diff options
Diffstat (limited to 'drivers/nvme/host/core.c')
| -rw-r--r-- | drivers/nvme/host/core.c | 5247 |
1 files changed, 3128 insertions, 2119 deletions
diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index c2c5bc4fb702..7bf228df6001 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -4,8 +4,10 @@ * Copyright (c) 2011-2014, Intel Corporation. */ +#include <linux/async.h> #include <linux/blkdev.h> #include <linux/blk-mq.h> +#include <linux/blk-integrity.h> #include <linux/compat.h> #include <linux/delay.h> #include <linux/errno.h> @@ -13,23 +15,39 @@ #include <linux/kernel.h> #include <linux/module.h> #include <linux/backing-dev.h> -#include <linux/list_sort.h> #include <linux/slab.h> #include <linux/types.h> #include <linux/pr.h> #include <linux/ptrace.h> #include <linux/nvme_ioctl.h> #include <linux/pm_qos.h> -#include <asm/unaligned.h> +#include <linux/ratelimit.h> +#include <linux/unaligned.h> #include "nvme.h" #include "fabrics.h" +#include <linux/nvme-auth.h> #define CREATE_TRACE_POINTS #include "trace.h" #define NVME_MINORS (1U << MINORBITS) +struct nvme_ns_info { + struct nvme_ns_ids ids; + u32 nsid; + __le32 anagrpid; + u8 pi_offset; + u16 endgid; + u64 runs; + bool is_shared; + bool is_readonly; + bool is_ready; + bool is_removed; + bool is_rotational; + bool no_vwc; +}; + unsigned int admin_timeout = 60; module_param(admin_timeout, uint, 0644); MODULE_PARM_DESC(admin_timeout, "timeout in seconds for admin commands"); @@ -57,9 +75,36 @@ static bool force_apst; module_param(force_apst, bool, 0644); MODULE_PARM_DESC(force_apst, "allow APST for newly enumerated devices even if quirked off"); -static bool streams; -module_param(streams, bool, 0644); -MODULE_PARM_DESC(streams, "turn on support for Streams write directives"); +static unsigned long apst_primary_timeout_ms = 100; +module_param(apst_primary_timeout_ms, ulong, 0644); +MODULE_PARM_DESC(apst_primary_timeout_ms, + "primary APST timeout in ms"); + +static unsigned long apst_secondary_timeout_ms = 2000; +module_param(apst_secondary_timeout_ms, ulong, 0644); +MODULE_PARM_DESC(apst_secondary_timeout_ms, + "secondary APST timeout in ms"); + +static unsigned long apst_primary_latency_tol_us = 15000; +module_param(apst_primary_latency_tol_us, ulong, 0644); +MODULE_PARM_DESC(apst_primary_latency_tol_us, + "primary APST latency tolerance in us"); + +static unsigned long apst_secondary_latency_tol_us = 100000; +module_param(apst_secondary_latency_tol_us, ulong, 0644); +MODULE_PARM_DESC(apst_secondary_latency_tol_us, + "secondary APST latency tolerance in us"); + +/* + * Older kernels didn't enable protection information if it was at an offset. + * Newer kernels do, so it breaks reads on the upgrade if such formats were + * used in prior kernels since the metadata written did not contain a valid + * checksum. + */ +static bool disable_pi_offsets = false; +module_param(disable_pi_offsets, bool, 0444); +MODULE_PARM_DESC(disable_pi_offsets, + "disable protection information if it has an offset"); /* * nvme_wq - hosts nvme related works that are not reset or delete @@ -82,41 +127,40 @@ struct workqueue_struct *nvme_delete_wq; EXPORT_SYMBOL_GPL(nvme_delete_wq); static LIST_HEAD(nvme_subsystems); -static DEFINE_MUTEX(nvme_subsystems_lock); +DEFINE_MUTEX(nvme_subsystems_lock); static DEFINE_IDA(nvme_instance_ida); -static dev_t nvme_chr_devt; -static struct class *nvme_class; -static struct class *nvme_subsys_class; +static dev_t nvme_ctrl_base_chr_devt; +static int nvme_class_uevent(const struct device *dev, struct kobj_uevent_env *env); +static const struct class nvme_class = { + .name = "nvme", + .dev_uevent = nvme_class_uevent, +}; + +static const struct class nvme_subsys_class = { + .name = "nvme-subsystem", +}; + +static DEFINE_IDA(nvme_ns_chr_minor_ida); +static dev_t nvme_ns_chr_devt; +static const struct class nvme_ns_chr_class = { + .name = "nvme-generic", +}; -static int nvme_revalidate_disk(struct gendisk *disk); static void nvme_put_subsystem(struct nvme_subsystem *subsys); static void nvme_remove_invalid_namespaces(struct nvme_ctrl *ctrl, unsigned nsid); +static void nvme_update_keep_alive(struct nvme_ctrl *ctrl, + struct nvme_command *cmd); +static int nvme_get_log_lsi(struct nvme_ctrl *ctrl, u32 nsid, u8 log_page, + u8 lsp, u8 csi, void *log, size_t size, u64 offset, u16 lsi); -static void nvme_set_queue_dying(struct nvme_ns *ns) -{ - /* - * Revalidating a dead namespace sets capacity to 0. This will end - * buffered writers dirtying pages that can't be synced. - */ - if (!ns->disk || test_and_set_bit(NVME_NS_DEAD, &ns->flags)) - return; - blk_set_queue_dying(ns->queue); - /* Forcibly unquiesce queues to avoid blocking dispatch */ - blk_mq_unquiesce_queue(ns->queue); - /* - * Revalidate after unblocking dispatchers that may be holding bd_butex - */ - revalidate_disk(ns->disk); -} - -static void nvme_queue_scan(struct nvme_ctrl *ctrl) +void nvme_queue_scan(struct nvme_ctrl *ctrl) { /* * Only new queue scan work when admin and IO queues are both alive */ - if (ctrl->state == NVME_CTRL_LIVE && ctrl->tagset) + if (nvme_ctrl_state(ctrl) == NVME_CTRL_LIVE && ctrl->tagset) queue_work(nvme_wq, &ctrl->scan_work); } @@ -128,7 +172,7 @@ static void nvme_queue_scan(struct nvme_ctrl *ctrl) */ int nvme_try_sched_reset(struct nvme_ctrl *ctrl) { - if (ctrl->state != NVME_CTRL_RESETTING) + if (nvme_ctrl_state(ctrl) != NVME_CTRL_RESETTING) return -EBUSY; if (!queue_work(nvme_reset_wq, &ctrl->reset_work)) return -EBUSY; @@ -136,6 +180,38 @@ int nvme_try_sched_reset(struct nvme_ctrl *ctrl) } EXPORT_SYMBOL_GPL(nvme_try_sched_reset); +static void nvme_failfast_work(struct work_struct *work) +{ + struct nvme_ctrl *ctrl = container_of(to_delayed_work(work), + struct nvme_ctrl, failfast_work); + + if (nvme_ctrl_state(ctrl) != NVME_CTRL_CONNECTING) + return; + + set_bit(NVME_CTRL_FAILFAST_EXPIRED, &ctrl->flags); + dev_info(ctrl->device, "failfast expired\n"); + nvme_kick_requeue_lists(ctrl); +} + +static inline void nvme_start_failfast_work(struct nvme_ctrl *ctrl) +{ + if (!ctrl->opts || ctrl->opts->fast_io_fail_tmo == -1) + return; + + schedule_delayed_work(&ctrl->failfast_work, + ctrl->opts->fast_io_fail_tmo * HZ); +} + +static inline void nvme_stop_failfast_work(struct nvme_ctrl *ctrl) +{ + if (!ctrl->opts) + return; + + cancel_delayed_work_sync(&ctrl->failfast_work); + clear_bit(NVME_CTRL_FAILFAST_EXPIRED, &ctrl->flags); +} + + int nvme_reset_ctrl(struct nvme_ctrl *ctrl) { if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_RESETTING)) @@ -153,18 +229,17 @@ int nvme_reset_ctrl_sync(struct nvme_ctrl *ctrl) ret = nvme_reset_ctrl(ctrl); if (!ret) { flush_work(&ctrl->reset_work); - if (ctrl->state != NVME_CTRL_LIVE) + if (nvme_ctrl_state(ctrl) != NVME_CTRL_LIVE) ret = -ENETRESET; } return ret; } -EXPORT_SYMBOL_GPL(nvme_reset_ctrl_sync); static void nvme_do_delete_ctrl(struct nvme_ctrl *ctrl) { dev_info(ctrl->device, - "Removing ctrl: NQN \"%s\"\n", ctrl->opts->subsysnqn); + "Removing ctrl: NQN \"%s\"\n", nvmf_ctrl_subsysnqn(ctrl)); flush_work(&ctrl->reset_work); nvme_stop_ctrl(ctrl); @@ -191,7 +266,7 @@ int nvme_delete_ctrl(struct nvme_ctrl *ctrl) } EXPORT_SYMBOL_GPL(nvme_delete_ctrl); -static void nvme_delete_ctrl_sync(struct nvme_ctrl *ctrl) +void nvme_delete_ctrl_sync(struct nvme_ctrl *ctrl) { /* * Keep a reference until nvme_do_delete_ctrl() complete, @@ -205,7 +280,7 @@ static void nvme_delete_ctrl_sync(struct nvme_ctrl *ctrl) static blk_status_t nvme_error_status(u16 status) { - switch (status & 0x7ff) { + switch (status & NVME_SCT_SC_MASK) { case NVME_SC_SUCCESS: return BLK_STS_OK; case NVME_SC_CAP_EXCEEDED: @@ -215,7 +290,6 @@ static blk_status_t nvme_error_status(u16 status) case NVME_SC_NS_NOT_READY: return BLK_STS_TARGET; case NVME_SC_BAD_ATTRIBUTES: - case NVME_SC_ONCS_NOT_SUPPORTED: case NVME_SC_INVALID_OPCODE: case NVME_SC_INVALID_FIELD: case NVME_SC_INVALID_NS: @@ -233,82 +307,253 @@ static blk_status_t nvme_error_status(u16 status) case NVME_SC_INVALID_PI: return BLK_STS_PROTECTION; case NVME_SC_RESERVATION_CONFLICT: - return BLK_STS_NEXUS; + return BLK_STS_RESV_CONFLICT; case NVME_SC_HOST_PATH_ERROR: return BLK_STS_TRANSPORT; + case NVME_SC_ZONE_TOO_MANY_ACTIVE: + return BLK_STS_ZONE_ACTIVE_RESOURCE; + case NVME_SC_ZONE_TOO_MANY_OPEN: + return BLK_STS_ZONE_OPEN_RESOURCE; default: return BLK_STS_IOERR; } } -static inline bool nvme_req_needs_retry(struct request *req) -{ - if (blk_noretry_request(req)) - return false; - if (nvme_req(req)->status & NVME_SC_DNR) - return false; - if (nvme_req(req)->retries >= nvme_max_retries) - return false; - return true; -} - static void nvme_retry_req(struct request *req) { - struct nvme_ns *ns = req->q->queuedata; unsigned long delay = 0; u16 crd; /* The mask and shift result must be <= 3 */ - crd = (nvme_req(req)->status & NVME_SC_CRD) >> 11; - if (ns && crd) - delay = ns->ctrl->crdt[crd - 1] * 100; + crd = (nvme_req(req)->status & NVME_STATUS_CRD) >> 11; + if (crd) + delay = nvme_req(req)->ctrl->crdt[crd - 1] * 100; nvme_req(req)->retries++; blk_mq_requeue_request(req, false); blk_mq_delay_kick_requeue_list(req->q, delay); } -void nvme_complete_rq(struct request *req) +static void nvme_log_error(struct request *req) { - blk_status_t status = nvme_error_status(nvme_req(req)->status); + struct nvme_ns *ns = req->q->queuedata; + struct nvme_request *nr = nvme_req(req); - trace_nvme_complete_rq(req); + if (ns) { + pr_err_ratelimited("%s: %s(0x%x) @ LBA %llu, %u blocks, %s (sct 0x%x / sc 0x%x) %s%s\n", + ns->disk ? ns->disk->disk_name : "?", + nvme_get_opcode_str(nr->cmd->common.opcode), + nr->cmd->common.opcode, + nvme_sect_to_lba(ns->head, blk_rq_pos(req)), + blk_rq_bytes(req) >> ns->head->lba_shift, + nvme_get_error_status_str(nr->status), + NVME_SCT(nr->status), /* Status Code Type */ + nr->status & NVME_SC_MASK, /* Status Code */ + nr->status & NVME_STATUS_MORE ? "MORE " : "", + nr->status & NVME_STATUS_DNR ? "DNR " : ""); + return; + } - nvme_cleanup_cmd(req); + pr_err_ratelimited("%s: %s(0x%x), %s (sct 0x%x / sc 0x%x) %s%s\n", + dev_name(nr->ctrl->device), + nvme_get_admin_opcode_str(nr->cmd->common.opcode), + nr->cmd->common.opcode, + nvme_get_error_status_str(nr->status), + NVME_SCT(nr->status), /* Status Code Type */ + nr->status & NVME_SC_MASK, /* Status Code */ + nr->status & NVME_STATUS_MORE ? "MORE " : "", + nr->status & NVME_STATUS_DNR ? "DNR " : ""); +} - if (nvme_req(req)->ctrl->kas) - nvme_req(req)->ctrl->comp_seen = true; +static void nvme_log_err_passthru(struct request *req) +{ + struct nvme_ns *ns = req->q->queuedata; + struct nvme_request *nr = nvme_req(req); + + pr_err_ratelimited("%s: %s(0x%x), %s (sct 0x%x / sc 0x%x) %s%s" + "cdw10=0x%x cdw11=0x%x cdw12=0x%x cdw13=0x%x cdw14=0x%x cdw15=0x%x\n", + ns ? ns->disk->disk_name : dev_name(nr->ctrl->device), + ns ? nvme_get_opcode_str(nr->cmd->common.opcode) : + nvme_get_admin_opcode_str(nr->cmd->common.opcode), + nr->cmd->common.opcode, + nvme_get_error_status_str(nr->status), + NVME_SCT(nr->status), /* Status Code Type */ + nr->status & NVME_SC_MASK, /* Status Code */ + nr->status & NVME_STATUS_MORE ? "MORE " : "", + nr->status & NVME_STATUS_DNR ? "DNR " : "", + le32_to_cpu(nr->cmd->common.cdw10), + le32_to_cpu(nr->cmd->common.cdw11), + le32_to_cpu(nr->cmd->common.cdw12), + le32_to_cpu(nr->cmd->common.cdw13), + le32_to_cpu(nr->cmd->common.cdw14), + le32_to_cpu(nr->cmd->common.cdw15)); +} + +enum nvme_disposition { + COMPLETE, + RETRY, + FAILOVER, + AUTHENTICATE, +}; - if (unlikely(status != BLK_STS_OK && nvme_req_needs_retry(req))) { - if ((req->cmd_flags & REQ_NVME_MPATH) && nvme_failover_req(req)) - return; +static inline enum nvme_disposition nvme_decide_disposition(struct request *req) +{ + if (likely(nvme_req(req)->status == 0)) + return COMPLETE; - if (!blk_queue_dying(req->q)) { - nvme_retry_req(req); - return; - } + if (blk_noretry_request(req) || + (nvme_req(req)->status & NVME_STATUS_DNR) || + nvme_req(req)->retries >= nvme_max_retries) + return COMPLETE; + + if ((nvme_req(req)->status & NVME_SCT_SC_MASK) == NVME_SC_AUTH_REQUIRED) + return AUTHENTICATE; + + if (req->cmd_flags & REQ_NVME_MPATH) { + if (nvme_is_path_error(nvme_req(req)->status) || + blk_queue_dying(req->q)) + return FAILOVER; + } else { + if (blk_queue_dying(req->q)) + return COMPLETE; } - nvme_trace_bio_complete(req, status); + return RETRY; +} + +static inline void nvme_end_req_zoned(struct request *req) +{ + if (IS_ENABLED(CONFIG_BLK_DEV_ZONED) && + req_op(req) == REQ_OP_ZONE_APPEND) { + struct nvme_ns *ns = req->q->queuedata; + + req->__sector = nvme_lba_to_sect(ns->head, + le64_to_cpu(nvme_req(req)->result.u64)); + } +} + +static inline void __nvme_end_req(struct request *req) +{ + if (unlikely(nvme_req(req)->status && !(req->rq_flags & RQF_QUIET))) { + if (blk_rq_is_passthrough(req)) + nvme_log_err_passthru(req); + else + nvme_log_error(req); + } + nvme_end_req_zoned(req); + nvme_trace_bio_complete(req); + if (req->cmd_flags & REQ_NVME_MPATH) + nvme_mpath_end_request(req); +} + +void nvme_end_req(struct request *req) +{ + blk_status_t status = nvme_error_status(nvme_req(req)->status); + + __nvme_end_req(req); blk_mq_end_request(req, status); } + +void nvme_complete_rq(struct request *req) +{ + struct nvme_ctrl *ctrl = nvme_req(req)->ctrl; + + trace_nvme_complete_rq(req); + nvme_cleanup_cmd(req); + + /* + * Completions of long-running commands should not be able to + * defer sending of periodic keep alives, since the controller + * may have completed processing such commands a long time ago + * (arbitrarily close to command submission time). + * req->deadline - req->timeout is the command submission time + * in jiffies. + */ + if (ctrl->kas && + req->deadline - req->timeout >= ctrl->ka_last_check_time) + ctrl->comp_seen = true; + + switch (nvme_decide_disposition(req)) { + case COMPLETE: + nvme_end_req(req); + return; + case RETRY: + nvme_retry_req(req); + return; + case FAILOVER: + nvme_failover_req(req); + return; + case AUTHENTICATE: +#ifdef CONFIG_NVME_HOST_AUTH + queue_work(nvme_wq, &ctrl->dhchap_auth_work); + nvme_retry_req(req); +#else + nvme_end_req(req); +#endif + return; + } +} EXPORT_SYMBOL_GPL(nvme_complete_rq); -bool nvme_cancel_request(struct request *req, void *data, bool reserved) +void nvme_complete_batch_req(struct request *req) +{ + trace_nvme_complete_rq(req); + nvme_cleanup_cmd(req); + __nvme_end_req(req); +} +EXPORT_SYMBOL_GPL(nvme_complete_batch_req); + +/* + * Called to unwind from ->queue_rq on a failed command submission so that the + * multipathing code gets called to potentially failover to another path. + * The caller needs to unwind all transport specific resource allocations and + * must return propagate the return value. + */ +blk_status_t nvme_host_path_error(struct request *req) +{ + nvme_req(req)->status = NVME_SC_HOST_PATH_ERROR; + blk_mq_set_request_complete(req); + nvme_complete_rq(req); + return BLK_STS_OK; +} +EXPORT_SYMBOL_GPL(nvme_host_path_error); + +bool nvme_cancel_request(struct request *req, void *data) { dev_dbg_ratelimited(((struct nvme_ctrl *) data)->device, "Cancelling I/O %d", req->tag); - /* don't abort one completed request */ - if (blk_mq_request_completed(req)) + /* don't abort one completed or idle request */ + if (blk_mq_rq_state(req) != MQ_RQ_IN_FLIGHT) return true; nvme_req(req)->status = NVME_SC_HOST_ABORTED_CMD; - blk_mq_force_complete_rq(req); + nvme_req(req)->flags |= NVME_REQ_CANCELLED; + blk_mq_complete_request(req); return true; } EXPORT_SYMBOL_GPL(nvme_cancel_request); +void nvme_cancel_tagset(struct nvme_ctrl *ctrl) +{ + if (ctrl->tagset) { + blk_mq_tagset_busy_iter(ctrl->tagset, + nvme_cancel_request, ctrl); + blk_mq_tagset_wait_completed_request(ctrl->tagset); + } +} +EXPORT_SYMBOL_GPL(nvme_cancel_tagset); + +void nvme_cancel_admin_tagset(struct nvme_ctrl *ctrl) +{ + if (ctrl->admin_tagset) { + blk_mq_tagset_busy_iter(ctrl->admin_tagset, + nvme_cancel_request, ctrl); + blk_mq_tagset_wait_completed_request(ctrl->admin_tagset); + } +} +EXPORT_SYMBOL_GPL(nvme_cancel_admin_tagset); + bool nvme_change_ctrl_state(struct nvme_ctrl *ctrl, enum nvme_ctrl_state new_state) { @@ -318,15 +563,13 @@ bool nvme_change_ctrl_state(struct nvme_ctrl *ctrl, spin_lock_irqsave(&ctrl->lock, flags); - old_state = ctrl->state; + old_state = nvme_ctrl_state(ctrl); switch (new_state) { case NVME_CTRL_LIVE: switch (old_state) { - case NVME_CTRL_NEW: - case NVME_CTRL_RESETTING: case NVME_CTRL_CONNECTING: changed = true; - /* FALLTHRU */ + fallthrough; default: break; } @@ -336,7 +579,7 @@ bool nvme_change_ctrl_state(struct nvme_ctrl *ctrl, case NVME_CTRL_NEW: case NVME_CTRL_LIVE: changed = true; - /* FALLTHRU */ + fallthrough; default: break; } @@ -346,7 +589,7 @@ bool nvme_change_ctrl_state(struct nvme_ctrl *ctrl, case NVME_CTRL_NEW: case NVME_CTRL_RESETTING: changed = true; - /* FALLTHRU */ + fallthrough; default: break; } @@ -357,7 +600,17 @@ bool nvme_change_ctrl_state(struct nvme_ctrl *ctrl, case NVME_CTRL_RESETTING: case NVME_CTRL_CONNECTING: changed = true; - /* FALLTHRU */ + fallthrough; + default: + break; + } + break; + case NVME_CTRL_DELETING_NOIO: + switch (old_state) { + case NVME_CTRL_DELETING: + case NVME_CTRL_DEAD: + changed = true; + fallthrough; default: break; } @@ -366,7 +619,7 @@ bool nvme_change_ctrl_state(struct nvme_ctrl *ctrl, switch (old_state) { case NVME_CTRL_DELETING: changed = true; - /* FALLTHRU */ + fallthrough; default: break; } @@ -376,38 +629,27 @@ bool nvme_change_ctrl_state(struct nvme_ctrl *ctrl, } if (changed) { - ctrl->state = new_state; + WRITE_ONCE(ctrl->state, new_state); wake_up_all(&ctrl->state_wq); } spin_unlock_irqrestore(&ctrl->lock, flags); - if (changed && ctrl->state == NVME_CTRL_LIVE) + if (!changed) + return false; + + if (new_state == NVME_CTRL_LIVE) { + if (old_state == NVME_CTRL_CONNECTING) + nvme_stop_failfast_work(ctrl); nvme_kick_requeue_lists(ctrl); + } else if (new_state == NVME_CTRL_CONNECTING && + old_state == NVME_CTRL_RESETTING) { + nvme_start_failfast_work(ctrl); + } return changed; } EXPORT_SYMBOL_GPL(nvme_change_ctrl_state); /* - * Returns true for sink states that can't ever transition back to live. - */ -static bool nvme_state_terminal(struct nvme_ctrl *ctrl) -{ - switch (ctrl->state) { - case NVME_CTRL_NEW: - case NVME_CTRL_LIVE: - case NVME_CTRL_RESETTING: - case NVME_CTRL_CONNECTING: - return false; - case NVME_CTRL_DELETING: - case NVME_CTRL_DEAD: - return true; - default: - WARN_ONCE(1, "Unhandled ctrl state:%d", ctrl->state); - return true; - } -} - -/* * Waits for the controller state to be resetting, or returns false if it is * not possible to ever transition to that state. */ @@ -416,7 +658,7 @@ bool nvme_wait_reset(struct nvme_ctrl *ctrl) wait_event(ctrl->state_wq, nvme_change_ctrl_state(ctrl, NVME_CTRL_RESETTING) || nvme_state_terminal(ctrl)); - return ctrl->state == NVME_CTRL_RESETTING; + return nvme_ctrl_state(ctrl) == NVME_CTRL_RESETTING; } EXPORT_SYMBOL_GPL(nvme_wait_reset); @@ -425,14 +667,20 @@ static void nvme_free_ns_head(struct kref *ref) struct nvme_ns_head *head = container_of(ref, struct nvme_ns_head, ref); - nvme_mpath_remove_disk(head); - ida_simple_remove(&head->subsys->ns_ida, head->instance); + nvme_mpath_put_disk(head); + ida_free(&head->subsys->ns_ida, head->instance); cleanup_srcu_struct(&head->srcu); nvme_put_subsystem(head->subsys); + kfree(head->plids); kfree(head); } -static void nvme_put_ns_head(struct nvme_ns_head *head) +bool nvme_tryget_ns_head(struct nvme_ns_head *head) +{ + return kref_get_unless_zero(&head->ref); +} + +void nvme_put_ns_head(struct nvme_ns_head *head) { kref_put(&head->ref, nvme_free_ns_head); } @@ -441,157 +689,134 @@ static void nvme_free_ns(struct kref *kref) { struct nvme_ns *ns = container_of(kref, struct nvme_ns, kref); - if (ns->ndev) - nvme_nvm_unregister(ns); - put_disk(ns->disk); nvme_put_ns_head(ns->head); nvme_put_ctrl(ns->ctrl); kfree(ns); } -static void nvme_put_ns(struct nvme_ns *ns) +bool nvme_get_ns(struct nvme_ns *ns) { - kref_put(&ns->kref, nvme_free_ns); + return kref_get_unless_zero(&ns->kref); } -static inline void nvme_clear_nvme_request(struct request *req) +void nvme_put_ns(struct nvme_ns *ns) { - if (!(req->rq_flags & RQF_DONTPREP)) { - nvme_req(req)->retries = 0; - nvme_req(req)->flags = 0; - req->rq_flags |= RQF_DONTPREP; - } -} - -struct request *nvme_alloc_request(struct request_queue *q, - struct nvme_command *cmd, blk_mq_req_flags_t flags, int qid) -{ - unsigned op = nvme_is_write(cmd) ? REQ_OP_DRV_OUT : REQ_OP_DRV_IN; - struct request *req; - - if (qid == NVME_QID_ANY) { - req = blk_mq_alloc_request(q, op, flags); - } else { - req = blk_mq_alloc_request_hctx(q, op, flags, - qid ? qid - 1 : 0); - } - if (IS_ERR(req)) - return req; - - req->cmd_flags |= REQ_FAILFAST_DRIVER; - nvme_clear_nvme_request(req); - nvme_req(req)->cmd = cmd; - - return req; + kref_put(&ns->kref, nvme_free_ns); } -EXPORT_SYMBOL_GPL(nvme_alloc_request); +EXPORT_SYMBOL_NS_GPL(nvme_put_ns, "NVME_TARGET_PASSTHRU"); -static int nvme_toggle_streams(struct nvme_ctrl *ctrl, bool enable) +static inline void nvme_clear_nvme_request(struct request *req) { - struct nvme_command c; - - memset(&c, 0, sizeof(c)); - - c.directive.opcode = nvme_admin_directive_send; - c.directive.nsid = cpu_to_le32(NVME_NSID_ALL); - c.directive.doper = NVME_DIR_SND_ID_OP_ENABLE; - c.directive.dtype = NVME_DIR_IDENTIFY; - c.directive.tdtype = NVME_DIR_STREAMS; - c.directive.endir = enable ? NVME_DIR_ENDIR : 0; - - return nvme_submit_sync_cmd(ctrl->admin_q, &c, NULL, 0); + nvme_req(req)->status = 0; + nvme_req(req)->retries = 0; + nvme_req(req)->flags = 0; + req->rq_flags |= RQF_DONTPREP; } -static int nvme_disable_streams(struct nvme_ctrl *ctrl) +/* initialize a passthrough request */ +void nvme_init_request(struct request *req, struct nvme_command *cmd) { - return nvme_toggle_streams(ctrl, false); -} + struct nvme_request *nr = nvme_req(req); + bool logging_enabled; -static int nvme_enable_streams(struct nvme_ctrl *ctrl) -{ - return nvme_toggle_streams(ctrl, true); -} + if (req->q->queuedata) { + struct nvme_ns *ns = req->q->disk->private_data; -static int nvme_get_stream_params(struct nvme_ctrl *ctrl, - struct streams_directive_params *s, u32 nsid) -{ - struct nvme_command c; + logging_enabled = ns->head->passthru_err_log_enabled; + req->timeout = NVME_IO_TIMEOUT; + } else { /* no queuedata implies admin queue */ + logging_enabled = nr->ctrl->passthru_err_log_enabled; + req->timeout = NVME_ADMIN_TIMEOUT; + } - memset(&c, 0, sizeof(c)); - memset(s, 0, sizeof(*s)); + if (!logging_enabled) + req->rq_flags |= RQF_QUIET; - c.directive.opcode = nvme_admin_directive_recv; - c.directive.nsid = cpu_to_le32(nsid); - c.directive.numd = cpu_to_le32(nvme_bytes_to_numd(sizeof(*s))); - c.directive.doper = NVME_DIR_RCV_ST_OP_PARAM; - c.directive.dtype = NVME_DIR_STREAMS; + /* passthru commands should let the driver set the SGL flags */ + cmd->common.flags &= ~NVME_CMD_SGL_ALL; - return nvme_submit_sync_cmd(ctrl->admin_q, &c, s, sizeof(*s)); + req->cmd_flags |= REQ_FAILFAST_DRIVER; + if (req->mq_hctx->type == HCTX_TYPE_POLL) + req->cmd_flags |= REQ_POLLED; + nvme_clear_nvme_request(req); + memcpy(nr->cmd, cmd, sizeof(*cmd)); } +EXPORT_SYMBOL_GPL(nvme_init_request); -static int nvme_configure_directives(struct nvme_ctrl *ctrl) +/* + * For something we're not in a state to send to the device the default action + * is to busy it and retry it after the controller state is recovered. However, + * if the controller is deleting or if anything is marked for failfast or + * nvme multipath it is immediately failed. + * + * Note: commands used to initialize the controller will be marked for failfast. + * Note: nvme cli/ioctl commands are marked for failfast. + */ +blk_status_t nvme_fail_nonready_command(struct nvme_ctrl *ctrl, + struct request *rq) { - struct streams_directive_params s; - int ret; + enum nvme_ctrl_state state = nvme_ctrl_state(ctrl); - if (!(ctrl->oacs & NVME_CTRL_OACS_DIRECTIVES)) - return 0; - if (!streams) - return 0; + if (state != NVME_CTRL_DELETING_NOIO && + state != NVME_CTRL_DELETING && + state != NVME_CTRL_DEAD && + !test_bit(NVME_CTRL_FAILFAST_EXPIRED, &ctrl->flags) && + !blk_noretry_request(rq) && !(rq->cmd_flags & REQ_NVME_MPATH)) + return BLK_STS_RESOURCE; - ret = nvme_enable_streams(ctrl); - if (ret) - return ret; - - ret = nvme_get_stream_params(ctrl, &s, NVME_NSID_ALL); - if (ret) - goto out_disable_stream; - - ctrl->nssa = le16_to_cpu(s.nssa); - if (ctrl->nssa < BLK_MAX_WRITE_HINTS - 1) { - dev_info(ctrl->device, "too few streams (%u) available\n", - ctrl->nssa); - goto out_disable_stream; - } - - ctrl->nr_streams = min_t(unsigned, ctrl->nssa, BLK_MAX_WRITE_HINTS - 1); - dev_info(ctrl->device, "Using %u streams\n", ctrl->nr_streams); - return 0; + if (!(rq->rq_flags & RQF_DONTPREP)) + nvme_clear_nvme_request(rq); -out_disable_stream: - nvme_disable_streams(ctrl); - return ret; + return nvme_host_path_error(rq); } +EXPORT_SYMBOL_GPL(nvme_fail_nonready_command); -/* - * Check if 'req' has a write hint associated with it. If it does, assign - * a valid namespace stream to the write. - */ -static void nvme_assign_write_stream(struct nvme_ctrl *ctrl, - struct request *req, u16 *control, - u32 *dsmgmt) +bool __nvme_check_ready(struct nvme_ctrl *ctrl, struct request *rq, + bool queue_live, enum nvme_ctrl_state state) { - enum rw_hint streamid = req->write_hint; + struct nvme_request *req = nvme_req(rq); - if (streamid == WRITE_LIFE_NOT_SET || streamid == WRITE_LIFE_NONE) - streamid = 0; - else { - streamid--; - if (WARN_ON_ONCE(streamid > ctrl->nr_streams)) - return; + /* + * currently we have a problem sending passthru commands + * on the admin_q if the controller is not LIVE because we can't + * make sure that they are going out after the admin connect, + * controller enable and/or other commands in the initialization + * sequence. until the controller will be LIVE, fail with + * BLK_STS_RESOURCE so that they will be rescheduled. + */ + if (rq->q == ctrl->admin_q && (req->flags & NVME_REQ_USERCMD)) + return false; - *control |= NVME_RW_DTYPE_STREAMS; - *dsmgmt |= streamid << 16; + if (ctrl->ops->flags & NVME_F_FABRICS) { + /* + * Only allow commands on a live queue, except for the connect + * command, which is require to set the queue live in the + * appropinquate states. + */ + switch (state) { + case NVME_CTRL_CONNECTING: + if (blk_rq_is_passthrough(rq) && nvme_is_fabrics(req->cmd) && + (req->cmd->fabrics.fctype == nvme_fabrics_type_connect || + req->cmd->fabrics.fctype == nvme_fabrics_type_auth_send || + req->cmd->fabrics.fctype == nvme_fabrics_type_auth_receive)) + return true; + break; + default: + break; + case NVME_CTRL_DEAD: + return false; + } } - if (streamid < ARRAY_SIZE(req->q->write_hints)) - req->q->write_hints[streamid] += blk_rq_bytes(req) >> 9; + return queue_live; } +EXPORT_SYMBOL_GPL(__nvme_check_ready); static inline void nvme_setup_flush(struct nvme_ns *ns, struct nvme_command *cmnd) { + memset(cmnd, 0, sizeof(*cmnd)); cmnd->common.opcode = nvme_cmd_flush; cmnd->common.nsid = cpu_to_le32(ns->head->ns_id); } @@ -623,16 +848,27 @@ static blk_status_t nvme_setup_discard(struct nvme_ns *ns, struct request *req, range = page_address(ns->ctrl->discard_page); } - __rq_for_each_bio(bio, req) { - u64 slba = nvme_sect_to_lba(ns, bio->bi_iter.bi_sector); - u32 nlb = bio->bi_iter.bi_size >> ns->lba_shift; + if (queue_max_discard_segments(req->q) == 1) { + u64 slba = nvme_sect_to_lba(ns->head, blk_rq_pos(req)); + u32 nlb = blk_rq_sectors(req) >> (ns->head->lba_shift - 9); - if (n < segments) { - range[n].cattr = cpu_to_le32(0); - range[n].nlb = cpu_to_le32(nlb); - range[n].slba = cpu_to_le64(slba); + range[0].cattr = cpu_to_le32(0); + range[0].nlb = cpu_to_le32(nlb); + range[0].slba = cpu_to_le64(slba); + n = 1; + } else { + __rq_for_each_bio(bio, req) { + u64 slba = nvme_sect_to_lba(ns->head, + bio->bi_iter.bi_sector); + u32 nlb = bio->bi_iter.bi_size >> ns->head->lba_shift; + + if (n < segments) { + range[n].cattr = cpu_to_le32(0); + range[n].nlb = cpu_to_le32(nlb); + range[n].slba = cpu_to_le64(slba); + } + n++; } - n++; } if (WARN_ON_ONCE(n != segments)) { @@ -643,39 +879,118 @@ static blk_status_t nvme_setup_discard(struct nvme_ns *ns, struct request *req, return BLK_STS_IOERR; } + memset(cmnd, 0, sizeof(*cmnd)); cmnd->dsm.opcode = nvme_cmd_dsm; cmnd->dsm.nsid = cpu_to_le32(ns->head->ns_id); cmnd->dsm.nr = cpu_to_le32(segments - 1); cmnd->dsm.attributes = cpu_to_le32(NVME_DSMGMT_AD); - req->special_vec.bv_page = virt_to_page(range); - req->special_vec.bv_offset = offset_in_page(range); - req->special_vec.bv_len = alloc_size; + bvec_set_virt(&req->special_vec, range, alloc_size); req->rq_flags |= RQF_SPECIAL_PAYLOAD; return BLK_STS_OK; } +static void nvme_set_app_tag(struct request *req, struct nvme_command *cmnd) +{ + cmnd->rw.lbat = cpu_to_le16(bio_integrity(req->bio)->app_tag); + cmnd->rw.lbatm = cpu_to_le16(0xffff); +} + +static void nvme_set_ref_tag(struct nvme_ns *ns, struct nvme_command *cmnd, + struct request *req) +{ + u32 upper, lower; + u64 ref48; + + /* only type1 and type 2 PI formats have a reftag */ + switch (ns->head->pi_type) { + case NVME_NS_DPS_PI_TYPE1: + case NVME_NS_DPS_PI_TYPE2: + break; + default: + return; + } + + /* both rw and write zeroes share the same reftag format */ + switch (ns->head->guard_type) { + case NVME_NVM_NS_16B_GUARD: + cmnd->rw.reftag = cpu_to_le32(t10_pi_ref_tag(req)); + break; + case NVME_NVM_NS_64B_GUARD: + ref48 = ext_pi_ref_tag(req); + lower = lower_32_bits(ref48); + upper = upper_32_bits(ref48); + + cmnd->rw.reftag = cpu_to_le32(lower); + cmnd->rw.cdw3 = cpu_to_le32(upper); + break; + default: + break; + } +} + static inline blk_status_t nvme_setup_write_zeroes(struct nvme_ns *ns, struct request *req, struct nvme_command *cmnd) { + memset(cmnd, 0, sizeof(*cmnd)); + if (ns->ctrl->quirks & NVME_QUIRK_DEALLOCATE_ZEROES) return nvme_setup_discard(ns, req, cmnd); cmnd->write_zeroes.opcode = nvme_cmd_write_zeroes; cmnd->write_zeroes.nsid = cpu_to_le32(ns->head->ns_id); cmnd->write_zeroes.slba = - cpu_to_le64(nvme_sect_to_lba(ns, blk_rq_pos(req))); + cpu_to_le64(nvme_sect_to_lba(ns->head, blk_rq_pos(req))); cmnd->write_zeroes.length = - cpu_to_le16((blk_rq_bytes(req) >> ns->lba_shift) - 1); - cmnd->write_zeroes.control = 0; + cpu_to_le16((blk_rq_bytes(req) >> ns->head->lba_shift) - 1); + + if (!(req->cmd_flags & REQ_NOUNMAP) && + (ns->head->features & NVME_NS_DEAC)) + cmnd->write_zeroes.control |= cpu_to_le16(NVME_WZ_DEAC); + + if (nvme_ns_has_pi(ns->head)) { + cmnd->write_zeroes.control |= cpu_to_le16(NVME_RW_PRINFO_PRACT); + nvme_set_ref_tag(ns, cmnd, req); + } + return BLK_STS_OK; } +/* + * NVMe does not support a dedicated command to issue an atomic write. A write + * which does adhere to the device atomic limits will silently be executed + * non-atomically. The request issuer should ensure that the write is within + * the queue atomic writes limits, but just validate this in case it is not. + */ +static bool nvme_valid_atomic_write(struct request *req) +{ + struct request_queue *q = req->q; + u32 boundary_bytes = queue_atomic_write_boundary_bytes(q); + + if (blk_rq_bytes(req) > queue_atomic_write_unit_max_bytes(q)) + return false; + + if (boundary_bytes) { + u64 mask = boundary_bytes - 1, imask = ~mask; + u64 start = blk_rq_pos(req) << SECTOR_SHIFT; + u64 end = start + blk_rq_bytes(req) - 1; + + /* If greater then must be crossing a boundary */ + if (blk_rq_bytes(req) > boundary_bytes) + return false; + + if ((start & imask) != (end & imask)) + return false; + } + + return true; +} + static inline blk_status_t nvme_setup_rw(struct nvme_ns *ns, - struct request *req, struct nvme_command *cmnd) + struct request *req, struct nvme_command *cmnd, + enum nvme_opcode op) { - struct nvme_ctrl *ctrl = ns->ctrl; u16 control = 0; u32 dsmgmt = 0; @@ -687,37 +1002,60 @@ static inline blk_status_t nvme_setup_rw(struct nvme_ns *ns, if (req->cmd_flags & REQ_RAHEAD) dsmgmt |= NVME_RW_DSM_FREQ_PREFETCH; - cmnd->rw.opcode = (rq_data_dir(req) ? nvme_cmd_write : nvme_cmd_read); - cmnd->rw.nsid = cpu_to_le32(ns->head->ns_id); - cmnd->rw.slba = cpu_to_le64(nvme_sect_to_lba(ns, blk_rq_pos(req))); - cmnd->rw.length = cpu_to_le16((blk_rq_bytes(req) >> ns->lba_shift) - 1); + if (op == nvme_cmd_write && ns->head->nr_plids) { + u16 write_stream = req->bio->bi_write_stream; + + if (WARN_ON_ONCE(write_stream > ns->head->nr_plids)) + return BLK_STS_INVAL; - if (req_op(req) == REQ_OP_WRITE && ctrl->nr_streams) - nvme_assign_write_stream(ctrl, req, &control, &dsmgmt); + if (write_stream) { + dsmgmt |= ns->head->plids[write_stream - 1] << 16; + control |= NVME_RW_DTYPE_DPLCMT; + } + } - if (ns->ms) { + if (req->cmd_flags & REQ_ATOMIC && !nvme_valid_atomic_write(req)) + return BLK_STS_INVAL; + + cmnd->rw.opcode = op; + cmnd->rw.flags = 0; + cmnd->rw.nsid = cpu_to_le32(ns->head->ns_id); + cmnd->rw.cdw2 = 0; + cmnd->rw.cdw3 = 0; + cmnd->rw.metadata = 0; + cmnd->rw.slba = + cpu_to_le64(nvme_sect_to_lba(ns->head, blk_rq_pos(req))); + cmnd->rw.length = + cpu_to_le16((blk_rq_bytes(req) >> ns->head->lba_shift) - 1); + cmnd->rw.reftag = 0; + cmnd->rw.lbat = 0; + cmnd->rw.lbatm = 0; + + if (ns->head->ms) { /* - * If formated with metadata, the block layer always provides a + * If formatted with metadata, the block layer always provides a * metadata buffer if CONFIG_BLK_DEV_INTEGRITY is enabled. Else * we enable the PRACT bit for protection information or set the * namespace capacity to zero to prevent any I/O. */ if (!blk_integrity_rq(req)) { - if (WARN_ON_ONCE(!nvme_ns_has_pi(ns))) + if (WARN_ON_ONCE(!nvme_ns_has_pi(ns->head))) return BLK_STS_NOTSUPP; control |= NVME_RW_PRINFO_PRACT; + nvme_set_ref_tag(ns, cmnd, req); } - switch (ns->pi_type) { - case NVME_NS_DPS_PI_TYPE3: + if (bio_integrity_flagged(req->bio, BIP_CHECK_GUARD)) control |= NVME_RW_PRINFO_PRCHK_GUARD; - break; - case NVME_NS_DPS_PI_TYPE1: - case NVME_NS_DPS_PI_TYPE2: - control |= NVME_RW_PRINFO_PRCHK_GUARD | - NVME_RW_PRINFO_PRCHK_REF; - cmnd->rw.reftag = cpu_to_le32(t10_pi_ref_tag(req)); - break; + if (bio_integrity_flagged(req->bio, BIP_CHECK_REFTAG)) { + control |= NVME_RW_PRINFO_PRCHK_REF; + if (op == nvme_cmd_zone_append) + control |= NVME_RW_APPEND_PIREMAP; + nvme_set_ref_tag(ns, cmnd, req); + } + if (bio_integrity_flagged(req->bio, BIP_CHECK_APPTAG)) { + control |= NVME_RW_PRINFO_PRCHK_APP; + nvme_set_app_tag(req, cmnd); } } @@ -729,33 +1067,46 @@ static inline blk_status_t nvme_setup_rw(struct nvme_ns *ns, void nvme_cleanup_cmd(struct request *req) { if (req->rq_flags & RQF_SPECIAL_PAYLOAD) { - struct nvme_ns *ns = req->rq_disk->private_data; - struct page *page = req->special_vec.bv_page; + struct nvme_ctrl *ctrl = nvme_req(req)->ctrl; - if (page == ns->ctrl->discard_page) - clear_bit_unlock(0, &ns->ctrl->discard_page_busy); + if (req->special_vec.bv_page == ctrl->discard_page) + clear_bit_unlock(0, &ctrl->discard_page_busy); else - kfree(page_address(page) + req->special_vec.bv_offset); + kfree(bvec_virt(&req->special_vec)); + req->rq_flags &= ~RQF_SPECIAL_PAYLOAD; } } EXPORT_SYMBOL_GPL(nvme_cleanup_cmd); -blk_status_t nvme_setup_cmd(struct nvme_ns *ns, struct request *req, - struct nvme_command *cmd) +blk_status_t nvme_setup_cmd(struct nvme_ns *ns, struct request *req) { + struct nvme_command *cmd = nvme_req(req)->cmd; blk_status_t ret = BLK_STS_OK; - nvme_clear_nvme_request(req); + if (!(req->rq_flags & RQF_DONTPREP)) + nvme_clear_nvme_request(req); - memset(cmd, 0, sizeof(*cmd)); switch (req_op(req)) { case REQ_OP_DRV_IN: case REQ_OP_DRV_OUT: - memcpy(cmd, nvme_req(req)->cmd, sizeof(*cmd)); + /* these are setup prior to execution in nvme_init_request() */ break; case REQ_OP_FLUSH: nvme_setup_flush(ns, cmd); break; + case REQ_OP_ZONE_RESET_ALL: + case REQ_OP_ZONE_RESET: + ret = nvme_setup_zone_mgmt_send(ns, req, cmd, NVME_ZONE_RESET); + break; + case REQ_OP_ZONE_OPEN: + ret = nvme_setup_zone_mgmt_send(ns, req, cmd, NVME_ZONE_OPEN); + break; + case REQ_OP_ZONE_CLOSE: + ret = nvme_setup_zone_mgmt_send(ns, req, cmd, NVME_ZONE_CLOSE); + break; + case REQ_OP_ZONE_FINISH: + ret = nvme_setup_zone_mgmt_send(ns, req, cmd, NVME_ZONE_FINISH); + break; case REQ_OP_WRITE_ZEROES: ret = nvme_setup_write_zeroes(ns, req, cmd); break; @@ -763,44 +1114,43 @@ blk_status_t nvme_setup_cmd(struct nvme_ns *ns, struct request *req, ret = nvme_setup_discard(ns, req, cmd); break; case REQ_OP_READ: + ret = nvme_setup_rw(ns, req, cmd, nvme_cmd_read); + break; case REQ_OP_WRITE: - ret = nvme_setup_rw(ns, req, cmd); + ret = nvme_setup_rw(ns, req, cmd, nvme_cmd_write); + break; + case REQ_OP_ZONE_APPEND: + ret = nvme_setup_rw(ns, req, cmd, nvme_cmd_zone_append); break; default: WARN_ON_ONCE(1); return BLK_STS_IOERR; } - cmd->common.command_id = req->tag; + cmd->common.command_id = nvme_cid(req); trace_nvme_setup_cmd(req, cmd); return ret; } EXPORT_SYMBOL_GPL(nvme_setup_cmd); -static void nvme_end_sync_rq(struct request *rq, blk_status_t error) -{ - struct completion *waiting = rq->end_io_data; - - rq->end_io_data = NULL; - complete(waiting); -} - -static void nvme_execute_rq_polled(struct request_queue *q, - struct gendisk *bd_disk, struct request *rq, int at_head) +/* + * Return values: + * 0: success + * >0: nvme controller's cqe status response + * <0: kernel error in lieu of controller response + */ +int nvme_execute_rq(struct request *rq, bool at_head) { - DECLARE_COMPLETION_ONSTACK(wait); - - WARN_ON_ONCE(!test_bit(QUEUE_FLAG_POLL, &q->queue_flags)); + blk_status_t status; - rq->cmd_flags |= REQ_HIPRI; - rq->end_io_data = &wait; - blk_execute_rq_nowait(q, bd_disk, rq, at_head, nvme_end_sync_rq); - - while (!completion_done(&wait)) { - blk_poll(q, request_to_qc_t(rq->mq_hctx, rq), true); - cond_resched(); - } + status = blk_execute_rq(rq, at_head); + if (nvme_req(rq)->flags & NVME_REQ_CANCELLED) + return -EINTR; + if (nvme_req(rq)->status) + return nvme_req(rq)->status; + return blk_status_to_errno(status); } +EXPORT_SYMBOL_NS_GPL(nvme_execute_rq, "NVME_TARGET_PASSTHRU"); /* * Returns 0 on success. If the result is negative, it's a Linux error code; @@ -808,34 +1158,37 @@ static void nvme_execute_rq_polled(struct request_queue *q, */ int __nvme_submit_sync_cmd(struct request_queue *q, struct nvme_command *cmd, union nvme_result *result, void *buffer, unsigned bufflen, - unsigned timeout, int qid, int at_head, - blk_mq_req_flags_t flags, bool poll) + int qid, nvme_submit_flags_t flags) { struct request *req; int ret; + blk_mq_req_flags_t blk_flags = 0; + + if (flags & NVME_SUBMIT_NOWAIT) + blk_flags |= BLK_MQ_REQ_NOWAIT; + if (flags & NVME_SUBMIT_RESERVED) + blk_flags |= BLK_MQ_REQ_RESERVED; + if (qid == NVME_QID_ANY) + req = blk_mq_alloc_request(q, nvme_req_op(cmd), blk_flags); + else + req = blk_mq_alloc_request_hctx(q, nvme_req_op(cmd), blk_flags, + qid - 1); - req = nvme_alloc_request(q, cmd, flags, qid); if (IS_ERR(req)) return PTR_ERR(req); - - req->timeout = timeout ? timeout : ADMIN_TIMEOUT; + nvme_init_request(req, cmd); + if (flags & NVME_SUBMIT_RETRY) + req->cmd_flags &= ~REQ_FAILFAST_DRIVER; if (buffer && bufflen) { - ret = blk_rq_map_kern(q, req, buffer, bufflen, GFP_KERNEL); + ret = blk_rq_map_kern(req, buffer, bufflen, GFP_KERNEL); if (ret) goto out; } - if (poll) - nvme_execute_rq_polled(req->q, NULL, req, at_head); - else - blk_execute_rq(req->q, NULL, req, at_head); - if (result) + ret = nvme_execute_rq(req, flags & NVME_SUBMIT_AT_HEAD); + if (result && ret >= 0) *result = nvme_req(req)->result; - if (nvme_req(req)->flags & NVME_REQ_CANCELLED) - ret = -EINTR; - else - ret = nvme_req(req)->status; out: blk_mq_free_request(req); return ret; @@ -845,108 +1198,159 @@ EXPORT_SYMBOL_GPL(__nvme_submit_sync_cmd); int nvme_submit_sync_cmd(struct request_queue *q, struct nvme_command *cmd, void *buffer, unsigned bufflen) { - return __nvme_submit_sync_cmd(q, cmd, NULL, buffer, bufflen, 0, - NVME_QID_ANY, 0, 0, false); + return __nvme_submit_sync_cmd(q, cmd, NULL, buffer, bufflen, + NVME_QID_ANY, 0); } EXPORT_SYMBOL_GPL(nvme_submit_sync_cmd); -static void *nvme_add_user_metadata(struct bio *bio, void __user *ubuf, - unsigned len, u32 seed, bool write) +u32 nvme_command_effects(struct nvme_ctrl *ctrl, struct nvme_ns *ns, u8 opcode) { - struct bio_integrity_payload *bip; - int ret = -ENOMEM; - void *buf; + u32 effects = 0; - buf = kmalloc(len, GFP_KERNEL); - if (!buf) - goto out; + if (ns) { + effects = le32_to_cpu(ns->head->effects->iocs[opcode]); + if (effects & ~(NVME_CMD_EFFECTS_CSUPP | NVME_CMD_EFFECTS_LBCC)) + dev_warn_once(ctrl->device, + "IO command:%02x has unusual effects:%08x\n", + opcode, effects); - ret = -EFAULT; - if (write && copy_from_user(buf, ubuf, len)) - goto out_free_meta; - - bip = bio_integrity_alloc(bio, GFP_KERNEL, 1); - if (IS_ERR(bip)) { - ret = PTR_ERR(bip); - goto out_free_meta; - } - - bip->bip_iter.bi_size = len; - bip->bip_iter.bi_sector = seed; - ret = bio_integrity_add_page(bio, virt_to_page(buf), len, - offset_in_page(buf)); - if (ret == len) - return buf; - ret = -ENOMEM; -out_free_meta: - kfree(buf); -out: - return ERR_PTR(ret); + /* + * NVME_CMD_EFFECTS_CSE_MASK causes a freeze all I/O queues, + * which would deadlock when done on an I/O command. Note that + * We already warn about an unusual effect above. + */ + effects &= ~NVME_CMD_EFFECTS_CSE_MASK; + } else { + effects = le32_to_cpu(ctrl->effects->acs[opcode]); + + /* Ignore execution restrictions if any relaxation bits are set */ + if (effects & NVME_CMD_EFFECTS_CSER_MASK) + effects &= ~NVME_CMD_EFFECTS_CSE_MASK; + } + + return effects; } +EXPORT_SYMBOL_NS_GPL(nvme_command_effects, "NVME_TARGET_PASSTHRU"); -static int nvme_submit_user_cmd(struct request_queue *q, - struct nvme_command *cmd, void __user *ubuffer, - unsigned bufflen, void __user *meta_buffer, unsigned meta_len, - u32 meta_seed, u64 *result, unsigned timeout) +u32 nvme_passthru_start(struct nvme_ctrl *ctrl, struct nvme_ns *ns, u8 opcode) { - bool write = nvme_is_write(cmd); - struct nvme_ns *ns = q->queuedata; - struct gendisk *disk = ns ? ns->disk : NULL; - struct request *req; - struct bio *bio = NULL; - void *meta = NULL; - int ret; + u32 effects = nvme_command_effects(ctrl, ns, opcode); - req = nvme_alloc_request(q, cmd, 0, NVME_QID_ANY); - if (IS_ERR(req)) - return PTR_ERR(req); + /* + * For simplicity, IO to all namespaces is quiesced even if the command + * effects say only one namespace is affected. + */ + if (effects & NVME_CMD_EFFECTS_CSE_MASK) { + mutex_lock(&ctrl->scan_lock); + mutex_lock(&ctrl->subsys->lock); + nvme_mpath_start_freeze(ctrl->subsys); + nvme_mpath_wait_freeze(ctrl->subsys); + nvme_start_freeze(ctrl); + nvme_wait_freeze(ctrl); + } + return effects; +} +EXPORT_SYMBOL_NS_GPL(nvme_passthru_start, "NVME_TARGET_PASSTHRU"); - req->timeout = timeout ? timeout : ADMIN_TIMEOUT; - nvme_req(req)->flags |= NVME_REQ_USERCMD; +void nvme_passthru_end(struct nvme_ctrl *ctrl, struct nvme_ns *ns, u32 effects, + struct nvme_command *cmd, int status) +{ + if (effects & NVME_CMD_EFFECTS_CSE_MASK) { + nvme_unfreeze(ctrl); + nvme_mpath_unfreeze(ctrl->subsys); + mutex_unlock(&ctrl->subsys->lock); + mutex_unlock(&ctrl->scan_lock); + } + if (effects & NVME_CMD_EFFECTS_CCC) { + if (!test_and_set_bit(NVME_CTRL_DIRTY_CAPABILITY, + &ctrl->flags)) { + dev_info(ctrl->device, +"controller capabilities changed, reset may be required to take effect.\n"); + } + } + if (effects & (NVME_CMD_EFFECTS_NIC | NVME_CMD_EFFECTS_NCC)) { + nvme_queue_scan(ctrl); + flush_work(&ctrl->scan_work); + } + if (ns) + return; - if (ubuffer && bufflen) { - ret = blk_rq_map_user(q, req, NULL, ubuffer, bufflen, - GFP_KERNEL); - if (ret) - goto out; - bio = req->bio; - bio->bi_disk = disk; - if (disk && meta_buffer && meta_len) { - meta = nvme_add_user_metadata(bio, meta_buffer, meta_len, - meta_seed, write); - if (IS_ERR(meta)) { - ret = PTR_ERR(meta); - goto out_unmap; - } - req->cmd_flags |= REQ_INTEGRITY; + switch (cmd->common.opcode) { + case nvme_admin_set_features: + switch (le32_to_cpu(cmd->common.cdw10) & 0xFF) { + case NVME_FEAT_KATO: + /* + * Keep alive commands interval on the host should be + * updated when KATO is modified by Set Features + * commands. + */ + if (!status) + nvme_update_keep_alive(ctrl, cmd); + break; + default: + break; } + break; + default: + break; } +} +EXPORT_SYMBOL_NS_GPL(nvme_passthru_end, "NVME_TARGET_PASSTHRU"); - blk_execute_rq(req->q, disk, req, 0); - if (nvme_req(req)->flags & NVME_REQ_CANCELLED) - ret = -EINTR; +/* + * Recommended frequency for KATO commands per NVMe 1.4 section 7.12.1: + * + * The host should send Keep Alive commands at half of the Keep Alive Timeout + * accounting for transport roundtrip times [..]. + */ +static unsigned long nvme_keep_alive_work_period(struct nvme_ctrl *ctrl) +{ + unsigned long delay = ctrl->kato * HZ / 2; + + /* + * When using Traffic Based Keep Alive, we need to run + * nvme_keep_alive_work at twice the normal frequency, as one + * command completion can postpone sending a keep alive command + * by up to twice the delay between runs. + */ + if (ctrl->ctratt & NVME_CTRL_ATTR_TBKAS) + delay /= 2; + return delay; +} + +static void nvme_queue_keep_alive_work(struct nvme_ctrl *ctrl) +{ + unsigned long now = jiffies; + unsigned long delay = nvme_keep_alive_work_period(ctrl); + unsigned long ka_next_check_tm = ctrl->ka_last_check_time + delay; + + if (time_after(now, ka_next_check_tm)) + delay = 0; else - ret = nvme_req(req)->status; - if (result) - *result = le64_to_cpu(nvme_req(req)->result.u64); - if (meta && !ret && !write) { - if (copy_to_user(meta_buffer, meta, meta_len)) - ret = -EFAULT; - } - kfree(meta); - out_unmap: - if (bio) - blk_rq_unmap_user(bio); - out: - blk_mq_free_request(req); - return ret; + delay = ka_next_check_tm - now; + + queue_delayed_work(nvme_wq, &ctrl->ka_work, delay); } -static void nvme_keep_alive_end_io(struct request *rq, blk_status_t status) +static enum rq_end_io_ret nvme_keep_alive_end_io(struct request *rq, + blk_status_t status) { struct nvme_ctrl *ctrl = rq->end_io_data; - unsigned long flags; - bool startka = false; + unsigned long rtt = jiffies - (rq->deadline - rq->timeout); + unsigned long delay = nvme_keep_alive_work_period(ctrl); + enum nvme_ctrl_state state = nvme_ctrl_state(ctrl); + + /* + * Subtract off the keepalive RTT so nvme_keep_alive_work runs + * at the desired frequency. + */ + if (rtt <= delay) { + delay -= rtt; + } else { + dev_warn(ctrl->device, "long keepalive RTT (%u ms)\n", + jiffies_to_msecs(rtt)); + delay = 0; + } blk_mq_free_request(rq); @@ -954,34 +1358,14 @@ static void nvme_keep_alive_end_io(struct request *rq, blk_status_t status) dev_err(ctrl->device, "failed nvme_keep_alive_end_io error=%d\n", status); - return; + return RQ_END_IO_NONE; } + ctrl->ka_last_check_time = jiffies; ctrl->comp_seen = false; - spin_lock_irqsave(&ctrl->lock, flags); - if (ctrl->state == NVME_CTRL_LIVE || - ctrl->state == NVME_CTRL_CONNECTING) - startka = true; - spin_unlock_irqrestore(&ctrl->lock, flags); - if (startka) - queue_delayed_work(nvme_wq, &ctrl->ka_work, ctrl->kato * HZ); -} - -static int nvme_keep_alive(struct nvme_ctrl *ctrl) -{ - struct request *rq; - - rq = nvme_alloc_request(ctrl->admin_q, &ctrl->ka_cmd, BLK_MQ_REQ_RESERVED, - NVME_QID_ANY); - if (IS_ERR(rq)) - return PTR_ERR(rq); - - rq->timeout = ctrl->kato * HZ; - rq->end_io_data = ctrl; - - blk_execute_rq_nowait(rq->q, NULL, rq, 0, nvme_keep_alive_end_io); - - return 0; + if (state == NVME_CTRL_LIVE || state == NVME_CTRL_CONNECTING) + queue_delayed_work(nvme_wq, &ctrl->ka_work, delay); + return RQ_END_IO_NONE; } static void nvme_keep_alive_work(struct work_struct *work) @@ -989,21 +1373,32 @@ static void nvme_keep_alive_work(struct work_struct *work) struct nvme_ctrl *ctrl = container_of(to_delayed_work(work), struct nvme_ctrl, ka_work); bool comp_seen = ctrl->comp_seen; + struct request *rq; + + ctrl->ka_last_check_time = jiffies; if ((ctrl->ctratt & NVME_CTRL_ATTR_TBKAS) && comp_seen) { dev_dbg(ctrl->device, "reschedule traffic based keep-alive timer\n"); ctrl->comp_seen = false; - queue_delayed_work(nvme_wq, &ctrl->ka_work, ctrl->kato * HZ); + nvme_queue_keep_alive_work(ctrl); return; } - if (nvme_keep_alive(ctrl)) { + rq = blk_mq_alloc_request(ctrl->admin_q, nvme_req_op(&ctrl->ka_cmd), + BLK_MQ_REQ_RESERVED | BLK_MQ_REQ_NOWAIT); + if (IS_ERR(rq)) { /* allocation failure, reset the controller */ - dev_err(ctrl->device, "keep-alive failed\n"); + dev_err(ctrl->device, "keep-alive failed: %ld\n", PTR_ERR(rq)); nvme_reset_ctrl(ctrl); return; } + nvme_init_request(rq, &ctrl->ka_cmd); + + rq->timeout = ctrl->kato * HZ; + rq->end_io = nvme_keep_alive_end_io; + rq->end_io_data = ctrl; + blk_execute_rq_nowait(rq, false); } static void nvme_start_keep_alive(struct nvme_ctrl *ctrl) @@ -1011,7 +1406,7 @@ static void nvme_start_keep_alive(struct nvme_ctrl *ctrl) if (unlikely(ctrl->kato == 0)) return; - queue_delayed_work(nvme_wq, &ctrl->ka_work, ctrl->kato * HZ); + nvme_queue_keep_alive_work(ctrl); } void nvme_stop_keep_alive(struct nvme_ctrl *ctrl) @@ -1023,17 +1418,45 @@ void nvme_stop_keep_alive(struct nvme_ctrl *ctrl) } EXPORT_SYMBOL_GPL(nvme_stop_keep_alive); -/* - * In NVMe 1.0 the CNS field was just a binary controller or namespace - * flag, thus sending any new CNS opcodes has a big chance of not working. - * Qemu unfortunately had that bug after reporting a 1.1 version compliance - * (but not for any later version). - */ -static bool nvme_ctrl_limited_cns(struct nvme_ctrl *ctrl) +static void nvme_update_keep_alive(struct nvme_ctrl *ctrl, + struct nvme_command *cmd) { - if (ctrl->quirks & NVME_QUIRK_IDENTIFY_CNS) - return ctrl->vs < NVME_VS(1, 2, 0); - return ctrl->vs < NVME_VS(1, 1, 0); + unsigned int new_kato = + DIV_ROUND_UP(le32_to_cpu(cmd->common.cdw11), 1000); + + dev_info(ctrl->device, + "keep alive interval updated from %u ms to %u ms\n", + ctrl->kato * 1000 / 2, new_kato * 1000 / 2); + + nvme_stop_keep_alive(ctrl); + ctrl->kato = new_kato; + nvme_start_keep_alive(ctrl); +} + +static bool nvme_id_cns_ok(struct nvme_ctrl *ctrl, u8 cns) +{ + /* + * The CNS field occupies a full byte starting with NVMe 1.2 + */ + if (ctrl->vs >= NVME_VS(1, 2, 0)) + return true; + + /* + * NVMe 1.1 expanded the CNS value to two bits, which means values + * larger than that could get truncated and treated as an incorrect + * value. + * + * Qemu implemented 1.0 behavior for controllers claiming 1.1 + * compliance, so they need to be quirked here. + */ + if (ctrl->vs >= NVME_VS(1, 1, 0) && + !(ctrl->quirks & NVME_QUIRK_IDENTIFY_CNS)) + return cns <= 3; + + /* + * NVMe 1.0 used a single bit for the CNS value. + */ + return cns <= 1; } static int nvme_identify_ctrl(struct nvme_ctrl *dev, struct nvme_id_ctrl **id) @@ -1051,13 +1474,15 @@ static int nvme_identify_ctrl(struct nvme_ctrl *dev, struct nvme_id_ctrl **id) error = nvme_submit_sync_cmd(dev->admin_q, &c, *id, sizeof(struct nvme_id_ctrl)); - if (error) + if (error) { kfree(*id); + *id = NULL; + } return error; } static int nvme_process_ns_desc(struct nvme_ctrl *ctrl, struct nvme_ns_ids *ids, - struct nvme_ns_id_desc *cur) + struct nvme_ns_id_desc *cur, bool *csi_seen) { const char *warn_str = "ctrl returned bogus length:"; void *data = cur; @@ -1069,6 +1494,8 @@ static int nvme_process_ns_desc(struct nvme_ctrl *ctrl, struct nvme_ns_ids *ids, warn_str, cur->nidl); return -1; } + if (ctrl->quirks & NVME_QUIRK_BOGUS_NID) + return NVME_NIDT_EUI64_LEN; memcpy(ids->eui64, data + sizeof(*cur), NVME_NIDT_EUI64_LEN); return NVME_NIDT_EUI64_LEN; case NVME_NIDT_NGUID: @@ -1077,6 +1504,8 @@ static int nvme_process_ns_desc(struct nvme_ctrl *ctrl, struct nvme_ns_ids *ids, warn_str, cur->nidl); return -1; } + if (ctrl->quirks & NVME_QUIRK_BOGUS_NID) + return NVME_NIDT_NGUID_LEN; memcpy(ids->nguid, data + sizeof(*cur), NVME_NIDT_NGUID_LEN); return NVME_NIDT_NGUID_LEN; case NVME_NIDT_UUID: @@ -1085,25 +1514,40 @@ static int nvme_process_ns_desc(struct nvme_ctrl *ctrl, struct nvme_ns_ids *ids, warn_str, cur->nidl); return -1; } + if (ctrl->quirks & NVME_QUIRK_BOGUS_NID) + return NVME_NIDT_UUID_LEN; uuid_copy(&ids->uuid, data + sizeof(*cur)); return NVME_NIDT_UUID_LEN; + case NVME_NIDT_CSI: + if (cur->nidl != NVME_NIDT_CSI_LEN) { + dev_warn(ctrl->device, "%s %d for NVME_NIDT_CSI\n", + warn_str, cur->nidl); + return -1; + } + memcpy(&ids->csi, data + sizeof(*cur), NVME_NIDT_CSI_LEN); + *csi_seen = true; + return NVME_NIDT_CSI_LEN; default: /* Skip unknown types */ return cur->nidl; } } -static int nvme_identify_ns_descs(struct nvme_ctrl *ctrl, unsigned nsid, - struct nvme_ns_ids *ids) +static int nvme_identify_ns_descs(struct nvme_ctrl *ctrl, + struct nvme_ns_info *info) { struct nvme_command c = { }; - int status; + bool csi_seen = false; + int status, pos, len; void *data; - int pos; - int len; + + if (ctrl->vs < NVME_VS(1, 3, 0) && !nvme_multi_css(ctrl)) + return 0; + if (ctrl->quirks & NVME_QUIRK_NO_NS_DESC_LIST) + return 0; c.identify.opcode = nvme_admin_identify; - c.identify.nsid = cpu_to_le32(nsid); + c.identify.nsid = cpu_to_le32(info->nsid); c.identify.cns = NVME_ID_CNS_NS_DESC_LIST; data = kzalloc(NVME_IDENTIFY_DATA_SIZE, GFP_KERNEL); @@ -1114,13 +1558,8 @@ static int nvme_identify_ns_descs(struct nvme_ctrl *ctrl, unsigned nsid, NVME_IDENTIFY_DATA_SIZE); if (status) { dev_warn(ctrl->device, - "Identify Descriptors failed (%d)\n", status); - /* - * Don't treat an error as fatal, as we potentially already - * have a NGUID or EUI-64. - */ - if (status > 0 && !(status & NVME_SC_DNR)) - status = 0; + "Identify Descriptors failed (nsid=%u, status=0x%x)\n", + info->nsid, status); goto free_data; } @@ -1130,30 +1569,26 @@ static int nvme_identify_ns_descs(struct nvme_ctrl *ctrl, unsigned nsid, if (cur->nidl == 0) break; - len = nvme_process_ns_desc(ctrl, ids, cur); + len = nvme_process_ns_desc(ctrl, &info->ids, cur, &csi_seen); if (len < 0) - goto free_data; + break; len += sizeof(*cur); } + + if (nvme_multi_css(ctrl) && !csi_seen) { + dev_warn(ctrl->device, "Command set not reported for nsid:%d\n", + info->nsid); + status = -EINVAL; + } + free_data: kfree(data); return status; } -static int nvme_identify_ns_list(struct nvme_ctrl *dev, unsigned nsid, __le32 *ns_list) -{ - struct nvme_command c = { }; - - c.identify.opcode = nvme_admin_identify; - c.identify.cns = NVME_ID_CNS_NS_ACTIVE_LIST; - c.identify.nsid = cpu_to_le32(nsid); - return nvme_submit_sync_cmd(dev->admin_q, &c, ns_list, - NVME_IDENTIFY_DATA_SIZE); -} - -static int nvme_identify_ns(struct nvme_ctrl *ctrl, - unsigned nsid, struct nvme_id_ns **id) +int nvme_identify_ns(struct nvme_ctrl *ctrl, unsigned nsid, + struct nvme_id_ns **id) { struct nvme_command c = { }; int error; @@ -1171,25 +1606,93 @@ static int nvme_identify_ns(struct nvme_ctrl *ctrl, if (error) { dev_warn(ctrl->device, "Identify namespace failed (%d)\n", error); kfree(*id); + *id = NULL; } - return error; } +static int nvme_ns_info_from_identify(struct nvme_ctrl *ctrl, + struct nvme_ns_info *info) +{ + struct nvme_ns_ids *ids = &info->ids; + struct nvme_id_ns *id; + int ret; + + ret = nvme_identify_ns(ctrl, info->nsid, &id); + if (ret) + return ret; + + if (id->ncap == 0) { + /* namespace not allocated or attached */ + info->is_removed = true; + ret = -ENODEV; + goto error; + } + + info->anagrpid = id->anagrpid; + info->is_shared = id->nmic & NVME_NS_NMIC_SHARED; + info->is_readonly = id->nsattr & NVME_NS_ATTR_RO; + info->is_ready = true; + info->endgid = le16_to_cpu(id->endgid); + if (ctrl->quirks & NVME_QUIRK_BOGUS_NID) { + dev_info(ctrl->device, + "Ignoring bogus Namespace Identifiers\n"); + } else { + if (ctrl->vs >= NVME_VS(1, 1, 0) && + !memchr_inv(ids->eui64, 0, sizeof(ids->eui64))) + memcpy(ids->eui64, id->eui64, sizeof(ids->eui64)); + if (ctrl->vs >= NVME_VS(1, 2, 0) && + !memchr_inv(ids->nguid, 0, sizeof(ids->nguid))) + memcpy(ids->nguid, id->nguid, sizeof(ids->nguid)); + } + +error: + kfree(id); + return ret; +} + +static int nvme_ns_info_from_id_cs_indep(struct nvme_ctrl *ctrl, + struct nvme_ns_info *info) +{ + struct nvme_id_ns_cs_indep *id; + struct nvme_command c = { + .identify.opcode = nvme_admin_identify, + .identify.nsid = cpu_to_le32(info->nsid), + .identify.cns = NVME_ID_CNS_NS_CS_INDEP, + }; + int ret; + + id = kmalloc(sizeof(*id), GFP_KERNEL); + if (!id) + return -ENOMEM; + + ret = nvme_submit_sync_cmd(ctrl->admin_q, &c, id, sizeof(*id)); + if (!ret) { + info->anagrpid = id->anagrpid; + info->is_shared = id->nmic & NVME_NS_NMIC_SHARED; + info->is_readonly = id->nsattr & NVME_NS_ATTR_RO; + info->is_ready = id->nstat & NVME_NSTAT_NRDY; + info->is_rotational = id->nsfeat & NVME_NS_ROTATIONAL; + info->no_vwc = id->nsfeat & NVME_NS_VWC_NOT_PRESENT; + info->endgid = le16_to_cpu(id->endgid); + } + kfree(id); + return ret; +} + static int nvme_features(struct nvme_ctrl *dev, u8 op, unsigned int fid, unsigned int dword11, void *buffer, size_t buflen, u32 *result) { union nvme_result res = { 0 }; - struct nvme_command c; + struct nvme_command c = { }; int ret; - memset(&c, 0, sizeof(c)); c.features.opcode = op; c.features.fid = cpu_to_le32(fid); c.features.dword11 = cpu_to_le32(dword11); ret = __nvme_submit_sync_cmd(dev->admin_q, &c, &res, - buffer, buflen, 0, NVME_QID_ANY, 0, 0, false); + buffer, buflen, NVME_QID_ANY, 0); if (ret >= 0 && result) *result = le32_to_cpu(res.u32); return ret; @@ -1197,7 +1700,7 @@ static int nvme_features(struct nvme_ctrl *dev, u8 op, unsigned int fid, int nvme_set_features(struct nvme_ctrl *dev, unsigned int fid, unsigned int dword11, void *buffer, size_t buflen, - u32 *result) + void *result) { return nvme_features(dev, nvme_admin_set_features, fid, dword11, buffer, buflen, result); @@ -1206,7 +1709,7 @@ EXPORT_SYMBOL_GPL(nvme_set_features); int nvme_get_features(struct nvme_ctrl *dev, unsigned int fid, unsigned int dword11, void *buffer, size_t buflen, - u32 *result) + void *result) { return nvme_features(dev, nvme_admin_get_features, fid, dword11, buffer, buflen, result); @@ -1221,7 +1724,13 @@ int nvme_set_queue_count(struct nvme_ctrl *ctrl, int *count) status = nvme_set_features(ctrl, NVME_FEAT_NUM_QUEUES, q_count, NULL, 0, &result); - if (status < 0) + + /* + * It's either a kernel error or the host observed a connection + * lost. In either case it's not possible communicate with the + * controller and thus enter the error code path. + */ + if (status < 0 || status == NVME_SC_HOST_PATH_ERROR) return status; /* @@ -1262,880 +1771,802 @@ static void nvme_enable_aen(struct nvme_ctrl *ctrl) queue_work(nvme_wq, &ctrl->async_event_work); } -/* - * Convert integer values from ioctl structures to user pointers, silently - * ignoring the upper bits in the compat case to match behaviour of 32-bit - * kernels. - */ -static void __user *nvme_to_user_ptr(uintptr_t ptrval) +static int nvme_ns_open(struct nvme_ns *ns) { - if (in_compat_syscall()) - ptrval = (compat_uptr_t)ptrval; - return (void __user *)ptrval; -} -static int nvme_submit_io(struct nvme_ns *ns, struct nvme_user_io __user *uio) -{ - struct nvme_user_io io; - struct nvme_command c; - unsigned length, meta_len; - void __user *metadata; + /* should never be called due to GENHD_FL_HIDDEN */ + if (WARN_ON_ONCE(nvme_ns_head_multipath(ns->head))) + goto fail; + if (!nvme_get_ns(ns)) + goto fail; + if (!try_module_get(ns->ctrl->ops->module)) + goto fail_put_ns; - if (copy_from_user(&io, uio, sizeof(io))) - return -EFAULT; - if (io.flags) - return -EINVAL; + return 0; - switch (io.opcode) { - case nvme_cmd_write: - case nvme_cmd_read: - case nvme_cmd_compare: - break; - default: - return -EINVAL; - } +fail_put_ns: + nvme_put_ns(ns); +fail: + return -ENXIO; +} - length = (io.nblocks + 1) << ns->lba_shift; - meta_len = (io.nblocks + 1) * ns->ms; - metadata = nvme_to_user_ptr(io.metadata); +static void nvme_ns_release(struct nvme_ns *ns) +{ - if (ns->features & NVME_NS_EXT_LBAS) { - length += meta_len; - meta_len = 0; - } else if (meta_len) { - if ((io.metadata & 3) || !io.metadata) - return -EINVAL; - } + module_put(ns->ctrl->ops->module); + nvme_put_ns(ns); +} - memset(&c, 0, sizeof(c)); - c.rw.opcode = io.opcode; - c.rw.flags = io.flags; - c.rw.nsid = cpu_to_le32(ns->head->ns_id); - c.rw.slba = cpu_to_le64(io.slba); - c.rw.length = cpu_to_le16(io.nblocks); - c.rw.control = cpu_to_le16(io.control); - c.rw.dsmgmt = cpu_to_le32(io.dsmgmt); - c.rw.reftag = cpu_to_le32(io.reftag); - c.rw.apptag = cpu_to_le16(io.apptag); - c.rw.appmask = cpu_to_le16(io.appmask); +static int nvme_open(struct gendisk *disk, blk_mode_t mode) +{ + return nvme_ns_open(disk->private_data); +} - return nvme_submit_user_cmd(ns->queue, &c, - nvme_to_user_ptr(io.addr), length, - metadata, meta_len, lower_32_bits(io.slba), NULL, 0); +static void nvme_release(struct gendisk *disk) +{ + nvme_ns_release(disk->private_data); } -static u32 nvme_known_admin_effects(u8 opcode) +int nvme_getgeo(struct gendisk *disk, struct hd_geometry *geo) { - switch (opcode) { - case nvme_admin_format_nvm: - return NVME_CMD_EFFECTS_CSUPP | NVME_CMD_EFFECTS_LBCC | - NVME_CMD_EFFECTS_CSE_MASK; - case nvme_admin_sanitize_nvm: - return NVME_CMD_EFFECTS_CSE_MASK; - default: - break; - } + /* some standard values */ + geo->heads = 1 << 6; + geo->sectors = 1 << 5; + geo->cylinders = get_capacity(disk) >> 11; return 0; } -static u32 nvme_passthru_start(struct nvme_ctrl *ctrl, struct nvme_ns *ns, - u8 opcode) +static bool nvme_init_integrity(struct nvme_ns_head *head, + struct queue_limits *lim, struct nvme_ns_info *info) { - u32 effects = 0; + struct blk_integrity *bi = &lim->integrity; - if (ns) { - if (ctrl->effects) - effects = le32_to_cpu(ctrl->effects->iocs[opcode]); - if (effects & ~(NVME_CMD_EFFECTS_CSUPP | NVME_CMD_EFFECTS_LBCC)) - dev_warn(ctrl->device, - "IO command:%02x has unhandled effects:%08x\n", - opcode, effects); - return 0; - } + memset(bi, 0, sizeof(*bi)); - if (ctrl->effects) - effects = le32_to_cpu(ctrl->effects->acs[opcode]); - effects |= nvme_known_admin_effects(opcode); + if (!head->ms) + return true; /* - * For simplicity, IO to all namespaces is quiesced even if the command - * effects say only one namespace is affected. + * PI can always be supported as we can ask the controller to simply + * insert/strip it, which is not possible for other kinds of metadata. */ - if (effects & (NVME_CMD_EFFECTS_LBCC | NVME_CMD_EFFECTS_CSE_MASK)) { - mutex_lock(&ctrl->scan_lock); - mutex_lock(&ctrl->subsys->lock); - nvme_mpath_start_freeze(ctrl->subsys); - nvme_mpath_wait_freeze(ctrl->subsys); - nvme_start_freeze(ctrl); - nvme_wait_freeze(ctrl); - } - return effects; -} - -static void nvme_update_formats(struct nvme_ctrl *ctrl) -{ - struct nvme_ns *ns; - - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) - if (ns->disk && nvme_revalidate_disk(ns->disk)) - nvme_set_queue_dying(ns); - up_read(&ctrl->namespaces_rwsem); -} + if (!IS_ENABLED(CONFIG_BLK_DEV_INTEGRITY) || + !(head->features & NVME_NS_METADATA_SUPPORTED)) + return nvme_ns_has_pi(head); -static void nvme_passthru_end(struct nvme_ctrl *ctrl, u32 effects) -{ - /* - * Revalidate LBA changes prior to unfreezing. This is necessary to - * prevent memory corruption if a logical block size was changed by - * this command. - */ - if (effects & NVME_CMD_EFFECTS_LBCC) - nvme_update_formats(ctrl); - if (effects & (NVME_CMD_EFFECTS_LBCC | NVME_CMD_EFFECTS_CSE_MASK)) { - nvme_unfreeze(ctrl); - nvme_mpath_unfreeze(ctrl->subsys); - mutex_unlock(&ctrl->subsys->lock); - nvme_remove_invalid_namespaces(ctrl, NVME_NSID_ALL); - mutex_unlock(&ctrl->scan_lock); + switch (head->pi_type) { + case NVME_NS_DPS_PI_TYPE3: + switch (head->guard_type) { + case NVME_NVM_NS_16B_GUARD: + bi->csum_type = BLK_INTEGRITY_CSUM_CRC; + bi->tag_size = sizeof(u16) + sizeof(u32); + bi->flags |= BLK_INTEGRITY_DEVICE_CAPABLE; + break; + case NVME_NVM_NS_64B_GUARD: + bi->csum_type = BLK_INTEGRITY_CSUM_CRC64; + bi->tag_size = sizeof(u16) + 6; + bi->flags |= BLK_INTEGRITY_DEVICE_CAPABLE; + break; + default: + break; + } + break; + case NVME_NS_DPS_PI_TYPE1: + case NVME_NS_DPS_PI_TYPE2: + switch (head->guard_type) { + case NVME_NVM_NS_16B_GUARD: + bi->csum_type = BLK_INTEGRITY_CSUM_CRC; + bi->tag_size = sizeof(u16); + bi->flags |= BLK_INTEGRITY_DEVICE_CAPABLE | + BLK_INTEGRITY_REF_TAG; + break; + case NVME_NVM_NS_64B_GUARD: + bi->csum_type = BLK_INTEGRITY_CSUM_CRC64; + bi->tag_size = sizeof(u16); + bi->flags |= BLK_INTEGRITY_DEVICE_CAPABLE | + BLK_INTEGRITY_REF_TAG; + break; + default: + break; + } + break; + default: + break; } - if (effects & NVME_CMD_EFFECTS_CCC) - nvme_init_identify(ctrl); - if (effects & (NVME_CMD_EFFECTS_NIC | NVME_CMD_EFFECTS_NCC)) { - nvme_queue_scan(ctrl); - flush_work(&ctrl->scan_work); + + bi->metadata_size = head->ms; + if (bi->csum_type) { + bi->pi_tuple_size = head->pi_size; + bi->pi_offset = info->pi_offset; } + return true; } -static int nvme_user_cmd(struct nvme_ctrl *ctrl, struct nvme_ns *ns, - struct nvme_passthru_cmd __user *ucmd) +static void nvme_config_discard(struct nvme_ns *ns, struct queue_limits *lim) { - struct nvme_passthru_cmd cmd; - struct nvme_command c; - unsigned timeout = 0; - u32 effects; - u64 result; - int status; + struct nvme_ctrl *ctrl = ns->ctrl; - if (!capable(CAP_SYS_ADMIN)) - return -EACCES; - if (copy_from_user(&cmd, ucmd, sizeof(cmd))) - return -EFAULT; - if (cmd.flags) - return -EINVAL; + if (ctrl->dmrsl && ctrl->dmrsl <= nvme_sect_to_lba(ns->head, UINT_MAX)) + lim->max_hw_discard_sectors = + nvme_lba_to_sect(ns->head, ctrl->dmrsl); + else if (ctrl->oncs & NVME_CTRL_ONCS_DSM) + lim->max_hw_discard_sectors = UINT_MAX; + else + lim->max_hw_discard_sectors = 0; - memset(&c, 0, sizeof(c)); - c.common.opcode = cmd.opcode; - c.common.flags = cmd.flags; - c.common.nsid = cpu_to_le32(cmd.nsid); - c.common.cdw2[0] = cpu_to_le32(cmd.cdw2); - c.common.cdw2[1] = cpu_to_le32(cmd.cdw3); - c.common.cdw10 = cpu_to_le32(cmd.cdw10); - c.common.cdw11 = cpu_to_le32(cmd.cdw11); - c.common.cdw12 = cpu_to_le32(cmd.cdw12); - c.common.cdw13 = cpu_to_le32(cmd.cdw13); - c.common.cdw14 = cpu_to_le32(cmd.cdw14); - c.common.cdw15 = cpu_to_le32(cmd.cdw15); - - if (cmd.timeout_ms) - timeout = msecs_to_jiffies(cmd.timeout_ms); - - effects = nvme_passthru_start(ctrl, ns, cmd.opcode); - status = nvme_submit_user_cmd(ns ? ns->queue : ctrl->admin_q, &c, - nvme_to_user_ptr(cmd.addr), cmd.data_len, - nvme_to_user_ptr(cmd.metadata), cmd.metadata_len, - 0, &result, timeout); - nvme_passthru_end(ctrl, effects); - - if (status >= 0) { - if (put_user(result, &ucmd->result)) - return -EFAULT; - } + lim->discard_granularity = lim->logical_block_size; - return status; + if (ctrl->dmrl) + lim->max_discard_segments = ctrl->dmrl; + else + lim->max_discard_segments = NVME_DSM_MAX_RANGES; } -static int nvme_user_cmd64(struct nvme_ctrl *ctrl, struct nvme_ns *ns, - struct nvme_passthru_cmd64 __user *ucmd) +static bool nvme_ns_ids_equal(struct nvme_ns_ids *a, struct nvme_ns_ids *b) { - struct nvme_passthru_cmd64 cmd; - struct nvme_command c; - unsigned timeout = 0; - u32 effects; - int status; - - if (!capable(CAP_SYS_ADMIN)) - return -EACCES; - if (copy_from_user(&cmd, ucmd, sizeof(cmd))) - return -EFAULT; - if (cmd.flags) - return -EINVAL; - - memset(&c, 0, sizeof(c)); - c.common.opcode = cmd.opcode; - c.common.flags = cmd.flags; - c.common.nsid = cpu_to_le32(cmd.nsid); - c.common.cdw2[0] = cpu_to_le32(cmd.cdw2); - c.common.cdw2[1] = cpu_to_le32(cmd.cdw3); - c.common.cdw10 = cpu_to_le32(cmd.cdw10); - c.common.cdw11 = cpu_to_le32(cmd.cdw11); - c.common.cdw12 = cpu_to_le32(cmd.cdw12); - c.common.cdw13 = cpu_to_le32(cmd.cdw13); - c.common.cdw14 = cpu_to_le32(cmd.cdw14); - c.common.cdw15 = cpu_to_le32(cmd.cdw15); - - if (cmd.timeout_ms) - timeout = msecs_to_jiffies(cmd.timeout_ms); - - effects = nvme_passthru_start(ctrl, ns, cmd.opcode); - status = nvme_submit_user_cmd(ns ? ns->queue : ctrl->admin_q, &c, - nvme_to_user_ptr(cmd.addr), cmd.data_len, - nvme_to_user_ptr(cmd.metadata), cmd.metadata_len, - 0, &cmd.result, timeout); - nvme_passthru_end(ctrl, effects); - - if (status >= 0) { - if (put_user(cmd.result, &ucmd->result)) - return -EFAULT; - } - - return status; + return uuid_equal(&a->uuid, &b->uuid) && + memcmp(&a->nguid, &b->nguid, sizeof(a->nguid)) == 0 && + memcmp(&a->eui64, &b->eui64, sizeof(a->eui64)) == 0 && + a->csi == b->csi; } -/* - * Issue ioctl requests on the first available path. Note that unlike normal - * block layer requests we will not retry failed request on another controller. - */ -static struct nvme_ns *nvme_get_ns_from_disk(struct gendisk *disk, - struct nvme_ns_head **head, int *srcu_idx) +static int nvme_identify_ns_nvm(struct nvme_ctrl *ctrl, unsigned int nsid, + struct nvme_id_ns_nvm **nvmp) { -#ifdef CONFIG_NVME_MULTIPATH - if (disk->fops == &nvme_ns_head_ops) { - struct nvme_ns *ns; + struct nvme_command c = { + .identify.opcode = nvme_admin_identify, + .identify.nsid = cpu_to_le32(nsid), + .identify.cns = NVME_ID_CNS_CS_NS, + .identify.csi = NVME_CSI_NVM, + }; + struct nvme_id_ns_nvm *nvm; + int ret; - *head = disk->private_data; - *srcu_idx = srcu_read_lock(&(*head)->srcu); - ns = nvme_find_path(*head); - if (!ns) - srcu_read_unlock(&(*head)->srcu, *srcu_idx); - return ns; - } -#endif - *head = NULL; - *srcu_idx = -1; - return disk->private_data; -} + nvm = kzalloc(sizeof(*nvm), GFP_KERNEL); + if (!nvm) + return -ENOMEM; -static void nvme_put_ns_from_disk(struct nvme_ns_head *head, int idx) -{ - if (head) - srcu_read_unlock(&head->srcu, idx); + ret = nvme_submit_sync_cmd(ctrl->admin_q, &c, nvm, sizeof(*nvm)); + if (ret) + kfree(nvm); + else + *nvmp = nvm; + return ret; } -static bool is_ctrl_ioctl(unsigned int cmd) +static void nvme_configure_pi_elbas(struct nvme_ns_head *head, + struct nvme_id_ns *id, struct nvme_id_ns_nvm *nvm) { - if (cmd == NVME_IOCTL_ADMIN_CMD || cmd == NVME_IOCTL_ADMIN64_CMD) - return true; - if (is_sed_ioctl(cmd)) - return true; - return false; -} + u32 elbaf = le32_to_cpu(nvm->elbaf[nvme_lbaf_index(id->flbas)]); + u8 guard_type; -static int nvme_handle_ctrl_ioctl(struct nvme_ns *ns, unsigned int cmd, - void __user *argp, - struct nvme_ns_head *head, - int srcu_idx) -{ - struct nvme_ctrl *ctrl = ns->ctrl; - int ret; + /* no support for storage tag formats right now */ + if (nvme_elbaf_sts(elbaf)) + return; - nvme_get_ctrl(ns->ctrl); - nvme_put_ns_from_disk(head, srcu_idx); + guard_type = nvme_elbaf_guard_type(elbaf); + if ((nvm->pic & NVME_ID_NS_NVM_QPIFS) && + guard_type == NVME_NVM_NS_QTYPE_GUARD) + guard_type = nvme_elbaf_qualified_guard_type(elbaf); - switch (cmd) { - case NVME_IOCTL_ADMIN_CMD: - ret = nvme_user_cmd(ctrl, NULL, argp); + head->guard_type = guard_type; + switch (head->guard_type) { + case NVME_NVM_NS_64B_GUARD: + head->pi_size = sizeof(struct crc64_pi_tuple); break; - case NVME_IOCTL_ADMIN64_CMD: - ret = nvme_user_cmd64(ctrl, NULL, argp); + case NVME_NVM_NS_16B_GUARD: + head->pi_size = sizeof(struct t10_pi_tuple); break; default: - ret = sed_ioctl(ctrl->opal_dev, cmd, argp); break; } - nvme_put_ctrl(ctrl); - return ret; } -static int nvme_ioctl(struct block_device *bdev, fmode_t mode, - unsigned int cmd, unsigned long arg) +static void nvme_configure_metadata(struct nvme_ctrl *ctrl, + struct nvme_ns_head *head, struct nvme_id_ns *id, + struct nvme_id_ns_nvm *nvm, struct nvme_ns_info *info) { - struct nvme_ns_head *head = NULL; - void __user *argp = (void __user *)arg; - struct nvme_ns *ns; - int srcu_idx, ret; - - ns = nvme_get_ns_from_disk(bdev->bd_disk, &head, &srcu_idx); - if (unlikely(!ns)) - return -EWOULDBLOCK; + head->features &= ~(NVME_NS_METADATA_SUPPORTED | NVME_NS_EXT_LBAS); + head->pi_type = 0; + head->pi_size = 0; + head->ms = le16_to_cpu(id->lbaf[nvme_lbaf_index(id->flbas)].ms); + if (!head->ms || !(ctrl->ops->flags & NVME_F_METADATA_SUPPORTED)) + return; - /* - * Handle ioctls that apply to the controller instead of the namespace - * seperately and drop the ns SRCU reference early. This avoids a - * deadlock when deleting namespaces using the passthrough interface. - */ - if (is_ctrl_ioctl(cmd)) - return nvme_handle_ctrl_ioctl(ns, cmd, argp, head, srcu_idx); + if (nvm && (ctrl->ctratt & NVME_CTRL_ATTR_ELBAS)) { + nvme_configure_pi_elbas(head, id, nvm); + } else { + head->pi_size = sizeof(struct t10_pi_tuple); + head->guard_type = NVME_NVM_NS_16B_GUARD; + } - switch (cmd) { - case NVME_IOCTL_ID: - force_successful_syscall_return(); - ret = ns->head->ns_id; - break; - case NVME_IOCTL_IO_CMD: - ret = nvme_user_cmd(ns->ctrl, ns, argp); - break; - case NVME_IOCTL_SUBMIT_IO: - ret = nvme_submit_io(ns, argp); - break; - case NVME_IOCTL_IO64_CMD: - ret = nvme_user_cmd64(ns->ctrl, ns, argp); - break; - default: - if (ns->ndev) - ret = nvme_nvm_ioctl(ns, cmd, arg); + if (head->pi_size && head->ms >= head->pi_size) + head->pi_type = id->dps & NVME_NS_DPS_PI_MASK; + if (!(id->dps & NVME_NS_DPS_PI_FIRST)) { + if (disable_pi_offsets) + head->pi_type = 0; else - ret = -ENOTTY; + info->pi_offset = head->ms - head->pi_size; } - nvme_put_ns_from_disk(head, srcu_idx); - return ret; -} - -#ifdef CONFIG_COMPAT -struct nvme_user_io32 { - __u8 opcode; - __u8 flags; - __u16 control; - __u16 nblocks; - __u16 rsvd; - __u64 metadata; - __u64 addr; - __u64 slba; - __u32 dsmgmt; - __u32 reftag; - __u16 apptag; - __u16 appmask; -} __attribute__((__packed__)); + if (ctrl->ops->flags & NVME_F_FABRICS) { + /* + * The NVMe over Fabrics specification only supports metadata as + * part of the extended data LBA. We rely on HCA/HBA support to + * remap the separate metadata buffer from the block layer. + */ + if (WARN_ON_ONCE(!(id->flbas & NVME_NS_FLBAS_META_EXT))) + return; -#define NVME_IOCTL_SUBMIT_IO32 _IOW('N', 0x42, struct nvme_user_io32) + head->features |= NVME_NS_EXT_LBAS; -static int nvme_compat_ioctl(struct block_device *bdev, fmode_t mode, - unsigned int cmd, unsigned long arg) -{ - /* - * Corresponds to the difference of NVME_IOCTL_SUBMIT_IO - * between 32 bit programs and 64 bit kernel. - * The cause is that the results of sizeof(struct nvme_user_io), - * which is used to define NVME_IOCTL_SUBMIT_IO, - * are not same between 32 bit compiler and 64 bit compiler. - * NVME_IOCTL_SUBMIT_IO32 is for 64 bit kernel handling - * NVME_IOCTL_SUBMIT_IO issued from 32 bit programs. - * Other IOCTL numbers are same between 32 bit and 64 bit. - * So there is nothing to do regarding to other IOCTL numbers. - */ - if (cmd == NVME_IOCTL_SUBMIT_IO32) - return nvme_ioctl(bdev, mode, NVME_IOCTL_SUBMIT_IO, arg); - - return nvme_ioctl(bdev, mode, cmd, arg); + /* + * The current fabrics transport drivers support namespace + * metadata formats only if nvme_ns_has_pi() returns true. + * Suppress support for all other formats so the namespace will + * have a 0 capacity and not be usable through the block stack. + * + * Note, this check will need to be modified if any drivers + * gain the ability to use other metadata formats. + */ + if (ctrl->max_integrity_segments && nvme_ns_has_pi(head)) + head->features |= NVME_NS_METADATA_SUPPORTED; + } else { + /* + * For PCIe controllers, we can't easily remap the separate + * metadata buffer from the block layer and thus require a + * separate metadata buffer for block layer metadata/PI support. + * We allow extended LBAs for the passthrough interface, though. + */ + if (id->flbas & NVME_NS_FLBAS_META_EXT) + head->features |= NVME_NS_EXT_LBAS; + else + head->features |= NVME_NS_METADATA_SUPPORTED; + } } -#else -#define nvme_compat_ioctl NULL -#endif /* CONFIG_COMPAT */ -static int nvme_open(struct block_device *bdev, fmode_t mode) + +static u32 nvme_configure_atomic_write(struct nvme_ns *ns, + struct nvme_id_ns *id, struct queue_limits *lim, u32 bs) { - struct nvme_ns *ns = bdev->bd_disk->private_data; + u32 atomic_bs, boundary = 0; -#ifdef CONFIG_NVME_MULTIPATH - /* should never be called due to GENHD_FL_HIDDEN */ - if (WARN_ON_ONCE(ns->head->disk)) - goto fail; -#endif - if (!kref_get_unless_zero(&ns->kref)) - goto fail; - if (!try_module_get(ns->ctrl->ops->module)) - goto fail_put_ns; + /* + * We do not support an offset for the atomic boundaries. + */ + if (id->nabo) + return bs; - return 0; + if ((id->nsfeat & NVME_NS_FEAT_ATOMICS) && id->nawupf) { + /* + * Use the per-namespace atomic write unit when available. + */ + atomic_bs = (1 + le16_to_cpu(id->nawupf)) * bs; + if (id->nabspf) + boundary = (le16_to_cpu(id->nabspf) + 1) * bs; + } else { + /* + * Use the controller wide atomic write unit. This sucks + * because the limit is defined in terms of logical blocks while + * namespaces can have different formats, and because there is + * no clear language in the specification prohibiting different + * values for different controllers in the subsystem. + */ + atomic_bs = (1 + ns->ctrl->subsys->awupf) * bs; + } -fail_put_ns: - nvme_put_ns(ns); -fail: - return -ENXIO; + lim->atomic_write_hw_max = atomic_bs; + lim->atomic_write_hw_boundary = boundary; + lim->atomic_write_hw_unit_min = bs; + lim->atomic_write_hw_unit_max = rounddown_pow_of_two(atomic_bs); + lim->features |= BLK_FEAT_ATOMIC_WRITES; + return atomic_bs; } -static void nvme_release(struct gendisk *disk, fmode_t mode) +static u32 nvme_max_drv_segments(struct nvme_ctrl *ctrl) { - struct nvme_ns *ns = disk->private_data; - - module_put(ns->ctrl->ops->module); - nvme_put_ns(ns); + return ctrl->max_hw_sectors / (NVME_CTRL_PAGE_SIZE >> SECTOR_SHIFT) + 1; } -static int nvme_getgeo(struct block_device *bdev, struct hd_geometry *geo) +static void nvme_set_ctrl_limits(struct nvme_ctrl *ctrl, + struct queue_limits *lim, bool is_admin) { - /* some standard values */ - geo->heads = 1 << 6; - geo->sectors = 1 << 5; - geo->cylinders = get_capacity(bdev->bd_disk) >> 11; - return 0; + lim->max_hw_sectors = ctrl->max_hw_sectors; + lim->max_segments = min_t(u32, USHRT_MAX, + min_not_zero(nvme_max_drv_segments(ctrl), ctrl->max_segments)); + lim->max_integrity_segments = ctrl->max_integrity_segments; + lim->virt_boundary_mask = ctrl->ops->get_virt_boundary(ctrl, is_admin); + lim->max_segment_size = UINT_MAX; + lim->dma_alignment = 3; } -#ifdef CONFIG_BLK_DEV_INTEGRITY -static void nvme_init_integrity(struct gendisk *disk, u16 ms, u8 pi_type, - u32 max_integrity_segments) +static bool nvme_update_disk_info(struct nvme_ns *ns, struct nvme_id_ns *id, + struct queue_limits *lim) { - struct blk_integrity integrity; + struct nvme_ns_head *head = ns->head; + u32 bs = 1U << head->lba_shift; + u32 atomic_bs, phys_bs, io_opt = 0; + bool valid = true; - memset(&integrity, 0, sizeof(integrity)); - switch (pi_type) { - case NVME_NS_DPS_PI_TYPE3: - integrity.profile = &t10_pi_type3_crc; - integrity.tag_size = sizeof(u16) + sizeof(u32); - integrity.flags |= BLK_INTEGRITY_DEVICE_CAPABLE; - break; - case NVME_NS_DPS_PI_TYPE1: - case NVME_NS_DPS_PI_TYPE2: - integrity.profile = &t10_pi_type1_crc; - integrity.tag_size = sizeof(u16); - integrity.flags |= BLK_INTEGRITY_DEVICE_CAPABLE; - break; - default: - integrity.profile = NULL; - break; + /* + * The block layer can't support LBA sizes larger than the page size + * or smaller than a sector size yet, so catch this early and don't + * allow block I/O. + */ + if (blk_validate_block_size(bs)) { + bs = (1 << 9); + valid = false; + } + + phys_bs = bs; + atomic_bs = nvme_configure_atomic_write(ns, id, lim, bs); + + if (id->nsfeat & NVME_NS_FEAT_IO_OPT) { + /* NPWG = Namespace Preferred Write Granularity */ + phys_bs = bs * (1 + le16_to_cpu(id->npwg)); + /* NOWS = Namespace Optimal Write Size */ + if (id->nows) + io_opt = bs * (1 + le16_to_cpu(id->nows)); } - integrity.tuple_size = ms; - blk_integrity_register(disk, &integrity); - blk_queue_max_integrity_segments(disk->queue, max_integrity_segments); + + /* + * Linux filesystems assume writing a single physical block is + * an atomic operation. Hence limit the physical block size to the + * value of the Atomic Write Unit Power Fail parameter. + */ + lim->logical_block_size = bs; + lim->physical_block_size = min(phys_bs, atomic_bs); + lim->io_min = phys_bs; + lim->io_opt = io_opt; + if ((ns->ctrl->quirks & NVME_QUIRK_DEALLOCATE_ZEROES) && + (ns->ctrl->oncs & NVME_CTRL_ONCS_DSM)) + lim->max_write_zeroes_sectors = UINT_MAX; + else + lim->max_write_zeroes_sectors = ns->ctrl->max_zeroes_sectors; + return valid; } -#else -static void nvme_init_integrity(struct gendisk *disk, u16 ms, u8 pi_type, - u32 max_integrity_segments) + +static bool nvme_ns_is_readonly(struct nvme_ns *ns, struct nvme_ns_info *info) { + return info->is_readonly || test_bit(NVME_NS_FORCE_RO, &ns->flags); } -#endif /* CONFIG_BLK_DEV_INTEGRITY */ -static void nvme_config_discard(struct gendisk *disk, struct nvme_ns *ns) +static inline bool nvme_first_scan(struct gendisk *disk) +{ + /* nvme_alloc_ns() scans the disk prior to adding it */ + return !disk_live(disk); +} + +static void nvme_set_chunk_sectors(struct nvme_ns *ns, struct nvme_id_ns *id, + struct queue_limits *lim) { struct nvme_ctrl *ctrl = ns->ctrl; - struct request_queue *queue = disk->queue; - u32 size = queue_logical_block_size(queue); + u32 iob; + + if ((ctrl->quirks & NVME_QUIRK_STRIPE_SIZE) && + is_power_of_2(ctrl->max_hw_sectors)) + iob = ctrl->max_hw_sectors; + else + iob = nvme_lba_to_sect(ns->head, le16_to_cpu(id->noiob)); - if (!(ctrl->oncs & NVME_CTRL_ONCS_DSM)) { - blk_queue_flag_clear(QUEUE_FLAG_DISCARD, queue); + if (!iob) + return; + + if (!is_power_of_2(iob)) { + if (nvme_first_scan(ns->disk)) + pr_warn("%s: ignoring unaligned IO boundary:%u\n", + ns->disk->disk_name, iob); return; } - if (ctrl->nr_streams && ns->sws && ns->sgs) - size *= ns->sws * ns->sgs; + if (blk_queue_is_zoned(ns->disk->queue)) { + if (nvme_first_scan(ns->disk)) + pr_warn("%s: ignoring zoned namespace IO boundary\n", + ns->disk->disk_name); + return; + } - BUILD_BUG_ON(PAGE_SIZE / sizeof(struct nvme_dsm_range) < - NVME_DSM_MAX_RANGES); + lim->chunk_sectors = iob; +} - queue->limits.discard_alignment = 0; - queue->limits.discard_granularity = size; +static int nvme_update_ns_info_generic(struct nvme_ns *ns, + struct nvme_ns_info *info) +{ + struct queue_limits lim; + unsigned int memflags; + int ret; - /* If discard is already enabled, don't reset queue limits */ - if (blk_queue_flag_test_and_set(QUEUE_FLAG_DISCARD, queue)) - return; + lim = queue_limits_start_update(ns->disk->queue); + nvme_set_ctrl_limits(ns->ctrl, &lim, false); - blk_queue_max_discard_sectors(queue, UINT_MAX); - blk_queue_max_discard_segments(queue, NVME_DSM_MAX_RANGES); + memflags = blk_mq_freeze_queue(ns->disk->queue); + ret = queue_limits_commit_update(ns->disk->queue, &lim); + set_disk_ro(ns->disk, nvme_ns_is_readonly(ns, info)); + blk_mq_unfreeze_queue(ns->disk->queue, memflags); - if (ctrl->quirks & NVME_QUIRK_DEALLOCATE_ZEROES) - blk_queue_max_write_zeroes_sectors(queue, UINT_MAX); + /* Hide the block-interface for these devices */ + if (!ret) + ret = -ENODEV; + return ret; } -static void nvme_config_write_zeroes(struct gendisk *disk, struct nvme_ns *ns) +static int nvme_query_fdp_granularity(struct nvme_ctrl *ctrl, + struct nvme_ns_info *info, u8 fdp_idx) { - u64 max_blocks; + struct nvme_fdp_config_log hdr, *h; + struct nvme_fdp_config_desc *desc; + size_t size = sizeof(hdr); + void *log, *end; + int i, n, ret; - if (!(ns->ctrl->oncs & NVME_CTRL_ONCS_WRITE_ZEROES) || - (ns->ctrl->quirks & NVME_QUIRK_DISABLE_WRITE_ZEROES)) - return; - /* - * Even though NVMe spec explicitly states that MDTS is not - * applicable to the write-zeroes:- "The restriction does not apply to - * commands that do not transfer data between the host and the - * controller (e.g., Write Uncorrectable ro Write Zeroes command).". - * In order to be more cautious use controller's max_hw_sectors value - * to configure the maximum sectors for the write-zeroes which is - * configured based on the controller's MDTS field in the - * nvme_init_identify() if available. - */ - if (ns->ctrl->max_hw_sectors == UINT_MAX) - max_blocks = (u64)USHRT_MAX + 1; - else - max_blocks = ns->ctrl->max_hw_sectors + 1; + ret = nvme_get_log_lsi(ctrl, 0, NVME_LOG_FDP_CONFIGS, 0, + NVME_CSI_NVM, &hdr, size, 0, info->endgid); + if (ret) { + dev_warn(ctrl->device, + "FDP configs log header status:0x%x endgid:%d\n", ret, + info->endgid); + return ret; + } - blk_queue_max_write_zeroes_sectors(disk->queue, - nvme_lba_to_sect(ns, max_blocks)); -} + size = le32_to_cpu(hdr.sze); + if (size > PAGE_SIZE * MAX_ORDER_NR_PAGES) { + dev_warn(ctrl->device, "FDP config size too large:%zu\n", + size); + return 0; + } -static int nvme_report_ns_ids(struct nvme_ctrl *ctrl, unsigned int nsid, - struct nvme_id_ns *id, struct nvme_ns_ids *ids) -{ - memset(ids, 0, sizeof(*ids)); + h = kvmalloc(size, GFP_KERNEL); + if (!h) + return -ENOMEM; - if (ctrl->vs >= NVME_VS(1, 1, 0)) - memcpy(ids->eui64, id->eui64, sizeof(id->eui64)); - if (ctrl->vs >= NVME_VS(1, 2, 0)) - memcpy(ids->nguid, id->nguid, sizeof(id->nguid)); - if (ctrl->vs >= NVME_VS(1, 3, 0)) - return nvme_identify_ns_descs(ctrl, nsid, ids); - return 0; -} + ret = nvme_get_log_lsi(ctrl, 0, NVME_LOG_FDP_CONFIGS, 0, + NVME_CSI_NVM, h, size, 0, info->endgid); + if (ret) { + dev_warn(ctrl->device, + "FDP configs log status:0x%x endgid:%d\n", ret, + info->endgid); + goto out; + } -static bool nvme_ns_ids_valid(struct nvme_ns_ids *ids) -{ - return !uuid_is_null(&ids->uuid) || - memchr_inv(ids->nguid, 0, sizeof(ids->nguid)) || - memchr_inv(ids->eui64, 0, sizeof(ids->eui64)); -} + n = le16_to_cpu(h->numfdpc) + 1; + if (fdp_idx > n) { + dev_warn(ctrl->device, "FDP index:%d out of range:%d\n", + fdp_idx, n); + /* Proceed without registering FDP streams */ + ret = 0; + goto out; + } -static bool nvme_ns_ids_equal(struct nvme_ns_ids *a, struct nvme_ns_ids *b) -{ - return uuid_equal(&a->uuid, &b->uuid) && - memcmp(&a->nguid, &b->nguid, sizeof(a->nguid)) == 0 && - memcmp(&a->eui64, &b->eui64, sizeof(a->eui64)) == 0; + log = h + 1; + desc = log; + end = log + size - sizeof(*h); + for (i = 0; i < fdp_idx; i++) { + log += le16_to_cpu(desc->dsze); + desc = log; + if (log >= end) { + dev_warn(ctrl->device, + "FDP invalid config descriptor list\n"); + ret = 0; + goto out; + } + } + + if (le32_to_cpu(desc->nrg) > 1) { + dev_warn(ctrl->device, "FDP NRG > 1 not supported\n"); + ret = 0; + goto out; + } + + info->runs = le64_to_cpu(desc->runs); +out: + kvfree(h); + return ret; } -static int nvme_setup_streams_ns(struct nvme_ctrl *ctrl, struct nvme_ns *ns, - u32 *phys_bs, u32 *io_opt) +static int nvme_query_fdp_info(struct nvme_ns *ns, struct nvme_ns_info *info) { - struct streams_directive_params s; - int ret; + struct nvme_ns_head *head = ns->head; + struct nvme_ctrl *ctrl = ns->ctrl; + struct nvme_fdp_ruh_status *ruhs; + struct nvme_fdp_config fdp; + struct nvme_command c = {}; + size_t size; + int i, ret; - if (!ctrl->nr_streams) + /* + * The FDP configuration is static for the lifetime of the namespace, + * so return immediately if we've already registered this namespace's + * streams. + */ + if (head->nr_plids) return 0; - ret = nvme_get_stream_params(ctrl, &s, ns->head->ns_id); - if (ret) + ret = nvme_get_features(ctrl, NVME_FEAT_FDP, info->endgid, NULL, 0, + &fdp); + if (ret) { + dev_warn(ctrl->device, "FDP get feature status:0x%x\n", ret); return ret; + } - ns->sws = le32_to_cpu(s.sws); - ns->sgs = le16_to_cpu(s.sgs); + if (!(fdp.flags & FDPCFG_FDPE)) + return 0; + + ret = nvme_query_fdp_granularity(ctrl, info, fdp.fdpcidx); + if (!info->runs) + return ret; - if (ns->sws) { - *phys_bs = ns->sws * (1 << ns->lba_shift); - if (ns->sgs) - *io_opt = *phys_bs * ns->sgs; + size = struct_size(ruhs, ruhsd, S8_MAX - 1); + ruhs = kzalloc(size, GFP_KERNEL); + if (!ruhs) + return -ENOMEM; + + c.imr.opcode = nvme_cmd_io_mgmt_recv; + c.imr.nsid = cpu_to_le32(head->ns_id); + c.imr.mo = NVME_IO_MGMT_RECV_MO_RUHS; + c.imr.numd = cpu_to_le32(nvme_bytes_to_numd(size)); + ret = nvme_submit_sync_cmd(ns->queue, &c, ruhs, size); + if (ret) { + dev_warn(ctrl->device, "FDP io-mgmt status:0x%x\n", ret); + goto free; } - return 0; + head->nr_plids = le16_to_cpu(ruhs->nruhsd); + if (!head->nr_plids) + goto free; + + head->plids = kcalloc(head->nr_plids, sizeof(*head->plids), + GFP_KERNEL); + if (!head->plids) { + dev_warn(ctrl->device, + "failed to allocate %u FDP placement IDs\n", + head->nr_plids); + head->nr_plids = 0; + ret = -ENOMEM; + goto free; + } + + for (i = 0; i < head->nr_plids; i++) + head->plids[i] = le16_to_cpu(ruhs->ruhsd[i].pid); +free: + kfree(ruhs); + return ret; } -static void nvme_update_disk_info(struct gendisk *disk, - struct nvme_ns *ns, struct nvme_id_ns *id) +static int nvme_update_ns_info_block(struct nvme_ns *ns, + struct nvme_ns_info *info) { - sector_t capacity = nvme_lba_to_sect(ns, le64_to_cpu(id->nsze)); - unsigned short bs = 1 << ns->lba_shift; - u32 atomic_bs, phys_bs, io_opt = 0; + struct queue_limits lim; + struct nvme_id_ns_nvm *nvm = NULL; + struct nvme_zone_info zi = {}; + struct nvme_id_ns *id; + unsigned int memflags; + sector_t capacity; + unsigned lbaf; + int ret; - if (ns->lba_shift > PAGE_SHIFT) { - /* unsupported block size, set capacity to 0 later */ - bs = (1 << 9); + ret = nvme_identify_ns(ns->ctrl, info->nsid, &id); + if (ret) + return ret; + + if (id->ncap == 0) { + /* namespace not allocated or attached */ + info->is_removed = true; + ret = -ENXIO; + goto out; } - blk_mq_freeze_queue(disk->queue); - blk_integrity_unregister(disk); + lbaf = nvme_lbaf_index(id->flbas); - atomic_bs = phys_bs = bs; - nvme_setup_streams_ns(ns->ctrl, ns, &phys_bs, &io_opt); - if (id->nabo == 0) { - /* - * Bit 1 indicates whether NAWUPF is defined for this namespace - * and whether it should be used instead of AWUPF. If NAWUPF == - * 0 then AWUPF must be used instead. - */ - if (id->nsfeat & NVME_NS_FEAT_ATOMICS && id->nawupf) - atomic_bs = (1 + le16_to_cpu(id->nawupf)) * bs; - else - atomic_bs = (1 + ns->ctrl->subsys->awupf) * bs; + if (ns->ctrl->ctratt & NVME_CTRL_ATTR_ELBAS) { + ret = nvme_identify_ns_nvm(ns->ctrl, info->nsid, &nvm); + if (ret < 0) + goto out; } - if (id->nsfeat & NVME_NS_FEAT_IO_OPT) { - /* NPWG = Namespace Preferred Write Granularity */ - phys_bs = bs * (1 + le16_to_cpu(id->npwg)); - /* NOWS = Namespace Optimal Write Size */ - io_opt = bs * (1 + le16_to_cpu(id->nows)); + if (IS_ENABLED(CONFIG_BLK_DEV_ZONED) && + ns->head->ids.csi == NVME_CSI_ZNS) { + ret = nvme_query_zone_info(ns, lbaf, &zi); + if (ret < 0) + goto out; } - blk_queue_logical_block_size(disk->queue, bs); - /* - * Linux filesystems assume writing a single physical block is - * an atomic operation. Hence limit the physical block size to the - * value of the Atomic Write Unit Power Fail parameter. - */ - blk_queue_physical_block_size(disk->queue, min(phys_bs, atomic_bs)); - blk_queue_io_min(disk->queue, phys_bs); - blk_queue_io_opt(disk->queue, io_opt); + if (ns->ctrl->ctratt & NVME_CTRL_ATTR_FDPS) { + ret = nvme_query_fdp_info(ns, info); + if (ret < 0) + goto out; + } - /* - * The block layer can't support LBA sizes larger than the page size - * yet, so catch this early and don't allow block I/O. - */ - if (ns->lba_shift > PAGE_SHIFT) + lim = queue_limits_start_update(ns->disk->queue); + + memflags = blk_mq_freeze_queue(ns->disk->queue); + ns->head->lba_shift = id->lbaf[lbaf].ds; + ns->head->nuse = le64_to_cpu(id->nuse); + capacity = nvme_lba_to_sect(ns->head, le64_to_cpu(id->nsze)); + nvme_set_ctrl_limits(ns->ctrl, &lim, false); + nvme_configure_metadata(ns->ctrl, ns->head, id, nvm, info); + nvme_set_chunk_sectors(ns, id, &lim); + if (!nvme_update_disk_info(ns, id, &lim)) capacity = 0; + nvme_config_discard(ns, &lim); + if (IS_ENABLED(CONFIG_BLK_DEV_ZONED) && + ns->head->ids.csi == NVME_CSI_ZNS) + nvme_update_zone_info(ns, &lim, &zi); + + if ((ns->ctrl->vwc & NVME_CTRL_VWC_PRESENT) && !info->no_vwc) + lim.features |= BLK_FEAT_WRITE_CACHE | BLK_FEAT_FUA; + else + lim.features &= ~(BLK_FEAT_WRITE_CACHE | BLK_FEAT_FUA); + + if (info->is_rotational) + lim.features |= BLK_FEAT_ROTATIONAL; + /* * Register a metadata profile for PI, or the plain non-integrity NVMe * metadata masquerading as Type 0 if supported, otherwise reject block * I/O to namespaces with metadata except when the namespace supports * PI, as it can strip/insert in that case. */ - if (ns->ms) { - if (IS_ENABLED(CONFIG_BLK_DEV_INTEGRITY) && - (ns->features & NVME_NS_METADATA_SUPPORTED)) - nvme_init_integrity(disk, ns->ms, ns->pi_type, - ns->ctrl->max_integrity_segments); - else if (!nvme_ns_has_pi(ns)) - capacity = 0; - } - - set_capacity_revalidate_and_notify(disk, capacity, false); - - nvme_config_discard(disk, ns); - nvme_config_write_zeroes(disk, ns); + if (!nvme_init_integrity(ns->head, &lim, info)) + capacity = 0; - if (id->nsattr & NVME_NS_ATTR_RO) - set_disk_ro(disk, true); + lim.max_write_streams = ns->head->nr_plids; + if (lim.max_write_streams) + lim.write_stream_granularity = min(info->runs, U32_MAX); else - set_disk_ro(disk, false); - - blk_mq_unfreeze_queue(disk->queue); -} - -static int __nvme_revalidate_disk(struct gendisk *disk, struct nvme_id_ns *id) -{ - struct nvme_ns *ns = disk->private_data; - struct nvme_ctrl *ctrl = ns->ctrl; - u32 iob; + lim.write_stream_granularity = 0; /* - * If identify namespace failed, use default 512 byte block size so - * block layer can use before failing read/write for 0 capacity. + * Only set the DEAC bit if the device guarantees that reads from + * deallocated data return zeroes. While the DEAC bit does not + * require that, it must be a no-op if reads from deallocated data + * do not return zeroes. */ - ns->lba_shift = id->lbaf[id->flbas & NVME_NS_FLBAS_LBA_MASK].ds; - if (ns->lba_shift == 0) - ns->lba_shift = 9; - - if ((ctrl->quirks & NVME_QUIRK_STRIPE_SIZE) && - is_power_of_2(ctrl->max_hw_sectors)) - iob = ctrl->max_hw_sectors; - else - iob = nvme_lba_to_sect(ns, le16_to_cpu(id->noiob)); - - ns->features = 0; - ns->ms = le16_to_cpu(id->lbaf[id->flbas & NVME_NS_FLBAS_LBA_MASK].ms); - /* the PI implementation requires metadata equal t10 pi tuple size */ - if (ns->ms == sizeof(struct t10_pi_tuple)) - ns->pi_type = id->dps & NVME_NS_DPS_PI_MASK; - else - ns->pi_type = 0; - - if (ns->ms) { - /* - * For PCIe only the separate metadata pointer is supported, - * as the block layer supplies metadata in a separate bio_vec - * chain. For Fabrics, only metadata as part of extended data - * LBA is supported on the wire per the Fabrics specification, - * but the HBA/HCA will do the remapping from the separate - * metadata buffers for us. - */ - if (id->flbas & NVME_NS_FLBAS_META_EXT) { - ns->features |= NVME_NS_EXT_LBAS; - if ((ctrl->ops->flags & NVME_F_FABRICS) && - (ctrl->ops->flags & NVME_F_METADATA_SUPPORTED) && - ctrl->max_integrity_segments) - ns->features |= NVME_NS_METADATA_SUPPORTED; - } else { - if (WARN_ON_ONCE(ctrl->ops->flags & NVME_F_FABRICS)) - return -EINVAL; - if (ctrl->ops->flags & NVME_F_METADATA_SUPPORTED) - ns->features |= NVME_NS_METADATA_SUPPORTED; - } + if ((id->dlfeat & 0x7) == 0x1 && (id->dlfeat & (1 << 3))) { + ns->head->features |= NVME_NS_DEAC; + lim.max_hw_wzeroes_unmap_sectors = lim.max_write_zeroes_sectors; } - if (iob) - blk_queue_chunk_sectors(ns->queue, rounddown_pow_of_two(iob)); - nvme_update_disk_info(disk, ns, id); -#ifdef CONFIG_NVME_MULTIPATH - if (ns->head->disk) { - nvme_update_disk_info(ns->head->disk, ns, id); - blk_queue_stack_limits(ns->head->disk->queue, ns->queue); - revalidate_disk(ns->head->disk); + ret = queue_limits_commit_update(ns->disk->queue, &lim); + if (ret) { + blk_mq_unfreeze_queue(ns->disk->queue, memflags); + goto out; } -#endif - return 0; -} -static int nvme_revalidate_disk(struct gendisk *disk) -{ - struct nvme_ns *ns = disk->private_data; - struct nvme_ctrl *ctrl = ns->ctrl; - struct nvme_id_ns *id; - struct nvme_ns_ids ids; - int ret = 0; + set_capacity_and_notify(ns->disk, capacity); + set_disk_ro(ns->disk, nvme_ns_is_readonly(ns, info)); + set_bit(NVME_NS_READY, &ns->flags); + blk_mq_unfreeze_queue(ns->disk->queue, memflags); - if (test_bit(NVME_NS_DEAD, &ns->flags)) { - set_capacity(disk, 0); - return -ENODEV; + if (blk_queue_is_zoned(ns->queue)) { + ret = blk_revalidate_disk_zones(ns->disk); + if (ret && !nvme_first_scan(ns->disk)) + goto out; } - ret = nvme_identify_ns(ctrl, ns->head->ns_id, &id); - if (ret) - goto out; - - if (id->ncap == 0) { - ret = -ENODEV; - goto free_id; - } + ret = 0; +out: + kfree(nvm); + kfree(id); + return ret; +} - ret = nvme_report_ns_ids(ctrl, ns->head->ns_id, id, &ids); - if (ret) - goto free_id; +static int nvme_update_ns_info(struct nvme_ns *ns, struct nvme_ns_info *info) +{ + bool unsupported = false; + int ret; - if (!nvme_ns_ids_equal(&ns->head->ids, &ids)) { - dev_err(ctrl->device, - "identifiers changed for nsid %d\n", ns->head->ns_id); - ret = -ENODEV; - goto free_id; + switch (info->ids.csi) { + case NVME_CSI_ZNS: + if (!IS_ENABLED(CONFIG_BLK_DEV_ZONED)) { + dev_info(ns->ctrl->device, + "block device for nsid %u not supported without CONFIG_BLK_DEV_ZONED\n", + info->nsid); + ret = nvme_update_ns_info_generic(ns, info); + break; + } + ret = nvme_update_ns_info_block(ns, info); + break; + case NVME_CSI_NVM: + ret = nvme_update_ns_info_block(ns, info); + break; + default: + dev_info(ns->ctrl->device, + "block device for nsid %u not supported (csi %u)\n", + info->nsid, info->ids.csi); + ret = nvme_update_ns_info_generic(ns, info); + break; } - ret = __nvme_revalidate_disk(disk, id); -free_id: - kfree(id); -out: /* - * Only fail the function if we got a fatal error back from the - * device, otherwise ignore the error and just move on. + * If probing fails due an unsupported feature, hide the block device, + * but still allow other access. */ - if (ret == -ENOMEM || (ret > 0 && !(ret & NVME_SC_DNR))) + if (ret == -ENODEV) { + ns->disk->flags |= GENHD_FL_HIDDEN; + set_bit(NVME_NS_READY, &ns->flags); + unsupported = true; ret = 0; - else if (ret > 0) - ret = blk_status_to_errno(nvme_error_status(ret)); - return ret; -} - -static char nvme_pr_type(enum pr_type type) -{ - switch (type) { - case PR_WRITE_EXCLUSIVE: - return 1; - case PR_EXCLUSIVE_ACCESS: - return 2; - case PR_WRITE_EXCLUSIVE_REG_ONLY: - return 3; - case PR_EXCLUSIVE_ACCESS_REG_ONLY: - return 4; - case PR_WRITE_EXCLUSIVE_ALL_REGS: - return 5; - case PR_EXCLUSIVE_ACCESS_ALL_REGS: - return 6; - default: - return 0; } -}; -static int nvme_pr_command(struct block_device *bdev, u32 cdw10, - u64 key, u64 sa_key, u8 op) -{ - struct nvme_ns_head *head = NULL; - struct nvme_ns *ns; - struct nvme_command c; - int srcu_idx, ret; - u8 data[16] = { 0, }; + if (!ret && nvme_ns_head_multipath(ns->head)) { + struct queue_limits *ns_lim = &ns->disk->queue->limits; + struct queue_limits lim; + unsigned int memflags; - ns = nvme_get_ns_from_disk(bdev->bd_disk, &head, &srcu_idx); - if (unlikely(!ns)) - return -EWOULDBLOCK; + lim = queue_limits_start_update(ns->head->disk->queue); + memflags = blk_mq_freeze_queue(ns->head->disk->queue); + /* + * queue_limits mixes values that are the hardware limitations + * for bio splitting with what is the device configuration. + * + * For NVMe the device configuration can change after e.g. a + * Format command, and we really want to pick up the new format + * value here. But we must still stack the queue limits to the + * least common denominator for multipathing to split the bios + * properly. + * + * To work around this, we explicitly set the device + * configuration to those that we just queried, but only stack + * the splitting limits in to make sure we still obey possibly + * lower limitations of other controllers. + */ + lim.logical_block_size = ns_lim->logical_block_size; + lim.physical_block_size = ns_lim->physical_block_size; + lim.io_min = ns_lim->io_min; + lim.io_opt = ns_lim->io_opt; + queue_limits_stack_bdev(&lim, ns->disk->part0, 0, + ns->head->disk->disk_name); + if (unsupported) + ns->head->disk->flags |= GENHD_FL_HIDDEN; + else + nvme_init_integrity(ns->head, &lim, info); + lim.max_write_streams = ns_lim->max_write_streams; + lim.write_stream_granularity = ns_lim->write_stream_granularity; + ret = queue_limits_commit_update(ns->head->disk->queue, &lim); - put_unaligned_le64(key, &data[0]); - put_unaligned_le64(sa_key, &data[8]); + set_capacity_and_notify(ns->head->disk, get_capacity(ns->disk)); + set_disk_ro(ns->head->disk, nvme_ns_is_readonly(ns, info)); + nvme_mpath_revalidate_paths(ns); - memset(&c, 0, sizeof(c)); - c.common.opcode = op; - c.common.nsid = cpu_to_le32(ns->head->ns_id); - c.common.cdw10 = cpu_to_le32(cdw10); + blk_mq_unfreeze_queue(ns->head->disk->queue, memflags); + } - ret = nvme_submit_sync_cmd(ns->queue, &c, data, 16); - nvme_put_ns_from_disk(head, srcu_idx); return ret; } -static int nvme_pr_register(struct block_device *bdev, u64 old, - u64 new, unsigned flags) +int nvme_ns_get_unique_id(struct nvme_ns *ns, u8 id[16], + enum blk_unique_id type) { - u32 cdw10; + struct nvme_ns_ids *ids = &ns->head->ids; - if (flags & ~PR_FL_IGNORE_KEY) - return -EOPNOTSUPP; - - cdw10 = old ? 2 : 0; - cdw10 |= (flags & PR_FL_IGNORE_KEY) ? 1 << 3 : 0; - cdw10 |= (1 << 30) | (1 << 31); /* PTPL=1 */ - return nvme_pr_command(bdev, cdw10, old, new, nvme_cmd_resv_register); -} - -static int nvme_pr_reserve(struct block_device *bdev, u64 key, - enum pr_type type, unsigned flags) -{ - u32 cdw10; - - if (flags & ~PR_FL_IGNORE_KEY) - return -EOPNOTSUPP; - - cdw10 = nvme_pr_type(type) << 8; - cdw10 |= ((flags & PR_FL_IGNORE_KEY) ? 1 << 3 : 0); - return nvme_pr_command(bdev, cdw10, key, 0, nvme_cmd_resv_acquire); -} + if (type != BLK_UID_EUI64) + return -EINVAL; -static int nvme_pr_preempt(struct block_device *bdev, u64 old, u64 new, - enum pr_type type, bool abort) -{ - u32 cdw10 = nvme_pr_type(type) << 8 | (abort ? 2 : 1); - return nvme_pr_command(bdev, cdw10, old, new, nvme_cmd_resv_acquire); -} + if (memchr_inv(ids->nguid, 0, sizeof(ids->nguid))) { + memcpy(id, &ids->nguid, sizeof(ids->nguid)); + return sizeof(ids->nguid); + } + if (memchr_inv(ids->eui64, 0, sizeof(ids->eui64))) { + memcpy(id, &ids->eui64, sizeof(ids->eui64)); + return sizeof(ids->eui64); + } -static int nvme_pr_clear(struct block_device *bdev, u64 key) -{ - u32 cdw10 = 1 | (key ? 1 << 3 : 0); - return nvme_pr_command(bdev, cdw10, key, 0, nvme_cmd_resv_register); + return -EINVAL; } -static int nvme_pr_release(struct block_device *bdev, u64 key, enum pr_type type) +static int nvme_get_unique_id(struct gendisk *disk, u8 id[16], + enum blk_unique_id type) { - u32 cdw10 = nvme_pr_type(type) << 8 | (key ? 1 << 3 : 0); - return nvme_pr_command(bdev, cdw10, key, 0, nvme_cmd_resv_release); + return nvme_ns_get_unique_id(disk->private_data, id, type); } -static const struct pr_ops nvme_pr_ops = { - .pr_register = nvme_pr_register, - .pr_reserve = nvme_pr_reserve, - .pr_release = nvme_pr_release, - .pr_preempt = nvme_pr_preempt, - .pr_clear = nvme_pr_clear, -}; - #ifdef CONFIG_BLK_SED_OPAL -int nvme_sec_submit(void *data, u16 spsp, u8 secp, void *buffer, size_t len, +static int nvme_sec_submit(void *data, u16 spsp, u8 secp, void *buffer, size_t len, bool send) { struct nvme_ctrl *ctrl = data; - struct nvme_command cmd; + struct nvme_command cmd = { }; - memset(&cmd, 0, sizeof(cmd)); if (send) cmd.common.opcode = nvme_admin_security_send; else @@ -2145,68 +2576,69 @@ int nvme_sec_submit(void *data, u16 spsp, u8 secp, void *buffer, size_t len, cmd.common.cdw11 = cpu_to_le32(len); return __nvme_submit_sync_cmd(ctrl->admin_q, &cmd, NULL, buffer, len, - ADMIN_TIMEOUT, NVME_QID_ANY, 1, 0, false); + NVME_QID_ANY, NVME_SUBMIT_AT_HEAD); } -EXPORT_SYMBOL_GPL(nvme_sec_submit); -#endif /* CONFIG_BLK_SED_OPAL */ - -static const struct block_device_operations nvme_fops = { - .owner = THIS_MODULE, - .ioctl = nvme_ioctl, - .compat_ioctl = nvme_compat_ioctl, - .open = nvme_open, - .release = nvme_release, - .getgeo = nvme_getgeo, - .revalidate_disk= nvme_revalidate_disk, - .pr_ops = &nvme_pr_ops, -}; -#ifdef CONFIG_NVME_MULTIPATH -static int nvme_ns_head_open(struct block_device *bdev, fmode_t mode) +static void nvme_configure_opal(struct nvme_ctrl *ctrl, bool was_suspended) +{ + if (ctrl->oacs & NVME_CTRL_OACS_SEC_SUPP) { + if (!ctrl->opal_dev) + ctrl->opal_dev = init_opal_dev(ctrl, &nvme_sec_submit); + else if (was_suspended) + opal_unlock_from_suspend(ctrl->opal_dev); + } else { + free_opal_dev(ctrl->opal_dev); + ctrl->opal_dev = NULL; + } +} +#else +static void nvme_configure_opal(struct nvme_ctrl *ctrl, bool was_suspended) { - struct nvme_ns_head *head = bdev->bd_disk->private_data; - - if (!kref_get_unless_zero(&head->ref)) - return -ENXIO; - return 0; } +#endif /* CONFIG_BLK_SED_OPAL */ -static void nvme_ns_head_release(struct gendisk *disk, fmode_t mode) +#ifdef CONFIG_BLK_DEV_ZONED +static int nvme_report_zones(struct gendisk *disk, sector_t sector, + unsigned int nr_zones, struct blk_report_zones_args *args) { - nvme_put_ns_head(disk->private_data); + return nvme_ns_report_zones(disk->private_data, sector, nr_zones, args); } +#else +#define nvme_report_zones NULL +#endif /* CONFIG_BLK_DEV_ZONED */ -const struct block_device_operations nvme_ns_head_ops = { +const struct block_device_operations nvme_bdev_ops = { .owner = THIS_MODULE, - .open = nvme_ns_head_open, - .release = nvme_ns_head_release, .ioctl = nvme_ioctl, - .compat_ioctl = nvme_compat_ioctl, + .compat_ioctl = blkdev_compat_ptr_ioctl, + .open = nvme_open, + .release = nvme_release, .getgeo = nvme_getgeo, + .get_unique_id = nvme_get_unique_id, + .report_zones = nvme_report_zones, .pr_ops = &nvme_pr_ops, }; -#endif /* CONFIG_NVME_MULTIPATH */ -static int nvme_wait_ready(struct nvme_ctrl *ctrl, u64 cap, bool enabled) +static int nvme_wait_ready(struct nvme_ctrl *ctrl, u32 mask, u32 val, + u32 timeout, const char *op) { - unsigned long timeout = - ((NVME_CAP_TIMEOUT(cap) + 1) * HZ / 2) + jiffies; - u32 csts, bit = enabled ? NVME_CSTS_RDY : 0; + unsigned long timeout_jiffies = jiffies + timeout * HZ; + u32 csts; int ret; while ((ret = ctrl->ops->reg_read32(ctrl, NVME_REG_CSTS, &csts)) == 0) { if (csts == ~0) return -ENODEV; - if ((csts & NVME_CSTS_RDY) == bit) + if ((csts & mask) == val) break; usleep_range(1000, 2000); if (fatal_signal_pending(current)) return -EINTR; - if (time_after(jiffies, timeout)) { + if (time_after(jiffies, timeout_jiffies)) { dev_err(ctrl->device, "Device not ready; aborting %s, CSTS=0x%x\n", - enabled ? "initialisation" : "reset", csts); + op, csts); return -ENODEV; } } @@ -2214,38 +2646,36 @@ static int nvme_wait_ready(struct nvme_ctrl *ctrl, u64 cap, bool enabled) return ret; } -/* - * If the device has been passed off to us in an enabled state, just clear - * the enabled bit. The spec says we should set the 'shutdown notification - * bits', but doing so may cause the device to complete commands to the - * admin queue ... and we don't know what memory that might be pointing at! - */ -int nvme_disable_ctrl(struct nvme_ctrl *ctrl) +int nvme_disable_ctrl(struct nvme_ctrl *ctrl, bool shutdown) { int ret; ctrl->ctrl_config &= ~NVME_CC_SHN_MASK; - ctrl->ctrl_config &= ~NVME_CC_ENABLE; + if (shutdown) + ctrl->ctrl_config |= NVME_CC_SHN_NORMAL; + else + ctrl->ctrl_config &= ~NVME_CC_ENABLE; ret = ctrl->ops->reg_write32(ctrl, NVME_REG_CC, ctrl->ctrl_config); if (ret) return ret; + if (shutdown) { + return nvme_wait_ready(ctrl, NVME_CSTS_SHST_MASK, + NVME_CSTS_SHST_CMPLT, + ctrl->shutdown_timeout, "shutdown"); + } if (ctrl->quirks & NVME_QUIRK_DELAY_BEFORE_CHK_RDY) msleep(NVME_QUIRK_DELAY_AMOUNT); - - return nvme_wait_ready(ctrl, ctrl->cap, false); + return nvme_wait_ready(ctrl, NVME_CSTS_RDY, 0, + (NVME_CAP_TIMEOUT(ctrl->cap) + 1) / 2, "reset"); } EXPORT_SYMBOL_GPL(nvme_disable_ctrl); int nvme_enable_ctrl(struct nvme_ctrl *ctrl) { - /* - * Default to a 4K page size, with the intention to update this - * path in the future to accomodate architectures with differing - * kernel and IO page sizes. - */ - unsigned dev_page_min, page_shift = 12; + unsigned dev_page_min; + u32 timeout; int ret; ret = ctrl->ops->reg_read64(ctrl, NVME_REG_CAP, &ctrl->cap); @@ -2255,78 +2685,71 @@ int nvme_enable_ctrl(struct nvme_ctrl *ctrl) } dev_page_min = NVME_CAP_MPSMIN(ctrl->cap) + 12; - if (page_shift < dev_page_min) { + if (NVME_CTRL_PAGE_SHIFT < dev_page_min) { dev_err(ctrl->device, "Minimum device page size %u too large for host (%u)\n", - 1 << dev_page_min, 1 << page_shift); + 1 << dev_page_min, 1 << NVME_CTRL_PAGE_SHIFT); return -ENODEV; } - ctrl->page_size = 1 << page_shift; + if (NVME_CAP_CSS(ctrl->cap) & NVME_CAP_CSS_CSI) + ctrl->ctrl_config = NVME_CC_CSS_CSI; + else + ctrl->ctrl_config = NVME_CC_CSS_NVM; - ctrl->ctrl_config = NVME_CC_CSS_NVM; - ctrl->ctrl_config |= (page_shift - 12) << NVME_CC_MPS_SHIFT; + /* + * Setting CRIME results in CSTS.RDY before the media is ready. This + * makes it possible for media related commands to return the error + * NVME_SC_ADMIN_COMMAND_MEDIA_NOT_READY. Until the driver is + * restructured to handle retries, disable CC.CRIME. + */ + ctrl->ctrl_config &= ~NVME_CC_CRIME; + + ctrl->ctrl_config |= (NVME_CTRL_PAGE_SHIFT - 12) << NVME_CC_MPS_SHIFT; ctrl->ctrl_config |= NVME_CC_AMS_RR | NVME_CC_SHN_NONE; ctrl->ctrl_config |= NVME_CC_IOSQES | NVME_CC_IOCQES; - ctrl->ctrl_config |= NVME_CC_ENABLE; - ret = ctrl->ops->reg_write32(ctrl, NVME_REG_CC, ctrl->ctrl_config); if (ret) return ret; - return nvme_wait_ready(ctrl, ctrl->cap, true); -} -EXPORT_SYMBOL_GPL(nvme_enable_ctrl); -int nvme_shutdown_ctrl(struct nvme_ctrl *ctrl) -{ - unsigned long timeout = jiffies + (ctrl->shutdown_timeout * HZ); - u32 csts; - int ret; - - ctrl->ctrl_config &= ~NVME_CC_SHN_MASK; - ctrl->ctrl_config |= NVME_CC_SHN_NORMAL; - - ret = ctrl->ops->reg_write32(ctrl, NVME_REG_CC, ctrl->ctrl_config); + /* CAP value may change after initial CC write */ + ret = ctrl->ops->reg_read64(ctrl, NVME_REG_CAP, &ctrl->cap); if (ret) return ret; - while ((ret = ctrl->ops->reg_read32(ctrl, NVME_REG_CSTS, &csts)) == 0) { - if ((csts & NVME_CSTS_SHST_MASK) == NVME_CSTS_SHST_CMPLT) - break; + timeout = NVME_CAP_TIMEOUT(ctrl->cap); + if (ctrl->cap & NVME_CAP_CRMS_CRWMS) { + u32 crto, ready_timeout; - msleep(100); - if (fatal_signal_pending(current)) - return -EINTR; - if (time_after(jiffies, timeout)) { - dev_err(ctrl->device, - "Device shutdown incomplete; abort shutdown\n"); - return -ENODEV; + ret = ctrl->ops->reg_read32(ctrl, NVME_REG_CRTO, &crto); + if (ret) { + dev_err(ctrl->device, "Reading CRTO failed (%d)\n", + ret); + return ret; } - } - - return ret; -} -EXPORT_SYMBOL_GPL(nvme_shutdown_ctrl); -static void nvme_set_queue_limits(struct nvme_ctrl *ctrl, - struct request_queue *q) -{ - bool vwc = false; - - if (ctrl->max_hw_sectors) { - u32 max_segments = - (ctrl->max_hw_sectors / (ctrl->page_size >> 9)) + 1; + /* + * CRTO should always be greater or equal to CAP.TO, but some + * devices are known to get this wrong. Use the larger of the + * two values. + */ + ready_timeout = NVME_CRTO_CRWMT(crto); - max_segments = min_not_zero(max_segments, ctrl->max_segments); - blk_queue_max_hw_sectors(q, ctrl->max_hw_sectors); - blk_queue_max_segments(q, min_t(u32, max_segments, USHRT_MAX)); + if (ready_timeout < timeout) + dev_warn_once(ctrl->device, "bad crto:%x cap:%llx\n", + crto, ctrl->cap); + else + timeout = ready_timeout; } - blk_queue_virt_boundary(q, ctrl->page_size - 1); - blk_queue_dma_alignment(q, 7); - if (ctrl->vwc & NVME_CTRL_VWC_PRESENT) - vwc = true; - blk_queue_write_cache(q, vwc, vwc); + + ctrl->ctrl_config |= NVME_CC_ENABLE; + ret = ctrl->ops->reg_write32(ctrl, NVME_REG_CC, ctrl->ctrl_config); + if (ret) + return ret; + return nvme_wait_ready(ctrl, NVME_CSTS_RDY, NVME_CSTS_RDY, + (timeout + 1) / 2, "initialisation"); } +EXPORT_SYMBOL_GPL(nvme_enable_ctrl); static int nvme_configure_timestamp(struct nvme_ctrl *ctrl) { @@ -2345,49 +2768,97 @@ static int nvme_configure_timestamp(struct nvme_ctrl *ctrl) return ret; } -static int nvme_configure_acre(struct nvme_ctrl *ctrl) +static int nvme_configure_host_options(struct nvme_ctrl *ctrl) { struct nvme_feat_host_behavior *host; + u8 acre = 0, lbafee = 0; int ret; /* Don't bother enabling the feature if retry delay is not reported */ - if (!ctrl->crdt[0]) + if (ctrl->crdt[0]) + acre = NVME_ENABLE_ACRE; + if (ctrl->ctratt & NVME_CTRL_ATTR_ELBAS) + lbafee = NVME_ENABLE_LBAFEE; + + if (!acre && !lbafee) return 0; host = kzalloc(sizeof(*host), GFP_KERNEL); if (!host) return 0; - host->acre = NVME_ENABLE_ACRE; + host->acre = acre; + host->lbafee = lbafee; ret = nvme_set_features(ctrl, NVME_FEAT_HOST_BEHAVIOR, 0, host, sizeof(*host), NULL); kfree(host); return ret; } -static int nvme_configure_apst(struct nvme_ctrl *ctrl) +/* + * The function checks whether the given total (exlat + enlat) latency of + * a power state allows the latter to be used as an APST transition target. + * It does so by comparing the latency to the primary and secondary latency + * tolerances defined by module params. If there's a match, the corresponding + * timeout value is returned and the matching tolerance index (1 or 2) is + * reported. + */ +static bool nvme_apst_get_transition_time(u64 total_latency, + u64 *transition_time, unsigned *last_index) { - /* - * APST (Autonomous Power State Transition) lets us program a - * table of power state transitions that the controller will - * perform automatically. We configure it with a simple - * heuristic: we are willing to spend at most 2% of the time - * transitioning between power states. Therefore, when running - * in any given state, we will enter the next lower-power - * non-operational state after waiting 50 * (enlat + exlat) - * microseconds, as long as that state's exit latency is under - * the requested maximum latency. - * - * We will not autonomously enter any non-operational state for - * which the total latency exceeds ps_max_latency_us. Users - * can set ps_max_latency_us to zero to turn off APST. - */ + if (total_latency <= apst_primary_latency_tol_us) { + if (*last_index == 1) + return false; + *last_index = 1; + *transition_time = apst_primary_timeout_ms; + return true; + } + if (apst_secondary_timeout_ms && + total_latency <= apst_secondary_latency_tol_us) { + if (*last_index <= 2) + return false; + *last_index = 2; + *transition_time = apst_secondary_timeout_ms; + return true; + } + return false; +} - unsigned apste; +/* + * APST (Autonomous Power State Transition) lets us program a table of power + * state transitions that the controller will perform automatically. + * + * Depending on module params, one of the two supported techniques will be used: + * + * - If the parameters provide explicit timeouts and tolerances, they will be + * used to build a table with up to 2 non-operational states to transition to. + * The default parameter values were selected based on the values used by + * Microsoft's and Intel's NVMe drivers. Yet, since we don't implement dynamic + * regeneration of the APST table in the event of switching between external + * and battery power, the timeouts and tolerances reflect a compromise + * between values used by Microsoft for AC and battery scenarios. + * - If not, we'll configure the table with a simple heuristic: we are willing + * to spend at most 2% of the time transitioning between power states. + * Therefore, when running in any given state, we will enter the next + * lower-power non-operational state after waiting 50 * (enlat + exlat) + * microseconds, as long as that state's exit latency is under the requested + * maximum latency. + * + * We will not autonomously enter any non-operational state for which the total + * latency exceeds ps_max_latency_us. + * + * Users can set ps_max_latency_us to zero to turn off APST. + */ +static int nvme_configure_apst(struct nvme_ctrl *ctrl) +{ struct nvme_feat_auto_pst *table; + unsigned apste = 0; u64 max_lat_us = 0; + __le64 target = 0; int max_ps = -1; + int state; int ret; + unsigned last_lt_index = UINT_MAX; /* * If APST isn't supported or if we haven't been initialized yet, @@ -2407,83 +2878,78 @@ static int nvme_configure_apst(struct nvme_ctrl *ctrl) if (!ctrl->apst_enabled || ctrl->ps_max_latency_us == 0) { /* Turn off APST. */ - apste = 0; dev_dbg(ctrl->device, "APST disabled\n"); - } else { - __le64 target = cpu_to_le64(0); - int state; + goto done; + } + + /* + * Walk through all states from lowest- to highest-power. + * According to the spec, lower-numbered states use more power. NPSS, + * despite the name, is the index of the lowest-power state, not the + * number of states. + */ + for (state = (int)ctrl->npss; state >= 0; state--) { + u64 total_latency_us, exit_latency_us, transition_ms; + + if (target) + table->entries[state] = target; /* - * Walk through all states from lowest- to highest-power. - * According to the spec, lower-numbered states use more - * power. NPSS, despite the name, is the index of the - * lowest-power state, not the number of states. + * Don't allow transitions to the deepest state if it's quirked + * off. */ - for (state = (int)ctrl->npss; state >= 0; state--) { - u64 total_latency_us, exit_latency_us, transition_ms; + if (state == ctrl->npss && + (ctrl->quirks & NVME_QUIRK_NO_DEEPEST_PS)) + continue; - if (target) - table->entries[state] = target; + /* + * Is this state a useful non-operational state for higher-power + * states to autonomously transition to? + */ + if (!(ctrl->psd[state].flags & NVME_PS_FLAGS_NON_OP_STATE)) + continue; - /* - * Don't allow transitions to the deepest state - * if it's quirked off. - */ - if (state == ctrl->npss && - (ctrl->quirks & NVME_QUIRK_NO_DEEPEST_PS)) - continue; + exit_latency_us = (u64)le32_to_cpu(ctrl->psd[state].exit_lat); + if (exit_latency_us > ctrl->ps_max_latency_us) + continue; - /* - * Is this state a useful non-operational state for - * higher-power states to autonomously transition to? - */ - if (!(ctrl->psd[state].flags & - NVME_PS_FLAGS_NON_OP_STATE)) - continue; + total_latency_us = exit_latency_us + + le32_to_cpu(ctrl->psd[state].entry_lat); - exit_latency_us = - (u64)le32_to_cpu(ctrl->psd[state].exit_lat); - if (exit_latency_us > ctrl->ps_max_latency_us) + /* + * This state is good. It can be used as the APST idle target + * for higher power states. + */ + if (apst_primary_timeout_ms && apst_primary_latency_tol_us) { + if (!nvme_apst_get_transition_time(total_latency_us, + &transition_ms, &last_lt_index)) continue; - - total_latency_us = - exit_latency_us + - le32_to_cpu(ctrl->psd[state].entry_lat); - - /* - * This state is good. Use it as the APST idle - * target for higher power states. - */ + } else { transition_ms = total_latency_us + 19; do_div(transition_ms, 20); if (transition_ms > (1 << 24) - 1) transition_ms = (1 << 24) - 1; - - target = cpu_to_le64((state << 3) | - (transition_ms << 8)); - - if (max_ps == -1) - max_ps = state; - - if (total_latency_us > max_lat_us) - max_lat_us = total_latency_us; } - apste = 1; - - if (max_ps == -1) { - dev_dbg(ctrl->device, "APST enabled but no non-operational states are available\n"); - } else { - dev_dbg(ctrl->device, "APST enabled: max PS = %d, max round-trip latency = %lluus, table = %*phN\n", - max_ps, max_lat_us, (int)sizeof(*table), table); - } + target = cpu_to_le64((state << 3) | (transition_ms << 8)); + if (max_ps == -1) + max_ps = state; + if (total_latency_us > max_lat_us) + max_lat_us = total_latency_us; } + if (max_ps == -1) + dev_dbg(ctrl->device, "APST enabled but no non-operational states are available\n"); + else + dev_dbg(ctrl->device, "APST enabled: max PS = %d, max round-trip latency = %lluus, table = %*phN\n", + max_ps, max_lat_us, (int)sizeof(*table), table); + apste = 1; + +done: ret = nvme_set_features(ctrl, NVME_FEAT_AUTO_PST, apste, table, sizeof(*table), NULL); if (ret) dev_err(ctrl->device, "failed to set APST feature (%d)\n", ret); - kfree(table); return ret; } @@ -2505,7 +2971,8 @@ static void nvme_set_latency_tolerance(struct device *dev, s32 val) if (ctrl->ps_max_latency_us != latency) { ctrl->ps_max_latency_us = latency; - nvme_configure_apst(ctrl); + if (nvme_ctrl_state(ctrl) == NVME_CTRL_LIVE) + nvme_configure_apst(ctrl); } } @@ -2540,6 +3007,34 @@ static const struct nvme_core_quirk_entry core_quirks[] = { .vid = 0x14a4, .fr = "22301111", .quirks = NVME_QUIRK_SIMPLE_SUSPEND, + }, + { + /* + * This Kioxia CD6-V Series / HPE PE8030 device times out and + * aborts I/O during any load, but more easily reproducible + * with discards (fstrim). + * + * The device is left in a state where it is also not possible + * to use "nvme set-feature" to disable APST, but booting with + * nvme_core.default_ps_max_latency=0 works. + */ + .vid = 0x1e0f, + .mn = "KCD6XVUL6T40", + .quirks = NVME_QUIRK_NO_APST, + }, + { + /* + * The external Samsung X5 SSD fails initialization without a + * delay before checking if it is ready and has a whole set of + * other problems. To make this even more interesting, it + * shares the PCI ID with internal Samsung 970 Evo Plus that + * does not need or want these quirks. + */ + .vid = 0x144d, + .mn = "Samsung Portable SSD X5", + .quirks = NVME_QUIRK_DELAY_BEFORE_CHK_RDY | + NVME_QUIRK_NO_DEEPEST_PS | + NVME_QUIRK_IGNORE_DEV_SUBNQN, } }; @@ -2581,7 +3076,7 @@ static void nvme_init_subnqn(struct nvme_subsystem *subsys, struct nvme_ctrl *ct if(!(ctrl->quirks & NVME_QUIRK_IGNORE_DEV_SUBNQN)) { nqnlen = strnlen(id->subnqn, NVMF_NQN_SIZE); if (nqnlen > 0 && nqnlen < NVMF_NQN_SIZE) { - strlcpy(subsys->subnqn, id->subnqn, NVMF_NQN_SIZE); + strscpy(subsys->subnqn, id->subnqn, NVMF_NQN_SIZE); return; } @@ -2589,7 +3084,11 @@ static void nvme_init_subnqn(struct nvme_subsystem *subsys, struct nvme_ctrl *ct dev_warn(ctrl->device, "missing or invalid SUBNQN field.\n"); } - /* Generate a "fake" NQN per Figure 254 in NVMe 1.3 + ECN 001 */ + /* + * Generate a "fake" NQN similar to the one in Section 4.5 of the NVMe + * Base Specification 2.0. It is slightly different from the format + * specified there due to historic reasons, and we can't change it now. + */ off = snprintf(subsys->subnqn, NVMF_NQN_SIZE, "nqn.2014.08.org.nvmexpress:%04x%04x", le16_to_cpu(id->vid), le16_to_cpu(id->ssvid)); @@ -2606,7 +3105,7 @@ static void nvme_release_subsystem(struct device *dev) container_of(dev, struct nvme_subsystem, dev); if (subsys->instance >= 0) - ida_simple_remove(&nvme_instance_ida, subsys->instance); + ida_free(&nvme_instance_ida, subsys->instance); kfree(subsys); } @@ -2657,55 +3156,20 @@ static struct nvme_subsystem *__nvme_find_get_subsystem(const char *subsysnqn) return NULL; } -#define SUBSYS_ATTR_RO(_name, _mode, _show) \ - struct device_attribute subsys_attr_##_name = \ - __ATTR(_name, _mode, _show, NULL) - -static ssize_t nvme_subsys_show_nqn(struct device *dev, - struct device_attribute *attr, - char *buf) +static inline bool nvme_discovery_ctrl(struct nvme_ctrl *ctrl) { - struct nvme_subsystem *subsys = - container_of(dev, struct nvme_subsystem, dev); - - return snprintf(buf, PAGE_SIZE, "%s\n", subsys->subnqn); -} -static SUBSYS_ATTR_RO(subsysnqn, S_IRUGO, nvme_subsys_show_nqn); - -#define nvme_subsys_show_str_function(field) \ -static ssize_t subsys_##field##_show(struct device *dev, \ - struct device_attribute *attr, char *buf) \ -{ \ - struct nvme_subsystem *subsys = \ - container_of(dev, struct nvme_subsystem, dev); \ - return sprintf(buf, "%.*s\n", \ - (int)sizeof(subsys->field), subsys->field); \ -} \ -static SUBSYS_ATTR_RO(field, S_IRUGO, subsys_##field##_show); - -nvme_subsys_show_str_function(model); -nvme_subsys_show_str_function(serial); -nvme_subsys_show_str_function(firmware_rev); - -static struct attribute *nvme_subsys_attrs[] = { - &subsys_attr_model.attr, - &subsys_attr_serial.attr, - &subsys_attr_firmware_rev.attr, - &subsys_attr_subsysnqn.attr, -#ifdef CONFIG_NVME_MULTIPATH - &subsys_attr_iopolicy.attr, -#endif - NULL, -}; + return ctrl->opts && ctrl->opts->discovery_nqn; +} -static struct attribute_group nvme_subsys_attrs_group = { - .attrs = nvme_subsys_attrs, -}; +static inline bool nvme_admin_ctrl(struct nvme_ctrl *ctrl) +{ + return ctrl->cntrltype == NVME_CTRL_ADMIN; +} -static const struct attribute_group *nvme_subsys_attrs_groups[] = { - &nvme_subsys_attrs_group, - NULL, -}; +static inline bool nvme_is_io_ctrl(struct nvme_ctrl *ctrl) +{ + return !nvme_discovery_ctrl(ctrl) && !nvme_admin_ctrl(ctrl); +} static bool nvme_validate_cntlid(struct nvme_subsystem *subsys, struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id) @@ -2720,13 +3184,14 @@ static bool nvme_validate_cntlid(struct nvme_subsystem *subsys, if (tmp->cntlid == ctrl->cntlid) { dev_err(ctrl->device, - "Duplicate cntlid %u with %s, rejecting\n", - ctrl->cntlid, dev_name(tmp->device)); + "Duplicate cntlid %u with %s, subsys %s, rejecting\n", + ctrl->cntlid, dev_name(tmp->device), + subsys->subnqn); return false; } if ((id->cmic & NVME_CTRL_CMIC_MULTI_CTRL) || - (ctrl->opts && ctrl->opts->discovery_nqn)) + nvme_discovery_ctrl(ctrl)) continue; dev_err(ctrl->device, @@ -2754,15 +3219,27 @@ static int nvme_init_subsystem(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id) nvme_init_subnqn(subsys, ctrl, id); memcpy(subsys->serial, id->sn, sizeof(subsys->serial)); memcpy(subsys->model, id->mn, sizeof(subsys->model)); - memcpy(subsys->firmware_rev, id->fr, sizeof(subsys->firmware_rev)); subsys->vendor_id = le16_to_cpu(id->vid); subsys->cmic = id->cmic; subsys->awupf = le16_to_cpu(id->awupf); -#ifdef CONFIG_NVME_MULTIPATH - subsys->iopolicy = NVME_IOPOLICY_NUMA; -#endif - subsys->dev.class = nvme_subsys_class; + /* Versions prior to 1.4 don't necessarily report a valid type */ + if (id->cntrltype == NVME_CTRL_DISC || + !strcmp(subsys->subnqn, NVME_DISC_SUBSYS_NAME)) + subsys->subtype = NVME_NQN_DISC; + else + subsys->subtype = NVME_NQN_NVME; + + if (nvme_discovery_ctrl(ctrl) && subsys->subtype != NVME_NQN_DISC) { + dev_err(ctrl->device, + "Subsystem %s is not a discovery controller", + subsys->subnqn); + kfree(subsys); + return -EINVAL; + } + nvme_mpath_default_iopolicy(subsys); + + subsys->dev.class = &nvme_subsys_class; subsys->dev.release = nvme_release_subsystem; subsys->dev.groups = nvme_subsys_attrs_groups; dev_set_name(&subsys->dev, "nvme-subsys%d", ctrl->instance); @@ -2812,8 +3289,8 @@ out_unlock: return ret; } -int nvme_get_log(struct nvme_ctrl *ctrl, u32 nsid, u8 log_page, u8 lsp, - void *log, size_t size, u64 offset) +static int nvme_get_log_lsi(struct nvme_ctrl *ctrl, u32 nsid, u8 log_page, + u8 lsp, u8 csi, void *log, size_t size, u64 offset, u16 lsi) { struct nvme_command c = { }; u32 dwlen = nvme_bytes_to_numd(size); @@ -2826,51 +3303,228 @@ int nvme_get_log(struct nvme_ctrl *ctrl, u32 nsid, u8 log_page, u8 lsp, c.get_log_page.numdu = cpu_to_le16(dwlen >> 16); c.get_log_page.lpol = cpu_to_le32(lower_32_bits(offset)); c.get_log_page.lpou = cpu_to_le32(upper_32_bits(offset)); + c.get_log_page.csi = csi; + c.get_log_page.lsi = cpu_to_le16(lsi); return nvme_submit_sync_cmd(ctrl->admin_q, &c, log, size); } -static int nvme_get_effects_log(struct nvme_ctrl *ctrl) +int nvme_get_log(struct nvme_ctrl *ctrl, u32 nsid, u8 log_page, u8 lsp, u8 csi, + void *log, size_t size, u64 offset) +{ + return nvme_get_log_lsi(ctrl, nsid, log_page, lsp, csi, log, size, + offset, 0); +} + +static int nvme_get_effects_log(struct nvme_ctrl *ctrl, u8 csi, + struct nvme_effects_log **log) { + struct nvme_effects_log *old, *cel = xa_load(&ctrl->cels, csi); int ret; - if (!ctrl->effects) - ctrl->effects = kzalloc(sizeof(*ctrl->effects), GFP_KERNEL); + if (cel) + goto out; - if (!ctrl->effects) - return 0; + cel = kzalloc(sizeof(*cel), GFP_KERNEL); + if (!cel) + return -ENOMEM; - ret = nvme_get_log(ctrl, NVME_NSID_ALL, NVME_LOG_CMD_EFFECTS, 0, - ctrl->effects, sizeof(*ctrl->effects), 0); + ret = nvme_get_log(ctrl, 0x00, NVME_LOG_CMD_EFFECTS, 0, csi, + cel, sizeof(*cel), 0); if (ret) { - kfree(ctrl->effects); - ctrl->effects = NULL; + kfree(cel); + return ret; } + + old = xa_store(&ctrl->cels, csi, cel, GFP_KERNEL); + if (xa_is_err(old)) { + kfree(cel); + return xa_err(old); + } +out: + *log = cel; + return 0; +} + +static inline u32 nvme_mps_to_sectors(struct nvme_ctrl *ctrl, u32 units) +{ + u32 page_shift = NVME_CAP_MPSMIN(ctrl->cap) + 12, val; + + if (check_shl_overflow(1U, units + page_shift - 9, &val)) + return UINT_MAX; + return val; +} + +static int nvme_init_non_mdts_limits(struct nvme_ctrl *ctrl) +{ + struct nvme_command c = { }; + struct nvme_id_ctrl_nvm *id; + int ret; + + /* + * Even though NVMe spec explicitly states that MDTS is not applicable + * to the write-zeroes, we are cautious and limit the size to the + * controllers max_hw_sectors value, which is based on the MDTS field + * and possibly other limiting factors. + */ + if ((ctrl->oncs & NVME_CTRL_ONCS_WRITE_ZEROES) && + !(ctrl->quirks & NVME_QUIRK_DISABLE_WRITE_ZEROES)) + ctrl->max_zeroes_sectors = ctrl->max_hw_sectors; + else + ctrl->max_zeroes_sectors = 0; + + if (!nvme_is_io_ctrl(ctrl) || + !nvme_id_cns_ok(ctrl, NVME_ID_CNS_CS_CTRL) || + test_bit(NVME_CTRL_SKIP_ID_CNS_CS, &ctrl->flags)) + return 0; + + id = kzalloc(sizeof(*id), GFP_KERNEL); + if (!id) + return -ENOMEM; + + c.identify.opcode = nvme_admin_identify; + c.identify.cns = NVME_ID_CNS_CS_CTRL; + c.identify.csi = NVME_CSI_NVM; + + ret = nvme_submit_sync_cmd(ctrl->admin_q, &c, id, sizeof(*id)); + if (ret) + goto free_data; + + ctrl->dmrl = id->dmrl; + ctrl->dmrsl = le32_to_cpu(id->dmrsl); + if (id->wzsl) + ctrl->max_zeroes_sectors = nvme_mps_to_sectors(ctrl, id->wzsl); + +free_data: + if (ret > 0) + set_bit(NVME_CTRL_SKIP_ID_CNS_CS, &ctrl->flags); + kfree(id); return ret; } -/* - * Initialize the cached copies of the Identify data and various controller - * register in our nvme_ctrl structure. This should be called as soon as - * the admin queue is fully up and running. - */ -int nvme_init_identify(struct nvme_ctrl *ctrl) +static int nvme_init_effects_log(struct nvme_ctrl *ctrl, + u8 csi, struct nvme_effects_log **log) { - struct nvme_id_ctrl *id; - int ret, page_shift; - u32 max_hw_sectors; - bool prev_apst_enabled; + struct nvme_effects_log *effects, *old; - ret = ctrl->ops->reg_read32(ctrl, NVME_REG_VS, &ctrl->vs); - if (ret) { - dev_err(ctrl->device, "Reading VS failed (%d)\n", ret); - return ret; + effects = kzalloc(sizeof(*effects), GFP_KERNEL); + if (!effects) + return -ENOMEM; + + old = xa_store(&ctrl->cels, csi, effects, GFP_KERNEL); + if (xa_is_err(old)) { + kfree(effects); + return xa_err(old); } - page_shift = NVME_CAP_MPSMIN(ctrl->cap) + 12; - ctrl->sqsize = min_t(int, NVME_CAP_MQES(ctrl->cap), ctrl->sqsize); - if (ctrl->vs >= NVME_VS(1, 1, 0)) - ctrl->subsystem = NVME_CAP_NSSRC(ctrl->cap); + *log = effects; + return 0; +} + +static void nvme_init_known_nvm_effects(struct nvme_ctrl *ctrl) +{ + struct nvme_effects_log *log = ctrl->effects; + + log->acs[nvme_admin_format_nvm] |= cpu_to_le32(NVME_CMD_EFFECTS_LBCC | + NVME_CMD_EFFECTS_NCC | + NVME_CMD_EFFECTS_CSE_MASK); + log->acs[nvme_admin_sanitize_nvm] |= cpu_to_le32(NVME_CMD_EFFECTS_LBCC | + NVME_CMD_EFFECTS_CSE_MASK); + + /* + * The spec says the result of a security receive command depends on + * the previous security send command. As such, many vendors log this + * command as one to submitted only when no other commands to the same + * namespace are outstanding. The intention is to tell the host to + * prevent mixing security send and receive. + * + * This driver can only enforce such exclusive access against IO + * queues, though. We are not readily able to enforce such a rule for + * two commands to the admin queue, which is the only queue that + * matters for this command. + * + * Rather than blindly freezing the IO queues for this effect that + * doesn't even apply to IO, mask it off. + */ + log->acs[nvme_admin_security_recv] &= cpu_to_le32(~NVME_CMD_EFFECTS_CSE_MASK); + + log->iocs[nvme_cmd_write] |= cpu_to_le32(NVME_CMD_EFFECTS_LBCC); + log->iocs[nvme_cmd_write_zeroes] |= cpu_to_le32(NVME_CMD_EFFECTS_LBCC); + log->iocs[nvme_cmd_write_uncor] |= cpu_to_le32(NVME_CMD_EFFECTS_LBCC); +} + +static int nvme_init_effects(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id) +{ + int ret = 0; + + if (ctrl->effects) + return 0; + + if (id->lpa & NVME_CTRL_LPA_CMD_EFFECTS_LOG) { + ret = nvme_get_effects_log(ctrl, NVME_CSI_NVM, &ctrl->effects); + if (ret < 0) + return ret; + } + + if (!ctrl->effects) { + ret = nvme_init_effects_log(ctrl, NVME_CSI_NVM, &ctrl->effects); + if (ret < 0) + return ret; + } + + nvme_init_known_nvm_effects(ctrl); + return 0; +} + +static int nvme_check_ctrl_fabric_info(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id) +{ + /* + * In fabrics we need to verify the cntlid matches the + * admin connect + */ + if (ctrl->cntlid != le16_to_cpu(id->cntlid)) { + dev_err(ctrl->device, + "Mismatching cntlid: Connect %u vs Identify %u, rejecting\n", + ctrl->cntlid, le16_to_cpu(id->cntlid)); + return -EINVAL; + } + + if (!nvme_discovery_ctrl(ctrl) && !ctrl->kas) { + dev_err(ctrl->device, + "keep-alive support is mandatory for fabrics\n"); + return -EINVAL; + } + + if (nvme_is_io_ctrl(ctrl) && ctrl->ioccsz < 4) { + dev_err(ctrl->device, + "I/O queue command capsule supported size %d < 4\n", + ctrl->ioccsz); + return -EINVAL; + } + + if (nvme_is_io_ctrl(ctrl) && ctrl->iorcsz < 1) { + dev_err(ctrl->device, + "I/O queue response capsule supported size %d < 1\n", + ctrl->iorcsz); + return -EINVAL; + } + + if (!ctrl->maxcmd) { + dev_warn(ctrl->device, + "Firmware bug: maximum outstanding commands is 0\n"); + ctrl->maxcmd = ctrl->sqsize + 1; + } + + return 0; +} + +static int nvme_init_identify(struct nvme_ctrl *ctrl) +{ + struct queue_limits lim; + struct nvme_id_ctrl *id; + u32 max_hw_sectors; + bool prev_apst_enabled; + int ret; ret = nvme_identify_ctrl(ctrl, &id); if (ret) { @@ -2878,21 +3532,11 @@ int nvme_init_identify(struct nvme_ctrl *ctrl) return -EIO; } - if (id->lpa & NVME_CTRL_LPA_CMD_EFFECTS_LOG) { - ret = nvme_get_effects_log(ctrl); - if (ret < 0) - goto out_free; - } - if (!(ctrl->ops->flags & NVME_F_FABRICS)) ctrl->cntlid = le16_to_cpu(id->cntlid); if (!ctrl->identified) { - int i; - - ret = nvme_init_subsystem(ctrl, id); - if (ret) - goto out_free; + unsigned int i; /* * Check for quirks. Quirk can depend on firmware version, @@ -2906,7 +3550,17 @@ int nvme_init_identify(struct nvme_ctrl *ctrl) if (quirk_matches(id, &core_quirks[i])) ctrl->quirks |= core_quirks[i].quirks; } + + ret = nvme_init_subsystem(ctrl, id); + if (ret) + goto out_free; + + ret = nvme_init_effects(ctrl, id); + if (ret) + goto out_free; } + memcpy(ctrl->subsys->firmware_rev, id->fr, + sizeof(ctrl->subsys->firmware_rev)); if (force_apst && (ctrl->quirks & NVME_QUIRK_NO_DEEPEST_PS)) { dev_warn(ctrl->device, "forcibly allowing all power states due to nvme_core.force_apst -- use at your own risk\n"); @@ -2927,28 +3581,36 @@ int nvme_init_identify(struct nvme_ctrl *ctrl) atomic_set(&ctrl->abort_limit, id->acl + 1); ctrl->vwc = id->vwc; if (id->mdts) - max_hw_sectors = 1 << (id->mdts + page_shift - 9); + max_hw_sectors = nvme_mps_to_sectors(ctrl, id->mdts); else max_hw_sectors = UINT_MAX; ctrl->max_hw_sectors = min_not_zero(ctrl->max_hw_sectors, max_hw_sectors); - nvme_set_queue_limits(ctrl, ctrl->admin_q); + lim = queue_limits_start_update(ctrl->admin_q); + nvme_set_ctrl_limits(ctrl, &lim, true); + ret = queue_limits_commit_update(ctrl->admin_q, &lim); + if (ret) + goto out_free; + ctrl->sgls = le32_to_cpu(id->sgls); ctrl->kas = le16_to_cpu(id->kas); ctrl->max_namespaces = le32_to_cpu(id->mnan); ctrl->ctratt = le32_to_cpu(id->ctratt); + ctrl->cntrltype = id->cntrltype; + ctrl->dctype = id->dctype; + if (id->rtd3e) { /* us -> s */ - u32 transition_time = le32_to_cpu(id->rtd3e) / 1000000; + u32 transition_time = le32_to_cpu(id->rtd3e) / USEC_PER_SEC; ctrl->shutdown_timeout = clamp_t(unsigned int, transition_time, shutdown_timeout, 60); if (ctrl->shutdown_timeout != shutdown_timeout) dev_info(ctrl->device, - "Shutdown timeout set to %u seconds\n", + "D3 entry latency set to %u seconds\n", ctrl->shutdown_timeout); } else ctrl->shutdown_timeout = shutdown_timeout; @@ -2974,25 +3636,9 @@ int nvme_init_identify(struct nvme_ctrl *ctrl) ctrl->iorcsz = le32_to_cpu(id->iorcsz); ctrl->maxcmd = le16_to_cpu(id->maxcmd); - /* - * In fabrics we need to verify the cntlid matches the - * admin connect - */ - if (ctrl->cntlid != le16_to_cpu(id->cntlid)) { - dev_err(ctrl->device, - "Mismatching cntlid: Connect %u vs Identify " - "%u, rejecting\n", - ctrl->cntlid, le16_to_cpu(id->cntlid)); - ret = -EINVAL; - goto out_free; - } - - if (!ctrl->opts->discovery_nqn && !ctrl->kas) { - dev_err(ctrl->device, - "keep-alive support is mandatory for fabrics\n"); - ret = -EINVAL; + ret = nvme_check_ctrl_fabric_info(ctrl, id); + if (ret) goto out_free; - } } else { ctrl->hmpre = le32_to_cpu(id->hmpre); ctrl->hmmin = le32_to_cpu(id->hmmin); @@ -3000,486 +3646,243 @@ int nvme_init_identify(struct nvme_ctrl *ctrl) ctrl->hmmaxd = le16_to_cpu(id->hmmaxd); } - ret = nvme_mpath_init(ctrl, id); - kfree(id); - + ret = nvme_mpath_init_identify(ctrl, id); if (ret < 0) - return ret; + goto out_free; if (ctrl->apst_enabled && !prev_apst_enabled) dev_pm_qos_expose_latency_tolerance(ctrl->device); else if (!ctrl->apst_enabled && prev_apst_enabled) dev_pm_qos_hide_latency_tolerance(ctrl->device); +out_free: + kfree(id); + return ret; +} + +/* + * Initialize the cached copies of the Identify data and various controller + * register in our nvme_ctrl structure. This should be called as soon as + * the admin queue is fully up and running. + */ +int nvme_init_ctrl_finish(struct nvme_ctrl *ctrl, bool was_suspended) +{ + int ret; + + ret = ctrl->ops->reg_read32(ctrl, NVME_REG_VS, &ctrl->vs); + if (ret) { + dev_err(ctrl->device, "Reading VS failed (%d)\n", ret); + return ret; + } + + ctrl->sqsize = min_t(u16, NVME_CAP_MQES(ctrl->cap), ctrl->sqsize); + + if (ctrl->vs >= NVME_VS(1, 1, 0)) + ctrl->subsystem = NVME_CAP_NSSRC(ctrl->cap); + + ret = nvme_init_identify(ctrl); + if (ret) + return ret; + + if (nvme_admin_ctrl(ctrl)) { + /* + * An admin controller has one admin queue, but no I/O queues. + * Override queue_count so it only creates an admin queue. + */ + dev_dbg(ctrl->device, + "Subsystem %s is an administrative controller", + ctrl->subsys->subnqn); + ctrl->queue_count = 1; + } ret = nvme_configure_apst(ctrl); if (ret < 0) return ret; - + ret = nvme_configure_timestamp(ctrl); if (ret < 0) return ret; - ret = nvme_configure_directives(ctrl); + ret = nvme_configure_host_options(ctrl); if (ret < 0) return ret; - ret = nvme_configure_acre(ctrl); - if (ret < 0) - return ret; + nvme_configure_opal(ctrl, was_suspended); - if (!ctrl->identified) - nvme_hwmon_init(ctrl); + if (!ctrl->identified && !nvme_discovery_ctrl(ctrl)) { + /* + * Do not return errors unless we are in a controller reset, + * the controller works perfectly fine without hwmon. + */ + ret = nvme_hwmon_init(ctrl); + if (ret == -EINTR) + return ret; + } + clear_bit(NVME_CTRL_DIRTY_CAPABILITY, &ctrl->flags); ctrl->identified = true; - return 0; + nvme_start_keep_alive(ctrl); -out_free: - kfree(id); - return ret; + return 0; } -EXPORT_SYMBOL_GPL(nvme_init_identify); +EXPORT_SYMBOL_GPL(nvme_init_ctrl_finish); static int nvme_dev_open(struct inode *inode, struct file *file) { struct nvme_ctrl *ctrl = container_of(inode->i_cdev, struct nvme_ctrl, cdev); - switch (ctrl->state) { + switch (nvme_ctrl_state(ctrl)) { case NVME_CTRL_LIVE: break; default: return -EWOULDBLOCK; } + nvme_get_ctrl(ctrl); + if (!try_module_get(ctrl->ops->module)) { + nvme_put_ctrl(ctrl); + return -EINVAL; + } + file->private_data = ctrl; return 0; } -static int nvme_dev_user_cmd(struct nvme_ctrl *ctrl, void __user *argp) +static int nvme_dev_release(struct inode *inode, struct file *file) { - struct nvme_ns *ns; - int ret; - - down_read(&ctrl->namespaces_rwsem); - if (list_empty(&ctrl->namespaces)) { - ret = -ENOTTY; - goto out_unlock; - } - - ns = list_first_entry(&ctrl->namespaces, struct nvme_ns, list); - if (ns != list_last_entry(&ctrl->namespaces, struct nvme_ns, list)) { - dev_warn(ctrl->device, - "NVME_IOCTL_IO_CMD not supported when multiple namespaces present!\n"); - ret = -EINVAL; - goto out_unlock; - } - - dev_warn(ctrl->device, - "using deprecated NVME_IOCTL_IO_CMD ioctl on the char device!\n"); - kref_get(&ns->kref); - up_read(&ctrl->namespaces_rwsem); - - ret = nvme_user_cmd(ctrl, ns, argp); - nvme_put_ns(ns); - return ret; - -out_unlock: - up_read(&ctrl->namespaces_rwsem); - return ret; -} + struct nvme_ctrl *ctrl = + container_of(inode->i_cdev, struct nvme_ctrl, cdev); -static long nvme_dev_ioctl(struct file *file, unsigned int cmd, - unsigned long arg) -{ - struct nvme_ctrl *ctrl = file->private_data; - void __user *argp = (void __user *)arg; - - switch (cmd) { - case NVME_IOCTL_ADMIN_CMD: - return nvme_user_cmd(ctrl, NULL, argp); - case NVME_IOCTL_ADMIN64_CMD: - return nvme_user_cmd64(ctrl, NULL, argp); - case NVME_IOCTL_IO_CMD: - return nvme_dev_user_cmd(ctrl, argp); - case NVME_IOCTL_RESET: - dev_warn(ctrl->device, "resetting controller\n"); - return nvme_reset_ctrl_sync(ctrl); - case NVME_IOCTL_SUBSYS_RESET: - return nvme_reset_subsystem(ctrl); - case NVME_IOCTL_RESCAN: - nvme_queue_scan(ctrl); - return 0; - default: - return -ENOTTY; - } + module_put(ctrl->ops->module); + nvme_put_ctrl(ctrl); + return 0; } static const struct file_operations nvme_dev_fops = { .owner = THIS_MODULE, .open = nvme_dev_open, + .release = nvme_dev_release, .unlocked_ioctl = nvme_dev_ioctl, .compat_ioctl = compat_ptr_ioctl, + .uring_cmd = nvme_dev_uring_cmd, }; -static ssize_t nvme_sysfs_reset(struct device *dev, - struct device_attribute *attr, const char *buf, - size_t count) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - int ret; - - ret = nvme_reset_ctrl_sync(ctrl); - if (ret < 0) - return ret; - return count; -} -static DEVICE_ATTR(reset_controller, S_IWUSR, NULL, nvme_sysfs_reset); - -static ssize_t nvme_sysfs_rescan(struct device *dev, - struct device_attribute *attr, const char *buf, - size_t count) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - nvme_queue_scan(ctrl); - return count; -} -static DEVICE_ATTR(rescan_controller, S_IWUSR, NULL, nvme_sysfs_rescan); - -static inline struct nvme_ns_head *dev_to_ns_head(struct device *dev) -{ - struct gendisk *disk = dev_to_disk(dev); - - if (disk->fops == &nvme_fops) - return nvme_get_ns_from_dev(dev)->head; - else - return disk->private_data; -} - -static ssize_t wwid_show(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct nvme_ns_head *head = dev_to_ns_head(dev); - struct nvme_ns_ids *ids = &head->ids; - struct nvme_subsystem *subsys = head->subsys; - int serial_len = sizeof(subsys->serial); - int model_len = sizeof(subsys->model); - - if (!uuid_is_null(&ids->uuid)) - return sprintf(buf, "uuid.%pU\n", &ids->uuid); - - if (memchr_inv(ids->nguid, 0, sizeof(ids->nguid))) - return sprintf(buf, "eui.%16phN\n", ids->nguid); - - if (memchr_inv(ids->eui64, 0, sizeof(ids->eui64))) - return sprintf(buf, "eui.%8phN\n", ids->eui64); - - while (serial_len > 0 && (subsys->serial[serial_len - 1] == ' ' || - subsys->serial[serial_len - 1] == '\0')) - serial_len--; - while (model_len > 0 && (subsys->model[model_len - 1] == ' ' || - subsys->model[model_len - 1] == '\0')) - model_len--; - - return sprintf(buf, "nvme.%04x-%*phN-%*phN-%08x\n", subsys->vendor_id, - serial_len, subsys->serial, model_len, subsys->model, - head->ns_id); -} -static DEVICE_ATTR_RO(wwid); - -static ssize_t nguid_show(struct device *dev, struct device_attribute *attr, - char *buf) +static struct nvme_ns_head *nvme_find_ns_head(struct nvme_ctrl *ctrl, + unsigned nsid) { - return sprintf(buf, "%pU\n", dev_to_ns_head(dev)->ids.nguid); -} -static DEVICE_ATTR_RO(nguid); + struct nvme_ns_head *h; -static ssize_t uuid_show(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct nvme_ns_ids *ids = &dev_to_ns_head(dev)->ids; + lockdep_assert_held(&ctrl->subsys->lock); - /* For backward compatibility expose the NGUID to userspace if - * we have no UUID set - */ - if (uuid_is_null(&ids->uuid)) { - printk_ratelimited(KERN_WARNING - "No UUID available providing old NGUID\n"); - return sprintf(buf, "%pU\n", ids->nguid); + list_for_each_entry(h, &ctrl->subsys->nsheads, entry) { + /* + * Private namespaces can share NSIDs under some conditions. + * In that case we can't use the same ns_head for namespaces + * with the same NSID. + */ + if (h->ns_id != nsid || !nvme_is_unique_nsid(ctrl, h)) + continue; + if (nvme_tryget_ns_head(h)) + return h; } - return sprintf(buf, "%pU\n", &ids->uuid); -} -static DEVICE_ATTR_RO(uuid); -static ssize_t eui_show(struct device *dev, struct device_attribute *attr, - char *buf) -{ - return sprintf(buf, "%8ph\n", dev_to_ns_head(dev)->ids.eui64); + return NULL; } -static DEVICE_ATTR_RO(eui); -static ssize_t nsid_show(struct device *dev, struct device_attribute *attr, - char *buf) +static int nvme_subsys_check_duplicate_ids(struct nvme_subsystem *subsys, + struct nvme_ns_ids *ids) { - return sprintf(buf, "%d\n", dev_to_ns_head(dev)->ns_id); -} -static DEVICE_ATTR_RO(nsid); - -static struct attribute *nvme_ns_id_attrs[] = { - &dev_attr_wwid.attr, - &dev_attr_uuid.attr, - &dev_attr_nguid.attr, - &dev_attr_eui.attr, - &dev_attr_nsid.attr, -#ifdef CONFIG_NVME_MULTIPATH - &dev_attr_ana_grpid.attr, - &dev_attr_ana_state.attr, -#endif - NULL, -}; + bool has_uuid = !uuid_is_null(&ids->uuid); + bool has_nguid = memchr_inv(ids->nguid, 0, sizeof(ids->nguid)); + bool has_eui64 = memchr_inv(ids->eui64, 0, sizeof(ids->eui64)); + struct nvme_ns_head *h; -static umode_t nvme_ns_id_attrs_are_visible(struct kobject *kobj, - struct attribute *a, int n) -{ - struct device *dev = container_of(kobj, struct device, kobj); - struct nvme_ns_ids *ids = &dev_to_ns_head(dev)->ids; + lockdep_assert_held(&subsys->lock); - if (a == &dev_attr_uuid.attr) { - if (uuid_is_null(&ids->uuid) && - !memchr_inv(ids->nguid, 0, sizeof(ids->nguid))) - return 0; - } - if (a == &dev_attr_nguid.attr) { - if (!memchr_inv(ids->nguid, 0, sizeof(ids->nguid))) - return 0; - } - if (a == &dev_attr_eui.attr) { - if (!memchr_inv(ids->eui64, 0, sizeof(ids->eui64))) - return 0; - } -#ifdef CONFIG_NVME_MULTIPATH - if (a == &dev_attr_ana_grpid.attr || a == &dev_attr_ana_state.attr) { - if (dev_to_disk(dev)->fops != &nvme_fops) /* per-path attr */ - return 0; - if (!nvme_ctrl_use_ana(nvme_get_ns_from_dev(dev)->ctrl)) - return 0; + list_for_each_entry(h, &subsys->nsheads, entry) { + if (has_uuid && uuid_equal(&ids->uuid, &h->ids.uuid)) + return -EINVAL; + if (has_nguid && + memcmp(&ids->nguid, &h->ids.nguid, sizeof(ids->nguid)) == 0) + return -EINVAL; + if (has_eui64 && + memcmp(&ids->eui64, &h->ids.eui64, sizeof(ids->eui64)) == 0) + return -EINVAL; } -#endif - return a->mode; -} - -static const struct attribute_group nvme_ns_id_attr_group = { - .attrs = nvme_ns_id_attrs, - .is_visible = nvme_ns_id_attrs_are_visible, -}; - -const struct attribute_group *nvme_ns_id_attr_groups[] = { - &nvme_ns_id_attr_group, -#ifdef CONFIG_NVM - &nvme_nvm_attr_group, -#endif - NULL, -}; - -#define nvme_show_str_function(field) \ -static ssize_t field##_show(struct device *dev, \ - struct device_attribute *attr, char *buf) \ -{ \ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); \ - return sprintf(buf, "%.*s\n", \ - (int)sizeof(ctrl->subsys->field), ctrl->subsys->field); \ -} \ -static DEVICE_ATTR(field, S_IRUGO, field##_show, NULL); - -nvme_show_str_function(model); -nvme_show_str_function(serial); -nvme_show_str_function(firmware_rev); - -#define nvme_show_int_function(field) \ -static ssize_t field##_show(struct device *dev, \ - struct device_attribute *attr, char *buf) \ -{ \ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); \ - return sprintf(buf, "%d\n", ctrl->field); \ -} \ -static DEVICE_ATTR(field, S_IRUGO, field##_show, NULL); - -nvme_show_int_function(cntlid); -nvme_show_int_function(numa_node); -nvme_show_int_function(queue_count); -nvme_show_int_function(sqsize); - -static ssize_t nvme_sysfs_delete(struct device *dev, - struct device_attribute *attr, const char *buf, - size_t count) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - /* Can't delete non-created controllers */ - if (!ctrl->created) - return -EBUSY; - - if (device_remove_file_self(dev, attr)) - nvme_delete_ctrl_sync(ctrl); - return count; + return 0; } -static DEVICE_ATTR(delete_controller, S_IWUSR, NULL, nvme_sysfs_delete); -static ssize_t nvme_sysfs_show_transport(struct device *dev, - struct device_attribute *attr, - char *buf) +static void nvme_cdev_rel(struct device *dev) { - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - return snprintf(buf, PAGE_SIZE, "%s\n", ctrl->ops->name); + ida_free(&nvme_ns_chr_minor_ida, MINOR(dev->devt)); } -static DEVICE_ATTR(transport, S_IRUGO, nvme_sysfs_show_transport, NULL); -static ssize_t nvme_sysfs_show_state(struct device *dev, - struct device_attribute *attr, - char *buf) +void nvme_cdev_del(struct cdev *cdev, struct device *cdev_device) { - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - static const char *const state_name[] = { - [NVME_CTRL_NEW] = "new", - [NVME_CTRL_LIVE] = "live", - [NVME_CTRL_RESETTING] = "resetting", - [NVME_CTRL_CONNECTING] = "connecting", - [NVME_CTRL_DELETING] = "deleting", - [NVME_CTRL_DEAD] = "dead", - }; - - if ((unsigned)ctrl->state < ARRAY_SIZE(state_name) && - state_name[ctrl->state]) - return sprintf(buf, "%s\n", state_name[ctrl->state]); - - return sprintf(buf, "unknown state\n"); + cdev_device_del(cdev, cdev_device); + put_device(cdev_device); } -static DEVICE_ATTR(state, S_IRUGO, nvme_sysfs_show_state, NULL); - -static ssize_t nvme_sysfs_show_subsysnqn(struct device *dev, - struct device_attribute *attr, - char *buf) +int nvme_cdev_add(struct cdev *cdev, struct device *cdev_device, + const struct file_operations *fops, struct module *owner) { - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - return snprintf(buf, PAGE_SIZE, "%s\n", ctrl->subsys->subnqn); -} -static DEVICE_ATTR(subsysnqn, S_IRUGO, nvme_sysfs_show_subsysnqn, NULL); + int minor, ret; -static ssize_t nvme_sysfs_show_hostnqn(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + minor = ida_alloc(&nvme_ns_chr_minor_ida, GFP_KERNEL); + if (minor < 0) + return minor; + cdev_device->devt = MKDEV(MAJOR(nvme_ns_chr_devt), minor); + cdev_device->class = &nvme_ns_chr_class; + cdev_device->release = nvme_cdev_rel; + device_initialize(cdev_device); + cdev_init(cdev, fops); + cdev->owner = owner; + ret = cdev_device_add(cdev, cdev_device); + if (ret) + put_device(cdev_device); - return snprintf(buf, PAGE_SIZE, "%s\n", ctrl->opts->host->nqn); + return ret; } -static DEVICE_ATTR(hostnqn, S_IRUGO, nvme_sysfs_show_hostnqn, NULL); -static ssize_t nvme_sysfs_show_hostid(struct device *dev, - struct device_attribute *attr, - char *buf) +static int nvme_ns_chr_open(struct inode *inode, struct file *file) { - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - return snprintf(buf, PAGE_SIZE, "%pU\n", &ctrl->opts->host->id); + return nvme_ns_open(container_of(inode->i_cdev, struct nvme_ns, cdev)); } -static DEVICE_ATTR(hostid, S_IRUGO, nvme_sysfs_show_hostid, NULL); - -static ssize_t nvme_sysfs_show_address(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - return ctrl->ops->get_address(ctrl, buf, PAGE_SIZE); -} -static DEVICE_ATTR(address, S_IRUGO, nvme_sysfs_show_address, NULL); - -static struct attribute *nvme_dev_attrs[] = { - &dev_attr_reset_controller.attr, - &dev_attr_rescan_controller.attr, - &dev_attr_model.attr, - &dev_attr_serial.attr, - &dev_attr_firmware_rev.attr, - &dev_attr_cntlid.attr, - &dev_attr_delete_controller.attr, - &dev_attr_transport.attr, - &dev_attr_subsysnqn.attr, - &dev_attr_address.attr, - &dev_attr_state.attr, - &dev_attr_numa_node.attr, - &dev_attr_queue_count.attr, - &dev_attr_sqsize.attr, - &dev_attr_hostnqn.attr, - &dev_attr_hostid.attr, - NULL -}; -static umode_t nvme_dev_attrs_are_visible(struct kobject *kobj, - struct attribute *a, int n) +static int nvme_ns_chr_release(struct inode *inode, struct file *file) { - struct device *dev = container_of(kobj, struct device, kobj); - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - if (a == &dev_attr_delete_controller.attr && !ctrl->ops->delete_ctrl) - return 0; - if (a == &dev_attr_address.attr && !ctrl->ops->get_address) - return 0; - if (a == &dev_attr_hostnqn.attr && !ctrl->opts) - return 0; - if (a == &dev_attr_hostid.attr && !ctrl->opts) - return 0; - - return a->mode; + nvme_ns_release(container_of(inode->i_cdev, struct nvme_ns, cdev)); + return 0; } -static struct attribute_group nvme_dev_attrs_group = { - .attrs = nvme_dev_attrs, - .is_visible = nvme_dev_attrs_are_visible, -}; - -static const struct attribute_group *nvme_dev_attr_groups[] = { - &nvme_dev_attrs_group, - NULL, +static const struct file_operations nvme_ns_chr_fops = { + .owner = THIS_MODULE, + .open = nvme_ns_chr_open, + .release = nvme_ns_chr_release, + .unlocked_ioctl = nvme_ns_chr_ioctl, + .compat_ioctl = compat_ptr_ioctl, + .uring_cmd = nvme_ns_chr_uring_cmd, + .uring_cmd_iopoll = nvme_ns_chr_uring_cmd_iopoll, }; -static struct nvme_ns_head *nvme_find_ns_head(struct nvme_subsystem *subsys, - unsigned nsid) +static int nvme_add_ns_cdev(struct nvme_ns *ns) { - struct nvme_ns_head *h; - - lockdep_assert_held(&subsys->lock); - - list_for_each_entry(h, &subsys->nsheads, entry) { - if (h->ns_id == nsid && kref_get_unless_zero(&h->ref)) - return h; - } - - return NULL; -} - -static int __nvme_check_ids(struct nvme_subsystem *subsys, - struct nvme_ns_head *new) -{ - struct nvme_ns_head *h; - - lockdep_assert_held(&subsys->lock); + int ret; - list_for_each_entry(h, &subsys->nsheads, entry) { - if (nvme_ns_ids_valid(&new->ids) && - nvme_ns_ids_equal(&new->ids, &h->ids)) - return -EINVAL; - } + ns->cdev_device.parent = ns->ctrl->device; + ret = dev_set_name(&ns->cdev_device, "ng%dn%d", + ns->ctrl->instance, ns->head->instance); + if (ret) + return ret; - return 0; + return nvme_cdev_add(&ns->cdev, &ns->cdev_device, &nvme_ns_chr_fops, + ns->ctrl->ops->module); } static struct nvme_ns_head *nvme_alloc_ns_head(struct nvme_ctrl *ctrl, - unsigned nsid, struct nvme_ns_ids *ids) + struct nvme_ns_info *info) { struct nvme_ns_head *head; size_t size = sizeof(*head); @@ -3492,7 +3895,7 @@ static struct nvme_ns_head *nvme_alloc_ns_head(struct nvme_ctrl *ctrl, head = kzalloc(size, GFP_KERNEL); if (!head) goto out; - ret = ida_simple_get(&ctrl->subsys->ns_ida, 1, 0, GFP_KERNEL); + ret = ida_alloc_min(&ctrl->subsys->ns_ida, 1, GFP_KERNEL); if (ret < 0) goto out_free_head; head->instance = ret; @@ -3501,16 +3904,20 @@ static struct nvme_ns_head *nvme_alloc_ns_head(struct nvme_ctrl *ctrl, if (ret) goto out_ida_remove; head->subsys = ctrl->subsys; - head->ns_id = nsid; - head->ids = *ids; + head->ns_id = info->nsid; + head->ids = info->ids; + head->shared = info->is_shared; + head->rotational = info->is_rotational; + ratelimit_state_init(&head->rs_nuse, 5 * HZ, 1); + ratelimit_set_flags(&head->rs_nuse, RATELIMIT_MSG_ON_RELEASE); kref_init(&head->ref); - ret = __nvme_check_ids(ctrl->subsys, head); - if (ret) { - dev_err(ctrl->device, - "duplicate IDs for nsid %d\n", nsid); - goto out_cleanup_srcu; - } + if (head->ids.csi) { + ret = nvme_get_effects_log(ctrl, head->ids.csi, &head->effects); + if (ret) + goto out_cleanup_srcu; + } else + head->effects = ctrl->effects; ret = nvme_mpath_alloc_disk(ctrl, head); if (ret) @@ -3524,7 +3931,7 @@ static struct nvme_ns_head *nvme_alloc_ns_head(struct nvme_ctrl *ctrl, out_cleanup_srcu: cleanup_srcu_struct(&head->srcu); out_ida_remove: - ida_simple_remove(&ctrl->subsys->ns_ida, head->instance); + ida_free(&ctrl->subsys->ns_ida, head->instance); out_free_head: kfree(head); out: @@ -3533,49 +3940,123 @@ out: return ERR_PTR(ret); } -static int nvme_init_ns_head(struct nvme_ns *ns, unsigned nsid, - struct nvme_id_ns *id) +static int nvme_global_check_duplicate_ids(struct nvme_subsystem *this, + struct nvme_ns_ids *ids) +{ + struct nvme_subsystem *s; + int ret = 0; + + /* + * Note that this check is racy as we try to avoid holding the global + * lock over the whole ns_head creation. But it is only intended as + * a sanity check anyway. + */ + mutex_lock(&nvme_subsystems_lock); + list_for_each_entry(s, &nvme_subsystems, entry) { + if (s == this) + continue; + mutex_lock(&s->lock); + ret = nvme_subsys_check_duplicate_ids(s, ids); + mutex_unlock(&s->lock); + if (ret) + break; + } + mutex_unlock(&nvme_subsystems_lock); + + return ret; +} + +static int nvme_init_ns_head(struct nvme_ns *ns, struct nvme_ns_info *info) { struct nvme_ctrl *ctrl = ns->ctrl; - bool is_shared = id->nmic & NVME_NS_NMIC_SHARED; struct nvme_ns_head *head = NULL; - struct nvme_ns_ids ids; - int ret = 0; + int ret; - ret = nvme_report_ns_ids(ctrl, nsid, id, &ids); + ret = nvme_global_check_duplicate_ids(ctrl->subsys, &info->ids); if (ret) { - if (ret < 0) + /* + * We've found two different namespaces on two different + * subsystems that report the same ID. This is pretty nasty + * for anything that actually requires unique device + * identification. In the kernel we need this for multipathing, + * and in user space the /dev/disk/by-id/ links rely on it. + * + * If the device also claims to be multi-path capable back off + * here now and refuse the probe the second device as this is a + * recipe for data corruption. If not this is probably a + * cheap consumer device if on the PCIe bus, so let the user + * proceed and use the shiny toy, but warn that with changing + * probing order (which due to our async probing could just be + * device taking longer to startup) the other device could show + * up at any time. + */ + nvme_print_device_info(ctrl); + if ((ns->ctrl->ops->flags & NVME_F_FABRICS) || /* !PCIe */ + ((ns->ctrl->subsys->cmic & NVME_CTRL_CMIC_MULTI_CTRL) && + info->is_shared)) { + dev_err(ctrl->device, + "ignoring nsid %d because of duplicate IDs\n", + info->nsid); return ret; - return blk_status_to_errno(nvme_error_status(ret)); + } + + dev_err(ctrl->device, + "clearing duplicate IDs for nsid %d\n", info->nsid); + dev_err(ctrl->device, + "use of /dev/disk/by-id/ may cause data corruption\n"); + memset(&info->ids.nguid, 0, sizeof(info->ids.nguid)); + memset(&info->ids.uuid, 0, sizeof(info->ids.uuid)); + memset(&info->ids.eui64, 0, sizeof(info->ids.eui64)); + ctrl->quirks |= NVME_QUIRK_BOGUS_NID; } mutex_lock(&ctrl->subsys->lock); - head = nvme_find_ns_head(ctrl->subsys, nsid); + head = nvme_find_ns_head(ctrl, info->nsid); if (!head) { - head = nvme_alloc_ns_head(ctrl, nsid, &ids); + ret = nvme_subsys_check_duplicate_ids(ctrl->subsys, &info->ids); + if (ret) { + dev_err(ctrl->device, + "duplicate IDs in subsystem for nsid %d\n", + info->nsid); + goto out_unlock; + } + head = nvme_alloc_ns_head(ctrl, info); if (IS_ERR(head)) { ret = PTR_ERR(head); goto out_unlock; } - head->shared = is_shared; } else { ret = -EINVAL; - if (!is_shared || !head->shared) { + if ((!info->is_shared || !head->shared) && + !list_empty(&head->list)) { dev_err(ctrl->device, - "Duplicate unshared namespace %d\n", nsid); + "Duplicate unshared namespace %d\n", + info->nsid); goto out_put_ns_head; } - if (!nvme_ns_ids_equal(&head->ids, &ids)) { + if (!nvme_ns_ids_equal(&head->ids, &info->ids)) { dev_err(ctrl->device, "IDs don't match for shared namespace %d\n", - nsid); + info->nsid); goto out_put_ns_head; } + + if (!multipath) { + dev_warn(ctrl->device, + "Found shared namespace %d, but multipathing not supported.\n", + info->nsid); + dev_warn_once(ctrl->device, + "Shared namespace support requires core_nvme.multipath=Y.\n"); + } } - list_add_tail(&ns->siblings, &head->list); + list_add_tail_rcu(&ns->siblings, &head->list); ns->head = head; mutex_unlock(&ctrl->subsys->lock); + +#ifdef CONFIG_NVME_MULTIPATH + cancel_delayed_work(&head->remove_work); +#endif return 0; out_put_ns_head: @@ -3585,22 +4066,16 @@ out_unlock: return ret; } -static int ns_cmp(void *priv, struct list_head *a, struct list_head *b) -{ - struct nvme_ns *nsa = container_of(a, struct nvme_ns, list); - struct nvme_ns *nsb = container_of(b, struct nvme_ns, list); - - return nsa->head->ns_id - nsb->head->ns_id; -} - -static struct nvme_ns *nvme_find_get_ns(struct nvme_ctrl *ctrl, unsigned nsid) +struct nvme_ns *nvme_find_get_ns(struct nvme_ctrl *ctrl, unsigned nsid) { struct nvme_ns *ns, *ret = NULL; + int srcu_idx; - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) { + srcu_idx = srcu_read_lock(&ctrl->srcu); + list_for_each_entry_srcu(ns, &ctrl->namespaces, list, + srcu_read_lock_held(&ctrl->srcu)) { if (ns->head->ns_id == nsid) { - if (!kref_get_unless_zero(&ns->kref)) + if (!nvme_get_ns(ns)) continue; ret = ns; break; @@ -3608,138 +4083,196 @@ static struct nvme_ns *nvme_find_get_ns(struct nvme_ctrl *ctrl, unsigned nsid) if (ns->head->ns_id > nsid) break; } - up_read(&ctrl->namespaces_rwsem); + srcu_read_unlock(&ctrl->srcu, srcu_idx); return ret; } +EXPORT_SYMBOL_NS_GPL(nvme_find_get_ns, "NVME_TARGET_PASSTHRU"); + +/* + * Add the namespace to the controller list while keeping the list ordered. + */ +static void nvme_ns_add_to_ctrl_list(struct nvme_ns *ns) +{ + struct nvme_ns *tmp; + + list_for_each_entry_reverse(tmp, &ns->ctrl->namespaces, list) { + if (tmp->head->ns_id < ns->head->ns_id) { + list_add_rcu(&ns->list, &tmp->list); + return; + } + } + list_add_rcu(&ns->list, &ns->ctrl->namespaces); +} -static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid) +static void nvme_alloc_ns(struct nvme_ctrl *ctrl, struct nvme_ns_info *info) { + struct queue_limits lim = { }; struct nvme_ns *ns; struct gendisk *disk; - struct nvme_id_ns *id; - char disk_name[DISK_NAME_LEN]; - int node = ctrl->numa_node, flags = GENHD_FL_EXT_DEVT, ret; + int node = ctrl->numa_node; + bool last_path = false; ns = kzalloc_node(sizeof(*ns), GFP_KERNEL, node); if (!ns) return; - ns->queue = blk_mq_init_queue(ctrl->tagset); - if (IS_ERR(ns->queue)) - goto out_free_ns; - if (ctrl->opts && ctrl->opts->data_digest) - ns->queue->backing_dev_info->capabilities - |= BDI_CAP_STABLE_WRITES; + lim.features |= BLK_FEAT_STABLE_WRITES; + if (ctrl->ops->supports_pci_p2pdma && + ctrl->ops->supports_pci_p2pdma(ctrl)) + lim.features |= BLK_FEAT_PCI_P2PDMA; - blk_queue_flag_set(QUEUE_FLAG_NONROT, ns->queue); - if (ctrl->ops->flags & NVME_F_PCI_P2PDMA) - blk_queue_flag_set(QUEUE_FLAG_PCI_P2PDMA, ns->queue); + disk = blk_mq_alloc_disk(ctrl->tagset, &lim, ns); + if (IS_ERR(disk)) + goto out_free_ns; + disk->fops = &nvme_bdev_ops; + disk->private_data = ns; - ns->queue->queuedata = ns; + ns->disk = disk; + ns->queue = disk->queue; ns->ctrl = ctrl; - kref_init(&ns->kref); - ns->lba_shift = 9; /* set to a default value for 512 until disk is validated */ - - blk_queue_logical_block_size(ns->queue, 1 << ns->lba_shift); - nvme_set_queue_limits(ctrl, ns->queue); - - ret = nvme_identify_ns(ctrl, nsid, &id); - if (ret) - goto out_free_queue; - if (id->ncap == 0) /* no namespace (legacy quirk) */ - goto out_free_id; + if (nvme_init_ns_head(ns, info)) + goto out_cleanup_disk; - ret = nvme_init_ns_head(ns, nsid, id); - if (ret) - goto out_free_id; - nvme_set_disk_name(disk_name, ns, ctrl, &flags); + /* + * If multipathing is enabled, the device name for all disks and not + * just those that represent shared namespaces needs to be based on the + * subsystem instance. Using the controller instance for private + * namespaces could lead to naming collisions between shared and private + * namespaces if they don't use a common numbering scheme. + * + * If multipathing is not enabled, disk names must use the controller + * instance as shared namespaces will show up as multiple block + * devices. + */ + if (nvme_ns_head_multipath(ns->head)) { + sprintf(disk->disk_name, "nvme%dc%dn%d", ctrl->subsys->instance, + ctrl->instance, ns->head->instance); + disk->flags |= GENHD_FL_HIDDEN; + } else if (multipath) { + sprintf(disk->disk_name, "nvme%dn%d", ctrl->subsys->instance, + ns->head->instance); + } else { + sprintf(disk->disk_name, "nvme%dn%d", ctrl->instance, + ns->head->instance); + } - disk = alloc_disk_node(0, node); - if (!disk) + if (nvme_update_ns_info(ns, info)) goto out_unlink_ns; - disk->fops = &nvme_fops; - disk->private_data = ns; - disk->queue = ns->queue; - disk->flags = flags; - memcpy(disk->disk_name, disk_name, DISK_NAME_LEN); - ns->disk = disk; - - if (__nvme_revalidate_disk(disk, id)) - goto out_put_disk; - - if ((ctrl->quirks & NVME_QUIRK_LIGHTNVM) && id->vs[0] == 0x1) { - ret = nvme_nvm_register(ns, disk_name, node); - if (ret) { - dev_warn(ctrl->device, "LightNVM init failure\n"); - goto out_put_disk; - } + mutex_lock(&ctrl->namespaces_lock); + /* + * Ensure that no namespaces are added to the ctrl list after the queues + * are frozen, thereby avoiding a deadlock between scan and reset. + */ + if (test_bit(NVME_CTRL_FROZEN, &ctrl->flags)) { + mutex_unlock(&ctrl->namespaces_lock); + goto out_unlink_ns; } - - down_write(&ctrl->namespaces_rwsem); - list_add_tail(&ns->list, &ctrl->namespaces); - up_write(&ctrl->namespaces_rwsem); - + nvme_ns_add_to_ctrl_list(ns); + mutex_unlock(&ctrl->namespaces_lock); + synchronize_srcu(&ctrl->srcu); nvme_get_ctrl(ctrl); - device_add_disk(ctrl->device, ns->disk, nvme_ns_id_attr_groups); + if (device_add_disk(ctrl->device, ns->disk, nvme_ns_attr_groups)) + goto out_cleanup_ns_from_list; - nvme_mpath_add_disk(ns, id); + if (!nvme_ns_head_multipath(ns->head)) + nvme_add_ns_cdev(ns); + + nvme_mpath_add_disk(ns, info->anagrpid); nvme_fault_inject_init(&ns->fault_inject, ns->disk->disk_name); - kfree(id); + + /* + * Set ns->disk->device->driver_data to ns so we can access + * ns->head->passthru_err_log_enabled in + * nvme_io_passthru_err_log_enabled_[store | show](). + */ + dev_set_drvdata(disk_to_dev(ns->disk), ns); return; - out_put_disk: - /* prevent double queue cleanup */ - ns->disk->queue = NULL; - put_disk(ns->disk); + + out_cleanup_ns_from_list: + nvme_put_ctrl(ctrl); + mutex_lock(&ctrl->namespaces_lock); + list_del_rcu(&ns->list); + mutex_unlock(&ctrl->namespaces_lock); + synchronize_srcu(&ctrl->srcu); out_unlink_ns: mutex_lock(&ctrl->subsys->lock); list_del_rcu(&ns->siblings); - if (list_empty(&ns->head->list)) + if (list_empty(&ns->head->list)) { list_del_init(&ns->head->entry); + /* + * If multipath is not configured, we still create a namespace + * head (nshead), but head->disk is not initialized in that + * case. As a result, only a single reference to nshead is held + * (via kref_init()) when it is created. Therefore, ensure that + * we do not release the reference to nshead twice if head->disk + * is not present. + */ + if (ns->head->disk) + last_path = true; + } mutex_unlock(&ctrl->subsys->lock); + if (last_path) + nvme_put_ns_head(ns->head); nvme_put_ns_head(ns->head); - out_free_id: - kfree(id); - out_free_queue: - blk_cleanup_queue(ns->queue); + out_cleanup_disk: + put_disk(disk); out_free_ns: kfree(ns); } static void nvme_ns_remove(struct nvme_ns *ns) { + bool last_path = false; + if (test_and_set_bit(NVME_NS_REMOVING, &ns->flags)) return; + clear_bit(NVME_NS_READY, &ns->flags); + set_capacity(ns->disk, 0); nvme_fault_inject_fini(&ns->fault_inject); + /* + * Ensure that !NVME_NS_READY is seen by other threads to prevent + * this ns going back into current_path. + */ + synchronize_srcu(&ns->head->srcu); + + /* wait for concurrent submissions */ + if (nvme_mpath_clear_current_path(ns)) + synchronize_srcu(&ns->head->srcu); + mutex_lock(&ns->ctrl->subsys->lock); list_del_rcu(&ns->siblings); - if (list_empty(&ns->head->list)) - list_del_init(&ns->head->entry); + if (list_empty(&ns->head->list)) { + if (!nvme_mpath_queue_if_no_path(ns->head)) + list_del_init(&ns->head->entry); + last_path = true; + } mutex_unlock(&ns->ctrl->subsys->lock); - synchronize_rcu(); /* guarantee not available in head->list */ - nvme_mpath_clear_current_path(ns); - synchronize_srcu(&ns->head->srcu); /* wait for concurrent submissions */ + /* guarantee not available in head->list */ + synchronize_srcu(&ns->head->srcu); - if (ns->disk && ns->disk->flags & GENHD_FL_UP) { - del_gendisk(ns->disk); - blk_cleanup_queue(ns->queue); - if (blk_get_integrity(ns->disk)) - blk_integrity_unregister(ns->disk); - } + if (!nvme_ns_head_multipath(ns->head)) + nvme_cdev_del(&ns->cdev, &ns->cdev_device); - down_write(&ns->ctrl->namespaces_rwsem); - list_del_init(&ns->list); - up_write(&ns->ctrl->namespaces_rwsem); + nvme_mpath_remove_sysfs_link(ns); - nvme_mpath_check_last_path(ns); + del_gendisk(ns->disk); + + mutex_lock(&ns->ctrl->namespaces_lock); + list_del_rcu(&ns->list); + mutex_unlock(&ns->ctrl->namespaces_lock); + synchronize_srcu(&ns->ctrl->srcu); + + if (last_path) + nvme_mpath_remove_disk(ns->head); nvme_put_ns(ns); } @@ -3753,17 +4286,101 @@ static void nvme_ns_remove_by_nsid(struct nvme_ctrl *ctrl, u32 nsid) } } -static void nvme_validate_ns(struct nvme_ctrl *ctrl, unsigned nsid) +static void nvme_validate_ns(struct nvme_ns *ns, struct nvme_ns_info *info) +{ + int ret = NVME_SC_INVALID_NS | NVME_STATUS_DNR; + + if (!nvme_ns_ids_equal(&ns->head->ids, &info->ids)) { + dev_err(ns->ctrl->device, + "identifiers changed for nsid %d\n", ns->head->ns_id); + goto out; + } + + ret = nvme_update_ns_info(ns, info); +out: + /* + * Only remove the namespace if we got a fatal error back from the + * device, otherwise ignore the error and just move on. + * + * TODO: we should probably schedule a delayed retry here. + */ + if (ret > 0 && (ret & NVME_STATUS_DNR)) + nvme_ns_remove(ns); +} + +static void nvme_scan_ns(struct nvme_ctrl *ctrl, unsigned nsid) { + struct nvme_ns_info info = { .nsid = nsid }; struct nvme_ns *ns; + int ret = 1; + + if (nvme_identify_ns_descs(ctrl, &info)) + return; + + if (info.ids.csi != NVME_CSI_NVM && !nvme_multi_css(ctrl)) { + dev_warn(ctrl->device, + "command set not reported for nsid: %d\n", nsid); + return; + } + + /* + * If available try to use the Command Set Independent Identify Namespace + * data structure to find all the generic information that is needed to + * set up a namespace. If not fall back to the legacy version. + */ + if ((ctrl->cap & NVME_CAP_CRMS_CRIMS) || + (info.ids.csi != NVME_CSI_NVM && info.ids.csi != NVME_CSI_ZNS) || + ctrl->vs >= NVME_VS(2, 0, 0)) + ret = nvme_ns_info_from_id_cs_indep(ctrl, &info); + if (ret > 0) + ret = nvme_ns_info_from_identify(ctrl, &info); + + if (info.is_removed) + nvme_ns_remove_by_nsid(ctrl, nsid); + + /* + * Ignore the namespace if it is not ready. We will get an AEN once it + * becomes ready and restart the scan. + */ + if (ret || !info.is_ready) + return; ns = nvme_find_get_ns(ctrl, nsid); if (ns) { - if (ns->disk && revalidate_disk(ns->disk)) - nvme_ns_remove(ns); + nvme_validate_ns(ns, &info); nvme_put_ns(ns); - } else - nvme_alloc_ns(ctrl, nsid); + } else { + nvme_alloc_ns(ctrl, &info); + } +} + +/** + * struct async_scan_info - keeps track of controller & NSIDs to scan + * @ctrl: Controller on which namespaces are being scanned + * @next_nsid: Index of next NSID to scan in ns_list + * @ns_list: Pointer to list of NSIDs to scan + * + * Note: There is a single async_scan_info structure shared by all instances + * of nvme_scan_ns_async() scanning a given controller, so the atomic + * operations on next_nsid are critical to ensure each instance scans a unique + * NSID. + */ +struct async_scan_info { + struct nvme_ctrl *ctrl; + atomic_t next_nsid; + __le32 *ns_list; +}; + +static void nvme_scan_ns_async(void *data, async_cookie_t cookie) +{ + struct async_scan_info *scan_info = data; + int idx; + u32 nsid; + + idx = (u32)atomic_fetch_inc(&scan_info->next_nsid); + nsid = le32_to_cpu(scan_info->ns_list[idx]); + + nvme_scan_ns(scan_info->ctrl, nsid); } static void nvme_remove_invalid_namespaces(struct nvme_ctrl *ctrl, @@ -3772,16 +4389,18 @@ static void nvme_remove_invalid_namespaces(struct nvme_ctrl *ctrl, struct nvme_ns *ns, *next; LIST_HEAD(rm_list); - down_write(&ctrl->namespaces_rwsem); + mutex_lock(&ctrl->namespaces_lock); list_for_each_entry_safe(ns, next, &ctrl->namespaces, list) { - if (ns->head->ns_id > nsid || test_bit(NVME_NS_DEAD, &ns->flags)) - list_move_tail(&ns->list, &rm_list); + if (ns->head->ns_id > nsid) { + list_del_rcu(&ns->list); + synchronize_srcu(&ctrl->srcu); + list_add_tail_rcu(&ns->list, &rm_list); + } } - up_write(&ctrl->namespaces_rwsem); + mutex_unlock(&ctrl->namespaces_lock); list_for_each_entry_safe(ns, next, &rm_list, list) nvme_ns_remove(ns); - } static int nvme_scan_ns_list(struct nvme_ctrl *ctrl) @@ -3790,32 +4409,47 @@ static int nvme_scan_ns_list(struct nvme_ctrl *ctrl) __le32 *ns_list; u32 prev = 0; int ret = 0, i; - - if (nvme_ctrl_limited_cns(ctrl)) - return -EOPNOTSUPP; + ASYNC_DOMAIN(domain); + struct async_scan_info scan_info; ns_list = kzalloc(NVME_IDENTIFY_DATA_SIZE, GFP_KERNEL); if (!ns_list) return -ENOMEM; + scan_info.ctrl = ctrl; + scan_info.ns_list = ns_list; for (;;) { - ret = nvme_identify_ns_list(ctrl, prev, ns_list); - if (ret) + struct nvme_command cmd = { + .identify.opcode = nvme_admin_identify, + .identify.cns = NVME_ID_CNS_NS_ACTIVE_LIST, + .identify.nsid = cpu_to_le32(prev), + }; + + ret = nvme_submit_sync_cmd(ctrl->admin_q, &cmd, ns_list, + NVME_IDENTIFY_DATA_SIZE); + if (ret) { + dev_warn(ctrl->device, + "Identify NS List failed (status=0x%x)\n", ret); goto free; + } + atomic_set(&scan_info.next_nsid, 0); for (i = 0; i < nr_entries; i++) { u32 nsid = le32_to_cpu(ns_list[i]); if (!nsid) /* end of the list? */ goto out; - nvme_validate_ns(ctrl, nsid); + async_schedule_domain(nvme_scan_ns_async, &scan_info, + &domain); while (++prev < nsid) nvme_ns_remove_by_nsid(ctrl, prev); } + async_synchronize_full_domain(&domain); } out: nvme_remove_invalid_namespaces(ctrl, prev); free: + async_synchronize_full_domain(&domain); kfree(ns_list); return ret; } @@ -3831,7 +4465,7 @@ static void nvme_scan_ns_sequential(struct nvme_ctrl *ctrl) kfree(id); for (i = 1; i <= nn; i++) - nvme_validate_ns(ctrl, i); + nvme_scan_ns(ctrl, i); nvme_remove_invalid_namespaces(ctrl, nn); } @@ -3852,8 +4486,8 @@ static void nvme_clear_changed_ns_log(struct nvme_ctrl *ctrl) * raced with us in reading the log page, which could cause us to miss * updates. */ - error = nvme_get_log(ctrl, NVME_NSID_ALL, NVME_LOG_CHANGED_NS, 0, log, - log_size, 0); + error = nvme_get_log(ctrl, NVME_NSID_ALL, NVME_LOG_CHANGED_NS, 0, + NVME_CSI_NVM, log, log_size, 0); if (error) dev_warn(ctrl->device, "reading changed ns log failed: %d\n", error); @@ -3865,24 +4499,54 @@ static void nvme_scan_work(struct work_struct *work) { struct nvme_ctrl *ctrl = container_of(work, struct nvme_ctrl, scan_work); + int ret; /* No tagset on a live ctrl means IO queues could not created */ - if (ctrl->state != NVME_CTRL_LIVE || !ctrl->tagset) + if (nvme_ctrl_state(ctrl) != NVME_CTRL_LIVE || !ctrl->tagset) return; + /* + * Identify controller limits can change at controller reset due to + * new firmware download, even though it is not common we cannot ignore + * such scenario. Controller's non-mdts limits are reported in the unit + * of logical blocks that is dependent on the format of attached + * namespace. Hence re-read the limits at the time of ns allocation. + */ + ret = nvme_init_non_mdts_limits(ctrl); + if (ret < 0) { + dev_warn(ctrl->device, + "reading non-mdts-limits failed: %d\n", ret); + return; + } + if (test_and_clear_bit(NVME_AER_NOTICE_NS_CHANGED, &ctrl->events)) { dev_info(ctrl->device, "rescanning namespaces.\n"); nvme_clear_changed_ns_log(ctrl); } mutex_lock(&ctrl->scan_lock); - if (nvme_scan_ns_list(ctrl) != 0) + if (!nvme_id_cns_ok(ctrl, NVME_ID_CNS_NS_ACTIVE_LIST)) { nvme_scan_ns_sequential(ctrl); + } else { + /* + * Fall back to sequential scan if DNR is set to handle broken + * devices which should support Identify NS List (as per the VS + * they report) but don't actually support it. + */ + ret = nvme_scan_ns_list(ctrl); + if (ret > 0 && ret & NVME_STATUS_DNR) + nvme_scan_ns_sequential(ctrl); + } mutex_unlock(&ctrl->scan_lock); - down_write(&ctrl->namespaces_rwsem); - list_sort(NULL, &ctrl->namespaces, ns_cmp); - up_write(&ctrl->namespaces_rwsem); + /* Requeue if we have missed AENs */ + if (test_bit(NVME_AER_NOTICE_NS_CHANGED, &ctrl->events)) + nvme_queue_scan(ctrl); +#ifdef CONFIG_NVME_MULTIPATH + else if (ctrl->ana_log_buf) + /* Re-read the ANA log page to not miss updates */ + queue_work(nvme_wq, &ctrl->ana_work); +#endif } /* @@ -3902,6 +4566,12 @@ void nvme_remove_namespaces(struct nvme_ctrl *ctrl) */ nvme_mpath_clear_ctrl_paths(ctrl); + /* + * Unquiesce io queues so any pending IO won't hang, especially + * those submitted from scan work + */ + nvme_unquiesce_io_queues(ctrl); + /* prevent racing with ns scanning */ flush_work(&ctrl->scan_work); @@ -3911,21 +4581,25 @@ void nvme_remove_namespaces(struct nvme_ctrl *ctrl) * removing the namespaces' disks; fail all the queues now to avoid * potentially having to clean up the failed sync later. */ - if (ctrl->state == NVME_CTRL_DEAD) - nvme_kill_queues(ctrl); + if (nvme_ctrl_state(ctrl) == NVME_CTRL_DEAD) + nvme_mark_namespaces_dead(ctrl); + + /* this is a no-op when called from the controller reset handler */ + nvme_change_ctrl_state(ctrl, NVME_CTRL_DELETING_NOIO); - down_write(&ctrl->namespaces_rwsem); - list_splice_init(&ctrl->namespaces, &ns_list); - up_write(&ctrl->namespaces_rwsem); + mutex_lock(&ctrl->namespaces_lock); + list_splice_init_rcu(&ctrl->namespaces, &ns_list, synchronize_rcu); + mutex_unlock(&ctrl->namespaces_lock); + synchronize_srcu(&ctrl->srcu); list_for_each_entry_safe(ns, next, &ns_list, list) nvme_ns_remove(ns); } EXPORT_SYMBOL_GPL(nvme_remove_namespaces); -static int nvme_class_uevent(struct device *dev, struct kobj_uevent_env *env) +static int nvme_class_uevent(const struct device *dev, struct kobj_uevent_env *env) { - struct nvme_ctrl *ctrl = + const struct nvme_ctrl *ctrl = container_of(dev, struct nvme_ctrl, ctrl_device); struct nvmf_ctrl_options *opts = ctrl->opts; int ret; @@ -3946,10 +4620,22 @@ static int nvme_class_uevent(struct device *dev, struct kobj_uevent_env *env) ret = add_uevent_var(env, "NVME_HOST_TRADDR=%s", opts->host_traddr ?: "none"); + if (ret) + return ret; + + ret = add_uevent_var(env, "NVME_HOST_IFACE=%s", + opts->host_iface ?: "none"); } return ret; } +static void nvme_change_uevent(struct nvme_ctrl *ctrl, char *envdata) +{ + char *envp[2] = { envdata, NULL }; + + kobject_uevent_env(&ctrl->device->kobj, KOBJ_CHANGE, envp); +} + static void nvme_aen_uevent(struct nvme_ctrl *ctrl) { char *envp[2] = { NULL, NULL }; @@ -3972,7 +4658,14 @@ static void nvme_async_event_work(struct work_struct *work) container_of(work, struct nvme_ctrl, async_event_work); nvme_aen_uevent(ctrl); - ctrl->ops->submit_async_event(ctrl); + + /* + * The transport drivers must guarantee AER submission here is safe by + * flushing ctrl async_event_work after changing the controller state + * from LIVE and before freeing the admin queue. + */ + if (nvme_ctrl_state(ctrl) == NVME_CTRL_LIVE) + ctrl->ops->submit_async_event(ctrl); } static bool nvme_ctrl_pp_status(struct nvme_ctrl *ctrl) @@ -3992,14 +4685,30 @@ static bool nvme_ctrl_pp_status(struct nvme_ctrl *ctrl) static void nvme_get_fw_slot_info(struct nvme_ctrl *ctrl) { struct nvme_fw_slot_info_log *log; + u8 next_fw_slot, cur_fw_slot; log = kmalloc(sizeof(*log), GFP_KERNEL); if (!log) return; - if (nvme_get_log(ctrl, NVME_NSID_ALL, NVME_LOG_FW_SLOT, 0, log, - sizeof(*log), 0)) + if (nvme_get_log(ctrl, NVME_NSID_ALL, NVME_LOG_FW_SLOT, 0, NVME_CSI_NVM, + log, sizeof(*log), 0)) { dev_warn(ctrl->device, "Get FW SLOT INFO log error\n"); + goto out_free_log; + } + + cur_fw_slot = log->afi & 0x7; + next_fw_slot = (log->afi & 0x70) >> 4; + if (!cur_fw_slot || (next_fw_slot && (cur_fw_slot != next_fw_slot))) { + dev_info(ctrl->device, + "Firmware is activated after next Controller Level Reset\n"); + goto out_free_log; + } + + memcpy(ctrl->subsys->firmware_rev, &log->frs[cur_fw_slot - 1], + sizeof(ctrl->subsys->firmware_rev)); + +out_free_log: kfree(log); } @@ -4009,14 +4718,14 @@ static void nvme_fw_act_work(struct work_struct *work) struct nvme_ctrl, fw_act_work); unsigned long fw_act_timeout; + nvme_auth_stop(ctrl); + if (ctrl->mtfa) - fw_act_timeout = jiffies + - msecs_to_jiffies(ctrl->mtfa * 100); + fw_act_timeout = jiffies + msecs_to_jiffies(ctrl->mtfa * 100); else - fw_act_timeout = jiffies + - msecs_to_jiffies(admin_timeout * 1000); + fw_act_timeout = jiffies + secs_to_jiffies(admin_timeout); - nvme_stop_queues(ctrl); + nvme_quiesce_io_queues(ctrl); while (nvme_ctrl_pp_status(ctrl)) { if (time_after(jiffies, fw_act_timeout)) { dev_warn(ctrl->device, @@ -4027,19 +4736,31 @@ static void nvme_fw_act_work(struct work_struct *work) msleep(100); } - if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_LIVE)) + if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_CONNECTING) || + !nvme_change_ctrl_state(ctrl, NVME_CTRL_LIVE)) return; - nvme_start_queues(ctrl); + nvme_unquiesce_io_queues(ctrl); /* read FW slot information to clear the AER */ nvme_get_fw_slot_info(ctrl); + + queue_work(nvme_wq, &ctrl->async_event_work); +} + +static u32 nvme_aer_type(u32 result) +{ + return result & 0x7; } -static void nvme_handle_aen_notice(struct nvme_ctrl *ctrl, u32 result) +static u32 nvme_aer_subtype(u32 result) { - u32 aer_notice_type = (result & 0xff00) >> 8; + return (result & 0xff00) >> 8; +} - trace_nvme_async_event(ctrl, aer_notice_type); +static bool nvme_handle_aen_notice(struct nvme_ctrl *ctrl, u32 result) +{ + u32 aer_notice_type = nvme_aer_subtype(result); + bool requeue = true; switch (aer_notice_type) { case NVME_AER_NOTICE_NS_CHANGED: @@ -4052,8 +4773,10 @@ static void nvme_handle_aen_notice(struct nvme_ctrl *ctrl, u32 result) * recovery actions from interfering with the controller's * firmware activation. */ - if (nvme_change_ctrl_state(ctrl, NVME_CTRL_RESETTING)) + if (nvme_change_ctrl_state(ctrl, NVME_CTRL_RESETTING)) { + requeue = false; queue_work(nvme_wq, &ctrl->fw_act_work); + } break; #ifdef CONFIG_NVME_MULTIPATH case NVME_AER_NOTICE_ANA: @@ -4068,61 +4791,232 @@ static void nvme_handle_aen_notice(struct nvme_ctrl *ctrl, u32 result) default: dev_warn(ctrl->device, "async event result %08x\n", result); } + return requeue; +} + +static void nvme_handle_aer_persistent_error(struct nvme_ctrl *ctrl) +{ + dev_warn(ctrl->device, + "resetting controller due to persistent internal error\n"); + nvme_reset_ctrl(ctrl); } void nvme_complete_async_event(struct nvme_ctrl *ctrl, __le16 status, volatile union nvme_result *res) { u32 result = le32_to_cpu(res->u32); - u32 aer_type = result & 0x07; + u32 aer_type = nvme_aer_type(result); + u32 aer_subtype = nvme_aer_subtype(result); + bool requeue = true; if (le16_to_cpu(status) >> 1 != NVME_SC_SUCCESS) return; + trace_nvme_async_event(ctrl, result); switch (aer_type) { case NVME_AER_NOTICE: - nvme_handle_aen_notice(ctrl, result); + requeue = nvme_handle_aen_notice(ctrl, result); break; case NVME_AER_ERROR: + /* + * For a persistent internal error, don't run async_event_work + * to submit a new AER. The controller reset will do it. + */ + if (aer_subtype == NVME_AER_ERROR_PERSIST_INT_ERR) { + nvme_handle_aer_persistent_error(ctrl); + return; + } + fallthrough; case NVME_AER_SMART: case NVME_AER_CSS: case NVME_AER_VS: - trace_nvme_async_event(ctrl, aer_type); ctrl->aen_result = result; break; default: break; } - queue_work(nvme_wq, &ctrl->async_event_work); + + if (requeue) + queue_work(nvme_wq, &ctrl->async_event_work); } EXPORT_SYMBOL_GPL(nvme_complete_async_event); +int nvme_alloc_admin_tag_set(struct nvme_ctrl *ctrl, struct blk_mq_tag_set *set, + const struct blk_mq_ops *ops, unsigned int cmd_size) +{ + struct queue_limits lim = {}; + int ret; + + memset(set, 0, sizeof(*set)); + set->ops = ops; + set->queue_depth = NVME_AQ_MQ_TAG_DEPTH; + if (ctrl->ops->flags & NVME_F_FABRICS) + /* Reserved for fabric connect and keep alive */ + set->reserved_tags = 2; + set->numa_node = ctrl->numa_node; + if (ctrl->ops->flags & NVME_F_BLOCKING) + set->flags |= BLK_MQ_F_BLOCKING; + set->cmd_size = cmd_size; + set->driver_data = ctrl; + set->nr_hw_queues = 1; + set->timeout = NVME_ADMIN_TIMEOUT; + ret = blk_mq_alloc_tag_set(set); + if (ret) + return ret; + + ctrl->admin_q = blk_mq_alloc_queue(set, &lim, NULL); + if (IS_ERR(ctrl->admin_q)) { + ret = PTR_ERR(ctrl->admin_q); + goto out_free_tagset; + } + + if (ctrl->ops->flags & NVME_F_FABRICS) { + ctrl->fabrics_q = blk_mq_alloc_queue(set, NULL, NULL); + if (IS_ERR(ctrl->fabrics_q)) { + ret = PTR_ERR(ctrl->fabrics_q); + goto out_cleanup_admin_q; + } + } + + ctrl->admin_tagset = set; + return 0; + +out_cleanup_admin_q: + blk_mq_destroy_queue(ctrl->admin_q); + blk_put_queue(ctrl->admin_q); +out_free_tagset: + blk_mq_free_tag_set(set); + ctrl->admin_q = NULL; + ctrl->fabrics_q = NULL; + return ret; +} +EXPORT_SYMBOL_GPL(nvme_alloc_admin_tag_set); + +void nvme_remove_admin_tag_set(struct nvme_ctrl *ctrl) +{ + /* + * As we're about to destroy the queue and free tagset + * we can not have keep-alive work running. + */ + nvme_stop_keep_alive(ctrl); + blk_mq_destroy_queue(ctrl->admin_q); + if (ctrl->ops->flags & NVME_F_FABRICS) { + blk_mq_destroy_queue(ctrl->fabrics_q); + blk_put_queue(ctrl->fabrics_q); + } + blk_mq_free_tag_set(ctrl->admin_tagset); +} +EXPORT_SYMBOL_GPL(nvme_remove_admin_tag_set); + +int nvme_alloc_io_tag_set(struct nvme_ctrl *ctrl, struct blk_mq_tag_set *set, + const struct blk_mq_ops *ops, unsigned int nr_maps, + unsigned int cmd_size) +{ + int ret; + + memset(set, 0, sizeof(*set)); + set->ops = ops; + set->queue_depth = min_t(unsigned, ctrl->sqsize, BLK_MQ_MAX_DEPTH - 1); + /* + * Some Apple controllers requires tags to be unique across admin and + * the (only) I/O queue, so reserve the first 32 tags of the I/O queue. + */ + if (ctrl->quirks & NVME_QUIRK_SHARED_TAGS) + set->reserved_tags = NVME_AQ_DEPTH; + else if (ctrl->ops->flags & NVME_F_FABRICS) + /* Reserved for fabric connect */ + set->reserved_tags = 1; + set->numa_node = ctrl->numa_node; + if (ctrl->ops->flags & NVME_F_BLOCKING) + set->flags |= BLK_MQ_F_BLOCKING; + set->cmd_size = cmd_size; + set->driver_data = ctrl; + set->nr_hw_queues = ctrl->queue_count - 1; + set->timeout = NVME_IO_TIMEOUT; + set->nr_maps = nr_maps; + ret = blk_mq_alloc_tag_set(set); + if (ret) + return ret; + + if (ctrl->ops->flags & NVME_F_FABRICS) { + struct queue_limits lim = { + .features = BLK_FEAT_SKIP_TAGSET_QUIESCE, + }; + + ctrl->connect_q = blk_mq_alloc_queue(set, &lim, NULL); + if (IS_ERR(ctrl->connect_q)) { + ret = PTR_ERR(ctrl->connect_q); + goto out_free_tag_set; + } + } + + ctrl->tagset = set; + return 0; + +out_free_tag_set: + blk_mq_free_tag_set(set); + ctrl->connect_q = NULL; + return ret; +} +EXPORT_SYMBOL_GPL(nvme_alloc_io_tag_set); + +void nvme_remove_io_tag_set(struct nvme_ctrl *ctrl) +{ + if (ctrl->ops->flags & NVME_F_FABRICS) { + blk_mq_destroy_queue(ctrl->connect_q); + blk_put_queue(ctrl->connect_q); + } + blk_mq_free_tag_set(ctrl->tagset); +} +EXPORT_SYMBOL_GPL(nvme_remove_io_tag_set); + void nvme_stop_ctrl(struct nvme_ctrl *ctrl) { nvme_mpath_stop(ctrl); - nvme_stop_keep_alive(ctrl); + nvme_auth_stop(ctrl); + nvme_stop_failfast_work(ctrl); flush_work(&ctrl->async_event_work); cancel_work_sync(&ctrl->fw_act_work); + if (ctrl->ops->stop_ctrl) + ctrl->ops->stop_ctrl(ctrl); } EXPORT_SYMBOL_GPL(nvme_stop_ctrl); void nvme_start_ctrl(struct nvme_ctrl *ctrl) { - if (ctrl->kato) - nvme_start_keep_alive(ctrl); - nvme_enable_aen(ctrl); + /* + * persistent discovery controllers need to send indication to userspace + * to re-read the discovery log page to learn about possible changes + * that were missed. We identify persistent discovery controllers by + * checking that they started once before, hence are reconnecting back. + */ + if (test_bit(NVME_CTRL_STARTED_ONCE, &ctrl->flags) && + nvme_discovery_ctrl(ctrl)) { + if (!ctrl->kato) { + nvme_stop_keep_alive(ctrl); + ctrl->kato = NVME_DEFAULT_KATO; + nvme_start_keep_alive(ctrl); + } + nvme_change_uevent(ctrl, "NVME_EVENT=rediscover"); + } + if (ctrl->queue_count > 1) { nvme_queue_scan(ctrl); - nvme_start_queues(ctrl); + nvme_unquiesce_io_queues(ctrl); + nvme_mpath_update(ctrl); } - ctrl->created = true; + + nvme_change_uevent(ctrl, "NVME_EVENT=connected"); + set_bit(NVME_CTRL_STARTED_ONCE, &ctrl->flags); } EXPORT_SYMBOL_GPL(nvme_start_ctrl); void nvme_uninit_ctrl(struct nvme_ctrl *ctrl) { + nvme_stop_keep_alive(ctrl); + nvme_hwmon_exit(ctrl); nvme_fault_inject_fini(&ctrl->fault_inject); dev_pm_qos_hide_latency_tolerance(ctrl->device); cdev_device_del(&ctrl->cdev, ctrl->device); @@ -4130,18 +5024,36 @@ void nvme_uninit_ctrl(struct nvme_ctrl *ctrl) } EXPORT_SYMBOL_GPL(nvme_uninit_ctrl); +static void nvme_free_cels(struct nvme_ctrl *ctrl) +{ + struct nvme_effects_log *cel; + unsigned long i; + + xa_for_each(&ctrl->cels, i, cel) { + xa_erase(&ctrl->cels, i); + kfree(cel); + } + + xa_destroy(&ctrl->cels); +} + static void nvme_free_ctrl(struct device *dev) { struct nvme_ctrl *ctrl = container_of(dev, struct nvme_ctrl, ctrl_device); struct nvme_subsystem *subsys = ctrl->subsys; - if (subsys && ctrl->instance != subsys->instance) - ida_simple_remove(&nvme_instance_ida, ctrl->instance); - - kfree(ctrl->effects); + if (ctrl->admin_q) + blk_put_queue(ctrl->admin_q); + if (!subsys || ctrl->instance != subsys->instance) + ida_free(&nvme_instance_ida, ctrl->instance); + nvme_free_cels(ctrl); nvme_mpath_uninit(ctrl); + cleanup_srcu_struct(&ctrl->srcu); + nvme_auth_stop(ctrl); + nvme_auth_free(ctrl); __free_page(ctrl->discard_page); + free_opal_dev(ctrl->opal_dev); if (subsys) { mutex_lock(&nvme_subsystems_lock); @@ -4160,20 +5072,32 @@ static void nvme_free_ctrl(struct device *dev) * Initialize a NVMe controller structures. This needs to be called during * earliest initialization so that we have the initialized structured around * during probing. + * + * On success, the caller must use the nvme_put_ctrl() to release this when + * needed, which also invokes the ops->free_ctrl() callback. */ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev, const struct nvme_ctrl_ops *ops, unsigned long quirks) { int ret; - ctrl->state = NVME_CTRL_NEW; + WRITE_ONCE(ctrl->state, NVME_CTRL_NEW); + ctrl->passthru_err_log_enabled = false; + clear_bit(NVME_CTRL_FAILFAST_EXPIRED, &ctrl->flags); spin_lock_init(&ctrl->lock); + mutex_init(&ctrl->namespaces_lock); + + ret = init_srcu_struct(&ctrl->srcu); + if (ret) + return ret; + mutex_init(&ctrl->scan_lock); INIT_LIST_HEAD(&ctrl->namespaces); - init_rwsem(&ctrl->namespaces_rwsem); + xa_init(&ctrl->cels); ctrl->dev = dev; ctrl->ops = ops; ctrl->quirks = quirks; + ctrl->numa_node = NUMA_NO_NODE; INIT_WORK(&ctrl->scan_work, nvme_scan_work); INIT_WORK(&ctrl->async_event_work, nvme_async_event_work); INIT_WORK(&ctrl->fw_act_work, nvme_fw_act_work); @@ -4181,8 +5105,10 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev, init_waitqueue_head(&ctrl->state_wq); INIT_DELAYED_WORK(&ctrl->ka_work, nvme_keep_alive_work); + INIT_DELAYED_WORK(&ctrl->failfast_work, nvme_failfast_work); memset(&ctrl->ka_cmd, 0, sizeof(ctrl->ka_cmd)); ctrl->ka_cmd.common.opcode = nvme_admin_keep_alive; + ctrl->ka_last_check_time = jiffies; BUILD_BUG_ON(NVME_DSM_MAX_RANGES * sizeof(struct nvme_dsm_range) > PAGE_SIZE); @@ -4192,29 +5118,59 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev, goto out; } - ret = ida_simple_get(&nvme_instance_ida, 0, 0, GFP_KERNEL); + ret = ida_alloc(&nvme_instance_ida, GFP_KERNEL); if (ret < 0) goto out; ctrl->instance = ret; + ret = nvme_auth_init_ctrl(ctrl); + if (ret) + goto out_release_instance; + + nvme_mpath_init_ctrl(ctrl); + device_initialize(&ctrl->ctrl_device); ctrl->device = &ctrl->ctrl_device; - ctrl->device->devt = MKDEV(MAJOR(nvme_chr_devt), ctrl->instance); - ctrl->device->class = nvme_class; + ctrl->device->devt = MKDEV(MAJOR(nvme_ctrl_base_chr_devt), + ctrl->instance); + ctrl->device->class = &nvme_class; ctrl->device->parent = ctrl->dev; - ctrl->device->groups = nvme_dev_attr_groups; + if (ops->dev_attr_groups) + ctrl->device->groups = ops->dev_attr_groups; + else + ctrl->device->groups = nvme_dev_attr_groups; ctrl->device->release = nvme_free_ctrl; dev_set_drvdata(ctrl->device, ctrl); + + return ret; + +out_release_instance: + ida_free(&nvme_instance_ida, ctrl->instance); +out: + if (ctrl->discard_page) + __free_page(ctrl->discard_page); + cleanup_srcu_struct(&ctrl->srcu); + return ret; +} +EXPORT_SYMBOL_GPL(nvme_init_ctrl); + +/* + * On success, returns with an elevated controller reference and caller must + * use nvme_uninit_ctrl() to properly free resources associated with the ctrl. + */ +int nvme_add_ctrl(struct nvme_ctrl *ctrl) +{ + int ret; + ret = dev_set_name(ctrl->device, "nvme%d", ctrl->instance); if (ret) - goto out_release_instance; + return ret; - nvme_get_ctrl(ctrl); cdev_init(&ctrl->cdev, &nvme_dev_fops); - ctrl->cdev.owner = ops->module; + ctrl->cdev.owner = ctrl->ops->module; ret = cdev_device_add(&ctrl->cdev, ctrl->device); if (ret) - goto out_free_name; + return ret; /* * Initialize latency tolerance controls. The sysfs files won't @@ -4225,128 +5181,154 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev, min(default_ps_max_latency_us, (unsigned long)S32_MAX)); nvme_fault_inject_init(&ctrl->fault_inject, dev_name(ctrl->device)); + nvme_get_ctrl(ctrl); return 0; -out_free_name: - nvme_put_ctrl(ctrl); - kfree_const(ctrl->device->kobj.name); -out_release_instance: - ida_simple_remove(&nvme_instance_ida, ctrl->instance); -out: - if (ctrl->discard_page) - __free_page(ctrl->discard_page); - return ret; } -EXPORT_SYMBOL_GPL(nvme_init_ctrl); +EXPORT_SYMBOL_GPL(nvme_add_ctrl); -/** - * nvme_kill_queues(): Ends all namespace queues - * @ctrl: the dead controller that needs to end - * - * Call this function when the driver determines it is unable to get the - * controller in a state capable of servicing IO. - */ -void nvme_kill_queues(struct nvme_ctrl *ctrl) +/* let I/O to all namespaces fail in preparation for surprise removal */ +void nvme_mark_namespaces_dead(struct nvme_ctrl *ctrl) { struct nvme_ns *ns; + int srcu_idx; - down_read(&ctrl->namespaces_rwsem); - - /* Forcibly unquiesce queues to avoid blocking dispatch */ - if (ctrl->admin_q && !blk_queue_dying(ctrl->admin_q)) - blk_mq_unquiesce_queue(ctrl->admin_q); - - list_for_each_entry(ns, &ctrl->namespaces, list) - nvme_set_queue_dying(ns); - - up_read(&ctrl->namespaces_rwsem); + srcu_idx = srcu_read_lock(&ctrl->srcu); + list_for_each_entry_srcu(ns, &ctrl->namespaces, list, + srcu_read_lock_held(&ctrl->srcu)) + blk_mark_disk_dead(ns->disk); + srcu_read_unlock(&ctrl->srcu, srcu_idx); } -EXPORT_SYMBOL_GPL(nvme_kill_queues); +EXPORT_SYMBOL_GPL(nvme_mark_namespaces_dead); void nvme_unfreeze(struct nvme_ctrl *ctrl) { struct nvme_ns *ns; + int srcu_idx; - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) - blk_mq_unfreeze_queue(ns->queue); - up_read(&ctrl->namespaces_rwsem); + srcu_idx = srcu_read_lock(&ctrl->srcu); + list_for_each_entry_srcu(ns, &ctrl->namespaces, list, + srcu_read_lock_held(&ctrl->srcu)) + blk_mq_unfreeze_queue_non_owner(ns->queue); + srcu_read_unlock(&ctrl->srcu, srcu_idx); + clear_bit(NVME_CTRL_FROZEN, &ctrl->flags); } EXPORT_SYMBOL_GPL(nvme_unfreeze); -void nvme_wait_freeze_timeout(struct nvme_ctrl *ctrl, long timeout) +int nvme_wait_freeze_timeout(struct nvme_ctrl *ctrl, long timeout) { struct nvme_ns *ns; + int srcu_idx; - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) { + srcu_idx = srcu_read_lock(&ctrl->srcu); + list_for_each_entry_srcu(ns, &ctrl->namespaces, list, + srcu_read_lock_held(&ctrl->srcu)) { timeout = blk_mq_freeze_queue_wait_timeout(ns->queue, timeout); if (timeout <= 0) break; } - up_read(&ctrl->namespaces_rwsem); + srcu_read_unlock(&ctrl->srcu, srcu_idx); + return timeout; } EXPORT_SYMBOL_GPL(nvme_wait_freeze_timeout); void nvme_wait_freeze(struct nvme_ctrl *ctrl) { struct nvme_ns *ns; + int srcu_idx; - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) + srcu_idx = srcu_read_lock(&ctrl->srcu); + list_for_each_entry_srcu(ns, &ctrl->namespaces, list, + srcu_read_lock_held(&ctrl->srcu)) blk_mq_freeze_queue_wait(ns->queue); - up_read(&ctrl->namespaces_rwsem); + srcu_read_unlock(&ctrl->srcu, srcu_idx); } EXPORT_SYMBOL_GPL(nvme_wait_freeze); void nvme_start_freeze(struct nvme_ctrl *ctrl) { struct nvme_ns *ns; + int srcu_idx; - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) - blk_freeze_queue_start(ns->queue); - up_read(&ctrl->namespaces_rwsem); + set_bit(NVME_CTRL_FROZEN, &ctrl->flags); + srcu_idx = srcu_read_lock(&ctrl->srcu); + list_for_each_entry_srcu(ns, &ctrl->namespaces, list, + srcu_read_lock_held(&ctrl->srcu)) + /* + * Typical non_owner use case is from pci driver, in which + * start_freeze is called from timeout work function, but + * unfreeze is done in reset work context + */ + blk_freeze_queue_start_non_owner(ns->queue); + srcu_read_unlock(&ctrl->srcu, srcu_idx); } EXPORT_SYMBOL_GPL(nvme_start_freeze); -void nvme_stop_queues(struct nvme_ctrl *ctrl) +void nvme_quiesce_io_queues(struct nvme_ctrl *ctrl) { - struct nvme_ns *ns; - - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) - blk_mq_quiesce_queue(ns->queue); - up_read(&ctrl->namespaces_rwsem); + if (!ctrl->tagset) + return; + if (!test_and_set_bit(NVME_CTRL_STOPPED, &ctrl->flags)) + blk_mq_quiesce_tagset(ctrl->tagset); + else + blk_mq_wait_quiesce_done(ctrl->tagset); } -EXPORT_SYMBOL_GPL(nvme_stop_queues); +EXPORT_SYMBOL_GPL(nvme_quiesce_io_queues); -void nvme_start_queues(struct nvme_ctrl *ctrl) +void nvme_unquiesce_io_queues(struct nvme_ctrl *ctrl) { - struct nvme_ns *ns; + if (!ctrl->tagset) + return; + if (test_and_clear_bit(NVME_CTRL_STOPPED, &ctrl->flags)) + blk_mq_unquiesce_tagset(ctrl->tagset); +} +EXPORT_SYMBOL_GPL(nvme_unquiesce_io_queues); - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) - blk_mq_unquiesce_queue(ns->queue); - up_read(&ctrl->namespaces_rwsem); +void nvme_quiesce_admin_queue(struct nvme_ctrl *ctrl) +{ + if (!test_and_set_bit(NVME_CTRL_ADMIN_Q_STOPPED, &ctrl->flags)) + blk_mq_quiesce_queue(ctrl->admin_q); + else + blk_mq_wait_quiesce_done(ctrl->admin_q->tag_set); } -EXPORT_SYMBOL_GPL(nvme_start_queues); +EXPORT_SYMBOL_GPL(nvme_quiesce_admin_queue); +void nvme_unquiesce_admin_queue(struct nvme_ctrl *ctrl) +{ + if (test_and_clear_bit(NVME_CTRL_ADMIN_Q_STOPPED, &ctrl->flags)) + blk_mq_unquiesce_queue(ctrl->admin_q); +} +EXPORT_SYMBOL_GPL(nvme_unquiesce_admin_queue); -void nvme_sync_queues(struct nvme_ctrl *ctrl) +void nvme_sync_io_queues(struct nvme_ctrl *ctrl) { struct nvme_ns *ns; + int srcu_idx; - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) + srcu_idx = srcu_read_lock(&ctrl->srcu); + list_for_each_entry_srcu(ns, &ctrl->namespaces, list, + srcu_read_lock_held(&ctrl->srcu)) blk_sync_queue(ns->queue); - up_read(&ctrl->namespaces_rwsem); + srcu_read_unlock(&ctrl->srcu, srcu_idx); +} +EXPORT_SYMBOL_GPL(nvme_sync_io_queues); +void nvme_sync_queues(struct nvme_ctrl *ctrl) +{ + nvme_sync_io_queues(ctrl); if (ctrl->admin_q) blk_sync_queue(ctrl->admin_q); } EXPORT_SYMBOL_GPL(nvme_sync_queues); +struct nvme_ctrl *nvme_ctrl_from_file(struct file *file) +{ + if (file->f_op != &nvme_dev_fops) + return NULL; + return file->private_data; +} +EXPORT_SYMBOL_NS_GPL(nvme_ctrl_from_file, "NVME_TARGET_PASSTHRU"); + /* * Check we didn't inadvertently grow the command structure sizes: */ @@ -4365,56 +5347,78 @@ static inline void _nvme_check_size(void) BUILD_BUG_ON(sizeof(struct nvme_command) != 64); BUILD_BUG_ON(sizeof(struct nvme_id_ctrl) != NVME_IDENTIFY_DATA_SIZE); BUILD_BUG_ON(sizeof(struct nvme_id_ns) != NVME_IDENTIFY_DATA_SIZE); + BUILD_BUG_ON(sizeof(struct nvme_id_ns_cs_indep) != + NVME_IDENTIFY_DATA_SIZE); + BUILD_BUG_ON(sizeof(struct nvme_id_ns_zns) != NVME_IDENTIFY_DATA_SIZE); + BUILD_BUG_ON(sizeof(struct nvme_id_ns_nvm) != NVME_IDENTIFY_DATA_SIZE); + BUILD_BUG_ON(sizeof(struct nvme_id_ctrl_zns) != NVME_IDENTIFY_DATA_SIZE); + BUILD_BUG_ON(sizeof(struct nvme_id_ctrl_nvm) != NVME_IDENTIFY_DATA_SIZE); BUILD_BUG_ON(sizeof(struct nvme_lba_range_type) != 64); BUILD_BUG_ON(sizeof(struct nvme_smart_log) != 512); + BUILD_BUG_ON(sizeof(struct nvme_endurance_group_log) != 512); + BUILD_BUG_ON(sizeof(struct nvme_rotational_media_log) != 512); BUILD_BUG_ON(sizeof(struct nvme_dbbuf) != 64); BUILD_BUG_ON(sizeof(struct nvme_directive_cmd) != 64); + BUILD_BUG_ON(sizeof(struct nvme_feat_host_behavior) != 512); } static int __init nvme_core_init(void) { + unsigned int wq_flags = WQ_UNBOUND | WQ_MEM_RECLAIM | WQ_SYSFS; int result = -ENOMEM; _nvme_check_size(); - nvme_wq = alloc_workqueue("nvme-wq", - WQ_UNBOUND | WQ_MEM_RECLAIM | WQ_SYSFS, 0); + nvme_wq = alloc_workqueue("nvme-wq", wq_flags, 0); if (!nvme_wq) goto out; - nvme_reset_wq = alloc_workqueue("nvme-reset-wq", - WQ_UNBOUND | WQ_MEM_RECLAIM | WQ_SYSFS, 0); + nvme_reset_wq = alloc_workqueue("nvme-reset-wq", wq_flags, 0); if (!nvme_reset_wq) goto destroy_wq; - nvme_delete_wq = alloc_workqueue("nvme-delete-wq", - WQ_UNBOUND | WQ_MEM_RECLAIM | WQ_SYSFS, 0); + nvme_delete_wq = alloc_workqueue("nvme-delete-wq", wq_flags, 0); if (!nvme_delete_wq) goto destroy_reset_wq; - result = alloc_chrdev_region(&nvme_chr_devt, 0, NVME_MINORS, "nvme"); + result = alloc_chrdev_region(&nvme_ctrl_base_chr_devt, 0, + NVME_MINORS, "nvme"); if (result < 0) goto destroy_delete_wq; - nvme_class = class_create(THIS_MODULE, "nvme"); - if (IS_ERR(nvme_class)) { - result = PTR_ERR(nvme_class); + result = class_register(&nvme_class); + if (result) goto unregister_chrdev; - } - nvme_class->dev_uevent = nvme_class_uevent; - nvme_subsys_class = class_create(THIS_MODULE, "nvme-subsystem"); - if (IS_ERR(nvme_subsys_class)) { - result = PTR_ERR(nvme_subsys_class); + result = class_register(&nvme_subsys_class); + if (result) goto destroy_class; - } + + result = alloc_chrdev_region(&nvme_ns_chr_devt, 0, NVME_MINORS, + "nvme-generic"); + if (result < 0) + goto destroy_subsys_class; + + result = class_register(&nvme_ns_chr_class); + if (result) + goto unregister_generic_ns; + + result = nvme_init_auth(); + if (result) + goto destroy_ns_chr; return 0; +destroy_ns_chr: + class_unregister(&nvme_ns_chr_class); +unregister_generic_ns: + unregister_chrdev_region(nvme_ns_chr_devt, NVME_MINORS); +destroy_subsys_class: + class_unregister(&nvme_subsys_class); destroy_class: - class_destroy(nvme_class); + class_unregister(&nvme_class); unregister_chrdev: - unregister_chrdev_region(nvme_chr_devt, NVME_MINORS); + unregister_chrdev_region(nvme_ctrl_base_chr_devt, NVME_MINORS); destroy_delete_wq: destroy_workqueue(nvme_delete_wq); destroy_reset_wq: @@ -4427,16 +5431,21 @@ out: static void __exit nvme_core_exit(void) { - class_destroy(nvme_subsys_class); - class_destroy(nvme_class); - unregister_chrdev_region(nvme_chr_devt, NVME_MINORS); + nvme_exit_auth(); + class_unregister(&nvme_ns_chr_class); + class_unregister(&nvme_subsys_class); + class_unregister(&nvme_class); + unregister_chrdev_region(nvme_ns_chr_devt, NVME_MINORS); + unregister_chrdev_region(nvme_ctrl_base_chr_devt, NVME_MINORS); destroy_workqueue(nvme_delete_wq); destroy_workqueue(nvme_reset_wq); destroy_workqueue(nvme_wq); + ida_destroy(&nvme_ns_chr_minor_ida); ida_destroy(&nvme_instance_ida); } MODULE_LICENSE("GPL"); MODULE_VERSION("1.0"); +MODULE_DESCRIPTION("NVMe host core framework"); module_init(nvme_core_init); module_exit(nvme_core_exit); |