Merge branch 'linus' into sched/core, to pick up fixes

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

kernel/cgroup.c

@@ -6270,6 +6270,12 @@ void cgroup_sk_alloc(struct sock_cgroup_data *skcd)
 	if (cgroup_sk_alloc_disabled)
 		return;
 
+	/* Socket clone path */
+	if (skcd->val) {
+		cgroup_get(sock_cgroup_ptr(skcd));
+		return;
+	}
+
 	rcu_read_lock();
 
 	while (true) {

kernel/events/core.c

@@ -2496,11 +2496,11 @@ static int __perf_event_stop(void *info)
 	return 0;
 }
 
-static int perf_event_restart(struct perf_event *event)
+static int perf_event_stop(struct perf_event *event, int restart)
 {
 	struct stop_event_data sd = {
 		.event		= event,
-		.restart	= 1,
+		.restart	= restart,
 	};
 	int ret = 0;
 
@@ -3549,10 +3549,18 @@ static int perf_event_read(struct perf_event *event, bool group)
 			.group = group,
 			.ret = 0,
 		};
-		ret = smp_call_function_single(event->oncpu, __perf_event_read, &data, 1);
-		/* The event must have been read from an online CPU: */
-		WARN_ON_ONCE(ret);
-		ret = ret ? : data.ret;
+		/*
+		 * Purposely ignore the smp_call_function_single() return
+		 * value.
+		 *
+		 * If event->oncpu isn't a valid CPU it means the event got
+		 * scheduled out and that will have updated the event count.
+		 *
+		 * Therefore, either way, we'll have an up-to-date event count
+		 * after this.
+		 */
+		(void)smp_call_function_single(event->oncpu, __perf_event_read, &data, 1);
+		ret = data.ret;
 	} else if (event->state == PERF_EVENT_STATE_INACTIVE) {
 		struct perf_event_context *ctx = event->ctx;
 		unsigned long flags;
@@ -4837,6 +4845,19 @@ static void ring_buffer_attach(struct perf_event *event,
 		spin_unlock_irqrestore(&rb->event_lock, flags);
 	}
 
+	/*
+	 * Avoid racing with perf_mmap_close(AUX): stop the event
+	 * before swizzling the event::rb pointer; if it's getting
+	 * unmapped, its aux_mmap_count will be 0 and it won't
+	 * restart. See the comment in __perf_pmu_output_stop().
+	 *
+	 * Data will inevitably be lost when set_output is done in
+	 * mid-air, but then again, whoever does it like this is
+	 * not in for the data anyway.
+	 */
+	if (has_aux(event))
+		perf_event_stop(event, 0);
+
 	rcu_assign_pointer(event->rb, rb);
 
 	if (old_rb) {
@@ -6112,7 +6133,7 @@ static void perf_event_addr_filters_exec(struct perf_event *event, void *data)
 	raw_spin_unlock_irqrestore(&ifh->lock, flags);
 
 	if (restart)
-		perf_event_restart(event);
+		perf_event_stop(event, 1);
 }
 
 void perf_event_exec(void)
@@ -6156,7 +6177,13 @@ static void __perf_event_output_stop(struct perf_event *event, void *data)
 
 	/*
 	 * In case of inheritance, it will be the parent that links to the
-	 * ring-buffer, but it will be the child that's actually using it:
+	 * ring-buffer, but it will be the child that's actually using it.
+	 *
+	 * We are using event::rb to determine if the event should be stopped,
+	 * however this may race with ring_buffer_attach() (through set_output),
+	 * which will make us skip the event that actually needs to be stopped.
+	 * So ring_buffer_attach() has to stop an aux event before re-assigning
+	 * its rb pointer.
 	 */
 	if (rcu_dereference(parent->rb) == rb)
 		ro->err = __perf_event_stop(&sd);
@@ -6670,7 +6697,7 @@ static void __perf_addr_filters_adjust(struct perf_event *event, void *data)
 	raw_spin_unlock_irqrestore(&ifh->lock, flags);
 
 	if (restart)
-		perf_event_restart(event);
+		perf_event_stop(event, 1);
 }
 
 /*
@@ -7859,7 +7886,7 @@ static void perf_event_addr_filters_apply(struct perf_event *event)
 	mmput(mm);
 
 restart:
-	perf_event_restart(event);
+	perf_event_stop(event, 1);
 }
 
 /*

kernel/events/ring_buffer.c

@@ -330,15 +330,22 @@ void *perf_aux_output_begin(struct perf_output_handle *handle,
 	if (!rb)
 		return NULL;
 
-	if (!rb_has_aux(rb) || !atomic_inc_not_zero(&rb->aux_refcount))
+	if (!rb_has_aux(rb))
 		goto err;
 
 	/*
-	 * If rb::aux_mmap_count is zero (and rb_has_aux() above went through),
-	 * the aux buffer is in perf_mmap_close(), about to get freed.
+	 * If aux_mmap_count is zero, the aux buffer is in perf_mmap_close(),
+	 * about to get freed, so we leave immediately.
+	 *
+	 * Checking rb::aux_mmap_count and rb::refcount has to be done in
+	 * the same order, see perf_mmap_close. Otherwise we end up freeing
+	 * aux pages in this path, which is a bug, because in_atomic().
 	 */
 	if (!atomic_read(&rb->aux_mmap_count))
-		goto err_put;
+		goto err;
+
+	if (!atomic_inc_not_zero(&rb->aux_refcount))
+		goto err;
 
 	/*
 	 * Nesting is not supported for AUX area, make sure nested

kernel/memremap.c

@@ -247,6 +247,7 @@ static void devm_memremap_pages_release(struct device *dev, void *data)
 	align_start = res->start & ~(SECTION_SIZE - 1);
 	align_size = ALIGN(resource_size(res), SECTION_SIZE);
 	arch_remove_memory(align_start, align_size);
+	untrack_pfn(NULL, PHYS_PFN(align_start), align_size);
 	pgmap_radix_release(res);
 	dev_WARN_ONCE(dev, pgmap->altmap && pgmap->altmap->alloc,
 			"%s: failed to free all reserved pages\n", __func__);
@@ -282,6 +283,7 @@ void *devm_memremap_pages(struct device *dev, struct resource *res,
 		struct percpu_ref *ref, struct vmem_altmap *altmap)
 {
 	resource_size_t key, align_start, align_size, align_end;
+	pgprot_t pgprot = PAGE_KERNEL;
 	struct dev_pagemap *pgmap;
 	struct page_map *page_map;
 	int error, nid, is_ram;
@@ -351,6 +353,11 @@ void *devm_memremap_pages(struct device *dev, struct resource *res,
 	if (nid < 0)
 		nid = numa_mem_id();
 
+	error = track_pfn_remap(NULL, &pgprot, PHYS_PFN(align_start), 0,
+			align_size);
+	if (error)
+		goto err_pfn_remap;
+
 	error = arch_add_memory(nid, align_start, align_size, true);
 	if (error)
 		goto err_add_memory;
@@ -371,6 +378,8 @@ void *devm_memremap_pages(struct device *dev, struct resource *res,
 	return __va(res->start);
 
  err_add_memory:
+	untrack_pfn(NULL, PHYS_PFN(align_start), align_size);
+ err_pfn_remap:
  err_radix:
 	pgmap_radix_release(res);
 	devres_free(page_map);

kernel/power/qos.c

@@ -482,7 +482,16 @@ void pm_qos_update_request(struct pm_qos_request *req,
 		return;
 	}
 
-	cancel_delayed_work_sync(&req->work);
+	/*
+	 * This function may be called very early during boot, for example,
+	 * from of_clk_init(), where irq needs to stay disabled.
+	 * cancel_delayed_work_sync() assumes that irq is enabled on
+	 * invocation and re-enables it on return.  Avoid calling it until
+	 * workqueue is initialized.
+	 */
+	if (keventd_up())
+		cancel_delayed_work_sync(&req->work);
+
 	__pm_qos_update_request(req, new_value);
 }
 EXPORT_SYMBOL_GPL(pm_qos_update_request);