summaryrefslogtreecommitdiff
path: root/drivers
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2018-05-05 09:41:44 +0300
committerLinus Torvalds <torvalds@linux-foundation.org>2018-05-05 09:41:44 +0300
commit2f50037a1c687ac928bbd47b6eb959b39f748ada (patch)
tree3b1e97aa8b1e4605822f9736588914809711092c /drivers
parent2e171ffcdf62a90ea7a0192728f81c1ac288de50 (diff)
parentb8b784958eccbf8f51ebeee65282ca3fd59ea391 (diff)
downloadlinux-2f50037a1c687ac928bbd47b6eb959b39f748ada.tar.xz
Merge tag 'for-linus-20180504' of git://git.kernel.dk/linux-block
Pull block fixes from Jens Axboe: "A collection of fixes that should to into this release. This contains: - Set of bcache fixes from Coly, fixing regression in patches that went into this series. - Set of NVMe fixes by way of Keith. - Set of bdi related fixes, one from Jan and two from Tetsuo Handa, fixing various issues around device addition/removal. - Two block inflight fixes from Omar, fixing issues around the transition to using tags for blk-mq inflight accounting that we did a few releases ago" * tag 'for-linus-20180504' of git://git.kernel.dk/linux-block: bdi: Fix oops in wb_workfn() nvmet: switch loopback target state to connecting when resetting nvme/multipath: Fix multipath disabled naming collisions nvme/multipath: Disable runtime writable enabling parameter nvme: Set integrity flag for user passthrough commands nvme: fix potential memory leak in option parsing bdi: Fix use after free bug in debugfs_remove() bdi: wake up concurrent wb_shutdown() callers. bcache: use pr_info() to inform duplicated CACHE_SET_IO_DISABLE set bcache: set dc->io_disable to true in conditional_stop_bcache_device() bcache: add wait_for_kthread_stop() in bch_allocator_thread() bcache: count backing device I/O error for writeback I/O bcache: set CACHE_SET_IO_DISABLE in bch_cached_dev_error() bcache: store disk name in struct cache and struct cached_dev blk-mq: fix sysfs inflight counter blk-mq: count allocated but not started requests in iostats inflight
Diffstat (limited to 'drivers')
-rw-r--r--drivers/md/bcache/alloc.c5
-rw-r--r--drivers/md/bcache/bcache.h4
-rw-r--r--drivers/md/bcache/debug.c3
-rw-r--r--drivers/md/bcache/io.c8
-rw-r--r--drivers/md/bcache/request.c5
-rw-r--r--drivers/md/bcache/super.c75
-rw-r--r--drivers/md/bcache/writeback.c4
-rw-r--r--drivers/nvme/host/core.c27
-rw-r--r--drivers/nvme/host/fabrics.c6
-rw-r--r--drivers/nvme/host/multipath.c24
-rw-r--r--drivers/nvme/host/nvme.h12
-rw-r--r--drivers/nvme/target/loop.c6
12 files changed, 117 insertions, 62 deletions
diff --git a/drivers/md/bcache/alloc.c b/drivers/md/bcache/alloc.c
index 004cc3cc6123..7fa2631b422c 100644
--- a/drivers/md/bcache/alloc.c
+++ b/drivers/md/bcache/alloc.c
@@ -290,7 +290,7 @@ do { \
if (kthread_should_stop() || \
test_bit(CACHE_SET_IO_DISABLE, &ca->set->flags)) { \
set_current_state(TASK_RUNNING); \
- return 0; \
+ goto out; \
} \
\
schedule(); \
@@ -378,6 +378,9 @@ retry_invalidate:
bch_prio_write(ca);
}
}
+out:
+ wait_for_kthread_stop();
+ return 0;
}
/* Allocation */
diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h
index d338b7086013..3a0cfb237af9 100644
--- a/drivers/md/bcache/bcache.h
+++ b/drivers/md/bcache/bcache.h
@@ -392,6 +392,8 @@ struct cached_dev {
#define DEFAULT_CACHED_DEV_ERROR_LIMIT 64
atomic_t io_errors;
unsigned error_limit;
+
+ char backing_dev_name[BDEVNAME_SIZE];
};
enum alloc_reserve {
@@ -464,6 +466,8 @@ struct cache {
atomic_long_t meta_sectors_written;
atomic_long_t btree_sectors_written;
atomic_long_t sectors_written;
+
+ char cache_dev_name[BDEVNAME_SIZE];
};
struct gc_stat {
diff --git a/drivers/md/bcache/debug.c b/drivers/md/bcache/debug.c
index 028f7b386e01..4e63c6f6c04d 100644
--- a/drivers/md/bcache/debug.c
+++ b/drivers/md/bcache/debug.c
@@ -106,7 +106,6 @@ void bch_btree_verify(struct btree *b)
void bch_data_verify(struct cached_dev *dc, struct bio *bio)
{
- char name[BDEVNAME_SIZE];
struct bio *check;
struct bio_vec bv, cbv;
struct bvec_iter iter, citer = { 0 };
@@ -134,7 +133,7 @@ void bch_data_verify(struct cached_dev *dc, struct bio *bio)
bv.bv_len),
dc->disk.c,
"verify failed at dev %s sector %llu",
- bdevname(dc->bdev, name),
+ dc->backing_dev_name,
(uint64_t) bio->bi_iter.bi_sector);
kunmap_atomic(p1);
diff --git a/drivers/md/bcache/io.c b/drivers/md/bcache/io.c
index 7fac97ae036e..2ddf8515e6a5 100644
--- a/drivers/md/bcache/io.c
+++ b/drivers/md/bcache/io.c
@@ -52,7 +52,6 @@ void bch_submit_bbio(struct bio *bio, struct cache_set *c,
/* IO errors */
void bch_count_backing_io_errors(struct cached_dev *dc, struct bio *bio)
{
- char buf[BDEVNAME_SIZE];
unsigned errors;
WARN_ONCE(!dc, "NULL pointer of struct cached_dev");
@@ -60,7 +59,7 @@ void bch_count_backing_io_errors(struct cached_dev *dc, struct bio *bio)
errors = atomic_add_return(1, &dc->io_errors);
if (errors < dc->error_limit)
pr_err("%s: IO error on backing device, unrecoverable",
- bio_devname(bio, buf));
+ dc->backing_dev_name);
else
bch_cached_dev_error(dc);
}
@@ -105,19 +104,18 @@ void bch_count_io_errors(struct cache *ca,
}
if (error) {
- char buf[BDEVNAME_SIZE];
unsigned errors = atomic_add_return(1 << IO_ERROR_SHIFT,
&ca->io_errors);
errors >>= IO_ERROR_SHIFT;
if (errors < ca->set->error_limit)
pr_err("%s: IO error on %s%s",
- bdevname(ca->bdev, buf), m,
+ ca->cache_dev_name, m,
is_read ? ", recovering." : ".");
else
bch_cache_set_error(ca->set,
"%s: too many IO errors %s",
- bdevname(ca->bdev, buf), m);
+ ca->cache_dev_name, m);
}
}
diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c
index a65e3365eeb9..8e3e8655ed63 100644
--- a/drivers/md/bcache/request.c
+++ b/drivers/md/bcache/request.c
@@ -649,11 +649,8 @@ static void backing_request_endio(struct bio *bio)
*/
if (unlikely(s->iop.writeback &&
bio->bi_opf & REQ_PREFLUSH)) {
- char buf[BDEVNAME_SIZE];
-
- bio_devname(bio, buf);
pr_err("Can't flush %s: returned bi_status %i",
- buf, bio->bi_status);
+ dc->backing_dev_name, bio->bi_status);
} else {
/* set to orig_bio->bi_status in bio_complete() */
s->iop.status = bio->bi_status;
diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c
index d90d9e59ca00..3dea06b41d43 100644
--- a/drivers/md/bcache/super.c
+++ b/drivers/md/bcache/super.c
@@ -936,7 +936,6 @@ static void cancel_writeback_rate_update_dwork(struct cached_dev *dc)
static void cached_dev_detach_finish(struct work_struct *w)
{
struct cached_dev *dc = container_of(w, struct cached_dev, detach);
- char buf[BDEVNAME_SIZE];
struct closure cl;
closure_init_stack(&cl);
@@ -967,7 +966,7 @@ static void cached_dev_detach_finish(struct work_struct *w)
mutex_unlock(&bch_register_lock);
- pr_info("Caching disabled for %s", bdevname(dc->bdev, buf));
+ pr_info("Caching disabled for %s", dc->backing_dev_name);
/* Drop ref we took in cached_dev_detach() */
closure_put(&dc->disk.cl);
@@ -999,29 +998,28 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c,
{
uint32_t rtime = cpu_to_le32(get_seconds());
struct uuid_entry *u;
- char buf[BDEVNAME_SIZE];
struct cached_dev *exist_dc, *t;
- bdevname(dc->bdev, buf);
-
if ((set_uuid && memcmp(set_uuid, c->sb.set_uuid, 16)) ||
(!set_uuid && memcmp(dc->sb.set_uuid, c->sb.set_uuid, 16)))
return -ENOENT;
if (dc->disk.c) {
- pr_err("Can't attach %s: already attached", buf);
+ pr_err("Can't attach %s: already attached",
+ dc->backing_dev_name);
return -EINVAL;
}
if (test_bit(CACHE_SET_STOPPING, &c->flags)) {
- pr_err("Can't attach %s: shutting down", buf);
+ pr_err("Can't attach %s: shutting down",
+ dc->backing_dev_name);
return -EINVAL;
}
if (dc->sb.block_size < c->sb.block_size) {
/* Will die */
pr_err("Couldn't attach %s: block size less than set's block size",
- buf);
+ dc->backing_dev_name);
return -EINVAL;
}
@@ -1029,7 +1027,7 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c,
list_for_each_entry_safe(exist_dc, t, &c->cached_devs, list) {
if (!memcmp(dc->sb.uuid, exist_dc->sb.uuid, 16)) {
pr_err("Tried to attach %s but duplicate UUID already attached",
- buf);
+ dc->backing_dev_name);
return -EINVAL;
}
@@ -1047,13 +1045,15 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c,
if (!u) {
if (BDEV_STATE(&dc->sb) == BDEV_STATE_DIRTY) {
- pr_err("Couldn't find uuid for %s in set", buf);
+ pr_err("Couldn't find uuid for %s in set",
+ dc->backing_dev_name);
return -ENOENT;
}
u = uuid_find_empty(c);
if (!u) {
- pr_err("Not caching %s, no room for UUID", buf);
+ pr_err("Not caching %s, no room for UUID",
+ dc->backing_dev_name);
return -EINVAL;
}
}
@@ -1112,7 +1112,8 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c,
up_write(&dc->writeback_lock);
pr_info("Caching %s as %s on set %pU",
- bdevname(dc->bdev, buf), dc->disk.disk->disk_name,
+ dc->backing_dev_name,
+ dc->disk.disk->disk_name,
dc->disk.c->sb.set_uuid);
return 0;
}
@@ -1225,10 +1226,10 @@ static void register_bdev(struct cache_sb *sb, struct page *sb_page,
struct block_device *bdev,
struct cached_dev *dc)
{
- char name[BDEVNAME_SIZE];
const char *err = "cannot allocate memory";
struct cache_set *c;
+ bdevname(bdev, dc->backing_dev_name);
memcpy(&dc->sb, sb, sizeof(struct cache_sb));
dc->bdev = bdev;
dc->bdev->bd_holder = dc;
@@ -1237,6 +1238,7 @@ static void register_bdev(struct cache_sb *sb, struct page *sb_page,
bio_first_bvec_all(&dc->sb_bio)->bv_page = sb_page;
get_page(sb_page);
+
if (cached_dev_init(dc, sb->block_size << 9))
goto err;
@@ -1247,7 +1249,7 @@ static void register_bdev(struct cache_sb *sb, struct page *sb_page,
if (bch_cache_accounting_add_kobjs(&dc->accounting, &dc->disk.kobj))
goto err;
- pr_info("registered backing device %s", bdevname(bdev, name));
+ pr_info("registered backing device %s", dc->backing_dev_name);
list_add(&dc->list, &uncached_devices);
list_for_each_entry(c, &bch_cache_sets, list)
@@ -1259,7 +1261,7 @@ static void register_bdev(struct cache_sb *sb, struct page *sb_page,
return;
err:
- pr_notice("error %s: %s", bdevname(bdev, name), err);
+ pr_notice("error %s: %s", dc->backing_dev_name, err);
bcache_device_stop(&dc->disk);
}
@@ -1367,7 +1369,7 @@ int bch_flash_dev_create(struct cache_set *c, uint64_t size)
bool bch_cached_dev_error(struct cached_dev *dc)
{
- char name[BDEVNAME_SIZE];
+ struct cache_set *c;
if (!dc || test_bit(BCACHE_DEV_CLOSING, &dc->disk.flags))
return false;
@@ -1377,7 +1379,22 @@ bool bch_cached_dev_error(struct cached_dev *dc)
smp_mb();
pr_err("stop %s: too many IO errors on backing device %s\n",
- dc->disk.disk->disk_name, bdevname(dc->bdev, name));
+ dc->disk.disk->disk_name, dc->backing_dev_name);
+
+ /*
+ * If the cached device is still attached to a cache set,
+ * even dc->io_disable is true and no more I/O requests
+ * accepted, cache device internal I/O (writeback scan or
+ * garbage collection) may still prevent bcache device from
+ * being stopped. So here CACHE_SET_IO_DISABLE should be
+ * set to c->flags too, to make the internal I/O to cache
+ * device rejected and stopped immediately.
+ * If c is NULL, that means the bcache device is not attached
+ * to any cache set, then no CACHE_SET_IO_DISABLE bit to set.
+ */
+ c = dc->disk.c;
+ if (c && test_and_set_bit(CACHE_SET_IO_DISABLE, &c->flags))
+ pr_info("CACHE_SET_IO_DISABLE already set");
bcache_device_stop(&dc->disk);
return true;
@@ -1395,7 +1412,7 @@ bool bch_cache_set_error(struct cache_set *c, const char *fmt, ...)
return false;
if (test_and_set_bit(CACHE_SET_IO_DISABLE, &c->flags))
- pr_warn("CACHE_SET_IO_DISABLE already set");
+ pr_info("CACHE_SET_IO_DISABLE already set");
/* XXX: we can be called from atomic context
acquire_console_sem();
@@ -1539,6 +1556,20 @@ static void conditional_stop_bcache_device(struct cache_set *c,
*/
pr_warn("stop_when_cache_set_failed of %s is \"auto\" and cache is dirty, stop it to avoid potential data corruption.",
d->disk->disk_name);
+ /*
+ * There might be a small time gap that cache set is
+ * released but bcache device is not. Inside this time
+ * gap, regular I/O requests will directly go into
+ * backing device as no cache set attached to. This
+ * behavior may also introduce potential inconsistence
+ * data in writeback mode while cache is dirty.
+ * Therefore before calling bcache_device_stop() due
+ * to a broken cache device, dc->io_disable should be
+ * explicitly set to true.
+ */
+ dc->io_disable = true;
+ /* make others know io_disable is true earlier */
+ smp_mb();
bcache_device_stop(d);
} else {
/*
@@ -2003,12 +2034,10 @@ static int cache_alloc(struct cache *ca)
static int register_cache(struct cache_sb *sb, struct page *sb_page,
struct block_device *bdev, struct cache *ca)
{
- char name[BDEVNAME_SIZE];
const char *err = NULL; /* must be set for any error case */
int ret = 0;
- bdevname(bdev, name);
-
+ bdevname(bdev, ca->cache_dev_name);
memcpy(&ca->sb, sb, sizeof(struct cache_sb));
ca->bdev = bdev;
ca->bdev->bd_holder = ca;
@@ -2045,14 +2074,14 @@ static int register_cache(struct cache_sb *sb, struct page *sb_page,
goto out;
}
- pr_info("registered cache device %s", name);
+ pr_info("registered cache device %s", ca->cache_dev_name);
out:
kobject_put(&ca->kobj);
err:
if (err)
- pr_notice("error %s: %s", name, err);
+ pr_notice("error %s: %s", ca->cache_dev_name, err);
return ret;
}
diff --git a/drivers/md/bcache/writeback.c b/drivers/md/bcache/writeback.c
index 4a9547cdcdc5..ad45ebe1a74b 100644
--- a/drivers/md/bcache/writeback.c
+++ b/drivers/md/bcache/writeback.c
@@ -244,8 +244,10 @@ static void dirty_endio(struct bio *bio)
struct keybuf_key *w = bio->bi_private;
struct dirty_io *io = w->private;
- if (bio->bi_status)
+ if (bio->bi_status) {
SET_KEY_DIRTY(&w->key, false);
+ bch_count_backing_io_errors(io->dc, bio);
+ }
closure_put(&io->cl);
}
diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
index 9df4f71e58ca..a3771c5729f5 100644
--- a/drivers/nvme/host/core.c
+++ b/drivers/nvme/host/core.c
@@ -764,6 +764,7 @@ static int nvme_submit_user_cmd(struct request_queue *q,
ret = PTR_ERR(meta);
goto out_unmap;
}
+ req->cmd_flags |= REQ_INTEGRITY;
}
}
@@ -2997,31 +2998,7 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
if (nvme_init_ns_head(ns, nsid, id))
goto out_free_id;
nvme_setup_streams_ns(ctrl, ns);
-
-#ifdef CONFIG_NVME_MULTIPATH
- /*
- * If multipathing is enabled we need to always use the subsystem
- * instance number for numbering our devices to avoid conflicts
- * between subsystems that have multiple controllers and thus use
- * the multipath-aware subsystem node and those that have a single
- * controller and use the controller node directly.
- */
- if (ns->head->disk) {
- sprintf(disk_name, "nvme%dc%dn%d", ctrl->subsys->instance,
- ctrl->cntlid, ns->head->instance);
- flags = GENHD_FL_HIDDEN;
- } else {
- sprintf(disk_name, "nvme%dn%d", ctrl->subsys->instance,
- ns->head->instance);
- }
-#else
- /*
- * But without the multipath code enabled, multiple controller per
- * subsystems are visible as devices and thus we cannot use the
- * subsystem instance.
- */
- sprintf(disk_name, "nvme%dn%d", ctrl->instance, ns->head->instance);
-#endif
+ nvme_set_disk_name(disk_name, ns, ctrl, &flags);
if ((ctrl->quirks & NVME_QUIRK_LIGHTNVM) && id->vs[0] == 0x1) {
if (nvme_nvm_register(ns, disk_name, node)) {
diff --git a/drivers/nvme/host/fabrics.c b/drivers/nvme/host/fabrics.c
index 124c458806df..7ae732a77fe8 100644
--- a/drivers/nvme/host/fabrics.c
+++ b/drivers/nvme/host/fabrics.c
@@ -668,6 +668,7 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts,
ret = -ENOMEM;
goto out;
}
+ kfree(opts->transport);
opts->transport = p;
break;
case NVMF_OPT_NQN:
@@ -676,6 +677,7 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts,
ret = -ENOMEM;
goto out;
}
+ kfree(opts->subsysnqn);
opts->subsysnqn = p;
nqnlen = strlen(opts->subsysnqn);
if (nqnlen >= NVMF_NQN_SIZE) {
@@ -698,6 +700,7 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts,
ret = -ENOMEM;
goto out;
}
+ kfree(opts->traddr);
opts->traddr = p;
break;
case NVMF_OPT_TRSVCID:
@@ -706,6 +709,7 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts,
ret = -ENOMEM;
goto out;
}
+ kfree(opts->trsvcid);
opts->trsvcid = p;
break;
case NVMF_OPT_QUEUE_SIZE:
@@ -792,6 +796,7 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts,
ret = -EINVAL;
goto out;
}
+ nvmf_host_put(opts->host);
opts->host = nvmf_host_add(p);
kfree(p);
if (!opts->host) {
@@ -817,6 +822,7 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts,
ret = -ENOMEM;
goto out;
}
+ kfree(opts->host_traddr);
opts->host_traddr = p;
break;
case NVMF_OPT_HOST_ID:
diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c
index 956e0b8e9c4d..d7b664ae5923 100644
--- a/drivers/nvme/host/multipath.c
+++ b/drivers/nvme/host/multipath.c
@@ -15,10 +15,32 @@
#include "nvme.h"
static bool multipath = true;
-module_param(multipath, bool, 0644);
+module_param(multipath, bool, 0444);
MODULE_PARM_DESC(multipath,
"turn on native support for multiple controllers per subsystem");
+/*
+ * If multipathing is enabled we need to always use the subsystem instance
+ * number for numbering our devices to avoid conflicts between subsystems that
+ * have multiple controllers and thus use the multipath-aware subsystem node
+ * and those that have a single controller and use the controller node
+ * directly.
+ */
+void nvme_set_disk_name(char *disk_name, struct nvme_ns *ns,
+ struct nvme_ctrl *ctrl, int *flags)
+{
+ if (!multipath) {
+ sprintf(disk_name, "nvme%dn%d", ctrl->instance, ns->head->instance);
+ } else if (ns->head->disk) {
+ sprintf(disk_name, "nvme%dc%dn%d", ctrl->subsys->instance,
+ ctrl->cntlid, ns->head->instance);
+ *flags = GENHD_FL_HIDDEN;
+ } else {
+ sprintf(disk_name, "nvme%dn%d", ctrl->subsys->instance,
+ ns->head->instance);
+ }
+}
+
void nvme_failover_req(struct request *req)
{
struct nvme_ns *ns = req->q->queuedata;
diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h
index 061fecfd44f5..7ded7a51c430 100644
--- a/drivers/nvme/host/nvme.h
+++ b/drivers/nvme/host/nvme.h
@@ -436,6 +436,8 @@ extern const struct attribute_group nvme_ns_id_attr_group;
extern const struct block_device_operations nvme_ns_head_ops;
#ifdef CONFIG_NVME_MULTIPATH
+void nvme_set_disk_name(char *disk_name, struct nvme_ns *ns,
+ struct nvme_ctrl *ctrl, int *flags);
void nvme_failover_req(struct request *req);
bool nvme_req_needs_failover(struct request *req, blk_status_t error);
void nvme_kick_requeue_lists(struct nvme_ctrl *ctrl);
@@ -461,6 +463,16 @@ static inline void nvme_mpath_check_last_path(struct nvme_ns *ns)
}
#else
+/*
+ * Without the multipath code enabled, multiple controller per subsystems are
+ * visible as devices and thus we cannot use the subsystem instance.
+ */
+static inline void nvme_set_disk_name(char *disk_name, struct nvme_ns *ns,
+ struct nvme_ctrl *ctrl, int *flags)
+{
+ sprintf(disk_name, "nvme%dn%d", ctrl->instance, ns->head->instance);
+}
+
static inline void nvme_failover_req(struct request *req)
{
}
diff --git a/drivers/nvme/target/loop.c b/drivers/nvme/target/loop.c
index 31fdfba556a8..27a8561c0cb9 100644
--- a/drivers/nvme/target/loop.c
+++ b/drivers/nvme/target/loop.c
@@ -469,6 +469,12 @@ static void nvme_loop_reset_ctrl_work(struct work_struct *work)
nvme_stop_ctrl(&ctrl->ctrl);
nvme_loop_shutdown_ctrl(ctrl);
+ if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING)) {
+ /* state change failure should never happen */
+ WARN_ON_ONCE(1);
+ return;
+ }
+
ret = nvme_loop_configure_admin_queue(ctrl);
if (ret)
goto out_disable;