All of lore.kernel.org
 help / color / mirror / Atom feed
From: Taehee Yoo <ap420073@gmail.com>
To: linux-nvme@lists.infradead.org, kbusch@kernel.org, axboe@fb.com,
	hch@lst.de, sagi@grimberg.me, kch@nvidia.com
Cc: james.p.freyensee@intel.com, ming.l@ssi.samsung.com,
	larrystevenwise@gmail.com, anthony.j.knapp@intel.com,
	pizhenwei@bytedance.com, ap420073@gmail.com
Subject: [PATCH 3/4] nvmet: fix hang in nvmet_ns_disable()
Date: Tue,  3 Jan 2023 10:03:56 +0000	[thread overview]
Message-ID: <20230103100357.875854-4-ap420073@gmail.com> (raw)
In-Reply-To: <20230103100357.875854-1-ap420073@gmail.com>

nvme target namespace is enabled or disabled by nvmet_ns_enable() or
nvmet_ns_disable().
The subsys->lock is used to disallow to use namespace data while
nvmet_ns_enable() or nvmet_ns_disable() are working.
The ns->enabled boolean variable prevents using namespace data in wrong
state such as uninitialized state.

nvmet_ns_disable() acquires ns->lock and set ns->enabled false.
Then, it releases ns->lock for a while to wait ns->disable_done completion.
At this point, nvmet_ns_enable() can be worked concurrently and it calls
percpu_ref_init().
So, ns->disable_done will never be completed.
Therefore hang would occur at this point.

   CPU0                                     CPU1
   nvmet_ns_disable();
   mutex_lock(&subsys->lock);               nvmet_ns_enable();
                                            mutex_lock(&subsys->lock);
   ns->enabled = false;
   mutex_unlock(&subsys->lock);
                                            percpu_ref_init();
   wait_for_completion(&ns->disable_done);  <-- infinite wait

   mutex_lock(&subsys->lock);
   mutex_unlock(&subsys->lock);

INFO: task bash:926 blocked for more than 30 seconds.
      Tainted: G        W          6.1.0+ #17
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
task:bash            state:D stack:27200 pid:926   ppid:911
flags:0x00004000
Call Trace:
 <TASK>
 __schedule+0xafc/0x2930
 ? io_schedule_timeout+0x160/0x160
 ? _raw_spin_unlock_irq+0x24/0x50
 ? __wait_for_common+0x39b/0x5c0
 ? usleep_range_state+0x190/0x190
 schedule+0x130/0x230
 schedule_timeout+0x18a/0x240
 ? usleep_range_state+0x190/0x190
 ? rcu_read_lock_sched_held+0x12/0x80
 ? lock_downgrade+0x700/0x700
 ? do_raw_spin_trylock+0xb5/0x180
 ? lock_contended+0xdf0/0xdf0
 ? _raw_spin_unlock_irq+0x24/0x50
 ? trace_hardirqs_on+0x3c/0x190
 __wait_for_common+0x1ca/0x5c0
 ? usleep_range_state+0x190/0x190
 ? bit_wait_io+0xf0/0xf0
 ? _raw_spin_unlock_irqrestore+0x59/0x70
 nvmet_ns_disable+0x288/0x490
 ? nvmet_ns_enable+0x970/0x970
 ? lockdep_hardirqs_on_prepare+0x410/0x410
 ? rcu_read_lock_sched_held+0x12/0x80
 ? configfs_write_iter+0x1df/0x480
 ? nvmet_ns_revalidate_size_store+0x220/0x220
 nvmet_ns_enable_store+0x85/0xe0
[ ... ]

Fixes: a07b4970f464 ("nvmet: add a generic NVMe target")
Signed-off-by: Taehee Yoo <ap420073@gmail.com>
---
 drivers/nvme/target/configfs.c | 14 +++++++-------
 drivers/nvme/target/core.c     | 10 ++++++----
 drivers/nvme/target/nvmet.h    |  8 +++++++-
 3 files changed, 20 insertions(+), 12 deletions(-)

diff --git a/drivers/nvme/target/configfs.c b/drivers/nvme/target/configfs.c
index 907143870da5..d878c4231d65 100644
--- a/drivers/nvme/target/configfs.c
+++ b/drivers/nvme/target/configfs.c
@@ -348,7 +348,7 @@ static ssize_t nvmet_ns_device_path_store(struct config_item *item,
 
 	mutex_lock(&subsys->lock);
 	ret = -EBUSY;
-	if (ns->enabled)
+	if (ns->state != NVMET_NS_DISABLED)
 		goto out_unlock;
 
 	ret = -EINVAL;
@@ -390,7 +390,7 @@ static ssize_t nvmet_ns_p2pmem_store(struct config_item *item,
 	int error;
 
 	mutex_lock(&ns->subsys->lock);
-	if (ns->enabled) {
+	if (ns->state != NVMET_NS_DISABLED) {
 		ret = -EBUSY;
 		goto out_unlock;
 	}
@@ -427,7 +427,7 @@ static ssize_t nvmet_ns_device_uuid_store(struct config_item *item,
 	int ret = 0;
 
 	mutex_lock(&subsys->lock);
-	if (ns->enabled) {
+	if (ns->state != NVMET_NS_DISABLED) {
 		ret = -EBUSY;
 		goto out_unlock;
 	}
@@ -458,7 +458,7 @@ static ssize_t nvmet_ns_device_nguid_store(struct config_item *item,
 	int ret = 0;
 
 	mutex_lock(&subsys->lock);
-	if (ns->enabled) {
+	if (ns->state != NVMET_NS_DISABLED) {
 		ret = -EBUSY;
 		goto out_unlock;
 	}
@@ -523,7 +523,7 @@ CONFIGFS_ATTR(nvmet_ns_, ana_grpid);
 
 static ssize_t nvmet_ns_enable_show(struct config_item *item, char *page)
 {
-	return sprintf(page, "%d\n", to_nvmet_ns(item)->enabled);
+	return sprintf(page, "%d\n", !!to_nvmet_ns(item)->state);
 }
 
 static ssize_t nvmet_ns_enable_store(struct config_item *item,
@@ -561,7 +561,7 @@ static ssize_t nvmet_ns_buffered_io_store(struct config_item *item,
 		return -EINVAL;
 
 	mutex_lock(&ns->subsys->lock);
-	if (ns->enabled) {
+	if (ns->state != NVMET_NS_DISABLED) {
 		pr_err("disable ns before setting buffered_io value.\n");
 		mutex_unlock(&ns->subsys->lock);
 		return -EINVAL;
@@ -587,7 +587,7 @@ static ssize_t nvmet_ns_revalidate_size_store(struct config_item *item,
 		return -EINVAL;
 
 	mutex_lock(&ns->subsys->lock);
-	if (!ns->enabled) {
+	if (ns->state != NVMET_NS_ENABLED) {
 		pr_err("enable ns before revalidate.\n");
 		mutex_unlock(&ns->subsys->lock);
 		return -EINVAL;
diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c
index f66ed13d7c11..58a91fb9c2f7 100644
--- a/drivers/nvme/target/core.c
+++ b/drivers/nvme/target/core.c
@@ -563,7 +563,7 @@ int nvmet_ns_enable(struct nvmet_ns *ns)
 		goto out_unlock;
 	}
 
-	if (ns->enabled)
+	if (ns->state != NVMET_NS_DISABLED)
 		goto out_unlock;
 
 	ret = -EMFILE;
@@ -598,7 +598,7 @@ int nvmet_ns_enable(struct nvmet_ns *ns)
 	subsys->nr_namespaces++;
 
 	nvmet_ns_changed(subsys, ns->nsid);
-	ns->enabled = true;
+	ns->state = NVMET_NS_ENABLED;
 	ret = 0;
 out_unlock:
 	mutex_unlock(&subsys->lock);
@@ -621,10 +621,10 @@ void nvmet_ns_disable(struct nvmet_ns *ns)
 	struct nvmet_ctrl *ctrl;
 
 	mutex_lock(&subsys->lock);
-	if (!ns->enabled)
+	if (ns->state != NVMET_NS_ENABLED)
 		goto out_unlock;
 
-	ns->enabled = false;
+	ns->state = NVMET_NS_DISABLING;
 	xa_erase(&ns->subsys->namespaces, ns->nsid);
 	if (ns->nsid == subsys->max_nsid)
 		subsys->max_nsid = nvmet_max_nsid(subsys);
@@ -652,6 +652,7 @@ void nvmet_ns_disable(struct nvmet_ns *ns)
 	subsys->nr_namespaces--;
 	nvmet_ns_changed(subsys, ns->nsid);
 	nvmet_ns_dev_disable(ns);
+	ns->state = NVMET_NS_DISABLED;
 out_unlock:
 	mutex_unlock(&subsys->lock);
 }
@@ -689,6 +690,7 @@ struct nvmet_ns *nvmet_ns_alloc(struct nvmet_subsys *subsys, u32 nsid)
 	uuid_gen(&ns->uuid);
 	ns->buffered_io = false;
 	ns->csi = NVME_CSI_NVM;
+	ns->state = NVMET_NS_DISABLED;
 
 	return ns;
 }
diff --git a/drivers/nvme/target/nvmet.h b/drivers/nvme/target/nvmet.h
index 89bedfcd974c..e609787577c6 100644
--- a/drivers/nvme/target/nvmet.h
+++ b/drivers/nvme/target/nvmet.h
@@ -56,6 +56,12 @@
 #define IPO_IATTR_CONNECT_SQE(x)	\
 	(cpu_to_le32(offsetof(struct nvmf_connect_command, x)))
 
+enum nvmet_ns_state {
+	NVMET_NS_ENABLED,
+	NVMET_NS_DISABLING,
+	NVMET_NS_DISABLED
+};
+
 struct nvmet_ns {
 	struct percpu_ref	ref;
 	struct block_device	*bdev;
@@ -69,7 +75,7 @@ struct nvmet_ns {
 	u32			anagrpid;
 
 	bool			buffered_io;
-	bool			enabled;
+	enum nvmet_ns_state	state;
 	struct nvmet_subsys	*subsys;
 	const char		*device_path;
 
-- 
2.34.1



  parent reply	other threads:[~2023-01-03 10:15 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-01-03 10:03 [PATCH 0/4] nvme: fix several bugs in nvme-fabric Taehee Yoo
2023-01-03 10:03 ` [PATCH 1/4] nvme: fix delete uninitialized controller Taehee Yoo
2023-01-03 10:30   ` Sagi Grimberg
2023-01-04  0:24     ` Chaitanya Kulkarni
2023-01-04  2:42       ` Taehee Yoo
2023-01-03 10:03 ` [PATCH 2/4] nvme: fix reset " Taehee Yoo
2023-01-03 10:32   ` Sagi Grimberg
2023-01-03 10:03 ` Taehee Yoo [this message]
2023-01-03 10:58   ` [PATCH 3/4] nvmet: fix hang in nvmet_ns_disable() Sagi Grimberg
2023-01-04  0:32   ` Chaitanya Kulkarni
2023-01-04  8:56     ` Taehee Yoo
2023-01-03 10:03 ` [PATCH 4/4] nvmet-tcp: fix memory leak in nvmet_tcp_free_cmd_data_in_buffers() Taehee Yoo
2023-01-03 10:54   ` Sagi Grimberg
2023-01-04  8:44     ` Taehee Yoo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20230103100357.875854-4-ap420073@gmail.com \
    --to=ap420073@gmail.com \
    --cc=anthony.j.knapp@intel.com \
    --cc=axboe@fb.com \
    --cc=hch@lst.de \
    --cc=james.p.freyensee@intel.com \
    --cc=kbusch@kernel.org \
    --cc=kch@nvidia.com \
    --cc=larrystevenwise@gmail.com \
    --cc=linux-nvme@lists.infradead.org \
    --cc=ming.l@ssi.samsung.com \
    --cc=pizhenwei@bytedance.com \
    --cc=sagi@grimberg.me \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.