Merge branch 'for-linus' of git://git.kernel.dk/linux-block
Pull NVMe fix from Jens Axboe:
"Late addition here, it's basically a revert of a patch that was added
in this merge window, but has proven to cause problems.
This is swapping out the RCU based namespace protection with a good
old mutex instead"
* 'for-linus' of git://git.kernel.dk/linux-block:
nvme: Remove RCU namespace protection
diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
index 1a51584..d5fb55c 100644
--- a/drivers/nvme/host/core.c
+++ b/drivers/nvme/host/core.c
@@ -1394,19 +1394,22 @@
return nsa->ns_id - nsb->ns_id;
}
-static struct nvme_ns *nvme_find_ns(struct nvme_ctrl *ctrl, unsigned nsid)
+static struct nvme_ns *nvme_find_get_ns(struct nvme_ctrl *ctrl, unsigned nsid)
{
- struct nvme_ns *ns;
+ struct nvme_ns *ns, *ret = NULL;
- lockdep_assert_held(&ctrl->namespaces_mutex);
-
+ mutex_lock(&ctrl->namespaces_mutex);
list_for_each_entry(ns, &ctrl->namespaces, list) {
- if (ns->ns_id == nsid)
- return ns;
+ if (ns->ns_id == nsid) {
+ kref_get(&ns->kref);
+ ret = ns;
+ break;
+ }
if (ns->ns_id > nsid)
break;
}
- return NULL;
+ mutex_unlock(&ctrl->namespaces_mutex);
+ return ret;
}
static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
@@ -1415,8 +1418,6 @@
struct gendisk *disk;
int node = dev_to_node(ctrl->dev);
- lockdep_assert_held(&ctrl->namespaces_mutex);
-
ns = kzalloc_node(sizeof(*ns), GFP_KERNEL, node);
if (!ns)
return;
@@ -1457,7 +1458,10 @@
if (nvme_revalidate_disk(ns->disk))
goto out_free_disk;
- list_add_tail_rcu(&ns->list, &ctrl->namespaces);
+ mutex_lock(&ctrl->namespaces_mutex);
+ list_add_tail(&ns->list, &ctrl->namespaces);
+ mutex_unlock(&ctrl->namespaces_mutex);
+
kref_get(&ctrl->kref);
if (ns->type == NVME_NS_LIGHTNVM)
return;
@@ -1480,8 +1484,6 @@
static void nvme_ns_remove(struct nvme_ns *ns)
{
- lockdep_assert_held(&ns->ctrl->namespaces_mutex);
-
if (test_and_set_bit(NVME_NS_REMOVING, &ns->flags))
return;
@@ -1494,8 +1496,11 @@
blk_mq_abort_requeue_list(ns->queue);
blk_cleanup_queue(ns->queue);
}
+
+ mutex_lock(&ns->ctrl->namespaces_mutex);
list_del_init(&ns->list);
- synchronize_rcu();
+ mutex_unlock(&ns->ctrl->namespaces_mutex);
+
nvme_put_ns(ns);
}
@@ -1503,10 +1508,11 @@
{
struct nvme_ns *ns;
- ns = nvme_find_ns(ctrl, nsid);
+ ns = nvme_find_get_ns(ctrl, nsid);
if (ns) {
if (revalidate_disk(ns->disk))
nvme_ns_remove(ns);
+ nvme_put_ns(ns);
} else
nvme_alloc_ns(ctrl, nsid);
}
@@ -1535,9 +1541,11 @@
nvme_validate_ns(ctrl, nsid);
while (++prev < nsid) {
- ns = nvme_find_ns(ctrl, prev);
- if (ns)
+ ns = nvme_find_get_ns(ctrl, prev);
+ if (ns) {
nvme_ns_remove(ns);
+ nvme_put_ns(ns);
+ }
}
}
nn -= j;
@@ -1552,8 +1560,6 @@
struct nvme_ns *ns, *next;
unsigned i;
- lockdep_assert_held(&ctrl->namespaces_mutex);
-
for (i = 1; i <= nn; i++)
nvme_validate_ns(ctrl, i);
@@ -1576,7 +1582,6 @@
if (nvme_identify_ctrl(ctrl, &id))
return;
- mutex_lock(&ctrl->namespaces_mutex);
nn = le32_to_cpu(id->nn);
if (ctrl->vs >= NVME_VS(1, 1) &&
!(ctrl->quirks & NVME_QUIRK_IDENTIFY_CNS)) {
@@ -1585,6 +1590,7 @@
}
nvme_scan_ns_sequential(ctrl, nn);
done:
+ mutex_lock(&ctrl->namespaces_mutex);
list_sort(NULL, &ctrl->namespaces, ns_cmp);
mutex_unlock(&ctrl->namespaces_mutex);
kfree(id);
@@ -1604,6 +1610,11 @@
}
EXPORT_SYMBOL_GPL(nvme_queue_scan);
+/*
+ * This function iterates the namespace list unlocked to allow recovery from
+ * controller failure. It is up to the caller to ensure the namespace list is
+ * not modified by scan work while this function is executing.
+ */
void nvme_remove_namespaces(struct nvme_ctrl *ctrl)
{
struct nvme_ns *ns, *next;
@@ -1617,10 +1628,8 @@
if (ctrl->state == NVME_CTRL_DEAD)
nvme_kill_queues(ctrl);
- mutex_lock(&ctrl->namespaces_mutex);
list_for_each_entry_safe(ns, next, &ctrl->namespaces, list)
nvme_ns_remove(ns);
- mutex_unlock(&ctrl->namespaces_mutex);
}
EXPORT_SYMBOL_GPL(nvme_remove_namespaces);
@@ -1791,11 +1800,8 @@
{
struct nvme_ns *ns;
- rcu_read_lock();
- list_for_each_entry_rcu(ns, &ctrl->namespaces, list) {
- if (!kref_get_unless_zero(&ns->kref))
- continue;
-
+ mutex_lock(&ctrl->namespaces_mutex);
+ list_for_each_entry(ns, &ctrl->namespaces, list) {
/*
* Revalidating a dead namespace sets capacity to 0. This will
* end buffered writers dirtying pages that can't be synced.
@@ -1806,10 +1812,8 @@
blk_set_queue_dying(ns->queue);
blk_mq_abort_requeue_list(ns->queue);
blk_mq_start_stopped_hw_queues(ns->queue, true);
-
- nvme_put_ns(ns);
}
- rcu_read_unlock();
+ mutex_unlock(&ctrl->namespaces_mutex);
}
EXPORT_SYMBOL_GPL(nvme_kill_queues);
@@ -1817,8 +1821,8 @@
{
struct nvme_ns *ns;
- rcu_read_lock();
- list_for_each_entry_rcu(ns, &ctrl->namespaces, list) {
+ mutex_lock(&ctrl->namespaces_mutex);
+ list_for_each_entry(ns, &ctrl->namespaces, list) {
spin_lock_irq(ns->queue->queue_lock);
queue_flag_set(QUEUE_FLAG_STOPPED, ns->queue);
spin_unlock_irq(ns->queue->queue_lock);
@@ -1826,7 +1830,7 @@
blk_mq_cancel_requeue_work(ns->queue);
blk_mq_stop_hw_queues(ns->queue);
}
- rcu_read_unlock();
+ mutex_unlock(&ctrl->namespaces_mutex);
}
EXPORT_SYMBOL_GPL(nvme_stop_queues);
@@ -1834,13 +1838,13 @@
{
struct nvme_ns *ns;
- rcu_read_lock();
- list_for_each_entry_rcu(ns, &ctrl->namespaces, list) {
+ mutex_lock(&ctrl->namespaces_mutex);
+ list_for_each_entry(ns, &ctrl->namespaces, list) {
queue_flag_clear_unlocked(QUEUE_FLAG_STOPPED, ns->queue);
blk_mq_start_stopped_hw_queues(ns->queue, true);
blk_mq_kick_requeue_list(ns->queue);
}
- rcu_read_unlock();
+ mutex_unlock(&ctrl->namespaces_mutex);
}
EXPORT_SYMBOL_GPL(nvme_start_queues);
