summaryrefslogtreecommitdiff
path: root/drivers/nvme/host
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/nvme/host')
-rw-r--r--drivers/nvme/host/Kconfig1
-rw-r--r--drivers/nvme/host/apple.c7
-rw-r--r--drivers/nvme/host/constants.c2
-rw-r--r--drivers/nvme/host/core.c87
-rw-r--r--drivers/nvme/host/fc.c68
-rw-r--r--drivers/nvme/host/ioctl.c44
-rw-r--r--drivers/nvme/host/multipath.c2
-rw-r--r--drivers/nvme/host/nvme.h7
-rw-r--r--drivers/nvme/host/pci.c245
-rw-r--r--drivers/nvme/host/pr.c2
-rw-r--r--drivers/nvme/host/sysfs.c2
-rw-r--r--drivers/nvme/host/tcp.c217
12 files changed, 514 insertions, 170 deletions
diff --git a/drivers/nvme/host/Kconfig b/drivers/nvme/host/Kconfig
index 486afe598184..09ed1f61c9a8 100644
--- a/drivers/nvme/host/Kconfig
+++ b/drivers/nvme/host/Kconfig
@@ -97,6 +97,7 @@ config NVME_TCP_TLS
depends on NVME_TCP
select NET_HANDSHAKE
select KEYS
+ select TLS
help
Enables TLS encryption for NVMe TCP using the netlink handshake API.
diff --git a/drivers/nvme/host/apple.c b/drivers/nvme/host/apple.c
index b1387dc459a3..328f5a103628 100644
--- a/drivers/nvme/host/apple.c
+++ b/drivers/nvme/host/apple.c
@@ -599,7 +599,8 @@ static inline void apple_nvme_handle_cqe(struct apple_nvme_queue *q,
}
if (!nvme_try_complete_req(req, cqe->status, cqe->result) &&
- !blk_mq_add_to_batch(req, iob, nvme_req(req)->status,
+ !blk_mq_add_to_batch(req, iob,
+ nvme_req(req)->status != NVME_SC_SUCCESS,
apple_nvme_complete_batch))
apple_nvme_complete_rq(req);
}
@@ -649,7 +650,7 @@ static bool apple_nvme_handle_cq(struct apple_nvme_queue *q, bool force)
found = apple_nvme_poll_cq(q, &iob);
- if (!rq_list_empty(iob.req_list))
+ if (!rq_list_empty(&iob.req_list))
apple_nvme_complete_batch(&iob);
return found;
@@ -1518,6 +1519,7 @@ static struct apple_nvme *apple_nvme_alloc(struct platform_device *pdev)
return anv;
put_dev:
+ apple_nvme_detach_genpd(anv);
put_device(anv->dev);
return ERR_PTR(ret);
}
@@ -1551,6 +1553,7 @@ out_uninit_ctrl:
nvme_uninit_ctrl(&anv->ctrl);
out_put_ctrl:
nvme_put_ctrl(&anv->ctrl);
+ apple_nvme_detach_genpd(anv);
return ret;
}
diff --git a/drivers/nvme/host/constants.c b/drivers/nvme/host/constants.c
index 2b9e6cfaf2a8..1a0058be5821 100644
--- a/drivers/nvme/host/constants.c
+++ b/drivers/nvme/host/constants.c
@@ -145,7 +145,7 @@ static const char * const nvme_statuses[] = {
[NVME_SC_BAD_ATTRIBUTES] = "Conflicting Attributes",
[NVME_SC_INVALID_PI] = "Invalid Protection Information",
[NVME_SC_READ_ONLY] = "Attempted Write to Read Only Range",
- [NVME_SC_ONCS_NOT_SUPPORTED] = "ONCS Not Supported",
+ [NVME_SC_CMD_SIZE_LIM_EXCEEDED ] = "Command Size Limits Exceeded",
[NVME_SC_ZONE_BOUNDARY_ERROR] = "Zoned Boundary Error",
[NVME_SC_ZONE_FULL] = "Zone Is Full",
[NVME_SC_ZONE_READ_ONLY] = "Zone Is Read Only",
diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
index 249914b90dbf..9e223574db7f 100644
--- a/drivers/nvme/host/core.c
+++ b/drivers/nvme/host/core.c
@@ -284,7 +284,6 @@ 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:
@@ -376,12 +375,12 @@ static void nvme_log_err_passthru(struct request *req)
nr->status & NVME_SC_MASK, /* Status Code */
nr->status & NVME_STATUS_MORE ? "MORE " : "",
nr->status & NVME_STATUS_DNR ? "DNR " : "",
- nr->cmd->common.cdw10,
- nr->cmd->common.cdw11,
- nr->cmd->common.cdw12,
- nr->cmd->common.cdw13,
- nr->cmd->common.cdw14,
- nr->cmd->common.cdw14);
+ 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 {
@@ -429,6 +428,12 @@ static inline void nvme_end_req_zoned(struct request *req)
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)
@@ -439,12 +444,6 @@ void nvme_end_req(struct request *req)
{
blk_status_t status = nvme_error_status(nvme_req(req)->status);
- 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(req);
blk_mq_end_request(req, status);
}
@@ -562,8 +561,6 @@ bool nvme_change_ctrl_state(struct nvme_ctrl *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;
fallthrough;
@@ -760,6 +757,10 @@ blk_status_t nvme_fail_nonready_command(struct nvme_ctrl *ctrl,
!test_bit(NVME_CTRL_FAILFAST_EXPIRED, &ctrl->flags) &&
!blk_noretry_request(rq) && !(rq->cmd_flags & REQ_NVME_MPATH))
return BLK_STS_RESOURCE;
+
+ if (!(rq->rq_flags & RQF_DONTPREP))
+ nvme_clear_nvme_request(rq);
+
return nvme_host_path_error(rq);
}
EXPORT_SYMBOL_GPL(nvme_fail_nonready_command);
@@ -1691,7 +1692,13 @@ 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;
/*
@@ -3085,7 +3092,7 @@ int nvme_get_log(struct nvme_ctrl *ctrl, u32 nsid, u8 log_page, u8 lsp, u8 csi,
static int nvme_get_effects_log(struct nvme_ctrl *ctrl, u8 csi,
struct nvme_effects_log **log)
{
- struct nvme_effects_log *cel = xa_load(&ctrl->cels, csi);
+ struct nvme_effects_log *old, *cel = xa_load(&ctrl->cels, csi);
int ret;
if (cel)
@@ -3102,7 +3109,11 @@ static int nvme_get_effects_log(struct nvme_ctrl *ctrl, u8 csi,
return ret;
}
- xa_store(&ctrl->cels, csi, cel, GFP_KERNEL);
+ 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;
@@ -3164,6 +3175,25 @@ free_data:
return ret;
}
+static int nvme_init_effects_log(struct nvme_ctrl *ctrl,
+ u8 csi, struct nvme_effects_log **log)
+{
+ struct nvme_effects_log *effects, *old;
+
+ 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);
+ }
+
+ *log = effects;
+ return 0;
+}
+
static void nvme_init_known_nvm_effects(struct nvme_ctrl *ctrl)
{
struct nvme_effects_log *log = ctrl->effects;
@@ -3210,10 +3240,9 @@ static int nvme_init_effects(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id)
}
if (!ctrl->effects) {
- ctrl->effects = kzalloc(sizeof(*ctrl->effects), GFP_KERNEL);
- if (!ctrl->effects)
- return -ENOMEM;
- xa_store(&ctrl->cels, NVME_CSI_NVM, ctrl->effects, GFP_KERNEL);
+ ret = nvme_init_effects_log(ctrl, NVME_CSI_NVM, &ctrl->effects);
+ if (ret < 0)
+ return ret;
}
nvme_init_known_nvm_effects(ctrl);
@@ -3829,7 +3858,7 @@ static void nvme_ns_add_to_ctrl_list(struct nvme_ns *ns)
return;
}
}
- list_add(&ns->list, &ns->ctrl->namespaces);
+ list_add_rcu(&ns->list, &ns->ctrl->namespaces);
}
static void nvme_alloc_ns(struct nvme_ctrl *ctrl, struct nvme_ns_info *info)
@@ -4247,6 +4276,15 @@ static void nvme_scan_work(struct work_struct *work)
nvme_scan_ns_sequential(ctrl);
}
mutex_unlock(&ctrl->scan_lock);
+
+ /* 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
}
/*
@@ -4438,7 +4476,8 @@ 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_unquiesce_io_queues(ctrl);
diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c
index b81af7919e94..7c13a400071e 100644
--- a/drivers/nvme/host/fc.c
+++ b/drivers/nvme/host/fc.c
@@ -786,49 +786,8 @@ nvme_fc_ctrl_connectivity_loss(struct nvme_fc_ctrl *ctrl)
"NVME-FC{%d}: controller connectivity lost. Awaiting "
"Reconnect", ctrl->cnum);
- switch (nvme_ctrl_state(&ctrl->ctrl)) {
- case NVME_CTRL_NEW:
- case NVME_CTRL_LIVE:
- /*
- * Schedule a controller reset. The reset will terminate the
- * association and schedule the reconnect timer. Reconnects
- * will be attempted until either the ctlr_loss_tmo
- * (max_retries * connect_delay) expires or the remoteport's
- * dev_loss_tmo expires.
- */
- if (nvme_reset_ctrl(&ctrl->ctrl)) {
- dev_warn(ctrl->ctrl.device,
- "NVME-FC{%d}: Couldn't schedule reset.\n",
- ctrl->cnum);
- nvme_delete_ctrl(&ctrl->ctrl);
- }
- break;
-
- case NVME_CTRL_CONNECTING:
- /*
- * The association has already been terminated and the
- * controller is attempting reconnects. No need to do anything
- * futher. Reconnects will be attempted until either the
- * ctlr_loss_tmo (max_retries * connect_delay) expires or the
- * remoteport's dev_loss_tmo expires.
- */
- break;
-
- case NVME_CTRL_RESETTING:
- /*
- * Controller is already in the process of terminating the
- * association. No need to do anything further. The reconnect
- * step will kick in naturally after the association is
- * terminated.
- */
- break;
-
- case NVME_CTRL_DELETING:
- case NVME_CTRL_DELETING_NOIO:
- default:
- /* no action to take - let it delete */
- break;
- }
+ set_bit(ASSOC_FAILED, &ctrl->flags);
+ nvme_reset_ctrl(&ctrl->ctrl);
}
/**
@@ -2080,7 +2039,8 @@ done:
nvme_fc_complete_rq(rq);
check_error:
- if (terminate_assoc && ctrl->ctrl.state != NVME_CTRL_RESETTING)
+ if (terminate_assoc &&
+ nvme_ctrl_state(&ctrl->ctrl) != NVME_CTRL_RESETTING)
queue_work(nvme_reset_wq, &ctrl->ioerr_work);
}
@@ -2534,6 +2494,8 @@ __nvme_fc_abort_outstanding_ios(struct nvme_fc_ctrl *ctrl, bool start_queues)
static void
nvme_fc_error_recovery(struct nvme_fc_ctrl *ctrl, char *errmsg)
{
+ enum nvme_ctrl_state state = nvme_ctrl_state(&ctrl->ctrl);
+
/*
* if an error (io timeout, etc) while (re)connecting, the remote
* port requested terminating of the association (disconnect_ls)
@@ -2541,9 +2503,8 @@ nvme_fc_error_recovery(struct nvme_fc_ctrl *ctrl, char *errmsg)
* the controller. Abort any ios on the association and let the
* create_association error path resolve things.
*/
- if (ctrl->ctrl.state == NVME_CTRL_CONNECTING) {
+ if (state == NVME_CTRL_CONNECTING) {
__nvme_fc_abort_outstanding_ios(ctrl, true);
- set_bit(ASSOC_FAILED, &ctrl->flags);
dev_warn(ctrl->ctrl.device,
"NVME-FC{%d}: transport error during (re)connect\n",
ctrl->cnum);
@@ -2551,7 +2512,7 @@ nvme_fc_error_recovery(struct nvme_fc_ctrl *ctrl, char *errmsg)
}
/* Otherwise, only proceed if in LIVE state - e.g. on first error */
- if (ctrl->ctrl.state != NVME_CTRL_LIVE)
+ if (state != NVME_CTRL_LIVE)
return;
dev_warn(ctrl->ctrl.device,
@@ -3062,7 +3023,6 @@ nvme_fc_create_association(struct nvme_fc_ctrl *ctrl)
struct nvmefc_ls_rcv_op *disls = NULL;
unsigned long flags;
int ret;
- bool changed;
++ctrl->ctrl.nr_reconnects;
@@ -3173,12 +3133,13 @@ nvme_fc_create_association(struct nvme_fc_ctrl *ctrl)
if (ret)
goto out_term_aen_ops;
- changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE);
+ if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE)) {
+ ret = -EIO;
+ goto out_term_aen_ops;
+ }
ctrl->ctrl.nr_reconnects = 0;
-
- if (changed)
- nvme_start_ctrl(&ctrl->ctrl);
+ nvme_start_ctrl(&ctrl->ctrl);
return 0; /* Success */
@@ -3579,8 +3540,7 @@ nvme_fc_init_ctrl(struct device *dev, struct nvmf_ctrl_options *opts,
list_add_tail(&ctrl->ctrl_list, &rport->ctrl_list);
spin_unlock_irqrestore(&rport->lock, flags);
- if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_RESETTING) ||
- !nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING)) {
+ if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING)) {
dev_err(ctrl->ctrl.device,
"NVME-FC{%d}: failed to init ctrl state\n", ctrl->cnum);
goto fail_ctrl;
diff --git a/drivers/nvme/host/ioctl.c b/drivers/nvme/host/ioctl.c
index a96976b22fa7..64ae8af01d9a 100644
--- a/drivers/nvme/host/ioctl.c
+++ b/drivers/nvme/host/ioctl.c
@@ -120,19 +120,31 @@ static int nvme_map_user_request(struct request *req, u64 ubuffer,
struct nvme_ns *ns = q->queuedata;
struct block_device *bdev = ns ? ns->disk->part0 : NULL;
bool supports_metadata = bdev && blk_get_integrity(bdev->bd_disk);
+ struct nvme_ctrl *ctrl = nvme_req(req)->ctrl;
bool has_metadata = meta_buffer && meta_len;
struct bio *bio = NULL;
int ret;
- if (has_metadata && !supports_metadata)
- return -EINVAL;
+ if (!nvme_ctrl_sgl_supported(ctrl))
+ dev_warn_once(ctrl->device, "using unchecked data buffer\n");
+ if (has_metadata) {
+ if (!supports_metadata) {
+ ret = -EINVAL;
+ goto out;
+ }
+ if (!nvme_ctrl_meta_sgl_supported(ctrl))
+ dev_warn_once(ctrl->device,
+ "using unchecked metadata buffer\n");
+ }
if (ioucmd && (ioucmd->flags & IORING_URING_CMD_FIXED)) {
struct iov_iter iter;
/* fixedbufs is only for non-vectored io */
- if (WARN_ON_ONCE(flags & NVME_IOCTL_VEC))
- return -EINVAL;
+ if (flags & NVME_IOCTL_VEC) {
+ ret = -EINVAL;
+ goto out;
+ }
ret = io_uring_cmd_import_fixed(ubuffer, bufflen,
rq_data_dir(req), &iter, ioucmd);
if (ret < 0)
@@ -276,8 +288,7 @@ static bool nvme_validate_passthru_nsid(struct nvme_ctrl *ctrl,
{
if (ns && nsid != ns->head->ns_id) {
dev_err(ctrl->device,
- "%s: nsid (%u) in cmd does not match nsid (%u)"
- "of namespace\n",
+ "%s: nsid (%u) in cmd does not match nsid (%u) of namespace\n",
current->comm, nsid, ns->head->ns_id);
return false;
}
@@ -431,21 +442,14 @@ static enum rq_end_io_ret nvme_uring_cmd_end_io(struct request *req,
pdu->result = le64_to_cpu(nvme_req(req)->result.u64);
/*
- * For iopoll, complete it directly. Note that using the uring_cmd
- * helper for this is safe only because we check blk_rq_is_poll().
- * As that returns false if we're NOT on a polled queue, then it's
- * safe to use the polled completion helper.
- *
- * Otherwise, move the completion to task work.
+ * IOPOLL could potentially complete this request directly, but
+ * if multiple rings are polling on the same queue, then it's possible
+ * for one ring to find completions for another ring. Punting the
+ * completion via task_work will always direct it to the right
+ * location, rather than potentially complete requests for ringA
+ * under iopoll invocations from ringB.
*/
- if (blk_rq_is_poll(req)) {
- if (pdu->bio)
- blk_rq_unmap_user(pdu->bio);
- io_uring_cmd_iopoll_done(ioucmd, pdu->result, pdu->status);
- } else {
- io_uring_cmd_do_in_task_lazy(ioucmd, nvme_uring_task_cb);
- }
-
+ io_uring_cmd_do_in_task_lazy(ioucmd, nvme_uring_task_cb);
return RQ_END_IO_FREE;
}
diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c
index f25582e4d88b..561dd08022c0 100644
--- a/drivers/nvme/host/multipath.c
+++ b/drivers/nvme/host/multipath.c
@@ -427,7 +427,7 @@ static bool nvme_available_path(struct nvme_ns_head *head)
struct nvme_ns *ns;
if (!test_bit(NVME_NSHEAD_DISK_LIVE, &head->flags))
- return NULL;
+ return false;
list_for_each_entry_srcu(ns, &head->list, siblings,
srcu_read_lock_held(&head->srcu)) {
diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h
index 61bba5513de0..dcdce7d12e44 100644
--- a/drivers/nvme/host/nvme.h
+++ b/drivers/nvme/host/nvme.h
@@ -1130,6 +1130,13 @@ static inline bool nvme_ctrl_sgl_supported(struct nvme_ctrl *ctrl)
return ctrl->sgls & ((1 << 0) | (1 << 1));
}
+static inline bool nvme_ctrl_meta_sgl_supported(struct nvme_ctrl *ctrl)
+{
+ if (ctrl->ops->flags & NVME_F_FABRICS)
+ return true;
+ return ctrl->sgls & NVME_CTRL_SGLS_MSDS;
+}
+
#ifdef CONFIG_NVME_HOST_AUTH
int __init nvme_init_auth(void);
void __exit nvme_exit_auth(void);
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index 76b3f7b396c8..2bddc9f60fec 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -43,6 +43,7 @@
*/
#define NVME_MAX_KB_SZ 8192
#define NVME_MAX_SEGS 128
+#define NVME_MAX_META_SEGS 15
#define NVME_MAX_NR_ALLOCATIONS 5
static int use_threaded_interrupts;
@@ -143,6 +144,7 @@ struct nvme_dev {
bool hmb;
mempool_t *iod_mempool;
+ mempool_t *iod_meta_mempool;
/* shadow doorbell buffer support: */
__le32 *dbbuf_dbs;
@@ -238,6 +240,8 @@ struct nvme_iod {
dma_addr_t first_dma;
dma_addr_t meta_dma;
struct sg_table sgt;
+ struct sg_table meta_sgt;
+ union nvme_descriptor meta_list;
union nvme_descriptor list[NVME_MAX_NR_ALLOCATIONS];
};
@@ -386,7 +390,7 @@ static bool nvme_dbbuf_update_and_check_event(u16 value, __le32 *dbbuf_db,
* as it only leads to a small amount of wasted memory for the lifetime of
* the I/O.
*/
-static int nvme_pci_npages_prp(void)
+static __always_inline int nvme_pci_npages_prp(void)
{
unsigned max_bytes = (NVME_MAX_KB_SZ * 1024) + NVME_CTRL_PAGE_SIZE;
unsigned nprps = DIV_ROUND_UP(max_bytes, NVME_CTRL_PAGE_SIZE);
@@ -505,6 +509,15 @@ static void nvme_commit_rqs(struct blk_mq_hw_ctx *hctx)
spin_unlock(&nvmeq->sq_lock);
}
+static inline bool nvme_pci_metadata_use_sgls(struct nvme_dev *dev,
+ struct request *req)
+{
+ if (!nvme_ctrl_meta_sgl_supported(&dev->ctrl))
+ return false;
+ return req->nr_integrity_segments > 1 ||
+ nvme_req(req)->flags & NVME_REQ_USERCMD;
+}
+
static inline bool nvme_pci_use_sgls(struct nvme_dev *dev, struct request *req,
int nseg)
{
@@ -517,8 +530,10 @@ static inline bool nvme_pci_use_sgls(struct nvme_dev *dev, struct request *req,
return false;
if (!nvmeq->qid)
return false;
+ if (nvme_pci_metadata_use_sgls(dev, req))
+ return true;
if (!sgl_threshold || avg_seg_size < sgl_threshold)
- return false;
+ return nvme_req(req)->flags & NVME_REQ_USERCMD;
return true;
}
@@ -779,7 +794,8 @@ static blk_status_t nvme_map_data(struct nvme_dev *dev, struct request *req,
struct bio_vec bv = req_bvec(req);
if (!is_pci_p2pdma_page(bv.bv_page)) {
- if ((bv.bv_offset & (NVME_CTRL_PAGE_SIZE - 1)) +
+ if (!nvme_pci_metadata_use_sgls(dev, req) &&
+ (bv.bv_offset & (NVME_CTRL_PAGE_SIZE - 1)) +
bv.bv_len <= NVME_CTRL_PAGE_SIZE * 2)
return nvme_setup_prp_simple(dev, req,
&cmnd->rw, &bv);
@@ -823,11 +839,69 @@ out_free_sg:
return ret;
}
-static blk_status_t nvme_map_metadata(struct nvme_dev *dev, struct request *req,
- struct nvme_command *cmnd)
+static blk_status_t nvme_pci_setup_meta_sgls(struct nvme_dev *dev,
+ struct request *req)
+{
+ struct nvme_iod *iod = blk_mq_rq_to_pdu(req);
+ struct nvme_rw_command *cmnd = &iod->cmd.rw;
+ struct nvme_sgl_desc *sg_list;
+ struct scatterlist *sgl, *sg;
+ unsigned int entries;
+ dma_addr_t sgl_dma;
+ int rc, i;
+
+ iod->meta_sgt.sgl = mempool_alloc(dev->iod_meta_mempool, GFP_ATOMIC);
+ if (!iod->meta_sgt.sgl)
+ return BLK_STS_RESOURCE;
+
+ sg_init_table(iod->meta_sgt.sgl, req->nr_integrity_segments);
+ iod->meta_sgt.orig_nents = blk_rq_map_integrity_sg(req,
+ iod->meta_sgt.sgl);
+ if (!iod->meta_sgt.orig_nents)
+ goto out_free_sg;
+
+ rc = dma_map_sgtable(dev->dev, &iod->meta_sgt, rq_dma_dir(req),
+ DMA_ATTR_NO_WARN);
+ if (rc)
+ goto out_free_sg;
+
+ sg_list = dma_pool_alloc(dev->prp_small_pool, GFP_ATOMIC, &sgl_dma);
+ if (!sg_list)
+ goto out_unmap_sg;
+
+ entries = iod->meta_sgt.nents;
+ iod->meta_list.sg_list = sg_list;
+ iod->meta_dma = sgl_dma;
+
+ cmnd->flags = NVME_CMD_SGL_METASEG;
+ cmnd->metadata = cpu_to_le64(sgl_dma);
+
+ sgl = iod->meta_sgt.sgl;
+ if (entries == 1) {
+ nvme_pci_sgl_set_data(sg_list, sgl);
+ return BLK_STS_OK;
+ }
+
+ sgl_dma += sizeof(*sg_list);
+ nvme_pci_sgl_set_seg(sg_list, sgl_dma, entries);
+ for_each_sg(sgl, sg, entries, i)
+ nvme_pci_sgl_set_data(&sg_list[i + 1], sg);
+
+ return BLK_STS_OK;
+
+out_unmap_sg:
+ dma_unmap_sgtable(dev->dev, &iod->meta_sgt, rq_dma_dir(req), 0);
+out_free_sg:
+ mempool_free(iod->meta_sgt.sgl, dev->iod_meta_mempool);
+ return BLK_STS_RESOURCE;
+}
+
+static blk_status_t nvme_pci_setup_meta_mptr(struct nvme_dev *dev,
+ struct request *req)
{
struct nvme_iod *iod = blk_mq_rq_to_pdu(req);
struct bio_vec bv = rq_integrity_vec(req);
+ struct nvme_command *cmnd = &iod->cmd;
iod->meta_dma = dma_map_bvec(dev->dev, &bv, rq_dma_dir(req), 0);
if (dma_mapping_error(dev->dev, iod->meta_dma))
@@ -836,6 +910,13 @@ static blk_status_t nvme_map_metadata(struct nvme_dev *dev, struct request *req,
return BLK_STS_OK;
}
+static blk_status_t nvme_map_metadata(struct nvme_dev *dev, struct request *req)
+{
+ if (nvme_pci_metadata_use_sgls(dev, req))
+ return nvme_pci_setup_meta_sgls(dev, req);
+ return nvme_pci_setup_meta_mptr(dev, req);
+}
+
static blk_status_t nvme_prep_rq(struct nvme_dev *dev, struct request *req)
{
struct nvme_iod *iod = blk_mq_rq_to_pdu(req);
@@ -844,6 +925,7 @@ static blk_status_t nvme_prep_rq(struct nvme_dev *dev, struct request *req)
iod->aborted = false;
iod->nr_allocations = -1;
iod->sgt.nents = 0;
+ iod->meta_sgt.nents = 0;
ret = nvme_setup_cmd(req->q->queuedata, req);
if (ret)
@@ -856,7 +938,7 @@ static blk_status_t nvme_prep_rq(struct nvme_dev *dev, struct request *req)
}
if (blk_integrity_rq(req)) {
- ret = nvme_map_metadata(dev, req, &iod->cmd);
+ ret = nvme_map_metadata(dev, req);
if (ret)
goto out_unmap_data;
}
@@ -903,7 +985,7 @@ static blk_status_t nvme_queue_rq(struct blk_mq_hw_ctx *hctx,
return BLK_STS_OK;
}
-static void nvme_submit_cmds(struct nvme_queue *nvmeq, struct request **rqlist)
+static void nvme_submit_cmds(struct nvme_queue *nvmeq, struct rq_list *rqlist)
{
struct request *req;
@@ -931,11 +1013,10 @@ static bool nvme_prep_rq_batch(struct nvme_queue *nvmeq, struct request *req)
return nvme_prep_rq(nvmeq->dev, req) == BLK_STS_OK;
}
-static void nvme_queue_rqs(struct request **rqlist)
+static void nvme_queue_rqs(struct rq_list *rqlist)
{
- struct request *submit_list = NULL;
- struct request *requeue_list = NULL;
- struct request **requeue_lastp = &requeue_list;
+ struct rq_list submit_list = { };
+ struct rq_list requeue_list = { };
struct nvme_queue *nvmeq = NULL;
struct request *req;
@@ -945,9 +1026,9 @@ static void nvme_queue_rqs(struct request **rqlist)
nvmeq = req->mq_hctx->driver_data;
if (nvme_prep_rq_batch(nvmeq, req))
- rq_list_add(&submit_list, req); /* reverse order */
+ rq_list_add_tail(&submit_list, req);
else
- rq_list_add_tail(&requeue_lastp, req);
+ rq_list_add_tail(&requeue_list, req);
}
if (nvmeq)
@@ -955,17 +1036,31 @@ static void nvme_queue_rqs(struct request **rqlist)
*rqlist = requeue_list;
}
+static __always_inline void nvme_unmap_metadata(struct nvme_dev *dev,
+ struct request *req)
+{
+ struct nvme_iod *iod = blk_mq_rq_to_pdu(req);
+
+ if (!iod->meta_sgt.nents) {
+ dma_unmap_page(dev->dev, iod->meta_dma,
+ rq_integrity_vec(req).bv_len,
+ rq_dma_dir(req));
+ return;
+ }
+
+ dma_pool_free(dev->prp_small_pool, iod->meta_list.sg_list,
+ iod->meta_dma);
+ dma_unmap_sgtable(dev->dev, &iod->meta_sgt, rq_dma_dir(req), 0);
+ mempool_free(iod->meta_sgt.sgl, dev->iod_meta_mempool);
+}
+
static __always_inline void nvme_pci_unmap_rq(struct request *req)
{
struct nvme_queue *nvmeq = req->mq_hctx->driver_data;
struct nvme_dev *dev = nvmeq->dev;
- if (blk_integrity_rq(req)) {
- struct nvme_iod *iod = blk_mq_rq_to_pdu(req);
-
- dma_unmap_page(dev->dev, iod->meta_dma,
- rq_integrity_vec(req).bv_len, rq_dma_dir(req));
- }
+ if (blk_integrity_rq(req))
+ nvme_unmap_metadata(dev, req);
if (blk_rq_nr_phys_segments(req))
nvme_unmap_data(dev, req);
@@ -1035,8 +1130,9 @@ static inline void nvme_handle_cqe(struct nvme_queue *nvmeq,
trace_nvme_sq(req, cqe->sq_head, nvmeq->sq_tail);
if (!nvme_try_complete_req(req, cqe->status, cqe->result) &&
- !blk_mq_add_to_batch(req, iob, nvme_req(req)->status,
- nvme_pci_complete_batch))
+ !blk_mq_add_to_batch(req, iob,
+ nvme_req(req)->status != NVME_SC_SUCCESS,
+ nvme_pci_complete_batch))
nvme_pci_complete_rq(req);
}
@@ -1079,7 +1175,7 @@ static irqreturn_t nvme_irq(int irq, void *data)
DEFINE_IO_COMP_BATCH(iob);
if (nvme_poll_cq(nvmeq, &iob)) {
- if (!rq_list_empty(iob.req_list))
+ if (!rq_list_empty(&iob.req_list))
nvme_pci_complete_batch(&iob);
return IRQ_HANDLED;
}
@@ -1106,7 +1202,9 @@ static void nvme_poll_irqdisable(struct nvme_queue *nvmeq)
WARN_ON_ONCE(test_bit(NVMEQ_POLLED, &nvmeq->flags));
disable_irq(pci_irq_vector(pdev, nvmeq->cq_vector));
+ spin_lock(&nvmeq->cq_poll_lock);
nvme_poll_cq(nvmeq, NULL);
+ spin_unlock(&nvmeq->cq_poll_lock);
enable_irq(pci_irq_vector(pdev, nvmeq->cq_vector));
}
@@ -1316,9 +1414,20 @@ static enum blk_eh_timer_return nvme_timeout(struct request *req)
struct nvme_dev *dev = nvmeq->dev;
struct request *abort_req;
struct nvme_command cmd = { };
+ struct pci_dev *pdev = to_pci_dev(dev->dev);
u32 csts = readl(dev->bar + NVME_REG_CSTS);
u8 opcode;
+ /*
+ * Shutdown the device immediately if we see it is disconnected. This
+ * unblocks PCIe error handling if the nvme driver is waiting in
+ * error_resume for a device that has been removed. We can't unbind the
+ * driver while the driver's error callback is waiting to complete, so
+ * we're relying on a timeout to break that deadlock if a removal
+ * occurs while reset work is running.
+ */
+ if (pci_dev_is_disconnected(pdev))
+ nvme_change_ctrl_state(&dev->ctrl, NVME_CTRL_DELETING);
if (nvme_state_terminal(&dev->ctrl))
goto disable;
@@ -1326,7 +1435,7 @@ static enum blk_eh_timer_return nvme_timeout(struct request *req)
* the recovery mechanism will surely fail.
*/
mb();
- if (pci_channel_offline(to_pci_dev(dev->dev)))
+ if (pci_channel_offline(pdev))
return BLK_EH_RESET_TIMER;
/*
@@ -1779,8 +1888,28 @@ static int nvme_pci_configure_admin_queue(struct nvme_dev *dev)
* might be pointing at!
*/
result = nvme_disable_ctrl(&dev->ctrl, false);
- if (result < 0)
- return result;
+ if (result < 0) {
+ struct pci_dev *pdev = to_pci_dev(dev->dev);
+
+ /*
+ * The NVMe Controller Reset method did not get an expected
+ * CSTS.RDY transition, so something with the device appears to
+ * be stuck. Use the lower level and bigger hammer PCIe
+ * Function Level Reset to attempt restoring the device to its
+ * initial state, and try again.
+ */
+ result = pcie_reset_flr(pdev, false);
+ if (result < 0)
+ return result;
+
+ pci_restore_state(pdev);
+ result = nvme_disable_ctrl(&dev->ctrl, false);
+ if (result < 0)
+ return result;
+
+ dev_info(dev->ctrl.device,
+ "controller reset completed after pcie flr\n");
+ }
result = nvme_alloc_queue(dev, 0, NVME_AQ_DEPTH);
if (result)
@@ -1888,6 +2017,18 @@ static void nvme_map_cmb(struct nvme_dev *dev)
return;
/*
+ * Controllers may support a CMB size larger than their BAR, for
+ * example, due to being behind a bridge. Reduce the CMB to the
+ * reported size of the BAR
+ */
+ size = min(size, bar_size - offset);
+
+ if (!IS_ALIGNED(size, memremap_compat_align()) ||
+ !IS_ALIGNED(pci_resource_start(pdev, bar),
+ memremap_compat_align()))
+ return;
+
+ /*
* Tell the controller about the host side address mapping the CMB,
* and enable CMB decoding for the NVMe 1.4+ scheme:
*/
@@ -1897,17 +2038,10 @@ static void nvme_map_cmb(struct nvme_dev *dev)
dev->bar + NVME_REG_CMBMSC);
}
- /*
- * Controllers may support a CMB size larger than their BAR,
- * for example, due to being behind a bridge. Reduce the CMB to
- * the reported size of the BAR
- */
- if (size > bar_size - offset)
- size = bar_size - offset;
-
if (pci_p2pdma_add_resource(pdev, bar, size, offset)) {
dev_warn(dev->ctrl.device,
"failed to register the CMB\n");
+ hi_lo_writeq(0, dev->bar + NVME_REG_CMBMSC);
return;
}
@@ -2719,6 +2853,7 @@ static void nvme_release_prp_pools(struct nvme_dev *dev)
static int nvme_pci_alloc_iod_mempool(struct nvme_dev *dev)
{
+ size_t meta_size = sizeof(struct scatterlist) * (NVME_MAX_META_SEGS + 1);
size_t alloc_size = sizeof(struct scatterlist) * NVME_MAX_SEGS;
dev->iod_mempool = mempool_create_node(1,
@@ -2727,7 +2862,18 @@ static int nvme_pci_alloc_iod_mempool(struct nvme_dev *dev)
dev_to_node(dev->dev));
if (!dev->iod_mempool)
return -ENOMEM;
+
+ dev->iod_meta_mempool = mempool_create_node(1,
+ mempool_kmalloc, mempool_kfree,
+ (void *)meta_size, GFP_KERNEL,
+ dev_to_node(dev->dev));
+ if (!dev->iod_meta_mempool)
+ goto free;
+
return 0;
+free:
+ mempool_destroy(dev->iod_mempool);
+ return -ENOMEM;
}
static void nvme_free_tagset(struct nvme_dev *dev)
@@ -2792,6 +2938,11 @@ static void nvme_reset_work(struct work_struct *work)
if (result)
goto out;
+ if (nvme_ctrl_meta_sgl_supported(&dev->ctrl))
+ dev->ctrl.max_integrity_segments = NVME_MAX_META_SEGS;
+ else
+ dev->ctrl.max_integrity_segments = 1;
+
nvme_dbbuf_dma_alloc(dev);
result = nvme_setup_host_mem(dev);
@@ -2987,7 +3138,9 @@ static unsigned long check_vendor_combination_bug(struct pci_dev *pdev)
* because of high power consumption (> 2 Watt) in s2idle
* sleep. Only some boards with Intel CPU are affected.
*/
- if (dmi_match(DMI_BOARD_NAME, "GMxPXxx") ||
+ if (dmi_match(DMI_BOARD_NAME, "DN50Z-140HC-YD") ||
+ dmi_match(DMI_BOARD_NAME, "GMxPXxx") ||
+ dmi_match(DMI_BOARD_NAME, "GXxMRXx") ||
dmi_match(DMI_BOARD_NAME, "PH4PG31") ||
dmi_match(DMI_BOARD_NAME, "PH4PRX1_PH6PRX1") ||
dmi_match(DMI_BOARD_NAME, "PH6PG01_PH6PG71"))
@@ -3059,11 +3212,6 @@ static struct nvme_dev *nvme_pci_alloc_dev(struct pci_dev *pdev,
dev->ctrl.max_hw_sectors = min_t(u32,
NVME_MAX_KB_SZ << 1, dma_opt_mapping_size(&pdev->dev) >> 9);
dev->ctrl.max_segments = NVME_MAX_SEGS;
-
- /*
- * There is no support for SGLs for metadata (yet), so we are limited to
- * a single integrity segment for the separate metadata pointer.
- */
dev->ctrl.max_integrity_segments = 1;
return dev;
@@ -3126,6 +3274,11 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id)
if (result)
goto out_disable;
+ if (nvme_ctrl_meta_sgl_supported(&dev->ctrl))
+ dev->ctrl.max_integrity_segments = NVME_MAX_META_SEGS;
+ else
+ dev->ctrl.max_integrity_segments = 1;
+
nvme_dbbuf_dma_alloc(dev);
result = nvme_setup_host_mem(dev);
@@ -3168,6 +3321,7 @@ out_disable:
nvme_free_queues(dev, 0);
out_release_iod_mempool:
mempool_destroy(dev->iod_mempool);
+ mempool_destroy(dev->iod_meta_mempool);
out_release_prp_pools:
nvme_release_prp_pools(dev);
out_dev_unmap:
@@ -3233,6 +3387,7 @@ static void nvme_remove(struct pci_dev *pdev)
nvme_dbbuf_dma_free(dev);
nvme_free_queues(dev, 0);
mempool_destroy(dev->iod_mempool);
+ mempool_destroy(dev->iod_meta_mempool);
nvme_release_prp_pools(dev);
nvme_dev_unmap(dev);
nvme_uninit_ctrl(&dev->ctrl);
@@ -3402,7 +3557,7 @@ static pci_ers_result_t nvme_slot_reset(struct pci_dev *pdev)
dev_info(dev->ctrl.device, "restart after slot reset\n");
pci_restore_state(pdev);
- if (!nvme_try_sched_reset(&dev->ctrl))
+ if (nvme_try_sched_reset(&dev->ctrl))
nvme_unquiesce_io_queues(&dev->ctrl);
return PCI_ERS_RESULT_RECOVERED;
}
@@ -3452,6 +3607,9 @@ static const struct pci_device_id nvme_id_table[] = {
.driver_data = NVME_QUIRK_BOGUS_NID, },
{ PCI_DEVICE(0x1217, 0x8760), /* O2 Micro 64GB Steam Deck */
.driver_data = NVME_QUIRK_DMAPOOL_ALIGN_512, },
+ { PCI_DEVICE(0x126f, 0x1001), /* Silicon Motion generic */
+ .driver_data = NVME_QUIRK_NO_DEEPEST_PS |
+ NVME_QUIRK_IGNORE_DEV_SUBNQN, },
{ PCI_DEVICE(0x126f, 0x2262), /* Silicon Motion generic */
.driver_data = NVME_QUIRK_NO_DEEPEST_PS |
NVME_QUIRK_BOGUS_NID, },
@@ -3475,6 +3633,9 @@ static const struct pci_device_id nvme_id_table[] = {
NVME_QUIRK_IGNORE_DEV_SUBNQN, },
{ PCI_DEVICE(0x15b7, 0x5008), /* Sandisk SN530 */
.driver_data = NVME_QUIRK_BROKEN_MSI },
+ { PCI_DEVICE(0x15b7, 0x5009), /* Sandisk SN550 */
+ .driver_data = NVME_QUIRK_BROKEN_MSI |
+ NVME_QUIRK_NO_DEEPEST_PS },
{ PCI_DEVICE(0x1987, 0x5012), /* Phison E12 */
.driver_data = NVME_QUIRK_BOGUS_NID, },
{ PCI_DEVICE(0x1987, 0x5016), /* Phison E16 */
@@ -3552,12 +3713,16 @@ static const struct pci_device_id nvme_id_table[] = {
.driver_data = NVME_QUIRK_BOGUS_NID, },
{ PCI_DEVICE(0x1cc1, 0x5350), /* ADATA XPG GAMMIX S50 */
.driver_data = NVME_QUIRK_BOGUS_NID, },
+ { PCI_DEVICE(0x1dbe, 0x5216), /* Acer/INNOGRIT FA100/5216 NVMe SSD */
+ .driver_data = NVME_QUIRK_BOGUS_NID, },
{ PCI_DEVICE(0x1dbe, 0x5236), /* ADATA XPG GAMMIX S70 */
.driver_data = NVME_QUIRK_BOGUS_NID, },
{ PCI_DEVICE(0x1e49, 0x0021), /* ZHITAI TiPro5000 NVMe SSD */
.driver_data = NVME_QUIRK_NO_DEEPEST_PS, },
{ PCI_DEVICE(0x1e49, 0x0041), /* ZHITAI TiPro7000 NVMe SSD */
.driver_data = NVME_QUIRK_NO_DEEPEST_PS, },
+ { PCI_DEVICE(0x025e, 0xf1ac), /* SOLIDIGM P44 pro SSDPFKKW020X7 */
+ .driver_data = NVME_QUIRK_NO_DEEPEST_PS, },
{ PCI_DEVICE(0xc0a9, 0x540a), /* Crucial P2 */
.driver_data = NVME_QUIRK_BOGUS_NID, },
{ PCI_DEVICE(0x1d97, 0x2263), /* Lexar NM610 */
diff --git a/drivers/nvme/host/pr.c b/drivers/nvme/host/pr.c
index dc7922f22600..80dd09aa01a3 100644
--- a/drivers/nvme/host/pr.c
+++ b/drivers/nvme/host/pr.c
@@ -82,8 +82,6 @@ static int nvme_status_to_pr_err(int status)
return PR_STS_SUCCESS;
case NVME_SC_RESERVATION_CONFLICT:
return PR_STS_RESERVATION_CONFLICT;
- case NVME_SC_ONCS_NOT_SUPPORTED:
- return -EOPNOTSUPP;
case NVME_SC_BAD_ATTRIBUTES:
case NVME_SC_INVALID_OPCODE:
case NVME_SC_INVALID_FIELD:
diff --git a/drivers/nvme/host/sysfs.c b/drivers/nvme/host/sysfs.c
index b68a9e5f1ea3..3a41b9ab0f13 100644
--- a/drivers/nvme/host/sysfs.c
+++ b/drivers/nvme/host/sysfs.c
@@ -792,7 +792,7 @@ static umode_t nvme_tls_attrs_are_visible(struct kobject *kobj,
return a->mode;
}
-const struct attribute_group nvme_tls_attrs_group = {
+static const struct attribute_group nvme_tls_attrs_group = {
.attrs = nvme_tls_attrs,
.is_visible = nvme_tls_attrs_are_visible,
};
diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c
index 55abfe5e1d25..83a6b18b01ad 100644
--- a/drivers/nvme/host/tcp.c
+++ b/drivers/nvme/host/tcp.c
@@ -54,6 +54,8 @@ MODULE_PARM_DESC(tls_handshake_timeout,
"nvme TLS handshake timeout in seconds (default 10)");
#endif
+static atomic_t nvme_tcp_cpu_queues[NR_CPUS];
+
#ifdef CONFIG_DEBUG_LOCK_ALLOC
/* lockdep can detect a circular dependency of the form
* sk_lock -> mmap_lock (page fault) -> fs locks -> sk_lock
@@ -127,6 +129,7 @@ enum nvme_tcp_queue_flags {
NVME_TCP_Q_ALLOCATED = 0,
NVME_TCP_Q_LIVE = 1,
NVME_TCP_Q_POLLING = 2,
+ NVME_TCP_Q_IO_CPU_SET = 3,
};
enum nvme_tcp_recv_state {
@@ -214,6 +217,19 @@ static inline int nvme_tcp_queue_id(struct nvme_tcp_queue *queue)
return queue - queue->ctrl->queues;
}
+static inline bool nvme_tcp_recv_pdu_supported(enum nvme_tcp_pdu_type type)
+{
+ switch (type) {
+ case nvme_tcp_c2h_term:
+ case nvme_tcp_c2h_data:
+ case nvme_tcp_r2t:
+ case nvme_tcp_rsp:
+ return true;
+ default:
+ return false;
+ }
+}
+
/*
* Check if the queue is TLS encrypted
*/
@@ -437,7 +453,8 @@ nvme_tcp_fetch_request(struct nvme_tcp_queue *queue)
return NULL;
}
- list_del(&req->entry);
+ list_del_init(&req->entry);
+ init_llist_node(&req->lentry);
return req;
}
@@ -545,6 +562,8 @@ static int nvme_tcp_init_request(struct blk_mq_tag_set *set,
req->queue = queue;
nvme_req(rq)->ctrl = &ctrl->ctrl;
nvme_req(rq)->cmd = &pdu->cmd;
+ init_llist_node(&req->lentry);
+ INIT_LIST_HEAD(&req->entry);
return 0;
}
@@ -749,6 +768,14 @@ static int nvme_tcp_handle_r2t(struct nvme_tcp_queue *queue,
return -EPROTO;
}
+ if (llist_on_list(&req->lentry) ||
+ !list_empty(&req->entry)) {
+ dev_err(queue->ctrl->ctrl.device,
+ "req %d unexpected r2t while processing request\n",
+ rq->tag);
+ return -EPROTO;
+ }
+
req->pdu_len = 0;
req->h2cdata_left = r2t_length;
req->h2cdata_offset = r2t_offset;
@@ -760,6 +787,40 @@ static int nvme_tcp_handle_r2t(struct nvme_tcp_queue *queue,
return 0;
}
+static void nvme_tcp_handle_c2h_term(struct nvme_tcp_queue *queue,
+ struct nvme_tcp_term_pdu *pdu)
+{
+ u16 fes;
+ const char *msg;
+ u32 plen = le32_to_cpu(pdu->hdr.plen);
+
+ static const char * const msg_table[] = {
+ [NVME_TCP_FES_INVALID_PDU_HDR] = "Invalid PDU Header Field",
+ [NVME_TCP_FES_PDU_SEQ_ERR] = "PDU Sequence Error",
+ [NVME_TCP_FES_HDR_DIGEST_ERR] = "Header Digest Error",
+ [NVME_TCP_FES_DATA_OUT_OF_RANGE] = "Data Transfer Out Of Range",
+ [NVME_TCP_FES_DATA_LIMIT_EXCEEDED] = "Data Transfer Limit Exceeded",
+ [NVME_TCP_FES_UNSUPPORTED_PARAM] = "Unsupported Parameter",
+ };
+
+ if (plen < NVME_TCP_MIN_C2HTERM_PLEN ||
+ plen > NVME_TCP_MAX_C2HTERM_PLEN) {
+ dev_err(queue->ctrl->ctrl.device,
+ "Received a malformed C2HTermReq PDU (plen = %u)\n",
+ plen);
+ return;
+ }
+
+ fes = le16_to_cpu(pdu->fes);
+ if (fes && fes < ARRAY_SIZE(msg_table))
+ msg = msg_table[fes];
+ else
+ msg = "Unknown";
+
+ dev_err(queue->ctrl->ctrl.device,
+ "Received C2HTermReq (FES = %s)\n", msg);
+}
+
static int nvme_tcp_recv_pdu(struct nvme_tcp_queue *queue, struct sk_buff *skb,
unsigned int *offset, size_t *len)
{
@@ -781,6 +842,25 @@ static int nvme_tcp_recv_pdu(struct nvme_tcp_queue *queue, struct sk_buff *skb,
return 0;
hdr = queue->pdu;
+ if (unlikely(hdr->hlen != sizeof(struct nvme_tcp_rsp_pdu))) {
+ if (!nvme_tcp_recv_pdu_supported(hdr->type))
+ goto unsupported_pdu;
+
+ dev_err(queue->ctrl->ctrl.device,
+ "pdu type %d has unexpected header length (%d)\n",
+ hdr->type, hdr->hlen);
+ return -EPROTO;
+ }
+
+ if (unlikely(hdr->type == nvme_tcp_c2h_term)) {
+ /*
+ * C2HTermReq never includes Header or Data digests.
+ * Skip the checks.
+ */
+ nvme_tcp_handle_c2h_term(queue, (void *)queue->pdu);
+ return -EINVAL;
+ }
+
if (queue->hdr_digest) {
ret = nvme_tcp_verify_hdgst(queue, queue->pdu, hdr->hlen);
if (unlikely(ret))
@@ -804,10 +884,13 @@ static int nvme_tcp_recv_pdu(struct nvme_tcp_queue *queue, struct sk_buff *skb,
nvme_tcp_init_recv_ctx(queue);
return nvme_tcp_handle_r2t(queue, (void *)queue->pdu);
default:
- dev_err(queue->ctrl->ctrl.device,
- "unsupported pdu type (%d)\n", hdr->type);
- return -EINVAL;
+ goto unsupported_pdu;
}
+
+unsupported_pdu:
+ dev_err(queue->ctrl->ctrl.device,
+ "unsupported pdu type (%d)\n", hdr->type);
+ return -EINVAL;
}
static inline void nvme_tcp_end_request(struct request *rq, u16 status)
@@ -1277,7 +1360,7 @@ static int nvme_tcp_try_recv(struct nvme_tcp_queue *queue)
queue->nr_cqe = 0;
consumed = sock->ops->read_sock(sk, &rd_desc, nvme_tcp_recv_skb);
release_sock(sk);
- return consumed;
+ return consumed == -EAGAIN ? 0 : consumed;
}
static void nvme_tcp_io_work(struct work_struct *w)
@@ -1305,6 +1388,11 @@ static void nvme_tcp_io_work(struct work_struct *w)
else if (unlikely(result < 0))
return;
+ /* did we get some space after spending time in recv? */
+ if (nvme_tcp_queue_has_pending(queue) &&
+ sk_stream_is_writeable(queue->sock->sk))
+ pending = true;
+
if (!pending || !queue->rd_enabled)
return;
@@ -1446,8 +1534,11 @@ static int nvme_tcp_init_connection(struct nvme_tcp_queue *queue)
msg.msg_control = cbuf;
msg.msg_controllen = sizeof(cbuf);
}
+ msg.msg_flags = MSG_WAITALL;
ret = kernel_recvmsg(queue->sock, &msg, &iov, 1,
iov.iov_len, msg.msg_flags);
+ if (ret >= 0 && ret < sizeof(*icresp))
+ ret = -ECONNRESET;
if (ret < 0) {
pr_warn("queue %d: failed to receive icresp, error %d\n",
nvme_tcp_queue_id(queue), ret);
@@ -1562,23 +1653,56 @@ static bool nvme_tcp_poll_queue(struct nvme_tcp_queue *queue)
ctrl->io_queues[HCTX_TYPE_POLL];
}
+/*
+ * Track the number of queues assigned to each cpu using a global per-cpu
+ * counter and select the least used cpu from the mq_map. Our goal is to spread
+ * different controllers I/O threads across different cpu cores.
+ *
+ * Note that the accounting is not 100% perfect, but we don't need to be, we're
+ * simply putting our best effort to select the best candidate cpu core that we
+ * find at any given point.
+ */
static void nvme_tcp_set_queue_io_cpu(struct nvme_tcp_queue *queue)
{
struct nvme_tcp_ctrl *ctrl = queue->ctrl;
- int qid = nvme_tcp_queue_id(queue);
- int n = 0;
+ struct blk_mq_tag_set *set = &ctrl->tag_set;
+ int qid = nvme_tcp_queue_id(queue) - 1;
+ unsigned int *mq_map = NULL;
+ int cpu, min_queues = INT_MAX, io_cpu;
+
+ if (wq_unbound)
+ goto out;
if (nvme_tcp_default_queue(queue))
- n = qid - 1;
+ mq_map = set->map[HCTX_TYPE_DEFAULT].mq_map;
else if (nvme_tcp_read_queue(queue))
- n = qid - ctrl->io_queues[HCTX_TYPE_DEFAULT] - 1;
+ mq_map = set->map[HCTX_TYPE_READ].mq_map;
else if (nvme_tcp_poll_queue(queue))
- n = qid - ctrl->io_queues[HCTX_TYPE_DEFAULT] -
- ctrl->io_queues[HCTX_TYPE_READ] - 1;
- if (wq_unbound)
- queue->io_cpu = WORK_CPU_UNBOUND;
- else
- queue->io_cpu = cpumask_next_wrap(n - 1, cpu_online_mask, -1, false);
+ mq_map = set->map[HCTX_TYPE_POLL].mq_map;
+
+ if (WARN_ON(!mq_map))
+ goto out;
+
+ /* Search for the least used cpu from the mq_map */
+ io_cpu = WORK_CPU_UNBOUND;
+ for_each_online_cpu(cpu) {
+ int num_queues = atomic_read(&nvme_tcp_cpu_queues[cpu]);
+
+ if (mq_map[cpu] != qid)
+ continue;
+ if (num_queues < min_queues) {
+ io_cpu = cpu;
+ min_queues = num_queues;
+ }
+ }
+ if (io_cpu != WORK_CPU_UNBOUND) {
+ queue->io_cpu = io_cpu;
+ atomic_inc(&nvme_tcp_cpu_queues[io_cpu]);
+ set_bit(NVME_TCP_Q_IO_CPU_SET, &queue->flags);
+ }
+out:
+ dev_dbg(ctrl->ctrl.device, "queue %d: using cpu %d\n",
+ qid, queue->io_cpu);
}
static void nvme_tcp_tls_done(void *data, int status, key_serial_t pskid)
@@ -1653,9 +1777,14 @@ static int nvme_tcp_start_tls(struct nvme_ctrl *nctrl,
qid, ret);
tls_handshake_cancel(queue->sock->sk);
} else {
- dev_dbg(nctrl->device,
- "queue %d: TLS handshake complete, error %d\n",
- qid, queue->tls_err);
+ if (queue->tls_err) {
+ dev_err(nctrl->device,
+ "queue %d: TLS handshake complete, error %d\n",
+ qid, queue->tls_err);
+ } else {
+ dev_dbg(nctrl->device,
+ "queue %d: TLS handshake complete\n", qid);
+ }
ret = queue->tls_err;
}
return ret;
@@ -1722,7 +1851,7 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl, int qid,
queue->sock->sk->sk_allocation = GFP_ATOMIC;
queue->sock->sk->sk_use_task_frag = false;
- nvme_tcp_set_queue_io_cpu(queue);
+ queue->io_cpu = WORK_CPU_UNBOUND;
queue->request = NULL;
queue->data_remaining = 0;
queue->ddgst_remaining = 0;
@@ -1836,7 +1965,7 @@ static void __nvme_tcp_stop_queue(struct nvme_tcp_queue *queue)
cancel_work_sync(&queue->io_work);
}
-static void nvme_tcp_stop_queue(struct nvme_ctrl *nctrl, int qid)
+static void nvme_tcp_stop_queue_nowait(struct nvme_ctrl *nctrl, int qid)
{
struct nvme_tcp_ctrl *ctrl = to_tcp_ctrl(nctrl);
struct nvme_tcp_queue *queue = &ctrl->queues[qid];
@@ -1844,6 +1973,9 @@ static void nvme_tcp_stop_queue(struct nvme_ctrl *nctrl, int qid)
if (!test_bit(NVME_TCP_Q_ALLOCATED, &queue->flags))
return;
+ if (test_and_clear_bit(NVME_TCP_Q_IO_CPU_SET, &queue->flags))
+ atomic_dec(&nvme_tcp_cpu_queues[queue->io_cpu]);
+
mutex_lock(&queue->queue_lock);
if (test_and_clear_bit(NVME_TCP_Q_LIVE, &queue->flags))
__nvme_tcp_stop_queue(queue);
@@ -1852,6 +1984,31 @@ static void nvme_tcp_stop_queue(struct nvme_ctrl *nctrl, int qid)
mutex_unlock(&queue->queue_lock);
}
+static void nvme_tcp_wait_queue(struct nvme_ctrl *nctrl, int qid)
+{
+ struct nvme_tcp_ctrl *ctrl = to_tcp_ctrl(nctrl);
+ struct nvme_tcp_queue *queue = &ctrl->queues[qid];
+ int timeout = 100;
+
+ while (timeout > 0) {
+ if (!test_bit(NVME_TCP_Q_ALLOCATED, &queue->flags) ||
+ !sk_wmem_alloc_get(queue->sock->sk))
+ return;
+ msleep(2);
+ timeout -= 2;
+ }
+ dev_warn(nctrl->device,
+ "qid %d: timeout draining sock wmem allocation expired\n",
+ qid);
+}
+
+static void nvme_tcp_stop_queue(struct nvme_ctrl *nctrl, int qid)
+{
+ nvme_tcp_stop_queue_nowait(nctrl, qid);
+ nvme_tcp_wait_queue(nctrl, qid);
+}
+
+
static void nvme_tcp_setup_sock_ops(struct nvme_tcp_queue *queue)
{
write_lock_bh(&queue->sock->sk->sk_callback_lock);
@@ -1878,9 +2035,10 @@ static int nvme_tcp_start_queue(struct nvme_ctrl *nctrl, int idx)
nvme_tcp_init_recv_ctx(queue);
nvme_tcp_setup_sock_ops(queue);
- if (idx)
+ if (idx) {
+ nvme_tcp_set_queue_io_cpu(queue);
ret = nvmf_connect_io_queue(nctrl, idx);
- else
+ } else
ret = nvmf_connect_admin_queue(nctrl);
if (!ret) {
@@ -1918,7 +2076,9 @@ static void nvme_tcp_stop_io_queues(struct nvme_ctrl *ctrl)
int i;
for (i = 1; i < ctrl->queue_count; i++)
- nvme_tcp_stop_queue(ctrl, i);
+ nvme_tcp_stop_queue_nowait(ctrl, i);
+ for (i = 1; i < ctrl->queue_count; i++)
+ nvme_tcp_wait_queue(ctrl, i);
}
static int nvme_tcp_start_io_queues(struct nvme_ctrl *ctrl,
@@ -2455,6 +2615,8 @@ static void nvme_tcp_submit_async_event(struct nvme_ctrl *arg)
ctrl->async_req.offset = 0;
ctrl->async_req.curr_bio = NULL;
ctrl->async_req.data_len = 0;
+ init_llist_node(&ctrl->async_req.lentry);
+ INIT_LIST_HEAD(&ctrl->async_req.entry);
nvme_tcp_queue_request(&ctrl->async_req, true, true);
}
@@ -2624,6 +2786,7 @@ static int nvme_tcp_poll(struct blk_mq_hw_ctx *hctx, struct io_comp_batch *iob)
{
struct nvme_tcp_queue *queue = hctx->driver_data;
struct sock *sk = queue->sock->sk;
+ int ret;
if (!test_bit(NVME_TCP_Q_LIVE, &queue->flags))
return 0;
@@ -2631,9 +2794,9 @@ static int nvme_tcp_poll(struct blk_mq_hw_ctx *hctx, struct io_comp_batch *iob)
set_bit(NVME_TCP_Q_POLLING, &queue->flags);
if (sk_can_busy_loop(sk) && skb_queue_empty_lockless(&sk->sk_receive_queue))
sk_busy_loop(sk, true);
- nvme_tcp_try_recv(queue);
+ ret = nvme_tcp_try_recv(queue);
clear_bit(NVME_TCP_Q_POLLING, &queue->flags);
- return queue->nr_cqe;
+ return ret < 0 ? ret : queue->nr_cqe;
}
static int nvme_tcp_get_address(struct nvme_ctrl *ctrl, char *buf, int size)
@@ -2856,6 +3019,7 @@ static struct nvmf_transport_ops nvme_tcp_transport = {
static int __init nvme_tcp_init_module(void)
{
unsigned int wq_flags = WQ_MEM_RECLAIM | WQ_HIGHPRI | WQ_SYSFS;
+ int cpu;
BUILD_BUG_ON(sizeof(struct nvme_tcp_hdr) != 8);
BUILD_BUG_ON(sizeof(struct nvme_tcp_cmd_pdu) != 72);
@@ -2873,6 +3037,9 @@ static int __init nvme_tcp_init_module(void)
if (!nvme_tcp_wq)
return -ENOMEM;
+ for_each_possible_cpu(cpu)
+ atomic_set(&nvme_tcp_cpu_queues[cpu], 0);
+
nvmf_register_transport(&nvme_tcp_transport);
return 0;
}
generated by cgit v1.2.3 (git 2.39.0) at 2025-08-29 18:25:55 +0300