Merge tag 'kmemleak' of git://git.kernel.org/pub/scm/linux/kernel/git/cmarinas/linux

Kmemleak patches

Main features:
- Handle percpu memory allocations (only scanning them, not actually
  reporting).
- Memory hotplug support.

Usability improvements:
- Show the origin of early allocations.
- Report previously found leaks even if kmemleak has been disabled by
  some error.

* tag 'kmemleak' of git://git.kernel.org/pub/scm/linux/kernel/git/cmarinas/linux:
  kmemleak: Add support for memory hotplug
  kmemleak: Handle percpu memory allocation
  kmemleak: Report previously found leaks even after an error
  kmemleak: When the early log buffer is exceeded, report the actual number
  kmemleak: Show where early_log issues come from

Documentation/kmemleak.txt

@@ -127,7 +127,10 @@ See the include/linux/kmemleak.h header for the functions prototype.
 
 kmemleak_init		 - initialize kmemleak
 kmemleak_alloc		 - notify of a memory block allocation
+kmemleak_alloc_percpu	 - notify of a percpu memory block allocation
 kmemleak_free		 - notify of a memory block freeing
+kmemleak_free_part	 - notify of a partial memory block freeing
+kmemleak_free_percpu	 - notify of a percpu memory block freeing
 kmemleak_not_leak	 - mark an object as not a leak
 kmemleak_ignore		 - do not scan or report an object as leak
 kmemleak_scan_area	 - add scan areas inside a memory block

include/linux/kmemleak.h

@@ -26,8 +26,10 @@
 extern void kmemleak_init(void) __ref;
 extern void kmemleak_alloc(const void *ptr, size_t size, int min_count,
 			   gfp_t gfp) __ref;
+extern void kmemleak_alloc_percpu(const void __percpu *ptr, size_t size) __ref;
 extern void kmemleak_free(const void *ptr) __ref;
 extern void kmemleak_free_part(const void *ptr, size_t size) __ref;
+extern void kmemleak_free_percpu(const void __percpu *ptr) __ref;
 extern void kmemleak_padding(const void *ptr, unsigned long offset,
 			     size_t size) __ref;
 extern void kmemleak_not_leak(const void *ptr) __ref;
@@ -68,6 +70,9 @@ static inline void kmemleak_alloc_recursive(const void *ptr, size_t size,
 					    gfp_t gfp)
 {
 }
+static inline void kmemleak_alloc_percpu(const void __percpu *ptr, size_t size)
+{
+}
 static inline void kmemleak_free(const void *ptr)
 {
 }
@@ -77,6 +82,9 @@ static inline void kmemleak_free_part(const void *ptr, size_t size)
 static inline void kmemleak_free_recursive(const void *ptr, unsigned long flags)
 {
 }
+static inline void kmemleak_free_percpu(const void __percpu *ptr)
+{
+}
 static inline void kmemleak_not_leak(const void *ptr)
 {
 }

lib/Kconfig.debug

@@ -414,7 +414,7 @@ config SLUB_STATS
 
 config DEBUG_KMEMLEAK
 	bool "Kernel memory leak detector"
-	depends on DEBUG_KERNEL && EXPERIMENTAL && !MEMORY_HOTPLUG && \
+	depends on DEBUG_KERNEL && EXPERIMENTAL && \
 		(X86 || ARM || PPC || MIPS || S390 || SPARC64 || SUPERH || MICROBLAZE || TILE)
 
 	select DEBUG_FS

mm/kmemleak.c

@@ -100,6 +100,7 @@
 
 #include <linux/kmemcheck.h>
 #include <linux/kmemleak.h>
+#include <linux/memory_hotplug.h>
 
 /*
  * Kmemleak configuration and common defines.
@@ -196,7 +197,9 @@ static atomic_t kmemleak_enabled = ATOMIC_INIT(0);
 static atomic_t kmemleak_initialized = ATOMIC_INIT(0);
 /* enables or disables early logging of the memory operations */
 static atomic_t kmemleak_early_log = ATOMIC_INIT(1);
-/* set if a fata kmemleak error has occurred */
+/* set if a kmemleak warning was issued */
+static atomic_t kmemleak_warning = ATOMIC_INIT(0);
+/* set if a fatal kmemleak error has occurred */
 static atomic_t kmemleak_error = ATOMIC_INIT(0);
 
 /* minimum and maximum address that may be valid pointers */
@@ -228,8 +231,10 @@ static int kmemleak_skip_disable;
 /* kmemleak operation type for early logging */
 enum {
 	KMEMLEAK_ALLOC,
+	KMEMLEAK_ALLOC_PERCPU,
 	KMEMLEAK_FREE,
 	KMEMLEAK_FREE_PART,
+	KMEMLEAK_FREE_PERCPU,
 	KMEMLEAK_NOT_LEAK,
 	KMEMLEAK_IGNORE,
 	KMEMLEAK_SCAN_AREA,
@@ -262,6 +267,7 @@ static void kmemleak_disable(void);
 #define kmemleak_warn(x...)	do {		\
 	pr_warning(x);				\
 	dump_stack();				\
+	atomic_set(&kmemleak_warning, 1);	\
 } while (0)
 
 /*
@@ -403,8 +409,8 @@ static struct kmemleak_object *lookup_object(unsigned long ptr, int alias)
 		object = prio_tree_entry(node, struct kmemleak_object,
 					 tree_node);
 		if (!alias && object->pointer != ptr) {
-			pr_warning("Found object by alias at 0x%08lx\n", ptr);
-			dump_stack();
+			kmemleak_warn("Found object by alias at 0x%08lx\n",
+				      ptr);
 			dump_object_info(object);
 			object = NULL;
 		}
@@ -794,9 +800,13 @@ static void __init log_early(int op_type, const void *ptr, size_t size,
 	unsigned long flags;
 	struct early_log *log;
 
+	if (atomic_read(&kmemleak_error)) {
+		/* kmemleak stopped recording, just count the requests */
+		crt_early_log++;
+		return;
+	}
+
 	if (crt_early_log >= ARRAY_SIZE(early_log)) {
-		pr_warning("Early log buffer exceeded, "
-			   "please increase DEBUG_KMEMLEAK_EARLY_LOG_SIZE\n");
 		kmemleak_disable();
 		return;
 	}
@@ -811,7 +821,6 @@ static void __init log_early(int op_type, const void *ptr, size_t size,
 	log->ptr = ptr;
 	log->size = size;
 	log->min_count = min_count;
-	if (op_type == KMEMLEAK_ALLOC)
 	log->trace_len = __save_stack_trace(log->trace);
 	crt_early_log++;
 	local_irq_restore(flags);
@@ -846,6 +855,20 @@ static void early_alloc(struct early_log *log)
 	rcu_read_unlock();
 }
 
+/*
+ * Log an early allocated block and populate the stack trace.
+ */
+static void early_alloc_percpu(struct early_log *log)
+{
+	unsigned int cpu;
+	const void __percpu *ptr = log->ptr;
+
+	for_each_possible_cpu(cpu) {
+		log->ptr = per_cpu_ptr(ptr, cpu);
+		early_alloc(log);
+	}
+}
+
 /**
  * kmemleak_alloc - register a newly allocated object
  * @ptr:	pointer to beginning of the object
@@ -872,6 +895,34 @@ void __ref kmemleak_alloc(const void *ptr, size_t size, int min_count,
 }
 EXPORT_SYMBOL_GPL(kmemleak_alloc);
 
+/**
+ * kmemleak_alloc_percpu - register a newly allocated __percpu object
+ * @ptr:	__percpu pointer to beginning of the object
+ * @size:	size of the object
+ *
+ * This function is called from the kernel percpu allocator when a new object
+ * (memory block) is allocated (alloc_percpu). It assumes GFP_KERNEL
+ * allocation.
+ */
+void __ref kmemleak_alloc_percpu(const void __percpu *ptr, size_t size)
+{
+	unsigned int cpu;
+
+	pr_debug("%s(0x%p, %zu)\n", __func__, ptr, size);
+
+	/*
+	 * Percpu allocations are only scanned and not reported as leaks
+	 * (min_count is set to 0).
+	 */
+	if (atomic_read(&kmemleak_enabled) && ptr && !IS_ERR(ptr))
+		for_each_possible_cpu(cpu)
+			create_object((unsigned long)per_cpu_ptr(ptr, cpu),
+				      size, 0, GFP_KERNEL);
+	else if (atomic_read(&kmemleak_early_log))
+		log_early(KMEMLEAK_ALLOC_PERCPU, ptr, size, 0);
+}
+EXPORT_SYMBOL_GPL(kmemleak_alloc_percpu);
+
 /**
  * kmemleak_free - unregister a previously registered object
  * @ptr:	pointer to beginning of the object
@@ -910,6 +961,28 @@ void __ref kmemleak_free_part(const void *ptr, size_t size)
 }
 EXPORT_SYMBOL_GPL(kmemleak_free_part);
 
+/**
+ * kmemleak_free_percpu - unregister a previously registered __percpu object
+ * @ptr:	__percpu pointer to beginning of the object
+ *
+ * This function is called from the kernel percpu allocator when an object
+ * (memory block) is freed (free_percpu).
+ */
+void __ref kmemleak_free_percpu(const void __percpu *ptr)
+{
+	unsigned int cpu;
+
+	pr_debug("%s(0x%p)\n", __func__, ptr);
+
+	if (atomic_read(&kmemleak_enabled) && ptr && !IS_ERR(ptr))
+		for_each_possible_cpu(cpu)
+			delete_object_full((unsigned long)per_cpu_ptr(ptr,
+								      cpu));
+	else if (atomic_read(&kmemleak_early_log))
+		log_early(KMEMLEAK_FREE_PERCPU, ptr, 0, 0);
+}
+EXPORT_SYMBOL_GPL(kmemleak_free_percpu);
+
 /**
  * kmemleak_not_leak - mark an allocated object as false positive
  * @ptr:	pointer to beginning of the object
@@ -1220,9 +1293,9 @@ static void kmemleak_scan(void)
 #endif
 
 	/*
-	 * Struct page scanning for each node. The code below is not yet safe
-	 * with MEMORY_HOTPLUG.
+	 * Struct page scanning for each node.
 	 */
+	lock_memory_hotplug();
 	for_each_online_node(i) {
 		pg_data_t *pgdat = NODE_DATA(i);
 		unsigned long start_pfn = pgdat->node_start_pfn;
@@ -1241,6 +1314,7 @@ static void kmemleak_scan(void)
 			scan_block(page, page + 1, NULL, 1);
 		}
 	}
+	unlock_memory_hotplug();
 
 	/*
 	 * Scanning the task stacks (may introduce false negatives).
@@ -1467,9 +1541,6 @@ static const struct seq_operations kmemleak_seq_ops = {
 
 static int kmemleak_open(struct inode *inode, struct file *file)
 {
-	if (!atomic_read(&kmemleak_enabled))
-		return -EBUSY;
-
 	return seq_open(file, &kmemleak_seq_ops);
 }
 
@@ -1543,6 +1614,9 @@ static ssize_t kmemleak_write(struct file *file, const char __user *user_buf,
 	int buf_size;
 	int ret;
 
+	if (!atomic_read(&kmemleak_enabled))
+		return -EBUSY;
+
 	buf_size = min(size, (sizeof(buf) - 1));
 	if (strncpy_from_user(buf, user_buf, buf_size) < 0)
 		return -EFAULT;
@@ -1602,20 +1676,24 @@ static const struct file_operations kmemleak_fops = {
 };
 
 /*
- * Perform the freeing of the kmemleak internal objects after waiting for any
- * current memory scan to complete.
+ * Stop the memory scanning thread and free the kmemleak internal objects if
+ * no previous scan thread (otherwise, kmemleak may still have some useful
+ * information on memory leaks).
  */
 static void kmemleak_do_cleanup(struct work_struct *work)
 {
 	struct kmemleak_object *object;
+	bool cleanup = scan_thread == NULL;
 
 	mutex_lock(&scan_mutex);
 	stop_scan_thread();
 
+	if (cleanup) {
 		rcu_read_lock();
 		list_for_each_entry_rcu(object, &object_list, object_list)
 			delete_object_full(object->pointer);
 		rcu_read_unlock();
+	}
 	mutex_unlock(&scan_mutex);
 }
 
@@ -1632,7 +1710,6 @@ static void kmemleak_disable(void)
 		return;
 
 	/* stop any memory operation tracing */
-	atomic_set(&kmemleak_early_log, 0);
 	atomic_set(&kmemleak_enabled, 0);
 
 	/* check whether it is too early for a kernel thread */
@@ -1659,6 +1736,17 @@ static int kmemleak_boot_config(char *str)
 }
 early_param("kmemleak", kmemleak_boot_config);
 
+static void __init print_log_trace(struct early_log *log)
+{
+	struct stack_trace trace;
+
+	trace.nr_entries = log->trace_len;
+	trace.entries = log->trace;
+
+	pr_notice("Early log backtrace:\n");
+	print_stack_trace(&trace, 2);
+}
+
 /*
  * Kmemleak initialization.
  */
@@ -1681,12 +1769,18 @@ void __init kmemleak_init(void)
 	scan_area_cache = KMEM_CACHE(kmemleak_scan_area, SLAB_NOLEAKTRACE);
 	INIT_PRIO_TREE_ROOT(&object_tree_root);
 
+	if (crt_early_log >= ARRAY_SIZE(early_log))
+		pr_warning("Early log buffer exceeded (%d), please increase "
+			   "DEBUG_KMEMLEAK_EARLY_LOG_SIZE\n", crt_early_log);
+
 	/* the kernel is still in UP mode, so disabling the IRQs is enough */
 	local_irq_save(flags);
-	if (!atomic_read(&kmemleak_error)) {
-		atomic_set(&kmemleak_enabled, 1);
 	atomic_set(&kmemleak_early_log, 0);
-	}
+	if (atomic_read(&kmemleak_error)) {
+		local_irq_restore(flags);
+		return;
+	} else
+		atomic_set(&kmemleak_enabled, 1);
 	local_irq_restore(flags);
 
 	/*
@@ -1701,12 +1795,18 @@ void __init kmemleak_init(void)
 		case KMEMLEAK_ALLOC:
 			early_alloc(log);
 			break;
+		case KMEMLEAK_ALLOC_PERCPU:
+			early_alloc_percpu(log);
+			break;
 		case KMEMLEAK_FREE:
 			kmemleak_free(log->ptr);
 			break;
 		case KMEMLEAK_FREE_PART:
 			kmemleak_free_part(log->ptr, log->size);
 			break;
+		case KMEMLEAK_FREE_PERCPU:
+			kmemleak_free_percpu(log->ptr);
+			break;
 		case KMEMLEAK_NOT_LEAK:
 			kmemleak_not_leak(log->ptr);
 			break;
@@ -1720,7 +1820,13 @@ void __init kmemleak_init(void)
 			kmemleak_no_scan(log->ptr);
 			break;
 		default:
-			WARN_ON(1);
+			kmemleak_warn("Unknown early log operation: %d\n",
+				      log->op_type);
+		}
+
+		if (atomic_read(&kmemleak_warning)) {
+			print_log_trace(log);
+			atomic_set(&kmemleak_warning, 0);
 		}
 	}
 }

mm/percpu.c

@@ -67,6 +67,7 @@
 #include <linux/spinlock.h>
 #include <linux/vmalloc.h>
 #include <linux/workqueue.h>
+#include <linux/kmemleak.h>
 
 #include <asm/cacheflush.h>
 #include <asm/sections.h>
@@ -710,6 +711,7 @@ static void __percpu *pcpu_alloc(size_t size, size_t align, bool reserved)
 	const char *err;
 	int slot, off, new_alloc;
 	unsigned long flags;
+	void __percpu *ptr;
 
 	if (unlikely(!size || size > PCPU_MIN_UNIT_SIZE || align > PAGE_SIZE)) {
 		WARN(true, "illegal size (%zu) or align (%zu) for "
@@ -802,7 +804,9 @@ static void __percpu *pcpu_alloc(size_t size, size_t align, bool reserved)
 	mutex_unlock(&pcpu_alloc_mutex);
 
 	/* return address relative to base address */
-	return __addr_to_pcpu_ptr(chunk->base_addr + off);
+	ptr = __addr_to_pcpu_ptr(chunk->base_addr + off);
+	kmemleak_alloc_percpu(ptr, size);
+	return ptr;
 
 fail_unlock:
 	spin_unlock_irqrestore(&pcpu_lock, flags);
@@ -916,6 +920,8 @@ void free_percpu(void __percpu *ptr)
 	if (!ptr)
 		return;
 
+	kmemleak_free_percpu(ptr);
+
 	addr = __pcpu_ptr_to_addr(ptr);
 
 	spin_lock_irqsave(&pcpu_lock, flags);
@@ -1639,6 +1645,8 @@ int __init pcpu_embed_first_chunk(size_t reserved_size, size_t dyn_size,
 			rc = -ENOMEM;
 			goto out_free_areas;
 		}
+		/* kmemleak tracks the percpu allocations separately */
+		kmemleak_free(ptr);
 		areas[group] = ptr;
 
 		base = min(ptr, base);
@@ -1753,6 +1761,8 @@ int __init pcpu_page_first_chunk(size_t reserved_size,
 					   "for cpu%u\n", psize_str, cpu);
 				goto enomem;
 			}
+			/* kmemleak tracks the percpu allocations separately */
+			kmemleak_free(ptr);
 			pages[j++] = virt_to_page(ptr);
 		}