summaryrefslogtreecommitdiff
path: root/drivers/md
diff options
context:
space:
mode:
authorMikulas Patocka <mpatocka@redhat.com>2014-01-14 04:37:54 +0400
committerMike Snitzer <snitzer@redhat.com>2014-01-15 08:23:04 +0400
commit2995fa78e423d7193f3b57835f6c1c75006a0315 (patch)
tree9f107c0503a3c0ba58559a62847e9236f3fd1951 /drivers/md
parent55b082e614e219fb5199a6f93e648ed35d3c96d5 (diff)
downloadlinux-2995fa78e423d7193f3b57835f6c1c75006a0315.tar.xz
dm sysfs: fix a module unload race
This reverts commit be35f48610 ("dm: wait until embedded kobject is released before destroying a device") and provides an improved fix. The kobject release code that calls the completion must be placed in a non-module file, otherwise there is a module unload race (if the process calling dm_kobject_release is preempted and the DM module unloaded after the completion is triggered, but before dm_kobject_release returns). To fix this race, this patch moves the completion code to dm-builtin.c which is always compiled directly into the kernel if BLK_DEV_DM is selected. The patch introduces a new dm_kobject_holder structure, its purpose is to keep the completion and kobject in one place, so that it can be accessed from non-module code without the need to export the layout of struct mapped_device to that code. Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Signed-off-by: Mike Snitzer <snitzer@redhat.com> Cc: stable@vger.kernel.org
Diffstat (limited to 'drivers/md')
-rw-r--r--drivers/md/Kconfig4
-rw-r--r--drivers/md/Makefile1
-rw-r--r--drivers/md/dm-builtin.c48
-rw-r--r--drivers/md/dm-sysfs.c5
-rw-r--r--drivers/md/dm.c20
-rw-r--r--drivers/md/dm.h17
6 files changed, 74 insertions, 21 deletions
diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig
index 39b540a13369..9a06fe883766 100644
--- a/drivers/md/Kconfig
+++ b/drivers/md/Kconfig
@@ -176,8 +176,12 @@ config MD_FAULTY
source "drivers/md/bcache/Kconfig"
+config BLK_DEV_DM_BUILTIN
+ boolean
+
config BLK_DEV_DM
tristate "Device mapper support"
+ select BLK_DEV_DM_BUILTIN
---help---
Device-mapper is a low level volume manager. It works by allowing
people to specify mappings for ranges of logical sectors. Various
diff --git a/drivers/md/Makefile b/drivers/md/Makefile
index 2acc43fe0229..f26d83292579 100644
--- a/drivers/md/Makefile
+++ b/drivers/md/Makefile
@@ -32,6 +32,7 @@ obj-$(CONFIG_MD_FAULTY) += faulty.o
obj-$(CONFIG_BCACHE) += bcache/
obj-$(CONFIG_BLK_DEV_MD) += md-mod.o
obj-$(CONFIG_BLK_DEV_DM) += dm-mod.o
+obj-$(CONFIG_BLK_DEV_DM_BUILTIN) += dm-builtin.o
obj-$(CONFIG_DM_BUFIO) += dm-bufio.o
obj-$(CONFIG_DM_BIO_PRISON) += dm-bio-prison.o
obj-$(CONFIG_DM_CRYPT) += dm-crypt.o
diff --git a/drivers/md/dm-builtin.c b/drivers/md/dm-builtin.c
new file mode 100644
index 000000000000..6c9049c51b2b
--- /dev/null
+++ b/drivers/md/dm-builtin.c
@@ -0,0 +1,48 @@
+#include "dm.h"
+
+/*
+ * The kobject release method must not be placed in the module itself,
+ * otherwise we are subject to module unload races.
+ *
+ * The release method is called when the last reference to the kobject is
+ * dropped. It may be called by any other kernel code that drops the last
+ * reference.
+ *
+ * The release method suffers from module unload race. We may prevent the
+ * module from being unloaded at the start of the release method (using
+ * increased module reference count or synchronizing against the release
+ * method), however there is no way to prevent the module from being
+ * unloaded at the end of the release method.
+ *
+ * If this code were placed in the dm module, the following race may
+ * happen:
+ * 1. Some other process takes a reference to dm kobject
+ * 2. The user issues ioctl function to unload the dm device
+ * 3. dm_sysfs_exit calls kobject_put, however the object is not released
+ * because of the other reference taken at step 1
+ * 4. dm_sysfs_exit waits on the completion
+ * 5. The other process that took the reference in step 1 drops it,
+ * dm_kobject_release is called from this process
+ * 6. dm_kobject_release calls complete()
+ * 7. a reschedule happens before dm_kobject_release returns
+ * 8. dm_sysfs_exit continues, the dm device is unloaded, module reference
+ * count is decremented
+ * 9. The user unloads the dm module
+ * 10. The other process that was rescheduled in step 7 continues to run,
+ * it is now executing code in unloaded module, so it crashes
+ *
+ * Note that if the process that takes the foreign reference to dm kobject
+ * has a low priority and the system is sufficiently loaded with
+ * higher-priority processes that prevent the low-priority process from
+ * being scheduled long enough, this bug may really happen.
+ *
+ * In order to fix this module unload race, we place the release method
+ * into a helper code that is compiled directly into the kernel.
+ */
+
+void dm_kobject_release(struct kobject *kobj)
+{
+ complete(dm_get_completion_from_kobject(kobj));
+}
+
+EXPORT_SYMBOL(dm_kobject_release);
diff --git a/drivers/md/dm-sysfs.c b/drivers/md/dm-sysfs.c
index e0cc5d6a9e46..c62c5ab6aed5 100644
--- a/drivers/md/dm-sysfs.c
+++ b/drivers/md/dm-sysfs.c
@@ -79,11 +79,6 @@ static const struct sysfs_ops dm_sysfs_ops = {
.show = dm_attr_show,
};
-static void dm_kobject_release(struct kobject *kobj)
-{
- complete(dm_get_completion_from_kobject(kobj));
-}
-
/*
* dm kobject is embedded in mapped_device structure
* no need to define release function here
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index e290e72922a4..b49c76284241 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -200,11 +200,8 @@ struct mapped_device {
/* forced geometry settings */
struct hd_geometry geometry;
- /* sysfs handle */
- struct kobject kobj;
-
- /* wait until the kobject is released */
- struct completion kobj_completion;
+ /* kobject and completion */
+ struct dm_kobject_holder kobj_holder;
/* zero-length flush that will be cloned and submitted to targets */
struct bio flush_bio;
@@ -2044,7 +2041,7 @@ static struct mapped_device *alloc_dev(int minor)
init_waitqueue_head(&md->wait);
INIT_WORK(&md->work, dm_wq_work);
init_waitqueue_head(&md->eventq);
- init_completion(&md->kobj_completion);
+ init_completion(&md->kobj_holder.completion);
md->disk->major = _major;
md->disk->first_minor = minor;
@@ -2906,14 +2903,14 @@ struct gendisk *dm_disk(struct mapped_device *md)
struct kobject *dm_kobject(struct mapped_device *md)
{
- return &md->kobj;
+ return &md->kobj_holder.kobj;
}
struct mapped_device *dm_get_from_kobject(struct kobject *kobj)
{
struct mapped_device *md;
- md = container_of(kobj, struct mapped_device, kobj);
+ md = container_of(kobj, struct mapped_device, kobj_holder.kobj);
if (test_bit(DMF_FREEING, &md->flags) ||
dm_deleting_md(md))
@@ -2923,13 +2920,6 @@ struct mapped_device *dm_get_from_kobject(struct kobject *kobj)
return md;
}
-struct completion *dm_get_completion_from_kobject(struct kobject *kobj)
-{
- struct mapped_device *md = container_of(kobj, struct mapped_device, kobj);
-
- return &md->kobj_completion;
-}
-
int dm_suspended_md(struct mapped_device *md)
{
return test_bit(DMF_SUSPENDED, &md->flags);
diff --git a/drivers/md/dm.h b/drivers/md/dm.h
index 1ab2028559ca..c4569f02f50f 100644
--- a/drivers/md/dm.h
+++ b/drivers/md/dm.h
@@ -16,6 +16,7 @@
#include <linux/blkdev.h>
#include <linux/hdreg.h>
#include <linux/completion.h>
+#include <linux/kobject.h>
#include "dm-stats.h"
@@ -149,11 +150,25 @@ void dm_interface_exit(void);
/*
* sysfs interface
*/
+struct dm_kobject_holder {
+ struct kobject kobj;
+ struct completion completion;
+};
+
+static inline struct completion *dm_get_completion_from_kobject(struct kobject *kobj)
+{
+ return &container_of(kobj, struct dm_kobject_holder, kobj)->completion;
+}
+
int dm_sysfs_init(struct mapped_device *md);
void dm_sysfs_exit(struct mapped_device *md);
struct kobject *dm_kobject(struct mapped_device *md);
struct mapped_device *dm_get_from_kobject(struct kobject *kobj);
-struct completion *dm_get_completion_from_kobject(struct kobject *kobj);
+
+/*
+ * The kobject helper
+ */
+void dm_kobject_release(struct kobject *kobj);
/*
* Targets for linear and striped mappings