/*
* Copyright (C) 2001, 2002 Sistina Software (UK) Limited.
- * Copyright (C) 2004 - 2005 Red Hat, Inc. All rights reserved.
+ * Copyright (C) 2004 - 2006 Red Hat, Inc. All rights reserved.
*
* This file is released under the GPL.
*/
#include <linux/init.h>
#include <linux/wait.h>
#include <linux/slab.h>
-#include <linux/devfs_fs_kernel.h>
#include <linux/dm-ioctl.h>
+#include <linux/hdreg.h>
+#include <linux/compat.h>
#include <asm/uaccess.h>
+#define DM_MSG_PREFIX "ioctl"
#define DM_DRIVER_EMAIL "dm-devel@redhat.com"
/*-----------------------------------------------------------------
static struct list_head _name_buckets[NUM_BUCKETS];
static struct list_head _uuid_buckets[NUM_BUCKETS];
-static void dm_hash_remove_all(void);
+static void dm_hash_remove_all(int keep_open_devices);
/*
* Guards access to both hash tables.
*/
static DECLARE_RWSEM(_hash_lock);
+/*
+ * Protects use of mdptr to obtain hash cell name and uuid from mapped device.
+ */
+static DEFINE_MUTEX(dm_hash_cells_mutex);
+
static void init_buckets(struct list_head *buckets)
{
unsigned int i;
{
init_buckets(_name_buckets);
init_buckets(_uuid_buckets);
- devfs_mk_dir(DM_DIR);
return 0;
}
static void dm_hash_exit(void)
{
- dm_hash_remove_all();
- devfs_remove(DM_DIR);
+ dm_hash_remove_all(0);
}
/*-----------------------------------------------------------------
unsigned int h = hash_str(str);
list_for_each_entry (hc, _name_buckets + h, name_list)
- if (!strcmp(hc->name, str))
+ if (!strcmp(hc->name, str)) {
+ dm_get(hc->md);
return hc;
+ }
return NULL;
}
unsigned int h = hash_str(str);
list_for_each_entry (hc, _uuid_buckets + h, uuid_list)
- if (!strcmp(hc->uuid, str))
+ if (!strcmp(hc->uuid, str)) {
+ dm_get(hc->md);
return hc;
+ }
return NULL;
}
}
/*
- * devfs stuff.
- */
-static int register_with_devfs(struct hash_cell *hc)
-{
- struct gendisk *disk = dm_disk(hc->md);
-
- devfs_mk_bdev(MKDEV(disk->major, disk->first_minor),
- S_IFBLK | S_IRUSR | S_IWUSR | S_IRGRP,
- DM_DIR "/%s", hc->name);
- return 0;
-}
-
-static int unregister_with_devfs(struct hash_cell *hc)
-{
- devfs_remove(DM_DIR"/%s", hc->name);
- return 0;
-}
-
-/*
* The kdev_t and uuid of a device can never change once it is
* initially inserted.
*/
static int dm_hash_insert(const char *name, const char *uuid, struct mapped_device *md)
{
- struct hash_cell *cell;
+ struct hash_cell *cell, *hc;
/*
* Allocate the new cells.
* Insert the cell into both hash tables.
*/
down_write(&_hash_lock);
- if (__get_name_cell(name))
+ hc = __get_name_cell(name);
+ if (hc) {
+ dm_put(hc->md);
goto bad;
+ }
list_add(&cell->name_list, _name_buckets + hash_str(name));
if (uuid) {
- if (__get_uuid_cell(uuid)) {
+ hc = __get_uuid_cell(uuid);
+ if (hc) {
list_del(&cell->name_list);
+ dm_put(hc->md);
goto bad;
}
list_add(&cell->uuid_list, _uuid_buckets + hash_str(uuid));
}
- register_with_devfs(cell);
dm_get(md);
+ mutex_lock(&dm_hash_cells_mutex);
dm_set_mdptr(md, cell);
+ mutex_unlock(&dm_hash_cells_mutex);
up_write(&_hash_lock);
return 0;
/* remove from the dev hash */
list_del(&hc->uuid_list);
list_del(&hc->name_list);
- unregister_with_devfs(hc);
+ mutex_lock(&dm_hash_cells_mutex);
dm_set_mdptr(hc->md, NULL);
+ mutex_unlock(&dm_hash_cells_mutex);
- table = dm_get_table(hc->md);
+ table = dm_get_live_table(hc->md);
if (table) {
dm_table_event(table);
dm_table_put(table);
}
- dm_put(hc->md);
if (hc->new_map)
- dm_table_put(hc->new_map);
+ dm_table_destroy(hc->new_map);
+ dm_put(hc->md);
free_cell(hc);
}
-static void dm_hash_remove_all(void)
+static void dm_hash_remove_all(int keep_open_devices)
{
- int i;
+ int i, dev_skipped, dev_removed;
struct hash_cell *hc;
struct list_head *tmp, *n;
down_write(&_hash_lock);
+
+retry:
+ dev_skipped = dev_removed = 0;
for (i = 0; i < NUM_BUCKETS; i++) {
list_for_each_safe (tmp, n, _name_buckets + i) {
hc = list_entry(tmp, struct hash_cell, name_list);
+
+ if (keep_open_devices &&
+ dm_lock_for_deletion(hc->md)) {
+ dev_skipped++;
+ continue;
+ }
__hash_remove(hc);
+ dev_removed = 1;
}
}
+
+ /*
+ * Some mapped devices may be using other mapped devices, so if any
+ * still exist, repeat until we make no further progress.
+ */
+ if (dev_skipped) {
+ if (dev_removed)
+ goto retry;
+
+ DMWARN("remove_all left %d open device(s)", dev_skipped);
+ }
+
up_write(&_hash_lock);
}
-static int dm_hash_rename(const char *old, const char *new)
+static int dm_hash_rename(uint32_t cookie, const char *old, const char *new)
{
char *new_name, *old_name;
struct hash_cell *hc;
if (hc) {
DMWARN("asked to rename to an already existing name %s -> %s",
old, new);
+ dm_put(hc->md);
up_write(&_hash_lock);
kfree(new_name);
return -EBUSY;
/*
* rename and move the name cell.
*/
- unregister_with_devfs(hc);
-
list_del(&hc->name_list);
old_name = hc->name;
+ mutex_lock(&dm_hash_cells_mutex);
hc->name = new_name;
+ mutex_unlock(&dm_hash_cells_mutex);
list_add(&hc->name_list, _name_buckets + hash_str(new_name));
- /* rename the device node in devfs */
- register_with_devfs(hc);
-
/*
* Wake up any dm event waiters.
*/
- table = dm_get_table(hc->md);
+ table = dm_get_live_table(hc->md);
if (table) {
dm_table_event(table);
dm_table_put(table);
}
+ dm_kobject_uevent(hc->md, KOBJ_CHANGE, cookie);
+
+ dm_put(hc->md);
up_write(&_hash_lock);
kfree(old_name);
return 0;
static int remove_all(struct dm_ioctl *param, size_t param_size)
{
- dm_hash_remove_all();
+ dm_hash_remove_all(1);
param->data_size = 0;
return 0;
}
old_nl->next = (uint32_t) ((void *) nl -
(void *) old_nl);
disk = dm_disk(hc->md);
- nl->dev = huge_encode_dev(MKDEV(disk->major, disk->first_minor));
+ nl->dev = huge_encode_dev(disk_devt(disk));
nl->next = 0;
strcpy(nl->name, hc->name);
{
struct gendisk *disk = dm_disk(md);
struct dm_table *table;
- struct block_device *bdev;
param->flags &= ~(DM_SUSPEND_FLAG | DM_READONLY_FLAG |
DM_ACTIVE_PRESENT_FLAG);
if (dm_suspended(md))
param->flags |= DM_SUSPEND_FLAG;
- param->dev = huge_encode_dev(MKDEV(disk->major, disk->first_minor));
-
- if (!(param->flags & DM_SKIP_BDGET_FLAG)) {
- bdev = bdget_disk(disk, 0);
- if (!bdev)
- return -ENXIO;
+ param->dev = huge_encode_dev(disk_devt(disk));
- /*
- * Yes, this will be out of date by the time it gets back
- * to userland, but it is still very useful for
- * debugging.
- */
- param->open_count = bdev->bd_openers;
- bdput(bdev);
- } else
- param->open_count = -1;
+ /*
+ * Yes, this will be out of date by the time it gets back
+ * to userland, but it is still very useful for
+ * debugging.
+ */
+ param->open_count = dm_open_count(md);
- if (disk->policy)
+ if (get_disk_ro(disk))
param->flags |= DM_READONLY_FLAG;
param->event_nr = dm_get_event_nr(md);
- table = dm_get_table(md);
+ table = dm_get_live_table(md);
if (table) {
param->flags |= DM_ACTIVE_PRESENT_FLAG;
param->target_count = dm_table_get_num_targets(table);
static int dev_create(struct dm_ioctl *param, size_t param_size)
{
- int r;
+ int r, m = DM_ANY_MINOR;
struct mapped_device *md;
r = check_name(param->name);
return r;
if (param->flags & DM_PERSISTENT_DEV_FLAG)
- r = dm_create_with_minor(MINOR(huge_decode_dev(param->dev)), &md);
- else
- r = dm_create(&md);
+ m = MINOR(huge_decode_dev(param->dev));
+ r = dm_create(m, &md);
if (r)
return r;
/*
* Always use UUID for lookups if it's present, otherwise use name or dev.
*/
-static inline struct hash_cell *__find_device_hash_cell(struct dm_ioctl *param)
+static struct hash_cell *__find_device_hash_cell(struct dm_ioctl *param)
{
+ struct mapped_device *md;
+ void *mdptr = NULL;
+
if (*param->uuid)
return __get_uuid_cell(param->uuid);
- else if (*param->name)
+
+ if (*param->name)
return __get_name_cell(param->name);
- else
- return dm_get_mdptr(huge_decode_dev(param->dev));
+
+ md = dm_get_md(huge_decode_dev(param->dev));
+ if (!md)
+ goto out;
+
+ mdptr = dm_get_mdptr(md);
+ if (!mdptr)
+ dm_put(md);
+
+out:
+ return mdptr;
}
-static inline struct mapped_device *find_device(struct dm_ioctl *param)
+static struct mapped_device *find_device(struct dm_ioctl *param)
{
struct hash_cell *hc;
struct mapped_device *md = NULL;
hc = __find_device_hash_cell(param);
if (hc) {
md = hc->md;
- dm_get(md);
/*
* Sneakily write in both the name and the uuid
* while we have the cell.
*/
- strncpy(param->name, hc->name, sizeof(param->name));
+ strlcpy(param->name, hc->name, sizeof(param->name));
if (hc->uuid)
- strncpy(param->uuid, hc->uuid, sizeof(param->uuid)-1);
+ strlcpy(param->uuid, hc->uuid, sizeof(param->uuid));
else
param->uuid[0] = '\0';
static int dev_remove(struct dm_ioctl *param, size_t param_size)
{
struct hash_cell *hc;
+ struct mapped_device *md;
+ int r;
down_write(&_hash_lock);
hc = __find_device_hash_cell(param);
return -ENXIO;
}
+ md = hc->md;
+
+ /*
+ * Ensure the device is not open and nothing further can open it.
+ */
+ r = dm_lock_for_deletion(md);
+ if (r) {
+ DMWARN("unable to remove open device %s", hc->name);
+ up_write(&_hash_lock);
+ dm_put(md);
+ return r;
+ }
+
__hash_remove(hc);
up_write(&_hash_lock);
+
+ dm_kobject_uevent(md, KOBJ_REMOVE, param->event_nr);
+
+ dm_put(md);
param->data_size = 0;
return 0;
}
int r;
char *new_name = (char *) param + param->data_start;
- if (new_name < (char *) (param + 1) ||
- invalid_str(new_name, (void *) param + param_size)) {
+ if (new_name < param->data ||
+ invalid_str(new_name, (void *) param + param_size) ||
+ strlen(new_name) > DM_NAME_LEN - 1) {
DMWARN("Invalid new logical volume name supplied.");
return -EINVAL;
}
return r;
param->data_size = 0;
- return dm_hash_rename(param->name, new_name);
+ return dm_hash_rename(param->event_nr, param->name, new_name);
+}
+
+static int dev_set_geometry(struct dm_ioctl *param, size_t param_size)
+{
+ int r = -EINVAL, x;
+ struct mapped_device *md;
+ struct hd_geometry geometry;
+ unsigned long indata[4];
+ char *geostr = (char *) param + param->data_start;
+
+ md = find_device(param);
+ if (!md)
+ return -ENXIO;
+
+ if (geostr < param->data ||
+ invalid_str(geostr, (void *) param + param_size)) {
+ DMWARN("Invalid geometry supplied.");
+ goto out;
+ }
+
+ x = sscanf(geostr, "%lu %lu %lu %lu", indata,
+ indata + 1, indata + 2, indata + 3);
+
+ if (x != 4) {
+ DMWARN("Unable to interpret geometry settings.");
+ goto out;
+ }
+
+ if (indata[0] > 65535 || indata[1] > 255 ||
+ indata[2] > 255 || indata[3] > ULONG_MAX) {
+ DMWARN("Geometry exceeds range limits.");
+ goto out;
+ }
+
+ geometry.cylinders = indata[0];
+ geometry.heads = indata[1];
+ geometry.sectors = indata[2];
+ geometry.start = indata[3];
+
+ r = dm_set_geometry(md, &geometry);
+ if (!r)
+ r = __dev_status(md, param);
+
+ param->data_size = 0;
+
+out:
+ dm_put(md);
+ return r;
}
static int do_suspend(struct dm_ioctl *param)
{
int r = 0;
+ unsigned suspend_flags = DM_SUSPEND_LOCKFS_FLAG;
struct mapped_device *md;
md = find_device(param);
if (!md)
return -ENXIO;
+ if (param->flags & DM_SKIP_LOCKFS_FLAG)
+ suspend_flags &= ~DM_SUSPEND_LOCKFS_FLAG;
+ if (param->flags & DM_NOFLUSH_FLAG)
+ suspend_flags |= DM_SUSPEND_NOFLUSH_FLAG;
+
if (!dm_suspended(md))
- r = dm_suspend(md);
+ r = dm_suspend(md, suspend_flags);
if (!r)
r = __dev_status(md, param);
static int do_resume(struct dm_ioctl *param)
{
int r = 0;
+ unsigned suspend_flags = DM_SUSPEND_LOCKFS_FLAG;
struct hash_cell *hc;
struct mapped_device *md;
struct dm_table *new_map;
}
md = hc->md;
- dm_get(md);
new_map = hc->new_map;
hc->new_map = NULL;
/* Do we need to load a new map ? */
if (new_map) {
/* Suspend if it isn't already suspended */
+ if (param->flags & DM_SKIP_LOCKFS_FLAG)
+ suspend_flags &= ~DM_SUSPEND_LOCKFS_FLAG;
+ if (param->flags & DM_NOFLUSH_FLAG)
+ suspend_flags |= DM_SUSPEND_NOFLUSH_FLAG;
if (!dm_suspended(md))
- dm_suspend(md);
+ dm_suspend(md, suspend_flags);
r = dm_swap_table(md, new_map);
if (r) {
+ dm_table_destroy(new_map);
dm_put(md);
- dm_table_put(new_map);
return r;
}
set_disk_ro(dm_disk(md), 0);
else
set_disk_ro(dm_disk(md), 1);
-
- dm_table_put(new_map);
}
if (dm_suspended(md))
r = dm_resume(md);
- if (!r)
+
+ if (!r) {
+ dm_kobject_uevent(md, KOBJ_CHANGE, param->event_nr);
r = __dev_status(md, param);
+ }
dm_put(md);
return r;
if (r)
goto out;
- table = dm_get_table(md);
+ table = dm_get_live_table(md);
if (table) {
retrieve_status(table, param, param_size);
dm_table_put(table);
return r;
}
-static inline int get_mode(struct dm_ioctl *param)
+static inline fmode_t get_mode(struct dm_ioctl *param)
{
- int mode = FMODE_READ | FMODE_WRITE;
+ fmode_t mode = FMODE_READ | FMODE_WRITE;
if (param->flags & DM_READONLY_FLAG)
mode = FMODE_READ;
next = spec->next;
}
+ r = dm_table_set_type(table);
+ if (r) {
+ DMWARN("unable to set table type");
+ return r;
+ }
+
return dm_table_complete(table);
}
+static int table_prealloc_integrity(struct dm_table *t,
+ struct mapped_device *md)
+{
+ struct list_head *devices = dm_table_get_devices(t);
+ struct dm_dev_internal *dd;
+
+ list_for_each_entry(dd, devices, list)
+ if (bdev_get_integrity(dd->dm_dev.bdev))
+ return blk_integrity_register(dm_disk(md), NULL);
+
+ return 0;
+}
+
static int table_load(struct dm_ioctl *param, size_t param_size)
{
int r;
struct hash_cell *hc;
struct dm_table *t;
+ struct mapped_device *md;
- r = dm_table_create(&t, get_mode(param), param->target_count);
+ md = find_device(param);
+ if (!md)
+ return -ENXIO;
+
+ r = dm_table_create(&t, get_mode(param), param->target_count, md);
if (r)
- return r;
+ goto out;
r = populate_table(t, param, param_size);
if (r) {
- dm_table_put(t);
- return r;
+ dm_table_destroy(t);
+ goto out;
+ }
+
+ r = table_prealloc_integrity(t, md);
+ if (r) {
+ DMERR("%s: could not register integrity profile.",
+ dm_device_name(md));
+ dm_table_destroy(t);
+ goto out;
+ }
+
+ r = dm_table_alloc_md_mempools(t);
+ if (r) {
+ DMWARN("unable to allocate mempools for this table");
+ dm_table_destroy(t);
+ goto out;
}
down_write(&_hash_lock);
- hc = __find_device_hash_cell(param);
- if (!hc) {
- DMWARN("device doesn't appear to be in the dev hash table.");
+ hc = dm_get_mdptr(md);
+ if (!hc || hc->md != md) {
+ DMWARN("device has been removed from the dev hash table.");
+ dm_table_destroy(t);
up_write(&_hash_lock);
- dm_table_put(t);
- return -ENXIO;
+ r = -ENXIO;
+ goto out;
}
if (hc->new_map)
- dm_table_put(hc->new_map);
+ dm_table_destroy(hc->new_map);
hc->new_map = t;
+ up_write(&_hash_lock);
+
param->flags |= DM_INACTIVE_PRESENT_FLAG;
+ r = __dev_status(md, param);
+
+out:
+ dm_put(md);
- r = __dev_status(hc->md, param);
- up_write(&_hash_lock);
return r;
}
{
int r;
struct hash_cell *hc;
+ struct mapped_device *md;
down_write(&_hash_lock);
}
if (hc->new_map) {
- dm_table_put(hc->new_map);
+ dm_table_destroy(hc->new_map);
hc->new_map = NULL;
}
param->flags &= ~DM_INACTIVE_PRESENT_FLAG;
r = __dev_status(hc->md, param);
+ md = hc->md;
up_write(&_hash_lock);
+ dm_put(md);
return r;
}
unsigned int count = 0;
struct list_head *tmp;
size_t len, needed;
- struct dm_dev *dd;
+ struct dm_dev_internal *dd;
struct dm_target_deps *deps;
deps = get_result_buffer(param, param_size, &len);
deps->count = count;
count = 0;
list_for_each_entry (dd, dm_table_get_devices(table), list)
- deps->dev[count++] = huge_encode_dev(dd->bdev->bd_dev);
+ deps->dev[count++] = huge_encode_dev(dd->dm_dev.bdev->bd_dev);
param->data_size = param->data_start + needed;
}
if (r)
goto out;
- table = dm_get_table(md);
+ table = dm_get_live_table(md);
if (table) {
retrieve_deps(table, param, param_size);
dm_table_put(table);
if (r)
goto out;
- table = dm_get_table(md);
+ table = dm_get_live_table(md);
if (table) {
retrieve_status(table, param, param_size);
dm_table_put(table);
if (r)
goto out;
- if (tmsg < (struct dm_target_msg *) (param + 1) ||
+ if (tmsg < (struct dm_target_msg *) param->data ||
invalid_str(tmsg->message, (void *) param + param_size)) {
DMWARN("Invalid target message parameters.");
r = -EINVAL;
goto out;
}
- table = dm_get_table(md);
+ table = dm_get_live_table(md);
if (!table)
goto out_argv;
- if (tmsg->sector >= dm_table_get_size(table)) {
- DMWARN("Target message sector outside device.");
- r = -EINVAL;
+ if (dm_deleting_md(md)) {
+ r = -ENXIO;
goto out_table;
}
ti = dm_table_find_target(table, tmsg->sector);
- if (ti->type->message)
+ if (!dm_target_is_valid(ti)) {
+ DMWARN("Target message sector outside device.");
+ r = -EINVAL;
+ } else if (ti->type->message)
r = ti->type->message(ti, argc, argv);
else {
DMWARN("Target type does not support messages");
{DM_LIST_VERSIONS_CMD, list_versions},
- {DM_TARGET_MSG_CMD, target_message}
+ {DM_TARGET_MSG_CMD, target_message},
+ {DM_DEV_SET_GEOMETRY_CMD, dev_set_geometry}
};
return (cmd >= ARRAY_SIZE(_ioctls)) ? NULL : _ioctls[cmd].fn;
{
struct dm_ioctl tmp, *dmi;
- if (copy_from_user(&tmp, user, sizeof(tmp)))
+ if (copy_from_user(&tmp, user, sizeof(tmp) - sizeof(tmp.data)))
return -EFAULT;
- if (tmp.data_size < sizeof(tmp))
+ if (tmp.data_size < (sizeof(tmp) - sizeof(tmp.data)))
return -EINVAL;
- dmi = (struct dm_ioctl *) vmalloc(tmp.data_size);
+ dmi = vmalloc(tmp.data_size);
if (!dmi)
return -ENOMEM;
return 0;
}
-static int ctl_ioctl(struct inode *inode, struct file *file,
- uint command, ulong u)
+static int ctl_ioctl(uint command, struct dm_ioctl __user *user)
{
int r = 0;
unsigned int cmd;
- struct dm_ioctl *param;
- struct dm_ioctl __user *user = (struct dm_ioctl __user *) u;
+ struct dm_ioctl *uninitialized_var(param);
ioctl_fn fn = NULL;
size_t param_size;
* Copy the parameters into kernel space.
*/
r = copy_params(user, ¶m);
- if (r) {
- current->flags &= ~PF_MEMALLOC;
- return r;
- }
- /*
- * FIXME: eventually we will remove the PF_MEMALLOC flag
- * here. However the tools still do nasty things like
- * 'load' while a device is suspended.
- */
+ current->flags &= ~PF_MEMALLOC;
+
+ if (r)
+ return r;
r = validate_params(cmd, param);
if (r)
out:
free_params(param);
- current->flags &= ~PF_MEMALLOC;
return r;
}
-static struct file_operations _ctl_fops = {
- .ioctl = ctl_ioctl,
+static long dm_ctl_ioctl(struct file *file, uint command, ulong u)
+{
+ return (long)ctl_ioctl(command, (struct dm_ioctl __user *)u);
+}
+
+#ifdef CONFIG_COMPAT
+static long dm_compat_ctl_ioctl(struct file *file, uint command, ulong u)
+{
+ return (long)dm_ctl_ioctl(file, command, (ulong) compat_ptr(u));
+}
+#else
+#define dm_compat_ctl_ioctl NULL
+#endif
+
+static const struct file_operations _ctl_fops = {
+ .unlocked_ioctl = dm_ctl_ioctl,
+ .compat_ioctl = dm_compat_ctl_ioctl,
.owner = THIS_MODULE,
};
static struct miscdevice _dm_misc = {
.minor = MISC_DYNAMIC_MINOR,
.name = DM_NAME,
- .devfs_name = "mapper/control",
+ .nodename = "mapper/control",
.fops = &_ctl_fops
};
dm_hash_exit();
}
+
+/**
+ * dm_copy_name_and_uuid - Copy mapped device name & uuid into supplied buffers
+ * @md: Pointer to mapped_device
+ * @name: Buffer (size DM_NAME_LEN) for name
+ * @uuid: Buffer (size DM_UUID_LEN) for uuid or empty string if uuid not defined
+ */
+int dm_copy_name_and_uuid(struct mapped_device *md, char *name, char *uuid)
+{
+ int r = 0;
+ struct hash_cell *hc;
+
+ if (!md)
+ return -ENXIO;
+
+ mutex_lock(&dm_hash_cells_mutex);
+ hc = dm_get_mdptr(md);
+ if (!hc || hc->md != md) {
+ r = -ENXIO;
+ goto out;
+ }
+
+ if (name)
+ strcpy(name, hc->name);
+ if (uuid)
+ strcpy(uuid, hc->uuid ? : "");
+
+out:
+ mutex_unlock(&dm_hash_cells_mutex);
+
+ return r;
+}