linux/drivers/md/dm-uevent.c

207 lines
4.7 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-or-later
/*
* Device Mapper Uevent Support (dm-uevent)
*
* Copyright IBM Corporation, 2007
* Author: Mike Anderson <andmike@linux.vnet.ibm.com>
*/
#include <linux/list.h>
#include <linux/slab.h>
#include <linux/kobject.h>
#include <linux/dm-ioctl.h>
#include <linux/export.h>
#include "dm.h"
#include "dm-uevent.h"
#define DM_MSG_PREFIX "uevent"
static const struct {
enum dm_uevent_type type;
enum kobject_action action;
char *name;
} _dm_uevent_type_names[] = {
{DM_UEVENT_PATH_FAILED, KOBJ_CHANGE, "PATH_FAILED"},
{DM_UEVENT_PATH_REINSTATED, KOBJ_CHANGE, "PATH_REINSTATED"},
};
static struct kmem_cache *_dm_event_cache;
struct dm_uevent {
struct mapped_device *md;
enum kobject_action action;
struct kobj_uevent_env ku_env;
struct list_head elist;
char name[DM_NAME_LEN];
char uuid[DM_UUID_LEN];
};
static void dm_uevent_free(struct dm_uevent *event)
{
kmem_cache_free(_dm_event_cache, event);
}
static struct dm_uevent *dm_uevent_alloc(struct mapped_device *md)
{
struct dm_uevent *event;
event = kmem_cache_zalloc(_dm_event_cache, GFP_ATOMIC);
if (!event)
return NULL;
INIT_LIST_HEAD(&event->elist);
event->md = md;
return event;
}
static struct dm_uevent *dm_build_path_uevent(struct mapped_device *md,
struct dm_target *ti,
enum kobject_action action,
const char *dm_action,
const char *path,
unsigned nr_valid_paths)
{
struct dm_uevent *event;
event = dm_uevent_alloc(md);
if (!event) {
DMERR("%s: dm_uevent_alloc() failed", __func__);
goto err_nomem;
}
event->action = action;
if (add_uevent_var(&event->ku_env, "DM_TARGET=%s", ti->type->name)) {
DMERR("%s: add_uevent_var() for DM_TARGET failed",
__func__);
goto err_add;
}
if (add_uevent_var(&event->ku_env, "DM_ACTION=%s", dm_action)) {
DMERR("%s: add_uevent_var() for DM_ACTION failed",
__func__);
goto err_add;
}
if (add_uevent_var(&event->ku_env, "DM_SEQNUM=%u",
dm_next_uevent_seq(md))) {
DMERR("%s: add_uevent_var() for DM_SEQNUM failed",
__func__);
goto err_add;
}
if (add_uevent_var(&event->ku_env, "DM_PATH=%s", path)) {
DMERR("%s: add_uevent_var() for DM_PATH failed", __func__);
goto err_add;
}
if (add_uevent_var(&event->ku_env, "DM_NR_VALID_PATHS=%d",
nr_valid_paths)) {
DMERR("%s: add_uevent_var() for DM_NR_VALID_PATHS failed",
__func__);
goto err_add;
}
return event;
err_add:
dm_uevent_free(event);
err_nomem:
return ERR_PTR(-ENOMEM);
}
/**
* dm_send_uevents - send uevents for given list
*
* @events: list of events to send
* @kobj: kobject generating event
*
*/
void dm_send_uevents(struct list_head *events, struct kobject *kobj)
{
int r;
struct dm_uevent *event, *next;
list_for_each_entry_safe(event, next, events, elist) {
list_del_init(&event->elist);
/*
dm: avoid _hash_lock deadlock Fix a reported deadlock if there are still unprocessed multipath events on a device that is being removed. _hash_lock is held during dev_remove while trying to send the outstanding events. Sending the events requests the _hash_lock again in dm_copy_name_and_uuid. This patch introduces a separate lock around regions that modify the link to the hash table (dm_set_mdptr) or the name or uuid so that dm_copy_name_and_uuid no longer needs _hash_lock. Additionally, dm_copy_name_and_uuid can only be called if md exists so we can drop the dm_get() and dm_put() which can lead to a BUG() while md is being freed. The deadlock: #0 [ffff8106298dfb48] schedule at ffffffff80063035 #1 [ffff8106298dfc20] __down_read at ffffffff8006475d #2 [ffff8106298dfc60] dm_copy_name_and_uuid at ffffffff8824f740 #3 [ffff8106298dfc90] dm_send_uevents at ffffffff88252685 #4 [ffff8106298dfcd0] event_callback at ffffffff8824c678 #5 [ffff8106298dfd00] dm_table_event at ffffffff8824dd01 #6 [ffff8106298dfd10] __hash_remove at ffffffff882507ad #7 [ffff8106298dfd30] dev_remove at ffffffff88250865 #8 [ffff8106298dfd60] ctl_ioctl at ffffffff88250d80 #9 [ffff8106298dfee0] do_ioctl at ffffffff800418c4 #10 [ffff8106298dff00] vfs_ioctl at ffffffff8002fab9 #11 [ffff8106298dff40] sys_ioctl at ffffffff8004bdaf #12 [ffff8106298dff80] tracesys at ffffffff8005d28d (via system_call) Cc: stable@kernel.org Reported-by: guy keren <choo@actcom.co.il> Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
2009-12-11 07:51:52 +08:00
* When a device is being removed this copy fails and we
* discard these unsent events.
*/
if (dm_copy_name_and_uuid(event->md, event->name,
event->uuid)) {
dm: avoid _hash_lock deadlock Fix a reported deadlock if there are still unprocessed multipath events on a device that is being removed. _hash_lock is held during dev_remove while trying to send the outstanding events. Sending the events requests the _hash_lock again in dm_copy_name_and_uuid. This patch introduces a separate lock around regions that modify the link to the hash table (dm_set_mdptr) or the name or uuid so that dm_copy_name_and_uuid no longer needs _hash_lock. Additionally, dm_copy_name_and_uuid can only be called if md exists so we can drop the dm_get() and dm_put() which can lead to a BUG() while md is being freed. The deadlock: #0 [ffff8106298dfb48] schedule at ffffffff80063035 #1 [ffff8106298dfc20] __down_read at ffffffff8006475d #2 [ffff8106298dfc60] dm_copy_name_and_uuid at ffffffff8824f740 #3 [ffff8106298dfc90] dm_send_uevents at ffffffff88252685 #4 [ffff8106298dfcd0] event_callback at ffffffff8824c678 #5 [ffff8106298dfd00] dm_table_event at ffffffff8824dd01 #6 [ffff8106298dfd10] __hash_remove at ffffffff882507ad #7 [ffff8106298dfd30] dev_remove at ffffffff88250865 #8 [ffff8106298dfd60] ctl_ioctl at ffffffff88250d80 #9 [ffff8106298dfee0] do_ioctl at ffffffff800418c4 #10 [ffff8106298dff00] vfs_ioctl at ffffffff8002fab9 #11 [ffff8106298dff40] sys_ioctl at ffffffff8004bdaf #12 [ffff8106298dff80] tracesys at ffffffff8005d28d (via system_call) Cc: stable@kernel.org Reported-by: guy keren <choo@actcom.co.il> Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
2009-12-11 07:51:52 +08:00
DMINFO("%s: skipping sending uevent for lost device",
__func__);
goto uevent_free;
}
if (add_uevent_var(&event->ku_env, "DM_NAME=%s", event->name)) {
DMERR("%s: add_uevent_var() for DM_NAME failed",
__func__);
goto uevent_free;
}
if (add_uevent_var(&event->ku_env, "DM_UUID=%s", event->uuid)) {
DMERR("%s: add_uevent_var() for DM_UUID failed",
__func__);
goto uevent_free;
}
r = kobject_uevent_env(kobj, event->action, event->ku_env.envp);
if (r)
DMERR("%s: kobject_uevent_env failed", __func__);
uevent_free:
dm_uevent_free(event);
}
}
EXPORT_SYMBOL_GPL(dm_send_uevents);
/**
* dm_path_uevent - called to create a new path event and queue it
*
* @event_type: path event type enum
* @ti: pointer to a dm_target
* @path: string containing pathname
* @nr_valid_paths: number of valid paths remaining
*
*/
void dm_path_uevent(enum dm_uevent_type event_type, struct dm_target *ti,
const char *path, unsigned nr_valid_paths)
{
struct mapped_device *md = dm_table_get_md(ti->table);
struct dm_uevent *event;
if (event_type >= ARRAY_SIZE(_dm_uevent_type_names)) {
DMERR("%s: Invalid event_type %d", __func__, event_type);
dm table: remove dm_get from dm_table_get_md Remove the dm_get() in dm_table_get_md() because dm_table_get_md() could be called from presuspend/postsuspend, which are called while mapped_device is in DMF_FREEING state, where dm_get() is not allowed. Justification for that is the lifetime of both objects: As far as the current dm design/implementation, mapped_device is never freed while targets are doing something, because dm core waits for targets to become quiet in dm_put() using presuspend/postsuspend. So targets should be able to touch mapped_device without holding reference count of the mapped_device, and we should allow targets to touch mapped_device even if it is in DMF_FREEING state. Backgrounds: I'm trying to remove the multipath internal queue, since dm core now has a generic queue for request-based dm. In the patch-set, the multipath target wants to request dm core to start/stop queue. One of such start/stop requests can happen during postsuspend() while the target waits for pg-init to complete, because the target stops queue when starting pg-init and tries to restart it when completing pg-init. Since queue belongs to mapped_device, it involves calling dm_table_get_md() and dm_put(). On the other hand, postsuspend() is called in dm_put() for mapped_device which is in DMF_FREEING state, and that triggers BUG_ON(DMF_FREEING) in the 2nd dm_put(). I had tried to solve this problem by changing only multipath not to touch mapped_device which is in DMF_FREEING state, but I couldn't and I came up with a question why we need dm_get() in dm_table_get_md(). Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com> Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
2010-03-06 10:29:52 +08:00
return;
}
event = dm_build_path_uevent(md, ti,
_dm_uevent_type_names[event_type].action,
_dm_uevent_type_names[event_type].name,
path, nr_valid_paths);
if (IS_ERR(event))
dm table: remove dm_get from dm_table_get_md Remove the dm_get() in dm_table_get_md() because dm_table_get_md() could be called from presuspend/postsuspend, which are called while mapped_device is in DMF_FREEING state, where dm_get() is not allowed. Justification for that is the lifetime of both objects: As far as the current dm design/implementation, mapped_device is never freed while targets are doing something, because dm core waits for targets to become quiet in dm_put() using presuspend/postsuspend. So targets should be able to touch mapped_device without holding reference count of the mapped_device, and we should allow targets to touch mapped_device even if it is in DMF_FREEING state. Backgrounds: I'm trying to remove the multipath internal queue, since dm core now has a generic queue for request-based dm. In the patch-set, the multipath target wants to request dm core to start/stop queue. One of such start/stop requests can happen during postsuspend() while the target waits for pg-init to complete, because the target stops queue when starting pg-init and tries to restart it when completing pg-init. Since queue belongs to mapped_device, it involves calling dm_table_get_md() and dm_put(). On the other hand, postsuspend() is called in dm_put() for mapped_device which is in DMF_FREEING state, and that triggers BUG_ON(DMF_FREEING) in the 2nd dm_put(). I had tried to solve this problem by changing only multipath not to touch mapped_device which is in DMF_FREEING state, but I couldn't and I came up with a question why we need dm_get() in dm_table_get_md(). Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com> Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
2010-03-06 10:29:52 +08:00
return;
dm_uevent_add(md, &event->elist);
}
EXPORT_SYMBOL_GPL(dm_path_uevent);
int dm_uevent_init(void)
{
_dm_event_cache = KMEM_CACHE(dm_uevent, 0);
if (!_dm_event_cache)
return -ENOMEM;
DMINFO("version 1.0.3");
return 0;
}
void dm_uevent_exit(void)
{
kmem_cache_destroy(_dm_event_cache);
}