kernel_optimize_test/drivers/base
Muchun Song ac43432cb1 driver core: Fix use-after-free and double free on glue directory
There is a race condition between removing glue directory and adding a new
device under the glue dir. It can be reproduced in following test:

CPU1:                                         CPU2:

device_add()
  get_device_parent()
    class_dir_create_and_add()
      kobject_add_internal()
        create_dir()    // create glue_dir

                                              device_add()
                                                get_device_parent()
                                                  kobject_get() // get glue_dir

device_del()
  cleanup_glue_dir()
    kobject_del(glue_dir)

                                                kobject_add()
                                                  kobject_add_internal()
                                                    create_dir() // in glue_dir
                                                      sysfs_create_dir_ns()
                                                        kernfs_create_dir_ns(sd)

      sysfs_remove_dir() // glue_dir->sd=NULL
      sysfs_put()        // free glue_dir->sd

                                                          // sd is freed
                                                          kernfs_new_node(sd)
                                                            kernfs_get(glue_dir)
                                                            kernfs_add_one()
                                                            kernfs_put()

Before CPU1 remove last child device under glue dir, if CPU2 add a new
device under glue dir, the glue_dir kobject reference count will be
increase to 2 via kobject_get() in get_device_parent(). And CPU2 has
been called kernfs_create_dir_ns(), but not call kernfs_new_node().
Meanwhile, CPU1 call sysfs_remove_dir() and sysfs_put(). This result in
glue_dir->sd is freed and it's reference count will be 0. Then CPU2 call
kernfs_get(glue_dir) will trigger a warning in kernfs_get() and increase
it's reference count to 1. Because glue_dir->sd is freed by CPU1, the next
call kernfs_add_one() by CPU2 will fail(This is also use-after-free)
and call kernfs_put() to decrease reference count. Because the reference
count is decremented to 0, it will also call kmem_cache_free() to free
the glue_dir->sd again. This will result in double free.

In order to avoid this happening, we also should make sure that kernfs_node
for glue_dir is released in CPU1 only when refcount for glue_dir kobj is
1 to fix this race.

The following calltrace is captured in kernel 4.14 with the following patch
applied:

commit 726e410979 ("drivers: core: Remove glue dirs from sysfs earlier")

--------------------------------------------------------------------------
[    3.633703] WARNING: CPU: 4 PID: 513 at .../fs/kernfs/dir.c:494
                Here is WARN_ON(!atomic_read(&kn->count) in kernfs_get().
....
[    3.633986] Call trace:
[    3.633991]  kernfs_create_dir_ns+0xa8/0xb0
[    3.633994]  sysfs_create_dir_ns+0x54/0xe8
[    3.634001]  kobject_add_internal+0x22c/0x3f0
[    3.634005]  kobject_add+0xe4/0x118
[    3.634011]  device_add+0x200/0x870
[    3.634017]  _request_firmware+0x958/0xc38
[    3.634020]  request_firmware_into_buf+0x4c/0x70
....
[    3.634064] kernel BUG at .../mm/slub.c:294!
                Here is BUG_ON(object == fp) in set_freepointer().
....
[    3.634346] Call trace:
[    3.634351]  kmem_cache_free+0x504/0x6b8
[    3.634355]  kernfs_put+0x14c/0x1d8
[    3.634359]  kernfs_create_dir_ns+0x88/0xb0
[    3.634362]  sysfs_create_dir_ns+0x54/0xe8
[    3.634366]  kobject_add_internal+0x22c/0x3f0
[    3.634370]  kobject_add+0xe4/0x118
[    3.634374]  device_add+0x200/0x870
[    3.634378]  _request_firmware+0x958/0xc38
[    3.634381]  request_firmware_into_buf+0x4c/0x70
--------------------------------------------------------------------------

Fixes: 726e410979 ("drivers: core: Remove glue dirs from sysfs earlier")
Signed-off-by: Muchun Song <smuchun@gmail.com>
Reviewed-by: Mukesh Ojha <mojha@codeaurora.org>
Signed-off-by: Prateek Sood <prsood@codeaurora.org>
Link: https://lore.kernel.org/r/20190727032122.24639-1-smuchun@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2019-07-30 18:44:47 +02:00
..
firmware_loader proc/sysctl: add shared variables for range check 2019-07-18 17:08:07 -07:00
power Merge branch 'pm-cpufreq' 2019-07-18 09:49:30 +02:00
regmap Merge branch 'regmap-5.3' into regmap-next 2019-07-04 17:33:59 +01:00
test drivers: base: test: add proper SPDX identifier to Makefile 2019-04-04 20:03:40 +02:00
arch_topology.c Driver Core and debugfs changes for 5.3-rc1 2019-07-12 12:24:03 -07:00
attribute_container.c driver core: Remove redundant license text 2017-12-07 18:36:44 +01:00
base.h driver core: Probe devices asynchronously instead of the driver 2019-01-31 14:20:53 +01:00
bus.c bus_find_device: Unify the match callback with class_find_device 2019-06-24 05:22:31 +02:00
cacheinfo.c Driver Core and debugfs changes for 5.3-rc1 2019-07-12 12:24:03 -07:00
class.c driver core: move device->knode_class to device_private 2019-01-18 16:55:48 +01:00
component.c drivers/component: kerneldoc polish 2019-02-19 13:20:35 +01:00
container.c driver core: Remove redundant license text 2017-12-07 18:36:44 +01:00
core.c driver core: Fix use-after-free and double free on glue directory 2019-07-30 18:44:47 +02:00
cpu.c Merge branch 'x86-mds-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2019-05-14 07:57:29 -07:00
dd.c driver: core: Allow subsystems to continue deferring probe 2019-07-03 21:28:20 +02:00
devcon.c Driver Core and debugfs changes for 5.3-rc1 2019-07-12 12:24:03 -07:00
devcoredump.c driver core: Remove redundant license text 2017-12-07 18:36:44 +01:00
devres.c drivers/base/devres: introduce devm_release_action() 2019-06-13 17:34:56 -10:00
devtmpfs.c constify ksys_mount() string arguments 2019-07-04 22:01:59 -04:00
driver.c driver_find_device: Unify the match function with class_find_device() 2019-06-24 05:22:31 +02:00
firmware.c driver core: Remove redundant license text 2017-12-07 18:36:44 +01:00
hypervisor.c driver core: Remove redundant license text 2017-12-07 18:36:44 +01:00
init.c base: fix order of OF initialization 2018-07-07 17:54:29 +02:00
isa.c Merge 4.15-rc3 into driver-core-next 2017-12-11 08:50:05 +01:00
Kconfig Driver core/kobject patches for 5.2-rc1 2019-05-07 13:01:40 -07:00
Makefile drivers: base: Introducing software nodes to the firmware node framework 2018-11-26 18:19:11 +01:00
map.c driver core: Remove redundant license text 2017-12-07 18:36:44 +01:00
memory.c drivers/base/memory.c: get rid of find_memory_block_hinted() 2019-07-18 17:08:07 -07:00
module.c driver core: Remove redundant license text 2017-12-07 18:36:44 +01:00
node.c mm/memory_hotplug: rename walk_memory_range() and pass start+size instead of pfns 2019-07-18 17:08:06 -07:00
pinctrl.c driver core: Remove redundant license text 2017-12-07 18:36:44 +01:00
platform-msi.c platform-msi: Free descriptors in platform_msi_domain_free() 2018-12-13 09:35:31 +00:00
platform.c driver core: platform: return -ENXIO for missing GpioInt 2019-07-30 13:46:15 +02:00
property.c device property: Introduce fwnode_find_reference() 2019-06-03 10:55:38 +02:00
soc.c base: soc: use put_device() instead of kfree() 2018-03-15 14:37:03 +01:00
swnode.c software node: Add software_node_get_reference_args() 2019-06-03 10:55:38 +02:00
syscore.c treewide: Switch printk users from %pf and %pF to %ps and %pS, respectively 2019-04-09 14:19:06 +02:00
topology.c topology: Create core_cpus and die_cpus sysfs attributes 2019-05-23 10:08:34 +02:00
transport_class.c driver core: Remove redundant license text 2017-12-07 18:36:44 +01:00