diff options
Diffstat (limited to 'drivers/infiniband/ulp/isert/ib_isert.c')
| -rw-r--r-- | drivers/infiniband/ulp/isert/ib_isert.c | 2765 |
1 files changed, 1483 insertions, 1282 deletions
diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index 3f62041222f2..af811d060cc8 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -1,19 +1,11 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /******************************************************************************* * This file contains iSCSI extentions for RDMA (iSER) Verbs * - * (c) Copyright 2013 RisingTide Systems LLC. + * (c) Copyright 2013 Datera, Inc. * * Nicholas A. Bellinger <nab@linux-iscsi.org> * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. ****************************************************************************/ #include <linux/string.h> @@ -23,300 +15,241 @@ #include <linux/in.h> #include <linux/in6.h> #include <rdma/ib_verbs.h> +#include <rdma/ib_cm.h> #include <rdma/rdma_cm.h> #include <target/target_core_base.h> #include <target/target_core_fabric.h> #include <target/iscsi/iscsi_transport.h> +#include <linux/semaphore.h> -#include "isert_proto.h" #include "ib_isert.h" -#define ISERT_MAX_CONN 8 -#define ISER_MAX_RX_CQ_LEN (ISERT_QP_MAX_RECV_DTOS * ISERT_MAX_CONN) -#define ISER_MAX_TX_CQ_LEN (ISERT_QP_MAX_REQ_DTOS * ISERT_MAX_CONN) +static int isert_debug_level; +module_param_named(debug_level, isert_debug_level, int, 0644); +MODULE_PARM_DESC(debug_level, "Enable debug tracing if > 0 (default:0)"); + +static int isert_sg_tablesize_set(const char *val, + const struct kernel_param *kp); +static const struct kernel_param_ops sg_tablesize_ops = { + .set = isert_sg_tablesize_set, + .get = param_get_int, +}; + +static int isert_sg_tablesize = ISCSI_ISER_MIN_SG_TABLESIZE; +module_param_cb(sg_tablesize, &sg_tablesize_ops, &isert_sg_tablesize, 0644); +MODULE_PARM_DESC(sg_tablesize, + "Number of gather/scatter entries in a single scsi command, should >= 128 (default: 128, max: 4096)"); static DEFINE_MUTEX(device_list_mutex); static LIST_HEAD(device_list); -static struct workqueue_struct *isert_rx_wq; +static struct workqueue_struct *isert_login_wq; static struct workqueue_struct *isert_comp_wq; -static struct kmem_cache *isert_cmd_cache; +static struct workqueue_struct *isert_release_wq; + +static int +isert_put_response(struct iscsit_conn *conn, struct iscsit_cmd *cmd); +static int +isert_login_post_recv(struct isert_conn *isert_conn); +static int +isert_rdma_accept(struct isert_conn *isert_conn); +struct rdma_cm_id *isert_setup_id(struct isert_np *isert_np); + +static void isert_release_work(struct work_struct *work); +static void isert_recv_done(struct ib_cq *cq, struct ib_wc *wc); +static void isert_send_done(struct ib_cq *cq, struct ib_wc *wc); +static void isert_login_recv_done(struct ib_cq *cq, struct ib_wc *wc); +static void isert_login_send_done(struct ib_cq *cq, struct ib_wc *wc); + +static int isert_sg_tablesize_set(const char *val, const struct kernel_param *kp) +{ + int n = 0, ret; + + ret = kstrtoint(val, 10, &n); + if (ret != 0 || n < ISCSI_ISER_MIN_SG_TABLESIZE || + n > ISCSI_ISER_MAX_SG_TABLESIZE) + return -EINVAL; + + return param_set_int(val, kp); +} + +static inline bool +isert_prot_cmd(struct isert_conn *conn, struct se_cmd *cmd) +{ + return (conn->pi_support && + cmd->prot_op != TARGET_PROT_NORMAL); +} static void isert_qp_event_callback(struct ib_event *e, void *context) { - struct isert_conn *isert_conn = (struct isert_conn *)context; + struct isert_conn *isert_conn = context; + + isert_err("%s (%d): conn %p\n", + ib_event_msg(e->event), e->event, isert_conn); - pr_err("isert_qp_event_callback event: %d\n", e->event); switch (e->event) { case IB_EVENT_COMM_EST: - rdma_notify(isert_conn->conn_cm_id, IB_EVENT_COMM_EST); - break; - case IB_EVENT_QP_LAST_WQE_REACHED: - pr_warn("Reached TX IB_EVENT_QP_LAST_WQE_REACHED:\n"); + rdma_notify(isert_conn->cm_id, IB_EVENT_COMM_EST); break; default: break; } } -static int -isert_query_device(struct ib_device *ib_dev, struct ib_device_attr *devattr) +static struct ib_qp * +isert_create_qp(struct isert_conn *isert_conn, + struct rdma_cm_id *cma_id) { - int ret; - - ret = ib_query_device(ib_dev, devattr); - if (ret) { - pr_err("ib_query_device() failed: %d\n", ret); - return ret; - } - pr_debug("devattr->max_sge: %d\n", devattr->max_sge); - pr_debug("devattr->max_sge_rd: %d\n", devattr->max_sge_rd); - - return 0; -} - -static int -isert_conn_setup_qp(struct isert_conn *isert_conn, struct rdma_cm_id *cma_id) -{ - struct isert_device *device = isert_conn->conn_device; + u32 cq_size = ISERT_QP_MAX_REQ_DTOS + ISERT_QP_MAX_RECV_DTOS + 2; + struct isert_device *device = isert_conn->device; + struct ib_device *ib_dev = device->ib_device; struct ib_qp_init_attr attr; - struct ib_device_attr devattr; - int ret, index, min_index = 0; + int ret, factor; - memset(&devattr, 0, sizeof(struct ib_device_attr)); - ret = isert_query_device(cma_id->device, &devattr); - if (ret) - return ret; - - mutex_lock(&device_list_mutex); - for (index = 0; index < device->cqs_used; index++) - if (device->cq_active_qps[index] < - device->cq_active_qps[min_index]) - min_index = index; - device->cq_active_qps[min_index]++; - pr_debug("isert_conn_setup_qp: Using min_index: %d\n", min_index); - mutex_unlock(&device_list_mutex); + isert_conn->cq = ib_cq_pool_get(ib_dev, cq_size, -1, IB_POLL_WORKQUEUE); + if (IS_ERR(isert_conn->cq)) { + isert_err("Unable to allocate cq\n"); + ret = PTR_ERR(isert_conn->cq); + return ERR_PTR(ret); + } + isert_conn->cq_size = cq_size; memset(&attr, 0, sizeof(struct ib_qp_init_attr)); attr.event_handler = isert_qp_event_callback; attr.qp_context = isert_conn; - attr.send_cq = device->dev_tx_cq[min_index]; - attr.recv_cq = device->dev_rx_cq[min_index]; - attr.cap.max_send_wr = ISERT_QP_MAX_REQ_DTOS; - attr.cap.max_recv_wr = ISERT_QP_MAX_RECV_DTOS; - /* - * FIXME: Use devattr.max_sge - 2 for max_send_sge as - * work-around for RDMA_READ.. - */ - attr.cap.max_send_sge = devattr.max_sge - 2; - isert_conn->max_sge = attr.cap.max_send_sge; - + attr.send_cq = isert_conn->cq; + attr.recv_cq = isert_conn->cq; + attr.cap.max_send_wr = ISERT_QP_MAX_REQ_DTOS + 1; + attr.cap.max_recv_wr = ISERT_QP_MAX_RECV_DTOS + 1; + factor = rdma_rw_mr_factor(device->ib_device, cma_id->port_num, + isert_sg_tablesize); + attr.cap.max_rdma_ctxs = ISCSI_DEF_XMIT_CMDS_MAX * factor; + attr.cap.max_send_sge = device->ib_device->attrs.max_send_sge; attr.cap.max_recv_sge = 1; attr.sq_sig_type = IB_SIGNAL_REQ_WR; attr.qp_type = IB_QPT_RC; + if (device->pi_capable) + attr.create_flags |= IB_QP_CREATE_INTEGRITY_EN; - pr_debug("isert_conn_setup_qp cma_id->device: %p\n", - cma_id->device); - pr_debug("isert_conn_setup_qp conn_pd->device: %p\n", - isert_conn->conn_pd->device); - - ret = rdma_create_qp(cma_id, isert_conn->conn_pd, &attr); + ret = rdma_create_qp(cma_id, device->pd, &attr); if (ret) { - pr_err("rdma_create_qp failed for cma_id %d\n", ret); - return ret; - } - isert_conn->conn_qp = cma_id->qp; - pr_debug("rdma_create_qp() returned success >>>>>>>>>>>>>>>>>>>>>>>>>.\n"); + isert_err("rdma_create_qp failed for cma_id %d\n", ret); + ib_cq_pool_put(isert_conn->cq, isert_conn->cq_size); - return 0; -} + return ERR_PTR(ret); + } -static void -isert_cq_event_callback(struct ib_event *e, void *context) -{ - pr_debug("isert_cq_event_callback event: %d\n", e->event); + return cma_id->qp; } static int isert_alloc_rx_descriptors(struct isert_conn *isert_conn) { - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; + struct isert_device *device = isert_conn->device; + struct ib_device *ib_dev = device->ib_device; struct iser_rx_desc *rx_desc; struct ib_sge *rx_sg; u64 dma_addr; int i, j; - isert_conn->conn_rx_descs = kzalloc(ISERT_QP_MAX_RECV_DTOS * - sizeof(struct iser_rx_desc), GFP_KERNEL); - if (!isert_conn->conn_rx_descs) - goto fail; + isert_conn->rx_descs = kcalloc(ISERT_QP_MAX_RECV_DTOS, + sizeof(struct iser_rx_desc), + GFP_KERNEL); + if (!isert_conn->rx_descs) + return -ENOMEM; - rx_desc = isert_conn->conn_rx_descs; + rx_desc = isert_conn->rx_descs; for (i = 0; i < ISERT_QP_MAX_RECV_DTOS; i++, rx_desc++) { - dma_addr = ib_dma_map_single(ib_dev, (void *)rx_desc, - ISER_RX_PAYLOAD_SIZE, DMA_FROM_DEVICE); + dma_addr = ib_dma_map_single(ib_dev, rx_desc->buf, + ISER_RX_SIZE, DMA_FROM_DEVICE); if (ib_dma_mapping_error(ib_dev, dma_addr)) goto dma_map_fail; rx_desc->dma_addr = dma_addr; rx_sg = &rx_desc->rx_sg; - rx_sg->addr = rx_desc->dma_addr; + rx_sg->addr = rx_desc->dma_addr + isert_get_hdr_offset(rx_desc); rx_sg->length = ISER_RX_PAYLOAD_SIZE; - rx_sg->lkey = isert_conn->conn_mr->lkey; + rx_sg->lkey = device->pd->local_dma_lkey; + rx_desc->rx_cqe.done = isert_recv_done; } - isert_conn->conn_rx_desc_head = 0; return 0; dma_map_fail: - rx_desc = isert_conn->conn_rx_descs; + rx_desc = isert_conn->rx_descs; for (j = 0; j < i; j++, rx_desc++) { ib_dma_unmap_single(ib_dev, rx_desc->dma_addr, - ISER_RX_PAYLOAD_SIZE, DMA_FROM_DEVICE); + ISER_RX_SIZE, DMA_FROM_DEVICE); } - kfree(isert_conn->conn_rx_descs); - isert_conn->conn_rx_descs = NULL; -fail: + kfree(isert_conn->rx_descs); + isert_conn->rx_descs = NULL; + isert_err("conn %p failed to allocate rx descriptors\n", isert_conn); return -ENOMEM; } static void isert_free_rx_descriptors(struct isert_conn *isert_conn) { - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; + struct ib_device *ib_dev = isert_conn->device->ib_device; struct iser_rx_desc *rx_desc; int i; - if (!isert_conn->conn_rx_descs) + if (!isert_conn->rx_descs) return; - rx_desc = isert_conn->conn_rx_descs; + rx_desc = isert_conn->rx_descs; for (i = 0; i < ISERT_QP_MAX_RECV_DTOS; i++, rx_desc++) { ib_dma_unmap_single(ib_dev, rx_desc->dma_addr, - ISER_RX_PAYLOAD_SIZE, DMA_FROM_DEVICE); + ISER_RX_SIZE, DMA_FROM_DEVICE); } - kfree(isert_conn->conn_rx_descs); - isert_conn->conn_rx_descs = NULL; + kfree(isert_conn->rx_descs); + isert_conn->rx_descs = NULL; } -static void isert_cq_tx_callback(struct ib_cq *, void *); -static void isert_cq_rx_callback(struct ib_cq *, void *); - static int isert_create_device_ib_res(struct isert_device *device) { struct ib_device *ib_dev = device->ib_device; - struct isert_cq_desc *cq_desc; - int ret = 0, i, j; - - device->cqs_used = min_t(int, num_online_cpus(), - device->ib_device->num_comp_vectors); - device->cqs_used = min(ISERT_MAX_CQ, device->cqs_used); - pr_debug("Using %d CQs, device %s supports %d vectors\n", - device->cqs_used, device->ib_device->name, - device->ib_device->num_comp_vectors); - device->cq_desc = kzalloc(sizeof(struct isert_cq_desc) * - device->cqs_used, GFP_KERNEL); - if (!device->cq_desc) { - pr_err("Unable to allocate device->cq_desc\n"); - return -ENOMEM; - } - cq_desc = device->cq_desc; - - device->dev_pd = ib_alloc_pd(ib_dev); - if (IS_ERR(device->dev_pd)) { - ret = PTR_ERR(device->dev_pd); - pr_err("ib_alloc_pd failed for dev_pd: %d\n", ret); - goto out_cq_desc; - } - - for (i = 0; i < device->cqs_used; i++) { - cq_desc[i].device = device; - cq_desc[i].cq_index = i; - - device->dev_rx_cq[i] = ib_create_cq(device->ib_device, - isert_cq_rx_callback, - isert_cq_event_callback, - (void *)&cq_desc[i], - ISER_MAX_RX_CQ_LEN, i); - if (IS_ERR(device->dev_rx_cq[i])) - goto out_cq; - - device->dev_tx_cq[i] = ib_create_cq(device->ib_device, - isert_cq_tx_callback, - isert_cq_event_callback, - (void *)&cq_desc[i], - ISER_MAX_TX_CQ_LEN, i); - if (IS_ERR(device->dev_tx_cq[i])) - goto out_cq; + int ret; - if (ib_req_notify_cq(device->dev_rx_cq[i], IB_CQ_NEXT_COMP)) - goto out_cq; + isert_dbg("devattr->max_send_sge: %d devattr->max_recv_sge %d\n", + ib_dev->attrs.max_send_sge, ib_dev->attrs.max_recv_sge); + isert_dbg("devattr->max_sge_rd: %d\n", ib_dev->attrs.max_sge_rd); - if (ib_req_notify_cq(device->dev_tx_cq[i], IB_CQ_NEXT_COMP)) - goto out_cq; + device->pd = ib_alloc_pd(ib_dev, 0); + if (IS_ERR(device->pd)) { + ret = PTR_ERR(device->pd); + isert_err("failed to allocate pd, device %p, ret=%d\n", + device, ret); + return ret; } - device->dev_mr = ib_get_dma_mr(device->dev_pd, IB_ACCESS_LOCAL_WRITE); - if (IS_ERR(device->dev_mr)) { - ret = PTR_ERR(device->dev_mr); - pr_err("ib_get_dma_mr failed for dev_mr: %d\n", ret); - goto out_cq; - } + /* Check signature cap */ + if (ib_dev->attrs.kernel_cap_flags & IBK_INTEGRITY_HANDOVER) + device->pi_capable = true; + else + device->pi_capable = false; return 0; - -out_cq: - for (j = 0; j < i; j++) { - cq_desc = &device->cq_desc[j]; - - if (device->dev_rx_cq[j]) { - cancel_work_sync(&cq_desc->cq_rx_work); - ib_destroy_cq(device->dev_rx_cq[j]); - } - if (device->dev_tx_cq[j]) { - cancel_work_sync(&cq_desc->cq_tx_work); - ib_destroy_cq(device->dev_tx_cq[j]); - } - } - ib_dealloc_pd(device->dev_pd); - -out_cq_desc: - kfree(device->cq_desc); - - return ret; } static void isert_free_device_ib_res(struct isert_device *device) { - struct isert_cq_desc *cq_desc; - int i; - - for (i = 0; i < device->cqs_used; i++) { - cq_desc = &device->cq_desc[i]; + isert_info("device %p\n", device); - cancel_work_sync(&cq_desc->cq_rx_work); - cancel_work_sync(&cq_desc->cq_tx_work); - ib_destroy_cq(device->dev_rx_cq[i]); - ib_destroy_cq(device->dev_tx_cq[i]); - device->dev_rx_cq[i] = NULL; - device->dev_tx_cq[i] = NULL; - } - - ib_dereg_mr(device->dev_mr); - ib_dealloc_pd(device->dev_pd); - kfree(device->cq_desc); + ib_dealloc_pd(device->pd); } static void -isert_device_try_release(struct isert_device *device) +isert_device_put(struct isert_device *device) { mutex_lock(&device_list_mutex); device->refcount--; + isert_info("device %p refcount %d\n", device, device->refcount); if (!device->refcount) { isert_free_device_ib_res(device); list_del(&device->dev_node); @@ -326,7 +259,7 @@ isert_device_try_release(struct isert_device *device) } static struct isert_device * -isert_device_find_by_ib_dev(struct rdma_cm_id *cma_id) +isert_device_get(struct rdma_cm_id *cma_id) { struct isert_device *device; int ret; @@ -335,6 +268,8 @@ isert_device_find_by_ib_dev(struct rdma_cm_id *cma_id) list_for_each_entry(device, &device_list, dev_node) { if (device->ib_device->node_guid == cma_id->device->node_guid) { device->refcount++; + isert_info("Found iser device %p refcount %d\n", + device, device->refcount); mutex_unlock(&device_list_mutex); return device; } @@ -358,172 +293,251 @@ isert_device_find_by_ib_dev(struct rdma_cm_id *cma_id) device->refcount++; list_add_tail(&device->dev_node, &device_list); + isert_info("Created a new iser device %p refcount %d\n", + device, device->refcount); mutex_unlock(&device_list_mutex); return device; } -static int -isert_connect_request(struct rdma_cm_id *cma_id, struct rdma_cm_event *event) +static void +isert_init_conn(struct isert_conn *isert_conn) { - struct iscsi_np *np = cma_id->context; - struct isert_np *isert_np = np->np_context; - struct isert_conn *isert_conn; - struct isert_device *device; - struct ib_device *ib_dev = cma_id->device; - int ret = 0; + isert_conn->state = ISER_CONN_INIT; + INIT_LIST_HEAD(&isert_conn->node); + init_completion(&isert_conn->login_comp); + init_completion(&isert_conn->login_req_comp); + init_waitqueue_head(&isert_conn->rem_wait); + kref_init(&isert_conn->kref); + mutex_init(&isert_conn->mutex); + INIT_WORK(&isert_conn->release_work, isert_release_work); +} - pr_debug("Entering isert_connect_request cma_id: %p, context: %p\n", - cma_id, cma_id->context); +static void +isert_free_login_buf(struct isert_conn *isert_conn) +{ + struct ib_device *ib_dev = isert_conn->device->ib_device; - isert_conn = kzalloc(sizeof(struct isert_conn), GFP_KERNEL); - if (!isert_conn) { - pr_err("Unable to allocate isert_conn\n"); - return -ENOMEM; - } - isert_conn->state = ISER_CONN_INIT; - INIT_LIST_HEAD(&isert_conn->conn_accept_node); - init_completion(&isert_conn->conn_login_comp); - init_waitqueue_head(&isert_conn->conn_wait); - init_waitqueue_head(&isert_conn->conn_wait_comp_err); - kref_init(&isert_conn->conn_kref); - kref_get(&isert_conn->conn_kref); - mutex_init(&isert_conn->conn_mutex); - - cma_id->context = isert_conn; - isert_conn->conn_cm_id = cma_id; - isert_conn->responder_resources = event->param.conn.responder_resources; - isert_conn->initiator_depth = event->param.conn.initiator_depth; - pr_debug("Using responder_resources: %u initiator_depth: %u\n", - isert_conn->responder_resources, isert_conn->initiator_depth); - - isert_conn->login_buf = kzalloc(ISCSI_DEF_MAX_RECV_SEG_LEN + - ISER_RX_LOGIN_SIZE, GFP_KERNEL); - if (!isert_conn->login_buf) { - pr_err("Unable to allocate isert_conn->login_buf\n"); - ret = -ENOMEM; - goto out; - } + ib_dma_unmap_single(ib_dev, isert_conn->login_rsp_dma, + ISER_RX_PAYLOAD_SIZE, DMA_TO_DEVICE); + kfree(isert_conn->login_rsp_buf); - isert_conn->login_req_buf = isert_conn->login_buf; - isert_conn->login_rsp_buf = isert_conn->login_buf + - ISCSI_DEF_MAX_RECV_SEG_LEN; - pr_debug("Set login_buf: %p login_req_buf: %p login_rsp_buf: %p\n", - isert_conn->login_buf, isert_conn->login_req_buf, - isert_conn->login_rsp_buf); + ib_dma_unmap_single(ib_dev, isert_conn->login_desc->dma_addr, + ISER_RX_SIZE, DMA_FROM_DEVICE); + kfree(isert_conn->login_desc); +} - isert_conn->login_req_dma = ib_dma_map_single(ib_dev, - (void *)isert_conn->login_req_buf, - ISCSI_DEF_MAX_RECV_SEG_LEN, DMA_FROM_DEVICE); +static int +isert_alloc_login_buf(struct isert_conn *isert_conn, + struct ib_device *ib_dev) +{ + int ret; - ret = ib_dma_mapping_error(ib_dev, isert_conn->login_req_dma); + isert_conn->login_desc = kzalloc(sizeof(*isert_conn->login_desc), + GFP_KERNEL); + if (!isert_conn->login_desc) + return -ENOMEM; + + isert_conn->login_desc->dma_addr = ib_dma_map_single(ib_dev, + isert_conn->login_desc->buf, + ISER_RX_SIZE, DMA_FROM_DEVICE); + ret = ib_dma_mapping_error(ib_dev, isert_conn->login_desc->dma_addr); if (ret) { - pr_err("ib_dma_mapping_error failed for login_req_dma: %d\n", - ret); - isert_conn->login_req_dma = 0; - goto out_login_buf; + isert_err("login_desc dma mapping error: %d\n", ret); + isert_conn->login_desc->dma_addr = 0; + goto out_free_login_desc; } - isert_conn->login_rsp_dma = ib_dma_map_single(ib_dev, - (void *)isert_conn->login_rsp_buf, - ISER_RX_LOGIN_SIZE, DMA_TO_DEVICE); + isert_conn->login_rsp_buf = kzalloc(ISER_RX_PAYLOAD_SIZE, GFP_KERNEL); + if (!isert_conn->login_rsp_buf) { + ret = -ENOMEM; + goto out_unmap_login_desc; + } + isert_conn->login_rsp_dma = ib_dma_map_single(ib_dev, + isert_conn->login_rsp_buf, + ISER_RX_PAYLOAD_SIZE, DMA_TO_DEVICE); ret = ib_dma_mapping_error(ib_dev, isert_conn->login_rsp_dma); if (ret) { - pr_err("ib_dma_mapping_error failed for login_rsp_dma: %d\n", - ret); + isert_err("login_rsp_dma mapping error: %d\n", ret); isert_conn->login_rsp_dma = 0; - goto out_req_dma_map; + goto out_free_login_rsp_buf; + } + + return 0; + +out_free_login_rsp_buf: + kfree(isert_conn->login_rsp_buf); +out_unmap_login_desc: + ib_dma_unmap_single(ib_dev, isert_conn->login_desc->dma_addr, + ISER_RX_SIZE, DMA_FROM_DEVICE); +out_free_login_desc: + kfree(isert_conn->login_desc); + return ret; +} + +static void +isert_set_nego_params(struct isert_conn *isert_conn, + struct rdma_conn_param *param) +{ + struct ib_device_attr *attr = &isert_conn->device->ib_device->attrs; + + /* Set max inflight RDMA READ requests */ + isert_conn->initiator_depth = min_t(u8, param->initiator_depth, + attr->max_qp_init_rd_atom); + isert_dbg("Using initiator_depth: %u\n", isert_conn->initiator_depth); + + if (param->private_data) { + u8 flags = *(u8 *)param->private_data; + + /* + * use remote invalidation if the both initiator + * and the HCA support it + */ + isert_conn->snd_w_inv = !(flags & ISER_SEND_W_INV_NOT_SUP) && + (attr->device_cap_flags & + IB_DEVICE_MEM_MGT_EXTENSIONS); + if (isert_conn->snd_w_inv) + isert_info("Using remote invalidation\n"); + } +} + +static void +isert_destroy_qp(struct isert_conn *isert_conn) +{ + ib_destroy_qp(isert_conn->qp); + ib_cq_pool_put(isert_conn->cq, isert_conn->cq_size); +} + +static int +isert_connect_request(struct rdma_cm_id *cma_id, struct rdma_cm_event *event) +{ + struct isert_np *isert_np = cma_id->context; + struct iscsi_np *np = isert_np->np; + struct isert_conn *isert_conn; + struct isert_device *device; + int ret = 0; + + spin_lock_bh(&np->np_thread_lock); + if (!np->enabled) { + spin_unlock_bh(&np->np_thread_lock); + isert_dbg("iscsi_np is not enabled, reject connect request\n"); + return rdma_reject(cma_id, NULL, 0, IB_CM_REJ_CONSUMER_DEFINED); } + spin_unlock_bh(&np->np_thread_lock); + + isert_dbg("cma_id: %p, portal: %p\n", + cma_id, cma_id->context); + + isert_conn = kzalloc(sizeof(struct isert_conn), GFP_KERNEL); + if (!isert_conn) + return -ENOMEM; + + isert_init_conn(isert_conn); + isert_conn->cm_id = cma_id; - device = isert_device_find_by_ib_dev(cma_id); + device = isert_device_get(cma_id); if (IS_ERR(device)) { ret = PTR_ERR(device); + goto out; + } + isert_conn->device = device; + + ret = isert_alloc_login_buf(isert_conn, cma_id->device); + if (ret) + goto out_conn_dev; + + isert_set_nego_params(isert_conn, &event->param.conn); + + isert_conn->qp = isert_create_qp(isert_conn, cma_id); + if (IS_ERR(isert_conn->qp)) { + ret = PTR_ERR(isert_conn->qp); goto out_rsp_dma_map; } - isert_conn->conn_device = device; - isert_conn->conn_pd = device->dev_pd; - isert_conn->conn_mr = device->dev_mr; + ret = isert_login_post_recv(isert_conn); + if (ret) + goto out_destroy_qp; - ret = isert_conn_setup_qp(isert_conn, cma_id); + ret = isert_rdma_accept(isert_conn); if (ret) - goto out_conn_dev; + goto out_destroy_qp; - mutex_lock(&isert_np->np_accept_mutex); - list_add_tail(&isert_np->np_accept_list, &isert_conn->conn_accept_node); - mutex_unlock(&isert_np->np_accept_mutex); + mutex_lock(&isert_np->mutex); + list_add_tail(&isert_conn->node, &isert_np->accepted); + mutex_unlock(&isert_np->mutex); - pr_debug("isert_connect_request() waking up np_accept_wq: %p\n", np); - wake_up(&isert_np->np_accept_wq); return 0; -out_conn_dev: - isert_device_try_release(device); +out_destroy_qp: + isert_destroy_qp(isert_conn); out_rsp_dma_map: - ib_dma_unmap_single(ib_dev, isert_conn->login_rsp_dma, - ISER_RX_LOGIN_SIZE, DMA_TO_DEVICE); -out_req_dma_map: - ib_dma_unmap_single(ib_dev, isert_conn->login_req_dma, - ISCSI_DEF_MAX_RECV_SEG_LEN, DMA_FROM_DEVICE); -out_login_buf: - kfree(isert_conn->login_buf); + isert_free_login_buf(isert_conn); +out_conn_dev: + isert_device_put(device); out: kfree(isert_conn); + rdma_reject(cma_id, NULL, 0, IB_CM_REJ_CONSUMER_DEFINED); return ret; } static void isert_connect_release(struct isert_conn *isert_conn) { - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; - struct isert_device *device = isert_conn->conn_device; - int cq_index; - - pr_debug("Entering isert_connect_release(): >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\n"); + struct isert_device *device = isert_conn->device; - if (isert_conn->conn_qp) { - cq_index = ((struct isert_cq_desc *) - isert_conn->conn_qp->recv_cq->cq_context)->cq_index; - pr_debug("isert_connect_release: cq_index: %d\n", cq_index); - isert_conn->conn_device->cq_active_qps[cq_index]--; + isert_dbg("conn %p\n", isert_conn); - rdma_destroy_qp(isert_conn->conn_cm_id); - } + BUG_ON(!device); isert_free_rx_descriptors(isert_conn); - rdma_destroy_id(isert_conn->conn_cm_id); + if (isert_conn->cm_id && + !isert_conn->dev_removed) + rdma_destroy_id(isert_conn->cm_id); - if (isert_conn->login_buf) { - ib_dma_unmap_single(ib_dev, isert_conn->login_rsp_dma, - ISER_RX_LOGIN_SIZE, DMA_TO_DEVICE); - ib_dma_unmap_single(ib_dev, isert_conn->login_req_dma, - ISCSI_DEF_MAX_RECV_SEG_LEN, - DMA_FROM_DEVICE); - kfree(isert_conn->login_buf); - } - kfree(isert_conn); + if (isert_conn->qp) + isert_destroy_qp(isert_conn); - if (device) - isert_device_try_release(device); + if (isert_conn->login_desc) + isert_free_login_buf(isert_conn); - pr_debug("Leaving isert_connect_release >>>>>>>>>>>>\n"); + isert_device_put(device); + + if (isert_conn->dev_removed) + wake_up_interruptible(&isert_conn->rem_wait); + else + kfree(isert_conn); } static void isert_connected_handler(struct rdma_cm_id *cma_id) { - return; + struct isert_conn *isert_conn = cma_id->qp->qp_context; + struct isert_np *isert_np = cma_id->context; + + isert_info("conn %p\n", isert_conn); + + mutex_lock(&isert_conn->mutex); + isert_conn->state = ISER_CONN_UP; + kref_get(&isert_conn->kref); + mutex_unlock(&isert_conn->mutex); + + mutex_lock(&isert_np->mutex); + list_move_tail(&isert_conn->node, &isert_np->pending); + mutex_unlock(&isert_np->mutex); + + isert_info("np %p: Allow accept_np to continue\n", isert_np); + up(&isert_np->sem); } static void -isert_release_conn_kref(struct kref *kref) +isert_release_kref(struct kref *kref) { struct isert_conn *isert_conn = container_of(kref, - struct isert_conn, conn_kref); + struct isert_conn, kref); - pr_debug("Calling isert_connect_release for final kref %s/%d\n", - current->comm, current->pid); + isert_info("conn %p final kref %s/%d\n", isert_conn, current->comm, + current->pid); isert_connect_release(isert_conn); } @@ -531,254 +545,382 @@ isert_release_conn_kref(struct kref *kref) static void isert_put_conn(struct isert_conn *isert_conn) { - kref_put(&isert_conn->conn_kref, isert_release_conn_kref); + kref_put(&isert_conn->kref, isert_release_kref); } static void -isert_disconnect_work(struct work_struct *work) +isert_handle_unbound_conn(struct isert_conn *isert_conn) { - struct isert_conn *isert_conn = container_of(work, - struct isert_conn, conn_logout_work); + struct isert_np *isert_np = isert_conn->cm_id->context; - pr_debug("isert_disconnect_work(): >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\n"); - mutex_lock(&isert_conn->conn_mutex); - isert_conn->state = ISER_CONN_DOWN; - - if (isert_conn->post_recv_buf_count == 0 && - atomic_read(&isert_conn->post_send_buf_count) == 0) { - pr_debug("Calling wake_up(&isert_conn->conn_wait);\n"); - mutex_unlock(&isert_conn->conn_mutex); - goto wake_up; - } - if (!isert_conn->conn_cm_id) { - mutex_unlock(&isert_conn->conn_mutex); + mutex_lock(&isert_np->mutex); + if (!list_empty(&isert_conn->node)) { + /* + * This means iscsi doesn't know this connection + * so schedule a cleanup ourselves + */ + list_del_init(&isert_conn->node); isert_put_conn(isert_conn); + queue_work(isert_release_wq, &isert_conn->release_work); + } + mutex_unlock(&isert_np->mutex); +} + +/** + * isert_conn_terminate() - Initiate connection termination + * @isert_conn: isert connection struct + * + * Notes: + * In case the connection state is BOUND, move state + * to TEMINATING and start teardown sequence (rdma_disconnect). + * In case the connection state is UP, complete flush as well. + * + * This routine must be called with mutex held. Thus it is + * safe to call multiple times. + */ +static void +isert_conn_terminate(struct isert_conn *isert_conn) +{ + int err; + + if (isert_conn->state >= ISER_CONN_TERMINATING) return; + + isert_info("Terminating conn %p state %d\n", + isert_conn, isert_conn->state); + isert_conn->state = ISER_CONN_TERMINATING; + err = rdma_disconnect(isert_conn->cm_id); + if (err) + isert_warn("Failed rdma_disconnect isert_conn %p\n", + isert_conn); +} + +static int +isert_np_cma_handler(struct isert_np *isert_np, + enum rdma_cm_event_type event) +{ + isert_dbg("%s (%d): isert np %p\n", + rdma_event_msg(event), event, isert_np); + + switch (event) { + case RDMA_CM_EVENT_DEVICE_REMOVAL: + isert_np->cm_id = NULL; + break; + case RDMA_CM_EVENT_ADDR_CHANGE: + isert_np->cm_id = isert_setup_id(isert_np); + if (IS_ERR(isert_np->cm_id)) { + isert_err("isert np %p setup id failed: %ld\n", + isert_np, PTR_ERR(isert_np->cm_id)); + isert_np->cm_id = NULL; + } + break; + default: + isert_err("isert np %p Unexpected event %d\n", + isert_np, event); } - if (!isert_conn->logout_posted) { - pr_debug("Calling rdma_disconnect for !logout_posted from" - " isert_disconnect_work\n"); - rdma_disconnect(isert_conn->conn_cm_id); - mutex_unlock(&isert_conn->conn_mutex); + + return -1; +} + +static int +isert_disconnected_handler(struct rdma_cm_id *cma_id, + enum rdma_cm_event_type event) +{ + struct isert_conn *isert_conn = cma_id->qp->qp_context; + + mutex_lock(&isert_conn->mutex); + switch (isert_conn->state) { + case ISER_CONN_TERMINATING: + break; + case ISER_CONN_UP: + isert_conn_terminate(isert_conn); + ib_drain_qp(isert_conn->qp); + isert_handle_unbound_conn(isert_conn); + break; + case ISER_CONN_BOUND: + case ISER_CONN_FULL_FEATURE: /* FALLTHRU */ iscsit_cause_connection_reinstatement(isert_conn->conn, 0); - goto wake_up; + break; + default: + isert_warn("conn %p terminating in state %d\n", + isert_conn, isert_conn->state); } - mutex_unlock(&isert_conn->conn_mutex); + mutex_unlock(&isert_conn->mutex); -wake_up: - wake_up(&isert_conn->conn_wait); - isert_put_conn(isert_conn); + return 0; } -static void -isert_disconnected_handler(struct rdma_cm_id *cma_id) +static int +isert_connect_error(struct rdma_cm_id *cma_id) { - struct isert_conn *isert_conn = (struct isert_conn *)cma_id->context; + struct isert_conn *isert_conn = cma_id->qp->qp_context; + struct isert_np *isert_np = cma_id->context; + + ib_drain_qp(isert_conn->qp); + + mutex_lock(&isert_np->mutex); + list_del_init(&isert_conn->node); + mutex_unlock(&isert_np->mutex); + isert_conn->cm_id = NULL; + isert_put_conn(isert_conn); - INIT_WORK(&isert_conn->conn_logout_work, isert_disconnect_work); - schedule_work(&isert_conn->conn_logout_work); + return -1; } static int isert_cma_handler(struct rdma_cm_id *cma_id, struct rdma_cm_event *event) { + struct isert_np *isert_np = cma_id->context; + struct isert_conn *isert_conn; int ret = 0; - pr_debug("isert_cma_handler: event %d status %d conn %p id %p\n", - event->event, event->status, cma_id->context, cma_id); + isert_info("%s (%d): status %d id %p np %p\n", + rdma_event_msg(event->event), event->event, + event->status, cma_id, cma_id->context); + + if (isert_np->cm_id == cma_id) + return isert_np_cma_handler(cma_id->context, event->event); switch (event->event) { case RDMA_CM_EVENT_CONNECT_REQUEST: - pr_debug("RDMA_CM_EVENT_CONNECT_REQUEST: >>>>>>>>>>>>>>>\n"); ret = isert_connect_request(cma_id, event); + if (ret) + isert_err("failed handle connect request %d\n", ret); break; case RDMA_CM_EVENT_ESTABLISHED: - pr_debug("RDMA_CM_EVENT_ESTABLISHED >>>>>>>>>>>>>>\n"); isert_connected_handler(cma_id); break; + case RDMA_CM_EVENT_ADDR_CHANGE: case RDMA_CM_EVENT_DISCONNECTED: - pr_debug("RDMA_CM_EVENT_DISCONNECTED: >>>>>>>>>>>>>>\n"); - isert_disconnected_handler(cma_id); + case RDMA_CM_EVENT_TIMEWAIT_EXIT: /* FALLTHRU */ + ret = isert_disconnected_handler(cma_id, event->event); break; case RDMA_CM_EVENT_DEVICE_REMOVAL: - case RDMA_CM_EVENT_ADDR_CHANGE: - break; + isert_conn = cma_id->qp->qp_context; + isert_conn->dev_removed = true; + isert_disconnected_handler(cma_id, event->event); + wait_event_interruptible(isert_conn->rem_wait, + isert_conn->state == ISER_CONN_DOWN); + kfree(isert_conn); + /* + * return non-zero from the callback to destroy + * the rdma cm id + */ + return 1; + case RDMA_CM_EVENT_REJECTED: + isert_info("Connection rejected: %s\n", + rdma_reject_msg(cma_id, event->status)); + fallthrough; + case RDMA_CM_EVENT_UNREACHABLE: case RDMA_CM_EVENT_CONNECT_ERROR: + ret = isert_connect_error(cma_id); + break; default: - pr_err("Unknown RDMA CMA event: %d\n", event->event); + isert_err("Unhandled RDMA CMA event: %d\n", event->event); break; } - if (ret != 0) { - pr_err("isert_cma_handler failed RDMA_CM_EVENT: 0x%08x %d\n", - event->event, ret); - dump_stack(); - } - return ret; } static int -isert_post_recv(struct isert_conn *isert_conn, u32 count) +isert_post_recvm(struct isert_conn *isert_conn, u32 count) { - struct ib_recv_wr *rx_wr, *rx_wr_failed; + struct ib_recv_wr *rx_wr; int i, ret; - unsigned int rx_head = isert_conn->conn_rx_desc_head; struct iser_rx_desc *rx_desc; - for (rx_wr = isert_conn->conn_rx_wr, i = 0; i < count; i++, rx_wr++) { - rx_desc = &isert_conn->conn_rx_descs[rx_head]; - rx_wr->wr_id = (unsigned long)rx_desc; - rx_wr->sg_list = &rx_desc->rx_sg; - rx_wr->num_sge = 1; - rx_wr->next = rx_wr + 1; - rx_head = (rx_head + 1) & (ISERT_QP_MAX_RECV_DTOS - 1); - } + for (rx_wr = isert_conn->rx_wr, i = 0; i < count; i++, rx_wr++) { + rx_desc = &isert_conn->rx_descs[i]; + rx_wr->wr_cqe = &rx_desc->rx_cqe; + rx_wr->sg_list = &rx_desc->rx_sg; + rx_wr->num_sge = 1; + rx_wr->next = rx_wr + 1; + rx_desc->in_use = false; + } rx_wr--; rx_wr->next = NULL; /* mark end of work requests list */ - isert_conn->post_recv_buf_count += count; - ret = ib_post_recv(isert_conn->conn_qp, isert_conn->conn_rx_wr, - &rx_wr_failed); - if (ret) { - pr_err("ib_post_recv() failed with ret: %d\n", ret); - isert_conn->post_recv_buf_count -= count; - } else { - pr_debug("isert_post_recv(): Posted %d RX buffers\n", count); - isert_conn->conn_rx_desc_head = rx_head; + ret = ib_post_recv(isert_conn->qp, isert_conn->rx_wr, NULL); + if (ret) + isert_err("ib_post_recv() failed with ret: %d\n", ret); + + return ret; +} + +static int +isert_post_recv(struct isert_conn *isert_conn, struct iser_rx_desc *rx_desc) +{ + struct ib_recv_wr rx_wr; + int ret; + + if (!rx_desc->in_use) { + /* + * if the descriptor is not in-use we already reposted it + * for recv, so just silently return + */ + return 0; } + + rx_desc->in_use = false; + rx_wr.wr_cqe = &rx_desc->rx_cqe; + rx_wr.sg_list = &rx_desc->rx_sg; + rx_wr.num_sge = 1; + rx_wr.next = NULL; + + ret = ib_post_recv(isert_conn->qp, &rx_wr, NULL); + if (ret) + isert_err("ib_post_recv() failed with ret: %d\n", ret); + return ret; } static int -isert_post_send(struct isert_conn *isert_conn, struct iser_tx_desc *tx_desc) +isert_login_post_send(struct isert_conn *isert_conn, struct iser_tx_desc *tx_desc) { - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; - struct ib_send_wr send_wr, *send_wr_failed; + struct ib_device *ib_dev = isert_conn->cm_id->device; + struct ib_send_wr send_wr; int ret; ib_dma_sync_single_for_device(ib_dev, tx_desc->dma_addr, ISER_HEADERS_LEN, DMA_TO_DEVICE); + tx_desc->tx_cqe.done = isert_login_send_done; + send_wr.next = NULL; - send_wr.wr_id = (unsigned long)tx_desc; + send_wr.wr_cqe = &tx_desc->tx_cqe; send_wr.sg_list = tx_desc->tx_sg; send_wr.num_sge = tx_desc->num_sge; send_wr.opcode = IB_WR_SEND; send_wr.send_flags = IB_SEND_SIGNALED; - atomic_inc(&isert_conn->post_send_buf_count); - - ret = ib_post_send(isert_conn->conn_qp, &send_wr, &send_wr_failed); - if (ret) { - pr_err("ib_post_send() failed, ret: %d\n", ret); - atomic_dec(&isert_conn->post_send_buf_count); - } + ret = ib_post_send(isert_conn->qp, &send_wr, NULL); + if (ret) + isert_err("ib_post_send() failed, ret: %d\n", ret); return ret; } static void +__isert_create_send_desc(struct isert_device *device, + struct iser_tx_desc *tx_desc) +{ + + memset(&tx_desc->iser_header, 0, sizeof(struct iser_ctrl)); + tx_desc->iser_header.flags = ISCSI_CTRL; + + tx_desc->num_sge = 1; + + if (tx_desc->tx_sg[0].lkey != device->pd->local_dma_lkey) { + tx_desc->tx_sg[0].lkey = device->pd->local_dma_lkey; + isert_dbg("tx_desc %p lkey mismatch, fixing\n", tx_desc); + } +} + +static void isert_create_send_desc(struct isert_conn *isert_conn, struct isert_cmd *isert_cmd, struct iser_tx_desc *tx_desc) { - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; + struct isert_device *device = isert_conn->device; + struct ib_device *ib_dev = device->ib_device; ib_dma_sync_single_for_cpu(ib_dev, tx_desc->dma_addr, ISER_HEADERS_LEN, DMA_TO_DEVICE); - memset(&tx_desc->iser_header, 0, sizeof(struct iser_hdr)); - tx_desc->iser_header.flags = ISER_VER; - - tx_desc->num_sge = 1; - tx_desc->isert_cmd = isert_cmd; - - if (tx_desc->tx_sg[0].lkey != isert_conn->conn_mr->lkey) { - tx_desc->tx_sg[0].lkey = isert_conn->conn_mr->lkey; - pr_debug("tx_desc %p lkey mismatch, fixing\n", tx_desc); - } + __isert_create_send_desc(device, tx_desc); } static int isert_init_tx_hdrs(struct isert_conn *isert_conn, struct iser_tx_desc *tx_desc) { - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; + struct isert_device *device = isert_conn->device; + struct ib_device *ib_dev = device->ib_device; u64 dma_addr; dma_addr = ib_dma_map_single(ib_dev, (void *)tx_desc, ISER_HEADERS_LEN, DMA_TO_DEVICE); if (ib_dma_mapping_error(ib_dev, dma_addr)) { - pr_err("ib_dma_mapping_error() failed\n"); + isert_err("ib_dma_mapping_error() failed\n"); return -ENOMEM; } tx_desc->dma_addr = dma_addr; tx_desc->tx_sg[0].addr = tx_desc->dma_addr; tx_desc->tx_sg[0].length = ISER_HEADERS_LEN; - tx_desc->tx_sg[0].lkey = isert_conn->conn_mr->lkey; + tx_desc->tx_sg[0].lkey = device->pd->local_dma_lkey; - pr_debug("isert_init_tx_hdrs: Setup tx_sg[0].addr: 0x%llx length: %u" - " lkey: 0x%08x\n", tx_desc->tx_sg[0].addr, - tx_desc->tx_sg[0].length, tx_desc->tx_sg[0].lkey); + isert_dbg("Setup tx_sg[0].addr: 0x%llx length: %u lkey: 0x%x\n", + tx_desc->tx_sg[0].addr, tx_desc->tx_sg[0].length, + tx_desc->tx_sg[0].lkey); return 0; } static void -isert_init_send_wr(struct isert_cmd *isert_cmd, struct ib_send_wr *send_wr) +isert_init_send_wr(struct isert_conn *isert_conn, struct isert_cmd *isert_cmd, + struct ib_send_wr *send_wr) { - isert_cmd->rdma_wr.iser_ib_op = ISER_IB_SEND; - send_wr->wr_id = (unsigned long)&isert_cmd->tx_desc; - send_wr->opcode = IB_WR_SEND; - send_wr->send_flags = IB_SEND_SIGNALED; - send_wr->sg_list = &isert_cmd->tx_desc.tx_sg[0]; + struct iser_tx_desc *tx_desc = &isert_cmd->tx_desc; + + tx_desc->tx_cqe.done = isert_send_done; + send_wr->wr_cqe = &tx_desc->tx_cqe; + + if (isert_conn->snd_w_inv && isert_cmd->inv_rkey) { + send_wr->opcode = IB_WR_SEND_WITH_INV; + send_wr->ex.invalidate_rkey = isert_cmd->inv_rkey; + } else { + send_wr->opcode = IB_WR_SEND; + } + + send_wr->sg_list = &tx_desc->tx_sg[0]; send_wr->num_sge = isert_cmd->tx_desc.num_sge; + send_wr->send_flags = IB_SEND_SIGNALED; } static int -isert_rdma_post_recvl(struct isert_conn *isert_conn) +isert_login_post_recv(struct isert_conn *isert_conn) { - struct ib_recv_wr rx_wr, *rx_wr_fail; + struct ib_recv_wr rx_wr; struct ib_sge sge; int ret; memset(&sge, 0, sizeof(struct ib_sge)); - sge.addr = isert_conn->login_req_dma; - sge.length = ISER_RX_LOGIN_SIZE; - sge.lkey = isert_conn->conn_mr->lkey; + sge.addr = isert_conn->login_desc->dma_addr + + isert_get_hdr_offset(isert_conn->login_desc); + sge.length = ISER_RX_PAYLOAD_SIZE; + sge.lkey = isert_conn->device->pd->local_dma_lkey; - pr_debug("Setup sge: addr: %llx length: %d 0x%08x\n", + isert_dbg("Setup sge: addr: %llx length: %d 0x%08x\n", sge.addr, sge.length, sge.lkey); + isert_conn->login_desc->rx_cqe.done = isert_login_recv_done; + memset(&rx_wr, 0, sizeof(struct ib_recv_wr)); - rx_wr.wr_id = (unsigned long)isert_conn->login_req_buf; + rx_wr.wr_cqe = &isert_conn->login_desc->rx_cqe; rx_wr.sg_list = &sge; rx_wr.num_sge = 1; - isert_conn->post_recv_buf_count++; - ret = ib_post_recv(isert_conn->conn_qp, &rx_wr, &rx_wr_fail); - if (ret) { - pr_err("ib_post_recv() failed: %d\n", ret); - isert_conn->post_recv_buf_count--; - } + ret = ib_post_recv(isert_conn->qp, &rx_wr, NULL); + if (ret) + isert_err("ib_post_recv() failed: %d\n", ret); - pr_debug("ib_post_recv(): returned success >>>>>>>>>>>>>>>>>>>>>>>>\n"); return ret; } static int -isert_put_login_tx(struct iscsi_conn *conn, struct iscsi_login *login, +isert_put_login_tx(struct iscsit_conn *conn, struct iscsi_login *login, u32 length) { struct isert_conn *isert_conn = conn->context; - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; - struct iser_tx_desc *tx_desc = &isert_conn->conn_login_tx_desc; + struct isert_device *device = isert_conn->device; + struct ib_device *ib_dev = device->ib_device; + struct iser_tx_desc *tx_desc = &isert_conn->login_tx_desc; int ret; - isert_create_send_desc(isert_conn, NULL, tx_desc); + __isert_create_send_desc(device, tx_desc); memcpy(&tx_desc->iscsi_header, &login->rsp[0], sizeof(struct iscsi_hdr)); @@ -798,7 +940,7 @@ isert_put_login_tx(struct iscsi_conn *conn, struct iscsi_login *login, tx_dsg->addr = isert_conn->login_rsp_dma; tx_dsg->length = length; - tx_dsg->lkey = isert_conn->conn_mr->lkey; + tx_dsg->lkey = isert_conn->device->pd->local_dma_lkey; tx_desc->num_sge = 2; } if (!login->login_failed) { @@ -807,20 +949,24 @@ isert_put_login_tx(struct iscsi_conn *conn, struct iscsi_login *login, if (ret) return ret; - ret = isert_post_recv(isert_conn, ISERT_MIN_POSTED_RX); + ret = isert_post_recvm(isert_conn, + ISERT_QP_MAX_RECV_DTOS); if (ret) return ret; - isert_conn->state = ISER_CONN_UP; + /* Now we are in FULL_FEATURE phase */ + mutex_lock(&isert_conn->mutex); + isert_conn->state = ISER_CONN_FULL_FEATURE; + mutex_unlock(&isert_conn->mutex); goto post_send; } - ret = isert_rdma_post_recvl(isert_conn); + ret = isert_login_post_recv(isert_conn); if (ret) return ret; } post_send: - ret = isert_post_send(isert_conn, tx_desc); + ret = isert_login_post_send(isert_conn, tx_desc); if (ret) return ret; @@ -828,30 +974,28 @@ post_send: } static void -isert_rx_login_req(struct iser_rx_desc *rx_desc, int rx_buflen, - struct isert_conn *isert_conn) +isert_rx_login_req(struct isert_conn *isert_conn) { - struct iscsi_conn *conn = isert_conn->conn; + struct iser_rx_desc *rx_desc = isert_conn->login_desc; + int rx_buflen = isert_conn->login_req_len; + struct iscsit_conn *conn = isert_conn->conn; struct iscsi_login *login = conn->conn_login; int size; - if (!login) { - pr_err("conn->conn_login is NULL\n"); - dump_stack(); - return; - } + isert_info("conn %p\n", isert_conn); + + WARN_ON_ONCE(!login); if (login->first_request) { struct iscsi_login_req *login_req = - (struct iscsi_login_req *)&rx_desc->iscsi_header; + (struct iscsi_login_req *)isert_get_iscsi_hdr(rx_desc); /* * Setup the initial iscsi_login values from the leading * login request PDU. */ login->leading_connection = (!login_req->tsih) ? 1 : 0; - login->current_stage = - (login_req->flags & ISCSI_FLAG_LOGIN_CURRENT_STAGE_MASK) - >> 2; + login->current_stage = ISCSI_LOGIN_CURRENT_STAGE( + login_req->flags); login->version_min = login_req->min_version; login->version_max = login_req->max_version; memcpy(login->isid, login_req->isid, 6); @@ -862,58 +1006,51 @@ isert_rx_login_req(struct iser_rx_desc *rx_desc, int rx_buflen, login->tsih = be16_to_cpu(login_req->tsih); } - memcpy(&login->req[0], (void *)&rx_desc->iscsi_header, ISCSI_HDR_LEN); + memcpy(&login->req[0], isert_get_iscsi_hdr(rx_desc), ISCSI_HDR_LEN); size = min(rx_buflen, MAX_KEY_VALUE_PAIRS); - pr_debug("Using login payload size: %d, rx_buflen: %d MAX_KEY_VALUE_PAIRS: %d\n", - size, rx_buflen, MAX_KEY_VALUE_PAIRS); - memcpy(login->req_buf, &rx_desc->data[0], size); + isert_dbg("Using login payload size: %d, rx_buflen: %d " + "MAX_KEY_VALUE_PAIRS: %d\n", size, rx_buflen, + MAX_KEY_VALUE_PAIRS); + memcpy(login->req_buf, isert_get_data(rx_desc), size); - complete(&isert_conn->conn_login_comp); -} - -static void -isert_release_cmd(struct iscsi_cmd *cmd) -{ - struct isert_cmd *isert_cmd = container_of(cmd, struct isert_cmd, - iscsi_cmd); - - pr_debug("Entering isert_release_cmd %p >>>>>>>>>>>>>>>.\n", isert_cmd); - - kfree(cmd->buf_ptr); - kfree(cmd->tmr_req); - - kmem_cache_free(isert_cmd_cache, isert_cmd); + if (login->first_request) { + complete(&isert_conn->login_comp); + return; + } + queue_delayed_work(isert_login_wq, &conn->login_work, 0); } -static struct iscsi_cmd -*isert_alloc_cmd(struct iscsi_conn *conn, gfp_t gfp) +static struct iscsit_cmd +*isert_allocate_cmd(struct iscsit_conn *conn, struct iser_rx_desc *rx_desc) { - struct isert_conn *isert_conn = (struct isert_conn *)conn->context; + struct isert_conn *isert_conn = conn->context; struct isert_cmd *isert_cmd; + struct iscsit_cmd *cmd; - isert_cmd = kmem_cache_zalloc(isert_cmd_cache, gfp); - if (!isert_cmd) { - pr_err("Unable to allocate isert_cmd\n"); + cmd = iscsit_allocate_cmd(conn, TASK_INTERRUPTIBLE); + if (!cmd) { + isert_err("Unable to allocate iscsit_cmd + isert_cmd\n"); return NULL; } + isert_cmd = iscsit_priv_cmd(cmd); isert_cmd->conn = isert_conn; - isert_cmd->iscsi_cmd.release_cmd = &isert_release_cmd; + isert_cmd->iscsit_cmd = cmd; + isert_cmd->rx_desc = rx_desc; - return &isert_cmd->iscsi_cmd; + return cmd; } static int isert_handle_scsi_cmd(struct isert_conn *isert_conn, - struct isert_cmd *isert_cmd, struct iser_rx_desc *rx_desc, - unsigned char *buf) + struct isert_cmd *isert_cmd, struct iscsit_cmd *cmd, + struct iser_rx_desc *rx_desc, unsigned char *buf) { - struct iscsi_cmd *cmd = &isert_cmd->iscsi_cmd; - struct iscsi_conn *conn = isert_conn->conn; + struct iscsit_conn *conn = isert_conn->conn; struct iscsi_scsi_req *hdr = (struct iscsi_scsi_req *)buf; - struct scatterlist *sg; int imm_data, imm_data_len, unsol_data, sg_nents, rc; bool dump_payload = false; + unsigned int data_len; rc = iscsit_setup_scsi_cmd(conn, cmd, buf); if (rc < 0) @@ -922,7 +1059,10 @@ isert_handle_scsi_cmd(struct isert_conn *isert_conn, imm_data = cmd->immediate_data; imm_data_len = cmd->first_burst_len; unsol_data = cmd->unsolicited_data; + data_len = cmd->se_cmd.data_length; + if (imm_data && imm_data_len == data_len) + cmd->se_cmd.se_cmd_flags |= SCF_PASSTHROUGH_SG_TO_MEM_NOALLOC; rc = iscsit_process_scsi_cmd(conn, cmd, hdr); if (rc < 0) { return 0; @@ -934,13 +1074,21 @@ isert_handle_scsi_cmd(struct isert_conn *isert_conn, if (!imm_data) return 0; - sg = &cmd->se_cmd.t_data_sg[0]; - sg_nents = max(1UL, DIV_ROUND_UP(imm_data_len, PAGE_SIZE)); - - pr_debug("Copying Immediate SG: %p sg_nents: %u from %p imm_data_len: %d\n", - sg, sg_nents, &rx_desc->data[0], imm_data_len); - - sg_copy_from_buffer(sg, sg_nents, &rx_desc->data[0], imm_data_len); + if (imm_data_len != data_len) { + sg_nents = max(1UL, DIV_ROUND_UP(imm_data_len, PAGE_SIZE)); + sg_copy_from_buffer(cmd->se_cmd.t_data_sg, sg_nents, + isert_get_data(rx_desc), imm_data_len); + isert_dbg("Copy Immediate sg_nents: %u imm_data_len: %d\n", + sg_nents, imm_data_len); + } else { + sg_init_table(&isert_cmd->sg, 1); + cmd->se_cmd.t_data_sg = &isert_cmd->sg; + cmd->se_cmd.t_data_nents = 1; + sg_set_buf(&isert_cmd->sg, isert_get_data(rx_desc), + imm_data_len); + isert_dbg("Transfer Immediate imm_data_len: %d\n", + imm_data_len); + } cmd->write_data_done += imm_data_len; @@ -954,8 +1102,10 @@ isert_handle_scsi_cmd(struct isert_conn *isert_conn, sequence_cmd: rc = iscsit_sequence_cmd(conn, cmd, buf, hdr->cmdsn); - if (!rc && dump_payload == false && unsol_data) - iscsit_set_unsoliticed_dataout(cmd); + if (!rc && !dump_payload && unsol_data) + iscsit_set_unsolicited_dataout(cmd); + else if (dump_payload && imm_data) + target_put_sess_cmd(&cmd->se_cmd); return 0; } @@ -965,8 +1115,8 @@ isert_handle_iscsi_dataout(struct isert_conn *isert_conn, struct iser_rx_desc *rx_desc, unsigned char *buf) { struct scatterlist *sg_start; - struct iscsi_conn *conn = isert_conn->conn; - struct iscsi_cmd *cmd = NULL; + struct iscsit_conn *conn = isert_conn->conn; + struct iscsit_cmd *cmd = NULL; struct iscsi_data *hdr = (struct iscsi_data *)buf; u32 unsol_data_len = ntoh24(hdr->dlength); int rc, sg_nents, sg_off, page_off; @@ -980,13 +1130,15 @@ isert_handle_iscsi_dataout(struct isert_conn *isert_conn, * FIXME: Unexpected unsolicited_data out */ if (!cmd->unsolicited_data) { - pr_err("Received unexpected solicited data payload\n"); + isert_err("Received unexpected solicited data payload\n"); dump_stack(); return -1; } - pr_debug("Unsolicited DataOut unsol_data_len: %u, write_data_done: %u, data_length: %u\n", - unsol_data_len, cmd->write_data_done, cmd->se_cmd.data_length); + isert_dbg("Unsolicited DataOut unsol_data_len: %u, " + "write_data_done: %u, data_length: %u\n", + unsol_data_len, cmd->write_data_done, + cmd->se_cmd.data_length); sg_off = cmd->write_data_done / PAGE_SIZE; sg_start = &cmd->se_cmd.t_data_sg[sg_off]; @@ -996,29 +1148,34 @@ isert_handle_iscsi_dataout(struct isert_conn *isert_conn, * FIXME: Non page-aligned unsolicited_data out */ if (page_off) { - pr_err("Received unexpected non-page aligned data payload\n"); + isert_err("unexpected non-page aligned data payload\n"); dump_stack(); return -1; } - pr_debug("Copying DataOut: sg_start: %p, sg_off: %u sg_nents: %u from %p %u\n", - sg_start, sg_off, sg_nents, &rx_desc->data[0], unsol_data_len); + isert_dbg("Copying DataOut: sg_start: %p, sg_off: %u " + "sg_nents: %u from %p %u\n", sg_start, sg_off, + sg_nents, isert_get_data(rx_desc), unsol_data_len); - sg_copy_from_buffer(sg_start, sg_nents, &rx_desc->data[0], + sg_copy_from_buffer(sg_start, sg_nents, isert_get_data(rx_desc), unsol_data_len); rc = iscsit_check_dataout_payload(cmd, hdr, false); if (rc < 0) return rc; - return 0; + /* + * multiple data-outs on the same command can arrive - + * so post the buffer before hand + */ + return isert_post_recv(isert_conn, rx_desc); } static int isert_handle_nop_out(struct isert_conn *isert_conn, struct isert_cmd *isert_cmd, - struct iser_rx_desc *rx_desc, unsigned char *buf) + struct iscsit_cmd *cmd, struct iser_rx_desc *rx_desc, + unsigned char *buf) { - struct iscsi_cmd *cmd = &isert_cmd->iscsi_cmd; - struct iscsi_conn *conn = isert_conn->conn; + struct iscsit_conn *conn = isert_conn->conn; struct iscsi_nopout *hdr = (struct iscsi_nopout *)buf; int rc; @@ -1034,27 +1191,26 @@ isert_handle_nop_out(struct isert_conn *isert_conn, struct isert_cmd *isert_cmd, static int isert_handle_text_cmd(struct isert_conn *isert_conn, struct isert_cmd *isert_cmd, - struct iser_rx_desc *rx_desc, struct iscsi_text *hdr) + struct iscsit_cmd *cmd, struct iser_rx_desc *rx_desc, + struct iscsi_text *hdr) { - struct iscsi_cmd *cmd = &isert_cmd->iscsi_cmd; - struct iscsi_conn *conn = isert_conn->conn; + struct iscsit_conn *conn = isert_conn->conn; u32 payload_length = ntoh24(hdr->dlength); int rc; - unsigned char *text_in; + unsigned char *text_in = NULL; rc = iscsit_setup_text_cmd(conn, cmd, hdr); if (rc < 0) return rc; - text_in = kzalloc(payload_length, GFP_KERNEL); - if (!text_in) { - pr_err("Unable to allocate text_in of payload_length: %u\n", - payload_length); - return -ENOMEM; + if (payload_length) { + text_in = kzalloc(payload_length, GFP_KERNEL); + if (!text_in) + return -ENOMEM; } cmd->text_in_ptr = text_in; - memcpy(cmd->text_in_ptr, &rx_desc->data[0], payload_length); + memcpy(cmd->text_in_ptr, isert_get_data(rx_desc), payload_length); return iscsit_process_text_cmd(conn, cmd, hdr); } @@ -1064,43 +1220,43 @@ isert_rx_opcode(struct isert_conn *isert_conn, struct iser_rx_desc *rx_desc, uint32_t read_stag, uint64_t read_va, uint32_t write_stag, uint64_t write_va) { - struct iscsi_hdr *hdr = &rx_desc->iscsi_header; - struct iscsi_conn *conn = isert_conn->conn; - struct iscsi_session *sess = conn->sess; - struct iscsi_cmd *cmd; + struct iscsi_hdr *hdr = isert_get_iscsi_hdr(rx_desc); + struct iscsit_conn *conn = isert_conn->conn; + struct iscsit_cmd *cmd; struct isert_cmd *isert_cmd; int ret = -EINVAL; u8 opcode = (hdr->opcode & ISCSI_OPCODE_MASK); - if (sess->sess_ops->SessionType && + if (conn->sess->sess_ops->SessionType && (!(opcode & ISCSI_OP_TEXT) || !(opcode & ISCSI_OP_LOGOUT))) { - pr_err("Got illegal opcode: 0x%02x in SessionType=Discovery," - " ignoring\n", opcode); + isert_err("Got illegal opcode: 0x%02x in SessionType=Discovery," + " ignoring\n", opcode); return 0; } switch (opcode) { case ISCSI_OP_SCSI_CMD: - cmd = iscsit_allocate_cmd(conn, GFP_KERNEL); + cmd = isert_allocate_cmd(conn, rx_desc); if (!cmd) break; - isert_cmd = container_of(cmd, struct isert_cmd, iscsi_cmd); + isert_cmd = iscsit_priv_cmd(cmd); isert_cmd->read_stag = read_stag; isert_cmd->read_va = read_va; isert_cmd->write_stag = write_stag; isert_cmd->write_va = write_va; + isert_cmd->inv_rkey = read_stag ? read_stag : write_stag; - ret = isert_handle_scsi_cmd(isert_conn, isert_cmd, + ret = isert_handle_scsi_cmd(isert_conn, isert_cmd, cmd, rx_desc, (unsigned char *)hdr); break; case ISCSI_OP_NOOP_OUT: - cmd = iscsit_allocate_cmd(conn, GFP_KERNEL); + cmd = isert_allocate_cmd(conn, rx_desc); if (!cmd) break; - isert_cmd = container_of(cmd, struct isert_cmd, iscsi_cmd); - ret = isert_handle_nop_out(isert_conn, isert_cmd, + isert_cmd = iscsit_priv_cmd(cmd); + ret = isert_handle_nop_out(isert_conn, isert_cmd, cmd, rx_desc, (unsigned char *)hdr); break; case ISCSI_OP_SCSI_DATA_OUT: @@ -1108,7 +1264,7 @@ isert_rx_opcode(struct isert_conn *isert_conn, struct iser_rx_desc *rx_desc, (unsigned char *)hdr); break; case ISCSI_OP_SCSI_TMFUNC: - cmd = iscsit_allocate_cmd(conn, GFP_KERNEL); + cmd = isert_allocate_cmd(conn, rx_desc); if (!cmd) break; @@ -1116,27 +1272,27 @@ isert_rx_opcode(struct isert_conn *isert_conn, struct iser_rx_desc *rx_desc, (unsigned char *)hdr); break; case ISCSI_OP_LOGOUT: - cmd = iscsit_allocate_cmd(conn, GFP_KERNEL); + cmd = isert_allocate_cmd(conn, rx_desc); if (!cmd) break; ret = iscsit_handle_logout_cmd(conn, cmd, (unsigned char *)hdr); - if (ret > 0) - wait_for_completion_timeout(&conn->conn_logout_comp, - SECONDS_FOR_LOGOUT_COMP * - HZ); break; case ISCSI_OP_TEXT: - cmd = iscsit_allocate_cmd(conn, GFP_KERNEL); + if (be32_to_cpu(hdr->ttt) != 0xFFFFFFFF) + cmd = iscsit_find_cmd_from_itt(conn, hdr->itt); + else + cmd = isert_allocate_cmd(conn, rx_desc); + if (!cmd) break; - isert_cmd = container_of(cmd, struct isert_cmd, iscsi_cmd); - ret = isert_handle_text_cmd(isert_conn, isert_cmd, + isert_cmd = iscsit_priv_cmd(cmd); + ret = isert_handle_text_cmd(isert_conn, isert_cmd, cmd, rx_desc, (struct iscsi_text *)hdr); break; default: - pr_err("Got unknown iSCSI OpCode: 0x%02x\n", opcode); + isert_err("Got unknown iSCSI OpCode: 0x%02x\n", opcode); dump_stack(); break; } @@ -1145,144 +1301,169 @@ isert_rx_opcode(struct isert_conn *isert_conn, struct iser_rx_desc *rx_desc, } static void -isert_rx_do_work(struct iser_rx_desc *rx_desc, struct isert_conn *isert_conn) +isert_print_wc(struct ib_wc *wc, const char *type) { - struct iser_hdr *iser_hdr = &rx_desc->iser_header; + if (wc->status != IB_WC_WR_FLUSH_ERR) + isert_err("%s failure: %s (%d) vend_err %x\n", type, + ib_wc_status_msg(wc->status), wc->status, + wc->vendor_err); + else + isert_dbg("%s failure: %s (%d)\n", type, + ib_wc_status_msg(wc->status), wc->status); +} + +static void +isert_recv_done(struct ib_cq *cq, struct ib_wc *wc) +{ + struct isert_conn *isert_conn = wc->qp->qp_context; + struct ib_device *ib_dev = isert_conn->cm_id->device; + struct iser_rx_desc *rx_desc = cqe_to_rx_desc(wc->wr_cqe); + struct iscsi_hdr *hdr = isert_get_iscsi_hdr(rx_desc); + struct iser_ctrl *iser_ctrl = isert_get_iser_hdr(rx_desc); uint64_t read_va = 0, write_va = 0; uint32_t read_stag = 0, write_stag = 0; - int rc; - switch (iser_hdr->flags & 0xF0) { + if (unlikely(wc->status != IB_WC_SUCCESS)) { + isert_print_wc(wc, "recv"); + if (wc->status != IB_WC_WR_FLUSH_ERR) + iscsit_cause_connection_reinstatement(isert_conn->conn, 0); + return; + } + + rx_desc->in_use = true; + + ib_dma_sync_single_for_cpu(ib_dev, rx_desc->dma_addr, + ISER_RX_SIZE, DMA_FROM_DEVICE); + + isert_dbg("DMA: 0x%llx, iSCSI opcode: 0x%02x, ITT: 0x%08x, flags: 0x%02x dlen: %d\n", + rx_desc->dma_addr, hdr->opcode, hdr->itt, hdr->flags, + (int)(wc->byte_len - ISER_HEADERS_LEN)); + + switch (iser_ctrl->flags & 0xF0) { case ISCSI_CTRL: - if (iser_hdr->flags & ISER_RSV) { - read_stag = be32_to_cpu(iser_hdr->read_stag); - read_va = be64_to_cpu(iser_hdr->read_va); - pr_debug("ISER_RSV: read_stag: 0x%08x read_va: 0x%16llx\n", - read_stag, (unsigned long long)read_va); + if (iser_ctrl->flags & ISER_RSV) { + read_stag = be32_to_cpu(iser_ctrl->read_stag); + read_va = be64_to_cpu(iser_ctrl->read_va); + isert_dbg("ISER_RSV: read_stag: 0x%x read_va: 0x%llx\n", + read_stag, (unsigned long long)read_va); } - if (iser_hdr->flags & ISER_WSV) { - write_stag = be32_to_cpu(iser_hdr->write_stag); - write_va = be64_to_cpu(iser_hdr->write_va); - pr_debug("ISER_WSV: write__stag: 0x%08x write_va: 0x%16llx\n", - write_stag, (unsigned long long)write_va); + if (iser_ctrl->flags & ISER_WSV) { + write_stag = be32_to_cpu(iser_ctrl->write_stag); + write_va = be64_to_cpu(iser_ctrl->write_va); + isert_dbg("ISER_WSV: write_stag: 0x%x write_va: 0x%llx\n", + write_stag, (unsigned long long)write_va); } - pr_debug("ISER ISCSI_CTRL PDU\n"); + isert_dbg("ISER ISCSI_CTRL PDU\n"); break; case ISER_HELLO: - pr_err("iSER Hello message\n"); + isert_err("iSER Hello message\n"); break; default: - pr_warn("Unknown iSER hdr flags: 0x%02x\n", iser_hdr->flags); + isert_warn("Unknown iSER hdr flags: 0x%02x\n", iser_ctrl->flags); break; } - rc = isert_rx_opcode(isert_conn, rx_desc, - read_stag, read_va, write_stag, write_va); + isert_rx_opcode(isert_conn, rx_desc, + read_stag, read_va, write_stag, write_va); + + ib_dma_sync_single_for_device(ib_dev, rx_desc->dma_addr, + ISER_RX_SIZE, DMA_FROM_DEVICE); } static void -isert_rx_completion(struct iser_rx_desc *desc, struct isert_conn *isert_conn, - unsigned long xfer_len) -{ - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; - struct iscsi_hdr *hdr; - u64 rx_dma; - int rx_buflen, outstanding; - - if ((char *)desc == isert_conn->login_req_buf) { - rx_dma = isert_conn->login_req_dma; - rx_buflen = ISER_RX_LOGIN_SIZE; - pr_debug("ISER login_buf: Using rx_dma: 0x%llx, rx_buflen: %d\n", - rx_dma, rx_buflen); - } else { - rx_dma = desc->dma_addr; - rx_buflen = ISER_RX_PAYLOAD_SIZE; - pr_debug("ISER req_buf: Using rx_dma: 0x%llx, rx_buflen: %d\n", - rx_dma, rx_buflen); - } +isert_login_recv_done(struct ib_cq *cq, struct ib_wc *wc) +{ + struct isert_conn *isert_conn = wc->qp->qp_context; + struct ib_device *ib_dev = isert_conn->device->ib_device; - ib_dma_sync_single_for_cpu(ib_dev, rx_dma, rx_buflen, DMA_FROM_DEVICE); + if (unlikely(wc->status != IB_WC_SUCCESS)) { + isert_print_wc(wc, "login recv"); + return; + } - hdr = &desc->iscsi_header; - pr_debug("iSCSI opcode: 0x%02x, ITT: 0x%08x, flags: 0x%02x dlen: %d\n", - hdr->opcode, hdr->itt, hdr->flags, - (int)(xfer_len - ISER_HEADERS_LEN)); + ib_dma_sync_single_for_cpu(ib_dev, isert_conn->login_desc->dma_addr, + ISER_RX_SIZE, DMA_FROM_DEVICE); - if ((char *)desc == isert_conn->login_req_buf) - isert_rx_login_req(desc, xfer_len - ISER_HEADERS_LEN, - isert_conn); - else - isert_rx_do_work(desc, isert_conn); + isert_conn->login_req_len = wc->byte_len - ISER_HEADERS_LEN; - ib_dma_sync_single_for_device(ib_dev, rx_dma, rx_buflen, - DMA_FROM_DEVICE); + if (isert_conn->conn) { + struct iscsi_login *login = isert_conn->conn->conn_login; - isert_conn->post_recv_buf_count--; - pr_debug("iSERT: Decremented post_recv_buf_count: %d\n", - isert_conn->post_recv_buf_count); + if (login && !login->first_request) + isert_rx_login_req(isert_conn); + } - if ((char *)desc == isert_conn->login_req_buf) - return; + mutex_lock(&isert_conn->mutex); + complete(&isert_conn->login_req_comp); + mutex_unlock(&isert_conn->mutex); - outstanding = isert_conn->post_recv_buf_count; - if (outstanding + ISERT_MIN_POSTED_RX <= ISERT_QP_MAX_RECV_DTOS) { - int err, count = min(ISERT_QP_MAX_RECV_DTOS - outstanding, - ISERT_MIN_POSTED_RX); - err = isert_post_recv(isert_conn, count); - if (err) { - pr_err("isert_post_recv() count: %d failed, %d\n", - count, err); - } - } + ib_dma_sync_single_for_device(ib_dev, isert_conn->login_desc->dma_addr, + ISER_RX_SIZE, DMA_FROM_DEVICE); } static void -isert_unmap_cmd(struct isert_cmd *isert_cmd, struct isert_conn *isert_conn) +isert_rdma_rw_ctx_destroy(struct isert_cmd *cmd, struct isert_conn *conn) { - struct isert_rdma_wr *wr = &isert_cmd->rdma_wr; - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; + struct se_cmd *se_cmd = &cmd->iscsit_cmd->se_cmd; + enum dma_data_direction dir = target_reverse_dma_direction(se_cmd); - pr_debug("isert_unmap_cmd >>>>>>>>>>>>>>>>>>>>>>>\n"); + if (!cmd->rw.nr_ops) + return; - if (wr->sge) { - ib_dma_unmap_sg(ib_dev, wr->sge, wr->num_sge, DMA_TO_DEVICE); - wr->sge = NULL; + if (isert_prot_cmd(conn, se_cmd)) { + rdma_rw_ctx_destroy_signature(&cmd->rw, conn->qp, + conn->cm_id->port_num, se_cmd->t_data_sg, + se_cmd->t_data_nents, se_cmd->t_prot_sg, + se_cmd->t_prot_nents, dir); + } else { + rdma_rw_ctx_destroy(&cmd->rw, conn->qp, conn->cm_id->port_num, + se_cmd->t_data_sg, se_cmd->t_data_nents, dir); } - kfree(wr->send_wr); - wr->send_wr = NULL; - - kfree(isert_cmd->ib_sge); - isert_cmd->ib_sge = NULL; + cmd->rw.nr_ops = 0; } static void -isert_put_cmd(struct isert_cmd *isert_cmd) +isert_put_cmd(struct isert_cmd *isert_cmd, bool comp_err) { - struct iscsi_cmd *cmd = &isert_cmd->iscsi_cmd; + struct iscsit_cmd *cmd = isert_cmd->iscsit_cmd; struct isert_conn *isert_conn = isert_cmd->conn; - struct iscsi_conn *conn = isert_conn->conn; + struct iscsit_conn *conn = isert_conn->conn; + struct iscsi_text_rsp *hdr; - pr_debug("Entering isert_put_cmd: %p\n", isert_cmd); + isert_dbg("Cmd %p\n", isert_cmd); switch (cmd->iscsi_opcode) { case ISCSI_OP_SCSI_CMD: spin_lock_bh(&conn->cmd_lock); if (!list_empty(&cmd->i_conn_node)) - list_del(&cmd->i_conn_node); + list_del_init(&cmd->i_conn_node); spin_unlock_bh(&conn->cmd_lock); - if (cmd->data_direction == DMA_TO_DEVICE) + if (cmd->data_direction == DMA_TO_DEVICE) { iscsit_stop_dataout_timer(cmd); + /* + * Check for special case during comp_err where + * WRITE_PENDING has been handed off from core, + * but requires an extra target_put_sess_cmd() + * before transport_generic_free_cmd() below. + */ + if (comp_err && + cmd->se_cmd.t_state == TRANSPORT_WRITE_PENDING) { + struct se_cmd *se_cmd = &cmd->se_cmd; + + target_put_sess_cmd(se_cmd); + } + } - isert_unmap_cmd(isert_cmd, isert_conn); + isert_rdma_rw_ctx_destroy(isert_cmd, isert_conn); transport_generic_free_cmd(&cmd->se_cmd, 0); break; case ISCSI_OP_SCSI_TMFUNC: spin_lock_bh(&conn->cmd_lock); if (!list_empty(&cmd->i_conn_node)) - list_del(&cmd->i_conn_node); + list_del_init(&cmd->i_conn_node); spin_unlock_bh(&conn->cmd_lock); transport_generic_free_cmd(&cmd->se_cmd, 0); @@ -1290,9 +1471,14 @@ isert_put_cmd(struct isert_cmd *isert_cmd) case ISCSI_OP_REJECT: case ISCSI_OP_NOOP_OUT: case ISCSI_OP_TEXT: + hdr = (struct iscsi_text_rsp *)&isert_cmd->tx_desc.iscsi_header; + /* If the continue bit is on, keep the command alive */ + if (hdr->flags & ISCSI_FLAG_TEXT_CONTINUE) + break; + spin_lock_bh(&conn->cmd_lock); if (!list_empty(&cmd->i_conn_node)) - list_del(&cmd->i_conn_node); + list_del_init(&cmd->i_conn_node); spin_unlock_bh(&conn->cmd_lock); /* @@ -1301,17 +1487,14 @@ isert_put_cmd(struct isert_cmd *isert_cmd) * associated cmd->se_cmd needs to be released. */ if (cmd->se_cmd.se_tfo != NULL) { - pr_debug("Calling transport_generic_free_cmd from" - " isert_put_cmd for 0x%02x\n", + isert_dbg("Calling transport_generic_free_cmd for 0x%02x\n", cmd->iscsi_opcode); transport_generic_free_cmd(&cmd->se_cmd, 0); break; } - /* - * Fall-through - */ + fallthrough; default: - isert_release_cmd(cmd); + iscsit_release_cmd(cmd); break; } } @@ -1320,7 +1503,7 @@ static void isert_unmap_tx_desc(struct iser_tx_desc *tx_desc, struct ib_device *ib_dev) { if (tx_desc->dma_addr != 0) { - pr_debug("Calling ib_dma_unmap_single for tx_desc->dma_addr\n"); + isert_dbg("unmap single for tx_desc->dma_addr\n"); ib_dma_unmap_single(ib_dev, tx_desc->dma_addr, ISER_HEADERS_LEN, DMA_TO_DEVICE); tx_desc->dma_addr = 0; @@ -1329,301 +1512,254 @@ isert_unmap_tx_desc(struct iser_tx_desc *tx_desc, struct ib_device *ib_dev) static void isert_completion_put(struct iser_tx_desc *tx_desc, struct isert_cmd *isert_cmd, - struct ib_device *ib_dev) + struct ib_device *ib_dev, bool comp_err) { if (isert_cmd->pdu_buf_dma != 0) { - pr_debug("Calling ib_dma_unmap_single for isert_cmd->pdu_buf_dma\n"); + isert_dbg("unmap single for isert_cmd->pdu_buf_dma\n"); ib_dma_unmap_single(ib_dev, isert_cmd->pdu_buf_dma, isert_cmd->pdu_buf_len, DMA_TO_DEVICE); isert_cmd->pdu_buf_dma = 0; } isert_unmap_tx_desc(tx_desc, ib_dev); - isert_put_cmd(isert_cmd); + isert_put_cmd(isert_cmd, comp_err); } -static void -isert_completion_rdma_read(struct iser_tx_desc *tx_desc, - struct isert_cmd *isert_cmd) +static int +isert_check_pi_status(struct se_cmd *se_cmd, struct ib_mr *sig_mr) { - struct isert_rdma_wr *wr = &isert_cmd->rdma_wr; - struct iscsi_cmd *cmd = &isert_cmd->iscsi_cmd; - struct se_cmd *se_cmd = &cmd->se_cmd; - struct ib_device *ib_dev = isert_cmd->conn->conn_cm_id->device; - - iscsit_stop_dataout_timer(cmd); + struct ib_mr_status mr_status; + int ret; - if (wr->sge) { - pr_debug("isert_do_rdma_read_comp: Unmapping wr->sge from t_data_sg\n"); - ib_dma_unmap_sg(ib_dev, wr->sge, wr->num_sge, DMA_TO_DEVICE); - wr->sge = NULL; + ret = ib_check_mr_status(sig_mr, IB_MR_CHECK_SIG_STATUS, &mr_status); + if (ret) { + isert_err("ib_check_mr_status failed, ret %d\n", ret); + goto fail_mr_status; } - if (isert_cmd->ib_sge) { - pr_debug("isert_do_rdma_read_comp: Freeing isert_cmd->ib_sge\n"); - kfree(isert_cmd->ib_sge); - isert_cmd->ib_sge = NULL; - } + if (mr_status.fail_status & IB_MR_CHECK_SIG_STATUS) { + u64 sec_offset_err; + u32 block_size = se_cmd->se_dev->dev_attrib.block_size + 8; - cmd->write_data_done = se_cmd->data_length; + switch (mr_status.sig_err.err_type) { + case IB_SIG_BAD_GUARD: + se_cmd->pi_err = TCM_LOGICAL_BLOCK_GUARD_CHECK_FAILED; + break; + case IB_SIG_BAD_REFTAG: + se_cmd->pi_err = TCM_LOGICAL_BLOCK_REF_TAG_CHECK_FAILED; + break; + case IB_SIG_BAD_APPTAG: + se_cmd->pi_err = TCM_LOGICAL_BLOCK_APP_TAG_CHECK_FAILED; + break; + } + sec_offset_err = mr_status.sig_err.sig_err_offset; + do_div(sec_offset_err, block_size); + se_cmd->sense_info = sec_offset_err + se_cmd->t_task_lba; - pr_debug("isert_do_rdma_read_comp, calling target_execute_cmd\n"); - spin_lock_bh(&cmd->istate_lock); - cmd->cmd_flags |= ICF_GOT_LAST_DATAOUT; - cmd->i_state = ISTATE_RECEIVED_LAST_DATAOUT; - spin_unlock_bh(&cmd->istate_lock); + isert_err("PI error found type %d at sector 0x%llx " + "expected 0x%x vs actual 0x%x\n", + mr_status.sig_err.err_type, + (unsigned long long)se_cmd->sense_info, + mr_status.sig_err.expected, + mr_status.sig_err.actual); + ret = 1; + } - target_execute_cmd(se_cmd); +fail_mr_status: + return ret; } static void -isert_do_control_comp(struct work_struct *work) +isert_rdma_write_done(struct ib_cq *cq, struct ib_wc *wc) { - struct isert_cmd *isert_cmd = container_of(work, - struct isert_cmd, comp_work); - struct isert_conn *isert_conn = isert_cmd->conn; - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; - struct iscsi_cmd *cmd = &isert_cmd->iscsi_cmd; + struct isert_conn *isert_conn = wc->qp->qp_context; + struct isert_device *device = isert_conn->device; + struct iser_tx_desc *desc = cqe_to_tx_desc(wc->wr_cqe); + struct isert_cmd *isert_cmd = tx_desc_to_cmd(desc); + struct se_cmd *cmd = &isert_cmd->iscsit_cmd->se_cmd; + int ret = 0; - switch (cmd->i_state) { - case ISTATE_SEND_TASKMGTRSP: - pr_debug("Calling iscsit_tmr_post_handler >>>>>>>>>>>>>>>>>\n"); + if (unlikely(wc->status != IB_WC_SUCCESS)) { + isert_print_wc(wc, "rdma write"); + if (wc->status != IB_WC_WR_FLUSH_ERR) + iscsit_cause_connection_reinstatement(isert_conn->conn, 0); + isert_completion_put(desc, isert_cmd, device->ib_device, true); + return; + } - atomic_dec(&isert_conn->post_send_buf_count); - iscsit_tmr_post_handler(cmd, cmd->conn); + isert_dbg("Cmd %p\n", isert_cmd); - cmd->i_state = ISTATE_SENT_STATUS; - isert_completion_put(&isert_cmd->tx_desc, isert_cmd, ib_dev); - break; - case ISTATE_SEND_REJECT: - pr_debug("Got isert_do_control_comp ISTATE_SEND_REJECT: >>>\n"); - atomic_dec(&isert_conn->post_send_buf_count); + ret = isert_check_pi_status(cmd, isert_cmd->rw.reg->mr); + isert_rdma_rw_ctx_destroy(isert_cmd, isert_conn); - cmd->i_state = ISTATE_SENT_STATUS; - isert_completion_put(&isert_cmd->tx_desc, isert_cmd, ib_dev); - break; - case ISTATE_SEND_LOGOUTRSP: - pr_debug("Calling iscsit_logout_post_handler >>>>>>>>>>>>>>\n"); + if (ret) { /* - * Call atomic_dec(&isert_conn->post_send_buf_count) - * from isert_free_conn() + * transport_generic_request_failure() expects to have + * plus two references to handle queue-full, so re-add + * one here as target-core will have already dropped + * it after the first isert_put_datain() callback. */ - isert_conn->logout_posted = true; - iscsit_logout_post_handler(cmd, cmd->conn); - break; - case ISTATE_SEND_TEXTRSP: - atomic_dec(&isert_conn->post_send_buf_count); - cmd->i_state = ISTATE_SENT_STATUS; - isert_completion_put(&isert_cmd->tx_desc, isert_cmd, ib_dev); - break; - default: - pr_err("Unknown do_control_comp i_state %d\n", cmd->i_state); - dump_stack(); - break; + kref_get(&cmd->cmd_kref); + transport_generic_request_failure(cmd, cmd->pi_err); + } else { + /* + * XXX: isert_put_response() failure is not retried. + */ + ret = isert_put_response(isert_conn->conn, isert_cmd->iscsit_cmd); + if (ret) + pr_warn_ratelimited("isert_put_response() ret: %d\n", ret); } } static void -isert_response_completion(struct iser_tx_desc *tx_desc, - struct isert_cmd *isert_cmd, - struct isert_conn *isert_conn, - struct ib_device *ib_dev) +isert_rdma_read_done(struct ib_cq *cq, struct ib_wc *wc) { - struct iscsi_cmd *cmd = &isert_cmd->iscsi_cmd; - - if (cmd->i_state == ISTATE_SEND_TASKMGTRSP || - cmd->i_state == ISTATE_SEND_LOGOUTRSP || - cmd->i_state == ISTATE_SEND_REJECT || - cmd->i_state == ISTATE_SEND_TEXTRSP) { - isert_unmap_tx_desc(tx_desc, ib_dev); + struct isert_conn *isert_conn = wc->qp->qp_context; + struct isert_device *device = isert_conn->device; + struct iser_tx_desc *desc = cqe_to_tx_desc(wc->wr_cqe); + struct isert_cmd *isert_cmd = tx_desc_to_cmd(desc); + struct iscsit_cmd *cmd = isert_cmd->iscsit_cmd; + struct se_cmd *se_cmd = &cmd->se_cmd; + int ret = 0; - INIT_WORK(&isert_cmd->comp_work, isert_do_control_comp); - queue_work(isert_comp_wq, &isert_cmd->comp_work); + if (unlikely(wc->status != IB_WC_SUCCESS)) { + isert_print_wc(wc, "rdma read"); + if (wc->status != IB_WC_WR_FLUSH_ERR) + iscsit_cause_connection_reinstatement(isert_conn->conn, 0); + isert_completion_put(desc, isert_cmd, device->ib_device, true); return; } - atomic_dec(&isert_conn->post_send_buf_count); - cmd->i_state = ISTATE_SENT_STATUS; - isert_completion_put(tx_desc, isert_cmd, ib_dev); + isert_dbg("Cmd %p\n", isert_cmd); + + iscsit_stop_dataout_timer(cmd); + + if (isert_prot_cmd(isert_conn, se_cmd)) + ret = isert_check_pi_status(se_cmd, isert_cmd->rw.reg->mr); + isert_rdma_rw_ctx_destroy(isert_cmd, isert_conn); + cmd->write_data_done = 0; + + isert_dbg("Cmd: %p RDMA_READ comp calling execute_cmd\n", isert_cmd); + spin_lock_bh(&cmd->istate_lock); + cmd->cmd_flags |= ICF_GOT_LAST_DATAOUT; + cmd->i_state = ISTATE_RECEIVED_LAST_DATAOUT; + spin_unlock_bh(&cmd->istate_lock); + + /* + * transport_generic_request_failure() will drop the extra + * se_cmd->cmd_kref reference after T10-PI error, and handle + * any non-zero ->queue_status() callback error retries. + */ + if (ret) + transport_generic_request_failure(se_cmd, se_cmd->pi_err); + else + target_execute_cmd(se_cmd); } static void -isert_send_completion(struct iser_tx_desc *tx_desc, - struct isert_conn *isert_conn) +isert_do_control_comp(struct work_struct *work) { - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; - struct isert_cmd *isert_cmd = tx_desc->isert_cmd; - struct isert_rdma_wr *wr; + struct isert_cmd *isert_cmd = container_of(work, + struct isert_cmd, comp_work); + struct isert_conn *isert_conn = isert_cmd->conn; + struct ib_device *ib_dev = isert_conn->cm_id->device; + struct iscsit_cmd *cmd = isert_cmd->iscsit_cmd; - if (!isert_cmd) { - atomic_dec(&isert_conn->post_send_buf_count); - isert_unmap_tx_desc(tx_desc, ib_dev); - return; - } - wr = &isert_cmd->rdma_wr; + isert_dbg("Cmd %p i_state %d\n", isert_cmd, cmd->i_state); - switch (wr->iser_ib_op) { - case ISER_IB_RECV: - pr_err("isert_send_completion: Got ISER_IB_RECV\n"); - dump_stack(); - break; - case ISER_IB_SEND: - pr_debug("isert_send_completion: Got ISER_IB_SEND\n"); - isert_response_completion(tx_desc, isert_cmd, - isert_conn, ib_dev); - break; - case ISER_IB_RDMA_WRITE: - pr_err("isert_send_completion: Got ISER_IB_RDMA_WRITE\n"); - dump_stack(); + switch (cmd->i_state) { + case ISTATE_SEND_TASKMGTRSP: + iscsit_tmr_post_handler(cmd, cmd->conn); + fallthrough; + case ISTATE_SEND_REJECT: + case ISTATE_SEND_TEXTRSP: + cmd->i_state = ISTATE_SENT_STATUS; + isert_completion_put(&isert_cmd->tx_desc, isert_cmd, + ib_dev, false); break; - case ISER_IB_RDMA_READ: - pr_debug("isert_send_completion: Got ISER_IB_RDMA_READ:\n"); - - atomic_dec(&isert_conn->post_send_buf_count); - isert_completion_rdma_read(tx_desc, isert_cmd); + case ISTATE_SEND_LOGOUTRSP: + iscsit_logout_post_handler(cmd, cmd->conn); break; default: - pr_err("Unknown wr->iser_ib_op: 0x%02x\n", wr->iser_ib_op); + isert_err("Unknown i_state %d\n", cmd->i_state); dump_stack(); break; } } static void -isert_cq_comp_err(struct iser_tx_desc *tx_desc, struct isert_conn *isert_conn) +isert_login_send_done(struct ib_cq *cq, struct ib_wc *wc) { - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; - - if (tx_desc) { - struct isert_cmd *isert_cmd = tx_desc->isert_cmd; + struct isert_conn *isert_conn = wc->qp->qp_context; + struct ib_device *ib_dev = isert_conn->cm_id->device; + struct iser_tx_desc *tx_desc = cqe_to_tx_desc(wc->wr_cqe); - if (!isert_cmd) - isert_unmap_tx_desc(tx_desc, ib_dev); - else - isert_completion_put(tx_desc, isert_cmd, ib_dev); + if (unlikely(wc->status != IB_WC_SUCCESS)) { + isert_print_wc(wc, "login send"); + if (wc->status != IB_WC_WR_FLUSH_ERR) + iscsit_cause_connection_reinstatement(isert_conn->conn, 0); } - if (isert_conn->post_recv_buf_count == 0 && - atomic_read(&isert_conn->post_send_buf_count) == 0) { - pr_debug("isert_cq_comp_err >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\n"); - pr_debug("Calling wake_up from isert_cq_comp_err\n"); - - mutex_lock(&isert_conn->conn_mutex); - if (isert_conn->state != ISER_CONN_DOWN) - isert_conn->state = ISER_CONN_TERMINATING; - mutex_unlock(&isert_conn->conn_mutex); - - wake_up(&isert_conn->conn_wait_comp_err); - } + isert_unmap_tx_desc(tx_desc, ib_dev); } static void -isert_cq_tx_work(struct work_struct *work) -{ - struct isert_cq_desc *cq_desc = container_of(work, - struct isert_cq_desc, cq_tx_work); - struct isert_device *device = cq_desc->device; - int cq_index = cq_desc->cq_index; - struct ib_cq *tx_cq = device->dev_tx_cq[cq_index]; - struct isert_conn *isert_conn; - struct iser_tx_desc *tx_desc; - struct ib_wc wc; - - while (ib_poll_cq(tx_cq, 1, &wc) == 1) { - tx_desc = (struct iser_tx_desc *)(unsigned long)wc.wr_id; - isert_conn = wc.qp->qp_context; - - if (wc.status == IB_WC_SUCCESS) { - isert_send_completion(tx_desc, isert_conn); - } else { - pr_debug("TX wc.status != IB_WC_SUCCESS >>>>>>>>>>>>>>\n"); - pr_debug("TX wc.status: 0x%08x\n", wc.status); - pr_debug("TX wc.vendor_err: 0x%08x\n", wc.vendor_err); - atomic_dec(&isert_conn->post_send_buf_count); - isert_cq_comp_err(tx_desc, isert_conn); - } +isert_send_done(struct ib_cq *cq, struct ib_wc *wc) +{ + struct isert_conn *isert_conn = wc->qp->qp_context; + struct ib_device *ib_dev = isert_conn->cm_id->device; + struct iser_tx_desc *tx_desc = cqe_to_tx_desc(wc->wr_cqe); + struct isert_cmd *isert_cmd = tx_desc_to_cmd(tx_desc); + + if (unlikely(wc->status != IB_WC_SUCCESS)) { + isert_print_wc(wc, "send"); + if (wc->status != IB_WC_WR_FLUSH_ERR) + iscsit_cause_connection_reinstatement(isert_conn->conn, 0); + isert_completion_put(tx_desc, isert_cmd, ib_dev, true); + return; } - ib_req_notify_cq(tx_cq, IB_CQ_NEXT_COMP); -} - -static void -isert_cq_tx_callback(struct ib_cq *cq, void *context) -{ - struct isert_cq_desc *cq_desc = (struct isert_cq_desc *)context; + isert_dbg("Cmd %p\n", isert_cmd); - INIT_WORK(&cq_desc->cq_tx_work, isert_cq_tx_work); - queue_work(isert_comp_wq, &cq_desc->cq_tx_work); -} + switch (isert_cmd->iscsit_cmd->i_state) { + case ISTATE_SEND_TASKMGTRSP: + case ISTATE_SEND_LOGOUTRSP: + case ISTATE_SEND_REJECT: + case ISTATE_SEND_TEXTRSP: + isert_unmap_tx_desc(tx_desc, ib_dev); -static void -isert_cq_rx_work(struct work_struct *work) -{ - struct isert_cq_desc *cq_desc = container_of(work, - struct isert_cq_desc, cq_rx_work); - struct isert_device *device = cq_desc->device; - int cq_index = cq_desc->cq_index; - struct ib_cq *rx_cq = device->dev_rx_cq[cq_index]; - struct isert_conn *isert_conn; - struct iser_rx_desc *rx_desc; - struct ib_wc wc; - unsigned long xfer_len; - - while (ib_poll_cq(rx_cq, 1, &wc) == 1) { - rx_desc = (struct iser_rx_desc *)(unsigned long)wc.wr_id; - isert_conn = wc.qp->qp_context; - - if (wc.status == IB_WC_SUCCESS) { - xfer_len = (unsigned long)wc.byte_len; - isert_rx_completion(rx_desc, isert_conn, xfer_len); - } else { - pr_debug("RX wc.status != IB_WC_SUCCESS >>>>>>>>>>>>>>\n"); - if (wc.status != IB_WC_WR_FLUSH_ERR) { - pr_debug("RX wc.status: 0x%08x\n", wc.status); - pr_debug("RX wc.vendor_err: 0x%08x\n", - wc.vendor_err); - } - isert_conn->post_recv_buf_count--; - isert_cq_comp_err(NULL, isert_conn); - } + INIT_WORK(&isert_cmd->comp_work, isert_do_control_comp); + queue_work(isert_comp_wq, &isert_cmd->comp_work); + return; + default: + isert_cmd->iscsit_cmd->i_state = ISTATE_SENT_STATUS; + isert_completion_put(tx_desc, isert_cmd, ib_dev, false); + break; } - - ib_req_notify_cq(rx_cq, IB_CQ_NEXT_COMP); -} - -static void -isert_cq_rx_callback(struct ib_cq *cq, void *context) -{ - struct isert_cq_desc *cq_desc = (struct isert_cq_desc *)context; - - INIT_WORK(&cq_desc->cq_rx_work, isert_cq_rx_work); - queue_work(isert_rx_wq, &cq_desc->cq_rx_work); } static int isert_post_response(struct isert_conn *isert_conn, struct isert_cmd *isert_cmd) { - struct ib_send_wr *wr_failed; int ret; - atomic_inc(&isert_conn->post_send_buf_count); + ret = isert_post_recv(isert_conn, isert_cmd->rx_desc); + if (ret) + return ret; - ret = ib_post_send(isert_conn->conn_qp, &isert_cmd->tx_desc.send_wr, - &wr_failed); + ret = ib_post_send(isert_conn->qp, &isert_cmd->tx_desc.send_wr, NULL); if (ret) { - pr_err("ib_post_send failed with %d\n", ret); - atomic_dec(&isert_conn->post_send_buf_count); + isert_err("ib_post_send failed with %d\n", ret); return ret; } return ret; } static int -isert_put_response(struct iscsi_conn *conn, struct iscsi_cmd *cmd) +isert_put_response(struct iscsit_conn *conn, struct iscsit_cmd *cmd) { - struct isert_cmd *isert_cmd = container_of(cmd, - struct isert_cmd, iscsi_cmd); - struct isert_conn *isert_conn = (struct isert_conn *)conn->context; + struct isert_cmd *isert_cmd = iscsit_priv_cmd(cmd); + struct isert_conn *isert_conn = conn->context; struct ib_send_wr *send_wr = &isert_cmd->tx_desc.send_wr; struct iscsi_scsi_rsp *hdr = (struct iscsi_scsi_rsp *) &isert_cmd->tx_desc.iscsi_header; @@ -1637,7 +1773,8 @@ isert_put_response(struct iscsi_conn *conn, struct iscsi_cmd *cmd) if (cmd->se_cmd.sense_buffer && ((cmd->se_cmd.se_cmd_flags & SCF_TRANSPORT_TASK_SENSE) || (cmd->se_cmd.se_cmd_flags & SCF_EMULATED_TASK_SENSE))) { - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; + struct isert_device *device = isert_conn->device; + struct ib_device *ib_dev = device->ib_device; struct ib_sge *tx_dsg = &isert_cmd->tx_desc.tx_sg[1]; u32 padding, pdu_len; @@ -1652,28 +1789,65 @@ isert_put_response(struct iscsi_conn *conn, struct iscsi_cmd *cmd) isert_cmd->pdu_buf_dma = ib_dma_map_single(ib_dev, (void *)cmd->sense_buffer, pdu_len, DMA_TO_DEVICE); + if (ib_dma_mapping_error(ib_dev, isert_cmd->pdu_buf_dma)) + return -ENOMEM; isert_cmd->pdu_buf_len = pdu_len; tx_dsg->addr = isert_cmd->pdu_buf_dma; tx_dsg->length = pdu_len; - tx_dsg->lkey = isert_conn->conn_mr->lkey; + tx_dsg->lkey = device->pd->local_dma_lkey; isert_cmd->tx_desc.num_sge = 2; } - isert_init_send_wr(isert_cmd, send_wr); + isert_init_send_wr(isert_conn, isert_cmd, send_wr); - pr_debug("Posting SCSI Response IB_WR_SEND >>>>>>>>>>>>>>>>>>>>>>\n"); + isert_dbg("Posting SCSI Response\n"); return isert_post_response(isert_conn, isert_cmd); } +static void +isert_aborted_task(struct iscsit_conn *conn, struct iscsit_cmd *cmd) +{ + struct isert_cmd *isert_cmd = iscsit_priv_cmd(cmd); + struct isert_conn *isert_conn = conn->context; + + spin_lock_bh(&conn->cmd_lock); + if (!list_empty(&cmd->i_conn_node)) + list_del_init(&cmd->i_conn_node); + spin_unlock_bh(&conn->cmd_lock); + + if (cmd->data_direction == DMA_TO_DEVICE) + iscsit_stop_dataout_timer(cmd); + isert_rdma_rw_ctx_destroy(isert_cmd, isert_conn); +} + +static enum target_prot_op +isert_get_sup_prot_ops(struct iscsit_conn *conn) +{ + struct isert_conn *isert_conn = conn->context; + struct isert_device *device = isert_conn->device; + + if (conn->tpg->tpg_attrib.t10_pi) { + if (device->pi_capable) { + isert_info("conn %p PI offload enabled\n", isert_conn); + isert_conn->pi_support = true; + return TARGET_PROT_ALL; + } + } + + isert_info("conn %p PI offload disabled\n", isert_conn); + isert_conn->pi_support = false; + + return TARGET_PROT_NORMAL; +} + static int -isert_put_nopin(struct iscsi_cmd *cmd, struct iscsi_conn *conn, +isert_put_nopin(struct iscsit_cmd *cmd, struct iscsit_conn *conn, bool nopout_response) { - struct isert_cmd *isert_cmd = container_of(cmd, - struct isert_cmd, iscsi_cmd); - struct isert_conn *isert_conn = (struct isert_conn *)conn->context; + struct isert_cmd *isert_cmd = iscsit_priv_cmd(cmd); + struct isert_conn *isert_conn = conn->context; struct ib_send_wr *send_wr = &isert_cmd->tx_desc.send_wr; isert_create_send_desc(isert_conn, isert_cmd, &isert_cmd->tx_desc); @@ -1681,59 +1855,57 @@ isert_put_nopin(struct iscsi_cmd *cmd, struct iscsi_conn *conn, &isert_cmd->tx_desc.iscsi_header, nopout_response); isert_init_tx_hdrs(isert_conn, &isert_cmd->tx_desc); - isert_init_send_wr(isert_cmd, send_wr); + isert_init_send_wr(isert_conn, isert_cmd, send_wr); - pr_debug("Posting NOPIN Response IB_WR_SEND >>>>>>>>>>>>>>>>>>>>>>\n"); + isert_dbg("conn %p Posting NOPIN Response\n", isert_conn); return isert_post_response(isert_conn, isert_cmd); } static int -isert_put_logout_rsp(struct iscsi_cmd *cmd, struct iscsi_conn *conn) +isert_put_logout_rsp(struct iscsit_cmd *cmd, struct iscsit_conn *conn) { - struct isert_cmd *isert_cmd = container_of(cmd, - struct isert_cmd, iscsi_cmd); - struct isert_conn *isert_conn = (struct isert_conn *)conn->context; + struct isert_cmd *isert_cmd = iscsit_priv_cmd(cmd); + struct isert_conn *isert_conn = conn->context; struct ib_send_wr *send_wr = &isert_cmd->tx_desc.send_wr; isert_create_send_desc(isert_conn, isert_cmd, &isert_cmd->tx_desc); iscsit_build_logout_rsp(cmd, conn, (struct iscsi_logout_rsp *) &isert_cmd->tx_desc.iscsi_header); isert_init_tx_hdrs(isert_conn, &isert_cmd->tx_desc); - isert_init_send_wr(isert_cmd, send_wr); + isert_init_send_wr(isert_conn, isert_cmd, send_wr); - pr_debug("Posting Logout Response IB_WR_SEND >>>>>>>>>>>>>>>>>>>>>>\n"); + isert_dbg("conn %p Posting Logout Response\n", isert_conn); return isert_post_response(isert_conn, isert_cmd); } static int -isert_put_tm_rsp(struct iscsi_cmd *cmd, struct iscsi_conn *conn) +isert_put_tm_rsp(struct iscsit_cmd *cmd, struct iscsit_conn *conn) { - struct isert_cmd *isert_cmd = container_of(cmd, - struct isert_cmd, iscsi_cmd); - struct isert_conn *isert_conn = (struct isert_conn *)conn->context; + struct isert_cmd *isert_cmd = iscsit_priv_cmd(cmd); + struct isert_conn *isert_conn = conn->context; struct ib_send_wr *send_wr = &isert_cmd->tx_desc.send_wr; isert_create_send_desc(isert_conn, isert_cmd, &isert_cmd->tx_desc); iscsit_build_task_mgt_rsp(cmd, conn, (struct iscsi_tm_rsp *) &isert_cmd->tx_desc.iscsi_header); isert_init_tx_hdrs(isert_conn, &isert_cmd->tx_desc); - isert_init_send_wr(isert_cmd, send_wr); + isert_init_send_wr(isert_conn, isert_cmd, send_wr); - pr_debug("Posting Task Management Response IB_WR_SEND >>>>>>>>>>>>>>>>>>>>>>\n"); + isert_dbg("conn %p Posting Task Management Response\n", isert_conn); return isert_post_response(isert_conn, isert_cmd); } static int -isert_put_reject(struct iscsi_cmd *cmd, struct iscsi_conn *conn) +isert_put_reject(struct iscsit_cmd *cmd, struct iscsit_conn *conn) { - struct isert_cmd *isert_cmd = container_of(cmd, - struct isert_cmd, iscsi_cmd); - struct isert_conn *isert_conn = (struct isert_conn *)conn->context; + struct isert_cmd *isert_cmd = iscsit_priv_cmd(cmd); + struct isert_conn *isert_conn = conn->context; struct ib_send_wr *send_wr = &isert_cmd->tx_desc.send_wr; - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; + struct isert_device *device = isert_conn->device; + struct ib_device *ib_dev = device->ib_device; struct ib_sge *tx_dsg = &isert_cmd->tx_desc.tx_sg[1]; struct iscsi_reject *hdr = (struct iscsi_reject *)&isert_cmd->tx_desc.iscsi_header; @@ -1746,25 +1918,26 @@ isert_put_reject(struct iscsi_cmd *cmd, struct iscsi_conn *conn) isert_cmd->pdu_buf_dma = ib_dma_map_single(ib_dev, (void *)cmd->buf_ptr, ISCSI_HDR_LEN, DMA_TO_DEVICE); + if (ib_dma_mapping_error(ib_dev, isert_cmd->pdu_buf_dma)) + return -ENOMEM; isert_cmd->pdu_buf_len = ISCSI_HDR_LEN; tx_dsg->addr = isert_cmd->pdu_buf_dma; tx_dsg->length = ISCSI_HDR_LEN; - tx_dsg->lkey = isert_conn->conn_mr->lkey; + tx_dsg->lkey = device->pd->local_dma_lkey; isert_cmd->tx_desc.num_sge = 2; - isert_init_send_wr(isert_cmd, send_wr); + isert_init_send_wr(isert_conn, isert_cmd, send_wr); - pr_debug("Posting Reject IB_WR_SEND >>>>>>>>>>>>>>>>>>>>>>\n"); + isert_dbg("conn %p Posting Reject\n", isert_conn); return isert_post_response(isert_conn, isert_cmd); } static int -isert_put_text_rsp(struct iscsi_cmd *cmd, struct iscsi_conn *conn) +isert_put_text_rsp(struct iscsit_cmd *cmd, struct iscsit_conn *conn) { - struct isert_cmd *isert_cmd = container_of(cmd, - struct isert_cmd, iscsi_cmd); - struct isert_conn *isert_conn = (struct isert_conn *)conn->context; + struct isert_cmd *isert_cmd = iscsit_priv_cmd(cmd); + struct isert_conn *isert_conn = conn->context; struct ib_send_wr *send_wr = &isert_cmd->tx_desc.send_wr; struct iscsi_text_rsp *hdr = (struct iscsi_text_rsp *)&isert_cmd->tx_desc.iscsi_header; @@ -1772,7 +1945,7 @@ isert_put_text_rsp(struct iscsi_cmd *cmd, struct iscsi_conn *conn) int rc; isert_create_send_desc(isert_conn, isert_cmd, &isert_cmd->tx_desc); - rc = iscsit_build_text_rsp(cmd, conn, hdr); + rc = iscsit_build_text_rsp(cmd, conn, hdr, ISCSI_INFINIBAND); if (rc < 0) return rc; @@ -1780,295 +1953,218 @@ isert_put_text_rsp(struct iscsi_cmd *cmd, struct iscsi_conn *conn) isert_init_tx_hdrs(isert_conn, &isert_cmd->tx_desc); if (txt_rsp_len) { - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; + struct isert_device *device = isert_conn->device; + struct ib_device *ib_dev = device->ib_device; struct ib_sge *tx_dsg = &isert_cmd->tx_desc.tx_sg[1]; void *txt_rsp_buf = cmd->buf_ptr; isert_cmd->pdu_buf_dma = ib_dma_map_single(ib_dev, txt_rsp_buf, txt_rsp_len, DMA_TO_DEVICE); + if (ib_dma_mapping_error(ib_dev, isert_cmd->pdu_buf_dma)) + return -ENOMEM; isert_cmd->pdu_buf_len = txt_rsp_len; tx_dsg->addr = isert_cmd->pdu_buf_dma; tx_dsg->length = txt_rsp_len; - tx_dsg->lkey = isert_conn->conn_mr->lkey; + tx_dsg->lkey = device->pd->local_dma_lkey; isert_cmd->tx_desc.num_sge = 2; } - isert_init_send_wr(isert_cmd, send_wr); + isert_init_send_wr(isert_conn, isert_cmd, send_wr); - pr_debug("Posting Text Response IB_WR_SEND >>>>>>>>>>>>>>>>>>>>>>\n"); + isert_dbg("conn %p Text Response\n", isert_conn); return isert_post_response(isert_conn, isert_cmd); } -static int -isert_build_rdma_wr(struct isert_conn *isert_conn, struct isert_cmd *isert_cmd, - struct ib_sge *ib_sge, struct ib_send_wr *send_wr, - u32 data_left, u32 offset) +static inline void +isert_set_dif_domain(struct se_cmd *se_cmd, struct ib_sig_domain *domain) { - struct iscsi_cmd *cmd = &isert_cmd->iscsi_cmd; - struct scatterlist *sg_start, *tmp_sg; - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; - u32 sg_off, page_off; - int i = 0, sg_nents; - - sg_off = offset / PAGE_SIZE; - sg_start = &cmd->se_cmd.t_data_sg[sg_off]; - sg_nents = min(cmd->se_cmd.t_data_nents - sg_off, isert_conn->max_sge); - page_off = offset % PAGE_SIZE; - - send_wr->sg_list = ib_sge; - send_wr->num_sge = sg_nents; - send_wr->wr_id = (unsigned long)&isert_cmd->tx_desc; + domain->sig_type = IB_SIG_TYPE_T10_DIF; + domain->sig.dif.bg_type = IB_T10DIF_CRC; + domain->sig.dif.pi_interval = se_cmd->se_dev->dev_attrib.block_size; + domain->sig.dif.ref_tag = se_cmd->reftag_seed; /* - * Perform mapping of TCM scatterlist memory ib_sge dma_addr. + * At the moment we hard code those, but if in the future + * the target core would like to use it, we will take it + * from se_cmd. */ - for_each_sg(sg_start, tmp_sg, sg_nents, i) { - pr_debug("ISER RDMA from SGL dma_addr: 0x%16llx dma_len: %u, page_off: %u\n", - (unsigned long long)tmp_sg->dma_address, - tmp_sg->length, page_off); - - ib_sge->addr = ib_sg_dma_address(ib_dev, tmp_sg) + page_off; - ib_sge->length = min_t(u32, data_left, - ib_sg_dma_len(ib_dev, tmp_sg) - page_off); - ib_sge->lkey = isert_conn->conn_mr->lkey; - - pr_debug("RDMA ib_sge: addr: 0x%16llx length: %u\n", - ib_sge->addr, ib_sge->length); - page_off = 0; - data_left -= ib_sge->length; - ib_sge++; - pr_debug("Incrementing ib_sge pointer to %p\n", ib_sge); - } - - pr_debug("Set outgoing sg_list: %p num_sg: %u from TCM SGLs\n", - send_wr->sg_list, send_wr->num_sge); - - return sg_nents; + domain->sig.dif.apptag_check_mask = 0xffff; + domain->sig.dif.app_escape = true; + domain->sig.dif.ref_escape = true; + if (se_cmd->prot_type == TARGET_DIF_TYPE1_PROT || + se_cmd->prot_type == TARGET_DIF_TYPE2_PROT) + domain->sig.dif.ref_remap = true; } static int -isert_put_datain(struct iscsi_conn *conn, struct iscsi_cmd *cmd) +isert_set_sig_attrs(struct se_cmd *se_cmd, struct ib_sig_attrs *sig_attrs) { - struct se_cmd *se_cmd = &cmd->se_cmd; - struct isert_cmd *isert_cmd = container_of(cmd, - struct isert_cmd, iscsi_cmd); - struct isert_rdma_wr *wr = &isert_cmd->rdma_wr; - struct isert_conn *isert_conn = (struct isert_conn *)conn->context; - struct ib_send_wr *wr_failed, *send_wr; - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; - struct ib_sge *ib_sge; - struct scatterlist *sg; - u32 offset = 0, data_len, data_left, rdma_write_max; - int rc, ret = 0, count, sg_nents, i, ib_sge_cnt; - - pr_debug("RDMA_WRITE: data_length: %u\n", se_cmd->data_length); - - sg = &se_cmd->t_data_sg[0]; - sg_nents = se_cmd->t_data_nents; - - count = ib_dma_map_sg(ib_dev, sg, sg_nents, DMA_TO_DEVICE); - if (unlikely(!count)) { - pr_err("Unable to map put_datain SGs\n"); - return -EINVAL; - } - wr->sge = sg; - wr->num_sge = sg_nents; - pr_debug("Mapped IB count: %u sg: %p sg_nents: %u for RDMA_WRITE\n", - count, sg, sg_nents); + memset(sig_attrs, 0, sizeof(*sig_attrs)); - ib_sge = kzalloc(sizeof(struct ib_sge) * sg_nents, GFP_KERNEL); - if (!ib_sge) { - pr_warn("Unable to allocate datain ib_sge\n"); - ret = -ENOMEM; - goto unmap_sg; + switch (se_cmd->prot_op) { + case TARGET_PROT_DIN_INSERT: + case TARGET_PROT_DOUT_STRIP: + sig_attrs->mem.sig_type = IB_SIG_TYPE_NONE; + isert_set_dif_domain(se_cmd, &sig_attrs->wire); + break; + case TARGET_PROT_DOUT_INSERT: + case TARGET_PROT_DIN_STRIP: + sig_attrs->wire.sig_type = IB_SIG_TYPE_NONE; + isert_set_dif_domain(se_cmd, &sig_attrs->mem); + break; + case TARGET_PROT_DIN_PASS: + case TARGET_PROT_DOUT_PASS: + isert_set_dif_domain(se_cmd, &sig_attrs->wire); + isert_set_dif_domain(se_cmd, &sig_attrs->mem); + break; + default: + isert_err("Unsupported PI operation %d\n", se_cmd->prot_op); + return -EINVAL; } - isert_cmd->ib_sge = ib_sge; - pr_debug("Allocated ib_sge: %p from t_data_ents: %d for RDMA_WRITE\n", - ib_sge, se_cmd->t_data_nents); + if (se_cmd->prot_checks & TARGET_DIF_CHECK_GUARD) + sig_attrs->check_mask |= IB_SIG_CHECK_GUARD; + if (se_cmd->prot_checks & TARGET_DIF_CHECK_APPTAG) + sig_attrs->check_mask |= IB_SIG_CHECK_APPTAG; + if (se_cmd->prot_checks & TARGET_DIF_CHECK_REFTAG) + sig_attrs->check_mask |= IB_SIG_CHECK_REFTAG; - wr->send_wr_num = DIV_ROUND_UP(sg_nents, isert_conn->max_sge); - wr->send_wr = kzalloc(sizeof(struct ib_send_wr) * wr->send_wr_num, - GFP_KERNEL); - if (!wr->send_wr) { - pr_err("Unable to allocate wr->send_wr\n"); - ret = -ENOMEM; - goto unmap_sg; - } - pr_debug("Allocated wr->send_wr: %p wr->send_wr_num: %u\n", - wr->send_wr, wr->send_wr_num); - - iscsit_increment_maxcmdsn(cmd, conn->sess); - cmd->stat_sn = conn->stat_sn++; + return 0; +} - wr->isert_cmd = isert_cmd; - rdma_write_max = isert_conn->max_sge * PAGE_SIZE; - data_left = se_cmd->data_length; +static int +isert_rdma_rw_ctx_post(struct isert_cmd *cmd, struct isert_conn *conn, + struct ib_cqe *cqe, struct ib_send_wr *chain_wr) +{ + struct se_cmd *se_cmd = &cmd->iscsit_cmd->se_cmd; + enum dma_data_direction dir = target_reverse_dma_direction(se_cmd); + u8 port_num = conn->cm_id->port_num; + u64 addr; + u32 rkey, offset; + int ret; - for (i = 0; i < wr->send_wr_num; i++) { - send_wr = &isert_cmd->rdma_wr.send_wr[i]; - data_len = min(data_left, rdma_write_max); + if (cmd->ctx_init_done) + goto rdma_ctx_post; - send_wr->opcode = IB_WR_RDMA_WRITE; - send_wr->send_flags = 0; - send_wr->wr.rdma.remote_addr = isert_cmd->read_va + offset; - send_wr->wr.rdma.rkey = isert_cmd->read_stag; + if (dir == DMA_FROM_DEVICE) { + addr = cmd->write_va; + rkey = cmd->write_stag; + offset = cmd->iscsit_cmd->write_data_done; + } else { + addr = cmd->read_va; + rkey = cmd->read_stag; + offset = 0; + } - ib_sge_cnt = isert_build_rdma_wr(isert_conn, isert_cmd, ib_sge, - send_wr, data_len, offset); - ib_sge += ib_sge_cnt; + if (isert_prot_cmd(conn, se_cmd)) { + struct ib_sig_attrs sig_attrs; - if (i + 1 == wr->send_wr_num) - send_wr->next = &isert_cmd->tx_desc.send_wr; - else - send_wr->next = &wr->send_wr[i + 1]; + ret = isert_set_sig_attrs(se_cmd, &sig_attrs); + if (ret) + return ret; - offset += data_len; - data_left -= data_len; + WARN_ON_ONCE(offset); + ret = rdma_rw_ctx_signature_init(&cmd->rw, conn->qp, port_num, + se_cmd->t_data_sg, se_cmd->t_data_nents, + se_cmd->t_prot_sg, se_cmd->t_prot_nents, + &sig_attrs, addr, rkey, dir); + } else { + ret = rdma_rw_ctx_init(&cmd->rw, conn->qp, port_num, + se_cmd->t_data_sg, se_cmd->t_data_nents, + offset, addr, rkey, dir); } - /* - * Build isert_conn->tx_desc for iSCSI response PDU and attach - */ - isert_create_send_desc(isert_conn, isert_cmd, &isert_cmd->tx_desc); - iscsit_build_rsp_pdu(cmd, conn, false, (struct iscsi_scsi_rsp *) - &isert_cmd->tx_desc.iscsi_header); - isert_init_tx_hdrs(isert_conn, &isert_cmd->tx_desc); - isert_init_send_wr(isert_cmd, &isert_cmd->tx_desc.send_wr); - - atomic_inc(&isert_conn->post_send_buf_count); - rc = ib_post_send(isert_conn->conn_qp, wr->send_wr, &wr_failed); - if (rc) { - pr_warn("ib_post_send() failed for IB_WR_RDMA_WRITE\n"); - atomic_dec(&isert_conn->post_send_buf_count); + if (ret < 0) { + isert_err("Cmd: %p failed to prepare RDMA res\n", cmd); + return ret; } - pr_debug("Posted RDMA_WRITE + Response for iSER Data READ\n"); - return 1; -unmap_sg: - ib_dma_unmap_sg(ib_dev, sg, sg_nents, DMA_TO_DEVICE); + cmd->ctx_init_done = true; + +rdma_ctx_post: + ret = rdma_rw_ctx_post(&cmd->rw, conn->qp, port_num, cqe, chain_wr); + if (ret < 0) + isert_err("Cmd: %p failed to post RDMA res\n", cmd); return ret; } static int -isert_get_dataout(struct iscsi_conn *conn, struct iscsi_cmd *cmd, bool recovery) +isert_put_datain(struct iscsit_conn *conn, struct iscsit_cmd *cmd) { struct se_cmd *se_cmd = &cmd->se_cmd; - struct isert_cmd *isert_cmd = container_of(cmd, - struct isert_cmd, iscsi_cmd); - struct isert_rdma_wr *wr = &isert_cmd->rdma_wr; - struct isert_conn *isert_conn = (struct isert_conn *)conn->context; - struct ib_send_wr *wr_failed, *send_wr; - struct ib_sge *ib_sge; - struct ib_device *ib_dev = isert_conn->conn_cm_id->device; - struct scatterlist *sg_start; - u32 sg_off, sg_nents, page_off, va_offset = 0; - u32 offset = 0, data_len, data_left, rdma_write_max; - int rc, ret = 0, count, i, ib_sge_cnt; - - pr_debug("RDMA_READ: data_length: %u write_data_done: %u\n", - se_cmd->data_length, cmd->write_data_done); - - sg_off = cmd->write_data_done / PAGE_SIZE; - sg_start = &cmd->se_cmd.t_data_sg[sg_off]; - page_off = cmd->write_data_done % PAGE_SIZE; - - pr_debug("RDMA_READ: sg_off: %d, sg_start: %p page_off: %d\n", - sg_off, sg_start, page_off); + struct isert_cmd *isert_cmd = iscsit_priv_cmd(cmd); + struct isert_conn *isert_conn = conn->context; + struct ib_cqe *cqe = NULL; + struct ib_send_wr *chain_wr = NULL; + int rc; - data_left = se_cmd->data_length - cmd->write_data_done; - sg_nents = se_cmd->t_data_nents - sg_off; + isert_dbg("Cmd: %p RDMA_WRITE data_length: %u\n", + isert_cmd, se_cmd->data_length); - pr_debug("RDMA_READ: data_left: %d, sg_nents: %d\n", - data_left, sg_nents); + if (isert_prot_cmd(isert_conn, se_cmd)) { + isert_cmd->tx_desc.tx_cqe.done = isert_rdma_write_done; + cqe = &isert_cmd->tx_desc.tx_cqe; + } else { + /* + * Build isert_conn->tx_desc for iSCSI response PDU and attach + */ + isert_create_send_desc(isert_conn, isert_cmd, + &isert_cmd->tx_desc); + iscsit_build_rsp_pdu(cmd, conn, true, (struct iscsi_scsi_rsp *) + &isert_cmd->tx_desc.iscsi_header); + isert_init_tx_hdrs(isert_conn, &isert_cmd->tx_desc); + isert_init_send_wr(isert_conn, isert_cmd, + &isert_cmd->tx_desc.send_wr); - count = ib_dma_map_sg(ib_dev, sg_start, sg_nents, DMA_FROM_DEVICE); - if (unlikely(!count)) { - pr_err("Unable to map get_dataout SGs\n"); - return -EINVAL; - } - wr->sge = sg_start; - wr->num_sge = sg_nents; - pr_debug("Mapped IB count: %u sg_start: %p sg_nents: %u for RDMA_READ\n", - count, sg_start, sg_nents); + rc = isert_post_recv(isert_conn, isert_cmd->rx_desc); + if (rc) + return rc; - ib_sge = kzalloc(sizeof(struct ib_sge) * sg_nents, GFP_KERNEL); - if (!ib_sge) { - pr_warn("Unable to allocate dataout ib_sge\n"); - ret = -ENOMEM; - goto unmap_sg; + chain_wr = &isert_cmd->tx_desc.send_wr; } - isert_cmd->ib_sge = ib_sge; - - pr_debug("Using ib_sge: %p from sg_ents: %d for RDMA_READ\n", - ib_sge, sg_nents); - wr->send_wr_num = DIV_ROUND_UP(sg_nents, isert_conn->max_sge); - wr->send_wr = kzalloc(sizeof(struct ib_send_wr) * wr->send_wr_num, - GFP_KERNEL); - if (!wr->send_wr) { - pr_debug("Unable to allocate wr->send_wr\n"); - ret = -ENOMEM; - goto unmap_sg; - } - pr_debug("Allocated wr->send_wr: %p wr->send_wr_num: %u\n", - wr->send_wr, wr->send_wr_num); - - isert_cmd->tx_desc.isert_cmd = isert_cmd; - - wr->iser_ib_op = ISER_IB_RDMA_READ; - wr->isert_cmd = isert_cmd; - rdma_write_max = isert_conn->max_sge * PAGE_SIZE; - offset = cmd->write_data_done; - - for (i = 0; i < wr->send_wr_num; i++) { - send_wr = &isert_cmd->rdma_wr.send_wr[i]; - data_len = min(data_left, rdma_write_max); - - send_wr->opcode = IB_WR_RDMA_READ; - send_wr->wr.rdma.remote_addr = isert_cmd->write_va + va_offset; - send_wr->wr.rdma.rkey = isert_cmd->write_stag; - - ib_sge_cnt = isert_build_rdma_wr(isert_conn, isert_cmd, ib_sge, - send_wr, data_len, offset); - ib_sge += ib_sge_cnt; - - if (i + 1 == wr->send_wr_num) - send_wr->send_flags = IB_SEND_SIGNALED; - else - send_wr->next = &wr->send_wr[i + 1]; + rc = isert_rdma_rw_ctx_post(isert_cmd, isert_conn, cqe, chain_wr); + isert_dbg("Cmd: %p posted RDMA_WRITE for iSER Data READ rc: %d\n", + isert_cmd, rc); + return rc; +} - offset += data_len; - va_offset += data_len; - data_left -= data_len; - } +static int +isert_get_dataout(struct iscsit_conn *conn, struct iscsit_cmd *cmd, bool recovery) +{ + struct isert_cmd *isert_cmd = iscsit_priv_cmd(cmd); + int ret; - atomic_inc(&isert_conn->post_send_buf_count); + isert_dbg("Cmd: %p RDMA_READ data_length: %u write_data_done: %u\n", + isert_cmd, cmd->se_cmd.data_length, cmd->write_data_done); - rc = ib_post_send(isert_conn->conn_qp, wr->send_wr, &wr_failed); - if (rc) { - pr_warn("ib_post_send() failed for IB_WR_RDMA_READ\n"); - atomic_dec(&isert_conn->post_send_buf_count); - } - pr_debug("Posted RDMA_READ memory for ISER Data WRITE\n"); - return 0; + isert_cmd->tx_desc.tx_cqe.done = isert_rdma_read_done; + ret = isert_rdma_rw_ctx_post(isert_cmd, conn->context, + &isert_cmd->tx_desc.tx_cqe, NULL); -unmap_sg: - ib_dma_unmap_sg(ib_dev, sg_start, sg_nents, DMA_FROM_DEVICE); + isert_dbg("Cmd: %p posted RDMA_READ memory for ISER Data WRITE rc: %d\n", + isert_cmd, ret); return ret; } static int -isert_immediate_queue(struct iscsi_conn *conn, struct iscsi_cmd *cmd, int state) +isert_immediate_queue(struct iscsit_conn *conn, struct iscsit_cmd *cmd, int state) { - int ret; + struct isert_cmd *isert_cmd = iscsit_priv_cmd(cmd); + int ret = 0; switch (state) { + case ISTATE_REMOVE: + spin_lock_bh(&conn->cmd_lock); + list_del_init(&cmd->i_conn_node); + spin_unlock_bh(&conn->cmd_lock); + isert_put_cmd(isert_cmd, true); + break; case ISTATE_SEND_NOPIN_WANT_RESPONSE: ret = isert_put_nopin(cmd, conn, false); break; default: - pr_err("Unknown immediate state: 0x%02x\n", state); + isert_err("Unknown immediate state: 0x%02x\n", state); ret = -EINVAL; break; } @@ -2077,17 +2173,16 @@ isert_immediate_queue(struct iscsi_conn *conn, struct iscsi_cmd *cmd, int state) } static int -isert_response_queue(struct iscsi_conn *conn, struct iscsi_cmd *cmd, int state) +isert_response_queue(struct iscsit_conn *conn, struct iscsit_cmd *cmd, int state) { + struct isert_conn *isert_conn = conn->context; int ret; switch (state) { case ISTATE_SEND_LOGOUTRSP: ret = isert_put_logout_rsp(cmd, conn); - if (!ret) { - pr_debug("Returning iSER Logout -EAGAIN\n"); - ret = -EAGAIN; - } + if (!ret) + isert_conn->logout_posted = true; break; case ISTATE_SEND_NOPIN: ret = isert_put_nopin(cmd, conn, true); @@ -2109,7 +2204,7 @@ isert_response_queue(struct iscsi_conn *conn, struct iscsi_cmd *cmd, int state) ret = isert_put_response(conn, cmd); break; default: - pr_err("Unknown response state: 0x%02x\n", state); + isert_err("Unknown response state: 0x%02x\n", state); ret = -EINVAL; break; } @@ -2117,288 +2212,393 @@ isert_response_queue(struct iscsi_conn *conn, struct iscsi_cmd *cmd, int state) return ret; } +struct rdma_cm_id * +isert_setup_id(struct isert_np *isert_np) +{ + struct iscsi_np *np = isert_np->np; + struct rdma_cm_id *id; + struct sockaddr *sa; + int ret; + + sa = (struct sockaddr *)&np->np_sockaddr; + isert_dbg("ksockaddr: %p, sa: %p\n", &np->np_sockaddr, sa); + + id = rdma_create_id(&init_net, isert_cma_handler, isert_np, + RDMA_PS_TCP, IB_QPT_RC); + if (IS_ERR(id)) { + isert_err("rdma_create_id() failed: %ld\n", PTR_ERR(id)); + ret = PTR_ERR(id); + goto out; + } + isert_dbg("id %p context %p\n", id, id->context); + + /* + * Allow both IPv4 and IPv6 sockets to bind a single port + * at the same time. + */ + ret = rdma_set_afonly(id, 1); + if (ret) { + isert_err("rdma_set_afonly() failed: %d\n", ret); + goto out_id; + } + + ret = rdma_bind_addr(id, sa); + if (ret) { + isert_err("rdma_bind_addr() failed: %d\n", ret); + goto out_id; + } + + ret = rdma_listen(id, 0); + if (ret) { + isert_err("rdma_listen() failed: %d\n", ret); + goto out_id; + } + + return id; +out_id: + rdma_destroy_id(id); +out: + return ERR_PTR(ret); +} + static int isert_setup_np(struct iscsi_np *np, - struct __kernel_sockaddr_storage *ksockaddr) + struct sockaddr_storage *ksockaddr) { struct isert_np *isert_np; struct rdma_cm_id *isert_lid; - struct sockaddr *sa; int ret; isert_np = kzalloc(sizeof(struct isert_np), GFP_KERNEL); - if (!isert_np) { - pr_err("Unable to allocate struct isert_np\n"); + if (!isert_np) return -ENOMEM; - } - init_waitqueue_head(&isert_np->np_accept_wq); - mutex_init(&isert_np->np_accept_mutex); - INIT_LIST_HEAD(&isert_np->np_accept_list); - init_completion(&isert_np->np_login_comp); - sa = (struct sockaddr *)ksockaddr; - pr_debug("ksockaddr: %p, sa: %p\n", ksockaddr, sa); + sema_init(&isert_np->sem, 0); + mutex_init(&isert_np->mutex); + INIT_LIST_HEAD(&isert_np->accepted); + INIT_LIST_HEAD(&isert_np->pending); + isert_np->np = np; + /* * Setup the np->np_sockaddr from the passed sockaddr setup * in iscsi_target_configfs.c code.. */ memcpy(&np->np_sockaddr, ksockaddr, - sizeof(struct __kernel_sockaddr_storage)); + sizeof(struct sockaddr_storage)); - isert_lid = rdma_create_id(isert_cma_handler, np, RDMA_PS_TCP, - IB_QPT_RC); + isert_lid = isert_setup_id(isert_np); if (IS_ERR(isert_lid)) { - pr_err("rdma_create_id() for isert_listen_handler failed: %ld\n", - PTR_ERR(isert_lid)); ret = PTR_ERR(isert_lid); goto out; } - ret = rdma_bind_addr(isert_lid, sa); - if (ret) { - pr_err("rdma_bind_addr() for isert_lid failed: %d\n", ret); - goto out_lid; - } - - ret = rdma_listen(isert_lid, ISERT_RDMA_LISTEN_BACKLOG); - if (ret) { - pr_err("rdma_listen() for isert_lid failed: %d\n", ret); - goto out_lid; - } - - isert_np->np_cm_id = isert_lid; + isert_np->cm_id = isert_lid; np->np_context = isert_np; - pr_debug("Setup isert_lid->context: %p\n", isert_lid->context); return 0; -out_lid: - rdma_destroy_id(isert_lid); out: kfree(isert_np); - return ret; -} -static int -isert_check_accept_queue(struct isert_np *isert_np) -{ - int empty; - - mutex_lock(&isert_np->np_accept_mutex); - empty = list_empty(&isert_np->np_accept_list); - mutex_unlock(&isert_np->np_accept_mutex); - - return empty; + return ret; } static int isert_rdma_accept(struct isert_conn *isert_conn) { - struct rdma_cm_id *cm_id = isert_conn->conn_cm_id; + struct rdma_cm_id *cm_id = isert_conn->cm_id; struct rdma_conn_param cp; int ret; + struct iser_cm_hdr rsp_hdr; memset(&cp, 0, sizeof(struct rdma_conn_param)); - cp.responder_resources = isert_conn->responder_resources; cp.initiator_depth = isert_conn->initiator_depth; cp.retry_count = 7; cp.rnr_retry_count = 7; - pr_debug("Before rdma_accept >>>>>>>>>>>>>>>>>>>>.\n"); + memset(&rsp_hdr, 0, sizeof(rsp_hdr)); + rsp_hdr.flags = ISERT_ZBVA_NOT_USED; + if (!isert_conn->snd_w_inv) + rsp_hdr.flags = rsp_hdr.flags | ISERT_SEND_W_INV_NOT_USED; + cp.private_data = (void *)&rsp_hdr; + cp.private_data_len = sizeof(rsp_hdr); ret = rdma_accept(cm_id, &cp); if (ret) { - pr_err("rdma_accept() failed with: %d\n", ret); + isert_err("rdma_accept() failed with: %d\n", ret); return ret; } - pr_debug("After rdma_accept >>>>>>>>>>>>>>>>>>>>>.\n"); - return 0; } static int -isert_get_login_rx(struct iscsi_conn *conn, struct iscsi_login *login) +isert_get_login_rx(struct iscsit_conn *conn, struct iscsi_login *login) { - struct isert_conn *isert_conn = (struct isert_conn *)conn->context; + struct isert_conn *isert_conn = conn->context; int ret; - pr_debug("isert_get_login_rx before conn_login_comp conn: %p\n", conn); + isert_info("before login_req comp conn: %p\n", isert_conn); + ret = wait_for_completion_interruptible(&isert_conn->login_req_comp); + if (ret) { + isert_err("isert_conn %p interrupted before got login req\n", + isert_conn); + return ret; + } + reinit_completion(&isert_conn->login_req_comp); + + /* + * For login requests after the first PDU, isert_rx_login_req() will + * kick queue_delayed_work(isert_login_wq, &conn->login_work) as + * the packet is received, which turns this callback from + * iscsi_target_do_login_rx() into a NOP. + */ + if (!login->first_request) + return 0; + + isert_rx_login_req(isert_conn); - ret = wait_for_completion_interruptible(&isert_conn->conn_login_comp); + isert_info("before login_comp conn: %p\n", conn); + ret = wait_for_completion_interruptible(&isert_conn->login_comp); if (ret) return ret; - pr_debug("isert_get_login_rx processing login->req: %p\n", login->req); + isert_info("processing login->req: %p\n", login->req); + return 0; } static void -isert_set_conn_info(struct iscsi_np *np, struct iscsi_conn *conn, +isert_set_conn_info(struct iscsi_np *np, struct iscsit_conn *conn, struct isert_conn *isert_conn) { - struct rdma_cm_id *cm_id = isert_conn->conn_cm_id; + struct rdma_cm_id *cm_id = isert_conn->cm_id; struct rdma_route *cm_route = &cm_id->route; - struct sockaddr_in *sock_in; - struct sockaddr_in6 *sock_in6; conn->login_family = np->np_sockaddr.ss_family; - if (np->np_sockaddr.ss_family == AF_INET6) { - sock_in6 = (struct sockaddr_in6 *)&cm_route->addr.dst_addr; - snprintf(conn->login_ip, sizeof(conn->login_ip), "%pI6c", - &sock_in6->sin6_addr.in6_u); - conn->login_port = ntohs(sock_in6->sin6_port); - - sock_in6 = (struct sockaddr_in6 *)&cm_route->addr.src_addr; - snprintf(conn->local_ip, sizeof(conn->local_ip), "%pI6c", - &sock_in6->sin6_addr.in6_u); - conn->local_port = ntohs(sock_in6->sin6_port); - } else { - sock_in = (struct sockaddr_in *)&cm_route->addr.dst_addr; - sprintf(conn->login_ip, "%pI4", - &sock_in->sin_addr.s_addr); - conn->login_port = ntohs(sock_in->sin_port); - - sock_in = (struct sockaddr_in *)&cm_route->addr.src_addr; - sprintf(conn->local_ip, "%pI4", - &sock_in->sin_addr.s_addr); - conn->local_port = ntohs(sock_in->sin_port); - } + conn->login_sockaddr = cm_route->addr.dst_addr; + conn->local_sockaddr = cm_route->addr.src_addr; } static int -isert_accept_np(struct iscsi_np *np, struct iscsi_conn *conn) +isert_accept_np(struct iscsi_np *np, struct iscsit_conn *conn) { - struct isert_np *isert_np = (struct isert_np *)np->np_context; + struct isert_np *isert_np = np->np_context; struct isert_conn *isert_conn; - int max_accept = 0, ret; + int ret; accept_wait: - ret = wait_event_interruptible(isert_np->np_accept_wq, - !isert_check_accept_queue(isert_np) || - np->np_thread_state == ISCSI_NP_THREAD_RESET); - if (max_accept > 5) + ret = down_interruptible(&isert_np->sem); + if (ret) return -ENODEV; spin_lock_bh(&np->np_thread_lock); - if (np->np_thread_state == ISCSI_NP_THREAD_RESET) { + if (np->np_thread_state >= ISCSI_NP_THREAD_RESET) { spin_unlock_bh(&np->np_thread_lock); - pr_err("ISCSI_NP_THREAD_RESET for isert_accept_np\n"); + isert_dbg("np_thread_state %d\n", + np->np_thread_state); + /* + * No point in stalling here when np_thread + * is in state RESET/SHUTDOWN/EXIT - bail + */ return -ENODEV; } spin_unlock_bh(&np->np_thread_lock); - mutex_lock(&isert_np->np_accept_mutex); - if (list_empty(&isert_np->np_accept_list)) { - mutex_unlock(&isert_np->np_accept_mutex); - max_accept++; + mutex_lock(&isert_np->mutex); + if (list_empty(&isert_np->pending)) { + mutex_unlock(&isert_np->mutex); goto accept_wait; } - isert_conn = list_first_entry(&isert_np->np_accept_list, - struct isert_conn, conn_accept_node); - list_del_init(&isert_conn->conn_accept_node); - mutex_unlock(&isert_np->np_accept_mutex); + isert_conn = list_first_entry(&isert_np->pending, + struct isert_conn, node); + list_del_init(&isert_conn->node); + mutex_unlock(&isert_np->mutex); conn->context = isert_conn; isert_conn->conn = conn; - max_accept = 0; - - ret = isert_rdma_post_recvl(isert_conn); - if (ret) - return ret; - - ret = isert_rdma_accept(isert_conn); - if (ret) - return ret; + isert_conn->state = ISER_CONN_BOUND; isert_set_conn_info(np, conn, isert_conn); - pr_debug("Processing isert_accept_np: isert_conn: %p\n", isert_conn); + isert_dbg("Processing isert_conn: %p\n", isert_conn); + return 0; } static void isert_free_np(struct iscsi_np *np) { - struct isert_np *isert_np = (struct isert_np *)np->np_context; + struct isert_np *isert_np = np->np_context; + struct isert_conn *isert_conn, *n; + LIST_HEAD(drop_conn_list); - rdma_destroy_id(isert_np->np_cm_id); + if (isert_np->cm_id) + rdma_destroy_id(isert_np->cm_id); + + /* + * FIXME: At this point we don't have a good way to insure + * that at this point we don't have hanging connections that + * completed RDMA establishment but didn't start iscsi login + * process. So work-around this by cleaning up what ever piled + * up in accepted and pending lists. + */ + mutex_lock(&isert_np->mutex); + if (!list_empty(&isert_np->pending)) { + isert_info("Still have isert pending connections\n"); + list_for_each_entry_safe(isert_conn, n, + &isert_np->pending, + node) { + isert_info("cleaning isert_conn %p state (%d)\n", + isert_conn, isert_conn->state); + list_move_tail(&isert_conn->node, &drop_conn_list); + } + } + + if (!list_empty(&isert_np->accepted)) { + isert_info("Still have isert accepted connections\n"); + list_for_each_entry_safe(isert_conn, n, + &isert_np->accepted, + node) { + isert_info("cleaning isert_conn %p state (%d)\n", + isert_conn, isert_conn->state); + list_move_tail(&isert_conn->node, &drop_conn_list); + } + } + mutex_unlock(&isert_np->mutex); + + list_for_each_entry_safe(isert_conn, n, &drop_conn_list, node) { + list_del_init(&isert_conn->node); + isert_connect_release(isert_conn); + } np->np_context = NULL; kfree(isert_np); } -static int isert_check_state(struct isert_conn *isert_conn, int state) +static void isert_release_work(struct work_struct *work) { - int ret; + struct isert_conn *isert_conn = container_of(work, + struct isert_conn, + release_work); - mutex_lock(&isert_conn->conn_mutex); - ret = (isert_conn->state == state); - mutex_unlock(&isert_conn->conn_mutex); + isert_info("Starting release conn %p\n", isert_conn); - return ret; + mutex_lock(&isert_conn->mutex); + isert_conn->state = ISER_CONN_DOWN; + mutex_unlock(&isert_conn->mutex); + + isert_info("Destroying conn %p\n", isert_conn); + isert_put_conn(isert_conn); } -static void isert_free_conn(struct iscsi_conn *conn) +static void +isert_wait4logout(struct isert_conn *isert_conn) { - struct isert_conn *isert_conn = conn->context; + struct iscsit_conn *conn = isert_conn->conn; - pr_debug("isert_free_conn: Starting \n"); - /* - * Decrement post_send_buf_count for special case when called - * from isert_do_control_comp() -> iscsit_logout_post_handler() - */ - mutex_lock(&isert_conn->conn_mutex); - if (isert_conn->logout_posted) - atomic_dec(&isert_conn->post_send_buf_count); + isert_info("conn %p\n", isert_conn); - if (isert_conn->conn_cm_id && isert_conn->state != ISER_CONN_DOWN) { - pr_debug("Calling rdma_disconnect from isert_free_conn\n"); - rdma_disconnect(isert_conn->conn_cm_id); + if (isert_conn->logout_posted) { + isert_info("conn %p wait for conn_logout_comp\n", isert_conn); + wait_for_completion_timeout(&conn->conn_logout_comp, + SECONDS_FOR_LOGOUT_COMP * HZ); } - /* - * Only wait for conn_wait_comp_err if the isert_conn made it - * into full feature phase.. - */ - if (isert_conn->state == ISER_CONN_UP) { - pr_debug("isert_free_conn: Before wait_event comp_err %d\n", - isert_conn->state); - mutex_unlock(&isert_conn->conn_mutex); +} - wait_event(isert_conn->conn_wait_comp_err, - (isert_check_state(isert_conn, ISER_CONN_TERMINATING))); +static void +isert_wait4cmds(struct iscsit_conn *conn) +{ + isert_info("iscsit_conn %p\n", conn); - wait_event(isert_conn->conn_wait, - (isert_check_state(isert_conn, ISER_CONN_DOWN))); + if (conn->sess) { + target_stop_cmd_counter(conn->cmd_cnt); + target_wait_for_cmds(conn->cmd_cnt); + } +} - isert_put_conn(isert_conn); - return; +/** + * isert_put_unsol_pending_cmds() - Drop commands waiting for + * unsolicitate dataout + * @conn: iscsi connection + * + * We might still have commands that are waiting for unsolicited + * dataouts messages. We must put the extra reference on those + * before blocking on the target_wait_for_session_cmds + */ +static void +isert_put_unsol_pending_cmds(struct iscsit_conn *conn) +{ + struct iscsit_cmd *cmd, *tmp; + static LIST_HEAD(drop_cmd_list); + + spin_lock_bh(&conn->cmd_lock); + list_for_each_entry_safe(cmd, tmp, &conn->conn_cmd_list, i_conn_node) { + if ((cmd->cmd_flags & ICF_NON_IMMEDIATE_UNSOLICITED_DATA) && + (cmd->write_data_done < conn->sess->sess_ops->FirstBurstLength) && + (cmd->write_data_done < cmd->se_cmd.data_length)) + list_move_tail(&cmd->i_conn_node, &drop_cmd_list); } - if (isert_conn->state == ISER_CONN_INIT) { - mutex_unlock(&isert_conn->conn_mutex); - isert_put_conn(isert_conn); - return; + spin_unlock_bh(&conn->cmd_lock); + + list_for_each_entry_safe(cmd, tmp, &drop_cmd_list, i_conn_node) { + list_del_init(&cmd->i_conn_node); + if (cmd->i_state != ISTATE_REMOVE) { + struct isert_cmd *isert_cmd = iscsit_priv_cmd(cmd); + + isert_info("conn %p dropping cmd %p\n", conn, cmd); + isert_put_cmd(isert_cmd, true); + } } - pr_debug("isert_free_conn: wait_event conn_wait %d\n", - isert_conn->state); - mutex_unlock(&isert_conn->conn_mutex); +} - wait_event(isert_conn->conn_wait, - (isert_check_state(isert_conn, ISER_CONN_DOWN))); +static void isert_wait_conn(struct iscsit_conn *conn) +{ + struct isert_conn *isert_conn = conn->context; + isert_info("Starting conn %p\n", isert_conn); + + mutex_lock(&isert_conn->mutex); + isert_conn_terminate(isert_conn); + mutex_unlock(&isert_conn->mutex); + + ib_drain_qp(isert_conn->qp); + isert_put_unsol_pending_cmds(conn); + isert_wait4cmds(conn); + isert_wait4logout(isert_conn); + + queue_work(isert_release_wq, &isert_conn->release_work); +} + +static void isert_free_conn(struct iscsit_conn *conn) +{ + struct isert_conn *isert_conn = conn->context; + + ib_drain_qp(isert_conn->qp); isert_put_conn(isert_conn); } +static void isert_get_rx_pdu(struct iscsit_conn *conn) +{ + struct completion comp; + + init_completion(&comp); + + wait_for_completion_interruptible(&comp); +} + static struct iscsit_transport iser_target_transport = { .name = "IB/iSER", .transport_type = ISCSI_INFINIBAND, + .rdma_shutdown = true, + .priv_size = sizeof(struct isert_cmd), .owner = THIS_MODULE, .iscsit_setup_np = isert_setup_np, .iscsit_accept_np = isert_accept_np, .iscsit_free_np = isert_free_np, + .iscsit_wait_conn = isert_wait_conn, .iscsit_free_conn = isert_free_conn, - .iscsit_alloc_cmd = isert_alloc_cmd, .iscsit_get_login_rx = isert_get_login_rx, .iscsit_put_login_tx = isert_put_login_tx, .iscsit_immediate_queue = isert_immediate_queue, @@ -2406,56 +2606,57 @@ static struct iscsit_transport iser_target_transport = { .iscsit_get_dataout = isert_get_dataout, .iscsit_queue_data_in = isert_put_datain, .iscsit_queue_status = isert_put_response, + .iscsit_aborted_task = isert_aborted_task, + .iscsit_get_rx_pdu = isert_get_rx_pdu, + .iscsit_get_sup_prot_ops = isert_get_sup_prot_ops, }; static int __init isert_init(void) { - int ret; - - isert_rx_wq = alloc_workqueue("isert_rx_wq", 0, 0); - if (!isert_rx_wq) { - pr_err("Unable to allocate isert_rx_wq\n"); + isert_login_wq = alloc_workqueue("isert_login_wq", WQ_PERCPU, 0); + if (!isert_login_wq) { + isert_err("Unable to allocate isert_login_wq\n"); return -ENOMEM; } - isert_comp_wq = alloc_workqueue("isert_comp_wq", 0, 0); + isert_comp_wq = alloc_workqueue("isert_comp_wq", + WQ_UNBOUND | WQ_HIGHPRI, 0); if (!isert_comp_wq) { - pr_err("Unable to allocate isert_comp_wq\n"); - ret = -ENOMEM; - goto destroy_rx_wq; + isert_err("Unable to allocate isert_comp_wq\n"); + goto destroy_login_wq; } - isert_cmd_cache = kmem_cache_create("isert_cmd_cache", - sizeof(struct isert_cmd), __alignof__(struct isert_cmd), - 0, NULL); - if (!isert_cmd_cache) { - pr_err("Unable to create isert_cmd_cache\n"); - ret = -ENOMEM; - goto destroy_tx_cq; + isert_release_wq = alloc_workqueue("isert_release_wq", WQ_UNBOUND, + WQ_UNBOUND_MAX_ACTIVE); + if (!isert_release_wq) { + isert_err("Unable to allocate isert_release_wq\n"); + goto destroy_comp_wq; } iscsit_register_transport(&iser_target_transport); - pr_debug("iSER_TARGET[0] - Loaded iser_target_transport\n"); + isert_info("iSER_TARGET[0] - Loaded iser_target_transport\n"); + return 0; -destroy_tx_cq: +destroy_comp_wq: destroy_workqueue(isert_comp_wq); -destroy_rx_wq: - destroy_workqueue(isert_rx_wq); - return ret; +destroy_login_wq: + destroy_workqueue(isert_login_wq); + + return -ENOMEM; } static void __exit isert_exit(void) { - kmem_cache_destroy(isert_cmd_cache); + flush_workqueue(isert_login_wq); + destroy_workqueue(isert_release_wq); destroy_workqueue(isert_comp_wq); - destroy_workqueue(isert_rx_wq); iscsit_unregister_transport(&iser_target_transport); - pr_debug("iSER_TARGET[0] - Released iser_target_transport\n"); + isert_info("iSER_TARGET[0] - Released iser_target_transport\n"); + destroy_workqueue(isert_login_wq); } MODULE_DESCRIPTION("iSER-Target for mainline target infrastructure"); -MODULE_VERSION("0.1"); MODULE_AUTHOR("nab@Linux-iSCSI.org"); MODULE_LICENSE("GPL"); |