diff options
Diffstat (limited to 'drivers/infiniband')
66 files changed, 1753 insertions, 580 deletions
diff --git a/drivers/infiniband/core/Makefile b/drivers/infiniband/core/Makefile index 6ebd9ad95010..e3cdafff8ece 100644 --- a/drivers/infiniband/core/Makefile +++ b/drivers/infiniband/core/Makefile @@ -10,7 +10,8 @@ obj-$(CONFIG_INFINIBAND_USER_ACCESS) += ib_uverbs.o ib_ucm.o \ ib_core-y := packer.o ud_header.o verbs.o cq.o rw.o sysfs.o \ device.o fmr_pool.o cache.o netlink.o \ roce_gid_mgmt.o mr_pool.o addr.o sa_query.o \ - multicast.o mad.o smi.o agent.o mad_rmpp.o + multicast.o mad.o smi.o agent.o mad_rmpp.o \ + security.o ib_core-$(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 diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c index ece6926fa2e6..01236cef7bfb 100644 --- a/drivers/infiniband/core/addr.c +++ b/drivers/infiniband/core/addr.c @@ -179,8 +179,7 @@ static int ib_nl_ip_send_msg(struct rdma_dev_addr *dev_addr, } /* Construct the family header first */ - header = (struct rdma_ls_ip_resolve_header *) - skb_put(skb, NLMSG_ALIGN(sizeof(*header))); + header = skb_put(skb, NLMSG_ALIGN(sizeof(*header))); header->ifindex = dev_addr->bound_dev_if; nla_put(skb, attrtype, size, daddr); @@ -269,6 +268,7 @@ int rdma_translate_ip(const struct sockaddr *addr, return ret; ret = rdma_copy_addr(dev_addr, dev, NULL); + dev_addr->bound_dev_if = dev->ifindex; if (vlan_id) *vlan_id = rdma_vlan_dev_vlan_id(dev); dev_put(dev); @@ -281,6 +281,7 @@ int rdma_translate_ip(const struct sockaddr *addr, &((const struct sockaddr_in6 *)addr)->sin6_addr, dev, 1)) { ret = rdma_copy_addr(dev_addr, dev, NULL); + dev_addr->bound_dev_if = dev->ifindex; if (vlan_id) *vlan_id = rdma_vlan_dev_vlan_id(dev); break; @@ -406,10 +407,10 @@ static int addr4_resolve(struct sockaddr_in *src_in, fl4.saddr = src_ip; fl4.flowi4_oif = addr->bound_dev_if; rt = ip_route_output_key(addr->net, &fl4); - if (IS_ERR(rt)) { - ret = PTR_ERR(rt); - goto out; - } + ret = PTR_ERR_OR_ZERO(rt); + if (ret) + return ret; + src_in->sin_family = AF_INET; src_in->sin_addr.s_addr = fl4.saddr; @@ -424,8 +425,6 @@ static int addr4_resolve(struct sockaddr_in *src_in, *prt = rt; return 0; -out: - return ret; } #if IS_ENABLED(CONFIG_IPV6) @@ -510,6 +509,11 @@ static int addr_resolve(struct sockaddr *src_in, struct dst_entry *dst; int ret; + if (!addr->net) { + pr_warn_ratelimited("%s: missing namespace\n", __func__); + return -EINVAL; + } + if (src_in->sa_family == AF_INET) { struct rtable *rt = NULL; const struct sockaddr_in *dst_in4 = @@ -523,8 +527,12 @@ static int addr_resolve(struct sockaddr *src_in, if (resolve_neigh) ret = addr_resolve_neigh(&rt->dst, dst_in, addr, seq); - ndev = rt->dst.dev; - dev_hold(ndev); + if (addr->bound_dev_if) { + ndev = dev_get_by_index(addr->net, addr->bound_dev_if); + } else { + ndev = rt->dst.dev; + dev_hold(ndev); + } ip_rt_put(rt); } else { @@ -540,14 +548,27 @@ static int addr_resolve(struct sockaddr *src_in, if (resolve_neigh) ret = addr_resolve_neigh(dst, dst_in, addr, seq); - ndev = dst->dev; - dev_hold(ndev); + if (addr->bound_dev_if) { + ndev = dev_get_by_index(addr->net, addr->bound_dev_if); + } else { + ndev = dst->dev; + dev_hold(ndev); + } dst_release(dst); } - addr->bound_dev_if = ndev->ifindex; - addr->net = dev_net(ndev); + if (ndev->flags & IFF_LOOPBACK) { + ret = rdma_translate_ip(dst_in, addr, NULL); + /* + * Put the loopback device and get the translated + * device instead. + */ + dev_put(ndev); + ndev = dev_get_by_index(addr->net, addr->bound_dev_if); + } else { + addr->bound_dev_if = ndev->ifindex; + } dev_put(ndev); return ret; diff --git a/drivers/infiniband/core/cache.c b/drivers/infiniband/core/cache.c index b1371eb9f46c..efc94304dee3 100644 --- a/drivers/infiniband/core/cache.c +++ b/drivers/infiniband/core/cache.c @@ -53,6 +53,7 @@ struct ib_update_work { struct work_struct work; struct ib_device *device; u8 port_num; + bool enforce_security; }; union ib_gid zgid; @@ -911,6 +912,26 @@ int ib_get_cached_pkey(struct ib_device *device, } EXPORT_SYMBOL(ib_get_cached_pkey); +int ib_get_cached_subnet_prefix(struct ib_device *device, + u8 port_num, + u64 *sn_pfx) +{ + unsigned long flags; + int p; + + if (port_num < rdma_start_port(device) || + port_num > rdma_end_port(device)) + return -EINVAL; + + p = port_num - rdma_start_port(device); + read_lock_irqsave(&device->cache.lock, flags); + *sn_pfx = device->cache.ports[p].subnet_prefix; + read_unlock_irqrestore(&device->cache.lock, flags); + + return 0; +} +EXPORT_SYMBOL(ib_get_cached_subnet_prefix); + int ib_find_cached_pkey(struct ib_device *device, u8 port_num, u16 pkey, @@ -1022,7 +1043,8 @@ int ib_get_cached_port_state(struct ib_device *device, EXPORT_SYMBOL(ib_get_cached_port_state); static void ib_cache_update(struct ib_device *device, - u8 port) + u8 port, + bool enforce_security) { struct ib_port_attr *tprops = NULL; struct ib_pkey_cache *pkey_cache = NULL, *old_pkey_cache; @@ -1108,8 +1130,15 @@ static void ib_cache_update(struct ib_device *device, device->cache.ports[port - rdma_start_port(device)].port_state = tprops->state; + device->cache.ports[port - rdma_start_port(device)].subnet_prefix = + tprops->subnet_prefix; write_unlock_irq(&device->cache.lock); + if (enforce_security) + ib_security_cache_change(device, + port, + tprops->subnet_prefix); + kfree(gid_cache); kfree(old_pkey_cache); kfree(tprops); @@ -1126,7 +1155,9 @@ static void ib_cache_task(struct work_struct *_work) struct ib_update_work *work = container_of(_work, struct ib_update_work, work); - ib_cache_update(work->device, work->port_num); + ib_cache_update(work->device, + work->port_num, + work->enforce_security); kfree(work); } @@ -1147,6 +1178,12 @@ static void ib_cache_event(struct ib_event_handler *handler, INIT_WORK(&work->work, ib_cache_task); work->device = event->device; work->port_num = event->element.port_num; + if (event->event == IB_EVENT_PKEY_CHANGE || + event->event == IB_EVENT_GID_CHANGE) + work->enforce_security = true; + else + work->enforce_security = false; + queue_work(ib_wq, &work->work); } } @@ -1172,7 +1209,7 @@ int ib_cache_setup_one(struct ib_device *device) goto out; for (p = 0; p <= rdma_end_port(device) - rdma_start_port(device); ++p) - ib_cache_update(device, p + rdma_start_port(device)); + ib_cache_update(device, p + rdma_start_port(device), true); INIT_IB_EVENT_HANDLER(&device->cache.event_handler, device, ib_cache_event); diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index 31bb82d8ecd7..11aff923b633 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -623,22 +623,11 @@ static inline int cma_validate_port(struct ib_device *device, u8 port, if ((dev_type != ARPHRD_INFINIBAND) && rdma_protocol_ib(device, port)) return ret; - if (dev_type == ARPHRD_ETHER && rdma_protocol_roce(device, port)) { + if (dev_type == ARPHRD_ETHER && rdma_protocol_roce(device, port)) ndev = dev_get_by_index(&init_net, bound_if_index); - if (ndev && ndev->flags & IFF_LOOPBACK) { - pr_info("detected loopback device\n"); - dev_put(ndev); - - if (!device->get_netdev) - return -EOPNOTSUPP; - - ndev = device->get_netdev(device, port); - if (!ndev) - return -ENODEV; - } - } else { + else gid_type = IB_GID_TYPE_IB; - } + ret = ib_find_cached_gid_by_port(device, gid, gid_type, port, ndev, NULL); @@ -2569,21 +2558,6 @@ static int cma_resolve_iboe_route(struct rdma_id_private *id_priv) goto err2; } - if (ndev->flags & IFF_LOOPBACK) { - dev_put(ndev); - if (!id_priv->id.device->get_netdev) { - ret = -EOPNOTSUPP; - goto err2; - } - - ndev = id_priv->id.device->get_netdev(id_priv->id.device, - id_priv->id.port_num); - if (!ndev) { - ret = -ENODEV; - goto err2; - } - } - supported_gids = roce_gid_type_mask_support(id_priv->id.device, id_priv->id.port_num); gid_type = cma_route_gid_type(addr->dev_addr.network, diff --git a/drivers/infiniband/core/core_priv.h b/drivers/infiniband/core/core_priv.h index d92ab4eaa8f3..11ae67514e13 100644 --- a/drivers/infiniband/core/core_priv.h +++ b/drivers/infiniband/core/core_priv.h @@ -38,6 +38,16 @@ #include <linux/cgroup_rdma.h> #include <rdma/ib_verbs.h> +#include <rdma/ib_mad.h> +#include "mad_priv.h" + +struct pkey_index_qp_list { + struct list_head pkey_index_list; + u16 pkey_index; + /* Lock to hold while iterating the qp_list. */ + spinlock_t qp_list_lock; + struct list_head qp_list; +}; #if IS_ENABLED(CONFIG_INFINIBAND_ADDR_TRANS_CONFIGFS) int cma_configfs_init(void); @@ -186,4 +196,109 @@ int ib_nl_handle_set_timeout(struct sk_buff *skb, int ib_nl_handle_ip_res_resp(struct sk_buff *skb, struct netlink_callback *cb); +int ib_get_cached_subnet_prefix(struct ib_device *device, + u8 port_num, + u64 *sn_pfx); + +#ifdef CONFIG_SECURITY_INFINIBAND +int ib_security_pkey_access(struct ib_device *dev, + u8 port_num, + u16 pkey_index, + void *sec); + +void ib_security_destroy_port_pkey_list(struct ib_device *device); + +void ib_security_cache_change(struct ib_device *device, + u8 port_num, + u64 subnet_prefix); + +int ib_security_modify_qp(struct ib_qp *qp, + struct ib_qp_attr *qp_attr, + int qp_attr_mask, + struct ib_udata *udata); + +int ib_create_qp_security(struct ib_qp *qp, struct ib_device *dev); +void ib_destroy_qp_security_begin(struct ib_qp_security *sec); +void ib_destroy_qp_security_abort(struct ib_qp_security *sec); +void ib_destroy_qp_security_end(struct ib_qp_security *sec); +int ib_open_shared_qp_security(struct ib_qp *qp, struct ib_device *dev); +void ib_close_shared_qp_security(struct ib_qp_security *sec); +int ib_mad_agent_security_setup(struct ib_mad_agent *agent, + enum ib_qp_type qp_type); +void ib_mad_agent_security_cleanup(struct ib_mad_agent *agent); +int ib_mad_enforce_security(struct ib_mad_agent_private *map, u16 pkey_index); +#else +static inline int ib_security_pkey_access(struct ib_device *dev, + u8 port_num, + u16 pkey_index, + void *sec) +{ + return 0; +} + +static inline void ib_security_destroy_port_pkey_list(struct ib_device *device) +{ +} + +static inline void ib_security_cache_change(struct ib_device *device, + u8 port_num, + u64 subnet_prefix) +{ +} + +static inline int ib_security_modify_qp(struct ib_qp *qp, + struct ib_qp_attr *qp_attr, + int qp_attr_mask, + struct ib_udata *udata) +{ + return qp->device->modify_qp(qp->real_qp, + qp_attr, + qp_attr_mask, + udata); +} + +static inline int ib_create_qp_security(struct ib_qp *qp, + struct ib_device *dev) +{ + return 0; +} + +static inline void ib_destroy_qp_security_begin(struct ib_qp_security *sec) +{ +} + +static inline void ib_destroy_qp_security_abort(struct ib_qp_security *sec) +{ +} + +static inline void ib_destroy_qp_security_end(struct ib_qp_security *sec) +{ +} + +static inline int ib_open_shared_qp_security(struct ib_qp *qp, + struct ib_device *dev) +{ + return 0; +} + +static inline void ib_close_shared_qp_security(struct ib_qp_security *sec) +{ +} + +static inline int ib_mad_agent_security_setup(struct ib_mad_agent *agent, + enum ib_qp_type qp_type) +{ + return 0; +} + +static inline void ib_mad_agent_security_cleanup(struct ib_mad_agent *agent) +{ +} + +static inline int ib_mad_enforce_security(struct ib_mad_agent_private *map, + u16 pkey_index) +{ + return 0; +} +#endif #endif /* _CORE_PRIV_H */ diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c index 81d447da0048..a5dfab6adf49 100644 --- a/drivers/infiniband/core/device.c +++ b/drivers/infiniband/core/device.c @@ -39,6 +39,8 @@ #include <linux/init.h> #include <linux/mutex.h> #include <linux/netdevice.h> +#include <linux/security.h> +#include <linux/notifier.h> #include <rdma/rdma_netlink.h> #include <rdma/ib_addr.h> #include <rdma/ib_cache.h> @@ -82,6 +84,14 @@ static LIST_HEAD(client_list); static DEFINE_MUTEX(device_mutex); static DECLARE_RWSEM(lists_rwsem); +static int ib_security_change(struct notifier_block *nb, unsigned long event, + void *lsm_data); +static void ib_policy_change_task(struct work_struct *work); +static DECLARE_WORK(ib_policy_change_work, ib_policy_change_task); + +static struct notifier_block ibdev_lsm_nb = { + .notifier_call = ib_security_change, +}; static int ib_device_check_mandatory(struct ib_device *device) { @@ -325,6 +335,65 @@ void ib_get_device_fw_str(struct ib_device *dev, char *str, size_t str_len) } EXPORT_SYMBOL(ib_get_device_fw_str); +static int setup_port_pkey_list(struct ib_device *device) +{ + int i; + + /** + * device->port_pkey_list is indexed directly by the port number, + * Therefore it is declared as a 1 based array with potential empty + * slots at the beginning. + */ + device->port_pkey_list = kcalloc(rdma_end_port(device) + 1, + sizeof(*device->port_pkey_list), + GFP_KERNEL); + + if (!device->port_pkey_list) + return -ENOMEM; + + for (i = 0; i < (rdma_end_port(device) + 1); i++) { + spin_lock_init(&device->port_pkey_list[i].list_lock); + INIT_LIST_HEAD(&device->port_pkey_list[i].pkey_list); + } + + return 0; +} + +static void ib_policy_change_task(struct work_struct *work) +{ + struct ib_device *dev; + + down_read(&lists_rwsem); + list_for_each_entry(dev, &device_list, core_list) { + int i; + + for (i = rdma_start_port(dev); i <= rdma_end_port(dev); i++) { + u64 sp; + int ret = ib_get_cached_subnet_prefix(dev, + i, + &sp); + + WARN_ONCE(ret, + "ib_get_cached_subnet_prefix err: %d, this should never happen here\n", + ret); + if (!ret) + ib_security_cache_change(dev, i, sp); + } + } + up_read(&lists_rwsem); +} + +static int ib_security_change(struct notifier_block *nb, unsigned long event, + void *lsm_data) +{ + if (event != LSM_POLICY_CHANGE) + return NOTIFY_DONE; + + schedule_work(&ib_policy_change_work); + + return NOTIFY_OK; +} + /** * ib_register_device - Register an IB device with IB core * @device:Device to register @@ -385,6 +454,12 @@ int ib_register_device(struct ib_device *device, goto out; } + ret = setup_port_pkey_list(device); + if (ret) { + pr_warn("Couldn't create per port_pkey_list\n"); + goto out; + } + ret = ib_cache_setup_one(device); if (ret) { pr_warn("Couldn't set up InfiniBand P_Key/GID cache\n"); @@ -468,6 +543,9 @@ void ib_unregister_device(struct ib_device *device) ib_device_unregister_sysfs(device); ib_cache_cleanup_one(device); + ib_security_destroy_port_pkey_list(device); + kfree(device->port_pkey_list); + down_write(&lists_rwsem); spin_lock_irqsave(&device->client_data_lock, flags); list_for_each_entry_safe(context, tmp, &device->client_data_list, list) @@ -1082,10 +1160,18 @@ static int __init ib_core_init(void) goto err_sa; } + ret = register_lsm_notifier(&ibdev_lsm_nb); + if (ret) { + pr_warn("Couldn't register LSM notifier. ret %d\n", ret); + goto err_ibnl_clients; + } + ib_cache_setup(); return 0; +err_ibnl_clients: + ib_remove_ibnl_clients(); err_sa: ib_sa_cleanup(); err_mad: @@ -1105,6 +1191,7 @@ err: static void __exit ib_core_cleanup(void) { + unregister_lsm_notifier(&ibdev_lsm_nb); ib_cache_cleanup(); ib_remove_ibnl_clients(); ib_sa_cleanup(); diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c index 192ee3dafb80..f8f53bb90837 100644 --- a/drivers/infiniband/core/mad.c +++ b/drivers/infiniband/core/mad.c @@ -40,9 +40,11 @@ #include <linux/dma-mapping.h> #include <linux/slab.h> #include <linux/module.h> +#include <linux/security.h> #include <rdma/ib_cache.h> #include "mad_priv.h" +#include "core_priv.h" #include "mad_rmpp.h" #include "smi.h" #include "opa_smi.h" @@ -369,6 +371,12 @@ struct ib_mad_agent *ib_register_mad_agent(struct ib_device *device, atomic_set(&mad_agent_priv->refcount, 1); init_completion(&mad_agent_priv->comp); + ret2 = ib_mad_agent_security_setup(&mad_agent_priv->agent, qp_type); + if (ret2) { + ret = ERR_PTR(ret2); + goto error4; + } + spin_lock_irqsave(&port_priv->reg_lock, flags); mad_agent_priv->agent.hi_tid = ++ib_mad_client_id; @@ -386,7 +394,7 @@ struct ib_mad_agent *ib_register_mad_agent(struct ib_device *device, if (method) { if (method_in_use(&method, mad_reg_req)) - goto error4; + goto error5; } } ret2 = add_nonoui_reg_req(mad_reg_req, mad_agent_priv, @@ -402,14 +410,14 @@ struct ib_mad_agent *ib_register_mad_agent(struct ib_device *device, if (is_vendor_method_in_use( vendor_class, mad_reg_req)) - goto error4; + goto error5; } } ret2 = add_oui_reg_req(mad_reg_req, mad_agent_priv); } if (ret2) { ret = ERR_PTR(ret2); - goto error4; + goto error5; } } @@ -418,9 +426,10 @@ struct ib_mad_agent *ib_register_mad_agent(struct ib_device *device, spin_unlock_irqrestore(&port_priv->reg_lock, flags); return &mad_agent_priv->agent; - -error4: +error5: spin_unlock_irqrestore(&port_priv->reg_lock, flags); + ib_mad_agent_security_cleanup(&mad_agent_priv->agent); +error4: kfree(reg_req); error3: kfree(mad_agent_priv); @@ -491,6 +500,7 @@ struct ib_mad_agent *ib_register_mad_snoop(struct ib_device *device, struct ib_mad_agent *ret; struct ib_mad_snoop_private *mad_snoop_priv; int qpn; + int err; /* Validate parameters */ if ((is_snooping_sends(mad_snoop_flags) && !snoop_handler) || @@ -525,17 +535,25 @@ struct ib_mad_agent *ib_register_mad_snoop(struct ib_device *device, mad_snoop_priv->agent.port_num = port_num; mad_snoop_priv->mad_snoop_flags = mad_snoop_flags; init_completion(&mad_snoop_priv->comp); + + err = ib_mad_agent_security_setup(&mad_snoop_priv->agent, qp_type); + if (err) { + ret = ERR_PTR(err); + goto error2; + } + mad_snoop_priv->snoop_index = register_snoop_agent( &port_priv->qp_info[qpn], mad_snoop_priv); if (mad_snoop_priv->snoop_index < 0) { ret = ERR_PTR(mad_snoop_priv->snoop_index); - goto error2; + goto error3; } atomic_set(&mad_snoop_priv->refcount, 1); return &mad_snoop_priv->agent; - +error3: + ib_mad_agent_security_cleanup(&mad_snoop_priv->agent); error2: kfree(mad_snoop_priv); error1: @@ -581,6 +599,8 @@ static void unregister_mad_agent(struct ib_mad_agent_private *mad_agent_priv) deref_mad_agent(mad_agent_priv); wait_for_completion(&mad_agent_priv->comp); + ib_mad_agent_security_cleanup(&mad_agent_priv->agent); + kfree(mad_agent_priv->reg_req); kfree(mad_agent_priv); } @@ -599,6 +619,8 @@ static void unregister_mad_snoop(struct ib_mad_snoop_private *mad_snoop_priv) deref_snoop_agent(mad_snoop_priv); wait_for_completion(&mad_snoop_priv->comp); + ib_mad_agent_security_cleanup(&mad_snoop_priv->agent); + kfree(mad_snoop_priv); } @@ -1215,12 +1237,16 @@ int ib_post_send_mad(struct ib_mad_send_buf *send_buf, /* Walk list of send WRs and post each on send list */ for (; send_buf; send_buf = next_send_buf) { - mad_send_wr = container_of(send_buf, struct ib_mad_send_wr_private, send_buf); mad_agent_priv = mad_send_wr->mad_agent_priv; + ret = ib_mad_enforce_security(mad_agent_priv, + mad_send_wr->send_wr.pkey_index); + if (ret) + goto error; + if (!send_buf->mad_agent->send_handler || (send_buf->timeout_ms && !send_buf->mad_agent->recv_handler)) { @@ -1946,6 +1972,14 @@ static void ib_mad_complete_recv(struct ib_mad_agent_private *mad_agent_priv, struct ib_mad_send_wr_private *mad_send_wr; struct ib_mad_send_wc mad_send_wc; unsigned long flags; + int ret; + + ret = ib_mad_enforce_security(mad_agent_priv, + mad_recv_wc->wc->pkey_index); + if (ret) { + ib_free_recv_mad(mad_recv_wc); + deref_mad_agent(mad_agent_priv); + } INIT_LIST_HEAD(&mad_recv_wc->rmpp_list); list_add(&mad_recv_wc->recv_buf.list, &mad_recv_wc->rmpp_list); @@ -2003,6 +2037,8 @@ static void ib_mad_complete_recv(struct ib_mad_agent_private *mad_agent_priv, mad_recv_wc); deref_mad_agent(mad_agent_priv); } + + return; } static enum smi_action handle_ib_smi(const struct ib_mad_port_private *port_priv, diff --git a/drivers/infiniband/core/roce_gid_mgmt.c b/drivers/infiniband/core/roce_gid_mgmt.c index db958d3207ef..94a9eefb3cfc 100644 --- a/drivers/infiniband/core/roce_gid_mgmt.c +++ b/drivers/infiniband/core/roce_gid_mgmt.c @@ -42,6 +42,8 @@ #include <rdma/ib_cache.h> #include <rdma/ib_addr.h> +static struct workqueue_struct *gid_cache_wq; + enum gid_op_type { GID_DEL = 0, GID_ADD @@ -560,7 +562,7 @@ static int netdevice_queue_work(struct netdev_event_work_cmd *cmds, } INIT_WORK(&ndev_work->work, netdevice_event_work_handler); - queue_work(ib_wq, &ndev_work->work); + queue_work(gid_cache_wq, &ndev_work->work); return NOTIFY_DONE; } @@ -693,7 +695,7 @@ static int addr_event(struct notifier_block *this, unsigned long event, dev_hold(ndev); work->gid_attr.ndev = ndev; - queue_work(ib_wq, &work->work); + queue_work(gid_cache_wq, &work->work); return NOTIFY_DONE; } @@ -740,6 +742,10 @@ static struct notifier_block nb_inet6addr = { int __init roce_gid_mgmt_init(void) { + gid_cache_wq = alloc_ordered_workqueue("gid-cache-wq", 0); + if (!gid_cache_wq) + return -ENOMEM; + register_inetaddr_notifier(&nb_inetaddr); if (IS_ENABLED(CONFIG_IPV6)) register_inet6addr_notifier(&nb_inet6addr); @@ -764,4 +770,5 @@ void __exit roce_gid_mgmt_cleanup(void) * ib-core is removed, all physical devices have been removed, * so no issue with remaining hardware contexts. */ + destroy_workqueue(gid_cache_wq); } diff --git a/drivers/infiniband/core/sa_query.c b/drivers/infiniband/core/sa_query.c index fb7aec4047c8..70fa4cabe48e 100644 --- a/drivers/infiniband/core/sa_query.c +++ b/drivers/infiniband/core/sa_query.c @@ -759,8 +759,7 @@ static void ib_nl_set_path_rec_attrs(struct sk_buff *skb, query->mad_buf->context[1] = NULL; /* Construct the family header first */ - header = (struct rdma_ls_resolve_header *) - skb_put(skb, NLMSG_ALIGN(sizeof(*header))); + header = skb_put(skb, NLMSG_ALIGN(sizeof(*header))); memcpy(header->device_name, query->port->agent->device->name, LS_DEVICE_NAME_MAX); header->port_num = query->port->port_num; diff --git a/drivers/infiniband/core/security.c b/drivers/infiniband/core/security.c new file mode 100644 index 000000000000..70ad19c4c73e --- /dev/null +++ b/drivers/infiniband/core/security.c @@ -0,0 +1,709 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. 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. + */ + +#ifdef CONFIG_SECURITY_INFINIBAND + +#include <linux/security.h> +#include <linux/completion.h> +#include <linux/list.h> + +#include <rdma/ib_verbs.h> +#include <rdma/ib_cache.h> +#include "core_priv.h" +#include "mad_priv.h" + +static struct pkey_index_qp_list *get_pkey_idx_qp_list(struct ib_port_pkey *pp) +{ + struct pkey_index_qp_list *pkey = NULL; + struct pkey_index_qp_list *tmp_pkey; + struct ib_device *dev = pp->sec->dev; + + spin_lock(&dev->port_pkey_list[pp->port_num].list_lock); + list_for_each_entry(tmp_pkey, + &dev->port_pkey_list[pp->port_num].pkey_list, + pkey_index_list) { + if (tmp_pkey->pkey_index == pp->pkey_index) { + pkey = tmp_pkey; + break; + } + } + spin_unlock(&dev->port_pkey_list[pp->port_num].list_lock); + return pkey; +} + +static int get_pkey_and_subnet_prefix(struct ib_port_pkey *pp, + u16 *pkey, + u64 *subnet_prefix) +{ + struct ib_device *dev = pp->sec->dev; + int ret; + + ret = ib_get_cached_pkey(dev, pp->port_num, pp->pkey_index, pkey); + if (ret) + return ret; + + ret = ib_get_cached_subnet_prefix(dev, pp->port_num, subnet_prefix); + + return ret; +} + +static int enforce_qp_pkey_security(u16 pkey, + u64 subnet_prefix, + struct ib_qp_security *qp_sec) +{ + struct ib_qp_security *shared_qp_sec; + int ret; + + ret = security_ib_pkey_access(qp_sec->security, subnet_prefix, pkey); + if (ret) + return ret; + + if (qp_sec->qp == qp_sec->qp->real_qp) { + list_for_each_entry(shared_qp_sec, + &qp_sec->shared_qp_list, + shared_qp_list) { + ret = security_ib_pkey_access(shared_qp_sec->security, + subnet_prefix, + pkey); + if (ret) + return ret; + } + } + return 0; +} + +/* The caller of this function must hold the QP security + * mutex of the QP of the security structure in *pps. + * + * It takes separate ports_pkeys and security structure + * because in some cases the pps will be for a new settings + * or the pps will be for the real QP and security structure + * will be for a shared QP. + */ +static int check_qp_port_pkey_settings(struct ib_ports_pkeys *pps, + struct ib_qp_security *sec) +{ + u64 subnet_prefix; + u16 pkey; + int ret = 0; + + if (!pps) + return 0; + + if (pps->main.state != IB_PORT_PKEY_NOT_VALID) { + ret = get_pkey_and_subnet_prefix(&pps->main, + &pkey, + &subnet_prefix); + if (ret) + return ret; + + ret = enforce_qp_pkey_security(pkey, + subnet_prefix, + sec); + if (ret) + return ret; + } + + if (pps->alt.state != IB_PORT_PKEY_NOT_VALID) { + ret = get_pkey_and_subnet_prefix(&pps->alt, + &pkey, + &subnet_prefix); + if (ret) + return ret; + + ret = enforce_qp_pkey_security(pkey, + subnet_prefix, + sec); + } + + return ret; +} + +/* The caller of this function must hold the QP security + * mutex. + */ +static void qp_to_error(struct ib_qp_security *sec) +{ + struct ib_qp_security *shared_qp_sec; + struct ib_qp_attr attr = { + .qp_state = IB_QPS_ERR + }; + struct ib_event event = { + .event = IB_EVENT_QP_FATAL + }; + + /* If the QP is in the process of being destroyed + * the qp pointer in the security structure is + * undefined. It cannot be modified now. + */ + if (sec->destroying) + return; + + ib_modify_qp(sec->qp, + &attr, + IB_QP_STATE); + + if (sec->qp->event_handler && sec->qp->qp_context) { + event.element.qp = sec->qp; + sec->qp->event_handler(&event, + sec->qp->qp_context); + } + + list_for_each_entry(shared_qp_sec, + &sec->shared_qp_list, + shared_qp_list) { + struct ib_qp *qp = shared_qp_sec->qp; + + if (qp->event_handler && qp->qp_context) { + event.element.qp = qp; + event.device = qp->device; + qp->event_handler(&event, + qp->qp_context); + } + } +} + +static inline void check_pkey_qps(struct pkey_index_qp_list *pkey, + struct ib_device *device, + u8 port_num, + u64 subnet_prefix) +{ + struct ib_port_pkey *pp, *tmp_pp; + bool comp; + LIST_HEAD(to_error_list); + u16 pkey_val; + + if (!ib_get_cached_pkey(device, + port_num, + pkey->pkey_index, + &pkey_val)) { + spin_lock(&pkey->qp_list_lock); + list_for_each_entry(pp, &pkey->qp_list, qp_list) { + if (atomic_read(&pp->sec->error_list_count)) + continue; + + if (enforce_qp_pkey_security(pkey_val, + subnet_prefix, + pp->sec)) { + atomic_inc(&pp->sec->error_list_count); + list_add(&pp->to_error_list, + &to_error_list); + } + } + spin_unlock(&pkey->qp_list_lock); + } + + list_for_each_entry_safe(pp, + tmp_pp, + &to_error_list, + to_error_list) { + mutex_lock(&pp->sec->mutex); + qp_to_error(pp->sec); + list_del(&pp->to_error_list); + atomic_dec(&pp->sec->error_list_count); + comp = pp->sec->destroying; + mutex_unlock(&pp->sec->mutex); + + if (comp) + complete(&pp->sec->error_complete); + } +} + +/* The caller of this function must hold the QP security + * mutex. + */ +static int port_pkey_list_insert(struct ib_port_pkey *pp) +{ + struct pkey_index_qp_list *tmp_pkey; + struct pkey_index_qp_list *pkey; + struct ib_device *dev; + u8 port_num = pp->port_num; + int ret = 0; + + if (pp->state != IB_PORT_PKEY_VALID) + return 0; + + dev = pp->sec->dev; + + pkey = get_pkey_idx_qp_list(pp); + + if (!pkey) { + bool found = false; + + pkey = kzalloc(sizeof(*pkey), GFP_KERNEL); + if (!pkey) + return -ENOMEM; + + spin_lock(&dev->port_pkey_list[port_num].list_lock); + /* Check for the PKey again. A racing process may + * have created it. + */ + list_for_each_entry(tmp_pkey, + &dev->port_pkey_list[port_num].pkey_list, + pkey_index_list) { + if (tmp_pkey->pkey_index == pp->pkey_index) { + kfree(pkey); + pkey = tmp_pkey; + found = true; + break; + } + } + + if (!found) { + pkey->pkey_index = pp->pkey_index; + spin_lock_init(&pkey->qp_list_lock); + INIT_LIST_HEAD(&pkey->qp_list); + list_add(&pkey->pkey_index_list, + &dev->port_pkey_list[port_num].pkey_list); + } + spin_unlock(&dev->port_pkey_list[port_num].list_lock); + } + + spin_lock(&pkey->qp_list_lock); + list_add(&pp->qp_list, &pkey->qp_list); + spin_unlock(&pkey->qp_list_lock); + + pp->state = IB_PORT_PKEY_LISTED; + + return ret; +} + +/* The caller of this function must hold the QP security + * mutex. + */ +static void port_pkey_list_remove(struct ib_port_pkey *pp) +{ + struct pkey_index_qp_list *pkey; + + if (pp->state != IB_PORT_PKEY_LISTED) + return; + + pkey = get_pkey_idx_qp_list(pp); + + spin_lock(&pkey->qp_list_lock); + list_del(&pp->qp_list); + spin_unlock(&pkey->qp_list_lock); + + /* The setting may still be valid, i.e. after + * a destroy has failed for example. + */ + pp->state = IB_PORT_PKEY_VALID; +} + +static void destroy_qp_security(struct ib_qp_security *sec) +{ + security_ib_free_security(sec->security); + kfree(sec->ports_pkeys); + kfree(sec); +} + +/* The caller of this function must hold the QP security + * mutex. + */ +static struct ib_ports_pkeys *get_new_pps(const struct ib_qp *qp, + const struct ib_qp_attr *qp_attr, + int qp_attr_mask) +{ + struct ib_ports_pkeys *new_pps; + struct ib_ports_pkeys *qp_pps = qp->qp_sec->ports_pkeys; + + new_pps = kzalloc(sizeof(*new_pps), GFP_KERNEL); + if (!new_pps) + return NULL; + + if (qp_attr_mask & (IB_QP_PKEY_INDEX | IB_QP_PORT)) { + if (!qp_pps) { + new_pps->main.port_num = qp_attr->port_num; + new_pps->main.pkey_index = qp_attr->pkey_index; + } else { + new_pps->main.port_num = (qp_attr_mask & IB_QP_PORT) ? + qp_attr->port_num : + qp_pps->main.port_num; + + new_pps->main.pkey_index = + (qp_attr_mask & IB_QP_PKEY_INDEX) ? + qp_attr->pkey_index : + qp_pps->main.pkey_index; + } + new_pps->main.state = IB_PORT_PKEY_VALID; + } else if (qp_pps) { + new_pps->main.port_num = qp_pps->main.port_num; + new_pps->main.pkey_index = qp_pps->main.pkey_index; + if (qp_pps->main.state != IB_PORT_PKEY_NOT_VALID) + new_pps->main.state = IB_PORT_PKEY_VALID; + } + + if (qp_attr_mask & IB_QP_ALT_PATH) { + new_pps->alt.port_num = qp_attr->alt_port_num; + new_pps->alt.pkey_index = qp_attr->alt_pkey_index; + new_pps->alt.state = IB_PORT_PKEY_VALID; + } else if (qp_pps) { + new_pps->alt.port_num = qp_pps->alt.port_num; + new_pps->alt.pkey_index = qp_pps->alt.pkey_index; + if (qp_pps->alt.state != IB_PORT_PKEY_NOT_VALID) + new_pps->alt.state = IB_PORT_PKEY_VALID; + } + + new_pps->main.sec = qp->qp_sec; + new_pps->alt.sec = qp->qp_sec; + return new_pps; +} + +int ib_open_shared_qp_security(struct ib_qp *qp, struct ib_device *dev) +{ + struct ib_qp *real_qp = qp->real_qp; + int ret; + + ret = ib_create_qp_security(qp, dev); + + if (ret) + return ret; + + mutex_lock(&real_qp->qp_sec->mutex); + ret = check_qp_port_pkey_settings(real_qp->qp_sec->ports_pkeys, + qp->qp_sec); + + if (ret) + goto ret; + + if (qp != real_qp) + list_add(&qp->qp_sec->shared_qp_list, + &real_qp->qp_sec->shared_qp_list); +ret: + mutex_unlock(&real_qp->qp_sec->mutex); + if (ret) + destroy_qp_security(qp->qp_sec); + + return ret; +} + +void ib_close_shared_qp_security(struct ib_qp_security *sec) +{ + struct ib_qp *real_qp = sec->qp->real_qp; + + mutex_lock(&real_qp->qp_sec->mutex); + list_del(&sec->shared_qp_list); + mutex_unlock(&real_qp->qp_sec->mutex); + + destroy_qp_security(sec); +} + +int ib_create_qp_security(struct ib_qp *qp, struct ib_device *dev) +{ + int ret; + + qp->qp_sec = kzalloc(sizeof(*qp->qp_sec), GFP_KERNEL); + if (!qp->qp_sec) + return -ENOMEM; + + qp->qp_sec->qp = qp; + qp->qp_sec->dev = dev; + mutex_init(&qp->qp_sec->mutex); + INIT_LIST_HEAD(&qp->qp_sec->shared_qp_list); + atomic_set(&qp->qp_sec->error_list_count, 0); + init_completion(&qp->qp_sec->error_complete); + ret = security_ib_alloc_security(&qp->qp_sec->security); + if (ret) + kfree(qp->qp_sec); + + return ret; +} +EXPORT_SYMBOL(ib_create_qp_security); + +void ib_destroy_qp_security_begin(struct ib_qp_security *sec) +{ + mutex_lock(&sec->mutex); + + /* Remove the QP from the lists so it won't get added to + * a to_error_list during the destroy process. + */ + if (sec->ports_pkeys) { + port_pkey_list_remove(&sec->ports_pkeys->main); + port_pkey_list_remove(&sec->ports_pkeys->alt); + } + + /* If the QP is already in one or more of those lists + * the destroying flag will ensure the to error flow + * doesn't operate on an undefined QP. + */ + sec->destroying = true; + + /* Record the error list count to know how many completions + * to wait for. + */ + sec->error_comps_pending = atomic_read(&sec->error_list_count); + + mutex_unlock(&sec->mutex); +} + +void ib_destroy_qp_security_abort(struct ib_qp_security *sec) +{ + int ret; + int i; + + /* If a concurrent cache update is in progress this + * QP security could be marked for an error state + * transition. Wait for this to complete. + */ + for (i = 0; i < sec->error_comps_pending; i++) + wait_for_completion(&sec->error_complete); + + mutex_lock(&sec->mutex); + sec->destroying = false; + + /* Restore the position in the lists and verify + * access is still allowed in case a cache update + * occurred while attempting to destroy. + * + * Because these setting were listed already + * and removed during ib_destroy_qp_security_begin + * we know the pkey_index_qp_list for the PKey + * already exists so port_pkey_list_insert won't fail. + */ + if (sec->ports_pkeys) { + port_pkey_list_insert(&sec->ports_pkeys->main); + port_pkey_list_insert(&sec->ports_pkeys->alt); + } + + ret = check_qp_port_pkey_settings(sec->ports_pkeys, sec); + if (ret) + qp_to_error(sec); + + mutex_unlock(&sec->mutex); +} + +void ib_destroy_qp_security_end(struct ib_qp_security *sec) +{ + int i; + + /* If a concurrent cache update is occurring we must + * wait until this QP security structure is processed + * in the QP to error flow before destroying it because + * the to_error_list is in use. + */ + for (i = 0; i < sec->error_comps_pending; i++) + wait_for_completion(&sec->error_complete); + + destroy_qp_security(sec); +} + +void ib_security_cache_change(struct ib_device *device, + u8 port_num, + u64 subnet_prefix) +{ + struct pkey_index_qp_list *pkey; + + list_for_each_entry(pkey, + &device->port_pkey_list[port_num].pkey_list, + pkey_index_list) { + check_pkey_qps(pkey, + device, + port_num, + subnet_prefix); + } +} + +void ib_security_destroy_port_pkey_list(struct ib_device *device) +{ + struct pkey_index_qp_list *pkey, *tmp_pkey; + int i; + + for (i = rdma_start_port(device); i <= rdma_end_port(device); i++) { + spin_lock(&device->port_pkey_list[i].list_lock); + list_for_each_entry_safe(pkey, + tmp_pkey, + &device->port_pkey_list[i].pkey_list, + pkey_index_list) { + list_del(&pkey->pkey_index_list); + kfree(pkey); + } + spin_unlock(&device->port_pkey_list[i].list_lock); + } +} + +int ib_security_modify_qp(struct ib_qp *qp, + struct ib_qp_attr *qp_attr, + int qp_attr_mask, + struct ib_udata *udata) +{ + int ret = 0; + struct ib_ports_pkeys *tmp_pps; + struct ib_ports_pkeys *new_pps; + bool special_qp = (qp->qp_type == IB_QPT_SMI || + qp->qp_type == IB_QPT_GSI || + qp->qp_type >= IB_QPT_RESERVED1); + bool pps_change = ((qp_attr_mask & (IB_QP_PKEY_INDEX | IB_QP_PORT)) || + (qp_attr_mask & IB_QP_ALT_PATH)); + + if (pps_change && !special_qp) { + mutex_lock(&qp->qp_sec->mutex); + new_pps = get_new_pps(qp, + qp_attr, + qp_attr_mask); + + /* Add this QP to the lists for the new port + * and pkey settings before checking for permission + * in case there is a concurrent cache update + * occurring. Walking the list for a cache change + * doesn't acquire the security mutex unless it's + * sending the QP to error. + */ + ret = port_pkey_list_insert(&new_pps->main); + + if (!ret) + ret = port_pkey_list_insert(&new_pps->alt); + + if (!ret) + ret = check_qp_port_pkey_settings(new_pps, + qp->qp_sec); + } + + if (!ret) + ret = qp->device->modify_qp(qp->real_qp, + qp_attr, + qp_attr_mask, + udata); + + if (pps_change && !special_qp) { + /* Clean up the lists and free the appropriate + * ports_pkeys structure. + */ + if (ret) { + tmp_pps = new_pps; + } else { + tmp_pps = qp->qp_sec->ports_pkeys; + qp->qp_sec->ports_pkeys = new_pps; + } + + if (tmp_pps) { + port_pkey_list_remove(&tmp_pps->main); + port_pkey_list_remove(&tmp_pps->alt); + } + kfree(tmp_pps); + mutex_unlock(&qp->qp_sec->mutex); + } + return ret; +} +EXPORT_SYMBOL(ib_security_modify_qp); + +int ib_security_pkey_access(struct ib_device *dev, + u8 port_num, + u16 pkey_index, + void *sec) +{ + u64 subnet_prefix; + u16 pkey; + int ret; + + ret = ib_get_cached_pkey(dev, port_num, pkey_index, &pkey); + if (ret) + return ret; + + ret = ib_get_cached_subnet_prefix(dev, port_num, &subnet_prefix); + + if (ret) + return ret; + + return security_ib_pkey_access(sec, subnet_prefix, pkey); +} +EXPORT_SYMBOL(ib_security_pkey_access); + +static int ib_mad_agent_security_change(struct notifier_block *nb, + unsigned long event, + void *data) +{ + struct ib_mad_agent *ag = container_of(nb, struct ib_mad_agent, lsm_nb); + + if (event != LSM_POLICY_CHANGE) + return NOTIFY_DONE; + + ag->smp_allowed = !security_ib_endport_manage_subnet(ag->security, + ag->device->name, + ag->port_num); + + return NOTIFY_OK; +} + +int ib_mad_agent_security_setup(struct ib_mad_agent *agent, + enum ib_qp_type qp_type) +{ + int ret; + + ret = security_ib_alloc_security(&agent->security); + if (ret) + return ret; + + if (qp_type != IB_QPT_SMI) + return 0; + + ret = security_ib_endport_manage_subnet(agent->security, + agent->device->name, + agent->port_num); + if (ret) + return ret; + + agent->lsm_nb.notifier_call = ib_mad_agent_security_change; + ret = register_lsm_notifier(&agent->lsm_nb); + if (ret) + return ret; + + agent->smp_allowed = true; + agent->lsm_nb_reg = true; + return 0; +} + +void ib_mad_agent_security_cleanup(struct ib_mad_agent *agent) +{ + security_ib_free_security(agent->security); + if (agent->lsm_nb_reg) + unregister_lsm_notifier(&agent->lsm_nb); +} + +int ib_mad_enforce_security(struct ib_mad_agent_private *map, u16 pkey_index) +{ + int ret; + + if (map->agent.qp->qp_type == IB_QPT_SMI && !map->agent.smp_allowed) + return -EACCES; + + ret = ib_security_pkey_access(map->agent.device, + map->agent.port_num, + pkey_index, + map->agent.security); + + if (ret) + return ret; + + return 0; +} + +#endif /* CONFIG_SECURITY_INFINIBAND */ diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 70b7fb156414..3f55d18a3791 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -1508,6 +1508,10 @@ static int create_qp(struct ib_uverbs_file *file, } if (cmd->qp_type != IB_QPT_XRC_TGT) { + ret = ib_create_qp_security(qp, device); + if (ret) + goto err_cb; + qp->real_qp = qp; qp->device = device; qp->pd = pd; @@ -1931,6 +1935,11 @@ static int modify_qp(struct ib_uverbs_file *file, goto out; } + if (!rdma_is_port_valid(qp->device, cmd->base.port_num)) { + ret = -EINVAL; + goto release_qp; + } + attr->qp_state = cmd->base.qp_state; attr->cur_qp_state = cmd->base.cur_qp_state; attr->path_mtu = cmd->base.path_mtu; @@ -1996,25 +2005,13 @@ static int modify_qp(struct ib_uverbs_file *file, rdma_ah_set_port_num(&attr->alt_ah_attr, cmd->base.alt_dest.port_num); - if (qp->real_qp == 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->base.attr_mask), - udata); - } else { - ret = ib_modify_qp(qp, attr, - modify_qp_mask(qp->qp_type, - cmd->base.attr_mask)); - } + ret = ib_modify_qp_with_udata(qp, attr, + modify_qp_mask(qp->qp_type, + cmd->base.attr_mask), + udata); release_qp: uobj_put_obj_read(qp); - out: kfree(attr); @@ -2541,6 +2538,9 @@ ssize_t ib_uverbs_create_ah(struct ib_uverbs_file *file, if (copy_from_user(&cmd, buf, sizeof cmd)) return -EFAULT; + if (!rdma_is_port_valid(ib_dev, cmd.attr.port_num)) + return -EINVAL; + INIT_UDATA(&udata, buf + sizeof(cmd), (unsigned long)cmd.response + sizeof(resp), in_len - sizeof(cmd), out_len - sizeof(resp)); diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c index 4792f5209ac2..fb98ed67d5bc 100644 --- a/drivers/infiniband/core/verbs.c +++ b/drivers/infiniband/core/verbs.c @@ -44,6 +44,7 @@ #include <linux/in.h> #include <linux/in6.h> #include <net/addrconf.h> +#include <linux/security.h> #include <rdma/ib_verbs.h> #include <rdma/ib_cache.h> @@ -451,6 +452,19 @@ int ib_get_gids_from_rdma_hdr(const union rdma_network_hdr *hdr, } EXPORT_SYMBOL(ib_get_gids_from_rdma_hdr); +/* + * This function creates ah from the incoming packet. + * Incoming packet has dgid of the receiver node on which this code is + * getting executed and, sgid contains the GID of the sender. + * + * When resolving mac address of destination, the arrived dgid is used + * as sgid and, sgid is used as dgid because sgid contains destinations + * GID whom to respond to. + * + * This is why when calling rdma_addr_find_l2_eth_by_grh() function, the + * position of arguments dgid and sgid do not match the order of the + * parameters. + */ int ib_init_ah_from_wc(struct ib_device *device, u8 port_num, const struct ib_wc *wc, const struct ib_grh *grh, struct rdma_ah_attr *ah_attr) @@ -506,11 +520,6 @@ int ib_init_ah_from_wc(struct ib_device *device, u8 port_num, } resolved_dev = dev_get_by_index(&init_net, if_index); - if (resolved_dev->flags & IFF_LOOPBACK) { - dev_put(resolved_dev); - resolved_dev = idev; - dev_hold(resolved_dev); - } rcu_read_lock(); if (resolved_dev != idev && !rdma_is_upper_dev_rcu(idev, resolved_dev)) @@ -713,12 +722,20 @@ static struct ib_qp *__ib_open_qp(struct ib_qp *real_qp, { struct ib_qp *qp; unsigned long flags; + int err; qp = kzalloc(sizeof *qp, GFP_KERNEL); if (!qp) return ERR_PTR(-ENOMEM); qp->real_qp = real_qp; + err = ib_open_shared_qp_security(qp, real_qp->device); + if (err) { + kfree(qp); + return ERR_PTR(err); + } + + qp->real_qp = real_qp; atomic_inc(&real_qp->usecnt); qp->device = real_qp->device; qp->event_handler = event_handler; @@ -804,6 +821,12 @@ struct ib_qp *ib_create_qp(struct ib_pd *pd, if (IS_ERR(qp)) return qp; + ret = ib_create_qp_security(qp, device); + if (ret) { + ib_destroy_qp(qp); + return ERR_PTR(ret); + } + qp->device = device; qp->real_qp = qp; qp->uobject = NULL; @@ -872,6 +895,7 @@ static const struct { } qp_state_table[IB_QPS_ERR + 1][IB_QPS_ERR + 1] = { [IB_QPS_RESET] = { [IB_QPS_RESET] = { .valid = 1 }, + [IB_QPS_ERR] = { .valid = 1 }, [IB_QPS_INIT] = { .valid = 1, .req_param = { @@ -1253,20 +1277,36 @@ out: } EXPORT_SYMBOL(ib_resolve_eth_dmac); -int ib_modify_qp(struct ib_qp *qp, - struct ib_qp_attr *qp_attr, - int qp_attr_mask) +/** + * ib_modify_qp_with_udata - Modifies the attributes for the specified QP. + * @qp: The QP to modify. + * @attr: On input, specifies the QP attributes to modify. On output, + * the current values of selected QP attributes are returned. + * @attr_mask: A bit-mask used to specify which attributes of the QP + * are being modified. + * @udata: pointer to user's input output buffer information + * are being modified. + * It returns 0 on success and returns appropriate error code on error. + */ +int ib_modify_qp_with_udata(struct ib_qp *qp, struct ib_qp_attr *attr, + int attr_mask, struct ib_udata *udata) { + int ret; - if (qp_attr_mask & IB_QP_AV) { - int ret; - - ret = ib_resolve_eth_dmac(qp->device, &qp_attr->ah_attr); + if (attr_mask & IB_QP_AV) { + ret = ib_resolve_eth_dmac(qp->device, &attr->ah_attr); if (ret) return ret; } + return ib_security_modify_qp(qp, attr, attr_mask, udata); +} +EXPORT_SYMBOL(ib_modify_qp_with_udata); - return qp->device->modify_qp(qp->real_qp, qp_attr, qp_attr_mask, NULL); +int ib_modify_qp(struct ib_qp *qp, + struct ib_qp_attr *qp_attr, + int qp_attr_mask) +{ + return ib_modify_qp_with_udata(qp, qp_attr, qp_attr_mask, NULL); } EXPORT_SYMBOL(ib_modify_qp); @@ -1295,6 +1335,7 @@ int ib_close_qp(struct ib_qp *qp) spin_unlock_irqrestore(&real_qp->device->event_handler_lock, flags); atomic_dec(&real_qp->usecnt); + ib_close_shared_qp_security(qp->qp_sec); kfree(qp); return 0; @@ -1335,6 +1376,7 @@ int ib_destroy_qp(struct ib_qp *qp) struct ib_cq *scq, *rcq; struct ib_srq *srq; struct ib_rwq_ind_table *ind_tbl; + struct ib_qp_security *sec; int ret; WARN_ON_ONCE(qp->mrs_used > 0); @@ -1350,6 +1392,9 @@ int ib_destroy_qp(struct ib_qp *qp) rcq = qp->recv_cq; srq = qp->srq; ind_tbl = qp->rwq_ind_tbl; + sec = qp->qp_sec; + if (sec) + ib_destroy_qp_security_begin(sec); if (!qp->uobject) rdma_rw_cleanup_mrs(qp); @@ -1366,6 +1411,11 @@ int ib_destroy_qp(struct ib_qp *qp) atomic_dec(&srq->usecnt); if (ind_tbl) atomic_dec(&ind_tbl->usecnt); + if (sec) + ib_destroy_qp_security_end(sec); + } else { + if (sec) + ib_destroy_qp_security_abort(sec); } return ret; diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.c b/drivers/infiniband/hw/cxgb3/cxio_hal.c index 558d6a03375d..3eff6541bd6f 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_hal.c +++ b/drivers/infiniband/hw/cxgb3/cxio_hal.c @@ -142,8 +142,7 @@ static int cxio_hal_clear_qp_ctx(struct cxio_rdev *rdev_p, u32 qpid) pr_debug("%s alloc_skb failed\n", __func__); return -ENOMEM; } - wqe = (struct t3_modify_qp_wr *) skb_put(skb, sizeof(*wqe)); - memset(wqe, 0, sizeof(*wqe)); + wqe = skb_put_zero(skb, sizeof(*wqe)); build_fw_riwrh((struct fw_riwrh *) wqe, T3_WR_QP_MOD, T3_COMPLETION_FLAG | T3_NOTIFY_FLAG, 0, qpid, 7, T3_SOPEOP); @@ -561,8 +560,7 @@ static int cxio_hal_init_ctrl_qp(struct cxio_rdev *rdev_p) ctx1 |= ((u64) (V_EC_BASE_HI((u32) base_addr & 0xf) | V_EC_RESPQ(0) | V_EC_TYPE(0) | V_EC_GEN(1) | V_EC_UP_TOKEN(T3_CTL_QP_TID) | F_EC_VALID)) << 32; - wqe = (struct t3_modify_qp_wr *) skb_put(skb, sizeof(*wqe)); - memset(wqe, 0, sizeof(*wqe)); + wqe = skb_put_zero(skb, sizeof(*wqe)); build_fw_riwrh((struct fw_riwrh *) wqe, T3_WR_QP_MOD, 0, 0, T3_CTL_QP_TID, 7, T3_SOPEOP); wqe->flags = cpu_to_be32(MODQP_WRITE_EC); @@ -837,7 +835,7 @@ int cxio_rdma_init(struct cxio_rdev *rdev_p, struct t3_rdma_init_attr *attr) if (!skb) return -ENOMEM; pr_debug("%s rdev_p %p\n", __func__, rdev_p); - wqe = (struct t3_rdma_init_wr *) __skb_put(skb, sizeof(*wqe)); + wqe = __skb_put(skb, sizeof(*wqe)); wqe->wrh.op_seop_flags = cpu_to_be32(V_FW_RIWR_OP(T3_WR_INIT)); wqe->wrh.gen_tid_len = cpu_to_be32(V_FW_RIWR_TID(attr->tid) | V_FW_RIWR_LEN(sizeof(*wqe) >> 3)); diff --git a/drivers/infiniband/hw/cxgb3/iwch_cm.c b/drivers/infiniband/hw/cxgb3/iwch_cm.c index b61630eba912..86975370a4c0 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_cm.c +++ b/drivers/infiniband/hw/cxgb3/iwch_cm.c @@ -175,7 +175,7 @@ static void release_tid(struct t3cdev *tdev, u32 hwtid, struct sk_buff *skb) skb = get_skb(skb, sizeof *req, GFP_KERNEL); if (!skb) return; - req = (struct cpl_tid_release *) skb_put(skb, sizeof(*req)); + req = skb_put(skb, sizeof(*req)); req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD)); OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_TID_RELEASE, hwtid)); skb->priority = CPL_PRIORITY_SETUP; @@ -190,7 +190,7 @@ int iwch_quiesce_tid(struct iwch_ep *ep) if (!skb) return -ENOMEM; - req = (struct cpl_set_tcb_field *) skb_put(skb, sizeof(*req)); + req = skb_put(skb, sizeof(*req)); req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD)); req->wr.wr_lo = htonl(V_WR_TID(ep->hwtid)); OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_SET_TCB_FIELD, ep->hwtid)); @@ -211,7 +211,7 @@ int iwch_resume_tid(struct iwch_ep *ep) if (!skb) return -ENOMEM; - req = (struct cpl_set_tcb_field *) skb_put(skb, sizeof(*req)); + req = skb_put(skb, sizeof(*req)); req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD)); req->wr.wr_lo = htonl(V_WR_TID(ep->hwtid)); OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_SET_TCB_FIELD, ep->hwtid)); @@ -398,7 +398,7 @@ static int send_halfclose(struct iwch_ep *ep, gfp_t gfp) } skb->priority = CPL_PRIORITY_DATA; set_arp_failure_handler(skb, arp_failure_discard); - req = (struct cpl_close_con_req *) skb_put(skb, sizeof(*req)); + req = skb_put(skb, sizeof(*req)); req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_CLOSE_CON)); req->wr.wr_lo = htonl(V_WR_TID(ep->hwtid)); OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_CLOSE_CON_REQ, ep->hwtid)); @@ -417,8 +417,7 @@ static int send_abort(struct iwch_ep *ep, struct sk_buff *skb, gfp_t gfp) } skb->priority = CPL_PRIORITY_DATA; set_arp_failure_handler(skb, abort_arp_failure); - req = (struct cpl_abort_req *) skb_put(skb, sizeof(*req)); - memset(req, 0, sizeof(*req)); + req = skb_put_zero(skb, sizeof(*req)); req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_HOST_ABORT_CON_REQ)); req->wr.wr_lo = htonl(V_WR_TID(ep->hwtid)); OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_ABORT_REQ, ep->hwtid)); @@ -456,7 +455,7 @@ static int send_connect(struct iwch_ep *ep) skb->priority = CPL_PRIORITY_SETUP; set_arp_failure_handler(skb, act_open_req_arp_failure); - req = (struct cpl_act_open_req *) skb_put(skb, sizeof(*req)); + req = skb_put(skb, sizeof(*req)); req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD)); OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_ACT_OPEN_REQ, ep->atid)); req->local_port = ep->com.local_addr.sin_port; @@ -514,7 +513,7 @@ static void send_mpa_req(struct iwch_ep *ep, struct sk_buff *skb) set_arp_failure_handler(skb, arp_failure_discard); skb_reset_transport_header(skb); len = skb->len; - req = (struct tx_data_wr *) skb_push(skb, sizeof(*req)); + req = skb_push(skb, sizeof(*req)); req->wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_TX_DATA)|F_WR_COMPL); req->wr_lo = htonl(V_WR_TID(ep->hwtid)); req->len = htonl(len); @@ -547,7 +546,7 @@ static int send_mpa_reject(struct iwch_ep *ep, const void *pdata, u8 plen) return -ENOMEM; } skb_reserve(skb, sizeof(*req)); - mpa = (struct mpa_message *) skb_put(skb, mpalen); + mpa = skb_put(skb, mpalen); memset(mpa, 0, sizeof(*mpa)); memcpy(mpa->key, MPA_KEY_REP, sizeof(mpa->key)); mpa->flags = MPA_REJECT; @@ -565,7 +564,7 @@ static int send_mpa_reject(struct iwch_ep *ep, const void *pdata, u8 plen) skb->priority = CPL_PRIORITY_DATA; set_arp_failure_handler(skb, arp_failure_discard); skb_reset_transport_header(skb); - req = (struct tx_data_wr *) skb_push(skb, sizeof(*req)); + req = skb_push(skb, sizeof(*req)); req->wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_TX_DATA)|F_WR_COMPL); req->wr_lo = htonl(V_WR_TID(ep->hwtid)); req->len = htonl(mpalen); @@ -597,7 +596,7 @@ static int send_mpa_reply(struct iwch_ep *ep, const void *pdata, u8 plen) } skb->priority = CPL_PRIORITY_DATA; skb_reserve(skb, sizeof(*req)); - mpa = (struct mpa_message *) skb_put(skb, mpalen); + mpa = skb_put(skb, mpalen); memset(mpa, 0, sizeof(*mpa)); memcpy(mpa->key, MPA_KEY_REP, sizeof(mpa->key)); mpa->flags = (ep->mpa_attr.crc_enabled ? MPA_CRC : 0) | @@ -616,7 +615,7 @@ static int send_mpa_reply(struct iwch_ep *ep, const void *pdata, u8 plen) set_arp_failure_handler(skb, arp_failure_discard); skb_reset_transport_header(skb); len = skb->len; - req = (struct tx_data_wr *) skb_push(skb, sizeof(*req)); + req = skb_push(skb, sizeof(*req)); req->wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_TX_DATA)|F_WR_COMPL); req->wr_lo = htonl(V_WR_TID(ep->hwtid)); req->len = htonl(len); @@ -801,7 +800,7 @@ static int update_rx_credits(struct iwch_ep *ep, u32 credits) return 0; } - req = (struct cpl_rx_data_ack *) skb_put(skb, sizeof(*req)); + req = skb_put(skb, sizeof(*req)); req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD)); OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_RX_DATA_ACK, ep->hwtid)); req->credit_dack = htonl(V_RX_CREDITS(credits) | V_RX_FORCE_ACK(1)); @@ -1206,7 +1205,7 @@ static int listen_start(struct iwch_listen_ep *ep) return -ENOMEM; } - req = (struct cpl_pass_open_req *) skb_put(skb, sizeof(*req)); + req = skb_put(skb, sizeof(*req)); req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD)); OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_PASS_OPEN_REQ, ep->stid)); req->local_port = ep->com.local_addr.sin_port; @@ -1247,7 +1246,7 @@ static int listen_stop(struct iwch_listen_ep *ep) pr_err("%s - failed to alloc skb\n", __func__); return -ENOMEM; } - req = (struct cpl_close_listserv_req *) skb_put(skb, sizeof(*req)); + req = skb_put(skb, sizeof(*req)); req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD)); req->cpu_idx = 0; OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_CLOSE_LISTSRV_REQ, ep->stid)); @@ -1615,7 +1614,7 @@ static int peer_abort(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) goto out; } rpl_skb->priority = CPL_PRIORITY_DATA; - rpl = (struct cpl_abort_rpl *) skb_put(rpl_skb, sizeof(*rpl)); + rpl = skb_put(rpl_skb, sizeof(*rpl)); rpl->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_HOST_ABORT_CON_RPL)); rpl->wr.wr_lo = htonl(V_WR_TID(ep->hwtid)); OPCODE_TID(rpl) = htonl(MK_OPCODE_TID(CPL_ABORT_RPL, ep->hwtid)); diff --git a/drivers/infiniband/hw/cxgb3/iwch_qp.c b/drivers/infiniband/hw/cxgb3/iwch_qp.c index ba6d5d281b03..7f633da0185d 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_qp.c +++ b/drivers/infiniband/hw/cxgb3/iwch_qp.c @@ -670,8 +670,7 @@ int iwch_post_zb_read(struct iwch_ep *ep) pr_err("%s cannot send zb_read!!\n", __func__); return -ENOMEM; } - wqe = (union t3_wr *)skb_put(skb, sizeof(struct t3_rdma_read_wr)); - memset(wqe, 0, sizeof(struct t3_rdma_read_wr)); + wqe = skb_put_zero(skb, sizeof(struct t3_rdma_read_wr)); wqe->read.rdmaop = T3_READ_REQ; wqe->read.reserved[0] = 0; wqe->read.reserved[1] = 0; @@ -702,8 +701,7 @@ int iwch_post_terminate(struct iwch_qp *qhp, struct respQ_msg_t *rsp_msg) pr_err("%s cannot send TERMINATE!\n", __func__); return -ENOMEM; } - wqe = (union t3_wr *)skb_put(skb, 40); - memset(wqe, 0, 40); + wqe = skb_put_zero(skb, 40); wqe->send.rdmaop = T3_TERMINATE; /* immediate data length */ diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index 0910faf3587b..e49b34c3b136 100644 --- a/drivers/infiniband/hw/cxgb4/cm.c +++ b/drivers/infiniband/hw/cxgb4/cm.c @@ -398,7 +398,8 @@ void _c4iw_free_ep(struct kref *kref) (const u32 *)&sin6->sin6_addr.s6_addr, 1); } - cxgb4_remove_tid(ep->com.dev->rdev.lldi.tids, 0, ep->hwtid); + cxgb4_remove_tid(ep->com.dev->rdev.lldi.tids, 0, ep->hwtid, + ep->com.local_addr.ss_family); dst_release(ep->dst); cxgb4_l2t_release(ep->l2t); if (ep->mpa_skb) @@ -596,7 +597,7 @@ static int send_flowc(struct c4iw_ep *ep) else nparams = 9; - flowc = (struct fw_flowc_wr *)__skb_put(skb, FLOWC_LEN); + flowc = __skb_put(skb, FLOWC_LEN); flowc->op_to_nparams = cpu_to_be32(FW_WR_OP_V(FW_FLOWC_WR) | FW_FLOWC_WR_NPARAMS_V(nparams)); @@ -786,18 +787,16 @@ static int send_connect(struct c4iw_ep *ep) if (ep->com.remote_addr.ss_family == AF_INET) { switch (CHELSIO_CHIP_VERSION(adapter_type)) { case CHELSIO_T4: - req = (struct cpl_act_open_req *)skb_put(skb, wrlen); + req = skb_put(skb, wrlen); INIT_TP_WR(req, 0); break; case CHELSIO_T5: - t5req = (struct cpl_t5_act_open_req *)skb_put(skb, - wrlen); + t5req = skb_put(skb, wrlen); INIT_TP_WR(t5req, 0); req = (struct cpl_act_open_req *)t5req; break; case CHELSIO_T6: - t6req = (struct cpl_t6_act_open_req *)skb_put(skb, - wrlen); + t6req = skb_put(skb, wrlen); INIT_TP_WR(t6req, 0); req = (struct cpl_act_open_req *)t6req; t5req = (struct cpl_t5_act_open_req *)t6req; @@ -838,18 +837,16 @@ static int send_connect(struct c4iw_ep *ep) } else { switch (CHELSIO_CHIP_VERSION(adapter_type)) { case CHELSIO_T4: - req6 = (struct cpl_act_open_req6 *)skb_put(skb, wrlen); + req6 = skb_put(skb, wrlen); INIT_TP_WR(req6, 0); break; case CHELSIO_T5: - t5req6 = (struct cpl_t5_act_open_req6 *)skb_put(skb, - wrlen); + t5req6 = skb_put(skb, wrlen); INIT_TP_WR(t5req6, 0); req6 = (struct cpl_act_open_req6 *)t5req6; break; case CHELSIO_T6: - t6req6 = (struct cpl_t6_act_open_req6 *)skb_put(skb, - wrlen); + t6req6 = skb_put(skb, wrlen); INIT_TP_WR(t6req6, 0); req6 = (struct cpl_act_open_req6 *)t6req6; t5req6 = (struct cpl_t5_act_open_req6 *)t6req6; @@ -926,8 +923,7 @@ static int send_mpa_req(struct c4iw_ep *ep, struct sk_buff *skb, } set_wr_txq(skb, CPL_PRIORITY_DATA, ep->txq_idx); - req = (struct fw_ofld_tx_data_wr *)skb_put(skb, wrlen); - memset(req, 0, wrlen); + req = skb_put_zero(skb, wrlen); req->op_to_immdlen = cpu_to_be32( FW_WR_OP_V(FW_OFLD_TX_DATA_WR) | FW_WR_COMPL_F | @@ -1033,8 +1029,7 @@ static int send_mpa_reject(struct c4iw_ep *ep, const void *pdata, u8 plen) } set_wr_txq(skb, CPL_PRIORITY_DATA, ep->txq_idx); - req = (struct fw_ofld_tx_data_wr *)skb_put(skb, wrlen); - memset(req, 0, wrlen); + req = skb_put_zero(skb, wrlen); req->op_to_immdlen = cpu_to_be32( FW_WR_OP_V(FW_OFLD_TX_DATA_WR) | FW_WR_COMPL_F | @@ -1114,8 +1109,7 @@ static int send_mpa_reply(struct c4iw_ep *ep, const void *pdata, u8 plen) } set_wr_txq(skb, CPL_PRIORITY_DATA, ep->txq_idx); - req = (struct fw_ofld_tx_data_wr *) skb_put(skb, wrlen); - memset(req, 0, wrlen); + req = skb_put_zero(skb, wrlen); req->op_to_immdlen = cpu_to_be32( FW_WR_OP_V(FW_OFLD_TX_DATA_WR) | FW_WR_COMPL_F | @@ -1199,7 +1193,7 @@ static int act_establish(struct c4iw_dev *dev, struct sk_buff *skb) /* setup the hwtid for this connection */ ep->hwtid = tid; - cxgb4_insert_tid(t, ep, tid); + cxgb4_insert_tid(t, ep, tid, ep->com.local_addr.ss_family); insert_ep_tid(ep); ep->snd_seq = be32_to_cpu(req->snd_isn); @@ -1906,8 +1900,7 @@ static int send_fw_act_open_req(struct c4iw_ep *ep, unsigned int atid) int win; skb = get_skb(NULL, sizeof(*req), GFP_KERNEL); - req = (struct fw_ofld_connection_wr *)__skb_put(skb, sizeof(*req)); - memset(req, 0, sizeof(*req)); + req = __skb_put_zero(skb, sizeof(*req)); req->op_compl = htonl(WR_OP_V(FW_OFLD_CONNECTION_WR)); req->len16_pkd = htonl(FW_WR_LEN16_V(DIV_ROUND_UP(sizeof(*req), 16))); req->le.filter = cpu_to_be32(cxgb4_select_ntuple( @@ -2304,7 +2297,8 @@ fail: (const u32 *)&sin6->sin6_addr.s6_addr, 1); } if (status && act_open_has_tid(status)) - cxgb4_remove_tid(ep->com.dev->rdev.lldi.tids, 0, GET_TID(rpl)); + cxgb4_remove_tid(ep->com.dev->rdev.lldi.tids, 0, GET_TID(rpl), + ep->com.local_addr.ss_family); remove_handle(ep->com.dev, &ep->com.dev->atid_idr, atid); cxgb4_free_atid(t, atid); @@ -2581,7 +2575,8 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb) child_ep->tx_chan, child_ep->smac_idx, child_ep->rss_qid); init_timer(&child_ep->timer); - cxgb4_insert_tid(t, child_ep, hwtid); + cxgb4_insert_tid(t, child_ep, hwtid, + child_ep->com.local_addr.ss_family); insert_ep_tid(child_ep); if (accept_cr(child_ep, skb, req)) { c4iw_put_ep(&parent_ep->com); @@ -2849,7 +2844,8 @@ out: 1); } remove_handle(ep->com.dev, &ep->com.dev->hwtid_idr, ep->hwtid); - cxgb4_remove_tid(ep->com.dev->rdev.lldi.tids, 0, ep->hwtid); + cxgb4_remove_tid(ep->com.dev->rdev.lldi.tids, 0, ep->hwtid, + ep->com.local_addr.ss_family); dst_release(ep->dst); cxgb4_l2t_release(ep->l2t); c4iw_reconnect(ep); @@ -3752,9 +3748,9 @@ static void build_cpl_pass_accept_req(struct sk_buff *skb, int stid , u8 tos) */ memset(&tmp_opt, 0, sizeof(tmp_opt)); tcp_clear_options(&tmp_opt); - tcp_parse_options(skb, &tmp_opt, 0, NULL); + tcp_parse_options(&init_net, skb, &tmp_opt, 0, NULL); - req = (struct cpl_pass_accept_req *)__skb_push(skb, sizeof(*req)); + req = __skb_push(skb, sizeof(*req)); memset(req, 0, sizeof(*req)); req->l2info = cpu_to_be16(SYN_INTF_V(intf) | SYN_MAC_IDX_V(RX_MACIDX_G( @@ -3806,8 +3802,7 @@ static void send_fw_pass_open_req(struct c4iw_dev *dev, struct sk_buff *skb, req_skb = alloc_skb(sizeof(struct fw_ofld_connection_wr), GFP_KERNEL); if (!req_skb) return; - req = (struct fw_ofld_connection_wr *)__skb_put(req_skb, sizeof(*req)); - memset(req, 0, sizeof(*req)); + req = __skb_put_zero(req_skb, sizeof(*req)); req->op_compl = htonl(WR_OP_V(FW_OFLD_CONNECTION_WR) | FW_WR_COMPL_F); req->len16_pkd = htonl(FW_WR_LEN16_V(DIV_ROUND_UP(sizeof(*req), 16))); req->le.version_cpl = htonl(FW_OFLD_CONNECTION_WR_CPL_F); diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c index 14de5bde1b63..e16fcaf6b5a3 100644 --- a/drivers/infiniband/hw/cxgb4/cq.c +++ b/drivers/infiniband/hw/cxgb4/cq.c @@ -44,8 +44,7 @@ static int destroy_cq(struct c4iw_rdev *rdev, struct t4_cq *cq, wr_len = sizeof *res_wr + sizeof *res; set_wr_txq(skb, CPL_PRIORITY_CONTROL, 0); - res_wr = (struct fw_ri_res_wr *)__skb_put(skb, wr_len); - memset(res_wr, 0, wr_len); + res_wr = __skb_put_zero(skb, wr_len); res_wr->op_nres = cpu_to_be32( FW_WR_OP_V(FW_RI_RES_WR) | FW_RI_RES_WR_NRES_V(1) | @@ -114,8 +113,7 @@ static int create_cq(struct c4iw_rdev *rdev, struct t4_cq *cq, } set_wr_txq(skb, CPL_PRIORITY_CONTROL, 0); - res_wr = (struct fw_ri_res_wr *)__skb_put(skb, wr_len); - memset(res_wr, 0, wr_len); + res_wr = __skb_put_zero(skb, wr_len); res_wr->op_nres = cpu_to_be32( FW_WR_OP_V(FW_RI_RES_WR) | FW_RI_RES_WR_NRES_V(1) | diff --git a/drivers/infiniband/hw/cxgb4/mem.c b/drivers/infiniband/hw/cxgb4/mem.c index 3ee7f43e419a..5332f06b99ba 100644 --- a/drivers/infiniband/hw/cxgb4/mem.c +++ b/drivers/infiniband/hw/cxgb4/mem.c @@ -81,8 +81,7 @@ static int _c4iw_write_mem_dma_aligned(struct c4iw_rdev *rdev, u32 addr, } set_wr_txq(skb, CPL_PRIORITY_CONTROL, 0); - req = (struct ulp_mem_io *)__skb_put(skb, wr_len); - memset(req, 0, wr_len); + req = __skb_put_zero(skb, wr_len); INIT_ULPTX_WR(req, wr_len, 0, 0); req->wr.wr_hi = cpu_to_be32(FW_WR_OP_V(FW_ULPTX_WR) | (wait ? FW_WR_COMPL_F : 0)); @@ -142,8 +141,7 @@ static int _c4iw_write_mem_inline(struct c4iw_rdev *rdev, u32 addr, u32 len, } set_wr_txq(skb, CPL_PRIORITY_CONTROL, 0); - req = (struct ulp_mem_io *)__skb_put(skb, wr_len); - memset(req, 0, wr_len); + req = __skb_put_zero(skb, wr_len); INIT_ULPTX_WR(req, wr_len, 0, 0); if (i == (num_wqe-1)) { diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c index 8e4154b4253e..bfc77596acbe 100644 --- a/drivers/infiniband/hw/cxgb4/qp.c +++ b/drivers/infiniband/hw/cxgb4/qp.c @@ -293,8 +293,7 @@ static int create_qp(struct c4iw_rdev *rdev, struct t4_wq *wq, } set_wr_txq(skb, CPL_PRIORITY_CONTROL, 0); - res_wr = (struct fw_ri_res_wr *)__skb_put(skb, wr_len); - memset(res_wr, 0, wr_len); + res_wr = __skb_put_zero(skb, wr_len); res_wr->op_nres = cpu_to_be32( FW_WR_OP_V(FW_RI_RES_WR) | FW_RI_RES_WR_NRES_V(2) | @@ -1228,7 +1227,7 @@ static void post_terminate(struct c4iw_qp *qhp, struct t4_cqe *err_cqe, set_wr_txq(skb, CPL_PRIORITY_DATA, qhp->ep->txq_idx); - wqe = (struct fw_ri_wr *)__skb_put(skb, sizeof(*wqe)); + wqe = __skb_put(skb, sizeof(*wqe)); memset(wqe, 0, sizeof *wqe); wqe->op_compl = cpu_to_be32(FW_WR_OP_V(FW_RI_INIT_WR)); wqe->flowid_len16 = cpu_to_be32( @@ -1350,7 +1349,7 @@ static int rdma_fini(struct c4iw_dev *rhp, struct c4iw_qp *qhp, set_wr_txq(skb, CPL_PRIORITY_DATA, ep->txq_idx); - wqe = (struct fw_ri_wr *)__skb_put(skb, sizeof(*wqe)); + wqe = __skb_put(skb, sizeof(*wqe)); memset(wqe, 0, sizeof *wqe); wqe->op_compl = cpu_to_be32( FW_WR_OP_V(FW_RI_INIT_WR) | @@ -1419,7 +1418,7 @@ static int rdma_init(struct c4iw_dev *rhp, struct c4iw_qp *qhp) } set_wr_txq(skb, CPL_PRIORITY_DATA, qhp->ep->txq_idx); - wqe = (struct fw_ri_wr *)__skb_put(skb, sizeof(*wqe)); + wqe = __skb_put(skb, sizeof(*wqe)); memset(wqe, 0, sizeof *wqe); wqe->op_compl = cpu_to_be32( FW_WR_OP_V(FW_RI_INIT_WR) | diff --git a/drivers/infiniband/hw/hfi1/chip.c b/drivers/infiniband/hw/hfi1/chip.c index 2ba00b89df6a..94b54850ec75 100644 --- a/drivers/infiniband/hw/hfi1/chip.c +++ b/drivers/infiniband/hw/hfi1/chip.c @@ -12847,7 +12847,12 @@ static void remap_intr(struct hfi1_devdata *dd, int isrc, int msix_intr) /* clear from the handled mask of the general interrupt */ m = isrc / 64; n = isrc % 64; - dd->gi_mask[m] &= ~((u64)1 << n); + if (likely(m < CCE_NUM_INT_CSRS)) { + dd->gi_mask[m] &= ~((u64)1 << n); + } else { + dd_dev_err(dd, "remap interrupt err\n"); + return; + } /* direct the chip source to the given MSI-X interrupt */ m = isrc / 8; diff --git a/drivers/infiniband/hw/hfi1/qp.c b/drivers/infiniband/hw/hfi1/qp.c index 650305cc0373..1a7af9f60c13 100644 --- a/drivers/infiniband/hw/hfi1/qp.c +++ b/drivers/infiniband/hw/hfi1/qp.c @@ -647,18 +647,17 @@ void qp_iter_print(struct seq_file *s, struct qp_iter *iter) qp->pid); } -void *qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp, - gfp_t gfp) +void *qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp) { struct hfi1_qp_priv *priv; - priv = kzalloc_node(sizeof(*priv), gfp, rdi->dparms.node); + priv = kzalloc_node(sizeof(*priv), GFP_KERNEL, rdi->dparms.node); if (!priv) return ERR_PTR(-ENOMEM); priv->owner = qp; - priv->s_ahg = kzalloc_node(sizeof(*priv->s_ahg), gfp, + priv->s_ahg = kzalloc_node(sizeof(*priv->s_ahg), GFP_KERNEL, rdi->dparms.node); if (!priv->s_ahg) { kfree(priv); diff --git a/drivers/infiniband/hw/hfi1/qp.h b/drivers/infiniband/hw/hfi1/qp.h index 1eb9cd7b8c19..6fe542b6a927 100644 --- a/drivers/infiniband/hw/hfi1/qp.h +++ b/drivers/infiniband/hw/hfi1/qp.h @@ -123,8 +123,7 @@ void hfi1_migrate_qp(struct rvt_qp *qp); /* * Functions provided by hfi1 driver for rdmavt to use */ -void *qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp, - gfp_t gfp); +void *qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp); void qp_priv_free(struct rvt_dev_info *rdi, struct rvt_qp *qp); unsigned free_all_qps(struct rvt_dev_info *rdi); void notify_qp_reset(struct rvt_qp *qp); diff --git a/drivers/infiniband/hw/hfi1/verbs.c b/drivers/infiniband/hw/hfi1/verbs.c index 90e7b77d68e8..2d19f9bb434d 100644 --- a/drivers/infiniband/hw/hfi1/verbs.c +++ b/drivers/infiniband/hw/hfi1/verbs.c @@ -1779,7 +1779,6 @@ int hfi1_register_ib_device(struct hfi1_devdata *dd) ibdev->alloc_hw_stats = alloc_hw_stats; ibdev->get_hw_stats = get_hw_stats; ibdev->alloc_rdma_netdev = hfi1_vnic_alloc_rn; - ibdev->free_rdma_netdev = hfi1_vnic_free_rn; /* keep process mad in the driver */ ibdev->process_mad = hfi1_process_mad; diff --git a/drivers/infiniband/hw/hfi1/vnic.h b/drivers/infiniband/hw/hfi1/vnic.h index e2c455299b53..4a621cde4abb 100644 --- a/drivers/infiniband/hw/hfi1/vnic.h +++ b/drivers/infiniband/hw/hfi1/vnic.h @@ -176,7 +176,6 @@ struct net_device *hfi1_vnic_alloc_rn(struct ib_device *device, const char *name, unsigned char name_assign_type, void (*setup)(struct net_device *)); -void hfi1_vnic_free_rn(struct net_device *netdev); int hfi1_vnic_send_dma(struct hfi1_devdata *dd, u8 q_idx, struct hfi1_vnic_vport_info *vinfo, struct sk_buff *skb, u64 pbc, u8 plen); diff --git a/drivers/infiniband/hw/hfi1/vnic_main.c b/drivers/infiniband/hw/hfi1/vnic_main.c index b601c2929f8f..339f0cdd56d6 100644 --- a/drivers/infiniband/hw/hfi1/vnic_main.c +++ b/drivers/infiniband/hw/hfi1/vnic_main.c @@ -833,6 +833,15 @@ static const struct net_device_ops hfi1_netdev_ops = { .ndo_get_stats64 = hfi1_vnic_get_stats64, }; +static void hfi1_vnic_free_rn(struct net_device *netdev) +{ + struct hfi1_vnic_vport_info *vinfo = opa_vnic_dev_priv(netdev); + + hfi1_vnic_deinit(vinfo); + mutex_destroy(&vinfo->lock); + free_netdev(netdev); +} + struct net_device *hfi1_vnic_alloc_rn(struct ib_device *device, u8 port_num, enum rdma_netdev_t type, @@ -864,6 +873,7 @@ struct net_device *hfi1_vnic_alloc_rn(struct ib_device *device, vinfo->num_tx_q = dd->chip_sdma_engines; vinfo->num_rx_q = HFI1_NUM_VNIC_CTXT; vinfo->netdev = netdev; + rn->free_rdma_netdev = hfi1_vnic_free_rn; rn->set_id = hfi1_vnic_set_vesw_id; netdev->features = NETIF_F_HIGHDMA | NETIF_F_SG; @@ -892,12 +902,3 @@ init_fail: free_netdev(netdev); return ERR_PTR(rc); } - -void hfi1_vnic_free_rn(struct net_device *netdev) -{ - struct hfi1_vnic_vport_info *vinfo = opa_vnic_dev_priv(netdev); - - hfi1_vnic_deinit(vinfo); - mutex_destroy(&vinfo->lock); - free_netdev(netdev); -} diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c index 37d5d29597a4..23fad6d96944 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c @@ -228,14 +228,14 @@ int hns_roce_v1_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, switch (wr->opcode) { case IB_WR_RDMA_READ: ps_opcode = HNS_ROCE_WQE_OPCODE_RDMA_READ; - set_raddr_seg(wqe, atomic_wr(wr)->remote_addr, - atomic_wr(wr)->rkey); + set_raddr_seg(wqe, rdma_wr(wr)->remote_addr, + rdma_wr(wr)->rkey); break; case IB_WR_RDMA_WRITE: case IB_WR_RDMA_WRITE_WITH_IMM: ps_opcode = HNS_ROCE_WQE_OPCODE_RDMA_WRITE; - set_raddr_seg(wqe, atomic_wr(wr)->remote_addr, - atomic_wr(wr)->rkey); + set_raddr_seg(wqe, rdma_wr(wr)->remote_addr, + rdma_wr(wr)->rkey); break; case IB_WR_SEND: case IB_WR_SEND_WITH_INV: @@ -661,9 +661,11 @@ static int hns_roce_v1_rsv_lp_qp(struct hns_roce_dev *hr_dev) union ib_gid dgid; u64 subnet_prefix; int attr_mask = 0; - int i; + int i, j; int ret; + u8 queue_en[HNS_ROCE_V1_RESV_QP] = { 0 }; u8 phy_port; + u8 port = 0; u8 sl; priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; @@ -709,11 +711,27 @@ static int hns_roce_v1_rsv_lp_qp(struct hns_roce_dev *hr_dev) attr.rnr_retry = 7; attr.timeout = 0x12; attr.path_mtu = IB_MTU_256; + attr.ah_attr.type = RDMA_AH_ATTR_TYPE_ROCE; rdma_ah_set_grh(&attr.ah_attr, NULL, 0, 0, 1, 0); rdma_ah_set_static_rate(&attr.ah_attr, 3); subnet_prefix = cpu_to_be64(0xfe80000000000000LL); for (i = 0; i < HNS_ROCE_V1_RESV_QP; i++) { + phy_port = (i >= HNS_ROCE_MAX_PORTS) ? (i - 2) : + (i % HNS_ROCE_MAX_PORTS); + sl = i / HNS_ROCE_MAX_PORTS; + + for (j = 0; j < caps->num_ports; j++) { + if (hr_dev->iboe.phy_port[j] == phy_port) { + queue_en[i] = 1; + port = j; + break; + } + } + + if (!queue_en[i]) + continue; + 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"); @@ -721,15 +739,7 @@ static int hns_roce_v1_rsv_lp_qp(struct hns_roce_dev *hr_dev) } 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->port = port; hr_qp->phy_port = phy_port; hr_qp->ibqp.qp_type = IB_QPT_RC; hr_qp->ibqp.device = &hr_dev->ib_dev; @@ -739,23 +749,22 @@ static int hns_roce_v1_rsv_lp_qp(struct hns_roce_dev *hr_dev) hr_qp->ibqp.recv_cq = cq; hr_qp->ibqp.send_cq = cq; - rdma_ah_set_port_num(&attr.ah_attr, phy_port + 1); - rdma_ah_set_sl(&attr.ah_attr, phy_port + 1); - attr.port_num = phy_port + 1; + rdma_ah_set_port_num(&attr.ah_attr, port + 1); + rdma_ah_set_sl(&attr.ah_attr, sl); + attr.port_num = port + 1; attr.dest_qp_num = hr_qp->qpn; memcpy(rdma_ah_retrieve_dmac(&attr.ah_attr), - hr_dev->dev_addr[phy_port], + hr_dev->dev_addr[port], MAC_ADDR_OCTET_NUM); memcpy(&dgid.raw, &subnet_prefix, sizeof(u64)); - memcpy(&dgid.raw[8], hr_dev->dev_addr[phy_port], 3); - memcpy(&dgid.raw[13], hr_dev->dev_addr[phy_port] + 3, 3); + memcpy(&dgid.raw[8], hr_dev->dev_addr[port], 3); + memcpy(&dgid.raw[13], hr_dev->dev_addr[port] + 3, 3); dgid.raw[11] = 0xff; dgid.raw[12] = 0xfe; dgid.raw[8] ^= 2; rdma_ah_set_dgid_raw(&attr.ah_attr, dgid.raw); - attr_mask |= IB_QP_PORT; ret = hr_dev->hw->modify_qp(&hr_qp->ibqp, &attr, attr_mask, IB_QPS_RESET, IB_QPS_INIT); @@ -812,6 +821,9 @@ static void hns_roce_v1_release_lp_qp(struct hns_roce_dev *hr_dev) for (i = 0; i < HNS_ROCE_V1_RESV_QP; i++) { hr_qp = free_mr->mr_free_qp[i]; + if (!hr_qp) + continue; + ret = hns_roce_v1_destroy_qp(&hr_qp->ibqp); if (ret) dev_err(dev, "Destroy qp %d for mr free failed(%d)!\n", @@ -963,7 +975,7 @@ static void hns_roce_v1_mr_free_work_fn(struct work_struct *work) msecs_to_jiffies(HNS_ROCE_V1_FREE_MR_TIMEOUT_MSECS) + jiffies; int i; int ret; - int ne; + int ne = 0; mr_work = container_of(work, struct hns_roce_mr_free_work, work); hr_mr = (struct hns_roce_mr *)mr_work->mr; @@ -976,6 +988,10 @@ static void hns_roce_v1_mr_free_work_fn(struct work_struct *work) for (i = 0; i < HNS_ROCE_V1_RESV_QP; i++) { hr_qp = free_mr->mr_free_qp[i]; + if (!hr_qp) + continue; + ne++; + ret = hns_roce_v1_send_lp_wqe(hr_qp); if (ret) { dev_err(dev, @@ -985,7 +1001,6 @@ static void hns_roce_v1_mr_free_work_fn(struct work_struct *work) } } - ne = HNS_ROCE_V1_RESV_QP; do { ret = hns_roce_v1_poll_cq(&mr_free_cq->ib_cq, ne, wc); if (ret < 0) { @@ -995,7 +1010,8 @@ static void hns_roce_v1_mr_free_work_fn(struct work_struct *work) goto free_work; } ne -= ret; - msleep(HNS_ROCE_V1_FREE_MR_WAIT_VALUE); + usleep_range(HNS_ROCE_V1_FREE_MR_WAIT_VALUE * 1000, + (1 + HNS_ROCE_V1_FREE_MR_WAIT_VALUE) * 1000); } while (ne && time_before_eq(jiffies, end)); if (ne != 0) @@ -2181,7 +2197,7 @@ static int hns_roce_v1_poll_one(struct hns_roce_cq *hr_cq, } wc->wr_id = wq->wrid[wq->tail & (wq->wqe_cnt - 1)]; ++wq->tail; - } else { + } else { /* RQ conrespond to CQE */ wc->byte_len = le32_to_cpu(cqe->byte_cnt); opcode = roce_get_field(cqe->cqe_byte_4, @@ -3533,10 +3549,12 @@ static int check_qp_db_process_status(struct hns_roce_dev *hr_dev, 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) + if (cur_cnt - old_cnt > + SDB_ST_CMP_VAL) { success_flags = 1; - else { - send_ptr = roce_get_field(old_send, + } 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, @@ -3641,6 +3659,7 @@ static void hns_roce_v1_destroy_qp_work_fn(struct work_struct *work) struct hns_roce_dev *hr_dev; struct hns_roce_qp *hr_qp; struct device *dev; + unsigned long qpn; int ret; qp_work_entry = container_of(work, struct hns_roce_qp_work, work); @@ -3648,8 +3667,9 @@ static void hns_roce_v1_destroy_qp_work_fn(struct work_struct *work) dev = &hr_dev->pdev->dev; priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; hr_qp = qp_work_entry->qp; + qpn = hr_qp->qpn; - dev_dbg(dev, "Schedule destroy QP(0x%lx) work.\n", hr_qp->qpn); + dev_dbg(dev, "Schedule destroy QP(0x%lx) work.\n", qpn); qp_work_entry->sche_cnt++; @@ -3660,7 +3680,7 @@ static void hns_roce_v1_destroy_qp_work_fn(struct work_struct *work) &qp_work_entry->db_wait_stage); if (ret) { dev_err(dev, "Check QP(0x%lx) db process status failed!\n", - hr_qp->qpn); + qpn); return; } @@ -3674,7 +3694,7 @@ static void hns_roce_v1_destroy_qp_work_fn(struct work_struct *work) 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); + dev_err(dev, "Modify QP(0x%lx) to RST failed!\n", qpn); return; } @@ -3683,14 +3703,14 @@ static void hns_roce_v1_destroy_qp_work_fn(struct work_struct *work) if (hr_qp->ibqp.qp_type == IB_QPT_RC) { /* RC QP, release QPN */ - hns_roce_release_range_qp(hr_dev, hr_qp->qpn, 1); + hns_roce_release_range_qp(hr_dev, 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); + dev_dbg(dev, "Accomplished destroy QP(0x%lx) work.\n", qpn); } int hns_roce_v1_destroy_qp(struct ib_qp *ibqp) diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c index c3b41f95e70a..d9777b662eba 100644 --- a/drivers/infiniband/hw/hns/hns_roce_main.c +++ b/drivers/infiniband/hw/hns/hns_roce_main.c @@ -125,8 +125,6 @@ static int handle_en_event(struct hns_roce_dev *hr_dev, u8 port, return -ENODEV; } - spin_lock_bh(&hr_dev->iboe.lock); - switch (event) { case NETDEV_UP: case NETDEV_CHANGE: @@ -144,7 +142,6 @@ static int handle_en_event(struct hns_roce_dev *hr_dev, u8 port, break; } - spin_unlock_bh(&hr_dev->iboe.lock); return 0; } diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c index 4f5a143fc0a7..ff931c580557 100644 --- a/drivers/infiniband/hw/mlx4/cq.c +++ b/drivers/infiniband/hw/mlx4/cq.c @@ -102,7 +102,7 @@ static int mlx4_ib_alloc_cq_buf(struct mlx4_ib_dev *dev, struct mlx4_ib_cq_buf * int err; err = mlx4_buf_alloc(dev->dev, nent * dev->dev->caps.cqe_size, - PAGE_SIZE * 2, &buf->buf, GFP_KERNEL); + PAGE_SIZE * 2, &buf->buf); if (err) goto out; @@ -113,7 +113,7 @@ static int mlx4_ib_alloc_cq_buf(struct mlx4_ib_dev *dev, struct mlx4_ib_cq_buf * if (err) goto err_buf; - err = mlx4_buf_write_mtt(dev->dev, &buf->mtt, &buf->buf, GFP_KERNEL); + err = mlx4_buf_write_mtt(dev->dev, &buf->mtt, &buf->buf); if (err) goto err_mtt; @@ -219,7 +219,7 @@ struct ib_cq *mlx4_ib_create_cq(struct ib_device *ibdev, uar = &to_mucontext(context)->uar; } else { - err = mlx4_db_alloc(dev->dev, &cq->db, 1, GFP_KERNEL); + err = mlx4_db_alloc(dev->dev, &cq->db, 1); if (err) goto err_cq; diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index 521d0def2d9e..d1b43cbbfea7 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -61,8 +61,7 @@ #include <rdma/mlx4-abi.h> #define DRV_NAME MLX4_IB_DRV_NAME -#define DRV_VERSION "2.2-1" -#define DRV_RELDATE "Feb 2014" +#define DRV_VERSION "4.0-0" #define MLX4_IB_FLOW_MAX_PRIO 0xFFF #define MLX4_IB_FLOW_QPN_MASK 0xFFFFFF @@ -79,7 +78,7 @@ MODULE_PARM_DESC(sm_guid_assign, "Enable SM alias_GUID assignment if sm_guid_ass static const char mlx4_ib_version[] = DRV_NAME ": Mellanox ConnectX InfiniBand driver v" - DRV_VERSION " (" DRV_RELDATE ")\n"; + DRV_VERSION "\n"; static void do_slave_init(struct mlx4_ib_dev *ibdev, int slave, int do_init); @@ -1156,7 +1155,7 @@ static void mlx4_ib_disassociate_ucontext(struct ib_ucontext *ibcontext) * call to mlx4_ib_vma_close. */ put_task_struct(owning_process); - msleep(1); + usleep_range(1000, 2000); owning_process = get_pid_task(ibcontext->tgid, PIDTYPE_PID); if (!owning_process || diff --git a/drivers/infiniband/hw/mlx4/mcg.c b/drivers/infiniband/hw/mlx4/mcg.c index 3405e947dc1e..b73f89700ef9 100644 --- a/drivers/infiniband/hw/mlx4/mcg.c +++ b/drivers/infiniband/hw/mlx4/mcg.c @@ -1091,7 +1091,7 @@ static void _mlx4_ib_mcg_port_cleanup(struct mlx4_ib_demux_ctx *ctx, int destroy if (!count) break; - msleep(1); + usleep_range(1000, 2000); } while (time_after(end, jiffies)); flush_workqueue(ctx->mcg_wq); diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h index c2b9cbf4da05..9db82e67e959 100644 --- a/drivers/infiniband/hw/mlx4/mlx4_ib.h +++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h @@ -185,7 +185,6 @@ enum mlx4_ib_qp_flags { MLX4_IB_QP_LSO = IB_QP_CREATE_IPOIB_UD_LSO, MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK = IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK, MLX4_IB_QP_NETIF = IB_QP_CREATE_NETIF_QP, - MLX4_IB_QP_CREATE_USE_GFP_NOIO = IB_QP_CREATE_USE_GFP_NOIO, /* Mellanox specific flags start from IB_QP_CREATE_RESERVED_START */ MLX4_IB_ROCE_V2_GSI_QP = MLX4_IB_QP_CREATE_ROCE_V2_GSI, diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c index 996e9058e515..75c0e6c5dd56 100644 --- a/drivers/infiniband/hw/mlx4/qp.c +++ b/drivers/infiniband/hw/mlx4/qp.c @@ -634,8 +634,8 @@ static void mlx4_ib_free_qp_counter(struct mlx4_ib_dev *dev, static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, struct ib_qp_init_attr *init_attr, - struct ib_udata *udata, int sqpn, struct mlx4_ib_qp **caller_qp, - gfp_t gfp) + struct ib_udata *udata, int sqpn, + struct mlx4_ib_qp **caller_qp) { int qpn; int err; @@ -691,14 +691,14 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, if (qp_type == MLX4_IB_QPT_SMI || qp_type == MLX4_IB_QPT_GSI || (qp_type & (MLX4_IB_QPT_PROXY_SMI | MLX4_IB_QPT_PROXY_SMI_OWNER | MLX4_IB_QPT_PROXY_GSI | MLX4_IB_QPT_TUN_SMI_OWNER))) { - sqp = kzalloc(sizeof (struct mlx4_ib_sqp), gfp); + sqp = kzalloc(sizeof(struct mlx4_ib_sqp), GFP_KERNEL); if (!sqp) return -ENOMEM; qp = &sqp->qp; qp->pri.vid = 0xFFFF; qp->alt.vid = 0xFFFF; } else { - qp = kzalloc(sizeof (struct mlx4_ib_qp), gfp); + qp = kzalloc(sizeof(struct mlx4_ib_qp), GFP_KERNEL); if (!qp) return -ENOMEM; qp->pri.vid = 0xFFFF; @@ -780,7 +780,7 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, goto err; if (qp_has_rq(init_attr)) { - err = mlx4_db_alloc(dev->dev, &qp->db, 0, gfp); + err = mlx4_db_alloc(dev->dev, &qp->db, 0); if (err) goto err; @@ -788,7 +788,7 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, } if (mlx4_buf_alloc(dev->dev, qp->buf_size, qp->buf_size, - &qp->buf, gfp)) { + &qp->buf)) { memcpy(&init_attr->cap, &backup_cap, sizeof(backup_cap)); err = set_kernel_sq_size(dev, &init_attr->cap, qp_type, @@ -797,7 +797,7 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, goto err_db; if (mlx4_buf_alloc(dev->dev, qp->buf_size, - PAGE_SIZE * 2, &qp->buf, gfp)) { + PAGE_SIZE * 2, &qp->buf)) { err = -ENOMEM; goto err_db; } @@ -808,20 +808,20 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, if (err) goto err_buf; - err = mlx4_buf_write_mtt(dev->dev, &qp->mtt, &qp->buf, gfp); + err = mlx4_buf_write_mtt(dev->dev, &qp->mtt, &qp->buf); if (err) goto err_mtt; qp->sq.wrid = kmalloc_array(qp->sq.wqe_cnt, sizeof(u64), - gfp | __GFP_NOWARN); + GFP_KERNEL | __GFP_NOWARN); if (!qp->sq.wrid) qp->sq.wrid = __vmalloc(qp->sq.wqe_cnt * sizeof(u64), - gfp, PAGE_KERNEL); + GFP_KERNEL, PAGE_KERNEL); qp->rq.wrid = kmalloc_array(qp->rq.wqe_cnt, sizeof(u64), - gfp | __GFP_NOWARN); + GFP_KERNEL | __GFP_NOWARN); if (!qp->rq.wrid) qp->rq.wrid = __vmalloc(qp->rq.wqe_cnt * sizeof(u64), - gfp, PAGE_KERNEL); + GFP_KERNEL, PAGE_KERNEL); if (!qp->sq.wrid || !qp->rq.wrid) { err = -ENOMEM; goto err_wrid; @@ -859,7 +859,7 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, if (init_attr->create_flags & IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK) qp->flags |= MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK; - err = mlx4_qp_alloc(dev->dev, qpn, &qp->mqp, gfp); + err = mlx4_qp_alloc(dev->dev, qpn, &qp->mqp); if (err) goto err_qpn; @@ -1127,10 +1127,7 @@ static struct ib_qp *_mlx4_ib_create_qp(struct ib_pd *pd, int err; int sup_u_create_flags = MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK; u16 xrcdn = 0; - gfp_t gfp; - gfp = (init_attr->create_flags & MLX4_IB_QP_CREATE_USE_GFP_NOIO) ? - GFP_NOIO : GFP_KERNEL; /* * We only support LSO, vendor flag1, and multicast loopback blocking, * and only for kernel UD QPs. @@ -1140,8 +1137,7 @@ static struct ib_qp *_mlx4_ib_create_qp(struct ib_pd *pd, MLX4_IB_SRIOV_TUNNEL_QP | MLX4_IB_SRIOV_SQP | MLX4_IB_QP_NETIF | - MLX4_IB_QP_CREATE_ROCE_V2_GSI | - MLX4_IB_QP_CREATE_USE_GFP_NOIO)) + MLX4_IB_QP_CREATE_ROCE_V2_GSI)) return ERR_PTR(-EINVAL); if (init_attr->create_flags & IB_QP_CREATE_NETIF_QP) { @@ -1154,7 +1150,6 @@ static struct ib_qp *_mlx4_ib_create_qp(struct ib_pd *pd, return ERR_PTR(-EINVAL); if ((init_attr->create_flags & ~(MLX4_IB_SRIOV_SQP | - MLX4_IB_QP_CREATE_USE_GFP_NOIO | MLX4_IB_QP_CREATE_ROCE_V2_GSI | MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK) && init_attr->qp_type != IB_QPT_UD) || @@ -1179,7 +1174,7 @@ static struct ib_qp *_mlx4_ib_create_qp(struct ib_pd *pd, case IB_QPT_RC: case IB_QPT_UC: case IB_QPT_RAW_PACKET: - qp = kzalloc(sizeof *qp, gfp); + qp = kzalloc(sizeof(*qp), GFP_KERNEL); if (!qp) return ERR_PTR(-ENOMEM); qp->pri.vid = 0xFFFF; @@ -1188,7 +1183,7 @@ static struct ib_qp *_mlx4_ib_create_qp(struct ib_pd *pd, case IB_QPT_UD: { err = create_qp_common(to_mdev(pd->device), pd, init_attr, - udata, 0, &qp, gfp); + udata, 0, &qp); if (err) { kfree(qp); return ERR_PTR(err); @@ -1217,8 +1212,7 @@ static struct ib_qp *_mlx4_ib_create_qp(struct ib_pd *pd, } err = create_qp_common(to_mdev(pd->device), pd, init_attr, udata, - sqpn, - &qp, gfp); + sqpn, &qp); if (err) return ERR_PTR(err); diff --git a/drivers/infiniband/hw/mlx4/srq.c b/drivers/infiniband/hw/mlx4/srq.c index e32dd58937a8..0facaf5f6d23 100644 --- a/drivers/infiniband/hw/mlx4/srq.c +++ b/drivers/infiniband/hw/mlx4/srq.c @@ -135,14 +135,14 @@ struct ib_srq *mlx4_ib_create_srq(struct ib_pd *pd, if (err) goto err_mtt; } else { - err = mlx4_db_alloc(dev->dev, &srq->db, 0, GFP_KERNEL); + err = mlx4_db_alloc(dev->dev, &srq->db, 0); if (err) goto err_srq; *srq->db.db = 0; - if (mlx4_buf_alloc(dev->dev, buf_size, PAGE_SIZE * 2, &srq->buf, - GFP_KERNEL)) { + if (mlx4_buf_alloc(dev->dev, buf_size, PAGE_SIZE * 2, + &srq->buf)) { err = -ENOMEM; goto err_db; } @@ -167,7 +167,7 @@ struct ib_srq *mlx4_ib_create_srq(struct ib_pd *pd, if (err) goto err_buf; - err = mlx4_buf_write_mtt(dev->dev, &srq->mtt, &srq->buf, GFP_KERNEL); + err = mlx4_buf_write_mtt(dev->dev, &srq->mtt, &srq->buf); if (err) goto err_mtt; diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c index 94c049b62c2f..a384d72ea3cd 100644 --- a/drivers/infiniband/hw/mlx5/cq.c +++ b/drivers/infiniband/hw/mlx5/cq.c @@ -788,7 +788,7 @@ static int create_cq_user(struct mlx5_ib_dev *dev, struct ib_udata *udata, *inlen = MLX5_ST_SZ_BYTES(create_cq_in) + MLX5_FLD_SZ_BYTES(create_cq_in, pas[0]) * ncont; - *cqb = mlx5_vzalloc(*inlen); + *cqb = kvzalloc(*inlen, GFP_KERNEL); if (!*cqb) { err = -ENOMEM; goto err_db; @@ -884,7 +884,7 @@ static int create_cq_kernel(struct mlx5_ib_dev *dev, struct mlx5_ib_cq *cq, *inlen = MLX5_ST_SZ_BYTES(create_cq_in) + MLX5_FLD_SZ_BYTES(create_cq_in, pas[0]) * cq->buf.buf.npages; - *cqb = mlx5_vzalloc(*inlen); + *cqb = kvzalloc(*inlen, GFP_KERNEL); if (!*cqb) { err = -ENOMEM; goto err_buf; @@ -1314,7 +1314,7 @@ int mlx5_ib_resize_cq(struct ib_cq *ibcq, int entries, struct ib_udata *udata) inlen = MLX5_ST_SZ_BYTES(modify_cq_in) + MLX5_FLD_SZ_BYTES(modify_cq_in, pas[0]) * npas; - in = mlx5_vzalloc(inlen); + in = kvzalloc(inlen, GFP_KERNEL); if (!in) { err = -ENOMEM; goto ex_resize; diff --git a/drivers/infiniband/hw/mlx5/mad.c b/drivers/infiniband/hw/mlx5/mad.c index f1b56de64871..95db929bdc34 100644 --- a/drivers/infiniband/hw/mlx5/mad.c +++ b/drivers/infiniband/hw/mlx5/mad.c @@ -218,7 +218,7 @@ static int process_pma_cmd(struct ib_device *ibdev, u8 port_num, (struct ib_pma_portcounters_ext *)(out_mad->data + 40); int sz = MLX5_ST_SZ_BYTES(query_vport_counter_out); - out_cnt = mlx5_vzalloc(sz); + out_cnt = kvzalloc(sz, GFP_KERNEL); if (!out_cnt) return IB_MAD_RESULT_FAILURE; @@ -231,7 +231,7 @@ static int process_pma_cmd(struct ib_device *ibdev, u8 port_num, (struct ib_pma_portcounters *)(out_mad->data + 40); int sz = MLX5_ST_SZ_BYTES(ppcnt_reg); - out_cnt = mlx5_vzalloc(sz); + out_cnt = kvzalloc(sz, GFP_KERNEL); if (!out_cnt) return IB_MAD_RESULT_FAILURE; diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 9ecc089d4529..a7f2e60085c4 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -60,8 +60,7 @@ #include "cmd.h" #define DRIVER_NAME "mlx5_ib" -#define DRIVER_VERSION "2.2-1" -#define DRIVER_RELDATE "Feb 2014" +#define DRIVER_VERSION "5.0-0" MODULE_AUTHOR("Eli Cohen <eli@mellanox.com>"); MODULE_DESCRIPTION("Mellanox Connect-IB HCA IB driver"); @@ -70,7 +69,7 @@ MODULE_VERSION(DRIVER_VERSION); static char mlx5_version[] = DRIVER_NAME ": Mellanox Connect-IB Infiniband driver v" - DRIVER_VERSION " (" DRIVER_RELDATE ")\n"; + DRIVER_VERSION "\n"; enum { MLX5_ATOMIC_SIZE_QP_8BYTES = 1 << 3, @@ -224,8 +223,8 @@ static int translate_eth_proto_oper(u32 eth_proto_oper, u8 *active_speed, return 0; } -static void mlx5_query_port_roce(struct ib_device *device, u8 port_num, - struct ib_port_attr *props) +static int mlx5_query_port_roce(struct ib_device *device, u8 port_num, + struct ib_port_attr *props) { struct mlx5_ib_dev *dev = to_mdev(device); struct mlx5_core_dev *mdev = dev->mdev; @@ -233,12 +232,14 @@ static void mlx5_query_port_roce(struct ib_device *device, u8 port_num, enum ib_mtu ndev_ib_mtu; u16 qkey_viol_cntr; u32 eth_prot_oper; + int err; /* Possible bad flows are checked before filling out props so in case * of an error it will still be zeroed out. */ - if (mlx5_query_port_eth_proto_oper(mdev, ð_prot_oper, port_num)) - return; + err = mlx5_query_port_eth_proto_oper(mdev, ð_prot_oper, port_num); + if (err) + return err; translate_eth_proto_oper(eth_prot_oper, &props->active_speed, &props->active_width); @@ -259,7 +260,7 @@ static void mlx5_query_port_roce(struct ib_device *device, u8 port_num, ndev = mlx5_ib_get_netdev(device, port_num); if (!ndev) - return; + return 0; if (mlx5_lag_is_active(dev->mdev)) { rcu_read_lock(); @@ -282,75 +283,49 @@ static void mlx5_query_port_roce(struct ib_device *device, u8 port_num, dev_put(ndev); props->active_mtu = min(props->max_mtu, ndev_ib_mtu); + return 0; } -static void ib_gid_to_mlx5_roce_addr(const union ib_gid *gid, - const struct ib_gid_attr *attr, - void *mlx5_addr) +static int set_roce_addr(struct mlx5_ib_dev *dev, u8 port_num, + unsigned int index, const union ib_gid *gid, + const struct ib_gid_attr *attr) { -#define MLX5_SET_RA(p, f, v) MLX5_SET(roce_addr_layout, p, f, v) - char *mlx5_addr_l3_addr = MLX5_ADDR_OF(roce_addr_layout, mlx5_addr, - source_l3_address); - void *mlx5_addr_mac = MLX5_ADDR_OF(roce_addr_layout, mlx5_addr, - source_mac_47_32); - - if (!gid) - return; + enum ib_gid_type gid_type = IB_GID_TYPE_IB; + u8 roce_version = 0; + u8 roce_l3_type = 0; + bool vlan = false; + u8 mac[ETH_ALEN]; + u16 vlan_id = 0; - ether_addr_copy(mlx5_addr_mac, attr->ndev->dev_addr); + if (gid) { + gid_type = attr->gid_type; + ether_addr_copy(mac, attr->ndev->dev_addr); - if (is_vlan_dev(attr->ndev)) { - MLX5_SET_RA(mlx5_addr, vlan_valid, 1); - MLX5_SET_RA(mlx5_addr, vlan_id, vlan_dev_vlan_id(attr->ndev)); + if (is_vlan_dev(attr->ndev)) { + vlan = true; + vlan_id = vlan_dev_vlan_id(attr->ndev); + } } - switch (attr->gid_type) { + switch (gid_type) { case IB_GID_TYPE_IB: - MLX5_SET_RA(mlx5_addr, roce_version, MLX5_ROCE_VERSION_1); + roce_version = MLX5_ROCE_VERSION_1; break; case IB_GID_TYPE_ROCE_UDP_ENCAP: - MLX5_SET_RA(mlx5_addr, roce_version, MLX5_ROCE_VERSION_2); + roce_version = MLX5_ROCE_VERSION_2; + if (ipv6_addr_v4mapped((void *)gid)) + roce_l3_type = MLX5_ROCE_L3_TYPE_IPV4; + else + roce_l3_type = MLX5_ROCE_L3_TYPE_IPV6; break; default: - WARN_ON(true); + mlx5_ib_warn(dev, "Unexpected GID type %u\n", gid_type); } - if (attr->gid_type != IB_GID_TYPE_IB) { - if (ipv6_addr_v4mapped((void *)gid)) - MLX5_SET_RA(mlx5_addr, roce_l3_type, - MLX5_ROCE_L3_TYPE_IPV4); - else - MLX5_SET_RA(mlx5_addr, roce_l3_type, - MLX5_ROCE_L3_TYPE_IPV6); - } - - if ((attr->gid_type == IB_GID_TYPE_IB) || - !ipv6_addr_v4mapped((void *)gid)) - memcpy(mlx5_addr_l3_addr, gid, sizeof(*gid)); - else - memcpy(&mlx5_addr_l3_addr[12], &gid->raw[12], 4); -} - -static int set_roce_addr(struct ib_device *device, u8 port_num, - unsigned int index, - const union ib_gid *gid, - const struct ib_gid_attr *attr) -{ - struct mlx5_ib_dev *dev = to_mdev(device); - u32 in[MLX5_ST_SZ_DW(set_roce_address_in)] = {0}; - u32 out[MLX5_ST_SZ_DW(set_roce_address_out)] = {0}; - void *in_addr = MLX5_ADDR_OF(set_roce_address_in, in, roce_address); - enum rdma_link_layer ll = mlx5_ib_port_link_layer(device, port_num); - - if (ll != IB_LINK_LAYER_ETHERNET) - return -EINVAL; - - ib_gid_to_mlx5_roce_addr(gid, attr, in_addr); - - MLX5_SET(set_roce_address_in, in, roce_address_index, index); - MLX5_SET(set_roce_address_in, in, opcode, MLX5_CMD_OP_SET_ROCE_ADDRESS); - return mlx5_cmd_exec(dev->mdev, in, sizeof(in), out, sizeof(out)); + return mlx5_core_roce_gid_set(dev->mdev, index, roce_version, + roce_l3_type, gid->raw, mac, vlan, + vlan_id); } static int mlx5_ib_add_gid(struct ib_device *device, u8 port_num, @@ -358,13 +333,13 @@ static int mlx5_ib_add_gid(struct ib_device *device, u8 port_num, const struct ib_gid_attr *attr, __always_unused void **context) { - return set_roce_addr(device, port_num, index, gid, attr); + return set_roce_addr(to_mdev(device), port_num, index, gid, attr); } static int mlx5_ib_del_gid(struct ib_device *device, u8 port_num, unsigned int index, __always_unused void **context) { - return set_roce_addr(device, port_num, index, NULL, NULL); + return set_roce_addr(to_mdev(device), port_num, index, NULL, NULL); } __be16 mlx5_get_roce_udp_sport(struct mlx5_ib_dev *dev, u8 port_num, @@ -440,7 +415,7 @@ static void get_atomic_caps(struct mlx5_ib_dev *dev, u8 atomic_operations = MLX5_CAP_ATOMIC(dev->mdev, atomic_operations); u8 atomic_size_qp = MLX5_CAP_ATOMIC(dev->mdev, atomic_size_qp); u8 atomic_req_8B_endianness_mode = - MLX5_CAP_ATOMIC(dev->mdev, atomic_req_8B_endianess_mode); + MLX5_CAP_ATOMIC(dev->mdev, atomic_req_8B_endianness_mode); /* Check if HW supports 8 bytes standard atomic operations and capable * of host endianness respond @@ -979,20 +954,31 @@ out: int mlx5_ib_query_port(struct ib_device *ibdev, u8 port, struct ib_port_attr *props) { + unsigned int count; + int ret; + switch (mlx5_get_vport_access_method(ibdev)) { case MLX5_VPORT_ACCESS_METHOD_MAD: - return mlx5_query_mad_ifc_port(ibdev, port, props); + ret = mlx5_query_mad_ifc_port(ibdev, port, props); + break; case MLX5_VPORT_ACCESS_METHOD_HCA: - return mlx5_query_hca_port(ibdev, port, props); + ret = mlx5_query_hca_port(ibdev, port, props); + break; case MLX5_VPORT_ACCESS_METHOD_NIC: - mlx5_query_port_roce(ibdev, port, props); - return 0; + ret = mlx5_query_port_roce(ibdev, port, props); + break; default: - return -EINVAL; + ret = -EINVAL; + } + + if (!ret && props) { + count = mlx5_core_reserved_gids_count(to_mdev(ibdev)->mdev); + props->gid_tbl_len -= count; } + return ret; } static int mlx5_ib_query_gid(struct ib_device *ibdev, u8 port, int index, @@ -2263,7 +2249,7 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, if (!is_valid_attr(dev->mdev, flow_attr)) return ERR_PTR(-EINVAL); - spec = mlx5_vzalloc(sizeof(*spec)); + spec = kvzalloc(sizeof(*spec), GFP_KERNEL); handler = kzalloc(sizeof(*handler), GFP_KERNEL); if (!handler || !spec) { err = -ENOMEM; @@ -3468,7 +3454,7 @@ static int mlx5_ib_query_q_counters(struct mlx5_ib_dev *dev, __be32 val; int ret, i; - out = mlx5_vzalloc(outlen); + out = kvzalloc(outlen, GFP_KERNEL); if (!out) return -ENOMEM; @@ -3497,7 +3483,7 @@ static int mlx5_ib_query_cong_counters(struct mlx5_ib_dev *dev, int ret, i; int offset = port->cnts.num_q_counters; - out = mlx5_vzalloc(outlen); + out = kvzalloc(outlen, GFP_KERNEL); if (!out) return -ENOMEM; @@ -3542,6 +3528,11 @@ static int mlx5_ib_get_hw_stats(struct ib_device *ibdev, return num_counters; } +static void mlx5_ib_free_rdma_netdev(struct net_device *netdev) +{ + return mlx5_rdma_netdev_free(netdev); +} + static struct net_device* mlx5_ib_alloc_rdma_netdev(struct ib_device *hca, u8 port_num, @@ -3550,16 +3541,19 @@ mlx5_ib_alloc_rdma_netdev(struct ib_device *hca, unsigned char name_assign_type, void (*setup)(struct net_device *)) { + struct net_device *netdev; + struct rdma_netdev *rn; + if (type != RDMA_NETDEV_IPOIB) return ERR_PTR(-EOPNOTSUPP); - return mlx5_rdma_netdev_alloc(to_mdev(hca)->mdev, hca, - name, setup); -} - -static void mlx5_ib_free_rdma_netdev(struct net_device *netdev) -{ - return mlx5_rdma_netdev_free(netdev); + netdev = mlx5_rdma_netdev_alloc(to_mdev(hca)->mdev, hca, + name, setup); + if (likely(!IS_ERR_OR_NULL(netdev))) { + rn = netdev_priv(netdev); + rn->free_rdma_netdev = mlx5_ib_free_rdma_netdev; + } + return netdev; } static void *mlx5_ib_add(struct mlx5_core_dev *mdev) @@ -3692,10 +3686,9 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev) dev->ib_dev.check_mr_status = mlx5_ib_check_mr_status; dev->ib_dev.get_port_immutable = mlx5_port_immutable; dev->ib_dev.get_dev_fw_str = get_dev_fw_str; - if (MLX5_CAP_GEN(mdev, ipoib_enhanced_offloads)) { + if (MLX5_CAP_GEN(mdev, ipoib_enhanced_offloads)) dev->ib_dev.alloc_rdma_netdev = mlx5_ib_alloc_rdma_netdev; - dev->ib_dev.free_rdma_netdev = mlx5_ib_free_rdma_netdev; - } + if (mlx5_core_is_pf(mdev)) { dev->ib_dev.get_vf_config = mlx5_ib_get_vf_config; dev->ib_dev.set_vf_link_state = mlx5_ib_set_vf_link_state; diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c index 366433f71b58..8ab2f1360a45 100644 --- a/drivers/infiniband/hw/mlx5/mr.c +++ b/drivers/infiniband/hw/mlx5/mr.c @@ -582,6 +582,15 @@ static void clean_keys(struct mlx5_ib_dev *dev, int c) } } +static void mlx5_mr_cache_debugfs_cleanup(struct mlx5_ib_dev *dev) +{ + if (!mlx5_debugfs_root) + return; + + debugfs_remove_recursive(dev->cache.root); + dev->cache.root = NULL; +} + static int mlx5_mr_cache_debugfs_init(struct mlx5_ib_dev *dev) { struct mlx5_mr_cache *cache = &dev->cache; @@ -600,38 +609,34 @@ static int mlx5_mr_cache_debugfs_init(struct mlx5_ib_dev *dev) sprintf(ent->name, "%d", ent->order); ent->dir = debugfs_create_dir(ent->name, cache->root); if (!ent->dir) - return -ENOMEM; + goto err; ent->fsize = debugfs_create_file("size", 0600, ent->dir, ent, &size_fops); if (!ent->fsize) - return -ENOMEM; + goto err; ent->flimit = debugfs_create_file("limit", 0600, ent->dir, ent, &limit_fops); if (!ent->flimit) - return -ENOMEM; + goto err; ent->fcur = debugfs_create_u32("cur", 0400, ent->dir, &ent->cur); if (!ent->fcur) - return -ENOMEM; + goto err; ent->fmiss = debugfs_create_u32("miss", 0600, ent->dir, &ent->miss); if (!ent->fmiss) - return -ENOMEM; + goto err; } return 0; -} - -static void mlx5_mr_cache_debugfs_cleanup(struct mlx5_ib_dev *dev) -{ - if (!mlx5_debugfs_root) - return; +err: + mlx5_mr_cache_debugfs_cleanup(dev); - debugfs_remove_recursive(dev->cache.root); + return -ENOMEM; } static void delay_time_func(unsigned long ctx) @@ -692,6 +697,11 @@ int mlx5_mr_cache_init(struct mlx5_ib_dev *dev) if (err) mlx5_ib_warn(dev, "cache debugfs failure\n"); + /* + * We don't want to fail driver if debugfs failed to initialize, + * so we are not forwarding error to the user. + */ + return 0; } @@ -1110,7 +1120,7 @@ static struct mlx5_ib_mr *reg_create(struct ib_mr *ibmr, struct ib_pd *pd, inlen = MLX5_ST_SZ_BYTES(create_mkey_in) + sizeof(*pas) * ((npages + 1) / 2) * 2; - in = mlx5_vzalloc(inlen); + in = kvzalloc(inlen, GFP_KERNEL); if (!in) { err = -ENOMEM; goto err_1; @@ -1779,7 +1789,7 @@ mlx5_ib_sg_to_klms(struct mlx5_ib_mr *mr, mr->ndescs = sg_nents; for_each_sg(sgl, sg, sg_nents, i) { - if (unlikely(i > mr->max_descs)) + if (unlikely(i >= mr->max_descs)) break; klms[i].va = cpu_to_be64(sg_dma_address(sg) + sg_offset); klms[i].bcount = cpu_to_be32(sg_dma_len(sg) - sg_offset); diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index ebb6768684de..0889ff367c86 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -823,7 +823,7 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd, *inlen = MLX5_ST_SZ_BYTES(create_qp_in) + MLX5_FLD_SZ_BYTES(create_qp_in, pas[0]) * ncont; - *in = mlx5_vzalloc(*inlen); + *in = kvzalloc(*inlen, GFP_KERNEL); if (!*in) { err = -ENOMEM; goto err_umem; @@ -931,7 +931,7 @@ static int create_kernel_qp(struct mlx5_ib_dev *dev, qp->sq.qend = mlx5_get_send_wqe(qp, qp->sq.wqe_cnt); *inlen = MLX5_ST_SZ_BYTES(create_qp_in) + MLX5_FLD_SZ_BYTES(create_qp_in, pas[0]) * qp->buf.npages; - *in = mlx5_vzalloc(*inlen); + *in = kvzalloc(*inlen, GFP_KERNEL); if (!*in) { err = -ENOMEM; goto err_buf; @@ -1060,7 +1060,7 @@ static int create_raw_packet_qp_sq(struct mlx5_ib_dev *dev, return err; inlen = MLX5_ST_SZ_BYTES(create_sq_in) + sizeof(u64) * ncont; - in = mlx5_vzalloc(inlen); + in = kvzalloc(inlen, GFP_KERNEL); if (!in) { err = -ENOMEM; goto err_umem; @@ -1140,7 +1140,7 @@ static int create_raw_packet_qp_rq(struct mlx5_ib_dev *dev, u32 rq_pas_size = get_rq_pas_size(qpc); inlen = MLX5_ST_SZ_BYTES(create_rq_in) + rq_pas_size; - in = mlx5_vzalloc(inlen); + in = kvzalloc(inlen, GFP_KERNEL); if (!in) return -ENOMEM; @@ -1193,7 +1193,7 @@ static int create_raw_packet_qp_tir(struct mlx5_ib_dev *dev, int err; inlen = MLX5_ST_SZ_BYTES(create_tir_in); - in = mlx5_vzalloc(inlen); + in = kvzalloc(inlen, GFP_KERNEL); if (!in) return -ENOMEM; @@ -1372,7 +1372,7 @@ static int create_rss_raw_qp_tir(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp, } inlen = MLX5_ST_SZ_BYTES(create_tir_in); - in = mlx5_vzalloc(inlen); + in = kvzalloc(inlen, GFP_KERNEL); if (!in) return -ENOMEM; @@ -1633,7 +1633,7 @@ static int create_qp_common(struct mlx5_ib_dev *dev, struct ib_pd *pd, if (err) return err; } else { - in = mlx5_vzalloc(inlen); + in = kvzalloc(inlen, GFP_KERNEL); if (!in) return -ENOMEM; @@ -2164,7 +2164,7 @@ static int modify_raw_packet_eth_prio(struct mlx5_core_dev *dev, int err; inlen = MLX5_ST_SZ_BYTES(modify_tis_in); - in = mlx5_vzalloc(inlen); + in = kvzalloc(inlen, GFP_KERNEL); if (!in) return -ENOMEM; @@ -2189,7 +2189,7 @@ static int modify_raw_packet_tx_affinity(struct mlx5_core_dev *dev, int err; inlen = MLX5_ST_SZ_BYTES(modify_tis_in); - in = mlx5_vzalloc(inlen); + in = kvzalloc(inlen, GFP_KERNEL); if (!in) return -ENOMEM; @@ -2434,7 +2434,7 @@ static int modify_raw_packet_qp_rq(struct mlx5_ib_dev *dev, int err; inlen = MLX5_ST_SZ_BYTES(modify_rq_in); - in = mlx5_vzalloc(inlen); + in = kvzalloc(inlen, GFP_KERNEL); if (!in) return -ENOMEM; @@ -2479,7 +2479,7 @@ static int modify_raw_packet_qp_sq(struct mlx5_core_dev *dev, int err; inlen = MLX5_ST_SZ_BYTES(modify_sq_in); - in = mlx5_vzalloc(inlen); + in = kvzalloc(inlen, GFP_KERNEL); if (!in) return -ENOMEM; @@ -4281,7 +4281,7 @@ static int query_raw_packet_qp_sq_state(struct mlx5_ib_dev *dev, int err; inlen = MLX5_ST_SZ_BYTES(query_sq_out); - out = mlx5_vzalloc(inlen); + out = kvzalloc(inlen, GFP_KERNEL); if (!out) return -ENOMEM; @@ -4308,7 +4308,7 @@ static int query_raw_packet_qp_rq_state(struct mlx5_ib_dev *dev, int err; inlen = MLX5_ST_SZ_BYTES(query_rq_out); - out = mlx5_vzalloc(inlen); + out = kvzalloc(inlen, GFP_KERNEL); if (!out) return -ENOMEM; @@ -4612,7 +4612,7 @@ static int create_rq(struct mlx5_ib_rwq *rwq, struct ib_pd *pd, dev = to_mdev(pd->device); inlen = MLX5_ST_SZ_BYTES(create_rq_in) + sizeof(u64) * rwq->rq_num_pas; - in = mlx5_vzalloc(inlen); + in = kvzalloc(inlen, GFP_KERNEL); if (!in) return -ENOMEM; @@ -4842,7 +4842,7 @@ struct ib_rwq_ind_table *mlx5_ib_create_rwq_ind_table(struct ib_device *device, return ERR_PTR(-ENOMEM); inlen = MLX5_ST_SZ_BYTES(create_rqt_in) + sizeof(u32) * sz; - in = mlx5_vzalloc(inlen); + in = kvzalloc(inlen, GFP_KERNEL); if (!in) { err = -ENOMEM; goto err; @@ -4921,7 +4921,7 @@ int mlx5_ib_modify_wq(struct ib_wq *wq, struct ib_wq_attr *wq_attr, return -EOPNOTSUPP; inlen = MLX5_ST_SZ_BYTES(modify_rq_in); - in = mlx5_vzalloc(inlen); + in = kvzalloc(inlen, GFP_KERNEL); if (!in) return -ENOMEM; diff --git a/drivers/infiniband/hw/mlx5/srq.c b/drivers/infiniband/hw/mlx5/srq.c index 7cb145f9a6db..43707b101f47 100644 --- a/drivers/infiniband/hw/mlx5/srq.c +++ b/drivers/infiniband/hw/mlx5/srq.c @@ -127,7 +127,7 @@ static int create_srq_user(struct ib_pd *pd, struct mlx5_ib_srq *srq, goto err_umem; } - in->pas = mlx5_vzalloc(sizeof(*in->pas) * ncont); + in->pas = kvzalloc(sizeof(*in->pas) * ncont, GFP_KERNEL); if (!in->pas) { err = -ENOMEM; goto err_umem; @@ -189,7 +189,7 @@ static int create_srq_kernel(struct mlx5_ib_dev *dev, struct mlx5_ib_srq *srq, } mlx5_ib_dbg(dev, "srq->buf.page_shift = %d\n", srq->buf.page_shift); - in->pas = mlx5_vzalloc(sizeof(*in->pas) * srq->buf.npages); + in->pas = kvzalloc(sizeof(*in->pas) * srq->buf.npages, GFP_KERNEL); if (!in->pas) { err = -ENOMEM; goto err_buf; diff --git a/drivers/infiniband/hw/nes/nes.c b/drivers/infiniband/hw/nes/nes.c index 5b9601014f0c..a30aa6527f7e 100644 --- a/drivers/infiniband/hw/nes/nes.c +++ b/drivers/infiniband/hw/nes/nes.c @@ -815,7 +815,7 @@ static struct pci_driver nes_pci_driver = { .remove = nes_remove, }; -static ssize_t nes_show_adapter(struct device_driver *ddp, char *buf) +static ssize_t adapter_show(struct device_driver *ddp, char *buf) { unsigned int devfn = 0xffffffff; unsigned char bus_number = 0xff; @@ -834,7 +834,7 @@ static ssize_t nes_show_adapter(struct device_driver *ddp, char *buf) return snprintf(buf, PAGE_SIZE, "%x:%x\n", bus_number, devfn); } -static ssize_t nes_store_adapter(struct device_driver *ddp, +static ssize_t adapter_store(struct device_driver *ddp, const char *buf, size_t count) { char *p = (char *)buf; @@ -843,7 +843,7 @@ static ssize_t nes_store_adapter(struct device_driver *ddp, return strnlen(buf, count); } -static ssize_t nes_show_ee_cmd(struct device_driver *ddp, char *buf) +static ssize_t eeprom_cmd_show(struct device_driver *ddp, char *buf) { u32 eeprom_cmd = 0xdead; u32 i = 0; @@ -859,7 +859,7 @@ static ssize_t nes_show_ee_cmd(struct device_driver *ddp, char *buf) return snprintf(buf, PAGE_SIZE, "0x%x\n", eeprom_cmd); } -static ssize_t nes_store_ee_cmd(struct device_driver *ddp, +static ssize_t eeprom_cmd_store(struct device_driver *ddp, const char *buf, size_t count) { char *p = (char *)buf; @@ -880,7 +880,7 @@ static ssize_t nes_store_ee_cmd(struct device_driver *ddp, return strnlen(buf, count); } -static ssize_t nes_show_ee_data(struct device_driver *ddp, char *buf) +static ssize_t eeprom_data_show(struct device_driver *ddp, char *buf) { u32 eeprom_data = 0xdead; u32 i = 0; @@ -897,7 +897,7 @@ static ssize_t nes_show_ee_data(struct device_driver *ddp, char *buf) return snprintf(buf, PAGE_SIZE, "0x%x\n", eeprom_data); } -static ssize_t nes_store_ee_data(struct device_driver *ddp, +static ssize_t eeprom_data_store(struct device_driver *ddp, const char *buf, size_t count) { char *p = (char *)buf; @@ -918,7 +918,7 @@ static ssize_t nes_store_ee_data(struct device_driver *ddp, return strnlen(buf, count); } -static ssize_t nes_show_flash_cmd(struct device_driver *ddp, char *buf) +static ssize_t flash_cmd_show(struct device_driver *ddp, char *buf) { u32 flash_cmd = 0xdead; u32 i = 0; @@ -935,7 +935,7 @@ static ssize_t nes_show_flash_cmd(struct device_driver *ddp, char *buf) return snprintf(buf, PAGE_SIZE, "0x%x\n", flash_cmd); } -static ssize_t nes_store_flash_cmd(struct device_driver *ddp, +static ssize_t flash_cmd_store(struct device_driver *ddp, const char *buf, size_t count) { char *p = (char *)buf; @@ -956,7 +956,7 @@ static ssize_t nes_store_flash_cmd(struct device_driver *ddp, return strnlen(buf, count); } -static ssize_t nes_show_flash_data(struct device_driver *ddp, char *buf) +static ssize_t flash_data_show(struct device_driver *ddp, char *buf) { u32 flash_data = 0xdead; u32 i = 0; @@ -973,7 +973,7 @@ static ssize_t nes_show_flash_data(struct device_driver *ddp, char *buf) return snprintf(buf, PAGE_SIZE, "0x%x\n", flash_data); } -static ssize_t nes_store_flash_data(struct device_driver *ddp, +static ssize_t flash_data_store(struct device_driver *ddp, const char *buf, size_t count) { char *p = (char *)buf; @@ -994,12 +994,12 @@ static ssize_t nes_store_flash_data(struct device_driver *ddp, return strnlen(buf, count); } -static ssize_t nes_show_nonidx_addr(struct device_driver *ddp, char *buf) +static ssize_t nonidx_addr_show(struct device_driver *ddp, char *buf) { return snprintf(buf, PAGE_SIZE, "0x%x\n", sysfs_nonidx_addr); } -static ssize_t nes_store_nonidx_addr(struct device_driver *ddp, +static ssize_t nonidx_addr_store(struct device_driver *ddp, const char *buf, size_t count) { char *p = (char *)buf; @@ -1010,7 +1010,7 @@ static ssize_t nes_store_nonidx_addr(struct device_driver *ddp, return strnlen(buf, count); } -static ssize_t nes_show_nonidx_data(struct device_driver *ddp, char *buf) +static ssize_t nonidx_data_show(struct device_driver *ddp, char *buf) { u32 nonidx_data = 0xdead; u32 i = 0; @@ -1027,7 +1027,7 @@ static ssize_t nes_show_nonidx_data(struct device_driver *ddp, char *buf) return snprintf(buf, PAGE_SIZE, "0x%x\n", nonidx_data); } -static ssize_t nes_store_nonidx_data(struct device_driver *ddp, +static ssize_t nonidx_data_store(struct device_driver *ddp, const char *buf, size_t count) { char *p = (char *)buf; @@ -1048,12 +1048,12 @@ static ssize_t nes_store_nonidx_data(struct device_driver *ddp, return strnlen(buf, count); } -static ssize_t nes_show_idx_addr(struct device_driver *ddp, char *buf) +static ssize_t idx_addr_show(struct device_driver *ddp, char *buf) { return snprintf(buf, PAGE_SIZE, "0x%x\n", sysfs_idx_addr); } -static ssize_t nes_store_idx_addr(struct device_driver *ddp, +static ssize_t idx_addr_store(struct device_driver *ddp, const char *buf, size_t count) { char *p = (char *)buf; @@ -1064,7 +1064,7 @@ static ssize_t nes_store_idx_addr(struct device_driver *ddp, return strnlen(buf, count); } -static ssize_t nes_show_idx_data(struct device_driver *ddp, char *buf) +static ssize_t idx_data_show(struct device_driver *ddp, char *buf) { u32 idx_data = 0xdead; u32 i = 0; @@ -1081,7 +1081,7 @@ static ssize_t nes_show_idx_data(struct device_driver *ddp, char *buf) return snprintf(buf, PAGE_SIZE, "0x%x\n", idx_data); } -static ssize_t nes_store_idx_data(struct device_driver *ddp, +static ssize_t idx_data_store(struct device_driver *ddp, const char *buf, size_t count) { char *p = (char *)buf; @@ -1102,11 +1102,7 @@ static ssize_t nes_store_idx_data(struct device_driver *ddp, return strnlen(buf, count); } - -/** - * nes_show_wqm_quanta - */ -static ssize_t nes_show_wqm_quanta(struct device_driver *ddp, char *buf) +static ssize_t wqm_quanta_show(struct device_driver *ddp, char *buf) { u32 wqm_quanta_value = 0xdead; u32 i = 0; @@ -1123,12 +1119,8 @@ static ssize_t nes_show_wqm_quanta(struct device_driver *ddp, char *buf) return snprintf(buf, PAGE_SIZE, "0x%X\n", wqm_quanta_value); } - -/** - * nes_store_wqm_quanta - */ -static ssize_t nes_store_wqm_quanta(struct device_driver *ddp, - const char *buf, size_t count) +static ssize_t wqm_quanta_store(struct device_driver *ddp, const char *buf, + size_t count) { unsigned long wqm_quanta_value; u32 wqm_config1; @@ -1153,26 +1145,16 @@ static ssize_t nes_store_wqm_quanta(struct device_driver *ddp, return strnlen(buf, count); } -static DRIVER_ATTR(adapter, S_IRUSR | S_IWUSR, - nes_show_adapter, nes_store_adapter); -static DRIVER_ATTR(eeprom_cmd, S_IRUSR | S_IWUSR, - nes_show_ee_cmd, nes_store_ee_cmd); -static DRIVER_ATTR(eeprom_data, S_IRUSR | S_IWUSR, - nes_show_ee_data, nes_store_ee_data); -static DRIVER_ATTR(flash_cmd, S_IRUSR | S_IWUSR, - nes_show_flash_cmd, nes_store_flash_cmd); -static DRIVER_ATTR(flash_data, S_IRUSR | S_IWUSR, - nes_show_flash_data, nes_store_flash_data); -static DRIVER_ATTR(nonidx_addr, S_IRUSR | S_IWUSR, - nes_show_nonidx_addr, nes_store_nonidx_addr); -static DRIVER_ATTR(nonidx_data, S_IRUSR | S_IWUSR, - nes_show_nonidx_data, nes_store_nonidx_data); -static DRIVER_ATTR(idx_addr, S_IRUSR | S_IWUSR, - nes_show_idx_addr, nes_store_idx_addr); -static DRIVER_ATTR(idx_data, S_IRUSR | S_IWUSR, - nes_show_idx_data, nes_store_idx_data); -static DRIVER_ATTR(wqm_quanta, S_IRUSR | S_IWUSR, - nes_show_wqm_quanta, nes_store_wqm_quanta); +static DRIVER_ATTR_RW(adapter); +static DRIVER_ATTR_RW(eeprom_cmd); +static DRIVER_ATTR_RW(eeprom_data); +static DRIVER_ATTR_RW(flash_cmd); +static DRIVER_ATTR_RW(flash_data); +static DRIVER_ATTR_RW(nonidx_addr); +static DRIVER_ATTR_RW(nonidx_data); +static DRIVER_ATTR_RW(idx_addr); +static DRIVER_ATTR_RW(idx_data); +static DRIVER_ATTR_RW(wqm_quanta); static int nes_create_driver_sysfs(struct pci_driver *drv) { diff --git a/drivers/infiniband/hw/nes/nes_cm.c b/drivers/infiniband/hw/nes/nes_cm.c index 30b256a2c54e..de4025deaa4a 100644 --- a/drivers/infiniband/hw/nes/nes_cm.c +++ b/drivers/infiniband/hw/nes/nes_cm.c @@ -742,7 +742,7 @@ int schedule_nes_timer(struct nes_cm_node *cm_node, struct sk_buff *skb, if (type == NES_TIMER_TYPE_SEND) { new_send->seq_num = ntohl(tcp_hdr(skb)->seq); - atomic_inc(&new_send->skb->users); + refcount_inc(&new_send->skb->users); spin_lock_irqsave(&cm_node->retrans_list_lock, flags); cm_node->send_entry = new_send; add_ref_cm_node(cm_node); @@ -924,7 +924,7 @@ static void nes_cm_timer_tick(unsigned long pass) flags); break; } - atomic_inc(&send_entry->skb->users); + refcount_inc(&send_entry->skb->users); cm_packets_retrans++; nes_debug(NES_DBG_CM, "Retransmitting send_entry %p " "for node %p, jiffies = %lu, time to send = " diff --git a/drivers/infiniband/hw/nes/nes_hw.c b/drivers/infiniband/hw/nes/nes_hw.c index 8f9d8b4ad583..b0adf65e4bdb 100644 --- a/drivers/infiniband/hw/nes/nes_hw.c +++ b/drivers/infiniband/hw/nes/nes_hw.c @@ -551,7 +551,7 @@ struct nes_adapter *nes_init_adapter(struct nes_device *nesdev, u8 hw_rev) { if ((0x0F000100 == (pcs_control_status0 & 0x0F000100)) || (0x0F000100 == (pcs_control_status1 & 0x0F000100))) int_cnt++; - msleep(1); + usleep_range(1000, 2000); } if (int_cnt > 1) { spin_lock_irqsave(&nesadapter->phy_lock, flags); @@ -592,7 +592,7 @@ struct nes_adapter *nes_init_adapter(struct nes_device *nesdev, u8 hw_rev) { break; } } - msleep(1); + usleep_range(1000, 2000); } } } diff --git a/drivers/infiniband/hw/qedr/main.c b/drivers/infiniband/hw/qedr/main.c index 6a72095d6c7a..b5851fd67d4f 100644 --- a/drivers/infiniband/hw/qedr/main.c +++ b/drivers/infiniband/hw/qedr/main.c @@ -37,7 +37,7 @@ #include <linux/iommu.h> #include <linux/pci.h> #include <net/addrconf.h> -#include <linux/qed/qede_roce.h> + #include <linux/qed/qed_chain.h> #include <linux/qed/qed_if.h> #include "qedr.h" @@ -276,7 +276,7 @@ static int qedr_alloc_resources(struct qedr_dev *dev) QED_CHAIN_CNT_TYPE_U16, n_entries, sizeof(struct regpair *), - &cnq->pbl); + &cnq->pbl, NULL); if (rc) goto err4; @@ -886,9 +886,9 @@ static void qedr_mac_address_change(struct qedr_dev *dev) memcpy(&sgid->raw[8], guid, sizeof(guid)); /* Update LL2 */ - rc = dev->ops->roce_ll2_set_mac_filter(dev->cdev, - dev->gsi_ll2_mac_address, - dev->ndev->dev_addr); + rc = dev->ops->ll2_set_mac_filter(dev->cdev, + dev->gsi_ll2_mac_address, + dev->ndev->dev_addr); ether_addr_copy(dev->gsi_ll2_mac_address, dev->ndev->dev_addr); @@ -902,7 +902,7 @@ static void qedr_mac_address_change(struct qedr_dev *dev) * initialization done before RoCE driver notifies * event to stack. */ -static void qedr_notify(struct qedr_dev *dev, enum qede_roce_event event) +static void qedr_notify(struct qedr_dev *dev, enum qede_rdma_event event) { switch (event) { case QEDE_UP: @@ -931,12 +931,12 @@ static struct qedr_driver qedr_drv = { static int __init qedr_init_module(void) { - return qede_roce_register_driver(&qedr_drv); + return qede_rdma_register_driver(&qedr_drv); } static void __exit qedr_exit_module(void) { - qede_roce_unregister_driver(&qedr_drv); + qede_rdma_unregister_driver(&qedr_drv); } module_init(qedr_init_module); diff --git a/drivers/infiniband/hw/qedr/qedr.h b/drivers/infiniband/hw/qedr/qedr.h index d961f79b317c..ab7784bfdac6 100644 --- a/drivers/infiniband/hw/qedr/qedr.h +++ b/drivers/infiniband/hw/qedr/qedr.h @@ -36,8 +36,8 @@ #include <rdma/ib_addr.h> #include <linux/qed/qed_if.h> #include <linux/qed/qed_chain.h> -#include <linux/qed/qed_roce_if.h> -#include <linux/qed/qede_roce.h> +#include <linux/qed/qed_rdma_if.h> +#include <linux/qed/qede_rdma.h> #include <linux/qed/roce_common.h> #include "qedr_hsi_rdma.h" @@ -153,6 +153,8 @@ struct qedr_dev { u32 dp_module; u8 dp_level; u8 num_hwfns; + u8 gsi_ll2_handle; + uint wq_multiplier; u8 gsi_ll2_mac_address[ETH_ALEN]; int gsi_qp_created; diff --git a/drivers/infiniband/hw/qedr/qedr_cm.c b/drivers/infiniband/hw/qedr/qedr_cm.c index d86dbe814d98..4689e802b332 100644 --- a/drivers/infiniband/hw/qedr/qedr_cm.c +++ b/drivers/infiniband/hw/qedr/qedr_cm.c @@ -44,7 +44,7 @@ #include <rdma/ib_cache.h> #include <linux/qed/qed_if.h> -#include <linux/qed/qed_roce_if.h> +#include <linux/qed/qed_rdma_if.h> #include "qedr.h" #include "verbs.h" #include <rdma/qedr-abi.h> @@ -64,9 +64,14 @@ void qedr_store_gsi_qp_cq(struct qedr_dev *dev, struct qedr_qp *qp, dev->gsi_qp = qp; } -void qedr_ll2_tx_cb(void *_qdev, struct qed_roce_ll2_packet *pkt) +void qedr_ll2_complete_tx_packet(void *cxt, + u8 connection_handle, + void *cookie, + dma_addr_t first_frag_addr, + bool b_last_fragment, bool b_last_packet) { - struct qedr_dev *dev = (struct qedr_dev *)_qdev; + struct qedr_dev *dev = (struct qedr_dev *)cxt; + struct qed_roce_ll2_packet *pkt = cookie; struct qedr_cq *cq = dev->gsi_sqcq; struct qedr_qp *qp = dev->gsi_qp; unsigned long flags; @@ -88,20 +93,26 @@ void qedr_ll2_tx_cb(void *_qdev, struct qed_roce_ll2_packet *pkt) (*cq->ibcq.comp_handler) (&cq->ibcq, cq->ibcq.cq_context); } -void qedr_ll2_rx_cb(void *_dev, struct qed_roce_ll2_packet *pkt, - struct qed_roce_ll2_rx_params *params) +void qedr_ll2_complete_rx_packet(void *cxt, + struct qed_ll2_comp_rx_data *data) { - struct qedr_dev *dev = (struct qedr_dev *)_dev; + struct qedr_dev *dev = (struct qedr_dev *)cxt; struct qedr_cq *cq = dev->gsi_rqcq; struct qedr_qp *qp = dev->gsi_qp; unsigned long flags; spin_lock_irqsave(&qp->q_lock, flags); - qp->rqe_wr_id[qp->rq.gsi_cons].rc = params->rc; - qp->rqe_wr_id[qp->rq.gsi_cons].vlan_id = params->vlan_id; - qp->rqe_wr_id[qp->rq.gsi_cons].sg_list[0].length = pkt->payload[0].len; - ether_addr_copy(qp->rqe_wr_id[qp->rq.gsi_cons].smac, params->smac); + qp->rqe_wr_id[qp->rq.gsi_cons].rc = data->u.data_length_error ? + -EINVAL : 0; + qp->rqe_wr_id[qp->rq.gsi_cons].vlan_id = data->vlan; + /* note: length stands for data length i.e. GRH is excluded */ + qp->rqe_wr_id[qp->rq.gsi_cons].sg_list[0].length = + data->length.data_length; + *((u32 *)&qp->rqe_wr_id[qp->rq.gsi_cons].smac[0]) = + ntohl(data->opaque_data_0); + *((u16 *)&qp->rqe_wr_id[qp->rq.gsi_cons].smac[4]) = + ntohs((u16)data->opaque_data_1); qedr_inc_sw_gsi_cons(&qp->rq); @@ -111,6 +122,14 @@ void qedr_ll2_rx_cb(void *_dev, struct qed_roce_ll2_packet *pkt, (*cq->ibcq.comp_handler) (&cq->ibcq, cq->ibcq.cq_context); } +void qedr_ll2_release_rx_packet(void *cxt, + u8 connection_handle, + void *cookie, + dma_addr_t rx_buf_addr, bool b_last_packet) +{ + /* Do nothing... */ +} + static void qedr_destroy_gsi_cq(struct qedr_dev *dev, struct ib_qp_init_attr *attrs) { @@ -159,27 +178,159 @@ static inline int qedr_check_gsi_qp_attrs(struct qedr_dev *dev, return 0; } +static int qedr_ll2_post_tx(struct qedr_dev *dev, + struct qed_roce_ll2_packet *pkt) +{ + enum qed_ll2_roce_flavor_type roce_flavor; + struct qed_ll2_tx_pkt_info ll2_tx_pkt; + int rc; + int i; + + memset(&ll2_tx_pkt, 0, sizeof(ll2_tx_pkt)); + + roce_flavor = (pkt->roce_mode == ROCE_V1) ? + QED_LL2_ROCE : QED_LL2_RROCE; + + if (pkt->roce_mode == ROCE_V2_IPV4) + ll2_tx_pkt.enable_ip_cksum = 1; + + ll2_tx_pkt.num_of_bds = 1 /* hdr */ + pkt->n_seg; + ll2_tx_pkt.vlan = 0; + ll2_tx_pkt.tx_dest = pkt->tx_dest; + ll2_tx_pkt.qed_roce_flavor = roce_flavor; + ll2_tx_pkt.first_frag = pkt->header.baddr; + ll2_tx_pkt.first_frag_len = pkt->header.len; + ll2_tx_pkt.cookie = pkt; + + /* tx header */ + rc = dev->ops->ll2_prepare_tx_packet(dev->rdma_ctx, + dev->gsi_ll2_handle, + &ll2_tx_pkt, 1); + if (rc) { + /* TX failed while posting header - release resources */ + dma_free_coherent(&dev->pdev->dev, pkt->header.len, + pkt->header.vaddr, pkt->header.baddr); + kfree(pkt); + + DP_ERR(dev, "roce ll2 tx: header failed (rc=%d)\n", rc); + return rc; + } + + /* tx payload */ + for (i = 0; i < pkt->n_seg; i++) { + rc = dev->ops->ll2_set_fragment_of_tx_packet( + dev->rdma_ctx, + dev->gsi_ll2_handle, + pkt->payload[i].baddr, + pkt->payload[i].len); + + if (rc) { + /* if failed not much to do here, partial packet has + * been posted we can't free memory, will need to wait + * for completion + */ + DP_ERR(dev, "ll2 tx: payload failed (rc=%d)\n", rc); + return rc; + } + } + + return 0; +} + +int qedr_ll2_stop(struct qedr_dev *dev) +{ + int rc; + + if (dev->gsi_ll2_handle == QED_LL2_UNUSED_HANDLE) + return 0; + + /* remove LL2 MAC address filter */ + rc = dev->ops->ll2_set_mac_filter(dev->cdev, + dev->gsi_ll2_mac_address, NULL); + + rc = dev->ops->ll2_terminate_connection(dev->rdma_ctx, + dev->gsi_ll2_handle); + if (rc) + DP_ERR(dev, "Failed to terminate LL2 connection (rc=%d)\n", rc); + + dev->ops->ll2_release_connection(dev->rdma_ctx, dev->gsi_ll2_handle); + + dev->gsi_ll2_handle = QED_LL2_UNUSED_HANDLE; + + return rc; +} + +int qedr_ll2_start(struct qedr_dev *dev, + struct ib_qp_init_attr *attrs, struct qedr_qp *qp) +{ + struct qed_ll2_acquire_data data; + struct qed_ll2_cbs cbs; + int rc; + + /* configure and start LL2 */ + cbs.rx_comp_cb = qedr_ll2_complete_rx_packet; + cbs.tx_comp_cb = qedr_ll2_complete_tx_packet; + cbs.rx_release_cb = qedr_ll2_release_rx_packet; + cbs.tx_release_cb = qedr_ll2_complete_tx_packet; + cbs.cookie = dev; + + memset(&data, 0, sizeof(data)); + data.input.conn_type = QED_LL2_TYPE_ROCE; + data.input.mtu = dev->ndev->mtu; + data.input.rx_num_desc = attrs->cap.max_recv_wr; + data.input.rx_drop_ttl0_flg = true; + data.input.rx_vlan_removal_en = false; + data.input.tx_num_desc = attrs->cap.max_send_wr; + data.input.tx_tc = 0; + data.input.tx_dest = QED_LL2_TX_DEST_NW; + data.input.ai_err_packet_too_big = QED_LL2_DROP_PACKET; + data.input.ai_err_no_buf = QED_LL2_DROP_PACKET; + data.input.gsi_enable = 1; + data.p_connection_handle = &dev->gsi_ll2_handle; + data.cbs = &cbs; + + rc = dev->ops->ll2_acquire_connection(dev->rdma_ctx, &data); + if (rc) { + DP_ERR(dev, + "ll2 start: failed to acquire LL2 connection (rc=%d)\n", + rc); + return rc; + } + + rc = dev->ops->ll2_establish_connection(dev->rdma_ctx, + dev->gsi_ll2_handle); + if (rc) { + DP_ERR(dev, + "ll2 start: failed to establish LL2 connection (rc=%d)\n", + rc); + goto err1; + } + + rc = dev->ops->ll2_set_mac_filter(dev->cdev, NULL, dev->ndev->dev_addr); + if (rc) + goto err2; + + return 0; + +err2: + dev->ops->ll2_terminate_connection(dev->rdma_ctx, dev->gsi_ll2_handle); +err1: + dev->ops->ll2_release_connection(dev->rdma_ctx, dev->gsi_ll2_handle); + + return rc; +} + struct ib_qp *qedr_create_gsi_qp(struct qedr_dev *dev, struct ib_qp_init_attr *attrs, struct qedr_qp *qp) { - struct qed_roce_ll2_params ll2_params; int rc; rc = qedr_check_gsi_qp_attrs(dev, attrs); if (rc) return ERR_PTR(rc); - /* configure and start LL2 */ - memset(&ll2_params, 0, sizeof(ll2_params)); - ll2_params.max_tx_buffers = attrs->cap.max_send_wr; - ll2_params.max_rx_buffers = attrs->cap.max_recv_wr; - ll2_params.cbs.tx_cb = qedr_ll2_tx_cb; - ll2_params.cbs.rx_cb = qedr_ll2_rx_cb; - ll2_params.cb_cookie = (void *)dev; - ll2_params.mtu = dev->ndev->mtu; - ether_addr_copy(ll2_params.mac_address, dev->ndev->dev_addr); - rc = dev->ops->roce_ll2_start(dev->cdev, &ll2_params); + rc = qedr_ll2_start(dev, attrs, qp); if (rc) { DP_ERR(dev, "create gsi qp: failed on ll2 start. rc=%d\n", rc); return ERR_PTR(rc); @@ -214,7 +365,7 @@ struct ib_qp *qedr_create_gsi_qp(struct qedr_dev *dev, err: kfree(qp->rqe_wr_id); - rc = dev->ops->roce_ll2_stop(dev->cdev); + rc = qedr_ll2_stop(dev); if (rc) DP_ERR(dev, "create gsi qp: failed destroy on create\n"); @@ -223,15 +374,7 @@ err: int qedr_destroy_gsi_qp(struct qedr_dev *dev) { - int rc; - - rc = dev->ops->roce_ll2_stop(dev->cdev); - if (rc) - DP_ERR(dev, "destroy gsi qp: failed (rc=%d)\n", rc); - else - DP_DEBUG(dev, QEDR_MSG_GSI, "destroy gsi qp: success\n"); - - return rc; + return qedr_ll2_stop(dev); } #define QEDR_MAX_UD_HEADER_SIZE (100) @@ -421,7 +564,6 @@ int qedr_gsi_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, { struct qed_roce_ll2_packet *pkt = NULL; struct qedr_qp *qp = get_qedr_qp(ibqp); - struct qed_roce_ll2_tx_params params; struct qedr_dev *dev = qp->dev; unsigned long flags; int rc; @@ -449,8 +591,6 @@ int qedr_gsi_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, goto err; } - memset(¶ms, 0, sizeof(params)); - spin_lock_irqsave(&qp->q_lock, flags); rc = qedr_gsi_build_packet(dev, qp, wr, &pkt); @@ -459,7 +599,8 @@ int qedr_gsi_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, goto err; } - rc = dev->ops->roce_ll2_tx(dev->cdev, pkt, ¶ms); + rc = qedr_ll2_post_tx(dev, pkt); + if (!rc) { qp->wqe_wr_id[qp->sq.prod].wr_id = wr->wr_id; qedr_inc_sw_prod(&qp->sq); @@ -467,17 +608,6 @@ int qedr_gsi_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, "gsi post send: opcode=%d, in_irq=%ld, irqs_disabled=%d, wr_id=%llx\n", wr->opcode, in_irq(), irqs_disabled(), wr->wr_id); } else { - if (rc == QED_ROCE_TX_HEAD_FAILURE) { - /* TX failed while posting header - release resources */ - dma_free_coherent(&dev->pdev->dev, pkt->header.len, - pkt->header.vaddr, pkt->header.baddr); - kfree(pkt); - } else if (rc == QED_ROCE_TX_FRAG_FAILURE) { - /* NTD since TX failed while posting a fragment. We will - * release the resources on TX callback - */ - } - DP_ERR(dev, "gsi post send: failed to transmit (rc=%d)\n", rc); rc = -EAGAIN; *bad_wr = wr; @@ -504,10 +634,8 @@ int qedr_gsi_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr, { struct qedr_dev *dev = get_qedr_dev(ibqp->device); struct qedr_qp *qp = get_qedr_qp(ibqp); - struct qed_roce_ll2_buffer buf; unsigned long flags; - int status = 0; - int rc; + int rc = 0; if ((qp->state != QED_ROCE_QP_STATE_RTR) && (qp->state != QED_ROCE_QP_STATE_RTS)) { @@ -518,8 +646,6 @@ int qedr_gsi_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr, return -EINVAL; } - memset(&buf, 0, sizeof(buf)); - spin_lock_irqsave(&qp->q_lock, flags); while (wr) { @@ -530,10 +656,12 @@ int qedr_gsi_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr, goto err; } - buf.baddr = wr->sg_list[0].addr; - buf.len = wr->sg_list[0].length; - - rc = dev->ops->roce_ll2_post_rx_buffer(dev->cdev, &buf, 0, 1); + rc = dev->ops->ll2_post_rx_buffer(dev->rdma_ctx, + dev->gsi_ll2_handle, + wr->sg_list[0].addr, + wr->sg_list[0].length, + 0 /* cookie */, + 1 /* notify_fw */); if (rc) { DP_ERR(dev, "gsi post recv: failed to post rx buffer (rc=%d)\n", @@ -553,7 +681,7 @@ int qedr_gsi_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr, spin_unlock_irqrestore(&qp->q_lock, flags); - return status; + return rc; err: spin_unlock_irqrestore(&qp->q_lock, flags); *bad_wr = wr; diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c index d6723c365c7f..548e4d1e998f 100644 --- a/drivers/infiniband/hw/qedr/verbs.c +++ b/drivers/infiniband/hw/qedr/verbs.c @@ -935,7 +935,7 @@ struct ib_cq *qedr_create_cq(struct ib_device *ibdev, QED_CHAIN_CNT_TYPE_U32, chain_entries, sizeof(union rdma_cqe), - &cq->pbl); + &cq->pbl, NULL); if (rc) goto err1; @@ -1423,7 +1423,7 @@ qedr_roce_create_kernel_qp(struct qedr_dev *dev, QED_CHAIN_CNT_TYPE_U32, n_sq_elems, QEDR_SQE_ELEMENT_SIZE, - &qp->sq.pbl); + &qp->sq.pbl, NULL); if (rc) return rc; @@ -1437,7 +1437,7 @@ qedr_roce_create_kernel_qp(struct qedr_dev *dev, QED_CHAIN_CNT_TYPE_U32, n_rq_elems, QEDR_RQE_ELEMENT_SIZE, - &qp->rq.pbl); + &qp->rq.pbl, NULL); if (rc) return rc; diff --git a/drivers/infiniband/hw/qib/qib_qp.c b/drivers/infiniband/hw/qib/qib_qp.c index 5984981e7dd4..a343e3b5d4cb 100644 --- a/drivers/infiniband/hw/qib/qib_qp.c +++ b/drivers/infiniband/hw/qib/qib_qp.c @@ -104,10 +104,9 @@ const struct rvt_operation_params qib_post_parms[RVT_OPERATION_MAX] = { }; -static void get_map_page(struct rvt_qpn_table *qpt, struct rvt_qpn_map *map, - gfp_t gfp) +static void get_map_page(struct rvt_qpn_table *qpt, struct rvt_qpn_map *map) { - unsigned long page = get_zeroed_page(gfp); + unsigned long page = get_zeroed_page(GFP_KERNEL); /* * Free the page if someone raced with us installing it. @@ -126,7 +125,7 @@ static void get_map_page(struct rvt_qpn_table *qpt, struct rvt_qpn_map *map, * zero/one for QP type IB_QPT_SMI/IB_QPT_GSI. */ int qib_alloc_qpn(struct rvt_dev_info *rdi, struct rvt_qpn_table *qpt, - enum ib_qp_type type, u8 port, gfp_t gfp) + enum ib_qp_type type, u8 port) { u32 i, offset, max_scan, qpn; struct rvt_qpn_map *map; @@ -160,7 +159,7 @@ int qib_alloc_qpn(struct rvt_dev_info *rdi, struct rvt_qpn_table *qpt, max_scan = qpt->nmaps - !offset; for (i = 0;;) { if (unlikely(!map->page)) { - get_map_page(qpt, map, gfp); + get_map_page(qpt, map); if (unlikely(!map->page)) break; } @@ -317,16 +316,16 @@ u32 qib_mtu_from_qp(struct rvt_dev_info *rdi, struct rvt_qp *qp, u32 pmtu) return ib_mtu_enum_to_int(pmtu); } -void *qib_qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp, gfp_t gfp) +void *qib_qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp) { struct qib_qp_priv *priv; - priv = kzalloc(sizeof(*priv), gfp); + priv = kzalloc(sizeof(*priv), GFP_KERNEL); if (!priv) return ERR_PTR(-ENOMEM); priv->owner = qp; - priv->s_hdr = kzalloc(sizeof(*priv->s_hdr), gfp); + priv->s_hdr = kzalloc(sizeof(*priv->s_hdr), GFP_KERNEL); if (!priv->s_hdr) { kfree(priv); return ERR_PTR(-ENOMEM); diff --git a/drivers/infiniband/hw/qib/qib_verbs.h b/drivers/infiniband/hw/qib/qib_verbs.h index da0db5485ddc..a52fc67b40d7 100644 --- a/drivers/infiniband/hw/qib/qib_verbs.h +++ b/drivers/infiniband/hw/qib/qib_verbs.h @@ -274,11 +274,11 @@ int qib_get_counters(struct qib_pportdata *ppd, * Functions provided by qib driver for rdmavt to use */ unsigned qib_free_all_qps(struct rvt_dev_info *rdi); -void *qib_qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp, gfp_t gfp); +void *qib_qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp); 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); + enum ib_qp_type type, u8 port); void qib_restart_rc(struct rvt_qp *qp, u32 psn, int wait); #ifdef CONFIG_DEBUG_FS diff --git a/drivers/infiniband/sw/rdmavt/qp.c b/drivers/infiniband/sw/rdmavt/qp.c index 727e81cc2c8f..459865439a0b 100644 --- a/drivers/infiniband/sw/rdmavt/qp.c +++ b/drivers/infiniband/sw/rdmavt/qp.c @@ -118,10 +118,9 @@ const int ib_rvt_state_ops[IB_QPS_ERR + 1] = { EXPORT_SYMBOL(ib_rvt_state_ops); static void get_map_page(struct rvt_qpn_table *qpt, - struct rvt_qpn_map *map, - gfp_t gfp) + struct rvt_qpn_map *map) { - unsigned long page = get_zeroed_page(gfp); + unsigned long page = get_zeroed_page(GFP_KERNEL); /* * Free the page if someone raced with us installing it. @@ -173,7 +172,7 @@ static int init_qpn_table(struct rvt_dev_info *rdi, struct rvt_qpn_table *qpt) rdi->dparms.qpn_res_start, rdi->dparms.qpn_res_end); for (i = rdi->dparms.qpn_res_start; i <= rdi->dparms.qpn_res_end; i++) { if (!map->page) { - get_map_page(qpt, map, GFP_KERNEL); + get_map_page(qpt, map); if (!map->page) { ret = -ENOMEM; break; @@ -342,14 +341,14 @@ static inline unsigned mk_qpn(struct rvt_qpn_table *qpt, * Return: The queue pair number */ static int alloc_qpn(struct rvt_dev_info *rdi, struct rvt_qpn_table *qpt, - enum ib_qp_type type, u8 port_num, gfp_t gfp) + enum ib_qp_type type, u8 port_num) { u32 i, offset, max_scan, qpn; struct rvt_qpn_map *map; u32 ret; if (rdi->driver_f.alloc_qpn) - return rdi->driver_f.alloc_qpn(rdi, qpt, type, port_num, gfp); + return rdi->driver_f.alloc_qpn(rdi, qpt, type, port_num); if (type == IB_QPT_SMI || type == IB_QPT_GSI) { unsigned n; @@ -374,7 +373,7 @@ static int alloc_qpn(struct rvt_dev_info *rdi, struct rvt_qpn_table *qpt, max_scan = qpt->nmaps - !offset; for (i = 0;;) { if (unlikely(!map->page)) { - get_map_page(qpt, map, gfp); + get_map_page(qpt, map); if (unlikely(!map->page)) break; } @@ -672,7 +671,6 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, struct ib_qp *ret = ERR_PTR(-ENOMEM); struct rvt_dev_info *rdi = ib_to_rvt(ibpd->device); void *priv = NULL; - gfp_t gfp; size_t sqsize; if (!rdi) @@ -680,18 +678,9 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, if (init_attr->cap.max_send_sge > rdi->dparms.props.max_sge || init_attr->cap.max_send_wr > rdi->dparms.props.max_qp_wr || - init_attr->create_flags & ~(IB_QP_CREATE_USE_GFP_NOIO)) + init_attr->create_flags) return ERR_PTR(-EINVAL); - /* GFP_NOIO is applicable to RC QP's only */ - - if (init_attr->create_flags & IB_QP_CREATE_USE_GFP_NOIO && - init_attr->qp_type != IB_QPT_RC) - return ERR_PTR(-EINVAL); - - gfp = init_attr->create_flags & IB_QP_CREATE_USE_GFP_NOIO ? - GFP_NOIO : GFP_KERNEL; - /* Check receive queue parameters if no SRQ is specified. */ if (!init_attr->srq) { if (init_attr->cap.max_recv_sge > rdi->dparms.props.max_sge || @@ -719,14 +708,7 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, sz = sizeof(struct rvt_sge) * init_attr->cap.max_send_sge + sizeof(struct rvt_swqe); - if (gfp == GFP_NOIO) - swq = __vmalloc( - sqsize * sz, - gfp | __GFP_ZERO, PAGE_KERNEL); - else - swq = vzalloc_node( - sqsize * sz, - rdi->dparms.node); + swq = vzalloc_node(sqsize * sz, rdi->dparms.node); if (!swq) return ERR_PTR(-ENOMEM); @@ -741,7 +723,8 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, } else if (init_attr->cap.max_recv_sge > 1) sg_list_sz = sizeof(*qp->r_sg_list) * (init_attr->cap.max_recv_sge - 1); - qp = kzalloc_node(sz + sg_list_sz, gfp, rdi->dparms.node); + qp = kzalloc_node(sz + sg_list_sz, GFP_KERNEL, + rdi->dparms.node); if (!qp) goto bail_swq; @@ -751,7 +734,7 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, kzalloc_node( sizeof(*qp->s_ack_queue) * rvt_max_atomic(rdi), - gfp, + GFP_KERNEL, rdi->dparms.node); if (!qp->s_ack_queue) goto bail_qp; @@ -766,7 +749,7 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, * Driver needs to set up it's private QP structure and do any * initialization that is needed. */ - priv = rdi->driver_f.qp_priv_alloc(rdi, qp, gfp); + priv = rdi->driver_f.qp_priv_alloc(rdi, qp); if (IS_ERR(priv)) { ret = priv; goto bail_qp; @@ -786,11 +769,6 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, qp->r_rq.wq = vmalloc_user( sizeof(struct rvt_rwq) + qp->r_rq.size * sz); - else if (gfp == GFP_NOIO) - qp->r_rq.wq = __vmalloc( - sizeof(struct rvt_rwq) + - qp->r_rq.size * sz, - gfp | __GFP_ZERO, PAGE_KERNEL); else qp->r_rq.wq = vzalloc_node( sizeof(struct rvt_rwq) + @@ -824,7 +802,7 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, err = alloc_qpn(rdi, &rdi->qp_dev->qpn_table, init_attr->qp_type, - init_attr->port_num, gfp); + init_attr->port_num); if (err < 0) { ret = ERR_PTR(err); goto bail_rq_wq; diff --git a/drivers/infiniband/sw/rxe/rxe_net.c b/drivers/infiniband/sw/rxe/rxe_net.c index c3a140ed4df2..08f3f90d2912 100644 --- a/drivers/infiniband/sw/rxe/rxe_net.c +++ b/drivers/infiniband/sw/rxe/rxe_net.c @@ -441,6 +441,8 @@ static void rxe_skb_tx_dtor(struct sk_buff *skb) if (unlikely(qp->need_req_skb && skb_out < RXE_INFLIGHT_SKBS_PER_QP_LOW)) rxe_run_task(&qp->req.task, 1); + + rxe_drop_ref(qp); } int rxe_send(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, struct sk_buff *skb) @@ -473,6 +475,7 @@ int rxe_send(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, struct sk_buff *skb) return -EAGAIN; } + rxe_add_ref(pkt->qp); atomic_inc(&pkt->qp->skb_out); kfree_skb(skb); diff --git a/drivers/infiniband/sw/rxe/rxe_resp.c b/drivers/infiniband/sw/rxe/rxe_resp.c index 23039768f541..be944d5aa9af 100644 --- a/drivers/infiniband/sw/rxe/rxe_resp.c +++ b/drivers/infiniband/sw/rxe/rxe_resp.c @@ -995,7 +995,9 @@ static int send_atomic_ack(struct rxe_qp *qp, struct rxe_pkt_info *pkt, free_rd_atomic_resource(qp, res); rxe_advance_resp_resource(qp); - memcpy(SKB_TO_PKT(skb), &ack_pkt, sizeof(skb->cb)); + memcpy(SKB_TO_PKT(skb), &ack_pkt, sizeof(ack_pkt)); + memset((unsigned char *)SKB_TO_PKT(skb) + sizeof(ack_pkt), 0, + sizeof(skb->cb) - sizeof(ack_pkt)); res->type = RXE_ATOMIC_MASK; res->atomic.skb = skb; diff --git a/drivers/infiniband/sw/rxe/rxe_verbs.c b/drivers/infiniband/sw/rxe/rxe_verbs.c index 073e66783f1d..07511718d98d 100644 --- a/drivers/infiniband/sw/rxe/rxe_verbs.c +++ b/drivers/infiniband/sw/rxe/rxe_verbs.c @@ -1240,6 +1240,8 @@ int rxe_register_device(struct rxe_dev *rxe) addrconf_addr_eui48((unsigned char *)&dev->node_guid, rxe->ndev->dev_addr); dev->dev.dma_ops = &dma_virt_ops; + dma_coerce_mask_and_coherent(&dev->dev, + dma_get_required_mask(dev->dev.parent)); 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/ulp/ipoib/ipoib_cm.c b/drivers/infiniband/ulp/ipoib/ipoib_cm.c index 7cbcfdac6529..f87d104837dc 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_cm.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_cm.c @@ -39,6 +39,7 @@ #include <linux/vmalloc.h> #include <linux/moduleparam.h> #include <linux/sched/signal.h> +#include <linux/sched/mm.h> #include "ipoib.h" @@ -954,7 +955,7 @@ void ipoib_cm_dev_stop(struct net_device *dev) break; } spin_unlock_irq(&priv->lock); - msleep(1); + usleep_range(1000, 2000); ipoib_drain_cq(dev); spin_lock_irq(&priv->lock); } @@ -1047,9 +1048,8 @@ static struct ib_qp *ipoib_cm_create_tx_qp(struct net_device *dev, struct ipoib_ .sq_sig_type = IB_SIGNAL_ALL_WR, .qp_type = IB_QPT_RC, .qp_context = tx, - .create_flags = IB_QP_CREATE_USE_GFP_NOIO + .create_flags = 0 }; - struct ib_qp *tx_qp; if (dev->features & NETIF_F_SG) @@ -1057,10 +1057,6 @@ static struct ib_qp *ipoib_cm_create_tx_qp(struct net_device *dev, struct ipoib_ min_t(u32, priv->ca->attrs.max_sge, MAX_SKB_FRAGS + 1); tx_qp = ib_create_qp(priv->pd, &attr); - if (PTR_ERR(tx_qp) == -EINVAL) { - attr.create_flags &= ~IB_QP_CREATE_USE_GFP_NOIO; - tx_qp = ib_create_qp(priv->pd, &attr); - } tx->max_send_sge = attr.cap.max_send_sge; return tx_qp; } @@ -1131,10 +1127,11 @@ static int ipoib_cm_tx_init(struct ipoib_cm_tx *p, u32 qpn, struct sa_path_rec *pathrec) { struct ipoib_dev_priv *priv = ipoib_priv(p->dev); + unsigned int noio_flag; int ret; - p->tx_ring = __vmalloc(ipoib_sendq_size * sizeof *p->tx_ring, - GFP_NOIO, PAGE_KERNEL); + noio_flag = memalloc_noio_save(); + p->tx_ring = vzalloc(ipoib_sendq_size * sizeof(*p->tx_ring)); if (!p->tx_ring) { ret = -ENOMEM; goto err_tx; @@ -1142,9 +1139,10 @@ static int ipoib_cm_tx_init(struct ipoib_cm_tx *p, u32 qpn, memset(p->tx_ring, 0, ipoib_sendq_size * sizeof *p->tx_ring); p->qp = ipoib_cm_create_tx_qp(p->dev, p); + memalloc_noio_restore(noio_flag); if (IS_ERR(p->qp)) { ret = PTR_ERR(p->qp); - ipoib_warn(priv, "failed to allocate tx qp: %d\n", ret); + ipoib_warn(priv, "failed to create tx qp: %d\n", ret); goto err_qp; } @@ -1206,7 +1204,7 @@ static void ipoib_cm_tx_destroy(struct ipoib_cm_tx *p) goto timeout; } - msleep(1); + usleep_range(1000, 2000); } } diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c index efe7402f4885..57a9655e844d 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c @@ -770,7 +770,7 @@ int ipoib_ib_dev_stop_default(struct net_device *dev) ipoib_drain_cq(dev); - msleep(1); + usleep_range(1000, 2000); } ipoib_dbg(priv, "All sends and receives done.\n"); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 1015a63de6ae..70dacaf9044e 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -233,6 +233,7 @@ static netdev_features_t ipoib_fix_features(struct net_device *dev, netdev_featu static int ipoib_change_mtu(struct net_device *dev, int new_mtu) { struct ipoib_dev_priv *priv = ipoib_priv(dev); + int ret = 0; /* dev->mtu > 2K ==> connected mode */ if (ipoib_cm_admin_enabled(dev)) { @@ -256,9 +257,34 @@ static int ipoib_change_mtu(struct net_device *dev, int new_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); + new_mtu = min(priv->mcast_mtu, priv->admin_mtu); - return 0; + if (priv->rn_ops->ndo_change_mtu) { + bool carrier_status = netif_carrier_ok(dev); + + netif_carrier_off(dev); + + /* notify lower level on the real mtu */ + ret = priv->rn_ops->ndo_change_mtu(dev, new_mtu); + + if (carrier_status) + netif_carrier_on(dev); + } else { + dev->mtu = new_mtu; + } + + return ret; +} + +static void ipoib_get_stats(struct net_device *dev, + struct rtnl_link_stats64 *stats) +{ + struct ipoib_dev_priv *priv = ipoib_priv(dev); + + if (priv->rn_ops->ndo_get_stats64) + priv->rn_ops->ndo_get_stats64(dev, stats); + else + netdev_stats_to_stats64(stats, &dev->stats); } /* Called with an RCU read lock taken */ @@ -681,7 +707,7 @@ 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)); + phdr = skb_push(skb, sizeof(*phdr)); memcpy(phdr->hwaddr, daddr, INFINIBAND_ALEN); } @@ -1129,7 +1155,7 @@ static int ipoib_hard_header(struct sk_buff *skb, { struct ipoib_header *header; - header = (struct ipoib_header *) skb_push(skb, sizeof *header); + header = skb_push(skb, sizeof *header); header->proto = htons(type); header->reserved = 0; @@ -1808,6 +1834,7 @@ static const struct net_device_ops ipoib_netdev_ops_pf = { .ndo_get_vf_stats = ipoib_get_vf_stats, .ndo_set_vf_guid = ipoib_set_vf_guid, .ndo_set_mac_address = ipoib_set_mac, + .ndo_get_stats64 = ipoib_get_stats, }; static const struct net_device_ops ipoib_netdev_ops_vf = { @@ -1893,6 +1920,7 @@ static struct net_device rn->send = ipoib_send; rn->attach_mcast = ipoib_mcast_attach; rn->detach_mcast = ipoib_mcast_detach; + rn->free_rdma_netdev = free_netdev; rn->hca = hca; dev->netdev_ops = &ipoib_netdev_default_pf; @@ -2288,6 +2316,8 @@ static void ipoib_remove_one(struct ib_device *device, void *client_data) return; list_for_each_entry_safe(priv, tmp, dev_list, list) { + struct rdma_netdev *rn = netdev_priv(priv->dev); + ib_unregister_event_handler(&priv->event_handler); flush_workqueue(ipoib_workqueue); @@ -2304,10 +2334,7 @@ static void ipoib_remove_one(struct ib_device *device, void *client_data) flush_workqueue(priv->wq); unregister_netdev(priv->dev); - if (device->free_rdma_netdev) - device->free_rdma_netdev(priv->dev); - else - free_netdev(priv->dev); + rn->free_rdma_netdev(priv->dev); list_for_each_entry_safe(cpriv, tcpriv, &priv->child_intfs, list) kfree(cpriv); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_netlink.c b/drivers/infiniband/ulp/ipoib/ipoib_netlink.c index 28884781311b..3e44087935ae 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_netlink.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_netlink.c @@ -64,8 +64,9 @@ nla_put_failure: return -EMSGSIZE; } -static int ipoib_changelink(struct net_device *dev, - struct nlattr *tb[], struct nlattr *data[]) +static int ipoib_changelink(struct net_device *dev, struct nlattr *tb[], + struct nlattr *data[], + struct netlink_ext_ack *extack) { u16 mode, umcast; int ret = 0; @@ -93,7 +94,8 @@ out_err: } static int ipoib_new_child_link(struct net *src_net, struct net_device *dev, - struct nlattr *tb[], struct nlattr *data[]) + struct nlattr *tb[], struct nlattr *data[], + struct netlink_ext_ack *extack) { struct net_device *pdev; struct ipoib_dev_priv *ppriv; @@ -133,7 +135,7 @@ static int ipoib_new_child_link(struct net *src_net, struct net_device *dev, child_pkey, IPOIB_RTNL_CHILD); if (!err && data) - err = ipoib_changelink(dev, tb, data); + err = ipoib_changelink(dev, tb, data, extack); return err; } diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c index 5a887efb4bdf..37b33d708c2d 100644 --- a/drivers/infiniband/ulp/iser/iscsi_iser.c +++ b/drivers/infiniband/ulp/iser/iscsi_iser.c @@ -83,6 +83,7 @@ static struct scsi_host_template iscsi_iser_sht; static struct iscsi_transport iscsi_iser_transport; static struct scsi_transport_template *iscsi_iser_scsi_transport; static struct workqueue_struct *release_wq; +static DEFINE_MUTEX(unbind_iser_conn_mutex); struct iser_global ig; int iser_debug_level = 0; @@ -550,12 +551,14 @@ iscsi_iser_conn_stop(struct iscsi_cls_conn *cls_conn, int flag) */ if (iser_conn) { mutex_lock(&iser_conn->state_mutex); + mutex_lock(&unbind_iser_conn_mutex); iser_conn_terminate(iser_conn); iscsi_conn_stop(cls_conn, flag); /* unbind */ iser_conn->iscsi_conn = NULL; conn->dd_data = NULL; + mutex_unlock(&unbind_iser_conn_mutex); complete(&iser_conn->stop_completion); mutex_unlock(&iser_conn->state_mutex); @@ -977,13 +980,21 @@ static int iscsi_iser_slave_alloc(struct scsi_device *sdev) struct iser_conn *iser_conn; struct ib_device *ib_dev; + mutex_lock(&unbind_iser_conn_mutex); + session = starget_to_session(scsi_target(sdev))->dd_data; iser_conn = session->leadconn->dd_data; + if (!iser_conn) { + mutex_unlock(&unbind_iser_conn_mutex); + return -ENOTCONN; + } ib_dev = iser_conn->ib_conn.device->ib_device; if (!(ib_dev->attrs.device_cap_flags & IB_DEVICE_SG_GAPS_REG)) blk_queue_virt_boundary(sdev->request_queue, ~MASK_4K); + mutex_unlock(&unbind_iser_conn_mutex); + return 0; } diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c index c538a38c91ce..26a004e97ae0 100644 --- a/drivers/infiniband/ulp/iser/iser_verbs.c +++ b/drivers/infiniband/ulp/iser/iser_verbs.c @@ -708,8 +708,14 @@ iser_calc_scsi_params(struct iser_conn *iser_conn, unsigned short sg_tablesize, sup_sg_tablesize; sg_tablesize = DIV_ROUND_UP(max_sectors * 512, SIZE_4K); - sup_sg_tablesize = min_t(unsigned, ISCSI_ISER_MAX_SG_TABLESIZE, - device->ib_device->attrs.max_fast_reg_page_list_len); + if (device->ib_device->attrs.device_cap_flags & + IB_DEVICE_MEM_MGT_EXTENSIONS) + sup_sg_tablesize = + min_t( + uint, ISCSI_ISER_MAX_SG_TABLESIZE, + device->ib_device->attrs.max_fast_reg_page_list_len); + else + sup_sg_tablesize = ISCSI_ISER_MAX_SG_TABLESIZE; iser_conn->scsi_sg_tablesize = min(sg_tablesize, sup_sg_tablesize); } diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index fcbed35e95a8..0e662656ef42 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -1452,7 +1452,7 @@ static void isert_login_recv_done(struct ib_cq *cq, struct ib_wc *wc) { struct isert_conn *isert_conn = wc->qp->qp_context; - struct ib_device *ib_dev = isert_conn->cm_id->device; + struct ib_device *ib_dev = isert_conn->device->ib_device; if (unlikely(wc->status != IB_WC_SUCCESS)) { isert_print_wc(wc, "login recv"); diff --git a/drivers/infiniband/ulp/opa_vnic/opa_vnic_encap.c b/drivers/infiniband/ulp/opa_vnic/opa_vnic_encap.c index 2e8fee982436..afa938bd26d6 100644 --- a/drivers/infiniband/ulp/opa_vnic/opa_vnic_encap.c +++ b/drivers/infiniband/ulp/opa_vnic/opa_vnic_encap.c @@ -460,7 +460,7 @@ void opa_vnic_encap_skb(struct opa_vnic_adapter *adapter, struct sk_buff *skb) sc = opa_vnic_get_sc(info, skb); l4_hdr = info->vesw.vesw_id; - mdata = (struct opa_vnic_skb_mdata *)skb_push(skb, sizeof(*mdata)); + mdata = skb_push(skb, sizeof(*mdata)); mdata->vl = opa_vnic_get_vl(adapter, skb); mdata->entropy = entropy; mdata->flags = 0; diff --git a/drivers/infiniband/ulp/opa_vnic/opa_vnic_ethtool.c b/drivers/infiniband/ulp/opa_vnic/opa_vnic_ethtool.c index d66540e24885..62390e9e0023 100644 --- a/drivers/infiniband/ulp/opa_vnic/opa_vnic_ethtool.c +++ b/drivers/infiniband/ulp/opa_vnic/opa_vnic_ethtool.c @@ -146,15 +146,15 @@ static void vnic_get_ethtool_stats(struct net_device *netdev, int i; memset(&vstats, 0, sizeof(vstats)); - mutex_lock(&adapter->stats_lock); + spin_lock(&adapter->stats_lock); adapter->rn_ops->ndo_get_stats64(netdev, &vstats.netstats); + spin_unlock(&adapter->stats_lock); for (i = 0; i < VNIC_STATS_LEN; i++) { char *p = (char *)&vstats + vnic_gstrings_stats[i].stat_offset; data[i] = (vnic_gstrings_stats[i].sizeof_stat == sizeof(u64)) ? *(u64 *)p : *(u32 *)p; } - mutex_unlock(&adapter->stats_lock); } /* vnic_get_strings - get strings */ diff --git a/drivers/infiniband/ulp/opa_vnic/opa_vnic_internal.h b/drivers/infiniband/ulp/opa_vnic/opa_vnic_internal.h index 6bba886bec1f..ca29e6d5aedc 100644 --- a/drivers/infiniband/ulp/opa_vnic/opa_vnic_internal.h +++ b/drivers/infiniband/ulp/opa_vnic/opa_vnic_internal.h @@ -214,7 +214,7 @@ struct opa_vnic_adapter { struct mutex mactbl_lock; /* Lock used to protect access to vnic counters */ - struct mutex stats_lock; + spinlock_t stats_lock; u8 flow_tbl[OPA_VNIC_FLOW_TBL_SIZE]; diff --git a/drivers/infiniband/ulp/opa_vnic/opa_vnic_netdev.c b/drivers/infiniband/ulp/opa_vnic/opa_vnic_netdev.c index 905f39dda5aa..1a3c25364b64 100644 --- a/drivers/infiniband/ulp/opa_vnic/opa_vnic_netdev.c +++ b/drivers/infiniband/ulp/opa_vnic/opa_vnic_netdev.c @@ -69,9 +69,9 @@ static void opa_vnic_get_stats64(struct net_device *netdev, struct opa_vnic_stats vstats; memset(&vstats, 0, sizeof(vstats)); - mutex_lock(&adapter->stats_lock); + spin_lock(&adapter->stats_lock); adapter->rn_ops->ndo_get_stats64(netdev, &vstats.netstats); - mutex_unlock(&adapter->stats_lock); + spin_unlock(&adapter->stats_lock); memcpy(stats, &vstats.netstats, sizeof(*stats)); } @@ -103,7 +103,7 @@ static u16 opa_vnic_select_queue(struct net_device *netdev, struct sk_buff *skb, int rc; /* pass entropy and vl as metadata in skb */ - mdata = (struct opa_vnic_skb_mdata *)skb_push(skb, sizeof(*mdata)); + mdata = skb_push(skb, sizeof(*mdata)); mdata->entropy = opa_vnic_calc_entropy(adapter, skb); mdata->vl = opa_vnic_get_vl(adapter, skb); rc = adapter->rn_ops->ndo_select_queue(netdev, skb, @@ -323,13 +323,13 @@ struct opa_vnic_adapter *opa_vnic_add_netdev(struct ib_device *ibdev, else if (IS_ERR(netdev)) return ERR_CAST(netdev); + rn = netdev_priv(netdev); adapter = kzalloc(sizeof(*adapter), GFP_KERNEL); if (!adapter) { rc = -ENOMEM; goto adapter_err; } - rn = netdev_priv(netdev); rn->clnt_priv = adapter; rn->hca = ibdev; rn->port_num = port_num; @@ -344,7 +344,7 @@ struct opa_vnic_adapter *opa_vnic_add_netdev(struct ib_device *ibdev, netdev->hard_header_len += OPA_VNIC_SKB_HEADROOM; mutex_init(&adapter->lock); mutex_init(&adapter->mactbl_lock); - mutex_init(&adapter->stats_lock); + spin_lock_init(&adapter->stats_lock); SET_NETDEV_DEV(netdev, ibdev->dev.parent); @@ -364,10 +364,9 @@ struct opa_vnic_adapter *opa_vnic_add_netdev(struct ib_device *ibdev, netdev_err: mutex_destroy(&adapter->lock); mutex_destroy(&adapter->mactbl_lock); - mutex_destroy(&adapter->stats_lock); kfree(adapter); adapter_err: - ibdev->free_rdma_netdev(netdev); + rn->free_rdma_netdev(netdev); return ERR_PTR(rc); } @@ -376,14 +375,13 @@ adapter_err: void opa_vnic_rem_netdev(struct opa_vnic_adapter *adapter) { struct net_device *netdev = adapter->netdev; - struct ib_device *ibdev = adapter->ibdev; + struct rdma_netdev *rn = netdev_priv(netdev); v_info("removing\n"); unregister_netdev(netdev); opa_vnic_release_mac_tbl(adapter); mutex_destroy(&adapter->lock); mutex_destroy(&adapter->mactbl_lock); - mutex_destroy(&adapter->stats_lock); kfree(adapter); - ibdev->free_rdma_netdev(netdev); + rn->free_rdma_netdev(netdev); } diff --git a/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c b/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c index 875694f9a7f9..cf768dd78d1b 100644 --- a/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c +++ b/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c @@ -794,7 +794,7 @@ void opa_vnic_vema_send_trap(struct opa_vnic_adapter *adapter, send_buf = ib_create_send_mad(port->mad_agent, 1, pkey_idx, 0, IB_MGMT_VENDOR_HDR, IB_MGMT_MAD_DATA, - GFP_KERNEL, OPA_MGMT_BASE_VERSION); + GFP_ATOMIC, OPA_MGMT_BASE_VERSION); if (IS_ERR(send_buf)) { c_err("%s:Couldn't allocate send buf\n", __func__); goto err_sndbuf; diff --git a/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema_iface.c b/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema_iface.c index a51bf977f4d6..c2733964379c 100644 --- a/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema_iface.c +++ b/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema_iface.c @@ -89,9 +89,9 @@ void opa_vnic_get_summary_counters(struct opa_vnic_adapter *adapter, u64 *src; memset(&vstats, 0, sizeof(vstats)); - mutex_lock(&adapter->stats_lock); + spin_lock(&adapter->stats_lock); adapter->rn_ops->ndo_get_stats64(adapter->netdev, &vstats.netstats); - mutex_unlock(&adapter->stats_lock); + spin_unlock(&adapter->stats_lock); cntrs->vp_instance = cpu_to_be16(adapter->vport_num); cntrs->vesw_id = cpu_to_be16(adapter->info.vesw.vesw_id); @@ -128,9 +128,9 @@ void opa_vnic_get_error_counters(struct opa_vnic_adapter *adapter, struct opa_vnic_stats vstats; memset(&vstats, 0, sizeof(vstats)); - mutex_lock(&adapter->stats_lock); + spin_lock(&adapter->stats_lock); adapter->rn_ops->ndo_get_stats64(adapter->netdev, &vstats.netstats); - mutex_unlock(&adapter->stats_lock); + spin_unlock(&adapter->stats_lock); cntrs->vp_instance = cpu_to_be16(adapter->vport_num); cntrs->vesw_id = cpu_to_be16(adapter->info.vesw.vesw_id); diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.c b/drivers/infiniband/ulp/srpt/ib_srpt.c index 1ced0731c140..402275be0931 100644 --- a/drivers/infiniband/ulp/srpt/ib_srpt.c +++ b/drivers/infiniband/ulp/srpt/ib_srpt.c @@ -1157,8 +1157,8 @@ static int srpt_abort_cmd(struct srpt_send_ioctx *ioctx) } spin_unlock_irqrestore(&ioctx->spinlock, flags); - pr_debug("Aborting cmd with state %d and tag %lld\n", state, - ioctx->cmd.tag); + pr_debug("Aborting cmd with state %d -> %d and tag %lld\n", state, + ioctx->state, ioctx->cmd.tag); switch (state) { case SRPT_STATE_NEW: |
