#include <linux/time.h>
#include <linux/workqueue.h>
#include <scsi/scsi_dh.h>
-#include <asm/atomic.h>
+#include <linux/atomic.h>
#define DM_MSG_PREFIX "multipath"
-#define MESG_STR(x) x, sizeof(x)
#define DM_PG_INIT_DELAY_MSECS 2000
#define DM_PG_INIT_DELAY_DEFAULT ((unsigned) -1)
struct list_head pgpaths;
};
+#define FEATURE_NO_PARTITIONS 1
+
/* Multipath context */
struct multipath {
struct list_head list;
unsigned pg_init_retries; /* Number of times to retry pg_init */
unsigned pg_init_count; /* Number of times pg_init called */
unsigned pg_init_delay_msecs; /* Number of msecs before pg_init retry */
+ unsigned features; /* Additional selected features */
struct work_struct process_queued_ios;
struct list_head queued_ios;
static void free_pgpaths(struct list_head *pgpaths, struct dm_target *ti)
{
struct pgpath *pgpath, *tmp;
- struct multipath *m = ti->private;
list_for_each_entry_safe(pgpath, tmp, pgpaths, list) {
list_del(&pgpath->list);
- if (m->hw_handler_name)
- scsi_dh_detach(bdev_get_queue(pgpath->path.dev->bdev));
dm_put_device(ti, pgpath->path.dev);
free_pgpath(pgpath);
}
kfree(m);
}
+static int set_mapinfo(struct multipath *m, union map_info *info)
+{
+ struct dm_mpath_io *mpio;
+
+ mpio = mempool_alloc(m->mpio_pool, GFP_ATOMIC);
+ if (!mpio)
+ return -ENOMEM;
+
+ memset(mpio, 0, sizeof(*mpio));
+ info->ptr = mpio;
+
+ return 0;
+}
+
+static void clear_mapinfo(struct multipath *m, union map_info *info)
+{
+ struct dm_mpath_io *mpio = info->ptr;
+
+ info->ptr = NULL;
+ mempool_free(mpio, m->mpio_pool);
+}
/*-----------------------------------------------
* Path selection
m->current_pgpath = path_to_pgpath(path);
+ if (!m->current_pgpath->path.dev) {
+ m->current_pgpath = NULL;
+ return -ENODEV;
+ }
+
if (m->current_pg != pg)
__switch_pg(m, m->current_pgpath);
}
static int map_io(struct multipath *m, struct request *clone,
- struct dm_mpath_io *mpio, unsigned was_queued)
+ union map_info *map_context, unsigned was_queued)
{
int r = DM_MAPIO_REMAPPED;
size_t nr_bytes = blk_rq_bytes(clone);
unsigned long flags;
struct pgpath *pgpath;
struct block_device *bdev;
+ struct dm_mpath_io *mpio = map_context->ptr;
spin_lock_irqsave(&m->lock, flags);
{
int r;
unsigned long flags;
- struct dm_mpath_io *mpio;
union map_info *info;
struct request *clone, *n;
LIST_HEAD(cl);
list_del_init(&clone->queuelist);
info = dm_get_rq_mapinfo(clone);
- mpio = info->ptr;
- r = map_io(m, clone, mpio, 1);
+ r = map_io(m, clone, info, 1);
if (r < 0) {
- mempool_free(mpio, m->mpio_pool);
+ clear_mapinfo(m, info);
dm_kill_unmapped_request(clone, r);
} else if (r == DM_MAPIO_REMAPPED)
dm_dispatch_request(clone);
else if (r == DM_MAPIO_REQUEUE) {
- mempool_free(mpio, m->mpio_pool);
+ clear_mapinfo(m, info);
dm_requeue_unmapped_request(clone);
}
}
* <#paths> <#per-path selector args>
* [<path> [<arg>]* ]+ ]+
*---------------------------------------------------------------*/
-struct param {
- unsigned min;
- unsigned max;
- char *error;
-};
-
-static int read_param(struct param *param, char *str, unsigned *v, char **error)
-{
- if (!str ||
- (sscanf(str, "%u", v) != 1) ||
- (*v < param->min) ||
- (*v > param->max)) {
- *error = param->error;
- return -EINVAL;
- }
-
- return 0;
-}
-
-struct arg_set {
- unsigned argc;
- char **argv;
-};
-
-static char *shift(struct arg_set *as)
-{
- char *r;
-
- if (as->argc) {
- as->argc--;
- r = *as->argv;
- as->argv++;
- return r;
- }
-
- return NULL;
-}
-
-static void consume(struct arg_set *as, unsigned n)
-{
- BUG_ON (as->argc < n);
- as->argc -= n;
- as->argv += n;
-}
-
-static int parse_path_selector(struct arg_set *as, struct priority_group *pg,
+static int parse_path_selector(struct dm_arg_set *as, struct priority_group *pg,
struct dm_target *ti)
{
int r;
struct path_selector_type *pst;
unsigned ps_argc;
- static struct param _params[] = {
+ static struct dm_arg _args[] = {
{0, 1024, "invalid number of path selector args"},
};
- pst = dm_get_path_selector(shift(as));
+ pst = dm_get_path_selector(dm_shift_arg(as));
if (!pst) {
ti->error = "unknown path selector type";
return -EINVAL;
}
- r = read_param(_params, shift(as), &ps_argc, &ti->error);
+ r = dm_read_arg_group(_args, as, &ps_argc, &ti->error);
if (r) {
dm_put_path_selector(pst);
return -EINVAL;
}
- if (ps_argc > as->argc) {
- dm_put_path_selector(pst);
- ti->error = "not enough arguments for path selector";
- return -EINVAL;
- }
-
r = pst->create(&pg->ps, ps_argc, as->argv);
if (r) {
dm_put_path_selector(pst);
}
pg->ps.type = pst;
- consume(as, ps_argc);
+ dm_consume_args(as, ps_argc);
return 0;
}
-static struct pgpath *parse_path(struct arg_set *as, struct path_selector *ps,
+static struct pgpath *parse_path(struct dm_arg_set *as, struct path_selector *ps,
struct dm_target *ti)
{
int r;
struct pgpath *p;
+ char *path;
struct multipath *m = ti->private;
/* we need at least a path arg */
if (!p)
return ERR_PTR(-ENOMEM);
- r = dm_get_device(ti, shift(as), dm_table_get_mode(ti->table),
+ path = dm_shift_arg(as);
+ r = dm_get_device(ti, path, dm_table_get_mode(ti->table),
&p->path.dev);
if (r) {
- ti->error = "error getting device";
- goto bad;
+ unsigned major, minor;
+
+ /* Try to add a failed device */
+ if (r == -ENXIO && sscanf(path, "%u:%u", &major, &minor) == 2) {
+ dev_t dev;
+
+ /* Extract the major/minor numbers */
+ dev = MKDEV(major, minor);
+ if (MAJOR(dev) != major || MINOR(dev) != minor) {
+ /* Nice try, didn't work */
+ DMWARN("Invalid device path %s", path);
+ ti->error = "error converting devnum";
+ goto bad;
+ }
+ DMWARN("adding disabled device %d:%d", major, minor);
+ p->path.dev = NULL;
+ format_dev_t(p->path.pdev, dev);
+ p->is_active = 0;
+ } else {
+ ti->error = "error getting device";
+ goto bad;
+ }
+ } else {
+ memcpy(p->path.pdev, p->path.dev->name, 16);
}
- if (m->hw_handler_name) {
+ if (p->path.dev) {
struct request_queue *q = bdev_get_queue(p->path.dev->bdev);
- r = scsi_dh_attach(q, m->hw_handler_name);
- if (r == -EBUSY) {
- /*
- * Already attached to different hw_handler,
- * try to reattach with correct one.
- */
- scsi_dh_detach(q);
+ if (m->hw_handler_name) {
r = scsi_dh_attach(q, m->hw_handler_name);
- }
-
- if (r < 0) {
- ti->error = "error attaching hardware handler";
- dm_put_device(ti, p->path.dev);
- goto bad;
+ if (r == -EBUSY) {
+ /*
+ * Already attached to different hw_handler,
+ * try to reattach with correct one.
+ */
+ scsi_dh_detach(q);
+ r = scsi_dh_attach(q, m->hw_handler_name);
+ }
+ if (r < 0) {
+ ti->error = "error attaching hardware handler";
+ dm_put_device(ti, p->path.dev);
+ goto bad;
+ }
+ } else {
+ /* Play safe and detach hardware handler */
+ scsi_dh_detach(q);
}
if (m->hw_handler_params) {
goto bad;
}
+ if (!p->is_active) {
+ ps->type->fail_path(ps, &p->path);
+ p->fail_count++;
+ m->nr_valid_paths--;
+ }
return p;
bad:
return ERR_PTR(r);
}
-static struct priority_group *parse_priority_group(struct arg_set *as,
+static struct priority_group *parse_priority_group(struct dm_arg_set *as,
struct multipath *m)
{
- static struct param _params[] = {
+ static struct dm_arg _args[] = {
{1, 1024, "invalid number of paths"},
{0, 1024, "invalid number of selector args"}
};
int r;
- unsigned i, nr_selector_args, nr_params;
+ unsigned i, nr_selector_args, nr_args;
struct priority_group *pg;
struct dm_target *ti = m->ti;
/*
* read the paths
*/
- r = read_param(_params, shift(as), &pg->nr_pgpaths, &ti->error);
+ r = dm_read_arg(_args, as, &pg->nr_pgpaths, &ti->error);
if (r)
goto bad;
- r = read_param(_params + 1, shift(as), &nr_selector_args, &ti->error);
+ r = dm_read_arg(_args + 1, as, &nr_selector_args, &ti->error);
if (r)
goto bad;
- nr_params = 1 + nr_selector_args;
+ nr_args = 1 + nr_selector_args;
for (i = 0; i < pg->nr_pgpaths; i++) {
struct pgpath *pgpath;
- struct arg_set path_args;
+ struct dm_arg_set path_args;
- if (as->argc < nr_params) {
+ if (as->argc < nr_args) {
ti->error = "not enough path parameters";
r = -EINVAL;
goto bad;
}
- path_args.argc = nr_params;
+ path_args.argc = nr_args;
path_args.argv = as->argv;
pgpath = parse_path(&path_args, &pg->ps, ti);
pgpath->pg = pg;
list_add_tail(&pgpath->list, &pg->pgpaths);
- consume(as, nr_params);
+ dm_consume_args(as, nr_args);
}
return pg;
return ERR_PTR(r);
}
-static int parse_hw_handler(struct arg_set *as, struct multipath *m)
+static int parse_hw_handler(struct dm_arg_set *as, struct multipath *m)
{
unsigned hw_argc;
int ret;
struct dm_target *ti = m->ti;
- static struct param _params[] = {
+ static struct dm_arg _args[] = {
{0, 1024, "invalid number of hardware handler args"},
};
- if (read_param(_params, shift(as), &hw_argc, &ti->error))
+ if (dm_read_arg_group(_args, as, &hw_argc, &ti->error))
return -EINVAL;
if (!hw_argc)
return 0;
- if (hw_argc > as->argc) {
- ti->error = "not enough arguments for hardware handler";
- return -EINVAL;
- }
-
- m->hw_handler_name = kstrdup(shift(as), GFP_KERNEL);
- request_module("scsi_dh_%s", m->hw_handler_name);
- if (scsi_dh_handler_exist(m->hw_handler_name) == 0) {
+ m->hw_handler_name = kstrdup(dm_shift_arg(as), GFP_KERNEL);
+ if (!try_then_request_module(scsi_dh_handler_exist(m->hw_handler_name),
+ "scsi_dh_%s", m->hw_handler_name)) {
ti->error = "unknown hardware handler type";
ret = -EINVAL;
goto fail;
for (i = 0, p+=j+1; i <= hw_argc - 2; i++, p+=j+1)
j = sprintf(p, "%s", as->argv[i]);
}
- consume(as, hw_argc - 1);
+ dm_consume_args(as, hw_argc - 1);
return 0;
fail:
return ret;
}
-static int parse_features(struct arg_set *as, struct multipath *m)
+static int parse_features(struct dm_arg_set *as, struct multipath *m)
{
int r;
unsigned argc;
struct dm_target *ti = m->ti;
- const char *param_name;
+ const char *arg_name;
- static struct param _params[] = {
+ static struct dm_arg _args[] = {
{0, 5, "invalid number of feature args"},
{1, 50, "pg_init_retries must be between 1 and 50"},
{0, 60000, "pg_init_delay_msecs must be between 0 and 60000"},
};
- r = read_param(_params, shift(as), &argc, &ti->error);
+ r = dm_read_arg_group(_args, as, &argc, &ti->error);
if (r)
return -EINVAL;
return 0;
do {
- param_name = shift(as);
+ arg_name = dm_shift_arg(as);
argc--;
- if (!strnicmp(param_name, MESG_STR("queue_if_no_path"))) {
+ if (!strcasecmp(arg_name, "queue_if_no_path")) {
r = queue_if_no_path(m, 1, 0);
continue;
}
- if (!strnicmp(param_name, MESG_STR("pg_init_retries")) &&
+ if (!strcasecmp(arg_name, "no_partitions")) {
+ m->features |= FEATURE_NO_PARTITIONS;
+ continue;
+ }
+ if (!strcasecmp(arg_name, "pg_init_retries") &&
(argc >= 1)) {
- r = read_param(_params + 1, shift(as),
- &m->pg_init_retries, &ti->error);
+ r = dm_read_arg(_args + 1, as, &m->pg_init_retries, &ti->error);
argc--;
continue;
}
- if (!strnicmp(param_name, MESG_STR("pg_init_delay_msecs")) &&
+ if (!strcasecmp(arg_name, "pg_init_delay_msecs") &&
(argc >= 1)) {
- r = read_param(_params + 2, shift(as),
- &m->pg_init_delay_msecs, &ti->error);
+ r = dm_read_arg(_args + 2, as, &m->pg_init_delay_msecs, &ti->error);
argc--;
continue;
}
static int multipath_ctr(struct dm_target *ti, unsigned int argc,
char **argv)
{
- /* target parameters */
- static struct param _params[] = {
- {1, 1024, "invalid number of priority groups"},
- {1, 1024, "invalid initial priority group number"},
+ /* target arguments */
+ static struct dm_arg _args[] = {
+ {0, 1024, "invalid number of priority groups"},
+ {0, 1024, "invalid initial priority group number"},
};
int r;
struct multipath *m;
- struct arg_set as;
+ struct dm_arg_set as;
unsigned pg_count = 0;
unsigned next_pg_num;
if (r)
goto bad;
- r = read_param(_params, shift(&as), &m->nr_priority_groups, &ti->error);
+ r = dm_read_arg(_args, &as, &m->nr_priority_groups, &ti->error);
if (r)
goto bad;
- r = read_param(_params + 1, shift(&as), &next_pg_num, &ti->error);
+ r = dm_read_arg(_args + 1, &as, &next_pg_num, &ti->error);
if (r)
goto bad;
+ if ((!m->nr_priority_groups && next_pg_num) ||
+ (m->nr_priority_groups && !next_pg_num)) {
+ ti->error = "invalid initial priority group";
+ r = -EINVAL;
+ goto bad;
+ }
+
/* parse the priority groups */
while (as.argc) {
struct priority_group *pg;
union map_info *map_context)
{
int r;
- struct dm_mpath_io *mpio;
struct multipath *m = (struct multipath *) ti->private;
- mpio = mempool_alloc(m->mpio_pool, GFP_ATOMIC);
- if (!mpio)
+ if (set_mapinfo(m, map_context) < 0)
/* ENOMEM, requeue */
return DM_MAPIO_REQUEUE;
- memset(mpio, 0, sizeof(*mpio));
- map_context->ptr = mpio;
clone->cmd_flags |= REQ_FAILFAST_TRANSPORT;
- r = map_io(m, clone, mpio, 0);
+ r = map_io(m, clone, map_context, 0);
if (r < 0 || r == DM_MAPIO_REQUEUE)
- mempool_free(mpio, m->mpio_pool);
+ clear_mapinfo(m, map_context);
return r;
}
if (!pgpath->is_active)
goto out;
- DMWARN("Failing path %s.", pgpath->path.dev->name);
+ DMWARN("Failing path %s.", pgpath->path.pdev);
pgpath->pg->ps.type->fail_path(&pgpath->pg->ps, &pgpath->path);
pgpath->is_active = 0;
m->current_pgpath = NULL;
dm_path_uevent(DM_UEVENT_PATH_FAILED, m->ti,
- pgpath->path.dev->name, m->nr_valid_paths);
+ pgpath->path.pdev, m->nr_valid_paths);
schedule_work(&m->trigger_event);
if (pgpath->is_active)
goto out;
+ if (!pgpath->path.dev) {
+ DMWARN("Cannot reinstate disabled path %s", pgpath->path.pdev);
+ r = -ENODEV;
+ goto out;
+ }
+
if (!pgpath->pg->ps.type->reinstate_path) {
DMWARN("Reinstate path not supported by path selector %s",
pgpath->pg->ps.type->name);
}
dm_path_uevent(DM_UEVENT_PATH_REINSTATED, m->ti,
- pgpath->path.dev->name, m->nr_valid_paths);
+ pgpath->path.pdev, m->nr_valid_paths);
schedule_work(&m->trigger_event);
struct pgpath *pgpath;
struct priority_group *pg;
+ if (!dev)
+ return 0;
+
list_for_each_entry(pg, &m->priority_groups, list) {
list_for_each_entry(pgpath, &pg->pgpaths, list) {
if (pgpath->path.dev == dev)
struct priority_group *pg;
unsigned pgnum;
unsigned long flags;
+ char dummy;
- if (!pgstr || (sscanf(pgstr, "%u", &pgnum) != 1) || !pgnum ||
+ if (!pgstr || (sscanf(pgstr, "%u%c", &pgnum, &dummy) != 1) || !pgnum ||
(pgnum > m->nr_priority_groups)) {
DMWARN("invalid PG number supplied to switch_pg_num");
return -EINVAL;
{
struct priority_group *pg;
unsigned pgnum;
+ char dummy;
- if (!pgstr || (sscanf(pgstr, "%u", &pgnum) != 1) || !pgnum ||
+ if (!pgstr || (sscanf(pgstr, "%u%c", &pgnum, &dummy) != 1) || !pgnum ||
(pgnum > m->nr_priority_groups)) {
DMWARN("invalid PG number supplied to bypass_pg");
return -EINVAL;
errors = 0;
break;
}
- DMERR("Could not failover the device: Handler scsi_dh_%s "
- "Error %d.", m->hw_handler_name, errors);
+ DMERR("Count not failover device %s: Handler scsi_dh_%s "
+ "was not loaded.", pgpath->path.dev->name,
+ m->hw_handler_name);
/*
* Fail path for now, so we do not ping pong
*/
*/
bypass_pg(m, pg, 1);
break;
+ case SCSI_DH_DEV_OFFLINED:
+ DMWARN("Device %s offlined.", pgpath->path.dev->name);
+ errors = 0;
+ break;
case SCSI_DH_RETRY:
/* Wait before retrying. */
delay_retry = 1;
spin_lock_irqsave(&m->lock, flags);
if (errors) {
if (pgpath == m->current_pgpath) {
- DMERR("Could not failover device. Error %d.", errors);
+ DMERR("Could not failover device %s, error %d.",
+ pgpath->path.dev->name, errors);
m->current_pgpath = NULL;
m->current_pg = NULL;
}
struct pgpath *pgpath =
container_of(work, struct pgpath, activate_path.work);
- scsi_dh_activate(bdev_get_queue(pgpath->path.dev->bdev),
- pg_init_done, pgpath);
+ if (pgpath->path.dev)
+ scsi_dh_activate(bdev_get_queue(pgpath->path.dev->bdev),
+ pg_init_done, pgpath);
}
/*
if (!error && !clone->errors)
return 0; /* I/O complete */
- if (error == -EOPNOTSUPP || error == -EREMOTEIO)
+ if (error == -EOPNOTSUPP || error == -EREMOTEIO || error == -EILSEQ)
return error;
if (mpio->pgpath)
struct path_selector *ps;
int r;
+ BUG_ON(!mpio);
+
r = do_end_io(m, clone, error, mpio);
if (pgpath) {
ps = &pgpath->pg->ps;
if (ps->type->end_io)
ps->type->end_io(ps, &pgpath->path, mpio->nr_bytes);
}
- mempool_free(mpio, m->mpio_pool);
+ clear_mapinfo(m, map_context);
return r;
}
else {
DMEMIT("%u ", m->queue_if_no_path +
(m->pg_init_retries > 0) * 2 +
- (m->pg_init_delay_msecs != DM_PG_INIT_DELAY_DEFAULT) * 2);
+ (m->pg_init_delay_msecs != DM_PG_INIT_DELAY_DEFAULT) * 2 +
+ (m->features & FEATURE_NO_PARTITIONS));
if (m->queue_if_no_path)
DMEMIT("queue_if_no_path ");
if (m->pg_init_retries)
DMEMIT("pg_init_retries %u ", m->pg_init_retries);
+ if (m->features & FEATURE_NO_PARTITIONS)
+ DMEMIT("no_partitions ");
if (m->pg_init_delay_msecs != DM_PG_INIT_DELAY_DEFAULT)
DMEMIT("pg_init_delay_msecs %u ", m->pg_init_delay_msecs);
}
else if (m->current_pg)
pg_num = m->current_pg->pg_num;
else
- pg_num = 1;
+ pg_num = (m->nr_priority_groups ? 1 : 0);
DMEMIT("%u ", pg_num);
pg->ps.type->info_args);
list_for_each_entry(p, &pg->pgpaths, list) {
- DMEMIT("%s %s %u ", p->path.dev->name,
+ DMEMIT("%s %s %u ", p->path.pdev,
p->is_active ? "A" : "F",
p->fail_count);
if (pg->ps.type->status)
pg->ps.type->table_args);
list_for_each_entry(p, &pg->pgpaths, list) {
- DMEMIT("%s ", p->path.dev->name);
+ DMEMIT("%s ", p->path.pdev);
if (pg->ps.type->status)
sz += pg->ps.type->status(&pg->ps,
&p->path, type, result + sz,
}
if (argc == 1) {
- if (!strnicmp(argv[0], MESG_STR("queue_if_no_path"))) {
+ if (!strcasecmp(argv[0], "queue_if_no_path")) {
r = queue_if_no_path(m, 1, 0);
goto out;
- } else if (!strnicmp(argv[0], MESG_STR("fail_if_no_path"))) {
+ } else if (!strcasecmp(argv[0], "fail_if_no_path")) {
r = queue_if_no_path(m, 0, 0);
goto out;
}
goto out;
}
- if (!strnicmp(argv[0], MESG_STR("disable_group"))) {
+ if (!strcasecmp(argv[0], "disable_group")) {
r = bypass_pg_num(m, argv[1], 1);
goto out;
- } else if (!strnicmp(argv[0], MESG_STR("enable_group"))) {
+ } else if (!strcasecmp(argv[0], "enable_group")) {
r = bypass_pg_num(m, argv[1], 0);
goto out;
- } else if (!strnicmp(argv[0], MESG_STR("switch_group"))) {
+ } else if (!strcasecmp(argv[0], "switch_group")) {
r = switch_pg_num(m, argv[1]);
goto out;
- } else if (!strnicmp(argv[0], MESG_STR("reinstate_path")))
+ } else if (!strcasecmp(argv[0], "reinstate_path"))
action = reinstate_path;
- else if (!strnicmp(argv[0], MESG_STR("fail_path")))
+ else if (!strcasecmp(argv[0], "fail_path"))
action = fail_path;
else {
DMWARN("Unrecognised multipath message received.");
if (!m->current_pgpath)
__choose_pgpath(m, 0);
- if (m->current_pgpath) {
+ if (m->current_pgpath && m->current_pgpath->path.dev) {
bdev = m->current_pgpath->path.dev->bdev;
mode = m->current_pgpath->path.dev->mode;
}
spin_unlock_irqrestore(&m->lock, flags);
+ /*
+ * Only pass ioctls through if the device sizes match exactly.
+ */
+ if (!r && ti->len != i_size_read(bdev->bd_inode) >> SECTOR_SHIFT)
+ r = scsi_verify_blk_ioctl(NULL, cmd);
+
return r ? : __blkdev_driver_ioctl(bdev, mode, cmd, arg);
}
projects / linux-flexiantxendom0-3.2.10.git / blobdiff