAL;
 
 	if (attr->sample_type & ~(PERF_SAMPLE_MAX-1))
@@ -10711,9 +10953,11 @@ static int perf_copy_attr(struct perf_event_attr __user *uattr,
 			attr->branch_sample_type = mask;
 		}
 		/* privileged levels capture (kernel, hv): check permissions */
-		if ((mask & PERF_SAMPLE_BRANCH_PERM_PLM)
-		    && perf_paranoid_kernel() && !capable(CAP_SYS_ADMIN))
-			return -EACCES;
+		if (mask & PERF_SAMPLE_BRANCH_PERM_PLM) {
+			ret = perf_allow_kernel(attr);
+			if (ret)
+				return ret;
+		}
 	}
 
 	if (attr->sample_type & PERF_SAMPLE_REGS_USER) {
@@ -10926,13 +11170,19 @@ SYSCALL_DEFINE5(perf_event_open,
 	if (flags & ~PERF_FLAG_ALL)
 		return -EINVAL;
 
+	/* Do we allow access to perf_event_open(2) ? */
+	err = security_perf_event_open(&attr, PERF_SECURITY_OPEN);
+	if (err)
+		return err;
+
 	err = perf_copy_attr(attr_uptr, &attr);
 	if (err)
 		return err;
 
 	if (!attr.exclude_kernel) {
-		if (perf_paranoid_kernel() && !capable(CAP_SYS_ADMIN))
-			return -EACCES;
+		err = perf_allow_kernel(&attr);
+		if (err)
+			return err;
 	}
 
 	if (attr.namespaces) {
@@ -10949,9 +11199,11 @@ SYSCALL_DEFINE5(perf_event_open,
 	}
 
 	/* Only privileged users can get physical addresses */
-	if ((attr.sample_type & PERF_SAMPLE_PHYS_ADDR) &&
-	    perf_paranoid_kernel() && !capable(CAP_SYS_ADMIN))
-		return -EACCES;
+	if ((attr.sample_type & PERF_SAMPLE_PHYS_ADDR)) {
+		err = perf_allow_kernel(&attr);
+		if (err)
+			return err;
+	}
 
 	err = security_locked_down(LOCKDOWN_PERF);
 	if (err && (attr.sample_type & PERF_SAMPLE_REGS_INTR))
@@ -11213,7 +11465,7 @@ SYSCALL_DEFINE5(perf_event_open,
 		}
 	}
 
-	if (event->attr.aux_output && !perf_get_aux_event(event, group_leader))
+	if (perf_need_aux_event(event) && !perf_get_aux_event(event, group_leader))
 		goto err_locked;
 
 	/*