Merge branch 'perf/urgent' into perf/core, to pick up fixes

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

7 files changed, 99 insertions, 61 deletions

diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c
index 229a5d5df977..3d55d95dcf49 100644
--- a/kernel/bpf/arraymap.c
+++ b/kernel/bpf/arraymap.c
@@ -11,7 +11,6 @@
  */
 #include <linux/bpf.h>
 #include <linux/err.h>
-#include <linux/vmalloc.h>
 #include <linux/slab.h>
 #include <linux/mm.h>
 #include <linux/filter.h>
@@ -74,14 +73,10 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr)
 	if (array_size >= U32_MAX - PAGE_SIZE)
 		return ERR_PTR(-ENOMEM);
 
-
 	/* allocate all map elements and zero-initialize them */
-	array = kzalloc(array_size, GFP_USER | __GFP_NOWARN);
-	if (!array) {
-		array = vzalloc(array_size);
-		if (!array)
-			return ERR_PTR(-ENOMEM);
-	}
+	array = bpf_map_area_alloc(array_size);
+	if (!array)
+		return ERR_PTR(-ENOMEM);
 
 	/* copy mandatory map attributes */
 	array->map.map_type = attr->map_type;
@@ -97,7 +92,7 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr)
 
 	if (array_size >= U32_MAX - PAGE_SIZE ||
 	    elem_size > PCPU_MIN_UNIT_SIZE || bpf_array_alloc_percpu(array)) {
-		kvfree(array);
+		bpf_map_area_free(array);
 		return ERR_PTR(-ENOMEM);
 	}
 out:
@@ -262,7 +257,7 @@ static void array_map_free(struct bpf_map *map)
 	if (array->map.map_type == BPF_MAP_TYPE_PERCPU_ARRAY)
 		bpf_array_free_percpu(array);
 
-	kvfree(array);
+	bpf_map_area_free(array);
 }
 
 static const struct bpf_map_ops array_ops = {
@@ -319,7 +314,8 @@ static void fd_array_map_free(struct bpf_map *map)
 	/* make sure it's empty */
 	for (i = 0; i < array->map.max_entries; i++)
 		BUG_ON(array->ptrs[i] != NULL);
-	kvfree(array);
+
+	bpf_map_area_free(array);
 }
 
 static void *fd_array_map_lookup_elem(struct bpf_map *map, void *key)
diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index 3f2bb58952d8..a753bbe7df0a 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -13,7 +13,6 @@
 #include <linux/bpf.h>
 #include <linux/jhash.h>
 #include <linux/filter.h>
-#include <linux/vmalloc.h>
 #include "percpu_freelist.h"
 #include "bpf_lru_list.h"
 
@@ -103,7 +102,7 @@ static void htab_free_elems(struct bpf_htab *htab)
 		free_percpu(pptr);
 	}
 free_elems:
-	vfree(htab->elems);
+	bpf_map_area_free(htab->elems);
 }
 
 static struct htab_elem *prealloc_lru_pop(struct bpf_htab *htab, void *key,
@@ -125,7 +124,8 @@ static int prealloc_init(struct bpf_htab *htab)
 {
 	int err = -ENOMEM, i;
 
-	htab->elems = vzalloc(htab->elem_size * htab->map.max_entries);
+	htab->elems = bpf_map_area_alloc(htab->elem_size *
+					 htab->map.max_entries);
 	if (!htab->elems)
 		return -ENOMEM;
 
@@ -320,14 +320,10 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr)
 		goto free_htab;
 
 	err = -ENOMEM;
-	htab->buckets = kmalloc_array(htab->n_buckets, sizeof(struct bucket),
-				      GFP_USER | __GFP_NOWARN);
-
-	if (!htab->buckets) {
-		htab->buckets = vmalloc(htab->n_buckets * sizeof(struct bucket));
-		if (!htab->buckets)
-			goto free_htab;
-	}
+	htab->buckets = bpf_map_area_alloc(htab->n_buckets *
+					   sizeof(struct bucket));
+	if (!htab->buckets)
+		goto free_htab;
 
 	for (i = 0; i < htab->n_buckets; i++) {
 		INIT_HLIST_HEAD(&htab->buckets[i].head);
@@ -354,7 +350,7 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr)
 free_extra_elems:
 	free_percpu(htab->extra_elems);
 free_buckets:
-	kvfree(htab->buckets);
+	bpf_map_area_free(htab->buckets);
 free_htab:
 	kfree(htab);
 	return ERR_PTR(err);
@@ -1014,7 +1010,7 @@ static void htab_map_free(struct bpf_map *map)
 		prealloc_destroy(htab);
 
 	free_percpu(htab->extra_elems);
-	kvfree(htab->buckets);
+	bpf_map_area_free(htab->buckets);
 	kfree(htab);
 }
 
diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c
index 732ae16d12b7..be8519148c25 100644
--- a/kernel/bpf/stackmap.c
+++ b/kernel/bpf/stackmap.c
@@ -7,7 +7,6 @@
 #include <linux/bpf.h>
 #include <linux/jhash.h>
 #include <linux/filter.h>
-#include <linux/vmalloc.h>
 #include <linux/stacktrace.h>
 #include <linux/perf_event.h>
 #include "percpu_freelist.h"
@@ -32,7 +31,7 @@ static int prealloc_elems_and_freelist(struct bpf_stack_map *smap)
 	u32 elem_size = sizeof(struct stack_map_bucket) + smap->map.value_size;
 	int err;
 
-	smap->elems = vzalloc(elem_size * smap->map.max_entries);
+	smap->elems = bpf_map_area_alloc(elem_size * smap->map.max_entries);
 	if (!smap->elems)
 		return -ENOMEM;
 
@@ -45,7 +44,7 @@ static int prealloc_elems_and_freelist(struct bpf_stack_map *smap)
 	return 0;
 
 free_elems:
-	vfree(smap->elems);
+	bpf_map_area_free(smap->elems);
 	return err;
 }
 
@@ -76,12 +75,9 @@ static struct bpf_map *stack_map_alloc(union bpf_attr *attr)
 	if (cost >= U32_MAX - PAGE_SIZE)
 		return ERR_PTR(-E2BIG);
 
-	smap = kzalloc(cost, GFP_USER | __GFP_NOWARN);
-	if (!smap) {
-		smap = vzalloc(cost);
-		if (!smap)
-			return ERR_PTR(-ENOMEM);
-	}
+	smap = bpf_map_area_alloc(cost);
+	if (!smap)
+		return ERR_PTR(-ENOMEM);
 
 	err = -E2BIG;
 	cost += n_buckets * (value_size + sizeof(struct stack_map_bucket));
@@ -112,7 +108,7 @@ static struct bpf_map *stack_map_alloc(union bpf_attr *attr)
 put_buffers:
 	put_callchain_buffers();
 free_smap:
-	kvfree(smap);
+	bpf_map_area_free(smap);
 	return ERR_PTR(err);
 }
 
@@ -262,9 +258,9 @@ static void stack_map_free(struct bpf_map *map)
 	/* wait for bpf programs to complete before freeing stack map */
 	synchronize_rcu();
 
-	vfree(smap->elems);
+	bpf_map_area_free(smap->elems);
 	pcpu_freelist_destroy(&smap->freelist);
-	kvfree(smap);
+	bpf_map_area_free(smap);
 	put_callchain_buffers();
 }
 
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 1d6b29e4e2c3..19b6129eab23 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -12,6 +12,8 @@
 #include <linux/bpf.h>
 #include <linux/syscalls.h>
 #include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/mmzone.h>
 #include <linux/anon_inodes.h>
 #include <linux/file.h>
 #include <linux/license.h>
@@ -49,6 +51,30 @@ void bpf_register_map_type(struct bpf_map_type_list *tl)
 	list_add(&tl->list_node, &bpf_map_types);
 }
 
+void *bpf_map_area_alloc(size_t size)
+{
+	/* We definitely need __GFP_NORETRY, so OOM killer doesn't
+	 * trigger under memory pressure as we really just want to
+	 * fail instead.
+	 */
+	const gfp_t flags = __GFP_NOWARN | __GFP_NORETRY | __GFP_ZERO;
+	void *area;
+
+	if (size <= (PAGE_SIZE << PAGE_ALLOC_COSTLY_ORDER)) {
+		area = kmalloc(size, GFP_USER | flags);
+		if (area != NULL)
+			return area;
+	}
+
+	return __vmalloc(size, GFP_KERNEL | __GFP_HIGHMEM | flags,
+			 PAGE_KERNEL);
+}
+
+void bpf_map_area_free(void *area)
+{
+	kvfree(area);
+}
+
 int bpf_map_precharge_memlock(u32 pages)
 {
 	struct user_struct *user = get_current_user();
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 110b38a58493..e5aaa806702d 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -1469,7 +1469,6 @@ ctx_group_list(struct perf_event *event, struct perf_event_context *ctx)
 static void
 list_add_event(struct perf_event *event, struct perf_event_context *ctx)
 {
-
 	lockdep_assert_held(&ctx->lock);
 
 	WARN_ON_ONCE(event->attach_state & PERF_ATTACH_CONTEXT);
@@ -1624,6 +1623,8 @@ static void perf_group_attach(struct perf_event *event)
 {
 	struct perf_event *group_leader = event->group_leader, *pos;
 
+	lockdep_assert_held(&event->ctx->lock);
+
 	/*
 	 * We can have double attach due to group movement in perf_event_open.
 	 */
@@ -1697,6 +1698,8 @@ static void perf_group_detach(struct perf_event *event)
 	struct perf_event *sibling, *tmp;
 	struct list_head *list = NULL;
 
+	lockdep_assert_held(&event->ctx->lock);
+
 	/*
 	 * We can have double detach due to exit/hot-unplug + close.
 	 */
@@ -1895,9 +1898,29 @@ __perf_remove_from_context(struct perf_event *event,
  */
 static void perf_remove_from_context(struct perf_event *event, unsigned long flags)
 {
-	lockdep_assert_held(&event->ctx->mutex);
+	struct perf_event_context *ctx = event->ctx;
+
+	lockdep_assert_held(&ctx->mutex);
 
 	event_function_call(event, __perf_remove_from_context, (void *)flags);
+
+	/*
+	 * The above event_function_call() can NO-OP when it hits
+	 * TASK_TOMBSTONE. In that case we must already have been detached
+	 * from the context (by perf_event_exit_event()) but the grouping
+	 * might still be in-tact.
+	 */
+	WARN_ON_ONCE(event->attach_state & PERF_ATTACH_CONTEXT);
+	if ((flags & DETACH_GROUP) &&
+	    (event->attach_state & PERF_ATTACH_GROUP)) {
+		/*
+		 * Since in that case we cannot possibly be scheduled, simply
+		 * detach now.
+		 */
+		raw_spin_lock_irq(&ctx->lock);
+		perf_group_detach(event);
+		raw_spin_unlock_irq(&ctx->lock);
+	}
 }
 
 /*
@@ -6609,6 +6632,27 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event)
 	char *buf = NULL;
 	char *name;
 
+	if (vma->vm_flags & VM_READ)
+		prot |= PROT_READ;
+	if (vma->vm_flags & VM_WRITE)
+		prot |= PROT_WRITE;
+	if (vma->vm_flags & VM_EXEC)
+		prot |= PROT_EXEC;
+
+	if (vma->vm_flags & VM_MAYSHARE)
+		flags = MAP_SHARED;
+	else
+		flags = MAP_PRIVATE;
+
+	if (vma->vm_flags & VM_DENYWRITE)
+		flags |= MAP_DENYWRITE;
+	if (vma->vm_flags & VM_MAYEXEC)
+		flags |= MAP_EXECUTABLE;
+	if (vma->vm_flags & VM_LOCKED)
+		flags |= MAP_LOCKED;
+	if (vma->vm_flags & VM_HUGETLB)
+		flags |= MAP_HUGETLB;
+
 	if (file) {
 		struct inode *inode;
 		dev_t dev;
@@ -6635,27 +6679,6 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event)
 		maj = MAJOR(dev);
 		min = MINOR(dev);
 
-		if (vma->vm_flags & VM_READ)
-			prot |= PROT_READ;
-		if (vma->vm_flags & VM_WRITE)
-			prot |= PROT_WRITE;
-		if (vma->vm_flags & VM_EXEC)
-			prot |= PROT_EXEC;
-
-		if (vma->vm_flags & VM_MAYSHARE)
-			flags = MAP_SHARED;
-		else
-			flags = MAP_PRIVATE;
-
-		if (vma->vm_flags & VM_DENYWRITE)
-			flags |= MAP_DENYWRITE;
-		if (vma->vm_flags & VM_MAYEXEC)
-			flags |= MAP_EXECUTABLE;
-		if (vma->vm_flags & VM_LOCKED)
-			flags |= MAP_LOCKED;
-		if (vma->vm_flags & VM_HUGETLB)
-			flags |= MAP_HUGETLB;
-
 		goto got_name;
 	} else {
 		if (vma->vm_ops && vma->vm_ops->name) {
diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c
index f67ceb7768b8..15e6baef5c73 100644
--- a/kernel/power/suspend.c
+++ b/kernel/power/suspend.c
@@ -46,7 +46,7 @@ static const char * const mem_sleep_labels[] = {
 const char *mem_sleep_states[PM_SUSPEND_MAX];
 
 suspend_state_t mem_sleep_current = PM_SUSPEND_FREEZE;
-suspend_state_t mem_sleep_default = PM_SUSPEND_MAX;
+static suspend_state_t mem_sleep_default = PM_SUSPEND_MEM;
 
 unsigned int pm_suspend_global_flags;
 EXPORT_SYMBOL_GPL(pm_suspend_global_flags);
@@ -168,7 +168,7 @@ void suspend_set_ops(const struct platform_suspend_ops *ops)
 	}
 	if (valid_state(PM_SUSPEND_MEM)) {
 		mem_sleep_states[PM_SUSPEND_MEM] = mem_sleep_labels[PM_SUSPEND_MEM];
-		if (mem_sleep_default >= PM_SUSPEND_MEM)
+		if (mem_sleep_default == PM_SUSPEND_MEM)
 			mem_sleep_current = PM_SUSPEND_MEM;
 	}
 
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 8dbaec0e4f7f..1aea594a54db 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -2475,6 +2475,7 @@ static int __do_proc_doulongvec_minmax(void *data, struct ctl_table *table, int
 				break;
 			if (neg)
 				continue;
+			val = convmul * val / convdiv;
 			if ((min && val < *min) || (max && val > *max))
 				continue;
 			*i = val;