From mboxrd@z Thu Jan 1 00:00:00 1970 From: mgandhi@redhat.com (Milan P. Gandhi) Date: Thu, 9 Aug 2018 19:03:44 +0530 Subject: [PATCH] Remove extra goto label from nvmet_ns_enable Message-ID: <20180809133344.GA11707@machine1> Currently nvmet_ns_enable has two labels viz. out_unlock and out_dev_put, and the reverse call from out_dev_put to out_unlock is little bit confusing. We could simplify it by calling nvmet_ns_dev_disable before. This would eliminate need for 2nd label out_dev_put. Also, this function already initializes ret variable to 0, so there is no need to initialize it to 0 before out_unlock, so removing it. Signed-off-by: Milan P. Gandhi --- drivers/nvme/target/core.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c index 9838103f2d62..b40fb6d724b4 100644 --- a/drivers/nvme/target/core.c +++ b/drivers/nvme/target/core.c @@ -346,9 +346,10 @@ int nvmet_ns_enable(struct nvmet_ns *ns) ret = percpu_ref_init(&ns->ref, nvmet_destroy_namespace, 0, GFP_KERNEL); - if (ret) - goto out_dev_put; - + if (ret) { + nvmet_ns_dev_disable(ns); + goto out_unlock; + } if (ns->nsid > subsys->max_nsid) subsys->max_nsid = ns->nsid; @@ -372,13 +373,10 @@ int nvmet_ns_enable(struct nvmet_ns *ns) nvmet_ns_changed(subsys, ns->nsid); ns->enabled = true; - ret = 0; + out_unlock: mutex_unlock(&subsys->lock); return ret; -out_dev_put: - nvmet_ns_dev_disable(ns); - goto out_unlock; } void nvmet_ns_disable(struct nvmet_ns *ns) -- 2.14.3