Messages in this thread | | | Date | Fri, 5 Apr 2024 12:02:51 +0200 | From | Daniel Wagner <> | Subject | Re: [PATCH v4 3/5] nvmet: return DHCHAP status codes from nvmet_setup_auth() |
| |
On Fri, Apr 05, 2024 at 08:20:55AM +0200, Christoph Hellwig wrote: > > @@ -131,7 +131,6 @@ int nvmet_setup_auth(struct nvmet_ctrl *ctrl) > > int ret = 0; > > struct nvmet_host_link *p; > > struct nvmet_host *host = NULL; > > - const char *hash_name; > > > > down_read(&nvmet_config_sem); > > if (nvmet_is_disc_subsys(ctrl->subsys)) > > @@ -149,13 +148,16 @@ int nvmet_setup_auth(struct nvmet_ctrl *ctrl) > > } > > if (!host) { > > pr_debug("host %s not found\n", ctrl->hostnqn); > > - ret = -EPERM; > > + ret = NVME_AUTH_DHCHAP_FAILURE_FAILED; > > goto out_unlock; > > This is now returning returning random on the wire fields that aren't > even the NVMe status codes from a function otherwise returning Linux > errno values. I can't see how this works or is maintainable long term.
This is the target side and we generate the on wire return code here. Are you sure I should map this to errno codes and the back to NVME status codes? Sure, this is possible but don't really think it makes sense.
| |