diff options
Diffstat (limited to 'drivers/infiniband')
274 files changed, 29562 insertions, 7224 deletions
diff --git a/drivers/infiniband/Kconfig b/drivers/infiniband/Kconfig index fb3fb89640e5..66f86027ed47 100644 --- a/drivers/infiniband/Kconfig +++ b/drivers/infiniband/Kconfig @@ -73,6 +73,7 @@ source "drivers/infiniband/hw/mlx4/Kconfig" source "drivers/infiniband/hw/mlx5/Kconfig" source "drivers/infiniband/hw/nes/Kconfig" source "drivers/infiniband/hw/ocrdma/Kconfig" +source "drivers/infiniband/hw/vmw_pvrdma/Kconfig" source "drivers/infiniband/hw/usnic/Kconfig" source "drivers/infiniband/hw/hns/Kconfig" @@ -91,4 +92,6 @@ source "drivers/infiniband/hw/hfi1/Kconfig" source "drivers/infiniband/hw/qedr/Kconfig" +source "drivers/infiniband/hw/bnxt_re/Kconfig" + endif # INFINIBAND diff --git a/drivers/infiniband/core/Makefile b/drivers/infiniband/core/Makefile index edaae9f9853c..e426ac877d19 100644 --- a/drivers/infiniband/core/Makefile +++ b/drivers/infiniband/core/Makefile @@ -13,6 +13,7 @@ ib_core-y := packer.o ud_header.o verbs.o cq.o rw.o sysfs.o \ multicast.o mad.o smi.o agent.o mad_rmpp.o ib_core-$(CONFIG_INFINIBAND_USER_MEM) += umem.o ib_core-$(CONFIG_INFINIBAND_ON_DEMAND_PAGING) += umem_odp.o umem_rbtree.o +ib_core-$(CONFIG_CGROUP_RDMA) += cgroup.o ib_cm-y := cm.o diff --git a/drivers/infiniband/core/agent.c b/drivers/infiniband/core/agent.c index 4fa524dfb6cf..11dacd97a667 100644 --- a/drivers/infiniband/core/agent.c +++ b/drivers/infiniband/core/agent.c @@ -156,7 +156,6 @@ int ib_agent_port_open(struct ib_device *device, int port_num) /* Create new device info */ port_priv = kzalloc(sizeof *port_priv, GFP_KERNEL); if (!port_priv) { - dev_err(&device->dev, "No memory for ib_agent_port_private\n"); ret = -ENOMEM; goto error1; } diff --git a/drivers/infiniband/core/cache.c b/drivers/infiniband/core/cache.c index 1a2984c28b95..b1371eb9f46c 100644 --- a/drivers/infiniband/core/cache.c +++ b/drivers/infiniband/core/cache.c @@ -314,14 +314,13 @@ static void make_default_gid(struct net_device *dev, union ib_gid *gid) int ib_cache_gid_add(struct ib_device *ib_dev, u8 port, union ib_gid *gid, struct ib_gid_attr *attr) { - struct ib_gid_table **ports_table = ib_dev->cache.gid_cache; struct ib_gid_table *table; int ix; int ret = 0; struct net_device *idev; int empty; - table = ports_table[port - rdma_start_port(ib_dev)]; + table = ib_dev->cache.ports[port - rdma_start_port(ib_dev)].gid; if (!memcmp(gid, &zgid, sizeof(*gid))) return -EINVAL; @@ -369,11 +368,10 @@ out_unlock: int ib_cache_gid_del(struct ib_device *ib_dev, u8 port, union ib_gid *gid, struct ib_gid_attr *attr) { - struct ib_gid_table **ports_table = ib_dev->cache.gid_cache; struct ib_gid_table *table; int ix; - table = ports_table[port - rdma_start_port(ib_dev)]; + table = ib_dev->cache.ports[port - rdma_start_port(ib_dev)].gid; mutex_lock(&table->lock); write_lock_irq(&table->rwlock); @@ -399,12 +397,11 @@ out_unlock: int ib_cache_gid_del_all_netdev_gids(struct ib_device *ib_dev, u8 port, struct net_device *ndev) { - struct ib_gid_table **ports_table = ib_dev->cache.gid_cache; struct ib_gid_table *table; int ix; bool deleted = false; - table = ports_table[port - rdma_start_port(ib_dev)]; + table = ib_dev->cache.ports[port - rdma_start_port(ib_dev)].gid; mutex_lock(&table->lock); write_lock_irq(&table->rwlock); @@ -428,10 +425,9 @@ int ib_cache_gid_del_all_netdev_gids(struct ib_device *ib_dev, u8 port, static int __ib_cache_gid_get(struct ib_device *ib_dev, u8 port, int index, union ib_gid *gid, struct ib_gid_attr *attr) { - struct ib_gid_table **ports_table = ib_dev->cache.gid_cache; struct ib_gid_table *table; - table = ports_table[port - rdma_start_port(ib_dev)]; + table = ib_dev->cache.ports[port - rdma_start_port(ib_dev)].gid; if (index < 0 || index >= table->sz) return -EINVAL; @@ -455,14 +451,13 @@ static int _ib_cache_gid_table_find(struct ib_device *ib_dev, unsigned long mask, u8 *port, u16 *index) { - struct ib_gid_table **ports_table = ib_dev->cache.gid_cache; struct ib_gid_table *table; u8 p; int local_index; unsigned long flags; for (p = 0; p < ib_dev->phys_port_cnt; p++) { - table = ports_table[p]; + table = ib_dev->cache.ports[p].gid; read_lock_irqsave(&table->rwlock, flags); local_index = find_gid(table, gid, val, false, mask, NULL); if (local_index >= 0) { @@ -503,18 +498,16 @@ int ib_find_cached_gid_by_port(struct ib_device *ib_dev, u16 *index) { int local_index; - struct ib_gid_table **ports_table = ib_dev->cache.gid_cache; struct ib_gid_table *table; unsigned long mask = GID_ATTR_FIND_MASK_GID | GID_ATTR_FIND_MASK_GID_TYPE; struct ib_gid_attr val = {.ndev = ndev, .gid_type = gid_type}; unsigned long flags; - if (port < rdma_start_port(ib_dev) || - port > rdma_end_port(ib_dev)) + if (!rdma_is_port_valid(ib_dev, port)) return -ENOENT; - table = ports_table[port - rdma_start_port(ib_dev)]; + table = ib_dev->cache.ports[port - rdma_start_port(ib_dev)].gid; if (ndev) mask |= GID_ATTR_FIND_MASK_NETDEV; @@ -562,21 +555,17 @@ static int ib_cache_gid_find_by_filter(struct ib_device *ib_dev, void *context, u16 *index) { - struct ib_gid_table **ports_table = ib_dev->cache.gid_cache; struct ib_gid_table *table; unsigned int i; unsigned long flags; bool found = false; - if (!ports_table) - return -EOPNOTSUPP; - if (port < rdma_start_port(ib_dev) || - port > rdma_end_port(ib_dev) || + if (!rdma_is_port_valid(ib_dev, port) || !rdma_protocol_roce(ib_dev, port)) return -EPROTONOSUPPORT; - table = ports_table[port - rdma_start_port(ib_dev)]; + table = ib_dev->cache.ports[port - rdma_start_port(ib_dev)].gid; read_lock_irqsave(&table->rwlock, flags); for (i = 0; i < table->sz; i++) { @@ -668,14 +657,13 @@ void ib_cache_gid_set_default_gid(struct ib_device *ib_dev, u8 port, unsigned long gid_type_mask, enum ib_cache_gid_default_mode mode) { - struct ib_gid_table **ports_table = ib_dev->cache.gid_cache; union ib_gid gid; struct ib_gid_attr gid_attr; struct ib_gid_attr zattr_type = zattr; struct ib_gid_table *table; unsigned int gid_type; - table = ports_table[port - rdma_start_port(ib_dev)]; + table = ib_dev->cache.ports[port - rdma_start_port(ib_dev)].gid; make_default_gid(ndev, &gid); memset(&gid_attr, 0, sizeof(gid_attr)); @@ -766,75 +754,64 @@ static int gid_table_reserve_default(struct ib_device *ib_dev, u8 port, static int _gid_table_setup_one(struct ib_device *ib_dev) { u8 port; - struct ib_gid_table **table; + struct ib_gid_table *table; int err = 0; - table = kcalloc(ib_dev->phys_port_cnt, sizeof(*table), GFP_KERNEL); - - if (!table) { - pr_warn("failed to allocate ib gid cache for %s\n", - ib_dev->name); - return -ENOMEM; - } - for (port = 0; port < ib_dev->phys_port_cnt; port++) { u8 rdma_port = port + rdma_start_port(ib_dev); - table[port] = + table = alloc_gid_table( ib_dev->port_immutable[rdma_port].gid_tbl_len); - if (!table[port]) { + if (!table) { err = -ENOMEM; goto rollback_table_setup; } err = gid_table_reserve_default(ib_dev, port + rdma_start_port(ib_dev), - table[port]); + table); if (err) goto rollback_table_setup; + ib_dev->cache.ports[port].gid = table; } - ib_dev->cache.gid_cache = table; return 0; rollback_table_setup: for (port = 0; port < ib_dev->phys_port_cnt; port++) { + table = ib_dev->cache.ports[port].gid; + cleanup_gid_table_port(ib_dev, port + rdma_start_port(ib_dev), - table[port]); - release_gid_table(table[port]); + table); + release_gid_table(table); } - kfree(table); return err; } static void gid_table_release_one(struct ib_device *ib_dev) { - struct ib_gid_table **table = ib_dev->cache.gid_cache; + struct ib_gid_table *table; u8 port; - if (!table) - return; - - for (port = 0; port < ib_dev->phys_port_cnt; port++) - release_gid_table(table[port]); - - kfree(table); - ib_dev->cache.gid_cache = NULL; + for (port = 0; port < ib_dev->phys_port_cnt; port++) { + table = ib_dev->cache.ports[port].gid; + release_gid_table(table); + ib_dev->cache.ports[port].gid = NULL; + } } static void gid_table_cleanup_one(struct ib_device *ib_dev) { - struct ib_gid_table **table = ib_dev->cache.gid_cache; + struct ib_gid_table *table; u8 port; - if (!table) - return; - - for (port = 0; port < ib_dev->phys_port_cnt; port++) + for (port = 0; port < ib_dev->phys_port_cnt; port++) { + table = ib_dev->cache.ports[port].gid; cleanup_gid_table_port(ib_dev, port + rdma_start_port(ib_dev), - table[port]); + table); + } } static int gid_table_setup_one(struct ib_device *ib_dev) @@ -864,12 +841,12 @@ int ib_get_cached_gid(struct ib_device *device, { int res; unsigned long flags; - struct ib_gid_table **ports_table = device->cache.gid_cache; - struct ib_gid_table *table = ports_table[port_num - rdma_start_port(device)]; + struct ib_gid_table *table; - if (port_num < rdma_start_port(device) || port_num > rdma_end_port(device)) + if (!rdma_is_port_valid(device, port_num)) return -EINVAL; + table = device->cache.ports[port_num - rdma_start_port(device)].gid; read_lock_irqsave(&table->rwlock, flags); res = __ib_cache_gid_get(device, port_num, index, gid, gid_attr); read_unlock_irqrestore(&table->rwlock, flags); @@ -916,12 +893,12 @@ int ib_get_cached_pkey(struct ib_device *device, unsigned long flags; int ret = 0; - if (port_num < rdma_start_port(device) || port_num > rdma_end_port(device)) + if (!rdma_is_port_valid(device, port_num)) return -EINVAL; read_lock_irqsave(&device->cache.lock, flags); - cache = device->cache.pkey_cache[port_num - rdma_start_port(device)]; + cache = device->cache.ports[port_num - rdma_start_port(device)].pkey; if (index < 0 || index >= cache->table_len) ret = -EINVAL; @@ -945,12 +922,12 @@ int ib_find_cached_pkey(struct ib_device *device, int ret = -ENOENT; int partial_ix = -1; - if (port_num < rdma_start_port(device) || port_num > rdma_end_port(device)) + if (!rdma_is_port_valid(device, port_num)) return -EINVAL; read_lock_irqsave(&device->cache.lock, flags); - cache = device->cache.pkey_cache[port_num - rdma_start_port(device)]; + cache = device->cache.ports[port_num - rdma_start_port(device)].pkey; *index = -1; @@ -985,12 +962,12 @@ int ib_find_exact_cached_pkey(struct ib_device *device, int i; int ret = -ENOENT; - if (port_num < rdma_start_port(device) || port_num > rdma_end_port(device)) + if (!rdma_is_port_valid(device, port_num)) return -EINVAL; read_lock_irqsave(&device->cache.lock, flags); - cache = device->cache.pkey_cache[port_num - rdma_start_port(device)]; + cache = device->cache.ports[port_num - rdma_start_port(device)].pkey; *index = -1; @@ -1014,17 +991,36 @@ int ib_get_cached_lmc(struct ib_device *device, unsigned long flags; int ret = 0; - if (port_num < rdma_start_port(device) || port_num > rdma_end_port(device)) + if (!rdma_is_port_valid(device, port_num)) return -EINVAL; read_lock_irqsave(&device->cache.lock, flags); - *lmc = device->cache.lmc_cache[port_num - rdma_start_port(device)]; + *lmc = device->cache.ports[port_num - rdma_start_port(device)].lmc; read_unlock_irqrestore(&device->cache.lock, flags); return ret; } EXPORT_SYMBOL(ib_get_cached_lmc); +int ib_get_cached_port_state(struct ib_device *device, + u8 port_num, + enum ib_port_state *port_state) +{ + unsigned long flags; + int ret = 0; + + if (port_num < rdma_start_port(device) || port_num > rdma_end_port(device)) + return -EINVAL; + + read_lock_irqsave(&device->cache.lock, flags); + *port_state = device->cache.ports[port_num + - rdma_start_port(device)].port_state; + read_unlock_irqrestore(&device->cache.lock, flags); + + return ret; +} +EXPORT_SYMBOL(ib_get_cached_port_state); + static void ib_cache_update(struct ib_device *device, u8 port) { @@ -1037,14 +1033,13 @@ static void ib_cache_update(struct ib_device *device, int i; int ret; struct ib_gid_table *table; - struct ib_gid_table **ports_table = device->cache.gid_cache; bool use_roce_gid_table = rdma_cap_roce_gid_table(device, port); - if (port < rdma_start_port(device) || port > rdma_end_port(device)) + if (!rdma_is_port_valid(device, port)) return; - table = ports_table[port - rdma_start_port(device)]; + table = device->cache.ports[port - rdma_start_port(device)].gid; tprops = kmalloc(sizeof *tprops, GFP_KERNEL); if (!tprops) @@ -1096,9 +1091,10 @@ static void ib_cache_update(struct ib_device *device, write_lock_irq(&device->cache.lock); - old_pkey_cache = device->cache.pkey_cache[port - rdma_start_port(device)]; + old_pkey_cache = device->cache.ports[port - + rdma_start_port(device)].pkey; - device->cache.pkey_cache[port - rdma_start_port(device)] = pkey_cache; + device->cache.ports[port - rdma_start_port(device)].pkey = pkey_cache; if (!use_roce_gid_table) { write_lock(&table->rwlock); for (i = 0; i < gid_cache->table_len; i++) { @@ -1108,7 +1104,9 @@ static void ib_cache_update(struct ib_device *device, write_unlock(&table->rwlock); } - device->cache.lmc_cache[port - rdma_start_port(device)] = tprops->lmc; + device->cache.ports[port - rdma_start_port(device)].lmc = tprops->lmc; + device->cache.ports[port - rdma_start_port(device)].port_state = + tprops->state; write_unlock_irq(&device->cache.lock); @@ -1161,23 +1159,17 @@ int ib_cache_setup_one(struct ib_device *device) rwlock_init(&device->cache.lock); - device->cache.pkey_cache = - kzalloc(sizeof *device->cache.pkey_cache * + device->cache.ports = + kzalloc(sizeof(*device->cache.ports) * (rdma_end_port(device) - rdma_start_port(device) + 1), GFP_KERNEL); - device->cache.lmc_cache = kmalloc(sizeof *device->cache.lmc_cache * - (rdma_end_port(device) - - rdma_start_port(device) + 1), - GFP_KERNEL); - if (!device->cache.pkey_cache || - !device->cache.lmc_cache) { - pr_warn("Couldn't allocate cache for %s\n", device->name); - return -ENOMEM; + if (!device->cache.ports) { + err = -ENOMEM; + goto out; } err = gid_table_setup_one(device); if (err) - /* Allocated memory will be cleaned in the release function */ - return err; + goto out; for (p = 0; p <= rdma_end_port(device) - rdma_start_port(device); ++p) ib_cache_update(device, p + rdma_start_port(device)); @@ -1192,6 +1184,7 @@ int ib_cache_setup_one(struct ib_device *device) err: gid_table_cleanup_one(device); +out: return err; } @@ -1205,14 +1198,11 @@ void ib_cache_release_one(struct ib_device *device) * all the device's resources when the cache could no * longer be accessed. */ - if (device->cache.pkey_cache) - for (p = 0; - p <= rdma_end_port(device) - rdma_start_port(device); ++p) - kfree(device->cache.pkey_cache[p]); + for (p = 0; p <= rdma_end_port(device) - rdma_start_port(device); ++p) + kfree(device->cache.ports[p].pkey); gid_table_release_one(device); - kfree(device->cache.pkey_cache); - kfree(device->cache.lmc_cache); + kfree(device->cache.ports); } void ib_cache_cleanup_one(struct ib_device *device) diff --git a/drivers/infiniband/core/cgroup.c b/drivers/infiniband/core/cgroup.c new file mode 100644 index 000000000000..126ac5f99db7 --- /dev/null +++ b/drivers/infiniband/core/cgroup.c @@ -0,0 +1,62 @@ +/* + * Copyright (C) 2016 Parav Pandit <pandit.parav@gmail.com> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope 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 "core_priv.h" + +/** + * ib_device_register_rdmacg - register with rdma cgroup. + * @device: device to register to participate in resource + * accounting by rdma cgroup. + * + * Register with the rdma cgroup. Should be called before + * exposing rdma device to user space applications to avoid + * resource accounting leak. + * Returns 0 on success or otherwise failure code. + */ +int ib_device_register_rdmacg(struct ib_device *device) +{ + device->cg_device.name = device->name; + return rdmacg_register_device(&device->cg_device); +} + +/** + * ib_device_unregister_rdmacg - unregister with rdma cgroup. + * @device: device to unregister. + * + * Unregister with the rdma cgroup. Should be called after + * all the resources are deallocated, and after a stage when any + * other resource allocation by user application cannot be done + * for this device to avoid any leak in accounting. + */ +void ib_device_unregister_rdmacg(struct ib_device *device) +{ + rdmacg_unregister_device(&device->cg_device); +} + +int ib_rdmacg_try_charge(struct ib_rdmacg_object *cg_obj, + struct ib_device *device, + enum rdmacg_resource_type resource_index) +{ + return rdmacg_try_charge(&cg_obj->cg, &device->cg_device, + resource_index); +} +EXPORT_SYMBOL(ib_rdmacg_try_charge); + +void ib_rdmacg_uncharge(struct ib_rdmacg_object *cg_obj, + struct ib_device *device, + enum rdmacg_resource_type resource_index) +{ + rdmacg_uncharge(cg_obj->cg, &device->cg_device, + resource_index); +} +EXPORT_SYMBOL(ib_rdmacg_uncharge); diff --git a/drivers/infiniband/core/cm.c b/drivers/infiniband/core/cm.c index 71c7c4c328ef..6535f09dc575 100644 --- a/drivers/infiniband/core/cm.c +++ b/drivers/infiniband/core/cm.c @@ -57,6 +57,54 @@ MODULE_AUTHOR("Sean Hefty"); MODULE_DESCRIPTION("InfiniBand CM"); MODULE_LICENSE("Dual BSD/GPL"); +static const char * const ibcm_rej_reason_strs[] = { + [IB_CM_REJ_NO_QP] = "no QP", + [IB_CM_REJ_NO_EEC] = "no EEC", + [IB_CM_REJ_NO_RESOURCES] = "no resources", + [IB_CM_REJ_TIMEOUT] = "timeout", + [IB_CM_REJ_UNSUPPORTED] = "unsupported", + [IB_CM_REJ_INVALID_COMM_ID] = "invalid comm ID", + [IB_CM_REJ_INVALID_COMM_INSTANCE] = "invalid comm instance", + [IB_CM_REJ_INVALID_SERVICE_ID] = "invalid service ID", + [IB_CM_REJ_INVALID_TRANSPORT_TYPE] = "invalid transport type", + [IB_CM_REJ_STALE_CONN] = "stale conn", + [IB_CM_REJ_RDC_NOT_EXIST] = "RDC not exist", + [IB_CM_REJ_INVALID_GID] = "invalid GID", + [IB_CM_REJ_INVALID_LID] = "invalid LID", + [IB_CM_REJ_INVALID_SL] = "invalid SL", + [IB_CM_REJ_INVALID_TRAFFIC_CLASS] = "invalid traffic class", + [IB_CM_REJ_INVALID_HOP_LIMIT] = "invalid hop limit", + [IB_CM_REJ_INVALID_PACKET_RATE] = "invalid packet rate", + [IB_CM_REJ_INVALID_ALT_GID] = "invalid alt GID", + [IB_CM_REJ_INVALID_ALT_LID] = "invalid alt LID", + [IB_CM_REJ_INVALID_ALT_SL] = "invalid alt SL", + [IB_CM_REJ_INVALID_ALT_TRAFFIC_CLASS] = "invalid alt traffic class", + [IB_CM_REJ_INVALID_ALT_HOP_LIMIT] = "invalid alt hop limit", + [IB_CM_REJ_INVALID_ALT_PACKET_RATE] = "invalid alt packet rate", + [IB_CM_REJ_PORT_CM_REDIRECT] = "port CM redirect", + [IB_CM_REJ_PORT_REDIRECT] = "port redirect", + [IB_CM_REJ_INVALID_MTU] = "invalid MTU", + [IB_CM_REJ_INSUFFICIENT_RESP_RESOURCES] = "insufficient resp resources", + [IB_CM_REJ_CONSUMER_DEFINED] = "consumer defined", + [IB_CM_REJ_INVALID_RNR_RETRY] = "invalid RNR retry", + [IB_CM_REJ_DUPLICATE_LOCAL_COMM_ID] = "duplicate local comm ID", + [IB_CM_REJ_INVALID_CLASS_VERSION] = "invalid class version", + [IB_CM_REJ_INVALID_FLOW_LABEL] = "invalid flow label", + [IB_CM_REJ_INVALID_ALT_FLOW_LABEL] = "invalid alt flow label", +}; + +const char *__attribute_const__ ibcm_reject_msg(int reason) +{ + size_t index = reason; + + if (index < ARRAY_SIZE(ibcm_rej_reason_strs) && + ibcm_rej_reason_strs[index]) + return ibcm_rej_reason_strs[index]; + else + return "unrecognized reason"; +} +EXPORT_SYMBOL(ibcm_reject_msg); + static void cm_add_one(struct ib_device *device); static void cm_remove_one(struct ib_device *device, void *client_data); @@ -1582,6 +1630,7 @@ static struct cm_id_private * cm_match_req(struct cm_work *work, struct cm_id_private *listen_cm_id_priv, *cur_cm_id_priv; struct cm_timewait_info *timewait_info; struct cm_req_msg *req_msg; + struct ib_cm_id *cm_id; req_msg = (struct cm_req_msg *)work->mad_recv_wc->recv_buf.mad; @@ -1603,10 +1652,18 @@ static struct cm_id_private * cm_match_req(struct cm_work *work, timewait_info = cm_insert_remote_qpn(cm_id_priv->timewait_info); if (timewait_info) { cm_cleanup_timewait(cm_id_priv->timewait_info); + cur_cm_id_priv = cm_get_id(timewait_info->work.local_id, + timewait_info->work.remote_id); + spin_unlock_irq(&cm.lock); cm_issue_rej(work->port, work->mad_recv_wc, IB_CM_REJ_STALE_CONN, CM_MSG_RESPONSE_REQ, NULL, 0); + if (cur_cm_id_priv) { + cm_id = &cur_cm_id_priv->id; + ib_send_cm_dreq(cm_id, NULL, 0); + cm_deref_id(cur_cm_id_priv); + } return NULL; } @@ -1984,6 +2041,9 @@ static int cm_rep_handler(struct cm_work *work) struct cm_id_private *cm_id_priv; struct cm_rep_msg *rep_msg; int ret; + struct cm_id_private *cur_cm_id_priv; + struct ib_cm_id *cm_id; + struct cm_timewait_info *timewait_info; rep_msg = (struct cm_rep_msg *)work->mad_recv_wc->recv_buf.mad; cm_id_priv = cm_acquire_id(rep_msg->remote_comm_id, 0); @@ -2018,16 +2078,26 @@ static int cm_rep_handler(struct cm_work *work) goto error; } /* Check for a stale connection. */ - if (cm_insert_remote_qpn(cm_id_priv->timewait_info)) { + timewait_info = cm_insert_remote_qpn(cm_id_priv->timewait_info); + if (timewait_info) { rb_erase(&cm_id_priv->timewait_info->remote_id_node, &cm.remote_id_table); cm_id_priv->timewait_info->inserted_remote_id = 0; + cur_cm_id_priv = cm_get_id(timewait_info->work.local_id, + timewait_info->work.remote_id); + spin_unlock(&cm.lock); spin_unlock_irq(&cm_id_priv->lock); cm_issue_rej(work->port, work->mad_recv_wc, IB_CM_REJ_STALE_CONN, CM_MSG_RESPONSE_REP, NULL, 0); ret = -EINVAL; + if (cur_cm_id_priv) { + cm_id = &cur_cm_id_priv->id; + ib_send_cm_dreq(cm_id, NULL, 0); + cm_deref_id(cur_cm_id_priv); + } + goto error; } spin_unlock(&cm.lock); @@ -3339,6 +3409,8 @@ static void cm_process_send_error(struct ib_mad_send_buf *msg, if (msg != cm_id_priv->msg || state != cm_id_priv->id.state) goto discard; + pr_debug_ratelimited("CM: failed sending MAD in state %d. (%s)\n", + state, ib_wc_status_msg(wc_status)); switch (state) { case IB_CM_REQ_SENT: case IB_CM_MRA_REQ_RCVD: diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index 2a6fc47a1dfb..acd10d666f1c 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -101,6 +101,49 @@ const char *__attribute_const__ rdma_event_msg(enum rdma_cm_event_type event) } EXPORT_SYMBOL(rdma_event_msg); +const char *__attribute_const__ rdma_reject_msg(struct rdma_cm_id *id, + int reason) +{ + if (rdma_ib_or_roce(id->device, id->port_num)) + return ibcm_reject_msg(reason); + + if (rdma_protocol_iwarp(id->device, id->port_num)) + return iwcm_reject_msg(reason); + + WARN_ON_ONCE(1); + return "unrecognized transport"; +} +EXPORT_SYMBOL(rdma_reject_msg); + +bool rdma_is_consumer_reject(struct rdma_cm_id *id, int reason) +{ + if (rdma_ib_or_roce(id->device, id->port_num)) + return reason == IB_CM_REJ_CONSUMER_DEFINED; + + if (rdma_protocol_iwarp(id->device, id->port_num)) + return reason == -ECONNREFUSED; + + WARN_ON_ONCE(1); + return false; +} +EXPORT_SYMBOL(rdma_is_consumer_reject); + +const void *rdma_consumer_reject_data(struct rdma_cm_id *id, + struct rdma_cm_event *ev, u8 *data_len) +{ + const void *p; + + if (rdma_is_consumer_reject(id, ev->status)) { + *data_len = ev->param.conn.private_data_len; + p = ev->param.conn.private_data; + } else { + *data_len = 0; + p = NULL; + } + return p; +} +EXPORT_SYMBOL(rdma_consumer_reject_data); + static void cma_add_one(struct ib_device *device); static void cma_remove_one(struct ib_device *device, void *client_data); @@ -116,7 +159,7 @@ static LIST_HEAD(dev_list); static LIST_HEAD(listen_any_list); static DEFINE_MUTEX(lock); static struct workqueue_struct *cma_wq; -static int cma_pernet_id; +static unsigned int cma_pernet_id; struct cma_pernet { struct idr tcp_ps; @@ -155,6 +198,7 @@ struct cma_device { atomic_t refcount; struct list_head id_list; enum ib_gid_type *default_gid_type; + u8 *default_roce_tos; }; struct rdma_bind_list { @@ -226,8 +270,7 @@ struct cma_device *cma_enum_devices_by_ibdev(cma_device_filter filter, int cma_get_default_gid_type(struct cma_device *cma_dev, unsigned int port) { - if (port < rdma_start_port(cma_dev->device) || - port > rdma_end_port(cma_dev->device)) + if (!rdma_is_port_valid(cma_dev->device, port)) return -EINVAL; return cma_dev->default_gid_type[port - rdma_start_port(cma_dev->device)]; @@ -239,8 +282,7 @@ int cma_set_default_gid_type(struct cma_device *cma_dev, { unsigned long supported_gids; - if (port < rdma_start_port(cma_dev->device) || - port > rdma_end_port(cma_dev->device)) + if (!rdma_is_port_valid(cma_dev->device, port)) return -EINVAL; supported_gids = roce_gid_type_mask_support(cma_dev->device, port); @@ -254,6 +296,25 @@ int cma_set_default_gid_type(struct cma_device *cma_dev, return 0; } +int cma_get_default_roce_tos(struct cma_device *cma_dev, unsigned int port) +{ + if (!rdma_is_port_valid(cma_dev->device, port)) + return -EINVAL; + + return cma_dev->default_roce_tos[port - rdma_start_port(cma_dev->device)]; +} + +int cma_set_default_roce_tos(struct cma_device *cma_dev, unsigned int port, + u8 default_roce_tos) +{ + if (!rdma_is_port_valid(cma_dev->device, port)) + return -EINVAL; + + cma_dev->default_roce_tos[port - rdma_start_port(cma_dev->device)] = + default_roce_tos; + + return 0; +} struct ib_device *cma_get_ib_dev(struct cma_device *cma_dev) { return cma_dev->device; @@ -300,6 +361,7 @@ struct rdma_id_private { u32 options; u8 srq; u8 tos; + bool tos_set; u8 reuseaddr; u8 afonly; enum ib_gid_type gid_type; @@ -666,6 +728,7 @@ static int cma_resolve_ib_dev(struct rdma_id_private *id_priv) union ib_gid gid, sgid, *dgid; u16 pkey, index; u8 p; + enum ib_port_state port_state; int i; cma_dev = NULL; @@ -681,6 +744,8 @@ static int cma_resolve_ib_dev(struct rdma_id_private *id_priv) if (ib_find_cached_pkey(cur_dev->device, p, pkey, &index)) continue; + if (ib_get_cached_port_state(cur_dev->device, p, &port_state)) + continue; for (i = 0; !ib_get_cached_gid(cur_dev->device, p, i, &gid, NULL); i++) { @@ -692,7 +757,8 @@ static int cma_resolve_ib_dev(struct rdma_id_private *id_priv) } if (!cma_dev && (gid.global.subnet_prefix == - dgid->global.subnet_prefix)) { + dgid->global.subnet_prefix) && + port_state == IB_PORT_ACTIVE) { cma_dev = cur_dev; sgid = gid; id_priv->id.port_num = p; @@ -735,6 +801,7 @@ struct rdma_cm_id *rdma_create_id(struct net *net, id_priv->id.event_handler = event_handler; id_priv->id.ps = ps; id_priv->id.qp_type = qp_type; + id_priv->tos_set = false; spin_lock_init(&id_priv->lock); mutex_init(&id_priv->qp_mutex); init_completion(&id_priv->comp); @@ -1646,6 +1713,7 @@ static int cma_rep_recv(struct rdma_id_private *id_priv) return 0; reject: + pr_debug_ratelimited("RDMA CM: CONNECT_ERROR: failed to handle reply. status %d\n", ret); cma_modify_qp_err(id_priv); ib_send_cm_rej(id_priv->cm_id.ib, IB_CM_REJ_CONSUMER_DEFINED, NULL, 0, NULL, 0); @@ -1717,6 +1785,8 @@ static int cma_ib_handler(struct ib_cm_id *cm_id, struct ib_cm_event *ib_event) /* ignore event */ goto out; case IB_CM_REJ_RECEIVED: + pr_debug_ratelimited("RDMA CM: REJECTED: %s\n", rdma_reject_msg(&id_priv->id, + ib_event->param.rej_rcvd.reason)); cma_modify_qp_err(id_priv); event.status = ib_event->param.rej_rcvd.reason; event.event = RDMA_CM_EVENT_REJECTED; @@ -2223,6 +2293,7 @@ void rdma_set_service_type(struct rdma_cm_id *id, int tos) id_priv = container_of(id, struct rdma_id_private, id); id_priv->tos = (u8) tos; + id_priv->tos_set = true; } EXPORT_SYMBOL(rdma_set_service_type); @@ -2242,6 +2313,8 @@ static void cma_query_handler(int status, struct ib_sa_path_rec *path_rec, work->new_state = RDMA_CM_ADDR_RESOLVED; work->event.event = RDMA_CM_EVENT_ROUTE_ERROR; work->event.status = status; + pr_debug_ratelimited("RDMA CM: ROUTE_ERROR: failed to query path. status %d\n", + status); } queue_work(cma_wq, &work->work); @@ -2424,14 +2497,12 @@ static int iboe_tos_to_sl(struct net_device *ndev, int tos) struct net_device *dev; prio = rt_tos2priority(tos); - dev = ndev->priv_flags & IFF_802_1Q_VLAN ? - vlan_dev_real_dev(ndev) : ndev; - + dev = is_vlan_dev(ndev) ? vlan_dev_real_dev(ndev) : ndev; if (dev->num_tc) return netdev_get_prio_tc_map(dev, prio); #if IS_ENABLED(CONFIG_VLAN_8021Q) - if (ndev->priv_flags & IFF_802_1Q_VLAN) + if (is_vlan_dev(ndev)) return (vlan_dev_get_egress_qos_mask(ndev, prio) & VLAN_PRIO_MASK) >> VLAN_PRIO_SHIFT; #endif @@ -2457,6 +2528,9 @@ static int cma_resolve_iboe_route(struct rdma_id_private *id_priv) struct cma_work *work; int ret; struct net_device *ndev = NULL; + u8 default_roce_tos = id_priv->cma_dev->default_roce_tos[id_priv->id.port_num - + rdma_start_port(id_priv->cma_dev->device)]; + u8 tos = id_priv->tos_set ? id_priv->tos : default_roce_tos; work = kzalloc(sizeof *work, GFP_KERNEL); @@ -2530,7 +2604,8 @@ static int cma_resolve_iboe_route(struct rdma_id_private *id_priv) route->path_rec->reversible = 1; route->path_rec->pkey = cpu_to_be16(0xffff); route->path_rec->mtu_selector = IB_SA_EQ; - route->path_rec->sl = iboe_tos_to_sl(ndev, id_priv->tos); + route->path_rec->sl = iboe_tos_to_sl(ndev, tos); + route->path_rec->traffic_class = tos; route->path_rec->mtu = iboe_get_mtu(ndev->mtu); route->path_rec->rate_selector = IB_SA_EQ; route->path_rec->rate = iboe_get_rate(ndev); @@ -2609,8 +2684,8 @@ static void cma_set_loopback(struct sockaddr *addr) static int cma_bind_loopback(struct rdma_id_private *id_priv) { struct cma_device *cma_dev, *cur_dev; - struct ib_port_attr port_attr; union ib_gid gid; + enum ib_port_state port_state; u16 pkey; int ret; u8 p; @@ -2626,8 +2701,8 @@ static int cma_bind_loopback(struct rdma_id_private *id_priv) cma_dev = cur_dev; for (p = 1; p <= cur_dev->device->phys_port_cnt; ++p) { - if (!ib_query_port(cur_dev->device, p, &port_attr) && - port_attr.state == IB_PORT_ACTIVE) { + if (!ib_get_cached_port_state(cur_dev->device, p, &port_state) && + port_state == IB_PORT_ACTIVE) { cma_dev = cur_dev; goto port_found; } @@ -2677,8 +2752,14 @@ static void addr_handler(int status, struct sockaddr *src_addr, goto out; memcpy(cma_src_addr(id_priv), src_addr, rdma_addr_size(src_addr)); - if (!status && !id_priv->cma_dev) + if (!status && !id_priv->cma_dev) { status = cma_acquire_dev(id_priv, NULL); + if (status) + pr_debug_ratelimited("RDMA CM: ADDR_ERROR: failed to acquire device. status %d\n", + status); + } else { + pr_debug_ratelimited("RDMA CM: ADDR_ERROR: failed to resolve IP. status %d\n", status); + } if (status) { if (!cma_comp_exch(id_priv, RDMA_CM_ADDR_RESOLVED, @@ -2768,7 +2849,8 @@ static int cma_bind_addr(struct rdma_cm_id *id, struct sockaddr *src_addr, if (!src_addr || !src_addr->sa_family) { src_addr = (struct sockaddr *) &id->route.addr.src_addr; src_addr->sa_family = dst_addr->sa_family; - if (dst_addr->sa_family == AF_INET6) { + if (IS_ENABLED(CONFIG_IPV6) && + dst_addr->sa_family == AF_INET6) { struct sockaddr_in6 *src_addr6 = (struct sockaddr_in6 *) src_addr; struct sockaddr_in6 *dst_addr6 = (struct sockaddr_in6 *) dst_addr; src_addr6->sin6_scope_id = dst_addr6->sin6_scope_id; @@ -2789,20 +2871,26 @@ int rdma_resolve_addr(struct rdma_cm_id *id, struct sockaddr *src_addr, int ret; id_priv = container_of(id, struct rdma_id_private, id); + memcpy(cma_dst_addr(id_priv), dst_addr, rdma_addr_size(dst_addr)); if (id_priv->state == RDMA_CM_IDLE) { ret = cma_bind_addr(id, src_addr, dst_addr); - if (ret) + if (ret) { + memset(cma_dst_addr(id_priv), 0, rdma_addr_size(dst_addr)); return ret; + } } - if (cma_family(id_priv) != dst_addr->sa_family) + if (cma_family(id_priv) != dst_addr->sa_family) { + memset(cma_dst_addr(id_priv), 0, rdma_addr_size(dst_addr)); return -EINVAL; + } - if (!cma_comp_exch(id_priv, RDMA_CM_ADDR_BOUND, RDMA_CM_ADDR_QUERY)) + if (!cma_comp_exch(id_priv, RDMA_CM_ADDR_BOUND, RDMA_CM_ADDR_QUERY)) { + memset(cma_dst_addr(id_priv), 0, rdma_addr_size(dst_addr)); return -EINVAL; + } atomic_inc(&id_priv->refcount); - memcpy(cma_dst_addr(id_priv), dst_addr, rdma_addr_size(dst_addr)); if (cma_any_addr(dst_addr)) { ret = cma_resolve_loopback(id_priv); } else { @@ -2918,6 +3006,43 @@ err: return ret == -ENOSPC ? -EADDRNOTAVAIL : ret; } +static int cma_port_is_unique(struct rdma_bind_list *bind_list, + struct rdma_id_private *id_priv) +{ + struct rdma_id_private *cur_id; + struct sockaddr *daddr = cma_dst_addr(id_priv); + struct sockaddr *saddr = cma_src_addr(id_priv); + __be16 dport = cma_port(daddr); + + hlist_for_each_entry(cur_id, &bind_list->owners, node) { + struct sockaddr *cur_daddr = cma_dst_addr(cur_id); + struct sockaddr *cur_saddr = cma_src_addr(cur_id); + __be16 cur_dport = cma_port(cur_daddr); + + if (id_priv == cur_id) + continue; + + /* different dest port -> unique */ + if (!cma_any_port(cur_daddr) && + (dport != cur_dport)) + continue; + + /* different src address -> unique */ + if (!cma_any_addr(saddr) && + !cma_any_addr(cur_saddr) && + cma_addr_cmp(saddr, cur_saddr)) + continue; + + /* different dst address -> unique */ + if (!cma_any_addr(cur_daddr) && + cma_addr_cmp(daddr, cur_daddr)) + continue; + + return -EADDRNOTAVAIL; + } + return 0; +} + static int cma_alloc_any_port(enum rdma_port_space ps, struct rdma_id_private *id_priv) { @@ -2930,9 +3055,19 @@ static int cma_alloc_any_port(enum rdma_port_space ps, remaining = (high - low) + 1; rover = prandom_u32() % remaining + low; retry: - if (last_used_port != rover && - !cma_ps_find(net, ps, (unsigned short)rover)) { - int ret = cma_alloc_port(ps, id_priv, rover); + if (last_used_port != rover) { + struct rdma_bind_list *bind_list; + int ret; + + bind_list = cma_ps_find(net, ps, (unsigned short)rover); + + if (!bind_list) { + ret = cma_alloc_port(ps, id_priv, rover); + } else { + ret = cma_port_is_unique(bind_list, id_priv); + if (!ret) + cma_bind_port(bind_list, id_priv); + } /* * Remember previously used port number in order to avoid * re-using same port immediately after it is closed. @@ -3161,6 +3296,7 @@ int rdma_bind_addr(struct rdma_cm_id *id, struct sockaddr *addr) { struct rdma_id_private *id_priv; int ret; + struct sockaddr *daddr; if (addr->sa_family != AF_INET && addr->sa_family != AF_INET6 && addr->sa_family != AF_IB) @@ -3200,6 +3336,9 @@ int rdma_bind_addr(struct rdma_cm_id *id, struct sockaddr *addr) if (ret) goto err2; + daddr = cma_dst_addr(id_priv); + daddr->sa_family = addr->sa_family; + return 0; err2: if (id_priv->cma_dev) @@ -3264,10 +3403,13 @@ static int cma_sidr_rep_handler(struct ib_cm_id *cm_id, if (rep->status != IB_SIDR_SUCCESS) { event.event = RDMA_CM_EVENT_UNREACHABLE; event.status = ib_event->param.sidr_rep_rcvd.status; + pr_debug_ratelimited("RDMA CM: UNREACHABLE: bad SIDR reply. status %d\n", + event.status); break; } ret = cma_set_qkey(id_priv, rep->qkey); if (ret) { + pr_debug_ratelimited("RDMA CM: ADDR_ERROR: failed to set qkey. status %d\n", ret); event.event = RDMA_CM_EVENT_ADDR_ERROR; event.status = ret; break; @@ -3539,6 +3681,9 @@ static int cma_accept_iw(struct rdma_id_private *id_priv, struct iw_cm_conn_param iw_param; int ret; + if (!conn_param) + return -EINVAL; + ret = cma_modify_qp_rtr(id_priv, conn_param); if (ret) return ret; @@ -3716,10 +3861,17 @@ static int cma_ib_mc_handler(int status, struct ib_sa_multicast *multicast) if (!status) status = cma_set_qkey(id_priv, be32_to_cpu(multicast->rec.qkey)); + else + pr_debug_ratelimited("RDMA CM: MULTICAST_ERROR: failed to join multicast. status %d\n", + status); mutex_lock(&id_priv->qp_mutex); - if (!status && id_priv->id.qp) + if (!status && id_priv->id.qp) { status = ib_attach_mcast(id_priv->id.qp, &multicast->rec.mgid, be16_to_cpu(multicast->rec.mlid)); + if (status) + pr_debug_ratelimited("RDMA CM: MULTICAST_ERROR: failed to attach QP. status %d\n", + status); + } mutex_unlock(&id_priv->qp_mutex); memset(&event, 0, sizeof event); @@ -4185,15 +4337,21 @@ static void cma_add_one(struct ib_device *device) cma_dev->default_gid_type = kcalloc(device->phys_port_cnt, sizeof(*cma_dev->default_gid_type), GFP_KERNEL); - if (!cma_dev->default_gid_type) { - kfree(cma_dev); - return; - } + if (!cma_dev->default_gid_type) + goto free_cma_dev; + + cma_dev->default_roce_tos = kcalloc(device->phys_port_cnt, + sizeof(*cma_dev->default_roce_tos), + GFP_KERNEL); + if (!cma_dev->default_roce_tos) + goto free_gid_type; + for (i = rdma_start_port(device); i <= rdma_end_port(device); i++) { supported_gids = roce_gid_type_mask_support(device, i); WARN_ON(!supported_gids); cma_dev->default_gid_type[i - rdma_start_port(device)] = find_first_bit(&supported_gids, BITS_PER_LONG); + cma_dev->default_roce_tos[i - rdma_start_port(device)] = 0; } init_completion(&cma_dev->comp); @@ -4206,6 +4364,16 @@ static void cma_add_one(struct ib_device *device) list_for_each_entry(id_priv, &listen_any_list, list) cma_listen_on_dev(id_priv, cma_dev); mutex_unlock(&lock); + + return; + +free_gid_type: + kfree(cma_dev->default_gid_type); + +free_cma_dev: + kfree(cma_dev); + + return; } static int cma_remove_id_dev(struct rdma_id_private *id_priv) @@ -4274,6 +4442,7 @@ static void cma_remove_one(struct ib_device *device, void *client_data) mutex_unlock(&lock); cma_process_remove(cma_dev); + kfree(cma_dev->default_roce_tos); kfree(cma_dev->default_gid_type); kfree(cma_dev); } diff --git a/drivers/infiniband/core/cma_configfs.c b/drivers/infiniband/core/cma_configfs.c index 41573df1d9fc..54076a3e8007 100644 --- a/drivers/infiniband/core/cma_configfs.c +++ b/drivers/infiniband/core/cma_configfs.c @@ -139,8 +139,50 @@ static ssize_t default_roce_mode_store(struct config_item *item, CONFIGFS_ATTR(, default_roce_mode); +static ssize_t default_roce_tos_show(struct config_item *item, char *buf) +{ + struct cma_device *cma_dev; + struct cma_dev_port_group *group; + ssize_t ret; + u8 tos; + + ret = cma_configfs_params_get(item, &cma_dev, &group); + if (ret) + return ret; + + tos = cma_get_default_roce_tos(cma_dev, group->port_num); + cma_configfs_params_put(cma_dev); + + return sprintf(buf, "%u\n", tos); +} + +static ssize_t default_roce_tos_store(struct config_item *item, + const char *buf, size_t count) +{ + struct cma_device *cma_dev; + struct cma_dev_port_group *group; + ssize_t ret; + u8 tos; + + ret = kstrtou8(buf, 0, &tos); + if (ret) + return ret; + + ret = cma_configfs_params_get(item, &cma_dev, &group); + if (ret) + return ret; + + ret = cma_set_default_roce_tos(cma_dev, group->port_num, tos); + cma_configfs_params_put(cma_dev); + + return ret ? ret : strnlen(buf, count); +} + +CONFIGFS_ATTR(, default_roce_tos); + static struct configfs_attribute *cma_configfs_attributes[] = { &attr_default_roce_mode, + &attr_default_roce_tos, NULL, }; diff --git a/drivers/infiniband/core/core_priv.h b/drivers/infiniband/core/core_priv.h index 19d499dcab76..cb7d372e4bdf 100644 --- a/drivers/infiniband/core/core_priv.h +++ b/drivers/infiniband/core/core_priv.h @@ -35,6 +35,7 @@ #include <linux/list.h> #include <linux/spinlock.h> +#include <linux/cgroup_rdma.h> #include <rdma/ib_verbs.h> @@ -62,6 +63,9 @@ int cma_get_default_gid_type(struct cma_device *cma_dev, int cma_set_default_gid_type(struct cma_device *cma_dev, unsigned int port, enum ib_gid_type default_gid_type); +int cma_get_default_roce_tos(struct cma_device *cma_dev, unsigned int port); +int cma_set_default_roce_tos(struct cma_device *a_dev, unsigned int port, + u8 default_roce_tos); struct ib_device *cma_get_ib_dev(struct cma_device *cma_dev); int ib_device_register_sysfs(struct ib_device *device, @@ -72,9 +76,6 @@ void ib_device_unregister_sysfs(struct ib_device *device); void ib_cache_setup(void); void ib_cache_cleanup(void); -int ib_resolve_eth_dmac(struct ib_qp *qp, - struct ib_qp_attr *qp_attr, int *qp_attr_mask); - typedef void (*roce_netdev_callback)(struct ib_device *device, u8 port, struct net_device *idev, void *cookie); @@ -124,17 +125,39 @@ int ib_cache_setup_one(struct ib_device *device); void ib_cache_cleanup_one(struct ib_device *device); void ib_cache_release_one(struct ib_device *device); +#ifdef CONFIG_CGROUP_RDMA +int ib_device_register_rdmacg(struct ib_device *device); +void ib_device_unregister_rdmacg(struct ib_device *device); + +int ib_rdmacg_try_charge(struct ib_rdmacg_object *cg_obj, + struct ib_device *device, + enum rdmacg_resource_type resource_index); + +void ib_rdmacg_uncharge(struct ib_rdmacg_object *cg_obj, + struct ib_device *device, + enum rdmacg_resource_type resource_index); +#else +static inline int ib_device_register_rdmacg(struct ib_device *device) +{ return 0; } + +static inline void ib_device_unregister_rdmacg(struct ib_device *device) +{ } + +static inline int ib_rdmacg_try_charge(struct ib_rdmacg_object *cg_obj, + struct ib_device *device, + enum rdmacg_resource_type resource_index) +{ return 0; } + +static inline void ib_rdmacg_uncharge(struct ib_rdmacg_object *cg_obj, + struct ib_device *device, + enum rdmacg_resource_type resource_index) +{ } +#endif + static inline bool rdma_is_upper_dev_rcu(struct net_device *dev, struct net_device *upper) { - struct net_device *_upper = NULL; - struct list_head *iter; - - netdev_for_each_all_upper_dev_rcu(dev, _upper, iter) - if (_upper == upper) - break; - - return _upper == upper; + return netdev_has_upper_dev_all_rcu(dev, upper); } int addr_init(void); diff --git a/drivers/infiniband/core/cq.c b/drivers/infiniband/core/cq.c index a754fc727de5..e95510117a6d 100644 --- a/drivers/infiniband/core/cq.c +++ b/drivers/infiniband/core/cq.c @@ -58,8 +58,8 @@ static int __ib_process_cq(struct ib_cq *cq, int budget) * %IB_POLL_DIRECT CQ. It does not offload CQ processing to a different * context and does not ask for completion interrupts from the HCA. * - * Note: for compatibility reasons -1 can be passed in %budget for unlimited - * polling. Do not use this feature in new code, it will be removed soon. + * Note: do not pass -1 as %budget unless it is guaranteed that the number + * of completions that will be processed is small. */ int ib_process_cq_direct(struct ib_cq *cq, int budget) { @@ -120,7 +120,7 @@ static void ib_cq_completion_workqueue(struct ib_cq *cq, void *private) * * This is the proper interface to allocate a CQ for in-kernel users. A * CQ allocated with this interface will automatically be polled from the - * specified context. The ULP needs must use wr->wr_cqe instead of wr->wr_id + * specified context. The ULP must use wr->wr_cqe instead of wr->wr_id * to use this CQ abstraction. */ struct ib_cq *ib_alloc_cq(struct ib_device *dev, void *private, diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c index 760ef603a468..593d2ce6ec7c 100644 --- a/drivers/infiniband/core/device.c +++ b/drivers/infiniband/core/device.c @@ -254,11 +254,8 @@ static int add_client_context(struct ib_device *device, struct ib_client *client unsigned long flags; context = kmalloc(sizeof *context, GFP_KERNEL); - if (!context) { - pr_warn("Couldn't allocate client context for %s/%s\n", - device->name, client->name); + if (!context) return -ENOMEM; - } context->client = client; context->data = NULL; @@ -336,6 +333,15 @@ int ib_register_device(struct ib_device *device, int ret; struct ib_client *client; struct ib_udata uhw = {.outlen = 0, .inlen = 0}; + struct device *parent = device->dev.parent; + + WARN_ON_ONCE(!parent); + if (!device->dev.dma_ops) + device->dev.dma_ops = parent->dma_ops; + if (!device->dev.dma_mask) + device->dev.dma_mask = parent->dma_mask; + if (!device->dev.coherent_dma_mask) + device->dev.coherent_dma_mask = parent->coherent_dma_mask; mutex_lock(&device_mutex); @@ -363,10 +369,18 @@ int ib_register_device(struct ib_device *device, goto out; } + ret = ib_device_register_rdmacg(device); + if (ret) { + pr_warn("Couldn't register device with rdma cgroup\n"); + ib_cache_cleanup_one(device); + goto out; + } + memset(&device->attrs, 0, sizeof(device->attrs)); ret = device->query_device(device, &device->attrs, &uhw); if (ret) { pr_warn("Couldn't query the device attributes\n"); + ib_device_unregister_rdmacg(device); ib_cache_cleanup_one(device); goto out; } @@ -375,6 +389,7 @@ int ib_register_device(struct ib_device *device, if (ret) { pr_warn("Couldn't register device %s with driver model\n", device->name); + ib_device_unregister_rdmacg(device); ib_cache_cleanup_one(device); goto out; } @@ -424,6 +439,7 @@ void ib_unregister_device(struct ib_device *device) mutex_unlock(&device_mutex); + ib_device_unregister_rdmacg(device); ib_device_unregister_sysfs(device); ib_cache_cleanup_one(device); @@ -662,7 +678,7 @@ int ib_query_port(struct ib_device *device, union ib_gid gid; int err; - if (port_num < rdma_start_port(device) || port_num > rdma_end_port(device)) + if (!rdma_is_port_valid(device, port_num)) return -EINVAL; memset(port_attr, 0, sizeof(*port_attr)); @@ -828,7 +844,7 @@ int ib_modify_port(struct ib_device *device, if (!device->modify_port) return -ENOSYS; - if (port_num < rdma_start_port(device) || port_num > rdma_end_port(device)) + if (!rdma_is_port_valid(device, port_num)) return -EINVAL; return device->modify_port(device, port_num, port_modify_mask, diff --git a/drivers/infiniband/core/fmr_pool.c b/drivers/infiniband/core/fmr_pool.c index cdbb1f1a6d97..cdfad5f26212 100644 --- a/drivers/infiniband/core/fmr_pool.c +++ b/drivers/infiniband/core/fmr_pool.c @@ -247,7 +247,6 @@ struct ib_fmr_pool *ib_create_fmr_pool(struct ib_pd *pd, kmalloc(IB_FMR_HASH_SIZE * sizeof *pool->cache_bucket, GFP_KERNEL); if (!pool->cache_bucket) { - pr_warn(PFX "Failed to allocate cache in pool\n"); ret = -ENOMEM; goto out_free_pool; } diff --git a/drivers/infiniband/core/iwcm.c b/drivers/infiniband/core/iwcm.c index 5495e22839a7..31661b5c1743 100644 --- a/drivers/infiniband/core/iwcm.c +++ b/drivers/infiniband/core/iwcm.c @@ -59,6 +59,27 @@ MODULE_AUTHOR("Tom Tucker"); MODULE_DESCRIPTION("iWARP CM"); MODULE_LICENSE("Dual BSD/GPL"); +static const char * const iwcm_rej_reason_strs[] = { + [ECONNRESET] = "reset by remote host", + [ECONNREFUSED] = "refused by remote application", + [ETIMEDOUT] = "setup timeout", +}; + +const char *__attribute_const__ iwcm_reject_msg(int reason) +{ + size_t index; + + /* iWARP uses negative errnos */ + index = -reason; + + if (index < ARRAY_SIZE(iwcm_rej_reason_strs) && + iwcm_rej_reason_strs[index]) + return iwcm_rej_reason_strs[index]; + else + return "unrecognized reason"; +} +EXPORT_SYMBOL(iwcm_reject_msg); + static struct ibnl_client_cbs iwcm_nl_cb_table[] = { [RDMA_NL_IWPM_REG_PID] = {.dump = iwpm_register_pid_cb}, [RDMA_NL_IWPM_ADD_MAPPING] = {.dump = iwpm_add_mapping_cb}, diff --git a/drivers/infiniband/core/iwpm_msg.c b/drivers/infiniband/core/iwpm_msg.c index 1c41b95cefec..a0e7c16d8bd8 100644 --- a/drivers/infiniband/core/iwpm_msg.c +++ b/drivers/infiniband/core/iwpm_msg.c @@ -604,7 +604,6 @@ int iwpm_remote_info_cb(struct sk_buff *skb, struct netlink_callback *cb) } rem_info = kzalloc(sizeof(struct iwpm_remote_info), GFP_ATOMIC); if (!rem_info) { - pr_err("%s: Unable to allocate a remote info\n", __func__); ret = -ENOMEM; return ret; } diff --git a/drivers/infiniband/core/iwpm_util.c b/drivers/infiniband/core/iwpm_util.c index ade71e7f0131..3ef51a96bbf1 100644 --- a/drivers/infiniband/core/iwpm_util.c +++ b/drivers/infiniband/core/iwpm_util.c @@ -62,7 +62,6 @@ int iwpm_init(u8 nl_client) sizeof(struct hlist_head), GFP_KERNEL); if (!iwpm_hash_bucket) { ret = -ENOMEM; - pr_err("%s Unable to create mapinfo hash table\n", __func__); goto init_exit; } iwpm_reminfo_bucket = kzalloc(IWPM_REMINFO_HASH_SIZE * @@ -70,7 +69,6 @@ int iwpm_init(u8 nl_client) if (!iwpm_reminfo_bucket) { kfree(iwpm_hash_bucket); ret = -ENOMEM; - pr_err("%s Unable to create reminfo hash table\n", __func__); goto init_exit; } } @@ -128,10 +126,9 @@ int iwpm_create_mapinfo(struct sockaddr_storage *local_sockaddr, if (!iwpm_valid_client(nl_client)) return ret; map_info = kzalloc(sizeof(struct iwpm_mapping_info), GFP_KERNEL); - if (!map_info) { - pr_err("%s: Unable to allocate a mapping info\n", __func__); + if (!map_info) return -ENOMEM; - } + memcpy(&map_info->local_sockaddr, local_sockaddr, sizeof(struct sockaddr_storage)); memcpy(&map_info->mapped_sockaddr, mapped_sockaddr, @@ -309,10 +306,9 @@ struct iwpm_nlmsg_request *iwpm_get_nlmsg_request(__u32 nlmsg_seq, unsigned long flags; nlmsg_request = kzalloc(sizeof(struct iwpm_nlmsg_request), gfp); - if (!nlmsg_request) { - pr_err("%s Unable to allocate a nlmsg_request\n", __func__); + if (!nlmsg_request) return NULL; - } + spin_lock_irqsave(&iwpm_nlmsg_req_lock, flags); list_add_tail(&nlmsg_request->inprocess_list, &iwpm_nlmsg_req_list); spin_unlock_irqrestore(&iwpm_nlmsg_req_lock, flags); diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c index 40cbd6bdb73b..57f231f1c721 100644 --- a/drivers/infiniband/core/mad.c +++ b/drivers/infiniband/core/mad.c @@ -316,7 +316,9 @@ struct ib_mad_agent *ib_register_mad_agent(struct ib_device *device, /* Validate device and port */ port_priv = ib_get_mad_port(device, port_num); if (!port_priv) { - dev_notice(&device->dev, "ib_register_mad_agent: Invalid port\n"); + dev_notice(&device->dev, + "ib_register_mad_agent: Invalid port %d\n", + port_num); ret = ERR_PTR(-ENODEV); goto error1; } @@ -769,7 +771,7 @@ static int handle_outgoing_dr_smp(struct ib_mad_agent_private *mad_agent_priv, * If we are at the start of the LID routed part, don't update the * hop_ptr or hop_cnt. See section 14.2.2, Vol 1 IB spec. */ - if (opa && smp->class_version == OPA_SMP_CLASS_VERSION) { + if (opa && smp->class_version == OPA_SM_CLASS_VERSION) { u32 opa_drslid; if ((opa_get_smp_direction(opa_smp) @@ -816,7 +818,6 @@ static int handle_outgoing_dr_smp(struct ib_mad_agent_private *mad_agent_priv, local = kmalloc(sizeof *local, GFP_ATOMIC); if (!local) { ret = -ENOMEM; - dev_err(&device->dev, "No memory for ib_mad_local_private\n"); goto out; } local->mad_priv = NULL; @@ -824,7 +825,6 @@ static int handle_outgoing_dr_smp(struct ib_mad_agent_private *mad_agent_priv, mad_priv = alloc_mad_private(mad_size, GFP_ATOMIC); if (!mad_priv) { ret = -ENOMEM; - dev_err(&device->dev, "No memory for local response MAD\n"); kfree(local); goto out; } @@ -947,9 +947,6 @@ static int alloc_send_rmpp_list(struct ib_mad_send_wr_private *send_wr, for (left = send_buf->data_len + pad; left > 0; left -= seg_size) { seg = kmalloc(sizeof (*seg) + seg_size, gfp_mask); if (!seg) { - dev_err(&send_buf->mad_agent->device->dev, - "alloc_send_rmpp_segs: RMPP mem alloc failed for len %zd, gfp %#x\n", - sizeof (*seg) + seg_size, gfp_mask); free_send_rmpp_list(send_wr); return -ENOMEM; } @@ -1362,12 +1359,7 @@ static int allocate_method_table(struct ib_mad_mgmt_method_table **method) { /* Allocate management method table */ *method = kzalloc(sizeof **method, GFP_ATOMIC); - if (!*method) { - pr_err("No memory for ib_mad_mgmt_method_table\n"); - return -ENOMEM; - } - - return 0; + return (*method) ? 0 : (-ENOMEM); } /* @@ -1458,8 +1450,6 @@ static int add_nonoui_reg_req(struct ib_mad_reg_req *mad_reg_req, /* Allocate management class table for "new" class version */ *class = kzalloc(sizeof **class, GFP_ATOMIC); if (!*class) { - dev_err(&agent_priv->agent.device->dev, - "No memory for ib_mad_mgmt_class_table\n"); ret = -ENOMEM; goto error1; } @@ -1524,22 +1514,16 @@ static int add_oui_reg_req(struct ib_mad_reg_req *mad_reg_req, if (!*vendor_table) { /* Allocate mgmt vendor class table for "new" class version */ vendor = kzalloc(sizeof *vendor, GFP_ATOMIC); - if (!vendor) { - dev_err(&agent_priv->agent.device->dev, - "No memory for ib_mad_mgmt_vendor_class_table\n"); + if (!vendor) goto error1; - } *vendor_table = vendor; } if (!(*vendor_table)->vendor_class[vclass]) { /* Allocate table for this management vendor class */ vendor_class = kzalloc(sizeof *vendor_class, GFP_ATOMIC); - if (!vendor_class) { - dev_err(&agent_priv->agent.device->dev, - "No memory for ib_mad_mgmt_vendor_class\n"); + if (!vendor_class) goto error2; - } (*vendor_table)->vendor_class[vclass] = vendor_class; } @@ -1746,7 +1730,7 @@ find_mad_agent(struct ib_mad_port_private *port_priv, if (!class) goto out; if (convert_mgmt_class(mad_hdr->mgmt_class) >= - IB_MGMT_MAX_METHODS) + ARRAY_SIZE(class->method_table)) goto out; method = class->method_table[convert_mgmt_class( mad_hdr->mgmt_class)]; @@ -2167,7 +2151,7 @@ handle_smi(struct ib_mad_port_private *port_priv, struct ib_mad_hdr *mad_hdr = (struct ib_mad_hdr *)recv->mad; if (opa && mad_hdr->base_version == OPA_MGMT_BASE_VERSION && - mad_hdr->class_version == OPA_SMI_CLASS_VERSION) + mad_hdr->class_version == OPA_SM_CLASS_VERSION) return handle_opa_smi(port_priv, qp_info, wc, port_num, recv, response); @@ -2238,11 +2222,8 @@ static void ib_mad_recv_done(struct ib_cq *cq, struct ib_wc *wc) mad_size = recv->mad_size; response = alloc_mad_private(mad_size, GFP_KERNEL); - if (!response) { - dev_err(&port_priv->device->dev, - "%s: no memory for response buffer\n", __func__); + if (!response) goto out; - } if (rdma_cap_ib_switch(port_priv->device)) port_num = wc->port_num; @@ -2869,8 +2850,6 @@ static int ib_mad_post_receive_mads(struct ib_mad_qp_info *qp_info, mad_priv = alloc_mad_private(port_mad_size(qp_info->port_priv), GFP_ATOMIC); if (!mad_priv) { - dev_err(&qp_info->port_priv->device->dev, - "No memory for receive buffer\n"); ret = -ENOMEM; break; } @@ -2961,11 +2940,8 @@ static int ib_mad_port_start(struct ib_mad_port_private *port_priv) u16 pkey_index; attr = kmalloc(sizeof *attr, GFP_KERNEL); - if (!attr) { - dev_err(&port_priv->device->dev, - "Couldn't kmalloc ib_qp_attr\n"); + if (!attr) return -ENOMEM; - } ret = ib_find_pkey(port_priv->device, port_priv->port_num, IB_DEFAULT_PKEY_FULL, &pkey_index); @@ -3135,10 +3111,8 @@ static int ib_mad_port_open(struct ib_device *device, /* Create new device info */ port_priv = kzalloc(sizeof *port_priv, GFP_KERNEL); - if (!port_priv) { - dev_err(&device->dev, "No memory for ib_mad_port_private\n"); + if (!port_priv) return -ENOMEM; - } port_priv->device = device; port_priv->port_num = port_num; diff --git a/drivers/infiniband/core/multicast.c b/drivers/infiniband/core/multicast.c index e51b739f6ea3..322cb67b07a9 100644 --- a/drivers/infiniband/core/multicast.c +++ b/drivers/infiniband/core/multicast.c @@ -518,8 +518,11 @@ static void join_handler(int status, struct ib_sa_mcmember_rec *rec, process_join_error(group, status); else { int mgids_changed, is_mgid0; - ib_find_pkey(group->port->dev->device, group->port->port_num, - be16_to_cpu(rec->pkey), &pkey_index); + + if (ib_find_pkey(group->port->dev->device, + group->port->port_num, be16_to_cpu(rec->pkey), + &pkey_index)) + pkey_index = MCAST_INVALID_PKEY_INDEX; spin_lock_irq(&group->port->lock); if (group->state == MCAST_BUSY && diff --git a/drivers/infiniband/core/roce_gid_mgmt.c b/drivers/infiniband/core/roce_gid_mgmt.c index 06556c34606d..db958d3207ef 100644 --- a/drivers/infiniband/core/roce_gid_mgmt.c +++ b/drivers/infiniband/core/roce_gid_mgmt.c @@ -144,7 +144,6 @@ static enum bonding_slave_state is_eth_active_slave_of_bonding_rcu(struct net_de static int is_eth_port_of_netdev(struct ib_device *ib_dev, u8 port, struct net_device *rdma_ndev, void *cookie) { - struct net_device *event_ndev = (struct net_device *)cookie; struct net_device *real_dev; int res; @@ -152,11 +151,11 @@ static int is_eth_port_of_netdev(struct ib_device *ib_dev, u8 port, return 0; rcu_read_lock(); - real_dev = rdma_vlan_dev_real_dev(event_ndev); + real_dev = rdma_vlan_dev_real_dev(cookie); if (!real_dev) - real_dev = event_ndev; + real_dev = cookie; - res = ((rdma_is_upper_dev_rcu(rdma_ndev, event_ndev) && + res = ((rdma_is_upper_dev_rcu(rdma_ndev, cookie) && (is_eth_active_slave_of_bonding_rcu(rdma_ndev, real_dev) & REQUIRED_BOND_STATES)) || real_dev == rdma_ndev); @@ -192,17 +191,16 @@ static int pass_all_filter(struct ib_device *ib_dev, u8 port, static int upper_device_filter(struct ib_device *ib_dev, u8 port, struct net_device *rdma_ndev, void *cookie) { - struct net_device *event_ndev = (struct net_device *)cookie; int res; if (!rdma_ndev) return 0; - if (rdma_ndev == event_ndev) + if (rdma_ndev == cookie) return 1; rcu_read_lock(); - res = rdma_is_upper_dev_rcu(rdma_ndev, event_ndev); + res = rdma_is_upper_dev_rcu(rdma_ndev, cookie); rcu_read_unlock(); return res; @@ -304,10 +302,9 @@ static void enum_netdev_ipv4_ips(struct ib_device *ib_dev, for_ifa(in_dev) { struct sin_list *entry = kzalloc(sizeof(*entry), GFP_ATOMIC); - if (!entry) { - pr_warn("roce_gid_mgmt: couldn't allocate entry for IPv4 update\n"); + if (!entry) continue; - } + entry->ip.sin_family = AF_INET; entry->ip.sin_addr.s_addr = ifa->ifa_address; list_add_tail(&entry->list, &sin_list); @@ -348,10 +345,8 @@ static void enum_netdev_ipv6_ips(struct ib_device *ib_dev, list_for_each_entry(ifp, &in6_dev->addr_list, if_list) { struct sin6_list *entry = kzalloc(sizeof(*entry), GFP_ATOMIC); - if (!entry) { - pr_warn("roce_gid_mgmt: couldn't allocate entry for IPv6 update\n"); + if (!entry) continue; - } entry->sin6.sin6_family = AF_INET6; entry->sin6.sin6_addr = ifp->addr; @@ -382,18 +377,14 @@ static void _add_netdev_ips(struct ib_device *ib_dev, u8 port, static void add_netdev_ips(struct ib_device *ib_dev, u8 port, struct net_device *rdma_ndev, void *cookie) { - struct net_device *event_ndev = (struct net_device *)cookie; - - enum_netdev_default_gids(ib_dev, port, event_ndev, rdma_ndev); - _add_netdev_ips(ib_dev, port, event_ndev); + enum_netdev_default_gids(ib_dev, port, cookie, rdma_ndev); + _add_netdev_ips(ib_dev, port, cookie); } static void del_netdev_ips(struct ib_device *ib_dev, u8 port, struct net_device *rdma_ndev, void *cookie) { - struct net_device *event_ndev = (struct net_device *)cookie; - - ib_cache_gid_del_all_netdev_gids(ib_dev, port, event_ndev); + ib_cache_gid_del_all_netdev_gids(ib_dev, port, cookie); } static void enum_all_gids_of_dev_cb(struct ib_device *ib_dev, @@ -437,37 +428,39 @@ static void callback_for_addr_gid_device_scan(struct ib_device *device, &parsed->gid_attr); } +struct upper_list { + struct list_head list; + struct net_device *upper; +}; + +static int netdev_upper_walk(struct net_device *upper, void *data) +{ + struct upper_list *entry = kmalloc(sizeof(*entry), GFP_ATOMIC); + struct list_head *upper_list = data; + + if (!entry) + return 0; + + list_add_tail(&entry->list, upper_list); + dev_hold(upper); + entry->upper = upper; + + return 0; +} + static void handle_netdev_upper(struct ib_device *ib_dev, u8 port, void *cookie, void (*handle_netdev)(struct ib_device *ib_dev, u8 port, struct net_device *ndev)) { - struct net_device *ndev = (struct net_device *)cookie; - struct upper_list { - struct list_head list; - struct net_device *upper; - }; - struct net_device *upper; - struct list_head *iter; + struct net_device *ndev = cookie; struct upper_list *upper_iter; struct upper_list *upper_temp; LIST_HEAD(upper_list); rcu_read_lock(); - netdev_for_each_all_upper_dev_rcu(ndev, upper, iter) { - struct upper_list *entry = kmalloc(sizeof(*entry), - GFP_ATOMIC); - - if (!entry) { - pr_info("roce_gid_mgmt: couldn't allocate entry to delete ndev\n"); - continue; - } - - list_add_tail(&entry->list, &upper_list); - dev_hold(upper); - entry->upper = upper; - } + netdev_walk_all_upper_dev_rcu(ndev, netdev_upper_walk, &upper_list); rcu_read_unlock(); handle_netdev(ib_dev, port, ndev); @@ -520,9 +513,7 @@ static void del_netdev_default_ips_join(struct ib_device *ib_dev, u8 port, static void del_netdev_default_ips(struct ib_device *ib_dev, u8 port, struct net_device *rdma_ndev, void *cookie) { - struct net_device *event_ndev = (struct net_device *)cookie; - - bond_delete_netdev_default_gids(ib_dev, port, event_ndev, rdma_ndev); + bond_delete_netdev_default_gids(ib_dev, port, cookie, rdma_ndev); } /* The following functions operate on all IB devices. netdevice_event and @@ -555,10 +546,8 @@ static int netdevice_queue_work(struct netdev_event_work_cmd *cmds, struct netdev_event_work *ndev_work = kmalloc(sizeof(*ndev_work), GFP_KERNEL); - if (!ndev_work) { - pr_warn("roce_gid_mgmt: can't allocate work for netdevice_event\n"); + if (!ndev_work) return NOTIFY_DONE; - } memcpy(ndev_work->cmds, cmds, sizeof(ndev_work->cmds)); for (i = 0; i < ARRAY_SIZE(ndev_work->cmds) && ndev_work->cmds[i].cb; i++) { @@ -692,10 +681,8 @@ static int addr_event(struct notifier_block *this, unsigned long event, } work = kmalloc(sizeof(*work), GFP_ATOMIC); - if (!work) { - pr_warn("roce_gid_mgmt: Couldn't allocate work for addr_event\n"); + if (!work) return NOTIFY_DONE; - } INIT_WORK(&work->work, update_gid_event_work_handler); diff --git a/drivers/infiniband/core/sysfs.c b/drivers/infiniband/core/sysfs.c index c1fb545e8d78..daadf3130c9f 100644 --- a/drivers/infiniband/core/sysfs.c +++ b/drivers/infiniband/core/sysfs.c @@ -1258,7 +1258,7 @@ int ib_device_register_sysfs(struct ib_device *device, int ret; int i; - device->dev.parent = device->dma_device; + WARN_ON_ONCE(!device->dev.parent); ret = dev_set_name(class_dev, "%s", device->name); if (ret) return ret; diff --git a/drivers/infiniband/core/ucm.c b/drivers/infiniband/core/ucm.c index 7713ef089c3c..cc0d51fb06e3 100644 --- a/drivers/infiniband/core/ucm.c +++ b/drivers/infiniband/core/ucm.c @@ -46,7 +46,7 @@ #include <linux/mutex.h> #include <linux/slab.h> -#include <asm/uaccess.h> +#include <linux/uaccess.h> #include <rdma/ib.h> #include <rdma/ib_cm.h> @@ -1104,8 +1104,11 @@ static ssize_t ib_ucm_write(struct file *filp, const char __user *buf, struct ib_ucm_cmd_hdr hdr; ssize_t result; - if (WARN_ON_ONCE(!ib_safe_file_access(filp))) + if (!ib_safe_file_access(filp)) { + pr_err_once("ucm_write: process %d (%s) changed security contexts after opening file descriptor, this is not allowed.\n", + task_tgid_vnr(current), current->comm); return -EACCES; + } if (len < sizeof(hdr)) return -EINVAL; @@ -1287,7 +1290,7 @@ static void ib_ucm_add_one(struct ib_device *device) goto err; ucm_dev->dev.class = &cm_class; - ucm_dev->dev.parent = device->dma_device; + ucm_dev->dev.parent = device->dev.parent; ucm_dev->dev.devt = ucm_dev->cdev.dev; ucm_dev->dev.release = ib_ucm_release_dev; dev_set_name(&ucm_dev->dev, "ucm%d", ucm_dev->devnum); diff --git a/drivers/infiniband/core/ucma.c b/drivers/infiniband/core/ucma.c index 9520154f1d7c..e12f8faf8c23 100644 --- a/drivers/infiniband/core/ucma.c +++ b/drivers/infiniband/core/ucma.c @@ -1584,8 +1584,11 @@ static ssize_t ucma_write(struct file *filp, const char __user *buf, struct rdma_ucm_cmd_hdr hdr; ssize_t ret; - if (WARN_ON_ONCE(!ib_safe_file_access(filp))) + if (!ib_safe_file_access(filp)) { + pr_err_once("ucma_write: process %d (%s) changed security contexts after opening file descriptor, this is not allowed.\n", + task_tgid_vnr(current), current->comm); return -EACCES; + } if (len < sizeof(hdr)) return -EINVAL; diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c index 84b4eff90395..446b56a5260b 100644 --- a/drivers/infiniband/core/umem.c +++ b/drivers/infiniband/core/umem.c @@ -51,7 +51,7 @@ static void __ib_umem_release(struct ib_device *dev, struct ib_umem *umem, int d if (umem->nmap > 0) ib_dma_unmap_sg(dev, umem->sg_head.sgl, - umem->nmap, + umem->npages, DMA_BIDIRECTIONAL); for_each_sg(umem->sg_head.sgl, sg, umem->npages, i) { @@ -99,9 +99,6 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, if (dmasync) dma_attrs |= DMA_ATTR_WRITE_BARRIER; - if (!size) - return ERR_PTR(-EINVAL); - /* * If the combination of the addr and size requested for this memory * region causes an integer overflow, return error. @@ -134,6 +131,7 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, IB_ACCESS_REMOTE_ATOMIC | IB_ACCESS_MW_BIND)); if (access & IB_ACCESS_ON_DEMAND) { + put_pid(umem->pid); ret = ib_umem_odp_get(context, umem); if (ret) { kfree(umem); @@ -149,6 +147,7 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, page_list = (struct page **) __get_free_page(GFP_KERNEL); if (!page_list) { + put_pid(umem->pid); kfree(umem); return ERR_PTR(-ENOMEM); } diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c index 1f0fe3217f23..f2fc0431512d 100644 --- a/drivers/infiniband/core/umem_odp.c +++ b/drivers/infiniband/core/umem_odp.c @@ -239,6 +239,71 @@ static const struct mmu_notifier_ops ib_umem_notifiers = { .invalidate_range_end = ib_umem_notifier_invalidate_range_end, }; +struct ib_umem *ib_alloc_odp_umem(struct ib_ucontext *context, + unsigned long addr, + size_t size) +{ + struct ib_umem *umem; + struct ib_umem_odp *odp_data; + int pages = size >> PAGE_SHIFT; + int ret; + + umem = kzalloc(sizeof(*umem), GFP_KERNEL); + if (!umem) + return ERR_PTR(-ENOMEM); + + umem->context = context; + umem->length = size; + umem->address = addr; + umem->page_size = PAGE_SIZE; + umem->writable = 1; + + odp_data = kzalloc(sizeof(*odp_data), GFP_KERNEL); + if (!odp_data) { + ret = -ENOMEM; + goto out_umem; + } + odp_data->umem = umem; + + mutex_init(&odp_data->umem_mutex); + init_completion(&odp_data->notifier_completion); + + odp_data->page_list = vzalloc(pages * sizeof(*odp_data->page_list)); + if (!odp_data->page_list) { + ret = -ENOMEM; + goto out_odp_data; + } + + odp_data->dma_list = vzalloc(pages * sizeof(*odp_data->dma_list)); + if (!odp_data->dma_list) { + ret = -ENOMEM; + goto out_page_list; + } + + down_write(&context->umem_rwsem); + context->odp_mrs_count++; + rbt_ib_umem_insert(&odp_data->interval_tree, &context->umem_tree); + if (likely(!atomic_read(&context->notifier_count))) + odp_data->mn_counters_active = true; + else + list_add(&odp_data->no_private_counters, + &context->no_private_counters); + up_write(&context->umem_rwsem); + + umem->odp_data = odp_data; + + return umem; + +out_page_list: + vfree(odp_data->page_list); +out_odp_data: + kfree(odp_data); +out_umem: + kfree(umem); + return ERR_PTR(ret); +} +EXPORT_SYMBOL(ib_alloc_odp_umem); + int ib_umem_odp_get(struct ib_ucontext *context, struct ib_umem *umem) { int ret_val; @@ -270,18 +335,20 @@ int ib_umem_odp_get(struct ib_ucontext *context, struct ib_umem *umem) init_completion(&umem->odp_data->notifier_completion); - umem->odp_data->page_list = vzalloc(ib_umem_num_pages(umem) * + if (ib_umem_num_pages(umem)) { + umem->odp_data->page_list = vzalloc(ib_umem_num_pages(umem) * sizeof(*umem->odp_data->page_list)); - if (!umem->odp_data->page_list) { - ret_val = -ENOMEM; - goto out_odp_data; - } + if (!umem->odp_data->page_list) { + ret_val = -ENOMEM; + goto out_odp_data; + } - umem->odp_data->dma_list = vzalloc(ib_umem_num_pages(umem) * + umem->odp_data->dma_list = vzalloc(ib_umem_num_pages(umem) * sizeof(*umem->odp_data->dma_list)); - if (!umem->odp_data->dma_list) { - ret_val = -ENOMEM; - goto out_page_list; + if (!umem->odp_data->dma_list) { + ret_val = -ENOMEM; + goto out_page_list; + } } /* @@ -466,6 +533,7 @@ static int ib_umem_odp_map_dma_single_page( } umem->odp_data->dma_list[page_index] = dma_addr | access_mask; umem->odp_data->page_list[page_index] = page; + umem->npages++; stored_page = 1; } else if (umem->odp_data->page_list[page_index] == page) { umem->odp_data->dma_list[page_index] |= access_mask; @@ -505,7 +573,8 @@ out: * for failure. * An -EAGAIN error code is returned when a concurrent mmu notifier prevents * the function from completing its task. - * + * An -ENOENT error code indicates that userspace process is being terminated + * and mm was already destroyed. * @umem: the umem to map and pin * @user_virt: the address from which we need to map. * @bcnt: the minimal number of bytes to pin and map. The mapping might be @@ -553,7 +622,7 @@ int ib_umem_odp_map_dma_pages(struct ib_umem *umem, u64 user_virt, u64 bcnt, owning_mm = get_task_mm(owning_process); if (owning_mm == NULL) { - ret = -EINVAL; + ret = -ENOENT; goto out_put_task; } @@ -578,7 +647,7 @@ int ib_umem_odp_map_dma_pages(struct ib_umem *umem, u64 user_virt, u64 bcnt, */ npages = get_user_pages_remote(owning_process, owning_mm, user_virt, gup_num_pages, - flags, local_page_list, NULL); + flags, local_page_list, NULL, NULL); up_read(&owning_mm->mmap_sem); if (npages < 0) @@ -665,6 +734,7 @@ void ib_umem_odp_unmap_dma_pages(struct ib_umem *umem, u64 virt, put_page(page); umem->odp_data->page_list[idx] = NULL; umem->odp_data->dma_list[idx] = 0; + umem->npages--; } } mutex_unlock(&umem->odp_data->umem_mutex); diff --git a/drivers/infiniband/core/umem_rbtree.c b/drivers/infiniband/core/umem_rbtree.c index 727d788448f5..d176597b4d78 100644 --- a/drivers/infiniband/core/umem_rbtree.c +++ b/drivers/infiniband/core/umem_rbtree.c @@ -78,17 +78,32 @@ int rbt_ib_umem_for_each_in_range(struct rb_root *root, void *cookie) { int ret_val = 0; - struct umem_odp_node *node; + struct umem_odp_node *node, *next; struct ib_umem_odp *umem; if (unlikely(start == last)) return ret_val; - for (node = rbt_ib_umem_iter_first(root, start, last - 1); node; - node = rbt_ib_umem_iter_next(node, start, last - 1)) { + for (node = rbt_ib_umem_iter_first(root, start, last - 1); + node; node = next) { + next = rbt_ib_umem_iter_next(node, start, last - 1); umem = container_of(node, struct ib_umem_odp, interval_tree); ret_val = cb(umem->umem, start, last, cookie) || ret_val; } return ret_val; } +EXPORT_SYMBOL(rbt_ib_umem_for_each_in_range); + +struct ib_umem_odp *rbt_ib_umem_lookup(struct rb_root *root, + u64 addr, u64 length) +{ + struct umem_odp_node *node; + + node = rbt_ib_umem_iter_first(root, addr, addr + length - 1); + if (node) + return container_of(node, struct ib_umem_odp, interval_tree); + return NULL; + +} +EXPORT_SYMBOL(rbt_ib_umem_lookup); diff --git a/drivers/infiniband/core/user_mad.c b/drivers/infiniband/core/user_mad.c index 415a3185cde7..aca7ff7abedc 100644 --- a/drivers/infiniband/core/user_mad.c +++ b/drivers/infiniband/core/user_mad.c @@ -50,7 +50,7 @@ #include <linux/semaphore.h> #include <linux/slab.h> -#include <asm/uaccess.h> +#include <linux/uaccess.h> #include <rdma/ib_mad.h> #include <rdma/ib_user_mad.h> @@ -1188,7 +1188,7 @@ static int ib_umad_init_port(struct ib_device *device, int port_num, if (cdev_add(&port->cdev, base, 1)) goto err_cdev; - port->dev = device_create(umad_class, device->dma_device, + port->dev = device_create(umad_class, device->dev.parent, port->cdev.dev, port, "umad%d", port->dev_num); if (IS_ERR(port->dev)) @@ -1207,7 +1207,7 @@ static int ib_umad_init_port(struct ib_device *device, int port_num, if (cdev_add(&port->sm_cdev, base, 1)) goto err_sm_cdev; - port->sm_dev = device_create(umad_class, device->dma_device, + port->sm_dev = device_create(umad_class, device->dev.parent, port->sm_cdev.dev, port, "issm%d", port->dev_num); if (IS_ERR(port->sm_dev)) diff --git a/drivers/infiniband/core/uverbs.h b/drivers/infiniband/core/uverbs.h index df26a741cda6..e1bedf0bac04 100644 --- a/drivers/infiniband/core/uverbs.h +++ b/drivers/infiniband/core/uverbs.h @@ -228,6 +228,7 @@ struct ib_uverbs_flow_spec { struct ib_uverbs_flow_spec_ipv4 ipv4; struct ib_uverbs_flow_spec_tcp_udp tcp_udp; struct ib_uverbs_flow_spec_ipv6 ipv6; + struct ib_uverbs_flow_spec_action_tag flow_tag; }; }; @@ -289,5 +290,6 @@ IB_UVERBS_DECLARE_EX_CMD(modify_wq); IB_UVERBS_DECLARE_EX_CMD(destroy_wq); IB_UVERBS_DECLARE_EX_CMD(create_rwq_ind_table); IB_UVERBS_DECLARE_EX_CMD(destroy_rwq_ind_table); +IB_UVERBS_DECLARE_EX_CMD(modify_qp); #endif /* UVERBS_H */ diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index cb3f515a2285..7b7a76e1279a 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -38,7 +38,7 @@ #include <linux/slab.h> #include <linux/sched.h> -#include <asm/uaccess.h> +#include <linux/uaccess.h> #include "uverbs.h" #include "core_priv.h" @@ -316,6 +316,7 @@ ssize_t ib_uverbs_get_context(struct ib_uverbs_file *file, struct ib_udata udata; struct ib_ucontext *ucontext; struct file *filp; + struct ib_rdmacg_object cg_obj; int ret; if (out_len < sizeof resp) @@ -335,13 +336,18 @@ ssize_t ib_uverbs_get_context(struct ib_uverbs_file *file, (unsigned long) cmd.response + sizeof resp, in_len - sizeof cmd, out_len - sizeof resp); + ret = ib_rdmacg_try_charge(&cg_obj, ib_dev, RDMACG_RESOURCE_HCA_HANDLE); + if (ret) + goto err; + ucontext = ib_dev->alloc_ucontext(ib_dev, &udata); if (IS_ERR(ucontext)) { ret = PTR_ERR(ucontext); - goto err; + goto err_alloc; } ucontext->device = ib_dev; + ucontext->cg_obj = cg_obj; INIT_LIST_HEAD(&ucontext->pd_list); INIT_LIST_HEAD(&ucontext->mr_list); INIT_LIST_HEAD(&ucontext->mw_list); @@ -407,6 +413,9 @@ err_free: put_pid(ucontext->tgid); ib_dev->dealloc_ucontext(ucontext); +err_alloc: + ib_rdmacg_uncharge(&cg_obj, ib_dev, RDMACG_RESOURCE_HCA_HANDLE); + err: mutex_unlock(&file->mutex); return ret; @@ -561,6 +570,13 @@ ssize_t ib_uverbs_alloc_pd(struct ib_uverbs_file *file, return -ENOMEM; init_uobj(uobj, 0, file->ucontext, &pd_lock_class); + ret = ib_rdmacg_try_charge(&uobj->cg_obj, ib_dev, + RDMACG_RESOURCE_HCA_OBJECT); + if (ret) { + kfree(uobj); + return ret; + } + down_write(&uobj->mutex); pd = ib_dev->alloc_pd(ib_dev, file->ucontext, &udata); @@ -605,6 +621,7 @@ err_idr: ib_dealloc_pd(pd); err: + ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); put_uobj_write(uobj); return ret; } @@ -637,6 +654,8 @@ ssize_t ib_uverbs_dealloc_pd(struct ib_uverbs_file *file, if (ret) goto err_put; + ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); + uobj->live = 0; put_uobj_write(uobj); @@ -1006,6 +1025,10 @@ ssize_t ib_uverbs_reg_mr(struct ib_uverbs_file *file, goto err_put; } } + ret = ib_rdmacg_try_charge(&uobj->cg_obj, ib_dev, + RDMACG_RESOURCE_HCA_OBJECT); + if (ret) + goto err_charge; mr = pd->device->reg_user_mr(pd, cmd.start, cmd.length, cmd.hca_va, cmd.access_flags, &udata); @@ -1054,6 +1077,9 @@ err_unreg: ib_dereg_mr(mr); err_put: + ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); + +err_charge: put_pd_read(pd); err_free: @@ -1178,6 +1204,8 @@ ssize_t ib_uverbs_dereg_mr(struct ib_uverbs_file *file, if (ret) return ret; + ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); + idr_remove_uobj(&ib_uverbs_mr_idr, uobj); mutex_lock(&file->mutex); @@ -1226,6 +1254,11 @@ ssize_t ib_uverbs_alloc_mw(struct ib_uverbs_file *file, in_len - sizeof(cmd) - sizeof(struct ib_uverbs_cmd_hdr), out_len - sizeof(resp)); + ret = ib_rdmacg_try_charge(&uobj->cg_obj, ib_dev, + RDMACG_RESOURCE_HCA_OBJECT); + if (ret) + goto err_charge; + mw = pd->device->alloc_mw(pd, cmd.mw_type, &udata); if (IS_ERR(mw)) { ret = PTR_ERR(mw); @@ -1271,6 +1304,9 @@ err_unalloc: uverbs_dealloc_mw(mw); err_put: + ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); + +err_charge: put_pd_read(pd); err_free: @@ -1306,6 +1342,8 @@ ssize_t ib_uverbs_dealloc_mw(struct ib_uverbs_file *file, if (ret) return ret; + ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); + idr_remove_uobj(&ib_uverbs_mw_idr, uobj); mutex_lock(&file->mutex); @@ -1405,6 +1443,11 @@ static struct ib_ucq_object *create_cq(struct ib_uverbs_file *file, if (cmd_sz > offsetof(typeof(*cmd), flags) + sizeof(cmd->flags)) attr.flags = cmd->flags; + ret = ib_rdmacg_try_charge(&obj->uobject.cg_obj, ib_dev, + RDMACG_RESOURCE_HCA_OBJECT); + if (ret) + goto err_charge; + cq = ib_dev->create_cq(ib_dev, &attr, file->ucontext, uhw); if (IS_ERR(cq)) { @@ -1452,6 +1495,10 @@ err_free: ib_destroy_cq(cq); err_file: + ib_rdmacg_uncharge(&obj->uobject.cg_obj, ib_dev, + RDMACG_RESOURCE_HCA_OBJECT); + +err_charge: if (ev_file) ib_uverbs_release_ucq(file, ev_file, obj); @@ -1732,6 +1779,8 @@ ssize_t ib_uverbs_destroy_cq(struct ib_uverbs_file *file, if (ret) return ret; + ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); + idr_remove_uobj(&ib_uverbs_cq_idr, uobj); mutex_lock(&file->mutex); @@ -1891,7 +1940,8 @@ static int create_qp(struct ib_uverbs_file *file, IB_QP_CREATE_CROSS_CHANNEL | IB_QP_CREATE_MANAGED_SEND | IB_QP_CREATE_MANAGED_RECV | - IB_QP_CREATE_SCATTER_FCS)) { + IB_QP_CREATE_SCATTER_FCS | + IB_QP_CREATE_CVLAN_STRIPPING)) { ret = -EINVAL; goto err_put; } @@ -1904,6 +1954,11 @@ static int create_qp(struct ib_uverbs_file *file, goto err_put; } + ret = ib_rdmacg_try_charge(&obj->uevent.uobject.cg_obj, device, + RDMACG_RESOURCE_HCA_OBJECT); + if (ret) + goto err_put; + if (cmd->qp_type == IB_QPT_XRC_TGT) qp = ib_create_qp(pd, &attr); else @@ -1911,7 +1966,7 @@ static int create_qp(struct ib_uverbs_file *file, if (IS_ERR(qp)) { ret = PTR_ERR(qp); - goto err_put; + goto err_create; } if (cmd->qp_type != IB_QPT_XRC_TGT) { @@ -1992,6 +2047,10 @@ err_cb: err_destroy: ib_destroy_qp(qp); +err_create: + ib_rdmacg_uncharge(&obj->uevent.uobject.cg_obj, device, + RDMACG_RESOURCE_HCA_OBJECT); + err_put: if (xrcd) put_xrcd_read(xrcd_uobj); @@ -2328,94 +2387,88 @@ static int modify_qp_mask(enum ib_qp_type qp_type, int mask) } } -ssize_t ib_uverbs_modify_qp(struct ib_uverbs_file *file, - struct ib_device *ib_dev, - const char __user *buf, int in_len, - int out_len) +static int modify_qp(struct ib_uverbs_file *file, + struct ib_uverbs_ex_modify_qp *cmd, struct ib_udata *udata) { - struct ib_uverbs_modify_qp cmd; - struct ib_udata udata; - struct ib_qp *qp; - struct ib_qp_attr *attr; - int ret; - - if (copy_from_user(&cmd, buf, sizeof cmd)) - return -EFAULT; - - INIT_UDATA(&udata, buf + sizeof cmd, NULL, in_len - sizeof cmd, - out_len); + struct ib_qp_attr *attr; + struct ib_qp *qp; + int ret; attr = kmalloc(sizeof *attr, GFP_KERNEL); if (!attr) return -ENOMEM; - qp = idr_read_qp(cmd.qp_handle, file->ucontext); + qp = idr_read_qp(cmd->base.qp_handle, file->ucontext); if (!qp) { ret = -EINVAL; goto out; } - attr->qp_state = cmd.qp_state; - attr->cur_qp_state = cmd.cur_qp_state; - attr->path_mtu = cmd.path_mtu; - attr->path_mig_state = cmd.path_mig_state; - attr->qkey = cmd.qkey; - attr->rq_psn = cmd.rq_psn; - attr->sq_psn = cmd.sq_psn; - attr->dest_qp_num = cmd.dest_qp_num; - attr->qp_access_flags = cmd.qp_access_flags; - attr->pkey_index = cmd.pkey_index; - attr->alt_pkey_index = cmd.alt_pkey_index; - attr->en_sqd_async_notify = cmd.en_sqd_async_notify; - attr->max_rd_atomic = cmd.max_rd_atomic; - attr->max_dest_rd_atomic = cmd.max_dest_rd_atomic; - attr->min_rnr_timer = cmd.min_rnr_timer; - attr->port_num = cmd.port_num; - attr->timeout = cmd.timeout; - attr->retry_cnt = cmd.retry_cnt; - attr->rnr_retry = cmd.rnr_retry; - attr->alt_port_num = cmd.alt_port_num; - attr->alt_timeout = cmd.alt_timeout; - - memcpy(attr->ah_attr.grh.dgid.raw, cmd.dest.dgid, 16); - attr->ah_attr.grh.flow_label = cmd.dest.flow_label; - attr->ah_attr.grh.sgid_index = cmd.dest.sgid_index; - attr->ah_attr.grh.hop_limit = cmd.dest.hop_limit; - attr->ah_attr.grh.traffic_class = cmd.dest.traffic_class; - attr->ah_attr.dlid = cmd.dest.dlid; - attr->ah_attr.sl = cmd.dest.sl; - attr->ah_attr.src_path_bits = cmd.dest.src_path_bits; - attr->ah_attr.static_rate = cmd.dest.static_rate; - attr->ah_attr.ah_flags = cmd.dest.is_global ? IB_AH_GRH : 0; - attr->ah_attr.port_num = cmd.dest.port_num; - - memcpy(attr->alt_ah_attr.grh.dgid.raw, cmd.alt_dest.dgid, 16); - attr->alt_ah_attr.grh.flow_label = cmd.alt_dest.flow_label; - attr->alt_ah_attr.grh.sgid_index = cmd.alt_dest.sgid_index; - attr->alt_ah_attr.grh.hop_limit = cmd.alt_dest.hop_limit; - attr->alt_ah_attr.grh.traffic_class = cmd.alt_dest.traffic_class; - attr->alt_ah_attr.dlid = cmd.alt_dest.dlid; - attr->alt_ah_attr.sl = cmd.alt_dest.sl; - attr->alt_ah_attr.src_path_bits = cmd.alt_dest.src_path_bits; - attr->alt_ah_attr.static_rate = cmd.alt_dest.static_rate; - attr->alt_ah_attr.ah_flags = cmd.alt_dest.is_global ? IB_AH_GRH : 0; - attr->alt_ah_attr.port_num = cmd.alt_dest.port_num; + attr->qp_state = cmd->base.qp_state; + attr->cur_qp_state = cmd->base.cur_qp_state; + attr->path_mtu = cmd->base.path_mtu; + attr->path_mig_state = cmd->base.path_mig_state; + attr->qkey = cmd->base.qkey; + attr->rq_psn = cmd->base.rq_psn; + attr->sq_psn = cmd->base.sq_psn; + attr->dest_qp_num = cmd->base.dest_qp_num; + attr->qp_access_flags = cmd->base.qp_access_flags; + attr->pkey_index = cmd->base.pkey_index; + attr->alt_pkey_index = cmd->base.alt_pkey_index; + attr->en_sqd_async_notify = cmd->base.en_sqd_async_notify; + attr->max_rd_atomic = cmd->base.max_rd_atomic; + attr->max_dest_rd_atomic = cmd->base.max_dest_rd_atomic; + attr->min_rnr_timer = cmd->base.min_rnr_timer; + attr->port_num = cmd->base.port_num; + attr->timeout = cmd->base.timeout; + attr->retry_cnt = cmd->base.retry_cnt; + attr->rnr_retry = cmd->base.rnr_retry; + attr->alt_port_num = cmd->base.alt_port_num; + attr->alt_timeout = cmd->base.alt_timeout; + attr->rate_limit = cmd->rate_limit; + + memcpy(attr->ah_attr.grh.dgid.raw, cmd->base.dest.dgid, 16); + attr->ah_attr.grh.flow_label = cmd->base.dest.flow_label; + attr->ah_attr.grh.sgid_index = cmd->base.dest.sgid_index; + attr->ah_attr.grh.hop_limit = cmd->base.dest.hop_limit; + attr->ah_attr.grh.traffic_class = cmd->base.dest.traffic_class; + attr->ah_attr.dlid = cmd->base.dest.dlid; + attr->ah_attr.sl = cmd->base.dest.sl; + attr->ah_attr.src_path_bits = cmd->base.dest.src_path_bits; + attr->ah_attr.static_rate = cmd->base.dest.static_rate; + attr->ah_attr.ah_flags = cmd->base.dest.is_global ? + IB_AH_GRH : 0; + attr->ah_attr.port_num = cmd->base.dest.port_num; + + memcpy(attr->alt_ah_attr.grh.dgid.raw, cmd->base.alt_dest.dgid, 16); + attr->alt_ah_attr.grh.flow_label = cmd->base.alt_dest.flow_label; + attr->alt_ah_attr.grh.sgid_index = cmd->base.alt_dest.sgid_index; + attr->alt_ah_attr.grh.hop_limit = cmd->base.alt_dest.hop_limit; + attr->alt_ah_attr.grh.traffic_class = cmd->base.alt_dest.traffic_class; + attr->alt_ah_attr.dlid = cmd->base.alt_dest.dlid; + attr->alt_ah_attr.sl = cmd->base.alt_dest.sl; + attr->alt_ah_attr.src_path_bits = cmd->base.alt_dest.src_path_bits; + attr->alt_ah_attr.static_rate = cmd->base.alt_dest.static_rate; + attr->alt_ah_attr.ah_flags = cmd->base.alt_dest.is_global ? + IB_AH_GRH : 0; + attr->alt_ah_attr.port_num = cmd->base.alt_dest.port_num; if (qp->real_qp == qp) { - ret = ib_resolve_eth_dmac(qp, attr, &cmd.attr_mask); - if (ret) - goto release_qp; + if (cmd->base.attr_mask & IB_QP_AV) { + ret = ib_resolve_eth_dmac(qp->device, &attr->ah_attr); + if (ret) + goto release_qp; + } ret = qp->device->modify_qp(qp, attr, - modify_qp_mask(qp->qp_type, cmd.attr_mask), &udata); + modify_qp_mask(qp->qp_type, + cmd->base.attr_mask), + udata); } else { - ret = ib_modify_qp(qp, attr, modify_qp_mask(qp->qp_type, cmd.attr_mask)); + ret = ib_modify_qp(qp, attr, + modify_qp_mask(qp->qp_type, + cmd->base.attr_mask)); } - if (ret) - goto release_qp; - - ret = in_len; - release_qp: put_qp_read(qp); @@ -2425,6 +2478,68 @@ out: return ret; } +ssize_t ib_uverbs_modify_qp(struct ib_uverbs_file *file, + struct ib_device *ib_dev, + const char __user *buf, int in_len, + int out_len) +{ + struct ib_uverbs_ex_modify_qp cmd = {}; + struct ib_udata udata; + int ret; + + if (copy_from_user(&cmd.base, buf, sizeof(cmd.base))) + return -EFAULT; + + if (cmd.base.attr_mask & + ~((IB_USER_LEGACY_LAST_QP_ATTR_MASK << 1) - 1)) + return -EOPNOTSUPP; + + INIT_UDATA(&udata, buf + sizeof(cmd.base), NULL, + in_len - sizeof(cmd.base), out_len); + + ret = modify_qp(file, &cmd, &udata); + if (ret) + return ret; + + return in_len; +} + +int ib_uverbs_ex_modify_qp(struct ib_uverbs_file *file, + struct ib_device *ib_dev, + struct ib_udata *ucore, + struct ib_udata *uhw) +{ + struct ib_uverbs_ex_modify_qp cmd = {}; + int ret; + + /* + * Last bit is reserved for extending the attr_mask by + * using another field. + */ + BUILD_BUG_ON(IB_USER_LAST_QP_ATTR_MASK == (1 << 31)); + + if (ucore->inlen < sizeof(cmd.base)) + return -EINVAL; + + ret = ib_copy_from_udata(&cmd, ucore, min(sizeof(cmd), ucore->inlen)); + if (ret) + return ret; + + if (cmd.base.attr_mask & + ~((IB_USER_LAST_QP_ATTR_MASK << 1) - 1)) + return -EOPNOTSUPP; + + if (ucore->inlen > sizeof(cmd)) { + if (ib_is_udata_cleared(ucore, sizeof(cmd), + ucore->inlen - sizeof(cmd))) + return -EOPNOTSUPP; + } + + ret = modify_qp(file, &cmd, uhw); + + return ret; +} + ssize_t ib_uverbs_destroy_qp(struct ib_uverbs_file *file, struct ib_device *ib_dev, const char __user *buf, int in_len, @@ -2462,6 +2577,8 @@ ssize_t ib_uverbs_destroy_qp(struct ib_uverbs_file *file, if (ret) return ret; + ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); + if (obj->uxrcd) atomic_dec(&obj->uxrcd->refcnt); @@ -2875,6 +2992,7 @@ ssize_t ib_uverbs_create_ah(struct ib_uverbs_file *file, struct ib_ah *ah; struct ib_ah_attr attr; int ret; + struct ib_udata udata; if (out_len < sizeof resp) return -ENOSPC; @@ -2882,6 +3000,10 @@ ssize_t ib_uverbs_create_ah(struct ib_uverbs_file *file, if (copy_from_user(&cmd, buf, sizeof cmd)) return -EFAULT; + INIT_UDATA(&udata, buf + sizeof(cmd), + (unsigned long)cmd.response + sizeof(resp), + in_len - sizeof(cmd), out_len - sizeof(resp)); + uobj = kmalloc(sizeof *uobj, GFP_KERNEL); if (!uobj) return -ENOMEM; @@ -2908,12 +3030,21 @@ ssize_t ib_uverbs_create_ah(struct ib_uverbs_file *file, memset(&attr.dmac, 0, sizeof(attr.dmac)); memcpy(attr.grh.dgid.raw, cmd.attr.grh.dgid, 16); - ah = ib_create_ah(pd, &attr); + ret = ib_rdmacg_try_charge(&uobj->cg_obj, ib_dev, + RDMACG_RESOURCE_HCA_OBJECT); + if (ret) + goto err_charge; + + ah = pd->device->create_ah(pd, &attr, &udata); + if (IS_ERR(ah)) { ret = PTR_ERR(ah); - goto err_put; + goto err_create; } + ah->device = pd->device; + ah->pd = pd; + atomic_inc(&pd->usecnt); ah->uobject = uobj; uobj->object = ah; @@ -2947,7 +3078,10 @@ err_copy: err_destroy: ib_destroy_ah(ah); -err_put: +err_create: + ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); + +err_charge: put_pd_read(pd); err: @@ -2981,6 +3115,8 @@ ssize_t ib_uverbs_destroy_ah(struct ib_uverbs_file *file, if (ret) return ret; + ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); + idr_remove_uobj(&ib_uverbs_ah_idr, uobj); mutex_lock(&file->mutex); @@ -3078,6 +3214,25 @@ out_put: return ret ? ret : in_len; } +static int kern_spec_to_ib_spec_action(struct ib_uverbs_flow_spec *kern_spec, + union ib_flow_spec *ib_spec) +{ + ib_spec->type = kern_spec->type; + switch (ib_spec->type) { + case IB_FLOW_SPEC_ACTION_TAG: + if (kern_spec->flow_tag.size != + sizeof(struct ib_uverbs_flow_spec_action_tag)) + return -EINVAL; + + ib_spec->flow_tag.size = sizeof(struct ib_flow_spec_action_tag); + ib_spec->flow_tag.tag_id = kern_spec->flow_tag.tag_id; + break; + default: + return -EINVAL; + } + return 0; +} + static size_t kern_spec_filter_sz(struct ib_uverbs_flow_spec_hdr *spec) { /* Returns user space filter size, includes padding */ @@ -3102,8 +3257,8 @@ static ssize_t spec_filter_size(void *kern_spec_filter, u16 kern_filter_size, return kern_filter_size; } -static int kern_spec_to_ib_spec(struct ib_uverbs_flow_spec *kern_spec, - union ib_flow_spec *ib_spec) +static int kern_spec_to_ib_spec_filter(struct ib_uverbs_flow_spec *kern_spec, + union ib_flow_spec *ib_spec) { ssize_t actual_filter_sz; ssize_t kern_filter_sz; @@ -3124,8 +3279,10 @@ static int kern_spec_to_ib_spec(struct ib_uverbs_flow_spec *kern_spec, kern_spec_val = (void *)kern_spec + sizeof(struct ib_uverbs_flow_spec_hdr); kern_spec_mask = kern_spec_val + kern_filter_sz; + if (ib_spec->type == (IB_FLOW_SPEC_INNER | IB_FLOW_SPEC_VXLAN_TUNNEL)) + return -EINVAL; - switch (ib_spec->type) { + switch (ib_spec->type & ~IB_FLOW_SPEC_INNER) { case IB_FLOW_SPEC_ETH: ib_filter_sz = offsetof(struct ib_flow_eth_filter, real_sz); actual_filter_sz = spec_filter_size(kern_spec_mask, @@ -3175,12 +3332,39 @@ static int kern_spec_to_ib_spec(struct ib_uverbs_flow_spec *kern_spec, memcpy(&ib_spec->tcp_udp.val, kern_spec_val, actual_filter_sz); memcpy(&ib_spec->tcp_udp.mask, kern_spec_mask, actual_filter_sz); break; + case IB_FLOW_SPEC_VXLAN_TUNNEL: + ib_filter_sz = offsetof(struct ib_flow_tunnel_filter, real_sz); + actual_filter_sz = spec_filter_size(kern_spec_mask, + kern_filter_sz, + ib_filter_sz); + if (actual_filter_sz <= 0) + return -EINVAL; + ib_spec->tunnel.size = sizeof(struct ib_flow_spec_tunnel); + memcpy(&ib_spec->tunnel.val, kern_spec_val, actual_filter_sz); + memcpy(&ib_spec->tunnel.mask, kern_spec_mask, actual_filter_sz); + + if ((ntohl(ib_spec->tunnel.mask.tunnel_id)) >= BIT(24) || + (ntohl(ib_spec->tunnel.val.tunnel_id)) >= BIT(24)) + return -EINVAL; + break; default: return -EINVAL; } return 0; } +static int kern_spec_to_ib_spec(struct ib_uverbs_flow_spec *kern_spec, + union ib_flow_spec *ib_spec) +{ + if (kern_spec->reserved) + return -EINVAL; + + if (kern_spec->type >= IB_FLOW_SPEC_ACTION_TAG) + return kern_spec_to_ib_spec_action(kern_spec, ib_spec); + else + return kern_spec_to_ib_spec_filter(kern_spec, ib_spec); +} + int ib_uverbs_ex_create_wq(struct ib_uverbs_file *file, struct ib_device *ib_dev, struct ib_udata *ucore, @@ -3243,6 +3427,9 @@ int ib_uverbs_ex_create_wq(struct ib_uverbs_file *file, wq_init_attr.wq_context = file; wq_init_attr.wq_type = cmd.wq_type; wq_init_attr.event_handler = ib_uverbs_wq_event_handler; + if (ucore->inlen >= (offsetof(typeof(cmd), create_flags) + + sizeof(cmd.create_flags))) + wq_init_attr.create_flags = cmd.create_flags; obj->uevent.events_reported = 0; INIT_LIST_HEAD(&obj->uevent.event_list); wq = pd->device->create_wq(pd, &wq_init_attr, uhw); @@ -3398,7 +3585,7 @@ int ib_uverbs_ex_modify_wq(struct ib_uverbs_file *file, if (!cmd.attr_mask) return -EINVAL; - if (cmd.attr_mask > (IB_WQ_STATE | IB_WQ_CUR_STATE)) + if (cmd.attr_mask > (IB_WQ_STATE | IB_WQ_CUR_STATE | IB_WQ_FLAGS)) return -EINVAL; wq = idr_read_wq(cmd.wq_handle, file->ucontext); @@ -3407,6 +3594,10 @@ int ib_uverbs_ex_modify_wq(struct ib_uverbs_file *file, wq_attr.curr_wq_state = cmd.curr_wq_state; wq_attr.wq_state = cmd.wq_state; + if (cmd.attr_mask & IB_WQ_FLAGS) { + wq_attr.flags = cmd.flags; + wq_attr.flags_mask = cmd.flags_mask; + } ret = wq->device->modify_wq(wq, &wq_attr, cmd.attr_mask, uhw); put_wq_read(wq); return ret; @@ -3740,12 +3931,17 @@ int ib_uverbs_ex_create_flow(struct ib_uverbs_file *file, err = -EINVAL; goto err_free; } + + err = ib_rdmacg_try_charge(&uobj->cg_obj, ib_dev, + RDMACG_RESOURCE_HCA_OBJECT); + if (err) + goto err_free; + flow_id = ib_create_flow(qp, flow_attr, IB_FLOW_DOMAIN_USER); if (IS_ERR(flow_id)) { err = PTR_ERR(flow_id); - goto err_free; + goto err_create; } - flow_id->qp = qp; flow_id->uobject = uobj; uobj->object = flow_id; @@ -3777,6 +3973,8 @@ err_copy: idr_remove_uobj(&ib_uverbs_rule_idr, uobj); destroy_flow: ib_destroy_flow(flow_id); +err_create: + ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); err_free: kfree(flow_attr); err_put: @@ -3816,8 +4014,11 @@ int ib_uverbs_ex_destroy_flow(struct ib_uverbs_file *file, flow_id = uobj->object; ret = ib_destroy_flow(flow_id); - if (!ret) + if (!ret) { + ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, + RDMACG_RESOURCE_HCA_OBJECT); uobj->live = 0; + } put_uobj_write(uobj); @@ -3885,6 +4086,11 @@ static int __uverbs_create_xsrq(struct ib_uverbs_file *file, obj->uevent.events_reported = 0; INIT_LIST_HEAD(&obj->uevent.event_list); + ret = ib_rdmacg_try_charge(&obj->uevent.uobject.cg_obj, ib_dev, + RDMACG_RESOURCE_HCA_OBJECT); + if (ret) + goto err_put_cq; + srq = pd->device->create_srq(pd, &attr, udata); if (IS_ERR(srq)) { ret = PTR_ERR(srq); @@ -3949,6 +4155,8 @@ err_destroy: ib_destroy_srq(srq); err_put: + ib_rdmacg_uncharge(&obj->uevent.uobject.cg_obj, ib_dev, + RDMACG_RESOURCE_HCA_OBJECT); put_pd_read(pd); err_put_cq: @@ -4135,6 +4343,8 @@ ssize_t ib_uverbs_destroy_srq(struct ib_uverbs_file *file, if (ret) return ret; + ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); + if (srq_type == IB_SRQT_XRC) { us = container_of(obj, struct ib_usrq_object, uevent); atomic_dec(&us->uxrcd->refcnt); @@ -4242,6 +4452,12 @@ int ib_uverbs_ex_query_device(struct ib_uverbs_file *file, resp.max_wq_type_rq = attr.max_wq_type_rq; resp.response_length += sizeof(resp.max_wq_type_rq); + + if (ucore->outlen < resp.response_length + sizeof(resp.raw_packet_caps)) + goto end; + + resp.raw_packet_caps = attr.raw_packet_caps; + resp.response_length += sizeof(resp.raw_packet_caps); end: err = ib_copy_to_udata(ucore, &resp, resp.response_length); return err; diff --git a/drivers/infiniband/core/uverbs_main.c b/drivers/infiniband/core/uverbs_main.c index 44b1104eb168..35c788a32e26 100644 --- a/drivers/infiniband/core/uverbs_main.c +++ b/drivers/infiniband/core/uverbs_main.c @@ -46,11 +46,12 @@ #include <linux/anon_inodes.h> #include <linux/slab.h> -#include <asm/uaccess.h> +#include <linux/uaccess.h> #include <rdma/ib.h> #include "uverbs.h" +#include "core_priv.h" MODULE_AUTHOR("Roland Dreier"); MODULE_DESCRIPTION("InfiniBand userspace verbs access"); @@ -137,6 +138,7 @@ static int (*uverbs_ex_cmd_table[])(struct ib_uverbs_file *file, [IB_USER_VERBS_EX_CMD_DESTROY_WQ] = ib_uverbs_ex_destroy_wq, [IB_USER_VERBS_EX_CMD_CREATE_RWQ_IND_TBL] = ib_uverbs_ex_create_rwq_ind_table, [IB_USER_VERBS_EX_CMD_DESTROY_RWQ_IND_TBL] = ib_uverbs_ex_destroy_rwq_ind_table, + [IB_USER_VERBS_EX_CMD_MODIFY_QP] = ib_uverbs_ex_modify_qp, }; static void ib_uverbs_add_one(struct ib_device *device); @@ -236,6 +238,8 @@ static int ib_uverbs_cleanup_ucontext(struct ib_uverbs_file *file, idr_remove_uobj(&ib_uverbs_ah_idr, uobj); ib_destroy_ah(ah); + ib_rdmacg_uncharge(&uobj->cg_obj, context->device, + RDMACG_RESOURCE_HCA_OBJECT); kfree(uobj); } @@ -245,6 +249,8 @@ static int ib_uverbs_cleanup_ucontext(struct ib_uverbs_file *file, idr_remove_uobj(&ib_uverbs_mw_idr, uobj); uverbs_dealloc_mw(mw); + ib_rdmacg_uncharge(&uobj->cg_obj, context->device, + RDMACG_RESOURCE_HCA_OBJECT); kfree(uobj); } @@ -253,6 +259,8 @@ static int ib_uverbs_cleanup_ucontext(struct ib_uverbs_file *file, idr_remove_uobj(&ib_uverbs_rule_idr, uobj); ib_destroy_flow(flow_id); + ib_rdmacg_uncharge(&uobj->cg_obj, context->device, + RDMACG_RESOURCE_HCA_OBJECT); kfree(uobj); } @@ -265,6 +273,8 @@ static int ib_uverbs_cleanup_ucontext(struct ib_uverbs_file *file, if (qp == qp->real_qp) ib_uverbs_detach_umcast(qp, uqp); ib_destroy_qp(qp); + ib_rdmacg_uncharge(&uobj->cg_obj, context->device, + RDMACG_RESOURCE_HCA_OBJECT); ib_uverbs_release_uevent(file, &uqp->uevent); kfree(uqp); } @@ -297,6 +307,8 @@ static int ib_uverbs_cleanup_ucontext(struct ib_uverbs_file *file, idr_remove_uobj(&ib_uverbs_srq_idr, uobj); ib_destroy_srq(srq); + ib_rdmacg_uncharge(&uobj->cg_obj, context->device, + RDMACG_RESOURCE_HCA_OBJECT); ib_uverbs_release_uevent(file, uevent); kfree(uevent); } @@ -309,6 +321,8 @@ static int ib_uverbs_cleanup_ucontext(struct ib_uverbs_file *file, idr_remove_uobj(&ib_uverbs_cq_idr, uobj); ib_destroy_cq(cq); + ib_rdmacg_uncharge(&uobj->cg_obj, context->device, + RDMACG_RESOURCE_HCA_OBJECT); ib_uverbs_release_ucq(file, ev_file, ucq); kfree(ucq); } @@ -318,6 +332,8 @@ static int ib_uverbs_cleanup_ucontext(struct ib_uverbs_file *file, idr_remove_uobj(&ib_uverbs_mr_idr, uobj); ib_dereg_mr(mr); + ib_rdmacg_uncharge(&uobj->cg_obj, context->device, + RDMACG_RESOURCE_HCA_OBJECT); kfree(uobj); } @@ -338,11 +354,16 @@ static int ib_uverbs_cleanup_ucontext(struct ib_uverbs_file *file, idr_remove_uobj(&ib_uverbs_pd_idr, uobj); ib_dealloc_pd(pd); + ib_rdmacg_uncharge(&uobj->cg_obj, context->device, + RDMACG_RESOURCE_HCA_OBJECT); kfree(uobj); } put_pid(context->tgid); + ib_rdmacg_uncharge(&context->cg_obj, context->device, + RDMACG_RESOURCE_HCA_HANDLE); + return context->device->dealloc_ucontext(context); } @@ -746,8 +767,11 @@ static ssize_t ib_uverbs_write(struct file *filp, const char __user *buf, int srcu_key; ssize_t ret; - if (WARN_ON_ONCE(!ib_safe_file_access(filp))) + if (!ib_safe_file_access(filp)) { + pr_err_once("uverbs_write: process %d (%s) changed security contexts after opening file descriptor, this is not allowed.\n", + task_tgid_vnr(current), current->comm); return -EACCES; + } if (count < sizeof hdr) return -EINVAL; @@ -1170,7 +1194,7 @@ static void ib_uverbs_add_one(struct ib_device *device) if (cdev_add(&uverbs_dev->cdev, base, 1)) goto err_cdev; - uverbs_dev->dev = device_create(uverbs_class, device->dma_device, + uverbs_dev->dev = device_create(uverbs_class, device->dev.parent, uverbs_dev->cdev.dev, uverbs_dev, "uverbs%d", uverbs_dev->devnum); if (IS_ERR(uverbs_dev->dev)) diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c index 83687646da68..85ed5051fdfd 100644 --- a/drivers/infiniband/core/verbs.c +++ b/drivers/infiniband/core/verbs.c @@ -315,7 +315,7 @@ struct ib_ah *ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr) { struct ib_ah *ah; - ah = pd->device->create_ah(pd, ah_attr); + ah = pd->device->create_ah(pd, ah_attr, NULL); if (!IS_ERR(ah)) { ah->device = pd->device; @@ -328,7 +328,7 @@ struct ib_ah *ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr) } EXPORT_SYMBOL(ib_create_ah); -static int ib_get_header_version(const union rdma_network_hdr *hdr) +int ib_get_rdma_header_version(const union rdma_network_hdr *hdr) { const struct iphdr *ip4h = (struct iphdr *)&hdr->roce4grh; struct iphdr ip4h_checked; @@ -359,6 +359,7 @@ static int ib_get_header_version(const union rdma_network_hdr *hdr) return 4; return 6; } +EXPORT_SYMBOL(ib_get_rdma_header_version); static enum rdma_network_type ib_get_net_type_by_grh(struct ib_device *device, u8 port_num, @@ -369,7 +370,7 @@ static enum rdma_network_type ib_get_net_type_by_grh(struct ib_device *device, if (rdma_protocol_ib(device, port_num)) return RDMA_NETWORK_IB; - grh_version = ib_get_header_version((union rdma_network_hdr *)grh); + grh_version = ib_get_rdma_header_version((union rdma_network_hdr *)grh); if (grh_version == 4) return RDMA_NETWORK_IPV4; @@ -415,9 +416,9 @@ static int get_sgid_index_from_eth(struct ib_device *device, u8 port_num, &context, gid_index); } -static int get_gids_from_rdma_hdr(union rdma_network_hdr *hdr, - enum rdma_network_type net_type, - union ib_gid *sgid, union ib_gid *dgid) +int ib_get_gids_from_rdma_hdr(const union rdma_network_hdr *hdr, + enum rdma_network_type net_type, + union ib_gid *sgid, union ib_gid *dgid) { struct sockaddr_in src_in; struct sockaddr_in dst_in; @@ -447,6 +448,7 @@ static int get_gids_from_rdma_hdr(union rdma_network_hdr *hdr, return -EINVAL; } } +EXPORT_SYMBOL(ib_get_gids_from_rdma_hdr); int ib_init_ah_from_wc(struct ib_device *device, u8 port_num, const struct ib_wc *wc, const struct ib_grh *grh, @@ -469,8 +471,8 @@ int ib_init_ah_from_wc(struct ib_device *device, u8 port_num, net_type = ib_get_net_type_by_grh(device, port_num, grh); gid_type = ib_network_to_gid_type(net_type); } - ret = get_gids_from_rdma_hdr((union rdma_network_hdr *)grh, net_type, - &sgid, &dgid); + ret = ib_get_gids_from_rdma_hdr((union rdma_network_hdr *)grh, net_type, + &sgid, &dgid); if (ret) return ret; @@ -1014,6 +1016,7 @@ static const struct { IB_QP_QKEY), [IB_QPT_GSI] = (IB_QP_CUR_STATE | IB_QP_QKEY), + [IB_QPT_RAW_PACKET] = IB_QP_RATE_LIMIT, } } }, @@ -1047,6 +1050,7 @@ static const struct { IB_QP_QKEY), [IB_QPT_GSI] = (IB_QP_CUR_STATE | IB_QP_QKEY), + [IB_QPT_RAW_PACKET] = IB_QP_RATE_LIMIT, } }, [IB_QPS_SQD] = { @@ -1196,66 +1200,65 @@ int ib_modify_qp_is_ok(enum ib_qp_state cur_state, enum ib_qp_state next_state, } EXPORT_SYMBOL(ib_modify_qp_is_ok); -int ib_resolve_eth_dmac(struct ib_qp *qp, - struct ib_qp_attr *qp_attr, int *qp_attr_mask) +int ib_resolve_eth_dmac(struct ib_device *device, + struct ib_ah_attr *ah_attr) { int ret = 0; - if (*qp_attr_mask & IB_QP_AV) { - if (qp_attr->ah_attr.port_num < rdma_start_port(qp->device) || - qp_attr->ah_attr.port_num > rdma_end_port(qp->device)) - return -EINVAL; - - if (!rdma_cap_eth_ah(qp->device, qp_attr->ah_attr.port_num)) - return 0; - - if (rdma_link_local_addr((struct in6_addr *)qp_attr->ah_attr.grh.dgid.raw)) { - rdma_get_ll_mac((struct in6_addr *)qp_attr->ah_attr.grh.dgid.raw, - qp_attr->ah_attr.dmac); - } else { - union ib_gid sgid; - struct ib_gid_attr sgid_attr; - int ifindex; - int hop_limit; - - ret = ib_query_gid(qp->device, - qp_attr->ah_attr.port_num, - qp_attr->ah_attr.grh.sgid_index, - &sgid, &sgid_attr); - - if (ret || !sgid_attr.ndev) { - if (!ret) - ret = -ENXIO; - goto out; - } + if (!rdma_is_port_valid(device, ah_attr->port_num)) + return -EINVAL; - ifindex = sgid_attr.ndev->ifindex; + if (!rdma_cap_eth_ah(device, ah_attr->port_num)) + return 0; - ret = rdma_addr_find_l2_eth_by_grh(&sgid, - &qp_attr->ah_attr.grh.dgid, - qp_attr->ah_attr.dmac, - NULL, &ifindex, &hop_limit); + if (rdma_link_local_addr((struct in6_addr *)ah_attr->grh.dgid.raw)) { + rdma_get_ll_mac((struct in6_addr *)ah_attr->grh.dgid.raw, + ah_attr->dmac); + } else { + union ib_gid sgid; + struct ib_gid_attr sgid_attr; + int ifindex; + int hop_limit; + + ret = ib_query_gid(device, + ah_attr->port_num, + ah_attr->grh.sgid_index, + &sgid, &sgid_attr); + + if (ret || !sgid_attr.ndev) { + if (!ret) + ret = -ENXIO; + goto out; + } - dev_put(sgid_attr.ndev); + ifindex = sgid_attr.ndev->ifindex; - qp_attr->ah_attr.grh.hop_limit = hop_limit; - } + ret = rdma_addr_find_l2_eth_by_grh(&sgid, + &ah_attr->grh.dgid, + ah_attr->dmac, + NULL, &ifindex, &hop_limit); + + dev_put(sgid_attr.ndev); + + ah_attr->grh.hop_limit = hop_limit; } out: return ret; } EXPORT_SYMBOL(ib_resolve_eth_dmac); - int ib_modify_qp(struct ib_qp *qp, struct ib_qp_attr *qp_attr, int qp_attr_mask) { - int ret; - ret = ib_resolve_eth_dmac(qp, qp_attr, &qp_attr_mask); - if (ret) - return ret; + if (qp_attr_mask & IB_QP_AV) { + int ret; + + ret = ib_resolve_eth_dmac(qp->device, &qp_attr->ah_attr); + if (ret) + return ret; + } return qp->device->modify_qp(qp->real_qp, qp_attr, qp_attr_mask, NULL); } @@ -1734,8 +1737,10 @@ struct ib_flow *ib_create_flow(struct ib_qp *qp, return ERR_PTR(-ENOSYS); flow_id = qp->device->create_flow(qp, flow_attr, domain); - if (!IS_ERR(flow_id)) + if (!IS_ERR(flow_id)) { atomic_inc(&qp->usecnt); + flow_id->qp = qp; + } return flow_id; } EXPORT_SYMBOL(ib_create_flow); @@ -1943,17 +1948,12 @@ static void ib_drain_qp_done(struct ib_cq *cq, struct ib_wc *wc) */ static void __ib_drain_sq(struct ib_qp *qp) { + struct ib_cq *cq = qp->send_cq; struct ib_qp_attr attr = { .qp_state = IB_QPS_ERR }; struct ib_drain_cqe sdrain; struct ib_send_wr swr = {}, *bad_swr; int ret; - if (qp->send_cq->poll_ctx == IB_POLL_DIRECT) { - WARN_ONCE(qp->send_cq->poll_ctx == IB_POLL_DIRECT, - "IB_POLL_DIRECT poll_ctx not supported for drain\n"); - return; - } - swr.wr_cqe = &sdrain.cqe; sdrain.cqe.done = ib_drain_qp_done; init_completion(&sdrain.done); @@ -1970,7 +1970,11 @@ static void __ib_drain_sq(struct ib_qp *qp) return; } - wait_for_completion(&sdrain.done); + if (cq->poll_ctx == IB_POLL_DIRECT) + while (wait_for_completion_timeout(&sdrain.done, HZ / 10) <= 0) + ib_process_cq_direct(cq, -1); + else + wait_for_completion(&sdrain.done); } /* @@ -1978,17 +1982,12 @@ static void __ib_drain_sq(struct ib_qp *qp) */ static void __ib_drain_rq(struct ib_qp *qp) { + struct ib_cq *cq = qp->recv_cq; struct ib_qp_attr attr = { .qp_state = IB_QPS_ERR }; struct ib_drain_cqe rdrain; struct ib_recv_wr rwr = {}, *bad_rwr; int ret; - if (qp->recv_cq->poll_ctx == IB_POLL_DIRECT) { - WARN_ONCE(qp->recv_cq->poll_ctx == IB_POLL_DIRECT, - "IB_POLL_DIRECT poll_ctx not supported for drain\n"); - return; - } - rwr.wr_cqe = &rdrain.cqe; rdrain.cqe.done = ib_drain_qp_done; init_completion(&rdrain.done); @@ -2005,7 +2004,11 @@ static void __ib_drain_rq(struct ib_qp *qp) return; } - wait_for_completion(&rdrain.done); + if (cq->poll_ctx == IB_POLL_DIRECT) + while (wait_for_completion_timeout(&rdrain.done, HZ / 10) <= 0) + ib_process_cq_direct(cq, -1); + else + wait_for_completion(&rdrain.done); } /** @@ -2022,8 +2025,7 @@ static void __ib_drain_rq(struct ib_qp *qp) * ensure there is room in the CQ and SQ for the drain work request and * completion. * - * allocate the CQ using ib_alloc_cq() and the CQ poll context cannot be - * IB_POLL_DIRECT. + * allocate the CQ using ib_alloc_cq(). * * ensure that there are no other contexts that are posting WRs concurrently. * Otherwise the drain is not guaranteed. @@ -2051,8 +2053,7 @@ EXPORT_SYMBOL(ib_drain_sq); * ensure there is room in the CQ and RQ for the drain work request and * completion. * - * allocate the CQ using ib_alloc_cq() and the CQ poll context cannot be - * IB_POLL_DIRECT. + * allocate the CQ using ib_alloc_cq(). * * ensure that there are no other contexts that are posting WRs concurrently. * Otherwise the drain is not guaranteed. @@ -2076,8 +2077,7 @@ EXPORT_SYMBOL(ib_drain_rq); * ensure there is room in the CQ(s), SQ, and RQ for drain work requests * and completions. * - * allocate the CQs using ib_alloc_cq() and the CQ poll context cannot be - * IB_POLL_DIRECT. + * allocate the CQs using ib_alloc_cq(). * * ensure that there are no other contexts that are posting WRs concurrently. * Otherwise the drain is not guaranteed. diff --git a/drivers/infiniband/hw/Makefile b/drivers/infiniband/hw/Makefile index e7a5ed9f6f3f..34c93abf0fe0 100644 --- a/drivers/infiniband/hw/Makefile +++ b/drivers/infiniband/hw/Makefile @@ -7,7 +7,9 @@ obj-$(CONFIG_MLX4_INFINIBAND) += mlx4/ obj-$(CONFIG_MLX5_INFINIBAND) += mlx5/ obj-$(CONFIG_INFINIBAND_NES) += nes/ obj-$(CONFIG_INFINIBAND_OCRDMA) += ocrdma/ +obj-$(CONFIG_INFINIBAND_VMWARE_PVRDMA) += vmw_pvrdma/ obj-$(CONFIG_INFINIBAND_USNIC) += usnic/ obj-$(CONFIG_INFINIBAND_HFI1) += hfi1/ obj-$(CONFIG_INFINIBAND_HNS) += hns/ obj-$(CONFIG_INFINIBAND_QEDR) += qedr/ +obj-$(CONFIG_INFINIBAND_BNXT_RE) += bnxt_re/ diff --git a/drivers/infiniband/hw/bnxt_re/Kconfig b/drivers/infiniband/hw/bnxt_re/Kconfig new file mode 100644 index 000000000000..19982a4a9bba --- /dev/null +++ b/drivers/infiniband/hw/bnxt_re/Kconfig @@ -0,0 +1,9 @@ +config INFINIBAND_BNXT_RE + tristate "Broadcom Netxtreme HCA support" + depends on ETHERNET && NETDEVICES && PCI && INET && DCB + select NET_VENDOR_BROADCOM + select BNXT + ---help--- + This driver supports Broadcom NetXtreme-E 10/25/40/50 gigabit + RoCE HCAs. To compile this driver as a module, choose M here: + the module will be called bnxt_re. diff --git a/drivers/infiniband/hw/bnxt_re/Makefile b/drivers/infiniband/hw/bnxt_re/Makefile new file mode 100644 index 000000000000..036f84efbc73 --- /dev/null +++ b/drivers/infiniband/hw/bnxt_re/Makefile @@ -0,0 +1,6 @@ + +ccflags-y := -Idrivers/net/ethernet/broadcom/bnxt +obj-$(CONFIG_INFINIBAND_BNXT_RE) += bnxt_re.o +bnxt_re-y := main.o ib_verbs.o \ + qplib_res.o qplib_rcfw.o \ + qplib_sp.o qplib_fp.o diff --git a/drivers/infiniband/hw/bnxt_re/bnxt_re.h b/drivers/infiniband/hw/bnxt_re/bnxt_re.h new file mode 100644 index 000000000000..ebf7be8d4139 --- /dev/null +++ b/drivers/infiniband/hw/bnxt_re/bnxt_re.h @@ -0,0 +1,146 @@ +/* + * Broadcom NetXtreme-E RoCE driver. + * + * Copyright (c) 2016 - 2017, Broadcom. All rights reserved. The term + * Broadcom refers to Broadcom Limited and/or its subsidiaries. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, + * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS + * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE + * OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN + * IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * Description: Slow Path Operators (header) + * + */ + +#ifndef __BNXT_RE_H__ +#define __BNXT_RE_H__ +#define ROCE_DRV_MODULE_NAME "bnxt_re" +#define ROCE_DRV_MODULE_VERSION "1.0.0" + +#define BNXT_RE_DESC "Broadcom NetXtreme-C/E RoCE Driver" + +#define BNXT_RE_PAGE_SIZE_4K BIT(12) +#define BNXT_RE_PAGE_SIZE_8K BIT(13) +#define BNXT_RE_PAGE_SIZE_64K BIT(16) +#define BNXT_RE_PAGE_SIZE_2M BIT(21) +#define BNXT_RE_PAGE_SIZE_8M BIT(23) +#define BNXT_RE_PAGE_SIZE_1G BIT(30) + +#define BNXT_RE_MAX_QPC_COUNT (64 * 1024) +#define BNXT_RE_MAX_MRW_COUNT (64 * 1024) +#define BNXT_RE_MAX_SRQC_COUNT (64 * 1024) +#define BNXT_RE_MAX_CQ_COUNT (64 * 1024) + +struct bnxt_re_work { + struct work_struct work; + unsigned long event; + struct bnxt_re_dev *rdev; + struct net_device *vlan_dev; +}; + +struct bnxt_re_sqp_entries { + struct bnxt_qplib_sge sge; + u64 wrid; + /* For storing the actual qp1 cqe */ + struct bnxt_qplib_cqe cqe; + struct bnxt_re_qp *qp1_qp; +}; + +#define BNXT_RE_MIN_MSIX 2 +#define BNXT_RE_MAX_MSIX 16 +#define BNXT_RE_AEQ_IDX 0 +#define BNXT_RE_NQ_IDX 1 + +struct bnxt_re_dev { + struct ib_device ibdev; + struct list_head list; + unsigned long flags; +#define BNXT_RE_FLAG_NETDEV_REGISTERED 0 +#define BNXT_RE_FLAG_IBDEV_REGISTERED 1 +#define BNXT_RE_FLAG_GOT_MSIX 2 +#define BNXT_RE_FLAG_RCFW_CHANNEL_EN 8 +#define BNXT_RE_FLAG_QOS_WORK_REG 16 + struct net_device *netdev; + unsigned int version, major, minor; + struct bnxt_en_dev *en_dev; + struct bnxt_msix_entry msix_entries[BNXT_RE_MAX_MSIX]; + int num_msix; + + int id; + + struct delayed_work worker; + u8 cur_prio_map; + + /* FP Notification Queue (CQ & SRQ) */ + struct tasklet_struct nq_task; + + /* RCFW Channel */ + struct bnxt_qplib_rcfw rcfw; + + /* NQ */ + struct bnxt_qplib_nq nq; + + /* Device Resources */ + struct bnxt_qplib_dev_attr dev_attr; + struct bnxt_qplib_ctx qplib_ctx; + struct bnxt_qplib_res qplib_res; + struct bnxt_qplib_dpi dpi_privileged; + + atomic_t qp_count; + struct mutex qp_lock; /* protect qp list */ + struct list_head qp_list; + + atomic_t cq_count; + atomic_t srq_count; + atomic_t mr_count; + atomic_t mw_count; + /* Max of 2 lossless traffic class supported per port */ + u16 cosq[2]; + + /* QP for for handling QP1 packets */ + u32 sqp_id; + struct bnxt_re_qp *qp1_sqp; + struct bnxt_re_ah *sqp_ah; + struct bnxt_re_sqp_entries sqp_tbl[1024]; +}; + +#define to_bnxt_re_dev(ptr, member) \ + container_of((ptr), struct bnxt_re_dev, member) + +#define BNXT_RE_ROCE_V1_PACKET 0 +#define BNXT_RE_ROCEV2_IPV4_PACKET 2 +#define BNXT_RE_ROCEV2_IPV6_PACKET 3 + +static inline struct device *rdev_to_dev(struct bnxt_re_dev *rdev) +{ + if (rdev) + return &rdev->ibdev.dev; + return NULL; +} + +#endif diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.c b/drivers/infiniband/hw/bnxt_re/ib_verbs.c new file mode 100644 index 000000000000..33af2e3de399 --- /dev/null +++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.c @@ -0,0 +1,3202 @@ +/* + * Broadcom NetXtreme-E RoCE driver. + * + * Copyright (c) 2016 - 2017, Broadcom. All rights reserved. The term + * Broadcom refers to Broadcom Limited and/or its subsidiaries. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, + * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS + * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE + * OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN + * IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * Description: IB Verbs interpreter + */ + +#include <linux/interrupt.h> +#include <linux/types.h> +#include <linux/pci.h> +#include <linux/netdevice.h> +#include <linux/if_ether.h> + +#include <rdma/ib_verbs.h> +#include <rdma/ib_user_verbs.h> +#include <rdma/ib_umem.h> +#include <rdma/ib_addr.h> +#include <rdma/ib_mad.h> +#include <rdma/ib_cache.h> + +#include "bnxt_ulp.h" + +#include "roce_hsi.h" +#include "qplib_res.h" +#include "qplib_sp.h" +#include "qplib_fp.h" +#include "qplib_rcfw.h" + +#include "bnxt_re.h" +#include "ib_verbs.h" +#include <rdma/bnxt_re-abi.h> + +static int bnxt_re_build_sgl(struct ib_sge *ib_sg_list, + struct bnxt_qplib_sge *sg_list, int num) +{ + int i, total = 0; + + for (i = 0; i < num; i++) { + sg_list[i].addr = ib_sg_list[i].addr; + sg_list[i].lkey = ib_sg_list[i].lkey; + sg_list[i].size = ib_sg_list[i].length; + total += sg_list[i].size; + } + return total; +} + +/* Device */ +struct net_device *bnxt_re_get_netdev(struct ib_device *ibdev, u8 port_num) +{ + struct bnxt_re_dev *rdev = to_bnxt_re_dev(ibdev, ibdev); + struct net_device *netdev = NULL; + + rcu_read_lock(); + if (rdev) + netdev = rdev->netdev; + if (netdev) + dev_hold(netdev); + + rcu_read_unlock(); + return netdev; +} + +int bnxt_re_query_device(struct ib_device *ibdev, + struct ib_device_attr *ib_attr, + struct ib_udata *udata) +{ + struct bnxt_re_dev *rdev = to_bnxt_re_dev(ibdev, ibdev); + struct bnxt_qplib_dev_attr *dev_attr = &rdev->dev_attr; + + memset(ib_attr, 0, sizeof(*ib_attr)); + + ib_attr->fw_ver = (u64)(unsigned long)(dev_attr->fw_ver); + bnxt_qplib_get_guid(rdev->netdev->dev_addr, + (u8 *)&ib_attr->sys_image_guid); + ib_attr->max_mr_size = ~0ull; + ib_attr->page_size_cap = BNXT_RE_PAGE_SIZE_4K | BNXT_RE_PAGE_SIZE_8K | + BNXT_RE_PAGE_SIZE_64K | BNXT_RE_PAGE_SIZE_2M | + BNXT_RE_PAGE_SIZE_8M | BNXT_RE_PAGE_SIZE_1G; + + ib_attr->vendor_id = rdev->en_dev->pdev->vendor; + ib_attr->vendor_part_id = rdev->en_dev->pdev->device; + ib_attr->hw_ver = rdev->en_dev->pdev->subsystem_device; + ib_attr->max_qp = dev_attr->max_qp; + ib_attr->max_qp_wr = dev_attr->max_qp_wqes; + ib_attr->device_cap_flags = + IB_DEVICE_CURR_QP_STATE_MOD + | IB_DEVICE_RC_RNR_NAK_GEN + | IB_DEVICE_SHUTDOWN_PORT + | IB_DEVICE_SYS_IMAGE_GUID + | IB_DEVICE_LOCAL_DMA_LKEY + | IB_DEVICE_RESIZE_MAX_WR + | IB_DEVICE_PORT_ACTIVE_EVENT + | IB_DEVICE_N_NOTIFY_CQ + | IB_DEVICE_MEM_WINDOW + | IB_DEVICE_MEM_WINDOW_TYPE_2B + | IB_DEVICE_MEM_MGT_EXTENSIONS; + ib_attr->max_sge = dev_attr->max_qp_sges; + ib_attr->max_sge_rd = dev_attr->max_qp_sges; + ib_attr->max_cq = dev_attr->max_cq; + ib_attr->max_cqe = dev_attr->max_cq_wqes; + ib_attr->max_mr = dev_attr->max_mr; + ib_attr->max_pd = dev_attr->max_pd; + ib_attr->max_qp_rd_atom = dev_attr->max_qp_rd_atom; + ib_attr->max_qp_init_rd_atom = dev_attr->max_qp_rd_atom; + ib_attr->atomic_cap = IB_ATOMIC_HCA; + ib_attr->masked_atomic_cap = IB_ATOMIC_HCA; + + ib_attr->max_ee_rd_atom = 0; + ib_attr->max_res_rd_atom = 0; + ib_attr->max_ee_init_rd_atom = 0; + ib_attr->max_ee = 0; + ib_attr->max_rdd = 0; + ib_attr->max_mw = dev_attr->max_mw; + ib_attr->max_raw_ipv6_qp = 0; + ib_attr->max_raw_ethy_qp = dev_attr->max_raw_ethy_qp; + ib_attr->max_mcast_grp = 0; + ib_attr->max_mcast_qp_attach = 0; + ib_attr->max_total_mcast_qp_attach = 0; + ib_attr->max_ah = dev_attr->max_ah; + + ib_attr->max_fmr = dev_attr->max_fmr; + ib_attr->max_map_per_fmr = 1; /* ? */ + + ib_attr->max_srq = dev_attr->max_srq; + ib_attr->max_srq_wr = dev_attr->max_srq_wqes; + ib_attr->max_srq_sge = dev_attr->max_srq_sges; + + ib_attr->max_fast_reg_page_list_len = MAX_PBL_LVL_1_PGS; + + ib_attr->max_pkeys = 1; + ib_attr->local_ca_ack_delay = 0; + return 0; +} + +int bnxt_re_modify_device(struct ib_device *ibdev, + int device_modify_mask, + struct ib_device_modify *device_modify) +{ + switch (device_modify_mask) { + case IB_DEVICE_MODIFY_SYS_IMAGE_GUID: + /* Modify the GUID requires the modification of the GID table */ + /* GUID should be made as READ-ONLY */ + break; + case IB_DEVICE_MODIFY_NODE_DESC: + /* Node Desc should be made as READ-ONLY */ + break; + default: + break; + } + return 0; +} + +static void __to_ib_speed_width(struct net_device *netdev, u8 *speed, u8 *width) +{ + struct ethtool_link_ksettings lksettings; + u32 espeed; + + if (netdev->ethtool_ops && netdev->ethtool_ops->get_link_ksettings) { + memset(&lksettings, 0, sizeof(lksettings)); + rtnl_lock(); + netdev->ethtool_ops->get_link_ksettings(netdev, &lksettings); + rtnl_unlock(); + espeed = lksettings.base.speed; + } else { + espeed = SPEED_UNKNOWN; + } + switch (espeed) { + case SPEED_1000: + *speed = IB_SPEED_SDR; + *width = IB_WIDTH_1X; + break; + case SPEED_10000: + *speed = IB_SPEED_QDR; + *width = IB_WIDTH_1X; + break; + case SPEED_20000: + *speed = IB_SPEED_DDR; + *width = IB_WIDTH_4X; + break; + case SPEED_25000: + *speed = IB_SPEED_EDR; + *width = IB_WIDTH_1X; + break; + case SPEED_40000: + *speed = IB_SPEED_QDR; + *width = IB_WIDTH_4X; + break; + case SPEED_50000: + break; + default: + *speed = IB_SPEED_SDR; + *width = IB_WIDTH_1X; + break; + } +} + +/* Port */ +int bnxt_re_query_port(struct ib_device *ibdev, u8 port_num, + struct ib_port_attr *port_attr) +{ + struct bnxt_re_dev *rdev = to_bnxt_re_dev(ibdev, ibdev); + struct bnxt_qplib_dev_attr *dev_attr = &rdev->dev_attr; + + memset(port_attr, 0, sizeof(*port_attr)); + + if (netif_running(rdev->netdev) && netif_carrier_ok(rdev->netdev)) { + port_attr->state = IB_PORT_ACTIVE; + port_attr->phys_state = 5; + } else { + port_attr->state = IB_PORT_DOWN; + port_attr->phys_state = 3; + } + port_attr->max_mtu = IB_MTU_4096; + port_attr->active_mtu = iboe_get_mtu(rdev->netdev->mtu); + port_attr->gid_tbl_len = dev_attr->max_sgid; + port_attr->port_cap_flags = IB_PORT_CM_SUP | IB_PORT_REINIT_SUP | + IB_PORT_DEVICE_MGMT_SUP | + IB_PORT_VENDOR_CLASS_SUP | + IB_PORT_IP_BASED_GIDS; + + /* Max MSG size set to 2G for now */ + port_attr->max_msg_sz = 0x80000000; + port_attr->bad_pkey_cntr = 0; + port_attr->qkey_viol_cntr = 0; + port_attr->pkey_tbl_len = dev_attr->max_pkey; + port_attr->lid = 0; + port_attr->sm_lid = 0; + port_attr->lmc = 0; + port_attr->max_vl_num = 4; + port_attr->sm_sl = 0; + port_attr->subnet_timeout = 0; + port_attr->init_type_reply = 0; + /* call the underlying netdev's ethtool hooks to query speed settings + * for which we acquire rtnl_lock _only_ if it's registered with + * IB stack to avoid race in the NETDEV_UNREG path + */ + if (test_bit(BNXT_RE_FLAG_IBDEV_REGISTERED, &rdev->flags)) + __to_ib_speed_width(rdev->netdev, &port_attr->active_speed, + &port_attr->active_width); + return 0; +} + +int bnxt_re_modify_port(struct ib_device *ibdev, u8 port_num, + int port_modify_mask, + struct ib_port_modify *port_modify) +{ + switch (port_modify_mask) { + case IB_PORT_SHUTDOWN: + break; + case IB_PORT_INIT_TYPE: + break; + case IB_PORT_RESET_QKEY_CNTR: + break; + default: + break; + } + return 0; +} + +int bnxt_re_get_port_immutable(struct ib_device *ibdev, u8 port_num, + struct ib_port_immutable *immutable) +{ + struct ib_port_attr port_attr; + + if (bnxt_re_query_port(ibdev, port_num, &port_attr)) + return -EINVAL; + + immutable->pkey_tbl_len = port_attr.pkey_tbl_len; + immutable->gid_tbl_len = port_attr.gid_tbl_len; + immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE; + immutable->core_cap_flags |= RDMA_CORE_CAP_PROT_ROCE_UDP_ENCAP; + immutable->max_mad_size = IB_MGMT_MAD_SIZE; + return 0; +} + +int bnxt_re_query_pkey(struct ib_device *ibdev, u8 port_num, + u16 index, u16 *pkey) +{ + struct bnxt_re_dev *rdev = to_bnxt_re_dev(ibdev, ibdev); + + /* Ignore port_num */ + + memset(pkey, 0, sizeof(*pkey)); + return bnxt_qplib_get_pkey(&rdev->qplib_res, + &rdev->qplib_res.pkey_tbl, index, pkey); +} + +int bnxt_re_query_gid(struct ib_device *ibdev, u8 port_num, + int index, union ib_gid *gid) +{ + struct bnxt_re_dev *rdev = to_bnxt_re_dev(ibdev, ibdev); + int rc = 0; + + /* Ignore port_num */ + memset(gid, 0, sizeof(*gid)); + rc = bnxt_qplib_get_sgid(&rdev->qplib_res, + &rdev->qplib_res.sgid_tbl, index, + (struct bnxt_qplib_gid *)gid); + return rc; +} + +int bnxt_re_del_gid(struct ib_device *ibdev, u8 port_num, + unsigned int index, void **context) +{ + int rc = 0; + struct bnxt_re_gid_ctx *ctx, **ctx_tbl; + struct bnxt_re_dev *rdev = to_bnxt_re_dev(ibdev, ibdev); + struct bnxt_qplib_sgid_tbl *sgid_tbl = &rdev->qplib_res.sgid_tbl; + + /* Delete the entry from the hardware */ + ctx = *context; + if (!ctx) + return -EINVAL; + + if (sgid_tbl && sgid_tbl->active) { + if (ctx->idx >= sgid_tbl->max) + return -EINVAL; + ctx->refcnt--; + if (!ctx->refcnt) { + rc = bnxt_qplib_del_sgid + (sgid_tbl, + &sgid_tbl->tbl[ctx->idx], true); + if (rc) + dev_err(rdev_to_dev(rdev), + "Failed to remove GID: %#x", rc); + ctx_tbl = sgid_tbl->ctx; + ctx_tbl[ctx->idx] = NULL; + kfree(ctx); + } + } else { + return -EINVAL; + } + return rc; +} + +int bnxt_re_add_gid(struct ib_device *ibdev, u8 port_num, + unsigned int index, const union ib_gid *gid, + const struct ib_gid_attr *attr, void **context) +{ + int rc; + u32 tbl_idx = 0; + u16 vlan_id = 0xFFFF; + struct bnxt_re_gid_ctx *ctx, **ctx_tbl; + struct bnxt_re_dev *rdev = to_bnxt_re_dev(ibdev, ibdev); + struct bnxt_qplib_sgid_tbl *sgid_tbl = &rdev->qplib_res.sgid_tbl; + + if ((attr->ndev) && is_vlan_dev(attr->ndev)) + vlan_id = vlan_dev_vlan_id(attr->ndev); + + rc = bnxt_qplib_add_sgid(sgid_tbl, (struct bnxt_qplib_gid *)gid, + rdev->qplib_res.netdev->dev_addr, + vlan_id, true, &tbl_idx); + if (rc == -EALREADY) { + ctx_tbl = sgid_tbl->ctx; + ctx_tbl[tbl_idx]->refcnt++; + *context = ctx_tbl[tbl_idx]; + return 0; + } + + if (rc < 0) { + dev_err(rdev_to_dev(rdev), "Failed to add GID: %#x", rc); + return rc; + } + + ctx = kmalloc(sizeof(*ctx), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + ctx_tbl = sgid_tbl->ctx; + ctx->idx = tbl_idx; + ctx->refcnt = 1; + ctx_tbl[tbl_idx] = ctx; + + return rc; +} + +enum rdma_link_layer bnxt_re_get_link_layer(struct ib_device *ibdev, + u8 port_num) +{ + return IB_LINK_LAYER_ETHERNET; +} + +/* Protection Domains */ +int bnxt_re_dealloc_pd(struct ib_pd *ib_pd) +{ + struct bnxt_re_pd *pd = container_of(ib_pd, struct bnxt_re_pd, ib_pd); + struct bnxt_re_dev *rdev = pd->rdev; + int rc; + + if (ib_pd->uobject && pd->dpi.dbr) { + struct ib_ucontext *ib_uctx = ib_pd->uobject->context; + struct bnxt_re_ucontext *ucntx; + + /* Free DPI only if this is the first PD allocated by the + * application and mark the context dpi as NULL + */ + ucntx = container_of(ib_uctx, struct bnxt_re_ucontext, ib_uctx); + + rc = bnxt_qplib_dealloc_dpi(&rdev->qplib_res, + &rdev->qplib_res.dpi_tbl, + &pd->dpi); + if (rc) + dev_err(rdev_to_dev(rdev), "Failed to deallocate HW DPI"); + /* Don't fail, continue*/ + ucntx->dpi = NULL; + } + + rc = bnxt_qplib_dealloc_pd(&rdev->qplib_res, + &rdev->qplib_res.pd_tbl, + &pd->qplib_pd); + if (rc) { + dev_err(rdev_to_dev(rdev), "Failed to deallocate HW PD"); + return rc; + } + + kfree(pd); + return 0; +} + +struct ib_pd *bnxt_re_alloc_pd(struct ib_device *ibdev, + struct ib_ucontext *ucontext, + struct ib_udata *udata) +{ + struct bnxt_re_dev *rdev = to_bnxt_re_dev(ibdev, ibdev); + struct bnxt_re_ucontext *ucntx = container_of(ucontext, + struct bnxt_re_ucontext, + ib_uctx); + struct bnxt_re_pd *pd; + int rc; + + pd = kzalloc(sizeof(*pd), GFP_KERNEL); + if (!pd) + return ERR_PTR(-ENOMEM); + + pd->rdev = rdev; + if (bnxt_qplib_alloc_pd(&rdev->qplib_res.pd_tbl, &pd->qplib_pd)) { + dev_err(rdev_to_dev(rdev), "Failed to allocate HW PD"); + rc = -ENOMEM; + goto fail; + } + + if (udata) { + struct bnxt_re_pd_resp resp; + + if (!ucntx->dpi) { + /* Allocate DPI in alloc_pd to avoid failing of + * ibv_devinfo and family of application when DPIs + * are depleted. + */ + if (bnxt_qplib_alloc_dpi(&rdev->qplib_res.dpi_tbl, + &pd->dpi, ucntx)) { + rc = -ENOMEM; + goto dbfail; + } + ucntx->dpi = &pd->dpi; + } + + resp.pdid = pd->qplib_pd.id; + /* Still allow mapping this DBR to the new user PD. */ + resp.dpi = ucntx->dpi->dpi; + resp.dbr = (u64)ucntx->dpi->umdbr; + + rc = ib_copy_to_udata(udata, &resp, sizeof(resp)); + if (rc) { + dev_err(rdev_to_dev(rdev), + "Failed to copy user response\n"); + goto dbfail; + } + } + + return &pd->ib_pd; +dbfail: + (void)bnxt_qplib_dealloc_pd(&rdev->qplib_res, &rdev->qplib_res.pd_tbl, + &pd->qplib_pd); +fail: + kfree(pd); + return ERR_PTR(rc); +} + +/* Address Handles */ +int bnxt_re_destroy_ah(struct ib_ah *ib_ah) +{ + struct bnxt_re_ah *ah = container_of(ib_ah, struct bnxt_re_ah, ib_ah); + struct bnxt_re_dev *rdev = ah->rdev; + int rc; + + rc = bnxt_qplib_destroy_ah(&rdev->qplib_res, &ah->qplib_ah); + if (rc) { + dev_err(rdev_to_dev(rdev), "Failed to destroy HW AH"); + return rc; + } + kfree(ah); + return 0; +} + +struct ib_ah *bnxt_re_create_ah(struct ib_pd *ib_pd, + struct ib_ah_attr *ah_attr, + struct ib_udata *udata) +{ + struct bnxt_re_pd *pd = container_of(ib_pd, struct bnxt_re_pd, ib_pd); + struct bnxt_re_dev *rdev = pd->rdev; + struct bnxt_re_ah *ah; + int rc; + u16 vlan_tag; + u8 nw_type; + + struct ib_gid_attr sgid_attr; + + if (!(ah_attr->ah_flags & IB_AH_GRH)) { + dev_err(rdev_to_dev(rdev), "Failed to alloc AH: GRH not set"); + return ERR_PTR(-EINVAL); + } + ah = kzalloc(sizeof(*ah), GFP_ATOMIC); + if (!ah) + return ERR_PTR(-ENOMEM); + + ah->rdev = rdev; + ah->qplib_ah.pd = &pd->qplib_pd; + + /* Supply the configuration for the HW */ + memcpy(ah->qplib_ah.dgid.data, ah_attr->grh.dgid.raw, + sizeof(union ib_gid)); + /* + * If RoCE V2 is enabled, stack will have two entries for + * each GID entry. Avoiding this duplicte entry in HW. Dividing + * the GID index by 2 for RoCE V2 + */ + ah->qplib_ah.sgid_index = ah_attr->grh.sgid_index / 2; + ah->qplib_ah.host_sgid_index = ah_attr->grh.sgid_index; + ah->qplib_ah.traffic_class = ah_attr->grh.traffic_class; + ah->qplib_ah.flow_label = ah_attr->grh.flow_label; + ah->qplib_ah.hop_limit = ah_attr->grh.hop_limit; + ah->qplib_ah.sl = ah_attr->sl; + if (ib_pd->uobject && + !rdma_is_multicast_addr((struct in6_addr *) + ah_attr->grh.dgid.raw) && + !rdma_link_local_addr((struct in6_addr *) + ah_attr->grh.dgid.raw)) { + union ib_gid sgid; + + rc = ib_get_cached_gid(&rdev->ibdev, 1, + ah_attr->grh.sgid_index, &sgid, + &sgid_attr); + if (rc) { + dev_err(rdev_to_dev(rdev), + "Failed to query gid at index %d", + ah_attr->grh.sgid_index); + goto fail; + } + if (sgid_attr.ndev) { + if (is_vlan_dev(sgid_attr.ndev)) + vlan_tag = vlan_dev_vlan_id(sgid_attr.ndev); + dev_put(sgid_attr.ndev); + } + /* Get network header type for this GID */ + nw_type = ib_gid_to_network_type(sgid_attr.gid_type, &sgid); + switch (nw_type) { + case RDMA_NETWORK_IPV4: + ah->qplib_ah.nw_type = CMDQ_CREATE_AH_TYPE_V2IPV4; + break; + case RDMA_NETWORK_IPV6: + ah->qplib_ah.nw_type = CMDQ_CREATE_AH_TYPE_V2IPV6; + break; + default: + ah->qplib_ah.nw_type = CMDQ_CREATE_AH_TYPE_V1; + break; + } + rc = rdma_addr_find_l2_eth_by_grh(&sgid, &ah_attr->grh.dgid, + ah_attr->dmac, &vlan_tag, + &sgid_attr.ndev->ifindex, + NULL); + if (rc) { + dev_err(rdev_to_dev(rdev), "Failed to get dmac\n"); + goto fail; + } + } + + memcpy(ah->qplib_ah.dmac, ah_attr->dmac, ETH_ALEN); + rc = bnxt_qplib_create_ah(&rdev->qplib_res, &ah->qplib_ah); + if (rc) { + dev_err(rdev_to_dev(rdev), "Failed to allocate HW AH"); + goto fail; + } + + /* Write AVID to shared page. */ + if (ib_pd->uobject) { + struct ib_ucontext *ib_uctx = ib_pd->uobject->context; + struct bnxt_re_ucontext *uctx; + unsigned long flag; + u32 *wrptr; + + uctx = container_of(ib_uctx, struct bnxt_re_ucontext, ib_uctx); + spin_lock_irqsave(&uctx->sh_lock, flag); + wrptr = (u32 *)(uctx->shpg + BNXT_RE_AVID_OFFT); + *wrptr = ah->qplib_ah.id; + wmb(); /* make sure cache is updated. */ + spin_unlock_irqrestore(&uctx->sh_lock, flag); + } + + return &ah->ib_ah; + +fail: + kfree(ah); + return ERR_PTR(rc); +} + +int bnxt_re_modify_ah(struct ib_ah *ib_ah, struct ib_ah_attr *ah_attr) +{ + return 0; +} + +int bnxt_re_query_ah(struct ib_ah *ib_ah, struct ib_ah_attr *ah_attr) +{ + struct bnxt_re_ah *ah = container_of(ib_ah, struct bnxt_re_ah, ib_ah); + + memcpy(ah_attr->grh.dgid.raw, ah->qplib_ah.dgid.data, + sizeof(union ib_gid)); + ah_attr->grh.sgid_index = ah->qplib_ah.host_sgid_index; + ah_attr->grh.traffic_class = ah->qplib_ah.traffic_class; + ah_attr->sl = ah->qplib_ah.sl; + memcpy(ah_attr->dmac, ah->qplib_ah.dmac, ETH_ALEN); + ah_attr->ah_flags = IB_AH_GRH; + ah_attr->port_num = 1; + ah_attr->static_rate = 0; + return 0; +} + +/* Queue Pairs */ +int bnxt_re_destroy_qp(struct ib_qp *ib_qp) +{ + struct bnxt_re_qp *qp = container_of(ib_qp, struct bnxt_re_qp, ib_qp); + struct bnxt_re_dev *rdev = qp->rdev; + int rc; + + rc = bnxt_qplib_destroy_qp(&rdev->qplib_res, &qp->qplib_qp); + if (rc) { + dev_err(rdev_to_dev(rdev), "Failed to destroy HW QP"); + return rc; + } + if (ib_qp->qp_type == IB_QPT_GSI && rdev->qp1_sqp) { + rc = bnxt_qplib_destroy_ah(&rdev->qplib_res, + &rdev->sqp_ah->qplib_ah); + if (rc) { + dev_err(rdev_to_dev(rdev), + "Failed to destroy HW AH for shadow QP"); + return rc; + } + + rc = bnxt_qplib_destroy_qp(&rdev->qplib_res, + &rdev->qp1_sqp->qplib_qp); + if (rc) { + dev_err(rdev_to_dev(rdev), + "Failed to destroy Shadow QP"); + return rc; + } + mutex_lock(&rdev->qp_lock); + list_del(&rdev->qp1_sqp->list); + atomic_dec(&rdev->qp_count); + mutex_unlock(&rdev->qp_lock); + + kfree(rdev->sqp_ah); + kfree(rdev->qp1_sqp); + } + + if (qp->rumem && !IS_ERR(qp->rumem)) + ib_umem_release(qp->rumem); + if (qp->sumem && !IS_ERR(qp->sumem)) + ib_umem_release(qp->sumem); + + mutex_lock(&rdev->qp_lock); + list_del(&qp->list); + atomic_dec(&rdev->qp_count); + mutex_unlock(&rdev->qp_lock); + kfree(qp); + return 0; +} + +static u8 __from_ib_qp_type(enum ib_qp_type type) +{ + switch (type) { + case IB_QPT_GSI: + return CMDQ_CREATE_QP1_TYPE_GSI; + case IB_QPT_RC: + return CMDQ_CREATE_QP_TYPE_RC; + case IB_QPT_UD: + return CMDQ_CREATE_QP_TYPE_UD; + default: + return IB_QPT_MAX; + } +} + +static int bnxt_re_init_user_qp(struct bnxt_re_dev *rdev, struct bnxt_re_pd *pd, + struct bnxt_re_qp *qp, struct ib_udata *udata) +{ + struct bnxt_re_qp_req ureq; + struct bnxt_qplib_qp *qplib_qp = &qp->qplib_qp; + struct ib_umem *umem; + int bytes = 0; + struct ib_ucontext *context = pd->ib_pd.uobject->context; + struct bnxt_re_ucontext *cntx = container_of(context, + struct bnxt_re_ucontext, + ib_uctx); + if (ib_copy_from_udata(&ureq, udata, sizeof(ureq))) + return -EFAULT; + + bytes = (qplib_qp->sq.max_wqe * BNXT_QPLIB_MAX_SQE_ENTRY_SIZE); + /* Consider mapping PSN search memory only for RC QPs. */ + if (qplib_qp->type == CMDQ_CREATE_QP_TYPE_RC) + bytes += (qplib_qp->sq.max_wqe * sizeof(struct sq_psn_search)); + bytes = PAGE_ALIGN(bytes); + umem = ib_umem_get(context, ureq.qpsva, bytes, + IB_ACCESS_LOCAL_WRITE, 1); + if (IS_ERR(umem)) + return PTR_ERR(umem); + + qp->sumem = umem; + qplib_qp->sq.sglist = umem->sg_head.sgl; + qplib_qp->sq.nmap = umem->nmap; + qplib_qp->qp_handle = ureq.qp_handle; + + if (!qp->qplib_qp.srq) { + bytes = (qplib_qp->rq.max_wqe * BNXT_QPLIB_MAX_RQE_ENTRY_SIZE); + bytes = PAGE_ALIGN(bytes); + umem = ib_umem_get(context, ureq.qprva, bytes, + IB_ACCESS_LOCAL_WRITE, 1); + if (IS_ERR(umem)) + goto rqfail; + qp->rumem = umem; + qplib_qp->rq.sglist = umem->sg_head.sgl; + qplib_qp->rq.nmap = umem->nmap; + } + + qplib_qp->dpi = cntx->dpi; + return 0; +rqfail: + ib_umem_release(qp->sumem); + qp->sumem = NULL; + qplib_qp->sq.sglist = NULL; + qplib_qp->sq.nmap = 0; + + return PTR_ERR(umem); +} + +static struct bnxt_re_ah *bnxt_re_create_shadow_qp_ah + (struct bnxt_re_pd *pd, + struct bnxt_qplib_res *qp1_res, + struct bnxt_qplib_qp *qp1_qp) +{ + struct bnxt_re_dev *rdev = pd->rdev; + struct bnxt_re_ah *ah; + union ib_gid sgid; + int rc; + + ah = kzalloc(sizeof(*ah), GFP_KERNEL); + if (!ah) + return NULL; + + memset(ah, 0, sizeof(*ah)); + ah->rdev = rdev; + ah->qplib_ah.pd = &pd->qplib_pd; + + rc = bnxt_re_query_gid(&rdev->ibdev, 1, 0, &sgid); + if (rc) + goto fail; + + /* supply the dgid data same as sgid */ + memcpy(ah->qplib_ah.dgid.data, &sgid.raw, + sizeof(union ib_gid)); + ah->qplib_ah.sgid_index = 0; + + ah->qplib_ah.traffic_class = 0; + ah->qplib_ah.flow_label = 0; + ah->qplib_ah.hop_limit = 1; + ah->qplib_ah.sl = 0; + /* Have DMAC same as SMAC */ + ether_addr_copy(ah->qplib_ah.dmac, rdev->netdev->dev_addr); + + rc = bnxt_qplib_create_ah(&rdev->qplib_res, &ah->qplib_ah); + if (rc) { + dev_err(rdev_to_dev(rdev), + "Failed to allocate HW AH for Shadow QP"); + goto fail; + } + + return ah; + +fail: + kfree(ah); + return NULL; +} + +static struct bnxt_re_qp *bnxt_re_create_shadow_qp + (struct bnxt_re_pd *pd, + struct bnxt_qplib_res *qp1_res, + struct bnxt_qplib_qp *qp1_qp) +{ + struct bnxt_re_dev *rdev = pd->rdev; + struct bnxt_re_qp *qp; + int rc; + + qp = kzalloc(sizeof(*qp), GFP_KERNEL); + if (!qp) + return NULL; + + memset(qp, 0, sizeof(*qp)); + qp->rdev = rdev; + + /* Initialize the shadow QP structure from the QP1 values */ + ether_addr_copy(qp->qplib_qp.smac, rdev->netdev->dev_addr); + + qp->qplib_qp.pd = &pd->qplib_pd; + qp->qplib_qp.qp_handle = (u64)(unsigned long)(&qp->qplib_qp); + qp->qplib_qp.type = IB_QPT_UD; + + qp->qplib_qp.max_inline_data = 0; + qp->qplib_qp.sig_type = true; + + /* Shadow QP SQ depth should be same as QP1 RQ depth */ + qp->qplib_qp.sq.max_wqe = qp1_qp->rq.max_wqe; + qp->qplib_qp.sq.max_sge = 2; + + qp->qplib_qp.scq = qp1_qp->scq; + qp->qplib_qp.rcq = qp1_qp->rcq; + + qp->qplib_qp.rq.max_wqe = qp1_qp->rq.max_wqe; + qp->qplib_qp.rq.max_sge = qp1_qp->rq.max_sge; + + qp->qplib_qp.mtu = qp1_qp->mtu; + + qp->qplib_qp.sq_hdr_buf_size = 0; + qp->qplib_qp.rq_hdr_buf_size = BNXT_QPLIB_MAX_GRH_HDR_SIZE_IPV6; + qp->qplib_qp.dpi = &rdev->dpi_privileged; + + rc = bnxt_qplib_create_qp(qp1_res, &qp->qplib_qp); + if (rc) + goto fail; + + rdev->sqp_id = qp->qplib_qp.id; + + spin_lock_init(&qp->sq_lock); + INIT_LIST_HEAD(&qp->list); + mutex_lock(&rdev->qp_lock); + list_add_tail(&qp->list, &rdev->qp_list); + atomic_inc(&rdev->qp_count); + mutex_unlock(&rdev->qp_lock); + return qp; +fail: + kfree(qp); + return NULL; +} + +struct ib_qp *bnxt_re_create_qp(struct ib_pd *ib_pd, + struct ib_qp_init_attr *qp_init_attr, + struct ib_udata *udata) +{ + struct bnxt_re_pd *pd = container_of(ib_pd, struct bnxt_re_pd, ib_pd); + struct bnxt_re_dev *rdev = pd->rdev; + struct bnxt_qplib_dev_attr *dev_attr = &rdev->dev_attr; + struct bnxt_re_qp *qp; + struct bnxt_re_cq *cq; + int rc, entries; + + if ((qp_init_attr->cap.max_send_wr > dev_attr->max_qp_wqes) || + (qp_init_attr->cap.max_recv_wr > dev_attr->max_qp_wqes) || + (qp_init_attr->cap.max_send_sge > dev_attr->max_qp_sges) || + (qp_init_attr->cap.max_recv_sge > dev_attr->max_qp_sges) || + (qp_init_attr->cap.max_inline_data > dev_attr->max_inline_data)) + return ERR_PTR(-EINVAL); + + qp = kzalloc(sizeof(*qp), GFP_KERNEL); + if (!qp) + return ERR_PTR(-ENOMEM); + + qp->rdev = rdev; + ether_addr_copy(qp->qplib_qp.smac, rdev->netdev->dev_addr); + qp->qplib_qp.pd = &pd->qplib_pd; + qp->qplib_qp.qp_handle = (u64)(unsigned long)(&qp->qplib_qp); + qp->qplib_qp.type = __from_ib_qp_type(qp_init_attr->qp_type); + if (qp->qplib_qp.type == IB_QPT_MAX) { + dev_err(rdev_to_dev(rdev), "QP type 0x%x not supported", + qp->qplib_qp.type); + rc = -EINVAL; + goto fail; + } + qp->qplib_qp.max_inline_data = qp_init_attr->cap.max_inline_data; + qp->qplib_qp.sig_type = ((qp_init_attr->sq_sig_type == + IB_SIGNAL_ALL_WR) ? true : false); + + entries = roundup_pow_of_two(qp_init_attr->cap.max_send_wr + 1); + qp->qplib_qp.sq.max_wqe = min_t(u32, entries, + dev_attr->max_qp_wqes + 1); + + qp->qplib_qp.sq.max_sge = qp_init_attr->cap.max_send_sge; + if (qp->qplib_qp.sq.max_sge > dev_attr->max_qp_sges) + qp->qplib_qp.sq.max_sge = dev_attr->max_qp_sges; + + if (qp_init_attr->send_cq) { + cq = container_of(qp_init_attr->send_cq, struct bnxt_re_cq, + ib_cq); + if (!cq) { + dev_err(rdev_to_dev(rdev), "Send CQ not found"); + rc = -EINVAL; + goto fail; + } + qp->qplib_qp.scq = &cq->qplib_cq; + } + + if (qp_init_attr->recv_cq) { + cq = container_of(qp_init_attr->recv_cq, struct bnxt_re_cq, + ib_cq); + if (!cq) { + dev_err(rdev_to_dev(rdev), "Receive CQ not found"); + rc = -EINVAL; + goto fail; + } + qp->qplib_qp.rcq = &cq->qplib_cq; + } + + if (qp_init_attr->srq) { + dev_err(rdev_to_dev(rdev), "SRQ not supported"); + rc = -ENOTSUPP; + goto fail; + } else { + /* Allocate 1 more than what's provided so posting max doesn't + * mean empty + */ + entries = roundup_pow_of_two(qp_init_attr->cap.max_recv_wr + 1); + qp->qplib_qp.rq.max_wqe = min_t(u32, entries, + dev_attr->max_qp_wqes + 1); + + qp->qplib_qp.rq.max_sge = qp_init_attr->cap.max_recv_sge; + if (qp->qplib_qp.rq.max_sge > dev_attr->max_qp_sges) + qp->qplib_qp.rq.max_sge = dev_attr->max_qp_sges; + } + + qp->qplib_qp.mtu = ib_mtu_enum_to_int(iboe_get_mtu(rdev->netdev->mtu)); + + if (qp_init_attr->qp_type == IB_QPT_GSI) { + qp->qplib_qp.rq.max_sge = dev_attr->max_qp_sges; + if (qp->qplib_qp.rq.max_sge > dev_attr->max_qp_sges) + qp->qplib_qp.rq.max_sge = dev_attr->max_qp_sges; + qp->qplib_qp.sq.max_sge++; + if (qp->qplib_qp.sq.max_sge > dev_attr->max_qp_sges) + qp->qplib_qp.sq.max_sge = dev_attr->max_qp_sges; + + qp->qplib_qp.rq_hdr_buf_size = + BNXT_QPLIB_MAX_QP1_RQ_HDR_SIZE_V2; + + qp->qplib_qp.sq_hdr_buf_size = + BNXT_QPLIB_MAX_QP1_SQ_HDR_SIZE_V2; + qp->qplib_qp.dpi = &rdev->dpi_privileged; + rc = bnxt_qplib_create_qp1(&rdev->qplib_res, &qp->qplib_qp); + if (rc) { + dev_err(rdev_to_dev(rdev), "Failed to create HW QP1"); + goto fail; + } + /* Create a shadow QP to handle the QP1 traffic */ + rdev->qp1_sqp = bnxt_re_create_shadow_qp(pd, &rdev->qplib_res, + &qp->qplib_qp); + if (!rdev->qp1_sqp) { + rc = -EINVAL; + dev_err(rdev_to_dev(rdev), + "Failed to create Shadow QP for QP1"); + goto qp_destroy; + } + rdev->sqp_ah = bnxt_re_create_shadow_qp_ah(pd, &rdev->qplib_res, + &qp->qplib_qp); + if (!rdev->sqp_ah) { + bnxt_qplib_destroy_qp(&rdev->qplib_res, + &rdev->qp1_sqp->qplib_qp); + rc = -EINVAL; + dev_err(rdev_to_dev(rdev), + "Failed to create AH entry for ShadowQP"); + goto qp_destroy; + } + + } else { + qp->qplib_qp.max_rd_atomic = dev_attr->max_qp_rd_atom; + qp->qplib_qp.max_dest_rd_atomic = dev_attr->max_qp_init_rd_atom; + if (udata) { + rc = bnxt_re_init_user_qp(rdev, pd, qp, udata); + if (rc) + goto fail; + } else { + qp->qplib_qp.dpi = &rdev->dpi_privileged; + } + + rc = bnxt_qplib_create_qp(&rdev->qplib_res, &qp->qplib_qp); + if (rc) { + dev_err(rdev_to_dev(rdev), "Failed to create HW QP"); + goto fail; + } + } + + qp->ib_qp.qp_num = qp->qplib_qp.id; + spin_lock_init(&qp->sq_lock); + + if (udata) { + struct bnxt_re_qp_resp resp; + + resp.qpid = qp->ib_qp.qp_num; + resp.rsvd = 0; + rc = ib_copy_to_udata(udata, &resp, sizeof(resp)); + if (rc) { + dev_err(rdev_to_dev(rdev), "Failed to copy QP udata"); + goto qp_destroy; + } + } + INIT_LIST_HEAD(&qp->list); + mutex_lock(&rdev->qp_lock); + list_add_tail(&qp->list, &rdev->qp_list); + atomic_inc(&rdev->qp_count); + mutex_unlock(&rdev->qp_lock); + + return &qp->ib_qp; +qp_destroy: + bnxt_qplib_destroy_qp(&rdev->qplib_res, &qp->qplib_qp); +fail: + kfree(qp); + return ERR_PTR(rc); +} + +static u8 __from_ib_qp_state(enum ib_qp_state state) +{ + switch (state) { + case IB_QPS_RESET: + return CMDQ_MODIFY_QP_NEW_STATE_RESET; + case IB_QPS_INIT: + return CMDQ_MODIFY_QP_NEW_STATE_INIT; + case IB_QPS_RTR: + return CMDQ_MODIFY_QP_NEW_STATE_RTR; + case IB_QPS_RTS: + return CMDQ_MODIFY_QP_NEW_STATE_RTS; + case IB_QPS_SQD: + return CMDQ_MODIFY_QP_NEW_STATE_SQD; + case IB_QPS_SQE: + return CMDQ_MODIFY_QP_NEW_STATE_SQE; + case IB_QPS_ERR: + default: + return CMDQ_MODIFY_QP_NEW_STATE_ERR; + } +} + +static enum ib_qp_state __to_ib_qp_state(u8 state) +{ + switch (state) { + case CMDQ_MODIFY_QP_NEW_STATE_RESET: + return IB_QPS_RESET; + case CMDQ_MODIFY_QP_NEW_STATE_INIT: + return IB_QPS_INIT; + case CMDQ_MODIFY_QP_NEW_STATE_RTR: + return IB_QPS_RTR; + case CMDQ_MODIFY_QP_NEW_STATE_RTS: + return IB_QPS_RTS; + case CMDQ_MODIFY_QP_NEW_STATE_SQD: + return IB_QPS_SQD; + case CMDQ_MODIFY_QP_NEW_STATE_SQE: + return IB_QPS_SQE; + case CMDQ_MODIFY_QP_NEW_STATE_ERR: + default: + return IB_QPS_ERR; + } +} + +static u32 __from_ib_mtu(enum ib_mtu mtu) +{ + switch (mtu) { + case IB_MTU_256: + return CMDQ_MODIFY_QP_PATH_MTU_MTU_256; + case IB_MTU_512: + return CMDQ_MODIFY_QP_PATH_MTU_MTU_512; + case IB_MTU_1024: + return CMDQ_MODIFY_QP_PATH_MTU_MTU_1024; + case IB_MTU_2048: + return CMDQ_MODIFY_QP_PATH_MTU_MTU_2048; + case IB_MTU_4096: + return CMDQ_MODIFY_QP_PATH_MTU_MTU_4096; + default: + return CMDQ_MODIFY_QP_PATH_MTU_MTU_2048; + } +} + +static enum ib_mtu __to_ib_mtu(u32 mtu) +{ + switch (mtu & CREQ_QUERY_QP_RESP_SB_PATH_MTU_MASK) { + case CMDQ_MODIFY_QP_PATH_MTU_MTU_256: + return IB_MTU_256; + case CMDQ_MODIFY_QP_PATH_MTU_MTU_512: + return IB_MTU_512; + case CMDQ_MODIFY_QP_PATH_MTU_MTU_1024: + return IB_MTU_1024; + case CMDQ_MODIFY_QP_PATH_MTU_MTU_2048: + return IB_MTU_2048; + case CMDQ_MODIFY_QP_PATH_MTU_MTU_4096: + return IB_MTU_4096; + default: + return IB_MTU_2048; + } +} + +static int __from_ib_access_flags(int iflags) +{ + int qflags = 0; + + if (iflags & IB_ACCESS_LOCAL_WRITE) + qflags |= BNXT_QPLIB_ACCESS_LOCAL_WRITE; + if (iflags & IB_ACCESS_REMOTE_READ) + qflags |= BNXT_QPLIB_ACCESS_REMOTE_READ; + if (iflags & IB_ACCESS_REMOTE_WRITE) + qflags |= BNXT_QPLIB_ACCESS_REMOTE_WRITE; + if (iflags & IB_ACCESS_REMOTE_ATOMIC) + qflags |= BNXT_QPLIB_ACCESS_REMOTE_ATOMIC; + if (iflags & IB_ACCESS_MW_BIND) + qflags |= BNXT_QPLIB_ACCESS_MW_BIND; + if (iflags & IB_ZERO_BASED) + qflags |= BNXT_QPLIB_ACCESS_ZERO_BASED; + if (iflags & IB_ACCESS_ON_DEMAND) + qflags |= BNXT_QPLIB_ACCESS_ON_DEMAND; + return qflags; +}; + +static enum ib_access_flags __to_ib_access_flags(int qflags) +{ + enum ib_access_flags iflags = 0; + + if (qflags & BNXT_QPLIB_ACCESS_LOCAL_WRITE) + iflags |= IB_ACCESS_LOCAL_WRITE; + if (qflags & BNXT_QPLIB_ACCESS_REMOTE_WRITE) + iflags |= IB_ACCESS_REMOTE_WRITE; + if (qflags & BNXT_QPLIB_ACCESS_REMOTE_READ) + iflags |= IB_ACCESS_REMOTE_READ; + if (qflags & BNXT_QPLIB_ACCESS_REMOTE_ATOMIC) + iflags |= IB_ACCESS_REMOTE_ATOMIC; + if (qflags & BNXT_QPLIB_ACCESS_MW_BIND) + iflags |= IB_ACCESS_MW_BIND; + if (qflags & BNXT_QPLIB_ACCESS_ZERO_BASED) + iflags |= IB_ZERO_BASED; + if (qflags & BNXT_QPLIB_ACCESS_ON_DEMAND) + iflags |= IB_ACCESS_ON_DEMAND; + return iflags; +}; + +static int bnxt_re_modify_shadow_qp(struct bnxt_re_dev *rdev, + struct bnxt_re_qp *qp1_qp, + int qp_attr_mask) +{ + struct bnxt_re_qp *qp = rdev->qp1_sqp; + int rc = 0; + + if (qp_attr_mask & IB_QP_STATE) { + qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_STATE; + qp->qplib_qp.state = qp1_qp->qplib_qp.state; + } + if (qp_attr_mask & IB_QP_PKEY_INDEX) { + qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_PKEY; + qp->qplib_qp.pkey_index = qp1_qp->qplib_qp.pkey_index; + } + + if (qp_attr_mask & IB_QP_QKEY) { + qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_QKEY; + /* Using a Random QKEY */ + qp->qplib_qp.qkey = 0x81818181; + } + if (qp_attr_mask & IB_QP_SQ_PSN) { + qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_SQ_PSN; + qp->qplib_qp.sq.psn = qp1_qp->qplib_qp.sq.psn; + } + + rc = bnxt_qplib_modify_qp(&rdev->qplib_res, &qp->qplib_qp); + if (rc) + dev_err(rdev_to_dev(rdev), + "Failed to modify Shadow QP for QP1"); + return rc; +} + +int bnxt_re_modify_qp(struct ib_qp *ib_qp, struct ib_qp_attr *qp_attr, + int qp_attr_mask, struct ib_udata *udata) +{ + struct bnxt_re_qp *qp = container_of(ib_qp, struct bnxt_re_qp, ib_qp); + struct bnxt_re_dev *rdev = qp->rdev; + struct bnxt_qplib_dev_attr *dev_attr = &rdev->dev_attr; + enum ib_qp_state curr_qp_state, new_qp_state; + int rc, entries; + int status; + union ib_gid sgid; + struct ib_gid_attr sgid_attr; + u8 nw_type; + + qp->qplib_qp.modify_flags = 0; + if (qp_attr_mask & IB_QP_STATE) { + curr_qp_state = __to_ib_qp_state(qp->qplib_qp.cur_qp_state); + new_qp_state = qp_attr->qp_state; + if (!ib_modify_qp_is_ok(curr_qp_state, new_qp_state, + ib_qp->qp_type, qp_attr_mask, + IB_LINK_LAYER_ETHERNET)) { + dev_err(rdev_to_dev(rdev), + "Invalid attribute mask: %#x specified ", + qp_attr_mask); + dev_err(rdev_to_dev(rdev), + "for qpn: %#x type: %#x", + ib_qp->qp_num, ib_qp->qp_type); + dev_err(rdev_to_dev(rdev), + "curr_qp_state=0x%x, new_qp_state=0x%x\n", + curr_qp_state, new_qp_state); + return -EINVAL; + } + qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_STATE; + qp->qplib_qp.state = __from_ib_qp_state(qp_attr->qp_state); + } + if (qp_attr_mask & IB_QP_EN_SQD_ASYNC_NOTIFY) { + qp->qplib_qp.modify_flags |= + CMDQ_MODIFY_QP_MODIFY_MASK_EN_SQD_ASYNC_NOTIFY; + qp->qplib_qp.en_sqd_async_notify = true; + } + if (qp_attr_mask & IB_QP_ACCESS_FLAGS) { + qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_ACCESS; + qp->qplib_qp.access = + __from_ib_access_flags(qp_attr->qp_access_flags); + /* LOCAL_WRITE access must be set to allow RC receive */ + qp->qplib_qp.access |= BNXT_QPLIB_ACCESS_LOCAL_WRITE; + } + if (qp_attr_mask & IB_QP_PKEY_INDEX) { + qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_PKEY; + qp->qplib_qp.pkey_index = qp_attr->pkey_index; + } + if (qp_attr_mask & IB_QP_QKEY) { + qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_QKEY; + qp->qplib_qp.qkey = qp_attr->qkey; + } + if (qp_attr_mask & IB_QP_AV) { + qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_DGID | + CMDQ_MODIFY_QP_MODIFY_MASK_FLOW_LABEL | + CMDQ_MODIFY_QP_MODIFY_MASK_SGID_INDEX | + CMDQ_MODIFY_QP_MODIFY_MASK_HOP_LIMIT | + CMDQ_MODIFY_QP_MODIFY_MASK_TRAFFIC_CLASS | + CMDQ_MODIFY_QP_MODIFY_MASK_DEST_MAC | + CMDQ_MODIFY_QP_MODIFY_MASK_VLAN_ID; + memcpy(qp->qplib_qp.ah.dgid.data, qp_attr->ah_attr.grh.dgid.raw, + sizeof(qp->qplib_qp.ah.dgid.data)); + qp->qplib_qp.ah.flow_label = qp_attr->ah_attr.grh.flow_label; + /* If RoCE V2 is enabled, stack will have two entries for + * each GID entry. Avoiding this duplicte entry in HW. Dividing + * the GID index by 2 for RoCE V2 + */ + qp->qplib_qp.ah.sgid_index = + qp_attr->ah_attr.grh.sgid_index / 2; + qp->qplib_qp.ah.host_sgid_index = + qp_attr->ah_attr.grh.sgid_index; + qp->qplib_qp.ah.hop_limit = qp_attr->ah_attr.grh.hop_limit; + qp->qplib_qp.ah.traffic_class = + qp_attr->ah_attr.grh.traffic_class; + qp->qplib_qp.ah.sl = qp_attr->ah_attr.sl; + ether_addr_copy(qp->qplib_qp.ah.dmac, qp_attr->ah_attr.dmac); + + status = ib_get_cached_gid(&rdev->ibdev, 1, + qp_attr->ah_attr.grh.sgid_index, + &sgid, &sgid_attr); + if (!status && sgid_attr.ndev) { + memcpy(qp->qplib_qp.smac, sgid_attr.ndev->dev_addr, + ETH_ALEN); + dev_put(sgid_attr.ndev); + nw_type = ib_gid_to_network_type(sgid_attr.gid_type, + &sgid); + switch (nw_type) { + case RDMA_NETWORK_IPV4: + qp->qplib_qp.nw_type = + CMDQ_MODIFY_QP_NETWORK_TYPE_ROCEV2_IPV4; + break; + case RDMA_NETWORK_IPV6: + qp->qplib_qp.nw_type = + CMDQ_MODIFY_QP_NETWORK_TYPE_ROCEV2_IPV6; + break; + default: + qp->qplib_qp.nw_type = + CMDQ_MODIFY_QP_NETWORK_TYPE_ROCEV1; + break; + } + } + } + + if (qp_attr_mask & IB_QP_PATH_MTU) { + qp->qplib_qp.modify_flags |= + CMDQ_MODIFY_QP_MODIFY_MASK_PATH_MTU; + qp->qplib_qp.path_mtu = __from_ib_mtu(qp_attr->path_mtu); + } else if (qp_attr->qp_state == IB_QPS_RTR) { + qp->qplib_qp.modify_flags |= + CMDQ_MODIFY_QP_MODIFY_MASK_PATH_MTU; + qp->qplib_qp.path_mtu = + __from_ib_mtu(iboe_get_mtu(rdev->netdev->mtu)); + } + + if (qp_attr_mask & IB_QP_TIMEOUT) { + qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_TIMEOUT; + qp->qplib_qp.timeout = qp_attr->timeout; + } + if (qp_attr_mask & IB_QP_RETRY_CNT) { + qp->qplib_qp.modify_flags |= + CMDQ_MODIFY_QP_MODIFY_MASK_RETRY_CNT; + qp->qplib_qp.retry_cnt = qp_attr->retry_cnt; + } + if (qp_attr_mask & IB_QP_RNR_RETRY) { + qp->qplib_qp.modify_flags |= + CMDQ_MODIFY_QP_MODIFY_MASK_RNR_RETRY; + qp->qplib_qp.rnr_retry = qp_attr->rnr_retry; + } + if (qp_attr_mask & IB_QP_MIN_RNR_TIMER) { + qp->qplib_qp.modify_flags |= + CMDQ_MODIFY_QP_MODIFY_MASK_MIN_RNR_TIMER; + qp->qplib_qp.min_rnr_timer = qp_attr->min_rnr_timer; + } + if (qp_attr_mask & IB_QP_RQ_PSN) { + qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_RQ_PSN; + qp->qplib_qp.rq.psn = qp_attr->rq_psn; + } + if (qp_attr_mask & IB_QP_MAX_QP_RD_ATOMIC) { + qp->qplib_qp.modify_flags |= + CMDQ_MODIFY_QP_MODIFY_MASK_MAX_RD_ATOMIC; + qp->qplib_qp.max_rd_atomic = qp_attr->max_rd_atomic; + } + if (qp_attr_mask & IB_QP_SQ_PSN) { + qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_SQ_PSN; + qp->qplib_qp.sq.psn = qp_attr->sq_psn; + } + if (qp_attr_mask & IB_QP_MAX_DEST_RD_ATOMIC) { + qp->qplib_qp.modify_flags |= + CMDQ_MODIFY_QP_MODIFY_MASK_MAX_DEST_RD_ATOMIC; + qp->qplib_qp.max_dest_rd_atomic = qp_attr->max_dest_rd_atomic; + } + if (qp_attr_mask & IB_QP_CAP) { + qp->qplib_qp.modify_flags |= + CMDQ_MODIFY_QP_MODIFY_MASK_SQ_SIZE | + CMDQ_MODIFY_QP_MODIFY_MASK_RQ_SIZE | + CMDQ_MODIFY_QP_MODIFY_MASK_SQ_SGE | + CMDQ_MODIFY_QP_MODIFY_MASK_RQ_SGE | + CMDQ_MODIFY_QP_MODIFY_MASK_MAX_INLINE_DATA; + if ((qp_attr->cap.max_send_wr >= dev_attr->max_qp_wqes) || + (qp_attr->cap.max_recv_wr >= dev_attr->max_qp_wqes) || + (qp_attr->cap.max_send_sge >= dev_attr->max_qp_sges) || + (qp_attr->cap.max_recv_sge >= dev_attr->max_qp_sges) || + (qp_attr->cap.max_inline_data >= + dev_attr->max_inline_data)) { + dev_err(rdev_to_dev(rdev), + "Create QP failed - max exceeded"); + return -EINVAL; + } + entries = roundup_pow_of_two(qp_attr->cap.max_send_wr); + qp->qplib_qp.sq.max_wqe = min_t(u32, entries, + dev_attr->max_qp_wqes + 1); + qp->qplib_qp.sq.max_sge = qp_attr->cap.max_send_sge; + if (qp->qplib_qp.rq.max_wqe) { + entries = roundup_pow_of_two(qp_attr->cap.max_recv_wr); + qp->qplib_qp.rq.max_wqe = + min_t(u32, entries, dev_attr->max_qp_wqes + 1); + qp->qplib_qp.rq.max_sge = qp_attr->cap.max_recv_sge; + } else { + /* SRQ was used prior, just ignore the RQ caps */ + } + } + if (qp_attr_mask & IB_QP_DEST_QPN) { + qp->qplib_qp.modify_flags |= + CMDQ_MODIFY_QP_MODIFY_MASK_DEST_QP_ID; + qp->qplib_qp.dest_qpn = qp_attr->dest_qp_num; + } + rc = bnxt_qplib_modify_qp(&rdev->qplib_res, &qp->qplib_qp); + if (rc) { + dev_err(rdev_to_dev(rdev), "Failed to modify HW QP"); + return rc; + } + if (ib_qp->qp_type == IB_QPT_GSI && rdev->qp1_sqp) + rc = bnxt_re_modify_shadow_qp(rdev, qp, qp_attr_mask); + return rc; +} + +int bnxt_re_query_qp(struct ib_qp *ib_qp, struct ib_qp_attr *qp_attr, + int qp_attr_mask, struct ib_qp_init_attr *qp_init_attr) +{ + struct bnxt_re_qp *qp = container_of(ib_qp, struct bnxt_re_qp, ib_qp); + struct bnxt_re_dev *rdev = qp->rdev; + struct bnxt_qplib_qp qplib_qp; + int rc; + + memset(&qplib_qp, 0, sizeof(struct bnxt_qplib_qp)); + qplib_qp.id = qp->qplib_qp.id; + qplib_qp.ah.host_sgid_index = qp->qplib_qp.ah.host_sgid_index; + + rc = bnxt_qplib_query_qp(&rdev->qplib_res, &qplib_qp); + if (rc) { + dev_err(rdev_to_dev(rdev), "Failed to query HW QP"); + return rc; + } + qp_attr->qp_state = __to_ib_qp_state(qplib_qp.state); + qp_attr->en_sqd_async_notify = qplib_qp.en_sqd_async_notify ? 1 : 0; + qp_attr->qp_access_flags = __to_ib_access_flags(qplib_qp.access); + qp_attr->pkey_index = qplib_qp.pkey_index; + qp_attr->qkey = qplib_qp.qkey; + memcpy(qp_attr->ah_attr.grh.dgid.raw, qplib_qp.ah.dgid.data, + sizeof(qplib_qp.ah.dgid.data)); + qp_attr->ah_attr.grh.flow_label = qplib_qp.ah.flow_label; + qp_attr->ah_attr.grh.sgid_index = qplib_qp.ah.host_sgid_index; + qp_attr->ah_attr.grh.hop_limit = qplib_qp.ah.hop_limit; + qp_attr->ah_attr.grh.traffic_class = qplib_qp.ah.traffic_class; + qp_attr->ah_attr.sl = qplib_qp.ah.sl; + ether_addr_copy(qp_attr->ah_attr.dmac, qplib_qp.ah.dmac); + qp_attr->path_mtu = __to_ib_mtu(qplib_qp.path_mtu); + qp_attr->timeout = qplib_qp.timeout; + qp_attr->retry_cnt = qplib_qp.retry_cnt; + qp_attr->rnr_retry = qplib_qp.rnr_retry; + qp_attr->min_rnr_timer = qplib_qp.min_rnr_timer; + qp_attr->rq_psn = qplib_qp.rq.psn; + qp_attr->max_rd_atomic = qplib_qp.max_rd_atomic; + qp_attr->sq_psn = qplib_qp.sq.psn; + qp_attr->max_dest_rd_atomic = qplib_qp.max_dest_rd_atomic; + qp_init_attr->sq_sig_type = qplib_qp.sig_type ? IB_SIGNAL_ALL_WR : + IB_SIGNAL_REQ_WR; + qp_attr->dest_qp_num = qplib_qp.dest_qpn; + + qp_attr->cap.max_send_wr = qp->qplib_qp.sq.max_wqe; + qp_attr->cap.max_send_sge = qp->qplib_qp.sq.max_sge; + qp_attr->cap.max_recv_wr = qp->qplib_qp.rq.max_wqe; + qp_attr->cap.max_recv_sge = qp->qplib_qp.rq.max_sge; + qp_attr->cap.max_inline_data = qp->qplib_qp.max_inline_data; + qp_init_attr->cap = qp_attr->cap; + + return 0; +} + +/* Routine for sending QP1 packets for RoCE V1 an V2 + */ +static int bnxt_re_build_qp1_send_v2(struct bnxt_re_qp *qp, + struct ib_send_wr *wr, + struct bnxt_qplib_swqe *wqe, + int payload_size) +{ + struct ib_device *ibdev = &qp->rdev->ibdev; + struct bnxt_re_ah *ah = container_of(ud_wr(wr)->ah, struct bnxt_re_ah, + ib_ah); + struct bnxt_qplib_ah *qplib_ah = &ah->qplib_ah; + struct bnxt_qplib_sge sge; + union ib_gid sgid; + u8 nw_type; + u16 ether_type; + struct ib_gid_attr sgid_attr; + union ib_gid dgid; + bool is_eth = false; + bool is_vlan = false; + bool is_grh = false; + bool is_udp = false; + u8 ip_version = 0; + u16 vlan_id = 0xFFFF; + void *buf; + int i, rc = 0, size; + + memset(&qp->qp1_hdr, 0, sizeof(qp->qp1_hdr)); + + rc = ib_get_cached_gid(ibdev, 1, + qplib_ah->host_sgid_index, &sgid, + &sgid_attr); + if (rc) { + dev_err(rdev_to_dev(qp->rdev), + "Failed to query gid at index %d", + qplib_ah->host_sgid_index); + return rc; + } + if (sgid_attr.ndev) { + if (is_vlan_dev(sgid_attr.ndev)) + vlan_id = vlan_dev_vlan_id(sgid_attr.ndev); + dev_put(sgid_attr.ndev); + } + /* Get network header type for this GID */ + nw_type = ib_gid_to_network_type(sgid_attr.gid_type, &sgid); + switch (nw_type) { + case RDMA_NETWORK_IPV4: + nw_type = BNXT_RE_ROCEV2_IPV4_PACKET; + break; + case RDMA_NETWORK_IPV6: + nw_type = BNXT_RE_ROCEV2_IPV6_PACKET; + break; + default: + nw_type = BNXT_RE_ROCE_V1_PACKET; + break; + } + memcpy(&dgid.raw, &qplib_ah->dgid, 16); + is_udp = sgid_attr.gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP; + if (is_udp) { + if (ipv6_addr_v4mapped((struct in6_addr *)&sgid)) { + ip_version = 4; + ether_type = ETH_P_IP; + } else { + ip_version = 6; + ether_type = ETH_P_IPV6; + } + is_grh = false; + } else { + ether_type = ETH_P_IBOE; + is_grh = true; + } + + is_eth = true; + is_vlan = (vlan_id && (vlan_id < 0x1000)) ? true : false; + + ib_ud_header_init(payload_size, !is_eth, is_eth, is_vlan, is_grh, + ip_version, is_udp, 0, &qp->qp1_hdr); + + /* ETH */ + ether_addr_copy(qp->qp1_hdr.eth.dmac_h, ah->qplib_ah.dmac); + ether_addr_copy(qp->qp1_hdr.eth.smac_h, qp->qplib_qp.smac); + + /* For vlan, check the sgid for vlan existence */ + + if (!is_vlan) { + qp->qp1_hdr.eth.type = cpu_to_be16(ether_type); + } else { + qp->qp1_hdr.vlan.type = cpu_to_be16(ether_type); + qp->qp1_hdr.vlan.tag = cpu_to_be16(vlan_id); + } + + if (is_grh || (ip_version == 6)) { + memcpy(qp->qp1_hdr.grh.source_gid.raw, sgid.raw, sizeof(sgid)); + memcpy(qp->qp1_hdr.grh.destination_gid.raw, qplib_ah->dgid.data, + sizeof(sgid)); + qp->qp1_hdr.grh.hop_limit = qplib_ah->hop_limit; + } + + if (ip_version == 4) { + qp->qp1_hdr.ip4.tos = 0; + qp->qp1_hdr.ip4.id = 0; + qp->qp1_hdr.ip4.frag_off = htons(IP_DF); + qp->qp1_hdr.ip4.ttl = qplib_ah->hop_limit; + + memcpy(&qp->qp1_hdr.ip4.saddr, sgid.raw + 12, 4); + memcpy(&qp->qp1_hdr.ip4.daddr, qplib_ah->dgid.data + 12, 4); + qp->qp1_hdr.ip4.check = ib_ud_ip4_csum(&qp->qp1_hdr); + } + + if (is_udp) { + qp->qp1_hdr.udp.dport = htons(ROCE_V2_UDP_DPORT); + qp->qp1_hdr.udp.sport = htons(0x8CD1); + qp->qp1_hdr.udp.csum = 0; + } + + /* BTH */ + if (wr->opcode == IB_WR_SEND_WITH_IMM) { + qp->qp1_hdr.bth.opcode = IB_OPCODE_UD_SEND_ONLY_WITH_IMMEDIATE; + qp->qp1_hdr.immediate_present = 1; + } else { + qp->qp1_hdr.bth.opcode = IB_OPCODE_UD_SEND_ONLY; + } + if (wr->send_flags & IB_SEND_SOLICITED) + qp->qp1_hdr.bth.solicited_event = 1; + /* pad_count */ + qp->qp1_hdr.bth.pad_count = (4 - payload_size) & 3; + + /* P_key for QP1 is for all members */ + qp->qp1_hdr.bth.pkey = cpu_to_be16(0xFFFF); + qp->qp1_hdr.bth.destination_qpn = IB_QP1; + qp->qp1_hdr.bth.ack_req = 0; + qp->send_psn++; + qp->send_psn &= BTH_PSN_MASK; + qp->qp1_hdr.bth.psn = cpu_to_be32(qp->send_psn); + /* DETH */ + /* Use the priviledged Q_Key for QP1 */ + qp->qp1_hdr.deth.qkey = cpu_to_be32(IB_QP1_QKEY); + qp->qp1_hdr.deth.source_qpn = IB_QP1; + + /* Pack the QP1 to the transmit buffer */ + buf = bnxt_qplib_get_qp1_sq_buf(&qp->qplib_qp, &sge); + if (buf) { + size = ib_ud_header_pack(&qp->qp1_hdr, buf); + for (i = wqe->num_sge; i; i--) { + wqe->sg_list[i].addr = wqe->sg_list[i - 1].addr; + wqe->sg_list[i].lkey = wqe->sg_list[i - 1].lkey; + wqe->sg_list[i].size = wqe->sg_list[i - 1].size; + } + + /* + * Max Header buf size for IPV6 RoCE V2 is 86, + * which is same as the QP1 SQ header buffer. + * Header buf size for IPV4 RoCE V2 can be 66. + * ETH(14) + VLAN(4)+ IP(20) + UDP (8) + BTH(20). + * Subtract 20 bytes from QP1 SQ header buf size + */ + if (is_udp && ip_version == 4) + sge.size -= 20; + /* + * Max Header buf size for RoCE V1 is 78. + * ETH(14) + VLAN(4) + GRH(40) + BTH(20). + * Subtract 8 bytes from QP1 SQ header buf size + */ + if (!is_udp) + sge.size -= 8; + + /* Subtract 4 bytes for non vlan packets */ + if (!is_vlan) + sge.size -= 4; + + wqe->sg_list[0].addr = sge.addr; + wqe->sg_list[0].lkey = sge.lkey; + wqe->sg_list[0].size = sge.size; + wqe->num_sge++; + + } else { + dev_err(rdev_to_dev(qp->rdev), "QP1 buffer is empty!"); + rc = -ENOMEM; + } + return rc; +} + +/* For the MAD layer, it only provides the recv SGE the size of + * ib_grh + MAD datagram. No Ethernet headers, Ethertype, BTH, DETH, + * nor RoCE iCRC. The Cu+ solution must provide buffer for the entire + * receive packet (334 bytes) with no VLAN and then copy the GRH + * and the MAD datagram out to the provided SGE. + */ +static int bnxt_re_build_qp1_shadow_qp_recv(struct bnxt_re_qp *qp, + struct ib_recv_wr *wr, + struct bnxt_qplib_swqe *wqe, + int payload_size) +{ + struct bnxt_qplib_sge ref, sge; + u32 rq_prod_index; + struct bnxt_re_sqp_entries *sqp_entry; + + rq_prod_index = bnxt_qplib_get_rq_prod_index(&qp->qplib_qp); + + if (!bnxt_qplib_get_qp1_rq_buf(&qp->qplib_qp, &sge)) + return -ENOMEM; + + /* Create 1 SGE to receive the entire + * ethernet packet + */ + /* Save the reference from ULP */ + ref.addr = wqe->sg_list[0].addr; + ref.lkey = wqe->sg_list[0].lkey; + ref.size = wqe->sg_list[0].size; + + sqp_entry = &qp->rdev->sqp_tbl[rq_prod_index]; + + /* SGE 1 */ + wqe->sg_list[0].addr = sge.addr; + wqe->sg_list[0].lkey = sge.lkey; + wqe->sg_list[0].size = BNXT_QPLIB_MAX_QP1_RQ_HDR_SIZE_V2; + sge.size -= wqe->sg_list[0].size; + + sqp_entry->sge.addr = ref.addr; + sqp_entry->sge.lkey = ref.lkey; + sqp_entry->sge.size = ref.size; + /* Store the wrid for reporting completion */ + sqp_entry->wrid = wqe->wr_id; + /* change the wqe->wrid to table index */ + wqe->wr_id = rq_prod_index; + return 0; +} + +static int is_ud_qp(struct bnxt_re_qp *qp) +{ + return qp->qplib_qp.type == CMDQ_CREATE_QP_TYPE_UD; +} + +static int bnxt_re_build_send_wqe(struct bnxt_re_qp *qp, + struct ib_send_wr *wr, + struct bnxt_qplib_swqe *wqe) +{ + struct bnxt_re_ah *ah = NULL; + + if (is_ud_qp(qp)) { + ah = container_of(ud_wr(wr)->ah, struct bnxt_re_ah, ib_ah); + wqe->send.q_key = ud_wr(wr)->remote_qkey; + wqe->send.dst_qp = ud_wr(wr)->remote_qpn; + wqe->send.avid = ah->qplib_ah.id; + } + switch (wr->opcode) { + case IB_WR_SEND: + wqe->type = BNXT_QPLIB_SWQE_TYPE_SEND; + break; + case IB_WR_SEND_WITH_IMM: + wqe->type = BNXT_QPLIB_SWQE_TYPE_SEND_WITH_IMM; + wqe->send.imm_data = wr->ex.imm_data; + break; + case IB_WR_SEND_WITH_INV: + wqe->type = BNXT_QPLIB_SWQE_TYPE_SEND_WITH_INV; + wqe->send.inv_key = wr->ex.invalidate_rkey; + break; + default: + return -EINVAL; + } + if (wr->send_flags & IB_SEND_SIGNALED) + wqe->flags |= BNXT_QPLIB_SWQE_FLAGS_SIGNAL_COMP; + if (wr->send_flags & IB_SEND_FENCE) + wqe->flags |= BNXT_QPLIB_SWQE_FLAGS_UC_FENCE; + if (wr->send_flags & IB_SEND_SOLICITED) + wqe->flags |= BNXT_QPLIB_SWQE_FLAGS_SOLICIT_EVENT; + if (wr->send_flags & IB_SEND_INLINE) + wqe->flags |= BNXT_QPLIB_SWQE_FLAGS_INLINE; + + return 0; +} + +static int bnxt_re_build_rdma_wqe(struct ib_send_wr *wr, + struct bnxt_qplib_swqe *wqe) +{ + switch (wr->opcode) { + case IB_WR_RDMA_WRITE: + wqe->type = BNXT_QPLIB_SWQE_TYPE_RDMA_WRITE; + break; + case IB_WR_RDMA_WRITE_WITH_IMM: + wqe->type = BNXT_QPLIB_SWQE_TYPE_RDMA_WRITE_WITH_IMM; + wqe->rdma.imm_data = wr->ex.imm_data; + break; + case IB_WR_RDMA_READ: + wqe->type = BNXT_QPLIB_SWQE_TYPE_RDMA_READ; + wqe->rdma.inv_key = wr->ex.invalidate_rkey; + break; + default: + return -EINVAL; + } + wqe->rdma.remote_va = rdma_wr(wr)->remote_addr; + wqe->rdma.r_key = rdma_wr(wr)->rkey; + if (wr->send_flags & IB_SEND_SIGNALED) + wqe->flags |= BNXT_QPLIB_SWQE_FLAGS_SIGNAL_COMP; + if (wr->send_flags & IB_SEND_FENCE) + wqe->flags |= BNXT_QPLIB_SWQE_FLAGS_UC_FENCE; + if (wr->send_flags & IB_SEND_SOLICITED) + wqe->flags |= BNXT_QPLIB_SWQE_FLAGS_SOLICIT_EVENT; + if (wr->send_flags & IB_SEND_INLINE) + wqe->flags |= BNXT_QPLIB_SWQE_FLAGS_INLINE; + + return 0; +} + +static int bnxt_re_build_atomic_wqe(struct ib_send_wr *wr, + struct bnxt_qplib_swqe *wqe) +{ + switch (wr->opcode) { + case IB_WR_ATOMIC_CMP_AND_SWP: + wqe->type = BNXT_QPLIB_SWQE_TYPE_ATOMIC_CMP_AND_SWP; + wqe->atomic.swap_data = atomic_wr(wr)->swap; + break; + case IB_WR_ATOMIC_FETCH_AND_ADD: + wqe->type = BNXT_QPLIB_SWQE_TYPE_ATOMIC_FETCH_AND_ADD; + wqe->atomic.cmp_data = atomic_wr(wr)->compare_add; + break; + default: + return -EINVAL; + } + wqe->atomic.remote_va = atomic_wr(wr)->remote_addr; + wqe->atomic.r_key = atomic_wr(wr)->rkey; + if (wr->send_flags & IB_SEND_SIGNALED) + wqe->flags |= BNXT_QPLIB_SWQE_FLAGS_SIGNAL_COMP; + if (wr->send_flags & IB_SEND_FENCE) + wqe->flags |= BNXT_QPLIB_SWQE_FLAGS_UC_FENCE; + if (wr->send_flags & IB_SEND_SOLICITED) + wqe->flags |= BNXT_QPLIB_SWQE_FLAGS_SOLICIT_EVENT; + return 0; +} + +static int bnxt_re_build_inv_wqe(struct ib_send_wr *wr, + struct bnxt_qplib_swqe *wqe) +{ + wqe->type = BNXT_QPLIB_SWQE_TYPE_LOCAL_INV; + wqe->local_inv.inv_l_key = wr->ex.invalidate_rkey; + + if (wr->send_flags & IB_SEND_SIGNALED) + wqe->flags |= BNXT_QPLIB_SWQE_FLAGS_SIGNAL_COMP; + if (wr->send_flags & IB_SEND_FENCE) + wqe->flags |= BNXT_QPLIB_SWQE_FLAGS_UC_FENCE; + if (wr->send_flags & IB_SEND_SOLICITED) + wqe->flags |= BNXT_QPLIB_SWQE_FLAGS_SOLICIT_EVENT; + + return 0; +} + +static int bnxt_re_build_reg_wqe(struct ib_reg_wr *wr, + struct bnxt_qplib_swqe *wqe) +{ + struct bnxt_re_mr *mr = container_of(wr->mr, struct bnxt_re_mr, ib_mr); + struct bnxt_qplib_frpl *qplib_frpl = &mr->qplib_frpl; + int access = wr->access; + + wqe->frmr.pbl_ptr = (__le64 *)qplib_frpl->hwq.pbl_ptr[0]; + wqe->frmr.pbl_dma_ptr = qplib_frpl->hwq.pbl_dma_ptr[0]; + wqe->frmr.page_list = mr->pages; + wqe->frmr.page_list_len = mr->npages; + wqe->frmr.levels = qplib_frpl->hwq.level + 1; + wqe->type = BNXT_QPLIB_SWQE_TYPE_REG_MR; + + if (wr->wr.send_flags & IB_SEND_FENCE) + wqe->flags |= BNXT_QPLIB_SWQE_FLAGS_UC_FENCE; + if (wr->wr.send_flags & IB_SEND_SIGNALED) + wqe->flags |= BNXT_QPLIB_SWQE_FLAGS_SIGNAL_COMP; + + if (access & IB_ACCESS_LOCAL_WRITE) + wqe->frmr.access_cntl |= SQ_FR_PMR_ACCESS_CNTL_LOCAL_WRITE; + if (access & IB_ACCESS_REMOTE_READ) + wqe->frmr.access_cntl |= SQ_FR_PMR_ACCESS_CNTL_REMOTE_READ; + if (access & IB_ACCESS_REMOTE_WRITE) + wqe->frmr.access_cntl |= SQ_FR_PMR_ACCESS_CNTL_REMOTE_WRITE; + if (access & IB_ACCESS_REMOTE_ATOMIC) + wqe->frmr.access_cntl |= SQ_FR_PMR_ACCESS_CNTL_REMOTE_ATOMIC; + if (access & IB_ACCESS_MW_BIND) + wqe->frmr.access_cntl |= SQ_FR_PMR_ACCESS_CNTL_WINDOW_BIND; + + wqe->frmr.l_key = wr->key; + wqe->frmr.length = wr->mr->length; + wqe->frmr.pbl_pg_sz_log = (wr->mr->page_size >> PAGE_SHIFT_4K) - 1; + wqe->frmr.va = wr->mr->iova; + return 0; +} + +static int bnxt_re_copy_inline_data(struct bnxt_re_dev *rdev, + struct ib_send_wr *wr, + struct bnxt_qplib_swqe *wqe) +{ + /* Copy the inline data to the data field */ + u8 *in_data; + u32 i, sge_len; + void *sge_addr; + + in_data = wqe->inline_data; + for (i = 0; i < wr->num_sge; i++) { + sge_addr = (void *)(unsigned long) + wr->sg_list[i].addr; + sge_len = wr->sg_list[i].length; + + if ((sge_len + wqe->inline_len) > + BNXT_QPLIB_SWQE_MAX_INLINE_LENGTH) { + dev_err(rdev_to_dev(rdev), + "Inline data size requested > supported value"); + return -EINVAL; + } + sge_len = wr->sg_list[i].length; + + memcpy(in_data, sge_addr, sge_len); + in_data += wr->sg_list[i].length; + wqe->inline_len += wr->sg_list[i].length; + } + return wqe->inline_len; +} + +static int bnxt_re_copy_wr_payload(struct bnxt_re_dev *rdev, + struct ib_send_wr *wr, + struct bnxt_qplib_swqe *wqe) +{ + int payload_sz = 0; + + if (wr->send_flags & IB_SEND_INLINE) + payload_sz = bnxt_re_copy_inline_data(rdev, wr, wqe); + else + payload_sz = bnxt_re_build_sgl(wr->sg_list, wqe->sg_list, + wqe->num_sge); + + return payload_sz; +} + +static int bnxt_re_post_send_shadow_qp(struct bnxt_re_dev *rdev, + struct bnxt_re_qp *qp, + struct ib_send_wr *wr) +{ + struct bnxt_qplib_swqe wqe; + int rc = 0, payload_sz = 0; + unsigned long flags; + + spin_lock_irqsave(&qp->sq_lock, flags); + memset(&wqe, 0, sizeof(wqe)); + while (wr) { + /* House keeping */ + memset(&wqe, 0, sizeof(wqe)); + + /* Common */ + wqe.num_sge = wr->num_sge; + if (wr->num_sge > qp->qplib_qp.sq.max_sge) { + dev_err(rdev_to_dev(rdev), + "Limit exceeded for Send SGEs"); + rc = -EINVAL; + goto bad; + } + + payload_sz = bnxt_re_copy_wr_payload(qp->rdev, wr, &wqe); + if (payload_sz < 0) { + rc = -EINVAL; + goto bad; + } + wqe.wr_id = wr->wr_id; + + wqe.type = BNXT_QPLIB_SWQE_TYPE_SEND; + + rc = bnxt_re_build_send_wqe(qp, wr, &wqe); + if (!rc) + rc = bnxt_qplib_post_send(&qp->qplib_qp, &wqe); +bad: + if (rc) { + dev_err(rdev_to_dev(rdev), + "Post send failed opcode = %#x rc = %d", + wr->opcode, rc); + break; + } + wr = wr->next; + } + bnxt_qplib_post_send_db(&qp->qplib_qp); + spin_unlock_irqrestore(&qp->sq_lock, flags); + return rc; +} + +int bnxt_re_post_send(struct ib_qp *ib_qp, struct ib_send_wr *wr, + struct ib_send_wr **bad_wr) +{ + struct bnxt_re_qp *qp = container_of(ib_qp, struct bnxt_re_qp, ib_qp); + struct bnxt_qplib_swqe wqe; + int rc = 0, payload_sz = 0; + unsigned long flags; + + spin_lock_irqsave(&qp->sq_lock, flags); + while (wr) { + /* House keeping */ + memset(&wqe, 0, sizeof(wqe)); + + /* Common */ + wqe.num_sge = wr->num_sge; + if (wr->num_sge > qp->qplib_qp.sq.max_sge) { + dev_err(rdev_to_dev(qp->rdev), + "Limit exceeded for Send SGEs"); + rc = -EINVAL; + goto bad; + } + + payload_sz = bnxt_re_copy_wr_payload(qp->rdev, wr, &wqe); + if (payload_sz < 0) { + rc = -EINVAL; + goto bad; + } + wqe.wr_id = wr->wr_id; + + switch (wr->opcode) { + case IB_WR_SEND: + case IB_WR_SEND_WITH_IMM: + if (ib_qp->qp_type == IB_QPT_GSI) { + rc = bnxt_re_build_qp1_send_v2(qp, wr, &wqe, + payload_sz); + if (rc) + goto bad; + wqe.rawqp1.lflags |= + SQ_SEND_RAWETH_QP1_LFLAGS_ROCE_CRC; + } + switch (wr->send_flags) { + case IB_SEND_IP_CSUM: + wqe.rawqp1.lflags |= + SQ_SEND_RAWETH_QP1_LFLAGS_IP_CHKSUM; + break; + default: + break; + } + /* Fall thru to build the wqe */ + case IB_WR_SEND_WITH_INV: + rc = bnxt_re_build_send_wqe(qp, wr, &wqe); + break; + case IB_WR_RDMA_WRITE: + case IB_WR_RDMA_WRITE_WITH_IMM: + case IB_WR_RDMA_READ: + rc = bnxt_re_build_rdma_wqe(wr, &wqe); + break; + case IB_WR_ATOMIC_CMP_AND_SWP: + case IB_WR_ATOMIC_FETCH_AND_ADD: + rc = bnxt_re_build_atomic_wqe(wr, &wqe); + break; + case IB_WR_RDMA_READ_WITH_INV: + dev_err(rdev_to_dev(qp->rdev), + "RDMA Read with Invalidate is not supported"); + rc = -EINVAL; + goto bad; + case IB_WR_LOCAL_INV: + rc = bnxt_re_build_inv_wqe(wr, &wqe); + break; + case IB_WR_REG_MR: + rc = bnxt_re_build_reg_wqe(reg_wr(wr), &wqe); + break; + default: + /* Unsupported WRs */ + dev_err(rdev_to_dev(qp->rdev), + "WR (%#x) is not supported", wr->opcode); + rc = -EINVAL; + goto bad; + } + if (!rc) + rc = bnxt_qplib_post_send(&qp->qplib_qp, &wqe); +bad: + if (rc) { + dev_err(rdev_to_dev(qp->rdev), + "post_send failed op:%#x qps = %#x rc = %d\n", + wr->opcode, qp->qplib_qp.state, rc); + *bad_wr = wr; + break; + } + wr = wr->next; + } + bnxt_qplib_post_send_db(&qp->qplib_qp); + spin_unlock_irqrestore(&qp->sq_lock, flags); + + return rc; +} + +static int bnxt_re_post_recv_shadow_qp(struct bnxt_re_dev *rdev, + struct bnxt_re_qp *qp, + struct ib_recv_wr *wr) +{ + struct bnxt_qplib_swqe wqe; + int rc = 0, payload_sz = 0; + + memset(&wqe, 0, sizeof(wqe)); + while (wr) { + /* House keeping */ + memset(&wqe, 0, sizeof(wqe)); + + /* Common */ + wqe.num_sge = wr->num_sge; + if (wr->num_sge > qp->qplib_qp.rq.max_sge) { + dev_err(rdev_to_dev(rdev), + "Limit exceeded for Receive SGEs"); + rc = -EINVAL; + break; + } + payload_sz = bnxt_re_build_sgl(wr->sg_list, wqe.sg_list, + wr->num_sge); + wqe.wr_id = wr->wr_id; + wqe.type = BNXT_QPLIB_SWQE_TYPE_RECV; + + rc = bnxt_qplib_post_recv(&qp->qplib_qp, &wqe); + if (rc) + break; + + wr = wr->next; + } + if (!rc) + bnxt_qplib_post_recv_db(&qp->qplib_qp); + return rc; +} + +int bnxt_re_post_recv(struct ib_qp *ib_qp, struct ib_recv_wr *wr, + struct ib_recv_wr **bad_wr) +{ + struct bnxt_re_qp *qp = container_of(ib_qp, struct bnxt_re_qp, ib_qp); + struct bnxt_qplib_swqe wqe; + int rc = 0, payload_sz = 0; + + while (wr) { + /* House keeping */ + memset(&wqe, 0, sizeof(wqe)); + + /* Common */ + wqe.num_sge = wr->num_sge; + if (wr->num_sge > qp->qplib_qp.rq.max_sge) { + dev_err(rdev_to_dev(qp->rdev), + "Limit exceeded for Receive SGEs"); + rc = -EINVAL; + *bad_wr = wr; + break; + } + + payload_sz = bnxt_re_build_sgl(wr->sg_list, wqe.sg_list, + wr->num_sge); + wqe.wr_id = wr->wr_id; + wqe.type = BNXT_QPLIB_SWQE_TYPE_RECV; + + if (ib_qp->qp_type == IB_QPT_GSI) + rc = bnxt_re_build_qp1_shadow_qp_recv(qp, wr, &wqe, + payload_sz); + if (!rc) + rc = bnxt_qplib_post_recv(&qp->qplib_qp, &wqe); + if (rc) { + *bad_wr = wr; + break; + } + wr = wr->next; + } + bnxt_qplib_post_recv_db(&qp->qplib_qp); + return rc; +} + +/* Completion Queues */ +int bnxt_re_destroy_cq(struct ib_cq *ib_cq) +{ + struct bnxt_re_cq *cq = container_of(ib_cq, struct bnxt_re_cq, ib_cq); + struct bnxt_re_dev *rdev = cq->rdev; + int rc; + + rc = bnxt_qplib_destroy_cq(&rdev->qplib_res, &cq->qplib_cq); + if (rc) { + dev_err(rdev_to_dev(rdev), "Failed to destroy HW CQ"); + return rc; + } + if (cq->umem && !IS_ERR(cq->umem)) + ib_umem_release(cq->umem); + + if (cq) { + kfree(cq->cql); + kfree(cq); + } + atomic_dec(&rdev->cq_count); + rdev->nq.budget--; + return 0; +} + +struct ib_cq *bnxt_re_create_cq(struct ib_device *ibdev, + const struct ib_cq_init_attr *attr, + struct ib_ucontext *context, + struct ib_udata *udata) +{ + struct bnxt_re_dev *rdev = to_bnxt_re_dev(ibdev, ibdev); + struct bnxt_qplib_dev_attr *dev_attr = &rdev->dev_attr; + struct bnxt_re_cq *cq = NULL; + int rc, entries; + int cqe = attr->cqe; + + /* Validate CQ fields */ + if (cqe < 1 || cqe > dev_attr->max_cq_wqes) { + dev_err(rdev_to_dev(rdev), "Failed to create CQ -max exceeded"); + return ERR_PTR(-EINVAL); + } + cq = kzalloc(sizeof(*cq), GFP_KERNEL); + if (!cq) + return ERR_PTR(-ENOMEM); + + cq->rdev = rdev; + cq->qplib_cq.cq_handle = (u64)(unsigned long)(&cq->qplib_cq); + + entries = roundup_pow_of_two(cqe + 1); + if (entries > dev_attr->max_cq_wqes + 1) + entries = dev_attr->max_cq_wqes + 1; + + if (context) { + struct bnxt_re_cq_req req; + struct bnxt_re_ucontext *uctx = container_of + (context, + struct bnxt_re_ucontext, + ib_uctx); + if (ib_copy_from_udata(&req, udata, sizeof(req))) { + rc = -EFAULT; + goto fail; + } + + cq->umem = ib_umem_get(context, req.cq_va, + entries * sizeof(struct cq_base), + IB_ACCESS_LOCAL_WRITE, 1); + if (IS_ERR(cq->umem)) { + rc = PTR_ERR(cq->umem); + goto fail; + } + cq->qplib_cq.sghead = cq->umem->sg_head.sgl; + cq->qplib_cq.nmap = cq->umem->nmap; + cq->qplib_cq.dpi = uctx->dpi; + } else { + cq->max_cql = min_t(u32, entries, MAX_CQL_PER_POLL); + cq->cql = kcalloc(cq->max_cql, sizeof(struct bnxt_qplib_cqe), + GFP_KERNEL); + if (!cq->cql) { + rc = -ENOMEM; + goto fail; + } + + cq->qplib_cq.dpi = &rdev->dpi_privileged; + cq->qplib_cq.sghead = NULL; + cq->qplib_cq.nmap = 0; + } + cq->qplib_cq.max_wqe = entries; + cq->qplib_cq.cnq_hw_ring_id = rdev->nq.ring_id; + + rc = bnxt_qplib_create_cq(&rdev->qplib_res, &cq->qplib_cq); + if (rc) { + dev_err(rdev_to_dev(rdev), "Failed to create HW CQ"); + goto fail; + } + + cq->ib_cq.cqe = entries; + cq->cq_period = cq->qplib_cq.period; + rdev->nq.budget++; + + atomic_inc(&rdev->cq_count); + + if (context) { + struct bnxt_re_cq_resp resp; + + resp.cqid = cq->qplib_cq.id; + resp.tail = cq->qplib_cq.hwq.cons; + resp.phase = cq->qplib_cq.period; + resp.rsvd = 0; + rc = ib_copy_to_udata(udata, &resp, sizeof(resp)); + if (rc) { + dev_err(rdev_to_dev(rdev), "Failed to copy CQ udata"); + bnxt_qplib_destroy_cq(&rdev->qplib_res, &cq->qplib_cq); + goto c2fail; + } + } + + return &cq->ib_cq; + +c2fail: + if (context) + ib_umem_release(cq->umem); +fail: + kfree(cq->cql); + kfree(cq); + return ERR_PTR(rc); +} + +static u8 __req_to_ib_wc_status(u8 qstatus) +{ + switch (qstatus) { + case CQ_REQ_STATUS_OK: + return IB_WC_SUCCESS; + case CQ_REQ_STATUS_BAD_RESPONSE_ERR: + return IB_WC_BAD_RESP_ERR; + case CQ_REQ_STATUS_LOCAL_LENGTH_ERR: + return IB_WC_LOC_LEN_ERR; + case CQ_REQ_STATUS_LOCAL_QP_OPERATION_ERR: + return IB_WC_LOC_QP_OP_ERR; + case CQ_REQ_STATUS_LOCAL_PROTECTION_ERR: + return IB_WC_LOC_PROT_ERR; + case CQ_REQ_STATUS_MEMORY_MGT_OPERATION_ERR: + return IB_WC_GENERAL_ERR; + case CQ_REQ_STATUS_REMOTE_INVALID_REQUEST_ERR: + return IB_WC_REM_INV_REQ_ERR; + case CQ_REQ_STATUS_REMOTE_ACCESS_ERR: + return IB_WC_REM_ACCESS_ERR; + case CQ_REQ_STATUS_REMOTE_OPERATION_ERR: + return IB_WC_REM_OP_ERR; + case CQ_REQ_STATUS_RNR_NAK_RETRY_CNT_ERR: + return IB_WC_RNR_RETRY_EXC_ERR; + case CQ_REQ_STATUS_TRANSPORT_RETRY_CNT_ERR: + return IB_WC_RETRY_EXC_ERR; + case CQ_REQ_STATUS_WORK_REQUEST_FLUSHED_ERR: + return IB_WC_WR_FLUSH_ERR; + default: + return IB_WC_GENERAL_ERR; + } + return 0; +} + +static u8 __rawqp1_to_ib_wc_status(u8 qstatus) +{ + switch (qstatus) { + case CQ_RES_RAWETH_QP1_STATUS_OK: + return IB_WC_SUCCESS; + case CQ_RES_RAWETH_QP1_STATUS_LOCAL_ACCESS_ERROR: + return IB_WC_LOC_ACCESS_ERR; + case CQ_RES_RAWETH_QP1_STATUS_HW_LOCAL_LENGTH_ERR: + return IB_WC_LOC_LEN_ERR; + case CQ_RES_RAWETH_QP1_STATUS_LOCAL_PROTECTION_ERR: + return IB_WC_LOC_PROT_ERR; + case CQ_RES_RAWETH_QP1_STATUS_LOCAL_QP_OPERATION_ERR: + return IB_WC_LOC_QP_OP_ERR; + case CQ_RES_RAWETH_QP1_STATUS_MEMORY_MGT_OPERATION_ERR: + return IB_WC_GENERAL_ERR; + case CQ_RES_RAWETH_QP1_STATUS_WORK_REQUEST_FLUSHED_ERR: + return IB_WC_WR_FLUSH_ERR; + case CQ_RES_RAWETH_QP1_STATUS_HW_FLUSH_ERR: + return IB_WC_WR_FLUSH_ERR; + default: + return IB_WC_GENERAL_ERR; + } +} + +static u8 __rc_to_ib_wc_status(u8 qstatus) +{ + switch (qstatus) { + case CQ_RES_RC_STATUS_OK: + return IB_WC_SUCCESS; + case CQ_RES_RC_STATUS_LOCAL_ACCESS_ERROR: + return IB_WC_LOC_ACCESS_ERR; + case CQ_RES_RC_STATUS_LOCAL_LENGTH_ERR: + return IB_WC_LOC_LEN_ERR; + case CQ_RES_RC_STATUS_LOCAL_PROTECTION_ERR: + return IB_WC_LOC_PROT_ERR; + case CQ_RES_RC_STATUS_LOCAL_QP_OPERATION_ERR: + return IB_WC_LOC_QP_OP_ERR; + case CQ_RES_RC_STATUS_MEMORY_MGT_OPERATION_ERR: + return IB_WC_GENERAL_ERR; + case CQ_RES_RC_STATUS_REMOTE_INVALID_REQUEST_ERR: + return IB_WC_REM_INV_REQ_ERR; + case CQ_RES_RC_STATUS_WORK_REQUEST_FLUSHED_ERR: + return IB_WC_WR_FLUSH_ERR; + case CQ_RES_RC_STATUS_HW_FLUSH_ERR: + return IB_WC_WR_FLUSH_ERR; + default: + return IB_WC_GENERAL_ERR; + } +} + +static void bnxt_re_process_req_wc(struct ib_wc *wc, struct bnxt_qplib_cqe *cqe) +{ + switch (cqe->type) { + case BNXT_QPLIB_SWQE_TYPE_SEND: + wc->opcode = IB_WC_SEND; + break; + case BNXT_QPLIB_SWQE_TYPE_SEND_WITH_IMM: + wc->opcode = IB_WC_SEND; + wc->wc_flags |= IB_WC_WITH_IMM; + break; + case BNXT_QPLIB_SWQE_TYPE_SEND_WITH_INV: + wc->opcode = IB_WC_SEND; + wc->wc_flags |= IB_WC_WITH_INVALIDATE; + break; + case BNXT_QPLIB_SWQE_TYPE_RDMA_WRITE: + wc->opcode = IB_WC_RDMA_WRITE; + break; + case BNXT_QPLIB_SWQE_TYPE_RDMA_WRITE_WITH_IMM: + wc->opcode = IB_WC_RDMA_WRITE; + wc->wc_flags |= IB_WC_WITH_IMM; + break; + case BNXT_QPLIB_SWQE_TYPE_RDMA_READ: + wc->opcode = IB_WC_RDMA_READ; + break; + case BNXT_QPLIB_SWQE_TYPE_ATOMIC_CMP_AND_SWP: + wc->opcode = IB_WC_COMP_SWAP; + break; + case BNXT_QPLIB_SWQE_TYPE_ATOMIC_FETCH_AND_ADD: + wc->opcode = IB_WC_FETCH_ADD; + break; + case BNXT_QPLIB_SWQE_TYPE_LOCAL_INV: + wc->opcode = IB_WC_LOCAL_INV; + break; + case BNXT_QPLIB_SWQE_TYPE_REG_MR: + wc->opcode = IB_WC_REG_MR; + break; + default: + wc->opcode = IB_WC_SEND; + break; + } + + wc->status = __req_to_ib_wc_status(cqe->status); +} + +static int bnxt_re_check_packet_type(u16 raweth_qp1_flags, + u16 raweth_qp1_flags2) +{ + bool is_udp = false, is_ipv6 = false, is_ipv4 = false; + + /* raweth_qp1_flags Bit 9-6 indicates itype */ + if ((raweth_qp1_flags & CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_ITYPE_ROCE) + != CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_ITYPE_ROCE) + return -1; + + if (raweth_qp1_flags2 & + CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS2_IP_CS_CALC && + raweth_qp1_flags2 & + CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS2_L4_CS_CALC) { + is_udp = true; + /* raweth_qp1_flags2 Bit 8 indicates ip_type. 0-v4 1 - v6 */ + (raweth_qp1_flags2 & + CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS2_IP_TYPE) ? + (is_ipv6 = true) : (is_ipv4 = true); + return ((is_ipv6) ? + BNXT_RE_ROCEV2_IPV6_PACKET : + BNXT_RE_ROCEV2_IPV4_PACKET); + } else { + return BNXT_RE_ROCE_V1_PACKET; + } +} + +static int bnxt_re_to_ib_nw_type(int nw_type) +{ + u8 nw_hdr_type = 0xFF; + + switch (nw_type) { + case BNXT_RE_ROCE_V1_PACKET: + nw_hdr_type = RDMA_NETWORK_ROCE_V1; + break; + case BNXT_RE_ROCEV2_IPV4_PACKET: + nw_hdr_type = RDMA_NETWORK_IPV4; + break; + case BNXT_RE_ROCEV2_IPV6_PACKET: + nw_hdr_type = RDMA_NETWORK_IPV6; + break; + } + return nw_hdr_type; +} + +static bool bnxt_re_is_loopback_packet(struct bnxt_re_dev *rdev, + void *rq_hdr_buf) +{ + u8 *tmp_buf = NULL; + struct ethhdr *eth_hdr; + u16 eth_type; + bool rc = false; + + tmp_buf = (u8 *)rq_hdr_buf; + /* + * If dest mac is not same as I/F mac, this could be a + * loopback address or multicast address, check whether + * it is a loopback packet + */ + if (!ether_addr_equal(tmp_buf, rdev->netdev->dev_addr)) { + tmp_buf += 4; + /* Check the ether type */ + eth_hdr = (struct ethhdr *)tmp_buf; + eth_type = ntohs(eth_hdr->h_proto); + switch (eth_type) { + case ETH_P_IBOE: + rc = true; + break; + case ETH_P_IP: + case ETH_P_IPV6: { + u32 len; + struct udphdr *udp_hdr; + + len = (eth_type == ETH_P_IP ? sizeof(struct iphdr) : + sizeof(struct ipv6hdr)); + tmp_buf += sizeof(struct ethhdr) + len; + udp_hdr = (struct udphdr *)tmp_buf; + if (ntohs(udp_hdr->dest) == + ROCE_V2_UDP_DPORT) + rc = true; + break; + } + default: + break; + } + } + + return rc; +} + +static int bnxt_re_process_raw_qp_pkt_rx(struct bnxt_re_qp *qp1_qp, + struct bnxt_qplib_cqe *cqe) +{ + struct bnxt_re_dev *rdev = qp1_qp->rdev; + struct bnxt_re_sqp_entries *sqp_entry = NULL; + struct bnxt_re_qp *qp = rdev->qp1_sqp; + struct ib_send_wr *swr; + struct ib_ud_wr udwr; + struct ib_recv_wr rwr; + int pkt_type = 0; + u32 tbl_idx; + void *rq_hdr_buf; + dma_addr_t rq_hdr_buf_map; + dma_addr_t shrq_hdr_buf_map; + u32 offset = 0; + u32 skip_bytes = 0; + struct ib_sge s_sge[2]; + struct ib_sge r_sge[2]; + int rc; + + memset(&udwr, 0, sizeof(udwr)); + memset(&rwr, 0, sizeof(rwr)); + memset(&s_sge, 0, sizeof(s_sge)); + memset(&r_sge, 0, sizeof(r_sge)); + + swr = &udwr.wr; + tbl_idx = cqe->wr_id; + + rq_hdr_buf = qp1_qp->qplib_qp.rq_hdr_buf + + (tbl_idx * qp1_qp->qplib_qp.rq_hdr_buf_size); + rq_hdr_buf_map = bnxt_qplib_get_qp_buf_from_index(&qp1_qp->qplib_qp, + tbl_idx); + + /* Shadow QP header buffer */ + shrq_hdr_buf_map = bnxt_qplib_get_qp_buf_from_index(&qp->qplib_qp, + tbl_idx); + sqp_entry = &rdev->sqp_tbl[tbl_idx]; + + /* Store this cqe */ + memcpy(&sqp_entry->cqe, cqe, sizeof(struct bnxt_qplib_cqe)); + sqp_entry->qp1_qp = qp1_qp; + + /* Find packet type from the cqe */ + + pkt_type = bnxt_re_check_packet_type(cqe->raweth_qp1_flags, + cqe->raweth_qp1_flags2); + if (pkt_type < 0) { + dev_err(rdev_to_dev(rdev), "Invalid packet\n"); + return -EINVAL; + } + + /* Adjust the offset for the user buffer and post in the rq */ + + if (pkt_type == BNXT_RE_ROCEV2_IPV4_PACKET) + offset = 20; + + /* + * QP1 loopback packet has 4 bytes of internal header before + * ether header. Skip these four bytes. + */ + if (bnxt_re_is_loopback_packet(rdev, rq_hdr_buf)) + skip_bytes = 4; + + /* First send SGE . Skip the ether header*/ + s_sge[0].addr = rq_hdr_buf_map + BNXT_QPLIB_MAX_QP1_RQ_ETH_HDR_SIZE + + skip_bytes; + s_sge[0].lkey = 0xFFFFFFFF; + s_sge[0].length = offset ? BNXT_QPLIB_MAX_GRH_HDR_SIZE_IPV4 : + BNXT_QPLIB_MAX_GRH_HDR_SIZE_IPV6; + + /* Second Send SGE */ + s_sge[1].addr = s_sge[0].addr + s_sge[0].length + + BNXT_QPLIB_MAX_QP1_RQ_BDETH_HDR_SIZE; + if (pkt_type != BNXT_RE_ROCE_V1_PACKET) + s_sge[1].addr += 8; + s_sge[1].lkey = 0xFFFFFFFF; + s_sge[1].length = 256; + + /* First recv SGE */ + + r_sge[0].addr = shrq_hdr_buf_map; + r_sge[0].lkey = 0xFFFFFFFF; + r_sge[0].length = 40; + + r_sge[1].addr = sqp_entry->sge.addr + offset; + r_sge[1].lkey = sqp_entry->sge.lkey; + r_sge[1].length = BNXT_QPLIB_MAX_GRH_HDR_SIZE_IPV6 + 256 - offset; + + /* Create receive work request */ + rwr.num_sge = 2; + rwr.sg_list = r_sge; + rwr.wr_id = tbl_idx; + rwr.next = NULL; + + rc = bnxt_re_post_recv_shadow_qp(rdev, qp, &rwr); + if (rc) { + dev_err(rdev_to_dev(rdev), + "Failed to post Rx buffers to shadow QP"); + return -ENOMEM; + } + + swr->num_sge = 2; + swr->sg_list = s_sge; + swr->wr_id = tbl_idx; + swr->opcode = IB_WR_SEND; + swr->next = NULL; + + udwr.ah = &rdev->sqp_ah->ib_ah; + udwr.remote_qpn = rdev->qp1_sqp->qplib_qp.id; + udwr.remote_qkey = rdev->qp1_sqp->qplib_qp.qkey; + + /* post data received in the send queue */ + rc = bnxt_re_post_send_shadow_qp(rdev, qp, swr); + + return 0; +} + +static void bnxt_re_process_res_rawqp1_wc(struct ib_wc *wc, + struct bnxt_qplib_cqe *cqe) +{ + wc->opcode = IB_WC_RECV; + wc->status = __rawqp1_to_ib_wc_status(cqe->status); + wc->wc_flags |= IB_WC_GRH; +} + +static void bnxt_re_process_res_rc_wc(struct ib_wc *wc, + struct bnxt_qplib_cqe *cqe) +{ + wc->opcode = IB_WC_RECV; + wc->status = __rc_to_ib_wc_status(cqe->status); + + if (cqe->flags & CQ_RES_RC_FLAGS_IMM) + wc->wc_flags |= IB_WC_WITH_IMM; + if (cqe->flags & CQ_RES_RC_FLAGS_INV) + wc->wc_flags |= IB_WC_WITH_INVALIDATE; + if ((cqe->flags & (CQ_RES_RC_FLAGS_RDMA | CQ_RES_RC_FLAGS_IMM)) == + (CQ_RES_RC_FLAGS_RDMA | CQ_RES_RC_FLAGS_IMM)) + wc->opcode = IB_WC_RECV_RDMA_WITH_IMM; +} + +static void bnxt_re_process_res_shadow_qp_wc(struct bnxt_re_qp *qp, + struct ib_wc *wc, + struct bnxt_qplib_cqe *cqe) +{ + u32 tbl_idx; + struct bnxt_re_dev *rdev = qp->rdev; + struct bnxt_re_qp *qp1_qp = NULL; + struct bnxt_qplib_cqe *orig_cqe = NULL; + struct bnxt_re_sqp_entries *sqp_entry = NULL; + int nw_type; + + tbl_idx = cqe->wr_id; + + sqp_entry = &rdev->sqp_tbl[tbl_idx]; + qp1_qp = sqp_entry->qp1_qp; + orig_cqe = &sqp_entry->cqe; + + wc->wr_id = sqp_entry->wrid; + wc->byte_len = orig_cqe->length; + wc->qp = &qp1_qp->ib_qp; + + wc->ex.imm_data = orig_cqe->immdata; + wc->src_qp = orig_cqe->src_qp; + memcpy(wc->smac, orig_cqe->smac, ETH_ALEN); + wc->port_num = 1; + wc->vendor_err = orig_cqe->status; + + wc->opcode = IB_WC_RECV; + wc->status = __rawqp1_to_ib_wc_status(orig_cqe->status); + wc->wc_flags |= IB_WC_GRH; + + nw_type = bnxt_re_check_packet_type(orig_cqe->raweth_qp1_flags, + orig_cqe->raweth_qp1_flags2); + if (nw_type >= 0) { + wc->network_hdr_type = bnxt_re_to_ib_nw_type(nw_type); + wc->wc_flags |= IB_WC_WITH_NETWORK_HDR_TYPE; + } +} + +static void bnxt_re_process_res_ud_wc(struct ib_wc *wc, + struct bnxt_qplib_cqe *cqe) +{ + wc->opcode = IB_WC_RECV; + wc->status = __rc_to_ib_wc_status(cqe->status); + + if (cqe->flags & CQ_RES_RC_FLAGS_IMM) + wc->wc_flags |= IB_WC_WITH_IMM; + if (cqe->flags & CQ_RES_RC_FLAGS_INV) + wc->wc_flags |= IB_WC_WITH_INVALIDATE; + if ((cqe->flags & (CQ_RES_RC_FLAGS_RDMA | CQ_RES_RC_FLAGS_IMM)) == + (CQ_RES_RC_FLAGS_RDMA | CQ_RES_RC_FLAGS_IMM)) + wc->opcode = IB_WC_RECV_RDMA_WITH_IMM; +} + +int bnxt_re_poll_cq(struct ib_cq *ib_cq, int num_entries, struct ib_wc *wc) +{ + struct bnxt_re_cq *cq = container_of(ib_cq, struct bnxt_re_cq, ib_cq); + struct bnxt_re_qp *qp; + struct bnxt_qplib_cqe *cqe; + int i, ncqe, budget; + u32 tbl_idx; + struct bnxt_re_sqp_entries *sqp_entry = NULL; + unsigned long flags; + + spin_lock_irqsave(&cq->cq_lock, flags); + budget = min_t(u32, num_entries, cq->max_cql); + if (!cq->cql) { + dev_err(rdev_to_dev(cq->rdev), "POLL CQ : no CQL to use"); + goto exit; + } + cqe = &cq->cql[0]; + while (budget) { + ncqe = bnxt_qplib_poll_cq(&cq->qplib_cq, cqe, budget); + if (!ncqe) + break; + + for (i = 0; i < ncqe; i++, cqe++) { + /* Transcribe each qplib_wqe back to ib_wc */ + memset(wc, 0, sizeof(*wc)); + + wc->wr_id = cqe->wr_id; + wc->byte_len = cqe->length; + qp = container_of + ((struct bnxt_qplib_qp *) + (unsigned long)(cqe->qp_handle), + struct bnxt_re_qp, qplib_qp); + if (!qp) { + dev_err(rdev_to_dev(cq->rdev), + "POLL CQ : bad QP handle"); + continue; + } + wc->qp = &qp->ib_qp; + wc->ex.imm_data = cqe->immdata; + wc->src_qp = cqe->src_qp; + memcpy(wc->smac, cqe->smac, ETH_ALEN); + wc->port_num = 1; + wc->vendor_err = cqe->status; + + switch (cqe->opcode) { + case CQ_BASE_CQE_TYPE_REQ: + if (qp->qplib_qp.id == + qp->rdev->qp1_sqp->qplib_qp.id) { + /* Handle this completion with + * the stored completion + */ + memset(wc, 0, sizeof(*wc)); + continue; + } + bnxt_re_process_req_wc(wc, cqe); + break; + case CQ_BASE_CQE_TYPE_RES_RAWETH_QP1: + if (!cqe->status) { + int rc = 0; + + rc = bnxt_re_process_raw_qp_pkt_rx + (qp, cqe); + if (!rc) { + memset(wc, 0, sizeof(*wc)); + continue; + } + cqe->status = -1; + } + /* Errors need not be looped back. + * But change the wr_id to the one + * stored in the table + */ + tbl_idx = cqe->wr_id; + sqp_entry = &cq->rdev->sqp_tbl[tbl_idx]; + wc->wr_id = sqp_entry->wrid; + bnxt_re_process_res_rawqp1_wc(wc, cqe); + break; + case CQ_BASE_CQE_TYPE_RES_RC: + bnxt_re_process_res_rc_wc(wc, cqe); + break; + case CQ_BASE_CQE_TYPE_RES_UD: + if (qp->qplib_qp.id == + qp->rdev->qp1_sqp->qplib_qp.id) { + /* Handle this completion with + * the stored completion + */ + if (cqe->status) { + continue; + } else { + bnxt_re_process_res_shadow_qp_wc + (qp, wc, cqe); + break; + } + } + bnxt_re_process_res_ud_wc(wc, cqe); + break; + default: + dev_err(rdev_to_dev(cq->rdev), + "POLL CQ : type 0x%x not handled", + cqe->opcode); + continue; + } + wc++; + budget--; + } + } +exit: + spin_unlock_irqrestore(&cq->cq_lock, flags); + return num_entries - budget; +} + +int bnxt_re_req_notify_cq(struct ib_cq *ib_cq, + enum ib_cq_notify_flags ib_cqn_flags) +{ + struct bnxt_re_cq *cq = container_of(ib_cq, struct bnxt_re_cq, ib_cq); + int type = 0; + + /* Trigger on the very next completion */ + if (ib_cqn_flags & IB_CQ_NEXT_COMP) + type = DBR_DBR_TYPE_CQ_ARMALL; + /* Trigger on the next solicited completion */ + else if (ib_cqn_flags & IB_CQ_SOLICITED) + type = DBR_DBR_TYPE_CQ_ARMSE; + + bnxt_qplib_req_notify_cq(&cq->qplib_cq, type); + + return 0; +} + +/* Memory Regions */ +struct ib_mr *bnxt_re_get_dma_mr(struct ib_pd *ib_pd, int mr_access_flags) +{ + struct bnxt_re_pd *pd = container_of(ib_pd, struct bnxt_re_pd, ib_pd); + struct bnxt_re_dev *rdev = pd->rdev; + struct bnxt_re_mr *mr; + u64 pbl = 0; + int rc; + + mr = kzalloc(sizeof(*mr), GFP_KERNEL); + if (!mr) + return ERR_PTR(-ENOMEM); + + mr->rdev = rdev; + mr->qplib_mr.pd = &pd->qplib_pd; + mr->qplib_mr.flags = __from_ib_access_flags(mr_access_flags); + mr->qplib_mr.type = CMDQ_ALLOCATE_MRW_MRW_FLAGS_PMR; + + /* Allocate and register 0 as the address */ + rc = bnxt_qplib_alloc_mrw(&rdev->qplib_res, &mr->qplib_mr); + if (rc) + goto fail; + + mr->qplib_mr.hwq.level = PBL_LVL_MAX; + mr->qplib_mr.total_size = -1; /* Infinte length */ + rc = bnxt_qplib_reg_mr(&rdev->qplib_res, &mr->qplib_mr, &pbl, 0, false); + if (rc) + goto fail_mr; + + mr->ib_mr.lkey = mr->qplib_mr.lkey; + if (mr_access_flags & (IB_ACCESS_REMOTE_WRITE | IB_ACCESS_REMOTE_READ | + IB_ACCESS_REMOTE_ATOMIC)) + mr->ib_mr.rkey = mr->ib_mr.lkey; + atomic_inc(&rdev->mr_count); + + return &mr->ib_mr; + +fail_mr: + bnxt_qplib_free_mrw(&rdev->qplib_res, &mr->qplib_mr); +fail: + kfree(mr); + return ERR_PTR(rc); +} + +int bnxt_re_dereg_mr(struct ib_mr *ib_mr) +{ + struct bnxt_re_mr *mr = container_of(ib_mr, struct bnxt_re_mr, ib_mr); + struct bnxt_re_dev *rdev = mr->rdev; + int rc = 0; + + if (mr->npages && mr->pages) { + rc = bnxt_qplib_free_fast_reg_page_list(&rdev->qplib_res, + &mr->qplib_frpl); + kfree(mr->pages); + mr->npages = 0; + mr->pages = NULL; + } + rc = bnxt_qplib_free_mrw(&rdev->qplib_res, &mr->qplib_mr); + + if (!IS_ERR(mr->ib_umem) && mr->ib_umem) + ib_umem_release(mr->ib_umem); + + kfree(mr); + atomic_dec(&rdev->mr_count); + return rc; +} + +static int bnxt_re_set_page(struct ib_mr *ib_mr, u64 addr) +{ + struct bnxt_re_mr *mr = container_of(ib_mr, struct bnxt_re_mr, ib_mr); + + if (unlikely(mr->npages == mr->qplib_frpl.max_pg_ptrs)) + return -ENOMEM; + + mr->pages[mr->npages++] = addr; + return 0; +} + +int bnxt_re_map_mr_sg(struct ib_mr *ib_mr, struct scatterlist *sg, int sg_nents, + unsigned int *sg_offset) +{ + struct bnxt_re_mr *mr = container_of(ib_mr, struct bnxt_re_mr, ib_mr); + + mr->npages = 0; + return ib_sg_to_pages(ib_mr, sg, sg_nents, sg_offset, bnxt_re_set_page); +} + +struct ib_mr *bnxt_re_alloc_mr(struct ib_pd *ib_pd, enum ib_mr_type type, + u32 max_num_sg) +{ + struct bnxt_re_pd *pd = container_of(ib_pd, struct bnxt_re_pd, ib_pd); + struct bnxt_re_dev *rdev = pd->rdev; + struct bnxt_re_mr *mr = NULL; + int rc; + + if (type != IB_MR_TYPE_MEM_REG) { + dev_dbg(rdev_to_dev(rdev), "MR type 0x%x not supported", type); + return ERR_PTR(-EINVAL); + } + if (max_num_sg > MAX_PBL_LVL_1_PGS) + return ERR_PTR(-EINVAL); + + mr = kzalloc(sizeof(*mr), GFP_KERNEL); + if (!mr) + return ERR_PTR(-ENOMEM); + + mr->rdev = rdev; + mr->qplib_mr.pd = &pd->qplib_pd; + mr->qplib_mr.flags = BNXT_QPLIB_FR_PMR; + mr->qplib_mr.type = CMDQ_ALLOCATE_MRW_MRW_FLAGS_PMR; + + rc = bnxt_qplib_alloc_mrw(&rdev->qplib_res, &mr->qplib_mr); + if (rc) + goto fail; + + mr->ib_mr.lkey = mr->qplib_mr.lkey; + mr->ib_mr.rkey = mr->ib_mr.lkey; + + mr->pages = kcalloc(max_num_sg, sizeof(u64), GFP_KERNEL); + if (!mr->pages) { + rc = -ENOMEM; + goto fail; + } + rc = bnxt_qplib_alloc_fast_reg_page_list(&rdev->qplib_res, + &mr->qplib_frpl, max_num_sg); + if (rc) { + dev_err(rdev_to_dev(rdev), + "Failed to allocate HW FR page list"); + goto fail_mr; + } + + atomic_inc(&rdev->mr_count); + return &mr->ib_mr; + +fail_mr: + bnxt_qplib_free_mrw(&rdev->qplib_res, &mr->qplib_mr); +fail: + kfree(mr->pages); + kfree(mr); + return ERR_PTR(rc); +} + +/* Fast Memory Regions */ +struct ib_fmr *bnxt_re_alloc_fmr(struct ib_pd *ib_pd, int mr_access_flags, + struct ib_fmr_attr *fmr_attr) +{ + struct bnxt_re_pd *pd = container_of(ib_pd, struct bnxt_re_pd, ib_pd); + struct bnxt_re_dev *rdev = pd->rdev; + struct bnxt_re_fmr *fmr; + int rc; + + if (fmr_attr->max_pages > MAX_PBL_LVL_2_PGS || + fmr_attr->max_maps > rdev->dev_attr.max_map_per_fmr) { + dev_err(rdev_to_dev(rdev), "Allocate FMR exceeded Max limit"); + return ERR_PTR(-ENOMEM); + } + fmr = kzalloc(sizeof(*fmr), GFP_KERNEL); + if (!fmr) + return ERR_PTR(-ENOMEM); + + fmr->rdev = rdev; + fmr->qplib_fmr.pd = &pd->qplib_pd; + fmr->qplib_fmr.type = CMDQ_ALLOCATE_MRW_MRW_FLAGS_PMR; + + rc = bnxt_qplib_alloc_mrw(&rdev->qplib_res, &fmr->qplib_fmr); + if (rc) + goto fail; + + fmr->qplib_fmr.flags = __from_ib_access_flags(mr_access_flags); + fmr->ib_fmr.lkey = fmr->qplib_fmr.lkey; + fmr->ib_fmr.rkey = fmr->ib_fmr.lkey; + + atomic_inc(&rdev->mr_count); + return &fmr->ib_fmr; +fail: + kfree(fmr); + return ERR_PTR(rc); +} + +int bnxt_re_map_phys_fmr(struct ib_fmr *ib_fmr, u64 *page_list, int list_len, + u64 iova) +{ + struct bnxt_re_fmr *fmr = container_of(ib_fmr, struct bnxt_re_fmr, + ib_fmr); + struct bnxt_re_dev *rdev = fmr->rdev; + int rc; + + fmr->qplib_fmr.va = iova; + fmr->qplib_fmr.total_size = list_len * PAGE_SIZE; + + rc = bnxt_qplib_reg_mr(&rdev->qplib_res, &fmr->qplib_fmr, page_list, + list_len, true); + if (rc) + dev_err(rdev_to_dev(rdev), "Failed to map FMR for lkey = 0x%x!", + fmr->ib_fmr.lkey); + return rc; +} + +int bnxt_re_unmap_fmr(struct list_head *fmr_list) +{ + struct bnxt_re_dev *rdev; + struct bnxt_re_fmr *fmr; + struct ib_fmr *ib_fmr; + int rc = 0; + + /* Validate each FMRs inside the fmr_list */ + list_for_each_entry(ib_fmr, fmr_list, list) { + fmr = container_of(ib_fmr, struct bnxt_re_fmr, ib_fmr); + rdev = fmr->rdev; + + if (rdev) { + rc = bnxt_qplib_dereg_mrw(&rdev->qplib_res, + &fmr->qplib_fmr, true); + if (rc) + break; + } + } + return rc; +} + +int bnxt_re_dealloc_fmr(struct ib_fmr *ib_fmr) +{ + struct bnxt_re_fmr *fmr = container_of(ib_fmr, struct bnxt_re_fmr, + ib_fmr); + struct bnxt_re_dev *rdev = fmr->rdev; + int rc; + + rc = bnxt_qplib_free_mrw(&rdev->qplib_res, &fmr->qplib_fmr); + if (rc) + dev_err(rdev_to_dev(rdev), "Failed to free FMR"); + + kfree(fmr); + atomic_dec(&rdev->mr_count); + return rc; +} + +/* uverbs */ +struct ib_mr *bnxt_re_reg_user_mr(struct ib_pd *ib_pd, u64 start, u64 length, + u64 virt_addr, int mr_access_flags, + struct ib_udata *udata) +{ + struct bnxt_re_pd *pd = container_of(ib_pd, struct bnxt_re_pd, ib_pd); + struct bnxt_re_dev *rdev = pd->rdev; + struct bnxt_re_mr *mr; + struct ib_umem *umem; + u64 *pbl_tbl, *pbl_tbl_orig; + int i, umem_pgs, pages, page_shift, rc; + struct scatterlist *sg; + int entry; + + mr = kzalloc(sizeof(*mr), GFP_KERNEL); + if (!mr) + return ERR_PTR(-ENOMEM); + + mr->rdev = rdev; + mr->qplib_mr.pd = &pd->qplib_pd; + mr->qplib_mr.flags = __from_ib_access_flags(mr_access_flags); + mr->qplib_mr.type = CMDQ_ALLOCATE_MRW_MRW_FLAGS_MR; + + umem = ib_umem_get(ib_pd->uobject->context, start, length, + mr_access_flags, 0); + if (IS_ERR(umem)) { + dev_err(rdev_to_dev(rdev), "Failed to get umem"); + rc = -EFAULT; + goto free_mr; + } + mr->ib_umem = umem; + + rc = bnxt_qplib_alloc_mrw(&rdev->qplib_res, &mr->qplib_mr); + if (rc) { + dev_err(rdev_to_dev(rdev), "Failed to allocate MR"); + goto release_umem; + } + /* The fixed portion of the rkey is the same as the lkey */ + mr->ib_mr.rkey = mr->qplib_mr.rkey; + + mr->qplib_mr.va = virt_addr; + umem_pgs = ib_umem_page_count(umem); + if (!umem_pgs) { + dev_err(rdev_to_dev(rdev), "umem is invalid!"); + rc = -EINVAL; + goto free_mrw; + } + mr->qplib_mr.total_size = length; + + pbl_tbl = kcalloc(umem_pgs, sizeof(u64 *), GFP_KERNEL); + if (!pbl_tbl) { + rc = -EINVAL; + goto free_mrw; + } + pbl_tbl_orig = pbl_tbl; + + page_shift = ilog2(umem->page_size); + if (umem->hugetlb) { + dev_err(rdev_to_dev(rdev), "umem hugetlb not supported!"); + rc = -EFAULT; + goto fail; + } + if (umem->page_size != PAGE_SIZE) { + dev_err(rdev_to_dev(rdev), "umem page size unsupported!"); + rc = -EFAULT; + goto fail; + } + /* Map umem buf ptrs to the PBL */ + for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) { + pages = sg_dma_len(sg) >> page_shift; + for (i = 0; i < pages; i++, pbl_tbl++) + *pbl_tbl = sg_dma_address(sg) + (i << page_shift); + } + rc = bnxt_qplib_reg_mr(&rdev->qplib_res, &mr->qplib_mr, pbl_tbl_orig, + umem_pgs, false); + if (rc) { + dev_err(rdev_to_dev(rdev), "Failed to register user MR"); + goto fail; + } + + kfree(pbl_tbl_orig); + + mr->ib_mr.lkey = mr->qplib_mr.lkey; + mr->ib_mr.rkey = mr->qplib_mr.lkey; + atomic_inc(&rdev->mr_count); + + return &mr->ib_mr; +fail: + kfree(pbl_tbl_orig); +free_mrw: + bnxt_qplib_free_mrw(&rdev->qplib_res, &mr->qplib_mr); +release_umem: + ib_umem_release(umem); +free_mr: + kfree(mr); + return ERR_PTR(rc); +} + +struct ib_ucontext *bnxt_re_alloc_ucontext(struct ib_device *ibdev, + struct ib_udata *udata) +{ + struct bnxt_re_dev *rdev = to_bnxt_re_dev(ibdev, ibdev); + struct bnxt_re_uctx_resp resp; + struct bnxt_re_ucontext *uctx; + struct bnxt_qplib_dev_attr *dev_attr = &rdev->dev_attr; + int rc; + + dev_dbg(rdev_to_dev(rdev), "ABI version requested %d", + ibdev->uverbs_abi_ver); + + if (ibdev->uverbs_abi_ver != BNXT_RE_ABI_VERSION) { + dev_dbg(rdev_to_dev(rdev), " is different from the device %d ", + BNXT_RE_ABI_VERSION); + return ERR_PTR(-EPERM); + } + + uctx = kzalloc(sizeof(*uctx), GFP_KERNEL); + if (!uctx) + return ERR_PTR(-ENOMEM); + + uctx->rdev = rdev; + + uctx->shpg = (void *)__get_free_page(GFP_KERNEL); + if (!uctx->shpg) { + rc = -ENOMEM; + goto fail; + } + spin_lock_init(&uctx->sh_lock); + + resp.dev_id = rdev->en_dev->pdev->devfn; /*Temp, Use idr_alloc instead*/ + resp.max_qp = rdev->qplib_ctx.qpc_count; + resp.pg_size = PAGE_SIZE; + resp.cqe_sz = sizeof(struct cq_base); + resp.max_cqd = dev_attr->max_cq_wqes; + resp.rsvd = 0; + + rc = ib_copy_to_udata(udata, &resp, sizeof(resp)); + if (rc) { + dev_err(rdev_to_dev(rdev), "Failed to copy user context"); + rc = -EFAULT; + goto cfail; + } + + return &uctx->ib_uctx; +cfail: + free_page((unsigned long)uctx->shpg); + uctx->shpg = NULL; +fail: + kfree(uctx); + return ERR_PTR(rc); +} + +int bnxt_re_dealloc_ucontext(struct ib_ucontext *ib_uctx) +{ + struct bnxt_re_ucontext *uctx = container_of(ib_uctx, + struct bnxt_re_ucontext, + ib_uctx); + if (uctx->shpg) + free_page((unsigned long)uctx->shpg); + kfree(uctx); + return 0; +} + +/* Helper function to mmap the virtual memory from user app */ +int bnxt_re_mmap(struct ib_ucontext *ib_uctx, struct vm_area_struct *vma) +{ + struct bnxt_re_ucontext *uctx = container_of(ib_uctx, + struct bnxt_re_ucontext, + ib_uctx); + struct bnxt_re_dev *rdev = uctx->rdev; + u64 pfn; + + if (vma->vm_end - vma->vm_start != PAGE_SIZE) + return -EINVAL; + + if (vma->vm_pgoff) { + vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); + if (io_remap_pfn_range(vma, vma->vm_start, vma->vm_pgoff, + PAGE_SIZE, vma->vm_page_prot)) { + dev_err(rdev_to_dev(rdev), "Failed to map DPI"); + return -EAGAIN; + } + } else { + pfn = virt_to_phys(uctx->shpg) >> PAGE_SHIFT; + if (remap_pfn_range(vma, vma->vm_start, + pfn, PAGE_SIZE, vma->vm_page_prot)) { + dev_err(rdev_to_dev(rdev), + "Failed to map shared page"); + return -EAGAIN; + } + } + + return 0; +} diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.h b/drivers/infiniband/hw/bnxt_re/ib_verbs.h new file mode 100644 index 000000000000..b4084c252f06 --- /dev/null +++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.h @@ -0,0 +1,197 @@ +/* + * Broadcom NetXtreme-E RoCE driver. + * + * Copyright (c) 2016 - 2017, Broadcom. All rights reserved. The term + * Broadcom refers to Broadcom Limited and/or its subsidiaries. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, + * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS + * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE + * OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN + * IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * Description: IB Verbs interpreter (header) + */ + +#ifndef __BNXT_RE_IB_VERBS_H__ +#define __BNXT_RE_IB_VERBS_H__ + +struct bnxt_re_gid_ctx { + u32 idx; + u32 refcnt; +}; + +struct bnxt_re_pd { + struct bnxt_re_dev *rdev; + struct ib_pd ib_pd; + struct bnxt_qplib_pd qplib_pd; + struct bnxt_qplib_dpi dpi; +}; + +struct bnxt_re_ah { + struct bnxt_re_dev *rdev; + struct ib_ah ib_ah; + struct bnxt_qplib_ah qplib_ah; +}; + +struct bnxt_re_qp { + struct list_head list; + struct bnxt_re_dev *rdev; + struct ib_qp ib_qp; + spinlock_t sq_lock; /* protect sq */ + struct bnxt_qplib_qp qplib_qp; + struct ib_umem *sumem; + struct ib_umem *rumem; + /* QP1 */ + u32 send_psn; + struct ib_ud_header qp1_hdr; +}; + +struct bnxt_re_cq { + struct bnxt_re_dev *rdev; + spinlock_t cq_lock; /* protect cq */ + u16 cq_count; + u16 cq_period; + struct ib_cq ib_cq; + struct bnxt_qplib_cq qplib_cq; + struct bnxt_qplib_cqe *cql; +#define MAX_CQL_PER_POLL 1024 + u32 max_cql; + struct ib_umem *umem; +}; + +struct bnxt_re_mr { + struct bnxt_re_dev *rdev; + struct ib_mr ib_mr; + struct ib_umem *ib_umem; + struct bnxt_qplib_mrw qplib_mr; + u32 npages; + u64 *pages; + struct bnxt_qplib_frpl qplib_frpl; +}; + +struct bnxt_re_frpl { + struct bnxt_re_dev *rdev; + struct bnxt_qplib_frpl qplib_frpl; + u64 *page_list; +}; + +struct bnxt_re_fmr { + struct bnxt_re_dev *rdev; + struct ib_fmr ib_fmr; + struct bnxt_qplib_mrw qplib_fmr; +}; + +struct bnxt_re_mw { + struct bnxt_re_dev *rdev; + struct ib_mw ib_mw; + struct bnxt_qplib_mrw qplib_mw; +}; + +struct bnxt_re_ucontext { + struct bnxt_re_dev *rdev; + struct ib_ucontext ib_uctx; + struct bnxt_qplib_dpi *dpi; + void *shpg; + spinlock_t sh_lock; /* protect shpg */ +}; + +struct net_device *bnxt_re_get_netdev(struct ib_device *ibdev, u8 port_num); + +int bnxt_re_query_device(struct ib_device *ibdev, + struct ib_device_attr *ib_attr, + struct ib_udata *udata); +int bnxt_re_modify_device(struct ib_device *ibdev, + int device_modify_mask, + struct ib_device_modify *device_modify); +int bnxt_re_query_port(struct ib_device *ibdev, u8 port_num, + struct ib_port_attr *port_attr); +int bnxt_re_modify_port(struct ib_device *ibdev, u8 port_num, + int port_modify_mask, + struct ib_port_modify *port_modify); +int bnxt_re_get_port_immutable(struct ib_device *ibdev, u8 port_num, + struct ib_port_immutable *immutable); +int bnxt_re_query_pkey(struct ib_device *ibdev, u8 port_num, + u16 index, u16 *pkey); +int bnxt_re_del_gid(struct ib_device *ibdev, u8 port_num, + unsigned int index, void **context); +int bnxt_re_add_gid(struct ib_device *ibdev, u8 port_num, + unsigned int index, const union ib_gid *gid, + const struct ib_gid_attr *attr, void **context); +int bnxt_re_query_gid(struct ib_device *ibdev, u8 port_num, + int index, union ib_gid *gid); +enum rdma_link_layer bnxt_re_get_link_layer(struct ib_device *ibdev, + u8 port_num); +struct ib_pd *bnxt_re_alloc_pd(struct ib_device *ibdev, + struct ib_ucontext *context, + struct ib_udata *udata); +int bnxt_re_dealloc_pd(struct ib_pd *pd); +struct ib_ah *bnxt_re_create_ah(struct ib_pd *pd, + struct ib_ah_attr *ah_attr, + struct ib_udata *udata); +int bnxt_re_modify_ah(struct ib_ah *ah, struct ib_ah_attr *ah_attr); +int bnxt_re_query_ah(struct ib_ah *ah, struct ib_ah_attr *ah_attr); +int bnxt_re_destroy_ah(struct ib_ah *ah); +struct ib_qp *bnxt_re_create_qp(struct ib_pd *pd, + struct ib_qp_init_attr *qp_init_attr, + struct ib_udata *udata); +int bnxt_re_modify_qp(struct ib_qp *qp, struct ib_qp_attr *qp_attr, + int qp_attr_mask, struct ib_udata *udata); +int bnxt_re_query_qp(struct ib_qp *qp, struct ib_qp_attr *qp_attr, + int qp_attr_mask, struct ib_qp_init_attr *qp_init_attr); +int bnxt_re_destroy_qp(struct ib_qp *qp); +int bnxt_re_post_send(struct ib_qp *qp, struct ib_send_wr *send_wr, + struct ib_send_wr **bad_send_wr); +int bnxt_re_post_recv(struct ib_qp *qp, struct ib_recv_wr *recv_wr, + struct ib_recv_wr **bad_recv_wr); +struct ib_cq *bnxt_re_create_cq(struct ib_device *ibdev, + const struct ib_cq_init_attr *attr, + struct ib_ucontext *context, + struct ib_udata *udata); +int bnxt_re_destroy_cq(struct ib_cq *cq); +int bnxt_re_poll_cq(struct ib_cq *cq, int num_entries, struct ib_wc *wc); +int bnxt_re_req_notify_cq(struct ib_cq *cq, enum ib_cq_notify_flags flags); +struct ib_mr *bnxt_re_get_dma_mr(struct ib_pd *pd, int mr_access_flags); + +int bnxt_re_map_mr_sg(struct ib_mr *ib_mr, struct scatterlist *sg, int sg_nents, + unsigned int *sg_offset); +struct ib_mr *bnxt_re_alloc_mr(struct ib_pd *ib_pd, enum ib_mr_type mr_type, + u32 max_num_sg); +int bnxt_re_dereg_mr(struct ib_mr *mr); +struct ib_fmr *bnxt_re_alloc_fmr(struct ib_pd *pd, int mr_access_flags, + struct ib_fmr_attr *fmr_attr); +int bnxt_re_map_phys_fmr(struct ib_fmr *fmr, u64 *page_list, int list_len, + u64 iova); +int bnxt_re_unmap_fmr(struct list_head *fmr_list); +int bnxt_re_dealloc_fmr(struct ib_fmr *fmr); +struct ib_mr *bnxt_re_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, + u64 virt_addr, int mr_access_flags, + struct ib_udata *udata); +struct ib_ucontext *bnxt_re_alloc_ucontext(struct ib_device *ibdev, + struct ib_udata *udata); +int bnxt_re_dealloc_ucontext(struct ib_ucontext *context); +int bnxt_re_mmap(struct ib_ucontext *context, struct vm_area_struct *vma); +#endif /* __BNXT_RE_IB_VERBS_H__ */ diff --git a/drivers/infiniband/hw/bnxt_re/main.c b/drivers/infiniband/hw/bnxt_re/main.c new file mode 100644 index 000000000000..5d355401179b --- /dev/null +++ b/drivers/infiniband/hw/bnxt_re/main.c @@ -0,0 +1,1315 @@ +/* + * Broadcom NetXtreme-E RoCE driver. + * + * Copyright (c) 2016 - 2017, Broadcom. All rights reserved. The term + * Broadcom refers to Broadcom Limited and/or its subsidiaries. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, + * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS + * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE + * OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN + * IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * Description: Main component of the bnxt_re driver + */ + +#include <linux/module.h> +#include <linux/netdevice.h> +#include <linux/ethtool.h> +#include <linux/mutex.h> +#include <linux/list.h> +#include <linux/rculist.h> +#include <linux/spinlock.h> +#include <linux/pci.h> +#include <net/dcbnl.h> +#include <net/ipv6.h> +#include <net/addrconf.h> +#include <linux/if_ether.h> + +#include <rdma/ib_verbs.h> +#include <rdma/ib_user_verbs.h> +#include <rdma/ib_umem.h> +#include <rdma/ib_addr.h> + +#include "bnxt_ulp.h" +#include "roce_hsi.h" +#include "qplib_res.h" +#include "qplib_sp.h" +#include "qplib_fp.h" +#include "qplib_rcfw.h" +#include "bnxt_re.h" +#include "ib_verbs.h" +#include <rdma/bnxt_re-abi.h> +#include "bnxt.h" +static char version[] = + BNXT_RE_DESC " v" ROCE_DRV_MODULE_VERSION "\n"; + +MODULE_AUTHOR("Eddie Wai <eddie.wai@broadcom.com>"); +MODULE_DESCRIPTION(BNXT_RE_DESC " Driver"); +MODULE_LICENSE("Dual BSD/GPL"); +MODULE_VERSION(ROCE_DRV_MODULE_VERSION); + +/* globals */ +static struct list_head bnxt_re_dev_list = LIST_HEAD_INIT(bnxt_re_dev_list); +/* Mutex to protect the list of bnxt_re devices added */ +static DEFINE_MUTEX(bnxt_re_dev_lock); +static struct workqueue_struct *bnxt_re_wq; + +/* for handling bnxt_en callbacks later */ +static void bnxt_re_stop(void *p) +{ +} + +static void bnxt_re_start(void *p) +{ +} + +static void bnxt_re_sriov_config(void *p, int num_vfs) +{ +} + +static struct bnxt_ulp_ops bnxt_re_ulp_ops = { + .ulp_async_notifier = NULL, + .ulp_stop = bnxt_re_stop, + .ulp_start = bnxt_re_start, + .ulp_sriov_config = bnxt_re_sriov_config +}; + +/* RoCE -> Net driver */ + +/* Driver registration routines used to let the networking driver (bnxt_en) + * to know that the RoCE driver is now installed + */ +static int bnxt_re_unregister_netdev(struct bnxt_re_dev *rdev, bool lock_wait) +{ + struct bnxt_en_dev *en_dev; + int rc; + + if (!rdev) + return -EINVAL; + + en_dev = rdev->en_dev; + /* Acquire rtnl lock if it is not invokded from netdev event */ + if (lock_wait) + rtnl_lock(); + + rc = en_dev->en_ops->bnxt_unregister_device(rdev->en_dev, + BNXT_ROCE_ULP); + if (lock_wait) + rtnl_unlock(); + return rc; +} + +static int bnxt_re_register_netdev(struct bnxt_re_dev *rdev) +{ + struct bnxt_en_dev *en_dev; + int rc = 0; + + if (!rdev) + return -EINVAL; + + en_dev = rdev->en_dev; + + rtnl_lock(); + rc = en_dev->en_ops->bnxt_register_device(en_dev, BNXT_ROCE_ULP, + &bnxt_re_ulp_ops, rdev); + rtnl_unlock(); + return rc; +} + +static int bnxt_re_free_msix(struct bnxt_re_dev *rdev, bool lock_wait) +{ + struct bnxt_en_dev *en_dev; + int rc; + + if (!rdev) + return -EINVAL; + + en_dev = rdev->en_dev; + + if (lock_wait) + rtnl_lock(); + + rc = en_dev->en_ops->bnxt_free_msix(rdev->en_dev, BNXT_ROCE_ULP); + + if (lock_wait) + rtnl_unlock(); + return rc; +} + +static int bnxt_re_request_msix(struct bnxt_re_dev *rdev) +{ + int rc = 0, num_msix_want = BNXT_RE_MIN_MSIX, num_msix_got; + struct bnxt_en_dev *en_dev; + + if (!rdev) + return -EINVAL; + + en_dev = rdev->en_dev; + + rtnl_lock(); + num_msix_got = en_dev->en_ops->bnxt_request_msix(en_dev, BNXT_ROCE_ULP, + rdev->msix_entries, + num_msix_want); + if (num_msix_got < BNXT_RE_MIN_MSIX) { + rc = -EINVAL; + goto done; + } + if (num_msix_got != num_msix_want) { + dev_warn(rdev_to_dev(rdev), + "Requested %d MSI-X vectors, got %d\n", + num_msix_want, num_msix_got); + } + rdev->num_msix = num_msix_got; +done: + rtnl_unlock(); + return rc; +} + +static void bnxt_re_init_hwrm_hdr(struct bnxt_re_dev *rdev, struct input *hdr, + u16 opcd, u16 crid, u16 trid) +{ + hdr->req_type = cpu_to_le16(opcd); + hdr->cmpl_ring = cpu_to_le16(crid); + hdr->target_id = cpu_to_le16(trid); +} + +static void bnxt_re_fill_fw_msg(struct bnxt_fw_msg *fw_msg, void *msg, + int msg_len, void *resp, int resp_max_len, + int timeout) +{ + fw_msg->msg = msg; + fw_msg->msg_len = msg_len; + fw_msg->resp = resp; + fw_msg->resp_max_len = resp_max_len; + fw_msg->timeout = timeout; +} + +static int bnxt_re_net_ring_free(struct bnxt_re_dev *rdev, u16 fw_ring_id, + bool lock_wait) +{ + struct bnxt_en_dev *en_dev = rdev->en_dev; + struct hwrm_ring_free_input req = {0}; + struct hwrm_ring_free_output resp; + struct bnxt_fw_msg fw_msg; + bool do_unlock = false; + int rc = -EINVAL; + + if (!en_dev) + return rc; + + memset(&fw_msg, 0, sizeof(fw_msg)); + if (lock_wait) { + rtnl_lock(); + do_unlock = true; + } + + bnxt_re_init_hwrm_hdr(rdev, (void *)&req, HWRM_RING_FREE, -1, -1); + req.ring_type = RING_ALLOC_REQ_RING_TYPE_L2_CMPL; + req.ring_id = cpu_to_le16(fw_ring_id); + bnxt_re_fill_fw_msg(&fw_msg, (void *)&req, sizeof(req), (void *)&resp, + sizeof(resp), DFLT_HWRM_CMD_TIMEOUT); + rc = en_dev->en_ops->bnxt_send_fw_msg(en_dev, BNXT_ROCE_ULP, &fw_msg); + if (rc) + dev_err(rdev_to_dev(rdev), + "Failed to free HW ring:%d :%#x", req.ring_id, rc); + if (do_unlock) + rtnl_unlock(); + return rc; +} + +static int bnxt_re_net_ring_alloc(struct bnxt_re_dev *rdev, dma_addr_t *dma_arr, + int pages, int type, u32 ring_mask, + u32 map_index, u16 *fw_ring_id) +{ + struct bnxt_en_dev *en_dev = rdev->en_dev; + struct hwrm_ring_alloc_input req = {0}; + struct hwrm_ring_alloc_output resp; + struct bnxt_fw_msg fw_msg; + int rc = -EINVAL; + + if (!en_dev) + return rc; + + memset(&fw_msg, 0, sizeof(fw_msg)); + rtnl_lock(); + bnxt_re_init_hwrm_hdr(rdev, (void *)&req, HWRM_RING_ALLOC, -1, -1); + req.enables = 0; + req.page_tbl_addr = cpu_to_le64(dma_arr[0]); + if (pages > 1) { + /* Page size is in log2 units */ + req.page_size = BNXT_PAGE_SHIFT; + req.page_tbl_depth = 1; + } + req.fbo = 0; + /* Association of ring index with doorbell index and MSIX number */ + req.logical_id = cpu_to_le16(map_index); + req.length = cpu_to_le32(ring_mask + 1); + req.ring_type = RING_ALLOC_REQ_RING_TYPE_L2_CMPL; + req.int_mode = RING_ALLOC_REQ_INT_MODE_MSIX; + bnxt_re_fill_fw_msg(&fw_msg, (void *)&req, sizeof(req), (void *)&resp, + sizeof(resp), DFLT_HWRM_CMD_TIMEOUT); + rc = en_dev->en_ops->bnxt_send_fw_msg(en_dev, BNXT_ROCE_ULP, &fw_msg); + if (!rc) + *fw_ring_id = le16_to_cpu(resp.ring_id); + + rtnl_unlock(); + return rc; +} + +static int bnxt_re_net_stats_ctx_free(struct bnxt_re_dev *rdev, + u32 fw_stats_ctx_id, bool lock_wait) +{ + struct bnxt_en_dev *en_dev = rdev->en_dev; + struct hwrm_stat_ctx_free_input req = {0}; + struct bnxt_fw_msg fw_msg; + bool do_unlock = false; + int rc = -EINVAL; + + if (!en_dev) + return rc; + + memset(&fw_msg, 0, sizeof(fw_msg)); + if (lock_wait) { + rtnl_lock(); + do_unlock = true; + } + + bnxt_re_init_hwrm_hdr(rdev, (void *)&req, HWRM_STAT_CTX_FREE, -1, -1); + req.stat_ctx_id = cpu_to_le32(fw_stats_ctx_id); + bnxt_re_fill_fw_msg(&fw_msg, (void *)&req, sizeof(req), (void *)&req, + sizeof(req), DFLT_HWRM_CMD_TIMEOUT); + rc = en_dev->en_ops->bnxt_send_fw_msg(en_dev, BNXT_ROCE_ULP, &fw_msg); + if (rc) + dev_err(rdev_to_dev(rdev), + "Failed to free HW stats context %#x", rc); + + if (do_unlock) + rtnl_unlock(); + return rc; +} + +static int bnxt_re_net_stats_ctx_alloc(struct bnxt_re_dev *rdev, + dma_addr_t dma_map, + u32 *fw_stats_ctx_id) +{ + struct hwrm_stat_ctx_alloc_output resp = {0}; + struct hwrm_stat_ctx_alloc_input req = {0}; + struct bnxt_en_dev *en_dev = rdev->en_dev; + struct bnxt_fw_msg fw_msg; + int rc = -EINVAL; + + *fw_stats_ctx_id = INVALID_STATS_CTX_ID; + + if (!en_dev) + return rc; + + memset(&fw_msg, 0, sizeof(fw_msg)); + rtnl_lock(); + + bnxt_re_init_hwrm_hdr(rdev, (void *)&req, HWRM_STAT_CTX_ALLOC, -1, -1); + req.update_period_ms = cpu_to_le32(1000); + req.stats_dma_addr = cpu_to_le64(dma_map); + bnxt_re_fill_fw_msg(&fw_msg, (void *)&req, sizeof(req), (void *)&resp, + sizeof(resp), DFLT_HWRM_CMD_TIMEOUT); + rc = en_dev->en_ops->bnxt_send_fw_msg(en_dev, BNXT_ROCE_ULP, &fw_msg); + if (!rc) + *fw_stats_ctx_id = le32_to_cpu(resp.stat_ctx_id); + + rtnl_unlock(); + return rc; +} + +/* Device */ + +static bool is_bnxt_re_dev(struct net_device *netdev) +{ + struct ethtool_drvinfo drvinfo; + + if (netdev->ethtool_ops && netdev->ethtool_ops->get_drvinfo) { + memset(&drvinfo, 0, sizeof(drvinfo)); + netdev->ethtool_ops->get_drvinfo(netdev, &drvinfo); + + if (strcmp(drvinfo.driver, "bnxt_en")) + return false; + return true; + } + return false; +} + +static struct bnxt_re_dev *bnxt_re_from_netdev(struct net_device *netdev) +{ + struct bnxt_re_dev *rdev; + + rcu_read_lock(); + list_for_each_entry_rcu(rdev, &bnxt_re_dev_list, list) { + if (rdev->netdev == netdev) { + rcu_read_unlock(); + return rdev; + } + } + rcu_read_unlock(); + return NULL; +} + +static void bnxt_re_dev_unprobe(struct net_device *netdev, + struct bnxt_en_dev *en_dev) +{ + dev_put(netdev); + module_put(en_dev->pdev->driver->driver.owner); +} + +static struct bnxt_en_dev *bnxt_re_dev_probe(struct net_device *netdev) +{ + struct bnxt *bp = netdev_priv(netdev); + struct bnxt_en_dev *en_dev; + struct pci_dev *pdev; + + /* Call bnxt_en's RoCE probe via indirect API */ + if (!bp->ulp_probe) + return ERR_PTR(-EINVAL); + + en_dev = bp->ulp_probe(netdev); + if (IS_ERR(en_dev)) + return en_dev; + + pdev = en_dev->pdev; + if (!pdev) + return ERR_PTR(-EINVAL); + + if (!(en_dev->flags & BNXT_EN_FLAG_ROCE_CAP)) { + dev_dbg(&pdev->dev, + "%s: probe error: RoCE is not supported on this device", + ROCE_DRV_MODULE_NAME); + return ERR_PTR(-ENODEV); + } + + /* Bump net device reference count */ + if (!try_module_get(pdev->driver->driver.owner)) + return ERR_PTR(-ENODEV); + + dev_hold(netdev); + + return en_dev; +} + +static void bnxt_re_unregister_ib(struct bnxt_re_dev *rdev) +{ + ib_unregister_device(&rdev->ibdev); +} + +static int bnxt_re_register_ib(struct bnxt_re_dev *rdev) +{ + struct ib_device *ibdev = &rdev->ibdev; + + /* ib device init */ + ibdev->owner = THIS_MODULE; + ibdev->node_type = RDMA_NODE_IB_CA; + strlcpy(ibdev->name, "bnxt_re%d", IB_DEVICE_NAME_MAX); + strlcpy(ibdev->node_desc, BNXT_RE_DESC " HCA", + strlen(BNXT_RE_DESC) + 5); + ibdev->phys_port_cnt = 1; + + bnxt_qplib_get_guid(rdev->netdev->dev_addr, (u8 *)&ibdev->node_guid); + + ibdev->num_comp_vectors = 1; + ibdev->dev.parent = &rdev->en_dev->pdev->dev; + ibdev->local_dma_lkey = BNXT_QPLIB_RSVD_LKEY; + + /* User space */ + ibdev->uverbs_abi_ver = BNXT_RE_ABI_VERSION; + ibdev->uverbs_cmd_mask = + (1ull << IB_USER_VERBS_CMD_GET_CONTEXT) | + (1ull << IB_USER_VERBS_CMD_QUERY_DEVICE) | + (1ull << IB_USER_VERBS_CMD_QUERY_PORT) | + (1ull << IB_USER_VERBS_CMD_ALLOC_PD) | + (1ull << IB_USER_VERBS_CMD_DEALLOC_PD) | + (1ull << IB_USER_VERBS_CMD_REG_MR) | + (1ull << IB_USER_VERBS_CMD_REREG_MR) | + (1ull << IB_USER_VERBS_CMD_DEREG_MR) | + (1ull << IB_USER_VERBS_CMD_CREATE_COMP_CHANNEL) | + (1ull << IB_USER_VERBS_CMD_CREATE_CQ) | + (1ull << IB_USER_VERBS_CMD_RESIZE_CQ) | + (1ull << IB_USER_VERBS_CMD_DESTROY_CQ) | + (1ull << IB_USER_VERBS_CMD_CREATE_QP) | + (1ull << IB_USER_VERBS_CMD_MODIFY_QP) | + (1ull << IB_USER_VERBS_CMD_QUERY_QP) | + (1ull << IB_USER_VERBS_CMD_DESTROY_QP) | + (1ull << IB_USER_VERBS_CMD_CREATE_SRQ) | + (1ull << IB_USER_VERBS_CMD_MODIFY_SRQ) | + (1ull << IB_USER_VERBS_CMD_QUERY_SRQ) | + (1ull << IB_USER_VERBS_CMD_DESTROY_SRQ) | + (1ull << IB_USER_VERBS_CMD_CREATE_AH) | + (1ull << IB_USER_VERBS_CMD_MODIFY_AH) | + (1ull << IB_USER_VERBS_CMD_QUERY_AH) | + (1ull << IB_USER_VERBS_CMD_DESTROY_AH); + /* POLL_CQ and REQ_NOTIFY_CQ is directly handled in libbnxt_re */ + + /* Kernel verbs */ + ibdev->query_device = bnxt_re_query_device; + ibdev->modify_device = bnxt_re_modify_device; + + ibdev->query_port = bnxt_re_query_port; + ibdev->modify_port = bnxt_re_modify_port; + ibdev->get_port_immutable = bnxt_re_get_port_immutable; + ibdev->query_pkey = bnxt_re_query_pkey; + ibdev->query_gid = bnxt_re_query_gid; + ibdev->get_netdev = bnxt_re_get_netdev; + ibdev->add_gid = bnxt_re_add_gid; + ibdev->del_gid = bnxt_re_del_gid; + ibdev->get_link_layer = bnxt_re_get_link_layer; + + ibdev->alloc_pd = bnxt_re_alloc_pd; + ibdev->dealloc_pd = bnxt_re_dealloc_pd; + + ibdev->create_ah = bnxt_re_create_ah; + ibdev->modify_ah = bnxt_re_modify_ah; + ibdev->query_ah = bnxt_re_query_ah; + ibdev->destroy_ah = bnxt_re_destroy_ah; + + ibdev->create_qp = bnxt_re_create_qp; + ibdev->modify_qp = bnxt_re_modify_qp; + ibdev->query_qp = bnxt_re_query_qp; + ibdev->destroy_qp = bnxt_re_destroy_qp; + + ibdev->post_send = bnxt_re_post_send; + ibdev->post_recv = bnxt_re_post_recv; + + ibdev->create_cq = bnxt_re_create_cq; + ibdev->destroy_cq = bnxt_re_destroy_cq; + ibdev->poll_cq = bnxt_re_poll_cq; + ibdev->req_notify_cq = bnxt_re_req_notify_cq; + + ibdev->get_dma_mr = bnxt_re_get_dma_mr; + ibdev->dereg_mr = bnxt_re_dereg_mr; + ibdev->alloc_mr = bnxt_re_alloc_mr; + ibdev->map_mr_sg = bnxt_re_map_mr_sg; + ibdev->alloc_fmr = bnxt_re_alloc_fmr; + ibdev->map_phys_fmr = bnxt_re_map_phys_fmr; + ibdev->unmap_fmr = bnxt_re_unmap_fmr; + ibdev->dealloc_fmr = bnxt_re_dealloc_fmr; + + ibdev->reg_user_mr = bnxt_re_reg_user_mr; + ibdev->alloc_ucontext = bnxt_re_alloc_ucontext; + ibdev->dealloc_ucontext = bnxt_re_dealloc_ucontext; + ibdev->mmap = bnxt_re_mmap; + + return ib_register_device(ibdev, NULL); +} + +static ssize_t show_rev(struct device *device, struct device_attribute *attr, + char *buf) +{ + struct bnxt_re_dev *rdev = to_bnxt_re_dev(device, ibdev.dev); + + return scnprintf(buf, PAGE_SIZE, "0x%x\n", rdev->en_dev->pdev->vendor); +} + +static ssize_t show_fw_ver(struct device *device, struct device_attribute *attr, + char *buf) +{ + struct bnxt_re_dev *rdev = to_bnxt_re_dev(device, ibdev.dev); + + return scnprintf(buf, PAGE_SIZE, "%s\n", rdev->dev_attr.fw_ver); +} + +static ssize_t show_hca(struct device *device, struct device_attribute *attr, + char *buf) +{ + struct bnxt_re_dev *rdev = to_bnxt_re_dev(device, ibdev.dev); + + return scnprintf(buf, PAGE_SIZE, "%s\n", rdev->ibdev.node_desc); +} + +static DEVICE_ATTR(hw_rev, 0444, show_rev, NULL); +static DEVICE_ATTR(fw_rev, 0444, show_fw_ver, NULL); +static DEVICE_ATTR(hca_type, 0444, show_hca, NULL); + +static struct device_attribute *bnxt_re_attributes[] = { + &dev_attr_hw_rev, + &dev_attr_fw_rev, + &dev_attr_hca_type +}; + +static void bnxt_re_dev_remove(struct bnxt_re_dev *rdev) +{ + dev_put(rdev->netdev); + rdev->netdev = NULL; + + mutex_lock(&bnxt_re_dev_lock); + list_del_rcu(&rdev->list); + mutex_unlock(&bnxt_re_dev_lock); + + synchronize_rcu(); + flush_workqueue(bnxt_re_wq); + + ib_dealloc_device(&rdev->ibdev); + /* rdev is gone */ +} + +static struct bnxt_re_dev *bnxt_re_dev_add(struct net_device *netdev, + struct bnxt_en_dev *en_dev) +{ + struct bnxt_re_dev *rdev; + + /* Allocate bnxt_re_dev instance here */ + rdev = (struct bnxt_re_dev *)ib_alloc_device(sizeof(*rdev)); + if (!rdev) { + dev_err(NULL, "%s: bnxt_re_dev allocation failure!", + ROCE_DRV_MODULE_NAME); + return NULL; + } + /* Default values */ + rdev->netdev = netdev; + dev_hold(rdev->netdev); + rdev->en_dev = en_dev; + rdev->id = rdev->en_dev->pdev->devfn; + INIT_LIST_HEAD(&rdev->qp_list); + mutex_init(&rdev->qp_lock); + atomic_set(&rdev->qp_count, 0); + atomic_set(&rdev->cq_count, 0); + atomic_set(&rdev->srq_count, 0); + atomic_set(&rdev->mr_count, 0); + atomic_set(&rdev->mw_count, 0); + rdev->cosq[0] = 0xFFFF; + rdev->cosq[1] = 0xFFFF; + + mutex_lock(&bnxt_re_dev_lock); + list_add_tail_rcu(&rdev->list, &bnxt_re_dev_list); + mutex_unlock(&bnxt_re_dev_lock); + return rdev; +} + +static int bnxt_re_aeq_handler(struct bnxt_qplib_rcfw *rcfw, + struct creq_func_event *aeqe) +{ + switch (aeqe->event) { + case CREQ_FUNC_EVENT_EVENT_TX_WQE_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_TX_DATA_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_RX_WQE_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_RX_DATA_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_CQ_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_TQM_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_CFCQ_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_CFCS_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_CFCC_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_CFCM_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_TIM_ERROR: + break; + default: + return -EINVAL; + } + return 0; +} + +static int bnxt_re_cqn_handler(struct bnxt_qplib_nq *nq, + struct bnxt_qplib_cq *handle) +{ + struct bnxt_re_cq *cq = container_of(handle, struct bnxt_re_cq, + qplib_cq); + + if (!cq) { + dev_err(NULL, "%s: CQ is NULL, CQN not handled", + ROCE_DRV_MODULE_NAME); + return -EINVAL; + } + if (cq->ib_cq.comp_handler) { + /* Lock comp_handler? */ + (*cq->ib_cq.comp_handler)(&cq->ib_cq, cq->ib_cq.cq_context); + } + + return 0; +} + +static void bnxt_re_cleanup_res(struct bnxt_re_dev *rdev) +{ + if (rdev->nq.hwq.max_elements) + bnxt_qplib_disable_nq(&rdev->nq); + + if (rdev->qplib_res.rcfw) + bnxt_qplib_cleanup_res(&rdev->qplib_res); +} + +static int bnxt_re_init_res(struct bnxt_re_dev *rdev) +{ + int rc = 0; + + bnxt_qplib_init_res(&rdev->qplib_res); + + if (rdev->msix_entries[BNXT_RE_NQ_IDX].vector <= 0) + return -EINVAL; + + rc = bnxt_qplib_enable_nq(rdev->en_dev->pdev, &rdev->nq, + rdev->msix_entries[BNXT_RE_NQ_IDX].vector, + rdev->msix_entries[BNXT_RE_NQ_IDX].db_offset, + &bnxt_re_cqn_handler, + NULL); + + if (rc) + dev_err(rdev_to_dev(rdev), "Failed to enable NQ: %#x", rc); + + return rc; +} + +static void bnxt_re_free_res(struct bnxt_re_dev *rdev, bool lock_wait) +{ + if (rdev->nq.hwq.max_elements) { + bnxt_re_net_ring_free(rdev, rdev->nq.ring_id, lock_wait); + bnxt_qplib_free_nq(&rdev->nq); + } + if (rdev->qplib_res.dpi_tbl.max) { + bnxt_qplib_dealloc_dpi(&rdev->qplib_res, + &rdev->qplib_res.dpi_tbl, + &rdev->dpi_privileged); + } + if (rdev->qplib_res.rcfw) { + bnxt_qplib_free_res(&rdev->qplib_res); + rdev->qplib_res.rcfw = NULL; + } +} + +static int bnxt_re_alloc_res(struct bnxt_re_dev *rdev) +{ + int rc = 0; + + /* Configure and allocate resources for qplib */ + rdev->qplib_res.rcfw = &rdev->rcfw; + rc = bnxt_qplib_get_dev_attr(&rdev->rcfw, &rdev->dev_attr); + if (rc) + goto fail; + + rc = bnxt_qplib_alloc_res(&rdev->qplib_res, rdev->en_dev->pdev, + rdev->netdev, &rdev->dev_attr); + if (rc) + goto fail; + + rc = bnxt_qplib_alloc_dpi(&rdev->qplib_res.dpi_tbl, + &rdev->dpi_privileged, + rdev); + if (rc) + goto fail; + + rdev->nq.hwq.max_elements = BNXT_RE_MAX_CQ_COUNT + + BNXT_RE_MAX_SRQC_COUNT + 2; + rc = bnxt_qplib_alloc_nq(rdev->en_dev->pdev, &rdev->nq); + if (rc) { + dev_err(rdev_to_dev(rdev), + "Failed to allocate NQ memory: %#x", rc); + goto fail; + } + rc = bnxt_re_net_ring_alloc + (rdev, rdev->nq.hwq.pbl[PBL_LVL_0].pg_map_arr, + rdev->nq.hwq.pbl[rdev->nq.hwq.level].pg_count, + HWRM_RING_ALLOC_CMPL, BNXT_QPLIB_NQE_MAX_CNT - 1, + rdev->msix_entries[BNXT_RE_NQ_IDX].ring_idx, + &rdev->nq.ring_id); + if (rc) { + dev_err(rdev_to_dev(rdev), + "Failed to allocate NQ ring: %#x", rc); + goto free_nq; + } + return 0; +free_nq: + bnxt_qplib_free_nq(&rdev->nq); +fail: + rdev->qplib_res.rcfw = NULL; + return rc; +} + +static void bnxt_re_dispatch_event(struct ib_device *ibdev, struct ib_qp *qp, + u8 port_num, enum ib_event_type event) +{ + struct ib_event ib_event; + + ib_event.device = ibdev; + if (qp) + ib_event.element.qp = qp; + else + ib_event.element.port_num = port_num; + ib_event.event = event; + ib_dispatch_event(&ib_event); +} + +#define HWRM_QUEUE_PRI2COS_QCFG_INPUT_FLAGS_IVLAN 0x02 +static int bnxt_re_query_hwrm_pri2cos(struct bnxt_re_dev *rdev, u8 dir, + u64 *cid_map) +{ + struct hwrm_queue_pri2cos_qcfg_input req = {0}; + struct bnxt *bp = netdev_priv(rdev->netdev); + struct hwrm_queue_pri2cos_qcfg_output resp; + struct bnxt_en_dev *en_dev = rdev->en_dev; + struct bnxt_fw_msg fw_msg; + u32 flags = 0; + u8 *qcfgmap, *tmp_map; + int rc = 0, i; + + if (!cid_map) + return -EINVAL; + + memset(&fw_msg, 0, sizeof(fw_msg)); + bnxt_re_init_hwrm_hdr(rdev, (void *)&req, + HWRM_QUEUE_PRI2COS_QCFG, -1, -1); + flags |= (dir & 0x01); + flags |= HWRM_QUEUE_PRI2COS_QCFG_INPUT_FLAGS_IVLAN; + req.flags = cpu_to_le32(flags); + req.port_id = bp->pf.port_id; + + bnxt_re_fill_fw_msg(&fw_msg, (void *)&req, sizeof(req), (void *)&resp, + sizeof(resp), DFLT_HWRM_CMD_TIMEOUT); + rc = en_dev->en_ops->bnxt_send_fw_msg(en_dev, BNXT_ROCE_ULP, &fw_msg); + if (rc) + return rc; + + if (resp.queue_cfg_info) { + dev_warn(rdev_to_dev(rdev), + "Asymmetric cos queue configuration detected"); + dev_warn(rdev_to_dev(rdev), + " on device, QoS may not be fully functional\n"); + } + qcfgmap = &resp.pri0_cos_queue_id; + tmp_map = (u8 *)cid_map; + for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) + tmp_map[i] = qcfgmap[i]; + + return rc; +} + +static bool bnxt_re_is_qp1_or_shadow_qp(struct bnxt_re_dev *rdev, + struct bnxt_re_qp *qp) +{ + return (qp->ib_qp.qp_type == IB_QPT_GSI) || (qp == rdev->qp1_sqp); +} + +static void bnxt_re_dev_stop(struct bnxt_re_dev *rdev) +{ + int mask = IB_QP_STATE; + struct ib_qp_attr qp_attr; + struct bnxt_re_qp *qp; + + qp_attr.qp_state = IB_QPS_ERR; + mutex_lock(&rdev->qp_lock); + list_for_each_entry(qp, &rdev->qp_list, list) { + /* Modify the state of all QPs except QP1/Shadow QP */ + if (!bnxt_re_is_qp1_or_shadow_qp(rdev, qp)) { + if (qp->qplib_qp.state != + CMDQ_MODIFY_QP_NEW_STATE_RESET && + qp->qplib_qp.state != + CMDQ_MODIFY_QP_NEW_STATE_ERR) { + bnxt_re_dispatch_event(&rdev->ibdev, &qp->ib_qp, + 1, IB_EVENT_QP_FATAL); + bnxt_re_modify_qp(&qp->ib_qp, &qp_attr, mask, + NULL); + } + } + } + mutex_unlock(&rdev->qp_lock); +} + +static u32 bnxt_re_get_priority_mask(struct bnxt_re_dev *rdev) +{ + u32 prio_map = 0, tmp_map = 0; + struct net_device *netdev; + struct dcb_app app; + + netdev = rdev->netdev; + + memset(&app, 0, sizeof(app)); + app.selector = IEEE_8021QAZ_APP_SEL_ETHERTYPE; + app.protocol = ETH_P_IBOE; + tmp_map = dcb_ieee_getapp_mask(netdev, &app); + prio_map = tmp_map; + + app.selector = IEEE_8021QAZ_APP_SEL_DGRAM; + app.protocol = ROCE_V2_UDP_DPORT; + tmp_map = dcb_ieee_getapp_mask(netdev, &app); + prio_map |= tmp_map; + + if (!prio_map) + prio_map = -EFAULT; + return prio_map; +} + +static void bnxt_re_parse_cid_map(u8 prio_map, u8 *cid_map, u16 *cosq) +{ + u16 prio; + u8 id; + + for (prio = 0, id = 0; prio < 8; prio++) { + if (prio_map & (1 << prio)) { + cosq[id] = cid_map[prio]; + id++; + if (id == 2) /* Max 2 tcs supported */ + break; + } + } +} + +static int bnxt_re_setup_qos(struct bnxt_re_dev *rdev) +{ + u8 prio_map = 0; + u64 cid_map; + int rc; + + /* Get priority for roce */ + rc = bnxt_re_get_priority_mask(rdev); + if (rc < 0) + return rc; + prio_map = (u8)rc; + + if (prio_map == rdev->cur_prio_map) + return 0; + rdev->cur_prio_map = prio_map; + /* Get cosq id for this priority */ + rc = bnxt_re_query_hwrm_pri2cos(rdev, 0, &cid_map); + if (rc) { + dev_warn(rdev_to_dev(rdev), "no cos for p_mask %x\n", prio_map); + return rc; + } + /* Parse CoS IDs for app priority */ + bnxt_re_parse_cid_map(prio_map, (u8 *)&cid_map, rdev->cosq); + + /* Config BONO. */ + rc = bnxt_qplib_map_tc2cos(&rdev->qplib_res, rdev->cosq); + if (rc) { + dev_warn(rdev_to_dev(rdev), "no tc for cos{%x, %x}\n", + rdev->cosq[0], rdev->cosq[1]); + return rc; + } + + return 0; +} + +static void bnxt_re_ib_unreg(struct bnxt_re_dev *rdev, bool lock_wait) +{ + int i, rc; + + if (test_and_clear_bit(BNXT_RE_FLAG_IBDEV_REGISTERED, &rdev->flags)) { + for (i = 0; i < ARRAY_SIZE(bnxt_re_attributes); i++) + device_remove_file(&rdev->ibdev.dev, + bnxt_re_attributes[i]); + /* Cleanup ib dev */ + bnxt_re_unregister_ib(rdev); + } + if (test_and_clear_bit(BNXT_RE_FLAG_QOS_WORK_REG, &rdev->flags)) + cancel_delayed_work(&rdev->worker); + + bnxt_re_cleanup_res(rdev); + bnxt_re_free_res(rdev, lock_wait); + + if (test_and_clear_bit(BNXT_RE_FLAG_RCFW_CHANNEL_EN, &rdev->flags)) { + rc = bnxt_qplib_deinit_rcfw(&rdev->rcfw); + if (rc) + dev_warn(rdev_to_dev(rdev), + "Failed to deinitialize RCFW: %#x", rc); + bnxt_re_net_stats_ctx_free(rdev, rdev->qplib_ctx.stats.fw_id, + lock_wait); + bnxt_qplib_free_ctx(rdev->en_dev->pdev, &rdev->qplib_ctx); + bnxt_qplib_disable_rcfw_channel(&rdev->rcfw); + bnxt_re_net_ring_free(rdev, rdev->rcfw.creq_ring_id, lock_wait); + bnxt_qplib_free_rcfw_channel(&rdev->rcfw); + } + if (test_and_clear_bit(BNXT_RE_FLAG_GOT_MSIX, &rdev->flags)) { + rc = bnxt_re_free_msix(rdev, lock_wait); + if (rc) + dev_warn(rdev_to_dev(rdev), + "Failed to free MSI-X vectors: %#x", rc); + } + if (test_and_clear_bit(BNXT_RE_FLAG_NETDEV_REGISTERED, &rdev->flags)) { + rc = bnxt_re_unregister_netdev(rdev, lock_wait); + if (rc) + dev_warn(rdev_to_dev(rdev), + "Failed to unregister with netdev: %#x", rc); + } +} + +static void bnxt_re_set_resource_limits(struct bnxt_re_dev *rdev) +{ + u32 i; + + rdev->qplib_ctx.qpc_count = BNXT_RE_MAX_QPC_COUNT; + rdev->qplib_ctx.mrw_count = BNXT_RE_MAX_MRW_COUNT; + rdev->qplib_ctx.srqc_count = BNXT_RE_MAX_SRQC_COUNT; + rdev->qplib_ctx.cq_count = BNXT_RE_MAX_CQ_COUNT; + for (i = 0; i < MAX_TQM_ALLOC_REQ; i++) + rdev->qplib_ctx.tqm_count[i] = + rdev->dev_attr.tqm_alloc_reqs[i]; +} + +/* worker thread for polling periodic events. Now used for QoS programming*/ +static void bnxt_re_worker(struct work_struct *work) +{ + struct bnxt_re_dev *rdev = container_of(work, struct bnxt_re_dev, + worker.work); + + bnxt_re_setup_qos(rdev); + schedule_delayed_work(&rdev->worker, msecs_to_jiffies(30000)); +} + +static int bnxt_re_ib_reg(struct bnxt_re_dev *rdev) +{ + int i, j, rc; + + /* Registered a new RoCE device instance to netdev */ + rc = bnxt_re_register_netdev(rdev); + if (rc) { + pr_err("Failed to register with netedev: %#x\n", rc); + return -EINVAL; + } + set_bit(BNXT_RE_FLAG_NETDEV_REGISTERED, &rdev->flags); + + rc = bnxt_re_request_msix(rdev); + if (rc) { + pr_err("Failed to get MSI-X vectors: %#x\n", rc); + rc = -EINVAL; + goto fail; + } + set_bit(BNXT_RE_FLAG_GOT_MSIX, &rdev->flags); + + /* Establish RCFW Communication Channel to initialize the context + * memory for the function and all child VFs + */ + rc = bnxt_qplib_alloc_rcfw_channel(rdev->en_dev->pdev, &rdev->rcfw); + if (rc) + goto fail; + + rc = bnxt_re_net_ring_alloc + (rdev, rdev->rcfw.creq.pbl[PBL_LVL_0].pg_map_arr, + rdev->rcfw.creq.pbl[rdev->rcfw.creq.level].pg_count, + HWRM_RING_ALLOC_CMPL, BNXT_QPLIB_CREQE_MAX_CNT - 1, + rdev->msix_entries[BNXT_RE_AEQ_IDX].ring_idx, + &rdev->rcfw.creq_ring_id); + if (rc) { + pr_err("Failed to allocate CREQ: %#x\n", rc); + goto free_rcfw; + } + rc = bnxt_qplib_enable_rcfw_channel + (rdev->en_dev->pdev, &rdev->rcfw, + rdev->msix_entries[BNXT_RE_AEQ_IDX].vector, + rdev->msix_entries[BNXT_RE_AEQ_IDX].db_offset, + 0, &bnxt_re_aeq_handler); + if (rc) { + pr_err("Failed to enable RCFW channel: %#x\n", rc); + goto free_ring; + } + + rc = bnxt_qplib_get_dev_attr(&rdev->rcfw, &rdev->dev_attr); + if (rc) + goto disable_rcfw; + bnxt_re_set_resource_limits(rdev); + + rc = bnxt_qplib_alloc_ctx(rdev->en_dev->pdev, &rdev->qplib_ctx, 0); + if (rc) { + pr_err("Failed to allocate QPLIB context: %#x\n", rc); + goto disable_rcfw; + } + rc = bnxt_re_net_stats_ctx_alloc(rdev, + rdev->qplib_ctx.stats.dma_map, + &rdev->qplib_ctx.stats.fw_id); + if (rc) { + pr_err("Failed to allocate stats context: %#x\n", rc); + goto free_ctx; + } + + rc = bnxt_qplib_init_rcfw(&rdev->rcfw, &rdev->qplib_ctx, 0); + if (rc) { + pr_err("Failed to initialize RCFW: %#x\n", rc); + goto free_sctx; + } + set_bit(BNXT_RE_FLAG_RCFW_CHANNEL_EN, &rdev->flags); + + /* Resources based on the 'new' device caps */ + rc = bnxt_re_alloc_res(rdev); + if (rc) { + pr_err("Failed to allocate resources: %#x\n", rc); + goto fail; + } + rc = bnxt_re_init_res(rdev); + if (rc) { + pr_err("Failed to initialize resources: %#x\n", rc); + goto fail; + } + + rc = bnxt_re_setup_qos(rdev); + if (rc) + pr_info("RoCE priority not yet configured\n"); + + INIT_DELAYED_WORK(&rdev->worker, bnxt_re_worker); + set_bit(BNXT_RE_FLAG_QOS_WORK_REG, &rdev->flags); + schedule_delayed_work(&rdev->worker, msecs_to_jiffies(30000)); + + /* Register ib dev */ + rc = bnxt_re_register_ib(rdev); + if (rc) { + pr_err("Failed to register with IB: %#x\n", rc); + goto fail; + } + dev_info(rdev_to_dev(rdev), "Device registered successfully"); + for (i = 0; i < ARRAY_SIZE(bnxt_re_attributes); i++) { + rc = device_create_file(&rdev->ibdev.dev, + bnxt_re_attributes[i]); + if (rc) { + dev_err(rdev_to_dev(rdev), + "Failed to create IB sysfs: %#x", rc); + /* Must clean up all created device files */ + for (j = 0; j < i; j++) + device_remove_file(&rdev->ibdev.dev, + bnxt_re_attributes[j]); + bnxt_re_unregister_ib(rdev); + goto fail; + } + } + set_bit(BNXT_RE_FLAG_IBDEV_REGISTERED, &rdev->flags); + bnxt_re_dispatch_event(&rdev->ibdev, NULL, 1, IB_EVENT_PORT_ACTIVE); + bnxt_re_dispatch_event(&rdev->ibdev, NULL, 1, IB_EVENT_GID_CHANGE); + + return 0; +free_sctx: + bnxt_re_net_stats_ctx_free(rdev, rdev->qplib_ctx.stats.fw_id, true); +free_ctx: + bnxt_qplib_free_ctx(rdev->en_dev->pdev, &rdev->qplib_ctx); +disable_rcfw: + bnxt_qplib_disable_rcfw_channel(&rdev->rcfw); +free_ring: + bnxt_re_net_ring_free(rdev, rdev->rcfw.creq_ring_id, true); +free_rcfw: + bnxt_qplib_free_rcfw_channel(&rdev->rcfw); +fail: + bnxt_re_ib_unreg(rdev, true); + return rc; +} + +static void bnxt_re_dev_unreg(struct bnxt_re_dev *rdev) +{ + struct bnxt_en_dev *en_dev = rdev->en_dev; + struct net_device *netdev = rdev->netdev; + + bnxt_re_dev_remove(rdev); + + if (netdev) + bnxt_re_dev_unprobe(netdev, en_dev); +} + +static int bnxt_re_dev_reg(struct bnxt_re_dev **rdev, struct net_device *netdev) +{ + struct bnxt_en_dev *en_dev; + int rc = 0; + + if (!is_bnxt_re_dev(netdev)) + return -ENODEV; + + en_dev = bnxt_re_dev_probe(netdev); + if (IS_ERR(en_dev)) { + if (en_dev != ERR_PTR(-ENODEV)) + pr_err("%s: Failed to probe\n", ROCE_DRV_MODULE_NAME); + rc = PTR_ERR(en_dev); + goto exit; + } + *rdev = bnxt_re_dev_add(netdev, en_dev); + if (!*rdev) { + rc = -ENOMEM; + bnxt_re_dev_unprobe(netdev, en_dev); + goto exit; + } +exit: + return rc; +} + +static void bnxt_re_remove_one(struct bnxt_re_dev *rdev) +{ + pci_dev_put(rdev->en_dev->pdev); +} + +/* Handle all deferred netevents tasks */ +static void bnxt_re_task(struct work_struct *work) +{ + struct bnxt_re_work *re_work; + struct bnxt_re_dev *rdev; + int rc = 0; + + re_work = container_of(work, struct bnxt_re_work, work); + rdev = re_work->rdev; + + if (re_work->event != NETDEV_REGISTER && + !test_bit(BNXT_RE_FLAG_IBDEV_REGISTERED, &rdev->flags)) + return; + + switch (re_work->event) { + case NETDEV_REGISTER: + rc = bnxt_re_ib_reg(rdev); + if (rc) + dev_err(rdev_to_dev(rdev), + "Failed to register with IB: %#x", rc); + break; + case NETDEV_UP: + bnxt_re_dispatch_event(&rdev->ibdev, NULL, 1, + IB_EVENT_PORT_ACTIVE); + break; + case NETDEV_DOWN: + bnxt_re_dev_stop(rdev); + break; + case NETDEV_CHANGE: + if (!netif_carrier_ok(rdev->netdev)) + bnxt_re_dev_stop(rdev); + else if (netif_carrier_ok(rdev->netdev)) + bnxt_re_dispatch_event(&rdev->ibdev, NULL, 1, + IB_EVENT_PORT_ACTIVE); + break; + default: + break; + } + kfree(re_work); +} + +static void bnxt_re_init_one(struct bnxt_re_dev *rdev) +{ + pci_dev_get(rdev->en_dev->pdev); +} + +/* + * "Notifier chain callback can be invoked for the same chain from + * different CPUs at the same time". + * + * For cases when the netdev is already present, our call to the + * register_netdevice_notifier() will actually get the rtnl_lock() + * before sending NETDEV_REGISTER and (if up) NETDEV_UP + * events. + * + * But for cases when the netdev is not already present, the notifier + * chain is subjected to be invoked from different CPUs simultaneously. + * + * This is protected by the netdev_mutex. + */ +static int bnxt_re_netdev_event(struct notifier_block *notifier, + unsigned long event, void *ptr) +{ + struct net_device *real_dev, *netdev = netdev_notifier_info_to_dev(ptr); + struct bnxt_re_work *re_work; + struct bnxt_re_dev *rdev; + int rc = 0; + bool sch_work = false; + + real_dev = rdma_vlan_dev_real_dev(netdev); + if (!real_dev) + real_dev = netdev; + + rdev = bnxt_re_from_netdev(real_dev); + if (!rdev && event != NETDEV_REGISTER) + goto exit; + if (real_dev != netdev) + goto exit; + + switch (event) { + case NETDEV_REGISTER: + if (rdev) + break; + rc = bnxt_re_dev_reg(&rdev, real_dev); + if (rc == -ENODEV) + break; + if (rc) { + pr_err("Failed to register with the device %s: %#x\n", + real_dev->name, rc); + break; + } + bnxt_re_init_one(rdev); + sch_work = true; + break; + + case NETDEV_UNREGISTER: + bnxt_re_ib_unreg(rdev, false); + bnxt_re_remove_one(rdev); + bnxt_re_dev_unreg(rdev); + break; + + default: + sch_work = true; + break; + } + if (sch_work) { + /* Allocate for the deferred task */ + re_work = kzalloc(sizeof(*re_work), GFP_ATOMIC); + if (re_work) { + re_work->rdev = rdev; + re_work->event = event; + re_work->vlan_dev = (real_dev == netdev ? + NULL : netdev); + INIT_WORK(&re_work->work, bnxt_re_task); + queue_work(bnxt_re_wq, &re_work->work); + } + } + +exit: + return NOTIFY_DONE; +} + +static struct notifier_block bnxt_re_netdev_notifier = { + .notifier_call = bnxt_re_netdev_event +}; + +static int __init bnxt_re_mod_init(void) +{ + int rc = 0; + + pr_info("%s: %s", ROCE_DRV_MODULE_NAME, version); + + bnxt_re_wq = create_singlethread_workqueue("bnxt_re"); + if (!bnxt_re_wq) + return -ENOMEM; + + INIT_LIST_HEAD(&bnxt_re_dev_list); + + rc = register_netdevice_notifier(&bnxt_re_netdev_notifier); + if (rc) { + pr_err("%s: Cannot register to netdevice_notifier", + ROCE_DRV_MODULE_NAME); + goto err_netdev; + } + return 0; + +err_netdev: + destroy_workqueue(bnxt_re_wq); + + return rc; +} + +static void __exit bnxt_re_mod_exit(void) +{ + unregister_netdevice_notifier(&bnxt_re_netdev_notifier); + if (bnxt_re_wq) + destroy_workqueue(bnxt_re_wq); +} + +module_init(bnxt_re_mod_init); +module_exit(bnxt_re_mod_exit); diff --git a/drivers/infiniband/hw/bnxt_re/qplib_fp.c b/drivers/infiniband/hw/bnxt_re/qplib_fp.c new file mode 100644 index 000000000000..43d08b5e9085 --- /dev/null +++ b/drivers/infiniband/hw/bnxt_re/qplib_fp.c @@ -0,0 +1,2167 @@ +/* + * Broadcom NetXtreme-E RoCE driver. + * + * Copyright (c) 2016 - 2017, Broadcom. All rights reserved. The term + * Broadcom refers to Broadcom Limited and/or its subsidiaries. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, + * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS + * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE + * OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN + * IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * Description: Fast Path Operators + */ + +#include <linux/interrupt.h> +#include <linux/spinlock.h> +#include <linux/sched.h> +#include <linux/slab.h> +#include <linux/pci.h> +#include <linux/prefetch.h> + +#include "roce_hsi.h" + +#include "qplib_res.h" +#include "qplib_rcfw.h" +#include "qplib_sp.h" +#include "qplib_fp.h" + +static void bnxt_qplib_arm_cq_enable(struct bnxt_qplib_cq *cq); + +static void bnxt_qplib_free_qp_hdr_buf(struct bnxt_qplib_res *res, + struct bnxt_qplib_qp *qp) +{ + struct bnxt_qplib_q *rq = &qp->rq; + struct bnxt_qplib_q *sq = &qp->sq; + + if (qp->rq_hdr_buf) + dma_free_coherent(&res->pdev->dev, + rq->hwq.max_elements * qp->rq_hdr_buf_size, + qp->rq_hdr_buf, qp->rq_hdr_buf_map); + if (qp->sq_hdr_buf) + dma_free_coherent(&res->pdev->dev, + sq->hwq.max_elements * qp->sq_hdr_buf_size, + qp->sq_hdr_buf, qp->sq_hdr_buf_map); + qp->rq_hdr_buf = NULL; + qp->sq_hdr_buf = NULL; + qp->rq_hdr_buf_map = 0; + qp->sq_hdr_buf_map = 0; + qp->sq_hdr_buf_size = 0; + qp->rq_hdr_buf_size = 0; +} + +static int bnxt_qplib_alloc_qp_hdr_buf(struct bnxt_qplib_res *res, + struct bnxt_qplib_qp *qp) +{ + struct bnxt_qplib_q *rq = &qp->rq; + struct bnxt_qplib_q *sq = &qp->rq; + int rc = 0; + + if (qp->sq_hdr_buf_size && sq->hwq.max_elements) { + qp->sq_hdr_buf = dma_alloc_coherent(&res->pdev->dev, + sq->hwq.max_elements * + qp->sq_hdr_buf_size, + &qp->sq_hdr_buf_map, GFP_KERNEL); + if (!qp->sq_hdr_buf) { + rc = -ENOMEM; + dev_err(&res->pdev->dev, + "QPLIB: Failed to create sq_hdr_buf"); + goto fail; + } + } + + if (qp->rq_hdr_buf_size && rq->hwq.max_elements) { + qp->rq_hdr_buf = dma_alloc_coherent(&res->pdev->dev, + rq->hwq.max_elements * + qp->rq_hdr_buf_size, + &qp->rq_hdr_buf_map, + GFP_KERNEL); + if (!qp->rq_hdr_buf) { + rc = -ENOMEM; + dev_err(&res->pdev->dev, + "QPLIB: Failed to create rq_hdr_buf"); + goto fail; + } + } + return 0; + +fail: + bnxt_qplib_free_qp_hdr_buf(res, qp); + return rc; +} + +static void bnxt_qplib_service_nq(unsigned long data) +{ + struct bnxt_qplib_nq *nq = (struct bnxt_qplib_nq *)data; + struct bnxt_qplib_hwq *hwq = &nq->hwq; + struct nq_base *nqe, **nq_ptr; + int num_cqne_processed = 0; + u32 sw_cons, raw_cons; + u16 type; + int budget = nq->budget; + u64 q_handle; + + /* Service the NQ until empty */ + raw_cons = hwq->cons; + while (budget--) { + sw_cons = HWQ_CMP(raw_cons, hwq); + nq_ptr = (struct nq_base **)hwq->pbl_ptr; + nqe = &nq_ptr[NQE_PG(sw_cons)][NQE_IDX(sw_cons)]; + if (!NQE_CMP_VALID(nqe, raw_cons, hwq->max_elements)) + break; + + type = le16_to_cpu(nqe->info10_type) & NQ_BASE_TYPE_MASK; + switch (type) { + case NQ_BASE_TYPE_CQ_NOTIFICATION: + { + struct nq_cn *nqcne = (struct nq_cn *)nqe; + + q_handle = le32_to_cpu(nqcne->cq_handle_low); + q_handle |= (u64)le32_to_cpu(nqcne->cq_handle_high) + << 32; + bnxt_qplib_arm_cq_enable((struct bnxt_qplib_cq *) + ((unsigned long)q_handle)); + if (!nq->cqn_handler(nq, (struct bnxt_qplib_cq *) + ((unsigned long)q_handle))) + num_cqne_processed++; + else + dev_warn(&nq->pdev->dev, + "QPLIB: cqn - type 0x%x not handled", + type); + break; + } + case NQ_BASE_TYPE_DBQ_EVENT: + break; + default: + dev_warn(&nq->pdev->dev, + "QPLIB: nqe with type = 0x%x not handled", + type); + break; + } + raw_cons++; + } + if (hwq->cons != raw_cons) { + hwq->cons = raw_cons; + NQ_DB_REARM(nq->bar_reg_iomem, hwq->cons, hwq->max_elements); + } +} + +static irqreturn_t bnxt_qplib_nq_irq(int irq, void *dev_instance) +{ + struct bnxt_qplib_nq *nq = dev_instance; + struct bnxt_qplib_hwq *hwq = &nq->hwq; + struct nq_base **nq_ptr; + u32 sw_cons; + + /* Prefetch the NQ element */ + sw_cons = HWQ_CMP(hwq->cons, hwq); + nq_ptr = (struct nq_base **)nq->hwq.pbl_ptr; + prefetch(&nq_ptr[NQE_PG(sw_cons)][NQE_IDX(sw_cons)]); + + /* Fan out to CPU affinitized kthreads? */ + tasklet_schedule(&nq->worker); + + return IRQ_HANDLED; +} + +void bnxt_qplib_disable_nq(struct bnxt_qplib_nq *nq) +{ + /* Make sure the HW is stopped! */ + synchronize_irq(nq->vector); + tasklet_disable(&nq->worker); + tasklet_kill(&nq->worker); + + if (nq->requested) { + free_irq(nq->vector, nq); + nq->requested = false; + } + if (nq->bar_reg_iomem) + iounmap(nq->bar_reg_iomem); + nq->bar_reg_iomem = NULL; + + nq->cqn_handler = NULL; + nq->srqn_handler = NULL; + nq->vector = 0; +} + +int bnxt_qplib_enable_nq(struct pci_dev *pdev, struct bnxt_qplib_nq *nq, + int msix_vector, int bar_reg_offset, + int (*cqn_handler)(struct bnxt_qplib_nq *nq, + struct bnxt_qplib_cq *), + int (*srqn_handler)(struct bnxt_qplib_nq *nq, + void *, u8 event)) +{ + resource_size_t nq_base; + int rc; + + nq->pdev = pdev; + nq->vector = msix_vector; + + nq->cqn_handler = cqn_handler; + + nq->srqn_handler = srqn_handler; + + tasklet_init(&nq->worker, bnxt_qplib_service_nq, (unsigned long)nq); + + nq->requested = false; + rc = request_irq(nq->vector, bnxt_qplib_nq_irq, 0, "bnxt_qplib_nq", nq); + if (rc) { + dev_err(&nq->pdev->dev, + "Failed to request IRQ for NQ: %#x", rc); + bnxt_qplib_disable_nq(nq); + goto fail; + } + nq->requested = true; + nq->bar_reg = NQ_CONS_PCI_BAR_REGION; + nq->bar_reg_off = bar_reg_offset; + nq_base = pci_resource_start(pdev, nq->bar_reg); + if (!nq_base) { + rc = -ENOMEM; + goto fail; + } + nq->bar_reg_iomem = ioremap_nocache(nq_base + nq->bar_reg_off, 4); + if (!nq->bar_reg_iomem) { + rc = -ENOMEM; + goto fail; + } + NQ_DB_REARM(nq->bar_reg_iomem, nq->hwq.cons, nq->hwq.max_elements); + + return 0; +fail: + bnxt_qplib_disable_nq(nq); + return rc; +} + +void bnxt_qplib_free_nq(struct bnxt_qplib_nq *nq) +{ + if (nq->hwq.max_elements) + bnxt_qplib_free_hwq(nq->pdev, &nq->hwq); +} + +int bnxt_qplib_alloc_nq(struct pci_dev *pdev, struct bnxt_qplib_nq *nq) +{ + nq->pdev = pdev; + if (!nq->hwq.max_elements || + nq->hwq.max_elements > BNXT_QPLIB_NQE_MAX_CNT) + nq->hwq.max_elements = BNXT_QPLIB_NQE_MAX_CNT; + + if (bnxt_qplib_alloc_init_hwq(nq->pdev, &nq->hwq, NULL, 0, + &nq->hwq.max_elements, + BNXT_QPLIB_MAX_NQE_ENTRY_SIZE, 0, + PAGE_SIZE, HWQ_TYPE_L2_CMPL)) + return -ENOMEM; + + nq->budget = 8; + return 0; +} + +/* QP */ +int bnxt_qplib_create_qp1(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp) +{ + struct bnxt_qplib_rcfw *rcfw = res->rcfw; + struct cmdq_create_qp1 req; + struct creq_create_qp1_resp *resp; + struct bnxt_qplib_pbl *pbl; + struct bnxt_qplib_q *sq = &qp->sq; + struct bnxt_qplib_q *rq = &qp->rq; + int rc; + u16 cmd_flags = 0; + u32 qp_flags = 0; + + RCFW_CMD_PREP(req, CREATE_QP1, cmd_flags); + + /* General */ + req.type = qp->type; + req.dpi = cpu_to_le32(qp->dpi->dpi); + req.qp_handle = cpu_to_le64(qp->qp_handle); + + /* SQ */ + sq->hwq.max_elements = sq->max_wqe; + rc = bnxt_qplib_alloc_init_hwq(res->pdev, &sq->hwq, NULL, 0, + &sq->hwq.max_elements, + BNXT_QPLIB_MAX_SQE_ENTRY_SIZE, 0, + PAGE_SIZE, HWQ_TYPE_QUEUE); + if (rc) + goto exit; + + sq->swq = kcalloc(sq->hwq.max_elements, sizeof(*sq->swq), GFP_KERNEL); + if (!sq->swq) { + rc = -ENOMEM; + goto fail_sq; + } + pbl = &sq->hwq.pbl[PBL_LVL_0]; + req.sq_pbl = cpu_to_le64(pbl->pg_map_arr[0]); + req.sq_pg_size_sq_lvl = + ((sq->hwq.level & CMDQ_CREATE_QP1_SQ_LVL_MASK) + << CMDQ_CREATE_QP1_SQ_LVL_SFT) | + (pbl->pg_size == ROCE_PG_SIZE_4K ? + CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_4K : + pbl->pg_size == ROCE_PG_SIZE_8K ? + CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_8K : + pbl->pg_size == ROCE_PG_SIZE_64K ? + CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_64K : + pbl->pg_size == ROCE_PG_SIZE_2M ? + CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_2M : + pbl->pg_size == ROCE_PG_SIZE_8M ? + CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_8M : + pbl->pg_size == ROCE_PG_SIZE_1G ? + CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_1G : + CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_4K); + + if (qp->scq) + req.scq_cid = cpu_to_le32(qp->scq->id); + + qp_flags |= CMDQ_CREATE_QP1_QP_FLAGS_RESERVED_LKEY_ENABLE; + + /* RQ */ + if (rq->max_wqe) { + rq->hwq.max_elements = qp->rq.max_wqe; + rc = bnxt_qplib_alloc_init_hwq(res->pdev, &rq->hwq, NULL, 0, + &rq->hwq.max_elements, + BNXT_QPLIB_MAX_RQE_ENTRY_SIZE, 0, + PAGE_SIZE, HWQ_TYPE_QUEUE); + if (rc) + goto fail_sq; + + rq->swq = kcalloc(rq->hwq.max_elements, sizeof(*rq->swq), + GFP_KERNEL); + if (!rq->swq) { + rc = -ENOMEM; + goto fail_rq; + } + pbl = &rq->hwq.pbl[PBL_LVL_0]; + req.rq_pbl = cpu_to_le64(pbl->pg_map_arr[0]); + req.rq_pg_size_rq_lvl = + ((rq->hwq.level & CMDQ_CREATE_QP1_RQ_LVL_MASK) << + CMDQ_CREATE_QP1_RQ_LVL_SFT) | + (pbl->pg_size == ROCE_PG_SIZE_4K ? + CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_4K : + pbl->pg_size == ROCE_PG_SIZE_8K ? + CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_8K : + pbl->pg_size == ROCE_PG_SIZE_64K ? + CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_64K : + pbl->pg_size == ROCE_PG_SIZE_2M ? + CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_2M : + pbl->pg_size == ROCE_PG_SIZE_8M ? + CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_8M : + pbl->pg_size == ROCE_PG_SIZE_1G ? + CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_1G : + CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_4K); + if (qp->rcq) + req.rcq_cid = cpu_to_le32(qp->rcq->id); + } + + /* Header buffer - allow hdr_buf pass in */ + rc = bnxt_qplib_alloc_qp_hdr_buf(res, qp); + if (rc) { + rc = -ENOMEM; + goto fail; + } + req.qp_flags = cpu_to_le32(qp_flags); + req.sq_size = cpu_to_le32(sq->hwq.max_elements); + req.rq_size = cpu_to_le32(rq->hwq.max_elements); + + req.sq_fwo_sq_sge = + cpu_to_le16((sq->max_sge & CMDQ_CREATE_QP1_SQ_SGE_MASK) << + CMDQ_CREATE_QP1_SQ_SGE_SFT); + req.rq_fwo_rq_sge = + cpu_to_le16((rq->max_sge & CMDQ_CREATE_QP1_RQ_SGE_MASK) << + CMDQ_CREATE_QP1_RQ_SGE_SFT); + + req.pd_id = cpu_to_le32(qp->pd->id); + + resp = (struct creq_create_qp1_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, + NULL, 0); + if (!resp) { + dev_err(&res->pdev->dev, "QPLIB: FP: CREATE_QP1 send failed"); + rc = -EINVAL; + goto fail; + } + if (!bnxt_qplib_rcfw_wait_for_resp(rcfw, le16_to_cpu(req.cookie))) { + /* Cmd timed out */ + dev_err(&rcfw->pdev->dev, "QPLIB: FP: CREATE_QP1 timed out"); + rc = -ETIMEDOUT; + goto fail; + } + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&rcfw->pdev->dev, "QPLIB: FP: CREATE_QP1 failed "); + dev_err(&rcfw->pdev->dev, + "QPLIB: with status 0x%x cmdq 0x%x resp 0x%x", + resp->status, le16_to_cpu(req.cookie), + le16_to_cpu(resp->cookie)); + rc = -EINVAL; + goto fail; + } + qp->id = le32_to_cpu(resp->xid); + qp->cur_qp_state = CMDQ_MODIFY_QP_NEW_STATE_RESET; + sq->flush_in_progress = false; + rq->flush_in_progress = false; + + return 0; + +fail: + bnxt_qplib_free_qp_hdr_buf(res, qp); +fail_rq: + bnxt_qplib_free_hwq(res->pdev, &rq->hwq); + kfree(rq->swq); +fail_sq: + bnxt_qplib_free_hwq(res->pdev, &sq->hwq); + kfree(sq->swq); +exit: + return rc; +} + +int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp) +{ + struct bnxt_qplib_rcfw *rcfw = res->rcfw; + struct sq_send *hw_sq_send_hdr, **hw_sq_send_ptr; + struct cmdq_create_qp req; + struct creq_create_qp_resp *resp; + struct bnxt_qplib_pbl *pbl; + struct sq_psn_search **psn_search_ptr; + unsigned long int psn_search, poff = 0; + struct bnxt_qplib_q *sq = &qp->sq; + struct bnxt_qplib_q *rq = &qp->rq; + struct bnxt_qplib_hwq *xrrq; + int i, rc, req_size, psn_sz; + u16 cmd_flags = 0, max_ssge; + u32 sw_prod, qp_flags = 0; + + RCFW_CMD_PREP(req, CREATE_QP, cmd_flags); + + /* General */ + req.type = qp->type; + req.dpi = cpu_to_le32(qp->dpi->dpi); + req.qp_handle = cpu_to_le64(qp->qp_handle); + + /* SQ */ + psn_sz = (qp->type == CMDQ_CREATE_QP_TYPE_RC) ? + sizeof(struct sq_psn_search) : 0; + sq->hwq.max_elements = sq->max_wqe; + rc = bnxt_qplib_alloc_init_hwq(res->pdev, &sq->hwq, sq->sglist, + sq->nmap, &sq->hwq.max_elements, + BNXT_QPLIB_MAX_SQE_ENTRY_SIZE, + psn_sz, + PAGE_SIZE, HWQ_TYPE_QUEUE); + if (rc) + goto exit; + + sq->swq = kcalloc(sq->hwq.max_elements, sizeof(*sq->swq), GFP_KERNEL); + if (!sq->swq) { + rc = -ENOMEM; + goto fail_sq; + } + hw_sq_send_ptr = (struct sq_send **)sq->hwq.pbl_ptr; + if (psn_sz) { + psn_search_ptr = (struct sq_psn_search **) + &hw_sq_send_ptr[get_sqe_pg + (sq->hwq.max_elements)]; + psn_search = (unsigned long int) + &hw_sq_send_ptr[get_sqe_pg(sq->hwq.max_elements)] + [get_sqe_idx(sq->hwq.max_elements)]; + if (psn_search & ~PAGE_MASK) { + /* If the psn_search does not start on a page boundary, + * then calculate the offset + */ + poff = (psn_search & ~PAGE_MASK) / + BNXT_QPLIB_MAX_PSNE_ENTRY_SIZE; + } + for (i = 0; i < sq->hwq.max_elements; i++) + sq->swq[i].psn_search = + &psn_search_ptr[get_psne_pg(i + poff)] + [get_psne_idx(i + poff)]; + } + pbl = &sq->hwq.pbl[PBL_LVL_0]; + req.sq_pbl = cpu_to_le64(pbl->pg_map_arr[0]); + req.sq_pg_size_sq_lvl = + ((sq->hwq.level & CMDQ_CREATE_QP_SQ_LVL_MASK) + << CMDQ_CREATE_QP_SQ_LVL_SFT) | + (pbl->pg_size == ROCE_PG_SIZE_4K ? + CMDQ_CREATE_QP_SQ_PG_SIZE_PG_4K : + pbl->pg_size == ROCE_PG_SIZE_8K ? + CMDQ_CREATE_QP_SQ_PG_SIZE_PG_8K : + pbl->pg_size == ROCE_PG_SIZE_64K ? + CMDQ_CREATE_QP_SQ_PG_SIZE_PG_64K : + pbl->pg_size == ROCE_PG_SIZE_2M ? + CMDQ_CREATE_QP_SQ_PG_SIZE_PG_2M : + pbl->pg_size == ROCE_PG_SIZE_8M ? + CMDQ_CREATE_QP_SQ_PG_SIZE_PG_8M : + pbl->pg_size == ROCE_PG_SIZE_1G ? + CMDQ_CREATE_QP_SQ_PG_SIZE_PG_1G : + CMDQ_CREATE_QP_SQ_PG_SIZE_PG_4K); + + /* initialize all SQ WQEs to LOCAL_INVALID (sq prep for hw fetch) */ + hw_sq_send_ptr = (struct sq_send **)sq->hwq.pbl_ptr; + for (sw_prod = 0; sw_prod < sq->hwq.max_elements; sw_prod++) { + hw_sq_send_hdr = &hw_sq_send_ptr[get_sqe_pg(sw_prod)] + [get_sqe_idx(sw_prod)]; + hw_sq_send_hdr->wqe_type = SQ_BASE_WQE_TYPE_LOCAL_INVALID; + } + + if (qp->scq) + req.scq_cid = cpu_to_le32(qp->scq->id); + + qp_flags |= CMDQ_CREATE_QP_QP_FLAGS_RESERVED_LKEY_ENABLE; + qp_flags |= CMDQ_CREATE_QP_QP_FLAGS_FR_PMR_ENABLED; + if (qp->sig_type) + qp_flags |= CMDQ_CREATE_QP_QP_FLAGS_FORCE_COMPLETION; + + /* RQ */ + if (rq->max_wqe) { + rq->hwq.max_elements = rq->max_wqe; + rc = bnxt_qplib_alloc_init_hwq(res->pdev, &rq->hwq, rq->sglist, + rq->nmap, &rq->hwq.max_elements, + BNXT_QPLIB_MAX_RQE_ENTRY_SIZE, 0, + PAGE_SIZE, HWQ_TYPE_QUEUE); + if (rc) + goto fail_sq; + + rq->swq = kcalloc(rq->hwq.max_elements, sizeof(*rq->swq), + GFP_KERNEL); + if (!rq->swq) { + rc = -ENOMEM; + goto fail_rq; + } + pbl = &rq->hwq.pbl[PBL_LVL_0]; + req.rq_pbl = cpu_to_le64(pbl->pg_map_arr[0]); + req.rq_pg_size_rq_lvl = + ((rq->hwq.level & CMDQ_CREATE_QP_RQ_LVL_MASK) << + CMDQ_CREATE_QP_RQ_LVL_SFT) | + (pbl->pg_size == ROCE_PG_SIZE_4K ? + CMDQ_CREATE_QP_RQ_PG_SIZE_PG_4K : + pbl->pg_size == ROCE_PG_SIZE_8K ? + CMDQ_CREATE_QP_RQ_PG_SIZE_PG_8K : + pbl->pg_size == ROCE_PG_SIZE_64K ? + CMDQ_CREATE_QP_RQ_PG_SIZE_PG_64K : + pbl->pg_size == ROCE_PG_SIZE_2M ? + CMDQ_CREATE_QP_RQ_PG_SIZE_PG_2M : + pbl->pg_size == ROCE_PG_SIZE_8M ? + CMDQ_CREATE_QP_RQ_PG_SIZE_PG_8M : + pbl->pg_size == ROCE_PG_SIZE_1G ? + CMDQ_CREATE_QP_RQ_PG_SIZE_PG_1G : + CMDQ_CREATE_QP_RQ_PG_SIZE_PG_4K); + } + + if (qp->rcq) + req.rcq_cid = cpu_to_le32(qp->rcq->id); + req.qp_flags = cpu_to_le32(qp_flags); + req.sq_size = cpu_to_le32(sq->hwq.max_elements); + req.rq_size = cpu_to_le32(rq->hwq.max_elements); + qp->sq_hdr_buf = NULL; + qp->rq_hdr_buf = NULL; + + rc = bnxt_qplib_alloc_qp_hdr_buf(res, qp); + if (rc) + goto fail_rq; + + /* CTRL-22434: Irrespective of the requested SGE count on the SQ + * always create the QP with max send sges possible if the requested + * inline size is greater than 0. + */ + max_ssge = qp->max_inline_data ? 6 : sq->max_sge; + req.sq_fwo_sq_sge = cpu_to_le16( + ((max_ssge & CMDQ_CREATE_QP_SQ_SGE_MASK) + << CMDQ_CREATE_QP_SQ_SGE_SFT) | 0); + req.rq_fwo_rq_sge = cpu_to_le16( + ((rq->max_sge & CMDQ_CREATE_QP_RQ_SGE_MASK) + << CMDQ_CREATE_QP_RQ_SGE_SFT) | 0); + /* ORRQ and IRRQ */ + if (psn_sz) { + xrrq = &qp->orrq; + xrrq->max_elements = + ORD_LIMIT_TO_ORRQ_SLOTS(qp->max_rd_atomic); + req_size = xrrq->max_elements * + BNXT_QPLIB_MAX_ORRQE_ENTRY_SIZE + PAGE_SIZE - 1; + req_size &= ~(PAGE_SIZE - 1); + rc = bnxt_qplib_alloc_init_hwq(res->pdev, xrrq, NULL, 0, + &xrrq->max_elements, + BNXT_QPLIB_MAX_ORRQE_ENTRY_SIZE, + 0, req_size, HWQ_TYPE_CTX); + if (rc) + goto fail_buf_free; + pbl = &xrrq->pbl[PBL_LVL_0]; + req.orrq_addr = cpu_to_le64(pbl->pg_map_arr[0]); + + xrrq = &qp->irrq; + xrrq->max_elements = IRD_LIMIT_TO_IRRQ_SLOTS( + qp->max_dest_rd_atomic); + req_size = xrrq->max_elements * + BNXT_QPLIB_MAX_IRRQE_ENTRY_SIZE + PAGE_SIZE - 1; + req_size &= ~(PAGE_SIZE - 1); + + rc = bnxt_qplib_alloc_init_hwq(res->pdev, xrrq, NULL, 0, + &xrrq->max_elements, + BNXT_QPLIB_MAX_IRRQE_ENTRY_SIZE, + 0, req_size, HWQ_TYPE_CTX); + if (rc) + goto fail_orrq; + + pbl = &xrrq->pbl[PBL_LVL_0]; + req.irrq_addr = cpu_to_le64(pbl->pg_map_arr[0]); + } + req.pd_id = cpu_to_le32(qp->pd->id); + + resp = (struct creq_create_qp_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, + NULL, 0); + if (!resp) { + dev_err(&rcfw->pdev->dev, "QPLIB: FP: CREATE_QP send failed"); + rc = -EINVAL; + goto fail; + } + if (!bnxt_qplib_rcfw_wait_for_resp(rcfw, le16_to_cpu(req.cookie))) { + /* Cmd timed out */ + dev_err(&rcfw->pdev->dev, "QPLIB: FP: CREATE_QP timed out"); + rc = -ETIMEDOUT; + goto fail; + } + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&rcfw->pdev->dev, "QPLIB: FP: CREATE_QP failed "); + dev_err(&rcfw->pdev->dev, + "QPLIB: with status 0x%x cmdq 0x%x resp 0x%x", + resp->status, le16_to_cpu(req.cookie), + le16_to_cpu(resp->cookie)); + rc = -EINVAL; + goto fail; + } + qp->id = le32_to_cpu(resp->xid); + qp->cur_qp_state = CMDQ_MODIFY_QP_NEW_STATE_RESET; + sq->flush_in_progress = false; + rq->flush_in_progress = false; + + return 0; + +fail: + if (qp->irrq.max_elements) + bnxt_qplib_free_hwq(res->pdev, &qp->irrq); +fail_orrq: + if (qp->orrq.max_elements) + bnxt_qplib_free_hwq(res->pdev, &qp->orrq); +fail_buf_free: + bnxt_qplib_free_qp_hdr_buf(res, qp); +fail_rq: + bnxt_qplib_free_hwq(res->pdev, &rq->hwq); + kfree(rq->swq); +fail_sq: + bnxt_qplib_free_hwq(res->pdev, &sq->hwq); + kfree(sq->swq); +exit: + return rc; +} + +static void __modify_flags_from_init_state(struct bnxt_qplib_qp *qp) +{ + switch (qp->state) { + case CMDQ_MODIFY_QP_NEW_STATE_RTR: + /* INIT->RTR, configure the path_mtu to the default + * 2048 if not being requested + */ + if (!(qp->modify_flags & + CMDQ_MODIFY_QP_MODIFY_MASK_PATH_MTU)) { + qp->modify_flags |= + CMDQ_MODIFY_QP_MODIFY_MASK_PATH_MTU; + qp->path_mtu = + CMDQ_MODIFY_QP_PATH_MTU_MTU_2048; + } + qp->modify_flags &= + ~CMDQ_MODIFY_QP_MODIFY_MASK_VLAN_ID; + /* Bono FW require the max_dest_rd_atomic to be >= 1 */ + if (qp->max_dest_rd_atomic < 1) + qp->max_dest_rd_atomic = 1; + qp->modify_flags &= ~CMDQ_MODIFY_QP_MODIFY_MASK_SRC_MAC; + /* Bono FW 20.6.5 requires SGID_INDEX configuration */ + if (!(qp->modify_flags & + CMDQ_MODIFY_QP_MODIFY_MASK_SGID_INDEX)) { + qp->modify_flags |= + CMDQ_MODIFY_QP_MODIFY_MASK_SGID_INDEX; + qp->ah.sgid_index = 0; + } + break; + default: + break; + } +} + +static void __modify_flags_from_rtr_state(struct bnxt_qplib_qp *qp) +{ + switch (qp->state) { + case CMDQ_MODIFY_QP_NEW_STATE_RTS: + /* Bono FW requires the max_rd_atomic to be >= 1 */ + if (qp->max_rd_atomic < 1) + qp->max_rd_atomic = 1; + /* Bono FW does not allow PKEY_INDEX, + * DGID, FLOW_LABEL, SGID_INDEX, HOP_LIMIT, + * TRAFFIC_CLASS, DEST_MAC, PATH_MTU, RQ_PSN, + * MIN_RNR_TIMER, MAX_DEST_RD_ATOMIC, DEST_QP_ID + * modification + */ + qp->modify_flags &= + ~(CMDQ_MODIFY_QP_MODIFY_MASK_PKEY | + CMDQ_MODIFY_QP_MODIFY_MASK_DGID | + CMDQ_MODIFY_QP_MODIFY_MASK_FLOW_LABEL | + CMDQ_MODIFY_QP_MODIFY_MASK_SGID_INDEX | + CMDQ_MODIFY_QP_MODIFY_MASK_HOP_LIMIT | + CMDQ_MODIFY_QP_MODIFY_MASK_TRAFFIC_CLASS | + CMDQ_MODIFY_QP_MODIFY_MASK_DEST_MAC | + CMDQ_MODIFY_QP_MODIFY_MASK_PATH_MTU | + CMDQ_MODIFY_QP_MODIFY_MASK_RQ_PSN | + CMDQ_MODIFY_QP_MODIFY_MASK_MIN_RNR_TIMER | + CMDQ_MODIFY_QP_MODIFY_MASK_MAX_DEST_RD_ATOMIC | + CMDQ_MODIFY_QP_MODIFY_MASK_DEST_QP_ID); + break; + default: + break; + } +} + +static void __filter_modify_flags(struct bnxt_qplib_qp *qp) +{ + switch (qp->cur_qp_state) { + case CMDQ_MODIFY_QP_NEW_STATE_RESET: + break; + case CMDQ_MODIFY_QP_NEW_STATE_INIT: + __modify_flags_from_init_state(qp); + break; + case CMDQ_MODIFY_QP_NEW_STATE_RTR: + __modify_flags_from_rtr_state(qp); + break; + case CMDQ_MODIFY_QP_NEW_STATE_RTS: + break; + case CMDQ_MODIFY_QP_NEW_STATE_SQD: + break; + case CMDQ_MODIFY_QP_NEW_STATE_SQE: + break; + case CMDQ_MODIFY_QP_NEW_STATE_ERR: + break; + default: + break; + } +} + +int bnxt_qplib_modify_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp) +{ + struct bnxt_qplib_rcfw *rcfw = res->rcfw; + struct cmdq_modify_qp req; + struct creq_modify_qp_resp *resp; + u16 cmd_flags = 0, pkey; + u32 temp32[4]; + u32 bmask; + + RCFW_CMD_PREP(req, MODIFY_QP, cmd_flags); + + /* Filter out the qp_attr_mask based on the state->new transition */ + __filter_modify_flags(qp); + bmask = qp->modify_flags; + req.modify_mask = cpu_to_le32(qp->modify_flags); + req.qp_cid = cpu_to_le32(qp->id); + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_STATE) { + req.network_type_en_sqd_async_notify_new_state = + (qp->state & CMDQ_MODIFY_QP_NEW_STATE_MASK) | + (qp->en_sqd_async_notify ? + CMDQ_MODIFY_QP_EN_SQD_ASYNC_NOTIFY : 0); + } + req.network_type_en_sqd_async_notify_new_state |= qp->nw_type; + + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_ACCESS) + req.access = qp->access; + + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_PKEY) { + if (!bnxt_qplib_get_pkey(res, &res->pkey_tbl, + qp->pkey_index, &pkey)) + req.pkey = cpu_to_le16(pkey); + } + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_QKEY) + req.qkey = cpu_to_le32(qp->qkey); + + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_DGID) { + memcpy(temp32, qp->ah.dgid.data, sizeof(struct bnxt_qplib_gid)); + req.dgid[0] = cpu_to_le32(temp32[0]); + req.dgid[1] = cpu_to_le32(temp32[1]); + req.dgid[2] = cpu_to_le32(temp32[2]); + req.dgid[3] = cpu_to_le32(temp32[3]); + } + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_FLOW_LABEL) + req.flow_label = cpu_to_le32(qp->ah.flow_label); + + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_SGID_INDEX) + req.sgid_index = cpu_to_le16(res->sgid_tbl.hw_id + [qp->ah.sgid_index]); + + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_HOP_LIMIT) + req.hop_limit = qp->ah.hop_limit; + + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_TRAFFIC_CLASS) + req.traffic_class = qp->ah.traffic_class; + + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_DEST_MAC) + memcpy(req.dest_mac, qp->ah.dmac, 6); + + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_PATH_MTU) + req.path_mtu = qp->path_mtu; + + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_TIMEOUT) + req.timeout = qp->timeout; + + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_RETRY_CNT) + req.retry_cnt = qp->retry_cnt; + + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_RNR_RETRY) + req.rnr_retry = qp->rnr_retry; + + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_MIN_RNR_TIMER) + req.min_rnr_timer = qp->min_rnr_timer; + + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_RQ_PSN) + req.rq_psn = cpu_to_le32(qp->rq.psn); + + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_SQ_PSN) + req.sq_psn = cpu_to_le32(qp->sq.psn); + + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_MAX_RD_ATOMIC) + req.max_rd_atomic = + ORD_LIMIT_TO_ORRQ_SLOTS(qp->max_rd_atomic); + + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_MAX_DEST_RD_ATOMIC) + req.max_dest_rd_atomic = + IRD_LIMIT_TO_IRRQ_SLOTS(qp->max_dest_rd_atomic); + + req.sq_size = cpu_to_le32(qp->sq.hwq.max_elements); + req.rq_size = cpu_to_le32(qp->rq.hwq.max_elements); + req.sq_sge = cpu_to_le16(qp->sq.max_sge); + req.rq_sge = cpu_to_le16(qp->rq.max_sge); + req.max_inline_data = cpu_to_le32(qp->max_inline_data); + if (bmask & CMDQ_MODIFY_QP_MODIFY_MASK_DEST_QP_ID) + req.dest_qp_id = cpu_to_le32(qp->dest_qpn); + + req.vlan_pcp_vlan_dei_vlan_id = cpu_to_le16(qp->vlan_id); + + resp = (struct creq_modify_qp_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, + NULL, 0); + if (!resp) { + dev_err(&rcfw->pdev->dev, "QPLIB: FP: MODIFY_QP send failed"); + return -EINVAL; + } + if (!bnxt_qplib_rcfw_wait_for_resp(rcfw, le16_to_cpu(req.cookie))) { + /* Cmd timed out */ + dev_err(&rcfw->pdev->dev, "QPLIB: FP: MODIFY_QP timed out"); + return -ETIMEDOUT; + } + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&rcfw->pdev->dev, "QPLIB: FP: MODIFY_QP failed "); + dev_err(&rcfw->pdev->dev, + "QPLIB: with status 0x%x cmdq 0x%x resp 0x%x", + resp->status, le16_to_cpu(req.cookie), + le16_to_cpu(resp->cookie)); + return -EINVAL; + } + qp->cur_qp_state = qp->state; + return 0; +} + +int bnxt_qplib_query_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp) +{ + struct bnxt_qplib_rcfw *rcfw = res->rcfw; + struct cmdq_query_qp req; + struct creq_query_qp_resp *resp; + struct creq_query_qp_resp_sb *sb; + u16 cmd_flags = 0; + u32 temp32[4]; + int i; + + RCFW_CMD_PREP(req, QUERY_QP, cmd_flags); + + req.qp_cid = cpu_to_le32(qp->id); + req.resp_size = sizeof(*sb) / BNXT_QPLIB_CMDQE_UNITS; + resp = (struct creq_query_qp_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, + (void **)&sb, 0); + if (!resp) { + dev_err(&rcfw->pdev->dev, "QPLIB: FP: QUERY_QP send failed"); + return -EINVAL; + } + if (!bnxt_qplib_rcfw_wait_for_resp(rcfw, le16_to_cpu(req.cookie))) { + /* Cmd timed out */ + dev_err(&rcfw->pdev->dev, "QPLIB: FP: QUERY_QP timed out"); + return -ETIMEDOUT; + } + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&rcfw->pdev->dev, "QPLIB: FP: QUERY_QP failed "); + dev_err(&rcfw->pdev->dev, + "QPLIB: with status 0x%x cmdq 0x%x resp 0x%x", + resp->status, le16_to_cpu(req.cookie), + le16_to_cpu(resp->cookie)); + return -EINVAL; + } + /* Extract the context from the side buffer */ + qp->state = sb->en_sqd_async_notify_state & + CREQ_QUERY_QP_RESP_SB_STATE_MASK; + qp->en_sqd_async_notify = sb->en_sqd_async_notify_state & + CREQ_QUERY_QP_RESP_SB_EN_SQD_ASYNC_NOTIFY ? + true : false; + qp->access = sb->access; + qp->pkey_index = le16_to_cpu(sb->pkey); + qp->qkey = le32_to_cpu(sb->qkey); + + temp32[0] = le32_to_cpu(sb->dgid[0]); + temp32[1] = le32_to_cpu(sb->dgid[1]); + temp32[2] = le32_to_cpu(sb->dgid[2]); + temp32[3] = le32_to_cpu(sb->dgid[3]); + memcpy(qp->ah.dgid.data, temp32, sizeof(qp->ah.dgid.data)); + + qp->ah.flow_label = le32_to_cpu(sb->flow_label); + + qp->ah.sgid_index = 0; + for (i = 0; i < res->sgid_tbl.max; i++) { + if (res->sgid_tbl.hw_id[i] == le16_to_cpu(sb->sgid_index)) { + qp->ah.sgid_index = i; + break; + } + } + if (i == res->sgid_tbl.max) + dev_warn(&res->pdev->dev, "QPLIB: SGID not found??"); + + qp->ah.hop_limit = sb->hop_limit; + qp->ah.traffic_class = sb->traffic_class; + memcpy(qp->ah.dmac, sb->dest_mac, 6); + qp->ah.vlan_id = (le16_to_cpu(sb->path_mtu_dest_vlan_id) & + CREQ_QUERY_QP_RESP_SB_VLAN_ID_MASK) >> + CREQ_QUERY_QP_RESP_SB_VLAN_ID_SFT; + qp->path_mtu = (le16_to_cpu(sb->path_mtu_dest_vlan_id) & + CREQ_QUERY_QP_RESP_SB_PATH_MTU_MASK) >> + CREQ_QUERY_QP_RESP_SB_PATH_MTU_SFT; + qp->timeout = sb->timeout; + qp->retry_cnt = sb->retry_cnt; + qp->rnr_retry = sb->rnr_retry; + qp->min_rnr_timer = sb->min_rnr_timer; + qp->rq.psn = le32_to_cpu(sb->rq_psn); + qp->max_rd_atomic = ORRQ_SLOTS_TO_ORD_LIMIT(sb->max_rd_atomic); + qp->sq.psn = le32_to_cpu(sb->sq_psn); + qp->max_dest_rd_atomic = + IRRQ_SLOTS_TO_IRD_LIMIT(sb->max_dest_rd_atomic); + qp->sq.max_wqe = qp->sq.hwq.max_elements; + qp->rq.max_wqe = qp->rq.hwq.max_elements; + qp->sq.max_sge = le16_to_cpu(sb->sq_sge); + qp->rq.max_sge = le16_to_cpu(sb->rq_sge); + qp->max_inline_data = le32_to_cpu(sb->max_inline_data); + qp->dest_qpn = le32_to_cpu(sb->dest_qp_id); + memcpy(qp->smac, sb->src_mac, 6); + qp->vlan_id = le16_to_cpu(sb->vlan_pcp_vlan_dei_vlan_id); + return 0; +} + +static void __clean_cq(struct bnxt_qplib_cq *cq, u64 qp) +{ + struct bnxt_qplib_hwq *cq_hwq = &cq->hwq; + struct cq_base *hw_cqe, **hw_cqe_ptr; + int i; + + for (i = 0; i < cq_hwq->max_elements; i++) { + hw_cqe_ptr = (struct cq_base **)cq_hwq->pbl_ptr; + hw_cqe = &hw_cqe_ptr[CQE_PG(i)][CQE_IDX(i)]; + if (!CQE_CMP_VALID(hw_cqe, i, cq_hwq->max_elements)) + continue; + switch (hw_cqe->cqe_type_toggle & CQ_BASE_CQE_TYPE_MASK) { + case CQ_BASE_CQE_TYPE_REQ: + case CQ_BASE_CQE_TYPE_TERMINAL: + { + struct cq_req *cqe = (struct cq_req *)hw_cqe; + + if (qp == le64_to_cpu(cqe->qp_handle)) + cqe->qp_handle = 0; + break; + } + case CQ_BASE_CQE_TYPE_RES_RC: + case CQ_BASE_CQE_TYPE_RES_UD: + case CQ_BASE_CQE_TYPE_RES_RAWETH_QP1: + { + struct cq_res_rc *cqe = (struct cq_res_rc *)hw_cqe; + + if (qp == le64_to_cpu(cqe->qp_handle)) + cqe->qp_handle = 0; + break; + } + default: + break; + } + } +} + +int bnxt_qplib_destroy_qp(struct bnxt_qplib_res *res, + struct bnxt_qplib_qp *qp) +{ + struct bnxt_qplib_rcfw *rcfw = res->rcfw; + struct cmdq_destroy_qp req; + struct creq_destroy_qp_resp *resp; + unsigned long flags; + u16 cmd_flags = 0; + + RCFW_CMD_PREP(req, DESTROY_QP, cmd_flags); + + req.qp_cid = cpu_to_le32(qp->id); + resp = (struct creq_destroy_qp_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, + NULL, 0); + if (!resp) { + dev_err(&rcfw->pdev->dev, "QPLIB: FP: DESTROY_QP send failed"); + return -EINVAL; + } + if (!bnxt_qplib_rcfw_wait_for_resp(rcfw, le16_to_cpu(req.cookie))) { + /* Cmd timed out */ + dev_err(&rcfw->pdev->dev, "QPLIB: FP: DESTROY_QP timed out"); + return -ETIMEDOUT; + } + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&rcfw->pdev->dev, "QPLIB: FP: DESTROY_QP failed "); + dev_err(&rcfw->pdev->dev, + "QPLIB: with status 0x%x cmdq 0x%x resp 0x%x", + resp->status, le16_to_cpu(req.cookie), + le16_to_cpu(resp->cookie)); + return -EINVAL; + } + + /* Must walk the associated CQs to nullified the QP ptr */ + spin_lock_irqsave(&qp->scq->hwq.lock, flags); + + __clean_cq(qp->scq, (u64)(unsigned long)qp); + + if (qp->rcq && qp->rcq != qp->scq) { + spin_lock(&qp->rcq->hwq.lock); + __clean_cq(qp->rcq, (u64)(unsigned long)qp); + spin_unlock(&qp->rcq->hwq.lock); + } + + spin_unlock_irqrestore(&qp->scq->hwq.lock, flags); + + bnxt_qplib_free_qp_hdr_buf(res, qp); + bnxt_qplib_free_hwq(res->pdev, &qp->sq.hwq); + kfree(qp->sq.swq); + + bnxt_qplib_free_hwq(res->pdev, &qp->rq.hwq); + kfree(qp->rq.swq); + + if (qp->irrq.max_elements) + bnxt_qplib_free_hwq(res->pdev, &qp->irrq); + if (qp->orrq.max_elements) + bnxt_qplib_free_hwq(res->pdev, &qp->orrq); + + return 0; +} + +void *bnxt_qplib_get_qp1_sq_buf(struct bnxt_qplib_qp *qp, + struct bnxt_qplib_sge *sge) +{ + struct bnxt_qplib_q *sq = &qp->sq; + u32 sw_prod; + + memset(sge, 0, sizeof(*sge)); + + if (qp->sq_hdr_buf) { + sw_prod = HWQ_CMP(sq->hwq.prod, &sq->hwq); + sge->addr = (dma_addr_t)(qp->sq_hdr_buf_map + + sw_prod * qp->sq_hdr_buf_size); + sge->lkey = 0xFFFFFFFF; + sge->size = qp->sq_hdr_buf_size; + return qp->sq_hdr_buf + sw_prod * sge->size; + } + return NULL; +} + +u32 bnxt_qplib_get_rq_prod_index(struct bnxt_qplib_qp *qp) +{ + struct bnxt_qplib_q *rq = &qp->rq; + + return HWQ_CMP(rq->hwq.prod, &rq->hwq); +} + +dma_addr_t bnxt_qplib_get_qp_buf_from_index(struct bnxt_qplib_qp *qp, u32 index) +{ + return (qp->rq_hdr_buf_map + index * qp->rq_hdr_buf_size); +} + +void *bnxt_qplib_get_qp1_rq_buf(struct bnxt_qplib_qp *qp, + struct bnxt_qplib_sge *sge) +{ + struct bnxt_qplib_q *rq = &qp->rq; + u32 sw_prod; + + memset(sge, 0, sizeof(*sge)); + + if (qp->rq_hdr_buf) { + sw_prod = HWQ_CMP(rq->hwq.prod, &rq->hwq); + sge->addr = (dma_addr_t)(qp->rq_hdr_buf_map + + sw_prod * qp->rq_hdr_buf_size); + sge->lkey = 0xFFFFFFFF; + sge->size = qp->rq_hdr_buf_size; + return qp->rq_hdr_buf + sw_prod * sge->size; + } + return NULL; +} + +void bnxt_qplib_post_send_db(struct bnxt_qplib_qp *qp) +{ + struct bnxt_qplib_q *sq = &qp->sq; + struct dbr_dbr db_msg = { 0 }; + u32 sw_prod; + + sw_prod = HWQ_CMP(sq->hwq.prod, &sq->hwq); + + db_msg.index = cpu_to_le32((sw_prod << DBR_DBR_INDEX_SFT) & + DBR_DBR_INDEX_MASK); + db_msg.type_xid = + cpu_to_le32(((qp->id << DBR_DBR_XID_SFT) & DBR_DBR_XID_MASK) | + DBR_DBR_TYPE_SQ); + /* Flush all the WQE writes to HW */ + wmb(); + __iowrite64_copy(qp->dpi->dbr, &db_msg, sizeof(db_msg) / sizeof(u64)); +} + +int bnxt_qplib_post_send(struct bnxt_qplib_qp *qp, + struct bnxt_qplib_swqe *wqe) +{ + struct bnxt_qplib_q *sq = &qp->sq; + struct bnxt_qplib_swq *swq; + struct sq_send *hw_sq_send_hdr, **hw_sq_send_ptr; + struct sq_sge *hw_sge; + u32 sw_prod; + u8 wqe_size16; + int i, rc = 0, data_len = 0, pkt_num = 0; + __le32 temp32; + + if (qp->state != CMDQ_MODIFY_QP_NEW_STATE_RTS) { + rc = -EINVAL; + goto done; + } + if (HWQ_CMP((sq->hwq.prod + 1), &sq->hwq) == + HWQ_CMP(sq->hwq.cons, &sq->hwq)) { + rc = -ENOMEM; + goto done; + } + sw_prod = HWQ_CMP(sq->hwq.prod, &sq->hwq); + swq = &sq->swq[sw_prod]; + swq->wr_id = wqe->wr_id; + swq->type = wqe->type; + swq->flags = wqe->flags; + if (qp->sig_type) + swq->flags |= SQ_SEND_FLAGS_SIGNAL_COMP; + swq->start_psn = sq->psn & BTH_PSN_MASK; + + hw_sq_send_ptr = (struct sq_send **)sq->hwq.pbl_ptr; + hw_sq_send_hdr = &hw_sq_send_ptr[get_sqe_pg(sw_prod)] + [get_sqe_idx(sw_prod)]; + + memset(hw_sq_send_hdr, 0, BNXT_QPLIB_MAX_SQE_ENTRY_SIZE); + + if (wqe->flags & BNXT_QPLIB_SWQE_FLAGS_INLINE) { + /* Copy the inline data */ + if (wqe->inline_len > BNXT_QPLIB_SWQE_MAX_INLINE_LENGTH) { + dev_warn(&sq->hwq.pdev->dev, + "QPLIB: Inline data length > 96 detected"); + data_len = BNXT_QPLIB_SWQE_MAX_INLINE_LENGTH; + } else { + data_len = wqe->inline_len; + } + memcpy(hw_sq_send_hdr->data, wqe->inline_data, data_len); + wqe_size16 = (data_len + 15) >> 4; + } else { + for (i = 0, hw_sge = (struct sq_sge *)hw_sq_send_hdr->data; + i < wqe->num_sge; i++, hw_sge++) { + hw_sge->va_or_pa = cpu_to_le64(wqe->sg_list[i].addr); + hw_sge->l_key = cpu_to_le32(wqe->sg_list[i].lkey); + hw_sge->size = cpu_to_le32(wqe->sg_list[i].size); + data_len += wqe->sg_list[i].size; + } + /* Each SGE entry = 1 WQE size16 */ + wqe_size16 = wqe->num_sge; + } + + /* Specifics */ + switch (wqe->type) { + case BNXT_QPLIB_SWQE_TYPE_SEND: + if (qp->type == CMDQ_CREATE_QP1_TYPE_GSI) { + /* Assemble info for Raw Ethertype QPs */ + struct sq_send_raweth_qp1 *sqe = + (struct sq_send_raweth_qp1 *)hw_sq_send_hdr; + + sqe->wqe_type = wqe->type; + sqe->flags = wqe->flags; + sqe->wqe_size = wqe_size16 + + ((offsetof(typeof(*sqe), data) + 15) >> 4); + sqe->cfa_action = cpu_to_le16(wqe->rawqp1.cfa_action); + sqe->lflags = cpu_to_le16(wqe->rawqp1.lflags); + sqe->length = cpu_to_le32(data_len); + sqe->cfa_meta = cpu_to_le32((wqe->rawqp1.cfa_meta & + SQ_SEND_RAWETH_QP1_CFA_META_VLAN_VID_MASK) << + SQ_SEND_RAWETH_QP1_CFA_META_VLAN_VID_SFT); + + break; + } + /* else, just fall thru */ + case BNXT_QPLIB_SWQE_TYPE_SEND_WITH_IMM: + case BNXT_QPLIB_SWQE_TYPE_SEND_WITH_INV: + { + struct sq_send *sqe = (struct sq_send *)hw_sq_send_hdr; + + sqe->wqe_type = wqe->type; + sqe->flags = wqe->flags; + sqe->wqe_size = wqe_size16 + + ((offsetof(typeof(*sqe), data) + 15) >> 4); + sqe->inv_key_or_imm_data = cpu_to_le32( + wqe->send.inv_key); + if (qp->type == CMDQ_CREATE_QP_TYPE_UD) { + sqe->q_key = cpu_to_le32(wqe->send.q_key); + sqe->dst_qp = cpu_to_le32( + wqe->send.dst_qp & SQ_SEND_DST_QP_MASK); + sqe->length = cpu_to_le32(data_len); + sqe->avid = cpu_to_le32(wqe->send.avid & + SQ_SEND_AVID_MASK); + sq->psn = (sq->psn + 1) & BTH_PSN_MASK; + } else { + sqe->length = cpu_to_le32(data_len); + sqe->dst_qp = 0; + sqe->avid = 0; + if (qp->mtu) + pkt_num = (data_len + qp->mtu - 1) / qp->mtu; + if (!pkt_num) + pkt_num = 1; + sq->psn = (sq->psn + pkt_num) & BTH_PSN_MASK; + } + break; + } + case BNXT_QPLIB_SWQE_TYPE_RDMA_WRITE: + case BNXT_QPLIB_SWQE_TYPE_RDMA_WRITE_WITH_IMM: + case BNXT_QPLIB_SWQE_TYPE_RDMA_READ: + { + struct sq_rdma *sqe = (struct sq_rdma *)hw_sq_send_hdr; + + sqe->wqe_type = wqe->type; + sqe->flags = wqe->flags; + sqe->wqe_size = wqe_size16 + + ((offsetof(typeof(*sqe), data) + 15) >> 4); + sqe->imm_data = cpu_to_le32(wqe->rdma.inv_key); + sqe->length = cpu_to_le32((u32)data_len); + sqe->remote_va = cpu_to_le64(wqe->rdma.remote_va); + sqe->remote_key = cpu_to_le32(wqe->rdma.r_key); + if (qp->mtu) + pkt_num = (data_len + qp->mtu - 1) / qp->mtu; + if (!pkt_num) + pkt_num = 1; + sq->psn = (sq->psn + pkt_num) & BTH_PSN_MASK; + break; + } + case BNXT_QPLIB_SWQE_TYPE_ATOMIC_CMP_AND_SWP: + case BNXT_QPLIB_SWQE_TYPE_ATOMIC_FETCH_AND_ADD: + { + struct sq_atomic *sqe = (struct sq_atomic *)hw_sq_send_hdr; + + sqe->wqe_type = wqe->type; + sqe->flags = wqe->flags; + sqe->remote_key = cpu_to_le32(wqe->atomic.r_key); + sqe->remote_va = cpu_to_le64(wqe->atomic.remote_va); + sqe->swap_data = cpu_to_le64(wqe->atomic.swap_data); + sqe->cmp_data = cpu_to_le64(wqe->atomic.cmp_data); + if (qp->mtu) + pkt_num = (data_len + qp->mtu - 1) / qp->mtu; + if (!pkt_num) + pkt_num = 1; + sq->psn = (sq->psn + pkt_num) & BTH_PSN_MASK; + break; + } + case BNXT_QPLIB_SWQE_TYPE_LOCAL_INV: + { + struct sq_localinvalidate *sqe = + (struct sq_localinvalidate *)hw_sq_send_hdr; + + sqe->wqe_type = wqe->type; + sqe->flags = wqe->flags; + sqe->inv_l_key = cpu_to_le32(wqe->local_inv.inv_l_key); + + break; + } + case BNXT_QPLIB_SWQE_TYPE_FAST_REG_MR: + { + struct sq_fr_pmr *sqe = (struct sq_fr_pmr *)hw_sq_send_hdr; + + sqe->wqe_type = wqe->type; + sqe->flags = wqe->flags; + sqe->access_cntl = wqe->frmr.access_cntl | + SQ_FR_PMR_ACCESS_CNTL_LOCAL_WRITE; + sqe->zero_based_page_size_log = + (wqe->frmr.pg_sz_log & SQ_FR_PMR_PAGE_SIZE_LOG_MASK) << + SQ_FR_PMR_PAGE_SIZE_LOG_SFT | + (wqe->frmr.zero_based ? SQ_FR_PMR_ZERO_BASED : 0); + sqe->l_key = cpu_to_le32(wqe->frmr.l_key); + temp32 = cpu_to_le32(wqe->frmr.length); + memcpy(sqe->length, &temp32, sizeof(wqe->frmr.length)); + sqe->numlevels_pbl_page_size_log = + ((wqe->frmr.pbl_pg_sz_log << + SQ_FR_PMR_PBL_PAGE_SIZE_LOG_SFT) & + SQ_FR_PMR_PBL_PAGE_SIZE_LOG_MASK) | + ((wqe->frmr.levels << SQ_FR_PMR_NUMLEVELS_SFT) & + SQ_FR_PMR_NUMLEVELS_MASK); + + for (i = 0; i < wqe->frmr.page_list_len; i++) + wqe->frmr.pbl_ptr[i] = cpu_to_le64( + wqe->frmr.page_list[i] | + PTU_PTE_VALID); + sqe->pblptr = cpu_to_le64(wqe->frmr.pbl_dma_ptr); + sqe->va = cpu_to_le64(wqe->frmr.va); + + break; + } + case BNXT_QPLIB_SWQE_TYPE_BIND_MW: + { + struct sq_bind *sqe = (struct sq_bind *)hw_sq_send_hdr; + + sqe->wqe_type = wqe->type; + sqe->flags = wqe->flags; + sqe->access_cntl = wqe->bind.access_cntl; + sqe->mw_type_zero_based = wqe->bind.mw_type | + (wqe->bind.zero_based ? SQ_BIND_ZERO_BASED : 0); + sqe->parent_l_key = cpu_to_le32(wqe->bind.parent_l_key); + sqe->l_key = cpu_to_le32(wqe->bind.r_key); + sqe->va = cpu_to_le64(wqe->bind.va); + temp32 = cpu_to_le32(wqe->bind.length); + memcpy(&sqe->length, &temp32, sizeof(wqe->bind.length)); + break; + } + default: + /* Bad wqe, return error */ + rc = -EINVAL; + goto done; + } + swq->next_psn = sq->psn & BTH_PSN_MASK; + if (swq->psn_search) { + swq->psn_search->opcode_start_psn = cpu_to_le32( + ((swq->start_psn << SQ_PSN_SEARCH_START_PSN_SFT) & + SQ_PSN_SEARCH_START_PSN_MASK) | + ((wqe->type << SQ_PSN_SEARCH_OPCODE_SFT) & + SQ_PSN_SEARCH_OPCODE_MASK)); + swq->psn_search->flags_next_psn = cpu_to_le32( + ((swq->next_psn << SQ_PSN_SEARCH_NEXT_PSN_SFT) & + SQ_PSN_SEARCH_NEXT_PSN_MASK)); + } + + sq->hwq.prod++; +done: + return rc; +} + +void bnxt_qplib_post_recv_db(struct bnxt_qplib_qp *qp) +{ + struct bnxt_qplib_q *rq = &qp->rq; + struct dbr_dbr db_msg = { 0 }; + u32 sw_prod; + + sw_prod = HWQ_CMP(rq->hwq.prod, &rq->hwq); + db_msg.index = cpu_to_le32((sw_prod << DBR_DBR_INDEX_SFT) & + DBR_DBR_INDEX_MASK); + db_msg.type_xid = + cpu_to_le32(((qp->id << DBR_DBR_XID_SFT) & DBR_DBR_XID_MASK) | + DBR_DBR_TYPE_RQ); + + /* Flush the writes to HW Rx WQE before the ringing Rx DB */ + wmb(); + __iowrite64_copy(qp->dpi->dbr, &db_msg, sizeof(db_msg) / sizeof(u64)); +} + +int bnxt_qplib_post_recv(struct bnxt_qplib_qp *qp, + struct bnxt_qplib_swqe *wqe) +{ + struct bnxt_qplib_q *rq = &qp->rq; + struct rq_wqe *rqe, **rqe_ptr; + struct sq_sge *hw_sge; + u32 sw_prod; + int i, rc = 0; + + if (qp->state == CMDQ_MODIFY_QP_NEW_STATE_ERR) { + dev_err(&rq->hwq.pdev->dev, + "QPLIB: FP: QP (0x%x) is in the 0x%x state", + qp->id, qp->state); + rc = -EINVAL; + goto done; + } + if (HWQ_CMP((rq->hwq.prod + 1), &rq->hwq) == + HWQ_CMP(rq->hwq.cons, &rq->hwq)) { + dev_err(&rq->hwq.pdev->dev, + "QPLIB: FP: QP (0x%x) RQ is full!", qp->id); + rc = -EINVAL; + goto done; + } + sw_prod = HWQ_CMP(rq->hwq.prod, &rq->hwq); + rq->swq[sw_prod].wr_id = wqe->wr_id; + + rqe_ptr = (struct rq_wqe **)rq->hwq.pbl_ptr; + rqe = &rqe_ptr[RQE_PG(sw_prod)][RQE_IDX(sw_prod)]; + + memset(rqe, 0, BNXT_QPLIB_MAX_RQE_ENTRY_SIZE); + + /* Calculate wqe_size16 and data_len */ + for (i = 0, hw_sge = (struct sq_sge *)rqe->data; + i < wqe->num_sge; i++, hw_sge++) { + hw_sge->va_or_pa = cpu_to_le64(wqe->sg_list[i].addr); + hw_sge->l_key = cpu_to_le32(wqe->sg_list[i].lkey); + hw_sge->size = cpu_to_le32(wqe->sg_list[i].size); + } + rqe->wqe_type = wqe->type; + rqe->flags = wqe->flags; + rqe->wqe_size = wqe->num_sge + + ((offsetof(typeof(*rqe), data) + 15) >> 4); + + /* Supply the rqe->wr_id index to the wr_id_tbl for now */ + rqe->wr_id[0] = cpu_to_le32(sw_prod); + + rq->hwq.prod++; +done: + return rc; +} + +/* CQ */ + +/* Spinlock must be held */ +static void bnxt_qplib_arm_cq_enable(struct bnxt_qplib_cq *cq) +{ + struct dbr_dbr db_msg = { 0 }; + + db_msg.type_xid = + cpu_to_le32(((cq->id << DBR_DBR_XID_SFT) & DBR_DBR_XID_MASK) | + DBR_DBR_TYPE_CQ_ARMENA); + /* Flush memory writes before enabling the CQ */ + wmb(); + __iowrite64_copy(cq->dbr_base, &db_msg, sizeof(db_msg) / sizeof(u64)); +} + +static void bnxt_qplib_arm_cq(struct bnxt_qplib_cq *cq, u32 arm_type) +{ + struct bnxt_qplib_hwq *cq_hwq = &cq->hwq; + struct dbr_dbr db_msg = { 0 }; + u32 sw_cons; + + /* Ring DB */ + sw_cons = HWQ_CMP(cq_hwq->cons, cq_hwq); + db_msg.index = cpu_to_le32((sw_cons << DBR_DBR_INDEX_SFT) & + DBR_DBR_INDEX_MASK); + db_msg.type_xid = + cpu_to_le32(((cq->id << DBR_DBR_XID_SFT) & DBR_DBR_XID_MASK) | + arm_type); + /* flush memory writes before arming the CQ */ + wmb(); + __iowrite64_copy(cq->dpi->dbr, &db_msg, sizeof(db_msg) / sizeof(u64)); +} + +int bnxt_qplib_create_cq(struct bnxt_qplib_res *res, struct bnxt_qplib_cq *cq) +{ + struct bnxt_qplib_rcfw *rcfw = res->rcfw; + struct cmdq_create_cq req; + struct creq_create_cq_resp *resp; + struct bnxt_qplib_pbl *pbl; + u16 cmd_flags = 0; + int rc; + + cq->hwq.max_elements = cq->max_wqe; + rc = bnxt_qplib_alloc_init_hwq(res->pdev, &cq->hwq, cq->sghead, + cq->nmap, &cq->hwq.max_elements, + BNXT_QPLIB_MAX_CQE_ENTRY_SIZE, 0, + PAGE_SIZE, HWQ_TYPE_QUEUE); + if (rc) + goto exit; + + RCFW_CMD_PREP(req, CREATE_CQ, cmd_flags); + + if (!cq->dpi) { + dev_err(&rcfw->pdev->dev, + "QPLIB: FP: CREATE_CQ failed due to NULL DPI"); + return -EINVAL; + } + req.dpi = cpu_to_le32(cq->dpi->dpi); + req.cq_handle = cpu_to_le64(cq->cq_handle); + + req.cq_size = cpu_to_le32(cq->hwq.max_elements); + pbl = &cq->hwq.pbl[PBL_LVL_0]; + req.pg_size_lvl = cpu_to_le32( + ((cq->hwq.level & CMDQ_CREATE_CQ_LVL_MASK) << + CMDQ_CREATE_CQ_LVL_SFT) | + (pbl->pg_size == ROCE_PG_SIZE_4K ? CMDQ_CREATE_CQ_PG_SIZE_PG_4K : + pbl->pg_size == ROCE_PG_SIZE_8K ? CMDQ_CREATE_CQ_PG_SIZE_PG_8K : + pbl->pg_size == ROCE_PG_SIZE_64K ? CMDQ_CREATE_CQ_PG_SIZE_PG_64K : + pbl->pg_size == ROCE_PG_SIZE_2M ? CMDQ_CREATE_CQ_PG_SIZE_PG_2M : + pbl->pg_size == ROCE_PG_SIZE_8M ? CMDQ_CREATE_CQ_PG_SIZE_PG_8M : + pbl->pg_size == ROCE_PG_SIZE_1G ? CMDQ_CREATE_CQ_PG_SIZE_PG_1G : + CMDQ_CREATE_CQ_PG_SIZE_PG_4K)); + + req.pbl = cpu_to_le64(pbl->pg_map_arr[0]); + + req.cq_fco_cnq_id = cpu_to_le32( + (cq->cnq_hw_ring_id & CMDQ_CREATE_CQ_CNQ_ID_MASK) << + CMDQ_CREATE_CQ_CNQ_ID_SFT); + + resp = (struct creq_create_cq_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, + NULL, 0); + if (!resp) { + dev_err(&rcfw->pdev->dev, "QPLIB: FP: CREATE_CQ send failed"); + return -EINVAL; + } + if (!bnxt_qplib_rcfw_wait_for_resp(rcfw, le16_to_cpu(req.cookie))) { + /* Cmd timed out */ + dev_err(&rcfw->pdev->dev, "QPLIB: FP: CREATE_CQ timed out"); + rc = -ETIMEDOUT; + goto fail; + } + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&rcfw->pdev->dev, "QPLIB: FP: CREATE_CQ failed "); + dev_err(&rcfw->pdev->dev, + "QPLIB: with status 0x%x cmdq 0x%x resp 0x%x", + resp->status, le16_to_cpu(req.cookie), + le16_to_cpu(resp->cookie)); + rc = -EINVAL; + goto fail; + } + cq->id = le32_to_cpu(resp->xid); + cq->dbr_base = res->dpi_tbl.dbr_bar_reg_iomem; + cq->period = BNXT_QPLIB_QUEUE_START_PERIOD; + init_waitqueue_head(&cq->waitq); + + bnxt_qplib_arm_cq_enable(cq); + return 0; + +fail: + bnxt_qplib_free_hwq(res->pdev, &cq->hwq); +exit: + return rc; +} + +int bnxt_qplib_destroy_cq(struct bnxt_qplib_res *res, struct bnxt_qplib_cq *cq) +{ + struct bnxt_qplib_rcfw *rcfw = res->rcfw; + struct cmdq_destroy_cq req; + struct creq_destroy_cq_resp *resp; + u16 cmd_flags = 0; + + RCFW_CMD_PREP(req, DESTROY_CQ, cmd_flags); + + req.cq_cid = cpu_to_le32(cq->id); + resp = (struct creq_destroy_cq_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, + NULL, 0); + if (!resp) { + dev_err(&rcfw->pdev->dev, "QPLIB: FP: DESTROY_CQ send failed"); + return -EINVAL; + } + if (!bnxt_qplib_rcfw_wait_for_resp(rcfw, le16_to_cpu(req.cookie))) { + /* Cmd timed out */ + dev_err(&rcfw->pdev->dev, "QPLIB: FP: DESTROY_CQ timed out"); + return -ETIMEDOUT; + } + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&rcfw->pdev->dev, "QPLIB: FP: DESTROY_CQ failed "); + dev_err(&rcfw->pdev->dev, + "QPLIB: with status 0x%x cmdq 0x%x resp 0x%x", + resp->status, le16_to_cpu(req.cookie), + le16_to_cpu(resp->cookie)); + return -EINVAL; + } + bnxt_qplib_free_hwq(res->pdev, &cq->hwq); + return 0; +} + +static int __flush_sq(struct bnxt_qplib_q *sq, struct bnxt_qplib_qp *qp, + struct bnxt_qplib_cqe **pcqe, int *budget) +{ + u32 sw_prod, sw_cons; + struct bnxt_qplib_cqe *cqe; + int rc = 0; + + /* Now complete all outstanding SQEs with FLUSHED_ERR */ + sw_prod = HWQ_CMP(sq->hwq.prod, &sq->hwq); + cqe = *pcqe; + while (*budget) { + sw_cons = HWQ_CMP(sq->hwq.cons, &sq->hwq); + if (sw_cons == sw_prod) { + sq->flush_in_progress = false; + break; + } + memset(cqe, 0, sizeof(*cqe)); + cqe->status = CQ_REQ_STATUS_WORK_REQUEST_FLUSHED_ERR; + cqe->opcode = CQ_BASE_CQE_TYPE_REQ; + cqe->qp_handle = (u64)(unsigned long)qp; + cqe->wr_id = sq->swq[sw_cons].wr_id; + cqe->src_qp = qp->id; + cqe->type = sq->swq[sw_cons].type; + cqe++; + (*budget)--; + sq->hwq.cons++; + } + *pcqe = cqe; + if (!(*budget) && HWQ_CMP(sq->hwq.cons, &sq->hwq) != sw_prod) + /* Out of budget */ + rc = -EAGAIN; + + return rc; +} + +static int __flush_rq(struct bnxt_qplib_q *rq, struct bnxt_qplib_qp *qp, + int opcode, struct bnxt_qplib_cqe **pcqe, int *budget) +{ + struct bnxt_qplib_cqe *cqe; + u32 sw_prod, sw_cons; + int rc = 0; + + /* Flush the rest of the RQ */ + sw_prod = HWQ_CMP(rq->hwq.prod, &rq->hwq); + cqe = *pcqe; + while (*budget) { + sw_cons = HWQ_CMP(rq->hwq.cons, &rq->hwq); + if (sw_cons == sw_prod) + break; + memset(cqe, 0, sizeof(*cqe)); + cqe->status = + CQ_RES_RC_STATUS_WORK_REQUEST_FLUSHED_ERR; + cqe->opcode = opcode; + cqe->qp_handle = (unsigned long)qp; + cqe->wr_id = rq->swq[sw_cons].wr_id; + cqe++; + (*budget)--; + rq->hwq.cons++; + } + *pcqe = cqe; + if (!*budget && HWQ_CMP(rq->hwq.cons, &rq->hwq) != sw_prod) + /* Out of budget */ + rc = -EAGAIN; + + return rc; +} + +static int bnxt_qplib_cq_process_req(struct bnxt_qplib_cq *cq, + struct cq_req *hwcqe, + struct bnxt_qplib_cqe **pcqe, int *budget) +{ + struct bnxt_qplib_qp *qp; + struct bnxt_qplib_q *sq; + struct bnxt_qplib_cqe *cqe; + u32 sw_cons, cqe_cons; + int rc = 0; + + qp = (struct bnxt_qplib_qp *)((unsigned long) + le64_to_cpu(hwcqe->qp_handle)); + if (!qp) { + dev_err(&cq->hwq.pdev->dev, + "QPLIB: FP: Process Req qp is NULL"); + return -EINVAL; + } + sq = &qp->sq; + + cqe_cons = HWQ_CMP(le16_to_cpu(hwcqe->sq_cons_idx), &sq->hwq); + if (cqe_cons > sq->hwq.max_elements) { + dev_err(&cq->hwq.pdev->dev, + "QPLIB: FP: CQ Process req reported "); + dev_err(&cq->hwq.pdev->dev, + "QPLIB: sq_cons_idx 0x%x which exceeded max 0x%x", + cqe_cons, sq->hwq.max_elements); + return -EINVAL; + } + /* If we were in the middle of flushing the SQ, continue */ + if (sq->flush_in_progress) + goto flush; + + /* Require to walk the sq's swq to fabricate CQEs for all previously + * signaled SWQEs due to CQE aggregation from the current sq cons + * to the cqe_cons + */ + cqe = *pcqe; + while (*budget) { + sw_cons = HWQ_CMP(sq->hwq.cons, &sq->hwq); + if (sw_cons == cqe_cons) + break; + memset(cqe, 0, sizeof(*cqe)); + cqe->opcode = CQ_BASE_CQE_TYPE_REQ; + cqe->qp_handle = (u64)(unsigned long)qp; + cqe->src_qp = qp->id; + cqe->wr_id = sq->swq[sw_cons].wr_id; + cqe->type = sq->swq[sw_cons].type; + + /* For the last CQE, check for status. For errors, regardless + * of the request being signaled or not, it must complete with + * the hwcqe error status + */ + if (HWQ_CMP((sw_cons + 1), &sq->hwq) == cqe_cons && + hwcqe->status != CQ_REQ_STATUS_OK) { + cqe->status = hwcqe->status; + dev_err(&cq->hwq.pdev->dev, + "QPLIB: FP: CQ Processed Req "); + dev_err(&cq->hwq.pdev->dev, + "QPLIB: wr_id[%d] = 0x%llx with status 0x%x", + sw_cons, cqe->wr_id, cqe->status); + cqe++; + (*budget)--; + sq->flush_in_progress = true; + /* Must block new posting of SQ and RQ */ + qp->state = CMDQ_MODIFY_QP_NEW_STATE_ERR; + } else { + if (sq->swq[sw_cons].flags & + SQ_SEND_FLAGS_SIGNAL_COMP) { + cqe->status = CQ_REQ_STATUS_OK; + cqe++; + (*budget)--; + } + } + sq->hwq.cons++; + } + *pcqe = cqe; + if (!*budget && HWQ_CMP(sq->hwq.cons, &sq->hwq) != cqe_cons) { + /* Out of budget */ + rc = -EAGAIN; + goto done; + } + if (!sq->flush_in_progress) + goto done; +flush: + /* Require to walk the sq's swq to fabricate CQEs for all + * previously posted SWQEs due to the error CQE received + */ + rc = __flush_sq(sq, qp, pcqe, budget); + if (!rc) + sq->flush_in_progress = false; +done: + return rc; +} + +static int bnxt_qplib_cq_process_res_rc(struct bnxt_qplib_cq *cq, + struct cq_res_rc *hwcqe, + struct bnxt_qplib_cqe **pcqe, + int *budget) +{ + struct bnxt_qplib_qp *qp; + struct bnxt_qplib_q *rq; + struct bnxt_qplib_cqe *cqe; + u32 wr_id_idx; + int rc = 0; + + qp = (struct bnxt_qplib_qp *)((unsigned long) + le64_to_cpu(hwcqe->qp_handle)); + if (!qp) { + dev_err(&cq->hwq.pdev->dev, "QPLIB: process_cq RC qp is NULL"); + return -EINVAL; + } + cqe = *pcqe; + cqe->opcode = hwcqe->cqe_type_toggle & CQ_BASE_CQE_TYPE_MASK; + cqe->length = le32_to_cpu(hwcqe->length); + cqe->invrkey = le32_to_cpu(hwcqe->imm_data_or_inv_r_key); + cqe->mr_handle = le64_to_cpu(hwcqe->mr_handle); + cqe->flags = le16_to_cpu(hwcqe->flags); + cqe->status = hwcqe->status; + cqe->qp_handle = (u64)(unsigned long)qp; + + wr_id_idx = le32_to_cpu(hwcqe->srq_or_rq_wr_id) & + CQ_RES_RC_SRQ_OR_RQ_WR_ID_MASK; + rq = &qp->rq; + if (wr_id_idx > rq->hwq.max_elements) { + dev_err(&cq->hwq.pdev->dev, "QPLIB: FP: CQ Process RC "); + dev_err(&cq->hwq.pdev->dev, + "QPLIB: wr_id idx 0x%x exceeded RQ max 0x%x", + wr_id_idx, rq->hwq.max_elements); + return -EINVAL; + } + if (rq->flush_in_progress) + goto flush_rq; + + cqe->wr_id = rq->swq[wr_id_idx].wr_id; + cqe++; + (*budget)--; + rq->hwq.cons++; + *pcqe = cqe; + + if (hwcqe->status != CQ_RES_RC_STATUS_OK) { + rq->flush_in_progress = true; +flush_rq: + rc = __flush_rq(rq, qp, CQ_BASE_CQE_TYPE_RES_RC, pcqe, budget); + if (!rc) + rq->flush_in_progress = false; + } + return rc; +} + +static int bnxt_qplib_cq_process_res_ud(struct bnxt_qplib_cq *cq, + struct cq_res_ud *hwcqe, + struct bnxt_qplib_cqe **pcqe, + int *budget) +{ + struct bnxt_qplib_qp *qp; + struct bnxt_qplib_q *rq; + struct bnxt_qplib_cqe *cqe; + u32 wr_id_idx; + int rc = 0; + + qp = (struct bnxt_qplib_qp *)((unsigned long) + le64_to_cpu(hwcqe->qp_handle)); + if (!qp) { + dev_err(&cq->hwq.pdev->dev, "QPLIB: process_cq UD qp is NULL"); + return -EINVAL; + } + cqe = *pcqe; + cqe->opcode = hwcqe->cqe_type_toggle & CQ_BASE_CQE_TYPE_MASK; + cqe->length = le32_to_cpu(hwcqe->length); + cqe->invrkey = le32_to_cpu(hwcqe->imm_data); + cqe->flags = le16_to_cpu(hwcqe->flags); + cqe->status = hwcqe->status; + cqe->qp_handle = (u64)(unsigned long)qp; + memcpy(cqe->smac, hwcqe->src_mac, 6); + wr_id_idx = le32_to_cpu(hwcqe->src_qp_high_srq_or_rq_wr_id) + & CQ_RES_UD_SRQ_OR_RQ_WR_ID_MASK; + cqe->src_qp = le16_to_cpu(hwcqe->src_qp_low) | + ((le32_to_cpu( + hwcqe->src_qp_high_srq_or_rq_wr_id) & + CQ_RES_UD_SRC_QP_HIGH_MASK) >> 8); + + rq = &qp->rq; + if (wr_id_idx > rq->hwq.max_elements) { + dev_err(&cq->hwq.pdev->dev, "QPLIB: FP: CQ Process UD "); + dev_err(&cq->hwq.pdev->dev, + "QPLIB: wr_id idx %#x exceeded RQ max %#x", + wr_id_idx, rq->hwq.max_elements); + return -EINVAL; + } + if (rq->flush_in_progress) + goto flush_rq; + + cqe->wr_id = rq->swq[wr_id_idx].wr_id; + cqe++; + (*budget)--; + rq->hwq.cons++; + *pcqe = cqe; + + if (hwcqe->status != CQ_RES_RC_STATUS_OK) { + rq->flush_in_progress = true; +flush_rq: + rc = __flush_rq(rq, qp, CQ_BASE_CQE_TYPE_RES_UD, pcqe, budget); + if (!rc) + rq->flush_in_progress = false; + } + return rc; +} + +static int bnxt_qplib_cq_process_res_raweth_qp1(struct bnxt_qplib_cq *cq, + struct cq_res_raweth_qp1 *hwcqe, + struct bnxt_qplib_cqe **pcqe, + int *budget) +{ + struct bnxt_qplib_qp *qp; + struct bnxt_qplib_q *rq; + struct bnxt_qplib_cqe *cqe; + u32 wr_id_idx; + int rc = 0; + + qp = (struct bnxt_qplib_qp *)((unsigned long) + le64_to_cpu(hwcqe->qp_handle)); + if (!qp) { + dev_err(&cq->hwq.pdev->dev, + "QPLIB: process_cq Raw/QP1 qp is NULL"); + return -EINVAL; + } + cqe = *pcqe; + cqe->opcode = hwcqe->cqe_type_toggle & CQ_BASE_CQE_TYPE_MASK; + cqe->flags = le16_to_cpu(hwcqe->flags); + cqe->qp_handle = (u64)(unsigned long)qp; + + wr_id_idx = + le32_to_cpu(hwcqe->raweth_qp1_payload_offset_srq_or_rq_wr_id) + & CQ_RES_RAWETH_QP1_SRQ_OR_RQ_WR_ID_MASK; + cqe->src_qp = qp->id; + if (qp->id == 1 && !cqe->length) { + /* Add workaround for the length misdetection */ + cqe->length = 296; + } else { + cqe->length = le16_to_cpu(hwcqe->length); + } + cqe->pkey_index = qp->pkey_index; + memcpy(cqe->smac, qp->smac, 6); + + cqe->raweth_qp1_flags = le16_to_cpu(hwcqe->raweth_qp1_flags); + cqe->raweth_qp1_flags2 = le32_to_cpu(hwcqe->raweth_qp1_flags2); + + rq = &qp->rq; + if (wr_id_idx > rq->hwq.max_elements) { + dev_err(&cq->hwq.pdev->dev, "QPLIB: FP: CQ Process Raw/QP1 RQ wr_id "); + dev_err(&cq->hwq.pdev->dev, "QPLIB: ix 0x%x exceeded RQ max 0x%x", + wr_id_idx, rq->hwq.max_elements); + return -EINVAL; + } + if (rq->flush_in_progress) + goto flush_rq; + + cqe->wr_id = rq->swq[wr_id_idx].wr_id; + cqe++; + (*budget)--; + rq->hwq.cons++; + *pcqe = cqe; + + if (hwcqe->status != CQ_RES_RC_STATUS_OK) { + rq->flush_in_progress = true; +flush_rq: + rc = __flush_rq(rq, qp, CQ_BASE_CQE_TYPE_RES_RAWETH_QP1, pcqe, + budget); + if (!rc) + rq->flush_in_progress = false; + } + return rc; +} + +static int bnxt_qplib_cq_process_terminal(struct bnxt_qplib_cq *cq, + struct cq_terminal *hwcqe, + struct bnxt_qplib_cqe **pcqe, + int *budget) +{ + struct bnxt_qplib_qp *qp; + struct bnxt_qplib_q *sq, *rq; + struct bnxt_qplib_cqe *cqe; + u32 sw_cons = 0, cqe_cons; + int rc = 0; + u8 opcode = 0; + + /* Check the Status */ + if (hwcqe->status != CQ_TERMINAL_STATUS_OK) + dev_warn(&cq->hwq.pdev->dev, + "QPLIB: FP: CQ Process Terminal Error status = 0x%x", + hwcqe->status); + + qp = (struct bnxt_qplib_qp *)((unsigned long) + le64_to_cpu(hwcqe->qp_handle)); + if (!qp) { + dev_err(&cq->hwq.pdev->dev, + "QPLIB: FP: CQ Process terminal qp is NULL"); + return -EINVAL; + } + /* Must block new posting of SQ and RQ */ + qp->state = CMDQ_MODIFY_QP_NEW_STATE_ERR; + + sq = &qp->sq; + rq = &qp->rq; + + cqe_cons = le16_to_cpu(hwcqe->sq_cons_idx); + if (cqe_cons == 0xFFFF) + goto do_rq; + + if (cqe_cons > sq->hwq.max_elements) { + dev_err(&cq->hwq.pdev->dev, + "QPLIB: FP: CQ Process terminal reported "); + dev_err(&cq->hwq.pdev->dev, + "QPLIB: sq_cons_idx 0x%x which exceeded max 0x%x", + cqe_cons, sq->hwq.max_elements); + goto do_rq; + } + /* If we were in the middle of flushing, continue */ + if (sq->flush_in_progress) + goto flush_sq; + + /* Terminal CQE can also include aggregated successful CQEs prior. + * So we must complete all CQEs from the current sq's cons to the + * cq_cons with status OK + */ + cqe = *pcqe; + while (*budget) { + sw_cons = HWQ_CMP(sq->hwq.cons, &sq->hwq); + if (sw_cons == cqe_cons) + break; + if (sq->swq[sw_cons].flags & SQ_SEND_FLAGS_SIGNAL_COMP) { + memset(cqe, 0, sizeof(*cqe)); + cqe->status = CQ_REQ_STATUS_OK; + cqe->opcode = CQ_BASE_CQE_TYPE_REQ; + cqe->qp_handle = (u64)(unsigned long)qp; + cqe->src_qp = qp->id; + cqe->wr_id = sq->swq[sw_cons].wr_id; + cqe->type = sq->swq[sw_cons].type; + cqe++; + (*budget)--; + } + sq->hwq.cons++; + } + *pcqe = cqe; + if (!(*budget) && sw_cons != cqe_cons) { + /* Out of budget */ + rc = -EAGAIN; + goto sq_done; + } + sq->flush_in_progress = true; +flush_sq: + rc = __flush_sq(sq, qp, pcqe, budget); + if (!rc) + sq->flush_in_progress = false; +sq_done: + if (rc) + return rc; +do_rq: + cqe_cons = le16_to_cpu(hwcqe->rq_cons_idx); + if (cqe_cons == 0xFFFF) { + goto done; + } else if (cqe_cons > rq->hwq.max_elements) { + dev_err(&cq->hwq.pdev->dev, + "QPLIB: FP: CQ Processed terminal "); + dev_err(&cq->hwq.pdev->dev, + "QPLIB: reported rq_cons_idx 0x%x exceeds max 0x%x", + cqe_cons, rq->hwq.max_elements); + goto done; + } + /* Terminal CQE requires all posted RQEs to complete with FLUSHED_ERR + * from the current rq->cons to the rq->prod regardless what the + * rq->cons the terminal CQE indicates + */ + rq->flush_in_progress = true; + switch (qp->type) { + case CMDQ_CREATE_QP1_TYPE_GSI: + opcode = CQ_BASE_CQE_TYPE_RES_RAWETH_QP1; + break; + case CMDQ_CREATE_QP_TYPE_RC: + opcode = CQ_BASE_CQE_TYPE_RES_RC; + break; + case CMDQ_CREATE_QP_TYPE_UD: + opcode = CQ_BASE_CQE_TYPE_RES_UD; + break; + } + + rc = __flush_rq(rq, qp, opcode, pcqe, budget); + if (!rc) + rq->flush_in_progress = false; +done: + return rc; +} + +static int bnxt_qplib_cq_process_cutoff(struct bnxt_qplib_cq *cq, + struct cq_cutoff *hwcqe) +{ + /* Check the Status */ + if (hwcqe->status != CQ_CUTOFF_STATUS_OK) { + dev_err(&cq->hwq.pdev->dev, + "QPLIB: FP: CQ Process Cutoff Error status = 0x%x", + hwcqe->status); + return -EINVAL; + } + clear_bit(CQ_FLAGS_RESIZE_IN_PROG, &cq->flags); + wake_up_interruptible(&cq->waitq); + + return 0; +} + +int bnxt_qplib_poll_cq(struct bnxt_qplib_cq *cq, struct bnxt_qplib_cqe *cqe, + int num_cqes) +{ + struct cq_base *hw_cqe, **hw_cqe_ptr; + unsigned long flags; + u32 sw_cons, raw_cons; + int budget, rc = 0; + + spin_lock_irqsave(&cq->hwq.lock, flags); + raw_cons = cq->hwq.cons; + budget = num_cqes; + + while (budget) { + sw_cons = HWQ_CMP(raw_cons, &cq->hwq); + hw_cqe_ptr = (struct cq_base **)cq->hwq.pbl_ptr; + hw_cqe = &hw_cqe_ptr[CQE_PG(sw_cons)][CQE_IDX(sw_cons)]; + + /* Check for Valid bit */ + if (!CQE_CMP_VALID(hw_cqe, raw_cons, cq->hwq.max_elements)) + break; + + /* From the device's respective CQE format to qplib_wc*/ + switch (hw_cqe->cqe_type_toggle & CQ_BASE_CQE_TYPE_MASK) { + case CQ_BASE_CQE_TYPE_REQ: + rc = bnxt_qplib_cq_process_req(cq, + (struct cq_req *)hw_cqe, + &cqe, &budget); + break; + case CQ_BASE_CQE_TYPE_RES_RC: + rc = bnxt_qplib_cq_process_res_rc(cq, + (struct cq_res_rc *) + hw_cqe, &cqe, + &budget); + break; + case CQ_BASE_CQE_TYPE_RES_UD: + rc = bnxt_qplib_cq_process_res_ud + (cq, (struct cq_res_ud *)hw_cqe, &cqe, + &budget); + break; + case CQ_BASE_CQE_TYPE_RES_RAWETH_QP1: + rc = bnxt_qplib_cq_process_res_raweth_qp1 + (cq, (struct cq_res_raweth_qp1 *) + hw_cqe, &cqe, &budget); + break; + case CQ_BASE_CQE_TYPE_TERMINAL: + rc = bnxt_qplib_cq_process_terminal + (cq, (struct cq_terminal *)hw_cqe, + &cqe, &budget); + break; + case CQ_BASE_CQE_TYPE_CUT_OFF: + bnxt_qplib_cq_process_cutoff + (cq, (struct cq_cutoff *)hw_cqe); + /* Done processing this CQ */ + goto exit; + default: + dev_err(&cq->hwq.pdev->dev, + "QPLIB: process_cq unknown type 0x%lx", + hw_cqe->cqe_type_toggle & + CQ_BASE_CQE_TYPE_MASK); + rc = -EINVAL; + break; + } + if (rc < 0) { + if (rc == -EAGAIN) + break; + /* Error while processing the CQE, just skip to the + * next one + */ + dev_err(&cq->hwq.pdev->dev, + "QPLIB: process_cqe error rc = 0x%x", rc); + } + raw_cons++; + } + if (cq->hwq.cons != raw_cons) { + cq->hwq.cons = raw_cons; + bnxt_qplib_arm_cq(cq, DBR_DBR_TYPE_CQ); + } +exit: + spin_unlock_irqrestore(&cq->hwq.lock, flags); + return num_cqes - budget; +} + +void bnxt_qplib_req_notify_cq(struct bnxt_qplib_cq *cq, u32 arm_type) +{ + unsigned long flags; + + spin_lock_irqsave(&cq->hwq.lock, flags); + if (arm_type) + bnxt_qplib_arm_cq(cq, arm_type); + + spin_unlock_irqrestore(&cq->hwq.lock, flags); +} diff --git a/drivers/infiniband/hw/bnxt_re/qplib_fp.h b/drivers/infiniband/hw/bnxt_re/qplib_fp.h new file mode 100644 index 000000000000..f0150f8da1e3 --- /dev/null +++ b/drivers/infiniband/hw/bnxt_re/qplib_fp.h @@ -0,0 +1,439 @@ +/* + * Broadcom NetXtreme-E RoCE driver. + * + * Copyright (c) 2016 - 2017, Broadcom. All rights reserved. The term + * Broadcom refers to Broadcom Limited and/or its subsidiaries. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, + * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS + * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE + * OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN + * IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * Description: Fast Path Operators (header) + */ + +#ifndef __BNXT_QPLIB_FP_H__ +#define __BNXT_QPLIB_FP_H__ + +struct bnxt_qplib_sge { + u64 addr; + u32 lkey; + u32 size; +}; + +#define BNXT_QPLIB_MAX_SQE_ENTRY_SIZE sizeof(struct sq_send) + +#define SQE_CNT_PER_PG (PAGE_SIZE / BNXT_QPLIB_MAX_SQE_ENTRY_SIZE) +#define SQE_MAX_IDX_PER_PG (SQE_CNT_PER_PG - 1) + +static inline u32 get_sqe_pg(u32 val) +{ + return ((val & ~SQE_MAX_IDX_PER_PG) / SQE_CNT_PER_PG); +} + +static inline u32 get_sqe_idx(u32 val) +{ + return (val & SQE_MAX_IDX_PER_PG); +} + +#define BNXT_QPLIB_MAX_PSNE_ENTRY_SIZE sizeof(struct sq_psn_search) + +#define PSNE_CNT_PER_PG (PAGE_SIZE / BNXT_QPLIB_MAX_PSNE_ENTRY_SIZE) +#define PSNE_MAX_IDX_PER_PG (PSNE_CNT_PER_PG - 1) + +static inline u32 get_psne_pg(u32 val) +{ + return ((val & ~PSNE_MAX_IDX_PER_PG) / PSNE_CNT_PER_PG); +} + +static inline u32 get_psne_idx(u32 val) +{ + return (val & PSNE_MAX_IDX_PER_PG); +} + +#define BNXT_QPLIB_QP_MAX_SGL 6 + +struct bnxt_qplib_swq { + u64 wr_id; + u8 type; + u8 flags; + u32 start_psn; + u32 next_psn; + struct sq_psn_search *psn_search; +}; + +struct bnxt_qplib_swqe { + /* General */ + u64 wr_id; + u8 reqs_type; + u8 type; +#define BNXT_QPLIB_SWQE_TYPE_SEND 0 +#define BNXT_QPLIB_SWQE_TYPE_SEND_WITH_IMM 1 +#define BNXT_QPLIB_SWQE_TYPE_SEND_WITH_INV 2 +#define BNXT_QPLIB_SWQE_TYPE_RDMA_WRITE 4 +#define BNXT_QPLIB_SWQE_TYPE_RDMA_WRITE_WITH_IMM 5 +#define BNXT_QPLIB_SWQE_TYPE_RDMA_READ 6 +#define BNXT_QPLIB_SWQE_TYPE_ATOMIC_CMP_AND_SWP 8 +#define BNXT_QPLIB_SWQE_TYPE_ATOMIC_FETCH_AND_ADD 11 +#define BNXT_QPLIB_SWQE_TYPE_LOCAL_INV 12 +#define BNXT_QPLIB_SWQE_TYPE_FAST_REG_MR 13 +#define BNXT_QPLIB_SWQE_TYPE_REG_MR 13 +#define BNXT_QPLIB_SWQE_TYPE_BIND_MW 14 +#define BNXT_QPLIB_SWQE_TYPE_RECV 128 +#define BNXT_QPLIB_SWQE_TYPE_RECV_RDMA_IMM 129 + u8 flags; +#define BNXT_QPLIB_SWQE_FLAGS_SIGNAL_COMP BIT(0) +#define BNXT_QPLIB_SWQE_FLAGS_RD_ATOMIC_FENCE BIT(1) +#define BNXT_QPLIB_SWQE_FLAGS_UC_FENCE BIT(2) +#define BNXT_QPLIB_SWQE_FLAGS_SOLICIT_EVENT BIT(3) +#define BNXT_QPLIB_SWQE_FLAGS_INLINE BIT(4) + struct bnxt_qplib_sge sg_list[BNXT_QPLIB_QP_MAX_SGL]; + int num_sge; + /* Max inline data is 96 bytes */ + u32 inline_len; +#define BNXT_QPLIB_SWQE_MAX_INLINE_LENGTH 96 + u8 inline_data[BNXT_QPLIB_SWQE_MAX_INLINE_LENGTH]; + + union { + /* Send, with imm, inval key */ + struct { + union { + __be32 imm_data; + u32 inv_key; + }; + u32 q_key; + u32 dst_qp; + u16 avid; + } send; + + /* Send Raw Ethernet and QP1 */ + struct { + u16 lflags; + u16 cfa_action; + u32 cfa_meta; + } rawqp1; + + /* RDMA write, with imm, read */ + struct { + union { + __be32 imm_data; + u32 inv_key; + }; + u64 remote_va; + u32 r_key; + } rdma; + + /* Atomic cmp/swap, fetch/add */ + struct { + u64 remote_va; + u32 r_key; + u64 swap_data; + u64 cmp_data; + } atomic; + + /* Local Invalidate */ + struct { + u32 inv_l_key; + } local_inv; + + /* FR-PMR */ + struct { + u8 access_cntl; + u8 pg_sz_log; + bool zero_based; + u32 l_key; + u32 length; + u8 pbl_pg_sz_log; +#define BNXT_QPLIB_SWQE_PAGE_SIZE_4K 0 +#define BNXT_QPLIB_SWQE_PAGE_SIZE_8K 1 +#define BNXT_QPLIB_SWQE_PAGE_SIZE_64K 4 +#define BNXT_QPLIB_SWQE_PAGE_SIZE_256K 6 +#define BNXT_QPLIB_SWQE_PAGE_SIZE_1M 8 +#define BNXT_QPLIB_SWQE_PAGE_SIZE_2M 9 +#define BNXT_QPLIB_SWQE_PAGE_SIZE_4M 10 +#define BNXT_QPLIB_SWQE_PAGE_SIZE_1G 18 + u8 levels; +#define PAGE_SHIFT_4K 12 + __le64 *pbl_ptr; + dma_addr_t pbl_dma_ptr; + u64 *page_list; + u16 page_list_len; + u64 va; + } frmr; + + /* Bind */ + struct { + u8 access_cntl; +#define BNXT_QPLIB_BIND_SWQE_ACCESS_LOCAL_WRITE BIT(0) +#define BNXT_QPLIB_BIND_SWQE_ACCESS_REMOTE_READ BIT(1) +#define BNXT_QPLIB_BIND_SWQE_ACCESS_REMOTE_WRITE BIT(2) +#define BNXT_QPLIB_BIND_SWQE_ACCESS_REMOTE_ATOMIC BIT(3) +#define BNXT_QPLIB_BIND_SWQE_ACCESS_WINDOW_BIND BIT(4) + bool zero_based; + u8 mw_type; + u32 parent_l_key; + u32 r_key; + u64 va; + u32 length; + } bind; + }; +}; + +#define BNXT_QPLIB_MAX_RQE_ENTRY_SIZE sizeof(struct rq_wqe) + +#define RQE_CNT_PER_PG (PAGE_SIZE / BNXT_QPLIB_MAX_RQE_ENTRY_SIZE) +#define RQE_MAX_IDX_PER_PG (RQE_CNT_PER_PG - 1) +#define RQE_PG(x) (((x) & ~RQE_MAX_IDX_PER_PG) / RQE_CNT_PER_PG) +#define RQE_IDX(x) ((x) & RQE_MAX_IDX_PER_PG) + +struct bnxt_qplib_q { + struct bnxt_qplib_hwq hwq; + struct bnxt_qplib_swq *swq; + struct scatterlist *sglist; + u32 nmap; + u32 max_wqe; + u16 max_sge; + u32 psn; + bool flush_in_progress; +}; + +struct bnxt_qplib_qp { + struct bnxt_qplib_pd *pd; + struct bnxt_qplib_dpi *dpi; + u64 qp_handle; + u32 id; + u8 type; + u8 sig_type; + u32 modify_flags; + u8 state; + u8 cur_qp_state; + u32 max_inline_data; + u32 mtu; + u8 path_mtu; + bool en_sqd_async_notify; + u16 pkey_index; + u32 qkey; + u32 dest_qp_id; + u8 access; + u8 timeout; + u8 retry_cnt; + u8 rnr_retry; + u32 min_rnr_timer; + u32 max_rd_atomic; + u32 max_dest_rd_atomic; + u32 dest_qpn; + u8 smac[6]; + u16 vlan_id; + u8 nw_type; + struct bnxt_qplib_ah ah; + +#define BTH_PSN_MASK ((1 << 24) - 1) + /* SQ */ + struct bnxt_qplib_q sq; + /* RQ */ + struct bnxt_qplib_q rq; + /* SRQ */ + struct bnxt_qplib_srq *srq; + /* CQ */ + struct bnxt_qplib_cq *scq; + struct bnxt_qplib_cq *rcq; + /* IRRQ and ORRQ */ + struct bnxt_qplib_hwq irrq; + struct bnxt_qplib_hwq orrq; + /* Header buffer for QP1 */ + int sq_hdr_buf_size; + int rq_hdr_buf_size; +/* + * Buffer space for ETH(14), IP or GRH(40), UDP header(8) + * and ib_bth + ib_deth (20). + * Max required is 82 when RoCE V2 is enabled + */ +#define BNXT_QPLIB_MAX_QP1_SQ_HDR_SIZE_V2 86 + /* Ethernet header = 14 */ + /* ib_grh = 40 (provided by MAD) */ + /* ib_bth + ib_deth = 20 */ + /* MAD = 256 (provided by MAD) */ + /* iCRC = 4 */ +#define BNXT_QPLIB_MAX_QP1_RQ_ETH_HDR_SIZE 14 +#define BNXT_QPLIB_MAX_QP1_RQ_HDR_SIZE_V2 512 +#define BNXT_QPLIB_MAX_GRH_HDR_SIZE_IPV4 20 +#define BNXT_QPLIB_MAX_GRH_HDR_SIZE_IPV6 40 +#define BNXT_QPLIB_MAX_QP1_RQ_BDETH_HDR_SIZE 20 + void *sq_hdr_buf; + dma_addr_t sq_hdr_buf_map; + void *rq_hdr_buf; + dma_addr_t rq_hdr_buf_map; +}; + +#define BNXT_QPLIB_MAX_CQE_ENTRY_SIZE sizeof(struct cq_base) + +#define CQE_CNT_PER_PG (PAGE_SIZE / BNXT_QPLIB_MAX_CQE_ENTRY_SIZE) +#define CQE_MAX_IDX_PER_PG (CQE_CNT_PER_PG - 1) +#define CQE_PG(x) (((x) & ~CQE_MAX_IDX_PER_PG) / CQE_CNT_PER_PG) +#define CQE_IDX(x) ((x) & CQE_MAX_IDX_PER_PG) + +#define ROCE_CQE_CMP_V 0 +#define CQE_CMP_VALID(hdr, raw_cons, cp_bit) \ + (!!((hdr)->cqe_type_toggle & CQ_BASE_TOGGLE) == \ + !((raw_cons) & (cp_bit))) + +struct bnxt_qplib_cqe { + u8 status; + u8 type; + u8 opcode; + u32 length; + u64 wr_id; + union { + __be32 immdata; + u32 invrkey; + }; + u64 qp_handle; + u64 mr_handle; + u16 flags; + u8 smac[6]; + u32 src_qp; + u16 raweth_qp1_flags; + u16 raweth_qp1_errors; + u16 raweth_qp1_cfa_code; + u32 raweth_qp1_flags2; + u32 raweth_qp1_metadata; + u8 raweth_qp1_payload_offset; + u16 pkey_index; +}; + +#define BNXT_QPLIB_QUEUE_START_PERIOD 0x01 +struct bnxt_qplib_cq { + struct bnxt_qplib_dpi *dpi; + void __iomem *dbr_base; + u32 max_wqe; + u32 id; + u16 count; + u16 period; + struct bnxt_qplib_hwq hwq; + u32 cnq_hw_ring_id; + bool resize_in_progress; + struct scatterlist *sghead; + u32 nmap; + u64 cq_handle; + +#define CQ_RESIZE_WAIT_TIME_MS 500 + unsigned long flags; +#define CQ_FLAGS_RESIZE_IN_PROG 1 + wait_queue_head_t waitq; +}; + +#define BNXT_QPLIB_MAX_IRRQE_ENTRY_SIZE sizeof(struct xrrq_irrq) +#define BNXT_QPLIB_MAX_ORRQE_ENTRY_SIZE sizeof(struct xrrq_orrq) +#define IRD_LIMIT_TO_IRRQ_SLOTS(x) (2 * (x) + 2) +#define IRRQ_SLOTS_TO_IRD_LIMIT(s) (((s) >> 1) - 1) +#define ORD_LIMIT_TO_ORRQ_SLOTS(x) ((x) + 1) +#define ORRQ_SLOTS_TO_ORD_LIMIT(s) ((s) - 1) + +#define BNXT_QPLIB_MAX_NQE_ENTRY_SIZE sizeof(struct nq_base) + +#define NQE_CNT_PER_PG (PAGE_SIZE / BNXT_QPLIB_MAX_NQE_ENTRY_SIZE) +#define NQE_MAX_IDX_PER_PG (NQE_CNT_PER_PG - 1) +#define NQE_PG(x) (((x) & ~NQE_MAX_IDX_PER_PG) / NQE_CNT_PER_PG) +#define NQE_IDX(x) ((x) & NQE_MAX_IDX_PER_PG) + +#define NQE_CMP_VALID(hdr, raw_cons, cp_bit) \ + (!!(le32_to_cpu((hdr)->info63_v[0]) & NQ_BASE_V) == \ + !((raw_cons) & (cp_bit))) + +#define BNXT_QPLIB_NQE_MAX_CNT (128 * 1024) + +#define NQ_CONS_PCI_BAR_REGION 2 +#define NQ_DB_KEY_CP (0x2 << CMPL_DOORBELL_KEY_SFT) +#define NQ_DB_IDX_VALID CMPL_DOORBELL_IDX_VALID +#define NQ_DB_IRQ_DIS CMPL_DOORBELL_MASK +#define NQ_DB_CP_FLAGS_REARM (NQ_DB_KEY_CP | \ + NQ_DB_IDX_VALID) +#define NQ_DB_CP_FLAGS (NQ_DB_KEY_CP | \ + NQ_DB_IDX_VALID | \ + NQ_DB_IRQ_DIS) +#define NQ_DB_REARM(db, raw_cons, cp_bit) \ + writel(NQ_DB_CP_FLAGS_REARM | ((raw_cons) & ((cp_bit) - 1)), db) +#define NQ_DB(db, raw_cons, cp_bit) \ + writel(NQ_DB_CP_FLAGS | ((raw_cons) & ((cp_bit) - 1)), db) + +struct bnxt_qplib_nq { + struct pci_dev *pdev; + + int vector; + int budget; + bool requested; + struct tasklet_struct worker; + struct bnxt_qplib_hwq hwq; + + u16 bar_reg; + u16 bar_reg_off; + u16 ring_id; + void __iomem *bar_reg_iomem; + + int (*cqn_handler) + (struct bnxt_qplib_nq *nq, + struct bnxt_qplib_cq *cq); + int (*srqn_handler) + (struct bnxt_qplib_nq *nq, + void *srq, + u8 event); +}; + +void bnxt_qplib_disable_nq(struct bnxt_qplib_nq *nq); +int bnxt_qplib_enable_nq(struct pci_dev *pdev, struct bnxt_qplib_nq *nq, + int msix_vector, int bar_reg_offset, + int (*cqn_handler)(struct bnxt_qplib_nq *nq, + struct bnxt_qplib_cq *cq), + int (*srqn_handler)(struct bnxt_qplib_nq *nq, + void *srq, + u8 event)); +int bnxt_qplib_create_qp1(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp); +int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp); +int bnxt_qplib_modify_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp); +int bnxt_qplib_query_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp); +int bnxt_qplib_destroy_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp); +void *bnxt_qplib_get_qp1_sq_buf(struct bnxt_qplib_qp *qp, + struct bnxt_qplib_sge *sge); +void *bnxt_qplib_get_qp1_rq_buf(struct bnxt_qplib_qp *qp, + struct bnxt_qplib_sge *sge); +u32 bnxt_qplib_get_rq_prod_index(struct bnxt_qplib_qp *qp); +dma_addr_t bnxt_qplib_get_qp_buf_from_index(struct bnxt_qplib_qp *qp, + u32 index); +void bnxt_qplib_post_send_db(struct bnxt_qplib_qp *qp); +int bnxt_qplib_post_send(struct bnxt_qplib_qp *qp, + struct bnxt_qplib_swqe *wqe); +void bnxt_qplib_post_recv_db(struct bnxt_qplib_qp *qp); +int bnxt_qplib_post_recv(struct bnxt_qplib_qp *qp, + struct bnxt_qplib_swqe *wqe); +int bnxt_qplib_create_cq(struct bnxt_qplib_res *res, struct bnxt_qplib_cq *cq); +int bnxt_qplib_destroy_cq(struct bnxt_qplib_res *res, struct bnxt_qplib_cq *cq); +int bnxt_qplib_poll_cq(struct bnxt_qplib_cq *cq, struct bnxt_qplib_cqe *cqe, + int num); +void bnxt_qplib_req_notify_cq(struct bnxt_qplib_cq *cq, u32 arm_type); +void bnxt_qplib_free_nq(struct bnxt_qplib_nq *nq); +int bnxt_qplib_alloc_nq(struct pci_dev *pdev, struct bnxt_qplib_nq *nq); +#endif /* __BNXT_QPLIB_FP_H__ */ diff --git a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c new file mode 100644 index 000000000000..23fb7260662b --- /dev/null +++ b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c @@ -0,0 +1,694 @@ +/* + * Broadcom NetXtreme-E RoCE driver. + * + * Copyright (c) 2016 - 2017, Broadcom. All rights reserved. The term + * Broadcom refers to Broadcom Limited and/or its subsidiaries. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, + * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS + * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE + * OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN + * IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * Description: RDMA Controller HW interface + */ +#include <linux/interrupt.h> +#include <linux/spinlock.h> +#include <linux/pci.h> +#include <linux/prefetch.h> +#include "roce_hsi.h" +#include "qplib_res.h" +#include "qplib_rcfw.h" +static void bnxt_qplib_service_creq(unsigned long data); + +/* Hardware communication channel */ +int bnxt_qplib_rcfw_wait_for_resp(struct bnxt_qplib_rcfw *rcfw, u16 cookie) +{ + u16 cbit; + int rc; + + cookie &= RCFW_MAX_COOKIE_VALUE; + cbit = cookie % RCFW_MAX_OUTSTANDING_CMD; + if (!test_bit(cbit, rcfw->cmdq_bitmap)) + dev_warn(&rcfw->pdev->dev, + "QPLIB: CMD bit %d for cookie 0x%x is not set?", + cbit, cookie); + + rc = wait_event_timeout(rcfw->waitq, + !test_bit(cbit, rcfw->cmdq_bitmap), + msecs_to_jiffies(RCFW_CMD_WAIT_TIME_MS)); + if (!rc) { + dev_warn(&rcfw->pdev->dev, + "QPLIB: Bono Error: timeout %d msec, msg {0x%x}\n", + RCFW_CMD_WAIT_TIME_MS, cookie); + } + + return rc; +}; + +int bnxt_qplib_rcfw_block_for_resp(struct bnxt_qplib_rcfw *rcfw, u16 cookie) +{ + u32 count = -1; + u16 cbit; + + cookie &= RCFW_MAX_COOKIE_VALUE; + cbit = cookie % RCFW_MAX_OUTSTANDING_CMD; + if (!test_bit(cbit, rcfw->cmdq_bitmap)) + goto done; + do { + bnxt_qplib_service_creq((unsigned long)rcfw); + } while (test_bit(cbit, rcfw->cmdq_bitmap) && --count); +done: + return count; +}; + +void *bnxt_qplib_rcfw_send_message(struct bnxt_qplib_rcfw *rcfw, + struct cmdq_base *req, void **crsbe, + u8 is_block) +{ + struct bnxt_qplib_crsq *crsq = &rcfw->crsq; + struct bnxt_qplib_cmdqe *cmdqe, **cmdq_ptr; + struct bnxt_qplib_hwq *cmdq = &rcfw->cmdq; + struct bnxt_qplib_hwq *crsb = &rcfw->crsb; + struct bnxt_qplib_crsqe *crsqe = NULL; + struct bnxt_qplib_crsbe **crsb_ptr; + u32 sw_prod, cmdq_prod; + u8 retry_cnt = 0xFF; + dma_addr_t dma_addr; + unsigned long flags; + u32 size, opcode; + u16 cookie, cbit; + int pg, idx; + u8 *preq; + +retry: + opcode = req->opcode; + if (!test_bit(FIRMWARE_INITIALIZED_FLAG, &rcfw->flags) && + (opcode != CMDQ_BASE_OPCODE_QUERY_FUNC && + opcode != CMDQ_BASE_OPCODE_INITIALIZE_FW)) { + dev_err(&rcfw->pdev->dev, + "QPLIB: RCFW not initialized, reject opcode 0x%x", + opcode); + return NULL; + } + + if (test_bit(FIRMWARE_INITIALIZED_FLAG, &rcfw->flags) && + opcode == CMDQ_BASE_OPCODE_INITIALIZE_FW) { + dev_err(&rcfw->pdev->dev, "QPLIB: RCFW already initialized!"); + return NULL; + } + + /* Cmdq are in 16-byte units, each request can consume 1 or more + * cmdqe + */ + spin_lock_irqsave(&cmdq->lock, flags); + if (req->cmd_size > cmdq->max_elements - + ((HWQ_CMP(cmdq->prod, cmdq) - HWQ_CMP(cmdq->cons, cmdq)) & + (cmdq->max_elements - 1))) { + dev_err(&rcfw->pdev->dev, "QPLIB: RCFW: CMDQ is full!"); + spin_unlock_irqrestore(&cmdq->lock, flags); + + if (!retry_cnt--) + return NULL; + goto retry; + } + + retry_cnt = 0xFF; + + cookie = atomic_inc_return(&rcfw->seq_num) & RCFW_MAX_COOKIE_VALUE; + cbit = cookie % RCFW_MAX_OUTSTANDING_CMD; + if (is_block) + cookie |= RCFW_CMD_IS_BLOCKING; + req->cookie = cpu_to_le16(cookie); + if (test_and_set_bit(cbit, rcfw->cmdq_bitmap)) { + dev_err(&rcfw->pdev->dev, + "QPLIB: RCFW MAX outstanding cmd reached!"); + atomic_dec(&rcfw->seq_num); + spin_unlock_irqrestore(&cmdq->lock, flags); + + if (!retry_cnt--) + return NULL; + goto retry; + } + /* Reserve a resp buffer slot if requested */ + if (req->resp_size && crsbe) { + spin_lock(&crsb->lock); + sw_prod = HWQ_CMP(crsb->prod, crsb); + crsb_ptr = (struct bnxt_qplib_crsbe **)crsb->pbl_ptr; + *crsbe = (void *)&crsb_ptr[get_crsb_pg(sw_prod)] + [get_crsb_idx(sw_prod)]; + bnxt_qplib_crsb_dma_next(crsb->pbl_dma_ptr, sw_prod, &dma_addr); + req->resp_addr = cpu_to_le64(dma_addr); + crsb->prod++; + spin_unlock(&crsb->lock); + + req->resp_size = (sizeof(struct bnxt_qplib_crsbe) + + BNXT_QPLIB_CMDQE_UNITS - 1) / + BNXT_QPLIB_CMDQE_UNITS; + } + cmdq_ptr = (struct bnxt_qplib_cmdqe **)cmdq->pbl_ptr; + preq = (u8 *)req; + size = req->cmd_size * BNXT_QPLIB_CMDQE_UNITS; + do { + pg = 0; + idx = 0; + + /* Locate the next cmdq slot */ + sw_prod = HWQ_CMP(cmdq->prod, cmdq); + cmdqe = &cmdq_ptr[get_cmdq_pg(sw_prod)][get_cmdq_idx(sw_prod)]; + if (!cmdqe) { + dev_err(&rcfw->pdev->dev, + "QPLIB: RCFW request failed with no cmdqe!"); + goto done; + } + /* Copy a segment of the req cmd to the cmdq */ + memset(cmdqe, 0, sizeof(*cmdqe)); + memcpy(cmdqe, preq, min_t(u32, size, sizeof(*cmdqe))); + preq += min_t(u32, size, sizeof(*cmdqe)); + size -= min_t(u32, size, sizeof(*cmdqe)); + cmdq->prod++; + } while (size > 0); + + cmdq_prod = cmdq->prod; + if (rcfw->flags & FIRMWARE_FIRST_FLAG) { + /* The very first doorbell write is required to set this flag + * which prompts the FW to reset its internal pointers + */ + cmdq_prod |= FIRMWARE_FIRST_FLAG; + rcfw->flags &= ~FIRMWARE_FIRST_FLAG; + } + sw_prod = HWQ_CMP(crsq->prod, crsq); + crsqe = &crsq->crsq[sw_prod]; + memset(crsqe, 0, sizeof(*crsqe)); + crsq->prod++; + crsqe->req_size = req->cmd_size; + + /* ring CMDQ DB */ + writel(cmdq_prod, rcfw->cmdq_bar_reg_iomem + + rcfw->cmdq_bar_reg_prod_off); + writel(RCFW_CMDQ_TRIG_VAL, rcfw->cmdq_bar_reg_iomem + + rcfw->cmdq_bar_reg_trig_off); +done: + spin_unlock_irqrestore(&cmdq->lock, flags); + /* Return the CREQ response pointer */ + return crsqe ? &crsqe->qp_event : NULL; +} + +/* Completions */ +static int bnxt_qplib_process_func_event(struct bnxt_qplib_rcfw *rcfw, + struct creq_func_event *func_event) +{ + switch (func_event->event) { + case CREQ_FUNC_EVENT_EVENT_TX_WQE_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_TX_DATA_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_RX_WQE_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_RX_DATA_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_CQ_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_TQM_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_CFCQ_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_CFCS_ERROR: + /* SRQ ctx error, call srq_handler?? + * But there's no SRQ handle! + */ + break; + case CREQ_FUNC_EVENT_EVENT_CFCC_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_CFCM_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_TIM_ERROR: + break; + case CREQ_FUNC_EVENT_EVENT_VF_COMM_REQUEST: + break; + case CREQ_FUNC_EVENT_EVENT_RESOURCE_EXHAUSTED: + break; + default: + return -EINVAL; + } + return 0; +} + +static int bnxt_qplib_process_qp_event(struct bnxt_qplib_rcfw *rcfw, + struct creq_qp_event *qp_event) +{ + struct bnxt_qplib_crsq *crsq = &rcfw->crsq; + struct bnxt_qplib_hwq *cmdq = &rcfw->cmdq; + struct bnxt_qplib_crsqe *crsqe; + u16 cbit, cookie, blocked = 0; + unsigned long flags; + u32 sw_cons; + + switch (qp_event->event) { + case CREQ_QP_EVENT_EVENT_QP_ERROR_NOTIFICATION: + dev_dbg(&rcfw->pdev->dev, + "QPLIB: Received QP error notification"); + break; + default: + /* Command Response */ + spin_lock_irqsave(&cmdq->lock, flags); + sw_cons = HWQ_CMP(crsq->cons, crsq); + crsqe = &crsq->crsq[sw_cons]; + crsq->cons++; + memcpy(&crsqe->qp_event, qp_event, sizeof(crsqe->qp_event)); + + cookie = le16_to_cpu(crsqe->qp_event.cookie); + blocked = cookie & RCFW_CMD_IS_BLOCKING; + cookie &= RCFW_MAX_COOKIE_VALUE; + cbit = cookie % RCFW_MAX_OUTSTANDING_CMD; + if (!test_and_clear_bit(cbit, rcfw->cmdq_bitmap)) + dev_warn(&rcfw->pdev->dev, + "QPLIB: CMD bit %d was not requested", cbit); + + cmdq->cons += crsqe->req_size; + spin_unlock_irqrestore(&cmdq->lock, flags); + if (!blocked) + wake_up(&rcfw->waitq); + break; + } + return 0; +} + +/* SP - CREQ Completion handlers */ +static void bnxt_qplib_service_creq(unsigned long data) +{ + struct bnxt_qplib_rcfw *rcfw = (struct bnxt_qplib_rcfw *)data; + struct bnxt_qplib_hwq *creq = &rcfw->creq; + struct creq_base *creqe, **creq_ptr; + u32 sw_cons, raw_cons; + unsigned long flags; + u32 type; + + /* Service the CREQ until empty */ + spin_lock_irqsave(&creq->lock, flags); + raw_cons = creq->cons; + while (1) { + sw_cons = HWQ_CMP(raw_cons, creq); + creq_ptr = (struct creq_base **)creq->pbl_ptr; + creqe = &creq_ptr[get_creq_pg(sw_cons)][get_creq_idx(sw_cons)]; + if (!CREQ_CMP_VALID(creqe, raw_cons, creq->max_elements)) + break; + + type = creqe->type & CREQ_BASE_TYPE_MASK; + switch (type) { + case CREQ_BASE_TYPE_QP_EVENT: + if (!bnxt_qplib_process_qp_event + (rcfw, (struct creq_qp_event *)creqe)) + rcfw->creq_qp_event_processed++; + else { + dev_warn(&rcfw->pdev->dev, "QPLIB: crsqe with"); + dev_warn(&rcfw->pdev->dev, + "QPLIB: type = 0x%x not handled", + type); + } + break; + case CREQ_BASE_TYPE_FUNC_EVENT: + if (!bnxt_qplib_process_func_event + (rcfw, (struct creq_func_event *)creqe)) + rcfw->creq_func_event_processed++; + else + dev_warn + (&rcfw->pdev->dev, "QPLIB:aeqe:%#x Not handled", + type); + break; + default: + dev_warn(&rcfw->pdev->dev, "QPLIB: creqe with "); + dev_warn(&rcfw->pdev->dev, + "QPLIB: op_event = 0x%x not handled", type); + break; + } + raw_cons++; + } + if (creq->cons != raw_cons) { + creq->cons = raw_cons; + CREQ_DB_REARM(rcfw->creq_bar_reg_iomem, raw_cons, + creq->max_elements); + } + spin_unlock_irqrestore(&creq->lock, flags); +} + +static irqreturn_t bnxt_qplib_creq_irq(int irq, void *dev_instance) +{ + struct bnxt_qplib_rcfw *rcfw = dev_instance; + struct bnxt_qplib_hwq *creq = &rcfw->creq; + struct creq_base **creq_ptr; + u32 sw_cons; + + /* Prefetch the CREQ element */ + sw_cons = HWQ_CMP(creq->cons, creq); + creq_ptr = (struct creq_base **)rcfw->creq.pbl_ptr; + prefetch(&creq_ptr[get_creq_pg(sw_cons)][get_creq_idx(sw_cons)]); + + tasklet_schedule(&rcfw->worker); + + return IRQ_HANDLED; +} + +/* RCFW */ +int bnxt_qplib_deinit_rcfw(struct bnxt_qplib_rcfw *rcfw) +{ + struct creq_deinitialize_fw_resp *resp; + struct cmdq_deinitialize_fw req; + u16 cmd_flags = 0; + + RCFW_CMD_PREP(req, DEINITIALIZE_FW, cmd_flags); + resp = (struct creq_deinitialize_fw_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, + NULL, 0); + if (!resp) + return -EINVAL; + + if (!bnxt_qplib_rcfw_wait_for_resp(rcfw, le16_to_cpu(req.cookie))) + return -ETIMEDOUT; + + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) + return -EFAULT; + + clear_bit(FIRMWARE_INITIALIZED_FLAG, &rcfw->flags); + return 0; +} + +static int __get_pbl_pg_idx(struct bnxt_qplib_pbl *pbl) +{ + return (pbl->pg_size == ROCE_PG_SIZE_4K ? + CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_4K : + pbl->pg_size == ROCE_PG_SIZE_8K ? + CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_8K : + pbl->pg_size == ROCE_PG_SIZE_64K ? + CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_64K : + pbl->pg_size == ROCE_PG_SIZE_2M ? + CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_2M : + pbl->pg_size == ROCE_PG_SIZE_8M ? + CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_8M : + pbl->pg_size == ROCE_PG_SIZE_1G ? + CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_1G : + CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_4K); +} + +int bnxt_qplib_init_rcfw(struct bnxt_qplib_rcfw *rcfw, + struct bnxt_qplib_ctx *ctx, int is_virtfn) +{ + struct creq_initialize_fw_resp *resp; + struct cmdq_initialize_fw req; + u16 cmd_flags = 0, level; + + RCFW_CMD_PREP(req, INITIALIZE_FW, cmd_flags); + + /* + * VFs need not setup the HW context area, PF + * shall setup this area for VF. Skipping the + * HW programming + */ + if (is_virtfn) + goto skip_ctx_setup; + + level = ctx->qpc_tbl.level; + req.qpc_pg_size_qpc_lvl = (level << CMDQ_INITIALIZE_FW_QPC_LVL_SFT) | + __get_pbl_pg_idx(&ctx->qpc_tbl.pbl[level]); + level = ctx->mrw_tbl.level; + req.mrw_pg_size_mrw_lvl = (level << CMDQ_INITIALIZE_FW_MRW_LVL_SFT) | + __get_pbl_pg_idx(&ctx->mrw_tbl.pbl[level]); + level = ctx->srqc_tbl.level; + req.srq_pg_size_srq_lvl = (level << CMDQ_INITIALIZE_FW_SRQ_LVL_SFT) | + __get_pbl_pg_idx(&ctx->srqc_tbl.pbl[level]); + level = ctx->cq_tbl.level; + req.cq_pg_size_cq_lvl = (level << CMDQ_INITIALIZE_FW_CQ_LVL_SFT) | + __get_pbl_pg_idx(&ctx->cq_tbl.pbl[level]); + level = ctx->srqc_tbl.level; + req.srq_pg_size_srq_lvl = (level << CMDQ_INITIALIZE_FW_SRQ_LVL_SFT) | + __get_pbl_pg_idx(&ctx->srqc_tbl.pbl[level]); + level = ctx->cq_tbl.level; + req.cq_pg_size_cq_lvl = (level << CMDQ_INITIALIZE_FW_CQ_LVL_SFT) | + __get_pbl_pg_idx(&ctx->cq_tbl.pbl[level]); + level = ctx->tim_tbl.level; + req.tim_pg_size_tim_lvl = (level << CMDQ_INITIALIZE_FW_TIM_LVL_SFT) | + __get_pbl_pg_idx(&ctx->tim_tbl.pbl[level]); + level = ctx->tqm_pde_level; + req.tqm_pg_size_tqm_lvl = (level << CMDQ_INITIALIZE_FW_TQM_LVL_SFT) | + __get_pbl_pg_idx(&ctx->tqm_pde.pbl[level]); + + req.qpc_page_dir = + cpu_to_le64(ctx->qpc_tbl.pbl[PBL_LVL_0].pg_map_arr[0]); + req.mrw_page_dir = + cpu_to_le64(ctx->mrw_tbl.pbl[PBL_LVL_0].pg_map_arr[0]); + req.srq_page_dir = + cpu_to_le64(ctx->srqc_tbl.pbl[PBL_LVL_0].pg_map_arr[0]); + req.cq_page_dir = + cpu_to_le64(ctx->cq_tbl.pbl[PBL_LVL_0].pg_map_arr[0]); + req.tim_page_dir = + cpu_to_le64(ctx->tim_tbl.pbl[PBL_LVL_0].pg_map_arr[0]); + req.tqm_page_dir = + cpu_to_le64(ctx->tqm_pde.pbl[PBL_LVL_0].pg_map_arr[0]); + + req.number_of_qp = cpu_to_le32(ctx->qpc_tbl.max_elements); + req.number_of_mrw = cpu_to_le32(ctx->mrw_tbl.max_elements); + req.number_of_srq = cpu_to_le32(ctx->srqc_tbl.max_elements); + req.number_of_cq = cpu_to_le32(ctx->cq_tbl.max_elements); + + req.max_qp_per_vf = cpu_to_le32(ctx->vf_res.max_qp_per_vf); + req.max_mrw_per_vf = cpu_to_le32(ctx->vf_res.max_mrw_per_vf); + req.max_srq_per_vf = cpu_to_le32(ctx->vf_res.max_srq_per_vf); + req.max_cq_per_vf = cpu_to_le32(ctx->vf_res.max_cq_per_vf); + req.max_gid_per_vf = cpu_to_le32(ctx->vf_res.max_gid_per_vf); + +skip_ctx_setup: + req.stat_ctx_id = cpu_to_le32(ctx->stats.fw_id); + resp = (struct creq_initialize_fw_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, + NULL, 0); + if (!resp) { + dev_err(&rcfw->pdev->dev, + "QPLIB: RCFW: INITIALIZE_FW send failed"); + return -EINVAL; + } + if (!bnxt_qplib_rcfw_wait_for_resp(rcfw, le16_to_cpu(req.cookie))) { + /* Cmd timed out */ + dev_err(&rcfw->pdev->dev, + "QPLIB: RCFW: INITIALIZE_FW timed out"); + return -ETIMEDOUT; + } + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&rcfw->pdev->dev, + "QPLIB: RCFW: INITIALIZE_FW failed"); + return -EINVAL; + } + set_bit(FIRMWARE_INITIALIZED_FLAG, &rcfw->flags); + return 0; +} + +void bnxt_qplib_free_rcfw_channel(struct bnxt_qplib_rcfw *rcfw) +{ + bnxt_qplib_free_hwq(rcfw->pdev, &rcfw->crsb); + kfree(rcfw->crsq.crsq); + bnxt_qplib_free_hwq(rcfw->pdev, &rcfw->cmdq); + bnxt_qplib_free_hwq(rcfw->pdev, &rcfw->creq); + + rcfw->pdev = NULL; +} + +int bnxt_qplib_alloc_rcfw_channel(struct pci_dev *pdev, + struct bnxt_qplib_rcfw *rcfw) +{ + rcfw->pdev = pdev; + rcfw->creq.max_elements = BNXT_QPLIB_CREQE_MAX_CNT; + if (bnxt_qplib_alloc_init_hwq(rcfw->pdev, &rcfw->creq, NULL, 0, + &rcfw->creq.max_elements, + BNXT_QPLIB_CREQE_UNITS, 0, PAGE_SIZE, + HWQ_TYPE_L2_CMPL)) { + dev_err(&rcfw->pdev->dev, + "QPLIB: HW channel CREQ allocation failed"); + goto fail; + } + rcfw->cmdq.max_elements = BNXT_QPLIB_CMDQE_MAX_CNT; + if (bnxt_qplib_alloc_init_hwq(rcfw->pdev, &rcfw->cmdq, NULL, 0, + &rcfw->cmdq.max_elements, + BNXT_QPLIB_CMDQE_UNITS, 0, PAGE_SIZE, + HWQ_TYPE_CTX)) { + dev_err(&rcfw->pdev->dev, + "QPLIB: HW channel CMDQ allocation failed"); + goto fail; + } + + rcfw->crsq.max_elements = rcfw->cmdq.max_elements; + rcfw->crsq.crsq = kcalloc(rcfw->crsq.max_elements, + sizeof(*rcfw->crsq.crsq), GFP_KERNEL); + if (!rcfw->crsq.crsq) + goto fail; + + rcfw->crsb.max_elements = BNXT_QPLIB_CRSBE_MAX_CNT; + if (bnxt_qplib_alloc_init_hwq(rcfw->pdev, &rcfw->crsb, NULL, 0, + &rcfw->crsb.max_elements, + BNXT_QPLIB_CRSBE_UNITS, 0, PAGE_SIZE, + HWQ_TYPE_CTX)) { + dev_err(&rcfw->pdev->dev, + "QPLIB: HW channel CRSB allocation failed"); + goto fail; + } + return 0; + +fail: + bnxt_qplib_free_rcfw_channel(rcfw); + return -ENOMEM; +} + +void bnxt_qplib_disable_rcfw_channel(struct bnxt_qplib_rcfw *rcfw) +{ + unsigned long indx; + + /* Make sure the HW channel is stopped! */ + synchronize_irq(rcfw->vector); + tasklet_disable(&rcfw->worker); + tasklet_kill(&rcfw->worker); + + if (rcfw->requested) { + free_irq(rcfw->vector, rcfw); + rcfw->requested = false; + } + if (rcfw->cmdq_bar_reg_iomem) + iounmap(rcfw->cmdq_bar_reg_iomem); + rcfw->cmdq_bar_reg_iomem = NULL; + + if (rcfw->creq_bar_reg_iomem) + iounmap(rcfw->creq_bar_reg_iomem); + rcfw->creq_bar_reg_iomem = NULL; + + indx = find_first_bit(rcfw->cmdq_bitmap, rcfw->bmap_size); + if (indx != rcfw->bmap_size) + dev_err(&rcfw->pdev->dev, + "QPLIB: disabling RCFW with pending cmd-bit %lx", indx); + kfree(rcfw->cmdq_bitmap); + rcfw->bmap_size = 0; + + rcfw->aeq_handler = NULL; + rcfw->vector = 0; +} + +int bnxt_qplib_enable_rcfw_channel(struct pci_dev *pdev, + struct bnxt_qplib_rcfw *rcfw, + int msix_vector, + int cp_bar_reg_off, int virt_fn, + int (*aeq_handler)(struct bnxt_qplib_rcfw *, + struct creq_func_event *)) +{ + resource_size_t res_base; + struct cmdq_init init; + u16 bmap_size; + int rc; + + /* General */ + atomic_set(&rcfw->seq_num, 0); + rcfw->flags = FIRMWARE_FIRST_FLAG; + bmap_size = BITS_TO_LONGS(RCFW_MAX_OUTSTANDING_CMD * + sizeof(unsigned long)); + rcfw->cmdq_bitmap = kzalloc(bmap_size, GFP_KERNEL); + if (!rcfw->cmdq_bitmap) + return -ENOMEM; + rcfw->bmap_size = bmap_size; + + /* CMDQ */ + rcfw->cmdq_bar_reg = RCFW_COMM_PCI_BAR_REGION; + res_base = pci_resource_start(pdev, rcfw->cmdq_bar_reg); + if (!res_base) + return -ENOMEM; + + rcfw->cmdq_bar_reg_iomem = ioremap_nocache(res_base + + RCFW_COMM_BASE_OFFSET, + RCFW_COMM_SIZE); + if (!rcfw->cmdq_bar_reg_iomem) { + dev_err(&rcfw->pdev->dev, + "QPLIB: CMDQ BAR region %d mapping failed", + rcfw->cmdq_bar_reg); + return -ENOMEM; + } + + rcfw->cmdq_bar_reg_prod_off = virt_fn ? RCFW_VF_COMM_PROD_OFFSET : + RCFW_PF_COMM_PROD_OFFSET; + + rcfw->cmdq_bar_reg_trig_off = RCFW_COMM_TRIG_OFFSET; + + /* CRSQ */ + rcfw->crsq.prod = 0; + rcfw->crsq.cons = 0; + + /* CREQ */ + rcfw->creq_bar_reg = RCFW_COMM_CONS_PCI_BAR_REGION; + res_base = pci_resource_start(pdev, rcfw->creq_bar_reg); + if (!res_base) + dev_err(&rcfw->pdev->dev, + "QPLIB: CREQ BAR region %d resc start is 0!", + rcfw->creq_bar_reg); + rcfw->creq_bar_reg_iomem = ioremap_nocache(res_base + cp_bar_reg_off, + 4); + if (!rcfw->creq_bar_reg_iomem) { + dev_err(&rcfw->pdev->dev, + "QPLIB: CREQ BAR region %d mapping failed", + rcfw->creq_bar_reg); + return -ENOMEM; + } + rcfw->creq_qp_event_processed = 0; + rcfw->creq_func_event_processed = 0; + + rcfw->vector = msix_vector; + if (aeq_handler) + rcfw->aeq_handler = aeq_handler; + + tasklet_init(&rcfw->worker, bnxt_qplib_service_creq, + (unsigned long)rcfw); + + rcfw->requested = false; + rc = request_irq(rcfw->vector, bnxt_qplib_creq_irq, 0, + "bnxt_qplib_creq", rcfw); + if (rc) { + dev_err(&rcfw->pdev->dev, + "QPLIB: Failed to request IRQ for CREQ rc = 0x%x", rc); + bnxt_qplib_disable_rcfw_channel(rcfw); + return rc; + } + rcfw->requested = true; + + init_waitqueue_head(&rcfw->waitq); + + CREQ_DB_REARM(rcfw->creq_bar_reg_iomem, 0, rcfw->creq.max_elements); + + init.cmdq_pbl = cpu_to_le64(rcfw->cmdq.pbl[PBL_LVL_0].pg_map_arr[0]); + init.cmdq_size_cmdq_lvl = cpu_to_le16( + ((BNXT_QPLIB_CMDQE_MAX_CNT << CMDQ_INIT_CMDQ_SIZE_SFT) & + CMDQ_INIT_CMDQ_SIZE_MASK) | + ((rcfw->cmdq.level << CMDQ_INIT_CMDQ_LVL_SFT) & + CMDQ_INIT_CMDQ_LVL_MASK)); + init.creq_ring_id = cpu_to_le16(rcfw->creq_ring_id); + + /* Write to the Bono mailbox register */ + __iowrite32_copy(rcfw->cmdq_bar_reg_iomem, &init, sizeof(init) / 4); + return 0; +} diff --git a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h new file mode 100644 index 000000000000..d3567d75bf58 --- /dev/null +++ b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h @@ -0,0 +1,231 @@ +/* + * Broadcom NetXtreme-E RoCE driver. + * + * Copyright (c) 2016 - 2017, Broadcom. All rights reserved. The term + * Broadcom refers to Broadcom Limited and/or its subsidiaries. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, + * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS + * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE + * OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN + * IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * Description: RDMA Controller HW interface (header) + */ + +#ifndef __BNXT_QPLIB_RCFW_H__ +#define __BNXT_QPLIB_RCFW_H__ + +#define RCFW_CMDQ_TRIG_VAL 1 +#define RCFW_COMM_PCI_BAR_REGION 0 +#define RCFW_COMM_CONS_PCI_BAR_REGION 2 +#define RCFW_COMM_BASE_OFFSET 0x600 +#define RCFW_PF_COMM_PROD_OFFSET 0xc +#define RCFW_VF_COMM_PROD_OFFSET 0xc +#define RCFW_COMM_TRIG_OFFSET 0x100 +#define RCFW_COMM_SIZE 0x104 + +#define RCFW_DBR_PCI_BAR_REGION 2 + +#define RCFW_CMD_PREP(req, CMD, cmd_flags) \ + do { \ + memset(&(req), 0, sizeof((req))); \ + (req).opcode = CMDQ_BASE_OPCODE_##CMD; \ + (req).cmd_size = (sizeof((req)) + \ + BNXT_QPLIB_CMDQE_UNITS - 1) / \ + BNXT_QPLIB_CMDQE_UNITS; \ + (req).flags = cpu_to_le16(cmd_flags); \ + } while (0) + +#define RCFW_CMD_WAIT_TIME_MS 20000 /* 20 Seconds timeout */ + +/* CMDQ elements */ +#define BNXT_QPLIB_CMDQE_MAX_CNT 256 +#define BNXT_QPLIB_CMDQE_UNITS sizeof(struct bnxt_qplib_cmdqe) +#define BNXT_QPLIB_CMDQE_CNT_PER_PG (PAGE_SIZE / BNXT_QPLIB_CMDQE_UNITS) + +#define MAX_CMDQ_IDX (BNXT_QPLIB_CMDQE_MAX_CNT - 1) +#define MAX_CMDQ_IDX_PER_PG (BNXT_QPLIB_CMDQE_CNT_PER_PG - 1) + +#define RCFW_MAX_OUTSTANDING_CMD BNXT_QPLIB_CMDQE_MAX_CNT +#define RCFW_MAX_COOKIE_VALUE 0x7FFF +#define RCFW_CMD_IS_BLOCKING 0x8000 + +/* Cmdq contains a fix number of a 16-Byte slots */ +struct bnxt_qplib_cmdqe { + u8 data[16]; +}; + +static inline u32 get_cmdq_pg(u32 val) +{ + return (val & ~MAX_CMDQ_IDX_PER_PG) / BNXT_QPLIB_CMDQE_CNT_PER_PG; +} + +static inline u32 get_cmdq_idx(u32 val) +{ + return val & MAX_CMDQ_IDX_PER_PG; +} + +/* Crsq buf is 1024-Byte */ +struct bnxt_qplib_crsbe { + u8 data[1024]; +}; + +/* CRSQ SB */ +#define BNXT_QPLIB_CRSBE_MAX_CNT 4 +#define BNXT_QPLIB_CRSBE_UNITS sizeof(struct bnxt_qplib_crsbe) +#define BNXT_QPLIB_CRSBE_CNT_PER_PG (PAGE_SIZE / BNXT_QPLIB_CRSBE_UNITS) + +#define MAX_CRSB_IDX (BNXT_QPLIB_CRSBE_MAX_CNT - 1) +#define MAX_CRSB_IDX_PER_PG (BNXT_QPLIB_CRSBE_CNT_PER_PG - 1) + +static inline u32 get_crsb_pg(u32 val) +{ + return (val & ~MAX_CRSB_IDX_PER_PG) / BNXT_QPLIB_CRSBE_CNT_PER_PG; +} + +static inline u32 get_crsb_idx(u32 val) +{ + return val & MAX_CRSB_IDX_PER_PG; +} + +static inline void bnxt_qplib_crsb_dma_next(dma_addr_t *pg_map_arr, + u32 prod, dma_addr_t *dma_addr) +{ + *dma_addr = pg_map_arr[(prod) / BNXT_QPLIB_CRSBE_CNT_PER_PG]; + *dma_addr += ((prod) % BNXT_QPLIB_CRSBE_CNT_PER_PG) * + BNXT_QPLIB_CRSBE_UNITS; +} + +/* CREQ */ +/* Allocate 1 per QP for async error notification for now */ +#define BNXT_QPLIB_CREQE_MAX_CNT (64 * 1024) +#define BNXT_QPLIB_CREQE_UNITS 16 /* 16-Bytes per prod unit */ +#define BNXT_QPLIB_CREQE_CNT_PER_PG (PAGE_SIZE / BNXT_QPLIB_CREQE_UNITS) + +#define MAX_CREQ_IDX (BNXT_QPLIB_CREQE_MAX_CNT - 1) +#define MAX_CREQ_IDX_PER_PG (BNXT_QPLIB_CREQE_CNT_PER_PG - 1) + +static inline u32 get_creq_pg(u32 val) +{ + return (val & ~MAX_CREQ_IDX_PER_PG) / BNXT_QPLIB_CREQE_CNT_PER_PG; +} + +static inline u32 get_creq_idx(u32 val) +{ + return val & MAX_CREQ_IDX_PER_PG; +} + +#define BNXT_QPLIB_CREQE_PER_PG (PAGE_SIZE / sizeof(struct creq_base)) + +#define CREQ_CMP_VALID(hdr, raw_cons, cp_bit) \ + (!!((hdr)->v & CREQ_BASE_V) == \ + !((raw_cons) & (cp_bit))) + +#define CREQ_DB_KEY_CP (0x2 << CMPL_DOORBELL_KEY_SFT) +#define CREQ_DB_IDX_VALID CMPL_DOORBELL_IDX_VALID +#define CREQ_DB_IRQ_DIS CMPL_DOORBELL_MASK +#define CREQ_DB_CP_FLAGS_REARM (CREQ_DB_KEY_CP | \ + CREQ_DB_IDX_VALID) +#define CREQ_DB_CP_FLAGS (CREQ_DB_KEY_CP | \ + CREQ_DB_IDX_VALID | \ + CREQ_DB_IRQ_DIS) +#define CREQ_DB_REARM(db, raw_cons, cp_bit) \ + writel(CREQ_DB_CP_FLAGS_REARM | ((raw_cons) & ((cp_bit) - 1)), db) +#define CREQ_DB(db, raw_cons, cp_bit) \ + writel(CREQ_DB_CP_FLAGS | ((raw_cons) & ((cp_bit) - 1)), db) + +/* HWQ */ +struct bnxt_qplib_crsqe { + struct creq_qp_event qp_event; + u32 req_size; +}; + +struct bnxt_qplib_crsq { + struct bnxt_qplib_crsqe *crsq; + u32 prod; + u32 cons; + u32 max_elements; +}; + +/* RCFW Communication Channels */ +struct bnxt_qplib_rcfw { + struct pci_dev *pdev; + int vector; + struct tasklet_struct worker; + bool requested; + unsigned long *cmdq_bitmap; + u32 bmap_size; + unsigned long flags; +#define FIRMWARE_INITIALIZED_FLAG 1 +#define FIRMWARE_FIRST_FLAG BIT(31) + wait_queue_head_t waitq; + int (*aeq_handler)(struct bnxt_qplib_rcfw *, + struct creq_func_event *); + atomic_t seq_num; + + /* Bar region info */ + void __iomem *cmdq_bar_reg_iomem; + u16 cmdq_bar_reg; + u16 cmdq_bar_reg_prod_off; + u16 cmdq_bar_reg_trig_off; + u16 creq_ring_id; + u16 creq_bar_reg; + void __iomem *creq_bar_reg_iomem; + + /* Cmd-Resp and Async Event notification queue */ + struct bnxt_qplib_hwq creq; + u64 creq_qp_event_processed; + u64 creq_func_event_processed; + + /* Actual Cmd and Resp Queues */ + struct bnxt_qplib_hwq cmdq; + struct bnxt_qplib_crsq crsq; + struct bnxt_qplib_hwq crsb; +}; + +void bnxt_qplib_free_rcfw_channel(struct bnxt_qplib_rcfw *rcfw); +int bnxt_qplib_alloc_rcfw_channel(struct pci_dev *pdev, + struct bnxt_qplib_rcfw *rcfw); +void bnxt_qplib_disable_rcfw_channel(struct bnxt_qplib_rcfw *rcfw); +int bnxt_qplib_enable_rcfw_channel(struct pci_dev *pdev, + struct bnxt_qplib_rcfw *rcfw, + int msix_vector, + int cp_bar_reg_off, int virt_fn, + int (*aeq_handler) + (struct bnxt_qplib_rcfw *, + struct creq_func_event *)); + +int bnxt_qplib_rcfw_block_for_resp(struct bnxt_qplib_rcfw *rcfw, u16 cookie); +int bnxt_qplib_rcfw_wait_for_resp(struct bnxt_qplib_rcfw *rcfw, u16 cookie); +void *bnxt_qplib_rcfw_send_message(struct bnxt_qplib_rcfw *rcfw, + struct cmdq_base *req, void **crsbe, + u8 is_block); + +int bnxt_qplib_deinit_rcfw(struct bnxt_qplib_rcfw *rcfw); +int bnxt_qplib_init_rcfw(struct bnxt_qplib_rcfw *rcfw, + struct bnxt_qplib_ctx *ctx, int is_virtfn); +#endif /* __BNXT_QPLIB_RCFW_H__ */ diff --git a/drivers/infiniband/hw/bnxt_re/qplib_res.c b/drivers/infiniband/hw/bnxt_re/qplib_res.c new file mode 100644 index 000000000000..62447b3badec --- /dev/null +++ b/drivers/infiniband/hw/bnxt_re/qplib_res.c @@ -0,0 +1,825 @@ +/* + * Broadcom NetXtreme-E RoCE driver. + * + * Copyright (c) 2016 - 2017, Broadcom. All rights reserved. The term + * Broadcom refers to Broadcom Limited and/or its subsidiaries. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, + * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS + * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE + * OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN + * IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * Description: QPLib resource manager + */ + +#include <linux/spinlock.h> +#include <linux/pci.h> +#include <linux/interrupt.h> +#include <linux/inetdevice.h> +#include <linux/dma-mapping.h> +#include <linux/if_vlan.h> +#include "roce_hsi.h" +#include "qplib_res.h" +#include "qplib_sp.h" +#include "qplib_rcfw.h" + +static void bnxt_qplib_free_stats_ctx(struct pci_dev *pdev, + struct bnxt_qplib_stats *stats); +static int bnxt_qplib_alloc_stats_ctx(struct pci_dev *pdev, + struct bnxt_qplib_stats *stats); + +/* PBL */ +static void __free_pbl(struct pci_dev *pdev, struct bnxt_qplib_pbl *pbl, + bool is_umem) +{ + int i; + + if (!is_umem) { + for (i = 0; i < pbl->pg_count; i++) { + if (pbl->pg_arr[i]) + dma_free_coherent(&pdev->dev, pbl->pg_size, + (void *)((unsigned long) + pbl->pg_arr[i] & + PAGE_MASK), + pbl->pg_map_arr[i]); + else + dev_warn(&pdev->dev, + "QPLIB: PBL free pg_arr[%d] empty?!", + i); + pbl->pg_arr[i] = NULL; + } + } + kfree(pbl->pg_arr); + pbl->pg_arr = NULL; + kfree(pbl->pg_map_arr); + pbl->pg_map_arr = NULL; + pbl->pg_count = 0; + pbl->pg_size = 0; +} + +static int __alloc_pbl(struct pci_dev *pdev, struct bnxt_qplib_pbl *pbl, + struct scatterlist *sghead, u32 pages, u32 pg_size) +{ + struct scatterlist *sg; + bool is_umem = false; + int i; + + /* page ptr arrays */ + pbl->pg_arr = kcalloc(pages, sizeof(void *), GFP_KERNEL); + if (!pbl->pg_arr) + return -ENOMEM; + + pbl->pg_map_arr = kcalloc(pages, sizeof(dma_addr_t), GFP_KERNEL); + if (!pbl->pg_map_arr) { + kfree(pbl->pg_arr); + pbl->pg_arr = NULL; + return -ENOMEM; + } + pbl->pg_count = 0; + pbl->pg_size = pg_size; + + if (!sghead) { + for (i = 0; i < pages; i++) { + pbl->pg_arr[i] = dma_alloc_coherent(&pdev->dev, + pbl->pg_size, + &pbl->pg_map_arr[i], + GFP_KERNEL); + if (!pbl->pg_arr[i]) + goto fail; + memset(pbl->pg_arr[i], 0, pbl->pg_size); + pbl->pg_count++; + } + } else { + i = 0; + is_umem = true; + for_each_sg(sghead, sg, pages, i) { + pbl->pg_map_arr[i] = sg_dma_address(sg); + pbl->pg_arr[i] = sg_virt(sg); + if (!pbl->pg_arr[i]) + goto fail; + + pbl->pg_count++; + } + } + + return 0; + +fail: + __free_pbl(pdev, pbl, is_umem); + return -ENOMEM; +} + +/* HWQ */ +void bnxt_qplib_free_hwq(struct pci_dev *pdev, struct bnxt_qplib_hwq *hwq) +{ + int i; + + if (!hwq->max_elements) + return; + if (hwq->level >= PBL_LVL_MAX) + return; + + for (i = 0; i < hwq->level + 1; i++) { + if (i == hwq->level) + __free_pbl(pdev, &hwq->pbl[i], hwq->is_user); + else + __free_pbl(pdev, &hwq->pbl[i], false); + } + + hwq->level = PBL_LVL_MAX; + hwq->max_elements = 0; + hwq->element_size = 0; + hwq->prod = 0; + hwq->cons = 0; + hwq->cp_bit = 0; +} + +/* All HWQs are power of 2 in size */ +int bnxt_qplib_alloc_init_hwq(struct pci_dev *pdev, struct bnxt_qplib_hwq *hwq, + struct scatterlist *sghead, int nmap, + u32 *elements, u32 element_size, u32 aux, + u32 pg_size, enum bnxt_qplib_hwq_type hwq_type) +{ + u32 pages, slots, size, aux_pages = 0, aux_size = 0; + dma_addr_t *src_phys_ptr, **dst_virt_ptr; + int i, rc; + + hwq->level = PBL_LVL_MAX; + + slots = roundup_pow_of_two(*elements); + if (aux) { + aux_size = roundup_pow_of_two(aux); + aux_pages = (slots * aux_size) / pg_size; + if ((slots * aux_size) % pg_size) + aux_pages++; + } + size = roundup_pow_of_two(element_size); + + if (!sghead) { + hwq->is_user = false; + pages = (slots * size) / pg_size + aux_pages; + if ((slots * size) % pg_size) + pages++; + if (!pages) + return -EINVAL; + } else { + hwq->is_user = true; + pages = nmap; + } + + /* Alloc the 1st memory block; can be a PDL/PTL/PBL */ + if (sghead && (pages == MAX_PBL_LVL_0_PGS)) + rc = __alloc_pbl(pdev, &hwq->pbl[PBL_LVL_0], sghead, + pages, pg_size); + else + rc = __alloc_pbl(pdev, &hwq->pbl[PBL_LVL_0], NULL, 1, pg_size); + if (rc) + goto fail; + + hwq->level = PBL_LVL_0; + + if (pages > MAX_PBL_LVL_0_PGS) { + if (pages > MAX_PBL_LVL_1_PGS) { + /* 2 levels of indirection */ + rc = __alloc_pbl(pdev, &hwq->pbl[PBL_LVL_1], NULL, + MAX_PBL_LVL_1_PGS_FOR_LVL_2, pg_size); + if (rc) + goto fail; + /* Fill in lvl0 PBL */ + dst_virt_ptr = + (dma_addr_t **)hwq->pbl[PBL_LVL_0].pg_arr; + src_phys_ptr = hwq->pbl[PBL_LVL_1].pg_map_arr; + for (i = 0; i < hwq->pbl[PBL_LVL_1].pg_count; i++) + dst_virt_ptr[PTR_PG(i)][PTR_IDX(i)] = + src_phys_ptr[i] | PTU_PDE_VALID; + hwq->level = PBL_LVL_1; + + rc = __alloc_pbl(pdev, &hwq->pbl[PBL_LVL_2], sghead, + pages, pg_size); + if (rc) + goto fail; + + /* Fill in lvl1 PBL */ + dst_virt_ptr = + (dma_addr_t **)hwq->pbl[PBL_LVL_1].pg_arr; + src_phys_ptr = hwq->pbl[PBL_LVL_2].pg_map_arr; + for (i = 0; i < hwq->pbl[PBL_LVL_2].pg_count; i++) { + dst_virt_ptr[PTR_PG(i)][PTR_IDX(i)] = + src_phys_ptr[i] | PTU_PTE_VALID; + } + if (hwq_type == HWQ_TYPE_QUEUE) { + /* Find the last pg of the size */ + i = hwq->pbl[PBL_LVL_2].pg_count; + dst_virt_ptr[PTR_PG(i - 1)][PTR_IDX(i - 1)] |= + PTU_PTE_LAST; + if (i > 1) + dst_virt_ptr[PTR_PG(i - 2)] + [PTR_IDX(i - 2)] |= + PTU_PTE_NEXT_TO_LAST; + } + hwq->level = PBL_LVL_2; + } else { + u32 flag = hwq_type == HWQ_TYPE_L2_CMPL ? 0 : + PTU_PTE_VALID; + + /* 1 level of indirection */ + rc = __alloc_pbl(pdev, &hwq->pbl[PBL_LVL_1], sghead, + pages, pg_size); + if (rc) + goto fail; + /* Fill in lvl0 PBL */ + dst_virt_ptr = + (dma_addr_t **)hwq->pbl[PBL_LVL_0].pg_arr; + src_phys_ptr = hwq->pbl[PBL_LVL_1].pg_map_arr; + for (i = 0; i < hwq->pbl[PBL_LVL_1].pg_count; i++) { + dst_virt_ptr[PTR_PG(i)][PTR_IDX(i)] = + src_phys_ptr[i] | flag; + } + if (hwq_type == HWQ_TYPE_QUEUE) { + /* Find the last pg of the size */ + i = hwq->pbl[PBL_LVL_1].pg_count; + dst_virt_ptr[PTR_PG(i - 1)][PTR_IDX(i - 1)] |= + PTU_PTE_LAST; + if (i > 1) + dst_virt_ptr[PTR_PG(i - 2)] + [PTR_IDX(i - 2)] |= + PTU_PTE_NEXT_TO_LAST; + } + hwq->level = PBL_LVL_1; + } + } + hwq->pdev = pdev; + spin_lock_init(&hwq->lock); + hwq->prod = 0; + hwq->cons = 0; + *elements = hwq->max_elements = slots; + hwq->element_size = size; + + /* For direct access to the elements */ + hwq->pbl_ptr = hwq->pbl[hwq->level].pg_arr; + hwq->pbl_dma_ptr = hwq->pbl[hwq->level].pg_map_arr; + + return 0; + +fail: + bnxt_qplib_free_hwq(pdev, hwq); + return -ENOMEM; +} + +/* Context Tables */ +void bnxt_qplib_free_ctx(struct pci_dev *pdev, + struct bnxt_qplib_ctx *ctx) +{ + int i; + + bnxt_qplib_free_hwq(pdev, &ctx->qpc_tbl); + bnxt_qplib_free_hwq(pdev, &ctx->mrw_tbl); + bnxt_qplib_free_hwq(pdev, &ctx->srqc_tbl); + bnxt_qplib_free_hwq(pdev, &ctx->cq_tbl); + bnxt_qplib_free_hwq(pdev, &ctx->tim_tbl); + for (i = 0; i < MAX_TQM_ALLOC_REQ; i++) + bnxt_qplib_free_hwq(pdev, &ctx->tqm_tbl[i]); + bnxt_qplib_free_hwq(pdev, &ctx->tqm_pde); + bnxt_qplib_free_stats_ctx(pdev, &ctx->stats); +} + +/* + * Routine: bnxt_qplib_alloc_ctx + * Description: + * Context tables are memories which are used by the chip fw. + * The 6 tables defined are: + * QPC ctx - holds QP states + * MRW ctx - holds memory region and window + * SRQ ctx - holds shared RQ states + * CQ ctx - holds completion queue states + * TQM ctx - holds Tx Queue Manager context + * TIM ctx - holds timer context + * Depending on the size of the tbl requested, either a 1 Page Buffer List + * or a 1-to-2-stage indirection Page Directory List + 1 PBL is used + * instead. + * Table might be employed as follows: + * For 0 < ctx size <= 1 PAGE, 0 level of ind is used + * For 1 PAGE < ctx size <= 512 entries size, 1 level of ind is used + * For 512 < ctx size <= MAX, 2 levels of ind is used + * Returns: + * 0 if success, else -ERRORS + */ +int bnxt_qplib_alloc_ctx(struct pci_dev *pdev, + struct bnxt_qplib_ctx *ctx, + bool virt_fn) +{ + int i, j, k, rc = 0; + int fnz_idx = -1; + __le64 **pbl_ptr; + + if (virt_fn) + goto stats_alloc; + + /* QPC Tables */ + ctx->qpc_tbl.max_elements = ctx->qpc_count; + rc = bnxt_qplib_alloc_init_hwq(pdev, &ctx->qpc_tbl, NULL, 0, + &ctx->qpc_tbl.max_elements, + BNXT_QPLIB_MAX_QP_CTX_ENTRY_SIZE, 0, + PAGE_SIZE, HWQ_TYPE_CTX); + if (rc) + goto fail; + + /* MRW Tables */ + ctx->mrw_tbl.max_elements = ctx->mrw_count; + rc = bnxt_qplib_alloc_init_hwq(pdev, &ctx->mrw_tbl, NULL, 0, + &ctx->mrw_tbl.max_elements, + BNXT_QPLIB_MAX_MRW_CTX_ENTRY_SIZE, 0, + PAGE_SIZE, HWQ_TYPE_CTX); + if (rc) + goto fail; + + /* SRQ Tables */ + ctx->srqc_tbl.max_elements = ctx->srqc_count; + rc = bnxt_qplib_alloc_init_hwq(pdev, &ctx->srqc_tbl, NULL, 0, + &ctx->srqc_tbl.max_elements, + BNXT_QPLIB_MAX_SRQ_CTX_ENTRY_SIZE, 0, + PAGE_SIZE, HWQ_TYPE_CTX); + if (rc) + goto fail; + + /* CQ Tables */ + ctx->cq_tbl.max_elements = ctx->cq_count; + rc = bnxt_qplib_alloc_init_hwq(pdev, &ctx->cq_tbl, NULL, 0, + &ctx->cq_tbl.max_elements, + BNXT_QPLIB_MAX_CQ_CTX_ENTRY_SIZE, 0, + PAGE_SIZE, HWQ_TYPE_CTX); + if (rc) + goto fail; + + /* TQM Buffer */ + ctx->tqm_pde.max_elements = 512; + rc = bnxt_qplib_alloc_init_hwq(pdev, &ctx->tqm_pde, NULL, 0, + &ctx->tqm_pde.max_elements, sizeof(u64), + 0, PAGE_SIZE, HWQ_TYPE_CTX); + if (rc) + goto fail; + + for (i = 0; i < MAX_TQM_ALLOC_REQ; i++) { + if (!ctx->tqm_count[i]) + continue; + ctx->tqm_tbl[i].max_elements = ctx->qpc_count * + ctx->tqm_count[i]; + rc = bnxt_qplib_alloc_init_hwq(pdev, &ctx->tqm_tbl[i], NULL, 0, + &ctx->tqm_tbl[i].max_elements, 1, + 0, PAGE_SIZE, HWQ_TYPE_CTX); + if (rc) + goto fail; + } + pbl_ptr = (__le64 **)ctx->tqm_pde.pbl_ptr; + for (i = 0, j = 0; i < MAX_TQM_ALLOC_REQ; + i++, j += MAX_TQM_ALLOC_BLK_SIZE) { + if (!ctx->tqm_tbl[i].max_elements) + continue; + if (fnz_idx == -1) + fnz_idx = i; + switch (ctx->tqm_tbl[i].level) { + case PBL_LVL_2: + for (k = 0; k < ctx->tqm_tbl[i].pbl[PBL_LVL_1].pg_count; + k++) + pbl_ptr[PTR_PG(j + k)][PTR_IDX(j + k)] = + cpu_to_le64( + ctx->tqm_tbl[i].pbl[PBL_LVL_1].pg_map_arr[k] + | PTU_PTE_VALID); + break; + case PBL_LVL_1: + case PBL_LVL_0: + default: + pbl_ptr[PTR_PG(j)][PTR_IDX(j)] = cpu_to_le64( + ctx->tqm_tbl[i].pbl[PBL_LVL_0].pg_map_arr[0] | + PTU_PTE_VALID); + break; + } + } + if (fnz_idx == -1) + fnz_idx = 0; + ctx->tqm_pde_level = ctx->tqm_tbl[fnz_idx].level == PBL_LVL_2 ? + PBL_LVL_2 : ctx->tqm_tbl[fnz_idx].level + 1; + + /* TIM Buffer */ + ctx->tim_tbl.max_elements = ctx->qpc_count * 16; + rc = bnxt_qplib_alloc_init_hwq(pdev, &ctx->tim_tbl, NULL, 0, + &ctx->tim_tbl.max_elements, 1, + 0, PAGE_SIZE, HWQ_TYPE_CTX); + if (rc) + goto fail; + +stats_alloc: + /* Stats */ + rc = bnxt_qplib_alloc_stats_ctx(pdev, &ctx->stats); + if (rc) + goto fail; + + return 0; + +fail: + bnxt_qplib_free_ctx(pdev, ctx); + return rc; +} + +/* GUID */ +void bnxt_qplib_get_guid(u8 *dev_addr, u8 *guid) +{ + u8 mac[ETH_ALEN]; + + /* MAC-48 to EUI-64 mapping */ + memcpy(mac, dev_addr, ETH_ALEN); + guid[0] = mac[0] ^ 2; + guid[1] = mac[1]; + guid[2] = mac[2]; + guid[3] = 0xff; + guid[4] = 0xfe; + guid[5] = mac[3]; + guid[6] = mac[4]; + guid[7] = mac[5]; +} + +static void bnxt_qplib_free_sgid_tbl(struct bnxt_qplib_res *res, + struct bnxt_qplib_sgid_tbl *sgid_tbl) +{ + kfree(sgid_tbl->tbl); + kfree(sgid_tbl->hw_id); + kfree(sgid_tbl->ctx); + sgid_tbl->tbl = NULL; + sgid_tbl->hw_id = NULL; + sgid_tbl->ctx = NULL; + sgid_tbl->max = 0; + sgid_tbl->active = 0; +} + +static int bnxt_qplib_alloc_sgid_tbl(struct bnxt_qplib_res *res, + struct bnxt_qplib_sgid_tbl *sgid_tbl, + u16 max) +{ + sgid_tbl->tbl = kcalloc(max, sizeof(struct bnxt_qplib_gid), GFP_KERNEL); + if (!sgid_tbl->tbl) + return -ENOMEM; + + sgid_tbl->hw_id = kcalloc(max, sizeof(u16), GFP_KERNEL); + if (!sgid_tbl->hw_id) + goto out_free1; + + sgid_tbl->ctx = kcalloc(max, sizeof(void *), GFP_KERNEL); + if (!sgid_tbl->ctx) + goto out_free2; + + sgid_tbl->max = max; + return 0; +out_free2: + kfree(sgid_tbl->hw_id); + sgid_tbl->hw_id = NULL; +out_free1: + kfree(sgid_tbl->tbl); + sgid_tbl->tbl = NULL; + return -ENOMEM; +}; + +static void bnxt_qplib_cleanup_sgid_tbl(struct bnxt_qplib_res *res, + struct bnxt_qplib_sgid_tbl *sgid_tbl) +{ + int i; + + for (i = 0; i < sgid_tbl->max; i++) { + if (memcmp(&sgid_tbl->tbl[i], &bnxt_qplib_gid_zero, + sizeof(bnxt_qplib_gid_zero))) + bnxt_qplib_del_sgid(sgid_tbl, &sgid_tbl->tbl[i], true); + } + memset(sgid_tbl->tbl, 0, sizeof(struct bnxt_qplib_gid) * sgid_tbl->max); + memset(sgid_tbl->hw_id, -1, sizeof(u16) * sgid_tbl->max); + sgid_tbl->active = 0; +} + +static void bnxt_qplib_init_sgid_tbl(struct bnxt_qplib_sgid_tbl *sgid_tbl, + struct net_device *netdev) +{ + memset(sgid_tbl->tbl, 0, sizeof(struct bnxt_qplib_gid) * sgid_tbl->max); + memset(sgid_tbl->hw_id, -1, sizeof(u16) * sgid_tbl->max); +} + +static void bnxt_qplib_free_pkey_tbl(struct bnxt_qplib_res *res, + struct bnxt_qplib_pkey_tbl *pkey_tbl) +{ + if (!pkey_tbl->tbl) + dev_dbg(&res->pdev->dev, "QPLIB: PKEY tbl not present"); + else + kfree(pkey_tbl->tbl); + + pkey_tbl->tbl = NULL; + pkey_tbl->max = 0; + pkey_tbl->active = 0; +} + +static int bnxt_qplib_alloc_pkey_tbl(struct bnxt_qplib_res *res, + struct bnxt_qplib_pkey_tbl *pkey_tbl, + u16 max) +{ + pkey_tbl->tbl = kcalloc(max, sizeof(u16), GFP_KERNEL); + if (!pkey_tbl->tbl) + return -ENOMEM; + + pkey_tbl->max = max; + return 0; +}; + +/* PDs */ +int bnxt_qplib_alloc_pd(struct bnxt_qplib_pd_tbl *pdt, struct bnxt_qplib_pd *pd) +{ + u32 bit_num; + + bit_num = find_first_bit(pdt->tbl, pdt->max); + if (bit_num == pdt->max) + return -ENOMEM; + + /* Found unused PD */ + clear_bit(bit_num, pdt->tbl); + pd->id = bit_num; + return 0; +} + +int bnxt_qplib_dealloc_pd(struct bnxt_qplib_res *res, + struct bnxt_qplib_pd_tbl *pdt, + struct bnxt_qplib_pd *pd) +{ + if (test_and_set_bit(pd->id, pdt->tbl)) { + dev_warn(&res->pdev->dev, "Freeing an unused PD? pdn = %d", + pd->id); + return -EINVAL; + } + pd->id = 0; + return 0; +} + +static void bnxt_qplib_free_pd_tbl(struct bnxt_qplib_pd_tbl *pdt) +{ + kfree(pdt->tbl); + pdt->tbl = NULL; + pdt->max = 0; +} + +static int bnxt_qplib_alloc_pd_tbl(struct bnxt_qplib_res *res, + struct bnxt_qplib_pd_tbl *pdt, + u32 max) +{ + u32 bytes; + + bytes = max >> 3; + if (!bytes) + bytes = 1; + pdt->tbl = kmalloc(bytes, GFP_KERNEL); + if (!pdt->tbl) + return -ENOMEM; + + pdt->max = max; + memset((u8 *)pdt->tbl, 0xFF, bytes); + + return 0; +} + +/* DPIs */ +int bnxt_qplib_alloc_dpi(struct bnxt_qplib_dpi_tbl *dpit, + struct bnxt_qplib_dpi *dpi, + void *app) +{ + u32 bit_num; + + bit_num = find_first_bit(dpit->tbl, dpit->max); + if (bit_num == dpit->max) + return -ENOMEM; + + /* Found unused DPI */ + clear_bit(bit_num, dpit->tbl); + dpit->app_tbl[bit_num] = app; + + dpi->dpi = bit_num; + dpi->dbr = dpit->dbr_bar_reg_iomem + (bit_num * PAGE_SIZE); + dpi->umdbr = dpit->unmapped_dbr + (bit_num * PAGE_SIZE); + + return 0; +} + +int bnxt_qplib_dealloc_dpi(struct bnxt_qplib_res *res, + struct bnxt_qplib_dpi_tbl *dpit, + struct bnxt_qplib_dpi *dpi) +{ + if (dpi->dpi >= dpit->max) { + dev_warn(&res->pdev->dev, "Invalid DPI? dpi = %d", dpi->dpi); + return -EINVAL; + } + if (test_and_set_bit(dpi->dpi, dpit->tbl)) { + dev_warn(&res->pdev->dev, "Freeing an unused DPI? dpi = %d", + dpi->dpi); + return -EINVAL; + } + if (dpit->app_tbl) + dpit->app_tbl[dpi->dpi] = NULL; + memset(dpi, 0, sizeof(*dpi)); + + return 0; +} + +static void bnxt_qplib_free_dpi_tbl(struct bnxt_qplib_res *res, + struct bnxt_qplib_dpi_tbl *dpit) +{ + kfree(dpit->tbl); + kfree(dpit->app_tbl); + if (dpit->dbr_bar_reg_iomem) + pci_iounmap(res->pdev, dpit->dbr_bar_reg_iomem); + memset(dpit, 0, sizeof(*dpit)); +} + +static int bnxt_qplib_alloc_dpi_tbl(struct bnxt_qplib_res *res, + struct bnxt_qplib_dpi_tbl *dpit, + u32 dbr_offset) +{ + u32 dbr_bar_reg = RCFW_DBR_PCI_BAR_REGION; + resource_size_t bar_reg_base; + u32 dbr_len, bytes; + + if (dpit->dbr_bar_reg_iomem) { + dev_err(&res->pdev->dev, + "QPLIB: DBR BAR region %d already mapped", dbr_bar_reg); + return -EALREADY; + } + + bar_reg_base = pci_resource_start(res->pdev, dbr_bar_reg); + if (!bar_reg_base) { + dev_err(&res->pdev->dev, + "QPLIB: BAR region %d resc start failed", dbr_bar_reg); + return -ENOMEM; + } + + dbr_len = pci_resource_len(res->pdev, dbr_bar_reg) - dbr_offset; + if (!dbr_len || ((dbr_len & (PAGE_SIZE - 1)) != 0)) { + dev_err(&res->pdev->dev, "QPLIB: Invalid DBR length %d", + dbr_len); + return -ENOMEM; + } + + dpit->dbr_bar_reg_iomem = ioremap_nocache(bar_reg_base + dbr_offset, + dbr_len); + if (!dpit->dbr_bar_reg_iomem) { + dev_err(&res->pdev->dev, + "QPLIB: FP: DBR BAR region %d mapping failed", + dbr_bar_reg); + return -ENOMEM; + } + + dpit->unmapped_dbr = bar_reg_base + dbr_offset; + dpit->max = dbr_len / PAGE_SIZE; + + dpit->app_tbl = kcalloc(dpit->max, sizeof(void *), GFP_KERNEL); + if (!dpit->app_tbl) { + pci_iounmap(res->pdev, dpit->dbr_bar_reg_iomem); + dev_err(&res->pdev->dev, + "QPLIB: DPI app tbl allocation failed"); + return -ENOMEM; + } + + bytes = dpit->max >> 3; + if (!bytes) + bytes = 1; + + dpit->tbl = kmalloc(bytes, GFP_KERNEL); + if (!dpit->tbl) { + pci_iounmap(res->pdev, dpit->dbr_bar_reg_iomem); + kfree(dpit->app_tbl); + dpit->app_tbl = NULL; + dev_err(&res->pdev->dev, + "QPLIB: DPI tbl allocation failed for size = %d", + bytes); + return -ENOMEM; + } + + memset((u8 *)dpit->tbl, 0xFF, bytes); + + return 0; +} + +/* PKEYs */ +static void bnxt_qplib_cleanup_pkey_tbl(struct bnxt_qplib_pkey_tbl *pkey_tbl) +{ + memset(pkey_tbl->tbl, 0, sizeof(u16) * pkey_tbl->max); + pkey_tbl->active = 0; +} + +static void bnxt_qplib_init_pkey_tbl(struct bnxt_qplib_res *res, + struct bnxt_qplib_pkey_tbl *pkey_tbl) +{ + u16 pkey = 0xFFFF; + + memset(pkey_tbl->tbl, 0, sizeof(u16) * pkey_tbl->max); + + /* pkey default = 0xFFFF */ + bnxt_qplib_add_pkey(res, pkey_tbl, &pkey, false); +} + +/* Stats */ +static void bnxt_qplib_free_stats_ctx(struct pci_dev *pdev, + struct bnxt_qplib_stats *stats) +{ + if (stats->dma) { + dma_free_coherent(&pdev->dev, stats->size, + stats->dma, stats->dma_map); + } + memset(stats, 0, sizeof(*stats)); + stats->fw_id = -1; +} + +static int bnxt_qplib_alloc_stats_ctx(struct pci_dev *pdev, + struct bnxt_qplib_stats *stats) +{ + memset(stats, 0, sizeof(*stats)); + stats->fw_id = -1; + stats->size = sizeof(struct ctx_hw_stats); + stats->dma = dma_alloc_coherent(&pdev->dev, stats->size, + &stats->dma_map, GFP_KERNEL); + if (!stats->dma) { + dev_err(&pdev->dev, "QPLIB: Stats DMA allocation failed"); + return -ENOMEM; + } + return 0; +} + +void bnxt_qplib_cleanup_res(struct bnxt_qplib_res *res) +{ + bnxt_qplib_cleanup_pkey_tbl(&res->pkey_tbl); + bnxt_qplib_cleanup_sgid_tbl(res, &res->sgid_tbl); +} + +int bnxt_qplib_init_res(struct bnxt_qplib_res *res) +{ + bnxt_qplib_init_sgid_tbl(&res->sgid_tbl, res->netdev); + bnxt_qplib_init_pkey_tbl(res, &res->pkey_tbl); + + return 0; +} + +void bnxt_qplib_free_res(struct bnxt_qplib_res *res) +{ + bnxt_qplib_free_pkey_tbl(res, &res->pkey_tbl); + bnxt_qplib_free_sgid_tbl(res, &res->sgid_tbl); + bnxt_qplib_free_pd_tbl(&res->pd_tbl); + bnxt_qplib_free_dpi_tbl(res, &res->dpi_tbl); + + res->netdev = NULL; + res->pdev = NULL; +} + +int bnxt_qplib_alloc_res(struct bnxt_qplib_res *res, struct pci_dev *pdev, + struct net_device *netdev, + struct bnxt_qplib_dev_attr *dev_attr) +{ + int rc = 0; + + res->pdev = pdev; + res->netdev = netdev; + + rc = bnxt_qplib_alloc_sgid_tbl(res, &res->sgid_tbl, dev_attr->max_sgid); + if (rc) + goto fail; + + rc = bnxt_qplib_alloc_pkey_tbl(res, &res->pkey_tbl, dev_attr->max_pkey); + if (rc) + goto fail; + + rc = bnxt_qplib_alloc_pd_tbl(res, &res->pd_tbl, dev_attr->max_pd); + if (rc) + goto fail; + + rc = bnxt_qplib_alloc_dpi_tbl(res, &res->dpi_tbl, dev_attr->l2_db_size); + if (rc) + goto fail; + + return 0; +fail: + bnxt_qplib_free_res(res); + return rc; +} diff --git a/drivers/infiniband/hw/bnxt_re/qplib_res.h b/drivers/infiniband/hw/bnxt_re/qplib_res.h new file mode 100644 index 000000000000..6277d802ca4b --- /dev/null +++ b/drivers/infiniband/hw/bnxt_re/qplib_res.h @@ -0,0 +1,223 @@ +/* + * Broadcom NetXtreme-E RoCE driver. + * + * Copyright (c) 2016 - 2017, Broadcom. All rights reserved. The term + * Broadcom refers to Broadcom Limited and/or its subsidiaries. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, + * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS + * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE + * OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN + * IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * Description: QPLib resource manager (header) + */ + +#ifndef __BNXT_QPLIB_RES_H__ +#define __BNXT_QPLIB_RES_H__ + +extern const struct bnxt_qplib_gid bnxt_qplib_gid_zero; + +#define PTR_CNT_PER_PG (PAGE_SIZE / sizeof(void *)) +#define PTR_MAX_IDX_PER_PG (PTR_CNT_PER_PG - 1) +#define PTR_PG(x) (((x) & ~PTR_MAX_IDX_PER_PG) / PTR_CNT_PER_PG) +#define PTR_IDX(x) ((x) & PTR_MAX_IDX_PER_PG) + +#define HWQ_CMP(idx, hwq) ((idx) & ((hwq)->max_elements - 1)) + +enum bnxt_qplib_hwq_type { + HWQ_TYPE_CTX, + HWQ_TYPE_QUEUE, + HWQ_TYPE_L2_CMPL +}; + +#define MAX_PBL_LVL_0_PGS 1 +#define MAX_PBL_LVL_1_PGS 512 +#define MAX_PBL_LVL_1_PGS_SHIFT 9 +#define MAX_PBL_LVL_1_PGS_FOR_LVL_2 256 +#define MAX_PBL_LVL_2_PGS (256 * 512) + +enum bnxt_qplib_pbl_lvl { + PBL_LVL_0, + PBL_LVL_1, + PBL_LVL_2, + PBL_LVL_MAX +}; + +#define ROCE_PG_SIZE_4K (4 * 1024) +#define ROCE_PG_SIZE_8K (8 * 1024) +#define ROCE_PG_SIZE_64K (64 * 1024) +#define ROCE_PG_SIZE_2M (2 * 1024 * 1024) +#define ROCE_PG_SIZE_8M (8 * 1024 * 1024) +#define ROCE_PG_SIZE_1G (1024 * 1024 * 1024) + +struct bnxt_qplib_pbl { + u32 pg_count; + u32 pg_size; + void **pg_arr; + dma_addr_t *pg_map_arr; +}; + +struct bnxt_qplib_hwq { + struct pci_dev *pdev; + /* lock to protect qplib_hwq */ + spinlock_t lock; + struct bnxt_qplib_pbl pbl[PBL_LVL_MAX]; + enum bnxt_qplib_pbl_lvl level; /* 0, 1, or 2 */ + /* ptr for easy access to the PBL entries */ + void **pbl_ptr; + /* ptr for easy access to the dma_addr */ + dma_addr_t *pbl_dma_ptr; + u32 max_elements; + u16 element_size; /* Size of each entry */ + + u32 prod; /* raw */ + u32 cons; /* raw */ + u8 cp_bit; + u8 is_user; +}; + +/* Tables */ +struct bnxt_qplib_pd_tbl { + unsigned long *tbl; + u32 max; +}; + +struct bnxt_qplib_sgid_tbl { + struct bnxt_qplib_gid *tbl; + u16 *hw_id; + u16 max; + u16 active; + void *ctx; +}; + +struct bnxt_qplib_pkey_tbl { + u16 *tbl; + u16 max; + u16 active; +}; + +struct bnxt_qplib_dpi { + u32 dpi; + void __iomem *dbr; + u64 umdbr; +}; + +struct bnxt_qplib_dpi_tbl { + void **app_tbl; + unsigned long *tbl; + u16 max; + void __iomem *dbr_bar_reg_iomem; + u64 unmapped_dbr; +}; + +struct bnxt_qplib_stats { + dma_addr_t dma_map; + void *dma; + u32 size; + u32 fw_id; +}; + +struct bnxt_qplib_vf_res { + u32 max_qp_per_vf; + u32 max_mrw_per_vf; + u32 max_srq_per_vf; + u32 max_cq_per_vf; + u32 max_gid_per_vf; +}; + +#define BNXT_QPLIB_MAX_QP_CTX_ENTRY_SIZE 448 +#define BNXT_QPLIB_MAX_SRQ_CTX_ENTRY_SIZE 64 +#define BNXT_QPLIB_MAX_CQ_CTX_ENTRY_SIZE 64 +#define BNXT_QPLIB_MAX_MRW_CTX_ENTRY_SIZE 128 + +struct bnxt_qplib_ctx { + u32 qpc_count; + struct bnxt_qplib_hwq qpc_tbl; + u32 mrw_count; + struct bnxt_qplib_hwq mrw_tbl; + u32 srqc_count; + struct bnxt_qplib_hwq srqc_tbl; + u32 cq_count; + struct bnxt_qplib_hwq cq_tbl; + struct bnxt_qplib_hwq tim_tbl; +#define MAX_TQM_ALLOC_REQ 32 +#define MAX_TQM_ALLOC_BLK_SIZE 8 + u8 tqm_count[MAX_TQM_ALLOC_REQ]; + struct bnxt_qplib_hwq tqm_pde; + u32 tqm_pde_level; + struct bnxt_qplib_hwq tqm_tbl[MAX_TQM_ALLOC_REQ]; + struct bnxt_qplib_stats stats; + struct bnxt_qplib_vf_res vf_res; +}; + +struct bnxt_qplib_res { + struct pci_dev *pdev; + struct net_device *netdev; + + struct bnxt_qplib_rcfw *rcfw; + + struct bnxt_qplib_pd_tbl pd_tbl; + struct bnxt_qplib_sgid_tbl sgid_tbl; + struct bnxt_qplib_pkey_tbl pkey_tbl; + struct bnxt_qplib_dpi_tbl dpi_tbl; +}; + +#define to_bnxt_qplib(ptr, type, member) \ + container_of(ptr, type, member) + +struct bnxt_qplib_pd; +struct bnxt_qplib_dev_attr; + +void bnxt_qplib_free_hwq(struct pci_dev *pdev, struct bnxt_qplib_hwq *hwq); +int bnxt_qplib_alloc_init_hwq(struct pci_dev *pdev, struct bnxt_qplib_hwq *hwq, + struct scatterlist *sl, int nmap, u32 *elements, + u32 elements_per_page, u32 aux, u32 pg_size, + enum bnxt_qplib_hwq_type hwq_type); +void bnxt_qplib_get_guid(u8 *dev_addr, u8 *guid); +int bnxt_qplib_alloc_pd(struct bnxt_qplib_pd_tbl *pd_tbl, + struct bnxt_qplib_pd *pd); +int bnxt_qplib_dealloc_pd(struct bnxt_qplib_res *res, + struct bnxt_qplib_pd_tbl *pd_tbl, + struct bnxt_qplib_pd *pd); +int bnxt_qplib_alloc_dpi(struct bnxt_qplib_dpi_tbl *dpit, + struct bnxt_qplib_dpi *dpi, + void *app); +int bnxt_qplib_dealloc_dpi(struct bnxt_qplib_res *res, + struct bnxt_qplib_dpi_tbl *dpi_tbl, + struct bnxt_qplib_dpi *dpi); +void bnxt_qplib_cleanup_res(struct bnxt_qplib_res *res); +int bnxt_qplib_init_res(struct bnxt_qplib_res *res); +void bnxt_qplib_free_res(struct bnxt_qplib_res *res); +int bnxt_qplib_alloc_res(struct bnxt_qplib_res *res, struct pci_dev *pdev, + struct net_device *netdev, + struct bnxt_qplib_dev_attr *dev_attr); +void bnxt_qplib_free_ctx(struct pci_dev *pdev, + struct bnxt_qplib_ctx *ctx); +int bnxt_qplib_alloc_ctx(struct pci_dev *pdev, + struct bnxt_qplib_ctx *ctx, + bool virt_fn); +#endif /* __BNXT_QPLIB_RES_H__ */ diff --git a/drivers/infiniband/hw/bnxt_re/qplib_sp.c b/drivers/infiniband/hw/bnxt_re/qplib_sp.c new file mode 100644 index 000000000000..7b31eccedf11 --- /dev/null +++ b/drivers/infiniband/hw/bnxt_re/qplib_sp.c @@ -0,0 +1,838 @@ +/* + * Broadcom NetXtreme-E RoCE driver. + * + * Copyright (c) 2016 - 2017, Broadcom. All rights reserved. The term + * Broadcom refers to Broadcom Limited and/or its subsidiaries. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, + * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS + * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE + * OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN + * IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * Description: Slow Path Operators + */ + +#include <linux/interrupt.h> +#include <linux/spinlock.h> +#include <linux/sched.h> +#include <linux/pci.h> + +#include "roce_hsi.h" + +#include "qplib_res.h" +#include "qplib_rcfw.h" +#include "qplib_sp.h" + +const struct bnxt_qplib_gid bnxt_qplib_gid_zero = {{ 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0 } }; + +/* Device */ +int bnxt_qplib_get_dev_attr(struct bnxt_qplib_rcfw *rcfw, + struct bnxt_qplib_dev_attr *attr) +{ + struct cmdq_query_func req; + struct creq_query_func_resp *resp; + struct creq_query_func_resp_sb *sb; + u16 cmd_flags = 0; + u32 temp; + u8 *tqm_alloc; + int i; + + RCFW_CMD_PREP(req, QUERY_FUNC, cmd_flags); + + req.resp_size = sizeof(*sb) / BNXT_QPLIB_CMDQE_UNITS; + resp = (struct creq_query_func_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, (void **)&sb, + 0); + if (!resp) { + dev_err(&rcfw->pdev->dev, "QPLIB: SP: QUERY_FUNC send failed"); + return -EINVAL; + } + if (!bnxt_qplib_rcfw_wait_for_resp(rcfw, le16_to_cpu(req.cookie))) { + /* Cmd timed out */ + dev_err(&rcfw->pdev->dev, "QPLIB: SP: QUERY_FUNC timed out"); + return -ETIMEDOUT; + } + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&rcfw->pdev->dev, "QPLIB: SP: QUERY_FUNC failed "); + dev_err(&rcfw->pdev->dev, + "QPLIB: with status 0x%x cmdq 0x%x resp 0x%x", + resp->status, le16_to_cpu(req.cookie), + le16_to_cpu(resp->cookie)); + return -EINVAL; + } + /* Extract the context from the side buffer */ + attr->max_qp = le32_to_cpu(sb->max_qp); + attr->max_qp_rd_atom = + sb->max_qp_rd_atom > BNXT_QPLIB_MAX_OUT_RD_ATOM ? + BNXT_QPLIB_MAX_OUT_RD_ATOM : sb->max_qp_rd_atom; + attr->max_qp_init_rd_atom = + sb->max_qp_init_rd_atom > BNXT_QPLIB_MAX_OUT_RD_ATOM ? + BNXT_QPLIB_MAX_OUT_RD_ATOM : sb->max_qp_init_rd_atom; + attr->max_qp_wqes = le16_to_cpu(sb->max_qp_wr); + attr->max_qp_sges = sb->max_sge; + attr->max_cq = le32_to_cpu(sb->max_cq); + attr->max_cq_wqes = le32_to_cpu(sb->max_cqe); + attr->max_cq_sges = attr->max_qp_sges; + attr->max_mr = le32_to_cpu(sb->max_mr); + attr->max_mw = le32_to_cpu(sb->max_mw); + + attr->max_mr_size = le64_to_cpu(sb->max_mr_size); + attr->max_pd = 64 * 1024; + attr->max_raw_ethy_qp = le32_to_cpu(sb->max_raw_eth_qp); + attr->max_ah = le32_to_cpu(sb->max_ah); + + attr->max_fmr = le32_to_cpu(sb->max_fmr); + attr->max_map_per_fmr = sb->max_map_per_fmr; + + attr->max_srq = le16_to_cpu(sb->max_srq); + attr->max_srq_wqes = le32_to_cpu(sb->max_srq_wr) - 1; + attr->max_srq_sges = sb->max_srq_sge; + /* Bono only reports 1 PKEY for now, but it can support > 1 */ + attr->max_pkey = le32_to_cpu(sb->max_pkeys); + + attr->max_inline_data = le32_to_cpu(sb->max_inline_data); + attr->l2_db_size = (sb->l2_db_space_size + 1) * PAGE_SIZE; + attr->max_sgid = le32_to_cpu(sb->max_gid); + + strlcpy(attr->fw_ver, "20.6.28.0", sizeof(attr->fw_ver)); + + for (i = 0; i < MAX_TQM_ALLOC_REQ / 4; i++) { + temp = le32_to_cpu(sb->tqm_alloc_reqs[i]); + tqm_alloc = (u8 *)&temp; + attr->tqm_alloc_reqs[i * 4] = *tqm_alloc; + attr->tqm_alloc_reqs[i * 4 + 1] = *(++tqm_alloc); + attr->tqm_alloc_reqs[i * 4 + 2] = *(++tqm_alloc); + attr->tqm_alloc_reqs[i * 4 + 3] = *(++tqm_alloc); + } + return 0; +} + +/* SGID */ +int bnxt_qplib_get_sgid(struct bnxt_qplib_res *res, + struct bnxt_qplib_sgid_tbl *sgid_tbl, int index, + struct bnxt_qplib_gid *gid) +{ + if (index > sgid_tbl->max) { + dev_err(&res->pdev->dev, + "QPLIB: Index %d exceeded SGID table max (%d)", + index, sgid_tbl->max); + return -EINVAL; + } + memcpy(gid, &sgid_tbl->tbl[index], sizeof(*gid)); + return 0; +} + +int bnxt_qplib_del_sgid(struct bnxt_qplib_sgid_tbl *sgid_tbl, + struct bnxt_qplib_gid *gid, bool update) +{ + struct bnxt_qplib_res *res = to_bnxt_qplib(sgid_tbl, + struct bnxt_qplib_res, + sgid_tbl); + struct bnxt_qplib_rcfw *rcfw = res->rcfw; + int index; + + if (!sgid_tbl) { + dev_err(&res->pdev->dev, "QPLIB: SGID table not allocated"); + return -EINVAL; + } + /* Do we need a sgid_lock here? */ + if (!sgid_tbl->active) { + dev_err(&res->pdev->dev, + "QPLIB: SGID table has no active entries"); + return -ENOMEM; + } + for (index = 0; index < sgid_tbl->max; index++) { + if (!memcmp(&sgid_tbl->tbl[index], gid, sizeof(*gid))) + break; + } + if (index == sgid_tbl->max) { + dev_warn(&res->pdev->dev, "GID not found in the SGID table"); + return 0; + } + /* Remove GID from the SGID table */ + if (update) { + struct cmdq_delete_gid req; + struct creq_delete_gid_resp *resp; + u16 cmd_flags = 0; + + RCFW_CMD_PREP(req, DELETE_GID, cmd_flags); + if (sgid_tbl->hw_id[index] == 0xFFFF) { + dev_err(&res->pdev->dev, + "QPLIB: GID entry contains an invalid HW id"); + return -EINVAL; + } + req.gid_index = cpu_to_le16(sgid_tbl->hw_id[index]); + resp = (struct creq_delete_gid_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, NULL, + 0); + if (!resp) { + dev_err(&res->pdev->dev, + "QPLIB: SP: DELETE_GID send failed"); + return -EINVAL; + } + if (!bnxt_qplib_rcfw_wait_for_resp(rcfw, + le16_to_cpu(req.cookie))) { + /* Cmd timed out */ + dev_err(&res->pdev->dev, + "QPLIB: SP: DELETE_GID timed out"); + return -ETIMEDOUT; + } + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&res->pdev->dev, + "QPLIB: SP: DELETE_GID failed "); + dev_err(&res->pdev->dev, + "QPLIB: with status 0x%x cmdq 0x%x resp 0x%x", + resp->status, le16_to_cpu(req.cookie), + le16_to_cpu(resp->cookie)); + return -EINVAL; + } + } + memcpy(&sgid_tbl->tbl[index], &bnxt_qplib_gid_zero, + sizeof(bnxt_qplib_gid_zero)); + sgid_tbl->active--; + dev_dbg(&res->pdev->dev, + "QPLIB: SGID deleted hw_id[0x%x] = 0x%x active = 0x%x", + index, sgid_tbl->hw_id[index], sgid_tbl->active); + sgid_tbl->hw_id[index] = (u16)-1; + + /* unlock */ + return 0; +} + +int bnxt_qplib_add_sgid(struct bnxt_qplib_sgid_tbl *sgid_tbl, + struct bnxt_qplib_gid *gid, u8 *smac, u16 vlan_id, + bool update, u32 *index) +{ + struct bnxt_qplib_res *res = to_bnxt_qplib(sgid_tbl, + struct bnxt_qplib_res, + sgid_tbl); + struct bnxt_qplib_rcfw *rcfw = res->rcfw; + int i, free_idx, rc = 0; + + if (!sgid_tbl) { + dev_err(&res->pdev->dev, "QPLIB: SGID table not allocated"); + return -EINVAL; + } + /* Do we need a sgid_lock here? */ + if (sgid_tbl->active == sgid_tbl->max) { + dev_err(&res->pdev->dev, "QPLIB: SGID table is full"); + return -ENOMEM; + } + free_idx = sgid_tbl->max; + for (i = 0; i < sgid_tbl->max; i++) { + if (!memcmp(&sgid_tbl->tbl[i], gid, sizeof(*gid))) { + dev_dbg(&res->pdev->dev, + "QPLIB: SGID entry already exist in entry %d!", + i); + *index = i; + return -EALREADY; + } else if (!memcmp(&sgid_tbl->tbl[i], &bnxt_qplib_gid_zero, + sizeof(bnxt_qplib_gid_zero)) && + free_idx == sgid_tbl->max) { + free_idx = i; + } + } + if (free_idx == sgid_tbl->max) { + dev_err(&res->pdev->dev, + "QPLIB: SGID table is FULL but count is not MAX??"); + return -ENOMEM; + } + if (update) { + struct cmdq_add_gid req; + struct creq_add_gid_resp *resp; + u16 cmd_flags = 0; + u32 temp32[4]; + u16 temp16[3]; + + RCFW_CMD_PREP(req, ADD_GID, cmd_flags); + + memcpy(temp32, gid->data, sizeof(struct bnxt_qplib_gid)); + req.gid[0] = cpu_to_be32(temp32[3]); + req.gid[1] = cpu_to_be32(temp32[2]); + req.gid[2] = cpu_to_be32(temp32[1]); + req.gid[3] = cpu_to_be32(temp32[0]); + if (vlan_id != 0xFFFF) + req.vlan = cpu_to_le16((vlan_id & + CMDQ_ADD_GID_VLAN_VLAN_ID_MASK) | + CMDQ_ADD_GID_VLAN_TPID_TPID_8100 | + CMDQ_ADD_GID_VLAN_VLAN_EN); + + /* MAC in network format */ + memcpy(temp16, smac, 6); + req.src_mac[0] = cpu_to_be16(temp16[0]); + req.src_mac[1] = cpu_to_be16(temp16[1]); + req.src_mac[2] = cpu_to_be16(temp16[2]); + + resp = (struct creq_add_gid_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, + NULL, 0); + if (!resp) { + dev_err(&res->pdev->dev, + "QPLIB: SP: ADD_GID send failed"); + return -EINVAL; + } + if (!bnxt_qplib_rcfw_wait_for_resp(rcfw, + le16_to_cpu(req.cookie))) { + /* Cmd timed out */ + dev_err(&res->pdev->dev, + "QPIB: SP: ADD_GID timed out"); + return -ETIMEDOUT; + } + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&res->pdev->dev, "QPLIB: SP: ADD_GID failed "); + dev_err(&res->pdev->dev, + "QPLIB: with status 0x%x cmdq 0x%x resp 0x%x", + resp->status, le16_to_cpu(req.cookie), + le16_to_cpu(resp->cookie)); + return -EINVAL; + } + sgid_tbl->hw_id[free_idx] = le32_to_cpu(resp->xid); + } + /* Add GID to the sgid_tbl */ + memcpy(&sgid_tbl->tbl[free_idx], gid, sizeof(*gid)); + sgid_tbl->active++; + dev_dbg(&res->pdev->dev, + "QPLIB: SGID added hw_id[0x%x] = 0x%x active = 0x%x", + free_idx, sgid_tbl->hw_id[free_idx], sgid_tbl->active); + + *index = free_idx; + /* unlock */ + return rc; +} + +/* pkeys */ +int bnxt_qplib_get_pkey(struct bnxt_qplib_res *res, + struct bnxt_qplib_pkey_tbl *pkey_tbl, u16 index, + u16 *pkey) +{ + if (index == 0xFFFF) { + *pkey = 0xFFFF; + return 0; + } + if (index > pkey_tbl->max) { + dev_err(&res->pdev->dev, + "QPLIB: Index %d exceeded PKEY table max (%d)", + index, pkey_tbl->max); + return -EINVAL; + } + memcpy(pkey, &pkey_tbl->tbl[index], sizeof(*pkey)); + return 0; +} + +int bnxt_qplib_del_pkey(struct bnxt_qplib_res *res, + struct bnxt_qplib_pkey_tbl *pkey_tbl, u16 *pkey, + bool update) +{ + int i, rc = 0; + + if (!pkey_tbl) { + dev_err(&res->pdev->dev, "QPLIB: PKEY table not allocated"); + return -EINVAL; + } + + /* Do we need a pkey_lock here? */ + if (!pkey_tbl->active) { + dev_err(&res->pdev->dev, + "QPLIB: PKEY table has no active entries"); + return -ENOMEM; + } + for (i = 0; i < pkey_tbl->max; i++) { + if (!memcmp(&pkey_tbl->tbl[i], pkey, sizeof(*pkey))) + break; + } + if (i == pkey_tbl->max) { + dev_err(&res->pdev->dev, + "QPLIB: PKEY 0x%04x not found in the pkey table", + *pkey); + return -ENOMEM; + } + memset(&pkey_tbl->tbl[i], 0, sizeof(*pkey)); + pkey_tbl->active--; + + /* unlock */ + return rc; +} + +int bnxt_qplib_add_pkey(struct bnxt_qplib_res *res, + struct bnxt_qplib_pkey_tbl *pkey_tbl, u16 *pkey, + bool update) +{ + int i, free_idx, rc = 0; + + if (!pkey_tbl) { + dev_err(&res->pdev->dev, "QPLIB: PKEY table not allocated"); + return -EINVAL; + } + + /* Do we need a pkey_lock here? */ + if (pkey_tbl->active == pkey_tbl->max) { + dev_err(&res->pdev->dev, "QPLIB: PKEY table is full"); + return -ENOMEM; + } + free_idx = pkey_tbl->max; + for (i = 0; i < pkey_tbl->max; i++) { + if (!memcmp(&pkey_tbl->tbl[i], pkey, sizeof(*pkey))) + return -EALREADY; + else if (!pkey_tbl->tbl[i] && free_idx == pkey_tbl->max) + free_idx = i; + } + if (free_idx == pkey_tbl->max) { + dev_err(&res->pdev->dev, + "QPLIB: PKEY table is FULL but count is not MAX??"); + return -ENOMEM; + } + /* Add PKEY to the pkey_tbl */ + memcpy(&pkey_tbl->tbl[free_idx], pkey, sizeof(*pkey)); + pkey_tbl->active++; + + /* unlock */ + return rc; +} + +/* AH */ +int bnxt_qplib_create_ah(struct bnxt_qplib_res *res, struct bnxt_qplib_ah *ah) +{ + struct bnxt_qplib_rcfw *rcfw = res->rcfw; + struct cmdq_create_ah req; + struct creq_create_ah_resp *resp; + u16 cmd_flags = 0; + u32 temp32[4]; + u16 temp16[3]; + + RCFW_CMD_PREP(req, CREATE_AH, cmd_flags); + + memcpy(temp32, ah->dgid.data, sizeof(struct bnxt_qplib_gid)); + req.dgid[0] = cpu_to_le32(temp32[0]); + req.dgid[1] = cpu_to_le32(temp32[1]); + req.dgid[2] = cpu_to_le32(temp32[2]); + req.dgid[3] = cpu_to_le32(temp32[3]); + + req.type = ah->nw_type; + req.hop_limit = ah->hop_limit; + req.sgid_index = cpu_to_le16(res->sgid_tbl.hw_id[ah->sgid_index]); + req.dest_vlan_id_flow_label = cpu_to_le32((ah->flow_label & + CMDQ_CREATE_AH_FLOW_LABEL_MASK) | + CMDQ_CREATE_AH_DEST_VLAN_ID_MASK); + req.pd_id = cpu_to_le32(ah->pd->id); + req.traffic_class = ah->traffic_class; + + /* MAC in network format */ + memcpy(temp16, ah->dmac, 6); + req.dest_mac[0] = cpu_to_le16(temp16[0]); + req.dest_mac[1] = cpu_to_le16(temp16[1]); + req.dest_mac[2] = cpu_to_le16(temp16[2]); + + resp = (struct creq_create_ah_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, + NULL, 1); + if (!resp) { + dev_err(&rcfw->pdev->dev, "QPLIB: SP: CREATE_AH send failed"); + return -EINVAL; + } + if (!bnxt_qplib_rcfw_block_for_resp(rcfw, le16_to_cpu(req.cookie))) { + /* Cmd timed out */ + dev_err(&rcfw->pdev->dev, "QPLIB: SP: CREATE_AH timed out"); + return -ETIMEDOUT; + } + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&rcfw->pdev->dev, "QPLIB: SP: CREATE_AH failed "); + dev_err(&rcfw->pdev->dev, + "QPLIB: with status 0x%x cmdq 0x%x resp 0x%x", + resp->status, le16_to_cpu(req.cookie), + le16_to_cpu(resp->cookie)); + return -EINVAL; + } + ah->id = le32_to_cpu(resp->xid); + return 0; +} + +int bnxt_qplib_destroy_ah(struct bnxt_qplib_res *res, struct bnxt_qplib_ah *ah) +{ + struct bnxt_qplib_rcfw *rcfw = res->rcfw; + struct cmdq_destroy_ah req; + struct creq_destroy_ah_resp *resp; + u16 cmd_flags = 0; + + /* Clean up the AH table in the device */ + RCFW_CMD_PREP(req, DESTROY_AH, cmd_flags); + + req.ah_cid = cpu_to_le32(ah->id); + + resp = (struct creq_destroy_ah_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, + NULL, 1); + if (!resp) { + dev_err(&rcfw->pdev->dev, "QPLIB: SP: DESTROY_AH send failed"); + return -EINVAL; + } + if (!bnxt_qplib_rcfw_block_for_resp(rcfw, le16_to_cpu(req.cookie))) { + /* Cmd timed out */ + dev_err(&rcfw->pdev->dev, "QPLIB: SP: DESTROY_AH timed out"); + return -ETIMEDOUT; + } + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&rcfw->pdev->dev, "QPLIB: SP: DESTROY_AH failed "); + dev_err(&rcfw->pdev->dev, + "QPLIB: with status 0x%x cmdq 0x%x resp 0x%x", + resp->status, le16_to_cpu(req.cookie), + le16_to_cpu(resp->cookie)); + return -EINVAL; + } + return 0; +} + +/* MRW */ +int bnxt_qplib_free_mrw(struct bnxt_qplib_res *res, struct bnxt_qplib_mrw *mrw) +{ + struct bnxt_qplib_rcfw *rcfw = res->rcfw; + struct cmdq_deallocate_key req; + struct creq_deallocate_key_resp *resp; + u16 cmd_flags = 0; + + if (mrw->lkey == 0xFFFFFFFF) { + dev_info(&res->pdev->dev, + "QPLIB: SP: Free a reserved lkey MRW"); + return 0; + } + + RCFW_CMD_PREP(req, DEALLOCATE_KEY, cmd_flags); + + req.mrw_flags = mrw->type; + + if ((mrw->type == CMDQ_ALLOCATE_MRW_MRW_FLAGS_MW_TYPE1) || + (mrw->type == CMDQ_ALLOCATE_MRW_MRW_FLAGS_MW_TYPE2A) || + (mrw->type == CMDQ_ALLOCATE_MRW_MRW_FLAGS_MW_TYPE2B)) + req.key = cpu_to_le32(mrw->rkey); + else + req.key = cpu_to_le32(mrw->lkey); + + resp = (struct creq_deallocate_key_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, + NULL, 0); + if (!resp) { + dev_err(&res->pdev->dev, "QPLIB: SP: FREE_MR send failed"); + return -EINVAL; + } + if (!bnxt_qplib_rcfw_wait_for_resp(rcfw, le16_to_cpu(req.cookie))) { + /* Cmd timed out */ + dev_err(&res->pdev->dev, "QPLIB: SP: FREE_MR timed out"); + return -ETIMEDOUT; + } + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&res->pdev->dev, "QPLIB: SP: FREE_MR failed "); + dev_err(&res->pdev->dev, + "QPLIB: with status 0x%x cmdq 0x%x resp 0x%x", + resp->status, le16_to_cpu(req.cookie), + le16_to_cpu(resp->cookie)); + return -EINVAL; + } + /* Free the qplib's MRW memory */ + if (mrw->hwq.max_elements) + bnxt_qplib_free_hwq(res->pdev, &mrw->hwq); + + return 0; +} + +int bnxt_qplib_alloc_mrw(struct bnxt_qplib_res *res, struct bnxt_qplib_mrw *mrw) +{ + struct bnxt_qplib_rcfw *rcfw = res->rcfw; + struct cmdq_allocate_mrw req; + struct creq_allocate_mrw_resp *resp; + u16 cmd_flags = 0; + unsigned long tmp; + + RCFW_CMD_PREP(req, ALLOCATE_MRW, cmd_flags); + + req.pd_id = cpu_to_le32(mrw->pd->id); + req.mrw_flags = mrw->type; + if ((mrw->type == CMDQ_ALLOCATE_MRW_MRW_FLAGS_PMR && + mrw->flags & BNXT_QPLIB_FR_PMR) || + mrw->type == CMDQ_ALLOCATE_MRW_MRW_FLAGS_MW_TYPE2A || + mrw->type == CMDQ_ALLOCATE_MRW_MRW_FLAGS_MW_TYPE2B) + req.access = CMDQ_ALLOCATE_MRW_ACCESS_CONSUMER_OWNED_KEY; + tmp = (unsigned long)mrw; + req.mrw_handle = cpu_to_le64(tmp); + + resp = (struct creq_allocate_mrw_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, + NULL, 0); + if (!resp) { + dev_err(&rcfw->pdev->dev, "QPLIB: SP: ALLOC_MRW send failed"); + return -EINVAL; + } + if (!bnxt_qplib_rcfw_wait_for_resp(rcfw, le16_to_cpu(req.cookie))) { + /* Cmd timed out */ + dev_err(&rcfw->pdev->dev, "QPLIB: SP: ALLOC_MRW timed out"); + return -ETIMEDOUT; + } + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&rcfw->pdev->dev, "QPLIB: SP: ALLOC_MRW failed "); + dev_err(&rcfw->pdev->dev, + "QPLIB: with status 0x%x cmdq 0x%x resp 0x%x", + resp->status, le16_to_cpu(req.cookie), + le16_to_cpu(resp->cookie)); + return -EINVAL; + } + if ((mrw->type == CMDQ_ALLOCATE_MRW_MRW_FLAGS_MW_TYPE1) || + (mrw->type == CMDQ_ALLOCATE_MRW_MRW_FLAGS_MW_TYPE2A) || + (mrw->type == CMDQ_ALLOCATE_MRW_MRW_FLAGS_MW_TYPE2B)) + mrw->rkey = le32_to_cpu(resp->xid); + else + mrw->lkey = le32_to_cpu(resp->xid); + return 0; +} + +int bnxt_qplib_dereg_mrw(struct bnxt_qplib_res *res, struct bnxt_qplib_mrw *mrw, + bool block) +{ + struct bnxt_qplib_rcfw *rcfw = res->rcfw; + struct cmdq_deregister_mr req; + struct creq_deregister_mr_resp *resp; + u16 cmd_flags = 0; + int rc; + + RCFW_CMD_PREP(req, DEREGISTER_MR, cmd_flags); + + req.lkey = cpu_to_le32(mrw->lkey); + resp = (struct creq_deregister_mr_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, + NULL, block); + if (!resp) { + dev_err(&rcfw->pdev->dev, "QPLIB: SP: DEREG_MR send failed"); + return -EINVAL; + } + if (block) + rc = bnxt_qplib_rcfw_block_for_resp(rcfw, + le16_to_cpu(req.cookie)); + else + rc = bnxt_qplib_rcfw_wait_for_resp(rcfw, + le16_to_cpu(req.cookie)); + if (!rc) { + /* Cmd timed out */ + dev_err(&res->pdev->dev, "QPLIB: SP: DEREG_MR timed out"); + return -ETIMEDOUT; + } + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&rcfw->pdev->dev, "QPLIB: SP: DEREG_MR failed "); + dev_err(&rcfw->pdev->dev, + "QPLIB: with status 0x%x cmdq 0x%x resp 0x%x", + resp->status, le16_to_cpu(req.cookie), + le16_to_cpu(resp->cookie)); + return -EINVAL; + } + + /* Free the qplib's MR memory */ + if (mrw->hwq.max_elements) { + mrw->va = 0; + mrw->total_size = 0; + bnxt_qplib_free_hwq(res->pdev, &mrw->hwq); + } + + return 0; +} + +int bnxt_qplib_reg_mr(struct bnxt_qplib_res *res, struct bnxt_qplib_mrw *mr, + u64 *pbl_tbl, int num_pbls, bool block) +{ + struct bnxt_qplib_rcfw *rcfw = res->rcfw; + struct cmdq_register_mr req; + struct creq_register_mr_resp *resp; + u16 cmd_flags = 0, level; + int pg_ptrs, pages, i, rc; + dma_addr_t **pbl_ptr; + u32 pg_size; + + if (num_pbls) { + pg_ptrs = roundup_pow_of_two(num_pbls); + pages = pg_ptrs >> MAX_PBL_LVL_1_PGS_SHIFT; + if (!pages) + pages++; + + if (pages > MAX_PBL_LVL_1_PGS) { + dev_err(&res->pdev->dev, "QPLIB: SP: Reg MR pages "); + dev_err(&res->pdev->dev, + "requested (0x%x) exceeded max (0x%x)", + pages, MAX_PBL_LVL_1_PGS); + return -ENOMEM; + } + /* Free the hwq if it already exist, must be a rereg */ + if (mr->hwq.max_elements) + bnxt_qplib_free_hwq(res->pdev, &mr->hwq); + + mr->hwq.max_elements = pages; + rc = bnxt_qplib_alloc_init_hwq(res->pdev, &mr->hwq, NULL, 0, + &mr->hwq.max_elements, + PAGE_SIZE, 0, PAGE_SIZE, + HWQ_TYPE_CTX); + if (rc) { + dev_err(&res->pdev->dev, + "SP: Reg MR memory allocation failed"); + return -ENOMEM; + } + /* Write to the hwq */ + pbl_ptr = (dma_addr_t **)mr->hwq.pbl_ptr; + for (i = 0; i < num_pbls; i++) + pbl_ptr[PTR_PG(i)][PTR_IDX(i)] = + (pbl_tbl[i] & PAGE_MASK) | PTU_PTE_VALID; + } + + RCFW_CMD_PREP(req, REGISTER_MR, cmd_flags); + + /* Configure the request */ + if (mr->hwq.level == PBL_LVL_MAX) { + level = 0; + req.pbl = 0; + pg_size = PAGE_SIZE; + } else { + level = mr->hwq.level + 1; + req.pbl = cpu_to_le64(mr->hwq.pbl[PBL_LVL_0].pg_map_arr[0]); + pg_size = mr->hwq.pbl[PBL_LVL_0].pg_size; + } + req.log2_pg_size_lvl = (level << CMDQ_REGISTER_MR_LVL_SFT) | + ((ilog2(pg_size) << + CMDQ_REGISTER_MR_LOG2_PG_SIZE_SFT) & + CMDQ_REGISTER_MR_LOG2_PG_SIZE_MASK); + req.access = (mr->flags & 0xFFFF); + req.va = cpu_to_le64(mr->va); + req.key = cpu_to_le32(mr->lkey); + req.mr_size = cpu_to_le64(mr->total_size); + + resp = (struct creq_register_mr_resp *) + bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, + NULL, block); + if (!resp) { + dev_err(&res->pdev->dev, "SP: REG_MR send failed"); + rc = -EINVAL; + goto fail; + } + if (block) + rc = bnxt_qplib_rcfw_block_for_resp(rcfw, + le16_to_cpu(req.cookie)); + else + rc = bnxt_qplib_rcfw_wait_for_resp(rcfw, + le16_to_cpu(req.cookie)); + if (!rc) { + /* Cmd timed out */ + dev_err(&res->pdev->dev, "SP: REG_MR timed out"); + rc = -ETIMEDOUT; + goto fail; + } + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&res->pdev->dev, "QPLIB: SP: REG_MR failed "); + dev_err(&res->pdev->dev, + "QPLIB: SP: with status 0x%x cmdq 0x%x resp 0x%x", + resp->status, le16_to_cpu(req.cookie), + le16_to_cpu(resp->cookie)); + rc = -EINVAL; + goto fail; + } + return 0; + +fail: + if (mr->hwq.max_elements) + bnxt_qplib_free_hwq(res->pdev, &mr->hwq); + return rc; +} + +int bnxt_qplib_alloc_fast_reg_page_list(struct bnxt_qplib_res *res, + struct bnxt_qplib_frpl *frpl, + int max_pg_ptrs) +{ + int pg_ptrs, pages, rc; + + /* Re-calculate the max to fit the HWQ allocation model */ + pg_ptrs = roundup_pow_of_two(max_pg_ptrs); + pages = pg_ptrs >> MAX_PBL_LVL_1_PGS_SHIFT; + if (!pages) + pages++; + + if (pages > MAX_PBL_LVL_1_PGS) + return -ENOMEM; + + frpl->hwq.max_elements = pages; + rc = bnxt_qplib_alloc_init_hwq(res->pdev, &frpl->hwq, NULL, 0, + &frpl->hwq.max_elements, PAGE_SIZE, 0, + PAGE_SIZE, HWQ_TYPE_CTX); + if (!rc) + frpl->max_pg_ptrs = pg_ptrs; + + return rc; +} + +int bnxt_qplib_free_fast_reg_page_list(struct bnxt_qplib_res *res, + struct bnxt_qplib_frpl *frpl) +{ + bnxt_qplib_free_hwq(res->pdev, &frpl->hwq); + return 0; +} + +int bnxt_qplib_map_tc2cos(struct bnxt_qplib_res *res, u16 *cids) +{ + struct bnxt_qplib_rcfw *rcfw = res->rcfw; + struct cmdq_map_tc_to_cos req; + struct creq_map_tc_to_cos_resp *resp; + u16 cmd_flags = 0; + int tleft; + + RCFW_CMD_PREP(req, MAP_TC_TO_COS, cmd_flags); + req.cos0 = cpu_to_le16(cids[0]); + req.cos1 = cpu_to_le16(cids[1]); + + resp = bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, NULL, 0); + if (!resp) { + dev_err(&res->pdev->dev, "QPLIB: SP: MAP_TC2COS send failed"); + return -EINVAL; + } + + tleft = bnxt_qplib_rcfw_block_for_resp(rcfw, le16_to_cpu(req.cookie)); + if (!tleft) { + dev_err(&res->pdev->dev, "QPLIB: SP: MAP_TC2COS timed out"); + return -ETIMEDOUT; + } + + if (resp->status || + le16_to_cpu(resp->cookie) != le16_to_cpu(req.cookie)) { + dev_err(&res->pdev->dev, "QPLIB: SP: MAP_TC2COS failed "); + dev_err(&res->pdev->dev, + "QPLIB: with status 0x%x cmdq 0x%x resp 0x%x", + resp->status, le16_to_cpu(req.cookie), + le16_to_cpu(resp->cookie)); + return -EINVAL; + } + + return 0; +} diff --git a/drivers/infiniband/hw/bnxt_re/qplib_sp.h b/drivers/infiniband/hw/bnxt_re/qplib_sp.h new file mode 100644 index 000000000000..1442a617e968 --- /dev/null +++ b/drivers/infiniband/hw/bnxt_re/qplib_sp.h @@ -0,0 +1,160 @@ +/* + * Broadcom NetXtreme-E RoCE driver. + * + * Copyright (c) 2016 - 2017, Broadcom. All rights reserved. The term + * Broadcom refers to Broadcom Limited and/or its subsidiaries. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, + * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS + * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE + * OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN + * IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * Description: Slow Path Operators (header) + * + */ + +#ifndef __BNXT_QPLIB_SP_H__ +#define __BNXT_QPLIB_SP_H__ + +struct bnxt_qplib_dev_attr { + char fw_ver[32]; + u16 max_sgid; + u16 max_mrw; + u32 max_qp; +#define BNXT_QPLIB_MAX_OUT_RD_ATOM 126 + u32 max_qp_rd_atom; + u32 max_qp_init_rd_atom; + u32 max_qp_wqes; + u32 max_qp_sges; + u32 max_cq; + u32 max_cq_wqes; + u32 max_cq_sges; + u32 max_mr; + u64 max_mr_size; + u32 max_pd; + u32 max_mw; + u32 max_raw_ethy_qp; + u32 max_ah; + u32 max_fmr; + u32 max_map_per_fmr; + u32 max_srq; + u32 max_srq_wqes; + u32 max_srq_sges; + u32 max_pkey; + u32 max_inline_data; + u32 l2_db_size; + u8 tqm_alloc_reqs[MAX_TQM_ALLOC_REQ]; +}; + +struct bnxt_qplib_pd { + u32 id; +}; + +struct bnxt_qplib_gid { + u8 data[16]; +}; + +struct bnxt_qplib_ah { + struct bnxt_qplib_gid dgid; + struct bnxt_qplib_pd *pd; + u32 id; + u8 sgid_index; + /* For Query AH if the hw table and SW table are differnt */ + u8 host_sgid_index; + u8 traffic_class; + u32 flow_label; + u8 hop_limit; + u8 sl; + u8 dmac[6]; + u16 vlan_id; + u8 nw_type; +}; + +struct bnxt_qplib_mrw { + struct bnxt_qplib_pd *pd; + int type; + u32 flags; +#define BNXT_QPLIB_FR_PMR 0x80000000 + u32 lkey; + u32 rkey; +#define BNXT_QPLIB_RSVD_LKEY 0xFFFFFFFF + u64 va; + u64 total_size; + u32 npages; + u64 mr_handle; + struct bnxt_qplib_hwq hwq; +}; + +struct bnxt_qplib_frpl { + int max_pg_ptrs; + struct bnxt_qplib_hwq hwq; +}; + +#define BNXT_QPLIB_ACCESS_LOCAL_WRITE BIT(0) +#define BNXT_QPLIB_ACCESS_REMOTE_READ BIT(1) +#define BNXT_QPLIB_ACCESS_REMOTE_WRITE BIT(2) +#define BNXT_QPLIB_ACCESS_REMOTE_ATOMIC BIT(3) +#define BNXT_QPLIB_ACCESS_MW_BIND BIT(4) +#define BNXT_QPLIB_ACCESS_ZERO_BASED BIT(5) +#define BNXT_QPLIB_ACCESS_ON_DEMAND BIT(6) + +int bnxt_qplib_get_sgid(struct bnxt_qplib_res *res, + struct bnxt_qplib_sgid_tbl *sgid_tbl, int index, + struct bnxt_qplib_gid *gid); +int bnxt_qplib_del_sgid(struct bnxt_qplib_sgid_tbl *sgid_tbl, + struct bnxt_qplib_gid *gid, bool update); +int bnxt_qplib_add_sgid(struct bnxt_qplib_sgid_tbl *sgid_tbl, + struct bnxt_qplib_gid *gid, u8 *mac, u16 vlan_id, + bool update, u32 *index); +int bnxt_qplib_get_pkey(struct bnxt_qplib_res *res, + struct bnxt_qplib_pkey_tbl *pkey_tbl, u16 index, + u16 *pkey); +int bnxt_qplib_del_pkey(struct bnxt_qplib_res *res, + struct bnxt_qplib_pkey_tbl *pkey_tbl, u16 *pkey, + bool update); +int bnxt_qplib_add_pkey(struct bnxt_qplib_res *res, + struct bnxt_qplib_pkey_tbl *pkey_tbl, u16 *pkey, + bool update); +int bnxt_qplib_get_dev_attr(struct bnxt_qplib_rcfw *rcfw, + struct bnxt_qplib_dev_attr *attr); +int bnxt_qplib_create_ah(struct bnxt_qplib_res *res, struct bnxt_qplib_ah *ah); +int bnxt_qplib_destroy_ah(struct bnxt_qplib_res *res, struct bnxt_qplib_ah *ah); +int bnxt_qplib_alloc_mrw(struct bnxt_qplib_res *res, + struct bnxt_qplib_mrw *mrw); +int bnxt_qplib_dereg_mrw(struct bnxt_qplib_res *res, struct bnxt_qplib_mrw *mrw, + bool block); +int bnxt_qplib_reg_mr(struct bnxt_qplib_res *res, struct bnxt_qplib_mrw *mr, + u64 *pbl_tbl, int num_pbls, bool block); +int bnxt_qplib_free_mrw(struct bnxt_qplib_res *res, struct bnxt_qplib_mrw *mr); +int bnxt_qplib_alloc_fast_reg_mr(struct bnxt_qplib_res *res, + struct bnxt_qplib_mrw *mr, int max); +int bnxt_qplib_alloc_fast_reg_page_list(struct bnxt_qplib_res *res, + struct bnxt_qplib_frpl *frpl, int max); +int bnxt_qplib_free_fast_reg_page_list(struct bnxt_qplib_res *res, + struct bnxt_qplib_frpl *frpl); +int bnxt_qplib_map_tc2cos(struct bnxt_qplib_res *res, u16 *cids); +#endif /* __BNXT_QPLIB_SP_H__*/ diff --git a/drivers/infiniband/hw/bnxt_re/roce_hsi.h b/drivers/infiniband/hw/bnxt_re/roce_hsi.h new file mode 100644 index 000000000000..fc23477ac52f --- /dev/null +++ b/drivers/infiniband/hw/bnxt_re/roce_hsi.h @@ -0,0 +1,2821 @@ +/* + * Broadcom NetXtreme-E RoCE driver. + * + * Copyright (c) 2016 - 2017, Broadcom. All rights reserved. The term + * Broadcom refers to Broadcom Limited and/or its subsidiaries. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, + * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS + * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE + * OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN + * IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * Description: RoCE HSI File - Autogenerated + */ + +#ifndef __BNXT_RE_HSI_H__ +#define __BNXT_RE_HSI_H__ + +/* include bnxt_hsi.h from bnxt_en driver */ +#include "bnxt_hsi.h" + +/* CMP Door Bell Format (4 bytes) */ +struct cmpl_doorbell { + __le32 key_mask_valid_idx; + #define CMPL_DOORBELL_IDX_MASK 0xffffffUL + #define CMPL_DOORBELL_IDX_SFT 0 + #define CMPL_DOORBELL_RESERVED_MASK 0x3000000UL + #define CMPL_DOORBELL_RESERVED_SFT 24 + #define CMPL_DOORBELL_IDX_VALID 0x4000000UL + #define CMPL_DOORBELL_MASK 0x8000000UL + #define CMPL_DOORBELL_KEY_MASK 0xf0000000UL + #define CMPL_DOORBELL_KEY_SFT 28 + #define CMPL_DOORBELL_KEY_CMPL (0x2UL << 28) +}; + +/* Status Door Bell Format (4 bytes) */ +struct status_doorbell { + __le32 key_idx; + #define STATUS_DOORBELL_IDX_MASK 0xffffffUL + #define STATUS_DOORBELL_IDX_SFT 0 + #define STATUS_DOORBELL_RESERVED_MASK 0xf000000UL + #define STATUS_DOORBELL_RESERVED_SFT 24 + #define STATUS_DOORBELL_KEY_MASK 0xf0000000UL + #define STATUS_DOORBELL_KEY_SFT 28 + #define STATUS_DOORBELL_KEY_STAT (0x3UL << 28) +}; + +/* RoCE Host Structures */ + +/* Doorbell Structures */ +/* 64b Doorbell Format (8 bytes) */ +struct dbr_dbr { + __le32 index; + #define DBR_DBR_INDEX_MASK 0xfffffUL + #define DBR_DBR_INDEX_SFT 0 + #define DBR_DBR_RESERVED12_MASK 0xfff00000UL + #define DBR_DBR_RESERVED12_SFT 20 + __le32 type_xid; + #define DBR_DBR_XID_MASK 0xfffffUL + #define DBR_DBR_XID_SFT 0 + #define DBR_DBR_RESERVED8_MASK 0xff00000UL + #define DBR_DBR_RESERVED8_SFT 20 + #define DBR_DBR_TYPE_MASK 0xf0000000UL + #define DBR_DBR_TYPE_SFT 28 + #define DBR_DBR_TYPE_SQ (0x0UL << 28) + #define DBR_DBR_TYPE_RQ (0x1UL << 28) + #define DBR_DBR_TYPE_SRQ (0x2UL << 28) + #define DBR_DBR_TYPE_SRQ_ARM (0x3UL << 28) + #define DBR_DBR_TYPE_CQ (0x4UL << 28) + #define DBR_DBR_TYPE_CQ_ARMSE (0x5UL << 28) + #define DBR_DBR_TYPE_CQ_ARMALL (0x6UL << 28) + #define DBR_DBR_TYPE_CQ_ARMENA (0x7UL << 28) + #define DBR_DBR_TYPE_SRQ_ARMENA (0x8UL << 28) + #define DBR_DBR_TYPE_CQ_CUTOFF_ACK (0x9UL << 28) + #define DBR_DBR_TYPE_NULL (0xfUL << 28) +}; + +/* 32b Doorbell Format (4 bytes) */ +struct dbr_dbr32 { + __le32 type_abs_incr_xid; + #define DBR_DBR32_XID_MASK 0xfffffUL + #define DBR_DBR32_XID_SFT 0 + #define DBR_DBR32_RESERVED4_MASK 0xf00000UL + #define DBR_DBR32_RESERVED4_SFT 20 + #define DBR_DBR32_INCR_MASK 0xf000000UL + #define DBR_DBR32_INCR_SFT 24 + #define DBR_DBR32_ABS 0x10000000UL + #define DBR_DBR32_TYPE_MASK 0xe0000000UL + #define DBR_DBR32_TYPE_SFT 29 + #define DBR_DBR32_TYPE_SQ (0x0UL << 29) +}; + +/* SQ WQE Structures */ +/* Base SQ WQE (8 bytes) */ +struct sq_base { + u8 wqe_type; + #define SQ_BASE_WQE_TYPE_SEND 0x0UL + #define SQ_BASE_WQE_TYPE_SEND_W_IMMEAD 0x1UL + #define SQ_BASE_WQE_TYPE_SEND_W_INVALID 0x2UL + #define SQ_BASE_WQE_TYPE_WRITE_WQE 0x4UL + #define SQ_BASE_WQE_TYPE_WRITE_W_IMMEAD 0x5UL + #define SQ_BASE_WQE_TYPE_READ_WQE 0x6UL + #define SQ_BASE_WQE_TYPE_ATOMIC_CS 0x8UL + #define SQ_BASE_WQE_TYPE_ATOMIC_FA 0xbUL + #define SQ_BASE_WQE_TYPE_LOCAL_INVALID 0xcUL + #define SQ_BASE_WQE_TYPE_FR_PMR 0xdUL + #define SQ_BASE_WQE_TYPE_BIND 0xeUL + u8 unused_0[7]; +}; + +/* WQE SGE (16 bytes) */ +struct sq_sge { + __le64 va_or_pa; + __le32 l_key; + __le32 size; +}; + +/* PSN Search Structure (8 bytes) */ +struct sq_psn_search { + __le32 opcode_start_psn; + #define SQ_PSN_SEARCH_START_PSN_MASK 0xffffffUL + #define SQ_PSN_SEARCH_START_PSN_SFT 0 + #define SQ_PSN_SEARCH_OPCODE_MASK 0xff000000UL + #define SQ_PSN_SEARCH_OPCODE_SFT 24 + __le32 flags_next_psn; + #define SQ_PSN_SEARCH_NEXT_PSN_MASK 0xffffffUL + #define SQ_PSN_SEARCH_NEXT_PSN_SFT 0 + #define SQ_PSN_SEARCH_FLAGS_MASK 0xff000000UL + #define SQ_PSN_SEARCH_FLAGS_SFT 24 +}; + +/* Send SQ WQE (40 bytes) */ +struct sq_send { + u8 wqe_type; + #define SQ_SEND_WQE_TYPE_SEND 0x0UL + #define SQ_SEND_WQE_TYPE_SEND_W_IMMEAD 0x1UL + #define SQ_SEND_WQE_TYPE_SEND_W_INVALID 0x2UL + u8 flags; + #define SQ_SEND_FLAGS_SIGNAL_COMP 0x1UL + #define SQ_SEND_FLAGS_RD_OR_ATOMIC_FENCE 0x2UL + #define SQ_SEND_FLAGS_UC_FENCE 0x4UL + #define SQ_SEND_FLAGS_SE 0x8UL + #define SQ_SEND_FLAGS_INLINE 0x10UL + u8 wqe_size; + u8 reserved8_1; + __le32 inv_key_or_imm_data; + __le32 length; + __le32 q_key; + __le32 dst_qp; + #define SQ_SEND_DST_QP_MASK 0xffffffUL + #define SQ_SEND_DST_QP_SFT 0 + #define SQ_SEND_RESERVED8_2_MASK 0xff000000UL + #define SQ_SEND_RESERVED8_2_SFT 24 + __le32 avid; + #define SQ_SEND_AVID_MASK 0xfffffUL + #define SQ_SEND_AVID_SFT 0 + #define SQ_SEND_RESERVED_AVID_MASK 0xfff00000UL + #define SQ_SEND_RESERVED_AVID_SFT 20 + __le64 reserved64; + __le32 data[24]; +}; + +/* Send Raw Ethernet and QP1 SQ WQE (40 bytes) */ +struct sq_send_raweth_qp1 { + u8 wqe_type; + #define SQ_SEND_RAWETH_QP1_WQE_TYPE_SEND 0x0UL + u8 flags; + #define SQ_SEND_RAWETH_QP1_FLAGS_SIGNAL_COMP 0x1UL + #define SQ_SEND_RAWETH_QP1_FLAGS_RD_OR_ATOMIC_FENCE 0x2UL + #define SQ_SEND_RAWETH_QP1_FLAGS_UC_FENCE 0x4UL + #define SQ_SEND_RAWETH_QP1_FLAGS_SE 0x8UL + #define SQ_SEND_RAWETH_QP1_FLAGS_INLINE 0x10UL + u8 wqe_size; + u8 reserved8; + __le16 lflags; + #define SQ_SEND_RAWETH_QP1_LFLAGS_TCP_UDP_CHKSUM 0x1UL + #define SQ_SEND_RAWETH_QP1_LFLAGS_IP_CHKSUM 0x2UL + #define SQ_SEND_RAWETH_QP1_LFLAGS_NOCRC 0x4UL + #define SQ_SEND_RAWETH_QP1_LFLAGS_STAMP 0x8UL + #define SQ_SEND_RAWETH_QP1_LFLAGS_T_IP_CHKSUM 0x10UL + #define SQ_SEND_RAWETH_QP1_LFLAGS_RESERVED1_1 0x20UL + #define SQ_SEND_RAWETH_QP1_LFLAGS_RESERVED1_2 0x40UL + #define SQ_SEND_RAWETH_QP1_LFLAGS_RESERVED1_3 0x80UL + #define SQ_SEND_RAWETH_QP1_LFLAGS_ROCE_CRC 0x100UL + #define SQ_SEND_RAWETH_QP1_LFLAGS_FCOE_CRC 0x200UL + __le16 cfa_action; + __le32 length; + __le32 reserved32_1; + __le32 cfa_meta; + #define SQ_SEND_RAWETH_QP1_CFA_META_VLAN_VID_MASK 0xfffUL + #define SQ_SEND_RAWETH_QP1_CFA_META_VLAN_VID_SFT 0 + #define SQ_SEND_RAWETH_QP1_CFA_META_VLAN_DE 0x1000UL + #define SQ_SEND_RAWETH_QP1_CFA_META_VLAN_PRI_MASK 0xe000UL + #define SQ_SEND_RAWETH_QP1_CFA_META_VLAN_PRI_SFT 13 + #define SQ_SEND_RAWETH_QP1_CFA_META_VLAN_TPID_MASK 0x70000UL + #define SQ_SEND_RAWETH_QP1_CFA_META_VLAN_TPID_SFT 16 + #define SQ_SEND_RAWETH_QP1_CFA_META_VLAN_TPID_TPID88A8 (0x0UL << 16) + #define SQ_SEND_RAWETH_QP1_CFA_META_VLAN_TPID_TPID8100 (0x1UL << 16) + #define SQ_SEND_RAWETH_QP1_CFA_META_VLAN_TPID_TPID9100 (0x2UL << 16) + #define SQ_SEND_RAWETH_QP1_CFA_META_VLAN_TPID_TPID9200 (0x3UL << 16) + #define SQ_SEND_RAWETH_QP1_CFA_META_VLAN_TPID_TPID9300 (0x4UL << 16) + #define SQ_SEND_RAWETH_QP1_CFA_META_VLAN_TPID_TPIDCFG (0x5UL << 16) + #define SQ_SEND_RAWETH_QP1_CFA_META_VLAN_TPID_LAST \ + SQ_SEND_RAWETH_QP1_CFA_META_VLAN_TPID_TPIDCFG + #define SQ_SEND_RAWETH_QP1_CFA_META_VLAN_RESERVED_MASK 0xff80000UL + #define SQ_SEND_RAWETH_QP1_CFA_META_VLAN_RESERVED_SFT 19 + #define SQ_SEND_RAWETH_QP1_CFA_META_KEY_MASK 0xf0000000UL + #define SQ_SEND_RAWETH_QP1_CFA_META_KEY_SFT 28 + #define SQ_SEND_RAWETH_QP1_CFA_META_KEY_NONE (0x0UL << 28) + #define SQ_SEND_RAWETH_QP1_CFA_META_KEY_VLAN_TAG (0x1UL << 28) + #define SQ_SEND_RAWETH_QP1_CFA_META_KEY_LAST \ + SQ_SEND_RAWETH_QP1_CFA_META_KEY_VLAN_TAG + __le32 reserved32_2; + __le64 reserved64; + __le32 data[24]; +}; + +/* RDMA SQ WQE (40 bytes) */ +struct sq_rdma { + u8 wqe_type; + #define SQ_RDMA_WQE_TYPE_WRITE_WQE 0x4UL + #define SQ_RDMA_WQE_TYPE_WRITE_W_IMMEAD 0x5UL + #define SQ_RDMA_WQE_TYPE_READ_WQE 0x6UL + u8 flags; + #define SQ_RDMA_FLAGS_SIGNAL_COMP 0x1UL + #define SQ_RDMA_FLAGS_RD_OR_ATOMIC_FENCE 0x2UL + #define SQ_RDMA_FLAGS_UC_FENCE 0x4UL + #define SQ_RDMA_FLAGS_SE 0x8UL + #define SQ_RDMA_FLAGS_INLINE 0x10UL + u8 wqe_size; + u8 reserved8; + __le32 imm_data; + __le32 length; + __le32 reserved32_1; + __le64 remote_va; + __le32 remote_key; + __le32 reserved32_2; + __le32 data[24]; +}; + +/* Atomic SQ WQE (40 bytes) */ +struct sq_atomic { + u8 wqe_type; + #define SQ_ATOMIC_WQE_TYPE_ATOMIC_CS 0x8UL + #define SQ_ATOMIC_WQE_TYPE_ATOMIC_FA 0xbUL + u8 flags; + #define SQ_ATOMIC_FLAGS_SIGNAL_COMP 0x1UL + #define SQ_ATOMIC_FLAGS_RD_OR_ATOMIC_FENCE 0x2UL + #define SQ_ATOMIC_FLAGS_UC_FENCE 0x4UL + #define SQ_ATOMIC_FLAGS_SE 0x8UL + #define SQ_ATOMIC_FLAGS_INLINE 0x10UL + __le16 reserved16; + __le32 remote_key; + __le64 remote_va; + __le64 swap_data; + __le64 cmp_data; + __le32 data[24]; +}; + +/* Local Invalidate SQ WQE (40 bytes) */ +struct sq_localinvalidate { + u8 wqe_type; + #define SQ_LOCALINVALIDATE_WQE_TYPE_LOCAL_INVALID 0xcUL + u8 flags; + #define SQ_LOCALINVALIDATE_FLAGS_SIGNAL_COMP 0x1UL + #define SQ_LOCALINVALIDATE_FLAGS_RD_OR_ATOMIC_FENCE 0x2UL + #define SQ_LOCALINVALIDATE_FLAGS_UC_FENCE 0x4UL + #define SQ_LOCALINVALIDATE_FLAGS_SE 0x8UL + #define SQ_LOCALINVALIDATE_FLAGS_INLINE 0x10UL + __le16 reserved16; + __le32 inv_l_key; + __le64 reserved64; + __le32 reserved128[4]; + __le32 data[24]; +}; + +/* FR-PMR SQ WQE (40 bytes) */ +struct sq_fr_pmr { + u8 wqe_type; + #define SQ_FR_PMR_WQE_TYPE_FR_PMR 0xdUL + u8 flags; + #define SQ_FR_PMR_FLAGS_SIGNAL_COMP 0x1UL + #define SQ_FR_PMR_FLAGS_RD_OR_ATOMIC_FENCE 0x2UL + #define SQ_FR_PMR_FLAGS_UC_FENCE 0x4UL + #define SQ_FR_PMR_FLAGS_SE 0x8UL + #define SQ_FR_PMR_FLAGS_INLINE 0x10UL + u8 access_cntl; + #define SQ_FR_PMR_ACCESS_CNTL_LOCAL_WRITE 0x1UL + #define SQ_FR_PMR_ACCESS_CNTL_REMOTE_READ 0x2UL + #define SQ_FR_PMR_ACCESS_CNTL_REMOTE_WRITE 0x4UL + #define SQ_FR_PMR_ACCESS_CNTL_REMOTE_ATOMIC 0x8UL + #define SQ_FR_PMR_ACCESS_CNTL_WINDOW_BIND 0x10UL + u8 zero_based_page_size_log; + #define SQ_FR_PMR_PAGE_SIZE_LOG_MASK 0x1fUL + #define SQ_FR_PMR_PAGE_SIZE_LOG_SFT 0 + #define SQ_FR_PMR_PAGE_SIZE_LOG_PGSZ_4K 0x0UL + #define SQ_FR_PMR_PAGE_SIZE_LOG_PGSZ_8K 0x1UL + #define SQ_FR_PMR_PAGE_SIZE_LOG_PGSZ_64K 0x4UL + #define SQ_FR_PMR_PAGE_SIZE_LOG_PGSZ_256K 0x6UL + #define SQ_FR_PMR_PAGE_SIZE_LOG_PGSZ_1M 0x8UL + #define SQ_FR_PMR_PAGE_SIZE_LOG_PGSZ_2M 0x9UL + #define SQ_FR_PMR_PAGE_SIZE_LOG_PGSZ_4M 0xaUL + #define SQ_FR_PMR_PAGE_SIZE_LOG_PGSZ_1G 0x12UL + #define SQ_FR_PMR_ZERO_BASED 0x20UL + #define SQ_FR_PMR_RESERVED2_MASK 0xc0UL + #define SQ_FR_PMR_RESERVED2_SFT 6 + __le32 l_key; + u8 length[5]; + u8 reserved8_1; + u8 reserved8_2; + u8 numlevels_pbl_page_size_log; + #define SQ_FR_PMR_PBL_PAGE_SIZE_LOG_MASK 0x1fUL + #define SQ_FR_PMR_PBL_PAGE_SIZE_LOG_SFT 0 + #define SQ_FR_PMR_PBL_PAGE_SIZE_LOG_PGSZ_4K 0x0UL + #define SQ_FR_PMR_PBL_PAGE_SIZE_LOG_PGSZ_8K 0x1UL + #define SQ_FR_PMR_PBL_PAGE_SIZE_LOG_PGSZ_64K 0x4UL + #define SQ_FR_PMR_PBL_PAGE_SIZE_LOG_PGSZ_256K 0x6UL + #define SQ_FR_PMR_PBL_PAGE_SIZE_LOG_PGSZ_1M 0x8UL + #define SQ_FR_PMR_PBL_PAGE_SIZE_LOG_PGSZ_2M 0x9UL + #define SQ_FR_PMR_PBL_PAGE_SIZE_LOG_PGSZ_4M 0xaUL + #define SQ_FR_PMR_PBL_PAGE_SIZE_LOG_PGSZ_1G 0x12UL + #define SQ_FR_PMR_RESERVED1 0x20UL + #define SQ_FR_PMR_NUMLEVELS_MASK 0xc0UL + #define SQ_FR_PMR_NUMLEVELS_SFT 6 + #define SQ_FR_PMR_NUMLEVELS_PHYSICAL (0x0UL << 6) + #define SQ_FR_PMR_NUMLEVELS_LAYER1 (0x1UL << 6) + #define SQ_FR_PMR_NUMLEVELS_LAYER2 (0x2UL << 6) + __le64 pblptr; + __le64 va; + __le32 data[24]; +}; + +/* Bind SQ WQE (40 bytes) */ +struct sq_bind { + u8 wqe_type; + #define SQ_BIND_WQE_TYPE_BIND 0xeUL + u8 flags; + #define SQ_BIND_FLAGS_SIGNAL_COMP 0x1UL + #define SQ_BIND_FLAGS_RD_OR_ATOMIC_FENCE 0x2UL + #define SQ_BIND_FLAGS_UC_FENCE 0x4UL + #define SQ_BIND_FLAGS_SE 0x8UL + #define SQ_BIND_FLAGS_INLINE 0x10UL + u8 access_cntl; + #define SQ_BIND_ACCESS_CNTL_LOCAL_WRITE 0x1UL + #define SQ_BIND_ACCESS_CNTL_REMOTE_READ 0x2UL + #define SQ_BIND_ACCESS_CNTL_REMOTE_WRITE 0x4UL + #define SQ_BIND_ACCESS_CNTL_REMOTE_ATOMIC 0x8UL + #define SQ_BIND_ACCESS_CNTL_WINDOW_BIND 0x10UL + u8 reserved8_1; + u8 mw_type_zero_based; + #define SQ_BIND_ZERO_BASED 0x1UL + #define SQ_BIND_MW_TYPE 0x2UL + #define SQ_BIND_MW_TYPE_TYPE1 (0x0UL << 1) + #define SQ_BIND_MW_TYPE_TYPE2 (0x1UL << 1) + #define SQ_BIND_RESERVED6_MASK 0xfcUL + #define SQ_BIND_RESERVED6_SFT 2 + u8 reserved8_2; + __le16 reserved16; + __le32 parent_l_key; + __le32 l_key; + __le64 va; + u8 length[5]; + u8 data_reserved24[99]; + #define SQ_BIND_RESERVED24_MASK 0xffffff00UL + #define SQ_BIND_RESERVED24_SFT 8 + #define SQ_BIND_DATA_MASK 0xffffffffUL + #define SQ_BIND_DATA_SFT 0 +}; + +/* RQ/SRQ WQE Structures */ +/* RQ/SRQ WQE (40 bytes) */ +struct rq_wqe { + u8 wqe_type; + #define RQ_WQE_WQE_TYPE_RCV 0x80UL + u8 flags; + u8 wqe_size; + u8 reserved8; + __le32 reserved32; + __le32 wr_id[2]; + #define RQ_WQE_WR_ID_MASK 0xfffffUL + #define RQ_WQE_WR_ID_SFT 0 + #define RQ_WQE_RESERVED44_MASK 0xfff00000UL + #define RQ_WQE_RESERVED44_SFT 20 + __le32 reserved128[4]; + __le32 data[24]; +}; + +/* CQ CQE Structures */ +/* Base CQE (32 bytes) */ +struct cq_base { + __le64 reserved64_1; + __le64 reserved64_2; + __le64 reserved64_3; + u8 cqe_type_toggle; + #define CQ_BASE_TOGGLE 0x1UL + #define CQ_BASE_CQE_TYPE_MASK 0x1eUL + #define CQ_BASE_CQE_TYPE_SFT 1 + #define CQ_BASE_CQE_TYPE_REQ (0x0UL << 1) + #define CQ_BASE_CQE_TYPE_RES_RC (0x1UL << 1) + #define CQ_BASE_CQE_TYPE_RES_UD (0x2UL << 1) + #define CQ_BASE_CQE_TYPE_RES_RAWETH_QP1 (0x3UL << 1) + #define CQ_BASE_CQE_TYPE_TERMINAL (0xeUL << 1) + #define CQ_BASE_CQE_TYPE_CUT_OFF (0xfUL << 1) + #define CQ_BASE_RESERVED3_MASK 0xe0UL + #define CQ_BASE_RESERVED3_SFT 5 + u8 status; + __le16 reserved16; + __le32 reserved32; +}; + +/* Requester CQ CQE (32 bytes) */ +struct cq_req { + __le64 qp_handle; + __le16 sq_cons_idx; + __le16 reserved16_1; + __le32 reserved32_2; + __le64 reserved64; + u8 cqe_type_toggle; + #define CQ_REQ_TOGGLE 0x1UL + #define CQ_REQ_CQE_TYPE_MASK 0x1eUL + #define CQ_REQ_CQE_TYPE_SFT 1 + #define CQ_REQ_CQE_TYPE_REQ (0x0UL << 1) + #define CQ_REQ_RESERVED3_MASK 0xe0UL + #define CQ_REQ_RESERVED3_SFT 5 + u8 status; + #define CQ_REQ_STATUS_OK 0x0UL + #define CQ_REQ_STATUS_BAD_RESPONSE_ERR 0x1UL + #define CQ_REQ_STATUS_LOCAL_LENGTH_ERR 0x2UL + #define CQ_REQ_STATUS_LOCAL_QP_OPERATION_ERR 0x3UL + #define CQ_REQ_STATUS_LOCAL_PROTECTION_ERR 0x4UL + #define CQ_REQ_STATUS_MEMORY_MGT_OPERATION_ERR 0x5UL + #define CQ_REQ_STATUS_REMOTE_INVALID_REQUEST_ERR 0x6UL + #define CQ_REQ_STATUS_REMOTE_ACCESS_ERR 0x7UL + #define CQ_REQ_STATUS_REMOTE_OPERATION_ERR 0x8UL + #define CQ_REQ_STATUS_RNR_NAK_RETRY_CNT_ERR 0x9UL + #define CQ_REQ_STATUS_TRANSPORT_RETRY_CNT_ERR 0xaUL + #define CQ_REQ_STATUS_WORK_REQUEST_FLUSHED_ERR 0xbUL + __le16 reserved16_2; + __le32 reserved32_1; +}; + +/* Responder RC CQE (32 bytes) */ +struct cq_res_rc { + __le32 length; + __le32 imm_data_or_inv_r_key; + __le64 qp_handle; + __le64 mr_handle; + u8 cqe_type_toggle; + #define CQ_RES_RC_TOGGLE 0x1UL + #define CQ_RES_RC_CQE_TYPE_MASK 0x1eUL + #define CQ_RES_RC_CQE_TYPE_SFT 1 + #define CQ_RES_RC_CQE_TYPE_RES_RC (0x1UL << 1) + #define CQ_RES_RC_RESERVED3_MASK 0xe0UL + #define CQ_RES_RC_RESERVED3_SFT 5 + u8 status; + #define CQ_RES_RC_STATUS_OK 0x0UL + #define CQ_RES_RC_STATUS_LOCAL_ACCESS_ERROR 0x1UL + #define CQ_RES_RC_STATUS_LOCAL_LENGTH_ERR 0x2UL + #define CQ_RES_RC_STATUS_LOCAL_PROTECTION_ERR 0x3UL + #define CQ_RES_RC_STATUS_LOCAL_QP_OPERATION_ERR 0x4UL + #define CQ_RES_RC_STATUS_MEMORY_MGT_OPERATION_ERR 0x5UL + #define CQ_RES_RC_STATUS_REMOTE_INVALID_REQUEST_ERR 0x6UL + #define CQ_RES_RC_STATUS_WORK_REQUEST_FLUSHED_ERR 0x7UL + #define CQ_RES_RC_STATUS_HW_FLUSH_ERR 0x8UL + __le16 flags; + #define CQ_RES_RC_FLAGS_SRQ 0x1UL + #define CQ_RES_RC_FLAGS_SRQ_RQ (0x0UL << 0) + #define CQ_RES_RC_FLAGS_SRQ_SRQ (0x1UL << 0) + #define CQ_RES_RC_FLAGS_SRQ_LAST CQ_RES_RC_FLAGS_SRQ_SRQ + #define CQ_RES_RC_FLAGS_IMM 0x2UL + #define CQ_RES_RC_FLAGS_INV 0x4UL + #define CQ_RES_RC_FLAGS_RDMA 0x8UL + #define CQ_RES_RC_FLAGS_RDMA_SEND (0x0UL << 3) + #define CQ_RES_RC_FLAGS_RDMA_RDMA_WRITE (0x1UL << 3) + #define CQ_RES_RC_FLAGS_RDMA_LAST CQ_RES_RC_FLAGS_RDMA_RDMA_WRITE + __le32 srq_or_rq_wr_id; + #define CQ_RES_RC_SRQ_OR_RQ_WR_ID_MASK 0xfffffUL + #define CQ_RES_RC_SRQ_OR_RQ_WR_ID_SFT 0 + #define CQ_RES_RC_RESERVED12_MASK 0xfff00000UL + #define CQ_RES_RC_RESERVED12_SFT 20 +}; + +/* Responder UD CQE (32 bytes) */ +struct cq_res_ud { + __le32 length; + #define CQ_RES_UD_LENGTH_MASK 0x3fffUL + #define CQ_RES_UD_LENGTH_SFT 0 + #define CQ_RES_UD_RESERVED18_MASK 0xffffc000UL + #define CQ_RES_UD_RESERVED18_SFT 14 + __le32 imm_data; + __le64 qp_handle; + __le16 src_mac[3]; + __le16 src_qp_low; + u8 cqe_type_toggle; + #define CQ_RES_UD_TOGGLE 0x1UL + #define CQ_RES_UD_CQE_TYPE_MASK 0x1eUL + #define CQ_RES_UD_CQE_TYPE_SFT 1 + #define CQ_RES_UD_CQE_TYPE_RES_UD (0x2UL << 1) + #define CQ_RES_UD_RESERVED3_MASK 0xe0UL + #define CQ_RES_UD_RESERVED3_SFT 5 + u8 status; + #define CQ_RES_UD_STATUS_OK 0x0UL + #define CQ_RES_UD_STATUS_LOCAL_ACCESS_ERROR 0x1UL + #define CQ_RES_UD_STATUS_HW_LOCAL_LENGTH_ERR 0x2UL + #define CQ_RES_UD_STATUS_LOCAL_PROTECTION_ERR 0x3UL + #define CQ_RES_UD_STATUS_LOCAL_QP_OPERATION_ERR 0x4UL + #define CQ_RES_UD_STATUS_MEMORY_MGT_OPERATION_ERR 0x5UL + #define CQ_RES_UD_STATUS_WORK_REQUEST_FLUSHED_ERR 0x7UL + #define CQ_RES_UD_STATUS_HW_FLUSH_ERR 0x8UL + __le16 flags; + #define CQ_RES_UD_FLAGS_SRQ 0x1UL + #define CQ_RES_UD_FLAGS_SRQ_RQ (0x0UL << 0) + #define CQ_RES_UD_FLAGS_SRQ_SRQ (0x1UL << 0) + #define CQ_RES_UD_FLAGS_SRQ_LAST CQ_RES_UD_FLAGS_SRQ_SRQ + #define CQ_RES_UD_FLAGS_IMM 0x2UL + #define CQ_RES_UD_FLAGS_ROCE_IP_VER_MASK 0xcUL + #define CQ_RES_UD_FLAGS_ROCE_IP_VER_SFT 2 + #define CQ_RES_UD_FLAGS_ROCE_IP_VER_V1 (0x0UL << 2) + #define CQ_RES_UD_FLAGS_ROCE_IP_VER_V2IPV4 (0x2UL << 2) + #define CQ_RES_UD_FLAGS_ROCE_IP_VER_V2IPV6 (0x3UL << 2) + #define CQ_RES_UD_FLAGS_ROCE_IP_VER_LAST \ + CQ_RES_UD_FLAGS_ROCE_IP_VER_V2IPV6 + __le32 src_qp_high_srq_or_rq_wr_id; + #define CQ_RES_UD_SRQ_OR_RQ_WR_ID_MASK 0xfffffUL + #define CQ_RES_UD_SRQ_OR_RQ_WR_ID_SFT 0 + #define CQ_RES_UD_RESERVED4_MASK 0xf00000UL + #define CQ_RES_UD_RESERVED4_SFT 20 + #define CQ_RES_UD_SRC_QP_HIGH_MASK 0xff000000UL + #define CQ_RES_UD_SRC_QP_HIGH_SFT 24 +}; + +/* Responder RawEth and QP1 CQE (32 bytes) */ +struct cq_res_raweth_qp1 { + __le16 length; + #define CQ_RES_RAWETH_QP1_LENGTH_MASK 0x3fffUL + #define CQ_RES_RAWETH_QP1_LENGTH_SFT 0 + #define CQ_RES_RAWETH_QP1_RESERVED2_MASK 0xc000UL + #define CQ_RES_RAWETH_QP1_RESERVED2_SFT 14 + __le16 raweth_qp1_flags; + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_ERROR 0x1UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_RESERVED5_1_MASK 0x3eUL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_RESERVED5_1_SFT 1 + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_ITYPE_MASK 0x3c0UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_ITYPE_SFT 6 + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_ITYPE_NOT_KNOWN (0x0UL << 6) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_ITYPE_IP (0x1UL << 6) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_ITYPE_TCP (0x2UL << 6) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_ITYPE_UDP (0x3UL << 6) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_ITYPE_FCOE (0x4UL << 6) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_ITYPE_ROCE (0x5UL << 6) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_ITYPE_ICMP (0x7UL << 6) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_ITYPE_PTP_WO_TIMESTAMP \ + (0x8UL << 6) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_ITYPE_PTP_W_TIMESTAMP \ + (0x9UL << 6) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_ITYPE_LAST \ + CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_ITYPE_PTP_W_TIMESTAMP + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_MASK 0x3ffUL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS_SFT 0 + #define CQ_RES_RAWETH_QP1_RESERVED6_MASK 0xfc00UL + #define CQ_RES_RAWETH_QP1_RESERVED6_SFT 10 + __le16 raweth_qp1_errors; + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_RESERVED4_MASK 0xfUL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_RESERVED4_SFT 0 + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_IP_CS_ERROR 0x10UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_L4_CS_ERROR 0x20UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_T_IP_CS_ERROR 0x40UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_T_L4_CS_ERROR 0x80UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_CRC_ERROR 0x100UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_T_PKT_ERROR_MASK 0xe00UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_T_PKT_ERROR_SFT 9 + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_T_PKT_ERROR_NO_ERROR \ + (0x0UL << 9) + #define \ + CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_T_PKT_ERROR_T_L3_BAD_VERSION \ + (0x1UL << 9) + #define \ + CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_T_PKT_ERROR_T_L3_BAD_HDR_LEN \ + (0x2UL << 9) + #define \ + CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_T_PKT_ERROR_TUNNEL_TOTAL_ERROR \ + (0x3UL << 9) + #define \ + CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_T_PKT_ERROR_T_IP_TOTAL_ERROR \ + (0x4UL << 9) + #define \ + CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_T_PKT_ERROR_T_UDP_TOTAL_ERROR \ + (0x5UL << 9) + #define \ + CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_T_PKT_ERROR_T_L3_BAD_TTL \ + (0x6UL << 9) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_T_PKT_ERROR_LAST \ + CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_T_PKT_ERROR_T_L3_BAD_TTL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_PKT_ERROR_MASK 0xf000UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_PKT_ERROR_SFT 12 + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_PKT_ERROR_NO_ERROR \ + (0x0UL << 12) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_PKT_ERROR_L3_BAD_VERSION \ + (0x1UL << 12) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_PKT_ERROR_L3_BAD_HDR_LEN \ + (0x2UL << 12) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_PKT_ERROR_L3_BAD_TTL \ + (0x3UL << 12) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_PKT_ERROR_IP_TOTAL_ERROR \ + (0x4UL << 12) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_PKT_ERROR_UDP_TOTAL_ERROR \ + (0x5UL << 12) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_PKT_ERROR_L4_BAD_HDR_LEN \ + (0x6UL << 12) + #define \ + CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_PKT_ERROR_L4_BAD_HDR_LEN_TOO_SMALL\ + (0x7UL << 12) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_PKT_ERROR_L4_BAD_OPT_LEN \ + (0x8UL << 12) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_PKT_ERROR_LAST \ + CQ_RES_RAWETH_QP1_RAWETH_QP1_ERRORS_PKT_ERROR_L4_BAD_OPT_LEN + __le16 raweth_qp1_cfa_code; + __le64 qp_handle; + __le32 raweth_qp1_flags2; + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS2_IP_CS_CALC 0x1UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS2_L4_CS_CALC 0x2UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS2_T_IP_CS_CALC 0x4UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS2_T_L4_CS_CALC 0x8UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS2_META_FORMAT_MASK 0xf0UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS2_META_FORMAT_SFT 4 + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS2_META_FORMAT_NONE \ + (0x0UL << 4) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS2_META_FORMAT_VLAN \ + (0x1UL << 4) + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS2_META_FORMAT_LAST\ + CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS2_META_FORMAT_VLAN + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_FLAGS2_IP_TYPE 0x100UL + __le32 raweth_qp1_metadata; + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_METADATA_VID_MASK 0xfffUL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_METADATA_VID_SFT 0 + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_METADATA_DE 0x1000UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_METADATA_PRI_MASK 0xe000UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_METADATA_PRI_SFT 13 + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_METADATA_TPID_MASK 0xffff0000UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_METADATA_TPID_SFT 16 + u8 cqe_type_toggle; + #define CQ_RES_RAWETH_QP1_TOGGLE 0x1UL + #define CQ_RES_RAWETH_QP1_CQE_TYPE_MASK 0x1eUL + #define CQ_RES_RAWETH_QP1_CQE_TYPE_SFT 1 + #define CQ_RES_RAWETH_QP1_CQE_TYPE_RES_RAWETH_QP1 (0x3UL << 1) + #define CQ_RES_RAWETH_QP1_RESERVED3_MASK 0xe0UL + #define CQ_RES_RAWETH_QP1_RESERVED3_SFT 5 + u8 status; + #define CQ_RES_RAWETH_QP1_STATUS_OK 0x0UL + #define CQ_RES_RAWETH_QP1_STATUS_LOCAL_ACCESS_ERROR 0x1UL + #define CQ_RES_RAWETH_QP1_STATUS_HW_LOCAL_LENGTH_ERR 0x2UL + #define CQ_RES_RAWETH_QP1_STATUS_LOCAL_PROTECTION_ERR 0x3UL + #define CQ_RES_RAWETH_QP1_STATUS_LOCAL_QP_OPERATION_ERR 0x4UL + #define CQ_RES_RAWETH_QP1_STATUS_MEMORY_MGT_OPERATION_ERR 0x5UL + #define CQ_RES_RAWETH_QP1_STATUS_WORK_REQUEST_FLUSHED_ERR 0x7UL + #define CQ_RES_RAWETH_QP1_STATUS_HW_FLUSH_ERR 0x8UL + __le16 flags; + #define CQ_RES_RAWETH_QP1_FLAGS_SRQ 0x1UL + #define CQ_RES_RAWETH_QP1_FLAGS_SRQ_RQ 0x0UL + #define CQ_RES_RAWETH_QP1_FLAGS_SRQ_SRQ 0x1UL + #define CQ_RES_RAWETH_QP1_FLAGS_SRQ_LAST \ + CQ_RES_RAWETH_QP1_FLAGS_SRQ_SRQ + __le32 raweth_qp1_payload_offset_srq_or_rq_wr_id; + #define CQ_RES_RAWETH_QP1_SRQ_OR_RQ_WR_ID_MASK 0xfffffUL + #define CQ_RES_RAWETH_QP1_SRQ_OR_RQ_WR_ID_SFT 0 + #define CQ_RES_RAWETH_QP1_RESERVED4_MASK 0xf00000UL + #define CQ_RES_RAWETH_QP1_RESERVED4_SFT 20 + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_PAYLOAD_OFFSET_MASK 0xff000000UL + #define CQ_RES_RAWETH_QP1_RAWETH_QP1_PAYLOAD_OFFSET_SFT 24 +}; + +/* Terminal CQE (32 bytes) */ +struct cq_terminal { + __le64 qp_handle; + __le16 sq_cons_idx; + __le16 rq_cons_idx; + __le32 reserved32_1; + __le64 reserved64_3; + u8 cqe_type_toggle; + #define CQ_TERMINAL_TOGGLE 0x1UL + #define CQ_TERMINAL_CQE_TYPE_MASK 0x1eUL + #define CQ_TERMINAL_CQE_TYPE_SFT 1 + #define CQ_TERMINAL_CQE_TYPE_TERMINAL (0xeUL << 1) + #define CQ_TERMINAL_RESERVED3_MASK 0xe0UL + #define CQ_TERMINAL_RESERVED3_SFT 5 + u8 status; + #define CQ_TERMINAL_STATUS_OK 0x0UL + __le16 reserved16; + __le32 reserved32_2; +}; + +/* Cutoff CQE (32 bytes) */ +struct cq_cutoff { + __le64 reserved64_1; + __le64 reserved64_2; + __le64 reserved64_3; + u8 cqe_type_toggle; + #define CQ_CUTOFF_TOGGLE 0x1UL + #define CQ_CUTOFF_CQE_TYPE_MASK 0x1eUL + #define CQ_CUTOFF_CQE_TYPE_SFT 1 + #define CQ_CUTOFF_CQE_TYPE_CUT_OFF (0xfUL << 1) + #define CQ_CUTOFF_RESERVED3_MASK 0xe0UL + #define CQ_CUTOFF_RESERVED3_SFT 5 + u8 status; + #define CQ_CUTOFF_STATUS_OK 0x0UL + __le16 reserved16; + __le32 reserved32; +}; + +/* Notification Queue (NQ) Structures */ +/* Base NQ Record (16 bytes) */ +struct nq_base { + __le16 info10_type; + #define NQ_BASE_TYPE_MASK 0x3fUL + #define NQ_BASE_TYPE_SFT 0 + #define NQ_BASE_TYPE_CQ_NOTIFICATION 0x30UL + #define NQ_BASE_TYPE_SRQ_EVENT 0x32UL + #define NQ_BASE_TYPE_DBQ_EVENT 0x34UL + #define NQ_BASE_TYPE_QP_EVENT 0x38UL + #define NQ_BASE_TYPE_FUNC_EVENT 0x3aUL + #define NQ_BASE_INFO10_MASK 0xffc0UL + #define NQ_BASE_INFO10_SFT 6 + __le16 info16; + __le32 info32; + __le32 info63_v[2]; + #define NQ_BASE_V 0x1UL + #define NQ_BASE_INFO63_MASK 0xfffffffeUL + #define NQ_BASE_INFO63_SFT 1 +}; + +/* Completion Queue Notification (16 bytes) */ +struct nq_cn { + __le16 type; + #define NQ_CN_TYPE_MASK 0x3fUL + #define NQ_CN_TYPE_SFT 0 + #define NQ_CN_TYPE_CQ_NOTIFICATION 0x30UL + #define NQ_CN_RESERVED9_MASK 0xffc0UL + #define NQ_CN_RESERVED9_SFT 6 + __le16 reserved16; + __le32 cq_handle_low; + __le32 v; + #define NQ_CN_V 0x1UL + #define NQ_CN_RESERVED31_MASK 0xfffffffeUL + #define NQ_CN_RESERVED31_SFT 1 + __le32 cq_handle_high; +}; + +/* SRQ Event Notification (16 bytes) */ +struct nq_srq_event { + u8 type; + #define NQ_SRQ_EVENT_TYPE_MASK 0x3fUL + #define NQ_SRQ_EVENT_TYPE_SFT 0 + #define NQ_SRQ_EVENT_TYPE_SRQ_EVENT 0x32UL + #define NQ_SRQ_EVENT_RESERVED1_MASK 0xc0UL + #define NQ_SRQ_EVENT_RESERVED1_SFT 6 + u8 event; + #define NQ_SRQ_EVENT_EVENT_SRQ_THRESHOLD_EVENT 0x1UL + __le16 reserved16; + __le32 srq_handle_low; + __le32 v; + #define NQ_SRQ_EVENT_V 0x1UL + #define NQ_SRQ_EVENT_RESERVED31_MASK 0xfffffffeUL + #define NQ_SRQ_EVENT_RESERVED31_SFT 1 + __le32 srq_handle_high; +}; + +/* DBQ Async Event Notification (16 bytes) */ +struct nq_dbq_event { + u8 type; + #define NQ_DBQ_EVENT_TYPE_MASK 0x3fUL + #define NQ_DBQ_EVENT_TYPE_SFT 0 + #define NQ_DBQ_EVENT_TYPE_DBQ_EVENT 0x34UL + #define NQ_DBQ_EVENT_RESERVED1_MASK 0xc0UL + #define NQ_DBQ_EVENT_RESERVED1_SFT 6 + u8 event; + #define NQ_DBQ_EVENT_EVENT_DBQ_THRESHOLD_EVENT 0x1UL + __le16 db_pfid; + #define NQ_DBQ_EVENT_DB_PFID_MASK 0xfUL + #define NQ_DBQ_EVENT_DB_PFID_SFT 0 + #define NQ_DBQ_EVENT_RESERVED12_MASK 0xfff0UL + #define NQ_DBQ_EVENT_RESERVED12_SFT 4 + __le32 db_dpi; + #define NQ_DBQ_EVENT_DB_DPI_MASK 0xfffffUL + #define NQ_DBQ_EVENT_DB_DPI_SFT 0 + #define NQ_DBQ_EVENT_RESERVED12_2_MASK 0xfff00000UL + #define NQ_DBQ_EVENT_RESERVED12_2_SFT 20 + __le32 v; + #define NQ_DBQ_EVENT_V 0x1UL + #define NQ_DBQ_EVENT_RESERVED32_MASK 0xfffffffeUL + #define NQ_DBQ_EVENT_RESERVED32_SFT 1 + __le32 db_type_db_xid; + #define NQ_DBQ_EVENT_DB_XID_MASK 0xfffffUL + #define NQ_DBQ_EVENT_DB_XID_SFT 0 + #define NQ_DBQ_EVENT_RESERVED8_MASK 0xff00000UL + #define NQ_DBQ_EVENT_RESERVED8_SFT 20 + #define NQ_DBQ_EVENT_DB_TYPE_MASK 0xf0000000UL + #define NQ_DBQ_EVENT_DB_TYPE_SFT 28 +}; + +/* Read Request/Response Queue Structures */ +/* Input Read Request Queue (IRRQ) Message (32 bytes) */ +struct xrrq_irrq { + __le16 credits_type; + #define XRRQ_IRRQ_TYPE 0x1UL + #define XRRQ_IRRQ_TYPE_READ_REQ 0x0UL + #define XRRQ_IRRQ_TYPE_ATOMIC_REQ 0x1UL + #define XRRQ_IRRQ_RESERVED10_MASK 0x7feUL + #define XRRQ_IRRQ_RESERVED10_SFT 1 + #define XRRQ_IRRQ_CREDITS_MASK 0xf800UL + #define XRRQ_IRRQ_CREDITS_SFT 11 + __le16 reserved16; + __le32 reserved32; + __le32 psn; + #define XRRQ_IRRQ_PSN_MASK 0xffffffUL + #define XRRQ_IRRQ_PSN_SFT 0 + #define XRRQ_IRRQ_RESERVED8_1_MASK 0xff000000UL + #define XRRQ_IRRQ_RESERVED8_1_SFT 24 + __le32 msn; + #define XRRQ_IRRQ_MSN_MASK 0xffffffUL + #define XRRQ_IRRQ_MSN_SFT 0 + #define XRRQ_IRRQ_RESERVED8_2_MASK 0xff000000UL + #define XRRQ_IRRQ_RESERVED8_2_SFT 24 + __le64 va_or_atomic_result; + __le32 rdma_r_key; + __le32 length; +}; + +/* Output Read Request Queue (ORRQ) Message (32 bytes) */ +struct xrrq_orrq { + __le16 num_sges_type; + #define XRRQ_ORRQ_TYPE 0x1UL + #define XRRQ_ORRQ_TYPE_READ_REQ 0x0UL + #define XRRQ_ORRQ_TYPE_ATOMIC_REQ 0x1UL + #define XRRQ_ORRQ_RESERVED10_MASK 0x7feUL + #define XRRQ_ORRQ_RESERVED10_SFT 1 + #define XRRQ_ORRQ_NUM_SGES_MASK 0xf800UL + #define XRRQ_ORRQ_NUM_SGES_SFT 11 + __le16 reserved16; + __le32 length; + __le32 psn; + #define XRRQ_ORRQ_PSN_MASK 0xffffffUL + #define XRRQ_ORRQ_PSN_SFT 0 + #define XRRQ_ORRQ_RESERVED8_1_MASK 0xff000000UL + #define XRRQ_ORRQ_RESERVED8_1_SFT 24 + __le32 end_psn; + #define XRRQ_ORRQ_END_PSN_MASK 0xffffffUL + #define XRRQ_ORRQ_END_PSN_SFT 0 + #define XRRQ_ORRQ_RESERVED8_2_MASK 0xff000000UL + #define XRRQ_ORRQ_RESERVED8_2_SFT 24 + __le64 first_sge_phy_or_sing_sge_va; + __le32 single_sge_l_key; + __le32 single_sge_size; +}; + +/* Page Buffer List Memory Structures (PBL) */ +/* Page Table Entry (PTE) (8 bytes) */ +struct ptu_pte { + __le32 page_next_to_last_last_valid[2]; + #define PTU_PTE_VALID 0x1UL + #define PTU_PTE_LAST 0x2UL + #define PTU_PTE_NEXT_TO_LAST 0x4UL + #define PTU_PTE_PAGE_MASK 0xfffff000UL + #define PTU_PTE_PAGE_SFT 12 +}; + +/* Page Directory Entry (PDE) (8 bytes) */ +struct ptu_pde { + __le32 page_valid[2]; + #define PTU_PDE_VALID 0x1UL + #define PTU_PDE_PAGE_MASK 0xfffff000UL + #define PTU_PDE_PAGE_SFT 12 +}; + +/* RoCE Fastpath Host Structures */ +/* Command Queue (CMDQ) Interface */ +/* Init CMDQ (16 bytes) */ +struct cmdq_init { + __le64 cmdq_pbl; + __le16 cmdq_size_cmdq_lvl; + #define CMDQ_INIT_CMDQ_LVL_MASK 0x3UL + #define CMDQ_INIT_CMDQ_LVL_SFT 0 + #define CMDQ_INIT_CMDQ_SIZE_MASK 0xfffcUL + #define CMDQ_INIT_CMDQ_SIZE_SFT 2 + __le16 creq_ring_id; + __le32 prod_idx; +}; + +/* Update CMDQ producer index (16 bytes) */ +struct cmdq_update { + __le64 reserved64; + __le32 reserved32; + __le32 prod_idx; +}; + +/* CMDQ common header structure (16 bytes) */ +struct cmdq_base { + u8 opcode; + #define CMDQ_BASE_OPCODE_CREATE_QP 0x1UL + #define CMDQ_BASE_OPCODE_DESTROY_QP 0x2UL + #define CMDQ_BASE_OPCODE_MODIFY_QP 0x3UL + #define CMDQ_BASE_OPCODE_QUERY_QP 0x4UL + #define CMDQ_BASE_OPCODE_CREATE_SRQ 0x5UL + #define CMDQ_BASE_OPCODE_DESTROY_SRQ 0x6UL + #define CMDQ_BASE_OPCODE_QUERY_SRQ 0x8UL + #define CMDQ_BASE_OPCODE_CREATE_CQ 0x9UL + #define CMDQ_BASE_OPCODE_DESTROY_CQ 0xaUL + #define CMDQ_BASE_OPCODE_RESIZE_CQ 0xcUL + #define CMDQ_BASE_OPCODE_ALLOCATE_MRW 0xdUL + #define CMDQ_BASE_OPCODE_DEALLOCATE_KEY 0xeUL + #define CMDQ_BASE_OPCODE_REGISTER_MR 0xfUL + #define CMDQ_BASE_OPCODE_DEREGISTER_MR 0x10UL + #define CMDQ_BASE_OPCODE_ADD_GID 0x11UL + #define CMDQ_BASE_OPCODE_DELETE_GID 0x12UL + #define CMDQ_BASE_OPCODE_MODIFY_GID 0x17UL + #define CMDQ_BASE_OPCODE_QUERY_GID 0x18UL + #define CMDQ_BASE_OPCODE_CREATE_QP1 0x13UL + #define CMDQ_BASE_OPCODE_DESTROY_QP1 0x14UL + #define CMDQ_BASE_OPCODE_CREATE_AH 0x15UL + #define CMDQ_BASE_OPCODE_DESTROY_AH 0x16UL + #define CMDQ_BASE_OPCODE_INITIALIZE_FW 0x80UL + #define CMDQ_BASE_OPCODE_DEINITIALIZE_FW 0x81UL + #define CMDQ_BASE_OPCODE_STOP_FUNC 0x82UL + #define CMDQ_BASE_OPCODE_QUERY_FUNC 0x83UL + #define CMDQ_BASE_OPCODE_SET_FUNC_RESOURCES 0x84UL + #define CMDQ_BASE_OPCODE_READ_CONTEXT 0x85UL + #define CMDQ_BASE_OPCODE_VF_BACKCHANNEL_REQUEST 0x86UL + #define CMDQ_BASE_OPCODE_READ_VF_MEMORY 0x87UL + #define CMDQ_BASE_OPCODE_COMPLETE_VF_REQUEST 0x88UL + #define CMDQ_BASE_OPCODE_EXTEND_CONTEXT_ARRRAY 0x89UL + #define CMDQ_BASE_OPCODE_MAP_TC_TO_COS 0x8aUL + #define CMDQ_BASE_OPCODE_QUERY_VERSION 0x8bUL + #define CMDQ_BASE_OPCODE_MODIFY_CC 0x8cUL + #define CMDQ_BASE_OPCODE_QUERY_CC 0x8dUL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; +}; + +/* Create QP command (96 bytes) */ +struct cmdq_create_qp { + u8 opcode; + #define CMDQ_CREATE_QP_OPCODE_CREATE_QP 0x1UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le64 qp_handle; + __le32 qp_flags; + #define CMDQ_CREATE_QP_QP_FLAGS_SRQ_USED 0x1UL + #define CMDQ_CREATE_QP_QP_FLAGS_FORCE_COMPLETION 0x2UL + #define CMDQ_CREATE_QP_QP_FLAGS_RESERVED_LKEY_ENABLE 0x4UL + #define CMDQ_CREATE_QP_QP_FLAGS_FR_PMR_ENABLED 0x8UL + u8 type; + #define CMDQ_CREATE_QP_TYPE_RC 0x2UL + #define CMDQ_CREATE_QP_TYPE_UD 0x4UL + #define CMDQ_CREATE_QP_TYPE_RAW_ETHERTYPE 0x6UL + u8 sq_pg_size_sq_lvl; + #define CMDQ_CREATE_QP_SQ_LVL_MASK 0xfUL + #define CMDQ_CREATE_QP_SQ_LVL_SFT 0 + #define CMDQ_CREATE_QP_SQ_LVL_LVL_0 0x0UL + #define CMDQ_CREATE_QP_SQ_LVL_LVL_1 0x1UL + #define CMDQ_CREATE_QP_SQ_LVL_LVL_2 0x2UL + #define CMDQ_CREATE_QP_SQ_PG_SIZE_MASK 0xf0UL + #define CMDQ_CREATE_QP_SQ_PG_SIZE_SFT 4 + #define CMDQ_CREATE_QP_SQ_PG_SIZE_PG_4K (0x0UL << 4) + #define CMDQ_CREATE_QP_SQ_PG_SIZE_PG_8K (0x1UL << 4) + #define CMDQ_CREATE_QP_SQ_PG_SIZE_PG_64K (0x2UL << 4) + #define CMDQ_CREATE_QP_SQ_PG_SIZE_PG_2M (0x3UL << 4) + #define CMDQ_CREATE_QP_SQ_PG_SIZE_PG_8M (0x4UL << 4) + #define CMDQ_CREATE_QP_SQ_PG_SIZE_PG_1G (0x5UL << 4) + u8 rq_pg_size_rq_lvl; + #define CMDQ_CREATE_QP_RQ_LVL_MASK 0xfUL + #define CMDQ_CREATE_QP_RQ_LVL_SFT 0 + #define CMDQ_CREATE_QP_RQ_LVL_LVL_0 0x0UL + #define CMDQ_CREATE_QP_RQ_LVL_LVL_1 0x1UL + #define CMDQ_CREATE_QP_RQ_LVL_LVL_2 0x2UL + #define CMDQ_CREATE_QP_RQ_PG_SIZE_MASK 0xf0UL + #define CMDQ_CREATE_QP_RQ_PG_SIZE_SFT 4 + #define CMDQ_CREATE_QP_RQ_PG_SIZE_PG_4K (0x0UL << 4) + #define CMDQ_CREATE_QP_RQ_PG_SIZE_PG_8K (0x1UL << 4) + #define CMDQ_CREATE_QP_RQ_PG_SIZE_PG_64K (0x2UL << 4) + #define CMDQ_CREATE_QP_RQ_PG_SIZE_PG_2M (0x3UL << 4) + #define CMDQ_CREATE_QP_RQ_PG_SIZE_PG_8M (0x4UL << 4) + #define CMDQ_CREATE_QP_RQ_PG_SIZE_PG_1G (0x5UL << 4) + u8 unused_0; + __le32 dpi; + __le32 sq_size; + __le32 rq_size; + __le16 sq_fwo_sq_sge; + #define CMDQ_CREATE_QP_SQ_SGE_MASK 0xfUL + #define CMDQ_CREATE_QP_SQ_SGE_SFT 0 + #define CMDQ_CREATE_QP_SQ_FWO_MASK 0xfff0UL + #define CMDQ_CREATE_QP_SQ_FWO_SFT 4 + __le16 rq_fwo_rq_sge; + #define CMDQ_CREATE_QP_RQ_SGE_MASK 0xfUL + #define CMDQ_CREATE_QP_RQ_SGE_SFT 0 + #define CMDQ_CREATE_QP_RQ_FWO_MASK 0xfff0UL + #define CMDQ_CREATE_QP_RQ_FWO_SFT 4 + __le32 scq_cid; + __le32 rcq_cid; + __le32 srq_cid; + __le32 pd_id; + __le64 sq_pbl; + __le64 rq_pbl; + __le64 irrq_addr; + __le64 orrq_addr; +}; + +/* Destroy QP command (24 bytes) */ +struct cmdq_destroy_qp { + u8 opcode; + #define CMDQ_DESTROY_QP_OPCODE_DESTROY_QP 0x2UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le32 qp_cid; + __le32 unused_0; +}; + +/* Modify QP command (112 bytes) */ +struct cmdq_modify_qp { + u8 opcode; + #define CMDQ_MODIFY_QP_OPCODE_MODIFY_QP 0x3UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le32 modify_mask; + #define CMDQ_MODIFY_QP_MODIFY_MASK_STATE 0x1UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_EN_SQD_ASYNC_NOTIFY 0x2UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_ACCESS 0x4UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_PKEY 0x8UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_QKEY 0x10UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_DGID 0x20UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_FLOW_LABEL 0x40UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_SGID_INDEX 0x80UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_HOP_LIMIT 0x100UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_TRAFFIC_CLASS 0x200UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_DEST_MAC 0x400UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_PATH_MTU 0x1000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_TIMEOUT 0x2000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_RETRY_CNT 0x4000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_RNR_RETRY 0x8000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_RQ_PSN 0x10000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_MAX_RD_ATOMIC 0x20000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_MIN_RNR_TIMER 0x40000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_SQ_PSN 0x80000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_MAX_DEST_RD_ATOMIC 0x100000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_SQ_SIZE 0x200000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_RQ_SIZE 0x400000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_SQ_SGE 0x800000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_RQ_SGE 0x1000000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_MAX_INLINE_DATA 0x2000000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_DEST_QP_ID 0x4000000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_SRC_MAC 0x8000000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_VLAN_ID 0x10000000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_ENABLE_CC 0x20000000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_TOS_ECN 0x40000000UL + #define CMDQ_MODIFY_QP_MODIFY_MASK_TOS_DSCP 0x80000000UL + __le32 qp_cid; + u8 network_type_en_sqd_async_notify_new_state; + #define CMDQ_MODIFY_QP_NEW_STATE_MASK 0xfUL + #define CMDQ_MODIFY_QP_NEW_STATE_SFT 0 + #define CMDQ_MODIFY_QP_NEW_STATE_RESET 0x0UL + #define CMDQ_MODIFY_QP_NEW_STATE_INIT 0x1UL + #define CMDQ_MODIFY_QP_NEW_STATE_RTR 0x2UL + #define CMDQ_MODIFY_QP_NEW_STATE_RTS 0x3UL + #define CMDQ_MODIFY_QP_NEW_STATE_SQD 0x4UL + #define CMDQ_MODIFY_QP_NEW_STATE_SQE 0x5UL + #define CMDQ_MODIFY_QP_NEW_STATE_ERR 0x6UL + #define CMDQ_MODIFY_QP_EN_SQD_ASYNC_NOTIFY 0x10UL + #define CMDQ_MODIFY_QP_NETWORK_TYPE_MASK 0xc0UL + #define CMDQ_MODIFY_QP_NETWORK_TYPE_SFT 6 + #define CMDQ_MODIFY_QP_NETWORK_TYPE_ROCEV1 (0x0UL << 6) + #define CMDQ_MODIFY_QP_NETWORK_TYPE_ROCEV2_IPV4 (0x2UL << 6) + #define CMDQ_MODIFY_QP_NETWORK_TYPE_ROCEV2_IPV6 (0x3UL << 6) + u8 access; + #define CMDQ_MODIFY_QP_ACCESS_LOCAL_WRITE 0x1UL + #define CMDQ_MODIFY_QP_ACCESS_REMOTE_WRITE 0x2UL + #define CMDQ_MODIFY_QP_ACCESS_REMOTE_READ 0x4UL + #define CMDQ_MODIFY_QP_ACCESS_REMOTE_ATOMIC 0x8UL + __le16 pkey; + __le32 qkey; + __le32 dgid[4]; + __le32 flow_label; + __le16 sgid_index; + u8 hop_limit; + u8 traffic_class; + __le16 dest_mac[3]; + u8 tos_dscp_tos_ecn; + #define CMDQ_MODIFY_QP_TOS_ECN_MASK 0x3UL + #define CMDQ_MODIFY_QP_TOS_ECN_SFT 0 + #define CMDQ_MODIFY_QP_TOS_DSCP_MASK 0xfcUL + #define CMDQ_MODIFY_QP_TOS_DSCP_SFT 2 + u8 path_mtu; + #define CMDQ_MODIFY_QP_PATH_MTU_MASK 0xf0UL + #define CMDQ_MODIFY_QP_PATH_MTU_SFT 4 + #define CMDQ_MODIFY_QP_PATH_MTU_MTU_256 (0x0UL << 4) + #define CMDQ_MODIFY_QP_PATH_MTU_MTU_512 (0x1UL << 4) + #define CMDQ_MODIFY_QP_PATH_MTU_MTU_1024 (0x2UL << 4) + #define CMDQ_MODIFY_QP_PATH_MTU_MTU_2048 (0x3UL << 4) + #define CMDQ_MODIFY_QP_PATH_MTU_MTU_4096 (0x4UL << 4) + #define CMDQ_MODIFY_QP_PATH_MTU_MTU_8192 (0x5UL << 4) + u8 timeout; + u8 retry_cnt; + u8 rnr_retry; + u8 min_rnr_timer; + __le32 rq_psn; + __le32 sq_psn; + u8 max_rd_atomic; + u8 max_dest_rd_atomic; + __le16 enable_cc; + #define CMDQ_MODIFY_QP_ENABLE_CC 0x1UL + __le32 sq_size; + __le32 rq_size; + __le16 sq_sge; + __le16 rq_sge; + __le32 max_inline_data; + __le32 dest_qp_id; + __le32 unused_3; + __le16 src_mac[3]; + __le16 vlan_pcp_vlan_dei_vlan_id; + #define CMDQ_MODIFY_QP_VLAN_ID_MASK 0xfffUL + #define CMDQ_MODIFY_QP_VLAN_ID_SFT 0 + #define CMDQ_MODIFY_QP_VLAN_DEI 0x1000UL + #define CMDQ_MODIFY_QP_VLAN_PCP_MASK 0xe000UL + #define CMDQ_MODIFY_QP_VLAN_PCP_SFT 13 +}; + +/* Query QP command (24 bytes) */ +struct cmdq_query_qp { + u8 opcode; + #define CMDQ_QUERY_QP_OPCODE_QUERY_QP 0x4UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le32 qp_cid; + __le32 unused_0; +}; + +/* Create SRQ command (48 bytes) */ +struct cmdq_create_srq { + u8 opcode; + #define CMDQ_CREATE_SRQ_OPCODE_CREATE_SRQ 0x5UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le64 srq_handle; + __le16 pg_size_lvl; + #define CMDQ_CREATE_SRQ_LVL_MASK 0x3UL + #define CMDQ_CREATE_SRQ_LVL_SFT 0 + #define CMDQ_CREATE_SRQ_LVL_LVL_0 0x0UL + #define CMDQ_CREATE_SRQ_LVL_LVL_1 0x1UL + #define CMDQ_CREATE_SRQ_LVL_LVL_2 0x2UL + #define CMDQ_CREATE_SRQ_PG_SIZE_MASK 0x1cUL + #define CMDQ_CREATE_SRQ_PG_SIZE_SFT 2 + #define CMDQ_CREATE_SRQ_PG_SIZE_PG_4K (0x0UL << 2) + #define CMDQ_CREATE_SRQ_PG_SIZE_PG_8K (0x1UL << 2) + #define CMDQ_CREATE_SRQ_PG_SIZE_PG_64K (0x2UL << 2) + #define CMDQ_CREATE_SRQ_PG_SIZE_PG_2M (0x3UL << 2) + #define CMDQ_CREATE_SRQ_PG_SIZE_PG_8M (0x4UL << 2) + #define CMDQ_CREATE_SRQ_PG_SIZE_PG_1G (0x5UL << 2) + __le16 eventq_id; + #define CMDQ_CREATE_SRQ_EVENTQ_ID_MASK 0xfffUL + #define CMDQ_CREATE_SRQ_EVENTQ_ID_SFT 0 + __le16 srq_size; + __le16 srq_fwo; + __le32 dpi; + __le32 pd_id; + __le64 pbl; +}; + +/* Destroy SRQ command (24 bytes) */ +struct cmdq_destroy_srq { + u8 opcode; + #define CMDQ_DESTROY_SRQ_OPCODE_DESTROY_SRQ 0x6UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le32 srq_cid; + __le32 unused_0; +}; + +/* Query SRQ command (24 bytes) */ +struct cmdq_query_srq { + u8 opcode; + #define CMDQ_QUERY_SRQ_OPCODE_QUERY_SRQ 0x8UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le32 srq_cid; + __le32 unused_0; +}; + +/* Create CQ command (48 bytes) */ +struct cmdq_create_cq { + u8 opcode; + #define CMDQ_CREATE_CQ_OPCODE_CREATE_CQ 0x9UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le64 cq_handle; + __le32 pg_size_lvl; + #define CMDQ_CREATE_CQ_LVL_MASK 0x3UL + #define CMDQ_CREATE_CQ_LVL_SFT 0 + #define CMDQ_CREATE_CQ_LVL_LVL_0 0x0UL + #define CMDQ_CREATE_CQ_LVL_LVL_1 0x1UL + #define CMDQ_CREATE_CQ_LVL_LVL_2 0x2UL + #define CMDQ_CREATE_CQ_PG_SIZE_MASK 0x1cUL + #define CMDQ_CREATE_CQ_PG_SIZE_SFT 2 + #define CMDQ_CREATE_CQ_PG_SIZE_PG_4K (0x0UL << 2) + #define CMDQ_CREATE_CQ_PG_SIZE_PG_8K (0x1UL << 2) + #define CMDQ_CREATE_CQ_PG_SIZE_PG_64K (0x2UL << 2) + #define CMDQ_CREATE_CQ_PG_SIZE_PG_2M (0x3UL << 2) + #define CMDQ_CREATE_CQ_PG_SIZE_PG_8M (0x4UL << 2) + #define CMDQ_CREATE_CQ_PG_SIZE_PG_1G (0x5UL << 2) + __le32 cq_fco_cnq_id; + #define CMDQ_CREATE_CQ_CNQ_ID_MASK 0xfffUL + #define CMDQ_CREATE_CQ_CNQ_ID_SFT 0 + #define CMDQ_CREATE_CQ_CQ_FCO_MASK 0xfffff000UL + #define CMDQ_CREATE_CQ_CQ_FCO_SFT 12 + __le32 dpi; + __le32 cq_size; + __le64 pbl; +}; + +/* Destroy CQ command (24 bytes) */ +struct cmdq_destroy_cq { + u8 opcode; + #define CMDQ_DESTROY_CQ_OPCODE_DESTROY_CQ 0xaUL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le32 cq_cid; + __le32 unused_0; +}; + +/* Resize CQ command (40 bytes) */ +struct cmdq_resize_cq { + u8 opcode; + #define CMDQ_RESIZE_CQ_OPCODE_RESIZE_CQ 0xcUL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le32 cq_cid; + __le32 new_cq_size_pg_size_lvl; + #define CMDQ_RESIZE_CQ_LVL_MASK 0x3UL + #define CMDQ_RESIZE_CQ_LVL_SFT 0 + #define CMDQ_RESIZE_CQ_LVL_LVL_0 0x0UL + #define CMDQ_RESIZE_CQ_LVL_LVL_1 0x1UL + #define CMDQ_RESIZE_CQ_LVL_LVL_2 0x2UL + #define CMDQ_RESIZE_CQ_PG_SIZE_MASK 0x1cUL + #define CMDQ_RESIZE_CQ_PG_SIZE_SFT 2 + #define CMDQ_RESIZE_CQ_PG_SIZE_PG_4K (0x0UL << 2) + #define CMDQ_RESIZE_CQ_PG_SIZE_PG_8K (0x1UL << 2) + #define CMDQ_RESIZE_CQ_PG_SIZE_PG_64K (0x2UL << 2) + #define CMDQ_RESIZE_CQ_PG_SIZE_PG_2M (0x3UL << 2) + #define CMDQ_RESIZE_CQ_PG_SIZE_PG_8M (0x4UL << 2) + #define CMDQ_RESIZE_CQ_PG_SIZE_PG_1G (0x5UL << 2) + #define CMDQ_RESIZE_CQ_NEW_CQ_SIZE_MASK 0x1fffe0UL + #define CMDQ_RESIZE_CQ_NEW_CQ_SIZE_SFT 5 + __le64 new_pbl; + __le32 new_cq_fco; + __le32 unused_2; +}; + +/* Allocate MRW command (32 bytes) */ +struct cmdq_allocate_mrw { + u8 opcode; + #define CMDQ_ALLOCATE_MRW_OPCODE_ALLOCATE_MRW 0xdUL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le64 mrw_handle; + u8 mrw_flags; + #define CMDQ_ALLOCATE_MRW_MRW_FLAGS_MASK 0xfUL + #define CMDQ_ALLOCATE_MRW_MRW_FLAGS_SFT 0 + #define CMDQ_ALLOCATE_MRW_MRW_FLAGS_MR 0x0UL + #define CMDQ_ALLOCATE_MRW_MRW_FLAGS_PMR 0x1UL + #define CMDQ_ALLOCATE_MRW_MRW_FLAGS_MW_TYPE1 0x2UL + #define CMDQ_ALLOCATE_MRW_MRW_FLAGS_MW_TYPE2A 0x3UL + #define CMDQ_ALLOCATE_MRW_MRW_FLAGS_MW_TYPE2B 0x4UL + u8 access; + #define CMDQ_ALLOCATE_MRW_ACCESS_RESERVED_MASK 0x1fUL + #define CMDQ_ALLOCATE_MRW_ACCESS_RESERVED_SFT 0 + #define CMDQ_ALLOCATE_MRW_ACCESS_CONSUMER_OWNED_KEY 0x20UL + __le16 unused_1; + __le32 pd_id; +}; + +/* De-allocate key command (24 bytes) */ +struct cmdq_deallocate_key { + u8 opcode; + #define CMDQ_DEALLOCATE_KEY_OPCODE_DEALLOCATE_KEY 0xeUL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + u8 mrw_flags; + #define CMDQ_DEALLOCATE_KEY_MRW_FLAGS_MASK 0xfUL + #define CMDQ_DEALLOCATE_KEY_MRW_FLAGS_SFT 0 + #define CMDQ_DEALLOCATE_KEY_MRW_FLAGS_MR 0x0UL + #define CMDQ_DEALLOCATE_KEY_MRW_FLAGS_PMR 0x1UL + #define CMDQ_DEALLOCATE_KEY_MRW_FLAGS_MW_TYPE1 0x2UL + #define CMDQ_DEALLOCATE_KEY_MRW_FLAGS_MW_TYPE2A 0x3UL + #define CMDQ_DEALLOCATE_KEY_MRW_FLAGS_MW_TYPE2B 0x4UL + u8 unused_1[3]; + __le32 key; +}; + +/* Register MR command (48 bytes) */ +struct cmdq_register_mr { + u8 opcode; + #define CMDQ_REGISTER_MR_OPCODE_REGISTER_MR 0xfUL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + u8 log2_pg_size_lvl; + #define CMDQ_REGISTER_MR_LVL_MASK 0x3UL + #define CMDQ_REGISTER_MR_LVL_SFT 0 + #define CMDQ_REGISTER_MR_LVL_LVL_0 0x0UL + #define CMDQ_REGISTER_MR_LVL_LVL_1 0x1UL + #define CMDQ_REGISTER_MR_LVL_LVL_2 0x2UL + #define CMDQ_REGISTER_MR_LOG2_PG_SIZE_MASK 0x7cUL + #define CMDQ_REGISTER_MR_LOG2_PG_SIZE_SFT 2 + u8 access; + #define CMDQ_REGISTER_MR_ACCESS_LOCAL_WRITE 0x1UL + #define CMDQ_REGISTER_MR_ACCESS_REMOTE_READ 0x2UL + #define CMDQ_REGISTER_MR_ACCESS_REMOTE_WRITE 0x4UL + #define CMDQ_REGISTER_MR_ACCESS_REMOTE_ATOMIC 0x8UL + #define CMDQ_REGISTER_MR_ACCESS_MW_BIND 0x10UL + #define CMDQ_REGISTER_MR_ACCESS_ZERO_BASED 0x20UL + __le16 unused_1; + __le32 key; + __le64 pbl; + __le64 va; + __le64 mr_size; +}; + +/* Deregister MR command (24 bytes) */ +struct cmdq_deregister_mr { + u8 opcode; + #define CMDQ_DEREGISTER_MR_OPCODE_DEREGISTER_MR 0x10UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le32 lkey; + __le32 unused_0; +}; + +/* Add GID command (48 bytes) */ +struct cmdq_add_gid { + u8 opcode; + #define CMDQ_ADD_GID_OPCODE_ADD_GID 0x11UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __be32 gid[4]; + __be16 src_mac[3]; + __le16 vlan; + #define CMDQ_ADD_GID_VLAN_VLAN_ID_MASK 0xfffUL + #define CMDQ_ADD_GID_VLAN_VLAN_ID_SFT 0 + #define CMDQ_ADD_GID_VLAN_TPID_MASK 0x7000UL + #define CMDQ_ADD_GID_VLAN_TPID_SFT 12 + #define CMDQ_ADD_GID_VLAN_TPID_TPID_88A8 (0x0UL << 12) + #define CMDQ_ADD_GID_VLAN_TPID_TPID_8100 (0x1UL << 12) + #define CMDQ_ADD_GID_VLAN_TPID_TPID_9100 (0x2UL << 12) + #define CMDQ_ADD_GID_VLAN_TPID_TPID_9200 (0x3UL << 12) + #define CMDQ_ADD_GID_VLAN_TPID_TPID_9300 (0x4UL << 12) + #define CMDQ_ADD_GID_VLAN_TPID_TPID_CFG1 (0x5UL << 12) + #define CMDQ_ADD_GID_VLAN_TPID_TPID_CFG2 (0x6UL << 12) + #define CMDQ_ADD_GID_VLAN_TPID_TPID_CFG3 (0x7UL << 12) + #define CMDQ_ADD_GID_VLAN_TPID_LAST CMDQ_ADD_GID_VLAN_TPID_TPID_CFG3 + #define CMDQ_ADD_GID_VLAN_VLAN_EN 0x8000UL + __le16 ipid; + __le16 stats_ctx; + #define CMDQ_ADD_GID_STATS_CTX_STATS_CTX_ID_MASK 0x7fffUL + #define CMDQ_ADD_GID_STATS_CTX_STATS_CTX_ID_SFT 0 + #define CMDQ_ADD_GID_STATS_CTX_STATS_CTX_VALID 0x8000UL + __le32 unused_0; +}; + +/* Delete GID command (24 bytes) */ +struct cmdq_delete_gid { + u8 opcode; + #define CMDQ_DELETE_GID_OPCODE_DELETE_GID 0x12UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le16 gid_index; + __le16 unused_0; + __le32 unused_1; +}; + +/* Modify GID command (48 bytes) */ +struct cmdq_modify_gid { + u8 opcode; + #define CMDQ_MODIFY_GID_OPCODE_MODIFY_GID 0x17UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le32 gid[4]; + __le16 src_mac[3]; + __le16 vlan; + #define CMDQ_MODIFY_GID_VLAN_VLAN_ID_MASK 0xfffUL + #define CMDQ_MODIFY_GID_VLAN_VLAN_ID_SFT 0 + #define CMDQ_MODIFY_GID_VLAN_TPID_MASK 0x7000UL + #define CMDQ_MODIFY_GID_VLAN_TPID_SFT 12 + #define CMDQ_MODIFY_GID_VLAN_TPID_TPID_88A8 (0x0UL << 12) + #define CMDQ_MODIFY_GID_VLAN_TPID_TPID_8100 (0x1UL << 12) + #define CMDQ_MODIFY_GID_VLAN_TPID_TPID_9100 (0x2UL << 12) + #define CMDQ_MODIFY_GID_VLAN_TPID_TPID_9200 (0x3UL << 12) + #define CMDQ_MODIFY_GID_VLAN_TPID_TPID_9300 (0x4UL << 12) + #define CMDQ_MODIFY_GID_VLAN_TPID_TPID_CFG1 (0x5UL << 12) + #define CMDQ_MODIFY_GID_VLAN_TPID_TPID_CFG2 (0x6UL << 12) + #define CMDQ_MODIFY_GID_VLAN_TPID_TPID_CFG3 (0x7UL << 12) + #define CMDQ_MODIFY_GID_VLAN_TPID_LAST \ + CMDQ_MODIFY_GID_VLAN_TPID_TPID_CFG3 + #define CMDQ_MODIFY_GID_VLAN_VLAN_EN 0x8000UL + __le16 ipid; + __le16 gid_index; + __le16 stats_ctx; + #define CMDQ_MODIFY_GID_STATS_CTX_STATS_CTX_ID_MASK 0x7fffUL + #define CMDQ_MODIFY_GID_STATS_CTX_STATS_CTX_ID_SFT 0 + #define CMDQ_MODIFY_GID_STATS_CTX_STATS_CTX_VALID 0x8000UL + __le16 unused_0; +}; + +/* Query GID command (24 bytes) */ +struct cmdq_query_gid { + u8 opcode; + #define CMDQ_QUERY_GID_OPCODE_QUERY_GID 0x18UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le16 gid_index; + __le16 unused_0; + __le32 unused_1; +}; + +/* Create QP1 command (80 bytes) */ +struct cmdq_create_qp1 { + u8 opcode; + #define CMDQ_CREATE_QP1_OPCODE_CREATE_QP1 0x13UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le64 qp_handle; + __le32 qp_flags; + #define CMDQ_CREATE_QP1_QP_FLAGS_SRQ_USED 0x1UL + #define CMDQ_CREATE_QP1_QP_FLAGS_FORCE_COMPLETION 0x2UL + #define CMDQ_CREATE_QP1_QP_FLAGS_RESERVED_LKEY_ENABLE 0x4UL + u8 type; + #define CMDQ_CREATE_QP1_TYPE_GSI 0x1UL + u8 sq_pg_size_sq_lvl; + #define CMDQ_CREATE_QP1_SQ_LVL_MASK 0xfUL + #define CMDQ_CREATE_QP1_SQ_LVL_SFT 0 + #define CMDQ_CREATE_QP1_SQ_LVL_LVL_0 0x0UL + #define CMDQ_CREATE_QP1_SQ_LVL_LVL_1 0x1UL + #define CMDQ_CREATE_QP1_SQ_LVL_LVL_2 0x2UL + #define CMDQ_CREATE_QP1_SQ_PG_SIZE_MASK 0xf0UL + #define CMDQ_CREATE_QP1_SQ_PG_SIZE_SFT 4 + #define CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_4K (0x0UL << 4) + #define CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_8K (0x1UL << 4) + #define CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_64K (0x2UL << 4) + #define CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_2M (0x3UL << 4) + #define CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_8M (0x4UL << 4) + #define CMDQ_CREATE_QP1_SQ_PG_SIZE_PG_1G (0x5UL << 4) + u8 rq_pg_size_rq_lvl; + #define CMDQ_CREATE_QP1_RQ_LVL_MASK 0xfUL + #define CMDQ_CREATE_QP1_RQ_LVL_SFT 0 + #define CMDQ_CREATE_QP1_RQ_LVL_LVL_0 0x0UL + #define CMDQ_CREATE_QP1_RQ_LVL_LVL_1 0x1UL + #define CMDQ_CREATE_QP1_RQ_LVL_LVL_2 0x2UL + #define CMDQ_CREATE_QP1_RQ_PG_SIZE_MASK 0xf0UL + #define CMDQ_CREATE_QP1_RQ_PG_SIZE_SFT 4 + #define CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_4K (0x0UL << 4) + #define CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_8K (0x1UL << 4) + #define CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_64K (0x2UL << 4) + #define CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_2M (0x3UL << 4) + #define CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_8M (0x4UL << 4) + #define CMDQ_CREATE_QP1_RQ_PG_SIZE_PG_1G (0x5UL << 4) + u8 unused_0; + __le32 dpi; + __le32 sq_size; + __le32 rq_size; + __le16 sq_fwo_sq_sge; + #define CMDQ_CREATE_QP1_SQ_SGE_MASK 0xfUL + #define CMDQ_CREATE_QP1_SQ_SGE_SFT 0 + #define CMDQ_CREATE_QP1_SQ_FWO_MASK 0xfff0UL + #define CMDQ_CREATE_QP1_SQ_FWO_SFT 4 + __le16 rq_fwo_rq_sge; + #define CMDQ_CREATE_QP1_RQ_SGE_MASK 0xfUL + #define CMDQ_CREATE_QP1_RQ_SGE_SFT 0 + #define CMDQ_CREATE_QP1_RQ_FWO_MASK 0xfff0UL + #define CMDQ_CREATE_QP1_RQ_FWO_SFT 4 + __le32 scq_cid; + __le32 rcq_cid; + __le32 srq_cid; + __le32 pd_id; + __le64 sq_pbl; + __le64 rq_pbl; +}; + +/* Destroy QP1 command (24 bytes) */ +struct cmdq_destroy_qp1 { + u8 opcode; + #define CMDQ_DESTROY_QP1_OPCODE_DESTROY_QP1 0x14UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le32 qp1_cid; + __le32 unused_0; +}; + +/* Create AH command (64 bytes) */ +struct cmdq_create_ah { + u8 opcode; + #define CMDQ_CREATE_AH_OPCODE_CREATE_AH 0x15UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le64 ah_handle; + __le32 dgid[4]; + u8 type; + #define CMDQ_CREATE_AH_TYPE_V1 0x0UL + #define CMDQ_CREATE_AH_TYPE_V2IPV4 0x2UL + #define CMDQ_CREATE_AH_TYPE_V2IPV6 0x3UL + u8 hop_limit; + __le16 sgid_index; + __le32 dest_vlan_id_flow_label; + #define CMDQ_CREATE_AH_FLOW_LABEL_MASK 0xfffffUL + #define CMDQ_CREATE_AH_FLOW_LABEL_SFT 0 + #define CMDQ_CREATE_AH_DEST_VLAN_ID_MASK 0xfff00000UL + #define CMDQ_CREATE_AH_DEST_VLAN_ID_SFT 20 + __le32 pd_id; + __le32 unused_0; + __le16 dest_mac[3]; + u8 traffic_class; + u8 unused_1; +}; + +/* Destroy AH command (24 bytes) */ +struct cmdq_destroy_ah { + u8 opcode; + #define CMDQ_DESTROY_AH_OPCODE_DESTROY_AH 0x16UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le32 ah_cid; + __le32 unused_0; +}; + +/* Initialize Firmware command (112 bytes) */ +struct cmdq_initialize_fw { + u8 opcode; + #define CMDQ_INITIALIZE_FW_OPCODE_INITIALIZE_FW 0x80UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + u8 qpc_pg_size_qpc_lvl; + #define CMDQ_INITIALIZE_FW_QPC_LVL_MASK 0xfUL + #define CMDQ_INITIALIZE_FW_QPC_LVL_SFT 0 + #define CMDQ_INITIALIZE_FW_QPC_LVL_LVL_0 0x0UL + #define CMDQ_INITIALIZE_FW_QPC_LVL_LVL_1 0x1UL + #define CMDQ_INITIALIZE_FW_QPC_LVL_LVL_2 0x2UL + #define CMDQ_INITIALIZE_FW_QPC_PG_SIZE_MASK 0xf0UL + #define CMDQ_INITIALIZE_FW_QPC_PG_SIZE_SFT 4 + #define CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_4K (0x0UL << 4) + #define CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_8K (0x1UL << 4) + #define CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_64K (0x2UL << 4) + #define CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_2M (0x3UL << 4) + #define CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_8M (0x4UL << 4) + #define CMDQ_INITIALIZE_FW_QPC_PG_SIZE_PG_1G (0x5UL << 4) + u8 mrw_pg_size_mrw_lvl; + #define CMDQ_INITIALIZE_FW_MRW_LVL_MASK 0xfUL + #define CMDQ_INITIALIZE_FW_MRW_LVL_SFT 0 + #define CMDQ_INITIALIZE_FW_MRW_LVL_LVL_0 0x0UL + #define CMDQ_INITIALIZE_FW_MRW_LVL_LVL_1 0x1UL + #define CMDQ_INITIALIZE_FW_MRW_LVL_LVL_2 0x2UL + #define CMDQ_INITIALIZE_FW_MRW_PG_SIZE_MASK 0xf0UL + #define CMDQ_INITIALIZE_FW_MRW_PG_SIZE_SFT 4 + #define CMDQ_INITIALIZE_FW_MRW_PG_SIZE_PG_4K (0x0UL << 4) + #define CMDQ_INITIALIZE_FW_MRW_PG_SIZE_PG_8K (0x1UL << 4) + #define CMDQ_INITIALIZE_FW_MRW_PG_SIZE_PG_64K (0x2UL << 4) + #define CMDQ_INITIALIZE_FW_MRW_PG_SIZE_PG_2M (0x3UL << 4) + #define CMDQ_INITIALIZE_FW_MRW_PG_SIZE_PG_8M (0x4UL << 4) + #define CMDQ_INITIALIZE_FW_MRW_PG_SIZE_PG_1G (0x5UL << 4) + u8 srq_pg_size_srq_lvl; + #define CMDQ_INITIALIZE_FW_SRQ_LVL_MASK 0xfUL + #define CMDQ_INITIALIZE_FW_SRQ_LVL_SFT 0 + #define CMDQ_INITIALIZE_FW_SRQ_LVL_LVL_0 0x0UL + #define CMDQ_INITIALIZE_FW_SRQ_LVL_LVL_1 0x1UL + #define CMDQ_INITIALIZE_FW_SRQ_LVL_LVL_2 0x2UL + #define CMDQ_INITIALIZE_FW_SRQ_PG_SIZE_MASK 0xf0UL + #define CMDQ_INITIALIZE_FW_SRQ_PG_SIZE_SFT 4 + #define CMDQ_INITIALIZE_FW_SRQ_PG_SIZE_PG_4K (0x0UL << 4) + #define CMDQ_INITIALIZE_FW_SRQ_PG_SIZE_PG_8K (0x1UL << 4) + #define CMDQ_INITIALIZE_FW_SRQ_PG_SIZE_PG_64K (0x2UL << 4) + #define CMDQ_INITIALIZE_FW_SRQ_PG_SIZE_PG_2M (0x3UL << 4) + #define CMDQ_INITIALIZE_FW_SRQ_PG_SIZE_PG_8M (0x4UL << 4) + #define CMDQ_INITIALIZE_FW_SRQ_PG_SIZE_PG_1G (0x5UL << 4) + u8 cq_pg_size_cq_lvl; + #define CMDQ_INITIALIZE_FW_CQ_LVL_MASK 0xfUL + #define CMDQ_INITIALIZE_FW_CQ_LVL_SFT 0 + #define CMDQ_INITIALIZE_FW_CQ_LVL_LVL_0 0x0UL + #define CMDQ_INITIALIZE_FW_CQ_LVL_LVL_1 0x1UL + #define CMDQ_INITIALIZE_FW_CQ_LVL_LVL_2 0x2UL + #define CMDQ_INITIALIZE_FW_CQ_PG_SIZE_MASK 0xf0UL + #define CMDQ_INITIALIZE_FW_CQ_PG_SIZE_SFT 4 + #define CMDQ_INITIALIZE_FW_CQ_PG_SIZE_PG_4K (0x0UL << 4) + #define CMDQ_INITIALIZE_FW_CQ_PG_SIZE_PG_8K (0x1UL << 4) + #define CMDQ_INITIALIZE_FW_CQ_PG_SIZE_PG_64K (0x2UL << 4) + #define CMDQ_INITIALIZE_FW_CQ_PG_SIZE_PG_2M (0x3UL << 4) + #define CMDQ_INITIALIZE_FW_CQ_PG_SIZE_PG_8M (0x4UL << 4) + #define CMDQ_INITIALIZE_FW_CQ_PG_SIZE_PG_1G (0x5UL << 4) + u8 tqm_pg_size_tqm_lvl; + #define CMDQ_INITIALIZE_FW_TQM_LVL_MASK 0xfUL + #define CMDQ_INITIALIZE_FW_TQM_LVL_SFT 0 + #define CMDQ_INITIALIZE_FW_TQM_LVL_LVL_0 0x0UL + #define CMDQ_INITIALIZE_FW_TQM_LVL_LVL_1 0x1UL + #define CMDQ_INITIALIZE_FW_TQM_LVL_LVL_2 0x2UL + #define CMDQ_INITIALIZE_FW_TQM_PG_SIZE_MASK 0xf0UL + #define CMDQ_INITIALIZE_FW_TQM_PG_SIZE_SFT 4 + #define CMDQ_INITIALIZE_FW_TQM_PG_SIZE_PG_4K (0x0UL << 4) + #define CMDQ_INITIALIZE_FW_TQM_PG_SIZE_PG_8K (0x1UL << 4) + #define CMDQ_INITIALIZE_FW_TQM_PG_SIZE_PG_64K (0x2UL << 4) + #define CMDQ_INITIALIZE_FW_TQM_PG_SIZE_PG_2M (0x3UL << 4) + #define CMDQ_INITIALIZE_FW_TQM_PG_SIZE_PG_8M (0x4UL << 4) + #define CMDQ_INITIALIZE_FW_TQM_PG_SIZE_PG_1G (0x5UL << 4) + u8 tim_pg_size_tim_lvl; + #define CMDQ_INITIALIZE_FW_TIM_LVL_MASK 0xfUL + #define CMDQ_INITIALIZE_FW_TIM_LVL_SFT 0 + #define CMDQ_INITIALIZE_FW_TIM_LVL_LVL_0 0x0UL + #define CMDQ_INITIALIZE_FW_TIM_LVL_LVL_1 0x1UL + #define CMDQ_INITIALIZE_FW_TIM_LVL_LVL_2 0x2UL + #define CMDQ_INITIALIZE_FW_TIM_PG_SIZE_MASK 0xf0UL + #define CMDQ_INITIALIZE_FW_TIM_PG_SIZE_SFT 4 + #define CMDQ_INITIALIZE_FW_TIM_PG_SIZE_PG_4K (0x0UL << 4) + #define CMDQ_INITIALIZE_FW_TIM_PG_SIZE_PG_8K (0x1UL << 4) + #define CMDQ_INITIALIZE_FW_TIM_PG_SIZE_PG_64K (0x2UL << 4) + #define CMDQ_INITIALIZE_FW_TIM_PG_SIZE_PG_2M (0x3UL << 4) + #define CMDQ_INITIALIZE_FW_TIM_PG_SIZE_PG_8M (0x4UL << 4) + #define CMDQ_INITIALIZE_FW_TIM_PG_SIZE_PG_1G (0x5UL << 4) + __le16 reserved16; + __le64 qpc_page_dir; + __le64 mrw_page_dir; + __le64 srq_page_dir; + __le64 cq_page_dir; + __le64 tqm_page_dir; + __le64 tim_page_dir; + __le32 number_of_qp; + __le32 number_of_mrw; + __le32 number_of_srq; + __le32 number_of_cq; + __le32 max_qp_per_vf; + __le32 max_mrw_per_vf; + __le32 max_srq_per_vf; + __le32 max_cq_per_vf; + __le32 max_gid_per_vf; + __le32 stat_ctx_id; +}; + +/* De-initialize Firmware command (16 bytes) */ +struct cmdq_deinitialize_fw { + u8 opcode; + #define CMDQ_DEINITIALIZE_FW_OPCODE_DEINITIALIZE_FW 0x81UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; +}; + +/* Stop function command (16 bytes) */ +struct cmdq_stop_func { + u8 opcode; + #define CMDQ_STOP_FUNC_OPCODE_STOP_FUNC 0x82UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; +}; + +/* Query function command (16 bytes) */ +struct cmdq_query_func { + u8 opcode; + #define CMDQ_QUERY_FUNC_OPCODE_QUERY_FUNC 0x83UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; +}; + +/* Set function resources command (16 bytes) */ +struct cmdq_set_func_resources { + u8 opcode; + #define CMDQ_SET_FUNC_RESOURCES_OPCODE_SET_FUNC_RESOURCES 0x84UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; +}; + +/* Read hardware resource context command (24 bytes) */ +struct cmdq_read_context { + u8 opcode; + #define CMDQ_READ_CONTEXT_OPCODE_READ_CONTEXT 0x85UL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le32 type_xid; + #define CMDQ_READ_CONTEXT_XID_MASK 0xffffffUL + #define CMDQ_READ_CONTEXT_XID_SFT 0 + #define CMDQ_READ_CONTEXT_TYPE_MASK 0xff000000UL + #define CMDQ_READ_CONTEXT_TYPE_SFT 24 + #define CMDQ_READ_CONTEXT_TYPE_QPC (0x0UL << 24) + #define CMDQ_READ_CONTEXT_TYPE_CQ (0x1UL << 24) + #define CMDQ_READ_CONTEXT_TYPE_MRW (0x2UL << 24) + #define CMDQ_READ_CONTEXT_TYPE_SRQ (0x3UL << 24) + __le32 unused_0; +}; + +/* Map TC to COS. Can only be issued from a PF (24 bytes) */ +struct cmdq_map_tc_to_cos { + u8 opcode; + #define CMDQ_MAP_TC_TO_COS_OPCODE_MAP_TC_TO_COS 0x8aUL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; + __le16 cos0; + #define CMDQ_MAP_TC_TO_COS_COS0_NO_CHANGE 0xffffUL + __le16 cos1; + #define CMDQ_MAP_TC_TO_COS_COS1_DISABLE 0x8000UL + #define CMDQ_MAP_TC_TO_COS_COS1_NO_CHANGE 0xffffUL + __le32 unused_0; +}; + +/* Query version command (16 bytes) */ +struct cmdq_query_version { + u8 opcode; + #define CMDQ_QUERY_VERSION_OPCODE_QUERY_VERSION 0x8bUL + u8 cmd_size; + __le16 flags; + __le16 cookie; + u8 resp_size; + u8 reserved8; + __le64 resp_addr; +}; + +/* Command-Response Event Queue (CREQ) Structures */ +/* Base CREQ Record (16 bytes) */ +struct creq_base { + u8 type; + #define CREQ_BASE_TYPE_MASK 0x3fUL + #define CREQ_BASE_TYPE_SFT 0 + #define CREQ_BASE_TYPE_QP_EVENT 0x38UL + #define CREQ_BASE_TYPE_FUNC_EVENT 0x3aUL + #define CREQ_BASE_RESERVED2_MASK 0xc0UL + #define CREQ_BASE_RESERVED2_SFT 6 + u8 reserved56[7]; + u8 v; + #define CREQ_BASE_V 0x1UL + #define CREQ_BASE_RESERVED7_MASK 0xfeUL + #define CREQ_BASE_RESERVED7_SFT 1 + u8 event; + __le16 reserved48[3]; +}; + +/* RoCE Function Async Event Notification (16 bytes) */ +struct creq_func_event { + u8 type; + #define CREQ_FUNC_EVENT_TYPE_MASK 0x3fUL + #define CREQ_FUNC_EVENT_TYPE_SFT 0 + #define CREQ_FUNC_EVENT_TYPE_FUNC_EVENT 0x3aUL + #define CREQ_FUNC_EVENT_RESERVED2_MASK 0xc0UL + #define CREQ_FUNC_EVENT_RESERVED2_SFT 6 + u8 reserved56[7]; + u8 v; + #define CREQ_FUNC_EVENT_V 0x1UL + #define CREQ_FUNC_EVENT_RESERVED7_MASK 0xfeUL + #define CREQ_FUNC_EVENT_RESERVED7_SFT 1 + u8 event; + #define CREQ_FUNC_EVENT_EVENT_TX_WQE_ERROR 0x1UL + #define CREQ_FUNC_EVENT_EVENT_TX_DATA_ERROR 0x2UL + #define CREQ_FUNC_EVENT_EVENT_RX_WQE_ERROR 0x3UL + #define CREQ_FUNC_EVENT_EVENT_RX_DATA_ERROR 0x4UL + #define CREQ_FUNC_EVENT_EVENT_CQ_ERROR 0x5UL + #define CREQ_FUNC_EVENT_EVENT_TQM_ERROR 0x6UL + #define CREQ_FUNC_EVENT_EVENT_CFCQ_ERROR 0x7UL + #define CREQ_FUNC_EVENT_EVENT_CFCS_ERROR 0x8UL + #define CREQ_FUNC_EVENT_EVENT_CFCC_ERROR 0x9UL + #define CREQ_FUNC_EVENT_EVENT_CFCM_ERROR 0xaUL + #define CREQ_FUNC_EVENT_EVENT_TIM_ERROR 0xbUL + #define CREQ_FUNC_EVENT_EVENT_VF_COMM_REQUEST 0x80UL + #define CREQ_FUNC_EVENT_EVENT_RESOURCE_EXHAUSTED 0x81UL + __le16 reserved48[3]; +}; + +/* RoCE Slowpath Command Completion (16 bytes) */ +struct creq_qp_event { + u8 type; + #define CREQ_QP_EVENT_TYPE_MASK 0x3fUL + #define CREQ_QP_EVENT_TYPE_SFT 0 + #define CREQ_QP_EVENT_TYPE_QP_EVENT 0x38UL + #define CREQ_QP_EVENT_RESERVED2_MASK 0xc0UL + #define CREQ_QP_EVENT_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 reserved32; + u8 v; + #define CREQ_QP_EVENT_V 0x1UL + #define CREQ_QP_EVENT_RESERVED7_MASK 0xfeUL + #define CREQ_QP_EVENT_RESERVED7_SFT 1 + u8 event; + #define CREQ_QP_EVENT_EVENT_CREATE_QP 0x1UL + #define CREQ_QP_EVENT_EVENT_DESTROY_QP 0x2UL + #define CREQ_QP_EVENT_EVENT_MODIFY_QP 0x3UL + #define CREQ_QP_EVENT_EVENT_QUERY_QP 0x4UL + #define CREQ_QP_EVENT_EVENT_CREATE_SRQ 0x5UL + #define CREQ_QP_EVENT_EVENT_DESTROY_SRQ 0x6UL + #define CREQ_QP_EVENT_EVENT_QUERY_SRQ 0x8UL + #define CREQ_QP_EVENT_EVENT_CREATE_CQ 0x9UL + #define CREQ_QP_EVENT_EVENT_DESTROY_CQ 0xaUL + #define CREQ_QP_EVENT_EVENT_RESIZE_CQ 0xcUL + #define CREQ_QP_EVENT_EVENT_ALLOCATE_MRW 0xdUL + #define CREQ_QP_EVENT_EVENT_DEALLOCATE_KEY 0xeUL + #define CREQ_QP_EVENT_EVENT_REGISTER_MR 0xfUL + #define CREQ_QP_EVENT_EVENT_DEREGISTER_MR 0x10UL + #define CREQ_QP_EVENT_EVENT_ADD_GID 0x11UL + #define CREQ_QP_EVENT_EVENT_DELETE_GID 0x12UL + #define CREQ_QP_EVENT_EVENT_MODIFY_GID 0x17UL + #define CREQ_QP_EVENT_EVENT_QUERY_GID 0x18UL + #define CREQ_QP_EVENT_EVENT_CREATE_QP1 0x13UL + #define CREQ_QP_EVENT_EVENT_DESTROY_QP1 0x14UL + #define CREQ_QP_EVENT_EVENT_CREATE_AH 0x15UL + #define CREQ_QP_EVENT_EVENT_DESTROY_AH 0x16UL + #define CREQ_QP_EVENT_EVENT_INITIALIZE_FW 0x80UL + #define CREQ_QP_EVENT_EVENT_DEINITIALIZE_FW 0x81UL + #define CREQ_QP_EVENT_EVENT_STOP_FUNC 0x82UL + #define CREQ_QP_EVENT_EVENT_QUERY_FUNC 0x83UL + #define CREQ_QP_EVENT_EVENT_SET_FUNC_RESOURCES 0x84UL + #define CREQ_QP_EVENT_EVENT_MAP_TC_TO_COS 0x8aUL + #define CREQ_QP_EVENT_EVENT_QUERY_VERSION 0x8bUL + #define CREQ_QP_EVENT_EVENT_MODIFY_CC 0x8cUL + #define CREQ_QP_EVENT_EVENT_QUERY_CC 0x8dUL + #define CREQ_QP_EVENT_EVENT_QP_ERROR_NOTIFICATION 0xc0UL + __le16 reserved48[3]; +}; + +/* Create QP command response (16 bytes) */ +struct creq_create_qp_resp { + u8 type; + #define CREQ_CREATE_QP_RESP_TYPE_MASK 0x3fUL + #define CREQ_CREATE_QP_RESP_TYPE_SFT 0 + #define CREQ_CREATE_QP_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_CREATE_QP_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_CREATE_QP_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_CREATE_QP_RESP_V 0x1UL + #define CREQ_CREATE_QP_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_CREATE_QP_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_CREATE_QP_RESP_EVENT_CREATE_QP 0x1UL + __le16 reserved48[3]; +}; + +/* Destroy QP command response (16 bytes) */ +struct creq_destroy_qp_resp { + u8 type; + #define CREQ_DESTROY_QP_RESP_TYPE_MASK 0x3fUL + #define CREQ_DESTROY_QP_RESP_TYPE_SFT 0 + #define CREQ_DESTROY_QP_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_DESTROY_QP_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_DESTROY_QP_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_DESTROY_QP_RESP_V 0x1UL + #define CREQ_DESTROY_QP_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_DESTROY_QP_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_DESTROY_QP_RESP_EVENT_DESTROY_QP 0x2UL + __le16 reserved48[3]; +}; + +/* Modify QP command response (16 bytes) */ +struct creq_modify_qp_resp { + u8 type; + #define CREQ_MODIFY_QP_RESP_TYPE_MASK 0x3fUL + #define CREQ_MODIFY_QP_RESP_TYPE_SFT 0 + #define CREQ_MODIFY_QP_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_MODIFY_QP_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_MODIFY_QP_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_MODIFY_QP_RESP_V 0x1UL + #define CREQ_MODIFY_QP_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_MODIFY_QP_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_MODIFY_QP_RESP_EVENT_MODIFY_QP 0x3UL + __le16 reserved48[3]; +}; + +/* Query QP command response (16 bytes) */ +struct creq_query_qp_resp { + u8 type; + #define CREQ_QUERY_QP_RESP_TYPE_MASK 0x3fUL + #define CREQ_QUERY_QP_RESP_TYPE_SFT 0 + #define CREQ_QUERY_QP_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_QUERY_QP_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_QUERY_QP_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 size; + u8 v; + #define CREQ_QUERY_QP_RESP_V 0x1UL + #define CREQ_QUERY_QP_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_QUERY_QP_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_QUERY_QP_RESP_EVENT_QUERY_QP 0x4UL + __le16 reserved48[3]; +}; + +/* Query QP command response side buffer structure (104 bytes) */ +struct creq_query_qp_resp_sb { + u8 opcode; + #define CREQ_QUERY_QP_RESP_SB_OPCODE_QUERY_QP 0x4UL + u8 status; + __le16 cookie; + __le16 flags; + u8 resp_size; + u8 reserved8; + __le32 xid; + u8 en_sqd_async_notify_state; + #define CREQ_QUERY_QP_RESP_SB_STATE_MASK 0xfUL + #define CREQ_QUERY_QP_RESP_SB_STATE_SFT 0 + #define CREQ_QUERY_QP_RESP_SB_STATE_RESET 0x0UL + #define CREQ_QUERY_QP_RESP_SB_STATE_INIT 0x1UL + #define CREQ_QUERY_QP_RESP_SB_STATE_RTR 0x2UL + #define CREQ_QUERY_QP_RESP_SB_STATE_RTS 0x3UL + #define CREQ_QUERY_QP_RESP_SB_STATE_SQD 0x4UL + #define CREQ_QUERY_QP_RESP_SB_STATE_SQE 0x5UL + #define CREQ_QUERY_QP_RESP_SB_STATE_ERR 0x6UL + #define CREQ_QUERY_QP_RESP_SB_EN_SQD_ASYNC_NOTIFY 0x10UL + u8 access; + #define CREQ_QUERY_QP_RESP_SB_ACCESS_LOCAL_WRITE 0x1UL + #define CREQ_QUERY_QP_RESP_SB_ACCESS_REMOTE_WRITE 0x2UL + #define CREQ_QUERY_QP_RESP_SB_ACCESS_REMOTE_READ 0x4UL + #define CREQ_QUERY_QP_RESP_SB_ACCESS_REMOTE_ATOMIC 0x8UL + __le16 pkey; + __le32 qkey; + __le32 reserved32; + __le32 dgid[4]; + __le32 flow_label; + __le16 sgid_index; + u8 hop_limit; + u8 traffic_class; + __le16 dest_mac[3]; + __le16 path_mtu_dest_vlan_id; + #define CREQ_QUERY_QP_RESP_SB_DEST_VLAN_ID_MASK 0xfffUL + #define CREQ_QUERY_QP_RESP_SB_DEST_VLAN_ID_SFT 0 + #define CREQ_QUERY_QP_RESP_SB_PATH_MTU_MASK 0xf000UL + #define CREQ_QUERY_QP_RESP_SB_PATH_MTU_SFT 12 + #define CREQ_QUERY_QP_RESP_SB_PATH_MTU_MTU_256 (0x0UL << 12) + #define CREQ_QUERY_QP_RESP_SB_PATH_MTU_MTU_512 (0x1UL << 12) + #define CREQ_QUERY_QP_RESP_SB_PATH_MTU_MTU_1024 (0x2UL << 12) + #define CREQ_QUERY_QP_RESP_SB_PATH_MTU_MTU_2048 (0x3UL << 12) + #define CREQ_QUERY_QP_RESP_SB_PATH_MTU_MTU_4096 (0x4UL << 12) + #define CREQ_QUERY_QP_RESP_SB_PATH_MTU_MTU_8192 (0x5UL << 12) + u8 timeout; + u8 retry_cnt; + u8 rnr_retry; + u8 min_rnr_timer; + __le32 rq_psn; + __le32 sq_psn; + u8 max_rd_atomic; + u8 max_dest_rd_atomic; + u8 tos_dscp_tos_ecn; + #define CREQ_QUERY_QP_RESP_SB_TOS_ECN_MASK 0x3UL + #define CREQ_QUERY_QP_RESP_SB_TOS_ECN_SFT 0 + #define CREQ_QUERY_QP_RESP_SB_TOS_DSCP_MASK 0xfcUL + #define CREQ_QUERY_QP_RESP_SB_TOS_DSCP_SFT 2 + u8 enable_cc; + #define CREQ_QUERY_QP_RESP_SB_ENABLE_CC 0x1UL + #define CREQ_QUERY_QP_RESP_SB_RESERVED7_MASK 0xfeUL + #define CREQ_QUERY_QP_RESP_SB_RESERVED7_SFT 1 + __le32 sq_size; + __le32 rq_size; + __le16 sq_sge; + __le16 rq_sge; + __le32 max_inline_data; + __le32 dest_qp_id; + __le32 unused_1; + __le16 src_mac[3]; + __le16 vlan_pcp_vlan_dei_vlan_id; + #define CREQ_QUERY_QP_RESP_SB_VLAN_ID_MASK 0xfffUL + #define CREQ_QUERY_QP_RESP_SB_VLAN_ID_SFT 0 + #define CREQ_QUERY_QP_RESP_SB_VLAN_DEI 0x1000UL + #define CREQ_QUERY_QP_RESP_SB_VLAN_PCP_MASK 0xe000UL + #define CREQ_QUERY_QP_RESP_SB_VLAN_PCP_SFT 13 +}; + +/* Create SRQ command response (16 bytes) */ +struct creq_create_srq_resp { + u8 type; + #define CREQ_CREATE_SRQ_RESP_TYPE_MASK 0x3fUL + #define CREQ_CREATE_SRQ_RESP_TYPE_SFT 0 + #define CREQ_CREATE_SRQ_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_CREATE_SRQ_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_CREATE_SRQ_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_CREATE_SRQ_RESP_V 0x1UL + #define CREQ_CREATE_SRQ_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_CREATE_SRQ_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_CREATE_SRQ_RESP_EVENT_CREATE_SRQ 0x5UL + __le16 reserved48[3]; +}; + +/* Destroy SRQ command response (16 bytes) */ +struct creq_destroy_srq_resp { + u8 type; + #define CREQ_DESTROY_SRQ_RESP_TYPE_MASK 0x3fUL + #define CREQ_DESTROY_SRQ_RESP_TYPE_SFT 0 + #define CREQ_DESTROY_SRQ_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_DESTROY_SRQ_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_DESTROY_SRQ_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_DESTROY_SRQ_RESP_V 0x1UL + #define CREQ_DESTROY_SRQ_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_DESTROY_SRQ_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_DESTROY_SRQ_RESP_EVENT_DESTROY_SRQ 0x6UL + __le16 enable_for_arm[3]; + #define CREQ_DESTROY_SRQ_RESP_ENABLE_FOR_ARM_MASK 0x30000UL + #define CREQ_DESTROY_SRQ_RESP_ENABLE_FOR_ARM_SFT 16 + #define CREQ_DESTROY_SRQ_RESP_RESERVED46_MASK 0xfffc0000UL + #define CREQ_DESTROY_SRQ_RESP_RESERVED46_SFT 18 +}; + +/* Query SRQ command response (16 bytes) */ +struct creq_query_srq_resp { + u8 type; + #define CREQ_QUERY_SRQ_RESP_TYPE_MASK 0x3fUL + #define CREQ_QUERY_SRQ_RESP_TYPE_SFT 0 + #define CREQ_QUERY_SRQ_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_QUERY_SRQ_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_QUERY_SRQ_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 size; + u8 v; + #define CREQ_QUERY_SRQ_RESP_V 0x1UL + #define CREQ_QUERY_SRQ_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_QUERY_SRQ_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_QUERY_SRQ_RESP_EVENT_QUERY_SRQ 0x8UL + __le16 reserved48[3]; +}; + +/* Query SRQ command response side buffer structure (24 bytes) */ +struct creq_query_srq_resp_sb { + u8 opcode; + #define CREQ_QUERY_SRQ_RESP_SB_OPCODE_QUERY_SRQ 0x8UL + u8 status; + __le16 cookie; + __le16 flags; + u8 resp_size; + u8 reserved8; + __le32 xid; + __le16 srq_limit; + __le16 reserved16; + __le32 data[4]; +}; + +/* Create CQ command Response (16 bytes) */ +struct creq_create_cq_resp { + u8 type; + #define CREQ_CREATE_CQ_RESP_TYPE_MASK 0x3fUL + #define CREQ_CREATE_CQ_RESP_TYPE_SFT 0 + #define CREQ_CREATE_CQ_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_CREATE_CQ_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_CREATE_CQ_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_CREATE_CQ_RESP_V 0x1UL + #define CREQ_CREATE_CQ_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_CREATE_CQ_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_CREATE_CQ_RESP_EVENT_CREATE_CQ 0x9UL + __le16 reserved48[3]; +}; + +/* Destroy CQ command response (16 bytes) */ +struct creq_destroy_cq_resp { + u8 type; + #define CREQ_DESTROY_CQ_RESP_TYPE_MASK 0x3fUL + #define CREQ_DESTROY_CQ_RESP_TYPE_SFT 0 + #define CREQ_DESTROY_CQ_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_DESTROY_CQ_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_DESTROY_CQ_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_DESTROY_CQ_RESP_V 0x1UL + #define CREQ_DESTROY_CQ_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_DESTROY_CQ_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_DESTROY_CQ_RESP_EVENT_DESTROY_CQ 0xaUL + __le16 cq_arm_lvl; + #define CREQ_DESTROY_CQ_RESP_CQ_ARM_LVL_MASK 0x3UL + #define CREQ_DESTROY_CQ_RESP_CQ_ARM_LVL_SFT 0 + #define CREQ_DESTROY_CQ_RESP_RESERVED14_MASK 0xfffcUL + #define CREQ_DESTROY_CQ_RESP_RESERVED14_SFT 2 + __le16 total_cnq_events; + __le16 reserved16; +}; + +/* Resize CQ command response (16 bytes) */ +struct creq_resize_cq_resp { + u8 type; + #define CREQ_RESIZE_CQ_RESP_TYPE_MASK 0x3fUL + #define CREQ_RESIZE_CQ_RESP_TYPE_SFT 0 + #define CREQ_RESIZE_CQ_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_RESIZE_CQ_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_RESIZE_CQ_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_RESIZE_CQ_RESP_V 0x1UL + #define CREQ_RESIZE_CQ_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_RESIZE_CQ_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_RESIZE_CQ_RESP_EVENT_RESIZE_CQ 0xcUL + __le16 reserved48[3]; +}; + +/* Allocate MRW command response (16 bytes) */ +struct creq_allocate_mrw_resp { + u8 type; + #define CREQ_ALLOCATE_MRW_RESP_TYPE_MASK 0x3fUL + #define CREQ_ALLOCATE_MRW_RESP_TYPE_SFT 0 + #define CREQ_ALLOCATE_MRW_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_ALLOCATE_MRW_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_ALLOCATE_MRW_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_ALLOCATE_MRW_RESP_V 0x1UL + #define CREQ_ALLOCATE_MRW_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_ALLOCATE_MRW_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_ALLOCATE_MRW_RESP_EVENT_ALLOCATE_MRW 0xdUL + __le16 reserved48[3]; +}; + +/* De-allocate key command response (16 bytes) */ +struct creq_deallocate_key_resp { + u8 type; + #define CREQ_DEALLOCATE_KEY_RESP_TYPE_MASK 0x3fUL + #define CREQ_DEALLOCATE_KEY_RESP_TYPE_SFT 0 + #define CREQ_DEALLOCATE_KEY_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_DEALLOCATE_KEY_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_DEALLOCATE_KEY_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_DEALLOCATE_KEY_RESP_V 0x1UL + #define CREQ_DEALLOCATE_KEY_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_DEALLOCATE_KEY_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_DEALLOCATE_KEY_RESP_EVENT_DEALLOCATE_KEY 0xeUL + __le16 reserved16; + __le32 bound_window_info; +}; + +/* Register MR command response (16 bytes) */ +struct creq_register_mr_resp { + u8 type; + #define CREQ_REGISTER_MR_RESP_TYPE_MASK 0x3fUL + #define CREQ_REGISTER_MR_RESP_TYPE_SFT 0 + #define CREQ_REGISTER_MR_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_REGISTER_MR_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_REGISTER_MR_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_REGISTER_MR_RESP_V 0x1UL + #define CREQ_REGISTER_MR_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_REGISTER_MR_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_REGISTER_MR_RESP_EVENT_REGISTER_MR 0xfUL + __le16 reserved48[3]; +}; + +/* Deregister MR command response (16 bytes) */ +struct creq_deregister_mr_resp { + u8 type; + #define CREQ_DEREGISTER_MR_RESP_TYPE_MASK 0x3fUL + #define CREQ_DEREGISTER_MR_RESP_TYPE_SFT 0 + #define CREQ_DEREGISTER_MR_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_DEREGISTER_MR_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_DEREGISTER_MR_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_DEREGISTER_MR_RESP_V 0x1UL + #define CREQ_DEREGISTER_MR_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_DEREGISTER_MR_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_DEREGISTER_MR_RESP_EVENT_DEREGISTER_MR 0x10UL + __le16 reserved16; + __le32 bound_windows; +}; + +/* Add GID command response (16 bytes) */ +struct creq_add_gid_resp { + u8 type; + #define CREQ_ADD_GID_RESP_TYPE_MASK 0x3fUL + #define CREQ_ADD_GID_RESP_TYPE_SFT 0 + #define CREQ_ADD_GID_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_ADD_GID_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_ADD_GID_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_ADD_GID_RESP_V 0x1UL + #define CREQ_ADD_GID_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_ADD_GID_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_ADD_GID_RESP_EVENT_ADD_GID 0x11UL + __le16 reserved48[3]; +}; + +/* Delete GID command response (16 bytes) */ +struct creq_delete_gid_resp { + u8 type; + #define CREQ_DELETE_GID_RESP_TYPE_MASK 0x3fUL + #define CREQ_DELETE_GID_RESP_TYPE_SFT 0 + #define CREQ_DELETE_GID_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_DELETE_GID_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_DELETE_GID_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_DELETE_GID_RESP_V 0x1UL + #define CREQ_DELETE_GID_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_DELETE_GID_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_DELETE_GID_RESP_EVENT_DELETE_GID 0x12UL + __le16 reserved48[3]; +}; + +/* Modify GID command response (16 bytes) */ +struct creq_modify_gid_resp { + u8 type; + #define CREQ_MODIFY_GID_RESP_TYPE_MASK 0x3fUL + #define CREQ_MODIFY_GID_RESP_TYPE_SFT 0 + #define CREQ_MODIFY_GID_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_MODIFY_GID_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_MODIFY_GID_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_MODIFY_GID_RESP_V 0x1UL + #define CREQ_MODIFY_GID_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_MODIFY_GID_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_MODIFY_GID_RESP_EVENT_ADD_GID 0x11UL + __le16 reserved48[3]; +}; + +/* Query GID command response (16 bytes) */ +struct creq_query_gid_resp { + u8 type; + #define CREQ_QUERY_GID_RESP_TYPE_MASK 0x3fUL + #define CREQ_QUERY_GID_RESP_TYPE_SFT 0 + #define CREQ_QUERY_GID_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_QUERY_GID_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_QUERY_GID_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 size; + u8 v; + #define CREQ_QUERY_GID_RESP_V 0x1UL + #define CREQ_QUERY_GID_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_QUERY_GID_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_QUERY_GID_RESP_EVENT_QUERY_GID 0x18UL + __le16 reserved48[3]; +}; + +/* Query GID command response side buffer structure (40 bytes) */ +struct creq_query_gid_resp_sb { + u8 opcode; + #define CREQ_QUERY_GID_RESP_SB_OPCODE_QUERY_GID 0x18UL + u8 status; + __le16 cookie; + __le16 flags; + u8 resp_size; + u8 reserved8; + __le32 gid[4]; + __le16 src_mac[3]; + __le16 vlan; + #define CREQ_QUERY_GID_RESP_SB_VLAN_VLAN_ID_MASK 0xfffUL + #define CREQ_QUERY_GID_RESP_SB_VLAN_VLAN_ID_SFT 0 + #define CREQ_QUERY_GID_RESP_SB_VLAN_TPID_MASK 0x7000UL + #define CREQ_QUERY_GID_RESP_SB_VLAN_TPID_SFT 12 + #define CREQ_QUERY_GID_RESP_SB_VLAN_TPID_TPID_88A8 (0x0UL << 12) + #define CREQ_QUERY_GID_RESP_SB_VLAN_TPID_TPID_8100 (0x1UL << 12) + #define CREQ_QUERY_GID_RESP_SB_VLAN_TPID_TPID_9100 (0x2UL << 12) + #define CREQ_QUERY_GID_RESP_SB_VLAN_TPID_TPID_9200 (0x3UL << 12) + #define CREQ_QUERY_GID_RESP_SB_VLAN_TPID_TPID_9300 (0x4UL << 12) + #define CREQ_QUERY_GID_RESP_SB_VLAN_TPID_TPID_CFG1 (0x5UL << 12) + #define CREQ_QUERY_GID_RESP_SB_VLAN_TPID_TPID_CFG2 (0x6UL << 12) + #define CREQ_QUERY_GID_RESP_SB_VLAN_TPID_TPID_CFG3 (0x7UL << 12) + #define CREQ_QUERY_GID_RESP_SB_VLAN_TPID_LAST \ + CREQ_QUERY_GID_RESP_SB_VLAN_TPID_TPID_CFG3 + #define CREQ_QUERY_GID_RESP_SB_VLAN_VLAN_EN 0x8000UL + __le16 ipid; + __le16 gid_index; + __le32 unused_0; +}; + +/* Create QP1 command response (16 bytes) */ +struct creq_create_qp1_resp { + u8 type; + #define CREQ_CREATE_QP1_RESP_TYPE_MASK 0x3fUL + #define CREQ_CREATE_QP1_RESP_TYPE_SFT 0 + #define CREQ_CREATE_QP1_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_CREATE_QP1_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_CREATE_QP1_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_CREATE_QP1_RESP_V 0x1UL + #define CREQ_CREATE_QP1_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_CREATE_QP1_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_CREATE_QP1_RESP_EVENT_CREATE_QP1 0x13UL + __le16 reserved48[3]; +}; + +/* Destroy QP1 command response (16 bytes) */ +struct creq_destroy_qp1_resp { + u8 type; + #define CREQ_DESTROY_QP1_RESP_TYPE_MASK 0x3fUL + #define CREQ_DESTROY_QP1_RESP_TYPE_SFT 0 + #define CREQ_DESTROY_QP1_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_DESTROY_QP1_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_DESTROY_QP1_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_DESTROY_QP1_RESP_V 0x1UL + #define CREQ_DESTROY_QP1_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_DESTROY_QP1_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_DESTROY_QP1_RESP_EVENT_DESTROY_QP1 0x14UL + __le16 reserved48[3]; +}; + +/* Create AH command response (16 bytes) */ +struct creq_create_ah_resp { + u8 type; + #define CREQ_CREATE_AH_RESP_TYPE_MASK 0x3fUL + #define CREQ_CREATE_AH_RESP_TYPE_SFT 0 + #define CREQ_CREATE_AH_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_CREATE_AH_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_CREATE_AH_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_CREATE_AH_RESP_V 0x1UL + #define CREQ_CREATE_AH_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_CREATE_AH_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_CREATE_AH_RESP_EVENT_CREATE_AH 0x15UL + __le16 reserved48[3]; +}; + +/* Destroy AH command response (16 bytes) */ +struct creq_destroy_ah_resp { + u8 type; + #define CREQ_DESTROY_AH_RESP_TYPE_MASK 0x3fUL + #define CREQ_DESTROY_AH_RESP_TYPE_SFT 0 + #define CREQ_DESTROY_AH_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_DESTROY_AH_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_DESTROY_AH_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 xid; + u8 v; + #define CREQ_DESTROY_AH_RESP_V 0x1UL + #define CREQ_DESTROY_AH_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_DESTROY_AH_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_DESTROY_AH_RESP_EVENT_DESTROY_AH 0x16UL + __le16 reserved48[3]; +}; + +/* Initialize Firmware command response (16 bytes) */ +struct creq_initialize_fw_resp { + u8 type; + #define CREQ_INITIALIZE_FW_RESP_TYPE_MASK 0x3fUL + #define CREQ_INITIALIZE_FW_RESP_TYPE_SFT 0 + #define CREQ_INITIALIZE_FW_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_INITIALIZE_FW_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_INITIALIZE_FW_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 reserved32; + u8 v; + #define CREQ_INITIALIZE_FW_RESP_V 0x1UL + #define CREQ_INITIALIZE_FW_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_INITIALIZE_FW_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_INITIALIZE_FW_RESP_EVENT_INITIALIZE_FW 0x80UL + __le16 reserved48[3]; +}; + +/* De-initialize Firmware command response (16 bytes) */ +struct creq_deinitialize_fw_resp { + u8 type; + #define CREQ_DEINITIALIZE_FW_RESP_TYPE_MASK 0x3fUL + #define CREQ_DEINITIALIZE_FW_RESP_TYPE_SFT 0 + #define CREQ_DEINITIALIZE_FW_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_DEINITIALIZE_FW_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_DEINITIALIZE_FW_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 reserved32; + u8 v; + #define CREQ_DEINITIALIZE_FW_RESP_V 0x1UL + #define CREQ_DEINITIALIZE_FW_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_DEINITIALIZE_FW_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_DEINITIALIZE_FW_RESP_EVENT_DEINITIALIZE_FW 0x81UL + __le16 reserved48[3]; +}; + +/* Stop function command response (16 bytes) */ +struct creq_stop_func_resp { + u8 type; + #define CREQ_STOP_FUNC_RESP_TYPE_MASK 0x3fUL + #define CREQ_STOP_FUNC_RESP_TYPE_SFT 0 + #define CREQ_STOP_FUNC_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_STOP_FUNC_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_STOP_FUNC_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 reserved32; + u8 v; + #define CREQ_STOP_FUNC_RESP_V 0x1UL + #define CREQ_STOP_FUNC_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_STOP_FUNC_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_STOP_FUNC_RESP_EVENT_STOP_FUNC 0x82UL + __le16 reserved48[3]; +}; + +/* Query function command response (16 bytes) */ +struct creq_query_func_resp { + u8 type; + #define CREQ_QUERY_FUNC_RESP_TYPE_MASK 0x3fUL + #define CREQ_QUERY_FUNC_RESP_TYPE_SFT 0 + #define CREQ_QUERY_FUNC_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_QUERY_FUNC_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_QUERY_FUNC_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 size; + u8 v; + #define CREQ_QUERY_FUNC_RESP_V 0x1UL + #define CREQ_QUERY_FUNC_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_QUERY_FUNC_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_QUERY_FUNC_RESP_EVENT_QUERY_FUNC 0x83UL + __le16 reserved48[3]; +}; + +/* Query function command response side buffer structure (88 bytes) */ +struct creq_query_func_resp_sb { + u8 opcode; + #define CREQ_QUERY_FUNC_RESP_SB_OPCODE_QUERY_FUNC 0x83UL + u8 status; + __le16 cookie; + __le16 flags; + u8 resp_size; + u8 reserved8; + __le64 max_mr_size; + __le32 max_qp; + __le16 max_qp_wr; + __le16 dev_cap_flags; + #define CREQ_QUERY_FUNC_RESP_SB_DEV_CAP_FLAGS_RESIZE_QP 0x1UL + __le32 max_cq; + __le32 max_cqe; + __le32 max_pd; + u8 max_sge; + u8 max_srq_sge; + u8 max_qp_rd_atom; + u8 max_qp_init_rd_atom; + __le32 max_mr; + __le32 max_mw; + __le32 max_raw_eth_qp; + __le32 max_ah; + __le32 max_fmr; + __le32 max_srq_wr; + __le32 max_pkeys; + __le32 max_inline_data; + u8 max_map_per_fmr; + u8 l2_db_space_size; + __le16 max_srq; + __le32 max_gid; + __le32 tqm_alloc_reqs[8]; +}; + +/* Set resources command response (16 bytes) */ +struct creq_set_func_resources_resp { + u8 type; + #define CREQ_SET_FUNC_RESOURCES_RESP_TYPE_MASK 0x3fUL + #define CREQ_SET_FUNC_RESOURCES_RESP_TYPE_SFT 0 + #define CREQ_SET_FUNC_RESOURCES_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_SET_FUNC_RESOURCES_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_SET_FUNC_RESOURCES_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 reserved32; + u8 v; + #define CREQ_SET_FUNC_RESOURCES_RESP_V 0x1UL + #define CREQ_SET_FUNC_RESOURCES_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_SET_FUNC_RESOURCES_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_SET_FUNC_RESOURCES_RESP_EVENT_SET_FUNC_RESOURCES 0x84UL + __le16 reserved48[3]; +}; + +/* Map TC to COS response (16 bytes) */ +struct creq_map_tc_to_cos_resp { + u8 type; + #define CREQ_MAP_TC_TO_COS_RESP_TYPE_MASK 0x3fUL + #define CREQ_MAP_TC_TO_COS_RESP_TYPE_SFT 0 + #define CREQ_MAP_TC_TO_COS_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_MAP_TC_TO_COS_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_MAP_TC_TO_COS_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 reserved32; + u8 v; + #define CREQ_MAP_TC_TO_COS_RESP_V 0x1UL + #define CREQ_MAP_TC_TO_COS_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_MAP_TC_TO_COS_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_MAP_TC_TO_COS_RESP_EVENT_MAP_TC_TO_COS 0x8aUL + __le16 reserved48[3]; +}; + +/* Query version response (16 bytes) */ +struct creq_query_version_resp { + u8 type; + #define CREQ_QUERY_VERSION_RESP_TYPE_MASK 0x3fUL + #define CREQ_QUERY_VERSION_RESP_TYPE_SFT 0 + #define CREQ_QUERY_VERSION_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_QUERY_VERSION_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_QUERY_VERSION_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + u8 fw_maj; + u8 fw_minor; + u8 fw_bld; + u8 fw_rsvd; + u8 v; + #define CREQ_QUERY_VERSION_RESP_V 0x1UL + #define CREQ_QUERY_VERSION_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_QUERY_VERSION_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_QUERY_VERSION_RESP_EVENT_QUERY_VERSION 0x8bUL + __le16 reserved16; + u8 intf_maj; + u8 intf_minor; + u8 intf_bld; + u8 intf_rsvd; +}; + +/* Modify congestion control command response (16 bytes) */ +struct creq_modify_cc_resp { + u8 type; + #define CREQ_MODIFY_CC_RESP_TYPE_MASK 0x3fUL + #define CREQ_MODIFY_CC_RESP_TYPE_SFT 0 + #define CREQ_MODIFY_CC_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_MODIFY_CC_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_MODIFY_CC_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 reserved32; + u8 v; + #define CREQ_MODIFY_CC_RESP_V 0x1UL + #define CREQ_MODIFY_CC_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_MODIFY_CC_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_MODIFY_CC_RESP_EVENT_MODIFY_CC 0x8cUL + __le16 reserved48[3]; +}; + +/* Query congestion control command response (16 bytes) */ +struct creq_query_cc_resp { + u8 type; + #define CREQ_QUERY_CC_RESP_TYPE_MASK 0x3fUL + #define CREQ_QUERY_CC_RESP_TYPE_SFT 0 + #define CREQ_QUERY_CC_RESP_TYPE_QP_EVENT 0x38UL + #define CREQ_QUERY_CC_RESP_RESERVED2_MASK 0xc0UL + #define CREQ_QUERY_CC_RESP_RESERVED2_SFT 6 + u8 status; + __le16 cookie; + __le32 size; + u8 v; + #define CREQ_QUERY_CC_RESP_V 0x1UL + #define CREQ_QUERY_CC_RESP_RESERVED7_MASK 0xfeUL + #define CREQ_QUERY_CC_RESP_RESERVED7_SFT 1 + u8 event; + #define CREQ_QUERY_CC_RESP_EVENT_QUERY_CC 0x8dUL + __le16 reserved48[3]; +}; + +/* Query congestion control command response side buffer structure (32 bytes) */ +struct creq_query_cc_resp_sb { + u8 opcode; + #define CREQ_QUERY_CC_RESP_SB_OPCODE_QUERY_CC 0x8dUL + u8 status; + __le16 cookie; + __le16 flags; + u8 resp_size; + u8 reserved8; + u8 enable_cc; + #define CREQ_QUERY_CC_RESP_SB_ENABLE_CC 0x1UL + u8 g; + #define CREQ_QUERY_CC_RESP_SB_G_MASK 0x7UL + #define CREQ_QUERY_CC_RESP_SB_G_SFT 0 + u8 num_phases_per_state; + __le16 init_cr; + u8 unused_2; + __le16 unused_3; + u8 unused_4; + __le16 init_tr; + u8 tos_dscp_tos_ecn; + #define CREQ_QUERY_CC_RESP_SB_TOS_ECN_MASK 0x3UL + #define CREQ_QUERY_CC_RESP_SB_TOS_ECN_SFT 0 + #define CREQ_QUERY_CC_RESP_SB_TOS_DSCP_MASK 0xfcUL + #define CREQ_QUERY_CC_RESP_SB_TOS_DSCP_SFT 2 + __le64 reserved64; + __le64 reserved64_1; +}; + +/* QP error notification event (16 bytes) */ +struct creq_qp_error_notification { + u8 type; + #define CREQ_QP_ERROR_NOTIFICATION_TYPE_MASK 0x3fUL + #define CREQ_QP_ERROR_NOTIFICATION_TYPE_SFT 0 + #define CREQ_QP_ERROR_NOTIFICATION_TYPE_QP_EVENT 0x38UL + #define CREQ_QP_ERROR_NOTIFICATION_RESERVED2_MASK 0xc0UL + #define CREQ_QP_ERROR_NOTIFICATION_RESERVED2_SFT 6 + u8 status; + u8 req_slow_path_state; + u8 req_err_state_reason; + __le32 xid; + u8 v; + #define CREQ_QP_ERROR_NOTIFICATION_V 0x1UL + #define CREQ_QP_ERROR_NOTIFICATION_RESERVED7_MASK 0xfeUL + #define CREQ_QP_ERROR_NOTIFICATION_RESERVED7_SFT 1 + u8 event; + #define CREQ_QP_ERROR_NOTIFICATION_EVENT_QP_ERROR_NOTIFICATION 0xc0UL + u8 res_slow_path_state; + u8 res_err_state_reason; + __le16 sq_cons_idx; + __le16 rq_cons_idx; +}; + +/* RoCE Slowpath HSI Specification 1.6.0 */ +#define ROCE_SP_HSI_VERSION_MAJOR 1 +#define ROCE_SP_HSI_VERSION_MINOR 6 +#define ROCE_SP_HSI_VERSION_UPDATE 0 + +#define ROCE_SP_HSI_VERSION_STR "1.6.0" +/* + * Following is the signature for ROCE_SP_HSI message field that indicates not + * applicable (All F's). Need to cast it the size of the field if needed. + */ +#define ROCE_SP_HSI_NA_SIGNATURE ((__le32)(-1)) +#endif /* __BNXT_RE_HSI_H__ */ diff --git a/drivers/infiniband/hw/cxgb3/cxio_dbg.c b/drivers/infiniband/hw/cxgb3/cxio_dbg.c index 8bca6b4ec9af..445e89e5e7cf 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_dbg.c +++ b/drivers/infiniband/hw/cxgb3/cxio_dbg.c @@ -45,10 +45,9 @@ void cxio_dump_tpt(struct cxio_rdev *rdev, u32 stag) int size = 32; m = kmalloc(sizeof(*m) + size, GFP_ATOMIC); - if (!m) { - PDBG("%s couldn't allocate memory.\n", __func__); + if (!m) return; - } + m->mem_id = MEM_PMRX; m->addr = (stag>>8) * 32 + rdev->rnic_info.tpt_base; m->len = size; @@ -82,10 +81,9 @@ void cxio_dump_pbl(struct cxio_rdev *rdev, u32 pbl_addr, uint len, u8 shift) size = npages * sizeof(u64); m = kmalloc(sizeof(*m) + size, GFP_ATOMIC); - if (!m) { - PDBG("%s couldn't allocate memory.\n", __func__); + if (!m) return; - } + m->mem_id = MEM_PMRX; m->addr = pbl_addr; m->len = size; @@ -144,10 +142,9 @@ void cxio_dump_rqt(struct cxio_rdev *rdev, u32 hwtid, int nents) int rc; m = kmalloc(sizeof(*m) + size, GFP_ATOMIC); - if (!m) { - PDBG("%s couldn't allocate memory.\n", __func__); + if (!m) return; - } + m->mem_id = MEM_PMRX; m->addr = ((hwtid)<<10) + rdev->rnic_info.rqt_base; m->len = size; @@ -177,10 +174,9 @@ void cxio_dump_tcb(struct cxio_rdev *rdev, u32 hwtid) int rc; m = kmalloc(sizeof(*m) + size, GFP_ATOMIC); - if (!m) { - PDBG("%s couldn't allocate memory.\n", __func__); + if (!m) return; - } + m->mem_id = MEM_CM; m->addr = hwtid * size; m->len = size; diff --git a/drivers/infiniband/hw/cxgb3/iwch_cm.h b/drivers/infiniband/hw/cxgb3/iwch_cm.h index b9efadfffb4f..e66e75921797 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_cm.h +++ b/drivers/infiniband/hw/cxgb3/iwch_cm.h @@ -55,14 +55,14 @@ #define put_ep(ep) { \ PDBG("put_ep (via %s:%u) ep %p refcnt %d\n", __func__, __LINE__, \ - ep, atomic_read(&((ep)->kref.refcount))); \ - WARN_ON(atomic_read(&((ep)->kref.refcount)) < 1); \ + ep, kref_read(&((ep)->kref))); \ + WARN_ON(kref_read(&((ep)->kref)) < 1); \ kref_put(&((ep)->kref), __free_ep); \ } #define get_ep(ep) { \ PDBG("get_ep (via %s:%u) ep %p, refcnt %d\n", __func__, __LINE__, \ - ep, atomic_read(&((ep)->kref.refcount))); \ + ep, kref_read(&((ep)->kref))); \ kref_get(&((ep)->kref)); \ } diff --git a/drivers/infiniband/hw/cxgb3/iwch_provider.c b/drivers/infiniband/hw/cxgb3/iwch_provider.c index cba57bb53dba..318ec5267bdf 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_provider.c +++ b/drivers/infiniband/hw/cxgb3/iwch_provider.c @@ -62,7 +62,8 @@ #include "common.h" static struct ib_ah *iwch_ah_create(struct ib_pd *pd, - struct ib_ah_attr *ah_attr) + struct ib_ah_attr *ah_attr, + struct ib_udata *udata) { return ERR_PTR(-ENOSYS); } @@ -1132,18 +1133,9 @@ static int iwch_query_port(struct ib_device *ibdev, dev = to_iwch_dev(ibdev); netdev = dev->rdev.port_info.lldevs[port-1]; - memset(props, 0, sizeof(struct ib_port_attr)); + /* props being zeroed by the caller, avoid zeroing it here */ props->max_mtu = IB_MTU_4096; - if (netdev->mtu >= 4096) - props->active_mtu = IB_MTU_4096; - else if (netdev->mtu >= 2048) - props->active_mtu = IB_MTU_2048; - else if (netdev->mtu >= 1024) - props->active_mtu = IB_MTU_1024; - else if (netdev->mtu >= 512) - props->active_mtu = IB_MTU_512; - else - props->active_mtu = IB_MTU_256; + props->active_mtu = ib_mtu_int_to_enum(netdev->mtu); if (!netif_carrier_ok(netdev)) props->state = IB_PORT_DOWN; @@ -1337,13 +1329,14 @@ static int iwch_port_immutable(struct ib_device *ibdev, u8 port_num, struct ib_port_attr attr; int err; - err = iwch_query_port(ibdev, port_num, &attr); + immutable->core_cap_flags = RDMA_CORE_PORT_IWARP; + + err = ib_query_port(ibdev, port_num, &attr); if (err) return err; immutable->pkey_tbl_len = attr.pkey_tbl_len; immutable->gid_tbl_len = attr.gid_tbl_len; - immutable->core_cap_flags = RDMA_CORE_PORT_IWARP; return 0; } @@ -1400,7 +1393,7 @@ int iwch_register_device(struct iwch_dev *dev) memcpy(dev->ibdev.node_desc, IWCH_NODE_DESC, sizeof(IWCH_NODE_DESC)); dev->ibdev.phys_port_cnt = dev->rdev.port_info.nports; dev->ibdev.num_comp_vectors = 1; - dev->ibdev.dma_device = &(dev->rdev.rnic_info.pdev->dev); + dev->ibdev.dev.parent = &dev->rdev.rnic_info.pdev->dev; dev->ibdev.query_device = iwch_query_device; dev->ibdev.query_port = iwch_query_port; dev->ibdev.query_pkey = iwch_query_pkey; diff --git a/drivers/infiniband/hw/cxgb3/iwch_qp.c b/drivers/infiniband/hw/cxgb3/iwch_qp.c index d939980a708f..a9194db7f9b8 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_qp.c +++ b/drivers/infiniband/hw/cxgb3/iwch_qp.c @@ -961,7 +961,7 @@ int iwch_modify_qp(struct iwch_dev *rhp, struct iwch_qp *qhp, case IWCH_QP_STATE_RTS: switch (attrs->next_state) { case IWCH_QP_STATE_CLOSING: - BUG_ON(atomic_read(&qhp->ep->com.kref.refcount) < 2); + BUG_ON(kref_read(&qhp->ep->com.kref) < 2); qhp->attr.state = IWCH_QP_STATE_CLOSING; if (!internal) { abort=0; diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index f1510cc76d2d..03a1b0e64fc3 100644 --- a/drivers/infiniband/hw/cxgb4/cm.c +++ b/drivers/infiniband/hw/cxgb4/cm.c @@ -692,6 +692,10 @@ static int send_connect(struct c4iw_ep *ep) int ret; enum chip_type adapter_type = ep->com.dev->rdev.lldi.adapter_type; u32 isn = (prandom_u32() & ~7UL) - 1; + struct net_device *netdev; + u64 params; + + netdev = ep->com.dev->rdev.lldi.ports[0]; switch (CHELSIO_CHIP_VERSION(adapter_type)) { case CHELSIO_T4: @@ -768,6 +772,8 @@ static int send_connect(struct c4iw_ep *ep) opt2 |= T5_ISS_F; } + params = cxgb4_select_ntuple(netdev, ep->l2t); + if (ep->com.remote_addr.ss_family == AF_INET6) cxgb4_clip_get(ep->com.dev->rdev.lldi.ports[0], (const u32 *)&la6->sin6_addr.s6_addr, 1); @@ -809,18 +815,22 @@ static int send_connect(struct c4iw_ep *ep) req->opt0 = cpu_to_be64(opt0); if (is_t4(ep->com.dev->rdev.lldi.adapter_type)) { - req->params = cpu_to_be32(cxgb4_select_ntuple( - ep->com.dev->rdev.lldi.ports[0], - ep->l2t)); + req->params = cpu_to_be32(params); req->opt2 = cpu_to_be32(opt2); } else { - t5req->params = cpu_to_be64(FILTER_TUPLE_V( - cxgb4_select_ntuple( - ep->com.dev->rdev.lldi.ports[0], - ep->l2t))); - t5req->rsvd = cpu_to_be32(isn); - PDBG("%s snd_isn %u\n", __func__, t5req->rsvd); - t5req->opt2 = cpu_to_be32(opt2); + if (is_t5(ep->com.dev->rdev.lldi.adapter_type)) { + t5req->params = + cpu_to_be64(FILTER_TUPLE_V(params)); + t5req->rsvd = cpu_to_be32(isn); + PDBG("%s snd_isn %u\n", __func__, t5req->rsvd); + t5req->opt2 = cpu_to_be32(opt2); + } else { + t6req->params = + cpu_to_be64(FILTER_TUPLE_V(params)); + t6req->rsvd = cpu_to_be32(isn); + PDBG("%s snd_isn %u\n", __func__, t6req->rsvd); + t6req->opt2 = cpu_to_be32(opt2); + } } } else { switch (CHELSIO_CHIP_VERSION(adapter_type)) { @@ -859,18 +869,24 @@ static int send_connect(struct c4iw_ep *ep) req6->opt0 = cpu_to_be64(opt0); if (is_t4(ep->com.dev->rdev.lldi.adapter_type)) { - req6->params = cpu_to_be32(cxgb4_select_ntuple( - ep->com.dev->rdev.lldi.ports[0], - ep->l2t)); + req6->params = cpu_to_be32(cxgb4_select_ntuple(netdev, + ep->l2t)); req6->opt2 = cpu_to_be32(opt2); } else { - t5req6->params = cpu_to_be64(FILTER_TUPLE_V( - cxgb4_select_ntuple( - ep->com.dev->rdev.lldi.ports[0], - ep->l2t))); - t5req6->rsvd = cpu_to_be32(isn); - PDBG("%s snd_isn %u\n", __func__, t5req6->rsvd); - t5req6->opt2 = cpu_to_be32(opt2); + if (is_t5(ep->com.dev->rdev.lldi.adapter_type)) { + t5req6->params = + cpu_to_be64(FILTER_TUPLE_V(params)); + t5req6->rsvd = cpu_to_be32(isn); + PDBG("%s snd_isn %u\n", __func__, t5req6->rsvd); + t5req6->opt2 = cpu_to_be32(opt2); + } else { + t6req6->params = + cpu_to_be64(FILTER_TUPLE_V(params)); + t6req6->rsvd = cpu_to_be32(isn); + PDBG("%s snd_isn %u\n", __func__, t6req6->rsvd); + t6req6->opt2 = cpu_to_be32(opt2); + } + } } @@ -1804,20 +1820,21 @@ static int rx_data(struct c4iw_dev *dev, struct sk_buff *skb) skb_trim(skb, dlen); mutex_lock(&ep->com.mutex); - /* update RX credits */ - update_rx_credits(ep, dlen); - switch (ep->com.state) { case MPA_REQ_SENT: + update_rx_credits(ep, dlen); ep->rcv_seq += dlen; disconnect = process_mpa_reply(ep, skb); break; case MPA_REQ_WAIT: + update_rx_credits(ep, dlen); ep->rcv_seq += dlen; disconnect = process_mpa_request(ep, skb); break; case FPDU_MODE: { struct c4iw_qp_attributes attrs; + + update_rx_credits(ep, dlen); BUG_ON(!ep->com.qp); if (status) pr_err("%s Unexpected streaming data." \ @@ -2516,18 +2533,18 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb) struct sockaddr_in *sin = (struct sockaddr_in *) &child_ep->com.local_addr; - sin->sin_family = PF_INET; + sin->sin_family = AF_INET; sin->sin_port = local_port; sin->sin_addr.s_addr = *(__be32 *)local_ip; sin = (struct sockaddr_in *)&child_ep->com.local_addr; - sin->sin_family = PF_INET; + sin->sin_family = AF_INET; sin->sin_port = ((struct sockaddr_in *) &parent_ep->com.local_addr)->sin_port; sin->sin_addr.s_addr = *(__be32 *)local_ip; sin = (struct sockaddr_in *)&child_ep->com.remote_addr; - sin->sin_family = PF_INET; + sin->sin_family = AF_INET; sin->sin_port = peer_port; sin->sin_addr.s_addr = *(__be32 *)peer_ip; } else { diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c index 19c6477af19f..bec82a600d77 100644 --- a/drivers/infiniband/hw/cxgb4/cq.c +++ b/drivers/infiniband/hw/cxgb4/cq.c @@ -505,6 +505,15 @@ static int poll_cq(struct t4_wq *wq, struct t4_cq *cq, struct t4_cqe *cqe, } /* + * Special cqe for drain WR completions... + */ + if (CQE_OPCODE(hw_cqe) == C4IW_DRAIN_OPCODE) { + *cookie = CQE_DRAIN_COOKIE(hw_cqe); + *cqe = *hw_cqe; + goto skip_cqe; + } + + /* * Gotta tweak READ completions: * 1) the cqe doesn't contain the sq_wptr from the wr. * 2) opcode not reflected from the wr. @@ -753,6 +762,9 @@ static int c4iw_poll_cq_one(struct c4iw_cq *chp, struct ib_wc *wc) c4iw_invalidate_mr(qhp->rhp, CQE_WRID_FR_STAG(&cqe)); break; + case C4IW_DRAIN_OPCODE: + wc->opcode = IB_WC_SEND; + break; default: printk(KERN_ERR MOD "Unexpected opcode %d " "in the CQE received for QPID=0x%0x\n", @@ -817,15 +829,8 @@ static int c4iw_poll_cq_one(struct c4iw_cq *chp, struct ib_wc *wc) } } out: - if (wq) { - if (unlikely(qhp->attr.state != C4IW_QP_STATE_RTS)) { - if (t4_sq_empty(wq)) - complete(&qhp->sq_drained); - if (t4_rq_empty(wq)) - complete(&qhp->rq_drained); - } + if (wq) spin_unlock(&qhp->lock); - } return ret; } diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c index 93e3d270a98a..4e4f1a732b01 100644 --- a/drivers/infiniband/hw/cxgb4/device.c +++ b/drivers/infiniband/hw/cxgb4/device.c @@ -214,6 +214,52 @@ static const struct file_operations wr_log_debugfs_fops = { .write = wr_log_clear, }; +static struct sockaddr_in zero_sin = { + .sin_family = AF_INET, +}; + +static struct sockaddr_in6 zero_sin6 = { + .sin6_family = AF_INET6, +}; + +static void set_ep_sin_addrs(struct c4iw_ep *ep, + struct sockaddr_in **lsin, + struct sockaddr_in **rsin, + struct sockaddr_in **m_lsin, + struct sockaddr_in **m_rsin) +{ + struct iw_cm_id *id = ep->com.cm_id; + + *lsin = (struct sockaddr_in *)&ep->com.local_addr; + *rsin = (struct sockaddr_in *)&ep->com.remote_addr; + if (id) { + *m_lsin = (struct sockaddr_in *)&id->m_local_addr; + *m_rsin = (struct sockaddr_in *)&id->m_remote_addr; + } else { + *m_lsin = &zero_sin; + *m_rsin = &zero_sin; + } +} + +static void set_ep_sin6_addrs(struct c4iw_ep *ep, + struct sockaddr_in6 **lsin6, + struct sockaddr_in6 **rsin6, + struct sockaddr_in6 **m_lsin6, + struct sockaddr_in6 **m_rsin6) +{ + struct iw_cm_id *id = ep->com.cm_id; + + *lsin6 = (struct sockaddr_in6 *)&ep->com.local_addr; + *rsin6 = (struct sockaddr_in6 *)&ep->com.remote_addr; + if (id) { + *m_lsin6 = (struct sockaddr_in6 *)&id->m_local_addr; + *m_rsin6 = (struct sockaddr_in6 *)&id->m_remote_addr; + } else { + *m_lsin6 = &zero_sin6; + *m_rsin6 = &zero_sin6; + } +} + static int dump_qp(int id, void *p, void *data) { struct c4iw_qp *qp = p; @@ -229,16 +275,15 @@ static int dump_qp(int id, void *p, void *data) return 1; if (qp->ep) { - if (qp->ep->com.local_addr.ss_family == AF_INET) { - struct sockaddr_in *lsin = (struct sockaddr_in *) - &qp->ep->com.cm_id->local_addr; - struct sockaddr_in *rsin = (struct sockaddr_in *) - &qp->ep->com.cm_id->remote_addr; - struct sockaddr_in *mapped_lsin = (struct sockaddr_in *) - &qp->ep->com.cm_id->m_local_addr; - struct sockaddr_in *mapped_rsin = (struct sockaddr_in *) - &qp->ep->com.cm_id->m_remote_addr; + struct c4iw_ep *ep = qp->ep; + + if (ep->com.local_addr.ss_family == AF_INET) { + struct sockaddr_in *lsin; + struct sockaddr_in *rsin; + struct sockaddr_in *m_lsin; + struct sockaddr_in *m_rsin; + set_ep_sin_addrs(ep, &lsin, &rsin, &m_lsin, &m_rsin); cc = snprintf(qpd->buf + qpd->pos, space, "rc qp sq id %u rq id %u state %u " "onchip %u ep tid %u state %u " @@ -246,23 +291,19 @@ static int dump_qp(int id, void *p, void *data) qp->wq.sq.qid, qp->wq.rq.qid, (int)qp->attr.state, qp->wq.sq.flags & T4_SQ_ONCHIP, - qp->ep->hwtid, (int)qp->ep->com.state, + ep->hwtid, (int)ep->com.state, &lsin->sin_addr, ntohs(lsin->sin_port), - ntohs(mapped_lsin->sin_port), + ntohs(m_lsin->sin_port), &rsin->sin_addr, ntohs(rsin->sin_port), - ntohs(mapped_rsin->sin_port)); + ntohs(m_rsin->sin_port)); } else { - struct sockaddr_in6 *lsin6 = (struct sockaddr_in6 *) - &qp->ep->com.cm_id->local_addr; - struct sockaddr_in6 *rsin6 = (struct sockaddr_in6 *) - &qp->ep->com.cm_id->remote_addr; - struct sockaddr_in6 *mapped_lsin6 = - (struct sockaddr_in6 *) - &qp->ep->com.cm_id->m_local_addr; - struct sockaddr_in6 *mapped_rsin6 = - (struct sockaddr_in6 *) - &qp->ep->com.cm_id->m_remote_addr; + struct sockaddr_in6 *lsin6; + struct sockaddr_in6 *rsin6; + struct sockaddr_in6 *m_lsin6; + struct sockaddr_in6 *m_rsin6; + set_ep_sin6_addrs(ep, &lsin6, &rsin6, &m_lsin6, + &m_rsin6); cc = snprintf(qpd->buf + qpd->pos, space, "rc qp sq id %u rq id %u state %u " "onchip %u ep tid %u state %u " @@ -270,13 +311,13 @@ static int dump_qp(int id, void *p, void *data) qp->wq.sq.qid, qp->wq.rq.qid, (int)qp->attr.state, qp->wq.sq.flags & T4_SQ_ONCHIP, - qp->ep->hwtid, (int)qp->ep->com.state, + ep->hwtid, (int)ep->com.state, &lsin6->sin6_addr, ntohs(lsin6->sin6_port), - ntohs(mapped_lsin6->sin6_port), + ntohs(m_lsin6->sin6_port), &rsin6->sin6_addr, ntohs(rsin6->sin6_port), - ntohs(mapped_rsin6->sin6_port)); + ntohs(m_rsin6->sin6_port)); } } else cc = snprintf(qpd->buf + qpd->pos, space, @@ -533,15 +574,12 @@ static int dump_ep(int id, void *p, void *data) return 1; if (ep->com.local_addr.ss_family == AF_INET) { - struct sockaddr_in *lsin = (struct sockaddr_in *) - &ep->com.cm_id->local_addr; - struct sockaddr_in *rsin = (struct sockaddr_in *) - &ep->com.cm_id->remote_addr; - struct sockaddr_in *mapped_lsin = (struct sockaddr_in *) - &ep->com.cm_id->m_local_addr; - struct sockaddr_in *mapped_rsin = (struct sockaddr_in *) - &ep->com.cm_id->m_remote_addr; + struct sockaddr_in *lsin; + struct sockaddr_in *rsin; + struct sockaddr_in *m_lsin; + struct sockaddr_in *m_rsin; + set_ep_sin_addrs(ep, &lsin, &rsin, &m_lsin, &m_rsin); cc = snprintf(epd->buf + epd->pos, space, "ep %p cm_id %p qp %p state %d flags 0x%lx " "history 0x%lx hwtid %d atid %d " @@ -553,19 +591,16 @@ static int dump_ep(int id, void *p, void *data) ep->stats.connect_neg_adv, ep->stats.abort_neg_adv, &lsin->sin_addr, ntohs(lsin->sin_port), - ntohs(mapped_lsin->sin_port), + ntohs(m_lsin->sin_port), &rsin->sin_addr, ntohs(rsin->sin_port), - ntohs(mapped_rsin->sin_port)); + ntohs(m_rsin->sin_port)); } else { - struct sockaddr_in6 *lsin6 = (struct sockaddr_in6 *) - &ep->com.cm_id->local_addr; - struct sockaddr_in6 *rsin6 = (struct sockaddr_in6 *) - &ep->com.cm_id->remote_addr; - struct sockaddr_in6 *mapped_lsin6 = (struct sockaddr_in6 *) - &ep->com.cm_id->m_local_addr; - struct sockaddr_in6 *mapped_rsin6 = (struct sockaddr_in6 *) - &ep->com.cm_id->m_remote_addr; + struct sockaddr_in6 *lsin6; + struct sockaddr_in6 *rsin6; + struct sockaddr_in6 *m_lsin6; + struct sockaddr_in6 *m_rsin6; + set_ep_sin6_addrs(ep, &lsin6, &rsin6, &m_lsin6, &m_rsin6); cc = snprintf(epd->buf + epd->pos, space, "ep %p cm_id %p qp %p state %d flags 0x%lx " "history 0x%lx hwtid %d atid %d " @@ -577,9 +612,9 @@ static int dump_ep(int id, void *p, void *data) ep->stats.connect_neg_adv, ep->stats.abort_neg_adv, &lsin6->sin6_addr, ntohs(lsin6->sin6_port), - ntohs(mapped_lsin6->sin6_port), + ntohs(m_lsin6->sin6_port), &rsin6->sin6_addr, ntohs(rsin6->sin6_port), - ntohs(mapped_rsin6->sin6_port)); + ntohs(m_rsin6->sin6_port)); } if (cc < space) epd->pos += cc; @@ -600,7 +635,7 @@ static int dump_listen_ep(int id, void *p, void *data) if (ep->com.local_addr.ss_family == AF_INET) { struct sockaddr_in *lsin = (struct sockaddr_in *) &ep->com.cm_id->local_addr; - struct sockaddr_in *mapped_lsin = (struct sockaddr_in *) + struct sockaddr_in *m_lsin = (struct sockaddr_in *) &ep->com.cm_id->m_local_addr; cc = snprintf(epd->buf + epd->pos, space, @@ -609,11 +644,11 @@ static int dump_listen_ep(int id, void *p, void *data) ep, ep->com.cm_id, (int)ep->com.state, ep->com.flags, ep->stid, ep->backlog, &lsin->sin_addr, ntohs(lsin->sin_port), - ntohs(mapped_lsin->sin_port)); + ntohs(m_lsin->sin_port)); } else { struct sockaddr_in6 *lsin6 = (struct sockaddr_in6 *) &ep->com.cm_id->local_addr; - struct sockaddr_in6 *mapped_lsin6 = (struct sockaddr_in6 *) + struct sockaddr_in6 *m_lsin6 = (struct sockaddr_in6 *) &ep->com.cm_id->m_local_addr; cc = snprintf(epd->buf + epd->pos, space, @@ -622,7 +657,7 @@ static int dump_listen_ep(int id, void *p, void *data) ep, ep->com.cm_id, (int)ep->com.state, ep->com.flags, ep->stid, ep->backlog, &lsin6->sin6_addr, ntohs(lsin6->sin6_port), - ntohs(mapped_lsin6->sin6_port)); + ntohs(m_lsin6->sin6_port)); } if (cc < space) epd->pos += cc; @@ -828,8 +863,10 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev) } rdev->status_page = (struct t4_dev_status_page *) __get_free_page(GFP_KERNEL); - if (!rdev->status_page) + if (!rdev->status_page) { + err = -ENOMEM; goto destroy_ocqp_pool; + } rdev->status_page->qp_start = rdev->lldi.vr->qp.start; rdev->status_page->qp_size = rdev->lldi.vr->qp.size; rdev->status_page->cq_start = rdev->lldi.vr->cq.start; @@ -841,14 +878,20 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev) if (rdev->wr_log) { rdev->wr_log_size = 1 << c4iw_wr_log_size_order; atomic_set(&rdev->wr_log_idx, 0); - } else { - pr_err(MOD "error allocating wr_log. Logging disabled\n"); } } + rdev->free_workq = create_singlethread_workqueue("iw_cxgb4_free"); + if (!rdev->free_workq) { + err = -ENOMEM; + goto err_free_status_page; + } + rdev->status_page->db_off = 0; return 0; +err_free_status_page: + free_page((unsigned long)rdev->status_page); destroy_ocqp_pool: c4iw_ocqp_pool_destroy(rdev); destroy_rqtpool: @@ -862,6 +905,7 @@ destroy_resource: static void c4iw_rdev_close(struct c4iw_rdev *rdev) { + destroy_workqueue(rdev->free_workq); kfree(rdev->wr_log); free_page((unsigned long)rdev->status_page); c4iw_pblpool_destroy(rdev); @@ -1424,8 +1468,6 @@ static void recover_queues(struct uld_ctx *ctx) qp_list.qps = kzalloc(count * sizeof *qp_list.qps, GFP_ATOMIC); if (!qp_list.qps) { - printk(KERN_ERR MOD "%s: Fatal error - DB overflow recovery failed\n", - pci_name(ctx->lldi.pdev)); spin_unlock_irq(&ctx->dev->lock); return; } @@ -1481,6 +1523,7 @@ static int c4iw_uld_control(void *handle, enum cxgb4_control control, ...) static struct cxgb4_uld_info c4iw_uld_info = { .name = DRV_NAME, .nrxq = MAX_ULD_QSETS, + .ntxq = MAX_ULD_QSETS, .rxq_size = 511, .ciq = true, .lro = false, diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h index 4788e1a46fde..d19662f635b1 100644 --- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h +++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h @@ -45,6 +45,7 @@ #include <linux/kref.h> #include <linux/timer.h> #include <linux/io.h> +#include <linux/workqueue.h> #include <asm/byteorder.h> @@ -107,6 +108,7 @@ struct c4iw_dev_ucontext { struct list_head qpids; struct list_head cqids; struct mutex lock; + struct kref kref; }; enum c4iw_rdev_flags { @@ -183,6 +185,7 @@ struct c4iw_rdev { atomic_t wr_log_idx; struct wr_log_entry *wr_log; int wr_log_size; + struct workqueue_struct *free_workq; }; static inline int c4iw_fatal_error(struct c4iw_rdev *rdev) @@ -480,8 +483,8 @@ struct c4iw_qp { wait_queue_head_t wait; struct timer_list timer; int sq_sig_all; - struct completion rq_drained; - struct completion sq_drained; + struct work_struct free_work; + struct c4iw_ucontext *ucontext; }; static inline struct c4iw_qp *to_c4iw_qp(struct ib_qp *ibqp) @@ -495,6 +498,7 @@ struct c4iw_ucontext { u32 key; spinlock_t mmap_lock; struct list_head mmaps; + struct kref kref; }; static inline struct c4iw_ucontext *to_c4iw_ucontext(struct ib_ucontext *c) @@ -502,6 +506,18 @@ static inline struct c4iw_ucontext *to_c4iw_ucontext(struct ib_ucontext *c) return container_of(c, struct c4iw_ucontext, ibucontext); } +void _c4iw_free_ucontext(struct kref *kref); + +static inline void c4iw_put_ucontext(struct c4iw_ucontext *ucontext) +{ + kref_put(&ucontext->kref, _c4iw_free_ucontext); +} + +static inline void c4iw_get_ucontext(struct c4iw_ucontext *ucontext) +{ + kref_get(&ucontext->kref); +} + struct c4iw_mm_entry { struct list_head entry; u64 addr; @@ -615,6 +631,8 @@ static inline int to_ib_qp_state(int c4iw_qp_state) return IB_QPS_ERR; } +#define C4IW_DRAIN_OPCODE FW_RI_SGE_EC_CR_RETURN + static inline u32 c4iw_ib_to_tpt_access(int a) { return (a & IB_ACCESS_REMOTE_WRITE ? FW_RI_MEM_ACCESS_REM_WRITE : 0) | @@ -654,14 +672,14 @@ enum c4iw_mmid_state { #define c4iw_put_ep(ep) { \ PDBG("put_ep (via %s:%u) ep %p refcnt %d\n", __func__, __LINE__, \ - ep, atomic_read(&((ep)->kref.refcount))); \ - WARN_ON(atomic_read(&((ep)->kref.refcount)) < 1); \ + ep, kref_read(&((ep)->kref))); \ + WARN_ON(kref_read(&((ep)->kref)) < 1); \ kref_put(&((ep)->kref), _c4iw_free_ep); \ } #define c4iw_get_ep(ep) { \ PDBG("get_ep (via %s:%u) ep %p, refcnt %d\n", __func__, __LINE__, \ - ep, atomic_read(&((ep)->kref.refcount))); \ + ep, kref_read(&((ep)->kref))); \ kref_get(&((ep)->kref)); \ } void _c4iw_free_ep(struct kref *kref); @@ -997,8 +1015,6 @@ extern int c4iw_wr_log; extern int db_fc_threshold; extern int db_coalescing_threshold; extern int use_dsgl; -void c4iw_drain_rq(struct ib_qp *qp); -void c4iw_drain_sq(struct ib_qp *qp); void c4iw_invalidate_mr(struct c4iw_dev *rhp, u32 rkey); #endif diff --git a/drivers/infiniband/hw/cxgb4/provider.c b/drivers/infiniband/hw/cxgb4/provider.c index 645e606a17c5..df64417ab6f2 100644 --- a/drivers/infiniband/hw/cxgb4/provider.c +++ b/drivers/infiniband/hw/cxgb4/provider.c @@ -59,7 +59,9 @@ module_param(fastreg_support, int, 0644); MODULE_PARM_DESC(fastreg_support, "Advertise fastreg support (default=1)"); static struct ib_ah *c4iw_ah_create(struct ib_pd *pd, - struct ib_ah_attr *ah_attr) + struct ib_ah_attr *ah_attr, + struct ib_udata *udata) + { return ERR_PTR(-ENOSYS); } @@ -91,17 +93,28 @@ static int c4iw_process_mad(struct ib_device *ibdev, int mad_flags, return -ENOSYS; } -static int c4iw_dealloc_ucontext(struct ib_ucontext *context) +void _c4iw_free_ucontext(struct kref *kref) { - struct c4iw_dev *rhp = to_c4iw_dev(context->device); - struct c4iw_ucontext *ucontext = to_c4iw_ucontext(context); + struct c4iw_ucontext *ucontext; + struct c4iw_dev *rhp; struct c4iw_mm_entry *mm, *tmp; - PDBG("%s context %p\n", __func__, context); + ucontext = container_of(kref, struct c4iw_ucontext, kref); + rhp = to_c4iw_dev(ucontext->ibucontext.device); + + PDBG("%s ucontext %p\n", __func__, ucontext); list_for_each_entry_safe(mm, tmp, &ucontext->mmaps, entry) kfree(mm); c4iw_release_dev_ucontext(&rhp->rdev, &ucontext->uctx); kfree(ucontext); +} + +static int c4iw_dealloc_ucontext(struct ib_ucontext *context) +{ + struct c4iw_ucontext *ucontext = to_c4iw_ucontext(context); + + PDBG("%s context %p\n", __func__, context); + c4iw_put_ucontext(ucontext); return 0; } @@ -125,6 +138,7 @@ static struct ib_ucontext *c4iw_alloc_ucontext(struct ib_device *ibdev, c4iw_init_dev_ucontext(&rhp->rdev, &context->uctx); INIT_LIST_HEAD(&context->mmaps); spin_lock_init(&context->mmap_lock); + kref_init(&context->kref); if (udata->outlen < sizeof(uresp) - sizeof(uresp.reserved)) { if (!warned++) @@ -356,19 +370,9 @@ static int c4iw_query_port(struct ib_device *ibdev, u8 port, dev = to_c4iw_dev(ibdev); netdev = dev->rdev.lldi.ports[port-1]; - - memset(props, 0, sizeof(struct ib_port_attr)); + /* props being zeroed by the caller, avoid zeroing it here */ props->max_mtu = IB_MTU_4096; - if (netdev->mtu >= 4096) - props->active_mtu = IB_MTU_4096; - else if (netdev->mtu >= 2048) - props->active_mtu = IB_MTU_2048; - else if (netdev->mtu >= 1024) - props->active_mtu = IB_MTU_1024; - else if (netdev->mtu >= 512) - props->active_mtu = IB_MTU_512; - else - props->active_mtu = IB_MTU_256; + props->active_mtu = ib_mtu_int_to_enum(netdev->mtu); if (!netif_carrier_ok(netdev)) props->state = IB_PORT_DOWN; @@ -503,13 +507,14 @@ static int c4iw_port_immutable(struct ib_device *ibdev, u8 port_num, struct ib_port_attr attr; int err; - err = c4iw_query_port(ibdev, port_num, &attr); + immutable->core_cap_flags = RDMA_CORE_PORT_IWARP; + + err = ib_query_port(ibdev, port_num, &attr); if (err) return err; immutable->pkey_tbl_len = attr.pkey_tbl_len; immutable->gid_tbl_len = attr.gid_tbl_len; - immutable->core_cap_flags = RDMA_CORE_PORT_IWARP; return 0; } @@ -567,7 +572,7 @@ int c4iw_register_device(struct c4iw_dev *dev) memcpy(dev->ibdev.node_desc, C4IW_NODE_DESC, sizeof(C4IW_NODE_DESC)); dev->ibdev.phys_port_cnt = dev->rdev.lldi.nports; dev->ibdev.num_comp_vectors = dev->rdev.lldi.nciq; - dev->ibdev.dma_device = &(dev->rdev.lldi.pdev->dev); + dev->ibdev.dev.parent = &dev->rdev.lldi.pdev->dev; dev->ibdev.query_device = c4iw_query_device; dev->ibdev.query_port = c4iw_query_port; dev->ibdev.query_pkey = c4iw_query_pkey; @@ -605,8 +610,6 @@ int c4iw_register_device(struct c4iw_dev *dev) dev->ibdev.uverbs_abi_ver = C4IW_UVERBS_ABI_VERSION; dev->ibdev.get_port_immutable = c4iw_port_immutable; dev->ibdev.get_dev_fw_str = get_dev_fw_str; - dev->ibdev.drain_sq = c4iw_drain_sq; - dev->ibdev.drain_rq = c4iw_drain_rq; dev->ibdev.iwcm = kmalloc(sizeof(struct iw_cm_verbs), GFP_KERNEL); if (!dev->ibdev.iwcm) diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c index b7ac97b27c88..d4fd2f5c8326 100644 --- a/drivers/infiniband/hw/cxgb4/qp.c +++ b/drivers/infiniband/hw/cxgb4/qp.c @@ -321,7 +321,8 @@ static int create_qp(struct c4iw_rdev *rdev, struct t4_wq *wq, FW_RI_RES_WR_DCAEN_V(0) | FW_RI_RES_WR_DCACPU_V(0) | FW_RI_RES_WR_FBMIN_V(2) | - FW_RI_RES_WR_FBMAX_V(2) | + (t4_sq_onchip(&wq->sq) ? FW_RI_RES_WR_FBMAX_V(2) : + FW_RI_RES_WR_FBMAX_V(3)) | FW_RI_RES_WR_CIDXFTHRESHO_V(0) | FW_RI_RES_WR_CIDXFTHRESH_V(0) | FW_RI_RES_WR_EQSIZE_V(eqsize)); @@ -345,7 +346,7 @@ static int create_qp(struct c4iw_rdev *rdev, struct t4_wq *wq, FW_RI_RES_WR_DCAEN_V(0) | FW_RI_RES_WR_DCACPU_V(0) | FW_RI_RES_WR_FBMIN_V(2) | - FW_RI_RES_WR_FBMAX_V(2) | + FW_RI_RES_WR_FBMAX_V(3) | FW_RI_RES_WR_CIDXFTHRESHO_V(0) | FW_RI_RES_WR_CIDXFTHRESH_V(0) | FW_RI_RES_WR_EQSIZE_V(eqsize)); @@ -714,13 +715,32 @@ static int build_inv_stag(union t4_wr *wqe, struct ib_send_wr *wr, u8 *len16) return 0; } -static void _free_qp(struct kref *kref) +static void free_qp_work(struct work_struct *work) +{ + struct c4iw_ucontext *ucontext; + struct c4iw_qp *qhp; + struct c4iw_dev *rhp; + + qhp = container_of(work, struct c4iw_qp, free_work); + ucontext = qhp->ucontext; + rhp = qhp->rhp; + + PDBG("%s qhp %p ucontext %p\n", __func__, qhp, ucontext); + destroy_qp(&rhp->rdev, &qhp->wq, + ucontext ? &ucontext->uctx : &rhp->rdev.uctx); + + if (ucontext) + c4iw_put_ucontext(ucontext); + kfree(qhp); +} + +static void queue_qp_free(struct kref *kref) { struct c4iw_qp *qhp; qhp = container_of(kref, struct c4iw_qp, kref); PDBG("%s qhp %p\n", __func__, qhp); - kfree(qhp); + queue_work(qhp->rhp->rdev.free_workq, &qhp->free_work); } void c4iw_qp_add_ref(struct ib_qp *qp) @@ -732,7 +752,7 @@ void c4iw_qp_add_ref(struct ib_qp *qp) void c4iw_qp_rem_ref(struct ib_qp *qp) { PDBG("%s ib_qp %p\n", __func__, qp); - kref_put(&to_c4iw_qp(qp)->kref, _free_qp); + kref_put(&to_c4iw_qp(qp)->kref, queue_qp_free); } static void add_to_fc_list(struct list_head *head, struct list_head *entry) @@ -775,6 +795,64 @@ static int ring_kernel_rq_db(struct c4iw_qp *qhp, u16 inc) return 0; } +static void complete_sq_drain_wr(struct c4iw_qp *qhp, struct ib_send_wr *wr) +{ + struct t4_cqe cqe = {}; + struct c4iw_cq *schp; + unsigned long flag; + struct t4_cq *cq; + + schp = to_c4iw_cq(qhp->ibqp.send_cq); + cq = &schp->cq; + + cqe.u.drain_cookie = wr->wr_id; + cqe.header = cpu_to_be32(CQE_STATUS_V(T4_ERR_SWFLUSH) | + CQE_OPCODE_V(C4IW_DRAIN_OPCODE) | + CQE_TYPE_V(1) | + CQE_SWCQE_V(1) | + CQE_QPID_V(qhp->wq.sq.qid)); + + spin_lock_irqsave(&schp->lock, flag); + cqe.bits_type_ts = cpu_to_be64(CQE_GENBIT_V((u64)cq->gen)); + cq->sw_queue[cq->sw_pidx] = cqe; + t4_swcq_produce(cq); + spin_unlock_irqrestore(&schp->lock, flag); + + spin_lock_irqsave(&schp->comp_handler_lock, flag); + (*schp->ibcq.comp_handler)(&schp->ibcq, + schp->ibcq.cq_context); + spin_unlock_irqrestore(&schp->comp_handler_lock, flag); +} + +static void complete_rq_drain_wr(struct c4iw_qp *qhp, struct ib_recv_wr *wr) +{ + struct t4_cqe cqe = {}; + struct c4iw_cq *rchp; + unsigned long flag; + struct t4_cq *cq; + + rchp = to_c4iw_cq(qhp->ibqp.recv_cq); + cq = &rchp->cq; + + cqe.u.drain_cookie = wr->wr_id; + cqe.header = cpu_to_be32(CQE_STATUS_V(T4_ERR_SWFLUSH) | + CQE_OPCODE_V(C4IW_DRAIN_OPCODE) | + CQE_TYPE_V(0) | + CQE_SWCQE_V(1) | + CQE_QPID_V(qhp->wq.sq.qid)); + + spin_lock_irqsave(&rchp->lock, flag); + cqe.bits_type_ts = cpu_to_be64(CQE_GENBIT_V((u64)cq->gen)); + cq->sw_queue[cq->sw_pidx] = cqe; + t4_swcq_produce(cq); + spin_unlock_irqrestore(&rchp->lock, flag); + + spin_lock_irqsave(&rchp->comp_handler_lock, flag); + (*rchp->ibcq.comp_handler)(&rchp->ibcq, + rchp->ibcq.cq_context); + spin_unlock_irqrestore(&rchp->comp_handler_lock, flag); +} + int c4iw_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, struct ib_send_wr **bad_wr) { @@ -793,8 +871,8 @@ int c4iw_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, spin_lock_irqsave(&qhp->lock, flag); if (t4_wq_in_error(&qhp->wq)) { spin_unlock_irqrestore(&qhp->lock, flag); - *bad_wr = wr; - return -EINVAL; + complete_sq_drain_wr(qhp, wr); + return err; } num_wrs = t4_sq_avail(&qhp->wq); if (num_wrs == 0) { @@ -936,8 +1014,8 @@ int c4iw_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr, spin_lock_irqsave(&qhp->lock, flag); if (t4_wq_in_error(&qhp->wq)) { spin_unlock_irqrestore(&qhp->lock, flag); - *bad_wr = wr; - return -EINVAL; + complete_rq_drain_wr(qhp, wr); + return err; } num_wrs = t4_rq_avail(&qhp->wq); if (num_wrs == 0) { @@ -1502,7 +1580,7 @@ int c4iw_modify_qp(struct c4iw_dev *rhp, struct c4iw_qp *qhp, case C4IW_QP_STATE_RTS: switch (attrs->next_state) { case C4IW_QP_STATE_CLOSING: - BUG_ON(atomic_read(&qhp->ep->com.kref.refcount) < 2); + BUG_ON(kref_read(&qhp->ep->com.kref) < 2); t4_set_wq_in_error(&qhp->wq); set_state(qhp, C4IW_QP_STATE_CLOSING); ep = qhp->ep; @@ -1549,7 +1627,12 @@ int c4iw_modify_qp(struct c4iw_dev *rhp, struct c4iw_qp *qhp, } break; case C4IW_QP_STATE_CLOSING: - if (!internal) { + + /* + * Allow kernel users to move to ERROR for qp draining. + */ + if (!internal && (qhp->ibqp.uobject || attrs->next_state != + C4IW_QP_STATE_ERROR)) { ret = -EINVAL; goto out; } @@ -1642,7 +1725,6 @@ int c4iw_destroy_qp(struct ib_qp *ib_qp) struct c4iw_dev *rhp; struct c4iw_qp *qhp; struct c4iw_qp_attributes attrs; - struct c4iw_ucontext *ucontext; qhp = to_c4iw_qp(ib_qp); rhp = qhp->rhp; @@ -1662,11 +1744,6 @@ int c4iw_destroy_qp(struct ib_qp *ib_qp) spin_unlock_irq(&rhp->lock); free_ird(rhp, qhp->attr.max_ird); - ucontext = ib_qp->uobject ? - to_c4iw_ucontext(ib_qp->uobject->context) : NULL; - destroy_qp(&rhp->rdev, &qhp->wq, - ucontext ? &ucontext->uctx : &rhp->rdev.uctx); - c4iw_qp_rem_ref(ib_qp); PDBG("%s ib_qp %p qpid 0x%0x\n", __func__, ib_qp, qhp->wq.sq.qid); @@ -1762,11 +1839,10 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs, qhp->attr.max_ird = 0; qhp->sq_sig_all = attrs->sq_sig_type == IB_SIGNAL_ALL_WR; spin_lock_init(&qhp->lock); - init_completion(&qhp->sq_drained); - init_completion(&qhp->rq_drained); mutex_init(&qhp->mutex); init_waitqueue_head(&qhp->wait); kref_init(&qhp->kref); + INIT_WORK(&qhp->free_work, free_qp_work); ret = insert_handle(rhp, &rhp->qpidr, qhp, qhp->wq.sq.qid); if (ret) @@ -1853,6 +1929,9 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs, ma_sync_key_mm->len = PAGE_SIZE; insert_mmap(ucontext, ma_sync_key_mm); } + + c4iw_get_ucontext(ucontext); + qhp->ucontext = ucontext; } qhp->ibqp.qp_num = qhp->wq.sq.qid; init_timer(&(qhp->timer)); @@ -1957,40 +2036,3 @@ int c4iw_ib_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, init_attr->sq_sig_type = qhp->sq_sig_all ? IB_SIGNAL_ALL_WR : 0; return 0; } - -static void move_qp_to_err(struct c4iw_qp *qp) -{ - struct c4iw_qp_attributes attrs = { .next_state = C4IW_QP_STATE_ERROR }; - - (void)c4iw_modify_qp(qp->rhp, qp, C4IW_QP_ATTR_NEXT_STATE, &attrs, 1); -} - -void c4iw_drain_sq(struct ib_qp *ibqp) -{ - struct c4iw_qp *qp = to_c4iw_qp(ibqp); - unsigned long flag; - bool need_to_wait; - - move_qp_to_err(qp); - spin_lock_irqsave(&qp->lock, flag); - need_to_wait = !t4_sq_empty(&qp->wq); - spin_unlock_irqrestore(&qp->lock, flag); - - if (need_to_wait) - wait_for_completion(&qp->sq_drained); -} - -void c4iw_drain_rq(struct ib_qp *ibqp) -{ - struct c4iw_qp *qp = to_c4iw_qp(ibqp); - unsigned long flag; - bool need_to_wait; - - move_qp_to_err(qp); - spin_lock_irqsave(&qp->lock, flag); - need_to_wait = !t4_rq_empty(&qp->wq); - spin_unlock_irqrestore(&qp->lock, flag); - - if (need_to_wait) - wait_for_completion(&qp->rq_drained); -} diff --git a/drivers/infiniband/hw/cxgb4/t4.h b/drivers/infiniband/hw/cxgb4/t4.h index 862381aa83c8..640d22148a3e 100644 --- a/drivers/infiniband/hw/cxgb4/t4.h +++ b/drivers/infiniband/hw/cxgb4/t4.h @@ -179,6 +179,7 @@ struct t4_cqe { __be32 wrid_hi; __be32 wrid_low; } gen; + u64 drain_cookie; } u; __be64 reserved; __be64 bits_type_ts; @@ -238,6 +239,7 @@ struct t4_cqe { /* generic accessor macros */ #define CQE_WRID_HI(x) (be32_to_cpu((x)->u.gen.wrid_hi)) #define CQE_WRID_LOW(x) (be32_to_cpu((x)->u.gen.wrid_low)) +#define CQE_DRAIN_COOKIE(x) ((x)->u.drain_cookie) /* macros for flit 3 of the cqe */ #define CQE_GENBIT_S 63 diff --git a/drivers/infiniband/hw/hfi1/affinity.c b/drivers/infiniband/hw/hfi1/affinity.c index 67ea85a56945..7a3d906b3671 100644 --- a/drivers/infiniband/hw/hfi1/affinity.c +++ b/drivers/infiniband/hw/hfi1/affinity.c @@ -125,6 +125,7 @@ int node_affinity_init(void) cpumask_weight(topology_sibling_cpumask( cpumask_first(&node_affinity.proc.mask) )); + node_affinity.num_possible_nodes = num_possible_nodes(); node_affinity.num_online_nodes = num_online_nodes(); node_affinity.num_online_cpus = num_online_cpus(); @@ -135,7 +136,7 @@ int node_affinity_init(void) */ init_real_cpu_mask(); - hfi1_per_node_cntr = kcalloc(num_possible_nodes(), + hfi1_per_node_cntr = kcalloc(node_affinity.num_possible_nodes, sizeof(*hfi1_per_node_cntr), GFP_KERNEL); if (!hfi1_per_node_cntr) return -ENOMEM; diff --git a/drivers/infiniband/hw/hfi1/affinity.h b/drivers/infiniband/hw/hfi1/affinity.h index 42e63316afd1..e78c7aa094e0 100644 --- a/drivers/infiniband/hw/hfi1/affinity.h +++ b/drivers/infiniband/hw/hfi1/affinity.h @@ -70,14 +70,6 @@ struct cpu_mask_set { uint gen; }; -struct hfi1_affinity { - struct cpu_mask_set def_intr; - struct cpu_mask_set rcv_intr; - struct cpumask real_cpu_mask; - /* spin lock to protect affinity struct */ - spinlock_t lock; -}; - struct hfi1_msix_entry; /* Initialize non-HT cpu cores mask */ @@ -115,6 +107,7 @@ struct hfi1_affinity_node_list { struct cpumask real_cpu_mask; struct cpu_mask_set proc; int num_core_siblings; + int num_possible_nodes; int num_online_nodes; int num_online_cpus; struct mutex lock; /* protects affinity nodes */ diff --git a/drivers/infiniband/hw/hfi1/chip.c b/drivers/infiniband/hw/hfi1/chip.c index 24d0820873cf..121a4c920f1b 100644 --- a/drivers/infiniband/hw/hfi1/chip.c +++ b/drivers/infiniband/hw/hfi1/chip.c @@ -7827,7 +7827,8 @@ static void handle_dcc_err(struct hfi1_devdata *dd, u32 unused, u64 reg) } /* just report this */ - dd_dev_info(dd, "DCC Error: fmconfig error: %s\n", extra); + dd_dev_info_ratelimited(dd, "DCC Error: fmconfig error: %s\n", + extra); reg &= ~DCC_ERR_FLG_FMCONFIG_ERR_SMASK; } @@ -7878,34 +7879,35 @@ static void handle_dcc_err(struct hfi1_devdata *dd, u32 unused, u64 reg) } /* just report this */ - dd_dev_info(dd, "DCC Error: PortRcv error: %s\n", extra); - dd_dev_info(dd, " hdr0 0x%llx, hdr1 0x%llx\n", - hdr0, hdr1); + dd_dev_info_ratelimited(dd, "DCC Error: PortRcv error: %s\n" + " hdr0 0x%llx, hdr1 0x%llx\n", + extra, hdr0, hdr1); reg &= ~DCC_ERR_FLG_RCVPORT_ERR_SMASK; } if (reg & DCC_ERR_FLG_EN_CSR_ACCESS_BLOCKED_UC_SMASK) { /* informative only */ - dd_dev_info(dd, "8051 access to LCB blocked\n"); + dd_dev_info_ratelimited(dd, "8051 access to LCB blocked\n"); reg &= ~DCC_ERR_FLG_EN_CSR_ACCESS_BLOCKED_UC_SMASK; } if (reg & DCC_ERR_FLG_EN_CSR_ACCESS_BLOCKED_HOST_SMASK) { /* informative only */ - dd_dev_info(dd, "host access to LCB blocked\n"); + dd_dev_info_ratelimited(dd, "host access to LCB blocked\n"); reg &= ~DCC_ERR_FLG_EN_CSR_ACCESS_BLOCKED_HOST_SMASK; } /* report any remaining errors */ if (reg) - dd_dev_info(dd, "DCC Error: %s\n", - dcc_err_string(buf, sizeof(buf), reg)); + dd_dev_info_ratelimited(dd, "DCC Error: %s\n", + dcc_err_string(buf, sizeof(buf), reg)); if (lcl_reason == 0) lcl_reason = OPA_LINKDOWN_REASON_UNKNOWN; if (do_bounce) { - dd_dev_info(dd, "%s: PortErrorAction bounce\n", __func__); + dd_dev_info_ratelimited(dd, "%s: PortErrorAction bounce\n", + __func__); set_link_down_reason(ppd, lcl_reason, 0, lcl_reason); queue_work(ppd->hfi1_wq, &ppd->link_bounce_work); } @@ -8477,7 +8479,10 @@ static int do_8051_command( */ if (type == HCMD_WRITE_LCB_CSR) { in_data |= ((*out_data) & 0xffffffffffull) << 8; - reg = ((((*out_data) >> 40) & 0xff) << + /* must preserve COMPLETED - it is tied to hardware */ + reg = read_csr(dd, DC_DC8051_CFG_EXT_DEV_0); + reg &= DC_DC8051_CFG_EXT_DEV_0_COMPLETED_SMASK; + reg |= ((((*out_data) >> 40) & 0xff) << DC_DC8051_CFG_EXT_DEV_0_RETURN_CODE_SHIFT) | ((((*out_data) >> 48) & 0xffff) << DC_DC8051_CFG_EXT_DEV_0_RSP_DATA_SHIFT); @@ -9556,11 +9561,11 @@ int bringup_serdes(struct hfi1_pportdata *ppd) if (HFI1_CAP_IS_KSET(EXTENDED_PSN)) add_rcvctrl(dd, RCV_CTRL_RCV_EXTENDED_PSN_ENABLE_SMASK); - guid = ppd->guid; + guid = ppd->guids[HFI1_PORT_GUID_INDEX]; if (!guid) { if (dd->base_guid) guid = dd->base_guid + ppd->port - 1; - ppd->guid = guid; + ppd->guids[HFI1_PORT_GUID_INDEX] = guid; } /* Set linkinit_reason on power up per OPA spec */ @@ -10505,16 +10510,18 @@ int set_link_state(struct hfi1_pportdata *ppd, u32 state) ppd->remote_link_down_reason = 0; } - ret1 = set_physical_link_state(dd, PLS_DISABLED); - if (ret1 != HCMD_SUCCESS) { - dd_dev_err(dd, - "Failed to transition to Disabled link state, return 0x%x\n", - ret1); - ret = -EINVAL; - break; + if (!dd->dc_shutdown) { + ret1 = set_physical_link_state(dd, PLS_DISABLED); + if (ret1 != HCMD_SUCCESS) { + dd_dev_err(dd, + "Failed to transition to Disabled link state, return 0x%x\n", + ret1); + ret = -EINVAL; + break; + } + dc_shutdown(dd); } ppd->host_link_state = HLS_DN_DISABLE; - dc_shutdown(dd); break; case HLS_DN_OFFLINE: if (ppd->host_link_state == HLS_DN_DISABLE) diff --git a/drivers/infiniband/hw/hfi1/chip_registers.h b/drivers/infiniband/hw/hfi1/chip_registers.h index 5b9993899789..5bfa839d1c48 100644 --- a/drivers/infiniband/hw/hfi1/chip_registers.h +++ b/drivers/infiniband/hw/hfi1/chip_registers.h @@ -415,6 +415,9 @@ #define ASIC_CFG_SBUS_REQUEST_DATA_IN_SHIFT 32 #define ASIC_CFG_SBUS_REQUEST_RECEIVER_ADDR_SHIFT 0 #define ASIC_CFG_SCRATCH (ASIC + 0x000000000020) +#define ASIC_CFG_SCRATCH_1 (ASIC_CFG_SCRATCH + 0x08) +#define ASIC_CFG_SCRATCH_2 (ASIC_CFG_SCRATCH + 0x10) +#define ASIC_CFG_SCRATCH_3 (ASIC_CFG_SCRATCH + 0x18) #define ASIC_CFG_THERM_POLL_EN (ASIC + 0x000000000050) #define ASIC_EEP_ADDR_CMD (ASIC + 0x000000000308) #define ASIC_EEP_ADDR_CMD_EP_ADDR_MASK 0xFFFFFFull diff --git a/drivers/infiniband/hw/hfi1/common.h b/drivers/infiniband/hw/hfi1/common.h index da7be21bedb4..1b783bbee4bb 100644 --- a/drivers/infiniband/hw/hfi1/common.h +++ b/drivers/infiniband/hw/hfi1/common.h @@ -331,10 +331,6 @@ struct diag_pkt { #define FULL_MGMT_P_KEY 0xFFFF #define DEFAULT_P_KEY LIM_MGMT_P_KEY -#define HFI1_AETH_CREDIT_SHIFT 24 -#define HFI1_AETH_CREDIT_MASK 0x1F -#define HFI1_AETH_CREDIT_INVAL 0x1F -#define HFI1_MSN_MASK 0xFFFFFF #define HFI1_FECN_SHIFT 31 #define HFI1_FECN_MASK 1 #define HFI1_FECN_SMASK BIT(HFI1_FECN_SHIFT) diff --git a/drivers/infiniband/hw/hfi1/debugfs.c b/drivers/infiniband/hw/hfi1/debugfs.c index 632ba21759ab..7fe9dd885746 100644 --- a/drivers/infiniband/hw/hfi1/debugfs.c +++ b/drivers/infiniband/hw/hfi1/debugfs.c @@ -50,6 +50,7 @@ #include <linux/kernel.h> #include <linux/export.h> #include <linux/module.h> +#include <linux/string.h> #include "hfi.h" #include "debugfs.h" @@ -503,18 +504,11 @@ static ssize_t asic_flags_write(struct file *file, const char __user *buf, ppd = private2ppd(file); dd = ppd->dd; - buff = kmalloc(count + 1, GFP_KERNEL); - if (!buff) - return -ENOMEM; - - ret = copy_from_user(buff, buf, count); - if (ret > 0) { - ret = -EFAULT; - goto do_free; - } - /* zero terminate and read the expected integer */ - buff[count] = 0; + buff = memdup_user_nul(buf, count); + if (IS_ERR(buff)) + return PTR_ERR(buff); + ret = kstrtoull(buff, 0, &value); if (ret) goto do_free; @@ -541,6 +535,114 @@ static ssize_t asic_flags_write(struct file *file, const char __user *buf, return ret; } +/* read the dc8051 memory */ +static ssize_t dc8051_memory_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) +{ + struct hfi1_pportdata *ppd = private2ppd(file); + ssize_t rval; + void *tmp; + loff_t start, end; + + /* the checks below expect the position to be positive */ + if (*ppos < 0) + return -EINVAL; + + tmp = kzalloc(DC8051_DATA_MEM_SIZE, GFP_KERNEL); + if (!tmp) + return -ENOMEM; + + /* + * Fill in the requested portion of the temporary buffer from the + * 8051 memory. The 8051 memory read is done in terms of 8 bytes. + * Adjust start and end to fit. Skip reading anything if out of + * range. + */ + start = *ppos & ~0x7; /* round down */ + if (start < DC8051_DATA_MEM_SIZE) { + end = (*ppos + count + 7) & ~0x7; /* round up */ + if (end > DC8051_DATA_MEM_SIZE) + end = DC8051_DATA_MEM_SIZE; + rval = read_8051_data(ppd->dd, start, end - start, + (u64 *)(tmp + start)); + if (rval) + goto done; + } + + rval = simple_read_from_buffer(buf, count, ppos, tmp, + DC8051_DATA_MEM_SIZE); +done: + kfree(tmp); + return rval; +} + +static ssize_t debugfs_lcb_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) +{ + struct hfi1_pportdata *ppd = private2ppd(file); + struct hfi1_devdata *dd = ppd->dd; + unsigned long total, csr_off; + u64 data; + + if (*ppos < 0) + return -EINVAL; + /* only read 8 byte quantities */ + if ((count % 8) != 0) + return -EINVAL; + /* offset must be 8-byte aligned */ + if ((*ppos % 8) != 0) + return -EINVAL; + /* do nothing if out of range or zero count */ + if (*ppos >= (LCB_END - LCB_START) || !count) + return 0; + /* reduce count if needed */ + if (*ppos + count > LCB_END - LCB_START) + count = (LCB_END - LCB_START) - *ppos; + + csr_off = LCB_START + *ppos; + for (total = 0; total < count; total += 8, csr_off += 8) { + if (read_lcb_csr(dd, csr_off, (u64 *)&data)) + break; /* failed */ + if (put_user(data, (unsigned long __user *)(buf + total))) + break; + } + *ppos += total; + return total; +} + +static ssize_t debugfs_lcb_write(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) +{ + struct hfi1_pportdata *ppd = private2ppd(file); + struct hfi1_devdata *dd = ppd->dd; + unsigned long total, csr_off, data; + + if (*ppos < 0) + return -EINVAL; + /* only write 8 byte quantities */ + if ((count % 8) != 0) + return -EINVAL; + /* offset must be 8-byte aligned */ + if ((*ppos % 8) != 0) + return -EINVAL; + /* do nothing if out of range or zero count */ + if (*ppos >= (LCB_END - LCB_START) || !count) + return 0; + /* reduce count if needed */ + if (*ppos + count > LCB_END - LCB_START) + count = (LCB_END - LCB_START) - *ppos; + + csr_off = LCB_START + *ppos; + for (total = 0; total < count; total += 8, csr_off += 8) { + if (get_user(data, (unsigned long __user *)(buf + total))) + break; + if (write_lcb_csr(dd, csr_off, data)) + break; /* failed */ + } + *ppos += total; + return total; +} + /* * read the per-port QSFP data for ppd */ @@ -584,15 +686,9 @@ static ssize_t __i2c_debugfs_write(struct file *file, const char __user *buf, if (i2c_addr == 0) return -EINVAL; - buff = kmalloc(count, GFP_KERNEL); - if (!buff) - return -ENOMEM; - - ret = copy_from_user(buff, buf, count); - if (ret > 0) { - ret = -EFAULT; - goto _free; - } + buff = memdup_user(buf, count); + if (IS_ERR(buff)) + return PTR_ERR(buff); total_written = i2c_write(ppd, target, i2c_addr, offset, buff, count); if (total_written < 0) { @@ -697,15 +793,10 @@ static ssize_t __qsfp_debugfs_write(struct file *file, const char __user *buf, ppd = private2ppd(file); - buff = kmalloc(count, GFP_KERNEL); - if (!buff) - return -ENOMEM; + buff = memdup_user(buf, count); + if (IS_ERR(buff)) + return PTR_ERR(buff); - ret = copy_from_user(buff, buf, count); - if (ret > 0) { - ret = -EFAULT; - goto _free; - } total_written = qsfp_write(ppd, target, *ppos, buff, count); if (total_written < 0) { ret = total_written; @@ -931,6 +1022,8 @@ static const struct counter_info port_cntr_ops[] = { DEBUGFS_XOPS("qsfp2", qsfp2_debugfs_read, qsfp2_debugfs_write, qsfp2_debugfs_open, qsfp2_debugfs_release), DEBUGFS_OPS("asic_flags", asic_flags_read, asic_flags_write), + DEBUGFS_OPS("dc8051_memory", dc8051_memory_read, NULL), + DEBUGFS_OPS("lcb", debugfs_lcb_read, debugfs_lcb_write), }; static void *_sdma_cpu_list_seq_start(struct seq_file *s, loff_t *pos) diff --git a/drivers/infiniband/hw/hfi1/dma.c b/drivers/infiniband/hw/hfi1/dma.c deleted file mode 100644 index 7e8dab892848..000000000000 --- a/drivers/infiniband/hw/hfi1/dma.c +++ /dev/null @@ -1,183 +0,0 @@ -/* - * Copyright(c) 2015, 2016 Intel Corporation. - * - * This file is provided under a dual BSD/GPLv2 license. When using or - * redistributing this file, you may do so under either license. - * - * GPL LICENSE SUMMARY - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * 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. - * - * BSD LICENSE - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * - * - Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * - Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in - * the documentation and/or other materials provided with the - * distribution. - * - Neither the name of Intel Corporation nor the names of its - * contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR - * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT - * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, - * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT - * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, - * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY - * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE - * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - * - */ -#include <linux/types.h> -#include <linux/scatterlist.h> - -#include "verbs.h" - -#define BAD_DMA_ADDRESS ((u64)0) - -/* - * The following functions implement driver specific replacements - * for the ib_dma_*() functions. - * - * These functions return kernel virtual addresses instead of - * device bus addresses since the driver uses the CPU to copy - * data instead of using hardware DMA. - */ - -static int hfi1_mapping_error(struct ib_device *dev, u64 dma_addr) -{ - return dma_addr == BAD_DMA_ADDRESS; -} - -static u64 hfi1_dma_map_single(struct ib_device *dev, void *cpu_addr, - size_t size, enum dma_data_direction direction) -{ - if (WARN_ON(!valid_dma_direction(direction))) - return BAD_DMA_ADDRESS; - - return (u64)cpu_addr; -} - -static void hfi1_dma_unmap_single(struct ib_device *dev, u64 addr, size_t size, - enum dma_data_direction direction) -{ - /* This is a stub, nothing to be done here */ -} - -static u64 hfi1_dma_map_page(struct ib_device *dev, struct page *page, - unsigned long offset, size_t size, - enum dma_data_direction direction) -{ - u64 addr; - - if (WARN_ON(!valid_dma_direction(direction))) - return BAD_DMA_ADDRESS; - - if (offset + size > PAGE_SIZE) - return BAD_DMA_ADDRESS; - - addr = (u64)page_address(page); - if (addr) - addr += offset; - - return addr; -} - -static void hfi1_dma_unmap_page(struct ib_device *dev, u64 addr, size_t size, - enum dma_data_direction direction) -{ - /* This is a stub, nothing to be done here */ -} - -static int hfi1_map_sg(struct ib_device *dev, struct scatterlist *sgl, - int nents, enum dma_data_direction direction) -{ - struct scatterlist *sg; - u64 addr; - int i; - int ret = nents; - - if (WARN_ON(!valid_dma_direction(direction))) - return BAD_DMA_ADDRESS; - - for_each_sg(sgl, sg, nents, i) { - addr = (u64)page_address(sg_page(sg)); - if (!addr) { - ret = 0; - break; - } - sg->dma_address = addr + sg->offset; -#ifdef CONFIG_NEED_SG_DMA_LENGTH - sg->dma_length = sg->length; -#endif - } - return ret; -} - -static void hfi1_unmap_sg(struct ib_device *dev, - struct scatterlist *sg, int nents, - enum dma_data_direction direction) -{ - /* This is a stub, nothing to be done here */ -} - -static void hfi1_sync_single_for_cpu(struct ib_device *dev, u64 addr, - size_t size, enum dma_data_direction dir) -{ -} - -static void hfi1_sync_single_for_device(struct ib_device *dev, u64 addr, - size_t size, - enum dma_data_direction dir) -{ -} - -static void *hfi1_dma_alloc_coherent(struct ib_device *dev, size_t size, - u64 *dma_handle, gfp_t flag) -{ - struct page *p; - void *addr = NULL; - - p = alloc_pages(flag, get_order(size)); - if (p) - addr = page_address(p); - if (dma_handle) - *dma_handle = (u64)addr; - return addr; -} - -static void hfi1_dma_free_coherent(struct ib_device *dev, size_t size, - void *cpu_addr, u64 dma_handle) -{ - free_pages((unsigned long)cpu_addr, get_order(size)); -} - -struct ib_dma_mapping_ops hfi1_dma_mapping_ops = { - .mapping_error = hfi1_mapping_error, - .map_single = hfi1_dma_map_single, - .unmap_single = hfi1_dma_unmap_single, - .map_page = hfi1_dma_map_page, - .unmap_page = hfi1_dma_unmap_page, - .map_sg = hfi1_map_sg, - .unmap_sg = hfi1_unmap_sg, - .sync_single_for_cpu = hfi1_sync_single_for_cpu, - .sync_single_for_device = hfi1_sync_single_for_device, - .alloc_coherent = hfi1_dma_alloc_coherent, - .free_coherent = hfi1_dma_free_coherent -}; diff --git a/drivers/infiniband/hw/hfi1/driver.c b/drivers/infiniband/hw/hfi1/driver.c index c5efff29c147..3881c951f6af 100644 --- a/drivers/infiniband/hw/hfi1/driver.c +++ b/drivers/infiniband/hw/hfi1/driver.c @@ -100,6 +100,11 @@ MODULE_VERSION(HFI1_DRIVER_VERSION); * MAX_PKT_RCV is the max # if packets processed per receive interrupt. */ #define MAX_PKT_RECV 64 +/* + * MAX_PKT_THREAD_RCV is the max # of packets processed before + * the qp_wait_list queue is flushed. + */ +#define MAX_PKT_RECV_THREAD (MAX_PKT_RECV * 4) #define EGR_HEAD_UPDATE_THRESHOLD 16 struct hfi1_ib_stats hfi1_stats; @@ -259,7 +264,7 @@ static inline void *get_egrbuf(const struct hfi1_ctxtdata *rcd, u64 rhf, * allowed size ranges for the respective type and, optionally, * return the proper encoding. */ -inline int hfi1_rcvbuf_validate(u32 size, u8 type, u16 *encoded) +int hfi1_rcvbuf_validate(u32 size, u8 type, u16 *encoded) { if (unlikely(!PAGE_ALIGNED(size))) return 0; @@ -279,7 +284,7 @@ static void rcv_hdrerr(struct hfi1_ctxtdata *rcd, struct hfi1_pportdata *ppd, struct ib_header *rhdr = packet->hdr; u32 rte = rhf_rcv_type_err(packet->rhf); int lnh = be16_to_cpu(rhdr->lrh[0]) & 3; - struct hfi1_ibport *ibp = &ppd->ibport_data; + struct hfi1_ibport *ibp = rcd_to_iport(rcd); struct hfi1_devdata *dd = ppd->dd; struct rvt_dev_info *rdi = &dd->verbs_dev.rdi; @@ -594,7 +599,7 @@ static void __prescan_rxq(struct hfi1_packet *packet) while (1) { struct hfi1_devdata *dd = rcd->dd; - struct hfi1_ibport *ibp = &rcd->ppd->ibport_data; + struct hfi1_ibport *ibp = rcd_to_iport(rcd); __le32 *rhf_addr = (__le32 *)rcd->rcvhdrq + mdata.ps_head + dd->rhf_offset; struct rvt_qp *qp; @@ -654,24 +659,68 @@ next: } } -static inline int skip_rcv_packet(struct hfi1_packet *packet, int thread) +static void process_rcv_qp_work(struct hfi1_ctxtdata *rcd) +{ + struct rvt_qp *qp, *nqp; + + /* + * Iterate over all QPs waiting to respond. + * The list won't change since the IRQ is only run on one CPU. + */ + list_for_each_entry_safe(qp, nqp, &rcd->qp_wait_list, rspwait) { + list_del_init(&qp->rspwait); + if (qp->r_flags & RVT_R_RSP_NAK) { + qp->r_flags &= ~RVT_R_RSP_NAK; + hfi1_send_rc_ack(rcd, qp, 0); + } + if (qp->r_flags & RVT_R_RSP_SEND) { + unsigned long flags; + + qp->r_flags &= ~RVT_R_RSP_SEND; + spin_lock_irqsave(&qp->s_lock, flags); + if (ib_rvt_state_ops[qp->state] & + RVT_PROCESS_OR_FLUSH_SEND) + hfi1_schedule_send(qp); + spin_unlock_irqrestore(&qp->s_lock, flags); + } + rvt_put_qp(qp); + } +} + +static noinline int max_packet_exceeded(struct hfi1_packet *packet, int thread) +{ + if (thread) { + if ((packet->numpkt & (MAX_PKT_RECV_THREAD - 1)) == 0) + /* allow defered processing */ + process_rcv_qp_work(packet->rcd); + cond_resched(); + return RCV_PKT_OK; + } else { + this_cpu_inc(*packet->rcd->dd->rcv_limit); + return RCV_PKT_LIMIT; + } +} + +static inline int check_max_packet(struct hfi1_packet *packet, int thread) { int ret = RCV_PKT_OK; + if (unlikely((packet->numpkt & (MAX_PKT_RECV - 1)) == 0)) + ret = max_packet_exceeded(packet, thread); + return ret; +} + +static noinline int skip_rcv_packet(struct hfi1_packet *packet, int thread) +{ + int ret; + /* Set up for the next packet */ packet->rhqoff += packet->rsize; if (packet->rhqoff >= packet->maxcnt) packet->rhqoff = 0; packet->numpkt++; - if (unlikely((packet->numpkt & (MAX_PKT_RECV - 1)) == 0)) { - if (thread) { - cond_resched(); - } else { - ret = RCV_PKT_LIMIT; - this_cpu_inc(*packet->rcd->dd->rcv_limit); - } - } + ret = check_max_packet(packet, thread); packet->rhf_addr = (__le32 *)packet->rcd->rcvhdrq + packet->rhqoff + packet->rcd->dd->rhf_offset; @@ -682,7 +731,7 @@ static inline int skip_rcv_packet(struct hfi1_packet *packet, int thread) static inline int process_rcv_packet(struct hfi1_packet *packet, int thread) { - int ret = RCV_PKT_OK; + int ret; packet->hdr = hfi1_get_msgheader(packet->rcd->dd, packet->rhf_addr); @@ -723,14 +772,7 @@ static inline int process_rcv_packet(struct hfi1_packet *packet, int thread) if (packet->rhqoff >= packet->maxcnt) packet->rhqoff = 0; - if (unlikely((packet->numpkt & (MAX_PKT_RECV - 1)) == 0)) { - if (thread) { - cond_resched(); - } else { - ret = RCV_PKT_LIMIT; - this_cpu_inc(*packet->rcd->dd->rcv_limit); - } - } + ret = check_max_packet(packet, thread); packet->rhf_addr = (__le32 *)packet->rcd->rcvhdrq + packet->rhqoff + packet->rcd->dd->rhf_offset; @@ -767,39 +809,6 @@ static inline void finish_packet(struct hfi1_packet *packet) packet->etail, rcv_intr_dynamic, packet->numpkt); } -static inline void process_rcv_qp_work(struct hfi1_packet *packet) -{ - struct hfi1_ctxtdata *rcd; - struct rvt_qp *qp, *nqp; - - rcd = packet->rcd; - rcd->head = packet->rhqoff; - - /* - * Iterate over all QPs waiting to respond. - * The list won't change since the IRQ is only run on one CPU. - */ - list_for_each_entry_safe(qp, nqp, &rcd->qp_wait_list, rspwait) { - list_del_init(&qp->rspwait); - if (qp->r_flags & RVT_R_RSP_NAK) { - qp->r_flags &= ~RVT_R_RSP_NAK; - hfi1_send_rc_ack(rcd, qp, 0); - } - if (qp->r_flags & RVT_R_RSP_SEND) { - unsigned long flags; - - qp->r_flags &= ~RVT_R_RSP_SEND; - spin_lock_irqsave(&qp->s_lock, flags); - if (ib_rvt_state_ops[qp->state] & - RVT_PROCESS_OR_FLUSH_SEND) - hfi1_schedule_send(qp); - spin_unlock_irqrestore(&qp->s_lock, flags); - } - if (atomic_dec_and_test(&qp->refcount)) - wake_up(&qp->wait); - } -} - /* * Handle receive interrupts when using the no dma rtail option. */ @@ -827,7 +836,8 @@ int handle_receive_interrupt_nodma_rtail(struct hfi1_ctxtdata *rcd, int thread) last = RCV_PKT_DONE; process_rcv_update(last, &packet); } - process_rcv_qp_work(&packet); + process_rcv_qp_work(rcd); + rcd->head = packet.rhqoff; bail: finish_packet(&packet); return last; @@ -855,7 +865,8 @@ int handle_receive_interrupt_dma_rtail(struct hfi1_ctxtdata *rcd, int thread) last = RCV_PKT_DONE; process_rcv_update(last, &packet); } - process_rcv_qp_work(&packet); + process_rcv_qp_work(rcd); + rcd->head = packet.rhqoff; bail: finish_packet(&packet); return last; @@ -1025,7 +1036,8 @@ int handle_receive_interrupt(struct hfi1_ctxtdata *rcd, int thread) process_rcv_update(last, &packet); } - process_rcv_qp_work(&packet); + process_rcv_qp_work(rcd); + rcd->head = packet.rhqoff; bail: /* diff --git a/drivers/infiniband/hw/hfi1/efivar.c b/drivers/infiniband/hw/hfi1/efivar.c index 106349fc1fb9..d106d23016ba 100644 --- a/drivers/infiniband/hw/hfi1/efivar.c +++ b/drivers/infiniband/hw/hfi1/efivar.c @@ -45,6 +45,7 @@ * */ +#include <linux/ctype.h> #include "efivar.h" /* GUID for HFI1 variables in EFI */ @@ -150,15 +151,32 @@ fail: int read_hfi1_efi_var(struct hfi1_devdata *dd, const char *kind, unsigned long *size, void **return_data) { + char prefix_name[64]; char name[64]; + int result; + int i; /* create a common prefix */ - snprintf(name, sizeof(name), "%04x:%02x:%02x.%x-%s", + snprintf(prefix_name, sizeof(prefix_name), "%04x:%02x:%02x.%x", pci_domain_nr(dd->pcidev->bus), dd->pcidev->bus->number, PCI_SLOT(dd->pcidev->devfn), - PCI_FUNC(dd->pcidev->devfn), - kind); + PCI_FUNC(dd->pcidev->devfn)); + snprintf(name, sizeof(name), "%s-%s", prefix_name, kind); + result = read_efi_var(name, size, return_data); + + /* + * If reading the lowercase EFI variable fail, read the uppercase + * variable. + */ + if (result) { + /* Converting to uppercase */ + for (i = 0; prefix_name[i]; i++) + if (isalpha(prefix_name[i])) + prefix_name[i] = toupper(prefix_name[i]); + snprintf(name, sizeof(name), "%s-%s", prefix_name, kind); + result = read_efi_var(name, size, return_data); + } - return read_efi_var(name, size, return_data); + return result; } diff --git a/drivers/infiniband/hw/hfi1/eprom.c b/drivers/infiniband/hw/hfi1/eprom.c index e70c223801b4..26da124c88e2 100644 --- a/drivers/infiniband/hw/hfi1/eprom.c +++ b/drivers/infiniband/hw/hfi1/eprom.c @@ -207,6 +207,40 @@ done_asic: /* magic character sequence that trails an image */ #define IMAGE_TRAIL_MAGIC "egamiAPO" +/* EPROM file types */ +#define HFI1_EFT_PLATFORM_CONFIG 2 + +/* segment size - 128 KiB */ +#define SEG_SIZE (128 * 1024) + +struct hfi1_eprom_footer { + u32 oprom_size; /* size of the oprom, in bytes */ + u16 num_table_entries; + u16 version; /* version of this footer */ + u32 magic; /* must be last */ +}; + +struct hfi1_eprom_table_entry { + u32 type; /* file type */ + u32 offset; /* file offset from start of EPROM */ + u32 size; /* file size, in bytes */ +}; + +/* + * Calculate the max number of table entries that will fit within a directory + * buffer of size 'dir_size'. + */ +#define MAX_TABLE_ENTRIES(dir_size) \ + (((dir_size) - sizeof(struct hfi1_eprom_footer)) / \ + sizeof(struct hfi1_eprom_table_entry)) + +#define DIRECTORY_SIZE(n) (sizeof(struct hfi1_eprom_footer) + \ + (sizeof(struct hfi1_eprom_table_entry) * (n))) + +#define MAGIC4(a, b, c, d) ((d) << 24 | (c) << 16 | (b) << 8 | (a)) +#define FOOTER_MAGIC MAGIC4('e', 'p', 'r', 'm') +#define FOOTER_VERSION 1 + /* * Read all of partition 1. The actual file is at the front. Adjust * the returned size if a trailing image magic is found. @@ -242,6 +276,167 @@ static int read_partition_platform_config(struct hfi1_devdata *dd, void **data, } /* + * The segment magic has been checked. There is a footer and table of + * contents present. + * + * directory is a u32 aligned buffer of size EP_PAGE_SIZE. + */ +static int read_segment_platform_config(struct hfi1_devdata *dd, + void *directory, void **data, u32 *size) +{ + struct hfi1_eprom_footer *footer; + struct hfi1_eprom_table_entry *table; + struct hfi1_eprom_table_entry *entry; + void *buffer = NULL; + void *table_buffer = NULL; + int ret, i; + u32 directory_size; + u32 seg_base, seg_offset; + u32 bytes_available, ncopied, to_copy; + + /* the footer is at the end of the directory */ + footer = (struct hfi1_eprom_footer *) + (directory + EP_PAGE_SIZE - sizeof(*footer)); + + /* make sure the structure version is supported */ + if (footer->version != FOOTER_VERSION) + return -EINVAL; + + /* oprom size cannot be larger than a segment */ + if (footer->oprom_size >= SEG_SIZE) + return -EINVAL; + + /* the file table must fit in a segment with the oprom */ + if (footer->num_table_entries > + MAX_TABLE_ENTRIES(SEG_SIZE - footer->oprom_size)) + return -EINVAL; + + /* find the file table start, which precedes the footer */ + directory_size = DIRECTORY_SIZE(footer->num_table_entries); + if (directory_size <= EP_PAGE_SIZE) { + /* the file table fits into the directory buffer handed in */ + table = (struct hfi1_eprom_table_entry *) + (directory + EP_PAGE_SIZE - directory_size); + } else { + /* need to allocate and read more */ + table_buffer = kmalloc(directory_size, GFP_KERNEL); + if (!table_buffer) + return -ENOMEM; + ret = read_length(dd, SEG_SIZE - directory_size, + directory_size, table_buffer); + if (ret) + goto done; + table = table_buffer; + } + + /* look for the platform configuration file in the table */ + for (entry = NULL, i = 0; i < footer->num_table_entries; i++) { + if (table[i].type == HFI1_EFT_PLATFORM_CONFIG) { + entry = &table[i]; + break; + } + } + if (!entry) { + ret = -ENOENT; + goto done; + } + + /* + * Sanity check on the configuration file size - it should never + * be larger than 4 KiB. + */ + if (entry->size > (4 * 1024)) { + dd_dev_err(dd, "Bad configuration file size 0x%x\n", + entry->size); + ret = -EINVAL; + goto done; + } + + /* check for bogus offset and size that wrap when added together */ + if (entry->offset + entry->size < entry->offset) { + dd_dev_err(dd, + "Bad configuration file start + size 0x%x+0x%x\n", + entry->offset, entry->size); + ret = -EINVAL; + goto done; + } + + /* allocate the buffer to return */ + buffer = kmalloc(entry->size, GFP_KERNEL); + if (!buffer) { + ret = -ENOMEM; + goto done; + } + + /* + * Extract the file by looping over segments until it is fully read. + */ + seg_offset = entry->offset % SEG_SIZE; + seg_base = entry->offset - seg_offset; + ncopied = 0; + while (ncopied < entry->size) { + /* calculate data bytes available in this segment */ + + /* start with the bytes from the current offset to the end */ + bytes_available = SEG_SIZE - seg_offset; + /* subtract off footer and table from segment 0 */ + if (seg_base == 0) { + /* + * Sanity check: should not have a starting point + * at or within the directory. + */ + if (bytes_available <= directory_size) { + dd_dev_err(dd, + "Bad configuration file - offset 0x%x within footer+table\n", + entry->offset); + ret = -EINVAL; + goto done; + } + bytes_available -= directory_size; + } + + /* calculate bytes wanted */ + to_copy = entry->size - ncopied; + + /* max out at the available bytes in this segment */ + if (to_copy > bytes_available) + to_copy = bytes_available; + + /* + * Read from the EPROM. + * + * The sanity check for entry->offset is done in read_length(). + * The EPROM offset is validated against what the hardware + * addressing supports. In addition, if the offset is larger + * than the actual EPROM, it silently wraps. It will work + * fine, though the reader may not get what they expected + * from the EPROM. + */ + ret = read_length(dd, seg_base + seg_offset, to_copy, + buffer + ncopied); + if (ret) + goto done; + + ncopied += to_copy; + + /* set up for next segment */ + seg_offset = footer->oprom_size; + seg_base += SEG_SIZE; + } + + /* success */ + ret = 0; + *data = buffer; + *size = entry->size; + +done: + kfree(table_buffer); + if (ret) + kfree(buffer); + return ret; +} + +/* * Read the platform configuration file from the EPROM. * * On success, an allocated buffer containing the data and its size are @@ -253,6 +448,7 @@ static int read_partition_platform_config(struct hfi1_devdata *dd, void **data, * -EBUSY - not able to acquire access to the EPROM * -ENOENT - no recognizable file written * -ENOMEM - buffer could not be allocated + * -EINVAL - invalid EPROM contentents found */ int eprom_read_platform_config(struct hfi1_devdata *dd, void **data, u32 *size) { @@ -266,21 +462,20 @@ int eprom_read_platform_config(struct hfi1_devdata *dd, void **data, u32 *size) if (ret) return -EBUSY; - /* read the last page of P0 for the EPROM format magic */ - ret = read_length(dd, P1_START - EP_PAGE_SIZE, EP_PAGE_SIZE, directory); + /* read the last page of the segment for the EPROM format magic */ + ret = read_length(dd, SEG_SIZE - EP_PAGE_SIZE, EP_PAGE_SIZE, directory); if (ret) goto done; - /* last dword of P0 contains a magic indicator */ - if (directory[EP_PAGE_DWORDS - 1] == 0) { + /* last dword of the segment contains a magic value */ + if (directory[EP_PAGE_DWORDS - 1] == FOOTER_MAGIC) { + /* segment format */ + ret = read_segment_platform_config(dd, directory, data, size); + } else { /* partition format */ ret = read_partition_platform_config(dd, data, size); - goto done; } - /* nothing recognized */ - ret = -ENOENT; - done: release_chip_resource(dd, CR_EPROM); return ret; diff --git a/drivers/infiniband/hw/hfi1/file_ops.c b/drivers/infiniband/hw/hfi1/file_ops.c index bd786b7bd30b..f46033984d07 100644 --- a/drivers/infiniband/hw/hfi1/file_ops.c +++ b/drivers/infiniband/hw/hfi1/file_ops.c @@ -92,7 +92,7 @@ static unsigned int poll_next(struct file *, struct poll_table_struct *); static int user_event_ack(struct hfi1_ctxtdata *, int, unsigned long); static int set_ctxt_pkey(struct hfi1_ctxtdata *, unsigned, u16); static int manage_rcvq(struct hfi1_ctxtdata *, unsigned, int); -static int vma_fault(struct vm_area_struct *, struct vm_fault *); +static int vma_fault(struct vm_fault *); static long hfi1_file_ioctl(struct file *fp, unsigned int cmd, unsigned long arg); @@ -695,7 +695,7 @@ done: * Local (non-chip) user memory is not mapped right away but as it is * accessed by the user-level code. */ -static int vma_fault(struct vm_area_struct *vma, struct vm_fault *vmf) +static int vma_fault(struct vm_fault *vmf) { struct page *page; diff --git a/drivers/infiniband/hw/hfi1/firmware.c b/drivers/infiniband/hw/hfi1/firmware.c index 13db8eb4f4ec..0dd50cdb039a 100644 --- a/drivers/infiniband/hw/hfi1/firmware.c +++ b/drivers/infiniband/hw/hfi1/firmware.c @@ -239,6 +239,16 @@ static const u8 all_fabric_serdes_broadcast = 0xe1; const u8 pcie_serdes_broadcast[2] = { 0xe2, 0xe3 }; static const u8 all_pcie_serdes_broadcast = 0xe0; +static const u32 platform_config_table_limits[PLATFORM_CONFIG_TABLE_MAX] = { + 0, + SYSTEM_TABLE_MAX, + PORT_TABLE_MAX, + RX_PRESET_TABLE_MAX, + TX_PRESET_TABLE_MAX, + QSFP_ATTEN_TABLE_MAX, + VARIABLE_SETTINGS_TABLE_MAX +}; + /* forwards */ static void dispose_one_firmware(struct firmware_details *fdet); static int load_fabric_serdes_firmware(struct hfi1_devdata *dd, @@ -263,11 +273,13 @@ static int __read_8051_data(struct hfi1_devdata *dd, u32 addr, u64 *result) u64 reg; int count; - /* start the read at the given address */ - reg = ((addr & DC_DC8051_CFG_RAM_ACCESS_CTRL_ADDRESS_MASK) - << DC_DC8051_CFG_RAM_ACCESS_CTRL_ADDRESS_SHIFT) - | DC_DC8051_CFG_RAM_ACCESS_CTRL_READ_ENA_SMASK; + /* step 1: set the address, clear enable */ + reg = (addr & DC_DC8051_CFG_RAM_ACCESS_CTRL_ADDRESS_MASK) + << DC_DC8051_CFG_RAM_ACCESS_CTRL_ADDRESS_SHIFT; write_csr(dd, DC_DC8051_CFG_RAM_ACCESS_CTRL, reg); + /* step 2: enable */ + write_csr(dd, DC_DC8051_CFG_RAM_ACCESS_CTRL, + reg | DC_DC8051_CFG_RAM_ACCESS_CTRL_READ_ENA_SMASK); /* wait until ACCESS_COMPLETED is set */ count = 0; @@ -707,6 +719,9 @@ static int obtain_firmware(struct hfi1_devdata *dd) &dd->pcidev->dev); if (err) { platform_config = NULL; + dd_dev_err(dd, + "%s: No default platform config file found\n", + __func__); goto done; } dd->platform_config.data = platform_config->data; @@ -1761,8 +1776,17 @@ int parse_platform_config(struct hfi1_devdata *dd) u32 record_idx = 0, table_type = 0, table_length_dwords = 0; int ret = -EINVAL; /* assume failure */ + /* + * For integrated devices that did not fall back to the default file, + * the SI tuning information for active channels is acquired from the + * scratch register bitmap, thus there is no platform config to parse. + * Skip parsing in these situations. + */ + if (is_integrated(dd) && !platform_config_load) + return 0; + if (!dd->platform_config.data) { - dd_dev_info(dd, "%s: Missing config file\n", __func__); + dd_dev_err(dd, "%s: Missing config file\n", __func__); goto bail; } ptr = (u32 *)dd->platform_config.data; @@ -1770,7 +1794,7 @@ int parse_platform_config(struct hfi1_devdata *dd) magic_num = *ptr; ptr++; if (magic_num != PLATFORM_CONFIG_MAGIC_NUM) { - dd_dev_info(dd, "%s: Bad config file\n", __func__); + dd_dev_err(dd, "%s: Bad config file\n", __func__); goto bail; } @@ -1797,9 +1821,9 @@ int parse_platform_config(struct hfi1_devdata *dd) header1 = *ptr; header2 = *(ptr + 1); if (header1 != ~header2) { - dd_dev_info(dd, "%s: Failed validation at offset %ld\n", - __func__, (ptr - (u32 *) - dd->platform_config.data)); + dd_dev_err(dd, "%s: Failed validation at offset %ld\n", + __func__, (ptr - (u32 *) + dd->platform_config.data)); goto bail; } @@ -1841,11 +1865,11 @@ int parse_platform_config(struct hfi1_devdata *dd) table_length_dwords; break; default: - dd_dev_info(dd, - "%s: Unknown data table %d, offset %ld\n", - __func__, table_type, - (ptr - (u32 *) - dd->platform_config.data)); + dd_dev_err(dd, + "%s: Unknown data table %d, offset %ld\n", + __func__, table_type, + (ptr - (u32 *) + dd->platform_config.data)); goto bail; /* We don't trust this file now */ } pcfgcache->config_tables[table_type].table = ptr; @@ -1865,11 +1889,11 @@ int parse_platform_config(struct hfi1_devdata *dd) case PLATFORM_CONFIG_VARIABLE_SETTINGS_TABLE: break; default: - dd_dev_info(dd, - "%s: Unknown meta table %d, offset %ld\n", - __func__, table_type, - (ptr - - (u32 *)dd->platform_config.data)); + dd_dev_err(dd, + "%s: Unknown meta table %d, offset %ld\n", + __func__, table_type, + (ptr - + (u32 *)dd->platform_config.data)); goto bail; /* We don't trust this file now */ } pcfgcache->config_tables[table_type].table_metadata = @@ -1884,10 +1908,9 @@ int parse_platform_config(struct hfi1_devdata *dd) /* Jump the table */ ptr += table_length_dwords; if (crc != *ptr) { - dd_dev_info(dd, "%s: Failed CRC check at offset %ld\n", - __func__, (ptr - - (u32 *) - dd->platform_config.data)); + dd_dev_err(dd, "%s: Failed CRC check at offset %ld\n", + __func__, (ptr - + (u32 *)dd->platform_config.data)); goto bail; } /* Jump the CRC DWORD */ @@ -1901,6 +1924,84 @@ bail: return ret; } +static void get_integrated_platform_config_field( + struct hfi1_devdata *dd, + enum platform_config_table_type_encoding table_type, + int field_index, u32 *data) +{ + struct hfi1_pportdata *ppd = dd->pport; + u8 *cache = ppd->qsfp_info.cache; + u32 tx_preset = 0; + + switch (table_type) { + case PLATFORM_CONFIG_SYSTEM_TABLE: + if (field_index == SYSTEM_TABLE_QSFP_POWER_CLASS_MAX) + *data = ppd->max_power_class; + else if (field_index == SYSTEM_TABLE_QSFP_ATTENUATION_DEFAULT_25G) + *data = ppd->default_atten; + break; + case PLATFORM_CONFIG_PORT_TABLE: + if (field_index == PORT_TABLE_PORT_TYPE) + *data = ppd->port_type; + else if (field_index == PORT_TABLE_LOCAL_ATTEN_25G) + *data = ppd->local_atten; + else if (field_index == PORT_TABLE_REMOTE_ATTEN_25G) + *data = ppd->remote_atten; + break; + case PLATFORM_CONFIG_RX_PRESET_TABLE: + if (field_index == RX_PRESET_TABLE_QSFP_RX_CDR_APPLY) + *data = (ppd->rx_preset & QSFP_RX_CDR_APPLY_SMASK) >> + QSFP_RX_CDR_APPLY_SHIFT; + else if (field_index == RX_PRESET_TABLE_QSFP_RX_EMP_APPLY) + *data = (ppd->rx_preset & QSFP_RX_EMP_APPLY_SMASK) >> + QSFP_RX_EMP_APPLY_SHIFT; + else if (field_index == RX_PRESET_TABLE_QSFP_RX_AMP_APPLY) + *data = (ppd->rx_preset & QSFP_RX_AMP_APPLY_SMASK) >> + QSFP_RX_AMP_APPLY_SHIFT; + else if (field_index == RX_PRESET_TABLE_QSFP_RX_CDR) + *data = (ppd->rx_preset & QSFP_RX_CDR_SMASK) >> + QSFP_RX_CDR_SHIFT; + else if (field_index == RX_PRESET_TABLE_QSFP_RX_EMP) + *data = (ppd->rx_preset & QSFP_RX_EMP_SMASK) >> + QSFP_RX_EMP_SHIFT; + else if (field_index == RX_PRESET_TABLE_QSFP_RX_AMP) + *data = (ppd->rx_preset & QSFP_RX_AMP_SMASK) >> + QSFP_RX_AMP_SHIFT; + break; + case PLATFORM_CONFIG_TX_PRESET_TABLE: + if (cache[QSFP_EQ_INFO_OFFS] & 0x4) + tx_preset = ppd->tx_preset_eq; + else + tx_preset = ppd->tx_preset_noeq; + if (field_index == TX_PRESET_TABLE_PRECUR) + *data = (tx_preset & TX_PRECUR_SMASK) >> + TX_PRECUR_SHIFT; + else if (field_index == TX_PRESET_TABLE_ATTN) + *data = (tx_preset & TX_ATTN_SMASK) >> + TX_ATTN_SHIFT; + else if (field_index == TX_PRESET_TABLE_POSTCUR) + *data = (tx_preset & TX_POSTCUR_SMASK) >> + TX_POSTCUR_SHIFT; + else if (field_index == TX_PRESET_TABLE_QSFP_TX_CDR_APPLY) + *data = (tx_preset & QSFP_TX_CDR_APPLY_SMASK) >> + QSFP_TX_CDR_APPLY_SHIFT; + else if (field_index == TX_PRESET_TABLE_QSFP_TX_EQ_APPLY) + *data = (tx_preset & QSFP_TX_EQ_APPLY_SMASK) >> + QSFP_TX_EQ_APPLY_SHIFT; + else if (field_index == TX_PRESET_TABLE_QSFP_TX_CDR) + *data = (tx_preset & QSFP_TX_CDR_SMASK) >> + QSFP_TX_CDR_SHIFT; + else if (field_index == TX_PRESET_TABLE_QSFP_TX_EQ) + *data = (tx_preset & QSFP_TX_EQ_SMASK) >> + QSFP_TX_EQ_SHIFT; + break; + case PLATFORM_CONFIG_QSFP_ATTEN_TABLE: + case PLATFORM_CONFIG_VARIABLE_SETTINGS_TABLE: + default: + break; + } +} + static int get_platform_fw_field_metadata(struct hfi1_devdata *dd, int table, int field, u32 *field_len_bits, u32 *field_start_bits) @@ -1976,6 +2077,15 @@ int get_platform_config_field(struct hfi1_devdata *dd, else return -EINVAL; + if (is_integrated(dd) && !platform_config_load) { + /* + * Use saved configuration from ppd for integrated platforms + */ + get_integrated_platform_config_field(dd, table_type, + field_index, data); + return 0; + } + ret = get_platform_fw_field_metadata(dd, table_type, field_index, &field_len_bits, &field_start_bits); diff --git a/drivers/infiniband/hw/hfi1/hfi.h b/drivers/infiniband/hw/hfi1/hfi.h index cc87fd4e534b..0808e3c3ba39 100644 --- a/drivers/infiniband/hw/hfi1/hfi.h +++ b/drivers/infiniband/hw/hfi1/hfi.h @@ -356,12 +356,11 @@ struct hfi1_packet { u64 rhf; u32 maxcnt; u32 rhqoff; - u32 hdrqtail; - int numpkt; u16 tlen; - u16 hlen; s16 etail; - u16 rsize; + u8 hlen; + u8 numpkt; + u8 rsize; u8 updegr; u8 rcv_flags; u8 etype; @@ -492,6 +491,9 @@ struct rvt_sge_state; #define HFI1_MIN_VLS_SUPPORTED 1 #define HFI1_MAX_VLS_SUPPORTED 8 +#define HFI1_GUIDS_PER_PORT 5 +#define HFI1_PORT_GUID_INDEX 0 + static inline void incr_cntr64(u64 *cntr) { if (*cntr < (u64)-1LL) @@ -559,11 +561,20 @@ struct hfi1_pportdata { struct kobject vl2mtu_kobj; /* PHY support */ - u32 port_type; struct qsfp_data qsfp_info; + /* Values for SI tuning of SerDes */ + u32 port_type; + u32 tx_preset_eq; + u32 tx_preset_noeq; + u32 rx_preset; + u8 local_atten; + u8 remote_atten; + u8 default_atten; + u8 max_power_class; + + /* GUIDs for this interface, in host order, guids[0] is a port guid */ + u64 guids[HFI1_GUIDS_PER_PORT]; - /* GUID for this interface, in host order */ - u64 guid; /* GUID for peer interface, in host order */ u64 neighbor_guid; @@ -826,32 +837,29 @@ struct hfi1_devdata { u8 __iomem *kregend; /* physical address of chip for io_remap, etc. */ resource_size_t physaddr; - /* receive context data */ - struct hfi1_ctxtdata **rcd; + /* Per VL data. Enough for all VLs but not all elements are set/used. */ + struct per_vl_data vld[PER_VL_SEND_CONTEXTS]; /* send context data */ struct send_context_info *send_contexts; /* map hardware send contexts to software index */ u8 *hw_to_sw; /* spinlock for allocating and releasing send context resources */ spinlock_t sc_lock; - /* Per VL data. Enough for all VLs but not all elements are set/used. */ - struct per_vl_data vld[PER_VL_SEND_CONTEXTS]; /* lock for pio_map */ spinlock_t pio_map_lock; + /* Send Context initialization lock. */ + spinlock_t sc_init_lock; + /* lock for sdma_map */ + spinlock_t sde_map_lock; /* array of kernel send contexts */ struct send_context **kernel_send_context; /* array of vl maps */ struct pio_vl_map __rcu *pio_map; - /* seqlock for sc2vl */ - seqlock_t sc2vl_lock; - u64 sc2vl[4]; - /* Send Context initialization lock. */ - spinlock_t sc_init_lock; + /* default flags to last descriptor */ + u64 default_desc1; /* fields common to all SDMA engines */ - /* default flags to last descriptor */ - u64 default_desc1; volatile __le64 *sdma_heads_dma; /* DMA'ed by chip */ dma_addr_t sdma_heads_phys; void *sdma_pad_dma; /* DMA'ed by chip */ @@ -862,8 +870,6 @@ struct hfi1_devdata { u32 chip_sdma_engines; /* num used */ u32 num_sdma; - /* lock for sdma_map */ - spinlock_t sde_map_lock; /* array of engines sized by num_sdma */ struct sdma_engine *per_sdma; /* array of vl maps */ @@ -872,14 +878,11 @@ struct hfi1_devdata { wait_queue_head_t sdma_unfreeze_wq; atomic_t sdma_unfreeze_count; + u32 lcb_access_count; /* count of LCB users */ + /* common data between shared ASIC HFIs in this OS */ struct hfi1_asic_data *asic_data; - /* hfi1_pportdata, points to array of (physical) port-specific - * data structs, indexed by pidx (0..n-1) - */ - struct hfi1_pportdata *pport; - /* mem-mapped pointer to base of PIO buffers */ void __iomem *piobase; /* @@ -896,20 +899,13 @@ struct hfi1_devdata { /* send context numbers and sizes for each type */ struct sc_config_sizes sc_sizes[SC_MAX]; - u32 lcb_access_count; /* count of LCB users */ - char *boardname; /* human readable board info */ - /* device (not port) flags, basically device capabilities */ - u32 flags; - /* reset value */ u64 z_int_counter; u64 z_rcv_limit; u64 z_send_schedule; - /* percpu int_counter */ - u64 __percpu *int_counter; - u64 __percpu *rcv_limit; + u64 __percpu *send_schedule; /* number of receive contexts in use by the driver */ u32 num_rcv_contexts; @@ -924,6 +920,7 @@ struct hfi1_devdata { /* base receive interrupt timeout, in CSR units */ u32 rcv_intr_timeout_csr; + u32 freezelen; /* max length of freezemsg */ u64 __iomem *egrtidbase; spinlock_t sendctrl_lock; /* protect changes to SendCtrl */ spinlock_t rcvctrl_lock; /* protect changes to RcvCtrl */ @@ -945,7 +942,6 @@ struct hfi1_devdata { * IB link status cheaply */ struct hfi1_status *status; - u32 freezelen; /* max length of freezemsg */ /* revision register shadow */ u64 revision; @@ -973,6 +969,8 @@ struct hfi1_devdata { u16 rcvegrbufsize_shift; /* both sides of the PCIe link are gen3 capable */ u8 link_gen3_capable; + /* default link down value (poll/sleep) */ + u8 link_default; /* localbus width (1, 2,4,8,16,32) from config space */ u32 lbus_width; /* localbus speed in MHz */ @@ -1008,8 +1006,6 @@ struct hfi1_devdata { u8 hfi1_id; /* implementation code */ u8 icode; - /* default link down value (poll/sleep) */ - u8 link_default; /* vAU of this device */ u8 vau; /* vCU of this device */ @@ -1020,27 +1016,17 @@ struct hfi1_devdata { u16 vl15_init; /* Misc small ints */ - /* Number of physical ports available */ - u8 num_pports; - /* Lowest context number which can be used by user processes */ - u8 first_user_ctxt; u8 n_krcv_queues; u8 qos_shift; - u8 qpn_mask; - u16 rhf_offset; /* offset of RHF within receive header entry */ u16 irev; /* implementation revision */ u16 dc8051_ver; /* 8051 firmware version */ + spinlock_t hfi1_diag_trans_lock; /* protect diag observer ops */ struct platform_config platform_config; struct platform_config_cache pcfg_cache; struct diag_client *diag_client; - spinlock_t hfi1_diag_trans_lock; /* protect diag observer ops */ - - u8 psxmitwait_supported; - /* cycle length of PS* counters in HW (in picoseconds) */ - u16 psxmitwait_check_rate; /* MSI-X information */ struct hfi1_msix_entry *msix_entries; @@ -1055,6 +1041,9 @@ struct hfi1_devdata { struct rcv_array_data rcv_entries; + /* cycle length of PS* counters in HW (in picoseconds) */ + u16 psxmitwait_check_rate; + /* * 64 bit synthetic counters */ @@ -1085,11 +1074,11 @@ struct hfi1_devdata { struct err_info_rcvport err_info_rcvport; struct err_info_constraint err_info_rcv_constraint; struct err_info_constraint err_info_xmit_constraint; - u8 err_info_uncorrectable; - u8 err_info_fmconfig; atomic_t drop_packet; u8 do_drop; + u8 err_info_uncorrectable; + u8 err_info_fmconfig; /* * Software counters for the status bits defined by the @@ -1112,40 +1101,60 @@ struct hfi1_devdata { u64 sw_cce_err_status_aggregate; /* Software counter that aggregates all bypass packet rcv errors */ u64 sw_rcv_bypass_packet_errors; - /* receive interrupt functions */ - rhf_rcv_function_ptr *rhf_rcv_function_map; + /* receive interrupt function */ rhf_rcv_function_ptr normal_rhf_rcv_functions[8]; + /* Save the enabled LCB error bits */ + u64 lcb_err_en; + /* * Capability to have different send engines simply by changing a * pointer value. */ - send_routine process_pio_send; + send_routine process_pio_send ____cacheline_aligned_in_smp; send_routine process_dma_send; void (*pio_inline_send)(struct hfi1_devdata *dd, struct pio_buf *pbuf, u64 pbc, const void *from, size_t count); + /* hfi1_pportdata, points to array of (physical) port-specific + * data structs, indexed by pidx (0..n-1) + */ + struct hfi1_pportdata *pport; + /* receive context data */ + struct hfi1_ctxtdata **rcd; + u64 __percpu *int_counter; + /* device (not port) flags, basically device capabilities */ + u16 flags; + /* Number of physical ports available */ + u8 num_pports; + /* Lowest context number which can be used by user processes */ + u8 first_user_ctxt; + /* adding a new field here would make it part of this cacheline */ + + /* seqlock for sc2vl */ + seqlock_t sc2vl_lock ____cacheline_aligned_in_smp; + u64 sc2vl[4]; + /* receive interrupt functions */ + rhf_rcv_function_ptr *rhf_rcv_function_map; + u64 __percpu *rcv_limit; + u16 rhf_offset; /* offset of RHF within receive header entry */ + /* adding a new field here would make it part of this cacheline */ /* OUI comes from the HW. Used everywhere as 3 separate bytes. */ u8 oui1; u8 oui2; u8 oui3; + u8 dc_shutdown; + /* Timer and counter used to detect RcvBufOvflCnt changes */ struct timer_list rcverr_timer; - u32 rcv_ovfl_cnt; wait_queue_head_t event_queue; - /* Save the enabled LCB error bits */ - u64 lcb_err_en; - u8 dc_shutdown; - /* receive context tail dummy address */ __le64 *rcvhdrtail_dummy_kvaddr; dma_addr_t rcvhdrtail_dummy_dma; - bool eprom_available; /* true if EPROM is available for this device */ - bool aspm_supported; /* Does HW support ASPM */ - bool aspm_enabled; /* ASPM state: enabled/disabled */ + u32 rcv_ovfl_cnt; /* Serialize ASPM enable/disable between multiple verbs contexts */ spinlock_t aspm_lock; /* Number of verbs contexts which have disabled ASPM */ @@ -1155,8 +1164,11 @@ struct hfi1_devdata { /* Used to wait for outstanding user space clients before dev removal */ struct completion user_comp; - struct hfi1_affinity *affinity; + bool eprom_available; /* true if EPROM is available for this device */ + bool aspm_supported; /* Does HW support ASPM */ + bool aspm_enabled; /* ASPM state: enabled/disabled */ struct rhashtable sdma_rht; + struct kobject kobj; }; @@ -1571,6 +1583,11 @@ static inline struct hfi1_ibport *to_iport(struct ib_device *ibdev, u8 port) return &dd->pport[pidx].ibport_data; } +static inline struct hfi1_ibport *rcd_to_iport(struct hfi1_ctxtdata *rcd) +{ + return &rcd->ppd->ibport_data; +} + void hfi1_process_ecn_slowpath(struct rvt_qp *qp, struct hfi1_packet *pkt, bool do_cnp); static inline bool process_ecn(struct rvt_qp *qp, struct hfi1_packet *pkt, @@ -1604,6 +1621,17 @@ static inline u16 hfi1_get_pkey(struct hfi1_ibport *ibp, unsigned index) } /* + * Return the indexed GUID from the port GUIDs table. + */ +static inline __be64 get_sguid(struct hfi1_ibport *ibp, unsigned int index) +{ + struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); + + WARN_ON(index >= HFI1_GUIDS_PER_PORT); + return cpu_to_be64(ppd->guids[index]); +} + +/* * Called by readers of cc_state only, must call under rcu_read_lock(). */ static inline struct cc_state *get_cc_state(struct hfi1_pportdata *ppd) @@ -1769,8 +1797,6 @@ int kdeth_process_expected(struct hfi1_packet *packet); int kdeth_process_eager(struct hfi1_packet *packet); int process_receive_invalid(struct hfi1_packet *packet); -void update_sge(struct rvt_sge_state *ss, u32 length); - /* global module parameter variables */ extern unsigned int hfi1_max_mtu; extern unsigned int hfi1_cu; @@ -1916,6 +1942,10 @@ static inline u64 hfi1_pkt_base_sdma_integrity(struct hfi1_devdata *dd) dev_info(&(dd)->pcidev->dev, "%s: " fmt, \ get_unit_name((dd)->unit), ##__VA_ARGS__) +#define dd_dev_info_ratelimited(dd, fmt, ...) \ + dev_info_ratelimited(&(dd)->pcidev->dev, "%s: " fmt, \ + get_unit_name((dd)->unit), ##__VA_ARGS__) + #define dd_dev_dbg(dd, fmt, ...) \ dev_dbg(&(dd)->pcidev->dev, "%s: " fmt, \ get_unit_name((dd)->unit), ##__VA_ARGS__) @@ -1982,6 +2012,12 @@ static inline u32 qsfp_resource(struct hfi1_devdata *dd) return i2c_target(dd->hfi1_id); } +/* Is this device integrated or discrete? */ +static inline bool is_integrated(struct hfi1_devdata *dd) +{ + return dd->pcidev->device == PCI_DEVICE_ID_INTEL1; +} + int hfi1_tempsense_rd(struct hfi1_devdata *dd, struct hfi1_temp *temp); #define DD_DEV_ENTRY(dd) __string(dev, dev_name(&(dd)->pcidev->dev)) diff --git a/drivers/infiniband/hw/hfi1/init.c b/drivers/infiniband/hw/hfi1/init.c index e3b5bc93bc70..f40864e9a3b2 100644 --- a/drivers/infiniband/hw/hfi1/init.c +++ b/drivers/infiniband/hw/hfi1/init.c @@ -297,14 +297,15 @@ struct hfi1_ctxtdata *hfi1_create_ctxtdata(struct hfi1_pportdata *ppd, u32 ctxt, * The resulting value will be rounded down to the closest * multiple of dd->rcv_entries.group_size. */ - rcd->egrbufs.buffers = kcalloc(rcd->egrbufs.count, - sizeof(*rcd->egrbufs.buffers), - GFP_KERNEL); + rcd->egrbufs.buffers = kzalloc_node( + rcd->egrbufs.count * sizeof(*rcd->egrbufs.buffers), + GFP_KERNEL, numa); if (!rcd->egrbufs.buffers) goto bail; - rcd->egrbufs.rcvtids = kcalloc(rcd->egrbufs.count, - sizeof(*rcd->egrbufs.rcvtids), - GFP_KERNEL); + rcd->egrbufs.rcvtids = kzalloc_node( + rcd->egrbufs.count * + sizeof(*rcd->egrbufs.rcvtids), + GFP_KERNEL, numa); if (!rcd->egrbufs.rcvtids) goto bail; rcd->egrbufs.size = eager_buffer_size; @@ -322,8 +323,8 @@ struct hfi1_ctxtdata *hfi1_create_ctxtdata(struct hfi1_pportdata *ppd, u32 ctxt, rcd->egrbufs.rcvtid_size = HFI1_MAX_EAGER_BUFFER_SIZE; if (ctxt < dd->first_user_ctxt) { /* N/A for PSM contexts */ - rcd->opstats = kzalloc(sizeof(*rcd->opstats), - GFP_KERNEL); + rcd->opstats = kzalloc_node(sizeof(*rcd->opstats), + GFP_KERNEL, numa); if (!rcd->opstats) goto bail; } diff --git a/drivers/infiniband/hw/hfi1/iowait.h b/drivers/infiniband/hw/hfi1/iowait.h index 2ec6ef38d389..d9740ddea6f1 100644 --- a/drivers/infiniband/hw/hfi1/iowait.h +++ b/drivers/infiniband/hw/hfi1/iowait.h @@ -64,6 +64,7 @@ struct sdma_engine; /** * struct iowait - linkage for delayed progress/waiting * @list: used to add/insert into QP/PQ wait lists + * @lock: uses to record the list head lock * @tx_head: overflow list of sdma_txreq's * @sleep: no space callback * @wakeup: space callback wakeup @@ -91,6 +92,11 @@ struct sdma_engine; * so sleeping is not allowed. * * The wait_dma member along with the iow + * + * The lock field is used by waiters to record + * the seqlock_t that guards the list head. + * Waiters explicity know that, but the destroy + * code that unwaits QPs does not. */ struct iowait { @@ -103,6 +109,7 @@ struct iowait { unsigned seq); void (*wakeup)(struct iowait *wait, int reason); void (*sdma_drained)(struct iowait *wait); + seqlock_t *lock; struct work_struct iowork; wait_queue_head_t wait_dma; wait_queue_head_t wait_pio; @@ -141,6 +148,7 @@ static inline void iowait_init( void (*sdma_drained)(struct iowait *wait)) { wait->count = 0; + wait->lock = NULL; INIT_LIST_HEAD(&wait->list); INIT_LIST_HEAD(&wait->tx_head); INIT_WORK(&wait->iowork, func); diff --git a/drivers/infiniband/hw/hfi1/mad.c b/drivers/infiniband/hw/hfi1/mad.c index 9487c9bb8920..09cda3c35e82 100644 --- a/drivers/infiniband/hw/hfi1/mad.c +++ b/drivers/infiniband/hw/hfi1/mad.c @@ -128,7 +128,7 @@ static void send_trap(struct hfi1_ibport *ibp, void *data, unsigned len) smp = send_buf->mad; smp->base_version = OPA_MGMT_BASE_VERSION; smp->mgmt_class = IB_MGMT_CLASS_SUBN_LID_ROUTED; - smp->class_version = OPA_SMI_CLASS_VERSION; + smp->class_version = OPA_SM_CLASS_VERSION; smp->method = IB_MGMT_METHOD_TRAP; ibp->rvp.tid++; smp->tid = cpu_to_be64(ibp->rvp.tid); @@ -336,20 +336,20 @@ static int __subn_get_opa_nodeinfo(struct opa_smp *smp, u32 am, u8 *data, ni = (struct opa_node_info *)data; /* GUID 0 is illegal */ - if (am || pidx >= dd->num_pports || dd->pport[pidx].guid == 0) { + if (am || pidx >= dd->num_pports || ibdev->node_guid == 0 || + get_sguid(to_iport(ibdev, port), HFI1_PORT_GUID_INDEX) == 0) { smp->status |= IB_SMP_INVALID_FIELD; return reply((struct ib_mad_hdr *)smp); } - ni->port_guid = cpu_to_be64(dd->pport[pidx].guid); + ni->port_guid = get_sguid(to_iport(ibdev, port), HFI1_PORT_GUID_INDEX); ni->base_version = OPA_MGMT_BASE_VERSION; - ni->class_version = OPA_SMI_CLASS_VERSION; + ni->class_version = OPA_SM_CLASS_VERSION; ni->node_type = 1; /* channel adapter */ ni->num_ports = ibdev->phys_port_cnt; /* This is already in network order */ ni->system_image_guid = ib_hfi1_sys_image_guid; - /* Use first-port GUID as node */ - ni->node_guid = cpu_to_be64(dd->pport->guid); + ni->node_guid = ibdev->node_guid; ni->partition_cap = cpu_to_be16(hfi1_get_npkeys(dd)); ni->device_id = cpu_to_be16(dd->pcidev->device); ni->revision = cpu_to_be32(dd->minrev); @@ -373,19 +373,20 @@ static int subn_get_nodeinfo(struct ib_smp *smp, struct ib_device *ibdev, /* GUID 0 is illegal */ if (smp->attr_mod || pidx >= dd->num_pports || - dd->pport[pidx].guid == 0) + ibdev->node_guid == 0 || + get_sguid(to_iport(ibdev, port), HFI1_PORT_GUID_INDEX) == 0) { smp->status |= IB_SMP_INVALID_FIELD; - else - nip->port_guid = cpu_to_be64(dd->pport[pidx].guid); + return reply((struct ib_mad_hdr *)smp); + } + nip->port_guid = get_sguid(to_iport(ibdev, port), HFI1_PORT_GUID_INDEX); nip->base_version = OPA_MGMT_BASE_VERSION; - nip->class_version = OPA_SMI_CLASS_VERSION; + nip->class_version = OPA_SM_CLASS_VERSION; nip->node_type = 1; /* channel adapter */ nip->num_ports = ibdev->phys_port_cnt; /* This is already in network order */ nip->sys_guid = ib_hfi1_sys_image_guid; - /* Use first-port GUID as node */ - nip->node_guid = cpu_to_be64(dd->pport->guid); + nip->node_guid = ibdev->node_guid; nip->partition_cap = cpu_to_be16(hfi1_get_npkeys(dd)); nip->device_id = cpu_to_be16(dd->pcidev->device); nip->revision = cpu_to_be32(dd->minrev); @@ -2302,7 +2303,7 @@ static int pma_get_opa_classportinfo(struct opa_pma_mad *pmp, pmp->mad_hdr.status |= IB_SMP_INVALID_FIELD; p->base_version = OPA_MGMT_BASE_VERSION; - p->class_version = OPA_SMI_CLASS_VERSION; + p->class_version = OPA_SM_CLASS_VERSION; /* * Expected response time is 4.096 usec. * 2^18 == 1.073741824 sec. */ @@ -4022,7 +4023,7 @@ static int process_subn_opa(struct ib_device *ibdev, int mad_flags, am = be32_to_cpu(smp->attr_mod); attr_id = smp->attr_id; - if (smp->class_version != OPA_SMI_CLASS_VERSION) { + if (smp->class_version != OPA_SM_CLASS_VERSION) { smp->status |= IB_SMP_UNSUP_VERSION; ret = reply((struct ib_mad_hdr *)smp); return ret; @@ -4232,7 +4233,7 @@ static int process_perf_opa(struct ib_device *ibdev, u8 port, *out_mad = *in_mad; - if (pmp->mad_hdr.class_version != OPA_SMI_CLASS_VERSION) { + if (pmp->mad_hdr.class_version != OPA_SM_CLASS_VERSION) { pmp->mad_hdr.status |= IB_SMP_UNSUP_VERSION; return reply((struct ib_mad_hdr *)pmp); } @@ -4405,7 +4406,7 @@ int hfi1_process_mad(struct ib_device *ibdev, int mad_flags, u8 port, switch (in_mad->base_version) { case OPA_MGMT_BASE_VERSION: if (unlikely(in_mad_size != sizeof(struct opa_mad))) { - dev_err(ibdev->dma_device, "invalid in_mad_size\n"); + dev_err(ibdev->dev.parent, "invalid in_mad_size\n"); return IB_MAD_RESULT_FAILURE; } return hfi1_process_opa_mad(ibdev, mad_flags, port, diff --git a/drivers/infiniband/hw/hfi1/mmu_rb.c b/drivers/infiniband/hw/hfi1/mmu_rb.c index 7ad30898fc19..ccbf52c8ff6f 100644 --- a/drivers/infiniband/hw/hfi1/mmu_rb.c +++ b/drivers/infiniband/hw/hfi1/mmu_rb.c @@ -81,7 +81,7 @@ static void do_remove(struct mmu_rb_handler *handler, struct list_head *del_list); static void handle_remove(struct work_struct *work); -static struct mmu_notifier_ops mn_opts = { +static const struct mmu_notifier_ops mn_opts = { .invalidate_page = mmu_notifier_page, .invalidate_range_start = mmu_notifier_range_start, }; diff --git a/drivers/infiniband/hw/hfi1/pcie.c b/drivers/infiniband/hw/hfi1/pcie.c index 4ac8f330c5cb..0829fce06172 100644 --- a/drivers/infiniband/hw/hfi1/pcie.c +++ b/drivers/infiniband/hw/hfi1/pcie.c @@ -598,15 +598,6 @@ pci_slot_reset(struct pci_dev *pdev) return PCI_ERS_RESULT_CAN_RECOVER; } -static pci_ers_result_t -pci_link_reset(struct pci_dev *pdev) -{ - struct hfi1_devdata *dd = pci_get_drvdata(pdev); - - dd_dev_info(dd, "HFI1 link_reset function called, ignored\n"); - return PCI_ERS_RESULT_CAN_RECOVER; -} - static void pci_resume(struct pci_dev *pdev) { @@ -625,7 +616,6 @@ pci_resume(struct pci_dev *pdev) const struct pci_error_handlers hfi1_pci_err_handler = { .error_detected = pci_error_detected, .mmio_enabled = pci_mmio_enabled, - .link_reset = pci_link_reset, .slot_reset = pci_slot_reset, .resume = pci_resume, }; @@ -673,12 +663,12 @@ MODULE_PARM_DESC(pcie_retry, "Driver will try this many times to reach requested #define UNSET_PSET 255 #define DEFAULT_DISCRETE_PSET 2 /* discrete HFI */ -#define DEFAULT_MCP_PSET 4 /* MCP HFI */ +#define DEFAULT_MCP_PSET 6 /* MCP HFI */ static uint pcie_pset = UNSET_PSET; module_param(pcie_pset, uint, S_IRUGO); MODULE_PARM_DESC(pcie_pset, "PCIe Eq Pset value to use, range is 0-10"); -static uint pcie_ctle = 1; /* discrete on, integrated off */ +static uint pcie_ctle = 3; /* discrete on, integrated on */ module_param(pcie_ctle, uint, S_IRUGO); MODULE_PARM_DESC(pcie_ctle, "PCIe static CTLE mode, bit 0 - discrete on/off, bit 1 - integrated on/off"); diff --git a/drivers/infiniband/hw/hfi1/pio.c b/drivers/infiniband/hw/hfi1/pio.c index d89b8745d4c1..615be68e40b3 100644 --- a/drivers/infiniband/hw/hfi1/pio.c +++ b/drivers/infiniband/hw/hfi1/pio.c @@ -758,6 +758,7 @@ struct send_context *sc_alloc(struct hfi1_devdata *dd, int type, sc->hw_context = hw_context; cr_group_addresses(sc, &dma); sc->credits = sci->credits; + sc->size = sc->credits * PIO_BLOCK_SIZE; /* PIO Send Memory Address details */ #define PIO_ADDR_CONTEXT_MASK 0xfful @@ -1242,6 +1243,7 @@ int sc_enable(struct send_context *sc) sc->free = 0; sc->alloc_free = 0; sc->fill = 0; + sc->fill_wrap = 0; sc->sr_head = 0; sc->sr_tail = 0; sc->flags = 0; @@ -1385,7 +1387,7 @@ struct pio_buf *sc_buffer_alloc(struct send_context *sc, u32 dw_len, unsigned long flags; unsigned long avail; unsigned long blocks = dwords_to_blocks(dw_len); - unsigned long start_fill; + u32 fill_wrap; int trycount = 0; u32 head, next; @@ -1410,9 +1412,7 @@ retry: (sc->fill - sc->alloc_free); if (blocks > avail) { /* still no room, actively update */ - spin_unlock_irqrestore(&sc->alloc_lock, flags); sc_release_update(sc); - spin_lock_irqsave(&sc->alloc_lock, flags); sc->alloc_free = ACCESS_ONCE(sc->free); trycount++; goto retry; @@ -1428,8 +1428,11 @@ retry: head = sc->sr_head; /* "allocate" the buffer */ - start_fill = sc->fill; sc->fill += blocks; + fill_wrap = sc->fill_wrap; + sc->fill_wrap += blocks; + if (sc->fill_wrap >= sc->credits) + sc->fill_wrap = sc->fill_wrap - sc->credits; /* * Fill the parts that the releaser looks at before moving the head. @@ -1458,11 +1461,8 @@ retry: spin_unlock_irqrestore(&sc->alloc_lock, flags); /* finish filling in the buffer outside the lock */ - pbuf->start = sc->base_addr + ((start_fill % sc->credits) - * PIO_BLOCK_SIZE); - pbuf->size = sc->credits * PIO_BLOCK_SIZE; - pbuf->end = sc->base_addr + pbuf->size; - pbuf->block_count = blocks; + pbuf->start = sc->base_addr + fill_wrap * PIO_BLOCK_SIZE; + pbuf->end = sc->base_addr + sc->size; pbuf->qw_written = 0; pbuf->carry_bytes = 0; pbuf->carry.val64 = 0; @@ -1573,6 +1573,7 @@ static void sc_piobufavail(struct send_context *sc) qp = iowait_to_qp(wait); priv = qp->priv; list_del_init(&priv->s_iowait.list); + priv->s_iowait.lock = NULL; /* refcount held until actual wake up */ qps[n++] = qp; } @@ -2028,29 +2029,17 @@ freesc15: int init_credit_return(struct hfi1_devdata *dd) { int ret; - int num_numa; int i; - num_numa = num_online_nodes(); - /* enforce the expectation that the numas are compact */ - for (i = 0; i < num_numa; i++) { - if (!node_online(i)) { - dd_dev_err(dd, "NUMA nodes are not compact\n"); - ret = -EINVAL; - goto done; - } - } - dd->cr_base = kcalloc( - num_numa, + node_affinity.num_possible_nodes, sizeof(struct credit_return_base), GFP_KERNEL); if (!dd->cr_base) { - dd_dev_err(dd, "Unable to allocate credit return base\n"); ret = -ENOMEM; goto done; } - for (i = 0; i < num_numa; i++) { + for_each_node_with_cpus(i) { int bytes = TXE_NUM_CONTEXTS * sizeof(struct credit_return); set_dev_node(&dd->pcidev->dev, i); @@ -2077,14 +2066,11 @@ done: void free_credit_return(struct hfi1_devdata *dd) { - int num_numa; int i; if (!dd->cr_base) return; - - num_numa = num_online_nodes(); - for (i = 0; i < num_numa; i++) { + for (i = 0; i < node_affinity.num_possible_nodes; i++) { if (dd->cr_base[i].va) { dma_free_coherent(&dd->pcidev->dev, TXE_NUM_CONTEXTS * diff --git a/drivers/infiniband/hw/hfi1/pio.h b/drivers/infiniband/hw/hfi1/pio.h index e709eaf743b5..867e5ffc3595 100644 --- a/drivers/infiniband/hw/hfi1/pio.h +++ b/drivers/infiniband/hw/hfi1/pio.h @@ -83,53 +83,55 @@ struct pio_buf { void *arg; /* argument for cb */ void __iomem *start; /* buffer start address */ void __iomem *end; /* context end address */ - unsigned long size; /* context size, in bytes */ unsigned long sent_at; /* buffer is sent when <= free */ - u32 block_count; /* size of buffer, in blocks */ - u32 qw_written; /* QW written so far */ - u32 carry_bytes; /* number of valid bytes in carry */ union mix carry; /* pending unwritten bytes */ + u16 qw_written; /* QW written so far */ + u8 carry_bytes; /* number of valid bytes in carry */ }; /* cache line aligned pio buffer array */ union pio_shadow_ring { struct pio_buf pbuf; - u64 unused[16]; /* cache line spacer */ } ____cacheline_aligned; /* per-NUMA send context */ struct send_context { /* read-only after init */ struct hfi1_devdata *dd; /* device */ - void __iomem *base_addr; /* start of PIO memory */ union pio_shadow_ring *sr; /* shadow ring */ + void __iomem *base_addr; /* start of PIO memory */ + u32 __percpu *buffers_allocated;/* count of buffers allocated */ + u32 size; /* context size, in bytes */ - volatile __le64 *hw_free; /* HW free counter */ - struct work_struct halt_work; /* halted context work queue entry */ - unsigned long flags; /* flags */ int node; /* context home node */ - int type; /* context type */ - u32 sw_index; /* software index number */ - u32 hw_context; /* hardware context number */ - u32 credits; /* number of blocks in context */ u32 sr_size; /* size of the shadow ring */ - u32 group; /* credit return group */ + u16 flags; /* flags */ + u8 type; /* context type */ + u8 sw_index; /* software index number */ + u8 hw_context; /* hardware context number */ + u8 group; /* credit return group */ + /* allocator fields */ spinlock_t alloc_lock ____cacheline_aligned_in_smp; + u32 sr_head; /* shadow ring head */ unsigned long fill; /* official alloc count */ unsigned long alloc_free; /* copy of free (less cache thrash) */ - u32 sr_head; /* shadow ring head */ + u32 fill_wrap; /* tracks fill within ring */ + u32 credits; /* number of blocks in context */ + /* adding a new field here would make it part of this cacheline */ + /* releaser fields */ spinlock_t release_lock ____cacheline_aligned_in_smp; - unsigned long free; /* official free count */ u32 sr_tail; /* shadow ring tail */ + unsigned long free; /* official free count */ + volatile __le64 *hw_free; /* HW free counter */ /* list for PIO waiters */ struct list_head piowait ____cacheline_aligned_in_smp; spinlock_t credit_ctrl_lock ____cacheline_aligned_in_smp; - u64 credit_ctrl; /* cache for credit control */ u32 credit_intr_count; /* count of credit intr users */ - u32 __percpu *buffers_allocated;/* count of buffers allocated */ + u64 credit_ctrl; /* cache for credit control */ wait_queue_head_t halt_wait; /* wait until kernel sees interrupt */ + struct work_struct halt_work; /* halted context work queue entry */ }; /* send context flags */ diff --git a/drivers/infiniband/hw/hfi1/pio_copy.c b/drivers/infiniband/hw/hfi1/pio_copy.c index aa7773643107..03024cec78dd 100644 --- a/drivers/infiniband/hw/hfi1/pio_copy.c +++ b/drivers/infiniband/hw/hfi1/pio_copy.c @@ -129,8 +129,8 @@ void pio_copy(struct hfi1_devdata *dd, struct pio_buf *pbuf, u64 pbc, dest += sizeof(u64); } - dest -= pbuf->size; - dend -= pbuf->size; + dest -= pbuf->sc->size; + dend -= pbuf->sc->size; } /* write 8-byte non-SOP, non-wrap chunk data */ @@ -361,8 +361,8 @@ void seg_pio_copy_start(struct pio_buf *pbuf, u64 pbc, dest += sizeof(u64); } - dest -= pbuf->size; - dend -= pbuf->size; + dest -= pbuf->sc->size; + dend -= pbuf->sc->size; } /* write 8-byte non-SOP, non-wrap chunk data */ @@ -458,8 +458,8 @@ static void mid_copy_mix(struct pio_buf *pbuf, const void *from, size_t nbytes) dest += sizeof(u64); } - dest -= pbuf->size; - dend -= pbuf->size; + dest -= pbuf->sc->size; + dend -= pbuf->sc->size; } /* write 8-byte non-SOP, non-wrap chunk data */ @@ -492,7 +492,7 @@ static void mid_copy_mix(struct pio_buf *pbuf, const void *from, size_t nbytes) */ /* adjust if we have wrapped */ if (dest >= pbuf->end) - dest -= pbuf->size; + dest -= pbuf->sc->size; /* jump to the SOP range if within the first block */ else if (pbuf->qw_written < PIO_BLOCK_QWS) dest += SOP_DISTANCE; @@ -584,8 +584,8 @@ static void mid_copy_straight(struct pio_buf *pbuf, dest += sizeof(u64); } - dest -= pbuf->size; - dend -= pbuf->size; + dest -= pbuf->sc->size; + dend -= pbuf->sc->size; } /* write 8-byte non-SOP, non-wrap chunk data */ @@ -666,7 +666,7 @@ void seg_pio_copy_mid(struct pio_buf *pbuf, const void *from, size_t nbytes) */ /* adjust if we've wrapped */ if (dest >= pbuf->end) - dest -= pbuf->size; + dest -= pbuf->sc->size; /* jump to SOP range if within the first block */ else if (pbuf->qw_written < PIO_BLOCK_QWS) dest += SOP_DISTANCE; @@ -719,7 +719,7 @@ void seg_pio_copy_end(struct pio_buf *pbuf) */ /* adjust if we have wrapped */ if (dest >= pbuf->end) - dest -= pbuf->size; + dest -= pbuf->sc->size; /* jump to the SOP range if within the first block */ else if (pbuf->qw_written < PIO_BLOCK_QWS) dest += SOP_DISTANCE; diff --git a/drivers/infiniband/hw/hfi1/platform.c b/drivers/infiniband/hw/hfi1/platform.c index 202433178864..838fe84e285a 100644 --- a/drivers/infiniband/hw/hfi1/platform.c +++ b/drivers/infiniband/hw/hfi1/platform.c @@ -49,6 +49,90 @@ #include "efivar.h" #include "eprom.h" +static int validate_scratch_checksum(struct hfi1_devdata *dd) +{ + u64 checksum = 0, temp_scratch = 0; + int i, j, version; + + temp_scratch = read_csr(dd, ASIC_CFG_SCRATCH); + version = (temp_scratch & BITMAP_VERSION_SMASK) >> BITMAP_VERSION_SHIFT; + + /* Prevent power on default of all zeroes from passing checksum */ + if (!version) + return 0; + + /* + * ASIC scratch 0 only contains the checksum and bitmap version as + * fields of interest, both of which are handled separately from the + * loop below, so skip it + */ + checksum += version; + for (i = 1; i < ASIC_NUM_SCRATCH; i++) { + temp_scratch = read_csr(dd, ASIC_CFG_SCRATCH + (8 * i)); + for (j = sizeof(u64); j != 0; j -= 2) { + checksum += (temp_scratch & 0xFFFF); + temp_scratch >>= 16; + } + } + + while (checksum >> 16) + checksum = (checksum & CHECKSUM_MASK) + (checksum >> 16); + + temp_scratch = read_csr(dd, ASIC_CFG_SCRATCH); + temp_scratch &= CHECKSUM_SMASK; + temp_scratch >>= CHECKSUM_SHIFT; + + if (checksum + temp_scratch == 0xFFFF) + return 1; + return 0; +} + +static void save_platform_config_fields(struct hfi1_devdata *dd) +{ + struct hfi1_pportdata *ppd = dd->pport; + u64 temp_scratch = 0, temp_dest = 0; + + temp_scratch = read_csr(dd, ASIC_CFG_SCRATCH_1); + + temp_dest = temp_scratch & + (dd->hfi1_id ? PORT1_PORT_TYPE_SMASK : + PORT0_PORT_TYPE_SMASK); + ppd->port_type = temp_dest >> + (dd->hfi1_id ? PORT1_PORT_TYPE_SHIFT : + PORT0_PORT_TYPE_SHIFT); + + temp_dest = temp_scratch & + (dd->hfi1_id ? PORT1_LOCAL_ATTEN_SMASK : + PORT0_LOCAL_ATTEN_SMASK); + ppd->local_atten = temp_dest >> + (dd->hfi1_id ? PORT1_LOCAL_ATTEN_SHIFT : + PORT0_LOCAL_ATTEN_SHIFT); + + temp_dest = temp_scratch & + (dd->hfi1_id ? PORT1_REMOTE_ATTEN_SMASK : + PORT0_REMOTE_ATTEN_SMASK); + ppd->remote_atten = temp_dest >> + (dd->hfi1_id ? PORT1_REMOTE_ATTEN_SHIFT : + PORT0_REMOTE_ATTEN_SHIFT); + + temp_dest = temp_scratch & + (dd->hfi1_id ? PORT1_DEFAULT_ATTEN_SMASK : + PORT0_DEFAULT_ATTEN_SMASK); + ppd->default_atten = temp_dest >> + (dd->hfi1_id ? PORT1_DEFAULT_ATTEN_SHIFT : + PORT0_DEFAULT_ATTEN_SHIFT); + + temp_scratch = read_csr(dd, dd->hfi1_id ? ASIC_CFG_SCRATCH_3 : + ASIC_CFG_SCRATCH_2); + + ppd->tx_preset_eq = (temp_scratch & TX_EQ_SMASK) >> TX_EQ_SHIFT; + ppd->tx_preset_noeq = (temp_scratch & TX_NO_EQ_SMASK) >> TX_NO_EQ_SHIFT; + ppd->rx_preset = (temp_scratch & RX_SMASK) >> RX_SHIFT; + + ppd->max_power_class = (temp_scratch & QSFP_MAX_POWER_SMASK) >> + QSFP_MAX_POWER_SHIFT; +} + void get_platform_config(struct hfi1_devdata *dd) { int ret = 0; @@ -56,38 +140,49 @@ void get_platform_config(struct hfi1_devdata *dd) u8 *temp_platform_config = NULL; u32 esize; - ret = eprom_read_platform_config(dd, (void **)&temp_platform_config, - &esize); - if (!ret) { - /* success */ - size = esize; - goto success; + if (is_integrated(dd)) { + if (validate_scratch_checksum(dd)) { + save_platform_config_fields(dd); + return; + } + dd_dev_err(dd, "%s: Config bitmap corrupted/uninitialized\n", + __func__); + dd_dev_err(dd, + "%s: Please update your BIOS to support active channels\n", + __func__); + } else { + ret = eprom_read_platform_config(dd, + (void **)&temp_platform_config, + &esize); + if (!ret) { + /* success */ + dd->platform_config.data = temp_platform_config; + dd->platform_config.size = esize; + return; + } + /* fail, try EFI variable */ + + ret = read_hfi1_efi_var(dd, "configuration", &size, + (void **)&temp_platform_config); + if (!ret) { + dd->platform_config.data = temp_platform_config; + dd->platform_config.size = size; + return; + } } - /* fail, try EFI variable */ - - ret = read_hfi1_efi_var(dd, "configuration", &size, - (void **)&temp_platform_config); - if (!ret) - goto success; - - dd_dev_info(dd, - "%s: Failed to get platform config from UEFI, falling back to request firmware\n", - __func__); + dd_dev_err(dd, + "%s: Failed to get platform config, falling back to sub-optimal default file\n", + __func__); /* fall back to request firmware */ platform_config_load = 1; - return; - -success: - dd->platform_config.data = temp_platform_config; - dd->platform_config.size = size; } void free_platform_config(struct hfi1_devdata *dd) { if (!platform_config_load) { /* - * was loaded from EFI, release memory - * allocated by read_efi_var + * was loaded from EFI or the EPROM, release memory + * allocated by read_efi_var/eprom_read_platform_config */ kfree(dd->platform_config.data); } @@ -100,12 +195,16 @@ void free_platform_config(struct hfi1_devdata *dd) void get_port_type(struct hfi1_pportdata *ppd) { int ret; + u32 temp; ret = get_platform_config_field(ppd->dd, PLATFORM_CONFIG_PORT_TABLE, 0, - PORT_TABLE_PORT_TYPE, &ppd->port_type, + PORT_TABLE_PORT_TYPE, &temp, 4); - if (ret) + if (ret) { ppd->port_type = PORT_TYPE_UNKNOWN; + return; + } + ppd->port_type = temp; } int set_qsfp_tx(struct hfi1_pportdata *ppd, int on) @@ -538,6 +637,38 @@ static void apply_tx_lanes(struct hfi1_pportdata *ppd, u8 field_id, } } +/* + * Return a special SerDes setting for low power AOC cables. The power class + * threshold and setting being used were all found by empirical testing. + * + * Summary of the logic: + * + * if (QSFP and QSFP_TYPE == AOC and QSFP_POWER_CLASS < 4) + * return 0xe + * return 0; // leave at default + */ +static u8 aoc_low_power_setting(struct hfi1_pportdata *ppd) +{ + u8 *cache = ppd->qsfp_info.cache; + int power_class; + + /* QSFP only */ + if (ppd->port_type != PORT_TYPE_QSFP) + return 0; /* leave at default */ + + /* active optical cables only */ + switch ((cache[QSFP_MOD_TECH_OFFS] & 0xF0) >> 4) { + case 0x0 ... 0x9: /* fallthrough */ + case 0xC: /* fallthrough */ + case 0xE: + /* active AOC */ + power_class = get_qsfp_power_class(cache[QSFP_MOD_PWR_OFFS]); + if (power_class < QSFP_POWER_CLASS_4) + return 0xe; + } + return 0; /* leave at default */ +} + static void apply_tunings( struct hfi1_pportdata *ppd, u32 tx_preset_index, u8 tuning_method, u32 total_atten, u8 limiting_active) @@ -606,7 +737,17 @@ static void apply_tunings( tx_preset_index, TX_PRESET_TABLE_POSTCUR, &tx_preset, 4); postcur = tx_preset; - config_data = precur | (attn << 8) | (postcur << 16); + /* + * NOTES: + * o The aoc_low_power_setting is applied to all lanes even + * though only lane 0's value is examined by the firmware. + * o A lingering low power setting after a cable swap does + * not occur. On cable unplug the 8051 is reset and + * restarted on cable insert. This resets all settings to + * their default, erasing any previous low power setting. + */ + config_data = precur | (attn << 8) | (postcur << 16) | + (aoc_low_power_setting(ppd) << 24); apply_tx_lanes(ppd, TX_EQ_SETTINGS, config_data, "Applying TX settings"); diff --git a/drivers/infiniband/hw/hfi1/platform.h b/drivers/infiniband/hw/hfi1/platform.h index e2c21613c326..eed0aa9124fa 100644 --- a/drivers/infiniband/hw/hfi1/platform.h +++ b/drivers/infiniband/hw/hfi1/platform.h @@ -168,16 +168,6 @@ struct platform_config_cache { struct platform_config_data config_tables[PLATFORM_CONFIG_TABLE_MAX]; }; -static const u32 platform_config_table_limits[PLATFORM_CONFIG_TABLE_MAX] = { - 0, - SYSTEM_TABLE_MAX, - PORT_TABLE_MAX, - RX_PRESET_TABLE_MAX, - TX_PRESET_TABLE_MAX, - QSFP_ATTEN_TABLE_MAX, - VARIABLE_SETTINGS_TABLE_MAX -}; - /* This section defines default values and encodings for the * fields defined for each table above */ @@ -295,6 +285,123 @@ enum link_tuning_encoding { OPA_UNKNOWN_TUNING }; +/* + * Shifts and masks for the link SI tuning values stuffed into the ASIC scratch + * registers for integrated platforms + */ +#define PORT0_PORT_TYPE_SHIFT 0 +#define PORT0_LOCAL_ATTEN_SHIFT 4 +#define PORT0_REMOTE_ATTEN_SHIFT 10 +#define PORT0_DEFAULT_ATTEN_SHIFT 32 + +#define PORT1_PORT_TYPE_SHIFT 16 +#define PORT1_LOCAL_ATTEN_SHIFT 20 +#define PORT1_REMOTE_ATTEN_SHIFT 26 +#define PORT1_DEFAULT_ATTEN_SHIFT 40 + +#define PORT0_PORT_TYPE_MASK 0xFUL +#define PORT0_LOCAL_ATTEN_MASK 0x3FUL +#define PORT0_REMOTE_ATTEN_MASK 0x3FUL +#define PORT0_DEFAULT_ATTEN_MASK 0xFFUL + +#define PORT1_PORT_TYPE_MASK 0xFUL +#define PORT1_LOCAL_ATTEN_MASK 0x3FUL +#define PORT1_REMOTE_ATTEN_MASK 0x3FUL +#define PORT1_DEFAULT_ATTEN_MASK 0xFFUL + +#define PORT0_PORT_TYPE_SMASK (PORT0_PORT_TYPE_MASK << \ + PORT0_PORT_TYPE_SHIFT) +#define PORT0_LOCAL_ATTEN_SMASK (PORT0_LOCAL_ATTEN_MASK << \ + PORT0_LOCAL_ATTEN_SHIFT) +#define PORT0_REMOTE_ATTEN_SMASK (PORT0_REMOTE_ATTEN_MASK << \ + PORT0_REMOTE_ATTEN_SHIFT) +#define PORT0_DEFAULT_ATTEN_SMASK (PORT0_DEFAULT_ATTEN_MASK << \ + PORT0_DEFAULT_ATTEN_SHIFT) + +#define PORT1_PORT_TYPE_SMASK (PORT1_PORT_TYPE_MASK << \ + PORT1_PORT_TYPE_SHIFT) +#define PORT1_LOCAL_ATTEN_SMASK (PORT1_LOCAL_ATTEN_MASK << \ + PORT1_LOCAL_ATTEN_SHIFT) +#define PORT1_REMOTE_ATTEN_SMASK (PORT1_REMOTE_ATTEN_MASK << \ + PORT1_REMOTE_ATTEN_SHIFT) +#define PORT1_DEFAULT_ATTEN_SMASK (PORT1_DEFAULT_ATTEN_MASK << \ + PORT1_DEFAULT_ATTEN_SHIFT) + +#define QSFP_MAX_POWER_SHIFT 0 +#define TX_NO_EQ_SHIFT 4 +#define TX_EQ_SHIFT 25 +#define RX_SHIFT 46 + +#define QSFP_MAX_POWER_MASK 0xFUL +#define TX_NO_EQ_MASK 0x1FFFFFUL +#define TX_EQ_MASK 0x1FFFFFUL +#define RX_MASK 0xFFFFUL + +#define QSFP_MAX_POWER_SMASK (QSFP_MAX_POWER_MASK << \ + QSFP_MAX_POWER_SHIFT) +#define TX_NO_EQ_SMASK (TX_NO_EQ_MASK << TX_NO_EQ_SHIFT) +#define TX_EQ_SMASK (TX_EQ_MASK << TX_EQ_SHIFT) +#define RX_SMASK (RX_MASK << RX_SHIFT) + +#define TX_PRECUR_SHIFT 0 +#define TX_ATTN_SHIFT 4 +#define QSFP_TX_CDR_APPLY_SHIFT 9 +#define QSFP_TX_EQ_APPLY_SHIFT 10 +#define QSFP_TX_CDR_SHIFT 11 +#define QSFP_TX_EQ_SHIFT 12 +#define TX_POSTCUR_SHIFT 16 + +#define TX_PRECUR_MASK 0xFUL +#define TX_ATTN_MASK 0x1FUL +#define QSFP_TX_CDR_APPLY_MASK 0x1UL +#define QSFP_TX_EQ_APPLY_MASK 0x1UL +#define QSFP_TX_CDR_MASK 0x1UL +#define QSFP_TX_EQ_MASK 0xFUL +#define TX_POSTCUR_MASK 0x1FUL + +#define TX_PRECUR_SMASK (TX_PRECUR_MASK << TX_PRECUR_SHIFT) +#define TX_ATTN_SMASK (TX_ATTN_MASK << TX_ATTN_SHIFT) +#define QSFP_TX_CDR_APPLY_SMASK (QSFP_TX_CDR_APPLY_MASK << \ + QSFP_TX_CDR_APPLY_SHIFT) +#define QSFP_TX_EQ_APPLY_SMASK (QSFP_TX_EQ_APPLY_MASK << \ + QSFP_TX_EQ_APPLY_SHIFT) +#define QSFP_TX_CDR_SMASK (QSFP_TX_CDR_MASK << QSFP_TX_CDR_SHIFT) +#define QSFP_TX_EQ_SMASK (QSFP_TX_EQ_MASK << QSFP_TX_EQ_SHIFT) +#define TX_POSTCUR_SMASK (TX_POSTCUR_MASK << TX_POSTCUR_SHIFT) + +#define QSFP_RX_CDR_APPLY_SHIFT 0 +#define QSFP_RX_EMP_APPLY_SHIFT 1 +#define QSFP_RX_AMP_APPLY_SHIFT 2 +#define QSFP_RX_CDR_SHIFT 3 +#define QSFP_RX_EMP_SHIFT 4 +#define QSFP_RX_AMP_SHIFT 8 + +#define QSFP_RX_CDR_APPLY_MASK 0x1UL +#define QSFP_RX_EMP_APPLY_MASK 0x1UL +#define QSFP_RX_AMP_APPLY_MASK 0x1UL +#define QSFP_RX_CDR_MASK 0x1UL +#define QSFP_RX_EMP_MASK 0xFUL +#define QSFP_RX_AMP_MASK 0x3UL + +#define QSFP_RX_CDR_APPLY_SMASK (QSFP_RX_CDR_APPLY_MASK << \ + QSFP_RX_CDR_APPLY_SHIFT) +#define QSFP_RX_EMP_APPLY_SMASK (QSFP_RX_EMP_APPLY_MASK << \ + QSFP_RX_EMP_APPLY_SHIFT) +#define QSFP_RX_AMP_APPLY_SMASK (QSFP_RX_AMP_APPLY_MASK << \ + QSFP_RX_AMP_APPLY_SHIFT) +#define QSFP_RX_CDR_SMASK (QSFP_RX_CDR_MASK << QSFP_RX_CDR_SHIFT) +#define QSFP_RX_EMP_SMASK (QSFP_RX_EMP_MASK << QSFP_RX_EMP_SHIFT) +#define QSFP_RX_AMP_SMASK (QSFP_RX_AMP_MASK << QSFP_RX_AMP_SHIFT) + +#define BITMAP_VERSION 1 +#define BITMAP_VERSION_SHIFT 44 +#define BITMAP_VERSION_MASK 0xFUL +#define BITMAP_VERSION_SMASK (BITMAP_VERSION_MASK << \ + BITMAP_VERSION_SHIFT) +#define CHECKSUM_SHIFT 48 +#define CHECKSUM_MASK 0xFFFFUL +#define CHECKSUM_SMASK (CHECKSUM_MASK << CHECKSUM_SHIFT) + /* platform.c */ void get_platform_config(struct hfi1_devdata *dd); void free_platform_config(struct hfi1_devdata *dd); diff --git a/drivers/infiniband/hw/hfi1/qp.c b/drivers/infiniband/hw/hfi1/qp.c index 9fc75e7e8781..c4ebd097b20f 100644 --- a/drivers/infiniband/hw/hfi1/qp.c +++ b/drivers/infiniband/hw/hfi1/qp.c @@ -79,43 +79,6 @@ static inline unsigned mk_qpn(struct rvt_qpn_table *qpt, return (map - qpt->map) * RVT_BITS_PER_PAGE + off; } -/* - * Convert the AETH credit code into the number of credits. - */ -static const u16 credit_table[31] = { - 0, /* 0 */ - 1, /* 1 */ - 2, /* 2 */ - 3, /* 3 */ - 4, /* 4 */ - 6, /* 5 */ - 8, /* 6 */ - 12, /* 7 */ - 16, /* 8 */ - 24, /* 9 */ - 32, /* A */ - 48, /* B */ - 64, /* C */ - 96, /* D */ - 128, /* E */ - 192, /* F */ - 256, /* 10 */ - 384, /* 11 */ - 512, /* 12 */ - 768, /* 13 */ - 1024, /* 14 */ - 1536, /* 15 */ - 2048, /* 16 */ - 3072, /* 17 */ - 4096, /* 18 */ - 6144, /* 19 */ - 8192, /* 1A */ - 12288, /* 1B */ - 16384, /* 1C */ - 24576, /* 1D */ - 32768 /* 1E */ -}; - const struct rvt_operation_params hfi1_post_parms[RVT_OPERATION_MAX] = { [IB_WR_RDMA_WRITE] = { .length = sizeof(struct ib_rdma_wr), @@ -196,15 +159,18 @@ static void flush_tx_list(struct rvt_qp *qp) static void flush_iowait(struct rvt_qp *qp) { struct hfi1_qp_priv *priv = qp->priv; - struct hfi1_ibdev *dev = to_idev(qp->ibqp.device); unsigned long flags; + seqlock_t *lock = priv->s_iowait.lock; - write_seqlock_irqsave(&dev->iowait_lock, flags); + if (!lock) + return; + write_seqlock_irqsave(lock, flags); if (!list_empty(&priv->s_iowait.list)) { list_del_init(&priv->s_iowait.list); + priv->s_iowait.lock = NULL; rvt_put_qp(qp); } - write_sequnlock_irqrestore(&dev->iowait_lock, flags); + write_sequnlock_irqrestore(lock, flags); } static inline int opa_mtu_enum_to_int(int mtu) @@ -337,68 +303,6 @@ int hfi1_check_send_wqe(struct rvt_qp *qp, } /** - * hfi1_compute_aeth - compute the AETH (syndrome + MSN) - * @qp: the queue pair to compute the AETH for - * - * Returns the AETH. - */ -__be32 hfi1_compute_aeth(struct rvt_qp *qp) -{ - u32 aeth = qp->r_msn & HFI1_MSN_MASK; - - if (qp->ibqp.srq) { - /* - * Shared receive queues don't generate credits. - * Set the credit field to the invalid value. - */ - aeth |= HFI1_AETH_CREDIT_INVAL << HFI1_AETH_CREDIT_SHIFT; - } else { - u32 min, max, x; - u32 credits; - struct rvt_rwq *wq = qp->r_rq.wq; - u32 head; - u32 tail; - - /* sanity check pointers before trusting them */ - head = wq->head; - if (head >= qp->r_rq.size) - head = 0; - tail = wq->tail; - if (tail >= qp->r_rq.size) - tail = 0; - /* - * Compute the number of credits available (RWQEs). - * There is a small chance that the pair of reads are - * not atomic, which is OK, since the fuzziness is - * resolved as further ACKs go out. - */ - credits = head - tail; - if ((int)credits < 0) - credits += qp->r_rq.size; - /* - * Binary search the credit table to find the code to - * use. - */ - min = 0; - max = 31; - for (;;) { - x = (min + max) / 2; - if (credit_table[x] == credits) - break; - if (credit_table[x] > credits) { - max = x; - } else { - if (min == x) - break; - min = x; - } - } - aeth |= x << HFI1_AETH_CREDIT_SHIFT; - } - return cpu_to_be32(aeth); -} - -/** * _hfi1_schedule_send - schedule progress * @qp: the QP * @@ -454,44 +358,6 @@ void hfi1_schedule_send(struct rvt_qp *qp) _hfi1_schedule_send(qp); } -/** - * hfi1_get_credit - handle credit in aeth - * @qp: the qp - * @aeth: the Acknowledge Extended Transport Header - * - * The QP s_lock should be held. - */ -void hfi1_get_credit(struct rvt_qp *qp, u32 aeth) -{ - u32 credit = (aeth >> HFI1_AETH_CREDIT_SHIFT) & HFI1_AETH_CREDIT_MASK; - - lockdep_assert_held(&qp->s_lock); - /* - * If the credit is invalid, we can send - * as many packets as we like. Otherwise, we have to - * honor the credit field. - */ - if (credit == HFI1_AETH_CREDIT_INVAL) { - if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT)) { - qp->s_flags |= RVT_S_UNLIMITED_CREDIT; - if (qp->s_flags & RVT_S_WAIT_SSN_CREDIT) { - qp->s_flags &= ~RVT_S_WAIT_SSN_CREDIT; - hfi1_schedule_send(qp); - } - } - } else if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT)) { - /* Compute new LSN (i.e., MSN + credit) */ - credit = (aeth + credit_table[credit]) & HFI1_MSN_MASK; - if (cmp_msn(credit, qp->s_lsn) > 0) { - qp->s_lsn = credit; - if (qp->s_flags & RVT_S_WAIT_SSN_CREDIT) { - qp->s_flags &= ~RVT_S_WAIT_SSN_CREDIT; - hfi1_schedule_send(qp); - } - } - } -} - void hfi1_qp_wakeup(struct rvt_qp *qp, u32 flag) { unsigned long flags; @@ -543,6 +409,7 @@ static int iowait_sleep( ibp->rvp.n_dmawait++; qp->s_flags |= RVT_S_WAIT_DMA_DESC; list_add_tail(&priv->s_iowait.list, &sde->dmawait); + priv->s_iowait.lock = &dev->iowait_lock; trace_hfi1_qpsleep(qp, RVT_S_WAIT_DMA_DESC); rvt_get_qp(qp); } @@ -740,7 +607,7 @@ void qp_iter_print(struct seq_file *s, struct qp_iter *iter) wqe = rvt_get_swqe_ptr(qp, qp->s_last); send_context = qp_to_send_context(qp, priv->s_sc); seq_printf(s, - "N %d %s QP %x R %u %s %u %u %u f=%x %u %u %u %u %u %u PSN %x %x %x %x %x (%u %u %u %u %u %u %u) RQP %x LID %x SL %u MTU %u %u %u %u SDE %p,%u SC %p,%u SCQ %u %u PID %d\n", + "N %d %s QP %x R %u %s %u %u %u f=%x %u %u %u %u %u %u SPSN %x %x %x %x %x RPSN %x (%u %u %u %u %u %u %u) RQP %x LID %x SL %u MTU %u %u %u %u %u SDE %p,%u SC %p,%u SCQ %u %u PID %d\n", iter->n, qp_idle(qp) ? "I" : "B", qp->ibqp.qp_num, @@ -759,6 +626,7 @@ void qp_iter_print(struct seq_file *s, struct qp_iter *iter) qp->s_last_psn, qp->s_psn, qp->s_next_psn, qp->s_sending_psn, qp->s_sending_hpsn, + qp->r_psn, qp->s_last, qp->s_acked, qp->s_cur, qp->s_tail, qp->s_head, qp->s_size, qp->s_avail, @@ -769,6 +637,7 @@ void qp_iter_print(struct seq_file *s, struct qp_iter *iter) qp->s_retry, qp->s_retry_cnt, qp->s_rnr_retry_cnt, + qp->s_rnr_retry, sde, sde ? sde->this_idx : 0, send_context, @@ -778,19 +647,6 @@ void qp_iter_print(struct seq_file *s, struct qp_iter *iter) qp->pid); } -void qp_comm_est(struct rvt_qp *qp) -{ - qp->r_flags |= RVT_R_COMM_EST; - if (qp->ibqp.event_handler) { - struct ib_event ev; - - ev.device = qp->ibqp.device; - ev.element.qp = &qp->ibqp; - ev.event = IB_EVENT_COMM_EST; - qp->ibqp.event_handler(&ev, qp->ibqp.qp_context); - } -} - void *qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp, gfp_t gfp) { @@ -815,8 +671,6 @@ void *qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp, iowait_sleep, iowait_wakeup, iowait_sdma_drained); - setup_timer(&priv->s_rnr_timer, hfi1_rc_rnr_retry, (unsigned long)qp); - qp->s_timer.function = hfi1_rc_timeout; return priv; } @@ -857,7 +711,6 @@ void flush_qp_waiters(struct rvt_qp *qp) { lockdep_assert_held(&qp->s_lock); flush_iowait(qp); - hfi1_stop_rc_timers(qp); } void stop_send_queue(struct rvt_qp *qp) @@ -865,7 +718,6 @@ void stop_send_queue(struct rvt_qp *qp) struct hfi1_qp_priv *priv = qp->priv; cancel_work_sync(&priv->s_iowait.iowork); - hfi1_del_timers_sync(qp); } void quiesce_qp(struct rvt_qp *qp) @@ -957,16 +809,20 @@ int get_pmtu_from_attr(struct rvt_dev_info *rdi, struct rvt_qp *qp, void notify_error_qp(struct rvt_qp *qp) { - struct hfi1_ibdev *dev = to_idev(qp->ibqp.device); struct hfi1_qp_priv *priv = qp->priv; - - write_seqlock(&dev->iowait_lock); - if (!list_empty(&priv->s_iowait.list) && !(qp->s_flags & RVT_S_BUSY)) { - qp->s_flags &= ~RVT_S_ANY_WAIT_IO; - list_del_init(&priv->s_iowait.list); - rvt_put_qp(qp); + seqlock_t *lock = priv->s_iowait.lock; + + if (lock) { + write_seqlock(lock); + if (!list_empty(&priv->s_iowait.list) && + !(qp->s_flags & RVT_S_BUSY)) { + qp->s_flags &= ~RVT_S_ANY_WAIT_IO; + list_del_init(&priv->s_iowait.list); + priv->s_iowait.lock = NULL; + rvt_put_qp(qp); + } + write_sequnlock(lock); } - write_sequnlock(&dev->iowait_lock); if (!(qp->s_flags & RVT_S_BUSY)) { qp->s_hdrwords = 0; diff --git a/drivers/infiniband/hw/hfi1/qp.h b/drivers/infiniband/hw/hfi1/qp.h index 587d84d65bb8..1eb9cd7b8c19 100644 --- a/drivers/infiniband/hw/hfi1/qp.h +++ b/drivers/infiniband/hw/hfi1/qp.h @@ -71,14 +71,6 @@ static inline void clear_ahg(struct rvt_qp *qp) } /** - * hfi1_compute_aeth - compute the AETH (syndrome + MSN) - * @qp: the queue pair to compute the AETH for - * - * Returns the AETH. - */ -__be32 hfi1_compute_aeth(struct rvt_qp *qp); - -/** * hfi1_create_qp - create a queue pair for a device * @ibpd: the protection domain who's device we create the queue pair for * @init_attr: the attributes of the queue pair @@ -91,14 +83,6 @@ __be32 hfi1_compute_aeth(struct rvt_qp *qp); struct ib_qp *hfi1_create_qp(struct ib_pd *ibpd, struct ib_qp_init_attr *init_attr, struct ib_udata *udata); -/** - * hfi1_get_credit - flush the send work queue of a QP - * @qp: the qp who's send work queue to flush - * @aeth: the Acknowledge Extended Transport Header - * - * The QP s_lock should be held. - */ -void hfi1_get_credit(struct rvt_qp *qp, u32 aeth); /** * hfi1_qp_wakeup - wake up on the indicated event @@ -131,12 +115,6 @@ int qp_iter_next(struct qp_iter *iter); */ void qp_iter_print(struct seq_file *s, struct qp_iter *iter); -/** - * qp_comm_est - handle trap with QP established - * @qp: the QP - */ -void qp_comm_est(struct rvt_qp *qp); - void _hfi1_schedule_send(struct rvt_qp *qp); void hfi1_schedule_send(struct rvt_qp *qp); diff --git a/drivers/infiniband/hw/hfi1/rc.c b/drivers/infiniband/hw/hfi1/rc.c index 83198a8a8797..7382be11afca 100644 --- a/drivers/infiniband/hw/hfi1/rc.c +++ b/drivers/infiniband/hw/hfi1/rc.c @@ -57,133 +57,6 @@ /* cut down ridiculously long IB macro names */ #define OP(x) RC_OP(x) -/** - * hfi1_add_retry_timer - add/start a retry timer - * @qp - the QP - * - * add a retry timer on the QP - */ -static inline void hfi1_add_retry_timer(struct rvt_qp *qp) -{ - struct ib_qp *ibqp = &qp->ibqp; - struct rvt_dev_info *rdi = ib_to_rvt(ibqp->device); - - lockdep_assert_held(&qp->s_lock); - qp->s_flags |= RVT_S_TIMER; - /* 4.096 usec. * (1 << qp->timeout) */ - qp->s_timer.expires = jiffies + qp->timeout_jiffies + - rdi->busy_jiffies; - add_timer(&qp->s_timer); -} - -/** - * hfi1_add_rnr_timer - add/start an rnr timer - * @qp - the QP - * @to - timeout in usecs - * - * add an rnr timer on the QP - */ -void hfi1_add_rnr_timer(struct rvt_qp *qp, u32 to) -{ - struct hfi1_qp_priv *priv = qp->priv; - - lockdep_assert_held(&qp->s_lock); - qp->s_flags |= RVT_S_WAIT_RNR; - priv->s_rnr_timer.expires = jiffies + usecs_to_jiffies(to); - add_timer(&priv->s_rnr_timer); -} - -/** - * hfi1_mod_retry_timer - mod a retry timer - * @qp - the QP - * - * Modify a potentially already running retry - * timer - */ -static inline void hfi1_mod_retry_timer(struct rvt_qp *qp) -{ - struct ib_qp *ibqp = &qp->ibqp; - struct rvt_dev_info *rdi = ib_to_rvt(ibqp->device); - - lockdep_assert_held(&qp->s_lock); - qp->s_flags |= RVT_S_TIMER; - /* 4.096 usec. * (1 << qp->timeout) */ - mod_timer(&qp->s_timer, jiffies + qp->timeout_jiffies + - rdi->busy_jiffies); -} - -/** - * hfi1_stop_retry_timer - stop a retry timer - * @qp - the QP - * - * stop a retry timer and return if the timer - * had been pending. - */ -static inline int hfi1_stop_retry_timer(struct rvt_qp *qp) -{ - int rval = 0; - - lockdep_assert_held(&qp->s_lock); - /* Remove QP from retry */ - if (qp->s_flags & RVT_S_TIMER) { - qp->s_flags &= ~RVT_S_TIMER; - rval = del_timer(&qp->s_timer); - } - return rval; -} - -/** - * hfi1_stop_rc_timers - stop all timers - * @qp - the QP - * - * stop any pending timers - */ -void hfi1_stop_rc_timers(struct rvt_qp *qp) -{ - struct hfi1_qp_priv *priv = qp->priv; - - lockdep_assert_held(&qp->s_lock); - /* Remove QP from all timers */ - if (qp->s_flags & (RVT_S_TIMER | RVT_S_WAIT_RNR)) { - qp->s_flags &= ~(RVT_S_TIMER | RVT_S_WAIT_RNR); - del_timer(&qp->s_timer); - del_timer(&priv->s_rnr_timer); - } -} - -/** - * hfi1_stop_rnr_timer - stop an rnr timer - * @qp - the QP - * - * stop an rnr timer and return if the timer - * had been pending. - */ -static inline int hfi1_stop_rnr_timer(struct rvt_qp *qp) -{ - int rval = 0; - struct hfi1_qp_priv *priv = qp->priv; - - lockdep_assert_held(&qp->s_lock); - /* Remove QP from rnr timer */ - if (qp->s_flags & RVT_S_WAIT_RNR) { - qp->s_flags &= ~RVT_S_WAIT_RNR; - rval = del_timer(&priv->s_rnr_timer); - } - return rval; -} - -/** - * hfi1_del_timers_sync - wait for any timeout routines to exit - * @qp - the QP - */ -void hfi1_del_timers_sync(struct rvt_qp *qp) -{ - struct hfi1_qp_priv *priv = qp->priv; - - del_timer_sync(&qp->s_timer); - del_timer_sync(&priv->s_rnr_timer); -} - static u32 restart_sge(struct rvt_sge_state *ss, struct rvt_swqe *wqe, u32 psn, u32 pmtu) { @@ -194,7 +67,7 @@ static u32 restart_sge(struct rvt_sge_state *ss, struct rvt_swqe *wqe, ss->sg_list = wqe->sg_list + 1; ss->num_sge = wqe->wr.num_sge; ss->total_len = wqe->length; - hfi1_skip_sge(ss, len, 0); + rvt_skip_sge(ss, len, false); return wqe->length - len; } @@ -276,7 +149,7 @@ static int make_rc_ack(struct hfi1_ibdev *dev, struct rvt_qp *qp, rvt_get_mr(ps->s_txreq->mr); qp->s_ack_rdma_sge.sge = e->rdma_sge; qp->s_ack_rdma_sge.num_sge = 1; - qp->s_cur_sge = &qp->s_ack_rdma_sge; + ps->s_txreq->ss = &qp->s_ack_rdma_sge; if (len > pmtu) { len = pmtu; qp->s_ack_state = OP(RDMA_READ_RESPONSE_FIRST); @@ -284,16 +157,16 @@ static int make_rc_ack(struct hfi1_ibdev *dev, struct rvt_qp *qp, qp->s_ack_state = OP(RDMA_READ_RESPONSE_ONLY); e->sent = 1; } - ohdr->u.aeth = hfi1_compute_aeth(qp); + ohdr->u.aeth = rvt_compute_aeth(qp); hwords++; qp->s_ack_rdma_psn = e->psn; bth2 = mask_psn(qp->s_ack_rdma_psn++); } else { /* COMPARE_SWAP or FETCH_ADD */ - qp->s_cur_sge = NULL; + ps->s_txreq->ss = NULL; len = 0; qp->s_ack_state = OP(ATOMIC_ACKNOWLEDGE); - ohdr->u.at.aeth = hfi1_compute_aeth(qp); + ohdr->u.at.aeth = rvt_compute_aeth(qp); ib_u64_put(e->atomic_data, &ohdr->u.at.atomic_ack_eth); hwords += sizeof(ohdr->u.at) / sizeof(u32); bth2 = mask_psn(e->psn); @@ -306,7 +179,7 @@ static int make_rc_ack(struct hfi1_ibdev *dev, struct rvt_qp *qp, qp->s_ack_state = OP(RDMA_READ_RESPONSE_MIDDLE); /* FALLTHROUGH */ case OP(RDMA_READ_RESPONSE_MIDDLE): - qp->s_cur_sge = &qp->s_ack_rdma_sge; + ps->s_txreq->ss = &qp->s_ack_rdma_sge; ps->s_txreq->mr = qp->s_ack_rdma_sge.sge.mr; if (ps->s_txreq->mr) rvt_get_mr(ps->s_txreq->mr); @@ -315,7 +188,7 @@ static int make_rc_ack(struct hfi1_ibdev *dev, struct rvt_qp *qp, len = pmtu; middle = HFI1_CAP_IS_KSET(SDMA_AHG); } else { - ohdr->u.aeth = hfi1_compute_aeth(qp); + ohdr->u.aeth = rvt_compute_aeth(qp); hwords++; qp->s_ack_state = OP(RDMA_READ_RESPONSE_LAST); e = &qp->s_ack_queue[qp->s_tail_ack_queue]; @@ -335,14 +208,14 @@ normal: */ qp->s_ack_state = OP(SEND_ONLY); qp->s_flags &= ~RVT_S_ACK_PENDING; - qp->s_cur_sge = NULL; + ps->s_txreq->ss = NULL; if (qp->s_nak_state) ohdr->u.aeth = - cpu_to_be32((qp->r_msn & HFI1_MSN_MASK) | + cpu_to_be32((qp->r_msn & IB_MSN_MASK) | (qp->s_nak_state << - HFI1_AETH_CREDIT_SHIFT)); + IB_AETH_CREDIT_SHIFT)); else - ohdr->u.aeth = hfi1_compute_aeth(qp); + ohdr->u.aeth = rvt_compute_aeth(qp); hwords++; len = 0; bth0 = OP(ACKNOWLEDGE) << 24; @@ -351,7 +224,7 @@ normal: qp->s_rdma_ack_cnt++; qp->s_hdrwords = hwords; ps->s_txreq->sde = priv->s_sde; - qp->s_cur_size = len; + ps->s_txreq->s_cur_size = len; hfi1_make_ruc_header(qp, ohdr, bth0, bth2, middle, ps); /* pbc */ ps->s_txreq->hdr_dwords = qp->s_hdrwords + 2; @@ -414,7 +287,7 @@ int hfi1_make_rc_req(struct rvt_qp *qp, struct hfi1_pkt_state *ps) goto bail; /* We are in the error state, flush the work request. */ smp_read_barrier_depends(); /* see post_one_send() */ - if (qp->s_last == ACCESS_ONCE(qp->s_head)) + if (qp->s_last == READ_ONCE(qp->s_head)) goto bail; /* If DMAs are in progress, we can't flush immediately. */ if (iowait_sdma_pending(&priv->s_iowait)) { @@ -457,7 +330,8 @@ int hfi1_make_rc_req(struct rvt_qp *qp, struct hfi1_pkt_state *ps) newreq = 0; if (qp->s_cur == qp->s_tail) { /* Check if send work queue is empty. */ - if (qp->s_tail == qp->s_head) { + smp_read_barrier_depends(); /* see post_one_send() */ + if (qp->s_tail == READ_ONCE(qp->s_head)) { clear_ahg(qp); goto bail; } @@ -518,7 +392,7 @@ int hfi1_make_rc_req(struct rvt_qp *qp, struct hfi1_pkt_state *ps) case IB_WR_SEND_WITH_INV: /* If no credit, return. */ if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT) && - cmp_msn(wqe->ssn, qp->s_lsn + 1) > 0) { + rvt_cmp_msn(wqe->ssn, qp->s_lsn + 1) > 0) { qp->s_flags |= RVT_S_WAIT_SSN_CREDIT; goto bail; } @@ -555,7 +429,7 @@ int hfi1_make_rc_req(struct rvt_qp *qp, struct hfi1_pkt_state *ps) case IB_WR_RDMA_WRITE_WITH_IMM: /* If no credit, return. */ if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT) && - cmp_msn(wqe->ssn, qp->s_lsn + 1) > 0) { + rvt_cmp_msn(wqe->ssn, qp->s_lsn + 1) > 0) { qp->s_flags |= RVT_S_WAIT_SSN_CREDIT; goto bail; } @@ -801,8 +675,8 @@ int hfi1_make_rc_req(struct rvt_qp *qp, struct hfi1_pkt_state *ps) qp->s_len -= len; qp->s_hdrwords = hwords; ps->s_txreq->sde = priv->s_sde; - qp->s_cur_sge = ss; - qp->s_cur_size = len; + ps->s_txreq->ss = ss; + ps->s_txreq->s_cur_size = len; hfi1_make_ruc_header( qp, ohdr, @@ -840,7 +714,7 @@ bail_no_tx: void hfi1_send_rc_ack(struct hfi1_ctxtdata *rcd, struct rvt_qp *qp, int is_fecn) { - struct hfi1_ibport *ibp = to_iport(qp->ibqp.device, qp->port_num); + struct hfi1_ibport *ibp = rcd_to_iport(rcd); struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); u64 pbc, pbc_flags = 0; u16 lrh0; @@ -853,6 +727,10 @@ void hfi1_send_rc_ack(struct hfi1_ctxtdata *rcd, struct rvt_qp *qp, struct ib_header hdr; struct ib_other_headers *ohdr; unsigned long flags; + struct hfi1_qp_priv *priv = qp->priv; + + /* clear the defer count */ + priv->r_adefered = 0; /* Don't send ACK or NAK if a RDMA read or atomic is pending. */ if (qp->s_flags & RVT_S_RESP_PENDING) @@ -880,11 +758,11 @@ void hfi1_send_rc_ack(struct hfi1_ctxtdata *rcd, struct rvt_qp *qp, if (qp->s_mig_state == IB_MIG_MIGRATED) bth0 |= IB_BTH_MIG_REQ; if (qp->r_nak_state) - ohdr->u.aeth = cpu_to_be32((qp->r_msn & HFI1_MSN_MASK) | + ohdr->u.aeth = cpu_to_be32((qp->r_msn & IB_MSN_MASK) | (qp->r_nak_state << - HFI1_AETH_CREDIT_SHIFT)); + IB_AETH_CREDIT_SHIFT)); else - ohdr->u.aeth = hfi1_compute_aeth(qp); + ohdr->u.aeth = rvt_compute_aeth(qp); sc5 = ibp->sl_to_sc[qp->remote_ah_attr.sl]; /* set PBC_DC_INFO bit (aka SC[4]) in pbc_flags */ pbc_flags |= ((!!(sc5 & 0x10)) << PBC_DC_INFO_SHIFT); @@ -1038,7 +916,7 @@ done: * Back up requester to resend the last un-ACKed request. * The QP r_lock and s_lock should be held and interrupts disabled. */ -static void restart_rc(struct rvt_qp *qp, u32 psn, int wait) +void hfi1_restart_rc(struct rvt_qp *qp, u32 psn, int wait) { struct rvt_swqe *wqe = rvt_get_swqe_ptr(qp, qp->s_acked); struct hfi1_ibport *ibp; @@ -1075,44 +953,6 @@ static void restart_rc(struct rvt_qp *qp, u32 psn, int wait) } /* - * This is called from s_timer for missing responses. - */ -void hfi1_rc_timeout(unsigned long arg) -{ - struct rvt_qp *qp = (struct rvt_qp *)arg; - struct hfi1_ibport *ibp; - unsigned long flags; - - spin_lock_irqsave(&qp->r_lock, flags); - spin_lock(&qp->s_lock); - if (qp->s_flags & RVT_S_TIMER) { - ibp = to_iport(qp->ibqp.device, qp->port_num); - ibp->rvp.n_rc_timeouts++; - qp->s_flags &= ~RVT_S_TIMER; - del_timer(&qp->s_timer); - trace_hfi1_timeout(qp, qp->s_last_psn + 1); - restart_rc(qp, qp->s_last_psn + 1, 1); - hfi1_schedule_send(qp); - } - spin_unlock(&qp->s_lock); - spin_unlock_irqrestore(&qp->r_lock, flags); -} - -/* - * This is called from s_timer for RNR timeouts. - */ -void hfi1_rc_rnr_retry(unsigned long arg) -{ - struct rvt_qp *qp = (struct rvt_qp *)arg; - unsigned long flags; - - spin_lock_irqsave(&qp->s_lock, flags); - hfi1_stop_rnr_timer(qp); - hfi1_schedule_send(qp); - spin_unlock_irqrestore(&qp->s_lock, flags); -} - -/* * Set qp->s_sending_psn to the next PSN after the given one. * This would be psn+1 except when RDMA reads are present. */ @@ -1146,13 +986,11 @@ void hfi1_rc_send_complete(struct rvt_qp *qp, struct ib_header *hdr) { struct ib_other_headers *ohdr; struct rvt_swqe *wqe; - struct ib_wc wc; - unsigned i; u32 opcode; u32 psn; lockdep_assert_held(&qp->s_lock); - if (!(ib_rvt_state_ops[qp->state] & RVT_PROCESS_OR_FLUSH_SEND)) + if (!(ib_rvt_state_ops[qp->state] & RVT_SEND_OR_FLUSH_OR_RECV_OK)) return; /* Find out where the BTH is */ @@ -1180,7 +1018,7 @@ void hfi1_rc_send_complete(struct rvt_qp *qp, struct ib_header *hdr) !(qp->s_flags & (RVT_S_TIMER | RVT_S_WAIT_RNR | RVT_S_WAIT_PSN)) && (ib_rvt_state_ops[qp->state] & RVT_PROCESS_RECV_OK)) - hfi1_add_retry_timer(qp); + rvt_add_retry_timer(qp); while (qp->s_last != qp->s_acked) { u32 s_last; @@ -1195,22 +1033,8 @@ void hfi1_rc_send_complete(struct rvt_qp *qp, struct ib_header *hdr) qp->s_last = s_last; /* see post_send() */ barrier(); - for (i = 0; i < wqe->wr.num_sge; i++) { - struct rvt_sge *sge = &wqe->sg_list[i]; - - rvt_put_mr(sge->mr); - } - /* Post a send completion queue entry if requested. */ - if (!(qp->s_flags & RVT_S_SIGNAL_REQ_WR) || - (wqe->wr.send_flags & IB_SEND_SIGNALED)) { - memset(&wc, 0, sizeof(wc)); - wc.wr_id = wqe->wr.wr_id; - wc.status = IB_WC_SUCCESS; - wc.opcode = ib_hfi1_wc_opcode[wqe->wr.opcode]; - wc.byte_len = wqe->length; - wc.qp = &qp->ibqp; - rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.send_cq), &wc, 0); - } + rvt_put_swqe(wqe); + rvt_qp_swqe_complete(qp, wqe, IB_WC_SUCCESS); } /* * If we were waiting for sends to complete before re-sending, @@ -1240,9 +1064,6 @@ static struct rvt_swqe *do_rc_completion(struct rvt_qp *qp, struct rvt_swqe *wqe, struct hfi1_ibport *ibp) { - struct ib_wc wc; - unsigned i; - lockdep_assert_held(&qp->s_lock); /* * Don't decrement refcount and don't generate a @@ -1253,28 +1074,14 @@ static struct rvt_swqe *do_rc_completion(struct rvt_qp *qp, cmp_psn(qp->s_sending_psn, qp->s_sending_hpsn) > 0) { u32 s_last; - for (i = 0; i < wqe->wr.num_sge; i++) { - struct rvt_sge *sge = &wqe->sg_list[i]; - - rvt_put_mr(sge->mr); - } + rvt_put_swqe(wqe); s_last = qp->s_last; if (++s_last >= qp->s_size) s_last = 0; qp->s_last = s_last; /* see post_send() */ barrier(); - /* Post a send completion queue entry if requested. */ - if (!(qp->s_flags & RVT_S_SIGNAL_REQ_WR) || - (wqe->wr.send_flags & IB_SEND_SIGNALED)) { - memset(&wc, 0, sizeof(wc)); - wc.wr_id = wqe->wr.wr_id; - wc.status = IB_WC_SUCCESS; - wc.opcode = ib_hfi1_wc_opcode[wqe->wr.opcode]; - wc.byte_len = wqe->length; - wc.qp = &qp->ibqp; - rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.send_cq), &wc, 0); - } + rvt_qp_swqe_complete(qp, wqe, IB_WC_SUCCESS); } else { struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); @@ -1341,7 +1148,6 @@ static int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode, int ret = 0; u32 ack_psn; int diff; - unsigned long to; lockdep_assert_held(&qp->s_lock); /* @@ -1351,10 +1157,10 @@ static int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode, * request but will include an ACK'ed request(s). */ ack_psn = psn; - if (aeth >> 29) + if (aeth >> IB_AETH_NAK_SHIFT) ack_psn--; wqe = rvt_get_swqe_ptr(qp, qp->s_acked); - ibp = to_iport(qp->ibqp.device, qp->port_num); + ibp = rcd_to_iport(rcd); /* * The MSN might be for a later WQE than the PSN indicates so @@ -1390,7 +1196,7 @@ static int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode, /* Retry this request. */ if (!(qp->r_flags & RVT_R_RDMAR_SEQ)) { qp->r_flags |= RVT_R_RDMAR_SEQ; - restart_rc(qp, qp->s_last_psn + 1, 0); + hfi1_restart_rc(qp, qp->s_last_psn + 1, 0); if (list_empty(&qp->rspwait)) { qp->r_flags |= RVT_R_RSP_SEND; rvt_get_qp(qp); @@ -1431,7 +1237,7 @@ static int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode, break; } - switch (aeth >> 29) { + switch (aeth >> IB_AETH_NAK_SHIFT) { case 0: /* ACK */ this_cpu_inc(*ibp->rvp.rc_acks); if (qp->s_acked != qp->s_tail) { @@ -1439,7 +1245,7 @@ static int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode, * We are expecting more ACKs so * mod the retry timer. */ - hfi1_mod_retry_timer(qp); + rvt_mod_retry_timer(qp); /* * We can stop re-sending the earlier packets and * continue with the next packet the receiver wants. @@ -1448,7 +1254,7 @@ static int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode, reset_psn(qp, psn + 1); } else { /* No more acks - kill all timers */ - hfi1_stop_rc_timers(qp); + rvt_stop_rc_timers(qp); if (cmp_psn(qp->s_psn, psn) <= 0) { qp->s_state = OP(SEND_LAST); qp->s_psn = psn + 1; @@ -1458,7 +1264,7 @@ static int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode, qp->s_flags &= ~RVT_S_WAIT_ACK; hfi1_schedule_send(qp); } - hfi1_get_credit(qp, aeth); + rvt_get_credit(qp, aeth); qp->s_rnr_retry = qp->s_rnr_retry_cnt; qp->s_retry = qp->s_retry_cnt; update_last_psn(qp, psn); @@ -1485,11 +1291,8 @@ static int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode, reset_psn(qp, psn); qp->s_flags &= ~(RVT_S_WAIT_SSN_CREDIT | RVT_S_WAIT_ACK); - hfi1_stop_rc_timers(qp); - to = - ib_hfi1_rnr_table[(aeth >> HFI1_AETH_CREDIT_SHIFT) & - HFI1_AETH_CREDIT_MASK]; - hfi1_add_rnr_timer(qp, to); + rvt_stop_rc_timers(qp); + rvt_add_rnr_timer(qp, aeth); return 0; case 3: /* NAK */ @@ -1497,8 +1300,8 @@ static int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode, goto bail_stop; /* The last valid PSN is the previous PSN. */ update_last_psn(qp, psn - 1); - switch ((aeth >> HFI1_AETH_CREDIT_SHIFT) & - HFI1_AETH_CREDIT_MASK) { + switch ((aeth >> IB_AETH_CREDIT_SHIFT) & + IB_AETH_CREDIT_MASK) { case 0: /* PSN sequence error */ ibp->rvp.n_seq_naks++; /* @@ -1507,7 +1310,7 @@ static int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode, * RDMA READ response which terminates the RDMA * READ. */ - restart_rc(qp, psn, 0); + hfi1_restart_rc(qp, psn, 0); hfi1_schedule_send(qp); break; @@ -1546,7 +1349,7 @@ reserved: } /* cannot be reached */ bail_stop: - hfi1_stop_rc_timers(qp); + rvt_stop_rc_timers(qp); return ret; } @@ -1561,7 +1364,7 @@ static void rdma_seq_err(struct rvt_qp *qp, struct hfi1_ibport *ibp, u32 psn, lockdep_assert_held(&qp->s_lock); /* Remove QP from retry timer */ - hfi1_stop_rc_timers(qp); + rvt_stop_rc_timers(qp); wqe = rvt_get_swqe_ptr(qp, qp->s_acked); @@ -1575,7 +1378,7 @@ static void rdma_seq_err(struct rvt_qp *qp, struct hfi1_ibport *ibp, u32 psn, ibp->rvp.n_rdma_seq++; qp->r_flags |= RVT_R_RDMAR_SEQ; - restart_rc(qp, qp->s_last_psn + 1, 0); + hfi1_restart_rc(qp, qp->s_last_psn + 1, 0); if (list_empty(&qp->rspwait)) { qp->r_flags |= RVT_R_RSP_SEND; rvt_get_qp(qp); @@ -1619,7 +1422,7 @@ static void rc_rcv_resp(struct hfi1_ibport *ibp, /* Ignore invalid responses. */ smp_read_barrier_depends(); /* see post_one_send */ - if (cmp_psn(psn, ACCESS_ONCE(qp->s_next_psn)) >= 0) + if (cmp_psn(psn, READ_ONCE(qp->s_next_psn)) >= 0) goto ack_done; /* Ignore duplicate responses. */ @@ -1628,8 +1431,8 @@ static void rc_rcv_resp(struct hfi1_ibport *ibp, /* Update credits for "ghost" ACKs */ if (diff == 0 && opcode == OP(ACKNOWLEDGE)) { aeth = be32_to_cpu(ohdr->u.aeth); - if ((aeth >> 29) == 0) - hfi1_get_credit(qp, aeth); + if ((aeth >> IB_AETH_NAK_SHIFT) == 0) + rvt_get_credit(qp, aeth); } goto ack_done; } @@ -1689,8 +1492,7 @@ read_middle: * We got a response so update the timeout. * 4.096 usec. * (1 << qp->timeout) */ - qp->s_flags |= RVT_S_TIMER; - mod_timer(&qp->s_timer, jiffies + qp->timeout_jiffies); + rvt_mod_retry_timer(qp); if (qp->s_flags & RVT_S_WAIT_ACK) { qp->s_flags &= ~RVT_S_WAIT_ACK; hfi1_schedule_send(qp); @@ -1706,7 +1508,7 @@ read_middle: qp->s_rdma_read_len -= pmtu; update_last_psn(qp, psn); spin_unlock_irqrestore(&qp->s_lock, flags); - hfi1_copy_sge(&qp->s_rdma_read_sge, data, pmtu, 0, 0); + hfi1_copy_sge(&qp->s_rdma_read_sge, data, pmtu, false, false); goto bail; case OP(RDMA_READ_RESPONSE_ONLY): @@ -1750,7 +1552,7 @@ read_last: if (unlikely(tlen != qp->s_rdma_read_len)) goto ack_len_err; aeth = be32_to_cpu(ohdr->u.aeth); - hfi1_copy_sge(&qp->s_rdma_read_sge, data, tlen, 0, 0); + hfi1_copy_sge(&qp->s_rdma_read_sge, data, tlen, false, false); WARN_ON(qp->s_rdma_read_sge.num_sge); (void)do_rc_ack(qp, aeth, psn, OP(RDMA_READ_RESPONSE_LAST), 0, rcd); @@ -1819,7 +1621,7 @@ static noinline int rc_rcv_error(struct ib_other_headers *ohdr, void *data, struct rvt_qp *qp, u32 opcode, u32 psn, int diff, struct hfi1_ctxtdata *rcd) { - struct hfi1_ibport *ibp = to_iport(qp->ibqp.device, qp->port_num); + struct hfi1_ibport *ibp = rcd_to_iport(rcd); struct rvt_ack_entry *e; unsigned long flags; u8 i, prev; @@ -1994,25 +1796,6 @@ send_ack: return 0; } -void hfi1_rc_error(struct rvt_qp *qp, enum ib_wc_status err) -{ - unsigned long flags; - int lastwqe; - - spin_lock_irqsave(&qp->s_lock, flags); - lastwqe = rvt_error_qp(qp, err); - spin_unlock_irqrestore(&qp->s_lock, flags); - - if (lastwqe) { - struct ib_event ev; - - ev.device = qp->ibqp.device; - ev.element.qp = &qp->ibqp; - ev.event = IB_EVENT_QP_LAST_WQE_REACHED; - qp->ibqp.event_handler(&ev, qp->ibqp.qp_context); - } -} - static inline void update_ack_queue(struct rvt_qp *qp, unsigned n) { unsigned next; @@ -2128,7 +1911,7 @@ void hfi1_rc_rcv(struct hfi1_packet *packet) void *data = packet->ebuf; u32 tlen = packet->tlen; struct rvt_qp *qp = packet->qp; - struct hfi1_ibport *ibp = to_iport(qp->ibqp.device, qp->port_num); + struct hfi1_ibport *ibp = rcd_to_iport(rcd); struct ib_other_headers *ohdr = packet->ohdr; u32 bth0, opcode; u32 hdrsize = packet->hlen; @@ -2140,7 +1923,7 @@ void hfi1_rc_rcv(struct hfi1_packet *packet) struct ib_reth *reth; unsigned long flags; int ret, is_fecn = 0; - int copy_last = 0; + bool copy_last = false; u32 rkey; lockdep_assert_held(&qp->r_lock); @@ -2213,7 +1996,7 @@ void hfi1_rc_rcv(struct hfi1_packet *packet) } if (qp->state == IB_QPS_RTR && !(qp->r_flags & RVT_R_COMM_EST)) - qp_comm_est(qp); + rvt_comm_est(qp); /* OK, process the packet. */ switch (opcode) { @@ -2234,7 +2017,7 @@ send_middle: qp->r_rcv_len += pmtu; if (unlikely(qp->r_rcv_len > qp->r_len)) goto nack_inv; - hfi1_copy_sge(&qp->r_sge, data, pmtu, 1, 0); + hfi1_copy_sge(&qp->r_sge, data, pmtu, true, false); break; case OP(RDMA_WRITE_LAST_WITH_IMMEDIATE): @@ -2274,7 +2057,7 @@ send_last_inv: wc.wc_flags = IB_WC_WITH_INVALIDATE; goto send_last; case OP(RDMA_WRITE_LAST): - copy_last = ibpd_to_rvtpd(qp->ibqp.pd)->user; + copy_last = rvt_is_user_qp(qp); /* fall through */ case OP(SEND_LAST): no_immediate_data: @@ -2292,10 +2075,10 @@ send_last: wc.byte_len = tlen + qp->r_rcv_len; if (unlikely(wc.byte_len > qp->r_len)) goto nack_inv; - hfi1_copy_sge(&qp->r_sge, data, tlen, 1, copy_last); + hfi1_copy_sge(&qp->r_sge, data, tlen, true, copy_last); rvt_put_ss(&qp->r_sge); qp->r_msn++; - if (!test_and_clear_bit(RVT_R_WRID_VALID, &qp->r_aflags)) + if (!__test_and_clear_bit(RVT_R_WRID_VALID, &qp->r_aflags)) break; wc.wr_id = qp->r_wr_id; wc.status = IB_WC_SUCCESS; @@ -2330,7 +2113,7 @@ send_last: break; case OP(RDMA_WRITE_ONLY): - copy_last = 1; + copy_last = rvt_is_user_qp(qp); /* fall through */ case OP(RDMA_WRITE_FIRST): case OP(RDMA_WRITE_ONLY_WITH_IMMEDIATE): @@ -2410,8 +2193,7 @@ send_last: * Update the next expected PSN. We add 1 later * below, so only add the remainder here. */ - if (len > pmtu) - qp->r_psn += (len - 1) / pmtu; + qp->r_psn += rvt_div_mtu(qp, len - 1); } else { e->rdma_sge.mr = NULL; e->rdma_sge.vaddr = NULL; @@ -2546,7 +2328,7 @@ rnr_nak: return; nack_op_err: - hfi1_rc_error(qp, IB_WC_LOC_QP_OP_ERR); + rvt_rc_error(qp, IB_WC_LOC_QP_OP_ERR); qp->r_nak_state = IB_NAK_REMOTE_OPERATIONAL_ERROR; qp->r_ack_psn = qp->r_psn; /* Queue NAK for later */ @@ -2556,7 +2338,7 @@ nack_op_err: nack_inv_unlck: spin_unlock_irqrestore(&qp->s_lock, flags); nack_inv: - hfi1_rc_error(qp, IB_WC_LOC_QP_OP_ERR); + rvt_rc_error(qp, IB_WC_LOC_QP_OP_ERR); qp->r_nak_state = IB_NAK_INVALID_REQUEST; qp->r_ack_psn = qp->r_psn; /* Queue NAK for later */ @@ -2566,7 +2348,7 @@ nack_inv: nack_acc_unlck: spin_unlock_irqrestore(&qp->s_lock, flags); nack_acc: - hfi1_rc_error(qp, IB_WC_LOC_PROT_ERR); + rvt_rc_error(qp, IB_WC_LOC_PROT_ERR); qp->r_nak_state = IB_NAK_REMOTE_ACCESS_ERROR; qp->r_ack_psn = qp->r_psn; send_ack: @@ -2581,7 +2363,7 @@ void hfi1_rc_hdrerr( { int has_grh = rcv_flags & HFI1_HAS_GRH; struct ib_other_headers *ohdr; - struct hfi1_ibport *ibp = to_iport(qp->ibqp.device, qp->port_num); + struct hfi1_ibport *ibp = rcd_to_iport(rcd); int diff; u32 opcode; u32 psn, bth0; diff --git a/drivers/infiniband/hw/hfi1/ruc.c b/drivers/infiniband/hw/hfi1/ruc.c index a1576aea4756..aa15bcbfb079 100644 --- a/drivers/infiniband/hw/hfi1/ruc.c +++ b/drivers/infiniband/hw/hfi1/ruc.c @@ -54,44 +54,6 @@ #include "trace.h" /* - * Convert the AETH RNR timeout code into the number of microseconds. - */ -const u32 ib_hfi1_rnr_table[32] = { - 655360, /* 00: 655.36 */ - 10, /* 01: .01 */ - 20, /* 02 .02 */ - 30, /* 03: .03 */ - 40, /* 04: .04 */ - 60, /* 05: .06 */ - 80, /* 06: .08 */ - 120, /* 07: .12 */ - 160, /* 08: .16 */ - 240, /* 09: .24 */ - 320, /* 0A: .32 */ - 480, /* 0B: .48 */ - 640, /* 0C: .64 */ - 960, /* 0D: .96 */ - 1280, /* 0E: 1.28 */ - 1920, /* 0F: 1.92 */ - 2560, /* 10: 2.56 */ - 3840, /* 11: 3.84 */ - 5120, /* 12: 5.12 */ - 7680, /* 13: 7.68 */ - 10240, /* 14: 10.24 */ - 15360, /* 15: 15.36 */ - 20480, /* 16: 20.48 */ - 30720, /* 17: 30.72 */ - 40960, /* 18: 40.96 */ - 61440, /* 19: 61.44 */ - 81920, /* 1A: 81.92 */ - 122880, /* 1B: 122.88 */ - 163840, /* 1C: 163.84 */ - 245760, /* 1D: 245.76 */ - 327680, /* 1E: 327.68 */ - 491520 /* 1F: 491.52 */ -}; - -/* * Validate a RWQE and fill in the SGE state. * Return 1 if OK. */ @@ -239,16 +201,6 @@ bail: return ret; } -static __be64 get_sguid(struct hfi1_ibport *ibp, unsigned index) -{ - if (!index) { - struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); - - return cpu_to_be64(ppd->guid); - } - return ibp->guids[index - 1]; -} - static int gid_ok(union ib_gid *gid, __be64 gid_prefix, __be64 id) { return (gid->global.interface_id == id && @@ -368,10 +320,9 @@ static void ruc_loopback(struct rvt_qp *sqp) u64 sdata; atomic64_t *maddr; enum ib_wc_status send_status; - int release; + bool release; int ret; - int copy_last = 0; - u32 to; + bool copy_last = false; int local_ops = 0; rcu_read_lock(); @@ -435,7 +386,7 @@ again: memset(&wc, 0, sizeof(wc)); send_status = IB_WC_SUCCESS; - release = 1; + release = true; sqp->s_sge.sge = wqe->sg_list[0]; sqp->s_sge.sg_list = wqe->sg_list + 1; sqp->s_sge.num_sge = wqe->wr.num_sge; @@ -486,7 +437,7 @@ send: /* skip copy_last set and qp_access_flags recheck */ goto do_write; case IB_WR_RDMA_WRITE: - copy_last = ibpd_to_rvtpd(qp->ibqp.pd)->user; + copy_last = rvt_is_user_qp(qp); if (unlikely(!(qp->qp_access_flags & IB_ACCESS_REMOTE_WRITE))) goto inv_err; do_write: @@ -510,7 +461,7 @@ do_write: wqe->rdma_wr.rkey, IB_ACCESS_REMOTE_READ))) goto acc_err; - release = 0; + release = false; sqp->s_sge.sg_list = NULL; sqp->s_sge.num_sge = 1; qp->r_sge.sge = wqe->sg_list[0]; @@ -628,8 +579,8 @@ rnr_nak: spin_lock_irqsave(&sqp->s_lock, flags); if (!(ib_rvt_state_ops[sqp->state] & RVT_PROCESS_RECV_OK)) goto clr_busy; - to = ib_hfi1_rnr_table[qp->r_min_rnr_timer]; - hfi1_add_rnr_timer(sqp, to); + rvt_add_rnr_timer(sqp, qp->r_min_rnr_timer << + IB_AETH_CREDIT_SHIFT); goto clr_busy; op_err: @@ -647,7 +598,7 @@ acc_err: wc.status = IB_WC_LOC_PROT_ERR; err: /* responder goes to error state */ - hfi1_rc_error(qp, wc.status); + rvt_rc_error(qp, wc.status); serr: spin_lock_irqsave(&sqp->s_lock, flags); @@ -699,9 +650,9 @@ u32 hfi1_make_grh(struct hfi1_ibport *ibp, struct ib_grh *hdr, /* The SGID is 32-bit aligned. */ hdr->sgid.global.subnet_prefix = ibp->rvp.gid_prefix; hdr->sgid.global.interface_id = - grh->sgid_index && grh->sgid_index < ARRAY_SIZE(ibp->guids) ? - ibp->guids[grh->sgid_index - 1] : - cpu_to_be64(ppd_from_ibp(ibp)->guid); + grh->sgid_index < HFI1_GUIDS_PER_PORT ? + get_sguid(ibp, grh->sgid_index) : + get_sguid(ibp, HFI1_PORT_GUID_INDEX); hdr->dgid = grh->dgid; /* GRH header size in 32-bit words. */ @@ -777,8 +728,8 @@ void hfi1_make_ruc_header(struct rvt_qp *qp, struct ib_other_headers *ohdr, u32 bth1; /* Construct the header. */ - extra_bytes = -qp->s_cur_size & 3; - nwords = (qp->s_cur_size + extra_bytes) >> 2; + extra_bytes = -ps->s_txreq->s_cur_size & 3; + nwords = (ps->s_txreq->s_cur_size + extra_bytes) >> 2; lrh0 = HFI1_LRH_BTH; if (unlikely(qp->remote_ah_attr.ah_flags & IB_AH_GRH)) { qp->s_hdrwords += hfi1_make_grh(ibp, @@ -952,7 +903,6 @@ void hfi1_send_complete(struct rvt_qp *qp, struct rvt_swqe *wqe, enum ib_wc_status status) { u32 old_last, last; - unsigned i; if (!(ib_rvt_state_ops[qp->state] & RVT_PROCESS_OR_FLUSH_SEND)) return; @@ -964,32 +914,13 @@ void hfi1_send_complete(struct rvt_qp *qp, struct rvt_swqe *wqe, qp->s_last = last; /* See post_send() */ barrier(); - for (i = 0; i < wqe->wr.num_sge; i++) { - struct rvt_sge *sge = &wqe->sg_list[i]; - - rvt_put_mr(sge->mr); - } + rvt_put_swqe(wqe); if (qp->ibqp.qp_type == IB_QPT_UD || qp->ibqp.qp_type == IB_QPT_SMI || qp->ibqp.qp_type == IB_QPT_GSI) atomic_dec(&ibah_to_rvtah(wqe->ud_wr.ah)->refcount); - /* See ch. 11.2.4.1 and 10.7.3.1 */ - if (!(qp->s_flags & RVT_S_SIGNAL_REQ_WR) || - (wqe->wr.send_flags & IB_SEND_SIGNALED) || - status != IB_WC_SUCCESS) { - struct ib_wc wc; - - memset(&wc, 0, sizeof(wc)); - wc.wr_id = wqe->wr.wr_id; - wc.status = status; - wc.opcode = ib_hfi1_wc_opcode[wqe->wr.opcode]; - wc.qp = &qp->ibqp; - if (status == IB_WC_SUCCESS) - wc.byte_len = wqe->length; - rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.send_cq), &wc, - status != IB_WC_SUCCESS); - } + rvt_qp_swqe_complete(qp, wqe, status); if (qp->s_acked == old_last) qp->s_acked = last; diff --git a/drivers/infiniband/hw/hfi1/sdma.c b/drivers/infiniband/hw/hfi1/sdma.c index 9cbe52d21077..1d81cac1fa6c 100644 --- a/drivers/infiniband/hw/hfi1/sdma.c +++ b/drivers/infiniband/hw/hfi1/sdma.c @@ -375,7 +375,7 @@ static inline void complete_tx(struct sdma_engine *sde, sde->head_sn, tx->sn); sde->head_sn++; #endif - sdma_txclean(sde->dd, tx); + __sdma_txclean(sde->dd, tx); if (complete) (*complete)(tx, res); if (wait && iowait_sdma_dec(wait)) @@ -1643,7 +1643,7 @@ static inline u8 ahg_mode(struct sdma_txreq *tx) } /** - * sdma_txclean() - clean tx of mappings, descp *kmalloc's + * __sdma_txclean() - clean tx of mappings, descp *kmalloc's * @dd: hfi1_devdata for unmapping * @tx: tx request to clean * @@ -1653,7 +1653,7 @@ static inline u8 ahg_mode(struct sdma_txreq *tx) * The code can be called multiple times without issue. * */ -void sdma_txclean( +void __sdma_txclean( struct hfi1_devdata *dd, struct sdma_txreq *tx) { @@ -3065,7 +3065,7 @@ static int _extend_sdma_tx_descs(struct hfi1_devdata *dd, struct sdma_txreq *tx) tx->descp[i] = tx->descs[i]; return 0; enomem: - sdma_txclean(dd, tx); + __sdma_txclean(dd, tx); return -ENOMEM; } @@ -3094,14 +3094,14 @@ int ext_coal_sdma_tx_descs(struct hfi1_devdata *dd, struct sdma_txreq *tx, rval = _extend_sdma_tx_descs(dd, tx); if (rval) { - sdma_txclean(dd, tx); + __sdma_txclean(dd, tx); return rval; } /* If coalesce buffer is allocated, copy data into it */ if (tx->coalesce_buf) { if (type == SDMA_MAP_NONE) { - sdma_txclean(dd, tx); + __sdma_txclean(dd, tx); return -EINVAL; } @@ -3109,7 +3109,7 @@ int ext_coal_sdma_tx_descs(struct hfi1_devdata *dd, struct sdma_txreq *tx, kvaddr = kmap(page); kvaddr += offset; } else if (WARN_ON(!kvaddr)) { - sdma_txclean(dd, tx); + __sdma_txclean(dd, tx); return -EINVAL; } @@ -3139,7 +3139,7 @@ int ext_coal_sdma_tx_descs(struct hfi1_devdata *dd, struct sdma_txreq *tx, DMA_TO_DEVICE); if (unlikely(dma_mapping_error(&dd->pcidev->dev, addr))) { - sdma_txclean(dd, tx); + __sdma_txclean(dd, tx); return -ENOSPC; } @@ -3181,7 +3181,7 @@ int _pad_sdma_tx_descs(struct hfi1_devdata *dd, struct sdma_txreq *tx) if ((unlikely(tx->num_desc == tx->desc_limit))) { rval = _extend_sdma_tx_descs(dd, tx); if (rval) { - sdma_txclean(dd, tx); + __sdma_txclean(dd, tx); return rval; } } diff --git a/drivers/infiniband/hw/hfi1/sdma.h b/drivers/infiniband/hw/hfi1/sdma.h index 56257ea3598f..21f1e2834f37 100644 --- a/drivers/infiniband/hw/hfi1/sdma.h +++ b/drivers/infiniband/hw/hfi1/sdma.h @@ -667,7 +667,13 @@ int ext_coal_sdma_tx_descs(struct hfi1_devdata *dd, struct sdma_txreq *tx, int type, void *kvaddr, struct page *page, unsigned long offset, u16 len); int _pad_sdma_tx_descs(struct hfi1_devdata *, struct sdma_txreq *); -void sdma_txclean(struct hfi1_devdata *, struct sdma_txreq *); +void __sdma_txclean(struct hfi1_devdata *, struct sdma_txreq *); + +static inline void sdma_txclean(struct hfi1_devdata *dd, struct sdma_txreq *tx) +{ + if (tx->num_desc) + __sdma_txclean(dd, tx); +} /* helpers used by public routines */ static inline void _sdma_close_tx(struct hfi1_devdata *dd, @@ -753,7 +759,7 @@ static inline int sdma_txadd_page( DMA_TO_DEVICE); if (unlikely(dma_mapping_error(&dd->pcidev->dev, addr))) { - sdma_txclean(dd, tx); + __sdma_txclean(dd, tx); return -ENOSPC; } @@ -834,7 +840,7 @@ static inline int sdma_txadd_kvaddr( DMA_TO_DEVICE); if (unlikely(dma_mapping_error(&dd->pcidev->dev, addr))) { - sdma_txclean(dd, tx); + __sdma_txclean(dd, tx); return -ENOSPC; } diff --git a/drivers/infiniband/hw/hfi1/trace.c b/drivers/infiniband/hw/hfi1/trace.c index 01f525cd985a..e86798af6903 100644 --- a/drivers/infiniband/hw/hfi1/trace.c +++ b/drivers/infiniband/hw/hfi1/trace.c @@ -130,14 +130,14 @@ const char *parse_everbs_hdrs( case OP(RC, ACKNOWLEDGE): trace_seq_printf(p, AETH_PRN, be32_to_cpu(eh->aeth) >> 24, parse_syndrome(be32_to_cpu(eh->aeth) >> 24), - be32_to_cpu(eh->aeth) & HFI1_MSN_MASK); + be32_to_cpu(eh->aeth) & IB_MSN_MASK); break; /* aeth + atomicacketh */ case OP(RC, ATOMIC_ACKNOWLEDGE): trace_seq_printf(p, AETH_PRN " " ATOMICACKETH_PRN, be32_to_cpu(eh->at.aeth) >> 24, parse_syndrome(be32_to_cpu(eh->at.aeth) >> 24), - be32_to_cpu(eh->at.aeth) & HFI1_MSN_MASK, + be32_to_cpu(eh->at.aeth) & IB_MSN_MASK, ib_u64_get(&eh->at.atomic_ack_eth)); break; /* atomiceth */ diff --git a/drivers/infiniband/hw/hfi1/uc.c b/drivers/infiniband/hw/hfi1/uc.c index 5e6d1bac4914..4b2a8400c823 100644 --- a/drivers/infiniband/hw/hfi1/uc.c +++ b/drivers/infiniband/hw/hfi1/uc.c @@ -258,8 +258,8 @@ int hfi1_make_uc_req(struct rvt_qp *qp, struct hfi1_pkt_state *ps) qp->s_len -= len; qp->s_hdrwords = hwords; ps->s_txreq->sde = priv->s_sde; - qp->s_cur_sge = &qp->s_sge; - qp->s_cur_size = len; + ps->s_txreq->ss = &qp->s_sge; + ps->s_txreq->s_cur_size = len; hfi1_make_ruc_header(qp, ohdr, bth0 | (qp->s_state << 24), mask_psn(qp->s_psn++), middle, ps); /* pbc */ @@ -296,7 +296,7 @@ bail_no_tx: */ void hfi1_uc_rcv(struct hfi1_packet *packet) { - struct hfi1_ibport *ibp = &packet->rcd->ppd->ibport_data; + struct hfi1_ibport *ibp = rcd_to_iport(packet->rcd); struct ib_header *hdr = packet->hdr; u32 rcv_flags = packet->rcv_flags; void *data = packet->ebuf; @@ -384,7 +384,7 @@ inv: } if (qp->state == IB_QPS_RTR && !(qp->r_flags & RVT_R_COMM_EST)) - qp_comm_est(qp); + rvt_comm_est(qp); /* OK, process the packet. */ switch (opcode) { @@ -419,7 +419,7 @@ send_first: qp->r_rcv_len += pmtu; if (unlikely(qp->r_rcv_len > qp->r_len)) goto rewind; - hfi1_copy_sge(&qp->r_sge, data, pmtu, 0, 0); + hfi1_copy_sge(&qp->r_sge, data, pmtu, false, false); break; case OP(SEND_LAST_WITH_IMMEDIATE): @@ -444,7 +444,7 @@ send_last: if (unlikely(wc.byte_len > qp->r_len)) goto rewind; wc.opcode = IB_WC_RECV; - hfi1_copy_sge(&qp->r_sge, data, tlen, 0, 0); + hfi1_copy_sge(&qp->r_sge, data, tlen, false, false); rvt_put_ss(&qp->s_rdma_read_sge); last_imm: wc.wr_id = qp->r_wr_id; @@ -519,7 +519,7 @@ rdma_first: qp->r_rcv_len += pmtu; if (unlikely(qp->r_rcv_len > qp->r_len)) goto drop; - hfi1_copy_sge(&qp->r_sge, data, pmtu, 1, 0); + hfi1_copy_sge(&qp->r_sge, data, pmtu, true, false); break; case OP(RDMA_WRITE_LAST_WITH_IMMEDIATE): @@ -548,7 +548,7 @@ rdma_last_imm: } wc.byte_len = qp->r_len; wc.opcode = IB_WC_RECV_RDMA_WITH_IMM; - hfi1_copy_sge(&qp->r_sge, data, tlen, 1, 0); + hfi1_copy_sge(&qp->r_sge, data, tlen, true, false); rvt_put_ss(&qp->r_sge); goto last_imm; @@ -564,7 +564,7 @@ rdma_last: tlen -= (hdrsize + pad + 4); if (unlikely(tlen + qp->r_rcv_len != qp->r_len)) goto drop; - hfi1_copy_sge(&qp->r_sge, data, tlen, 1, 0); + hfi1_copy_sge(&qp->r_sge, data, tlen, true, false); rvt_put_ss(&qp->r_sge); break; @@ -584,5 +584,5 @@ drop: return; op_err: - hfi1_rc_error(qp, IB_WC_LOC_QP_OP_ERR); + rvt_rc_error(qp, IB_WC_LOC_QP_OP_ERR); } diff --git a/drivers/infiniband/hw/hfi1/ud.c b/drivers/infiniband/hw/hfi1/ud.c index 97ae24b6314c..13ea4eb6ef3d 100644 --- a/drivers/infiniband/hw/hfi1/ud.c +++ b/drivers/infiniband/hw/hfi1/ud.c @@ -167,7 +167,7 @@ static void ud_loopback(struct rvt_qp *sqp, struct rvt_swqe *swqe) ret = hfi1_rvt_get_rwqe(qp, 0); if (ret < 0) { - hfi1_rc_error(qp, IB_WC_LOC_QP_OP_ERR); + rvt_rc_error(qp, IB_WC_LOC_QP_OP_ERR); goto bail_unlock; } if (!ret) { @@ -189,10 +189,10 @@ static void ud_loopback(struct rvt_qp *sqp, struct rvt_swqe *swqe) hfi1_make_grh(ibp, &grh, &grd, 0, 0); hfi1_copy_sge(&qp->r_sge, &grh, - sizeof(grh), 1, 0); + sizeof(grh), true, false); wc.wc_flags |= IB_WC_GRH; } else { - hfi1_skip_sge(&qp->r_sge, sizeof(struct ib_grh), 1); + rvt_skip_sge(&qp->r_sge, sizeof(struct ib_grh), true); } ssge.sg_list = swqe->sg_list + 1; ssge.sge = *swqe->sg_list; @@ -206,7 +206,7 @@ static void ud_loopback(struct rvt_qp *sqp, struct rvt_swqe *swqe) if (len > sge->sge_length) len = sge->sge_length; WARN_ON_ONCE(len == 0); - hfi1_copy_sge(&qp->r_sge, sge->vaddr, len, 1, 0); + hfi1_copy_sge(&qp->r_sge, sge->vaddr, len, true, false); sge->vaddr += len; sge->length -= len; sge->sge_length -= len; @@ -354,8 +354,8 @@ int hfi1_make_ud_req(struct rvt_qp *qp, struct hfi1_pkt_state *ps) /* header size in 32-bit words LRH+BTH+DETH = (8+12+8)/4. */ qp->s_hdrwords = 7; - qp->s_cur_size = wqe->length; - qp->s_cur_sge = &qp->s_sge; + ps->s_txreq->s_cur_size = wqe->length; + ps->s_txreq->ss = &qp->s_sge; qp->s_srate = ah_attr->static_rate; qp->srate_mbps = ib_rate_to_mbps(qp->s_srate); qp->s_wqe = wqe; @@ -672,7 +672,7 @@ void hfi1_ud_rcv(struct hfi1_packet *packet) u32 src_qp; u16 dlid, pkey; int mgmt_pkey_idx = -1; - struct hfi1_ibport *ibp = &packet->rcd->ppd->ibport_data; + struct hfi1_ibport *ibp = rcd_to_iport(packet->rcd); struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); struct ib_header *hdr = packet->hdr; u32 rcv_flags = packet->rcv_flags; @@ -796,7 +796,7 @@ void hfi1_ud_rcv(struct hfi1_packet *packet) ret = hfi1_rvt_get_rwqe(qp, 0); if (ret < 0) { - hfi1_rc_error(qp, IB_WC_LOC_QP_OP_ERR); + rvt_rc_error(qp, IB_WC_LOC_QP_OP_ERR); return; } if (!ret) { @@ -812,13 +812,13 @@ void hfi1_ud_rcv(struct hfi1_packet *packet) } if (has_grh) { hfi1_copy_sge(&qp->r_sge, &hdr->u.l.grh, - sizeof(struct ib_grh), 1, 0); + sizeof(struct ib_grh), true, false); wc.wc_flags |= IB_WC_GRH; } else { - hfi1_skip_sge(&qp->r_sge, sizeof(struct ib_grh), 1); + rvt_skip_sge(&qp->r_sge, sizeof(struct ib_grh), true); } hfi1_copy_sge(&qp->r_sge, data, wc.byte_len - sizeof(struct ib_grh), - 1, 0); + true, false); rvt_put_ss(&qp->r_sge); if (!test_and_clear_bit(RVT_R_WRID_VALID, &qp->r_aflags)) return; diff --git a/drivers/infiniband/hw/hfi1/user_exp_rcv.c b/drivers/infiniband/hw/hfi1/user_exp_rcv.c index 64d26525435a..4a8295399e71 100644 --- a/drivers/infiniband/hw/hfi1/user_exp_rcv.c +++ b/drivers/infiniband/hw/hfi1/user_exp_rcv.c @@ -45,6 +45,7 @@ * */ #include <asm/page.h> +#include <linux/string.h> #include "user_exp_rcv.h" #include "trace.h" @@ -577,16 +578,10 @@ int hfi1_user_exp_rcv_clear(struct file *fp, struct hfi1_tid_info *tinfo) u32 *tidinfo; unsigned tididx; - tidinfo = kcalloc(tinfo->tidcnt, sizeof(*tidinfo), GFP_KERNEL); - if (!tidinfo) - return -ENOMEM; - - if (copy_from_user(tidinfo, (void __user *)(unsigned long) - tinfo->tidlist, sizeof(tidinfo[0]) * - tinfo->tidcnt)) { - ret = -EFAULT; - goto done; - } + tidinfo = memdup_user((void __user *)(unsigned long)tinfo->tidlist, + sizeof(tidinfo[0]) * tinfo->tidcnt); + if (IS_ERR(tidinfo)) + return PTR_ERR(tidinfo); mutex_lock(&uctxt->exp_lock); for (tididx = 0; tididx < tinfo->tidcnt; tididx++) { @@ -602,7 +597,7 @@ int hfi1_user_exp_rcv_clear(struct file *fp, struct hfi1_tid_info *tinfo) spin_unlock(&fd->tid_lock); tinfo->tidcnt = tididx; mutex_unlock(&uctxt->exp_lock); -done: + kfree(tidinfo); return ret; } diff --git a/drivers/infiniband/hw/hfi1/user_sdma.c b/drivers/infiniband/hw/hfi1/user_sdma.c index 77697d690f3e..e6811c4edc73 100644 --- a/drivers/infiniband/hw/hfi1/user_sdma.c +++ b/drivers/infiniband/hw/hfi1/user_sdma.c @@ -60,6 +60,7 @@ #include <linux/mmu_context.h> #include <linux/module.h> #include <linux/vmalloc.h> +#include <linux/string.h> #include "hfi.h" #include "sdma.h" @@ -115,6 +116,7 @@ MODULE_PARM_DESC(sdma_comp_size, "Size of User SDMA completion ring. Default: 12 #define KDETH_HCRC_LOWER_MASK 0xff #define AHG_KDETH_INTR_SHIFT 12 +#define AHG_KDETH_SH_SHIFT 13 #define PBC2LRH(x) ((((x) & 0xfff) << 2) - 4) #define LRH2PBC(x) ((((x) >> 2) + 1) & 0xfff) @@ -144,8 +146,9 @@ MODULE_PARM_DESC(sdma_comp_size, "Size of User SDMA completion ring. Default: 12 #define KDETH_OM_LARGE 64 #define KDETH_OM_MAX_SIZE (1 << ((KDETH_OM_LARGE / KDETH_OM_SMALL) + 1)) -/* Last packet in the request */ -#define TXREQ_FLAGS_REQ_LAST_PKT BIT(0) +/* Tx request flag bits */ +#define TXREQ_FLAGS_REQ_ACK BIT(0) /* Set the ACK bit in the header */ +#define TXREQ_FLAGS_REQ_DISABLE_SH BIT(1) /* Disable header suppression */ /* SDMA request flag bits */ #define SDMA_REQ_FOR_THREAD 1 @@ -723,30 +726,28 @@ int hfi1_user_sdma_process_request(struct file *fp, struct iovec *iovec, */ if (req_opcode(req->info.ctrl) == EXPECTED) { u16 ntids = iovec[idx].iov_len / sizeof(*req->tids); + u32 *tmp; if (!ntids || ntids > MAX_TID_PAIR_ENTRIES) { ret = -EINVAL; goto free_req; } - req->tids = kcalloc(ntids, sizeof(*req->tids), GFP_KERNEL); - if (!req->tids) { - ret = -ENOMEM; - goto free_req; - } + /* * We have to copy all of the tids because they may vary * in size and, therefore, the TID count might not be * equal to the pkt count. However, there is no way to * tell at this point. */ - ret = copy_from_user(req->tids, iovec[idx].iov_base, - ntids * sizeof(*req->tids)); - if (ret) { + tmp = memdup_user(iovec[idx].iov_base, + ntids * sizeof(*req->tids)); + if (IS_ERR(tmp)) { + ret = PTR_ERR(tmp); SDMA_DBG(req, "Failed to copy %d TIDs (%d)", ntids, ret); - ret = -EFAULT; goto free_req; } + req->tids = tmp; req->n_tids = ntids; idx++; } @@ -943,8 +944,13 @@ static int user_sdma_send_pkts(struct user_sdma_request *req, unsigned maxpkts) tx->busycount = 0; INIT_LIST_HEAD(&tx->list); + /* + * For the last packet set the ACK request + * and disable header suppression. + */ if (req->seqnum == req->info.npkts - 1) - tx->flags |= TXREQ_FLAGS_REQ_LAST_PKT; + tx->flags |= (TXREQ_FLAGS_REQ_ACK | + TXREQ_FLAGS_REQ_DISABLE_SH); /* * Calculate the payload size - this is min of the fragment @@ -963,11 +969,22 @@ static int user_sdma_send_pkts(struct user_sdma_request *req, unsigned maxpkts) } datalen = compute_data_length(req, tx); + + /* + * Disable header suppression for the payload <= 8DWS. + * If there is an uncorrectable error in the receive + * data FIFO when the received payload size is less than + * or equal to 8DWS then the RxDmaDataFifoRdUncErr is + * not reported.There is set RHF.EccErr if the header + * is not suppressed. + */ if (!datalen) { SDMA_DBG(req, "Request has data but pkt len is 0"); ret = -EFAULT; goto free_tx; + } else if (datalen <= 32) { + tx->flags |= TXREQ_FLAGS_REQ_DISABLE_SH; } } @@ -990,6 +1007,10 @@ static int user_sdma_send_pkts(struct user_sdma_request *req, unsigned maxpkts) LRH2PBC(lrhlen); tx->hdr.pbc[0] = cpu_to_le16(pbclen); } + ret = check_header_template(req, &tx->hdr, + lrhlen, datalen); + if (ret) + goto free_tx; ret = sdma_txinit_ahg(&tx->txreq, SDMA_TXREQ_F_AHG_COPY, sizeof(tx->hdr) + datalen, @@ -1351,7 +1372,7 @@ static int set_txreq_header(struct user_sdma_request *req, req->seqnum)); /* Set ACK request on last packet */ - if (unlikely(tx->flags & TXREQ_FLAGS_REQ_LAST_PKT)) + if (unlikely(tx->flags & TXREQ_FLAGS_REQ_ACK)) hdr->bth[2] |= cpu_to_be32(1UL << 31); /* Set the new offset */ @@ -1384,8 +1405,8 @@ static int set_txreq_header(struct user_sdma_request *req, /* Set KDETH.TID based on value for this TID */ KDETH_SET(hdr->kdeth.ver_tid_offset, TID, EXP_TID_GET(tidval, IDX)); - /* Clear KDETH.SH only on the last packet */ - if (unlikely(tx->flags & TXREQ_FLAGS_REQ_LAST_PKT)) + /* Clear KDETH.SH when DISABLE_SH flag is set */ + if (unlikely(tx->flags & TXREQ_FLAGS_REQ_DISABLE_SH)) KDETH_SET(hdr->kdeth.ver_tid_offset, SH, 0); /* * Set the KDETH.OFFSET and KDETH.OM based on size of @@ -1429,7 +1450,7 @@ static int set_txreq_header_ahg(struct user_sdma_request *req, /* BTH.PSN and BTH.A */ val32 = (be32_to_cpu(hdr->bth[2]) + req->seqnum) & (HFI1_CAP_IS_KSET(EXTENDED_PSN) ? 0x7fffffff : 0xffffff); - if (unlikely(tx->flags & TXREQ_FLAGS_REQ_LAST_PKT)) + if (unlikely(tx->flags & TXREQ_FLAGS_REQ_ACK)) val32 |= 1UL << 31; AHG_HEADER_SET(req->ahg, diff, 6, 0, 16, cpu_to_be16(val32 >> 16)); AHG_HEADER_SET(req->ahg, diff, 6, 16, 16, cpu_to_be16(val32 & 0xffff)); @@ -1468,19 +1489,23 @@ static int set_txreq_header_ahg(struct user_sdma_request *req, AHG_HEADER_SET(req->ahg, diff, 7, 0, 16, ((!!(req->omfactor - KDETH_OM_SMALL)) << 15 | ((req->tidoffset / req->omfactor) & 0x7fff))); - /* KDETH.TIDCtrl, KDETH.TID */ + /* KDETH.TIDCtrl, KDETH.TID, KDETH.Intr, KDETH.SH */ val = cpu_to_le16(((EXP_TID_GET(tidval, CTRL) & 0x3) << 10) | - (EXP_TID_GET(tidval, IDX) & 0x3ff)); - /* Clear KDETH.SH on last packet */ - if (unlikely(tx->flags & TXREQ_FLAGS_REQ_LAST_PKT)) { - val |= cpu_to_le16(KDETH_GET(hdr->kdeth.ver_tid_offset, - INTR) << - AHG_KDETH_INTR_SHIFT); - val &= cpu_to_le16(~(1U << 13)); - AHG_HEADER_SET(req->ahg, diff, 7, 16, 14, val); + (EXP_TID_GET(tidval, IDX) & 0x3ff)); + + if (unlikely(tx->flags & TXREQ_FLAGS_REQ_DISABLE_SH)) { + val |= cpu_to_le16((KDETH_GET(hdr->kdeth.ver_tid_offset, + INTR) << + AHG_KDETH_INTR_SHIFT)); } else { - AHG_HEADER_SET(req->ahg, diff, 7, 16, 12, val); + val |= KDETH_GET(hdr->kdeth.ver_tid_offset, SH) ? + cpu_to_le16(0x1 << AHG_KDETH_SH_SHIFT) : + cpu_to_le16((KDETH_GET(hdr->kdeth.ver_tid_offset, + INTR) << + AHG_KDETH_INTR_SHIFT)); } + + AHG_HEADER_SET(req->ahg, diff, 7, 16, 14, val); } trace_hfi1_sdma_user_header_ahg(pq->dd, pq->ctxt, pq->subctxt, diff --git a/drivers/infiniband/hw/hfi1/verbs.c b/drivers/infiniband/hw/hfi1/verbs.c index 4b7a16ceb362..222315fadab1 100644 --- a/drivers/infiniband/hw/hfi1/verbs.c +++ b/drivers/infiniband/hw/hfi1/verbs.c @@ -291,28 +291,12 @@ static void wss_insert(void *address) /* * Is the working set larger than the threshold? */ -static inline int wss_exceeds_threshold(void) +static inline bool wss_exceeds_threshold(void) { return atomic_read(&wss.total_count) >= wss.threshold; } /* - * Translate ib_wr_opcode into ib_wc_opcode. - */ -const enum ib_wc_opcode ib_hfi1_wc_opcode[] = { - [IB_WR_RDMA_WRITE] = IB_WC_RDMA_WRITE, - [IB_WR_RDMA_WRITE_WITH_IMM] = IB_WC_RDMA_WRITE, - [IB_WR_SEND] = IB_WC_SEND, - [IB_WR_SEND_WITH_IMM] = IB_WC_SEND, - [IB_WR_RDMA_READ] = IB_WC_RDMA_READ, - [IB_WR_ATOMIC_CMP_AND_SWP] = IB_WC_COMP_SWAP, - [IB_WR_ATOMIC_FETCH_AND_ADD] = IB_WC_FETCH_ADD, - [IB_WR_SEND_WITH_INV] = IB_WC_SEND, - [IB_WR_LOCAL_INV] = IB_WC_LOCAL_INV, - [IB_WR_REG_MR] = IB_WC_REG_MR -}; - -/* * Length of header by opcode, 0 --> not supported */ const u8 hdr_len_by_opcode[256] = { @@ -435,18 +419,19 @@ __be64 ib_hfi1_sys_image_guid; * @ss: the SGE state * @data: the data to copy * @length: the length of the data + * @release: boolean to release MR * @copy_last: do a separate copy of the last 8 bytes */ void hfi1_copy_sge( struct rvt_sge_state *ss, void *data, u32 length, - int release, - int copy_last) + bool release, + bool copy_last) { struct rvt_sge *sge = &ss->sge; - int in_last = 0; int i; - int cacheless_copy = 0; + bool in_last = false; + bool cacheless_copy = false; if (sge_copy_mode == COPY_CACHELESS) { cacheless_copy = length >= PAGE_SIZE; @@ -470,19 +455,15 @@ void hfi1_copy_sge( if (length > 8) { length -= 8; } else { - copy_last = 0; - in_last = 1; + copy_last = false; + in_last = true; } } again: while (length) { - u32 len = sge->length; + u32 len = rvt_get_sge_length(sge, length); - if (len > length) - len = length; - if (len > sge->sge_length) - len = sge->sge_length; WARN_ON_ONCE(len == 0); if (unlikely(in_last)) { /* enforce byte transfer ordering */ @@ -493,77 +474,19 @@ again: } else { memcpy(sge->vaddr, data, len); } - sge->vaddr += len; - sge->length -= len; - sge->sge_length -= len; - if (sge->sge_length == 0) { - if (release) - rvt_put_mr(sge->mr); - if (--ss->num_sge) - *sge = *ss->sg_list++; - } else if (sge->length == 0 && sge->mr->lkey) { - if (++sge->n >= RVT_SEGSZ) { - if (++sge->m >= sge->mr->mapsz) - break; - sge->n = 0; - } - sge->vaddr = - sge->mr->map[sge->m]->segs[sge->n].vaddr; - sge->length = - sge->mr->map[sge->m]->segs[sge->n].length; - } + rvt_update_sge(ss, len, release); data += len; length -= len; } if (copy_last) { - copy_last = 0; - in_last = 1; + copy_last = false; + in_last = true; length = 8; goto again; } } -/** - * hfi1_skip_sge - skip over SGE memory - * @ss: the SGE state - * @length: the number of bytes to skip - */ -void hfi1_skip_sge(struct rvt_sge_state *ss, u32 length, int release) -{ - struct rvt_sge *sge = &ss->sge; - - while (length) { - u32 len = sge->length; - - if (len > length) - len = length; - if (len > sge->sge_length) - len = sge->sge_length; - WARN_ON_ONCE(len == 0); - sge->vaddr += len; - sge->length -= len; - sge->sge_length -= len; - if (sge->sge_length == 0) { - if (release) - rvt_put_mr(sge->mr); - if (--ss->num_sge) - *sge = *ss->sg_list++; - } else if (sge->length == 0 && sge->mr->lkey) { - if (++sge->n >= RVT_SEGSZ) { - if (++sge->m >= sge->mr->mapsz) - break; - sge->n = 0; - } - sge->vaddr = - sge->mr->map[sge->m]->segs[sge->n].vaddr; - sge->length = - sge->mr->map[sge->m]->segs[sge->n].length; - } - length -= len; - } -} - /* * Make sure the QP is ready and able to accept the given opcode. */ @@ -592,7 +515,7 @@ void hfi1_ib_rcv(struct hfi1_packet *packet) struct ib_header *hdr = packet->hdr; u32 tlen = packet->tlen; struct hfi1_pportdata *ppd = rcd->ppd; - struct hfi1_ibport *ibp = &ppd->ibport_data; + struct hfi1_ibport *ibp = rcd_to_iport(rcd); struct rvt_dev_info *rdi = &ppd->dd->verbs_dev.rdi; opcode_handler packet_handler; unsigned long flags; @@ -694,6 +617,7 @@ static void mem_timer(unsigned long data) qp = iowait_to_qp(wait); priv = qp->priv; list_del_init(&priv->s_iowait.list); + priv->s_iowait.lock = NULL; /* refcount held until actual wake up */ if (!list_empty(list)) mod_timer(&dev->mem_timer, jiffies + 1); @@ -704,27 +628,6 @@ static void mem_timer(unsigned long data) hfi1_qp_wakeup(qp, RVT_S_WAIT_KMEM); } -void update_sge(struct rvt_sge_state *ss, u32 length) -{ - struct rvt_sge *sge = &ss->sge; - - sge->vaddr += length; - sge->length -= length; - sge->sge_length -= length; - if (sge->sge_length == 0) { - if (--ss->num_sge) - *sge = *ss->sg_list++; - } else if (sge->length == 0 && sge->mr->lkey) { - if (++sge->n >= RVT_SEGSZ) { - if (++sge->m >= sge->mr->mapsz) - return; - sge->n = 0; - } - sge->vaddr = sge->mr->map[sge->m]->segs[sge->n].vaddr; - sge->length = sge->mr->map[sge->m]->segs[sge->n].length; - } -} - /* * This is called with progress side lock held. */ @@ -769,6 +672,7 @@ static int wait_kmem(struct hfi1_ibdev *dev, mod_timer(&dev->mem_timer, jiffies + 1); qp->s_flags |= RVT_S_WAIT_KMEM; list_add_tail(&priv->s_iowait.list, &dev->memwait); + priv->s_iowait.lock = &dev->iowait_lock; trace_hfi1_qpsleep(qp, RVT_S_WAIT_KMEM); rvt_get_qp(qp); } @@ -788,10 +692,10 @@ static int wait_kmem(struct hfi1_ibdev *dev, */ static noinline int build_verbs_ulp_payload( struct sdma_engine *sde, - struct rvt_sge_state *ss, u32 length, struct verbs_txreq *tx) { + struct rvt_sge_state *ss = tx->ss; struct rvt_sge *sg_list = ss->sg_list; struct rvt_sge sge = ss->sge; u8 num_sge = ss->num_sge; @@ -812,7 +716,7 @@ static noinline int build_verbs_ulp_payload( len); if (ret) goto bail_txadd; - update_sge(ss, len); + rvt_update_sge(ss, len, false); length -= len; } return ret; @@ -835,7 +739,6 @@ bail_txadd: /* New API */ static int build_verbs_tx_desc( struct sdma_engine *sde, - struct rvt_sge_state *ss, u32 length, struct verbs_txreq *tx, struct hfi1_ahg_info *ahg_info, @@ -879,9 +782,9 @@ static int build_verbs_tx_desc( goto bail_txadd; } - /* add the ulp payload - if any. ss can be NULL for acks */ - if (ss) - ret = build_verbs_ulp_payload(sde, ss, length, tx); + /* add the ulp payload - if any. tx->ss can be NULL for acks */ + if (tx->ss) + ret = build_verbs_ulp_payload(sde, length, tx); bail_txadd: return ret; } @@ -892,8 +795,7 @@ int hfi1_verbs_send_dma(struct rvt_qp *qp, struct hfi1_pkt_state *ps, struct hfi1_qp_priv *priv = qp->priv; struct hfi1_ahg_info *ahg_info = priv->s_ahg; u32 hdrwords = qp->s_hdrwords; - struct rvt_sge_state *ss = qp->s_cur_sge; - u32 len = qp->s_cur_size; + u32 len = ps->s_txreq->s_cur_size; u32 plen = hdrwords + ((len + 3) >> 2) + 2; /* includes pbc */ struct hfi1_ibdev *dev = ps->dev; struct hfi1_pportdata *ppd = ps->ppd; @@ -918,7 +820,7 @@ int hfi1_verbs_send_dma(struct rvt_qp *qp, struct hfi1_pkt_state *ps, plen); } tx->wqe = qp->s_wqe; - ret = build_verbs_tx_desc(tx->sde, ss, len, tx, ahg_info, pbc); + ret = build_verbs_tx_desc(tx->sde, len, tx, ahg_info, pbc); if (unlikely(ret)) goto bail_build; } @@ -980,6 +882,7 @@ static int pio_wait(struct rvt_qp *qp, qp->s_flags |= flag; was_empty = list_empty(&sc->piowait); list_add_tail(&priv->s_iowait.list, &sc->piowait); + priv->s_iowait.lock = &dev->iowait_lock; trace_hfi1_qpsleep(qp, RVT_S_WAIT_PIO); rvt_get_qp(qp); /* counting: only call wantpiobuf_intr if first user */ @@ -1008,8 +911,8 @@ int hfi1_verbs_send_pio(struct rvt_qp *qp, struct hfi1_pkt_state *ps, { struct hfi1_qp_priv *priv = qp->priv; u32 hdrwords = qp->s_hdrwords; - struct rvt_sge_state *ss = qp->s_cur_sge; - u32 len = qp->s_cur_size; + struct rvt_sge_state *ss = ps->s_txreq->ss; + u32 len = ps->s_txreq->s_cur_size; u32 dwords = (len + 3) >> 2; u32 plen = hdrwords + dwords + 2; /* includes pbc */ struct hfi1_pportdata *ppd = ps->ppd; @@ -1088,7 +991,7 @@ int hfi1_verbs_send_pio(struct rvt_qp *qp, struct hfi1_pkt_state *ps, if (slen > len) slen = len; - update_sge(ss, slen); + rvt_update_sge(ss, slen, false); seg_pio_copy_mid(pbuf, addr, slen); len -= slen; } @@ -1237,7 +1140,7 @@ static inline send_routine get_send_routine(struct rvt_qp *qp, u8 op = get_opcode(h); if (piothreshold && - qp->s_cur_size <= min(piothreshold, qp->pmtu) && + tx->s_cur_size <= min(piothreshold, qp->pmtu) && (BIT(op & OPMASK) & pio_opmask[op >> 5]) && iowait_sdma_pending(&priv->s_iowait) == 0 && !sdma_txreq_built(&tx->txreq)) @@ -1399,6 +1302,7 @@ static int query_port(struct rvt_dev_info *rdi, u8 port_num, struct hfi1_pportdata *ppd = &dd->pport[port_num - 1]; u16 lid = ppd->lid; + /* props being zeroed by the caller, avoid zeroing it here */ props->lid = lid ? lid : 0; props->lmc = ppd->lmc; /* OPA logical states match IB logical states */ @@ -1483,15 +1387,11 @@ static int hfi1_get_guid_be(struct rvt_dev_info *rdi, struct rvt_ibport *rvp, int guid_index, __be64 *guid) { struct hfi1_ibport *ibp = container_of(rvp, struct hfi1_ibport, rvp); - struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); - if (guid_index == 0) - *guid = cpu_to_be64(ppd->guid); - else if (guid_index < HFI1_GUIDS_PER_PORT) - *guid = ibp->guids[guid_index - 1]; - else + if (guid_index >= HFI1_GUIDS_PER_PORT) return -EINVAL; + *guid = get_sguid(ibp, guid_index); return 0; } @@ -1610,6 +1510,154 @@ static void hfi1_get_dev_fw_str(struct ib_device *ibdev, char *str, dc8051_ver_min(ver)); } +static const char * const driver_cntr_names[] = { + /* must be element 0*/ + "DRIVER_KernIntr", + "DRIVER_ErrorIntr", + "DRIVER_Tx_Errs", + "DRIVER_Rcv_Errs", + "DRIVER_HW_Errs", + "DRIVER_NoPIOBufs", + "DRIVER_CtxtsOpen", + "DRIVER_RcvLen_Errs", + "DRIVER_EgrBufFull", + "DRIVER_EgrHdrFull" +}; + +static const char **dev_cntr_names; +static const char **port_cntr_names; +static int num_driver_cntrs = ARRAY_SIZE(driver_cntr_names); +static int num_dev_cntrs; +static int num_port_cntrs; +static int cntr_names_initialized; + +/* + * Convert a list of names separated by '\n' into an array of NULL terminated + * strings. Optionally some entries can be reserved in the array to hold extra + * external strings. + */ +static int init_cntr_names(const char *names_in, + const size_t names_len, + int num_extra_names, + int *num_cntrs, + const char ***cntr_names) +{ + char *names_out, *p, **q; + int i, n; + + n = 0; + for (i = 0; i < names_len; i++) + if (names_in[i] == '\n') + n++; + + names_out = kmalloc((n + num_extra_names) * sizeof(char *) + names_len, + GFP_KERNEL); + if (!names_out) { + *num_cntrs = 0; + *cntr_names = NULL; + return -ENOMEM; + } + + p = names_out + (n + num_extra_names) * sizeof(char *); + memcpy(p, names_in, names_len); + + q = (char **)names_out; + for (i = 0; i < n; i++) { + q[i] = p; + p = strchr(p, '\n'); + *p++ = '\0'; + } + + *num_cntrs = n; + *cntr_names = (const char **)names_out; + return 0; +} + +static struct rdma_hw_stats *alloc_hw_stats(struct ib_device *ibdev, + u8 port_num) +{ + int i, err; + + if (!cntr_names_initialized) { + struct hfi1_devdata *dd = dd_from_ibdev(ibdev); + + err = init_cntr_names(dd->cntrnames, + dd->cntrnameslen, + num_driver_cntrs, + &num_dev_cntrs, + &dev_cntr_names); + if (err) + return NULL; + + for (i = 0; i < num_driver_cntrs; i++) + dev_cntr_names[num_dev_cntrs + i] = + driver_cntr_names[i]; + + err = init_cntr_names(dd->portcntrnames, + dd->portcntrnameslen, + 0, + &num_port_cntrs, + &port_cntr_names); + if (err) { + kfree(dev_cntr_names); + dev_cntr_names = NULL; + return NULL; + } + cntr_names_initialized = 1; + } + + if (!port_num) + return rdma_alloc_hw_stats_struct( + dev_cntr_names, + num_dev_cntrs + num_driver_cntrs, + RDMA_HW_STATS_DEFAULT_LIFESPAN); + else + return rdma_alloc_hw_stats_struct( + port_cntr_names, + num_port_cntrs, + RDMA_HW_STATS_DEFAULT_LIFESPAN); +} + +static u64 hfi1_sps_ints(void) +{ + unsigned long flags; + struct hfi1_devdata *dd; + u64 sps_ints = 0; + + spin_lock_irqsave(&hfi1_devs_lock, flags); + list_for_each_entry(dd, &hfi1_dev_list, list) { + sps_ints += get_all_cpu_total(dd->int_counter); + } + spin_unlock_irqrestore(&hfi1_devs_lock, flags); + return sps_ints; +} + +static int get_hw_stats(struct ib_device *ibdev, struct rdma_hw_stats *stats, + u8 port, int index) +{ + u64 *values; + int count; + + if (!port) { + u64 *stats = (u64 *)&hfi1_stats; + int i; + + hfi1_read_cntrs(dd_from_ibdev(ibdev), NULL, &values); + values[num_dev_cntrs] = hfi1_sps_ints(); + for (i = 1; i < num_driver_cntrs; i++) + values[num_dev_cntrs + i] = stats[i]; + count = num_dev_cntrs + num_driver_cntrs; + } else { + struct hfi1_ibport *ibp = to_iport(ibdev, port); + + hfi1_read_portcntrs(ppd_from_ibp(ibp), NULL, &values); + count = num_port_cntrs; + } + + memcpy(stats->value, values, count * sizeof(u64)); + return count; +} + /** * hfi1_register_ib_device - register our device with the infiniband core * @dd: the device data structure @@ -1620,6 +1668,7 @@ int hfi1_register_ib_device(struct hfi1_devdata *dd) struct hfi1_ibdev *dev = &dd->verbs_dev; struct ib_device *ibdev = &dev->rdi.ibdev; struct hfi1_pportdata *ppd = dd->pport; + struct hfi1_ibport *ibp = &ppd->ibport_data; unsigned i; int ret; size_t lcpysz = IB_DEVICE_NAME_MAX; @@ -1632,6 +1681,7 @@ int hfi1_register_ib_device(struct hfi1_devdata *dd) setup_timer(&dev->mem_timer, mem_timer, (unsigned long)dev); seqlock_init(&dev->iowait_lock); + seqlock_init(&dev->txwait_lock); INIT_LIST_HEAD(&dev->txwait); INIT_LIST_HEAD(&dev->memwait); @@ -1639,20 +1689,24 @@ int hfi1_register_ib_device(struct hfi1_devdata *dd) if (ret) goto err_verbs_txreq; + /* Use first-port GUID as node guid */ + ibdev->node_guid = get_sguid(ibp, HFI1_PORT_GUID_INDEX); + /* * The system image GUID is supposed to be the same for all * HFIs in a single system but since there can be other * device types in the system, we can't be sure this is unique. */ if (!ib_hfi1_sys_image_guid) - ib_hfi1_sys_image_guid = cpu_to_be64(ppd->guid); + ib_hfi1_sys_image_guid = ibdev->node_guid; lcpysz = strlcpy(ibdev->name, class_name(), lcpysz); strlcpy(ibdev->name + lcpysz, "_%d", IB_DEVICE_NAME_MAX - lcpysz); ibdev->owner = THIS_MODULE; - ibdev->node_guid = cpu_to_be64(ppd->guid); ibdev->phys_port_cnt = dd->num_pports; - ibdev->dma_device = &dd->pcidev->dev; + ibdev->dev.parent = &dd->pcidev->dev; ibdev->modify_device = modify_device; + ibdev->alloc_hw_stats = alloc_hw_stats; + ibdev->get_hw_stats = get_hw_stats; /* keep process mad in the driver */ ibdev->process_mad = hfi1_process_mad; @@ -1710,6 +1764,7 @@ int hfi1_register_ib_device(struct hfi1_devdata *dd) dd->verbs_dev.rdi.driver_f.mtu_to_path_mtu = mtu_to_path_mtu; dd->verbs_dev.rdi.driver_f.check_modify_qp = hfi1_check_modify_qp; dd->verbs_dev.rdi.driver_f.modify_qp = hfi1_modify_qp; + dd->verbs_dev.rdi.driver_f.notify_restart_rc = hfi1_restart_rc; dd->verbs_dev.rdi.driver_f.check_send_wqe = hfi1_check_send_wqe; /* completeion queue */ @@ -1767,11 +1822,15 @@ void hfi1_unregister_ib_device(struct hfi1_devdata *dd) del_timer_sync(&dev->mem_timer); verbs_txreq_exit(dev); + + kfree(dev_cntr_names); + kfree(port_cntr_names); + cntr_names_initialized = 0; } void hfi1_cnp_rcv(struct hfi1_packet *packet) { - struct hfi1_ibport *ibp = &packet->rcd->ppd->ibport_data; + struct hfi1_ibport *ibp = rcd_to_iport(packet->rcd); struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); struct ib_header *hdr = packet->hdr; struct rvt_qp *qp = packet->qp; diff --git a/drivers/infiniband/hw/hfi1/verbs.h b/drivers/infiniband/hw/hfi1/verbs.h index 1c3815d89eb7..3a0b589e41c2 100644 --- a/drivers/infiniband/hw/hfi1/verbs.h +++ b/drivers/infiniband/hw/hfi1/verbs.h @@ -73,7 +73,6 @@ struct hfi1_packet; #include "iowait.h" #define HFI1_MAX_RDMA_ATOMIC 16 -#define HFI1_GUIDS_PER_PORT 5 /* * Increment this value if any changes that break userspace ABI @@ -128,7 +127,6 @@ struct hfi1_qp_priv { u8 s_sc; /* SC[0..4] for next packet */ u8 r_adefered; /* number of acks defered */ struct iowait s_iowait; - struct timer_list s_rnr_timer; struct rvt_qp *owner; }; @@ -169,8 +167,6 @@ struct hfi1_ibport { struct rvt_qp __rcu *qp[2]; struct rvt_ibport rvp; - __be64 guids[HFI1_GUIDS_PER_PORT - 1]; /* writable GUIDs */ - /* the first 16 entries are sl_to_vl for !OPA */ u8 sl_to_sc[32]; u8 sc_to_sl[32]; @@ -180,18 +176,19 @@ struct hfi1_ibdev { struct rvt_dev_info rdi; /* Must be first */ /* QP numbers are shared by all IB ports */ - /* protect wait lists */ - seqlock_t iowait_lock; + /* protect txwait list */ + seqlock_t txwait_lock ____cacheline_aligned_in_smp; struct list_head txwait; /* list for wait verbs_txreq */ struct list_head memwait; /* list for wait kernel memory */ - struct list_head txreq_free; struct kmem_cache *verbs_txreq_cache; - struct timer_list mem_timer; + u64 n_txwait; + u64 n_kmem_wait; + /* protect iowait lists */ + seqlock_t iowait_lock ____cacheline_aligned_in_smp; u64 n_piowait; u64 n_piodrain; - u64 n_txwait; - u64 n_kmem_wait; + struct timer_list mem_timer; #ifdef CONFIG_DEBUG_FS /* per HFI debugfs */ @@ -262,15 +259,6 @@ int hfi1_process_mad(struct ib_device *ibdev, int mad_flags, u8 port, #define PSN_MODIFY_MASK 0xFFFFFF /* - * Compare the lower 24 bits of the msn values. - * Returns an integer <, ==, or > than zero. - */ -static inline int cmp_msn(u32 a, u32 b) -{ - return (((int)a) - ((int)b)) << 8; -} - -/* * Compare two PSNs * Returns an integer <, ==, or > than zero. */ @@ -301,9 +289,7 @@ void hfi1_put_txreq(struct verbs_txreq *tx); int hfi1_verbs_send(struct rvt_qp *qp, struct hfi1_pkt_state *ps); void hfi1_copy_sge(struct rvt_sge_state *ss, void *data, u32 length, - int release, int copy_last); - -void hfi1_skip_sge(struct rvt_sge_state *ss, u32 length, int release); + bool release, bool copy_last); void hfi1_cnp_rcv(struct hfi1_packet *packet); @@ -321,16 +307,8 @@ u8 ah_to_sc(struct ib_device *ibdev, struct ib_ah_attr *ah_attr); struct ib_ah *hfi1_create_qp0_ah(struct hfi1_ibport *ibp, u16 dlid); -void hfi1_rc_rnr_retry(unsigned long arg); -void hfi1_add_rnr_timer(struct rvt_qp *qp, u32 to); -void hfi1_rc_timeout(unsigned long arg); -void hfi1_del_timers_sync(struct rvt_qp *qp); -void hfi1_stop_rc_timers(struct rvt_qp *qp); - void hfi1_rc_send_complete(struct rvt_qp *qp, struct ib_header *hdr); -void hfi1_rc_error(struct rvt_qp *qp, enum ib_wc_status err); - void hfi1_ud_rcv(struct hfi1_packet *packet); int hfi1_lookup_pkey_idx(struct hfi1_ibport *ibp, u16 pkey); @@ -344,7 +322,7 @@ int hfi1_check_modify_qp(struct rvt_qp *qp, struct ib_qp_attr *attr, void hfi1_modify_qp(struct rvt_qp *qp, struct ib_qp_attr *attr, int attr_mask, struct ib_udata *udata); - +void hfi1_restart_rc(struct rvt_qp *qp, u32 psn, int wait); int hfi1_check_send_wqe(struct rvt_qp *qp, struct rvt_swqe *wqe); extern const u32 rc_only_opcode; diff --git a/drivers/infiniband/hw/hfi1/verbs_txreq.c b/drivers/infiniband/hw/hfi1/verbs_txreq.c index 094ab829ec42..5d23172c470f 100644 --- a/drivers/infiniband/hw/hfi1/verbs_txreq.c +++ b/drivers/infiniband/hw/hfi1/verbs_txreq.c @@ -72,22 +72,22 @@ void hfi1_put_txreq(struct verbs_txreq *tx) kmem_cache_free(dev->verbs_txreq_cache, tx); do { - seq = read_seqbegin(&dev->iowait_lock); + seq = read_seqbegin(&dev->txwait_lock); if (!list_empty(&dev->txwait)) { struct iowait *wait; - write_seqlock_irqsave(&dev->iowait_lock, flags); + write_seqlock_irqsave(&dev->txwait_lock, flags); wait = list_first_entry(&dev->txwait, struct iowait, list); qp = iowait_to_qp(wait); priv = qp->priv; list_del_init(&priv->s_iowait.list); /* refcount held until actual wake up */ - write_sequnlock_irqrestore(&dev->iowait_lock, flags); + write_sequnlock_irqrestore(&dev->txwait_lock, flags); hfi1_qp_wakeup(qp, RVT_S_WAIT_TX); break; } - } while (read_seqretry(&dev->iowait_lock, seq)); + } while (read_seqretry(&dev->txwait_lock, seq)); } struct verbs_txreq *__get_txreq(struct hfi1_ibdev *dev, @@ -96,7 +96,7 @@ struct verbs_txreq *__get_txreq(struct hfi1_ibdev *dev, { struct verbs_txreq *tx = ERR_PTR(-EBUSY); - write_seqlock(&dev->iowait_lock); + write_seqlock(&dev->txwait_lock); if (ib_rvt_state_ops[qp->state] & RVT_PROCESS_RECV_OK) { struct hfi1_qp_priv *priv; @@ -108,13 +108,14 @@ struct verbs_txreq *__get_txreq(struct hfi1_ibdev *dev, dev->n_txwait++; qp->s_flags |= RVT_S_WAIT_TX; list_add_tail(&priv->s_iowait.list, &dev->txwait); + priv->s_iowait.lock = &dev->txwait_lock; trace_hfi1_qpsleep(qp, RVT_S_WAIT_TX); rvt_get_qp(qp); } qp->s_flags &= ~RVT_S_BUSY; } out: - write_sequnlock(&dev->iowait_lock); + write_sequnlock(&dev->txwait_lock); return tx; } diff --git a/drivers/infiniband/hw/hfi1/verbs_txreq.h b/drivers/infiniband/hw/hfi1/verbs_txreq.h index 5660897593ba..76216f2ef35a 100644 --- a/drivers/infiniband/hw/hfi1/verbs_txreq.h +++ b/drivers/infiniband/hw/hfi1/verbs_txreq.h @@ -65,6 +65,7 @@ struct verbs_txreq { struct sdma_engine *sde; struct send_context *psc; u16 hdr_dwords; + u16 s_cur_size; }; struct hfi1_ibdev; diff --git a/drivers/infiniband/hw/hns/hns_roce_ah.c b/drivers/infiniband/hw/hns/hns_roce_ah.c index 24f79ee39fdf..0ac294db3b29 100644 --- a/drivers/infiniband/hw/hns/hns_roce_ah.c +++ b/drivers/infiniband/hw/hns/hns_roce_ah.c @@ -39,7 +39,8 @@ #define HNS_ROCE_VLAN_SL_BIT_MASK 7 #define HNS_ROCE_VLAN_SL_SHIFT 13 -struct ib_ah *hns_roce_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *ah_attr) +struct ib_ah *hns_roce_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *ah_attr, + struct ib_udata *udata) { struct hns_roce_dev *hr_dev = to_hr_dev(ibpd->device); struct device *dev = &hr_dev->pdev->dev; diff --git a/drivers/infiniband/hw/hns/hns_roce_alloc.c b/drivers/infiniband/hw/hns/hns_roce_alloc.c index 863a17a2de40..605962f2828c 100644 --- a/drivers/infiniband/hw/hns/hns_roce_alloc.c +++ b/drivers/infiniband/hw/hns/hns_roce_alloc.c @@ -61,9 +61,10 @@ int hns_roce_bitmap_alloc(struct hns_roce_bitmap *bitmap, unsigned long *obj) return ret; } -void hns_roce_bitmap_free(struct hns_roce_bitmap *bitmap, unsigned long obj) +void hns_roce_bitmap_free(struct hns_roce_bitmap *bitmap, unsigned long obj, + int rr) { - hns_roce_bitmap_free_range(bitmap, obj, 1); + hns_roce_bitmap_free_range(bitmap, obj, 1, rr); } int hns_roce_bitmap_alloc_range(struct hns_roce_bitmap *bitmap, int cnt, @@ -106,7 +107,8 @@ int hns_roce_bitmap_alloc_range(struct hns_roce_bitmap *bitmap, int cnt, } void hns_roce_bitmap_free_range(struct hns_roce_bitmap *bitmap, - unsigned long obj, int cnt) + unsigned long obj, int cnt, + int rr) { int i; @@ -116,7 +118,8 @@ void hns_roce_bitmap_free_range(struct hns_roce_bitmap *bitmap, for (i = 0; i < cnt; i++) clear_bit(obj + i, bitmap->table); - bitmap->last = min(bitmap->last, obj); + if (!rr) + bitmap->last = min(bitmap->last, obj); bitmap->top = (bitmap->top + bitmap->max + bitmap->reserved_top) & bitmap->mask; spin_unlock(&bitmap->lock); diff --git a/drivers/infiniband/hw/hns/hns_roce_cmd.c b/drivers/infiniband/hw/hns/hns_roce_cmd.c index 2a0b6c05da5f..8c1f7a6f84d2 100644 --- a/drivers/infiniband/hw/hns/hns_roce_cmd.c +++ b/drivers/infiniband/hw/hns/hns_roce_cmd.c @@ -216,10 +216,10 @@ static int __hns_roce_cmd_mbox_wait(struct hns_roce_dev *hr_dev, u64 in_param, goto out; /* - * It is timeout when wait_for_completion_timeout return 0 - * The return value is the time limit set in advance - * how many seconds showing - */ + * It is timeout when wait_for_completion_timeout return 0 + * The return value is the time limit set in advance + * how many seconds showing + */ if (!wait_for_completion_timeout(&context->done, msecs_to_jiffies(timeout))) { dev_err(dev, "[cmd]wait_for_completion_timeout timeout\n"); diff --git a/drivers/infiniband/hw/hns/hns_roce_cmd.h b/drivers/infiniband/hw/hns/hns_roce_cmd.h index e3997d312c55..f5a9ee2fc53d 100644 --- a/drivers/infiniband/hw/hns/hns_roce_cmd.h +++ b/drivers/infiniband/hw/hns/hns_roce_cmd.h @@ -34,6 +34,7 @@ #define _HNS_ROCE_CMD_H #define HNS_ROCE_MAILBOX_SIZE 4096 +#define HNS_ROCE_CMD_TIMEOUT_MSECS 10000 enum { /* TPT commands */ @@ -57,17 +58,6 @@ enum { HNS_ROCE_CMD_QUERY_QP = 0x22, }; -enum { - HNS_ROCE_CMD_TIME_CLASS_A = 10000, - HNS_ROCE_CMD_TIME_CLASS_B = 10000, - HNS_ROCE_CMD_TIME_CLASS_C = 10000, -}; - -struct hns_roce_cmd_mailbox { - void *buf; - dma_addr_t dma; -}; - int hns_roce_cmd_mbox(struct hns_roce_dev *hr_dev, u64 in_param, u64 out_param, unsigned long in_modifier, u8 op_modifier, u16 op, unsigned long timeout); diff --git a/drivers/infiniband/hw/hns/hns_roce_common.h b/drivers/infiniband/hw/hns/hns_roce_common.h index 297016103aa7..4af403e1348c 100644 --- a/drivers/infiniband/hw/hns/hns_roce_common.h +++ b/drivers/infiniband/hw/hns/hns_roce_common.h @@ -57,6 +57,32 @@ #define roce_set_bit(origin, shift, val) \ roce_set_field((origin), (1ul << (shift)), (shift), (val)) +/* + * roce_hw_index_cmp_lt - Compare two hardware index values in hisilicon + * SOC, check if a is less than b. + * @a: hardware index value + * @b: hardware index value + * @bits: the number of bits of a and b, range: 0~31. + * + * Hardware index increases continuously till max value, and then restart + * from zero, again and again. Because the bits of reg field is often + * limited, the reg field can only hold the low bits of the hardware index + * in hisilicon SOC. + * In some scenes we need to compare two values(a,b) getted from two reg + * fields in this driver, for example: + * If a equals 0xfffe, b equals 0x1 and bits equals 16, we think b has + * incresed from 0xffff to 0x1 and a is less than b. + * If a equals 0xfffe, b equals 0x0xf001 and bits equals 16, we think a + * is bigger than b. + * + * Return true on a less than b, otherwise false. + */ +#define roce_hw_index_mask(bits) ((1ul << (bits)) - 1) +#define roce_hw_index_shift(bits) (32 - (bits)) +#define roce_hw_index_cmp_lt(a, b, bits) \ + ((int)((((a) - (b)) & roce_hw_index_mask(bits)) << \ + roce_hw_index_shift(bits)) < 0) + #define ROCEE_GLB_CFG_ROCEE_DB_SQ_MODE_S 3 #define ROCEE_GLB_CFG_ROCEE_DB_OTH_MODE_S 4 @@ -245,16 +271,26 @@ #define ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_M \ (((1UL << 28) - 1) << ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_S) +#define ROCEE_SDB_PTR_CMP_BITS 28 + #define ROCEE_SDB_INV_CNT_SDB_INV_CNT_S 0 #define ROCEE_SDB_INV_CNT_SDB_INV_CNT_M \ (((1UL << 16) - 1) << ROCEE_SDB_INV_CNT_SDB_INV_CNT_S) +#define ROCEE_SDB_RETRY_CNT_SDB_RETRY_CT_S 0 +#define ROCEE_SDB_RETRY_CNT_SDB_RETRY_CT_M \ + (((1UL << 16) - 1) << ROCEE_SDB_RETRY_CNT_SDB_RETRY_CT_S) + +#define ROCEE_SDB_CNT_CMP_BITS 16 + +#define ROCEE_TSP_BP_ST_QH_FIFO_ENTRY_S 20 + +#define ROCEE_CNT_CLR_CE_CNT_CLR_CE_S 0 + /*************ROCEE_REG DEFINITION****************/ #define ROCEE_VENDOR_ID_REG 0x0 #define ROCEE_VENDOR_PART_ID_REG 0x4 -#define ROCEE_HW_VERSION_REG 0x8 - #define ROCEE_SYS_IMAGE_GUID_L_REG 0xC #define ROCEE_SYS_IMAGE_GUID_H_REG 0x10 @@ -318,7 +354,11 @@ #define ROCEE_SDB_ISSUE_PTR_REG 0x758 #define ROCEE_SDB_SEND_PTR_REG 0x75C +#define ROCEE_CAEP_CQE_WCMD_EMPTY 0x850 +#define ROCEE_SCAEP_WR_CQE_CNT 0x8D0 #define ROCEE_SDB_INV_CNT_REG 0x9A4 +#define ROCEE_SDB_RETRY_CNT_REG 0x9AC +#define ROCEE_TSP_BP_ST_REG 0x9EC #define ROCEE_ECC_UCERR_ALM0_REG 0xB34 #define ROCEE_ECC_CERR_ALM0_REG 0xB40 diff --git a/drivers/infiniband/hw/hns/hns_roce_cq.c b/drivers/infiniband/hw/hns/hns_roce_cq.c index 097365932b09..589496c8fb9e 100644 --- a/drivers/infiniband/hw/hns/hns_roce_cq.c +++ b/drivers/infiniband/hw/hns/hns_roce_cq.c @@ -35,7 +35,7 @@ #include "hns_roce_device.h" #include "hns_roce_cmd.h" #include "hns_roce_hem.h" -#include "hns_roce_user.h" +#include <rdma/hns-abi.h> #include "hns_roce_common.h" static void hns_roce_ib_cq_comp(struct hns_roce_cq *hr_cq) @@ -77,7 +77,7 @@ static int hns_roce_sw2hw_cq(struct hns_roce_dev *dev, unsigned long cq_num) { return hns_roce_cmd_mbox(dev, mailbox->dma, 0, cq_num, 0, - HNS_ROCE_CMD_SW2HW_CQ, HNS_ROCE_CMD_TIME_CLASS_A); + HNS_ROCE_CMD_SW2HW_CQ, HNS_ROCE_CMD_TIMEOUT_MSECS); } static int hns_roce_cq_alloc(struct hns_roce_dev *hr_dev, int nent, @@ -166,7 +166,7 @@ err_put: hns_roce_table_put(hr_dev, &cq_table->table, hr_cq->cqn); err_out: - hns_roce_bitmap_free(&cq_table->bitmap, hr_cq->cqn); + hns_roce_bitmap_free(&cq_table->bitmap, hr_cq->cqn, BITMAP_NO_RR); return ret; } @@ -176,11 +176,10 @@ static int hns_roce_hw2sw_cq(struct hns_roce_dev *dev, { return hns_roce_cmd_mbox(dev, 0, mailbox ? mailbox->dma : 0, cq_num, mailbox ? 0 : 1, HNS_ROCE_CMD_HW2SW_CQ, - HNS_ROCE_CMD_TIME_CLASS_A); + HNS_ROCE_CMD_TIMEOUT_MSECS); } -static void hns_roce_free_cq(struct hns_roce_dev *hr_dev, - struct hns_roce_cq *hr_cq) +void hns_roce_free_cq(struct hns_roce_dev *hr_dev, struct hns_roce_cq *hr_cq) { struct hns_roce_cq_table *cq_table = &hr_dev->cq_table; struct device *dev = &hr_dev->pdev->dev; @@ -204,7 +203,7 @@ static void hns_roce_free_cq(struct hns_roce_dev *hr_dev, spin_unlock_irq(&cq_table->lock); hns_roce_table_put(hr_dev, &cq_table->table, hr_cq->cqn); - hns_roce_bitmap_free(&cq_table->bitmap, hr_cq->cqn); + hns_roce_bitmap_free(&cq_table->bitmap, hr_cq->cqn, BITMAP_NO_RR); } static int hns_roce_ib_get_cq_umem(struct hns_roce_dev *hr_dev, @@ -349,6 +348,15 @@ struct ib_cq *hns_roce_ib_create_cq(struct ib_device *ib_dev, goto err_mtt; } + /* + * For the QP created by kernel space, tptr value should be initialized + * to zero; For the QP created by user space, it will cause synchronous + * problems if tptr is set to zero here, so we initialze it in user + * space. + */ + if (!context) + *hr_cq->tptr_addr = 0; + /* Get created cq handler and carry out event */ hr_cq->comp = hns_roce_ib_cq_comp; hr_cq->event = hns_roce_ib_cq_event; @@ -383,19 +391,25 @@ int hns_roce_ib_destroy_cq(struct ib_cq *ib_cq) { struct hns_roce_dev *hr_dev = to_hr_dev(ib_cq->device); struct hns_roce_cq *hr_cq = to_hr_cq(ib_cq); + int ret = 0; - hns_roce_free_cq(hr_dev, hr_cq); - hns_roce_mtt_cleanup(hr_dev, &hr_cq->hr_buf.hr_mtt); + if (hr_dev->hw->destroy_cq) { + ret = hr_dev->hw->destroy_cq(ib_cq); + } else { + hns_roce_free_cq(hr_dev, hr_cq); + hns_roce_mtt_cleanup(hr_dev, &hr_cq->hr_buf.hr_mtt); - if (ib_cq->uobject) - ib_umem_release(hr_cq->umem); - else - /* Free the buff of stored cq */ - hns_roce_ib_free_cq_buf(hr_dev, &hr_cq->hr_buf, ib_cq->cqe); + if (ib_cq->uobject) + ib_umem_release(hr_cq->umem); + else + /* Free the buff of stored cq */ + hns_roce_ib_free_cq_buf(hr_dev, &hr_cq->hr_buf, + ib_cq->cqe); - kfree(hr_cq); + kfree(hr_cq); + } - return 0; + return ret; } void hns_roce_cq_completion(struct hns_roce_dev *hr_dev, u32 cqn) diff --git a/drivers/infiniband/hw/hns/hns_roce_device.h b/drivers/infiniband/hw/hns/hns_roce_device.h index 341731553a60..1a6cb5d7a0dd 100644 --- a/drivers/infiniband/hw/hns/hns_roce_device.h +++ b/drivers/infiniband/hw/hns/hns_roce_device.h @@ -37,6 +37,8 @@ #define DRV_NAME "hns_roce" +#define HNS_ROCE_HW_VER1 ('h' << 24 | 'i' << 16 | '0' << 8 | '6') + #define MAC_ADDR_OCTET_NUM 6 #define HNS_ROCE_MAX_MSG_LEN 0x80000000 @@ -54,6 +56,12 @@ #define HNS_ROCE_MAX_INNER_MTPT_NUM 0x7 #define HNS_ROCE_MAX_MTPT_PBL_NUM 0x100000 +#define HNS_ROCE_EACH_FREE_CQ_WAIT_MSECS 20 +#define HNS_ROCE_MAX_FREE_CQ_WAIT_CNT \ + (5000 / HNS_ROCE_EACH_FREE_CQ_WAIT_MSECS) +#define HNS_ROCE_CQE_WCMD_EMPTY_BIT 0x2 +#define HNS_ROCE_MIN_CQE_CNT 16 + #define HNS_ROCE_MAX_IRQ_NUM 34 #define HNS_ROCE_COMP_VEC_NUM 32 @@ -70,6 +78,9 @@ #define HNS_ROCE_MAX_GID_NUM 16 #define HNS_ROCE_GID_SIZE 16 +#define BITMAP_NO_RR 0 +#define BITMAP_RR 1 + #define MR_TYPE_MR 0x00 #define MR_TYPE_DMA 0x03 @@ -196,9 +207,9 @@ struct hns_roce_bitmap { /* Order = 0: bitmap is biggest, order = max bitmap is least (only a bit) */ /* Every bit repesent to a partner free/used status in bitmap */ /* -* Initial, bits of other bitmap are all 0 except that a bit of max_order is 1 -* Bit = 1 represent to idle and available; bit = 0: not available -*/ + * Initial, bits of other bitmap are all 0 except that a bit of max_order is 1 + * Bit = 1 represent to idle and available; bit = 0: not available + */ struct hns_roce_buddy { /* Members point to every order level bitmap */ unsigned long **bits; @@ -296,7 +307,7 @@ struct hns_roce_cq { u32 cq_depth; u32 cons_index; void __iomem *cq_db_l; - void __iomem *tptr_addr; + u16 *tptr_addr; unsigned long cqn; u32 vector; atomic_t refcount; @@ -360,29 +371,34 @@ struct hns_roce_cmdq { struct mutex hcr_mutex; struct semaphore poll_sem; /* - * Event mode: cmd register mutex protection, - * ensure to not exceed max_cmds and user use limit region - */ + * Event mode: cmd register mutex protection, + * ensure to not exceed max_cmds and user use limit region + */ struct semaphore event_sem; int max_cmds; spinlock_t context_lock; int free_head; struct hns_roce_cmd_context *context; /* - * Result of get integer part - * which max_comds compute according a power of 2 - */ + * Result of get integer part + * which max_comds compute according a power of 2 + */ u16 token_mask; /* - * Process whether use event mode, init default non-zero - * After the event queue of cmd event ready, - * can switch into event mode - * close device, switch into poll mode(non event mode) - */ + * Process whether use event mode, init default non-zero + * After the event queue of cmd event ready, + * can switch into event mode + * close device, switch into poll mode(non event mode) + */ u8 use_events; u8 toggle; }; +struct hns_roce_cmd_mailbox { + void *buf; + dma_addr_t dma; +}; + struct hns_roce_dev; struct hns_roce_qp { @@ -424,8 +440,6 @@ struct hns_roce_ib_iboe { struct net_device *netdevs[HNS_ROCE_MAX_PORTS]; struct notifier_block nb; struct notifier_block nb_inet; - /* 16 GID is shared by 6 port in v1 engine. */ - union ib_gid gid_table[HNS_ROCE_MAX_GID_NUM]; u8 phy_port[HNS_ROCE_MAX_PORTS]; }; @@ -519,6 +533,8 @@ struct hns_roce_hw { struct ib_recv_wr **bad_recv_wr); int (*req_notify_cq)(struct ib_cq *ibcq, enum ib_cq_notify_flags flags); int (*poll_cq)(struct ib_cq *ibcq, int num_entries, struct ib_wc *wc); + int (*dereg_mr)(struct hns_roce_dev *hr_dev, struct hns_roce_mr *mr); + int (*destroy_cq)(struct ib_cq *ibcq); void *priv; }; @@ -553,6 +569,8 @@ struct hns_roce_dev { int cmd_mod; int loop_idc; + dma_addr_t tptr_dma_addr; /*only for hw v1*/ + u32 tptr_size; /*only for hw v1*/ struct hns_roce_hw *hw; }; @@ -657,7 +675,8 @@ void hns_roce_cleanup_cq_table(struct hns_roce_dev *hr_dev); void hns_roce_cleanup_qp_table(struct hns_roce_dev *hr_dev); int hns_roce_bitmap_alloc(struct hns_roce_bitmap *bitmap, unsigned long *obj); -void hns_roce_bitmap_free(struct hns_roce_bitmap *bitmap, unsigned long obj); +void hns_roce_bitmap_free(struct hns_roce_bitmap *bitmap, unsigned long obj, + int rr); int hns_roce_bitmap_init(struct hns_roce_bitmap *bitmap, u32 num, u32 mask, u32 reserved_bot, u32 resetrved_top); void hns_roce_bitmap_cleanup(struct hns_roce_bitmap *bitmap); @@ -665,9 +684,11 @@ void hns_roce_cleanup_bitmap(struct hns_roce_dev *hr_dev); int hns_roce_bitmap_alloc_range(struct hns_roce_bitmap *bitmap, int cnt, int align, unsigned long *obj); void hns_roce_bitmap_free_range(struct hns_roce_bitmap *bitmap, - unsigned long obj, int cnt); + unsigned long obj, int cnt, + int rr); -struct ib_ah *hns_roce_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr); +struct ib_ah *hns_roce_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, + struct ib_udata *udata); int hns_roce_query_ah(struct ib_ah *ibah, struct ib_ah_attr *ah_attr); int hns_roce_destroy_ah(struct ib_ah *ah); @@ -681,6 +702,10 @@ struct ib_mr *hns_roce_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, u64 virt_addr, int access_flags, struct ib_udata *udata); int hns_roce_dereg_mr(struct ib_mr *ibmr); +int hns_roce_hw2sw_mpt(struct hns_roce_dev *hr_dev, + struct hns_roce_cmd_mailbox *mailbox, + unsigned long mpt_index); +unsigned long key_to_hw_index(u32 key); void hns_roce_buf_free(struct hns_roce_dev *hr_dev, u32 size, struct hns_roce_buf *buf); @@ -717,6 +742,7 @@ struct ib_cq *hns_roce_ib_create_cq(struct ib_device *ib_dev, struct ib_udata *udata); int hns_roce_ib_destroy_cq(struct ib_cq *ib_cq); +void hns_roce_free_cq(struct hns_roce_dev *hr_dev, struct hns_roce_cq *hr_cq); void hns_roce_cq_completion(struct hns_roce_dev *hr_dev, u32 cqn); void hns_roce_cq_event(struct hns_roce_dev *hr_dev, u32 cqn, int event_type); diff --git a/drivers/infiniband/hw/hns/hns_roce_eq.c b/drivers/infiniband/hw/hns/hns_roce_eq.c index 21e21b03cfb5..50f864935a0e 100644 --- a/drivers/infiniband/hw/hns/hns_roce_eq.c +++ b/drivers/infiniband/hw/hns/hns_roce_eq.c @@ -371,9 +371,9 @@ static int hns_roce_aeq_ovf_int(struct hns_roce_dev *hr_dev, int i = 0; /** - * AEQ overflow ECC mult bit err CEQ overflow alarm - * must clear interrupt, mask irq, clear irq, cancel mask operation - */ + * AEQ overflow ECC mult bit err CEQ overflow alarm + * must clear interrupt, mask irq, clear irq, cancel mask operation + */ aeshift_val = roce_read(hr_dev, ROCEE_CAEP_AEQC_AEQE_SHIFT_REG); if (roce_get_bit(aeshift_val, diff --git a/drivers/infiniband/hw/hns/hns_roce_hem.c b/drivers/infiniband/hw/hns/hns_roce_hem.c index 250d8f280390..c5104e0b2916 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hem.c +++ b/drivers/infiniband/hw/hns/hns_roce_hem.c @@ -80,9 +80,9 @@ struct hns_roce_hem *hns_roce_alloc_hem(struct hns_roce_dev *hr_dev, int npages, --order; /* - * Alloc memory one time. If failed, don't alloc small block - * memory, directly return fail. - */ + * Alloc memory one time. If failed, don't alloc small block + * memory, directly return fail. + */ mem = &chunk->mem[chunk->npages]; buf = dma_alloc_coherent(&hr_dev->pdev->dev, PAGE_SIZE << order, &sg_dma_address(mem), gfp_mask); diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c index 71232e5fabf6..b8111b0c8877 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c @@ -32,6 +32,7 @@ #include <linux/platform_device.h> #include <linux/acpi.h> +#include <linux/etherdevice.h> #include <rdma/ib_umem.h> #include "hns_roce_common.h" #include "hns_roce_device.h" @@ -72,6 +73,8 @@ int hns_roce_v1_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, int nreq = 0; u32 ind = 0; int ret = 0; + u8 *smac; + int loopback; if (unlikely(ibqp->qp_type != IB_QPT_GSI && ibqp->qp_type != IB_QPT_RC)) { @@ -129,6 +132,14 @@ int hns_roce_v1_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, UD_SEND_WQE_U32_8_DMAC_5_M, UD_SEND_WQE_U32_8_DMAC_5_S, ah->av.mac[5]); + + smac = (u8 *)hr_dev->dev_addr[qp->port]; + loopback = ether_addr_equal_unaligned(ah->av.mac, + smac) ? 1 : 0; + roce_set_bit(ud_sq_wqe->u32_8, + UD_SEND_WQE_U32_8_LOOPBACK_INDICATOR_S, + loopback); + roce_set_field(ud_sq_wqe->u32_8, UD_SEND_WQE_U32_8_OPERATION_TYPE_M, UD_SEND_WQE_U32_8_OPERATION_TYPE_S, @@ -284,6 +295,8 @@ out: roce_set_field(sq_db.u32_4, SQ_DOORBELL_U32_4_SQ_HEAD_M, SQ_DOORBELL_U32_4_SQ_HEAD_S, (qp->sq.head & ((qp->sq.wqe_cnt << 1) - 1))); + roce_set_field(sq_db.u32_4, SQ_DOORBELL_U32_4_SL_M, + SQ_DOORBELL_U32_4_SL_S, qp->sl); roce_set_field(sq_db.u32_4, SQ_DOORBELL_U32_4_PORT_M, SQ_DOORBELL_U32_4_PORT_S, qp->phy_port); roce_set_field(sq_db.u32_8, SQ_DOORBELL_U32_8_QPN_M, @@ -611,6 +624,213 @@ ext_sdb_buf_fail_out: return ret; } +static struct hns_roce_qp *hns_roce_v1_create_lp_qp(struct hns_roce_dev *hr_dev, + struct ib_pd *pd) +{ + struct device *dev = &hr_dev->pdev->dev; + struct ib_qp_init_attr init_attr; + struct ib_qp *qp; + + memset(&init_attr, 0, sizeof(struct ib_qp_init_attr)); + init_attr.qp_type = IB_QPT_RC; + init_attr.sq_sig_type = IB_SIGNAL_ALL_WR; + init_attr.cap.max_recv_wr = HNS_ROCE_MIN_WQE_NUM; + init_attr.cap.max_send_wr = HNS_ROCE_MIN_WQE_NUM; + + qp = hns_roce_create_qp(pd, &init_attr, NULL); + if (IS_ERR(qp)) { + dev_err(dev, "Create loop qp for mr free failed!"); + return NULL; + } + + return to_hr_qp(qp); +} + +static int hns_roce_v1_rsv_lp_qp(struct hns_roce_dev *hr_dev) +{ + struct hns_roce_caps *caps = &hr_dev->caps; + struct device *dev = &hr_dev->pdev->dev; + struct ib_cq_init_attr cq_init_attr; + struct hns_roce_free_mr *free_mr; + struct ib_qp_attr attr = { 0 }; + struct hns_roce_v1_priv *priv; + struct hns_roce_qp *hr_qp; + struct ib_cq *cq; + struct ib_pd *pd; + u64 subnet_prefix; + int attr_mask = 0; + int i; + int ret; + u8 phy_port; + u8 sl; + + priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; + free_mr = &priv->free_mr; + + /* Reserved cq for loop qp */ + cq_init_attr.cqe = HNS_ROCE_MIN_WQE_NUM * 2; + cq_init_attr.comp_vector = 0; + cq = hns_roce_ib_create_cq(&hr_dev->ib_dev, &cq_init_attr, NULL, NULL); + if (IS_ERR(cq)) { + dev_err(dev, "Create cq for reseved loop qp failed!"); + return -ENOMEM; + } + free_mr->mr_free_cq = to_hr_cq(cq); + free_mr->mr_free_cq->ib_cq.device = &hr_dev->ib_dev; + free_mr->mr_free_cq->ib_cq.uobject = NULL; + free_mr->mr_free_cq->ib_cq.comp_handler = NULL; + free_mr->mr_free_cq->ib_cq.event_handler = NULL; + free_mr->mr_free_cq->ib_cq.cq_context = NULL; + atomic_set(&free_mr->mr_free_cq->ib_cq.usecnt, 0); + + pd = hns_roce_alloc_pd(&hr_dev->ib_dev, NULL, NULL); + if (IS_ERR(pd)) { + dev_err(dev, "Create pd for reseved loop qp failed!"); + ret = -ENOMEM; + goto alloc_pd_failed; + } + free_mr->mr_free_pd = to_hr_pd(pd); + free_mr->mr_free_pd->ibpd.device = &hr_dev->ib_dev; + free_mr->mr_free_pd->ibpd.uobject = NULL; + atomic_set(&free_mr->mr_free_pd->ibpd.usecnt, 0); + + attr.qp_access_flags = IB_ACCESS_REMOTE_WRITE; + attr.pkey_index = 0; + attr.min_rnr_timer = 0; + /* Disable read ability */ + attr.max_dest_rd_atomic = 0; + attr.max_rd_atomic = 0; + /* Use arbitrary values as rq_psn and sq_psn */ + attr.rq_psn = 0x0808; + attr.sq_psn = 0x0808; + attr.retry_cnt = 7; + attr.rnr_retry = 7; + attr.timeout = 0x12; + attr.path_mtu = IB_MTU_256; + attr.ah_attr.ah_flags = 1; + attr.ah_attr.static_rate = 3; + attr.ah_attr.grh.sgid_index = 0; + attr.ah_attr.grh.hop_limit = 1; + attr.ah_attr.grh.flow_label = 0; + attr.ah_attr.grh.traffic_class = 0; + + subnet_prefix = cpu_to_be64(0xfe80000000000000LL); + for (i = 0; i < HNS_ROCE_V1_RESV_QP; i++) { + free_mr->mr_free_qp[i] = hns_roce_v1_create_lp_qp(hr_dev, pd); + if (IS_ERR(free_mr->mr_free_qp[i])) { + dev_err(dev, "Create loop qp failed!\n"); + goto create_lp_qp_failed; + } + hr_qp = free_mr->mr_free_qp[i]; + + sl = i / caps->num_ports; + + if (caps->num_ports == HNS_ROCE_MAX_PORTS) + phy_port = (i >= HNS_ROCE_MAX_PORTS) ? (i - 2) : + (i % caps->num_ports); + else + phy_port = i % caps->num_ports; + + hr_qp->port = phy_port + 1; + hr_qp->phy_port = phy_port; + hr_qp->ibqp.qp_type = IB_QPT_RC; + hr_qp->ibqp.device = &hr_dev->ib_dev; + hr_qp->ibqp.uobject = NULL; + atomic_set(&hr_qp->ibqp.usecnt, 0); + hr_qp->ibqp.pd = pd; + hr_qp->ibqp.recv_cq = cq; + hr_qp->ibqp.send_cq = cq; + + attr.ah_attr.port_num = phy_port + 1; + attr.ah_attr.sl = sl; + attr.port_num = phy_port + 1; + + attr.dest_qp_num = hr_qp->qpn; + memcpy(attr.ah_attr.dmac, hr_dev->dev_addr[phy_port], + MAC_ADDR_OCTET_NUM); + + memcpy(attr.ah_attr.grh.dgid.raw, + &subnet_prefix, sizeof(u64)); + memcpy(&attr.ah_attr.grh.dgid.raw[8], + hr_dev->dev_addr[phy_port], 3); + memcpy(&attr.ah_attr.grh.dgid.raw[13], + hr_dev->dev_addr[phy_port] + 3, 3); + attr.ah_attr.grh.dgid.raw[11] = 0xff; + attr.ah_attr.grh.dgid.raw[12] = 0xfe; + attr.ah_attr.grh.dgid.raw[8] ^= 2; + + attr_mask |= IB_QP_PORT; + + ret = hr_dev->hw->modify_qp(&hr_qp->ibqp, &attr, attr_mask, + IB_QPS_RESET, IB_QPS_INIT); + if (ret) { + dev_err(dev, "modify qp failed(%d)!\n", ret); + goto create_lp_qp_failed; + } + + ret = hr_dev->hw->modify_qp(&hr_qp->ibqp, &attr, attr_mask, + IB_QPS_INIT, IB_QPS_RTR); + if (ret) { + dev_err(dev, "modify qp failed(%d)!\n", ret); + goto create_lp_qp_failed; + } + + ret = hr_dev->hw->modify_qp(&hr_qp->ibqp, &attr, attr_mask, + IB_QPS_RTR, IB_QPS_RTS); + if (ret) { + dev_err(dev, "modify qp failed(%d)!\n", ret); + goto create_lp_qp_failed; + } + } + + return 0; + +create_lp_qp_failed: + for (i -= 1; i >= 0; i--) { + hr_qp = free_mr->mr_free_qp[i]; + if (hns_roce_v1_destroy_qp(&hr_qp->ibqp)) + dev_err(dev, "Destroy qp %d for mr free failed!\n", i); + } + + if (hns_roce_dealloc_pd(pd)) + dev_err(dev, "Destroy pd for create_lp_qp failed!\n"); + +alloc_pd_failed: + if (hns_roce_ib_destroy_cq(cq)) + dev_err(dev, "Destroy cq for create_lp_qp failed!\n"); + + return -EINVAL; +} + +static void hns_roce_v1_release_lp_qp(struct hns_roce_dev *hr_dev) +{ + struct device *dev = &hr_dev->pdev->dev; + struct hns_roce_free_mr *free_mr; + struct hns_roce_v1_priv *priv; + struct hns_roce_qp *hr_qp; + int ret; + int i; + + priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; + free_mr = &priv->free_mr; + + for (i = 0; i < HNS_ROCE_V1_RESV_QP; i++) { + hr_qp = free_mr->mr_free_qp[i]; + ret = hns_roce_v1_destroy_qp(&hr_qp->ibqp); + if (ret) + dev_err(dev, "Destroy qp %d for mr free failed(%d)!\n", + i, ret); + } + + ret = hns_roce_ib_destroy_cq(&free_mr->mr_free_cq->ib_cq); + if (ret) + dev_err(dev, "Destroy cq for mr_free failed(%d)!\n", ret); + + ret = hns_roce_dealloc_pd(&free_mr->mr_free_pd->ibpd); + if (ret) + dev_err(dev, "Destroy pd for mr_free failed(%d)!\n", ret); +} + static int hns_roce_db_init(struct hns_roce_dev *hr_dev) { struct device *dev = &hr_dev->pdev->dev; @@ -648,6 +868,223 @@ static int hns_roce_db_init(struct hns_roce_dev *hr_dev) return 0; } +void hns_roce_v1_recreate_lp_qp_work_fn(struct work_struct *work) +{ + struct hns_roce_recreate_lp_qp_work *lp_qp_work; + struct hns_roce_dev *hr_dev; + + lp_qp_work = container_of(work, struct hns_roce_recreate_lp_qp_work, + work); + hr_dev = to_hr_dev(lp_qp_work->ib_dev); + + hns_roce_v1_release_lp_qp(hr_dev); + + if (hns_roce_v1_rsv_lp_qp(hr_dev)) + dev_err(&hr_dev->pdev->dev, "create reserver qp failed\n"); + + if (lp_qp_work->comp_flag) + complete(lp_qp_work->comp); + + kfree(lp_qp_work); +} + +static int hns_roce_v1_recreate_lp_qp(struct hns_roce_dev *hr_dev) +{ + struct device *dev = &hr_dev->pdev->dev; + struct hns_roce_recreate_lp_qp_work *lp_qp_work; + struct hns_roce_free_mr *free_mr; + struct hns_roce_v1_priv *priv; + struct completion comp; + unsigned long end = + msecs_to_jiffies(HNS_ROCE_V1_RECREATE_LP_QP_TIMEOUT_MSECS) + jiffies; + + priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; + free_mr = &priv->free_mr; + + lp_qp_work = kzalloc(sizeof(struct hns_roce_recreate_lp_qp_work), + GFP_KERNEL); + + INIT_WORK(&(lp_qp_work->work), hns_roce_v1_recreate_lp_qp_work_fn); + + lp_qp_work->ib_dev = &(hr_dev->ib_dev); + lp_qp_work->comp = ∁ + lp_qp_work->comp_flag = 1; + + init_completion(lp_qp_work->comp); + + queue_work(free_mr->free_mr_wq, &(lp_qp_work->work)); + + while (time_before_eq(jiffies, end)) { + if (try_wait_for_completion(&comp)) + return 0; + msleep(HNS_ROCE_V1_RECREATE_LP_QP_WAIT_VALUE); + } + + lp_qp_work->comp_flag = 0; + if (try_wait_for_completion(&comp)) + return 0; + + dev_warn(dev, "recreate lp qp failed 20s timeout and return failed!\n"); + return -ETIMEDOUT; +} + +static int hns_roce_v1_send_lp_wqe(struct hns_roce_qp *hr_qp) +{ + struct hns_roce_dev *hr_dev = to_hr_dev(hr_qp->ibqp.device); + struct device *dev = &hr_dev->pdev->dev; + struct ib_send_wr send_wr, *bad_wr; + int ret; + + memset(&send_wr, 0, sizeof(send_wr)); + send_wr.next = NULL; + send_wr.num_sge = 0; + send_wr.send_flags = 0; + send_wr.sg_list = NULL; + send_wr.wr_id = (unsigned long long)&send_wr; + send_wr.opcode = IB_WR_RDMA_WRITE; + + ret = hns_roce_v1_post_send(&hr_qp->ibqp, &send_wr, &bad_wr); + if (ret) { + dev_err(dev, "Post write wqe for mr free failed(%d)!", ret); + return ret; + } + + return 0; +} + +static void hns_roce_v1_mr_free_work_fn(struct work_struct *work) +{ + struct hns_roce_mr_free_work *mr_work; + struct ib_wc wc[HNS_ROCE_V1_RESV_QP]; + struct hns_roce_free_mr *free_mr; + struct hns_roce_cq *mr_free_cq; + struct hns_roce_v1_priv *priv; + struct hns_roce_dev *hr_dev; + struct hns_roce_mr *hr_mr; + struct hns_roce_qp *hr_qp; + struct device *dev; + unsigned long end = + msecs_to_jiffies(HNS_ROCE_V1_FREE_MR_TIMEOUT_MSECS) + jiffies; + int i; + int ret; + int ne; + + mr_work = container_of(work, struct hns_roce_mr_free_work, work); + hr_mr = (struct hns_roce_mr *)mr_work->mr; + hr_dev = to_hr_dev(mr_work->ib_dev); + dev = &hr_dev->pdev->dev; + + priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; + free_mr = &priv->free_mr; + mr_free_cq = free_mr->mr_free_cq; + + for (i = 0; i < HNS_ROCE_V1_RESV_QP; i++) { + hr_qp = free_mr->mr_free_qp[i]; + ret = hns_roce_v1_send_lp_wqe(hr_qp); + if (ret) { + dev_err(dev, + "Send wqe (qp:0x%lx) for mr free failed(%d)!\n", + hr_qp->qpn, ret); + goto free_work; + } + } + + ne = HNS_ROCE_V1_RESV_QP; + do { + ret = hns_roce_v1_poll_cq(&mr_free_cq->ib_cq, ne, wc); + if (ret < 0) { + dev_err(dev, + "(qp:0x%lx) starts, Poll cqe failed(%d) for mr 0x%x free! Remain %d cqe\n", + hr_qp->qpn, ret, hr_mr->key, ne); + goto free_work; + } + ne -= ret; + msleep(HNS_ROCE_V1_FREE_MR_WAIT_VALUE); + } while (ne && time_before_eq(jiffies, end)); + + if (ne != 0) + dev_err(dev, + "Poll cqe for mr 0x%x free timeout! Remain %d cqe\n", + hr_mr->key, ne); + +free_work: + if (mr_work->comp_flag) + complete(mr_work->comp); + kfree(mr_work); +} + +int hns_roce_v1_dereg_mr(struct hns_roce_dev *hr_dev, struct hns_roce_mr *mr) +{ + struct device *dev = &hr_dev->pdev->dev; + struct hns_roce_mr_free_work *mr_work; + struct hns_roce_free_mr *free_mr; + struct hns_roce_v1_priv *priv; + struct completion comp; + unsigned long end = + msecs_to_jiffies(HNS_ROCE_V1_FREE_MR_TIMEOUT_MSECS) + jiffies; + unsigned long start = jiffies; + int npages; + int ret = 0; + + priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; + free_mr = &priv->free_mr; + + if (mr->enabled) { + if (hns_roce_hw2sw_mpt(hr_dev, NULL, key_to_hw_index(mr->key) + & (hr_dev->caps.num_mtpts - 1))) + dev_warn(dev, "HW2SW_MPT failed!\n"); + } + + mr_work = kzalloc(sizeof(*mr_work), GFP_KERNEL); + if (!mr_work) { + ret = -ENOMEM; + goto free_mr; + } + + INIT_WORK(&(mr_work->work), hns_roce_v1_mr_free_work_fn); + + mr_work->ib_dev = &(hr_dev->ib_dev); + mr_work->comp = ∁ + mr_work->comp_flag = 1; + mr_work->mr = (void *)mr; + init_completion(mr_work->comp); + + queue_work(free_mr->free_mr_wq, &(mr_work->work)); + + while (time_before_eq(jiffies, end)) { + if (try_wait_for_completion(&comp)) + goto free_mr; + msleep(HNS_ROCE_V1_FREE_MR_WAIT_VALUE); + } + + mr_work->comp_flag = 0; + if (try_wait_for_completion(&comp)) + goto free_mr; + + dev_warn(dev, "Free mr work 0x%x over 50s and failed!\n", mr->key); + ret = -ETIMEDOUT; + +free_mr: + dev_dbg(dev, "Free mr 0x%x use 0x%x us.\n", + mr->key, jiffies_to_usecs(jiffies) - jiffies_to_usecs(start)); + + if (mr->size != ~0ULL) { + npages = ib_umem_page_count(mr->umem); + dma_free_coherent(dev, npages * 8, mr->pbl_buf, + mr->pbl_dma_addr); + } + + hns_roce_bitmap_free(&hr_dev->mr_table.mtpt_bitmap, + key_to_hw_index(mr->key), 0); + + if (mr->umem) + ib_umem_release(mr->umem); + + kfree(mr); + + return ret; +} + static void hns_roce_db_free(struct hns_roce_dev *hr_dev) { struct device *dev = &hr_dev->pdev->dev; @@ -849,6 +1286,85 @@ static void hns_roce_bt_free(struct hns_roce_dev *hr_dev) priv->bt_table.qpc_buf.buf, priv->bt_table.qpc_buf.map); } +static int hns_roce_tptr_init(struct hns_roce_dev *hr_dev) +{ + struct device *dev = &hr_dev->pdev->dev; + struct hns_roce_buf_list *tptr_buf; + struct hns_roce_v1_priv *priv; + + priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; + tptr_buf = &priv->tptr_table.tptr_buf; + + /* + * This buffer will be used for CQ's tptr(tail pointer), also + * named ci(customer index). Every CQ will use 2 bytes to save + * cqe ci in hip06. Hardware will read this area to get new ci + * when the queue is almost full. + */ + tptr_buf->buf = dma_alloc_coherent(dev, HNS_ROCE_V1_TPTR_BUF_SIZE, + &tptr_buf->map, GFP_KERNEL); + if (!tptr_buf->buf) + return -ENOMEM; + + hr_dev->tptr_dma_addr = tptr_buf->map; + hr_dev->tptr_size = HNS_ROCE_V1_TPTR_BUF_SIZE; + + return 0; +} + +static void hns_roce_tptr_free(struct hns_roce_dev *hr_dev) +{ + struct device *dev = &hr_dev->pdev->dev; + struct hns_roce_buf_list *tptr_buf; + struct hns_roce_v1_priv *priv; + + priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; + tptr_buf = &priv->tptr_table.tptr_buf; + + dma_free_coherent(dev, HNS_ROCE_V1_TPTR_BUF_SIZE, + tptr_buf->buf, tptr_buf->map); +} + +static int hns_roce_free_mr_init(struct hns_roce_dev *hr_dev) +{ + struct device *dev = &hr_dev->pdev->dev; + struct hns_roce_free_mr *free_mr; + struct hns_roce_v1_priv *priv; + int ret = 0; + + priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; + free_mr = &priv->free_mr; + + free_mr->free_mr_wq = create_singlethread_workqueue("hns_roce_free_mr"); + if (!free_mr->free_mr_wq) { + dev_err(dev, "Create free mr workqueue failed!\n"); + return -ENOMEM; + } + + ret = hns_roce_v1_rsv_lp_qp(hr_dev); + if (ret) { + dev_err(dev, "Reserved loop qp failed(%d)!\n", ret); + flush_workqueue(free_mr->free_mr_wq); + destroy_workqueue(free_mr->free_mr_wq); + } + + return ret; +} + +static void hns_roce_free_mr_free(struct hns_roce_dev *hr_dev) +{ + struct hns_roce_free_mr *free_mr; + struct hns_roce_v1_priv *priv; + + priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; + free_mr = &priv->free_mr; + + flush_workqueue(free_mr->free_mr_wq); + destroy_workqueue(free_mr->free_mr_wq); + + hns_roce_v1_release_lp_qp(hr_dev); +} + /** * hns_roce_v1_reset - reset RoCE * @hr_dev: RoCE device struct pointer @@ -898,6 +1414,38 @@ int hns_roce_v1_reset(struct hns_roce_dev *hr_dev, bool dereset) return ret; } +static int hns_roce_des_qp_init(struct hns_roce_dev *hr_dev) +{ + struct device *dev = &hr_dev->pdev->dev; + struct hns_roce_v1_priv *priv; + struct hns_roce_des_qp *des_qp; + + priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; + des_qp = &priv->des_qp; + + des_qp->requeue_flag = 1; + des_qp->qp_wq = create_singlethread_workqueue("hns_roce_destroy_qp"); + if (!des_qp->qp_wq) { + dev_err(dev, "Create destroy qp workqueue failed!\n"); + return -ENOMEM; + } + + return 0; +} + +static void hns_roce_des_qp_free(struct hns_roce_dev *hr_dev) +{ + struct hns_roce_v1_priv *priv; + struct hns_roce_des_qp *des_qp; + + priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; + des_qp = &priv->des_qp; + + des_qp->requeue_flag = 0; + flush_workqueue(des_qp->qp_wq); + destroy_workqueue(des_qp->qp_wq); +} + void hns_roce_v1_profile(struct hns_roce_dev *hr_dev) { int i = 0; @@ -906,12 +1454,11 @@ void hns_roce_v1_profile(struct hns_roce_dev *hr_dev) hr_dev->vendor_id = le32_to_cpu(roce_read(hr_dev, ROCEE_VENDOR_ID_REG)); hr_dev->vendor_part_id = le32_to_cpu(roce_read(hr_dev, ROCEE_VENDOR_PART_ID_REG)); - hr_dev->hw_rev = le32_to_cpu(roce_read(hr_dev, ROCEE_HW_VERSION_REG)); - hr_dev->sys_image_guid = le32_to_cpu(roce_read(hr_dev, ROCEE_SYS_IMAGE_GUID_L_REG)) | ((u64)le32_to_cpu(roce_read(hr_dev, ROCEE_SYS_IMAGE_GUID_H_REG)) << 32); + hr_dev->hw_rev = HNS_ROCE_HW_VER1; caps->num_qps = HNS_ROCE_V1_MAX_QP_NUM; caps->max_wqes = HNS_ROCE_V1_MAX_WQE_NUM; @@ -1001,18 +1548,44 @@ int hns_roce_v1_init(struct hns_roce_dev *hr_dev) goto error_failed_raq_init; } - hns_roce_port_enable(hr_dev, HNS_ROCE_PORT_UP); - ret = hns_roce_bt_init(hr_dev); if (ret) { dev_err(dev, "bt init failed!\n"); goto error_failed_bt_init; } + ret = hns_roce_tptr_init(hr_dev); + if (ret) { + dev_err(dev, "tptr init failed!\n"); + goto error_failed_tptr_init; + } + + ret = hns_roce_des_qp_init(hr_dev); + if (ret) { + dev_err(dev, "des qp init failed!\n"); + goto error_failed_des_qp_init; + } + + ret = hns_roce_free_mr_init(hr_dev); + if (ret) { + dev_err(dev, "free mr init failed!\n"); + goto error_failed_free_mr_init; + } + + hns_roce_port_enable(hr_dev, HNS_ROCE_PORT_UP); + return 0; +error_failed_free_mr_init: + hns_roce_des_qp_free(hr_dev); + +error_failed_des_qp_init: + hns_roce_tptr_free(hr_dev); + +error_failed_tptr_init: + hns_roce_bt_free(hr_dev); + error_failed_bt_init: - hns_roce_port_enable(hr_dev, HNS_ROCE_PORT_DOWN); hns_roce_raq_free(hr_dev); error_failed_raq_init: @@ -1022,8 +1595,11 @@ error_failed_raq_init: void hns_roce_v1_exit(struct hns_roce_dev *hr_dev) { - hns_roce_bt_free(hr_dev); hns_roce_port_enable(hr_dev, HNS_ROCE_PORT_DOWN); + hns_roce_free_mr_free(hr_dev); + hns_roce_des_qp_free(hr_dev); + hns_roce_tptr_free(hr_dev); + hns_roce_bt_free(hr_dev); hns_roce_raq_free(hr_dev); hns_roce_db_free(hr_dev); } @@ -1061,6 +1637,14 @@ void hns_roce_v1_set_mac(struct hns_roce_dev *hr_dev, u8 phy_port, u8 *addr) u32 *p; u32 val; + /* + * When mac changed, loopback may fail + * because of smac not equal to dmac. + * We Need to release and create reserved qp again. + */ + if (hr_dev->hw->dereg_mr && hns_roce_v1_recreate_lp_qp(hr_dev)) + dev_warn(&hr_dev->pdev->dev, "recreate lp qp timeout!\n"); + p = (u32 *)(&addr[0]); reg_smac_l = *p; roce_raw_write(reg_smac_l, hr_dev->reg_base + ROCEE_SMAC_L_0_REG + @@ -1293,9 +1877,9 @@ static void __hns_roce_v1_cq_clean(struct hns_roce_cq *hr_cq, u32 qpn, } /* - * Now backwards through the CQ, removing CQ entries - * that match our QP by overwriting them with next entries. - */ + * Now backwards through the CQ, removing CQ entries + * that match our QP by overwriting them with next entries. + */ while ((int) --prod_index - (int) hr_cq->cons_index >= 0) { cqe = get_cqe(hr_cq, prod_index & hr_cq->ib_cq.cqe); if ((roce_get_field(cqe->cqe_byte_16, CQE_BYTE_16_LOCAL_QPN_M, @@ -1317,9 +1901,9 @@ static void __hns_roce_v1_cq_clean(struct hns_roce_cq *hr_cq, u32 qpn, if (nfreed) { hr_cq->cons_index += nfreed; /* - * Make sure update of buffer contents is done before - * updating consumer index. - */ + * Make sure update of buffer contents is done before + * updating consumer index. + */ wmb(); hns_roce_v1_cq_set_ci(hr_cq, hr_cq->cons_index); @@ -1339,14 +1923,21 @@ void hns_roce_v1_write_cqc(struct hns_roce_dev *hr_dev, dma_addr_t dma_handle, int nent, u32 vector) { struct hns_roce_cq_context *cq_context = NULL; - void __iomem *tptr_addr; + struct hns_roce_buf_list *tptr_buf; + struct hns_roce_v1_priv *priv; + dma_addr_t tptr_dma_addr; + int offset; + + priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; + tptr_buf = &priv->tptr_table.tptr_buf; cq_context = mb_buf; memset(cq_context, 0, sizeof(*cq_context)); - tptr_addr = 0; - hr_dev->priv_addr = tptr_addr; - hr_cq->tptr_addr = tptr_addr; + /* Get the tptr for this CQ. */ + offset = hr_cq->cqn * HNS_ROCE_V1_TPTR_ENTRY_SIZE; + tptr_dma_addr = tptr_buf->map + offset; + hr_cq->tptr_addr = (u16 *)(tptr_buf->buf + offset); /* Register cq_context members */ roce_set_field(cq_context->cqc_byte_4, @@ -1390,10 +1981,10 @@ void hns_roce_v1_write_cqc(struct hns_roce_dev *hr_dev, roce_set_field(cq_context->cqc_byte_20, CQ_CONTEXT_CQC_BYTE_20_CQE_TPTR_ADDR_H_M, CQ_CONTEXT_CQC_BYTE_20_CQE_TPTR_ADDR_H_S, - (u64)tptr_addr >> 44); + tptr_dma_addr >> 44); cq_context->cqc_byte_20 = cpu_to_le32(cq_context->cqc_byte_20); - cq_context->cqe_tptr_addr_l = (u32)((u64)tptr_addr >> 12); + cq_context->cqe_tptr_addr_l = (u32)(tptr_dma_addr >> 12); roce_set_field(cq_context->cqc_byte_32, CQ_CONTEXT_CQC_BYTE_32_CUR_CQE_BA1_H_M, @@ -1407,7 +1998,7 @@ void hns_roce_v1_write_cqc(struct hns_roce_dev *hr_dev, roce_set_bit(cq_context->cqc_byte_32, CQ_CQNTEXT_CQC_BYTE_32_TYPE_OF_COMPLETION_NOTIFICATION_S, 0); - /*The initial value of cq's ci is 0 */ + /* The initial value of cq's ci is 0 */ roce_set_field(cq_context->cqc_byte_32, CQ_CONTEXT_CQC_BYTE_32_CQ_CONS_IDX_M, CQ_CONTEXT_CQC_BYTE_32_CQ_CONS_IDX_S, 0); @@ -1424,9 +2015,9 @@ int hns_roce_v1_req_notify_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags flags) notification_flag = (flags & IB_CQ_SOLICITED_MASK) == IB_CQ_SOLICITED ? CQ_DB_REQ_NOT : CQ_DB_REQ_NOT_SOL; /* - * flags = 0; Notification Flag = 1, next - * flags = 1; Notification Flag = 0, solocited - */ + * flags = 0; Notification Flag = 1, next + * flags = 1; Notification Flag = 0, solocited + */ doorbell[0] = hr_cq->cons_index & ((hr_cq->cq_depth << 1) - 1); roce_set_bit(doorbell[1], ROCEE_DB_OTHERS_H_ROCEE_DB_OTH_HW_SYNS_S, 1); roce_set_field(doorbell[1], ROCEE_DB_OTHERS_H_ROCEE_DB_OTH_CMD_M, @@ -1581,10 +2172,10 @@ static int hns_roce_v1_poll_one(struct hns_roce_cq *hr_cq, wq = &(*cur_qp)->sq; if ((*cur_qp)->sq_signal_bits) { /* - * If sg_signal_bit is 1, - * firstly tail pointer updated to wqe - * which current cqe correspond to - */ + * If sg_signal_bit is 1, + * firstly tail pointer updated to wqe + * which current cqe correspond to + */ wqe_ctr = (u16)roce_get_field(cqe->cqe_byte_4, CQE_BYTE_4_WQE_INDEX_M, CQE_BYTE_4_WQE_INDEX_S); @@ -1659,8 +2250,14 @@ int hns_roce_v1_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *wc) break; } - if (npolled) + if (npolled) { + *hr_cq->tptr_addr = hr_cq->cons_index & + ((hr_cq->cq_depth << 1) - 1); + + /* Memroy barrier */ + wmb(); hns_roce_v1_cq_set_ci(hr_cq, hr_cq->cons_index); + } spin_unlock_irqrestore(&hr_cq->lock, flags); @@ -1799,12 +2396,12 @@ static int hns_roce_v1_qp_modify(struct hns_roce_dev *hr_dev, if (op[cur_state][new_state] == HNS_ROCE_CMD_2RST_QP) return hns_roce_cmd_mbox(hr_dev, 0, 0, hr_qp->qpn, 2, HNS_ROCE_CMD_2RST_QP, - HNS_ROCE_CMD_TIME_CLASS_A); + HNS_ROCE_CMD_TIMEOUT_MSECS); if (op[cur_state][new_state] == HNS_ROCE_CMD_2ERR_QP) return hns_roce_cmd_mbox(hr_dev, 0, 0, hr_qp->qpn, 2, HNS_ROCE_CMD_2ERR_QP, - HNS_ROCE_CMD_TIME_CLASS_A); + HNS_ROCE_CMD_TIMEOUT_MSECS); mailbox = hns_roce_alloc_cmd_mailbox(hr_dev); if (IS_ERR(mailbox)) @@ -1814,7 +2411,7 @@ static int hns_roce_v1_qp_modify(struct hns_roce_dev *hr_dev, ret = hns_roce_cmd_mbox(hr_dev, mailbox->dma, 0, hr_qp->qpn, 0, op[cur_state][new_state], - HNS_ROCE_CMD_TIME_CLASS_C); + HNS_ROCE_CMD_TIMEOUT_MSECS); hns_roce_free_cmd_mailbox(hr_dev, mailbox); return ret; @@ -2000,11 +2597,11 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr, } /* - *Reset to init - * Mandatory param: - * IB_QP_STATE | IB_QP_PKEY_INDEX | IB_QP_PORT | IB_QP_ACCESS_FLAGS - * Optional param: NA - */ + * Reset to init + * Mandatory param: + * IB_QP_STATE | IB_QP_PKEY_INDEX | IB_QP_PORT | IB_QP_ACCESS_FLAGS + * Optional param: NA + */ if (cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) { roce_set_field(context->qpc_bytes_4, QP_CONTEXT_QPC_BYTES_4_TRANSPORT_SERVICE_TYPE_M, @@ -2172,24 +2769,14 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr, QP_CONTEXT_QPC_BYTE_32_SIGNALING_TYPE_S, hr_qp->sq_signal_bits); - for (port = 0; port < hr_dev->caps.num_ports; port++) { - smac = (u8 *)hr_dev->dev_addr[port]; - dev_dbg(dev, "smac: %2x: %2x: %2x: %2x: %2x: %2x\n", - smac[0], smac[1], smac[2], smac[3], smac[4], - smac[5]); - if ((dmac[0] == smac[0]) && (dmac[1] == smac[1]) && - (dmac[2] == smac[2]) && (dmac[3] == smac[3]) && - (dmac[4] == smac[4]) && (dmac[5] == smac[5])) { - roce_set_bit(context->qpc_bytes_32, - QP_CONTEXT_QPC_BYTE_32_LOOPBACK_INDICATOR_S, - 1); - break; - } - } - - if (hr_dev->loop_idc == 0x1) + port = (attr_mask & IB_QP_PORT) ? (attr->port_num - 1) : + hr_qp->port; + smac = (u8 *)hr_dev->dev_addr[port]; + /* when dmac equals smac or loop_idc is 1, it should loopback */ + if (ether_addr_equal_unaligned(dmac, smac) || + hr_dev->loop_idc == 0x1) roce_set_bit(context->qpc_bytes_32, - QP_CONTEXT_QPC_BYTE_32_LOOPBACK_INDICATOR_S, 1); + QP_CONTEXT_QPC_BYTE_32_LOOPBACK_INDICATOR_S, 1); roce_set_bit(context->qpc_bytes_32, QP_CONTEXT_QPC_BYTE_32_GLOBAL_HEADER_S, @@ -2509,7 +3096,7 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr, /* Every status migrate must change state */ roce_set_field(context->qpc_bytes_144, QP_CONTEXT_QPC_BYTES_144_QP_STATE_M, - QP_CONTEXT_QPC_BYTES_144_QP_STATE_S, attr->qp_state); + QP_CONTEXT_QPC_BYTES_144_QP_STATE_S, new_state); /* SW pass context to HW */ ret = hns_roce_v1_qp_modify(hr_dev, &hr_qp->mtt, @@ -2522,9 +3109,9 @@ static int hns_roce_v1_m_qp(struct ib_qp *ibqp, const struct ib_qp_attr *attr, } /* - * Use rst2init to instead of init2init with drv, - * need to hw to flash RQ HEAD by DB again - */ + * Use rst2init to instead of init2init with drv, + * need to hw to flash RQ HEAD by DB again + */ if (cur_state == IB_QPS_INIT && new_state == IB_QPS_INIT) { /* Memory barrier */ wmb(); @@ -2619,7 +3206,7 @@ static int hns_roce_v1_query_qpc(struct hns_roce_dev *hr_dev, ret = hns_roce_cmd_mbox(hr_dev, 0, mailbox->dma, hr_qp->qpn, 0, HNS_ROCE_CMD_QUERY_QP, - HNS_ROCE_CMD_TIME_CLASS_A); + HNS_ROCE_CMD_TIMEOUT_MSECS); if (!ret) memcpy(hr_context, mailbox->buf, sizeof(*hr_context)); else @@ -2630,8 +3217,78 @@ static int hns_roce_v1_query_qpc(struct hns_roce_dev *hr_dev, return ret; } -int hns_roce_v1_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, - int qp_attr_mask, struct ib_qp_init_attr *qp_init_attr) +static int hns_roce_v1_q_sqp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, + int qp_attr_mask, + struct ib_qp_init_attr *qp_init_attr) +{ + struct hns_roce_dev *hr_dev = to_hr_dev(ibqp->device); + struct hns_roce_qp *hr_qp = to_hr_qp(ibqp); + struct hns_roce_sqp_context context; + u32 addr; + + mutex_lock(&hr_qp->mutex); + + if (hr_qp->state == IB_QPS_RESET) { + qp_attr->qp_state = IB_QPS_RESET; + goto done; + } + + addr = ROCEE_QP1C_CFG0_0_REG + + hr_qp->port * sizeof(struct hns_roce_sqp_context); + context.qp1c_bytes_4 = roce_read(hr_dev, addr); + context.sq_rq_bt_l = roce_read(hr_dev, addr + 1); + context.qp1c_bytes_12 = roce_read(hr_dev, addr + 2); + context.qp1c_bytes_16 = roce_read(hr_dev, addr + 3); + context.qp1c_bytes_20 = roce_read(hr_dev, addr + 4); + context.cur_rq_wqe_ba_l = roce_read(hr_dev, addr + 5); + context.qp1c_bytes_28 = roce_read(hr_dev, addr + 6); + context.qp1c_bytes_32 = roce_read(hr_dev, addr + 7); + context.cur_sq_wqe_ba_l = roce_read(hr_dev, addr + 8); + context.qp1c_bytes_40 = roce_read(hr_dev, addr + 9); + + hr_qp->state = roce_get_field(context.qp1c_bytes_4, + QP1C_BYTES_4_QP_STATE_M, + QP1C_BYTES_4_QP_STATE_S); + qp_attr->qp_state = hr_qp->state; + qp_attr->path_mtu = IB_MTU_256; + qp_attr->path_mig_state = IB_MIG_ARMED; + qp_attr->qkey = QKEY_VAL; + qp_attr->rq_psn = 0; + qp_attr->sq_psn = 0; + qp_attr->dest_qp_num = 1; + qp_attr->qp_access_flags = 6; + + qp_attr->pkey_index = roce_get_field(context.qp1c_bytes_20, + QP1C_BYTES_20_PKEY_IDX_M, + QP1C_BYTES_20_PKEY_IDX_S); + qp_attr->port_num = hr_qp->port + 1; + qp_attr->sq_draining = 0; + qp_attr->max_rd_atomic = 0; + qp_attr->max_dest_rd_atomic = 0; + qp_attr->min_rnr_timer = 0; + qp_attr->timeout = 0; + qp_attr->retry_cnt = 0; + qp_attr->rnr_retry = 0; + qp_attr->alt_timeout = 0; + +done: + qp_attr->cur_qp_state = qp_attr->qp_state; + qp_attr->cap.max_recv_wr = hr_qp->rq.wqe_cnt; + qp_attr->cap.max_recv_sge = hr_qp->rq.max_gs; + qp_attr->cap.max_send_wr = hr_qp->sq.wqe_cnt; + qp_attr->cap.max_send_sge = hr_qp->sq.max_gs; + qp_attr->cap.max_inline_data = 0; + qp_init_attr->cap = qp_attr->cap; + qp_init_attr->create_flags = 0; + + mutex_unlock(&hr_qp->mutex); + + return 0; +} + +static int hns_roce_v1_q_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, + int qp_attr_mask, + struct ib_qp_init_attr *qp_init_attr) { struct hns_roce_dev *hr_dev = to_hr_dev(ibqp->device); struct hns_roce_qp *hr_qp = to_hr_qp(ibqp); @@ -2725,9 +3382,7 @@ int hns_roce_v1_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, qp_attr->pkey_index = roce_get_field(context->qpc_bytes_12, QP_CONTEXT_QPC_BYTES_12_P_KEY_INDEX_M, QP_CONTEXT_QPC_BYTES_12_P_KEY_INDEX_S); - qp_attr->port_num = (u8)roce_get_field(context->qpc_bytes_156, - QP_CONTEXT_QPC_BYTES_156_PORT_NUM_M, - QP_CONTEXT_QPC_BYTES_156_PORT_NUM_S) + 1; + qp_attr->port_num = hr_qp->port + 1; qp_attr->sq_draining = 0; qp_attr->max_rd_atomic = roce_get_field(context->qpc_bytes_156, QP_CONTEXT_QPC_BYTES_156_INITIATOR_DEPTH_M, @@ -2767,134 +3422,397 @@ out: return ret; } -static void hns_roce_v1_destroy_qp_common(struct hns_roce_dev *hr_dev, - struct hns_roce_qp *hr_qp, - int is_user) +int hns_roce_v1_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, + int qp_attr_mask, struct ib_qp_init_attr *qp_init_attr) +{ + struct hns_roce_qp *hr_qp = to_hr_qp(ibqp); + + return hr_qp->doorbell_qpn <= 1 ? + hns_roce_v1_q_sqp(ibqp, qp_attr, qp_attr_mask, qp_init_attr) : + hns_roce_v1_q_qp(ibqp, qp_attr, qp_attr_mask, qp_init_attr); +} + +static int check_qp_db_process_status(struct hns_roce_dev *hr_dev, + struct hns_roce_qp *hr_qp, + u32 sdb_issue_ptr, + u32 *sdb_inv_cnt, + u32 *wait_stage) { - u32 sdbinvcnt; - unsigned long end = 0; - u32 sdbinvcnt_val; - u32 sdbsendptr_val; - u32 sdbisusepr_val; - struct hns_roce_cq *send_cq, *recv_cq; struct device *dev = &hr_dev->pdev->dev; + u32 sdb_retry_cnt, old_retry; + u32 sdb_send_ptr, old_send; + u32 success_flags = 0; + u32 cur_cnt, old_cnt; + unsigned long end; + u32 send_ptr; + u32 inv_cnt; + u32 tsp_st; + + if (*wait_stage > HNS_ROCE_V1_DB_STAGE2 || + *wait_stage < HNS_ROCE_V1_DB_STAGE1) { + dev_err(dev, "QP(0x%lx) db status wait stage(%d) error!\n", + hr_qp->qpn, *wait_stage); + return -EINVAL; + } - if (hr_qp->ibqp.qp_type == IB_QPT_RC) { - if (hr_qp->state != IB_QPS_RESET) { - /* - * Set qp to ERR, - * waiting for hw complete processing all dbs - */ - if (hns_roce_v1_qp_modify(hr_dev, NULL, - to_hns_roce_state( - (enum ib_qp_state)hr_qp->state), - HNS_ROCE_QP_STATE_ERR, NULL, - hr_qp)) - dev_err(dev, "modify QP %06lx to ERR failed.\n", - hr_qp->qpn); - - /* Record issued doorbell */ - sdbisusepr_val = roce_read(hr_dev, - ROCEE_SDB_ISSUE_PTR_REG); - /* - * Query db process status, - * until hw process completely - */ - end = msecs_to_jiffies( - HNS_ROCE_QP_DESTROY_TIMEOUT_MSECS) + jiffies; - do { - sdbsendptr_val = roce_read(hr_dev, + /* Calculate the total timeout for the entire verification process */ + end = msecs_to_jiffies(HNS_ROCE_V1_CHECK_DB_TIMEOUT_MSECS) + jiffies; + + if (*wait_stage == HNS_ROCE_V1_DB_STAGE1) { + /* Query db process status, until hw process completely */ + sdb_send_ptr = roce_read(hr_dev, ROCEE_SDB_SEND_PTR_REG); + while (roce_hw_index_cmp_lt(sdb_send_ptr, sdb_issue_ptr, + ROCEE_SDB_PTR_CMP_BITS)) { + if (!time_before(jiffies, end)) { + dev_dbg(dev, "QP(0x%lx) db process stage1 timeout. issue 0x%x send 0x%x.\n", + hr_qp->qpn, sdb_issue_ptr, + sdb_send_ptr); + return 0; + } + + msleep(HNS_ROCE_V1_CHECK_DB_SLEEP_MSECS); + sdb_send_ptr = roce_read(hr_dev, ROCEE_SDB_SEND_PTR_REG); - if (!time_before(jiffies, end)) { - dev_err(dev, "destroy qp(0x%lx) timeout!!!", - hr_qp->qpn); - break; - } - } while ((short)(roce_get_field(sdbsendptr_val, - ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_M, - ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_S) - - roce_get_field(sdbisusepr_val, - ROCEE_SDB_ISSUE_PTR_SDB_ISSUE_PTR_M, - ROCEE_SDB_ISSUE_PTR_SDB_ISSUE_PTR_S) - ) < 0); + } - /* Get list pointer */ - sdbinvcnt = roce_read(hr_dev, ROCEE_SDB_INV_CNT_REG); + if (roce_get_field(sdb_issue_ptr, + ROCEE_SDB_ISSUE_PTR_SDB_ISSUE_PTR_M, + ROCEE_SDB_ISSUE_PTR_SDB_ISSUE_PTR_S) == + roce_get_field(sdb_send_ptr, + ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_M, + ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_S)) { + old_send = roce_read(hr_dev, ROCEE_SDB_SEND_PTR_REG); + old_retry = roce_read(hr_dev, ROCEE_SDB_RETRY_CNT_REG); - /* Query db's list status, until hw reversal */ do { - sdbinvcnt_val = roce_read(hr_dev, - ROCEE_SDB_INV_CNT_REG); + tsp_st = roce_read(hr_dev, ROCEE_TSP_BP_ST_REG); + if (roce_get_bit(tsp_st, + ROCEE_TSP_BP_ST_QH_FIFO_ENTRY_S) == 1) { + *wait_stage = HNS_ROCE_V1_DB_WAIT_OK; + return 0; + } + if (!time_before(jiffies, end)) { - dev_err(dev, "destroy qp(0x%lx) timeout!!!", - hr_qp->qpn); - dev_err(dev, "SdbInvCnt = 0x%x\n", - sdbinvcnt_val); - break; + dev_dbg(dev, "QP(0x%lx) db process stage1 timeout when send ptr equals issue ptr.\n" + "issue 0x%x send 0x%x.\n", + hr_qp->qpn, sdb_issue_ptr, + sdb_send_ptr); + return 0; + } + + msleep(HNS_ROCE_V1_CHECK_DB_SLEEP_MSECS); + + sdb_send_ptr = roce_read(hr_dev, + ROCEE_SDB_SEND_PTR_REG); + sdb_retry_cnt = roce_read(hr_dev, + ROCEE_SDB_RETRY_CNT_REG); + cur_cnt = roce_get_field(sdb_send_ptr, + ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_M, + ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_S) + + roce_get_field(sdb_retry_cnt, + ROCEE_SDB_RETRY_CNT_SDB_RETRY_CT_M, + ROCEE_SDB_RETRY_CNT_SDB_RETRY_CT_S); + if (!roce_get_bit(tsp_st, + ROCEE_CNT_CLR_CE_CNT_CLR_CE_S)) { + old_cnt = roce_get_field(old_send, + ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_M, + ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_S) + + roce_get_field(old_retry, + ROCEE_SDB_RETRY_CNT_SDB_RETRY_CT_M, + ROCEE_SDB_RETRY_CNT_SDB_RETRY_CT_S); + if (cur_cnt - old_cnt > SDB_ST_CMP_VAL) + success_flags = 1; + } else { + old_cnt = roce_get_field(old_send, + ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_M, + ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_S); + if (cur_cnt - old_cnt > SDB_ST_CMP_VAL) + success_flags = 1; + else { + send_ptr = roce_get_field(old_send, + ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_M, + ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_S) + + roce_get_field(sdb_retry_cnt, + ROCEE_SDB_RETRY_CNT_SDB_RETRY_CT_M, + ROCEE_SDB_RETRY_CNT_SDB_RETRY_CT_S); + roce_set_field(old_send, + ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_M, + ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_S, + send_ptr); + } } - } while ((short)(roce_get_field(sdbinvcnt_val, - ROCEE_SDB_INV_CNT_SDB_INV_CNT_M, - ROCEE_SDB_INV_CNT_SDB_INV_CNT_S) - - (sdbinvcnt + SDB_INV_CNT_OFFSET)) < 0); - - /* Modify qp to reset before destroying qp */ - if (hns_roce_v1_qp_modify(hr_dev, NULL, - to_hns_roce_state( - (enum ib_qp_state)hr_qp->state), - HNS_ROCE_QP_STATE_RST, NULL, hr_qp)) - dev_err(dev, "modify QP %06lx to RESET failed.\n", - hr_qp->qpn); + } while (!success_flags); } + + *wait_stage = HNS_ROCE_V1_DB_STAGE2; + + /* Get list pointer */ + *sdb_inv_cnt = roce_read(hr_dev, ROCEE_SDB_INV_CNT_REG); + dev_dbg(dev, "QP(0x%lx) db process stage2. inv cnt = 0x%x.\n", + hr_qp->qpn, *sdb_inv_cnt); + } + + if (*wait_stage == HNS_ROCE_V1_DB_STAGE2) { + /* Query db's list status, until hw reversal */ + inv_cnt = roce_read(hr_dev, ROCEE_SDB_INV_CNT_REG); + while (roce_hw_index_cmp_lt(inv_cnt, + *sdb_inv_cnt + SDB_INV_CNT_OFFSET, + ROCEE_SDB_CNT_CMP_BITS)) { + if (!time_before(jiffies, end)) { + dev_dbg(dev, "QP(0x%lx) db process stage2 timeout. inv cnt 0x%x.\n", + hr_qp->qpn, inv_cnt); + return 0; + } + + msleep(HNS_ROCE_V1_CHECK_DB_SLEEP_MSECS); + inv_cnt = roce_read(hr_dev, ROCEE_SDB_INV_CNT_REG); + } + + *wait_stage = HNS_ROCE_V1_DB_WAIT_OK; + } + + return 0; +} + +static int check_qp_reset_state(struct hns_roce_dev *hr_dev, + struct hns_roce_qp *hr_qp, + struct hns_roce_qp_work *qp_work_entry, + int *is_timeout) +{ + struct device *dev = &hr_dev->pdev->dev; + u32 sdb_issue_ptr; + int ret; + + if (hr_qp->state != IB_QPS_RESET) { + /* Set qp to ERR, waiting for hw complete processing all dbs */ + ret = hns_roce_v1_modify_qp(&hr_qp->ibqp, NULL, 0, hr_qp->state, + IB_QPS_ERR); + if (ret) { + dev_err(dev, "Modify QP(0x%lx) to ERR failed!\n", + hr_qp->qpn); + return ret; + } + + /* Record issued doorbell */ + sdb_issue_ptr = roce_read(hr_dev, ROCEE_SDB_ISSUE_PTR_REG); + qp_work_entry->sdb_issue_ptr = sdb_issue_ptr; + qp_work_entry->db_wait_stage = HNS_ROCE_V1_DB_STAGE1; + + /* Query db process status, until hw process completely */ + ret = check_qp_db_process_status(hr_dev, hr_qp, sdb_issue_ptr, + &qp_work_entry->sdb_inv_cnt, + &qp_work_entry->db_wait_stage); + if (ret) { + dev_err(dev, "Check QP(0x%lx) db process status failed!\n", + hr_qp->qpn); + return ret; + } + + if (qp_work_entry->db_wait_stage != HNS_ROCE_V1_DB_WAIT_OK) { + qp_work_entry->sche_cnt = 0; + *is_timeout = 1; + return 0; + } + + /* Modify qp to reset before destroying qp */ + ret = hns_roce_v1_modify_qp(&hr_qp->ibqp, NULL, 0, hr_qp->state, + IB_QPS_RESET); + if (ret) { + dev_err(dev, "Modify QP(0x%lx) to RST failed!\n", + hr_qp->qpn); + return ret; + } + } + + return 0; +} + +static void hns_roce_v1_destroy_qp_work_fn(struct work_struct *work) +{ + struct hns_roce_qp_work *qp_work_entry; + struct hns_roce_v1_priv *priv; + struct hns_roce_dev *hr_dev; + struct hns_roce_qp *hr_qp; + struct device *dev; + int ret; + + qp_work_entry = container_of(work, struct hns_roce_qp_work, work); + hr_dev = to_hr_dev(qp_work_entry->ib_dev); + dev = &hr_dev->pdev->dev; + priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; + hr_qp = qp_work_entry->qp; + + dev_dbg(dev, "Schedule destroy QP(0x%lx) work.\n", hr_qp->qpn); + + qp_work_entry->sche_cnt++; + + /* Query db process status, until hw process completely */ + ret = check_qp_db_process_status(hr_dev, hr_qp, + qp_work_entry->sdb_issue_ptr, + &qp_work_entry->sdb_inv_cnt, + &qp_work_entry->db_wait_stage); + if (ret) { + dev_err(dev, "Check QP(0x%lx) db process status failed!\n", + hr_qp->qpn); + return; + } + + if (qp_work_entry->db_wait_stage != HNS_ROCE_V1_DB_WAIT_OK && + priv->des_qp.requeue_flag) { + queue_work(priv->des_qp.qp_wq, work); + return; + } + + /* Modify qp to reset before destroying qp */ + ret = hns_roce_v1_modify_qp(&hr_qp->ibqp, NULL, 0, hr_qp->state, + IB_QPS_RESET); + if (ret) { + dev_err(dev, "Modify QP(0x%lx) to RST failed!\n", hr_qp->qpn); + return; + } + + hns_roce_qp_remove(hr_dev, hr_qp); + hns_roce_qp_free(hr_dev, hr_qp); + + if (hr_qp->ibqp.qp_type == IB_QPT_RC) { + /* RC QP, release QPN */ + hns_roce_release_range_qp(hr_dev, hr_qp->qpn, 1); + kfree(hr_qp); + } else + kfree(hr_to_hr_sqp(hr_qp)); + + kfree(qp_work_entry); + + dev_dbg(dev, "Accomplished destroy QP(0x%lx) work.\n", hr_qp->qpn); +} + +int hns_roce_v1_destroy_qp(struct ib_qp *ibqp) +{ + struct hns_roce_dev *hr_dev = to_hr_dev(ibqp->device); + struct hns_roce_qp *hr_qp = to_hr_qp(ibqp); + struct device *dev = &hr_dev->pdev->dev; + struct hns_roce_qp_work qp_work_entry; + struct hns_roce_qp_work *qp_work; + struct hns_roce_v1_priv *priv; + struct hns_roce_cq *send_cq, *recv_cq; + int is_user = !!ibqp->pd->uobject; + int is_timeout = 0; + int ret; + + ret = check_qp_reset_state(hr_dev, hr_qp, &qp_work_entry, &is_timeout); + if (ret) { + dev_err(dev, "QP reset state check failed(%d)!\n", ret); + return ret; } send_cq = to_hr_cq(hr_qp->ibqp.send_cq); recv_cq = to_hr_cq(hr_qp->ibqp.recv_cq); hns_roce_lock_cqs(send_cq, recv_cq); - if (!is_user) { __hns_roce_v1_cq_clean(recv_cq, hr_qp->qpn, hr_qp->ibqp.srq ? to_hr_srq(hr_qp->ibqp.srq) : NULL); if (send_cq != recv_cq) __hns_roce_v1_cq_clean(send_cq, hr_qp->qpn, NULL); } - - hns_roce_qp_remove(hr_dev, hr_qp); - hns_roce_unlock_cqs(send_cq, recv_cq); - hns_roce_qp_free(hr_dev, hr_qp); + if (!is_timeout) { + hns_roce_qp_remove(hr_dev, hr_qp); + hns_roce_qp_free(hr_dev, hr_qp); - /* Not special_QP, free their QPN */ - if ((hr_qp->ibqp.qp_type == IB_QPT_RC) || - (hr_qp->ibqp.qp_type == IB_QPT_UC) || - (hr_qp->ibqp.qp_type == IB_QPT_UD)) - hns_roce_release_range_qp(hr_dev, hr_qp->qpn, 1); + /* RC QP, release QPN */ + if (hr_qp->ibqp.qp_type == IB_QPT_RC) + hns_roce_release_range_qp(hr_dev, hr_qp->qpn, 1); + } hns_roce_mtt_cleanup(hr_dev, &hr_qp->mtt); - if (is_user) { + if (is_user) ib_umem_release(hr_qp->umem); - } else { + else { kfree(hr_qp->sq.wrid); kfree(hr_qp->rq.wrid); + hns_roce_buf_free(hr_dev, hr_qp->buff_size, &hr_qp->hr_buf); } + + if (!is_timeout) { + if (hr_qp->ibqp.qp_type == IB_QPT_RC) + kfree(hr_qp); + else + kfree(hr_to_hr_sqp(hr_qp)); + } else { + qp_work = kzalloc(sizeof(*qp_work), GFP_KERNEL); + if (!qp_work) + return -ENOMEM; + + INIT_WORK(&qp_work->work, hns_roce_v1_destroy_qp_work_fn); + qp_work->ib_dev = &hr_dev->ib_dev; + qp_work->qp = hr_qp; + qp_work->db_wait_stage = qp_work_entry.db_wait_stage; + qp_work->sdb_issue_ptr = qp_work_entry.sdb_issue_ptr; + qp_work->sdb_inv_cnt = qp_work_entry.sdb_inv_cnt; + qp_work->sche_cnt = qp_work_entry.sche_cnt; + + priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; + queue_work(priv->des_qp.qp_wq, &qp_work->work); + dev_dbg(dev, "Begin destroy QP(0x%lx) work.\n", hr_qp->qpn); + } + + return 0; } -int hns_roce_v1_destroy_qp(struct ib_qp *ibqp) +int hns_roce_v1_destroy_cq(struct ib_cq *ibcq) { - struct hns_roce_dev *hr_dev = to_hr_dev(ibqp->device); - struct hns_roce_qp *hr_qp = to_hr_qp(ibqp); + struct hns_roce_dev *hr_dev = to_hr_dev(ibcq->device); + struct hns_roce_cq *hr_cq = to_hr_cq(ibcq); + struct device *dev = &hr_dev->pdev->dev; + u32 cqe_cnt_ori; + u32 cqe_cnt_cur; + u32 cq_buf_size; + int wait_time = 0; + int ret = 0; - hns_roce_v1_destroy_qp_common(hr_dev, hr_qp, !!ibqp->pd->uobject); + hns_roce_free_cq(hr_dev, hr_cq); - if (hr_qp->ibqp.qp_type == IB_QPT_GSI) - kfree(hr_to_hr_sqp(hr_qp)); - else - kfree(hr_qp); + /* + * Before freeing cq buffer, we need to ensure that the outstanding CQE + * have been written by checking the CQE counter. + */ + cqe_cnt_ori = roce_read(hr_dev, ROCEE_SCAEP_WR_CQE_CNT); + while (1) { + if (roce_read(hr_dev, ROCEE_CAEP_CQE_WCMD_EMPTY) & + HNS_ROCE_CQE_WCMD_EMPTY_BIT) + break; - return 0; + cqe_cnt_cur = roce_read(hr_dev, ROCEE_SCAEP_WR_CQE_CNT); + if ((cqe_cnt_cur - cqe_cnt_ori) >= HNS_ROCE_MIN_CQE_CNT) + break; + + msleep(HNS_ROCE_EACH_FREE_CQ_WAIT_MSECS); + if (wait_time > HNS_ROCE_MAX_FREE_CQ_WAIT_CNT) { + dev_warn(dev, "Destroy cq 0x%lx timeout!\n", + hr_cq->cqn); + ret = -ETIMEDOUT; + break; + } + wait_time++; + } + + hns_roce_mtt_cleanup(hr_dev, &hr_cq->hr_buf.hr_mtt); + + if (ibcq->uobject) + ib_umem_release(hr_cq->umem); + else { + /* Free the buff of stored cq */ + cq_buf_size = (ibcq->cqe + 1) * hr_dev->caps.cq_entry_sz; + hns_roce_buf_free(hr_dev, cq_buf_size, &hr_cq->hr_buf.hr_buf); + } + + kfree(hr_cq); + + return ret; } struct hns_roce_v1_priv hr_v1_priv; @@ -2917,5 +3835,7 @@ struct hns_roce_hw hns_roce_hw_v1 = { .post_recv = hns_roce_v1_post_recv, .req_notify_cq = hns_roce_v1_req_notify_cq, .poll_cq = hns_roce_v1_poll_cq, + .dereg_mr = hns_roce_v1_dereg_mr, + .destroy_cq = hns_roce_v1_destroy_cq, .priv = &hr_v1_priv, }; diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v1.h b/drivers/infiniband/hw/hns/hns_roce_hw_v1.h index 539b0a3b92b0..b213b5e6fef1 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v1.h +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v1.h @@ -58,6 +58,7 @@ #define HNS_ROCE_V1_PHY_UAR_NUM 8 #define HNS_ROCE_V1_GID_NUM 16 +#define HNS_ROCE_V1_RESV_QP 8 #define HNS_ROCE_V1_NUM_COMP_EQE 0x8000 #define HNS_ROCE_V1_NUM_ASYNC_EQE 0x400 @@ -102,8 +103,22 @@ #define HNS_ROCE_V1_EXT_ODB_ALFUL \ (HNS_ROCE_V1_EXT_ODB_DEPTH - HNS_ROCE_V1_DB_RSVD) +#define HNS_ROCE_V1_DB_WAIT_OK 0 +#define HNS_ROCE_V1_DB_STAGE1 1 +#define HNS_ROCE_V1_DB_STAGE2 2 +#define HNS_ROCE_V1_CHECK_DB_TIMEOUT_MSECS 10000 +#define HNS_ROCE_V1_CHECK_DB_SLEEP_MSECS 20 +#define HNS_ROCE_V1_FREE_MR_TIMEOUT_MSECS 50000 +#define HNS_ROCE_V1_RECREATE_LP_QP_TIMEOUT_MSECS 10000 +#define HNS_ROCE_V1_FREE_MR_WAIT_VALUE 5 +#define HNS_ROCE_V1_RECREATE_LP_QP_WAIT_VALUE 20 + #define HNS_ROCE_BT_RSV_BUF_SIZE (1 << 17) +#define HNS_ROCE_V1_TPTR_ENTRY_SIZE 2 +#define HNS_ROCE_V1_TPTR_BUF_SIZE \ + (HNS_ROCE_V1_TPTR_ENTRY_SIZE * HNS_ROCE_V1_MAX_CQ_NUM) + #define HNS_ROCE_ODB_POLL_MODE 0 #define HNS_ROCE_SDB_NORMAL_MODE 0 @@ -140,6 +155,7 @@ #define SQ_PSN_SHIFT 8 #define QKEY_VAL 0x80010000 #define SDB_INV_CNT_OFFSET 8 +#define SDB_ST_CMP_VAL 8 struct hns_roce_cq_context { u32 cqc_byte_4; @@ -436,6 +452,8 @@ struct hns_roce_ud_send_wqe { #define UD_SEND_WQE_U32_8_DMAC_5_M \ (((1UL << 8) - 1) << UD_SEND_WQE_U32_8_DMAC_5_S) +#define UD_SEND_WQE_U32_8_LOOPBACK_INDICATOR_S 22 + #define UD_SEND_WQE_U32_8_OPERATION_TYPE_S 16 #define UD_SEND_WQE_U32_8_OPERATION_TYPE_M \ (((1UL << 4) - 1) << UD_SEND_WQE_U32_8_OPERATION_TYPE_S) @@ -480,13 +498,17 @@ struct hns_roce_sqp_context { u32 qp1c_bytes_12; u32 qp1c_bytes_16; u32 qp1c_bytes_20; - u32 qp1c_bytes_28; u32 cur_rq_wqe_ba_l; + u32 qp1c_bytes_28; u32 qp1c_bytes_32; u32 cur_sq_wqe_ba_l; u32 qp1c_bytes_40; }; +#define QP1C_BYTES_4_QP_STATE_S 0 +#define QP1C_BYTES_4_QP_STATE_M \ + (((1UL << 3) - 1) << QP1C_BYTES_4_QP_STATE_S) + #define QP1C_BYTES_4_SQ_WQE_SHIFT_S 8 #define QP1C_BYTES_4_SQ_WQE_SHIFT_M \ (((1UL << 4) - 1) << QP1C_BYTES_4_SQ_WQE_SHIFT_S) @@ -952,6 +974,10 @@ struct hns_roce_sq_db { #define SQ_DOORBELL_U32_4_SQ_HEAD_M \ (((1UL << 15) - 1) << SQ_DOORBELL_U32_4_SQ_HEAD_S) +#define SQ_DOORBELL_U32_4_SL_S 16 +#define SQ_DOORBELL_U32_4_SL_M \ + (((1UL << 2) - 1) << SQ_DOORBELL_U32_4_SL_S) + #define SQ_DOORBELL_U32_4_PORT_S 18 #define SQ_DOORBELL_U32_4_PORT_M (((1UL << 3) - 1) << SQ_DOORBELL_U32_4_PORT_S) @@ -979,12 +1005,58 @@ struct hns_roce_bt_table { struct hns_roce_buf_list cqc_buf; }; +struct hns_roce_tptr_table { + struct hns_roce_buf_list tptr_buf; +}; + +struct hns_roce_qp_work { + struct work_struct work; + struct ib_device *ib_dev; + struct hns_roce_qp *qp; + u32 db_wait_stage; + u32 sdb_issue_ptr; + u32 sdb_inv_cnt; + u32 sche_cnt; +}; + +struct hns_roce_des_qp { + struct workqueue_struct *qp_wq; + int requeue_flag; +}; + +struct hns_roce_mr_free_work { + struct work_struct work; + struct ib_device *ib_dev; + struct completion *comp; + int comp_flag; + void *mr; +}; + +struct hns_roce_recreate_lp_qp_work { + struct work_struct work; + struct ib_device *ib_dev; + struct completion *comp; + int comp_flag; +}; + +struct hns_roce_free_mr { + struct workqueue_struct *free_mr_wq; + struct hns_roce_qp *mr_free_qp[HNS_ROCE_V1_RESV_QP]; + struct hns_roce_cq *mr_free_cq; + struct hns_roce_pd *mr_free_pd; +}; + struct hns_roce_v1_priv { struct hns_roce_db_table db_table; struct hns_roce_raq_table raq_table; struct hns_roce_bt_table bt_table; + struct hns_roce_tptr_table tptr_table; + struct hns_roce_des_qp des_qp; + struct hns_roce_free_mr free_mr; }; int hns_dsaf_roce_reset(struct fwnode_handle *dsaf_fwnode, bool dereset); +int hns_roce_v1_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *wc); +int hns_roce_v1_destroy_qp(struct ib_qp *ibqp); #endif diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c index 764e35a54457..c3b41f95e70a 100644 --- a/drivers/infiniband/hw/hns/hns_roce_main.c +++ b/drivers/infiniband/hw/hns/hns_roce_main.c @@ -32,55 +32,17 @@ */ #include <linux/acpi.h> #include <linux/of_platform.h> +#include <linux/module.h> #include <rdma/ib_addr.h> #include <rdma/ib_smi.h> #include <rdma/ib_user_verbs.h> +#include <rdma/ib_cache.h> #include "hns_roce_common.h" #include "hns_roce_device.h" -#include "hns_roce_user.h" +#include <rdma/hns-abi.h> #include "hns_roce_hem.h" /** - * hns_roce_addrconf_ifid_eui48 - Get default gid. - * @eui: eui. - * @vlan_id: gid - * @dev: net device - * Description: - * MAC convert to GID - * gid[0..7] = fe80 0000 0000 0000 - * gid[8] = mac[0] ^ 2 - * gid[9] = mac[1] - * gid[10] = mac[2] - * gid[11] = ff (VLAN ID high byte (4 MS bits)) - * gid[12] = fe (VLAN ID low byte) - * gid[13] = mac[3] - * gid[14] = mac[4] - * gid[15] = mac[5] - */ -static void hns_roce_addrconf_ifid_eui48(u8 *eui, u16 vlan_id, - struct net_device *dev) -{ - memcpy(eui, dev->dev_addr, 3); - memcpy(eui + 5, dev->dev_addr + 3, 3); - if (vlan_id < 0x1000) { - eui[3] = vlan_id >> 8; - eui[4] = vlan_id & 0xff; - } else { - eui[3] = 0xff; - eui[4] = 0xfe; - } - eui[0] ^= 2; -} - -static void hns_roce_make_default_gid(struct net_device *dev, union ib_gid *gid) -{ - memset(gid, 0, sizeof(*gid)); - gid->raw[0] = 0xFE; - gid->raw[1] = 0x80; - hns_roce_addrconf_ifid_eui48(&gid->raw[8], 0xffff, dev); -} - -/** * hns_get_gid_index - Get gid index. * @hr_dev: pointer to structure hns_roce_dev. * @port: port, value range: 0 ~ MAX @@ -96,30 +58,6 @@ int hns_get_gid_index(struct hns_roce_dev *hr_dev, u8 port, int gid_index) return gid_index * hr_dev->caps.num_ports + port; } -static int hns_roce_set_gid(struct hns_roce_dev *hr_dev, u8 port, int gid_index, - union ib_gid *gid) -{ - struct device *dev = &hr_dev->pdev->dev; - u8 gid_idx = 0; - - if (gid_index >= hr_dev->caps.gid_table_len[port]) { - dev_err(dev, "gid_index %d illegal, port %d gid range: 0~%d\n", - gid_index, port, hr_dev->caps.gid_table_len[port] - 1); - return -EINVAL; - } - - gid_idx = hns_get_gid_index(hr_dev, port, gid_index); - - if (!memcmp(gid, &hr_dev->iboe.gid_table[gid_idx], sizeof(*gid))) - return -EINVAL; - - memcpy(&hr_dev->iboe.gid_table[gid_idx], gid, sizeof(*gid)); - - hr_dev->hw->set_gid(hr_dev, port, gid_index, gid); - - return 0; -} - static void hns_roce_set_mac(struct hns_roce_dev *hr_dev, u8 port, u8 *addr) { u8 phy_port; @@ -135,27 +73,44 @@ static void hns_roce_set_mac(struct hns_roce_dev *hr_dev, u8 port, u8 *addr) hr_dev->hw->set_mac(hr_dev, phy_port, addr); } -static void hns_roce_set_mtu(struct hns_roce_dev *hr_dev, u8 port, int mtu) +static int hns_roce_add_gid(struct ib_device *device, u8 port_num, + unsigned int index, const union ib_gid *gid, + const struct ib_gid_attr *attr, void **context) { - u8 phy_port = hr_dev->iboe.phy_port[port]; - enum ib_mtu tmp; + struct hns_roce_dev *hr_dev = to_hr_dev(device); + u8 port = port_num - 1; + unsigned long flags; - tmp = iboe_get_mtu(mtu); - if (!tmp) - tmp = IB_MTU_256; + if (port >= hr_dev->caps.num_ports) + return -EINVAL; - hr_dev->hw->set_mtu(hr_dev, phy_port, tmp); + spin_lock_irqsave(&hr_dev->iboe.lock, flags); + + hr_dev->hw->set_gid(hr_dev, port, index, (union ib_gid *)gid); + + spin_unlock_irqrestore(&hr_dev->iboe.lock, flags); + + return 0; } -static void hns_roce_update_gids(struct hns_roce_dev *hr_dev, int port) +static int hns_roce_del_gid(struct ib_device *device, u8 port_num, + unsigned int index, void **context) { - struct ib_event event; + struct hns_roce_dev *hr_dev = to_hr_dev(device); + union ib_gid zgid = { {0} }; + u8 port = port_num - 1; + unsigned long flags; + + if (port >= hr_dev->caps.num_ports) + return -EINVAL; + + spin_lock_irqsave(&hr_dev->iboe.lock, flags); + + hr_dev->hw->set_gid(hr_dev, port, index, &zgid); - /* Refresh gid in ib_cache */ - event.device = &hr_dev->ib_dev; - event.element.port_num = port + 1; - event.event = IB_EVENT_GID_CHANGE; - ib_dispatch_event(&event); + spin_unlock_irqrestore(&hr_dev->iboe.lock, flags); + + return 0; } static int handle_en_event(struct hns_roce_dev *hr_dev, u8 port, @@ -163,9 +118,6 @@ static int handle_en_event(struct hns_roce_dev *hr_dev, u8 port, { struct device *dev = &hr_dev->pdev->dev; struct net_device *netdev; - unsigned long flags; - union ib_gid gid; - int ret = 0; netdev = hr_dev->iboe.netdevs[port]; if (!netdev) { @@ -173,7 +125,7 @@ static int handle_en_event(struct hns_roce_dev *hr_dev, u8 port, return -ENODEV; } - spin_lock_irqsave(&hr_dev->iboe.lock, flags); + spin_lock_bh(&hr_dev->iboe.lock); switch (event) { case NETDEV_UP: @@ -181,23 +133,19 @@ static int handle_en_event(struct hns_roce_dev *hr_dev, u8 port, case NETDEV_REGISTER: case NETDEV_CHANGEADDR: hns_roce_set_mac(hr_dev, port, netdev->dev_addr); - hns_roce_make_default_gid(netdev, &gid); - ret = hns_roce_set_gid(hr_dev, port, 0, &gid); - if (!ret) - hns_roce_update_gids(hr_dev, port); break; case NETDEV_DOWN: /* - * In v1 engine, only support all ports closed together. - */ + * In v1 engine, only support all ports closed together. + */ break; default: dev_dbg(dev, "NETDEV event = 0x%x!\n", (u32)(event)); break; } - spin_unlock_irqrestore(&hr_dev->iboe.lock, flags); - return ret; + spin_unlock_bh(&hr_dev->iboe.lock); + return 0; } static int hns_roce_netdev_event(struct notifier_block *self, @@ -224,118 +172,17 @@ static int hns_roce_netdev_event(struct notifier_block *self, return NOTIFY_DONE; } -static void hns_roce_addr_event(int event, struct net_device *event_netdev, - struct hns_roce_dev *hr_dev, union ib_gid *gid) +static int hns_roce_setup_mtu_mac(struct hns_roce_dev *hr_dev) { - struct hns_roce_ib_iboe *iboe = NULL; - int gid_table_len = 0; - unsigned long flags; - union ib_gid zgid; - u8 gid_idx = 0; - u8 port = 0; - int i = 0; - int free; - struct net_device *real_dev = rdma_vlan_dev_real_dev(event_netdev) ? - rdma_vlan_dev_real_dev(event_netdev) : - event_netdev; - - if (event != NETDEV_UP && event != NETDEV_DOWN) - return; - - iboe = &hr_dev->iboe; - while (port < hr_dev->caps.num_ports) { - if (real_dev == iboe->netdevs[port]) - break; - port++; - } - - if (port >= hr_dev->caps.num_ports) { - dev_dbg(&hr_dev->pdev->dev, "can't find netdev\n"); - return; - } - - memset(zgid.raw, 0, sizeof(zgid.raw)); - free = -1; - gid_table_len = hr_dev->caps.gid_table_len[port]; - - spin_lock_irqsave(&hr_dev->iboe.lock, flags); - - for (i = 0; i < gid_table_len; i++) { - gid_idx = hns_get_gid_index(hr_dev, port, i); - if (!memcmp(gid->raw, iboe->gid_table[gid_idx].raw, - sizeof(gid->raw))) - break; - if (free < 0 && !memcmp(zgid.raw, - iboe->gid_table[gid_idx].raw, sizeof(zgid.raw))) - free = i; - } - - if (i >= gid_table_len) { - if (free < 0) { - spin_unlock_irqrestore(&hr_dev->iboe.lock, flags); - dev_dbg(&hr_dev->pdev->dev, - "gid_index overflow, port(%d)\n", port); - return; - } - if (!hns_roce_set_gid(hr_dev, port, free, gid)) - hns_roce_update_gids(hr_dev, port); - } else if (event == NETDEV_DOWN) { - if (!hns_roce_set_gid(hr_dev, port, i, &zgid)) - hns_roce_update_gids(hr_dev, port); - } - - spin_unlock_irqrestore(&hr_dev->iboe.lock, flags); -} - -static int hns_roce_inet_event(struct notifier_block *self, unsigned long event, - void *ptr) -{ - struct in_ifaddr *ifa = ptr; - struct hns_roce_dev *hr_dev; - struct net_device *dev = ifa->ifa_dev->dev; - union ib_gid gid; - - ipv6_addr_set_v4mapped(ifa->ifa_address, (struct in6_addr *)&gid); - - hr_dev = container_of(self, struct hns_roce_dev, iboe.nb_inet); - - hns_roce_addr_event(event, dev, hr_dev, &gid); - - return NOTIFY_DONE; -} - -static int hns_roce_setup_mtu_gids(struct hns_roce_dev *hr_dev) -{ - struct in_ifaddr *ifa_list = NULL; - union ib_gid gid = {{0} }; - u32 ipaddr = 0; - int index = 0; - int ret = 0; - u8 i = 0; + u8 i; for (i = 0; i < hr_dev->caps.num_ports; i++) { - hns_roce_set_mtu(hr_dev, i, - ib_mtu_enum_to_int(hr_dev->caps.max_mtu)); + hr_dev->hw->set_mtu(hr_dev, hr_dev->iboe.phy_port[i], + hr_dev->caps.max_mtu); hns_roce_set_mac(hr_dev, i, hr_dev->iboe.netdevs[i]->dev_addr); - - if (hr_dev->iboe.netdevs[i]->ip_ptr) { - ifa_list = hr_dev->iboe.netdevs[i]->ip_ptr->ifa_list; - index = 1; - while (ifa_list) { - ipaddr = ifa_list->ifa_address; - ipv6_addr_set_v4mapped(ipaddr, - (struct in6_addr *)&gid); - ret = hns_roce_set_gid(hr_dev, i, index, &gid); - if (ret) - break; - index++; - ifa_list = ifa_list->ifa_next; - } - hns_roce_update_gids(hr_dev, i); - } } - return ret; + return 0; } static int hns_roce_query_device(struct ib_device *ib_dev, @@ -403,7 +250,7 @@ static int hns_roce_query_port(struct ib_device *ib_dev, u8 port_num, assert(port_num > 0); port = port_num - 1; - memset(props, 0, sizeof(*props)); + /* props being zeroed by the caller, avoid zeroing it here */ props->max_mtu = hr_dev->caps.max_mtu; props->gid_tbl_len = hr_dev->caps.gid_table_len[port]; @@ -444,31 +291,6 @@ static enum rdma_link_layer hns_roce_get_link_layer(struct ib_device *device, static int hns_roce_query_gid(struct ib_device *ib_dev, u8 port_num, int index, union ib_gid *gid) { - struct hns_roce_dev *hr_dev = to_hr_dev(ib_dev); - struct device *dev = &hr_dev->pdev->dev; - u8 gid_idx = 0; - u8 port; - - if (port_num < 1 || port_num > hr_dev->caps.num_ports || - index >= hr_dev->caps.gid_table_len[port_num - 1]) { - dev_err(dev, - "port_num %d index %d illegal! correct range: port_num 1~%d index 0~%d!\n", - port_num, index, hr_dev->caps.num_ports, - hr_dev->caps.gid_table_len[port_num - 1] - 1); - return -EINVAL; - } - - port = port_num - 1; - gid_idx = hns_get_gid_index(hr_dev, port, index); - if (gid_idx >= HNS_ROCE_MAX_GID_NUM) { - dev_err(dev, "port_num %d index %d illegal! total gid num %d!\n", - port_num, index, HNS_ROCE_MAX_GID_NUM); - return -EINVAL; - } - - memcpy(gid->raw, hr_dev->iboe.gid_table[gid_idx].raw, - HNS_ROCE_GID_SIZE); - return 0; } @@ -549,6 +371,8 @@ static int hns_roce_dealloc_ucontext(struct ib_ucontext *ibcontext) static int hns_roce_mmap(struct ib_ucontext *context, struct vm_area_struct *vma) { + struct hns_roce_dev *hr_dev = to_hr_dev(context->device); + if (((vma->vm_end - vma->vm_start) % PAGE_SIZE) != 0) return -EINVAL; @@ -558,10 +382,15 @@ static int hns_roce_mmap(struct ib_ucontext *context, to_hr_ucontext(context)->uar.pfn, PAGE_SIZE, vma->vm_page_prot)) return -EAGAIN; - - } else { + } else if (vma->vm_pgoff == 1 && hr_dev->hw_rev == HNS_ROCE_HW_VER1) { + /* vm_pgoff: 1 -- TPTR */ + if (io_remap_pfn_range(vma, vma->vm_start, + hr_dev->tptr_dma_addr >> PAGE_SHIFT, + hr_dev->tptr_size, + vma->vm_page_prot)) + return -EAGAIN; + } else return -EINVAL; - } return 0; } @@ -572,14 +401,15 @@ static int hns_roce_port_immutable(struct ib_device *ib_dev, u8 port_num, struct ib_port_attr attr; int ret; - ret = hns_roce_query_port(ib_dev, port_num, &attr); + immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE; + + ret = ib_query_port(ib_dev, port_num, &attr); if (ret) return ret; immutable->pkey_tbl_len = attr.pkey_tbl_len; immutable->gid_tbl_len = attr.gid_tbl_len; - immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE; immutable->max_mad_size = IB_MGMT_MAD_SIZE; return 0; @@ -605,11 +435,11 @@ static int hns_roce_register_device(struct hns_roce_dev *hr_dev) spin_lock_init(&iboe->lock); ib_dev = &hr_dev->ib_dev; - strlcpy(ib_dev->name, "hisi_%d", IB_DEVICE_NAME_MAX); + strlcpy(ib_dev->name, "hns_%d", IB_DEVICE_NAME_MAX); ib_dev->owner = THIS_MODULE; ib_dev->node_type = RDMA_NODE_IB_CA; - ib_dev->dma_device = dev; + ib_dev->dev.parent = dev; ib_dev->phys_port_cnt = hr_dev->caps.num_ports; ib_dev->local_dma_lkey = hr_dev->caps.reserved_lkey; @@ -639,6 +469,8 @@ static int hns_roce_register_device(struct hns_roce_dev *hr_dev) ib_dev->get_link_layer = hns_roce_get_link_layer; ib_dev->get_netdev = hns_roce_get_netdev; ib_dev->query_gid = hns_roce_query_gid; + ib_dev->add_gid = hns_roce_add_gid; + ib_dev->del_gid = hns_roce_del_gid; ib_dev->query_pkey = hns_roce_query_pkey; ib_dev->alloc_ucontext = hns_roce_alloc_ucontext; ib_dev->dealloc_ucontext = hns_roce_dealloc_ucontext; @@ -681,32 +513,22 @@ static int hns_roce_register_device(struct hns_roce_dev *hr_dev) return ret; } - ret = hns_roce_setup_mtu_gids(hr_dev); + ret = hns_roce_setup_mtu_mac(hr_dev); if (ret) { - dev_err(dev, "roce_setup_mtu_gids failed!\n"); - goto error_failed_setup_mtu_gids; + dev_err(dev, "setup_mtu_mac failed!\n"); + goto error_failed_setup_mtu_mac; } iboe->nb.notifier_call = hns_roce_netdev_event; ret = register_netdevice_notifier(&iboe->nb); if (ret) { dev_err(dev, "register_netdevice_notifier failed!\n"); - goto error_failed_setup_mtu_gids; - } - - iboe->nb_inet.notifier_call = hns_roce_inet_event; - ret = register_inetaddr_notifier(&iboe->nb_inet); - if (ret) { - dev_err(dev, "register inet addr notifier failed!\n"); - goto error_failed_register_inetaddr_notifier; + goto error_failed_setup_mtu_mac; } return 0; -error_failed_register_inetaddr_notifier: - unregister_netdevice_notifier(&iboe->nb); - -error_failed_setup_mtu_gids: +error_failed_setup_mtu_mac: ib_unregister_device(ib_dev); return ret; @@ -940,10 +762,10 @@ err_unmap_mtt: } /** -* hns_roce_setup_hca - setup host channel adapter -* @hr_dev: pointer to hns roce device -* Return : int -*/ + * hns_roce_setup_hca - setup host channel adapter + * @hr_dev: pointer to hns roce device + * Return : int + */ static int hns_roce_setup_hca(struct hns_roce_dev *hr_dev) { int ret; @@ -1008,11 +830,11 @@ err_uar_table_free: } /** -* hns_roce_probe - RoCE driver entrance -* @pdev: pointer to platform device -* Return : int -* -*/ + * hns_roce_probe - RoCE driver entrance + * @pdev: pointer to platform device + * Return : int + * + */ static int hns_roce_probe(struct platform_device *pdev) { int ret; @@ -1023,9 +845,6 @@ static int hns_roce_probe(struct platform_device *pdev) if (!hr_dev) return -ENOMEM; - memset((u8 *)hr_dev + sizeof(struct ib_device), 0, - sizeof(struct hns_roce_dev) - sizeof(struct ib_device)); - hr_dev->pdev = pdev; platform_set_drvdata(pdev, hr_dev); @@ -1125,9 +944,9 @@ error_failed_get_cfg: } /** -* hns_roce_remove - remove RoCE device -* @pdev: pointer to platform device -*/ + * hns_roce_remove - remove RoCE device + * @pdev: pointer to platform device + */ static int hns_roce_remove(struct platform_device *pdev) { struct hns_roce_dev *hr_dev = platform_get_drvdata(pdev); diff --git a/drivers/infiniband/hw/hns/hns_roce_mr.c b/drivers/infiniband/hw/hns/hns_roce_mr.c index fb87883ead34..4139abee3b54 100644 --- a/drivers/infiniband/hw/hns/hns_roce_mr.c +++ b/drivers/infiniband/hw/hns/hns_roce_mr.c @@ -42,7 +42,7 @@ static u32 hw_index_to_key(unsigned long ind) return (u32)(ind >> 24) | (ind << 8); } -static unsigned long key_to_hw_index(u32 key) +unsigned long key_to_hw_index(u32 key) { return (key << 24) | (key >> 8); } @@ -53,16 +53,16 @@ static int hns_roce_sw2hw_mpt(struct hns_roce_dev *hr_dev, { return hns_roce_cmd_mbox(hr_dev, mailbox->dma, 0, mpt_index, 0, HNS_ROCE_CMD_SW2HW_MPT, - HNS_ROCE_CMD_TIME_CLASS_B); + HNS_ROCE_CMD_TIMEOUT_MSECS); } -static int hns_roce_hw2sw_mpt(struct hns_roce_dev *hr_dev, +int hns_roce_hw2sw_mpt(struct hns_roce_dev *hr_dev, struct hns_roce_cmd_mailbox *mailbox, unsigned long mpt_index) { return hns_roce_cmd_mbox(hr_dev, 0, mailbox ? mailbox->dma : 0, mpt_index, !mailbox, HNS_ROCE_CMD_HW2SW_MPT, - HNS_ROCE_CMD_TIME_CLASS_B); + HNS_ROCE_CMD_TIMEOUT_MSECS); } static int hns_roce_buddy_alloc(struct hns_roce_buddy *buddy, int order, @@ -137,11 +137,13 @@ static int hns_roce_buddy_init(struct hns_roce_buddy *buddy, int max_order) for (i = 0; i <= buddy->max_order; ++i) { s = BITS_TO_LONGS(1 << (buddy->max_order - i)); - buddy->bits[i] = kmalloc_array(s, sizeof(long), GFP_KERNEL); - if (!buddy->bits[i]) - goto err_out_free; - - bitmap_zero(buddy->bits[i], 1 << (buddy->max_order - i)); + buddy->bits[i] = kcalloc(s, sizeof(long), GFP_KERNEL | + __GFP_NOWARN); + if (!buddy->bits[i]) { + buddy->bits[i] = vzalloc(s * sizeof(long)); + if (!buddy->bits[i]) + goto err_out_free; + } } set_bit(0, buddy->bits[buddy->max_order]); @@ -151,7 +153,7 @@ static int hns_roce_buddy_init(struct hns_roce_buddy *buddy, int max_order) err_out_free: for (i = 0; i <= buddy->max_order; ++i) - kfree(buddy->bits[i]); + kvfree(buddy->bits[i]); err_out: kfree(buddy->bits); @@ -164,7 +166,7 @@ static void hns_roce_buddy_cleanup(struct hns_roce_buddy *buddy) int i; for (i = 0; i <= buddy->max_order; ++i) - kfree(buddy->bits[i]); + kvfree(buddy->bits[i]); kfree(buddy->bits); kfree(buddy->num_free); @@ -287,7 +289,7 @@ static void hns_roce_mr_free(struct hns_roce_dev *hr_dev, } hns_roce_bitmap_free(&hr_dev->mr_table.mtpt_bitmap, - key_to_hw_index(mr->key)); + key_to_hw_index(mr->key), BITMAP_NO_RR); } static int hns_roce_mr_enable(struct hns_roce_dev *hr_dev, @@ -605,13 +607,20 @@ err_free: int hns_roce_dereg_mr(struct ib_mr *ibmr) { + struct hns_roce_dev *hr_dev = to_hr_dev(ibmr->device); struct hns_roce_mr *mr = to_hr_mr(ibmr); + int ret = 0; - hns_roce_mr_free(to_hr_dev(ibmr->device), mr); - if (mr->umem) - ib_umem_release(mr->umem); + if (hr_dev->hw->dereg_mr) { + ret = hr_dev->hw->dereg_mr(hr_dev, mr); + } else { + hns_roce_mr_free(hr_dev, mr); - kfree(mr); + if (mr->umem) + ib_umem_release(mr->umem); - return 0; + kfree(mr); + } + + return ret; } diff --git a/drivers/infiniband/hw/hns/hns_roce_pd.c b/drivers/infiniband/hw/hns/hns_roce_pd.c index 05db7d59812a..a64500fa1145 100644 --- a/drivers/infiniband/hw/hns/hns_roce_pd.c +++ b/drivers/infiniband/hw/hns/hns_roce_pd.c @@ -40,7 +40,7 @@ static int hns_roce_pd_alloc(struct hns_roce_dev *hr_dev, unsigned long *pdn) static void hns_roce_pd_free(struct hns_roce_dev *hr_dev, unsigned long pdn) { - hns_roce_bitmap_free(&hr_dev->pd_bitmap, pdn); + hns_roce_bitmap_free(&hr_dev->pd_bitmap, pdn, BITMAP_NO_RR); } int hns_roce_init_pd_table(struct hns_roce_dev *hr_dev) @@ -121,7 +121,8 @@ int hns_roce_uar_alloc(struct hns_roce_dev *hr_dev, struct hns_roce_uar *uar) void hns_roce_uar_free(struct hns_roce_dev *hr_dev, struct hns_roce_uar *uar) { - hns_roce_bitmap_free(&hr_dev->uar_table.bitmap, uar->index); + hns_roce_bitmap_free(&hr_dev->uar_table.bitmap, uar->index, + BITMAP_NO_RR); } int hns_roce_init_uar_table(struct hns_roce_dev *hr_dev) diff --git a/drivers/infiniband/hw/hns/hns_roce_qp.c b/drivers/infiniband/hw/hns/hns_roce_qp.c index e86dd8d06777..3f44f2f91f03 100644 --- a/drivers/infiniband/hw/hns/hns_roce_qp.c +++ b/drivers/infiniband/hw/hns/hns_roce_qp.c @@ -37,7 +37,7 @@ #include "hns_roce_common.h" #include "hns_roce_device.h" #include "hns_roce_hem.h" -#include "hns_roce_user.h" +#include <rdma/hns-abi.h> #define SQP_NUM (2 * HNS_ROCE_MAX_PORTS) @@ -101,7 +101,7 @@ static void hns_roce_ib_qp_event(struct hns_roce_qp *hr_qp, event.event = IB_EVENT_QP_ACCESS_ERR; break; default: - dev_dbg(ibqp->device->dma_device, "roce_ib: Unexpected event type %d on QP %06lx\n", + dev_dbg(ibqp->device->dev.parent, "roce_ib: Unexpected event type %d on QP %06lx\n", type, hr_qp->qpn); return; } @@ -250,7 +250,7 @@ void hns_roce_release_range_qp(struct hns_roce_dev *hr_dev, int base_qpn, if (base_qpn < SQP_NUM) return; - hns_roce_bitmap_free_range(&qp_table->bitmap, base_qpn, cnt); + hns_roce_bitmap_free_range(&qp_table->bitmap, base_qpn, cnt, BITMAP_RR); } static int hns_roce_set_rq_size(struct hns_roce_dev *hr_dev, diff --git a/drivers/infiniband/hw/i40iw/i40iw.h b/drivers/infiniband/hw/i40iw/i40iw.h index 8ec09e470f84..da2eb5a281fa 100644 --- a/drivers/infiniband/hw/i40iw/i40iw.h +++ b/drivers/infiniband/hw/i40iw/i40iw.h @@ -112,9 +112,12 @@ #define I40IW_DRV_OPT_MCAST_LOGPORT_MAP 0x00000800 #define IW_HMC_OBJ_TYPE_NUM ARRAY_SIZE(iw_hmc_obj_types) -#define IW_CFG_FPM_QP_COUNT 32768 -#define I40IW_MAX_PAGES_PER_FMR 512 -#define I40IW_MIN_PAGES_PER_FMR 1 +#define IW_CFG_FPM_QP_COUNT 32768 +#define I40IW_MAX_PAGES_PER_FMR 512 +#define I40IW_MIN_PAGES_PER_FMR 1 +#define I40IW_CQP_COMPL_RQ_WQE_FLUSHED 2 +#define I40IW_CQP_COMPL_SQ_WQE_FLUSHED 3 +#define I40IW_CQP_COMPL_RQ_SQ_WQE_FLUSHED 4 #define I40IW_MTU_TO_MSS 40 #define I40IW_DEFAULT_MSS 1460 @@ -210,6 +213,12 @@ struct i40iw_msix_vector { u32 ceq_id; }; +struct l2params_work { + struct work_struct work; + struct i40iw_device *iwdev; + struct i40iw_l2params l2params; +}; + #define I40IW_MSIX_TABLE_SIZE 65 struct virtchnl_work { @@ -227,6 +236,7 @@ struct i40iw_device { struct net_device *netdev; wait_queue_head_t vchnl_waitq; struct i40iw_sc_dev sc_dev; + struct i40iw_sc_vsi vsi; struct i40iw_handler *hdl; struct i40e_info *ldev; struct i40e_client *client; @@ -280,7 +290,6 @@ struct i40iw_device { u32 sd_type; struct workqueue_struct *param_wq; atomic_t params_busy; - u32 mss; enum init_completion_state init_state; u16 mac_ip_table_idx; atomic_t vchnl_msgs; @@ -297,6 +306,14 @@ struct i40iw_device { u32 mr_stagmask; u32 mpa_version; bool dcb; + bool closing; + bool reset; + u32 used_pds; + u32 used_cqs; + u32 used_mrs; + u32 used_qps; + wait_queue_head_t close_wq; + atomic64_t use_count; }; struct i40iw_ib_device { @@ -498,7 +515,7 @@ u32 i40iw_initialize_hw_resources(struct i40iw_device *iwdev); int i40iw_register_rdma_device(struct i40iw_device *iwdev); void i40iw_port_ibevent(struct i40iw_device *iwdev); -int i40iw_cm_disconn(struct i40iw_qp *); +void i40iw_cm_disconn(struct i40iw_qp *iwqp); void i40iw_cm_disconn_worker(void *); int mini_cm_recv_pkt(struct i40iw_cm_core *, struct i40iw_device *, struct sk_buff *); @@ -508,20 +525,26 @@ enum i40iw_status_code i40iw_handle_cqp_op(struct i40iw_device *iwdev, enum i40iw_status_code i40iw_add_mac_addr(struct i40iw_device *iwdev, u8 *mac_addr, u8 *mac_index); int i40iw_modify_qp(struct ib_qp *, struct ib_qp_attr *, int, struct ib_udata *); +void i40iw_cq_wq_destroy(struct i40iw_device *iwdev, struct i40iw_sc_cq *cq); void i40iw_rem_pdusecount(struct i40iw_pd *iwpd, struct i40iw_device *iwdev); void i40iw_add_pdusecount(struct i40iw_pd *iwpd); +void i40iw_rem_devusecount(struct i40iw_device *iwdev); +void i40iw_add_devusecount(struct i40iw_device *iwdev); void i40iw_hw_modify_qp(struct i40iw_device *iwdev, struct i40iw_qp *iwqp, struct i40iw_modify_qp_info *info, bool wait); +void i40iw_qp_suspend_resume(struct i40iw_sc_dev *dev, + struct i40iw_sc_qp *qp, + bool suspend); enum i40iw_status_code i40iw_manage_qhash(struct i40iw_device *iwdev, struct i40iw_cm_info *cminfo, enum i40iw_quad_entry_type etype, enum i40iw_quad_hash_manage_type mtype, void *cmnode, bool wait); -void i40iw_receive_ilq(struct i40iw_sc_dev *dev, struct i40iw_puda_buf *rbuf); -void i40iw_free_sqbuf(struct i40iw_sc_dev *dev, void *bufp); +void i40iw_receive_ilq(struct i40iw_sc_vsi *vsi, struct i40iw_puda_buf *rbuf); +void i40iw_free_sqbuf(struct i40iw_sc_vsi *vsi, void *bufp); void i40iw_free_qp_resources(struct i40iw_device *iwdev, struct i40iw_qp *iwqp, u32 qp_num); diff --git a/drivers/infiniband/hw/i40iw/i40iw_cm.c b/drivers/infiniband/hw/i40iw/i40iw_cm.c index 85637696f6e9..95a0586a4da8 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_cm.c +++ b/drivers/infiniband/hw/i40iw/i40iw_cm.c @@ -68,13 +68,13 @@ static void i40iw_disconnect_worker(struct work_struct *work); /** * i40iw_free_sqbuf - put back puda buffer if refcount = 0 - * @dev: FPK device + * @vsi: pointer to vsi structure * @buf: puda buffer to free */ -void i40iw_free_sqbuf(struct i40iw_sc_dev *dev, void *bufp) +void i40iw_free_sqbuf(struct i40iw_sc_vsi *vsi, void *bufp) { struct i40iw_puda_buf *buf = (struct i40iw_puda_buf *)bufp; - struct i40iw_puda_rsrc *ilq = dev->ilq; + struct i40iw_puda_rsrc *ilq = vsi->ilq; if (!atomic_dec_return(&buf->refcount)) i40iw_puda_ret_bufpool(ilq, buf); @@ -221,6 +221,7 @@ static void i40iw_get_addr_info(struct i40iw_cm_node *cm_node, memcpy(cm_info->rem_addr, cm_node->rem_addr, sizeof(cm_info->rem_addr)); cm_info->loc_port = cm_node->loc_port; cm_info->rem_port = cm_node->rem_port; + cm_info->user_pri = cm_node->user_pri; } /** @@ -271,6 +272,7 @@ static int i40iw_send_cm_event(struct i40iw_cm_node *cm_node, event.provider_data = (void *)cm_node; event.private_data = (void *)cm_node->pdata_buf; event.private_data_len = (u8)cm_node->pdata.size; + event.ird = cm_node->ird_size; break; case IW_CM_EVENT_CONNECT_REPLY: i40iw_get_cmevent_info(cm_node, cm_id, &event); @@ -335,13 +337,13 @@ static struct i40iw_cm_event *i40iw_create_event(struct i40iw_cm_node *cm_node, */ static void i40iw_free_retrans_entry(struct i40iw_cm_node *cm_node) { - struct i40iw_sc_dev *dev = cm_node->dev; + struct i40iw_device *iwdev = cm_node->iwdev; struct i40iw_timer_entry *send_entry; send_entry = cm_node->send_entry; if (send_entry) { cm_node->send_entry = NULL; - i40iw_free_sqbuf(dev, (void *)send_entry->sqbuf); + i40iw_free_sqbuf(&iwdev->vsi, (void *)send_entry->sqbuf); kfree(send_entry); atomic_dec(&cm_node->ref_count); } @@ -360,15 +362,6 @@ static void i40iw_cleanup_retrans_entry(struct i40iw_cm_node *cm_node) spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags); } -static bool is_remote_ne020_or_chelsio(struct i40iw_cm_node *cm_node) -{ - if ((cm_node->rem_mac[0] == 0x0) && - (((cm_node->rem_mac[1] == 0x12) && (cm_node->rem_mac[2] == 0x55)) || - ((cm_node->rem_mac[1] == 0x07 && (cm_node->rem_mac[2] == 0x43))))) - return true; - return false; -} - /** * i40iw_form_cm_frame - get a free packet and build frame * @cm_node: connection's node ionfo to use in frame @@ -384,7 +377,7 @@ static struct i40iw_puda_buf *i40iw_form_cm_frame(struct i40iw_cm_node *cm_node, u8 flags) { struct i40iw_puda_buf *sqbuf; - struct i40iw_sc_dev *dev = cm_node->dev; + struct i40iw_sc_vsi *vsi = &cm_node->iwdev->vsi; u8 *buf; struct tcphdr *tcph; @@ -396,8 +389,9 @@ static struct i40iw_puda_buf *i40iw_form_cm_frame(struct i40iw_cm_node *cm_node, u32 opts_len = 0; u32 pd_len = 0; u32 hdr_len = 0; + u16 vtag; - sqbuf = i40iw_puda_get_bufpool(dev->ilq); + sqbuf = i40iw_puda_get_bufpool(vsi->ilq); if (!sqbuf) return NULL; buf = sqbuf->mem.va; @@ -408,11 +402,8 @@ static struct i40iw_puda_buf *i40iw_form_cm_frame(struct i40iw_cm_node *cm_node, if (hdr) hdr_len = hdr->size; - if (pdata) { + if (pdata) pd_len = pdata->size; - if (!is_remote_ne020_or_chelsio(cm_node)) - pd_len += MPA_ZERO_PAD_LEN; - } if (cm_node->vlan_id < VLAN_TAG_PRESENT) eth_hlen += 4; @@ -445,7 +436,8 @@ static struct i40iw_puda_buf *i40iw_form_cm_frame(struct i40iw_cm_node *cm_node, ether_addr_copy(ethh->h_source, cm_node->loc_mac); if (cm_node->vlan_id < VLAN_TAG_PRESENT) { ((struct vlan_ethhdr *)ethh)->h_vlan_proto = htons(ETH_P_8021Q); - ((struct vlan_ethhdr *)ethh)->h_vlan_TCI = htons(cm_node->vlan_id); + vtag = (cm_node->user_pri << VLAN_PRIO_SHIFT) | cm_node->vlan_id; + ((struct vlan_ethhdr *)ethh)->h_vlan_TCI = htons(vtag); ((struct vlan_ethhdr *)ethh)->h_vlan_encapsulated_proto = htons(ETH_P_IP); } else { @@ -454,7 +446,7 @@ static struct i40iw_puda_buf *i40iw_form_cm_frame(struct i40iw_cm_node *cm_node, iph->version = IPVERSION; iph->ihl = 5; /* 5 * 4Byte words, IP headr len */ - iph->tos = 0; + iph->tos = cm_node->tos; iph->tot_len = htons(packetsize); iph->id = htons(++cm_node->tcp_cntxt.loc_id); @@ -474,13 +466,15 @@ static struct i40iw_puda_buf *i40iw_form_cm_frame(struct i40iw_cm_node *cm_node, ether_addr_copy(ethh->h_source, cm_node->loc_mac); if (cm_node->vlan_id < VLAN_TAG_PRESENT) { ((struct vlan_ethhdr *)ethh)->h_vlan_proto = htons(ETH_P_8021Q); - ((struct vlan_ethhdr *)ethh)->h_vlan_TCI = htons(cm_node->vlan_id); + vtag = (cm_node->user_pri << VLAN_PRIO_SHIFT) | cm_node->vlan_id; + ((struct vlan_ethhdr *)ethh)->h_vlan_TCI = htons(vtag); ((struct vlan_ethhdr *)ethh)->h_vlan_encapsulated_proto = htons(ETH_P_IPV6); } else { ethh->h_proto = htons(ETH_P_IPV6); } ip6h->version = 6; - ip6h->flow_lbl[0] = 0; + ip6h->priority = cm_node->tos >> 4; + ip6h->flow_lbl[0] = cm_node->tos << 4; ip6h->flow_lbl[1] = 0; ip6h->flow_lbl[2] = 0; ip6h->payload_len = htons(packetsize - sizeof(*ip6h)); @@ -1065,7 +1059,7 @@ int i40iw_schedule_cm_timer(struct i40iw_cm_node *cm_node, int send_retrans, int close_when_complete) { - struct i40iw_sc_dev *dev = cm_node->dev; + struct i40iw_sc_vsi *vsi = &cm_node->iwdev->vsi; struct i40iw_cm_core *cm_core = cm_node->cm_core; struct i40iw_timer_entry *new_send; int ret = 0; @@ -1074,7 +1068,7 @@ int i40iw_schedule_cm_timer(struct i40iw_cm_node *cm_node, new_send = kzalloc(sizeof(*new_send), GFP_ATOMIC); if (!new_send) { - i40iw_free_sqbuf(cm_node->dev, (void *)sqbuf); + i40iw_free_sqbuf(vsi, (void *)sqbuf); return -ENOMEM; } new_send->retrycount = I40IW_DEFAULT_RETRYS; @@ -1089,7 +1083,7 @@ int i40iw_schedule_cm_timer(struct i40iw_cm_node *cm_node, new_send->timetosend += (HZ / 10); if (cm_node->close_entry) { kfree(new_send); - i40iw_free_sqbuf(cm_node->dev, (void *)sqbuf); + i40iw_free_sqbuf(vsi, (void *)sqbuf); i40iw_pr_err("already close entry\n"); return -EINVAL; } @@ -1104,7 +1098,7 @@ int i40iw_schedule_cm_timer(struct i40iw_cm_node *cm_node, new_send->timetosend = jiffies + I40IW_RETRY_TIMEOUT; atomic_inc(&sqbuf->refcount); - i40iw_puda_send_buf(dev->ilq, sqbuf); + i40iw_puda_send_buf(vsi->ilq, sqbuf); if (!send_retrans) { i40iw_cleanup_retrans_entry(cm_node); if (close_when_complete) @@ -1201,6 +1195,7 @@ static void i40iw_cm_timer_tick(unsigned long pass) struct i40iw_cm_node *cm_node; struct i40iw_timer_entry *send_entry, *close_entry; struct list_head *list_core_temp; + struct i40iw_sc_vsi *vsi; struct list_head *list_node; struct i40iw_cm_core *cm_core = (struct i40iw_cm_core *)pass; u32 settimer = 0; @@ -1276,9 +1271,10 @@ static void i40iw_cm_timer_tick(unsigned long pass) cm_node->cm_core->stats_pkt_retrans++; spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags); + vsi = &cm_node->iwdev->vsi; dev = cm_node->dev; atomic_inc(&send_entry->sqbuf->refcount); - i40iw_puda_send_buf(dev->ilq, send_entry->sqbuf); + i40iw_puda_send_buf(vsi->ilq, send_entry->sqbuf); spin_lock_irqsave(&cm_node->retrans_list_lock, flags); if (send_entry->send_retrans) { send_entry->retranscount--; @@ -1379,10 +1375,11 @@ int i40iw_send_syn(struct i40iw_cm_node *cm_node, u32 sendack) static void i40iw_send_ack(struct i40iw_cm_node *cm_node) { struct i40iw_puda_buf *sqbuf; + struct i40iw_sc_vsi *vsi = &cm_node->iwdev->vsi; sqbuf = i40iw_form_cm_frame(cm_node, NULL, NULL, NULL, SET_ACK); if (sqbuf) - i40iw_puda_send_buf(cm_node->dev->ilq, sqbuf); + i40iw_puda_send_buf(vsi->ilq, sqbuf); else i40iw_pr_err("no sqbuf\n"); } @@ -1564,9 +1561,15 @@ static enum i40iw_status_code i40iw_del_multiple_qhash( memcpy(cm_info->loc_addr, child_listen_node->loc_addr, sizeof(cm_info->loc_addr)); cm_info->vlan_id = child_listen_node->vlan_id; - ret = i40iw_manage_qhash(iwdev, cm_info, - I40IW_QHASH_TYPE_TCP_SYN, - I40IW_QHASH_MANAGE_TYPE_DELETE, NULL, false); + if (child_listen_node->qhash_set) { + ret = i40iw_manage_qhash(iwdev, cm_info, + I40IW_QHASH_TYPE_TCP_SYN, + I40IW_QHASH_MANAGE_TYPE_DELETE, + NULL, false); + child_listen_node->qhash_set = false; + } else { + ret = I40IW_SUCCESS; + } i40iw_debug(&iwdev->sc_dev, I40IW_DEBUG_CM, "freed pointer = %p\n", @@ -1591,9 +1594,10 @@ static enum i40iw_status_code i40iw_del_multiple_qhash( static struct net_device *i40iw_netdev_vlan_ipv6(u32 *addr, u16 *vlan_id, u8 *mac) { struct net_device *ip_dev = NULL; -#if IS_ENABLED(CONFIG_IPV6) struct in6_addr laddr6; + if (!IS_ENABLED(CONFIG_IPV6)) + return NULL; i40iw_copy_ip_htonl(laddr6.in6_u.u6_addr32, addr); if (vlan_id) *vlan_id = I40IW_NO_VLAN; @@ -1610,7 +1614,6 @@ static struct net_device *i40iw_netdev_vlan_ipv6(u32 *addr, u16 *vlan_id, u8 *ma } } rcu_read_unlock(); -#endif return ip_dev; } @@ -1646,7 +1649,7 @@ static enum i40iw_status_code i40iw_add_mqh_6(struct i40iw_device *iwdev, { struct net_device *ip_dev; struct inet6_dev *idev; - struct inet6_ifaddr *ifp; + struct inet6_ifaddr *ifp, *tmp; enum i40iw_status_code ret = 0; struct i40iw_cm_listener *child_listen_node; unsigned long flags; @@ -1661,7 +1664,7 @@ static enum i40iw_status_code i40iw_add_mqh_6(struct i40iw_device *iwdev, i40iw_pr_err("idev == NULL\n"); break; } - list_for_each_entry(ifp, &idev->addr_list, if_list) { + list_for_each_entry_safe(ifp, tmp, &idev->addr_list, if_list) { i40iw_debug(&iwdev->sc_dev, I40IW_DEBUG_CM, "IP=%pI6, vlan_id=%d, MAC=%pM\n", @@ -1675,7 +1678,6 @@ static enum i40iw_status_code i40iw_add_mqh_6(struct i40iw_device *iwdev, "Allocating child listener %p\n", child_listen_node); if (!child_listen_node) { - i40iw_pr_err("listener memory allocation\n"); ret = I40IW_ERR_NO_MEMORY; goto exit; } @@ -1695,6 +1697,7 @@ static enum i40iw_status_code i40iw_add_mqh_6(struct i40iw_device *iwdev, I40IW_QHASH_MANAGE_TYPE_ADD, NULL, true); if (!ret) { + child_listen_node->qhash_set = true; spin_lock_irqsave(&iwdev->cm_core.listen_list_lock, flags); list_add(&child_listen_node->child_listen_list, &cm_parent_listen_node->child_listen_list); @@ -1751,7 +1754,6 @@ static enum i40iw_status_code i40iw_add_mqh_4( "Allocating child listener %p\n", child_listen_node); if (!child_listen_node) { - i40iw_pr_err("listener memory allocation\n"); in_dev_put(idev); ret = I40IW_ERR_NO_MEMORY; goto exit; @@ -1773,6 +1775,7 @@ static enum i40iw_status_code i40iw_add_mqh_4( NULL, true); if (!ret) { + child_listen_node->qhash_set = true; spin_lock_irqsave(&iwdev->cm_core.listen_list_lock, flags); list_add(&child_listen_node->child_listen_list, &cm_parent_listen_node->child_listen_list); @@ -1880,6 +1883,7 @@ static int i40iw_dec_refcnt_listen(struct i40iw_cm_core *cm_core, nfo.loc_port = listener->loc_port; nfo.ipv4 = listener->ipv4; nfo.vlan_id = listener->vlan_id; + nfo.user_pri = listener->user_pri; if (!list_empty(&listener->child_listen_list)) { i40iw_del_multiple_qhash(listener->iwdev, &nfo, listener); @@ -2138,6 +2142,20 @@ static struct i40iw_cm_node *i40iw_make_cm_node( /* set our node specific transport info */ cm_node->ipv4 = cm_info->ipv4; cm_node->vlan_id = cm_info->vlan_id; + if ((cm_node->vlan_id == I40IW_NO_VLAN) && iwdev->dcb) + cm_node->vlan_id = 0; + cm_node->tos = cm_info->tos; + cm_node->user_pri = cm_info->user_pri; + if (listener) { + if (listener->tos != cm_info->tos) + i40iw_debug(&iwdev->sc_dev, I40IW_DEBUG_DCB, + "application TOS[%d] and remote client TOS[%d] mismatch\n", + listener->tos, cm_info->tos); + cm_node->tos = max(listener->tos, cm_info->tos); + cm_node->user_pri = rt_tos2priority(cm_node->tos); + i40iw_debug(&iwdev->sc_dev, I40IW_DEBUG_DCB, "listener: TOS:[%d] UP:[%d]\n", + cm_node->tos, cm_node->user_pri); + } memcpy(cm_node->loc_addr, cm_info->loc_addr, sizeof(cm_node->loc_addr)); memcpy(cm_node->rem_addr, cm_info->rem_addr, sizeof(cm_node->rem_addr)); cm_node->loc_port = cm_info->loc_port; @@ -2162,7 +2180,7 @@ static struct i40iw_cm_node *i40iw_make_cm_node( I40IW_CM_DEFAULT_RCV_WND_SCALED >> I40IW_CM_DEFAULT_RCV_WND_SCALE; ts = current_kernel_time(); cm_node->tcp_cntxt.loc_seq_num = ts.tv_nsec; - cm_node->tcp_cntxt.mss = iwdev->mss; + cm_node->tcp_cntxt.mss = iwdev->vsi.mss; cm_node->iwdev = iwdev; cm_node->dev = &iwdev->sc_dev; @@ -2236,7 +2254,7 @@ static void i40iw_rem_ref_cm_node(struct i40iw_cm_node *cm_node) i40iw_dec_refcnt_listen(cm_core, cm_node->listener, 0, true); } else { if (!i40iw_listen_port_in_use(cm_core, cm_node->loc_port) && - cm_node->apbvt_set && cm_node->iwdev) { + cm_node->apbvt_set) { i40iw_manage_apbvt(cm_node->iwdev, cm_node->loc_port, I40IW_MANAGE_APBVT_DEL); @@ -2861,7 +2879,7 @@ static struct i40iw_cm_node *i40iw_create_cm_node( /* create a CM connection node */ cm_node = i40iw_make_cm_node(cm_core, iwdev, cm_info, NULL); if (!cm_node) - return NULL; + return ERR_PTR(-ENOMEM); /* set our node side to client (active) side */ cm_node->tcp_cntxt.client = 1; cm_node->tcp_cntxt.rcv_wscale = I40IW_CM_DEFAULT_RCV_WND_SCALE; @@ -2874,7 +2892,8 @@ static struct i40iw_cm_node *i40iw_create_cm_node( cm_node->vlan_id, I40IW_CM_LISTENER_ACTIVE_STATE); if (!loopback_remotelistener) { - i40iw_create_event(cm_node, I40IW_CM_EVENT_ABORTED); + i40iw_rem_ref_cm_node(cm_node); + return ERR_PTR(-ECONNREFUSED); } else { loopback_cm_info = *cm_info; loopback_cm_info.loc_port = cm_info->rem_port; @@ -2887,7 +2906,7 @@ static struct i40iw_cm_node *i40iw_create_cm_node( loopback_remotelistener); if (!loopback_remotenode) { i40iw_rem_ref_cm_node(cm_node); - return NULL; + return ERR_PTR(-ENOMEM); } cm_core->stats_loopbacks++; loopback_remotenode->loopbackpartner = cm_node; @@ -3041,10 +3060,10 @@ static int i40iw_cm_close(struct i40iw_cm_node *cm_node) /** * i40iw_receive_ilq - recv an ETHERNET packet, and process it * through CM - * @dev: FPK dev struct + * @vsi: pointer to the vsi structure * @rbuf: receive buffer */ -void i40iw_receive_ilq(struct i40iw_sc_dev *dev, struct i40iw_puda_buf *rbuf) +void i40iw_receive_ilq(struct i40iw_sc_vsi *vsi, struct i40iw_puda_buf *rbuf) { struct i40iw_cm_node *cm_node; struct i40iw_cm_listener *listener; @@ -3052,9 +3071,11 @@ void i40iw_receive_ilq(struct i40iw_sc_dev *dev, struct i40iw_puda_buf *rbuf) struct ipv6hdr *ip6h; struct tcphdr *tcph; struct i40iw_cm_info cm_info; + struct i40iw_sc_dev *dev = vsi->dev; struct i40iw_device *iwdev = (struct i40iw_device *)dev->back_dev; struct i40iw_cm_core *cm_core = &iwdev->cm_core; struct vlan_ethhdr *ethh; + u16 vtag; /* if vlan, then maclen = 18 else 14 */ iph = (struct iphdr *)rbuf->iph; @@ -3068,7 +3089,9 @@ void i40iw_receive_ilq(struct i40iw_sc_dev *dev, struct i40iw_puda_buf *rbuf) ethh = (struct vlan_ethhdr *)rbuf->mem.va; if (ethh->h_vlan_proto == htons(ETH_P_8021Q)) { - cm_info.vlan_id = ntohs(ethh->h_vlan_TCI) & VLAN_VID_MASK; + vtag = ntohs(ethh->h_vlan_TCI); + cm_info.user_pri = (vtag & VLAN_PRIO_MASK) >> VLAN_PRIO_SHIFT; + cm_info.vlan_id = vtag & VLAN_VID_MASK; i40iw_debug(cm_core->dev, I40IW_DEBUG_CM, "%s vlan_id=%d\n", @@ -3083,6 +3106,7 @@ void i40iw_receive_ilq(struct i40iw_sc_dev *dev, struct i40iw_puda_buf *rbuf) cm_info.loc_addr[0] = ntohl(iph->daddr); cm_info.rem_addr[0] = ntohl(iph->saddr); cm_info.ipv4 = true; + cm_info.tos = iph->tos; } else { ip6h = (struct ipv6hdr *)rbuf->iph; i40iw_copy_ip_ntohl(cm_info.loc_addr, @@ -3090,6 +3114,7 @@ void i40iw_receive_ilq(struct i40iw_sc_dev *dev, struct i40iw_puda_buf *rbuf) i40iw_copy_ip_ntohl(cm_info.rem_addr, ip6h->saddr.in6_u.u6_addr32); cm_info.ipv4 = false; + cm_info.tos = (ip6h->priority << 4) | (ip6h->flow_lbl[0] >> 4); } cm_info.loc_port = ntohs(tcph->dest); cm_info.rem_port = ntohs(tcph->source); @@ -3309,6 +3334,8 @@ static void i40iw_cm_init_tsa_conn(struct i40iw_qp *iwqp, ctx_info->tcp_info_valid = true; ctx_info->iwarp_info_valid = true; + ctx_info->add_to_qoslist = true; + ctx_info->user_pri = cm_node->user_pri; i40iw_init_tcp_ctx(cm_node, &tcp_info, iwqp); if (cm_node->snd_mark_en) { @@ -3320,33 +3347,47 @@ static void i40iw_cm_init_tsa_conn(struct i40iw_qp *iwqp, cm_node->state = I40IW_CM_STATE_OFFLOADED; tcp_info.tcp_state = I40IW_TCP_STATE_ESTABLISHED; tcp_info.src_mac_addr_idx = iwdev->mac_ip_table_idx; + tcp_info.tos = cm_node->tos; dev->iw_priv_qp_ops->qp_setctx(&iwqp->sc_qp, (u64 *)(iwqp->host_ctx.va), ctx_info); /* once tcp_info is set, no need to do it again */ ctx_info->tcp_info_valid = false; ctx_info->iwarp_info_valid = false; + ctx_info->add_to_qoslist = false; } /** * i40iw_cm_disconn - when a connection is being closed * @iwqp: associate qp for the connection */ -int i40iw_cm_disconn(struct i40iw_qp *iwqp) +void i40iw_cm_disconn(struct i40iw_qp *iwqp) { struct disconn_work *work; struct i40iw_device *iwdev = iwqp->iwdev; struct i40iw_cm_core *cm_core = &iwdev->cm_core; + unsigned long flags; work = kzalloc(sizeof(*work), GFP_ATOMIC); if (!work) - return -ENOMEM; /* Timer will clean up */ - + return; /* Timer will clean up */ + + spin_lock_irqsave(&iwdev->qptable_lock, flags); + if (!iwdev->qp_table[iwqp->ibqp.qp_num]) { + spin_unlock_irqrestore(&iwdev->qptable_lock, flags); + i40iw_debug(&iwdev->sc_dev, I40IW_DEBUG_CM, + "%s qp_id %d is already freed\n", + __func__, iwqp->ibqp.qp_num); + kfree(work); + return; + } i40iw_add_ref(&iwqp->ibqp); + spin_unlock_irqrestore(&iwdev->qptable_lock, flags); + work->iwqp = iwqp; INIT_WORK(&work->work, i40iw_disconnect_worker); queue_work(cm_core->disconn_wq, &work->work); - return 0; + return; } /** @@ -3432,7 +3473,7 @@ static void i40iw_cm_disconn_true(struct i40iw_qp *iwqp) *terminate-handler to issue cm_disconn which can re-free *a QP even after its refcnt=0. */ - del_timer(&iwqp->terminate_timer); + i40iw_terminate_del_timer(qp); if (!iwqp->flush_issued) { iwqp->flush_issued = 1; issue_flush = 1; @@ -3462,7 +3503,7 @@ static void i40iw_cm_disconn_true(struct i40iw_qp *iwqp) /* Flush the queues */ i40iw_flush_wqes(iwdev, iwqp); - if (qp->term_flags) { + if (qp->term_flags && iwqp->ibqp.event_handler) { ibevent.device = iwqp->ibqp.device; ibevent.event = (qp->eventtype == TERM_EVENT_QP_FATAL) ? IB_EVENT_QP_FATAL : IB_EVENT_QP_ACCESS_ERR; @@ -3571,7 +3612,7 @@ int i40iw_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) iwqp->cm_node = (void *)cm_node; cm_node->iwqp = iwqp; - buf_len = conn_param->private_data_len + I40IW_MAX_IETF_SIZE + MPA_ZERO_PAD_LEN; + buf_len = conn_param->private_data_len + I40IW_MAX_IETF_SIZE; status = i40iw_allocate_dma_mem(dev->hw, &iwqp->ietf_mem, buf_len, 1); @@ -3605,18 +3646,10 @@ int i40iw_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) iwqp->lsmm_mr = ibmr; if (iwqp->page) iwqp->sc_qp.qp_uk.sq_base = kmap(iwqp->page); - if (is_remote_ne020_or_chelsio(cm_node)) - dev->iw_priv_qp_ops->qp_send_lsmm( - &iwqp->sc_qp, + dev->iw_priv_qp_ops->qp_send_lsmm(&iwqp->sc_qp, iwqp->ietf_mem.va, (accept.size + conn_param->private_data_len), ibmr->lkey); - else - dev->iw_priv_qp_ops->qp_send_lsmm( - &iwqp->sc_qp, - iwqp->ietf_mem.va, - (accept.size + conn_param->private_data_len + MPA_ZERO_PAD_LEN), - ibmr->lkey); } else { if (iwqp->page) @@ -3714,6 +3747,7 @@ int i40iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) struct sockaddr_in6 *raddr6; bool qhash_set = false; int apbvt_set = 0; + int err = 0; enum i40iw_status_code status; ibqp = i40iw_get_qp(cm_id->device, conn_param->qpn); @@ -3759,6 +3793,10 @@ int i40iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) i40iw_netdev_vlan_ipv6(cm_info.loc_addr, &cm_info.vlan_id, NULL); } cm_info.cm_id = cm_id; + cm_info.tos = cm_id->tos; + cm_info.user_pri = rt_tos2priority(cm_id->tos); + i40iw_debug(&iwdev->sc_dev, I40IW_DEBUG_DCB, "%s TOS:[%d] UP:[%d]\n", + __func__, cm_id->tos, cm_info.user_pri); if ((cm_info.ipv4 && (laddr->sin_addr.s_addr != raddr->sin_addr.s_addr)) || (!cm_info.ipv4 && memcmp(laddr6->sin6_addr.in6_u.u6_addr32, raddr6->sin6_addr.in6_u.u6_addr32, @@ -3790,8 +3828,11 @@ int i40iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) conn_param->private_data_len, (void *)conn_param->private_data, &cm_info); - if (!cm_node) - goto err; + + if (IS_ERR(cm_node)) { + err = PTR_ERR(cm_node); + goto err_out; + } i40iw_record_ird_ord(cm_node, (u16)conn_param->ird, (u16)conn_param->ord); if (cm_node->send_rdma0_op == SEND_RDMA_READ_ZERO && @@ -3805,10 +3846,12 @@ int i40iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) iwqp->cm_id = cm_id; i40iw_add_ref(&iwqp->ibqp); - if (cm_node->state == I40IW_CM_STATE_SYN_SENT) { - if (i40iw_send_syn(cm_node, 0)) { + if (cm_node->state != I40IW_CM_STATE_OFFLOADED) { + cm_node->state = I40IW_CM_STATE_SYN_SENT; + err = i40iw_send_syn(cm_node, 0); + if (err) { i40iw_rem_ref_cm_node(cm_node); - goto err; + goto err_out; } } @@ -3820,24 +3863,25 @@ int i40iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) cm_node->cm_id); return 0; -err: - if (cm_node) { - if (cm_node->ipv4) - i40iw_debug(cm_node->dev, - I40IW_DEBUG_CM, - "Api - connect() FAILED: dest addr=%pI4", - cm_node->rem_addr); - else - i40iw_debug(cm_node->dev, I40IW_DEBUG_CM, - "Api - connect() FAILED: dest addr=%pI6", - cm_node->rem_addr); - } - i40iw_manage_qhash(iwdev, - &cm_info, - I40IW_QHASH_TYPE_TCP_ESTABLISHED, - I40IW_QHASH_MANAGE_TYPE_DELETE, - NULL, - false); +err_out: + if (cm_info.ipv4) + i40iw_debug(&iwdev->sc_dev, + I40IW_DEBUG_CM, + "Api - connect() FAILED: dest addr=%pI4", + cm_info.rem_addr); + else + i40iw_debug(&iwdev->sc_dev, + I40IW_DEBUG_CM, + "Api - connect() FAILED: dest addr=%pI6", + cm_info.rem_addr); + + if (qhash_set) + i40iw_manage_qhash(iwdev, + &cm_info, + I40IW_QHASH_TYPE_TCP_ESTABLISHED, + I40IW_QHASH_MANAGE_TYPE_DELETE, + NULL, + false); if (apbvt_set && !i40iw_listen_port_in_use(&iwdev->cm_core, cm_info.loc_port)) @@ -3846,7 +3890,7 @@ err: I40IW_MANAGE_APBVT_DEL); cm_id->rem_ref(cm_id); iwdev->cm_core.stats_connect_errs++; - return -ENOMEM; + return err; } /** @@ -3904,6 +3948,10 @@ int i40iw_create_listen(struct iw_cm_id *cm_id, int backlog) cm_id->provider_data = cm_listen_node; + cm_listen_node->tos = cm_id->tos; + cm_listen_node->user_pri = rt_tos2priority(cm_id->tos); + cm_info.user_pri = cm_listen_node->user_pri; + if (!cm_listen_node->reused_node) { if (wildcard) { if (cm_info.ipv4) @@ -4124,3 +4172,158 @@ static void i40iw_cm_post_event(struct i40iw_cm_event *event) queue_work(event->cm_node->cm_core->event_wq, &event->event_work); } + +/** + * i40iw_qhash_ctrl - enable/disable qhash for list + * @iwdev: device pointer + * @parent_listen_node: parent listen node + * @nfo: cm info node + * @ipaddr: Pointer to IPv4 or IPv6 address + * @ipv4: flag indicating IPv4 when true + * @ifup: flag indicating interface up when true + * + * Enables or disables the qhash for the node in the child + * listen list that matches ipaddr. If no matching IP was found + * it will allocate and add a new child listen node to the + * parent listen node. The listen_list_lock is assumed to be + * held when called. + */ +static void i40iw_qhash_ctrl(struct i40iw_device *iwdev, + struct i40iw_cm_listener *parent_listen_node, + struct i40iw_cm_info *nfo, + u32 *ipaddr, bool ipv4, bool ifup) +{ + struct list_head *child_listen_list = &parent_listen_node->child_listen_list; + struct i40iw_cm_listener *child_listen_node; + struct list_head *pos, *tpos; + enum i40iw_status_code ret; + bool node_allocated = false; + enum i40iw_quad_hash_manage_type op = + ifup ? I40IW_QHASH_MANAGE_TYPE_ADD : I40IW_QHASH_MANAGE_TYPE_DELETE; + + list_for_each_safe(pos, tpos, child_listen_list) { + child_listen_node = + list_entry(pos, + struct i40iw_cm_listener, + child_listen_list); + if (!memcmp(child_listen_node->loc_addr, ipaddr, ipv4 ? 4 : 16)) + goto set_qhash; + } + + /* if not found then add a child listener if interface is going up */ + if (!ifup) + return; + child_listen_node = kzalloc(sizeof(*child_listen_node), GFP_ATOMIC); + if (!child_listen_node) + return; + node_allocated = true; + memcpy(child_listen_node, parent_listen_node, sizeof(*child_listen_node)); + + memcpy(child_listen_node->loc_addr, ipaddr, ipv4 ? 4 : 16); + +set_qhash: + memcpy(nfo->loc_addr, + child_listen_node->loc_addr, + sizeof(nfo->loc_addr)); + nfo->vlan_id = child_listen_node->vlan_id; + ret = i40iw_manage_qhash(iwdev, nfo, + I40IW_QHASH_TYPE_TCP_SYN, + op, + NULL, false); + if (!ret) { + child_listen_node->qhash_set = ifup; + if (node_allocated) + list_add(&child_listen_node->child_listen_list, + &parent_listen_node->child_listen_list); + } else if (node_allocated) { + kfree(child_listen_node); + } +} + +/** + * i40iw_cm_disconnect_all - disconnect all connected qp's + * @iwdev: device pointer + */ +void i40iw_cm_disconnect_all(struct i40iw_device *iwdev) +{ + struct i40iw_cm_core *cm_core = &iwdev->cm_core; + struct list_head *list_core_temp; + struct list_head *list_node; + struct i40iw_cm_node *cm_node; + unsigned long flags; + struct list_head connected_list; + struct ib_qp_attr attr; + + INIT_LIST_HEAD(&connected_list); + spin_lock_irqsave(&cm_core->ht_lock, flags); + list_for_each_safe(list_node, list_core_temp, &cm_core->connected_nodes) { + cm_node = container_of(list_node, struct i40iw_cm_node, list); + atomic_inc(&cm_node->ref_count); + list_add(&cm_node->connected_entry, &connected_list); + } + spin_unlock_irqrestore(&cm_core->ht_lock, flags); + + list_for_each_safe(list_node, list_core_temp, &connected_list) { + cm_node = container_of(list_node, struct i40iw_cm_node, connected_entry); + attr.qp_state = IB_QPS_ERR; + i40iw_modify_qp(&cm_node->iwqp->ibqp, &attr, IB_QP_STATE, NULL); + i40iw_rem_ref_cm_node(cm_node); + } +} + +/** + * i40iw_ifdown_notify - process an ifdown on an interface + * @iwdev: device pointer + * @ipaddr: Pointer to IPv4 or IPv6 address + * @ipv4: flag indicating IPv4 when true + * @ifup: flag indicating interface up when true + */ +void i40iw_if_notify(struct i40iw_device *iwdev, struct net_device *netdev, + u32 *ipaddr, bool ipv4, bool ifup) +{ + struct i40iw_cm_core *cm_core = &iwdev->cm_core; + unsigned long flags; + struct i40iw_cm_listener *listen_node; + static const u32 ip_zero[4] = { 0, 0, 0, 0 }; + struct i40iw_cm_info nfo; + u16 vlan_id = rdma_vlan_dev_vlan_id(netdev); + enum i40iw_status_code ret; + enum i40iw_quad_hash_manage_type op = + ifup ? I40IW_QHASH_MANAGE_TYPE_ADD : I40IW_QHASH_MANAGE_TYPE_DELETE; + + /* Disable or enable qhash for listeners */ + spin_lock_irqsave(&cm_core->listen_list_lock, flags); + list_for_each_entry(listen_node, &cm_core->listen_nodes, list) { + if (vlan_id == listen_node->vlan_id && + (!memcmp(listen_node->loc_addr, ipaddr, ipv4 ? 4 : 16) || + !memcmp(listen_node->loc_addr, ip_zero, ipv4 ? 4 : 16))) { + memcpy(nfo.loc_addr, listen_node->loc_addr, + sizeof(nfo.loc_addr)); + nfo.loc_port = listen_node->loc_port; + nfo.ipv4 = listen_node->ipv4; + nfo.vlan_id = listen_node->vlan_id; + nfo.user_pri = listen_node->user_pri; + if (!list_empty(&listen_node->child_listen_list)) { + i40iw_qhash_ctrl(iwdev, + listen_node, + &nfo, + ipaddr, ipv4, ifup); + } else if (memcmp(listen_node->loc_addr, ip_zero, + ipv4 ? 4 : 16)) { + ret = i40iw_manage_qhash(iwdev, + &nfo, + I40IW_QHASH_TYPE_TCP_SYN, + op, + NULL, + false); + if (!ret) + listen_node->qhash_set = ifup; + } + } + } + spin_unlock_irqrestore(&cm_core->listen_list_lock, flags); + + /* disconnect any connected qp's on ifdown */ + if (!ifup) + i40iw_cm_disconnect_all(iwdev); +} diff --git a/drivers/infiniband/hw/i40iw/i40iw_cm.h b/drivers/infiniband/hw/i40iw/i40iw_cm.h index e9046d9f9645..2e52e38ffcf3 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_cm.h +++ b/drivers/infiniband/hw/i40iw/i40iw_cm.h @@ -56,8 +56,6 @@ #define I40IW_MAX_IETF_SIZE 32 -#define MPA_ZERO_PAD_LEN 4 - /* IETF RTR MSG Fields */ #define IETF_PEER_TO_PEER 0x8000 #define IETF_FLPDU_ZERO_LEN 0x4000 @@ -299,6 +297,7 @@ struct i40iw_cm_listener { enum i40iw_cm_listener_state listener_state; u32 reused_node; u8 user_pri; + u8 tos; u16 vlan_id; bool qhash_set; bool ipv4; @@ -341,9 +340,11 @@ struct i40iw_cm_node { int accept_pend; struct list_head timer_entry; struct list_head reset_entry; + struct list_head connected_entry; atomic_t passive_state; bool qhash_set; u8 user_pri; + u8 tos; bool ipv4; bool snd_mark_en; u16 lsmm_size; @@ -368,7 +369,8 @@ struct i40iw_cm_info { u32 rem_addr[4]; u16 vlan_id; int backlog; - u16 user_pri; + u8 user_pri; + u8 tos; bool ipv4; }; @@ -445,4 +447,7 @@ int i40iw_arp_table(struct i40iw_device *iwdev, u8 *mac_addr, u32 action); +void i40iw_if_notify(struct i40iw_device *iwdev, struct net_device *netdev, + u32 *ipaddr, bool ipv4, bool ifup); +void i40iw_cm_disconnect_all(struct i40iw_device *iwdev); #endif /* I40IW_CM_H */ diff --git a/drivers/infiniband/hw/i40iw/i40iw_ctrl.c b/drivers/infiniband/hw/i40iw/i40iw_ctrl.c index 2c4b4d072d6a..f82483b3d1e7 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_ctrl.c +++ b/drivers/infiniband/hw/i40iw/i40iw_ctrl.c @@ -103,6 +103,7 @@ static enum i40iw_status_code i40iw_cqp_poll_registers( if (newtail != tail) { /* SUCCESS */ I40IW_RING_MOVE_TAIL(cqp->sq_ring); + cqp->dev->cqp_cmd_stats[OP_COMPLETED_COMMANDS]++; return 0; } udelay(I40IW_SLEEP_COUNT); @@ -223,17 +224,150 @@ static enum i40iw_status_code i40iw_sc_parse_fpm_query_buf( } /** + * i40iw_fill_qos_list - Change all unknown qs handles to available ones + * @qs_list: list of qs_handles to be fixed with valid qs_handles + */ +static void i40iw_fill_qos_list(u16 *qs_list) +{ + u16 qshandle = qs_list[0]; + int i; + + for (i = 0; i < I40IW_MAX_USER_PRIORITY; i++) { + if (qs_list[i] == QS_HANDLE_UNKNOWN) + qs_list[i] = qshandle; + else + qshandle = qs_list[i]; + } +} + +/** + * i40iw_qp_from_entry - Given entry, get to the qp structure + * @entry: Points to list of qp structure + */ +static struct i40iw_sc_qp *i40iw_qp_from_entry(struct list_head *entry) +{ + if (!entry) + return NULL; + + return (struct i40iw_sc_qp *)((char *)entry - offsetof(struct i40iw_sc_qp, list)); +} + +/** + * i40iw_get_qp - get the next qp from the list given current qp + * @head: Listhead of qp's + * @qp: current qp + */ +static struct i40iw_sc_qp *i40iw_get_qp(struct list_head *head, struct i40iw_sc_qp *qp) +{ + struct list_head *entry = NULL; + struct list_head *lastentry; + + if (list_empty(head)) + return NULL; + + if (!qp) { + entry = head->next; + } else { + lastentry = &qp->list; + entry = (lastentry != head) ? lastentry->next : NULL; + } + + return i40iw_qp_from_entry(entry); +} + +/** + * i40iw_change_l2params - given the new l2 parameters, change all qp + * @vsi: pointer to the vsi structure + * @l2params: New paramaters from l2 + */ +void i40iw_change_l2params(struct i40iw_sc_vsi *vsi, struct i40iw_l2params *l2params) +{ + struct i40iw_sc_dev *dev = vsi->dev; + struct i40iw_sc_qp *qp = NULL; + bool qs_handle_change = false; + bool mss_change = false; + unsigned long flags; + u16 qs_handle; + int i; + + if (vsi->mss != l2params->mss) { + mss_change = true; + vsi->mss = l2params->mss; + } + + i40iw_fill_qos_list(l2params->qs_handle_list); + for (i = 0; i < I40IW_MAX_USER_PRIORITY; i++) { + qs_handle = l2params->qs_handle_list[i]; + if (vsi->qos[i].qs_handle != qs_handle) + qs_handle_change = true; + else if (!mss_change) + continue; /* no MSS nor qs handle change */ + spin_lock_irqsave(&vsi->qos[i].lock, flags); + qp = i40iw_get_qp(&vsi->qos[i].qplist, qp); + while (qp) { + if (mss_change) + i40iw_qp_mss_modify(dev, qp); + if (qs_handle_change) { + qp->qs_handle = qs_handle; + /* issue cqp suspend command */ + i40iw_qp_suspend_resume(dev, qp, true); + } + qp = i40iw_get_qp(&vsi->qos[i].qplist, qp); + } + spin_unlock_irqrestore(&vsi->qos[i].lock, flags); + vsi->qos[i].qs_handle = qs_handle; + } +} + +/** + * i40iw_qp_rem_qos - remove qp from qos lists during destroy qp + * @qp: qp to be removed from qos + */ +static void i40iw_qp_rem_qos(struct i40iw_sc_qp *qp) +{ + struct i40iw_sc_vsi *vsi = qp->vsi; + unsigned long flags; + + if (!qp->on_qoslist) + return; + spin_lock_irqsave(&vsi->qos[qp->user_pri].lock, flags); + list_del(&qp->list); + spin_unlock_irqrestore(&vsi->qos[qp->user_pri].lock, flags); +} + +/** + * i40iw_qp_add_qos - called during setctx fot qp to be added to qos + * @qp: qp to be added to qos + */ +void i40iw_qp_add_qos(struct i40iw_sc_qp *qp) +{ + struct i40iw_sc_vsi *vsi = qp->vsi; + unsigned long flags; + + if (qp->on_qoslist) + return; + spin_lock_irqsave(&vsi->qos[qp->user_pri].lock, flags); + qp->qs_handle = vsi->qos[qp->user_pri].qs_handle; + list_add(&qp->list, &vsi->qos[qp->user_pri].qplist); + qp->on_qoslist = true; + spin_unlock_irqrestore(&vsi->qos[qp->user_pri].lock, flags); +} + +/** * i40iw_sc_pd_init - initialize sc pd struct * @dev: sc device struct * @pd: sc pd ptr * @pd_id: pd_id for allocated pd + * @abi_ver: ABI version from user context, -1 if not valid */ static void i40iw_sc_pd_init(struct i40iw_sc_dev *dev, struct i40iw_sc_pd *pd, - u16 pd_id) + u16 pd_id, + int abi_ver) { pd->size = sizeof(*pd); pd->pd_id = pd_id; + pd->abi_ver = abi_ver; pd->dev = dev; } @@ -292,6 +426,9 @@ static enum i40iw_status_code i40iw_sc_cqp_init(struct i40iw_sc_cqp *cqp, info->dev->cqp = cqp; I40IW_RING_INIT(cqp->sq_ring, cqp->sq_size); + cqp->dev->cqp_cmd_stats[OP_REQUESTED_COMMANDS] = 0; + cqp->dev->cqp_cmd_stats[OP_COMPLETED_COMMANDS] = 0; + i40iw_debug(cqp->dev, I40IW_DEBUG_WQE, "%s: sq_size[%04d] hw_sq_size[%04d] sq_base[%p] sq_pa[%llxh] cqp[%p] polarity[x%04X]\n", __func__, cqp->sq_size, cqp->hw_sq_size, @@ -302,12 +439,10 @@ static enum i40iw_status_code i40iw_sc_cqp_init(struct i40iw_sc_cqp *cqp, /** * i40iw_sc_cqp_create - create cqp during bringup * @cqp: struct for cqp hw - * @disable_pfpdus: if pfpdu to be disabled * @maj_err: If error, major err number * @min_err: If error, minor err number */ static enum i40iw_status_code i40iw_sc_cqp_create(struct i40iw_sc_cqp *cqp, - bool disable_pfpdus, u16 *maj_err, u16 *min_err) { @@ -315,6 +450,9 @@ static enum i40iw_status_code i40iw_sc_cqp_create(struct i40iw_sc_cqp *cqp, u32 cnt = 0, p1, p2, val = 0, err_code; enum i40iw_status_code ret_code; + *maj_err = 0; + *min_err = 0; + ret_code = i40iw_allocate_dma_mem(cqp->dev->hw, &cqp->sdbuf, 128, @@ -326,9 +464,6 @@ static enum i40iw_status_code i40iw_sc_cqp_create(struct i40iw_sc_cqp *cqp, temp = LS_64(cqp->hw_sq_size, I40IW_CQPHC_SQSIZE) | LS_64(cqp->struct_ver, I40IW_CQPHC_SVER); - if (disable_pfpdus) - temp |= LS_64(1, I40IW_CQPHC_DISABLE_PFPDUS); - set_64bit_val(cqp->host_ctx, 0, temp); set_64bit_val(cqp->host_ctx, 8, cqp->sq_pa); temp = LS_64(cqp->enabled_vf_count, I40IW_CQPHC_ENABLED_VFS) | @@ -424,6 +559,7 @@ u64 *i40iw_sc_cqp_get_next_send_wqe(struct i40iw_sc_cqp *cqp, u64 scratch) return NULL; } I40IW_ATOMIC_RING_MOVE_HEAD(cqp->sq_ring, wqe_idx, ret_code); + cqp->dev->cqp_cmd_stats[OP_REQUESTED_COMMANDS]++; if (ret_code) return NULL; if (!wqe_idx) @@ -559,6 +695,8 @@ static enum i40iw_status_code i40iw_sc_ccq_get_cqe_info( I40IW_RING_GETCURRENT_HEAD(ccq->cq_uk.cq_ring)); wmb(); /* write shadow area before tail */ I40IW_RING_MOVE_TAIL(cqp->sq_ring); + ccq->dev->cqp_cmd_stats[OP_COMPLETED_COMMANDS]++; + return ret_code; } @@ -1051,6 +1189,7 @@ static enum i40iw_status_code i40iw_sc_manage_qhash_table_entry( u64 qw1 = 0; u64 qw2 = 0; u64 temp; + struct i40iw_sc_vsi *vsi = info->vsi; wqe = i40iw_sc_cqp_get_next_send_wqe(cqp, scratch); if (!wqe) @@ -1082,7 +1221,7 @@ static enum i40iw_status_code i40iw_sc_manage_qhash_table_entry( LS_64(info->dest_ip[2], I40IW_CQPSQ_QHASH_ADDR2) | LS_64(info->dest_ip[3], I40IW_CQPSQ_QHASH_ADDR3)); } - qw2 = LS_64(cqp->dev->qs_handle, I40IW_CQPSQ_QHASH_QS_HANDLE); + qw2 = LS_64(vsi->qos[info->user_pri].qs_handle, I40IW_CQPSQ_QHASH_QS_HANDLE); if (info->vlan_valid) qw2 |= LS_64(info->vlan_id, I40IW_CQPSQ_QHASH_VLANID); set_64bit_val(wqe, 16, qw2); @@ -2103,6 +2242,7 @@ static enum i40iw_status_code i40iw_sc_qp_init(struct i40iw_sc_qp *qp, u32 offset; qp->dev = info->pd->dev; + qp->vsi = info->vsi; qp->sq_pa = info->sq_pa; qp->rq_pa = info->rq_pa; qp->hw_host_ctx_pa = info->host_ctx_pa; @@ -2118,6 +2258,7 @@ static enum i40iw_status_code i40iw_sc_qp_init(struct i40iw_sc_qp *qp, offset); info->qp_uk_init_info.wqe_alloc_reg = wqe_alloc_reg; + info->qp_uk_init_info.abi_ver = qp->pd->abi_ver; ret_code = i40iw_qp_uk_init(&qp->qp_uk, &info->qp_uk_init_info); if (ret_code) return ret_code; @@ -2136,10 +2277,21 @@ static enum i40iw_status_code i40iw_sc_qp_init(struct i40iw_sc_qp *qp, false); i40iw_debug(qp->dev, I40IW_DEBUG_WQE, "%s: hw_sq_size[%04d] sq_ring.size[%04d]\n", __func__, qp->hw_sq_size, qp->qp_uk.sq_ring.size); - ret_code = i40iw_fragcnt_to_wqesize_rq(qp->qp_uk.max_rq_frag_cnt, - &wqe_size); - if (ret_code) - return ret_code; + + switch (qp->pd->abi_ver) { + case 4: + ret_code = i40iw_fragcnt_to_wqesize_rq(qp->qp_uk.max_rq_frag_cnt, + &wqe_size); + if (ret_code) + return ret_code; + break; + case 5: /* fallthrough until next ABI version */ + default: + if (qp->qp_uk.max_rq_frag_cnt > I40IW_MAX_WQ_FRAGMENT_COUNT) + return I40IW_ERR_INVALID_FRAG_COUNT; + wqe_size = I40IW_MAX_WQE_SIZE_RQ; + break; + } qp->hw_rq_size = i40iw_get_encoded_wqe_size(qp->qp_uk.rq_size * (wqe_size / I40IW_QP_WQE_MIN_SIZE), false); i40iw_debug(qp->dev, I40IW_DEBUG_WQE, @@ -2151,7 +2303,7 @@ static enum i40iw_status_code i40iw_sc_qp_init(struct i40iw_sc_qp *qp, qp->rq_tph_en = info->rq_tph_en; qp->rcv_tph_en = info->rcv_tph_en; qp->xmit_tph_en = info->xmit_tph_en; - qp->qs_handle = qp->pd->dev->qs_handle; + qp->qs_handle = qp->vsi->qos[qp->user_pri].qs_handle; qp->exception_lan_queue = qp->pd->dev->exception_lan_queue; return 0; @@ -2296,6 +2448,7 @@ static enum i40iw_status_code i40iw_sc_qp_destroy( struct i40iw_sc_cqp *cqp; u64 header; + i40iw_qp_rem_qos(qp); cqp = qp->pd->dev->cqp; wqe = i40iw_sc_cqp_get_next_send_wqe(cqp, scratch); if (!wqe) @@ -2443,10 +2596,20 @@ static enum i40iw_status_code i40iw_sc_qp_setctx( { struct i40iwarp_offload_info *iw; struct i40iw_tcp_offload_info *tcp; + struct i40iw_sc_vsi *vsi; + struct i40iw_sc_dev *dev; u64 qw0, qw3, qw7 = 0; iw = info->iwarp_info; tcp = info->tcp_info; + vsi = qp->vsi; + dev = qp->dev; + if (info->add_to_qoslist) { + qp->user_pri = info->user_pri; + i40iw_qp_add_qos(qp); + i40iw_debug(qp->dev, I40IW_DEBUG_DCB, "%s qp[%d] UP[%d] qset[%d]\n", + __func__, qp->qp_uk.qp_id, qp->user_pri, qp->qs_handle); + } qw0 = LS_64(qp->qp_uk.rq_wqe_size, I40IWQPC_RQWQESIZE) | LS_64(info->err_rq_idx_valid, I40IWQPC_ERR_RQ_IDX_VALID) | LS_64(qp->rcv_tph_en, I40IWQPC_RCVTPHEN) | @@ -2487,16 +2650,14 @@ static enum i40iw_status_code i40iw_sc_qp_setctx( LS_64(iw->rdmap_ver, I40IWQPC_RDMAP_VER); qw7 |= LS_64(iw->pd_id, I40IWQPC_PDIDX); - set_64bit_val(qp_ctx, 144, qp->q2_pa); + set_64bit_val(qp_ctx, + 144, + LS_64(qp->q2_pa, I40IWQPC_Q2ADDR) | + LS_64(vsi->fcn_id, I40IWQPC_STAT_INDEX)); set_64bit_val(qp_ctx, 152, LS_64(iw->last_byte_sent, I40IWQPC_LASTBYTESENT)); - /* - * Hard-code IRD_SIZE to hw-limit, 128, in qpctx, i.e matching an - *advertisable IRD of 64 - */ - iw->ird_size = I40IW_QPCTX_ENCD_MAXIRD; set_64bit_val(qp_ctx, 160, LS_64(iw->ord_size, I40IWQPC_ORDSIZE) | @@ -2507,6 +2668,9 @@ static enum i40iw_status_code i40iw_sc_qp_setctx( LS_64(iw->bind_en, I40IWQPC_BINDEN) | LS_64(iw->fast_reg_en, I40IWQPC_FASTREGEN) | LS_64(iw->priv_mode_en, I40IWQPC_PRIVEN) | + LS_64((((vsi->stats_fcn_id_alloc) && + (dev->is_pf) && (vsi->fcn_id >= I40IW_FIRST_NON_PF_STAT)) ? 1 : 0), + I40IWQPC_USESTATSINSTANCE) | LS_64(1, I40IWQPC_IWARPMODE) | LS_64(iw->rcv_mark_en, I40IWQPC_RCVMARKERS) | LS_64(iw->align_hdrs, I40IWQPC_ALIGNHDRS) | @@ -2623,7 +2787,9 @@ static enum i40iw_status_code i40iw_sc_alloc_stag( u64 *wqe; struct i40iw_sc_cqp *cqp; u64 header; + enum i40iw_page_size page_size; + page_size = (info->page_size == 0x200000) ? I40IW_PAGE_SIZE_2M : I40IW_PAGE_SIZE_4K; cqp = dev->cqp; wqe = i40iw_sc_cqp_get_next_send_wqe(cqp, scratch); if (!wqe) @@ -2643,7 +2809,7 @@ static enum i40iw_status_code i40iw_sc_alloc_stag( LS_64(1, I40IW_CQPSQ_STAG_MR) | LS_64(info->access_rights, I40IW_CQPSQ_STAG_ARIGHTS) | LS_64(info->chunk_size, I40IW_CQPSQ_STAG_LPBLSIZE) | - LS_64(info->page_size, I40IW_CQPSQ_STAG_HPAGESIZE) | + LS_64(page_size, I40IW_CQPSQ_STAG_HPAGESIZE) | LS_64(info->remote_access, I40IW_CQPSQ_STAG_REMACCENABLED) | LS_64(info->use_hmc_fcn_index, I40IW_CQPSQ_STAG_USEHMCFNIDX) | LS_64(info->use_pf_rid, I40IW_CQPSQ_STAG_USEPFRID) | @@ -2679,7 +2845,9 @@ static enum i40iw_status_code i40iw_sc_mr_reg_non_shared( u32 pble_obj_cnt; bool remote_access; u8 addr_type; + enum i40iw_page_size page_size; + page_size = (info->page_size == 0x200000) ? I40IW_PAGE_SIZE_2M : I40IW_PAGE_SIZE_4K; if (info->access_rights & (I40IW_ACCESS_FLAGS_REMOTEREAD_ONLY | I40IW_ACCESS_FLAGS_REMOTEWRITE_ONLY)) remote_access = true; @@ -2722,7 +2890,7 @@ static enum i40iw_status_code i40iw_sc_mr_reg_non_shared( header = LS_64(I40IW_CQP_OP_REG_MR, I40IW_CQPSQ_OPCODE) | LS_64(1, I40IW_CQPSQ_STAG_MR) | LS_64(info->chunk_size, I40IW_CQPSQ_STAG_LPBLSIZE) | - LS_64(info->page_size, I40IW_CQPSQ_STAG_HPAGESIZE) | + LS_64(page_size, I40IW_CQPSQ_STAG_HPAGESIZE) | LS_64(info->access_rights, I40IW_CQPSQ_STAG_ARIGHTS) | LS_64(remote_access, I40IW_CQPSQ_STAG_REMACCENABLED) | LS_64(addr_type, I40IW_CQPSQ_STAG_VABASEDTO) | @@ -2937,7 +3105,9 @@ enum i40iw_status_code i40iw_sc_mr_fast_register( u64 temp, header; u64 *wqe; u32 wqe_idx; + enum i40iw_page_size page_size; + page_size = (info->page_size == 0x200000) ? I40IW_PAGE_SIZE_2M : I40IW_PAGE_SIZE_4K; wqe = i40iw_qp_get_next_send_wqe(&qp->qp_uk, &wqe_idx, I40IW_QP_WQE_MIN_SIZE, 0, info->wr_id); if (!wqe) @@ -2964,7 +3134,7 @@ enum i40iw_status_code i40iw_sc_mr_fast_register( LS_64(info->stag_idx, I40IWQPSQ_STAGINDEX) | LS_64(I40IWQP_OP_FAST_REGISTER, I40IWQPSQ_OPCODE) | LS_64(info->chunk_size, I40IWQPSQ_LPBLSIZE) | - LS_64(info->page_size, I40IWQPSQ_HPAGESIZE) | + LS_64(page_size, I40IWQPSQ_HPAGESIZE) | LS_64(info->access_rights, I40IWQPSQ_STAGRIGHTS) | LS_64(info->addr_type, I40IWQPSQ_VABASEDTO) | LS_64(info->read_fence, I40IWQPSQ_READFENCE) | @@ -3959,7 +4129,7 @@ enum i40iw_status_code i40iw_process_cqp_cmd(struct i40iw_sc_dev *dev, struct cqp_commands_info *pcmdinfo) { enum i40iw_status_code status = 0; - unsigned long flags; + unsigned long flags; spin_lock_irqsave(&dev->cqp_lock, flags); if (list_empty(&dev->cqp_cmd_head) && !i40iw_ring_full(dev->cqp)) @@ -3978,7 +4148,7 @@ enum i40iw_status_code i40iw_process_bh(struct i40iw_sc_dev *dev) { enum i40iw_status_code status = 0; struct cqp_commands_info *pcmdinfo; - unsigned long flags; + unsigned long flags; spin_lock_irqsave(&dev->cqp_lock, flags); while (!list_empty(&dev->cqp_cmd_head) && !i40iw_ring_full(dev->cqp)) { @@ -4055,7 +4225,6 @@ static int i40iw_bld_terminate_hdr(struct i40iw_sc_qp *qp, u16 ddp_seg_len; int copy_len = 0; u8 is_tagged = 0; - enum i40iw_flush_opcode flush_code = FLUSH_INVALID; u32 opcode; struct i40iw_terminate_hdr *termhdr; @@ -4228,9 +4397,6 @@ static int i40iw_bld_terminate_hdr(struct i40iw_sc_qp *qp, if (copy_len) memcpy(termhdr + 1, pkt, copy_len); - if (flush_code && !info->in_rdrsp_wr) - qp->sq_flush = (info->sq) ? true : false; - return sizeof(struct i40iw_terminate_hdr) + copy_len; } @@ -4321,329 +4487,413 @@ void i40iw_terminate_received(struct i40iw_sc_qp *qp, struct i40iw_aeqe_info *in } /** - * i40iw_hw_stat_init - Initiliaze HW stats table - * @devstat: pestat struct + * i40iw_sc_vsi_init - Initialize virtual device + * @vsi: pointer to the vsi structure + * @info: parameters to initialize vsi + **/ +void i40iw_sc_vsi_init(struct i40iw_sc_vsi *vsi, struct i40iw_vsi_init_info *info) +{ + int i; + + vsi->dev = info->dev; + vsi->back_vsi = info->back_vsi; + vsi->mss = info->params->mss; + i40iw_fill_qos_list(info->params->qs_handle_list); + + for (i = 0; i < I40IW_MAX_USER_PRIORITY; i++) { + vsi->qos[i].qs_handle = info->params->qs_handle_list[i]; + i40iw_debug(vsi->dev, I40IW_DEBUG_DCB, "qset[%d]: %d\n", i, + vsi->qos[i].qs_handle); + spin_lock_init(&vsi->qos[i].lock); + INIT_LIST_HEAD(&vsi->qos[i].qplist); + } +} + +/** + * i40iw_hw_stats_init - Initiliaze HW stats table + * @stats: pestat struct * @fcn_idx: PCI fn id - * @hw: PF i40iw_hw structure. * @is_pf: Is it a PF? * - * Populate the HW stat table with register offset addr for each - * stat. And start the perioidic stats timer. + * Populate the HW stats table with register offset addr for each + * stats. And start the perioidic stats timer. */ -static void i40iw_hw_stat_init(struct i40iw_dev_pestat *devstat, - u8 fcn_idx, - struct i40iw_hw *hw, bool is_pf) +void i40iw_hw_stats_init(struct i40iw_vsi_pestat *stats, u8 fcn_idx, bool is_pf) { - u32 stat_reg_offset; - u32 stat_index; - struct i40iw_dev_hw_stat_offsets *stat_table = - &devstat->hw_stat_offsets; - struct i40iw_dev_hw_stats *last_rd_stats = &devstat->last_read_hw_stats; - - devstat->hw = hw; + u32 stats_reg_offset; + u32 stats_index; + struct i40iw_dev_hw_stats_offsets *stats_table = + &stats->hw_stats_offsets; + struct i40iw_dev_hw_stats *last_rd_stats = &stats->last_read_hw_stats; if (is_pf) { - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP4RXDISCARD] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP4RXDISCARD] = I40E_GLPES_PFIP4RXDISCARD(fcn_idx); - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP4RXTRUNC] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP4RXTRUNC] = I40E_GLPES_PFIP4RXTRUNC(fcn_idx); - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP4TXNOROUTE] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP4TXNOROUTE] = I40E_GLPES_PFIP4TXNOROUTE(fcn_idx); - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP6RXDISCARD] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP6RXDISCARD] = I40E_GLPES_PFIP6RXDISCARD(fcn_idx); - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP6RXTRUNC] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP6RXTRUNC] = I40E_GLPES_PFIP6RXTRUNC(fcn_idx); - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP6TXNOROUTE] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP6TXNOROUTE] = I40E_GLPES_PFIP6TXNOROUTE(fcn_idx); - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_TCPRTXSEG] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_TCPRTXSEG] = I40E_GLPES_PFTCPRTXSEG(fcn_idx); - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_TCPRXOPTERR] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_TCPRXOPTERR] = I40E_GLPES_PFTCPRXOPTERR(fcn_idx); - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_TCPRXPROTOERR] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_TCPRXPROTOERR] = I40E_GLPES_PFTCPRXPROTOERR(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4RXOCTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4RXOCTS] = I40E_GLPES_PFIP4RXOCTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4RXPKTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4RXPKTS] = I40E_GLPES_PFIP4RXPKTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4RXFRAGS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4RXFRAGS] = I40E_GLPES_PFIP4RXFRAGSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4RXMCPKTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4RXMCPKTS] = I40E_GLPES_PFIP4RXMCPKTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4TXOCTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4TXOCTS] = I40E_GLPES_PFIP4TXOCTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4TXPKTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4TXPKTS] = I40E_GLPES_PFIP4TXPKTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4TXFRAGS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4TXFRAGS] = I40E_GLPES_PFIP4TXFRAGSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4TXMCPKTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4TXMCPKTS] = I40E_GLPES_PFIP4TXMCPKTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6RXOCTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6RXOCTS] = I40E_GLPES_PFIP6RXOCTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6RXPKTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6RXPKTS] = I40E_GLPES_PFIP6RXPKTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6RXFRAGS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6RXFRAGS] = I40E_GLPES_PFIP6RXFRAGSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6RXMCPKTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6RXMCPKTS] = I40E_GLPES_PFIP6RXMCPKTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6TXOCTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6TXOCTS] = I40E_GLPES_PFIP6TXOCTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6TXPKTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6TXPKTS] = I40E_GLPES_PFIP6TXPKTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6TXPKTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6TXPKTS] = I40E_GLPES_PFIP6TXPKTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6TXFRAGS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6TXFRAGS] = I40E_GLPES_PFIP6TXFRAGSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_TCPRXSEGS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_TCPRXSEGS] = I40E_GLPES_PFTCPRXSEGSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_TCPTXSEG] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_TCPTXSEG] = I40E_GLPES_PFTCPTXSEGLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMARXRDS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMARXRDS] = I40E_GLPES_PFRDMARXRDSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMARXSNDS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMARXSNDS] = I40E_GLPES_PFRDMARXSNDSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMARXWRS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMARXWRS] = I40E_GLPES_PFRDMARXWRSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMATXRDS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMATXRDS] = I40E_GLPES_PFRDMATXRDSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMATXSNDS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMATXSNDS] = I40E_GLPES_PFRDMATXSNDSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMATXWRS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMATXWRS] = I40E_GLPES_PFRDMATXWRSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMAVBND] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMAVBND] = I40E_GLPES_PFRDMAVBNDLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMAVINV] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMAVINV] = I40E_GLPES_PFRDMAVINVLO(fcn_idx); } else { - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP4RXDISCARD] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP4RXDISCARD] = I40E_GLPES_VFIP4RXDISCARD(fcn_idx); - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP4RXTRUNC] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP4RXTRUNC] = I40E_GLPES_VFIP4RXTRUNC(fcn_idx); - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP4TXNOROUTE] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP4TXNOROUTE] = I40E_GLPES_VFIP4TXNOROUTE(fcn_idx); - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP6RXDISCARD] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP6RXDISCARD] = I40E_GLPES_VFIP6RXDISCARD(fcn_idx); - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP6RXTRUNC] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP6RXTRUNC] = I40E_GLPES_VFIP6RXTRUNC(fcn_idx); - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_IP6TXNOROUTE] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_IP6TXNOROUTE] = I40E_GLPES_VFIP6TXNOROUTE(fcn_idx); - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_TCPRTXSEG] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_TCPRTXSEG] = I40E_GLPES_VFTCPRTXSEG(fcn_idx); - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_TCPRXOPTERR] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_TCPRXOPTERR] = I40E_GLPES_VFTCPRXOPTERR(fcn_idx); - stat_table->stat_offset_32[I40IW_HW_STAT_INDEX_TCPRXPROTOERR] = + stats_table->stats_offset_32[I40IW_HW_STAT_INDEX_TCPRXPROTOERR] = I40E_GLPES_VFTCPRXPROTOERR(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4RXOCTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4RXOCTS] = I40E_GLPES_VFIP4RXOCTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4RXPKTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4RXPKTS] = I40E_GLPES_VFIP4RXPKTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4RXFRAGS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4RXFRAGS] = I40E_GLPES_VFIP4RXFRAGSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4RXMCPKTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4RXMCPKTS] = I40E_GLPES_VFIP4RXMCPKTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4TXOCTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4TXOCTS] = I40E_GLPES_VFIP4TXOCTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4TXPKTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4TXPKTS] = I40E_GLPES_VFIP4TXPKTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4TXFRAGS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4TXFRAGS] = I40E_GLPES_VFIP4TXFRAGSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP4TXMCPKTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP4TXMCPKTS] = I40E_GLPES_VFIP4TXMCPKTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6RXOCTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6RXOCTS] = I40E_GLPES_VFIP6RXOCTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6RXPKTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6RXPKTS] = I40E_GLPES_VFIP6RXPKTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6RXFRAGS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6RXFRAGS] = I40E_GLPES_VFIP6RXFRAGSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6RXMCPKTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6RXMCPKTS] = I40E_GLPES_VFIP6RXMCPKTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6TXOCTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6TXOCTS] = I40E_GLPES_VFIP6TXOCTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6TXPKTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6TXPKTS] = I40E_GLPES_VFIP6TXPKTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6TXPKTS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6TXPKTS] = I40E_GLPES_VFIP6TXPKTSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_IP6TXFRAGS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_IP6TXFRAGS] = I40E_GLPES_VFIP6TXFRAGSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_TCPRXSEGS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_TCPRXSEGS] = I40E_GLPES_VFTCPRXSEGSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_TCPTXSEG] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_TCPTXSEG] = I40E_GLPES_VFTCPTXSEGLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMARXRDS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMARXRDS] = I40E_GLPES_VFRDMARXRDSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMARXSNDS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMARXSNDS] = I40E_GLPES_VFRDMARXSNDSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMARXWRS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMARXWRS] = I40E_GLPES_VFRDMARXWRSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMATXRDS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMATXRDS] = I40E_GLPES_VFRDMATXRDSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMATXSNDS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMATXSNDS] = I40E_GLPES_VFRDMATXSNDSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMATXWRS] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMATXWRS] = I40E_GLPES_VFRDMATXWRSLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMAVBND] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMAVBND] = I40E_GLPES_VFRDMAVBNDLO(fcn_idx); - stat_table->stat_offset_64[I40IW_HW_STAT_INDEX_RDMAVINV] = + stats_table->stats_offset_64[I40IW_HW_STAT_INDEX_RDMAVINV] = I40E_GLPES_VFRDMAVINVLO(fcn_idx); } - for (stat_index = 0; stat_index < I40IW_HW_STAT_INDEX_MAX_64; - stat_index++) { - stat_reg_offset = stat_table->stat_offset_64[stat_index]; - last_rd_stats->stat_value_64[stat_index] = - readq(devstat->hw->hw_addr + stat_reg_offset); + for (stats_index = 0; stats_index < I40IW_HW_STAT_INDEX_MAX_64; + stats_index++) { + stats_reg_offset = stats_table->stats_offset_64[stats_index]; + last_rd_stats->stats_value_64[stats_index] = + readq(stats->hw->hw_addr + stats_reg_offset); } - for (stat_index = 0; stat_index < I40IW_HW_STAT_INDEX_MAX_32; - stat_index++) { - stat_reg_offset = stat_table->stat_offset_32[stat_index]; - last_rd_stats->stat_value_32[stat_index] = - i40iw_rd32(devstat->hw, stat_reg_offset); + for (stats_index = 0; stats_index < I40IW_HW_STAT_INDEX_MAX_32; + stats_index++) { + stats_reg_offset = stats_table->stats_offset_32[stats_index]; + last_rd_stats->stats_value_32[stats_index] = + i40iw_rd32(stats->hw, stats_reg_offset); } } /** - * i40iw_hw_stat_read_32 - Read 32-bit HW stat counters and accommodates for roll-overs. - * @devstat: pestat struct - * @index: index in HW stat table which contains offset reg-addr - * @value: hw stat value + * i40iw_hw_stats_read_32 - Read 32-bit HW stats counters and accommodates for roll-overs. + * @stat: pestat struct + * @index: index in HW stats table which contains offset reg-addr + * @value: hw stats value */ -static void i40iw_hw_stat_read_32(struct i40iw_dev_pestat *devstat, - enum i40iw_hw_stat_index_32b index, - u64 *value) +void i40iw_hw_stats_read_32(struct i40iw_vsi_pestat *stats, + enum i40iw_hw_stats_index_32b index, + u64 *value) { - struct i40iw_dev_hw_stat_offsets *stat_table = - &devstat->hw_stat_offsets; - struct i40iw_dev_hw_stats *last_rd_stats = &devstat->last_read_hw_stats; - struct i40iw_dev_hw_stats *hw_stats = &devstat->hw_stats; - u64 new_stat_value = 0; - u32 stat_reg_offset = stat_table->stat_offset_32[index]; - - new_stat_value = i40iw_rd32(devstat->hw, stat_reg_offset); + struct i40iw_dev_hw_stats_offsets *stats_table = + &stats->hw_stats_offsets; + struct i40iw_dev_hw_stats *last_rd_stats = &stats->last_read_hw_stats; + struct i40iw_dev_hw_stats *hw_stats = &stats->hw_stats; + u64 new_stats_value = 0; + u32 stats_reg_offset = stats_table->stats_offset_32[index]; + + new_stats_value = i40iw_rd32(stats->hw, stats_reg_offset); /*roll-over case */ - if (new_stat_value < last_rd_stats->stat_value_32[index]) - hw_stats->stat_value_32[index] += new_stat_value; + if (new_stats_value < last_rd_stats->stats_value_32[index]) + hw_stats->stats_value_32[index] += new_stats_value; else - hw_stats->stat_value_32[index] += - new_stat_value - last_rd_stats->stat_value_32[index]; - last_rd_stats->stat_value_32[index] = new_stat_value; - *value = hw_stats->stat_value_32[index]; + hw_stats->stats_value_32[index] += + new_stats_value - last_rd_stats->stats_value_32[index]; + last_rd_stats->stats_value_32[index] = new_stats_value; + *value = hw_stats->stats_value_32[index]; } /** - * i40iw_hw_stat_read_64 - Read HW stat counters (greater than 32-bit) and accommodates for roll-overs. - * @devstat: pestat struct - * @index: index in HW stat table which contains offset reg-addr - * @value: hw stat value + * i40iw_hw_stats_read_64 - Read HW stats counters (greater than 32-bit) and accommodates for roll-overs. + * @stats: pestat struct + * @index: index in HW stats table which contains offset reg-addr + * @value: hw stats value */ -static void i40iw_hw_stat_read_64(struct i40iw_dev_pestat *devstat, - enum i40iw_hw_stat_index_64b index, - u64 *value) +void i40iw_hw_stats_read_64(struct i40iw_vsi_pestat *stats, + enum i40iw_hw_stats_index_64b index, + u64 *value) { - struct i40iw_dev_hw_stat_offsets *stat_table = - &devstat->hw_stat_offsets; - struct i40iw_dev_hw_stats *last_rd_stats = &devstat->last_read_hw_stats; - struct i40iw_dev_hw_stats *hw_stats = &devstat->hw_stats; - u64 new_stat_value = 0; - u32 stat_reg_offset = stat_table->stat_offset_64[index]; - - new_stat_value = readq(devstat->hw->hw_addr + stat_reg_offset); + struct i40iw_dev_hw_stats_offsets *stats_table = + &stats->hw_stats_offsets; + struct i40iw_dev_hw_stats *last_rd_stats = &stats->last_read_hw_stats; + struct i40iw_dev_hw_stats *hw_stats = &stats->hw_stats; + u64 new_stats_value = 0; + u32 stats_reg_offset = stats_table->stats_offset_64[index]; + + new_stats_value = readq(stats->hw->hw_addr + stats_reg_offset); /*roll-over case */ - if (new_stat_value < last_rd_stats->stat_value_64[index]) - hw_stats->stat_value_64[index] += new_stat_value; + if (new_stats_value < last_rd_stats->stats_value_64[index]) + hw_stats->stats_value_64[index] += new_stats_value; else - hw_stats->stat_value_64[index] += - new_stat_value - last_rd_stats->stat_value_64[index]; - last_rd_stats->stat_value_64[index] = new_stat_value; - *value = hw_stats->stat_value_64[index]; + hw_stats->stats_value_64[index] += + new_stats_value - last_rd_stats->stats_value_64[index]; + last_rd_stats->stats_value_64[index] = new_stats_value; + *value = hw_stats->stats_value_64[index]; } /** - * i40iw_hw_stat_read_all - read all HW stat counters - * @devstat: pestat struct - * @stat_values: hw stats structure + * i40iw_hw_stats_read_all - read all HW stat counters + * @stats: pestat struct + * @stats_values: hw stats structure * * Read all the HW stat counters and populates hw_stats structure - * of passed-in dev's pestat as well as copy created in stat_values. + * of passed-in vsi's pestat as well as copy created in stat_values. */ -static void i40iw_hw_stat_read_all(struct i40iw_dev_pestat *devstat, - struct i40iw_dev_hw_stats *stat_values) +void i40iw_hw_stats_read_all(struct i40iw_vsi_pestat *stats, + struct i40iw_dev_hw_stats *stats_values) { - u32 stat_index; - - for (stat_index = 0; stat_index < I40IW_HW_STAT_INDEX_MAX_32; - stat_index++) - i40iw_hw_stat_read_32(devstat, stat_index, - &stat_values->stat_value_32[stat_index]); - for (stat_index = 0; stat_index < I40IW_HW_STAT_INDEX_MAX_64; - stat_index++) - i40iw_hw_stat_read_64(devstat, stat_index, - &stat_values->stat_value_64[stat_index]); + u32 stats_index; + unsigned long flags; + + spin_lock_irqsave(&stats->lock, flags); + + for (stats_index = 0; stats_index < I40IW_HW_STAT_INDEX_MAX_32; + stats_index++) + i40iw_hw_stats_read_32(stats, stats_index, + &stats_values->stats_value_32[stats_index]); + for (stats_index = 0; stats_index < I40IW_HW_STAT_INDEX_MAX_64; + stats_index++) + i40iw_hw_stats_read_64(stats, stats_index, + &stats_values->stats_value_64[stats_index]); + spin_unlock_irqrestore(&stats->lock, flags); } /** - * i40iw_hw_stat_refresh_all - Update all HW stat structs - * @devstat: pestat struct - * @stat_values: hw stats structure + * i40iw_hw_stats_refresh_all - Update all HW stats structs + * @stats: pestat struct * - * Read all the HW stat counters to refresh values in hw_stats structure + * Read all the HW stats counters to refresh values in hw_stats structure * of passed-in dev's pestat */ -static void i40iw_hw_stat_refresh_all(struct i40iw_dev_pestat *devstat) +void i40iw_hw_stats_refresh_all(struct i40iw_vsi_pestat *stats) +{ + u64 stats_value; + u32 stats_index; + unsigned long flags; + + spin_lock_irqsave(&stats->lock, flags); + + for (stats_index = 0; stats_index < I40IW_HW_STAT_INDEX_MAX_32; + stats_index++) + i40iw_hw_stats_read_32(stats, stats_index, &stats_value); + for (stats_index = 0; stats_index < I40IW_HW_STAT_INDEX_MAX_64; + stats_index++) + i40iw_hw_stats_read_64(stats, stats_index, &stats_value); + spin_unlock_irqrestore(&stats->lock, flags); +} + +/** + * i40iw_get_fcn_id - Return the function id + * @dev: pointer to the device + */ +static u8 i40iw_get_fcn_id(struct i40iw_sc_dev *dev) { - u64 stat_value; - u32 stat_index; - - for (stat_index = 0; stat_index < I40IW_HW_STAT_INDEX_MAX_32; - stat_index++) - i40iw_hw_stat_read_32(devstat, stat_index, &stat_value); - for (stat_index = 0; stat_index < I40IW_HW_STAT_INDEX_MAX_64; - stat_index++) - i40iw_hw_stat_read_64(devstat, stat_index, &stat_value); + u8 fcn_id = I40IW_INVALID_FCN_ID; + u8 i; + + for (i = I40IW_FIRST_NON_PF_STAT; i < I40IW_MAX_STATS_COUNT; i++) + if (!dev->fcn_id_array[i]) { + fcn_id = i; + dev->fcn_id_array[i] = true; + break; + } + return fcn_id; +} + +/** + * i40iw_vsi_stats_init - Initialize the vsi statistics + * @vsi: pointer to the vsi structure + * @info: The info structure used for initialization + */ +enum i40iw_status_code i40iw_vsi_stats_init(struct i40iw_sc_vsi *vsi, struct i40iw_vsi_stats_info *info) +{ + u8 fcn_id = info->fcn_id; + + if (info->alloc_fcn_id) + fcn_id = i40iw_get_fcn_id(vsi->dev); + + if (fcn_id == I40IW_INVALID_FCN_ID) + return I40IW_ERR_NOT_READY; + + vsi->pestat = info->pestat; + vsi->pestat->hw = vsi->dev->hw; + + if (info->stats_initialize) { + i40iw_hw_stats_init(vsi->pestat, fcn_id, true); + spin_lock_init(&vsi->pestat->lock); + i40iw_hw_stats_start_timer(vsi); + } + vsi->stats_fcn_id_alloc = info->alloc_fcn_id; + vsi->fcn_id = fcn_id; + return I40IW_SUCCESS; +} + +/** + * i40iw_vsi_stats_free - Free the vsi stats + * @vsi: pointer to the vsi structure + */ +void i40iw_vsi_stats_free(struct i40iw_sc_vsi *vsi) +{ + u8 fcn_id = vsi->fcn_id; + + if ((vsi->stats_fcn_id_alloc) && (fcn_id != I40IW_INVALID_FCN_ID)) + vsi->dev->fcn_id_array[fcn_id] = false; + i40iw_hw_stats_stop_timer(vsi); } static struct i40iw_cqp_ops iw_cqp_ops = { - i40iw_sc_cqp_init, - i40iw_sc_cqp_create, - i40iw_sc_cqp_post_sq, - i40iw_sc_cqp_get_next_send_wqe, - i40iw_sc_cqp_destroy, - i40iw_sc_poll_for_cqp_op_done + .cqp_init = i40iw_sc_cqp_init, + .cqp_create = i40iw_sc_cqp_create, + .cqp_post_sq = i40iw_sc_cqp_post_sq, + .cqp_get_next_send_wqe = i40iw_sc_cqp_get_next_send_wqe, + .cqp_destroy = i40iw_sc_cqp_destroy, + .poll_for_cqp_op_done = i40iw_sc_poll_for_cqp_op_done }; static struct i40iw_ccq_ops iw_ccq_ops = { - i40iw_sc_ccq_init, - i40iw_sc_ccq_create, - i40iw_sc_ccq_destroy, - i40iw_sc_ccq_create_done, - i40iw_sc_ccq_get_cqe_info, - i40iw_sc_ccq_arm + .ccq_init = i40iw_sc_ccq_init, + .ccq_create = i40iw_sc_ccq_create, + .ccq_destroy = i40iw_sc_ccq_destroy, + .ccq_create_done = i40iw_sc_ccq_create_done, + .ccq_get_cqe_info = i40iw_sc_ccq_get_cqe_info, + .ccq_arm = i40iw_sc_ccq_arm }; static struct i40iw_ceq_ops iw_ceq_ops = { - i40iw_sc_ceq_init, - i40iw_sc_ceq_create, - i40iw_sc_cceq_create_done, - i40iw_sc_cceq_destroy_done, - i40iw_sc_cceq_create, - i40iw_sc_ceq_destroy, - i40iw_sc_process_ceq + .ceq_init = i40iw_sc_ceq_init, + .ceq_create = i40iw_sc_ceq_create, + .cceq_create_done = i40iw_sc_cceq_create_done, + .cceq_destroy_done = i40iw_sc_cceq_destroy_done, + .cceq_create = i40iw_sc_cceq_create, + .ceq_destroy = i40iw_sc_ceq_destroy, + .process_ceq = i40iw_sc_process_ceq }; static struct i40iw_aeq_ops iw_aeq_ops = { - i40iw_sc_aeq_init, - i40iw_sc_aeq_create, - i40iw_sc_aeq_destroy, - i40iw_sc_get_next_aeqe, - i40iw_sc_repost_aeq_entries, - i40iw_sc_aeq_create_done, - i40iw_sc_aeq_destroy_done + .aeq_init = i40iw_sc_aeq_init, + .aeq_create = i40iw_sc_aeq_create, + .aeq_destroy = i40iw_sc_aeq_destroy, + .get_next_aeqe = i40iw_sc_get_next_aeqe, + .repost_aeq_entries = i40iw_sc_repost_aeq_entries, + .aeq_create_done = i40iw_sc_aeq_create_done, + .aeq_destroy_done = i40iw_sc_aeq_destroy_done }; /* iwarp pd ops */ static struct i40iw_pd_ops iw_pd_ops = { - i40iw_sc_pd_init, + .pd_init = i40iw_sc_pd_init, }; static struct i40iw_priv_qp_ops iw_priv_qp_ops = { @@ -4662,74 +4912,54 @@ static struct i40iw_priv_qp_ops iw_priv_qp_ops = { }; static struct i40iw_priv_cq_ops iw_priv_cq_ops = { - i40iw_sc_cq_init, - i40iw_sc_cq_create, - i40iw_sc_cq_destroy, - i40iw_sc_cq_modify, + .cq_init = i40iw_sc_cq_init, + .cq_create = i40iw_sc_cq_create, + .cq_destroy = i40iw_sc_cq_destroy, + .cq_modify = i40iw_sc_cq_modify, }; static struct i40iw_mr_ops iw_mr_ops = { - i40iw_sc_alloc_stag, - i40iw_sc_mr_reg_non_shared, - i40iw_sc_mr_reg_shared, - i40iw_sc_dealloc_stag, - i40iw_sc_query_stag, - i40iw_sc_mw_alloc + .alloc_stag = i40iw_sc_alloc_stag, + .mr_reg_non_shared = i40iw_sc_mr_reg_non_shared, + .mr_reg_shared = i40iw_sc_mr_reg_shared, + .dealloc_stag = i40iw_sc_dealloc_stag, + .query_stag = i40iw_sc_query_stag, + .mw_alloc = i40iw_sc_mw_alloc }; static struct i40iw_cqp_misc_ops iw_cqp_misc_ops = { - i40iw_sc_manage_push_page, - i40iw_sc_manage_hmc_pm_func_table, - i40iw_sc_set_hmc_resource_profile, - i40iw_sc_commit_fpm_values, - i40iw_sc_query_fpm_values, - i40iw_sc_static_hmc_pages_allocated, - i40iw_sc_add_arp_cache_entry, - i40iw_sc_del_arp_cache_entry, - i40iw_sc_query_arp_cache_entry, - i40iw_sc_manage_apbvt_entry, - i40iw_sc_manage_qhash_table_entry, - i40iw_sc_alloc_local_mac_ipaddr_entry, - i40iw_sc_add_local_mac_ipaddr_entry, - i40iw_sc_del_local_mac_ipaddr_entry, - i40iw_sc_cqp_nop, - i40iw_sc_commit_fpm_values_done, - i40iw_sc_query_fpm_values_done, - i40iw_sc_manage_hmc_pm_func_table_done, - i40iw_sc_suspend_qp, - i40iw_sc_resume_qp + .manage_push_page = i40iw_sc_manage_push_page, + .manage_hmc_pm_func_table = i40iw_sc_manage_hmc_pm_func_table, + .set_hmc_resource_profile = i40iw_sc_set_hmc_resource_profile, + .commit_fpm_values = i40iw_sc_commit_fpm_values, + .query_fpm_values = i40iw_sc_query_fpm_values, + .static_hmc_pages_allocated = i40iw_sc_static_hmc_pages_allocated, + .add_arp_cache_entry = i40iw_sc_add_arp_cache_entry, + .del_arp_cache_entry = i40iw_sc_del_arp_cache_entry, + .query_arp_cache_entry = i40iw_sc_query_arp_cache_entry, + .manage_apbvt_entry = i40iw_sc_manage_apbvt_entry, + .manage_qhash_table_entry = i40iw_sc_manage_qhash_table_entry, + .alloc_local_mac_ipaddr_table_entry = i40iw_sc_alloc_local_mac_ipaddr_entry, + .add_local_mac_ipaddr_entry = i40iw_sc_add_local_mac_ipaddr_entry, + .del_local_mac_ipaddr_entry = i40iw_sc_del_local_mac_ipaddr_entry, + .cqp_nop = i40iw_sc_cqp_nop, + .commit_fpm_values_done = i40iw_sc_commit_fpm_values_done, + .query_fpm_values_done = i40iw_sc_query_fpm_values_done, + .manage_hmc_pm_func_table_done = i40iw_sc_manage_hmc_pm_func_table_done, + .update_suspend_qp = i40iw_sc_suspend_qp, + .update_resume_qp = i40iw_sc_resume_qp }; static struct i40iw_hmc_ops iw_hmc_ops = { - i40iw_sc_init_iw_hmc, - i40iw_sc_parse_fpm_query_buf, - i40iw_sc_configure_iw_fpm, - i40iw_sc_parse_fpm_commit_buf, - i40iw_sc_create_hmc_obj, - i40iw_sc_del_hmc_obj, - NULL, - NULL -}; - -static const struct i40iw_device_pestat_ops iw_device_pestat_ops = { - i40iw_hw_stat_init, - i40iw_hw_stat_read_32, - i40iw_hw_stat_read_64, - i40iw_hw_stat_read_all, - i40iw_hw_stat_refresh_all + .init_iw_hmc = i40iw_sc_init_iw_hmc, + .parse_fpm_query_buf = i40iw_sc_parse_fpm_query_buf, + .configure_iw_fpm = i40iw_sc_configure_iw_fpm, + .parse_fpm_commit_buf = i40iw_sc_parse_fpm_commit_buf, + .create_hmc_object = i40iw_sc_create_hmc_obj, + .del_hmc_object = i40iw_sc_del_hmc_obj }; /** - * i40iw_device_init_pestat - Initialize the pestat structure - * @dev: pestat struct - */ -enum i40iw_status_code i40iw_device_init_pestat(struct i40iw_dev_pestat *devstat) -{ - devstat->ops = iw_device_pestat_ops; - return 0; -} - -/** * i40iw_device_init - Initialize IWARP device * @dev: IWARP device pointer * @info: IWARP init info @@ -4750,14 +4980,7 @@ enum i40iw_status_code i40iw_device_init(struct i40iw_sc_dev *dev, dev->debug_mask = info->debug_mask; - ret_code = i40iw_device_init_pestat(&dev->dev_pestat); - if (ret_code) { - i40iw_debug(dev, I40IW_DEBUG_DEV, - "%s: i40iw_device_init_pestat failed\n", __func__); - return ret_code; - } dev->hmc_fn_id = info->hmc_fn_id; - dev->qs_handle = info->qs_handle; dev->exception_lan_queue = info->exception_lan_queue; dev->is_pf = info->is_pf; @@ -4770,15 +4993,10 @@ enum i40iw_status_code i40iw_device_init(struct i40iw_sc_dev *dev, dev->hw = info->hw; dev->hw->hw_addr = info->bar0; - val = i40iw_rd32(dev->hw, I40E_GLPCI_DREVID); - dev->hw_rev = (u8)RS_32(val, I40E_GLPCI_DREVID_DEFAULT_REVID); - if (dev->is_pf) { - dev->dev_pestat.ops.iw_hw_stat_init(&dev->dev_pestat, - dev->hmc_fn_id, dev->hw, true); - spin_lock_init(&dev->dev_pestat.stats_lock); - /*start the periodic stats_timer */ - i40iw_hw_stats_start_timer(dev); + val = i40iw_rd32(dev->hw, I40E_GLPCI_DREVID); + dev->hw_rev = (u8)RS_32(val, I40E_GLPCI_DREVID_DEFAULT_REVID); + val = i40iw_rd32(dev->hw, I40E_GLPCI_LBARCTRL); db_size = (u8)RS_32(val, I40E_GLPCI_LBARCTRL_PE_DB_SIZE); if ((db_size != I40IW_PE_DB_SIZE_4M) && diff --git a/drivers/infiniband/hw/i40iw/i40iw_d.h b/drivers/infiniband/hw/i40iw/i40iw_d.h index 2fac1db0e0a0..a39ac12b6a7e 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_d.h +++ b/drivers/infiniband/hw/i40iw/i40iw_d.h @@ -35,6 +35,8 @@ #ifndef I40IW_D_H #define I40IW_D_H +#define I40IW_FIRST_USER_QP_ID 2 + #define I40IW_DB_ADDR_OFFSET (4 * 1024 * 1024 - 64 * 1024) #define I40IW_VF_DB_ADDR_OFFSET (64 * 1024) @@ -67,6 +69,9 @@ #define I40IW_STAG_TYPE_NONSHARED 1 #define I40IW_MAX_USER_PRIORITY 8 +#define I40IW_MAX_STATS_COUNT 16 +#define I40IW_FIRST_NON_PF_STAT 4 + #define LS_64_1(val, bits) ((u64)(uintptr_t)val << bits) #define RS_64_1(val, bits) ((u64)(uintptr_t)val >> bits) @@ -74,6 +79,8 @@ #define RS_32_1(val, bits) (u32)(val >> bits) #define I40E_HI_DWORD(x) ((u32)((((x) >> 16) >> 16) & 0xFFFFFFFF)) +#define QS_HANDLE_UNKNOWN 0xffff + #define LS_64(val, field) (((u64)val << field ## _SHIFT) & (field ## _MASK)) #define RS_64(val, field) ((u64)(val & field ## _MASK) >> field ## _SHIFT) @@ -1199,8 +1206,11 @@ #define I40IWQPC_RXCQNUM_SHIFT 32 #define I40IWQPC_RXCQNUM_MASK (0x1ffffULL << I40IWQPC_RXCQNUM_SHIFT) -#define I40IWQPC_Q2ADDR_SHIFT I40IW_CQPHC_QPCTX_SHIFT -#define I40IWQPC_Q2ADDR_MASK I40IW_CQPHC_QPCTX_MASK +#define I40IWQPC_STAT_INDEX_SHIFT 0 +#define I40IWQPC_STAT_INDEX_MASK (0x1fULL << I40IWQPC_STAT_INDEX_SHIFT) + +#define I40IWQPC_Q2ADDR_SHIFT 0 +#define I40IWQPC_Q2ADDR_MASK (0xffffffffffffff00ULL << I40IWQPC_Q2ADDR_SHIFT) #define I40IWQPC_LASTBYTESENT_SHIFT 0 #define I40IWQPC_LASTBYTESENT_MASK (0xffUL << I40IWQPC_LASTBYTESENT_SHIFT) @@ -1232,11 +1242,8 @@ #define I40IWQPC_PRIVEN_SHIFT 25 #define I40IWQPC_PRIVEN_MASK (1UL << I40IWQPC_PRIVEN_SHIFT) -#define I40IWQPC_LSMMPRESENT_SHIFT 26 -#define I40IWQPC_LSMMPRESENT_MASK (1UL << I40IWQPC_LSMMPRESENT_SHIFT) - -#define I40IWQPC_ADJUSTFORLSMM_SHIFT 27 -#define I40IWQPC_ADJUSTFORLSMM_MASK (1UL << I40IWQPC_ADJUSTFORLSMM_SHIFT) +#define I40IWQPC_USESTATSINSTANCE_SHIFT 26 +#define I40IWQPC_USESTATSINSTANCE_MASK (1UL << I40IWQPC_USESTATSINSTANCE_SHIFT) #define I40IWQPC_IWARPMODE_SHIFT 28 #define I40IWQPC_IWARPMODE_MASK (1UL << I40IWQPC_IWARPMODE_SHIFT) @@ -1713,6 +1720,8 @@ enum i40iw_alignment { #define OP_MANAGE_VF_PBLE_BP 28 #define OP_QUERY_FPM_VALUES 29 #define OP_COMMIT_FPM_VALUES 30 -#define OP_SIZE_CQP_STAT_ARRAY 31 +#define OP_REQUESTED_COMMANDS 31 +#define OP_COMPLETED_COMMANDS 32 +#define OP_SIZE_CQP_STAT_ARRAY 33 #endif diff --git a/drivers/infiniband/hw/i40iw/i40iw_hw.c b/drivers/infiniband/hw/i40iw/i40iw_hw.c index 0c92a40b3e86..476867a3f584 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_hw.c +++ b/drivers/infiniband/hw/i40iw/i40iw_hw.c @@ -62,7 +62,7 @@ u32 i40iw_initialize_hw_resources(struct i40iw_device *iwdev) max_mr = iwdev->sc_dev.hmc_info->hmc_obj[I40IW_HMC_IW_MR].cnt; arp_table_size = iwdev->sc_dev.hmc_info->hmc_obj[I40IW_HMC_IW_ARP].cnt; iwdev->max_cqe = 0xFFFFF; - num_pds = max_qp * 4; + num_pds = I40IW_MAX_PDS; resources_size = sizeof(struct i40iw_arp_entry) * arp_table_size; resources_size += sizeof(unsigned long) * BITS_TO_LONGS(max_qp); resources_size += sizeof(unsigned long) * BITS_TO_LONGS(max_mr); @@ -308,7 +308,9 @@ void i40iw_process_aeq(struct i40iw_device *iwdev) iwqp = iwdev->qp_table[info->qp_cq_id]; if (!iwqp) { spin_unlock_irqrestore(&iwdev->qptable_lock, flags); - i40iw_pr_err("qp_id %d is already freed\n", info->qp_cq_id); + i40iw_debug(dev, I40IW_DEBUG_AEQ, + "%s qp_id %d is already freed\n", + __func__, info->qp_cq_id); continue; } i40iw_add_ref(&iwqp->ibqp); @@ -359,6 +361,9 @@ void i40iw_process_aeq(struct i40iw_device *iwdev) continue; i40iw_cm_disconn(iwqp); break; + case I40IW_AE_QP_SUSPEND_COMPLETE: + i40iw_qp_suspend_resume(dev, &iwqp->sc_qp, false); + break; case I40IW_AE_TERMINATE_SENT: i40iw_terminate_send_fin(qp); break; @@ -404,19 +409,18 @@ void i40iw_process_aeq(struct i40iw_device *iwdev) case I40IW_AE_LCE_CQ_CATASTROPHIC: case I40IW_AE_UDA_XMIT_DGRAM_TOO_LONG: case I40IW_AE_UDA_XMIT_IPADDR_MISMATCH: - case I40IW_AE_QP_SUSPEND_COMPLETE: ctx_info->err_rq_idx_valid = false; default: - if (!info->sq && ctx_info->err_rq_idx_valid) { - ctx_info->err_rq_idx = info->wqe_idx; - ctx_info->tcp_info_valid = false; - ctx_info->iwarp_info_valid = false; - ret = dev->iw_priv_qp_ops->qp_setctx(&iwqp->sc_qp, - iwqp->host_ctx.va, - ctx_info); - } - i40iw_terminate_connection(qp, info); - break; + if (!info->sq && ctx_info->err_rq_idx_valid) { + ctx_info->err_rq_idx = info->wqe_idx; + ctx_info->tcp_info_valid = false; + ctx_info->iwarp_info_valid = false; + ret = dev->iw_priv_qp_ops->qp_setctx(&iwqp->sc_qp, + iwqp->host_ctx.va, + ctx_info); + } + i40iw_terminate_connection(qp, info); + break; } if (info->qp) i40iw_rem_ref(&iwqp->ibqp); @@ -538,6 +542,7 @@ enum i40iw_status_code i40iw_manage_qhash(struct i40iw_device *iwdev, { struct i40iw_qhash_table_info *info; struct i40iw_sc_dev *dev = &iwdev->sc_dev; + struct i40iw_sc_vsi *vsi = &iwdev->vsi; enum i40iw_status_code status; struct i40iw_cqp *iwcqp = &iwdev->cqp; struct i40iw_cqp_request *cqp_request; @@ -550,6 +555,7 @@ enum i40iw_status_code i40iw_manage_qhash(struct i40iw_device *iwdev, info = &cqp_info->in.u.manage_qhash_table_entry.info; memset(info, 0, sizeof(*info)); + info->vsi = &iwdev->vsi; info->manage = mtype; info->entry_type = etype; if (cminfo->vlan_id != 0xFFFF) { @@ -560,8 +566,9 @@ enum i40iw_status_code i40iw_manage_qhash(struct i40iw_device *iwdev, } info->ipv4_valid = cminfo->ipv4; + info->user_pri = cminfo->user_pri; ether_addr_copy(info->mac_addr, iwdev->netdev->dev_addr); - info->qp_num = cpu_to_le32(dev->ilq->qp_id); + info->qp_num = cpu_to_le32(vsi->ilq->qp_id); info->dest_port = cpu_to_le16(cminfo->loc_port); info->dest_ip[0] = cpu_to_le32(cminfo->loc_addr[0]); info->dest_ip[1] = cpu_to_le32(cminfo->loc_addr[1]); @@ -617,6 +624,7 @@ enum i40iw_status_code i40iw_hw_flush_wqes(struct i40iw_device *iwdev, struct i40iw_qp_flush_info *hw_info; struct i40iw_cqp_request *cqp_request; struct cqp_commands_info *cqp_info; + struct i40iw_qp *iwqp = (struct i40iw_qp *)qp->back_qp; cqp_request = i40iw_get_cqp_request(&iwdev->cqp, wait); if (!cqp_request) @@ -631,9 +639,30 @@ enum i40iw_status_code i40iw_hw_flush_wqes(struct i40iw_device *iwdev, cqp_info->in.u.qp_flush_wqes.qp = qp; cqp_info->in.u.qp_flush_wqes.scratch = (uintptr_t)cqp_request; status = i40iw_handle_cqp_op(iwdev, cqp_request); - if (status) + if (status) { i40iw_pr_err("CQP-OP Flush WQE's fail"); - return status; + complete(&iwqp->sq_drained); + complete(&iwqp->rq_drained); + return status; + } + if (!cqp_request->compl_info.maj_err_code) { + switch (cqp_request->compl_info.min_err_code) { + case I40IW_CQP_COMPL_RQ_WQE_FLUSHED: + complete(&iwqp->sq_drained); + break; + case I40IW_CQP_COMPL_SQ_WQE_FLUSHED: + complete(&iwqp->rq_drained); + break; + case I40IW_CQP_COMPL_RQ_SQ_WQE_FLUSHED: + break; + default: + complete(&iwqp->sq_drained); + complete(&iwqp->rq_drained); + break; + } + } + + return 0; } /** diff --git a/drivers/infiniband/hw/i40iw/i40iw_main.c b/drivers/infiniband/hw/i40iw/i40iw_main.c index ac2f3cd9478c..2728af3103ce 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_main.c +++ b/drivers/infiniband/hw/i40iw/i40iw_main.c @@ -237,14 +237,11 @@ static irqreturn_t i40iw_irq_handler(int irq, void *data) */ static void i40iw_destroy_cqp(struct i40iw_device *iwdev, bool free_hwcqp) { - enum i40iw_status_code status = 0; struct i40iw_sc_dev *dev = &iwdev->sc_dev; struct i40iw_cqp *cqp = &iwdev->cqp; - if (free_hwcqp && dev->cqp_ops->cqp_destroy) - status = dev->cqp_ops->cqp_destroy(dev->cqp); - if (status) - i40iw_pr_err("destroy cqp failed"); + if (free_hwcqp) + dev->cqp_ops->cqp_destroy(dev->cqp); i40iw_free_dma_mem(dev->hw, &cqp->sq); kfree(cqp->scratch_array); @@ -270,6 +267,7 @@ static void i40iw_disable_irq(struct i40iw_sc_dev *dev, i40iw_wr32(dev->hw, I40E_PFINT_DYN_CTLN(msix_vec->idx - 1), 0); else i40iw_wr32(dev->hw, I40E_VFINT_DYN_CTLN1(msix_vec->idx - 1), 0); + irq_set_affinity_hint(msix_vec->irq, NULL); free_irq(msix_vec->irq, dev_id); } @@ -603,7 +601,7 @@ static enum i40iw_status_code i40iw_create_cqp(struct i40iw_device *iwdev) i40iw_pr_err("cqp init status %d\n", status); goto exit; } - status = dev->cqp_ops->cqp_create(dev->cqp, true, &maj_err, &min_err); + status = dev->cqp_ops->cqp_create(dev->cqp, &maj_err, &min_err); if (status) { i40iw_pr_err("cqp create status %d maj_err %d min_err %d\n", status, maj_err, min_err); @@ -688,6 +686,7 @@ static enum i40iw_status_code i40iw_configure_ceq_vector(struct i40iw_device *iw struct i40iw_msix_vector *msix_vec) { enum i40iw_status_code status; + cpumask_t mask; if (iwdev->msix_shared && !ceq_id) { tasklet_init(&iwdev->dpc_tasklet, i40iw_dpc, (unsigned long)iwdev); @@ -697,12 +696,15 @@ static enum i40iw_status_code i40iw_configure_ceq_vector(struct i40iw_device *iw status = request_irq(msix_vec->irq, i40iw_ceq_handler, 0, "CEQ", iwceq); } + cpumask_clear(&mask); + cpumask_set_cpu(msix_vec->cpu_affinity, &mask); + irq_set_affinity_hint(msix_vec->irq, &mask); + if (status) { i40iw_pr_err("ceq irq config fail\n"); return I40IW_ERR_CONFIG; } msix_vec->ceq_id = ceq_id; - msix_vec->cpu_affinity = 0; return 0; } @@ -930,6 +932,7 @@ static enum i40iw_status_code i40iw_initialize_ilq(struct i40iw_device *iwdev) struct i40iw_puda_rsrc_info info; enum i40iw_status_code status; + memset(&info, 0, sizeof(info)); info.type = I40IW_PUDA_RSRC_TYPE_ILQ; info.cq_id = 1; info.qp_id = 0; @@ -939,10 +942,9 @@ static enum i40iw_status_code i40iw_initialize_ilq(struct i40iw_device *iwdev) info.rq_size = 8192; info.buf_size = 1024; info.tx_buf_cnt = 16384; - info.mss = iwdev->mss; info.receive = i40iw_receive_ilq; info.xmit_complete = i40iw_free_sqbuf; - status = i40iw_puda_create_rsrc(&iwdev->sc_dev, &info); + status = i40iw_puda_create_rsrc(&iwdev->vsi, &info); if (status) i40iw_pr_err("ilq create fail\n"); return status; @@ -959,6 +961,7 @@ static enum i40iw_status_code i40iw_initialize_ieq(struct i40iw_device *iwdev) struct i40iw_puda_rsrc_info info; enum i40iw_status_code status; + memset(&info, 0, sizeof(info)); info.type = I40IW_PUDA_RSRC_TYPE_IEQ; info.cq_id = 2; info.qp_id = iwdev->sc_dev.exception_lan_queue; @@ -967,9 +970,8 @@ static enum i40iw_status_code i40iw_initialize_ieq(struct i40iw_device *iwdev) info.sq_size = 8192; info.rq_size = 8192; info.buf_size = 2048; - info.mss = iwdev->mss; info.tx_buf_cnt = 16384; - status = i40iw_puda_create_rsrc(&iwdev->sc_dev, &info); + status = i40iw_puda_create_rsrc(&iwdev->vsi, &info); if (status) i40iw_pr_err("ieq create fail\n"); return status; @@ -1159,7 +1161,7 @@ static void i40iw_add_ipv6_addr(struct i40iw_device *iwdev) { struct net_device *ip_dev; struct inet6_dev *idev; - struct inet6_ifaddr *ifp; + struct inet6_ifaddr *ifp, *tmp; u32 local_ipaddr6[4]; rcu_read_lock(); @@ -1172,7 +1174,7 @@ static void i40iw_add_ipv6_addr(struct i40iw_device *iwdev) i40iw_pr_err("ipv6 inet device not found\n"); break; } - list_for_each_entry(ifp, &idev->addr_list, if_list) { + list_for_each_entry_safe(ifp, tmp, &idev->addr_list, if_list) { i40iw_pr_info("IP=%pI6, vlan_id=%d, MAC=%pM\n", &ifp->addr, rdma_vlan_dev_vlan_id(ip_dev), ip_dev->dev_addr); i40iw_copy_ip_ntohl(local_ipaddr6, @@ -1294,17 +1296,23 @@ static enum i40iw_status_code i40iw_initialize_dev(struct i40iw_device *iwdev, enum i40iw_status_code status; struct i40iw_sc_dev *dev = &iwdev->sc_dev; struct i40iw_device_init_info info; + struct i40iw_vsi_init_info vsi_info; struct i40iw_dma_mem mem; + struct i40iw_l2params l2params; u32 size; + struct i40iw_vsi_stats_info stats_info; + u16 last_qset = I40IW_NO_QSET; + u16 qset; + u32 i; + memset(&l2params, 0, sizeof(l2params)); memset(&info, 0, sizeof(info)); size = sizeof(struct i40iw_hmc_pble_rsrc) + sizeof(struct i40iw_hmc_info) + (sizeof(struct i40iw_hmc_obj_info) * I40IW_HMC_IW_MAX); iwdev->hmc_info_mem = kzalloc(size, GFP_KERNEL); - if (!iwdev->hmc_info_mem) { - i40iw_pr_err("memory alloc fail\n"); + if (!iwdev->hmc_info_mem) return I40IW_ERR_NO_MEMORY; - } + iwdev->pble_rsrc = (struct i40iw_hmc_pble_rsrc *)iwdev->hmc_info_mem; dev->hmc_info = &iwdev->hw.hmc; dev->hmc_info->hmc_obj = (struct i40iw_hmc_obj_info *)(iwdev->pble_rsrc + 1); @@ -1325,7 +1333,17 @@ static enum i40iw_status_code i40iw_initialize_dev(struct i40iw_device *iwdev, info.bar0 = ldev->hw_addr; info.hw = &iwdev->hw; info.debug_mask = debug; - info.qs_handle = ldev->params.qos.prio_qos[0].qs_handle; + l2params.mss = + (ldev->params.mtu) ? ldev->params.mtu - I40IW_MTU_TO_MSS : I40IW_DEFAULT_MSS; + for (i = 0; i < I40E_CLIENT_MAX_USER_PRIORITY; i++) { + qset = ldev->params.qos.prio_qos[i].qs_handle; + l2params.qs_handle_list[i] = qset; + if (last_qset == I40IW_NO_QSET) + last_qset = qset; + else if ((qset != last_qset) && (qset != I40IW_NO_QSET)) + iwdev->dcb = true; + } + i40iw_pr_info("DCB is set/clear = %d\n", iwdev->dcb); info.exception_lan_queue = 1; info.vchnl_send = i40iw_virtchnl_send; status = i40iw_device_init(&iwdev->sc_dev, &info); @@ -1334,6 +1352,20 @@ exit: kfree(iwdev->hmc_info_mem); iwdev->hmc_info_mem = NULL; } + memset(&vsi_info, 0, sizeof(vsi_info)); + vsi_info.dev = &iwdev->sc_dev; + vsi_info.back_vsi = (void *)iwdev; + vsi_info.params = &l2params; + i40iw_sc_vsi_init(&iwdev->vsi, &vsi_info); + + if (dev->is_pf) { + memset(&stats_info, 0, sizeof(stats_info)); + stats_info.fcn_id = ldev->fid; + stats_info.pestat = kzalloc(sizeof(*stats_info.pestat), GFP_KERNEL); + stats_info.stats_initialize = true; + if (stats_info.pestat) + i40iw_vsi_stats_init(&iwdev->vsi, &stats_info); + } return status; } @@ -1384,6 +1416,7 @@ static enum i40iw_status_code i40iw_save_msix_info(struct i40iw_device *iwdev, for (i = 0, ceq_idx = 0; i < iwdev->msix_count; i++, iw_qvinfo++) { iwdev->iw_msixtbl[i].idx = ldev->msix_entries[i].entry; iwdev->iw_msixtbl[i].irq = ldev->msix_entries[i].vector; + iwdev->iw_msixtbl[i].cpu_affinity = ceq_idx; if (i == 0) { iw_qvinfo->aeq_idx = 0; if (iwdev->msix_shared) @@ -1404,18 +1437,19 @@ static enum i40iw_status_code i40iw_save_msix_info(struct i40iw_device *iwdev, * i40iw_deinit_device - clean up the device resources * @iwdev: iwarp device * @reset: true if called before reset - * @del_hdl: true if delete hdl entry * * Destroy the ib device interface, remove the mac ip entry and ipv4/ipv6 addresses, * destroy the device queues and free the pble and the hmc objects */ -static void i40iw_deinit_device(struct i40iw_device *iwdev, bool reset, bool del_hdl) +static void i40iw_deinit_device(struct i40iw_device *iwdev, bool reset) { struct i40e_info *ldev = iwdev->ldev; struct i40iw_sc_dev *dev = &iwdev->sc_dev; i40iw_pr_info("state = %d\n", iwdev->init_state); + if (iwdev->param_wq) + destroy_workqueue(iwdev->param_wq); switch (iwdev->init_state) { case RDMA_DEV_REGISTERED: @@ -1441,10 +1475,10 @@ static void i40iw_deinit_device(struct i40iw_device *iwdev, bool reset, bool del i40iw_destroy_aeq(iwdev, reset); /* fallthrough */ case IEQ_CREATED: - i40iw_puda_dele_resources(dev, I40IW_PUDA_RSRC_TYPE_IEQ, reset); + i40iw_puda_dele_resources(&iwdev->vsi, I40IW_PUDA_RSRC_TYPE_IEQ, reset); /* fallthrough */ case ILQ_CREATED: - i40iw_puda_dele_resources(dev, I40IW_PUDA_RSRC_TYPE_ILQ, reset); + i40iw_puda_dele_resources(&iwdev->vsi, I40IW_PUDA_RSRC_TYPE_ILQ, reset); /* fallthrough */ case CCQ_CREATED: i40iw_destroy_ccq(iwdev, reset); @@ -1456,13 +1490,14 @@ static void i40iw_deinit_device(struct i40iw_device *iwdev, bool reset, bool del i40iw_del_hmc_objects(dev, dev->hmc_info, true, reset); /* fallthrough */ case CQP_CREATED: - i40iw_destroy_cqp(iwdev, !reset); + i40iw_destroy_cqp(iwdev, true); /* fallthrough */ case INITIAL_STATE: i40iw_cleanup_cm_core(&iwdev->cm_core); - if (dev->is_pf) - i40iw_hw_stats_del_timer(dev); - + if (iwdev->vsi.pestat) { + i40iw_vsi_stats_free(&iwdev->vsi); + kfree(iwdev->vsi.pestat); + } i40iw_del_init_mem(iwdev); break; case INVALID_STATE: @@ -1472,8 +1507,7 @@ static void i40iw_deinit_device(struct i40iw_device *iwdev, bool reset, bool del break; } - if (del_hdl) - i40iw_del_handler(i40iw_find_i40e_handler(ldev)); + i40iw_del_handler(i40iw_find_i40e_handler(ldev)); kfree(iwdev->hdl); } @@ -1508,7 +1542,6 @@ static enum i40iw_status_code i40iw_setup_init_state(struct i40iw_handler *hdl, iwdev->max_enabled_vfs = iwdev->max_rdma_vfs; iwdev->netdev = ldev->netdev; hdl->client = client; - iwdev->mss = (!ldev->params.mtu) ? I40IW_DEFAULT_MSS : ldev->params.mtu - I40IW_MTU_TO_MSS; if (!ldev->ftype) iwdev->db_start = pci_resource_start(ldev->pcidev, 0) + I40IW_DB_ADDR_OFFSET; else @@ -1528,6 +1561,7 @@ static enum i40iw_status_code i40iw_setup_init_state(struct i40iw_handler *hdl, init_waitqueue_head(&iwdev->vchnl_waitq); init_waitqueue_head(&dev->vf_reqs); + init_waitqueue_head(&iwdev->close_wq); status = i40iw_initialize_dev(iwdev, ldev); exit: @@ -1540,6 +1574,20 @@ exit: } /** + * i40iw_get_used_rsrc - determine resources used internally + * @iwdev: iwarp device + * + * Called after internal allocations + */ +static void i40iw_get_used_rsrc(struct i40iw_device *iwdev) +{ + iwdev->used_pds = find_next_zero_bit(iwdev->allocated_pds, iwdev->max_pd, 0); + iwdev->used_qps = find_next_zero_bit(iwdev->allocated_qps, iwdev->max_qp, 0); + iwdev->used_cqs = find_next_zero_bit(iwdev->allocated_cqs, iwdev->max_cq, 0); + iwdev->used_mrs = find_next_zero_bit(iwdev->allocated_mrs, iwdev->max_mr, 0); +} + +/** * i40iw_open - client interface operation open for iwarp/uda device * @ldev: lan device information * @client: iwarp client information, provided during registration @@ -1611,6 +1659,7 @@ static int i40iw_open(struct i40e_info *ldev, struct i40e_client *client) status = i40iw_initialize_hw_resources(iwdev); if (status) break; + i40iw_get_used_rsrc(iwdev); dev->ccq_ops->ccq_arm(dev->ccq); status = i40iw_hmc_init_pble(&iwdev->sc_dev, iwdev->pble_rsrc); if (status) @@ -1630,35 +1679,73 @@ static int i40iw_open(struct i40e_info *ldev, struct i40e_client *client) iwdev->init_state = RDMA_DEV_REGISTERED; iwdev->iw_status = 1; i40iw_port_ibevent(iwdev); + iwdev->param_wq = alloc_ordered_workqueue("l2params", WQ_MEM_RECLAIM); + if(iwdev->param_wq == NULL) + break; i40iw_pr_info("i40iw_open completed\n"); return 0; } while (0); i40iw_pr_err("status = %d last completion = %d\n", status, iwdev->init_state); - i40iw_deinit_device(iwdev, false, false); + i40iw_deinit_device(iwdev, false); return -ERESTART; } /** - * i40iw_l2param_change : handle qs handles for qos and mss change + * i40iw_l2params_worker - worker for l2 params change + * @work: work pointer for l2 params + */ +static void i40iw_l2params_worker(struct work_struct *work) +{ + struct l2params_work *dwork = + container_of(work, struct l2params_work, work); + struct i40iw_device *iwdev = dwork->iwdev; + + i40iw_change_l2params(&iwdev->vsi, &dwork->l2params); + atomic_dec(&iwdev->params_busy); + kfree(work); +} + +/** + * i40iw_l2param_change - handle qs handles for qos and mss change * @ldev: lan device information * @client: client for paramater change * @params: new parameters from L2 */ -static void i40iw_l2param_change(struct i40e_info *ldev, - struct i40e_client *client, +static void i40iw_l2param_change(struct i40e_info *ldev, struct i40e_client *client, struct i40e_params *params) { struct i40iw_handler *hdl; + struct i40iw_l2params *l2params; + struct l2params_work *work; struct i40iw_device *iwdev; + int i; hdl = i40iw_find_i40e_handler(ldev); if (!hdl) return; iwdev = &hdl->device; - if (params->mtu) - iwdev->mss = params->mtu - I40IW_MTU_TO_MSS; + + if (atomic_read(&iwdev->params_busy)) + return; + + + work = kzalloc(sizeof(*work), GFP_ATOMIC); + if (!work) + return; + + atomic_inc(&iwdev->params_busy); + + work->iwdev = iwdev; + l2params = &work->l2params; + for (i = 0; i < I40E_CLIENT_MAX_USER_PRIORITY; i++) + l2params->qs_handle_list[i] = params->qos.prio_qos[i].qs_handle; + + l2params->mss = (params->mtu) ? params->mtu - I40IW_MTU_TO_MSS : iwdev->vsi.mss; + + INIT_WORK(&work->work, i40iw_l2params_worker); + queue_work(iwdev->param_wq, &work->work); } /** @@ -1679,8 +1766,11 @@ static void i40iw_close(struct i40e_info *ldev, struct i40e_client *client, bool return; iwdev = &hdl->device; + iwdev->closing = true; + + i40iw_cm_disconnect_all(iwdev); destroy_workqueue(iwdev->virtchnl_wq); - i40iw_deinit_device(iwdev, reset, true); + i40iw_deinit_device(iwdev, reset); } /** @@ -1701,21 +1791,23 @@ static void i40iw_vf_reset(struct i40e_info *ldev, struct i40e_client *client, u struct i40iw_vfdev *tmp_vfdev; unsigned int i; unsigned long flags; + struct i40iw_device *iwdev; hdl = i40iw_find_i40e_handler(ldev); if (!hdl) return; dev = &hdl->device.sc_dev; + iwdev = (struct i40iw_device *)dev->back_dev; for (i = 0; i < I40IW_MAX_PE_ENABLED_VF_COUNT; i++) { if (!dev->vf_dev[i] || (dev->vf_dev[i]->vf_id != vf_id)) continue; /* free all resources allocated on behalf of vf */ tmp_vfdev = dev->vf_dev[i]; - spin_lock_irqsave(&dev->dev_pestat.stats_lock, flags); + spin_lock_irqsave(&iwdev->vsi.pestat->lock, flags); dev->vf_dev[i] = NULL; - spin_unlock_irqrestore(&dev->dev_pestat.stats_lock, flags); + spin_unlock_irqrestore(&iwdev->vsi.pestat->lock, flags); i40iw_del_hmc_objects(dev, &tmp_vfdev->hmc_info, false, false); /* remove vf hmc function */ memset(&hmc_fcn_info, 0, sizeof(hmc_fcn_info)); diff --git a/drivers/infiniband/hw/i40iw/i40iw_osdep.h b/drivers/infiniband/hw/i40iw/i40iw_osdep.h index 80f422bf3967..aa66c1c63dfa 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_osdep.h +++ b/drivers/infiniband/hw/i40iw/i40iw_osdep.h @@ -198,6 +198,8 @@ enum i40iw_status_code i40iw_cqp_manage_vf_pble_bp(struct i40iw_sc_dev *dev, void i40iw_cqp_spawn_worker(struct i40iw_sc_dev *dev, struct i40iw_virtchnl_work_info *work_info, u32 iw_vf_idx); void *i40iw_remove_head(struct list_head *list); +void i40iw_qp_suspend_resume(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp, bool suspend); +void i40iw_qp_mss_modify(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp); void i40iw_term_modify_qp(struct i40iw_sc_qp *qp, u8 next_state, u8 term, u8 term_len); void i40iw_terminate_done(struct i40iw_sc_qp *qp, int timeout_occurred); @@ -207,9 +209,9 @@ void i40iw_terminate_del_timer(struct i40iw_sc_qp *qp); enum i40iw_status_code i40iw_hw_manage_vf_pble_bp(struct i40iw_device *iwdev, struct i40iw_manage_vf_pble_info *info, bool wait); -struct i40iw_dev_pestat; -void i40iw_hw_stats_start_timer(struct i40iw_sc_dev *); -void i40iw_hw_stats_del_timer(struct i40iw_sc_dev *); +struct i40iw_sc_vsi; +void i40iw_hw_stats_start_timer(struct i40iw_sc_vsi *vsi); +void i40iw_hw_stats_stop_timer(struct i40iw_sc_vsi *vsi); #define i40iw_mmiowb() mmiowb() void i40iw_wr32(struct i40iw_hw *hw, u32 reg, u32 value); u32 i40iw_rd32(struct i40iw_hw *hw, u32 reg); diff --git a/drivers/infiniband/hw/i40iw/i40iw_p.h b/drivers/infiniband/hw/i40iw/i40iw_p.h index a0b8ca10d67e..28a92fee0822 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_p.h +++ b/drivers/infiniband/hw/i40iw/i40iw_p.h @@ -47,8 +47,6 @@ void i40iw_debug_buf(struct i40iw_sc_dev *dev, enum i40iw_debug_flag mask, enum i40iw_status_code i40iw_device_init(struct i40iw_sc_dev *dev, struct i40iw_device_init_info *info); -enum i40iw_status_code i40iw_device_init_pestat(struct i40iw_dev_pestat *); - void i40iw_sc_cqp_post_sq(struct i40iw_sc_cqp *cqp); u64 *i40iw_sc_cqp_get_next_send_wqe(struct i40iw_sc_cqp *cqp, u64 scratch); @@ -64,7 +62,24 @@ enum i40iw_status_code i40iw_sc_init_iw_hmc(struct i40iw_sc_dev *dev, enum i40iw_status_code i40iw_pf_init_vfhmc(struct i40iw_sc_dev *dev, u8 vf_hmc_fn_id, u32 *vf_cnt_array); -/* cqp misc functions */ +/* stats functions */ +void i40iw_hw_stats_refresh_all(struct i40iw_vsi_pestat *stats); +void i40iw_hw_stats_read_all(struct i40iw_vsi_pestat *stats, struct i40iw_dev_hw_stats *stats_values); +void i40iw_hw_stats_read_32(struct i40iw_vsi_pestat *stats, + enum i40iw_hw_stats_index_32b index, + u64 *value); +void i40iw_hw_stats_read_64(struct i40iw_vsi_pestat *stats, + enum i40iw_hw_stats_index_64b index, + u64 *value); +void i40iw_hw_stats_init(struct i40iw_vsi_pestat *stats, u8 index, bool is_pf); + +/* vsi misc functions */ +enum i40iw_status_code i40iw_vsi_stats_init(struct i40iw_sc_vsi *vsi, struct i40iw_vsi_stats_info *info); +void i40iw_vsi_stats_free(struct i40iw_sc_vsi *vsi); +void i40iw_sc_vsi_init(struct i40iw_sc_vsi *vsi, struct i40iw_vsi_init_info *info); + +void i40iw_change_l2params(struct i40iw_sc_vsi *vsi, struct i40iw_l2params *l2params); +void i40iw_qp_add_qos(struct i40iw_sc_qp *qp); void i40iw_terminate_send_fin(struct i40iw_sc_qp *qp); diff --git a/drivers/infiniband/hw/i40iw/i40iw_pble.c b/drivers/infiniband/hw/i40iw/i40iw_pble.c index 85993dc44f6e..c87ba1617087 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_pble.c +++ b/drivers/infiniband/hw/i40iw/i40iw_pble.c @@ -353,10 +353,6 @@ static enum i40iw_status_code add_pble_pool(struct i40iw_sc_dev *dev, pages = (idx->rel_pd_idx) ? (I40IW_HMC_PD_CNT_IN_SD - idx->rel_pd_idx) : I40IW_HMC_PD_CNT_IN_SD; pages = min(pages, pble_rsrc->unallocated_pble >> PBLE_512_SHIFT); - if (!pages) { - ret_code = I40IW_ERR_NO_PBLCHUNKS_AVAILABLE; - goto error; - } info.chunk = chunk; info.hmc_info = hmc_info; info.pages = pages; diff --git a/drivers/infiniband/hw/i40iw/i40iw_puda.c b/drivers/infiniband/hw/i40iw/i40iw_puda.c index c62d354f7810..db41ab40da9c 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_puda.c +++ b/drivers/infiniband/hw/i40iw/i40iw_puda.c @@ -42,12 +42,13 @@ #include "i40iw_p.h" #include "i40iw_puda.h" -static void i40iw_ieq_receive(struct i40iw_sc_dev *dev, +static void i40iw_ieq_receive(struct i40iw_sc_vsi *vsi, struct i40iw_puda_buf *buf); -static void i40iw_ieq_tx_compl(struct i40iw_sc_dev *dev, void *sqwrid); +static void i40iw_ieq_tx_compl(struct i40iw_sc_vsi *vsi, void *sqwrid); static void i40iw_ilq_putback_rcvbuf(struct i40iw_sc_qp *qp, u32 wqe_idx); static enum i40iw_status_code i40iw_puda_replenish_rq(struct i40iw_puda_rsrc *rsrc, bool initial); +static void i40iw_ieq_cleanup_qp(struct i40iw_puda_rsrc *ieq, struct i40iw_sc_qp *qp); /** * i40iw_puda_get_listbuf - get buffer from puda list * @list: list to use for buffers (ILQ or IEQ) @@ -292,7 +293,7 @@ enum i40iw_status_code i40iw_puda_poll_completion(struct i40iw_sc_dev *dev, unsigned long flags; if ((cq_type == I40IW_CQ_TYPE_ILQ) || (cq_type == I40IW_CQ_TYPE_IEQ)) { - rsrc = (cq_type == I40IW_CQ_TYPE_ILQ) ? dev->ilq : dev->ieq; + rsrc = (cq_type == I40IW_CQ_TYPE_ILQ) ? cq->vsi->ilq : cq->vsi->ieq; } else { i40iw_debug(dev, I40IW_DEBUG_PUDA, "%s qp_type error\n", __func__); return I40IW_ERR_BAD_PTR; @@ -335,7 +336,7 @@ enum i40iw_status_code i40iw_puda_poll_completion(struct i40iw_sc_dev *dev, rsrc->stats_pkt_rcvd++; rsrc->compl_rxwqe_idx = info.wqe_idx; i40iw_debug(dev, I40IW_DEBUG_PUDA, "%s RQ completion\n", __func__); - rsrc->receive(rsrc->dev, buf); + rsrc->receive(rsrc->vsi, buf); if (cq_type == I40IW_CQ_TYPE_ILQ) i40iw_ilq_putback_rcvbuf(&rsrc->qp, info.wqe_idx); else @@ -345,12 +346,12 @@ enum i40iw_status_code i40iw_puda_poll_completion(struct i40iw_sc_dev *dev, i40iw_debug(dev, I40IW_DEBUG_PUDA, "%s SQ completion\n", __func__); sqwrid = (void *)(uintptr_t)qp->sq_wrtrk_array[info.wqe_idx].wrid; I40IW_RING_SET_TAIL(qp->sq_ring, info.wqe_idx); - rsrc->xmit_complete(rsrc->dev, sqwrid); + rsrc->xmit_complete(rsrc->vsi, sqwrid); spin_lock_irqsave(&rsrc->bufpool_lock, flags); rsrc->tx_wqe_avail_cnt++; spin_unlock_irqrestore(&rsrc->bufpool_lock, flags); - if (!list_empty(&dev->ilq->txpend)) - i40iw_puda_send_buf(dev->ilq, NULL); + if (!list_empty(&rsrc->vsi->ilq->txpend)) + i40iw_puda_send_buf(rsrc->vsi->ilq, NULL); } done: @@ -513,10 +514,8 @@ static void i40iw_puda_qp_setctx(struct i40iw_puda_rsrc *rsrc) * i40iw_puda_qp_wqe - setup wqe for qp create * @rsrc: resource for qp */ -static enum i40iw_status_code i40iw_puda_qp_wqe(struct i40iw_puda_rsrc *rsrc) +static enum i40iw_status_code i40iw_puda_qp_wqe(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp) { - struct i40iw_sc_qp *qp = &rsrc->qp; - struct i40iw_sc_dev *dev = rsrc->dev; struct i40iw_sc_cqp *cqp; u64 *wqe; u64 header; @@ -582,6 +581,7 @@ static enum i40iw_status_code i40iw_puda_qp_create(struct i40iw_puda_rsrc *rsrc) qp->back_qp = (void *)rsrc; qp->sq_pa = mem->pa; qp->rq_pa = qp->sq_pa + sq_size; + qp->vsi = rsrc->vsi; ukqp->sq_base = mem->va; ukqp->rq_base = &ukqp->sq_base[rsrc->sq_size]; ukqp->shadow_area = ukqp->rq_base[rsrc->rq_size].elem; @@ -608,15 +608,63 @@ static enum i40iw_status_code i40iw_puda_qp_create(struct i40iw_puda_rsrc *rsrc) ukqp->wqe_alloc_reg = (u32 __iomem *)(i40iw_get_hw_addr(qp->pd->dev) + I40E_VFPE_WQEALLOC1); - qp->qs_handle = qp->dev->qs_handle; + qp->user_pri = 0; + i40iw_qp_add_qos(qp); i40iw_puda_qp_setctx(rsrc); - ret = i40iw_puda_qp_wqe(rsrc); + if (rsrc->ceq_valid) + ret = i40iw_cqp_qp_create_cmd(rsrc->dev, qp); + else + ret = i40iw_puda_qp_wqe(rsrc->dev, qp); if (ret) i40iw_free_dma_mem(rsrc->dev->hw, &rsrc->qpmem); return ret; } /** + * i40iw_puda_cq_wqe - setup wqe for cq create + * @rsrc: resource for cq + */ +static enum i40iw_status_code i40iw_puda_cq_wqe(struct i40iw_sc_dev *dev, struct i40iw_sc_cq *cq) +{ + u64 *wqe; + struct i40iw_sc_cqp *cqp; + u64 header; + struct i40iw_ccq_cqe_info compl_info; + enum i40iw_status_code status = 0; + + cqp = dev->cqp; + wqe = i40iw_sc_cqp_get_next_send_wqe(cqp, 0); + if (!wqe) + return I40IW_ERR_RING_FULL; + + set_64bit_val(wqe, 0, cq->cq_uk.cq_size); + set_64bit_val(wqe, 8, RS_64_1(cq, 1)); + set_64bit_val(wqe, 16, + LS_64(cq->shadow_read_threshold, + I40IW_CQPSQ_CQ_SHADOW_READ_THRESHOLD)); + set_64bit_val(wqe, 32, cq->cq_pa); + + set_64bit_val(wqe, 40, cq->shadow_area_pa); + + header = cq->cq_uk.cq_id | + LS_64(I40IW_CQP_OP_CREATE_CQ, I40IW_CQPSQ_OPCODE) | + LS_64(1, I40IW_CQPSQ_CQ_CHKOVERFLOW) | + LS_64(1, I40IW_CQPSQ_CQ_ENCEQEMASK) | + LS_64(1, I40IW_CQPSQ_CQ_CEQIDVALID) | + LS_64(cqp->polarity, I40IW_CQPSQ_WQEVALID); + set_64bit_val(wqe, 24, header); + + i40iw_debug_buf(dev, I40IW_DEBUG_PUDA, "PUDA CQE", + wqe, I40IW_CQP_WQE_SIZE * 8); + + i40iw_sc_cqp_post_sq(dev->cqp); + status = dev->cqp_ops->poll_for_cqp_op_done(dev->cqp, + I40IW_CQP_OP_CREATE_CQ, + &compl_info); + return status; +} + +/** * i40iw_puda_cq_create - create cq for resource * @rsrc: resource for which cq to create */ @@ -624,18 +672,13 @@ static enum i40iw_status_code i40iw_puda_cq_create(struct i40iw_puda_rsrc *rsrc) { struct i40iw_sc_dev *dev = rsrc->dev; struct i40iw_sc_cq *cq = &rsrc->cq; - u64 *wqe; - struct i40iw_sc_cqp *cqp; - u64 header; enum i40iw_status_code ret = 0; u32 tsize, cqsize; - u32 shadow_read_threshold = 128; struct i40iw_dma_mem *mem; - struct i40iw_ccq_cqe_info compl_info; struct i40iw_cq_init_info info; struct i40iw_cq_uk_init_info *init_info = &info.cq_uk_init_info; - cq->back_cq = (void *)rsrc; + cq->vsi = rsrc->vsi; cqsize = rsrc->cq_size * (sizeof(struct i40iw_cqe)); tsize = cqsize + sizeof(struct i40iw_cq_shadow_area); ret = i40iw_allocate_dma_mem(dev->hw, &rsrc->cqmem, tsize, @@ -656,43 +699,84 @@ static enum i40iw_status_code i40iw_puda_cq_create(struct i40iw_puda_rsrc *rsrc) init_info->shadow_area = (u64 *)((u8 *)mem->va + cqsize); init_info->cq_size = rsrc->cq_size; init_info->cq_id = rsrc->cq_id; + info.ceqe_mask = true; + info.ceq_id_valid = true; ret = dev->iw_priv_cq_ops->cq_init(cq, &info); if (ret) goto error; - cqp = dev->cqp; - wqe = i40iw_sc_cqp_get_next_send_wqe(cqp, 0); - if (!wqe) { - ret = I40IW_ERR_RING_FULL; - goto error; - } + if (rsrc->ceq_valid) + ret = i40iw_cqp_cq_create_cmd(dev, cq); + else + ret = i40iw_puda_cq_wqe(dev, cq); +error: + if (ret) + i40iw_free_dma_mem(dev->hw, &rsrc->cqmem); + return ret; +} - set_64bit_val(wqe, 0, rsrc->cq_size); - set_64bit_val(wqe, 8, RS_64_1(cq, 1)); - set_64bit_val(wqe, 16, LS_64(shadow_read_threshold, I40IW_CQPSQ_CQ_SHADOW_READ_THRESHOLD)); - set_64bit_val(wqe, 32, cq->cq_pa); +/** + * i40iw_puda_free_qp - free qp for resource + * @rsrc: resource for which qp to free + */ +static void i40iw_puda_free_qp(struct i40iw_puda_rsrc *rsrc) +{ + enum i40iw_status_code ret; + struct i40iw_ccq_cqe_info compl_info; + struct i40iw_sc_dev *dev = rsrc->dev; - set_64bit_val(wqe, 40, cq->shadow_area_pa); + if (rsrc->ceq_valid) { + i40iw_cqp_qp_destroy_cmd(dev, &rsrc->qp); + return; + } - header = rsrc->cq_id | - LS_64(I40IW_CQP_OP_CREATE_CQ, I40IW_CQPSQ_OPCODE) | - LS_64(1, I40IW_CQPSQ_CQ_CHKOVERFLOW) | - LS_64(1, I40IW_CQPSQ_CQ_ENCEQEMASK) | - LS_64(1, I40IW_CQPSQ_CQ_CEQIDVALID) | - LS_64(cqp->polarity, I40IW_CQPSQ_WQEVALID); - set_64bit_val(wqe, 24, header); + ret = dev->iw_priv_qp_ops->qp_destroy(&rsrc->qp, + 0, false, true, true); + if (ret) + i40iw_debug(dev, I40IW_DEBUG_PUDA, + "%s error puda qp destroy wqe\n", + __func__); - i40iw_debug_buf(dev, I40IW_DEBUG_PUDA, "PUDA CQE", - wqe, I40IW_CQP_WQE_SIZE * 8); + if (!ret) { + ret = dev->cqp_ops->poll_for_cqp_op_done(dev->cqp, + I40IW_CQP_OP_DESTROY_QP, + &compl_info); + if (ret) + i40iw_debug(dev, I40IW_DEBUG_PUDA, + "%s error puda qp destroy failed\n", + __func__); + } +} - i40iw_sc_cqp_post_sq(dev->cqp); - ret = dev->cqp_ops->poll_for_cqp_op_done(dev->cqp, - I40IW_CQP_OP_CREATE_CQ, - &compl_info); +/** + * i40iw_puda_free_cq - free cq for resource + * @rsrc: resource for which cq to free + */ +static void i40iw_puda_free_cq(struct i40iw_puda_rsrc *rsrc) +{ + enum i40iw_status_code ret; + struct i40iw_ccq_cqe_info compl_info; + struct i40iw_sc_dev *dev = rsrc->dev; + + if (rsrc->ceq_valid) { + i40iw_cqp_cq_destroy_cmd(dev, &rsrc->cq); + return; + } + ret = dev->iw_priv_cq_ops->cq_destroy(&rsrc->cq, 0, true); -error: if (ret) - i40iw_free_dma_mem(dev->hw, &rsrc->cqmem); - return ret; + i40iw_debug(dev, I40IW_DEBUG_PUDA, + "%s error ieq cq destroy\n", + __func__); + + if (!ret) { + ret = dev->cqp_ops->poll_for_cqp_op_done(dev->cqp, + I40IW_CQP_OP_DESTROY_CQ, + &compl_info); + if (ret) + i40iw_debug(dev, I40IW_DEBUG_PUDA, + "%s error ieq qp destroy done\n", + __func__); + } } /** @@ -701,25 +785,24 @@ error: * @type: type of resource to dele * @reset: true if reset chip */ -void i40iw_puda_dele_resources(struct i40iw_sc_dev *dev, +void i40iw_puda_dele_resources(struct i40iw_sc_vsi *vsi, enum puda_resource_type type, bool reset) { - struct i40iw_ccq_cqe_info compl_info; + struct i40iw_sc_dev *dev = vsi->dev; struct i40iw_puda_rsrc *rsrc; struct i40iw_puda_buf *buf = NULL; struct i40iw_puda_buf *nextbuf = NULL; struct i40iw_virt_mem *vmem; - enum i40iw_status_code ret; switch (type) { case I40IW_PUDA_RSRC_TYPE_ILQ: - rsrc = dev->ilq; - vmem = &dev->ilq_mem; + rsrc = vsi->ilq; + vmem = &vsi->ilq_mem; break; case I40IW_PUDA_RSRC_TYPE_IEQ: - rsrc = dev->ieq; - vmem = &dev->ieq_mem; + rsrc = vsi->ieq; + vmem = &vsi->ieq_mem; break; default: i40iw_debug(dev, I40IW_DEBUG_PUDA, "%s: error resource type = 0x%x\n", @@ -731,45 +814,14 @@ void i40iw_puda_dele_resources(struct i40iw_sc_dev *dev, case PUDA_HASH_CRC_COMPLETE: i40iw_free_hash_desc(rsrc->hash_desc); case PUDA_QP_CREATED: - do { - if (reset) - break; - ret = dev->iw_priv_qp_ops->qp_destroy(&rsrc->qp, - 0, false, true, true); - if (ret) - i40iw_debug(rsrc->dev, I40IW_DEBUG_PUDA, - "%s error ieq qp destroy\n", - __func__); - - ret = dev->cqp_ops->poll_for_cqp_op_done(dev->cqp, - I40IW_CQP_OP_DESTROY_QP, - &compl_info); - if (ret) - i40iw_debug(rsrc->dev, I40IW_DEBUG_PUDA, - "%s error ieq qp destroy done\n", - __func__); - } while (0); + if (!reset) + i40iw_puda_free_qp(rsrc); i40iw_free_dma_mem(dev->hw, &rsrc->qpmem); /* fallthrough */ case PUDA_CQ_CREATED: - do { - if (reset) - break; - ret = dev->iw_priv_cq_ops->cq_destroy(&rsrc->cq, 0, true); - if (ret) - i40iw_debug(rsrc->dev, I40IW_DEBUG_PUDA, - "%s error ieq cq destroy\n", - __func__); - - ret = dev->cqp_ops->poll_for_cqp_op_done(dev->cqp, - I40IW_CQP_OP_DESTROY_CQ, - &compl_info); - if (ret) - i40iw_debug(rsrc->dev, I40IW_DEBUG_PUDA, - "%s error ieq qp destroy done\n", - __func__); - } while (0); + if (!reset) + i40iw_puda_free_cq(rsrc); i40iw_free_dma_mem(dev->hw, &rsrc->cqmem); break; @@ -825,9 +877,10 @@ static enum i40iw_status_code i40iw_puda_allocbufs(struct i40iw_puda_rsrc *rsrc, * @dev: iwarp device * @info: resource information */ -enum i40iw_status_code i40iw_puda_create_rsrc(struct i40iw_sc_dev *dev, +enum i40iw_status_code i40iw_puda_create_rsrc(struct i40iw_sc_vsi *vsi, struct i40iw_puda_rsrc_info *info) { + struct i40iw_sc_dev *dev = vsi->dev; enum i40iw_status_code ret = 0; struct i40iw_puda_rsrc *rsrc; u32 pudasize; @@ -840,10 +893,10 @@ enum i40iw_status_code i40iw_puda_create_rsrc(struct i40iw_sc_dev *dev, rqwridsize = info->rq_size * 8; switch (info->type) { case I40IW_PUDA_RSRC_TYPE_ILQ: - vmem = &dev->ilq_mem; + vmem = &vsi->ilq_mem; break; case I40IW_PUDA_RSRC_TYPE_IEQ: - vmem = &dev->ieq_mem; + vmem = &vsi->ieq_mem; break; default: return I40IW_NOT_SUPPORTED; @@ -856,28 +909,28 @@ enum i40iw_status_code i40iw_puda_create_rsrc(struct i40iw_sc_dev *dev, rsrc = (struct i40iw_puda_rsrc *)vmem->va; spin_lock_init(&rsrc->bufpool_lock); if (info->type == I40IW_PUDA_RSRC_TYPE_ILQ) { - dev->ilq = (struct i40iw_puda_rsrc *)vmem->va; - dev->ilq_count = info->count; + vsi->ilq = (struct i40iw_puda_rsrc *)vmem->va; + vsi->ilq_count = info->count; rsrc->receive = info->receive; rsrc->xmit_complete = info->xmit_complete; } else { - vmem = &dev->ieq_mem; - dev->ieq_count = info->count; - dev->ieq = (struct i40iw_puda_rsrc *)vmem->va; + vmem = &vsi->ieq_mem; + vsi->ieq_count = info->count; + vsi->ieq = (struct i40iw_puda_rsrc *)vmem->va; rsrc->receive = i40iw_ieq_receive; rsrc->xmit_complete = i40iw_ieq_tx_compl; } + rsrc->ceq_valid = info->ceq_valid; rsrc->type = info->type; rsrc->sq_wrtrk_array = (struct i40iw_sq_uk_wr_trk_info *)((u8 *)vmem->va + pudasize); rsrc->rq_wrid_array = (u64 *)((u8 *)vmem->va + pudasize + sqwridsize); - rsrc->mss = info->mss; /* Initialize all ieq lists */ INIT_LIST_HEAD(&rsrc->bufpool); INIT_LIST_HEAD(&rsrc->txpend); rsrc->tx_wqe_avail_cnt = info->sq_size - 1; - dev->iw_pd_ops->pd_init(dev, &rsrc->sc_pd, info->pd_id); + dev->iw_pd_ops->pd_init(dev, &rsrc->sc_pd, info->pd_id, -1); rsrc->qp_id = info->qp_id; rsrc->cq_id = info->cq_id; rsrc->sq_size = info->sq_size; @@ -885,6 +938,7 @@ enum i40iw_status_code i40iw_puda_create_rsrc(struct i40iw_sc_dev *dev, rsrc->cq_size = info->rq_size + info->sq_size; rsrc->buf_size = info->buf_size; rsrc->dev = dev; + rsrc->vsi = vsi; ret = i40iw_puda_cq_create(rsrc); if (!ret) { @@ -919,7 +973,7 @@ enum i40iw_status_code i40iw_puda_create_rsrc(struct i40iw_sc_dev *dev, dev->ccq_ops->ccq_arm(&rsrc->cq); return ret; error: - i40iw_puda_dele_resources(dev, info->type, false); + i40iw_puda_dele_resources(vsi, info->type, false); return ret; } @@ -1131,7 +1185,7 @@ static enum i40iw_status_code i40iw_ieq_handle_partial(struct i40iw_puda_rsrc *i list_add(&buf->list, &pbufl); status = i40iw_ieq_create_pbufl(pfpdu, rxlist, &pbufl, buf, fpdu_len); - if (!status) + if (status) goto error; txbuf = i40iw_puda_get_bufpool(ieq); @@ -1332,7 +1386,7 @@ static void i40iw_ieq_handle_exception(struct i40iw_puda_rsrc *ieq, } if (pfpdu->mode && (fps != pfpdu->fps)) { /* clean up qp as it is new partial sequence */ - i40iw_ieq_cleanup_qp(ieq->dev, qp); + i40iw_ieq_cleanup_qp(ieq, qp); i40iw_debug(ieq->dev, I40IW_DEBUG_IEQ, "%s: restarting new partial\n", __func__); pfpdu->mode = false; @@ -1344,7 +1398,7 @@ static void i40iw_ieq_handle_exception(struct i40iw_puda_rsrc *ieq, pfpdu->rcv_nxt = fps; pfpdu->fps = fps; pfpdu->mode = true; - pfpdu->max_fpdu_data = ieq->mss; + pfpdu->max_fpdu_data = ieq->vsi->mss; pfpdu->pmode_count++; INIT_LIST_HEAD(rxlist); i40iw_ieq_check_first_buf(buf, fps); @@ -1379,14 +1433,14 @@ static void i40iw_ieq_handle_exception(struct i40iw_puda_rsrc *ieq, * @dev: iwarp device * @buf: exception buffer received */ -static void i40iw_ieq_receive(struct i40iw_sc_dev *dev, +static void i40iw_ieq_receive(struct i40iw_sc_vsi *vsi, struct i40iw_puda_buf *buf) { - struct i40iw_puda_rsrc *ieq = dev->ieq; + struct i40iw_puda_rsrc *ieq = vsi->ieq; struct i40iw_sc_qp *qp = NULL; u32 wqe_idx = ieq->compl_rxwqe_idx; - qp = i40iw_ieq_get_qp(dev, buf); + qp = i40iw_ieq_get_qp(vsi->dev, buf); if (!qp) { ieq->stats_bad_qp_id++; i40iw_puda_ret_bufpool(ieq, buf); @@ -1404,12 +1458,12 @@ static void i40iw_ieq_receive(struct i40iw_sc_dev *dev, /** * i40iw_ieq_tx_compl - put back after sending completed exception buffer - * @dev: iwarp device + * @vsi: pointer to the vsi structure * @sqwrid: pointer to puda buffer */ -static void i40iw_ieq_tx_compl(struct i40iw_sc_dev *dev, void *sqwrid) +static void i40iw_ieq_tx_compl(struct i40iw_sc_vsi *vsi, void *sqwrid) { - struct i40iw_puda_rsrc *ieq = dev->ieq; + struct i40iw_puda_rsrc *ieq = vsi->ieq; struct i40iw_puda_buf *buf = (struct i40iw_puda_buf *)sqwrid; i40iw_puda_ret_bufpool(ieq, buf); @@ -1421,15 +1475,14 @@ static void i40iw_ieq_tx_compl(struct i40iw_sc_dev *dev, void *sqwrid) /** * i40iw_ieq_cleanup_qp - qp is being destroyed - * @dev: iwarp device + * @ieq: ieq resource * @qp: all pending fpdu buffers */ -void i40iw_ieq_cleanup_qp(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp) +static void i40iw_ieq_cleanup_qp(struct i40iw_puda_rsrc *ieq, struct i40iw_sc_qp *qp) { struct i40iw_puda_buf *buf; struct i40iw_pfpdu *pfpdu = &qp->pfpdu; struct list_head *rxlist = &pfpdu->rxlist; - struct i40iw_puda_rsrc *ieq = dev->ieq; if (!pfpdu->mode) return; diff --git a/drivers/infiniband/hw/i40iw/i40iw_puda.h b/drivers/infiniband/hw/i40iw/i40iw_puda.h index 52bf7826ce4e..dba05ce7d392 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_puda.h +++ b/drivers/infiniband/hw/i40iw/i40iw_puda.h @@ -100,6 +100,7 @@ struct i40iw_puda_rsrc_info { enum puda_resource_type type; /* ILQ or IEQ */ u32 count; u16 pd_id; + bool ceq_valid; u32 cq_id; u32 qp_id; u32 sq_size; @@ -107,8 +108,8 @@ struct i40iw_puda_rsrc_info { u16 buf_size; u16 mss; u32 tx_buf_cnt; /* total bufs allocated will be rq_size + tx_buf_cnt */ - void (*receive)(struct i40iw_sc_dev *, struct i40iw_puda_buf *); - void (*xmit_complete)(struct i40iw_sc_dev *, void *); + void (*receive)(struct i40iw_sc_vsi *, struct i40iw_puda_buf *); + void (*xmit_complete)(struct i40iw_sc_vsi *, void *); }; struct i40iw_puda_rsrc { @@ -116,6 +117,7 @@ struct i40iw_puda_rsrc { struct i40iw_sc_qp qp; struct i40iw_sc_pd sc_pd; struct i40iw_sc_dev *dev; + struct i40iw_sc_vsi *vsi; struct i40iw_dma_mem cqmem; struct i40iw_dma_mem qpmem; struct i40iw_virt_mem ilq_mem; @@ -123,6 +125,7 @@ struct i40iw_puda_rsrc { enum puda_resource_type type; u16 buf_size; /*buffer must be max datalen + tcpip hdr + mac */ u16 mss; + bool ceq_valid; u32 cq_id; u32 qp_id; u32 sq_size; @@ -142,8 +145,8 @@ struct i40iw_puda_rsrc { u32 avail_buf_count; /* snapshot of currently available buffers */ spinlock_t bufpool_lock; struct i40iw_puda_buf *alloclist; - void (*receive)(struct i40iw_sc_dev *, struct i40iw_puda_buf *); - void (*xmit_complete)(struct i40iw_sc_dev *, void *); + void (*receive)(struct i40iw_sc_vsi *, struct i40iw_puda_buf *); + void (*xmit_complete)(struct i40iw_sc_vsi *, void *); /* puda stats */ u64 stats_buf_alloc_fail; u64 stats_pkt_rcvd; @@ -160,14 +163,13 @@ void i40iw_puda_send_buf(struct i40iw_puda_rsrc *rsrc, struct i40iw_puda_buf *buf); enum i40iw_status_code i40iw_puda_send(struct i40iw_sc_qp *qp, struct i40iw_puda_send_info *info); -enum i40iw_status_code i40iw_puda_create_rsrc(struct i40iw_sc_dev *dev, +enum i40iw_status_code i40iw_puda_create_rsrc(struct i40iw_sc_vsi *vsi, struct i40iw_puda_rsrc_info *info); -void i40iw_puda_dele_resources(struct i40iw_sc_dev *dev, +void i40iw_puda_dele_resources(struct i40iw_sc_vsi *vsi, enum puda_resource_type type, bool reset); enum i40iw_status_code i40iw_puda_poll_completion(struct i40iw_sc_dev *dev, struct i40iw_sc_cq *cq, u32 *compl_err); -void i40iw_ieq_cleanup_qp(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp); struct i40iw_sc_qp *i40iw_ieq_get_qp(struct i40iw_sc_dev *dev, struct i40iw_puda_buf *buf); @@ -180,4 +182,8 @@ void i40iw_ieq_mpa_crc_ae(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp); void i40iw_free_hash_desc(struct shash_desc *desc); void i40iw_ieq_update_tcpip_info(struct i40iw_puda_buf *buf, u16 length, u32 seqnum); +enum i40iw_status_code i40iw_cqp_qp_create_cmd(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp); +enum i40iw_status_code i40iw_cqp_cq_create_cmd(struct i40iw_sc_dev *dev, struct i40iw_sc_cq *cq); +void i40iw_cqp_qp_destroy_cmd(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp); +void i40iw_cqp_cq_destroy_cmd(struct i40iw_sc_dev *dev, struct i40iw_sc_cq *cq); #endif diff --git a/drivers/infiniband/hw/i40iw/i40iw_type.h b/drivers/infiniband/hw/i40iw/i40iw_type.h index 2b1a04e9ca3c..7b76259752b0 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_type.h +++ b/drivers/infiniband/hw/i40iw/i40iw_type.h @@ -61,7 +61,7 @@ struct i40iw_cq_shadow_area { struct i40iw_sc_dev; struct i40iw_hmc_info; -struct i40iw_dev_pestat; +struct i40iw_vsi_pestat; struct i40iw_cqp_ops; struct i40iw_ccq_ops; @@ -74,6 +74,11 @@ struct i40iw_priv_qp_ops; struct i40iw_priv_cq_ops; struct i40iw_hmc_ops; +enum i40iw_page_size { + I40IW_PAGE_SIZE_4K, + I40IW_PAGE_SIZE_2M +}; + enum i40iw_resource_indicator_type { I40IW_RSRC_INDICATOR_TYPE_ADAPTER = 0, I40IW_RSRC_INDICATOR_TYPE_CQ, @@ -186,7 +191,7 @@ enum i40iw_debug_flag { I40IW_DEBUG_ALL = 0xFFFFFFFF }; -enum i40iw_hw_stat_index_32b { +enum i40iw_hw_stats_index_32b { I40IW_HW_STAT_INDEX_IP4RXDISCARD = 0, I40IW_HW_STAT_INDEX_IP4RXTRUNC, I40IW_HW_STAT_INDEX_IP4TXNOROUTE, @@ -199,7 +204,7 @@ enum i40iw_hw_stat_index_32b { I40IW_HW_STAT_INDEX_MAX_32 }; -enum i40iw_hw_stat_index_64b { +enum i40iw_hw_stats_index_64b { I40IW_HW_STAT_INDEX_IP4RXOCTS = 0, I40IW_HW_STAT_INDEX_IP4RXPKTS, I40IW_HW_STAT_INDEX_IP4RXFRAGS, @@ -229,32 +234,23 @@ enum i40iw_hw_stat_index_64b { I40IW_HW_STAT_INDEX_MAX_64 }; -struct i40iw_dev_hw_stat_offsets { - u32 stat_offset_32[I40IW_HW_STAT_INDEX_MAX_32]; - u32 stat_offset_64[I40IW_HW_STAT_INDEX_MAX_64]; +struct i40iw_dev_hw_stats_offsets { + u32 stats_offset_32[I40IW_HW_STAT_INDEX_MAX_32]; + u32 stats_offset_64[I40IW_HW_STAT_INDEX_MAX_64]; }; struct i40iw_dev_hw_stats { - u64 stat_value_32[I40IW_HW_STAT_INDEX_MAX_32]; - u64 stat_value_64[I40IW_HW_STAT_INDEX_MAX_64]; -}; - -struct i40iw_device_pestat_ops { - void (*iw_hw_stat_init)(struct i40iw_dev_pestat *, u8, struct i40iw_hw *, bool); - void (*iw_hw_stat_read_32)(struct i40iw_dev_pestat *, enum i40iw_hw_stat_index_32b, u64 *); - void (*iw_hw_stat_read_64)(struct i40iw_dev_pestat *, enum i40iw_hw_stat_index_64b, u64 *); - void (*iw_hw_stat_read_all)(struct i40iw_dev_pestat *, struct i40iw_dev_hw_stats *); - void (*iw_hw_stat_refresh_all)(struct i40iw_dev_pestat *); + u64 stats_value_32[I40IW_HW_STAT_INDEX_MAX_32]; + u64 stats_value_64[I40IW_HW_STAT_INDEX_MAX_64]; }; -struct i40iw_dev_pestat { +struct i40iw_vsi_pestat { struct i40iw_hw *hw; - struct i40iw_device_pestat_ops ops; struct i40iw_dev_hw_stats hw_stats; struct i40iw_dev_hw_stats last_read_hw_stats; - struct i40iw_dev_hw_stat_offsets hw_stat_offsets; + struct i40iw_dev_hw_stats_offsets hw_stats_offsets; struct timer_list stats_timer; - spinlock_t stats_lock; /* rdma stats lock */ + spinlock_t lock; /* rdma stats lock */ }; struct i40iw_hw { @@ -284,6 +280,7 @@ struct i40iw_sc_pd { u32 size; struct i40iw_sc_dev *dev; u16 pd_id; + int abi_ver; }; struct i40iw_cqp_quanta { @@ -350,6 +347,7 @@ struct i40iw_sc_cq { u64 cq_pa; u64 shadow_area_pa; struct i40iw_sc_dev *dev; + struct i40iw_sc_vsi *vsi; void *pbl_list; void *back_cq; u32 ceq_id; @@ -373,6 +371,7 @@ struct i40iw_sc_qp { u64 shadow_area_pa; u64 q2_pa; struct i40iw_sc_dev *dev; + struct i40iw_sc_vsi *vsi; struct i40iw_sc_pd *pd; u64 *hw_host_ctx; void *llp_stream_handle; @@ -397,6 +396,9 @@ struct i40iw_sc_qp { bool virtual_map; bool flush_sq; bool flush_rq; + u8 user_pri; + struct list_head list; + bool on_qoslist; bool sq_flush; enum i40iw_flush_opcode flush_code; enum i40iw_term_eventtypes eventtype; @@ -424,10 +426,16 @@ struct i40iw_vchnl_vf_msg_buffer { char parm_buffer[I40IW_VCHNL_MAX_VF_MSG_SIZE - 1]; }; +struct i40iw_qos { + struct list_head qplist; + spinlock_t lock; /* qos list */ + u16 qs_handle; +}; + struct i40iw_vfdev { struct i40iw_sc_dev *pf_dev; u8 *hmc_info_mem; - struct i40iw_dev_pestat dev_pestat; + struct i40iw_vsi_pestat pestat; struct i40iw_hmc_pble_info *pble_info; struct i40iw_hmc_info hmc_info; struct i40iw_vchnl_vf_msg_buffer vf_msg_buffer; @@ -441,11 +449,28 @@ struct i40iw_vfdev { bool stats_initialized; }; +#define I40IW_INVALID_FCN_ID 0xff +struct i40iw_sc_vsi { + struct i40iw_sc_dev *dev; + void *back_vsi; /* Owned by OS */ + u32 ilq_count; + struct i40iw_virt_mem ilq_mem; + struct i40iw_puda_rsrc *ilq; + u32 ieq_count; + struct i40iw_virt_mem ieq_mem; + struct i40iw_puda_rsrc *ieq; + u16 mss; + u8 fcn_id; + bool stats_fcn_id_alloc; + struct i40iw_qos qos[I40IW_MAX_USER_PRIORITY]; + struct i40iw_vsi_pestat *pestat; +}; + struct i40iw_sc_dev { struct list_head cqp_cmd_head; /* head of the CQP command list */ spinlock_t cqp_lock; /* cqp list sync */ struct i40iw_dev_uk dev_uk; - struct i40iw_dev_pestat dev_pestat; + bool fcn_id_array[I40IW_MAX_STATS_COUNT]; struct i40iw_dma_mem vf_fpm_query_buf[I40IW_MAX_PE_ENABLED_VF_COUNT]; u64 fpm_query_buf_pa; u64 fpm_commit_buf_pa; @@ -472,17 +497,9 @@ struct i40iw_sc_dev { struct i40iw_cqp_misc_ops *cqp_misc_ops; struct i40iw_hmc_ops *hmc_ops; struct i40iw_vchnl_if vchnl_if; - u32 ilq_count; - struct i40iw_virt_mem ilq_mem; - struct i40iw_puda_rsrc *ilq; - u32 ieq_count; - struct i40iw_virt_mem ieq_mem; - struct i40iw_puda_rsrc *ieq; - const struct i40iw_vf_cqp_ops *iw_vf_cqp_ops; struct i40iw_hmc_fpm_misc hmc_fpm_misc; - u16 qs_handle; u32 debug_mask; u16 exception_lan_queue; u8 hmc_fn_id; @@ -556,6 +573,19 @@ struct i40iw_l2params { u16 mss; }; +struct i40iw_vsi_init_info { + struct i40iw_sc_dev *dev; + void *back_vsi; + struct i40iw_l2params *params; +}; + +struct i40iw_vsi_stats_info { + struct i40iw_vsi_pestat *pestat; + u8 fcn_id; + bool alloc_fcn_id; + bool stats_initialize; +}; + struct i40iw_device_init_info { u64 fpm_query_buf_pa; u64 fpm_commit_buf_pa; @@ -564,7 +594,6 @@ struct i40iw_device_init_info { struct i40iw_hw *hw; void __iomem *bar0; enum i40iw_status_code (*vchnl_send)(struct i40iw_sc_dev *, u32, u8 *, u16); - u16 qs_handle; u16 exception_lan_queue; u8 hmc_fn_id; bool is_pf; @@ -722,6 +751,8 @@ struct i40iw_qp_host_ctx_info { bool iwarp_info_valid; bool err_rq_idx_valid; u16 err_rq_idx; + bool add_to_qoslist; + u8 user_pri; }; struct i40iw_aeqe_info { @@ -814,6 +845,7 @@ struct i40iw_register_shared_stag { struct i40iw_qp_init_info { struct i40iw_qp_uk_init_info qp_uk_init_info; struct i40iw_sc_pd *pd; + struct i40iw_sc_vsi *vsi; u64 *host_ctx; u8 *q2; u64 sq_pa; @@ -821,6 +853,7 @@ struct i40iw_qp_init_info { u64 host_ctx_pa; u64 q2_pa; u64 shadow_area_pa; + int abi_ver; u8 sq_tph_val; u8 rq_tph_val; u8 type; @@ -880,13 +913,14 @@ enum i40iw_quad_hash_manage_type { }; struct i40iw_qhash_table_info { + struct i40iw_sc_vsi *vsi; enum i40iw_quad_hash_manage_type manage; enum i40iw_quad_entry_type entry_type; bool vlan_valid; bool ipv4_valid; u8 mac_addr[6]; u16 vlan_id; - u16 qs_handle; + u8 user_pri; u32 qp_num; u32 dest_ip[4]; u32 src_ip[4]; @@ -976,7 +1010,7 @@ struct i40iw_cqp_query_fpm_values { struct i40iw_cqp_ops { enum i40iw_status_code (*cqp_init)(struct i40iw_sc_cqp *, struct i40iw_cqp_init_info *); - enum i40iw_status_code (*cqp_create)(struct i40iw_sc_cqp *, bool, u16 *, u16 *); + enum i40iw_status_code (*cqp_create)(struct i40iw_sc_cqp *, u16 *, u16 *); void (*cqp_post_sq)(struct i40iw_sc_cqp *); u64 *(*cqp_get_next_send_wqe)(struct i40iw_sc_cqp *, u64 scratch); enum i40iw_status_code (*cqp_destroy)(struct i40iw_sc_cqp *); @@ -1019,7 +1053,7 @@ struct i40iw_aeq_ops { }; struct i40iw_pd_ops { - void (*pd_init)(struct i40iw_sc_dev *, struct i40iw_sc_pd *, u16); + void (*pd_init)(struct i40iw_sc_dev *, struct i40iw_sc_pd *, u16, int); }; struct i40iw_priv_qp_ops { diff --git a/drivers/infiniband/hw/i40iw/i40iw_ucontext.h b/drivers/infiniband/hw/i40iw/i40iw_ucontext.h index 12acd688def4..57d3f1d11ff1 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_ucontext.h +++ b/drivers/infiniband/hw/i40iw/i40iw_ucontext.h @@ -39,8 +39,8 @@ #include <linux/types.h> -#define I40IW_ABI_USERSPACE_VER 4 -#define I40IW_ABI_KERNEL_VER 4 +#define I40IW_ABI_VER 5 + struct i40iw_alloc_ucontext_req { __u32 reserved32; __u8 userspace_ver; diff --git a/drivers/infiniband/hw/i40iw/i40iw_uk.c b/drivers/infiniband/hw/i40iw/i40iw_uk.c index 4d28c3cb03cc..b0d3a0e8a9b5 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_uk.c +++ b/drivers/infiniband/hw/i40iw/i40iw_uk.c @@ -175,12 +175,10 @@ u64 *i40iw_qp_get_next_send_wqe(struct i40iw_qp_uk *qp, if (!*wqe_idx) qp->swqe_polarity = !qp->swqe_polarity; } - - for (i = 0; i < wqe_size / I40IW_QP_WQE_MIN_SIZE; i++) { - I40IW_RING_MOVE_HEAD(qp->sq_ring, ret_code); - if (ret_code) - return NULL; - } + I40IW_RING_MOVE_HEAD_BY_COUNT(qp->sq_ring, + wqe_size / I40IW_QP_WQE_MIN_SIZE, ret_code); + if (ret_code) + return NULL; wqe = qp->sq_base[*wqe_idx].elem; @@ -430,7 +428,7 @@ static enum i40iw_status_code i40iw_inline_rdma_write(struct i40iw_qp_uk *qp, struct i40iw_inline_rdma_write *op_info; u64 *push; u64 header = 0; - u32 i, wqe_idx; + u32 wqe_idx; enum i40iw_status_code ret_code; bool read_fence = false; u8 wqe_size; @@ -465,14 +463,12 @@ static enum i40iw_status_code i40iw_inline_rdma_write(struct i40iw_qp_uk *qp, src = (u8 *)(op_info->data); if (op_info->len <= 16) { - for (i = 0; i < op_info->len; i++, src++, dest++) - *dest = *src; + memcpy(dest, src, op_info->len); } else { - for (i = 0; i < 16; i++, src++, dest++) - *dest = *src; + memcpy(dest, src, 16); + src += 16; dest = (u8 *)wqe + 32; - for (; i < op_info->len; i++, src++, dest++) - *dest = *src; + memcpy(dest, src, op_info->len - 16); } wmb(); /* make sure WQE is populated before valid bit is set */ @@ -507,7 +503,7 @@ static enum i40iw_status_code i40iw_inline_send(struct i40iw_qp_uk *qp, u8 *dest, *src; struct i40iw_post_inline_send *op_info; u64 header; - u32 wqe_idx, i; + u32 wqe_idx; enum i40iw_status_code ret_code; bool read_fence = false; u8 wqe_size; @@ -540,14 +536,12 @@ static enum i40iw_status_code i40iw_inline_send(struct i40iw_qp_uk *qp, src = (u8 *)(op_info->data); if (op_info->len <= 16) { - for (i = 0; i < op_info->len; i++, src++, dest++) - *dest = *src; + memcpy(dest, src, op_info->len); } else { - for (i = 0; i < 16; i++, src++, dest++) - *dest = *src; + memcpy(dest, src, 16); + src += 16; dest = (u8 *)wqe + 32; - for (; i < op_info->len; i++, src++, dest++) - *dest = *src; + memcpy(dest, src, op_info->len - 16); } wmb(); /* make sure WQE is populated before valid bit is set */ @@ -919,29 +913,29 @@ enum i40iw_status_code i40iw_get_wqe_shift(u32 wqdepth, u32 sge, u32 inline_data } static struct i40iw_qp_uk_ops iw_qp_uk_ops = { - i40iw_qp_post_wr, - i40iw_qp_ring_push_db, - i40iw_rdma_write, - i40iw_rdma_read, - i40iw_send, - i40iw_inline_rdma_write, - i40iw_inline_send, - i40iw_stag_local_invalidate, - i40iw_mw_bind, - i40iw_post_receive, - i40iw_nop + .iw_qp_post_wr = i40iw_qp_post_wr, + .iw_qp_ring_push_db = i40iw_qp_ring_push_db, + .iw_rdma_write = i40iw_rdma_write, + .iw_rdma_read = i40iw_rdma_read, + .iw_send = i40iw_send, + .iw_inline_rdma_write = i40iw_inline_rdma_write, + .iw_inline_send = i40iw_inline_send, + .iw_stag_local_invalidate = i40iw_stag_local_invalidate, + .iw_mw_bind = i40iw_mw_bind, + .iw_post_receive = i40iw_post_receive, + .iw_post_nop = i40iw_nop }; static struct i40iw_cq_ops iw_cq_ops = { - i40iw_cq_request_notification, - i40iw_cq_poll_completion, - i40iw_cq_post_entries, - i40iw_clean_cq + .iw_cq_request_notification = i40iw_cq_request_notification, + .iw_cq_poll_completion = i40iw_cq_poll_completion, + .iw_cq_post_entries = i40iw_cq_post_entries, + .iw_cq_clean = i40iw_clean_cq }; static struct i40iw_device_uk_ops iw_device_uk_ops = { - i40iw_cq_uk_init, - i40iw_qp_uk_init, + .iwarp_cq_uk_init = i40iw_cq_uk_init, + .iwarp_qp_uk_init = i40iw_qp_uk_init, }; /** @@ -972,10 +966,6 @@ enum i40iw_status_code i40iw_qp_uk_init(struct i40iw_qp_uk *qp, if (ret_code) return ret_code; - ret_code = i40iw_get_wqe_shift(info->rq_size, info->max_rq_frag_cnt, 0, &rqshift); - if (ret_code) - return ret_code; - qp->sq_base = info->sq; qp->rq_base = info->rq; qp->shadow_area = info->shadow_area; @@ -1004,8 +994,19 @@ enum i40iw_status_code i40iw_qp_uk_init(struct i40iw_qp_uk *qp, if (!qp->use_srq) { qp->rq_size = info->rq_size; qp->max_rq_frag_cnt = info->max_rq_frag_cnt; - qp->rq_wqe_size = rqshift; I40IW_RING_INIT(qp->rq_ring, qp->rq_size); + switch (info->abi_ver) { + case 4: + ret_code = i40iw_get_wqe_shift(info->rq_size, info->max_rq_frag_cnt, 0, &rqshift); + if (ret_code) + return ret_code; + break; + case 5: /* fallthrough until next ABI version */ + default: + rqshift = I40IW_MAX_RQ_WQE_SHIFT; + break; + } + qp->rq_wqe_size = rqshift; qp->rq_wqe_size_multiplier = 4 << rqshift; } qp->ops = iw_qp_uk_ops; @@ -1190,12 +1191,8 @@ enum i40iw_status_code i40iw_inline_data_size_to_wqesize(u32 data_size, if (data_size <= 16) *wqe_size = I40IW_QP_WQE_MIN_SIZE; - else if (data_size <= 48) - *wqe_size = 64; - else if (data_size <= 80) - *wqe_size = 96; else - *wqe_size = 128; + *wqe_size = 64; return 0; } diff --git a/drivers/infiniband/hw/i40iw/i40iw_user.h b/drivers/infiniband/hw/i40iw/i40iw_user.h index 276bcefffd7e..84be6f13b9c5 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_user.h +++ b/drivers/infiniband/hw/i40iw/i40iw_user.h @@ -72,12 +72,13 @@ enum i40iw_device_capabilities_const { I40IW_MAX_SQ_PAYLOAD_SIZE = 2145386496, I40IW_MAX_INLINE_DATA_SIZE = 48, I40IW_MAX_PUSHMODE_INLINE_DATA_SIZE = 48, - I40IW_MAX_IRD_SIZE = 32, - I40IW_QPCTX_ENCD_MAXIRD = 3, + I40IW_MAX_IRD_SIZE = 63, + I40IW_MAX_ORD_SIZE = 127, I40IW_MAX_WQ_ENTRIES = 2048, - I40IW_MAX_ORD_SIZE = 32, I40IW_Q2_BUFFER_SIZE = (248 + 100), - I40IW_QP_CTX_SIZE = 248 + I40IW_MAX_WQE_SIZE_RQ = 128, + I40IW_QP_CTX_SIZE = 248, + I40IW_MAX_PDS = 32768 }; #define i40iw_handle void * @@ -96,13 +97,8 @@ enum i40iw_device_capabilities_const { #define i40iw_physical_fragment u64 #define i40iw_address_list u64 * -#define I40IW_CREATE_STAG(index, key) (((index) << 8) + (key)) - -#define I40IW_STAG_KEY_FROM_STAG(stag) ((stag) && 0x000000FF) - -#define I40IW_STAG_INDEX_FROM_STAG(stag) (((stag) && 0xFFFFFF00) >> 8) - #define I40IW_MAX_MR_SIZE 0x10000000000L +#define I40IW_MAX_RQ_WQE_SHIFT 2 struct i40iw_qp_uk; struct i40iw_cq_uk; @@ -411,7 +407,7 @@ struct i40iw_qp_uk_init_info { u32 max_sq_frag_cnt; u32 max_rq_frag_cnt; u32 max_inline_data; - + int abi_ver; }; struct i40iw_cq_uk_init_info { diff --git a/drivers/infiniband/hw/i40iw/i40iw_utils.c b/drivers/infiniband/hw/i40iw/i40iw_utils.c index 6fd043b1d714..0f5d43d1f5fc 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_utils.c +++ b/drivers/infiniband/hw/i40iw/i40iw_utils.c @@ -153,6 +153,7 @@ int i40iw_inetaddr_event(struct notifier_block *notifier, struct i40iw_device *iwdev; struct i40iw_handler *hdl; u32 local_ipaddr; + u32 action = I40IW_ARP_ADD; hdl = i40iw_find_netdev(event_netdev); if (!hdl) @@ -164,44 +165,25 @@ int i40iw_inetaddr_event(struct notifier_block *notifier, if (netdev != event_netdev) return NOTIFY_DONE; + if (upper_dev) + local_ipaddr = ntohl( + ((struct in_device *)upper_dev->ip_ptr)->ifa_list->ifa_address); + else + local_ipaddr = ntohl(ifa->ifa_address); switch (event) { case NETDEV_DOWN: - if (upper_dev) - local_ipaddr = ntohl( - ((struct in_device *)upper_dev->ip_ptr)->ifa_list->ifa_address); - else - local_ipaddr = ntohl(ifa->ifa_address); - i40iw_manage_arp_cache(iwdev, - netdev->dev_addr, - &local_ipaddr, - true, - I40IW_ARP_DELETE); - return NOTIFY_OK; + action = I40IW_ARP_DELETE; + /* Fall through */ case NETDEV_UP: - if (upper_dev) - local_ipaddr = ntohl( - ((struct in_device *)upper_dev->ip_ptr)->ifa_list->ifa_address); - else - local_ipaddr = ntohl(ifa->ifa_address); - i40iw_manage_arp_cache(iwdev, - netdev->dev_addr, - &local_ipaddr, - true, - I40IW_ARP_ADD); - break; + /* Fall through */ case NETDEV_CHANGEADDR: - /* Add the address to the IP table */ - if (upper_dev) - local_ipaddr = ntohl( - ((struct in_device *)upper_dev->ip_ptr)->ifa_list->ifa_address); - else - local_ipaddr = ntohl(ifa->ifa_address); - i40iw_manage_arp_cache(iwdev, netdev->dev_addr, &local_ipaddr, true, - I40IW_ARP_ADD); + action); + i40iw_if_notify(iwdev, netdev, &local_ipaddr, true, + (action == I40IW_ARP_ADD) ? true : false); break; default: break; @@ -225,6 +207,7 @@ int i40iw_inet6addr_event(struct notifier_block *notifier, struct i40iw_device *iwdev; struct i40iw_handler *hdl; u32 local_ipaddr6[4]; + u32 action = I40IW_ARP_ADD; hdl = i40iw_find_netdev(event_netdev); if (!hdl) @@ -235,24 +218,21 @@ int i40iw_inet6addr_event(struct notifier_block *notifier, if (netdev != event_netdev) return NOTIFY_DONE; + i40iw_copy_ip_ntohl(local_ipaddr6, ifa->addr.in6_u.u6_addr32); switch (event) { case NETDEV_DOWN: - i40iw_copy_ip_ntohl(local_ipaddr6, ifa->addr.in6_u.u6_addr32); - i40iw_manage_arp_cache(iwdev, - netdev->dev_addr, - local_ipaddr6, - false, - I40IW_ARP_DELETE); - return NOTIFY_OK; + action = I40IW_ARP_DELETE; + /* Fall through */ case NETDEV_UP: /* Fall through */ case NETDEV_CHANGEADDR: - i40iw_copy_ip_ntohl(local_ipaddr6, ifa->addr.in6_u.u6_addr32); i40iw_manage_arp_cache(iwdev, netdev->dev_addr, local_ipaddr6, false, - I40IW_ARP_ADD); + action); + i40iw_if_notify(iwdev, netdev, local_ipaddr6, false, + (action == I40IW_ARP_ADD) ? true : false); break; default: break; @@ -392,6 +372,7 @@ static void i40iw_free_qp(struct i40iw_cqp_request *cqp_request, u32 num) i40iw_rem_pdusecount(iwqp->iwpd, iwdev); i40iw_free_qp_resources(iwdev, iwqp, qp_num); + i40iw_rem_devusecount(iwdev); } /** @@ -415,7 +396,10 @@ static int i40iw_wait_event(struct i40iw_device *iwdev, i40iw_pr_err("error cqp command 0x%x timed out ret = %d\n", info->cqp_cmd, timeout_ret); err_code = -ETIME; - i40iw_request_reset(iwdev); + if (!iwdev->reset) { + iwdev->reset = true; + i40iw_request_reset(iwdev); + } goto done; } cqp_error = cqp_request->compl_info.error; @@ -445,6 +429,11 @@ enum i40iw_status_code i40iw_handle_cqp_op(struct i40iw_device *iwdev, struct cqp_commands_info *info = &cqp_request->info; int err_code = 0; + if (iwdev->reset) { + i40iw_free_cqp_request(&iwdev->cqp, cqp_request); + return I40IW_ERR_CQP_COMPL_ERROR; + } + status = i40iw_process_cqp_cmd(dev, info); if (status) { i40iw_pr_err("error cqp command 0x%x failed\n", info->cqp_cmd); @@ -459,6 +448,26 @@ enum i40iw_status_code i40iw_handle_cqp_op(struct i40iw_device *iwdev, } /** + * i40iw_add_devusecount - add dev refcount + * @iwdev: dev for refcount + */ +void i40iw_add_devusecount(struct i40iw_device *iwdev) +{ + atomic64_inc(&iwdev->use_count); +} + +/** + * i40iw_rem_devusecount - decrement refcount for dev + * @iwdev: device + */ +void i40iw_rem_devusecount(struct i40iw_device *iwdev) +{ + if (!atomic64_dec_and_test(&iwdev->use_count)) + return; + wake_up(&iwdev->close_wq); +} + +/** * i40iw_add_pdusecount - add pd refcount * @iwpd: pd for refcount */ @@ -712,6 +721,51 @@ enum i40iw_status_code i40iw_cqp_sds_cmd(struct i40iw_sc_dev *dev, } /** + * i40iw_qp_suspend_resume - cqp command for suspend/resume + * @dev: hardware control device structure + * @qp: hardware control qp + * @suspend: flag if suspend or resume + */ +void i40iw_qp_suspend_resume(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp, bool suspend) +{ + struct i40iw_device *iwdev = (struct i40iw_device *)dev->back_dev; + struct i40iw_cqp_request *cqp_request; + struct i40iw_sc_cqp *cqp = dev->cqp; + struct cqp_commands_info *cqp_info; + enum i40iw_status_code status; + + cqp_request = i40iw_get_cqp_request(&iwdev->cqp, false); + if (!cqp_request) + return; + + cqp_info = &cqp_request->info; + cqp_info->cqp_cmd = (suspend) ? OP_SUSPEND : OP_RESUME; + cqp_info->in.u.suspend_resume.cqp = cqp; + cqp_info->in.u.suspend_resume.qp = qp; + cqp_info->in.u.suspend_resume.scratch = (uintptr_t)cqp_request; + status = i40iw_handle_cqp_op(iwdev, cqp_request); + if (status) + i40iw_pr_err("CQP-OP QP Suspend/Resume fail"); +} + +/** + * i40iw_qp_mss_modify - modify mss for qp + * @dev: hardware control device structure + * @qp: hardware control qp + */ +void i40iw_qp_mss_modify(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp) +{ + struct i40iw_device *iwdev = (struct i40iw_device *)dev->back_dev; + struct i40iw_qp *iwqp = (struct i40iw_qp *)qp->back_qp; + struct i40iw_modify_qp_info info; + + memset(&info, 0, sizeof(info)); + info.mss_change = true; + info.new_mss = qp->vsi->mss; + i40iw_hw_modify_qp(iwdev, iwqp, &info, false); +} + +/** * i40iw_term_modify_qp - modify qp for term message * @qp: hardware control qp * @next_state: qp's next state @@ -769,6 +823,7 @@ static void i40iw_terminate_timeout(unsigned long context) struct i40iw_sc_qp *qp = (struct i40iw_sc_qp *)&iwqp->sc_qp; i40iw_terminate_done(qp, 1); + i40iw_rem_ref(&iwqp->ibqp); } /** @@ -780,6 +835,7 @@ void i40iw_terminate_start_timer(struct i40iw_sc_qp *qp) struct i40iw_qp *iwqp; iwqp = (struct i40iw_qp *)qp->back_qp; + i40iw_add_ref(&iwqp->ibqp); init_timer(&iwqp->terminate_timer); iwqp->terminate_timer.function = i40iw_terminate_timeout; iwqp->terminate_timer.expires = jiffies + HZ; @@ -796,7 +852,8 @@ void i40iw_terminate_del_timer(struct i40iw_sc_qp *qp) struct i40iw_qp *iwqp; iwqp = (struct i40iw_qp *)qp->back_qp; - del_timer(&iwqp->terminate_timer); + if (del_timer(&iwqp->terminate_timer)) + i40iw_rem_ref(&iwqp->ibqp); } /** @@ -1011,6 +1068,116 @@ enum i40iw_status_code i40iw_vf_wait_vchnl_resp(struct i40iw_sc_dev *dev) } /** + * i40iw_cqp_cq_create_cmd - create a cq for the cqp + * @dev: device pointer + * @cq: pointer to created cq + */ +enum i40iw_status_code i40iw_cqp_cq_create_cmd(struct i40iw_sc_dev *dev, + struct i40iw_sc_cq *cq) +{ + struct i40iw_device *iwdev = (struct i40iw_device *)dev->back_dev; + struct i40iw_cqp *iwcqp = &iwdev->cqp; + struct i40iw_cqp_request *cqp_request; + struct cqp_commands_info *cqp_info; + enum i40iw_status_code status; + + cqp_request = i40iw_get_cqp_request(iwcqp, true); + if (!cqp_request) + return I40IW_ERR_NO_MEMORY; + + cqp_info = &cqp_request->info; + cqp_info->cqp_cmd = OP_CQ_CREATE; + cqp_info->post_sq = 1; + cqp_info->in.u.cq_create.cq = cq; + cqp_info->in.u.cq_create.scratch = (uintptr_t)cqp_request; + status = i40iw_handle_cqp_op(iwdev, cqp_request); + if (status) + i40iw_pr_err("CQP-OP Create QP fail"); + + return status; +} + +/** + * i40iw_cqp_qp_create_cmd - create a qp for the cqp + * @dev: device pointer + * @qp: pointer to created qp + */ +enum i40iw_status_code i40iw_cqp_qp_create_cmd(struct i40iw_sc_dev *dev, + struct i40iw_sc_qp *qp) +{ + struct i40iw_device *iwdev = (struct i40iw_device *)dev->back_dev; + struct i40iw_cqp *iwcqp = &iwdev->cqp; + struct i40iw_cqp_request *cqp_request; + struct cqp_commands_info *cqp_info; + struct i40iw_create_qp_info *qp_info; + enum i40iw_status_code status; + + cqp_request = i40iw_get_cqp_request(iwcqp, true); + if (!cqp_request) + return I40IW_ERR_NO_MEMORY; + + cqp_info = &cqp_request->info; + qp_info = &cqp_request->info.in.u.qp_create.info; + + memset(qp_info, 0, sizeof(*qp_info)); + + qp_info->cq_num_valid = true; + qp_info->next_iwarp_state = I40IW_QP_STATE_RTS; + + cqp_info->cqp_cmd = OP_QP_CREATE; + cqp_info->post_sq = 1; + cqp_info->in.u.qp_create.qp = qp; + cqp_info->in.u.qp_create.scratch = (uintptr_t)cqp_request; + status = i40iw_handle_cqp_op(iwdev, cqp_request); + if (status) + i40iw_pr_err("CQP-OP QP create fail"); + return status; +} + +/** + * i40iw_cqp_cq_destroy_cmd - destroy the cqp cq + * @dev: device pointer + * @cq: pointer to cq + */ +void i40iw_cqp_cq_destroy_cmd(struct i40iw_sc_dev *dev, struct i40iw_sc_cq *cq) +{ + struct i40iw_device *iwdev = (struct i40iw_device *)dev->back_dev; + + i40iw_cq_wq_destroy(iwdev, cq); +} + +/** + * i40iw_cqp_qp_destroy_cmd - destroy the cqp + * @dev: device pointer + * @qp: pointer to qp + */ +void i40iw_cqp_qp_destroy_cmd(struct i40iw_sc_dev *dev, struct i40iw_sc_qp *qp) +{ + struct i40iw_device *iwdev = (struct i40iw_device *)dev->back_dev; + struct i40iw_cqp *iwcqp = &iwdev->cqp; + struct i40iw_cqp_request *cqp_request; + struct cqp_commands_info *cqp_info; + enum i40iw_status_code status; + + cqp_request = i40iw_get_cqp_request(iwcqp, true); + if (!cqp_request) + return; + + cqp_info = &cqp_request->info; + memset(cqp_info, 0, sizeof(*cqp_info)); + + cqp_info->cqp_cmd = OP_QP_DESTROY; + cqp_info->post_sq = 1; + cqp_info->in.u.qp_destroy.qp = qp; + cqp_info->in.u.qp_destroy.scratch = (uintptr_t)cqp_request; + cqp_info->in.u.qp_destroy.remove_hash_idx = true; + status = i40iw_handle_cqp_op(iwdev, cqp_request); + if (status) + i40iw_pr_err("CQP QP_DESTROY fail"); +} + + +/** * i40iw_ieq_mpa_crc_ae - generate AE for crc error * @dev: hardware control device structure * @qp: hardware control qp @@ -1208,7 +1375,7 @@ enum i40iw_status_code i40iw_puda_get_tcpip_info(struct i40iw_puda_completion_in buf->totallen = pkt_len + buf->maclen; - if (info->payload_len < buf->totallen - 4) { + if (info->payload_len < buf->totallen) { i40iw_pr_err("payload_len = 0x%x totallen expected0x%x\n", info->payload_len, buf->totallen); return I40IW_ERR_INVALID_SIZE; @@ -1224,27 +1391,29 @@ enum i40iw_status_code i40iw_puda_get_tcpip_info(struct i40iw_puda_completion_in /** * i40iw_hw_stats_timeout - Stats timer-handler which updates all HW stats - * @dev: hardware control device structure + * @vsi: pointer to the vsi structure */ -static void i40iw_hw_stats_timeout(unsigned long dev) +static void i40iw_hw_stats_timeout(unsigned long vsi) { - struct i40iw_sc_dev *pf_dev = (struct i40iw_sc_dev *)dev; - struct i40iw_dev_pestat *pf_devstat = &pf_dev->dev_pestat; - struct i40iw_dev_pestat *vf_devstat = NULL; + struct i40iw_sc_vsi *sc_vsi = (struct i40iw_sc_vsi *)vsi; + struct i40iw_sc_dev *pf_dev = sc_vsi->dev; + struct i40iw_vsi_pestat *pf_devstat = sc_vsi->pestat; + struct i40iw_vsi_pestat *vf_devstat = NULL; u16 iw_vf_idx; unsigned long flags; /*PF*/ - pf_devstat->ops.iw_hw_stat_read_all(pf_devstat, &pf_devstat->hw_stats); + i40iw_hw_stats_read_all(pf_devstat, &pf_devstat->hw_stats); + for (iw_vf_idx = 0; iw_vf_idx < I40IW_MAX_PE_ENABLED_VF_COUNT; iw_vf_idx++) { - spin_lock_irqsave(&pf_devstat->stats_lock, flags); + spin_lock_irqsave(&pf_devstat->lock, flags); if (pf_dev->vf_dev[iw_vf_idx]) { if (pf_dev->vf_dev[iw_vf_idx]->stats_initialized) { - vf_devstat = &pf_dev->vf_dev[iw_vf_idx]->dev_pestat; - vf_devstat->ops.iw_hw_stat_read_all(vf_devstat, &vf_devstat->hw_stats); + vf_devstat = &pf_dev->vf_dev[iw_vf_idx]->pestat; + i40iw_hw_stats_read_all(vf_devstat, &vf_devstat->hw_stats); } } - spin_unlock_irqrestore(&pf_devstat->stats_lock, flags); + spin_unlock_irqrestore(&pf_devstat->lock, flags); } mod_timer(&pf_devstat->stats_timer, @@ -1253,26 +1422,26 @@ static void i40iw_hw_stats_timeout(unsigned long dev) /** * i40iw_hw_stats_start_timer - Start periodic stats timer - * @dev: hardware control device structure + * @vsi: pointer to the vsi structure */ -void i40iw_hw_stats_start_timer(struct i40iw_sc_dev *dev) +void i40iw_hw_stats_start_timer(struct i40iw_sc_vsi *vsi) { - struct i40iw_dev_pestat *devstat = &dev->dev_pestat; + struct i40iw_vsi_pestat *devstat = vsi->pestat; init_timer(&devstat->stats_timer); devstat->stats_timer.function = i40iw_hw_stats_timeout; - devstat->stats_timer.data = (unsigned long)dev; + devstat->stats_timer.data = (unsigned long)vsi; mod_timer(&devstat->stats_timer, jiffies + msecs_to_jiffies(STATS_TIMER_DELAY)); } /** - * i40iw_hw_stats_del_timer - Delete periodic stats timer - * @dev: hardware control device structure + * i40iw_hw_stats_stop_timer - Delete periodic stats timer + * @vsi: pointer to the vsi structure */ -void i40iw_hw_stats_del_timer(struct i40iw_sc_dev *dev) +void i40iw_hw_stats_stop_timer(struct i40iw_sc_vsi *vsi) { - struct i40iw_dev_pestat *devstat = &dev->dev_pestat; + struct i40iw_vsi_pestat *devstat = vsi->pestat; del_timer_sync(&devstat->stats_timer); } diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.c b/drivers/infiniband/hw/i40iw/i40iw_verbs.c index 6329c971c22f..9b2849979756 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_verbs.c +++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.c @@ -37,6 +37,7 @@ #include <linux/random.h> #include <linux/highmem.h> #include <linux/time.h> +#include <linux/hugetlb.h> #include <asm/byteorder.h> #include <net/ip.h> #include <rdma/ib_verbs.h> @@ -67,13 +68,13 @@ static int i40iw_query_device(struct ib_device *ibdev, props->vendor_part_id = iwdev->ldev->pcidev->device; props->hw_ver = (u32)iwdev->sc_dev.hw_rev; props->max_mr_size = I40IW_MAX_OUTBOUND_MESSAGE_SIZE; - props->max_qp = iwdev->max_qp; + props->max_qp = iwdev->max_qp - iwdev->used_qps; props->max_qp_wr = (I40IW_MAX_WQ_ENTRIES >> 2) - 1; props->max_sge = I40IW_MAX_WQ_FRAGMENT_COUNT; - props->max_cq = iwdev->max_cq; + props->max_cq = iwdev->max_cq - iwdev->used_cqs; props->max_cqe = iwdev->max_cqe; - props->max_mr = iwdev->max_mr; - props->max_pd = iwdev->max_pd; + props->max_mr = iwdev->max_mr - iwdev->used_mrs; + props->max_pd = iwdev->max_pd - iwdev->used_pds; props->max_sge_rd = I40IW_MAX_SGE_RD; props->max_qp_rd_atom = I40IW_MAX_IRD_SIZE; props->max_qp_init_rd_atom = props->max_qp_rd_atom; @@ -96,19 +97,9 @@ static int i40iw_query_port(struct ib_device *ibdev, struct i40iw_device *iwdev = to_iwdev(ibdev); struct net_device *netdev = iwdev->netdev; - memset(props, 0, sizeof(*props)); - + /* props being zeroed by the caller, avoid zeroing it here */ props->max_mtu = IB_MTU_4096; - if (netdev->mtu >= 4096) - props->active_mtu = IB_MTU_4096; - else if (netdev->mtu >= 2048) - props->active_mtu = IB_MTU_2048; - else if (netdev->mtu >= 1024) - props->active_mtu = IB_MTU_1024; - else if (netdev->mtu >= 512) - props->active_mtu = IB_MTU_512; - else - props->active_mtu = IB_MTU_256; + props->active_mtu = ib_mtu_int_to_enum(netdev->mtu); props->lid = 1; if (netif_carrier_ok(iwdev->netdev)) @@ -144,9 +135,8 @@ static struct ib_ucontext *i40iw_alloc_ucontext(struct ib_device *ibdev, if (ib_copy_from_udata(&req, udata, sizeof(req))) return ERR_PTR(-EINVAL); - if (req.userspace_ver != I40IW_ABI_USERSPACE_VER) { - i40iw_pr_err("Invalid userspace driver version detected. Detected version %d, should be %d\n", - req.userspace_ver, I40IW_ABI_USERSPACE_VER); + if (req.userspace_ver < 4 || req.userspace_ver > I40IW_ABI_VER) { + i40iw_pr_err("Unsupported provider library version %u.\n", req.userspace_ver); return ERR_PTR(-EINVAL); } @@ -154,13 +144,14 @@ static struct ib_ucontext *i40iw_alloc_ucontext(struct ib_device *ibdev, uresp.max_qps = iwdev->max_qp; uresp.max_pds = iwdev->max_pd; uresp.wq_size = iwdev->max_qp_wr * 2; - uresp.kernel_ver = I40IW_ABI_KERNEL_VER; + uresp.kernel_ver = req.userspace_ver; ucontext = kzalloc(sizeof(*ucontext), GFP_KERNEL); if (!ucontext) return ERR_PTR(-ENOMEM); ucontext->iwdev = iwdev; + ucontext->abi_ver = req.userspace_ver; if (ib_copy_to_udata(udata, &uresp, sizeof(uresp))) { kfree(ucontext); @@ -254,7 +245,6 @@ static void i40iw_alloc_push_page(struct i40iw_device *iwdev, struct i40iw_sc_qp { struct i40iw_cqp_request *cqp_request; struct cqp_commands_info *cqp_info; - struct i40iw_sc_dev *dev = &iwdev->sc_dev; enum i40iw_status_code status; if (qp->push_idx != I40IW_INVALID_PUSH_PAGE_INDEX) @@ -270,7 +260,7 @@ static void i40iw_alloc_push_page(struct i40iw_device *iwdev, struct i40iw_sc_qp cqp_info->cqp_cmd = OP_MANAGE_PUSH_PAGE; cqp_info->post_sq = 1; - cqp_info->in.u.manage_push_page.info.qs_handle = dev->qs_handle; + cqp_info->in.u.manage_push_page.info.qs_handle = qp->qs_handle; cqp_info->in.u.manage_push_page.info.free_page = 0; cqp_info->in.u.manage_push_page.cqp = &iwdev->cqp.sc_cqp; cqp_info->in.u.manage_push_page.scratch = (uintptr_t)cqp_request; @@ -292,7 +282,6 @@ static void i40iw_dealloc_push_page(struct i40iw_device *iwdev, struct i40iw_sc_ { struct i40iw_cqp_request *cqp_request; struct cqp_commands_info *cqp_info; - struct i40iw_sc_dev *dev = &iwdev->sc_dev; enum i40iw_status_code status; if (qp->push_idx == I40IW_INVALID_PUSH_PAGE_INDEX) @@ -307,7 +296,7 @@ static void i40iw_dealloc_push_page(struct i40iw_device *iwdev, struct i40iw_sc_ cqp_info->post_sq = 1; cqp_info->in.u.manage_push_page.info.push_idx = qp->push_idx; - cqp_info->in.u.manage_push_page.info.qs_handle = dev->qs_handle; + cqp_info->in.u.manage_push_page.info.qs_handle = qp->qs_handle; cqp_info->in.u.manage_push_page.info.free_page = 1; cqp_info->in.u.manage_push_page.cqp = &iwdev->cqp.sc_cqp; cqp_info->in.u.manage_push_page.scratch = (uintptr_t)cqp_request; @@ -334,9 +323,13 @@ static struct ib_pd *i40iw_alloc_pd(struct ib_device *ibdev, struct i40iw_sc_dev *dev = &iwdev->sc_dev; struct i40iw_alloc_pd_resp uresp; struct i40iw_sc_pd *sc_pd; + struct i40iw_ucontext *ucontext; u32 pd_id = 0; int err; + if (iwdev->closing) + return ERR_PTR(-ENODEV); + err = i40iw_alloc_resource(iwdev, iwdev->allocated_pds, iwdev->max_pd, &pd_id, &iwdev->next_pd); if (err) { @@ -351,15 +344,18 @@ static struct ib_pd *i40iw_alloc_pd(struct ib_device *ibdev, } sc_pd = &iwpd->sc_pd; - dev->iw_pd_ops->pd_init(dev, sc_pd, pd_id); if (context) { + ucontext = to_ucontext(context); + dev->iw_pd_ops->pd_init(dev, sc_pd, pd_id, ucontext->abi_ver); memset(&uresp, 0, sizeof(uresp)); uresp.pd_id = pd_id; if (ib_copy_to_udata(udata, &uresp, sizeof(uresp))) { err = -EFAULT; goto error; } + } else { + dev->iw_pd_ops->pd_init(dev, sc_pd, pd_id, -1); } i40iw_add_pdusecount(iwpd); @@ -516,7 +512,7 @@ static int i40iw_setup_kmode_qp(struct i40iw_device *iwdev, struct i40iw_dma_mem *mem = &iwqp->kqp.dma_mem; u32 sqdepth, rqdepth; u32 sq_size, rq_size; - u8 sqshift, rqshift; + u8 sqshift; u32 size; enum i40iw_status_code status; struct i40iw_qp_uk_init_info *ukinfo = &info->qp_uk_init_info; @@ -525,14 +521,11 @@ static int i40iw_setup_kmode_qp(struct i40iw_device *iwdev, rq_size = i40iw_qp_roundup(ukinfo->rq_size + 1); status = i40iw_get_wqe_shift(sq_size, ukinfo->max_sq_frag_cnt, ukinfo->max_inline_data, &sqshift); - if (!status) - status = i40iw_get_wqe_shift(rq_size, ukinfo->max_rq_frag_cnt, 0, &rqshift); - if (status) return -ENOMEM; sqdepth = sq_size << sqshift; - rqdepth = rq_size << rqshift; + rqdepth = rq_size << I40IW_MAX_RQ_WQE_SHIFT; size = sqdepth * sizeof(struct i40iw_sq_uk_wr_trk_info) + (rqdepth << 3); iwqp->kqp.wrid_mem = kzalloc(size, GFP_KERNEL); @@ -602,6 +595,9 @@ static struct ib_qp *i40iw_create_qp(struct ib_pd *ibpd, struct i40iwarp_offload_info *iwarp_info; unsigned long flags; + if (iwdev->closing) + return ERR_PTR(-ENODEV); + if (init_attr->create_flags) return ERR_PTR(-EINVAL); if (init_attr->cap.max_inline_data > I40IW_MAX_INLINE_DATA_SIZE) @@ -610,11 +606,15 @@ static struct ib_qp *i40iw_create_qp(struct ib_pd *ibpd, if (init_attr->cap.max_send_sge > I40IW_MAX_WQ_FRAGMENT_COUNT) init_attr->cap.max_send_sge = I40IW_MAX_WQ_FRAGMENT_COUNT; + if (init_attr->cap.max_recv_sge > I40IW_MAX_WQ_FRAGMENT_COUNT) + init_attr->cap.max_recv_sge = I40IW_MAX_WQ_FRAGMENT_COUNT; + memset(&init_info, 0, sizeof(init_info)); sq_size = init_attr->cap.max_send_wr; rq_size = init_attr->cap.max_recv_wr; + init_info.vsi = &iwdev->vsi; init_info.qp_uk_init_info.sq_size = sq_size; init_info.qp_uk_init_info.rq_size = rq_size; init_info.qp_uk_init_info.max_sq_frag_cnt = init_attr->cap.max_send_sge; @@ -774,6 +774,7 @@ static struct ib_qp *i40iw_create_qp(struct ib_pd *ibpd, iwqp->sig_all = (init_attr->sq_sig_type == IB_SIGNAL_ALL_WR) ? 1 : 0; iwdev->qp_table[qp_num] = iwqp; i40iw_add_pdusecount(iwqp->iwpd); + i40iw_add_devusecount(iwdev); if (ibpd->uobject && udata) { memset(&uresp, 0, sizeof(uresp)); uresp.actual_sq_size = sq_size; @@ -815,8 +816,9 @@ static int i40iw_query_qp(struct ib_qp *ibqp, attr->qp_access_flags = 0; attr->cap.max_send_wr = qp->qp_uk.sq_size; attr->cap.max_recv_wr = qp->qp_uk.rq_size; - attr->cap.max_recv_sge = 1; attr->cap.max_inline_data = I40IW_MAX_INLINE_DATA_SIZE; + attr->cap.max_send_sge = I40IW_MAX_WQ_FRAGMENT_COUNT; + attr->cap.max_recv_sge = I40IW_MAX_WQ_FRAGMENT_COUNT; init_attr->event_handler = iwqp->ibqp.event_handler; init_attr->qp_context = iwqp->ibqp.qp_context; init_attr->send_cq = iwqp->ibqp.send_cq; @@ -884,6 +886,11 @@ int i40iw_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, spin_lock_irqsave(&iwqp->lock, flags); if (attr_mask & IB_QP_STATE) { + if (iwdev->closing && attr->qp_state != IB_QPS_ERR) { + err = -EINVAL; + goto exit; + } + switch (attr->qp_state) { case IB_QPS_INIT: case IB_QPS_RTR: @@ -944,7 +951,7 @@ int i40iw_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, goto exit; } if (iwqp->sc_qp.term_flags) - del_timer(&iwqp->terminate_timer); + i40iw_terminate_del_timer(&iwqp->sc_qp); info.next_iwarp_state = I40IW_QP_STATE_ERROR; if ((iwqp->hw_tcp_state > I40IW_TCP_STATE_CLOSED) && iwdev->iw_status && @@ -1037,11 +1044,11 @@ static void cq_free_resources(struct i40iw_device *iwdev, struct i40iw_cq *iwcq) } /** - * cq_wq_destroy - send cq destroy cqp + * i40iw_cq_wq_destroy - send cq destroy cqp * @iwdev: iwarp device * @cq: hardware control cq */ -static void cq_wq_destroy(struct i40iw_device *iwdev, struct i40iw_sc_cq *cq) +void i40iw_cq_wq_destroy(struct i40iw_device *iwdev, struct i40iw_sc_cq *cq) { enum i40iw_status_code status; struct i40iw_cqp_request *cqp_request; @@ -1080,9 +1087,10 @@ static int i40iw_destroy_cq(struct ib_cq *ib_cq) iwcq = to_iwcq(ib_cq); iwdev = to_iwdev(ib_cq->device); cq = &iwcq->sc_cq; - cq_wq_destroy(iwdev, cq); + i40iw_cq_wq_destroy(iwdev, cq); cq_free_resources(iwdev, iwcq); kfree(iwcq); + i40iw_rem_devusecount(iwdev); return 0; } @@ -1113,6 +1121,9 @@ static struct ib_cq *i40iw_create_cq(struct ib_device *ibdev, int err_code; int entries = attr->cqe; + if (iwdev->closing) + return ERR_PTR(-ENODEV); + if (entries > iwdev->max_cqe) return ERR_PTR(-EINVAL); @@ -1137,7 +1148,8 @@ static struct ib_cq *i40iw_create_cq(struct ib_device *ibdev, ukinfo->cq_id = cq_num; iwcq->ibcq.cqe = info.cq_uk_init_info.cq_size; info.ceqe_mask = 0; - info.ceq_id = 0; + if (attr->comp_vector < iwdev->ceqs_count) + info.ceq_id = attr->comp_vector; info.ceq_id_valid = true; info.ceqe_mask = 1; info.type = I40IW_CQ_TYPE_IWARP; @@ -1229,10 +1241,11 @@ static struct ib_cq *i40iw_create_cq(struct ib_device *ibdev, } } + i40iw_add_devusecount(iwdev); return (struct ib_cq *)iwcq; cq_destroy: - cq_wq_destroy(iwdev, cq); + i40iw_cq_wq_destroy(iwdev, cq); cq_free_resources: cq_free_resources(iwdev, iwcq); error: @@ -1266,6 +1279,7 @@ static void i40iw_free_stag(struct i40iw_device *iwdev, u32 stag) stag_idx = (stag & iwdev->mr_stagmask) >> I40IW_CQPSQ_STAG_IDX_SHIFT; i40iw_free_resource(iwdev, iwdev->allocated_mrs, stag_idx); + i40iw_rem_devusecount(iwdev); } /** @@ -1296,19 +1310,18 @@ static u32 i40iw_create_stag(struct i40iw_device *iwdev) stag = stag_index << I40IW_CQPSQ_STAG_IDX_SHIFT; stag |= driver_key; stag += (u32)consumer_key; + i40iw_add_devusecount(iwdev); } return stag; } /** * i40iw_next_pbl_addr - Get next pbl address - * @palloc: Poiner to allocated pbles * @pbl: pointer to a pble * @pinfo: info pointer * @idx: index */ -static inline u64 *i40iw_next_pbl_addr(struct i40iw_pble_alloc *palloc, - u64 *pbl, +static inline u64 *i40iw_next_pbl_addr(u64 *pbl, struct i40iw_pble_info **pinfo, u32 *idx) { @@ -1336,9 +1349,11 @@ static void i40iw_copy_user_pgaddrs(struct i40iw_mr *iwmr, struct i40iw_pble_alloc *palloc = &iwpbl->pble_alloc; struct i40iw_pble_info *pinfo; struct scatterlist *sg; + u64 pg_addr = 0; u32 idx = 0; pinfo = (level == I40IW_LEVEL_1) ? NULL : palloc->level2.leaf; + pg_shift = ffs(region->page_size) - 1; for_each_sg(region->sg_head.sgl, sg, region->nmap, entry) { chunk_pages = sg_dma_len(sg) >> pg_shift; @@ -1346,17 +1361,96 @@ static void i40iw_copy_user_pgaddrs(struct i40iw_mr *iwmr, !iwpbl->qp_mr.sq_page) iwpbl->qp_mr.sq_page = sg_page(sg); for (i = 0; i < chunk_pages; i++) { - *pbl = cpu_to_le64(sg_dma_address(sg) + region->page_size * i); - pbl = i40iw_next_pbl_addr(palloc, pbl, &pinfo, &idx); + pg_addr = sg_dma_address(sg) + region->page_size * i; + + if ((entry + i) == 0) + *pbl = cpu_to_le64(pg_addr & iwmr->page_msk); + else if (!(pg_addr & ~iwmr->page_msk)) + *pbl = cpu_to_le64(pg_addr); + else + continue; + pbl = i40iw_next_pbl_addr(pbl, &pinfo, &idx); } } } /** + * i40iw_set_hugetlb_params - set MR pg size and mask to huge pg values. + * @addr: virtual address + * @iwmr: mr pointer for this memory registration + */ +static void i40iw_set_hugetlb_values(u64 addr, struct i40iw_mr *iwmr) +{ + struct vm_area_struct *vma; + struct hstate *h; + + vma = find_vma(current->mm, addr); + if (vma && is_vm_hugetlb_page(vma)) { + h = hstate_vma(vma); + if (huge_page_size(h) == 0x200000) { + iwmr->page_size = huge_page_size(h); + iwmr->page_msk = huge_page_mask(h); + } + } +} + +/** + * i40iw_check_mem_contiguous - check if pbls stored in arr are contiguous + * @arr: lvl1 pbl array + * @npages: page count + * pg_size: page size + * + */ +static bool i40iw_check_mem_contiguous(u64 *arr, u32 npages, u32 pg_size) +{ + u32 pg_idx; + + for (pg_idx = 0; pg_idx < npages; pg_idx++) { + if ((*arr + (pg_size * pg_idx)) != arr[pg_idx]) + return false; + } + return true; +} + +/** + * i40iw_check_mr_contiguous - check if MR is physically contiguous + * @palloc: pbl allocation struct + * pg_size: page size + */ +static bool i40iw_check_mr_contiguous(struct i40iw_pble_alloc *palloc, u32 pg_size) +{ + struct i40iw_pble_level2 *lvl2 = &palloc->level2; + struct i40iw_pble_info *leaf = lvl2->leaf; + u64 *arr = NULL; + u64 *start_addr = NULL; + int i; + bool ret; + + if (palloc->level == I40IW_LEVEL_1) { + arr = (u64 *)palloc->level1.addr; + ret = i40iw_check_mem_contiguous(arr, palloc->total_cnt, pg_size); + return ret; + } + + start_addr = (u64 *)leaf->addr; + + for (i = 0; i < lvl2->leaf_cnt; i++, leaf++) { + arr = (u64 *)leaf->addr; + if ((*start_addr + (i * pg_size * PBLE_PER_PAGE)) != *arr) + return false; + ret = i40iw_check_mem_contiguous(arr, leaf->cnt, pg_size); + if (!ret) + return false; + } + + return true; +} + +/** * i40iw_setup_pbles - copy user pg address to pble's * @iwdev: iwarp device * @iwmr: mr pointer for this memory registration - * @use_pbles: flag if to use pble's or memory (level 0) + * @use_pbles: flag if to use pble's */ static int i40iw_setup_pbles(struct i40iw_device *iwdev, struct i40iw_mr *iwmr, @@ -1369,9 +1463,6 @@ static int i40iw_setup_pbles(struct i40iw_device *iwdev, enum i40iw_status_code status; enum i40iw_pble_level level = I40IW_LEVEL_1; - if (!use_pbles && (iwmr->page_cnt > MAX_SAVE_PAGE_ADDRS)) - return -ENOMEM; - if (use_pbles) { mutex_lock(&iwdev->pbl_mutex); status = i40iw_get_pble(&iwdev->sc_dev, iwdev->pble_rsrc, palloc, iwmr->page_cnt); @@ -1388,6 +1479,10 @@ static int i40iw_setup_pbles(struct i40iw_device *iwdev, } i40iw_copy_user_pgaddrs(iwmr, pbl, level); + + if (use_pbles) + iwmr->pgaddrmem[0] = *pbl; + return 0; } @@ -1409,14 +1504,18 @@ static int i40iw_handle_q_mem(struct i40iw_device *iwdev, struct i40iw_cq_mr *cqmr = &iwpbl->cq_mr; struct i40iw_hmc_pble *hmc_p; u64 *arr = iwmr->pgaddrmem; + u32 pg_size; int err; int total; + bool ret = true; total = req->sq_pages + req->rq_pages + req->cq_pages; + pg_size = iwmr->page_size; err = i40iw_setup_pbles(iwdev, iwmr, use_pbles); if (err) return err; + if (use_pbles && (palloc->level != I40IW_LEVEL_1)) { i40iw_free_pble(iwdev->pble_rsrc, palloc); iwpbl->pbl_allocated = false; @@ -1425,26 +1524,44 @@ static int i40iw_handle_q_mem(struct i40iw_device *iwdev, if (use_pbles) arr = (u64 *)palloc->level1.addr; - if (req->reg_type == IW_MEMREG_TYPE_QP) { + + if (iwmr->type == IW_MEMREG_TYPE_QP) { hmc_p = &qpmr->sq_pbl; qpmr->shadow = (dma_addr_t)arr[total]; + if (use_pbles) { + ret = i40iw_check_mem_contiguous(arr, req->sq_pages, pg_size); + if (ret) + ret = i40iw_check_mem_contiguous(&arr[req->sq_pages], req->rq_pages, pg_size); + } + + if (!ret) { hmc_p->idx = palloc->level1.idx; hmc_p = &qpmr->rq_pbl; hmc_p->idx = palloc->level1.idx + req->sq_pages; } else { hmc_p->addr = arr[0]; hmc_p = &qpmr->rq_pbl; - hmc_p->addr = arr[1]; + hmc_p->addr = arr[req->sq_pages]; } } else { /* CQ */ hmc_p = &cqmr->cq_pbl; cqmr->shadow = (dma_addr_t)arr[total]; + if (use_pbles) + ret = i40iw_check_mem_contiguous(arr, req->cq_pages, pg_size); + + if (!ret) hmc_p->idx = palloc->level1.idx; else hmc_p->addr = arr[0]; } + + if (use_pbles && ret) { + i40iw_free_pble(iwdev->pble_rsrc, palloc); + iwpbl->pbl_allocated = false; + } + return err; } @@ -1642,8 +1759,9 @@ static int i40iw_hwreg_mr(struct i40iw_device *iwdev, stag_info->access_rights = access; stag_info->pd_id = iwpd->sc_pd.pd_id; stag_info->addr_type = I40IW_ADDR_TYPE_VA_BASED; + stag_info->page_size = iwmr->page_size; - if (iwmr->page_cnt > 1) { + if (iwpbl->pbl_allocated) { if (palloc->level == I40IW_LEVEL_1) { stag_info->first_pm_pbl_index = palloc->level1.idx; stag_info->chunk_size = 1; @@ -1699,6 +1817,11 @@ static struct ib_mr *i40iw_reg_user_mr(struct ib_pd *pd, bool use_pbles = false; unsigned long flags; int err = -ENOSYS; + int ret; + int pg_shift; + + if (iwdev->closing) + return ERR_PTR(-ENODEV); if (length > I40IW_MAX_MR_SIZE) return ERR_PTR(-EINVAL); @@ -1723,9 +1846,17 @@ static struct ib_mr *i40iw_reg_user_mr(struct ib_pd *pd, iwmr->ibmr.pd = pd; iwmr->ibmr.device = pd->device; ucontext = to_ucontext(pd->uobject->context); - region_length = region->length + (start & 0xfff); - pbl_depth = region_length >> 12; - pbl_depth += (region_length & (4096 - 1)) ? 1 : 0; + + iwmr->page_size = region->page_size; + iwmr->page_msk = PAGE_MASK; + + if (region->hugetlb && (req.reg_type == IW_MEMREG_TYPE_MEM)) + i40iw_set_hugetlb_values(start, iwmr); + + region_length = region->length + (start & (iwmr->page_size - 1)); + pg_shift = ffs(iwmr->page_size) - 1; + pbl_depth = region_length >> pg_shift; + pbl_depth += (region_length & (iwmr->page_size - 1)) ? 1 : 0; iwmr->length = region->length; iwpbl->user_base = virt; @@ -1755,13 +1886,21 @@ static struct ib_mr *i40iw_reg_user_mr(struct ib_pd *pd, spin_unlock_irqrestore(&ucontext->cq_reg_mem_list_lock, flags); break; case IW_MEMREG_TYPE_MEM: + use_pbles = (iwmr->page_cnt != 1); access = I40IW_ACCESS_FLAGS_LOCALREAD; - use_pbles = (iwmr->page_cnt != 1); err = i40iw_setup_pbles(iwdev, iwmr, use_pbles); if (err) goto error; + if (use_pbles) { + ret = i40iw_check_mr_contiguous(palloc, iwmr->page_size); + if (ret) { + i40iw_free_pble(iwdev->pble_rsrc, palloc); + iwpbl->pbl_allocated = false; + } + } + access |= i40iw_get_user_access(acc); stag = i40iw_create_stag(iwdev); if (!stag) { @@ -1778,6 +1917,7 @@ static struct ib_mr *i40iw_reg_user_mr(struct ib_pd *pd, i40iw_free_stag(iwdev, stag); goto error; } + break; default: goto error; @@ -1789,7 +1929,7 @@ static struct ib_mr *i40iw_reg_user_mr(struct ib_pd *pd, return &iwmr->ibmr; error: - if (palloc->level != I40IW_LEVEL_0) + if (palloc->level != I40IW_LEVEL_0 && iwpbl->pbl_allocated) i40iw_free_pble(iwdev->pble_rsrc, palloc); ib_umem_release(region); kfree(iwmr); @@ -2142,7 +2282,6 @@ static int i40iw_post_send(struct ib_qp *ibqp, case IB_WR_REG_MR: { struct i40iw_mr *iwmr = to_iwmr(reg_wr(ib_wr)->mr); - int page_shift = ilog2(reg_wr(ib_wr)->mr->page_size); int flags = reg_wr(ib_wr)->access; struct i40iw_pble_alloc *palloc = &iwmr->iwpbl.pble_alloc; struct i40iw_sc_dev *dev = &iwqp->iwdev->sc_dev; @@ -2153,6 +2292,7 @@ static int i40iw_post_send(struct ib_qp *ibqp, info.access_rights |= i40iw_get_user_access(flags); info.stag_key = reg_wr(ib_wr)->key & 0xff; info.stag_idx = reg_wr(ib_wr)->key >> 8; + info.page_size = reg_wr(ib_wr)->mr->page_size; info.wr_id = ib_wr->wr_id; info.addr_type = I40IW_ADDR_TYPE_VA_BASED; @@ -2166,9 +2306,6 @@ static int i40iw_post_send(struct ib_qp *ibqp, if (iwmr->npages > I40IW_MIN_PAGES_PER_FMR) info.chunk_size = 1; - if (page_shift == 21) - info.page_size = 1; /* 2M page */ - ret = dev->iw_priv_qp_ops->iw_mr_fast_register(&iwqp->sc_qp, &info, true); if (ret) err = -ENOMEM; @@ -2359,14 +2496,15 @@ static int i40iw_port_immutable(struct ib_device *ibdev, u8 port_num, struct ib_port_attr attr; int err; - err = i40iw_query_port(ibdev, port_num, &attr); + immutable->core_cap_flags = RDMA_CORE_PORT_IWARP; + + err = ib_query_port(ibdev, port_num, &attr); if (err) return err; immutable->pkey_tbl_len = attr.pkey_tbl_len; immutable->gid_tbl_len = attr.gid_tbl_len; - immutable->core_cap_flags = RDMA_CORE_PORT_IWARP; return 0; } @@ -2487,21 +2625,17 @@ static int i40iw_get_hw_stats(struct ib_device *ibdev, { struct i40iw_device *iwdev = to_iwdev(ibdev); struct i40iw_sc_dev *dev = &iwdev->sc_dev; - struct i40iw_dev_pestat *devstat = &dev->dev_pestat; + struct i40iw_vsi_pestat *devstat = iwdev->vsi.pestat; struct i40iw_dev_hw_stats *hw_stats = &devstat->hw_stats; - unsigned long flags; if (dev->is_pf) { - spin_lock_irqsave(&devstat->stats_lock, flags); - devstat->ops.iw_hw_stat_read_all(devstat, - &devstat->hw_stats); - spin_unlock_irqrestore(&devstat->stats_lock, flags); + i40iw_hw_stats_read_all(devstat, &devstat->hw_stats); } else { if (i40iw_vchnl_vf_get_pe_stats(dev, &devstat->hw_stats)) return -ENOSYS; } - memcpy(&stats->value[0], &hw_stats, sizeof(*hw_stats)); + memcpy(&stats->value[0], hw_stats, sizeof(*hw_stats)); return stats->num_counters; } @@ -2562,7 +2696,9 @@ static int i40iw_query_pkey(struct ib_device *ibdev, * @ah_attr: address handle attributes */ static struct ib_ah *i40iw_create_ah(struct ib_pd *ibpd, - struct ib_ah_attr *attr) + struct ib_ah_attr *attr, + struct ib_udata *udata) + { return ERR_PTR(-ENOSYS); } @@ -2621,8 +2757,7 @@ static struct i40iw_ib_device *i40iw_init_rdma_device(struct i40iw_device *iwdev (1ull << IB_USER_VERBS_CMD_POST_RECV) | (1ull << IB_USER_VERBS_CMD_POST_SEND); iwibdev->ibdev.phys_port_cnt = 1; - iwibdev->ibdev.num_comp_vectors = 1; - iwibdev->ibdev.dma_device = &pcidev->dev; + iwibdev->ibdev.num_comp_vectors = iwdev->ceqs_count; iwibdev->ibdev.dev.parent = &pcidev->dev; iwibdev->ibdev.query_port = i40iw_query_port; iwibdev->ibdev.modify_port = i40iw_modify_port; @@ -2654,7 +2789,6 @@ static struct i40iw_ib_device *i40iw_init_rdma_device(struct i40iw_device *iwdev iwibdev->ibdev.iwcm = kzalloc(sizeof(*iwibdev->ibdev.iwcm), GFP_KERNEL); if (!iwibdev->ibdev.iwcm) { ib_dealloc_device(&iwibdev->ibdev); - i40iw_pr_err("iwcm == NULL\n"); return NULL; } @@ -2719,6 +2853,9 @@ void i40iw_destroy_rdma_device(struct i40iw_ib_device *iwibdev) i40iw_unregister_rdma_device(iwibdev); kfree(iwibdev->ibdev.iwcm); iwibdev->ibdev.iwcm = NULL; + wait_event_timeout(iwibdev->iwdev->close_wq, + !atomic64_read(&iwibdev->iwdev->use_count), + I40IW_EVENT_TIMEOUT); ib_dealloc_device(&iwibdev->ibdev); } diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.h b/drivers/infiniband/hw/i40iw/i40iw_verbs.h index 0069be8a5a38..07c3fec77de6 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_verbs.h +++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.h @@ -42,6 +42,7 @@ struct i40iw_ucontext { spinlock_t cq_reg_mem_list_lock; /* memory list for cq's */ struct list_head qp_reg_mem_list; spinlock_t qp_reg_mem_list_lock; /* memory list for qp's */ + int abi_ver; }; struct i40iw_pd { @@ -92,6 +93,8 @@ struct i40iw_mr { struct ib_umem *region; u16 type; u32 page_cnt; + u32 page_size; + u64 page_msk; u32 npages; u32 stag; u64 length; diff --git a/drivers/infiniband/hw/i40iw/i40iw_virtchnl.c b/drivers/infiniband/hw/i40iw/i40iw_virtchnl.c index 3041003c94d2..f4d13683a403 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_virtchnl.c +++ b/drivers/infiniband/hw/i40iw/i40iw_virtchnl.c @@ -403,6 +403,19 @@ del_out: } /** + * i40iw_vf_init_pestat - Initialize stats for VF + * @devL pointer to the VF Device + * @stats: Statistics structure pointer + * @index: Stats index + */ +static void i40iw_vf_init_pestat(struct i40iw_sc_dev *dev, struct i40iw_vsi_pestat *stats, u16 index) +{ + stats->hw = dev->hw; + i40iw_hw_stats_init(stats, (u8)index, false); + spin_lock_init(&stats->lock); +} + +/** * i40iw_vchnl_recv_pf - Receive PF virtual channel messages * @dev: IWARP device pointer * @vf_id: Virtual function ID associated with the message @@ -421,9 +434,8 @@ enum i40iw_status_code i40iw_vchnl_recv_pf(struct i40iw_sc_dev *dev, u16 first_avail_iw_vf = I40IW_MAX_PE_ENABLED_VF_COUNT; struct i40iw_virt_mem vf_dev_mem; struct i40iw_virtchnl_work_info work_info; - struct i40iw_dev_pestat *devstat; + struct i40iw_vsi_pestat *stats; enum i40iw_status_code ret_code; - unsigned long flags; if (!dev || !msg || !len) return I40IW_ERR_PARAM; @@ -496,14 +508,7 @@ enum i40iw_status_code i40iw_vchnl_recv_pf(struct i40iw_sc_dev *dev, i40iw_debug(dev, I40IW_DEBUG_VIRT, "VF%u error CQP HMC Function operation.\n", vf_id); - ret_code = i40iw_device_init_pestat(&vf_dev->dev_pestat); - if (ret_code) - i40iw_debug(dev, I40IW_DEBUG_VIRT, - "VF%u - i40iw_device_init_pestat failed\n", - vf_id); - vf_dev->dev_pestat.ops.iw_hw_stat_init(&vf_dev->dev_pestat, - (u8)vf_dev->pmf_index, - dev->hw, false); + i40iw_vf_init_pestat(dev, &vf_dev->pestat, vf_dev->pmf_index); vf_dev->stats_initialized = true; } else { if (vf_dev) { @@ -534,12 +539,10 @@ enum i40iw_status_code i40iw_vchnl_recv_pf(struct i40iw_sc_dev *dev, case I40IW_VCHNL_OP_GET_STATS: if (!vf_dev) return I40IW_ERR_BAD_PTR; - devstat = &vf_dev->dev_pestat; - spin_lock_irqsave(&dev->dev_pestat.stats_lock, flags); - devstat->ops.iw_hw_stat_read_all(devstat, &devstat->hw_stats); - spin_unlock_irqrestore(&dev->dev_pestat.stats_lock, flags); + stats = &vf_dev->pestat; + i40iw_hw_stats_read_all(stats, &stats->hw_stats); vf_dev->msg_count--; - vchnl_pf_send_get_pe_stats_resp(dev, vf_id, vchnl_msg, &devstat->hw_stats); + vchnl_pf_send_get_pe_stats_resp(dev, vf_id, vchnl_msg, &stats->hw_stats); break; default: i40iw_debug(dev, I40IW_DEBUG_VIRT, diff --git a/drivers/infiniband/hw/mlx4/ah.c b/drivers/infiniband/hw/mlx4/ah.c index b9bf0759f10a..077c33d2dc75 100644 --- a/drivers/infiniband/hw/mlx4/ah.c +++ b/drivers/infiniband/hw/mlx4/ah.c @@ -114,7 +114,9 @@ static struct ib_ah *create_iboe_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr !(1 << ah->av.eth.stat_rate & dev->caps.stat_rate_support)) --ah->av.eth.stat_rate; } - + ah->av.eth.sl_tclass_flowlabel |= + cpu_to_be32((ah_attr->grh.traffic_class << 20) | + ah_attr->grh.flow_label); /* * HW requires multicast LID so we just choose one. */ @@ -122,12 +124,14 @@ static struct ib_ah *create_iboe_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr ah->av.ib.dlid = cpu_to_be16(0xc000); memcpy(ah->av.eth.dgid, ah_attr->grh.dgid.raw, 16); - ah->av.eth.sl_tclass_flowlabel = cpu_to_be32(ah_attr->sl << 29); + ah->av.eth.sl_tclass_flowlabel |= cpu_to_be32(ah_attr->sl << 29); return &ah->ibah; } -struct ib_ah *mlx4_ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr) +struct ib_ah *mlx4_ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, + struct ib_udata *udata) + { struct mlx4_ib_ah *ah; struct ib_ah *ret; diff --git a/drivers/infiniband/hw/mlx4/alias_GUID.c b/drivers/infiniband/hw/mlx4/alias_GUID.c index 5e9939045852..ea24230ea0d4 100644 --- a/drivers/infiniband/hw/mlx4/alias_GUID.c +++ b/drivers/infiniband/hw/mlx4/alias_GUID.c @@ -499,6 +499,7 @@ static int set_guid_rec(struct ib_device *ibdev, struct list_head *head = &dev->sriov.alias_guid.ports_guid[port - 1].cb_list; + memset(&attr, 0, sizeof(attr)); err = __mlx4_ib_query_port(ibdev, port, &attr, 1); if (err) { pr_debug("mlx4_ib_query_port failed (err: %d), port: %d\n", @@ -755,10 +756,8 @@ static void alias_guid_work(struct work_struct *work) struct mlx4_ib_dev *dev = container_of(ib_sriov, struct mlx4_ib_dev, sriov); rec = kzalloc(sizeof *rec, GFP_KERNEL); - if (!rec) { - pr_err("alias_guid_work: No Memory\n"); + if (!rec) return; - } pr_debug("starting [port: %d]...\n", sriov_alias_port->port + 1); ret = get_next_record_to_update(dev, sriov_alias_port->port, rec); diff --git a/drivers/infiniband/hw/mlx4/cm.c b/drivers/infiniband/hw/mlx4/cm.c index 39a488889fc7..d64845335e87 100644 --- a/drivers/infiniband/hw/mlx4/cm.c +++ b/drivers/infiniband/hw/mlx4/cm.c @@ -247,10 +247,8 @@ id_map_alloc(struct ib_device *ibdev, int slave_id, u32 sl_cm_id) struct mlx4_ib_sriov *sriov = &to_mdev(ibdev)->sriov; ent = kmalloc(sizeof (struct id_map_entry), GFP_KERNEL); - if (!ent) { - mlx4_ib_warn(ibdev, "Couldn't allocate id cache entry - out of memory\n"); + if (!ent) return ERR_PTR(-ENOMEM); - } ent->sl_cm_id = sl_cm_id; ent->slave_id = slave_id; diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c index 1672907ff219..db564ccc0f92 100644 --- a/drivers/infiniband/hw/mlx4/mad.c +++ b/drivers/infiniband/hw/mlx4/mad.c @@ -39,6 +39,8 @@ #include <linux/mlx4/cmd.h> #include <linux/gfp.h> #include <rdma/ib_pma.h> +#include <linux/ip.h> +#include <net/ipv6.h> #include <linux/mlx4/driver.h> #include "mlx4_ib.h" @@ -480,6 +482,23 @@ static int find_slave_port_pkey_ix(struct mlx4_ib_dev *dev, int slave, return -EINVAL; } +static int get_gids_from_l3_hdr(struct ib_grh *grh, union ib_gid *sgid, + union ib_gid *dgid) +{ + int version = ib_get_rdma_header_version((const union rdma_network_hdr *)grh); + enum rdma_network_type net_type; + + if (version == 4) + net_type = RDMA_NETWORK_IPV4; + else if (version == 6) + net_type = RDMA_NETWORK_IPV6; + else + return -EINVAL; + + return ib_get_gids_from_rdma_hdr((union rdma_network_hdr *)grh, net_type, + sgid, dgid); +} + int mlx4_ib_send_to_slave(struct mlx4_ib_dev *dev, int slave, u8 port, enum ib_qp_type dest_qpt, struct ib_wc *wc, struct ib_grh *grh, struct ib_mad *mad) @@ -538,7 +557,10 @@ int mlx4_ib_send_to_slave(struct mlx4_ib_dev *dev, int slave, u8 port, memset(&attr, 0, sizeof attr); attr.port_num = port; if (is_eth) { - memcpy(&attr.grh.dgid.raw[0], &grh->dgid.raw[0], 16); + union ib_gid sgid; + + if (get_gids_from_l3_hdr(grh, &sgid, &attr.grh.dgid)) + return -EINVAL; attr.ah_flags = IB_AH_GRH; } ah = ib_create_ah(tun_ctx->pd, &attr); @@ -651,6 +673,11 @@ static int mlx4_ib_demux_mad(struct ib_device *ibdev, u8 port, is_eth = 1; if (is_eth) { + union ib_gid dgid; + union ib_gid sgid; + + if (get_gids_from_l3_hdr(grh, &sgid, &dgid)) + return -EINVAL; if (!(wc->wc_flags & IB_WC_GRH)) { mlx4_ib_warn(ibdev, "RoCE grh not present.\n"); return -EINVAL; @@ -659,10 +686,10 @@ static int mlx4_ib_demux_mad(struct ib_device *ibdev, u8 port, mlx4_ib_warn(ibdev, "RoCE mgmt class is not CM\n"); return -EINVAL; } - err = mlx4_get_slave_from_roce_gid(dev->dev, port, grh->dgid.raw, &slave); + err = mlx4_get_slave_from_roce_gid(dev->dev, port, dgid.raw, &slave); if (err && mlx4_is_mf_bonded(dev->dev)) { other_port = (port == 1) ? 2 : 1; - err = mlx4_get_slave_from_roce_gid(dev->dev, other_port, grh->dgid.raw, &slave); + err = mlx4_get_slave_from_roce_gid(dev->dev, other_port, dgid.raw, &slave); if (!err) { port = other_port; pr_debug("resolved slave %d from gid %pI6 wire port %d other %d\n", @@ -702,10 +729,18 @@ static int mlx4_ib_demux_mad(struct ib_device *ibdev, u8 port, /* If a grh is present, we demux according to it */ if (wc->wc_flags & IB_WC_GRH) { - slave = mlx4_ib_find_real_gid(ibdev, port, grh->dgid.global.interface_id); - if (slave < 0) { - mlx4_ib_warn(ibdev, "failed matching grh\n"); - return -ENOENT; + if (grh->dgid.global.interface_id == + cpu_to_be64(IB_SA_WELL_KNOWN_GUID) && + grh->dgid.global.subnet_prefix == cpu_to_be64( + atomic64_read(&dev->sriov.demux[port - 1].subnet_prefix))) { + slave = 0; + } else { + slave = mlx4_ib_find_real_gid(ibdev, port, + grh->dgid.global.interface_id); + if (slave < 0) { + mlx4_ib_warn(ibdev, "failed matching grh\n"); + return -ENOENT; + } } } /* Class-specific handling */ @@ -1102,10 +1137,8 @@ static void handle_slaves_guid_change(struct mlx4_ib_dev *dev, u8 port_num, in_mad = kmalloc(sizeof *in_mad, GFP_KERNEL); out_mad = kmalloc(sizeof *out_mad, GFP_KERNEL); - if (!in_mad || !out_mad) { - mlx4_ib_warn(&dev->ib_dev, "failed to allocate memory for guid info mads\n"); + if (!in_mad || !out_mad) goto out; - } guid_tbl_blk_num *= 4; @@ -1916,11 +1949,8 @@ static int alloc_pv_object(struct mlx4_ib_dev *dev, int slave, int port, *ret_ctx = NULL; ctx = kzalloc(sizeof (struct mlx4_ib_demux_pv_ctx), GFP_KERNEL); - if (!ctx) { - pr_err("failed allocating pv resource context " - "for port %d, slave %d\n", port, slave); + if (!ctx) return -ENOMEM; - } ctx->ib_dev = &dev->ib_dev; ctx->port = port; diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index b597e8227591..88608906ce25 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -430,7 +430,7 @@ static int mlx4_ib_query_device(struct ib_device *ibdev, struct mlx4_ib_dev *dev = to_mdev(ibdev); struct ib_smp *in_mad = NULL; struct ib_smp *out_mad = NULL; - int err = -ENOMEM; + int err; int have_ib_ports; struct mlx4_uverbs_ex_query_device cmd; struct mlx4_uverbs_ex_query_device_resp resp = {.comp_mask = 0}; @@ -455,6 +455,7 @@ static int mlx4_ib_query_device(struct ib_device *ibdev, sizeof(resp.response_length); in_mad = kzalloc(sizeof *in_mad, GFP_KERNEL); out_mad = kmalloc(sizeof *out_mad, GFP_KERNEL); + err = -ENOMEM; if (!in_mad || !out_mad) goto out; @@ -547,6 +548,7 @@ static int mlx4_ib_query_device(struct ib_device *ibdev, props->max_map_per_fmr = dev->dev->caps.max_fmr_maps; props->hca_core_clock = dev->dev->caps.hca_core_clock * 1000UL; props->timestamp_mask = 0xFFFFFFFFFFFFULL; + props->max_ah = INT_MAX; if (!mlx4_is_slave(dev->dev)) err = mlx4_get_internal_clock_params(dev->dev, &clock_params); @@ -676,7 +678,7 @@ static u8 state_to_phys_state(enum ib_port_state state) } static int eth_link_query_port(struct ib_device *ibdev, u8 port, - struct ib_port_attr *props, int netw_view) + struct ib_port_attr *props) { struct mlx4_ib_dev *mdev = to_mdev(ibdev); @@ -697,9 +699,11 @@ static int eth_link_query_port(struct ib_device *ibdev, u8 port, if (err) goto out; - props->active_width = (((u8 *)mailbox->buf)[5] == 0x40) ? - IB_WIDTH_4X : IB_WIDTH_1X; - props->active_speed = IB_SPEED_QDR; + props->active_width = (((u8 *)mailbox->buf)[5] == 0x40) || + (((u8 *)mailbox->buf)[5] == 0x20 /*56Gb*/) ? + IB_WIDTH_4X : IB_WIDTH_1X; + props->active_speed = (((u8 *)mailbox->buf)[5] == 0x20 /*56Gb*/) ? + IB_SPEED_FDR : IB_SPEED_QDR; props->port_cap_flags = IB_PORT_CM_SUP | IB_PORT_IP_BASED_GIDS; props->gid_tbl_len = mdev->dev->caps.gid_table_len[port]; props->max_msg_sz = mdev->dev->caps.max_msg_sz; @@ -737,11 +741,11 @@ int __mlx4_ib_query_port(struct ib_device *ibdev, u8 port, { int err; - memset(props, 0, sizeof *props); + /* props being zeroed by the caller, avoid zeroing it here */ err = mlx4_ib_port_link_layer(ibdev, port) == IB_LINK_LAYER_INFINIBAND ? ib_link_query_port(ibdev, port, props, netw_view) : - eth_link_query_port(ibdev, port, props, netw_view); + eth_link_query_port(ibdev, port, props); return err; } @@ -1010,7 +1014,7 @@ static int mlx4_ib_modify_port(struct ib_device *ibdev, u8 port, int mask, mutex_lock(&mdev->cap_mask_mutex); - err = mlx4_ib_query_port(ibdev, port, &attr); + err = ib_query_port(ibdev, port, &attr); if (err) goto out; @@ -1678,9 +1682,19 @@ static int __mlx4_ib_create_flow(struct ib_qp *qp, struct ib_flow_attr *flow_att size += ret; } + if (mlx4_is_master(mdev->dev) && flow_type == MLX4_FS_REGULAR && + flow_attr->num_of_specs == 1) { + struct _rule_hw *rule_header = (struct _rule_hw *)(ctrl + 1); + enum ib_flow_spec_type header_spec = + ((union ib_flow_spec *)(flow_attr + 1))->type; + + if (header_spec == IB_FLOW_SPEC_ETH) + mlx4_handle_eth_header_mcast_prio(ctrl, rule_header); + } + ret = mlx4_cmd_imm(mdev->dev, mailbox->dma, reg_id, size >> 2, 0, MLX4_QP_FLOW_STEERING_ATTACH, MLX4_CMD_TIME_CLASS_A, - MLX4_CMD_WRAPPED); + MLX4_CMD_NATIVE); if (ret == -ENOMEM) pr_err("mcg table is full. Fail to register network rule.\n"); else if (ret == -ENXIO) @@ -1697,7 +1711,7 @@ static int __mlx4_ib_destroy_flow(struct mlx4_dev *dev, u64 reg_id) int err; err = mlx4_cmd(dev, reg_id, 0, 0, MLX4_QP_FLOW_STEERING_DETACH, MLX4_CMD_TIME_CLASS_A, - MLX4_CMD_WRAPPED); + MLX4_CMD_NATIVE); if (err) pr_err("Fail to detach network rule. registration id = 0x%llx\n", reg_id); @@ -2523,24 +2537,27 @@ static int mlx4_port_immutable(struct ib_device *ibdev, u8 port_num, struct mlx4_ib_dev *mdev = to_mdev(ibdev); int err; - err = mlx4_ib_query_port(ibdev, port_num, &attr); - if (err) - return err; - - immutable->pkey_tbl_len = attr.pkey_tbl_len; - immutable->gid_tbl_len = attr.gid_tbl_len; - if (mlx4_ib_port_link_layer(ibdev, port_num) == IB_LINK_LAYER_INFINIBAND) { immutable->core_cap_flags = RDMA_CORE_PORT_IBA_IB; + immutable->max_mad_size = IB_MGMT_MAD_SIZE; } else { if (mdev->dev->caps.flags & MLX4_DEV_CAP_FLAG_IBOE) immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE; if (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ROCE_V1_V2) immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE | RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP; + immutable->core_cap_flags |= RDMA_CORE_PORT_RAW_PACKET; + if (immutable->core_cap_flags & (RDMA_CORE_PORT_IBA_ROCE | + RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP)) + immutable->max_mad_size = IB_MGMT_MAD_SIZE; } - immutable->max_mad_size = IB_MGMT_MAD_SIZE; + err = ib_query_port(ibdev, port_num, &attr); + if (err) + return err; + + immutable->pkey_tbl_len = attr.pkey_tbl_len; + immutable->gid_tbl_len = attr.gid_tbl_len; return 0; } @@ -2611,7 +2628,7 @@ static void *mlx4_ib_add(struct mlx4_dev *dev) ibdev->ib_dev.phys_port_cnt = mlx4_is_bonded(dev) ? 1 : ibdev->num_ports; ibdev->ib_dev.num_comp_vectors = dev->caps.num_comp_vectors; - ibdev->ib_dev.dma_device = &dev->persist->pdev->dev; + ibdev->ib_dev.dev.parent = &dev->persist->pdev->dev; ibdev->ib_dev.get_netdev = mlx4_ib_get_netdev; ibdev->ib_dev.add_gid = mlx4_ib_add_gid; ibdev->ib_dev.del_gid = mlx4_ib_del_gid; @@ -2814,20 +2831,22 @@ static void *mlx4_ib_add(struct mlx4_dev *dev) kmalloc(BITS_TO_LONGS(ibdev->steer_qpn_count) * sizeof(long), GFP_KERNEL); - if (!ibdev->ib_uc_qpns_bitmap) { - dev_err(&dev->persist->pdev->dev, - "bit map alloc failed\n"); + if (!ibdev->ib_uc_qpns_bitmap) goto err_steer_qp_release; - } - - bitmap_zero(ibdev->ib_uc_qpns_bitmap, ibdev->steer_qpn_count); - err = mlx4_FLOW_STEERING_IB_UC_QP_RANGE( - dev, ibdev->steer_qpn_base, - ibdev->steer_qpn_base + - ibdev->steer_qpn_count - 1); - if (err) - goto err_steer_free_bitmap; + if (dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_DMFS_IPOIB) { + bitmap_zero(ibdev->ib_uc_qpns_bitmap, + ibdev->steer_qpn_count); + err = mlx4_FLOW_STEERING_IB_UC_QP_RANGE( + dev, ibdev->steer_qpn_base, + ibdev->steer_qpn_base + + ibdev->steer_qpn_count - 1); + if (err) + goto err_steer_free_bitmap; + } else { + bitmap_fill(ibdev->ib_uc_qpns_bitmap, + ibdev->steer_qpn_count); + } } for (j = 1; j <= ibdev->dev->caps.num_ports; j++) @@ -3055,15 +3074,12 @@ static void do_slave_init(struct mlx4_ib_dev *ibdev, int slave, int do_init) first_port = find_first_bit(actv_ports.ports, dev->caps.num_ports); dm = kcalloc(ports, sizeof(*dm), GFP_ATOMIC); - if (!dm) { - pr_err("failed to allocate memory for tunneling qp update\n"); + if (!dm) return; - } for (i = 0; i < ports; i++) { dm[i] = kmalloc(sizeof (struct mlx4_ib_demux_work), GFP_ATOMIC); if (!dm[i]) { - pr_err("failed to allocate memory for tunneling qp update work struct\n"); while (--i >= 0) kfree(dm[i]); goto out; @@ -3223,8 +3239,6 @@ void mlx4_sched_ib_sl2vl_update_work(struct mlx4_ib_dev *ibdev, ew->port = port; ew->ib_dev = ibdev; queue_work(wq, &ew->work); - } else { - pr_err("failed to allocate memory for sl2vl update work\n"); } } @@ -3284,10 +3298,8 @@ static void mlx4_ib_event(struct mlx4_dev *dev, void *ibdev_ptr, case MLX4_DEV_EVENT_PORT_MGMT_CHANGE: ew = kmalloc(sizeof *ew, GFP_ATOMIC); - if (!ew) { - pr_err("failed to allocate memory for events work\n"); + if (!ew) break; - } INIT_WORK(&ew->work, handle_port_mgmt_change_event); memcpy(&ew->ib_eqe, eqe, sizeof *eqe); diff --git a/drivers/infiniband/hw/mlx4/mcg.c b/drivers/infiniband/hw/mlx4/mcg.c index a21d37f02f35..e010fe459e67 100644 --- a/drivers/infiniband/hw/mlx4/mcg.c +++ b/drivers/infiniband/hw/mlx4/mcg.c @@ -1142,7 +1142,6 @@ void mlx4_ib_mcg_port_cleanup(struct mlx4_ib_demux_ctx *ctx, int destroy_wq) work = kmalloc(sizeof *work, GFP_KERNEL); if (!work) { ctx->flushing = 0; - mcg_warn("failed allocating work for cleanup\n"); return; } @@ -1202,10 +1201,8 @@ static int push_deleteing_req(struct mcast_group *group, int slave) return 0; req = kzalloc(sizeof *req, GFP_KERNEL); - if (!req) { - mcg_warn_group(group, "failed allocation - may leave stall groups\n"); + if (!req) return -ENOMEM; - } if (!list_empty(&group->func[slave].pending)) { pend_req = list_entry(group->func[slave].pending.prev, struct mcast_req, group_list); diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h index 35141f451e5c..64fed44b43a6 100644 --- a/drivers/infiniband/hw/mlx4/mlx4_ib.h +++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h @@ -55,7 +55,7 @@ #define pr_fmt(fmt) "<" MLX4_IB_DRV_NAME "> %s: " fmt, __func__ #define mlx4_ib_warn(ibdev, format, arg...) \ - dev_warn((ibdev)->dma_device, MLX4_IB_DRV_NAME ": " format, ## arg) + dev_warn((ibdev)->dev.parent, MLX4_IB_DRV_NAME ": " format, ## arg) enum { MLX4_IB_SQ_MIN_WQE_SHIFT = 6, @@ -742,7 +742,8 @@ int mlx4_ib_arm_cq(struct ib_cq *cq, enum ib_cq_notify_flags flags); void __mlx4_ib_cq_clean(struct mlx4_ib_cq *cq, u32 qpn, struct mlx4_ib_srq *srq); void mlx4_ib_cq_clean(struct mlx4_ib_cq *cq, u32 qpn, struct mlx4_ib_srq *srq); -struct ib_ah *mlx4_ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr); +struct ib_ah *mlx4_ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, + struct ib_udata *udata); int mlx4_ib_query_ah(struct ib_ah *ibah, struct ib_ah_attr *ah_attr); int mlx4_ib_destroy_ah(struct ib_ah *ah); diff --git a/drivers/infiniband/hw/mlx4/mr.c b/drivers/infiniband/hw/mlx4/mr.c index 5d73989d9771..433bcdbdd680 100644 --- a/drivers/infiniband/hw/mlx4/mr.c +++ b/drivers/infiniband/hw/mlx4/mr.c @@ -292,10 +292,10 @@ mlx4_alloc_priv_pages(struct ib_device *device, if (!mr->pages) return -ENOMEM; - mr->page_map = dma_map_single(device->dma_device, mr->pages, + mr->page_map = dma_map_single(device->dev.parent, mr->pages, mr->page_map_size, DMA_TO_DEVICE); - if (dma_mapping_error(device->dma_device, mr->page_map)) { + if (dma_mapping_error(device->dev.parent, mr->page_map)) { ret = -ENOMEM; goto err; } @@ -313,7 +313,7 @@ mlx4_free_priv_pages(struct mlx4_ib_mr *mr) if (mr->pages) { struct ib_device *device = mr->ibmr.device; - dma_unmap_single(device->dma_device, mr->page_map, + dma_unmap_single(device->dev.parent, mr->page_map, mr->page_map_size, DMA_TO_DEVICE); free_page((unsigned long)mr->pages); mr->pages = NULL; diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c index 570bc866b1d6..c34eebc7db65 100644 --- a/drivers/infiniband/hw/mlx4/qp.c +++ b/drivers/infiniband/hw/mlx4/qp.c @@ -76,10 +76,6 @@ enum { MLX4_IB_LSO_HEADER_SPARE = 128, }; -enum { - MLX4_IB_IBOE_ETHERTYPE = 0x8915 -}; - struct mlx4_ib_sqp { struct mlx4_ib_qp qp; int pkey_index; @@ -644,7 +640,7 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, int qpn; int err; struct ib_qp_cap backup_cap; - struct mlx4_ib_sqp *sqp; + struct mlx4_ib_sqp *sqp = NULL; struct mlx4_ib_qp *qp; enum mlx4_ib_qp_type qp_type = (enum mlx4_ib_qp_type) init_attr->qp_type; struct mlx4_ib_cq *mcq; @@ -933,7 +929,9 @@ err_db: mlx4_db_free(dev->dev, &qp->db); err: - if (!*caller_qp) + if (sqp) + kfree(sqp); + else if (!*caller_qp) kfree(qp); return err; } @@ -1280,7 +1278,8 @@ static int _mlx4_ib_destroy_qp(struct ib_qp *qp) if (is_qp0(dev, mqp)) mlx4_CLOSE_PORT(dev->dev, mqp->port); - if (dev->qp1_proxy[mqp->port - 1] == mqp) { + if (mqp->mlx4_ib_qp_type == MLX4_IB_QPT_PROXY_GSI && + dev->qp1_proxy[mqp->port - 1] == mqp) { mutex_lock(&dev->qp1_proxy_lock[mqp->port - 1]); dev->qp1_proxy[mqp->port - 1] = NULL; mutex_unlock(&dev->qp1_proxy_lock[mqp->port - 1]); @@ -1764,14 +1763,14 @@ static int __mlx4_ib_modify_qp(struct ib_qp *ibqp, u8 port_num = mlx4_is_bonded(to_mdev(ibqp->device)->dev) ? 1 : attr_mask & IB_QP_PORT ? attr->port_num : qp->port; union ib_gid gid; - struct ib_gid_attr gid_attr; + struct ib_gid_attr gid_attr = {.gid_type = IB_GID_TYPE_IB}; u16 vlan = 0xffff; u8 smac[ETH_ALEN]; int status = 0; int is_eth = rdma_cap_eth_ah(&dev->ib_dev, port_num) && attr->ah_attr.ah_flags & IB_AH_GRH; - if (is_eth) { + if (is_eth && attr->ah_attr.ah_flags & IB_AH_GRH) { int index = attr->ah_attr.grh.sgid_index; status = ib_get_cached_gid(ibqp->device, port_num, @@ -2421,11 +2420,31 @@ static u8 sl_to_vl(struct mlx4_ib_dev *dev, u8 sl, int port_num) return vl; } +static int fill_gid_by_hw_index(struct mlx4_ib_dev *ibdev, u8 port_num, + int index, union ib_gid *gid, + enum ib_gid_type *gid_type) +{ + struct mlx4_ib_iboe *iboe = &ibdev->iboe; + struct mlx4_port_gid_table *port_gid_table; + unsigned long flags; + + port_gid_table = &iboe->gids[port_num - 1]; + spin_lock_irqsave(&iboe->lock, flags); + memcpy(gid, &port_gid_table->gids[index].gid, sizeof(*gid)); + *gid_type = port_gid_table->gids[index].gid_type; + spin_unlock_irqrestore(&iboe->lock, flags); + if (!memcmp(gid, &zgid, sizeof(*gid))) + return -ENOENT; + + return 0; +} + #define MLX4_ROCEV2_QP1_SPORT 0xC000 static int build_mlx_header(struct mlx4_ib_sqp *sqp, struct ib_ud_wr *wr, void *wqe, unsigned *mlx_seg_len) { struct ib_device *ib_dev = sqp->qp.ibqp.device; + struct mlx4_ib_dev *ibdev = to_mdev(ib_dev); struct mlx4_wqe_mlx_seg *mlx = wqe; struct mlx4_wqe_ctrl_seg *ctrl = wqe; struct mlx4_wqe_inline_seg *inl = wqe + sizeof *mlx; @@ -2451,8 +2470,7 @@ static int build_mlx_header(struct mlx4_ib_sqp *sqp, struct ib_ud_wr *wr, is_eth = rdma_port_get_link_layer(sqp->qp.ibqp.device, sqp->qp.port) == IB_LINK_LAYER_ETHERNET; is_grh = mlx4_ib_ah_grh_present(ah); if (is_eth) { - struct ib_gid_attr gid_attr; - + enum ib_gid_type gid_type; if (mlx4_is_mfunc(to_mdev(ib_dev)->dev)) { /* When multi-function is enabled, the ib_core gid * indexes don't necessarily match the hw ones, so @@ -2463,18 +2481,11 @@ static int build_mlx_header(struct mlx4_ib_sqp *sqp, struct ib_ud_wr *wr, if (err) return err; } else { - err = ib_get_cached_gid(ib_dev, - be32_to_cpu(ah->av.ib.port_pd) >> 24, - ah->av.ib.gid_index, &sgid, - &gid_attr); + err = fill_gid_by_hw_index(ibdev, sqp->qp.port, + ah->av.ib.gid_index, + &sgid, &gid_type); if (!err) { - if (gid_attr.ndev) - dev_put(gid_attr.ndev); - if (!memcmp(&sgid, &zgid, sizeof(sgid))) - err = -ENOENT; - } - if (!err) { - is_udp = gid_attr.gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP; + is_udp = gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP; if (is_udp) { if (ipv6_addr_v4mapped((struct in6_addr *)&sgid)) ip_version = 4; @@ -2585,7 +2596,7 @@ static int build_mlx_header(struct mlx4_ib_sqp *sqp, struct ib_ud_wr *wr, u16 ether_type; u16 pcp = (be32_to_cpu(ah->av.ib.sl_tclass_flowlabel) >> 29) << 13; - ether_type = (!is_udp) ? MLX4_IB_IBOE_ETHERTYPE : + ether_type = (!is_udp) ? ETH_P_IBOE: (ip_version == 4 ? ETH_P_IP : ETH_P_IPV6); mlx->sched_prio = cpu_to_be16(pcp); @@ -2952,21 +2963,17 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, if (sqp->roce_v2_gsi) { struct mlx4_ib_ah *ah = to_mah(ud_wr(wr)->ah); - struct ib_gid_attr gid_attr; + enum ib_gid_type gid_type; union ib_gid gid; - if (!ib_get_cached_gid(ibqp->device, - be32_to_cpu(ah->av.ib.port_pd) >> 24, - ah->av.ib.gid_index, &gid, - &gid_attr)) { - if (gid_attr.ndev) - dev_put(gid_attr.ndev); - qp = (gid_attr.gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP) ? - to_mqp(sqp->roce_v2_gsi) : qp; - } else { + if (!fill_gid_by_hw_index(mdev, sqp->qp.port, + ah->av.ib.gid_index, + &gid, &gid_type)) + qp = (gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP) ? + to_mqp(sqp->roce_v2_gsi) : qp; + else pr_err("Failed to get gid at index %d. RoCEv2 will not work properly\n", ah->av.ib.gid_index); - } } } diff --git a/drivers/infiniband/hw/mlx4/sysfs.c b/drivers/infiniband/hw/mlx4/sysfs.c index 69fb5ba94d0f..0ba5ba7540c8 100644 --- a/drivers/infiniband/hw/mlx4/sysfs.c +++ b/drivers/infiniband/hw/mlx4/sysfs.c @@ -226,6 +226,7 @@ static int add_port_entries(struct mlx4_ib_dev *device, int port_num) int ret = 0 ; struct ib_port_attr attr; + memset(&attr, 0, sizeof(attr)); /* get the physical gid and pkey table sizes.*/ ret = __mlx4_ib_query_port(&device->ib_dev, port_num, &attr, 1); if (ret) diff --git a/drivers/infiniband/hw/mlx5/Makefile b/drivers/infiniband/hw/mlx5/Makefile index 7493a83acd28..90ad2adc752f 100644 --- a/drivers/infiniband/hw/mlx5/Makefile +++ b/drivers/infiniband/hw/mlx5/Makefile @@ -1,4 +1,4 @@ obj-$(CONFIG_MLX5_INFINIBAND) += mlx5_ib.o -mlx5_ib-y := main.o cq.o doorbell.o qp.o mem.o srq.o mr.o ah.o mad.o gsi.o ib_virt.o +mlx5_ib-y := main.o cq.o doorbell.o qp.o mem.o srq.o mr.o ah.o mad.o gsi.o ib_virt.o cmd.o mlx5_ib-$(CONFIG_INFINIBAND_ON_DEMAND_PAGING) += odp.o diff --git a/drivers/infiniband/hw/mlx5/ah.c b/drivers/infiniband/hw/mlx5/ah.c index 745efa4cfc71..d090e96f6f01 100644 --- a/drivers/infiniband/hw/mlx5/ah.c +++ b/drivers/infiniband/hw/mlx5/ah.c @@ -64,7 +64,9 @@ static struct ib_ah *create_ib_ah(struct mlx5_ib_dev *dev, return &ah->ibah; } -struct ib_ah *mlx5_ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr) +struct ib_ah *mlx5_ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, + struct ib_udata *udata) + { struct mlx5_ib_ah *ah; struct mlx5_ib_dev *dev = to_mdev(pd->device); @@ -75,6 +77,27 @@ struct ib_ah *mlx5_ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr) if (ll == IB_LINK_LAYER_ETHERNET && !(ah_attr->ah_flags & IB_AH_GRH)) return ERR_PTR(-EINVAL); + if (ll == IB_LINK_LAYER_ETHERNET && udata) { + int err; + struct mlx5_ib_create_ah_resp resp = {}; + u32 min_resp_len = offsetof(typeof(resp), dmac) + + sizeof(resp.dmac); + + if (udata->outlen < min_resp_len) + return ERR_PTR(-EINVAL); + + resp.response_length = min_resp_len; + + err = ib_resolve_eth_dmac(pd->device, ah_attr); + if (err) + return ERR_PTR(err); + + memcpy(resp.dmac, ah_attr->dmac, ETH_ALEN); + err = ib_copy_to_udata(udata, &resp, resp.response_length); + if (err) + return ERR_PTR(err); + } + ah = kzalloc(sizeof(*ah), GFP_ATOMIC); if (!ah) return ERR_PTR(-ENOMEM); diff --git a/drivers/infiniband/hw/mlx5/cmd.c b/drivers/infiniband/hw/mlx5/cmd.c new file mode 100644 index 000000000000..cdc2d3017da7 --- /dev/null +++ b/drivers/infiniband/hw/mlx5/cmd.c @@ -0,0 +1,48 @@ +/* + * Copyright (c) 2017, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "cmd.h" + +int mlx5_cmd_null_mkey(struct mlx5_core_dev *dev, u32 *null_mkey) +{ + u32 out[MLX5_ST_SZ_DW(query_special_contexts_out)] = {}; + u32 in[MLX5_ST_SZ_DW(query_special_contexts_in)] = {}; + int err; + + MLX5_SET(query_special_contexts_in, in, opcode, + MLX5_CMD_OP_QUERY_SPECIAL_CONTEXTS); + err = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out)); + if (!err) + *null_mkey = MLX5_GET(query_special_contexts_out, out, + null_mkey); + return err; +} diff --git a/drivers/infiniband/hw/hns/hns_roce_user.h b/drivers/infiniband/hw/mlx5/cmd.h index a28f761a9f65..7ca8a7b6434d 100644 --- a/drivers/infiniband/hw/hns/hns_roce_user.h +++ b/drivers/infiniband/hw/mlx5/cmd.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2016 Hisilicon Limited. + * Copyright (c) 2017, Mellanox Technologies. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU @@ -30,24 +30,11 @@ * SOFTWARE. */ -#ifndef _HNS_ROCE_USER_H -#define _HNS_ROCE_USER_H +#ifndef MLX5_IB_CMD_H +#define MLX5_IB_CMD_H -struct hns_roce_ib_create_cq { - __u64 buf_addr; -}; +#include <linux/kernel.h> +#include <linux/mlx5/driver.h> -struct hns_roce_ib_create_qp { - __u64 buf_addr; - __u64 db_addr; - __u8 log_sq_bb_count; - __u8 log_sq_stride; - __u8 sq_no_prefetch; - __u8 reserved[5]; -}; - -struct hns_roce_ib_alloc_ucontext_resp { - __u32 qp_tab_size; -}; - -#endif /*_HNS_ROCE_USER_H */ +int mlx5_cmd_null_mkey(struct mlx5_core_dev *dev, u32 *null_mkey); +#endif /* MLX5_IB_CMD_H */ diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c index fcd04b881ec1..31803b367104 100644 --- a/drivers/infiniband/hw/mlx5/cq.c +++ b/drivers/infiniband/hw/mlx5/cq.c @@ -689,7 +689,7 @@ int mlx5_ib_arm_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags flags) { struct mlx5_core_dev *mdev = to_mdev(ibcq->device)->mdev; struct mlx5_ib_cq *cq = to_mcq(ibcq); - void __iomem *uar_page = mdev->priv.uuari.uars[0].map; + void __iomem *uar_page = mdev->priv.uar->map; unsigned long irq_flags; int ret = 0; @@ -704,9 +704,7 @@ int mlx5_ib_arm_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags flags) mlx5_cq_arm(&cq->mcq, (flags & IB_CQ_SOLICITED_MASK) == IB_CQ_SOLICITED ? MLX5_CQ_DB_REQ_NOT_SOL : MLX5_CQ_DB_REQ_NOT, - uar_page, - MLX5_GET_DOORBELL_LOCK(&mdev->priv.cq_uar_lock), - to_mcq(ibcq)->mcq.cons_index); + uar_page, to_mcq(ibcq)->mcq.cons_index); return ret; } @@ -731,7 +729,7 @@ static int create_cq_user(struct mlx5_ib_dev *dev, struct ib_udata *udata, int entries, u32 **cqb, int *cqe_size, int *index, int *inlen) { - struct mlx5_ib_create_cq ucmd; + struct mlx5_ib_create_cq ucmd = {}; size_t ucmdlen; int page_shift; __be64 *pas; @@ -770,7 +768,7 @@ static int create_cq_user(struct mlx5_ib_dev *dev, struct ib_udata *udata, if (err) goto err_umem; - mlx5_ib_cont_pages(cq->buf.umem, ucmd.buf_addr, &npages, &page_shift, + mlx5_ib_cont_pages(cq->buf.umem, ucmd.buf_addr, 0, &npages, &page_shift, &ncont, NULL); mlx5_ib_dbg(dev, "addr 0x%llx, size %u, npages %d, page_shift %d, ncont %d\n", ucmd.buf_addr, entries * ucmd.cqe_size, npages, page_shift, ncont); @@ -790,10 +788,38 @@ static int create_cq_user(struct mlx5_ib_dev *dev, struct ib_udata *udata, MLX5_SET(cqc, cqc, log_page_size, page_shift - MLX5_ADAPTER_PAGE_SHIFT); - *index = to_mucontext(context)->uuari.uars[0].index; + *index = to_mucontext(context)->bfregi.sys_pages[0]; + + if (ucmd.cqe_comp_en == 1) { + if (unlikely((*cqe_size != 64) || + !MLX5_CAP_GEN(dev->mdev, cqe_compression))) { + err = -EOPNOTSUPP; + mlx5_ib_warn(dev, "CQE compression is not supported for size %d!\n", + *cqe_size); + goto err_cqb; + } + + if (unlikely(!ucmd.cqe_comp_res_format || + !(ucmd.cqe_comp_res_format < + MLX5_IB_CQE_RES_RESERVED) || + (ucmd.cqe_comp_res_format & + (ucmd.cqe_comp_res_format - 1)))) { + err = -EOPNOTSUPP; + mlx5_ib_warn(dev, "CQE compression res format %d is not supported!\n", + ucmd.cqe_comp_res_format); + goto err_cqb; + } + + MLX5_SET(cqc, cqc, cqe_comp_en, 1); + MLX5_SET(cqc, cqc, mini_cqe_res_format, + ilog2(ucmd.cqe_comp_res_format)); + } return 0; +err_cqb: + kfree(cqb); + err_db: mlx5_ib_db_unmap_user(to_mucontext(context), &cq->db); @@ -858,7 +884,7 @@ static int create_cq_kernel(struct mlx5_ib_dev *dev, struct mlx5_ib_cq *cq, MLX5_SET(cqc, cqc, log_page_size, cq->buf.buf.page_shift - MLX5_ADAPTER_PAGE_SHIFT); - *index = dev->mdev->priv.uuari.uars[0].index; + *index = dev->mdev->priv.uar->index; return 0; @@ -1124,7 +1150,7 @@ static int resize_user(struct mlx5_ib_dev *dev, struct mlx5_ib_cq *cq, return err; } - mlx5_ib_cont_pages(umem, ucmd.buf_addr, &npages, page_shift, + mlx5_ib_cont_pages(umem, ucmd.buf_addr, 0, &npages, page_shift, npas, NULL); cq->resize_umem = umem; diff --git a/drivers/infiniband/hw/mlx5/mad.c b/drivers/infiniband/hw/mlx5/mad.c index 39e58489dcc2..8dacb49eabd9 100644 --- a/drivers/infiniband/hw/mlx5/mad.c +++ b/drivers/infiniband/hw/mlx5/mad.c @@ -42,12 +42,24 @@ enum { MLX5_IB_VENDOR_CLASS2 = 0xa }; +static bool can_do_mad_ifc(struct mlx5_ib_dev *dev, u8 port_num, + struct ib_mad *in_mad) +{ + if (in_mad->mad_hdr.mgmt_class != IB_MGMT_CLASS_SUBN_LID_ROUTED && + in_mad->mad_hdr.mgmt_class != IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE) + return true; + return dev->mdev->port_caps[port_num - 1].has_smi; +} + int mlx5_MAD_IFC(struct mlx5_ib_dev *dev, int ignore_mkey, int ignore_bkey, u8 port, const struct ib_wc *in_wc, const struct ib_grh *in_grh, const void *in_mad, void *response_mad) { u8 op_modifier = 0; + if (!can_do_mad_ifc(dev, port, (struct ib_mad *)in_mad)) + return -EPERM; + /* Key check traps can't be generated unless we have in_wc to * tell us where to send the trap. */ @@ -515,7 +527,7 @@ int mlx5_query_mad_ifc_port(struct ib_device *ibdev, u8 port, if (!in_mad || !out_mad) goto out; - memset(props, 0, sizeof(*props)); + /* props being zeroed by the caller, avoid zeroing it here */ init_query_mad(in_mad); in_mad->attr_id = IB_SMP_ATTR_PORT_INFO; diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 32b09f059c84..5b3355268725 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -53,6 +53,7 @@ #include <linux/in.h> #include <linux/etherdevice.h> #include <linux/mlx5/fs.h> +#include <linux/mlx5/vport.h> #include "mlx5_ib.h" #define DRIVER_NAME "mlx5_ib" @@ -64,10 +65,6 @@ MODULE_DESCRIPTION("Mellanox Connect-IB HCA IB driver"); MODULE_LICENSE("Dual BSD/GPL"); MODULE_VERSION(DRIVER_VERSION); -static int deprecated_prof_sel = 2; -module_param_named(prof_sel, deprecated_prof_sel, int, 0444); -MODULE_PARM_DESC(prof_sel, "profile selector. Deprecated here. Moved to module mlx5_core"); - static char mlx5_version[] = DRIVER_NAME ": Mellanox Connect-IB Infiniband driver v" DRIVER_VERSION " (" DRIVER_RELDATE ")\n"; @@ -127,7 +124,7 @@ static int mlx5_netdev_event(struct notifier_block *this, if ((upper == ndev || (!upper && ndev == ibdev->roce.netdev)) && ibdev->ib_active) { - struct ib_event ibev = {0}; + struct ib_event ibev = { }; ibev.device = &ibdev->ib_dev; ibev.event = (event == NETDEV_UP) ? @@ -174,7 +171,7 @@ static int mlx5_query_port_roce(struct ib_device *device, u8 port_num, enum ib_mtu ndev_ib_mtu; u16 qkey_viol_cntr; - memset(props, 0, sizeof(*props)); + /* props being zeroed by the caller, avoid zeroing it here */ props->port_cap_flags |= IB_PORT_CM_SUP; props->port_cap_flags |= IB_PORT_IP_BASED_GIDS; @@ -325,6 +322,27 @@ __be16 mlx5_get_roce_udp_sport(struct mlx5_ib_dev *dev, u8 port_num, return cpu_to_be16(MLX5_CAP_ROCE(dev->mdev, r_roce_min_src_udp_port)); } +int mlx5_get_roce_gid_type(struct mlx5_ib_dev *dev, u8 port_num, + int index, enum ib_gid_type *gid_type) +{ + struct ib_gid_attr attr; + union ib_gid gid; + int ret; + + ret = ib_get_cached_gid(&dev->ib_dev, port_num, index, &gid, &attr); + if (ret) + return ret; + + if (!attr.ndev) + return -ENODEV; + + dev_put(attr.ndev); + + *gid_type = attr.gid_type; + + return 0; +} + static int mlx5_use_mad_ifc(struct mlx5_ib_dev *dev) { if (MLX5_CAP_GEN(dev->mdev, port_type) == MLX5_CAP_PORT_TYPE_IB) @@ -496,6 +514,7 @@ static int mlx5_ib_query_device(struct ib_device *ibdev, struct mlx5_ib_dev *dev = to_mdev(ibdev); struct mlx5_core_dev *mdev = dev->mdev; int err = -ENOMEM; + int max_sq_desc; int max_rq_sg; int max_sq_sg; u64 min_page_size = 1ull << MLX5_CAP_GEN(mdev, log_pg_sz); @@ -563,8 +582,15 @@ static int mlx5_ib_query_device(struct ib_device *ibdev, props->device_cap_flags |= IB_DEVICE_BLOCK_MULTICAST_LOOPBACK; if (MLX5_CAP_GEN(dev->mdev, eth_net_offloads)) { - if (MLX5_CAP_ETH(mdev, csum_cap)) + if (MLX5_CAP_ETH(mdev, csum_cap)) { + /* Legacy bit to support old userspace libraries */ props->device_cap_flags |= IB_DEVICE_RAW_IP_CSUM; + props->raw_packet_caps |= IB_RAW_PACKET_CAP_IP_CSUM; + } + + if (MLX5_CAP_ETH(dev->mdev, vlan_cap)) + props->raw_packet_caps |= + IB_RAW_PACKET_CAP_CVLAN_STRIPPING; if (field_avail(typeof(resp), tso_caps, uhw->outlen)) { max_tso = MLX5_CAP_ETH(mdev, max_lso_cap); @@ -603,8 +629,11 @@ static int mlx5_ib_query_device(struct ib_device *ibdev, } if (MLX5_CAP_GEN(dev->mdev, eth_net_offloads) && - MLX5_CAP_ETH(dev->mdev, scatter_fcs)) + MLX5_CAP_ETH(dev->mdev, scatter_fcs)) { + /* Legacy bit to support old userspace libraries */ props->device_cap_flags |= IB_DEVICE_RAW_SCATTER_FCS; + props->raw_packet_caps |= IB_RAW_PACKET_CAP_SCATTER_FCS; + } if (mlx5_get_flow_namespace(dev->mdev, MLX5_FLOW_NAMESPACE_BYPASS)) props->device_cap_flags |= IB_DEVICE_MANAGED_FLOW_STEERING; @@ -618,9 +647,10 @@ static int mlx5_ib_query_device(struct ib_device *ibdev, props->max_qp_wr = 1 << MLX5_CAP_GEN(mdev, log_max_qp_sz); max_rq_sg = MLX5_CAP_GEN(mdev, max_wqe_sz_rq) / sizeof(struct mlx5_wqe_data_seg); - max_sq_sg = (MLX5_CAP_GEN(mdev, max_wqe_sz_sq) - - sizeof(struct mlx5_wqe_ctrl_seg)) / - sizeof(struct mlx5_wqe_data_seg); + max_sq_desc = min_t(int, MLX5_CAP_GEN(mdev, max_wqe_sz_sq), 512); + max_sq_sg = (max_sq_desc - sizeof(struct mlx5_wqe_ctrl_seg) - + sizeof(struct mlx5_wqe_raddr_seg)) / + sizeof(struct mlx5_wqe_data_seg); props->max_sge = min(max_rq_sg, max_sq_sg); props->max_sge_rd = MLX5_MAX_SGE_RD; props->max_cq = 1 << MLX5_CAP_GEN(mdev, log_max_cq); @@ -643,6 +673,7 @@ static int mlx5_ib_query_device(struct ib_device *ibdev, props->max_total_mcast_qp_attach = props->max_mcast_qp_attach * props->max_mcast_grp; props->max_map_per_fmr = INT_MAX; /* no limit in ConnectIB */ + props->max_ah = INT_MAX; props->hca_core_clock = MLX5_CAP_GEN(mdev, device_frequency_khz); props->timestamp_mask = 0x7FFFFFFFFFFFFFFFULL; @@ -669,6 +700,40 @@ static int mlx5_ib_query_device(struct ib_device *ibdev, 1 << MLX5_CAP_GEN(dev->mdev, log_max_rq); } + if (field_avail(typeof(resp), cqe_comp_caps, uhw->outlen)) { + resp.cqe_comp_caps.max_num = + MLX5_CAP_GEN(dev->mdev, cqe_compression) ? + MLX5_CAP_GEN(dev->mdev, cqe_compression_max_num) : 0; + resp.cqe_comp_caps.supported_format = + MLX5_IB_CQE_RES_FORMAT_HASH | + MLX5_IB_CQE_RES_FORMAT_CSUM; + resp.response_length += sizeof(resp.cqe_comp_caps); + } + + if (field_avail(typeof(resp), packet_pacing_caps, uhw->outlen)) { + if (MLX5_CAP_QOS(mdev, packet_pacing) && + MLX5_CAP_GEN(mdev, qos)) { + resp.packet_pacing_caps.qp_rate_limit_max = + MLX5_CAP_QOS(mdev, packet_pacing_max_rate); + resp.packet_pacing_caps.qp_rate_limit_min = + MLX5_CAP_QOS(mdev, packet_pacing_min_rate); + resp.packet_pacing_caps.supported_qpts |= + 1 << IB_QPT_RAW_PACKET; + } + resp.response_length += sizeof(resp.packet_pacing_caps); + } + + if (field_avail(typeof(resp), mlx5_ib_support_multi_pkt_send_wqes, + uhw->outlen)) { + resp.mlx5_ib_support_multi_pkt_send_wqes = + MLX5_CAP_ETH(mdev, multi_pkt_send_wqe); + resp.response_length += + sizeof(resp.mlx5_ib_support_multi_pkt_send_wqes); + } + + if (field_avail(typeof(resp), reserved, uhw->outlen)) + resp.response_length += sizeof(resp.reserved); + if (uhw->outlen) { err = ib_copy_to_udata(uhw, &resp, resp.response_length); @@ -793,7 +858,7 @@ static int mlx5_query_hca_port(struct ib_device *ibdev, u8 port, goto out; } - memset(props, 0, sizeof(*props)); + /* props being zeroed by the caller, avoid zeroing it here */ err = mlx5_query_hca_vport_context(mdev, 0, port, 0, rep); if (err) @@ -931,6 +996,31 @@ static int mlx5_ib_modify_device(struct ib_device *ibdev, int mask, return err; } +static int set_port_caps_atomic(struct mlx5_ib_dev *dev, u8 port_num, u32 mask, + u32 value) +{ + struct mlx5_hca_vport_context ctx = {}; + int err; + + err = mlx5_query_hca_vport_context(dev->mdev, 0, + port_num, 0, &ctx); + if (err) + return err; + + if (~ctx.cap_mask1_perm & mask) { + mlx5_ib_warn(dev, "trying to change bitmask 0x%X but change supported 0x%X\n", + mask, ctx.cap_mask1_perm); + return -EINVAL; + } + + ctx.cap_mask1 = value; + ctx.cap_mask1_perm = mask; + err = mlx5_core_modify_hca_vport_context(dev->mdev, 0, + port_num, 0, &ctx); + + return err; +} + static int mlx5_ib_modify_port(struct ib_device *ibdev, u8 port, int mask, struct ib_port_modify *props) { @@ -938,10 +1028,20 @@ static int mlx5_ib_modify_port(struct ib_device *ibdev, u8 port, int mask, struct ib_port_attr attr; u32 tmp; int err; + u32 change_mask; + u32 value; + bool is_ib = (mlx5_ib_port_link_layer(ibdev, port) == + IB_LINK_LAYER_INFINIBAND); + + if (MLX5_CAP_GEN(dev->mdev, ib_virt) && is_ib) { + change_mask = props->clr_port_cap_mask | props->set_port_cap_mask; + value = ~props->clr_port_cap_mask | props->set_port_cap_mask; + return set_port_caps_atomic(dev, port, change_mask, value); + } mutex_lock(&dev->cap_mask_mutex); - err = mlx5_ib_query_port(ibdev, port, &attr); + err = ib_query_port(ibdev, port, &attr); if (err) goto out; @@ -955,6 +1055,86 @@ out: return err; } +static void print_lib_caps(struct mlx5_ib_dev *dev, u64 caps) +{ + mlx5_ib_dbg(dev, "MLX5_LIB_CAP_4K_UAR = %s\n", + caps & MLX5_LIB_CAP_4K_UAR ? "y" : "n"); +} + +static int calc_total_bfregs(struct mlx5_ib_dev *dev, bool lib_uar_4k, + struct mlx5_ib_alloc_ucontext_req_v2 *req, + u32 *num_sys_pages) +{ + int uars_per_sys_page; + int bfregs_per_sys_page; + int ref_bfregs = req->total_num_bfregs; + + if (req->total_num_bfregs == 0) + return -EINVAL; + + BUILD_BUG_ON(MLX5_MAX_BFREGS % MLX5_NON_FP_BFREGS_IN_PAGE); + BUILD_BUG_ON(MLX5_MAX_BFREGS < MLX5_NON_FP_BFREGS_IN_PAGE); + + if (req->total_num_bfregs > MLX5_MAX_BFREGS) + return -ENOMEM; + + uars_per_sys_page = get_uars_per_sys_page(dev, lib_uar_4k); + bfregs_per_sys_page = uars_per_sys_page * MLX5_NON_FP_BFREGS_PER_UAR; + req->total_num_bfregs = ALIGN(req->total_num_bfregs, bfregs_per_sys_page); + *num_sys_pages = req->total_num_bfregs / bfregs_per_sys_page; + + if (req->num_low_latency_bfregs > req->total_num_bfregs - 1) + return -EINVAL; + + mlx5_ib_dbg(dev, "uar_4k: fw support %s, lib support %s, user requested %d bfregs, alloated %d, using %d sys pages\n", + MLX5_CAP_GEN(dev->mdev, uar_4k) ? "yes" : "no", + lib_uar_4k ? "yes" : "no", ref_bfregs, + req->total_num_bfregs, *num_sys_pages); + + return 0; +} + +static int allocate_uars(struct mlx5_ib_dev *dev, struct mlx5_ib_ucontext *context) +{ + struct mlx5_bfreg_info *bfregi; + int err; + int i; + + bfregi = &context->bfregi; + for (i = 0; i < bfregi->num_sys_pages; i++) { + err = mlx5_cmd_alloc_uar(dev->mdev, &bfregi->sys_pages[i]); + if (err) + goto error; + + mlx5_ib_dbg(dev, "allocated uar %d\n", bfregi->sys_pages[i]); + } + return 0; + +error: + for (--i; i >= 0; i--) + if (mlx5_cmd_free_uar(dev->mdev, bfregi->sys_pages[i])) + mlx5_ib_warn(dev, "failed to free uar %d\n", i); + + return err; +} + +static int deallocate_uars(struct mlx5_ib_dev *dev, struct mlx5_ib_ucontext *context) +{ + struct mlx5_bfreg_info *bfregi; + int err; + int i; + + bfregi = &context->bfregi; + for (i = 0; i < bfregi->num_sys_pages; i++) { + err = mlx5_cmd_free_uar(dev->mdev, bfregi->sys_pages[i]); + if (err) { + mlx5_ib_warn(dev, "failed to free uar %d\n", i); + return err; + } + } + return 0; +} + static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev, struct ib_udata *udata) { @@ -962,17 +1142,13 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev, struct mlx5_ib_alloc_ucontext_req_v2 req = {}; struct mlx5_ib_alloc_ucontext_resp resp = {}; struct mlx5_ib_ucontext *context; - struct mlx5_uuar_info *uuari; - struct mlx5_uar *uars; - int gross_uuars; - int num_uars; + struct mlx5_bfreg_info *bfregi; int ver; - int uuarn; int err; - int i; size_t reqlen; size_t min_req_v2 = offsetof(struct mlx5_ib_alloc_ucontext_req_v2, max_cqe_version); + bool lib_uar_4k; if (!dev->ib_active) return ERR_PTR(-EAGAIN); @@ -995,27 +1171,14 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev, if (req.flags) return ERR_PTR(-EINVAL); - if (req.total_num_uuars > MLX5_MAX_UUARS) - return ERR_PTR(-ENOMEM); - - if (req.total_num_uuars == 0) - return ERR_PTR(-EINVAL); - if (req.comp_mask || req.reserved0 || req.reserved1 || req.reserved2) return ERR_PTR(-EOPNOTSUPP); - if (reqlen > sizeof(req) && - !ib_is_udata_cleared(udata, sizeof(req), - reqlen - sizeof(req))) - return ERR_PTR(-EOPNOTSUPP); - - req.total_num_uuars = ALIGN(req.total_num_uuars, - MLX5_NON_FP_BF_REGS_PER_PAGE); - if (req.num_low_latency_uuars > req.total_num_uuars - 1) + req.total_num_bfregs = ALIGN(req.total_num_bfregs, + MLX5_NON_FP_BFREGS_PER_UAR); + if (req.num_low_latency_bfregs > req.total_num_bfregs - 1) return ERR_PTR(-EINVAL); - num_uars = req.total_num_uuars / MLX5_NON_FP_BF_REGS_PER_PAGE; - gross_uuars = num_uars * MLX5_BF_REGS_PER_PAGE; resp.qp_tab_size = 1 << MLX5_CAP_GEN(dev->mdev, log_max_qp); if (mlx5_core_is_pf(dev->mdev) && MLX5_CAP_GEN(dev->mdev, bf)) resp.bf_reg_size = 1 << MLX5_CAP_GEN(dev->mdev, log_bf_reg_size); @@ -1028,6 +1191,10 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev, resp.cqe_version = min_t(__u8, (__u8)MLX5_CAP_GEN(dev->mdev, cqe_version), req.max_cqe_version); + resp.log_uar_size = MLX5_CAP_GEN(dev->mdev, uar_4k) ? + MLX5_ADAPTER_PAGE_SHIFT : PAGE_SHIFT; + resp.num_uars_per_page = MLX5_CAP_GEN(dev->mdev, uar_4k) ? + MLX5_CAP_GEN(dev->mdev, num_of_uars_per_page) : 1; resp.response_length = min(offsetof(typeof(resp), response_length) + sizeof(resp.response_length), udata->outlen); @@ -1035,94 +1202,109 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev, if (!context) return ERR_PTR(-ENOMEM); - uuari = &context->uuari; - mutex_init(&uuari->lock); - uars = kcalloc(num_uars, sizeof(*uars), GFP_KERNEL); - if (!uars) { - err = -ENOMEM; + lib_uar_4k = req.lib_caps & MLX5_LIB_CAP_4K_UAR; + bfregi = &context->bfregi; + + /* updates req->total_num_bfregs */ + err = calc_total_bfregs(dev, lib_uar_4k, &req, &bfregi->num_sys_pages); + if (err) goto out_ctx; - } - uuari->bitmap = kcalloc(BITS_TO_LONGS(gross_uuars), - sizeof(*uuari->bitmap), + mutex_init(&bfregi->lock); + bfregi->lib_uar_4k = lib_uar_4k; + bfregi->count = kcalloc(req.total_num_bfregs, sizeof(*bfregi->count), GFP_KERNEL); - if (!uuari->bitmap) { + if (!bfregi->count) { err = -ENOMEM; - goto out_uar_ctx; - } - /* - * clear all fast path uuars - */ - for (i = 0; i < gross_uuars; i++) { - uuarn = i & 3; - if (uuarn == 2 || uuarn == 3) - set_bit(i, uuari->bitmap); + goto out_ctx; } - uuari->count = kcalloc(gross_uuars, sizeof(*uuari->count), GFP_KERNEL); - if (!uuari->count) { + bfregi->sys_pages = kcalloc(bfregi->num_sys_pages, + sizeof(*bfregi->sys_pages), + GFP_KERNEL); + if (!bfregi->sys_pages) { err = -ENOMEM; - goto out_bitmap; + goto out_count; } - for (i = 0; i < num_uars; i++) { - err = mlx5_cmd_alloc_uar(dev->mdev, &uars[i].index); - if (err) - goto out_count; - } + err = allocate_uars(dev, context); + if (err) + goto out_sys_pages; #ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING context->ibucontext.invalidate_range = &mlx5_ib_invalidate_range; #endif + context->upd_xlt_page = __get_free_page(GFP_KERNEL); + if (!context->upd_xlt_page) { + err = -ENOMEM; + goto out_uars; + } + mutex_init(&context->upd_xlt_page_mutex); + if (MLX5_CAP_GEN(dev->mdev, log_max_transport_domain)) { err = mlx5_core_alloc_transport_domain(dev->mdev, &context->tdn); if (err) - goto out_uars; + goto out_page; } INIT_LIST_HEAD(&context->vma_private_list); INIT_LIST_HEAD(&context->db_page_list); mutex_init(&context->db_page_mutex); - resp.tot_uuars = req.total_num_uuars; + resp.tot_bfregs = req.total_num_bfregs; resp.num_ports = MLX5_CAP_GEN(dev->mdev, num_ports); if (field_avail(typeof(resp), cqe_version, udata->outlen)) resp.response_length += sizeof(resp.cqe_version); if (field_avail(typeof(resp), cmds_supp_uhw, udata->outlen)) { - resp.cmds_supp_uhw |= MLX5_USER_CMDS_SUPP_UHW_QUERY_DEVICE; + resp.cmds_supp_uhw |= MLX5_USER_CMDS_SUPP_UHW_QUERY_DEVICE | + MLX5_USER_CMDS_SUPP_UHW_CREATE_AH; resp.response_length += sizeof(resp.cmds_supp_uhw); } + if (field_avail(typeof(resp), eth_min_inline, udata->outlen)) { + if (mlx5_ib_port_link_layer(ibdev, 1) == IB_LINK_LAYER_ETHERNET) { + mlx5_query_min_inline(dev->mdev, &resp.eth_min_inline); + resp.eth_min_inline++; + } + resp.response_length += sizeof(resp.eth_min_inline); + } + /* * We don't want to expose information from the PCI bar that is located * after 4096 bytes, so if the arch only supports larger pages, let's * pretend we don't support reading the HCA's core clock. This is also * forced by mmap function. */ - if (PAGE_SIZE <= 4096 && - field_avail(typeof(resp), hca_core_clock_offset, udata->outlen)) { - resp.comp_mask |= - MLX5_IB_ALLOC_UCONTEXT_RESP_MASK_CORE_CLOCK_OFFSET; - resp.hca_core_clock_offset = - offsetof(struct mlx5_init_seg, internal_timer_h) % - PAGE_SIZE; + if (field_avail(typeof(resp), hca_core_clock_offset, udata->outlen)) { + if (PAGE_SIZE <= 4096) { + resp.comp_mask |= + MLX5_IB_ALLOC_UCONTEXT_RESP_MASK_CORE_CLOCK_OFFSET; + resp.hca_core_clock_offset = + offsetof(struct mlx5_init_seg, internal_timer_h) % PAGE_SIZE; + } resp.response_length += sizeof(resp.hca_core_clock_offset) + sizeof(resp.reserved2); } + if (field_avail(typeof(resp), log_uar_size, udata->outlen)) + resp.response_length += sizeof(resp.log_uar_size); + + if (field_avail(typeof(resp), num_uars_per_page, udata->outlen)) + resp.response_length += sizeof(resp.num_uars_per_page); + err = ib_copy_to_udata(udata, &resp, resp.response_length); if (err) goto out_td; - uuari->ver = ver; - uuari->num_low_latency_uuars = req.num_low_latency_uuars; - uuari->uars = uars; - uuari->num_uars = num_uars; + bfregi->ver = ver; + bfregi->num_low_latency_bfregs = req.num_low_latency_bfregs; context->cqe_version = resp.cqe_version; + context->lib_caps = req.lib_caps; + print_lib_caps(dev, context->lib_caps); return &context->ibucontext; @@ -1130,20 +1312,21 @@ out_td: if (MLX5_CAP_GEN(dev->mdev, log_max_transport_domain)) mlx5_core_dealloc_transport_domain(dev->mdev, context->tdn); +out_page: + free_page(context->upd_xlt_page); + out_uars: - for (i--; i >= 0; i--) - mlx5_cmd_free_uar(dev->mdev, uars[i].index); -out_count: - kfree(uuari->count); + deallocate_uars(dev, context); -out_bitmap: - kfree(uuari->bitmap); +out_sys_pages: + kfree(bfregi->sys_pages); -out_uar_ctx: - kfree(uars); +out_count: + kfree(bfregi->count); out_ctx: kfree(context); + return ERR_PTR(err); } @@ -1151,28 +1334,31 @@ static int mlx5_ib_dealloc_ucontext(struct ib_ucontext *ibcontext) { struct mlx5_ib_ucontext *context = to_mucontext(ibcontext); struct mlx5_ib_dev *dev = to_mdev(ibcontext->device); - struct mlx5_uuar_info *uuari = &context->uuari; - int i; + struct mlx5_bfreg_info *bfregi; + bfregi = &context->bfregi; if (MLX5_CAP_GEN(dev->mdev, log_max_transport_domain)) mlx5_core_dealloc_transport_domain(dev->mdev, context->tdn); - for (i = 0; i < uuari->num_uars; i++) { - if (mlx5_cmd_free_uar(dev->mdev, uuari->uars[i].index)) - mlx5_ib_warn(dev, "failed to free UAR 0x%x\n", uuari->uars[i].index); - } - - kfree(uuari->count); - kfree(uuari->bitmap); - kfree(uuari->uars); + free_page(context->upd_xlt_page); + deallocate_uars(dev, context); + kfree(bfregi->sys_pages); + kfree(bfregi->count); kfree(context); return 0; } -static phys_addr_t uar_index2pfn(struct mlx5_ib_dev *dev, int index) +static phys_addr_t uar_index2pfn(struct mlx5_ib_dev *dev, + struct mlx5_bfreg_info *bfregi, + int idx) { - return (pci_resource_start(dev->mdev->pdev, 0) >> PAGE_SHIFT) + index; + int fw_uars_per_page; + + fw_uars_per_page = MLX5_CAP_GEN(dev->mdev, uar_4k) ? MLX5_UARS_IN_PAGE : 1; + + return (pci_resource_start(dev->mdev->pdev, 0) >> PAGE_SHIFT) + + bfregi->sys_pages[idx] / fw_uars_per_page; } static int get_command(unsigned long offset) @@ -1327,11 +1513,23 @@ static int uar_mmap(struct mlx5_ib_dev *dev, enum mlx5_ib_mmap_cmd cmd, struct vm_area_struct *vma, struct mlx5_ib_ucontext *context) { - struct mlx5_uuar_info *uuari = &context->uuari; + struct mlx5_bfreg_info *bfregi = &context->bfregi; int err; unsigned long idx; phys_addr_t pfn, pa; pgprot_t prot; + int uars_per_page; + + if (vma->vm_end - vma->vm_start != PAGE_SIZE) + return -EINVAL; + + uars_per_page = get_uars_per_sys_page(dev, bfregi->lib_uar_4k); + idx = get_index(vma->vm_pgoff); + if (idx % uars_per_page || + idx * uars_per_page >= bfregi->num_sys_pages) { + mlx5_ib_warn(dev, "invalid uar index %lu\n", idx); + return -EINVAL; + } switch (cmd) { case MLX5_IB_MMAP_WC_PAGE: @@ -1354,14 +1552,7 @@ static int uar_mmap(struct mlx5_ib_dev *dev, enum mlx5_ib_mmap_cmd cmd, return -EINVAL; } - if (vma->vm_end - vma->vm_start != PAGE_SIZE) - return -EINVAL; - - idx = get_index(vma->vm_pgoff); - if (idx >= uuari->num_uars) - return -EINVAL; - - pfn = uar_index2pfn(dev, uuari->uars[idx].index); + pfn = uar_index2pfn(dev, bfregi, idx); mlx5_ib_dbg(dev, "uar idx 0x%lx, pfn %pa\n", idx, &pfn); vma->vm_page_prot = prot; @@ -1502,6 +1693,22 @@ static void set_proto(void *outer_c, void *outer_v, u8 mask, u8 val) MLX5_SET(fte_match_set_lyr_2_4, outer_v, ip_protocol, val); } +static void set_flow_label(void *misc_c, void *misc_v, u8 mask, u8 val, + bool inner) +{ + if (inner) { + MLX5_SET(fte_match_set_misc, + misc_c, inner_ipv6_flow_label, mask); + MLX5_SET(fte_match_set_misc, + misc_v, inner_ipv6_flow_label, val); + } else { + MLX5_SET(fte_match_set_misc, + misc_c, outer_ipv6_flow_label, mask); + MLX5_SET(fte_match_set_misc, + misc_v, outer_ipv6_flow_label, val); + } +} + static void set_tos(void *outer_c, void *outer_v, u8 mask, u8 val) { MLX5_SET(fte_match_set_lyr_2_4, outer_c, ip_ecn, mask); @@ -1515,6 +1722,8 @@ static void set_tos(void *outer_c, void *outer_v, u8 mask, u8 val) #define LAST_IPV4_FIELD tos #define LAST_IPV6_FIELD traffic_class #define LAST_TCP_UDP_FIELD src_port +#define LAST_TUNNEL_FIELD tunnel_id +#define LAST_FLOW_TAG_FIELD tag_id /* Field is the last supported field */ #define FIELDS_NOT_SUPPORTED(filter, field)\ @@ -1525,179 +1734,207 @@ static void set_tos(void *outer_c, void *outer_v, u8 mask, u8 val) sizeof(filter.field)) static int parse_flow_attr(u32 *match_c, u32 *match_v, - const union ib_flow_spec *ib_spec) + const union ib_flow_spec *ib_spec, u32 *tag_id) { - void *outer_headers_c = MLX5_ADDR_OF(fte_match_param, match_c, - outer_headers); - void *outer_headers_v = MLX5_ADDR_OF(fte_match_param, match_v, - outer_headers); void *misc_params_c = MLX5_ADDR_OF(fte_match_param, match_c, misc_parameters); void *misc_params_v = MLX5_ADDR_OF(fte_match_param, match_v, misc_parameters); + void *headers_c; + void *headers_v; + + if (ib_spec->type & IB_FLOW_SPEC_INNER) { + headers_c = MLX5_ADDR_OF(fte_match_param, match_c, + inner_headers); + headers_v = MLX5_ADDR_OF(fte_match_param, match_v, + inner_headers); + } else { + headers_c = MLX5_ADDR_OF(fte_match_param, match_c, + outer_headers); + headers_v = MLX5_ADDR_OF(fte_match_param, match_v, + outer_headers); + } - switch (ib_spec->type) { + switch (ib_spec->type & ~IB_FLOW_SPEC_INNER) { case IB_FLOW_SPEC_ETH: if (FIELDS_NOT_SUPPORTED(ib_spec->eth.mask, LAST_ETH_FIELD)) - return -ENOTSUPP; + return -EOPNOTSUPP; - ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_c, + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, dmac_47_16), ib_spec->eth.mask.dst_mac); - ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_v, + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, dmac_47_16), ib_spec->eth.val.dst_mac); - ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_c, + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, smac_47_16), ib_spec->eth.mask.src_mac); - ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_v, + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, smac_47_16), ib_spec->eth.val.src_mac); if (ib_spec->eth.mask.vlan_tag) { - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, - vlan_tag, 1); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, - vlan_tag, 1); + MLX5_SET(fte_match_set_lyr_2_4, headers_c, + cvlan_tag, 1); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, + cvlan_tag, 1); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + MLX5_SET(fte_match_set_lyr_2_4, headers_c, first_vid, ntohs(ib_spec->eth.mask.vlan_tag)); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + MLX5_SET(fte_match_set_lyr_2_4, headers_v, first_vid, ntohs(ib_spec->eth.val.vlan_tag)); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + MLX5_SET(fte_match_set_lyr_2_4, headers_c, first_cfi, ntohs(ib_spec->eth.mask.vlan_tag) >> 12); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + MLX5_SET(fte_match_set_lyr_2_4, headers_v, first_cfi, ntohs(ib_spec->eth.val.vlan_tag) >> 12); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + MLX5_SET(fte_match_set_lyr_2_4, headers_c, first_prio, ntohs(ib_spec->eth.mask.vlan_tag) >> 13); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + MLX5_SET(fte_match_set_lyr_2_4, headers_v, first_prio, ntohs(ib_spec->eth.val.vlan_tag) >> 13); } - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + MLX5_SET(fte_match_set_lyr_2_4, headers_c, ethertype, ntohs(ib_spec->eth.mask.ether_type)); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype, ntohs(ib_spec->eth.val.ether_type)); break; case IB_FLOW_SPEC_IPV4: if (FIELDS_NOT_SUPPORTED(ib_spec->ipv4.mask, LAST_IPV4_FIELD)) - return -ENOTSUPP; + return -EOPNOTSUPP; - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + MLX5_SET(fte_match_set_lyr_2_4, headers_c, ethertype, 0xffff); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype, ETH_P_IP); - memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_c, + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, src_ipv4_src_ipv6.ipv4_layout.ipv4), &ib_spec->ipv4.mask.src_ip, sizeof(ib_spec->ipv4.mask.src_ip)); - memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_v, + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, src_ipv4_src_ipv6.ipv4_layout.ipv4), &ib_spec->ipv4.val.src_ip, sizeof(ib_spec->ipv4.val.src_ip)); - memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_c, + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, dst_ipv4_dst_ipv6.ipv4_layout.ipv4), &ib_spec->ipv4.mask.dst_ip, sizeof(ib_spec->ipv4.mask.dst_ip)); - memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_v, + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, dst_ipv4_dst_ipv6.ipv4_layout.ipv4), &ib_spec->ipv4.val.dst_ip, sizeof(ib_spec->ipv4.val.dst_ip)); - set_tos(outer_headers_c, outer_headers_v, + set_tos(headers_c, headers_v, ib_spec->ipv4.mask.tos, ib_spec->ipv4.val.tos); - set_proto(outer_headers_c, outer_headers_v, + set_proto(headers_c, headers_v, ib_spec->ipv4.mask.proto, ib_spec->ipv4.val.proto); break; case IB_FLOW_SPEC_IPV6: if (FIELDS_NOT_SUPPORTED(ib_spec->ipv6.mask, LAST_IPV6_FIELD)) - return -ENOTSUPP; + return -EOPNOTSUPP; - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + MLX5_SET(fte_match_set_lyr_2_4, headers_c, ethertype, 0xffff); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype, ETH_P_IPV6); - memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_c, + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, src_ipv4_src_ipv6.ipv6_layout.ipv6), &ib_spec->ipv6.mask.src_ip, sizeof(ib_spec->ipv6.mask.src_ip)); - memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_v, + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, src_ipv4_src_ipv6.ipv6_layout.ipv6), &ib_spec->ipv6.val.src_ip, sizeof(ib_spec->ipv6.val.src_ip)); - memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_c, + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, dst_ipv4_dst_ipv6.ipv6_layout.ipv6), &ib_spec->ipv6.mask.dst_ip, sizeof(ib_spec->ipv6.mask.dst_ip)); - memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_v, + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v, dst_ipv4_dst_ipv6.ipv6_layout.ipv6), &ib_spec->ipv6.val.dst_ip, sizeof(ib_spec->ipv6.val.dst_ip)); - set_tos(outer_headers_c, outer_headers_v, + set_tos(headers_c, headers_v, ib_spec->ipv6.mask.traffic_class, ib_spec->ipv6.val.traffic_class); - set_proto(outer_headers_c, outer_headers_v, + set_proto(headers_c, headers_v, ib_spec->ipv6.mask.next_hdr, ib_spec->ipv6.val.next_hdr); - MLX5_SET(fte_match_set_misc, misc_params_c, - outer_ipv6_flow_label, - ntohl(ib_spec->ipv6.mask.flow_label)); - MLX5_SET(fte_match_set_misc, misc_params_v, - outer_ipv6_flow_label, - ntohl(ib_spec->ipv6.val.flow_label)); + set_flow_label(misc_params_c, misc_params_v, + ntohl(ib_spec->ipv6.mask.flow_label), + ntohl(ib_spec->ipv6.val.flow_label), + ib_spec->type & IB_FLOW_SPEC_INNER); + break; case IB_FLOW_SPEC_TCP: if (FIELDS_NOT_SUPPORTED(ib_spec->tcp_udp.mask, LAST_TCP_UDP_FIELD)) - return -ENOTSUPP; + return -EOPNOTSUPP; - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, ip_protocol, + MLX5_SET(fte_match_set_lyr_2_4, headers_c, ip_protocol, 0xff); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, ip_protocol, + MLX5_SET(fte_match_set_lyr_2_4, headers_v, ip_protocol, IPPROTO_TCP); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, tcp_sport, + MLX5_SET(fte_match_set_lyr_2_4, headers_c, tcp_sport, ntohs(ib_spec->tcp_udp.mask.src_port)); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, tcp_sport, + MLX5_SET(fte_match_set_lyr_2_4, headers_v, tcp_sport, ntohs(ib_spec->tcp_udp.val.src_port)); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, tcp_dport, + MLX5_SET(fte_match_set_lyr_2_4, headers_c, tcp_dport, ntohs(ib_spec->tcp_udp.mask.dst_port)); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, tcp_dport, + MLX5_SET(fte_match_set_lyr_2_4, headers_v, tcp_dport, ntohs(ib_spec->tcp_udp.val.dst_port)); break; case IB_FLOW_SPEC_UDP: if (FIELDS_NOT_SUPPORTED(ib_spec->tcp_udp.mask, LAST_TCP_UDP_FIELD)) - return -ENOTSUPP; + return -EOPNOTSUPP; - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, ip_protocol, + MLX5_SET(fte_match_set_lyr_2_4, headers_c, ip_protocol, 0xff); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, ip_protocol, + MLX5_SET(fte_match_set_lyr_2_4, headers_v, ip_protocol, IPPROTO_UDP); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, udp_sport, + MLX5_SET(fte_match_set_lyr_2_4, headers_c, udp_sport, ntohs(ib_spec->tcp_udp.mask.src_port)); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, udp_sport, + MLX5_SET(fte_match_set_lyr_2_4, headers_v, udp_sport, ntohs(ib_spec->tcp_udp.val.src_port)); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, udp_dport, + MLX5_SET(fte_match_set_lyr_2_4, headers_c, udp_dport, ntohs(ib_spec->tcp_udp.mask.dst_port)); - MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, udp_dport, + MLX5_SET(fte_match_set_lyr_2_4, headers_v, udp_dport, ntohs(ib_spec->tcp_udp.val.dst_port)); break; + case IB_FLOW_SPEC_VXLAN_TUNNEL: + if (FIELDS_NOT_SUPPORTED(ib_spec->tunnel.mask, + LAST_TUNNEL_FIELD)) + return -EOPNOTSUPP; + + MLX5_SET(fte_match_set_misc, misc_params_c, vxlan_vni, + ntohl(ib_spec->tunnel.mask.tunnel_id)); + MLX5_SET(fte_match_set_misc, misc_params_v, vxlan_vni, + ntohl(ib_spec->tunnel.val.tunnel_id)); + break; + case IB_FLOW_SPEC_ACTION_TAG: + if (FIELDS_NOT_SUPPORTED(ib_spec->flow_tag, + LAST_FLOW_TAG_FIELD)) + return -EOPNOTSUPP; + if (ib_spec->flow_tag.tag_id >= BIT(24)) + return -EINVAL; + + *tag_id = ib_spec->flow_tag.tag_id; + break; default: return -EINVAL; } @@ -1771,13 +2008,13 @@ static int mlx5_ib_destroy_flow(struct ib_flow *flow_id) mutex_lock(&dev->flow_db.lock); list_for_each_entry_safe(iter, tmp, &handler->list, list) { - mlx5_del_flow_rule(iter->rule); + mlx5_del_flow_rules(iter->rule); put_flow_table(dev, iter->prio, true); list_del(&iter->list); kfree(iter); } - mlx5_del_flow_rule(handler->rule); + mlx5_del_flow_rules(handler->rule); put_flow_table(dev, handler->prio, true); mutex_unlock(&dev->flow_db.lock); @@ -1857,7 +2094,7 @@ static struct mlx5_ib_flow_prio *get_flow_table(struct mlx5_ib_dev *dev, ft = mlx5_create_auto_grouped_flow_table(ns, priority, num_entries, num_groups, - 0); + 0, 0); if (!IS_ERR(ft)) { prio->refcount = 0; @@ -1877,10 +2114,11 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, { struct mlx5_flow_table *ft = ft_prio->flow_table; struct mlx5_ib_flow_handler *handler; + struct mlx5_flow_act flow_act = {0}; struct mlx5_flow_spec *spec; const void *ib_flow = (const void *)flow_attr + sizeof(*flow_attr); unsigned int spec_index; - u32 action; + u32 flow_tag = MLX5_FS_DEFAULT_FLOW_TAG; int err = 0; if (!is_valid_attr(flow_attr)) @@ -1897,7 +2135,7 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, for (spec_index = 0; spec_index < flow_attr->num_of_specs; spec_index++) { err = parse_flow_attr(spec->match_criteria, - spec->match_value, ib_flow); + spec->match_value, ib_flow, &flow_tag); if (err < 0) goto free; @@ -1905,12 +2143,21 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, } spec->match_criteria_enable = get_match_criteria_enable(spec->match_criteria); - action = dst ? MLX5_FLOW_CONTEXT_ACTION_FWD_DEST : + flow_act.action = dst ? MLX5_FLOW_CONTEXT_ACTION_FWD_DEST : MLX5_FLOW_CONTEXT_ACTION_FWD_NEXT_PRIO; - handler->rule = mlx5_add_flow_rule(ft, spec, - action, - MLX5_FS_DEFAULT_FLOW_TAG, - dst); + + if (flow_tag != MLX5_FS_DEFAULT_FLOW_TAG && + (flow_attr->type == IB_FLOW_ATTR_ALL_DEFAULT || + flow_attr->type == IB_FLOW_ATTR_MC_DEFAULT)) { + mlx5_ib_warn(dev, "Flow tag %u and attribute type %x isn't allowed in leftovers\n", + flow_tag, flow_attr->type); + err = -EINVAL; + goto free; + } + flow_act.flow_tag = flow_tag; + handler->rule = mlx5_add_flow_rules(ft, spec, + &flow_act, + dst, 1); if (IS_ERR(handler->rule)) { err = PTR_ERR(handler->rule); @@ -1941,7 +2188,7 @@ static struct mlx5_ib_flow_handler *create_dont_trap_rule(struct mlx5_ib_dev *de handler_dst = create_flow_rule(dev, ft_prio, flow_attr, dst); if (IS_ERR(handler_dst)) { - mlx5_del_flow_rule(handler->rule); + mlx5_del_flow_rules(handler->rule); ft_prio->refcount--; kfree(handler); handler = handler_dst; @@ -2004,7 +2251,7 @@ static struct mlx5_ib_flow_handler *create_leftovers_rule(struct mlx5_ib_dev *de &leftovers_specs[LEFTOVERS_UC].flow_attr, dst); if (IS_ERR(handler_ucast)) { - mlx5_del_flow_rule(handler->rule); + mlx5_del_flow_rules(handler->rule); ft_prio->refcount--; kfree(handler); handler = handler_ucast; @@ -2046,7 +2293,7 @@ static struct mlx5_ib_flow_handler *create_sniffer_rule(struct mlx5_ib_dev *dev, return handler_rx; err_tx: - mlx5_del_flow_rule(handler_rx->rule); + mlx5_del_flow_rules(handler_rx->rule); ft_rx->refcount--; kfree(handler_rx); err: @@ -2358,6 +2605,8 @@ static void mlx5_ib_event(struct mlx5_core_dev *dev, void *context, ibev.event = IB_EVENT_CLIENT_REREGISTER; port = (u8)param; break; + default: + return; } ibev.device = &ibdev->ib_dev; @@ -2375,6 +2624,35 @@ static void mlx5_ib_event(struct mlx5_core_dev *dev, void *context, ibdev->ib_active = false; } +static int set_has_smi_cap(struct mlx5_ib_dev *dev) +{ + struct mlx5_hca_vport_context vport_ctx; + int err; + int port; + + for (port = 1; port <= MLX5_CAP_GEN(dev->mdev, num_ports); port++) { + dev->mdev->port_caps[port - 1].has_smi = false; + if (MLX5_CAP_GEN(dev->mdev, port_type) == + MLX5_CAP_PORT_TYPE_IB) { + if (MLX5_CAP_GEN(dev->mdev, ib_virt)) { + err = mlx5_query_hca_vport_context(dev->mdev, 0, + port, 0, + &vport_ctx); + if (err) { + mlx5_ib_err(dev, "query_hca_vport_context for port=%d failed %d\n", + port, err); + return err; + } + dev->mdev->port_caps[port - 1].has_smi = + vport_ctx.has_smi; + } else { + dev->mdev->port_caps[port - 1].has_smi = true; + } + } + } + return 0; +} + static void get_ext_port_caps(struct mlx5_ib_dev *dev) { int port; @@ -2399,6 +2677,10 @@ static int get_port_caps(struct mlx5_ib_dev *dev) if (!dprops) goto out; + err = set_has_smi_cap(dev); + if (err) + goto out; + err = mlx5_ib_query_device(&dev->ib_dev, dprops, &uhw); if (err) { mlx5_ib_warn(dev, "query_device failed %d\n", err); @@ -2406,6 +2688,7 @@ static int get_port_caps(struct mlx5_ib_dev *dev) } for (port = 1; port <= MLX5_CAP_GEN(dev->mdev, num_ports); port++) { + memset(pprops, 0, sizeof(*pprops)); err = mlx5_ib_query_port(&dev->ib_dev, port, pprops); if (err) { mlx5_ib_warn(dev, "query_port %d failed %d\n", @@ -2700,11 +2983,13 @@ static u32 get_core_cap_flags(struct ib_device *ibdev) if (ll == IB_LINK_LAYER_INFINIBAND) return RDMA_CORE_PORT_IBA_IB; + ret = RDMA_CORE_PORT_RAW_PACKET; + if (!(l3_type_cap & MLX5_ROCE_L3_TYPE_IPV4_CAP)) - return 0; + return ret; if (!(l3_type_cap & MLX5_ROCE_L3_TYPE_IPV6_CAP)) - return 0; + return ret; if (roce_version_cap & MLX5_ROCE_VERSION_1_CAP) ret |= RDMA_CORE_PORT_IBA_ROCE; @@ -2719,16 +3004,21 @@ static int mlx5_port_immutable(struct ib_device *ibdev, u8 port_num, struct ib_port_immutable *immutable) { struct ib_port_attr attr; + struct mlx5_ib_dev *dev = to_mdev(ibdev); + enum rdma_link_layer ll = mlx5_ib_port_link_layer(ibdev, port_num); int err; - err = mlx5_ib_query_port(ibdev, port_num, &attr); + immutable->core_cap_flags = get_core_cap_flags(ibdev); + + err = ib_query_port(ibdev, port_num, &attr); if (err) return err; immutable->pkey_tbl_len = attr.pkey_tbl_len; immutable->gid_tbl_len = attr.gid_tbl_len; immutable->core_cap_flags = get_core_cap_flags(ibdev); - immutable->max_mad_size = IB_MGMT_MAD_SIZE; + if ((ll == IB_LINK_LAYER_INFINIBAND) || MLX5_CAP_GEN(dev->mdev, roce)) + immutable->max_mad_size = IB_MGMT_MAD_SIZE; return 0; } @@ -2742,7 +3032,7 @@ static void get_dev_fw_str(struct ib_device *ibdev, char *str, fw_rev_min(dev->mdev), fw_rev_sub(dev->mdev)); } -static int mlx5_roce_lag_init(struct mlx5_ib_dev *dev) +static int mlx5_eth_lag_init(struct mlx5_ib_dev *dev) { struct mlx5_core_dev *mdev = dev->mdev; struct mlx5_flow_namespace *ns = mlx5_get_flow_namespace(mdev, @@ -2771,7 +3061,7 @@ err_destroy_vport_lag: return err; } -static void mlx5_roce_lag_cleanup(struct mlx5_ib_dev *dev) +static void mlx5_eth_lag_cleanup(struct mlx5_ib_dev *dev) { struct mlx5_core_dev *mdev = dev->mdev; @@ -2783,7 +3073,21 @@ static void mlx5_roce_lag_cleanup(struct mlx5_ib_dev *dev) } } -static void mlx5_remove_roce_notifier(struct mlx5_ib_dev *dev) +static int mlx5_add_netdev_notifier(struct mlx5_ib_dev *dev) +{ + int err; + + dev->roce.nb.notifier_call = mlx5_netdev_event; + err = register_netdevice_notifier(&dev->roce.nb); + if (err) { + dev->roce.nb.notifier_call = NULL; + return err; + } + + return 0; +} + +static void mlx5_remove_netdev_notifier(struct mlx5_ib_dev *dev) { if (dev->roce.nb.notifier_call) { unregister_netdevice_notifier(&dev->roce.nb); @@ -2791,48 +3095,138 @@ static void mlx5_remove_roce_notifier(struct mlx5_ib_dev *dev) } } -static int mlx5_enable_roce(struct mlx5_ib_dev *dev) +static int mlx5_enable_eth(struct mlx5_ib_dev *dev) { int err; - dev->roce.nb.notifier_call = mlx5_netdev_event; - err = register_netdevice_notifier(&dev->roce.nb); - if (err) { - dev->roce.nb.notifier_call = NULL; + err = mlx5_add_netdev_notifier(dev); + if (err) return err; - } - err = mlx5_nic_vport_enable_roce(dev->mdev); - if (err) - goto err_unregister_netdevice_notifier; + if (MLX5_CAP_GEN(dev->mdev, roce)) { + err = mlx5_nic_vport_enable_roce(dev->mdev); + if (err) + goto err_unregister_netdevice_notifier; + } - err = mlx5_roce_lag_init(dev); + err = mlx5_eth_lag_init(dev); if (err) goto err_disable_roce; return 0; err_disable_roce: - mlx5_nic_vport_disable_roce(dev->mdev); + if (MLX5_CAP_GEN(dev->mdev, roce)) + mlx5_nic_vport_disable_roce(dev->mdev); err_unregister_netdevice_notifier: - mlx5_remove_roce_notifier(dev); + mlx5_remove_netdev_notifier(dev); return err; } -static void mlx5_disable_roce(struct mlx5_ib_dev *dev) +static void mlx5_disable_eth(struct mlx5_ib_dev *dev) { - mlx5_roce_lag_cleanup(dev); - mlx5_nic_vport_disable_roce(dev->mdev); + mlx5_eth_lag_cleanup(dev); + if (MLX5_CAP_GEN(dev->mdev, roce)) + mlx5_nic_vport_disable_roce(dev->mdev); } +struct mlx5_ib_q_counter { + const char *name; + size_t offset; +}; + +#define INIT_Q_COUNTER(_name) \ + { .name = #_name, .offset = MLX5_BYTE_OFF(query_q_counter_out, _name)} + +static const struct mlx5_ib_q_counter basic_q_cnts[] = { + INIT_Q_COUNTER(rx_write_requests), + INIT_Q_COUNTER(rx_read_requests), + INIT_Q_COUNTER(rx_atomic_requests), + INIT_Q_COUNTER(out_of_buffer), +}; + +static const struct mlx5_ib_q_counter out_of_seq_q_cnts[] = { + INIT_Q_COUNTER(out_of_sequence), +}; + +static const struct mlx5_ib_q_counter retrans_q_cnts[] = { + INIT_Q_COUNTER(duplicate_request), + INIT_Q_COUNTER(rnr_nak_retry_err), + INIT_Q_COUNTER(packet_seq_err), + INIT_Q_COUNTER(implied_nak_seq_err), + INIT_Q_COUNTER(local_ack_timeout_err), +}; + static void mlx5_ib_dealloc_q_counters(struct mlx5_ib_dev *dev) { unsigned int i; - for (i = 0; i < dev->num_ports; i++) + for (i = 0; i < dev->num_ports; i++) { mlx5_core_dealloc_q_counter(dev->mdev, - dev->port[i].q_cnt_id); + dev->port[i].q_cnts.set_id); + kfree(dev->port[i].q_cnts.names); + kfree(dev->port[i].q_cnts.offsets); + } +} + +static int __mlx5_ib_alloc_q_counters(struct mlx5_ib_dev *dev, + const char ***names, + size_t **offsets, + u32 *num) +{ + u32 num_counters; + + num_counters = ARRAY_SIZE(basic_q_cnts); + + if (MLX5_CAP_GEN(dev->mdev, out_of_seq_cnt)) + num_counters += ARRAY_SIZE(out_of_seq_q_cnts); + + if (MLX5_CAP_GEN(dev->mdev, retransmission_q_counters)) + num_counters += ARRAY_SIZE(retrans_q_cnts); + + *names = kcalloc(num_counters, sizeof(**names), GFP_KERNEL); + if (!*names) + return -ENOMEM; + + *offsets = kcalloc(num_counters, sizeof(**offsets), GFP_KERNEL); + if (!*offsets) + goto err_names; + + *num = num_counters; + + return 0; + +err_names: + kfree(*names); + return -ENOMEM; +} + +static void mlx5_ib_fill_q_counters(struct mlx5_ib_dev *dev, + const char **names, + size_t *offsets) +{ + int i; + int j = 0; + + for (i = 0; i < ARRAY_SIZE(basic_q_cnts); i++, j++) { + names[j] = basic_q_cnts[i].name; + offsets[j] = basic_q_cnts[i].offset; + } + + if (MLX5_CAP_GEN(dev->mdev, out_of_seq_cnt)) { + for (i = 0; i < ARRAY_SIZE(out_of_seq_q_cnts); i++, j++) { + names[j] = out_of_seq_q_cnts[i].name; + offsets[j] = out_of_seq_q_cnts[i].offset; + } + } + + if (MLX5_CAP_GEN(dev->mdev, retransmission_q_counters)) { + for (i = 0; i < ARRAY_SIZE(retrans_q_cnts); i++, j++) { + names[j] = retrans_q_cnts[i].name; + offsets[j] = retrans_q_cnts[i].offset; + } + } } static int mlx5_ib_alloc_q_counters(struct mlx5_ib_dev *dev) @@ -2841,14 +3235,26 @@ static int mlx5_ib_alloc_q_counters(struct mlx5_ib_dev *dev) int ret; for (i = 0; i < dev->num_ports; i++) { + struct mlx5_ib_port *port = &dev->port[i]; + ret = mlx5_core_alloc_q_counter(dev->mdev, - &dev->port[i].q_cnt_id); + &port->q_cnts.set_id); if (ret) { mlx5_ib_warn(dev, "couldn't allocate queue counter for port %d, err %d\n", i + 1, ret); goto dealloc_counters; } + + ret = __mlx5_ib_alloc_q_counters(dev, + &port->q_cnts.names, + &port->q_cnts.offsets, + &port->q_cnts.num_counters); + if (ret) + goto dealloc_counters; + + mlx5_ib_fill_q_counters(dev, port->q_cnts.names, + port->q_cnts.offsets); } return 0; @@ -2856,62 +3262,39 @@ static int mlx5_ib_alloc_q_counters(struct mlx5_ib_dev *dev) dealloc_counters: while (--i >= 0) mlx5_core_dealloc_q_counter(dev->mdev, - dev->port[i].q_cnt_id); + dev->port[i].q_cnts.set_id); return ret; } -static const char * const names[] = { - "rx_write_requests", - "rx_read_requests", - "rx_atomic_requests", - "out_of_buffer", - "out_of_sequence", - "duplicate_request", - "rnr_nak_retry_err", - "packet_seq_err", - "implied_nak_seq_err", - "local_ack_timeout_err", -}; - -static const size_t stats_offsets[] = { - MLX5_BYTE_OFF(query_q_counter_out, rx_write_requests), - MLX5_BYTE_OFF(query_q_counter_out, rx_read_requests), - MLX5_BYTE_OFF(query_q_counter_out, rx_atomic_requests), - MLX5_BYTE_OFF(query_q_counter_out, out_of_buffer), - MLX5_BYTE_OFF(query_q_counter_out, out_of_sequence), - MLX5_BYTE_OFF(query_q_counter_out, duplicate_request), - MLX5_BYTE_OFF(query_q_counter_out, rnr_nak_retry_err), - MLX5_BYTE_OFF(query_q_counter_out, packet_seq_err), - MLX5_BYTE_OFF(query_q_counter_out, implied_nak_seq_err), - MLX5_BYTE_OFF(query_q_counter_out, local_ack_timeout_err), -}; - static struct rdma_hw_stats *mlx5_ib_alloc_hw_stats(struct ib_device *ibdev, u8 port_num) { - BUILD_BUG_ON(ARRAY_SIZE(names) != ARRAY_SIZE(stats_offsets)); + struct mlx5_ib_dev *dev = to_mdev(ibdev); + struct mlx5_ib_port *port = &dev->port[port_num - 1]; /* We support only per port stats */ if (port_num == 0) return NULL; - return rdma_alloc_hw_stats_struct(names, ARRAY_SIZE(names), + return rdma_alloc_hw_stats_struct(port->q_cnts.names, + port->q_cnts.num_counters, RDMA_HW_STATS_DEFAULT_LIFESPAN); } static int mlx5_ib_get_hw_stats(struct ib_device *ibdev, struct rdma_hw_stats *stats, - u8 port, int index) + u8 port_num, int index) { struct mlx5_ib_dev *dev = to_mdev(ibdev); + struct mlx5_ib_port *port = &dev->port[port_num - 1]; int outlen = MLX5_ST_SZ_BYTES(query_q_counter_out); void *out; __be32 val; int ret; int i; - if (!port || !stats) + if (!stats) return -ENOSYS; out = mlx5_vzalloc(outlen); @@ -2919,18 +3302,19 @@ static int mlx5_ib_get_hw_stats(struct ib_device *ibdev, return -ENOMEM; ret = mlx5_core_query_q_counter(dev->mdev, - dev->port[port - 1].q_cnt_id, 0, + port->q_cnts.set_id, 0, out, outlen); if (ret) goto free; - for (i = 0; i < ARRAY_SIZE(names); i++) { - val = *(__be32 *)(out + stats_offsets[i]); + for (i = 0; i < port->q_cnts.num_counters; i++) { + val = *(__be32 *)(out + port->q_cnts.offsets[i]); stats->value[i] = (u64)be32_to_cpu(val); } + free: kvfree(out); - return ARRAY_SIZE(names); + return port->q_cnts.num_counters; } static void *mlx5_ib_add(struct mlx5_core_dev *mdev) @@ -2945,9 +3329,6 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev) port_type_cap = MLX5_CAP_GEN(mdev, port_type); ll = mlx5_port_type_cap_to_rdma_ll(port_type_cap); - if ((ll == IB_LINK_LAYER_ETHERNET) && !MLX5_CAP_GEN(mdev, roce)) - return NULL; - printk_once(KERN_INFO "%s", mlx5_version); dev = (struct mlx5_ib_dev *)ib_alloc_device(sizeof(*dev)); @@ -2969,8 +3350,6 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev) if (mlx5_use_mad_ifc(dev)) get_ext_port_caps(dev); - MLX5_INIT_DOORBELL_LOCK(&dev->uar_lock); - if (!mlx5_lag_is_active(mdev)) name = "mlx5_%d"; else @@ -2984,7 +3363,7 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev) dev->ib_dev.phys_port_cnt = dev->num_ports; dev->ib_dev.num_comp_vectors = dev->mdev->priv.eq_table.num_comp_vectors; - dev->ib_dev.dma_device = &mdev->pdev->dev; + dev->ib_dev.dev.parent = &mdev->pdev->dev; dev->ib_dev.uverbs_abi_ver = MLX5_IB_UVERBS_ABI_VERSION; dev->ib_dev.uverbs_cmd_mask = @@ -2993,6 +3372,8 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev) (1ull << IB_USER_VERBS_CMD_QUERY_PORT) | (1ull << IB_USER_VERBS_CMD_ALLOC_PD) | (1ull << IB_USER_VERBS_CMD_DEALLOC_PD) | + (1ull << IB_USER_VERBS_CMD_CREATE_AH) | + (1ull << IB_USER_VERBS_CMD_DESTROY_AH) | (1ull << IB_USER_VERBS_CMD_REG_MR) | (1ull << IB_USER_VERBS_CMD_REREG_MR) | (1ull << IB_USER_VERBS_CMD_DEREG_MR) | @@ -3015,7 +3396,8 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev) dev->ib_dev.uverbs_ex_cmd_mask = (1ull << IB_USER_VERBS_EX_CMD_QUERY_DEVICE) | (1ull << IB_USER_VERBS_EX_CMD_CREATE_CQ) | - (1ull << IB_USER_VERBS_EX_CMD_CREATE_QP); + (1ull << IB_USER_VERBS_EX_CMD_CREATE_QP) | + (1ull << IB_USER_VERBS_EX_CMD_MODIFY_QP); dev->ib_dev.query_device = mlx5_ib_query_device; dev->ib_dev.query_port = mlx5_ib_query_port; @@ -3084,8 +3466,7 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev) (1ull << IB_USER_VERBS_CMD_DEALLOC_MW); } - if (MLX5_CAP_GEN(dev->mdev, out_of_seq_cnt) && - MLX5_CAP_GEN(dev->mdev, retransmission_q_counters)) { + if (MLX5_CAP_GEN(dev->mdev, max_qp_cnt)) { dev->ib_dev.get_hw_stats = mlx5_ib_get_hw_stats; dev->ib_dev.alloc_hw_stats = mlx5_ib_alloc_hw_stats; } @@ -3126,26 +3507,40 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev) spin_lock_init(&dev->reset_flow_resource_lock); if (ll == IB_LINK_LAYER_ETHERNET) { - err = mlx5_enable_roce(dev); + err = mlx5_enable_eth(dev); if (err) goto err_free_port; } err = create_dev_resources(&dev->devr); if (err) - goto err_disable_roce; + goto err_disable_eth; err = mlx5_ib_odp_init_one(dev); if (err) goto err_rsrc; - err = mlx5_ib_alloc_q_counters(dev); + if (MLX5_CAP_GEN(dev->mdev, max_qp_cnt)) { + err = mlx5_ib_alloc_q_counters(dev); + if (err) + goto err_odp; + } + + dev->mdev->priv.uar = mlx5_get_uars_page(dev->mdev); + if (!dev->mdev->priv.uar) + goto err_q_cnt; + + err = mlx5_alloc_bfreg(dev->mdev, &dev->bfreg, false, false); if (err) - goto err_odp; + goto err_uar_page; + + err = mlx5_alloc_bfreg(dev->mdev, &dev->fp_bfreg, false, true); + if (err) + goto err_bfreg; err = ib_register_device(&dev->ib_dev, NULL); if (err) - goto err_q_cnt; + goto err_fp_bfreg; err = create_umr_res(dev); if (err) @@ -3168,8 +3563,18 @@ err_umrc: err_dev: ib_unregister_device(&dev->ib_dev); +err_fp_bfreg: + mlx5_free_bfreg(dev->mdev, &dev->fp_bfreg); + +err_bfreg: + mlx5_free_bfreg(dev->mdev, &dev->bfreg); + +err_uar_page: + mlx5_put_uars_page(dev->mdev, dev->mdev->priv.uar); + err_q_cnt: - mlx5_ib_dealloc_q_counters(dev); + if (MLX5_CAP_GEN(dev->mdev, max_qp_cnt)) + mlx5_ib_dealloc_q_counters(dev); err_odp: mlx5_ib_odp_remove_one(dev); @@ -3177,10 +3582,10 @@ err_odp: err_rsrc: destroy_dev_resources(&dev->devr); -err_disable_roce: +err_disable_eth: if (ll == IB_LINK_LAYER_ETHERNET) { - mlx5_disable_roce(dev); - mlx5_remove_roce_notifier(dev); + mlx5_disable_eth(dev); + mlx5_remove_netdev_notifier(dev); } err_free_port: @@ -3197,14 +3602,18 @@ static void mlx5_ib_remove(struct mlx5_core_dev *mdev, void *context) struct mlx5_ib_dev *dev = context; enum rdma_link_layer ll = mlx5_ib_port_link_layer(&dev->ib_dev, 1); - mlx5_remove_roce_notifier(dev); + mlx5_remove_netdev_notifier(dev); ib_unregister_device(&dev->ib_dev); - mlx5_ib_dealloc_q_counters(dev); + mlx5_free_bfreg(dev->mdev, &dev->fp_bfreg); + mlx5_free_bfreg(dev->mdev, &dev->bfreg); + mlx5_put_uars_page(dev->mdev, mdev->priv.uar); + if (MLX5_CAP_GEN(dev->mdev, max_qp_cnt)) + mlx5_ib_dealloc_q_counters(dev); destroy_umrc_res(dev); mlx5_ib_odp_remove_one(dev); destroy_dev_resources(&dev->devr); if (ll == IB_LINK_LAYER_ETHERNET) - mlx5_disable_roce(dev); + mlx5_disable_eth(dev); kfree(dev->port); ib_dealloc_device(&dev->ib_dev); } @@ -3213,6 +3622,9 @@ static struct mlx5_interface mlx5_ib_interface = { .add = mlx5_ib_add, .remove = mlx5_ib_remove, .event = mlx5_ib_event, +#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING + .pfault = mlx5_ib_pfault, +#endif .protocol = MLX5_INTERFACE_PROTOCOL_IB, }; @@ -3220,28 +3632,16 @@ static int __init mlx5_ib_init(void) { int err; - if (deprecated_prof_sel != 2) - pr_warn("prof_sel is deprecated for mlx5_ib, set it for mlx5_core\n"); - - err = mlx5_ib_odp_init(); - if (err) - return err; + mlx5_ib_odp_init(); err = mlx5_register_interface(&mlx5_ib_interface); - if (err) - goto clean_odp; - - return err; -clean_odp: - mlx5_ib_odp_cleanup(); return err; } static void __exit mlx5_ib_cleanup(void) { mlx5_unregister_interface(&mlx5_ib_interface); - mlx5_ib_odp_cleanup(); } module_init(mlx5_ib_init); diff --git a/drivers/infiniband/hw/mlx5/mem.c b/drivers/infiniband/hw/mlx5/mem.c index 996b54e366b0..778d8a18925f 100644 --- a/drivers/infiniband/hw/mlx5/mem.c +++ b/drivers/infiniband/hw/mlx5/mem.c @@ -37,12 +37,15 @@ /* @umem: umem object to scan * @addr: ib virtual address requested by the user + * @max_page_shift: high limit for page_shift - 0 means no limit * @count: number of PAGE_SIZE pages covered by umem * @shift: page shift for the compound pages found in the region * @ncont: number of compund pages * @order: log2 of the number of compound pages */ -void mlx5_ib_cont_pages(struct ib_umem *umem, u64 addr, int *count, int *shift, +void mlx5_ib_cont_pages(struct ib_umem *umem, u64 addr, + unsigned long max_page_shift, + int *count, int *shift, int *ncont, int *order) { unsigned long tmp; @@ -72,6 +75,8 @@ void mlx5_ib_cont_pages(struct ib_umem *umem, u64 addr, int *count, int *shift, addr = addr >> page_shift; tmp = (unsigned long)addr; m = find_first_bit(&tmp, BITS_PER_LONG); + if (max_page_shift) + m = min_t(unsigned long, max_page_shift - page_shift, m); skip = 1 << m; mask = skip - 1; i = 0; @@ -154,7 +159,7 @@ void __mlx5_ib_populate_pas(struct mlx5_ib_dev *dev, struct ib_umem *umem, unsigned long umem_page_shift = ilog2(umem->page_size); int shift = page_shift - umem_page_shift; int mask = (1 << shift) - 1; - int i, k; + int i, k, idx; u64 cur = 0; u64 base; int len; @@ -180,18 +185,36 @@ void __mlx5_ib_populate_pas(struct mlx5_ib_dev *dev, struct ib_umem *umem, for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) { len = sg_dma_len(sg) >> umem_page_shift; base = sg_dma_address(sg); - for (k = 0; k < len; k++) { + + /* Skip elements below offset */ + if (i + len < offset << shift) { + i += len; + continue; + } + + /* Skip pages below offset */ + if (i < offset << shift) { + k = (offset << shift) - i; + i = offset << shift; + } else { + k = 0; + } + + for (; k < len; k++) { if (!(i & mask)) { cur = base + (k << umem_page_shift); cur |= access_flags; + idx = (i >> shift) - offset; - pas[i >> shift] = cpu_to_be64(cur); + pas[idx] = cpu_to_be64(cur); mlx5_ib_dbg(dev, "pas[%d] 0x%llx\n", - i >> shift, be64_to_cpu(pas[i >> shift])); - } else - mlx5_ib_dbg(dev, "=====> 0x%llx\n", - base + (k << umem_page_shift)); + i >> shift, be64_to_cpu(pas[idx])); + } i++; + + /* Stop after num_pages reached */ + if (i >> shift >= offset + num_pages) + return; } } } diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h index 7d689903c87c..3cd064b5f0bf 100644 --- a/drivers/infiniband/hw/mlx5/mlx5_ib.h +++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h @@ -63,6 +63,8 @@ pr_warn("%s:%s:%d:(pid %d): " format, (dev)->ib_dev.name, __func__, \ #define MLX5_IB_DEFAULT_UIDX 0xffffff #define MLX5_USER_ASSIGNED_UIDX_MASK __mlx5_mask(qpc, user_index) +#define MLX5_MKEY_PAGE_SHIFT_MASK __mlx5_mask(mkc, log_page_size) + enum { MLX5_IB_MMAP_CMD_SHIFT = 8, MLX5_IB_MMAP_CMD_MASK = 0xff, @@ -88,7 +90,6 @@ enum mlx5_ib_latency_class { MLX5_IB_LATENCY_CLASS_LOW, MLX5_IB_LATENCY_CLASS_MEDIUM, MLX5_IB_LATENCY_CLASS_HIGH, - MLX5_IB_LATENCY_CLASS_FAST_PATH }; enum mlx5_ib_mad_ifc_flags { @@ -98,7 +99,7 @@ enum mlx5_ib_mad_ifc_flags { }; enum { - MLX5_CROSS_CHANNEL_UUAR = 0, + MLX5_CROSS_CHANNEL_BFREG = 0, }; enum { @@ -118,11 +119,16 @@ struct mlx5_ib_ucontext { /* protect doorbell record alloc/free */ struct mutex db_page_mutex; - struct mlx5_uuar_info uuari; + struct mlx5_bfreg_info bfregi; u8 cqe_version; /* Transport Domain number */ u32 tdn; struct list_head vma_private_list; + + unsigned long upd_xlt_page; + /* protect ODP/KSM */ + struct mutex upd_xlt_page_mutex; + u64 lib_caps; }; static inline struct mlx5_ib_ucontext *to_mucontext(struct ib_ucontext *ibucontext) @@ -153,7 +159,7 @@ struct mlx5_ib_flow_handler { struct list_head list; struct ib_flow ibflow; struct mlx5_ib_flow_prio *prio; - struct mlx5_flow_rule *rule; + struct mlx5_flow_handle *rule; }; struct mlx5_ib_flow_db { @@ -172,13 +178,12 @@ struct mlx5_ib_flow_db { * enum ib_send_flags and enum ib_qp_type for low-level driver */ -#define MLX5_IB_SEND_UMR_UNREG IB_SEND_RESERVED_START -#define MLX5_IB_SEND_UMR_FAIL_IF_FREE (IB_SEND_RESERVED_START << 1) -#define MLX5_IB_SEND_UMR_UPDATE_MTT (IB_SEND_RESERVED_START << 2) - -#define MLX5_IB_SEND_UMR_UPDATE_TRANSLATION (IB_SEND_RESERVED_START << 3) -#define MLX5_IB_SEND_UMR_UPDATE_PD (IB_SEND_RESERVED_START << 4) -#define MLX5_IB_SEND_UMR_UPDATE_ACCESS IB_SEND_RESERVED_END +#define MLX5_IB_SEND_UMR_ENABLE_MR (IB_SEND_RESERVED_START << 0) +#define MLX5_IB_SEND_UMR_DISABLE_MR (IB_SEND_RESERVED_START << 1) +#define MLX5_IB_SEND_UMR_FAIL_IF_FREE (IB_SEND_RESERVED_START << 2) +#define MLX5_IB_SEND_UMR_UPDATE_XLT (IB_SEND_RESERVED_START << 3) +#define MLX5_IB_SEND_UMR_UPDATE_TRANSLATION (IB_SEND_RESERVED_START << 4) +#define MLX5_IB_SEND_UMR_UPDATE_PD_ACCESS IB_SEND_RESERVED_END #define MLX5_IB_QPT_REG_UMR IB_QPT_RESERVED1 /* @@ -188,6 +193,17 @@ struct mlx5_ib_flow_db { #define MLX5_IB_QPT_HW_GSI IB_QPT_RESERVED2 #define MLX5_IB_WR_UMR IB_WR_RESERVED1 +#define MLX5_IB_UMR_OCTOWORD 16 +#define MLX5_IB_UMR_XLT_ALIGNMENT 64 + +#define MLX5_IB_UPD_XLT_ZAP BIT(0) +#define MLX5_IB_UPD_XLT_ENABLE BIT(1) +#define MLX5_IB_UPD_XLT_ATOMIC BIT(2) +#define MLX5_IB_UPD_XLT_ADDR BIT(3) +#define MLX5_IB_UPD_XLT_PD BIT(4) +#define MLX5_IB_UPD_XLT_ACCESS BIT(5) +#define MLX5_IB_UPD_XLT_INDIRECT BIT(6) + /* Private QP creation flags to be passed in ib_qp_init_attr.create_flags. * * These flags are intended for internal use by the mlx5_ib driver, and they @@ -205,6 +221,10 @@ struct wr_list { u16 next; }; +enum mlx5_ib_rq_flags { + MLX5_IB_RQ_CVLAN_STRIPPING = 1 << 0, +}; + struct mlx5_ib_wq { u64 *wrid; u32 *wr_data; @@ -262,29 +282,6 @@ struct mlx5_ib_rwq_ind_table { u32 rqtn; }; -/* - * Connect-IB can trigger up to four concurrent pagefaults - * per-QP. - */ -enum mlx5_ib_pagefault_context { - MLX5_IB_PAGEFAULT_RESPONDER_READ, - MLX5_IB_PAGEFAULT_REQUESTOR_READ, - MLX5_IB_PAGEFAULT_RESPONDER_WRITE, - MLX5_IB_PAGEFAULT_REQUESTOR_WRITE, - MLX5_IB_PAGEFAULT_CONTEXTS -}; - -static inline enum mlx5_ib_pagefault_context - mlx5_ib_get_pagefault_context(struct mlx5_pagefault *pagefault) -{ - return pagefault->flags & (MLX5_PFAULT_REQUESTOR | MLX5_PFAULT_WRITE); -} - -struct mlx5_ib_pfault { - struct work_struct work; - struct mlx5_pagefault mpfault; -}; - struct mlx5_ib_ubuffer { struct ib_umem *umem; int buf_size; @@ -316,6 +313,7 @@ struct mlx5_ib_rq { struct mlx5_db *doorbell; u32 tirn; u8 state; + u32 flags; }; struct mlx5_ib_sq { @@ -332,6 +330,12 @@ struct mlx5_ib_raw_packet_qp { struct mlx5_ib_rq rq; }; +struct mlx5_bf { + int buf_size; + unsigned long offset; + struct mlx5_sq_bfreg *bfreg; +}; + struct mlx5_ib_qp { struct ib_qp ibqp; union { @@ -357,36 +361,23 @@ struct mlx5_ib_qp { int wq_sig; int scat_cqe; int max_inline_data; - struct mlx5_bf *bf; + struct mlx5_bf bf; int has_rq; /* only for user space QPs. For kernel * we have it from the bf object */ - int uuarn; + int bfregn; int create_type; /* Store signature errors */ bool signature_en; -#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING - /* - * A flag that is true for QP's that are in a state that doesn't - * allow page faults, and shouldn't schedule any more faults. - */ - int disable_page_faults; - /* - * The disable_page_faults_lock protects a QP's disable_page_faults - * field, allowing for a thread to atomically check whether the QP - * allows page faults, and if so schedule a page fault. - */ - spinlock_t disable_page_faults_lock; - struct mlx5_ib_pfault pagefaults[MLX5_IB_PAGEFAULT_CONTEXTS]; -#endif struct list_head qps_list; struct list_head cq_recv_list; struct list_head cq_send_list; + u32 rate_limit; }; struct mlx5_ib_cq_buf { @@ -407,18 +398,17 @@ enum mlx5_ib_qp_flags { MLX5_IB_QP_SQPN_QP1 = 1 << 6, MLX5_IB_QP_CAP_SCATTER_FCS = 1 << 7, MLX5_IB_QP_RSS = 1 << 8, + MLX5_IB_QP_CVLAN_STRIPPING = 1 << 9, }; struct mlx5_umr_wr { struct ib_send_wr wr; - union { - u64 virt_addr; - u64 offset; - } target; + u64 virt_addr; + u64 offset; struct ib_pd *pd; unsigned int page_shift; - unsigned int npages; - u32 length; + unsigned int xlt_size; + u64 length; int access_flags; u32 mkey; }; @@ -514,6 +504,10 @@ struct mlx5_ib_mr { int live; void *descs_alloc; int access_flags; /* Needed for rereg MR */ + + struct mlx5_ib_mr *parent; + atomic_t num_leaf_free; + wait_queue_head_t q_leaf_free; }; struct mlx5_ib_mw { @@ -552,6 +546,10 @@ struct mlx5_cache_ent { struct dentry *dir; char name[4]; u32 order; + u32 xlt; + u32 access_mode; + u32 page; + u32 size; u32 cur; u32 miss; @@ -566,6 +564,7 @@ struct mlx5_cache_ent { struct work_struct work; struct delayed_work dwork; int pending; + struct completion compl; }; struct mlx5_mr_cache { @@ -596,8 +595,15 @@ struct mlx5_ib_resources { struct mutex mutex; }; +struct mlx5_ib_q_counters { + const char **names; + size_t *offsets; + u32 num_counters; + u16 set_id; +}; + struct mlx5_ib_port { - u16 q_cnt_id; + struct mlx5_ib_q_counters q_cnts; }; struct mlx5_roce { @@ -614,7 +620,6 @@ struct mlx5_ib_dev { struct ib_device ib_dev; struct mlx5_core_dev *mdev; struct mlx5_roce roce; - MLX5_DECLARE_DOORBELL_LOCK(uar_lock); int num_ports; /* serialize update of capability mask */ @@ -631,11 +636,13 @@ struct mlx5_ib_dev { int fill_delay; #ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING struct ib_odp_caps odp_caps; + u64 odp_max_size; /* * Sleepable RCU that prevents destruction of MRs while they are still * being used by a page fault handler. */ struct srcu_struct mr_srcu; + u32 null_mkey; #endif struct mlx5_ib_flow_db flow_db; /* protect resources needed as part of reset flow */ @@ -643,6 +650,8 @@ struct mlx5_ib_dev { struct list_head qp_list; /* Array with num_ports elements */ struct mlx5_ib_port *port; + struct mlx5_sq_bfreg bfreg; + struct mlx5_sq_bfreg fp_bfreg; }; static inline struct mlx5_ib_cq *to_mibcq(struct mlx5_core_cq *mcq) @@ -739,7 +748,8 @@ void mlx5_ib_free_srq_wqe(struct mlx5_ib_srq *srq, int wqe_index); int mlx5_MAD_IFC(struct mlx5_ib_dev *dev, int ignore_mkey, int ignore_bkey, u8 port, const struct ib_wc *in_wc, const struct ib_grh *in_grh, const void *in_mad, void *response_mad); -struct ib_ah *mlx5_ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr); +struct ib_ah *mlx5_ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, + struct ib_udata *udata); int mlx5_ib_query_ah(struct ib_ah *ibah, struct ib_ah_attr *ah_attr); int mlx5_ib_destroy_ah(struct ib_ah *ah); struct ib_srq *mlx5_ib_create_srq(struct ib_pd *pd, @@ -783,8 +793,11 @@ struct ib_mr *mlx5_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, struct ib_mw *mlx5_ib_alloc_mw(struct ib_pd *pd, enum ib_mw_type type, struct ib_udata *udata); int mlx5_ib_dealloc_mw(struct ib_mw *mw); -int mlx5_ib_update_mtt(struct mlx5_ib_mr *mr, u64 start_page_index, - int npages, int zap); +int mlx5_ib_update_xlt(struct mlx5_ib_mr *mr, u64 idx, int npages, + int page_shift, int flags); +struct mlx5_ib_mr *mlx5_ib_alloc_implicit_mr(struct mlx5_ib_pd *pd, + int access_flags); +void mlx5_ib_free_implicit_mr(struct mlx5_ib_mr *mr); int mlx5_ib_rereg_user_mr(struct ib_mr *ib_mr, int flags, u64 start, u64 length, u64 virt_addr, int access_flags, struct ib_pd *pd, struct ib_udata *udata); @@ -825,7 +838,9 @@ int mlx5_ib_query_port(struct ib_device *ibdev, u8 port, struct ib_port_attr *props); int mlx5_ib_init_fmr(struct mlx5_ib_dev *dev); void mlx5_ib_cleanup_fmr(struct mlx5_ib_dev *dev); -void mlx5_ib_cont_pages(struct ib_umem *umem, u64 addr, int *count, int *shift, +void mlx5_ib_cont_pages(struct ib_umem *umem, u64 addr, + unsigned long max_page_shift, + int *count, int *shift, int *ncont, int *order); void __mlx5_ib_populate_pas(struct mlx5_ib_dev *dev, struct ib_umem *umem, int page_shift, size_t offset, size_t num_pages, @@ -836,7 +851,9 @@ void mlx5_ib_copy_pas(u64 *old, u64 *new, int step, int num); int mlx5_ib_get_cqe_size(struct mlx5_ib_dev *dev, struct ib_cq *ibcq); int mlx5_mr_cache_init(struct mlx5_ib_dev *dev); int mlx5_mr_cache_cleanup(struct mlx5_ib_dev *dev); -int mlx5_mr_ib_cont_pages(struct ib_umem *umem, u64 addr, int *count, int *shift); + +struct mlx5_ib_mr *mlx5_mr_cache_alloc(struct mlx5_ib_dev *dev, int entry); +void mlx5_mr_cache_free(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr); int mlx5_ib_check_mr_status(struct ib_mr *ibmr, u32 check_mask, struct ib_mr_status *mr_status); struct ib_wq *mlx5_ib_create_wq(struct ib_pd *pd, @@ -851,33 +868,32 @@ struct ib_rwq_ind_table *mlx5_ib_create_rwq_ind_table(struct ib_device *device, int mlx5_ib_destroy_rwq_ind_table(struct ib_rwq_ind_table *wq_ind_table); #ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING -extern struct workqueue_struct *mlx5_ib_page_fault_wq; - void mlx5_ib_internal_fill_odp_caps(struct mlx5_ib_dev *dev); -void mlx5_ib_mr_pfault_handler(struct mlx5_ib_qp *qp, - struct mlx5_ib_pfault *pfault); -void mlx5_ib_odp_create_qp(struct mlx5_ib_qp *qp); +void mlx5_ib_pfault(struct mlx5_core_dev *mdev, void *context, + struct mlx5_pagefault *pfault); int mlx5_ib_odp_init_one(struct mlx5_ib_dev *ibdev); void mlx5_ib_odp_remove_one(struct mlx5_ib_dev *ibdev); int __init mlx5_ib_odp_init(void); void mlx5_ib_odp_cleanup(void); -void mlx5_ib_qp_disable_pagefaults(struct mlx5_ib_qp *qp); -void mlx5_ib_qp_enable_pagefaults(struct mlx5_ib_qp *qp); void mlx5_ib_invalidate_range(struct ib_umem *umem, unsigned long start, unsigned long end); +void mlx5_odp_init_mr_cache_entry(struct mlx5_cache_ent *ent); +void mlx5_odp_populate_klm(struct mlx5_klm *pklm, size_t offset, + size_t nentries, struct mlx5_ib_mr *mr, int flags); #else /* CONFIG_INFINIBAND_ON_DEMAND_PAGING */ static inline void mlx5_ib_internal_fill_odp_caps(struct mlx5_ib_dev *dev) { return; } -static inline void mlx5_ib_odp_create_qp(struct mlx5_ib_qp *qp) {} static inline int mlx5_ib_odp_init_one(struct mlx5_ib_dev *ibdev) { return 0; } -static inline void mlx5_ib_odp_remove_one(struct mlx5_ib_dev *ibdev) {} +static inline void mlx5_ib_odp_remove_one(struct mlx5_ib_dev *ibdev) {} static inline int mlx5_ib_odp_init(void) { return 0; } -static inline void mlx5_ib_odp_cleanup(void) {} -static inline void mlx5_ib_qp_disable_pagefaults(struct mlx5_ib_qp *qp) {} -static inline void mlx5_ib_qp_enable_pagefaults(struct mlx5_ib_qp *qp) {} +static inline void mlx5_ib_odp_cleanup(void) {} +static inline void mlx5_odp_init_mr_cache_entry(struct mlx5_cache_ent *ent) {} +static inline void mlx5_odp_populate_klm(struct mlx5_klm *pklm, size_t offset, + size_t nentries, struct mlx5_ib_mr *mr, + int flags) {} #endif /* CONFIG_INFINIBAND_ON_DEMAND_PAGING */ @@ -892,6 +908,8 @@ int mlx5_ib_set_vf_guid(struct ib_device *device, int vf, u8 port, __be16 mlx5_get_roce_udp_sport(struct mlx5_ib_dev *dev, u8 port_num, int index); +int mlx5_get_roce_gid_type(struct mlx5_ib_dev *dev, u8 port_num, + int index, enum ib_gid_type *gid_type); /* GSI QP helper functions */ struct ib_qp *mlx5_ib_gsi_create_qp(struct ib_pd *pd, @@ -995,4 +1013,17 @@ static inline int get_srq_user_index(struct mlx5_ib_ucontext *ucontext, return verify_assign_uidx(cqe_version, ucmd->uidx, user_index); } + +static inline int get_uars_per_sys_page(struct mlx5_ib_dev *dev, bool lib_support) +{ + return lib_support && MLX5_CAP_GEN(dev->mdev, uar_4k) ? + MLX5_UARS_IN_PAGE : 1; +} + +static inline int get_num_uars(struct mlx5_ib_dev *dev, + struct mlx5_bfreg_info *bfregi) +{ + return get_uars_per_sys_page(dev, bfregi->lib_uar_4k) * bfregi->num_sys_pages; +} + #endif /* MLX5_IB_H */ diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c index 4e9012463c37..b8f9382a8b7d 100644 --- a/drivers/infiniband/hw/mlx5/mr.c +++ b/drivers/infiniband/hw/mlx5/mr.c @@ -46,14 +46,10 @@ enum { }; #define MLX5_UMR_ALIGN 2048 -#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING -static __be64 mlx5_ib_update_mtt_emergency_buffer[ - MLX5_UMR_MTT_MIN_CHUNK_SIZE/sizeof(__be64)] - __aligned(MLX5_UMR_ALIGN); -static DEFINE_MUTEX(mlx5_ib_update_mtt_emergency_buffer_mutex); -#endif static int clean_mr(struct mlx5_ib_mr *mr); +static int use_umr(struct mlx5_ib_dev *dev, int order); +static int unreg_umr(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr); static int destroy_mkey(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr) { @@ -134,6 +130,7 @@ static void reg_mr_callback(int status, void *context) return; } + mr->mmkey.type = MLX5_MKEY_MR; spin_lock_irqsave(&dev->mdev->priv.mkey_lock, flags); key = dev->mdev->priv.mkey_key++; spin_unlock_irqrestore(&dev->mdev->priv.mkey_lock, flags); @@ -153,6 +150,9 @@ static void reg_mr_callback(int status, void *context) if (err) pr_err("Error inserting to mkey tree. 0x%x\n", -err); write_unlock_irqrestore(&table->lock, flags); + + if (!completion_done(&ent->compl)) + complete(&ent->compl); } static int add_keys(struct mlx5_ib_dev *dev, int c, int num) @@ -161,7 +161,6 @@ static int add_keys(struct mlx5_ib_dev *dev, int c, int num) struct mlx5_cache_ent *ent = &cache->ent[c]; int inlen = MLX5_ST_SZ_BYTES(create_mkey_in); struct mlx5_ib_mr *mr; - int npages = 1 << ent->order; void *mkc; u32 *in; int err = 0; @@ -189,11 +188,11 @@ static int add_keys(struct mlx5_ib_dev *dev, int c, int num) MLX5_SET(mkc, mkc, free, 1); MLX5_SET(mkc, mkc, umr_en, 1); - MLX5_SET(mkc, mkc, access_mode, MLX5_MKC_ACCESS_MODE_MTT); + MLX5_SET(mkc, mkc, access_mode, ent->access_mode); MLX5_SET(mkc, mkc, qpn, 0xffffff); - MLX5_SET(mkc, mkc, translations_octword_size, (npages + 1) / 2); - MLX5_SET(mkc, mkc, log_page_size, 12); + MLX5_SET(mkc, mkc, translations_octword_size, ent->xlt); + MLX5_SET(mkc, mkc, log_page_size, ent->page); spin_lock_irq(&ent->lock); ent->pending++; @@ -451,6 +450,42 @@ static void cache_work_func(struct work_struct *work) __cache_work_func(ent); } +struct mlx5_ib_mr *mlx5_mr_cache_alloc(struct mlx5_ib_dev *dev, int entry) +{ + struct mlx5_mr_cache *cache = &dev->cache; + struct mlx5_cache_ent *ent; + struct mlx5_ib_mr *mr; + int err; + + if (entry < 0 || entry >= MAX_MR_CACHE_ENTRIES) { + mlx5_ib_err(dev, "cache entry %d is out of range\n", entry); + return NULL; + } + + ent = &cache->ent[entry]; + while (1) { + spin_lock_irq(&ent->lock); + if (list_empty(&ent->head)) { + spin_unlock_irq(&ent->lock); + + err = add_keys(dev, entry, 1); + if (err && err != -EAGAIN) + return ERR_PTR(err); + + wait_for_completion(&ent->compl); + } else { + mr = list_first_entry(&ent->head, struct mlx5_ib_mr, + list); + list_del(&mr->list); + ent->cur--; + spin_unlock_irq(&ent->lock); + if (ent->cur < ent->limit) + queue_work(cache->wq, &ent->work); + return mr; + } + } +} + static struct mlx5_ib_mr *alloc_cached_mr(struct mlx5_ib_dev *dev, int order) { struct mlx5_mr_cache *cache = &dev->cache; @@ -460,12 +495,12 @@ static struct mlx5_ib_mr *alloc_cached_mr(struct mlx5_ib_dev *dev, int order) int i; c = order2idx(dev, order); - if (c < 0 || c >= MAX_MR_CACHE_ENTRIES) { + if (c < 0 || c > MAX_UMR_CACHE_ENTRY) { mlx5_ib_warn(dev, "order %d, cache index %d\n", order, c); return NULL; } - for (i = c; i < MAX_MR_CACHE_ENTRIES; i++) { + for (i = c; i < MAX_UMR_CACHE_ENTRY; i++) { ent = &cache->ent[i]; mlx5_ib_dbg(dev, "order %d, cache index %d\n", ent->order, i); @@ -492,7 +527,7 @@ static struct mlx5_ib_mr *alloc_cached_mr(struct mlx5_ib_dev *dev, int order) return mr; } -static void free_cached_mr(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr) +void mlx5_mr_cache_free(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr) { struct mlx5_mr_cache *cache = &dev->cache; struct mlx5_cache_ent *ent; @@ -504,6 +539,10 @@ static void free_cached_mr(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr) mlx5_ib_warn(dev, "order %d, cache index %d\n", mr->order, c); return; } + + if (unreg_umr(dev, mr)) + return; + ent = &cache->ent[c]; spin_lock_irq(&ent->lock); list_add_tail(&mr->list, &ent->head); @@ -606,7 +645,6 @@ int mlx5_mr_cache_init(struct mlx5_ib_dev *dev) { struct mlx5_mr_cache *cache = &dev->cache; struct mlx5_cache_ent *ent; - int limit; int err; int i; @@ -619,24 +657,35 @@ int mlx5_mr_cache_init(struct mlx5_ib_dev *dev) setup_timer(&dev->delay_timer, delay_time_func, (unsigned long)dev); for (i = 0; i < MAX_MR_CACHE_ENTRIES; i++) { - INIT_LIST_HEAD(&cache->ent[i].head); - spin_lock_init(&cache->ent[i].lock); - ent = &cache->ent[i]; INIT_LIST_HEAD(&ent->head); spin_lock_init(&ent->lock); ent->order = i + 2; ent->dev = dev; + ent->limit = 0; - if (dev->mdev->profile->mask & MLX5_PROF_MASK_MR_CACHE) - limit = dev->mdev->profile->mr_cache[i].limit; - else - limit = 0; - + init_completion(&ent->compl); INIT_WORK(&ent->work, cache_work_func); INIT_DELAYED_WORK(&ent->dwork, delayed_cache_work_func); - ent->limit = limit; queue_work(cache->wq, &ent->work); + + if (i > MAX_UMR_CACHE_ENTRY) { + mlx5_odp_init_mr_cache_entry(ent); + continue; + } + + if (!use_umr(dev, ent->order)) + continue; + + ent->page = PAGE_SHIFT; + ent->xlt = (1 << ent->order) * sizeof(struct mlx5_mtt) / + MLX5_IB_UMR_OCTOWORD; + ent->access_mode = MLX5_MKC_ACCESS_MODE_MTT; + if ((dev->mdev->profile->mask & MLX5_PROF_MASK_MR_CACHE) && + mlx5_core_is_pf(dev->mdev)) + ent->limit = dev->mdev->profile->mr_cache[i].limit; + else + ent->limit = 0; } err = mlx5_mr_cache_debugfs_init(dev); @@ -646,6 +695,33 @@ int mlx5_mr_cache_init(struct mlx5_ib_dev *dev) return 0; } +static void wait_for_async_commands(struct mlx5_ib_dev *dev) +{ + struct mlx5_mr_cache *cache = &dev->cache; + struct mlx5_cache_ent *ent; + int total = 0; + int i; + int j; + + for (i = 0; i < MAX_MR_CACHE_ENTRIES; i++) { + ent = &cache->ent[i]; + for (j = 0 ; j < 1000; j++) { + if (!ent->pending) + break; + msleep(50); + } + } + for (i = 0; i < MAX_MR_CACHE_ENTRIES; i++) { + ent = &cache->ent[i]; + total += ent->pending; + } + + if (total) + mlx5_ib_warn(dev, "aborted while there are %d pending mr requests\n", total); + else + mlx5_ib_warn(dev, "done with all pending requests\n"); +} + int mlx5_mr_cache_cleanup(struct mlx5_ib_dev *dev) { int i; @@ -659,6 +735,7 @@ int mlx5_mr_cache_cleanup(struct mlx5_ib_dev *dev) clean_keys(dev, i); destroy_workqueue(dev->cache.wq); + wait_for_async_commands(dev); del_timer_sync(&dev->delay_timer); return 0; @@ -703,6 +780,7 @@ struct ib_mr *mlx5_ib_get_dma_mr(struct ib_pd *pd, int acc) goto err_in; kfree(in); + mr->mmkey.type = MLX5_MKEY_MR; mr->ibmr.lkey = mr->mmkey.key; mr->ibmr.rkey = mr->mmkey.key; mr->umem = NULL; @@ -728,117 +806,41 @@ static int get_octo_len(u64 addr, u64 len, int page_size) return (npages + 1) / 2; } -static int use_umr(int order) +static int use_umr(struct mlx5_ib_dev *dev, int order) { + if (MLX5_CAP_GEN(dev->mdev, umr_extended_translation_offset)) + return order <= MAX_UMR_CACHE_ENTRY + 2; return order <= MLX5_MAX_UMR_SHIFT; } -static int dma_map_mr_pas(struct mlx5_ib_dev *dev, struct ib_umem *umem, - int npages, int page_shift, int *size, - __be64 **mr_pas, dma_addr_t *dma) -{ - __be64 *pas; - struct device *ddev = dev->ib_dev.dma_device; - - /* - * UMR copies MTTs in units of MLX5_UMR_MTT_ALIGNMENT bytes. - * To avoid copying garbage after the pas array, we allocate - * a little more. - */ - *size = ALIGN(sizeof(u64) * npages, MLX5_UMR_MTT_ALIGNMENT); - *mr_pas = kmalloc(*size + MLX5_UMR_ALIGN - 1, GFP_KERNEL); - if (!(*mr_pas)) - return -ENOMEM; - - pas = PTR_ALIGN(*mr_pas, MLX5_UMR_ALIGN); - mlx5_ib_populate_pas(dev, umem, page_shift, pas, MLX5_IB_MTT_PRESENT); - /* Clear padding after the actual pages. */ - memset(pas + npages, 0, *size - npages * sizeof(u64)); - - *dma = dma_map_single(ddev, pas, *size, DMA_TO_DEVICE); - if (dma_mapping_error(ddev, *dma)) { - kfree(*mr_pas); - return -ENOMEM; - } - - return 0; -} - -static void prep_umr_wqe_common(struct ib_pd *pd, struct ib_send_wr *wr, - struct ib_sge *sg, u64 dma, int n, u32 key, - int page_shift) +static int mr_umem_get(struct ib_pd *pd, u64 start, u64 length, + int access_flags, struct ib_umem **umem, + int *npages, int *page_shift, int *ncont, + int *order) { struct mlx5_ib_dev *dev = to_mdev(pd->device); - struct mlx5_umr_wr *umrwr = umr_wr(wr); - - sg->addr = dma; - sg->length = ALIGN(sizeof(u64) * n, 64); - sg->lkey = dev->umrc.pd->local_dma_lkey; - - wr->next = NULL; - wr->sg_list = sg; - if (n) - wr->num_sge = 1; - else - wr->num_sge = 0; - - wr->opcode = MLX5_IB_WR_UMR; - - umrwr->npages = n; - umrwr->page_shift = page_shift; - umrwr->mkey = key; -} - -static void prep_umr_reg_wqe(struct ib_pd *pd, struct ib_send_wr *wr, - struct ib_sge *sg, u64 dma, int n, u32 key, - int page_shift, u64 virt_addr, u64 len, - int access_flags) -{ - struct mlx5_umr_wr *umrwr = umr_wr(wr); - - prep_umr_wqe_common(pd, wr, sg, dma, n, key, page_shift); - - wr->send_flags = 0; - - umrwr->target.virt_addr = virt_addr; - umrwr->length = len; - umrwr->access_flags = access_flags; - umrwr->pd = pd; -} - -static void prep_umr_unreg_wqe(struct mlx5_ib_dev *dev, - struct ib_send_wr *wr, u32 key) -{ - struct mlx5_umr_wr *umrwr = umr_wr(wr); - - wr->send_flags = MLX5_IB_SEND_UMR_UNREG | MLX5_IB_SEND_UMR_FAIL_IF_FREE; - wr->opcode = MLX5_IB_WR_UMR; - umrwr->mkey = key; -} + int err; -static struct ib_umem *mr_umem_get(struct ib_pd *pd, u64 start, u64 length, - int access_flags, int *npages, - int *page_shift, int *ncont, int *order) -{ - struct mlx5_ib_dev *dev = to_mdev(pd->device); - struct ib_umem *umem = ib_umem_get(pd->uobject->context, start, length, - access_flags, 0); - if (IS_ERR(umem)) { + *umem = ib_umem_get(pd->uobject->context, start, length, + access_flags, 0); + err = PTR_ERR_OR_ZERO(*umem); + if (err < 0) { mlx5_ib_err(dev, "umem get failed (%ld)\n", PTR_ERR(umem)); - return (void *)umem; + return err; } - mlx5_ib_cont_pages(umem, start, npages, page_shift, ncont, order); + mlx5_ib_cont_pages(*umem, start, MLX5_MKEY_PAGE_SHIFT_MASK, npages, + page_shift, ncont, order); if (!*npages) { mlx5_ib_warn(dev, "avoid zero region\n"); - ib_umem_release(umem); - return ERR_PTR(-EINVAL); + ib_umem_release(*umem); + return -EINVAL; } mlx5_ib_dbg(dev, "npages %d, ncont %d, order %d, page_shift %d\n", *npages, *ncont, *order, *page_shift); - return umem; + return 0; } static void mlx5_ib_umr_done(struct ib_cq *cq, struct ib_wc *wc) @@ -857,21 +859,39 @@ static inline void mlx5_ib_init_umr_context(struct mlx5_ib_umr_context *context) init_completion(&context->done); } +static int mlx5_ib_post_send_wait(struct mlx5_ib_dev *dev, + struct mlx5_umr_wr *umrwr) +{ + struct umr_common *umrc = &dev->umrc; + struct ib_send_wr *bad; + int err; + struct mlx5_ib_umr_context umr_context; + + mlx5_ib_init_umr_context(&umr_context); + umrwr->wr.wr_cqe = &umr_context.cqe; + + down(&umrc->sem); + err = ib_post_send(umrc->qp, &umrwr->wr, &bad); + if (err) { + mlx5_ib_warn(dev, "UMR post send failed, err %d\n", err); + } else { + wait_for_completion(&umr_context.done); + if (umr_context.status != IB_WC_SUCCESS) { + mlx5_ib_warn(dev, "reg umr failed (%u)\n", + umr_context.status); + err = -EFAULT; + } + } + up(&umrc->sem); + return err; +} + static struct mlx5_ib_mr *reg_umr(struct ib_pd *pd, struct ib_umem *umem, u64 virt_addr, u64 len, int npages, int page_shift, int order, int access_flags) { struct mlx5_ib_dev *dev = to_mdev(pd->device); - struct device *ddev = dev->ib_dev.dma_device; - struct umr_common *umrc = &dev->umrc; - struct mlx5_ib_umr_context umr_context; - struct mlx5_umr_wr umrwr = {}; - struct ib_send_wr *bad; struct mlx5_ib_mr *mr; - struct ib_sge sg; - int size; - __be64 *mr_pas; - dma_addr_t dma; int err = 0; int i; @@ -890,173 +910,180 @@ static struct mlx5_ib_mr *reg_umr(struct ib_pd *pd, struct ib_umem *umem, if (!mr) return ERR_PTR(-EAGAIN); - err = dma_map_mr_pas(dev, umem, npages, page_shift, &size, &mr_pas, - &dma); - if (err) - goto free_mr; - - mlx5_ib_init_umr_context(&umr_context); - - umrwr.wr.wr_cqe = &umr_context.cqe; - prep_umr_reg_wqe(pd, &umrwr.wr, &sg, dma, npages, mr->mmkey.key, - page_shift, virt_addr, len, access_flags); - - down(&umrc->sem); - err = ib_post_send(umrc->qp, &umrwr.wr, &bad); - if (err) { - mlx5_ib_warn(dev, "post send failed, err %d\n", err); - goto unmap_dma; - } else { - wait_for_completion(&umr_context.done); - if (umr_context.status != IB_WC_SUCCESS) { - mlx5_ib_warn(dev, "reg umr failed\n"); - err = -EFAULT; - } - } - + mr->ibmr.pd = pd; + mr->umem = umem; + mr->access_flags = access_flags; + mr->desc_size = sizeof(struct mlx5_mtt); mr->mmkey.iova = virt_addr; mr->mmkey.size = len; mr->mmkey.pd = to_mpd(pd)->pdn; - mr->live = 1; - -unmap_dma: - up(&umrc->sem); - dma_unmap_single(ddev, dma, size, DMA_TO_DEVICE); - - kfree(mr_pas); + err = mlx5_ib_update_xlt(mr, 0, npages, page_shift, + MLX5_IB_UPD_XLT_ENABLE); -free_mr: if (err) { - free_cached_mr(dev, mr); + mlx5_mr_cache_free(dev, mr); return ERR_PTR(err); } + mr->live = 1; + return mr; } -#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING -int mlx5_ib_update_mtt(struct mlx5_ib_mr *mr, u64 start_page_index, int npages, - int zap) +static inline int populate_xlt(struct mlx5_ib_mr *mr, int idx, int npages, + void *xlt, int page_shift, size_t size, + int flags) { struct mlx5_ib_dev *dev = mr->dev; - struct device *ddev = dev->ib_dev.dma_device; - struct umr_common *umrc = &dev->umrc; - struct mlx5_ib_umr_context umr_context; struct ib_umem *umem = mr->umem; + if (flags & MLX5_IB_UPD_XLT_INDIRECT) { + mlx5_odp_populate_klm(xlt, idx, npages, mr, flags); + return npages; + } + + npages = min_t(size_t, npages, ib_umem_num_pages(umem) - idx); + + if (!(flags & MLX5_IB_UPD_XLT_ZAP)) { + __mlx5_ib_populate_pas(dev, umem, page_shift, + idx, npages, xlt, + MLX5_IB_MTT_PRESENT); + /* Clear padding after the pages + * brought from the umem. + */ + memset(xlt + (npages * sizeof(struct mlx5_mtt)), 0, + size - npages * sizeof(struct mlx5_mtt)); + } + + return npages; +} + +#define MLX5_MAX_UMR_CHUNK ((1 << (MLX5_MAX_UMR_SHIFT + 4)) - \ + MLX5_UMR_MTT_ALIGNMENT) +#define MLX5_SPARE_UMR_CHUNK 0x10000 + +int mlx5_ib_update_xlt(struct mlx5_ib_mr *mr, u64 idx, int npages, + int page_shift, int flags) +{ + struct mlx5_ib_dev *dev = mr->dev; + struct device *ddev = dev->ib_dev.dev.parent; + struct mlx5_ib_ucontext *uctx = NULL; int size; - __be64 *pas; + void *xlt; dma_addr_t dma; - struct ib_send_wr *bad; struct mlx5_umr_wr wr; struct ib_sge sg; int err = 0; - const int page_index_alignment = MLX5_UMR_MTT_ALIGNMENT / sizeof(u64); - const int page_index_mask = page_index_alignment - 1; + int desc_size = (flags & MLX5_IB_UPD_XLT_INDIRECT) + ? sizeof(struct mlx5_klm) + : sizeof(struct mlx5_mtt); + const int page_align = MLX5_UMR_MTT_ALIGNMENT / desc_size; + const int page_mask = page_align - 1; size_t pages_mapped = 0; size_t pages_to_map = 0; size_t pages_iter = 0; - int use_emergency_buf = 0; + gfp_t gfp; /* UMR copies MTTs in units of MLX5_UMR_MTT_ALIGNMENT bytes, - * so we need to align the offset and length accordingly */ - if (start_page_index & page_index_mask) { - npages += start_page_index & page_index_mask; - start_page_index &= ~page_index_mask; + * so we need to align the offset and length accordingly + */ + if (idx & page_mask) { + npages += idx & page_mask; + idx &= ~page_mask; } - pages_to_map = ALIGN(npages, page_index_alignment); + gfp = flags & MLX5_IB_UPD_XLT_ATOMIC ? GFP_ATOMIC : GFP_KERNEL; + gfp |= __GFP_ZERO | __GFP_NOWARN; - if (start_page_index + pages_to_map > MLX5_MAX_UMR_PAGES) - return -EINVAL; + pages_to_map = ALIGN(npages, page_align); + size = desc_size * pages_to_map; + size = min_t(int, size, MLX5_MAX_UMR_CHUNK); + + xlt = (void *)__get_free_pages(gfp, get_order(size)); + if (!xlt && size > MLX5_SPARE_UMR_CHUNK) { + mlx5_ib_dbg(dev, "Failed to allocate %d bytes of order %d. fallback to spare UMR allocation od %d bytes\n", + size, get_order(size), MLX5_SPARE_UMR_CHUNK); + + size = MLX5_SPARE_UMR_CHUNK; + xlt = (void *)__get_free_pages(gfp, get_order(size)); + } - size = sizeof(u64) * pages_to_map; - size = min_t(int, PAGE_SIZE, size); - /* We allocate with GFP_ATOMIC to avoid recursion into page-reclaim - * code, when we are called from an invalidation. The pas buffer must - * be 2k-aligned for Connect-IB. */ - pas = (__be64 *)get_zeroed_page(GFP_ATOMIC); - if (!pas) { - mlx5_ib_warn(dev, "unable to allocate memory during MTT update, falling back to slower chunked mechanism.\n"); - pas = mlx5_ib_update_mtt_emergency_buffer; - size = MLX5_UMR_MTT_MIN_CHUNK_SIZE; - use_emergency_buf = 1; - mutex_lock(&mlx5_ib_update_mtt_emergency_buffer_mutex); - memset(pas, 0, size); + if (!xlt) { + uctx = to_mucontext(mr->ibmr.uobject->context); + mlx5_ib_warn(dev, "Using XLT emergency buffer\n"); + size = PAGE_SIZE; + xlt = (void *)uctx->upd_xlt_page; + mutex_lock(&uctx->upd_xlt_page_mutex); + memset(xlt, 0, size); } - pages_iter = size / sizeof(u64); - dma = dma_map_single(ddev, pas, size, DMA_TO_DEVICE); + pages_iter = size / desc_size; + dma = dma_map_single(ddev, xlt, size, DMA_TO_DEVICE); if (dma_mapping_error(ddev, dma)) { - mlx5_ib_err(dev, "unable to map DMA during MTT update.\n"); + mlx5_ib_err(dev, "unable to map DMA during XLT update.\n"); err = -ENOMEM; - goto free_pas; + goto free_xlt; } + sg.addr = dma; + sg.lkey = dev->umrc.pd->local_dma_lkey; + + memset(&wr, 0, sizeof(wr)); + wr.wr.send_flags = MLX5_IB_SEND_UMR_UPDATE_XLT; + if (!(flags & MLX5_IB_UPD_XLT_ENABLE)) + wr.wr.send_flags |= MLX5_IB_SEND_UMR_FAIL_IF_FREE; + wr.wr.sg_list = &sg; + wr.wr.num_sge = 1; + wr.wr.opcode = MLX5_IB_WR_UMR; + + wr.pd = mr->ibmr.pd; + wr.mkey = mr->mmkey.key; + wr.length = mr->mmkey.size; + wr.virt_addr = mr->mmkey.iova; + wr.access_flags = mr->access_flags; + wr.page_shift = page_shift; + for (pages_mapped = 0; pages_mapped < pages_to_map && !err; - pages_mapped += pages_iter, start_page_index += pages_iter) { + pages_mapped += pages_iter, idx += pages_iter) { dma_sync_single_for_cpu(ddev, dma, size, DMA_TO_DEVICE); - - npages = min_t(size_t, - pages_iter, - ib_umem_num_pages(umem) - start_page_index); - - if (!zap) { - __mlx5_ib_populate_pas(dev, umem, PAGE_SHIFT, - start_page_index, npages, pas, - MLX5_IB_MTT_PRESENT); - /* Clear padding after the pages brought from the - * umem. */ - memset(pas + npages, 0, size - npages * sizeof(u64)); - } + npages = populate_xlt(mr, idx, pages_iter, xlt, + page_shift, size, flags); dma_sync_single_for_device(ddev, dma, size, DMA_TO_DEVICE); - mlx5_ib_init_umr_context(&umr_context); - - memset(&wr, 0, sizeof(wr)); - wr.wr.wr_cqe = &umr_context.cqe; - - sg.addr = dma; - sg.length = ALIGN(npages * sizeof(u64), - MLX5_UMR_MTT_ALIGNMENT); - sg.lkey = dev->umrc.pd->local_dma_lkey; + sg.length = ALIGN(npages * desc_size, + MLX5_UMR_MTT_ALIGNMENT); + + if (pages_mapped + pages_iter >= pages_to_map) { + if (flags & MLX5_IB_UPD_XLT_ENABLE) + wr.wr.send_flags |= + MLX5_IB_SEND_UMR_ENABLE_MR | + MLX5_IB_SEND_UMR_UPDATE_PD_ACCESS | + MLX5_IB_SEND_UMR_UPDATE_TRANSLATION; + if (flags & MLX5_IB_UPD_XLT_PD || + flags & MLX5_IB_UPD_XLT_ACCESS) + wr.wr.send_flags |= + MLX5_IB_SEND_UMR_UPDATE_PD_ACCESS; + if (flags & MLX5_IB_UPD_XLT_ADDR) + wr.wr.send_flags |= + MLX5_IB_SEND_UMR_UPDATE_TRANSLATION; + } - wr.wr.send_flags = MLX5_IB_SEND_UMR_FAIL_IF_FREE | - MLX5_IB_SEND_UMR_UPDATE_MTT; - wr.wr.sg_list = &sg; - wr.wr.num_sge = 1; - wr.wr.opcode = MLX5_IB_WR_UMR; - wr.npages = sg.length / sizeof(u64); - wr.page_shift = PAGE_SHIFT; - wr.mkey = mr->mmkey.key; - wr.target.offset = start_page_index; + wr.offset = idx * desc_size; + wr.xlt_size = sg.length; - down(&umrc->sem); - err = ib_post_send(umrc->qp, &wr.wr, &bad); - if (err) { - mlx5_ib_err(dev, "UMR post send failed, err %d\n", err); - } else { - wait_for_completion(&umr_context.done); - if (umr_context.status != IB_WC_SUCCESS) { - mlx5_ib_err(dev, "UMR completion failed, code %d\n", - umr_context.status); - err = -EFAULT; - } - } - up(&umrc->sem); + err = mlx5_ib_post_send_wait(dev, &wr); } dma_unmap_single(ddev, dma, size, DMA_TO_DEVICE); -free_pas: - if (!use_emergency_buf) - free_page((unsigned long)pas); +free_xlt: + if (uctx) + mutex_unlock(&uctx->upd_xlt_page_mutex); else - mutex_unlock(&mlx5_ib_update_mtt_emergency_buffer_mutex); + free_pages((unsigned long)xlt, get_order(size)); return err; } -#endif /* * If ibmr is NULL it will be allocated by reg_create. @@ -1088,8 +1115,9 @@ static struct mlx5_ib_mr *reg_create(struct ib_mr *ibmr, struct ib_pd *pd, goto err_1; } pas = (__be64 *)MLX5_ADDR_OF(create_mkey_in, in, klm_pas_mtt); - mlx5_ib_populate_pas(dev, umem, page_shift, pas, - pg_cap ? MLX5_IB_MTT_PRESENT : 0); + if (!(access_flags & IB_ACCESS_ON_DEMAND)) + mlx5_ib_populate_pas(dev, umem, page_shift, pas, + pg_cap ? MLX5_IB_MTT_PRESENT : 0); /* The pg_access bit allows setting the access flags * in the page list submitted with the command. */ @@ -1119,6 +1147,8 @@ static struct mlx5_ib_mr *reg_create(struct ib_mr *ibmr, struct ib_pd *pd, mlx5_ib_warn(dev, "create mkey failed\n"); goto err_2; } + mr->mmkey.type = MLX5_MKEY_MR; + mr->desc_size = sizeof(struct mlx5_mtt); mr->umem = umem; mr->dev = dev; mr->live = 1; @@ -1164,20 +1194,33 @@ struct ib_mr *mlx5_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, mlx5_ib_dbg(dev, "start 0x%llx, virt_addr 0x%llx, length 0x%llx, access_flags 0x%x\n", start, virt_addr, length, access_flags); - umem = mr_umem_get(pd, start, length, access_flags, &npages, + +#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING + if (!start && length == U64_MAX) { + if (!(access_flags & IB_ACCESS_ON_DEMAND) || + !(dev->odp_caps.general_caps & IB_ODP_SUPPORT_IMPLICIT)) + return ERR_PTR(-EINVAL); + + mr = mlx5_ib_alloc_implicit_mr(to_mpd(pd), access_flags); + return &mr->ibmr; + } +#endif + + err = mr_umem_get(pd, start, length, access_flags, &umem, &npages, &page_shift, &ncont, &order); - if (IS_ERR(umem)) - return (void *)umem; + if (err < 0) + return ERR_PTR(err); - if (use_umr(order)) { + if (use_umr(dev, order)) { mr = reg_umr(pd, umem, virt_addr, length, ncont, page_shift, order, access_flags); if (PTR_ERR(mr) == -EAGAIN) { mlx5_ib_dbg(dev, "cache empty for order %d", order); mr = NULL; } - } else if (access_flags & IB_ACCESS_ON_DEMAND) { + } else if (access_flags & IB_ACCESS_ON_DEMAND && + !MLX5_CAP_GEN(dev->mdev, umr_extended_translation_offset)) { err = -EINVAL; pr_err("Got MR registration for ODP MR > 512MB, not supported for Connect-IB"); goto error; @@ -1214,106 +1257,39 @@ error: static int unreg_umr(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr) { struct mlx5_core_dev *mdev = dev->mdev; - struct umr_common *umrc = &dev->umrc; - struct mlx5_ib_umr_context umr_context; struct mlx5_umr_wr umrwr = {}; - struct ib_send_wr *bad; - int err; if (mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR) return 0; - mlx5_ib_init_umr_context(&umr_context); - - umrwr.wr.wr_cqe = &umr_context.cqe; - prep_umr_unreg_wqe(dev, &umrwr.wr, mr->mmkey.key); - - down(&umrc->sem); - err = ib_post_send(umrc->qp, &umrwr.wr, &bad); - if (err) { - up(&umrc->sem); - mlx5_ib_dbg(dev, "err %d\n", err); - goto error; - } else { - wait_for_completion(&umr_context.done); - up(&umrc->sem); - } - if (umr_context.status != IB_WC_SUCCESS) { - mlx5_ib_warn(dev, "unreg umr failed\n"); - err = -EFAULT; - goto error; - } - return 0; + umrwr.wr.send_flags = MLX5_IB_SEND_UMR_DISABLE_MR | + MLX5_IB_SEND_UMR_FAIL_IF_FREE; + umrwr.wr.opcode = MLX5_IB_WR_UMR; + umrwr.mkey = mr->mmkey.key; -error: - return err; + return mlx5_ib_post_send_wait(dev, &umrwr); } -static int rereg_umr(struct ib_pd *pd, struct mlx5_ib_mr *mr, u64 virt_addr, - u64 length, int npages, int page_shift, int order, +static int rereg_umr(struct ib_pd *pd, struct mlx5_ib_mr *mr, int access_flags, int flags) { struct mlx5_ib_dev *dev = to_mdev(pd->device); - struct device *ddev = dev->ib_dev.dma_device; - struct mlx5_ib_umr_context umr_context; - struct ib_send_wr *bad; struct mlx5_umr_wr umrwr = {}; - struct ib_sge sg; - struct umr_common *umrc = &dev->umrc; - dma_addr_t dma = 0; - __be64 *mr_pas = NULL; - int size; int err; - mlx5_ib_init_umr_context(&umr_context); - - umrwr.wr.wr_cqe = &umr_context.cqe; umrwr.wr.send_flags = MLX5_IB_SEND_UMR_FAIL_IF_FREE; - if (flags & IB_MR_REREG_TRANS) { - err = dma_map_mr_pas(dev, mr->umem, npages, page_shift, &size, - &mr_pas, &dma); - if (err) - return err; - - umrwr.target.virt_addr = virt_addr; - umrwr.length = length; - umrwr.wr.send_flags |= MLX5_IB_SEND_UMR_UPDATE_TRANSLATION; - } - - prep_umr_wqe_common(pd, &umrwr.wr, &sg, dma, npages, mr->mmkey.key, - page_shift); + umrwr.wr.opcode = MLX5_IB_WR_UMR; + umrwr.mkey = mr->mmkey.key; - if (flags & IB_MR_REREG_PD) { + if (flags & IB_MR_REREG_PD || flags & IB_MR_REREG_ACCESS) { umrwr.pd = pd; - umrwr.wr.send_flags |= MLX5_IB_SEND_UMR_UPDATE_PD; - } - - if (flags & IB_MR_REREG_ACCESS) { umrwr.access_flags = access_flags; - umrwr.wr.send_flags |= MLX5_IB_SEND_UMR_UPDATE_ACCESS; + umrwr.wr.send_flags |= MLX5_IB_SEND_UMR_UPDATE_PD_ACCESS; } - /* post send request to UMR QP */ - down(&umrc->sem); - err = ib_post_send(umrc->qp, &umrwr.wr, &bad); + err = mlx5_ib_post_send_wait(dev, &umrwr); - if (err) { - mlx5_ib_warn(dev, "post send failed, err %d\n", err); - } else { - wait_for_completion(&umr_context.done); - if (umr_context.status != IB_WC_SUCCESS) { - mlx5_ib_warn(dev, "reg umr failed (%u)\n", - umr_context.status); - err = -EFAULT; - } - } - - up(&umrc->sem); - if (flags & IB_MR_REREG_TRANS) { - dma_unmap_single(ddev, dma, size, DMA_TO_DEVICE); - kfree(mr_pas); - } return err; } @@ -1330,6 +1306,7 @@ int mlx5_ib_rereg_user_mr(struct ib_mr *ib_mr, int flags, u64 start, u64 addr = (flags & IB_MR_REREG_TRANS) ? virt_addr : mr->umem->address; u64 len = (flags & IB_MR_REREG_TRANS) ? length : mr->umem->length; int page_shift = 0; + int upd_flags = 0; int npages = 0; int ncont = 0; int order = 0; @@ -1338,6 +1315,8 @@ int mlx5_ib_rereg_user_mr(struct ib_mr *ib_mr, int flags, u64 start, mlx5_ib_dbg(dev, "start 0x%llx, virt_addr 0x%llx, length 0x%llx, access_flags 0x%x\n", start, virt_addr, length, access_flags); + atomic_sub(mr->npages, &dev->mdev->priv.reg_pages); + if (flags != IB_MR_REREG_PD) { /* * Replace umem. This needs to be done whether or not UMR is @@ -1345,11 +1324,10 @@ int mlx5_ib_rereg_user_mr(struct ib_mr *ib_mr, int flags, u64 start, */ flags |= IB_MR_REREG_TRANS; ib_umem_release(mr->umem); - mr->umem = mr_umem_get(pd, addr, len, access_flags, &npages, - &page_shift, &ncont, &order); - if (IS_ERR(mr->umem)) { - err = PTR_ERR(mr->umem); - mr->umem = NULL; + err = mr_umem_get(pd, addr, len, access_flags, &mr->umem, + &npages, &page_shift, &ncont, &order); + if (err < 0) { + clean_mr(mr); return err; } } @@ -1381,32 +1359,37 @@ int mlx5_ib_rereg_user_mr(struct ib_mr *ib_mr, int flags, u64 start, /* * Send a UMR WQE */ - err = rereg_umr(pd, mr, addr, len, npages, page_shift, - order, access_flags, flags); + mr->ibmr.pd = pd; + mr->access_flags = access_flags; + mr->mmkey.iova = addr; + mr->mmkey.size = len; + mr->mmkey.pd = to_mpd(pd)->pdn; + + if (flags & IB_MR_REREG_TRANS) { + upd_flags = MLX5_IB_UPD_XLT_ADDR; + if (flags & IB_MR_REREG_PD) + upd_flags |= MLX5_IB_UPD_XLT_PD; + if (flags & IB_MR_REREG_ACCESS) + upd_flags |= MLX5_IB_UPD_XLT_ACCESS; + err = mlx5_ib_update_xlt(mr, 0, npages, page_shift, + upd_flags); + } else { + err = rereg_umr(pd, mr, access_flags, flags); + } + if (err) { mlx5_ib_warn(dev, "Failed to rereg UMR\n"); + ib_umem_release(mr->umem); + clean_mr(mr); return err; } } - if (flags & IB_MR_REREG_PD) { - ib_mr->pd = pd; - mr->mmkey.pd = to_mpd(pd)->pdn; - } - - if (flags & IB_MR_REREG_ACCESS) - mr->access_flags = access_flags; + set_mr_fileds(dev, mr, npages, len, access_flags); - if (flags & IB_MR_REREG_TRANS) { - atomic_sub(mr->npages, &dev->mdev->priv.reg_pages); - set_mr_fileds(dev, mr, npages, len, access_flags); - mr->mmkey.iova = addr; - mr->mmkey.size = len; - } #ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING update_odp_mr(mr); #endif - return 0; } @@ -1428,9 +1411,9 @@ mlx5_alloc_priv_descs(struct ib_device *device, mr->descs = PTR_ALIGN(mr->descs_alloc, MLX5_UMR_ALIGN); - mr->desc_map = dma_map_single(device->dma_device, mr->descs, + mr->desc_map = dma_map_single(device->dev.parent, mr->descs, size, DMA_TO_DEVICE); - if (dma_mapping_error(device->dma_device, mr->desc_map)) { + if (dma_mapping_error(device->dev.parent, mr->desc_map)) { ret = -ENOMEM; goto err; } @@ -1449,7 +1432,7 @@ mlx5_free_priv_descs(struct mlx5_ib_mr *mr) struct ib_device *device = mr->ibmr.device; int size = mr->max_descs * mr->desc_size; - dma_unmap_single(device->dma_device, mr->desc_map, + dma_unmap_single(device->dev.parent, mr->desc_map, size, DMA_TO_DEVICE); kfree(mr->descs_alloc); mr->descs = NULL; @@ -1485,12 +1468,7 @@ static int clean_mr(struct mlx5_ib_mr *mr) return err; } } else { - err = unreg_umr(dev, mr); - if (err) { - mlx5_ib_warn(dev, "failed unregister\n"); - return err; - } - free_cached_mr(dev, mr); + mlx5_mr_cache_free(dev, mr); } if (!umred) @@ -1513,8 +1491,11 @@ int mlx5_ib_dereg_mr(struct ib_mr *ibmr) /* Wait for all running page-fault handlers to finish. */ synchronize_srcu(&dev->mr_srcu); /* Destroy all page mappings */ - mlx5_ib_invalidate_range(umem, ib_umem_start(umem), - ib_umem_end(umem)); + if (umem->odp_data->page_list) + mlx5_ib_invalidate_range(umem, ib_umem_start(umem), + ib_umem_end(umem)); + else + mlx5_ib_free_implicit_mr(mr); /* * We kill the umem before the MR for ODP, * so that there will not be any invalidations in @@ -1570,11 +1551,11 @@ struct ib_mr *mlx5_ib_alloc_mr(struct ib_pd *pd, mr->access_mode = MLX5_MKC_ACCESS_MODE_MTT; MLX5_SET(mkc, mkc, log_page_size, PAGE_SHIFT); err = mlx5_alloc_priv_descs(pd->device, mr, - ndescs, sizeof(u64)); + ndescs, sizeof(struct mlx5_mtt)); if (err) goto err_free_in; - mr->desc_size = sizeof(u64); + mr->desc_size = sizeof(struct mlx5_mtt); mr->max_descs = ndescs; } else if (mr_type == IB_MR_TYPE_SG_GAPS) { mr->access_mode = MLX5_MKC_ACCESS_MODE_KLMS; @@ -1623,6 +1604,7 @@ struct ib_mr *mlx5_ib_alloc_mr(struct ib_pd *pd, if (err) goto err_destroy_psv; + mr->mmkey.type = MLX5_MKEY_MR; mr->ibmr.lkey = mr->mmkey.key; mr->ibmr.rkey = mr->mmkey.key; mr->umem = NULL; @@ -1703,6 +1685,7 @@ struct ib_mw *mlx5_ib_alloc_mw(struct ib_pd *pd, enum ib_mw_type type, if (err) goto free; + mw->mmkey.type = MLX5_MKEY_MW; mw->ibmw.rkey = mw->mmkey.key; resp.response_length = min(offsetof(typeof(resp), response_length) + diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c index cacb631a7b0a..d7b12f0750e2 100644 --- a/drivers/infiniband/hw/mlx5/odp.c +++ b/drivers/infiniband/hw/mlx5/odp.c @@ -34,6 +34,7 @@ #include <rdma/ib_umem_odp.h> #include "mlx5_ib.h" +#include "cmd.h" #define MAX_PREFETCH_LEN (4*1024*1024U) @@ -41,13 +42,146 @@ * a pagefault. */ #define MMU_NOTIFIER_TIMEOUT 1000 -struct workqueue_struct *mlx5_ib_page_fault_wq; +#define MLX5_IMR_MTT_BITS (30 - PAGE_SHIFT) +#define MLX5_IMR_MTT_SHIFT (MLX5_IMR_MTT_BITS + PAGE_SHIFT) +#define MLX5_IMR_MTT_ENTRIES BIT_ULL(MLX5_IMR_MTT_BITS) +#define MLX5_IMR_MTT_SIZE BIT_ULL(MLX5_IMR_MTT_SHIFT) +#define MLX5_IMR_MTT_MASK (~(MLX5_IMR_MTT_SIZE - 1)) + +#define MLX5_KSM_PAGE_SHIFT MLX5_IMR_MTT_SHIFT + +static u64 mlx5_imr_ksm_entries; + +static int check_parent(struct ib_umem_odp *odp, + struct mlx5_ib_mr *parent) +{ + struct mlx5_ib_mr *mr = odp->private; + + return mr && mr->parent == parent; +} + +static struct ib_umem_odp *odp_next(struct ib_umem_odp *odp) +{ + struct mlx5_ib_mr *mr = odp->private, *parent = mr->parent; + struct ib_ucontext *ctx = odp->umem->context; + struct rb_node *rb; + + down_read(&ctx->umem_rwsem); + while (1) { + rb = rb_next(&odp->interval_tree.rb); + if (!rb) + goto not_found; + odp = rb_entry(rb, struct ib_umem_odp, interval_tree.rb); + if (check_parent(odp, parent)) + goto end; + } +not_found: + odp = NULL; +end: + up_read(&ctx->umem_rwsem); + return odp; +} + +static struct ib_umem_odp *odp_lookup(struct ib_ucontext *ctx, + u64 start, u64 length, + struct mlx5_ib_mr *parent) +{ + struct ib_umem_odp *odp; + struct rb_node *rb; + + down_read(&ctx->umem_rwsem); + odp = rbt_ib_umem_lookup(&ctx->umem_tree, start, length); + if (!odp) + goto end; + + while (1) { + if (check_parent(odp, parent)) + goto end; + rb = rb_next(&odp->interval_tree.rb); + if (!rb) + goto not_found; + odp = rb_entry(rb, struct ib_umem_odp, interval_tree.rb); + if (ib_umem_start(odp->umem) > start + length) + goto not_found; + } +not_found: + odp = NULL; +end: + up_read(&ctx->umem_rwsem); + return odp; +} + +void mlx5_odp_populate_klm(struct mlx5_klm *pklm, size_t offset, + size_t nentries, struct mlx5_ib_mr *mr, int flags) +{ + struct ib_pd *pd = mr->ibmr.pd; + struct ib_ucontext *ctx = pd->uobject->context; + struct mlx5_ib_dev *dev = to_mdev(pd->device); + struct ib_umem_odp *odp; + unsigned long va; + int i; + + if (flags & MLX5_IB_UPD_XLT_ZAP) { + for (i = 0; i < nentries; i++, pklm++) { + pklm->bcount = cpu_to_be32(MLX5_IMR_MTT_SIZE); + pklm->key = cpu_to_be32(dev->null_mkey); + pklm->va = 0; + } + return; + } + + odp = odp_lookup(ctx, offset * MLX5_IMR_MTT_SIZE, + nentries * MLX5_IMR_MTT_SIZE, mr); + + for (i = 0; i < nentries; i++, pklm++) { + pklm->bcount = cpu_to_be32(MLX5_IMR_MTT_SIZE); + va = (offset + i) * MLX5_IMR_MTT_SIZE; + if (odp && odp->umem->address == va) { + struct mlx5_ib_mr *mtt = odp->private; + + pklm->key = cpu_to_be32(mtt->ibmr.lkey); + odp = odp_next(odp); + } else { + pklm->key = cpu_to_be32(dev->null_mkey); + } + mlx5_ib_dbg(dev, "[%d] va %lx key %x\n", + i, va, be32_to_cpu(pklm->key)); + } +} + +static void mr_leaf_free_action(struct work_struct *work) +{ + struct ib_umem_odp *odp = container_of(work, struct ib_umem_odp, work); + int idx = ib_umem_start(odp->umem) >> MLX5_IMR_MTT_SHIFT; + struct mlx5_ib_mr *mr = odp->private, *imr = mr->parent; + + mr->parent = NULL; + synchronize_srcu(&mr->dev->mr_srcu); + + if (!READ_ONCE(odp->dying)) { + mr->parent = imr; + if (atomic_dec_and_test(&imr->num_leaf_free)) + wake_up(&imr->q_leaf_free); + return; + } + + ib_umem_release(odp->umem); + if (imr->live) + mlx5_ib_update_xlt(imr, idx, 1, 0, + MLX5_IB_UPD_XLT_INDIRECT | + MLX5_IB_UPD_XLT_ATOMIC); + mlx5_mr_cache_free(mr->dev, mr); + + if (atomic_dec_and_test(&imr->num_leaf_free)) + wake_up(&imr->q_leaf_free); +} void mlx5_ib_invalidate_range(struct ib_umem *umem, unsigned long start, unsigned long end) { struct mlx5_ib_mr *mr; - const u64 umr_block_mask = (MLX5_UMR_MTT_ALIGNMENT / sizeof(u64)) - 1; + const u64 umr_block_mask = (MLX5_UMR_MTT_ALIGNMENT / + sizeof(struct mlx5_mtt)) - 1; u64 idx = 0, blk_start_idx = 0; int in_block = 0; u64 addr; @@ -90,16 +224,21 @@ void mlx5_ib_invalidate_range(struct ib_umem *umem, unsigned long start, u64 umr_offset = idx & umr_block_mask; if (in_block && umr_offset == 0) { - mlx5_ib_update_mtt(mr, blk_start_idx, - idx - blk_start_idx, 1); + mlx5_ib_update_xlt(mr, blk_start_idx, + idx - blk_start_idx, + PAGE_SHIFT, + MLX5_IB_UPD_XLT_ZAP | + MLX5_IB_UPD_XLT_ATOMIC); in_block = 0; } } } if (in_block) - mlx5_ib_update_mtt(mr, blk_start_idx, idx - blk_start_idx + 1, - 1); - + mlx5_ib_update_xlt(mr, blk_start_idx, + idx - blk_start_idx + 1, + PAGE_SHIFT, + MLX5_IB_UPD_XLT_ZAP | + MLX5_IB_UPD_XLT_ATOMIC); /* * We are now sure that the device will not access the * memory. We can safely unmap it, and mark it as dirty if @@ -107,6 +246,13 @@ void mlx5_ib_invalidate_range(struct ib_umem *umem, unsigned long start, */ ib_umem_odp_unmap_dma_pages(umem, start, end); + + if (unlikely(!umem->npages && mr->parent && + !umem->odp_data->dying)) { + WRITE_ONCE(umem->odp_data->dying, 1); + atomic_inc(&mr->parent->num_leaf_free); + schedule_work(&umem->odp_data->work); + } } void mlx5_ib_internal_fill_odp_caps(struct mlx5_ib_dev *dev) @@ -120,6 +266,11 @@ void mlx5_ib_internal_fill_odp_caps(struct mlx5_ib_dev *dev) caps->general_caps = IB_ODP_SUPPORT; + if (MLX5_CAP_GEN(dev->mdev, umr_extended_translation_offset)) + dev->odp_max_size = U64_MAX; + else + dev->odp_max_size = BIT_ULL(MLX5_MAX_UMR_SHIFT + PAGE_SHIFT); + if (MLX5_CAP_ODP(dev->mdev, ud_odp_caps.send)) caps->per_transport_caps.ud_odp_caps |= IB_ODP_SUPPORT_SEND; @@ -135,6 +286,14 @@ void mlx5_ib_internal_fill_odp_caps(struct mlx5_ib_dev *dev) if (MLX5_CAP_ODP(dev->mdev, rc_odp_caps.read)) caps->per_transport_caps.rc_odp_caps |= IB_ODP_SUPPORT_READ; + if (MLX5_CAP_ODP(dev->mdev, rc_odp_caps.atomic)) + caps->per_transport_caps.rc_odp_caps |= IB_ODP_SUPPORT_ATOMIC; + + if (MLX5_CAP_GEN(dev->mdev, fixed_buffer_size) && + MLX5_CAP_GEN(dev->mdev, null_mkey) && + MLX5_CAP_GEN(dev->mdev, umr_extended_translation_offset)) + caps->general_caps |= IB_ODP_SUPPORT_IMPLICIT; + return; } @@ -143,151 +302,386 @@ static struct mlx5_ib_mr *mlx5_ib_odp_find_mr_lkey(struct mlx5_ib_dev *dev, { u32 base_key = mlx5_base_mkey(key); struct mlx5_core_mkey *mmkey = __mlx5_mr_lookup(dev->mdev, base_key); - struct mlx5_ib_mr *mr = container_of(mmkey, struct mlx5_ib_mr, mmkey); + struct mlx5_ib_mr *mr; + + if (!mmkey || mmkey->key != key || mmkey->type != MLX5_MKEY_MR) + return NULL; - if (!mmkey || mmkey->key != key || !mr->live) + mr = container_of(mmkey, struct mlx5_ib_mr, mmkey); + + if (!mr->live) return NULL; return container_of(mmkey, struct mlx5_ib_mr, mmkey); } -static void mlx5_ib_page_fault_resume(struct mlx5_ib_qp *qp, - struct mlx5_ib_pfault *pfault, +static void mlx5_ib_page_fault_resume(struct mlx5_ib_dev *dev, + struct mlx5_pagefault *pfault, int error) { - struct mlx5_ib_dev *dev = to_mdev(qp->ibqp.pd->device); - u32 qpn = qp->trans_qp.base.mqp.qpn; + int wq_num = pfault->event_subtype == MLX5_PFAULT_SUBTYPE_WQE ? + pfault->wqe.wq_num : pfault->token; int ret = mlx5_core_page_fault_resume(dev->mdev, - qpn, - pfault->mpfault.flags, + pfault->token, + wq_num, + pfault->type, error); if (ret) - pr_err("Failed to resolve the page fault on QP 0x%x\n", qpn); + mlx5_ib_err(dev, "Failed to resolve the page fault on WQ 0x%x\n", + wq_num); +} + +static struct mlx5_ib_mr *implicit_mr_alloc(struct ib_pd *pd, + struct ib_umem *umem, + bool ksm, int access_flags) +{ + struct mlx5_ib_dev *dev = to_mdev(pd->device); + struct mlx5_ib_mr *mr; + int err; + + mr = mlx5_mr_cache_alloc(dev, ksm ? MLX5_IMR_KSM_CACHE_ENTRY : + MLX5_IMR_MTT_CACHE_ENTRY); + + if (IS_ERR(mr)) + return mr; + + mr->ibmr.pd = pd; + + mr->dev = dev; + mr->access_flags = access_flags; + mr->mmkey.iova = 0; + mr->umem = umem; + + if (ksm) { + err = mlx5_ib_update_xlt(mr, 0, + mlx5_imr_ksm_entries, + MLX5_KSM_PAGE_SHIFT, + MLX5_IB_UPD_XLT_INDIRECT | + MLX5_IB_UPD_XLT_ZAP | + MLX5_IB_UPD_XLT_ENABLE); + + } else { + err = mlx5_ib_update_xlt(mr, 0, + MLX5_IMR_MTT_ENTRIES, + PAGE_SHIFT, + MLX5_IB_UPD_XLT_ZAP | + MLX5_IB_UPD_XLT_ENABLE | + MLX5_IB_UPD_XLT_ATOMIC); + } + + if (err) + goto fail; + + mr->ibmr.lkey = mr->mmkey.key; + mr->ibmr.rkey = mr->mmkey.key; + + mr->live = 1; + + mlx5_ib_dbg(dev, "key %x dev %p mr %p\n", + mr->mmkey.key, dev->mdev, mr); + + return mr; + +fail: + mlx5_ib_err(dev, "Failed to register MKEY %d\n", err); + mlx5_mr_cache_free(dev, mr); + + return ERR_PTR(err); +} + +static struct ib_umem_odp *implicit_mr_get_data(struct mlx5_ib_mr *mr, + u64 io_virt, size_t bcnt) +{ + struct ib_ucontext *ctx = mr->ibmr.pd->uobject->context; + struct mlx5_ib_dev *dev = to_mdev(mr->ibmr.pd->device); + struct ib_umem_odp *odp, *result = NULL; + u64 addr = io_virt & MLX5_IMR_MTT_MASK; + int nentries = 0, start_idx = 0, ret; + struct mlx5_ib_mr *mtt; + struct ib_umem *umem; + + mutex_lock(&mr->umem->odp_data->umem_mutex); + odp = odp_lookup(ctx, addr, 1, mr); + + mlx5_ib_dbg(dev, "io_virt:%llx bcnt:%zx addr:%llx odp:%p\n", + io_virt, bcnt, addr, odp); + +next_mr: + if (likely(odp)) { + if (nentries) + nentries++; + } else { + umem = ib_alloc_odp_umem(ctx, addr, MLX5_IMR_MTT_SIZE); + if (IS_ERR(umem)) { + mutex_unlock(&mr->umem->odp_data->umem_mutex); + return ERR_CAST(umem); + } + + mtt = implicit_mr_alloc(mr->ibmr.pd, umem, 0, mr->access_flags); + if (IS_ERR(mtt)) { + mutex_unlock(&mr->umem->odp_data->umem_mutex); + ib_umem_release(umem); + return ERR_CAST(mtt); + } + + odp = umem->odp_data; + odp->private = mtt; + mtt->umem = umem; + mtt->mmkey.iova = addr; + mtt->parent = mr; + INIT_WORK(&odp->work, mr_leaf_free_action); + + if (!nentries) + start_idx = addr >> MLX5_IMR_MTT_SHIFT; + nentries++; + } + + odp->dying = 0; + + /* Return first odp if region not covered by single one */ + if (likely(!result)) + result = odp; + + addr += MLX5_IMR_MTT_SIZE; + if (unlikely(addr < io_virt + bcnt)) { + odp = odp_next(odp); + if (odp && odp->umem->address != addr) + odp = NULL; + goto next_mr; + } + + if (unlikely(nentries)) { + ret = mlx5_ib_update_xlt(mr, start_idx, nentries, 0, + MLX5_IB_UPD_XLT_INDIRECT | + MLX5_IB_UPD_XLT_ATOMIC); + if (ret) { + mlx5_ib_err(dev, "Failed to update PAS\n"); + result = ERR_PTR(ret); + } + } + + mutex_unlock(&mr->umem->odp_data->umem_mutex); + return result; +} + +struct mlx5_ib_mr *mlx5_ib_alloc_implicit_mr(struct mlx5_ib_pd *pd, + int access_flags) +{ + struct ib_ucontext *ctx = pd->ibpd.uobject->context; + struct mlx5_ib_mr *imr; + struct ib_umem *umem; + + umem = ib_umem_get(ctx, 0, 0, IB_ACCESS_ON_DEMAND, 0); + if (IS_ERR(umem)) + return ERR_CAST(umem); + + imr = implicit_mr_alloc(&pd->ibpd, umem, 1, access_flags); + if (IS_ERR(imr)) { + ib_umem_release(umem); + return ERR_CAST(imr); + } + + imr->umem = umem; + init_waitqueue_head(&imr->q_leaf_free); + atomic_set(&imr->num_leaf_free, 0); + + return imr; +} + +static int mr_leaf_free(struct ib_umem *umem, u64 start, + u64 end, void *cookie) +{ + struct mlx5_ib_mr *mr = umem->odp_data->private, *imr = cookie; + + if (mr->parent != imr) + return 0; + + ib_umem_odp_unmap_dma_pages(umem, + ib_umem_start(umem), + ib_umem_end(umem)); + + if (umem->odp_data->dying) + return 0; + + WRITE_ONCE(umem->odp_data->dying, 1); + atomic_inc(&imr->num_leaf_free); + schedule_work(&umem->odp_data->work); + + return 0; +} + +void mlx5_ib_free_implicit_mr(struct mlx5_ib_mr *imr) +{ + struct ib_ucontext *ctx = imr->ibmr.pd->uobject->context; + + down_read(&ctx->umem_rwsem); + rbt_ib_umem_for_each_in_range(&ctx->umem_tree, 0, ULLONG_MAX, + mr_leaf_free, imr); + up_read(&ctx->umem_rwsem); + + wait_event(imr->q_leaf_free, !atomic_read(&imr->num_leaf_free)); } /* - * Handle a single data segment in a page-fault WQE. + * Handle a single data segment in a page-fault WQE or RDMA region. * - * Returns number of pages retrieved on success. The caller will continue to + * Returns number of pages retrieved on success. The caller may continue to * the next data segment. * Can return the following error codes: * -EAGAIN to designate a temporary error. The caller will abort handling the * page fault and resolve it. * -EFAULT when there's an error mapping the requested pages. The caller will - * abort the page fault handling and possibly move the QP to an error state. - * On other errors the QP should also be closed with an error. + * abort the page fault handling. */ -static int pagefault_single_data_segment(struct mlx5_ib_qp *qp, - struct mlx5_ib_pfault *pfault, +static int pagefault_single_data_segment(struct mlx5_ib_dev *dev, u32 key, u64 io_virt, size_t bcnt, + u32 *bytes_committed, u32 *bytes_mapped) { - struct mlx5_ib_dev *mib_dev = to_mdev(qp->ibqp.pd->device); int srcu_key; - unsigned int current_seq; + unsigned int current_seq = 0; u64 start_idx; int npages = 0, ret = 0; struct mlx5_ib_mr *mr; u64 access_mask = ODP_READ_ALLOWED_BIT; + struct ib_umem_odp *odp; + int implicit = 0; + size_t size; - srcu_key = srcu_read_lock(&mib_dev->mr_srcu); - mr = mlx5_ib_odp_find_mr_lkey(mib_dev, key); + srcu_key = srcu_read_lock(&dev->mr_srcu); + mr = mlx5_ib_odp_find_mr_lkey(dev, key); /* * If we didn't find the MR, it means the MR was closed while we were * handling the ODP event. In this case we return -EFAULT so that the * QP will be closed. */ if (!mr || !mr->ibmr.pd) { - pr_err("Failed to find relevant mr for lkey=0x%06x, probably the MR was destroyed\n", - key); + mlx5_ib_dbg(dev, "Failed to find relevant mr for lkey=0x%06x, probably the MR was destroyed\n", + key); ret = -EFAULT; goto srcu_unlock; } if (!mr->umem->odp_data) { - pr_debug("skipping non ODP MR (lkey=0x%06x) in page fault handler.\n", - key); + mlx5_ib_dbg(dev, "skipping non ODP MR (lkey=0x%06x) in page fault handler.\n", + key); if (bytes_mapped) *bytes_mapped += - (bcnt - pfault->mpfault.bytes_committed); + (bcnt - *bytes_committed); goto srcu_unlock; } - if (mr->ibmr.pd != qp->ibqp.pd) { - pr_err("Page-fault with different PDs for QP and MR.\n"); - ret = -EFAULT; - goto srcu_unlock; + + /* + * Avoid branches - this code will perform correctly + * in all iterations (in iteration 2 and above, + * bytes_committed == 0). + */ + io_virt += *bytes_committed; + bcnt -= *bytes_committed; + + if (!mr->umem->odp_data->page_list) { + odp = implicit_mr_get_data(mr, io_virt, bcnt); + + if (IS_ERR(odp)) { + ret = PTR_ERR(odp); + goto srcu_unlock; + } + mr = odp->private; + implicit = 1; + + } else { + odp = mr->umem->odp_data; } - current_seq = ACCESS_ONCE(mr->umem->odp_data->notifiers_seq); +next_mr: + current_seq = READ_ONCE(odp->notifiers_seq); /* * Ensure the sequence number is valid for some time before we call * gup. */ smp_rmb(); - /* - * Avoid branches - this code will perform correctly - * in all iterations (in iteration 2 and above, - * bytes_committed == 0). - */ - io_virt += pfault->mpfault.bytes_committed; - bcnt -= pfault->mpfault.bytes_committed; - + size = min_t(size_t, bcnt, ib_umem_end(odp->umem) - io_virt); start_idx = (io_virt - (mr->mmkey.iova & PAGE_MASK)) >> PAGE_SHIFT; if (mr->umem->writable) access_mask |= ODP_WRITE_ALLOWED_BIT; - npages = ib_umem_odp_map_dma_pages(mr->umem, io_virt, bcnt, - access_mask, current_seq); - if (npages < 0) { - ret = npages; + + ret = ib_umem_odp_map_dma_pages(mr->umem, io_virt, size, + access_mask, current_seq); + + if (ret < 0) goto srcu_unlock; - } - if (npages > 0) { - mutex_lock(&mr->umem->odp_data->umem_mutex); + if (ret > 0) { + int np = ret; + + mutex_lock(&odp->umem_mutex); if (!ib_umem_mmu_notifier_retry(mr->umem, current_seq)) { /* * No need to check whether the MTTs really belong to * this MR, since ib_umem_odp_map_dma_pages already * checks this. */ - ret = mlx5_ib_update_mtt(mr, start_idx, npages, 0); + ret = mlx5_ib_update_xlt(mr, start_idx, np, + PAGE_SHIFT, + MLX5_IB_UPD_XLT_ATOMIC); } else { ret = -EAGAIN; } - mutex_unlock(&mr->umem->odp_data->umem_mutex); + mutex_unlock(&odp->umem_mutex); if (ret < 0) { if (ret != -EAGAIN) - pr_err("Failed to update mkey page tables\n"); + mlx5_ib_err(dev, "Failed to update mkey page tables\n"); goto srcu_unlock; } if (bytes_mapped) { - u32 new_mappings = npages * PAGE_SIZE - + u32 new_mappings = np * PAGE_SIZE - (io_virt - round_down(io_virt, PAGE_SIZE)); - *bytes_mapped += min_t(u32, new_mappings, bcnt); + *bytes_mapped += min_t(u32, new_mappings, size); } + + npages += np; + } + + bcnt -= size; + if (unlikely(bcnt)) { + struct ib_umem_odp *next; + + io_virt += size; + next = odp_next(odp); + if (unlikely(!next || next->umem->address != io_virt)) { + mlx5_ib_dbg(dev, "next implicit leaf removed at 0x%llx. got %p\n", + io_virt, next); + ret = -EAGAIN; + goto srcu_unlock_no_wait; + } + odp = next; + mr = odp->private; + goto next_mr; } srcu_unlock: if (ret == -EAGAIN) { - if (!mr->umem->odp_data->dying) { - struct ib_umem_odp *odp_data = mr->umem->odp_data; + if (implicit || !odp->dying) { unsigned long timeout = msecs_to_jiffies(MMU_NOTIFIER_TIMEOUT); if (!wait_for_completion_timeout( - &odp_data->notifier_completion, + &odp->notifier_completion, timeout)) { - pr_warn("timeout waiting for mmu notifier completion\n"); + mlx5_ib_warn(dev, "timeout waiting for mmu notifier. seq %d against %d\n", + current_seq, odp->notifiers_seq); } } else { /* The MR is being killed, kill the QP as well. */ ret = -EFAULT; } } - srcu_read_unlock(&mib_dev->mr_srcu, srcu_key); - pfault->mpfault.bytes_committed = 0; + +srcu_unlock_no_wait: + srcu_read_unlock(&dev->mr_srcu, srcu_key); + *bytes_committed = 0; return ret ? ret : npages; } @@ -309,8 +703,9 @@ srcu_unlock: * Returns the number of pages loaded if positive, zero for an empty WQE, or a * negative error code. */ -static int pagefault_data_segments(struct mlx5_ib_qp *qp, - struct mlx5_ib_pfault *pfault, void *wqe, +static int pagefault_data_segments(struct mlx5_ib_dev *dev, + struct mlx5_pagefault *pfault, + struct mlx5_ib_qp *qp, void *wqe, void *wqe_end, u32 *bytes_mapped, u32 *total_wqe_bytes, int receive_queue) { @@ -354,22 +749,23 @@ static int pagefault_data_segments(struct mlx5_ib_qp *qp, if (!inline_segment && total_wqe_bytes) { *total_wqe_bytes += bcnt - min_t(size_t, bcnt, - pfault->mpfault.bytes_committed); + pfault->bytes_committed); } /* A zero length data segment designates a length of 2GB. */ if (bcnt == 0) bcnt = 1U << 31; - if (inline_segment || bcnt <= pfault->mpfault.bytes_committed) { - pfault->mpfault.bytes_committed -= + if (inline_segment || bcnt <= pfault->bytes_committed) { + pfault->bytes_committed -= min_t(size_t, bcnt, - pfault->mpfault.bytes_committed); + pfault->bytes_committed); continue; } - ret = pagefault_single_data_segment(qp, pfault, key, io_virt, - bcnt, bytes_mapped); + ret = pagefault_single_data_segment(dev, key, io_virt, bcnt, + &pfault->bytes_committed, + bytes_mapped); if (ret < 0) break; npages += ret; @@ -378,17 +774,29 @@ static int pagefault_data_segments(struct mlx5_ib_qp *qp, return ret < 0 ? ret : npages; } +static const u32 mlx5_ib_odp_opcode_cap[] = { + [MLX5_OPCODE_SEND] = IB_ODP_SUPPORT_SEND, + [MLX5_OPCODE_SEND_IMM] = IB_ODP_SUPPORT_SEND, + [MLX5_OPCODE_SEND_INVAL] = IB_ODP_SUPPORT_SEND, + [MLX5_OPCODE_RDMA_WRITE] = IB_ODP_SUPPORT_WRITE, + [MLX5_OPCODE_RDMA_WRITE_IMM] = IB_ODP_SUPPORT_WRITE, + [MLX5_OPCODE_RDMA_READ] = IB_ODP_SUPPORT_READ, + [MLX5_OPCODE_ATOMIC_CS] = IB_ODP_SUPPORT_ATOMIC, + [MLX5_OPCODE_ATOMIC_FA] = IB_ODP_SUPPORT_ATOMIC, +}; + /* * Parse initiator WQE. Advances the wqe pointer to point at the * scatter-gather list, and set wqe_end to the end of the WQE. */ static int mlx5_ib_mr_initiator_pfault_handler( - struct mlx5_ib_qp *qp, struct mlx5_ib_pfault *pfault, - void **wqe, void **wqe_end, int wqe_length) + struct mlx5_ib_dev *dev, struct mlx5_pagefault *pfault, + struct mlx5_ib_qp *qp, void **wqe, void **wqe_end, int wqe_length) { - struct mlx5_ib_dev *dev = to_mdev(qp->ibqp.pd->device); struct mlx5_wqe_ctrl_seg *ctrl = *wqe; - u16 wqe_index = pfault->mpfault.wqe.wqe_index; + u16 wqe_index = pfault->wqe.wqe_index; + u32 transport_caps; + struct mlx5_base_av *av; unsigned ds, opcode; #if defined(DEBUG) u32 ctrl_wqe_index, ctrl_qpn; @@ -434,53 +842,49 @@ static int mlx5_ib_mr_initiator_pfault_handler( opcode = be32_to_cpu(ctrl->opmod_idx_opcode) & MLX5_WQE_CTRL_OPCODE_MASK; + switch (qp->ibqp.qp_type) { case IB_QPT_RC: - switch (opcode) { - case MLX5_OPCODE_SEND: - case MLX5_OPCODE_SEND_IMM: - case MLX5_OPCODE_SEND_INVAL: - if (!(dev->odp_caps.per_transport_caps.rc_odp_caps & - IB_ODP_SUPPORT_SEND)) - goto invalid_transport_or_opcode; - break; - case MLX5_OPCODE_RDMA_WRITE: - case MLX5_OPCODE_RDMA_WRITE_IMM: - if (!(dev->odp_caps.per_transport_caps.rc_odp_caps & - IB_ODP_SUPPORT_WRITE)) - goto invalid_transport_or_opcode; - *wqe += sizeof(struct mlx5_wqe_raddr_seg); - break; - case MLX5_OPCODE_RDMA_READ: - if (!(dev->odp_caps.per_transport_caps.rc_odp_caps & - IB_ODP_SUPPORT_READ)) - goto invalid_transport_or_opcode; - *wqe += sizeof(struct mlx5_wqe_raddr_seg); - break; - default: - goto invalid_transport_or_opcode; - } + transport_caps = dev->odp_caps.per_transport_caps.rc_odp_caps; break; case IB_QPT_UD: - switch (opcode) { - case MLX5_OPCODE_SEND: - case MLX5_OPCODE_SEND_IMM: - if (!(dev->odp_caps.per_transport_caps.ud_odp_caps & - IB_ODP_SUPPORT_SEND)) - goto invalid_transport_or_opcode; - *wqe += sizeof(struct mlx5_wqe_datagram_seg); - break; - default: - goto invalid_transport_or_opcode; - } + transport_caps = dev->odp_caps.per_transport_caps.ud_odp_caps; break; default: -invalid_transport_or_opcode: - mlx5_ib_err(dev, "ODP fault on QP of an unsupported opcode or transport. transport: 0x%x opcode: 0x%x.\n", - qp->ibqp.qp_type, opcode); + mlx5_ib_err(dev, "ODP fault on QP of an unsupported transport 0x%x\n", + qp->ibqp.qp_type); + return -EFAULT; + } + + if (unlikely(opcode >= sizeof(mlx5_ib_odp_opcode_cap) / + sizeof(mlx5_ib_odp_opcode_cap[0]) || + !(transport_caps & mlx5_ib_odp_opcode_cap[opcode]))) { + mlx5_ib_err(dev, "ODP fault on QP of an unsupported opcode 0x%x\n", + opcode); return -EFAULT; } + if (qp->ibqp.qp_type != IB_QPT_RC) { + av = *wqe; + if (av->dqp_dct & be32_to_cpu(MLX5_WQE_AV_EXT)) + *wqe += sizeof(struct mlx5_av); + else + *wqe += sizeof(struct mlx5_base_av); + } + + switch (opcode) { + case MLX5_OPCODE_RDMA_WRITE: + case MLX5_OPCODE_RDMA_WRITE_IMM: + case MLX5_OPCODE_RDMA_READ: + *wqe += sizeof(struct mlx5_wqe_raddr_seg); + break; + case MLX5_OPCODE_ATOMIC_CS: + case MLX5_OPCODE_ATOMIC_FA: + *wqe += sizeof(struct mlx5_wqe_raddr_seg); + *wqe += sizeof(struct mlx5_wqe_atomic_seg); + break; + } + return 0; } @@ -489,10 +893,9 @@ invalid_transport_or_opcode: * scatter-gather list, and set wqe_end to the end of the WQE. */ static int mlx5_ib_mr_responder_pfault_handler( - struct mlx5_ib_qp *qp, struct mlx5_ib_pfault *pfault, - void **wqe, void **wqe_end, int wqe_length) + struct mlx5_ib_dev *dev, struct mlx5_pagefault *pfault, + struct mlx5_ib_qp *qp, void **wqe, void **wqe_end, int wqe_length) { - struct mlx5_ib_dev *dev = to_mdev(qp->ibqp.pd->device); struct mlx5_ib_wq *wq = &qp->rq; int wqe_size = 1 << wq->wqe_shift; @@ -529,70 +932,83 @@ invalid_transport_or_opcode: return 0; } -static void mlx5_ib_mr_wqe_pfault_handler(struct mlx5_ib_qp *qp, - struct mlx5_ib_pfault *pfault) +static struct mlx5_ib_qp *mlx5_ib_odp_find_qp(struct mlx5_ib_dev *dev, + u32 wq_num) +{ + struct mlx5_core_qp *mqp = __mlx5_qp_lookup(dev->mdev, wq_num); + + if (!mqp) { + mlx5_ib_err(dev, "QPN 0x%6x not found\n", wq_num); + return NULL; + } + + return to_mibqp(mqp); +} + +static void mlx5_ib_mr_wqe_pfault_handler(struct mlx5_ib_dev *dev, + struct mlx5_pagefault *pfault) { - struct mlx5_ib_dev *dev = to_mdev(qp->ibqp.pd->device); int ret; void *wqe, *wqe_end; u32 bytes_mapped, total_wqe_bytes; char *buffer = NULL; - int resume_with_error = 0; - u16 wqe_index = pfault->mpfault.wqe.wqe_index; - int requestor = pfault->mpfault.flags & MLX5_PFAULT_REQUESTOR; - u32 qpn = qp->trans_qp.base.mqp.qpn; + int resume_with_error = 1; + u16 wqe_index = pfault->wqe.wqe_index; + int requestor = pfault->type & MLX5_PFAULT_REQUESTOR; + struct mlx5_ib_qp *qp; buffer = (char *)__get_free_page(GFP_KERNEL); if (!buffer) { mlx5_ib_err(dev, "Error allocating memory for IO page fault handling.\n"); - resume_with_error = 1; goto resolve_page_fault; } + qp = mlx5_ib_odp_find_qp(dev, pfault->wqe.wq_num); + if (!qp) + goto resolve_page_fault; + ret = mlx5_ib_read_user_wqe(qp, requestor, wqe_index, buffer, PAGE_SIZE, &qp->trans_qp.base); if (ret < 0) { - mlx5_ib_err(dev, "Failed reading a WQE following page fault, error=%x, wqe_index=%x, qpn=%x\n", - -ret, wqe_index, qpn); - resume_with_error = 1; + mlx5_ib_err(dev, "Failed reading a WQE following page fault, error=%d, wqe_index=%x, qpn=%x\n", + ret, wqe_index, pfault->token); goto resolve_page_fault; } wqe = buffer; if (requestor) - ret = mlx5_ib_mr_initiator_pfault_handler(qp, pfault, &wqe, + ret = mlx5_ib_mr_initiator_pfault_handler(dev, pfault, qp, &wqe, &wqe_end, ret); else - ret = mlx5_ib_mr_responder_pfault_handler(qp, pfault, &wqe, + ret = mlx5_ib_mr_responder_pfault_handler(dev, pfault, qp, &wqe, &wqe_end, ret); - if (ret < 0) { - resume_with_error = 1; + if (ret < 0) goto resolve_page_fault; - } if (wqe >= wqe_end) { mlx5_ib_err(dev, "ODP fault on invalid WQE.\n"); - resume_with_error = 1; goto resolve_page_fault; } - ret = pagefault_data_segments(qp, pfault, wqe, wqe_end, &bytes_mapped, - &total_wqe_bytes, !requestor); + ret = pagefault_data_segments(dev, pfault, qp, wqe, wqe_end, + &bytes_mapped, &total_wqe_bytes, + !requestor); if (ret == -EAGAIN) { + resume_with_error = 0; goto resolve_page_fault; } else if (ret < 0 || total_wqe_bytes > bytes_mapped) { - mlx5_ib_err(dev, "Error getting user pages for page fault. Error: 0x%x\n", - -ret); - resume_with_error = 1; + if (ret != -ENOENT) + mlx5_ib_err(dev, "PAGE FAULT error: %d. QP 0x%x. type: 0x%x\n", + ret, pfault->wqe.wq_num, pfault->type); goto resolve_page_fault; } + resume_with_error = 0; resolve_page_fault: - mlx5_ib_page_fault_resume(qp, pfault, resume_with_error); - mlx5_ib_dbg(dev, "PAGE FAULT completed. QP 0x%x resume_with_error=%d, flags: 0x%x\n", - qpn, resume_with_error, - pfault->mpfault.flags); - + mlx5_ib_page_fault_resume(dev, pfault, resume_with_error); + mlx5_ib_dbg(dev, "PAGE FAULT completed. QP 0x%x resume_with_error=%d, type: 0x%x\n", + pfault->wqe.wq_num, resume_with_error, + pfault->type); free_page((unsigned long)buffer); } @@ -602,15 +1018,14 @@ static int pages_in_range(u64 address, u32 length) (address & PAGE_MASK)) >> PAGE_SHIFT; } -static void mlx5_ib_mr_rdma_pfault_handler(struct mlx5_ib_qp *qp, - struct mlx5_ib_pfault *pfault) +static void mlx5_ib_mr_rdma_pfault_handler(struct mlx5_ib_dev *dev, + struct mlx5_pagefault *pfault) { - struct mlx5_pagefault *mpfault = &pfault->mpfault; u64 address; u32 length; - u32 prefetch_len = mpfault->bytes_committed; + u32 prefetch_len = pfault->bytes_committed; int prefetch_activated = 0; - u32 rkey = mpfault->rdma.r_key; + u32 rkey = pfault->rdma.r_key; int ret; /* The RDMA responder handler handles the page fault in two parts. @@ -619,38 +1034,40 @@ static void mlx5_ib_mr_rdma_pfault_handler(struct mlx5_ib_qp *qp, * prefetches more pages. The second operation cannot use the pfault * context and therefore uses the dummy_pfault context allocated on * the stack */ - struct mlx5_ib_pfault dummy_pfault = {}; - - dummy_pfault.mpfault.bytes_committed = 0; - - mpfault->rdma.rdma_va += mpfault->bytes_committed; - mpfault->rdma.rdma_op_len -= min(mpfault->bytes_committed, - mpfault->rdma.rdma_op_len); - mpfault->bytes_committed = 0; + pfault->rdma.rdma_va += pfault->bytes_committed; + pfault->rdma.rdma_op_len -= min(pfault->bytes_committed, + pfault->rdma.rdma_op_len); + pfault->bytes_committed = 0; - address = mpfault->rdma.rdma_va; - length = mpfault->rdma.rdma_op_len; + address = pfault->rdma.rdma_va; + length = pfault->rdma.rdma_op_len; /* For some operations, the hardware cannot tell the exact message * length, and in those cases it reports zero. Use prefetch * logic. */ if (length == 0) { prefetch_activated = 1; - length = mpfault->rdma.packet_size; + length = pfault->rdma.packet_size; prefetch_len = min(MAX_PREFETCH_LEN, prefetch_len); } - ret = pagefault_single_data_segment(qp, pfault, rkey, address, length, - NULL); + ret = pagefault_single_data_segment(dev, rkey, address, length, + &pfault->bytes_committed, NULL); if (ret == -EAGAIN) { /* We're racing with an invalidation, don't prefetch */ prefetch_activated = 0; } else if (ret < 0 || pages_in_range(address, length) > ret) { - mlx5_ib_page_fault_resume(qp, pfault, 1); + mlx5_ib_page_fault_resume(dev, pfault, 1); + if (ret != -ENOENT) + mlx5_ib_warn(dev, "PAGE FAULT error %d. QP 0x%x, type: 0x%x\n", + ret, pfault->token, pfault->type); return; } - mlx5_ib_page_fault_resume(qp, pfault, 0); + mlx5_ib_page_fault_resume(dev, pfault, 0); + mlx5_ib_dbg(dev, "PAGE FAULT completed. QP 0x%x, type: 0x%x, prefetch_activated: %d\n", + pfault->token, pfault->type, + prefetch_activated); /* At this point, there might be a new pagefault already arriving in * the eq, switch to the dummy pagefault for the rest of the @@ -658,139 +1075,93 @@ static void mlx5_ib_mr_rdma_pfault_handler(struct mlx5_ib_qp *qp, * work-queue is being fenced. */ if (prefetch_activated) { - ret = pagefault_single_data_segment(qp, &dummy_pfault, rkey, - address, + u32 bytes_committed = 0; + + ret = pagefault_single_data_segment(dev, rkey, address, prefetch_len, - NULL); - if (ret < 0) { - pr_warn("Prefetch failed (ret = %d, prefetch_activated = %d) for QPN %d, address: 0x%.16llx, length = 0x%.16x\n", - ret, prefetch_activated, - qp->ibqp.qp_num, address, prefetch_len); + &bytes_committed, NULL); + if (ret < 0 && ret != -EAGAIN) { + mlx5_ib_warn(dev, "Prefetch failed. ret: %d, QP 0x%x, address: 0x%.16llx, length = 0x%.16x\n", + ret, pfault->token, address, prefetch_len); } } } -void mlx5_ib_mr_pfault_handler(struct mlx5_ib_qp *qp, - struct mlx5_ib_pfault *pfault) +void mlx5_ib_pfault(struct mlx5_core_dev *mdev, void *context, + struct mlx5_pagefault *pfault) { - u8 event_subtype = pfault->mpfault.event_subtype; + struct mlx5_ib_dev *dev = context; + u8 event_subtype = pfault->event_subtype; switch (event_subtype) { case MLX5_PFAULT_SUBTYPE_WQE: - mlx5_ib_mr_wqe_pfault_handler(qp, pfault); + mlx5_ib_mr_wqe_pfault_handler(dev, pfault); break; case MLX5_PFAULT_SUBTYPE_RDMA: - mlx5_ib_mr_rdma_pfault_handler(qp, pfault); + mlx5_ib_mr_rdma_pfault_handler(dev, pfault); break; default: - pr_warn("Invalid page fault event subtype: 0x%x\n", - event_subtype); - mlx5_ib_page_fault_resume(qp, pfault, 1); - break; + mlx5_ib_err(dev, "Invalid page fault event subtype: 0x%x\n", + event_subtype); + mlx5_ib_page_fault_resume(dev, pfault, 1); } } -static void mlx5_ib_qp_pfault_action(struct work_struct *work) +void mlx5_odp_init_mr_cache_entry(struct mlx5_cache_ent *ent) { - struct mlx5_ib_pfault *pfault = container_of(work, - struct mlx5_ib_pfault, - work); - enum mlx5_ib_pagefault_context context = - mlx5_ib_get_pagefault_context(&pfault->mpfault); - struct mlx5_ib_qp *qp = container_of(pfault, struct mlx5_ib_qp, - pagefaults[context]); - mlx5_ib_mr_pfault_handler(qp, pfault); -} - -void mlx5_ib_qp_disable_pagefaults(struct mlx5_ib_qp *qp) -{ - unsigned long flags; - - spin_lock_irqsave(&qp->disable_page_faults_lock, flags); - qp->disable_page_faults = 1; - spin_unlock_irqrestore(&qp->disable_page_faults_lock, flags); - - /* - * Note that at this point, we are guarenteed that no more - * work queue elements will be posted to the work queue with - * the QP we are closing. - */ - flush_workqueue(mlx5_ib_page_fault_wq); -} - -void mlx5_ib_qp_enable_pagefaults(struct mlx5_ib_qp *qp) -{ - unsigned long flags; - - spin_lock_irqsave(&qp->disable_page_faults_lock, flags); - qp->disable_page_faults = 0; - spin_unlock_irqrestore(&qp->disable_page_faults_lock, flags); -} - -static void mlx5_ib_pfault_handler(struct mlx5_core_qp *qp, - struct mlx5_pagefault *pfault) -{ - /* - * Note that we will only get one fault event per QP per context - * (responder/initiator, read/write), until we resolve the page fault - * with the mlx5_ib_page_fault_resume command. Since this function is - * called from within the work element, there is no risk of missing - * events. - */ - struct mlx5_ib_qp *mibqp = to_mibqp(qp); - enum mlx5_ib_pagefault_context context = - mlx5_ib_get_pagefault_context(pfault); - struct mlx5_ib_pfault *qp_pfault = &mibqp->pagefaults[context]; - - qp_pfault->mpfault = *pfault; - - /* No need to stop interrupts here since we are in an interrupt */ - spin_lock(&mibqp->disable_page_faults_lock); - if (!mibqp->disable_page_faults) - queue_work(mlx5_ib_page_fault_wq, &qp_pfault->work); - spin_unlock(&mibqp->disable_page_faults_lock); -} - -void mlx5_ib_odp_create_qp(struct mlx5_ib_qp *qp) -{ - int i; - - qp->disable_page_faults = 1; - spin_lock_init(&qp->disable_page_faults_lock); + if (!(ent->dev->odp_caps.general_caps & IB_ODP_SUPPORT_IMPLICIT)) + return; - qp->trans_qp.base.mqp.pfault_handler = mlx5_ib_pfault_handler; + switch (ent->order - 2) { + case MLX5_IMR_MTT_CACHE_ENTRY: + ent->page = PAGE_SHIFT; + ent->xlt = MLX5_IMR_MTT_ENTRIES * + sizeof(struct mlx5_mtt) / + MLX5_IB_UMR_OCTOWORD; + ent->access_mode = MLX5_MKC_ACCESS_MODE_MTT; + ent->limit = 0; + break; - for (i = 0; i < MLX5_IB_PAGEFAULT_CONTEXTS; ++i) - INIT_WORK(&qp->pagefaults[i].work, mlx5_ib_qp_pfault_action); + case MLX5_IMR_KSM_CACHE_ENTRY: + ent->page = MLX5_KSM_PAGE_SHIFT; + ent->xlt = mlx5_imr_ksm_entries * + sizeof(struct mlx5_klm) / + MLX5_IB_UMR_OCTOWORD; + ent->access_mode = MLX5_MKC_ACCESS_MODE_KSM; + ent->limit = 0; + break; + } } -int mlx5_ib_odp_init_one(struct mlx5_ib_dev *ibdev) +int mlx5_ib_odp_init_one(struct mlx5_ib_dev *dev) { int ret; - ret = init_srcu_struct(&ibdev->mr_srcu); + ret = init_srcu_struct(&dev->mr_srcu); if (ret) return ret; + if (dev->odp_caps.general_caps & IB_ODP_SUPPORT_IMPLICIT) { + ret = mlx5_cmd_null_mkey(dev->mdev, &dev->null_mkey); + if (ret) { + mlx5_ib_err(dev, "Error getting null_mkey %d\n", ret); + return ret; + } + } + return 0; } -void mlx5_ib_odp_remove_one(struct mlx5_ib_dev *ibdev) +void mlx5_ib_odp_remove_one(struct mlx5_ib_dev *dev) { - cleanup_srcu_struct(&ibdev->mr_srcu); + cleanup_srcu_struct(&dev->mr_srcu); } -int __init mlx5_ib_odp_init(void) +int mlx5_ib_odp_init(void) { - mlx5_ib_page_fault_wq = alloc_ordered_workqueue("mlx5_ib_page_faults", - WQ_MEM_RECLAIM); - if (!mlx5_ib_page_fault_wq) - return -ENOMEM; + mlx5_imr_ksm_entries = BIT_ULL(get_order(TASK_SIZE) - + MLX5_IMR_MTT_BITS); return 0; } -void mlx5_ib_odp_cleanup(void) -{ - destroy_workqueue(mlx5_ib_page_fault_wq); -} diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index d1e921816bfe..ad8a2638e339 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -77,12 +77,14 @@ struct mlx5_wqe_eth_pad { enum raw_qp_set_mask_map { MLX5_RAW_QP_MOD_SET_RQ_Q_CTR_ID = 1UL << 0, + MLX5_RAW_QP_RATE_LIMIT = 1UL << 1, }; struct mlx5_modify_raw_qp_param { u16 operation; u32 set_mask; /* raw_qp_set_mask_map */ + u32 rate_limit; u8 rq_q_ctr_id; }; @@ -351,6 +353,29 @@ static int calc_send_wqe(struct ib_qp_init_attr *attr) return ALIGN(max_t(int, inl_size, size), MLX5_SEND_WQE_BB); } +static int get_send_sge(struct ib_qp_init_attr *attr, int wqe_size) +{ + int max_sge; + + if (attr->qp_type == IB_QPT_RC) + max_sge = (min_t(int, wqe_size, 512) - + sizeof(struct mlx5_wqe_ctrl_seg) - + sizeof(struct mlx5_wqe_raddr_seg)) / + sizeof(struct mlx5_wqe_data_seg); + else if (attr->qp_type == IB_QPT_XRC_INI) + max_sge = (min_t(int, wqe_size, 512) - + sizeof(struct mlx5_wqe_ctrl_seg) - + sizeof(struct mlx5_wqe_xrc_seg) - + sizeof(struct mlx5_wqe_raddr_seg)) / + sizeof(struct mlx5_wqe_data_seg); + else + max_sge = (wqe_size - sq_overhead(attr)) / + sizeof(struct mlx5_wqe_data_seg); + + return min_t(int, max_sge, wqe_size - sq_overhead(attr) / + sizeof(struct mlx5_wqe_data_seg)); +} + static int calc_sq_size(struct mlx5_ib_dev *dev, struct ib_qp_init_attr *attr, struct mlx5_ib_qp *qp) { @@ -381,13 +406,18 @@ static int calc_sq_size(struct mlx5_ib_dev *dev, struct ib_qp_init_attr *attr, wq_size = roundup_pow_of_two(attr->cap.max_send_wr * wqe_size); qp->sq.wqe_cnt = wq_size / MLX5_SEND_WQE_BB; if (qp->sq.wqe_cnt > (1 << MLX5_CAP_GEN(dev->mdev, log_max_qp_sz))) { - mlx5_ib_dbg(dev, "wqe count(%d) exceeds limits(%d)\n", + mlx5_ib_dbg(dev, "send queue size (%d * %d / %d -> %d) exceeds limits(%d)\n", + attr->cap.max_send_wr, wqe_size, MLX5_SEND_WQE_BB, qp->sq.wqe_cnt, 1 << MLX5_CAP_GEN(dev->mdev, log_max_qp_sz)); return -ENOMEM; } qp->sq.wqe_shift = ilog2(MLX5_SEND_WQE_BB); - qp->sq.max_gs = attr->cap.max_send_sge; + qp->sq.max_gs = get_send_sge(attr, wqe_size); + if (qp->sq.max_gs < attr->cap.max_send_sge) + return -ENOMEM; + + attr->cap.max_send_sge = qp->sq.max_gs; qp->sq.max_post = wq_size / wqe_size; attr->cap.max_send_wr = qp->sq.max_post; @@ -445,60 +475,53 @@ static int qp_has_rq(struct ib_qp_init_attr *attr) return 1; } -static int first_med_uuar(void) +static int first_med_bfreg(void) { return 1; } -static int next_uuar(int n) -{ - n++; - - while (((n % 4) & 2)) - n++; +enum { + /* this is the first blue flame register in the array of bfregs assigned + * to a processes. Since we do not use it for blue flame but rather + * regular 64 bit doorbells, we do not need a lock for maintaiing + * "odd/even" order + */ + NUM_NON_BLUE_FLAME_BFREGS = 1, +}; - return n; +static int max_bfregs(struct mlx5_ib_dev *dev, struct mlx5_bfreg_info *bfregi) +{ + return get_num_uars(dev, bfregi) * MLX5_NON_FP_BFREGS_PER_UAR; } -static int num_med_uuar(struct mlx5_uuar_info *uuari) +static int num_med_bfreg(struct mlx5_ib_dev *dev, + struct mlx5_bfreg_info *bfregi) { int n; - n = uuari->num_uars * MLX5_NON_FP_BF_REGS_PER_PAGE - - uuari->num_low_latency_uuars - 1; + n = max_bfregs(dev, bfregi) - bfregi->num_low_latency_bfregs - + NUM_NON_BLUE_FLAME_BFREGS; return n >= 0 ? n : 0; } -static int max_uuari(struct mlx5_uuar_info *uuari) -{ - return uuari->num_uars * 4; -} - -static int first_hi_uuar(struct mlx5_uuar_info *uuari) +static int first_hi_bfreg(struct mlx5_ib_dev *dev, + struct mlx5_bfreg_info *bfregi) { int med; - int i; - int t; - med = num_med_uuar(uuari); - for (t = 0, i = first_med_uuar();; i = next_uuar(i)) { - t++; - if (t == med) - return next_uuar(i); - } - - return 0; + med = num_med_bfreg(dev, bfregi); + return ++med; } -static int alloc_high_class_uuar(struct mlx5_uuar_info *uuari) +static int alloc_high_class_bfreg(struct mlx5_ib_dev *dev, + struct mlx5_bfreg_info *bfregi) { int i; - for (i = first_hi_uuar(uuari); i < max_uuari(uuari); i = next_uuar(i)) { - if (!test_bit(i, uuari->bitmap)) { - set_bit(i, uuari->bitmap); - uuari->count[i]++; + for (i = first_hi_bfreg(dev, bfregi); i < max_bfregs(dev, bfregi); i++) { + if (!bfregi->count[i]) { + bfregi->count[i]++; return i; } } @@ -506,87 +529,61 @@ static int alloc_high_class_uuar(struct mlx5_uuar_info *uuari) return -ENOMEM; } -static int alloc_med_class_uuar(struct mlx5_uuar_info *uuari) +static int alloc_med_class_bfreg(struct mlx5_ib_dev *dev, + struct mlx5_bfreg_info *bfregi) { - int minidx = first_med_uuar(); + int minidx = first_med_bfreg(); int i; - for (i = first_med_uuar(); i < first_hi_uuar(uuari); i = next_uuar(i)) { - if (uuari->count[i] < uuari->count[minidx]) + for (i = first_med_bfreg(); i < first_hi_bfreg(dev, bfregi); i++) { + if (bfregi->count[i] < bfregi->count[minidx]) minidx = i; + if (!bfregi->count[minidx]) + break; } - uuari->count[minidx]++; + bfregi->count[minidx]++; return minidx; } -static int alloc_uuar(struct mlx5_uuar_info *uuari, - enum mlx5_ib_latency_class lat) +static int alloc_bfreg(struct mlx5_ib_dev *dev, + struct mlx5_bfreg_info *bfregi, + enum mlx5_ib_latency_class lat) { - int uuarn = -EINVAL; + int bfregn = -EINVAL; - mutex_lock(&uuari->lock); + mutex_lock(&bfregi->lock); switch (lat) { case MLX5_IB_LATENCY_CLASS_LOW: - uuarn = 0; - uuari->count[uuarn]++; + BUILD_BUG_ON(NUM_NON_BLUE_FLAME_BFREGS != 1); + bfregn = 0; + bfregi->count[bfregn]++; break; case MLX5_IB_LATENCY_CLASS_MEDIUM: - if (uuari->ver < 2) - uuarn = -ENOMEM; + if (bfregi->ver < 2) + bfregn = -ENOMEM; else - uuarn = alloc_med_class_uuar(uuari); + bfregn = alloc_med_class_bfreg(dev, bfregi); break; case MLX5_IB_LATENCY_CLASS_HIGH: - if (uuari->ver < 2) - uuarn = -ENOMEM; + if (bfregi->ver < 2) + bfregn = -ENOMEM; else - uuarn = alloc_high_class_uuar(uuari); - break; - - case MLX5_IB_LATENCY_CLASS_FAST_PATH: - uuarn = 2; + bfregn = alloc_high_class_bfreg(dev, bfregi); break; } - mutex_unlock(&uuari->lock); - - return uuarn; -} + mutex_unlock(&bfregi->lock); -static void free_med_class_uuar(struct mlx5_uuar_info *uuari, int uuarn) -{ - clear_bit(uuarn, uuari->bitmap); - --uuari->count[uuarn]; + return bfregn; } -static void free_high_class_uuar(struct mlx5_uuar_info *uuari, int uuarn) +static void free_bfreg(struct mlx5_ib_dev *dev, struct mlx5_bfreg_info *bfregi, int bfregn) { - clear_bit(uuarn, uuari->bitmap); - --uuari->count[uuarn]; -} - -static void free_uuar(struct mlx5_uuar_info *uuari, int uuarn) -{ - int nuuars = uuari->num_uars * MLX5_BF_REGS_PER_PAGE; - int high_uuar = nuuars - uuari->num_low_latency_uuars; - - mutex_lock(&uuari->lock); - if (uuarn == 0) { - --uuari->count[uuarn]; - goto out; - } - - if (uuarn < high_uuar) { - free_med_class_uuar(uuari, uuarn); - goto out; - } - - free_high_class_uuar(uuari, uuarn); - -out: - mutex_unlock(&uuari->lock); + mutex_lock(&bfregi->lock); + bfregi->count[bfregn]--; + mutex_unlock(&bfregi->lock); } static enum mlx5_qp_state to_mlx5_state(enum ib_qp_state state) @@ -627,9 +624,20 @@ static void mlx5_ib_lock_cqs(struct mlx5_ib_cq *send_cq, static void mlx5_ib_unlock_cqs(struct mlx5_ib_cq *send_cq, struct mlx5_ib_cq *recv_cq); -static int uuarn_to_uar_index(struct mlx5_uuar_info *uuari, int uuarn) +static int bfregn_to_uar_index(struct mlx5_ib_dev *dev, + struct mlx5_bfreg_info *bfregi, int bfregn) { - return uuari->uars[uuarn / MLX5_BF_REGS_PER_PAGE].index; + int bfregs_per_sys_page; + int index_of_sys_page; + int offset; + + bfregs_per_sys_page = get_uars_per_sys_page(dev, bfregi->lib_uar_4k) * + MLX5_NON_FP_BFREGS_PER_UAR; + index_of_sys_page = bfregn / bfregs_per_sys_page; + + offset = bfregn % bfregs_per_sys_page / MLX5_NON_FP_BFREGS_PER_UAR; + + return bfregi->sys_pages[index_of_sys_page] + offset; } static int mlx5_ib_umem_get(struct mlx5_ib_dev *dev, @@ -647,7 +655,7 @@ static int mlx5_ib_umem_get(struct mlx5_ib_dev *dev, return PTR_ERR(*umem); } - mlx5_ib_cont_pages(*umem, addr, npages, page_shift, ncont, NULL); + mlx5_ib_cont_pages(*umem, addr, 0, npages, page_shift, ncont, NULL); err = mlx5_ib_get_buf_offset(addr, *page_shift, offset); if (err) { @@ -700,7 +708,7 @@ static int create_user_rq(struct mlx5_ib_dev *dev, struct ib_pd *pd, return err; } - mlx5_ib_cont_pages(rwq->umem, ucmd->buf_addr, &npages, &page_shift, + mlx5_ib_cont_pages(rwq->umem, ucmd->buf_addr, 0, &npages, &page_shift, &ncont, NULL); err = mlx5_ib_get_buf_offset(ucmd->buf_addr, page_shift, &rwq->rq_page_offset); @@ -732,6 +740,13 @@ err_umem: return err; } +static int adjust_bfregn(struct mlx5_ib_dev *dev, + struct mlx5_bfreg_info *bfregi, int bfregn) +{ + return bfregn / MLX5_NON_FP_BFREGS_PER_UAR * MLX5_BFREGS_PER_UAR + + bfregn % MLX5_NON_FP_BFREGS_PER_UAR; +} + static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd, struct mlx5_ib_qp *qp, struct ib_udata *udata, struct ib_qp_init_attr *attr, @@ -746,7 +761,7 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd, int uar_index; int npages; u32 offset = 0; - int uuarn; + int bfregn; int ncont = 0; __be64 *pas; void *qpc; @@ -764,27 +779,27 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd, */ if (qp->flags & MLX5_IB_QP_CROSS_CHANNEL) /* In CROSS_CHANNEL CQ and QP must use the same UAR */ - uuarn = MLX5_CROSS_CHANNEL_UUAR; + bfregn = MLX5_CROSS_CHANNEL_BFREG; else { - uuarn = alloc_uuar(&context->uuari, MLX5_IB_LATENCY_CLASS_HIGH); - if (uuarn < 0) { - mlx5_ib_dbg(dev, "failed to allocate low latency UUAR\n"); + bfregn = alloc_bfreg(dev, &context->bfregi, MLX5_IB_LATENCY_CLASS_HIGH); + if (bfregn < 0) { + mlx5_ib_dbg(dev, "failed to allocate low latency BFREG\n"); mlx5_ib_dbg(dev, "reverting to medium latency\n"); - uuarn = alloc_uuar(&context->uuari, MLX5_IB_LATENCY_CLASS_MEDIUM); - if (uuarn < 0) { - mlx5_ib_dbg(dev, "failed to allocate medium latency UUAR\n"); + bfregn = alloc_bfreg(dev, &context->bfregi, MLX5_IB_LATENCY_CLASS_MEDIUM); + if (bfregn < 0) { + mlx5_ib_dbg(dev, "failed to allocate medium latency BFREG\n"); mlx5_ib_dbg(dev, "reverting to high latency\n"); - uuarn = alloc_uuar(&context->uuari, MLX5_IB_LATENCY_CLASS_LOW); - if (uuarn < 0) { - mlx5_ib_warn(dev, "uuar allocation failed\n"); - return uuarn; + bfregn = alloc_bfreg(dev, &context->bfregi, MLX5_IB_LATENCY_CLASS_LOW); + if (bfregn < 0) { + mlx5_ib_warn(dev, "bfreg allocation failed\n"); + return bfregn; } } } } - uar_index = uuarn_to_uar_index(&context->uuari, uuarn); - mlx5_ib_dbg(dev, "uuarn 0x%x, uar_index 0x%x\n", uuarn, uar_index); + uar_index = bfregn_to_uar_index(dev, &context->bfregi, bfregn); + mlx5_ib_dbg(dev, "bfregn 0x%x, uar_index 0x%x\n", bfregn, uar_index); qp->rq.offset = 0; qp->sq.wqe_shift = ilog2(MLX5_SEND_WQE_BB); @@ -792,7 +807,7 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd, err = set_user_buf_size(dev, qp, &ucmd, base, attr); if (err) - goto err_uuar; + goto err_bfreg; if (ucmd.buf_addr && ubuffer->buf_size) { ubuffer->buf_addr = ucmd.buf_addr; @@ -801,7 +816,7 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd, &ubuffer->umem, &npages, &page_shift, &ncont, &offset); if (err) - goto err_uuar; + goto err_bfreg; } else { ubuffer->umem = NULL; } @@ -824,8 +839,8 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd, MLX5_SET(qpc, qpc, page_offset, offset); MLX5_SET(qpc, qpc, uar_page, uar_index); - resp->uuar_index = uuarn; - qp->uuarn = uuarn; + resp->bfreg_index = adjust_bfregn(dev, &context->bfregi, bfregn); + qp->bfregn = bfregn; err = mlx5_ib_db_map_user(context, ucmd.db_addr, &qp->db); if (err) { @@ -852,13 +867,13 @@ err_umem: if (ubuffer->umem) ib_umem_release(ubuffer->umem); -err_uuar: - free_uuar(&context->uuari, uuarn); +err_bfreg: + free_bfreg(dev, &context->bfregi, bfregn); return err; } -static void destroy_qp_user(struct ib_pd *pd, struct mlx5_ib_qp *qp, - struct mlx5_ib_qp_base *base) +static void destroy_qp_user(struct mlx5_ib_dev *dev, struct ib_pd *pd, + struct mlx5_ib_qp *qp, struct mlx5_ib_qp_base *base) { struct mlx5_ib_ucontext *context; @@ -866,7 +881,7 @@ static void destroy_qp_user(struct ib_pd *pd, struct mlx5_ib_qp *qp, mlx5_ib_db_unmap_user(context, &qp->db); if (base->ubuffer.umem) ib_umem_release(base->ubuffer.umem); - free_uuar(&context->uuari, qp->uuarn); + free_bfreg(dev, &context->bfregi, qp->bfregn); } static int create_kernel_qp(struct mlx5_ib_dev *dev, @@ -875,14 +890,10 @@ static int create_kernel_qp(struct mlx5_ib_dev *dev, u32 **in, int *inlen, struct mlx5_ib_qp_base *base) { - enum mlx5_ib_latency_class lc = MLX5_IB_LATENCY_CLASS_LOW; - struct mlx5_uuar_info *uuari; int uar_index; void *qpc; - int uuarn; int err; - uuari = &dev->mdev->priv.uuari; if (init_attr->create_flags & ~(IB_QP_CREATE_SIGNATURE_EN | IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK | IB_QP_CREATE_IPOIB_UD_LSO | @@ -890,21 +901,20 @@ static int create_kernel_qp(struct mlx5_ib_dev *dev, return -EINVAL; if (init_attr->qp_type == MLX5_IB_QPT_REG_UMR) - lc = MLX5_IB_LATENCY_CLASS_FAST_PATH; - - uuarn = alloc_uuar(uuari, lc); - if (uuarn < 0) { - mlx5_ib_dbg(dev, "\n"); - return -ENOMEM; - } + qp->bf.bfreg = &dev->fp_bfreg; + else + qp->bf.bfreg = &dev->bfreg; - qp->bf = &uuari->bfs[uuarn]; - uar_index = qp->bf->uar->index; + /* We need to divide by two since each register is comprised of + * two buffers of identical size, namely odd and even + */ + qp->bf.buf_size = (1 << MLX5_CAP_GEN(dev->mdev, log_bf_reg_size)) / 2; + uar_index = qp->bf.bfreg->index; err = calc_sq_size(dev, init_attr, qp); if (err < 0) { mlx5_ib_dbg(dev, "err %d\n", err); - goto err_uuar; + return err; } qp->rq.offset = 0; @@ -914,7 +924,7 @@ static int create_kernel_qp(struct mlx5_ib_dev *dev, err = mlx5_buf_alloc(dev->mdev, base->ubuffer.buf_size, &qp->buf); if (err) { mlx5_ib_dbg(dev, "err %d\n", err); - goto err_uuar; + return err; } qp->sq.qend = mlx5_get_send_wqe(qp, qp->sq.wqe_cnt); @@ -964,34 +974,30 @@ static int create_kernel_qp(struct mlx5_ib_dev *dev, return 0; err_wrid: - mlx5_db_free(dev->mdev, &qp->db); kfree(qp->sq.wqe_head); kfree(qp->sq.w_list); kfree(qp->sq.wrid); kfree(qp->sq.wr_data); kfree(qp->rq.wrid); + mlx5_db_free(dev->mdev, &qp->db); err_free: kvfree(*in); err_buf: mlx5_buf_free(dev->mdev, &qp->buf); - -err_uuar: - free_uuar(&dev->mdev->priv.uuari, uuarn); return err; } static void destroy_qp_kernel(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp) { - mlx5_db_free(dev->mdev, &qp->db); kfree(qp->sq.wqe_head); kfree(qp->sq.w_list); kfree(qp->sq.wrid); kfree(qp->sq.wr_data); kfree(qp->rq.wrid); + mlx5_db_free(dev->mdev, &qp->db); mlx5_buf_free(dev->mdev, &qp->buf); - free_uuar(&dev->mdev->priv.uuari, qp->bf->uuarn); } static u32 get_rx_type(struct mlx5_ib_qp *qp, struct ib_qp_init_attr *attr) @@ -1138,7 +1144,8 @@ static int create_raw_packet_qp_rq(struct mlx5_ib_dev *dev, return -ENOMEM; rqc = MLX5_ADDR_OF(create_rq_in, in, ctx); - MLX5_SET(rqc, rqc, vsd, 1); + if (!(rq->flags & MLX5_IB_RQ_CVLAN_STRIPPING)) + MLX5_SET(rqc, rqc, vsd, 1); MLX5_SET(rqc, rqc, mem_rq_type, MLX5_RQC_MEM_RQ_TYPE_MEMORY_RQ_INLINE); MLX5_SET(rqc, rqc, state, MLX5_RQC_STATE_RST); MLX5_SET(rqc, rqc, flush_in_error_en, 1); @@ -1235,6 +1242,8 @@ static int create_raw_packet_qp(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp, if (qp->rq.wqe_cnt) { rq->base.container_mibqp = qp; + if (qp->flags & MLX5_IB_QP_CVLAN_STRIPPING) + rq->flags |= MLX5_IB_RQ_CVLAN_STRIPPING; err = create_raw_packet_qp_rq(dev, rq, in); if (err) goto err_destroy_sq; @@ -1323,7 +1332,7 @@ static int create_rss_raw_qp_tir(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp, if (init_attr->create_flags || init_attr->send_cq) return -EINVAL; - min_resp_len = offsetof(typeof(resp), uuar_index) + sizeof(resp.uuar_index); + min_resp_len = offsetof(typeof(resp), bfreg_index) + sizeof(resp.bfreg_index); if (udata->outlen < min_resp_len) return -EINVAL; @@ -1496,9 +1505,6 @@ static int create_qp_common(struct mlx5_ib_dev *dev, struct ib_pd *pd, &qp->raw_packet_qp.rq.base : &qp->trans_qp.base; - if (init_attr->qp_type != IB_QPT_RAW_PACKET) - mlx5_ib_odp_create_qp(qp); - mutex_init(&qp->mutex); spin_lock_init(&qp->sq.lock); spin_lock_init(&qp->rq.lock); @@ -1559,6 +1565,14 @@ static int create_qp_common(struct mlx5_ib_dev *dev, struct ib_pd *pd, if (init_attr->sq_sig_type == IB_SIGNAL_ALL_WR) qp->sq_signal_bits = MLX5_WQE_CTRL_CQ_UPDATE; + if (init_attr->create_flags & IB_QP_CREATE_CVLAN_STRIPPING) { + if (!(MLX5_CAP_GEN(dev->mdev, eth_net_offloads) && + MLX5_CAP_ETH(dev->mdev, vlan_cap)) || + (init_attr->qp_type != IB_QPT_RAW_PACKET)) + return -EOPNOTSUPP; + qp->flags |= MLX5_IB_QP_CVLAN_STRIPPING; + } + if (pd && pd->uobject) { if (ib_copy_from_udata(&ucmd, udata, sizeof(ucmd))) { mlx5_ib_dbg(dev, "copy failed\n"); @@ -1765,7 +1779,7 @@ static int create_qp_common(struct mlx5_ib_dev *dev, struct ib_pd *pd, err_create: if (qp->create_type == MLX5_QP_USER) - destroy_qp_user(pd, qp, base); + destroy_qp_user(dev, pd, qp, base); else if (qp->create_type == MLX5_QP_KERNEL) destroy_qp_kernel(dev, qp); @@ -1893,7 +1907,6 @@ static void destroy_qp_common(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp) if (qp->state != IB_QPS_RESET) { if (qp->ibqp.qp_type != IB_QPT_RAW_PACKET) { - mlx5_ib_qp_disable_pagefaults(qp); err = mlx5_core_qp_modify(dev->mdev, MLX5_CMD_OP_2RST_QP, 0, NULL, &base->mqp); @@ -1944,7 +1957,7 @@ static void destroy_qp_common(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp) if (qp->create_type == MLX5_QP_KERNEL) destroy_qp_kernel(dev, qp); else if (qp->create_type == MLX5_QP_USER) - destroy_qp_user(&get_pd(qp)->ibpd, qp, base); + destroy_qp_user(dev, &get_pd(qp)->ibpd, qp, base); } static const char *ib_qp_type_str(enum ib_qp_type type) @@ -2199,6 +2212,7 @@ static int mlx5_set_path(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp, { enum rdma_link_layer ll = rdma_port_get_link_layer(&dev->ib_dev, port); int err; + enum ib_gid_type gid_type; if (attr_mask & IB_QP_PKEY_INDEX) path->pkey_index = cpu_to_be16(alt ? attr->alt_pkey_index : @@ -2217,10 +2231,16 @@ static int mlx5_set_path(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp, if (ll == IB_LINK_LAYER_ETHERNET) { if (!(ah->ah_flags & IB_AH_GRH)) return -EINVAL; + err = mlx5_get_roce_gid_type(dev, port, ah->grh.sgid_index, + &gid_type); + if (err) + return err; memcpy(path->rmac, ah->dmac, sizeof(ah->dmac)); path->udp_sport = mlx5_get_roce_udp_sport(dev, port, ah->grh.sgid_index); path->dci_cfi_prio_sl = (ah->sl & 0x7) << 4; + if (gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP) + path->ecn_dscp = (ah->grh.traffic_class >> 2) & 0x3f; } else { path->fl_free_ar = (path_flags & MLX5_PATH_FLAG_FL) ? 0x80 : 0; path->fl_free_ar |= @@ -2423,7 +2443,7 @@ static int modify_raw_packet_qp_rq(struct mlx5_ib_dev *dev, if (raw_qp_param->set_mask & MLX5_RAW_QP_MOD_SET_RQ_Q_CTR_ID) { if (MLX5_CAP_GEN(dev->mdev, modify_rq_counter_set_id)) { MLX5_SET64(modify_rq_in, in, modify_bitmask, - MLX5_MODIFY_RQ_IN_MODIFY_BITMASK_MODIFY_RQ_COUNTER_SET_ID); + MLX5_MODIFY_RQ_IN_MODIFY_BITMASK_RQ_COUNTER_SET_ID); MLX5_SET(rqc, rqc, counter_set_id, raw_qp_param->rq_q_ctr_id); } else pr_info_once("%s: RAW PACKET QP counters are not supported on current FW\n", @@ -2442,8 +2462,14 @@ out: } static int modify_raw_packet_qp_sq(struct mlx5_core_dev *dev, - struct mlx5_ib_sq *sq, int new_state) + struct mlx5_ib_sq *sq, + int new_state, + const struct mlx5_modify_raw_qp_param *raw_qp_param) { + struct mlx5_ib_qp *ibqp = sq->base.container_mibqp; + u32 old_rate = ibqp->rate_limit; + u32 new_rate = old_rate; + u16 rl_index = 0; void *in; void *sqc; int inlen; @@ -2459,10 +2485,44 @@ static int modify_raw_packet_qp_sq(struct mlx5_core_dev *dev, sqc = MLX5_ADDR_OF(modify_sq_in, in, ctx); MLX5_SET(sqc, sqc, state, new_state); + if (raw_qp_param->set_mask & MLX5_RAW_QP_RATE_LIMIT) { + if (new_state != MLX5_SQC_STATE_RDY) + pr_warn("%s: Rate limit can only be changed when SQ is moving to RDY\n", + __func__); + else + new_rate = raw_qp_param->rate_limit; + } + + if (old_rate != new_rate) { + if (new_rate) { + err = mlx5_rl_add_rate(dev, new_rate, &rl_index); + if (err) { + pr_err("Failed configuring rate %u: %d\n", + new_rate, err); + goto out; + } + } + + MLX5_SET64(modify_sq_in, in, modify_bitmask, 1); + MLX5_SET(sqc, sqc, packet_pacing_rate_limit_index, rl_index); + } + err = mlx5_core_modify_sq(dev, sq->base.mqp.qpn, in, inlen); - if (err) + if (err) { + /* Remove new rate from table if failed */ + if (new_rate && + old_rate != new_rate) + mlx5_rl_remove_rate(dev, new_rate); goto out; + } + + /* Only remove the old rate after new rate was set */ + if ((old_rate && + (old_rate != new_rate)) || + (new_state != MLX5_SQC_STATE_RDY)) + mlx5_rl_remove_rate(dev, old_rate); + ibqp->rate_limit = new_rate; sq->state = new_state; out: @@ -2477,6 +2537,8 @@ static int modify_raw_packet_qp(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp, struct mlx5_ib_raw_packet_qp *raw_packet_qp = &qp->raw_packet_qp; struct mlx5_ib_rq *rq = &raw_packet_qp->rq; struct mlx5_ib_sq *sq = &raw_packet_qp->sq; + int modify_rq = !!qp->rq.wqe_cnt; + int modify_sq = !!qp->sq.wqe_cnt; int rq_state; int sq_state; int err; @@ -2494,10 +2556,18 @@ static int modify_raw_packet_qp(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp, rq_state = MLX5_RQC_STATE_RST; sq_state = MLX5_SQC_STATE_RST; break; - case MLX5_CMD_OP_INIT2INIT_QP: - case MLX5_CMD_OP_INIT2RTR_QP: case MLX5_CMD_OP_RTR2RTS_QP: case MLX5_CMD_OP_RTS2RTS_QP: + if (raw_qp_param->set_mask == + MLX5_RAW_QP_RATE_LIMIT) { + modify_rq = 0; + sq_state = sq->state; + } else { + return raw_qp_param->set_mask ? -EINVAL : 0; + } + break; + case MLX5_CMD_OP_INIT2INIT_QP: + case MLX5_CMD_OP_INIT2RTR_QP: if (raw_qp_param->set_mask) return -EINVAL; else @@ -2507,13 +2577,13 @@ static int modify_raw_packet_qp(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp, return -EINVAL; } - if (qp->rq.wqe_cnt) { - err = modify_raw_packet_qp_rq(dev, rq, rq_state, raw_qp_param); + if (modify_rq) { + err = modify_raw_packet_qp_rq(dev, rq, rq_state, raw_qp_param); if (err) return err; } - if (qp->sq.wqe_cnt) { + if (modify_sq) { if (tx_affinity) { err = modify_raw_packet_tx_affinity(dev->mdev, sq, tx_affinity); @@ -2521,7 +2591,7 @@ static int modify_raw_packet_qp(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp, return err; } - return modify_raw_packet_qp_sq(dev->mdev, sq, sq_state); + return modify_raw_packet_qp_sq(dev->mdev, sq, sq_state, raw_qp_param); } return 0; @@ -2577,7 +2647,6 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp, struct mlx5_ib_port *mibport = NULL; enum mlx5_qp_state mlx5_cur, mlx5_new; enum mlx5_qp_optpar optpar; - int sqd_event; int mlx5_st; int err; u16 op; @@ -2724,18 +2793,12 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp, if (qp->rq.wqe_cnt && cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) context->db_rec_addr = cpu_to_be64(qp->db.dma); - if (cur_state == IB_QPS_RTS && new_state == IB_QPS_SQD && - attr_mask & IB_QP_EN_SQD_ASYNC_NOTIFY && attr->en_sqd_async_notify) - sqd_event = 1; - else - sqd_event = 0; - if (cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) { u8 port_num = (attr_mask & IB_QP_PORT ? attr->port_num : qp->port) - 1; mibport = &dev->port[port_num]; context->qp_counter_set_usr_page |= - cpu_to_be32((u32)(mibport->q_cnt_id) << 24); + cpu_to_be32((u32)(mibport->q_cnts.set_id) << 24); } if (!ibqp->uobject && cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) @@ -2750,16 +2813,6 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp, if (mlx5_st < 0) goto out; - /* If moving to a reset or error state, we must disable page faults on - * this QP and flush all current page faults. Otherwise a stale page - * fault may attempt to work on this QP after it is reset and moved - * again to RTS, and may cause the driver and the device to get out of - * sync. */ - if (cur_state != IB_QPS_RESET && cur_state != IB_QPS_ERR && - (new_state == IB_QPS_RESET || new_state == IB_QPS_ERR) && - (qp->ibqp.qp_type != IB_QPT_RAW_PACKET)) - mlx5_ib_qp_disable_pagefaults(qp); - if (mlx5_cur >= MLX5_QP_NUM_STATE || mlx5_new >= MLX5_QP_NUM_STATE || !optab[mlx5_cur][mlx5_new]) goto out; @@ -2773,9 +2826,15 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp, raw_qp_param.operation = op; if (cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) { - raw_qp_param.rq_q_ctr_id = mibport->q_cnt_id; + raw_qp_param.rq_q_ctr_id = mibport->q_cnts.set_id; raw_qp_param.set_mask |= MLX5_RAW_QP_MOD_SET_RQ_Q_CTR_ID; } + + if (attr_mask & IB_QP_RATE_LIMIT) { + raw_qp_param.rate_limit = attr->rate_limit; + raw_qp_param.set_mask |= MLX5_RAW_QP_RATE_LIMIT; + } + err = modify_raw_packet_qp(dev, qp, &raw_qp_param, tx_affinity); } else { err = mlx5_core_qp_modify(dev->mdev, op, optpar, context, @@ -2785,10 +2844,6 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp, if (err) goto out; - if (cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT && - (qp->ibqp.qp_type != IB_QPT_RAW_PACKET)) - mlx5_ib_qp_enable_pagefaults(qp); - qp->state = new_state; if (attr_mask & IB_QP_ACCESS_FLAGS) @@ -2950,20 +3005,20 @@ static void *set_eth_seg(struct mlx5_wqe_eth_seg *eseg, if (wr->opcode == IB_WR_LSO) { struct ib_ud_wr *ud_wr = container_of(wr, struct ib_ud_wr, wr); - int size_of_inl_hdr_start = sizeof(eseg->inline_hdr_start); + int size_of_inl_hdr_start = sizeof(eseg->inline_hdr.start); u64 left, leftlen, copysz; void *pdata = ud_wr->header; left = ud_wr->hlen; eseg->mss = cpu_to_be16(ud_wr->mss); - eseg->inline_hdr_sz = cpu_to_be16(left); + eseg->inline_hdr.sz = cpu_to_be16(left); /* * check if there is space till the end of queue, if yes, * copy all in one shot, otherwise copy till the end of queue, * rollback and than the copy the left */ - leftlen = qend - (void *)eseg->inline_hdr_start; + leftlen = qend - (void *)eseg->inline_hdr.start; copysz = min_t(u64, leftlen, left); memcpy(seg - size_of_inl_hdr_start, pdata, copysz); @@ -3001,9 +3056,10 @@ static void set_data_ptr_seg(struct mlx5_wqe_data_seg *dseg, struct ib_sge *sg) dseg->addr = cpu_to_be64(sg->addr); } -static __be16 get_klm_octo(int npages) +static u64 get_xlt_octo(u64 bytes) { - return cpu_to_be16(ALIGN(npages, 8) / 2); + return ALIGN(bytes, MLX5_IB_UMR_XLT_ALIGNMENT) / + MLX5_IB_UMR_OCTOWORD; } static __be64 frwr_mkey_mask(void) @@ -3048,18 +3104,14 @@ static __be64 sig_mkey_mask(void) } static void set_reg_umr_seg(struct mlx5_wqe_umr_ctrl_seg *umr, - struct mlx5_ib_mr *mr) + struct mlx5_ib_mr *mr) { - int ndescs = mr->ndescs; + int size = mr->ndescs * mr->desc_size; memset(umr, 0, sizeof(*umr)); - if (mr->access_mode == MLX5_MKC_ACCESS_MODE_KLMS) - /* KLMs take twice the size of MTTs */ - ndescs *= 2; - umr->flags = MLX5_UMR_CHECK_NOT_FREE; - umr->klm_octowords = get_klm_octo(ndescs); + umr->xlt_octowords = cpu_to_be16(get_xlt_octo(size)); umr->mkey_mask = frwr_mkey_mask(); } @@ -3067,38 +3119,20 @@ static void set_linv_umr_seg(struct mlx5_wqe_umr_ctrl_seg *umr) { memset(umr, 0, sizeof(*umr)); umr->mkey_mask = cpu_to_be64(MLX5_MKEY_MASK_FREE); - umr->flags = 1 << 7; + umr->flags = MLX5_UMR_INLINE; } -static __be64 get_umr_reg_mr_mask(void) +static __be64 get_umr_enable_mr_mask(void) { u64 result; - result = MLX5_MKEY_MASK_LEN | - MLX5_MKEY_MASK_PAGE_SIZE | - MLX5_MKEY_MASK_START_ADDR | - MLX5_MKEY_MASK_PD | - MLX5_MKEY_MASK_LR | - MLX5_MKEY_MASK_LW | - MLX5_MKEY_MASK_KEY | - MLX5_MKEY_MASK_RR | - MLX5_MKEY_MASK_RW | - MLX5_MKEY_MASK_A | + result = MLX5_MKEY_MASK_KEY | MLX5_MKEY_MASK_FREE; return cpu_to_be64(result); } -static __be64 get_umr_unreg_mr_mask(void) -{ - u64 result; - - result = MLX5_MKEY_MASK_FREE; - - return cpu_to_be64(result); -} - -static __be64 get_umr_update_mtt_mask(void) +static __be64 get_umr_disable_mr_mask(void) { u64 result; @@ -3113,23 +3147,22 @@ static __be64 get_umr_update_translation_mask(void) result = MLX5_MKEY_MASK_LEN | MLX5_MKEY_MASK_PAGE_SIZE | - MLX5_MKEY_MASK_START_ADDR | - MLX5_MKEY_MASK_KEY | - MLX5_MKEY_MASK_FREE; + MLX5_MKEY_MASK_START_ADDR; return cpu_to_be64(result); } -static __be64 get_umr_update_access_mask(void) +static __be64 get_umr_update_access_mask(int atomic) { u64 result; - result = MLX5_MKEY_MASK_LW | + result = MLX5_MKEY_MASK_LR | + MLX5_MKEY_MASK_LW | MLX5_MKEY_MASK_RR | - MLX5_MKEY_MASK_RW | - MLX5_MKEY_MASK_A | - MLX5_MKEY_MASK_KEY | - MLX5_MKEY_MASK_FREE; + MLX5_MKEY_MASK_RW; + + if (atomic) + result |= MLX5_MKEY_MASK_A; return cpu_to_be64(result); } @@ -3138,15 +3171,13 @@ static __be64 get_umr_update_pd_mask(void) { u64 result; - result = MLX5_MKEY_MASK_PD | - MLX5_MKEY_MASK_KEY | - MLX5_MKEY_MASK_FREE; + result = MLX5_MKEY_MASK_PD; return cpu_to_be64(result); } static void set_reg_umr_segment(struct mlx5_wqe_umr_ctrl_seg *umr, - struct ib_send_wr *wr) + struct ib_send_wr *wr, int atomic) { struct mlx5_umr_wr *umrwr = umr_wr(wr); @@ -3157,24 +3188,24 @@ static void set_reg_umr_segment(struct mlx5_wqe_umr_ctrl_seg *umr, else umr->flags = MLX5_UMR_CHECK_NOT_FREE; /* fail if not free */ - if (!(wr->send_flags & MLX5_IB_SEND_UMR_UNREG)) { - umr->klm_octowords = get_klm_octo(umrwr->npages); - if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_MTT) { - umr->mkey_mask = get_umr_update_mtt_mask(); - umr->bsf_octowords = get_klm_octo(umrwr->target.offset); - umr->flags |= MLX5_UMR_TRANSLATION_OFFSET_EN; - } - if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_TRANSLATION) - umr->mkey_mask |= get_umr_update_translation_mask(); - if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_ACCESS) - umr->mkey_mask |= get_umr_update_access_mask(); - if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_PD) - umr->mkey_mask |= get_umr_update_pd_mask(); - if (!umr->mkey_mask) - umr->mkey_mask = get_umr_reg_mr_mask(); - } else { - umr->mkey_mask = get_umr_unreg_mr_mask(); + umr->xlt_octowords = cpu_to_be16(get_xlt_octo(umrwr->xlt_size)); + if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_XLT) { + u64 offset = get_xlt_octo(umrwr->offset); + + umr->xlt_offset = cpu_to_be16(offset & 0xffff); + umr->xlt_offset_47_16 = cpu_to_be32(offset >> 16); + umr->flags |= MLX5_UMR_TRANSLATION_OFFSET_EN; } + if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_TRANSLATION) + umr->mkey_mask |= get_umr_update_translation_mask(); + if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_PD_ACCESS) { + umr->mkey_mask |= get_umr_update_access_mask(atomic); + umr->mkey_mask |= get_umr_update_pd_mask(); + } + if (wr->send_flags & MLX5_IB_SEND_UMR_ENABLE_MR) + umr->mkey_mask |= get_umr_enable_mr_mask(); + if (wr->send_flags & MLX5_IB_SEND_UMR_DISABLE_MR) + umr->mkey_mask |= get_umr_disable_mr_mask(); if (!wr->num_sge) umr->flags |= MLX5_UMR_INLINE; @@ -3222,17 +3253,17 @@ static void set_reg_mkey_segment(struct mlx5_mkey_seg *seg, struct ib_send_wr *w struct mlx5_umr_wr *umrwr = umr_wr(wr); memset(seg, 0, sizeof(*seg)); - if (wr->send_flags & MLX5_IB_SEND_UMR_UNREG) { + if (wr->send_flags & MLX5_IB_SEND_UMR_DISABLE_MR) seg->status = MLX5_MKEY_STATUS_FREE; - return; - } seg->flags = convert_access(umrwr->access_flags); - if (!(wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_MTT)) { - if (umrwr->pd) - seg->flags_pd = cpu_to_be32(to_mpd(umrwr->pd)->pdn); - seg->start_addr = cpu_to_be64(umrwr->target.virt_addr); - } + if (umrwr->pd) + seg->flags_pd = cpu_to_be32(to_mpd(umrwr->pd)->pdn); + if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_TRANSLATION && + !umrwr->length) + seg->flags_pd |= cpu_to_be32(MLX5_MKEY_LEN64); + + seg->start_addr = cpu_to_be64(umrwr->virt_addr); seg->len = cpu_to_be64(umrwr->length); seg->log2_page_size = umrwr->page_shift; seg->qpn_mkey7_0 = cpu_to_be32(0xffffff00 | @@ -3530,7 +3561,7 @@ static int set_sig_data_segment(struct ib_sig_handover_wr *wr, } static void set_sig_mkey_segment(struct mlx5_mkey_seg *seg, - struct ib_sig_handover_wr *wr, u32 nelements, + struct ib_sig_handover_wr *wr, u32 size, u32 length, u32 pdn) { struct ib_mr *sig_mr = wr->sig_mr; @@ -3545,17 +3576,17 @@ static void set_sig_mkey_segment(struct mlx5_mkey_seg *seg, seg->flags_pd = cpu_to_be32(MLX5_MKEY_REMOTE_INVAL | sigerr << 26 | MLX5_MKEY_BSF_EN | pdn); seg->len = cpu_to_be64(length); - seg->xlt_oct_size = cpu_to_be32(be16_to_cpu(get_klm_octo(nelements))); + seg->xlt_oct_size = cpu_to_be32(get_xlt_octo(size)); seg->bsfs_octo_size = cpu_to_be32(MLX5_MKEY_BSF_OCTO_SIZE); } static void set_sig_umr_segment(struct mlx5_wqe_umr_ctrl_seg *umr, - u32 nelements) + u32 size) { memset(umr, 0, sizeof(*umr)); umr->flags = MLX5_FLAGS_INLINE | MLX5_FLAGS_CHECK_FREE; - umr->klm_octowords = get_klm_octo(nelements); + umr->xlt_octowords = cpu_to_be16(get_xlt_octo(size)); umr->bsf_octowords = cpu_to_be16(MLX5_MKEY_BSF_OCTO_SIZE); umr->mkey_mask = sig_mkey_mask(); } @@ -3567,7 +3598,7 @@ static int set_sig_umr_wr(struct ib_send_wr *send_wr, struct mlx5_ib_qp *qp, struct ib_sig_handover_wr *wr = sig_handover_wr(send_wr); struct mlx5_ib_mr *sig_mr = to_mmr(wr->sig_mr); u32 pdn = get_pd(qp)->pdn; - u32 klm_oct_size; + u32 xlt_size; int region_len, ret; if (unlikely(wr->wr.num_sge != 1) || @@ -3589,15 +3620,15 @@ static int set_sig_umr_wr(struct ib_send_wr *send_wr, struct mlx5_ib_qp *qp, * then we use strided block format (3 octowords), * else we use single KLM (1 octoword) **/ - klm_oct_size = wr->prot ? 3 : 1; + xlt_size = wr->prot ? 0x30 : sizeof(struct mlx5_klm); - set_sig_umr_segment(*seg, klm_oct_size); + set_sig_umr_segment(*seg, xlt_size); *seg += sizeof(struct mlx5_wqe_umr_ctrl_seg); *size += sizeof(struct mlx5_wqe_umr_ctrl_seg) / 16; if (unlikely((*seg == qp->sq.qend))) *seg = mlx5_get_send_wqe(qp, 0); - set_sig_mkey_segment(*seg, wr, klm_oct_size, region_len, pdn); + set_sig_mkey_segment(*seg, wr, xlt_size, region_len, pdn); *seg += sizeof(struct mlx5_mkey_seg); *size += sizeof(struct mlx5_mkey_seg) / 16; if (unlikely((*seg == qp->sq.qend))) @@ -3627,8 +3658,9 @@ static int set_psv_wr(struct ib_sig_domain *domain, psv_seg->ref_tag = cpu_to_be32(domain->sig.dif.ref_tag); break; default: - pr_err("Bad signature type given.\n"); - return 1; + pr_err("Bad signature type (%d) is given.\n", + domain->sig_type); + return -EINVAL; } *seg += sizeof(*psv_seg); @@ -3703,24 +3735,6 @@ static void dump_wqe(struct mlx5_ib_qp *qp, int idx, int size_16) } } -static void mlx5_bf_copy(u64 __iomem *dst, u64 *src, - unsigned bytecnt, struct mlx5_ib_qp *qp) -{ - while (bytecnt > 0) { - __iowrite64_copy(dst++, src++, 8); - __iowrite64_copy(dst++, src++, 8); - __iowrite64_copy(dst++, src++, 8); - __iowrite64_copy(dst++, src++, 8); - __iowrite64_copy(dst++, src++, 8); - __iowrite64_copy(dst++, src++, 8); - __iowrite64_copy(dst++, src++, 8); - __iowrite64_copy(dst++, src++, 8); - bytecnt -= 64; - if (unlikely(src == qp->sq.qend)) - src = mlx5_get_send_wqe(qp, 0); - } -} - static u8 get_fence(u8 fence, struct ib_send_wr *wr) { if (unlikely(wr->opcode == IB_WR_LOCAL_INV && @@ -3816,7 +3830,7 @@ int mlx5_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, return mlx5_ib_gsi_post_send(ibqp, wr, bad_wr); qp = to_mqp(ibqp); - bf = qp->bf; + bf = &qp->bf; qend = qp->sq.qend; spin_lock_irqsave(&qp->sq.lock, flags); @@ -3986,6 +4000,12 @@ int mlx5_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, break; case IB_QPT_SMI: + if (unlikely(!mdev->port_caps[qp->port - 1].has_smi)) { + mlx5_ib_warn(dev, "Send SMP MADs is not allowed\n"); + err = -EPERM; + *bad_wr = wr; + goto out; + } case MLX5_IB_QPT_HW_GSI: set_datagram_seg(seg, wr); seg += sizeof(struct mlx5_wqe_datagram_seg); @@ -4024,7 +4044,7 @@ int mlx5_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, } qp->sq.wr_data[idx] = MLX5_IB_WR_UMR; ctrl->imm = cpu_to_be32(umr_wr(wr)->mkey); - set_reg_umr_segment(seg, wr); + set_reg_umr_segment(seg, wr, !!(MLX5_CAP_GEN(mdev, atomic))); seg += sizeof(struct mlx5_wqe_umr_ctrl_seg); size += sizeof(struct mlx5_wqe_umr_ctrl_seg) / 16; if (unlikely((seg == qend))) @@ -4089,28 +4109,13 @@ out: * we hit doorbell */ wmb(); - if (bf->need_lock) - spin_lock(&bf->lock); - else - __acquire(&bf->lock); - - /* TBD enable WC */ - if (0 && nreq == 1 && bf->uuarn && inl && size > 1 && size <= bf->buf_size / 16) { - mlx5_bf_copy(bf->reg + bf->offset, (u64 *)ctrl, ALIGN(size * 16, 64), qp); - /* wc_wmb(); */ - } else { - mlx5_write64((__be32 *)ctrl, bf->regreg + bf->offset, - MLX5_GET_DOORBELL_LOCK(&bf->lock32)); - /* Make sure doorbells don't leak out of SQ spinlock - * and reach the HCA out of order. - */ - mmiowb(); - } + /* currently we support only regular doorbells */ + mlx5_write64((__be32 *)ctrl, bf->bfreg->map + bf->offset, NULL); + /* Make sure doorbells don't leak out of SQ spinlock + * and reach the HCA out of order. + */ + mmiowb(); bf->offset ^= bf->buf_size; - if (bf->need_lock) - spin_unlock(&bf->lock); - else - __release(&bf->lock); } spin_unlock_irqrestore(&qp->sq.lock, flags); @@ -4478,14 +4483,6 @@ int mlx5_ib_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, return mlx5_ib_gsi_query_qp(ibqp, qp_attr, qp_attr_mask, qp_init_attr); -#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING - /* - * Wait for any outstanding page faults, in case the user frees memory - * based upon this query's result. - */ - flush_workqueue(mlx5_ib_page_fault_wq); -#endif - mutex_lock(&qp->mutex); if (qp->ibqp.qp_type == IB_QPT_RAW_PACKET) { @@ -4610,6 +4607,7 @@ static int create_rq(struct mlx5_ib_rwq *rwq, struct ib_pd *pd, struct ib_wq_init_attr *init_attr) { struct mlx5_ib_dev *dev; + int has_net_offloads; __be64 *rq_pas0; void *in; void *rqc; @@ -4641,9 +4639,28 @@ static int create_rq(struct mlx5_ib_rwq *rwq, struct ib_pd *pd, MLX5_SET(wq, wq, log_wq_pg_sz, rwq->log_page_size); MLX5_SET(wq, wq, wq_signature, rwq->wq_sig); MLX5_SET64(wq, wq, dbr_addr, rwq->db.dma); + has_net_offloads = MLX5_CAP_GEN(dev->mdev, eth_net_offloads); + if (init_attr->create_flags & IB_WQ_FLAGS_CVLAN_STRIPPING) { + if (!(has_net_offloads && MLX5_CAP_ETH(dev->mdev, vlan_cap))) { + mlx5_ib_dbg(dev, "VLAN offloads are not supported\n"); + err = -EOPNOTSUPP; + goto out; + } + } else { + MLX5_SET(rqc, rqc, vsd, 1); + } + if (init_attr->create_flags & IB_WQ_FLAGS_SCATTER_FCS) { + if (!(has_net_offloads && MLX5_CAP_ETH(dev->mdev, scatter_fcs))) { + mlx5_ib_dbg(dev, "Scatter FCS is not supported\n"); + err = -EOPNOTSUPP; + goto out; + } + MLX5_SET(rqc, rqc, scatter_fcs, 1); + } rq_pas0 = (__be64 *)MLX5_ADDR_OF(wq, wq, pas); mlx5_ib_populate_pas(dev, rwq->umem, rwq->page_shift, rq_pas0, 0); err = mlx5_core_create_rq_tracked(dev->mdev, in, inlen, &rwq->core_qp); +out: kvfree(in); return err; } @@ -4927,10 +4944,37 @@ int mlx5_ib_modify_wq(struct ib_wq *wq, struct ib_wq_attr *wq_attr, MLX5_SET(modify_rq_in, in, rq_state, curr_wq_state); MLX5_SET(rqc, rqc, state, wq_state); + if (wq_attr_mask & IB_WQ_FLAGS) { + if (wq_attr->flags_mask & IB_WQ_FLAGS_CVLAN_STRIPPING) { + if (!(MLX5_CAP_GEN(dev->mdev, eth_net_offloads) && + MLX5_CAP_ETH(dev->mdev, vlan_cap))) { + mlx5_ib_dbg(dev, "VLAN offloads are not " + "supported\n"); + err = -EOPNOTSUPP; + goto out; + } + MLX5_SET64(modify_rq_in, in, modify_bitmask, + MLX5_MODIFY_RQ_IN_MODIFY_BITMASK_VSD); + MLX5_SET(rqc, rqc, vsd, + (wq_attr->flags & IB_WQ_FLAGS_CVLAN_STRIPPING) ? 0 : 1); + } + } + + if (curr_wq_state == IB_WQS_RESET && wq_state == IB_WQS_RDY) { + if (MLX5_CAP_GEN(dev->mdev, modify_rq_counter_set_id)) { + MLX5_SET64(modify_rq_in, in, modify_bitmask, + MLX5_MODIFY_RQ_IN_MODIFY_BITMASK_RQ_COUNTER_SET_ID); + MLX5_SET(rqc, rqc, counter_set_id, dev->port->q_cnts.set_id); + } else + pr_info_once("%s: Receive WQ counters are not supported on current FW\n", + dev->ib_dev.name); + } + err = mlx5_core_modify_rq(dev->mdev, rwq->core_qp.qpn, in, inlen); - kvfree(in); if (!err) rwq->ibwq.state = (wq_state == MLX5_RQC_STATE_ERR) ? IB_WQS_ERR : wq_state; +out: + kvfree(in); return err; } diff --git a/drivers/infiniband/hw/mlx5/srq.c b/drivers/infiniband/hw/mlx5/srq.c index 3857dbd9c956..7cb145f9a6db 100644 --- a/drivers/infiniband/hw/mlx5/srq.c +++ b/drivers/infiniband/hw/mlx5/srq.c @@ -118,7 +118,7 @@ static int create_srq_user(struct ib_pd *pd, struct mlx5_ib_srq *srq, return err; } - mlx5_ib_cont_pages(srq->umem, ucmd.buf_addr, &npages, + mlx5_ib_cont_pages(srq->umem, ucmd.buf_addr, 0, &npages, &page_shift, &ncont, NULL); err = mlx5_ib_get_buf_offset(ucmd.buf_addr, page_shift, &offset); @@ -165,8 +165,6 @@ static int create_srq_kernel(struct mlx5_ib_dev *dev, struct mlx5_ib_srq *srq, int err; int i; struct mlx5_wqe_srq_next_seg *next; - int page_shift; - int npages; err = mlx5_db_alloc(dev->mdev, &srq->db); if (err) { @@ -179,7 +177,6 @@ static int create_srq_kernel(struct mlx5_ib_dev *dev, struct mlx5_ib_srq *srq, err = -ENOMEM; goto err_db; } - page_shift = srq->buf.page_shift; srq->head = 0; srq->tail = srq->msrq.max - 1; @@ -191,10 +188,8 @@ static int create_srq_kernel(struct mlx5_ib_dev *dev, struct mlx5_ib_srq *srq, cpu_to_be16((i + 1) & (srq->msrq.max - 1)); } - npages = DIV_ROUND_UP(srq->buf.npages, 1 << (page_shift - PAGE_SHIFT)); - mlx5_ib_dbg(dev, "buf_size %d, page_shift %d, npages %d, calc npages %d\n", - buf_size, page_shift, srq->buf.npages, npages); - in->pas = mlx5_vzalloc(sizeof(*in->pas) * npages); + mlx5_ib_dbg(dev, "srq->buf.page_shift = %d\n", srq->buf.page_shift); + in->pas = mlx5_vzalloc(sizeof(*in->pas) * srq->buf.npages); if (!in->pas) { err = -ENOMEM; goto err_buf; @@ -203,14 +198,12 @@ static int create_srq_kernel(struct mlx5_ib_dev *dev, struct mlx5_ib_srq *srq, srq->wrid = kmalloc(srq->msrq.max * sizeof(u64), GFP_KERNEL); if (!srq->wrid) { - mlx5_ib_dbg(dev, "kmalloc failed %lu\n", - (unsigned long)(srq->msrq.max * sizeof(u64))); err = -ENOMEM; goto err_in; } srq->wq_sig = !!srq_signature; - in->log_page_size = page_shift - MLX5_ADAPTER_PAGE_SHIFT; + in->log_page_size = srq->buf.page_shift - MLX5_ADAPTER_PAGE_SHIFT; if (MLX5_CAP_GEN(dev->mdev, cqe_version) == MLX5_CQE_VERSION_V1 && in->type == IB_SRQT_XRC) in->user_index = MLX5_IB_DEFAULT_UIDX; @@ -282,6 +275,7 @@ struct ib_srq *mlx5_ib_create_srq(struct ib_pd *pd, mlx5_ib_dbg(dev, "desc_size 0x%x, req wr 0x%x, srq size 0x%x, max_gs 0x%x, max_avail_gather 0x%x\n", desc_size, init_attr->attr.max_wr, srq->msrq.max, srq->msrq.max_gs, srq->msrq.max_avail_gather); + in.type = init_attr->srq_type; if (pd->uobject) err = create_srq_user(pd, srq, &in, udata, buf_size); @@ -294,7 +288,6 @@ struct ib_srq *mlx5_ib_create_srq(struct ib_pd *pd, goto err_srq; } - in.type = init_attr->srq_type; in.log_size = ilog2(srq->msrq.max); in.wqe_shift = srq->msrq.wqe_shift - 4; if (srq->wq_sig) diff --git a/drivers/infiniband/hw/mthca/mthca_av.c b/drivers/infiniband/hw/mthca/mthca_av.c index bcac294042f5..c9f0f364f484 100644 --- a/drivers/infiniband/hw/mthca/mthca_av.c +++ b/drivers/infiniband/hw/mthca/mthca_av.c @@ -186,8 +186,8 @@ int mthca_create_ah(struct mthca_dev *dev, on_hca_fail: if (ah->type == MTHCA_AH_PCI_POOL) { - ah->av = pci_pool_alloc(dev->av_table.pool, - GFP_ATOMIC, &ah->avdma); + ah->av = pci_pool_zalloc(dev->av_table.pool, + GFP_ATOMIC, &ah->avdma); if (!ah->av) return -ENOMEM; @@ -196,8 +196,6 @@ on_hca_fail: ah->key = pd->ntmr.ibmr.lkey; - memset(av, 0, MTHCA_AV_SIZE); - av->port_pd = cpu_to_be32(pd->pd_num | (ah_attr->port_num << 24)); av->g_slid = ah_attr->src_path_bits; av->dlid = cpu_to_be16(ah_attr->dlid); diff --git a/drivers/infiniband/hw/mthca/mthca_main.c b/drivers/infiniband/hw/mthca/mthca_main.c index ded76c101dde..c309e5c96383 100644 --- a/drivers/infiniband/hw/mthca/mthca_main.c +++ b/drivers/infiniband/hw/mthca/mthca_main.c @@ -851,20 +851,18 @@ err_uar_table_free: static int mthca_enable_msi_x(struct mthca_dev *mdev) { - struct msix_entry entries[3]; int err; - entries[0].entry = 0; - entries[1].entry = 1; - entries[2].entry = 2; - - err = pci_enable_msix_exact(mdev->pdev, entries, ARRAY_SIZE(entries)); - if (err) + err = pci_alloc_irq_vectors(mdev->pdev, 3, 3, PCI_IRQ_MSIX); + if (err < 0) return err; - mdev->eq_table.eq[MTHCA_EQ_COMP ].msi_x_vector = entries[0].vector; - mdev->eq_table.eq[MTHCA_EQ_ASYNC].msi_x_vector = entries[1].vector; - mdev->eq_table.eq[MTHCA_EQ_CMD ].msi_x_vector = entries[2].vector; + mdev->eq_table.eq[MTHCA_EQ_COMP ].msi_x_vector = + pci_irq_vector(mdev->pdev, 0); + mdev->eq_table.eq[MTHCA_EQ_ASYNC].msi_x_vector = + pci_irq_vector(mdev->pdev, 1); + mdev->eq_table.eq[MTHCA_EQ_CMD ].msi_x_vector = + pci_irq_vector(mdev->pdev, 2); return 0; } @@ -1018,7 +1016,7 @@ static int __mthca_init_one(struct pci_dev *pdev, int hca_type) err = mthca_setup_hca(mdev); if (err == -EBUSY && (mdev->mthca_flags & MTHCA_FLAG_MSI_X)) { if (mdev->mthca_flags & MTHCA_FLAG_MSI_X) - pci_disable_msix(pdev); + pci_free_irq_vectors(pdev); mdev->mthca_flags &= ~MTHCA_FLAG_MSI_X; err = mthca_setup_hca(mdev); @@ -1062,7 +1060,7 @@ err_cleanup: err_close: if (mdev->mthca_flags & MTHCA_FLAG_MSI_X) - pci_disable_msix(pdev); + pci_free_irq_vectors(pdev); mthca_close_hca(mdev); @@ -1113,7 +1111,7 @@ static void __mthca_remove_one(struct pci_dev *pdev) mthca_cmd_cleanup(mdev); if (mdev->mthca_flags & MTHCA_FLAG_MSI_X) - pci_disable_msix(pdev); + pci_free_irq_vectors(pdev); ib_dealloc_device(&mdev->ib_dev); pci_release_regions(pdev); diff --git a/drivers/infiniband/hw/mthca/mthca_provider.c b/drivers/infiniband/hw/mthca/mthca_provider.c index 358930a41e36..22d0e6ee5af6 100644 --- a/drivers/infiniband/hw/mthca/mthca_provider.c +++ b/drivers/infiniband/hw/mthca/mthca_provider.c @@ -146,7 +146,7 @@ static int mthca_query_port(struct ib_device *ibdev, if (!in_mad || !out_mad) goto out; - memset(props, 0, sizeof *props); + /* props being zeroed by the caller, avoid zeroing it here */ init_query_mad(in_mad); in_mad->attr_id = IB_SMP_ATTR_PORT_INFO; @@ -212,7 +212,7 @@ static int mthca_modify_port(struct ib_device *ibdev, if (mutex_lock_interruptible(&to_mdev(ibdev)->cap_mask_mutex)) return -ERESTARTSYS; - err = mthca_query_port(ibdev, port, &attr); + err = ib_query_port(ibdev, port, &attr); if (err) goto out; @@ -410,7 +410,9 @@ static int mthca_dealloc_pd(struct ib_pd *pd) } static struct ib_ah *mthca_ah_create(struct ib_pd *pd, - struct ib_ah_attr *ah_attr) + struct ib_ah_attr *ah_attr, + struct ib_udata *udata) + { int err; struct mthca_ah *ah; @@ -1164,13 +1166,14 @@ static int mthca_port_immutable(struct ib_device *ibdev, u8 port_num, struct ib_port_attr attr; int err; - err = mthca_query_port(ibdev, port_num, &attr); + immutable->core_cap_flags = RDMA_CORE_PORT_IBA_IB; + + err = ib_query_port(ibdev, port_num, &attr); if (err) return err; immutable->pkey_tbl_len = attr.pkey_tbl_len; immutable->gid_tbl_len = attr.gid_tbl_len; - immutable->core_cap_flags = RDMA_CORE_PORT_IBA_IB; immutable->max_mad_size = IB_MGMT_MAD_SIZE; return 0; @@ -1221,7 +1224,7 @@ int mthca_register_device(struct mthca_dev *dev) dev->ib_dev.node_type = RDMA_NODE_IB_CA; dev->ib_dev.phys_port_cnt = dev->limits.num_ports; dev->ib_dev.num_comp_vectors = 1; - dev->ib_dev.dma_device = &dev->pdev->dev; + dev->ib_dev.dev.parent = &dev->pdev->dev; dev->ib_dev.query_device = mthca_query_device; dev->ib_dev.query_port = mthca_query_port; dev->ib_dev.modify_device = mthca_modify_device; diff --git a/drivers/infiniband/hw/mthca/mthca_reset.c b/drivers/infiniband/hw/mthca/mthca_reset.c index 6727af27c017..2a6979e4ae1c 100644 --- a/drivers/infiniband/hw/mthca/mthca_reset.c +++ b/drivers/infiniband/hw/mthca/mthca_reset.c @@ -96,8 +96,6 @@ int mthca_reset(struct mthca_dev *mdev) hca_header = kmalloc(256, GFP_KERNEL); if (!hca_header) { err = -ENOMEM; - mthca_err(mdev, "Couldn't allocate memory to save HCA " - "PCI header, aborting.\n"); goto put_dev; } @@ -119,8 +117,6 @@ int mthca_reset(struct mthca_dev *mdev) bridge_header = kmalloc(256, GFP_KERNEL); if (!bridge_header) { err = -ENOMEM; - mthca_err(mdev, "Couldn't allocate memory to save HCA " - "bridge PCI header, aborting.\n"); goto free_hca; } diff --git a/drivers/infiniband/hw/nes/nes.c b/drivers/infiniband/hw/nes/nes.c index 35cbb17bec12..5b9601014f0c 100644 --- a/drivers/infiniband/hw/nes/nes.c +++ b/drivers/infiniband/hw/nes/nes.c @@ -65,7 +65,6 @@ MODULE_DESCRIPTION("NetEffect RNIC Low-level iWARP Driver"); MODULE_LICENSE("Dual BSD/GPL"); MODULE_VERSION(DRV_VERSION); -int max_mtu = 9000; int interrupt_mod_interval = 0; /* Interoperability */ @@ -516,7 +515,6 @@ static int nes_probe(struct pci_dev *pcidev, const struct pci_device_id *ent) /* Allocate hardware structure */ nesdev = kzalloc(sizeof(struct nes_device), GFP_KERNEL); if (!nesdev) { - printk(KERN_ERR PFX "%s: Unable to alloc hardware struct\n", pci_name(pcidev)); ret = -ENOMEM; goto bail2; } diff --git a/drivers/infiniband/hw/nes/nes.h b/drivers/infiniband/hw/nes/nes.h index e7430c9254d3..85acd0843b50 100644 --- a/drivers/infiniband/hw/nes/nes.h +++ b/drivers/infiniband/hw/nes/nes.h @@ -83,6 +83,8 @@ #define NES_FIRST_QPN 64 #define NES_SW_CONTEXT_ALIGN 1024 +#define NES_MAX_MTU 9000 + #define NES_NIC_MAX_NICS 16 #define NES_MAX_ARP_TABLE_SIZE 4096 @@ -169,8 +171,6 @@ do { \ #include "nes_cm.h" #include "nes_mgt.h" -extern int max_mtu; -#define max_frame_len (max_mtu+ETH_HLEN) extern int interrupt_mod_interval; extern int nes_if_count; extern int mpa_version; diff --git a/drivers/infiniband/hw/nes/nes_cm.c b/drivers/infiniband/hw/nes/nes_cm.c index 57db9b332f44..fb983df7c157 100644 --- a/drivers/infiniband/hw/nes/nes_cm.c +++ b/drivers/infiniband/hw/nes/nes_cm.c @@ -135,17 +135,17 @@ static void record_ird_ord(struct nes_cm_node *, u16, u16); /* instance of function pointers for client API */ /* set address of this instance to cm_core->cm_ops at cm_core alloc */ static const struct nes_cm_ops nes_cm_api = { - mini_cm_accelerated, - mini_cm_listen, - mini_cm_del_listen, - mini_cm_connect, - mini_cm_close, - mini_cm_accept, - mini_cm_reject, - mini_cm_recv_pkt, - mini_cm_dealloc_core, - mini_cm_get, - mini_cm_set + .accelerated = mini_cm_accelerated, + .listen = mini_cm_listen, + .stop_listener = mini_cm_del_listen, + .connect = mini_cm_connect, + .close = mini_cm_close, + .accept = mini_cm_accept, + .reject = mini_cm_reject, + .recv_pkt = mini_cm_recv_pkt, + .destroy_cm_core = mini_cm_dealloc_core, + .get = mini_cm_get, + .set = mini_cm_set }; static struct nes_cm_core *g_cm_core; @@ -2282,10 +2282,8 @@ static struct nes_cm_listener *mini_cm_listen(struct nes_cm_core *cm_core, if (!listener) { /* create a CM listen node (1/2 node to compare incoming traffic to) */ listener = kzalloc(sizeof(*listener), GFP_ATOMIC); - if (!listener) { - nes_debug(NES_DBG_CM, "Not creating listener memory allocation failed\n"); + if (!listener) return NULL; - } listener->loc_addr = cm_info->loc_addr; listener->loc_port = cm_info->loc_port; diff --git a/drivers/infiniband/hw/nes/nes_hw.c b/drivers/infiniband/hw/nes/nes_hw.c index a1c6481d8038..19acd13c6cb1 100644 --- a/drivers/infiniband/hw/nes/nes_hw.c +++ b/drivers/infiniband/hw/nes/nes_hw.c @@ -351,9 +351,8 @@ struct nes_adapter *nes_init_adapter(struct nes_device *nesdev, u8 hw_rev) { /* allocate a new adapter struct */ nesadapter = kzalloc(adapter_size, GFP_KERNEL); - if (nesadapter == NULL) { + if (!nesadapter) return NULL; - } nes_debug(NES_DBG_INIT, "Allocating new nesadapter @ %p, size = %u (actual size = %u).\n", nesadapter, (u32)sizeof(struct nes_adapter), adapter_size); @@ -1007,8 +1006,7 @@ int nes_init_cqp(struct nes_device *nesdev) /* Allocate a twice the number of CQP requests as the SQ size */ nesdev->nes_cqp_requests = kzalloc(sizeof(struct nes_cqp_request) * 2 * NES_CQP_SQ_SIZE, GFP_KERNEL); - if (nesdev->nes_cqp_requests == NULL) { - nes_debug(NES_DBG_INIT, "Unable to allocate memory CQP request entries.\n"); + if (!nesdev->nes_cqp_requests) { pci_free_consistent(nesdev->pcidev, nesdev->cqp_mem_size, nesdev->cqp.sq_vbase, nesdev->cqp.sq_pbase); return -ENOMEM; diff --git a/drivers/infiniband/hw/nes/nes_mgt.c b/drivers/infiniband/hw/nes/nes_mgt.c index 416645259b0f..33624f17c347 100644 --- a/drivers/infiniband/hw/nes/nes_mgt.c +++ b/drivers/infiniband/hw/nes/nes_mgt.c @@ -320,8 +320,7 @@ static int get_fpdu_info(struct nes_device *nesdev, struct nes_qp *nesqp, /* Found one */ fpdu_info = kzalloc(sizeof(*fpdu_info), GFP_ATOMIC); - if (fpdu_info == NULL) { - nes_debug(NES_DBG_PAU, "Failed to alloc a fpdu_info.\n"); + if (!fpdu_info) { rc = -ENOMEM; goto out; } @@ -729,8 +728,7 @@ static int nes_change_quad_hash(struct nes_device *nesdev, } qh_chg = kmalloc(sizeof *qh_chg, GFP_ATOMIC); - if (qh_chg == NULL) { - nes_debug(NES_DBG_PAU, "Failed to get a cqp_request.\n"); + if (!qh_chg) { ret = -ENOMEM; goto chg_qh_err; } @@ -880,10 +878,8 @@ int nes_init_mgt_qp(struct nes_device *nesdev, struct net_device *netdev, struct /* Allocate space the all mgt QPs once */ mgtvnic = kzalloc(NES_MGT_QP_COUNT * sizeof(struct nes_vnic_mgt), GFP_KERNEL); - if (mgtvnic == NULL) { - nes_debug(NES_DBG_INIT, "Unable to allocate memory for mgt structure\n"); + if (!mgtvnic) return -ENOMEM; - } /* Allocate fragment, RQ, and CQ; Reuse CEQ based on the PCI function */ /* We are not sending from this NIC so sq is not allocated */ diff --git a/drivers/infiniband/hw/nes/nes_nic.c b/drivers/infiniband/hw/nes/nes_nic.c index 2b27d1351cf7..5921ea3d50ae 100644 --- a/drivers/infiniband/hw/nes/nes_nic.c +++ b/drivers/infiniband/hw/nes/nes_nic.c @@ -662,10 +662,14 @@ tso_sq_no_longer_full: nesnic->sq_head &= nesnic->sq_size-1; } } else { - nesvnic->linearized_skbs++; hoffset = skb_transport_header(skb) - skb->data; nhoffset = skb_network_header(skb) - skb->data; - skb_linearize(skb); + if (skb_linearize(skb)) { + nesvnic->tx_sw_dropped++; + kfree_skb(skb); + return NETDEV_TX_OK; + } + nesvnic->linearized_skbs++; skb_set_transport_header(skb, hoffset); skb_set_network_header(skb, nhoffset); if (!nes_nic_send(skb, netdev)) @@ -981,20 +985,16 @@ static int nes_netdev_change_mtu(struct net_device *netdev, int new_mtu) { struct nes_vnic *nesvnic = netdev_priv(netdev); struct nes_device *nesdev = nesvnic->nesdev; - int ret = 0; u8 jumbomode = 0; u32 nic_active; u32 nic_active_bit; u32 uc_all_active; u32 mc_all_active; - if ((new_mtu < ETH_ZLEN) || (new_mtu > max_mtu)) - return -EINVAL; - netdev->mtu = new_mtu; nesvnic->max_frame_size = new_mtu + VLAN_ETH_HLEN; - if (netdev->mtu > 1500) { + if (netdev->mtu > ETH_DATA_LEN) { jumbomode=1; } nes_nic_init_timer_defaults(nesdev, jumbomode); @@ -1020,7 +1020,7 @@ static int nes_netdev_change_mtu(struct net_device *netdev, int new_mtu) nes_write_indexed(nesdev, NES_IDX_NIC_UNICAST_ALL, nic_active); } - return ret; + return 0; } @@ -1465,7 +1465,8 @@ static int nes_netdev_set_pauseparam(struct net_device *netdev, /** * nes_netdev_get_settings */ -static int nes_netdev_get_settings(struct net_device *netdev, struct ethtool_cmd *et_cmd) +static int nes_netdev_get_link_ksettings(struct net_device *netdev, + struct ethtool_link_ksettings *cmd) { struct nes_vnic *nesvnic = netdev_priv(netdev); struct nes_device *nesdev = nesvnic->nesdev; @@ -1474,54 +1475,59 @@ static int nes_netdev_get_settings(struct net_device *netdev, struct ethtool_cmd u8 phy_type = nesadapter->phy_type[mac_index]; u8 phy_index = nesadapter->phy_index[mac_index]; u16 phy_data; + u32 supported, advertising; - et_cmd->duplex = DUPLEX_FULL; - et_cmd->port = PORT_MII; - et_cmd->maxtxpkt = 511; - et_cmd->maxrxpkt = 511; + cmd->base.duplex = DUPLEX_FULL; + cmd->base.port = PORT_MII; if (nesadapter->OneG_Mode) { - ethtool_cmd_speed_set(et_cmd, SPEED_1000); + cmd->base.speed = SPEED_1000; if (phy_type == NES_PHY_TYPE_PUMA_1G) { - et_cmd->supported = SUPPORTED_1000baseT_Full; - et_cmd->advertising = ADVERTISED_1000baseT_Full; - et_cmd->autoneg = AUTONEG_DISABLE; - et_cmd->transceiver = XCVR_INTERNAL; - et_cmd->phy_address = mac_index; + supported = SUPPORTED_1000baseT_Full; + advertising = ADVERTISED_1000baseT_Full; + cmd->base.autoneg = AUTONEG_DISABLE; + cmd->base.phy_address = mac_index; } else { unsigned long flags; - et_cmd->supported = SUPPORTED_1000baseT_Full - | SUPPORTED_Autoneg; - et_cmd->advertising = ADVERTISED_1000baseT_Full - | ADVERTISED_Autoneg; + + supported = SUPPORTED_1000baseT_Full + | SUPPORTED_Autoneg; + advertising = ADVERTISED_1000baseT_Full + | ADVERTISED_Autoneg; spin_lock_irqsave(&nesadapter->phy_lock, flags); nes_read_1G_phy_reg(nesdev, 0, phy_index, &phy_data); spin_unlock_irqrestore(&nesadapter->phy_lock, flags); if (phy_data & 0x1000) - et_cmd->autoneg = AUTONEG_ENABLE; + cmd->base.autoneg = AUTONEG_ENABLE; else - et_cmd->autoneg = AUTONEG_DISABLE; - et_cmd->transceiver = XCVR_EXTERNAL; - et_cmd->phy_address = phy_index; + cmd->base.autoneg = AUTONEG_DISABLE; + cmd->base.phy_address = phy_index; } + ethtool_convert_legacy_u32_to_link_mode( + cmd->link_modes.supported, supported); + ethtool_convert_legacy_u32_to_link_mode( + cmd->link_modes.advertising, advertising); return 0; } if ((phy_type == NES_PHY_TYPE_ARGUS) || (phy_type == NES_PHY_TYPE_SFP_D) || (phy_type == NES_PHY_TYPE_KR)) { - et_cmd->transceiver = XCVR_EXTERNAL; - et_cmd->port = PORT_FIBRE; - et_cmd->supported = SUPPORTED_FIBRE; - et_cmd->advertising = ADVERTISED_FIBRE; - et_cmd->phy_address = phy_index; + cmd->base.port = PORT_FIBRE; + supported = SUPPORTED_FIBRE; + advertising = ADVERTISED_FIBRE; + cmd->base.phy_address = phy_index; } else { - et_cmd->transceiver = XCVR_INTERNAL; - et_cmd->supported = SUPPORTED_10000baseT_Full; - et_cmd->advertising = ADVERTISED_10000baseT_Full; - et_cmd->phy_address = mac_index; + supported = SUPPORTED_10000baseT_Full; + advertising = ADVERTISED_10000baseT_Full; + cmd->base.phy_address = mac_index; } - ethtool_cmd_speed_set(et_cmd, SPEED_10000); - et_cmd->autoneg = AUTONEG_DISABLE; + cmd->base.speed = SPEED_10000; + cmd->base.autoneg = AUTONEG_DISABLE; + ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.supported, + supported); + ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.advertising, + advertising); + return 0; } @@ -1529,7 +1535,9 @@ static int nes_netdev_get_settings(struct net_device *netdev, struct ethtool_cmd /** * nes_netdev_set_settings */ -static int nes_netdev_set_settings(struct net_device *netdev, struct ethtool_cmd *et_cmd) +static int +nes_netdev_set_link_ksettings(struct net_device *netdev, + const struct ethtool_link_ksettings *cmd) { struct nes_vnic *nesvnic = netdev_priv(netdev); struct nes_device *nesdev = nesvnic->nesdev; @@ -1543,7 +1551,7 @@ static int nes_netdev_set_settings(struct net_device *netdev, struct ethtool_cmd spin_lock_irqsave(&nesadapter->phy_lock, flags); nes_read_1G_phy_reg(nesdev, 0, phy_index, &phy_data); - if (et_cmd->autoneg) { + if (cmd->base.autoneg) { /* Turn on Full duplex, Autoneg, and restart autonegotiation */ phy_data |= 0x1300; } else { @@ -1560,8 +1568,6 @@ static int nes_netdev_set_settings(struct net_device *netdev, struct ethtool_cmd static const struct ethtool_ops nes_ethtool_ops = { .get_link = ethtool_op_get_link, - .get_settings = nes_netdev_get_settings, - .set_settings = nes_netdev_set_settings, .get_strings = nes_netdev_get_strings, .get_sset_count = nes_netdev_get_sset_count, .get_ethtool_stats = nes_netdev_get_ethtool_stats, @@ -1570,6 +1576,8 @@ static const struct ethtool_ops nes_ethtool_ops = { .set_coalesce = nes_netdev_set_coalesce, .get_pauseparam = nes_netdev_get_pauseparam, .set_pauseparam = nes_netdev_set_pauseparam, + .get_link_ksettings = nes_netdev_get_link_ksettings, + .set_link_ksettings = nes_netdev_set_link_ksettings, }; static void nes_vlan_mode(struct net_device *netdev, struct nes_device *nesdev, netdev_features_t features) @@ -1658,7 +1666,7 @@ struct net_device *nes_netdev_init(struct nes_device *nesdev, netdev->watchdog_timeo = NES_TX_TIMEOUT; netdev->irq = nesdev->pcidev->irq; - netdev->mtu = ETH_DATA_LEN; + netdev->max_mtu = NES_MAX_MTU; netdev->hard_header_len = ETH_HLEN; netdev->addr_len = ETH_ALEN; netdev->type = ARPHRD_ETHER; diff --git a/drivers/infiniband/hw/nes/nes_verbs.c b/drivers/infiniband/hw/nes/nes_verbs.c index bd69125731c1..ccf0a4cffe9c 100644 --- a/drivers/infiniband/hw/nes/nes_verbs.c +++ b/drivers/infiniband/hw/nes/nes_verbs.c @@ -475,20 +475,10 @@ static int nes_query_port(struct ib_device *ibdev, u8 port, struct ib_port_attr struct nes_vnic *nesvnic = to_nesvnic(ibdev); struct net_device *netdev = nesvnic->netdev; - memset(props, 0, sizeof(*props)); + /* props being zeroed by the caller, avoid zeroing it here */ props->max_mtu = IB_MTU_4096; - - if (netdev->mtu >= 4096) - props->active_mtu = IB_MTU_4096; - else if (netdev->mtu >= 2048) - props->active_mtu = IB_MTU_2048; - else if (netdev->mtu >= 1024) - props->active_mtu = IB_MTU_1024; - else if (netdev->mtu >= 512) - props->active_mtu = IB_MTU_512; - else - props->active_mtu = IB_MTU_256; + props->active_mtu = ib_mtu_int_to_enum(netdev->mtu); props->lid = 1; props->lmc = 0; @@ -771,7 +761,8 @@ static int nes_dealloc_pd(struct ib_pd *ibpd) /** * nes_create_ah */ -static struct ib_ah *nes_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr) +static struct ib_ah *nes_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, + struct ib_udata *udata) { return ERR_PTR(-ENOSYS); } @@ -1075,7 +1066,6 @@ static struct ib_qp *nes_create_qp(struct ib_pd *ibpd, mem = kzalloc(sizeof(*nesqp)+NES_SW_CONTEXT_ALIGN-1, GFP_KERNEL); if (!mem) { nes_free_resource(nesadapter, nesadapter->allocated_qps, qp_num); - nes_debug(NES_DBG_QP, "Unable to allocate QP\n"); return ERR_PTR(-ENOMEM); } u64nesqp = (unsigned long)mem; @@ -1475,7 +1465,6 @@ static struct ib_cq *nes_create_cq(struct ib_device *ibdev, nescq = kzalloc(sizeof(struct nes_cq), GFP_KERNEL); if (!nescq) { nes_free_resource(nesadapter, nesadapter->allocated_cqs, cq_num); - nes_debug(NES_DBG_CQ, "Unable to allocate nes_cq struct\n"); return ERR_PTR(-ENOMEM); } @@ -2408,7 +2397,6 @@ static struct ib_mr *nes_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, } nespbl = kzalloc(sizeof(*nespbl), GFP_KERNEL); if (!nespbl) { - nes_debug(NES_DBG_MR, "Unable to allocate PBL\n"); ib_umem_release(region); return ERR_PTR(-ENOMEM); } @@ -2416,7 +2404,6 @@ static struct ib_mr *nes_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, if (!nesmr) { ib_umem_release(region); kfree(nespbl); - nes_debug(NES_DBG_MR, "Unable to allocate nesmr\n"); return ERR_PTR(-ENOMEM); } nesmr->region = region; @@ -3673,13 +3660,14 @@ static int nes_port_immutable(struct ib_device *ibdev, u8 port_num, struct ib_port_attr attr; int err; + immutable->core_cap_flags = RDMA_CORE_PORT_IWARP; + err = nes_query_port(ibdev, port_num, &attr); if (err) return err; immutable->pkey_tbl_len = attr.pkey_tbl_len; immutable->gid_tbl_len = attr.gid_tbl_len; - immutable->core_cap_flags = RDMA_CORE_PORT_IWARP; return 0; } @@ -3743,7 +3731,6 @@ struct nes_ib_device *nes_init_ofa_device(struct net_device *netdev) nesibdev->ibdev.phys_port_cnt = 1; nesibdev->ibdev.num_comp_vectors = 1; - nesibdev->ibdev.dma_device = &nesdev->pcidev->dev; nesibdev->ibdev.dev.parent = &nesdev->pcidev->dev; nesibdev->ibdev.query_device = nes_query_device; nesibdev->ibdev.query_port = nes_query_port; diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_ah.c b/drivers/infiniband/hw/ocrdma/ocrdma_ah.c index 797362a297b2..cd66e1e45dd7 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_ah.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_ah.c @@ -59,7 +59,7 @@ static u16 ocrdma_hdr_type_to_proto_num(int devid, u8 hdr_type) { switch (hdr_type) { case OCRDMA_L3_TYPE_IB_GRH: - return (u16)0x8915; + return (u16)ETH_P_IBOE; case OCRDMA_L3_TYPE_IPV4: return (u16)0x0800; case OCRDMA_L3_TYPE_IPV6: @@ -94,7 +94,7 @@ static inline int set_av_attr(struct ocrdma_dev *dev, struct ocrdma_ah *ah, proto_num = ocrdma_hdr_type_to_proto_num(dev->id, ah->hdr_type); if (!proto_num) return -EINVAL; - nxthdr = (proto_num == 0x8915) ? 0x1b : 0x11; + nxthdr = (proto_num == ETH_P_IBOE) ? 0x1b : 0x11; /* VLAN */ if (!vlan_tag || (vlan_tag > 0xFFF)) vlan_tag = dev->pvid; @@ -154,7 +154,8 @@ static inline int set_av_attr(struct ocrdma_dev *dev, struct ocrdma_ah *ah, return status; } -struct ib_ah *ocrdma_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *attr) +struct ib_ah *ocrdma_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *attr, + struct ib_udata *udata) { u32 *ahid_addr; int status; diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_ah.h b/drivers/infiniband/hw/ocrdma/ocrdma_ah.h index 3856dd4c7e3d..0704a24b17c8 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_ah.h +++ b/drivers/infiniband/hw/ocrdma/ocrdma_ah.h @@ -50,7 +50,9 @@ enum { OCRDMA_AH_L3_TYPE_MASK = 0x03, OCRDMA_AH_L3_TYPE_SHIFT = 0x1D /* 29 bits */ }; -struct ib_ah *ocrdma_create_ah(struct ib_pd *, struct ib_ah_attr *); + +struct ib_ah *ocrdma_create_ah(struct ib_pd *, struct ib_ah_attr *, + struct ib_udata *); int ocrdma_destroy_ah(struct ib_ah *); int ocrdma_query_ah(struct ib_ah *, struct ib_ah_attr *); int ocrdma_modify_ah(struct ib_ah *, struct ib_ah_attr *); diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_hw.c b/drivers/infiniband/hw/ocrdma/ocrdma_hw.c index 67fc0b6857e1..aa6967197620 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_hw.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_hw.c @@ -44,6 +44,7 @@ #include <linux/interrupt.h> #include <linux/log2.h> #include <linux/dma-mapping.h> +#include <linux/if_ether.h> #include <rdma/ib_verbs.h> #include <rdma/ib_user_verbs.h> @@ -1596,10 +1597,9 @@ void ocrdma_alloc_pd_pool(struct ocrdma_dev *dev) dev->pd_mgr = kzalloc(sizeof(struct ocrdma_pd_resource_mgr), GFP_KERNEL); - if (!dev->pd_mgr) { - pr_err("%s(%d)Memory allocation failure.\n", __func__, dev->id); + if (!dev->pd_mgr) return; - } + status = ocrdma_mbx_alloc_pd_range(dev); if (status) { pr_err("%s(%d) Unable to initialize PD pool, using default.\n", @@ -1642,7 +1642,7 @@ static int ocrdma_build_q_conf(u32 *num_entries, int entry_size, static int ocrdma_mbx_create_ah_tbl(struct ocrdma_dev *dev) { int i; - int status = 0; + int status = -ENOMEM; int max_ah; struct ocrdma_create_ah_tbl *cmd; struct ocrdma_create_ah_tbl_rsp *rsp; @@ -2985,7 +2985,7 @@ static int ocrdma_parse_dcbxcfg_rsp(struct ocrdma_dev *dev, int ptype, OCRDMA_APP_PARAM_APP_PROTO_MASK; if ( - valid && proto == OCRDMA_APP_PROTO_ROCE && + valid && proto == ETH_P_IBOE && proto_sel == OCRDMA_PROTO_SELECT_L2) { for (slindx = 0; slindx < OCRDMA_MAX_SERVICE_LEVEL_INDEX; slindx++) { diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_main.c b/drivers/infiniband/hw/ocrdma/ocrdma_main.c index 896071502739..57c9a2ad0260 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_main.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_main.c @@ -93,15 +93,16 @@ static int ocrdma_port_immutable(struct ib_device *ibdev, u8 port_num, int err; dev = get_ocrdma_dev(ibdev); - err = ocrdma_query_port(ibdev, port_num, &attr); + immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE; + if (ocrdma_is_udp_encap_supported(dev)) + immutable->core_cap_flags |= RDMA_CORE_CAP_PROT_ROCE_UDP_ENCAP; + + err = ib_query_port(ibdev, port_num, &attr); if (err) return err; immutable->pkey_tbl_len = attr.pkey_tbl_len; immutable->gid_tbl_len = attr.gid_tbl_len; - immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE; - if (ocrdma_is_udp_encap_supported(dev)) - immutable->core_cap_flags |= RDMA_CORE_CAP_PROT_ROCE_UDP_ENCAP; immutable->max_mad_size = IB_MGMT_MAD_SIZE; return 0; @@ -198,7 +199,7 @@ static int ocrdma_register_device(struct ocrdma_dev *dev) dev->ibdev.alloc_ucontext = ocrdma_alloc_ucontext; dev->ibdev.dealloc_ucontext = ocrdma_dealloc_ucontext; dev->ibdev.mmap = ocrdma_mmap; - dev->ibdev.dma_device = &dev->nic_info.pdev->dev; + dev->ibdev.dev.parent = &dev->nic_info.pdev->dev; dev->ibdev.process_mad = ocrdma_process_mad; dev->ibdev.get_port_immutable = ocrdma_port_immutable; diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_sli.h b/drivers/infiniband/hw/ocrdma/ocrdma_sli.h index 37df4481bb8f..6ef89c226ad8 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_sli.h +++ b/drivers/infiniband/hw/ocrdma/ocrdma_sli.h @@ -1901,7 +1901,6 @@ struct ocrdma_eth_vlan { u8 smac[6]; __be16 eth_type; __be16 vlan_tag; -#define OCRDMA_ROCE_ETH_TYPE 0x8915 __be16 roce_eth_type; } __packed; @@ -2179,10 +2178,6 @@ enum OCRDMA_DCBX_PARAM_TYPE { OCRDMA_PARAMETER_TYPE_PEER = 0x02 }; -enum OCRDMA_DCBX_APP_PROTO { - OCRDMA_APP_PROTO_ROCE = 0x8915 -}; - enum OCRDMA_DCBX_PROTO { OCRDMA_PROTO_SELECT_L2 = 0x00, OCRDMA_PROTO_SELECT_L4 = 0x01 diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_stats.c b/drivers/infiniband/hw/ocrdma/ocrdma_stats.c index 8bef09a8c49f..f8e4b0a6486f 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_stats.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_stats.c @@ -84,10 +84,8 @@ bool ocrdma_alloc_stats_resources(struct ocrdma_dev *dev) /* Alloc debugfs mem */ mem->debugfs_mem = kzalloc(OCRDMA_MAX_DBGFS_MEM, GFP_KERNEL); - if (!mem->debugfs_mem) { - pr_err("%s: stats debugfs mem allocation failed\n", __func__); + if (!mem->debugfs_mem) return false; - } return true; } diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c index 6af44f8db3d5..bc9fb144e57b 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c @@ -210,6 +210,7 @@ int ocrdma_query_port(struct ib_device *ibdev, struct ocrdma_dev *dev; struct net_device *netdev; + /* props being zeroed by the caller, avoid zeroing it here */ dev = get_ocrdma_dev(ibdev); if (port > 1) { pr_err("%s(%d) invalid_port=0x%x\n", __func__, @@ -1170,8 +1171,7 @@ int ocrdma_destroy_cq(struct ib_cq *ibcq) dev->cq_tbl[cq->id] = NULL; indx = ocrdma_get_eq_table_index(dev, cq->eqn); - if (indx == -EINVAL) - BUG(); + BUG_ON(indx == -EINVAL); eq = &dev->eq_tbl[indx]; irq = ocrdma_get_irq(dev, eq); @@ -1741,8 +1741,7 @@ static void ocrdma_discard_cqes(struct ocrdma_qp *qp, struct ocrdma_cq *cq) wqe_idx = (le32_to_cpu(cqe->rq.buftag_qpn) >> OCRDMA_CQE_BUFTAG_SHIFT) & qp->srq->rq.max_wqe_idx; - if (wqe_idx < 1) - BUG(); + BUG_ON(wqe_idx < 1); spin_lock_irqsave(&qp->srq->q_lock, flags); ocrdma_hwq_inc_tail(&qp->srq->rq); ocrdma_srq_toggle_bit(qp->srq, wqe_idx - 1); @@ -2388,15 +2387,13 @@ static int ocrdma_srq_get_idx(struct ocrdma_srq *srq) if (srq->idx_bit_fields[row]) { indx = ffs(srq->idx_bit_fields[row]); indx = (row * 32) + (indx - 1); - if (indx >= srq->rq.max_cnt) - BUG(); + BUG_ON(indx >= srq->rq.max_cnt); ocrdma_srq_toggle_bit(srq, indx); break; } } - if (row == srq->bit_fields_len) - BUG(); + BUG_ON(row == srq->bit_fields_len); return indx + 1; /* Use from index 1 */ } @@ -2754,8 +2751,7 @@ static void ocrdma_update_free_srq_cqe(struct ib_wc *ibwc, srq = get_ocrdma_srq(qp->ibqp.srq); wqe_idx = (le32_to_cpu(cqe->rq.buftag_qpn) >> OCRDMA_CQE_BUFTAG_SHIFT) & srq->rq.max_wqe_idx; - if (wqe_idx < 1) - BUG(); + BUG_ON(wqe_idx < 1); ibwc->wr_id = srq->rqe_wr_id_tbl[wqe_idx]; spin_lock_irqsave(&srq->q_lock, flags); diff --git a/drivers/infiniband/hw/qedr/main.c b/drivers/infiniband/hw/qedr/main.c index 7b74d09a8217..b9b47e5cc8b3 100644 --- a/drivers/infiniband/hw/qedr/main.c +++ b/drivers/infiniband/hw/qedr/main.c @@ -170,7 +170,7 @@ static int qedr_register_device(struct qedr_dev *dev) dev->ibdev.get_port_immutable = qedr_port_immutable; dev->ibdev.get_netdev = qedr_get_netdev; - dev->ibdev.dma_device = &dev->pdev->dev; + dev->ibdev.dev.parent = &dev->pdev->dev; dev->ibdev.get_link_layer = qedr_link_layer; dev->ibdev.get_dev_fw_str = qedr_get_dev_fw_str; @@ -576,8 +576,7 @@ static int qedr_set_device_attr(struct qedr_dev *dev) return 0; } -void qedr_unaffiliated_event(void *context, - u8 event_code) +void qedr_unaffiliated_event(void *context, u8 event_code) { pr_err("unaffiliated event not implemented yet\n"); } @@ -792,6 +791,9 @@ static struct qedr_dev *qedr_add(struct qed_dev *cdev, struct pci_dev *pdev, if (device_create_file(&dev->ibdev.dev, qedr_attributes[i])) goto sysfs_err; + if (!test_and_set_bit(QEDR_ENET_STATE_BIT, &dev->enet_state)) + qedr_ib_dispatch_event(dev, QEDR_PORT, IB_EVENT_PORT_ACTIVE); + DP_DEBUG(dev, QEDR_MSG_INIT, "qedr driver loaded successfully\n"); return dev; @@ -824,11 +826,10 @@ static void qedr_remove(struct qedr_dev *dev) ib_dealloc_device(&dev->ibdev); } -static int qedr_close(struct qedr_dev *dev) +static void qedr_close(struct qedr_dev *dev) { - qedr_ib_dispatch_event(dev, 1, IB_EVENT_PORT_ERR); - - return 0; + if (test_and_clear_bit(QEDR_ENET_STATE_BIT, &dev->enet_state)) + qedr_ib_dispatch_event(dev, QEDR_PORT, IB_EVENT_PORT_ERR); } static void qedr_shutdown(struct qedr_dev *dev) @@ -837,6 +838,12 @@ static void qedr_shutdown(struct qedr_dev *dev) qedr_remove(dev); } +static void qedr_open(struct qedr_dev *dev) +{ + if (!test_and_set_bit(QEDR_ENET_STATE_BIT, &dev->enet_state)) + qedr_ib_dispatch_event(dev, QEDR_PORT, IB_EVENT_PORT_ACTIVE); +} + static void qedr_mac_address_change(struct qedr_dev *dev) { union ib_gid *sgid = &dev->sgid_tbl[0]; @@ -863,7 +870,7 @@ static void qedr_mac_address_change(struct qedr_dev *dev) ether_addr_copy(dev->gsi_ll2_mac_address, dev->ndev->dev_addr); - qedr_ib_dispatch_event(dev, 1, IB_EVENT_GID_CHANGE); + qedr_ib_dispatch_event(dev, QEDR_PORT, IB_EVENT_GID_CHANGE); if (rc) DP_ERR(dev, "Error updating mac filter\n"); @@ -877,7 +884,7 @@ static void qedr_notify(struct qedr_dev *dev, enum qede_roce_event event) { switch (event) { case QEDE_UP: - qedr_ib_dispatch_event(dev, 1, IB_EVENT_PORT_ACTIVE); + qedr_open(dev); break; case QEDE_DOWN: qedr_close(dev); diff --git a/drivers/infiniband/hw/qedr/qedr.h b/drivers/infiniband/hw/qedr/qedr.h index 620badd7d4fb..bb32e4792ec9 100644 --- a/drivers/infiniband/hw/qedr/qedr.h +++ b/drivers/infiniband/hw/qedr/qedr.h @@ -113,6 +113,8 @@ struct qedr_device_attr { struct qed_rdma_events events; }; +#define QEDR_ENET_STATE_BIT (0) + struct qedr_dev { struct ib_device ibdev; struct qed_dev *cdev; @@ -153,6 +155,8 @@ struct qedr_dev { struct qedr_cq *gsi_sqcq; struct qedr_cq *gsi_rqcq; struct qedr_qp *gsi_qp; + + unsigned long enet_state; }; #define QEDR_MAX_SQ_PBL (0x8000) @@ -188,6 +192,7 @@ struct qedr_dev { #define QEDR_ROCE_MAX_CNQ_SIZE (0x4000) #define QEDR_MAX_PORT (1) +#define QEDR_PORT (1) #define QEDR_UVERBS(CMD_NAME) (1ull << IB_USER_VERBS_CMD_##CMD_NAME) @@ -251,9 +256,6 @@ struct qedr_cq { u16 icid; - /* Lock to protect completion handler */ - spinlock_t comp_handler_lock; - /* Lock to protect multiplem CQ's */ spinlock_t cq_lock; u8 arm_flags; diff --git a/drivers/infiniband/hw/qedr/qedr_cm.c b/drivers/infiniband/hw/qedr/qedr_cm.c index 63890ebb72bd..699632893dd9 100644 --- a/drivers/infiniband/hw/qedr/qedr_cm.c +++ b/drivers/infiniband/hw/qedr/qedr_cm.c @@ -87,11 +87,8 @@ void qedr_ll2_tx_cb(void *_qdev, struct qed_roce_ll2_packet *pkt) qedr_inc_sw_gsi_cons(&qp->sq); spin_unlock_irqrestore(&qp->q_lock, flags); - if (cq->ibcq.comp_handler) { - spin_lock_irqsave(&cq->comp_handler_lock, flags); + if (cq->ibcq.comp_handler) (*cq->ibcq.comp_handler) (&cq->ibcq, cq->ibcq.cq_context); - spin_unlock_irqrestore(&cq->comp_handler_lock, flags); - } } void qedr_ll2_rx_cb(void *_dev, struct qed_roce_ll2_packet *pkt, @@ -113,11 +110,8 @@ void qedr_ll2_rx_cb(void *_dev, struct qed_roce_ll2_packet *pkt, spin_unlock_irqrestore(&qp->q_lock, flags); - if (cq->ibcq.comp_handler) { - spin_lock_irqsave(&cq->comp_handler_lock, flags); + if (cq->ibcq.comp_handler) (*cq->ibcq.comp_handler) (&cq->ibcq, cq->ibcq.cq_context); - spin_unlock_irqrestore(&cq->comp_handler_lock, flags); - } } static void qedr_destroy_gsi_cq(struct qedr_dev *dev, @@ -293,7 +287,7 @@ static inline int qedr_gsi_build_header(struct qedr_dev *dev, has_udp = (sgid_attr.gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP); if (!has_udp) { /* RoCE v1 */ - ether_type = ETH_P_ROCE; + ether_type = ETH_P_IBOE; *roce_mode = ROCE_V1; } else if (ipv6_addr_v4mapped((struct in6_addr *)&sgid)) { /* RoCE v2 IPv4 */ @@ -404,9 +398,9 @@ static inline int qedr_gsi_build_packet(struct qedr_dev *dev, } if (ether_addr_equal(udh.eth.smac_h, udh.eth.dmac_h)) - packet->tx_dest = QED_ROCE_LL2_TX_DEST_NW; - else packet->tx_dest = QED_ROCE_LL2_TX_DEST_LB; + else + packet->tx_dest = QED_ROCE_LL2_TX_DEST_NW; packet->roce_mode = roce_mode; memcpy(packet->header.vaddr, ud_header_buffer, header_size); diff --git a/drivers/infiniband/hw/qedr/qedr_cm.h b/drivers/infiniband/hw/qedr/qedr_cm.h index 9ba6e15cd93f..78efb1b056d1 100644 --- a/drivers/infiniband/hw/qedr/qedr_cm.h +++ b/drivers/infiniband/hw/qedr/qedr_cm.h @@ -37,7 +37,6 @@ #define QEDR_GSI_MAX_RECV_SGE (1) /* LL2 FW limitation */ -#define ETH_P_ROCE (0x8915) #define QEDR_ROCE_V2_UDP_SPORT (0000) static inline u32 qedr_get_ipv4_from_gid(u8 *gid) diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c index a61514296767..6b3bb32803bd 100644 --- a/drivers/infiniband/hw/qedr/verbs.c +++ b/drivers/infiniband/hw/qedr/verbs.c @@ -238,8 +238,8 @@ int qedr_query_port(struct ib_device *ibdev, u8 port, struct ib_port_attr *attr) } rdma_port = dev->ops->rdma_query_port(dev->rdma_ctx); - memset(attr, 0, sizeof(*attr)); + /* *attr being zeroed by the caller, avoid zeroing it here */ if (rdma_port->port_state == QED_RDMA_PORT_UP) { attr->state = IB_PORT_ACTIVE; attr->phys_state = 5; @@ -471,8 +471,6 @@ struct ib_pd *qedr_alloc_pd(struct ib_device *ibdev, struct ib_ucontext *context, struct ib_udata *udata) { struct qedr_dev *dev = get_qedr_dev(ibdev); - struct qedr_ucontext *uctx = NULL; - struct qedr_alloc_pd_uresp uresp; struct qedr_pd *pd; u16 pd_id; int rc; @@ -489,21 +487,33 @@ struct ib_pd *qedr_alloc_pd(struct ib_device *ibdev, if (!pd) return ERR_PTR(-ENOMEM); - dev->ops->rdma_alloc_pd(dev->rdma_ctx, &pd_id); + rc = dev->ops->rdma_alloc_pd(dev->rdma_ctx, &pd_id); + if (rc) + goto err; - uresp.pd_id = pd_id; pd->pd_id = pd_id; if (udata && context) { + struct qedr_alloc_pd_uresp uresp; + + uresp.pd_id = pd_id; + rc = ib_copy_to_udata(udata, &uresp, sizeof(uresp)); - if (rc) + if (rc) { DP_ERR(dev, "copy error pd_id=0x%x.\n", pd_id); - uctx = get_qedr_ucontext(context); - uctx->pd = pd; - pd->uctx = uctx; + dev->ops->rdma_dealloc_pd(dev->rdma_ctx, pd_id); + goto err; + } + + pd->uctx = get_qedr_ucontext(context); + pd->uctx->pd = pd; } return &pd->ibpd; + +err: + kfree(pd); + return ERR_PTR(rc); } int qedr_dealloc_pd(struct ib_pd *ibpd) @@ -511,8 +521,10 @@ int qedr_dealloc_pd(struct ib_pd *ibpd) struct qedr_dev *dev = get_qedr_dev(ibpd->device); struct qedr_pd *pd = get_qedr_pd(ibpd); - if (!pd) + if (!pd) { pr_err("Invalid PD received in dealloc_pd\n"); + return -EINVAL; + } DP_DEBUG(dev, QEDR_MSG_INIT, "Deallocating PD %d\n", pd->pd_id); dev->ops->rdma_dealloc_pd(dev->rdma_ctx, pd->pd_id); @@ -759,8 +771,10 @@ static inline int qedr_init_user_queue(struct ib_ucontext *ib_ctx, goto err0; q->pbl_tbl = qedr_alloc_pbl_tbl(dev, &q->pbl_info, GFP_KERNEL); - if (IS_ERR_OR_NULL(q->pbl_tbl)) + if (IS_ERR(q->pbl_tbl)) { + rc = PTR_ERR(q->pbl_tbl); goto err0; + } qedr_populate_pbls(dev, q->umem, q->pbl_tbl, &q->pbl_info); @@ -888,6 +902,8 @@ struct ib_cq *qedr_create_cq(struct ib_device *ibdev, pbl_ptr = cq->q.pbl_tbl->pa; page_cnt = cq->q.pbl_info.num_pbes; + + cq->ibcq.cqe = chain_entries; } else { cq->cq_type = QEDR_CQ_TYPE_KERNEL; @@ -903,6 +919,7 @@ struct ib_cq *qedr_create_cq(struct ib_device *ibdev, page_cnt = qed_chain_get_page_cnt(&cq->pbl); pbl_ptr = qed_chain_get_pbl_phys(&cq->pbl); + cq->ibcq.cqe = cq->pbl.capacity; } qedr_init_cq_params(cq, ctx, dev, vector, chain_entries, page_cnt, @@ -980,8 +997,13 @@ int qedr_destroy_cq(struct ib_cq *ibcq) /* GSIs CQs are handled by driver, so they don't exist in the FW */ if (cq->cq_type != QEDR_CQ_TYPE_GSI) { + int rc; + iparams.icid = cq->icid; - dev->ops->rdma_destroy_cq(dev->rdma_ctx, &iparams, &oparams); + rc = dev->ops->rdma_destroy_cq(dev->rdma_ctx, &iparams, + &oparams); + if (rc) + return rc; dev->ops->common->chain_free(dev->cdev, &cq->pbl); } @@ -1066,30 +1088,6 @@ static inline int get_gid_info_from_table(struct ib_qp *ibqp, return 0; } -static void qedr_cleanup_user_sq(struct qedr_dev *dev, struct qedr_qp *qp) -{ - qedr_free_pbl(dev, &qp->usq.pbl_info, qp->usq.pbl_tbl); - ib_umem_release(qp->usq.umem); -} - -static void qedr_cleanup_user_rq(struct qedr_dev *dev, struct qedr_qp *qp) -{ - qedr_free_pbl(dev, &qp->urq.pbl_info, qp->urq.pbl_tbl); - ib_umem_release(qp->urq.umem); -} - -static void qedr_cleanup_kernel_sq(struct qedr_dev *dev, struct qedr_qp *qp) -{ - dev->ops->common->chain_free(dev->cdev, &qp->sq.pbl); - kfree(qp->wqe_wr_id); -} - -static void qedr_cleanup_kernel_rq(struct qedr_dev *dev, struct qedr_qp *qp) -{ - dev->ops->common->chain_free(dev->cdev, &qp->rq.pbl); - kfree(qp->rqe_wr_id); -} - static int qedr_check_qp_attrs(struct ib_pd *ibpd, struct qedr_dev *dev, struct ib_qp_init_attr *attrs) { @@ -1178,15 +1176,13 @@ static int qedr_copy_qp_uresp(struct qedr_dev *dev, return rc; } -static void qedr_set_qp_init_params(struct qedr_dev *dev, - struct qedr_qp *qp, - struct qedr_pd *pd, - struct ib_qp_init_attr *attrs) +static void qedr_set_common_qp_params(struct qedr_dev *dev, + struct qedr_qp *qp, + struct qedr_pd *pd, + struct ib_qp_init_attr *attrs) { - qp->pd = pd; - spin_lock_init(&qp->q_lock); - + qp->pd = pd; qp->qp_type = attrs->qp_type; qp->max_inline_data = attrs->cap.max_inline_data; qp->sq.max_sges = attrs->cap.max_send_sge; @@ -1195,103 +1191,161 @@ static void qedr_set_qp_init_params(struct qedr_dev *dev, qp->sq_cq = get_qedr_cq(attrs->send_cq); qp->rq_cq = get_qedr_cq(attrs->recv_cq); qp->dev = dev; + qp->rq.max_sges = attrs->cap.max_recv_sge; DP_DEBUG(dev, QEDR_MSG_QP, + "RQ params:\trq_max_sges = %d, rq_cq_id = %d\n", + qp->rq.max_sges, qp->rq_cq->icid); + DP_DEBUG(dev, QEDR_MSG_QP, "QP params:\tpd = %d, qp_type = %d, max_inline_data = %d, state = %d, signaled = %d, use_srq=%d\n", pd->pd_id, qp->qp_type, qp->max_inline_data, qp->state, qp->signaled, (attrs->srq) ? 1 : 0); DP_DEBUG(dev, QEDR_MSG_QP, "SQ params:\tsq_max_sges = %d, sq_cq_id = %d\n", qp->sq.max_sges, qp->sq_cq->icid); - qp->rq.max_sges = attrs->cap.max_recv_sge; - DP_DEBUG(dev, QEDR_MSG_QP, - "RQ params:\trq_max_sges = %d, rq_cq_id = %d\n", - qp->rq.max_sges, qp->rq_cq->icid); } -static inline void -qedr_init_qp_user_params(struct qed_rdma_create_qp_in_params *params, - struct qedr_create_qp_ureq *ureq) -{ - /* QP handle to be written in CQE */ - params->qp_handle_lo = ureq->qp_handle_lo; - params->qp_handle_hi = ureq->qp_handle_hi; -} - -static inline void -qedr_init_qp_kernel_doorbell_sq(struct qedr_dev *dev, struct qedr_qp *qp) +static void qedr_set_roce_db_info(struct qedr_dev *dev, struct qedr_qp *qp) { qp->sq.db = dev->db_addr + DB_ADDR_SHIFT(DQ_PWM_OFFSET_XCM_RDMA_SQ_PROD); qp->sq.db_data.data.icid = qp->icid + 1; + qp->rq.db = dev->db_addr + + DB_ADDR_SHIFT(DQ_PWM_OFFSET_TCM_ROCE_RQ_PROD); + qp->rq.db_data.data.icid = qp->icid; } static inline void -qedr_init_qp_kernel_doorbell_rq(struct qedr_dev *dev, struct qedr_qp *qp) +qedr_init_common_qp_in_params(struct qedr_dev *dev, + struct qedr_pd *pd, + struct qedr_qp *qp, + struct ib_qp_init_attr *attrs, + bool fmr_and_reserved_lkey, + struct qed_rdma_create_qp_in_params *params) { - qp->rq.db = dev->db_addr + - DB_ADDR_SHIFT(DQ_PWM_OFFSET_TCM_ROCE_RQ_PROD); - qp->rq.db_data.data.icid = qp->icid; + /* QP handle to be written in an async event */ + params->qp_handle_async_lo = lower_32_bits((uintptr_t) qp); + params->qp_handle_async_hi = upper_32_bits((uintptr_t) qp); + + params->signal_all = (attrs->sq_sig_type == IB_SIGNAL_ALL_WR); + params->fmr_and_reserved_lkey = fmr_and_reserved_lkey; + params->pd = pd->pd_id; + params->dpi = pd->uctx ? pd->uctx->dpi : dev->dpi; + params->sq_cq_id = get_qedr_cq(attrs->send_cq)->icid; + params->stats_queue = 0; + params->rq_cq_id = get_qedr_cq(attrs->recv_cq)->icid; + params->srq_id = 0; + params->use_srq = false; } -static inline int -qedr_init_qp_kernel_params_rq(struct qedr_dev *dev, - struct qedr_qp *qp, struct ib_qp_init_attr *attrs) +static inline void qedr_qp_user_print(struct qedr_dev *dev, struct qedr_qp *qp) { - /* Allocate driver internal RQ array */ - qp->rqe_wr_id = kcalloc(qp->rq.max_wr, sizeof(*qp->rqe_wr_id), - GFP_KERNEL); - if (!qp->rqe_wr_id) - return -ENOMEM; + DP_DEBUG(dev, QEDR_MSG_QP, "create qp: successfully created user QP. " + "qp=%p. " + "sq_addr=0x%llx, " + "sq_len=%zd, " + "rq_addr=0x%llx, " + "rq_len=%zd" + "\n", + qp, + qp->usq.buf_addr, + qp->usq.buf_len, qp->urq.buf_addr, qp->urq.buf_len); +} - DP_DEBUG(dev, QEDR_MSG_QP, "RQ max_wr set to %d.\n", qp->rq.max_wr); +static void qedr_cleanup_user(struct qedr_dev *dev, struct qedr_qp *qp) +{ + if (qp->usq.umem) + ib_umem_release(qp->usq.umem); + qp->usq.umem = NULL; - return 0; + if (qp->urq.umem) + ib_umem_release(qp->urq.umem); + qp->urq.umem = NULL; } -static inline int -qedr_init_qp_kernel_params_sq(struct qedr_dev *dev, - struct qedr_qp *qp, - struct ib_qp_init_attr *attrs, - struct qed_rdma_create_qp_in_params *params) +static int qedr_create_user_qp(struct qedr_dev *dev, + struct qedr_qp *qp, + struct ib_pd *ibpd, + struct ib_udata *udata, + struct ib_qp_init_attr *attrs) { - u32 temp_max_wr; + struct qed_rdma_create_qp_in_params in_params; + struct qed_rdma_create_qp_out_params out_params; + struct qedr_pd *pd = get_qedr_pd(ibpd); + struct ib_ucontext *ib_ctx = NULL; + struct qedr_ucontext *ctx = NULL; + struct qedr_create_qp_ureq ureq; + int rc = -EINVAL; - /* Allocate driver internal SQ array */ - temp_max_wr = attrs->cap.max_send_wr * dev->wq_multiplier; - temp_max_wr = min_t(u32, temp_max_wr, dev->attr.max_sqe); + ib_ctx = ibpd->uobject->context; + ctx = get_qedr_ucontext(ib_ctx); - /* temp_max_wr < attr->max_sqe < u16 so the casting is safe */ - qp->sq.max_wr = (u16)temp_max_wr; - qp->wqe_wr_id = kcalloc(qp->sq.max_wr, sizeof(*qp->wqe_wr_id), - GFP_KERNEL); - if (!qp->wqe_wr_id) - return -ENOMEM; + memset(&ureq, 0, sizeof(ureq)); + rc = ib_copy_from_udata(&ureq, udata, sizeof(ureq)); + if (rc) { + DP_ERR(dev, "Problem copying data from user space\n"); + return rc; + } - DP_DEBUG(dev, QEDR_MSG_QP, "SQ max_wr set to %d.\n", qp->sq.max_wr); + /* SQ - read access only (0), dma sync not required (0) */ + rc = qedr_init_user_queue(ib_ctx, dev, &qp->usq, ureq.sq_addr, + ureq.sq_len, 0, 0); + if (rc) + return rc; - /* QP handle to be written in CQE */ - params->qp_handle_lo = lower_32_bits((uintptr_t)qp); - params->qp_handle_hi = upper_32_bits((uintptr_t)qp); + /* RQ - read access only (0), dma sync not required (0) */ + rc = qedr_init_user_queue(ib_ctx, dev, &qp->urq, ureq.rq_addr, + ureq.rq_len, 0, 0); + + if (rc) + return rc; + + memset(&in_params, 0, sizeof(in_params)); + qedr_init_common_qp_in_params(dev, pd, qp, attrs, false, &in_params); + in_params.qp_handle_lo = ureq.qp_handle_lo; + in_params.qp_handle_hi = ureq.qp_handle_hi; + in_params.sq_num_pages = qp->usq.pbl_info.num_pbes; + in_params.sq_pbl_ptr = qp->usq.pbl_tbl->pa; + in_params.rq_num_pages = qp->urq.pbl_info.num_pbes; + in_params.rq_pbl_ptr = qp->urq.pbl_tbl->pa; + + qp->qed_qp = dev->ops->rdma_create_qp(dev->rdma_ctx, + &in_params, &out_params); + + if (!qp->qed_qp) { + rc = -ENOMEM; + goto err1; + } + + qp->qp_id = out_params.qp_id; + qp->icid = out_params.icid; + + rc = qedr_copy_qp_uresp(dev, qp, udata); + if (rc) + goto err; + + qedr_qp_user_print(dev, qp); return 0; +err: + rc = dev->ops->rdma_destroy_qp(dev->rdma_ctx, qp->qed_qp); + if (rc) + DP_ERR(dev, "create qp: fatal fault. rc=%d", rc); + +err1: + qedr_cleanup_user(dev, qp); + return rc; } -static inline int qedr_init_qp_kernel_sq(struct qedr_dev *dev, - struct qedr_qp *qp, - struct ib_qp_init_attr *attrs) +static int +qedr_roce_create_kernel_qp(struct qedr_dev *dev, + struct qedr_qp *qp, + struct qed_rdma_create_qp_in_params *in_params, + u32 n_sq_elems, u32 n_rq_elems) { - u32 n_sq_elems, n_sq_entries; + struct qed_rdma_create_qp_out_params out_params; int rc; - /* A single work request may take up to QEDR_MAX_SQ_WQE_SIZE elements in - * the ring. The ring should allow at least a single WR, even if the - * user requested none, due to allocation issues. - */ - n_sq_entries = attrs->cap.max_send_wr; - n_sq_entries = min_t(u32, n_sq_entries, dev->attr.max_sqe); - n_sq_entries = max_t(u32, n_sq_entries, 1); - n_sq_elems = n_sq_entries * QEDR_MAX_SQE_ELEMENTS_PER_SQE; rc = dev->ops->common->chain_alloc(dev->cdev, QED_CHAIN_USE_TO_PRODUCE, QED_CHAIN_MODE_PBL, @@ -1299,31 +1353,13 @@ static inline int qedr_init_qp_kernel_sq(struct qedr_dev *dev, n_sq_elems, QEDR_SQE_ELEMENT_SIZE, &qp->sq.pbl); - if (rc) { - DP_ERR(dev, "failed to allocate QP %p SQ\n", qp); - return rc; - } - DP_DEBUG(dev, QEDR_MSG_SQ, - "SQ Pbl base addr = %llx max_send_wr=%d max_wr=%d capacity=%d, rc=%d\n", - qed_chain_get_pbl_phys(&qp->sq.pbl), attrs->cap.max_send_wr, - n_sq_entries, qed_chain_get_capacity(&qp->sq.pbl), rc); - return 0; -} + if (rc) + return rc; -static inline int qedr_init_qp_kernel_rq(struct qedr_dev *dev, - struct qedr_qp *qp, - struct ib_qp_init_attr *attrs) -{ - u32 n_rq_elems, n_rq_entries; - int rc; + in_params->sq_num_pages = qed_chain_get_page_cnt(&qp->sq.pbl); + in_params->sq_pbl_ptr = qed_chain_get_pbl_phys(&qp->sq.pbl); - /* A single work request may take up to QEDR_MAX_RQ_WQE_SIZE elements in - * the ring. There ring should allow at least a single WR, even if the - * user requested none, due to allocation issues. - */ - n_rq_entries = max_t(u32, attrs->cap.max_recv_wr, 1); - n_rq_elems = n_rq_entries * QEDR_MAX_RQE_ELEMENTS_PER_RQE; rc = dev->ops->common->chain_alloc(dev->cdev, QED_CHAIN_USE_TO_CONSUME_PRODUCE, QED_CHAIN_MODE_PBL, @@ -1331,136 +1367,102 @@ static inline int qedr_init_qp_kernel_rq(struct qedr_dev *dev, n_rq_elems, QEDR_RQE_ELEMENT_SIZE, &qp->rq.pbl); + if (rc) + return rc; - if (rc) { - DP_ERR(dev, "failed to allocate memory for QP %p RQ\n", qp); - return -ENOMEM; - } - - DP_DEBUG(dev, QEDR_MSG_RQ, - "RQ Pbl base addr = %llx max_recv_wr=%d max_wr=%d capacity=%d, rc=%d\n", - qed_chain_get_pbl_phys(&qp->rq.pbl), attrs->cap.max_recv_wr, - n_rq_entries, qed_chain_get_capacity(&qp->rq.pbl), rc); + in_params->rq_num_pages = qed_chain_get_page_cnt(&qp->rq.pbl); + in_params->rq_pbl_ptr = qed_chain_get_pbl_phys(&qp->rq.pbl); - /* n_rq_entries < u16 so the casting is safe */ - qp->rq.max_wr = (u16)n_rq_entries; + qp->qed_qp = dev->ops->rdma_create_qp(dev->rdma_ctx, + in_params, &out_params); - return 0; -} + if (!qp->qed_qp) + return -EINVAL; -static inline void -qedr_init_qp_in_params_sq(struct qedr_dev *dev, - struct qedr_pd *pd, - struct qedr_qp *qp, - struct ib_qp_init_attr *attrs, - struct ib_udata *udata, - struct qed_rdma_create_qp_in_params *params) -{ - /* QP handle to be written in an async event */ - params->qp_handle_async_lo = lower_32_bits((uintptr_t)qp); - params->qp_handle_async_hi = upper_32_bits((uintptr_t)qp); + qp->qp_id = out_params.qp_id; + qp->icid = out_params.icid; - params->signal_all = (attrs->sq_sig_type == IB_SIGNAL_ALL_WR); - params->fmr_and_reserved_lkey = !udata; - params->pd = pd->pd_id; - params->dpi = pd->uctx ? pd->uctx->dpi : dev->dpi; - params->sq_cq_id = get_qedr_cq(attrs->send_cq)->icid; - params->max_sq_sges = 0; - params->stats_queue = 0; + qedr_set_roce_db_info(dev, qp); - if (udata) { - params->sq_num_pages = qp->usq.pbl_info.num_pbes; - params->sq_pbl_ptr = qp->usq.pbl_tbl->pa; - } else { - params->sq_num_pages = qed_chain_get_page_cnt(&qp->sq.pbl); - params->sq_pbl_ptr = qed_chain_get_pbl_phys(&qp->sq.pbl); - } + return 0; } -static inline void -qedr_init_qp_in_params_rq(struct qedr_qp *qp, - struct ib_qp_init_attr *attrs, - struct ib_udata *udata, - struct qed_rdma_create_qp_in_params *params) +static void qedr_cleanup_kernel(struct qedr_dev *dev, struct qedr_qp *qp) { - params->rq_cq_id = get_qedr_cq(attrs->recv_cq)->icid; - params->srq_id = 0; - params->use_srq = false; + dev->ops->common->chain_free(dev->cdev, &qp->sq.pbl); + kfree(qp->wqe_wr_id); - if (udata) { - params->rq_num_pages = qp->urq.pbl_info.num_pbes; - params->rq_pbl_ptr = qp->urq.pbl_tbl->pa; - } else { - params->rq_num_pages = qed_chain_get_page_cnt(&qp->rq.pbl); - params->rq_pbl_ptr = qed_chain_get_pbl_phys(&qp->rq.pbl); - } + dev->ops->common->chain_free(dev->cdev, &qp->rq.pbl); + kfree(qp->rqe_wr_id); } -static inline void qedr_qp_user_print(struct qedr_dev *dev, struct qedr_qp *qp) +static int qedr_create_kernel_qp(struct qedr_dev *dev, + struct qedr_qp *qp, + struct ib_pd *ibpd, + struct ib_qp_init_attr *attrs) { - DP_DEBUG(dev, QEDR_MSG_QP, - "create qp: successfully created user QP. qp=%p, sq_addr=0x%llx, sq_len=%zd, rq_addr=0x%llx, rq_len=%zd\n", - qp, qp->usq.buf_addr, qp->usq.buf_len, qp->urq.buf_addr, - qp->urq.buf_len); -} + struct qed_rdma_create_qp_in_params in_params; + struct qedr_pd *pd = get_qedr_pd(ibpd); + int rc = -EINVAL; + u32 n_rq_elems; + u32 n_sq_elems; + u32 n_sq_entries; -static inline int qedr_init_user_qp(struct ib_ucontext *ib_ctx, - struct qedr_dev *dev, - struct qedr_qp *qp, - struct qedr_create_qp_ureq *ureq) -{ - int rc; + memset(&in_params, 0, sizeof(in_params)); - /* SQ - read access only (0), dma sync not required (0) */ - rc = qedr_init_user_queue(ib_ctx, dev, &qp->usq, ureq->sq_addr, - ureq->sq_len, 0, 0); - if (rc) - return rc; + /* A single work request may take up to QEDR_MAX_SQ_WQE_SIZE elements in + * the ring. The ring should allow at least a single WR, even if the + * user requested none, due to allocation issues. + * We should add an extra WR since the prod and cons indices of + * wqe_wr_id are managed in such a way that the WQ is considered full + * when (prod+1)%max_wr==cons. We currently don't do that because we + * double the number of entries due an iSER issue that pushes far more + * WRs than indicated. If we decline its ib_post_send() then we get + * error prints in the dmesg we'd like to avoid. + */ + qp->sq.max_wr = min_t(u32, attrs->cap.max_send_wr * dev->wq_multiplier, + dev->attr.max_sqe); - /* RQ - read access only (0), dma sync not required (0) */ - rc = qedr_init_user_queue(ib_ctx, dev, &qp->urq, ureq->rq_addr, - ureq->rq_len, 0, 0); + qp->wqe_wr_id = kzalloc(qp->sq.max_wr * sizeof(*qp->wqe_wr_id), + GFP_KERNEL); + if (!qp->wqe_wr_id) { + DP_ERR(dev, "create qp: failed SQ shadow memory allocation\n"); + return -ENOMEM; + } - if (rc) - qedr_cleanup_user_sq(dev, qp); - return rc; -} + /* QP handle to be written in CQE */ + in_params.qp_handle_lo = lower_32_bits((uintptr_t) qp); + in_params.qp_handle_hi = upper_32_bits((uintptr_t) qp); -static inline int -qedr_init_kernel_qp(struct qedr_dev *dev, - struct qedr_qp *qp, - struct ib_qp_init_attr *attrs, - struct qed_rdma_create_qp_in_params *params) -{ - int rc; + /* A single work request may take up to QEDR_MAX_RQ_WQE_SIZE elements in + * the ring. There ring should allow at least a single WR, even if the + * user requested none, due to allocation issues. + */ + qp->rq.max_wr = (u16) max_t(u32, attrs->cap.max_recv_wr, 1); - rc = qedr_init_qp_kernel_sq(dev, qp, attrs); - if (rc) { - DP_ERR(dev, "failed to init kernel QP %p SQ\n", qp); - return rc; + /* Allocate driver internal RQ array */ + qp->rqe_wr_id = kzalloc(qp->rq.max_wr * sizeof(*qp->rqe_wr_id), + GFP_KERNEL); + if (!qp->rqe_wr_id) { + DP_ERR(dev, + "create qp: failed RQ shadow memory allocation\n"); + kfree(qp->wqe_wr_id); + return -ENOMEM; } - rc = qedr_init_qp_kernel_params_sq(dev, qp, attrs, params); - if (rc) { - dev->ops->common->chain_free(dev->cdev, &qp->sq.pbl); - DP_ERR(dev, "failed to init kernel QP %p SQ params\n", qp); - return rc; - } + qedr_init_common_qp_in_params(dev, pd, qp, attrs, true, &in_params); - rc = qedr_init_qp_kernel_rq(dev, qp, attrs); - if (rc) { - qedr_cleanup_kernel_sq(dev, qp); - DP_ERR(dev, "failed to init kernel QP %p RQ\n", qp); - return rc; - } + n_sq_entries = attrs->cap.max_send_wr; + n_sq_entries = min_t(u32, n_sq_entries, dev->attr.max_sqe); + n_sq_entries = max_t(u32, n_sq_entries, 1); + n_sq_elems = n_sq_entries * QEDR_MAX_SQE_ELEMENTS_PER_SQE; - rc = qedr_init_qp_kernel_params_rq(dev, qp, attrs); - if (rc) { - DP_ERR(dev, "failed to init kernel QP %p RQ params\n", qp); - qedr_cleanup_kernel_sq(dev, qp); - dev->ops->common->chain_free(dev->cdev, &qp->rq.pbl); - return rc; - } + n_rq_elems = qp->rq.max_wr * QEDR_MAX_RQE_ELEMENTS_PER_RQE; + + rc = qedr_roce_create_kernel_qp(dev, qp, &in_params, + n_sq_elems, n_rq_elems); + if (rc) + qedr_cleanup_kernel(dev, qp); return rc; } @@ -1470,13 +1472,9 @@ struct ib_qp *qedr_create_qp(struct ib_pd *ibpd, struct ib_udata *udata) { struct qedr_dev *dev = get_qedr_dev(ibpd->device); - struct qed_rdma_create_qp_out_params out_params; - struct qed_rdma_create_qp_in_params in_params; struct qedr_pd *pd = get_qedr_pd(ibpd); - struct ib_ucontext *ib_ctx = NULL; - struct qedr_ucontext *ctx = NULL; - struct qedr_create_qp_ureq ureq; struct qedr_qp *qp; + struct ib_qp *ibqp; int rc = 0; DP_DEBUG(dev, QEDR_MSG_QP, "create qp: called from %s, pd=%p\n", @@ -1486,107 +1484,51 @@ struct ib_qp *qedr_create_qp(struct ib_pd *ibpd, if (rc) return ERR_PTR(rc); - qp = kzalloc(sizeof(*qp), GFP_KERNEL); - if (!qp) - return ERR_PTR(-ENOMEM); - if (attrs->srq) return ERR_PTR(-EINVAL); DP_DEBUG(dev, QEDR_MSG_QP, - "create qp: sq_cq=%p, sq_icid=%d, rq_cq=%p, rq_icid=%d\n", + "create qp: called from %s, event_handler=%p, eepd=%p sq_cq=%p, sq_icid=%d, rq_cq=%p, rq_icid=%d\n", + udata ? "user library" : "kernel", attrs->event_handler, pd, get_qedr_cq(attrs->send_cq), get_qedr_cq(attrs->send_cq)->icid, get_qedr_cq(attrs->recv_cq), get_qedr_cq(attrs->recv_cq)->icid); - qedr_set_qp_init_params(dev, qp, pd, attrs); - - if (attrs->qp_type == IB_QPT_GSI) { - if (udata) { - DP_ERR(dev, - "create qp: unexpected udata when creating GSI QP\n"); - goto err0; - } - return qedr_create_gsi_qp(dev, attrs, qp); + qp = kzalloc(sizeof(*qp), GFP_KERNEL); + if (!qp) { + DP_ERR(dev, "create qp: failed allocating memory\n"); + return ERR_PTR(-ENOMEM); } - memset(&in_params, 0, sizeof(in_params)); + qedr_set_common_qp_params(dev, qp, pd, attrs); - if (udata) { - if (!(udata && ibpd->uobject && ibpd->uobject->context)) - goto err0; - - ib_ctx = ibpd->uobject->context; - ctx = get_qedr_ucontext(ib_ctx); - - memset(&ureq, 0, sizeof(ureq)); - if (ib_copy_from_udata(&ureq, udata, sizeof(ureq))) { - DP_ERR(dev, - "create qp: problem copying data from user space\n"); - goto err0; - } - - rc = qedr_init_user_qp(ib_ctx, dev, qp, &ureq); - if (rc) - goto err0; - - qedr_init_qp_user_params(&in_params, &ureq); - } else { - rc = qedr_init_kernel_qp(dev, qp, attrs, &in_params); - if (rc) - goto err0; + if (attrs->qp_type == IB_QPT_GSI) { + ibqp = qedr_create_gsi_qp(dev, attrs, qp); + if (IS_ERR(ibqp)) + kfree(qp); + return ibqp; } - qedr_init_qp_in_params_sq(dev, pd, qp, attrs, udata, &in_params); - qedr_init_qp_in_params_rq(qp, attrs, udata, &in_params); - - qp->qed_qp = dev->ops->rdma_create_qp(dev->rdma_ctx, - &in_params, &out_params); + if (udata) + rc = qedr_create_user_qp(dev, qp, ibpd, udata, attrs); + else + rc = qedr_create_kernel_qp(dev, qp, ibpd, attrs); - if (!qp->qed_qp) - goto err1; + if (rc) + goto err; - qp->qp_id = out_params.qp_id; - qp->icid = out_params.icid; qp->ibqp.qp_num = qp->qp_id; - if (udata) { - rc = qedr_copy_qp_uresp(dev, qp, udata); - if (rc) - goto err2; - - qedr_qp_user_print(dev, qp); - } else { - qedr_init_qp_kernel_doorbell_sq(dev, qp); - qedr_init_qp_kernel_doorbell_rq(dev, qp); - } - - DP_DEBUG(dev, QEDR_MSG_QP, "created %s space QP %p\n", - udata ? "user" : "kernel", qp); - return &qp->ibqp; -err2: - rc = dev->ops->rdma_destroy_qp(dev->rdma_ctx, qp->qed_qp); - if (rc) - DP_ERR(dev, "create qp: fatal fault. rc=%d", rc); -err1: - if (udata) { - qedr_cleanup_user_sq(dev, qp); - qedr_cleanup_user_rq(dev, qp); - } else { - qedr_cleanup_kernel_sq(dev, qp); - qedr_cleanup_kernel_rq(dev, qp); - } - -err0: +err: kfree(qp); return ERR_PTR(-EFAULT); } -enum ib_qp_state qedr_get_ibqp_state(enum qed_roce_qp_state qp_state) +static enum ib_qp_state qedr_get_ibqp_state(enum qed_roce_qp_state qp_state) { switch (qp_state) { case QED_ROCE_QP_STATE_RESET: @@ -1607,7 +1549,8 @@ enum ib_qp_state qedr_get_ibqp_state(enum qed_roce_qp_state qp_state) return IB_QPS_ERR; } -enum qed_roce_qp_state qedr_get_state_from_ibqp(enum ib_qp_state qp_state) +static enum qed_roce_qp_state qedr_get_state_from_ibqp( + enum ib_qp_state qp_state) { switch (qp_state) { case IB_QPS_RESET: @@ -1643,7 +1586,7 @@ static int qedr_update_qp_state(struct qedr_dev *dev, int status = 0; if (new_state == qp->state) - return 1; + return 0; switch (qp->state) { case QED_ROCE_QP_STATE_RESET: @@ -1719,6 +1662,14 @@ static int qedr_update_qp_state(struct qedr_dev *dev, /* ERR->XXX */ switch (new_state) { case QED_ROCE_QP_STATE_RESET: + if ((qp->rq.prod != qp->rq.cons) || + (qp->sq.prod != qp->sq.cons)) { + DP_NOTICE(dev, + "Error->Reset with rq/sq not empty rq.prod=%x rq.cons=%x sq.prod=%x sq.cons=%x\n", + qp->rq.prod, qp->rq.cons, qp->sq.prod, + qp->sq.cons); + status = -EINVAL; + } break; default: status = -EINVAL; @@ -1851,7 +1802,6 @@ int qedr_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, qp_params.sgid.dwords[2], qp_params.sgid.dwords[3]); DP_DEBUG(dev, QEDR_MSG_QP, "remote_mac=[%pM]\n", qp_params.remote_mac_addr); -; qp_params.mtu = qp->mtu; qp_params.lb_indication = false; @@ -1960,7 +1910,7 @@ int qedr_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, if (attr_mask & IB_QP_STATE) { if ((qp->qp_type != IB_QPT_GSI) && (!udata)) - qedr_update_qp_state(dev, qp, qp_params.new_state); + rc = qedr_update_qp_state(dev, qp, qp_params.new_state); qp->state = qp_params.new_state; } @@ -2002,7 +1952,7 @@ int qedr_query_qp(struct ib_qp *ibqp, qp_attr->qp_state = qedr_get_ibqp_state(params.state); qp_attr->cur_qp_state = qedr_get_ibqp_state(params.state); - qp_attr->path_mtu = iboe_get_mtu(params.mtu); + qp_attr->path_mtu = ib_mtu_int_to_enum(params.mtu); qp_attr->path_mig_state = IB_MIG_MIGRATED; qp_attr->rq_psn = params.rq_psn; qp_attr->sq_psn = params.sq_psn; @@ -2014,7 +1964,7 @@ int qedr_query_qp(struct ib_qp *ibqp, qp_attr->cap.max_recv_wr = qp->rq.max_wr; qp_attr->cap.max_send_sge = qp->sq.max_sges; qp_attr->cap.max_recv_sge = qp->rq.max_sges; - qp_attr->cap.max_inline_data = qp->max_inline_data; + qp_attr->cap.max_inline_data = ROCE_REQ_MAX_INLINE_DATA_SIZE; qp_init_attr->cap = qp_attr->cap; memcpy(&qp_attr->ah_attr.grh.dgid.raw[0], ¶ms.dgid.bytes[0], @@ -2053,6 +2003,24 @@ err: return rc; } +int qedr_free_qp_resources(struct qedr_dev *dev, struct qedr_qp *qp) +{ + int rc = 0; + + if (qp->qp_type != IB_QPT_GSI) { + rc = dev->ops->rdma_destroy_qp(dev->rdma_ctx, qp->qed_qp); + if (rc) + return rc; + } + + if (qp->ibqp.uobject && qp->ibqp.uobject->context) + qedr_cleanup_user(dev, qp); + else + qedr_cleanup_kernel(dev, qp); + + return 0; +} + int qedr_destroy_qp(struct ib_qp *ibqp) { struct qedr_qp *qp = get_qedr_qp(ibqp); @@ -2064,8 +2032,10 @@ int qedr_destroy_qp(struct ib_qp *ibqp) DP_DEBUG(dev, QEDR_MSG_QP, "destroy qp: destroying %p, qp type=%d\n", qp, qp->qp_type); - if (qp->state != (QED_ROCE_QP_STATE_RESET | QED_ROCE_QP_STATE_ERR | - QED_ROCE_QP_STATE_INIT)) { + if ((qp->state != QED_ROCE_QP_STATE_RESET) && + (qp->state != QED_ROCE_QP_STATE_ERR) && + (qp->state != QED_ROCE_QP_STATE_INIT)) { + attr.qp_state = IB_QPS_ERR; attr_mask |= IB_QP_STATE; @@ -2073,28 +2043,18 @@ int qedr_destroy_qp(struct ib_qp *ibqp) qedr_modify_qp(ibqp, &attr, attr_mask, NULL); } - if (qp->qp_type != IB_QPT_GSI) { - rc = dev->ops->rdma_destroy_qp(dev->rdma_ctx, qp->qed_qp); - if (rc) - return rc; - } else { + if (qp->qp_type == IB_QPT_GSI) qedr_destroy_gsi_qp(dev); - } - if (ibqp->uobject && ibqp->uobject->context) { - qedr_cleanup_user_sq(dev, qp); - qedr_cleanup_user_rq(dev, qp); - } else { - qedr_cleanup_kernel_sq(dev, qp); - qedr_cleanup_kernel_rq(dev, qp); - } + qedr_free_qp_resources(dev, qp); kfree(qp); return rc; } -struct ib_ah *qedr_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *attr) +struct ib_ah *qedr_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *attr, + struct ib_udata *udata) { struct qedr_ah *ah; @@ -2147,8 +2107,8 @@ static int init_mr_info(struct qedr_dev *dev, struct mr_info *info, goto done; info->pbl_table = qedr_alloc_pbl_tbl(dev, &info->pbl_info, GFP_KERNEL); - if (!info->pbl_table) { - rc = -ENOMEM; + if (IS_ERR(info->pbl_table)) { + rc = PTR_ERR(info->pbl_table); goto done; } @@ -2159,7 +2119,7 @@ static int init_mr_info(struct qedr_dev *dev, struct mr_info *info, * list and allocating another one */ tmp = qedr_alloc_pbl_tbl(dev, &info->pbl_info, GFP_KERNEL); - if (!tmp) { + if (IS_ERR(tmp)) { DP_DEBUG(dev, QEDR_MSG_MR, "Extra PBL is not allocated\n"); goto done; } @@ -2285,7 +2245,8 @@ int qedr_dereg_mr(struct ib_mr *ib_mr) return rc; } -struct qedr_mr *__qedr_alloc_mr(struct ib_pd *ibpd, int max_page_list_len) +static struct qedr_mr *__qedr_alloc_mr(struct ib_pd *ibpd, + int max_page_list_len) { struct qedr_pd *pd = get_qedr_pd(ibpd); struct qedr_dev *dev = get_qedr_dev(ibpd->device); @@ -2413,8 +2374,7 @@ static void handle_completed_mrs(struct qedr_dev *dev, struct mr_info *info) */ pbl = list_first_entry(&info->inuse_pbl_list, struct qedr_pbl, list_entry); - list_del(&pbl->list_entry); - list_add_tail(&pbl->list_entry, &info->free_pbl_list); + list_move_tail(&pbl->list_entry, &info->free_pbl_list); info->completed_handled++; } } @@ -2620,7 +2580,9 @@ static u32 qedr_prepare_sq_rdma_data(struct qedr_dev *dev, rwqe2->r_key = cpu_to_le32(rdma_wr(wr)->rkey); DMA_REGPAIR_LE(rwqe2->remote_va, rdma_wr(wr)->remote_addr); - if (wr->send_flags & IB_SEND_INLINE) { + if (wr->send_flags & IB_SEND_INLINE && + (wr->opcode == IB_WR_RDMA_WRITE_WITH_IMM || + wr->opcode == IB_WR_RDMA_WRITE)) { u8 flags = 0; SET_FIELD2(flags, RDMA_SQ_RDMA_WQE_1ST_INLINE_FLG, 1); @@ -2686,7 +2648,7 @@ static int qedr_prepare_reg(struct qedr_qp *qp, return 0; } -enum ib_wc_opcode qedr_ib_to_wc_opcode(enum ib_wr_opcode opcode) +static enum ib_wc_opcode qedr_ib_to_wc_opcode(enum ib_wr_opcode opcode) { switch (opcode) { case IB_WR_RDMA_WRITE: @@ -2711,7 +2673,7 @@ enum ib_wc_opcode qedr_ib_to_wc_opcode(enum ib_wr_opcode opcode) } } -inline bool qedr_can_post_send(struct qedr_qp *qp, struct ib_send_wr *wr) +static inline bool qedr_can_post_send(struct qedr_qp *qp, struct ib_send_wr *wr) { int wq_is_full, err_wr, pbl_is_full; struct qedr_dev *dev = qp->dev; @@ -2748,7 +2710,7 @@ inline bool qedr_can_post_send(struct qedr_qp *qp, struct ib_send_wr *wr) return true; } -int __qedr_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, +static int __qedr_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, struct ib_send_wr **bad_wr) { struct qedr_dev *dev = get_qedr_dev(ibqp->device); @@ -2971,8 +2933,9 @@ int qedr_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, spin_lock_irqsave(&qp->q_lock, flags); - if ((qp->state == QED_ROCE_QP_STATE_RESET) || - (qp->state == QED_ROCE_QP_STATE_ERR)) { + if ((qp->state != QED_ROCE_QP_STATE_RTS) && + (qp->state != QED_ROCE_QP_STATE_ERR) && + (qp->state != QED_ROCE_QP_STATE_SQD)) { spin_unlock_irqrestore(&qp->q_lock, flags); *bad_wr = wr; DP_DEBUG(dev, QEDR_MSG_CQ, @@ -2981,11 +2944,6 @@ int qedr_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, return -EINVAL; } - if (!wr) { - DP_ERR(dev, "Got an empty post send.\n"); - return -EINVAL; - } - while (wr) { rc = __qedr_post_send(ibqp, wr, bad_wr); if (rc) @@ -3030,8 +2988,7 @@ int qedr_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr, spin_lock_irqsave(&qp->q_lock, flags); - if ((qp->state == QED_ROCE_QP_STATE_RESET) || - (qp->state == QED_ROCE_QP_STATE_ERR)) { + if (qp->state == QED_ROCE_QP_STATE_RESET) { spin_unlock_irqrestore(&qp->q_lock, flags); *bad_wr = wr; return -EINVAL; @@ -3173,6 +3130,7 @@ static int process_req(struct qedr_dev *dev, struct qedr_qp *qp, /* fill WC */ wc->status = status; + wc->vendor_err = 0; wc->wc_flags = 0; wc->src_qp = qp->id; wc->qp = &qp->ibqp; @@ -3220,11 +3178,12 @@ static int qedr_poll_cq_req(struct qedr_dev *dev, IB_WC_SUCCESS, 0); break; case RDMA_CQE_REQ_STS_WORK_REQUEST_FLUSHED_ERR: - DP_ERR(dev, - "Error: POLL CQ with RDMA_CQE_REQ_STS_WORK_REQUEST_FLUSHED_ERR. CQ icid=0x%x, QP icid=0x%x\n", - cq->icid, qp->icid); + if (qp->state != QED_ROCE_QP_STATE_ERR) + DP_ERR(dev, + "Error: POLL CQ with RDMA_CQE_REQ_STS_WORK_REQUEST_FLUSHED_ERR. CQ icid=0x%x, QP icid=0x%x\n", + cq->icid, qp->icid); cnt = process_req(dev, qp, cq, num_entries, wc, req->sq_cons, - IB_WC_WR_FLUSH_ERR, 0); + IB_WC_WR_FLUSH_ERR, 1); break; default: /* process all WQE before the cosumer */ @@ -3362,6 +3321,7 @@ static void __process_resp_one(struct qedr_dev *dev, struct qedr_qp *qp, /* fill WC */ wc->status = wc_status; + wc->vendor_err = 0; wc->src_qp = qp->id; wc->qp = &qp->ibqp; wc->wr_id = wr_id; @@ -3390,6 +3350,7 @@ static int process_resp_flush(struct qedr_qp *qp, struct qedr_cq *cq, while (num_entries && qp->rq.wqe_cons != hw_cons) { /* fill WC */ wc->status = IB_WC_WR_FLUSH_ERR; + wc->vendor_err = 0; wc->wc_flags = 0; wc->src_qp = qp->id; wc->byte_len = 0; @@ -3533,14 +3494,15 @@ int qedr_port_immutable(struct ib_device *ibdev, u8 port_num, struct ib_port_attr attr; int err; - err = qedr_query_port(ibdev, port_num, &attr); + immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE | + RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP; + + err = ib_query_port(ibdev, port_num, &attr); if (err) return err; immutable->pkey_tbl_len = attr.pkey_tbl_len; immutable->gid_tbl_len = attr.gid_tbl_len; - immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE | - RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP; immutable->max_mad_size = IB_MGMT_MAD_SIZE; return 0; diff --git a/drivers/infiniband/hw/qedr/verbs.h b/drivers/infiniband/hw/qedr/verbs.h index a9b5e67bb81e..070677ca4d19 100644 --- a/drivers/infiniband/hw/qedr/verbs.h +++ b/drivers/infiniband/hw/qedr/verbs.h @@ -70,7 +70,8 @@ int qedr_query_qp(struct ib_qp *, struct ib_qp_attr *qp_attr, int qp_attr_mask, struct ib_qp_init_attr *); int qedr_destroy_qp(struct ib_qp *ibqp); -struct ib_ah *qedr_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *attr); +struct ib_ah *qedr_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *attr, + struct ib_udata *udata); int qedr_destroy_ah(struct ib_ah *ibah); int qedr_dereg_mr(struct ib_mr *); diff --git a/drivers/infiniband/hw/qib/qib_common.h b/drivers/infiniband/hw/qib/qib_common.h index 1d6e63eb1146..a4a1f56ce824 100644 --- a/drivers/infiniband/hw/qib/qib_common.h +++ b/drivers/infiniband/hw/qib/qib_common.h @@ -742,11 +742,7 @@ struct qib_tid_session_member { #define SIZE_OF_CRC 1 #define QIB_DEFAULT_P_KEY 0xFFFF -#define QIB_AETH_CREDIT_SHIFT 24 -#define QIB_AETH_CREDIT_MASK 0x1F -#define QIB_AETH_CREDIT_INVAL 0x1F #define QIB_PSN_MASK 0xFFFFFF -#define QIB_MSN_MASK 0xFFFFFF #define QIB_EAGER_TID_ID QLOGIC_IB_I_TID_MASK #define QIB_MULTICAST_QPN 0xFFFFFF diff --git a/drivers/infiniband/hw/qib/qib_diag.c b/drivers/infiniband/hw/qib/qib_diag.c index 8c34b23e5bf6..775018b32b0d 100644 --- a/drivers/infiniband/hw/qib/qib_diag.c +++ b/drivers/infiniband/hw/qib/qib_diag.c @@ -609,8 +609,6 @@ static ssize_t qib_diagpkt_write(struct file *fp, tmpbuf = vmalloc(plen); if (!tmpbuf) { - qib_devinfo(dd->pcidev, - "Unable to allocate tmp buffer, failing\n"); ret = -ENOMEM; goto bail; } @@ -702,10 +700,8 @@ int qib_register_observer(struct qib_devdata *dd, if (!dd || !op) return -EINVAL; olp = vmalloc(sizeof(*olp)); - if (!olp) { - pr_err("vmalloc for observer failed\n"); + if (!olp) return -ENOMEM; - } spin_lock_irqsave(&dd->qib_diag_trans_lock, flags); olp->op = op; diff --git a/drivers/infiniband/hw/qib/qib_dma.c b/drivers/infiniband/hw/qib/qib_dma.c deleted file mode 100644 index 59fe092b4b0f..000000000000 --- a/drivers/infiniband/hw/qib/qib_dma.c +++ /dev/null @@ -1,169 +0,0 @@ -/* - * Copyright (c) 2006, 2009, 2010 QLogic, Corporation. All rights reserved. - * - * This software is available to you under a choice of one of two - * licenses. You may choose to be licensed under the terms of the GNU - * General Public License (GPL) Version 2, available from the file - * COPYING in the main directory of this source tree, or the - * OpenIB.org BSD license below: - * - * Redistribution and use in source and binary forms, with or - * without modification, are permitted provided that the following - * conditions are met: - * - * - Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. - * - * - Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, - * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF - * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND - * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS - * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN - * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN - * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ -#include <linux/types.h> -#include <linux/scatterlist.h> - -#include "qib_verbs.h" - -#define BAD_DMA_ADDRESS ((u64) 0) - -/* - * The following functions implement driver specific replacements - * for the ib_dma_*() functions. - * - * These functions return kernel virtual addresses instead of - * device bus addresses since the driver uses the CPU to copy - * data instead of using hardware DMA. - */ - -static int qib_mapping_error(struct ib_device *dev, u64 dma_addr) -{ - return dma_addr == BAD_DMA_ADDRESS; -} - -static u64 qib_dma_map_single(struct ib_device *dev, void *cpu_addr, - size_t size, enum dma_data_direction direction) -{ - BUG_ON(!valid_dma_direction(direction)); - return (u64) cpu_addr; -} - -static void qib_dma_unmap_single(struct ib_device *dev, u64 addr, size_t size, - enum dma_data_direction direction) -{ - BUG_ON(!valid_dma_direction(direction)); -} - -static u64 qib_dma_map_page(struct ib_device *dev, struct page *page, - unsigned long offset, size_t size, - enum dma_data_direction direction) -{ - u64 addr; - - BUG_ON(!valid_dma_direction(direction)); - - if (offset + size > PAGE_SIZE) { - addr = BAD_DMA_ADDRESS; - goto done; - } - - addr = (u64) page_address(page); - if (addr) - addr += offset; - /* TODO: handle highmem pages */ - -done: - return addr; -} - -static void qib_dma_unmap_page(struct ib_device *dev, u64 addr, size_t size, - enum dma_data_direction direction) -{ - BUG_ON(!valid_dma_direction(direction)); -} - -static int qib_map_sg(struct ib_device *dev, struct scatterlist *sgl, - int nents, enum dma_data_direction direction) -{ - struct scatterlist *sg; - u64 addr; - int i; - int ret = nents; - - BUG_ON(!valid_dma_direction(direction)); - - for_each_sg(sgl, sg, nents, i) { - addr = (u64) page_address(sg_page(sg)); - /* TODO: handle highmem pages */ - if (!addr) { - ret = 0; - break; - } - sg->dma_address = addr + sg->offset; -#ifdef CONFIG_NEED_SG_DMA_LENGTH - sg->dma_length = sg->length; -#endif - } - return ret; -} - -static void qib_unmap_sg(struct ib_device *dev, - struct scatterlist *sg, int nents, - enum dma_data_direction direction) -{ - BUG_ON(!valid_dma_direction(direction)); -} - -static void qib_sync_single_for_cpu(struct ib_device *dev, u64 addr, - size_t size, enum dma_data_direction dir) -{ -} - -static void qib_sync_single_for_device(struct ib_device *dev, u64 addr, - size_t size, - enum dma_data_direction dir) -{ -} - -static void *qib_dma_alloc_coherent(struct ib_device *dev, size_t size, - u64 *dma_handle, gfp_t flag) -{ - struct page *p; - void *addr = NULL; - - p = alloc_pages(flag, get_order(size)); - if (p) - addr = page_address(p); - if (dma_handle) - *dma_handle = (u64) addr; - return addr; -} - -static void qib_dma_free_coherent(struct ib_device *dev, size_t size, - void *cpu_addr, u64 dma_handle) -{ - free_pages((unsigned long) cpu_addr, get_order(size)); -} - -struct ib_dma_mapping_ops qib_dma_mapping_ops = { - .mapping_error = qib_mapping_error, - .map_single = qib_dma_map_single, - .unmap_single = qib_dma_unmap_single, - .map_page = qib_dma_map_page, - .unmap_page = qib_dma_unmap_page, - .map_sg = qib_map_sg, - .unmap_sg = qib_unmap_sg, - .sync_single_for_cpu = qib_sync_single_for_cpu, - .sync_single_for_device = qib_sync_single_for_device, - .alloc_coherent = qib_dma_alloc_coherent, - .free_coherent = qib_dma_free_coherent -}; diff --git a/drivers/infiniband/hw/qib/qib_driver.c b/drivers/infiniband/hw/qib/qib_driver.c index 728e0a030d2e..2b5982f743ef 100644 --- a/drivers/infiniband/hw/qib/qib_driver.c +++ b/drivers/infiniband/hw/qib/qib_driver.c @@ -420,8 +420,7 @@ static u32 qib_rcv_hdrerr(struct qib_ctxtdata *rcd, struct qib_pportdata *ppd, if (list_empty(&qp->rspwait)) { qp->r_flags |= RVT_R_RSP_NAK; - atomic_inc( - &qp->refcount); + rvt_get_qp(qp); list_add_tail( &qp->rspwait, &rcd->qp_wait_list); diff --git a/drivers/infiniband/hw/qib/qib_eeprom.c b/drivers/infiniband/hw/qib/qib_eeprom.c index 311ee6c3dd5e..33a2e74c8495 100644 --- a/drivers/infiniband/hw/qib/qib_eeprom.c +++ b/drivers/infiniband/hw/qib/qib_eeprom.c @@ -182,12 +182,8 @@ void qib_get_eeprom_info(struct qib_devdata *dd) * */ len = sizeof(struct qib_flash); buf = vmalloc(len); - if (!buf) { - qib_dev_err(dd, - "Couldn't allocate memory to read %u bytes from eeprom for GUID\n", - len); + if (!buf) goto bail; - } /* * Use "public" eeprom read function, which does locking and diff --git a/drivers/infiniband/hw/qib/qib_file_ops.c b/drivers/infiniband/hw/qib/qib_file_ops.c index 382466a90da7..9396c1807cc3 100644 --- a/drivers/infiniband/hw/qib/qib_file_ops.c +++ b/drivers/infiniband/hw/qib/qib_file_ops.c @@ -893,7 +893,7 @@ bail: /* * qib_file_vma_fault - handle a VMA page fault. */ -static int qib_file_vma_fault(struct vm_area_struct *vma, struct vm_fault *vmf) +static int qib_file_vma_fault(struct vm_fault *vmf) { struct page *page; @@ -2066,8 +2066,11 @@ static ssize_t qib_write(struct file *fp, const char __user *data, ssize_t ret = 0; void *dest; - if (WARN_ON_ONCE(!ib_safe_file_access(fp))) + if (!ib_safe_file_access(fp)) { + pr_err_once("qib_write: process %d (%s) changed security contexts after opening file descriptor, this is not allowed.\n", + task_tgid_vnr(current), current->comm); return -EACCES; + } if (count < sizeof(cmd.type)) { ret = -EINVAL; diff --git a/drivers/infiniband/hw/qib/qib_iba6120.c b/drivers/infiniband/hw/qib/qib_iba6120.c index a3733f25280f..92399d3ffd15 100644 --- a/drivers/infiniband/hw/qib/qib_iba6120.c +++ b/drivers/infiniband/hw/qib/qib_iba6120.c @@ -1759,9 +1759,7 @@ static void pe_boardname(struct qib_devdata *dd) } namelen = strlen(n) + 1; dd->boardname = kmalloc(namelen, GFP_KERNEL); - if (!dd->boardname) - qib_dev_err(dd, "Failed allocation for board name: %s\n", n); - else + if (dd->boardname) snprintf(dd->boardname, namelen, "%s", n); if (dd->majrev != 4 || !dd->minrev || dd->minrev > 2) @@ -2533,8 +2531,6 @@ static void init_6120_cntrnames(struct qib_devdata *dd) dd->cspec->cntrnamelen = 1 + s - cntr6120names; dd->cspec->cntrs = kmalloc(dd->cspec->ncntrs * sizeof(u64), GFP_KERNEL); - if (!dd->cspec->cntrs) - qib_dev_err(dd, "Failed allocation for counters\n"); for (i = 0, s = (char *)portcntr6120names; s; i++) s = strchr(s + 1, '\n'); @@ -2542,8 +2538,6 @@ static void init_6120_cntrnames(struct qib_devdata *dd) dd->cspec->portcntrnamelen = sizeof(portcntr6120names) - 1; dd->cspec->portcntrs = kmalloc(dd->cspec->nportcntrs * sizeof(u64), GFP_KERNEL); - if (!dd->cspec->portcntrs) - qib_dev_err(dd, "Failed allocation for portcounters\n"); } static u32 qib_read_6120cntrs(struct qib_devdata *dd, loff_t pos, char **namep, diff --git a/drivers/infiniband/hw/qib/qib_iba7220.c b/drivers/infiniband/hw/qib/qib_iba7220.c index 00b2af211157..e55e31a69195 100644 --- a/drivers/infiniband/hw/qib/qib_iba7220.c +++ b/drivers/infiniband/hw/qib/qib_iba7220.c @@ -2070,9 +2070,7 @@ static void qib_7220_boardname(struct qib_devdata *dd) namelen = strlen(n) + 1; dd->boardname = kmalloc(namelen, GFP_KERNEL); - if (!dd->boardname) - qib_dev_err(dd, "Failed allocation for board name: %s\n", n); - else + if (dd->boardname) snprintf(dd->boardname, namelen, "%s", n); if (dd->majrev != 5 || !dd->minrev || dd->minrev > 2) @@ -3179,8 +3177,6 @@ static void init_7220_cntrnames(struct qib_devdata *dd) dd->cspec->cntrnamelen = 1 + s - cntr7220names; dd->cspec->cntrs = kmalloc(dd->cspec->ncntrs * sizeof(u64), GFP_KERNEL); - if (!dd->cspec->cntrs) - qib_dev_err(dd, "Failed allocation for counters\n"); for (i = 0, s = (char *)portcntr7220names; s; i++) s = strchr(s + 1, '\n'); @@ -3188,8 +3184,6 @@ static void init_7220_cntrnames(struct qib_devdata *dd) dd->cspec->portcntrnamelen = sizeof(portcntr7220names) - 1; dd->cspec->portcntrs = kmalloc(dd->cspec->nportcntrs * sizeof(u64), GFP_KERNEL); - if (!dd->cspec->portcntrs) - qib_dev_err(dd, "Failed allocation for portcounters\n"); } static u32 qib_read_7220cntrs(struct qib_devdata *dd, loff_t pos, char **namep, diff --git a/drivers/infiniband/hw/qib/qib_iba7322.c b/drivers/infiniband/hw/qib/qib_iba7322.c index ded27172320e..9cc97bd42775 100644 --- a/drivers/infiniband/hw/qib/qib_iba7322.c +++ b/drivers/infiniband/hw/qib/qib_iba7322.c @@ -2893,7 +2893,6 @@ static void qib_setup_7322_cleanup(struct qib_devdata *dd) dd->cspec->gpio_mask &= ~mask; qib_write_kreg(dd, kr_gpio_mask, dd->cspec->gpio_mask); spin_unlock_irqrestore(&dd->cspec->gpio_lock, flags); - qib_qsfp_deinit(&dd->pport[i].cpspec->qsfp_data); } } } @@ -3627,9 +3626,7 @@ static unsigned qib_7322_boardname(struct qib_devdata *dd) namelen = strlen(n) + 1; dd->boardname = kmalloc(namelen, GFP_KERNEL); - if (!dd->boardname) - qib_dev_err(dd, "Failed allocation for board name: %s\n", n); - else + if (dd->boardname) snprintf(dd->boardname, namelen, "%s", n); snprintf(dd->boardversion, sizeof(dd->boardversion), @@ -3656,7 +3653,7 @@ static unsigned qib_7322_boardname(struct qib_devdata *dd) static int qib_do_7322_reset(struct qib_devdata *dd) { u64 val; - u64 *msix_vecsave; + u64 *msix_vecsave = NULL; int i, msix_entries, ret = 1; u16 cmdval; u8 int_line, clinesz; @@ -3677,10 +3674,7 @@ static int qib_do_7322_reset(struct qib_devdata *dd) /* can be up to 512 bytes, too big for stack */ msix_vecsave = kmalloc(2 * dd->cspec->num_msix_entries * sizeof(u64), GFP_KERNEL); - if (!msix_vecsave) - qib_dev_err(dd, "No mem to save MSIx data\n"); - } else - msix_vecsave = NULL; + } /* * Core PCI (as of 2.6.18) doesn't save or rewrite the full vector @@ -5043,8 +5037,6 @@ static void init_7322_cntrnames(struct qib_devdata *dd) dd->cspec->cntrnamelen = 1 + s - cntr7322names; dd->cspec->cntrs = kmalloc(dd->cspec->ncntrs * sizeof(u64), GFP_KERNEL); - if (!dd->cspec->cntrs) - qib_dev_err(dd, "Failed allocation for counters\n"); for (i = 0, s = (char *)portcntr7322names; s; i++) s = strchr(s + 1, '\n'); @@ -5053,9 +5045,6 @@ static void init_7322_cntrnames(struct qib_devdata *dd) for (i = 0; i < dd->num_pports; ++i) { dd->pport[i].cpspec->portcntrs = kmalloc(dd->cspec->nportcntrs * sizeof(u64), GFP_KERNEL); - if (!dd->pport[i].cpspec->portcntrs) - qib_dev_err(dd, - "Failed allocation for portcounters\n"); } } @@ -6461,7 +6450,6 @@ static int qib_init_7322_variables(struct qib_devdata *dd) sizeof(*dd->cspec->sendibchk), GFP_KERNEL); if (!dd->cspec->sendchkenable || !dd->cspec->sendgrhchk || !dd->cspec->sendibchk) { - qib_dev_err(dd, "Failed allocation for hdrchk bitmaps\n"); ret = -ENOMEM; goto bail; } @@ -7338,10 +7326,9 @@ struct qib_devdata *qib_init_iba7322_funcs(struct pci_dev *pdev, tabsize = actual_cnt; dd->cspec->msix_entries = kzalloc(tabsize * sizeof(struct qib_msix_entry), GFP_KERNEL); - if (!dd->cspec->msix_entries) { - qib_dev_err(dd, "No memory for MSIx table\n"); + if (!dd->cspec->msix_entries) tabsize = 0; - } + for (i = 0; i < tabsize; i++) dd->cspec->msix_entries[i].msix.entry = i; diff --git a/drivers/infiniband/hw/qib/qib_init.c b/drivers/infiniband/hw/qib/qib_init.c index 1730aa839a47..b50240b1d5a4 100644 --- a/drivers/infiniband/hw/qib/qib_init.c +++ b/drivers/infiniband/hw/qib/qib_init.c @@ -133,11 +133,8 @@ int qib_create_ctxts(struct qib_devdata *dd) * cleanup iterates across all possible ctxts. */ dd->rcd = kcalloc(dd->ctxtcnt, sizeof(*dd->rcd), GFP_KERNEL); - if (!dd->rcd) { - qib_dev_err(dd, - "Unable to allocate ctxtdata array, failing\n"); + if (!dd->rcd) return -ENOMEM; - } /* create (one or more) kctxt */ for (i = 0; i < dd->first_user_ctxt; ++i) { @@ -265,39 +262,23 @@ int qib_init_pportdata(struct qib_pportdata *ppd, struct qib_devdata *dd, size = IB_CC_TABLE_CAP_DEFAULT * sizeof(struct ib_cc_table_entry) * IB_CCT_ENTRIES; ppd->ccti_entries = kzalloc(size, GFP_KERNEL); - if (!ppd->ccti_entries) { - qib_dev_err(dd, - "failed to allocate congestion control table for port %d!\n", - port); + if (!ppd->ccti_entries) goto bail; - } size = IB_CC_CCS_ENTRIES * sizeof(struct ib_cc_congestion_entry); ppd->congestion_entries = kzalloc(size, GFP_KERNEL); - if (!ppd->congestion_entries) { - qib_dev_err(dd, - "failed to allocate congestion setting list for port %d!\n", - port); + if (!ppd->congestion_entries) goto bail_1; - } size = sizeof(struct cc_table_shadow); ppd->ccti_entries_shadow = kzalloc(size, GFP_KERNEL); - if (!ppd->ccti_entries_shadow) { - qib_dev_err(dd, - "failed to allocate shadow ccti list for port %d!\n", - port); + if (!ppd->ccti_entries_shadow) goto bail_2; - } size = sizeof(struct ib_cc_congestion_setting_attr); ppd->congestion_entries_shadow = kzalloc(size, GFP_KERNEL); - if (!ppd->congestion_entries_shadow) { - qib_dev_err(dd, - "failed to allocate shadow congestion setting list for port %d!\n", - port); + if (!ppd->congestion_entries_shadow) goto bail_3; - } return 0; @@ -391,18 +372,12 @@ static void init_shadow_tids(struct qib_devdata *dd) dma_addr_t *addrs; pages = vzalloc(dd->cfgctxts * dd->rcvtidcnt * sizeof(struct page *)); - if (!pages) { - qib_dev_err(dd, - "failed to allocate shadow page * array, no expected sends!\n"); + if (!pages) goto bail; - } addrs = vzalloc(dd->cfgctxts * dd->rcvtidcnt * sizeof(dma_addr_t)); - if (!addrs) { - qib_dev_err(dd, - "failed to allocate shadow dma handle array, no expected sends!\n"); + if (!addrs) goto bail_free; - } dd->pageshadow = pages; dd->physshadow = addrs; @@ -1026,11 +1001,8 @@ static void qib_verify_pioperf(struct qib_devdata *dd) cnt = 1024; addr = vmalloc(cnt); - if (!addr) { - qib_devinfo(dd->pcidev, - "Couldn't get memory for checking PIO perf, skipping\n"); + if (!addr) goto done; - } preempt_disable(); /* we want reasonably accurate elapsed time */ msecs = 1 + jiffies_to_msecs(jiffies); @@ -1172,9 +1144,6 @@ struct qib_devdata *qib_alloc_devdata(struct pci_dev *pdev, size_t extra) sizeof(long), GFP_KERNEL); if (qib_cpulist) qib_cpulist_count = count; - else - qib_early_err(&pdev->dev, - "Could not alloc cpulist info, cpu affinity might be wrong\n"); } #ifdef CONFIG_DEBUG_FS qib_dbg_ibdev_init(&dd->verbs_dev); diff --git a/drivers/infiniband/hw/qib/qib_keys.c b/drivers/infiniband/hw/qib/qib_keys.c index 2c3c93572c17..8fdf79f8d4e4 100644 --- a/drivers/infiniband/hw/qib/qib_keys.c +++ b/drivers/infiniband/hw/qib/qib_keys.c @@ -158,10 +158,7 @@ int qib_rkey_ok(struct rvt_qp *qp, struct rvt_sge *sge, unsigned n, m; size_t off; - /* - * We use RKEY == zero for kernel virtual addresses - * (see qib_get_dma_mr and qib_dma.c). - */ + /* We use RKEY == zero for kernel virtual addresses */ rcu_read_lock(); if (rkey == 0) { struct rvt_pd *pd = ibpd_to_rvtpd(qp->ibqp.pd); diff --git a/drivers/infiniband/hw/qib/qib_pcie.c b/drivers/infiniband/hw/qib/qib_pcie.c index 6abe1c621aa4..c379b8342a09 100644 --- a/drivers/infiniband/hw/qib/qib_pcie.c +++ b/drivers/infiniband/hw/qib/qib_pcie.c @@ -682,13 +682,6 @@ qib_pci_slot_reset(struct pci_dev *pdev) return PCI_ERS_RESULT_CAN_RECOVER; } -static pci_ers_result_t -qib_pci_link_reset(struct pci_dev *pdev) -{ - qib_devinfo(pdev, "QIB link_reset function called, ignored\n"); - return PCI_ERS_RESULT_CAN_RECOVER; -} - static void qib_pci_resume(struct pci_dev *pdev) { @@ -707,7 +700,6 @@ qib_pci_resume(struct pci_dev *pdev) const struct pci_error_handlers qib_pci_err_handler = { .error_detected = qib_pci_error_detected, .mmio_enabled = qib_pci_mmio_enabled, - .link_reset = qib_pci_link_reset, .slot_reset = qib_pci_slot_reset, .resume = qib_pci_resume, }; diff --git a/drivers/infiniband/hw/qib/qib_qp.c b/drivers/infiniband/hw/qib/qib_qp.c index 99d31efe4c2f..2ac0c0f79e74 100644 --- a/drivers/infiniband/hw/qib/qib_qp.c +++ b/drivers/infiniband/hw/qib/qib_qp.c @@ -61,43 +61,6 @@ static inline unsigned find_next_offset(struct rvt_qpn_table *qpt, return off; } -/* - * Convert the AETH credit code into the number of credits. - */ -static u32 credit_table[31] = { - 0, /* 0 */ - 1, /* 1 */ - 2, /* 2 */ - 3, /* 3 */ - 4, /* 4 */ - 6, /* 5 */ - 8, /* 6 */ - 12, /* 7 */ - 16, /* 8 */ - 24, /* 9 */ - 32, /* A */ - 48, /* B */ - 64, /* C */ - 96, /* D */ - 128, /* E */ - 192, /* F */ - 256, /* 10 */ - 384, /* 11 */ - 512, /* 12 */ - 768, /* 13 */ - 1024, /* 14 */ - 1536, /* 15 */ - 2048, /* 16 */ - 3072, /* 17 */ - 4096, /* 18 */ - 6144, /* 19 */ - 8192, /* 1A */ - 12288, /* 1B */ - 16384, /* 1C */ - 24576, /* 1D */ - 32768 /* 1E */ -}; - const struct rvt_operation_params qib_post_parms[RVT_OPERATION_MAX] = { [IB_WR_RDMA_WRITE] = { .length = sizeof(struct ib_rdma_wr), @@ -354,66 +317,6 @@ u32 qib_mtu_from_qp(struct rvt_dev_info *rdi, struct rvt_qp *qp, u32 pmtu) return ib_mtu_enum_to_int(pmtu); } -/** - * qib_compute_aeth - compute the AETH (syndrome + MSN) - * @qp: the queue pair to compute the AETH for - * - * Returns the AETH. - */ -__be32 qib_compute_aeth(struct rvt_qp *qp) -{ - u32 aeth = qp->r_msn & QIB_MSN_MASK; - - if (qp->ibqp.srq) { - /* - * Shared receive queues don't generate credits. - * Set the credit field to the invalid value. - */ - aeth |= QIB_AETH_CREDIT_INVAL << QIB_AETH_CREDIT_SHIFT; - } else { - u32 min, max, x; - u32 credits; - struct rvt_rwq *wq = qp->r_rq.wq; - u32 head; - u32 tail; - - /* sanity check pointers before trusting them */ - head = wq->head; - if (head >= qp->r_rq.size) - head = 0; - tail = wq->tail; - if (tail >= qp->r_rq.size) - tail = 0; - /* - * Compute the number of credits available (RWQEs). - * XXX Not holding the r_rq.lock here so there is a small - * chance that the pair of reads are not atomic. - */ - credits = head - tail; - if ((int)credits < 0) - credits += qp->r_rq.size; - /* - * Binary search the credit table to find the code to - * use. - */ - min = 0; - max = 31; - for (;;) { - x = (min + max) / 2; - if (credit_table[x] == credits) - break; - if (credit_table[x] > credits) - max = x; - else if (min == x) - break; - else - min = x; - } - aeth |= x << QIB_AETH_CREDIT_SHIFT; - } - return cpu_to_be32(aeth); -} - void *qib_qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp, gfp_t gfp) { struct qib_qp_priv *priv; @@ -448,7 +351,6 @@ void qib_stop_send_queue(struct rvt_qp *qp) struct qib_qp_priv *priv = qp->priv; cancel_work_sync(&priv->s_work); - del_timer_sync(&qp->s_timer); } void qib_quiesce_qp(struct rvt_qp *qp) @@ -474,43 +376,6 @@ void qib_flush_qp_waiters(struct rvt_qp *qp) } /** - * qib_get_credit - flush the send work queue of a QP - * @qp: the qp who's send work queue to flush - * @aeth: the Acknowledge Extended Transport Header - * - * The QP s_lock should be held. - */ -void qib_get_credit(struct rvt_qp *qp, u32 aeth) -{ - u32 credit = (aeth >> QIB_AETH_CREDIT_SHIFT) & QIB_AETH_CREDIT_MASK; - - /* - * If the credit is invalid, we can send - * as many packets as we like. Otherwise, we have to - * honor the credit field. - */ - if (credit == QIB_AETH_CREDIT_INVAL) { - if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT)) { - qp->s_flags |= RVT_S_UNLIMITED_CREDIT; - if (qp->s_flags & RVT_S_WAIT_SSN_CREDIT) { - qp->s_flags &= ~RVT_S_WAIT_SSN_CREDIT; - qib_schedule_send(qp); - } - } - } else if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT)) { - /* Compute new LSN (i.e., MSN + credit) */ - credit = (aeth + credit_table[credit]) & QIB_MSN_MASK; - if (qib_cmp24(credit, qp->s_lsn) > 0) { - qp->s_lsn = credit; - if (qp->s_flags & RVT_S_WAIT_SSN_CREDIT) { - qp->s_flags &= ~RVT_S_WAIT_SSN_CREDIT; - qib_schedule_send(qp); - } - } - } -} - -/** * qib_check_send_wqe - validate wr/wqe * @qp - The qp * @wqe - The built wqe diff --git a/drivers/infiniband/hw/qib/qib_qsfp.c b/drivers/infiniband/hw/qib/qib_qsfp.c index 4c7c3c84a741..295d40a83bb6 100644 --- a/drivers/infiniband/hw/qib/qib_qsfp.c +++ b/drivers/infiniband/hw/qib/qib_qsfp.c @@ -485,16 +485,6 @@ void qib_qsfp_init(struct qib_qsfp_data *qd, dd->f_gpio_mod(dd, mask, mask, mask); } -void qib_qsfp_deinit(struct qib_qsfp_data *qd) -{ - /* - * There is nothing to do here for now. our work is scheduled - * with queue_work(), and flush_workqueue() from remove_one - * will block until all work setup with queue_work() - * completes. - */ -} - int qib_qsfp_dump(struct qib_pportdata *ppd, char *buf, int len) { struct qib_qsfp_cache cd; diff --git a/drivers/infiniband/hw/qib/qib_qsfp.h b/drivers/infiniband/hw/qib/qib_qsfp.h index 91908f533a2b..ad8dbd6ac0cf 100644 --- a/drivers/infiniband/hw/qib/qib_qsfp.h +++ b/drivers/infiniband/hw/qib/qib_qsfp.h @@ -186,4 +186,3 @@ extern int qib_refresh_qsfp_cache(struct qib_pportdata *ppd, extern int qib_qsfp_mod_present(struct qib_pportdata *ppd); extern void qib_qsfp_init(struct qib_qsfp_data *qd, void (*fevent)(struct work_struct *)); -extern void qib_qsfp_deinit(struct qib_qsfp_data *qd); diff --git a/drivers/infiniband/hw/qib/qib_rc.c b/drivers/infiniband/hw/qib/qib_rc.c index 2097512e75aa..12658e3fe154 100644 --- a/drivers/infiniband/hw/qib/qib_rc.c +++ b/drivers/infiniband/hw/qib/qib_rc.c @@ -38,7 +38,6 @@ /* cut down ridiculously long IB macro names */ #define OP(x) IB_OPCODE_RC_##x -static void rc_timeout(unsigned long arg); static u32 restart_sge(struct rvt_sge_state *ss, struct rvt_swqe *wqe, u32 psn, u32 pmtu) @@ -50,19 +49,10 @@ static u32 restart_sge(struct rvt_sge_state *ss, struct rvt_swqe *wqe, ss->sg_list = wqe->sg_list + 1; ss->num_sge = wqe->wr.num_sge; ss->total_len = wqe->length; - qib_skip_sge(ss, len, 0); + rvt_skip_sge(ss, len, false); return wqe->length - len; } -static void start_timer(struct rvt_qp *qp) -{ - qp->s_flags |= RVT_S_TIMER; - qp->s_timer.function = rc_timeout; - /* 4.096 usec. * (1 << qp->timeout) */ - qp->s_timer.expires = jiffies + qp->timeout_jiffies; - add_timer(&qp->s_timer); -} - /** * qib_make_rc_ack - construct a response packet (ACK, NAK, or RDMA read) * @dev: the device for this QP @@ -144,7 +134,7 @@ static int qib_make_rc_ack(struct qib_ibdev *dev, struct rvt_qp *qp, qp->s_ack_state = OP(RDMA_READ_RESPONSE_ONLY); e->sent = 1; } - ohdr->u.aeth = qib_compute_aeth(qp); + ohdr->u.aeth = rvt_compute_aeth(qp); hwords++; qp->s_ack_rdma_psn = e->psn; bth2 = qp->s_ack_rdma_psn++ & QIB_PSN_MASK; @@ -153,7 +143,7 @@ static int qib_make_rc_ack(struct qib_ibdev *dev, struct rvt_qp *qp, qp->s_cur_sge = NULL; len = 0; qp->s_ack_state = OP(ATOMIC_ACKNOWLEDGE); - ohdr->u.at.aeth = qib_compute_aeth(qp); + ohdr->u.at.aeth = rvt_compute_aeth(qp); ib_u64_put(e->atomic_data, &ohdr->u.at.atomic_ack_eth); hwords += sizeof(ohdr->u.at) / sizeof(u32); bth2 = e->psn & QIB_PSN_MASK; @@ -174,7 +164,7 @@ static int qib_make_rc_ack(struct qib_ibdev *dev, struct rvt_qp *qp, if (len > pmtu) len = pmtu; else { - ohdr->u.aeth = qib_compute_aeth(qp); + ohdr->u.aeth = rvt_compute_aeth(qp); hwords++; qp->s_ack_state = OP(RDMA_READ_RESPONSE_LAST); e = &qp->s_ack_queue[qp->s_tail_ack_queue]; @@ -197,11 +187,11 @@ normal: qp->s_cur_sge = NULL; if (qp->s_nak_state) ohdr->u.aeth = - cpu_to_be32((qp->r_msn & QIB_MSN_MASK) | + cpu_to_be32((qp->r_msn & IB_MSN_MASK) | (qp->s_nak_state << - QIB_AETH_CREDIT_SHIFT)); + IB_AETH_CREDIT_SHIFT)); else - ohdr->u.aeth = qib_compute_aeth(qp); + ohdr->u.aeth = rvt_compute_aeth(qp); hwords++; len = 0; bth0 = OP(ACKNOWLEDGE) << 24; @@ -257,7 +247,7 @@ int qib_make_rc_req(struct rvt_qp *qp, unsigned long *flags) goto bail; /* We are in the error state, flush the work request. */ smp_read_barrier_depends(); /* see post_one_send() */ - if (qp->s_last == ACCESS_ONCE(qp->s_head)) + if (qp->s_last == READ_ONCE(qp->s_head)) goto bail; /* If DMAs are in progress, we can't flush immediately. */ if (atomic_read(&priv->s_dma_busy)) { @@ -303,7 +293,8 @@ int qib_make_rc_req(struct rvt_qp *qp, unsigned long *flags) newreq = 0; if (qp->s_cur == qp->s_tail) { /* Check if send work queue is empty. */ - if (qp->s_tail == qp->s_head) + smp_read_barrier_depends(); /* see post_one_send() */ + if (qp->s_tail == READ_ONCE(qp->s_head)) goto bail; /* * If a fence is requested, wait for previous @@ -330,7 +321,7 @@ int qib_make_rc_req(struct rvt_qp *qp, unsigned long *flags) case IB_WR_SEND_WITH_IMM: /* If no credit, return. */ if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT) && - qib_cmp24(wqe->ssn, qp->s_lsn + 1) > 0) { + rvt_cmp_msn(wqe->ssn, qp->s_lsn + 1) > 0) { qp->s_flags |= RVT_S_WAIT_SSN_CREDIT; goto bail; } @@ -361,7 +352,7 @@ int qib_make_rc_req(struct rvt_qp *qp, unsigned long *flags) case IB_WR_RDMA_WRITE_WITH_IMM: /* If no credit, return. */ if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT) && - qib_cmp24(wqe->ssn, qp->s_lsn + 1) > 0) { + rvt_cmp_msn(wqe->ssn, qp->s_lsn + 1) > 0) { qp->s_flags |= RVT_S_WAIT_SSN_CREDIT; goto bail; } @@ -657,11 +648,11 @@ void qib_send_rc_ack(struct rvt_qp *qp) if (qp->s_mig_state == IB_MIG_MIGRATED) bth0 |= IB_BTH_MIG_REQ; if (qp->r_nak_state) - ohdr->u.aeth = cpu_to_be32((qp->r_msn & QIB_MSN_MASK) | + ohdr->u.aeth = cpu_to_be32((qp->r_msn & IB_MSN_MASK) | (qp->r_nak_state << - QIB_AETH_CREDIT_SHIFT)); + IB_AETH_CREDIT_SHIFT)); else - ohdr->u.aeth = qib_compute_aeth(qp); + ohdr->u.aeth = rvt_compute_aeth(qp); lrh0 |= ibp->sl_to_vl[qp->remote_ah_attr.sl] << 12 | qp->remote_ah_attr.sl << 4; hdr.lrh[0] = cpu_to_be16(lrh0); @@ -836,7 +827,7 @@ done: * Back up requester to resend the last un-ACKed request. * The QP r_lock and s_lock should be held and interrupts disabled. */ -static void qib_restart_rc(struct rvt_qp *qp, u32 psn, int wait) +void qib_restart_rc(struct rvt_qp *qp, u32 psn, int wait) { struct rvt_swqe *wqe = rvt_get_swqe_ptr(qp, qp->s_acked); struct qib_ibport *ibp; @@ -869,46 +860,6 @@ static void qib_restart_rc(struct rvt_qp *qp, u32 psn, int wait) } /* - * This is called from s_timer for missing responses. - */ -static void rc_timeout(unsigned long arg) -{ - struct rvt_qp *qp = (struct rvt_qp *)arg; - struct qib_ibport *ibp; - unsigned long flags; - - spin_lock_irqsave(&qp->r_lock, flags); - spin_lock(&qp->s_lock); - if (qp->s_flags & RVT_S_TIMER) { - ibp = to_iport(qp->ibqp.device, qp->port_num); - ibp->rvp.n_rc_timeouts++; - qp->s_flags &= ~RVT_S_TIMER; - del_timer(&qp->s_timer); - qib_restart_rc(qp, qp->s_last_psn + 1, 1); - qib_schedule_send(qp); - } - spin_unlock(&qp->s_lock); - spin_unlock_irqrestore(&qp->r_lock, flags); -} - -/* - * This is called from s_timer for RNR timeouts. - */ -void qib_rc_rnr_retry(unsigned long arg) -{ - struct rvt_qp *qp = (struct rvt_qp *)arg; - unsigned long flags; - - spin_lock_irqsave(&qp->s_lock, flags); - if (qp->s_flags & RVT_S_WAIT_RNR) { - qp->s_flags &= ~RVT_S_WAIT_RNR; - del_timer(&qp->s_timer); - qib_schedule_send(qp); - } - spin_unlock_irqrestore(&qp->s_lock, flags); -} - -/* * Set qp->s_sending_psn to the next PSN after the given one. * This would be psn+1 except when RDMA reads are present. */ @@ -941,12 +892,10 @@ void qib_rc_send_complete(struct rvt_qp *qp, struct ib_header *hdr) { struct ib_other_headers *ohdr; struct rvt_swqe *wqe; - struct ib_wc wc; - unsigned i; u32 opcode; u32 psn; - if (!(ib_rvt_state_ops[qp->state] & RVT_PROCESS_OR_FLUSH_SEND)) + if (!(ib_rvt_state_ops[qp->state] & RVT_SEND_OR_FLUSH_OR_RECV_OK)) return; /* Find out where the BTH is */ @@ -973,7 +922,7 @@ void qib_rc_send_complete(struct rvt_qp *qp, struct ib_header *hdr) if ((psn & IB_BTH_REQ_ACK) && qp->s_acked != qp->s_tail && !(qp->s_flags & (RVT_S_TIMER | RVT_S_WAIT_RNR | RVT_S_WAIT_PSN)) && (ib_rvt_state_ops[qp->state] & RVT_PROCESS_RECV_OK)) - start_timer(qp); + rvt_add_retry_timer(qp); while (qp->s_last != qp->s_acked) { u32 s_last; @@ -988,22 +937,8 @@ void qib_rc_send_complete(struct rvt_qp *qp, struct ib_header *hdr) qp->s_last = s_last; /* see post_send() */ barrier(); - for (i = 0; i < wqe->wr.num_sge; i++) { - struct rvt_sge *sge = &wqe->sg_list[i]; - - rvt_put_mr(sge->mr); - } - /* Post a send completion queue entry if requested. */ - if (!(qp->s_flags & RVT_S_SIGNAL_REQ_WR) || - (wqe->wr.send_flags & IB_SEND_SIGNALED)) { - memset(&wc, 0, sizeof(wc)); - wc.wr_id = wqe->wr.wr_id; - wc.status = IB_WC_SUCCESS; - wc.opcode = ib_qib_wc_opcode[wqe->wr.opcode]; - wc.byte_len = wqe->length; - wc.qp = &qp->ibqp; - rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.send_cq), &wc, 0); - } + rvt_put_swqe(wqe); + rvt_qp_swqe_complete(qp, wqe, IB_WC_SUCCESS); } /* * If we were waiting for sends to complete before resending, @@ -1032,9 +967,6 @@ static struct rvt_swqe *do_rc_completion(struct rvt_qp *qp, struct rvt_swqe *wqe, struct qib_ibport *ibp) { - struct ib_wc wc; - unsigned i; - /* * Don't decrement refcount and don't generate a * completion if the SWQE is being resent until the send @@ -1044,28 +976,14 @@ static struct rvt_swqe *do_rc_completion(struct rvt_qp *qp, qib_cmp24(qp->s_sending_psn, qp->s_sending_hpsn) > 0) { u32 s_last; - for (i = 0; i < wqe->wr.num_sge; i++) { - struct rvt_sge *sge = &wqe->sg_list[i]; - - rvt_put_mr(sge->mr); - } + rvt_put_swqe(wqe); s_last = qp->s_last; if (++s_last >= qp->s_size) s_last = 0; qp->s_last = s_last; /* see post_send() */ barrier(); - /* Post a send completion queue entry if requested. */ - if (!(qp->s_flags & RVT_S_SIGNAL_REQ_WR) || - (wqe->wr.send_flags & IB_SEND_SIGNALED)) { - memset(&wc, 0, sizeof(wc)); - wc.wr_id = wqe->wr.wr_id; - wc.status = IB_WC_SUCCESS; - wc.opcode = ib_qib_wc_opcode[wqe->wr.opcode]; - wc.byte_len = wqe->length; - wc.qp = &qp->ibqp; - rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.send_cq), &wc, 0); - } + rvt_qp_swqe_complete(qp, wqe, IB_WC_SUCCESS); } else this_cpu_inc(*ibp->rvp.rc_delayed_comp); @@ -1117,12 +1035,6 @@ static int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode, u32 ack_psn; int diff; - /* Remove QP from retry timer */ - if (qp->s_flags & (RVT_S_TIMER | RVT_S_WAIT_RNR)) { - qp->s_flags &= ~(RVT_S_TIMER | RVT_S_WAIT_RNR); - del_timer(&qp->s_timer); - } - /* * Note that NAKs implicitly ACK outstanding SEND and RDMA write * requests and implicitly NAK RDMA read and atomic requests issued @@ -1130,7 +1042,7 @@ static int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode, * request but will include an ACK'ed request(s). */ ack_psn = psn; - if (aeth >> 29) + if (aeth >> IB_AETH_NAK_SHIFT) ack_psn--; wqe = rvt_get_swqe_ptr(qp, qp->s_acked); ibp = to_iport(qp->ibqp.device, qp->port_num); @@ -1210,7 +1122,7 @@ static int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode, break; } - switch (aeth >> 29) { + switch (aeth >> IB_AETH_NAK_SHIFT) { case 0: /* ACK */ this_cpu_inc(*ibp->rvp.rc_acks); if (qp->s_acked != qp->s_tail) { @@ -1218,27 +1130,30 @@ static int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode, * We are expecting more ACKs so * reset the retransmit timer. */ - start_timer(qp); + rvt_mod_retry_timer(qp); /* * We can stop resending the earlier packets and * continue with the next packet the receiver wants. */ if (qib_cmp24(qp->s_psn, psn) <= 0) reset_psn(qp, psn + 1); - } else if (qib_cmp24(qp->s_psn, psn) <= 0) { - qp->s_state = OP(SEND_LAST); - qp->s_psn = psn + 1; + } else { + /* No more acks - kill all timers */ + rvt_stop_rc_timers(qp); + if (qib_cmp24(qp->s_psn, psn) <= 0) { + qp->s_state = OP(SEND_LAST); + qp->s_psn = psn + 1; + } } if (qp->s_flags & RVT_S_WAIT_ACK) { qp->s_flags &= ~RVT_S_WAIT_ACK; qib_schedule_send(qp); } - qib_get_credit(qp, aeth); + rvt_get_credit(qp, aeth); qp->s_rnr_retry = qp->s_rnr_retry_cnt; qp->s_retry = qp->s_retry_cnt; update_last_psn(qp, psn); - ret = 1; - goto bail; + return 1; case 1: /* RNR NAK */ ibp->rvp.n_rnr_naks++; @@ -1261,21 +1176,17 @@ static int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode, reset_psn(qp, psn); qp->s_flags &= ~(RVT_S_WAIT_SSN_CREDIT | RVT_S_WAIT_ACK); - qp->s_flags |= RVT_S_WAIT_RNR; - qp->s_timer.function = qib_rc_rnr_retry; - qp->s_timer.expires = jiffies + usecs_to_jiffies( - ib_qib_rnr_table[(aeth >> QIB_AETH_CREDIT_SHIFT) & - QIB_AETH_CREDIT_MASK]); - add_timer(&qp->s_timer); - goto bail; + rvt_stop_rc_timers(qp); + rvt_add_rnr_timer(qp, aeth); + return 0; case 3: /* NAK */ if (qp->s_acked == qp->s_tail) goto bail; /* The last valid PSN is the previous PSN. */ update_last_psn(qp, psn - 1); - switch ((aeth >> QIB_AETH_CREDIT_SHIFT) & - QIB_AETH_CREDIT_MASK) { + switch ((aeth >> IB_AETH_CREDIT_SHIFT) & + IB_AETH_CREDIT_MASK) { case 0: /* PSN sequence error */ ibp->rvp.n_seq_naks++; /* @@ -1323,6 +1234,7 @@ reserved: } bail: + rvt_stop_rc_timers(qp); return ret; } @@ -1336,10 +1248,7 @@ static void rdma_seq_err(struct rvt_qp *qp, struct qib_ibport *ibp, u32 psn, struct rvt_swqe *wqe; /* Remove QP from retry timer */ - if (qp->s_flags & (RVT_S_TIMER | RVT_S_WAIT_RNR)) { - qp->s_flags &= ~(RVT_S_TIMER | RVT_S_WAIT_RNR); - del_timer(&qp->s_timer); - } + rvt_stop_rc_timers(qp); wqe = rvt_get_swqe_ptr(qp, qp->s_acked); @@ -1423,7 +1332,7 @@ static void qib_rc_rcv_resp(struct qib_ibport *ibp, /* Ignore invalid responses. */ smp_read_barrier_depends(); /* see post_one_send */ - if (qib_cmp24(psn, ACCESS_ONCE(qp->s_next_psn)) >= 0) + if (qib_cmp24(psn, READ_ONCE(qp->s_next_psn)) >= 0) goto ack_done; /* Ignore duplicate responses. */ @@ -1432,8 +1341,8 @@ static void qib_rc_rcv_resp(struct qib_ibport *ibp, /* Update credits for "ghost" ACKs */ if (diff == 0 && opcode == OP(ACKNOWLEDGE)) { aeth = be32_to_cpu(ohdr->u.aeth); - if ((aeth >> 29) == 0) - qib_get_credit(qp, aeth); + if ((aeth >> IB_AETH_NAK_SHIFT) == 0) + rvt_get_credit(qp, aeth); } goto ack_done; } @@ -1494,8 +1403,7 @@ read_middle: * We got a response so update the timeout. * 4.096 usec. * (1 << qp->timeout) */ - qp->s_flags |= RVT_S_TIMER; - mod_timer(&qp->s_timer, jiffies + qp->timeout_jiffies); + rvt_mod_retry_timer(qp); if (qp->s_flags & RVT_S_WAIT_ACK) { qp->s_flags &= ~RVT_S_WAIT_ACK; qib_schedule_send(qp); @@ -1797,25 +1705,6 @@ send_ack: return 0; } -void qib_rc_error(struct rvt_qp *qp, enum ib_wc_status err) -{ - unsigned long flags; - int lastwqe; - - spin_lock_irqsave(&qp->s_lock, flags); - lastwqe = rvt_error_qp(qp, err); - spin_unlock_irqrestore(&qp->s_lock, flags); - - if (lastwqe) { - struct ib_event ev; - - ev.device = qp->ibqp.device; - ev.element.qp = &qp->ibqp; - ev.event = IB_EVENT_QP_LAST_WQE_REACHED; - qp->ibqp.event_handler(&ev, qp->ibqp.qp_context); - } -} - static inline void qib_update_ack_queue(struct rvt_qp *qp, unsigned n) { unsigned next; @@ -1927,17 +1816,8 @@ void qib_rc_rcv(struct qib_ctxtdata *rcd, struct ib_header *hdr, break; } - if (qp->state == IB_QPS_RTR && !(qp->r_flags & RVT_R_COMM_EST)) { - qp->r_flags |= RVT_R_COMM_EST; - if (qp->ibqp.event_handler) { - struct ib_event ev; - - ev.device = qp->ibqp.device; - ev.element.qp = &qp->ibqp; - ev.event = IB_EVENT_COMM_EST; - qp->ibqp.event_handler(&ev, qp->ibqp.qp_context); - } - } + if (qp->state == IB_QPS_RTR && !(qp->r_flags & RVT_R_COMM_EST)) + rvt_comm_est(qp); /* OK, process the packet. */ switch (opcode) { @@ -2112,8 +1992,7 @@ send_last: * Update the next expected PSN. We add 1 later * below, so only add the remainder here. */ - if (len > pmtu) - qp->r_psn += (len - 1) / pmtu; + qp->r_psn += rvt_div_mtu(qp, len - 1); } else { e->rdma_sge.mr = NULL; e->rdma_sge.vaddr = NULL; @@ -2230,7 +2109,7 @@ rnr_nak: return; nack_op_err: - qib_rc_error(qp, IB_WC_LOC_QP_OP_ERR); + rvt_rc_error(qp, IB_WC_LOC_QP_OP_ERR); qp->r_nak_state = IB_NAK_REMOTE_OPERATIONAL_ERROR; qp->r_ack_psn = qp->r_psn; /* Queue NAK for later */ @@ -2244,7 +2123,7 @@ nack_op_err: nack_inv_unlck: spin_unlock_irqrestore(&qp->s_lock, flags); nack_inv: - qib_rc_error(qp, IB_WC_LOC_QP_OP_ERR); + rvt_rc_error(qp, IB_WC_LOC_QP_OP_ERR); qp->r_nak_state = IB_NAK_INVALID_REQUEST; qp->r_ack_psn = qp->r_psn; /* Queue NAK for later */ @@ -2258,7 +2137,7 @@ nack_inv: nack_acc_unlck: spin_unlock_irqrestore(&qp->s_lock, flags); nack_acc: - qib_rc_error(qp, IB_WC_LOC_PROT_ERR); + rvt_rc_error(qp, IB_WC_LOC_PROT_ERR); qp->r_nak_state = IB_NAK_REMOTE_ACCESS_ERROR; qp->r_ack_psn = qp->r_psn; send_ack: diff --git a/drivers/infiniband/hw/qib/qib_ruc.c b/drivers/infiniband/hw/qib/qib_ruc.c index de1bde5950f5..17655cc3e6fe 100644 --- a/drivers/infiniband/hw/qib/qib_ruc.c +++ b/drivers/infiniband/hw/qib/qib_ruc.c @@ -38,44 +38,6 @@ #include "qib_mad.h" /* - * Convert the AETH RNR timeout code into the number of microseconds. - */ -const u32 ib_qib_rnr_table[32] = { - 655360, /* 00: 655.36 */ - 10, /* 01: .01 */ - 20, /* 02 .02 */ - 30, /* 03: .03 */ - 40, /* 04: .04 */ - 60, /* 05: .06 */ - 80, /* 06: .08 */ - 120, /* 07: .12 */ - 160, /* 08: .16 */ - 240, /* 09: .24 */ - 320, /* 0A: .32 */ - 480, /* 0B: .48 */ - 640, /* 0C: .64 */ - 960, /* 0D: .96 */ - 1280, /* 0E: 1.28 */ - 1920, /* 0F: 1.92 */ - 2560, /* 10: 2.56 */ - 3840, /* 11: 3.84 */ - 5120, /* 12: 5.12 */ - 7680, /* 13: 7.68 */ - 10240, /* 14: 10.24 */ - 15360, /* 15: 15.36 */ - 20480, /* 16: 20.48 */ - 30720, /* 17: 30.72 */ - 40960, /* 18: 40.96 */ - 61440, /* 19: 61.44 */ - 81920, /* 1A: 81.92 */ - 122880, /* 1B: 122.88 */ - 163840, /* 1C: 163.84 */ - 245760, /* 1D: 245.76 */ - 327680, /* 1E: 327.68 */ - 491520 /* 1F: 491.52 */ -}; - -/* * Validate a RWQE and fill in the SGE state. * Return 1 if OK. */ @@ -599,11 +561,8 @@ rnr_nak: spin_lock_irqsave(&sqp->s_lock, flags); if (!(ib_rvt_state_ops[sqp->state] & RVT_PROCESS_RECV_OK)) goto clr_busy; - sqp->s_flags |= RVT_S_WAIT_RNR; - sqp->s_timer.function = qib_rc_rnr_retry; - sqp->s_timer.expires = jiffies + - usecs_to_jiffies(ib_qib_rnr_table[qp->r_min_rnr_timer]); - add_timer(&sqp->s_timer); + rvt_add_rnr_timer(sqp, qp->r_min_rnr_timer << + IB_AETH_CREDIT_SHIFT); goto clr_busy; op_err: @@ -621,7 +580,7 @@ acc_err: wc.status = IB_WC_LOC_PROT_ERR; err: /* responder goes to error state */ - qib_rc_error(qp, wc.status); + rvt_rc_error(qp, wc.status); serr: spin_lock_irqsave(&sqp->s_lock, flags); @@ -793,7 +752,6 @@ void qib_send_complete(struct rvt_qp *qp, struct rvt_swqe *wqe, enum ib_wc_status status) { u32 old_last, last; - unsigned i; if (!(ib_rvt_state_ops[qp->state] & RVT_PROCESS_OR_FLUSH_SEND)) return; @@ -805,32 +763,13 @@ void qib_send_complete(struct rvt_qp *qp, struct rvt_swqe *wqe, qp->s_last = last; /* See post_send() */ barrier(); - for (i = 0; i < wqe->wr.num_sge; i++) { - struct rvt_sge *sge = &wqe->sg_list[i]; - - rvt_put_mr(sge->mr); - } + rvt_put_swqe(wqe); if (qp->ibqp.qp_type == IB_QPT_UD || qp->ibqp.qp_type == IB_QPT_SMI || qp->ibqp.qp_type == IB_QPT_GSI) atomic_dec(&ibah_to_rvtah(wqe->ud_wr.ah)->refcount); - /* See ch. 11.2.4.1 and 10.7.3.1 */ - if (!(qp->s_flags & RVT_S_SIGNAL_REQ_WR) || - (wqe->wr.send_flags & IB_SEND_SIGNALED) || - status != IB_WC_SUCCESS) { - struct ib_wc wc; - - memset(&wc, 0, sizeof(wc)); - wc.wr_id = wqe->wr.wr_id; - wc.status = status; - wc.opcode = ib_qib_wc_opcode[wqe->wr.opcode]; - wc.qp = &qp->ibqp; - if (status == IB_WC_SUCCESS) - wc.byte_len = wqe->length; - rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.send_cq), &wc, - status != IB_WC_SUCCESS); - } + rvt_qp_swqe_complete(qp, wqe, status); if (qp->s_acked == old_last) qp->s_acked = last; diff --git a/drivers/infiniband/hw/qib/qib_uc.c b/drivers/infiniband/hw/qib/qib_uc.c index 5b2d483451ad..b337b60fc40d 100644 --- a/drivers/infiniband/hw/qib/qib_uc.c +++ b/drivers/infiniband/hw/qib/qib_uc.c @@ -325,17 +325,8 @@ inv: goto inv; } - if (qp->state == IB_QPS_RTR && !(qp->r_flags & RVT_R_COMM_EST)) { - qp->r_flags |= RVT_R_COMM_EST; - if (qp->ibqp.event_handler) { - struct ib_event ev; - - ev.device = qp->ibqp.device; - ev.element.qp = &qp->ibqp; - ev.event = IB_EVENT_COMM_EST; - qp->ibqp.event_handler(&ev, qp->ibqp.qp_context); - } - } + if (qp->state == IB_QPS_RTR && !(qp->r_flags & RVT_R_COMM_EST)) + rvt_comm_est(qp); /* OK, process the packet. */ switch (opcode) { @@ -527,7 +518,7 @@ drop: return; op_err: - qib_rc_error(qp, IB_WC_LOC_QP_OP_ERR); + rvt_rc_error(qp, IB_WC_LOC_QP_OP_ERR); return; } diff --git a/drivers/infiniband/hw/qib/qib_ud.c b/drivers/infiniband/hw/qib/qib_ud.c index f45cad1198b0..ddd4e7458750 100644 --- a/drivers/infiniband/hw/qib/qib_ud.c +++ b/drivers/infiniband/hw/qib/qib_ud.c @@ -152,7 +152,7 @@ static void qib_ud_loopback(struct rvt_qp *sqp, struct rvt_swqe *swqe) ret = qib_get_rwqe(qp, 0); if (ret < 0) { - qib_rc_error(qp, IB_WC_LOC_QP_OP_ERR); + rvt_rc_error(qp, IB_WC_LOC_QP_OP_ERR); goto bail_unlock; } if (!ret) { @@ -177,7 +177,7 @@ static void qib_ud_loopback(struct rvt_qp *sqp, struct rvt_swqe *swqe) sizeof(grh), 1); wc.wc_flags |= IB_WC_GRH; } else - qib_skip_sge(&qp->r_sge, sizeof(struct ib_grh), 1); + rvt_skip_sge(&qp->r_sge, sizeof(struct ib_grh), true); ssge.sg_list = swqe->sg_list + 1; ssge.sge = *swqe->sg_list; ssge.num_sge = swqe->wr.num_sge; @@ -548,7 +548,7 @@ void qib_ud_rcv(struct qib_ibport *ibp, struct ib_header *hdr, ret = qib_get_rwqe(qp, 0); if (ret < 0) { - qib_rc_error(qp, IB_WC_LOC_QP_OP_ERR); + rvt_rc_error(qp, IB_WC_LOC_QP_OP_ERR); return; } if (!ret) { @@ -567,7 +567,7 @@ void qib_ud_rcv(struct qib_ibport *ibp, struct ib_header *hdr, sizeof(struct ib_grh), 1); wc.wc_flags |= IB_WC_GRH; } else - qib_skip_sge(&qp->r_sge, sizeof(struct ib_grh), 1); + rvt_skip_sge(&qp->r_sge, sizeof(struct ib_grh), true); qib_copy_sge(&qp->r_sge, data, wc.byte_len - sizeof(struct ib_grh), 1); rvt_put_ss(&qp->r_sge); if (!test_and_clear_bit(RVT_R_WRID_VALID, &qp->r_aflags)) diff --git a/drivers/infiniband/hw/qib/qib_user_sdma.c b/drivers/infiniband/hw/qib/qib_user_sdma.c index 3e0677c51276..926f3c8eba69 100644 --- a/drivers/infiniband/hw/qib/qib_user_sdma.c +++ b/drivers/infiniband/hw/qib/qib_user_sdma.c @@ -144,8 +144,8 @@ qib_user_sdma_rb_search(struct rb_root *root, pid_t pid) struct rb_node *node = root->rb_node; while (node) { - sdma_rb_node = container_of(node, - struct qib_user_sdma_rb_node, node); + sdma_rb_node = rb_entry(node, struct qib_user_sdma_rb_node, + node); if (pid < sdma_rb_node->pid) node = node->rb_left; else if (pid > sdma_rb_node->pid) @@ -164,7 +164,7 @@ qib_user_sdma_rb_insert(struct rb_root *root, struct qib_user_sdma_rb_node *new) struct qib_user_sdma_rb_node *got; while (*node) { - got = container_of(*node, struct qib_user_sdma_rb_node, node); + got = rb_entry(*node, struct qib_user_sdma_rb_node, node); parent = *node; if (new->pid < got->pid) node = &((*node)->rb_left); diff --git a/drivers/infiniband/hw/qib/qib_verbs.c b/drivers/infiniband/hw/qib/qib_verbs.c index 954f15064514..83f8b5f24381 100644 --- a/drivers/infiniband/hw/qib/qib_verbs.c +++ b/drivers/infiniband/hw/qib/qib_verbs.c @@ -114,19 +114,6 @@ module_param_named(disable_sma, ib_qib_disable_sma, uint, S_IWUSR | S_IRUGO); MODULE_PARM_DESC(disable_sma, "Disable the SMA"); /* - * Translate ib_wr_opcode into ib_wc_opcode. - */ -const enum ib_wc_opcode ib_qib_wc_opcode[] = { - [IB_WR_RDMA_WRITE] = IB_WC_RDMA_WRITE, - [IB_WR_RDMA_WRITE_WITH_IMM] = IB_WC_RDMA_WRITE, - [IB_WR_SEND] = IB_WC_SEND, - [IB_WR_SEND_WITH_IMM] = IB_WC_SEND, - [IB_WR_RDMA_READ] = IB_WC_RDMA_READ, - [IB_WR_ATOMIC_CMP_AND_SWP] = IB_WC_COMP_SWAP, - [IB_WR_ATOMIC_FETCH_AND_ADD] = IB_WC_FETCH_ADD -}; - -/* * System image GUID. */ __be64 ib_qib_sys_image_guid; @@ -142,78 +129,16 @@ void qib_copy_sge(struct rvt_sge_state *ss, void *data, u32 length, int release) struct rvt_sge *sge = &ss->sge; while (length) { - u32 len = sge->length; + u32 len = rvt_get_sge_length(sge, length); - if (len > length) - len = length; - if (len > sge->sge_length) - len = sge->sge_length; - BUG_ON(len == 0); + WARN_ON_ONCE(len == 0); memcpy(sge->vaddr, data, len); - sge->vaddr += len; - sge->length -= len; - sge->sge_length -= len; - if (sge->sge_length == 0) { - if (release) - rvt_put_mr(sge->mr); - if (--ss->num_sge) - *sge = *ss->sg_list++; - } else if (sge->length == 0 && sge->mr->lkey) { - if (++sge->n >= RVT_SEGSZ) { - if (++sge->m >= sge->mr->mapsz) - break; - sge->n = 0; - } - sge->vaddr = - sge->mr->map[sge->m]->segs[sge->n].vaddr; - sge->length = - sge->mr->map[sge->m]->segs[sge->n].length; - } + rvt_update_sge(ss, len, release); data += len; length -= len; } } -/** - * qib_skip_sge - skip over SGE memory - XXX almost dup of prev func - * @ss: the SGE state - * @length: the number of bytes to skip - */ -void qib_skip_sge(struct rvt_sge_state *ss, u32 length, int release) -{ - struct rvt_sge *sge = &ss->sge; - - while (length) { - u32 len = sge->length; - - if (len > length) - len = length; - if (len > sge->sge_length) - len = sge->sge_length; - BUG_ON(len == 0); - sge->vaddr += len; - sge->length -= len; - sge->sge_length -= len; - if (sge->sge_length == 0) { - if (release) - rvt_put_mr(sge->mr); - if (--ss->num_sge) - *sge = *ss->sg_list++; - } else if (sge->length == 0 && sge->mr->lkey) { - if (++sge->n >= RVT_SEGSZ) { - if (++sge->m >= sge->mr->mapsz) - break; - sge->n = 0; - } - sge->vaddr = - sge->mr->map[sge->m]->segs[sge->n].vaddr; - sge->length = - sge->mr->map[sge->m]->segs[sge->n].length; - } - length -= len; - } -} - /* * Count the number of DMA descriptors needed to send length bytes of data. * Don't modify the qib_sge_state to get the count. @@ -464,7 +389,7 @@ static void mem_timer(unsigned long data) priv = list_entry(list->next, struct qib_qp_priv, iowait); qp = priv->owner; list_del_init(&priv->iowait); - atomic_inc(&qp->refcount); + rvt_get_qp(qp); if (!list_empty(list)) mod_timer(&dev->mem_timer, jiffies + 1); } @@ -477,29 +402,7 @@ static void mem_timer(unsigned long data) qib_schedule_send(qp); } spin_unlock_irqrestore(&qp->s_lock, flags); - if (atomic_dec_and_test(&qp->refcount)) - wake_up(&qp->wait); - } -} - -static void update_sge(struct rvt_sge_state *ss, u32 length) -{ - struct rvt_sge *sge = &ss->sge; - - sge->vaddr += length; - sge->length -= length; - sge->sge_length -= length; - if (sge->sge_length == 0) { - if (--ss->num_sge) - *sge = *ss->sg_list++; - } else if (sge->length == 0 && sge->mr->lkey) { - if (++sge->n >= RVT_SEGSZ) { - if (++sge->m >= sge->mr->mapsz) - return; - sge->n = 0; - } - sge->vaddr = sge->mr->map[sge->m]->segs[sge->n].vaddr; - sge->length = sge->mr->map[sge->m]->segs[sge->n].length; + rvt_put_qp(qp); } } @@ -660,11 +563,11 @@ static void copy_io(u32 __iomem *piobuf, struct rvt_sge_state *ss, data = clear_upper_bytes(v, extra, 0); } } - update_sge(ss, len); + rvt_update_sge(ss, len, false); length -= len; } /* Update address before sending packet. */ - update_sge(ss, length); + rvt_update_sge(ss, length, false); if (flush_wc) { /* must flush early everything before trigger word */ qib_flush_wc(); @@ -762,7 +665,7 @@ void qib_put_txreq(struct qib_verbs_txreq *tx) iowait); qp = priv->owner; list_del_init(&priv->iowait); - atomic_inc(&qp->refcount); + rvt_get_qp(qp); spin_unlock_irqrestore(&dev->rdi.pending_lock, flags); spin_lock_irqsave(&qp->s_lock, flags); @@ -772,8 +675,7 @@ void qib_put_txreq(struct qib_verbs_txreq *tx) } spin_unlock_irqrestore(&qp->s_lock, flags); - if (atomic_dec_and_test(&qp->refcount)) - wake_up(&qp->wait); + rvt_put_qp(qp); } else spin_unlock_irqrestore(&dev->rdi.pending_lock, flags); } @@ -808,7 +710,7 @@ void qib_verbs_sdma_desc_avail(struct qib_pportdata *ppd, unsigned avail) break; avail -= qpp->s_tx->txreq.sg_count; list_del_init(&qpp->iowait); - atomic_inc(&qp->refcount); + rvt_get_qp(qp); qps[n++] = qp; } @@ -822,8 +724,7 @@ void qib_verbs_sdma_desc_avail(struct qib_pportdata *ppd, unsigned avail) qib_schedule_send(qp); } spin_unlock(&qp->s_lock); - if (atomic_dec_and_test(&qp->refcount)) - wake_up(&qp->wait); + rvt_put_qp(qp); } } @@ -1085,7 +986,7 @@ static int qib_verbs_send_pio(struct rvt_qp *qp, struct ib_header *ibhdr, u32 *addr = (u32 *) ss->sge.vaddr; /* Update address before sending packet. */ - update_sge(ss, len); + rvt_update_sge(ss, len, false); if (flush_wc) { qib_pio_copy(piobuf, addr, dwords - 1); /* must flush early everything before trigger word */ @@ -1288,7 +1189,7 @@ void qib_ib_piobufavail(struct qib_devdata *dd) priv = list_entry(list->next, struct qib_qp_priv, iowait); qp = priv->owner; list_del_init(&priv->iowait); - atomic_inc(&qp->refcount); + rvt_get_qp(qp); qps[n++] = qp; } dd->f_wantpiobuf_intr(dd, 0); @@ -1306,8 +1207,7 @@ full: spin_unlock_irqrestore(&qp->s_lock, flags); /* Notify qib_destroy_qp() if it is waiting. */ - if (atomic_dec_and_test(&qp->refcount)) - wake_up(&qp->wait); + rvt_put_qp(qp); } } @@ -1320,6 +1220,7 @@ static int qib_query_port(struct rvt_dev_info *rdi, u8 port_num, enum ib_mtu mtu; u16 lid = ppd->lid; + /* props being zeroed by the caller, avoid zeroing it here */ props->lid = lid ? lid : be16_to_cpu(IB_LID_PERMISSIVE); props->lmc = ppd->lmc; props->state = dd->f_iblink_state(ppd->lastibcstat); @@ -1649,7 +1550,7 @@ int qib_register_ib_device(struct qib_devdata *dd) ibdev->owner = THIS_MODULE; ibdev->node_guid = ppd->guid; ibdev->phys_port_cnt = dd->num_pports; - ibdev->dma_device = &dd->pcidev->dev; + ibdev->dev.parent = &dd->pcidev->dev; ibdev->modify_device = qib_modify_device; ibdev->process_mad = qib_process_mad; @@ -1676,6 +1577,7 @@ int qib_register_ib_device(struct qib_devdata *dd) dd->verbs_dev.rdi.driver_f.stop_send_queue = qib_stop_send_queue; dd->verbs_dev.rdi.driver_f.flush_qp_waiters = qib_flush_qp_waiters; dd->verbs_dev.rdi.driver_f.notify_error_qp = qib_notify_error_qp; + dd->verbs_dev.rdi.driver_f.notify_restart_rc = qib_restart_rc; dd->verbs_dev.rdi.driver_f.mtu_to_path_mtu = qib_mtu_to_path_mtu; dd->verbs_dev.rdi.driver_f.mtu_from_qp = qib_mtu_from_qp; dd->verbs_dev.rdi.driver_f.get_pmtu_from_attr = qib_get_pmtu_from_attr; diff --git a/drivers/infiniband/hw/qib/qib_verbs.h b/drivers/infiniband/hw/qib/qib_verbs.h index 94fd30fdedac..212e8ce71be8 100644 --- a/drivers/infiniband/hw/qib/qib_verbs.h +++ b/drivers/infiniband/hw/qib/qib_verbs.h @@ -270,8 +270,6 @@ int qib_snapshot_counters(struct qib_pportdata *ppd, u64 *swords, int qib_get_counters(struct qib_pportdata *ppd, struct qib_verbs_counters *cntrs); -__be32 qib_compute_aeth(struct rvt_qp *qp); - /* * Functions provided by qib driver for rdmavt to use */ @@ -281,7 +279,7 @@ void qib_qp_priv_free(struct rvt_dev_info *rdi, struct rvt_qp *qp); void qib_notify_qp_reset(struct rvt_qp *qp); int qib_alloc_qpn(struct rvt_dev_info *rdi, struct rvt_qpn_table *qpt, enum ib_qp_type type, u8 port, gfp_t gfp); - +void qib_restart_rc(struct rvt_qp *qp, u32 psn, int wait); #ifdef CONFIG_DEBUG_FS struct qib_qp_iter; @@ -294,8 +292,6 @@ void qib_qp_iter_print(struct seq_file *s, struct qib_qp_iter *iter); #endif -void qib_get_credit(struct rvt_qp *qp, u32 aeth); - unsigned qib_pkt_delay(u32 plen, u8 snd_mult, u8 rcv_mult); void qib_verbs_sdma_desc_avail(struct qib_pportdata *ppd, unsigned avail); @@ -308,8 +304,6 @@ int qib_verbs_send(struct rvt_qp *qp, struct ib_header *hdr, void qib_copy_sge(struct rvt_sge_state *ss, void *data, u32 length, int release); -void qib_skip_sge(struct rvt_sge_state *ss, u32 length, int release); - void qib_uc_rcv(struct qib_ibport *ibp, struct ib_header *hdr, int has_grh, void *data, u32 tlen, struct rvt_qp *qp); @@ -326,8 +320,6 @@ void qib_rc_rnr_retry(unsigned long arg); void qib_rc_send_complete(struct rvt_qp *qp, struct ib_header *hdr); -void qib_rc_error(struct rvt_qp *qp, enum ib_wc_status err); - int qib_post_ud_send(struct rvt_qp *qp, struct ib_send_wr *wr); void qib_ud_rcv(struct qib_ibport *ibp, struct ib_header *hdr, diff --git a/drivers/infiniband/hw/usnic/usnic_common_pkt_hdr.h b/drivers/infiniband/hw/usnic/usnic_common_pkt_hdr.h index 596e0ed49a8e..bf7d197a9f42 100644 --- a/drivers/infiniband/hw/usnic/usnic_common_pkt_hdr.h +++ b/drivers/infiniband/hw/usnic/usnic_common_pkt_hdr.h @@ -34,7 +34,6 @@ #ifndef USNIC_CMN_PKT_HDR_H #define USNIC_CMN_PKT_HDR_H -#define USNIC_ROCE_ETHERTYPE (0x8915) #define USNIC_ROCE_GRH_VER (8) #define USNIC_PROTO_VER (1) #define USNIC_ROCE_GRH_VER_SHIFT (4) diff --git a/drivers/infiniband/hw/usnic/usnic_fwd.h b/drivers/infiniband/hw/usnic/usnic_fwd.h index 3a8add9ddf46..b2ac22be0731 100644 --- a/drivers/infiniband/hw/usnic/usnic_fwd.h +++ b/drivers/infiniband/hw/usnic/usnic_fwd.h @@ -36,6 +36,7 @@ #include <linux/if.h> #include <linux/netdevice.h> +#include <linux/if_ether.h> #include <linux/pci.h> #include <linux/in.h> @@ -97,7 +98,7 @@ static inline void usnic_fwd_init_usnic_filter(struct filter *filter, uint32_t usnic_id) { filter->type = FILTER_USNIC_ID; - filter->u.usnic.ethtype = USNIC_ROCE_ETHERTYPE; + filter->u.usnic.ethtype = ETH_P_IBOE; filter->u.usnic.flags = FILTER_FIELD_USNIC_ETHTYPE | FILTER_FIELD_USNIC_ID | FILTER_FIELD_USNIC_PROTO; diff --git a/drivers/infiniband/hw/usnic/usnic_ib_main.c b/drivers/infiniband/hw/usnic/usnic_ib_main.c index 0a89a955550b..c0c1e8b027b1 100644 --- a/drivers/infiniband/hw/usnic/usnic_ib_main.c +++ b/drivers/infiniband/hw/usnic/usnic_ib_main.c @@ -321,7 +321,9 @@ static int usnic_port_immutable(struct ib_device *ibdev, u8 port_num, struct ib_port_attr attr; int err; - err = usnic_ib_query_port(ibdev, port_num, &attr); + immutable->core_cap_flags = RDMA_CORE_PORT_USNIC; + + err = ib_query_port(ibdev, port_num, &attr); if (err) return err; @@ -380,7 +382,7 @@ static void *usnic_ib_device_add(struct pci_dev *dev) us_ibdev->ib_dev.node_type = RDMA_NODE_USNIC_UDP; us_ibdev->ib_dev.phys_port_cnt = USNIC_IB_PORT_CNT; us_ibdev->ib_dev.num_comp_vectors = USNIC_IB_NUM_COMP_VECTORS; - us_ibdev->ib_dev.dma_device = &dev->dev; + us_ibdev->ib_dev.dev.parent = &dev->dev; us_ibdev->ib_dev.uverbs_abi_ver = USNIC_UVERBS_ABI_VERSION; strlcpy(us_ibdev->ib_dev.name, "usnic_%d", IB_DEVICE_NAME_MAX); diff --git a/drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c b/drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c index 5b0248adf4ce..092d4e11a633 100644 --- a/drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c +++ b/drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c @@ -117,10 +117,10 @@ static int enable_qp_grp(struct usnic_ib_qp_grp *qp_grp) vnic_idx = usnic_vnic_get_index(qp_grp->vf->vnic); res_chunk = get_qp_res_chunk(qp_grp); - if (IS_ERR_OR_NULL(res_chunk)) { + if (IS_ERR(res_chunk)) { usnic_err("Unable to get qp res with err %ld\n", PTR_ERR(res_chunk)); - return res_chunk ? PTR_ERR(res_chunk) : -ENOMEM; + return PTR_ERR(res_chunk); } for (i = 0; i < res_chunk->cnt; i++) { @@ -158,10 +158,10 @@ static int disable_qp_grp(struct usnic_ib_qp_grp *qp_grp) vnic_idx = usnic_vnic_get_index(qp_grp->vf->vnic); res_chunk = get_qp_res_chunk(qp_grp); - if (IS_ERR_OR_NULL(res_chunk)) { + if (IS_ERR(res_chunk)) { usnic_err("Unable to get qp res with err %ld\n", PTR_ERR(res_chunk)); - return res_chunk ? PTR_ERR(res_chunk) : -ENOMEM; + return PTR_ERR(res_chunk); } for (i = 0; i < res_chunk->cnt; i++) { @@ -186,11 +186,11 @@ static int init_filter_action(struct usnic_ib_qp_grp *qp_grp, struct usnic_vnic_res_chunk *res_chunk; res_chunk = usnic_ib_qp_grp_get_chunk(qp_grp, USNIC_VNIC_RES_TYPE_RQ); - if (IS_ERR_OR_NULL(res_chunk)) { + if (IS_ERR(res_chunk)) { usnic_err("Unable to get %s with err %ld\n", usnic_vnic_res_type_to_str(USNIC_VNIC_RES_TYPE_RQ), PTR_ERR(res_chunk)); - return res_chunk ? PTR_ERR(res_chunk) : -ENOMEM; + return PTR_ERR(res_chunk); } uaction->vnic_idx = usnic_vnic_get_index(qp_grp->vf->vnic); @@ -228,8 +228,6 @@ create_roce_custom_flow(struct usnic_ib_qp_grp *qp_grp, flow = usnic_fwd_alloc_flow(qp_grp->ufdev, &filter, &uaction); if (IS_ERR_OR_NULL(flow)) { - usnic_err("Unable to alloc flow failed with err %ld\n", - PTR_ERR(flow)); err = flow ? PTR_ERR(flow) : -EFAULT; goto out_unreserve_port; } @@ -303,8 +301,6 @@ create_udp_flow(struct usnic_ib_qp_grp *qp_grp, flow = usnic_fwd_alloc_flow(qp_grp->ufdev, &filter, &uaction); if (IS_ERR_OR_NULL(flow)) { - usnic_err("Unable to alloc flow failed with err %ld\n", - PTR_ERR(flow)); err = flow ? PTR_ERR(flow) : -EFAULT; goto out_put_sock; } @@ -694,18 +690,14 @@ usnic_ib_qp_grp_create(struct usnic_fwd_dev *ufdev, struct usnic_ib_vf *vf, } qp_grp = kzalloc(sizeof(*qp_grp), GFP_ATOMIC); - if (!qp_grp) { - usnic_err("Unable to alloc qp_grp - Out of memory\n"); + if (!qp_grp) return NULL; - } qp_grp->res_chunk_list = alloc_res_chunk_list(vf->vnic, res_spec, qp_grp); if (IS_ERR_OR_NULL(qp_grp->res_chunk_list)) { err = qp_grp->res_chunk_list ? PTR_ERR(qp_grp->res_chunk_list) : -ENOMEM; - usnic_err("Unable to alloc res for %d with err %d\n", - qp_grp->grp_id, err); goto out_free_qp_grp; } diff --git a/drivers/infiniband/hw/usnic/usnic_ib_sysfs.c b/drivers/infiniband/hw/usnic/usnic_ib_sysfs.c index 80ef3f8998c8..04443242e258 100644 --- a/drivers/infiniband/hw/usnic/usnic_ib_sysfs.c +++ b/drivers/infiniband/hw/usnic/usnic_ib_sysfs.c @@ -80,7 +80,7 @@ usnic_ib_show_config(struct device *device, struct device_attribute *attr, left = PAGE_SIZE; mutex_lock(&us_ibdev->usdev_lock); - if (atomic_read(&us_ibdev->vf_cnt.refcount) > 0) { + if (kref_read(&us_ibdev->vf_cnt) > 0) { char *busname; /* @@ -99,7 +99,7 @@ usnic_ib_show_config(struct device *device, struct device_attribute *attr, PCI_FUNC(us_ibdev->pdev->devfn), netdev_name(us_ibdev->netdev), us_ibdev->ufdev->mac, - atomic_read(&us_ibdev->vf_cnt.refcount)); + kref_read(&us_ibdev->vf_cnt)); UPDATE_PTR_LEFT(n, ptr, left); for (res_type = USNIC_VNIC_RES_TYPE_EOL; @@ -147,7 +147,7 @@ usnic_ib_show_max_vf(struct device *device, struct device_attribute *attr, us_ibdev = container_of(device, struct usnic_ib_dev, ib_dev.dev); return scnprintf(buf, PAGE_SIZE, "%u\n", - atomic_read(&us_ibdev->vf_cnt.refcount)); + kref_read(&us_ibdev->vf_cnt)); } static ssize_t diff --git a/drivers/infiniband/hw/usnic/usnic_ib_verbs.c b/drivers/infiniband/hw/usnic/usnic_ib_verbs.c index a5bfbba6bbac..3284730d3c09 100644 --- a/drivers/infiniband/hw/usnic/usnic_ib_verbs.c +++ b/drivers/infiniband/hw/usnic/usnic_ib_verbs.c @@ -87,12 +87,12 @@ static int usnic_ib_fill_create_qp_resp(struct usnic_ib_qp_grp *qp_grp, resp.bar_len = bar->len; chunk = usnic_ib_qp_grp_get_chunk(qp_grp, USNIC_VNIC_RES_TYPE_RQ); - if (IS_ERR_OR_NULL(chunk)) { + if (IS_ERR(chunk)) { usnic_err("Failed to get chunk %s for qp_grp %d with err %ld\n", usnic_vnic_res_type_to_str(USNIC_VNIC_RES_TYPE_RQ), qp_grp->grp_id, PTR_ERR(chunk)); - return chunk ? PTR_ERR(chunk) : -ENOMEM; + return PTR_ERR(chunk); } WARN_ON(chunk->type != USNIC_VNIC_RES_TYPE_RQ); @@ -101,12 +101,12 @@ static int usnic_ib_fill_create_qp_resp(struct usnic_ib_qp_grp *qp_grp, resp.rq_idx[i] = chunk->res[i]->vnic_idx; chunk = usnic_ib_qp_grp_get_chunk(qp_grp, USNIC_VNIC_RES_TYPE_WQ); - if (IS_ERR_OR_NULL(chunk)) { + if (IS_ERR(chunk)) { usnic_err("Failed to get chunk %s for qp_grp %d with err %ld\n", usnic_vnic_res_type_to_str(USNIC_VNIC_RES_TYPE_WQ), qp_grp->grp_id, PTR_ERR(chunk)); - return chunk ? PTR_ERR(chunk) : -ENOMEM; + return PTR_ERR(chunk); } WARN_ON(chunk->type != USNIC_VNIC_RES_TYPE_WQ); @@ -115,12 +115,12 @@ static int usnic_ib_fill_create_qp_resp(struct usnic_ib_qp_grp *qp_grp, resp.wq_idx[i] = chunk->res[i]->vnic_idx; chunk = usnic_ib_qp_grp_get_chunk(qp_grp, USNIC_VNIC_RES_TYPE_CQ); - if (IS_ERR_OR_NULL(chunk)) { + if (IS_ERR(chunk)) { usnic_err("Failed to get chunk %s for qp_grp %d with err %ld\n", usnic_vnic_res_type_to_str(USNIC_VNIC_RES_TYPE_CQ), qp_grp->grp_id, PTR_ERR(chunk)); - return chunk ? PTR_ERR(chunk) : -ENOMEM; + return PTR_ERR(chunk); } WARN_ON(chunk->type != USNIC_VNIC_RES_TYPE_CQ); @@ -291,11 +291,11 @@ int usnic_ib_query_device(struct ib_device *ibdev, qp_per_vf = max(us_ibdev->vf_res_cnt[USNIC_VNIC_RES_TYPE_WQ], us_ibdev->vf_res_cnt[USNIC_VNIC_RES_TYPE_RQ]); props->max_qp = qp_per_vf * - atomic_read(&us_ibdev->vf_cnt.refcount); + kref_read(&us_ibdev->vf_cnt); props->device_cap_flags = IB_DEVICE_PORT_ACTIVE_EVENT | IB_DEVICE_SYS_IMAGE_GUID | IB_DEVICE_BLOCK_MULTICAST_LOOPBACK; props->max_cq = us_ibdev->vf_res_cnt[USNIC_VNIC_RES_TYPE_CQ] * - atomic_read(&us_ibdev->vf_cnt.refcount); + kref_read(&us_ibdev->vf_cnt); props->max_pd = USNIC_UIOM_MAX_PD_CNT; props->max_mr = USNIC_UIOM_MAX_MR_CNT; props->local_ca_ack_delay = 0; @@ -330,7 +330,7 @@ int usnic_ib_query_port(struct ib_device *ibdev, u8 port, mutex_lock(&us_ibdev->usdev_lock); __ethtool_get_link_ksettings(us_ibdev->netdev, &cmd); - memset(props, 0, sizeof(*props)); + /* props being zeroed by the caller, avoid zeroing it here */ props->lid = 0; props->lmc = 1; @@ -738,7 +738,9 @@ int usnic_ib_mmap(struct ib_ucontext *context, /* In ib callbacks section - Start of stub funcs */ struct ib_ah *usnic_ib_create_ah(struct ib_pd *pd, - struct ib_ah_attr *ah_attr) + struct ib_ah_attr *ah_attr, + struct ib_udata *udata) + { usnic_dbg("\n"); return ERR_PTR(-EPERM); diff --git a/drivers/infiniband/hw/usnic/usnic_ib_verbs.h b/drivers/infiniband/hw/usnic/usnic_ib_verbs.h index 0d9d2e6a14d5..0ed8e072329e 100644 --- a/drivers/infiniband/hw/usnic/usnic_ib_verbs.h +++ b/drivers/infiniband/hw/usnic/usnic_ib_verbs.h @@ -75,7 +75,9 @@ int usnic_ib_dealloc_ucontext(struct ib_ucontext *ibcontext); int usnic_ib_mmap(struct ib_ucontext *context, struct vm_area_struct *vma); struct ib_ah *usnic_ib_create_ah(struct ib_pd *pd, - struct ib_ah_attr *ah_attr); + struct ib_ah_attr *ah_attr, + struct ib_udata *udata); + int usnic_ib_destroy_ah(struct ib_ah *ah); int usnic_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, struct ib_send_wr **bad_wr); diff --git a/drivers/infiniband/hw/usnic/usnic_vnic.c b/drivers/infiniband/hw/usnic/usnic_vnic.c index 887510718690..e7b0030254da 100644 --- a/drivers/infiniband/hw/usnic/usnic_vnic.c +++ b/drivers/infiniband/hw/usnic/usnic_vnic.c @@ -241,17 +241,12 @@ usnic_vnic_get_resources(struct usnic_vnic *vnic, enum usnic_vnic_res_type type, return ERR_PTR(-EINVAL); ret = kzalloc(sizeof(*ret), GFP_ATOMIC); - if (!ret) { - usnic_err("Failed to allocate chunk for %s - Out of memory\n", - usnic_vnic_pci_name(vnic)); + if (!ret) return ERR_PTR(-ENOMEM); - } if (cnt > 0) { ret->res = kcalloc(cnt, sizeof(*(ret->res)), GFP_ATOMIC); if (!ret->res) { - usnic_err("Failed to allocate resources for %s. Out of memory\n", - usnic_vnic_pci_name(vnic)); kfree(ret); return ERR_PTR(-ENOMEM); } @@ -311,8 +306,10 @@ static int usnic_vnic_alloc_res_chunk(struct usnic_vnic *vnic, struct usnic_vnic_res *res; cnt = vnic_dev_get_res_count(vnic->vdev, _to_vnic_res_type(type)); - if (cnt < 1) + if (cnt < 1) { + usnic_err("Wrong res count with cnt %d\n", cnt); return -EINVAL; + } chunk->cnt = chunk->free_cnt = cnt; chunk->res = kzalloc(sizeof(*(chunk->res))*cnt, GFP_KERNEL); @@ -384,12 +381,8 @@ static int usnic_vnic_discover_resources(struct pci_dev *pdev, res_type < USNIC_VNIC_RES_TYPE_MAX; res_type++) { err = usnic_vnic_alloc_res_chunk(vnic, res_type, &vnic->chunks[res_type]); - if (err) { - usnic_err("Failed to alloc res %s with err %d\n", - usnic_vnic_res_type_to_str(res_type), - err); + if (err) goto out_clean_chunks; - } } return 0; @@ -454,11 +447,8 @@ struct usnic_vnic *usnic_vnic_alloc(struct pci_dev *pdev) } vnic = kzalloc(sizeof(*vnic), GFP_KERNEL); - if (!vnic) { - usnic_err("Failed to alloc vnic for %s - out of memory\n", - pci_name(pdev)); + if (!vnic) return ERR_PTR(-ENOMEM); - } spin_lock_init(&vnic->res_lock); diff --git a/drivers/infiniband/hw/vmw_pvrdma/Kconfig b/drivers/infiniband/hw/vmw_pvrdma/Kconfig new file mode 100644 index 000000000000..5a9790ac0ede --- /dev/null +++ b/drivers/infiniband/hw/vmw_pvrdma/Kconfig @@ -0,0 +1,7 @@ +config INFINIBAND_VMWARE_PVRDMA + tristate "VMware Paravirtualized RDMA Driver" + depends on NETDEVICES && ETHERNET && PCI && INET && VMXNET3 + ---help--- + This driver provides low-level support for VMware Paravirtual + RDMA adapter. It interacts with the VMXNet3 driver to provide + Ethernet capabilities. diff --git a/drivers/infiniband/hw/vmw_pvrdma/Makefile b/drivers/infiniband/hw/vmw_pvrdma/Makefile new file mode 100644 index 000000000000..0194ed19f542 --- /dev/null +++ b/drivers/infiniband/hw/vmw_pvrdma/Makefile @@ -0,0 +1,3 @@ +obj-$(CONFIG_INFINIBAND_VMWARE_PVRDMA) += vmw_pvrdma.o + +vmw_pvrdma-y := pvrdma_cmd.o pvrdma_cq.o pvrdma_doorbell.o pvrdma_main.o pvrdma_misc.o pvrdma_mr.o pvrdma_qp.o pvrdma_verbs.o diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma.h b/drivers/infiniband/hw/vmw_pvrdma/pvrdma.h new file mode 100644 index 000000000000..3cd96c1b9502 --- /dev/null +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma.h @@ -0,0 +1,468 @@ +/* + * Copyright (c) 2012-2016 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of EITHER the GNU General Public License + * version 2 as published by the Free Software Foundation or the BSD + * 2-Clause License. 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 version 2 for more details at + * http://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html. + * + * You should have received a copy of the GNU General Public License + * along with this program available in the file COPYING in the main + * directory of this source tree. + * + * The BSD 2-Clause License + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE + * COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef __PVRDMA_H__ +#define __PVRDMA_H__ + +#include <linux/compiler.h> +#include <linux/interrupt.h> +#include <linux/list.h> +#include <linux/mutex.h> +#include <linux/pci.h> +#include <linux/semaphore.h> +#include <linux/workqueue.h> +#include <rdma/ib_umem.h> +#include <rdma/ib_verbs.h> +#include <rdma/vmw_pvrdma-abi.h> + +#include "pvrdma_ring.h" +#include "pvrdma_dev_api.h" +#include "pvrdma_verbs.h" + +/* NOT the same as BIT_MASK(). */ +#define PVRDMA_MASK(n) ((n << 1) - 1) + +/* + * VMware PVRDMA PCI device id. + */ +#define PCI_DEVICE_ID_VMWARE_PVRDMA 0x0820 + +struct pvrdma_dev; + +struct pvrdma_page_dir { + dma_addr_t dir_dma; + u64 *dir; + int ntables; + u64 **tables; + u64 npages; + void **pages; +}; + +struct pvrdma_cq { + struct ib_cq ibcq; + int offset; + spinlock_t cq_lock; /* Poll lock. */ + struct pvrdma_uar_map *uar; + struct ib_umem *umem; + struct pvrdma_ring_state *ring_state; + struct pvrdma_page_dir pdir; + u32 cq_handle; + bool is_kernel; + atomic_t refcnt; + wait_queue_head_t wait; +}; + +struct pvrdma_id_table { + u32 last; + u32 top; + u32 max; + u32 mask; + spinlock_t lock; /* Table lock. */ + unsigned long *table; +}; + +struct pvrdma_uar_map { + unsigned long pfn; + void __iomem *map; + int index; +}; + +struct pvrdma_uar_table { + struct pvrdma_id_table tbl; + int size; +}; + +struct pvrdma_ucontext { + struct ib_ucontext ibucontext; + struct pvrdma_dev *dev; + struct pvrdma_uar_map uar; + u64 ctx_handle; +}; + +struct pvrdma_pd { + struct ib_pd ibpd; + u32 pdn; + u32 pd_handle; + int privileged; +}; + +struct pvrdma_mr { + u32 mr_handle; + u64 iova; + u64 size; +}; + +struct pvrdma_user_mr { + struct ib_mr ibmr; + struct ib_umem *umem; + struct pvrdma_mr mmr; + struct pvrdma_page_dir pdir; + u64 *pages; + u32 npages; + u32 max_pages; + u32 page_shift; +}; + +struct pvrdma_wq { + struct pvrdma_ring *ring; + spinlock_t lock; /* Work queue lock. */ + int wqe_cnt; + int wqe_size; + int max_sg; + int offset; +}; + +struct pvrdma_ah { + struct ib_ah ibah; + struct pvrdma_av av; +}; + +struct pvrdma_qp { + struct ib_qp ibqp; + u32 qp_handle; + u32 qkey; + struct pvrdma_wq sq; + struct pvrdma_wq rq; + struct ib_umem *rumem; + struct ib_umem *sumem; + struct pvrdma_page_dir pdir; + int npages; + int npages_send; + int npages_recv; + u32 flags; + u8 port; + u8 state; + bool is_kernel; + struct mutex mutex; /* QP state mutex. */ + atomic_t refcnt; + wait_queue_head_t wait; +}; + +struct pvrdma_dev { + /* PCI device-related information. */ + struct ib_device ib_dev; + struct pci_dev *pdev; + void __iomem *regs; + struct pvrdma_device_shared_region *dsr; /* Shared region pointer */ + dma_addr_t dsrbase; /* Shared region base address */ + void *cmd_slot; + void *resp_slot; + unsigned long flags; + struct list_head device_link; + + /* Locking and interrupt information. */ + spinlock_t cmd_lock; /* Command lock. */ + struct semaphore cmd_sema; + struct completion cmd_done; + unsigned int nr_vectors; + + /* RDMA-related device information. */ + union ib_gid *sgid_tbl; + struct pvrdma_ring_state *async_ring_state; + struct pvrdma_page_dir async_pdir; + struct pvrdma_ring_state *cq_ring_state; + struct pvrdma_page_dir cq_pdir; + struct pvrdma_cq **cq_tbl; + spinlock_t cq_tbl_lock; + struct pvrdma_qp **qp_tbl; + spinlock_t qp_tbl_lock; + struct pvrdma_uar_table uar_table; + struct pvrdma_uar_map driver_uar; + __be64 sys_image_guid; + spinlock_t desc_lock; /* Device modification lock. */ + u32 port_cap_mask; + struct mutex port_mutex; /* Port modification mutex. */ + bool ib_active; + atomic_t num_qps; + atomic_t num_cqs; + atomic_t num_pds; + atomic_t num_ahs; + + /* Network device information. */ + struct net_device *netdev; + struct notifier_block nb_netdev; +}; + +struct pvrdma_netdevice_work { + struct work_struct work; + struct net_device *event_netdev; + unsigned long event; +}; + +static inline struct pvrdma_dev *to_vdev(struct ib_device *ibdev) +{ + return container_of(ibdev, struct pvrdma_dev, ib_dev); +} + +static inline struct +pvrdma_ucontext *to_vucontext(struct ib_ucontext *ibucontext) +{ + return container_of(ibucontext, struct pvrdma_ucontext, ibucontext); +} + +static inline struct pvrdma_pd *to_vpd(struct ib_pd *ibpd) +{ + return container_of(ibpd, struct pvrdma_pd, ibpd); +} + +static inline struct pvrdma_cq *to_vcq(struct ib_cq *ibcq) +{ + return container_of(ibcq, struct pvrdma_cq, ibcq); +} + +static inline struct pvrdma_user_mr *to_vmr(struct ib_mr *ibmr) +{ + return container_of(ibmr, struct pvrdma_user_mr, ibmr); +} + +static inline struct pvrdma_qp *to_vqp(struct ib_qp *ibqp) +{ + return container_of(ibqp, struct pvrdma_qp, ibqp); +} + +static inline struct pvrdma_ah *to_vah(struct ib_ah *ibah) +{ + return container_of(ibah, struct pvrdma_ah, ibah); +} + +static inline void pvrdma_write_reg(struct pvrdma_dev *dev, u32 reg, u32 val) +{ + writel(cpu_to_le32(val), dev->regs + reg); +} + +static inline u32 pvrdma_read_reg(struct pvrdma_dev *dev, u32 reg) +{ + return le32_to_cpu(readl(dev->regs + reg)); +} + +static inline void pvrdma_write_uar_cq(struct pvrdma_dev *dev, u32 val) +{ + writel(cpu_to_le32(val), dev->driver_uar.map + PVRDMA_UAR_CQ_OFFSET); +} + +static inline void pvrdma_write_uar_qp(struct pvrdma_dev *dev, u32 val) +{ + writel(cpu_to_le32(val), dev->driver_uar.map + PVRDMA_UAR_QP_OFFSET); +} + +static inline void *pvrdma_page_dir_get_ptr(struct pvrdma_page_dir *pdir, + u64 offset) +{ + return pdir->pages[offset / PAGE_SIZE] + (offset % PAGE_SIZE); +} + +static inline enum pvrdma_mtu ib_mtu_to_pvrdma(enum ib_mtu mtu) +{ + return (enum pvrdma_mtu)mtu; +} + +static inline enum ib_mtu pvrdma_mtu_to_ib(enum pvrdma_mtu mtu) +{ + return (enum ib_mtu)mtu; +} + +static inline enum pvrdma_port_state ib_port_state_to_pvrdma( + enum ib_port_state state) +{ + return (enum pvrdma_port_state)state; +} + +static inline enum ib_port_state pvrdma_port_state_to_ib( + enum pvrdma_port_state state) +{ + return (enum ib_port_state)state; +} + +static inline int ib_port_cap_flags_to_pvrdma(int flags) +{ + return flags & PVRDMA_MASK(PVRDMA_PORT_CAP_FLAGS_MAX); +} + +static inline int pvrdma_port_cap_flags_to_ib(int flags) +{ + return flags; +} + +static inline enum pvrdma_port_width ib_port_width_to_pvrdma( + enum ib_port_width width) +{ + return (enum pvrdma_port_width)width; +} + +static inline enum ib_port_width pvrdma_port_width_to_ib( + enum pvrdma_port_width width) +{ + return (enum ib_port_width)width; +} + +static inline enum pvrdma_port_speed ib_port_speed_to_pvrdma( + enum ib_port_speed speed) +{ + return (enum pvrdma_port_speed)speed; +} + +static inline enum ib_port_speed pvrdma_port_speed_to_ib( + enum pvrdma_port_speed speed) +{ + return (enum ib_port_speed)speed; +} + +static inline int pvrdma_qp_attr_mask_to_ib(int attr_mask) +{ + return attr_mask; +} + +static inline int ib_qp_attr_mask_to_pvrdma(int attr_mask) +{ + return attr_mask & PVRDMA_MASK(PVRDMA_QP_ATTR_MASK_MAX); +} + +static inline enum pvrdma_mig_state ib_mig_state_to_pvrdma( + enum ib_mig_state state) +{ + return (enum pvrdma_mig_state)state; +} + +static inline enum ib_mig_state pvrdma_mig_state_to_ib( + enum pvrdma_mig_state state) +{ + return (enum ib_mig_state)state; +} + +static inline int ib_access_flags_to_pvrdma(int flags) +{ + return flags; +} + +static inline int pvrdma_access_flags_to_ib(int flags) +{ + return flags & PVRDMA_MASK(PVRDMA_ACCESS_FLAGS_MAX); +} + +static inline enum pvrdma_qp_type ib_qp_type_to_pvrdma(enum ib_qp_type type) +{ + return (enum pvrdma_qp_type)type; +} + +static inline enum ib_qp_type pvrdma_qp_type_to_ib(enum pvrdma_qp_type type) +{ + return (enum ib_qp_type)type; +} + +static inline enum pvrdma_qp_state ib_qp_state_to_pvrdma(enum ib_qp_state state) +{ + return (enum pvrdma_qp_state)state; +} + +static inline enum ib_qp_state pvrdma_qp_state_to_ib(enum pvrdma_qp_state state) +{ + return (enum ib_qp_state)state; +} + +static inline enum pvrdma_wr_opcode ib_wr_opcode_to_pvrdma(enum ib_wr_opcode op) +{ + return (enum pvrdma_wr_opcode)op; +} + +static inline enum ib_wc_status pvrdma_wc_status_to_ib( + enum pvrdma_wc_status status) +{ + return (enum ib_wc_status)status; +} + +static inline int pvrdma_wc_opcode_to_ib(int opcode) +{ + return opcode; +} + +static inline int pvrdma_wc_flags_to_ib(int flags) +{ + return flags; +} + +static inline int ib_send_flags_to_pvrdma(int flags) +{ + return flags & PVRDMA_MASK(PVRDMA_SEND_FLAGS_MAX); +} + +void pvrdma_qp_cap_to_ib(struct ib_qp_cap *dst, + const struct pvrdma_qp_cap *src); +void ib_qp_cap_to_pvrdma(struct pvrdma_qp_cap *dst, + const struct ib_qp_cap *src); +void pvrdma_gid_to_ib(union ib_gid *dst, const union pvrdma_gid *src); +void ib_gid_to_pvrdma(union pvrdma_gid *dst, const union ib_gid *src); +void pvrdma_global_route_to_ib(struct ib_global_route *dst, + const struct pvrdma_global_route *src); +void ib_global_route_to_pvrdma(struct pvrdma_global_route *dst, + const struct ib_global_route *src); +void pvrdma_ah_attr_to_ib(struct ib_ah_attr *dst, + const struct pvrdma_ah_attr *src); +void ib_ah_attr_to_pvrdma(struct pvrdma_ah_attr *dst, + const struct ib_ah_attr *src); + +int pvrdma_uar_table_init(struct pvrdma_dev *dev); +void pvrdma_uar_table_cleanup(struct pvrdma_dev *dev); + +int pvrdma_uar_alloc(struct pvrdma_dev *dev, struct pvrdma_uar_map *uar); +void pvrdma_uar_free(struct pvrdma_dev *dev, struct pvrdma_uar_map *uar); + +void _pvrdma_flush_cqe(struct pvrdma_qp *qp, struct pvrdma_cq *cq); + +int pvrdma_page_dir_init(struct pvrdma_dev *dev, struct pvrdma_page_dir *pdir, + u64 npages, bool alloc_pages); +void pvrdma_page_dir_cleanup(struct pvrdma_dev *dev, + struct pvrdma_page_dir *pdir); +int pvrdma_page_dir_insert_dma(struct pvrdma_page_dir *pdir, u64 idx, + dma_addr_t daddr); +int pvrdma_page_dir_insert_umem(struct pvrdma_page_dir *pdir, + struct ib_umem *umem, u64 offset); +dma_addr_t pvrdma_page_dir_get_dma(struct pvrdma_page_dir *pdir, u64 idx); +int pvrdma_page_dir_insert_page_list(struct pvrdma_page_dir *pdir, + u64 *page_list, int num_pages); + +int pvrdma_cmd_post(struct pvrdma_dev *dev, union pvrdma_cmd_req *req, + union pvrdma_cmd_resp *rsp, unsigned resp_code); + +#endif /* __PVRDMA_H__ */ diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_cmd.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_cmd.c new file mode 100644 index 000000000000..4a78c537d8a1 --- /dev/null +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_cmd.c @@ -0,0 +1,119 @@ +/* + * Copyright (c) 2012-2016 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of EITHER the GNU General Public License + * version 2 as published by the Free Software Foundation or the BSD + * 2-Clause License. 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 version 2 for more details at + * http://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html. + * + * You should have received a copy of the GNU General Public License + * along with this program available in the file COPYING in the main + * directory of this source tree. + * + * The BSD 2-Clause License + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE + * COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <linux/list.h> + +#include "pvrdma.h" + +#define PVRDMA_CMD_TIMEOUT 10000 /* ms */ + +static inline int pvrdma_cmd_recv(struct pvrdma_dev *dev, + union pvrdma_cmd_resp *resp, + unsigned resp_code) +{ + int err; + + dev_dbg(&dev->pdev->dev, "receive response from device\n"); + + err = wait_for_completion_interruptible_timeout(&dev->cmd_done, + msecs_to_jiffies(PVRDMA_CMD_TIMEOUT)); + if (err == 0 || err == -ERESTARTSYS) { + dev_warn(&dev->pdev->dev, + "completion timeout or interrupted\n"); + return -ETIMEDOUT; + } + + spin_lock(&dev->cmd_lock); + memcpy(resp, dev->resp_slot, sizeof(*resp)); + spin_unlock(&dev->cmd_lock); + + if (resp->hdr.ack != resp_code) { + dev_warn(&dev->pdev->dev, + "unknown response %#x expected %#x\n", + resp->hdr.ack, resp_code); + return -EFAULT; + } + + return 0; +} + +int +pvrdma_cmd_post(struct pvrdma_dev *dev, union pvrdma_cmd_req *req, + union pvrdma_cmd_resp *resp, unsigned resp_code) +{ + int err; + + dev_dbg(&dev->pdev->dev, "post request to device\n"); + + /* Serializiation */ + down(&dev->cmd_sema); + + BUILD_BUG_ON(sizeof(union pvrdma_cmd_req) != + sizeof(struct pvrdma_cmd_modify_qp)); + + spin_lock(&dev->cmd_lock); + memcpy(dev->cmd_slot, req, sizeof(*req)); + spin_unlock(&dev->cmd_lock); + + init_completion(&dev->cmd_done); + pvrdma_write_reg(dev, PVRDMA_REG_REQUEST, 0); + + /* Make sure the request is written before reading status. */ + mb(); + + err = pvrdma_read_reg(dev, PVRDMA_REG_ERR); + if (err == 0) { + if (resp != NULL) + err = pvrdma_cmd_recv(dev, resp, resp_code); + } else { + dev_warn(&dev->pdev->dev, + "failed to write request error reg: %d\n", err); + err = -EFAULT; + } + + up(&dev->cmd_sema); + + return err; +} diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_cq.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_cq.c new file mode 100644 index 000000000000..69bda611d313 --- /dev/null +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_cq.c @@ -0,0 +1,425 @@ +/* + * Copyright (c) 2012-2016 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of EITHER the GNU General Public License + * version 2 as published by the Free Software Foundation or the BSD + * 2-Clause License. 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 version 2 for more details at + * http://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html. + * + * You should have received a copy of the GNU General Public License + * along with this program available in the file COPYING in the main + * directory of this source tree. + * + * The BSD 2-Clause License + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE + * COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <asm/page.h> +#include <linux/io.h> +#include <linux/wait.h> +#include <rdma/ib_addr.h> +#include <rdma/ib_smi.h> +#include <rdma/ib_user_verbs.h> + +#include "pvrdma.h" + +/** + * pvrdma_req_notify_cq - request notification for a completion queue + * @ibcq: the completion queue + * @notify_flags: notification flags + * + * @return: 0 for success. + */ +int pvrdma_req_notify_cq(struct ib_cq *ibcq, + enum ib_cq_notify_flags notify_flags) +{ + struct pvrdma_dev *dev = to_vdev(ibcq->device); + struct pvrdma_cq *cq = to_vcq(ibcq); + u32 val = cq->cq_handle; + + val |= (notify_flags & IB_CQ_SOLICITED_MASK) == IB_CQ_SOLICITED ? + PVRDMA_UAR_CQ_ARM_SOL : PVRDMA_UAR_CQ_ARM; + + pvrdma_write_uar_cq(dev, val); + + return 0; +} + +/** + * pvrdma_create_cq - create completion queue + * @ibdev: the device + * @attr: completion queue attributes + * @context: user context + * @udata: user data + * + * @return: ib_cq completion queue pointer on success, + * otherwise returns negative errno. + */ +struct ib_cq *pvrdma_create_cq(struct ib_device *ibdev, + const struct ib_cq_init_attr *attr, + struct ib_ucontext *context, + struct ib_udata *udata) +{ + int entries = attr->cqe; + struct pvrdma_dev *dev = to_vdev(ibdev); + struct pvrdma_cq *cq; + int ret; + int npages; + unsigned long flags; + union pvrdma_cmd_req req; + union pvrdma_cmd_resp rsp; + struct pvrdma_cmd_create_cq *cmd = &req.create_cq; + struct pvrdma_cmd_create_cq_resp *resp = &rsp.create_cq_resp; + struct pvrdma_create_cq ucmd; + + BUILD_BUG_ON(sizeof(struct pvrdma_cqe) != 64); + + entries = roundup_pow_of_two(entries); + if (entries < 1 || entries > dev->dsr->caps.max_cqe) + return ERR_PTR(-EINVAL); + + if (!atomic_add_unless(&dev->num_cqs, 1, dev->dsr->caps.max_cq)) + return ERR_PTR(-ENOMEM); + + cq = kzalloc(sizeof(*cq), GFP_KERNEL); + if (!cq) { + atomic_dec(&dev->num_cqs); + return ERR_PTR(-ENOMEM); + } + + cq->ibcq.cqe = entries; + + if (context) { + if (ib_copy_from_udata(&ucmd, udata, sizeof(ucmd))) { + ret = -EFAULT; + goto err_cq; + } + + cq->umem = ib_umem_get(context, ucmd.buf_addr, ucmd.buf_size, + IB_ACCESS_LOCAL_WRITE, 1); + if (IS_ERR(cq->umem)) { + ret = PTR_ERR(cq->umem); + goto err_cq; + } + + npages = ib_umem_page_count(cq->umem); + } else { + cq->is_kernel = true; + + /* One extra page for shared ring state */ + npages = 1 + (entries * sizeof(struct pvrdma_cqe) + + PAGE_SIZE - 1) / PAGE_SIZE; + + /* Skip header page. */ + cq->offset = PAGE_SIZE; + } + + if (npages < 0 || npages > PVRDMA_PAGE_DIR_MAX_PAGES) { + dev_warn(&dev->pdev->dev, + "overflow pages in completion queue\n"); + ret = -EINVAL; + goto err_umem; + } + + ret = pvrdma_page_dir_init(dev, &cq->pdir, npages, cq->is_kernel); + if (ret) { + dev_warn(&dev->pdev->dev, + "could not allocate page directory\n"); + goto err_umem; + } + + /* Ring state is always the first page. Set in library for user cq. */ + if (cq->is_kernel) + cq->ring_state = cq->pdir.pages[0]; + else + pvrdma_page_dir_insert_umem(&cq->pdir, cq->umem, 0); + + atomic_set(&cq->refcnt, 1); + init_waitqueue_head(&cq->wait); + spin_lock_init(&cq->cq_lock); + + memset(cmd, 0, sizeof(*cmd)); + cmd->hdr.cmd = PVRDMA_CMD_CREATE_CQ; + cmd->nchunks = npages; + cmd->ctx_handle = (context) ? + (u64)to_vucontext(context)->ctx_handle : 0; + cmd->cqe = entries; + cmd->pdir_dma = cq->pdir.dir_dma; + ret = pvrdma_cmd_post(dev, &req, &rsp, PVRDMA_CMD_CREATE_CQ_RESP); + if (ret < 0) { + dev_warn(&dev->pdev->dev, + "could not create completion queue, error: %d\n", ret); + goto err_page_dir; + } + + cq->ibcq.cqe = resp->cqe; + cq->cq_handle = resp->cq_handle; + spin_lock_irqsave(&dev->cq_tbl_lock, flags); + dev->cq_tbl[cq->cq_handle % dev->dsr->caps.max_cq] = cq; + spin_unlock_irqrestore(&dev->cq_tbl_lock, flags); + + if (context) { + cq->uar = &(to_vucontext(context)->uar); + + /* Copy udata back. */ + if (ib_copy_to_udata(udata, &cq->cq_handle, sizeof(__u32))) { + dev_warn(&dev->pdev->dev, + "failed to copy back udata\n"); + pvrdma_destroy_cq(&cq->ibcq); + return ERR_PTR(-EINVAL); + } + } + + return &cq->ibcq; + +err_page_dir: + pvrdma_page_dir_cleanup(dev, &cq->pdir); +err_umem: + if (context) + ib_umem_release(cq->umem); +err_cq: + atomic_dec(&dev->num_cqs); + kfree(cq); + + return ERR_PTR(ret); +} + +static void pvrdma_free_cq(struct pvrdma_dev *dev, struct pvrdma_cq *cq) +{ + atomic_dec(&cq->refcnt); + wait_event(cq->wait, !atomic_read(&cq->refcnt)); + + if (!cq->is_kernel) + ib_umem_release(cq->umem); + + pvrdma_page_dir_cleanup(dev, &cq->pdir); + kfree(cq); +} + +/** + * pvrdma_destroy_cq - destroy completion queue + * @cq: the completion queue to destroy. + * + * @return: 0 for success. + */ +int pvrdma_destroy_cq(struct ib_cq *cq) +{ + struct pvrdma_cq *vcq = to_vcq(cq); + union pvrdma_cmd_req req; + struct pvrdma_cmd_destroy_cq *cmd = &req.destroy_cq; + struct pvrdma_dev *dev = to_vdev(cq->device); + unsigned long flags; + int ret; + + memset(cmd, 0, sizeof(*cmd)); + cmd->hdr.cmd = PVRDMA_CMD_DESTROY_CQ; + cmd->cq_handle = vcq->cq_handle; + + ret = pvrdma_cmd_post(dev, &req, NULL, 0); + if (ret < 0) + dev_warn(&dev->pdev->dev, + "could not destroy completion queue, error: %d\n", + ret); + + /* free cq's resources */ + spin_lock_irqsave(&dev->cq_tbl_lock, flags); + dev->cq_tbl[vcq->cq_handle] = NULL; + spin_unlock_irqrestore(&dev->cq_tbl_lock, flags); + + pvrdma_free_cq(dev, vcq); + atomic_dec(&dev->num_cqs); + + return ret; +} + +/** + * pvrdma_modify_cq - modify the CQ moderation parameters + * @ibcq: the CQ to modify + * @cq_count: number of CQEs that will trigger an event + * @cq_period: max period of time in usec before triggering an event + * + * @return: -EOPNOTSUPP as CQ resize is not supported. + */ +int pvrdma_modify_cq(struct ib_cq *cq, u16 cq_count, u16 cq_period) +{ + return -EOPNOTSUPP; +} + +static inline struct pvrdma_cqe *get_cqe(struct pvrdma_cq *cq, int i) +{ + return (struct pvrdma_cqe *)pvrdma_page_dir_get_ptr( + &cq->pdir, + cq->offset + + sizeof(struct pvrdma_cqe) * i); +} + +void _pvrdma_flush_cqe(struct pvrdma_qp *qp, struct pvrdma_cq *cq) +{ + int head; + int has_data; + + if (!cq->is_kernel) + return; + + /* Lock held */ + has_data = pvrdma_idx_ring_has_data(&cq->ring_state->rx, + cq->ibcq.cqe, &head); + if (unlikely(has_data > 0)) { + int items; + int curr; + int tail = pvrdma_idx(&cq->ring_state->rx.prod_tail, + cq->ibcq.cqe); + struct pvrdma_cqe *cqe; + struct pvrdma_cqe *curr_cqe; + + items = (tail > head) ? (tail - head) : + (cq->ibcq.cqe - head + tail); + curr = --tail; + while (items-- > 0) { + if (curr < 0) + curr = cq->ibcq.cqe - 1; + if (tail < 0) + tail = cq->ibcq.cqe - 1; + curr_cqe = get_cqe(cq, curr); + if ((curr_cqe->qp & 0xFFFF) != qp->qp_handle) { + if (curr != tail) { + cqe = get_cqe(cq, tail); + *cqe = *curr_cqe; + } + tail--; + } else { + pvrdma_idx_ring_inc( + &cq->ring_state->rx.cons_head, + cq->ibcq.cqe); + } + curr--; + } + } +} + +static int pvrdma_poll_one(struct pvrdma_cq *cq, struct pvrdma_qp **cur_qp, + struct ib_wc *wc) +{ + struct pvrdma_dev *dev = to_vdev(cq->ibcq.device); + int has_data; + unsigned int head; + bool tried = false; + struct pvrdma_cqe *cqe; + +retry: + has_data = pvrdma_idx_ring_has_data(&cq->ring_state->rx, + cq->ibcq.cqe, &head); + if (has_data == 0) { + if (tried) + return -EAGAIN; + + pvrdma_write_uar_cq(dev, cq->cq_handle | PVRDMA_UAR_CQ_POLL); + + tried = true; + goto retry; + } else if (has_data == PVRDMA_INVALID_IDX) { + dev_err(&dev->pdev->dev, "CQ ring state invalid\n"); + return -EAGAIN; + } + + cqe = get_cqe(cq, head); + + /* Ensure cqe is valid. */ + rmb(); + if (dev->qp_tbl[cqe->qp & 0xffff]) + *cur_qp = (struct pvrdma_qp *)dev->qp_tbl[cqe->qp & 0xffff]; + else + return -EAGAIN; + + wc->opcode = pvrdma_wc_opcode_to_ib(cqe->opcode); + wc->status = pvrdma_wc_status_to_ib(cqe->status); + wc->wr_id = cqe->wr_id; + wc->qp = &(*cur_qp)->ibqp; + wc->byte_len = cqe->byte_len; + wc->ex.imm_data = cqe->imm_data; + wc->src_qp = cqe->src_qp; + wc->wc_flags = pvrdma_wc_flags_to_ib(cqe->wc_flags); + wc->pkey_index = cqe->pkey_index; + wc->slid = cqe->slid; + wc->sl = cqe->sl; + wc->dlid_path_bits = cqe->dlid_path_bits; + wc->port_num = cqe->port_num; + wc->vendor_err = cqe->vendor_err; + + /* Update shared ring state */ + pvrdma_idx_ring_inc(&cq->ring_state->rx.cons_head, cq->ibcq.cqe); + + return 0; +} + +/** + * pvrdma_poll_cq - poll for work completion queue entries + * @ibcq: completion queue + * @num_entries: the maximum number of entries + * @entry: pointer to work completion array + * + * @return: number of polled completion entries + */ +int pvrdma_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *wc) +{ + struct pvrdma_cq *cq = to_vcq(ibcq); + struct pvrdma_qp *cur_qp = NULL; + unsigned long flags; + int npolled; + + if (num_entries < 1 || wc == NULL) + return 0; + + spin_lock_irqsave(&cq->cq_lock, flags); + for (npolled = 0; npolled < num_entries; ++npolled) { + if (pvrdma_poll_one(cq, &cur_qp, wc + npolled)) + break; + } + + spin_unlock_irqrestore(&cq->cq_lock, flags); + + /* Ensure we do not return errors from poll_cq */ + return npolled; +} + +/** + * pvrdma_resize_cq - resize CQ + * @ibcq: the completion queue + * @entries: CQ entries + * @udata: user data + * + * @return: -EOPNOTSUPP as CQ resize is not supported. + */ +int pvrdma_resize_cq(struct ib_cq *ibcq, int entries, struct ib_udata *udata) +{ + return -EOPNOTSUPP; +} diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_dev_api.h b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_dev_api.h new file mode 100644 index 000000000000..e69d6f3cae32 --- /dev/null +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_dev_api.h @@ -0,0 +1,580 @@ +/* + * Copyright (c) 2012-2016 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of EITHER the GNU General Public License + * version 2 as published by the Free Software Foundation or the BSD + * 2-Clause License. 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 version 2 for more details at + * http://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html. + * + * You should have received a copy of the GNU General Public License + * along with this program available in the file COPYING in the main + * directory of this source tree. + * + * The BSD 2-Clause License + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE + * COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef __PVRDMA_DEV_API_H__ +#define __PVRDMA_DEV_API_H__ + +#include <linux/types.h> + +#include "pvrdma_verbs.h" + +#define PVRDMA_VERSION 17 +#define PVRDMA_BOARD_ID 1 +#define PVRDMA_REV_ID 1 + +/* + * Masks and accessors for page directory, which is a two-level lookup: + * page directory -> page table -> page. Only one directory for now, but we + * could expand that easily. 9 bits for tables, 9 bits for pages, gives one + * gigabyte for memory regions and so forth. + */ + +#define PVRDMA_PDIR_SHIFT 18 +#define PVRDMA_PTABLE_SHIFT 9 +#define PVRDMA_PAGE_DIR_DIR(x) (((x) >> PVRDMA_PDIR_SHIFT) & 0x1) +#define PVRDMA_PAGE_DIR_TABLE(x) (((x) >> PVRDMA_PTABLE_SHIFT) & 0x1ff) +#define PVRDMA_PAGE_DIR_PAGE(x) ((x) & 0x1ff) +#define PVRDMA_PAGE_DIR_MAX_PAGES (1 * 512 * 512) +#define PVRDMA_MAX_FAST_REG_PAGES 128 + +/* + * Max MSI-X vectors. + */ + +#define PVRDMA_MAX_INTERRUPTS 3 + +/* Register offsets within PCI resource on BAR1. */ +#define PVRDMA_REG_VERSION 0x00 /* R: Version of device. */ +#define PVRDMA_REG_DSRLOW 0x04 /* W: Device shared region low PA. */ +#define PVRDMA_REG_DSRHIGH 0x08 /* W: Device shared region high PA. */ +#define PVRDMA_REG_CTL 0x0c /* W: PVRDMA_DEVICE_CTL */ +#define PVRDMA_REG_REQUEST 0x10 /* W: Indicate device request. */ +#define PVRDMA_REG_ERR 0x14 /* R: Device error. */ +#define PVRDMA_REG_ICR 0x18 /* R: Interrupt cause. */ +#define PVRDMA_REG_IMR 0x1c /* R/W: Interrupt mask. */ +#define PVRDMA_REG_MACL 0x20 /* R/W: MAC address low. */ +#define PVRDMA_REG_MACH 0x24 /* R/W: MAC address high. */ + +/* Object flags. */ +#define PVRDMA_CQ_FLAG_ARMED_SOL BIT(0) /* Armed for solicited-only. */ +#define PVRDMA_CQ_FLAG_ARMED BIT(1) /* Armed. */ +#define PVRDMA_MR_FLAG_DMA BIT(0) /* DMA region. */ +#define PVRDMA_MR_FLAG_FRMR BIT(1) /* Fast reg memory region. */ + +/* + * Atomic operation capability (masked versions are extended atomic + * operations. + */ + +#define PVRDMA_ATOMIC_OP_COMP_SWAP BIT(0) /* Compare and swap. */ +#define PVRDMA_ATOMIC_OP_FETCH_ADD BIT(1) /* Fetch and add. */ +#define PVRDMA_ATOMIC_OP_MASK_COMP_SWAP BIT(2) /* Masked compare and swap. */ +#define PVRDMA_ATOMIC_OP_MASK_FETCH_ADD BIT(3) /* Masked fetch and add. */ + +/* + * Base Memory Management Extension flags to support Fast Reg Memory Regions + * and Fast Reg Work Requests. Each flag represents a verb operation and we + * must support all of them to qualify for the BMME device cap. + */ + +#define PVRDMA_BMME_FLAG_LOCAL_INV BIT(0) /* Local Invalidate. */ +#define PVRDMA_BMME_FLAG_REMOTE_INV BIT(1) /* Remote Invalidate. */ +#define PVRDMA_BMME_FLAG_FAST_REG_WR BIT(2) /* Fast Reg Work Request. */ + +/* + * GID types. The interpretation of the gid_types bit field in the device + * capabilities will depend on the device mode. For now, the device only + * supports RoCE as mode, so only the different GID types for RoCE are + * defined. + */ + +#define PVRDMA_GID_TYPE_FLAG_ROCE_V1 BIT(0) +#define PVRDMA_GID_TYPE_FLAG_ROCE_V2 BIT(1) + +enum pvrdma_pci_resource { + PVRDMA_PCI_RESOURCE_MSIX, /* BAR0: MSI-X, MMIO. */ + PVRDMA_PCI_RESOURCE_REG, /* BAR1: Registers, MMIO. */ + PVRDMA_PCI_RESOURCE_UAR, /* BAR2: UAR pages, MMIO, 64-bit. */ + PVRDMA_PCI_RESOURCE_LAST, /* Last. */ +}; + +enum pvrdma_device_ctl { + PVRDMA_DEVICE_CTL_ACTIVATE, /* Activate device. */ + PVRDMA_DEVICE_CTL_QUIESCE, /* Quiesce device. */ + PVRDMA_DEVICE_CTL_RESET, /* Reset device. */ +}; + +enum pvrdma_intr_vector { + PVRDMA_INTR_VECTOR_RESPONSE, /* Command response. */ + PVRDMA_INTR_VECTOR_ASYNC, /* Async events. */ + PVRDMA_INTR_VECTOR_CQ, /* CQ notification. */ + /* Additional CQ notification vectors. */ +}; + +enum pvrdma_intr_cause { + PVRDMA_INTR_CAUSE_RESPONSE = (1 << PVRDMA_INTR_VECTOR_RESPONSE), + PVRDMA_INTR_CAUSE_ASYNC = (1 << PVRDMA_INTR_VECTOR_ASYNC), + PVRDMA_INTR_CAUSE_CQ = (1 << PVRDMA_INTR_VECTOR_CQ), +}; + +enum pvrdma_gos_bits { + PVRDMA_GOS_BITS_UNK, /* Unknown. */ + PVRDMA_GOS_BITS_32, /* 32-bit. */ + PVRDMA_GOS_BITS_64, /* 64-bit. */ +}; + +enum pvrdma_gos_type { + PVRDMA_GOS_TYPE_UNK, /* Unknown. */ + PVRDMA_GOS_TYPE_LINUX, /* Linux. */ +}; + +enum pvrdma_device_mode { + PVRDMA_DEVICE_MODE_ROCE, /* RoCE. */ + PVRDMA_DEVICE_MODE_IWARP, /* iWarp. */ + PVRDMA_DEVICE_MODE_IB, /* InfiniBand. */ +}; + +struct pvrdma_gos_info { + u32 gos_bits:2; /* W: PVRDMA_GOS_BITS_ */ + u32 gos_type:4; /* W: PVRDMA_GOS_TYPE_ */ + u32 gos_ver:16; /* W: Guest OS version. */ + u32 gos_misc:10; /* W: Other. */ + u32 pad; /* Pad to 8-byte alignment. */ +}; + +struct pvrdma_device_caps { + u64 fw_ver; /* R: Query device. */ + __be64 node_guid; + __be64 sys_image_guid; + u64 max_mr_size; + u64 page_size_cap; + u64 atomic_arg_sizes; /* EX verbs. */ + u32 ex_comp_mask; /* EX verbs. */ + u32 device_cap_flags2; /* EX verbs. */ + u32 max_fa_bit_boundary; /* EX verbs. */ + u32 log_max_atomic_inline_arg; /* EX verbs. */ + u32 vendor_id; + u32 vendor_part_id; + u32 hw_ver; + u32 max_qp; + u32 max_qp_wr; + u32 device_cap_flags; + u32 max_sge; + u32 max_sge_rd; + u32 max_cq; + u32 max_cqe; + u32 max_mr; + u32 max_pd; + u32 max_qp_rd_atom; + u32 max_ee_rd_atom; + u32 max_res_rd_atom; + u32 max_qp_init_rd_atom; + u32 max_ee_init_rd_atom; + u32 max_ee; + u32 max_rdd; + u32 max_mw; + u32 max_raw_ipv6_qp; + u32 max_raw_ethy_qp; + u32 max_mcast_grp; + u32 max_mcast_qp_attach; + u32 max_total_mcast_qp_attach; + u32 max_ah; + u32 max_fmr; + u32 max_map_per_fmr; + u32 max_srq; + u32 max_srq_wr; + u32 max_srq_sge; + u32 max_uar; + u32 gid_tbl_len; + u16 max_pkeys; + u8 local_ca_ack_delay; + u8 phys_port_cnt; + u8 mode; /* PVRDMA_DEVICE_MODE_ */ + u8 atomic_ops; /* PVRDMA_ATOMIC_OP_* bits */ + u8 bmme_flags; /* FRWR Mem Mgmt Extensions */ + u8 gid_types; /* PVRDMA_GID_TYPE_FLAG_ */ + u8 reserved[4]; +}; + +struct pvrdma_ring_page_info { + u32 num_pages; /* Num pages incl. header. */ + u32 reserved; /* Reserved. */ + u64 pdir_dma; /* Page directory PA. */ +}; + +#pragma pack(push, 1) + +struct pvrdma_device_shared_region { + u32 driver_version; /* W: Driver version. */ + u32 pad; /* Pad to 8-byte align. */ + struct pvrdma_gos_info gos_info; /* W: Guest OS information. */ + u64 cmd_slot_dma; /* W: Command slot address. */ + u64 resp_slot_dma; /* W: Response slot address. */ + struct pvrdma_ring_page_info async_ring_pages; + /* W: Async ring page info. */ + struct pvrdma_ring_page_info cq_ring_pages; + /* W: CQ ring page info. */ + u32 uar_pfn; /* W: UAR pageframe. */ + u32 pad2; /* Pad to 8-byte align. */ + struct pvrdma_device_caps caps; /* R: Device capabilities. */ +}; + +#pragma pack(pop) + +/* Event types. Currently a 1:1 mapping with enum ib_event. */ +enum pvrdma_eqe_type { + PVRDMA_EVENT_CQ_ERR, + PVRDMA_EVENT_QP_FATAL, + PVRDMA_EVENT_QP_REQ_ERR, + PVRDMA_EVENT_QP_ACCESS_ERR, + PVRDMA_EVENT_COMM_EST, + PVRDMA_EVENT_SQ_DRAINED, + PVRDMA_EVENT_PATH_MIG, + PVRDMA_EVENT_PATH_MIG_ERR, + PVRDMA_EVENT_DEVICE_FATAL, + PVRDMA_EVENT_PORT_ACTIVE, + PVRDMA_EVENT_PORT_ERR, + PVRDMA_EVENT_LID_CHANGE, + PVRDMA_EVENT_PKEY_CHANGE, + PVRDMA_EVENT_SM_CHANGE, + PVRDMA_EVENT_SRQ_ERR, + PVRDMA_EVENT_SRQ_LIMIT_REACHED, + PVRDMA_EVENT_QP_LAST_WQE_REACHED, + PVRDMA_EVENT_CLIENT_REREGISTER, + PVRDMA_EVENT_GID_CHANGE, +}; + +/* Event queue element. */ +struct pvrdma_eqe { + u32 type; /* Event type. */ + u32 info; /* Handle, other. */ +}; + +/* CQ notification queue element. */ +struct pvrdma_cqne { + u32 info; /* Handle */ +}; + +enum { + PVRDMA_CMD_FIRST, + PVRDMA_CMD_QUERY_PORT = PVRDMA_CMD_FIRST, + PVRDMA_CMD_QUERY_PKEY, + PVRDMA_CMD_CREATE_PD, + PVRDMA_CMD_DESTROY_PD, + PVRDMA_CMD_CREATE_MR, + PVRDMA_CMD_DESTROY_MR, + PVRDMA_CMD_CREATE_CQ, + PVRDMA_CMD_RESIZE_CQ, + PVRDMA_CMD_DESTROY_CQ, + PVRDMA_CMD_CREATE_QP, + PVRDMA_CMD_MODIFY_QP, + PVRDMA_CMD_QUERY_QP, + PVRDMA_CMD_DESTROY_QP, + PVRDMA_CMD_CREATE_UC, + PVRDMA_CMD_DESTROY_UC, + PVRDMA_CMD_CREATE_BIND, + PVRDMA_CMD_DESTROY_BIND, + PVRDMA_CMD_MAX, +}; + +enum { + PVRDMA_CMD_FIRST_RESP = (1 << 31), + PVRDMA_CMD_QUERY_PORT_RESP = PVRDMA_CMD_FIRST_RESP, + PVRDMA_CMD_QUERY_PKEY_RESP, + PVRDMA_CMD_CREATE_PD_RESP, + PVRDMA_CMD_DESTROY_PD_RESP_NOOP, + PVRDMA_CMD_CREATE_MR_RESP, + PVRDMA_CMD_DESTROY_MR_RESP_NOOP, + PVRDMA_CMD_CREATE_CQ_RESP, + PVRDMA_CMD_RESIZE_CQ_RESP, + PVRDMA_CMD_DESTROY_CQ_RESP_NOOP, + PVRDMA_CMD_CREATE_QP_RESP, + PVRDMA_CMD_MODIFY_QP_RESP, + PVRDMA_CMD_QUERY_QP_RESP, + PVRDMA_CMD_DESTROY_QP_RESP, + PVRDMA_CMD_CREATE_UC_RESP, + PVRDMA_CMD_DESTROY_UC_RESP_NOOP, + PVRDMA_CMD_CREATE_BIND_RESP_NOOP, + PVRDMA_CMD_DESTROY_BIND_RESP_NOOP, + PVRDMA_CMD_MAX_RESP, +}; + +struct pvrdma_cmd_hdr { + u64 response; /* Key for response lookup. */ + u32 cmd; /* PVRDMA_CMD_ */ + u32 reserved; /* Reserved. */ +}; + +struct pvrdma_cmd_resp_hdr { + u64 response; /* From cmd hdr. */ + u32 ack; /* PVRDMA_CMD_XXX_RESP */ + u8 err; /* Error. */ + u8 reserved[3]; /* Reserved. */ +}; + +struct pvrdma_cmd_query_port { + struct pvrdma_cmd_hdr hdr; + u8 port_num; + u8 reserved[7]; +}; + +struct pvrdma_cmd_query_port_resp { + struct pvrdma_cmd_resp_hdr hdr; + struct pvrdma_port_attr attrs; +}; + +struct pvrdma_cmd_query_pkey { + struct pvrdma_cmd_hdr hdr; + u8 port_num; + u8 index; + u8 reserved[6]; +}; + +struct pvrdma_cmd_query_pkey_resp { + struct pvrdma_cmd_resp_hdr hdr; + u16 pkey; + u8 reserved[6]; +}; + +struct pvrdma_cmd_create_uc { + struct pvrdma_cmd_hdr hdr; + u32 pfn; /* UAR page frame number */ + u8 reserved[4]; +}; + +struct pvrdma_cmd_create_uc_resp { + struct pvrdma_cmd_resp_hdr hdr; + u32 ctx_handle; + u8 reserved[4]; +}; + +struct pvrdma_cmd_destroy_uc { + struct pvrdma_cmd_hdr hdr; + u32 ctx_handle; + u8 reserved[4]; +}; + +struct pvrdma_cmd_create_pd { + struct pvrdma_cmd_hdr hdr; + u32 ctx_handle; + u8 reserved[4]; +}; + +struct pvrdma_cmd_create_pd_resp { + struct pvrdma_cmd_resp_hdr hdr; + u32 pd_handle; + u8 reserved[4]; +}; + +struct pvrdma_cmd_destroy_pd { + struct pvrdma_cmd_hdr hdr; + u32 pd_handle; + u8 reserved[4]; +}; + +struct pvrdma_cmd_create_mr { + struct pvrdma_cmd_hdr hdr; + u64 start; + u64 length; + u64 pdir_dma; + u32 pd_handle; + u32 access_flags; + u32 flags; + u32 nchunks; +}; + +struct pvrdma_cmd_create_mr_resp { + struct pvrdma_cmd_resp_hdr hdr; + u32 mr_handle; + u32 lkey; + u32 rkey; + u8 reserved[4]; +}; + +struct pvrdma_cmd_destroy_mr { + struct pvrdma_cmd_hdr hdr; + u32 mr_handle; + u8 reserved[4]; +}; + +struct pvrdma_cmd_create_cq { + struct pvrdma_cmd_hdr hdr; + u64 pdir_dma; + u32 ctx_handle; + u32 cqe; + u32 nchunks; + u8 reserved[4]; +}; + +struct pvrdma_cmd_create_cq_resp { + struct pvrdma_cmd_resp_hdr hdr; + u32 cq_handle; + u32 cqe; +}; + +struct pvrdma_cmd_resize_cq { + struct pvrdma_cmd_hdr hdr; + u32 cq_handle; + u32 cqe; +}; + +struct pvrdma_cmd_resize_cq_resp { + struct pvrdma_cmd_resp_hdr hdr; + u32 cqe; + u8 reserved[4]; +}; + +struct pvrdma_cmd_destroy_cq { + struct pvrdma_cmd_hdr hdr; + u32 cq_handle; + u8 reserved[4]; +}; + +struct pvrdma_cmd_create_qp { + struct pvrdma_cmd_hdr hdr; + u64 pdir_dma; + u32 pd_handle; + u32 send_cq_handle; + u32 recv_cq_handle; + u32 srq_handle; + u32 max_send_wr; + u32 max_recv_wr; + u32 max_send_sge; + u32 max_recv_sge; + u32 max_inline_data; + u32 lkey; + u32 access_flags; + u16 total_chunks; + u16 send_chunks; + u16 max_atomic_arg; + u8 sq_sig_all; + u8 qp_type; + u8 is_srq; + u8 reserved[3]; +}; + +struct pvrdma_cmd_create_qp_resp { + struct pvrdma_cmd_resp_hdr hdr; + u32 qpn; + u32 max_send_wr; + u32 max_recv_wr; + u32 max_send_sge; + u32 max_recv_sge; + u32 max_inline_data; +}; + +struct pvrdma_cmd_modify_qp { + struct pvrdma_cmd_hdr hdr; + u32 qp_handle; + u32 attr_mask; + struct pvrdma_qp_attr attrs; +}; + +struct pvrdma_cmd_query_qp { + struct pvrdma_cmd_hdr hdr; + u32 qp_handle; + u32 attr_mask; +}; + +struct pvrdma_cmd_query_qp_resp { + struct pvrdma_cmd_resp_hdr hdr; + struct pvrdma_qp_attr attrs; +}; + +struct pvrdma_cmd_destroy_qp { + struct pvrdma_cmd_hdr hdr; + u32 qp_handle; + u8 reserved[4]; +}; + +struct pvrdma_cmd_destroy_qp_resp { + struct pvrdma_cmd_resp_hdr hdr; + u32 events_reported; + u8 reserved[4]; +}; + +struct pvrdma_cmd_create_bind { + struct pvrdma_cmd_hdr hdr; + u32 mtu; + u32 vlan; + u32 index; + u8 new_gid[16]; + u8 gid_type; + u8 reserved[3]; +}; + +struct pvrdma_cmd_destroy_bind { + struct pvrdma_cmd_hdr hdr; + u32 index; + u8 dest_gid[16]; + u8 reserved[4]; +}; + +union pvrdma_cmd_req { + struct pvrdma_cmd_hdr hdr; + struct pvrdma_cmd_query_port query_port; + struct pvrdma_cmd_query_pkey query_pkey; + struct pvrdma_cmd_create_uc create_uc; + struct pvrdma_cmd_destroy_uc destroy_uc; + struct pvrdma_cmd_create_pd create_pd; + struct pvrdma_cmd_destroy_pd destroy_pd; + struct pvrdma_cmd_create_mr create_mr; + struct pvrdma_cmd_destroy_mr destroy_mr; + struct pvrdma_cmd_create_cq create_cq; + struct pvrdma_cmd_resize_cq resize_cq; + struct pvrdma_cmd_destroy_cq destroy_cq; + struct pvrdma_cmd_create_qp create_qp; + struct pvrdma_cmd_modify_qp modify_qp; + struct pvrdma_cmd_query_qp query_qp; + struct pvrdma_cmd_destroy_qp destroy_qp; + struct pvrdma_cmd_create_bind create_bind; + struct pvrdma_cmd_destroy_bind destroy_bind; +}; + +union pvrdma_cmd_resp { + struct pvrdma_cmd_resp_hdr hdr; + struct pvrdma_cmd_query_port_resp query_port_resp; + struct pvrdma_cmd_query_pkey_resp query_pkey_resp; + struct pvrdma_cmd_create_uc_resp create_uc_resp; + struct pvrdma_cmd_create_pd_resp create_pd_resp; + struct pvrdma_cmd_create_mr_resp create_mr_resp; + struct pvrdma_cmd_create_cq_resp create_cq_resp; + struct pvrdma_cmd_resize_cq_resp resize_cq_resp; + struct pvrdma_cmd_create_qp_resp create_qp_resp; + struct pvrdma_cmd_query_qp_resp query_qp_resp; + struct pvrdma_cmd_destroy_qp_resp destroy_qp_resp; +}; + +#endif /* __PVRDMA_DEV_API_H__ */ diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_doorbell.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_doorbell.c new file mode 100644 index 000000000000..bf51357ea3aa --- /dev/null +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_doorbell.c @@ -0,0 +1,127 @@ +/* + * Copyright (c) 2012-2016 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of EITHER the GNU General Public License + * version 2 as published by the Free Software Foundation or the BSD + * 2-Clause License. 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 version 2 for more details at + * http://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html. + * + * You should have received a copy of the GNU General Public License + * along with this program available in the file COPYING in the main + * directory of this source tree. + * + * The BSD 2-Clause License + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE + * COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <linux/bitmap.h> +#include <linux/errno.h> +#include <linux/slab.h> + +#include "pvrdma.h" + +int pvrdma_uar_table_init(struct pvrdma_dev *dev) +{ + u32 num = dev->dsr->caps.max_uar; + u32 mask = num - 1; + struct pvrdma_id_table *tbl = &dev->uar_table.tbl; + + if (!is_power_of_2(num)) + return -EINVAL; + + tbl->last = 0; + tbl->top = 0; + tbl->max = num; + tbl->mask = mask; + spin_lock_init(&tbl->lock); + tbl->table = kcalloc(BITS_TO_LONGS(num), sizeof(long), GFP_KERNEL); + if (!tbl->table) + return -ENOMEM; + + /* 0th UAR is taken by the device. */ + set_bit(0, tbl->table); + + return 0; +} + +void pvrdma_uar_table_cleanup(struct pvrdma_dev *dev) +{ + struct pvrdma_id_table *tbl = &dev->uar_table.tbl; + + kfree(tbl->table); +} + +int pvrdma_uar_alloc(struct pvrdma_dev *dev, struct pvrdma_uar_map *uar) +{ + struct pvrdma_id_table *tbl; + unsigned long flags; + u32 obj; + + tbl = &dev->uar_table.tbl; + + spin_lock_irqsave(&tbl->lock, flags); + obj = find_next_zero_bit(tbl->table, tbl->max, tbl->last); + if (obj >= tbl->max) { + tbl->top = (tbl->top + tbl->max) & tbl->mask; + obj = find_first_zero_bit(tbl->table, tbl->max); + } + + if (obj >= tbl->max) { + spin_unlock_irqrestore(&tbl->lock, flags); + return -ENOMEM; + } + + set_bit(obj, tbl->table); + obj |= tbl->top; + + spin_unlock_irqrestore(&tbl->lock, flags); + + uar->index = obj; + uar->pfn = (pci_resource_start(dev->pdev, PVRDMA_PCI_RESOURCE_UAR) >> + PAGE_SHIFT) + uar->index; + + return 0; +} + +void pvrdma_uar_free(struct pvrdma_dev *dev, struct pvrdma_uar_map *uar) +{ + struct pvrdma_id_table *tbl = &dev->uar_table.tbl; + unsigned long flags; + u32 obj; + + obj = uar->index & (tbl->max - 1); + spin_lock_irqsave(&tbl->lock, flags); + clear_bit(obj, tbl->table); + tbl->last = min(tbl->last, obj); + tbl->top = (tbl->top + tbl->max) & tbl->mask; + spin_unlock_irqrestore(&tbl->lock, flags); +} diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_main.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_main.c new file mode 100644 index 000000000000..100bea5c42ff --- /dev/null +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_main.c @@ -0,0 +1,1114 @@ +/* + * Copyright (c) 2012-2016 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of EITHER the GNU General Public License + * version 2 as published by the Free Software Foundation or the BSD + * 2-Clause License. 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 version 2 for more details at + * http://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html. + * + * You should have received a copy of the GNU General Public License + * along with this program available in the file COPYING in the main + * directory of this source tree. + * + * The BSD 2-Clause License + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE + * COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <linux/errno.h> +#include <linux/inetdevice.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/slab.h> +#include <rdma/ib_addr.h> +#include <rdma/ib_smi.h> +#include <rdma/ib_user_verbs.h> +#include <net/addrconf.h> + +#include "pvrdma.h" + +#define DRV_NAME "vmw_pvrdma" +#define DRV_VERSION "1.0.0.0-k" + +static DEFINE_MUTEX(pvrdma_device_list_lock); +static LIST_HEAD(pvrdma_device_list); +static struct workqueue_struct *event_wq; + +static int pvrdma_add_gid(struct ib_device *ibdev, + u8 port_num, + unsigned int index, + const union ib_gid *gid, + const struct ib_gid_attr *attr, + void **context); +static int pvrdma_del_gid(struct ib_device *ibdev, + u8 port_num, + unsigned int index, + void **context); + + +static ssize_t show_hca(struct device *device, struct device_attribute *attr, + char *buf) +{ + return sprintf(buf, "VMW_PVRDMA-%s\n", DRV_VERSION); +} + +static ssize_t show_rev(struct device *device, struct device_attribute *attr, + char *buf) +{ + return sprintf(buf, "%d\n", PVRDMA_REV_ID); +} + +static ssize_t show_board(struct device *device, struct device_attribute *attr, + char *buf) +{ + return sprintf(buf, "%d\n", PVRDMA_BOARD_ID); +} + +static DEVICE_ATTR(hw_rev, S_IRUGO, show_rev, NULL); +static DEVICE_ATTR(hca_type, S_IRUGO, show_hca, NULL); +static DEVICE_ATTR(board_id, S_IRUGO, show_board, NULL); + +static struct device_attribute *pvrdma_class_attributes[] = { + &dev_attr_hw_rev, + &dev_attr_hca_type, + &dev_attr_board_id +}; + +static void pvrdma_get_fw_ver_str(struct ib_device *device, char *str, + size_t str_len) +{ + struct pvrdma_dev *dev = + container_of(device, struct pvrdma_dev, ib_dev); + snprintf(str, str_len, "%d.%d.%d\n", + (int) (dev->dsr->caps.fw_ver >> 32), + (int) (dev->dsr->caps.fw_ver >> 16) & 0xffff, + (int) dev->dsr->caps.fw_ver & 0xffff); +} + +static int pvrdma_init_device(struct pvrdma_dev *dev) +{ + /* Initialize some device related stuff */ + spin_lock_init(&dev->cmd_lock); + sema_init(&dev->cmd_sema, 1); + atomic_set(&dev->num_qps, 0); + atomic_set(&dev->num_cqs, 0); + atomic_set(&dev->num_pds, 0); + atomic_set(&dev->num_ahs, 0); + + return 0; +} + +static int pvrdma_port_immutable(struct ib_device *ibdev, u8 port_num, + struct ib_port_immutable *immutable) +{ + struct ib_port_attr attr; + int err; + + immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE; + + err = ib_query_port(ibdev, port_num, &attr); + if (err) + return err; + + immutable->pkey_tbl_len = attr.pkey_tbl_len; + immutable->gid_tbl_len = attr.gid_tbl_len; + immutable->max_mad_size = IB_MGMT_MAD_SIZE; + return 0; +} + +static struct net_device *pvrdma_get_netdev(struct ib_device *ibdev, + u8 port_num) +{ + struct net_device *netdev; + struct pvrdma_dev *dev = to_vdev(ibdev); + + if (port_num != 1) + return NULL; + + rcu_read_lock(); + netdev = dev->netdev; + if (netdev) + dev_hold(netdev); + rcu_read_unlock(); + + return netdev; +} + +static int pvrdma_register_device(struct pvrdma_dev *dev) +{ + int ret = -1; + int i = 0; + + strlcpy(dev->ib_dev.name, "vmw_pvrdma%d", IB_DEVICE_NAME_MAX); + dev->ib_dev.node_guid = dev->dsr->caps.node_guid; + dev->sys_image_guid = dev->dsr->caps.sys_image_guid; + dev->flags = 0; + dev->ib_dev.owner = THIS_MODULE; + dev->ib_dev.num_comp_vectors = 1; + dev->ib_dev.dev.parent = &dev->pdev->dev; + dev->ib_dev.uverbs_abi_ver = PVRDMA_UVERBS_ABI_VERSION; + dev->ib_dev.uverbs_cmd_mask = + (1ull << IB_USER_VERBS_CMD_GET_CONTEXT) | + (1ull << IB_USER_VERBS_CMD_QUERY_DEVICE) | + (1ull << IB_USER_VERBS_CMD_QUERY_PORT) | + (1ull << IB_USER_VERBS_CMD_ALLOC_PD) | + (1ull << IB_USER_VERBS_CMD_DEALLOC_PD) | + (1ull << IB_USER_VERBS_CMD_REG_MR) | + (1ull << IB_USER_VERBS_CMD_DEREG_MR) | + (1ull << IB_USER_VERBS_CMD_CREATE_COMP_CHANNEL) | + (1ull << IB_USER_VERBS_CMD_CREATE_CQ) | + (1ull << IB_USER_VERBS_CMD_POLL_CQ) | + (1ull << IB_USER_VERBS_CMD_REQ_NOTIFY_CQ) | + (1ull << IB_USER_VERBS_CMD_DESTROY_CQ) | + (1ull << IB_USER_VERBS_CMD_CREATE_QP) | + (1ull << IB_USER_VERBS_CMD_MODIFY_QP) | + (1ull << IB_USER_VERBS_CMD_QUERY_QP) | + (1ull << IB_USER_VERBS_CMD_DESTROY_QP) | + (1ull << IB_USER_VERBS_CMD_POST_SEND) | + (1ull << IB_USER_VERBS_CMD_POST_RECV) | + (1ull << IB_USER_VERBS_CMD_CREATE_AH) | + (1ull << IB_USER_VERBS_CMD_DESTROY_AH); + + dev->ib_dev.node_type = RDMA_NODE_IB_CA; + dev->ib_dev.phys_port_cnt = dev->dsr->caps.phys_port_cnt; + + dev->ib_dev.query_device = pvrdma_query_device; + dev->ib_dev.query_port = pvrdma_query_port; + dev->ib_dev.query_gid = pvrdma_query_gid; + dev->ib_dev.query_pkey = pvrdma_query_pkey; + dev->ib_dev.modify_port = pvrdma_modify_port; + dev->ib_dev.alloc_ucontext = pvrdma_alloc_ucontext; + dev->ib_dev.dealloc_ucontext = pvrdma_dealloc_ucontext; + dev->ib_dev.mmap = pvrdma_mmap; + dev->ib_dev.alloc_pd = pvrdma_alloc_pd; + dev->ib_dev.dealloc_pd = pvrdma_dealloc_pd; + dev->ib_dev.create_ah = pvrdma_create_ah; + dev->ib_dev.destroy_ah = pvrdma_destroy_ah; + dev->ib_dev.create_qp = pvrdma_create_qp; + dev->ib_dev.modify_qp = pvrdma_modify_qp; + dev->ib_dev.query_qp = pvrdma_query_qp; + dev->ib_dev.destroy_qp = pvrdma_destroy_qp; + dev->ib_dev.post_send = pvrdma_post_send; + dev->ib_dev.post_recv = pvrdma_post_recv; + dev->ib_dev.create_cq = pvrdma_create_cq; + dev->ib_dev.modify_cq = pvrdma_modify_cq; + dev->ib_dev.resize_cq = pvrdma_resize_cq; + dev->ib_dev.destroy_cq = pvrdma_destroy_cq; + dev->ib_dev.poll_cq = pvrdma_poll_cq; + dev->ib_dev.req_notify_cq = pvrdma_req_notify_cq; + dev->ib_dev.get_dma_mr = pvrdma_get_dma_mr; + dev->ib_dev.reg_user_mr = pvrdma_reg_user_mr; + dev->ib_dev.dereg_mr = pvrdma_dereg_mr; + dev->ib_dev.alloc_mr = pvrdma_alloc_mr; + dev->ib_dev.map_mr_sg = pvrdma_map_mr_sg; + dev->ib_dev.add_gid = pvrdma_add_gid; + dev->ib_dev.del_gid = pvrdma_del_gid; + dev->ib_dev.get_netdev = pvrdma_get_netdev; + dev->ib_dev.get_port_immutable = pvrdma_port_immutable; + dev->ib_dev.get_link_layer = pvrdma_port_link_layer; + dev->ib_dev.get_dev_fw_str = pvrdma_get_fw_ver_str; + + mutex_init(&dev->port_mutex); + spin_lock_init(&dev->desc_lock); + + dev->cq_tbl = kcalloc(dev->dsr->caps.max_cq, sizeof(void *), + GFP_KERNEL); + if (!dev->cq_tbl) + return ret; + spin_lock_init(&dev->cq_tbl_lock); + + dev->qp_tbl = kcalloc(dev->dsr->caps.max_qp, sizeof(void *), + GFP_KERNEL); + if (!dev->qp_tbl) + goto err_cq_free; + spin_lock_init(&dev->qp_tbl_lock); + + ret = ib_register_device(&dev->ib_dev, NULL); + if (ret) + goto err_qp_free; + + for (i = 0; i < ARRAY_SIZE(pvrdma_class_attributes); ++i) { + ret = device_create_file(&dev->ib_dev.dev, + pvrdma_class_attributes[i]); + if (ret) + goto err_class; + } + + dev->ib_active = true; + + return 0; + +err_class: + ib_unregister_device(&dev->ib_dev); +err_qp_free: + kfree(dev->qp_tbl); +err_cq_free: + kfree(dev->cq_tbl); + + return ret; +} + +static irqreturn_t pvrdma_intr0_handler(int irq, void *dev_id) +{ + u32 icr = PVRDMA_INTR_CAUSE_RESPONSE; + struct pvrdma_dev *dev = dev_id; + + dev_dbg(&dev->pdev->dev, "interrupt 0 (response) handler\n"); + + if (!dev->pdev->msix_enabled) { + /* Legacy intr */ + icr = pvrdma_read_reg(dev, PVRDMA_REG_ICR); + if (icr == 0) + return IRQ_NONE; + } + + if (icr == PVRDMA_INTR_CAUSE_RESPONSE) + complete(&dev->cmd_done); + + return IRQ_HANDLED; +} + +static void pvrdma_qp_event(struct pvrdma_dev *dev, u32 qpn, int type) +{ + struct pvrdma_qp *qp; + unsigned long flags; + + spin_lock_irqsave(&dev->qp_tbl_lock, flags); + qp = dev->qp_tbl[qpn % dev->dsr->caps.max_qp]; + if (qp) + atomic_inc(&qp->refcnt); + spin_unlock_irqrestore(&dev->qp_tbl_lock, flags); + + if (qp && qp->ibqp.event_handler) { + struct ib_qp *ibqp = &qp->ibqp; + struct ib_event e; + + e.device = ibqp->device; + e.element.qp = ibqp; + e.event = type; /* 1:1 mapping for now. */ + ibqp->event_handler(&e, ibqp->qp_context); + } + if (qp) { + atomic_dec(&qp->refcnt); + if (atomic_read(&qp->refcnt) == 0) + wake_up(&qp->wait); + } +} + +static void pvrdma_cq_event(struct pvrdma_dev *dev, u32 cqn, int type) +{ + struct pvrdma_cq *cq; + unsigned long flags; + + spin_lock_irqsave(&dev->cq_tbl_lock, flags); + cq = dev->cq_tbl[cqn % dev->dsr->caps.max_cq]; + if (cq) + atomic_inc(&cq->refcnt); + spin_unlock_irqrestore(&dev->cq_tbl_lock, flags); + + if (cq && cq->ibcq.event_handler) { + struct ib_cq *ibcq = &cq->ibcq; + struct ib_event e; + + e.device = ibcq->device; + e.element.cq = ibcq; + e.event = type; /* 1:1 mapping for now. */ + ibcq->event_handler(&e, ibcq->cq_context); + } + if (cq) { + atomic_dec(&cq->refcnt); + if (atomic_read(&cq->refcnt) == 0) + wake_up(&cq->wait); + } +} + +static void pvrdma_dispatch_event(struct pvrdma_dev *dev, int port, + enum ib_event_type event) +{ + struct ib_event ib_event; + + memset(&ib_event, 0, sizeof(ib_event)); + ib_event.device = &dev->ib_dev; + ib_event.element.port_num = port; + ib_event.event = event; + ib_dispatch_event(&ib_event); +} + +static void pvrdma_dev_event(struct pvrdma_dev *dev, u8 port, int type) +{ + if (port < 1 || port > dev->dsr->caps.phys_port_cnt) { + dev_warn(&dev->pdev->dev, "event on port %d\n", port); + return; + } + + pvrdma_dispatch_event(dev, port, type); +} + +static inline struct pvrdma_eqe *get_eqe(struct pvrdma_dev *dev, unsigned int i) +{ + return (struct pvrdma_eqe *)pvrdma_page_dir_get_ptr( + &dev->async_pdir, + PAGE_SIZE + + sizeof(struct pvrdma_eqe) * i); +} + +static irqreturn_t pvrdma_intr1_handler(int irq, void *dev_id) +{ + struct pvrdma_dev *dev = dev_id; + struct pvrdma_ring *ring = &dev->async_ring_state->rx; + int ring_slots = (dev->dsr->async_ring_pages.num_pages - 1) * + PAGE_SIZE / sizeof(struct pvrdma_eqe); + unsigned int head; + + dev_dbg(&dev->pdev->dev, "interrupt 1 (async event) handler\n"); + + /* + * Don't process events until the IB device is registered. Otherwise + * we'll try to ib_dispatch_event() on an invalid device. + */ + if (!dev->ib_active) + return IRQ_HANDLED; + + while (pvrdma_idx_ring_has_data(ring, ring_slots, &head) > 0) { + struct pvrdma_eqe *eqe; + + eqe = get_eqe(dev, head); + + switch (eqe->type) { + case PVRDMA_EVENT_QP_FATAL: + case PVRDMA_EVENT_QP_REQ_ERR: + case PVRDMA_EVENT_QP_ACCESS_ERR: + case PVRDMA_EVENT_COMM_EST: + case PVRDMA_EVENT_SQ_DRAINED: + case PVRDMA_EVENT_PATH_MIG: + case PVRDMA_EVENT_PATH_MIG_ERR: + case PVRDMA_EVENT_QP_LAST_WQE_REACHED: + pvrdma_qp_event(dev, eqe->info, eqe->type); + break; + + case PVRDMA_EVENT_CQ_ERR: + pvrdma_cq_event(dev, eqe->info, eqe->type); + break; + + case PVRDMA_EVENT_SRQ_ERR: + case PVRDMA_EVENT_SRQ_LIMIT_REACHED: + break; + + case PVRDMA_EVENT_PORT_ACTIVE: + case PVRDMA_EVENT_PORT_ERR: + case PVRDMA_EVENT_LID_CHANGE: + case PVRDMA_EVENT_PKEY_CHANGE: + case PVRDMA_EVENT_SM_CHANGE: + case PVRDMA_EVENT_CLIENT_REREGISTER: + case PVRDMA_EVENT_GID_CHANGE: + pvrdma_dev_event(dev, eqe->info, eqe->type); + break; + + case PVRDMA_EVENT_DEVICE_FATAL: + pvrdma_dev_event(dev, 1, eqe->type); + break; + + default: + break; + } + + pvrdma_idx_ring_inc(&ring->cons_head, ring_slots); + } + + return IRQ_HANDLED; +} + +static inline struct pvrdma_cqne *get_cqne(struct pvrdma_dev *dev, + unsigned int i) +{ + return (struct pvrdma_cqne *)pvrdma_page_dir_get_ptr( + &dev->cq_pdir, + PAGE_SIZE + + sizeof(struct pvrdma_cqne) * i); +} + +static irqreturn_t pvrdma_intrx_handler(int irq, void *dev_id) +{ + struct pvrdma_dev *dev = dev_id; + struct pvrdma_ring *ring = &dev->cq_ring_state->rx; + int ring_slots = (dev->dsr->cq_ring_pages.num_pages - 1) * PAGE_SIZE / + sizeof(struct pvrdma_cqne); + unsigned int head; + unsigned long flags; + + dev_dbg(&dev->pdev->dev, "interrupt x (completion) handler\n"); + + while (pvrdma_idx_ring_has_data(ring, ring_slots, &head) > 0) { + struct pvrdma_cqne *cqne; + struct pvrdma_cq *cq; + + cqne = get_cqne(dev, head); + spin_lock_irqsave(&dev->cq_tbl_lock, flags); + cq = dev->cq_tbl[cqne->info % dev->dsr->caps.max_cq]; + if (cq) + atomic_inc(&cq->refcnt); + spin_unlock_irqrestore(&dev->cq_tbl_lock, flags); + + if (cq && cq->ibcq.comp_handler) + cq->ibcq.comp_handler(&cq->ibcq, cq->ibcq.cq_context); + if (cq) { + atomic_dec(&cq->refcnt); + if (atomic_read(&cq->refcnt)) + wake_up(&cq->wait); + } + pvrdma_idx_ring_inc(&ring->cons_head, ring_slots); + } + + return IRQ_HANDLED; +} + +static void pvrdma_free_irq(struct pvrdma_dev *dev) +{ + int i; + + dev_dbg(&dev->pdev->dev, "freeing interrupts\n"); + for (i = 0; i < dev->nr_vectors; i++) + free_irq(pci_irq_vector(dev->pdev, i), dev); +} + +static void pvrdma_enable_intrs(struct pvrdma_dev *dev) +{ + dev_dbg(&dev->pdev->dev, "enable interrupts\n"); + pvrdma_write_reg(dev, PVRDMA_REG_IMR, 0); +} + +static void pvrdma_disable_intrs(struct pvrdma_dev *dev) +{ + dev_dbg(&dev->pdev->dev, "disable interrupts\n"); + pvrdma_write_reg(dev, PVRDMA_REG_IMR, ~0); +} + +static int pvrdma_alloc_intrs(struct pvrdma_dev *dev) +{ + struct pci_dev *pdev = dev->pdev; + int ret = 0, i; + + ret = pci_alloc_irq_vectors(pdev, 1, PVRDMA_MAX_INTERRUPTS, + PCI_IRQ_MSIX); + if (ret < 0) { + ret = pci_alloc_irq_vectors(pdev, 1, 1, + PCI_IRQ_MSI | PCI_IRQ_LEGACY); + if (ret < 0) + return ret; + } + dev->nr_vectors = ret; + + ret = request_irq(pci_irq_vector(dev->pdev, 0), pvrdma_intr0_handler, + pdev->msix_enabled ? 0 : IRQF_SHARED, DRV_NAME, dev); + if (ret) { + dev_err(&dev->pdev->dev, + "failed to request interrupt 0\n"); + goto out_free_vectors; + } + + for (i = 1; i < dev->nr_vectors; i++) { + ret = request_irq(pci_irq_vector(dev->pdev, i), + i == 1 ? pvrdma_intr1_handler : + pvrdma_intrx_handler, + 0, DRV_NAME, dev); + if (ret) { + dev_err(&dev->pdev->dev, + "failed to request interrupt %d\n", i); + goto free_irqs; + } + } + + return 0; + +free_irqs: + while (--i >= 0) + free_irq(pci_irq_vector(dev->pdev, i), dev); +out_free_vectors: + pci_free_irq_vectors(pdev); + return ret; +} + +static void pvrdma_free_slots(struct pvrdma_dev *dev) +{ + struct pci_dev *pdev = dev->pdev; + + if (dev->resp_slot) + dma_free_coherent(&pdev->dev, PAGE_SIZE, dev->resp_slot, + dev->dsr->resp_slot_dma); + if (dev->cmd_slot) + dma_free_coherent(&pdev->dev, PAGE_SIZE, dev->cmd_slot, + dev->dsr->cmd_slot_dma); +} + +static int pvrdma_add_gid_at_index(struct pvrdma_dev *dev, + const union ib_gid *gid, + int index) +{ + int ret; + union pvrdma_cmd_req req; + struct pvrdma_cmd_create_bind *cmd_bind = &req.create_bind; + + if (!dev->sgid_tbl) { + dev_warn(&dev->pdev->dev, "sgid table not initialized\n"); + return -EINVAL; + } + + memset(cmd_bind, 0, sizeof(*cmd_bind)); + cmd_bind->hdr.cmd = PVRDMA_CMD_CREATE_BIND; + memcpy(cmd_bind->new_gid, gid->raw, 16); + cmd_bind->mtu = ib_mtu_enum_to_int(IB_MTU_1024); + cmd_bind->vlan = 0xfff; + cmd_bind->index = index; + cmd_bind->gid_type = PVRDMA_GID_TYPE_FLAG_ROCE_V1; + + ret = pvrdma_cmd_post(dev, &req, NULL, 0); + if (ret < 0) { + dev_warn(&dev->pdev->dev, + "could not create binding, error: %d\n", ret); + return -EFAULT; + } + memcpy(&dev->sgid_tbl[index], gid, sizeof(*gid)); + return 0; +} + +static int pvrdma_add_gid(struct ib_device *ibdev, + u8 port_num, + unsigned int index, + const union ib_gid *gid, + const struct ib_gid_attr *attr, + void **context) +{ + struct pvrdma_dev *dev = to_vdev(ibdev); + + return pvrdma_add_gid_at_index(dev, gid, index); +} + +static int pvrdma_del_gid_at_index(struct pvrdma_dev *dev, int index) +{ + int ret; + union pvrdma_cmd_req req; + struct pvrdma_cmd_destroy_bind *cmd_dest = &req.destroy_bind; + + /* Update sgid table. */ + if (!dev->sgid_tbl) { + dev_warn(&dev->pdev->dev, "sgid table not initialized\n"); + return -EINVAL; + } + + memset(cmd_dest, 0, sizeof(*cmd_dest)); + cmd_dest->hdr.cmd = PVRDMA_CMD_DESTROY_BIND; + memcpy(cmd_dest->dest_gid, &dev->sgid_tbl[index], 16); + cmd_dest->index = index; + + ret = pvrdma_cmd_post(dev, &req, NULL, 0); + if (ret < 0) { + dev_warn(&dev->pdev->dev, + "could not destroy binding, error: %d\n", ret); + return ret; + } + memset(&dev->sgid_tbl[index], 0, 16); + return 0; +} + +static int pvrdma_del_gid(struct ib_device *ibdev, + u8 port_num, + unsigned int index, + void **context) +{ + struct pvrdma_dev *dev = to_vdev(ibdev); + + dev_dbg(&dev->pdev->dev, "removing gid at index %u from %s", + index, dev->netdev->name); + + return pvrdma_del_gid_at_index(dev, index); +} + +static void pvrdma_netdevice_event_handle(struct pvrdma_dev *dev, + unsigned long event) +{ + switch (event) { + case NETDEV_REBOOT: + case NETDEV_DOWN: + pvrdma_dispatch_event(dev, 1, IB_EVENT_PORT_ERR); + break; + case NETDEV_UP: + pvrdma_dispatch_event(dev, 1, IB_EVENT_PORT_ACTIVE); + break; + default: + dev_dbg(&dev->pdev->dev, "ignore netdevice event %ld on %s\n", + event, dev->ib_dev.name); + break; + } +} + +static void pvrdma_netdevice_event_work(struct work_struct *work) +{ + struct pvrdma_netdevice_work *netdev_work; + struct pvrdma_dev *dev; + + netdev_work = container_of(work, struct pvrdma_netdevice_work, work); + + mutex_lock(&pvrdma_device_list_lock); + list_for_each_entry(dev, &pvrdma_device_list, device_link) { + if (dev->netdev == netdev_work->event_netdev) { + pvrdma_netdevice_event_handle(dev, netdev_work->event); + break; + } + } + mutex_unlock(&pvrdma_device_list_lock); + + kfree(netdev_work); +} + +static int pvrdma_netdevice_event(struct notifier_block *this, + unsigned long event, void *ptr) +{ + struct net_device *event_netdev = netdev_notifier_info_to_dev(ptr); + struct pvrdma_netdevice_work *netdev_work; + + netdev_work = kmalloc(sizeof(*netdev_work), GFP_ATOMIC); + if (!netdev_work) + return NOTIFY_BAD; + + INIT_WORK(&netdev_work->work, pvrdma_netdevice_event_work); + netdev_work->event_netdev = event_netdev; + netdev_work->event = event; + queue_work(event_wq, &netdev_work->work); + + return NOTIFY_DONE; +} + +static int pvrdma_pci_probe(struct pci_dev *pdev, + const struct pci_device_id *id) +{ + struct pci_dev *pdev_net; + struct pvrdma_dev *dev; + int ret; + unsigned long start; + unsigned long len; + unsigned int version; + dma_addr_t slot_dma = 0; + + dev_dbg(&pdev->dev, "initializing driver %s\n", pci_name(pdev)); + + /* Allocate zero-out device */ + dev = (struct pvrdma_dev *)ib_alloc_device(sizeof(*dev)); + if (!dev) { + dev_err(&pdev->dev, "failed to allocate IB device\n"); + return -ENOMEM; + } + + mutex_lock(&pvrdma_device_list_lock); + list_add(&dev->device_link, &pvrdma_device_list); + mutex_unlock(&pvrdma_device_list_lock); + + ret = pvrdma_init_device(dev); + if (ret) + goto err_free_device; + + dev->pdev = pdev; + pci_set_drvdata(pdev, dev); + + ret = pci_enable_device(pdev); + if (ret) { + dev_err(&pdev->dev, "cannot enable PCI device\n"); + goto err_free_device; + } + + dev_dbg(&pdev->dev, "PCI resource flags BAR0 %#lx\n", + pci_resource_flags(pdev, 0)); + dev_dbg(&pdev->dev, "PCI resource len %#llx\n", + (unsigned long long)pci_resource_len(pdev, 0)); + dev_dbg(&pdev->dev, "PCI resource start %#llx\n", + (unsigned long long)pci_resource_start(pdev, 0)); + dev_dbg(&pdev->dev, "PCI resource flags BAR1 %#lx\n", + pci_resource_flags(pdev, 1)); + dev_dbg(&pdev->dev, "PCI resource len %#llx\n", + (unsigned long long)pci_resource_len(pdev, 1)); + dev_dbg(&pdev->dev, "PCI resource start %#llx\n", + (unsigned long long)pci_resource_start(pdev, 1)); + + if (!(pci_resource_flags(pdev, 0) & IORESOURCE_MEM) || + !(pci_resource_flags(pdev, 1) & IORESOURCE_MEM)) { + dev_err(&pdev->dev, "PCI BAR region not MMIO\n"); + ret = -ENOMEM; + goto err_free_device; + } + + ret = pci_request_regions(pdev, DRV_NAME); + if (ret) { + dev_err(&pdev->dev, "cannot request PCI resources\n"); + goto err_disable_pdev; + } + + /* Enable 64-Bit DMA */ + if (pci_set_dma_mask(pdev, DMA_BIT_MASK(64)) == 0) { + ret = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64)); + if (ret != 0) { + dev_err(&pdev->dev, + "pci_set_consistent_dma_mask failed\n"); + goto err_free_resource; + } + } else { + ret = pci_set_dma_mask(pdev, DMA_BIT_MASK(32)); + if (ret != 0) { + dev_err(&pdev->dev, + "pci_set_dma_mask failed\n"); + goto err_free_resource; + } + } + + pci_set_master(pdev); + + /* Map register space */ + start = pci_resource_start(dev->pdev, PVRDMA_PCI_RESOURCE_REG); + len = pci_resource_len(dev->pdev, PVRDMA_PCI_RESOURCE_REG); + dev->regs = ioremap(start, len); + if (!dev->regs) { + dev_err(&pdev->dev, "register mapping failed\n"); + ret = -ENOMEM; + goto err_free_resource; + } + + /* Setup per-device UAR. */ + dev->driver_uar.index = 0; + dev->driver_uar.pfn = + pci_resource_start(dev->pdev, PVRDMA_PCI_RESOURCE_UAR) >> + PAGE_SHIFT; + dev->driver_uar.map = + ioremap(dev->driver_uar.pfn << PAGE_SHIFT, PAGE_SIZE); + if (!dev->driver_uar.map) { + dev_err(&pdev->dev, "failed to remap UAR pages\n"); + ret = -ENOMEM; + goto err_unmap_regs; + } + + version = pvrdma_read_reg(dev, PVRDMA_REG_VERSION); + dev_info(&pdev->dev, "device version %d, driver version %d\n", + version, PVRDMA_VERSION); + if (version < PVRDMA_VERSION) { + dev_err(&pdev->dev, "incompatible device version\n"); + goto err_uar_unmap; + } + + dev->dsr = dma_alloc_coherent(&pdev->dev, sizeof(*dev->dsr), + &dev->dsrbase, GFP_KERNEL); + if (!dev->dsr) { + dev_err(&pdev->dev, "failed to allocate shared region\n"); + ret = -ENOMEM; + goto err_uar_unmap; + } + + /* Setup the shared region */ + memset(dev->dsr, 0, sizeof(*dev->dsr)); + dev->dsr->driver_version = PVRDMA_VERSION; + dev->dsr->gos_info.gos_bits = sizeof(void *) == 4 ? + PVRDMA_GOS_BITS_32 : + PVRDMA_GOS_BITS_64; + dev->dsr->gos_info.gos_type = PVRDMA_GOS_TYPE_LINUX; + dev->dsr->gos_info.gos_ver = 1; + dev->dsr->uar_pfn = dev->driver_uar.pfn; + + /* Command slot. */ + dev->cmd_slot = dma_alloc_coherent(&pdev->dev, PAGE_SIZE, + &slot_dma, GFP_KERNEL); + if (!dev->cmd_slot) { + ret = -ENOMEM; + goto err_free_dsr; + } + + dev->dsr->cmd_slot_dma = (u64)slot_dma; + + /* Response slot. */ + dev->resp_slot = dma_alloc_coherent(&pdev->dev, PAGE_SIZE, + &slot_dma, GFP_KERNEL); + if (!dev->resp_slot) { + ret = -ENOMEM; + goto err_free_slots; + } + + dev->dsr->resp_slot_dma = (u64)slot_dma; + + /* Async event ring */ + dev->dsr->async_ring_pages.num_pages = 4; + ret = pvrdma_page_dir_init(dev, &dev->async_pdir, + dev->dsr->async_ring_pages.num_pages, true); + if (ret) + goto err_free_slots; + dev->async_ring_state = dev->async_pdir.pages[0]; + dev->dsr->async_ring_pages.pdir_dma = dev->async_pdir.dir_dma; + + /* CQ notification ring */ + dev->dsr->cq_ring_pages.num_pages = 4; + ret = pvrdma_page_dir_init(dev, &dev->cq_pdir, + dev->dsr->cq_ring_pages.num_pages, true); + if (ret) + goto err_free_async_ring; + dev->cq_ring_state = dev->cq_pdir.pages[0]; + dev->dsr->cq_ring_pages.pdir_dma = dev->cq_pdir.dir_dma; + + /* + * Write the PA of the shared region to the device. The writes must be + * ordered such that the high bits are written last. When the writes + * complete, the device will have filled out the capabilities. + */ + + pvrdma_write_reg(dev, PVRDMA_REG_DSRLOW, (u32)dev->dsrbase); + pvrdma_write_reg(dev, PVRDMA_REG_DSRHIGH, + (u32)((u64)(dev->dsrbase) >> 32)); + + /* Make sure the write is complete before reading status. */ + mb(); + + /* Currently, the driver only supports RoCE mode. */ + if (dev->dsr->caps.mode != PVRDMA_DEVICE_MODE_ROCE) { + dev_err(&pdev->dev, "unsupported transport %d\n", + dev->dsr->caps.mode); + ret = -EFAULT; + goto err_free_cq_ring; + } + + /* Currently, the driver only supports RoCE V1. */ + if (!(dev->dsr->caps.gid_types & PVRDMA_GID_TYPE_FLAG_ROCE_V1)) { + dev_err(&pdev->dev, "driver needs RoCE v1 support\n"); + ret = -EFAULT; + goto err_free_cq_ring; + } + + /* Paired vmxnet3 will have same bus, slot. But func will be 0 */ + pdev_net = pci_get_slot(pdev->bus, PCI_DEVFN(PCI_SLOT(pdev->devfn), 0)); + if (!pdev_net) { + dev_err(&pdev->dev, "failed to find paired net device\n"); + ret = -ENODEV; + goto err_free_cq_ring; + } + + if (pdev_net->vendor != PCI_VENDOR_ID_VMWARE || + pdev_net->device != PCI_DEVICE_ID_VMWARE_VMXNET3) { + dev_err(&pdev->dev, "failed to find paired vmxnet3 device\n"); + pci_dev_put(pdev_net); + ret = -ENODEV; + goto err_free_cq_ring; + } + + dev->netdev = pci_get_drvdata(pdev_net); + pci_dev_put(pdev_net); + if (!dev->netdev) { + dev_err(&pdev->dev, "failed to get vmxnet3 device\n"); + ret = -ENODEV; + goto err_free_cq_ring; + } + + dev_info(&pdev->dev, "paired device to %s\n", dev->netdev->name); + + /* Interrupt setup */ + ret = pvrdma_alloc_intrs(dev); + if (ret) { + dev_err(&pdev->dev, "failed to allocate interrupts\n"); + ret = -ENOMEM; + goto err_free_cq_ring; + } + + /* Allocate UAR table. */ + ret = pvrdma_uar_table_init(dev); + if (ret) { + dev_err(&pdev->dev, "failed to allocate UAR table\n"); + ret = -ENOMEM; + goto err_free_intrs; + } + + /* Allocate GID table */ + dev->sgid_tbl = kcalloc(dev->dsr->caps.gid_tbl_len, + sizeof(union ib_gid), GFP_KERNEL); + if (!dev->sgid_tbl) { + ret = -ENOMEM; + goto err_free_uar_table; + } + dev_dbg(&pdev->dev, "gid table len %d\n", dev->dsr->caps.gid_tbl_len); + + pvrdma_enable_intrs(dev); + + /* Activate pvrdma device */ + pvrdma_write_reg(dev, PVRDMA_REG_CTL, PVRDMA_DEVICE_CTL_ACTIVATE); + + /* Make sure the write is complete before reading status. */ + mb(); + + /* Check if device was successfully activated */ + ret = pvrdma_read_reg(dev, PVRDMA_REG_ERR); + if (ret != 0) { + dev_err(&pdev->dev, "failed to activate device\n"); + ret = -EFAULT; + goto err_disable_intr; + } + + /* Register IB device */ + ret = pvrdma_register_device(dev); + if (ret) { + dev_err(&pdev->dev, "failed to register IB device\n"); + goto err_disable_intr; + } + + dev->nb_netdev.notifier_call = pvrdma_netdevice_event; + ret = register_netdevice_notifier(&dev->nb_netdev); + if (ret) { + dev_err(&pdev->dev, "failed to register netdevice events\n"); + goto err_unreg_ibdev; + } + + dev_info(&pdev->dev, "attached to device\n"); + return 0; + +err_unreg_ibdev: + ib_unregister_device(&dev->ib_dev); +err_disable_intr: + pvrdma_disable_intrs(dev); + kfree(dev->sgid_tbl); +err_free_uar_table: + pvrdma_uar_table_cleanup(dev); +err_free_intrs: + pvrdma_free_irq(dev); + pci_free_irq_vectors(pdev); +err_free_cq_ring: + pvrdma_page_dir_cleanup(dev, &dev->cq_pdir); +err_free_async_ring: + pvrdma_page_dir_cleanup(dev, &dev->async_pdir); +err_free_slots: + pvrdma_free_slots(dev); +err_free_dsr: + dma_free_coherent(&pdev->dev, sizeof(*dev->dsr), dev->dsr, + dev->dsrbase); +err_uar_unmap: + iounmap(dev->driver_uar.map); +err_unmap_regs: + iounmap(dev->regs); +err_free_resource: + pci_release_regions(pdev); +err_disable_pdev: + pci_disable_device(pdev); + pci_set_drvdata(pdev, NULL); +err_free_device: + mutex_lock(&pvrdma_device_list_lock); + list_del(&dev->device_link); + mutex_unlock(&pvrdma_device_list_lock); + ib_dealloc_device(&dev->ib_dev); + return ret; +} + +static void pvrdma_pci_remove(struct pci_dev *pdev) +{ + struct pvrdma_dev *dev = pci_get_drvdata(pdev); + + if (!dev) + return; + + dev_info(&pdev->dev, "detaching from device\n"); + + unregister_netdevice_notifier(&dev->nb_netdev); + dev->nb_netdev.notifier_call = NULL; + + flush_workqueue(event_wq); + + /* Unregister ib device */ + ib_unregister_device(&dev->ib_dev); + + mutex_lock(&pvrdma_device_list_lock); + list_del(&dev->device_link); + mutex_unlock(&pvrdma_device_list_lock); + + pvrdma_disable_intrs(dev); + pvrdma_free_irq(dev); + pci_free_irq_vectors(pdev); + + /* Deactivate pvrdma device */ + pvrdma_write_reg(dev, PVRDMA_REG_CTL, PVRDMA_DEVICE_CTL_RESET); + pvrdma_page_dir_cleanup(dev, &dev->cq_pdir); + pvrdma_page_dir_cleanup(dev, &dev->async_pdir); + pvrdma_free_slots(dev); + + iounmap(dev->regs); + kfree(dev->sgid_tbl); + kfree(dev->cq_tbl); + kfree(dev->qp_tbl); + pvrdma_uar_table_cleanup(dev); + iounmap(dev->driver_uar.map); + + ib_dealloc_device(&dev->ib_dev); + + /* Free pci resources */ + pci_release_regions(pdev); + pci_disable_device(pdev); + pci_set_drvdata(pdev, NULL); +} + +static struct pci_device_id pvrdma_pci_table[] = { + { PCI_DEVICE(PCI_VENDOR_ID_VMWARE, PCI_DEVICE_ID_VMWARE_PVRDMA), }, + { 0 }, +}; + +MODULE_DEVICE_TABLE(pci, pvrdma_pci_table); + +static struct pci_driver pvrdma_driver = { + .name = DRV_NAME, + .id_table = pvrdma_pci_table, + .probe = pvrdma_pci_probe, + .remove = pvrdma_pci_remove, +}; + +static int __init pvrdma_init(void) +{ + int err; + + event_wq = alloc_ordered_workqueue("pvrdma_event_wq", WQ_MEM_RECLAIM); + if (!event_wq) + return -ENOMEM; + + err = pci_register_driver(&pvrdma_driver); + if (err) + destroy_workqueue(event_wq); + + return err; +} + +static void __exit pvrdma_cleanup(void) +{ + pci_unregister_driver(&pvrdma_driver); + + destroy_workqueue(event_wq); +} + +module_init(pvrdma_init); +module_exit(pvrdma_cleanup); + +MODULE_AUTHOR("VMware, Inc"); +MODULE_DESCRIPTION("VMware Paravirtual RDMA driver"); +MODULE_VERSION(DRV_VERSION); +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_misc.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_misc.c new file mode 100644 index 000000000000..948b5ccd2a70 --- /dev/null +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_misc.c @@ -0,0 +1,304 @@ +/* + * Copyright (c) 2012-2016 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of EITHER the GNU General Public License + * version 2 as published by the Free Software Foundation or the BSD + * 2-Clause License. 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 version 2 for more details at + * http://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html. + * + * You should have received a copy of the GNU General Public License + * along with this program available in the file COPYING in the main + * directory of this source tree. + * + * The BSD 2-Clause License + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE + * COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/bitmap.h> + +#include "pvrdma.h" + +int pvrdma_page_dir_init(struct pvrdma_dev *dev, struct pvrdma_page_dir *pdir, + u64 npages, bool alloc_pages) +{ + u64 i; + + if (npages > PVRDMA_PAGE_DIR_MAX_PAGES) + return -EINVAL; + + memset(pdir, 0, sizeof(*pdir)); + + pdir->dir = dma_alloc_coherent(&dev->pdev->dev, PAGE_SIZE, + &pdir->dir_dma, GFP_KERNEL); + if (!pdir->dir) + goto err; + + pdir->ntables = PVRDMA_PAGE_DIR_TABLE(npages - 1) + 1; + pdir->tables = kcalloc(pdir->ntables, sizeof(*pdir->tables), + GFP_KERNEL); + if (!pdir->tables) + goto err; + + for (i = 0; i < pdir->ntables; i++) { + pdir->tables[i] = dma_alloc_coherent(&dev->pdev->dev, PAGE_SIZE, + (dma_addr_t *)&pdir->dir[i], + GFP_KERNEL); + if (!pdir->tables[i]) + goto err; + } + + pdir->npages = npages; + + if (alloc_pages) { + pdir->pages = kcalloc(npages, sizeof(*pdir->pages), + GFP_KERNEL); + if (!pdir->pages) + goto err; + + for (i = 0; i < pdir->npages; i++) { + dma_addr_t page_dma; + + pdir->pages[i] = dma_alloc_coherent(&dev->pdev->dev, + PAGE_SIZE, + &page_dma, + GFP_KERNEL); + if (!pdir->pages[i]) + goto err; + + pvrdma_page_dir_insert_dma(pdir, i, page_dma); + } + } + + return 0; + +err: + pvrdma_page_dir_cleanup(dev, pdir); + + return -ENOMEM; +} + +static u64 *pvrdma_page_dir_table(struct pvrdma_page_dir *pdir, u64 idx) +{ + return pdir->tables[PVRDMA_PAGE_DIR_TABLE(idx)]; +} + +dma_addr_t pvrdma_page_dir_get_dma(struct pvrdma_page_dir *pdir, u64 idx) +{ + return pvrdma_page_dir_table(pdir, idx)[PVRDMA_PAGE_DIR_PAGE(idx)]; +} + +static void pvrdma_page_dir_cleanup_pages(struct pvrdma_dev *dev, + struct pvrdma_page_dir *pdir) +{ + if (pdir->pages) { + u64 i; + + for (i = 0; i < pdir->npages && pdir->pages[i]; i++) { + dma_addr_t page_dma = pvrdma_page_dir_get_dma(pdir, i); + + dma_free_coherent(&dev->pdev->dev, PAGE_SIZE, + pdir->pages[i], page_dma); + } + + kfree(pdir->pages); + } +} + +static void pvrdma_page_dir_cleanup_tables(struct pvrdma_dev *dev, + struct pvrdma_page_dir *pdir) +{ + if (pdir->tables) { + int i; + + pvrdma_page_dir_cleanup_pages(dev, pdir); + + for (i = 0; i < pdir->ntables; i++) { + u64 *table = pdir->tables[i]; + + if (table) + dma_free_coherent(&dev->pdev->dev, PAGE_SIZE, + table, pdir->dir[i]); + } + + kfree(pdir->tables); + } +} + +void pvrdma_page_dir_cleanup(struct pvrdma_dev *dev, + struct pvrdma_page_dir *pdir) +{ + if (pdir->dir) { + pvrdma_page_dir_cleanup_tables(dev, pdir); + dma_free_coherent(&dev->pdev->dev, PAGE_SIZE, + pdir->dir, pdir->dir_dma); + } +} + +int pvrdma_page_dir_insert_dma(struct pvrdma_page_dir *pdir, u64 idx, + dma_addr_t daddr) +{ + u64 *table; + + if (idx >= pdir->npages) + return -EINVAL; + + table = pvrdma_page_dir_table(pdir, idx); + table[PVRDMA_PAGE_DIR_PAGE(idx)] = daddr; + + return 0; +} + +int pvrdma_page_dir_insert_umem(struct pvrdma_page_dir *pdir, + struct ib_umem *umem, u64 offset) +{ + u64 i = offset; + int j, entry; + int ret = 0, len = 0; + struct scatterlist *sg; + + if (offset >= pdir->npages) + return -EINVAL; + + for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) { + len = sg_dma_len(sg) >> PAGE_SHIFT; + for (j = 0; j < len; j++) { + dma_addr_t addr = sg_dma_address(sg) + + umem->page_size * j; + + ret = pvrdma_page_dir_insert_dma(pdir, i, addr); + if (ret) + goto exit; + + i++; + } + } + +exit: + return ret; +} + +int pvrdma_page_dir_insert_page_list(struct pvrdma_page_dir *pdir, + u64 *page_list, + int num_pages) +{ + int i; + int ret; + + if (num_pages > pdir->npages) + return -EINVAL; + + for (i = 0; i < num_pages; i++) { + ret = pvrdma_page_dir_insert_dma(pdir, i, page_list[i]); + if (ret) + return ret; + } + + return 0; +} + +void pvrdma_qp_cap_to_ib(struct ib_qp_cap *dst, const struct pvrdma_qp_cap *src) +{ + dst->max_send_wr = src->max_send_wr; + dst->max_recv_wr = src->max_recv_wr; + dst->max_send_sge = src->max_send_sge; + dst->max_recv_sge = src->max_recv_sge; + dst->max_inline_data = src->max_inline_data; +} + +void ib_qp_cap_to_pvrdma(struct pvrdma_qp_cap *dst, const struct ib_qp_cap *src) +{ + dst->max_send_wr = src->max_send_wr; + dst->max_recv_wr = src->max_recv_wr; + dst->max_send_sge = src->max_send_sge; + dst->max_recv_sge = src->max_recv_sge; + dst->max_inline_data = src->max_inline_data; +} + +void pvrdma_gid_to_ib(union ib_gid *dst, const union pvrdma_gid *src) +{ + BUILD_BUG_ON(sizeof(union pvrdma_gid) != sizeof(union ib_gid)); + memcpy(dst, src, sizeof(*src)); +} + +void ib_gid_to_pvrdma(union pvrdma_gid *dst, const union ib_gid *src) +{ + BUILD_BUG_ON(sizeof(union pvrdma_gid) != sizeof(union ib_gid)); + memcpy(dst, src, sizeof(*src)); +} + +void pvrdma_global_route_to_ib(struct ib_global_route *dst, + const struct pvrdma_global_route *src) +{ + pvrdma_gid_to_ib(&dst->dgid, &src->dgid); + dst->flow_label = src->flow_label; + dst->sgid_index = src->sgid_index; + dst->hop_limit = src->hop_limit; + dst->traffic_class = src->traffic_class; +} + +void ib_global_route_to_pvrdma(struct pvrdma_global_route *dst, + const struct ib_global_route *src) +{ + ib_gid_to_pvrdma(&dst->dgid, &src->dgid); + dst->flow_label = src->flow_label; + dst->sgid_index = src->sgid_index; + dst->hop_limit = src->hop_limit; + dst->traffic_class = src->traffic_class; +} + +void pvrdma_ah_attr_to_ib(struct ib_ah_attr *dst, + const struct pvrdma_ah_attr *src) +{ + pvrdma_global_route_to_ib(&dst->grh, &src->grh); + dst->dlid = src->dlid; + dst->sl = src->sl; + dst->src_path_bits = src->src_path_bits; + dst->static_rate = src->static_rate; + dst->ah_flags = src->ah_flags; + dst->port_num = src->port_num; + memcpy(&dst->dmac, &src->dmac, sizeof(dst->dmac)); +} + +void ib_ah_attr_to_pvrdma(struct pvrdma_ah_attr *dst, + const struct ib_ah_attr *src) +{ + ib_global_route_to_pvrdma(&dst->grh, &src->grh); + dst->dlid = src->dlid; + dst->sl = src->sl; + dst->src_path_bits = src->src_path_bits; + dst->static_rate = src->static_rate; + dst->ah_flags = src->ah_flags; + dst->port_num = src->port_num; + memcpy(&dst->dmac, &src->dmac, sizeof(dst->dmac)); +} diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_mr.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_mr.c new file mode 100644 index 000000000000..8519f3212e52 --- /dev/null +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_mr.c @@ -0,0 +1,334 @@ +/* + * Copyright (c) 2012-2016 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of EITHER the GNU General Public License + * version 2 as published by the Free Software Foundation or the BSD + * 2-Clause License. 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 version 2 for more details at + * http://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html. + * + * You should have received a copy of the GNU General Public License + * along with this program available in the file COPYING in the main + * directory of this source tree. + * + * The BSD 2-Clause License + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE + * COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <linux/list.h> +#include <linux/slab.h> + +#include "pvrdma.h" + +/** + * pvrdma_get_dma_mr - get a DMA memory region + * @pd: protection domain + * @acc: access flags + * + * @return: ib_mr pointer on success, otherwise returns an errno. + */ +struct ib_mr *pvrdma_get_dma_mr(struct ib_pd *pd, int acc) +{ + struct pvrdma_dev *dev = to_vdev(pd->device); + struct pvrdma_user_mr *mr; + union pvrdma_cmd_req req; + union pvrdma_cmd_resp rsp; + struct pvrdma_cmd_create_mr *cmd = &req.create_mr; + struct pvrdma_cmd_create_mr_resp *resp = &rsp.create_mr_resp; + int ret; + + /* Support only LOCAL_WRITE flag for DMA MRs */ + if (acc & ~IB_ACCESS_LOCAL_WRITE) { + dev_warn(&dev->pdev->dev, + "unsupported dma mr access flags %#x\n", acc); + return ERR_PTR(-EOPNOTSUPP); + } + + mr = kzalloc(sizeof(*mr), GFP_KERNEL); + if (!mr) + return ERR_PTR(-ENOMEM); + + memset(cmd, 0, sizeof(*cmd)); + cmd->hdr.cmd = PVRDMA_CMD_CREATE_MR; + cmd->pd_handle = to_vpd(pd)->pd_handle; + cmd->access_flags = acc; + cmd->flags = PVRDMA_MR_FLAG_DMA; + + ret = pvrdma_cmd_post(dev, &req, &rsp, PVRDMA_CMD_CREATE_MR_RESP); + if (ret < 0) { + dev_warn(&dev->pdev->dev, + "could not get DMA mem region, error: %d\n", ret); + kfree(mr); + return ERR_PTR(ret); + } + + mr->mmr.mr_handle = resp->mr_handle; + mr->ibmr.lkey = resp->lkey; + mr->ibmr.rkey = resp->rkey; + + return &mr->ibmr; +} + +/** + * pvrdma_reg_user_mr - register a userspace memory region + * @pd: protection domain + * @start: starting address + * @length: length of region + * @virt_addr: I/O virtual address + * @access_flags: access flags for memory region + * @udata: user data + * + * @return: ib_mr pointer on success, otherwise returns an errno. + */ +struct ib_mr *pvrdma_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, + u64 virt_addr, int access_flags, + struct ib_udata *udata) +{ + struct pvrdma_dev *dev = to_vdev(pd->device); + struct pvrdma_user_mr *mr = NULL; + struct ib_umem *umem; + union pvrdma_cmd_req req; + union pvrdma_cmd_resp rsp; + struct pvrdma_cmd_create_mr *cmd = &req.create_mr; + struct pvrdma_cmd_create_mr_resp *resp = &rsp.create_mr_resp; + int nchunks; + int ret; + int entry; + struct scatterlist *sg; + + if (length == 0 || length > dev->dsr->caps.max_mr_size) { + dev_warn(&dev->pdev->dev, "invalid mem region length\n"); + return ERR_PTR(-EINVAL); + } + + umem = ib_umem_get(pd->uobject->context, start, + length, access_flags, 0); + if (IS_ERR(umem)) { + dev_warn(&dev->pdev->dev, + "could not get umem for mem region\n"); + return ERR_CAST(umem); + } + + nchunks = 0; + for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) + nchunks += sg_dma_len(sg) >> PAGE_SHIFT; + + if (nchunks < 0 || nchunks > PVRDMA_PAGE_DIR_MAX_PAGES) { + dev_warn(&dev->pdev->dev, "overflow %d pages in mem region\n", + nchunks); + ret = -EINVAL; + goto err_umem; + } + + mr = kzalloc(sizeof(*mr), GFP_KERNEL); + if (!mr) { + ret = -ENOMEM; + goto err_umem; + } + + mr->mmr.iova = virt_addr; + mr->mmr.size = length; + mr->umem = umem; + + ret = pvrdma_page_dir_init(dev, &mr->pdir, nchunks, false); + if (ret) { + dev_warn(&dev->pdev->dev, + "could not allocate page directory\n"); + goto err_umem; + } + + ret = pvrdma_page_dir_insert_umem(&mr->pdir, mr->umem, 0); + if (ret) + goto err_pdir; + + memset(cmd, 0, sizeof(*cmd)); + cmd->hdr.cmd = PVRDMA_CMD_CREATE_MR; + cmd->start = start; + cmd->length = length; + cmd->pd_handle = to_vpd(pd)->pd_handle; + cmd->access_flags = access_flags; + cmd->nchunks = nchunks; + cmd->pdir_dma = mr->pdir.dir_dma; + + ret = pvrdma_cmd_post(dev, &req, &rsp, PVRDMA_CMD_CREATE_MR_RESP); + if (ret < 0) { + dev_warn(&dev->pdev->dev, + "could not register mem region, error: %d\n", ret); + goto err_pdir; + } + + mr->mmr.mr_handle = resp->mr_handle; + mr->ibmr.lkey = resp->lkey; + mr->ibmr.rkey = resp->rkey; + + return &mr->ibmr; + +err_pdir: + pvrdma_page_dir_cleanup(dev, &mr->pdir); +err_umem: + ib_umem_release(umem); + kfree(mr); + + return ERR_PTR(ret); +} + +/** + * pvrdma_alloc_mr - allocate a memory region + * @pd: protection domain + * @mr_type: type of memory region + * @max_num_sg: maximum number of pages + * + * @return: ib_mr pointer on success, otherwise returns an errno. + */ +struct ib_mr *pvrdma_alloc_mr(struct ib_pd *pd, enum ib_mr_type mr_type, + u32 max_num_sg) +{ + struct pvrdma_dev *dev = to_vdev(pd->device); + struct pvrdma_user_mr *mr; + union pvrdma_cmd_req req; + union pvrdma_cmd_resp rsp; + struct pvrdma_cmd_create_mr *cmd = &req.create_mr; + struct pvrdma_cmd_create_mr_resp *resp = &rsp.create_mr_resp; + int size = max_num_sg * sizeof(u64); + int ret; + + if (mr_type != IB_MR_TYPE_MEM_REG || + max_num_sg > PVRDMA_MAX_FAST_REG_PAGES) + return ERR_PTR(-EINVAL); + + mr = kzalloc(sizeof(*mr), GFP_KERNEL); + if (!mr) + return ERR_PTR(-ENOMEM); + + mr->pages = kzalloc(size, GFP_KERNEL); + if (!mr->pages) { + ret = -ENOMEM; + goto freemr; + } + + ret = pvrdma_page_dir_init(dev, &mr->pdir, max_num_sg, false); + if (ret) { + dev_warn(&dev->pdev->dev, + "failed to allocate page dir for mr\n"); + ret = -ENOMEM; + goto freepages; + } + + memset(cmd, 0, sizeof(*cmd)); + cmd->hdr.cmd = PVRDMA_CMD_CREATE_MR; + cmd->pd_handle = to_vpd(pd)->pd_handle; + cmd->access_flags = 0; + cmd->flags = PVRDMA_MR_FLAG_FRMR; + cmd->nchunks = max_num_sg; + + ret = pvrdma_cmd_post(dev, &req, &rsp, PVRDMA_CMD_CREATE_MR_RESP); + if (ret < 0) { + dev_warn(&dev->pdev->dev, + "could not create FR mem region, error: %d\n", ret); + goto freepdir; + } + + mr->max_pages = max_num_sg; + mr->mmr.mr_handle = resp->mr_handle; + mr->ibmr.lkey = resp->lkey; + mr->ibmr.rkey = resp->rkey; + mr->page_shift = PAGE_SHIFT; + mr->umem = NULL; + + return &mr->ibmr; + +freepdir: + pvrdma_page_dir_cleanup(dev, &mr->pdir); +freepages: + kfree(mr->pages); +freemr: + kfree(mr); + return ERR_PTR(ret); +} + +/** + * pvrdma_dereg_mr - deregister a memory region + * @ibmr: memory region + * + * @return: 0 on success. + */ +int pvrdma_dereg_mr(struct ib_mr *ibmr) +{ + struct pvrdma_user_mr *mr = to_vmr(ibmr); + struct pvrdma_dev *dev = to_vdev(ibmr->device); + union pvrdma_cmd_req req; + struct pvrdma_cmd_destroy_mr *cmd = &req.destroy_mr; + int ret; + + memset(cmd, 0, sizeof(*cmd)); + cmd->hdr.cmd = PVRDMA_CMD_DESTROY_MR; + cmd->mr_handle = mr->mmr.mr_handle; + ret = pvrdma_cmd_post(dev, &req, NULL, 0); + if (ret < 0) + dev_warn(&dev->pdev->dev, + "could not deregister mem region, error: %d\n", ret); + + pvrdma_page_dir_cleanup(dev, &mr->pdir); + if (mr->umem) + ib_umem_release(mr->umem); + + kfree(mr->pages); + kfree(mr); + + return 0; +} + +static int pvrdma_set_page(struct ib_mr *ibmr, u64 addr) +{ + struct pvrdma_user_mr *mr = to_vmr(ibmr); + + if (mr->npages == mr->max_pages) + return -ENOMEM; + + mr->pages[mr->npages++] = addr; + return 0; +} + +int pvrdma_map_mr_sg(struct ib_mr *ibmr, struct scatterlist *sg, int sg_nents, + unsigned int *sg_offset) +{ + struct pvrdma_user_mr *mr = to_vmr(ibmr); + struct pvrdma_dev *dev = to_vdev(ibmr->device); + int ret; + + mr->npages = 0; + + ret = ib_sg_to_pages(ibmr, sg, sg_nents, sg_offset, pvrdma_set_page); + if (ret < 0) + dev_warn(&dev->pdev->dev, "could not map sg to pages\n"); + + return ret; +} diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_qp.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_qp.c new file mode 100644 index 000000000000..dbbfd35e7da7 --- /dev/null +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_qp.c @@ -0,0 +1,971 @@ +/* + * Copyright (c) 2012-2016 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of EITHER the GNU General Public License + * version 2 as published by the Free Software Foundation or the BSD + * 2-Clause License. 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 version 2 for more details at + * http://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html. + * + * You should have received a copy of the GNU General Public License + * along with this program available in the file COPYING in the main + * directory of this source tree. + * + * The BSD 2-Clause License + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE + * COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <asm/page.h> +#include <linux/io.h> +#include <linux/wait.h> +#include <rdma/ib_addr.h> +#include <rdma/ib_smi.h> +#include <rdma/ib_user_verbs.h> + +#include "pvrdma.h" + +static inline void get_cqs(struct pvrdma_qp *qp, struct pvrdma_cq **send_cq, + struct pvrdma_cq **recv_cq) +{ + *send_cq = to_vcq(qp->ibqp.send_cq); + *recv_cq = to_vcq(qp->ibqp.recv_cq); +} + +static void pvrdma_lock_cqs(struct pvrdma_cq *scq, struct pvrdma_cq *rcq, + unsigned long *scq_flags, + unsigned long *rcq_flags) + __acquires(scq->cq_lock) __acquires(rcq->cq_lock) +{ + if (scq == rcq) { + spin_lock_irqsave(&scq->cq_lock, *scq_flags); + __acquire(rcq->cq_lock); + } else if (scq->cq_handle < rcq->cq_handle) { + spin_lock_irqsave(&scq->cq_lock, *scq_flags); + spin_lock_irqsave_nested(&rcq->cq_lock, *rcq_flags, + SINGLE_DEPTH_NESTING); + } else { + spin_lock_irqsave(&rcq->cq_lock, *rcq_flags); + spin_lock_irqsave_nested(&scq->cq_lock, *scq_flags, + SINGLE_DEPTH_NESTING); + } +} + +static void pvrdma_unlock_cqs(struct pvrdma_cq *scq, struct pvrdma_cq *rcq, + unsigned long *scq_flags, + unsigned long *rcq_flags) + __releases(scq->cq_lock) __releases(rcq->cq_lock) +{ + if (scq == rcq) { + __release(rcq->cq_lock); + spin_unlock_irqrestore(&scq->cq_lock, *scq_flags); + } else if (scq->cq_handle < rcq->cq_handle) { + spin_unlock_irqrestore(&rcq->cq_lock, *rcq_flags); + spin_unlock_irqrestore(&scq->cq_lock, *scq_flags); + } else { + spin_unlock_irqrestore(&scq->cq_lock, *scq_flags); + spin_unlock_irqrestore(&rcq->cq_lock, *rcq_flags); + } +} + +static void pvrdma_reset_qp(struct pvrdma_qp *qp) +{ + struct pvrdma_cq *scq, *rcq; + unsigned long scq_flags, rcq_flags; + + /* Clean up cqes */ + get_cqs(qp, &scq, &rcq); + pvrdma_lock_cqs(scq, rcq, &scq_flags, &rcq_flags); + + _pvrdma_flush_cqe(qp, scq); + if (scq != rcq) + _pvrdma_flush_cqe(qp, rcq); + + pvrdma_unlock_cqs(scq, rcq, &scq_flags, &rcq_flags); + + /* + * Reset queuepair. The checks are because usermode queuepairs won't + * have kernel ringstates. + */ + if (qp->rq.ring) { + atomic_set(&qp->rq.ring->cons_head, 0); + atomic_set(&qp->rq.ring->prod_tail, 0); + } + if (qp->sq.ring) { + atomic_set(&qp->sq.ring->cons_head, 0); + atomic_set(&qp->sq.ring->prod_tail, 0); + } +} + +static int pvrdma_set_rq_size(struct pvrdma_dev *dev, + struct ib_qp_cap *req_cap, + struct pvrdma_qp *qp) +{ + if (req_cap->max_recv_wr > dev->dsr->caps.max_qp_wr || + req_cap->max_recv_sge > dev->dsr->caps.max_sge) { + dev_warn(&dev->pdev->dev, "recv queue size invalid\n"); + return -EINVAL; + } + + qp->rq.wqe_cnt = roundup_pow_of_two(max(1U, req_cap->max_recv_wr)); + qp->rq.max_sg = roundup_pow_of_two(max(1U, req_cap->max_recv_sge)); + + /* Write back */ + req_cap->max_recv_wr = qp->rq.wqe_cnt; + req_cap->max_recv_sge = qp->rq.max_sg; + + qp->rq.wqe_size = roundup_pow_of_two(sizeof(struct pvrdma_rq_wqe_hdr) + + sizeof(struct pvrdma_sge) * + qp->rq.max_sg); + qp->npages_recv = (qp->rq.wqe_cnt * qp->rq.wqe_size + PAGE_SIZE - 1) / + PAGE_SIZE; + + return 0; +} + +static int pvrdma_set_sq_size(struct pvrdma_dev *dev, struct ib_qp_cap *req_cap, + struct pvrdma_qp *qp) +{ + if (req_cap->max_send_wr > dev->dsr->caps.max_qp_wr || + req_cap->max_send_sge > dev->dsr->caps.max_sge) { + dev_warn(&dev->pdev->dev, "send queue size invalid\n"); + return -EINVAL; + } + + qp->sq.wqe_cnt = roundup_pow_of_two(max(1U, req_cap->max_send_wr)); + qp->sq.max_sg = roundup_pow_of_two(max(1U, req_cap->max_send_sge)); + + /* Write back */ + req_cap->max_send_wr = qp->sq.wqe_cnt; + req_cap->max_send_sge = qp->sq.max_sg; + + qp->sq.wqe_size = roundup_pow_of_two(sizeof(struct pvrdma_sq_wqe_hdr) + + sizeof(struct pvrdma_sge) * + qp->sq.max_sg); + /* Note: one extra page for the header. */ + qp->npages_send = 1 + (qp->sq.wqe_cnt * qp->sq.wqe_size + + PAGE_SIZE - 1) / PAGE_SIZE; + + return 0; +} + +/** + * pvrdma_create_qp - create queue pair + * @pd: protection domain + * @init_attr: queue pair attributes + * @udata: user data + * + * @return: the ib_qp pointer on success, otherwise returns an errno. + */ +struct ib_qp *pvrdma_create_qp(struct ib_pd *pd, + struct ib_qp_init_attr *init_attr, + struct ib_udata *udata) +{ + struct pvrdma_qp *qp = NULL; + struct pvrdma_dev *dev = to_vdev(pd->device); + union pvrdma_cmd_req req; + union pvrdma_cmd_resp rsp; + struct pvrdma_cmd_create_qp *cmd = &req.create_qp; + struct pvrdma_cmd_create_qp_resp *resp = &rsp.create_qp_resp; + struct pvrdma_create_qp ucmd; + unsigned long flags; + int ret; + + if (init_attr->create_flags) { + dev_warn(&dev->pdev->dev, + "invalid create queuepair flags %#x\n", + init_attr->create_flags); + return ERR_PTR(-EINVAL); + } + + if (init_attr->qp_type != IB_QPT_RC && + init_attr->qp_type != IB_QPT_UD && + init_attr->qp_type != IB_QPT_GSI) { + dev_warn(&dev->pdev->dev, "queuepair type %d not supported\n", + init_attr->qp_type); + return ERR_PTR(-EINVAL); + } + + if (!atomic_add_unless(&dev->num_qps, 1, dev->dsr->caps.max_qp)) + return ERR_PTR(-ENOMEM); + + switch (init_attr->qp_type) { + case IB_QPT_GSI: + if (init_attr->port_num == 0 || + init_attr->port_num > pd->device->phys_port_cnt || + udata) { + dev_warn(&dev->pdev->dev, "invalid queuepair attrs\n"); + ret = -EINVAL; + goto err_qp; + } + /* fall through */ + case IB_QPT_RC: + case IB_QPT_UD: + qp = kzalloc(sizeof(*qp), GFP_KERNEL); + if (!qp) { + ret = -ENOMEM; + goto err_qp; + } + + spin_lock_init(&qp->sq.lock); + spin_lock_init(&qp->rq.lock); + mutex_init(&qp->mutex); + atomic_set(&qp->refcnt, 1); + init_waitqueue_head(&qp->wait); + + qp->state = IB_QPS_RESET; + + if (pd->uobject && udata) { + dev_dbg(&dev->pdev->dev, + "create queuepair from user space\n"); + + if (ib_copy_from_udata(&ucmd, udata, sizeof(ucmd))) { + ret = -EFAULT; + goto err_qp; + } + + /* set qp->sq.wqe_cnt, shift, buf_size.. */ + qp->rumem = ib_umem_get(pd->uobject->context, + ucmd.rbuf_addr, + ucmd.rbuf_size, 0, 0); + if (IS_ERR(qp->rumem)) { + ret = PTR_ERR(qp->rumem); + goto err_qp; + } + + qp->sumem = ib_umem_get(pd->uobject->context, + ucmd.sbuf_addr, + ucmd.sbuf_size, 0, 0); + if (IS_ERR(qp->sumem)) { + ib_umem_release(qp->rumem); + ret = PTR_ERR(qp->sumem); + goto err_qp; + } + + qp->npages_send = ib_umem_page_count(qp->sumem); + qp->npages_recv = ib_umem_page_count(qp->rumem); + qp->npages = qp->npages_send + qp->npages_recv; + } else { + qp->is_kernel = true; + + ret = pvrdma_set_sq_size(to_vdev(pd->device), + &init_attr->cap, qp); + if (ret) + goto err_qp; + + ret = pvrdma_set_rq_size(to_vdev(pd->device), + &init_attr->cap, qp); + if (ret) + goto err_qp; + + qp->npages = qp->npages_send + qp->npages_recv; + + /* Skip header page. */ + qp->sq.offset = PAGE_SIZE; + + /* Recv queue pages are after send pages. */ + qp->rq.offset = qp->npages_send * PAGE_SIZE; + } + + if (qp->npages < 0 || qp->npages > PVRDMA_PAGE_DIR_MAX_PAGES) { + dev_warn(&dev->pdev->dev, + "overflow pages in queuepair\n"); + ret = -EINVAL; + goto err_umem; + } + + ret = pvrdma_page_dir_init(dev, &qp->pdir, qp->npages, + qp->is_kernel); + if (ret) { + dev_warn(&dev->pdev->dev, + "could not allocate page directory\n"); + goto err_umem; + } + + if (!qp->is_kernel) { + pvrdma_page_dir_insert_umem(&qp->pdir, qp->sumem, 0); + pvrdma_page_dir_insert_umem(&qp->pdir, qp->rumem, + qp->npages_send); + } else { + /* Ring state is always the first page. */ + qp->sq.ring = qp->pdir.pages[0]; + qp->rq.ring = &qp->sq.ring[1]; + } + break; + default: + ret = -EINVAL; + goto err_qp; + } + + /* Not supported */ + init_attr->cap.max_inline_data = 0; + + memset(cmd, 0, sizeof(*cmd)); + cmd->hdr.cmd = PVRDMA_CMD_CREATE_QP; + cmd->pd_handle = to_vpd(pd)->pd_handle; + cmd->send_cq_handle = to_vcq(init_attr->send_cq)->cq_handle; + cmd->recv_cq_handle = to_vcq(init_attr->recv_cq)->cq_handle; + cmd->max_send_wr = init_attr->cap.max_send_wr; + cmd->max_recv_wr = init_attr->cap.max_recv_wr; + cmd->max_send_sge = init_attr->cap.max_send_sge; + cmd->max_recv_sge = init_attr->cap.max_recv_sge; + cmd->max_inline_data = init_attr->cap.max_inline_data; + cmd->sq_sig_all = (init_attr->sq_sig_type == IB_SIGNAL_ALL_WR) ? 1 : 0; + cmd->qp_type = ib_qp_type_to_pvrdma(init_attr->qp_type); + cmd->access_flags = IB_ACCESS_LOCAL_WRITE; + cmd->total_chunks = qp->npages; + cmd->send_chunks = qp->npages_send - 1; + cmd->pdir_dma = qp->pdir.dir_dma; + + dev_dbg(&dev->pdev->dev, "create queuepair with %d, %d, %d, %d\n", + cmd->max_send_wr, cmd->max_recv_wr, cmd->max_send_sge, + cmd->max_recv_sge); + + ret = pvrdma_cmd_post(dev, &req, &rsp, PVRDMA_CMD_CREATE_QP_RESP); + if (ret < 0) { + dev_warn(&dev->pdev->dev, + "could not create queuepair, error: %d\n", ret); + goto err_pdir; + } + + /* max_send_wr/_recv_wr/_send_sge/_recv_sge/_inline_data */ + qp->qp_handle = resp->qpn; + qp->port = init_attr->port_num; + qp->ibqp.qp_num = resp->qpn; + spin_lock_irqsave(&dev->qp_tbl_lock, flags); + dev->qp_tbl[qp->qp_handle % dev->dsr->caps.max_qp] = qp; + spin_unlock_irqrestore(&dev->qp_tbl_lock, flags); + + return &qp->ibqp; + +err_pdir: + pvrdma_page_dir_cleanup(dev, &qp->pdir); +err_umem: + if (pd->uobject && udata) { + if (qp->rumem) + ib_umem_release(qp->rumem); + if (qp->sumem) + ib_umem_release(qp->sumem); + } +err_qp: + kfree(qp); + atomic_dec(&dev->num_qps); + + return ERR_PTR(ret); +} + +static void pvrdma_free_qp(struct pvrdma_qp *qp) +{ + struct pvrdma_dev *dev = to_vdev(qp->ibqp.device); + struct pvrdma_cq *scq; + struct pvrdma_cq *rcq; + unsigned long flags, scq_flags, rcq_flags; + + /* In case cq is polling */ + get_cqs(qp, &scq, &rcq); + pvrdma_lock_cqs(scq, rcq, &scq_flags, &rcq_flags); + + _pvrdma_flush_cqe(qp, scq); + if (scq != rcq) + _pvrdma_flush_cqe(qp, rcq); + + spin_lock_irqsave(&dev->qp_tbl_lock, flags); + dev->qp_tbl[qp->qp_handle] = NULL; + spin_unlock_irqrestore(&dev->qp_tbl_lock, flags); + + pvrdma_unlock_cqs(scq, rcq, &scq_flags, &rcq_flags); + + atomic_dec(&qp->refcnt); + wait_event(qp->wait, !atomic_read(&qp->refcnt)); + + pvrdma_page_dir_cleanup(dev, &qp->pdir); + + kfree(qp); + + atomic_dec(&dev->num_qps); +} + +/** + * pvrdma_destroy_qp - destroy a queue pair + * @qp: the queue pair to destroy + * + * @return: 0 on success. + */ +int pvrdma_destroy_qp(struct ib_qp *qp) +{ + struct pvrdma_qp *vqp = to_vqp(qp); + union pvrdma_cmd_req req; + struct pvrdma_cmd_destroy_qp *cmd = &req.destroy_qp; + int ret; + + memset(cmd, 0, sizeof(*cmd)); + cmd->hdr.cmd = PVRDMA_CMD_DESTROY_QP; + cmd->qp_handle = vqp->qp_handle; + + ret = pvrdma_cmd_post(to_vdev(qp->device), &req, NULL, 0); + if (ret < 0) + dev_warn(&to_vdev(qp->device)->pdev->dev, + "destroy queuepair failed, error: %d\n", ret); + + pvrdma_free_qp(vqp); + + return 0; +} + +/** + * pvrdma_modify_qp - modify queue pair attributes + * @ibqp: the queue pair + * @attr: the new queue pair's attributes + * @attr_mask: attributes mask + * @udata: user data + * + * @returns 0 on success, otherwise returns an errno. + */ +int pvrdma_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, + int attr_mask, struct ib_udata *udata) +{ + struct pvrdma_dev *dev = to_vdev(ibqp->device); + struct pvrdma_qp *qp = to_vqp(ibqp); + union pvrdma_cmd_req req; + union pvrdma_cmd_resp rsp; + struct pvrdma_cmd_modify_qp *cmd = &req.modify_qp; + int cur_state, next_state; + int ret; + + /* Sanity checking. Should need lock here */ + mutex_lock(&qp->mutex); + cur_state = (attr_mask & IB_QP_CUR_STATE) ? attr->cur_qp_state : + qp->state; + next_state = (attr_mask & IB_QP_STATE) ? attr->qp_state : cur_state; + + if (!ib_modify_qp_is_ok(cur_state, next_state, ibqp->qp_type, + attr_mask, IB_LINK_LAYER_ETHERNET)) { + ret = -EINVAL; + goto out; + } + + if (attr_mask & IB_QP_PORT) { + if (attr->port_num == 0 || + attr->port_num > ibqp->device->phys_port_cnt) { + ret = -EINVAL; + goto out; + } + } + + if (attr_mask & IB_QP_MIN_RNR_TIMER) { + if (attr->min_rnr_timer > 31) { + ret = -EINVAL; + goto out; + } + } + + if (attr_mask & IB_QP_PKEY_INDEX) { + if (attr->pkey_index >= dev->dsr->caps.max_pkeys) { + ret = -EINVAL; + goto out; + } + } + + if (attr_mask & IB_QP_QKEY) + qp->qkey = attr->qkey; + + if (cur_state == next_state && cur_state == IB_QPS_RESET) { + ret = 0; + goto out; + } + + qp->state = next_state; + memset(cmd, 0, sizeof(*cmd)); + cmd->hdr.cmd = PVRDMA_CMD_MODIFY_QP; + cmd->qp_handle = qp->qp_handle; + cmd->attr_mask = ib_qp_attr_mask_to_pvrdma(attr_mask); + cmd->attrs.qp_state = ib_qp_state_to_pvrdma(attr->qp_state); + cmd->attrs.cur_qp_state = + ib_qp_state_to_pvrdma(attr->cur_qp_state); + cmd->attrs.path_mtu = ib_mtu_to_pvrdma(attr->path_mtu); + cmd->attrs.path_mig_state = + ib_mig_state_to_pvrdma(attr->path_mig_state); + cmd->attrs.qkey = attr->qkey; + cmd->attrs.rq_psn = attr->rq_psn; + cmd->attrs.sq_psn = attr->sq_psn; + cmd->attrs.dest_qp_num = attr->dest_qp_num; + cmd->attrs.qp_access_flags = + ib_access_flags_to_pvrdma(attr->qp_access_flags); + cmd->attrs.pkey_index = attr->pkey_index; + cmd->attrs.alt_pkey_index = attr->alt_pkey_index; + cmd->attrs.en_sqd_async_notify = attr->en_sqd_async_notify; + cmd->attrs.sq_draining = attr->sq_draining; + cmd->attrs.max_rd_atomic = attr->max_rd_atomic; + cmd->attrs.max_dest_rd_atomic = attr->max_dest_rd_atomic; + cmd->attrs.min_rnr_timer = attr->min_rnr_timer; + cmd->attrs.port_num = attr->port_num; + cmd->attrs.timeout = attr->timeout; + cmd->attrs.retry_cnt = attr->retry_cnt; + cmd->attrs.rnr_retry = attr->rnr_retry; + cmd->attrs.alt_port_num = attr->alt_port_num; + cmd->attrs.alt_timeout = attr->alt_timeout; + ib_qp_cap_to_pvrdma(&cmd->attrs.cap, &attr->cap); + ib_ah_attr_to_pvrdma(&cmd->attrs.ah_attr, &attr->ah_attr); + ib_ah_attr_to_pvrdma(&cmd->attrs.alt_ah_attr, &attr->alt_ah_attr); + + ret = pvrdma_cmd_post(dev, &req, &rsp, PVRDMA_CMD_MODIFY_QP_RESP); + if (ret < 0) { + dev_warn(&dev->pdev->dev, + "could not modify queuepair, error: %d\n", ret); + } else if (rsp.hdr.err > 0) { + dev_warn(&dev->pdev->dev, + "cannot modify queuepair, error: %d\n", rsp.hdr.err); + ret = -EINVAL; + } + + if (ret == 0 && next_state == IB_QPS_RESET) + pvrdma_reset_qp(qp); + +out: + mutex_unlock(&qp->mutex); + + return ret; +} + +static inline void *get_sq_wqe(struct pvrdma_qp *qp, int n) +{ + return pvrdma_page_dir_get_ptr(&qp->pdir, + qp->sq.offset + n * qp->sq.wqe_size); +} + +static inline void *get_rq_wqe(struct pvrdma_qp *qp, int n) +{ + return pvrdma_page_dir_get_ptr(&qp->pdir, + qp->rq.offset + n * qp->rq.wqe_size); +} + +static int set_reg_seg(struct pvrdma_sq_wqe_hdr *wqe_hdr, struct ib_reg_wr *wr) +{ + struct pvrdma_user_mr *mr = to_vmr(wr->mr); + + wqe_hdr->wr.fast_reg.iova_start = mr->ibmr.iova; + wqe_hdr->wr.fast_reg.pl_pdir_dma = mr->pdir.dir_dma; + wqe_hdr->wr.fast_reg.page_shift = mr->page_shift; + wqe_hdr->wr.fast_reg.page_list_len = mr->npages; + wqe_hdr->wr.fast_reg.length = mr->ibmr.length; + wqe_hdr->wr.fast_reg.access_flags = wr->access; + wqe_hdr->wr.fast_reg.rkey = wr->key; + + return pvrdma_page_dir_insert_page_list(&mr->pdir, mr->pages, + mr->npages); +} + +/** + * pvrdma_post_send - post send work request entries on a QP + * @ibqp: the QP + * @wr: work request list to post + * @bad_wr: the first bad WR returned + * + * @return: 0 on success, otherwise errno returned. + */ +int pvrdma_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, + struct ib_send_wr **bad_wr) +{ + struct pvrdma_qp *qp = to_vqp(ibqp); + struct pvrdma_dev *dev = to_vdev(ibqp->device); + unsigned long flags; + struct pvrdma_sq_wqe_hdr *wqe_hdr; + struct pvrdma_sge *sge; + int i, index; + int nreq; + int ret; + + /* + * In states lower than RTS, we can fail immediately. In other states, + * just post and let the device figure it out. + */ + if (qp->state < IB_QPS_RTS) { + *bad_wr = wr; + return -EINVAL; + } + + spin_lock_irqsave(&qp->sq.lock, flags); + + index = pvrdma_idx(&qp->sq.ring->prod_tail, qp->sq.wqe_cnt); + for (nreq = 0; wr; nreq++, wr = wr->next) { + unsigned int tail; + + if (unlikely(!pvrdma_idx_ring_has_space( + qp->sq.ring, qp->sq.wqe_cnt, &tail))) { + dev_warn_ratelimited(&dev->pdev->dev, + "send queue is full\n"); + *bad_wr = wr; + ret = -ENOMEM; + goto out; + } + + if (unlikely(wr->num_sge > qp->sq.max_sg || wr->num_sge < 0)) { + dev_warn_ratelimited(&dev->pdev->dev, + "send SGE overflow\n"); + *bad_wr = wr; + ret = -EINVAL; + goto out; + } + + if (unlikely(wr->opcode < 0)) { + dev_warn_ratelimited(&dev->pdev->dev, + "invalid send opcode\n"); + *bad_wr = wr; + ret = -EINVAL; + goto out; + } + + /* + * Only support UD, RC. + * Need to check opcode table for thorough checking. + * opcode _UD _UC _RC + * _SEND x x x + * _SEND_WITH_IMM x x x + * _RDMA_WRITE x x + * _RDMA_WRITE_WITH_IMM x x + * _LOCAL_INV x x + * _SEND_WITH_INV x x + * _RDMA_READ x + * _ATOMIC_CMP_AND_SWP x + * _ATOMIC_FETCH_AND_ADD x + * _MASK_ATOMIC_CMP_AND_SWP x + * _MASK_ATOMIC_FETCH_AND_ADD x + * _REG_MR x + * + */ + if (qp->ibqp.qp_type != IB_QPT_UD && + qp->ibqp.qp_type != IB_QPT_RC && + wr->opcode != IB_WR_SEND) { + dev_warn_ratelimited(&dev->pdev->dev, + "unsupported queuepair type\n"); + *bad_wr = wr; + ret = -EINVAL; + goto out; + } else if (qp->ibqp.qp_type == IB_QPT_UD || + qp->ibqp.qp_type == IB_QPT_GSI) { + if (wr->opcode != IB_WR_SEND && + wr->opcode != IB_WR_SEND_WITH_IMM) { + dev_warn_ratelimited(&dev->pdev->dev, + "invalid send opcode\n"); + *bad_wr = wr; + ret = -EINVAL; + goto out; + } + } + + wqe_hdr = (struct pvrdma_sq_wqe_hdr *)get_sq_wqe(qp, index); + memset(wqe_hdr, 0, sizeof(*wqe_hdr)); + wqe_hdr->wr_id = wr->wr_id; + wqe_hdr->num_sge = wr->num_sge; + wqe_hdr->opcode = ib_wr_opcode_to_pvrdma(wr->opcode); + wqe_hdr->send_flags = ib_send_flags_to_pvrdma(wr->send_flags); + if (wr->opcode == IB_WR_SEND_WITH_IMM || + wr->opcode == IB_WR_RDMA_WRITE_WITH_IMM) + wqe_hdr->ex.imm_data = wr->ex.imm_data; + + switch (qp->ibqp.qp_type) { + case IB_QPT_GSI: + case IB_QPT_UD: + if (unlikely(!ud_wr(wr)->ah)) { + dev_warn_ratelimited(&dev->pdev->dev, + "invalid address handle\n"); + *bad_wr = wr; + ret = -EINVAL; + goto out; + } + + /* + * Use qkey from qp context if high order bit set, + * otherwise from work request. + */ + wqe_hdr->wr.ud.remote_qpn = ud_wr(wr)->remote_qpn; + wqe_hdr->wr.ud.remote_qkey = + ud_wr(wr)->remote_qkey & 0x80000000 ? + qp->qkey : ud_wr(wr)->remote_qkey; + wqe_hdr->wr.ud.av = to_vah(ud_wr(wr)->ah)->av; + + break; + case IB_QPT_RC: + switch (wr->opcode) { + case IB_WR_RDMA_READ: + case IB_WR_RDMA_WRITE: + case IB_WR_RDMA_WRITE_WITH_IMM: + wqe_hdr->wr.rdma.remote_addr = + rdma_wr(wr)->remote_addr; + wqe_hdr->wr.rdma.rkey = rdma_wr(wr)->rkey; + break; + case IB_WR_LOCAL_INV: + case IB_WR_SEND_WITH_INV: + wqe_hdr->ex.invalidate_rkey = + wr->ex.invalidate_rkey; + break; + case IB_WR_ATOMIC_CMP_AND_SWP: + case IB_WR_ATOMIC_FETCH_AND_ADD: + wqe_hdr->wr.atomic.remote_addr = + atomic_wr(wr)->remote_addr; + wqe_hdr->wr.atomic.rkey = atomic_wr(wr)->rkey; + wqe_hdr->wr.atomic.compare_add = + atomic_wr(wr)->compare_add; + if (wr->opcode == IB_WR_ATOMIC_CMP_AND_SWP) + wqe_hdr->wr.atomic.swap = + atomic_wr(wr)->swap; + break; + case IB_WR_REG_MR: + ret = set_reg_seg(wqe_hdr, reg_wr(wr)); + if (ret < 0) { + dev_warn_ratelimited(&dev->pdev->dev, + "Failed to set fast register work request\n"); + *bad_wr = wr; + goto out; + } + break; + default: + break; + } + + break; + default: + dev_warn_ratelimited(&dev->pdev->dev, + "invalid queuepair type\n"); + ret = -EINVAL; + *bad_wr = wr; + goto out; + } + + sge = (struct pvrdma_sge *)(wqe_hdr + 1); + for (i = 0; i < wr->num_sge; i++) { + /* Need to check wqe_size 0 or max size */ + sge->addr = wr->sg_list[i].addr; + sge->length = wr->sg_list[i].length; + sge->lkey = wr->sg_list[i].lkey; + sge++; + } + + /* Make sure wqe is written before index update */ + smp_wmb(); + + index++; + if (unlikely(index >= qp->sq.wqe_cnt)) + index = 0; + /* Update shared sq ring */ + pvrdma_idx_ring_inc(&qp->sq.ring->prod_tail, + qp->sq.wqe_cnt); + } + + ret = 0; + +out: + spin_unlock_irqrestore(&qp->sq.lock, flags); + + if (!ret) + pvrdma_write_uar_qp(dev, PVRDMA_UAR_QP_SEND | qp->qp_handle); + + return ret; +} + +/** + * pvrdma_post_receive - post receive work request entries on a QP + * @ibqp: the QP + * @wr: the work request list to post + * @bad_wr: the first bad WR returned + * + * @return: 0 on success, otherwise errno returned. + */ +int pvrdma_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr, + struct ib_recv_wr **bad_wr) +{ + struct pvrdma_dev *dev = to_vdev(ibqp->device); + unsigned long flags; + struct pvrdma_qp *qp = to_vqp(ibqp); + struct pvrdma_rq_wqe_hdr *wqe_hdr; + struct pvrdma_sge *sge; + int index, nreq; + int ret = 0; + int i; + + /* + * In the RESET state, we can fail immediately. For other states, + * just post and let the device figure it out. + */ + if (qp->state == IB_QPS_RESET) { + *bad_wr = wr; + return -EINVAL; + } + + spin_lock_irqsave(&qp->rq.lock, flags); + + index = pvrdma_idx(&qp->rq.ring->prod_tail, qp->rq.wqe_cnt); + for (nreq = 0; wr; nreq++, wr = wr->next) { + unsigned int tail; + + if (unlikely(wr->num_sge > qp->rq.max_sg || + wr->num_sge < 0)) { + ret = -EINVAL; + *bad_wr = wr; + dev_warn_ratelimited(&dev->pdev->dev, + "recv SGE overflow\n"); + goto out; + } + + if (unlikely(!pvrdma_idx_ring_has_space( + qp->rq.ring, qp->rq.wqe_cnt, &tail))) { + ret = -ENOMEM; + *bad_wr = wr; + dev_warn_ratelimited(&dev->pdev->dev, + "recv queue full\n"); + goto out; + } + + wqe_hdr = (struct pvrdma_rq_wqe_hdr *)get_rq_wqe(qp, index); + wqe_hdr->wr_id = wr->wr_id; + wqe_hdr->num_sge = wr->num_sge; + wqe_hdr->total_len = 0; + + sge = (struct pvrdma_sge *)(wqe_hdr + 1); + for (i = 0; i < wr->num_sge; i++) { + sge->addr = wr->sg_list[i].addr; + sge->length = wr->sg_list[i].length; + sge->lkey = wr->sg_list[i].lkey; + sge++; + } + + /* Make sure wqe is written before index update */ + smp_wmb(); + + index++; + if (unlikely(index >= qp->rq.wqe_cnt)) + index = 0; + /* Update shared rq ring */ + pvrdma_idx_ring_inc(&qp->rq.ring->prod_tail, + qp->rq.wqe_cnt); + } + + spin_unlock_irqrestore(&qp->rq.lock, flags); + + pvrdma_write_uar_qp(dev, PVRDMA_UAR_QP_RECV | qp->qp_handle); + + return ret; + +out: + spin_unlock_irqrestore(&qp->rq.lock, flags); + + return ret; +} + +/** + * pvrdma_query_qp - query a queue pair's attributes + * @ibqp: the queue pair to query + * @attr: the queue pair's attributes + * @attr_mask: attributes mask + * @init_attr: initial queue pair attributes + * + * @returns 0 on success, otherwise returns an errno. + */ +int pvrdma_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, + int attr_mask, struct ib_qp_init_attr *init_attr) +{ + struct pvrdma_dev *dev = to_vdev(ibqp->device); + struct pvrdma_qp *qp = to_vqp(ibqp); + union pvrdma_cmd_req req; + union pvrdma_cmd_resp rsp; + struct pvrdma_cmd_query_qp *cmd = &req.query_qp; + struct pvrdma_cmd_query_qp_resp *resp = &rsp.query_qp_resp; + int ret = 0; + + mutex_lock(&qp->mutex); + + if (qp->state == IB_QPS_RESET) { + attr->qp_state = IB_QPS_RESET; + goto out; + } + + memset(cmd, 0, sizeof(*cmd)); + cmd->hdr.cmd = PVRDMA_CMD_QUERY_QP; + cmd->qp_handle = qp->qp_handle; + cmd->attr_mask = ib_qp_attr_mask_to_pvrdma(attr_mask); + + ret = pvrdma_cmd_post(dev, &req, &rsp, PVRDMA_CMD_QUERY_QP_RESP); + if (ret < 0) { + dev_warn(&dev->pdev->dev, + "could not query queuepair, error: %d\n", ret); + goto out; + } + + attr->qp_state = pvrdma_qp_state_to_ib(resp->attrs.qp_state); + attr->cur_qp_state = + pvrdma_qp_state_to_ib(resp->attrs.cur_qp_state); + attr->path_mtu = pvrdma_mtu_to_ib(resp->attrs.path_mtu); + attr->path_mig_state = + pvrdma_mig_state_to_ib(resp->attrs.path_mig_state); + attr->qkey = resp->attrs.qkey; + attr->rq_psn = resp->attrs.rq_psn; + attr->sq_psn = resp->attrs.sq_psn; + attr->dest_qp_num = resp->attrs.dest_qp_num; + attr->qp_access_flags = + pvrdma_access_flags_to_ib(resp->attrs.qp_access_flags); + attr->pkey_index = resp->attrs.pkey_index; + attr->alt_pkey_index = resp->attrs.alt_pkey_index; + attr->en_sqd_async_notify = resp->attrs.en_sqd_async_notify; + attr->sq_draining = resp->attrs.sq_draining; + attr->max_rd_atomic = resp->attrs.max_rd_atomic; + attr->max_dest_rd_atomic = resp->attrs.max_dest_rd_atomic; + attr->min_rnr_timer = resp->attrs.min_rnr_timer; + attr->port_num = resp->attrs.port_num; + attr->timeout = resp->attrs.timeout; + attr->retry_cnt = resp->attrs.retry_cnt; + attr->rnr_retry = resp->attrs.rnr_retry; + attr->alt_port_num = resp->attrs.alt_port_num; + attr->alt_timeout = resp->attrs.alt_timeout; + pvrdma_qp_cap_to_ib(&attr->cap, &resp->attrs.cap); + pvrdma_ah_attr_to_ib(&attr->ah_attr, &resp->attrs.ah_attr); + pvrdma_ah_attr_to_ib(&attr->alt_ah_attr, &resp->attrs.alt_ah_attr); + + qp->state = attr->qp_state; + + ret = 0; + +out: + attr->cur_qp_state = attr->qp_state; + + init_attr->event_handler = qp->ibqp.event_handler; + init_attr->qp_context = qp->ibqp.qp_context; + init_attr->send_cq = qp->ibqp.send_cq; + init_attr->recv_cq = qp->ibqp.recv_cq; + init_attr->srq = qp->ibqp.srq; + init_attr->xrcd = NULL; + init_attr->cap = attr->cap; + init_attr->sq_sig_type = 0; + init_attr->qp_type = qp->ibqp.qp_type; + init_attr->create_flags = 0; + init_attr->port_num = qp->port; + + mutex_unlock(&qp->mutex); + return ret; +} diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_ring.h b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_ring.h new file mode 100644 index 000000000000..ed9022a91a1d --- /dev/null +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_ring.h @@ -0,0 +1,131 @@ +/* + * Copyright (c) 2012-2016 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of EITHER the GNU General Public License + * version 2 as published by the Free Software Foundation or the BSD + * 2-Clause License. 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 version 2 for more details at + * http://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html. + * + * You should have received a copy of the GNU General Public License + * along with this program available in the file COPYING in the main + * directory of this source tree. + * + * The BSD 2-Clause License + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE + * COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef __PVRDMA_RING_H__ +#define __PVRDMA_RING_H__ + +#include <linux/types.h> + +#define PVRDMA_INVALID_IDX -1 /* Invalid index. */ + +struct pvrdma_ring { + atomic_t prod_tail; /* Producer tail. */ + atomic_t cons_head; /* Consumer head. */ +}; + +struct pvrdma_ring_state { + struct pvrdma_ring tx; /* Tx ring. */ + struct pvrdma_ring rx; /* Rx ring. */ +}; + +static inline int pvrdma_idx_valid(__u32 idx, __u32 max_elems) +{ + /* Generates fewer instructions than a less-than. */ + return (idx & ~((max_elems << 1) - 1)) == 0; +} + +static inline __s32 pvrdma_idx(atomic_t *var, __u32 max_elems) +{ + const unsigned int idx = atomic_read(var); + + if (pvrdma_idx_valid(idx, max_elems)) + return idx & (max_elems - 1); + return PVRDMA_INVALID_IDX; +} + +static inline void pvrdma_idx_ring_inc(atomic_t *var, __u32 max_elems) +{ + __u32 idx = atomic_read(var) + 1; /* Increment. */ + + idx &= (max_elems << 1) - 1; /* Modulo size, flip gen. */ + atomic_set(var, idx); +} + +static inline __s32 pvrdma_idx_ring_has_space(const struct pvrdma_ring *r, + __u32 max_elems, __u32 *out_tail) +{ + const __u32 tail = atomic_read(&r->prod_tail); + const __u32 head = atomic_read(&r->cons_head); + + if (pvrdma_idx_valid(tail, max_elems) && + pvrdma_idx_valid(head, max_elems)) { + *out_tail = tail & (max_elems - 1); + return tail != (head ^ max_elems); + } + return PVRDMA_INVALID_IDX; +} + +static inline __s32 pvrdma_idx_ring_has_data(const struct pvrdma_ring *r, + __u32 max_elems, __u32 *out_head) +{ + const __u32 tail = atomic_read(&r->prod_tail); + const __u32 head = atomic_read(&r->cons_head); + + if (pvrdma_idx_valid(tail, max_elems) && + pvrdma_idx_valid(head, max_elems)) { + *out_head = head & (max_elems - 1); + return tail != head; + } + return PVRDMA_INVALID_IDX; +} + +static inline bool pvrdma_idx_ring_is_valid_idx(const struct pvrdma_ring *r, + __u32 max_elems, __u32 *idx) +{ + const __u32 tail = atomic_read(&r->prod_tail); + const __u32 head = atomic_read(&r->cons_head); + + if (pvrdma_idx_valid(tail, max_elems) && + pvrdma_idx_valid(head, max_elems) && + pvrdma_idx_valid(*idx, max_elems)) { + if (tail > head && (*idx < tail && *idx >= head)) + return true; + else if (head > tail && (*idx >= head || *idx < tail)) + return true; + } + return false; +} + +#endif /* __PVRDMA_RING_H__ */ diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c new file mode 100644 index 000000000000..fec17c49103b --- /dev/null +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c @@ -0,0 +1,579 @@ +/* + * Copyright (c) 2012-2016 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of EITHER the GNU General Public License + * version 2 as published by the Free Software Foundation or the BSD + * 2-Clause License. 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 version 2 for more details at + * http://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html. + * + * You should have received a copy of the GNU General Public License + * along with this program available in the file COPYING in the main + * directory of this source tree. + * + * The BSD 2-Clause License + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE + * COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <asm/page.h> +#include <linux/inet.h> +#include <linux/io.h> +#include <rdma/ib_addr.h> +#include <rdma/ib_smi.h> +#include <rdma/ib_user_verbs.h> +#include <rdma/vmw_pvrdma-abi.h> + +#include "pvrdma.h" + +/** + * pvrdma_query_device - query device + * @ibdev: the device to query + * @props: the device properties + * @uhw: user data + * + * @return: 0 on success, otherwise negative errno + */ +int pvrdma_query_device(struct ib_device *ibdev, + struct ib_device_attr *props, + struct ib_udata *uhw) +{ + struct pvrdma_dev *dev = to_vdev(ibdev); + + if (uhw->inlen || uhw->outlen) + return -EINVAL; + + memset(props, 0, sizeof(*props)); + + props->fw_ver = dev->dsr->caps.fw_ver; + props->sys_image_guid = dev->dsr->caps.sys_image_guid; + props->max_mr_size = dev->dsr->caps.max_mr_size; + props->page_size_cap = dev->dsr->caps.page_size_cap; + props->vendor_id = dev->dsr->caps.vendor_id; + props->vendor_part_id = dev->pdev->device; + props->hw_ver = dev->dsr->caps.hw_ver; + props->max_qp = dev->dsr->caps.max_qp; + props->max_qp_wr = dev->dsr->caps.max_qp_wr; + props->device_cap_flags = dev->dsr->caps.device_cap_flags; + props->max_sge = dev->dsr->caps.max_sge; + props->max_cq = dev->dsr->caps.max_cq; + props->max_cqe = dev->dsr->caps.max_cqe; + props->max_mr = dev->dsr->caps.max_mr; + props->max_pd = dev->dsr->caps.max_pd; + props->max_qp_rd_atom = dev->dsr->caps.max_qp_rd_atom; + props->max_qp_init_rd_atom = dev->dsr->caps.max_qp_init_rd_atom; + props->atomic_cap = + dev->dsr->caps.atomic_ops & + (PVRDMA_ATOMIC_OP_COMP_SWAP | PVRDMA_ATOMIC_OP_FETCH_ADD) ? + IB_ATOMIC_HCA : IB_ATOMIC_NONE; + props->masked_atomic_cap = props->atomic_cap; + props->max_ah = dev->dsr->caps.max_ah; + props->max_pkeys = dev->dsr->caps.max_pkeys; + props->local_ca_ack_delay = dev->dsr->caps.local_ca_ack_delay; + if ((dev->dsr->caps.bmme_flags & PVRDMA_BMME_FLAG_LOCAL_INV) && + (dev->dsr->caps.bmme_flags & PVRDMA_BMME_FLAG_REMOTE_INV) && + (dev->dsr->caps.bmme_flags & PVRDMA_BMME_FLAG_FAST_REG_WR)) { + props->device_cap_flags |= IB_DEVICE_MEM_MGT_EXTENSIONS; + } + + return 0; +} + +/** + * pvrdma_query_port - query device port attributes + * @ibdev: the device to query + * @port: the port number + * @props: the device properties + * + * @return: 0 on success, otherwise negative errno + */ +int pvrdma_query_port(struct ib_device *ibdev, u8 port, + struct ib_port_attr *props) +{ + struct pvrdma_dev *dev = to_vdev(ibdev); + union pvrdma_cmd_req req; + union pvrdma_cmd_resp rsp; + struct pvrdma_cmd_query_port *cmd = &req.query_port; + struct pvrdma_cmd_query_port_resp *resp = &rsp.query_port_resp; + int err; + + memset(cmd, 0, sizeof(*cmd)); + cmd->hdr.cmd = PVRDMA_CMD_QUERY_PORT; + cmd->port_num = port; + + err = pvrdma_cmd_post(dev, &req, &rsp, PVRDMA_CMD_QUERY_PORT_RESP); + if (err < 0) { + dev_warn(&dev->pdev->dev, + "could not query port, error: %d\n", err); + return err; + } + + /* props being zeroed by the caller, avoid zeroing it here */ + + props->state = pvrdma_port_state_to_ib(resp->attrs.state); + props->max_mtu = pvrdma_mtu_to_ib(resp->attrs.max_mtu); + props->active_mtu = pvrdma_mtu_to_ib(resp->attrs.active_mtu); + props->gid_tbl_len = resp->attrs.gid_tbl_len; + props->port_cap_flags = + pvrdma_port_cap_flags_to_ib(resp->attrs.port_cap_flags); + props->max_msg_sz = resp->attrs.max_msg_sz; + props->bad_pkey_cntr = resp->attrs.bad_pkey_cntr; + props->qkey_viol_cntr = resp->attrs.qkey_viol_cntr; + props->pkey_tbl_len = resp->attrs.pkey_tbl_len; + props->lid = resp->attrs.lid; + props->sm_lid = resp->attrs.sm_lid; + props->lmc = resp->attrs.lmc; + props->max_vl_num = resp->attrs.max_vl_num; + props->sm_sl = resp->attrs.sm_sl; + props->subnet_timeout = resp->attrs.subnet_timeout; + props->init_type_reply = resp->attrs.init_type_reply; + props->active_width = pvrdma_port_width_to_ib(resp->attrs.active_width); + props->active_speed = pvrdma_port_speed_to_ib(resp->attrs.active_speed); + props->phys_state = resp->attrs.phys_state; + + return 0; +} + +/** + * pvrdma_query_gid - query device gid + * @ibdev: the device to query + * @port: the port number + * @index: the index + * @gid: the device gid value + * + * @return: 0 on success, otherwise negative errno + */ +int pvrdma_query_gid(struct ib_device *ibdev, u8 port, int index, + union ib_gid *gid) +{ + struct pvrdma_dev *dev = to_vdev(ibdev); + + if (index >= dev->dsr->caps.gid_tbl_len) + return -EINVAL; + + memcpy(gid, &dev->sgid_tbl[index], sizeof(union ib_gid)); + + return 0; +} + +/** + * pvrdma_query_pkey - query device port's P_Key table + * @ibdev: the device to query + * @port: the port number + * @index: the index + * @pkey: the device P_Key value + * + * @return: 0 on success, otherwise negative errno + */ +int pvrdma_query_pkey(struct ib_device *ibdev, u8 port, u16 index, + u16 *pkey) +{ + int err = 0; + union pvrdma_cmd_req req; + union pvrdma_cmd_resp rsp; + struct pvrdma_cmd_query_pkey *cmd = &req.query_pkey; + + memset(cmd, 0, sizeof(*cmd)); + cmd->hdr.cmd = PVRDMA_CMD_QUERY_PKEY; + cmd->port_num = port; + cmd->index = index; + + err = pvrdma_cmd_post(to_vdev(ibdev), &req, &rsp, + PVRDMA_CMD_QUERY_PKEY_RESP); + if (err < 0) { + dev_warn(&to_vdev(ibdev)->pdev->dev, + "could not query pkey, error: %d\n", err); + return err; + } + + *pkey = rsp.query_pkey_resp.pkey; + + return 0; +} + +enum rdma_link_layer pvrdma_port_link_layer(struct ib_device *ibdev, + u8 port) +{ + return IB_LINK_LAYER_ETHERNET; +} + +int pvrdma_modify_device(struct ib_device *ibdev, int mask, + struct ib_device_modify *props) +{ + unsigned long flags; + + if (mask & ~(IB_DEVICE_MODIFY_SYS_IMAGE_GUID | + IB_DEVICE_MODIFY_NODE_DESC)) { + dev_warn(&to_vdev(ibdev)->pdev->dev, + "unsupported device modify mask %#x\n", mask); + return -EOPNOTSUPP; + } + + if (mask & IB_DEVICE_MODIFY_NODE_DESC) { + spin_lock_irqsave(&to_vdev(ibdev)->desc_lock, flags); + memcpy(ibdev->node_desc, props->node_desc, 64); + spin_unlock_irqrestore(&to_vdev(ibdev)->desc_lock, flags); + } + + if (mask & IB_DEVICE_MODIFY_SYS_IMAGE_GUID) { + mutex_lock(&to_vdev(ibdev)->port_mutex); + to_vdev(ibdev)->sys_image_guid = + cpu_to_be64(props->sys_image_guid); + mutex_unlock(&to_vdev(ibdev)->port_mutex); + } + + return 0; +} + +/** + * pvrdma_modify_port - modify device port attributes + * @ibdev: the device to modify + * @port: the port number + * @mask: attributes to modify + * @props: the device properties + * + * @return: 0 on success, otherwise negative errno + */ +int pvrdma_modify_port(struct ib_device *ibdev, u8 port, int mask, + struct ib_port_modify *props) +{ + struct ib_port_attr attr; + struct pvrdma_dev *vdev = to_vdev(ibdev); + int ret; + + if (mask & ~IB_PORT_SHUTDOWN) { + dev_warn(&vdev->pdev->dev, + "unsupported port modify mask %#x\n", mask); + return -EOPNOTSUPP; + } + + mutex_lock(&vdev->port_mutex); + ret = ib_query_port(ibdev, port, &attr); + if (ret) + goto out; + + vdev->port_cap_mask |= props->set_port_cap_mask; + vdev->port_cap_mask &= ~props->clr_port_cap_mask; + + if (mask & IB_PORT_SHUTDOWN) + vdev->ib_active = false; + +out: + mutex_unlock(&vdev->port_mutex); + return ret; +} + +/** + * pvrdma_alloc_ucontext - allocate ucontext + * @ibdev: the IB device + * @udata: user data + * + * @return: the ib_ucontext pointer on success, otherwise errno. + */ +struct ib_ucontext *pvrdma_alloc_ucontext(struct ib_device *ibdev, + struct ib_udata *udata) +{ + struct pvrdma_dev *vdev = to_vdev(ibdev); + struct pvrdma_ucontext *context; + union pvrdma_cmd_req req; + union pvrdma_cmd_resp rsp; + struct pvrdma_cmd_create_uc *cmd = &req.create_uc; + struct pvrdma_cmd_create_uc_resp *resp = &rsp.create_uc_resp; + struct pvrdma_alloc_ucontext_resp uresp = {0}; + int ret; + void *ptr; + + if (!vdev->ib_active) + return ERR_PTR(-EAGAIN); + + context = kmalloc(sizeof(*context), GFP_KERNEL); + if (!context) + return ERR_PTR(-ENOMEM); + + context->dev = vdev; + ret = pvrdma_uar_alloc(vdev, &context->uar); + if (ret) { + kfree(context); + return ERR_PTR(-ENOMEM); + } + + /* get ctx_handle from host */ + memset(cmd, 0, sizeof(*cmd)); + cmd->pfn = context->uar.pfn; + cmd->hdr.cmd = PVRDMA_CMD_CREATE_UC; + ret = pvrdma_cmd_post(vdev, &req, &rsp, PVRDMA_CMD_CREATE_UC_RESP); + if (ret < 0) { + dev_warn(&vdev->pdev->dev, + "could not create ucontext, error: %d\n", ret); + ptr = ERR_PTR(ret); + goto err; + } + + context->ctx_handle = resp->ctx_handle; + + /* copy back to user */ + uresp.qp_tab_size = vdev->dsr->caps.max_qp; + ret = ib_copy_to_udata(udata, &uresp, sizeof(uresp)); + if (ret) { + pvrdma_uar_free(vdev, &context->uar); + context->ibucontext.device = ibdev; + pvrdma_dealloc_ucontext(&context->ibucontext); + return ERR_PTR(-EFAULT); + } + + return &context->ibucontext; + +err: + pvrdma_uar_free(vdev, &context->uar); + kfree(context); + return ptr; +} + +/** + * pvrdma_dealloc_ucontext - deallocate ucontext + * @ibcontext: the ucontext + * + * @return: 0 on success, otherwise errno. + */ +int pvrdma_dealloc_ucontext(struct ib_ucontext *ibcontext) +{ + struct pvrdma_ucontext *context = to_vucontext(ibcontext); + union pvrdma_cmd_req req; + struct pvrdma_cmd_destroy_uc *cmd = &req.destroy_uc; + int ret; + + memset(cmd, 0, sizeof(*cmd)); + cmd->hdr.cmd = PVRDMA_CMD_DESTROY_UC; + cmd->ctx_handle = context->ctx_handle; + + ret = pvrdma_cmd_post(context->dev, &req, NULL, 0); + if (ret < 0) + dev_warn(&context->dev->pdev->dev, + "destroy ucontext failed, error: %d\n", ret); + + /* Free the UAR even if the device command failed */ + pvrdma_uar_free(to_vdev(ibcontext->device), &context->uar); + kfree(context); + + return ret; +} + +/** + * pvrdma_mmap - create mmap region + * @ibcontext: the user context + * @vma: the VMA + * + * @return: 0 on success, otherwise errno. + */ +int pvrdma_mmap(struct ib_ucontext *ibcontext, struct vm_area_struct *vma) +{ + struct pvrdma_ucontext *context = to_vucontext(ibcontext); + unsigned long start = vma->vm_start; + unsigned long size = vma->vm_end - vma->vm_start; + unsigned long offset = vma->vm_pgoff << PAGE_SHIFT; + + dev_dbg(&context->dev->pdev->dev, "create mmap region\n"); + + if ((size != PAGE_SIZE) || (offset & ~PAGE_MASK)) { + dev_warn(&context->dev->pdev->dev, + "invalid params for mmap region\n"); + return -EINVAL; + } + + /* Map UAR to kernel space, VM_LOCKED? */ + vma->vm_flags |= VM_DONTCOPY | VM_DONTEXPAND; + vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); + if (io_remap_pfn_range(vma, start, context->uar.pfn, size, + vma->vm_page_prot)) + return -EAGAIN; + + return 0; +} + +/** + * pvrdma_alloc_pd - allocate protection domain + * @ibdev: the IB device + * @context: user context + * @udata: user data + * + * @return: the ib_pd protection domain pointer on success, otherwise errno. + */ +struct ib_pd *pvrdma_alloc_pd(struct ib_device *ibdev, + struct ib_ucontext *context, + struct ib_udata *udata) +{ + struct pvrdma_pd *pd; + struct pvrdma_dev *dev = to_vdev(ibdev); + union pvrdma_cmd_req req; + union pvrdma_cmd_resp rsp; + struct pvrdma_cmd_create_pd *cmd = &req.create_pd; + struct pvrdma_cmd_create_pd_resp *resp = &rsp.create_pd_resp; + int ret; + void *ptr; + + /* Check allowed max pds */ + if (!atomic_add_unless(&dev->num_pds, 1, dev->dsr->caps.max_pd)) + return ERR_PTR(-ENOMEM); + + pd = kmalloc(sizeof(*pd), GFP_KERNEL); + if (!pd) { + ptr = ERR_PTR(-ENOMEM); + goto err; + } + + memset(cmd, 0, sizeof(*cmd)); + cmd->hdr.cmd = PVRDMA_CMD_CREATE_PD; + cmd->ctx_handle = (context) ? to_vucontext(context)->ctx_handle : 0; + ret = pvrdma_cmd_post(dev, &req, &rsp, PVRDMA_CMD_CREATE_PD_RESP); + if (ret < 0) { + dev_warn(&dev->pdev->dev, + "failed to allocate protection domain, error: %d\n", + ret); + ptr = ERR_PTR(ret); + goto freepd; + } + + pd->privileged = !context; + pd->pd_handle = resp->pd_handle; + pd->pdn = resp->pd_handle; + + if (context) { + if (ib_copy_to_udata(udata, &pd->pdn, sizeof(__u32))) { + dev_warn(&dev->pdev->dev, + "failed to copy back protection domain\n"); + pvrdma_dealloc_pd(&pd->ibpd); + return ERR_PTR(-EFAULT); + } + } + + /* u32 pd handle */ + return &pd->ibpd; + +freepd: + kfree(pd); +err: + atomic_dec(&dev->num_pds); + return ptr; +} + +/** + * pvrdma_dealloc_pd - deallocate protection domain + * @pd: the protection domain to be released + * + * @return: 0 on success, otherwise errno. + */ +int pvrdma_dealloc_pd(struct ib_pd *pd) +{ + struct pvrdma_dev *dev = to_vdev(pd->device); + union pvrdma_cmd_req req; + struct pvrdma_cmd_destroy_pd *cmd = &req.destroy_pd; + int ret; + + memset(cmd, 0, sizeof(*cmd)); + cmd->hdr.cmd = PVRDMA_CMD_DESTROY_PD; + cmd->pd_handle = to_vpd(pd)->pd_handle; + + ret = pvrdma_cmd_post(dev, &req, NULL, 0); + if (ret) + dev_warn(&dev->pdev->dev, + "could not dealloc protection domain, error: %d\n", + ret); + + kfree(to_vpd(pd)); + atomic_dec(&dev->num_pds); + + return 0; +} + +/** + * pvrdma_create_ah - create an address handle + * @pd: the protection domain + * @ah_attr: the attributes of the AH + * @udata: user data blob + * + * @return: the ib_ah pointer on success, otherwise errno. + */ +struct ib_ah *pvrdma_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, + struct ib_udata *udata) +{ + struct pvrdma_dev *dev = to_vdev(pd->device); + struct pvrdma_ah *ah; + enum rdma_link_layer ll; + + if (!(ah_attr->ah_flags & IB_AH_GRH)) + return ERR_PTR(-EINVAL); + + ll = rdma_port_get_link_layer(pd->device, ah_attr->port_num); + + if (ll != IB_LINK_LAYER_ETHERNET || + rdma_is_multicast_addr((struct in6_addr *)ah_attr->grh.dgid.raw)) + return ERR_PTR(-EINVAL); + + if (!atomic_add_unless(&dev->num_ahs, 1, dev->dsr->caps.max_ah)) + return ERR_PTR(-ENOMEM); + + ah = kzalloc(sizeof(*ah), GFP_KERNEL); + if (!ah) { + atomic_dec(&dev->num_ahs); + return ERR_PTR(-ENOMEM); + } + + ah->av.port_pd = to_vpd(pd)->pd_handle | (ah_attr->port_num << 24); + ah->av.src_path_bits = ah_attr->src_path_bits; + ah->av.src_path_bits |= 0x80; + ah->av.gid_index = ah_attr->grh.sgid_index; + ah->av.hop_limit = ah_attr->grh.hop_limit; + ah->av.sl_tclass_flowlabel = (ah_attr->grh.traffic_class << 20) | + ah_attr->grh.flow_label; + memcpy(ah->av.dgid, ah_attr->grh.dgid.raw, 16); + memcpy(ah->av.dmac, ah_attr->dmac, 6); + + ah->ibah.device = pd->device; + ah->ibah.pd = pd; + ah->ibah.uobject = NULL; + + return &ah->ibah; +} + +/** + * pvrdma_destroy_ah - destroy an address handle + * @ah: the address handle to destroyed + * + * @return: 0 on success. + */ +int pvrdma_destroy_ah(struct ib_ah *ah) +{ + struct pvrdma_dev *dev = to_vdev(ah->device); + + kfree(to_vah(ah)); + atomic_dec(&dev->num_ahs); + + return 0; +} diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.h b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.h new file mode 100644 index 000000000000..bfbe96b56255 --- /dev/null +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.h @@ -0,0 +1,436 @@ +/* + * Copyright (c) 2012-2016 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of EITHER the GNU General Public License + * version 2 as published by the Free Software Foundation or the BSD + * 2-Clause License. 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 version 2 for more details at + * http://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html. + * + * You should have received a copy of the GNU General Public License + * along with this program available in the file COPYING in the main + * directory of this source tree. + * + * The BSD 2-Clause License + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE + * COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, + * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef __PVRDMA_VERBS_H__ +#define __PVRDMA_VERBS_H__ + +#include <linux/types.h> + +union pvrdma_gid { + u8 raw[16]; + struct { + __be64 subnet_prefix; + __be64 interface_id; + } global; +}; + +enum pvrdma_link_layer { + PVRDMA_LINK_LAYER_UNSPECIFIED, + PVRDMA_LINK_LAYER_INFINIBAND, + PVRDMA_LINK_LAYER_ETHERNET, +}; + +enum pvrdma_mtu { + PVRDMA_MTU_256 = 1, + PVRDMA_MTU_512 = 2, + PVRDMA_MTU_1024 = 3, + PVRDMA_MTU_2048 = 4, + PVRDMA_MTU_4096 = 5, +}; + +static inline int pvrdma_mtu_enum_to_int(enum pvrdma_mtu mtu) +{ + switch (mtu) { + case PVRDMA_MTU_256: return 256; + case PVRDMA_MTU_512: return 512; + case PVRDMA_MTU_1024: return 1024; + case PVRDMA_MTU_2048: return 2048; + case PVRDMA_MTU_4096: return 4096; + default: return -1; + } +} + +static inline enum pvrdma_mtu pvrdma_mtu_int_to_enum(int mtu) +{ + switch (mtu) { + case 256: return PVRDMA_MTU_256; + case 512: return PVRDMA_MTU_512; + case 1024: return PVRDMA_MTU_1024; + case 2048: return PVRDMA_MTU_2048; + case 4096: + default: return PVRDMA_MTU_4096; + } +} + +enum pvrdma_port_state { + PVRDMA_PORT_NOP = 0, + PVRDMA_PORT_DOWN = 1, + PVRDMA_PORT_INIT = 2, + PVRDMA_PORT_ARMED = 3, + PVRDMA_PORT_ACTIVE = 4, + PVRDMA_PORT_ACTIVE_DEFER = 5, +}; + +enum pvrdma_port_cap_flags { + PVRDMA_PORT_SM = 1 << 1, + PVRDMA_PORT_NOTICE_SUP = 1 << 2, + PVRDMA_PORT_TRAP_SUP = 1 << 3, + PVRDMA_PORT_OPT_IPD_SUP = 1 << 4, + PVRDMA_PORT_AUTO_MIGR_SUP = 1 << 5, + PVRDMA_PORT_SL_MAP_SUP = 1 << 6, + PVRDMA_PORT_MKEY_NVRAM = 1 << 7, + PVRDMA_PORT_PKEY_NVRAM = 1 << 8, + PVRDMA_PORT_LED_INFO_SUP = 1 << 9, + PVRDMA_PORT_SM_DISABLED = 1 << 10, + PVRDMA_PORT_SYS_IMAGE_GUID_SUP = 1 << 11, + PVRDMA_PORT_PKEY_SW_EXT_PORT_TRAP_SUP = 1 << 12, + PVRDMA_PORT_EXTENDED_SPEEDS_SUP = 1 << 14, + PVRDMA_PORT_CM_SUP = 1 << 16, + PVRDMA_PORT_SNMP_TUNNEL_SUP = 1 << 17, + PVRDMA_PORT_REINIT_SUP = 1 << 18, + PVRDMA_PORT_DEVICE_MGMT_SUP = 1 << 19, + PVRDMA_PORT_VENDOR_CLASS_SUP = 1 << 20, + PVRDMA_PORT_DR_NOTICE_SUP = 1 << 21, + PVRDMA_PORT_CAP_MASK_NOTICE_SUP = 1 << 22, + PVRDMA_PORT_BOOT_MGMT_SUP = 1 << 23, + PVRDMA_PORT_LINK_LATENCY_SUP = 1 << 24, + PVRDMA_PORT_CLIENT_REG_SUP = 1 << 25, + PVRDMA_PORT_IP_BASED_GIDS = 1 << 26, + PVRDMA_PORT_CAP_FLAGS_MAX = PVRDMA_PORT_IP_BASED_GIDS, +}; + +enum pvrdma_port_width { + PVRDMA_WIDTH_1X = 1, + PVRDMA_WIDTH_4X = 2, + PVRDMA_WIDTH_8X = 4, + PVRDMA_WIDTH_12X = 8, +}; + +static inline int pvrdma_width_enum_to_int(enum pvrdma_port_width width) +{ + switch (width) { + case PVRDMA_WIDTH_1X: return 1; + case PVRDMA_WIDTH_4X: return 4; + case PVRDMA_WIDTH_8X: return 8; + case PVRDMA_WIDTH_12X: return 12; + default: return -1; + } +} + +enum pvrdma_port_speed { + PVRDMA_SPEED_SDR = 1, + PVRDMA_SPEED_DDR = 2, + PVRDMA_SPEED_QDR = 4, + PVRDMA_SPEED_FDR10 = 8, + PVRDMA_SPEED_FDR = 16, + PVRDMA_SPEED_EDR = 32, +}; + +struct pvrdma_port_attr { + enum pvrdma_port_state state; + enum pvrdma_mtu max_mtu; + enum pvrdma_mtu active_mtu; + u32 gid_tbl_len; + u32 port_cap_flags; + u32 max_msg_sz; + u32 bad_pkey_cntr; + u32 qkey_viol_cntr; + u16 pkey_tbl_len; + u16 lid; + u16 sm_lid; + u8 lmc; + u8 max_vl_num; + u8 sm_sl; + u8 subnet_timeout; + u8 init_type_reply; + u8 active_width; + u8 active_speed; + u8 phys_state; + u8 reserved[2]; +}; + +struct pvrdma_global_route { + union pvrdma_gid dgid; + u32 flow_label; + u8 sgid_index; + u8 hop_limit; + u8 traffic_class; + u8 reserved; +}; + +struct pvrdma_grh { + __be32 version_tclass_flow; + __be16 paylen; + u8 next_hdr; + u8 hop_limit; + union pvrdma_gid sgid; + union pvrdma_gid dgid; +}; + +enum pvrdma_ah_flags { + PVRDMA_AH_GRH = 1, +}; + +enum pvrdma_rate { + PVRDMA_RATE_PORT_CURRENT = 0, + PVRDMA_RATE_2_5_GBPS = 2, + PVRDMA_RATE_5_GBPS = 5, + PVRDMA_RATE_10_GBPS = 3, + PVRDMA_RATE_20_GBPS = 6, + PVRDMA_RATE_30_GBPS = 4, + PVRDMA_RATE_40_GBPS = 7, + PVRDMA_RATE_60_GBPS = 8, + PVRDMA_RATE_80_GBPS = 9, + PVRDMA_RATE_120_GBPS = 10, + PVRDMA_RATE_14_GBPS = 11, + PVRDMA_RATE_56_GBPS = 12, + PVRDMA_RATE_112_GBPS = 13, + PVRDMA_RATE_168_GBPS = 14, + PVRDMA_RATE_25_GBPS = 15, + PVRDMA_RATE_100_GBPS = 16, + PVRDMA_RATE_200_GBPS = 17, + PVRDMA_RATE_300_GBPS = 18, +}; + +struct pvrdma_ah_attr { + struct pvrdma_global_route grh; + u16 dlid; + u16 vlan_id; + u8 sl; + u8 src_path_bits; + u8 static_rate; + u8 ah_flags; + u8 port_num; + u8 dmac[6]; + u8 reserved; +}; + +enum pvrdma_cq_notify_flags { + PVRDMA_CQ_SOLICITED = 1 << 0, + PVRDMA_CQ_NEXT_COMP = 1 << 1, + PVRDMA_CQ_SOLICITED_MASK = PVRDMA_CQ_SOLICITED | + PVRDMA_CQ_NEXT_COMP, + PVRDMA_CQ_REPORT_MISSED_EVENTS = 1 << 2, +}; + +struct pvrdma_qp_cap { + u32 max_send_wr; + u32 max_recv_wr; + u32 max_send_sge; + u32 max_recv_sge; + u32 max_inline_data; + u32 reserved; +}; + +enum pvrdma_sig_type { + PVRDMA_SIGNAL_ALL_WR, + PVRDMA_SIGNAL_REQ_WR, +}; + +enum pvrdma_qp_type { + PVRDMA_QPT_SMI, + PVRDMA_QPT_GSI, + PVRDMA_QPT_RC, + PVRDMA_QPT_UC, + PVRDMA_QPT_UD, + PVRDMA_QPT_RAW_IPV6, + PVRDMA_QPT_RAW_ETHERTYPE, + PVRDMA_QPT_RAW_PACKET = 8, + PVRDMA_QPT_XRC_INI = 9, + PVRDMA_QPT_XRC_TGT, + PVRDMA_QPT_MAX, +}; + +enum pvrdma_qp_create_flags { + PVRDMA_QP_CREATE_IPOPVRDMA_UD_LSO = 1 << 0, + PVRDMA_QP_CREATE_BLOCK_MULTICAST_LOOPBACK = 1 << 1, +}; + +enum pvrdma_qp_attr_mask { + PVRDMA_QP_STATE = 1 << 0, + PVRDMA_QP_CUR_STATE = 1 << 1, + PVRDMA_QP_EN_SQD_ASYNC_NOTIFY = 1 << 2, + PVRDMA_QP_ACCESS_FLAGS = 1 << 3, + PVRDMA_QP_PKEY_INDEX = 1 << 4, + PVRDMA_QP_PORT = 1 << 5, + PVRDMA_QP_QKEY = 1 << 6, + PVRDMA_QP_AV = 1 << 7, + PVRDMA_QP_PATH_MTU = 1 << 8, + PVRDMA_QP_TIMEOUT = 1 << 9, + PVRDMA_QP_RETRY_CNT = 1 << 10, + PVRDMA_QP_RNR_RETRY = 1 << 11, + PVRDMA_QP_RQ_PSN = 1 << 12, + PVRDMA_QP_MAX_QP_RD_ATOMIC = 1 << 13, + PVRDMA_QP_ALT_PATH = 1 << 14, + PVRDMA_QP_MIN_RNR_TIMER = 1 << 15, + PVRDMA_QP_SQ_PSN = 1 << 16, + PVRDMA_QP_MAX_DEST_RD_ATOMIC = 1 << 17, + PVRDMA_QP_PATH_MIG_STATE = 1 << 18, + PVRDMA_QP_CAP = 1 << 19, + PVRDMA_QP_DEST_QPN = 1 << 20, + PVRDMA_QP_ATTR_MASK_MAX = PVRDMA_QP_DEST_QPN, +}; + +enum pvrdma_qp_state { + PVRDMA_QPS_RESET, + PVRDMA_QPS_INIT, + PVRDMA_QPS_RTR, + PVRDMA_QPS_RTS, + PVRDMA_QPS_SQD, + PVRDMA_QPS_SQE, + PVRDMA_QPS_ERR, +}; + +enum pvrdma_mig_state { + PVRDMA_MIG_MIGRATED, + PVRDMA_MIG_REARM, + PVRDMA_MIG_ARMED, +}; + +enum pvrdma_mw_type { + PVRDMA_MW_TYPE_1 = 1, + PVRDMA_MW_TYPE_2 = 2, +}; + +struct pvrdma_qp_attr { + enum pvrdma_qp_state qp_state; + enum pvrdma_qp_state cur_qp_state; + enum pvrdma_mtu path_mtu; + enum pvrdma_mig_state path_mig_state; + u32 qkey; + u32 rq_psn; + u32 sq_psn; + u32 dest_qp_num; + u32 qp_access_flags; + u16 pkey_index; + u16 alt_pkey_index; + u8 en_sqd_async_notify; + u8 sq_draining; + u8 max_rd_atomic; + u8 max_dest_rd_atomic; + u8 min_rnr_timer; + u8 port_num; + u8 timeout; + u8 retry_cnt; + u8 rnr_retry; + u8 alt_port_num; + u8 alt_timeout; + u8 reserved[5]; + struct pvrdma_qp_cap cap; + struct pvrdma_ah_attr ah_attr; + struct pvrdma_ah_attr alt_ah_attr; +}; + +enum pvrdma_send_flags { + PVRDMA_SEND_FENCE = 1 << 0, + PVRDMA_SEND_SIGNALED = 1 << 1, + PVRDMA_SEND_SOLICITED = 1 << 2, + PVRDMA_SEND_INLINE = 1 << 3, + PVRDMA_SEND_IP_CSUM = 1 << 4, + PVRDMA_SEND_FLAGS_MAX = PVRDMA_SEND_IP_CSUM, +}; + +enum pvrdma_access_flags { + PVRDMA_ACCESS_LOCAL_WRITE = 1 << 0, + PVRDMA_ACCESS_REMOTE_WRITE = 1 << 1, + PVRDMA_ACCESS_REMOTE_READ = 1 << 2, + PVRDMA_ACCESS_REMOTE_ATOMIC = 1 << 3, + PVRDMA_ACCESS_MW_BIND = 1 << 4, + PVRDMA_ZERO_BASED = 1 << 5, + PVRDMA_ACCESS_ON_DEMAND = 1 << 6, + PVRDMA_ACCESS_FLAGS_MAX = PVRDMA_ACCESS_ON_DEMAND, +}; + +int pvrdma_query_device(struct ib_device *ibdev, + struct ib_device_attr *props, + struct ib_udata *udata); +int pvrdma_query_port(struct ib_device *ibdev, u8 port, + struct ib_port_attr *props); +int pvrdma_query_gid(struct ib_device *ibdev, u8 port, + int index, union ib_gid *gid); +int pvrdma_query_pkey(struct ib_device *ibdev, u8 port, + u16 index, u16 *pkey); +enum rdma_link_layer pvrdma_port_link_layer(struct ib_device *ibdev, + u8 port); +int pvrdma_modify_device(struct ib_device *ibdev, int mask, + struct ib_device_modify *props); +int pvrdma_modify_port(struct ib_device *ibdev, u8 port, + int mask, struct ib_port_modify *props); +int pvrdma_mmap(struct ib_ucontext *context, struct vm_area_struct *vma); +struct ib_ucontext *pvrdma_alloc_ucontext(struct ib_device *ibdev, + struct ib_udata *udata); +int pvrdma_dealloc_ucontext(struct ib_ucontext *context); +struct ib_pd *pvrdma_alloc_pd(struct ib_device *ibdev, + struct ib_ucontext *context, + struct ib_udata *udata); +int pvrdma_dealloc_pd(struct ib_pd *ibpd); +struct ib_mr *pvrdma_get_dma_mr(struct ib_pd *pd, int acc); +struct ib_mr *pvrdma_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, + u64 virt_addr, int access_flags, + struct ib_udata *udata); +int pvrdma_dereg_mr(struct ib_mr *mr); +struct ib_mr *pvrdma_alloc_mr(struct ib_pd *pd, enum ib_mr_type mr_type, + u32 max_num_sg); +int pvrdma_map_mr_sg(struct ib_mr *ibmr, struct scatterlist *sg, + int sg_nents, unsigned int *sg_offset); +int pvrdma_modify_cq(struct ib_cq *cq, u16 cq_count, u16 cq_period); +int pvrdma_resize_cq(struct ib_cq *ibcq, int entries, + struct ib_udata *udata); +struct ib_cq *pvrdma_create_cq(struct ib_device *ibdev, + const struct ib_cq_init_attr *attr, + struct ib_ucontext *context, + struct ib_udata *udata); +int pvrdma_resize_cq(struct ib_cq *ibcq, int entries, + struct ib_udata *udata); +int pvrdma_destroy_cq(struct ib_cq *cq); +int pvrdma_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *wc); +int pvrdma_req_notify_cq(struct ib_cq *cq, enum ib_cq_notify_flags flags); +struct ib_ah *pvrdma_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr, + struct ib_udata *udata); +int pvrdma_destroy_ah(struct ib_ah *ah); +struct ib_qp *pvrdma_create_qp(struct ib_pd *pd, + struct ib_qp_init_attr *init_attr, + struct ib_udata *udata); +int pvrdma_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, + int attr_mask, struct ib_udata *udata); +int pvrdma_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, + int qp_attr_mask, struct ib_qp_init_attr *qp_init_attr); +int pvrdma_destroy_qp(struct ib_qp *qp); +int pvrdma_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, + struct ib_send_wr **bad_wr); +int pvrdma_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr, + struct ib_recv_wr **bad_wr); + +#endif /* __PVRDMA_VERBS_H__ */ diff --git a/drivers/infiniband/sw/rdmavt/Kconfig b/drivers/infiniband/sw/rdmavt/Kconfig index 1da8d01a6855..fdd001ce13d8 100644 --- a/drivers/infiniband/sw/rdmavt/Kconfig +++ b/drivers/infiniband/sw/rdmavt/Kconfig @@ -1,5 +1,6 @@ config INFINIBAND_RDMAVT tristate "RDMA verbs transport library" depends on 64BIT + select DMA_VIRT_OPS ---help--- This is a common software verbs provider for RDMA networks. diff --git a/drivers/infiniband/sw/rdmavt/Makefile b/drivers/infiniband/sw/rdmavt/Makefile index ccaa7992ac97..78b276a90401 100644 --- a/drivers/infiniband/sw/rdmavt/Makefile +++ b/drivers/infiniband/sw/rdmavt/Makefile @@ -7,7 +7,7 @@ # obj-$(CONFIG_INFINIBAND_RDMAVT) += rdmavt.o -rdmavt-y := vt.o ah.o cq.o dma.o mad.o mcast.o mmap.o mr.o pd.o qp.o srq.o \ - trace.o +rdmavt-y := vt.o ah.o cq.o mad.o mcast.o mmap.o mr.o pd.o qp.o \ + rc.o srq.o trace.o CFLAGS_trace.o = -I$(src) diff --git a/drivers/infiniband/sw/rdmavt/cq.c b/drivers/infiniband/sw/rdmavt/cq.c index 6d9904a4a0ab..7aa7a4e312f1 100644 --- a/drivers/infiniband/sw/rdmavt/cq.c +++ b/drivers/infiniband/sw/rdmavt/cq.c @@ -119,18 +119,17 @@ void rvt_cq_enter(struct rvt_cq *cq, struct ib_wc *entry, bool solicited) if (cq->notify == IB_CQ_NEXT_COMP || (cq->notify == IB_CQ_SOLICITED && (solicited || entry->status != IB_WC_SUCCESS))) { - struct kthread_worker *worker; /* * This will cause send_complete() to be called in * another thread. */ - smp_read_barrier_depends(); /* see rvt_cq_exit */ - worker = cq->rdi->worker; - if (likely(worker)) { + spin_lock(&cq->rdi->n_cqs_lock); + if (likely(cq->rdi->worker)) { cq->notify = RVT_CQ_NONE; cq->triggered++; - kthread_queue_work(worker, &cq->comptask); + kthread_queue_work(cq->rdi->worker, &cq->comptask); } + spin_unlock(&cq->rdi->n_cqs_lock); } spin_unlock_irqrestore(&cq->lock, flags); @@ -240,15 +239,15 @@ struct ib_cq *rvt_create_cq(struct ib_device *ibdev, } } - spin_lock(&rdi->n_cqs_lock); + spin_lock_irq(&rdi->n_cqs_lock); if (rdi->n_cqs_allocated == rdi->dparms.props.max_cq) { - spin_unlock(&rdi->n_cqs_lock); + spin_unlock_irq(&rdi->n_cqs_lock); ret = ERR_PTR(-ENOMEM); goto bail_ip; } rdi->n_cqs_allocated++; - spin_unlock(&rdi->n_cqs_lock); + spin_unlock_irq(&rdi->n_cqs_lock); if (cq->ip) { spin_lock_irq(&rdi->pending_lock); @@ -296,9 +295,9 @@ int rvt_destroy_cq(struct ib_cq *ibcq) struct rvt_dev_info *rdi = cq->rdi; kthread_flush_work(&cq->comptask); - spin_lock(&rdi->n_cqs_lock); + spin_lock_irq(&rdi->n_cqs_lock); rdi->n_cqs_allocated--; - spin_unlock(&rdi->n_cqs_lock); + spin_unlock_irq(&rdi->n_cqs_lock); if (cq->ip) kref_put(&cq->ip->ref, rvt_release_mmap_info); else @@ -504,33 +503,23 @@ int rvt_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *entry) */ int rvt_driver_cq_init(struct rvt_dev_info *rdi) { - int ret = 0; int cpu; - struct task_struct *task; + struct kthread_worker *worker; if (rdi->worker) return 0; + spin_lock_init(&rdi->n_cqs_lock); - rdi->worker = kzalloc(sizeof(*rdi->worker), GFP_KERNEL); - if (!rdi->worker) - return -ENOMEM; - kthread_init_worker(rdi->worker); - task = kthread_create_on_node( - kthread_worker_fn, - rdi->worker, - rdi->dparms.node, - "%s", rdi->dparms.cq_name); - if (IS_ERR(task)) { - kfree(rdi->worker); - rdi->worker = NULL; - return PTR_ERR(task); - } - set_user_nice(task, MIN_NICE); cpu = cpumask_first(cpumask_of_node(rdi->dparms.node)); - kthread_bind(task, cpu); - wake_up_process(task); - return ret; + worker = kthread_create_worker_on_cpu(cpu, 0, + "%s", rdi->dparms.cq_name); + if (IS_ERR(worker)) + return PTR_ERR(worker); + + set_user_nice(worker->task, MIN_NICE); + rdi->worker = worker; + return 0; } /** @@ -541,13 +530,15 @@ void rvt_cq_exit(struct rvt_dev_info *rdi) { struct kthread_worker *worker; + /* block future queuing from send_complete() */ + spin_lock_irq(&rdi->n_cqs_lock); worker = rdi->worker; - if (!worker) + if (!worker) { + spin_unlock_irq(&rdi->n_cqs_lock); return; - /* blocks future queuing from send_complete() */ + } rdi->worker = NULL; - smp_wmb(); /* See rdi_cq_enter */ - kthread_flush_worker(worker); - kthread_stop(worker->task); - kfree(worker); + spin_unlock_irq(&rdi->n_cqs_lock); + + kthread_destroy_worker(worker); } diff --git a/drivers/infiniband/sw/rdmavt/dma.c b/drivers/infiniband/sw/rdmavt/dma.c deleted file mode 100644 index f2cefb0d9180..000000000000 --- a/drivers/infiniband/sw/rdmavt/dma.c +++ /dev/null @@ -1,198 +0,0 @@ -/* - * Copyright(c) 2016 Intel Corporation. - * - * This file is provided under a dual BSD/GPLv2 license. When using or - * redistributing this file, you may do so under either license. - * - * GPL LICENSE SUMMARY - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * 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. - * - * BSD LICENSE - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * - * - Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * - Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in - * the documentation and/or other materials provided with the - * distribution. - * - Neither the name of Intel Corporation nor the names of its - * contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR - * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT - * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, - * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT - * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, - * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY - * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE - * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - * - */ -#include <linux/types.h> -#include <linux/scatterlist.h> -#include <rdma/ib_verbs.h> - -#include "dma.h" - -#define BAD_DMA_ADDRESS ((u64)0) - -/* - * The following functions implement driver specific replacements - * for the ib_dma_*() functions. - * - * These functions return kernel virtual addresses instead of - * device bus addresses since the driver uses the CPU to copy - * data instead of using hardware DMA. - */ - -static int rvt_mapping_error(struct ib_device *dev, u64 dma_addr) -{ - return dma_addr == BAD_DMA_ADDRESS; -} - -static u64 rvt_dma_map_single(struct ib_device *dev, void *cpu_addr, - size_t size, enum dma_data_direction direction) -{ - if (WARN_ON(!valid_dma_direction(direction))) - return BAD_DMA_ADDRESS; - - return (u64)cpu_addr; -} - -static void rvt_dma_unmap_single(struct ib_device *dev, u64 addr, size_t size, - enum dma_data_direction direction) -{ - /* This is a stub, nothing to be done here */ -} - -static u64 rvt_dma_map_page(struct ib_device *dev, struct page *page, - unsigned long offset, size_t size, - enum dma_data_direction direction) -{ - u64 addr; - - if (WARN_ON(!valid_dma_direction(direction))) - return BAD_DMA_ADDRESS; - - addr = (u64)page_address(page); - if (addr) - addr += offset; - - return addr; -} - -static void rvt_dma_unmap_page(struct ib_device *dev, u64 addr, size_t size, - enum dma_data_direction direction) -{ - /* This is a stub, nothing to be done here */ -} - -static int rvt_map_sg(struct ib_device *dev, struct scatterlist *sgl, - int nents, enum dma_data_direction direction) -{ - struct scatterlist *sg; - u64 addr; - int i; - int ret = nents; - - if (WARN_ON(!valid_dma_direction(direction))) - return 0; - - for_each_sg(sgl, sg, nents, i) { - addr = (u64)page_address(sg_page(sg)); - if (!addr) { - ret = 0; - break; - } - sg->dma_address = addr + sg->offset; -#ifdef CONFIG_NEED_SG_DMA_LENGTH - sg->dma_length = sg->length; -#endif - } - return ret; -} - -static void rvt_unmap_sg(struct ib_device *dev, - struct scatterlist *sg, int nents, - enum dma_data_direction direction) -{ - /* This is a stub, nothing to be done here */ -} - -static int rvt_map_sg_attrs(struct ib_device *dev, struct scatterlist *sgl, - int nents, enum dma_data_direction direction, - unsigned long attrs) -{ - return rvt_map_sg(dev, sgl, nents, direction); -} - -static void rvt_unmap_sg_attrs(struct ib_device *dev, - struct scatterlist *sg, int nents, - enum dma_data_direction direction, - unsigned long attrs) -{ - return rvt_unmap_sg(dev, sg, nents, direction); -} - -static void rvt_sync_single_for_cpu(struct ib_device *dev, u64 addr, - size_t size, enum dma_data_direction dir) -{ -} - -static void rvt_sync_single_for_device(struct ib_device *dev, u64 addr, - size_t size, - enum dma_data_direction dir) -{ -} - -static void *rvt_dma_alloc_coherent(struct ib_device *dev, size_t size, - u64 *dma_handle, gfp_t flag) -{ - struct page *p; - void *addr = NULL; - - p = alloc_pages(flag, get_order(size)); - if (p) - addr = page_address(p); - if (dma_handle) - *dma_handle = (u64)addr; - return addr; -} - -static void rvt_dma_free_coherent(struct ib_device *dev, size_t size, - void *cpu_addr, u64 dma_handle) -{ - free_pages((unsigned long)cpu_addr, get_order(size)); -} - -struct ib_dma_mapping_ops rvt_default_dma_mapping_ops = { - .mapping_error = rvt_mapping_error, - .map_single = rvt_dma_map_single, - .unmap_single = rvt_dma_unmap_single, - .map_page = rvt_dma_map_page, - .unmap_page = rvt_dma_unmap_page, - .map_sg = rvt_map_sg, - .unmap_sg = rvt_unmap_sg, - .map_sg_attrs = rvt_map_sg_attrs, - .unmap_sg_attrs = rvt_unmap_sg_attrs, - .sync_single_for_cpu = rvt_sync_single_for_cpu, - .sync_single_for_device = rvt_sync_single_for_device, - .alloc_coherent = rvt_dma_alloc_coherent, - .free_coherent = rvt_dma_free_coherent -}; diff --git a/drivers/infiniband/sw/rdmavt/mcast.c b/drivers/infiniband/sw/rdmavt/mcast.c index 983d319ac976..05c8c2afb0e3 100644 --- a/drivers/infiniband/sw/rdmavt/mcast.c +++ b/drivers/infiniband/sw/rdmavt/mcast.c @@ -81,7 +81,7 @@ static struct rvt_mcast_qp *rvt_mcast_qp_alloc(struct rvt_qp *qp) goto bail; mqp->qp = qp; - atomic_inc(&qp->refcount); + rvt_get_qp(qp); bail: return mqp; @@ -92,8 +92,7 @@ static void rvt_mcast_qp_free(struct rvt_mcast_qp *mqp) struct rvt_qp *qp = mqp->qp; /* Notify hfi1_destroy_qp() if it is waiting. */ - if (atomic_dec_and_test(&qp->refcount)) - wake_up(&qp->wait); + rvt_put_qp(qp); kfree(mqp); } diff --git a/drivers/infiniband/sw/rdmavt/mr.c b/drivers/infiniband/sw/rdmavt/mr.c index 46b64970058e..ae30b6838d79 100644 --- a/drivers/infiniband/sw/rdmavt/mr.c +++ b/drivers/infiniband/sw/rdmavt/mr.c @@ -51,6 +51,7 @@ #include <rdma/rdma_vt.h> #include "vt.h" #include "mr.h" +#include "trace.h" /** * rvt_driver_mr_init - Init MR resources per driver @@ -84,6 +85,7 @@ int rvt_driver_mr_init(struct rvt_dev_info *rdi) lkey_table_size = rdi->dparms.lkey_table_size; } rdi->lkey_table.max = 1 << lkey_table_size; + rdi->lkey_table.shift = 32 - lkey_table_size; lk_tab_size = rdi->lkey_table.max * sizeof(*rdi->lkey_table.table); rdi->lkey_table.table = (struct rvt_mregion __rcu **) vmalloc_node(lk_tab_size, rdi->dparms.node); @@ -118,10 +120,19 @@ static void rvt_deinit_mregion(struct rvt_mregion *mr) mr->mapsz = 0; while (i) kfree(mr->map[--i]); + percpu_ref_exit(&mr->refcount); +} + +static void __rvt_mregion_complete(struct percpu_ref *ref) +{ + struct rvt_mregion *mr = container_of(ref, struct rvt_mregion, + refcount); + + complete(&mr->comp); } static int rvt_init_mregion(struct rvt_mregion *mr, struct ib_pd *pd, - int count) + int count, unsigned int percpu_flags) { int m, i = 0; struct rvt_dev_info *dev = ib_to_rvt(pd->device); @@ -131,19 +142,23 @@ static int rvt_init_mregion(struct rvt_mregion *mr, struct ib_pd *pd, for (; i < m; i++) { mr->map[i] = kzalloc_node(sizeof(*mr->map[0]), GFP_KERNEL, dev->dparms.node); - if (!mr->map[i]) { - rvt_deinit_mregion(mr); - return -ENOMEM; - } + if (!mr->map[i]) + goto bail; mr->mapsz++; } init_completion(&mr->comp); /* count returning the ptr to user */ - atomic_set(&mr->refcount, 1); + if (percpu_ref_init(&mr->refcount, &__rvt_mregion_complete, + percpu_flags, GFP_KERNEL)) + goto bail; + atomic_set(&mr->lkey_invalid, 0); mr->pd = pd; mr->max_segs = count; return 0; +bail: + rvt_deinit_mregion(mr); + return -ENOMEM; } /** @@ -178,8 +193,7 @@ static int rvt_alloc_lkey(struct rvt_mregion *mr, int dma_region) if (!tmr) { rcu_assign_pointer(dev->dma_mr, mr); mr->lkey_published = 1; - } else { - rvt_put_mr(mr); + rvt_get_mr(mr); } goto success; } @@ -237,11 +251,14 @@ static void rvt_free_lkey(struct rvt_mregion *mr) int freed = 0; spin_lock_irqsave(&rkt->lock, flags); - if (!mr->lkey_published) - goto out; - if (lkey == 0) { - RCU_INIT_POINTER(dev->dma_mr, NULL); + if (!lkey) { + if (mr->lkey_published) { + RCU_INIT_POINTER(dev->dma_mr, NULL); + rvt_put_mr(mr); + } } else { + if (!mr->lkey_published) + goto out; r = lkey >> (32 - dev->dparms.lkey_table_size); RCU_INIT_POINTER(rkt->table[r], NULL); } @@ -251,7 +268,7 @@ out: spin_unlock_irqrestore(&rkt->lock, flags); if (freed) { synchronize_rcu(); - rvt_put_mr(mr); + percpu_ref_kill(&mr->refcount); } } @@ -267,7 +284,7 @@ static struct rvt_mr *__rvt_alloc_mr(int count, struct ib_pd *pd) if (!mr) goto bail; - rval = rvt_init_mregion(&mr->mr, pd, count); + rval = rvt_init_mregion(&mr->mr, pd, count, 0); if (rval) goto bail; /* @@ -292,8 +309,8 @@ bail: static void __rvt_free_mr(struct rvt_mr *mr) { - rvt_deinit_mregion(&mr->mr); rvt_free_lkey(&mr->mr); + rvt_deinit_mregion(&mr->mr); kfree(mr); } @@ -303,8 +320,8 @@ static void __rvt_free_mr(struct rvt_mr *mr) * @acc: access flags * * Return: the memory region on success, otherwise returns an errno. - * Note that all DMA addresses should be created via the - * struct ib_dma_mapping_ops functions (see dma.c). + * Note that all DMA addresses should be created via the functions in + * struct dma_virt_ops. */ struct ib_mr *rvt_get_dma_mr(struct ib_pd *pd, int acc) { @@ -321,7 +338,7 @@ struct ib_mr *rvt_get_dma_mr(struct ib_pd *pd, int acc) goto bail; } - rval = rvt_init_mregion(&mr->mr, pd, 0); + rval = rvt_init_mregion(&mr->mr, pd, 0, 0); if (rval) { ret = ERR_PTR(rval); goto bail; @@ -402,6 +419,7 @@ struct ib_mr *rvt_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, } mr->mr.map[m]->segs[n].vaddr = vaddr; mr->mr.map[m]->segs[n].length = umem->page_size; + trace_rvt_mr_user_seg(&mr->mr, m, n, vaddr, umem->page_size); n++; if (n == RVT_SEGSZ) { m++; @@ -442,8 +460,8 @@ int rvt_dereg_mr(struct ib_mr *ibmr) timeout = wait_for_completion_timeout(&mr->mr.comp, 5 * HZ); if (!timeout) { rvt_pr_err(rdi, - "rvt_dereg_mr timeout mr %p pd %p refcount %u\n", - mr, mr->mr.pd, atomic_read(&mr->mr.refcount)); + "rvt_dereg_mr timeout mr %p pd %p\n", + mr, mr->mr.pd); rvt_get_mr(&mr->mr); ret = -EBUSY; goto out; @@ -506,6 +524,7 @@ static int rvt_set_page(struct ib_mr *ibmr, u64 addr) n = mapped_segs % RVT_SEGSZ; mr->mr.map[m]->segs[n].vaddr = (void *)addr; mr->mr.map[m]->segs[n].length = ps; + trace_rvt_mr_page_seg(&mr->mr, m, n, (void *)addr, ps); mr->mr.length += ps; return 0; @@ -619,7 +638,8 @@ struct ib_fmr *rvt_alloc_fmr(struct ib_pd *pd, int mr_access_flags, if (!fmr) goto bail; - rval = rvt_init_mregion(&fmr->mr, pd, fmr_attr->max_pages); + rval = rvt_init_mregion(&fmr->mr, pd, fmr_attr->max_pages, + PERCPU_REF_INIT_ATOMIC); if (rval) goto bail; @@ -670,11 +690,12 @@ int rvt_map_phys_fmr(struct ib_fmr *ibfmr, u64 *page_list, struct rvt_fmr *fmr = to_ifmr(ibfmr); struct rvt_lkey_table *rkt; unsigned long flags; - int m, n, i; + int m, n; + unsigned long i; u32 ps; struct rvt_dev_info *rdi = ib_to_rvt(ibfmr->device); - i = atomic_read(&fmr->mr.refcount); + i = atomic_long_read(&fmr->mr.refcount.count); if (i > 2) return -EBUSY; @@ -692,6 +713,7 @@ int rvt_map_phys_fmr(struct ib_fmr *ibfmr, u64 *page_list, for (i = 0; i < list_len; i++) { fmr->mr.map[m]->segs[n].vaddr = (void *)page_list[i]; fmr->mr.map[m]->segs[n].length = ps; + trace_rvt_mr_fmr_seg(&fmr->mr, m, n, (void *)page_list[i], ps); if (++n == RVT_SEGSZ) { m++; n = 0; @@ -774,20 +796,21 @@ int rvt_lkey_ok(struct rvt_lkey_table *rkt, struct rvt_pd *pd, struct rvt_mregion *mr; unsigned n, m; size_t off; - struct rvt_dev_info *dev = ib_to_rvt(pd->ibpd.device); /* * We use LKEY == zero for kernel virtual addresses - * (see rvt_get_dma_mr and dma.c). + * (see rvt_get_dma_mr() and dma_virt_ops). */ rcu_read_lock(); if (sge->lkey == 0) { + struct rvt_dev_info *dev = ib_to_rvt(pd->ibpd.device); + if (pd->user) goto bail; mr = rcu_dereference(dev->dma_mr); if (!mr) goto bail; - atomic_inc(&mr->refcount); + rvt_get_mr(mr); rcu_read_unlock(); isge->mr = mr; @@ -798,8 +821,7 @@ int rvt_lkey_ok(struct rvt_lkey_table *rkt, struct rvt_pd *pd, isge->n = 0; goto ok; } - mr = rcu_dereference( - rkt->table[(sge->lkey >> (32 - dev->dparms.lkey_table_size))]); + mr = rcu_dereference(rkt->table[sge->lkey >> rkt->shift]); if (unlikely(!mr || atomic_read(&mr->lkey_invalid) || mr->lkey != sge->lkey || mr->pd != &pd->ibpd)) goto bail; @@ -809,7 +831,7 @@ int rvt_lkey_ok(struct rvt_lkey_table *rkt, struct rvt_pd *pd, off + sge->length > mr->length || (mr->access_flags & acc) != acc)) goto bail; - atomic_inc(&mr->refcount); + rvt_get_mr(mr); rcu_read_unlock(); off += mr->offset; @@ -875,7 +897,7 @@ int rvt_rkey_ok(struct rvt_qp *qp, struct rvt_sge *sge, /* * We use RKEY == zero for kernel virtual addresses - * (see rvt_get_dma_mr and dma.c). + * (see rvt_get_dma_mr() and dma_virt_ops). */ rcu_read_lock(); if (rkey == 0) { @@ -887,7 +909,7 @@ int rvt_rkey_ok(struct rvt_qp *qp, struct rvt_sge *sge, mr = rcu_dereference(rdi->dma_mr); if (!mr) goto bail; - atomic_inc(&mr->refcount); + rvt_get_mr(mr); rcu_read_unlock(); sge->mr = mr; @@ -899,8 +921,7 @@ int rvt_rkey_ok(struct rvt_qp *qp, struct rvt_sge *sge, goto ok; } - mr = rcu_dereference( - rkt->table[(rkey >> (32 - dev->dparms.lkey_table_size))]); + mr = rcu_dereference(rkt->table[rkey >> rkt->shift]); if (unlikely(!mr || atomic_read(&mr->lkey_invalid) || mr->lkey != rkey || qp->ibqp.pd != mr->pd)) goto bail; @@ -909,7 +930,7 @@ int rvt_rkey_ok(struct rvt_qp *qp, struct rvt_sge *sge, if (unlikely(vaddr < mr->iova || off + len > mr->length || (mr->access_flags & acc) == 0)) goto bail; - atomic_inc(&mr->refcount); + rvt_get_mr(mr); rcu_read_unlock(); off += mr->offset; diff --git a/drivers/infiniband/sw/rdmavt/pd.c b/drivers/infiniband/sw/rdmavt/pd.c index d1292f324c67..8a89afff3363 100644 --- a/drivers/infiniband/sw/rdmavt/pd.c +++ b/drivers/infiniband/sw/rdmavt/pd.c @@ -90,7 +90,7 @@ struct ib_pd *rvt_alloc_pd(struct ib_device *ibdev, spin_unlock(&dev->n_pds_lock); /* ib_alloc_pd() will initialize pd->ibpd. */ - pd->user = udata ? 1 : 0; + pd->user = !!udata; ret = &pd->ibpd; diff --git a/drivers/infiniband/sw/rdmavt/qp.c b/drivers/infiniband/sw/rdmavt/qp.c index 6500c3b5a89c..f5ad8d4bfb39 100644 --- a/drivers/infiniband/sw/rdmavt/qp.c +++ b/drivers/infiniband/sw/rdmavt/qp.c @@ -51,10 +51,51 @@ #include <linux/vmalloc.h> #include <linux/slab.h> #include <rdma/ib_verbs.h> +#include <rdma/ib_hdrs.h> #include "qp.h" #include "vt.h" #include "trace.h" +static void rvt_rc_timeout(unsigned long arg); + +/* + * Convert the AETH RNR timeout code into the number of microseconds. + */ +static const u32 ib_rvt_rnr_table[32] = { + 655360, /* 00: 655.36 */ + 10, /* 01: .01 */ + 20, /* 02 .02 */ + 30, /* 03: .03 */ + 40, /* 04: .04 */ + 60, /* 05: .06 */ + 80, /* 06: .08 */ + 120, /* 07: .12 */ + 160, /* 08: .16 */ + 240, /* 09: .24 */ + 320, /* 0A: .32 */ + 480, /* 0B: .48 */ + 640, /* 0C: .64 */ + 960, /* 0D: .96 */ + 1280, /* 0E: 1.28 */ + 1920, /* 0F: 1.92 */ + 2560, /* 10: 2.56 */ + 3840, /* 11: 3.84 */ + 5120, /* 12: 5.12 */ + 7680, /* 13: 7.68 */ + 10240, /* 14: 10.24 */ + 15360, /* 15: 15.36 */ + 20480, /* 16: 20.48 */ + 30720, /* 17: 30.72 */ + 40960, /* 18: 40.96 */ + 61440, /* 19: 61.44 */ + 81920, /* 1A: 81.92 */ + 122880, /* 1B: 122.88 */ + 163840, /* 1C: 163.84 */ + 245760, /* 1D: 245.76 */ + 327680, /* 1E: 327.68 */ + 491520 /* 1F: 491.52 */ +}; + /* * Note that it is OK to post send work requests in the SQE and ERR * states; rvt_do_send() will process them and generate error @@ -76,6 +117,23 @@ const int ib_rvt_state_ops[IB_QPS_ERR + 1] = { }; EXPORT_SYMBOL(ib_rvt_state_ops); +/* + * Translate ib_wr_opcode into ib_wc_opcode. + */ +const enum ib_wc_opcode ib_rvt_wc_opcode[] = { + [IB_WR_RDMA_WRITE] = IB_WC_RDMA_WRITE, + [IB_WR_RDMA_WRITE_WITH_IMM] = IB_WC_RDMA_WRITE, + [IB_WR_SEND] = IB_WC_SEND, + [IB_WR_SEND_WITH_IMM] = IB_WC_SEND, + [IB_WR_RDMA_READ] = IB_WC_RDMA_READ, + [IB_WR_ATOMIC_CMP_AND_SWP] = IB_WC_COMP_SWAP, + [IB_WR_ATOMIC_FETCH_AND_ADD] = IB_WC_FETCH_ADD, + [IB_WR_SEND_WITH_INV] = IB_WC_SEND, + [IB_WR_LOCAL_INV] = IB_WC_LOCAL_INV, + [IB_WR_REG_MR] = IB_WC_REG_MR +}; +EXPORT_SYMBOL(ib_rvt_wc_opcode); + static void get_map_page(struct rvt_qpn_table *qpt, struct rvt_qpn_map *map, gfp_t gfp) @@ -183,7 +241,8 @@ int rvt_driver_qp_init(struct rvt_dev_info *rdi) if (!rdi->driver_f.free_all_qps || !rdi->driver_f.qp_priv_alloc || !rdi->driver_f.qp_priv_free || - !rdi->driver_f.notify_qp_reset) + !rdi->driver_f.notify_qp_reset || + !rdi->driver_f.notify_restart_rc) return -EINVAL; /* allocate parent object */ @@ -570,6 +629,7 @@ static void rvt_reset_qp(struct rvt_dev_info *rdi, struct rvt_qp *qp, /* Let drivers flush their waitlist */ rdi->driver_f.flush_qp_waiters(qp); + rvt_stop_rc_timers(qp); qp->s_flags &= ~(RVT_S_TIMER | RVT_S_ANY_WAIT); spin_unlock(&qp->s_lock); spin_unlock(&qp->s_hlock); @@ -577,7 +637,7 @@ static void rvt_reset_qp(struct rvt_dev_info *rdi, struct rvt_qp *qp, /* Stop the send queue and the retry timer */ rdi->driver_f.stop_send_queue(qp); - + rvt_del_timers_sync(qp); /* Wait for things to stop */ rdi->driver_f.quiesce_qp(qp); @@ -713,6 +773,11 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, if (!qp->s_ack_queue) goto bail_qp; } + /* initialize timers needed for rc qp */ + setup_timer(&qp->s_timer, rvt_rc_timeout, (unsigned long)qp); + hrtimer_init(&qp->s_rnr_timer, CLOCK_MONOTONIC, + HRTIMER_MODE_REL); + qp->s_rnr_timer.function = rvt_rc_rnr_retry; /* * Driver needs to set up it's private QP structure and do any @@ -884,7 +949,8 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, return ret; bail_ip: - kref_put(&qp->ip->ref, rvt_release_mmap_info); + if (qp->ip) + kref_put(&qp->ip->ref, rvt_release_mmap_info); bail_qpn: free_qpn(&rdi->qp_dev->qpn_table, qp->ibqp.qp_num); @@ -1850,3 +1916,184 @@ int rvt_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr, } return 0; } + +/** + * qp_comm_est - handle trap with QP established + * @qp: the QP + */ +void rvt_comm_est(struct rvt_qp *qp) +{ + qp->r_flags |= RVT_R_COMM_EST; + if (qp->ibqp.event_handler) { + struct ib_event ev; + + ev.device = qp->ibqp.device; + ev.element.qp = &qp->ibqp; + ev.event = IB_EVENT_COMM_EST; + qp->ibqp.event_handler(&ev, qp->ibqp.qp_context); + } +} +EXPORT_SYMBOL(rvt_comm_est); + +void rvt_rc_error(struct rvt_qp *qp, enum ib_wc_status err) +{ + unsigned long flags; + int lastwqe; + + spin_lock_irqsave(&qp->s_lock, flags); + lastwqe = rvt_error_qp(qp, err); + spin_unlock_irqrestore(&qp->s_lock, flags); + + if (lastwqe) { + struct ib_event ev; + + ev.device = qp->ibqp.device; + ev.element.qp = &qp->ibqp; + ev.event = IB_EVENT_QP_LAST_WQE_REACHED; + qp->ibqp.event_handler(&ev, qp->ibqp.qp_context); + } +} +EXPORT_SYMBOL(rvt_rc_error); + +/* + * rvt_rnr_tbl_to_usec - return index into ib_rvt_rnr_table + * @index - the index + * return usec from an index into ib_rvt_rnr_table + */ +unsigned long rvt_rnr_tbl_to_usec(u32 index) +{ + return ib_rvt_rnr_table[(index & IB_AETH_CREDIT_MASK)]; +} +EXPORT_SYMBOL(rvt_rnr_tbl_to_usec); + +static inline unsigned long rvt_aeth_to_usec(u32 aeth) +{ + return ib_rvt_rnr_table[(aeth >> IB_AETH_CREDIT_SHIFT) & + IB_AETH_CREDIT_MASK]; +} + +/* + * rvt_add_retry_timer - add/start a retry timer + * @qp - the QP + * add a retry timer on the QP + */ +void rvt_add_retry_timer(struct rvt_qp *qp) +{ + struct ib_qp *ibqp = &qp->ibqp; + struct rvt_dev_info *rdi = ib_to_rvt(ibqp->device); + + lockdep_assert_held(&qp->s_lock); + qp->s_flags |= RVT_S_TIMER; + /* 4.096 usec. * (1 << qp->timeout) */ + qp->s_timer.expires = jiffies + qp->timeout_jiffies + + rdi->busy_jiffies; + add_timer(&qp->s_timer); +} +EXPORT_SYMBOL(rvt_add_retry_timer); + +/** + * rvt_add_rnr_timer - add/start an rnr timer + * @qp - the QP + * @aeth - aeth of RNR timeout, simulated aeth for loopback + * add an rnr timer on the QP + */ +void rvt_add_rnr_timer(struct rvt_qp *qp, u32 aeth) +{ + u32 to; + + lockdep_assert_held(&qp->s_lock); + qp->s_flags |= RVT_S_WAIT_RNR; + to = rvt_aeth_to_usec(aeth); + hrtimer_start(&qp->s_rnr_timer, + ns_to_ktime(1000 * to), HRTIMER_MODE_REL); +} +EXPORT_SYMBOL(rvt_add_rnr_timer); + +/** + * rvt_stop_rc_timers - stop all timers + * @qp - the QP + * stop any pending timers + */ +void rvt_stop_rc_timers(struct rvt_qp *qp) +{ + lockdep_assert_held(&qp->s_lock); + /* Remove QP from all timers */ + if (qp->s_flags & (RVT_S_TIMER | RVT_S_WAIT_RNR)) { + qp->s_flags &= ~(RVT_S_TIMER | RVT_S_WAIT_RNR); + del_timer(&qp->s_timer); + hrtimer_try_to_cancel(&qp->s_rnr_timer); + } +} +EXPORT_SYMBOL(rvt_stop_rc_timers); + +/** + * rvt_stop_rnr_timer - stop an rnr timer + * @qp - the QP + * + * stop an rnr timer and return if the timer + * had been pending. + */ +static int rvt_stop_rnr_timer(struct rvt_qp *qp) +{ + int rval = 0; + + lockdep_assert_held(&qp->s_lock); + /* Remove QP from rnr timer */ + if (qp->s_flags & RVT_S_WAIT_RNR) { + qp->s_flags &= ~RVT_S_WAIT_RNR; + rval = hrtimer_try_to_cancel(&qp->s_rnr_timer); + } + return rval; +} + +/** + * rvt_del_timers_sync - wait for any timeout routines to exit + * @qp - the QP + */ +void rvt_del_timers_sync(struct rvt_qp *qp) +{ + del_timer_sync(&qp->s_timer); + hrtimer_cancel(&qp->s_rnr_timer); +} +EXPORT_SYMBOL(rvt_del_timers_sync); + +/** + * This is called from s_timer for missing responses. + */ +static void rvt_rc_timeout(unsigned long arg) +{ + struct rvt_qp *qp = (struct rvt_qp *)arg; + struct rvt_dev_info *rdi = ib_to_rvt(qp->ibqp.device); + unsigned long flags; + + spin_lock_irqsave(&qp->r_lock, flags); + spin_lock(&qp->s_lock); + if (qp->s_flags & RVT_S_TIMER) { + qp->s_flags &= ~RVT_S_TIMER; + del_timer(&qp->s_timer); + if (rdi->driver_f.notify_restart_rc) + rdi->driver_f.notify_restart_rc(qp, + qp->s_last_psn + 1, + 1); + rdi->driver_f.schedule_send(qp); + } + spin_unlock(&qp->s_lock); + spin_unlock_irqrestore(&qp->r_lock, flags); +} + +/* + * This is called from s_timer for RNR timeouts. + */ +enum hrtimer_restart rvt_rc_rnr_retry(struct hrtimer *t) +{ + struct rvt_qp *qp = container_of(t, struct rvt_qp, s_rnr_timer); + struct rvt_dev_info *rdi = ib_to_rvt(qp->ibqp.device); + unsigned long flags; + + spin_lock_irqsave(&qp->s_lock, flags); + rvt_stop_rnr_timer(qp); + rdi->driver_f.schedule_send(qp); + spin_unlock_irqrestore(&qp->s_lock, flags); + return HRTIMER_NORESTART; +} +EXPORT_SYMBOL(rvt_rc_rnr_retry); diff --git a/drivers/infiniband/sw/rdmavt/rc.c b/drivers/infiniband/sw/rdmavt/rc.c new file mode 100644 index 000000000000..6131cc558bdb --- /dev/null +++ b/drivers/infiniband/sw/rdmavt/rc.c @@ -0,0 +1,189 @@ +/* + * Copyright(c) 2016 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * 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. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ + +#include <rdma/rdma_vt.h> +#include <rdma/ib_hdrs.h> + +/* + * Convert the AETH credit code into the number of credits. + */ +static const u16 credit_table[31] = { + 0, /* 0 */ + 1, /* 1 */ + 2, /* 2 */ + 3, /* 3 */ + 4, /* 4 */ + 6, /* 5 */ + 8, /* 6 */ + 12, /* 7 */ + 16, /* 8 */ + 24, /* 9 */ + 32, /* A */ + 48, /* B */ + 64, /* C */ + 96, /* D */ + 128, /* E */ + 192, /* F */ + 256, /* 10 */ + 384, /* 11 */ + 512, /* 12 */ + 768, /* 13 */ + 1024, /* 14 */ + 1536, /* 15 */ + 2048, /* 16 */ + 3072, /* 17 */ + 4096, /* 18 */ + 6144, /* 19 */ + 8192, /* 1A */ + 12288, /* 1B */ + 16384, /* 1C */ + 24576, /* 1D */ + 32768 /* 1E */ +}; + +/** + * rvt_compute_aeth - compute the AETH (syndrome + MSN) + * @qp: the queue pair to compute the AETH for + * + * Returns the AETH. + */ +__be32 rvt_compute_aeth(struct rvt_qp *qp) +{ + u32 aeth = qp->r_msn & IB_MSN_MASK; + + if (qp->ibqp.srq) { + /* + * Shared receive queues don't generate credits. + * Set the credit field to the invalid value. + */ + aeth |= IB_AETH_CREDIT_INVAL << IB_AETH_CREDIT_SHIFT; + } else { + u32 min, max, x; + u32 credits; + struct rvt_rwq *wq = qp->r_rq.wq; + u32 head; + u32 tail; + + /* sanity check pointers before trusting them */ + head = wq->head; + if (head >= qp->r_rq.size) + head = 0; + tail = wq->tail; + if (tail >= qp->r_rq.size) + tail = 0; + /* + * Compute the number of credits available (RWQEs). + * There is a small chance that the pair of reads are + * not atomic, which is OK, since the fuzziness is + * resolved as further ACKs go out. + */ + credits = head - tail; + if ((int)credits < 0) + credits += qp->r_rq.size; + /* + * Binary search the credit table to find the code to + * use. + */ + min = 0; + max = 31; + for (;;) { + x = (min + max) / 2; + if (credit_table[x] == credits) + break; + if (credit_table[x] > credits) { + max = x; + } else { + if (min == x) + break; + min = x; + } + } + aeth |= x << IB_AETH_CREDIT_SHIFT; + } + return cpu_to_be32(aeth); +} +EXPORT_SYMBOL(rvt_compute_aeth); + +/** + * rvt_get_credit - flush the send work queue of a QP + * @qp: the qp who's send work queue to flush + * @aeth: the Acknowledge Extended Transport Header + * + * The QP s_lock should be held. + */ +void rvt_get_credit(struct rvt_qp *qp, u32 aeth) +{ + struct rvt_dev_info *rdi = ib_to_rvt(qp->ibqp.device); + u32 credit = (aeth >> IB_AETH_CREDIT_SHIFT) & IB_AETH_CREDIT_MASK; + + lockdep_assert_held(&qp->s_lock); + /* + * If the credit is invalid, we can send + * as many packets as we like. Otherwise, we have to + * honor the credit field. + */ + if (credit == IB_AETH_CREDIT_INVAL) { + if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT)) { + qp->s_flags |= RVT_S_UNLIMITED_CREDIT; + if (qp->s_flags & RVT_S_WAIT_SSN_CREDIT) { + qp->s_flags &= ~RVT_S_WAIT_SSN_CREDIT; + rdi->driver_f.schedule_send(qp); + } + } + } else if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT)) { + /* Compute new LSN (i.e., MSN + credit) */ + credit = (aeth + credit_table[credit]) & IB_MSN_MASK; + if (rvt_cmp_msn(credit, qp->s_lsn) > 0) { + qp->s_lsn = credit; + if (qp->s_flags & RVT_S_WAIT_SSN_CREDIT) { + qp->s_flags &= ~RVT_S_WAIT_SSN_CREDIT; + rdi->driver_f.schedule_send(qp); + } + } + } +} +EXPORT_SYMBOL(rvt_get_credit); diff --git a/drivers/infiniband/sw/rdmavt/trace.h b/drivers/infiniband/sw/rdmavt/trace.h index 6c0457db5499..e2d23acb6a7d 100644 --- a/drivers/infiniband/sw/rdmavt/trace.h +++ b/drivers/infiniband/sw/rdmavt/trace.h @@ -45,143 +45,10 @@ * */ -#undef TRACE_SYSTEM_VAR -#define TRACE_SYSTEM_VAR rdmavt - -#if !defined(__RDMAVT_TRACE_H) || defined(TRACE_HEADER_MULTI_READ) -#define __RDMAVT_TRACE_H - -#include <linux/tracepoint.h> -#include <linux/trace_seq.h> - -#include <rdma/ib_verbs.h> -#include <rdma/rdma_vt.h> - #define RDI_DEV_ENTRY(rdi) __string(dev, rdi->driver_f.get_card_name(rdi)) #define RDI_DEV_ASSIGN(rdi) __assign_str(dev, rdi->driver_f.get_card_name(rdi)) -#undef TRACE_SYSTEM -#define TRACE_SYSTEM rdmavt - -TRACE_EVENT(rvt_dbg, - TP_PROTO(struct rvt_dev_info *rdi, - const char *msg), - TP_ARGS(rdi, msg), - TP_STRUCT__entry( - RDI_DEV_ENTRY(rdi) - __string(msg, msg) - ), - TP_fast_assign( - RDI_DEV_ASSIGN(rdi); - __assign_str(msg, msg); - ), - TP_printk("[%s]: %s", __get_str(dev), __get_str(msg)) -); - -#undef TRACE_SYSTEM -#define TRACE_SYSTEM rvt_qphash -DECLARE_EVENT_CLASS(rvt_qphash_template, - TP_PROTO(struct rvt_qp *qp, u32 bucket), - TP_ARGS(qp, bucket), - TP_STRUCT__entry( - RDI_DEV_ENTRY(ib_to_rvt(qp->ibqp.device)) - __field(u32, qpn) - __field(u32, bucket) - ), - TP_fast_assign( - RDI_DEV_ASSIGN(ib_to_rvt(qp->ibqp.device)) - __entry->qpn = qp->ibqp.qp_num; - __entry->bucket = bucket; - ), - TP_printk( - "[%s] qpn 0x%x bucket %u", - __get_str(dev), - __entry->qpn, - __entry->bucket - ) -); - -DEFINE_EVENT(rvt_qphash_template, rvt_qpinsert, - TP_PROTO(struct rvt_qp *qp, u32 bucket), - TP_ARGS(qp, bucket)); - -DEFINE_EVENT(rvt_qphash_template, rvt_qpremove, - TP_PROTO(struct rvt_qp *qp, u32 bucket), - TP_ARGS(qp, bucket)); - -#undef TRACE_SYSTEM -#define TRACE_SYSTEM rvt_tx - -#define wr_opcode_name(opcode) { IB_WR_##opcode, #opcode } -#define show_wr_opcode(opcode) \ -__print_symbolic(opcode, \ - wr_opcode_name(RDMA_WRITE), \ - wr_opcode_name(RDMA_WRITE_WITH_IMM), \ - wr_opcode_name(SEND), \ - wr_opcode_name(SEND_WITH_IMM), \ - wr_opcode_name(RDMA_READ), \ - wr_opcode_name(ATOMIC_CMP_AND_SWP), \ - wr_opcode_name(ATOMIC_FETCH_AND_ADD), \ - wr_opcode_name(LSO), \ - wr_opcode_name(SEND_WITH_INV), \ - wr_opcode_name(RDMA_READ_WITH_INV), \ - wr_opcode_name(LOCAL_INV), \ - wr_opcode_name(MASKED_ATOMIC_CMP_AND_SWP), \ - wr_opcode_name(MASKED_ATOMIC_FETCH_AND_ADD)) - -#define POS_PRN \ -"[%s] wr_id %llx qpn %x psn 0x%x lpsn 0x%x length %u opcode 0x%.2x,%s size %u avail %u head %u last %u" - -TRACE_EVENT( - rvt_post_one_wr, - TP_PROTO(struct rvt_qp *qp, struct rvt_swqe *wqe), - TP_ARGS(qp, wqe), - TP_STRUCT__entry( - RDI_DEV_ENTRY(ib_to_rvt(qp->ibqp.device)) - __field(u64, wr_id) - __field(u32, qpn) - __field(u32, psn) - __field(u32, lpsn) - __field(u32, length) - __field(u32, opcode) - __field(u32, size) - __field(u32, avail) - __field(u32, head) - __field(u32, last) - ), - TP_fast_assign( - RDI_DEV_ASSIGN(ib_to_rvt(qp->ibqp.device)) - __entry->wr_id = wqe->wr.wr_id; - __entry->qpn = qp->ibqp.qp_num; - __entry->psn = wqe->psn; - __entry->lpsn = wqe->lpsn; - __entry->length = wqe->length; - __entry->opcode = wqe->wr.opcode; - __entry->size = qp->s_size; - __entry->avail = qp->s_avail; - __entry->head = qp->s_head; - __entry->last = qp->s_last; - ), - TP_printk( - POS_PRN, - __get_str(dev), - __entry->wr_id, - __entry->qpn, - __entry->psn, - __entry->lpsn, - __entry->length, - __entry->opcode, show_wr_opcode(__entry->opcode), - __entry->size, - __entry->avail, - __entry->head, - __entry->last - ) -); - -#endif /* __RDMAVT_TRACE_H */ - -#undef TRACE_INCLUDE_PATH -#undef TRACE_INCLUDE_FILE -#define TRACE_INCLUDE_PATH . -#define TRACE_INCLUDE_FILE trace -#include <trace/define_trace.h> +#include "trace_rvt.h" +#include "trace_qp.h" +#include "trace_tx.h" +#include "trace_mr.h" diff --git a/drivers/infiniband/sw/rdmavt/trace_mr.h b/drivers/infiniband/sw/rdmavt/trace_mr.h new file mode 100644 index 000000000000..3318a6c36373 --- /dev/null +++ b/drivers/infiniband/sw/rdmavt/trace_mr.h @@ -0,0 +1,112 @@ +/* + * Copyright(c) 2016 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * 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. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ +#if !defined(__RVT_TRACE_MR_H) || defined(TRACE_HEADER_MULTI_READ) +#define __RVT_TRACE_MR_H + +#include <linux/tracepoint.h> +#include <linux/trace_seq.h> + +#include <rdma/ib_verbs.h> +#include <rdma/rdma_vt.h> +#include <rdma/rdmavt_mr.h> + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM rvt_mr +DECLARE_EVENT_CLASS( + rvt_mr_template, + TP_PROTO(struct rvt_mregion *mr, u16 m, u16 n, void *v, size_t len), + TP_ARGS(mr, m, n, v, len), + TP_STRUCT__entry( + RDI_DEV_ENTRY(ib_to_rvt(mr->pd->device)) + __field(void *, vaddr) + __field(struct page *, page) + __field(size_t, len) + __field(u32, lkey) + __field(u16, m) + __field(u16, n) + ), + TP_fast_assign( + RDI_DEV_ASSIGN(ib_to_rvt(mr->pd->device)); + __entry->vaddr = v; + __entry->page = virt_to_page(v); + __entry->m = m; + __entry->n = n; + __entry->len = len; + ), + TP_printk( + "[%s] vaddr %p page %p m %u n %u len %ld", + __get_str(dev), + __entry->vaddr, + __entry->page, + __entry->m, + __entry->n, + __entry->len + ) +); + +DEFINE_EVENT( + rvt_mr_template, rvt_mr_page_seg, + TP_PROTO(struct rvt_mregion *mr, u16 m, u16 n, void *v, size_t len), + TP_ARGS(mr, m, n, v, len)); + +DEFINE_EVENT( + rvt_mr_template, rvt_mr_fmr_seg, + TP_PROTO(struct rvt_mregion *mr, u16 m, u16 n, void *v, size_t len), + TP_ARGS(mr, m, n, v, len)); + +DEFINE_EVENT( + rvt_mr_template, rvt_mr_user_seg, + TP_PROTO(struct rvt_mregion *mr, u16 m, u16 n, void *v, size_t len), + TP_ARGS(mr, m, n, v, len)); + +#endif /* __RVT_TRACE_MR_H */ + +#undef TRACE_INCLUDE_PATH +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_PATH . +#define TRACE_INCLUDE_FILE trace_mr +#include <trace/define_trace.h> diff --git a/drivers/infiniband/sw/rdmavt/trace_qp.h b/drivers/infiniband/sw/rdmavt/trace_qp.h new file mode 100644 index 000000000000..4c77a3119bda --- /dev/null +++ b/drivers/infiniband/sw/rdmavt/trace_qp.h @@ -0,0 +1,96 @@ +/* + * Copyright(c) 2016 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * 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. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ +#if !defined(__RVT_TRACE_QP_H) || defined(TRACE_HEADER_MULTI_READ) +#define __RVT_TRACE_QP_H + +#include <linux/tracepoint.h> +#include <linux/trace_seq.h> + +#include <rdma/ib_verbs.h> +#include <rdma/rdma_vt.h> + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM rvt_qp + +DECLARE_EVENT_CLASS(rvt_qphash_template, + TP_PROTO(struct rvt_qp *qp, u32 bucket), + TP_ARGS(qp, bucket), + TP_STRUCT__entry( + RDI_DEV_ENTRY(ib_to_rvt(qp->ibqp.device)) + __field(u32, qpn) + __field(u32, bucket) + ), + TP_fast_assign( + RDI_DEV_ASSIGN(ib_to_rvt(qp->ibqp.device)) + __entry->qpn = qp->ibqp.qp_num; + __entry->bucket = bucket; + ), + TP_printk( + "[%s] qpn 0x%x bucket %u", + __get_str(dev), + __entry->qpn, + __entry->bucket + ) +); + +DEFINE_EVENT(rvt_qphash_template, rvt_qpinsert, + TP_PROTO(struct rvt_qp *qp, u32 bucket), + TP_ARGS(qp, bucket)); + +DEFINE_EVENT(rvt_qphash_template, rvt_qpremove, + TP_PROTO(struct rvt_qp *qp, u32 bucket), + TP_ARGS(qp, bucket)); + + +#endif /* __RVT_TRACE_QP_H */ + +#undef TRACE_INCLUDE_PATH +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_PATH . +#define TRACE_INCLUDE_FILE trace_qp +#include <trace/define_trace.h> + diff --git a/drivers/infiniband/sw/rdmavt/dma.h b/drivers/infiniband/sw/rdmavt/trace_rvt.h index 979f07e09195..746f33461d9a 100644 --- a/drivers/infiniband/sw/rdmavt/dma.h +++ b/drivers/infiniband/sw/rdmavt/trace_rvt.h @@ -1,6 +1,3 @@ -#ifndef DEF_RDMAVTDMA_H -#define DEF_RDMAVTDMA_H - /* * Copyright(c) 2016 Intel Corporation. * @@ -47,7 +44,38 @@ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. * */ +#if !defined(__RVT_TRACE_RVT_H) || defined(TRACE_HEADER_MULTI_READ) +#define __RVT_TRACE_RVT_H + +#include <linux/tracepoint.h> +#include <linux/trace_seq.h> + +#include <rdma/ib_verbs.h> +#include <rdma/rdma_vt.h> + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM rvt + +TRACE_EVENT(rvt_dbg, + TP_PROTO(struct rvt_dev_info *rdi, + const char *msg), + TP_ARGS(rdi, msg), + TP_STRUCT__entry( + RDI_DEV_ENTRY(rdi) + __string(msg, msg) + ), + TP_fast_assign( + RDI_DEV_ASSIGN(rdi); + __assign_str(msg, msg); + ), + TP_printk("[%s]: %s", __get_str(dev), __get_str(msg)) +); + +#endif /* __RVT_TRACE_MISC_H */ -extern struct ib_dma_mapping_ops rvt_default_dma_mapping_ops; +#undef TRACE_INCLUDE_PATH +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_PATH . +#define TRACE_INCLUDE_FILE trace_rvt +#include <trace/define_trace.h> -#endif /* DEF_RDMAVTDMA_H */ diff --git a/drivers/infiniband/sw/rdmavt/trace_tx.h b/drivers/infiniband/sw/rdmavt/trace_tx.h new file mode 100644 index 000000000000..0e03173662d8 --- /dev/null +++ b/drivers/infiniband/sw/rdmavt/trace_tx.h @@ -0,0 +1,132 @@ +/* + * Copyright(c) 2016 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * 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. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ +#if !defined(__RVT_TRACE_TX_H) || defined(TRACE_HEADER_MULTI_READ) +#define __RVT_TRACE_TX_H + +#include <linux/tracepoint.h> +#include <linux/trace_seq.h> + +#include <rdma/ib_verbs.h> +#include <rdma/rdma_vt.h> + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM rvt_tx + +#define wr_opcode_name(opcode) { IB_WR_##opcode, #opcode } +#define show_wr_opcode(opcode) \ +__print_symbolic(opcode, \ + wr_opcode_name(RDMA_WRITE), \ + wr_opcode_name(RDMA_WRITE_WITH_IMM), \ + wr_opcode_name(SEND), \ + wr_opcode_name(SEND_WITH_IMM), \ + wr_opcode_name(RDMA_READ), \ + wr_opcode_name(ATOMIC_CMP_AND_SWP), \ + wr_opcode_name(ATOMIC_FETCH_AND_ADD), \ + wr_opcode_name(LSO), \ + wr_opcode_name(SEND_WITH_INV), \ + wr_opcode_name(RDMA_READ_WITH_INV), \ + wr_opcode_name(LOCAL_INV), \ + wr_opcode_name(MASKED_ATOMIC_CMP_AND_SWP), \ + wr_opcode_name(MASKED_ATOMIC_FETCH_AND_ADD)) + +#define POS_PRN \ +"[%s] wr_id %llx qpn %x psn 0x%x lpsn 0x%x length %u opcode 0x%.2x,%s size %u avail %u head %u last %u" + +TRACE_EVENT( + rvt_post_one_wr, + TP_PROTO(struct rvt_qp *qp, struct rvt_swqe *wqe), + TP_ARGS(qp, wqe), + TP_STRUCT__entry( + RDI_DEV_ENTRY(ib_to_rvt(qp->ibqp.device)) + __field(u64, wr_id) + __field(u32, qpn) + __field(u32, psn) + __field(u32, lpsn) + __field(u32, length) + __field(u32, opcode) + __field(u32, size) + __field(u32, avail) + __field(u32, head) + __field(u32, last) + ), + TP_fast_assign( + RDI_DEV_ASSIGN(ib_to_rvt(qp->ibqp.device)) + __entry->wr_id = wqe->wr.wr_id; + __entry->qpn = qp->ibqp.qp_num; + __entry->psn = wqe->psn; + __entry->lpsn = wqe->lpsn; + __entry->length = wqe->length; + __entry->opcode = wqe->wr.opcode; + __entry->size = qp->s_size; + __entry->avail = qp->s_avail; + __entry->head = qp->s_head; + __entry->last = qp->s_last; + ), + TP_printk( + POS_PRN, + __get_str(dev), + __entry->wr_id, + __entry->qpn, + __entry->psn, + __entry->lpsn, + __entry->length, + __entry->opcode, show_wr_opcode(__entry->opcode), + __entry->size, + __entry->avail, + __entry->head, + __entry->last + ) +); + +#endif /* __RVT_TRACE_TX_H */ + +#undef TRACE_INCLUDE_PATH +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_PATH . +#define TRACE_INCLUDE_FILE trace_tx +#include <trace/define_trace.h> + diff --git a/drivers/infiniband/sw/rdmavt/vt.c b/drivers/infiniband/sw/rdmavt/vt.c index d430c2f7cec4..0d7c6bb551d9 100644 --- a/drivers/infiniband/sw/rdmavt/vt.c +++ b/drivers/infiniband/sw/rdmavt/vt.c @@ -47,6 +47,7 @@ #include <linux/module.h> #include <linux/kernel.h> +#include <linux/dma-mapping.h> #include "vt.h" #include "trace.h" @@ -165,7 +166,7 @@ static int rvt_query_port(struct ib_device *ibdev, u8 port_num, return -EINVAL; rvp = rdi->ports[port_index]; - memset(props, 0, sizeof(*props)); + /* props being zeroed by the caller, avoid zeroing it here */ props->sm_lid = rvp->sm_lid; props->sm_sl = rvp->sm_sl; props->port_cap_flags = rvp->port_cap_flags; @@ -326,13 +327,14 @@ static int rvt_get_port_immutable(struct ib_device *ibdev, u8 port_num, if (port_index < 0) return -EINVAL; - err = rvt_query_port(ibdev, port_num, &attr); + immutable->core_cap_flags = rdi->dparms.core_cap_flags; + + err = ib_query_port(ibdev, port_num, &attr); if (err) return err; immutable->pkey_tbl_len = attr.pkey_tbl_len; immutable->gid_tbl_len = attr.gid_tbl_len; - immutable->core_cap_flags = rdi->dparms.core_cap_flags; immutable->max_mad_size = rdi->dparms.max_mad_size; return 0; @@ -777,8 +779,7 @@ int rvt_register_device(struct rvt_dev_info *rdi) } /* DMA Operations */ - rdi->ibdev.dma_ops = - rdi->ibdev.dma_ops ? : &rvt_default_dma_mapping_ops; + rdi->ibdev.dev.dma_ops = rdi->ibdev.dev.dma_ops ? : &dma_virt_ops; /* Protection Domain */ spin_lock_init(&rdi->n_pds_lock); diff --git a/drivers/infiniband/sw/rdmavt/vt.h b/drivers/infiniband/sw/rdmavt/vt.h index 6b01eaa4461b..f363505312be 100644 --- a/drivers/infiniband/sw/rdmavt/vt.h +++ b/drivers/infiniband/sw/rdmavt/vt.h @@ -50,7 +50,6 @@ #include <rdma/rdma_vt.h> #include <linux/pci.h> -#include "dma.h" #include "pd.h" #include "qp.h" #include "ah.h" diff --git a/drivers/infiniband/sw/rxe/Kconfig b/drivers/infiniband/sw/rxe/Kconfig index 1e4e628fe7b0..7d1ac27ed251 100644 --- a/drivers/infiniband/sw/rxe/Kconfig +++ b/drivers/infiniband/sw/rxe/Kconfig @@ -2,6 +2,7 @@ config RDMA_RXE tristate "Software RDMA over Ethernet (RoCE) driver" depends on INET && PCI && INFINIBAND depends on NET_UDP_TUNNEL + select DMA_VIRT_OPS ---help--- This driver implements the InfiniBand RDMA transport over the Linux network stack. It enables a system with a diff --git a/drivers/infiniband/sw/rxe/Makefile b/drivers/infiniband/sw/rxe/Makefile index 3b3fb9d1c470..ec35ff022a42 100644 --- a/drivers/infiniband/sw/rxe/Makefile +++ b/drivers/infiniband/sw/rxe/Makefile @@ -14,7 +14,6 @@ rdma_rxe-y := \ rxe_qp.o \ rxe_cq.o \ rxe_mr.o \ - rxe_dma.o \ rxe_opcode.o \ rxe_mmap.o \ rxe_icrc.o \ diff --git a/drivers/infiniband/sw/rxe/rxe.c b/drivers/infiniband/sw/rxe/rxe.c index ab6c3c25d7ff..b12dd9b5a89d 100644 --- a/drivers/infiniband/sw/rxe/rxe.c +++ b/drivers/infiniband/sw/rxe/rxe.c @@ -178,7 +178,7 @@ static int rxe_init_ports(struct rxe_dev *rxe) return -ENOMEM; port->pkey_tbl[0] = 0xffff; - port->port_guid = rxe->ifc_ops->port_guid(rxe); + port->port_guid = rxe_port_guid(rxe); spin_lock_init(&port->port_lock); diff --git a/drivers/infiniband/sw/rxe/rxe_comp.c b/drivers/infiniband/sw/rxe/rxe_comp.c index 6c5e29db88e3..4cd55d5617f7 100644 --- a/drivers/infiniband/sw/rxe/rxe_comp.c +++ b/drivers/infiniband/sw/rxe/rxe_comp.c @@ -224,7 +224,7 @@ static inline enum comp_state check_psn(struct rxe_qp *qp, else return COMPST_DONE; } else if ((diff > 0) && (wqe->mask & WR_ATOMIC_OR_READ_MASK)) { - return COMPST_ERROR_RETRY; + return COMPST_DONE; } else { return COMPST_CHECK_ACK; } @@ -254,7 +254,7 @@ static inline enum comp_state check_ack(struct rxe_qp *qp, } break; default: - WARN_ON(1); + WARN_ON_ONCE(1); } /* Check operation validity. */ @@ -412,19 +412,28 @@ static void make_send_cqe(struct rxe_qp *qp, struct rxe_send_wqe *wqe, } } +/* + * IBA Spec. Section 10.7.3.1 SIGNALED COMPLETIONS + * ---------8<---------8<------------- + * ...Note that if a completion error occurs, a Work Completion + * will always be generated, even if the signaling + * indicator requests an Unsignaled Completion. + * ---------8<---------8<------------- + */ static void do_complete(struct rxe_qp *qp, struct rxe_send_wqe *wqe) { struct rxe_cqe cqe; if ((qp->sq_sig_type == IB_SIGNAL_ALL_WR) || (wqe->wr.send_flags & IB_SEND_SIGNALED) || - (qp->req.state == QP_STATE_ERROR)) { + wqe->status != IB_WC_SUCCESS) { make_send_cqe(qp, wqe, &cqe); + advance_consumer(qp->sq.queue); rxe_cq_post(qp->scq, &cqe, 0); + } else { + advance_consumer(qp->sq.queue); } - advance_consumer(qp->sq.queue); - /* * we completed something so let req run again * if it is trying to fence @@ -502,55 +511,40 @@ static inline enum comp_state complete_wqe(struct rxe_qp *qp, return COMPST_GET_WQE; } -int rxe_completer(void *arg) +static void rxe_drain_resp_pkts(struct rxe_qp *qp, bool notify) { - struct rxe_qp *qp = (struct rxe_qp *)arg; - struct rxe_send_wqe *wqe = wqe; - struct sk_buff *skb = NULL; - struct rxe_pkt_info *pkt = NULL; - enum comp_state state; - - if (!qp->valid) { - while ((skb = skb_dequeue(&qp->resp_pkts))) { - rxe_drop_ref(qp); - kfree_skb(skb); - } - skb = NULL; - pkt = NULL; - - while (queue_head(qp->sq.queue)) - advance_consumer(qp->sq.queue); + struct sk_buff *skb; + struct rxe_send_wqe *wqe; - goto exit; + while ((skb = skb_dequeue(&qp->resp_pkts))) { + rxe_drop_ref(qp); + kfree_skb(skb); } - if (qp->req.state == QP_STATE_ERROR) { - while ((skb = skb_dequeue(&qp->resp_pkts))) { - rxe_drop_ref(qp); - kfree_skb(skb); - } - skb = NULL; - pkt = NULL; - - while ((wqe = queue_head(qp->sq.queue))) { + while ((wqe = queue_head(qp->sq.queue))) { + if (notify) { wqe->status = IB_WC_WR_FLUSH_ERR; do_complete(qp, wqe); + } else { + advance_consumer(qp->sq.queue); } - - goto exit; } +} - if (qp->req.state == QP_STATE_RESET) { - while ((skb = skb_dequeue(&qp->resp_pkts))) { - rxe_drop_ref(qp); - kfree_skb(skb); - } - skb = NULL; - pkt = NULL; +int rxe_completer(void *arg) +{ + struct rxe_qp *qp = (struct rxe_qp *)arg; + struct rxe_send_wqe *wqe = wqe; + struct sk_buff *skb = NULL; + struct rxe_pkt_info *pkt = NULL; + enum comp_state state; - while (queue_head(qp->sq.queue)) - advance_consumer(qp->sq.queue); + rxe_add_ref(qp); + if (!qp->valid || qp->req.state == QP_STATE_ERROR || + qp->req.state == QP_STATE_RESET) { + rxe_drain_resp_pkts(qp, qp->valid && + qp->req.state == QP_STATE_ERROR); goto exit; } @@ -636,6 +630,7 @@ int rxe_completer(void *arg) if (pkt) { rxe_drop_ref(pkt->qp); kfree_skb(skb); + skb = NULL; } goto done; @@ -659,6 +654,7 @@ int rxe_completer(void *arg) qp->qp_timeout_jiffies) mod_timer(&qp->retrans_timer, jiffies + qp->qp_timeout_jiffies); + WARN_ON_ONCE(skb); goto exit; case COMPST_ERROR_RETRY: @@ -671,8 +667,10 @@ int rxe_completer(void *arg) */ /* there is nothing to retry in this case */ - if (!wqe || (wqe->state == wqe_state_posted)) + if (!wqe || (wqe->state == wqe_state_posted)) { + WARN_ON_ONCE(skb); goto exit; + } if (qp->comp.retry_cnt > 0) { if (qp->comp.retry_cnt != 7) @@ -694,8 +692,10 @@ int rxe_completer(void *arg) if (pkt) { rxe_drop_ref(pkt->qp); kfree_skb(skb); + skb = NULL; } + WARN_ON_ONCE(skb); goto exit; } else { @@ -715,6 +715,9 @@ int rxe_completer(void *arg) mod_timer(&qp->rnr_nak_timer, jiffies + rnrnak_jiffies(aeth_syn(pkt) & ~AETH_TYPE_MASK)); + rxe_drop_ref(pkt->qp); + kfree_skb(skb); + skb = NULL; goto exit; } else { wqe->status = IB_WC_RNR_RETRY_EXC_ERR; @@ -723,14 +726,17 @@ int rxe_completer(void *arg) break; case COMPST_ERROR: + WARN_ON_ONCE(wqe->status == IB_WC_SUCCESS); do_complete(qp, wqe); rxe_qp_error(qp); if (pkt) { rxe_drop_ref(pkt->qp); kfree_skb(skb); + skb = NULL; } + WARN_ON_ONCE(skb); goto exit; } } @@ -739,11 +745,15 @@ exit: /* we come here if we are done with processing and want the task to * exit from the loop calling us */ + WARN_ON_ONCE(skb); + rxe_drop_ref(qp); return -EAGAIN; done: /* we come here if we have processed a packet we want the task to call * us again to see if there is anything else to do */ + WARN_ON_ONCE(skb); + rxe_drop_ref(qp); return 0; } diff --git a/drivers/infiniband/sw/rxe/rxe_cq.c b/drivers/infiniband/sw/rxe/rxe_cq.c index e5e6a5e7dee9..49fe42c23f4d 100644 --- a/drivers/infiniband/sw/rxe/rxe_cq.c +++ b/drivers/infiniband/sw/rxe/rxe_cq.c @@ -156,9 +156,9 @@ int rxe_cq_post(struct rxe_cq *cq, struct rxe_cqe *cqe, int solicited) return 0; } -void rxe_cq_cleanup(void *arg) +void rxe_cq_cleanup(struct rxe_pool_entry *arg) { - struct rxe_cq *cq = arg; + struct rxe_cq *cq = container_of(arg, typeof(*cq), pelem); if (cq->queue) rxe_queue_cleanup(cq->queue); diff --git a/drivers/infiniband/sw/rxe/rxe_dma.c b/drivers/infiniband/sw/rxe/rxe_dma.c deleted file mode 100644 index a0f8af5851ae..000000000000 --- a/drivers/infiniband/sw/rxe/rxe_dma.c +++ /dev/null @@ -1,183 +0,0 @@ -/* - * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. - * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. - * - * This software is available to you under a choice of one of two - * licenses. You may choose to be licensed under the terms of the GNU - * General Public License (GPL) Version 2, available from the file - * COPYING in the main directory of this source tree, or the - * OpenIB.org BSD license below: - * - * Redistribution and use in source and binary forms, with or - * without modification, are permitted provided that the following - * conditions are met: - * - * - Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. - * - * - Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, - * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF - * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND - * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS - * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN - * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN - * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. - */ - -#include "rxe.h" -#include "rxe_loc.h" - -#define DMA_BAD_ADDER ((u64)0) - -static int rxe_mapping_error(struct ib_device *dev, u64 dma_addr) -{ - return dma_addr == DMA_BAD_ADDER; -} - -static u64 rxe_dma_map_single(struct ib_device *dev, - void *cpu_addr, size_t size, - enum dma_data_direction direction) -{ - WARN_ON(!valid_dma_direction(direction)); - return (uintptr_t)cpu_addr; -} - -static void rxe_dma_unmap_single(struct ib_device *dev, - u64 addr, size_t size, - enum dma_data_direction direction) -{ - WARN_ON(!valid_dma_direction(direction)); -} - -static u64 rxe_dma_map_page(struct ib_device *dev, - struct page *page, - unsigned long offset, - size_t size, enum dma_data_direction direction) -{ - u64 addr; - - WARN_ON(!valid_dma_direction(direction)); - - if (offset + size > PAGE_SIZE) { - addr = DMA_BAD_ADDER; - goto done; - } - - addr = (uintptr_t)page_address(page); - if (addr) - addr += offset; - -done: - return addr; -} - -static void rxe_dma_unmap_page(struct ib_device *dev, - u64 addr, size_t size, - enum dma_data_direction direction) -{ - WARN_ON(!valid_dma_direction(direction)); -} - -static int rxe_map_sg(struct ib_device *dev, struct scatterlist *sgl, - int nents, enum dma_data_direction direction) -{ - struct scatterlist *sg; - u64 addr; - int i; - int ret = nents; - - WARN_ON(!valid_dma_direction(direction)); - - for_each_sg(sgl, sg, nents, i) { - addr = (uintptr_t)page_address(sg_page(sg)); - if (!addr) { - ret = 0; - break; - } - sg->dma_address = addr + sg->offset; -#ifdef CONFIG_NEED_SG_DMA_LENGTH - sg->dma_length = sg->length; -#endif - } - - return ret; -} - -static void rxe_unmap_sg(struct ib_device *dev, - struct scatterlist *sg, int nents, - enum dma_data_direction direction) -{ - WARN_ON(!valid_dma_direction(direction)); -} - -static int rxe_map_sg_attrs(struct ib_device *dev, struct scatterlist *sgl, - int nents, enum dma_data_direction direction, - unsigned long attrs) -{ - return rxe_map_sg(dev, sgl, nents, direction); -} - -static void rxe_unmap_sg_attrs(struct ib_device *dev, - struct scatterlist *sg, int nents, - enum dma_data_direction direction, - unsigned long attrs) -{ - rxe_unmap_sg(dev, sg, nents, direction); -} - -static void rxe_sync_single_for_cpu(struct ib_device *dev, - u64 addr, - size_t size, enum dma_data_direction dir) -{ -} - -static void rxe_sync_single_for_device(struct ib_device *dev, - u64 addr, - size_t size, enum dma_data_direction dir) -{ -} - -static void *rxe_dma_alloc_coherent(struct ib_device *dev, size_t size, - u64 *dma_handle, gfp_t flag) -{ - struct page *p; - void *addr = NULL; - - p = alloc_pages(flag, get_order(size)); - if (p) - addr = page_address(p); - - if (dma_handle) - *dma_handle = (uintptr_t)addr; - - return addr; -} - -static void rxe_dma_free_coherent(struct ib_device *dev, size_t size, - void *cpu_addr, u64 dma_handle) -{ - free_pages((unsigned long)cpu_addr, get_order(size)); -} - -struct ib_dma_mapping_ops rxe_dma_mapping_ops = { - .mapping_error = rxe_mapping_error, - .map_single = rxe_dma_map_single, - .unmap_single = rxe_dma_unmap_single, - .map_page = rxe_dma_map_page, - .unmap_page = rxe_dma_unmap_page, - .map_sg = rxe_map_sg, - .unmap_sg = rxe_unmap_sg, - .map_sg_attrs = rxe_map_sg_attrs, - .unmap_sg_attrs = rxe_unmap_sg_attrs, - .sync_single_for_cpu = rxe_sync_single_for_cpu, - .sync_single_for_device = rxe_sync_single_for_device, - .alloc_coherent = rxe_dma_alloc_coherent, - .free_coherent = rxe_dma_free_coherent -}; diff --git a/drivers/infiniband/sw/rxe/rxe_hdr.h b/drivers/infiniband/sw/rxe/rxe_hdr.h index d57b5e956ceb..6cb18406f5b8 100644 --- a/drivers/infiniband/sw/rxe/rxe_hdr.h +++ b/drivers/infiniband/sw/rxe/rxe_hdr.h @@ -53,8 +53,16 @@ struct rxe_pkt_info { }; /* Macros should be used only for received skb */ -#define SKB_TO_PKT(skb) ((struct rxe_pkt_info *)(skb)->cb) -#define PKT_TO_SKB(pkt) container_of((void *)(pkt), struct sk_buff, cb) +static inline struct rxe_pkt_info *SKB_TO_PKT(struct sk_buff *skb) +{ + BUILD_BUG_ON(sizeof(struct rxe_pkt_info) > sizeof(skb->cb)); + return (void *)skb->cb; +} + +static inline struct sk_buff *PKT_TO_SKB(struct rxe_pkt_info *pkt) +{ + return container_of((void *)pkt, struct sk_buff, cb); +} /* * IBA header types and methods diff --git a/drivers/infiniband/sw/rxe/rxe_loc.h b/drivers/infiniband/sw/rxe/rxe_loc.h index 73849a5a91b3..183a9d379b41 100644 --- a/drivers/infiniband/sw/rxe/rxe_loc.h +++ b/drivers/infiniband/sw/rxe/rxe_loc.h @@ -64,7 +64,7 @@ int rxe_cq_resize_queue(struct rxe_cq *cq, int new_cqe, struct ib_udata *udata); int rxe_cq_post(struct rxe_cq *cq, struct rxe_cqe *cqe, int solicited); -void rxe_cq_cleanup(void *arg); +void rxe_cq_cleanup(struct rxe_pool_entry *arg); /* rxe_mcast.c */ int rxe_mcast_get_grp(struct rxe_dev *rxe, union ib_gid *mgid, @@ -78,7 +78,7 @@ int rxe_mcast_drop_grp_elem(struct rxe_dev *rxe, struct rxe_qp *qp, void rxe_drop_all_mcast_groups(struct rxe_qp *qp); -void rxe_mc_cleanup(void *arg); +void rxe_mc_cleanup(struct rxe_pool_entry *arg); /* rxe_mmap.c */ struct rxe_mmap_info { @@ -137,10 +137,26 @@ int mem_check_range(struct rxe_mem *mem, u64 iova, size_t length); int rxe_mem_map_pages(struct rxe_dev *rxe, struct rxe_mem *mem, u64 *page, int num_pages, u64 iova); -void rxe_mem_cleanup(void *arg); +void rxe_mem_cleanup(struct rxe_pool_entry *arg); int advance_dma_data(struct rxe_dma_info *dma, unsigned int length); +/* rxe_net.c */ +int rxe_loopback(struct sk_buff *skb); +int rxe_send(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, + struct sk_buff *skb); +__be64 rxe_port_guid(struct rxe_dev *rxe); +struct sk_buff *rxe_init_packet(struct rxe_dev *rxe, struct rxe_av *av, + int paylen, struct rxe_pkt_info *pkt); +int rxe_prepare(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, + struct sk_buff *skb, u32 *crc); +enum rdma_link_layer rxe_link_layer(struct rxe_dev *rxe, unsigned int port_num); +const char *rxe_parent_name(struct rxe_dev *rxe, unsigned int port_num); +struct device *rxe_dma_device(struct rxe_dev *rxe); +__be64 rxe_node_guid(struct rxe_dev *rxe); +int rxe_mcast_add(struct rxe_dev *rxe, union ib_gid *mgid); +int rxe_mcast_delete(struct rxe_dev *rxe, union ib_gid *mgid); + /* rxe_qp.c */ int rxe_qp_chk_init(struct rxe_dev *rxe, struct ib_qp_init_attr *init); @@ -162,7 +178,7 @@ void rxe_qp_error(struct rxe_qp *qp); void rxe_qp_destroy(struct rxe_qp *qp); -void rxe_qp_cleanup(void *arg); +void rxe_qp_cleanup(struct rxe_pool_entry *arg); static inline int qp_num(struct rxe_qp *qp) { @@ -221,10 +237,9 @@ int rxe_srq_from_attr(struct rxe_dev *rxe, struct rxe_srq *srq, struct ib_srq_attr *attr, enum ib_srq_attr_mask mask, struct ib_udata *udata); -extern struct ib_dma_mapping_ops rxe_dma_mapping_ops; - void rxe_release(struct kref *kref); +void rxe_drain_req_pkts(struct rxe_qp *qp, bool notify); int rxe_completer(void *arg); int rxe_requester(void *arg); int rxe_responder(void *arg); @@ -256,9 +271,9 @@ static inline int rxe_xmit_packet(struct rxe_dev *rxe, struct rxe_qp *qp, if (pkt->mask & RXE_LOOPBACK_MASK) { memcpy(SKB_TO_PKT(skb), pkt, sizeof(*pkt)); - err = rxe->ifc_ops->loopback(skb); + err = rxe_loopback(skb); } else { - err = rxe->ifc_ops->send(rxe, pkt, skb); + err = rxe_send(rxe, pkt, skb); } if (err) { @@ -266,8 +281,6 @@ static inline int rxe_xmit_packet(struct rxe_dev *rxe, struct rxe_qp *qp, return err; } - atomic_inc(&qp->skb_out); - if ((qp_type(qp) != IB_QPT_RC) && (pkt->mask & RXE_END_MASK)) { pkt->wqe->state = wqe_state_done; diff --git a/drivers/infiniband/sw/rxe/rxe_mcast.c b/drivers/infiniband/sw/rxe/rxe_mcast.c index fa95544ca7e0..522a7942c56c 100644 --- a/drivers/infiniband/sw/rxe/rxe_mcast.c +++ b/drivers/infiniband/sw/rxe/rxe_mcast.c @@ -61,7 +61,7 @@ int rxe_mcast_get_grp(struct rxe_dev *rxe, union ib_gid *mgid, rxe_add_key(grp, mgid); - err = rxe->ifc_ops->mcast_add(rxe, mgid); + err = rxe_mcast_add(rxe, mgid); if (err) goto err2; @@ -180,11 +180,11 @@ void rxe_drop_all_mcast_groups(struct rxe_qp *qp) } } -void rxe_mc_cleanup(void *arg) +void rxe_mc_cleanup(struct rxe_pool_entry *arg) { - struct rxe_mc_grp *grp = arg; + struct rxe_mc_grp *grp = container_of(arg, typeof(*grp), pelem); struct rxe_dev *rxe = grp->rxe; rxe_drop_key(grp); - rxe->ifc_ops->mcast_delete(rxe, &grp->mgid); + rxe_mcast_delete(rxe, &grp->mgid); } diff --git a/drivers/infiniband/sw/rxe/rxe_mr.c b/drivers/infiniband/sw/rxe/rxe_mr.c index 1869152f1d23..37eea7441ca4 100644 --- a/drivers/infiniband/sw/rxe/rxe_mr.c +++ b/drivers/infiniband/sw/rxe/rxe_mr.c @@ -59,9 +59,11 @@ int mem_check_range(struct rxe_mem *mem, u64 iova, size_t length) case RXE_MEM_TYPE_MR: case RXE_MEM_TYPE_FMR: - return ((iova < mem->iova) || - ((iova + length) > (mem->iova + mem->length))) ? - -EFAULT : 0; + if (iova < mem->iova || + length > mem->length || + iova > mem->iova + mem->length - length) + return -EFAULT; + return 0; default: return -EFAULT; @@ -89,9 +91,9 @@ static void rxe_mem_init(int access, struct rxe_mem *mem) mem->map_shift = ilog2(RXE_BUF_PER_MAP); } -void rxe_mem_cleanup(void *arg) +void rxe_mem_cleanup(struct rxe_pool_entry *arg) { - struct rxe_mem *mem = arg; + struct rxe_mem *mem = container_of(arg, typeof(*mem), pelem); int i; if (mem->umem) @@ -123,7 +125,7 @@ static int rxe_mem_alloc(struct rxe_dev *rxe, struct rxe_mem *mem, int num_buf) goto err2; } - WARN_ON(!is_power_of_2(RXE_BUF_PER_MAP)); + BUILD_BUG_ON(!is_power_of_2(RXE_BUF_PER_MAP)); mem->map_shift = ilog2(RXE_BUF_PER_MAP); mem->map_mask = RXE_BUF_PER_MAP - 1; @@ -189,7 +191,7 @@ int rxe_mem_init_user(struct rxe_dev *rxe, struct rxe_pd *pd, u64 start, goto err1; } - WARN_ON(!is_power_of_2(umem->page_size)); + WARN_ON_ONCE(!is_power_of_2(umem->page_size)); mem->page_shift = ilog2(umem->page_size); mem->page_mask = umem->page_size - 1; @@ -355,6 +357,9 @@ int rxe_mem_copy(struct rxe_mem *mem, u64 iova, void *addr, int length, size_t offset; u32 crc = crcp ? (*crcp) : 0; + if (length == 0) + return 0; + if (mem->type == RXE_MEM_TYPE_DMA) { u8 *src, *dest; @@ -372,7 +377,7 @@ int rxe_mem_copy(struct rxe_mem *mem, u64 iova, void *addr, int length, return 0; } - WARN_ON(!mem->map); + WARN_ON_ONCE(!mem->map); err = mem_check_range(mem, iova, length); if (err) { diff --git a/drivers/infiniband/sw/rxe/rxe_net.c b/drivers/infiniband/sw/rxe/rxe_net.c index ffff5a54cb34..d8610960630a 100644 --- a/drivers/infiniband/sw/rxe/rxe_net.c +++ b/drivers/infiniband/sw/rxe/rxe_net.c @@ -46,7 +46,7 @@ #include "rxe_loc.h" static LIST_HEAD(rxe_dev_list); -static spinlock_t dev_list_lock; /* spinlock for device list */ +static DEFINE_SPINLOCK(dev_list_lock); /* spinlock for device list */ struct rxe_dev *net_to_rxe(struct net_device *ndev) { @@ -102,29 +102,29 @@ static __be64 rxe_mac_to_eui64(struct net_device *ndev) return eui64; } -static __be64 node_guid(struct rxe_dev *rxe) +__be64 rxe_node_guid(struct rxe_dev *rxe) { return rxe_mac_to_eui64(rxe->ndev); } -static __be64 port_guid(struct rxe_dev *rxe) +__be64 rxe_port_guid(struct rxe_dev *rxe) { return rxe_mac_to_eui64(rxe->ndev); } -static struct device *dma_device(struct rxe_dev *rxe) +struct device *rxe_dma_device(struct rxe_dev *rxe) { struct net_device *ndev; ndev = rxe->ndev; - if (ndev->priv_flags & IFF_802_1Q_VLAN) + if (is_vlan_dev(ndev)) ndev = vlan_dev_real_dev(ndev); return ndev->dev.parent; } -static int mcast_add(struct rxe_dev *rxe, union ib_gid *mgid) +int rxe_mcast_add(struct rxe_dev *rxe, union ib_gid *mgid) { int err; unsigned char ll_addr[ETH_ALEN]; @@ -135,7 +135,7 @@ static int mcast_add(struct rxe_dev *rxe, union ib_gid *mgid) return err; } -static int mcast_delete(struct rxe_dev *rxe, union ib_gid *mgid) +int rxe_mcast_delete(struct rxe_dev *rxe, union ib_gid *mgid) { int err; unsigned char ll_addr[ETH_ALEN]; @@ -243,8 +243,8 @@ static struct socket *rxe_setup_udp_tunnel(struct net *net, __be16 port, { int err; struct socket *sock; - struct udp_port_cfg udp_cfg = {0}; - struct udp_tunnel_sock_cfg tnl_cfg = {0}; + struct udp_port_cfg udp_cfg = { }; + struct udp_tunnel_sock_cfg tnl_cfg = { }; if (ipv6) { udp_cfg.family = AF_INET6; @@ -397,8 +397,8 @@ static int prepare6(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, return 0; } -static int prepare(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, - struct sk_buff *skb, u32 *crc) +int rxe_prepare(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, + struct sk_buff *skb, u32 *crc) { int err = 0; struct rxe_av *av = rxe_get_av(pkt); @@ -424,8 +424,7 @@ static void rxe_skb_tx_dtor(struct sk_buff *skb) rxe_run_task(&qp->req.task, 1); } -static int send(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, - struct sk_buff *skb) +int rxe_send(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, struct sk_buff *skb) { struct sk_buff *nskb; struct rxe_av *av; @@ -455,12 +454,13 @@ static int send(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, return -EAGAIN; } + atomic_inc(&pkt->qp->skb_out); kfree_skb(skb); return 0; } -static int loopback(struct sk_buff *skb) +int rxe_loopback(struct sk_buff *skb) { return rxe_rcv(skb); } @@ -470,8 +470,8 @@ static inline int addr_same(struct rxe_dev *rxe, struct rxe_av *av) return rxe->port.port_guid == av->grh.dgid.global.interface_id; } -static struct sk_buff *init_packet(struct rxe_dev *rxe, struct rxe_av *av, - int paylen, struct rxe_pkt_info *pkt) +struct sk_buff *rxe_init_packet(struct rxe_dev *rxe, struct rxe_av *av, + int paylen, struct rxe_pkt_info *pkt) { unsigned int hdr_len; struct sk_buff *skb; @@ -510,31 +510,16 @@ static struct sk_buff *init_packet(struct rxe_dev *rxe, struct rxe_av *av, * this is required by rxe_cfg to match rxe devices in * /sys/class/infiniband up with their underlying ethernet devices */ -static char *parent_name(struct rxe_dev *rxe, unsigned int port_num) +const char *rxe_parent_name(struct rxe_dev *rxe, unsigned int port_num) { return rxe->ndev->name; } -static enum rdma_link_layer link_layer(struct rxe_dev *rxe, - unsigned int port_num) +enum rdma_link_layer rxe_link_layer(struct rxe_dev *rxe, unsigned int port_num) { return IB_LINK_LAYER_ETHERNET; } -static struct rxe_ifc_ops ifc_ops = { - .node_guid = node_guid, - .port_guid = port_guid, - .dma_device = dma_device, - .mcast_add = mcast_add, - .mcast_delete = mcast_delete, - .prepare = prepare, - .send = send, - .loopback = loopback, - .init_packet = init_packet, - .parent_name = parent_name, - .link_layer = link_layer, -}; - struct rxe_dev *rxe_net_add(struct net_device *ndev) { int err; @@ -544,7 +529,6 @@ struct rxe_dev *rxe_net_add(struct net_device *ndev) if (!rxe) return NULL; - rxe->ifc_ops = &ifc_ops; rxe->ndev = ndev; err = rxe_add(rxe, ndev->mtu); @@ -554,7 +538,7 @@ struct rxe_dev *rxe_net_add(struct net_device *ndev) } spin_lock_bh(&dev_list_lock); - list_add_tail(&rxe_dev_list, &rxe->list); + list_add_tail(&rxe->list, &rxe_dev_list); spin_unlock_bh(&dev_list_lock); return rxe; } @@ -657,10 +641,8 @@ struct notifier_block rxe_net_notifier = { .notifier_call = rxe_notify, }; -int rxe_net_ipv4_init(void) +static int rxe_net_ipv4_init(void) { - spin_lock_init(&dev_list_lock); - recv_sockets.sk4 = rxe_setup_udp_tunnel(&init_net, htons(ROCE_V2_UDP_DPORT), false); if (IS_ERR(recv_sockets.sk4)) { @@ -672,12 +654,10 @@ int rxe_net_ipv4_init(void) return 0; } -int rxe_net_ipv6_init(void) +static int rxe_net_ipv6_init(void) { #if IS_ENABLED(CONFIG_IPV6) - spin_lock_init(&dev_list_lock); - recv_sockets.sk6 = rxe_setup_udp_tunnel(&init_net, htons(ROCE_V2_UDP_DPORT), true); if (IS_ERR(recv_sockets.sk6)) { diff --git a/drivers/infiniband/sw/rxe/rxe_param.h b/drivers/infiniband/sw/rxe/rxe_param.h index f459c43a77c8..13ed2cc6eaa2 100644 --- a/drivers/infiniband/sw/rxe/rxe_param.h +++ b/drivers/infiniband/sw/rxe/rxe_param.h @@ -82,7 +82,7 @@ enum rxe_device_param { RXE_MAX_SGE = 32, RXE_MAX_SGE_RD = 32, RXE_MAX_CQ = 16384, - RXE_MAX_LOG_CQE = 13, + RXE_MAX_LOG_CQE = 15, RXE_MAX_MR = 2 * 1024, RXE_MAX_PD = 0x7ffc, RXE_MAX_QP_RD_ATOM = 128, diff --git a/drivers/infiniband/sw/rxe/rxe_pool.c b/drivers/infiniband/sw/rxe/rxe_pool.c index 6bac0717c540..75d11ee635ec 100644 --- a/drivers/infiniband/sw/rxe/rxe_pool.c +++ b/drivers/infiniband/sw/rxe/rxe_pool.c @@ -102,7 +102,7 @@ struct rxe_type_info rxe_type_info[RXE_NUM_TYPES] = { }, }; -static inline char *pool_name(struct rxe_pool *pool) +static inline const char *pool_name(struct rxe_pool *pool) { return rxe_type_info[pool->type].name; } @@ -112,13 +112,6 @@ static inline struct kmem_cache *pool_cache(struct rxe_pool *pool) return rxe_type_info[pool->type].cache; } -static inline enum rxe_elem_type rxe_type(void *arg) -{ - struct rxe_pool_entry *elem = arg; - - return elem->pool->type; -} - int rxe_cache_init(void) { int err; @@ -180,7 +173,6 @@ static int rxe_pool_init_index(struct rxe_pool *pool, u32 max, u32 min) size = BITS_TO_LONGS(max - min + 1) * sizeof(long); pool->table = kmalloc(size, GFP_KERNEL); if (!pool->table) { - pr_warn("no memory for bit table\n"); err = -ENOMEM; goto out; } @@ -274,6 +266,7 @@ static u32 alloc_index(struct rxe_pool *pool) if (index >= range) index = find_first_zero_bit(pool->table, range); + WARN_ON_ONCE(index >= range); set_bit(index, pool->table); pool->last = index; return index + pool->min_index; @@ -462,7 +455,7 @@ void *rxe_pool_get_index(struct rxe_pool *pool, u32 index) out: spin_unlock_irqrestore(&pool->pool_lock, flags); - return node ? (void *)elem : NULL; + return node ? elem : NULL; } void *rxe_pool_get_key(struct rxe_pool *pool, void *key) @@ -498,5 +491,5 @@ void *rxe_pool_get_key(struct rxe_pool *pool, void *key) out: spin_unlock_irqrestore(&pool->pool_lock, flags); - return node ? ((void *)elem) : NULL; + return node ? elem : NULL; } diff --git a/drivers/infiniband/sw/rxe/rxe_pool.h b/drivers/infiniband/sw/rxe/rxe_pool.h index 4d04830adcae..47df28e43acf 100644 --- a/drivers/infiniband/sw/rxe/rxe_pool.h +++ b/drivers/infiniband/sw/rxe/rxe_pool.h @@ -57,10 +57,12 @@ enum rxe_elem_type { RXE_NUM_TYPES, /* keep me last */ }; +struct rxe_pool_entry; + struct rxe_type_info { - char *name; + const char *name; size_t size; - void (*cleanup)(void *obj); + void (*cleanup)(struct rxe_pool_entry *obj); enum rxe_pool_flags flags; u32 max_index; u32 min_index; @@ -91,7 +93,7 @@ struct rxe_pool { spinlock_t pool_lock; /* pool spinlock */ size_t elem_size; struct kref ref_cnt; - void (*cleanup)(void *obj); + void (*cleanup)(struct rxe_pool_entry *obj); enum rxe_pool_state state; enum rxe_pool_flags flags; enum rxe_elem_type type; diff --git a/drivers/infiniband/sw/rxe/rxe_qp.c b/drivers/infiniband/sw/rxe/rxe_qp.c index c3e60e4bde6e..f98a19e61a3d 100644 --- a/drivers/infiniband/sw/rxe/rxe_qp.c +++ b/drivers/infiniband/sw/rxe/rxe_qp.c @@ -273,13 +273,8 @@ static int rxe_qp_init_req(struct rxe_dev *rxe, struct rxe_qp *qp, rxe_init_task(rxe, &qp->comp.task, qp, rxe_completer, "comp"); - init_timer(&qp->rnr_nak_timer); - qp->rnr_nak_timer.function = rnr_nak_timer; - qp->rnr_nak_timer.data = (unsigned long)qp; - - init_timer(&qp->retrans_timer); - qp->retrans_timer.function = retransmit_timer; - qp->retrans_timer.data = (unsigned long)qp; + setup_timer(&qp->rnr_nak_timer, rnr_nak_timer, (unsigned long)qp); + setup_timer(&qp->retrans_timer, retransmit_timer, (unsigned long)qp); qp->qp_timeout_jiffies = 0; /* Can't be set for UD/UC in modify_qp */ return 0; @@ -813,8 +808,7 @@ void rxe_qp_destroy(struct rxe_qp *qp) del_timer_sync(&qp->rnr_nak_timer); rxe_cleanup_task(&qp->req.task); - if (qp_type(qp) == IB_QPT_RC) - rxe_cleanup_task(&qp->comp.task); + rxe_cleanup_task(&qp->comp.task); /* flush out any receive wr's or pending requests */ __rxe_do_task(&qp->req.task); @@ -825,9 +819,9 @@ void rxe_qp_destroy(struct rxe_qp *qp) } /* called when the last reference to the qp is dropped */ -void rxe_qp_cleanup(void *arg) +void rxe_qp_cleanup(struct rxe_pool_entry *arg) { - struct rxe_qp *qp = arg; + struct rxe_qp *qp = container_of(arg, typeof(*qp), pelem); rxe_drop_all_mcast_groups(qp); @@ -855,4 +849,5 @@ void rxe_qp_cleanup(void *arg) free_rd_atomic_resources(qp); kernel_sock_shutdown(qp->sk, SHUT_RDWR); + sock_release(qp->sk); } diff --git a/drivers/infiniband/sw/rxe/rxe_recv.c b/drivers/infiniband/sw/rxe/rxe_recv.c index 46f062842a9a..50886031096f 100644 --- a/drivers/infiniband/sw/rxe/rxe_recv.c +++ b/drivers/infiniband/sw/rxe/rxe_recv.c @@ -389,18 +389,17 @@ int rxe_rcv(struct sk_buff *skb) calc_icrc = rxe_icrc_hdr(pkt, skb); calc_icrc = crc32_le(calc_icrc, (u8 *)payload_addr(pkt), payload_size(pkt)); - calc_icrc = cpu_to_be32(~calc_icrc); + calc_icrc = (__force u32)cpu_to_be32(~calc_icrc); if (unlikely(calc_icrc != pack_icrc)) { - char saddr[sizeof(struct in6_addr)]; - if (skb->protocol == htons(ETH_P_IPV6)) - sprintf(saddr, "%pI6", &ipv6_hdr(skb)->saddr); + pr_warn_ratelimited("bad ICRC from %pI6c\n", + &ipv6_hdr(skb)->saddr); else if (skb->protocol == htons(ETH_P_IP)) - sprintf(saddr, "%pI4", &ip_hdr(skb)->saddr); + pr_warn_ratelimited("bad ICRC from %pI4\n", + &ip_hdr(skb)->saddr); else - sprintf(saddr, "unknown"); + pr_warn_ratelimited("bad ICRC from unknown\n"); - pr_warn_ratelimited("bad ICRC from %s\n", saddr); goto drop; } diff --git a/drivers/infiniband/sw/rxe/rxe_req.c b/drivers/infiniband/sw/rxe/rxe_req.c index 22bd9630dcd9..dbfde0dc6ff7 100644 --- a/drivers/infiniband/sw/rxe/rxe_req.c +++ b/drivers/infiniband/sw/rxe/rxe_req.c @@ -361,19 +361,14 @@ static inline int check_init_depth(struct rxe_qp *qp, struct rxe_send_wqe *wqe) return -EAGAIN; } -static inline int get_mtu(struct rxe_qp *qp, struct rxe_send_wqe *wqe) +static inline int get_mtu(struct rxe_qp *qp) { struct rxe_dev *rxe = to_rdev(qp->ibqp.device); - struct rxe_port *port; - struct rxe_av *av; if ((qp_type(qp) == IB_QPT_RC) || (qp_type(qp) == IB_QPT_UC)) return qp->mtu; - av = &wqe->av; - port = &rxe->port; - - return port->mtu_cap; + return rxe->port.mtu_cap; } static struct sk_buff *init_req_packet(struct rxe_qp *qp, @@ -409,7 +404,7 @@ static struct sk_buff *init_req_packet(struct rxe_qp *qp, /* init skb */ av = rxe_get_av(pkt); - skb = rxe->ifc_ops->init_packet(rxe, av, paylen, pkt); + skb = rxe_init_packet(rxe, av, paylen, pkt); if (unlikely(!skb)) return NULL; @@ -480,7 +475,7 @@ static int fill_packet(struct rxe_qp *qp, struct rxe_send_wqe *wqe, u32 *p; int err; - err = rxe->ifc_ops->prepare(rxe, pkt, skb, &crc); + err = rxe_prepare(rxe, pkt, skb, &crc); if (err) return err; @@ -548,23 +543,23 @@ static void update_wqe_psn(struct rxe_qp *qp, static void save_state(struct rxe_send_wqe *wqe, struct rxe_qp *qp, struct rxe_send_wqe *rollback_wqe, - struct rxe_qp *rollback_qp) + u32 *rollback_psn) { rollback_wqe->state = wqe->state; rollback_wqe->first_psn = wqe->first_psn; rollback_wqe->last_psn = wqe->last_psn; - rollback_qp->req.psn = qp->req.psn; + *rollback_psn = qp->req.psn; } static void rollback_state(struct rxe_send_wqe *wqe, struct rxe_qp *qp, struct rxe_send_wqe *rollback_wqe, - struct rxe_qp *rollback_qp) + u32 rollback_psn) { wqe->state = rollback_wqe->state; wqe->first_psn = rollback_wqe->first_psn; wqe->last_psn = rollback_wqe->last_psn; - qp->req.psn = rollback_qp->req.psn; + qp->req.psn = rollback_psn; } static void update_state(struct rxe_qp *qp, struct rxe_send_wqe *wqe, @@ -593,13 +588,20 @@ int rxe_requester(void *arg) int mtu; int opcode; int ret; - struct rxe_qp rollback_qp; struct rxe_send_wqe rollback_wqe; + u32 rollback_psn; + + rxe_add_ref(qp); next_wqe: - if (unlikely(!qp->valid || qp->req.state == QP_STATE_ERROR)) + if (unlikely(!qp->valid)) goto exit; + if (unlikely(qp->req.state == QP_STATE_ERROR)) { + rxe_drain_req_pkts(qp, true); + goto exit; + } + if (unlikely(qp->req.state == QP_STATE_RESET)) { qp->req.wqe_index = consumer_index(qp->sq.queue); qp->req.opcode = -1; @@ -633,6 +635,7 @@ next_wqe: goto exit; } rmr->state = RXE_MEM_STATE_FREE; + rxe_drop_ref(rmr); wqe->state = wqe_state_done; wqe->status = IB_WC_SUCCESS; } else if (wqe->wr.opcode == IB_WR_REG_MR) { @@ -677,7 +680,7 @@ next_wqe: goto exit; } - mtu = get_mtu(qp, wqe); + mtu = get_mtu(qp); payload = (mask & RXE_WRITE_OR_SEND) ? wqe->dma.resid : 0; if (payload > mtu) { if (qp_type(qp) == IB_QPT_UD) { @@ -697,6 +700,7 @@ next_wqe: wqe->state = wqe_state_done; wqe->status = IB_WC_SUCCESS; __rxe_do_task(&qp->comp.task); + rxe_drop_ref(qp); return 0; } payload = mtu; @@ -719,7 +723,7 @@ next_wqe: * rxe_xmit_packet(). * Otherwise, completer might initiate an unjustified retry flow. */ - save_state(wqe, qp, &rollback_wqe, &rollback_qp); + save_state(wqe, qp, &rollback_wqe, &rollback_psn); update_wqe_state(qp, wqe, &pkt); update_wqe_psn(qp, wqe, &pkt, payload); ret = rxe_xmit_packet(to_rdev(qp->ibqp.device), qp, &pkt, skb); @@ -727,7 +731,7 @@ next_wqe: qp->need_req_skb = 1; kfree_skb(skb); - rollback_state(wqe, qp, &rollback_wqe, &rollback_qp); + rollback_state(wqe, qp, &rollback_wqe, rollback_psn); if (ret == -EAGAIN) { rxe_run_task(&qp->req.task, 1); @@ -745,19 +749,9 @@ err: kfree_skb(skb); wqe->status = IB_WC_LOC_PROT_ERR; wqe->state = wqe_state_error; - - /* - * IBA Spec. Section 10.7.3.1 SIGNALED COMPLETIONS - * ---------8<---------8<------------- - * ...Note that if a completion error occurs, a Work Completion - * will always be generated, even if the signaling - * indicator requests an Unsignaled Completion. - * ---------8<---------8<------------- - */ - wqe->wr.send_flags |= IB_SEND_SIGNALED; __rxe_do_task(&qp->comp.task); - return -EAGAIN; exit: + rxe_drop_ref(qp); return -EAGAIN; } diff --git a/drivers/infiniband/sw/rxe/rxe_resp.c b/drivers/infiniband/sw/rxe/rxe_resp.c index dd3d88adc003..d404a8aba7af 100644 --- a/drivers/infiniband/sw/rxe/rxe_resp.c +++ b/drivers/infiniband/sw/rxe/rxe_resp.c @@ -307,7 +307,7 @@ static enum resp_states check_op_valid(struct rxe_qp *qp, break; default: - WARN_ON(1); + WARN_ON_ONCE(1); break; } @@ -418,7 +418,7 @@ static enum resp_states check_length(struct rxe_qp *qp, static enum resp_states check_rkey(struct rxe_qp *qp, struct rxe_pkt_info *pkt) { - struct rxe_mem *mem; + struct rxe_mem *mem = NULL; u64 va; u32 rkey; u32 resid; @@ -444,6 +444,13 @@ static enum resp_states check_rkey(struct rxe_qp *qp, return RESPST_EXECUTE; } + /* A zero-byte op is not required to set an addr or rkey. */ + if ((pkt->mask & (RXE_READ_MASK | RXE_WRITE_OR_SEND)) && + (pkt->mask & RXE_RETH_MASK) && + reth_len(pkt) == 0) { + return RESPST_EXECUTE; + } + va = qp->resp.va; rkey = qp->resp.rkey; resid = qp->resp.resid; @@ -452,50 +459,50 @@ static enum resp_states check_rkey(struct rxe_qp *qp, mem = lookup_mem(qp->pd, access, rkey, lookup_remote); if (!mem) { state = RESPST_ERR_RKEY_VIOLATION; - goto err1; + goto err; } if (unlikely(mem->state == RXE_MEM_STATE_FREE)) { state = RESPST_ERR_RKEY_VIOLATION; - goto err1; + goto err; } if (mem_check_range(mem, va, resid)) { state = RESPST_ERR_RKEY_VIOLATION; - goto err2; + goto err; } if (pkt->mask & RXE_WRITE_MASK) { if (resid > mtu) { if (pktlen != mtu || bth_pad(pkt)) { state = RESPST_ERR_LENGTH; - goto err2; + goto err; } - resid = mtu; + qp->resp.resid = mtu; } else { if (pktlen != resid) { state = RESPST_ERR_LENGTH; - goto err2; + goto err; } if ((bth_pad(pkt) != (0x3 & (-resid)))) { /* This case may not be exactly that * but nothing else fits. */ state = RESPST_ERR_LENGTH; - goto err2; + goto err; } } } - WARN_ON(qp->resp.mr); + WARN_ON_ONCE(qp->resp.mr); qp->resp.mr = mem; return RESPST_EXECUTE; -err2: - rxe_drop_ref(mem); -err1: +err: + if (mem) + rxe_drop_ref(mem); return state; } @@ -601,7 +608,7 @@ static struct sk_buff *prepare_ack_packet(struct rxe_qp *qp, pad = (-payload) & 0x3; paylen = rxe_opcode[opcode].length + payload + pad + RXE_ICRC_SIZE; - skb = rxe->ifc_ops->init_packet(rxe, &qp->pri_av, paylen, ack); + skb = rxe_init_packet(rxe, &qp->pri_av, paylen, ack); if (!skb) return NULL; @@ -630,7 +637,7 @@ static struct sk_buff *prepare_ack_packet(struct rxe_qp *qp, if (ack->mask & RXE_ATMACK_MASK) atmack_set_orig(ack, qp->resp.atomic_orig); - err = rxe->ifc_ops->prepare(rxe, ack, skb, &crc); + err = rxe_prepare(rxe, ack, skb, &crc); if (err) { kfree_skb(skb); return NULL; @@ -680,9 +687,14 @@ static enum resp_states read_reply(struct rxe_qp *qp, res->read.va_org = qp->resp.va; res->first_psn = req_pkt->psn; - res->last_psn = req_pkt->psn + - (reth_len(req_pkt) + mtu - 1) / - mtu - 1; + + if (reth_len(req_pkt)) { + res->last_psn = (req_pkt->psn + + (reth_len(req_pkt) + mtu - 1) / + mtu - 1) & BTH_PSN_MASK; + } else { + res->last_psn = res->first_psn; + } res->cur_psn = req_pkt->psn; res->read.resid = qp->resp.resid; @@ -742,7 +754,8 @@ static enum resp_states read_reply(struct rxe_qp *qp, } else { qp->resp.res = NULL; qp->resp.opcode = -1; - qp->resp.psn = res->cur_psn; + if (psn_compare(res->cur_psn, qp->resp.psn) >= 0) + qp->resp.psn = res->cur_psn; state = RESPST_CLEANUP; } @@ -795,9 +808,10 @@ static enum resp_states execute(struct rxe_qp *qp, struct rxe_pkt_info *pkt) err = process_atomic(qp, pkt); if (err) return err; - } else + } else { /* Unreachable */ - WARN_ON(1); + WARN_ON_ONCE(1); + } /* We successfully processed this new request. */ qp->resp.msn++; @@ -893,6 +907,7 @@ static enum resp_states do_complete(struct rxe_qp *qp, return RESPST_ERROR; } rmr->state = RXE_MEM_STATE_FREE; + rxe_drop_ref(rmr); } wc->qp = &qp->ibqp; @@ -1057,12 +1072,13 @@ static enum resp_states duplicate_request(struct rxe_qp *qp, struct rxe_pkt_info *pkt) { enum resp_states rc; + u32 prev_psn = (qp->resp.psn - 1) & BTH_PSN_MASK; if (pkt->mask & RXE_SEND_MASK || pkt->mask & RXE_WRITE_MASK) { /* SEND. Ack again and cleanup. C9-105. */ if (bth_ack(pkt)) - send_ack(qp, pkt, AETH_ACK_UNLIMITED, qp->resp.psn - 1); + send_ack(qp, pkt, AETH_ACK_UNLIMITED, prev_psn); rc = RESPST_CLEANUP; goto out; } else if (pkt->mask & RXE_READ_MASK) { @@ -1132,6 +1148,7 @@ static enum resp_states duplicate_request(struct rxe_qp *qp, pkt, skb_copy); if (rc) { pr_err("Failed resending result. This flow is not handled - skb ignored\n"); + rxe_drop_ref(qp); kfree_skb(skb_copy); rc = RESPST_CLEANUP; goto out; @@ -1191,6 +1208,19 @@ static enum resp_states do_class_d1e_error(struct rxe_qp *qp) } } +void rxe_drain_req_pkts(struct rxe_qp *qp, bool notify) +{ + struct sk_buff *skb; + + while ((skb = skb_dequeue(&qp->req_pkts))) { + rxe_drop_ref(qp); + kfree_skb(skb); + } + + while (!qp->srq && qp->rq.queue && queue_head(qp->rq.queue)) + advance_consumer(qp->rq.queue); +} + int rxe_responder(void *arg) { struct rxe_qp *qp = (struct rxe_qp *)arg; @@ -1198,6 +1228,8 @@ int rxe_responder(void *arg) struct rxe_pkt_info *pkt = NULL; int ret = 0; + rxe_add_ref(qp); + qp->resp.aeth_syndrome = AETH_ACK_UNLIMITED; if (!qp->valid) { @@ -1356,21 +1388,10 @@ int rxe_responder(void *arg) goto exit; - case RESPST_RESET: { - struct sk_buff *skb; - - while ((skb = skb_dequeue(&qp->req_pkts))) { - rxe_drop_ref(qp); - kfree_skb(skb); - } - - while (!qp->srq && qp->rq.queue && - queue_head(qp->rq.queue)) - advance_consumer(qp->rq.queue); - + case RESPST_RESET: + rxe_drain_req_pkts(qp, false); qp->resp.wqe = NULL; goto exit; - } case RESPST_ERROR: qp->resp.goto_error = 0; @@ -1379,12 +1400,13 @@ int rxe_responder(void *arg) goto exit; default: - WARN_ON(1); + WARN_ON_ONCE(1); } } exit: ret = -EAGAIN; done: + rxe_drop_ref(qp); return ret; } diff --git a/drivers/infiniband/sw/rxe/rxe_srq.c b/drivers/infiniband/sw/rxe/rxe_srq.c index 2a6e3cd2d4e8..efc832a2d7c6 100644 --- a/drivers/infiniband/sw/rxe/rxe_srq.c +++ b/drivers/infiniband/sw/rxe/rxe_srq.c @@ -169,7 +169,7 @@ int rxe_srq_from_attr(struct rxe_dev *rxe, struct rxe_srq *srq, } } - err = rxe_queue_resize(q, (unsigned int *)&attr->max_wr, + err = rxe_queue_resize(q, &attr->max_wr, rcv_wqe_size(srq->rq.max_sge), srq->rq.queue->ip ? srq->rq.queue->ip->context : diff --git a/drivers/infiniband/sw/rxe/rxe_task.c b/drivers/infiniband/sw/rxe/rxe_task.c index 1e19bf828a6e..d2a14a1bdc7f 100644 --- a/drivers/infiniband/sw/rxe/rxe_task.c +++ b/drivers/infiniband/sw/rxe/rxe_task.c @@ -121,6 +121,7 @@ int rxe_init_task(void *obj, struct rxe_task *task, task->arg = arg; task->func = func; snprintf(task->name, sizeof(task->name), "%s", name); + task->destroyed = false; tasklet_init(&task->tasklet, rxe_do_task, (unsigned long)task); @@ -132,11 +133,29 @@ int rxe_init_task(void *obj, struct rxe_task *task, void rxe_cleanup_task(struct rxe_task *task) { + unsigned long flags; + bool idle; + + /* + * Mark the task, then wait for it to finish. It might be + * running in a non-tasklet (direct call) context. + */ + task->destroyed = true; + + do { + spin_lock_irqsave(&task->state_lock, flags); + idle = (task->state == TASK_STATE_START); + spin_unlock_irqrestore(&task->state_lock, flags); + } while (!idle); + tasklet_kill(&task->tasklet); } void rxe_run_task(struct rxe_task *task, int sched) { + if (task->destroyed) + return; + if (sched) tasklet_schedule(&task->tasklet); else diff --git a/drivers/infiniband/sw/rxe/rxe_task.h b/drivers/infiniband/sw/rxe/rxe_task.h index d14aa6daed05..08ff42d451c6 100644 --- a/drivers/infiniband/sw/rxe/rxe_task.h +++ b/drivers/infiniband/sw/rxe/rxe_task.h @@ -54,6 +54,7 @@ struct rxe_task { int (*func)(void *arg); int ret; char name[16]; + bool destroyed; }; /* diff --git a/drivers/infiniband/sw/rxe/rxe_verbs.c b/drivers/infiniband/sw/rxe/rxe_verbs.c index 19841c863daf..5113e502f6f9 100644 --- a/drivers/infiniband/sw/rxe/rxe_verbs.c +++ b/drivers/infiniband/sw/rxe/rxe_verbs.c @@ -31,6 +31,7 @@ * SOFTWARE. */ +#include <linux/dma-mapping.h> #include "rxe.h" #include "rxe_loc.h" #include "rxe_queue.h" @@ -86,6 +87,7 @@ static int rxe_query_port(struct ib_device *dev, port = &rxe->port; + /* *attr being zeroed by the caller, avoid zeroing it here */ *attr = port->attr; mutex_lock(&rxe->usdev_lock); @@ -168,7 +170,7 @@ static int rxe_query_pkey(struct ib_device *device, struct rxe_port *port; if (unlikely(port_num != 1)) { - dev_warn(device->dma_device, "invalid port_num = %d\n", + dev_warn(device->dev.parent, "invalid port_num = %d\n", port_num); goto err1; } @@ -176,7 +178,7 @@ static int rxe_query_pkey(struct ib_device *device, port = &rxe->port; if (unlikely(index >= port->attr.pkey_tbl_len)) { - dev_warn(device->dma_device, "invalid index = %d\n", + dev_warn(device->dev.parent, "invalid index = %d\n", index); goto err1; } @@ -234,7 +236,7 @@ static enum rdma_link_layer rxe_get_link_layer(struct ib_device *dev, { struct rxe_dev *rxe = to_rdev(dev); - return rxe->ifc_ops->link_layer(rxe, port_num); + return rxe_link_layer(rxe, port_num); } static struct ib_ucontext *rxe_alloc_ucontext(struct ib_device *dev, @@ -261,13 +263,14 @@ static int rxe_port_immutable(struct ib_device *dev, u8 port_num, int err; struct ib_port_attr attr; - err = rxe_query_port(dev, port_num, &attr); + immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP; + + err = ib_query_port(dev, port_num, &attr); if (err) return err; immutable->pkey_tbl_len = attr.pkey_tbl_len; immutable->gid_tbl_len = attr.gid_tbl_len; - immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP; immutable->max_mad_size = IB_MGMT_MAD_SIZE; return 0; @@ -316,7 +319,9 @@ static int rxe_init_av(struct rxe_dev *rxe, struct ib_ah_attr *attr, return err; } -static struct ib_ah *rxe_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *attr) +static struct ib_ah *rxe_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *attr, + struct ib_udata *udata) + { int err; struct rxe_dev *rxe = to_rdev(ibpd->device); @@ -564,7 +569,7 @@ static struct ib_qp *rxe_create_qp(struct ib_pd *ibpd, if (udata) { if (udata->inlen) { err = -EINVAL; - goto err1; + goto err2; } qp->is_user = 1; } @@ -573,12 +578,13 @@ static struct ib_qp *rxe_create_qp(struct ib_pd *ibpd, err = rxe_qp_from_init(rxe, qp, pd, init, udata, ibpd); if (err) - goto err2; + goto err3; return &qp->ibqp; -err2: +err3: rxe_drop_index(qp); +err2: rxe_drop_ref(qp); err1: return ERR_PTR(err); @@ -1007,11 +1013,19 @@ static int rxe_peek_cq(struct ib_cq *ibcq, int wc_cnt) static int rxe_req_notify_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags flags) { struct rxe_cq *cq = to_rcq(ibcq); + unsigned long irq_flags; + int ret = 0; + spin_lock_irqsave(&cq->cq_lock, irq_flags); if (cq->notify != IB_CQ_NEXT_COMP) cq->notify = flags & IB_CQ_SOLICITED_MASK; - return 0; + if ((flags & IB_CQ_REPORT_MISSED_EVENTS) && !queue_empty(cq->queue)) + ret = 1; + + spin_unlock_irqrestore(&cq->cq_lock, irq_flags); + + return ret; } static struct ib_mr *rxe_get_dma_mr(struct ib_pd *ibpd, int access) @@ -1198,10 +1212,8 @@ static ssize_t rxe_show_parent(struct device *device, { struct rxe_dev *rxe = container_of(device, struct rxe_dev, ib_dev.dev); - char *name; - name = rxe->ifc_ops->parent_name(rxe, 1); - return snprintf(buf, 16, "%s\n", name); + return snprintf(buf, 16, "%s\n", rxe_parent_name(rxe, 1)); } static DEVICE_ATTR(parent, S_IRUGO, rxe_show_parent, NULL); @@ -1223,10 +1235,10 @@ int rxe_register_device(struct rxe_dev *rxe) dev->node_type = RDMA_NODE_IB_CA; dev->phys_port_cnt = 1; dev->num_comp_vectors = RXE_NUM_COMP_VECTORS; - dev->dma_device = rxe->ifc_ops->dma_device(rxe); + dev->dev.parent = rxe_dma_device(rxe); dev->local_dma_lkey = 0; - dev->node_guid = rxe->ifc_ops->node_guid(rxe); - dev->dma_ops = &rxe_dma_mapping_ops; + dev->node_guid = rxe_node_guid(rxe); + dev->dev.dma_ops = &dma_virt_ops; dev->uverbs_abi_ver = RXE_UVERBS_ABI_VERSION; dev->uverbs_cmd_mask = BIT_ULL(IB_USER_VERBS_CMD_GET_CONTEXT) diff --git a/drivers/infiniband/sw/rxe/rxe_verbs.h b/drivers/infiniband/sw/rxe/rxe_verbs.h index cac1d52a08f0..e100c500ae85 100644 --- a/drivers/infiniband/sw/rxe/rxe_verbs.h +++ b/drivers/infiniband/sw/rxe/rxe_verbs.h @@ -372,26 +372,6 @@ struct rxe_port { u32 qp_gsi_index; }; -/* callbacks from rdma_rxe to network interface layer */ -struct rxe_ifc_ops { - void (*release)(struct rxe_dev *rxe); - __be64 (*node_guid)(struct rxe_dev *rxe); - __be64 (*port_guid)(struct rxe_dev *rxe); - struct device *(*dma_device)(struct rxe_dev *rxe); - int (*mcast_add)(struct rxe_dev *rxe, union ib_gid *mgid); - int (*mcast_delete)(struct rxe_dev *rxe, union ib_gid *mgid); - int (*prepare)(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, - struct sk_buff *skb, u32 *crc); - int (*send)(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, - struct sk_buff *skb); - int (*loopback)(struct sk_buff *skb); - struct sk_buff *(*init_packet)(struct rxe_dev *rxe, struct rxe_av *av, - int paylen, struct rxe_pkt_info *pkt); - char *(*parent_name)(struct rxe_dev *rxe, unsigned int port_num); - enum rdma_link_layer (*link_layer)(struct rxe_dev *rxe, - unsigned int port_num); -}; - struct rxe_dev { struct ib_device ib_dev; struct ib_device_attr attr; @@ -400,8 +380,6 @@ struct rxe_dev { struct kref ref_cnt; struct mutex usdev_lock; - struct rxe_ifc_ops *ifc_ops; - struct net_device *ndev; int xmit_errors; @@ -475,6 +453,6 @@ static inline struct rxe_mem *to_rmw(struct ib_mw *mw) int rxe_register_device(struct rxe_dev *rxe); int rxe_unregister_device(struct rxe_dev *rxe); -void rxe_mc_cleanup(void *arg); +void rxe_mc_cleanup(struct rxe_pool_entry *arg); #endif /* RXE_VERBS_H */ diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h index da12717a3eb7..bed233bf45c3 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib.h +++ b/drivers/infiniband/ulp/ipoib/ipoib.h @@ -500,9 +500,9 @@ void ipoib_pkey_event(struct work_struct *work); void ipoib_ib_dev_cleanup(struct net_device *dev); int ipoib_ib_dev_open(struct net_device *dev); -int ipoib_ib_dev_up(struct net_device *dev); -int ipoib_ib_dev_down(struct net_device *dev); -int ipoib_ib_dev_stop(struct net_device *dev); +void ipoib_ib_dev_up(struct net_device *dev); +void ipoib_ib_dev_down(struct net_device *dev); +void ipoib_ib_dev_stop(struct net_device *dev); void ipoib_pkey_dev_check_presence(struct net_device *dev); int ipoib_dev_init(struct net_device *dev, struct ib_device *ca, int port); @@ -513,7 +513,7 @@ void ipoib_mcast_carrier_on_task(struct work_struct *work); void ipoib_mcast_send(struct net_device *dev, u8 *daddr, struct sk_buff *skb); void ipoib_mcast_restart_task(struct work_struct *work); -int ipoib_mcast_start_thread(struct net_device *dev); +void ipoib_mcast_start_thread(struct net_device *dev); int ipoib_mcast_stop_thread(struct net_device *dev); void ipoib_mcast_dev_down(struct net_device *dev); @@ -593,7 +593,7 @@ void ipoib_pkey_open(struct ipoib_dev_priv *priv); void ipoib_drain_cq(struct net_device *dev); void ipoib_set_ethtool_ops(struct net_device *dev); -int ipoib_set_dev_features(struct ipoib_dev_priv *priv, struct ib_device *hca); +void ipoib_set_dev_features(struct ipoib_dev_priv *priv, struct ib_device *hca); #define IPOIB_FLAGS_RC 0x80 #define IPOIB_FLAGS_UC 0x40 diff --git a/drivers/infiniband/ulp/ipoib/ipoib_cm.c b/drivers/infiniband/ulp/ipoib/ipoib_cm.c index 339a1eecdfe3..a6d6c617b597 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_cm.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_cm.c @@ -357,16 +357,13 @@ static int ipoib_cm_nonsrq_init_rx(struct net_device *dev, struct ib_cm_id *cm_i int i; rx->rx_ring = vzalloc(ipoib_recvq_size * sizeof *rx->rx_ring); - if (!rx->rx_ring) { - printk(KERN_WARNING "%s: failed to allocate CM non-SRQ ring (%d entries)\n", - priv->ca->name, ipoib_recvq_size); + if (!rx->rx_ring) return -ENOMEM; - } t = kmalloc(sizeof *t, GFP_KERNEL); if (!t) { ret = -ENOMEM; - goto err_free; + goto err_free_1; } ipoib_cm_init_rx_wr(dev, &t->wr, t->sge); @@ -413,6 +410,8 @@ err_count: err_free: kfree(t); + +err_free_1: ipoib_cm_free_rx_ring(dev, rx->rx_ring); return ret; @@ -823,9 +822,12 @@ void ipoib_cm_handle_tx_wc(struct net_device *dev, struct ib_wc *wc) wc->status != IB_WC_WR_FLUSH_ERR) { struct ipoib_neigh *neigh; - ipoib_dbg(priv, "failed cm send event " - "(status=%d, wrid=%d vend_err %x)\n", - wc->status, wr_id, wc->vendor_err); + if (wc->status != IB_WC_RNR_RETRY_EXC_ERR) + ipoib_warn(priv, "failed cm send event (status=%d, wrid=%d vend_err %x)\n", + wc->status, wr_id, wc->vendor_err); + else + ipoib_dbg(priv, "failed cm send event (status=%d, wrid=%d vend_err %x)\n", + wc->status, wr_id, wc->vendor_err); spin_lock_irqsave(&priv->lock, flags); neigh = tx->neigh; @@ -1018,9 +1020,10 @@ static int ipoib_cm_rep_handler(struct ib_cm_id *cm_id, struct ib_cm_event *even while ((skb = __skb_dequeue(&skqueue))) { skb->dev = p->dev; - if (dev_queue_xmit(skb)) - ipoib_warn(priv, "dev_queue_xmit failed " - "to requeue packet\n"); + ret = dev_queue_xmit(skb); + if (ret) + ipoib_warn(priv, "%s:dev_queue_xmit failed to re-queue packet, ret:%d\n", + __func__, ret); } ret = ib_send_cm_rtu(cm_id, NULL, 0); @@ -1054,8 +1057,6 @@ static struct ib_qp *ipoib_cm_create_tx_qp(struct net_device *dev, struct ipoib_ tx_qp = ib_create_qp(priv->pd, &attr); if (PTR_ERR(tx_qp) == -EINVAL) { - ipoib_warn(priv, "can't use GFP_NOIO for QPs on device %s, using GFP_KERNEL\n", - priv->ca->name); attr.create_flags &= ~IB_QP_CREATE_USE_GFP_NOIO; tx_qp = ib_create_qp(priv->pd, &attr); } @@ -1134,7 +1135,6 @@ static int ipoib_cm_tx_init(struct ipoib_cm_tx *p, u32 qpn, p->tx_ring = __vmalloc(ipoib_sendq_size * sizeof *p->tx_ring, GFP_NOIO, PAGE_KERNEL); if (!p->tx_ring) { - ipoib_warn(priv, "failed to allocate tx ring\n"); ret = -ENOMEM; goto err_tx; } @@ -1157,13 +1157,13 @@ static int ipoib_cm_tx_init(struct ipoib_cm_tx *p, u32 qpn, ret = ipoib_cm_modify_tx_init(p->dev, p->id, p->qp); if (ret) { ipoib_warn(priv, "failed to modify tx qp to rtr: %d\n", ret); - goto err_modify; + goto err_modify_send; } ret = ipoib_cm_send_req(p->dev, p->id, p->qp, qpn, pathrec); if (ret) { ipoib_warn(priv, "failed to send cm req: %d\n", ret); - goto err_send_cm; + goto err_modify_send; } ipoib_dbg(priv, "Request connection 0x%x for gid %pI6 qpn 0x%x\n", @@ -1171,8 +1171,7 @@ static int ipoib_cm_tx_init(struct ipoib_cm_tx *p, u32 qpn, return 0; -err_send_cm: -err_modify: +err_modify_send: ib_destroy_cm_id(p->id); err_id: p->id = NULL; @@ -1394,7 +1393,7 @@ static void ipoib_cm_tx_reap(struct work_struct *work) while (!list_empty(&priv->cm.reap_list)) { p = list_entry(priv->cm.reap_list.next, typeof(*p), list); - list_del(&p->list); + list_del_init(&p->list); spin_unlock_irqrestore(&priv->lock, flags); netif_tx_unlock_bh(dev); ipoib_cm_tx_destroy(p); @@ -1513,12 +1512,14 @@ static ssize_t set_mode(struct device *d, struct device_attribute *attr, ret = ipoib_set_mode(dev, buf); - rtnl_unlock(); - - if (!ret) - return count; + /* The assumption is that the function ipoib_set_mode returned + * with the rtnl held by it, if not the value -EBUSY returned, + * then no need to rtnl_unlock + */ + if (ret != -EBUSY) + rtnl_unlock(); - return ret; + return (!ret || ret == -EBUSY) ? count : ret; } static DEVICE_ATTR(mode, S_IWUSR | S_IRUGO, show_mode, set_mode); @@ -1550,8 +1551,6 @@ static void ipoib_cm_create_srq(struct net_device *dev, int max_sge) priv->cm.srq_ring = vzalloc(ipoib_recvq_size * sizeof *priv->cm.srq_ring); if (!priv->cm.srq_ring) { - printk(KERN_WARNING "%s: failed to allocate CM SRQ ring (%d entries)\n", - priv->ca->name, ipoib_recvq_size); ib_destroy_srq(priv->cm.srq); priv->cm.srq = NULL; return; diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c b/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c index 7b6d40ff1acf..bac455a1942d 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c @@ -65,7 +65,7 @@ static void ipoib_get_drvinfo(struct net_device *netdev, ib_get_device_fw_str(priv->ca, drvinfo->fw_version, sizeof(drvinfo->fw_version)); - strlcpy(drvinfo->bus_info, dev_name(priv->ca->dma_device), + strlcpy(drvinfo->bus_info, dev_name(priv->ca->dev.parent), sizeof(drvinfo->bus_info)); strlcpy(drvinfo->version, ipoib_driver_version, diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c index 830fecb6934c..12c4f84a6639 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c @@ -416,11 +416,8 @@ static void ipoib_ib_handle_tx_wc(struct net_device *dev, struct ib_wc *wc) "(status=%d, wrid=%d vend_err %x)\n", wc->status, wr_id, wc->vendor_err); qp_work = kzalloc(sizeof(*qp_work), GFP_ATOMIC); - if (!qp_work) { - ipoib_warn(priv, "%s Failed alloc ipoib_qp_state_validate for qp: 0x%x\n", - __func__, priv->qp->qp_num); + if (!qp_work) return; - } INIT_WORK(&qp_work->work, ipoib_qp_state_validate_work); qp_work->priv = priv; @@ -758,7 +755,7 @@ void ipoib_pkey_dev_check_presence(struct net_device *dev) set_bit(IPOIB_PKEY_ASSIGNED, &priv->flags); } -int ipoib_ib_dev_up(struct net_device *dev) +void ipoib_ib_dev_up(struct net_device *dev) { struct ipoib_dev_priv *priv = netdev_priv(dev); @@ -766,15 +763,15 @@ int ipoib_ib_dev_up(struct net_device *dev) if (!test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags)) { ipoib_dbg(priv, "PKEY is not assigned.\n"); - return 0; + return; } set_bit(IPOIB_FLAG_OPER_UP, &priv->flags); - return ipoib_mcast_start_thread(dev); + ipoib_mcast_start_thread(dev); } -int ipoib_ib_dev_down(struct net_device *dev) +void ipoib_ib_dev_down(struct net_device *dev) { struct ipoib_dev_priv *priv = netdev_priv(dev); @@ -787,8 +784,6 @@ int ipoib_ib_dev_down(struct net_device *dev) ipoib_mcast_dev_flush(dev); ipoib_flush_paths(dev); - - return 0; } static int recvs_pending(struct net_device *dev) @@ -843,7 +838,7 @@ void ipoib_drain_cq(struct net_device *dev) local_bh_enable(); } -int ipoib_ib_dev_stop(struct net_device *dev) +void ipoib_ib_dev_stop(struct net_device *dev) { struct ipoib_dev_priv *priv = netdev_priv(dev); struct ib_qp_attr qp_attr; @@ -916,8 +911,6 @@ timeout: ipoib_flush_ah(dev); ib_req_notify_cq(priv->recv_cq, IB_CQ_NEXT_COMP); - - return 0; } int ipoib_ib_dev_init(struct net_device *dev, struct ib_device *ca, int port) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index b58d9dca5c93..d1d3fb7a6127 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -126,8 +126,7 @@ int ipoib_open(struct net_device *dev) goto err_disable; } - if (ipoib_ib_dev_up(dev)) - goto err_stop; + ipoib_ib_dev_up(dev); if (!test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags)) { struct ipoib_dev_priv *cpriv; @@ -150,9 +149,6 @@ int ipoib_open(struct net_device *dev) return 0; -err_stop: - ipoib_ib_dev_stop(dev); - err_disable: clear_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags); @@ -229,6 +225,10 @@ static int ipoib_change_mtu(struct net_device *dev, int new_mtu) priv->admin_mtu = new_mtu; + if (priv->mcast_mtu < priv->admin_mtu) + ipoib_dbg(priv, "MTU must be smaller than the underlying " + "link layer MTU - 4 (%u)\n", priv->mcast_mtu); + dev->mtu = min(priv->mcast_mtu, priv->admin_mtu); return 0; @@ -292,6 +292,25 @@ static struct net_device *ipoib_get_master_net_dev(struct net_device *dev) return dev; } +struct ipoib_walk_data { + const struct sockaddr *addr; + struct net_device *result; +}; + +static int ipoib_upper_walk(struct net_device *upper, void *_data) +{ + struct ipoib_walk_data *data = _data; + int ret = 0; + + if (ipoib_is_dev_match_addr_rcu(data->addr, upper)) { + dev_hold(upper); + data->result = upper; + ret = 1; + } + + return ret; +} + /** * Find a net_device matching the given address, which is an upper device of * the given net_device. @@ -304,27 +323,21 @@ static struct net_device *ipoib_get_master_net_dev(struct net_device *dev) static struct net_device *ipoib_get_net_dev_match_addr( const struct sockaddr *addr, struct net_device *dev) { - struct net_device *upper, - *result = NULL; - struct list_head *iter; + struct ipoib_walk_data data = { + .addr = addr, + }; rcu_read_lock(); if (ipoib_is_dev_match_addr_rcu(addr, dev)) { dev_hold(dev); - result = dev; + data.result = dev; goto out; } - netdev_for_each_all_upper_dev_rcu(dev, upper, iter) { - if (ipoib_is_dev_match_addr_rcu(addr, upper)) { - dev_hold(upper); - result = upper; - break; - } - } + netdev_walk_all_upper_dev_rcu(dev, ipoib_upper_walk, &data); out: rcu_read_unlock(); - return result; + return data.result; } /* returns the number of IPoIB netdevs on top a given ipoib device matching a @@ -457,6 +470,13 @@ int ipoib_set_mode(struct net_device *dev, const char *buf) { struct ipoib_dev_priv *priv = netdev_priv(dev); + if ((test_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags) && + !strcmp(buf, "connected\n")) || + (!test_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags) && + !strcmp(buf, "datagram\n"))) { + return 0; + } + /* flush paths if we switch modes so that connections are restarted */ if (IPOIB_CM_SUPPORTED(dev->dev_addr) && !strcmp(buf, "connected\n")) { set_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags); @@ -468,8 +488,7 @@ int ipoib_set_mode(struct net_device *dev, const char *buf) priv->tx_wr.wr.send_flags &= ~IB_SEND_IP_CSUM; ipoib_flush_paths(dev); - rtnl_lock(); - return 0; + return (!rtnl_trylock()) ? -EBUSY : 0; } if (!strcmp(buf, "datagram\n")) { @@ -478,8 +497,7 @@ int ipoib_set_mode(struct net_device *dev, const char *buf) dev_set_mtu(dev, min(priv->mcast_mtu, dev->mtu)); rtnl_unlock(); ipoib_flush_paths(dev); - rtnl_lock(); - return 0; + return (!rtnl_trylock()) ? -EBUSY : 0; } return -EINVAL; @@ -703,6 +721,14 @@ int ipoib_check_sm_sendonly_fullmember_support(struct ipoib_dev_priv *priv) return ret; } +static void push_pseudo_header(struct sk_buff *skb, const char *daddr) +{ + struct ipoib_pseudo_header *phdr; + + phdr = (struct ipoib_pseudo_header *)skb_push(skb, sizeof(*phdr)); + memcpy(phdr->hwaddr, daddr, INFINIBAND_ALEN); +} + void ipoib_flush_paths(struct net_device *dev) { struct ipoib_dev_priv *priv = netdev_priv(dev); @@ -821,10 +847,12 @@ static void path_rec_completion(int status, ipoib_put_ah(old_ah); while ((skb = __skb_dequeue(&skqueue))) { + int ret; skb->dev = dev; - if (dev_queue_xmit(skb)) - ipoib_warn(priv, "dev_queue_xmit failed " - "to requeue packet\n"); + ret = dev_queue_xmit(skb); + if (ret) + ipoib_warn(priv, "%s: dev_queue_xmit failed to re-queue packet, ret:%d\n", + __func__, ret); } } @@ -927,8 +955,7 @@ static void neigh_add_path(struct sk_buff *skb, u8 *daddr, } if (skb_queue_len(&neigh->queue) < IPOIB_MAX_PATH_REC_QUEUE) { - /* put pseudoheader back on for next time */ - skb_push(skb, IPOIB_PSEUDO_LEN); + push_pseudo_header(skb, neigh->daddr); __skb_queue_tail(&neigh->queue, skb); } else { ipoib_warn(priv, "queue length limit %d. Packet drop.\n", @@ -946,10 +973,12 @@ static void neigh_add_path(struct sk_buff *skb, u8 *daddr, if (!path->query && path_rec_start(dev, path)) goto err_path; - if (skb_queue_len(&neigh->queue) < IPOIB_MAX_PATH_REC_QUEUE) + if (skb_queue_len(&neigh->queue) < IPOIB_MAX_PATH_REC_QUEUE) { + push_pseudo_header(skb, neigh->daddr); __skb_queue_tail(&neigh->queue, skb); - else + } else { goto err_drop; + } } spin_unlock_irqrestore(&priv->lock, flags); @@ -985,8 +1014,7 @@ static void unicast_arp_send(struct sk_buff *skb, struct net_device *dev, } if (path) { if (skb_queue_len(&path->queue) < IPOIB_MAX_PATH_REC_QUEUE) { - /* put pseudoheader back on for next time */ - skb_push(skb, IPOIB_PSEUDO_LEN); + push_pseudo_header(skb, phdr->hwaddr); __skb_queue_tail(&path->queue, skb); } else { ++dev->stats.tx_dropped; @@ -1018,8 +1046,7 @@ static void unicast_arp_send(struct sk_buff *skb, struct net_device *dev, return; } else if ((path->query || !path_rec_start(dev, path)) && skb_queue_len(&path->queue) < IPOIB_MAX_PATH_REC_QUEUE) { - /* put pseudoheader back on for next time */ - skb_push(skb, IPOIB_PSEUDO_LEN); + push_pseudo_header(skb, phdr->hwaddr); __skb_queue_tail(&path->queue, skb); } else { ++dev->stats.tx_dropped; @@ -1100,8 +1127,7 @@ send_using_neigh: } if (skb_queue_len(&neigh->queue) < IPOIB_MAX_PATH_REC_QUEUE) { - /* put pseudoheader back on for next time */ - skb_push(skb, sizeof(*phdr)); + push_pseudo_header(skb, phdr->hwaddr); spin_lock_irqsave(&priv->lock, flags); __skb_queue_tail(&neigh->queue, skb); spin_unlock_irqrestore(&priv->lock, flags); @@ -1133,7 +1159,6 @@ static int ipoib_hard_header(struct sk_buff *skb, unsigned short type, const void *daddr, const void *saddr, unsigned len) { - struct ipoib_pseudo_header *phdr; struct ipoib_header *header; header = (struct ipoib_header *) skb_push(skb, sizeof *header); @@ -1146,8 +1171,7 @@ static int ipoib_hard_header(struct sk_buff *skb, * destination address into skb hard header so we can figure out where * to send the packet later. */ - phdr = (struct ipoib_pseudo_header *) skb_push(skb, sizeof(*phdr)); - memcpy(phdr->hwaddr, daddr, INFINIBAND_ALEN); + push_pseudo_header(skb, daddr); return IPOIB_HARD_LEN; } @@ -1273,7 +1297,7 @@ static void __ipoib_reap_neigh(struct ipoib_dev_priv *priv) rcu_dereference_protected(neigh->hnext, lockdep_is_held(&priv->lock))); /* remove from path/mc list */ - list_del(&neigh->list); + list_del_init(&neigh->list); call_rcu(&neigh->rcu, ipoib_neigh_reclaim); } else { np = &neigh->hnext; @@ -1437,7 +1461,7 @@ void ipoib_neigh_free(struct ipoib_neigh *neigh) rcu_dereference_protected(neigh->hnext, lockdep_is_held(&priv->lock))); /* remove from parent list */ - list_del(&neigh->list); + list_del_init(&neigh->list); call_rcu(&neigh->rcu, ipoib_neigh_reclaim); return; } else { @@ -1522,7 +1546,7 @@ void ipoib_del_neighs_by_gid(struct net_device *dev, u8 *gid) rcu_dereference_protected(neigh->hnext, lockdep_is_held(&priv->lock))); /* remove from parent list */ - list_del(&neigh->list); + list_del_init(&neigh->list); call_rcu(&neigh->rcu, ipoib_neigh_reclaim); } else { np = &neigh->hnext; @@ -1564,7 +1588,7 @@ static void ipoib_flush_neighs(struct ipoib_dev_priv *priv) rcu_dereference_protected(neigh->hnext, lockdep_is_held(&priv->lock))); /* remove from path/mc list */ - list_del(&neigh->list); + list_del_init(&neigh->list); call_rcu(&neigh->rcu, ipoib_neigh_reclaim); } } @@ -1606,11 +1630,8 @@ int ipoib_dev_init(struct net_device *dev, struct ib_device *ca, int port) /* Allocate RX/TX "rings" to hold queued skbs */ priv->rx_ring = kzalloc(ipoib_recvq_size * sizeof *priv->rx_ring, GFP_KERNEL); - if (!priv->rx_ring) { - printk(KERN_WARNING "%s: failed to allocate RX ring (%d entries)\n", - ca->name, ipoib_recvq_size); + if (!priv->rx_ring) goto out; - } priv->tx_ring = vzalloc(ipoib_sendq_size * sizeof *priv->tx_ring); if (!priv->tx_ring) { @@ -1974,7 +1995,7 @@ int ipoib_add_pkey_attr(struct net_device *dev) return device_create_file(&dev->dev, &dev_attr_pkey); } -int ipoib_set_dev_features(struct ipoib_dev_priv *priv, struct ib_device *hca) +void ipoib_set_dev_features(struct ipoib_dev_priv *priv, struct ib_device *hca) { priv->hca_caps = hca->attrs.device_cap_flags; @@ -1986,8 +2007,6 @@ int ipoib_set_dev_features(struct ipoib_dev_priv *priv, struct ib_device *hca) priv->dev->features |= priv->dev->hw_features; } - - return 0; } static struct net_device *ipoib_add_port(const char *format, @@ -2001,7 +2020,7 @@ static struct net_device *ipoib_add_port(const char *format, if (!priv) goto alloc_mem_failed; - SET_NETDEV_DEV(priv->dev, hca->dma_device); + SET_NETDEV_DEV(priv->dev, hca->dev.parent); priv->dev->dev_id = port - 1; result = ib_query_port(hca, port, &attr); @@ -2016,6 +2035,7 @@ static struct net_device *ipoib_add_port(const char *format, /* MTU will be reset when mcast join happens */ priv->dev->mtu = IPOIB_UD_MTU(priv->max_ib_mtu); priv->mcast_mtu = priv->admin_mtu = priv->dev->mtu; + priv->dev->max_mtu = IPOIB_CM_MTU; priv->dev->neigh_priv_len = sizeof(struct ipoib_neigh); @@ -2026,9 +2046,7 @@ static struct net_device *ipoib_add_port(const char *format, goto device_init_failed; } - result = ipoib_set_dev_features(priv, hca); - if (result) - goto device_init_failed; + ipoib_set_dev_features(priv, hca); /* * Set the full membership bit, so that we join the right diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c index 1909dd252c94..69e146cdc306 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c @@ -314,9 +314,11 @@ static int ipoib_mcast_join_finish(struct ipoib_mcast *mcast, netif_tx_unlock_bh(dev); skb->dev = dev; - if (dev_queue_xmit(skb)) - ipoib_warn(priv, "dev_queue_xmit failed to requeue packet\n"); + ret = dev_queue_xmit(skb); + if (ret) + ipoib_warn(priv, "%s:dev_queue_xmit failed to re-queue packet, ret:%d\n", + __func__, ret); netif_tx_lock_bh(dev); } netif_tx_unlock_bh(dev); @@ -575,8 +577,11 @@ void ipoib_mcast_join_task(struct work_struct *work) if (!test_bit(IPOIB_FLAG_OPER_UP, &priv->flags)) return; - if (ib_query_port(priv->ca, priv->port, &port_attr) || - port_attr.state != IB_PORT_ACTIVE) { + if (ib_query_port(priv->ca, priv->port, &port_attr)) { + ipoib_dbg(priv, "ib_query_port() failed\n"); + return; + } + if (port_attr.state != IB_PORT_ACTIVE) { ipoib_dbg(priv, "port state is not ACTIVE (state = %d) suspending join task\n", port_attr.state); return; @@ -671,7 +676,7 @@ out: spin_unlock_irq(&priv->lock); } -int ipoib_mcast_start_thread(struct net_device *dev) +void ipoib_mcast_start_thread(struct net_device *dev) { struct ipoib_dev_priv *priv = netdev_priv(dev); unsigned long flags; @@ -681,8 +686,6 @@ int ipoib_mcast_start_thread(struct net_device *dev) spin_lock_irqsave(&priv->lock, flags); __ipoib_mcast_schedule_join_thread(priv, NULL, 0); spin_unlock_irqrestore(&priv->lock, flags); - - return 0; } int ipoib_mcast_stop_thread(struct net_device *dev) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c index a2f9f29c6ab5..deedb6fc1b05 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c @@ -35,7 +35,7 @@ #include <linux/init.h> #include <linux/seq_file.h> -#include <asm/uaccess.h> +#include <linux/uaccess.h> #include "ipoib.h" @@ -61,9 +61,7 @@ int __ipoib_vlan_add(struct ipoib_dev_priv *ppriv, struct ipoib_dev_priv *priv, priv->parent = ppriv->dev; set_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags); - result = ipoib_set_dev_features(priv, ppriv->ca); - if (result) - goto err; + ipoib_set_dev_features(priv, ppriv->ca); priv->pkey = pkey; @@ -168,11 +166,11 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey) out: up_write(&ppriv->vlan_rwsem); + rtnl_unlock(); + if (result) free_netdev(priv->dev); - rtnl_unlock(); - return result; } @@ -196,7 +194,6 @@ int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey) list_for_each_entry_safe(priv, tpriv, &ppriv->child_intfs, list) { if (priv->pkey == pkey && priv->child_type == IPOIB_LEGACY_CHILD) { - unregister_netdevice(priv->dev); list_del(&priv->list); dev = priv->dev; break; @@ -204,6 +201,11 @@ int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey) } up_write(&ppriv->vlan_rwsem); + if (dev) { + ipoib_dbg(ppriv, "delete child vlan %s\n", dev->name); + unregister_netdevice(dev); + } + rtnl_unlock(); if (dev) { diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c index 64b3d11dcf1e..5a887efb4bdf 100644 --- a/drivers/infiniband/ulp/iser/iscsi_iser.c +++ b/drivers/infiniband/ulp/iser/iscsi_iser.c @@ -62,7 +62,7 @@ #include <net/sock.h> -#include <asm/uaccess.h> +#include <linux/uaccess.h> #include <scsi/scsi_cmnd.h> #include <scsi/scsi_device.h> @@ -651,15 +651,8 @@ iscsi_iser_session_create(struct iscsi_endpoint *ep, SHOST_DIX_GUARD_CRC); } - /* - * Limit the sg_tablesize and max_sectors based on the device - * max fastreg page list length. - */ - shost->sg_tablesize = min_t(unsigned short, shost->sg_tablesize, - ib_conn->device->ib_device->attrs.max_fast_reg_page_list_len); - if (iscsi_host_add(shost, - ib_conn->device->ib_device->dma_device)) { + ib_conn->device->ib_device->dev.parent)) { mutex_unlock(&iser_conn->state_mutex); goto free_host; } @@ -679,6 +672,10 @@ iscsi_iser_session_create(struct iscsi_endpoint *ep, max_fr_sectors = ((shost->sg_tablesize - 1) * PAGE_SIZE) >> 9; shost->max_sectors = min(iser_max_sectors, max_fr_sectors); + iser_dbg("iser_conn %p, sg_tablesize %u, max_sectors %u\n", + iser_conn, shost->sg_tablesize, + shost->max_sectors); + if (cmds_max > max_cmds) { iser_info("cmds_max changed from %u to %u\n", cmds_max, max_cmds); @@ -997,6 +994,7 @@ static struct scsi_host_template iscsi_iser_sht = { .change_queue_depth = scsi_change_queue_depth, .sg_tablesize = ISCSI_ISER_DEF_SG_TABLESIZE, .cmd_per_lun = ISER_DEF_CMD_PER_LUN, + .eh_timed_out = iscsi_eh_cmd_timed_out, .eh_abort_handler = iscsi_eh_abort, .eh_device_reset_handler= iscsi_eh_device_reset, .eh_target_reset_handler = iscsi_eh_recover_target, diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.h b/drivers/infiniband/ulp/iser/iscsi_iser.h index 0be6a7c5ddb5..9d0b22ad58c1 100644 --- a/drivers/infiniband/ulp/iser/iscsi_iser.h +++ b/drivers/infiniband/ulp/iser/iscsi_iser.h @@ -496,7 +496,6 @@ struct ib_conn { * @rx_descs: rx buffers array (cyclic buffer) * @num_rx_descs: number of rx descriptors * @scsi_sg_tablesize: scsi host sg_tablesize - * @scsi_max_sectors: scsi host max sectors */ struct iser_conn { struct ib_conn ib_conn; @@ -519,7 +518,6 @@ struct iser_conn { struct iser_rx_desc *rx_descs; u32 num_rx_descs; unsigned short scsi_sg_tablesize; - unsigned int scsi_max_sectors; bool snd_w_inv; }; diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c index a4b791dfaa1d..30b622f2ab73 100644 --- a/drivers/infiniband/ulp/iser/iser_verbs.c +++ b/drivers/infiniband/ulp/iser/iser_verbs.c @@ -597,7 +597,9 @@ static void iser_free_ib_conn_res(struct iser_conn *iser_conn, iser_conn, ib_conn->cma_id, ib_conn->qp); if (ib_conn->qp != NULL) { + mutex_lock(&ig.connlist_mutex); ib_conn->comp->active_qps--; + mutex_unlock(&ig.connlist_mutex); rdma_destroy_qp(ib_conn->cma_id); ib_conn->qp = NULL; } @@ -707,18 +709,7 @@ iser_calc_scsi_params(struct iser_conn *iser_conn, sup_sg_tablesize = min_t(unsigned, ISCSI_ISER_MAX_SG_TABLESIZE, device->ib_device->attrs.max_fast_reg_page_list_len); - if (sg_tablesize > sup_sg_tablesize) { - sg_tablesize = sup_sg_tablesize; - iser_conn->scsi_max_sectors = sg_tablesize * SIZE_4K / 512; - } else { - iser_conn->scsi_max_sectors = max_sectors; - } - - iser_conn->scsi_sg_tablesize = sg_tablesize; - - iser_dbg("iser_conn %p, sg_tablesize %u, max_sectors %u\n", - iser_conn, iser_conn->scsi_sg_tablesize, - iser_conn->scsi_max_sectors); + iser_conn->scsi_sg_tablesize = min(sg_tablesize, sup_sg_tablesize); } /** @@ -890,11 +881,14 @@ static int iser_cma_handler(struct rdma_cm_id *cma_id, struct rdma_cm_event *eve case RDMA_CM_EVENT_ESTABLISHED: iser_connected_handler(cma_id, event->param.conn.private_data); break; + case RDMA_CM_EVENT_REJECTED: + iser_info("Connection rejected: %s\n", + rdma_reject_msg(cma_id, event->status)); + /* FALLTHROUGH */ case RDMA_CM_EVENT_ADDR_ERROR: case RDMA_CM_EVENT_ROUTE_ERROR: case RDMA_CM_EVENT_CONNECT_ERROR: case RDMA_CM_EVENT_UNREACHABLE: - case RDMA_CM_EVENT_REJECTED: iser_connect_error(cma_id); break; case RDMA_CM_EVENT_DISCONNECTED: diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index 6dd43f63238e..91cbe86b25c8 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -184,7 +184,7 @@ isert_alloc_rx_descriptors(struct isert_conn *isert_conn) isert_conn->rx_descs = kzalloc(ISERT_QP_MAX_RECV_DTOS * sizeof(struct iser_rx_desc), GFP_KERNEL); if (!isert_conn->rx_descs) - goto fail; + return -ENOMEM; rx_desc = isert_conn->rx_descs; @@ -213,9 +213,7 @@ dma_map_fail: } kfree(isert_conn->rx_descs); isert_conn->rx_descs = NULL; -fail: isert_err("conn %p failed to allocate rx descriptors\n", isert_conn); - return -ENOMEM; } @@ -269,10 +267,8 @@ isert_alloc_comps(struct isert_device *device) device->comps = kcalloc(device->comps_used, sizeof(struct isert_comp), GFP_KERNEL); - if (!device->comps) { - isert_err("Unable to allocate completion contexts\n"); + if (!device->comps) return -ENOMEM; - } max_cqe = min(ISER_MAX_CQ_LEN, device->ib_device->attrs.max_cqe); @@ -432,10 +428,8 @@ isert_alloc_login_buf(struct isert_conn *isert_conn, isert_conn->login_req_buf = kzalloc(sizeof(*isert_conn->login_req_buf), GFP_KERNEL); - if (!isert_conn->login_req_buf) { - isert_err("Unable to allocate isert_conn->login_buf\n"); + if (!isert_conn->login_req_buf) return -ENOMEM; - } isert_conn->login_req_dma = ib_dma_map_single(ib_dev, isert_conn->login_req_buf, @@ -734,7 +728,7 @@ isert_disconnected_handler(struct rdma_cm_id *cma_id, iscsit_cause_connection_reinstatement(isert_conn->conn, 0); break; default: - isert_warn("conn %p teminating in state %d\n", + isert_warn("conn %p terminating in state %d\n", isert_conn, isert_conn->state); } mutex_unlock(&isert_conn->mutex); @@ -795,6 +789,8 @@ isert_cma_handler(struct rdma_cm_id *cma_id, struct rdma_cm_event *event) */ return 1; case RDMA_CM_EVENT_REJECTED: /* FALLTHRU */ + isert_info("Connection rejected: %s\n", + rdma_reject_msg(cma_id, event->status)); case RDMA_CM_EVENT_UNREACHABLE: /* FALLTHRU */ case RDMA_CM_EVENT_CONNECT_ERROR: ret = isert_connect_error(cma_id); @@ -1276,11 +1272,8 @@ isert_handle_text_cmd(struct isert_conn *isert_conn, struct isert_cmd *isert_cmd if (payload_length) { text_in = kzalloc(payload_length, GFP_KERNEL); - if (!text_in) { - isert_err("Unable to allocate text_in of payload_length: %u\n", - payload_length); + if (!text_in) return -ENOMEM; - } } cmd->text_in_ptr = text_in; @@ -1851,6 +1844,8 @@ 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; @@ -1978,6 +1973,8 @@ 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; @@ -2018,6 +2015,8 @@ isert_put_text_rsp(struct iscsi_cmd *cmd, struct iscsi_conn *conn) 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; @@ -2307,10 +2306,9 @@ isert_setup_np(struct iscsi_np *np, int ret; isert_np = kzalloc(sizeof(struct isert_np), GFP_KERNEL); - if (!isert_np) { - isert_err("Unable to allocate struct isert_np\n"); + if (!isert_np) return -ENOMEM; - } + sema_init(&isert_np->sem, 0); mutex_init(&isert_np->mutex); INIT_LIST_HEAD(&isert_np->accepted); @@ -2651,7 +2649,6 @@ static int __init isert_init(void) WQ_UNBOUND | WQ_HIGHPRI, 0); if (!isert_comp_wq) { isert_err("Unable to allocate isert_comp_wq\n"); - ret = -ENOMEM; return -ENOMEM; } diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c index d980fb458ad4..cee46266f434 100644 --- a/drivers/infiniband/ulp/srp/ib_srp.c +++ b/drivers/infiniband/ulp/srp/ib_srp.c @@ -40,6 +40,7 @@ #include <linux/parser.h> #include <linux/random.h> #include <linux/jiffies.h> +#include <linux/lockdep.h> #include <rdma/ib_cache.h> #include <linux/atomic.h> @@ -64,6 +65,11 @@ MODULE_LICENSE("Dual BSD/GPL"); MODULE_VERSION(DRV_VERSION); MODULE_INFO(release_date, DRV_RELDATE); +#if !defined(CONFIG_DYNAMIC_DEBUG) +#define DEFINE_DYNAMIC_DEBUG_METADATA(name, fmt) +#define DYNAMIC_DEBUG_BRANCH(descriptor) false +#endif + static unsigned int srp_sg_tablesize; static unsigned int cmd_sg_entries; static unsigned int indirect_sg_entries; @@ -384,6 +390,9 @@ static struct srp_fr_pool *srp_create_fr_pool(struct ib_device *device, max_page_list_len); if (IS_ERR(mr)) { ret = PTR_ERR(mr); + if (ret == -ENOMEM) + pr_info("%s: ib_alloc_mr() failed. Try to reduce max_cmd_per_lun, max_sect or ch_count\n", + dev_name(&device->dev)); goto destroy_pool; } d->mr = mr; @@ -457,9 +466,13 @@ static struct srp_fr_pool *srp_alloc_fr_pool(struct srp_target_port *target) * completion handler can access the queue pair while it is * being destroyed. */ -static void srp_destroy_qp(struct ib_qp *qp) +static void srp_destroy_qp(struct srp_rdma_ch *ch, struct ib_qp *qp) { - ib_drain_rq(qp); + spin_lock_irq(&ch->lock); + ib_process_cq_direct(ch->send_cq, -1); + spin_unlock_irq(&ch->lock); + + ib_drain_qp(qp); ib_destroy_qp(qp); } @@ -533,7 +546,7 @@ static int srp_create_ch_ib(struct srp_rdma_ch *ch) } if (ch->qp) - srp_destroy_qp(ch->qp); + srp_destroy_qp(ch, ch->qp); if (ch->recv_cq) ib_free_cq(ch->recv_cq); if (ch->send_cq) @@ -557,7 +570,7 @@ static int srp_create_ch_ib(struct srp_rdma_ch *ch) return 0; err_qp: - srp_destroy_qp(qp); + srp_destroy_qp(ch, qp); err_send_cq: ib_free_cq(send_cq); @@ -600,7 +613,7 @@ static void srp_free_ch_ib(struct srp_target_port *target, ib_destroy_fmr_pool(ch->fmr_pool); } - srp_destroy_qp(ch->qp); + srp_destroy_qp(ch, ch->qp); ib_free_cq(ch->send_cq); ib_free_cq(ch->recv_cq); @@ -1266,8 +1279,12 @@ static int srp_map_finish_fmr(struct srp_map_state *state, struct ib_pool_fmr *fmr; u64 io_addr = 0; - if (state->fmr.next >= state->fmr.end) + if (state->fmr.next >= state->fmr.end) { + shost_printk(KERN_ERR, ch->target->scsi_host, + PFX "Out of MRs (mr_per_cmd = %d)\n", + ch->target->mr_per_cmd); return -ENOMEM; + } WARN_ON_ONCE(!dev->use_fmr); @@ -1323,8 +1340,12 @@ static int srp_map_finish_fr(struct srp_map_state *state, u32 rkey; int n, err; - if (state->fr.next >= state->fr.end) + if (state->fr.next >= state->fr.end) { + shost_printk(KERN_ERR, ch->target->scsi_host, + PFX "Out of MRs (mr_per_cmd = %d)\n", + ch->target->mr_per_cmd); return -ENOMEM; + } WARN_ON_ONCE(!dev->use_fast_reg); @@ -1556,7 +1577,6 @@ static int srp_map_idb(struct srp_rdma_ch *ch, struct srp_request *req, return 0; } -#if defined(DYNAMIC_DATA_DEBUG) static void srp_check_mapping(struct srp_map_state *state, struct srp_rdma_ch *ch, struct srp_request *req, struct scatterlist *scat, int count) @@ -1580,7 +1600,6 @@ static void srp_check_mapping(struct srp_map_state *state, scsi_bufflen(req->scmnd), desc_len, mr_len, state->ndesc, state->nmdesc); } -#endif /** * srp_map_data() - map SCSI data buffer onto an SRP request @@ -1669,14 +1688,12 @@ static int srp_map_data(struct scsi_cmnd *scmnd, struct srp_rdma_ch *ch, if (ret < 0) goto unmap; -#if defined(DYNAMIC_DEBUG) { DEFINE_DYNAMIC_DEBUG_METADATA(ddm, "Memory mapping consistency check"); - if (unlikely(ddm.flags & _DPRINTK_FLAGS_PRINT)) + if (DYNAMIC_DEBUG_BRANCH(ddm)) srp_check_mapping(&state, ch, req, scat, count); } -#endif /* We've mapped the request, now pull as much of the indirect * descriptor table as we can into the command buffer. If this @@ -1787,6 +1804,8 @@ static struct srp_iu *__srp_get_tx_iu(struct srp_rdma_ch *ch, s32 rsv = (iu_type == SRP_IU_TSK_MGMT) ? 0 : SRP_TSK_MGMT_SQ_SIZE; struct srp_iu *iu; + lockdep_assert_held(&ch->lock); + ib_process_cq_direct(ch->send_cq, -1); if (list_empty(&ch->free_tx)) @@ -1807,6 +1826,11 @@ static struct srp_iu *__srp_get_tx_iu(struct srp_rdma_ch *ch, return iu; } +/* + * Note: if this function is called from inside ib_drain_sq() then it will + * be called without ch->lock being held. If ib_drain_sq() dequeues a WQE + * with status IB_WC_SUCCESS then that's a bug. + */ static void srp_send_done(struct ib_cq *cq, struct ib_wc *wc) { struct srp_iu *iu = container_of(wc->wr_cqe, struct srp_iu, cqe); @@ -1817,6 +1841,8 @@ static void srp_send_done(struct ib_cq *cq, struct ib_wc *wc) return; } + lockdep_assert_held(&ch->lock); + list_add(&iu->list, &ch->free_tx); } @@ -1872,17 +1898,24 @@ static void srp_process_rsp(struct srp_rdma_ch *ch, struct srp_rsp *rsp) if (unlikely(rsp->tag & SRP_TAG_TSK_MGMT)) { spin_lock_irqsave(&ch->lock, flags); ch->req_lim += be32_to_cpu(rsp->req_lim_delta); + if (rsp->tag == ch->tsk_mgmt_tag) { + ch->tsk_mgmt_status = -1; + if (be32_to_cpu(rsp->resp_data_len) >= 4) + ch->tsk_mgmt_status = rsp->data[3]; + complete(&ch->tsk_mgmt_done); + } else { + shost_printk(KERN_ERR, target->scsi_host, + "Received tsk mgmt response too late for tag %#llx\n", + rsp->tag); + } spin_unlock_irqrestore(&ch->lock, flags); - - ch->tsk_mgmt_status = -1; - if (be32_to_cpu(rsp->resp_data_len) >= 4) - ch->tsk_mgmt_status = rsp->data[3]; - complete(&ch->tsk_mgmt_done); } else { scmnd = scsi_host_find_tag(target->scsi_host, rsp->tag); - if (scmnd) { + if (scmnd && scmnd->host_scribble) { req = (void *)scmnd->host_scribble; scmnd = srp_claim_req(ch, req, NULL, scmnd); + } else { + scmnd = NULL; } if (!scmnd) { shost_printk(KERN_ERR, target->scsi_host, @@ -2514,19 +2547,18 @@ srp_change_queue_depth(struct scsi_device *sdev, int qdepth) } static int srp_send_tsk_mgmt(struct srp_rdma_ch *ch, u64 req_tag, u64 lun, - u8 func) + u8 func, u8 *status) { struct srp_target_port *target = ch->target; struct srp_rport *rport = target->rport; struct ib_device *dev = target->srp_host->srp_dev->dev; struct srp_iu *iu; struct srp_tsk_mgmt *tsk_mgmt; + int res; if (!ch->connected || target->qp_in_error) return -1; - init_completion(&ch->tsk_mgmt_done); - /* * Lock the rport mutex to avoid that srp_create_ch_ib() is * invoked while a task management function is being sent. @@ -2549,10 +2581,16 @@ static int srp_send_tsk_mgmt(struct srp_rdma_ch *ch, u64 req_tag, u64 lun, tsk_mgmt->opcode = SRP_TSK_MGMT; int_to_scsilun(lun, &tsk_mgmt->lun); - tsk_mgmt->tag = req_tag | SRP_TAG_TSK_MGMT; tsk_mgmt->tsk_mgmt_func = func; tsk_mgmt->task_tag = req_tag; + spin_lock_irq(&ch->lock); + ch->tsk_mgmt_tag = (ch->tsk_mgmt_tag + 1) | SRP_TAG_TSK_MGMT; + tsk_mgmt->tag = ch->tsk_mgmt_tag; + spin_unlock_irq(&ch->lock); + + init_completion(&ch->tsk_mgmt_done); + ib_dma_sync_single_for_device(dev, iu->dma, sizeof *tsk_mgmt, DMA_TO_DEVICE); if (srp_post_send(ch, iu, sizeof(*tsk_mgmt))) { @@ -2561,13 +2599,15 @@ static int srp_send_tsk_mgmt(struct srp_rdma_ch *ch, u64 req_tag, u64 lun, return -1; } + res = wait_for_completion_timeout(&ch->tsk_mgmt_done, + msecs_to_jiffies(SRP_ABORT_TIMEOUT_MS)); + if (res > 0 && status) + *status = ch->tsk_mgmt_status; mutex_unlock(&rport->mutex); - if (!wait_for_completion_timeout(&ch->tsk_mgmt_done, - msecs_to_jiffies(SRP_ABORT_TIMEOUT_MS))) - return -1; + WARN_ON_ONCE(res < 0); - return 0; + return res > 0 ? 0 : -1; } static int srp_abort(struct scsi_cmnd *scmnd) @@ -2593,7 +2633,7 @@ static int srp_abort(struct scsi_cmnd *scmnd) shost_printk(KERN_ERR, target->scsi_host, "Sending SRP abort for tag %#x\n", tag); if (srp_send_tsk_mgmt(ch, tag, scmnd->device->lun, - SRP_TSK_ABORT_TASK) == 0) + SRP_TSK_ABORT_TASK, NULL) == 0) ret = SUCCESS; else if (target->rport->state == SRP_RPORT_LOST) ret = FAST_IO_FAIL; @@ -2611,14 +2651,15 @@ static int srp_reset_device(struct scsi_cmnd *scmnd) struct srp_target_port *target = host_to_target(scmnd->device->host); struct srp_rdma_ch *ch; int i; + u8 status; shost_printk(KERN_ERR, target->scsi_host, "SRP reset_device called\n"); ch = &target->ch[0]; if (srp_send_tsk_mgmt(ch, SRP_TAG_NO_REQ, scmnd->device->lun, - SRP_TSK_LUN_RESET)) + SRP_TSK_LUN_RESET, &status)) return FAILED; - if (ch->tsk_mgmt_status) + if (status) return FAILED; for (i = 0; i < target->ch_count; i++) { @@ -2647,9 +2688,8 @@ static int srp_slave_alloc(struct scsi_device *sdev) struct Scsi_Host *shost = sdev->host; struct srp_target_port *target = host_to_target(shost); struct srp_device *srp_dev = target->srp_host->srp_dev; - struct ib_device *ibdev = srp_dev->dev; - if (!(ibdev->attrs.device_cap_flags & IB_DEVICE_SG_GAPS_REG)) + if (true) blk_queue_virt_boundary(sdev->request_queue, ~srp_dev->mr_page_mask); @@ -2852,6 +2892,7 @@ static struct scsi_host_template srp_template = { .info = srp_target_info, .queuecommand = srp_queuecommand, .change_queue_depth = srp_change_queue_depth, + .eh_timed_out = srp_timed_out, .eh_abort_handler = srp_abort, .eh_device_reset_handler = srp_reset_device, .eh_host_reset_handler = srp_reset_host, @@ -2892,7 +2933,7 @@ static int srp_add_target(struct srp_host *host, struct srp_target_port *target) sprintf(target->target_name, "SRP.T10:%016llX", be64_to_cpu(target->id_ext)); - if (scsi_add_host(target->scsi_host, host->srp_dev->dev->dma_device)) + if (scsi_add_host(target->scsi_host, host->srp_dev->dev->dev.parent)) return -ENODEV; memcpy(ids.port_id, &target->id_ext, 8); @@ -3287,7 +3328,9 @@ static ssize_t srp_create_target(struct device *dev, */ scsi_host_get(target->scsi_host); - mutex_lock(&host->add_target_mutex); + ret = mutex_lock_interruptible(&host->add_target_mutex); + if (ret < 0) + goto put; ret = srp_parse_options(buf, target); if (ret) @@ -3402,11 +3445,12 @@ static ssize_t srp_create_target(struct device *dev, ret = srp_connect_ch(ch, multich); if (ret) { shost_printk(KERN_ERR, target->scsi_host, - PFX "Connection %d/%d failed\n", + PFX "Connection %d/%d to %pI6 failed\n", ch_start + cpu_idx, - target->ch_count); + target->ch_count, + ch->target->orig_dgid.raw); if (node_idx == 0 && cpu_idx == 0) { - goto err_disconnect; + goto free_ch; } else { srp_free_ch_ib(target, ch); srp_free_req_data(target, ch); @@ -3443,6 +3487,7 @@ connected: out: mutex_unlock(&host->add_target_mutex); +put: scsi_host_put(target->scsi_host); if (ret < 0) scsi_host_put(target->scsi_host); @@ -3452,6 +3497,7 @@ out: err_disconnect: srp_disconnect_target(target); +free_ch: for (i = 0; i < target->ch_count; i++) { ch = &target->ch[i]; srp_free_ch_ib(target, ch); @@ -3500,7 +3546,7 @@ static struct srp_host *srp_add_port(struct srp_device *device, u8 port) host->port = port; host->dev.class = &srp_class; - host->dev.parent = device->dev->dma_device; + host->dev.parent = device->dev->dev.parent; dev_set_name(&host->dev, "srp-%s-%d", device->dev->name, port); if (device_register(&host->dev)) @@ -3526,6 +3572,7 @@ free_host: static void srp_add_one(struct ib_device *device) { struct srp_device *srp_dev; + struct ib_device_attr *attr = &device->attrs; struct srp_host *host; int mr_page_shift, p; u64 max_pages_per_mr; @@ -3540,25 +3587,25 @@ static void srp_add_one(struct ib_device *device) * minimum of 4096 bytes. We're unlikely to build large sglists * out of smaller entries. */ - mr_page_shift = max(12, ffs(device->attrs.page_size_cap) - 1); + mr_page_shift = max(12, ffs(attr->page_size_cap) - 1); srp_dev->mr_page_size = 1 << mr_page_shift; srp_dev->mr_page_mask = ~((u64) srp_dev->mr_page_size - 1); - max_pages_per_mr = device->attrs.max_mr_size; + max_pages_per_mr = attr->max_mr_size; do_div(max_pages_per_mr, srp_dev->mr_page_size); pr_debug("%s: %llu / %u = %llu <> %u\n", __func__, - device->attrs.max_mr_size, srp_dev->mr_page_size, + attr->max_mr_size, srp_dev->mr_page_size, max_pages_per_mr, SRP_MAX_PAGES_PER_MR); srp_dev->max_pages_per_mr = min_t(u64, SRP_MAX_PAGES_PER_MR, max_pages_per_mr); srp_dev->has_fmr = (device->alloc_fmr && device->dealloc_fmr && device->map_phys_fmr && device->unmap_fmr); - srp_dev->has_fr = (device->attrs.device_cap_flags & + srp_dev->has_fr = (attr->device_cap_flags & IB_DEVICE_MEM_MGT_EXTENSIONS); if (!never_register && !srp_dev->has_fmr && !srp_dev->has_fr) { dev_warn(&device->dev, "neither FMR nor FR is supported\n"); } else if (!never_register && - device->attrs.max_mr_size >= 2 * srp_dev->mr_page_size) { + attr->max_mr_size >= 2 * srp_dev->mr_page_size) { srp_dev->use_fast_reg = (srp_dev->has_fr && (!srp_dev->has_fmr || prefer_fr)); srp_dev->use_fmr = !srp_dev->use_fast_reg && srp_dev->has_fmr; @@ -3571,13 +3618,13 @@ static void srp_add_one(struct ib_device *device) if (srp_dev->use_fast_reg) { srp_dev->max_pages_per_mr = min_t(u32, srp_dev->max_pages_per_mr, - device->attrs.max_fast_reg_page_list_len); + attr->max_fast_reg_page_list_len); } srp_dev->mr_max_size = srp_dev->mr_page_size * srp_dev->max_pages_per_mr; pr_debug("%s: mr_page_shift = %d, device->max_mr_size = %#llx, device->max_fast_reg_page_list_len = %u, max_pages_per_mr = %d, mr_max_size = %#x\n", - device->name, mr_page_shift, device->attrs.max_mr_size, - device->attrs.max_fast_reg_page_list_len, + device->name, mr_page_shift, attr->max_mr_size, + attr->max_fast_reg_page_list_len, srp_dev->max_pages_per_mr, srp_dev->mr_max_size); INIT_LIST_HEAD(&srp_dev->dev_list); @@ -3678,6 +3725,12 @@ static int __init srp_init_module(void) indirect_sg_entries = cmd_sg_entries; } + if (indirect_sg_entries > SG_MAX_SEGMENTS) { + pr_warn("Clamping indirect_sg_entries to %u\n", + SG_MAX_SEGMENTS); + indirect_sg_entries = SG_MAX_SEGMENTS; + } + srp_remove_wq = create_workqueue("srp_remove"); if (!srp_remove_wq) { ret = -ENOMEM; diff --git a/drivers/infiniband/ulp/srp/ib_srp.h b/drivers/infiniband/ulp/srp/ib_srp.h index 21c69695f9d4..32ed40db3ca2 100644 --- a/drivers/infiniband/ulp/srp/ib_srp.h +++ b/drivers/infiniband/ulp/srp/ib_srp.h @@ -163,6 +163,7 @@ struct srp_rdma_ch { int max_ti_iu_len; int comp_vector; + u64 tsk_mgmt_tag; struct completion tsk_mgmt_done; u8 tsk_mgmt_status; bool connected; diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.c b/drivers/infiniband/ulp/srpt/ib_srpt.c index 0b1f69ed2e92..7e314c2f2071 100644 --- a/drivers/infiniband/ulp/srpt/ib_srpt.c +++ b/drivers/infiniband/ulp/srpt/ib_srpt.c @@ -500,6 +500,7 @@ static int srpt_refresh_port(struct srpt_port *sport) struct ib_mad_reg_req reg_req; struct ib_port_modify port_modify; struct ib_port_attr port_attr; + __be16 *guid; int ret; memset(&port_modify, 0, sizeof(port_modify)); @@ -522,10 +523,17 @@ static int srpt_refresh_port(struct srpt_port *sport) if (ret) goto err_query_port; + sport->port_guid_wwn.priv = sport; + guid = (__be16 *)&sport->gid.global.interface_id; snprintf(sport->port_guid, sizeof(sport->port_guid), - "0x%016llx%016llx", - be64_to_cpu(sport->gid.global.subnet_prefix), - be64_to_cpu(sport->gid.global.interface_id)); + "%04x:%04x:%04x:%04x", + be16_to_cpu(guid[0]), be16_to_cpu(guid[1]), + be16_to_cpu(guid[2]), be16_to_cpu(guid[3])); + sport->port_gid_wwn.priv = sport; + snprintf(sport->port_gid, sizeof(sport->port_gid), + "0x%016llx%016llx", + be64_to_cpu(sport->gid.global.subnet_prefix), + be64_to_cpu(sport->gid.global.interface_id)); if (!sport->mad_agent) { memset(®_req, 0, sizeof(reg_req)); @@ -1838,9 +1846,9 @@ static int srpt_cm_req_recv(struct ib_cm_id *cm_id, struct srp_login_rej *rej; struct ib_cm_rep_param *rep_param; struct srpt_rdma_ch *ch, *tmp_ch; + __be16 *guid; u32 it_iu_len; int i, ret = 0; - unsigned char *p; WARN_ON_ONCE(irqs_disabled()); @@ -1984,31 +1992,32 @@ static int srpt_cm_req_recv(struct ib_cm_id *cm_id, goto destroy_ib; } - /* - * Use the initator port identifier as the session name, when - * checking against se_node_acl->initiatorname[] this can be - * with or without preceeding '0x'. - */ + guid = (__be16 *)¶m->primary_path->sgid.global.interface_id; + snprintf(ch->ini_guid, sizeof(ch->ini_guid), "%04x:%04x:%04x:%04x", + be16_to_cpu(guid[0]), be16_to_cpu(guid[1]), + be16_to_cpu(guid[2]), be16_to_cpu(guid[3])); snprintf(ch->sess_name, sizeof(ch->sess_name), "0x%016llx%016llx", be64_to_cpu(*(__be64 *)ch->i_port_id), be64_to_cpu(*(__be64 *)(ch->i_port_id + 8))); pr_debug("registering session %s\n", ch->sess_name); - p = &ch->sess_name[0]; - -try_again: - ch->sess = target_alloc_session(&sport->port_tpg_1, 0, 0, - TARGET_PROT_NORMAL, p, ch, NULL); - if (IS_ERR(ch->sess)) { - pr_info("Rejected login because no ACL has been" - " configured yet for initiator %s.\n", p); - /* - * XXX: Hack to retry of ch->i_port_id without leading '0x' - */ - if (p == &ch->sess_name[0]) { - p += 2; - goto try_again; - } + + if (sport->port_guid_tpg.se_tpg_wwn) + ch->sess = target_alloc_session(&sport->port_guid_tpg, 0, 0, + TARGET_PROT_NORMAL, + ch->ini_guid, ch, NULL); + if (sport->port_gid_tpg.se_tpg_wwn && IS_ERR_OR_NULL(ch->sess)) + ch->sess = target_alloc_session(&sport->port_gid_tpg, 0, 0, + TARGET_PROT_NORMAL, ch->sess_name, ch, + NULL); + /* Retry without leading "0x" */ + if (sport->port_gid_tpg.se_tpg_wwn && IS_ERR_OR_NULL(ch->sess)) + ch->sess = target_alloc_session(&sport->port_gid_tpg, 0, 0, + TARGET_PROT_NORMAL, + ch->sess_name + 2, ch, NULL); + if (IS_ERR_OR_NULL(ch->sess)) { + pr_info("Rejected login because no ACL has been configured yet for initiator %s.\n", + ch->sess_name); rej->reason = cpu_to_be32((PTR_ERR(ch->sess) == -ENOMEM) ? SRP_LOGIN_REJ_INSUFFICIENT_RESOURCES : SRP_LOGIN_REJ_CHANNEL_LIMIT_REACHED); @@ -2424,7 +2433,7 @@ static int srpt_release_sdev(struct srpt_device *sdev) return 0; } -static struct srpt_port *__srpt_lookup_port(const char *name) +static struct se_wwn *__srpt_lookup_wwn(const char *name) { struct ib_device *dev; struct srpt_device *sdev; @@ -2439,23 +2448,25 @@ static struct srpt_port *__srpt_lookup_port(const char *name) for (i = 0; i < dev->phys_port_cnt; i++) { sport = &sdev->port[i]; - if (!strcmp(sport->port_guid, name)) - return sport; + if (strcmp(sport->port_guid, name) == 0) + return &sport->port_guid_wwn; + if (strcmp(sport->port_gid, name) == 0) + return &sport->port_gid_wwn; } } return NULL; } -static struct srpt_port *srpt_lookup_port(const char *name) +static struct se_wwn *srpt_lookup_wwn(const char *name) { - struct srpt_port *sport; + struct se_wwn *wwn; spin_lock(&srpt_dev_lock); - sport = __srpt_lookup_port(name); + wwn = __srpt_lookup_wwn(name); spin_unlock(&srpt_dev_lock); - return sport; + return wwn; } /** @@ -2468,8 +2479,7 @@ static void srpt_add_one(struct ib_device *device) struct ib_srq_init_attr srq_attr; int i; - pr_debug("device = %p, device->dma_ops = %p\n", device, - device->dma_ops); + pr_debug("device = %p\n", device); sdev = kzalloc(sizeof(*sdev), GFP_KERNEL); if (!sdev) @@ -2647,11 +2657,19 @@ static char *srpt_get_fabric_name(void) return "srpt"; } +static struct srpt_port *srpt_tpg_to_sport(struct se_portal_group *tpg) +{ + return tpg->se_tpg_wwn->priv; +} + static char *srpt_get_fabric_wwn(struct se_portal_group *tpg) { - struct srpt_port *sport = container_of(tpg, struct srpt_port, port_tpg_1); + struct srpt_port *sport = srpt_tpg_to_sport(tpg); - return sport->port_guid; + WARN_ON_ONCE(tpg != &sport->port_guid_tpg && + tpg != &sport->port_gid_tpg); + return tpg == &sport->port_guid_tpg ? sport->port_guid : + sport->port_gid; } static u16 srpt_get_tag(struct se_portal_group *tpg) @@ -2741,6 +2759,19 @@ static int srpt_get_tcm_cmd_state(struct se_cmd *se_cmd) return srpt_get_cmd_state(ioctx); } +static int srpt_parse_guid(u64 *guid, const char *name) +{ + u16 w[4]; + int ret = -EINVAL; + + if (sscanf(name, "%hx:%hx:%hx:%hx", &w[0], &w[1], &w[2], &w[3]) != 4) + goto out; + *guid = get_unaligned_be64(w); + ret = 0; +out: + return ret; +} + /** * srpt_parse_i_port_id() - Parse an initiator port ID. * @name: ASCII representation of a 128-bit initiator port ID. @@ -2776,20 +2807,23 @@ out: */ static int srpt_init_nodeacl(struct se_node_acl *se_nacl, const char *name) { + u64 guid; u8 i_port_id[16]; + int ret; - if (srpt_parse_i_port_id(i_port_id, name) < 0) { + ret = srpt_parse_guid(&guid, name); + if (ret < 0) + ret = srpt_parse_i_port_id(i_port_id, name); + if (ret < 0) pr_err("invalid initiator port ID %s\n", name); - return -EINVAL; - } - return 0; + return ret; } static ssize_t srpt_tpg_attrib_srp_max_rdma_size_show(struct config_item *item, char *page) { struct se_portal_group *se_tpg = attrib_to_tpg(item); - struct srpt_port *sport = container_of(se_tpg, struct srpt_port, port_tpg_1); + struct srpt_port *sport = srpt_tpg_to_sport(se_tpg); return sprintf(page, "%u\n", sport->port_attrib.srp_max_rdma_size); } @@ -2798,7 +2832,7 @@ static ssize_t srpt_tpg_attrib_srp_max_rdma_size_store(struct config_item *item, const char *page, size_t count) { struct se_portal_group *se_tpg = attrib_to_tpg(item); - struct srpt_port *sport = container_of(se_tpg, struct srpt_port, port_tpg_1); + struct srpt_port *sport = srpt_tpg_to_sport(se_tpg); unsigned long val; int ret; @@ -2826,7 +2860,7 @@ static ssize_t srpt_tpg_attrib_srp_max_rsp_size_show(struct config_item *item, char *page) { struct se_portal_group *se_tpg = attrib_to_tpg(item); - struct srpt_port *sport = container_of(se_tpg, struct srpt_port, port_tpg_1); + struct srpt_port *sport = srpt_tpg_to_sport(se_tpg); return sprintf(page, "%u\n", sport->port_attrib.srp_max_rsp_size); } @@ -2835,7 +2869,7 @@ static ssize_t srpt_tpg_attrib_srp_max_rsp_size_store(struct config_item *item, const char *page, size_t count) { struct se_portal_group *se_tpg = attrib_to_tpg(item); - struct srpt_port *sport = container_of(se_tpg, struct srpt_port, port_tpg_1); + struct srpt_port *sport = srpt_tpg_to_sport(se_tpg); unsigned long val; int ret; @@ -2863,7 +2897,7 @@ static ssize_t srpt_tpg_attrib_srp_sq_size_show(struct config_item *item, char *page) { struct se_portal_group *se_tpg = attrib_to_tpg(item); - struct srpt_port *sport = container_of(se_tpg, struct srpt_port, port_tpg_1); + struct srpt_port *sport = srpt_tpg_to_sport(se_tpg); return sprintf(page, "%u\n", sport->port_attrib.srp_sq_size); } @@ -2872,7 +2906,7 @@ static ssize_t srpt_tpg_attrib_srp_sq_size_store(struct config_item *item, const char *page, size_t count) { struct se_portal_group *se_tpg = attrib_to_tpg(item); - struct srpt_port *sport = container_of(se_tpg, struct srpt_port, port_tpg_1); + struct srpt_port *sport = srpt_tpg_to_sport(se_tpg); unsigned long val; int ret; @@ -2910,7 +2944,7 @@ static struct configfs_attribute *srpt_tpg_attrib_attrs[] = { static ssize_t srpt_tpg_enable_show(struct config_item *item, char *page) { struct se_portal_group *se_tpg = to_tpg(item); - struct srpt_port *sport = container_of(se_tpg, struct srpt_port, port_tpg_1); + struct srpt_port *sport = srpt_tpg_to_sport(se_tpg); return snprintf(page, PAGE_SIZE, "%d\n", (sport->enabled) ? 1: 0); } @@ -2919,7 +2953,7 @@ static ssize_t srpt_tpg_enable_store(struct config_item *item, const char *page, size_t count) { struct se_portal_group *se_tpg = to_tpg(item); - struct srpt_port *sport = container_of(se_tpg, struct srpt_port, port_tpg_1); + struct srpt_port *sport = srpt_tpg_to_sport(se_tpg); struct srpt_device *sdev = sport->sdev; struct srpt_rdma_ch *ch; unsigned long tmp; @@ -2971,15 +3005,19 @@ static struct se_portal_group *srpt_make_tpg(struct se_wwn *wwn, struct config_group *group, const char *name) { - struct srpt_port *sport = container_of(wwn, struct srpt_port, port_wwn); + struct srpt_port *sport = wwn->priv; + static struct se_portal_group *tpg; int res; - /* Initialize sport->port_wwn and sport->port_tpg_1 */ - res = core_tpg_register(&sport->port_wwn, &sport->port_tpg_1, SCSI_PROTOCOL_SRP); + WARN_ON_ONCE(wwn != &sport->port_guid_wwn && + wwn != &sport->port_gid_wwn); + tpg = wwn == &sport->port_guid_wwn ? &sport->port_guid_tpg : + &sport->port_gid_tpg; + res = core_tpg_register(wwn, tpg, SCSI_PROTOCOL_SRP); if (res) return ERR_PTR(res); - return &sport->port_tpg_1; + return tpg; } /** @@ -2988,11 +3026,10 @@ static struct se_portal_group *srpt_make_tpg(struct se_wwn *wwn, */ static void srpt_drop_tpg(struct se_portal_group *tpg) { - struct srpt_port *sport = container_of(tpg, - struct srpt_port, port_tpg_1); + struct srpt_port *sport = srpt_tpg_to_sport(tpg); sport->enabled = false; - core_tpg_deregister(&sport->port_tpg_1); + core_tpg_deregister(tpg); } /** @@ -3003,19 +3040,7 @@ static struct se_wwn *srpt_make_tport(struct target_fabric_configfs *tf, struct config_group *group, const char *name) { - struct srpt_port *sport; - int ret; - - sport = srpt_lookup_port(name); - pr_debug("make_tport(%s)\n", name); - ret = -EINVAL; - if (!sport) - goto err; - - return &sport->port_wwn; - -err: - return ERR_PTR(ret); + return srpt_lookup_wwn(name) ? : ERR_PTR(-EINVAL); } /** @@ -3024,9 +3049,6 @@ err: */ static void srpt_drop_tport(struct se_wwn *wwn) { - struct srpt_port *sport = container_of(wwn, struct srpt_port, port_wwn); - - pr_debug("drop_tport(%s\n", config_item_name(&sport->port_wwn.wwn_group.cg_item)); } static ssize_t srpt_wwn_version_show(struct config_item *item, char *buf) diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.h b/drivers/infiniband/ulp/srpt/ib_srpt.h index 581878782854..cc1183851af5 100644 --- a/drivers/infiniband/ulp/srpt/ib_srpt.h +++ b/drivers/infiniband/ulp/srpt/ib_srpt.h @@ -258,6 +258,7 @@ enum rdma_ch_state { * against concurrent modification by the cm_id spinlock. * @sess: Session information associated with this SRP channel. * @sess_name: Session name. + * @ini_guid: Initiator port GUID. * @release_work: Allows scheduling of srpt_release_channel(). * @release_done: Enables waiting for srpt_release_channel() completion. */ @@ -284,6 +285,7 @@ struct srpt_rdma_ch { struct list_head cmd_wait_list; struct se_session *sess; u8 sess_name[36]; + u8 ini_guid[24]; struct work_struct release_work; struct completion *release_done; }; @@ -306,28 +308,34 @@ struct srpt_port_attrib { * @mad_agent: per-port management datagram processing information. * @enabled: Whether or not this target port is enabled. * @port_guid: ASCII representation of Port GUID + * @port_gid: ASCII representation of Port GID * @port: one-based port number. * @sm_lid: cached value of the port's sm_lid. * @lid: cached value of the port's lid. * @gid: cached value of the port's gid. * @port_acl_lock spinlock for port_acl_list: * @work: work structure for refreshing the aforementioned cached values. - * @port_tpg_1 Target portal group = 1 data. - * @port_wwn: Target core WWN data. + * @port_guid_tpg: TPG associated with target port GUID. + * @port_guid_wwn: WWN associated with target port GUID. + * @port_gid_tpg: TPG associated with target port GID. + * @port_gid_wwn: WWN associated with target port GID. * @port_acl_list: Head of the list with all node ACLs for this port. */ struct srpt_port { struct srpt_device *sdev; struct ib_mad_agent *mad_agent; bool enabled; - u8 port_guid[64]; + u8 port_guid[24]; + u8 port_gid[64]; u8 port; u16 sm_lid; u16 lid; union ib_gid gid; struct work_struct work; - struct se_portal_group port_tpg_1; - struct se_wwn port_wwn; + struct se_portal_group port_guid_tpg; + struct se_wwn port_guid_wwn; + struct se_portal_group port_gid_tpg; + struct se_wwn port_gid_wwn; struct srpt_port_attrib port_attrib; }; |