#include "dm-core.h"
#include "dm-rq.h"
#include <linux/module.h>
#include <linux/vmalloc.h>
#include <linux/blkdev.h>
#include <linux/blk-integrity.h>
#include <linux/namei.h>
#include <linux/ctype.h>
#include <linux/string.h>
#include <linux/slab.h>
#include <linux/interrupt.h>
#include <linux/mutex.h>
#include <linux/delay.h>
#include <linux/atomic.h>
#include <linux/blk-mq.h>
#include <linux/mount.h>
#include <linux/dax.h>
#define DM_MSG_PREFIX "table"
#define NODE_SIZE L1_CACHE_BYTES
#define KEYS_PER_NODE (NODE_SIZE / sizeof(sector_t))
#define CHILDREN_PER_NODE (KEYS_PER_NODE + 1)
static unsigned int int_log(unsigned int n, unsigned int base)
{
int result = 0;
while (n > 1) {
n = dm_div_up(n, base);
result++;
}
return result;
}
static inline unsigned int get_child(unsigned int n, unsigned int k)
{
return (n * CHILDREN_PER_NODE) + k;
}
static inline sector_t *get_node(struct dm_table *t,
unsigned int l, unsigned int n)
{
return t->index[l] + (n * KEYS_PER_NODE);
}
static sector_t high(struct dm_table *t, unsigned int l, unsigned int n)
{
for (; l < t->depth - 1; l++)
n = get_child(n, CHILDREN_PER_NODE - 1);
if (n >= t->counts[l])
return (sector_t) -1;
return get_node(t, l, n)[KEYS_PER_NODE - 1];
}
static int setup_btree_index(unsigned int l, struct dm_table *t)
{
unsigned int n, k;
sector_t *node;
for (n = 0U; n < t->counts[l]; n++) {
node = get_node(t, l, n);
for (k = 0U; k < KEYS_PER_NODE; k++)
node[k] = high(t, l + 1, get_child(n, k));
}
return 0;
}
static int alloc_targets(struct dm_table *t, unsigned int num)
{
sector_t *n_highs;
struct dm_target *n_targets;
n_highs = kvcalloc(num, sizeof(struct dm_target) + sizeof(sector_t),
GFP_KERNEL);
if (!n_highs)
return -ENOMEM;
n_targets = (struct dm_target *) (n_highs + num);
memset(n_highs, -1, sizeof(*n_highs) * num);
kvfree(t->highs);
t->num_allocated = num;
t->highs = n_highs;
t->targets = n_targets;
return 0;
}
int dm_table_create(struct dm_table **result, blk_mode_t mode,
unsigned int num_targets, struct mapped_device *md)
{
struct dm_table *t = kzalloc(sizeof(*t), GFP_KERNEL);
if (!t)
return -ENOMEM;
INIT_LIST_HEAD(&t->devices);
init_rwsem(&t->devices_lock);
if (!num_targets)
num_targets = KEYS_PER_NODE;
num_targets = dm_round_up(num_targets, KEYS_PER_NODE);
if (!num_targets) {
kfree(t);
return -ENOMEM;
}
if (alloc_targets(t, num_targets)) {
kfree(t);
return -ENOMEM;
}
t->type = DM_TYPE_NONE;
t->mode = mode;
t->md = md;
*result = t;
return 0;
}
static void free_devices(struct list_head *devices, struct mapped_device *md)
{
struct list_head *tmp, *next;
list_for_each_safe(tmp, next, devices) {
struct dm_dev_internal *dd =
list_entry(tmp, struct dm_dev_internal, list);
DMWARN("%s: dm_table_destroy: dm_put_device call missing for %s",
dm_device_name(md), dd->dm_dev->name);
dm_put_table_device(md, dd->dm_dev);
kfree(dd);
}
}
static void dm_table_destroy_crypto_profile(struct dm_table *t);
void dm_table_destroy(struct dm_table *t)
{
if (!t)
return;
if (t->depth >= 2)
kvfree(t->index[t->depth - 2]);
for (unsigned int i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
if (ti->type->dtr)
ti->type->dtr(ti);
dm_put_target_type(ti->type);
}
kvfree(t->highs);
free_devices(&t->devices, t->md);
dm_free_md_mempools(t->mempools);
dm_table_destroy_crypto_profile(t);
kfree(t);
}
static struct dm_dev_internal *find_device(struct list_head *l, dev_t dev)
{
struct dm_dev_internal *dd;
list_for_each_entry(dd, l, list)
if (dd->dm_dev->bdev->bd_dev == dev)
return dd;
return NULL;
}
static int device_area_is_invalid(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
struct queue_limits *limits = data;
struct block_device *bdev = dev->bdev;
sector_t dev_size = bdev_nr_sectors(bdev);
unsigned short logical_block_size_sectors =
limits->logical_block_size >> SECTOR_SHIFT;
if (!dev_size)
return 0;
if ((start >= dev_size) || (start + len > dev_size)) {
DMERR("%s: %pg too small for target: start=%llu, len=%llu, dev_size=%llu",
dm_device_name(ti->table->md), bdev,
(unsigned long long)start,
(unsigned long long)len,
(unsigned long long)dev_size);
return 1;
}
if (bdev_is_zoned(bdev)) {
unsigned int zone_sectors = bdev_zone_sectors(bdev);
if (start & (zone_sectors - 1)) {
DMERR("%s: start=%llu not aligned to h/w zone size %u of %pg",
dm_device_name(ti->table->md),
(unsigned long long)start,
zone_sectors, bdev);
return 1;
}
if (len & (zone_sectors - 1)) {
DMERR("%s: len=%llu not aligned to h/w zone size %u of %pg",
dm_device_name(ti->table->md),
(unsigned long long)len,
zone_sectors, bdev);
return 1;
}
}
if (logical_block_size_sectors <= 1)
return 0;
if (start & (logical_block_size_sectors - 1)) {
DMERR("%s: start=%llu not aligned to h/w logical block size %u of %pg",
dm_device_name(ti->table->md),
(unsigned long long)start,
limits->logical_block_size, bdev);
return 1;
}
if (len & (logical_block_size_sectors - 1)) {
DMERR("%s: len=%llu not aligned to h/w logical block size %u of %pg",
dm_device_name(ti->table->md),
(unsigned long long)len,
limits->logical_block_size, bdev);
return 1;
}
return 0;
}
static int upgrade_mode(struct dm_dev_internal *dd, blk_mode_t new_mode,
struct mapped_device *md)
{
int r;
struct dm_dev *old_dev, *new_dev;
old_dev = dd->dm_dev;
r = dm_get_table_device(md, dd->dm_dev->bdev->bd_dev,
dd->dm_dev->mode | new_mode, &new_dev);
if (r)
return r;
dd->dm_dev = new_dev;
dm_put_table_device(md, old_dev);
return 0;
}
int __ref dm_get_device(struct dm_target *ti, const char *path, blk_mode_t mode,
struct dm_dev **result)
{
int r;
dev_t dev;
unsigned int major, minor;
char dummy;
struct dm_dev_internal *dd;
struct dm_table *t = ti->table;
BUG_ON(!t);
if (sscanf(path, "%u:%u%c", &major, &minor, &dummy) == 2) {
dev = MKDEV(major, minor);
if (MAJOR(dev) != major || MINOR(dev) != minor)
return -EOVERFLOW;
} else {
r = lookup_bdev(path, &dev);
#ifndef MODULE
if (r && system_state < SYSTEM_RUNNING)
r = early_lookup_bdev(path, &dev);
#endif
if (r)
return r;
}
if (dev == disk_devt(t->md->disk))
return -EINVAL;
down_write(&t->devices_lock);
dd = find_device(&t->devices, dev);
if (!dd) {
dd = kmalloc(sizeof(*dd), GFP_KERNEL);
if (!dd) {
r = -ENOMEM;
goto unlock_ret_r;
}
r = dm_get_table_device(t->md, dev, mode, &dd->dm_dev);
if (r) {
kfree(dd);
goto unlock_ret_r;
}
refcount_set(&dd->count, 1);
list_add(&dd->list, &t->devices);
goto out;
} else if (dd->dm_dev->mode != (mode | dd->dm_dev->mode)) {
r = upgrade_mode(dd, mode, t->md);
if (r)
goto unlock_ret_r;
}
refcount_inc(&dd->count);
out:
up_write(&t->devices_lock);
*result = dd->dm_dev;
return 0;
unlock_ret_r:
up_write(&t->devices_lock);
return r;
}
EXPORT_SYMBOL(dm_get_device);
static int dm_set_device_limits(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
struct queue_limits *limits = data;
struct block_device *bdev = dev->bdev;
struct request_queue *q = bdev_get_queue(bdev);
if (unlikely(!q)) {
DMWARN("%s: Cannot set limits for nonexistent device %pg",
dm_device_name(ti->table->md), bdev);
return 0;
}
if (blk_stack_limits(limits, &q->limits,
get_start_sect(bdev) + start) < 0)
DMWARN("%s: adding target device %pg caused an alignment inconsistency: "
"physical_block_size=%u, logical_block_size=%u, "
"alignment_offset=%u, start=%llu",
dm_device_name(ti->table->md), bdev,
q->limits.physical_block_size,
q->limits.logical_block_size,
q->limits.alignment_offset,
(unsigned long long) start << SECTOR_SHIFT);
return 0;
}
void dm_put_device(struct dm_target *ti, struct dm_dev *d)
{
int found = 0;
struct dm_table *t = ti->table;
struct list_head *devices = &t->devices;
struct dm_dev_internal *dd;
down_write(&t->devices_lock);
list_for_each_entry(dd, devices, list) {
if (dd->dm_dev == d) {
found = 1;
break;
}
}
if (!found) {
DMERR("%s: device %s not in table devices list",
dm_device_name(t->md), d->name);
goto unlock_ret;
}
if (refcount_dec_and_test(&dd->count)) {
dm_put_table_device(t->md, d);
list_del(&dd->list);
kfree(dd);
}
unlock_ret:
up_write(&t->devices_lock);
}
EXPORT_SYMBOL(dm_put_device);
static int adjoin(struct dm_table *t, struct dm_target *ti)
{
struct dm_target *prev;
if (!t->num_targets)
return !ti->begin;
prev = &t->targets[t->num_targets - 1];
return (ti->begin == (prev->begin + prev->len));
}
static char **realloc_argv(unsigned int *size, char **old_argv)
{
char **argv;
unsigned int new_size;
gfp_t gfp;
if (*size) {
new_size = *size * 2;
gfp = GFP_KERNEL;
} else {
new_size = 8;
gfp = GFP_NOIO;
}
argv = kmalloc_array(new_size, sizeof(*argv), gfp);
if (argv && old_argv) {
memcpy(argv, old_argv, *size * sizeof(*argv));
*size = new_size;
}
kfree(old_argv);
return argv;
}
int dm_split_args(int *argc, char ***argvp, char *input)
{
char *start, *end = input, *out, **argv = NULL;
unsigned int array_size = 0;
*argc = 0;
if (!input) {
*argvp = NULL;
return 0;
}
argv = realloc_argv(&array_size, argv);
if (!argv)
return -ENOMEM;
while (1) {
start = skip_spaces(end);
if (!*start)
break;
end = out = start;
while (*end) {
if (*end == '\\' && *(end + 1)) {
*out++ = *(end + 1);
end += 2;
continue;
}
if (isspace(*end))
break;
*out++ = *end++;
}
if ((*argc + 1) > array_size) {
argv = realloc_argv(&array_size, argv);
if (!argv)
return -ENOMEM;
}
if (*end)
end++;
*out = '\0';
argv[*argc] = start;
(*argc)++;
}
*argvp = argv;
return 0;
}
static int validate_hardware_logical_block_alignment(struct dm_table *t,
struct queue_limits *limits)
{
unsigned short device_logical_block_size_sects =
limits->logical_block_size >> SECTOR_SHIFT;
unsigned short next_target_start = 0;
unsigned short remaining = 0;
struct dm_target *ti;
struct queue_limits ti_limits;
unsigned int i;
for (i = 0; i < t->num_targets; i++) {
ti = dm_table_get_target(t, i);
blk_set_stacking_limits(&ti_limits);
if (ti->type->iterate_devices)
ti->type->iterate_devices(ti, dm_set_device_limits,
&ti_limits);
if (remaining < ti->len &&
remaining & ((ti_limits.logical_block_size >>
SECTOR_SHIFT) - 1))
break;
next_target_start =
(unsigned short) ((next_target_start + ti->len) &
(device_logical_block_size_sects - 1));
remaining = next_target_start ?
device_logical_block_size_sects - next_target_start : 0;
}
if (remaining) {
DMERR("%s: table line %u (start sect %llu len %llu) "
"not aligned to h/w logical block size %u",
dm_device_name(t->md), i,
(unsigned long long) ti->begin,
(unsigned long long) ti->len,
limits->logical_block_size);
return -EINVAL;
}
return 0;
}
int dm_table_add_target(struct dm_table *t, const char *type,
sector_t start, sector_t len, char *params)
{
int r = -EINVAL, argc;
char **argv;
struct dm_target *ti;
if (t->singleton) {
DMERR("%s: target type %s must appear alone in table",
dm_device_name(t->md), t->targets->type->name);
return -EINVAL;
}
BUG_ON(t->num_targets >= t->num_allocated);
ti = t->targets + t->num_targets;
memset(ti, 0, sizeof(*ti));
if (!len) {
DMERR("%s: zero-length target", dm_device_name(t->md));
return -EINVAL;
}
ti->type = dm_get_target_type(type);
if (!ti->type) {
DMERR("%s: %s: unknown target type", dm_device_name(t->md), type);
return -EINVAL;
}
if (dm_target_needs_singleton(ti->type)) {
if (t->num_targets) {
ti->error = "singleton target type must appear alone in table";
goto bad;
}
t->singleton = true;
}
if (dm_target_always_writeable(ti->type) &&
!(t->mode & BLK_OPEN_WRITE)) {
ti->error = "target type may not be included in a read-only table";
goto bad;
}
if (t->immutable_target_type) {
if (t->immutable_target_type != ti->type) {
ti->error = "immutable target type cannot be mixed with other target types";
goto bad;
}
} else if (dm_target_is_immutable(ti->type)) {
if (t->num_targets) {
ti->error = "immutable target type cannot be mixed with other target types";
goto bad;
}
t->immutable_target_type = ti->type;
}
if (dm_target_has_integrity(ti->type))
t->integrity_added = 1;
ti->table = t;
ti->begin = start;
ti->len = len;
ti->error = "Unknown error";
if (!adjoin(t, ti)) {
ti->error = "Gap in table";
goto bad;
}
r = dm_split_args(&argc, &argv, params);
if (r) {
ti->error = "couldn't split parameters";
goto bad;
}
r = ti->type->ctr(ti, argc, argv);
kfree(argv);
if (r)
goto bad;
t->highs[t->num_targets++] = ti->begin + ti->len - 1;
if (!ti->num_discard_bios && ti->discards_supported)
DMWARN("%s: %s: ignoring discards_supported because num_discard_bios is zero.",
dm_device_name(t->md), type);
if (ti->limit_swap_bios && !static_key_enabled(&swap_bios_enabled.key))
static_branch_enable(&swap_bios_enabled);
return 0;
bad:
DMERR("%s: %s: %s (%pe)", dm_device_name(t->md), type, ti->error, ERR_PTR(r));
dm_put_target_type(ti->type);
return r;
}
static int validate_next_arg(const struct dm_arg *arg, struct dm_arg_set *arg_set,
unsigned int *value, char **error, unsigned int grouped)
{
const char *arg_str = dm_shift_arg(arg_set);
char dummy;
if (!arg_str ||
(sscanf(arg_str, "%u%c", value, &dummy) != 1) ||
(*value < arg->min) ||
(*value > arg->max) ||
(grouped && arg_set->argc < *value)) {
*error = arg->error;
return -EINVAL;
}
return 0;
}
int dm_read_arg(const struct dm_arg *arg, struct dm_arg_set *arg_set,
unsigned int *value, char **error)
{
return validate_next_arg(arg, arg_set, value, error, 0);
}
EXPORT_SYMBOL(dm_read_arg);
int dm_read_arg_group(const struct dm_arg *arg, struct dm_arg_set *arg_set,
unsigned int *value, char **error)
{
return validate_next_arg(arg, arg_set, value, error, 1);
}
EXPORT_SYMBOL(dm_read_arg_group);
const char *dm_shift_arg(struct dm_arg_set *as)
{
char *r;
if (as->argc) {
as->argc--;
r = *as->argv;
as->argv++;
return r;
}
return NULL;
}
EXPORT_SYMBOL(dm_shift_arg);
void dm_consume_args(struct dm_arg_set *as, unsigned int num_args)
{
BUG_ON(as->argc < num_args);
as->argc -= num_args;
as->argv += num_args;
}
EXPORT_SYMBOL(dm_consume_args);
static bool __table_type_bio_based(enum dm_queue_mode table_type)
{
return (table_type == DM_TYPE_BIO_BASED ||
table_type == DM_TYPE_DAX_BIO_BASED);
}
static bool __table_type_request_based(enum dm_queue_mode table_type)
{
return table_type == DM_TYPE_REQUEST_BASED;
}
void dm_table_set_type(struct dm_table *t, enum dm_queue_mode type)
{
t->type = type;
}
EXPORT_SYMBOL_GPL(dm_table_set_type);
static int device_not_dax_capable(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
if (dev->dax_dev)
return false;
DMDEBUG("%pg: error: dax unsupported by block device", dev->bdev);
return true;
}
static int device_not_dax_synchronous_capable(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
return !dev->dax_dev || !dax_synchronous(dev->dax_dev);
}
static bool dm_table_supports_dax(struct dm_table *t,
iterate_devices_callout_fn iterate_fn)
{
for (unsigned int i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
if (!ti->type->direct_access)
return false;
if (!ti->type->iterate_devices ||
ti->type->iterate_devices(ti, iterate_fn, NULL))
return false;
}
return true;
}
static int device_is_rq_stackable(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
struct block_device *bdev = dev->bdev;
struct request_queue *q = bdev_get_queue(bdev);
if (bdev_is_partition(bdev))
return false;
return queue_is_mq(q);
}
static int dm_table_determine_type(struct dm_table *t)
{
unsigned int bio_based = 0, request_based = 0, hybrid = 0;
struct dm_target *ti;
struct list_head *devices = dm_table_get_devices(t);
enum dm_queue_mode live_md_type = dm_get_md_type(t->md);
if (t->type != DM_TYPE_NONE) {
if (t->type == DM_TYPE_BIO_BASED) {
goto verify_bio_based;
}
BUG_ON(t->type == DM_TYPE_DAX_BIO_BASED);
goto verify_rq_based;
}
for (unsigned int i = 0; i < t->num_targets; i++) {
ti = dm_table_get_target(t, i);
if (dm_target_hybrid(ti))
hybrid = 1;
else if (dm_target_request_based(ti))
request_based = 1;
else
bio_based = 1;
if (bio_based && request_based) {
DMERR("Inconsistent table: different target types can't be mixed up");
return -EINVAL;
}
}
if (hybrid && !bio_based && !request_based) {
if (__table_type_request_based(live_md_type))
request_based = 1;
else
bio_based = 1;
}
if (bio_based) {
verify_bio_based:
t->type = DM_TYPE_BIO_BASED;
if (dm_table_supports_dax(t, device_not_dax_capable) ||
(list_empty(devices) && live_md_type == DM_TYPE_DAX_BIO_BASED)) {
t->type = DM_TYPE_DAX_BIO_BASED;
}
return 0;
}
BUG_ON(!request_based);
t->type = DM_TYPE_REQUEST_BASED;
verify_rq_based:
if (t->num_targets > 1) {
DMERR("request-based DM doesn't support multiple targets");
return -EINVAL;
}
if (list_empty(devices)) {
int srcu_idx;
struct dm_table *live_table = dm_get_live_table(t->md, &srcu_idx);
if (live_table)
t->type = live_table->type;
dm_put_live_table(t->md, srcu_idx);
return 0;
}
ti = dm_table_get_immutable_target(t);
if (!ti) {
DMERR("table load rejected: immutable target is required");
return -EINVAL;
} else if (ti->max_io_len) {
DMERR("table load rejected: immutable target that splits IO is not supported");
return -EINVAL;
}
if (!ti->type->iterate_devices ||
!ti->type->iterate_devices(ti, device_is_rq_stackable, NULL)) {
DMERR("table load rejected: including non-request-stackable devices");
return -EINVAL;
}
return 0;
}
enum dm_queue_mode dm_table_get_type(struct dm_table *t)
{
return t->type;
}
struct target_type *dm_table_get_immutable_target_type(struct dm_table *t)
{
return t->immutable_target_type;
}
struct dm_target *dm_table_get_immutable_target(struct dm_table *t)
{
if (t->num_targets > 1 ||
!dm_target_is_immutable(t->targets[0].type))
return NULL;
return t->targets;
}
struct dm_target *dm_table_get_wildcard_target(struct dm_table *t)
{
for (unsigned int i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
if (dm_target_is_wildcard(ti->type))
return ti;
}
return NULL;
}
bool dm_table_bio_based(struct dm_table *t)
{
return __table_type_bio_based(dm_table_get_type(t));
}
bool dm_table_request_based(struct dm_table *t)
{
return __table_type_request_based(dm_table_get_type(t));
}
static bool dm_table_supports_poll(struct dm_table *t);
static int dm_table_alloc_md_mempools(struct dm_table *t, struct mapped_device *md)
{
enum dm_queue_mode type = dm_table_get_type(t);
unsigned int per_io_data_size = 0, front_pad, io_front_pad;
unsigned int min_pool_size = 0, pool_size;
struct dm_md_mempools *pools;
if (unlikely(type == DM_TYPE_NONE)) {
DMERR("no table type is set, can't allocate mempools");
return -EINVAL;
}
pools = kzalloc_node(sizeof(*pools), GFP_KERNEL, md->numa_node_id);
if (!pools)
return -ENOMEM;
if (type == DM_TYPE_REQUEST_BASED) {
pool_size = dm_get_reserved_rq_based_ios();
front_pad = offsetof(struct dm_rq_clone_bio_info, clone);
goto init_bs;
}
for (unsigned int i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
per_io_data_size = max(per_io_data_size, ti->per_io_data_size);
min_pool_size = max(min_pool_size, ti->num_flush_bios);
}
pool_size = max(dm_get_reserved_bio_based_ios(), min_pool_size);
front_pad = roundup(per_io_data_size,
__alignof__(struct dm_target_io)) + DM_TARGET_IO_BIO_OFFSET;
io_front_pad = roundup(per_io_data_size,
__alignof__(struct dm_io)) + DM_IO_BIO_OFFSET;
if (bioset_init(&pools->io_bs, pool_size, io_front_pad,
dm_table_supports_poll(t) ? BIOSET_PERCPU_CACHE : 0))
goto out_free_pools;
if (t->integrity_supported &&
bioset_integrity_create(&pools->io_bs, pool_size))
goto out_free_pools;
init_bs:
if (bioset_init(&pools->bs, pool_size, front_pad, 0))
goto out_free_pools;
if (t->integrity_supported &&
bioset_integrity_create(&pools->bs, pool_size))
goto out_free_pools;
t->mempools = pools;
return 0;
out_free_pools:
dm_free_md_mempools(pools);
return -ENOMEM;
}
static int setup_indexes(struct dm_table *t)
{
int i;
unsigned int total = 0;
sector_t *indexes;
for (i = t->depth - 2; i >= 0; i--) {
t->counts[i] = dm_div_up(t->counts[i + 1], CHILDREN_PER_NODE);
total += t->counts[i];
}
indexes = kvcalloc(total, NODE_SIZE, GFP_KERNEL);
if (!indexes)
return -ENOMEM;
for (i = t->depth - 2; i >= 0; i--) {
t->index[i] = indexes;
indexes += (KEYS_PER_NODE * t->counts[i]);
setup_btree_index(i, t);
}
return 0;
}
static int dm_table_build_index(struct dm_table *t)
{
int r = 0;
unsigned int leaf_nodes;
leaf_nodes = dm_div_up(t->num_targets, KEYS_PER_NODE);
t->depth = 1 + int_log(leaf_nodes, CHILDREN_PER_NODE);
t->counts[t->depth - 1] = leaf_nodes;
t->index[t->depth - 1] = t->highs;
if (t->depth >= 2)
r = setup_indexes(t);
return r;
}
static bool integrity_profile_exists(struct gendisk *disk)
{
return !!blk_get_integrity(disk);
}
static struct gendisk *dm_table_get_integrity_disk(struct dm_table *t)
{
struct list_head *devices = dm_table_get_devices(t);
struct dm_dev_internal *dd = NULL;
struct gendisk *prev_disk = NULL, *template_disk = NULL;
for (unsigned int i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
if (!dm_target_passes_integrity(ti->type))
goto no_integrity;
}
list_for_each_entry(dd, devices, list) {
template_disk = dd->dm_dev->bdev->bd_disk;
if (!integrity_profile_exists(template_disk))
goto no_integrity;
else if (prev_disk &&
blk_integrity_compare(prev_disk, template_disk) < 0)
goto no_integrity;
prev_disk = template_disk;
}
return template_disk;
no_integrity:
if (prev_disk)
DMWARN("%s: integrity not set: %s and %s profile mismatch",
dm_device_name(t->md),
prev_disk->disk_name,
template_disk->disk_name);
return NULL;
}
static int dm_table_register_integrity(struct dm_table *t)
{
struct mapped_device *md = t->md;
struct gendisk *template_disk = NULL;
if (t->integrity_added)
return 0;
template_disk = dm_table_get_integrity_disk(t);
if (!template_disk)
return 0;
if (!integrity_profile_exists(dm_disk(md))) {
t->integrity_supported = true;
blk_integrity_register(dm_disk(md),
blk_get_integrity(template_disk));
return 0;
}
if (blk_integrity_compare(dm_disk(md), template_disk) < 0) {
DMERR("%s: conflict with existing integrity profile: %s profile mismatch",
dm_device_name(t->md),
template_disk->disk_name);
return 1;
}
t->integrity_supported = true;
return 0;
}
#ifdef CONFIG_BLK_INLINE_ENCRYPTION
struct dm_crypto_profile {
struct blk_crypto_profile profile;
struct mapped_device *md;
};
static int dm_keyslot_evict_callback(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
const struct blk_crypto_key *key = data;
blk_crypto_evict_key(dev->bdev, key);
return 0;
}
static int dm_keyslot_evict(struct blk_crypto_profile *profile,
const struct blk_crypto_key *key, unsigned int slot)
{
struct mapped_device *md =
container_of(profile, struct dm_crypto_profile, profile)->md;
struct dm_table *t;
int srcu_idx;
t = dm_get_live_table(md, &srcu_idx);
if (!t)
return 0;
for (unsigned int i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
if (!ti->type->iterate_devices)
continue;
ti->type->iterate_devices(ti, dm_keyslot_evict_callback,
(void *)key);
}
dm_put_live_table(md, srcu_idx);
return 0;
}
static int
device_intersect_crypto_capabilities(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
struct blk_crypto_profile *parent = data;
struct blk_crypto_profile *child =
bdev_get_queue(dev->bdev)->crypto_profile;
blk_crypto_intersect_capabilities(parent, child);
return 0;
}
void dm_destroy_crypto_profile(struct blk_crypto_profile *profile)
{
struct dm_crypto_profile *dmcp = container_of(profile,
struct dm_crypto_profile,
profile);
if (!profile)
return;
blk_crypto_profile_destroy(profile);
kfree(dmcp);
}
static void dm_table_destroy_crypto_profile(struct dm_table *t)
{
dm_destroy_crypto_profile(t->crypto_profile);
t->crypto_profile = NULL;
}
static int dm_table_construct_crypto_profile(struct dm_table *t)
{
struct dm_crypto_profile *dmcp;
struct blk_crypto_profile *profile;
unsigned int i;
bool empty_profile = true;
dmcp = kmalloc(sizeof(*dmcp), GFP_KERNEL);
if (!dmcp)
return -ENOMEM;
dmcp->md = t->md;
profile = &dmcp->profile;
blk_crypto_profile_init(profile, 0);
profile->ll_ops.keyslot_evict = dm_keyslot_evict;
profile->max_dun_bytes_supported = UINT_MAX;
memset(profile->modes_supported, 0xFF,
sizeof(profile->modes_supported));
for (i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
if (!dm_target_passes_crypto(ti->type)) {
blk_crypto_intersect_capabilities(profile, NULL);
break;
}
if (!ti->type->iterate_devices)
continue;
ti->type->iterate_devices(ti,
device_intersect_crypto_capabilities,
profile);
}
if (t->md->queue &&
!blk_crypto_has_capabilities(profile,
t->md->queue->crypto_profile)) {
DMERR("Inline encryption capabilities of new DM table were more restrictive than the old table's. This is not supported!");
dm_destroy_crypto_profile(profile);
return -EINVAL;
}
for (i = 0; i < ARRAY_SIZE(profile->modes_supported); i++) {
if (profile->modes_supported[i]) {
empty_profile = false;
break;
}
}
if (empty_profile) {
dm_destroy_crypto_profile(profile);
profile = NULL;
}
t->crypto_profile = profile;
return 0;
}
static void dm_update_crypto_profile(struct request_queue *q,
struct dm_table *t)
{
if (!t->crypto_profile)
return;
if (!q->crypto_profile) {
blk_crypto_register(t->crypto_profile, q);
} else {
blk_crypto_update_capabilities(q->crypto_profile,
t->crypto_profile);
dm_destroy_crypto_profile(t->crypto_profile);
}
t->crypto_profile = NULL;
}
#else /* CONFIG_BLK_INLINE_ENCRYPTION */
static int dm_table_construct_crypto_profile(struct dm_table *t)
{
return 0;
}
void dm_destroy_crypto_profile(struct blk_crypto_profile *profile)
{
}
static void dm_table_destroy_crypto_profile(struct dm_table *t)
{
}
static void dm_update_crypto_profile(struct request_queue *q,
struct dm_table *t)
{
}
#endif /* !CONFIG_BLK_INLINE_ENCRYPTION */
int dm_table_complete(struct dm_table *t)
{
int r;
r = dm_table_determine_type(t);
if (r) {
DMERR("unable to determine table type");
return r;
}
r = dm_table_build_index(t);
if (r) {
DMERR("unable to build btrees");
return r;
}
r = dm_table_register_integrity(t);
if (r) {
DMERR("could not register integrity profile.");
return r;
}
r = dm_table_construct_crypto_profile(t);
if (r) {
DMERR("could not construct crypto profile.");
return r;
}
r = dm_table_alloc_md_mempools(t, t->md);
if (r)
DMERR("unable to allocate mempools");
return r;
}
static DEFINE_MUTEX(_event_lock);
void dm_table_event_callback(struct dm_table *t,
void (*fn)(void *), void *context)
{
mutex_lock(&_event_lock);
t->event_fn = fn;
t->event_context = context;
mutex_unlock(&_event_lock);
}
void dm_table_event(struct dm_table *t)
{
mutex_lock(&_event_lock);
if (t->event_fn)
t->event_fn(t->event_context);
mutex_unlock(&_event_lock);
}
EXPORT_SYMBOL(dm_table_event);
inline sector_t dm_table_get_size(struct dm_table *t)
{
return t->num_targets ? (t->highs[t->num_targets - 1] + 1) : 0;
}
EXPORT_SYMBOL(dm_table_get_size);
struct dm_target *dm_table_find_target(struct dm_table *t, sector_t sector)
{
unsigned int l, n = 0, k = 0;
sector_t *node;
if (unlikely(sector >= dm_table_get_size(t)))
return NULL;
for (l = 0; l < t->depth; l++) {
n = get_child(n, k);
node = get_node(t, l, n);
for (k = 0; k < KEYS_PER_NODE; k++)
if (node[k] >= sector)
break;
}
return &t->targets[(KEYS_PER_NODE * n) + k];
}
static int device_not_poll_capable(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
struct request_queue *q = bdev_get_queue(dev->bdev);
return !test_bit(QUEUE_FLAG_POLL, &q->queue_flags);
}
static bool dm_table_any_dev_attr(struct dm_table *t,
iterate_devices_callout_fn func, void *data)
{
for (unsigned int i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
if (ti->type->iterate_devices &&
ti->type->iterate_devices(ti, func, data))
return true;
}
return false;
}
static int count_device(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
unsigned int *num_devices = data;
(*num_devices)++;
return 0;
}
static bool dm_table_supports_poll(struct dm_table *t)
{
for (unsigned int i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
if (!ti->type->iterate_devices ||
ti->type->iterate_devices(ti, device_not_poll_capable, NULL))
return false;
}
return true;
}
bool dm_table_has_no_data_devices(struct dm_table *t)
{
for (unsigned int i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
unsigned int num_devices = 0;
if (!ti->type->iterate_devices)
return false;
ti->type->iterate_devices(ti, count_device, &num_devices);
if (num_devices)
return false;
}
return true;
}
static int device_not_zoned_model(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
struct request_queue *q = bdev_get_queue(dev->bdev);
enum blk_zoned_model *zoned_model = data;
return blk_queue_zoned_model(q) != *zoned_model;
}
static bool dm_table_supports_zoned_model(struct dm_table *t,
enum blk_zoned_model zoned_model)
{
for (unsigned int i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
if (dm_target_supports_zoned_hm(ti->type)) {
if (!ti->type->iterate_devices ||
ti->type->iterate_devices(ti, device_not_zoned_model,
&zoned_model))
return false;
} else if (!dm_target_supports_mixed_zoned_model(ti->type)) {
if (zoned_model == BLK_ZONED_HM)
return false;
}
}
return true;
}
static int device_not_matches_zone_sectors(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
unsigned int *zone_sectors = data;
if (!bdev_is_zoned(dev->bdev))
return 0;
return bdev_zone_sectors(dev->bdev) != *zone_sectors;
}
static int validate_hardware_zoned_model(struct dm_table *t,
enum blk_zoned_model zoned_model,
unsigned int zone_sectors)
{
if (zoned_model == BLK_ZONED_NONE)
return 0;
if (!dm_table_supports_zoned_model(t, zoned_model)) {
DMERR("%s: zoned model is not consistent across all devices",
dm_device_name(t->md));
return -EINVAL;
}
if (!zone_sectors || !is_power_of_2(zone_sectors))
return -EINVAL;
if (dm_table_any_dev_attr(t, device_not_matches_zone_sectors, &zone_sectors)) {
DMERR("%s: zone sectors is not consistent across all zoned devices",
dm_device_name(t->md));
return -EINVAL;
}
return 0;
}
int dm_calculate_queue_limits(struct dm_table *t,
struct queue_limits *limits)
{
struct queue_limits ti_limits;
enum blk_zoned_model zoned_model = BLK_ZONED_NONE;
unsigned int zone_sectors = 0;
blk_set_stacking_limits(limits);
for (unsigned int i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
blk_set_stacking_limits(&ti_limits);
if (!ti->type->iterate_devices) {
if (ti->type->io_hints)
ti->type->io_hints(ti, &ti_limits);
goto combine_limits;
}
ti->type->iterate_devices(ti, dm_set_device_limits,
&ti_limits);
if (zoned_model == BLK_ZONED_NONE && ti_limits.zoned != BLK_ZONED_NONE) {
zoned_model = ti_limits.zoned;
zone_sectors = ti_limits.chunk_sectors;
}
if (ti->type->io_hints)
ti->type->io_hints(ti, &ti_limits);
if (ti->type->iterate_devices(ti, device_area_is_invalid,
&ti_limits))
return -EINVAL;
combine_limits:
if (blk_stack_limits(limits, &ti_limits, 0) < 0)
DMWARN("%s: adding target device (start sect %llu len %llu) "
"caused an alignment inconsistency",
dm_device_name(t->md),
(unsigned long long) ti->begin,
(unsigned long long) ti->len);
}
if (limits->zoned != BLK_ZONED_NONE) {
zoned_model = limits->zoned;
zone_sectors = limits->chunk_sectors;
}
if (validate_hardware_zoned_model(t, zoned_model, zone_sectors))
return -EINVAL;
return validate_hardware_logical_block_alignment(t, limits);
}
static void dm_table_verify_integrity(struct dm_table *t)
{
struct gendisk *template_disk = NULL;
if (t->integrity_added)
return;
if (t->integrity_supported) {
template_disk = dm_table_get_integrity_disk(t);
if (template_disk &&
blk_integrity_compare(dm_disk(t->md), template_disk) >= 0)
return;
}
if (integrity_profile_exists(dm_disk(t->md))) {
DMWARN("%s: unable to establish an integrity profile",
dm_device_name(t->md));
blk_integrity_unregister(dm_disk(t->md));
}
}
static int device_flush_capable(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
unsigned long flush = (unsigned long) data;
struct request_queue *q = bdev_get_queue(dev->bdev);
return (q->queue_flags & flush);
}
static bool dm_table_supports_flush(struct dm_table *t, unsigned long flush)
{
for (unsigned int i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
if (!ti->num_flush_bios)
continue;
if (ti->flush_supported)
return true;
if (ti->type->iterate_devices &&
ti->type->iterate_devices(ti, device_flush_capable, (void *) flush))
return true;
}
return false;
}
static int device_dax_write_cache_enabled(struct dm_target *ti,
struct dm_dev *dev, sector_t start,
sector_t len, void *data)
{
struct dax_device *dax_dev = dev->dax_dev;
if (!dax_dev)
return false;
if (dax_write_cache_enabled(dax_dev))
return true;
return false;
}
static int device_is_rotational(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
return !bdev_nonrot(dev->bdev);
}
static int device_is_not_random(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
struct request_queue *q = bdev_get_queue(dev->bdev);
return !blk_queue_add_random(q);
}
static int device_not_write_zeroes_capable(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
struct request_queue *q = bdev_get_queue(dev->bdev);
return !q->limits.max_write_zeroes_sectors;
}
static bool dm_table_supports_write_zeroes(struct dm_table *t)
{
for (unsigned int i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
if (!ti->num_write_zeroes_bios)
return false;
if (!ti->type->iterate_devices ||
ti->type->iterate_devices(ti, device_not_write_zeroes_capable, NULL))
return false;
}
return true;
}
static int device_not_nowait_capable(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
return !bdev_nowait(dev->bdev);
}
static bool dm_table_supports_nowait(struct dm_table *t)
{
for (unsigned int i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
if (!dm_target_supports_nowait(ti->type))
return false;
if (!ti->type->iterate_devices ||
ti->type->iterate_devices(ti, device_not_nowait_capable, NULL))
return false;
}
return true;
}
static int device_not_discard_capable(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
return !bdev_max_discard_sectors(dev->bdev);
}
static bool dm_table_supports_discards(struct dm_table *t)
{
for (unsigned int i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
if (!ti->num_discard_bios)
return false;
if (!ti->discards_supported &&
(!ti->type->iterate_devices ||
ti->type->iterate_devices(ti, device_not_discard_capable, NULL)))
return false;
}
return true;
}
static int device_not_secure_erase_capable(struct dm_target *ti,
struct dm_dev *dev, sector_t start,
sector_t len, void *data)
{
return !bdev_max_secure_erase_sectors(dev->bdev);
}
static bool dm_table_supports_secure_erase(struct dm_table *t)
{
for (unsigned int i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
if (!ti->num_secure_erase_bios)
return false;
if (!ti->type->iterate_devices ||
ti->type->iterate_devices(ti, device_not_secure_erase_capable, NULL))
return false;
}
return true;
}
static int device_requires_stable_pages(struct dm_target *ti,
struct dm_dev *dev, sector_t start,
sector_t len, void *data)
{
return bdev_stable_writes(dev->bdev);
}
int dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
struct queue_limits *limits)
{
bool wc = false, fua = false;
int r;
q->limits = *limits;
if (dm_table_supports_nowait(t))
blk_queue_flag_set(QUEUE_FLAG_NOWAIT, q);
else
blk_queue_flag_clear(QUEUE_FLAG_NOWAIT, q);
if (!dm_table_supports_discards(t)) {
q->limits.max_discard_sectors = 0;
q->limits.max_hw_discard_sectors = 0;
q->limits.discard_granularity = 0;
q->limits.discard_alignment = 0;
q->limits.discard_misaligned = 0;
}
if (!dm_table_supports_secure_erase(t))
q->limits.max_secure_erase_sectors = 0;
if (dm_table_supports_flush(t, (1UL << QUEUE_FLAG_WC))) {
wc = true;
if (dm_table_supports_flush(t, (1UL << QUEUE_FLAG_FUA)))
fua = true;
}
blk_queue_write_cache(q, wc, fua);
if (dm_table_supports_dax(t, device_not_dax_capable)) {
blk_queue_flag_set(QUEUE_FLAG_DAX, q);
if (dm_table_supports_dax(t, device_not_dax_synchronous_capable))
set_dax_synchronous(t->md->dax_dev);
} else
blk_queue_flag_clear(QUEUE_FLAG_DAX, q);
if (dm_table_any_dev_attr(t, device_dax_write_cache_enabled, NULL))
dax_write_cache(t->md->dax_dev, true);
if (dm_table_any_dev_attr(t, device_is_rotational, NULL))
blk_queue_flag_clear(QUEUE_FLAG_NONROT, q);
else
blk_queue_flag_set(QUEUE_FLAG_NONROT, q);
if (!dm_table_supports_write_zeroes(t))
q->limits.max_write_zeroes_sectors = 0;
dm_table_verify_integrity(t);
if (dm_table_any_dev_attr(t, device_requires_stable_pages, NULL))
blk_queue_flag_set(QUEUE_FLAG_STABLE_WRITES, q);
else
blk_queue_flag_clear(QUEUE_FLAG_STABLE_WRITES, q);
if (blk_queue_add_random(q) &&
dm_table_any_dev_attr(t, device_is_not_random, NULL))
blk_queue_flag_clear(QUEUE_FLAG_ADD_RANDOM, q);
if (blk_queue_is_zoned(q)) {
r = dm_set_zones_restrictions(t, q);
if (r)
return r;
if (!static_key_enabled(&zoned_enabled.key))
static_branch_enable(&zoned_enabled);
}
dm_update_crypto_profile(q, t);
disk_update_readahead(t->md->disk);
if (__table_type_bio_based(t->type)) {
if (dm_table_supports_poll(t))
blk_queue_flag_set(QUEUE_FLAG_POLL, q);
else
blk_queue_flag_clear(QUEUE_FLAG_POLL, q);
}
return 0;
}
struct list_head *dm_table_get_devices(struct dm_table *t)
{
return &t->devices;
}
blk_mode_t dm_table_get_mode(struct dm_table *t)
{
return t->mode;
}
EXPORT_SYMBOL(dm_table_get_mode);
enum suspend_mode {
PRESUSPEND,
PRESUSPEND_UNDO,
POSTSUSPEND,
};
static void suspend_targets(struct dm_table *t, enum suspend_mode mode)
{
lockdep_assert_held(&t->md->suspend_lock);
for (unsigned int i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
switch (mode) {
case PRESUSPEND:
if (ti->type->presuspend)
ti->type->presuspend(ti);
break;
case PRESUSPEND_UNDO:
if (ti->type->presuspend_undo)
ti->type->presuspend_undo(ti);
break;
case POSTSUSPEND:
if (ti->type->postsuspend)
ti->type->postsuspend(ti);
break;
}
}
}
void dm_table_presuspend_targets(struct dm_table *t)
{
if (!t)
return;
suspend_targets(t, PRESUSPEND);
}
void dm_table_presuspend_undo_targets(struct dm_table *t)
{
if (!t)
return;
suspend_targets(t, PRESUSPEND_UNDO);
}
void dm_table_postsuspend_targets(struct dm_table *t)
{
if (!t)
return;
suspend_targets(t, POSTSUSPEND);
}
int dm_table_resume_targets(struct dm_table *t)
{
unsigned int i;
int r = 0;
lockdep_assert_held(&t->md->suspend_lock);
for (i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
if (!ti->type->preresume)
continue;
r = ti->type->preresume(ti);
if (r) {
DMERR("%s: %s: preresume failed, error = %d",
dm_device_name(t->md), ti->type->name, r);
return r;
}
}
for (i = 0; i < t->num_targets; i++) {
struct dm_target *ti = dm_table_get_target(t, i);
if (ti->type->resume)
ti->type->resume(ti);
}
return 0;
}
struct mapped_device *dm_table_get_md(struct dm_table *t)
{
return t->md;
}
EXPORT_SYMBOL(dm_table_get_md);
const char *dm_table_device_name(struct dm_table *t)
{
return dm_device_name(t->md);
}
EXPORT_SYMBOL_GPL(dm_table_device_name);
void dm_table_run_md_queue_async(struct dm_table *t)
{
if (!dm_table_request_based(t))
return;
if (t->md->queue)
blk_mq_run_hw_queues(t->md->queue, true);
}
EXPORT_SYMBOL