summaryrefslogtreecommitdiff
path: root/drivers/nvme/target
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/nvme/target')
-rw-r--r--drivers/nvme/target/Kconfig3
-rw-r--r--drivers/nvme/target/admin-cmd.c33
-rw-r--r--drivers/nvme/target/auth.c92
-rw-r--r--drivers/nvme/target/core.c155
-rw-r--r--drivers/nvme/target/debugfs.c29
-rw-r--r--drivers/nvme/target/discovery.c2
-rw-r--r--drivers/nvme/target/fabrics-cmd-auth.c60
-rw-r--r--drivers/nvme/target/fabrics-cmd.c33
-rw-r--r--drivers/nvme/target/fc.c172
-rw-r--r--drivers/nvme/target/fcloop.c497
-rw-r--r--drivers/nvme/target/io-cmd-bdev.c11
-rw-r--r--drivers/nvme/target/loop.c31
-rw-r--r--drivers/nvme/target/nvmet.h79
-rw-r--r--drivers/nvme/target/passthru.c2
-rw-r--r--drivers/nvme/target/pci-epf.c281
-rw-r--r--drivers/nvme/target/rdma.c43
-rw-r--r--drivers/nvme/target/tcp.c152
17 files changed, 1070 insertions, 605 deletions
diff --git a/drivers/nvme/target/Kconfig b/drivers/nvme/target/Kconfig
index fb7446d6d682..4904097dfd49 100644
--- a/drivers/nvme/target/Kconfig
+++ b/drivers/nvme/target/Kconfig
@@ -3,7 +3,7 @@
config NVME_TARGET
tristate "NVMe Target support"
depends on BLOCK
- depends on CONFIGFS_FS
+ select CONFIGFS_FS
select NVME_KEYRING if NVME_TARGET_TCP_TLS
select KEYS if NVME_TARGET_TCP_TLS
select SGL_ALLOC
@@ -98,6 +98,7 @@ config NVME_TARGET_TCP_TLS
bool "NVMe over Fabrics TCP target TLS encryption support"
depends on NVME_TARGET_TCP
select NET_HANDSHAKE
+ select TLS
help
Enables TLS encryption for the NVMe TCP target using the netlink handshake API.
diff --git a/drivers/nvme/target/admin-cmd.c b/drivers/nvme/target/admin-cmd.c
index acc138bbf8f2..3e378153a781 100644
--- a/drivers/nvme/target/admin-cmd.c
+++ b/drivers/nvme/target/admin-cmd.c
@@ -63,14 +63,9 @@ static void nvmet_execute_create_sq(struct nvmet_req *req)
if (status != NVME_SC_SUCCESS)
goto complete;
- /*
- * Note: The NVMe specification allows multiple SQs to use the same CQ.
- * However, the target code does not really support that. So for now,
- * prevent this and fail the command if sqid and cqid are different.
- */
- if (!cqid || cqid != sqid) {
- pr_err("SQ %u: Unsupported CQID %u\n", sqid, cqid);
- status = NVME_SC_CQ_INVALID | NVME_STATUS_DNR;
+ status = nvmet_check_io_cqid(ctrl, cqid, false);
+ if (status != NVME_SC_SUCCESS) {
+ pr_err("SQ %u: Invalid CQID %u\n", sqid, cqid);
goto complete;
}
@@ -79,7 +74,7 @@ static void nvmet_execute_create_sq(struct nvmet_req *req)
goto complete;
}
- status = ctrl->ops->create_sq(ctrl, sqid, sq_flags, qsize, prp1);
+ status = ctrl->ops->create_sq(ctrl, sqid, cqid, sq_flags, qsize, prp1);
complete:
nvmet_req_complete(req, status);
@@ -96,14 +91,15 @@ static void nvmet_execute_delete_cq(struct nvmet_req *req)
goto complete;
}
- if (!cqid) {
- status = NVME_SC_QID_INVALID | NVME_STATUS_DNR;
+ status = nvmet_check_io_cqid(ctrl, cqid, false);
+ if (status != NVME_SC_SUCCESS)
goto complete;
- }
- status = nvmet_check_cqid(ctrl, cqid);
- if (status != NVME_SC_SUCCESS)
+ if (!ctrl->cqs[cqid] || nvmet_cq_in_use(ctrl->cqs[cqid])) {
+ /* Some SQs are still using this CQ */
+ status = NVME_SC_QID_INVALID | NVME_STATUS_DNR;
goto complete;
+ }
status = ctrl->ops->delete_cq(ctrl, cqid);
@@ -127,12 +123,7 @@ static void nvmet_execute_create_cq(struct nvmet_req *req)
goto complete;
}
- if (!cqid) {
- status = NVME_SC_QID_INVALID | NVME_STATUS_DNR;
- goto complete;
- }
-
- status = nvmet_check_cqid(ctrl, cqid);
+ status = nvmet_check_io_cqid(ctrl, cqid, true);
if (status != NVME_SC_SUCCESS)
goto complete;
@@ -1174,7 +1165,7 @@ static void nvmet_execute_identify(struct nvmet_req *req)
* A "minimum viable" abort implementation: the command is mandatory in the
* spec, but we are not required to do any useful work. We couldn't really
* do a useful abort, so don't bother even with waiting for the command
- * to be exectuted and return immediately telling the command to abort
+ * to be executed and return immediately telling the command to abort
* wasn't found.
*/
static void nvmet_execute_abort(struct nvmet_req *req)
diff --git a/drivers/nvme/target/auth.c b/drivers/nvme/target/auth.c
index b47d675232d2..b340380f3892 100644
--- a/drivers/nvme/target/auth.c
+++ b/drivers/nvme/target/auth.c
@@ -15,6 +15,7 @@
#include <linux/ctype.h>
#include <linux/random.h>
#include <linux/nvme-auth.h>
+#include <linux/nvme-keyring.h>
#include <linux/unaligned.h>
#include "nvmet.h"
@@ -139,7 +140,7 @@ int nvmet_setup_dhgroup(struct nvmet_ctrl *ctrl, u8 dhgroup_id)
return ret;
}
-u8 nvmet_setup_auth(struct nvmet_ctrl *ctrl)
+u8 nvmet_setup_auth(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq)
{
int ret = 0;
struct nvmet_host_link *p;
@@ -165,6 +166,11 @@ u8 nvmet_setup_auth(struct nvmet_ctrl *ctrl)
goto out_unlock;
}
+ if (nvmet_queue_tls_keyid(sq)) {
+ pr_debug("host %s tls enabled\n", ctrl->hostnqn);
+ goto out_unlock;
+ }
+
ret = nvmet_setup_dhgroup(ctrl, host->dhchap_dhgroup_id);
if (ret < 0) {
pr_warn("Failed to setup DH group");
@@ -233,6 +239,9 @@ out_unlock:
void nvmet_auth_sq_free(struct nvmet_sq *sq)
{
cancel_delayed_work(&sq->auth_expired_work);
+#ifdef CONFIG_NVME_TARGET_TCP_TLS
+ sq->tls_key = NULL;
+#endif
kfree(sq->dhchap_c1);
sq->dhchap_c1 = NULL;
kfree(sq->dhchap_c2);
@@ -261,13 +270,22 @@ void nvmet_destroy_auth(struct nvmet_ctrl *ctrl)
nvme_auth_free_key(ctrl->ctrl_key);
ctrl->ctrl_key = NULL;
}
+#ifdef CONFIG_NVME_TARGET_TCP_TLS
+ if (ctrl->tls_key) {
+ key_put(ctrl->tls_key);
+ ctrl->tls_key = NULL;
+ }
+#endif
}
bool nvmet_check_auth_status(struct nvmet_req *req)
{
- if (req->sq->ctrl->host_key &&
- !req->sq->authenticated)
- return false;
+ if (req->sq->ctrl->host_key) {
+ if (req->sq->qid > 0)
+ return true;
+ if (!req->sq->authenticated)
+ return false;
+ }
return true;
}
@@ -275,7 +293,7 @@ int nvmet_auth_host_hash(struct nvmet_req *req, u8 *response,
unsigned int shash_len)
{
struct crypto_shash *shash_tfm;
- struct shash_desc *shash;
+ SHASH_DESC_ON_STACK(shash, shash_tfm);
struct nvmet_ctrl *ctrl = req->sq->ctrl;
const char *hash_name;
u8 *challenge = req->sq->dhchap_c1;
@@ -327,19 +345,13 @@ int nvmet_auth_host_hash(struct nvmet_req *req, u8 *response,
req->sq->dhchap_c1,
challenge, shash_len);
if (ret)
- goto out_free_challenge;
+ goto out;
}
pr_debug("ctrl %d qid %d host response seq %u transaction %d\n",
ctrl->cntlid, req->sq->qid, req->sq->dhchap_s1,
req->sq->dhchap_tid);
- shash = kzalloc(sizeof(*shash) + crypto_shash_descsize(shash_tfm),
- GFP_KERNEL);
- if (!shash) {
- ret = -ENOMEM;
- goto out_free_challenge;
- }
shash->tfm = shash_tfm;
ret = crypto_shash_init(shash);
if (ret)
@@ -374,8 +386,6 @@ int nvmet_auth_host_hash(struct nvmet_req *req, u8 *response,
goto out;
ret = crypto_shash_final(shash, response);
out:
- kfree(shash);
-out_free_challenge:
if (challenge != req->sq->dhchap_c1)
kfree(challenge);
out_free_response:
@@ -542,3 +552,57 @@ int nvmet_auth_ctrl_sesskey(struct nvmet_req *req,
return ret;
}
+
+void nvmet_auth_insert_psk(struct nvmet_sq *sq)
+{
+ int hash_len = nvme_auth_hmac_hash_len(sq->ctrl->shash_id);
+ u8 *psk, *digest, *tls_psk;
+ size_t psk_len;
+ int ret;
+#ifdef CONFIG_NVME_TARGET_TCP_TLS
+ struct key *tls_key = NULL;
+#endif
+
+ ret = nvme_auth_generate_psk(sq->ctrl->shash_id,
+ sq->dhchap_skey,
+ sq->dhchap_skey_len,
+ sq->dhchap_c1, sq->dhchap_c2,
+ hash_len, &psk, &psk_len);
+ if (ret) {
+ pr_warn("%s: ctrl %d qid %d failed to generate PSK, error %d\n",
+ __func__, sq->ctrl->cntlid, sq->qid, ret);
+ return;
+ }
+ ret = nvme_auth_generate_digest(sq->ctrl->shash_id, psk, psk_len,
+ sq->ctrl->subsysnqn,
+ sq->ctrl->hostnqn, &digest);
+ if (ret) {
+ pr_warn("%s: ctrl %d qid %d failed to generate digest, error %d\n",
+ __func__, sq->ctrl->cntlid, sq->qid, ret);
+ goto out_free_psk;
+ }
+ ret = nvme_auth_derive_tls_psk(sq->ctrl->shash_id, psk, psk_len,
+ digest, &tls_psk);
+ if (ret) {
+ pr_warn("%s: ctrl %d qid %d failed to derive TLS PSK, error %d\n",
+ __func__, sq->ctrl->cntlid, sq->qid, ret);
+ goto out_free_digest;
+ }
+#ifdef CONFIG_NVME_TARGET_TCP_TLS
+ tls_key = nvme_tls_psk_refresh(NULL, sq->ctrl->hostnqn, sq->ctrl->subsysnqn,
+ sq->ctrl->shash_id, tls_psk, psk_len, digest);
+ if (IS_ERR(tls_key)) {
+ pr_warn("%s: ctrl %d qid %d failed to refresh key, error %ld\n",
+ __func__, sq->ctrl->cntlid, sq->qid, PTR_ERR(tls_key));
+ tls_key = NULL;
+ }
+ if (sq->ctrl->tls_key)
+ key_put(sq->ctrl->tls_key);
+ sq->ctrl->tls_key = tls_key;
+#endif
+ kfree_sensitive(tls_psk);
+out_free_digest:
+ kfree_sensitive(digest);
+out_free_psk:
+ kfree_sensitive(psk);
+}
diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c
index cdc4a09a6e8a..175c5b6d4dd5 100644
--- a/drivers/nvme/target/core.c
+++ b/drivers/nvme/target/core.c
@@ -62,14 +62,7 @@ inline u16 errno_to_nvme_status(struct nvmet_req *req, int errno)
return NVME_SC_LBA_RANGE | NVME_STATUS_DNR;
case -EOPNOTSUPP:
req->error_loc = offsetof(struct nvme_common_command, opcode);
- switch (req->cmd->common.opcode) {
- case nvme_cmd_dsm:
- case nvme_cmd_write_zeroes:
- return NVME_SC_ONCS_NOT_SUPPORTED | NVME_STATUS_DNR;
- default:
- return NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR;
- }
- break;
+ return NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR;
case -ENODATA:
req->error_loc = offsetof(struct nvme_rw_command, nsid);
return NVME_SC_ACCESS_DENIED;
@@ -324,6 +317,9 @@ int nvmet_enable_port(struct nvmet_port *port)
lockdep_assert_held(&nvmet_config_sem);
+ if (port->disc_addr.trtype == NVMF_TRTYPE_MAX)
+ return -EINVAL;
+
ops = nvmet_transports[port->disc_addr.trtype];
if (!ops) {
up_write(&nvmet_config_sem);
@@ -606,6 +602,9 @@ int nvmet_ns_enable(struct nvmet_ns *ns)
goto out_dev_put;
}
+ if (percpu_ref_init(&ns->ref, nvmet_destroy_namespace, 0, GFP_KERNEL))
+ goto out_pr_exit;
+
nvmet_ns_changed(subsys, ns->nsid);
ns->enabled = true;
xa_set_mark(&subsys->namespaces, ns->nsid, NVMET_NS_ENABLED);
@@ -613,6 +612,9 @@ int nvmet_ns_enable(struct nvmet_ns *ns)
out_unlock:
mutex_unlock(&subsys->lock);
return ret;
+out_pr_exit:
+ if (ns->pr.enable)
+ nvmet_pr_exit_ns(ns);
out_dev_put:
list_for_each_entry(ctrl, &subsys->ctrls, subsys_entry)
pci_dev_put(radix_tree_delete(&ctrl->p2p_ns_map, ns->nsid));
@@ -638,6 +640,19 @@ void nvmet_ns_disable(struct nvmet_ns *ns)
mutex_unlock(&subsys->lock);
+ /*
+ * Now that we removed the namespaces from the lookup list, we
+ * can kill the per_cpu ref and wait for any remaining references
+ * to be dropped, as well as a RCU grace period for anyone only
+ * using the namespace under rcu_read_lock(). Note that we can't
+ * use call_rcu here as we need to ensure the namespaces have
+ * been fully destroyed before unloading the module.
+ */
+ percpu_ref_kill(&ns->ref);
+ synchronize_rcu();
+ wait_for_completion(&ns->disable_done);
+ percpu_ref_exit(&ns->ref);
+
if (ns->pr.enable)
nvmet_pr_exit_ns(ns);
@@ -660,22 +675,6 @@ void nvmet_ns_free(struct nvmet_ns *ns)
if (ns->nsid == subsys->max_nsid)
subsys->max_nsid = nvmet_max_nsid(subsys);
- mutex_unlock(&subsys->lock);
-
- /*
- * Now that we removed the namespaces from the lookup list, we
- * can kill the per_cpu ref and wait for any remaining references
- * to be dropped, as well as a RCU grace period for anyone only
- * using the namepace under rcu_read_lock(). Note that we can't
- * use call_rcu here as we need to ensure the namespaces have
- * been fully destroyed before unloading the module.
- */
- percpu_ref_kill(&ns->ref);
- synchronize_rcu();
- wait_for_completion(&ns->disable_done);
- percpu_ref_exit(&ns->ref);
-
- mutex_lock(&subsys->lock);
subsys->nr_namespaces--;
mutex_unlock(&subsys->lock);
@@ -705,9 +704,6 @@ struct nvmet_ns *nvmet_ns_alloc(struct nvmet_subsys *subsys, u32 nsid)
ns->nsid = nsid;
ns->subsys = subsys;
- if (percpu_ref_init(&ns->ref, nvmet_destroy_namespace, 0, GFP_KERNEL))
- goto out_free;
-
if (ns->nsid > subsys->max_nsid)
subsys->max_nsid = nsid;
@@ -730,8 +726,6 @@ struct nvmet_ns *nvmet_ns_alloc(struct nvmet_subsys *subsys, u32 nsid)
return ns;
out_exit:
subsys->max_nsid = nvmet_max_nsid(subsys);
- percpu_ref_exit(&ns->ref);
-out_free:
kfree(ns);
out_unlock:
mutex_unlock(&subsys->lock);
@@ -812,11 +806,43 @@ void nvmet_req_complete(struct nvmet_req *req, u16 status)
}
EXPORT_SYMBOL_GPL(nvmet_req_complete);
+void nvmet_cq_init(struct nvmet_cq *cq)
+{
+ refcount_set(&cq->ref, 1);
+}
+EXPORT_SYMBOL_GPL(nvmet_cq_init);
+
+bool nvmet_cq_get(struct nvmet_cq *cq)
+{
+ return refcount_inc_not_zero(&cq->ref);
+}
+EXPORT_SYMBOL_GPL(nvmet_cq_get);
+
+void nvmet_cq_put(struct nvmet_cq *cq)
+{
+ if (refcount_dec_and_test(&cq->ref))
+ nvmet_cq_destroy(cq);
+}
+EXPORT_SYMBOL_GPL(nvmet_cq_put);
+
void nvmet_cq_setup(struct nvmet_ctrl *ctrl, struct nvmet_cq *cq,
u16 qid, u16 size)
{
cq->qid = qid;
cq->size = size;
+
+ ctrl->cqs[qid] = cq;
+}
+
+void nvmet_cq_destroy(struct nvmet_cq *cq)
+{
+ struct nvmet_ctrl *ctrl = cq->ctrl;
+
+ if (ctrl) {
+ ctrl->cqs[cq->qid] = NULL;
+ nvmet_ctrl_put(cq->ctrl);
+ cq->ctrl = NULL;
+ }
}
void nvmet_sq_setup(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq,
@@ -836,37 +862,47 @@ static void nvmet_confirm_sq(struct percpu_ref *ref)
complete(&sq->confirm_done);
}
-u16 nvmet_check_cqid(struct nvmet_ctrl *ctrl, u16 cqid)
+u16 nvmet_check_cqid(struct nvmet_ctrl *ctrl, u16 cqid, bool create)
{
- if (!ctrl->sqs)
+ if (!ctrl->cqs)
return NVME_SC_INTERNAL | NVME_STATUS_DNR;
if (cqid > ctrl->subsys->max_qid)
return NVME_SC_QID_INVALID | NVME_STATUS_DNR;
- /*
- * Note: For PCI controllers, the NVMe specifications allows multiple
- * SQs to share a single CQ. However, we do not support this yet, so
- * check that there is no SQ defined for a CQ. If one exist, then the
- * CQ ID is invalid for creation as well as when the CQ is being
- * deleted (as that would mean that the SQ was not deleted before the
- * CQ).
- */
- if (ctrl->sqs[cqid])
+ if ((create && ctrl->cqs[cqid]) || (!create && !ctrl->cqs[cqid]))
return NVME_SC_QID_INVALID | NVME_STATUS_DNR;
return NVME_SC_SUCCESS;
}
+u16 nvmet_check_io_cqid(struct nvmet_ctrl *ctrl, u16 cqid, bool create)
+{
+ if (!cqid)
+ return NVME_SC_QID_INVALID | NVME_STATUS_DNR;
+ return nvmet_check_cqid(ctrl, cqid, create);
+}
+
+bool nvmet_cq_in_use(struct nvmet_cq *cq)
+{
+ return refcount_read(&cq->ref) > 1;
+}
+EXPORT_SYMBOL_GPL(nvmet_cq_in_use);
+
u16 nvmet_cq_create(struct nvmet_ctrl *ctrl, struct nvmet_cq *cq,
u16 qid, u16 size)
{
u16 status;
- status = nvmet_check_cqid(ctrl, qid);
+ status = nvmet_check_cqid(ctrl, qid, true);
if (status != NVME_SC_SUCCESS)
return status;
+ if (!kref_get_unless_zero(&ctrl->ref))
+ return NVME_SC_INTERNAL | NVME_STATUS_DNR;
+ cq->ctrl = ctrl;
+
+ nvmet_cq_init(cq);
nvmet_cq_setup(ctrl, cq, qid, size);
return NVME_SC_SUCCESS;
@@ -890,7 +926,7 @@ u16 nvmet_check_sqid(struct nvmet_ctrl *ctrl, u16 sqid,
}
u16 nvmet_sq_create(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq,
- u16 sqid, u16 size)
+ struct nvmet_cq *cq, u16 sqid, u16 size)
{
u16 status;
int ret;
@@ -902,7 +938,7 @@ u16 nvmet_sq_create(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq,
if (status != NVME_SC_SUCCESS)
return status;
- ret = nvmet_sq_init(sq);
+ ret = nvmet_sq_init(sq, cq);
if (ret) {
status = NVME_SC_INTERNAL | NVME_STATUS_DNR;
goto ctrl_put;
@@ -934,6 +970,7 @@ void nvmet_sq_destroy(struct nvmet_sq *sq)
wait_for_completion(&sq->free_done);
percpu_ref_exit(&sq->ref);
nvmet_auth_sq_free(sq);
+ nvmet_cq_put(sq->cq);
/*
* we must reference the ctrl again after waiting for inflight IO
@@ -966,18 +1003,23 @@ static void nvmet_sq_free(struct percpu_ref *ref)
complete(&sq->free_done);
}
-int nvmet_sq_init(struct nvmet_sq *sq)
+int nvmet_sq_init(struct nvmet_sq *sq, struct nvmet_cq *cq)
{
int ret;
+ if (!nvmet_cq_get(cq))
+ return -EINVAL;
+
ret = percpu_ref_init(&sq->ref, nvmet_sq_free, 0, GFP_KERNEL);
if (ret) {
pr_err("percpu_ref init failed!\n");
+ nvmet_cq_put(cq);
return ret;
}
init_completion(&sq->free_done);
init_completion(&sq->confirm_done);
nvmet_auth_sq_init(sq);
+ sq->cq = cq;
return 0;
}
@@ -1107,13 +1149,13 @@ static u16 nvmet_parse_io_cmd(struct nvmet_req *req)
return ret;
}
-bool nvmet_req_init(struct nvmet_req *req, struct nvmet_cq *cq,
- struct nvmet_sq *sq, const struct nvmet_fabrics_ops *ops)
+bool nvmet_req_init(struct nvmet_req *req, struct nvmet_sq *sq,
+ const struct nvmet_fabrics_ops *ops)
{
u8 flags = req->cmd->common.flags;
u16 status;
- req->cq = cq;
+ req->cq = sq->cq;
req->sq = sq;
req->ops = ops;
req->sg = NULL;
@@ -1611,17 +1653,20 @@ struct nvmet_ctrl *nvmet_alloc_ctrl(struct nvmet_alloc_ctrl_args *args)
if (!ctrl->sqs)
goto out_free_changed_ns_list;
+ ctrl->cqs = kcalloc(subsys->max_qid + 1, sizeof(struct nvmet_cq *),
+ GFP_KERNEL);
+ if (!ctrl->cqs)
+ goto out_free_sqs;
+
ret = ida_alloc_range(&cntlid_ida,
subsys->cntlid_min, subsys->cntlid_max,
GFP_KERNEL);
if (ret < 0) {
args->status = NVME_SC_CONNECT_CTRL_BUSY | NVME_STATUS_DNR;
- goto out_free_sqs;
+ goto out_free_cqs;
}
ctrl->cntlid = ret;
- uuid_copy(&ctrl->hostid, args->hostid);
-
/*
* Discovery controllers may use some arbitrary high value
* in order to cleanup stale discovery sessions
@@ -1649,7 +1694,7 @@ struct nvmet_ctrl *nvmet_alloc_ctrl(struct nvmet_alloc_ctrl_args *args)
if (args->hostid)
uuid_copy(&ctrl->hostid, args->hostid);
- dhchap_status = nvmet_setup_auth(ctrl);
+ dhchap_status = nvmet_setup_auth(ctrl, args->sq);
if (dhchap_status) {
pr_err("Failed to setup authentication, dhchap status %u\n",
dhchap_status);
@@ -1664,11 +1709,12 @@ struct nvmet_ctrl *nvmet_alloc_ctrl(struct nvmet_alloc_ctrl_args *args)
args->status = NVME_SC_SUCCESS;
- pr_info("Created %s controller %d for subsystem %s for NQN %s%s%s.\n",
+ pr_info("Created %s controller %d for subsystem %s for NQN %s%s%s%s.\n",
nvmet_is_disc_subsys(ctrl->subsys) ? "discovery" : "nvm",
ctrl->cntlid, ctrl->subsys->subsysnqn, ctrl->hostnqn,
ctrl->pi_support ? " T10-PI is enabled" : "",
- nvmet_has_auth(ctrl) ? " with DH-HMAC-CHAP" : "");
+ nvmet_has_auth(ctrl, args->sq) ? " with DH-HMAC-CHAP" : "",
+ nvmet_queue_tls_keyid(args->sq) ? ", TLS" : "");
return ctrl;
@@ -1676,6 +1722,8 @@ init_pr_fail:
mutex_unlock(&subsys->lock);
nvmet_stop_keep_alive_timer(ctrl);
ida_free(&cntlid_ida, ctrl->cntlid);
+out_free_cqs:
+ kfree(ctrl->cqs);
out_free_sqs:
kfree(ctrl->sqs);
out_free_changed_ns_list:
@@ -1712,6 +1760,7 @@ static void nvmet_ctrl_free(struct kref *ref)
nvmet_async_events_free(ctrl);
kfree(ctrl->sqs);
+ kfree(ctrl->cqs);
kfree(ctrl->changed_ns_list);
kfree(ctrl);
diff --git a/drivers/nvme/target/debugfs.c b/drivers/nvme/target/debugfs.c
index 220c7391fc19..5dcbd5aa86e1 100644
--- a/drivers/nvme/target/debugfs.c
+++ b/drivers/nvme/target/debugfs.c
@@ -78,7 +78,7 @@ static int nvmet_ctrl_state_show(struct seq_file *m, void *p)
bool sep = false;
int i;
- for (i = 0; i < 7; i++) {
+ for (i = 0; i < ARRAY_SIZE(csts_state_names); i++) {
int state = BIT(i);
if (!(ctrl->csts & state))
@@ -132,6 +132,27 @@ static int nvmet_ctrl_host_traddr_show(struct seq_file *m, void *p)
}
NVMET_DEBUGFS_ATTR(nvmet_ctrl_host_traddr);
+#ifdef CONFIG_NVME_TARGET_TCP_TLS
+static int nvmet_ctrl_tls_key_show(struct seq_file *m, void *p)
+{
+ struct nvmet_ctrl *ctrl = m->private;
+ key_serial_t keyid = nvmet_queue_tls_keyid(ctrl->sqs[0]);
+
+ seq_printf(m, "%08x\n", keyid);
+ return 0;
+}
+NVMET_DEBUGFS_ATTR(nvmet_ctrl_tls_key);
+
+static int nvmet_ctrl_tls_concat_show(struct seq_file *m, void *p)
+{
+ struct nvmet_ctrl *ctrl = m->private;
+
+ seq_printf(m, "%d\n", ctrl->concat);
+ return 0;
+}
+NVMET_DEBUGFS_ATTR(nvmet_ctrl_tls_concat);
+#endif
+
int nvmet_debugfs_ctrl_setup(struct nvmet_ctrl *ctrl)
{
char name[32];
@@ -157,6 +178,12 @@ int nvmet_debugfs_ctrl_setup(struct nvmet_ctrl *ctrl)
&nvmet_ctrl_state_fops);
debugfs_create_file("host_traddr", S_IRUSR, ctrl->debugfs_dir, ctrl,
&nvmet_ctrl_host_traddr_fops);
+#ifdef CONFIG_NVME_TARGET_TCP_TLS
+ debugfs_create_file("tls_concat", S_IRUSR, ctrl->debugfs_dir, ctrl,
+ &nvmet_ctrl_tls_concat_fops);
+ debugfs_create_file("tls_key", S_IRUSR, ctrl->debugfs_dir, ctrl,
+ &nvmet_ctrl_tls_key_fops);
+#endif
return 0;
}
diff --git a/drivers/nvme/target/discovery.c b/drivers/nvme/target/discovery.c
index df7207640506..c06f3e04296c 100644
--- a/drivers/nvme/target/discovery.c
+++ b/drivers/nvme/target/discovery.c
@@ -119,7 +119,7 @@ static void nvmet_format_discovery_entry(struct nvmf_disc_rsp_page_hdr *hdr,
memcpy(e->trsvcid, port->disc_addr.trsvcid, NVMF_TRSVCID_SIZE);
memcpy(e->traddr, traddr, NVMF_TRADDR_SIZE);
memcpy(e->tsas.common, port->disc_addr.tsas.common, NVMF_TSAS_SIZE);
- strncpy(e->subnqn, subsys_nqn, NVMF_NQN_SIZE);
+ strscpy(e->subnqn, subsys_nqn, NVMF_NQN_SIZE);
}
/*
diff --git a/drivers/nvme/target/fabrics-cmd-auth.c b/drivers/nvme/target/fabrics-cmd-auth.c
index 2022757f08dc..bf01ec414c55 100644
--- a/drivers/nvme/target/fabrics-cmd-auth.c
+++ b/drivers/nvme/target/fabrics-cmd-auth.c
@@ -43,8 +43,26 @@ static u8 nvmet_auth_negotiate(struct nvmet_req *req, void *d)
data->auth_protocol[0].dhchap.halen,
data->auth_protocol[0].dhchap.dhlen);
req->sq->dhchap_tid = le16_to_cpu(data->t_id);
- if (data->sc_c)
- return NVME_AUTH_DHCHAP_FAILURE_CONCAT_MISMATCH;
+ if (data->sc_c != NVME_AUTH_SECP_NOSC) {
+ if (!IS_ENABLED(CONFIG_NVME_TARGET_TCP_TLS))
+ return NVME_AUTH_DHCHAP_FAILURE_CONCAT_MISMATCH;
+ /* Secure concatenation can only be enabled on the admin queue */
+ if (req->sq->qid)
+ return NVME_AUTH_DHCHAP_FAILURE_CONCAT_MISMATCH;
+ switch (data->sc_c) {
+ case NVME_AUTH_SECP_NEWTLSPSK:
+ if (nvmet_queue_tls_keyid(req->sq))
+ return NVME_AUTH_DHCHAP_FAILURE_CONCAT_MISMATCH;
+ break;
+ case NVME_AUTH_SECP_REPLACETLSPSK:
+ if (!nvmet_queue_tls_keyid(req->sq))
+ return NVME_AUTH_DHCHAP_FAILURE_CONCAT_MISMATCH;
+ break;
+ default:
+ return NVME_AUTH_DHCHAP_FAILURE_CONCAT_MISMATCH;
+ }
+ ctrl->concat = true;
+ }
if (data->napd != 1)
return NVME_AUTH_DHCHAP_FAILURE_HASH_UNUSABLE;
@@ -103,6 +121,12 @@ static u8 nvmet_auth_negotiate(struct nvmet_req *req, void *d)
nvme_auth_dhgroup_name(fallback_dhgid));
ctrl->dh_gid = fallback_dhgid;
}
+ if (ctrl->dh_gid == NVME_AUTH_DHGROUP_NULL && ctrl->concat) {
+ pr_debug("%s: ctrl %d qid %d: NULL DH group invalid "
+ "for secure channel concatenation\n", __func__,
+ ctrl->cntlid, req->sq->qid);
+ return NVME_AUTH_DHCHAP_FAILURE_CONCAT_MISMATCH;
+ }
pr_debug("%s: ctrl %d qid %d: selected DH group %s (%d)\n",
__func__, ctrl->cntlid, req->sq->qid,
nvme_auth_dhgroup_name(ctrl->dh_gid), ctrl->dh_gid);
@@ -148,12 +172,22 @@ static u8 nvmet_auth_reply(struct nvmet_req *req, void *d)
if (memcmp(data->rval, response, data->hl)) {
pr_info("ctrl %d qid %d host response mismatch\n",
ctrl->cntlid, req->sq->qid);
+ pr_debug("ctrl %d qid %d rval %*ph\n",
+ ctrl->cntlid, req->sq->qid, data->hl, data->rval);
+ pr_debug("ctrl %d qid %d response %*ph\n",
+ ctrl->cntlid, req->sq->qid, data->hl, response);
kfree(response);
return NVME_AUTH_DHCHAP_FAILURE_FAILED;
}
kfree(response);
pr_debug("%s: ctrl %d qid %d host authenticated\n",
__func__, ctrl->cntlid, req->sq->qid);
+ if (!data->cvalid && ctrl->concat) {
+ pr_debug("%s: ctrl %d qid %d invalid challenge\n",
+ __func__, ctrl->cntlid, req->sq->qid);
+ return NVME_AUTH_DHCHAP_FAILURE_FAILED;
+ }
+ req->sq->dhchap_s2 = le32_to_cpu(data->seqnum);
if (data->cvalid) {
req->sq->dhchap_c2 = kmemdup(data->rval + data->hl, data->hl,
GFP_KERNEL);
@@ -163,11 +197,23 @@ static u8 nvmet_auth_reply(struct nvmet_req *req, void *d)
pr_debug("%s: ctrl %d qid %d challenge %*ph\n",
__func__, ctrl->cntlid, req->sq->qid, data->hl,
req->sq->dhchap_c2);
- } else {
+ }
+ /*
+ * NVMe Base Spec 2.2 section 8.3.4.5.4: DH-HMAC-CHAP_Reply message
+ * Sequence Number (SEQNUM): [ .. ]
+ * The value 0h is used to indicate that bidirectional authentication
+ * is not performed, but a challenge value C2 is carried in order to
+ * generate a pre-shared key (PSK) for subsequent establishment of a
+ * secure channel.
+ */
+ if (req->sq->dhchap_s2 == 0) {
+ if (ctrl->concat)
+ nvmet_auth_insert_psk(req->sq);
req->sq->authenticated = true;
+ kfree(req->sq->dhchap_c2);
req->sq->dhchap_c2 = NULL;
- }
- req->sq->dhchap_s2 = le32_to_cpu(data->seqnum);
+ } else if (!data->cvalid)
+ req->sq->authenticated = true;
return 0;
}
@@ -246,7 +292,7 @@ void nvmet_execute_auth_send(struct nvmet_req *req)
pr_debug("%s: ctrl %d qid %d reset negotiation\n",
__func__, ctrl->cntlid, req->sq->qid);
if (!req->sq->qid) {
- dhchap_status = nvmet_setup_auth(ctrl);
+ dhchap_status = nvmet_setup_auth(ctrl, req->sq);
if (dhchap_status) {
pr_err("ctrl %d qid 0 failed to setup re-authentication\n",
ctrl->cntlid);
@@ -303,6 +349,8 @@ void nvmet_execute_auth_send(struct nvmet_req *req)
}
goto done_kfree;
case NVME_AUTH_DHCHAP_MESSAGE_SUCCESS2:
+ if (ctrl->concat)
+ nvmet_auth_insert_psk(req->sq);
req->sq->authenticated = true;
pr_debug("%s: ctrl %d qid %d ctrl authenticated\n",
__func__, ctrl->cntlid, req->sq->qid);
diff --git a/drivers/nvme/target/fabrics-cmd.c b/drivers/nvme/target/fabrics-cmd.c
index eb406c90c167..7b8d8b397802 100644
--- a/drivers/nvme/target/fabrics-cmd.c
+++ b/drivers/nvme/target/fabrics-cmd.c
@@ -208,6 +208,14 @@ static u16 nvmet_install_queue(struct nvmet_ctrl *ctrl, struct nvmet_req *req)
return NVME_SC_CONNECT_CTRL_BUSY | NVME_STATUS_DNR;
}
+ kref_get(&ctrl->ref);
+ old = cmpxchg(&req->cq->ctrl, NULL, ctrl);
+ if (old) {
+ pr_warn("queue already connected!\n");
+ req->error_loc = offsetof(struct nvmf_connect_command, opcode);
+ return NVME_SC_CONNECT_CTRL_BUSY | NVME_STATUS_DNR;
+ }
+
/* note: convert queue size from 0's-based value to 1's-based value */
nvmet_cq_setup(ctrl, req->cq, qid, sqsize + 1);
nvmet_sq_setup(ctrl, req->sq, qid, sqsize + 1);
@@ -234,10 +242,26 @@ err:
return ret;
}
-static u32 nvmet_connect_result(struct nvmet_ctrl *ctrl)
+static u32 nvmet_connect_result(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq)
{
+ bool needs_auth = nvmet_has_auth(ctrl, sq);
+ key_serial_t keyid = nvmet_queue_tls_keyid(sq);
+
+ /* Do not authenticate I/O queues */
+ if (sq->qid)
+ needs_auth = false;
+
+ if (keyid)
+ pr_debug("%s: ctrl %d qid %d should %sauthenticate, tls psk %08x\n",
+ __func__, ctrl->cntlid, sq->qid,
+ needs_auth ? "" : "not ", keyid);
+ else
+ pr_debug("%s: ctrl %d qid %d should %sauthenticate%s\n",
+ __func__, ctrl->cntlid, sq->qid,
+ needs_auth ? "" : "not ",
+ ctrl->concat ? ", secure concatenation" : "");
return (u32)ctrl->cntlid |
- (nvmet_has_auth(ctrl) ? NVME_CONNECT_AUTHREQ_ATR : 0);
+ (needs_auth ? NVME_CONNECT_AUTHREQ_ATR : 0);
}
static void nvmet_execute_admin_connect(struct nvmet_req *req)
@@ -247,6 +271,7 @@ static void nvmet_execute_admin_connect(struct nvmet_req *req)
struct nvmet_ctrl *ctrl = NULL;
struct nvmet_alloc_ctrl_args args = {
.port = req->port,
+ .sq = req->sq,
.ops = req->ops,
.p2p_client = req->p2p_client,
.kato = le32_to_cpu(c->kato),
@@ -299,7 +324,7 @@ static void nvmet_execute_admin_connect(struct nvmet_req *req)
goto out;
}
- args.result = cpu_to_le32(nvmet_connect_result(ctrl));
+ args.result = cpu_to_le32(nvmet_connect_result(ctrl, req->sq));
out:
kfree(d);
complete:
@@ -357,7 +382,7 @@ static void nvmet_execute_io_connect(struct nvmet_req *req)
goto out_ctrl_put;
pr_debug("adding queue %d to ctrl %d.\n", qid, ctrl->cntlid);
- req->cqe->result.u32 = cpu_to_le32(nvmet_connect_result(ctrl));
+ req->cqe->result.u32 = cpu_to_le32(nvmet_connect_result(ctrl, req->sq));
out:
kfree(d);
complete:
diff --git a/drivers/nvme/target/fc.c b/drivers/nvme/target/fc.c
index 3ef4beacde32..25598a46bf0d 100644
--- a/drivers/nvme/target/fc.c
+++ b/drivers/nvme/target/fc.c
@@ -172,20 +172,6 @@ struct nvmet_fc_tgt_assoc {
struct work_struct del_work;
};
-
-static inline int
-nvmet_fc_iodnum(struct nvmet_fc_ls_iod *iodptr)
-{
- return (iodptr - iodptr->tgtport->iod);
-}
-
-static inline int
-nvmet_fc_fodnum(struct nvmet_fc_fcp_iod *fodptr)
-{
- return (fodptr - fodptr->queue->fod);
-}
-
-
/*
* Association and Connection IDs:
*
@@ -830,7 +816,8 @@ nvmet_fc_alloc_target_queue(struct nvmet_fc_tgt_assoc *assoc,
nvmet_fc_prep_fcp_iodlist(assoc->tgtport, queue);
- ret = nvmet_sq_init(&queue->nvme_sq);
+ nvmet_cq_init(&queue->nvme_cq);
+ ret = nvmet_sq_init(&queue->nvme_sq, &queue->nvme_cq);
if (ret)
goto out_fail_iodlist;
@@ -840,6 +827,7 @@ nvmet_fc_alloc_target_queue(struct nvmet_fc_tgt_assoc *assoc,
return queue;
out_fail_iodlist:
+ nvmet_cq_put(&queue->nvme_cq);
nvmet_fc_destroy_fcp_iodlist(assoc->tgtport, queue);
destroy_workqueue(queue->work_q);
out_free_queue:
@@ -948,6 +936,7 @@ nvmet_fc_delete_target_queue(struct nvmet_fc_tgt_queue *queue)
flush_workqueue(queue->work_q);
nvmet_sq_destroy(&queue->nvme_sq);
+ nvmet_cq_put(&queue->nvme_cq);
nvmet_fc_tgt_q_put(queue);
}
@@ -1009,16 +998,6 @@ nvmet_fc_hostport_get(struct nvmet_fc_hostport *hostport)
return kref_get_unless_zero(&hostport->ref);
}
-static void
-nvmet_fc_free_hostport(struct nvmet_fc_hostport *hostport)
-{
- /* if LLDD not implemented, leave as NULL */
- if (!hostport || !hostport->hosthandle)
- return;
-
- nvmet_fc_hostport_put(hostport);
-}
-
static struct nvmet_fc_hostport *
nvmet_fc_match_hostport(struct nvmet_fc_tgtport *tgtport, void *hosthandle)
{
@@ -1042,33 +1021,24 @@ nvmet_fc_alloc_hostport(struct nvmet_fc_tgtport *tgtport, void *hosthandle)
struct nvmet_fc_hostport *newhost, *match = NULL;
unsigned long flags;
+ /*
+ * Caller holds a reference on tgtport.
+ */
+
/* if LLDD not implemented, leave as NULL */
if (!hosthandle)
return NULL;
- /*
- * take reference for what will be the newly allocated hostport if
- * we end up using a new allocation
- */
- if (!nvmet_fc_tgtport_get(tgtport))
- return ERR_PTR(-EINVAL);
-
spin_lock_irqsave(&tgtport->lock, flags);
match = nvmet_fc_match_hostport(tgtport, hosthandle);
spin_unlock_irqrestore(&tgtport->lock, flags);
- if (match) {
- /* no new allocation - release reference */
- nvmet_fc_tgtport_put(tgtport);
+ if (match)
return match;
- }
newhost = kzalloc(sizeof(*newhost), GFP_KERNEL);
- if (!newhost) {
- /* no new allocation - release reference */
- nvmet_fc_tgtport_put(tgtport);
+ if (!newhost)
return ERR_PTR(-ENOMEM);
- }
spin_lock_irqsave(&tgtport->lock, flags);
match = nvmet_fc_match_hostport(tgtport, hosthandle);
@@ -1077,6 +1047,7 @@ nvmet_fc_alloc_hostport(struct nvmet_fc_tgtport *tgtport, void *hosthandle)
kfree(newhost);
newhost = match;
} else {
+ nvmet_fc_tgtport_get(tgtport);
newhost->tgtport = tgtport;
newhost->hosthandle = hosthandle;
INIT_LIST_HEAD(&newhost->host_list);
@@ -1090,20 +1061,14 @@ nvmet_fc_alloc_hostport(struct nvmet_fc_tgtport *tgtport, void *hosthandle)
}
static void
-nvmet_fc_delete_assoc(struct nvmet_fc_tgt_assoc *assoc)
-{
- nvmet_fc_delete_target_assoc(assoc);
- nvmet_fc_tgt_a_put(assoc);
-}
-
-static void
nvmet_fc_delete_assoc_work(struct work_struct *work)
{
struct nvmet_fc_tgt_assoc *assoc =
container_of(work, struct nvmet_fc_tgt_assoc, del_work);
struct nvmet_fc_tgtport *tgtport = assoc->tgtport;
- nvmet_fc_delete_assoc(assoc);
+ nvmet_fc_delete_target_assoc(assoc);
+ nvmet_fc_tgt_a_put(assoc);
nvmet_fc_tgtport_put(tgtport);
}
@@ -1111,7 +1076,8 @@ static void
nvmet_fc_schedule_delete_assoc(struct nvmet_fc_tgt_assoc *assoc)
{
nvmet_fc_tgtport_get(assoc->tgtport);
- queue_work(nvmet_wq, &assoc->del_work);
+ if (!queue_work(nvmet_wq, &assoc->del_work))
+ nvmet_fc_tgtport_put(assoc->tgtport);
}
static bool
@@ -1157,6 +1123,7 @@ nvmet_fc_alloc_target_assoc(struct nvmet_fc_tgtport *tgtport, void *hosthandle)
goto out_ida;
assoc->tgtport = tgtport;
+ nvmet_fc_tgtport_get(tgtport);
assoc->a_id = idx;
INIT_LIST_HEAD(&assoc->a_list);
kref_init(&assoc->ref);
@@ -1204,7 +1171,7 @@ nvmet_fc_target_assoc_free(struct kref *ref)
/* Send Disconnect now that all i/o has completed */
nvmet_fc_xmt_disconnect_assoc(assoc);
- nvmet_fc_free_hostport(assoc->hostport);
+ nvmet_fc_hostport_put(assoc->hostport);
spin_lock_irqsave(&tgtport->lock, flags);
oldls = assoc->rcv_disconn;
spin_unlock_irqrestore(&tgtport->lock, flags);
@@ -1258,6 +1225,8 @@ nvmet_fc_delete_target_assoc(struct nvmet_fc_tgt_assoc *assoc)
dev_info(tgtport->dev,
"{%d:%d} Association deleted\n",
tgtport->fc_target_port.port_num, assoc->a_id);
+
+ nvmet_fc_tgtport_put(tgtport);
}
static struct nvmet_fc_tgt_assoc *
@@ -1288,6 +1257,7 @@ nvmet_fc_portentry_bind(struct nvmet_fc_tgtport *tgtport,
{
lockdep_assert_held(&nvmet_fc_tgtlock);
+ nvmet_fc_tgtport_get(tgtport);
pe->tgtport = tgtport;
tgtport->pe = pe;
@@ -1307,8 +1277,10 @@ nvmet_fc_portentry_unbind(struct nvmet_fc_port_entry *pe)
unsigned long flags;
spin_lock_irqsave(&nvmet_fc_tgtlock, flags);
- if (pe->tgtport)
+ if (pe->tgtport) {
+ nvmet_fc_tgtport_put(pe->tgtport);
pe->tgtport->pe = NULL;
+ }
list_del(&pe->pe_list);
spin_unlock_irqrestore(&nvmet_fc_tgtlock, flags);
}
@@ -1326,8 +1298,10 @@ nvmet_fc_portentry_unbind_tgt(struct nvmet_fc_tgtport *tgtport)
spin_lock_irqsave(&nvmet_fc_tgtlock, flags);
pe = tgtport->pe;
- if (pe)
+ if (pe) {
+ nvmet_fc_tgtport_put(pe->tgtport);
pe->tgtport = NULL;
+ }
tgtport->pe = NULL;
spin_unlock_irqrestore(&nvmet_fc_tgtlock, flags);
}
@@ -1350,6 +1324,9 @@ nvmet_fc_portentry_rebind_tgt(struct nvmet_fc_tgtport *tgtport)
list_for_each_entry(pe, &nvmet_fc_portentry_list, pe_list) {
if (tgtport->fc_target_port.node_name == pe->node_name &&
tgtport->fc_target_port.port_name == pe->port_name) {
+ if (!nvmet_fc_tgtport_get(tgtport))
+ continue;
+
WARN_ON(pe->tgtport);
tgtport->pe = pe;
pe->tgtport = tgtport;
@@ -1362,7 +1339,7 @@ nvmet_fc_portentry_rebind_tgt(struct nvmet_fc_tgtport *tgtport)
/**
* nvmet_fc_register_targetport - transport entry point called by an
* LLDD to register the existence of a local
- * NVME subystem FC port.
+ * NVME subsystem FC port.
* @pinfo: pointer to information about the port to be registered
* @template: LLDD entrypoints and operational parameters for the port
* @dev: physical hardware device node port corresponds to. Will be
@@ -1469,11 +1446,6 @@ nvmet_fc_free_tgtport(struct kref *ref)
struct nvmet_fc_tgtport *tgtport =
container_of(ref, struct nvmet_fc_tgtport, ref);
struct device *dev = tgtport->dev;
- unsigned long flags;
-
- spin_lock_irqsave(&nvmet_fc_tgtlock, flags);
- list_del(&tgtport->tgt_list);
- spin_unlock_irqrestore(&nvmet_fc_tgtlock, flags);
nvmet_fc_free_ls_iodlist(tgtport);
@@ -1619,6 +1591,39 @@ nvmet_fc_delete_ctrl(struct nvmet_ctrl *ctrl)
spin_unlock_irqrestore(&nvmet_fc_tgtlock, flags);
}
+static void
+nvmet_fc_free_pending_reqs(struct nvmet_fc_tgtport *tgtport)
+{
+ struct nvmet_fc_ls_req_op *lsop;
+ struct nvmefc_ls_req *lsreq;
+ struct nvmet_fc_ls_iod *iod;
+ int i;
+
+ iod = tgtport->iod;
+ for (i = 0; i < NVMET_LS_CTX_COUNT; iod++, i++)
+ cancel_work(&iod->work);
+
+ /*
+ * After this point the connection is lost and thus any pending
+ * request can't be processed by the normal completion path. This
+ * is likely a request from nvmet_fc_send_ls_req_async.
+ */
+ while ((lsop = list_first_entry_or_null(&tgtport->ls_req_list,
+ struct nvmet_fc_ls_req_op, lsreq_list))) {
+ list_del(&lsop->lsreq_list);
+
+ if (!lsop->req_queued)
+ continue;
+
+ lsreq = &lsop->ls_req;
+ fc_dma_unmap_single(tgtport->dev, lsreq->rqstdma,
+ (lsreq->rqstlen + lsreq->rsplen),
+ DMA_BIDIRECTIONAL);
+ nvmet_fc_tgtport_put(tgtport);
+ kfree(lsop);
+ }
+}
+
/**
* nvmet_fc_unregister_targetport - transport entry point called by an
* LLDD to deregister/remove a previously
@@ -1634,6 +1639,11 @@ int
nvmet_fc_unregister_targetport(struct nvmet_fc_target_port *target_port)
{
struct nvmet_fc_tgtport *tgtport = targetport_to_tgtport(target_port);
+ unsigned long flags;
+
+ spin_lock_irqsave(&nvmet_fc_tgtlock, flags);
+ list_del(&tgtport->tgt_list);
+ spin_unlock_irqrestore(&nvmet_fc_tgtlock, flags);
nvmet_fc_portentry_unbind_tgt(tgtport);
@@ -1642,13 +1652,7 @@ nvmet_fc_unregister_targetport(struct nvmet_fc_target_port *target_port)
flush_workqueue(nvmet_wq);
- /*
- * should terminate LS's as well. However, LS's will be generated
- * at the tail end of association termination, so they likely don't
- * exist yet. And even if they did, it's worthwhile to just let
- * them finish and targetport ref counting will clean things up.
- */
-
+ nvmet_fc_free_pending_reqs(tgtport);
nvmet_fc_tgtport_put(tgtport);
return 0;
@@ -2565,10 +2569,8 @@ nvmet_fc_handle_fcp_rqst(struct nvmet_fc_tgtport *tgtport,
fod->data_sg = NULL;
fod->data_sg_cnt = 0;
- ret = nvmet_req_init(&fod->req,
- &fod->queue->nvme_cq,
- &fod->queue->nvme_sq,
- &nvmet_fc_tgt_fcp_ops);
+ ret = nvmet_req_init(&fod->req, &fod->queue->nvme_sq,
+ &nvmet_fc_tgt_fcp_ops);
if (!ret) {
/* bad SQE content or invalid ctrl state */
/* nvmet layer has already called op done to send rsp. */
@@ -2894,12 +2896,17 @@ nvmet_fc_add_port(struct nvmet_port *port)
list_for_each_entry(tgtport, &nvmet_fc_target_list, tgt_list) {
if ((tgtport->fc_target_port.node_name == traddr.nn) &&
(tgtport->fc_target_port.port_name == traddr.pn)) {
+ if (!nvmet_fc_tgtport_get(tgtport))
+ continue;
+
/* a FC port can only be 1 nvmet port id */
if (!tgtport->pe) {
nvmet_fc_portentry_bind(tgtport, pe, port);
ret = 0;
} else
ret = -EALREADY;
+
+ nvmet_fc_tgtport_put(tgtport);
break;
}
}
@@ -2915,11 +2922,21 @@ static void
nvmet_fc_remove_port(struct nvmet_port *port)
{
struct nvmet_fc_port_entry *pe = port->priv;
+ struct nvmet_fc_tgtport *tgtport = NULL;
+ unsigned long flags;
+
+ spin_lock_irqsave(&nvmet_fc_tgtlock, flags);
+ if (pe->tgtport && nvmet_fc_tgtport_get(pe->tgtport))
+ tgtport = pe->tgtport;
+ spin_unlock_irqrestore(&nvmet_fc_tgtlock, flags);
nvmet_fc_portentry_unbind(pe);
- /* terminate any outstanding associations */
- __nvmet_fc_free_assocs(pe->tgtport);
+ if (tgtport) {
+ /* terminate any outstanding associations */
+ __nvmet_fc_free_assocs(tgtport);
+ nvmet_fc_tgtport_put(tgtport);
+ }
kfree(pe);
}
@@ -2928,10 +2945,21 @@ static void
nvmet_fc_discovery_chg(struct nvmet_port *port)
{
struct nvmet_fc_port_entry *pe = port->priv;
- struct nvmet_fc_tgtport *tgtport = pe->tgtport;
+ struct nvmet_fc_tgtport *tgtport = NULL;
+ unsigned long flags;
+
+ spin_lock_irqsave(&nvmet_fc_tgtlock, flags);
+ if (pe->tgtport && nvmet_fc_tgtport_get(pe->tgtport))
+ tgtport = pe->tgtport;
+ spin_unlock_irqrestore(&nvmet_fc_tgtlock, flags);
+
+ if (!tgtport)
+ return;
if (tgtport && tgtport->ops->discovery_event)
tgtport->ops->discovery_event(&tgtport->fc_target_port);
+
+ nvmet_fc_tgtport_put(tgtport);
}
static ssize_t
diff --git a/drivers/nvme/target/fcloop.c b/drivers/nvme/target/fcloop.c
index e1abb27927ff..257b497d515a 100644
--- a/drivers/nvme/target/fcloop.c
+++ b/drivers/nvme/target/fcloop.c
@@ -207,13 +207,16 @@ static LIST_HEAD(fcloop_nports);
struct fcloop_lport {
struct nvme_fc_local_port *localport;
struct list_head lport_list;
- struct completion unreg_done;
+ refcount_t ref;
};
struct fcloop_lport_priv {
struct fcloop_lport *lport;
};
+/* The port is already being removed, avoid double free */
+#define PORT_DELETED 0
+
struct fcloop_rport {
struct nvme_fc_remote_port *remoteport;
struct nvmet_fc_target_port *targetport;
@@ -222,6 +225,7 @@ struct fcloop_rport {
spinlock_t lock;
struct list_head ls_list;
struct work_struct ls_work;
+ unsigned long flags;
};
struct fcloop_tport {
@@ -232,6 +236,7 @@ struct fcloop_tport {
spinlock_t lock;
struct list_head ls_list;
struct work_struct ls_work;
+ unsigned long flags;
};
struct fcloop_nport {
@@ -239,7 +244,7 @@ struct fcloop_nport {
struct fcloop_tport *tport;
struct fcloop_lport *lport;
struct list_head nport_list;
- struct kref ref;
+ refcount_t ref;
u64 node_name;
u64 port_name;
u32 port_role;
@@ -274,7 +279,7 @@ struct fcloop_fcpreq {
u32 inistate;
bool active;
bool aborted;
- struct kref ref;
+ refcount_t ref;
struct work_struct fcp_rcv_work;
struct work_struct abort_rcv_work;
struct work_struct tio_done_work;
@@ -287,6 +292,9 @@ struct fcloop_ini_fcpreq {
spinlock_t inilock;
};
+/* SLAB cache for fcloop_lsreq structures */
+static struct kmem_cache *lsreq_cache;
+
static inline struct fcloop_lsreq *
ls_rsp_to_lsreq(struct nvmefc_ls_rsp *lsrsp)
{
@@ -337,6 +345,7 @@ fcloop_rport_lsrqst_work(struct work_struct *work)
* callee may free memory containing tls_req.
* do not reference lsreq after this.
*/
+ kmem_cache_free(lsreq_cache, tls_req);
spin_lock(&rport->lock);
}
@@ -348,10 +357,13 @@ fcloop_h2t_ls_req(struct nvme_fc_local_port *localport,
struct nvme_fc_remote_port *remoteport,
struct nvmefc_ls_req *lsreq)
{
- struct fcloop_lsreq *tls_req = lsreq->private;
struct fcloop_rport *rport = remoteport->private;
+ struct fcloop_lsreq *tls_req;
int ret = 0;
+ tls_req = kmem_cache_alloc(lsreq_cache, GFP_KERNEL);
+ if (!tls_req)
+ return -ENOMEM;
tls_req->lsreq = lsreq;
INIT_LIST_HEAD(&tls_req->ls_list);
@@ -388,14 +400,17 @@ fcloop_h2t_xmt_ls_rsp(struct nvmet_fc_target_port *targetport,
lsrsp->done(lsrsp);
- if (remoteport) {
- rport = remoteport->private;
- spin_lock(&rport->lock);
- list_add_tail(&tls_req->ls_list, &rport->ls_list);
- spin_unlock(&rport->lock);
- queue_work(nvmet_wq, &rport->ls_work);
+ if (!remoteport) {
+ kmem_cache_free(lsreq_cache, tls_req);
+ return 0;
}
+ rport = remoteport->private;
+ spin_lock(&rport->lock);
+ list_add_tail(&tls_req->ls_list, &rport->ls_list);
+ spin_unlock(&rport->lock);
+ queue_work(nvmet_wq, &rport->ls_work);
+
return 0;
}
@@ -421,6 +436,7 @@ fcloop_tport_lsrqst_work(struct work_struct *work)
* callee may free memory containing tls_req.
* do not reference lsreq after this.
*/
+ kmem_cache_free(lsreq_cache, tls_req);
spin_lock(&tport->lock);
}
@@ -431,8 +447,8 @@ static int
fcloop_t2h_ls_req(struct nvmet_fc_target_port *targetport, void *hosthandle,
struct nvmefc_ls_req *lsreq)
{
- struct fcloop_lsreq *tls_req = lsreq->private;
struct fcloop_tport *tport = targetport->private;
+ struct fcloop_lsreq *tls_req;
int ret = 0;
/*
@@ -440,6 +456,10 @@ fcloop_t2h_ls_req(struct nvmet_fc_target_port *targetport, void *hosthandle,
* hosthandle ignored as fcloop currently is
* 1:1 tgtport vs remoteport
*/
+
+ tls_req = kmem_cache_alloc(lsreq_cache, GFP_KERNEL);
+ if (!tls_req)
+ return -ENOMEM;
tls_req->lsreq = lsreq;
INIT_LIST_HEAD(&tls_req->ls_list);
@@ -456,6 +476,9 @@ fcloop_t2h_ls_req(struct nvmet_fc_target_port *targetport, void *hosthandle,
ret = nvme_fc_rcv_ls_req(tport->remoteport, &tls_req->ls_rsp,
lsreq->rqstaddr, lsreq->rqstlen);
+ if (ret)
+ kmem_cache_free(lsreq_cache, tls_req);
+
return ret;
}
@@ -470,18 +493,30 @@ fcloop_t2h_xmt_ls_rsp(struct nvme_fc_local_port *localport,
struct nvmet_fc_target_port *targetport = rport->targetport;
struct fcloop_tport *tport;
+ if (!targetport) {
+ /*
+ * The target port is gone. The target doesn't expect any
+ * response anymore and the ->done call is not valid
+ * because the resources have been freed by
+ * nvmet_fc_free_pending_reqs.
+ *
+ * We end up here from delete association exchange:
+ * nvmet_fc_xmt_disconnect_assoc sends an async request.
+ */
+ kmem_cache_free(lsreq_cache, tls_req);
+ return 0;
+ }
+
memcpy(lsreq->rspaddr, lsrsp->rspbuf,
((lsreq->rsplen < lsrsp->rsplen) ?
lsreq->rsplen : lsrsp->rsplen));
lsrsp->done(lsrsp);
- if (targetport) {
- tport = targetport->private;
- spin_lock(&tport->lock);
- list_add_tail(&tport->ls_list, &tls_req->ls_list);
- spin_unlock(&tport->lock);
- queue_work(nvmet_wq, &tport->ls_work);
- }
+ tport = targetport->private;
+ spin_lock(&tport->lock);
+ list_add_tail(&tls_req->ls_list, &tport->ls_list);
+ spin_unlock(&tport->lock);
+ queue_work(nvmet_wq, &tport->ls_work);
return 0;
}
@@ -534,24 +569,18 @@ fcloop_tgt_discovery_evt(struct nvmet_fc_target_port *tgtport)
}
static void
-fcloop_tfcp_req_free(struct kref *ref)
+fcloop_tfcp_req_put(struct fcloop_fcpreq *tfcp_req)
{
- struct fcloop_fcpreq *tfcp_req =
- container_of(ref, struct fcloop_fcpreq, ref);
+ if (!refcount_dec_and_test(&tfcp_req->ref))
+ return;
kfree(tfcp_req);
}
-static void
-fcloop_tfcp_req_put(struct fcloop_fcpreq *tfcp_req)
-{
- kref_put(&tfcp_req->ref, fcloop_tfcp_req_free);
-}
-
static int
fcloop_tfcp_req_get(struct fcloop_fcpreq *tfcp_req)
{
- return kref_get_unless_zero(&tfcp_req->ref);
+ return refcount_inc_not_zero(&tfcp_req->ref);
}
static void
@@ -571,7 +600,8 @@ fcloop_call_host_done(struct nvmefc_fcp_req *fcpreq,
}
/* release original io reference on tgt struct */
- fcloop_tfcp_req_put(tfcp_req);
+ if (tfcp_req)
+ fcloop_tfcp_req_put(tfcp_req);
}
static bool drop_fabric_opcode;
@@ -623,12 +653,13 @@ fcloop_fcp_recv_work(struct work_struct *work)
{
struct fcloop_fcpreq *tfcp_req =
container_of(work, struct fcloop_fcpreq, fcp_rcv_work);
- struct nvmefc_fcp_req *fcpreq = tfcp_req->fcpreq;
+ struct nvmefc_fcp_req *fcpreq;
unsigned long flags;
int ret = 0;
bool aborted = false;
spin_lock_irqsave(&tfcp_req->reqlock, flags);
+ fcpreq = tfcp_req->fcpreq;
switch (tfcp_req->inistate) {
case INI_IO_START:
tfcp_req->inistate = INI_IO_ACTIVE;
@@ -643,16 +674,19 @@ fcloop_fcp_recv_work(struct work_struct *work)
}
spin_unlock_irqrestore(&tfcp_req->reqlock, flags);
- if (unlikely(aborted))
- ret = -ECANCELED;
- else {
- if (likely(!check_for_drop(tfcp_req)))
- ret = nvmet_fc_rcv_fcp_req(tfcp_req->tport->targetport,
- &tfcp_req->tgt_fcp_req,
- fcpreq->cmdaddr, fcpreq->cmdlen);
- else
- pr_info("%s: dropped command ********\n", __func__);
+ if (unlikely(aborted)) {
+ /* the abort handler will call fcloop_call_host_done */
+ return;
+ }
+
+ if (unlikely(check_for_drop(tfcp_req))) {
+ pr_info("%s: dropped command ********\n", __func__);
+ return;
}
+
+ ret = nvmet_fc_rcv_fcp_req(tfcp_req->tport->targetport,
+ &tfcp_req->tgt_fcp_req,
+ fcpreq->cmdaddr, fcpreq->cmdlen);
if (ret)
fcloop_call_host_done(fcpreq, tfcp_req, ret);
}
@@ -667,15 +701,17 @@ fcloop_fcp_abort_recv_work(struct work_struct *work)
unsigned long flags;
spin_lock_irqsave(&tfcp_req->reqlock, flags);
- fcpreq = tfcp_req->fcpreq;
switch (tfcp_req->inistate) {
case INI_IO_ABORTED:
+ fcpreq = tfcp_req->fcpreq;
+ tfcp_req->fcpreq = NULL;
break;
case INI_IO_COMPLETED:
completed = true;
break;
default:
spin_unlock_irqrestore(&tfcp_req->reqlock, flags);
+ fcloop_tfcp_req_put(tfcp_req);
WARN_ON(1);
return;
}
@@ -691,10 +727,6 @@ fcloop_fcp_abort_recv_work(struct work_struct *work)
nvmet_fc_rcv_fcp_abort(tfcp_req->tport->targetport,
&tfcp_req->tgt_fcp_req);
- spin_lock_irqsave(&tfcp_req->reqlock, flags);
- tfcp_req->fcpreq = NULL;
- spin_unlock_irqrestore(&tfcp_req->reqlock, flags);
-
fcloop_call_host_done(fcpreq, tfcp_req, -ECANCELED);
/* call_host_done releases reference for abort downcall */
}
@@ -748,7 +780,7 @@ fcloop_fcp_req(struct nvme_fc_local_port *localport,
INIT_WORK(&tfcp_req->fcp_rcv_work, fcloop_fcp_recv_work);
INIT_WORK(&tfcp_req->abort_rcv_work, fcloop_fcp_abort_recv_work);
INIT_WORK(&tfcp_req->tio_done_work, fcloop_tgt_fcprqst_done_work);
- kref_init(&tfcp_req->ref);
+ refcount_set(&tfcp_req->ref, 1);
queue_work(nvmet_wq, &tfcp_req->fcp_rcv_work);
@@ -963,13 +995,16 @@ fcloop_fcp_abort(struct nvme_fc_local_port *localport,
spin_lock(&inireq->inilock);
tfcp_req = inireq->tfcp_req;
- if (tfcp_req)
- fcloop_tfcp_req_get(tfcp_req);
+ if (tfcp_req) {
+ if (!fcloop_tfcp_req_get(tfcp_req))
+ tfcp_req = NULL;
+ }
spin_unlock(&inireq->inilock);
- if (!tfcp_req)
+ if (!tfcp_req) {
/* abort has already been called */
- return;
+ goto out_host_done;
+ }
/* break initiator/target relationship for io */
spin_lock_irqsave(&tfcp_req->reqlock, flags);
@@ -984,7 +1019,7 @@ fcloop_fcp_abort(struct nvme_fc_local_port *localport,
default:
spin_unlock_irqrestore(&tfcp_req->reqlock, flags);
WARN_ON(1);
- return;
+ goto out_host_done;
}
spin_unlock_irqrestore(&tfcp_req->reqlock, flags);
@@ -998,27 +1033,56 @@ fcloop_fcp_abort(struct nvme_fc_local_port *localport,
*/
fcloop_tfcp_req_put(tfcp_req);
}
+
+ return;
+
+out_host_done:
+ fcloop_call_host_done(fcpreq, tfcp_req, -ECANCELED);
}
static void
-fcloop_nport_free(struct kref *ref)
+fcloop_lport_put(struct fcloop_lport *lport)
{
- struct fcloop_nport *nport =
- container_of(ref, struct fcloop_nport, ref);
+ unsigned long flags;
- kfree(nport);
+ if (!refcount_dec_and_test(&lport->ref))
+ return;
+
+ spin_lock_irqsave(&fcloop_lock, flags);
+ list_del(&lport->lport_list);
+ spin_unlock_irqrestore(&fcloop_lock, flags);
+
+ kfree(lport);
+}
+
+static int
+fcloop_lport_get(struct fcloop_lport *lport)
+{
+ return refcount_inc_not_zero(&lport->ref);
}
static void
fcloop_nport_put(struct fcloop_nport *nport)
{
- kref_put(&nport->ref, fcloop_nport_free);
+ unsigned long flags;
+
+ if (!refcount_dec_and_test(&nport->ref))
+ return;
+
+ spin_lock_irqsave(&fcloop_lock, flags);
+ list_del(&nport->nport_list);
+ spin_unlock_irqrestore(&fcloop_lock, flags);
+
+ if (nport->lport)
+ fcloop_lport_put(nport->lport);
+
+ kfree(nport);
}
static int
fcloop_nport_get(struct fcloop_nport *nport)
{
- return kref_get_unless_zero(&nport->ref);
+ return refcount_inc_not_zero(&nport->ref);
}
static void
@@ -1027,26 +1091,45 @@ fcloop_localport_delete(struct nvme_fc_local_port *localport)
struct fcloop_lport_priv *lport_priv = localport->private;
struct fcloop_lport *lport = lport_priv->lport;
- /* release any threads waiting for the unreg to complete */
- complete(&lport->unreg_done);
+ fcloop_lport_put(lport);
}
static void
fcloop_remoteport_delete(struct nvme_fc_remote_port *remoteport)
{
struct fcloop_rport *rport = remoteport->private;
+ bool put_port = false;
+ unsigned long flags;
flush_work(&rport->ls_work);
- fcloop_nport_put(rport->nport);
+
+ spin_lock_irqsave(&fcloop_lock, flags);
+ if (!test_and_set_bit(PORT_DELETED, &rport->flags))
+ put_port = true;
+ rport->nport->rport = NULL;
+ spin_unlock_irqrestore(&fcloop_lock, flags);
+
+ if (put_port)
+ fcloop_nport_put(rport->nport);
}
static void
fcloop_targetport_delete(struct nvmet_fc_target_port *targetport)
{
struct fcloop_tport *tport = targetport->private;
+ bool put_port = false;
+ unsigned long flags;
flush_work(&tport->ls_work);
- fcloop_nport_put(tport->nport);
+
+ spin_lock_irqsave(&fcloop_lock, flags);
+ if (!test_and_set_bit(PORT_DELETED, &tport->flags))
+ put_port = true;
+ tport->nport->tport = NULL;
+ spin_unlock_irqrestore(&fcloop_lock, flags);
+
+ if (put_port)
+ fcloop_nport_put(tport->nport);
}
#define FCLOOP_HW_QUEUES 4
@@ -1070,7 +1153,6 @@ static struct nvme_fc_port_template fctemplate = {
/* sizes of additional private data for data structures */
.local_priv_sz = sizeof(struct fcloop_lport_priv),
.remote_priv_sz = sizeof(struct fcloop_rport),
- .lsrqst_priv_sz = sizeof(struct fcloop_lsreq),
.fcprqst_priv_sz = sizeof(struct fcloop_ini_fcpreq),
};
@@ -1093,7 +1175,6 @@ static struct nvmet_fc_target_template tgttemplate = {
.target_features = 0,
/* sizes of additional private data for data structures */
.target_priv_sz = sizeof(struct fcloop_tport),
- .lsrqst_priv_sz = sizeof(struct fcloop_lsreq),
};
static ssize_t
@@ -1140,6 +1221,7 @@ fcloop_create_local_port(struct device *dev, struct device_attribute *attr,
lport->localport = localport;
INIT_LIST_HEAD(&lport->lport_list);
+ refcount_set(&lport->ref, 1);
spin_lock_irqsave(&fcloop_lock, flags);
list_add_tail(&lport->lport_list, &fcloop_lports);
@@ -1156,60 +1238,94 @@ out_free_lport:
return ret ? ret : count;
}
+static int
+__localport_unreg(struct fcloop_lport *lport)
+{
+ return nvme_fc_unregister_localport(lport->localport);
+}
-static void
-__unlink_local_port(struct fcloop_lport *lport)
+static struct fcloop_nport *
+__fcloop_nport_lookup(u64 node_name, u64 port_name)
{
- list_del(&lport->lport_list);
+ struct fcloop_nport *nport;
+
+ list_for_each_entry(nport, &fcloop_nports, nport_list) {
+ if (nport->node_name != node_name ||
+ nport->port_name != port_name)
+ continue;
+
+ if (fcloop_nport_get(nport))
+ return nport;
+
+ break;
+ }
+
+ return NULL;
}
-static int
-__wait_localport_unreg(struct fcloop_lport *lport)
+static struct fcloop_nport *
+fcloop_nport_lookup(u64 node_name, u64 port_name)
{
- int ret;
+ struct fcloop_nport *nport;
+ unsigned long flags;
- init_completion(&lport->unreg_done);
+ spin_lock_irqsave(&fcloop_lock, flags);
+ nport = __fcloop_nport_lookup(node_name, port_name);
+ spin_unlock_irqrestore(&fcloop_lock, flags);
- ret = nvme_fc_unregister_localport(lport->localport);
+ return nport;
+}
- if (!ret)
- wait_for_completion(&lport->unreg_done);
+static struct fcloop_lport *
+__fcloop_lport_lookup(u64 node_name, u64 port_name)
+{
+ struct fcloop_lport *lport;
- kfree(lport);
+ list_for_each_entry(lport, &fcloop_lports, lport_list) {
+ if (lport->localport->node_name != node_name ||
+ lport->localport->port_name != port_name)
+ continue;
- return ret;
+ if (fcloop_lport_get(lport))
+ return lport;
+
+ break;
+ }
+
+ return NULL;
}
+static struct fcloop_lport *
+fcloop_lport_lookup(u64 node_name, u64 port_name)
+{
+ struct fcloop_lport *lport;
+ unsigned long flags;
+
+ spin_lock_irqsave(&fcloop_lock, flags);
+ lport = __fcloop_lport_lookup(node_name, port_name);
+ spin_unlock_irqrestore(&fcloop_lock, flags);
+
+ return lport;
+}
static ssize_t
fcloop_delete_local_port(struct device *dev, struct device_attribute *attr,
const char *buf, size_t count)
{
- struct fcloop_lport *tlport, *lport = NULL;
+ struct fcloop_lport *lport;
u64 nodename, portname;
- unsigned long flags;
int ret;
ret = fcloop_parse_nm_options(dev, &nodename, &portname, buf);
if (ret)
return ret;
- spin_lock_irqsave(&fcloop_lock, flags);
-
- list_for_each_entry(tlport, &fcloop_lports, lport_list) {
- if (tlport->localport->node_name == nodename &&
- tlport->localport->port_name == portname) {
- lport = tlport;
- __unlink_local_port(lport);
- break;
- }
- }
- spin_unlock_irqrestore(&fcloop_lock, flags);
-
+ lport = fcloop_lport_lookup(nodename, portname);
if (!lport)
return -ENOENT;
- ret = __wait_localport_unreg(lport);
+ ret = __localport_unreg(lport);
+ fcloop_lport_put(lport);
return ret ? ret : count;
}
@@ -1217,8 +1333,8 @@ fcloop_delete_local_port(struct device *dev, struct device_attribute *attr,
static struct fcloop_nport *
fcloop_alloc_nport(const char *buf, size_t count, bool remoteport)
{
- struct fcloop_nport *newnport, *nport = NULL;
- struct fcloop_lport *tmplport, *lport = NULL;
+ struct fcloop_nport *newnport, *nport;
+ struct fcloop_lport *lport;
struct fcloop_ctrl_options *opts;
unsigned long flags;
u32 opts_mask = (remoteport) ? RPORT_OPTS : TGTPORT_OPTS;
@@ -1233,10 +1349,8 @@ fcloop_alloc_nport(const char *buf, size_t count, bool remoteport)
goto out_free_opts;
/* everything there ? */
- if ((opts->mask & opts_mask) != opts_mask) {
- ret = -EINVAL;
+ if ((opts->mask & opts_mask) != opts_mask)
goto out_free_opts;
- }
newnport = kzalloc(sizeof(*newnport), GFP_KERNEL);
if (!newnport)
@@ -1249,63 +1363,64 @@ fcloop_alloc_nport(const char *buf, size_t count, bool remoteport)
newnport->port_role = opts->roles;
if (opts->mask & NVMF_OPT_FCADDR)
newnport->port_id = opts->fcaddr;
- kref_init(&newnport->ref);
+ refcount_set(&newnport->ref, 1);
spin_lock_irqsave(&fcloop_lock, flags);
-
- list_for_each_entry(tmplport, &fcloop_lports, lport_list) {
- if (tmplport->localport->node_name == opts->wwnn &&
- tmplport->localport->port_name == opts->wwpn)
- goto out_invalid_opts;
-
- if (tmplport->localport->node_name == opts->lpwwnn &&
- tmplport->localport->port_name == opts->lpwwpn)
- lport = tmplport;
+ lport = __fcloop_lport_lookup(opts->wwnn, opts->wwpn);
+ if (lport) {
+ /* invalid configuration */
+ fcloop_lport_put(lport);
+ goto out_free_newnport;
}
if (remoteport) {
- if (!lport)
- goto out_invalid_opts;
- newnport->lport = lport;
- }
-
- list_for_each_entry(nport, &fcloop_nports, nport_list) {
- if (nport->node_name == opts->wwnn &&
- nport->port_name == opts->wwpn) {
- if ((remoteport && nport->rport) ||
- (!remoteport && nport->tport)) {
- nport = NULL;
- goto out_invalid_opts;
- }
-
- fcloop_nport_get(nport);
-
- spin_unlock_irqrestore(&fcloop_lock, flags);
-
- if (remoteport)
- nport->lport = lport;
- if (opts->mask & NVMF_OPT_ROLES)
- nport->port_role = opts->roles;
- if (opts->mask & NVMF_OPT_FCADDR)
- nport->port_id = opts->fcaddr;
+ lport = __fcloop_lport_lookup(opts->lpwwnn, opts->lpwwpn);
+ if (!lport) {
+ /* invalid configuration */
goto out_free_newnport;
}
}
- list_add_tail(&newnport->nport_list, &fcloop_nports);
+ nport = __fcloop_nport_lookup(opts->wwnn, opts->wwpn);
+ if (nport) {
+ if ((remoteport && nport->rport) ||
+ (!remoteport && nport->tport)) {
+ /* invalid configuration */
+ goto out_put_nport;
+ }
+
+ /* found existing nport, discard the new nport */
+ kfree(newnport);
+ } else {
+ list_add_tail(&newnport->nport_list, &fcloop_nports);
+ nport = newnport;
+ }
+ if (opts->mask & NVMF_OPT_ROLES)
+ nport->port_role = opts->roles;
+ if (opts->mask & NVMF_OPT_FCADDR)
+ nport->port_id = opts->fcaddr;
+ if (lport) {
+ if (!nport->lport)
+ nport->lport = lport;
+ else
+ fcloop_lport_put(lport);
+ }
spin_unlock_irqrestore(&fcloop_lock, flags);
kfree(opts);
- return newnport;
+ return nport;
-out_invalid_opts:
- spin_unlock_irqrestore(&fcloop_lock, flags);
+out_put_nport:
+ if (lport)
+ fcloop_lport_put(lport);
+ fcloop_nport_put(nport);
out_free_newnport:
+ spin_unlock_irqrestore(&fcloop_lock, flags);
kfree(newnport);
out_free_opts:
kfree(opts);
- return nport;
+ return NULL;
}
static ssize_t
@@ -1346,6 +1461,7 @@ fcloop_create_remote_port(struct device *dev, struct device_attribute *attr,
rport->nport = nport;
rport->lport = nport->lport;
nport->rport = rport;
+ rport->flags = 0;
spin_lock_init(&rport->lock);
INIT_WORK(&rport->ls_work, fcloop_rport_lsrqst_work);
INIT_LIST_HEAD(&rport->ls_list);
@@ -1359,21 +1475,18 @@ __unlink_remote_port(struct fcloop_nport *nport)
{
struct fcloop_rport *rport = nport->rport;
+ lockdep_assert_held(&fcloop_lock);
+
if (rport && nport->tport)
nport->tport->remoteport = NULL;
nport->rport = NULL;
- list_del(&nport->nport_list);
-
return rport;
}
static int
__remoteport_unreg(struct fcloop_nport *nport, struct fcloop_rport *rport)
{
- if (!rport)
- return -EALREADY;
-
return nvme_fc_unregister_remoteport(rport->remoteport);
}
@@ -1381,8 +1494,8 @@ static ssize_t
fcloop_delete_remote_port(struct device *dev, struct device_attribute *attr,
const char *buf, size_t count)
{
- struct fcloop_nport *nport = NULL, *tmpport;
- static struct fcloop_rport *rport;
+ struct fcloop_nport *nport;
+ struct fcloop_rport *rport;
u64 nodename, portname;
unsigned long flags;
int ret;
@@ -1391,24 +1504,24 @@ fcloop_delete_remote_port(struct device *dev, struct device_attribute *attr,
if (ret)
return ret;
- spin_lock_irqsave(&fcloop_lock, flags);
-
- list_for_each_entry(tmpport, &fcloop_nports, nport_list) {
- if (tmpport->node_name == nodename &&
- tmpport->port_name == portname && tmpport->rport) {
- nport = tmpport;
- rport = __unlink_remote_port(nport);
- break;
- }
- }
+ nport = fcloop_nport_lookup(nodename, portname);
+ if (!nport)
+ return -ENOENT;
+ spin_lock_irqsave(&fcloop_lock, flags);
+ rport = __unlink_remote_port(nport);
spin_unlock_irqrestore(&fcloop_lock, flags);
- if (!nport)
- return -ENOENT;
+ if (!rport) {
+ ret = -ENOENT;
+ goto out_nport_put;
+ }
ret = __remoteport_unreg(nport, rport);
+out_nport_put:
+ fcloop_nport_put(nport);
+
return ret ? ret : count;
}
@@ -1446,6 +1559,7 @@ fcloop_create_target_port(struct device *dev, struct device_attribute *attr,
tport->nport = nport;
tport->lport = nport->lport;
nport->tport = tport;
+ tport->flags = 0;
spin_lock_init(&tport->lock);
INIT_WORK(&tport->ls_work, fcloop_tport_lsrqst_work);
INIT_LIST_HEAD(&tport->ls_list);
@@ -1459,6 +1573,8 @@ __unlink_target_port(struct fcloop_nport *nport)
{
struct fcloop_tport *tport = nport->tport;
+ lockdep_assert_held(&fcloop_lock);
+
if (tport && nport->rport)
nport->rport->targetport = NULL;
nport->tport = NULL;
@@ -1469,9 +1585,6 @@ __unlink_target_port(struct fcloop_nport *nport)
static int
__targetport_unreg(struct fcloop_nport *nport, struct fcloop_tport *tport)
{
- if (!tport)
- return -EALREADY;
-
return nvmet_fc_unregister_targetport(tport->targetport);
}
@@ -1479,8 +1592,8 @@ static ssize_t
fcloop_delete_target_port(struct device *dev, struct device_attribute *attr,
const char *buf, size_t count)
{
- struct fcloop_nport *nport = NULL, *tmpport;
- struct fcloop_tport *tport = NULL;
+ struct fcloop_nport *nport;
+ struct fcloop_tport *tport;
u64 nodename, portname;
unsigned long flags;
int ret;
@@ -1489,24 +1602,24 @@ fcloop_delete_target_port(struct device *dev, struct device_attribute *attr,
if (ret)
return ret;
- spin_lock_irqsave(&fcloop_lock, flags);
-
- list_for_each_entry(tmpport, &fcloop_nports, nport_list) {
- if (tmpport->node_name == nodename &&
- tmpport->port_name == portname && tmpport->tport) {
- nport = tmpport;
- tport = __unlink_target_port(nport);
- break;
- }
- }
+ nport = fcloop_nport_lookup(nodename, portname);
+ if (!nport)
+ return -ENOENT;
+ spin_lock_irqsave(&fcloop_lock, flags);
+ tport = __unlink_target_port(nport);
spin_unlock_irqrestore(&fcloop_lock, flags);
- if (!nport)
- return -ENOENT;
+ if (!tport) {
+ ret = -ENOENT;
+ goto out_nport_put;
+ }
ret = __targetport_unreg(nport, tport);
+out_nport_put:
+ fcloop_nport_put(nport);
+
return ret ? ret : count;
}
@@ -1572,15 +1685,20 @@ static const struct class fcloop_class = {
};
static struct device *fcloop_device;
-
static int __init fcloop_init(void)
{
int ret;
+ lsreq_cache = kmem_cache_create("lsreq_cache",
+ sizeof(struct fcloop_lsreq), 0,
+ 0, NULL);
+ if (!lsreq_cache)
+ return -ENOMEM;
+
ret = class_register(&fcloop_class);
if (ret) {
pr_err("couldn't register class fcloop\n");
- return ret;
+ goto out_destroy_cache;
}
fcloop_device = device_create_with_groups(
@@ -1598,13 +1716,15 @@ static int __init fcloop_init(void)
out_destroy_class:
class_unregister(&fcloop_class);
+out_destroy_cache:
+ kmem_cache_destroy(lsreq_cache);
return ret;
}
static void __exit fcloop_exit(void)
{
- struct fcloop_lport *lport = NULL;
- struct fcloop_nport *nport = NULL;
+ struct fcloop_lport *lport;
+ struct fcloop_nport *nport;
struct fcloop_tport *tport;
struct fcloop_rport *rport;
unsigned long flags;
@@ -1615,7 +1735,7 @@ static void __exit fcloop_exit(void)
for (;;) {
nport = list_first_entry_or_null(&fcloop_nports,
typeof(*nport), nport_list);
- if (!nport)
+ if (!nport || !fcloop_nport_get(nport))
break;
tport = __unlink_target_port(nport);
@@ -1623,13 +1743,21 @@ static void __exit fcloop_exit(void)
spin_unlock_irqrestore(&fcloop_lock, flags);
- ret = __targetport_unreg(nport, tport);
- if (ret)
- pr_warn("%s: Failed deleting target port\n", __func__);
+ if (tport) {
+ ret = __targetport_unreg(nport, tport);
+ if (ret)
+ pr_warn("%s: Failed deleting target port\n",
+ __func__);
+ }
- ret = __remoteport_unreg(nport, rport);
- if (ret)
- pr_warn("%s: Failed deleting remote port\n", __func__);
+ if (rport) {
+ ret = __remoteport_unreg(nport, rport);
+ if (ret)
+ pr_warn("%s: Failed deleting remote port\n",
+ __func__);
+ }
+
+ fcloop_nport_put(nport);
spin_lock_irqsave(&fcloop_lock, flags);
}
@@ -1637,17 +1765,17 @@ static void __exit fcloop_exit(void)
for (;;) {
lport = list_first_entry_or_null(&fcloop_lports,
typeof(*lport), lport_list);
- if (!lport)
+ if (!lport || !fcloop_lport_get(lport))
break;
- __unlink_local_port(lport);
-
spin_unlock_irqrestore(&fcloop_lock, flags);
- ret = __wait_localport_unreg(lport);
+ ret = __localport_unreg(lport);
if (ret)
pr_warn("%s: Failed deleting local port\n", __func__);
+ fcloop_lport_put(lport);
+
spin_lock_irqsave(&fcloop_lock, flags);
}
@@ -1657,6 +1785,7 @@ static void __exit fcloop_exit(void)
device_destroy(&fcloop_class, MKDEV(0, 0));
class_unregister(&fcloop_class);
+ kmem_cache_destroy(lsreq_cache);
}
module_init(fcloop_init);
diff --git a/drivers/nvme/target/io-cmd-bdev.c b/drivers/nvme/target/io-cmd-bdev.c
index 83be0657e6df..eba42df2f821 100644
--- a/drivers/nvme/target/io-cmd-bdev.c
+++ b/drivers/nvme/target/io-cmd-bdev.c
@@ -133,7 +133,7 @@ u16 blk_to_nvme_status(struct nvmet_req *req, blk_status_t blk_sts)
* Right now there exists M : 1 mapping between block layer error
* to the NVMe status code (see nvme_error_status()). For consistency,
* when we reverse map we use most appropriate NVMe Status code from
- * the group of the NVMe staus codes used in the nvme_error_status().
+ * the group of the NVMe status codes used in the nvme_error_status().
*/
switch (blk_sts) {
case BLK_STS_NOSPC:
@@ -145,15 +145,8 @@ u16 blk_to_nvme_status(struct nvmet_req *req, blk_status_t blk_sts)
req->error_loc = offsetof(struct nvme_rw_command, slba);
break;
case BLK_STS_NOTSUPP:
+ status = NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR;
req->error_loc = offsetof(struct nvme_common_command, opcode);
- switch (req->cmd->common.opcode) {
- case nvme_cmd_dsm:
- case nvme_cmd_write_zeroes:
- status = NVME_SC_ONCS_NOT_SUPPORTED | NVME_STATUS_DNR;
- break;
- default:
- status = NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR;
- }
break;
case BLK_STS_MEDIUM:
status = NVME_SC_ACCESS_DENIED;
diff --git a/drivers/nvme/target/loop.c b/drivers/nvme/target/loop.c
index a9d112d34d4f..f85a8441bcc6 100644
--- a/drivers/nvme/target/loop.c
+++ b/drivers/nvme/target/loop.c
@@ -33,10 +33,12 @@ struct nvme_loop_ctrl {
struct list_head list;
struct blk_mq_tag_set tag_set;
- struct nvme_loop_iod async_event_iod;
struct nvme_ctrl ctrl;
struct nvmet_port *port;
+
+ /* Must be last --ends in a flexible-array member. */
+ struct nvme_loop_iod async_event_iod;
};
static inline struct nvme_loop_ctrl *to_loop_ctrl(struct nvme_ctrl *ctrl)
@@ -148,8 +150,7 @@ static blk_status_t nvme_loop_queue_rq(struct blk_mq_hw_ctx *hctx,
nvme_start_request(req);
iod->cmd.common.flags |= NVME_CMD_SGL_METABUF;
iod->req.port = queue->ctrl->port;
- if (!nvmet_req_init(&iod->req, &queue->nvme_cq,
- &queue->nvme_sq, &nvme_loop_ops))
+ if (!nvmet_req_init(&iod->req, &queue->nvme_sq, &nvme_loop_ops))
return BLK_STS_OK;
if (blk_rq_nr_phys_segments(req)) {
@@ -162,7 +163,7 @@ static blk_status_t nvme_loop_queue_rq(struct blk_mq_hw_ctx *hctx,
}
iod->req.sg = iod->sg_table.sgl;
- iod->req.sg_cnt = blk_rq_map_sg(req->q, req, iod->sg_table.sgl);
+ iod->req.sg_cnt = blk_rq_map_sg(req, iod->sg_table.sgl);
iod->req.transfer_len = blk_rq_payload_bytes(req);
}
@@ -181,8 +182,7 @@ static void nvme_loop_submit_async_event(struct nvme_ctrl *arg)
iod->cmd.common.command_id = NVME_AQ_BLK_MQ_DEPTH;
iod->cmd.common.flags |= NVME_CMD_SGL_METABUF;
- if (!nvmet_req_init(&iod->req, &queue->nvme_cq, &queue->nvme_sq,
- &nvme_loop_ops)) {
+ if (!nvmet_req_init(&iod->req, &queue->nvme_sq, &nvme_loop_ops)) {
dev_err(ctrl->ctrl.device, "failed async event work\n");
return;
}
@@ -273,6 +273,7 @@ static void nvme_loop_destroy_admin_queue(struct nvme_loop_ctrl *ctrl)
nvme_unquiesce_admin_queue(&ctrl->ctrl);
nvmet_sq_destroy(&ctrl->queues[0].nvme_sq);
+ nvmet_cq_put(&ctrl->queues[0].nvme_cq);
nvme_remove_admin_tag_set(&ctrl->ctrl);
}
@@ -302,6 +303,7 @@ static void nvme_loop_destroy_io_queues(struct nvme_loop_ctrl *ctrl)
for (i = 1; i < ctrl->ctrl.queue_count; i++) {
clear_bit(NVME_LOOP_Q_LIVE, &ctrl->queues[i].flags);
nvmet_sq_destroy(&ctrl->queues[i].nvme_sq);
+ nvmet_cq_put(&ctrl->queues[i].nvme_cq);
}
ctrl->ctrl.queue_count = 1;
/*
@@ -327,9 +329,13 @@ static int nvme_loop_init_io_queues(struct nvme_loop_ctrl *ctrl)
for (i = 1; i <= nr_io_queues; i++) {
ctrl->queues[i].ctrl = ctrl;
- ret = nvmet_sq_init(&ctrl->queues[i].nvme_sq);
- if (ret)
+ nvmet_cq_init(&ctrl->queues[i].nvme_cq);
+ ret = nvmet_sq_init(&ctrl->queues[i].nvme_sq,
+ &ctrl->queues[i].nvme_cq);
+ if (ret) {
+ nvmet_cq_put(&ctrl->queues[i].nvme_cq);
goto out_destroy_queues;
+ }
ctrl->ctrl.queue_count++;
}
@@ -360,9 +366,13 @@ static int nvme_loop_configure_admin_queue(struct nvme_loop_ctrl *ctrl)
int error;
ctrl->queues[0].ctrl = ctrl;
- error = nvmet_sq_init(&ctrl->queues[0].nvme_sq);
- if (error)
+ nvmet_cq_init(&ctrl->queues[0].nvme_cq);
+ error = nvmet_sq_init(&ctrl->queues[0].nvme_sq,
+ &ctrl->queues[0].nvme_cq);
+ if (error) {
+ nvmet_cq_put(&ctrl->queues[0].nvme_cq);
return error;
+ }
ctrl->ctrl.queue_count = 1;
error = nvme_alloc_admin_tag_set(&ctrl->ctrl, &ctrl->admin_tag_set,
@@ -401,6 +411,7 @@ out_cleanup_tagset:
nvme_remove_admin_tag_set(&ctrl->ctrl);
out_free_sq:
nvmet_sq_destroy(&ctrl->queues[0].nvme_sq);
+ nvmet_cq_put(&ctrl->queues[0].nvme_cq);
return error;
}
diff --git a/drivers/nvme/target/nvmet.h b/drivers/nvme/target/nvmet.h
index 4be8d22d2d8d..df69a9dee71c 100644
--- a/drivers/nvme/target/nvmet.h
+++ b/drivers/nvme/target/nvmet.h
@@ -141,13 +141,16 @@ static inline struct device *nvmet_ns_dev(struct nvmet_ns *ns)
}
struct nvmet_cq {
+ struct nvmet_ctrl *ctrl;
u16 qid;
u16 size;
+ refcount_t ref;
};
struct nvmet_sq {
struct nvmet_ctrl *ctrl;
struct percpu_ref ref;
+ struct nvmet_cq *cq;
u16 qid;
u16 size;
u32 sqhd;
@@ -165,6 +168,9 @@ struct nvmet_sq {
u8 *dhchap_skey;
int dhchap_skey_len;
#endif
+#ifdef CONFIG_NVME_TARGET_TCP_TLS
+ struct key *tls_key;
+#endif
struct completion free_done;
struct completion confirm_done;
};
@@ -244,6 +250,7 @@ struct nvmet_pr_log_mgr {
struct nvmet_ctrl {
struct nvmet_subsys *subsys;
struct nvmet_sq **sqs;
+ struct nvmet_cq **cqs;
void *drvdata;
@@ -289,6 +296,7 @@ struct nvmet_ctrl {
u64 err_counter;
struct nvme_error_slot slots[NVMET_ERROR_LOG_SLOTS];
bool pi_support;
+ bool concat;
#ifdef CONFIG_NVME_TARGET_AUTH
struct nvme_dhchap_key *host_key;
struct nvme_dhchap_key *ctrl_key;
@@ -298,6 +306,9 @@ struct nvmet_ctrl {
u8 *dh_key;
size_t dh_keysize;
#endif
+#ifdef CONFIG_NVME_TARGET_TCP_TLS
+ struct key *tls_key;
+#endif
struct nvmet_pr_log_mgr pr_log_mgr;
};
@@ -417,7 +428,7 @@ struct nvmet_fabrics_ops {
u16 (*get_max_queue_size)(const struct nvmet_ctrl *ctrl);
/* Operations mandatory for PCI target controllers */
- u16 (*create_sq)(struct nvmet_ctrl *ctrl, u16 sqid, u16 flags,
+ u16 (*create_sq)(struct nvmet_ctrl *ctrl, u16 sqid, u16 cqid, u16 flags,
u16 qsize, u64 prp1);
u16 (*delete_sq)(struct nvmet_ctrl *ctrl, u16 sqid);
u16 (*create_cq)(struct nvmet_ctrl *ctrl, u16 cqid, u16 flags,
@@ -550,8 +561,8 @@ u32 nvmet_fabrics_admin_cmd_data_len(struct nvmet_req *req);
u16 nvmet_parse_fabrics_io_cmd(struct nvmet_req *req);
u32 nvmet_fabrics_io_cmd_data_len(struct nvmet_req *req);
-bool nvmet_req_init(struct nvmet_req *req, struct nvmet_cq *cq,
- struct nvmet_sq *sq, const struct nvmet_fabrics_ops *ops);
+bool nvmet_req_init(struct nvmet_req *req, struct nvmet_sq *sq,
+ const struct nvmet_fabrics_ops *ops);
void nvmet_req_uninit(struct nvmet_req *req);
size_t nvmet_req_transfer_len(struct nvmet_req *req);
bool nvmet_check_transfer_len(struct nvmet_req *req, size_t len);
@@ -564,18 +575,24 @@ void nvmet_execute_set_features(struct nvmet_req *req);
void nvmet_execute_get_features(struct nvmet_req *req);
void nvmet_execute_keep_alive(struct nvmet_req *req);
-u16 nvmet_check_cqid(struct nvmet_ctrl *ctrl, u16 cqid);
+u16 nvmet_check_cqid(struct nvmet_ctrl *ctrl, u16 cqid, bool create);
+u16 nvmet_check_io_cqid(struct nvmet_ctrl *ctrl, u16 cqid, bool create);
+void nvmet_cq_init(struct nvmet_cq *cq);
void nvmet_cq_setup(struct nvmet_ctrl *ctrl, struct nvmet_cq *cq, u16 qid,
u16 size);
u16 nvmet_cq_create(struct nvmet_ctrl *ctrl, struct nvmet_cq *cq, u16 qid,
u16 size);
+void nvmet_cq_destroy(struct nvmet_cq *cq);
+bool nvmet_cq_get(struct nvmet_cq *cq);
+void nvmet_cq_put(struct nvmet_cq *cq);
+bool nvmet_cq_in_use(struct nvmet_cq *cq);
u16 nvmet_check_sqid(struct nvmet_ctrl *ctrl, u16 sqid, bool create);
void nvmet_sq_setup(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq, u16 qid,
u16 size);
-u16 nvmet_sq_create(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq, u16 qid,
- u16 size);
+u16 nvmet_sq_create(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq,
+ struct nvmet_cq *cq, u16 qid, u16 size);
void nvmet_sq_destroy(struct nvmet_sq *sq);
-int nvmet_sq_init(struct nvmet_sq *sq);
+int nvmet_sq_init(struct nvmet_sq *sq, struct nvmet_cq *cq);
void nvmet_ctrl_fatal_error(struct nvmet_ctrl *ctrl);
@@ -583,6 +600,7 @@ void nvmet_update_cc(struct nvmet_ctrl *ctrl, u32 new);
struct nvmet_alloc_ctrl_args {
struct nvmet_port *port;
+ struct nvmet_sq *sq;
char *subsysnqn;
char *hostnqn;
uuid_t *hostid;
@@ -647,7 +665,6 @@ void nvmet_subsys_disc_changed(struct nvmet_subsys *subsys,
struct nvmet_host *host);
void nvmet_add_async_event(struct nvmet_ctrl *ctrl, u8 event_type,
u8 event_info, u8 log_page);
-bool nvmet_subsys_nsid_exists(struct nvmet_subsys *subsys, u32 nsid);
#define NVMET_MIN_QUEUE_SIZE 16
#define NVMET_MAX_QUEUE_SIZE 1024
@@ -784,43 +801,43 @@ u16 nvmet_report_invalid_opcode(struct nvmet_req *req);
static inline bool nvmet_cc_en(u32 cc)
{
- return (cc >> NVME_CC_EN_SHIFT) & 0x1;
+ return (cc & NVME_CC_ENABLE) >> NVME_CC_EN_SHIFT;
}
static inline u8 nvmet_cc_css(u32 cc)
{
- return (cc >> NVME_CC_CSS_SHIFT) & 0x7;
+ return (cc & NVME_CC_CSS_MASK) >> NVME_CC_CSS_SHIFT;
}
static inline u8 nvmet_cc_mps(u32 cc)
{
- return (cc >> NVME_CC_MPS_SHIFT) & 0xf;
+ return (cc & NVME_CC_MPS_MASK) >> NVME_CC_MPS_SHIFT;
}
static inline u8 nvmet_cc_ams(u32 cc)
{
- return (cc >> NVME_CC_AMS_SHIFT) & 0x7;
+ return (cc & NVME_CC_AMS_MASK) >> NVME_CC_AMS_SHIFT;
}
static inline u8 nvmet_cc_shn(u32 cc)
{
- return (cc >> NVME_CC_SHN_SHIFT) & 0x3;
+ return (cc & NVME_CC_SHN_MASK) >> NVME_CC_SHN_SHIFT;
}
static inline u8 nvmet_cc_iosqes(u32 cc)
{
- return (cc >> NVME_CC_IOSQES_SHIFT) & 0xf;
+ return (cc & NVME_CC_IOSQES_MASK) >> NVME_CC_IOSQES_SHIFT;
}
static inline u8 nvmet_cc_iocqes(u32 cc)
{
- return (cc >> NVME_CC_IOCQES_SHIFT) & 0xf;
+ return (cc & NVME_CC_IOCQES_MASK) >> NVME_CC_IOCQES_SHIFT;
}
/* Convert a 32-bit number to a 16-bit 0's based number */
static inline __le16 to0based(u32 a)
{
- return cpu_to_le16(max(1U, min(1U << 16, a)) - 1);
+ return cpu_to_le16(clamp(a, 1U, 1U << 16) - 1);
}
static inline bool nvmet_ns_has_pi(struct nvmet_ns *ns)
@@ -852,6 +869,22 @@ static inline void nvmet_req_bio_put(struct nvmet_req *req, struct bio *bio)
bio_put(bio);
}
+#ifdef CONFIG_NVME_TARGET_TCP_TLS
+static inline key_serial_t nvmet_queue_tls_keyid(struct nvmet_sq *sq)
+{
+ return sq->tls_key ? key_serial(sq->tls_key) : 0;
+}
+static inline void nvmet_sq_put_tls_key(struct nvmet_sq *sq)
+{
+ if (sq->tls_key) {
+ key_put(sq->tls_key);
+ sq->tls_key = NULL;
+ }
+}
+#else
+static inline key_serial_t nvmet_queue_tls_keyid(struct nvmet_sq *sq) { return 0; }
+static inline void nvmet_sq_put_tls_key(struct nvmet_sq *sq) {}
+#endif
#ifdef CONFIG_NVME_TARGET_AUTH
u32 nvmet_auth_send_data_len(struct nvmet_req *req);
void nvmet_execute_auth_send(struct nvmet_req *req);
@@ -860,7 +893,7 @@ void nvmet_execute_auth_receive(struct nvmet_req *req);
int nvmet_auth_set_key(struct nvmet_host *host, const char *secret,
bool set_ctrl);
int nvmet_auth_set_host_hash(struct nvmet_host *host, const char *hash);
-u8 nvmet_setup_auth(struct nvmet_ctrl *ctrl);
+u8 nvmet_setup_auth(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq);
void nvmet_auth_sq_init(struct nvmet_sq *sq);
void nvmet_destroy_auth(struct nvmet_ctrl *ctrl);
void nvmet_auth_sq_free(struct nvmet_sq *sq);
@@ -870,16 +903,18 @@ int nvmet_auth_host_hash(struct nvmet_req *req, u8 *response,
unsigned int hash_len);
int nvmet_auth_ctrl_hash(struct nvmet_req *req, u8 *response,
unsigned int hash_len);
-static inline bool nvmet_has_auth(struct nvmet_ctrl *ctrl)
+static inline bool nvmet_has_auth(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq)
{
- return ctrl->host_key != NULL;
+ return ctrl->host_key != NULL && !nvmet_queue_tls_keyid(sq);
}
int nvmet_auth_ctrl_exponential(struct nvmet_req *req,
u8 *buf, int buf_size);
int nvmet_auth_ctrl_sesskey(struct nvmet_req *req,
u8 *buf, int buf_size);
+void nvmet_auth_insert_psk(struct nvmet_sq *sq);
#else
-static inline u8 nvmet_setup_auth(struct nvmet_ctrl *ctrl)
+static inline u8 nvmet_setup_auth(struct nvmet_ctrl *ctrl,
+ struct nvmet_sq *sq)
{
return 0;
}
@@ -892,11 +927,13 @@ static inline bool nvmet_check_auth_status(struct nvmet_req *req)
{
return true;
}
-static inline bool nvmet_has_auth(struct nvmet_ctrl *ctrl)
+static inline bool nvmet_has_auth(struct nvmet_ctrl *ctrl,
+ struct nvmet_sq *sq)
{
return false;
}
static inline const char *nvmet_dhchap_dhgroup_name(u8 dhgid) { return NULL; }
+static inline void nvmet_auth_insert_psk(struct nvmet_sq *sq) {};
#endif
int nvmet_pr_init_ns(struct nvmet_ns *ns);
diff --git a/drivers/nvme/target/passthru.c b/drivers/nvme/target/passthru.c
index 26e2907ce8bb..b7515c53829b 100644
--- a/drivers/nvme/target/passthru.c
+++ b/drivers/nvme/target/passthru.c
@@ -99,7 +99,7 @@ static u16 nvmet_passthru_override_id_ctrl(struct nvmet_req *req)
/*
* The passthru NVMe driver may have a limit on the number of segments
- * which depends on the host's memory fragementation. To solve this,
+ * which depends on the host's memory fragmentation. To solve this,
* ensure mdts is limited to the pages equal to the number of segments.
*/
max_hw_sectors = min_not_zero(pctrl->max_segments << PAGE_SECTORS_SHIFT,
diff --git a/drivers/nvme/target/pci-epf.c b/drivers/nvme/target/pci-epf.c
index ac30b42cc622..a4295a5b8d28 100644
--- a/drivers/nvme/target/pci-epf.c
+++ b/drivers/nvme/target/pci-epf.c
@@ -46,7 +46,7 @@ static DEFINE_MUTEX(nvmet_pci_epf_ports_mutex);
/*
* BAR CC register and SQ polling intervals.
*/
-#define NVMET_PCI_EPF_CC_POLL_INTERVAL msecs_to_jiffies(5)
+#define NVMET_PCI_EPF_CC_POLL_INTERVAL msecs_to_jiffies(10)
#define NVMET_PCI_EPF_SQ_POLL_INTERVAL msecs_to_jiffies(5)
#define NVMET_PCI_EPF_SQ_POLL_IDLE msecs_to_jiffies(5000)
@@ -62,8 +62,7 @@ static DEFINE_MUTEX(nvmet_pci_epf_ports_mutex);
#define NVMET_PCI_EPF_CQ_RETRY_INTERVAL msecs_to_jiffies(1)
enum nvmet_pci_epf_queue_flags {
- NVMET_PCI_EPF_Q_IS_SQ = 0, /* The queue is a submission queue */
- NVMET_PCI_EPF_Q_LIVE, /* The queue is live */
+ NVMET_PCI_EPF_Q_LIVE = 0, /* The queue is live */
NVMET_PCI_EPF_Q_IRQ_ENABLED, /* IRQ is enabled for this queue */
};
@@ -596,9 +595,6 @@ static bool nvmet_pci_epf_should_raise_irq(struct nvmet_pci_epf_ctrl *ctrl,
struct nvmet_pci_epf_irq_vector *iv = cq->iv;
bool ret;
- if (!test_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags))
- return false;
-
/* IRQ coalescing for the admin queue is not allowed. */
if (!cq->qid)
return true;
@@ -625,7 +621,8 @@ static void nvmet_pci_epf_raise_irq(struct nvmet_pci_epf_ctrl *ctrl,
struct pci_epf *epf = nvme_epf->epf;
int ret = 0;
- if (!test_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags))
+ if (!test_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags) ||
+ !test_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags))
return;
mutex_lock(&ctrl->irq_lock);
@@ -636,14 +633,16 @@ static void nvmet_pci_epf_raise_irq(struct nvmet_pci_epf_ctrl *ctrl,
switch (nvme_epf->irq_type) {
case PCI_IRQ_MSIX:
case PCI_IRQ_MSI:
+ /*
+ * If we fail to raise an MSI or MSI-X interrupt, it is likely
+ * because the host is using legacy INTX IRQs (e.g. BIOS,
+ * grub), but we can fallback to the INTX type only if the
+ * endpoint controller supports this type.
+ */
ret = pci_epc_raise_irq(epf->epc, epf->func_no, epf->vfunc_no,
nvme_epf->irq_type, cq->vector + 1);
- if (!ret)
+ if (!ret || !nvme_epf->epc_features->intx_capable)
break;
- /*
- * If we got an error, it is likely because the host is using
- * legacy IRQs (e.g. BIOS, grub).
- */
fallthrough;
case PCI_IRQ_INTX:
ret = pci_epc_raise_irq(epf->epc, epf->func_no, epf->vfunc_no,
@@ -656,7 +655,9 @@ static void nvmet_pci_epf_raise_irq(struct nvmet_pci_epf_ctrl *ctrl,
}
if (ret)
- dev_err(ctrl->dev, "Failed to raise IRQ (err=%d)\n", ret);
+ dev_err_ratelimited(ctrl->dev,
+ "CQ[%u]: Failed to raise IRQ (err=%d)\n",
+ cq->qid, ret);
unlock:
mutex_unlock(&ctrl->irq_lock);
@@ -1264,16 +1265,14 @@ static u16 nvmet_pci_epf_create_cq(struct nvmet_ctrl *tctrl,
struct nvmet_pci_epf_ctrl *ctrl = tctrl->drvdata;
struct nvmet_pci_epf_queue *cq = &ctrl->cq[cqid];
u16 status;
+ int ret;
- if (test_and_set_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags))
+ if (test_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags))
return NVME_SC_QID_INVALID | NVME_STATUS_DNR;
if (!(flags & NVME_QUEUE_PHYS_CONTIG))
return NVME_SC_INVALID_QUEUE | NVME_STATUS_DNR;
- if (flags & NVME_CQ_IRQ_ENABLED)
- set_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags);
-
cq->pci_addr = pci_addr;
cq->qid = cqid;
cq->depth = qsize + 1;
@@ -1290,24 +1289,55 @@ static u16 nvmet_pci_epf_create_cq(struct nvmet_ctrl *tctrl,
cq->qes = ctrl->io_cqes;
cq->pci_size = cq->qes * cq->depth;
- cq->iv = nvmet_pci_epf_add_irq_vector(ctrl, vector);
- if (!cq->iv) {
- status = NVME_SC_INTERNAL | NVME_STATUS_DNR;
- goto err;
+ if (flags & NVME_CQ_IRQ_ENABLED) {
+ cq->iv = nvmet_pci_epf_add_irq_vector(ctrl, vector);
+ if (!cq->iv)
+ return NVME_SC_INTERNAL | NVME_STATUS_DNR;
+ set_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags);
}
status = nvmet_cq_create(tctrl, &cq->nvme_cq, cqid, cq->depth);
if (status != NVME_SC_SUCCESS)
goto err;
- dev_dbg(ctrl->dev, "CQ[%u]: %u entries of %zu B, IRQ vector %u\n",
- cqid, qsize, cq->qes, cq->vector);
+ /*
+ * Map the CQ PCI address space and since PCI endpoint controllers may
+ * return a partial mapping, check that the mapping is large enough.
+ */
+ ret = nvmet_pci_epf_mem_map(ctrl->nvme_epf, cq->pci_addr, cq->pci_size,
+ &cq->pci_map);
+ if (ret) {
+ dev_err(ctrl->dev, "Failed to map CQ %u (err=%d)\n",
+ cq->qid, ret);
+ goto err_internal;
+ }
+
+ if (cq->pci_map.pci_size < cq->pci_size) {
+ dev_err(ctrl->dev, "Invalid partial mapping of queue %u\n",
+ cq->qid);
+ goto err_unmap_queue;
+ }
+
+ set_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags);
+
+ if (test_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags))
+ dev_dbg(ctrl->dev,
+ "CQ[%u]: %u entries of %zu B, IRQ vector %u\n",
+ cqid, qsize, cq->qes, cq->vector);
+ else
+ dev_dbg(ctrl->dev,
+ "CQ[%u]: %u entries of %zu B, IRQ disabled\n",
+ cqid, qsize, cq->qes);
return NVME_SC_SUCCESS;
+err_unmap_queue:
+ nvmet_pci_epf_mem_unmap(ctrl->nvme_epf, &cq->pci_map);
+err_internal:
+ status = NVME_SC_INTERNAL | NVME_STATUS_DNR;
err:
- clear_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags);
- clear_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags);
+ if (test_and_clear_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags))
+ nvmet_pci_epf_remove_irq_vector(ctrl, cq->vector);
return status;
}
@@ -1321,19 +1351,23 @@ static u16 nvmet_pci_epf_delete_cq(struct nvmet_ctrl *tctrl, u16 cqid)
cancel_delayed_work_sync(&cq->work);
nvmet_pci_epf_drain_queue(cq);
- nvmet_pci_epf_remove_irq_vector(ctrl, cq->vector);
+ if (test_and_clear_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags))
+ nvmet_pci_epf_remove_irq_vector(ctrl, cq->vector);
+ nvmet_pci_epf_mem_unmap(ctrl->nvme_epf, &cq->pci_map);
+ nvmet_cq_put(&cq->nvme_cq);
return NVME_SC_SUCCESS;
}
static u16 nvmet_pci_epf_create_sq(struct nvmet_ctrl *tctrl,
- u16 sqid, u16 flags, u16 qsize, u64 pci_addr)
+ u16 sqid, u16 cqid, u16 flags, u16 qsize, u64 pci_addr)
{
struct nvmet_pci_epf_ctrl *ctrl = tctrl->drvdata;
struct nvmet_pci_epf_queue *sq = &ctrl->sq[sqid];
+ struct nvmet_pci_epf_queue *cq = &ctrl->cq[cqid];
u16 status;
- if (test_and_set_bit(NVMET_PCI_EPF_Q_LIVE, &sq->flags))
+ if (test_bit(NVMET_PCI_EPF_Q_LIVE, &sq->flags))
return NVME_SC_QID_INVALID | NVME_STATUS_DNR;
if (!(flags & NVME_QUEUE_PHYS_CONTIG))
@@ -1353,9 +1387,10 @@ static u16 nvmet_pci_epf_create_sq(struct nvmet_ctrl *tctrl,
sq->qes = ctrl->io_sqes;
sq->pci_size = sq->qes * sq->depth;
- status = nvmet_sq_create(tctrl, &sq->nvme_sq, sqid, sq->depth);
+ status = nvmet_sq_create(tctrl, &sq->nvme_sq, &cq->nvme_cq, sqid,
+ sq->depth);
if (status != NVME_SC_SUCCESS)
- goto out_clear_bit;
+ return status;
sq->iod_wq = alloc_workqueue("sq%d_wq", WQ_UNBOUND,
min_t(int, sq->depth, WQ_MAX_ACTIVE), sqid);
@@ -1365,6 +1400,8 @@ static u16 nvmet_pci_epf_create_sq(struct nvmet_ctrl *tctrl,
goto out_destroy_sq;
}
+ set_bit(NVMET_PCI_EPF_Q_LIVE, &sq->flags);
+
dev_dbg(ctrl->dev, "SQ[%u]: %u entries of %zu B\n",
sqid, qsize, sq->qes);
@@ -1372,8 +1409,6 @@ static u16 nvmet_pci_epf_create_sq(struct nvmet_ctrl *tctrl,
out_destroy_sq:
nvmet_sq_destroy(&sq->nvme_sq);
-out_clear_bit:
- clear_bit(NVMET_PCI_EPF_Q_LIVE, &sq->flags);
return status;
}
@@ -1385,7 +1420,6 @@ static u16 nvmet_pci_epf_delete_sq(struct nvmet_ctrl *tctrl, u16 sqid)
if (!test_and_clear_bit(NVMET_PCI_EPF_Q_LIVE, &sq->flags))
return NVME_SC_QID_INVALID | NVME_STATUS_DNR;
- flush_workqueue(sq->iod_wq);
destroy_workqueue(sq->iod_wq);
sq->iod_wq = NULL;
@@ -1510,7 +1544,6 @@ static void nvmet_pci_epf_init_queue(struct nvmet_pci_epf_ctrl *ctrl,
if (sq) {
queue = &ctrl->sq[qid];
- set_bit(NVMET_PCI_EPF_Q_IS_SQ, &queue->flags);
} else {
queue = &ctrl->cq[qid];
INIT_DELAYED_WORK(&queue->work, nvmet_pci_epf_cq_work);
@@ -1554,36 +1587,6 @@ static void nvmet_pci_epf_free_queues(struct nvmet_pci_epf_ctrl *ctrl)
ctrl->cq = NULL;
}
-static int nvmet_pci_epf_map_queue(struct nvmet_pci_epf_ctrl *ctrl,
- struct nvmet_pci_epf_queue *queue)
-{
- struct nvmet_pci_epf *nvme_epf = ctrl->nvme_epf;
- int ret;
-
- ret = nvmet_pci_epf_mem_map(nvme_epf, queue->pci_addr,
- queue->pci_size, &queue->pci_map);
- if (ret) {
- dev_err(ctrl->dev, "Failed to map queue %u (err=%d)\n",
- queue->qid, ret);
- return ret;
- }
-
- if (queue->pci_map.pci_size < queue->pci_size) {
- dev_err(ctrl->dev, "Invalid partial mapping of queue %u\n",
- queue->qid);
- nvmet_pci_epf_mem_unmap(nvme_epf, &queue->pci_map);
- return -ENOMEM;
- }
-
- return 0;
-}
-
-static inline void nvmet_pci_epf_unmap_queue(struct nvmet_pci_epf_ctrl *ctrl,
- struct nvmet_pci_epf_queue *queue)
-{
- nvmet_pci_epf_mem_unmap(ctrl->nvme_epf, &queue->pci_map);
-}
-
static void nvmet_pci_epf_exec_iod_work(struct work_struct *work)
{
struct nvmet_pci_epf_iod *iod =
@@ -1601,8 +1604,7 @@ static void nvmet_pci_epf_exec_iod_work(struct work_struct *work)
goto complete;
}
- if (!nvmet_req_init(req, &iod->cq->nvme_cq, &iod->sq->nvme_sq,
- &nvmet_pci_epf_fabrics_ops))
+ if (!nvmet_req_init(req, &iod->sq->nvme_sq, &nvmet_pci_epf_fabrics_ops))
goto complete;
iod->data_len = nvmet_req_transfer_len(req);
@@ -1655,16 +1657,17 @@ static int nvmet_pci_epf_process_sq(struct nvmet_pci_epf_ctrl *ctrl,
{
struct nvmet_pci_epf_iod *iod;
int ret, n = 0;
+ u16 head = sq->head;
sq->tail = nvmet_pci_epf_bar_read32(ctrl, sq->db);
- while (sq->head != sq->tail && (!ctrl->sq_ab || n < ctrl->sq_ab)) {
+ while (head != sq->tail && (!ctrl->sq_ab || n < ctrl->sq_ab)) {
iod = nvmet_pci_epf_alloc_iod(sq);
if (!iod)
break;
/* Get the NVMe command submitted by the host. */
ret = nvmet_pci_epf_transfer(ctrl, &iod->cmd,
- sq->pci_addr + sq->head * sq->qes,
+ sq->pci_addr + head * sq->qes,
sq->qes, DMA_FROM_DEVICE);
if (ret) {
/* Not much we can do... */
@@ -1673,12 +1676,13 @@ static int nvmet_pci_epf_process_sq(struct nvmet_pci_epf_ctrl *ctrl,
}
dev_dbg(ctrl->dev, "SQ[%u]: head %u, tail %u, command %s\n",
- sq->qid, sq->head, sq->tail,
+ sq->qid, head, sq->tail,
nvmet_pci_epf_iod_name(iod));
- sq->head++;
- if (sq->head == sq->depth)
- sq->head = 0;
+ head++;
+ if (head == sq->depth)
+ head = 0;
+ WRITE_ONCE(sq->head, head);
n++;
queue_work_on(WORK_CPU_UNBOUND, sq->iod_wq, &iod->work);
@@ -1694,6 +1698,7 @@ static void nvmet_pci_epf_poll_sqs_work(struct work_struct *work)
struct nvmet_pci_epf_ctrl *ctrl =
container_of(work, struct nvmet_pci_epf_ctrl, poll_sqs.work);
struct nvmet_pci_epf_queue *sq;
+ unsigned long limit = jiffies;
unsigned long last = 0;
int i, nr_sqs;
@@ -1708,6 +1713,16 @@ static void nvmet_pci_epf_poll_sqs_work(struct work_struct *work)
nr_sqs++;
}
+ /*
+ * If we have been running for a while, reschedule to let other
+ * tasks run and to avoid RCU stalls.
+ */
+ if (time_is_before_jiffies(limit + secs_to_jiffies(1))) {
+ cond_resched();
+ limit = jiffies;
+ continue;
+ }
+
if (nr_sqs) {
last = jiffies;
continue;
@@ -1736,11 +1751,7 @@ static void nvmet_pci_epf_cq_work(struct work_struct *work)
struct nvme_completion *cqe;
struct nvmet_pci_epf_iod *iod;
unsigned long flags;
- int ret, n = 0;
-
- ret = nvmet_pci_epf_map_queue(ctrl, cq);
- if (ret)
- goto again;
+ int ret = 0, n = 0;
while (test_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags) && ctrl->link_up) {
@@ -1761,8 +1772,17 @@ static void nvmet_pci_epf_cq_work(struct work_struct *work)
if (!iod)
break;
- /* Post the IOD completion entry. */
+ /*
+ * Post the IOD completion entry. If the IOD request was
+ * executed (req->execute() called), the CQE is already
+ * initialized. However, the IOD may have been failed before
+ * that, leaving the CQE not properly initialized. So always
+ * initialize it here.
+ */
cqe = &iod->cqe;
+ cqe->sq_head = cpu_to_le16(READ_ONCE(iod->sq->head));
+ cqe->sq_id = cpu_to_le16(iod->sq->qid);
+ cqe->command_id = iod->cmd.common.command_id;
cqe->status = cpu_to_le16((iod->status << 1) | cq->phase);
dev_dbg(ctrl->dev,
@@ -1787,8 +1807,6 @@ static void nvmet_pci_epf_cq_work(struct work_struct *work)
n++;
}
- nvmet_pci_epf_unmap_queue(ctrl, cq);
-
/*
* We do not support precise IRQ coalescing time (100ns units as per
* NVMe specifications). So if we have posted completion entries without
@@ -1797,12 +1815,26 @@ static void nvmet_pci_epf_cq_work(struct work_struct *work)
if (n)
nvmet_pci_epf_raise_irq(ctrl, cq, true);
-again:
if (ret < 0)
queue_delayed_work(system_highpri_wq, &cq->work,
NVMET_PCI_EPF_CQ_RETRY_INTERVAL);
}
+static void nvmet_pci_epf_clear_ctrl_config(struct nvmet_pci_epf_ctrl *ctrl)
+{
+ struct nvmet_ctrl *tctrl = ctrl->tctrl;
+
+ /* Initialize controller status. */
+ tctrl->csts = 0;
+ ctrl->csts = 0;
+ nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CSTS, ctrl->csts);
+
+ /* Initialize controller configuration and start polling. */
+ tctrl->cc = 0;
+ ctrl->cc = 0;
+ nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CC, ctrl->cc);
+}
+
static int nvmet_pci_epf_enable_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
{
u64 pci_addr, asq, acq;
@@ -1822,14 +1854,14 @@ static int nvmet_pci_epf_enable_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
if (ctrl->io_sqes < sizeof(struct nvme_command)) {
dev_err(ctrl->dev, "Unsupported I/O SQES %zu (need %zu)\n",
ctrl->io_sqes, sizeof(struct nvme_command));
- return -EINVAL;
+ goto err;
}
ctrl->io_cqes = 1UL << nvmet_cc_iocqes(ctrl->cc);
if (ctrl->io_cqes < sizeof(struct nvme_completion)) {
dev_err(ctrl->dev, "Unsupported I/O CQES %zu (need %zu)\n",
ctrl->io_sqes, sizeof(struct nvme_completion));
- return -EINVAL;
+ goto err;
}
/* Create the admin queue. */
@@ -1844,37 +1876,44 @@ static int nvmet_pci_epf_enable_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
qsize, pci_addr, 0);
if (status != NVME_SC_SUCCESS) {
dev_err(ctrl->dev, "Failed to create admin completion queue\n");
- return -EINVAL;
+ goto err;
}
qsize = aqa & 0x00000fff;
pci_addr = asq & GENMASK_ULL(63, 12);
- status = nvmet_pci_epf_create_sq(ctrl->tctrl, 0, NVME_QUEUE_PHYS_CONTIG,
- qsize, pci_addr);
+ status = nvmet_pci_epf_create_sq(ctrl->tctrl, 0, 0,
+ NVME_QUEUE_PHYS_CONTIG, qsize, pci_addr);
if (status != NVME_SC_SUCCESS) {
dev_err(ctrl->dev, "Failed to create admin submission queue\n");
nvmet_pci_epf_delete_cq(ctrl->tctrl, 0);
- return -EINVAL;
+ goto err;
}
ctrl->sq_ab = NVMET_PCI_EPF_SQ_AB;
ctrl->irq_vector_threshold = NVMET_PCI_EPF_IV_THRESHOLD;
ctrl->enabled = true;
+ ctrl->csts = NVME_CSTS_RDY;
/* Start polling the controller SQs. */
schedule_delayed_work(&ctrl->poll_sqs, 0);
return 0;
+
+err:
+ nvmet_pci_epf_clear_ctrl_config(ctrl);
+ return -EINVAL;
}
-static void nvmet_pci_epf_disable_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
+static void nvmet_pci_epf_disable_ctrl(struct nvmet_pci_epf_ctrl *ctrl,
+ bool shutdown)
{
int qid;
if (!ctrl->enabled)
return;
- dev_info(ctrl->dev, "Disabling controller\n");
+ dev_info(ctrl->dev, "%s controller\n",
+ shutdown ? "Shutting down" : "Disabling");
ctrl->enabled = false;
cancel_delayed_work_sync(&ctrl->poll_sqs);
@@ -1889,6 +1928,13 @@ static void nvmet_pci_epf_disable_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
/* Delete the admin queue last. */
nvmet_pci_epf_delete_sq(ctrl->tctrl, 0);
nvmet_pci_epf_delete_cq(ctrl->tctrl, 0);
+
+ ctrl->csts &= ~NVME_CSTS_RDY;
+ if (shutdown) {
+ ctrl->csts |= NVME_CSTS_SHST_CMPLT;
+ ctrl->cc &= ~NVME_CC_ENABLE;
+ nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CC, ctrl->cc);
+ }
}
static void nvmet_pci_epf_poll_cc_work(struct work_struct *work)
@@ -1903,24 +1949,22 @@ static void nvmet_pci_epf_poll_cc_work(struct work_struct *work)
old_cc = ctrl->cc;
new_cc = nvmet_pci_epf_bar_read32(ctrl, NVME_REG_CC);
+ if (new_cc == old_cc)
+ goto reschedule_work;
+
ctrl->cc = new_cc;
if (nvmet_cc_en(new_cc) && !nvmet_cc_en(old_cc)) {
ret = nvmet_pci_epf_enable_ctrl(ctrl);
if (ret)
- return;
- ctrl->csts |= NVME_CSTS_RDY;
+ goto reschedule_work;
}
- if (!nvmet_cc_en(new_cc) && nvmet_cc_en(old_cc)) {
- nvmet_pci_epf_disable_ctrl(ctrl);
- ctrl->csts &= ~NVME_CSTS_RDY;
- }
+ if (!nvmet_cc_en(new_cc) && nvmet_cc_en(old_cc))
+ nvmet_pci_epf_disable_ctrl(ctrl, false);
- if (nvmet_cc_shn(new_cc) && !nvmet_cc_shn(old_cc)) {
- nvmet_pci_epf_disable_ctrl(ctrl);
- ctrl->csts |= NVME_CSTS_SHST_CMPLT;
- }
+ if (nvmet_cc_shn(new_cc) && !nvmet_cc_shn(old_cc))
+ nvmet_pci_epf_disable_ctrl(ctrl, true);
if (!nvmet_cc_shn(new_cc) && nvmet_cc_shn(old_cc))
ctrl->csts &= ~NVME_CSTS_SHST_CMPLT;
@@ -1928,6 +1972,7 @@ static void nvmet_pci_epf_poll_cc_work(struct work_struct *work)
nvmet_update_cc(ctrl->tctrl, ctrl->cc);
nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CSTS, ctrl->csts);
+reschedule_work:
schedule_delayed_work(&ctrl->poll_cc, NVMET_PCI_EPF_CC_POLL_INTERVAL);
}
@@ -1958,16 +2003,10 @@ static void nvmet_pci_epf_init_bar(struct nvmet_pci_epf_ctrl *ctrl)
/* Clear Controller Memory Buffer Supported (CMBS). */
ctrl->cap &= ~(0x1ULL << 57);
- /* Controller configuration. */
- ctrl->cc = tctrl->cc & (~NVME_CC_ENABLE);
-
- /* Controller status. */
- ctrl->csts = ctrl->tctrl->csts;
-
nvmet_pci_epf_bar_write64(ctrl, NVME_REG_CAP, ctrl->cap);
nvmet_pci_epf_bar_write32(ctrl, NVME_REG_VS, tctrl->subsys->ver);
- nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CSTS, ctrl->csts);
- nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CC, ctrl->cc);
+
+ nvmet_pci_epf_clear_ctrl_config(ctrl);
}
static int nvmet_pci_epf_create_ctrl(struct nvmet_pci_epf *nvme_epf,
@@ -2065,14 +2104,22 @@ out_mempool_exit:
static void nvmet_pci_epf_start_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
{
+
+ dev_info(ctrl->dev, "PCI link up\n");
+ ctrl->link_up = true;
+
schedule_delayed_work(&ctrl->poll_cc, NVMET_PCI_EPF_CC_POLL_INTERVAL);
}
static void nvmet_pci_epf_stop_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
{
+ dev_info(ctrl->dev, "PCI link down\n");
+ ctrl->link_up = false;
+
cancel_delayed_work_sync(&ctrl->poll_cc);
- nvmet_pci_epf_disable_ctrl(ctrl);
+ nvmet_pci_epf_disable_ctrl(ctrl, false);
+ nvmet_pci_epf_clear_ctrl_config(ctrl);
}
static void nvmet_pci_epf_destroy_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
@@ -2110,8 +2157,15 @@ static int nvmet_pci_epf_configure_bar(struct nvmet_pci_epf *nvme_epf)
return -ENODEV;
}
- if (epc_features->bar[BAR_0].only_64bit)
- epf->bar[BAR_0].flags |= PCI_BASE_ADDRESS_MEM_TYPE_64;
+ /*
+ * While NVMe PCIe Transport Specification 1.1, section 2.1.10, claims
+ * that the BAR0 type is Implementation Specific, in NVMe 1.1, the type
+ * is required to be 64-bit. Thus, for interoperability, always set the
+ * type to 64-bit. In the rare case that the PCI EPC does not support
+ * configuring BAR0 as 64-bit, the call to pci_epc_set_bar() will fail,
+ * and we will return failure back to the user.
+ */
+ epf->bar[BAR_0].flags |= PCI_BASE_ADDRESS_MEM_TYPE_64;
/*
* Calculate the size of the register bar: NVMe registers first with
@@ -2288,10 +2342,8 @@ static int nvmet_pci_epf_epc_init(struct pci_epf *epf)
if (ret)
goto out_clear_bar;
- if (!epc_features->linkup_notifier) {
- ctrl->link_up = true;
+ if (!epc_features->linkup_notifier)
nvmet_pci_epf_start_ctrl(&nvme_epf->ctrl);
- }
return 0;
@@ -2307,7 +2359,6 @@ static void nvmet_pci_epf_epc_deinit(struct pci_epf *epf)
struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf);
struct nvmet_pci_epf_ctrl *ctrl = &nvme_epf->ctrl;
- ctrl->link_up = false;
nvmet_pci_epf_destroy_ctrl(ctrl);
nvmet_pci_epf_deinit_dma(nvme_epf);
@@ -2319,7 +2370,6 @@ static int nvmet_pci_epf_link_up(struct pci_epf *epf)
struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf);
struct nvmet_pci_epf_ctrl *ctrl = &nvme_epf->ctrl;
- ctrl->link_up = true;
nvmet_pci_epf_start_ctrl(ctrl);
return 0;
@@ -2330,7 +2380,6 @@ static int nvmet_pci_epf_link_down(struct pci_epf *epf)
struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf);
struct nvmet_pci_epf_ctrl *ctrl = &nvme_epf->ctrl;
- ctrl->link_up = false;
nvmet_pci_epf_stop_ctrl(ctrl);
return 0;
diff --git a/drivers/nvme/target/rdma.c b/drivers/nvme/target/rdma.c
index 1afd93026f9b..67f61c67c167 100644
--- a/drivers/nvme/target/rdma.c
+++ b/drivers/nvme/target/rdma.c
@@ -976,8 +976,7 @@ static void nvmet_rdma_handle_command(struct nvmet_rdma_queue *queue,
cmd->send_sge.addr, cmd->send_sge.length,
DMA_TO_DEVICE);
- if (!nvmet_req_init(&cmd->req, &queue->nvme_cq,
- &queue->nvme_sq, &nvmet_rdma_ops))
+ if (!nvmet_req_init(&cmd->req, &queue->nvme_sq, &nvmet_rdma_ops))
return;
status = nvmet_rdma_map_sgl(cmd);
@@ -996,6 +995,27 @@ out_err:
nvmet_req_complete(&cmd->req, status);
}
+static bool nvmet_rdma_recv_not_live(struct nvmet_rdma_queue *queue,
+ struct nvmet_rdma_rsp *rsp)
+{
+ unsigned long flags;
+ bool ret = true;
+
+ spin_lock_irqsave(&queue->state_lock, flags);
+ /*
+ * recheck queue state is not live to prevent a race condition
+ * with RDMA_CM_EVENT_ESTABLISHED handler.
+ */
+ if (queue->state == NVMET_RDMA_Q_LIVE)
+ ret = false;
+ else if (queue->state == NVMET_RDMA_Q_CONNECTING)
+ list_add_tail(&rsp->wait_list, &queue->rsp_wait_list);
+ else
+ nvmet_rdma_put_rsp(rsp);
+ spin_unlock_irqrestore(&queue->state_lock, flags);
+ return ret;
+}
+
static void nvmet_rdma_recv_done(struct ib_cq *cq, struct ib_wc *wc)
{
struct nvmet_rdma_cmd *cmd =
@@ -1038,17 +1058,9 @@ static void nvmet_rdma_recv_done(struct ib_cq *cq, struct ib_wc *wc)
rsp->n_rdma = 0;
rsp->invalidate_rkey = 0;
- if (unlikely(queue->state != NVMET_RDMA_Q_LIVE)) {
- unsigned long flags;
-
- spin_lock_irqsave(&queue->state_lock, flags);
- if (queue->state == NVMET_RDMA_Q_CONNECTING)
- list_add_tail(&rsp->wait_list, &queue->rsp_wait_list);
- else
- nvmet_rdma_put_rsp(rsp);
- spin_unlock_irqrestore(&queue->state_lock, flags);
+ if (unlikely(queue->state != NVMET_RDMA_Q_LIVE) &&
+ nvmet_rdma_recv_not_live(queue, rsp))
return;
- }
nvmet_rdma_handle_command(queue, rsp);
}
@@ -1340,6 +1352,7 @@ static void nvmet_rdma_free_queue(struct nvmet_rdma_queue *queue)
pr_debug("freeing queue %d\n", queue->idx);
nvmet_sq_destroy(&queue->nvme_sq);
+ nvmet_cq_put(&queue->nvme_cq);
nvmet_rdma_destroy_queue_ib(queue);
if (!queue->nsrq) {
@@ -1423,7 +1436,8 @@ nvmet_rdma_alloc_queue(struct nvmet_rdma_device *ndev,
goto out_reject;
}
- ret = nvmet_sq_init(&queue->nvme_sq);
+ nvmet_cq_init(&queue->nvme_cq);
+ ret = nvmet_sq_init(&queue->nvme_sq, &queue->nvme_cq);
if (ret) {
ret = NVME_RDMA_CM_NO_RSC;
goto out_free_queue;
@@ -1504,6 +1518,7 @@ out_ida_remove:
out_destroy_sq:
nvmet_sq_destroy(&queue->nvme_sq);
out_free_queue:
+ nvmet_cq_put(&queue->nvme_cq);
kfree(queue);
out_reject:
nvmet_rdma_cm_reject(cm_id, ret);
@@ -1986,7 +2001,7 @@ static void nvmet_rdma_disc_port_addr(struct nvmet_req *req,
struct nvmet_rdma_port *port = nport->priv;
struct rdma_cm_id *cm_id = port->cm_id;
- if (inet_addr_is_any((struct sockaddr *)&cm_id->route.addr.src_addr)) {
+ if (inet_addr_is_any(&cm_id->route.addr.src_addr)) {
struct nvmet_rdma_rsp *rsp =
container_of(req, struct nvmet_rdma_rsp, req);
struct rdma_cm_id *req_cm_id = rsp->queue->cm_id;
diff --git a/drivers/nvme/target/tcp.c b/drivers/nvme/target/tcp.c
index 7c51c2a8c109..688033b88d38 100644
--- a/drivers/nvme/target/tcp.c
+++ b/drivers/nvme/target/tcp.c
@@ -7,8 +7,8 @@
#include <linux/module.h>
#include <linux/init.h>
#include <linux/slab.h>
+#include <linux/crc32c.h>
#include <linux/err.h>
-#include <linux/key.h>
#include <linux/nvme-tcp.h>
#include <linux/nvme-keyring.h>
#include <net/sock.h>
@@ -18,7 +18,6 @@
#include <net/handshake.h>
#include <linux/inet.h>
#include <linux/llist.h>
-#include <crypto/hash.h>
#include <trace/events/sock.h>
#include "nvmet.h"
@@ -173,8 +172,6 @@ struct nvmet_tcp_queue {
/* digest state */
bool hdr_digest;
bool data_digest;
- struct ahash_request *snd_hash;
- struct ahash_request *rcv_hash;
/* TLS state */
key_serial_t tls_pskid;
@@ -295,14 +292,9 @@ static inline u8 nvmet_tcp_ddgst_len(struct nvmet_tcp_queue *queue)
return queue->data_digest ? NVME_TCP_DIGEST_LENGTH : 0;
}
-static inline void nvmet_tcp_hdgst(struct ahash_request *hash,
- void *pdu, size_t len)
+static inline void nvmet_tcp_hdgst(void *pdu, size_t len)
{
- struct scatterlist sg;
-
- sg_init_one(&sg, pdu, len);
- ahash_request_set_crypt(hash, &sg, pdu + len, len);
- crypto_ahash_digest(hash);
+ put_unaligned_le32(~crc32c(~0, pdu, len), pdu + len);
}
static int nvmet_tcp_verify_hdgst(struct nvmet_tcp_queue *queue,
@@ -319,7 +311,7 @@ static int nvmet_tcp_verify_hdgst(struct nvmet_tcp_queue *queue,
}
recv_digest = *(__le32 *)(pdu + hdr->hlen);
- nvmet_tcp_hdgst(queue->rcv_hash, pdu, len);
+ nvmet_tcp_hdgst(pdu, len);
exp_digest = *(__le32 *)(pdu + hdr->hlen);
if (recv_digest != exp_digest) {
pr_err("queue %d: header digest error: recv %#x expected %#x\n",
@@ -442,12 +434,24 @@ err:
return NVME_SC_INTERNAL;
}
-static void nvmet_tcp_calc_ddgst(struct ahash_request *hash,
- struct nvmet_tcp_cmd *cmd)
+static void nvmet_tcp_calc_ddgst(struct nvmet_tcp_cmd *cmd)
{
- ahash_request_set_crypt(hash, cmd->req.sg,
- (void *)&cmd->exp_ddgst, cmd->req.transfer_len);
- crypto_ahash_digest(hash);
+ size_t total_len = cmd->req.transfer_len;
+ struct scatterlist *sg = cmd->req.sg;
+ u32 crc = ~0;
+
+ while (total_len) {
+ size_t len = min_t(size_t, total_len, sg->length);
+
+ /*
+ * Note that the scatterlist does not contain any highmem pages,
+ * as it was allocated by sgl_alloc() with GFP_KERNEL.
+ */
+ crc = crc32c(crc, sg_virt(sg), len);
+ total_len -= len;
+ sg = sg_next(sg);
+ }
+ cmd->exp_ddgst = cpu_to_le32(~crc);
}
static void nvmet_setup_c2h_data_pdu(struct nvmet_tcp_cmd *cmd)
@@ -474,19 +478,18 @@ static void nvmet_setup_c2h_data_pdu(struct nvmet_tcp_cmd *cmd)
if (queue->data_digest) {
pdu->hdr.flags |= NVME_TCP_F_DDGST;
- nvmet_tcp_calc_ddgst(queue->snd_hash, cmd);
+ nvmet_tcp_calc_ddgst(cmd);
}
if (cmd->queue->hdr_digest) {
pdu->hdr.flags |= NVME_TCP_F_HDGST;
- nvmet_tcp_hdgst(queue->snd_hash, pdu, sizeof(*pdu));
+ nvmet_tcp_hdgst(pdu, sizeof(*pdu));
}
}
static void nvmet_setup_r2t_pdu(struct nvmet_tcp_cmd *cmd)
{
struct nvme_tcp_r2t_pdu *pdu = cmd->r2t_pdu;
- struct nvmet_tcp_queue *queue = cmd->queue;
u8 hdgst = nvmet_tcp_hdgst_len(cmd->queue);
cmd->offset = 0;
@@ -504,14 +507,13 @@ static void nvmet_setup_r2t_pdu(struct nvmet_tcp_cmd *cmd)
pdu->r2t_offset = cpu_to_le32(cmd->rbytes_done);
if (cmd->queue->hdr_digest) {
pdu->hdr.flags |= NVME_TCP_F_HDGST;
- nvmet_tcp_hdgst(queue->snd_hash, pdu, sizeof(*pdu));
+ nvmet_tcp_hdgst(pdu, sizeof(*pdu));
}
}
static void nvmet_setup_response_pdu(struct nvmet_tcp_cmd *cmd)
{
struct nvme_tcp_rsp_pdu *pdu = cmd->rsp_pdu;
- struct nvmet_tcp_queue *queue = cmd->queue;
u8 hdgst = nvmet_tcp_hdgst_len(cmd->queue);
cmd->offset = 0;
@@ -524,7 +526,7 @@ static void nvmet_setup_response_pdu(struct nvmet_tcp_cmd *cmd)
pdu->hdr.plen = cpu_to_le32(pdu->hdr.hlen + hdgst);
if (cmd->queue->hdr_digest) {
pdu->hdr.flags |= NVME_TCP_F_HDGST;
- nvmet_tcp_hdgst(queue->snd_hash, pdu, sizeof(*pdu));
+ nvmet_tcp_hdgst(pdu, sizeof(*pdu));
}
}
@@ -571,10 +573,16 @@ static void nvmet_tcp_queue_response(struct nvmet_req *req)
struct nvmet_tcp_cmd *cmd =
container_of(req, struct nvmet_tcp_cmd, req);
struct nvmet_tcp_queue *queue = cmd->queue;
+ enum nvmet_tcp_recv_state queue_state;
+ struct nvmet_tcp_cmd *queue_cmd;
struct nvme_sgl_desc *sgl;
u32 len;
- if (unlikely(cmd == queue->cmd)) {
+ /* Pairs with store_release in nvmet_prepare_receive_pdu() */
+ queue_state = smp_load_acquire(&queue->rcv_state);
+ queue_cmd = READ_ONCE(queue->cmd);
+
+ if (unlikely(cmd == queue_cmd)) {
sgl = &cmd->req.cmd->common.dptr.sgl;
len = le32_to_cpu(sgl->length);
@@ -583,7 +591,7 @@ static void nvmet_tcp_queue_response(struct nvmet_req *req)
* Avoid using helpers, this might happen before
* nvmet_req_init is completed.
*/
- if (queue->rcv_state == NVMET_TCP_RECV_PDU &&
+ if (queue_state == NVMET_TCP_RECV_PDU &&
len && len <= cmd->req.port->inline_data_size &&
nvme_is_write(cmd->req.cmd))
return;
@@ -847,46 +855,11 @@ static void nvmet_prepare_receive_pdu(struct nvmet_tcp_queue *queue)
{
queue->offset = 0;
queue->left = sizeof(struct nvme_tcp_hdr);
- queue->cmd = NULL;
- queue->rcv_state = NVMET_TCP_RECV_PDU;
-}
-
-static void nvmet_tcp_free_crypto(struct nvmet_tcp_queue *queue)
-{
- struct crypto_ahash *tfm = crypto_ahash_reqtfm(queue->rcv_hash);
-
- ahash_request_free(queue->rcv_hash);
- ahash_request_free(queue->snd_hash);
- crypto_free_ahash(tfm);
-}
-
-static int nvmet_tcp_alloc_crypto(struct nvmet_tcp_queue *queue)
-{
- struct crypto_ahash *tfm;
-
- tfm = crypto_alloc_ahash("crc32c", 0, CRYPTO_ALG_ASYNC);
- if (IS_ERR(tfm))
- return PTR_ERR(tfm);
-
- queue->snd_hash = ahash_request_alloc(tfm, GFP_KERNEL);
- if (!queue->snd_hash)
- goto free_tfm;
- ahash_request_set_callback(queue->snd_hash, 0, NULL, NULL);
-
- queue->rcv_hash = ahash_request_alloc(tfm, GFP_KERNEL);
- if (!queue->rcv_hash)
- goto free_snd_hash;
- ahash_request_set_callback(queue->rcv_hash, 0, NULL, NULL);
-
- return 0;
-free_snd_hash:
- ahash_request_free(queue->snd_hash);
-free_tfm:
- crypto_free_ahash(tfm);
- return -ENOMEM;
+ WRITE_ONCE(queue->cmd, NULL);
+ /* Ensure rcv_state is visible only after queue->cmd is set */
+ smp_store_release(&queue->rcv_state, NVMET_TCP_RECV_PDU);
}
-
static int nvmet_tcp_handle_icreq(struct nvmet_tcp_queue *queue)
{
struct nvme_tcp_icreq_pdu *icreq = &queue->pdu.icreq;
@@ -915,11 +888,6 @@ static int nvmet_tcp_handle_icreq(struct nvmet_tcp_queue *queue)
queue->hdr_digest = !!(icreq->digest & NVME_TCP_HDR_DIGEST_ENABLE);
queue->data_digest = !!(icreq->digest & NVME_TCP_DATA_DIGEST_ENABLE);
- if (queue->hdr_digest || queue->data_digest) {
- ret = nvmet_tcp_alloc_crypto(queue);
- if (ret)
- return ret;
- }
memset(icresp, 0, sizeof(*icresp));
icresp->hdr.type = nvme_tcp_icresp;
@@ -1071,12 +1039,12 @@ static int nvmet_tcp_done_recv_pdu(struct nvmet_tcp_queue *queue)
req = &queue->cmd->req;
memcpy(req->cmd, nvme_cmd, sizeof(*nvme_cmd));
- if (unlikely(!nvmet_req_init(req, &queue->nvme_cq,
- &queue->nvme_sq, &nvmet_tcp_ops))) {
- pr_err("failed cmd %p id %d opcode %d, data_len: %d\n",
+ if (unlikely(!nvmet_req_init(req, &queue->nvme_sq, &nvmet_tcp_ops))) {
+ pr_err("failed cmd %p id %d opcode %d, data_len: %d, status: %04x\n",
req->cmd, req->cmd->common.command_id,
req->cmd->common.opcode,
- le32_to_cpu(req->cmd->common.dptr.sgl.length));
+ le32_to_cpu(req->cmd->common.dptr.sgl.length),
+ le16_to_cpu(req->cqe->status));
nvmet_tcp_handle_req_failure(queue, queue->cmd, req);
return 0;
@@ -1240,7 +1208,7 @@ static void nvmet_tcp_prep_recv_ddgst(struct nvmet_tcp_cmd *cmd)
{
struct nvmet_tcp_queue *queue = cmd->queue;
- nvmet_tcp_calc_ddgst(queue->rcv_hash, cmd);
+ nvmet_tcp_calc_ddgst(cmd);
queue->offset = 0;
queue->left = NVME_TCP_DIGEST_LENGTH;
queue->rcv_state = NVMET_TCP_RECV_DDGST;
@@ -1553,6 +1521,9 @@ static void nvmet_tcp_restore_socket_callbacks(struct nvmet_tcp_queue *queue)
{
struct socket *sock = queue->sock;
+ if (!queue->state_change)
+ return;
+
write_lock_bh(&sock->sk->sk_callback_lock);
sock->sk->sk_data_ready = queue->data_ready;
sock->sk->sk_state_change = queue->state_change;
@@ -1602,15 +1573,15 @@ static void nvmet_tcp_release_queue_work(struct work_struct *w)
/* stop accepting incoming data */
queue->rcv_state = NVMET_TCP_RECV_ERR;
+ nvmet_sq_put_tls_key(&queue->nvme_sq);
nvmet_tcp_uninit_data_in_cmds(queue);
nvmet_sq_destroy(&queue->nvme_sq);
+ nvmet_cq_put(&queue->nvme_cq);
cancel_work_sync(&queue->io_work);
nvmet_tcp_free_cmd_data_in_buffers(queue);
/* ->sock will be released by fput() */
fput(queue->sock->file);
nvmet_tcp_free_cmds(queue);
- if (queue->hdr_digest || queue->data_digest)
- nvmet_tcp_free_crypto(queue);
ida_free(&nvmet_tcp_queue_ida, queue->idx);
page_frag_cache_drain(&queue->pf_cache);
kfree(queue);
@@ -1787,6 +1758,27 @@ static int nvmet_tcp_try_peek_pdu(struct nvmet_tcp_queue *queue)
return 0;
}
+static int nvmet_tcp_tls_key_lookup(struct nvmet_tcp_queue *queue,
+ key_serial_t peerid)
+{
+ struct key *tls_key = nvme_tls_key_lookup(peerid);
+ int status = 0;
+
+ if (IS_ERR(tls_key)) {
+ pr_warn("%s: queue %d failed to lookup key %x\n",
+ __func__, queue->idx, peerid);
+ spin_lock_bh(&queue->state_lock);
+ queue->state = NVMET_TCP_Q_FAILED;
+ spin_unlock_bh(&queue->state_lock);
+ status = PTR_ERR(tls_key);
+ } else {
+ pr_debug("%s: queue %d using TLS PSK %x\n",
+ __func__, queue->idx, peerid);
+ queue->nvme_sq.tls_key = tls_key;
+ }
+ return status;
+}
+
static void nvmet_tcp_tls_handshake_done(void *data, int status,
key_serial_t peerid)
{
@@ -1807,6 +1799,10 @@ static void nvmet_tcp_tls_handshake_done(void *data, int status,
spin_unlock_bh(&queue->state_lock);
cancel_delayed_work_sync(&queue->tls_handshake_tmo_work);
+
+ if (!status)
+ status = nvmet_tcp_tls_key_lookup(queue, peerid);
+
if (status)
nvmet_tcp_schedule_release_queue(queue);
else
@@ -1914,7 +1910,8 @@ static void nvmet_tcp_alloc_queue(struct nvmet_tcp_port *port,
if (ret)
goto out_ida_remove;
- ret = nvmet_sq_init(&queue->nvme_sq);
+ nvmet_cq_init(&queue->nvme_cq);
+ ret = nvmet_sq_init(&queue->nvme_sq, &queue->nvme_cq);
if (ret)
goto out_free_connect;
@@ -1957,6 +1954,7 @@ out_destroy_sq:
mutex_unlock(&nvmet_tcp_queue_mutex);
nvmet_sq_destroy(&queue->nvme_sq);
out_free_connect:
+ nvmet_cq_put(&queue->nvme_cq);
nvmet_tcp_free_cmd(&queue->connect);
out_ida_remove:
ida_free(&nvmet_tcp_queue_ida, queue->idx);
@@ -2158,7 +2156,7 @@ static void nvmet_tcp_disc_port_addr(struct nvmet_req *req,
{
struct nvmet_tcp_port *port = nport->priv;
- if (inet_addr_is_any((struct sockaddr *)&port->addr)) {
+ if (inet_addr_is_any(&port->addr)) {
struct nvmet_tcp_cmd *cmd =
container_of(req, struct nvmet_tcp_cmd, req);
struct nvmet_tcp_queue *queue = cmd->queue;
generated by cgit v1.2.3 (git 2.39.0) at 2025-08-21 08:25:19 +0300