summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2009-06-12 01:15:57 +0400
committerLinus Torvalds <torvalds@linux-foundation.org>2009-06-12 01:15:57 +0400
commit512626a04e72aca60effe111fa0333ed0b195d21 (patch)
treec22e23b0dcc2dd2ff5a9a96a007de6799e9223de
parent8a1ca8cedd108c8e76a6ab34079d0bbb4f244799 (diff)
parent3aa27bbe7a6536d1ec859d3a97caf3319b5081b7 (diff)
downloadlinux-512626a04e72aca60effe111fa0333ed0b195d21.tar.xz
Merge branch 'for-linus' of git://linux-arm.org/linux-2.6
* 'for-linus' of git://linux-arm.org/linux-2.6: kmemleak: Add the corresponding MAINTAINERS entry kmemleak: Simple testing module for kmemleak kmemleak: Enable the building of the memory leak detector kmemleak: Remove some of the kmemleak false positives kmemleak: Add modules support kmemleak: Add kmemleak_alloc callback from alloc_large_system_hash kmemleak: Add the vmalloc memory allocation/freeing hooks kmemleak: Add the slub memory allocation/freeing hooks kmemleak: Add the slob memory allocation/freeing hooks kmemleak: Add the slab memory allocation/freeing hooks kmemleak: Add documentation on the memory leak detector kmemleak: Add the base support Manual conflict resolution (with the slab/earlyboot changes) in: drivers/char/vt.c init/main.c mm/slab.c
-rw-r--r--Documentation/kernel-parameters.txt4
-rw-r--r--Documentation/kmemleak.txt142
-rw-r--r--MAINTAINERS6
-rw-r--r--drivers/char/vt.c1
-rw-r--r--fs/block_dev.c6
-rw-r--r--include/linux/kmemleak.h96
-rw-r--r--include/linux/percpu.h5
-rw-r--r--include/linux/slab.h2
-rw-r--r--init/main.c4
-rw-r--r--kernel/module.c56
-rw-r--r--lib/Kconfig.debug32
-rw-r--r--mm/Makefile2
-rw-r--r--mm/kmemleak-test.c111
-rw-r--r--mm/kmemleak.c1498
-rw-r--r--mm/page_alloc.c11
-rw-r--r--mm/slab.c32
-rw-r--r--mm/slob.c7
-rw-r--r--mm/slub.c5
-rw-r--r--mm/vmalloc.c30
19 files changed, 2043 insertions, 7 deletions
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 72d3bf08d79b..7bcdebffdab3 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1083,6 +1083,10 @@ and is between 256 and 4096 characters. It is defined in the file
Configure the RouterBoard 532 series on-chip
Ethernet adapter MAC address.
+ kmemleak= [KNL] Boot-time kmemleak enable/disable
+ Valid arguments: on, off
+ Default: on
+
kstack=N [X86] Print N words from the kernel stack
in oops dumps.
diff --git a/Documentation/kmemleak.txt b/Documentation/kmemleak.txt
new file mode 100644
index 000000000000..0112da3b9ab8
--- /dev/null
+++ b/Documentation/kmemleak.txt
@@ -0,0 +1,142 @@
+Kernel Memory Leak Detector
+===========================
+
+Introduction
+------------
+
+Kmemleak provides a way of detecting possible kernel memory leaks in a
+way similar to a tracing garbage collector
+(http://en.wikipedia.org/wiki/Garbage_collection_%28computer_science%29#Tracing_garbage_collectors),
+with the difference that the orphan objects are not freed but only
+reported via /sys/kernel/debug/kmemleak. A similar method is used by the
+Valgrind tool (memcheck --leak-check) to detect the memory leaks in
+user-space applications.
+
+Usage
+-----
+
+CONFIG_DEBUG_KMEMLEAK in "Kernel hacking" has to be enabled. A kernel
+thread scans the memory every 10 minutes (by default) and prints any new
+unreferenced objects found. To trigger an intermediate scan and display
+all the possible memory leaks:
+
+ # mount -t debugfs nodev /sys/kernel/debug/
+ # cat /sys/kernel/debug/kmemleak
+
+Note that the orphan objects are listed in the order they were allocated
+and one object at the beginning of the list may cause other subsequent
+objects to be reported as orphan.
+
+Memory scanning parameters can be modified at run-time by writing to the
+/sys/kernel/debug/kmemleak file. The following parameters are supported:
+
+ off - disable kmemleak (irreversible)
+ stack=on - enable the task stacks scanning
+ stack=off - disable the tasks stacks scanning
+ scan=on - start the automatic memory scanning thread
+ scan=off - stop the automatic memory scanning thread
+ scan=<secs> - set the automatic memory scanning period in seconds (0
+ to disable it)
+
+Kmemleak can also be disabled at boot-time by passing "kmemleak=off" on
+the kernel command line.
+
+Basic Algorithm
+---------------
+
+The memory allocations via kmalloc, vmalloc, kmem_cache_alloc and
+friends are traced and the pointers, together with additional
+information like size and stack trace, are stored in a prio search tree.
+The corresponding freeing function calls are tracked and the pointers
+removed from the kmemleak data structures.
+
+An allocated block of memory is considered orphan if no pointer to its
+start address or to any location inside the block can be found by
+scanning the memory (including saved registers). This means that there
+might be no way for the kernel to pass the address of the allocated
+block to a freeing function and therefore the block is considered a
+memory leak.
+
+The scanning algorithm steps:
+
+ 1. mark all objects as white (remaining white objects will later be
+ considered orphan)
+ 2. scan the memory starting with the data section and stacks, checking
+ the values against the addresses stored in the prio search tree. If
+ a pointer to a white object is found, the object is added to the
+ gray list
+ 3. scan the gray objects for matching addresses (some white objects
+ can become gray and added at the end of the gray list) until the
+ gray set is finished
+ 4. the remaining white objects are considered orphan and reported via
+ /sys/kernel/debug/kmemleak
+
+Some allocated memory blocks have pointers stored in the kernel's
+internal data structures and they cannot be detected as orphans. To
+avoid this, kmemleak can also store the number of values pointing to an
+address inside the block address range that need to be found so that the
+block is not considered a leak. One example is __vmalloc().
+
+Kmemleak API
+------------
+
+See the include/linux/kmemleak.h header for the functions prototype.
+
+kmemleak_init - initialize kmemleak
+kmemleak_alloc - notify of a memory block allocation
+kmemleak_free - notify of a 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
+kmemleak_no_scan - do not scan a memory block
+kmemleak_erase - erase an old value in a pointer variable
+kmemleak_alloc_recursive - as kmemleak_alloc but checks the recursiveness
+kmemleak_free_recursive - as kmemleak_free but checks the recursiveness
+
+Dealing with false positives/negatives
+--------------------------------------
+
+The false negatives are real memory leaks (orphan objects) but not
+reported by kmemleak because values found during the memory scanning
+point to such objects. To reduce the number of false negatives, kmemleak
+provides the kmemleak_ignore, kmemleak_scan_area, kmemleak_no_scan and
+kmemleak_erase functions (see above). The task stacks also increase the
+amount of false negatives and their scanning is not enabled by default.
+
+The false positives are objects wrongly reported as being memory leaks
+(orphan). For objects known not to be leaks, kmemleak provides the
+kmemleak_not_leak function. The kmemleak_ignore could also be used if
+the memory block is known not to contain other pointers and it will no
+longer be scanned.
+
+Some of the reported leaks are only transient, especially on SMP
+systems, because of pointers temporarily stored in CPU registers or
+stacks. Kmemleak defines MSECS_MIN_AGE (defaulting to 1000) representing
+the minimum age of an object to be reported as a memory leak.
+
+Limitations and Drawbacks
+-------------------------
+
+The main drawback is the reduced performance of memory allocation and
+freeing. To avoid other penalties, the memory scanning is only performed
+when the /sys/kernel/debug/kmemleak file is read. Anyway, this tool is
+intended for debugging purposes where the performance might not be the
+most important requirement.
+
+To keep the algorithm simple, kmemleak scans for values pointing to any
+address inside a block's address range. This may lead to an increased
+number of false negatives. However, it is likely that a real memory leak
+will eventually become visible.
+
+Another source of false negatives is the data stored in non-pointer
+values. In a future version, kmemleak could only scan the pointer
+members in the allocated structures. This feature would solve many of
+the false negative cases described above.
+
+The tool can report false positives. These are cases where an allocated
+block doesn't need to be freed (some cases in the init_call functions),
+the pointer is calculated by other methods than the usual container_of
+macro or the pointer is stored in a location not scanned by kmemleak.
+
+Page allocations and ioremap are not tracked. Only the ARM and x86
+architectures are currently supported.
diff --git a/MAINTAINERS b/MAINTAINERS
index 70f961d43d9c..1a0084e22cf3 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -3370,6 +3370,12 @@ F: Documentation/trace/kmemtrace.txt
F: include/trace/kmemtrace.h
F: kernel/trace/kmemtrace.c
+KMEMLEAK
+P: Catalin Marinas
+M: catalin.marinas@arm.com
+L: linux-kernel@vger.kernel.org
+S: Maintained
+
KPROBES
P: Ananth N Mavinakayanahalli
M: ananth@in.ibm.com
diff --git a/drivers/char/vt.c b/drivers/char/vt.c
index c796a86ab7f3..de9ebee8657b 100644
--- a/drivers/char/vt.c
+++ b/drivers/char/vt.c
@@ -103,6 +103,7 @@
#include <linux/io.h>
#include <asm/system.h>
#include <linux/uaccess.h>
+#include <linux/kmemleak.h>
#define MAX_NR_CON_DRIVER 16
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 2dfc6cdcebbe..931f6b8c4b2f 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -25,6 +25,7 @@
#include <linux/uio.h>
#include <linux/namei.h>
#include <linux/log2.h>
+#include <linux/kmemleak.h>
#include <asm/uaccess.h>
#include "internal.h"
@@ -492,6 +493,11 @@ void __init bdev_cache_init(void)
bd_mnt = kern_mount(&bd_type);
if (IS_ERR(bd_mnt))
panic("Cannot create bdev pseudo-fs");
+ /*
+ * This vfsmount structure is only used to obtain the
+ * blockdev_superblock, so tell kmemleak not to report it.
+ */
+ kmemleak_not_leak(bd_mnt);
blockdev_superblock = bd_mnt->mnt_sb; /* For writeback */
}
diff --git a/include/linux/kmemleak.h b/include/linux/kmemleak.h
new file mode 100644
index 000000000000..7796aed6cdd5
--- /dev/null
+++ b/include/linux/kmemleak.h
@@ -0,0 +1,96 @@
+/*
+ * include/linux/kmemleak.h
+ *
+ * Copyright (C) 2008 ARM Limited
+ * Written by Catalin Marinas <catalin.marinas@arm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#ifndef __KMEMLEAK_H
+#define __KMEMLEAK_H
+
+#ifdef CONFIG_DEBUG_KMEMLEAK
+
+extern void kmemleak_init(void);
+extern void kmemleak_alloc(const void *ptr, size_t size, int min_count,
+ gfp_t gfp);
+extern void kmemleak_free(const void *ptr);
+extern void kmemleak_padding(const void *ptr, unsigned long offset,
+ size_t size);
+extern void kmemleak_not_leak(const void *ptr);
+extern void kmemleak_ignore(const void *ptr);
+extern void kmemleak_scan_area(const void *ptr, unsigned long offset,
+ size_t length, gfp_t gfp);
+extern void kmemleak_no_scan(const void *ptr);
+
+static inline void kmemleak_alloc_recursive(const void *ptr, size_t size,
+ int min_count, unsigned long flags,
+ gfp_t gfp)
+{
+ if (!(flags & SLAB_NOLEAKTRACE))
+ kmemleak_alloc(ptr, size, min_count, gfp);
+}
+
+static inline void kmemleak_free_recursive(const void *ptr, unsigned long flags)
+{
+ if (!(flags & SLAB_NOLEAKTRACE))
+ kmemleak_free(ptr);
+}
+
+static inline void kmemleak_erase(void **ptr)
+{
+ *ptr = NULL;
+}
+
+#else
+
+static inline void kmemleak_init(void)
+{
+}
+static inline void kmemleak_alloc(const void *ptr, size_t size, int min_count,
+ gfp_t gfp)
+{
+}
+static inline void kmemleak_alloc_recursive(const void *ptr, size_t size,
+ int min_count, unsigned long flags,
+ gfp_t gfp)
+{
+}
+static inline void kmemleak_free(const void *ptr)
+{
+}
+static inline void kmemleak_free_recursive(const void *ptr, unsigned long flags)
+{
+}
+static inline void kmemleak_not_leak(const void *ptr)
+{
+}
+static inline void kmemleak_ignore(const void *ptr)
+{
+}
+static inline void kmemleak_scan_area(const void *ptr, unsigned long offset,
+ size_t length, gfp_t gfp)
+{
+}
+static inline void kmemleak_erase(void **ptr)
+{
+}
+static inline void kmemleak_no_scan(const void *ptr)
+{
+}
+
+#endif /* CONFIG_DEBUG_KMEMLEAK */
+
+#endif /* __KMEMLEAK_H */
diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index 1581ff235c7e..26fd9d12f050 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -86,7 +86,12 @@ struct percpu_data {
void *ptrs[1];
};
+/* pointer disguising messes up the kmemleak objects tracking */
+#ifndef CONFIG_DEBUG_KMEMLEAK
#define __percpu_disguise(pdata) (struct percpu_data *)~(unsigned long)(pdata)
+#else
+#define __percpu_disguise(pdata) (struct percpu_data *)(pdata)
+#endif
#define per_cpu_ptr(ptr, cpu) \
({ \
diff --git a/include/linux/slab.h b/include/linux/slab.h
index 24c5602bee99..48803064cedf 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -62,6 +62,8 @@
# define SLAB_DEBUG_OBJECTS 0x00000000UL
#endif
+#define SLAB_NOLEAKTRACE 0x00800000UL /* Avoid kmemleak tracing */
+
/* The following flags affect the page allocator grouping pages by mobility */
#define SLAB_RECLAIM_ACCOUNT 0x00020000UL /* Objects are reclaimable */
#define SLAB_TEMPORARY SLAB_RECLAIM_ACCOUNT /* Objects are short-lived */
diff --git a/init/main.c b/init/main.c
index 7917695bf71e..5616661eac01 100644
--- a/init/main.c
+++ b/init/main.c
@@ -56,6 +56,7 @@
#include <linux/debug_locks.h>
#include <linux/debugobjects.h>
#include <linux/lockdep.h>
+#include <linux/kmemleak.h>
#include <linux/pid_namespace.h>
#include <linux/device.h>
#include <linux/kthread.h>
@@ -621,6 +622,7 @@ asmlinkage void __init start_kernel(void)
/* init some links before init_ISA_irqs() */
early_irq_init();
init_IRQ();
+ prio_tree_init();
init_timers();
hrtimers_init();
softirq_init();
@@ -667,6 +669,7 @@ asmlinkage void __init start_kernel(void)
enable_debug_pagealloc();
cpu_hotplug_init();
kmemtrace_init();
+ kmemleak_init();
debug_objects_mem_init();
idr_init_cache();
setup_per_cpu_pageset();
@@ -676,7 +679,6 @@ asmlinkage void __init start_kernel(void)
calibrate_delay();
pidmap_init();
pgtable_cache_init();
- prio_tree_init();
anon_vma_init();
#ifdef CONFIG_X86
if (efi_enabled)
diff --git a/kernel/module.c b/kernel/module.c
index 278e9b6762bb..35f7de00bf0d 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -53,6 +53,7 @@
#include <linux/ftrace.h>
#include <linux/async.h>
#include <linux/percpu.h>
+#include <linux/kmemleak.h>
#if 0
#define DEBUGP printk
@@ -433,6 +434,7 @@ static void *percpu_modalloc(unsigned long size, unsigned long align,
unsigned long extra;
unsigned int i;
void *ptr;
+ int cpu;
if (align > PAGE_SIZE) {
printk(KERN_WARNING "%s: per-cpu alignment %li > %li\n",
@@ -462,6 +464,11 @@ static void *percpu_modalloc(unsigned long size, unsigned long align,
if (!split_block(i, size))
return NULL;
+ /* add the per-cpu scanning areas */
+ for_each_possible_cpu(cpu)
+ kmemleak_alloc(ptr + per_cpu_offset(cpu), size, 0,
+ GFP_KERNEL);
+
/* Mark allocated */
pcpu_size[i] = -pcpu_size[i];
return ptr;
@@ -476,6 +483,7 @@ static void percpu_modfree(void *freeme)
{
unsigned int i;
void *ptr = __per_cpu_start + block_size(pcpu_size[0]);
+ int cpu;
/* First entry is core kernel percpu data. */
for (i = 1; i < pcpu_num_used; ptr += block_size(pcpu_size[i]), i++) {
@@ -487,6 +495,10 @@ static void percpu_modfree(void *freeme)
BUG();
free:
+ /* remove the per-cpu scanning areas */
+ for_each_possible_cpu(cpu)
+ kmemleak_free(freeme + per_cpu_offset(cpu));
+
/* Merge with previous? */
if (pcpu_size[i-1] >= 0) {
pcpu_size[i-1] += pcpu_size[i];
@@ -1879,6 +1891,36 @@ static void *module_alloc_update_bounds(unsigned long size)
return ret;
}
+#ifdef CONFIG_DEBUG_KMEMLEAK
+static void kmemleak_load_module(struct module *mod, Elf_Ehdr *hdr,
+ Elf_Shdr *sechdrs, char *secstrings)
+{
+ unsigned int i;
+
+ /* only scan the sections containing data */
+ kmemleak_scan_area(mod->module_core, (unsigned long)mod -
+ (unsigned long)mod->module_core,
+ sizeof(struct module), GFP_KERNEL);
+
+ for (i = 1; i < hdr->e_shnum; i++) {
+ if (!(sechdrs[i].sh_flags & SHF_ALLOC))
+ continue;
+ if (strncmp(secstrings + sechdrs[i].sh_name, ".data", 5) != 0
+ && strncmp(secstrings + sechdrs[i].sh_name, ".bss", 4) != 0)
+ continue;
+
+ kmemleak_scan_area(mod->module_core, sechdrs[i].sh_addr -
+ (unsigned long)mod->module_core,
+ sechdrs[i].sh_size, GFP_KERNEL);
+ }
+}
+#else
+static inline void kmemleak_load_module(struct module *mod, Elf_Ehdr *hdr,
+ Elf_Shdr *sechdrs, char *secstrings)
+{
+}
+#endif
+
/* Allocate and load the module: note that size of section 0 is always
zero, and we rely on this for optional sections. */
static noinline struct module *load_module(void __user *umod,
@@ -2049,6 +2091,12 @@ static noinline struct module *load_module(void __user *umod,
/* Do the allocs. */
ptr = module_alloc_update_bounds(mod->core_size);
+ /*
+ * The pointer to this block is stored in the module structure
+ * which is inside the block. Just mark it as not being a
+ * leak.
+ */
+ kmemleak_not_leak(ptr);
if (!ptr) {
err = -ENOMEM;
goto free_percpu;
@@ -2057,6 +2105,13 @@ static noinline struct module *load_module(void __user *umod,
mod->module_core = ptr;
ptr = module_alloc_update_bounds(mod->init_size);
+ /*
+ * The pointer to this block is stored in the module structure
+ * which is inside the block. This block doesn't need to be
+ * scanned as it contains data and code that will be freed
+ * after the module is initialized.
+ */
+ kmemleak_ignore(ptr);
if (!ptr && mod->init_size) {
err = -ENOMEM;
goto free_core;
@@ -2087,6 +2142,7 @@ static noinline struct module *load_module(void __user *umod,
}
/* Module has been moved. */
mod = (void *)sechdrs[modindex].sh_addr;
+ kmemleak_load_module(mod, hdr, sechdrs, secstrings);
#if defined(CONFIG_MODULE_UNLOAD) && defined(CONFIG_SMP)
mod->refptr = percpu_modalloc(sizeof(local_t), __alignof__(local_t),
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 6cdcf38f2da9..116a35051be6 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -336,6 +336,38 @@ config SLUB_STATS
out which slabs are relevant to a particular load.
Try running: slabinfo -DA
+config DEBUG_KMEMLEAK
+ bool "Kernel memory leak detector"
+ depends on DEBUG_KERNEL && EXPERIMENTAL && (X86 || ARM) && \
+ !MEMORY_HOTPLUG
+ select DEBUG_SLAB if SLAB
+ select SLUB_DEBUG if SLUB
+ select DEBUG_FS if SYSFS
+ select STACKTRACE if STACKTRACE_SUPPORT
+ select KALLSYMS
+ help
+ Say Y here if you want to enable the memory leak
+ detector. The memory allocation/freeing is traced in a way
+ similar to the Boehm's conservative garbage collector, the
+ difference being that the orphan objects are not freed but
+ only shown in /sys/kernel/debug/kmemleak. Enabling this
+ feature will introduce an overhead to memory
+ allocations. See Documentation/kmemleak.txt for more
+ details.
+
+ In order to access the kmemleak file, debugfs needs to be
+ mounted (usually at /sys/kernel/debug).
+
+config DEBUG_KMEMLEAK_TEST
+ tristate "Simple test for the kernel memory leak detector"
+ depends on DEBUG_KMEMLEAK
+ help
+ Say Y or M here to build a test for the kernel memory leak
+ detector. This option enables a module that explicitly leaks
+ memory.
+
+ If unsure, say N.
+
config DEBUG_PREEMPT
bool "Debug preemptible kernel"
depends on DEBUG_KERNEL && PREEMPT && (TRACE_IRQFLAGS_SUPPORT || PPC64)
diff --git a/mm/Makefile b/mm/Makefile
index ec73c68b6015..e89acb090b4d 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -38,3 +38,5 @@ obj-$(CONFIG_SMP) += allocpercpu.o
endif
obj-$(CONFIG_QUICKLIST) += quicklist.o
obj-$(CONFIG_CGROUP_MEM_RES_CTLR) += memcontrol.o page_cgroup.o
+obj-$(CONFIG_DEBUG_KMEMLEAK) += kmemleak.o
+obj-$(CONFIG_DEBUG_KMEMLEAK_TEST) += kmemleak-test.o
diff --git a/mm/kmemleak-test.c b/mm/kmemleak-test.c
new file mode 100644
index 000000000000..d5292fc6f523
--- /dev/null
+++ b/mm/kmemleak-test.c
@@ -0,0 +1,111 @@
+/*
+ * mm/kmemleak-test.c
+ *
+ * Copyright (C) 2008 ARM Limited
+ * Written by Catalin Marinas <catalin.marinas@arm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/list.h>
+#include <linux/percpu.h>
+#include <linux/fdtable.h>
+
+#include <linux/kmemleak.h>
+
+struct test_node {
+ long header[25];
+ struct list_head list;
+ long footer[25];
+};
+
+static LIST_HEAD(test_list);
+static DEFINE_PER_CPU(void *, test_pointer);
+
+/*
+ * Some very simple testing. This function needs to be extended for
+ * proper testing.
+ */
+static int __init kmemleak_test_init(void)
+{
+ struct test_node *elem;
+ int i;
+
+ printk(KERN_INFO "Kmemleak testing\n");
+
+ /* make some orphan objects */
+ pr_info("kmemleak: kmalloc(32) = %p\n", kmalloc(32, GFP_KERNEL));
+ pr_info("kmemleak: kmalloc(32) = %p\n", kmalloc(32, GFP_KERNEL));
+ pr_info("kmemleak: kmalloc(1024) = %p\n", kmalloc(1024, GFP_KERNEL));
+ pr_info("kmemleak: kmalloc(1024) = %p\n", kmalloc(1024, GFP_KERNEL));
+ pr_info("kmemleak: kmalloc(2048) = %p\n", kmalloc(2048, GFP_KERNEL));
+ pr_info("kmemleak: kmalloc(2048) = %p\n", kmalloc(2048, GFP_KERNEL));
+ pr_info("kmemleak: kmalloc(4096) = %p\n", kmalloc(4096, GFP_KERNEL));
+ pr_info("kmemleak: kmalloc(4096) = %p\n", kmalloc(4096, GFP_KERNEL));
+#ifndef CONFIG_MODULES
+ pr_info("kmemleak: kmem_cache_alloc(files_cachep) = %p\n",
+ kmem_cache_alloc(files_cachep, GFP_KERNEL));
+ pr_info("kmemleak: kmem_cache_alloc(files_cachep) = %p\n",
+ kmem_cache_alloc(files_cachep, GFP_KERNEL));
+#endif
+ pr_info("kmemleak: vmalloc(64) = %p\n", vmalloc(64));
+ pr_info("kmemleak: vmalloc(64) = %p\n", vmalloc(64));
+ pr_info("kmemleak: vmalloc(64) = %p\n", vmalloc(64));
+ pr_info("kmemleak: vmalloc(64) = %p\n", vmalloc(64));
+ pr_info("kmemleak: vmalloc(64) = %p\n", vmalloc(64));
+
+ /*
+ * Add elements to a list. They should only appear as orphan
+ * after the module is removed.
+ */
+ for (i = 0; i < 10; i++) {
+ elem = kmalloc(sizeof(*elem), GFP_KERNEL);
+ pr_info("kmemleak: kmalloc(sizeof(*elem)) = %p\n", elem);
+ if (!elem)
+ return -ENOMEM;
+ memset(elem, 0, sizeof(*elem));
+ INIT_LIST_HEAD(&elem->list);
+
+ list_add_tail(&elem->list, &test_list);
+ }
+
+ for_each_possible_cpu(i) {
+ per_cpu(test_pointer, i) = kmalloc(129, GFP_KERNEL);
+ pr_info("kmemleak: kmalloc(129) = %p\n",
+ per_cpu(test_pointer, i));
+ }
+
+ return 0;
+}
+module_init(kmemleak_test_init);
+
+static void __exit kmemleak_test_exit(void)
+{
+ struct test_node *elem, *tmp;
+
+ /*
+ * Remove the list elements without actually freeing the
+ * memory.
+ */
+ list_for_each_entry_safe(elem, tmp, &test_list, list)
+ list_del(&elem->list);
+}
+module_exit(kmemleak_test_exit);
+
+MODULE_LICENSE("GPL");
diff --git a/mm/kmemleak.c b/mm/kmemleak.c
new file mode 100644
index 000000000000..58ec86c9e58a
--- /dev/null
+++ b/mm/kmemleak.c
@@ -0,0 +1,1498 @@
+/*
+ * mm/kmemleak.c
+ *
+ * Copyright (C) 2008 ARM Limited
+ * Written by Catalin Marinas <catalin.marinas@arm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ *
+ * For more information on the algorithm and kmemleak usage, please see
+ * Documentation/kmemleak.txt.
+ *
+ * Notes on locking
+ * ----------------
+ *
+ * The following locks and mutexes are used by kmemleak:
+ *
+ * - kmemleak_lock (rwlock): protects the object_list modifications and
+ * accesses to the object_tree_root. The object_list is the main list
+ * holding the metadata (struct kmemleak_object) for the allocated memory
+ * blocks. The object_tree_root is a priority search tree used to look-up
+ * metadata based on a pointer to the corresponding memory block. The
+ * kmemleak_object structures are added to the object_list and
+ * object_tree_root in the create_object() function called from the
+ * kmemleak_alloc() callback and removed in delete_object() called from the
+ * kmemleak_free() callback
+ * - kmemleak_object.lock (spinlock): protects a kmemleak_object. Accesses to
+ * the metadata (e.g. count) are protected by this lock. Note that some
+ * members of this structure may be protected by other means (atomic or
+ * kmemleak_lock). This lock is also held when scanning the corresponding
+ * memory block to avoid the kernel freeing it via the kmemleak_free()
+ * callback. This is less heavyweight than holding a global lock like
+ * kmemleak_lock during scanning
+ * - scan_mutex (mutex): ensures that only one thread may scan the memory for
+ * unreferenced objects at a time. The gray_list contains the objects which
+ * are already referenced or marked as false positives and need to be
+ * scanned. This list is only modified during a scanning episode when the
+ * scan_mutex is held. At the end of a scan, the gray_list is always empty.
+ * Note that the kmemleak_object.use_count is incremented when an object is
+ * added to the gray_list and therefore cannot be freed
+ * - kmemleak_mutex (mutex): prevents multiple users of the "kmemleak" debugfs
+ * file together with modifications to the memory scanning parameters
+ * including the scan_thread pointer
+ *
+ * The kmemleak_object structures have a use_count incremented or decremented
+ * using the get_object()/put_object() functions. When the use_count becomes
+ * 0, this count can no longer be incremented and put_object() schedules the
+ * kmemleak_object freeing via an RCU callback. All calls to the get_object()
+ * function must be protected by rcu_read_lock() to avoid accessing a freed
+ * structure.
+ */
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/list.h>
+#include <linux/sched.h>
+#include <linux/jiffies.h>
+#include <linux/delay.h>
+#include <linux/module.h>
+#include <linux/kthread.h>
+#include <linux/prio_tree.h>
+#include <linux/gfp.h>
+#include <linux/fs.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+#include <linux/cpumask.h>
+#include <linux/spinlock.h>
+#include <linux/mutex.h>
+#include <linux/rcupdate.h>
+#include <linux/stacktrace.h>
+#include <linux/cache.h>
+#include <linux/percpu.h>
+#include <linux/hardirq.h>
+#include <linux/mmzone.h>
+#include <linux/slab.h>
+#include <linux/thread_info.h>
+#include <linux/err.h>
+#include <linux/uaccess.h>
+#include <linux/string.h>
+#include <linux/nodemask.h>
+#include <linux/mm.h>
+
+#include <asm/sections.h>
+#include <asm/processor.h>
+#include <asm/atomic.h>
+
+#include <linux/kmemleak.h>
+
+/*
+ * Kmemleak configuration and common defines.
+ */
+#define MAX_TRACE 16 /* stack trace length */
+#define REPORTS_NR 50 /* maximum number of reported leaks */
+#define MSECS_MIN_AGE 5000 /* minimum object age for reporting */
+#define MSECS_SCAN_YIELD 10 /* CPU yielding period */
+#define SECS_FIRST_SCAN 60 /* delay before the first scan */
+#define SECS_SCAN_WAIT 600 /* subsequent auto scanning delay */
+
+#define BYTES_PER_POINTER sizeof(void *)
+
+/* scanning area inside a memory block */
+struct kmemleak_scan_area {
+ struct hlist_node node;
+ unsigned long offset;
+ size_t length;
+};
+
+/*
+ * Structure holding the metadata for each allocated memory block.
+ * Modifications to such objects should be made while holding the
+ * object->lock. Insertions or deletions from object_list, gray_list or
+ * tree_node are already protected by the corresponding locks or mutex (see
+ * the notes on locking above). These objects are reference-counted
+ * (use_count) and freed using the RCU mechanism.
+ */
+struct kmemleak_object {
+ spinlock_t lock;
+ unsigned long flags; /* object status flags */
+ struct list_head object_list;
+ struct list_head gray_list;
+ struct prio_tree_node tree_node;
+ struct rcu_head rcu; /* object_list lockless traversal */
+ /* object usage count; object freed when use_count == 0 */
+ atomic_t use_count;
+ unsigned long pointer;
+ size_t size;
+ /* minimum number of a pointers found before it is considered leak */
+ int min_count;
+ /* the total number of pointers found pointing to this object */
+ int count;
+ /* memory ranges to be scanned inside an object (empty for all) */
+ struct hlist_head area_list;
+ unsigned long trace[MAX_TRACE];
+ unsigned int trace_len;
+ unsigned long jiffies; /* creation timestamp */
+ pid_t pid; /* pid of the current task */
+ char comm[TASK_COMM_LEN]; /* executable name */
+};
+
+/* flag representing the memory block allocation status */
+#define OBJECT_ALLOCATED (1 << 0)
+/* flag set after the first reporting of an unreference object */
+#define OBJECT_REPORTED (1 << 1)
+/* flag set to not scan the object */
+#define OBJECT_NO_SCAN (1 << 2)
+
+/* the list of all allocated objects */
+static LIST_HEAD(object_list);
+/* the list of gray-colored objects (see color_gray comment below) */
+static LIST_HEAD(gray_list);
+/* prio search tree for object boundaries */
+static struct prio_tree_root object_tree_root;
+/* rw_lock protecting the access to object_list and prio_tree_root */
+static DEFINE_RWLOCK(kmemleak_lock);
+
+/* allocation caches for kmemleak internal data */
+static struct kmem_cache *object_cache;
+static struct kmem_cache *scan_area_cache;
+
+/* set if tracing memory operations is enabled */
+static atomic_t kmemleak_enabled = ATOMIC_INIT(0);
+/* set in the late_initcall if there were no errors */
+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 */
+static atomic_t kmemleak_error = ATOMIC_INIT(0);
+
+/* minimum and maximum address that may be valid pointers */
+static unsigned long min_addr = ULONG_MAX;
+static unsigned long max_addr;
+
+/* used for yielding the CPU to other tasks during scanning */
+static unsigned long next_scan_yield;
+static struct task_struct *scan_thread;
+static unsigned long jiffies_scan_yield;
+static unsigned long jiffies_min_age;
+/* delay between automatic memory scannings */
+static signed long jiffies_scan_wait;
+/* enables or disables the task stacks scanning */
+static int kmemleak_stack_scan;
+/* mutex protecting the memory scanning */
+static DEFINE_MUTEX(scan_mutex);
+/* mutex protecting the access to the /sys/kernel/debug/kmemleak file */
+static DEFINE_MUTEX(kmemleak_mutex);
+
+/* number of leaks reported (for limitation purposes) */
+static int reported_leaks;
+
+/*
+ * Early object allocation/freeing logging. Kkmemleak is initialized after the
+ * kernel allocator. However, both the kernel allocator and kmemleak may
+ * allocate memory blocks which need to be tracked. Kkmemleak defines an
+ * arbitrary buffer to hold the allocation/freeing information before it is
+ * fully initialized.
+ */
+
+/* kmemleak operation type for early logging */
+enum {
+ KMEMLEAK_ALLOC,
+ KMEMLEAK_FREE,
+ KMEMLEAK_NOT_LEAK,
+ KMEMLEAK_IGNORE,
+ KMEMLEAK_SCAN_AREA,
+ KMEMLEAK_NO_SCAN
+};
+
+/*
+ * Structure holding the information passed to kmemleak callbacks during the
+ * early logging.
+ */
+struct early_log {
+ int op_type; /* kmemleak operation type */
+ const void *ptr; /* allocated/freed memory block */
+ size_t size; /* memory block size */
+ int min_count; /* minimum reference count */
+ unsigned long offset; /* scan area offset */
+ size_t length; /* scan area length */
+};
+
+/* early logging buffer and current position */
+static struct early_log early_log[200];
+static int crt_early_log;
+
+static void kmemleak_disable(void);
+
+/*
+ * Print a warning and dump the stack trace.
+ */
+#define kmemleak_warn(x...) do { \
+ pr_warning(x); \
+ dump_stack(); \
+} while (0)
+
+/*
+ * Macro invoked when a serious kmemleak condition occured and cannot be
+ * recovered from. Kkmemleak will be disabled and further allocation/freeing
+ * tracing no longer available.
+ */
+#define kmemleak_panic(x...) do { \
+ kmemleak_warn(x); \
+ kmemleak_disable(); \
+} while (0)
+
+/*
+ * Object colors, encoded with count and min_count:
+ * - white - orphan object, not enough references to it (count < min_count)
+ * - gray - not orphan, not marked as false positive (min_count == 0) or
+ * sufficient references to it (count >= min_count)
+ * - black - ignore, it doesn't contain references (e.g. text section)
+ * (min_count == -1). No function defined for this color.
+ * Newly created objects don't have any color assigned (object->count == -1)
+ * before the next memory scan when they become white.
+ */
+static int color_white(const struct kmemleak_object *object)
+{
+ return object->count != -1 && object->count < object->min_count;
+}
+
+static int color_gray(const struct kmemleak_object *object)
+{
+ return object->min_count != -1 && object->count >= object->min_count;
+}
+
+/*
+ * Objects are considered referenced if their color is gray and they have not
+ * been deleted.
+ */
+static int referenced_object(struct kmemleak_object *object)
+{
+ return (object->flags & OBJECT_ALLOCATED) && color_gray(object);
+}
+
+/*
+ * Objects are considered unreferenced only if their color is white, they have
+ * not be deleted and have a minimum age to avoid false positives caused by
+ * pointers temporarily stored in CPU registers.
+ */
+static int unreferenced_object(struct kmemleak_object *object)
+{
+ return (object->flags & OBJECT_ALLOCATED) && color_white(object) &&
+ time_is_before_eq_jiffies(object->jiffies + jiffies_min_age);
+}
+
+/*
+ * Printing of the (un)referenced objects information, either to the seq file
+ * or to the kernel log. The print_referenced/print_unreferenced functions
+ * must be called with the object->lock held.
+ */
+#define print_helper(seq, x...) do { \
+ struct seq_file *s = (seq); \
+ if (s) \
+ seq_printf(s, x); \
+ else \
+ pr_info(x); \
+} while (0)
+
+static void print_referenced(struct kmemleak_object *object)
+{
+ pr_info("kmemleak: referenced object 0x%08lx (size %zu)\n",
+ object->pointer, object->size);
+}
+
+static void print_unreferenced(struct seq_file *seq,
+ struct kmemleak_object *object)
+{
+ int i;
+
+ print_helper(seq, "kmemleak: unreferenced object 0x%08lx (size %zu):\n",
+ object->pointer, object->size);
+ print_helper(seq, " comm \"%s\", pid %d, jiffies %lu\n",
+ object->comm, object->pid, object->jiffies);
+ print_helper(seq, " backtrace:\n");
+
+ for (i = 0; i < object->trace_len; i++) {
+ void *ptr = (void *)object->trace[i];
+ print_helper(seq, " [<%p>] %pS\n", ptr, ptr);
+ }
+}
+
+/*
+ * Print the kmemleak_object information. This function is used mainly for
+ * debugging special cases when kmemleak operations. It must be called with
+ * the object->lock held.
+ */
+static void dump_object_info(struct kmemleak_object *object)
+{
+ struct stack_trace trace;
+
+ trace.nr_entries = object->trace_len;
+ trace.entries = object->trace;
+
+ pr_notice("kmemleak: Object 0x%08lx (size %zu):\n",
+ object->tree_node.start, object->size);
+ pr_notice(" comm \"%s\", pid %d, jiffies %lu\n",
+ object->comm, object->pid, object->jiffies);
+ pr_notice(" min_count = %d\n", object->min_count);
+ pr_notice(" count = %d\n", object->count);
+ pr_notice(" backtrace:\n");
+ print_stack_trace(&trace, 4);
+}
+
+/*
+ * Look-up a memory block metadata (kmemleak_object) in the priority search
+ * tree based on a pointer value. If alias is 0, only values pointing to the
+ * beginning of the memory block are allowed. The kmemleak_lock must be held
+ * when calling this function.
+ */
+static struct kmemleak_object *lookup_object(unsigned long ptr, int alias)
+{
+ struct prio_tree_node *node;
+ struct prio_tree_iter iter;
+ struct kmemleak_object *object;
+
+ prio_tree_iter_init(&iter, &object_tree_root, ptr, ptr);
+ node = prio_tree_next(&iter);
+ if (node) {
+ object = prio_tree_entry(node, struct kmemleak_object,
+ tree_node);
+ if (!alias && object->pointer != ptr) {
+ kmemleak_warn("kmemleak: Found object by alias");
+ object = NULL;
+ }
+ } else
+ object = NULL;
+
+ return object;
+}
+
+/*
+ * Increment the object use_count. Return 1 if successful or 0 otherwise. Note
+ * that once an object's use_count reached 0, the RCU freeing was already
+ * registered and the object should no longer be used. This function must be
+ * called under the protection of rcu_read_lock().
+ */
+static int get_object(struct kmemleak_object *object)
+{
+ return atomic_inc_not_zero(&object->use_count);
+}
+
+/*
+ * RCU callback to free a kmemleak_object.
+ */
+static void free_object_rcu(struct rcu_head *rcu)
+{
+ struct hlist_node *elem, *tmp;
+ struct kmemleak_scan_area *area;
+ struct kmemleak_object *object =
+ container_of(rcu, struct kmemleak_object, rcu);
+
+ /*
+ * Once use_count is 0 (guaranteed by put_object), there is no other
+ * code accessing this object, hence no need for locking.
+ */
+ hlist_for_each_entry_safe(area, elem, tmp, &object->area_list, node) {
+ hlist_del(elem);
+ kmem_cache_free(scan_area_cache, area);
+ }
+ kmem_cache_free(object_cache, object);
+}
+
+/*
+ * Decrement the object use_count. Once the count is 0, free the object using
+ * an RCU callback. Since put_object() may be called via the kmemleak_free() ->
+ * delete_object() path, the delayed RCU freeing ensures that there is no
+ * recursive call to the kernel allocator. Lock-less RCU object_list traversal
+ * is also possible.
+ */
+static void put_object(struct kmemleak_object *object)
+{
+ if (!atomic_dec_and_test(&object->use_count))
+ return;
+
+ /* should only get here after delete_object was called */
+ WARN_ON(object->flags & OBJECT_ALLOCATED);
+
+ call_rcu(&object->rcu, free_object_rcu);
+}
+
+/*
+ * Look up an object in the prio search tree and increase its use_count.
+ */
+static struct kmemleak_object *find_and_get_object(unsigned long ptr, int alias)
+{
+ unsigned long flags;
+ struct kmemleak_object *object = NULL;
+
+ rcu_read_lock();
+ read_lock_irqsave(&kmemleak_lock, flags);
+ if (ptr >= min_addr && ptr < max_addr)
+ object = lookup_object(ptr, alias);
+ read_unlock_irqrestore(&kmemleak_lock, flags);
+
+ /* check whether the object is still available */
+ if (object && !get_object(object))
+ object = NULL;
+ rcu_read_unlock();
+
+ return object;
+}
+
+/*
+ * Create the metadata (struct kmemleak_object) corresponding to an allocated
+ * memory block and add it to the object_list and object_tree_root.
+ */
+static void create_object(unsigned long ptr, size_t size, int min_count,
+ gfp_t gfp)
+{
+ unsigned long flags;
+ struct kmemleak_object *object;
+ struct prio_tree_node *node;
+ struct stack_trace trace;
+
+ object = kmem_cache_alloc(object_cache, gfp & ~GFP_SLAB_BUG_MASK);
+ if (!object) {
+ kmemleak_panic("kmemleak: Cannot allocate a kmemleak_object "
+ "structure\n");
+ return;
+ }
+
+ INIT_LIST_HEAD(&object->object_list);
+ INIT_LIST_HEAD(&object->gray_list);
+ INIT_HLIST_HEAD(&object->area_list);
+ spin_lock_init(&object->lock);
+ atomic_set(&object->use_count, 1);
+ object->flags = OBJECT_ALLOCATED;
+ object->pointer = ptr;
+ object->size = size;
+ object->min_count = min_count;
+ object->count = -1; /* no color initially */
+ object->jiffies = jiffies;
+
+ /* task information */
+ if (in_irq()) {
+ object->pid = 0;
+ strncpy(object->comm, "hardirq", sizeof(object->comm));
+ } else if (in_softirq()) {
+ object->pid = 0;
+ strncpy(object->comm, "softirq", sizeof(object->comm));
+ } else {
+ object->pid = current->pid;
+ /*
+ * There is a small chance of a race with set_task_comm(),
+ * however using get_task_comm() here may cause locking
+ * dependency issues with current->alloc_lock. In the worst
+ * case, the command line is not correct.
+ */
+ strncpy(object->comm, current->comm, sizeof(object->comm));
+ }
+
+ /* kernel backtrace */
+ trace.max_entries = MAX_TRACE;
+ trace.nr_entries = 0;
+ trace.entries = object->trace;
+ trace.skip = 1;
+ save_stack_trace(&trace);
+ object->trace_len = trace.nr_entries;
+
+ INIT_PRIO_TREE_NODE(&object->tree_node);
+ object->tree_node.start = ptr;
+ object->tree_node.last = ptr + size - 1;
+
+ write_lock_irqsave(&kmemleak_lock, flags);
+ min_addr = min(min_addr, ptr);
+ max_addr = max(max_addr, ptr + size);
+ node = prio_tree_insert(&object_tree_root, &object->tree_node);
+ /*
+ * The code calling the kernel does not yet have the pointer to the
+ * memory block to be able to free it. However, we still hold the
+ * kmemleak_lock here in case parts of the kernel started freeing
+ * random memory blocks.
+ */
+ if (node != &object->tree_node) {
+ unsigned long flags;
+
+ kmemleak_panic("kmemleak: Cannot insert 0x%lx into the object "
+ "search tree (already existing)\n", ptr);
+ object = lookup_object(ptr, 1);
+ spin_lock_irqsave(&object->lock, flags);
+ dump_object_info(object);
+ spin_unlock_irqrestore(&object->lock, flags);
+
+ goto out;
+ }
+ list_add_tail_rcu(&object->object_list, &object_list);
+out:
+ write_unlock_irqrestore(&kmemleak_lock, flags);
+}
+
+/*
+ * Remove the metadata (struct kmemleak_object) for a memory block from the
+ * object_list and object_tree_root and decrement its use_count.
+ */
+static void delete_object(unsigned long ptr)
+{
+ unsigned long flags;
+ struct kmemleak_object *object;
+
+ write_lock_irqsave(&kmemleak_lock, flags);
+ object = lookup_object(ptr, 0);
+ if (!object) {
+ kmemleak_warn("kmemleak: Freeing unknown object at 0x%08lx\n",
+ ptr);
+ write_unlock_irqrestore(&kmemleak_lock, flags);
+ return;
+ }
+ prio_tree_remove(&object_tree_root, &object->tree_node);
+ list_del_rcu(&object->object_list);
+ write_unlock_irqrestore(&kmemleak_lock, flags);
+
+ WARN_ON(!(object->flags & OBJECT_ALLOCATED));
+ WARN_ON(atomic_read(&object->use_count) < 1);
+
+ /*
+ * Locking here also ensures that the corresponding memory block
+ * cannot be freed when it is being scanned.
+ */
+ spin_lock_irqsave(&object->lock, flags);
+ if (object->flags & OBJECT_REPORTED)
+ print_referenced(object);
+ object->flags &= ~OBJECT_ALLOCATED;
+ spin_unlock_irqrestore(&object->lock, flags);
+ put_object(object);
+}
+
+/*
+ * Make a object permanently as gray-colored so that it can no longer be
+ * reported as a leak. This is used in general to mark a false positive.
+ */
+static void make_gray_object(unsigned long ptr)
+{
+ unsigned long flags;
+ struct kmemleak_object *object;
+
+ object = find_and_get_object(ptr, 0);
+ if (!object) {
+ kmemleak_warn("kmemleak: Graying unknown object at 0x%08lx\n",
+ ptr);
+ return;
+ }
+
+ spin_lock_irqsave(&object->lock, flags);
+ object->min_count = 0;
+ spin_unlock_irqrestore(&object->lock, flags);
+ put_object(object);
+}
+
+/*
+ * Mark the object as black-colored so that it is ignored from scans and
+ * reporting.
+ */
+static void make_black_object(unsigned long ptr)
+{
+ unsigned long flags;
+ struct kmemleak_object *object;
+
+ object = find_and_get_object(ptr, 0);
+ if (!object) {
+ kmemleak_warn("kmemleak: Blacking unknown object at 0x%08lx\n",
+ ptr);
+ return;
+ }
+
+ spin_lock_irqsave(&object->lock, flags);
+ object->min_count = -1;
+ spin_unlock_irqrestore(&object->lock, flags);
+ put_object(object);
+}
+
+/*
+ * Add a scanning area to the object. If at least one such area is added,
+ * kmemleak will only scan these ranges rather than the whole memory block.
+ */
+static void add_scan_area(unsigned long ptr, unsigned long offset,
+ size_t length, gfp_t gfp)
+{
+ unsigned long flags;
+ struct kmemleak_object *object;
+ struct kmemleak_scan_area *area;
+
+ object = find_and_get_object(ptr, 0);
+ if (!object) {
+ kmemleak_warn("kmemleak: Adding scan area to unknown "
+ "object at 0x%08lx\n", ptr);
+ return;
+ }
+
+ area = kmem_cache_alloc(scan_area_cache, gfp & ~GFP_SLAB_BUG_MASK);
+ if (!area) {
+ kmemleak_warn("kmemleak: Cannot allocate a scan area\n");
+ goto out;
+ }
+
+ spin_lock_irqsave(&object->lock, flags);
+ if (offset + length > object->size) {
+ kmemleak_warn("kmemleak: Scan area larger than object "
+ "0x%08lx\n", ptr);
+ dump_object_info(object);
+ kmem_cache_free(scan_area_cache, area);
+ goto out_unlock;
+ }
+
+ INIT_HLIST_NODE(&area->node);
+ area->offset = offset;
+ area->length = length;
+
+ hlist_add_head(&area->node, &object->area_list);
+out_unlock:
+ spin_unlock_irqrestore(&object->lock, flags);
+out:
+ put_object(object);
+}
+
+/*
+ * Set the OBJECT_NO_SCAN flag for the object corresponding to the give
+ * pointer. Such object will not be scanned by kmemleak but references to it
+ * are searched.
+ */
+static void object_no_scan(unsigned long ptr)
+{
+ unsigned long flags;
+ struct kmemleak_object *object;
+
+ object = find_and_get_object(ptr, 0);
+ if (!object) {
+ kmemleak_warn("kmemleak: Not scanning unknown object at "
+ "0x%08lx\n", ptr);
+ return;
+ }
+
+ spin_lock_irqsave(&object->lock, flags);
+ object->flags |= OBJECT_NO_SCAN;
+ spin_unlock_irqrestore(&object->lock, flags);
+ put_object(object);
+}
+
+/*
+ * Log an early kmemleak_* call to the early_log buffer. These calls will be
+ * processed later once kmemleak is fully initialized.
+ */
+static void log_early(int op_type, const void *ptr, size_t size,
+ int min_count, unsigned long offset, size_t length)
+{
+ unsigned long flags;
+ struct early_log *log;
+
+ if (crt_early_log >= ARRAY_SIZE(early_log)) {
+ kmemleak_panic("kmemleak: Early log buffer exceeded\n");
+ return;
+ }
+
+ /*
+ * There is no need for locking since the kernel is still in UP mode
+ * at this stage. Disabling the IRQs is enough.
+ */
+ local_irq_save(flags);
+ log = &early_log[crt_early_log];
+ log->op_type = op_type;
+ log->ptr = ptr;
+ log->size = size;
+ log->min_count = min_count;
+ log->offset = offset;
+ log->length = length;
+ crt_early_log++;
+ local_irq_restore(flags);
+}
+
+/*
+ * Memory allocation function callback. This function is called from the
+ * kernel allocators when a new block is allocated (kmem_cache_alloc, kmalloc,
+ * vmalloc etc.).
+ */
+void kmemleak_alloc(const void *ptr, size_t size, int min_count, gfp_t gfp)
+{
+ pr_debug("%s(0x%p, %zu, %d)\n", __func__, ptr, size, min_count);
+
+ if (atomic_read(&kmemleak_enabled) && ptr && !IS_ERR(ptr))
+ create_object((unsigned long)ptr, size, min_count, gfp);
+ else if (atomic_read(&kmemleak_early_log))
+ log_early(KMEMLEAK_ALLOC, ptr, size, min_count, 0, 0);
+}
+EXPORT_SYMBOL_GPL(kmemleak_alloc);
+
+/*
+ * Memory freeing function callback. This function is called from the kernel
+ * allocators when a block is freed (kmem_cache_free, kfree, vfree etc.).
+ */
+void kmemleak_free(const void *ptr)
+{
+ pr_debug("%s(0x%p)\n", __func__, ptr);
+
+ if (atomic_read(&kmemleak_enabled) && ptr && !IS_ERR(ptr))
+ delete_object((unsigned long)ptr);
+ else if (atomic_read(&kmemleak_early_log))
+ log_early(KMEMLEAK_FREE, ptr, 0, 0, 0, 0);
+}
+EXPORT_SYMBOL_GPL(kmemleak_free);
+
+/*
+ * Mark an already allocated memory block as a false positive. This will cause
+ * the block to no longer be reported as leak and always be scanned.
+ */
+void kmemleak_not_leak(const void *ptr)
+{
+ pr_debug("%s(0x%p)\n", __func__, ptr);
+
+ if (atomic_read(&kmemleak_enabled) && ptr && !IS_ERR(ptr))
+ make_gray_object((unsigned long)ptr);
+ else if (atomic_read(&kmemleak_early_log))
+ log_early(KMEMLEAK_NOT_LEAK, ptr, 0, 0, 0, 0);
+}
+EXPORT_SYMBOL(kmemleak_not_leak);
+
+/*
+ * Ignore a memory block. This is usually done when it is known that the
+ * corresponding block is not a leak and does not contain any references to
+ * other allocated memory blocks.
+ */
+void kmemleak_ignore(const void *ptr)
+{
+ pr_debug("%s(0x%p)\n", __func__, ptr);
+
+ if (atomic_read(&kmemleak_enabled) && ptr && !IS_ERR(ptr))
+ make_black_object((unsigned long)ptr);
+ else if (atomic_read(&kmemleak_early_log))
+ log_early(KMEMLEAK_IGNORE, ptr, 0, 0, 0, 0);
+}
+EXPORT_SYMBOL(kmemleak_ignore);
+
+/*
+ * Limit the range to be scanned in an allocated memory block.
+ */
+void kmemleak_scan_area(const void *ptr, unsigned long offset, size_t length,
+ gfp_t gfp)
+{
+ pr_debug("%s(0x%p)\n", __func__, ptr);
+
+ if (atomic_read(&kmemleak_enabled) && ptr && !IS_ERR(ptr))
+ add_scan_area((unsigned long)ptr, offset, length, gfp);
+ else if (atomic_read(&kmemleak_early_log))
+ log_early(KMEMLEAK_SCAN_AREA, ptr, 0, 0, offset, length);
+}
+EXPORT_SYMBOL(kmemleak_scan_area);
+
+/*
+ * Inform kmemleak not to scan the given memory block.
+ */
+void kmemleak_no_scan(const void *ptr)
+{
+ pr_debug("%s(0x%p)\n", __func__, ptr);
+
+ if (atomic_read(&kmemleak_enabled) && ptr && !IS_ERR(ptr))
+ object_no_scan((unsigned long)ptr);
+ else if (atomic_read(&kmemleak_early_log))
+ log_early(KMEMLEAK_NO_SCAN, ptr, 0, 0, 0, 0);
+}
+EXPORT_SYMBOL(kmemleak_no_scan);
+
+/*
+ * Yield the CPU so that other tasks get a chance to run. The yielding is
+ * rate-limited to avoid excessive number of calls to the schedule() function
+ * during memory scanning.
+ */
+static void scan_yield(void)
+{
+ might_sleep();
+
+ if (time_is_before_eq_jiffies(next_scan_yield)) {
+ schedule();
+ next_scan_yield = jiffies + jiffies_scan_yield;
+ }
+}
+
+/*
+ * Memory scanning is a long process and it needs to be interruptable. This
+ * function checks whether such interrupt condition occured.
+ */
+static int scan_should_stop(void)
+{
+ if (!atomic_read(&kmemleak_enabled))
+ return 1;
+
+ /*
+ * This function may be called from either process or kthread context,
+ * hence the need to check for both stop conditions.
+ */
+ if (current->mm)
+ return signal_pending(current);
+ else
+ return kthread_should_stop();
+
+ return 0;
+}
+
+/*
+ * Scan a memory block (exclusive range) for valid pointers and add those
+ * found to the gray list.
+ */
+static void scan_block(void *_start, void *_end,
+ struct kmemleak_object *scanned)
+{
+ unsigned long *ptr;
+ unsigned long *start = PTR_ALIGN(_start, BYTES_PER_POINTER);
+ unsigned long *end = _end - (BYTES_PER_POINTER - 1);
+
+ for (ptr = start; ptr < end; ptr++) {
+ unsigned long flags;
+ unsigned long pointer = *ptr;
+ struct kmemleak_object *object;
+
+ if (scan_should_stop())
+ break;
+
+ /*
+ * When scanning a memory block with a corresponding
+ * kmemleak_object, the CPU yielding is handled in the calling
+ * code since it holds the object->lock to avoid the block
+ * freeing.
+ */
+ if (!scanned)
+ scan_yield();
+
+ object = find_and_get_object(pointer, 1);
+ if (!object)
+ continue;
+ if (object == scanned) {
+ /* self referenced, ignore */
+ put_object(object);
+ continue;
+ }
+
+ /*
+ * Avoid the lockdep recursive warning on object->lock being
+ * previously acquired in scan_object(). These locks are
+ * enclosed by scan_mutex.
+ */
+ spin_lock_irqsave_nested(&object->lock, flags,
+ SINGLE_DEPTH_NESTING);
+ if (!color_white(object)) {
+ /* non-orphan, ignored or new */
+ spin_unlock_irqrestore(&object->lock, flags);
+ put_object(object);
+ continue;
+ }
+
+ /*
+ * Increase the object's reference count (number of pointers
+ * to the memory block). If this count reaches the required
+ * minimum, the object's color will become gray and it will be
+ * added to the gray_list.
+ */
+ object->count++;
+ if (color_gray(object))
+ list_add_tail(&object->gray_list, &gray_list);
+ else
+ put_object(object);
+ spin_unlock_irqrestore(&object->lock, flags);
+ }
+}
+
+/*
+ * Scan a memory block corresponding to a kmemleak_object. A condition is
+ * that object->use_count >= 1.
+ */
+static void scan_object(struct kmemleak_object *object)
+{
+ struct kmemleak_scan_area *area;
+ struct hlist_node *elem;
+ unsigned long flags;
+
+ /*
+ * Once the object->lock is aquired, the corresponding memory block
+ * cannot be freed (the same lock is aquired in delete_object).
+ */
+ spin_lock_irqsave(&object->lock, flags);
+ if (object->flags & OBJECT_NO_SCAN)
+ goto out;
+ if (!(object->flags & OBJECT_ALLOCATED))
+ /* already freed object */
+ goto out;
+ if (hlist_empty(&object->area_list))
+ scan_block((void *)object->pointer,
+ (void *)(object->pointer + object->size), object);
+ else
+ hlist_for_each_entry(area, elem, &object->area_list, node)
+ scan_block((void *)(object->pointer + area->offset),
+ (void *)(object->pointer + area->offset
+ + area->length), object);
+out:
+ spin_unlock_irqrestore(&object->lock, flags);
+}
+
+/*
+ * Scan data sections and all the referenced memory blocks allocated via the
+ * kernel's standard allocators. This function must be called with the
+ * scan_mutex held.
+ */
+static void kmemleak_scan(void)
+{
+ unsigned long flags;
+ struct kmemleak_object *object, *tmp;
+ struct task_struct *task;
+ int i;
+
+ /* prepare the kmemleak_object's */
+ rcu_read_lock();
+ list_for_each_entry_rcu(object, &object_list, object_list) {
+ spin_lock_irqsave(&object->lock, flags);
+#ifdef DEBUG
+ /*
+ * With a few exceptions there should be a maximum of
+ * 1 reference to any object at this point.
+ */
+ if (atomic_read(&object->use_count) > 1) {
+ pr_debug("kmemleak: object->use_count = %d\n",
+ atomic_read(&object->use_count));
+ dump_object_info(object);
+ }
+#endif
+ /* reset the reference count (whiten the object) */
+ object->count = 0;
+ if (color_gray(object) && get_object(object))
+ list_add_tail(&object->gray_list, &gray_list);
+
+ spin_unlock_irqrestore(&object->lock, flags);
+ }
+ rcu_read_unlock();
+
+ /* data/bss scanning */
+ scan_block(_sdata, _edata, NULL);
+ scan_block(__bss_start, __bss_stop, NULL);
+
+#ifdef CONFIG_SMP
+ /* per-cpu sections scanning */
+ for_each_possible_cpu(i)
+ scan_block(__per_cpu_start + per_cpu_offset(i),
+ __per_cpu_end + per_cpu_offset(i), NULL);
+#endif
+
+ /*
+ * Struct page scanning for each node. The code below is not yet safe
+ * with MEMORY_HOTPLUG.
+ */
+ for_each_online_node(i) {
+ pg_data_t *pgdat = NODE_DATA(i);
+ unsigned long start_pfn = pgdat->node_start_pfn;
+ unsigned long end_pfn = start_pfn + pgdat->node_spanned_pages;
+ unsigned long pfn;
+
+ for (pfn = start_pfn; pfn < end_pfn; pfn++) {
+ struct page *page;
+
+ if (!pfn_valid(pfn))
+ continue;
+ page = pfn_to_page(pfn);
+ /* only scan if page is in use */
+ if (page_count(page) == 0)
+ continue;
+ scan_block(page, page + 1, NULL);
+ }
+ }
+
+ /*
+ * Scanning the task stacks may introduce false negatives and it is
+ * not enabled by default.
+ */
+ if (kmemleak_stack_scan) {
+ read_lock(&tasklist_lock);
+ for_each_process(task)
+ scan_block(task_stack_page(task),
+ task_stack_page(task) + THREAD_SIZE, NULL);
+ read_unlock(&tasklist_lock);
+ }
+
+ /*
+ * Scan the objects already referenced from the sections scanned
+ * above. More objects will be referenced and, if there are no memory
+ * leaks, all the objects will be scanned. The list traversal is safe
+ * for both tail additions and removals from inside the loop. The
+ * kmemleak objects cannot be freed from outside the loop because their
+ * use_count was increased.
+ */
+ object = list_entry(gray_list.next, typeof(*object), gray_list);
+ while (&object->gray_list != &gray_list) {
+ scan_yield();
+
+ /* may add new objects to the list */
+ if (!scan_should_stop())
+ scan_object(object);
+
+ tmp = list_entry(object->gray_list.next, typeof(*object),
+ gray_list);
+
+ /* remove the object from the list and release it */
+ list_del(&object->gray_list);
+ put_object(object);
+
+ object = tmp;
+ }
+ WARN_ON(!list_empty(&gray_list));
+}
+
+/*
+ * Thread function performing automatic memory scanning. Unreferenced objects
+ * at the end of a memory scan are reported but only the first time.
+ */
+static int kmemleak_scan_thread(void *arg)
+{
+ static int first_run = 1;
+
+ pr_info("kmemleak: Automatic memory scanning thread started\n");
+
+ /*
+ * Wait before the first scan to allow the system to fully initialize.
+ */
+ if (first_run) {
+ first_run = 0;
+ ssleep(SECS_FIRST_SCAN);
+ }
+
+ while (!kthread_should_stop()) {
+ struct kmemleak_object *object;
+ signed long timeout = jiffies_scan_wait;
+
+ mutex_lock(&scan_mutex);
+
+ kmemleak_scan();
+ reported_leaks = 0;
+
+ rcu_read_lock();
+ list_for_each_entry_rcu(object, &object_list, object_list) {
+ unsigned long flags;
+
+ if (reported_leaks >= REPORTS_NR)
+ break;
+ spin_lock_irqsave(&object->lock, flags);
+ if (!(object->flags & OBJECT_REPORTED) &&
+ unreferenced_object(object)) {
+ print_unreferenced(NULL, object);
+ object->flags |= OBJECT_REPORTED;
+ reported_leaks++;
+ } else if ((object->flags & OBJECT_REPORTED) &&
+ referenced_object(object)) {
+ print_referenced(object);
+ object->flags &= ~OBJECT_REPORTED;
+ }
+ spin_unlock_irqrestore(&object->lock, flags);
+ }
+ rcu_read_unlock();
+
+ mutex_unlock(&scan_mutex);
+ /* wait before the next scan */
+ while (timeout && !kthread_should_stop())
+ timeout = schedule_timeout_interruptible(timeout);
+ }
+
+ pr_info("kmemleak: Automatic memory scanning thread ended\n");
+
+ return 0;
+}
+
+/*
+ * Start the automatic memory scanning thread. This function must be called
+ * with the kmemleak_mutex held.
+ */
+void start_scan_thread(void)
+{
+ if (scan_thread)
+ return;
+ scan_thread = kthread_run(kmemleak_scan_thread, NULL, "kmemleak");
+ if (IS_ERR(scan_thread)) {
+ pr_warning("kmemleak: Failed to create the scan thread\n");
+ scan_thread = NULL;
+ }
+}
+
+/*
+ * Stop the automatic memory scanning thread. This function must be called
+ * with the kmemleak_mutex held.
+ */
+void stop_scan_thread(void)
+{
+ if (scan_thread) {
+ kthread_stop(scan_thread);
+ scan_thread = NULL;
+ }
+}
+
+/*
+ * Iterate over the object_list and return the first valid object at or after
+ * the required position with its use_count incremented. The function triggers
+ * a memory scanning when the pos argument points to the first position.
+ */
+static void *kmemleak_seq_start(struct seq_file *seq, loff_t *pos)
+{
+ struct kmemleak_object *object;
+ loff_t n = *pos;
+
+ if (!n) {
+ kmemleak_scan();
+ reported_leaks = 0;
+ }
+ if (reported_leaks >= REPORTS_NR)
+ return NULL;
+
+ rcu_read_lock();
+ list_for_each_entry_rcu(object, &object_list, object_list) {
+ if (n-- > 0)
+ continue;
+ if (get_object(object))
+ goto out;
+ }
+ object = NULL;
+out:
+ rcu_read_unlock();
+ return object;
+}
+
+/*
+ * Return the next object in the object_list. The function decrements the
+ * use_count of the previous object and increases that of the next one.
+ */
+static void *kmemleak_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+ struct kmemleak_object *prev_obj = v;
+ struct kmemleak_object *next_obj = NULL;
+ struct list_head *n = &prev_obj->object_list;
+
+ ++(*pos);
+ if (reported_leaks >= REPORTS_NR)
+ goto out;
+
+ rcu_read_lock();
+ list_for_each_continue_rcu(n, &object_list) {
+ next_obj = list_entry(n, struct kmemleak_object, object_list);
+ if (get_object(next_obj))
+ break;
+ }
+ rcu_read_unlock();
+out:
+ put_object(prev_obj);
+ return next_obj;
+}
+
+/*
+ * Decrement the use_count of the last object required, if any.
+ */
+static void kmemleak_seq_stop(struct seq_file *seq, void *v)
+{
+ if (v)
+ put_object(v);
+}
+
+/*
+ * Print the information for an unreferenced object to the seq file.
+ */
+static int kmemleak_seq_show(struct seq_file *seq, void *v)
+{
+ struct kmemleak_object *object = v;
+ unsigned long flags;
+
+ spin_lock_irqsave(&object->lock, flags);
+ if (!unreferenced_object(object))
+ goto out;
+ print_unreferenced(seq, object);
+ reported_leaks++;
+out:
+ spin_unlock_irqrestore(&object->lock, flags);
+ return 0;
+}
+
+static const struct seq_operations kmemleak_seq_ops = {
+ .start = kmemleak_seq_start,
+ .next = kmemleak_seq_next,
+ .stop = kmemleak_seq_stop,
+ .show = kmemleak_seq_show,
+};
+
+static int kmemleak_open(struct inode *inode, struct file *file)
+{
+ int ret = 0;
+
+ if (!atomic_read(&kmemleak_enabled))
+ return -EBUSY;
+
+ ret = mutex_lock_interruptible(&kmemleak_mutex);
+ if (ret < 0)
+ goto out;
+ if (file->f_mode & FMODE_READ) {
+ ret = mutex_lock_interruptible(&scan_mutex);
+ if (ret < 0)
+ goto kmemleak_unlock;
+ ret = seq_open(file, &kmemleak_seq_ops);
+ if (ret < 0)
+ goto scan_unlock;
+ }
+ return ret;
+
+scan_unlock:
+ mutex_unlock(&scan_mutex);
+kmemleak_unlock:
+ mutex_unlock(&kmemleak_mutex);
+out:
+ return ret;
+}
+
+static int kmemleak_release(struct inode *inode, struct file *file)
+{
+ int ret = 0;
+
+ if (file->f_mode & FMODE_READ) {
+ seq_release(inode, file);
+ mutex_unlock(&scan_mutex);
+ }
+ mutex_unlock(&kmemleak_mutex);
+
+ return ret;
+}
+
+/*
+ * File write operation to configure kmemleak at run-time. The following
+ * commands can be written to the /sys/kernel/debug/kmemleak file:
+ * off - disable kmemleak (irreversible)
+ * stack=on - enable the task stacks scanning
+ * stack=off - disable the tasks stacks scanning
+ * scan=on - start the automatic memory scanning thread
+ * scan=off - stop the automatic memory scanning thread
+ * scan=... - set the automatic memory scanning period in seconds (0 to
+ * disable it)
+ */
+static ssize_t kmemleak_write(struct file *file, const char __user *user_buf,
+ size_t size, loff_t *ppos)
+{
+ char buf[64];
+ int buf_size;
+
+ 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;
+ buf[buf_size] = 0;
+
+ if (strncmp(buf, "off", 3) == 0)
+ kmemleak_disable();
+ else if (strncmp(buf, "stack=on", 8) == 0)
+ kmemleak_stack_scan = 1;
+ else if (strncmp(buf, "stack=off", 9) == 0)
+ kmemleak_stack_scan = 0;
+ else if (strncmp(buf, "scan=on", 7) == 0)
+ start_scan_thread();
+ else if (strncmp(buf, "scan=off", 8) == 0)
+ stop_scan_thread();
+ else if (strncmp(buf, "scan=", 5) == 0) {
+ unsigned long secs;
+ int err;
+
+ err = strict_strtoul(buf + 5, 0, &secs);
+ if (err < 0)
+ return err;
+ stop_scan_thread();
+ if (secs) {
+ jiffies_scan_wait = msecs_to_jiffies(secs * 1000);
+ start_scan_thread();
+ }
+ } else
+ return -EINVAL;
+
+ /* ignore the rest of the buffer, only one command at a time */
+ *ppos += size;
+ return size;
+}
+
+static const struct file_operations kmemleak_fops = {
+ .owner = THIS_MODULE,
+ .open = kmemleak_open,
+ .read = seq_read,
+ .write = kmemleak_write,
+ .llseek = seq_lseek,
+ .release = kmemleak_release,
+};
+
+/*
+ * Perform the freeing of the kmemleak internal objects after waiting for any
+ * current memory scan to complete.
+ */
+static int kmemleak_cleanup_thread(void *arg)
+{
+ struct kmemleak_object *object;
+
+ mutex_lock(&kmemleak_mutex);
+ stop_scan_thread();
+ mutex_unlock(&kmemleak_mutex);
+
+ mutex_lock(&scan_mutex);
+ rcu_read_lock();
+ list_for_each_entry_rcu(object, &object_list, object_list)
+ delete_object(object->pointer);
+ rcu_read_unlock();
+ mutex_unlock(&scan_mutex);
+
+ return 0;
+}
+
+/*
+ * Start the clean-up thread.
+ */
+static void kmemleak_cleanup(void)
+{
+ struct task_struct *cleanup_thread;
+
+ cleanup_thread = kthread_run(kmemleak_cleanup_thread, NULL,
+ "kmemleak-clean");
+ if (IS_ERR(cleanup_thread))
+ pr_warning("kmemleak: Failed to create the clean-up thread\n");
+}
+
+/*
+ * Disable kmemleak. No memory allocation/freeing will be traced once this
+ * function is called. Disabling kmemleak is an irreversible operation.
+ */
+static void kmemleak_disable(void)
+{
+ /* atomically check whether it was already invoked */
+ if (atomic_cmpxchg(&kmemleak_error, 0, 1))
+ 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 */
+ if (atomic_read(&kmemleak_initialized))
+ kmemleak_cleanup();
+
+ pr_info("Kernel memory leak detector disabled\n");
+}
+
+/*
+ * Allow boot-time kmemleak disabling (enabled by default).
+ */
+static int kmemleak_boot_config(char *str)
+{
+ if (!str)
+ return -EINVAL;
+ if (strcmp(str, "off") == 0)
+ kmemleak_disable();
+ else if (strcmp(str, "on") != 0)
+ return -EINVAL;
+ return 0;
+}
+early_param("kmemleak", kmemleak_boot_config);
+
+/*
+ * Kkmemleak initialization.
+ */
+void __init kmemleak_init(void)
+{
+ int i;
+ unsigned long flags;
+
+ jiffies_scan_yield = msecs_to_jiffies(MSECS_SCAN_YIELD);
+ jiffies_min_age = msecs_to_jiffies(MSECS_MIN_AGE);
+ jiffies_scan_wait = msecs_to_jiffies(SECS_SCAN_WAIT * 1000);
+
+ object_cache = KMEM_CACHE(kmemleak_object, SLAB_NOLEAKTRACE);
+ scan_area_cache = KMEM_CACHE(kmemleak_scan_area, SLAB_NOLEAKTRACE);
+ INIT_PRIO_TREE_ROOT(&object_tree_root);
+
+ /* 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);
+ }
+ local_irq_restore(flags);
+
+ /*
+ * This is the point where tracking allocations is safe. Automatic
+ * scanning is started during the late initcall. Add the early logged
+ * callbacks to the kmemleak infrastructure.
+ */
+ for (i = 0; i < crt_early_log; i++) {
+ struct early_log *log = &early_log[i];
+
+ switch (log->op_type) {
+ case KMEMLEAK_ALLOC:
+ kmemleak_alloc(log->ptr, log->size, log->min_count,
+ GFP_KERNEL);
+ break;
+ case KMEMLEAK_FREE:
+ kmemleak_free(log->ptr);
+ break;
+ case KMEMLEAK_NOT_LEAK:
+ kmemleak_not_leak(log->ptr);
+ break;
+ case KMEMLEAK_IGNORE:
+ kmemleak_ignore(log->ptr);
+ break;
+ case KMEMLEAK_SCAN_AREA:
+ kmemleak_scan_area(log->ptr, log->offset, log->length,
+ GFP_KERNEL);
+ break;
+ case KMEMLEAK_NO_SCAN:
+ kmemleak_no_scan(log->ptr);
+ break;
+ default:
+ WARN_ON(1);
+ }
+ }
+}
+
+/*
+ * Late initialization function.
+ */
+static int __init kmemleak_late_init(void)
+{
+ struct dentry *dentry;
+
+ atomic_set(&kmemleak_initialized, 1);
+
+ if (atomic_read(&kmemleak_error)) {
+ /*
+ * Some error occured and kmemleak was disabled. There is a
+ * small chance that kmemleak_disable() was called immediately
+ * after setting kmemleak_initialized and we may end up with
+ * two clean-up threads but serialized by scan_mutex.
+ */
+ kmemleak_cleanup();
+ return -ENOMEM;
+ }
+
+ dentry = debugfs_create_file("kmemleak", S_IRUGO, NULL, NULL,
+ &kmemleak_fops);
+ if (!dentry)
+ pr_warning("kmemleak: Failed to create the debugfs kmemleak "
+ "file\n");
+ mutex_lock(&kmemleak_mutex);
+ start_scan_thread();
+ mutex_unlock(&kmemleak_mutex);
+
+ pr_info("Kernel memory leak detector initialized\n");
+
+ return 0;
+}
+late_initcall(kmemleak_late_init);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 474c7e9dd51a..17d5f539a9aa 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -46,6 +46,7 @@
#include <linux/page-isolation.h>
#include <linux/page_cgroup.h>
#include <linux/debugobjects.h>
+#include <linux/kmemleak.h>
#include <asm/tlbflush.h>
#include <asm/div64.h>
@@ -4546,6 +4547,16 @@ void *__init alloc_large_system_hash(const char *tablename,
if (_hash_mask)
*_hash_mask = (1 << log2qty) - 1;
+ /*
+ * If hashdist is set, the table allocation is done with __vmalloc()
+ * which invokes the kmemleak_alloc() callback. This function may also
+ * be called before the slab and kmemleak are initialised when
+ * kmemleak simply buffers the request to be executed later
+ * (GFP_ATOMIC flag ignored in this case).
+ */
+ if (!hashdist)
+ kmemleak_alloc(table, size, 1, GFP_ATOMIC);
+
return table;
}
diff --git a/mm/slab.c b/mm/slab.c
index 2bd611fa87bf..f46b65d124e5 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -107,6 +107,7 @@
#include <linux/string.h>
#include <linux/uaccess.h>
#include <linux/nodemask.h>
+#include <linux/kmemleak.h>
#include <linux/mempolicy.h>
#include <linux/mutex.h>
#include <linux/fault-inject.h>
@@ -178,13 +179,13 @@
SLAB_STORE_USER | \
SLAB_RECLAIM_ACCOUNT | SLAB_PANIC | \
SLAB_DESTROY_BY_RCU | SLAB_MEM_SPREAD | \
- SLAB_DEBUG_OBJECTS)
+ SLAB_DEBUG_OBJECTS | SLAB_NOLEAKTRACE)
#else
# define CREATE_MASK (SLAB_HWCACHE_ALIGN | \
SLAB_CACHE_DMA | \
SLAB_RECLAIM_ACCOUNT | SLAB_PANIC | \
SLAB_DESTROY_BY_RCU | SLAB_MEM_SPREAD | \
- SLAB_DEBUG_OBJECTS)
+ SLAB_DEBUG_OBJECTS | SLAB_NOLEAKTRACE)
#endif
/*
@@ -964,6 +965,14 @@ static struct array_cache *alloc_arraycache(int node, int entries,
struct array_cache *nc = NULL;
nc = kmalloc_node(memsize, gfp, node);
+ /*
+ * The array_cache structures contain pointers to free object.
+ * However, when such objects are allocated or transfered to another
+ * cache the pointers are not cleared and they could be counted as
+ * valid references during a kmemleak scan. Therefore, kmemleak must
+ * not scan such objects.
+ */
+ kmemleak_no_scan(nc);
if (nc) {
nc->avail = 0;
nc->limit = entries;
@@ -2625,6 +2634,14 @@ static struct slab *alloc_slabmgmt(struct kmem_cache *cachep, void *objp,
/* Slab management obj is off-slab. */
slabp = kmem_cache_alloc_node(cachep->slabp_cache,
local_flags, nodeid);
+ /*
+ * If the first object in the slab is leaked (it's allocated
+ * but no one has a reference to it), we want to make sure
+ * kmemleak does not treat the ->s_mem pointer as a reference
+ * to the object. Otherwise we will not report the leak.
+ */
+ kmemleak_scan_area(slabp, offsetof(struct slab, list),
+ sizeof(struct list_head), local_flags);
if (!slabp)
return NULL;
} else {
@@ -3145,6 +3162,12 @@ static inline void *____cache_alloc(struct kmem_cache *cachep, gfp_t flags)
STATS_INC_ALLOCMISS(cachep);
objp = cache_alloc_refill(cachep, flags);
}
+ /*
+ * To avoid a false negative, if an object that is in one of the
+ * per-CPU caches is leaked, we need to make sure kmemleak doesn't
+ * treat the array pointers as a reference to the object.
+ */
+ kmemleak_erase(&ac->entry[ac->avail]);
return objp;
}
@@ -3364,6 +3387,8 @@ __cache_alloc_node(struct kmem_cache *cachep, gfp_t flags, int nodeid,
out:
local_irq_restore(save_flags);
ptr = cache_alloc_debugcheck_after(cachep, flags, ptr, caller);
+ kmemleak_alloc_recursive(ptr, obj_size(cachep), 1, cachep->flags,
+ flags);
if (unlikely((flags & __GFP_ZERO) && ptr))
memset(ptr, 0, obj_size(cachep));
@@ -3419,6 +3444,8 @@ __cache_alloc(struct kmem_cache *cachep, gfp_t flags, void *caller)
objp = __do_cache_alloc(cachep, flags);
local_irq_restore(save_flags);
objp = cache_alloc_debugcheck_after(cachep, flags, objp, caller);
+ kmemleak_alloc_recursive(objp, obj_size(cachep), 1, cachep->flags,
+ flags);
prefetchw(objp);
if (unlikely((flags & __GFP_ZERO) && objp))
@@ -3534,6 +3561,7 @@ static inline void __cache_free(struct kmem_cache *cachep, void *objp)
struct array_cache *ac = cpu_cache_get(cachep);
check_irq_off();
+ kmemleak_free_recursive(objp, cachep->flags);
objp = cache_free_debugcheck(cachep, objp, __builtin_return_address(0));
/*
diff --git a/mm/slob.c b/mm/slob.c
index 9b1737b0787b..12f261499925 100644
--- a/mm/slob.c
+++ b/mm/slob.c
@@ -67,6 +67,7 @@
#include <linux/rcupdate.h>
#include <linux/list.h>
#include <linux/kmemtrace.h>
+#include <linux/kmemleak.h>
#include <asm/atomic.h>
/*
@@ -509,6 +510,7 @@ void *__kmalloc_node(size_t size, gfp_t gfp, int node)
size, PAGE_SIZE << order, gfp, node);
}
+ kmemleak_alloc(ret, size, 1, gfp);
return ret;
}
EXPORT_SYMBOL(__kmalloc_node);
@@ -521,6 +523,7 @@ void kfree(const void *block)
if (unlikely(ZERO_OR_NULL_PTR(block)))
return;
+ kmemleak_free(block);
sp = slob_page(block);
if (is_slob_page(sp)) {
@@ -584,12 +587,14 @@ struct kmem_cache *kmem_cache_create(const char *name, size_t size,
} else if (flags & SLAB_PANIC)
panic("Cannot create slab cache %s\n", name);
+ kmemleak_alloc(c, sizeof(struct kmem_cache), 1, GFP_KERNEL);
return c;
}
EXPORT_SYMBOL(kmem_cache_create);
void kmem_cache_destroy(struct kmem_cache *c)
{
+ kmemleak_free(c);
slob_free(c, sizeof(struct kmem_cache));
}
EXPORT_SYMBOL(kmem_cache_destroy);
@@ -613,6 +618,7 @@ void *kmem_cache_alloc_node(struct kmem_cache *c, gfp_t flags, int node)
if (c->ctor)
c->ctor(b);
+ kmemleak_alloc_recursive(b, c->size, 1, c->flags, flags);
return b;
}
EXPORT_SYMBOL(kmem_cache_alloc_node);
@@ -635,6 +641,7 @@ static void kmem_rcu_free(struct rcu_head *head)
void kmem_cache_free(struct kmem_cache *c, void *b)
{
+ kmemleak_free_recursive(b, c->flags);
if (unlikely(c->flags & SLAB_DESTROY_BY_RCU)) {
struct slob_rcu *slob_rcu;
slob_rcu = b + (c->size - sizeof(struct slob_rcu));
diff --git a/mm/slub.c b/mm/slub.c
index c1815a63807a..3964d3ce4c15 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -20,6 +20,7 @@
#include <linux/kmemtrace.h>
#include <linux/cpu.h>
#include <linux/cpuset.h>
+#include <linux/kmemleak.h>
#include <linux/mempolicy.h>
#include <linux/ctype.h>
#include <linux/debugobjects.h>
@@ -143,7 +144,7 @@
* Set of flags that will prevent slab merging
*/
#define SLUB_NEVER_MERGE (SLAB_RED_ZONE | SLAB_POISON | SLAB_STORE_USER | \
- SLAB_TRACE | SLAB_DESTROY_BY_RCU)
+ SLAB_TRACE | SLAB_DESTROY_BY_RCU | SLAB_NOLEAKTRACE)
#define SLUB_MERGE_SAME (SLAB_DEBUG_FREE | SLAB_RECLAIM_ACCOUNT | \
SLAB_CACHE_DMA)
@@ -1617,6 +1618,7 @@ static __always_inline void *slab_alloc(struct kmem_cache *s,
if (unlikely((gfpflags & __GFP_ZERO) && object))
memset(object, 0, objsize);
+ kmemleak_alloc_recursive(object, objsize, 1, s->flags, gfpflags);
return object;
}
@@ -1746,6 +1748,7 @@ static __always_inline void slab_free(struct kmem_cache *s,
struct kmem_cache_cpu *c;
unsigned long flags;
+ kmemleak_free_recursive(x, s->flags);
local_irq_save(flags);
c = get_cpu_slab(s, smp_processor_id());
debug_check_no_locks_freed(object, c->objsize);
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index 323513858c20..f8189a4b3e13 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -24,6 +24,7 @@
#include <linux/radix-tree.h>
#include <linux/rcupdate.h>
#include <linux/pfn.h>
+#include <linux/kmemleak.h>
#include <asm/atomic.h>
#include <asm/uaccess.h>
@@ -1326,6 +1327,9 @@ static void __vunmap(const void *addr, int deallocate_pages)
void vfree(const void *addr)
{
BUG_ON(in_interrupt());
+
+ kmemleak_free(addr);
+
__vunmap(addr, 1);
}
EXPORT_SYMBOL(vfree);
@@ -1438,8 +1442,17 @@ fail:
void *__vmalloc_area(struct vm_struct *area, gfp_t gfp_mask, pgprot_t prot)
{
- return __vmalloc_area_node(area, gfp_mask, prot, -1,
- __builtin_return_address(0));
+ void *addr = __vmalloc_area_node(area, gfp_mask, prot, -1,
+ __builtin_return_address(0));
+
+ /*
+ * A ref_count = 3 is needed because the vm_struct and vmap_area
+ * structures allocated in the __get_vm_area_node() function contain
+ * references to the virtual address of the vmalloc'ed block.
+ */
+ kmemleak_alloc(addr, area->size - PAGE_SIZE, 3, gfp_mask);
+
+ return addr;
}
/**
@@ -1458,6 +1471,8 @@ static void *__vmalloc_node(unsigned long size, gfp_t gfp_mask, pgprot_t prot,
int node, void *caller)
{
struct vm_struct *area;
+ void *addr;
+ unsigned long real_size = size;
size = PAGE_ALIGN(size);
if (!size || (size >> PAGE_SHIFT) > num_physpages)
@@ -1469,7 +1484,16 @@ static void *__vmalloc_node(unsigned long size, gfp_t gfp_mask, pgprot_t prot,
if (!area)
return NULL;
- return __vmalloc_area_node(area, gfp_mask, prot, node, caller);
+ addr = __vmalloc_area_node(area, gfp_mask, prot, node, caller);
+
+ /*
+ * A ref_count = 3 is needed because the vm_struct and vmap_area
+ * structures allocated in the __get_vm_area_node() function contain
+ * references to the virtual address of the vmalloc'ed block.
+ */
+ kmemleak_alloc(addr, real_size, 3, gfp_mask);
+
+ return addr;
}
void *__vmalloc(unsigned long size, gfp_t gfp_mask, pgprot_t prot)