diff options
| author | Stephen Rothwell <sfr@canb.auug.org.au> | 2017-05-01 10:37:57 +1000 |
|---|---|---|
| committer | Stephen Rothwell <sfr@canb.auug.org.au> | 2017-05-01 10:37:57 +1000 |
| commit | 6b762260b8b89fed6187d1bbcdafcfd44d65cc1d (patch) | |
| tree | 99c96b304687843ba55a469a70854a912ead9588 | |
| parent | 1c85247fa78a168be76588ad99d048fec4069d11 (diff) | |
| parent | 2196f2716292c39b35c86f5238ed16a8978f9ad4 (diff) | |
Merge remote-tracking branch 'rdma/for-next'
Updates for 4.12 kernel merge window
- idr usage and locking changes
- build fix for hns
- ipoib debug path record file fix
- hfi1 updates
- core RDMA netdev addition
- Intel VNIC driver addition
- Enhanced allerators for IPoIB addition
- Debug cleanups in cxgb3/4
- Trivial cleanups from SF Markus Elfring
- Misc rxe fixes from Mellanox
- Misc ipoib fixes from Mellanox
- Lots of mlx4/mlx5 changes from Mellanox
- Misc fixes across the RDMA subsystem
- ODP paging fixes and improvements
- qedr updates
- hfi1 updates
- OPA port info patches
# gpg: Signature made Sat 29 Apr 2017 09:30:17 AEST
# gpg: using RSA key B826A3330E572FDD
# gpg: Can't check signature: No public key
190 files changed, 12222 insertions, 4221 deletions
diff --git a/Documentation/infiniband/opa_vnic.txt b/Documentation/infiniband/opa_vnic.txt new file mode 100644 index 000000000000..282e17be798a --- /dev/null +++ b/Documentation/infiniband/opa_vnic.txt @@ -0,0 +1,153 @@ +Intel Omni-Path (OPA) Virtual Network Interface Controller (VNIC) feature +supports Ethernet functionality over Omni-Path fabric by encapsulating +the Ethernet packets between HFI nodes. + +Architecture +============= +The patterns of exchanges of Omni-Path encapsulated Ethernet packets +involves one or more virtual Ethernet switches overlaid on the Omni-Path +fabric topology. A subset of HFI nodes on the Omni-Path fabric are +permitted to exchange encapsulated Ethernet packets across a particular +virtual Ethernet switch. The virtual Ethernet switches are logical +abstractions achieved by configuring the HFI nodes on the fabric for +header generation and processing. In the simplest configuration all HFI +nodes across the fabric exchange encapsulated Ethernet packets over a +single virtual Ethernet switch. A virtual Ethernet switch, is effectively +an independent Ethernet network. The configuration is performed by an +Ethernet Manager (EM) which is part of the trusted Fabric Manager (FM) +application. HFI nodes can have multiple VNICs each connected to a +different virtual Ethernet switch. The below diagram presents a case +of two virtual Ethernet switches with two HFI nodes. + + +-------------------+ + | Subnet/ | + | Ethernet | + | Manager | + +-------------------+ + / / + / / + / / + / / ++-----------------------------+ +------------------------------+ +| Virtual Ethernet Switch | | Virtual Ethernet Switch | +| +---------+ +---------+ | | +---------+ +---------+ | +| | VPORT | | VPORT | | | | VPORT | | VPORT | | ++--+---------+----+---------+-+ +-+---------+----+---------+---+ + | \ / | + | \ / | + | \/ | + | / \ | + | / \ | + +-----------+------------+ +-----------+------------+ + | VNIC | VNIC | | VNIC | VNIC | + +-----------+------------+ +-----------+------------+ + | HFI | | HFI | + +------------------------+ +------------------------+ + + +The Omni-Path encapsulated Ethernet packet format is as described below. + +Bits Field +------------------------------------ +Quad Word 0: +0-19 SLID (lower 20 bits) +20-30 Length (in Quad Words) +31 BECN bit +32-51 DLID (lower 20 bits) +52-56 SC (Service Class) +57-59 RC (Routing Control) +60 FECN bit +61-62 L2 (=10, 16B format) +63 LT (=1, Link Transfer Head Flit) + +Quad Word 1: +0-7 L4 type (=0x78 ETHERNET) +8-11 SLID[23:20] +12-15 DLID[23:20] +16-31 PKEY +32-47 Entropy +48-63 Reserved + +Quad Word 2: +0-15 Reserved +16-31 L4 header +32-63 Ethernet Packet + +Quad Words 3 to N-1: +0-63 Ethernet packet (pad extended) + +Quad Word N (last): +0-23 Ethernet packet (pad extended) +24-55 ICRC +56-61 Tail +62-63 LT (=01, Link Transfer Tail Flit) + +Ethernet packet is padded on the transmit side to ensure that the VNIC OPA +packet is quad word aligned. The 'Tail' field contains the number of bytes +padded. On the receive side the 'Tail' field is read and the padding is +removed (along with ICRC, Tail and OPA header) before passing packet up +the network stack. + +The L4 header field contains the virtual Ethernet switch id the VNIC port +belongs to. On the receive side, this field is used to de-multiplex the +received VNIC packets to different VNIC ports. + +Driver Design +============== +Intel OPA VNIC software design is presented in the below diagram. +OPA VNIC functionality has a HW dependent component and a HW +independent component. + +The support has been added for IB device to allocate and free the RDMA +netdev devices. The RDMA netdev supports interfacing with the network +stack thus creating standard network interfaces. OPA_VNIC is an RDMA +netdev device type. + +The HW dependent VNIC functionality is part of the HFI1 driver. It +implements the verbs to allocate and free the OPA_VNIC RDMA netdev. +It involves HW resource allocation/management for VNIC functionality. +It interfaces with the network stack and implements the required +net_device_ops functions. It expects Omni-Path encapsulated Ethernet +packets in the transmit path and provides HW access to them. It strips +the Omni-Path header from the received packets before passing them up +the network stack. It also implements the RDMA netdev control operations. + +The OPA VNIC module implements the HW independent VNIC functionality. +It consists of two parts. The VNIC Ethernet Management Agent (VEMA) +registers itself with IB core as an IB client and interfaces with the +IB MAD stack. It exchanges the management information with the Ethernet +Manager (EM) and the VNIC netdev. The VNIC netdev part allocates and frees +the OPA_VNIC RDMA netdev devices. It overrides the net_device_ops functions +set by HW dependent VNIC driver where required to accommodate any control +operation. It also handles the encapsulation of Ethernet packets with an +Omni-Path header in the transmit path. For each VNIC interface, the +information required for encapsulation is configured by the EM via VEMA MAD +interface. It also passes any control information to the HW dependent driver +by invoking the RDMA netdev control operations. + + +-------------------+ +----------------------+ + | | | Linux | + | IB MAD | | Network | + | | | Stack | + +-------------------+ +----------------------+ + | | | + | | | + +----------------------------+ | + | | | + | OPA VNIC Module | | + | (OPA VNIC RDMA Netdev | | + | & EMA functions) | | + | | | + +----------------------------+ | + | | + | | + +------------------+ | + | IB core | | + +------------------+ | + | | + | | + +--------------------------------------------+ + | | + | HFI1 Driver with VNIC support | + | | + +--------------------------------------------+ diff --git a/MAINTAINERS b/MAINTAINERS index 7a9fd64fec3b..ff8da543faa2 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5899,6 +5899,13 @@ F: drivers/block/cciss* F: include/linux/cciss_ioctl.h F: include/uapi/linux/cciss_ioctl.h +OPA-VNIC DRIVER +M: Dennis Dalessandro <dennis.dalessandro@intel.com> +M: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com> +L: linux-rdma@vger.kernel.org +S: Supported +F: drivers/infiniband/ulp/opa_vnic + HFI1 DRIVER M: Mike Marciniszyn <mike.marciniszyn@intel.com> M: Dennis Dalessandro <dennis.dalessandro@intel.com> @@ -6507,6 +6514,7 @@ W: http://www.openfabrics.org/ Q: http://patchwork.kernel.org/project/linux-rdma/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma.git S: Supported +F: Documentation/devicetree/bindings/infiniband/ F: Documentation/infiniband/ F: drivers/infiniband/ F: include/uapi/linux/if_infiniband.h @@ -11501,11 +11509,11 @@ S: Supported F: drivers/net/ethernet/emulex/benet/ EMULEX ONECONNECT ROCE DRIVER -M: Selvin Xavier <selvin.xavier@avagotech.com> -M: Devesh Sharma <devesh.sharma@avagotech.com> +M: Selvin Xavier <selvin.xavier@broadcom.com> +M: Devesh Sharma <devesh.sharma@broadcom.com> L: linux-rdma@vger.kernel.org -W: http://www.emulex.com -S: Supported +W: http://www.broadcom.com +S: Odd Fixes F: drivers/infiniband/hw/ocrdma/ F: include/uapi/rdma/ocrdma-abi.h diff --git a/drivers/infiniband/Kconfig b/drivers/infiniband/Kconfig index 66f86027ed47..234fe01904e7 100644 --- a/drivers/infiniband/Kconfig +++ b/drivers/infiniband/Kconfig @@ -85,6 +85,7 @@ source "drivers/infiniband/ulp/srpt/Kconfig" source "drivers/infiniband/ulp/iser/Kconfig" source "drivers/infiniband/ulp/isert/Kconfig" +source "drivers/infiniband/ulp/opa_vnic/Kconfig" source "drivers/infiniband/sw/rdmavt/Kconfig" source "drivers/infiniband/sw/rxe/Kconfig" diff --git a/drivers/infiniband/core/Makefile b/drivers/infiniband/core/Makefile index e426ac877d19..6ebd9ad95010 100644 --- a/drivers/infiniband/core/Makefile +++ b/drivers/infiniband/core/Makefile @@ -29,4 +29,5 @@ ib_umad-y := user_mad.o ib_ucm-y := ucm.o -ib_uverbs-y := uverbs_main.o uverbs_cmd.o uverbs_marshall.o +ib_uverbs-y := uverbs_main.o uverbs_cmd.o uverbs_marshall.o \ + rdma_core.o uverbs_std_types.o diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c index 329d08c884f6..523d24320100 100644 --- a/drivers/infiniband/core/addr.c +++ b/drivers/infiniband/core/addr.c @@ -444,8 +444,8 @@ static int addr6_resolve(struct sockaddr_in6 *src_in, fl6.saddr = src_in->sin6_addr; fl6.flowi6_oif = addr->bound_dev_if; - dst = ip6_route_output(addr->net, NULL, &fl6); - if ((ret = dst->error)) + ret = ipv6_stub->ipv6_dst_lookup(addr->net, NULL, &dst, &fl6); + if (ret < 0) goto put; rt = (struct rt6_info *)dst; diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index acd10d666f1c..34dc81d30692 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -1755,6 +1755,9 @@ static int cma_ib_handler(struct ib_cm_id *cm_id, struct ib_cm_event *ib_event) event.status = -ETIMEDOUT; break; case IB_CM_REP_RECEIVED: + if (cma_comp(id_priv, RDMA_CM_CONNECT) && + (id_priv->id.qp_type != IB_QPT_UD)) + ib_send_cm_mra(cm_id, CMA_CM_MRA_SETTING, NULL, 0); if (id_priv->id.qp) { event.status = cma_rep_recv(id_priv); event.event = event.status ? RDMA_CM_EVENT_CONNECT_ERROR : @@ -3941,63 +3944,10 @@ static void cma_set_mgid(struct rdma_id_private *id_priv, } } -static void cma_query_sa_classport_info_cb(int status, - struct ib_class_port_info *rec, - void *context) -{ - struct class_port_info_context *cb_ctx = context; - - WARN_ON(!context); - - if (status || !rec) { - pr_debug("RDMA CM: %s port %u failed query ClassPortInfo status: %d\n", - cb_ctx->device->name, cb_ctx->port_num, status); - goto out; - } - - memcpy(cb_ctx->class_port_info, rec, sizeof(struct ib_class_port_info)); - -out: - complete(&cb_ctx->done); -} - -static int cma_query_sa_classport_info(struct ib_device *device, u8 port_num, - struct ib_class_port_info *class_port_info) -{ - struct class_port_info_context *cb_ctx; - int ret; - - cb_ctx = kmalloc(sizeof(*cb_ctx), GFP_KERNEL); - if (!cb_ctx) - return -ENOMEM; - - cb_ctx->device = device; - cb_ctx->class_port_info = class_port_info; - cb_ctx->port_num = port_num; - init_completion(&cb_ctx->done); - - ret = ib_sa_classport_info_rec_query(&sa_client, device, port_num, - CMA_QUERY_CLASSPORT_INFO_TIMEOUT, - GFP_KERNEL, cma_query_sa_classport_info_cb, - cb_ctx, &cb_ctx->sa_query); - if (ret < 0) { - pr_err("RDMA CM: %s port %u failed to send ClassPortInfo query, ret: %d\n", - device->name, port_num, ret); - goto out; - } - - wait_for_completion(&cb_ctx->done); - -out: - kfree(cb_ctx); - return ret; -} - static int cma_join_ib_multicast(struct rdma_id_private *id_priv, struct cma_multicast *mc) { struct ib_sa_mcmember_rec rec; - struct ib_class_port_info class_port_info; struct rdma_dev_addr *dev_addr = &id_priv->id.route.addr.dev_addr; ib_sa_comp_mask comp_mask; int ret; @@ -4018,21 +3968,14 @@ static int cma_join_ib_multicast(struct rdma_id_private *id_priv, rec.pkey = cpu_to_be16(ib_addr_get_pkey(dev_addr)); rec.join_state = mc->join_state; - if (rec.join_state == BIT(SENDONLY_FULLMEMBER_JOIN)) { - ret = cma_query_sa_classport_info(id_priv->id.device, - id_priv->id.port_num, - &class_port_info); - - if (ret) - return ret; - - if (!(ib_get_cpi_capmask2(&class_port_info) & - IB_SA_CAP_MASK2_SENDONLY_FULL_MEM_SUPPORT)) { - pr_warn("RDMA CM: %s port %u Unable to multicast join\n" - "RDMA CM: SM doesn't support Send Only Full Member option\n", - id_priv->id.device->name, id_priv->id.port_num); - return -EOPNOTSUPP; - } + if ((rec.join_state == BIT(SENDONLY_FULLMEMBER_JOIN)) && + (!ib_sa_sendonly_fullmem_support(&sa_client, + id_priv->id.device, + id_priv->id.port_num))) { + pr_warn("RDMA CM: %s port %u Unable to multicast join\n" + "RDMA CM: SM doesn't support Send Only Full Member option\n", + id_priv->id.device->name, id_priv->id.port_num); + return -EOPNOTSUPP; } comp_mask = IB_SA_MCMEMBER_REC_MGID | IB_SA_MCMEMBER_REC_PORT_GID | diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c index 7c9e34d679d3..81d447da0048 100644 --- a/drivers/infiniband/core/device.c +++ b/drivers/infiniband/core/device.c @@ -172,8 +172,16 @@ static void ib_device_release(struct device *device) { struct ib_device *dev = container_of(device, struct ib_device, dev); - ib_cache_release_one(dev); - kfree(dev->port_immutable); + WARN_ON(dev->reg_state == IB_DEV_REGISTERED); + if (dev->reg_state == IB_DEV_UNREGISTERED) { + /* + * In IB_DEV_UNINITIALIZED state, cache or port table + * is not even created. Free cache and port table only when + * device reaches UNREGISTERED state. + */ + ib_cache_release_one(dev); + kfree(dev->port_immutable); + } kfree(dev); } @@ -380,32 +388,27 @@ int ib_register_device(struct ib_device *device, ret = ib_cache_setup_one(device); if (ret) { pr_warn("Couldn't set up InfiniBand P_Key/GID cache\n"); - goto out; + goto port_cleanup; } ret = ib_device_register_rdmacg(device); if (ret) { pr_warn("Couldn't register device with rdma cgroup\n"); - ib_cache_cleanup_one(device); - goto out; + goto cache_cleanup; } memset(&device->attrs, 0, sizeof(device->attrs)); ret = device->query_device(device, &device->attrs, &uhw); if (ret) { pr_warn("Couldn't query the device attributes\n"); - ib_device_unregister_rdmacg(device); - ib_cache_cleanup_one(device); - goto out; + goto cache_cleanup; } ret = ib_device_register_sysfs(device, port_callback); if (ret) { pr_warn("Couldn't register device %s with driver model\n", device->name); - ib_device_unregister_rdmacg(device); - ib_cache_cleanup_one(device); - goto out; + goto cache_cleanup; } device->reg_state = IB_DEV_REGISTERED; @@ -417,6 +420,14 @@ int ib_register_device(struct ib_device *device, down_write(&lists_rwsem); list_add_tail(&device->core_list, &device_list); up_write(&lists_rwsem); + mutex_unlock(&device_mutex); + return 0; + +cache_cleanup: + ib_cache_cleanup_one(device); + ib_cache_release_one(device); +port_cleanup: + kfree(device->port_immutable); out: mutex_unlock(&device_mutex); return ret; diff --git a/drivers/infiniband/core/fmr_pool.c b/drivers/infiniband/core/fmr_pool.c index cdfad5f26212..84d2615b5d4b 100644 --- a/drivers/infiniband/core/fmr_pool.c +++ b/drivers/infiniband/core/fmr_pool.c @@ -96,7 +96,8 @@ struct ib_fmr_pool { void * arg); void *flush_arg; - struct task_struct *thread; + struct kthread_worker *worker; + struct kthread_work work; atomic_t req_ser; atomic_t flush_ser; @@ -174,29 +175,19 @@ static void ib_fmr_batch_release(struct ib_fmr_pool *pool) spin_unlock_irq(&pool->pool_lock); } -static int ib_fmr_cleanup_thread(void *pool_ptr) +static void ib_fmr_cleanup_func(struct kthread_work *work) { - struct ib_fmr_pool *pool = pool_ptr; + struct ib_fmr_pool *pool = container_of(work, struct ib_fmr_pool, work); - do { - if (atomic_read(&pool->flush_ser) - atomic_read(&pool->req_ser) < 0) { - ib_fmr_batch_release(pool); - - atomic_inc(&pool->flush_ser); - wake_up_interruptible(&pool->force_wait); - - if (pool->flush_function) - pool->flush_function(pool, pool->flush_arg); - } + ib_fmr_batch_release(pool); + atomic_inc(&pool->flush_ser); + wake_up_interruptible(&pool->force_wait); - set_current_state(TASK_INTERRUPTIBLE); - if (atomic_read(&pool->flush_ser) - atomic_read(&pool->req_ser) >= 0 && - !kthread_should_stop()) - schedule(); - __set_current_state(TASK_RUNNING); - } while (!kthread_should_stop()); + if (pool->flush_function) + pool->flush_function(pool, pool->flush_arg); - return 0; + if (atomic_read(&pool->flush_ser) - atomic_read(&pool->req_ser) < 0) + kthread_queue_work(pool->worker, &pool->work); } /** @@ -265,15 +256,13 @@ struct ib_fmr_pool *ib_create_fmr_pool(struct ib_pd *pd, atomic_set(&pool->flush_ser, 0); init_waitqueue_head(&pool->force_wait); - pool->thread = kthread_run(ib_fmr_cleanup_thread, - pool, - "ib_fmr(%s)", - device->name); - if (IS_ERR(pool->thread)) { - pr_warn(PFX "couldn't start cleanup thread\n"); - ret = PTR_ERR(pool->thread); + pool->worker = kthread_create_worker(0, "ib_fmr(%s)", device->name); + if (IS_ERR(pool->worker)) { + pr_warn(PFX "couldn't start cleanup kthread worker\n"); + ret = PTR_ERR(pool->worker); goto out_free_pool; } + kthread_init_work(&pool->work, ib_fmr_cleanup_func); { struct ib_pool_fmr *fmr; @@ -338,7 +327,7 @@ void ib_destroy_fmr_pool(struct ib_fmr_pool *pool) LIST_HEAD(fmr_list); int i; - kthread_stop(pool->thread); + kthread_destroy_worker(pool->worker); ib_fmr_batch_release(pool); i = 0; @@ -388,7 +377,7 @@ int ib_flush_fmr_pool(struct ib_fmr_pool *pool) spin_unlock_irq(&pool->pool_lock); serial = atomic_inc_return(&pool->req_ser); - wake_up_process(pool->thread); + kthread_queue_work(pool->worker, &pool->work); if (wait_event_interruptible(pool->force_wait, atomic_read(&pool->flush_ser) - serial >= 0)) @@ -502,7 +491,7 @@ int ib_fmr_pool_unmap(struct ib_pool_fmr *fmr) list_add_tail(&fmr->list, &pool->dirty_list); if (++pool->dirty_len >= pool->dirty_watermark) { atomic_inc(&pool->req_ser); - wake_up_process(pool->thread); + kthread_queue_work(pool->worker, &pool->work); } } } diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c index 57f231f1c721..8a2ceb4ddaa2 100644 --- a/drivers/infiniband/core/mad.c +++ b/drivers/infiniband/core/mad.c @@ -605,7 +605,7 @@ static void unregister_mad_snoop(struct ib_mad_snoop_private *mad_snoop_priv) /* * ib_unregister_mad_agent - Unregisters a client from using MAD services */ -int ib_unregister_mad_agent(struct ib_mad_agent *mad_agent) +void ib_unregister_mad_agent(struct ib_mad_agent *mad_agent) { struct ib_mad_agent_private *mad_agent_priv; struct ib_mad_snoop_private *mad_snoop_priv; @@ -622,7 +622,6 @@ int ib_unregister_mad_agent(struct ib_mad_agent *mad_agent) agent); unregister_mad_snoop(mad_snoop_priv); } - return 0; } EXPORT_SYMBOL(ib_unregister_mad_agent); diff --git a/drivers/infiniband/core/rdma_core.c b/drivers/infiniband/core/rdma_core.c new file mode 100644 index 000000000000..41c31a2bf093 --- /dev/null +++ b/drivers/infiniband/core/rdma_core.c @@ -0,0 +1,627 @@ +/* + * Copyright (c) 2016, Mellanox Technologies inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include <linux/file.h> +#include <linux/anon_inodes.h> +#include <rdma/ib_verbs.h> +#include <rdma/uverbs_types.h> +#include <linux/rcupdate.h> +#include "uverbs.h" +#include "core_priv.h" +#include "rdma_core.h" + +void uverbs_uobject_get(struct ib_uobject *uobject) +{ + kref_get(&uobject->ref); +} + +static void uverbs_uobject_free(struct kref *ref) +{ + struct ib_uobject *uobj = + container_of(ref, struct ib_uobject, ref); + + if (uobj->type->type_class->needs_kfree_rcu) + kfree_rcu(uobj, rcu); + else + kfree(uobj); +} + +void uverbs_uobject_put(struct ib_uobject *uobject) +{ + kref_put(&uobject->ref, uverbs_uobject_free); +} + +static int uverbs_try_lock_object(struct ib_uobject *uobj, bool exclusive) +{ + /* + * When a shared access is required, we use a positive counter. Each + * shared access request checks that the value != -1 and increment it. + * Exclusive access is required for operations like write or destroy. + * In exclusive access mode, we check that the counter is zero (nobody + * claimed this object) and we set it to -1. Releasing a shared access + * lock is done simply by decreasing the counter. As for exclusive + * access locks, since only a single one of them is is allowed + * concurrently, setting the counter to zero is enough for releasing + * this lock. + */ + if (!exclusive) + return __atomic_add_unless(&uobj->usecnt, 1, -1) == -1 ? + -EBUSY : 0; + + /* lock is either WRITE or DESTROY - should be exclusive */ + return atomic_cmpxchg(&uobj->usecnt, 0, -1) == 0 ? 0 : -EBUSY; +} + +static struct ib_uobject *alloc_uobj(struct ib_ucontext *context, + const struct uverbs_obj_type *type) +{ + struct ib_uobject *uobj = kzalloc(type->obj_size, GFP_KERNEL); + + if (!uobj) + return ERR_PTR(-ENOMEM); + /* + * user_handle should be filled by the handler, + * The object is added to the list in the commit stage. + */ + uobj->context = context; + uobj->type = type; + atomic_set(&uobj->usecnt, 0); + kref_init(&uobj->ref); + + return uobj; +} + +static int idr_add_uobj(struct ib_uobject *uobj) +{ + int ret; + + idr_preload(GFP_KERNEL); + spin_lock(&uobj->context->ufile->idr_lock); + + /* + * We start with allocating an idr pointing to NULL. This represents an + * object which isn't initialized yet. We'll replace it later on with + * the real object once we commit. + */ + ret = idr_alloc(&uobj->context->ufile->idr, NULL, 0, + min_t(unsigned long, U32_MAX - 1, INT_MAX), GFP_NOWAIT); + if (ret >= 0) + uobj->id = ret; + + spin_unlock(&uobj->context->ufile->idr_lock); + idr_preload_end(); + + return ret < 0 ? ret : 0; +} + +/* + * It only removes it from the uobjects list, uverbs_uobject_put() is still + * required. + */ +static void uverbs_idr_remove_uobj(struct ib_uobject *uobj) +{ + spin_lock(&uobj->context->ufile->idr_lock); + idr_remove(&uobj->context->ufile->idr, uobj->id); + spin_unlock(&uobj->context->ufile->idr_lock); +} + +/* Returns the ib_uobject or an error. The caller should check for IS_ERR. */ +static struct ib_uobject *lookup_get_idr_uobject(const struct uverbs_obj_type *type, + struct ib_ucontext *ucontext, + int id, bool exclusive) +{ + struct ib_uobject *uobj; + + rcu_read_lock(); + /* object won't be released as we're protected in rcu */ + uobj = idr_find(&ucontext->ufile->idr, id); + if (!uobj) { + uobj = ERR_PTR(-ENOENT); + goto free; + } + + uverbs_uobject_get(uobj); +free: + rcu_read_unlock(); + return uobj; +} + +static struct ib_uobject *lookup_get_fd_uobject(const struct uverbs_obj_type *type, + struct ib_ucontext *ucontext, + int id, bool exclusive) +{ + struct file *f; + struct ib_uobject *uobject; + const struct uverbs_obj_fd_type *fd_type = + container_of(type, struct uverbs_obj_fd_type, type); + + if (exclusive) + return ERR_PTR(-EOPNOTSUPP); + + f = fget(id); + if (!f) + return ERR_PTR(-EBADF); + + uobject = f->private_data; + /* + * fget(id) ensures we are not currently running uverbs_close_fd, + * and the caller is expected to ensure that uverbs_close_fd is never + * done while a call top lookup is possible. + */ + if (f->f_op != fd_type->fops) { + fput(f); + return ERR_PTR(-EBADF); + } + + uverbs_uobject_get(uobject); + return uobject; +} + +struct ib_uobject *rdma_lookup_get_uobject(const struct uverbs_obj_type *type, + struct ib_ucontext *ucontext, + int id, bool exclusive) +{ + struct ib_uobject *uobj; + int ret; + + uobj = type->type_class->lookup_get(type, ucontext, id, exclusive); + if (IS_ERR(uobj)) + return uobj; + + if (uobj->type != type) { + ret = -EINVAL; + goto free; + } + + ret = uverbs_try_lock_object(uobj, exclusive); + if (ret) { + WARN(ucontext->cleanup_reason, + "ib_uverbs: Trying to lookup_get while cleanup context\n"); + goto free; + } + + return uobj; +free: + uobj->type->type_class->lookup_put(uobj, exclusive); + uverbs_uobject_put(uobj); + return ERR_PTR(ret); +} + +static struct ib_uobject *alloc_begin_idr_uobject(const struct uverbs_obj_type *type, + struct ib_ucontext *ucontext) +{ + int ret; + struct ib_uobject *uobj; + + uobj = alloc_uobj(ucontext, type); + if (IS_ERR(uobj)) + return uobj; + + ret = idr_add_uobj(uobj); + if (ret) + goto uobj_put; + + ret = ib_rdmacg_try_charge(&uobj->cg_obj, ucontext->device, + RDMACG_RESOURCE_HCA_OBJECT); + if (ret) + goto idr_remove; + + return uobj; + +idr_remove: + uverbs_idr_remove_uobj(uobj); +uobj_put: + uverbs_uobject_put(uobj); + return ERR_PTR(ret); +} + +static struct ib_uobject *alloc_begin_fd_uobject(const struct uverbs_obj_type *type, + struct ib_ucontext *ucontext) +{ + const struct uverbs_obj_fd_type *fd_type = + container_of(type, struct uverbs_obj_fd_type, type); + int new_fd; + struct ib_uobject *uobj; + struct ib_uobject_file *uobj_file; + struct file *filp; + + new_fd = get_unused_fd_flags(O_CLOEXEC); + if (new_fd < 0) + return ERR_PTR(new_fd); + + uobj = alloc_uobj(ucontext, type); + if (IS_ERR(uobj)) { + put_unused_fd(new_fd); + return uobj; + } + + uobj_file = container_of(uobj, struct ib_uobject_file, uobj); + filp = anon_inode_getfile(fd_type->name, + fd_type->fops, + uobj_file, + fd_type->flags); + if (IS_ERR(filp)) { + put_unused_fd(new_fd); + uverbs_uobject_put(uobj); + return (void *)filp; + } + + uobj_file->uobj.id = new_fd; + uobj_file->uobj.object = filp; + uobj_file->ufile = ucontext->ufile; + INIT_LIST_HEAD(&uobj->list); + kref_get(&uobj_file->ufile->ref); + + return uobj; +} + +struct ib_uobject *rdma_alloc_begin_uobject(const struct uverbs_obj_type *type, + struct ib_ucontext *ucontext) +{ + return type->type_class->alloc_begin(type, ucontext); +} + +static void uverbs_uobject_add(struct ib_uobject *uobject) +{ + mutex_lock(&uobject->context->uobjects_lock); + list_add(&uobject->list, &uobject->context->uobjects); + mutex_unlock(&uobject->context->uobjects_lock); +} + +static int __must_check remove_commit_idr_uobject(struct ib_uobject *uobj, + enum rdma_remove_reason why) +{ + const struct uverbs_obj_idr_type *idr_type = + container_of(uobj->type, struct uverbs_obj_idr_type, + type); + int ret = idr_type->destroy_object(uobj, why); + + /* + * We can only fail gracefully if the user requested to destroy the + * object. In the rest of the cases, just remove whatever you can. + */ + if (why == RDMA_REMOVE_DESTROY && ret) + return ret; + + ib_rdmacg_uncharge(&uobj->cg_obj, uobj->context->device, + RDMACG_RESOURCE_HCA_OBJECT); + uverbs_idr_remove_uobj(uobj); + + return ret; +} + +static void alloc_abort_fd_uobject(struct ib_uobject *uobj) +{ + struct ib_uobject_file *uobj_file = + container_of(uobj, struct ib_uobject_file, uobj); + struct file *filp = uobj->object; + int id = uobj_file->uobj.id; + + /* Unsuccessful NEW */ + fput(filp); + put_unused_fd(id); +} + +static int __must_check remove_commit_fd_uobject(struct ib_uobject *uobj, + enum rdma_remove_reason why) +{ + const struct uverbs_obj_fd_type *fd_type = + container_of(uobj->type, struct uverbs_obj_fd_type, type); + struct ib_uobject_file *uobj_file = + container_of(uobj, struct ib_uobject_file, uobj); + int ret = fd_type->context_closed(uobj_file, why); + + if (why == RDMA_REMOVE_DESTROY && ret) + return ret; + + if (why == RDMA_REMOVE_DURING_CLEANUP) { + alloc_abort_fd_uobject(uobj); + return ret; + } + + uobj_file->uobj.context = NULL; + return ret; +} + +static void lockdep_check(struct ib_uobject *uobj, bool exclusive) +{ +#ifdef CONFIG_LOCKDEP + if (exclusive) + WARN_ON(atomic_read(&uobj->usecnt) > 0); + else + WARN_ON(atomic_read(&uobj->usecnt) == -1); +#endif +} + +static int __must_check _rdma_remove_commit_uobject(struct ib_uobject *uobj, + enum rdma_remove_reason why) +{ + int ret; + struct ib_ucontext *ucontext = uobj->context; + + ret = uobj->type->type_class->remove_commit(uobj, why); + if (ret && why == RDMA_REMOVE_DESTROY) { + /* We couldn't remove the object, so just unlock the uobject */ + atomic_set(&uobj->usecnt, 0); + uobj->type->type_class->lookup_put(uobj, true); + } else { + mutex_lock(&ucontext->uobjects_lock); + list_del(&uobj->list); + mutex_unlock(&ucontext->uobjects_lock); + /* put the ref we took when we created the object */ + uverbs_uobject_put(uobj); + } + + return ret; +} + +/* This is called only for user requested DESTROY reasons */ +int __must_check rdma_remove_commit_uobject(struct ib_uobject *uobj) +{ + int ret; + struct ib_ucontext *ucontext = uobj->context; + + /* put the ref count we took at lookup_get */ + uverbs_uobject_put(uobj); + /* Cleanup is running. Calling this should have been impossible */ + if (!down_read_trylock(&ucontext->cleanup_rwsem)) { + WARN(true, "ib_uverbs: Cleanup is running while removing an uobject\n"); + return 0; + } + lockdep_check(uobj, true); + ret = _rdma_remove_commit_uobject(uobj, RDMA_REMOVE_DESTROY); + + up_read(&ucontext->cleanup_rwsem); + return ret; +} + +static void alloc_commit_idr_uobject(struct ib_uobject *uobj) +{ + uverbs_uobject_add(uobj); + spin_lock(&uobj->context->ufile->idr_lock); + /* + * We already allocated this IDR with a NULL object, so + * this shouldn't fail. + */ + WARN_ON(idr_replace(&uobj->context->ufile->idr, + uobj, uobj->id)); + spin_unlock(&uobj->context->ufile->idr_lock); +} + +static void alloc_commit_fd_uobject(struct ib_uobject *uobj) +{ + struct ib_uobject_file *uobj_file = + container_of(uobj, struct ib_uobject_file, uobj); + + uverbs_uobject_add(&uobj_file->uobj); + fd_install(uobj_file->uobj.id, uobj->object); + /* This shouldn't be used anymore. Use the file object instead */ + uobj_file->uobj.id = 0; + /* Get another reference as we export this to the fops */ + uverbs_uobject_get(&uobj_file->uobj); +} + +int rdma_alloc_commit_uobject(struct ib_uobject *uobj) +{ + /* Cleanup is running. Calling this should have been impossible */ + if (!down_read_trylock(&uobj->context->cleanup_rwsem)) { + int ret; + + WARN(true, "ib_uverbs: Cleanup is running while allocating an uobject\n"); + ret = uobj->type->type_class->remove_commit(uobj, + RDMA_REMOVE_DURING_CLEANUP); + if (ret) + pr_warn("ib_uverbs: cleanup of idr object %d failed\n", + uobj->id); + return ret; + } + + uobj->type->type_class->alloc_commit(uobj); + up_read(&uobj->context->cleanup_rwsem); + + return 0; +} + +static void alloc_abort_idr_uobject(struct ib_uobject *uobj) +{ + uverbs_idr_remove_uobj(uobj); + ib_rdmacg_uncharge(&uobj->cg_obj, uobj->context->device, + RDMACG_RESOURCE_HCA_OBJECT); + uverbs_uobject_put(uobj); +} + +void rdma_alloc_abort_uobject(struct ib_uobject *uobj) +{ + uobj->type->type_class->alloc_abort(uobj); +} + +static void lookup_put_idr_uobject(struct ib_uobject *uobj, bool exclusive) +{ +} + +static void lookup_put_fd_uobject(struct ib_uobject *uobj, bool exclusive) +{ + struct file *filp = uobj->object; + + WARN_ON(exclusive); + /* This indirectly calls uverbs_close_fd and free the object */ + fput(filp); +} + +void rdma_lookup_put_uobject(struct ib_uobject *uobj, bool exclusive) +{ + lockdep_check(uobj, exclusive); + uobj->type->type_class->lookup_put(uobj, exclusive); + /* + * In order to unlock an object, either decrease its usecnt for + * read access or zero it in case of exclusive access. See + * uverbs_try_lock_object for locking schema information. + */ + if (!exclusive) + atomic_dec(&uobj->usecnt); + else + atomic_set(&uobj->usecnt, 0); + + uverbs_uobject_put(uobj); +} + +const struct uverbs_obj_type_class uverbs_idr_class = { + .alloc_begin = alloc_begin_idr_uobject, + .lookup_get = lookup_get_idr_uobject, + .alloc_commit = alloc_commit_idr_uobject, + .alloc_abort = alloc_abort_idr_uobject, + .lookup_put = lookup_put_idr_uobject, + .remove_commit = remove_commit_idr_uobject, + /* + * When we destroy an object, we first just lock it for WRITE and + * actually DESTROY it in the finalize stage. So, the problematic + * scenario is when we just started the finalize stage of the + * destruction (nothing was executed yet). Now, the other thread + * fetched the object for READ access, but it didn't lock it yet. + * The DESTROY thread continues and starts destroying the object. + * When the other thread continue - without the RCU, it would + * access freed memory. However, the rcu_read_lock delays the free + * until the rcu_read_lock of the READ operation quits. Since the + * exclusive lock of the object is still taken by the DESTROY flow, the + * READ operation will get -EBUSY and it'll just bail out. + */ + .needs_kfree_rcu = true, +}; + +static void _uverbs_close_fd(struct ib_uobject_file *uobj_file) +{ + struct ib_ucontext *ucontext; + struct ib_uverbs_file *ufile = uobj_file->ufile; + int ret; + + mutex_lock(&uobj_file->ufile->cleanup_mutex); + + /* uobject was either already cleaned up or is cleaned up right now anyway */ + if (!uobj_file->uobj.context || + !down_read_trylock(&uobj_file->uobj.context->cleanup_rwsem)) + goto unlock; + + ucontext = uobj_file->uobj.context; + ret = _rdma_remove_commit_uobject(&uobj_file->uobj, RDMA_REMOVE_CLOSE); + up_read(&ucontext->cleanup_rwsem); + if (ret) + pr_warn("uverbs: unable to clean up uobject file in uverbs_close_fd.\n"); +unlock: + mutex_unlock(&ufile->cleanup_mutex); +} + +void uverbs_close_fd(struct file *f) +{ + struct ib_uobject_file *uobj_file = f->private_data; + struct kref *uverbs_file_ref = &uobj_file->ufile->ref; + + _uverbs_close_fd(uobj_file); + uverbs_uobject_put(&uobj_file->uobj); + kref_put(uverbs_file_ref, ib_uverbs_release_file); +} + +void uverbs_cleanup_ucontext(struct ib_ucontext *ucontext, bool device_removed) +{ + enum rdma_remove_reason reason = device_removed ? + RDMA_REMOVE_DRIVER_REMOVE : RDMA_REMOVE_CLOSE; + unsigned int cur_order = 0; + + ucontext->cleanup_reason = reason; + /* + * Waits for all remove_commit and alloc_commit to finish. Logically, We + * want to hold this forever as the context is going to be destroyed, + * but we'll release it since it causes a "held lock freed" BUG message. + */ + down_write(&ucontext->cleanup_rwsem); + + while (!list_empty(&ucontext->uobjects)) { + struct ib_uobject *obj, *next_obj; + unsigned int next_order = UINT_MAX; + + /* + * This shouldn't run while executing other commands on this + * context. Thus, the only thing we should take care of is + * releasing a FD while traversing this list. The FD could be + * closed and released from the _release fop of this FD. + * In order to mitigate this, we add a lock. + * We take and release the lock per order traversal in order + * to let other threads (which might still use the FDs) chance + * to run. + */ + mutex_lock(&ucontext->uobjects_lock); + list_for_each_entry_safe(obj, next_obj, &ucontext->uobjects, + list) { + if (obj->type->destroy_order == cur_order) { + int ret; + + /* + * if we hit this WARN_ON, that means we are + * racing with a lookup_get. + */ + WARN_ON(uverbs_try_lock_object(obj, true)); + ret = obj->type->type_class->remove_commit(obj, + reason); + list_del(&obj->list); + if (ret) + pr_warn("ib_uverbs: failed to remove uobject id %d order %u\n", + obj->id, cur_order); + /* put the ref we took when we created the object */ + uverbs_uobject_put(obj); + } else { + next_order = min(next_order, + obj->type->destroy_order); + } + } + mutex_unlock(&ucontext->uobjects_lock); + cur_order = next_order; + } + up_write(&ucontext->cleanup_rwsem); +} + +void uverbs_initialize_ucontext(struct ib_ucontext *ucontext) +{ + ucontext->cleanup_reason = 0; + mutex_init(&ucontext->uobjects_lock); + INIT_LIST_HEAD(&ucontext->uobjects); + init_rwsem(&ucontext->cleanup_rwsem); +} + +const struct uverbs_obj_type_class uverbs_fd_class = { + .alloc_begin = alloc_begin_fd_uobject, + .lookup_get = lookup_get_fd_uobject, + .alloc_commit = alloc_commit_fd_uobject, + .alloc_abort = alloc_abort_fd_uobject, + .lookup_put = lookup_put_fd_uobject, + .remove_commit = remove_commit_fd_uobject, + .needs_kfree_rcu = false, +}; + diff --git a/drivers/infiniband/core/rdma_core.h b/drivers/infiniband/core/rdma_core.h new file mode 100644 index 000000000000..1b82e7ff7fe8 --- /dev/null +++ b/drivers/infiniband/core/rdma_core.h @@ -0,0 +1,78 @@ +/* + * Copyright (c) 2005 Topspin Communications. All rights reserved. + * Copyright (c) 2005, 2006 Cisco Systems. All rights reserved. + * Copyright (c) 2005-2017 Mellanox Technologies. All rights reserved. + * Copyright (c) 2005 Voltaire, Inc. All rights reserved. + * Copyright (c) 2005 PathScale, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef RDMA_CORE_H +#define RDMA_CORE_H + +#include <linux/idr.h> +#include <rdma/uverbs_types.h> +#include <rdma/ib_verbs.h> +#include <linux/mutex.h> + +/* + * These functions initialize the context and cleanups its uobjects. + * The context has a list of objects which is protected by a mutex + * on the context. initialize_ucontext should be called when we create + * a context. + * cleanup_ucontext removes all uobjects from the context and puts them. + */ +void uverbs_cleanup_ucontext(struct ib_ucontext *ucontext, bool device_removed); +void uverbs_initialize_ucontext(struct ib_ucontext *ucontext); + +/* + * uverbs_uobject_get is called in order to increase the reference count on + * an uobject. This is useful when a handler wants to keep the uobject's memory + * alive, regardless if this uobject is still alive in the context's objects + * repository. Objects are put via uverbs_uobject_put. + */ +void uverbs_uobject_get(struct ib_uobject *uobject); + +/* + * In order to indicate we no longer needs this uobject, uverbs_uobject_put + * is called. When the reference count is decreased, the uobject is freed. + * For example, this is used when attaching a completion channel to a CQ. + */ +void uverbs_uobject_put(struct ib_uobject *uobject); + +/* Indicate this fd is no longer used by this consumer, but its memory isn't + * necessarily released yet. When the last reference is put, we release the + * memory. After this call is executed, calling uverbs_uobject_get isn't + * allowed. + * This must be called from the release file_operations of the file! + */ +void uverbs_close_fd(struct file *f); + +#endif /* RDMA_CORE_H */ diff --git a/drivers/infiniband/core/sa_query.c b/drivers/infiniband/core/sa_query.c index ceae153997d0..b57bdc257fcf 100644 --- a/drivers/infiniband/core/sa_query.c +++ b/drivers/infiniband/core/sa_query.c @@ -56,6 +56,8 @@ #define IB_SA_LOCAL_SVC_TIMEOUT_MIN 100 #define IB_SA_LOCAL_SVC_TIMEOUT_DEFAULT 2000 #define IB_SA_LOCAL_SVC_TIMEOUT_MAX 200000 +#define IB_SA_CPI_MAX_RETRY_CNT 3 +#define IB_SA_CPI_RETRY_WAIT 1000 /*msecs */ static int sa_local_svc_timeout_ms = IB_SA_LOCAL_SVC_TIMEOUT_DEFAULT; struct ib_sa_sm_ah { @@ -65,9 +67,23 @@ struct ib_sa_sm_ah { u8 src_path_mask; }; +enum rdma_class_port_info_type { + RDMA_CLASS_PORT_INFO_IB, + RDMA_CLASS_PORT_INFO_OPA +}; + +struct rdma_class_port_info { + enum rdma_class_port_info_type type; + union { + struct ib_class_port_info ib; + struct opa_class_port_info opa; + }; +}; + struct ib_sa_classport_cache { bool valid; - struct ib_class_port_info data; + int retry_cnt; + struct rdma_class_port_info data; }; struct ib_sa_port { @@ -75,6 +91,7 @@ struct ib_sa_port { struct ib_sa_sm_ah *sm_ah; struct work_struct update_task; struct ib_sa_classport_cache classport_info; + struct delayed_work ib_cpi_work; spinlock_t classport_lock; /* protects class port info set */ spinlock_t ah_lock; u8 port_num; @@ -103,6 +120,7 @@ struct ib_sa_query { #define IB_SA_ENABLE_LOCAL_SERVICE 0x00000001 #define IB_SA_CANCEL 0x00000002 +#define IB_SA_QUERY_OPA 0x00000004 struct ib_sa_service_query { void (*callback)(int, struct ib_sa_service_rec *, void *); @@ -123,7 +141,7 @@ struct ib_sa_guidinfo_query { }; struct ib_sa_classport_info_query { - void (*callback)(int, struct ib_class_port_info *, void *); + void (*callback)(void *); void *context; struct ib_sa_query sa_query; }; @@ -406,7 +424,7 @@ static const struct ib_field service_rec_table[] = { .struct_size_bytes = sizeof((struct ib_class_port_info *)0)->field, \ .field_name = "ib_class_port_info:" #field -static const struct ib_field classport_info_rec_table[] = { +static const struct ib_field ib_classport_info_rec_table[] = { { CLASSPORTINFO_REC_FIELD(base_version), .offset_words = 0, .offset_bits = 0, @@ -477,6 +495,88 @@ static const struct ib_field classport_info_rec_table[] = { .size_bits = 32 }, }; +#define OPA_CLASSPORTINFO_REC_FIELD(field) \ + .struct_offset_bytes =\ + offsetof(struct opa_class_port_info, field), \ + .struct_size_bytes = \ + sizeof((struct opa_class_port_info *)0)->field, \ + .field_name = "opa_class_port_info:" #field + +static const struct ib_field opa_classport_info_rec_table[] = { + { OPA_CLASSPORTINFO_REC_FIELD(base_version), + .offset_words = 0, + .offset_bits = 0, + .size_bits = 8 }, + { OPA_CLASSPORTINFO_REC_FIELD(class_version), + .offset_words = 0, + .offset_bits = 8, + .size_bits = 8 }, + { OPA_CLASSPORTINFO_REC_FIELD(cap_mask), + .offset_words = 0, + .offset_bits = 16, + .size_bits = 16 }, + { OPA_CLASSPORTINFO_REC_FIELD(cap_mask2_resp_time), + .offset_words = 1, + .offset_bits = 0, + .size_bits = 32 }, + { OPA_CLASSPORTINFO_REC_FIELD(redirect_gid), + .offset_words = 2, + .offset_bits = 0, + .size_bits = 128 }, + { OPA_CLASSPORTINFO_REC_FIELD(redirect_tc_fl), + .offset_words = 6, + .offset_bits = 0, + .size_bits = 32 }, + { OPA_CLASSPORTINFO_REC_FIELD(redirect_lid), + .offset_words = 7, + .offset_bits = 0, + .size_bits = 32 }, + { OPA_CLASSPORTINFO_REC_FIELD(redirect_sl_qp), + .offset_words = 8, + .offset_bits = 0, + .size_bits = 32 }, + { OPA_CLASSPORTINFO_REC_FIELD(redirect_qkey), + .offset_words = 9, + .offset_bits = 0, + .size_bits = 32 }, + { OPA_CLASSPORTINFO_REC_FIELD(trap_gid), + .offset_words = 10, + .offset_bits = 0, + .size_bits = 128 }, + { OPA_CLASSPORTINFO_REC_FIELD(trap_tc_fl), + .offset_words = 14, + .offset_bits = 0, + .size_bits = 32 }, + { OPA_CLASSPORTINFO_REC_FIELD(trap_lid), + .offset_words = 15, + .offset_bits = 0, + .size_bits = 32 }, + { OPA_CLASSPORTINFO_REC_FIELD(trap_hl_qp), + .offset_words = 16, + .offset_bits = 0, + .size_bits = 32 }, + { OPA_CLASSPORTINFO_REC_FIELD(trap_qkey), + .offset_words = 17, + .offset_bits = 0, + .size_bits = 32 }, + { OPA_CLASSPORTINFO_REC_FIELD(trap_pkey), + .offset_words = 18, + .offset_bits = 0, + .size_bits = 16 }, + { OPA_CLASSPORTINFO_REC_FIELD(redirect_pkey), + .offset_words = 18, + .offset_bits = 16, + .size_bits = 16 }, + { OPA_CLASSPORTINFO_REC_FIELD(trap_sl_rsvd), + .offset_words = 19, + .offset_bits = 0, + .size_bits = 8 }, + { RESERVED, + .offset_words = 19, + .offset_bits = 8, + .size_bits = 24 }, +}; + #define GUIDINFO_REC_FIELD(field) \ .struct_offset_bytes = offsetof(struct ib_sa_guidinfo_rec, field), \ .struct_size_bytes = sizeof((struct ib_sa_guidinfo_rec *) 0)->field, \ @@ -931,92 +1031,6 @@ static void free_sm_ah(struct kref *kref) kfree(sm_ah); } -static void update_sm_ah(struct work_struct *work) -{ - struct ib_sa_port *port = - container_of(work, struct ib_sa_port, update_task); - struct ib_sa_sm_ah *new_ah; - struct ib_port_attr port_attr; - struct ib_ah_attr ah_attr; - - if (ib_query_port(port->agent->device, port->port_num, &port_attr)) { - pr_warn("Couldn't query port\n"); - return; - } - - new_ah = kmalloc(sizeof *new_ah, GFP_KERNEL); - if (!new_ah) { - return; - } - - kref_init(&new_ah->ref); - new_ah->src_path_mask = (1 << port_attr.lmc) - 1; - - new_ah->pkey_index = 0; - if (ib_find_pkey(port->agent->device, port->port_num, - IB_DEFAULT_PKEY_FULL, &new_ah->pkey_index)) - pr_err("Couldn't find index for default PKey\n"); - - memset(&ah_attr, 0, sizeof ah_attr); - ah_attr.dlid = port_attr.sm_lid; - ah_attr.sl = port_attr.sm_sl; - ah_attr.port_num = port->port_num; - if (port_attr.grh_required) { - ah_attr.ah_flags = IB_AH_GRH; - ah_attr.grh.dgid.global.subnet_prefix = cpu_to_be64(port_attr.subnet_prefix); - ah_attr.grh.dgid.global.interface_id = cpu_to_be64(IB_SA_WELL_KNOWN_GUID); - } - - new_ah->ah = ib_create_ah(port->agent->qp->pd, &ah_attr); - if (IS_ERR(new_ah->ah)) { - pr_warn("Couldn't create new SM AH\n"); - kfree(new_ah); - return; - } - - spin_lock_irq(&port->ah_lock); - if (port->sm_ah) - kref_put(&port->sm_ah->ref, free_sm_ah); - port->sm_ah = new_ah; - spin_unlock_irq(&port->ah_lock); - -} - -static void ib_sa_event(struct ib_event_handler *handler, struct ib_event *event) -{ - if (event->event == IB_EVENT_PORT_ERR || - event->event == IB_EVENT_PORT_ACTIVE || - event->event == IB_EVENT_LID_CHANGE || - event->event == IB_EVENT_PKEY_CHANGE || - event->event == IB_EVENT_SM_CHANGE || - event->event == IB_EVENT_CLIENT_REREGISTER) { - unsigned long flags; - struct ib_sa_device *sa_dev = - container_of(handler, typeof(*sa_dev), event_handler); - struct ib_sa_port *port = - &sa_dev->port[event->element.port_num - sa_dev->start_port]; - - if (!rdma_cap_ib_sa(handler->device, port->port_num)) - return; - - spin_lock_irqsave(&port->ah_lock, flags); - if (port->sm_ah) - kref_put(&port->sm_ah->ref, free_sm_ah); - port->sm_ah = NULL; - spin_unlock_irqrestore(&port->ah_lock, flags); - - if (event->event == IB_EVENT_SM_CHANGE || - event->event == IB_EVENT_CLIENT_REREGISTER || - event->event == IB_EVENT_LID_CHANGE) { - spin_lock_irqsave(&port->classport_lock, flags); - port->classport_info.valid = false; - spin_unlock_irqrestore(&port->classport_lock, flags); - } - queue_work(ib_wq, &sa_dev->port[event->element.port_num - - sa_dev->start_port].update_task); - } -} - void ib_sa_register_client(struct ib_sa_client *client) { atomic_set(&client->users, 1); @@ -1203,7 +1217,9 @@ static int alloc_mad(struct ib_sa_query *query, gfp_t gfp_mask) query->sm_ah->pkey_index, 0, IB_MGMT_SA_HDR, IB_MGMT_SA_DATA, gfp_mask, - IB_MGMT_BASE_VERSION); + ((query->flags & IB_SA_QUERY_OPA) ? + OPA_MGMT_BASE_VERSION : + IB_MGMT_BASE_VERSION)); if (IS_ERR(query->mad_buf)) { kref_put(&query->sm_ah->ref, free_sm_ah); return -ENOMEM; @@ -1220,16 +1236,21 @@ static void free_mad(struct ib_sa_query *query) kref_put(&query->sm_ah->ref, free_sm_ah); } -static void init_mad(struct ib_sa_mad *mad, struct ib_mad_agent *agent) +static void init_mad(struct ib_sa_query *query, struct ib_mad_agent *agent) { + struct ib_sa_mad *mad = query->mad_buf->mad; unsigned long flags; memset(mad, 0, sizeof *mad); - mad->mad_hdr.base_version = IB_MGMT_BASE_VERSION; + if (query->flags & IB_SA_QUERY_OPA) { + mad->mad_hdr.base_version = OPA_MGMT_BASE_VERSION; + mad->mad_hdr.class_version = OPA_SA_CLASS_VERSION; + } else { + mad->mad_hdr.base_version = IB_MGMT_BASE_VERSION; + mad->mad_hdr.class_version = IB_SA_CLASS_VERSION; + } mad->mad_hdr.mgmt_class = IB_MGMT_CLASS_SUBN_ADM; - mad->mad_hdr.class_version = IB_SA_CLASS_VERSION; - spin_lock_irqsave(&tid_lock, flags); mad->mad_hdr.tid = cpu_to_be64(((u64) agent->hi_tid) << 32 | tid++); @@ -1383,7 +1404,7 @@ int ib_sa_path_rec_get(struct ib_sa_client *client, query->context = context; mad = query->sa_query.mad_buf->mad; - init_mad(mad, agent); + init_mad(&query->sa_query, agent); query->sa_query.callback = callback ? ib_sa_path_rec_callback : NULL; query->sa_query.release = ib_sa_path_rec_release; @@ -1508,7 +1529,7 @@ int ib_sa_service_rec_query(struct ib_sa_client *client, query->context = context; mad = query->sa_query.mad_buf->mad; - init_mad(mad, agent); + init_mad(&query->sa_query, agent); query->sa_query.callback = callback ? ib_sa_service_rec_callback : NULL; query->sa_query.release = ib_sa_service_rec_release; @@ -1600,7 +1621,7 @@ int ib_sa_mcmember_rec_query(struct ib_sa_client *client, query->context = context; mad = query->sa_query.mad_buf->mad; - init_mad(mad, agent); + init_mad(&query->sa_query, agent); query->sa_query.callback = callback ? ib_sa_mcmember_rec_callback : NULL; query->sa_query.release = ib_sa_mcmember_rec_release; @@ -1697,7 +1718,7 @@ int ib_sa_guid_info_rec_query(struct ib_sa_client *client, query->context = context; mad = query->sa_query.mad_buf->mad; - init_mad(mad, agent); + init_mad(&query->sa_query, agent); query->sa_query.callback = callback ? ib_sa_guidinfo_rec_callback : NULL; query->sa_query.release = ib_sa_guidinfo_rec_release; @@ -1728,7 +1749,42 @@ err1: } EXPORT_SYMBOL(ib_sa_guid_info_rec_query); -/* Support get SA ClassPortInfo */ +bool ib_sa_sendonly_fullmem_support(struct ib_sa_client *client, + struct ib_device *device, + u8 port_num) +{ + struct ib_sa_device *sa_dev = ib_get_client_data(device, &sa_client); + struct ib_sa_port *port; + bool ret = false; + unsigned long flags; + + if (!sa_dev) + return ret; + + port = &sa_dev->port[port_num - sa_dev->start_port]; + + spin_lock_irqsave(&port->classport_lock, flags); + if ((port->classport_info.valid) && + (port->classport_info.data.type == RDMA_CLASS_PORT_INFO_IB)) + ret = ib_get_cpi_capmask2(&port->classport_info.data.ib) + & IB_SA_CAP_MASK2_SENDONLY_FULL_MEM_SUPPORT; + spin_unlock_irqrestore(&port->classport_lock, flags); + return ret; +} +EXPORT_SYMBOL(ib_sa_sendonly_fullmem_support); + +struct ib_classport_info_context { + struct completion done; + struct ib_sa_query *sa_query; +}; + +static void ib_classportinfo_cb(void *context) +{ + struct ib_classport_info_context *cb_ctx = context; + + complete(&cb_ctx->done); +} + static void ib_sa_classport_info_rec_callback(struct ib_sa_query *sa_query, int status, struct ib_sa_mad *mad) @@ -1736,91 +1792,91 @@ static void ib_sa_classport_info_rec_callback(struct ib_sa_query *sa_query, unsigned long flags; struct ib_sa_classport_info_query *query = container_of(sa_query, struct ib_sa_classport_info_query, sa_query); + struct ib_sa_classport_cache *info = &sa_query->port->classport_info; if (mad) { - struct ib_class_port_info rec; + if (sa_query->flags & IB_SA_QUERY_OPA) { + struct opa_class_port_info rec; - ib_unpack(classport_info_rec_table, - ARRAY_SIZE(classport_info_rec_table), - mad->data, &rec); + ib_unpack(opa_classport_info_rec_table, + ARRAY_SIZE(opa_classport_info_rec_table), + mad->data, &rec); - spin_lock_irqsave(&sa_query->port->classport_lock, flags); - if (!status && !sa_query->port->classport_info.valid) { - memcpy(&sa_query->port->classport_info.data, &rec, - sizeof(sa_query->port->classport_info.data)); + spin_lock_irqsave(&sa_query->port->classport_lock, + flags); + if (!status && !info->valid) { + memcpy(&info->data.opa, &rec, + sizeof(info->data.opa)); - sa_query->port->classport_info.valid = true; - } - spin_unlock_irqrestore(&sa_query->port->classport_lock, flags); + info->valid = true; + info->data.type = RDMA_CLASS_PORT_INFO_OPA; + } + spin_unlock_irqrestore(&sa_query->port->classport_lock, + flags); - query->callback(status, &rec, query->context); - } else { - query->callback(status, NULL, query->context); + } else { + struct ib_class_port_info rec; + + ib_unpack(ib_classport_info_rec_table, + ARRAY_SIZE(ib_classport_info_rec_table), + mad->data, &rec); + + spin_lock_irqsave(&sa_query->port->classport_lock, + flags); + if (!status && !info->valid) { + memcpy(&info->data.ib, &rec, + sizeof(info->data.ib)); + + info->valid = true; + info->data.type = RDMA_CLASS_PORT_INFO_IB; + } + spin_unlock_irqrestore(&sa_query->port->classport_lock, + flags); + } } + query->callback(query->context); } -static void ib_sa_portclass_info_rec_release(struct ib_sa_query *sa_query) +static void ib_sa_classport_info_rec_release(struct ib_sa_query *sa_query) { kfree(container_of(sa_query, struct ib_sa_classport_info_query, sa_query)); } -int ib_sa_classport_info_rec_query(struct ib_sa_client *client, - struct ib_device *device, u8 port_num, - int timeout_ms, gfp_t gfp_mask, - void (*callback)(int status, - struct ib_class_port_info *resp, - void *context), - void *context, - struct ib_sa_query **sa_query) +static int ib_sa_classport_info_rec_query(struct ib_sa_port *port, + int timeout_ms, + void (*callback)(void *context), + void *context, + struct ib_sa_query **sa_query) { - struct ib_sa_classport_info_query *query; - struct ib_sa_device *sa_dev = ib_get_client_data(device, &sa_client); - struct ib_sa_port *port; struct ib_mad_agent *agent; + struct ib_sa_classport_info_query *query; struct ib_sa_mad *mad; - struct ib_class_port_info cached_class_port_info; + gfp_t gfp_mask = GFP_KERNEL; int ret; - unsigned long flags; - - if (!sa_dev) - return -ENODEV; - port = &sa_dev->port[port_num - sa_dev->start_port]; agent = port->agent; - /* Use cached ClassPortInfo attribute if valid instead of sending mad */ - spin_lock_irqsave(&port->classport_lock, flags); - if (port->classport_info.valid && callback) { - memcpy(&cached_class_port_info, &port->classport_info.data, - sizeof(cached_class_port_info)); - spin_unlock_irqrestore(&port->classport_lock, flags); - callback(0, &cached_class_port_info, context); - return 0; - } - spin_unlock_irqrestore(&port->classport_lock, flags); - query = kzalloc(sizeof(*query), gfp_mask); if (!query) return -ENOMEM; query->sa_query.port = port; + query->sa_query.flags |= rdma_cap_opa_ah(port->agent->device, + port->port_num) ? + IB_SA_QUERY_OPA : 0; ret = alloc_mad(&query->sa_query, gfp_mask); if (ret) - goto err1; + goto err_free; - ib_sa_client_get(client); - query->sa_query.client = client; - query->callback = callback; - query->context = context; + query->callback = callback; + query->context = context; mad = query->sa_query.mad_buf->mad; - init_mad(mad, agent); - - query->sa_query.callback = callback ? ib_sa_classport_info_rec_callback : NULL; + init_mad(&query->sa_query, agent); - query->sa_query.release = ib_sa_portclass_info_rec_release; - /* support GET only */ + query->sa_query.callback = ib_sa_classport_info_rec_callback; + query->sa_query.release = ib_sa_classport_info_rec_release; mad->mad_hdr.method = IB_MGMT_METHOD_GET; mad->mad_hdr.attr_id = cpu_to_be16(IB_SA_ATTR_CLASS_PORTINFO); mad->sa_hdr.comp_mask = 0; @@ -1828,20 +1884,71 @@ int ib_sa_classport_info_rec_query(struct ib_sa_client *client, ret = send_mad(&query->sa_query, timeout_ms, gfp_mask); if (ret < 0) - goto err2; + goto err_free_mad; return ret; -err2: +err_free_mad: *sa_query = NULL; - ib_sa_client_put(query->sa_query.client); free_mad(&query->sa_query); -err1: +err_free: kfree(query); return ret; } -EXPORT_SYMBOL(ib_sa_classport_info_rec_query); + +static void update_ib_cpi(struct work_struct *work) +{ + struct ib_sa_port *port = + container_of(work, struct ib_sa_port, ib_cpi_work.work); + struct ib_classport_info_context *cb_context; + unsigned long flags; + int ret; + + /* If the classport info is valid, nothing + * to do here. + */ + spin_lock_irqsave(&port->classport_lock, flags); + if (port->classport_info.valid) { + spin_unlock_irqrestore(&port->classport_lock, flags); + return; + } + spin_unlock_irqrestore(&port->classport_lock, flags); + + cb_context = kmalloc(sizeof(*cb_context), GFP_KERNEL); + if (!cb_context) + goto err_nomem; + + init_completion(&cb_context->done); + + ret = ib_sa_classport_info_rec_query(port, 3000, + ib_classportinfo_cb, cb_context, + &cb_context->sa_query); + if (ret < 0) + goto free_cb_err; + wait_for_completion(&cb_context->done); +free_cb_err: + kfree(cb_context); + spin_lock_irqsave(&port->classport_lock, flags); + + /* If the classport info is still not valid, the query should have + * failed for some reason. Retry issuing the query + */ + if (!port->classport_info.valid) { + port->classport_info.retry_cnt++; + if (port->classport_info.retry_cnt <= + IB_SA_CPI_MAX_RETRY_CNT) { + unsigned long delay = + msecs_to_jiffies(IB_SA_CPI_RETRY_WAIT); + + queue_delayed_work(ib_wq, &port->ib_cpi_work, delay); + } + } + spin_unlock_irqrestore(&port->classport_lock, flags); + +err_nomem: + return; +} static void send_handler(struct ib_mad_agent *agent, struct ib_mad_send_wc *mad_send_wc) @@ -1870,7 +1977,8 @@ static void send_handler(struct ib_mad_agent *agent, spin_unlock_irqrestore(&idr_lock, flags); free_mad(query); - ib_sa_client_put(query->client); + if (query->client) + ib_sa_client_put(query->client); query->release(query); } @@ -1897,6 +2005,99 @@ static void recv_handler(struct ib_mad_agent *mad_agent, ib_free_recv_mad(mad_recv_wc); } +static void update_sm_ah(struct work_struct *work) +{ + struct ib_sa_port *port = + container_of(work, struct ib_sa_port, update_task); + struct ib_sa_sm_ah *new_ah; + struct ib_port_attr port_attr; + struct ib_ah_attr ah_attr; + + if (ib_query_port(port->agent->device, port->port_num, &port_attr)) { + pr_warn("Couldn't query port\n"); + return; + } + + new_ah = kmalloc(sizeof(*new_ah), GFP_KERNEL); + if (!new_ah) + return; + + kref_init(&new_ah->ref); + new_ah->src_path_mask = (1 << port_attr.lmc) - 1; + + new_ah->pkey_index = 0; + if (ib_find_pkey(port->agent->device, port->port_num, + IB_DEFAULT_PKEY_FULL, &new_ah->pkey_index)) + pr_err("Couldn't find index for default PKey\n"); + + memset(&ah_attr, 0, sizeof(ah_attr)); + ah_attr.dlid = port_attr.sm_lid; + ah_attr.sl = port_attr.sm_sl; + ah_attr.port_num = port->port_num; + if (port_attr.grh_required) { + ah_attr.ah_flags = IB_AH_GRH; + ah_attr.grh.dgid.global.subnet_prefix = + cpu_to_be64(port_attr.subnet_prefix); + ah_attr.grh.dgid.global.interface_id = + cpu_to_be64(IB_SA_WELL_KNOWN_GUID); + } + + new_ah->ah = ib_create_ah(port->agent->qp->pd, &ah_attr); + if (IS_ERR(new_ah->ah)) { + pr_warn("Couldn't create new SM AH\n"); + kfree(new_ah); + return; + } + + spin_lock_irq(&port->ah_lock); + if (port->sm_ah) + kref_put(&port->sm_ah->ref, free_sm_ah); + port->sm_ah = new_ah; + spin_unlock_irq(&port->ah_lock); +} + +static void ib_sa_event(struct ib_event_handler *handler, + struct ib_event *event) +{ + if (event->event == IB_EVENT_PORT_ERR || + event->event == IB_EVENT_PORT_ACTIVE || + event->event == IB_EVENT_LID_CHANGE || + event->event == IB_EVENT_PKEY_CHANGE || + event->event == IB_EVENT_SM_CHANGE || + event->event == IB_EVENT_CLIENT_REREGISTER) { + unsigned long flags; + struct ib_sa_device *sa_dev = + container_of(handler, typeof(*sa_dev), event_handler); + u8 port_num = event->element.port_num - sa_dev->start_port; + struct ib_sa_port *port = &sa_dev->port[port_num]; + + if (!rdma_cap_ib_sa(handler->device, port->port_num)) + return; + + spin_lock_irqsave(&port->ah_lock, flags); + if (port->sm_ah) + kref_put(&port->sm_ah->ref, free_sm_ah); + port->sm_ah = NULL; + spin_unlock_irqrestore(&port->ah_lock, flags); + + if (event->event == IB_EVENT_SM_CHANGE || + event->event == IB_EVENT_CLIENT_REREGISTER || + event->event == IB_EVENT_LID_CHANGE || + event->event == IB_EVENT_PORT_ACTIVE) { + unsigned long delay = + msecs_to_jiffies(IB_SA_CPI_RETRY_WAIT); + + spin_lock_irqsave(&port->classport_lock, flags); + port->classport_info.valid = false; + port->classport_info.retry_cnt = 0; + spin_unlock_irqrestore(&port->classport_lock, flags); + queue_delayed_work(ib_wq, + &port->ib_cpi_work, delay); + } + queue_work(ib_wq, &sa_dev->port[port_num].update_task); + } +} + static void ib_sa_add_one(struct ib_device *device) { struct ib_sa_device *sa_dev; @@ -1934,6 +2135,8 @@ static void ib_sa_add_one(struct ib_device *device) goto err; INIT_WORK(&sa_dev->port[i].update_task, update_sm_ah); + INIT_DELAYED_WORK(&sa_dev->port[i].ib_cpi_work, + update_ib_cpi); count++; } @@ -1980,11 +2183,11 @@ static void ib_sa_remove_one(struct ib_device *device, void *client_data) return; ib_unregister_event_handler(&sa_dev->event_handler); - flush_workqueue(ib_wq); for (i = 0; i <= sa_dev->end_port - sa_dev->start_port; ++i) { if (rdma_cap_ib_sa(device, i + 1)) { + cancel_delayed_work_sync(&sa_dev->port[i].ib_cpi_work); ib_unregister_mad_agent(sa_dev->port[i].agent); if (sa_dev->port[i].sm_ah) kref_put(&sa_dev->port[i].sm_ah->ref, free_sm_ah); diff --git a/drivers/infiniband/core/sysfs.c b/drivers/infiniband/core/sysfs.c index daadf3130c9f..7ebe1ef23652 100644 --- a/drivers/infiniband/core/sysfs.c +++ b/drivers/infiniband/core/sysfs.c @@ -253,6 +253,10 @@ static ssize_t rate_show(struct ib_port *p, struct port_attribute *unused, speed = " EDR"; rate = 250; break; + case IB_SPEED_HDR: + speed = " HDR"; + rate = 500; + break; case IB_SPEED_SDR: default: /* default to SDR for invalid rates */ rate = 25; @@ -1301,7 +1305,7 @@ err_put: free_port_list_attributes(device); err_unregister: - device_unregister(class_dev); + device_del(class_dev); err: return ret; diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c index 27f155d2df8d..3dbf811d3c51 100644 --- a/drivers/infiniband/core/umem.c +++ b/drivers/infiniband/core/umem.c @@ -115,11 +115,11 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, if (!umem) return ERR_PTR(-ENOMEM); - umem->context = context; - umem->length = size; - umem->address = addr; - umem->page_size = PAGE_SIZE; - umem->pid = get_task_pid(current, PIDTYPE_PID); + umem->context = context; + umem->length = size; + umem->address = addr; + umem->page_shift = PAGE_SHIFT; + umem->pid = get_task_pid(current, PIDTYPE_PID); /* * We ask for writable memory if any of the following * access flags are set. "Local write" and "remote write" @@ -133,7 +133,7 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, if (access & IB_ACCESS_ON_DEMAND) { put_pid(umem->pid); - ret = ib_umem_odp_get(context, umem); + ret = ib_umem_odp_get(context, umem, access); if (ret) { kfree(umem); return ERR_PTR(ret); @@ -315,7 +315,6 @@ EXPORT_SYMBOL(ib_umem_release); int ib_umem_page_count(struct ib_umem *umem) { - int shift; int i; int n; struct scatterlist *sg; @@ -323,11 +322,9 @@ int ib_umem_page_count(struct ib_umem *umem) if (umem->odp_data) return ib_umem_num_pages(umem); - shift = ilog2(umem->page_size); - n = 0; for_each_sg(umem->sg_head.sgl, sg, umem->nmap, i) - n += sg_dma_len(sg) >> shift; + n += sg_dma_len(sg) >> umem->page_shift; return n; } diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c index cb2742b548bb..0780b1afefa9 100644 --- a/drivers/infiniband/core/umem_odp.c +++ b/drivers/infiniband/core/umem_odp.c @@ -38,6 +38,7 @@ #include <linux/slab.h> #include <linux/export.h> #include <linux/vmalloc.h> +#include <linux/hugetlb.h> #include <rdma/ib_verbs.h> #include <rdma/ib_umem.h> @@ -254,11 +255,11 @@ struct ib_umem *ib_alloc_odp_umem(struct ib_ucontext *context, if (!umem) return ERR_PTR(-ENOMEM); - umem->context = context; - umem->length = size; - umem->address = addr; - umem->page_size = PAGE_SIZE; - umem->writable = 1; + umem->context = context; + umem->length = size; + umem->address = addr; + umem->page_shift = PAGE_SHIFT; + umem->writable = 1; odp_data = kzalloc(sizeof(*odp_data), GFP_KERNEL); if (!odp_data) { @@ -306,7 +307,8 @@ out_umem: } EXPORT_SYMBOL(ib_alloc_odp_umem); -int ib_umem_odp_get(struct ib_ucontext *context, struct ib_umem *umem) +int ib_umem_odp_get(struct ib_ucontext *context, struct ib_umem *umem, + int access) { int ret_val; struct pid *our_pid; @@ -315,6 +317,20 @@ int ib_umem_odp_get(struct ib_ucontext *context, struct ib_umem *umem) if (!mm) return -EINVAL; + if (access & IB_ACCESS_HUGETLB) { + struct vm_area_struct *vma; + struct hstate *h; + + vma = find_vma(mm, ib_umem_start(umem)); + if (!vma || !is_vm_hugetlb_page(vma)) + return -EINVAL; + h = hstate_vma(vma); + umem->page_shift = huge_page_shift(h); + umem->hugetlb = 1; + } else { + umem->hugetlb = 0; + } + /* Prevent creating ODP MRs in child processes */ rcu_read_lock(); our_pid = get_task_pid(current->group_leader, PIDTYPE_PID); @@ -325,7 +341,6 @@ int ib_umem_odp_get(struct ib_ucontext *context, struct ib_umem *umem) goto out_mm; } - umem->hugetlb = 0; umem->odp_data = kzalloc(sizeof(*umem->odp_data), GFP_KERNEL); if (!umem->odp_data) { ret_val = -ENOMEM; @@ -504,7 +519,6 @@ out: static int ib_umem_odp_map_dma_single_page( struct ib_umem *umem, int page_index, - u64 base_virt_addr, struct page *page, u64 access_mask, unsigned long current_seq) @@ -527,7 +541,7 @@ static int ib_umem_odp_map_dma_single_page( if (!(umem->odp_data->dma_list[page_index])) { dma_addr = ib_dma_map_page(dev, page, - 0, PAGE_SIZE, + 0, BIT(umem->page_shift), DMA_BIDIRECTIONAL); if (ib_dma_mapping_error(dev, dma_addr)) { ret = -EFAULT; @@ -555,8 +569,9 @@ out: if (remove_existing_mapping && umem->context->invalidate_range) { invalidate_page_trampoline( umem, - base_virt_addr + (page_index * PAGE_SIZE), - base_virt_addr + ((page_index+1)*PAGE_SIZE), + ib_umem_start(umem) + (page_index >> umem->page_shift), + ib_umem_start(umem) + ((page_index + 1) >> + umem->page_shift), NULL); ret = -EAGAIN; } @@ -595,10 +610,10 @@ int ib_umem_odp_map_dma_pages(struct ib_umem *umem, u64 user_virt, u64 bcnt, struct task_struct *owning_process = NULL; struct mm_struct *owning_mm = NULL; struct page **local_page_list = NULL; - u64 off; - int j, k, ret = 0, start_idx, npages = 0; - u64 base_virt_addr; + u64 page_mask, off; + int j, k, ret = 0, start_idx, npages = 0, page_shift; unsigned int flags = 0; + phys_addr_t p = 0; if (access_mask == 0) return -EINVAL; @@ -611,9 +626,10 @@ int ib_umem_odp_map_dma_pages(struct ib_umem *umem, u64 user_virt, u64 bcnt, if (!local_page_list) return -ENOMEM; - off = user_virt & (~PAGE_MASK); - user_virt = user_virt & PAGE_MASK; - base_virt_addr = user_virt; + page_shift = umem->page_shift; + page_mask = ~(BIT(page_shift) - 1); + off = user_virt & (~page_mask); + user_virt = user_virt & page_mask; bcnt += off; /* Charge for the first page offset as well. */ owning_process = get_pid_task(umem->context->tgid, PIDTYPE_PID); @@ -631,13 +647,13 @@ int ib_umem_odp_map_dma_pages(struct ib_umem *umem, u64 user_virt, u64 bcnt, if (access_mask & ODP_WRITE_ALLOWED_BIT) flags |= FOLL_WRITE; - start_idx = (user_virt - ib_umem_start(umem)) >> PAGE_SHIFT; + start_idx = (user_virt - ib_umem_start(umem)) >> page_shift; k = start_idx; while (bcnt > 0) { - const size_t gup_num_pages = - min_t(size_t, ALIGN(bcnt, PAGE_SIZE) / PAGE_SIZE, - PAGE_SIZE / sizeof(struct page *)); + const size_t gup_num_pages = min_t(size_t, + (bcnt + BIT(page_shift) - 1) >> page_shift, + PAGE_SIZE / sizeof(struct page *)); down_read(&owning_mm->mmap_sem); /* @@ -656,14 +672,25 @@ int ib_umem_odp_map_dma_pages(struct ib_umem *umem, u64 user_virt, u64 bcnt, break; bcnt -= min_t(size_t, npages << PAGE_SHIFT, bcnt); - user_virt += npages << PAGE_SHIFT; mutex_lock(&umem->odp_data->umem_mutex); - for (j = 0; j < npages; ++j) { + for (j = 0; j < npages; j++, user_virt += PAGE_SIZE) { + if (user_virt & ~page_mask) { + p += PAGE_SIZE; + if (page_to_phys(local_page_list[j]) != p) { + ret = -EFAULT; + break; + } + put_page(local_page_list[j]); + continue; + } + ret = ib_umem_odp_map_dma_single_page( - umem, k, base_virt_addr, local_page_list[j], - access_mask, current_seq); + umem, k, local_page_list[j], + access_mask, current_seq); if (ret < 0) break; + + p = page_to_phys(local_page_list[j]); k++; } mutex_unlock(&umem->odp_data->umem_mutex); @@ -707,8 +734,8 @@ void ib_umem_odp_unmap_dma_pages(struct ib_umem *umem, u64 virt, * invalidations, so we must make sure we free each page only * once. */ mutex_lock(&umem->odp_data->umem_mutex); - for (addr = virt; addr < bound; addr += (u64)umem->page_size) { - idx = (addr - ib_umem_start(umem)) / PAGE_SIZE; + for (addr = virt; addr < bound; addr += BIT(umem->page_shift)) { + idx = (addr - ib_umem_start(umem)) >> umem->page_shift; if (umem->odp_data->page_list[idx]) { struct page *page = umem->odp_data->page_list[idx]; dma_addr_t dma = umem->odp_data->dma_list[idx]; diff --git a/drivers/infiniband/core/uverbs.h b/drivers/infiniband/core/uverbs.h index e1bedf0bac04..64d494a64daf 100644 --- a/drivers/infiniband/core/uverbs.h +++ b/drivers/infiniband/core/uverbs.h @@ -76,12 +76,13 @@ * an asynchronous event queue file is created and released when the * event file is closed. * - * struct ib_uverbs_event_file: One reference is held by the VFS and - * released when the file is closed. For asynchronous event files, - * another reference is held by the corresponding main context file - * and released when that file is closed. For completion event files, - * a reference is taken when a CQ is created that uses the file, and - * released when the CQ is destroyed. + * struct ib_uverbs_event_queue: Base structure for + * struct ib_uverbs_async_event_file and struct ib_uverbs_completion_event_file. + * One reference is held by the VFS and released when the file is closed. + * For asynchronous event files, another reference is held by the corresponding + * main context file and released when that file is closed. For completion + * event files, a reference is taken when a CQ is created that uses the file, + * and released when the CQ is destroyed. */ struct ib_uverbs_device { @@ -101,18 +102,26 @@ struct ib_uverbs_device { struct list_head uverbs_events_file_list; }; -struct ib_uverbs_event_file { - struct kref ref; - int is_async; - struct ib_uverbs_file *uverbs_file; +struct ib_uverbs_event_queue { spinlock_t lock; int is_closed; wait_queue_head_t poll_wait; struct fasync_struct *async_queue; struct list_head event_list; +}; + +struct ib_uverbs_async_event_file { + struct ib_uverbs_event_queue ev_queue; + struct ib_uverbs_file *uverbs_file; + struct kref ref; struct list_head list; }; +struct ib_uverbs_completion_event_file { + struct ib_uobject_file uobj_file; + struct ib_uverbs_event_queue ev_queue; +}; + struct ib_uverbs_file { struct kref ref; struct mutex mutex; @@ -120,9 +129,13 @@ struct ib_uverbs_file { struct ib_uverbs_device *device; struct ib_ucontext *ucontext; struct ib_event_handler event_handler; - struct ib_uverbs_event_file *async_file; + struct ib_uverbs_async_event_file *async_file; struct list_head list; int is_closed; + + struct idr idr; + /* spinlock protects write access to idr */ + spinlock_t idr_lock; }; struct ib_uverbs_event { @@ -159,6 +172,8 @@ struct ib_usrq_object { struct ib_uqp_object { struct ib_uevent_object uevent; + /* lock for mcast list */ + struct mutex mcast_lock; struct list_head mcast_list; struct ib_uxrcd_object *uxrcd; }; @@ -176,32 +191,18 @@ struct ib_ucq_object { u32 async_events_reported; }; -extern spinlock_t ib_uverbs_idr_lock; -extern struct idr ib_uverbs_pd_idr; -extern struct idr ib_uverbs_mr_idr; -extern struct idr ib_uverbs_mw_idr; -extern struct idr ib_uverbs_ah_idr; -extern struct idr ib_uverbs_cq_idr; -extern struct idr ib_uverbs_qp_idr; -extern struct idr ib_uverbs_srq_idr; -extern struct idr ib_uverbs_xrcd_idr; -extern struct idr ib_uverbs_rule_idr; -extern struct idr ib_uverbs_wq_idr; -extern struct idr ib_uverbs_rwq_ind_tbl_idr; - -void idr_remove_uobj(struct idr *idp, struct ib_uobject *uobj); - -struct file *ib_uverbs_alloc_event_file(struct ib_uverbs_file *uverbs_file, - struct ib_device *ib_dev, - int is_async); +extern const struct file_operations uverbs_event_fops; +void ib_uverbs_init_event_queue(struct ib_uverbs_event_queue *ev_queue); +struct file *ib_uverbs_alloc_async_event_file(struct ib_uverbs_file *uverbs_file, + struct ib_device *ib_dev); void ib_uverbs_free_async_event_file(struct ib_uverbs_file *uverbs_file); -struct ib_uverbs_event_file *ib_uverbs_lookup_comp_file(int fd); void ib_uverbs_release_ucq(struct ib_uverbs_file *file, - struct ib_uverbs_event_file *ev_file, + struct ib_uverbs_completion_event_file *ev_file, struct ib_ucq_object *uobj); void ib_uverbs_release_uevent(struct ib_uverbs_file *file, struct ib_uevent_object *uobj); +void ib_uverbs_release_file(struct kref *ref); void ib_uverbs_comp_handler(struct ib_cq *cq, void *cq_context); void ib_uverbs_cq_event_handler(struct ib_event *event, void *context_ptr); @@ -210,9 +211,12 @@ void ib_uverbs_wq_event_handler(struct ib_event *event, void *context_ptr); void ib_uverbs_srq_event_handler(struct ib_event *event, void *context_ptr); void ib_uverbs_event_handler(struct ib_event_handler *handler, struct ib_event *event); -void ib_uverbs_dealloc_xrcd(struct ib_uverbs_device *dev, struct ib_xrcd *xrcd); +int ib_uverbs_dealloc_xrcd(struct ib_uverbs_device *dev, struct ib_xrcd *xrcd, + enum rdma_remove_reason why); int uverbs_dealloc_mw(struct ib_mw *mw); +void ib_uverbs_detach_umcast(struct ib_qp *qp, + struct ib_uqp_object *uobj); struct ib_uverbs_flow_spec { union { @@ -229,6 +233,7 @@ struct ib_uverbs_flow_spec { struct ib_uverbs_flow_spec_tcp_udp tcp_udp; struct ib_uverbs_flow_spec_ipv6 ipv6; struct ib_uverbs_flow_spec_action_tag flow_tag; + struct ib_uverbs_flow_spec_action_drop drop; }; }; diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 7b7a76e1279a..cb3c426c0dad 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -40,270 +40,29 @@ #include <linux/uaccess.h> +#include <rdma/uverbs_types.h> +#include <rdma/uverbs_std_types.h> +#include "rdma_core.h" + #include "uverbs.h" #include "core_priv.h" -struct uverbs_lock_class { - struct lock_class_key key; - char name[16]; -}; - -static struct uverbs_lock_class pd_lock_class = { .name = "PD-uobj" }; -static struct uverbs_lock_class mr_lock_class = { .name = "MR-uobj" }; -static struct uverbs_lock_class mw_lock_class = { .name = "MW-uobj" }; -static struct uverbs_lock_class cq_lock_class = { .name = "CQ-uobj" }; -static struct uverbs_lock_class qp_lock_class = { .name = "QP-uobj" }; -static struct uverbs_lock_class ah_lock_class = { .name = "AH-uobj" }; -static struct uverbs_lock_class srq_lock_class = { .name = "SRQ-uobj" }; -static struct uverbs_lock_class xrcd_lock_class = { .name = "XRCD-uobj" }; -static struct uverbs_lock_class rule_lock_class = { .name = "RULE-uobj" }; -static struct uverbs_lock_class wq_lock_class = { .name = "WQ-uobj" }; -static struct uverbs_lock_class rwq_ind_table_lock_class = { .name = "IND_TBL-uobj" }; - -/* - * The ib_uobject locking scheme is as follows: - * - * - ib_uverbs_idr_lock protects the uverbs idrs themselves, so it - * needs to be held during all idr write operations. When an object is - * looked up, a reference must be taken on the object's kref before - * dropping this lock. For read operations, the rcu_read_lock() - * and rcu_write_lock() but similarly the kref reference is grabbed - * before the rcu_read_unlock(). - * - * - Each object also has an rwsem. This rwsem must be held for - * reading while an operation that uses the object is performed. - * For example, while registering an MR, the associated PD's - * uobject.mutex must be held for reading. The rwsem must be held - * for writing while initializing or destroying an object. - * - * - In addition, each object has a "live" flag. If this flag is not - * set, then lookups of the object will fail even if it is found in - * the idr. This handles a reader that blocks and does not acquire - * the rwsem until after the object is destroyed. The destroy - * operation will set the live flag to 0 and then drop the rwsem; - * this will allow the reader to acquire the rwsem, see that the - * live flag is 0, and then drop the rwsem and its reference to - * object. The underlying storage will not be freed until the last - * reference to the object is dropped. - */ - -static void init_uobj(struct ib_uobject *uobj, u64 user_handle, - struct ib_ucontext *context, struct uverbs_lock_class *c) -{ - uobj->user_handle = user_handle; - uobj->context = context; - kref_init(&uobj->ref); - init_rwsem(&uobj->mutex); - lockdep_set_class_and_name(&uobj->mutex, &c->key, c->name); - uobj->live = 0; -} - -static void release_uobj(struct kref *kref) -{ - kfree_rcu(container_of(kref, struct ib_uobject, ref), rcu); -} - -static void put_uobj(struct ib_uobject *uobj) -{ - kref_put(&uobj->ref, release_uobj); -} - -static void put_uobj_read(struct ib_uobject *uobj) -{ - up_read(&uobj->mutex); - put_uobj(uobj); -} - -static void put_uobj_write(struct ib_uobject *uobj) -{ - up_write(&uobj->mutex); - put_uobj(uobj); -} - -static int idr_add_uobj(struct idr *idr, struct ib_uobject *uobj) -{ - int ret; - - idr_preload(GFP_KERNEL); - spin_lock(&ib_uverbs_idr_lock); - - ret = idr_alloc(idr, uobj, 0, 0, GFP_NOWAIT); - if (ret >= 0) - uobj->id = ret; - - spin_unlock(&ib_uverbs_idr_lock); - idr_preload_end(); - - return ret < 0 ? ret : 0; -} - -void idr_remove_uobj(struct idr *idr, struct ib_uobject *uobj) -{ - spin_lock(&ib_uverbs_idr_lock); - idr_remove(idr, uobj->id); - spin_unlock(&ib_uverbs_idr_lock); -} - -static struct ib_uobject *__idr_get_uobj(struct idr *idr, int id, - struct ib_ucontext *context) -{ - struct ib_uobject *uobj; - - rcu_read_lock(); - uobj = idr_find(idr, id); - if (uobj) { - if (uobj->context == context) - kref_get(&uobj->ref); - else - uobj = NULL; - } - rcu_read_unlock(); - - return uobj; -} - -static struct ib_uobject *idr_read_uobj(struct idr *idr, int id, - struct ib_ucontext *context, int nested) -{ - struct ib_uobject *uobj; - - uobj = __idr_get_uobj(idr, id, context); - if (!uobj) - return NULL; - - if (nested) - down_read_nested(&uobj->mutex, SINGLE_DEPTH_NESTING); - else - down_read(&uobj->mutex); - if (!uobj->live) { - put_uobj_read(uobj); - return NULL; - } - - return uobj; -} - -static struct ib_uobject *idr_write_uobj(struct idr *idr, int id, - struct ib_ucontext *context) -{ - struct ib_uobject *uobj; - - uobj = __idr_get_uobj(idr, id, context); - if (!uobj) - return NULL; - - down_write(&uobj->mutex); - if (!uobj->live) { - put_uobj_write(uobj); - return NULL; - } - - return uobj; -} - -static void *idr_read_obj(struct idr *idr, int id, struct ib_ucontext *context, - int nested) -{ - struct ib_uobject *uobj; - - uobj = idr_read_uobj(idr, id, context, nested); - return uobj ? uobj->object : NULL; -} - -static struct ib_pd *idr_read_pd(int pd_handle, struct ib_ucontext *context) +static struct ib_uverbs_completion_event_file * +ib_uverbs_lookup_comp_file(int fd, struct ib_ucontext *context) { - return idr_read_obj(&ib_uverbs_pd_idr, pd_handle, context, 0); -} + struct ib_uobject *uobj = uobj_get_read(uobj_get_type(comp_channel), + fd, context); + struct ib_uobject_file *uobj_file; -static void put_pd_read(struct ib_pd *pd) -{ - put_uobj_read(pd->uobject); -} + if (IS_ERR(uobj)) + return (void *)uobj; -static struct ib_cq *idr_read_cq(int cq_handle, struct ib_ucontext *context, int nested) -{ - return idr_read_obj(&ib_uverbs_cq_idr, cq_handle, context, nested); -} + uverbs_uobject_get(uobj); + uobj_put_read(uobj); -static void put_cq_read(struct ib_cq *cq) -{ - put_uobj_read(cq->uobject); -} - -static struct ib_ah *idr_read_ah(int ah_handle, struct ib_ucontext *context) -{ - return idr_read_obj(&ib_uverbs_ah_idr, ah_handle, context, 0); -} - -static void put_ah_read(struct ib_ah *ah) -{ - put_uobj_read(ah->uobject); -} - -static struct ib_qp *idr_read_qp(int qp_handle, struct ib_ucontext *context) -{ - return idr_read_obj(&ib_uverbs_qp_idr, qp_handle, context, 0); -} - -static struct ib_wq *idr_read_wq(int wq_handle, struct ib_ucontext *context) -{ - return idr_read_obj(&ib_uverbs_wq_idr, wq_handle, context, 0); -} - -static void put_wq_read(struct ib_wq *wq) -{ - put_uobj_read(wq->uobject); -} - -static struct ib_rwq_ind_table *idr_read_rwq_indirection_table(int ind_table_handle, - struct ib_ucontext *context) -{ - return idr_read_obj(&ib_uverbs_rwq_ind_tbl_idr, ind_table_handle, context, 0); -} - -static void put_rwq_indirection_table_read(struct ib_rwq_ind_table *ind_table) -{ - put_uobj_read(ind_table->uobject); -} - -static struct ib_qp *idr_write_qp(int qp_handle, struct ib_ucontext *context) -{ - struct ib_uobject *uobj; - - uobj = idr_write_uobj(&ib_uverbs_qp_idr, qp_handle, context); - return uobj ? uobj->object : NULL; -} - -static void put_qp_read(struct ib_qp *qp) -{ - put_uobj_read(qp->uobject); -} - -static void put_qp_write(struct ib_qp *qp) -{ - put_uobj_write(qp->uobject); -} - -static struct ib_srq *idr_read_srq(int srq_handle, struct ib_ucontext *context) -{ - return idr_read_obj(&ib_uverbs_srq_idr, srq_handle, context, 0); -} - -static void put_srq_read(struct ib_srq *srq) -{ - put_uobj_read(srq->uobject); -} - -static struct ib_xrcd *idr_read_xrcd(int xrcd_handle, struct ib_ucontext *context, - struct ib_uobject **uobj) -{ - *uobj = idr_read_uobj(&ib_uverbs_xrcd_idr, xrcd_handle, context, 0); - return *uobj ? (*uobj)->object : NULL; -} - -static void put_xrcd_read(struct ib_uobject *uobj) -{ - put_uobj_read(uobj); + uobj_file = container_of(uobj, struct ib_uobject_file, uobj); + return container_of(uobj_file, struct ib_uverbs_completion_event_file, + uobj_file); } ssize_t ib_uverbs_get_context(struct ib_uverbs_file *file, @@ -348,17 +107,10 @@ ssize_t ib_uverbs_get_context(struct ib_uverbs_file *file, ucontext->device = ib_dev; ucontext->cg_obj = cg_obj; - INIT_LIST_HEAD(&ucontext->pd_list); - INIT_LIST_HEAD(&ucontext->mr_list); - INIT_LIST_HEAD(&ucontext->mw_list); - INIT_LIST_HEAD(&ucontext->cq_list); - INIT_LIST_HEAD(&ucontext->qp_list); - INIT_LIST_HEAD(&ucontext->srq_list); - INIT_LIST_HEAD(&ucontext->ah_list); - INIT_LIST_HEAD(&ucontext->wq_list); - INIT_LIST_HEAD(&ucontext->rwq_ind_tbl_list); - INIT_LIST_HEAD(&ucontext->xrcd_list); - INIT_LIST_HEAD(&ucontext->rule_list); + /* ufile is required when some objects are released */ + ucontext->ufile = file; + uverbs_initialize_ucontext(ucontext); + rcu_read_lock(); ucontext->tgid = get_task_pid(current->group_leader, PIDTYPE_PID); rcu_read_unlock(); @@ -382,7 +134,7 @@ ssize_t ib_uverbs_get_context(struct ib_uverbs_file *file, goto err_free; resp.async_fd = ret; - filp = ib_uverbs_alloc_event_file(file, ib_dev, 1); + filp = ib_uverbs_alloc_async_event_file(file, ib_dev); if (IS_ERR(filp)) { ret = PTR_ERR(filp); goto err_fd; @@ -565,19 +317,9 @@ ssize_t ib_uverbs_alloc_pd(struct ib_uverbs_file *file, (unsigned long) cmd.response + sizeof resp, in_len - sizeof cmd, out_len - sizeof resp); - uobj = kmalloc(sizeof *uobj, GFP_KERNEL); - if (!uobj) - return -ENOMEM; - - init_uobj(uobj, 0, file->ucontext, &pd_lock_class); - ret = ib_rdmacg_try_charge(&uobj->cg_obj, ib_dev, - RDMACG_RESOURCE_HCA_OBJECT); - if (ret) { - kfree(uobj); - return ret; - } - - down_write(&uobj->mutex); + uobj = uobj_alloc(uobj_get_type(pd), file->ucontext); + if (IS_ERR(uobj)) + return PTR_ERR(uobj); pd = ib_dev->alloc_pd(ib_dev, file->ucontext, &udata); if (IS_ERR(pd)) { @@ -591,10 +333,6 @@ ssize_t ib_uverbs_alloc_pd(struct ib_uverbs_file *file, atomic_set(&pd->usecnt, 0); uobj->object = pd; - ret = idr_add_uobj(&ib_uverbs_pd_idr, uobj); - if (ret) - goto err_idr; - memset(&resp, 0, sizeof resp); resp.pd_handle = uobj->id; @@ -604,25 +342,15 @@ ssize_t ib_uverbs_alloc_pd(struct ib_uverbs_file *file, goto err_copy; } - mutex_lock(&file->mutex); - list_add_tail(&uobj->list, &file->ucontext->pd_list); - mutex_unlock(&file->mutex); - - uobj->live = 1; - - up_write(&uobj->mutex); + uobj_alloc_commit(uobj); return in_len; err_copy: - idr_remove_uobj(&ib_uverbs_pd_idr, uobj); - -err_idr: ib_dealloc_pd(pd); err: - ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); - put_uobj_write(uobj); + uobj_alloc_abort(uobj); return ret; } @@ -633,45 +361,19 @@ ssize_t ib_uverbs_dealloc_pd(struct ib_uverbs_file *file, { struct ib_uverbs_dealloc_pd cmd; struct ib_uobject *uobj; - struct ib_pd *pd; int ret; if (copy_from_user(&cmd, buf, sizeof cmd)) return -EFAULT; - uobj = idr_write_uobj(&ib_uverbs_pd_idr, cmd.pd_handle, file->ucontext); - if (!uobj) - return -EINVAL; - pd = uobj->object; - - if (atomic_read(&pd->usecnt)) { - ret = -EBUSY; - goto err_put; - } - - ret = pd->device->dealloc_pd(uobj->object); - WARN_ONCE(ret, "Infiniband HW driver failed dealloc_pd"); - if (ret) - goto err_put; - - ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); + uobj = uobj_get_write(uobj_get_type(pd), cmd.pd_handle, + file->ucontext); + if (IS_ERR(uobj)) + return PTR_ERR(uobj); - uobj->live = 0; - put_uobj_write(uobj); + ret = uobj_remove_commit(uobj); - idr_remove_uobj(&ib_uverbs_pd_idr, uobj); - - mutex_lock(&file->mutex); - list_del(&uobj->list); - mutex_unlock(&file->mutex); - - put_uobj(uobj); - - return in_len; - -err_put: - put_uobj_write(uobj); - return ret; + return ret ?: in_len; } struct xrcd_table_entry { @@ -808,16 +510,13 @@ ssize_t ib_uverbs_open_xrcd(struct ib_uverbs_file *file, } } - obj = kmalloc(sizeof *obj, GFP_KERNEL); - if (!obj) { - ret = -ENOMEM; + obj = (struct ib_uxrcd_object *)uobj_alloc(uobj_get_type(xrcd), + file->ucontext); + if (IS_ERR(obj)) { + ret = PTR_ERR(obj); goto err_tree_mutex_unlock; } - init_uobj(&obj->uobject, 0, file->ucontext, &xrcd_lock_class); - - down_write(&obj->uobject.mutex); - if (!xrcd) { xrcd = ib_dev->alloc_xrcd(ib_dev, file->ucontext, &udata); if (IS_ERR(xrcd)) { @@ -835,10 +534,6 @@ ssize_t ib_uverbs_open_xrcd(struct ib_uverbs_file *file, atomic_set(&obj->refcnt, 0); obj->uobject.object = xrcd; - ret = idr_add_uobj(&ib_uverbs_xrcd_idr, &obj->uobject); - if (ret) - goto err_idr; - memset(&resp, 0, sizeof resp); resp.xrcd_handle = obj->uobject.id; @@ -847,7 +542,7 @@ ssize_t ib_uverbs_open_xrcd(struct ib_uverbs_file *file, /* create new inode/xrcd table entry */ ret = xrcd_table_insert(file->device, inode, xrcd); if (ret) - goto err_insert_xrcd; + goto err_dealloc_xrcd; } atomic_inc(&xrcd->usecnt); } @@ -861,12 +556,7 @@ ssize_t ib_uverbs_open_xrcd(struct ib_uverbs_file *file, if (f.file) fdput(f); - mutex_lock(&file->mutex); - list_add_tail(&obj->uobject.list, &file->ucontext->xrcd_list); - mutex_unlock(&file->mutex); - - obj->uobject.live = 1; - up_write(&obj->uobject.mutex); + uobj_alloc_commit(&obj->uobject); mutex_unlock(&file->device->xrcd_tree_mutex); return in_len; @@ -878,14 +568,11 @@ err_copy: atomic_dec(&xrcd->usecnt); } -err_insert_xrcd: - idr_remove_uobj(&ib_uverbs_xrcd_idr, &obj->uobject); - -err_idr: +err_dealloc_xrcd: ib_dealloc_xrcd(xrcd); err: - put_uobj_write(&obj->uobject); + uobj_alloc_abort(&obj->uobject); err_tree_mutex_unlock: if (f.file) @@ -903,75 +590,41 @@ ssize_t ib_uverbs_close_xrcd(struct ib_uverbs_file *file, { struct ib_uverbs_close_xrcd cmd; struct ib_uobject *uobj; - struct ib_xrcd *xrcd = NULL; - struct inode *inode = NULL; - struct ib_uxrcd_object *obj; - int live; int ret = 0; if (copy_from_user(&cmd, buf, sizeof cmd)) return -EFAULT; - mutex_lock(&file->device->xrcd_tree_mutex); - uobj = idr_write_uobj(&ib_uverbs_xrcd_idr, cmd.xrcd_handle, file->ucontext); - if (!uobj) { - ret = -EINVAL; - goto out; + uobj = uobj_get_write(uobj_get_type(xrcd), cmd.xrcd_handle, + file->ucontext); + if (IS_ERR(uobj)) { + mutex_unlock(&file->device->xrcd_tree_mutex); + return PTR_ERR(uobj); } - xrcd = uobj->object; - inode = xrcd->inode; - obj = container_of(uobj, struct ib_uxrcd_object, uobject); - if (atomic_read(&obj->refcnt)) { - put_uobj_write(uobj); - ret = -EBUSY; - goto out; - } - - if (!inode || atomic_dec_and_test(&xrcd->usecnt)) { - ret = ib_dealloc_xrcd(uobj->object); - if (!ret) - uobj->live = 0; - } - - live = uobj->live; - if (inode && ret) - atomic_inc(&xrcd->usecnt); - - put_uobj_write(uobj); - - if (ret) - goto out; - - if (inode && !live) - xrcd_table_delete(file->device, inode); - - idr_remove_uobj(&ib_uverbs_xrcd_idr, uobj); - mutex_lock(&file->mutex); - list_del(&uobj->list); - mutex_unlock(&file->mutex); - - put_uobj(uobj); - ret = in_len; - -out: - mutex_unlock(&file->device->xrcd_tree_mutex); - return ret; + ret = uobj_remove_commit(uobj); + return ret ?: in_len; } -void ib_uverbs_dealloc_xrcd(struct ib_uverbs_device *dev, - struct ib_xrcd *xrcd) +int ib_uverbs_dealloc_xrcd(struct ib_uverbs_device *dev, + struct ib_xrcd *xrcd, + enum rdma_remove_reason why) { struct inode *inode; + int ret; inode = xrcd->inode; if (inode && !atomic_dec_and_test(&xrcd->usecnt)) - return; + return 0; - ib_dealloc_xrcd(xrcd); + ret = ib_dealloc_xrcd(xrcd); - if (inode) + if (why == RDMA_REMOVE_DESTROY && ret) + atomic_inc(&xrcd->usecnt); + else if (inode) xrcd_table_delete(dev, inode); + + return ret; } ssize_t ib_uverbs_reg_mr(struct ib_uverbs_file *file, @@ -1004,14 +657,11 @@ ssize_t ib_uverbs_reg_mr(struct ib_uverbs_file *file, if (ret) return ret; - uobj = kmalloc(sizeof *uobj, GFP_KERNEL); - if (!uobj) - return -ENOMEM; - - init_uobj(uobj, 0, file->ucontext, &mr_lock_class); - down_write(&uobj->mutex); + uobj = uobj_alloc(uobj_get_type(mr), file->ucontext); + if (IS_ERR(uobj)) + return PTR_ERR(uobj); - pd = idr_read_pd(cmd.pd_handle, file->ucontext); + pd = uobj_get_obj_read(pd, cmd.pd_handle, file->ucontext); if (!pd) { ret = -EINVAL; goto err_free; @@ -1025,10 +675,6 @@ ssize_t ib_uverbs_reg_mr(struct ib_uverbs_file *file, goto err_put; } } - ret = ib_rdmacg_try_charge(&uobj->cg_obj, ib_dev, - RDMACG_RESOURCE_HCA_OBJECT); - if (ret) - goto err_charge; mr = pd->device->reg_user_mr(pd, cmd.start, cmd.length, cmd.hca_va, cmd.access_flags, &udata); @@ -1043,9 +689,6 @@ ssize_t ib_uverbs_reg_mr(struct ib_uverbs_file *file, atomic_inc(&pd->usecnt); uobj->object = mr; - ret = idr_add_uobj(&ib_uverbs_mr_idr, uobj); - if (ret) - goto err_unreg; memset(&resp, 0, sizeof resp); resp.lkey = mr->lkey; @@ -1058,32 +701,20 @@ ssize_t ib_uverbs_reg_mr(struct ib_uverbs_file *file, goto err_copy; } - put_pd_read(pd); - - mutex_lock(&file->mutex); - list_add_tail(&uobj->list, &file->ucontext->mr_list); - mutex_unlock(&file->mutex); + uobj_put_obj_read(pd); - uobj->live = 1; - - up_write(&uobj->mutex); + uobj_alloc_commit(uobj); return in_len; err_copy: - idr_remove_uobj(&ib_uverbs_mr_idr, uobj); - -err_unreg: ib_dereg_mr(mr); err_put: - ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); - -err_charge: - put_pd_read(pd); + uobj_put_obj_read(pd); err_free: - put_uobj_write(uobj); + uobj_alloc_abort(uobj); return ret; } @@ -1119,11 +750,10 @@ ssize_t ib_uverbs_rereg_mr(struct ib_uverbs_file *file, (cmd.start & ~PAGE_MASK) != (cmd.hca_va & ~PAGE_MASK))) return -EINVAL; - uobj = idr_write_uobj(&ib_uverbs_mr_idr, cmd.mr_handle, - file->ucontext); - - if (!uobj) - return -EINVAL; + uobj = uobj_get_write(uobj_get_type(mr), cmd.mr_handle, + file->ucontext); + if (IS_ERR(uobj)) + return PTR_ERR(uobj); mr = uobj->object; @@ -1134,7 +764,7 @@ ssize_t ib_uverbs_rereg_mr(struct ib_uverbs_file *file, } if (cmd.flags & IB_MR_REREG_PD) { - pd = idr_read_pd(cmd.pd_handle, file->ucontext); + pd = uobj_get_obj_read(pd, cmd.pd_handle, file->ucontext); if (!pd) { ret = -EINVAL; goto put_uobjs; @@ -1167,11 +797,10 @@ ssize_t ib_uverbs_rereg_mr(struct ib_uverbs_file *file, put_uobj_pd: if (cmd.flags & IB_MR_REREG_PD) - put_pd_read(pd); + uobj_put_obj_read(pd); put_uobjs: - - put_uobj_write(mr->uobject); + uobj_put_write(uobj); return ret; } @@ -1182,39 +811,20 @@ ssize_t ib_uverbs_dereg_mr(struct ib_uverbs_file *file, int out_len) { struct ib_uverbs_dereg_mr cmd; - struct ib_mr *mr; struct ib_uobject *uobj; int ret = -EINVAL; if (copy_from_user(&cmd, buf, sizeof cmd)) return -EFAULT; - uobj = idr_write_uobj(&ib_uverbs_mr_idr, cmd.mr_handle, file->ucontext); - if (!uobj) - return -EINVAL; - - mr = uobj->object; - - ret = ib_dereg_mr(mr); - if (!ret) - uobj->live = 0; - - put_uobj_write(uobj); - - if (ret) - return ret; - - ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); + uobj = uobj_get_write(uobj_get_type(mr), cmd.mr_handle, + file->ucontext); + if (IS_ERR(uobj)) + return PTR_ERR(uobj); - idr_remove_uobj(&ib_uverbs_mr_idr, uobj); + ret = uobj_remove_commit(uobj); - mutex_lock(&file->mutex); - list_del(&uobj->list); - mutex_unlock(&file->mutex); - - put_uobj(uobj); - - return in_len; + return ret ?: in_len; } ssize_t ib_uverbs_alloc_mw(struct ib_uverbs_file *file, @@ -1236,14 +846,11 @@ ssize_t ib_uverbs_alloc_mw(struct ib_uverbs_file *file, if (copy_from_user(&cmd, buf, sizeof(cmd))) return -EFAULT; - uobj = kmalloc(sizeof(*uobj), GFP_KERNEL); - if (!uobj) - return -ENOMEM; - - init_uobj(uobj, 0, file->ucontext, &mw_lock_class); - down_write(&uobj->mutex); + uobj = uobj_alloc(uobj_get_type(mw), file->ucontext); + if (IS_ERR(uobj)) + return PTR_ERR(uobj); - pd = idr_read_pd(cmd.pd_handle, file->ucontext); + pd = uobj_get_obj_read(pd, cmd.pd_handle, file->ucontext); if (!pd) { ret = -EINVAL; goto err_free; @@ -1254,11 +861,6 @@ ssize_t ib_uverbs_alloc_mw(struct ib_uverbs_file *file, in_len - sizeof(cmd) - sizeof(struct ib_uverbs_cmd_hdr), out_len - sizeof(resp)); - ret = ib_rdmacg_try_charge(&uobj->cg_obj, ib_dev, - RDMACG_RESOURCE_HCA_OBJECT); - if (ret) - goto err_charge; - mw = pd->device->alloc_mw(pd, cmd.mw_type, &udata); if (IS_ERR(mw)) { ret = PTR_ERR(mw); @@ -1271,9 +873,6 @@ ssize_t ib_uverbs_alloc_mw(struct ib_uverbs_file *file, atomic_inc(&pd->usecnt); uobj->object = mw; - ret = idr_add_uobj(&ib_uverbs_mw_idr, uobj); - if (ret) - goto err_unalloc; memset(&resp, 0, sizeof(resp)); resp.rkey = mw->rkey; @@ -1285,32 +884,17 @@ ssize_t ib_uverbs_alloc_mw(struct ib_uverbs_file *file, goto err_copy; } - put_pd_read(pd); - - mutex_lock(&file->mutex); - list_add_tail(&uobj->list, &file->ucontext->mw_list); - mutex_unlock(&file->mutex); - - uobj->live = 1; - - up_write(&uobj->mutex); + uobj_put_obj_read(pd); + uobj_alloc_commit(uobj); return in_len; err_copy: - idr_remove_uobj(&ib_uverbs_mw_idr, uobj); - -err_unalloc: uverbs_dealloc_mw(mw); - err_put: - ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); - -err_charge: - put_pd_read(pd); - + uobj_put_obj_read(pd); err_free: - put_uobj_write(uobj); + uobj_alloc_abort(uobj); return ret; } @@ -1320,39 +904,19 @@ ssize_t ib_uverbs_dealloc_mw(struct ib_uverbs_file *file, int out_len) { struct ib_uverbs_dealloc_mw cmd; - struct ib_mw *mw; struct ib_uobject *uobj; int ret = -EINVAL; if (copy_from_user(&cmd, buf, sizeof(cmd))) return -EFAULT; - uobj = idr_write_uobj(&ib_uverbs_mw_idr, cmd.mw_handle, file->ucontext); - if (!uobj) - return -EINVAL; - - mw = uobj->object; - - ret = uverbs_dealloc_mw(mw); - if (!ret) - uobj->live = 0; - - put_uobj_write(uobj); - - if (ret) - return ret; - - ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); - - idr_remove_uobj(&ib_uverbs_mw_idr, uobj); - - mutex_lock(&file->mutex); - list_del(&uobj->list); - mutex_unlock(&file->mutex); - - put_uobj(uobj); + uobj = uobj_get_write(uobj_get_type(mw), cmd.mw_handle, + file->ucontext); + if (IS_ERR(uobj)) + return PTR_ERR(uobj); - return in_len; + ret = uobj_remove_commit(uobj); + return ret ?: in_len; } ssize_t ib_uverbs_create_comp_channel(struct ib_uverbs_file *file, @@ -1362,8 +926,8 @@ ssize_t ib_uverbs_create_comp_channel(struct ib_uverbs_file *file, { struct ib_uverbs_create_comp_channel cmd; struct ib_uverbs_create_comp_channel_resp resp; - struct file *filp; - int ret; + struct ib_uobject *uobj; + struct ib_uverbs_completion_event_file *ev_file; if (out_len < sizeof resp) return -ENOSPC; @@ -1371,25 +935,23 @@ ssize_t ib_uverbs_create_comp_channel(struct ib_uverbs_file *file, if (copy_from_user(&cmd, buf, sizeof cmd)) return -EFAULT; - ret = get_unused_fd_flags(O_CLOEXEC); - if (ret < 0) - return ret; - resp.fd = ret; + uobj = uobj_alloc(uobj_get_type(comp_channel), file->ucontext); + if (IS_ERR(uobj)) + return PTR_ERR(uobj); - filp = ib_uverbs_alloc_event_file(file, ib_dev, 0); - if (IS_ERR(filp)) { - put_unused_fd(resp.fd); - return PTR_ERR(filp); - } + resp.fd = uobj->id; + + ev_file = container_of(uobj, struct ib_uverbs_completion_event_file, + uobj_file.uobj); + ib_uverbs_init_event_queue(&ev_file->ev_queue); if (copy_to_user((void __user *) (unsigned long) cmd.response, &resp, sizeof resp)) { - put_unused_fd(resp.fd); - fput(filp); + uobj_alloc_abort(uobj); return -EFAULT; } - fd_install(resp.fd, filp); + uobj_alloc_commit(uobj); return in_len; } @@ -1407,7 +969,7 @@ static struct ib_ucq_object *create_cq(struct ib_uverbs_file *file, void *context) { struct ib_ucq_object *obj; - struct ib_uverbs_event_file *ev_file = NULL; + struct ib_uverbs_completion_event_file *ev_file = NULL; struct ib_cq *cq; int ret; struct ib_uverbs_ex_create_cq_resp resp; @@ -1416,21 +978,21 @@ static struct ib_ucq_object *create_cq(struct ib_uverbs_file *file, if (cmd->comp_vector >= file->device->num_comp_vectors) return ERR_PTR(-EINVAL); - obj = kmalloc(sizeof *obj, GFP_KERNEL); - if (!obj) - return ERR_PTR(-ENOMEM); - - init_uobj(&obj->uobject, cmd->user_handle, file->ucontext, &cq_lock_class); - down_write(&obj->uobject.mutex); + obj = (struct ib_ucq_object *)uobj_alloc(uobj_get_type(cq), + file->ucontext); + if (IS_ERR(obj)) + return obj; if (cmd->comp_channel >= 0) { - ev_file = ib_uverbs_lookup_comp_file(cmd->comp_channel); - if (!ev_file) { - ret = -EINVAL; + ev_file = ib_uverbs_lookup_comp_file(cmd->comp_channel, + file->ucontext); + if (IS_ERR(ev_file)) { + ret = PTR_ERR(ev_file); goto err; } } + obj->uobject.user_handle = cmd->user_handle; obj->uverbs_file = file; obj->comp_events_reported = 0; obj->async_events_reported = 0; @@ -1443,13 +1005,7 @@ static struct ib_ucq_object *create_cq(struct ib_uverbs_file *file, if (cmd_sz > offsetof(typeof(*cmd), flags) + sizeof(cmd->flags)) attr.flags = cmd->flags; - ret = ib_rdmacg_try_charge(&obj->uobject.cg_obj, ib_dev, - RDMACG_RESOURCE_HCA_OBJECT); - if (ret) - goto err_charge; - - cq = ib_dev->create_cq(ib_dev, &attr, - file->ucontext, uhw); + cq = ib_dev->create_cq(ib_dev, &attr, file->ucontext, uhw); if (IS_ERR(cq)) { ret = PTR_ERR(cq); goto err_file; @@ -1459,14 +1015,10 @@ static struct ib_ucq_object *create_cq(struct ib_uverbs_file *file, cq->uobject = &obj->uobject; cq->comp_handler = ib_uverbs_comp_handler; cq->event_handler = ib_uverbs_cq_event_handler; - cq->cq_context = ev_file; + cq->cq_context = &ev_file->ev_queue; atomic_set(&cq->usecnt, 0); obj->uobject.object = cq; - ret = idr_add_uobj(&ib_uverbs_cq_idr, &obj->uobject); - if (ret) - goto err_free; - memset(&resp, 0, sizeof resp); resp.base.cq_handle = obj->uobject.id; resp.base.cqe = cq->cqe; @@ -1478,32 +1030,19 @@ static struct ib_ucq_object *create_cq(struct ib_uverbs_file *file, if (ret) goto err_cb; - mutex_lock(&file->mutex); - list_add_tail(&obj->uobject.list, &file->ucontext->cq_list); - mutex_unlock(&file->mutex); - - obj->uobject.live = 1; - - up_write(&obj->uobject.mutex); + uobj_alloc_commit(&obj->uobject); return obj; err_cb: - idr_remove_uobj(&ib_uverbs_cq_idr, &obj->uobject); - -err_free: ib_destroy_cq(cq); err_file: - ib_rdmacg_uncharge(&obj->uobject.cg_obj, ib_dev, - RDMACG_RESOURCE_HCA_OBJECT); - -err_charge: if (ev_file) ib_uverbs_release_ucq(file, ev_file, obj); err: - put_uobj_write(&obj->uobject); + uobj_alloc_abort(&obj->uobject); return ERR_PTR(ret); } @@ -1626,7 +1165,7 @@ ssize_t ib_uverbs_resize_cq(struct ib_uverbs_file *file, (unsigned long) cmd.response + sizeof resp, in_len - sizeof cmd, out_len - sizeof resp); - cq = idr_read_cq(cmd.cq_handle, file->ucontext, 0); + cq = uobj_get_obj_read(cq, cmd.cq_handle, file->ucontext); if (!cq) return -EINVAL; @@ -1641,7 +1180,7 @@ ssize_t ib_uverbs_resize_cq(struct ib_uverbs_file *file, ret = -EFAULT; out: - put_cq_read(cq); + uobj_put_obj_read(cq); return ret ? ret : in_len; } @@ -1688,7 +1227,7 @@ ssize_t ib_uverbs_poll_cq(struct ib_uverbs_file *file, if (copy_from_user(&cmd, buf, sizeof cmd)) return -EFAULT; - cq = idr_read_cq(cmd.cq_handle, file->ucontext, 0); + cq = uobj_get_obj_read(cq, cmd.cq_handle, file->ucontext); if (!cq) return -EINVAL; @@ -1720,7 +1259,7 @@ ssize_t ib_uverbs_poll_cq(struct ib_uverbs_file *file, ret = in_len; out_put: - put_cq_read(cq); + uobj_put_obj_read(cq); return ret; } @@ -1735,14 +1274,14 @@ ssize_t ib_uverbs_req_notify_cq(struct ib_uverbs_file *file, if (copy_from_user(&cmd, buf, sizeof cmd)) return -EFAULT; - cq = idr_read_cq(cmd.cq_handle, file->ucontext, 0); + cq = uobj_get_obj_read(cq, cmd.cq_handle, file->ucontext); if (!cq) return -EINVAL; ib_req_notify_cq(cq, cmd.solicited_only ? IB_CQ_SOLICITED : IB_CQ_NEXT_COMP); - put_cq_read(cq); + uobj_put_obj_read(cq); return in_len; } @@ -1757,44 +1296,38 @@ ssize_t ib_uverbs_destroy_cq(struct ib_uverbs_file *file, struct ib_uobject *uobj; struct ib_cq *cq; struct ib_ucq_object *obj; - struct ib_uverbs_event_file *ev_file; + struct ib_uverbs_event_queue *ev_queue; int ret = -EINVAL; if (copy_from_user(&cmd, buf, sizeof cmd)) return -EFAULT; - uobj = idr_write_uobj(&ib_uverbs_cq_idr, cmd.cq_handle, file->ucontext); - if (!uobj) - return -EINVAL; + uobj = uobj_get_write(uobj_get_type(cq), cmd.cq_handle, + file->ucontext); + if (IS_ERR(uobj)) + return PTR_ERR(uobj); + + /* + * Make sure we don't free the memory in remove_commit as we still + * needs the uobject memory to create the response. + */ + uverbs_uobject_get(uobj); cq = uobj->object; - ev_file = cq->cq_context; + ev_queue = cq->cq_context; obj = container_of(cq->uobject, struct ib_ucq_object, uobject); - ret = ib_destroy_cq(cq); - if (!ret) - uobj->live = 0; - - put_uobj_write(uobj); + memset(&resp, 0, sizeof(resp)); - if (ret) + ret = uobj_remove_commit(uobj); + if (ret) { + uverbs_uobject_put(uobj); return ret; + } - ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); - - idr_remove_uobj(&ib_uverbs_cq_idr, uobj); - - mutex_lock(&file->mutex); - list_del(&uobj->list); - mutex_unlock(&file->mutex); - - ib_uverbs_release_ucq(file, ev_file, obj); - - memset(&resp, 0, sizeof resp); resp.comp_events_reported = obj->comp_events_reported; resp.async_events_reported = obj->async_events_reported; - put_uobj(uobj); - + uverbs_uobject_put(uobj); if (copy_to_user((void __user *) (unsigned long) cmd.response, &resp, sizeof resp)) return -EFAULT; @@ -1816,7 +1349,7 @@ static int create_qp(struct ib_uverbs_file *file, struct ib_device *device; struct ib_pd *pd = NULL; struct ib_xrcd *xrcd = NULL; - struct ib_uobject *uninitialized_var(xrcd_uobj); + struct ib_uobject *xrcd_uobj = ERR_PTR(-ENOENT); struct ib_cq *scq = NULL, *rcq = NULL; struct ib_srq *srq = NULL; struct ib_qp *qp; @@ -1830,18 +1363,20 @@ static int create_qp(struct ib_uverbs_file *file, if (cmd->qp_type == IB_QPT_RAW_PACKET && !capable(CAP_NET_RAW)) return -EPERM; - obj = kzalloc(sizeof *obj, GFP_KERNEL); - if (!obj) - return -ENOMEM; + obj = (struct ib_uqp_object *)uobj_alloc(uobj_get_type(qp), + file->ucontext); + if (IS_ERR(obj)) + return PTR_ERR(obj); + obj->uxrcd = NULL; + obj->uevent.uobject.user_handle = cmd->user_handle; + mutex_init(&obj->mcast_lock); - init_uobj(&obj->uevent.uobject, cmd->user_handle, file->ucontext, - &qp_lock_class); - down_write(&obj->uevent.uobject.mutex); if (cmd_sz >= offsetof(typeof(*cmd), rwq_ind_tbl_handle) + sizeof(cmd->rwq_ind_tbl_handle) && (cmd->comp_mask & IB_UVERBS_CREATE_QP_MASK_IND_TABLE)) { - ind_tbl = idr_read_rwq_indirection_table(cmd->rwq_ind_tbl_handle, - file->ucontext); + ind_tbl = uobj_get_obj_read(rwq_ind_table, + cmd->rwq_ind_tbl_handle, + file->ucontext); if (!ind_tbl) { ret = -EINVAL; goto err_put; @@ -1865,8 +1400,15 @@ static int create_qp(struct ib_uverbs_file *file, has_sq = false; if (cmd->qp_type == IB_QPT_XRC_TGT) { - xrcd = idr_read_xrcd(cmd->pd_handle, file->ucontext, - &xrcd_uobj); + xrcd_uobj = uobj_get_read(uobj_get_type(xrcd), cmd->pd_handle, + file->ucontext); + + if (IS_ERR(xrcd_uobj)) { + ret = -EINVAL; + goto err_put; + } + + xrcd = (struct ib_xrcd *)xrcd_uobj->object; if (!xrcd) { ret = -EINVAL; goto err_put; @@ -1878,8 +1420,8 @@ static int create_qp(struct ib_uverbs_file *file, cmd->max_recv_sge = 0; } else { if (cmd->is_srq) { - srq = idr_read_srq(cmd->srq_handle, - file->ucontext); + srq = uobj_get_obj_read(srq, cmd->srq_handle, + file->ucontext); if (!srq || srq->srq_type != IB_SRQT_BASIC) { ret = -EINVAL; goto err_put; @@ -1888,8 +1430,8 @@ static int create_qp(struct ib_uverbs_file *file, if (!ind_tbl) { if (cmd->recv_cq_handle != cmd->send_cq_handle) { - rcq = idr_read_cq(cmd->recv_cq_handle, - file->ucontext, 0); + rcq = uobj_get_obj_read(cq, cmd->recv_cq_handle, + file->ucontext); if (!rcq) { ret = -EINVAL; goto err_put; @@ -1899,10 +1441,11 @@ static int create_qp(struct ib_uverbs_file *file, } if (has_sq) - scq = idr_read_cq(cmd->send_cq_handle, file->ucontext, !!rcq); + scq = uobj_get_obj_read(cq, cmd->send_cq_handle, + file->ucontext); if (!ind_tbl) rcq = rcq ?: scq; - pd = idr_read_pd(cmd->pd_handle, file->ucontext); + pd = uobj_get_obj_read(pd, cmd->pd_handle, file->ucontext); if (!pd || (!scq && has_sq)) { ret = -EINVAL; goto err_put; @@ -1954,11 +1497,6 @@ static int create_qp(struct ib_uverbs_file *file, goto err_put; } - ret = ib_rdmacg_try_charge(&obj->uevent.uobject.cg_obj, device, - RDMACG_RESOURCE_HCA_OBJECT); - if (ret) - goto err_put; - if (cmd->qp_type == IB_QPT_XRC_TGT) qp = ib_create_qp(pd, &attr); else @@ -1966,7 +1504,7 @@ static int create_qp(struct ib_uverbs_file *file, if (IS_ERR(qp)) { ret = PTR_ERR(qp); - goto err_create; + goto err_put; } if (cmd->qp_type != IB_QPT_XRC_TGT) { @@ -1994,9 +1532,6 @@ static int create_qp(struct ib_uverbs_file *file, qp->uobject = &obj->uevent.uobject; obj->uevent.uobject.object = qp; - ret = idr_add_uobj(&ib_uverbs_qp_idr, &obj->uevent.uobject); - if (ret) - goto err_destroy; memset(&resp, 0, sizeof resp); resp.base.qpn = qp->qp_num; @@ -2018,54 +1553,41 @@ static int create_qp(struct ib_uverbs_file *file, obj->uxrcd = container_of(xrcd_uobj, struct ib_uxrcd_object, uobject); atomic_inc(&obj->uxrcd->refcnt); - put_xrcd_read(xrcd_uobj); + uobj_put_read(xrcd_uobj); } if (pd) - put_pd_read(pd); + uobj_put_obj_read(pd); if (scq) - put_cq_read(scq); + uobj_put_obj_read(scq); if (rcq && rcq != scq) - put_cq_read(rcq); + uobj_put_obj_read(rcq); if (srq) - put_srq_read(srq); + uobj_put_obj_read(srq); if (ind_tbl) - put_rwq_indirection_table_read(ind_tbl); - - mutex_lock(&file->mutex); - list_add_tail(&obj->uevent.uobject.list, &file->ucontext->qp_list); - mutex_unlock(&file->mutex); + uobj_put_obj_read(ind_tbl); - obj->uevent.uobject.live = 1; - - up_write(&obj->uevent.uobject.mutex); + uobj_alloc_commit(&obj->uevent.uobject); return 0; err_cb: - idr_remove_uobj(&ib_uverbs_qp_idr, &obj->uevent.uobject); - -err_destroy: ib_destroy_qp(qp); -err_create: - ib_rdmacg_uncharge(&obj->uevent.uobject.cg_obj, device, - RDMACG_RESOURCE_HCA_OBJECT); - err_put: - if (xrcd) - put_xrcd_read(xrcd_uobj); + if (!IS_ERR(xrcd_uobj)) + uobj_put_read(xrcd_uobj); if (pd) - put_pd_read(pd); + uobj_put_obj_read(pd); if (scq) - put_cq_read(scq); + uobj_put_obj_read(scq); if (rcq && rcq != scq) - put_cq_read(rcq); + uobj_put_obj_read(rcq); if (srq) - put_srq_read(srq); + uobj_put_obj_read(srq); if (ind_tbl) - put_rwq_indirection_table_read(ind_tbl); + uobj_put_obj_read(ind_tbl); - put_uobj_write(&obj->uevent.uobject); + uobj_alloc_abort(&obj->uevent.uobject); return ret; } @@ -2201,17 +1723,22 @@ ssize_t ib_uverbs_open_qp(struct ib_uverbs_file *file, (unsigned long) cmd.response + sizeof resp, in_len - sizeof cmd, out_len - sizeof resp); - obj = kmalloc(sizeof *obj, GFP_KERNEL); - if (!obj) - return -ENOMEM; + obj = (struct ib_uqp_object *)uobj_alloc(uobj_get_type(qp), + file->ucontext); + if (IS_ERR(obj)) + return PTR_ERR(obj); - init_uobj(&obj->uevent.uobject, cmd.user_handle, file->ucontext, &qp_lock_class); - down_write(&obj->uevent.uobject.mutex); + xrcd_uobj = uobj_get_read(uobj_get_type(xrcd), cmd.pd_handle, + file->ucontext); + if (IS_ERR(xrcd_uobj)) { + ret = -EINVAL; + goto err_put; + } - xrcd = idr_read_xrcd(cmd.pd_handle, file->ucontext, &xrcd_uobj); + xrcd = (struct ib_xrcd *)xrcd_uobj->object; if (!xrcd) { ret = -EINVAL; - goto err_put; + goto err_xrcd; } attr.event_handler = ib_uverbs_qp_event_handler; @@ -2226,15 +1753,11 @@ ssize_t ib_uverbs_open_qp(struct ib_uverbs_file *file, qp = ib_open_qp(xrcd, &attr); if (IS_ERR(qp)) { ret = PTR_ERR(qp); - goto err_put; + goto err_xrcd; } - qp->uobject = &obj->uevent.uobject; - obj->uevent.uobject.object = qp; - ret = idr_add_uobj(&ib_uverbs_qp_idr, &obj->uevent.uobject); - if (ret) - goto err_destroy; + obj->uevent.uobject.user_handle = cmd.user_handle; memset(&resp, 0, sizeof resp); resp.qpn = qp->qp_num; @@ -2243,32 +1766,25 @@ ssize_t ib_uverbs_open_qp(struct ib_uverbs_file *file, if (copy_to_user((void __user *) (unsigned long) cmd.response, &resp, sizeof resp)) { ret = -EFAULT; - goto err_remove; + goto err_destroy; } obj->uxrcd = container_of(xrcd_uobj, struct ib_uxrcd_object, uobject); atomic_inc(&obj->uxrcd->refcnt); - put_xrcd_read(xrcd_uobj); - - mutex_lock(&file->mutex); - list_add_tail(&obj->uevent.uobject.list, &file->ucontext->qp_list); - mutex_unlock(&file->mutex); + qp->uobject = &obj->uevent.uobject; + uobj_put_read(xrcd_uobj); - obj->uevent.uobject.live = 1; - up_write(&obj->uevent.uobject.mutex); + uobj_alloc_commit(&obj->uevent.uobject); return in_len; -err_remove: - idr_remove_uobj(&ib_uverbs_qp_idr, &obj->uevent.uobject); - err_destroy: ib_destroy_qp(qp); - +err_xrcd: + uobj_put_read(xrcd_uobj); err_put: - put_xrcd_read(xrcd_uobj); - put_uobj_write(&obj->uevent.uobject); + uobj_alloc_abort(&obj->uevent.uobject); return ret; } @@ -2294,7 +1810,7 @@ ssize_t ib_uverbs_query_qp(struct ib_uverbs_file *file, goto out; } - qp = idr_read_qp(cmd.qp_handle, file->ucontext); + qp = uobj_get_obj_read(qp, cmd.qp_handle, file->ucontext); if (!qp) { ret = -EINVAL; goto out; @@ -2302,7 +1818,7 @@ ssize_t ib_uverbs_query_qp(struct ib_uverbs_file *file, ret = ib_query_qp(qp, attr, cmd.attr_mask, init_attr); - put_qp_read(qp); + uobj_put_obj_read(qp); if (ret) goto out; @@ -2398,7 +1914,7 @@ static int modify_qp(struct ib_uverbs_file *file, if (!attr) return -ENOMEM; - qp = idr_read_qp(cmd->base.qp_handle, file->ucontext); + qp = uobj_get_obj_read(qp, cmd->base.qp_handle, file->ucontext); if (!qp) { ret = -EINVAL; goto out; @@ -2470,7 +1986,7 @@ static int modify_qp(struct ib_uverbs_file *file, } release_qp: - put_qp_read(qp); + uobj_put_obj_read(qp); out: kfree(attr); @@ -2557,42 +2073,27 @@ ssize_t ib_uverbs_destroy_qp(struct ib_uverbs_file *file, memset(&resp, 0, sizeof resp); - uobj = idr_write_uobj(&ib_uverbs_qp_idr, cmd.qp_handle, file->ucontext); - if (!uobj) - return -EINVAL; + uobj = uobj_get_write(uobj_get_type(qp), cmd.qp_handle, + file->ucontext); + if (IS_ERR(uobj)) + return PTR_ERR(uobj); + qp = uobj->object; obj = container_of(uobj, struct ib_uqp_object, uevent.uobject); + /* + * Make sure we don't free the memory in remove_commit as we still + * needs the uobject memory to create the response. + */ + uverbs_uobject_get(uobj); - if (!list_empty(&obj->mcast_list)) { - put_uobj_write(uobj); - return -EBUSY; - } - - ret = ib_destroy_qp(qp); - if (!ret) - uobj->live = 0; - - put_uobj_write(uobj); - - if (ret) + ret = uobj_remove_commit(uobj); + if (ret) { + uverbs_uobject_put(uobj); return ret; - - ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); - - if (obj->uxrcd) - atomic_dec(&obj->uxrcd->refcnt); - - idr_remove_uobj(&ib_uverbs_qp_idr, uobj); - - mutex_lock(&file->mutex); - list_del(&uobj->list); - mutex_unlock(&file->mutex); - - ib_uverbs_release_uevent(file, &obj->uevent); + } resp.events_reported = obj->uevent.events_reported; - - put_uobj(uobj); + uverbs_uobject_put(uobj); if (copy_to_user((void __user *) (unsigned long) cmd.response, &resp, sizeof resp)) @@ -2603,9 +2104,13 @@ ssize_t ib_uverbs_destroy_qp(struct ib_uverbs_file *file, static void *alloc_wr(size_t wr_size, __u32 num_sge) { + if (num_sge >= (U32_MAX - ALIGN(wr_size, sizeof (struct ib_sge))) / + sizeof (struct ib_sge)) + return NULL; + return kmalloc(ALIGN(wr_size, sizeof (struct ib_sge)) + num_sge * sizeof (struct ib_sge), GFP_KERNEL); -}; +} ssize_t ib_uverbs_post_send(struct ib_uverbs_file *file, struct ib_device *ib_dev, @@ -2636,7 +2141,7 @@ ssize_t ib_uverbs_post_send(struct ib_uverbs_file *file, if (!user_wr) return -ENOMEM; - qp = idr_read_qp(cmd.qp_handle, file->ucontext); + qp = uobj_get_obj_read(qp, cmd.qp_handle, file->ucontext); if (!qp) goto out; @@ -2672,7 +2177,8 @@ ssize_t ib_uverbs_post_send(struct ib_uverbs_file *file, goto out_put; } - ud->ah = idr_read_ah(user_wr->wr.ud.ah, file->ucontext); + ud->ah = uobj_get_obj_read(ah, user_wr->wr.ud.ah, + file->ucontext); if (!ud->ah) { kfree(ud); ret = -EINVAL; @@ -2779,11 +2285,11 @@ ssize_t ib_uverbs_post_send(struct ib_uverbs_file *file, ret = -EFAULT; out_put: - put_qp_read(qp); + uobj_put_obj_read(qp); while (wr) { if (is_ud && ud_wr(wr)->ah) - put_ah_read(ud_wr(wr)->ah); + uobj_put_obj_read(ud_wr(wr)->ah); next = wr->next; kfree(wr); wr = next; @@ -2832,6 +2338,13 @@ static struct ib_recv_wr *ib_uverbs_unmarshall_recv(const char __user *buf, goto err; } + if (user_wr->num_sge >= + (U32_MAX - ALIGN(sizeof *next, sizeof (struct ib_sge))) / + sizeof (struct ib_sge)) { + ret = -EINVAL; + goto err; + } + next = kmalloc(ALIGN(sizeof *next, sizeof (struct ib_sge)) + user_wr->num_sge * sizeof (struct ib_sge), GFP_KERNEL); @@ -2900,21 +2413,21 @@ ssize_t ib_uverbs_post_recv(struct ib_uverbs_file *file, if (IS_ERR(wr)) return PTR_ERR(wr); - qp = idr_read_qp(cmd.qp_handle, file->ucontext); + qp = uobj_get_obj_read(qp, cmd.qp_handle, file->ucontext); if (!qp) goto out; resp.bad_wr = 0; ret = qp->device->post_recv(qp->real_qp, wr, &bad_wr); - put_qp_read(qp); - - if (ret) + uobj_put_obj_read(qp); + if (ret) { for (next = wr; next; next = next->next) { ++resp.bad_wr; if (next == bad_wr) break; } + } if (copy_to_user((void __user *) (unsigned long) cmd.response, &resp, sizeof resp)) @@ -2950,14 +2463,14 @@ ssize_t ib_uverbs_post_srq_recv(struct ib_uverbs_file *file, if (IS_ERR(wr)) return PTR_ERR(wr); - srq = idr_read_srq(cmd.srq_handle, file->ucontext); + srq = uobj_get_obj_read(srq, cmd.srq_handle, file->ucontext); if (!srq) goto out; resp.bad_wr = 0; ret = srq->device->post_srq_recv(srq, wr, &bad_wr); - put_srq_read(srq); + uobj_put_obj_read(srq); if (ret) for (next = wr; next; next = next->next) { @@ -3004,14 +2517,11 @@ ssize_t ib_uverbs_create_ah(struct ib_uverbs_file *file, (unsigned long)cmd.response + sizeof(resp), in_len - sizeof(cmd), out_len - sizeof(resp)); - uobj = kmalloc(sizeof *uobj, GFP_KERNEL); - if (!uobj) - return -ENOMEM; - - init_uobj(uobj, cmd.user_handle, file->ucontext, &ah_lock_class); - down_write(&uobj->mutex); + uobj = uobj_alloc(uobj_get_type(ah), file->ucontext); + if (IS_ERR(uobj)) + return PTR_ERR(uobj); - pd = idr_read_pd(cmd.pd_handle, file->ucontext); + pd = uobj_get_obj_read(pd, cmd.pd_handle, file->ucontext); if (!pd) { ret = -EINVAL; goto err; @@ -3030,28 +2540,20 @@ ssize_t ib_uverbs_create_ah(struct ib_uverbs_file *file, memset(&attr.dmac, 0, sizeof(attr.dmac)); memcpy(attr.grh.dgid.raw, cmd.attr.grh.dgid, 16); - ret = ib_rdmacg_try_charge(&uobj->cg_obj, ib_dev, - RDMACG_RESOURCE_HCA_OBJECT); - if (ret) - goto err_charge; - ah = pd->device->create_ah(pd, &attr, &udata); if (IS_ERR(ah)) { ret = PTR_ERR(ah); - goto err_create; + goto err_put; } ah->device = pd->device; ah->pd = pd; atomic_inc(&pd->usecnt); ah->uobject = uobj; + uobj->user_handle = cmd.user_handle; uobj->object = ah; - ret = idr_add_uobj(&ib_uverbs_ah_idr, uobj); - if (ret) - goto err_destroy; - resp.ah_handle = uobj->id; if (copy_to_user((void __user *) (unsigned long) cmd.response, @@ -3060,32 +2562,19 @@ ssize_t ib_uverbs_create_ah(struct ib_uverbs_file *file, goto err_copy; } - put_pd_read(pd); - - mutex_lock(&file->mutex); - list_add_tail(&uobj->list, &file->ucontext->ah_list); - mutex_unlock(&file->mutex); - - uobj->live = 1; - - up_write(&uobj->mutex); + uobj_put_obj_read(pd); + uobj_alloc_commit(uobj); return in_len; err_copy: - idr_remove_uobj(&ib_uverbs_ah_idr, uobj); - -err_destroy: ib_destroy_ah(ah); -err_create: - ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); - -err_charge: - put_pd_read(pd); +err_put: + uobj_put_obj_read(pd); err: - put_uobj_write(uobj); + uobj_alloc_abort(uobj); return ret; } @@ -3094,38 +2583,19 @@ ssize_t ib_uverbs_destroy_ah(struct ib_uverbs_file *file, const char __user *buf, int in_len, int out_len) { struct ib_uverbs_destroy_ah cmd; - struct ib_ah *ah; struct ib_uobject *uobj; int ret; if (copy_from_user(&cmd, buf, sizeof cmd)) return -EFAULT; - uobj = idr_write_uobj(&ib_uverbs_ah_idr, cmd.ah_handle, file->ucontext); - if (!uobj) - return -EINVAL; - ah = uobj->object; - - ret = ib_destroy_ah(ah); - if (!ret) - uobj->live = 0; - - put_uobj_write(uobj); - - if (ret) - return ret; - - ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); - - idr_remove_uobj(&ib_uverbs_ah_idr, uobj); + uobj = uobj_get_write(uobj_get_type(ah), cmd.ah_handle, + file->ucontext); + if (IS_ERR(uobj)) + return PTR_ERR(uobj); - mutex_lock(&file->mutex); - list_del(&uobj->list); - mutex_unlock(&file->mutex); - - put_uobj(uobj); - - return in_len; + ret = uobj_remove_commit(uobj); + return ret ?: in_len; } ssize_t ib_uverbs_attach_mcast(struct ib_uverbs_file *file, @@ -3142,12 +2612,13 @@ ssize_t ib_uverbs_attach_mcast(struct ib_uverbs_file *file, if (copy_from_user(&cmd, buf, sizeof cmd)) return -EFAULT; - qp = idr_write_qp(cmd.qp_handle, file->ucontext); + qp = uobj_get_obj_read(qp, cmd.qp_handle, file->ucontext); if (!qp) return -EINVAL; obj = container_of(qp->uobject, struct ib_uqp_object, uevent.uobject); + mutex_lock(&obj->mcast_lock); list_for_each_entry(mcast, &obj->mcast_list, list) if (cmd.mlid == mcast->lid && !memcmp(cmd.gid, mcast->gid.raw, sizeof mcast->gid.raw)) { @@ -3171,7 +2642,8 @@ ssize_t ib_uverbs_attach_mcast(struct ib_uverbs_file *file, kfree(mcast); out_put: - put_qp_write(qp); + mutex_unlock(&obj->mcast_lock); + uobj_put_obj_read(qp); return ret ? ret : in_len; } @@ -3186,31 +2658,37 @@ ssize_t ib_uverbs_detach_mcast(struct ib_uverbs_file *file, struct ib_qp *qp; struct ib_uverbs_mcast_entry *mcast; int ret = -EINVAL; + bool found = false; if (copy_from_user(&cmd, buf, sizeof cmd)) return -EFAULT; - qp = idr_write_qp(cmd.qp_handle, file->ucontext); + qp = uobj_get_obj_read(qp, cmd.qp_handle, file->ucontext); if (!qp) return -EINVAL; - ret = ib_detach_mcast(qp, (union ib_gid *) cmd.gid, cmd.mlid); - if (ret) - goto out_put; - obj = container_of(qp->uobject, struct ib_uqp_object, uevent.uobject); + mutex_lock(&obj->mcast_lock); list_for_each_entry(mcast, &obj->mcast_list, list) if (cmd.mlid == mcast->lid && !memcmp(cmd.gid, mcast->gid.raw, sizeof mcast->gid.raw)) { list_del(&mcast->list); kfree(mcast); + found = true; break; } -out_put: - put_qp_write(qp); + if (!found) { + ret = -EINVAL; + goto out_put; + } + + ret = ib_detach_mcast(qp, (union ib_gid *)cmd.gid, cmd.mlid); +out_put: + mutex_unlock(&obj->mcast_lock); + uobj_put_obj_read(qp); return ret ? ret : in_len; } @@ -3227,6 +2705,13 @@ static int kern_spec_to_ib_spec_action(struct ib_uverbs_flow_spec *kern_spec, ib_spec->flow_tag.size = sizeof(struct ib_flow_spec_action_tag); ib_spec->flow_tag.tag_id = kern_spec->flow_tag.tag_id; break; + case IB_FLOW_SPEC_ACTION_DROP: + if (kern_spec->drop.size != + sizeof(struct ib_uverbs_flow_spec_action_drop)) + return -EINVAL; + + ib_spec->drop.size = sizeof(struct ib_flow_spec_action_drop); + break; default: return -EINVAL; } @@ -3402,20 +2887,18 @@ int ib_uverbs_ex_create_wq(struct ib_uverbs_file *file, if (cmd.comp_mask) return -EOPNOTSUPP; - obj = kmalloc(sizeof(*obj), GFP_KERNEL); - if (!obj) - return -ENOMEM; + obj = (struct ib_uwq_object *)uobj_alloc(uobj_get_type(wq), + file->ucontext); + if (IS_ERR(obj)) + return PTR_ERR(obj); - init_uobj(&obj->uevent.uobject, cmd.user_handle, file->ucontext, - &wq_lock_class); - down_write(&obj->uevent.uobject.mutex); - pd = idr_read_pd(cmd.pd_handle, file->ucontext); + pd = uobj_get_obj_read(pd, cmd.pd_handle, file->ucontext); if (!pd) { err = -EINVAL; goto err_uobj; } - cq = idr_read_cq(cmd.cq_handle, file->ucontext, 0); + cq = uobj_get_obj_read(cq, cmd.cq_handle, file->ucontext); if (!cq) { err = -EINVAL; goto err_put_pd; @@ -3450,9 +2933,6 @@ int ib_uverbs_ex_create_wq(struct ib_uverbs_file *file, atomic_inc(&cq->usecnt); wq->uobject = &obj->uevent.uobject; obj->uevent.uobject.object = wq; - err = idr_add_uobj(&ib_uverbs_wq_idr, &obj->uevent.uobject); - if (err) - goto destroy_wq; memset(&resp, 0, sizeof(resp)); resp.wq_handle = obj->uevent.uobject.id; @@ -3465,27 +2945,19 @@ int ib_uverbs_ex_create_wq(struct ib_uverbs_file *file, if (err) goto err_copy; - put_pd_read(pd); - put_cq_read(cq); - - mutex_lock(&file->mutex); - list_add_tail(&obj->uevent.uobject.list, &file->ucontext->wq_list); - mutex_unlock(&file->mutex); - - obj->uevent.uobject.live = 1; - up_write(&obj->uevent.uobject.mutex); + uobj_put_obj_read(pd); + uobj_put_obj_read(cq); + uobj_alloc_commit(&obj->uevent.uobject); return 0; err_copy: - idr_remove_uobj(&ib_uverbs_wq_idr, &obj->uevent.uobject); -destroy_wq: ib_destroy_wq(wq); err_put_cq: - put_cq_read(cq); + uobj_put_obj_read(cq); err_put_pd: - put_pd_read(pd); + uobj_put_obj_read(pd); err_uobj: - put_uobj_write(&obj->uevent.uobject); + uobj_alloc_abort(&obj->uevent.uobject); return err; } @@ -3526,36 +2998,26 @@ int ib_uverbs_ex_destroy_wq(struct ib_uverbs_file *file, return -EOPNOTSUPP; resp.response_length = required_resp_len; - uobj = idr_write_uobj(&ib_uverbs_wq_idr, cmd.wq_handle, - file->ucontext); - if (!uobj) - return -EINVAL; + uobj = uobj_get_write(uobj_get_type(wq), cmd.wq_handle, + file->ucontext); + if (IS_ERR(uobj)) + return PTR_ERR(uobj); wq = uobj->object; obj = container_of(uobj, struct ib_uwq_object, uevent.uobject); - ret = ib_destroy_wq(wq); - if (!ret) - uobj->live = 0; - - put_uobj_write(uobj); - if (ret) - return ret; - - idr_remove_uobj(&ib_uverbs_wq_idr, uobj); - - mutex_lock(&file->mutex); - list_del(&uobj->list); - mutex_unlock(&file->mutex); + /* + * Make sure we don't free the memory in remove_commit as we still + * needs the uobject memory to create the response. + */ + uverbs_uobject_get(uobj); - ib_uverbs_release_uevent(file, &obj->uevent); + ret = uobj_remove_commit(uobj); resp.events_reported = obj->uevent.events_reported; - put_uobj(uobj); - - ret = ib_copy_to_udata(ucore, &resp, resp.response_length); + uverbs_uobject_put(uobj); if (ret) return ret; - return 0; + return ib_copy_to_udata(ucore, &resp, resp.response_length); } int ib_uverbs_ex_modify_wq(struct ib_uverbs_file *file, @@ -3588,7 +3050,7 @@ int ib_uverbs_ex_modify_wq(struct ib_uverbs_file *file, if (cmd.attr_mask > (IB_WQ_STATE | IB_WQ_CUR_STATE | IB_WQ_FLAGS)) return -EINVAL; - wq = idr_read_wq(cmd.wq_handle, file->ucontext); + wq = uobj_get_obj_read(wq, cmd.wq_handle, file->ucontext); if (!wq) return -EINVAL; @@ -3599,7 +3061,7 @@ int ib_uverbs_ex_modify_wq(struct ib_uverbs_file *file, wq_attr.flags_mask = cmd.flags_mask; } ret = wq->device->modify_wq(wq, &wq_attr, cmd.attr_mask, uhw); - put_wq_read(wq); + uobj_put_obj_read(wq); return ret; } @@ -3677,7 +3139,8 @@ int ib_uverbs_ex_create_rwq_ind_table(struct ib_uverbs_file *file, for (num_read_wqs = 0; num_read_wqs < num_wq_handles; num_read_wqs++) { - wq = idr_read_wq(wqs_handles[num_read_wqs], file->ucontext); + wq = uobj_get_obj_read(wq, wqs_handles[num_read_wqs], + file->ucontext); if (!wq) { err = -EINVAL; goto put_wqs; @@ -3686,14 +3149,12 @@ int ib_uverbs_ex_create_rwq_ind_table(struct ib_uverbs_file *file, wqs[num_read_wqs] = wq; } - uobj = kmalloc(sizeof(*uobj), GFP_KERNEL); - if (!uobj) { - err = -ENOMEM; + uobj = uobj_alloc(uobj_get_type(rwq_ind_table), file->ucontext); + if (IS_ERR(uobj)) { + err = PTR_ERR(uobj); goto put_wqs; } - init_uobj(uobj, 0, file->ucontext, &rwq_ind_table_lock_class); - down_write(&uobj->mutex); init_attr.log_ind_tbl_size = cmd.log_ind_tbl_size; init_attr.ind_tbl = wqs; rwq_ind_tbl = ib_dev->create_rwq_ind_table(ib_dev, &init_attr, uhw); @@ -3713,10 +3174,6 @@ int ib_uverbs_ex_create_rwq_ind_table(struct ib_uverbs_file *file, for (i = 0; i < num_wq_handles; i++) atomic_inc(&wqs[i]->usecnt); - err = idr_add_uobj(&ib_uverbs_rwq_ind_tbl_idr, uobj); - if (err) - goto destroy_ind_tbl; - resp.ind_tbl_handle = uobj->id; resp.ind_tbl_num = rwq_ind_tbl->ind_tbl_num; resp.response_length = required_resp_len; @@ -3729,26 +3186,18 @@ int ib_uverbs_ex_create_rwq_ind_table(struct ib_uverbs_file *file, kfree(wqs_handles); for (j = 0; j < num_read_wqs; j++) - put_wq_read(wqs[j]); - - mutex_lock(&file->mutex); - list_add_tail(&uobj->list, &file->ucontext->rwq_ind_tbl_list); - mutex_unlock(&file->mutex); - - uobj->live = 1; + uobj_put_obj_read(wqs[j]); - up_write(&uobj->mutex); + uobj_alloc_commit(uobj); return 0; err_copy: - idr_remove_uobj(&ib_uverbs_rwq_ind_tbl_idr, uobj); -destroy_ind_tbl: ib_destroy_rwq_ind_table(rwq_ind_tbl); err_uobj: - put_uobj_write(uobj); + uobj_alloc_abort(uobj); put_wqs: for (j = 0; j < num_read_wqs; j++) - put_wq_read(wqs[j]); + uobj_put_obj_read(wqs[j]); err_free: kfree(wqs_handles); kfree(wqs); @@ -3761,10 +3210,8 @@ int ib_uverbs_ex_destroy_rwq_ind_table(struct ib_uverbs_file *file, struct ib_udata *uhw) { struct ib_uverbs_ex_destroy_rwq_ind_table cmd = {}; - struct ib_rwq_ind_table *rwq_ind_tbl; struct ib_uobject *uobj; int ret; - struct ib_wq **ind_tbl; size_t required_cmd_sz; required_cmd_sz = offsetof(typeof(cmd), ind_tbl_handle) + sizeof(cmd.ind_tbl_handle); @@ -3784,31 +3231,12 @@ int ib_uverbs_ex_destroy_rwq_ind_table(struct ib_uverbs_file *file, if (cmd.comp_mask) return -EOPNOTSUPP; - uobj = idr_write_uobj(&ib_uverbs_rwq_ind_tbl_idr, cmd.ind_tbl_handle, - file->ucontext); - if (!uobj) - return -EINVAL; - rwq_ind_tbl = uobj->object; - ind_tbl = rwq_ind_tbl->ind_tbl; - - ret = ib_destroy_rwq_ind_table(rwq_ind_tbl); - if (!ret) - uobj->live = 0; - - put_uobj_write(uobj); - - if (ret) - return ret; - - idr_remove_uobj(&ib_uverbs_rwq_ind_tbl_idr, uobj); - - mutex_lock(&file->mutex); - list_del(&uobj->list); - mutex_unlock(&file->mutex); + uobj = uobj_get_write(uobj_get_type(rwq_ind_table), cmd.ind_tbl_handle, + file->ucontext); + if (IS_ERR(uobj)) + return PTR_ERR(uobj); - put_uobj(uobj); - kfree(ind_tbl); - return ret; + return uobj_remove_commit(uobj); } int ib_uverbs_ex_create_flow(struct ib_uverbs_file *file, @@ -3882,15 +3310,13 @@ int ib_uverbs_ex_create_flow(struct ib_uverbs_file *file, kern_flow_attr = &cmd.flow_attr; } - uobj = kmalloc(sizeof(*uobj), GFP_KERNEL); - if (!uobj) { - err = -ENOMEM; + uobj = uobj_alloc(uobj_get_type(flow), file->ucontext); + if (IS_ERR(uobj)) { + err = PTR_ERR(uobj); goto err_free_attr; } - init_uobj(uobj, 0, file->ucontext, &rule_lock_class); - down_write(&uobj->mutex); - qp = idr_read_qp(cmd.qp_handle, file->ucontext); + qp = uobj_get_obj_read(qp, cmd.qp_handle, file->ucontext); if (!qp) { err = -EINVAL; goto err_uobj; @@ -3931,24 +3357,14 @@ int ib_uverbs_ex_create_flow(struct ib_uverbs_file *file, err = -EINVAL; goto err_free; } - - err = ib_rdmacg_try_charge(&uobj->cg_obj, ib_dev, - RDMACG_RESOURCE_HCA_OBJECT); - if (err) - goto err_free; - flow_id = ib_create_flow(qp, flow_attr, IB_FLOW_DOMAIN_USER); if (IS_ERR(flow_id)) { err = PTR_ERR(flow_id); - goto err_create; + goto err_free; } flow_id->uobject = uobj; uobj->object = flow_id; - err = idr_add_uobj(&ib_uverbs_rule_idr, uobj); - if (err) - goto destroy_flow; - memset(&resp, 0, sizeof(resp)); resp.flow_handle = uobj->id; @@ -3957,30 +3373,20 @@ int ib_uverbs_ex_create_flow(struct ib_uverbs_file *file, if (err) goto err_copy; - put_qp_read(qp); - mutex_lock(&file->mutex); - list_add_tail(&uobj->list, &file->ucontext->rule_list); - mutex_unlock(&file->mutex); - - uobj->live = 1; - - up_write(&uobj->mutex); + uobj_put_obj_read(qp); + uobj_alloc_commit(uobj); kfree(flow_attr); if (cmd.flow_attr.num_of_specs) kfree(kern_flow_attr); return 0; err_copy: - idr_remove_uobj(&ib_uverbs_rule_idr, uobj); -destroy_flow: ib_destroy_flow(flow_id); -err_create: - ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); err_free: kfree(flow_attr); err_put: - put_qp_read(qp); + uobj_put_obj_read(qp); err_uobj: - put_uobj_write(uobj); + uobj_alloc_abort(uobj); err_free_attr: if (cmd.flow_attr.num_of_specs) kfree(kern_flow_attr); @@ -3993,7 +3399,6 @@ int ib_uverbs_ex_destroy_flow(struct ib_uverbs_file *file, struct ib_udata *uhw) { struct ib_uverbs_destroy_flow cmd; - struct ib_flow *flow_id; struct ib_uobject *uobj; int ret; @@ -4007,29 +3412,12 @@ int ib_uverbs_ex_destroy_flow(struct ib_uverbs_file *file, if (cmd.comp_mask) return -EINVAL; - uobj = idr_write_uobj(&ib_uverbs_rule_idr, cmd.flow_handle, - file->ucontext); - if (!uobj) - return -EINVAL; - flow_id = uobj->object; - - ret = ib_destroy_flow(flow_id); - if (!ret) { - ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, - RDMACG_RESOURCE_HCA_OBJECT); - uobj->live = 0; - } - - put_uobj_write(uobj); - - idr_remove_uobj(&ib_uverbs_rule_idr, uobj); - - mutex_lock(&file->mutex); - list_del(&uobj->list); - mutex_unlock(&file->mutex); - - put_uobj(uobj); + uobj = uobj_get_write(uobj_get_type(flow), cmd.flow_handle, + file->ucontext); + if (IS_ERR(uobj)) + return PTR_ERR(uobj); + ret = uobj_remove_commit(uobj); return ret; } @@ -4046,31 +3434,37 @@ static int __uverbs_create_xsrq(struct ib_uverbs_file *file, struct ib_srq_init_attr attr; int ret; - obj = kmalloc(sizeof *obj, GFP_KERNEL); - if (!obj) - return -ENOMEM; - - init_uobj(&obj->uevent.uobject, cmd->user_handle, file->ucontext, &srq_lock_class); - down_write(&obj->uevent.uobject.mutex); + obj = (struct ib_usrq_object *)uobj_alloc(uobj_get_type(srq), + file->ucontext); + if (IS_ERR(obj)) + return PTR_ERR(obj); if (cmd->srq_type == IB_SRQT_XRC) { - attr.ext.xrc.xrcd = idr_read_xrcd(cmd->xrcd_handle, file->ucontext, &xrcd_uobj); - if (!attr.ext.xrc.xrcd) { + xrcd_uobj = uobj_get_read(uobj_get_type(xrcd), cmd->xrcd_handle, + file->ucontext); + if (IS_ERR(xrcd_uobj)) { ret = -EINVAL; goto err; } + attr.ext.xrc.xrcd = (struct ib_xrcd *)xrcd_uobj->object; + if (!attr.ext.xrc.xrcd) { + ret = -EINVAL; + goto err_put_xrcd; + } + obj->uxrcd = container_of(xrcd_uobj, struct ib_uxrcd_object, uobject); atomic_inc(&obj->uxrcd->refcnt); - attr.ext.xrc.cq = idr_read_cq(cmd->cq_handle, file->ucontext, 0); + attr.ext.xrc.cq = uobj_get_obj_read(cq, cmd->cq_handle, + file->ucontext); if (!attr.ext.xrc.cq) { ret = -EINVAL; goto err_put_xrcd; } } - pd = idr_read_pd(cmd->pd_handle, file->ucontext); + pd = uobj_get_obj_read(pd, cmd->pd_handle, file->ucontext); if (!pd) { ret = -EINVAL; goto err_put_cq; @@ -4086,11 +3480,6 @@ static int __uverbs_create_xsrq(struct ib_uverbs_file *file, obj->uevent.events_reported = 0; INIT_LIST_HEAD(&obj->uevent.event_list); - ret = ib_rdmacg_try_charge(&obj->uevent.uobject.cg_obj, ib_dev, - RDMACG_RESOURCE_HCA_OBJECT); - if (ret) - goto err_put_cq; - srq = pd->device->create_srq(pd, &attr, udata); if (IS_ERR(srq)) { ret = PTR_ERR(srq); @@ -4115,9 +3504,7 @@ static int __uverbs_create_xsrq(struct ib_uverbs_file *file, atomic_set(&srq->usecnt, 0); obj->uevent.uobject.object = srq; - ret = idr_add_uobj(&ib_uverbs_srq_idr, &obj->uevent.uobject); - if (ret) - goto err_destroy; + obj->uevent.uobject.user_handle = cmd->user_handle; memset(&resp, 0, sizeof resp); resp.srq_handle = obj->uevent.uobject.id; @@ -4133,44 +3520,32 @@ static int __uverbs_create_xsrq(struct ib_uverbs_file *file, } if (cmd->srq_type == IB_SRQT_XRC) { - put_uobj_read(xrcd_uobj); - put_cq_read(attr.ext.xrc.cq); + uobj_put_read(xrcd_uobj); + uobj_put_obj_read(attr.ext.xrc.cq); } - put_pd_read(pd); - - mutex_lock(&file->mutex); - list_add_tail(&obj->uevent.uobject.list, &file->ucontext->srq_list); - mutex_unlock(&file->mutex); - - obj->uevent.uobject.live = 1; - - up_write(&obj->uevent.uobject.mutex); + uobj_put_obj_read(pd); + uobj_alloc_commit(&obj->uevent.uobject); return 0; err_copy: - idr_remove_uobj(&ib_uverbs_srq_idr, &obj->uevent.uobject); - -err_destroy: ib_destroy_srq(srq); err_put: - ib_rdmacg_uncharge(&obj->uevent.uobject.cg_obj, ib_dev, - RDMACG_RESOURCE_HCA_OBJECT); - put_pd_read(pd); + uobj_put_obj_read(pd); err_put_cq: if (cmd->srq_type == IB_SRQT_XRC) - put_cq_read(attr.ext.xrc.cq); + uobj_put_obj_read(attr.ext.xrc.cq); err_put_xrcd: if (cmd->srq_type == IB_SRQT_XRC) { atomic_dec(&obj->uxrcd->refcnt); - put_uobj_read(xrcd_uobj); + uobj_put_read(xrcd_uobj); } err: - put_uobj_write(&obj->uevent.uobject); + uobj_alloc_abort(&obj->uevent.uobject); return ret; } @@ -4255,7 +3630,7 @@ ssize_t ib_uverbs_modify_srq(struct ib_uverbs_file *file, INIT_UDATA(&udata, buf + sizeof cmd, NULL, in_len - sizeof cmd, out_len); - srq = idr_read_srq(cmd.srq_handle, file->ucontext); + srq = uobj_get_obj_read(srq, cmd.srq_handle, file->ucontext); if (!srq) return -EINVAL; @@ -4264,7 +3639,7 @@ ssize_t ib_uverbs_modify_srq(struct ib_uverbs_file *file, ret = srq->device->modify_srq(srq, &attr, cmd.attr_mask, &udata); - put_srq_read(srq); + uobj_put_obj_read(srq); return ret ? ret : in_len; } @@ -4286,13 +3661,13 @@ ssize_t ib_uverbs_query_srq(struct ib_uverbs_file *file, if (copy_from_user(&cmd, buf, sizeof cmd)) return -EFAULT; - srq = idr_read_srq(cmd.srq_handle, file->ucontext); + srq = uobj_get_obj_read(srq, cmd.srq_handle, file->ucontext); if (!srq) return -EINVAL; ret = ib_query_srq(srq, &attr); - put_srq_read(srq); + uobj_put_obj_read(srq); if (ret) return ret; @@ -4321,53 +3696,39 @@ ssize_t ib_uverbs_destroy_srq(struct ib_uverbs_file *file, struct ib_srq *srq; struct ib_uevent_object *obj; int ret = -EINVAL; - struct ib_usrq_object *us; enum ib_srq_type srq_type; if (copy_from_user(&cmd, buf, sizeof cmd)) return -EFAULT; - uobj = idr_write_uobj(&ib_uverbs_srq_idr, cmd.srq_handle, file->ucontext); - if (!uobj) - return -EINVAL; + uobj = uobj_get_write(uobj_get_type(srq), cmd.srq_handle, + file->ucontext); + if (IS_ERR(uobj)) + return PTR_ERR(uobj); + srq = uobj->object; obj = container_of(uobj, struct ib_uevent_object, uobject); srq_type = srq->srq_type; + /* + * Make sure we don't free the memory in remove_commit as we still + * needs the uobject memory to create the response. + */ + uverbs_uobject_get(uobj); - ret = ib_destroy_srq(srq); - if (!ret) - uobj->live = 0; - - put_uobj_write(uobj); + memset(&resp, 0, sizeof(resp)); - if (ret) + ret = uobj_remove_commit(uobj); + if (ret) { + uverbs_uobject_put(uobj); return ret; - - ib_rdmacg_uncharge(&uobj->cg_obj, ib_dev, RDMACG_RESOURCE_HCA_OBJECT); - - if (srq_type == IB_SRQT_XRC) { - us = container_of(obj, struct ib_usrq_object, uevent); - atomic_dec(&us->uxrcd->refcnt); } - - idr_remove_uobj(&ib_uverbs_srq_idr, uobj); - - mutex_lock(&file->mutex); - list_del(&uobj->list); - mutex_unlock(&file->mutex); - - ib_uverbs_release_uevent(file, obj); - - memset(&resp, 0, sizeof resp); resp.events_reported = obj->events_reported; + uverbs_uobject_put(uobj); + if (copy_to_user((void __user *)(unsigned long)cmd.response, + &resp, sizeof(resp))) + return -EFAULT; - put_uobj(uobj); - - if (copy_to_user((void __user *) (unsigned long) cmd.response, - &resp, sizeof resp)) - ret = -EFAULT; - - return ret ? ret : in_len; + return in_len; } int ib_uverbs_ex_query_device(struct ib_uverbs_file *file, diff --git a/drivers/infiniband/core/uverbs_main.c b/drivers/infiniband/core/uverbs_main.c index 35c788a32e26..3a9883d1257e 100644 --- a/drivers/infiniband/core/uverbs_main.c +++ b/drivers/infiniband/core/uverbs_main.c @@ -52,6 +52,7 @@ #include "uverbs.h" #include "core_priv.h" +#include "rdma_core.h" MODULE_AUTHOR("Roland Dreier"); MODULE_DESCRIPTION("InfiniBand userspace verbs access"); @@ -67,19 +68,6 @@ enum { static struct class *uverbs_class; -DEFINE_SPINLOCK(ib_uverbs_idr_lock); -DEFINE_IDR(ib_uverbs_pd_idr); -DEFINE_IDR(ib_uverbs_mr_idr); -DEFINE_IDR(ib_uverbs_mw_idr); -DEFINE_IDR(ib_uverbs_ah_idr); -DEFINE_IDR(ib_uverbs_cq_idr); -DEFINE_IDR(ib_uverbs_qp_idr); -DEFINE_IDR(ib_uverbs_srq_idr); -DEFINE_IDR(ib_uverbs_xrcd_idr); -DEFINE_IDR(ib_uverbs_rule_idr); -DEFINE_IDR(ib_uverbs_wq_idr); -DEFINE_IDR(ib_uverbs_rwq_ind_tbl_idr); - static DEFINE_SPINLOCK(map_lock); static DECLARE_BITMAP(dev_map, IB_UVERBS_MAX_DEVICES); @@ -168,37 +156,37 @@ static struct kobj_type ib_uverbs_dev_ktype = { .release = ib_uverbs_release_dev, }; -static void ib_uverbs_release_event_file(struct kref *ref) +static void ib_uverbs_release_async_event_file(struct kref *ref) { - struct ib_uverbs_event_file *file = - container_of(ref, struct ib_uverbs_event_file, ref); + struct ib_uverbs_async_event_file *file = + container_of(ref, struct ib_uverbs_async_event_file, ref); kfree(file); } void ib_uverbs_release_ucq(struct ib_uverbs_file *file, - struct ib_uverbs_event_file *ev_file, + struct ib_uverbs_completion_event_file *ev_file, struct ib_ucq_object *uobj) { struct ib_uverbs_event *evt, *tmp; if (ev_file) { - spin_lock_irq(&ev_file->lock); + spin_lock_irq(&ev_file->ev_queue.lock); list_for_each_entry_safe(evt, tmp, &uobj->comp_list, obj_list) { list_del(&evt->list); kfree(evt); } - spin_unlock_irq(&ev_file->lock); + spin_unlock_irq(&ev_file->ev_queue.lock); - kref_put(&ev_file->ref, ib_uverbs_release_event_file); + uverbs_uobject_put(&ev_file->uobj_file.uobj); } - spin_lock_irq(&file->async_file->lock); + spin_lock_irq(&file->async_file->ev_queue.lock); list_for_each_entry_safe(evt, tmp, &uobj->async_list, obj_list) { list_del(&evt->list); kfree(evt); } - spin_unlock_irq(&file->async_file->lock); + spin_unlock_irq(&file->async_file->ev_queue.lock); } void ib_uverbs_release_uevent(struct ib_uverbs_file *file, @@ -206,16 +194,16 @@ void ib_uverbs_release_uevent(struct ib_uverbs_file *file, { struct ib_uverbs_event *evt, *tmp; - spin_lock_irq(&file->async_file->lock); + spin_lock_irq(&file->async_file->ev_queue.lock); list_for_each_entry_safe(evt, tmp, &uobj->event_list, obj_list) { list_del(&evt->list); kfree(evt); } - spin_unlock_irq(&file->async_file->lock); + spin_unlock_irq(&file->async_file->ev_queue.lock); } -static void ib_uverbs_detach_umcast(struct ib_qp *qp, - struct ib_uqp_object *uobj) +void ib_uverbs_detach_umcast(struct ib_qp *qp, + struct ib_uqp_object *uobj) { struct ib_uverbs_mcast_entry *mcast, *tmp; @@ -227,138 +215,11 @@ static void ib_uverbs_detach_umcast(struct ib_qp *qp, } static int ib_uverbs_cleanup_ucontext(struct ib_uverbs_file *file, - struct ib_ucontext *context) + struct ib_ucontext *context, + bool device_removed) { - struct ib_uobject *uobj, *tmp; - context->closing = 1; - - list_for_each_entry_safe(uobj, tmp, &context->ah_list, list) { - struct ib_ah *ah = uobj->object; - - idr_remove_uobj(&ib_uverbs_ah_idr, uobj); - ib_destroy_ah(ah); - ib_rdmacg_uncharge(&uobj->cg_obj, context->device, - RDMACG_RESOURCE_HCA_OBJECT); - kfree(uobj); - } - - /* Remove MWs before QPs, in order to support type 2A MWs. */ - list_for_each_entry_safe(uobj, tmp, &context->mw_list, list) { - struct ib_mw *mw = uobj->object; - - idr_remove_uobj(&ib_uverbs_mw_idr, uobj); - uverbs_dealloc_mw(mw); - ib_rdmacg_uncharge(&uobj->cg_obj, context->device, - RDMACG_RESOURCE_HCA_OBJECT); - kfree(uobj); - } - - list_for_each_entry_safe(uobj, tmp, &context->rule_list, list) { - struct ib_flow *flow_id = uobj->object; - - idr_remove_uobj(&ib_uverbs_rule_idr, uobj); - ib_destroy_flow(flow_id); - ib_rdmacg_uncharge(&uobj->cg_obj, context->device, - RDMACG_RESOURCE_HCA_OBJECT); - kfree(uobj); - } - - list_for_each_entry_safe(uobj, tmp, &context->qp_list, list) { - struct ib_qp *qp = uobj->object; - struct ib_uqp_object *uqp = - container_of(uobj, struct ib_uqp_object, uevent.uobject); - - idr_remove_uobj(&ib_uverbs_qp_idr, uobj); - if (qp == qp->real_qp) - ib_uverbs_detach_umcast(qp, uqp); - ib_destroy_qp(qp); - ib_rdmacg_uncharge(&uobj->cg_obj, context->device, - RDMACG_RESOURCE_HCA_OBJECT); - ib_uverbs_release_uevent(file, &uqp->uevent); - kfree(uqp); - } - - list_for_each_entry_safe(uobj, tmp, &context->rwq_ind_tbl_list, list) { - struct ib_rwq_ind_table *rwq_ind_tbl = uobj->object; - struct ib_wq **ind_tbl = rwq_ind_tbl->ind_tbl; - - idr_remove_uobj(&ib_uverbs_rwq_ind_tbl_idr, uobj); - ib_destroy_rwq_ind_table(rwq_ind_tbl); - kfree(ind_tbl); - kfree(uobj); - } - - list_for_each_entry_safe(uobj, tmp, &context->wq_list, list) { - struct ib_wq *wq = uobj->object; - struct ib_uwq_object *uwq = - container_of(uobj, struct ib_uwq_object, uevent.uobject); - - idr_remove_uobj(&ib_uverbs_wq_idr, uobj); - ib_destroy_wq(wq); - ib_uverbs_release_uevent(file, &uwq->uevent); - kfree(uwq); - } - - list_for_each_entry_safe(uobj, tmp, &context->srq_list, list) { - struct ib_srq *srq = uobj->object; - struct ib_uevent_object *uevent = - container_of(uobj, struct ib_uevent_object, uobject); - - idr_remove_uobj(&ib_uverbs_srq_idr, uobj); - ib_destroy_srq(srq); - ib_rdmacg_uncharge(&uobj->cg_obj, context->device, - RDMACG_RESOURCE_HCA_OBJECT); - ib_uverbs_release_uevent(file, uevent); - kfree(uevent); - } - - list_for_each_entry_safe(uobj, tmp, &context->cq_list, list) { - struct ib_cq *cq = uobj->object; - struct ib_uverbs_event_file *ev_file = cq->cq_context; - struct ib_ucq_object *ucq = - container_of(uobj, struct ib_ucq_object, uobject); - - idr_remove_uobj(&ib_uverbs_cq_idr, uobj); - ib_destroy_cq(cq); - ib_rdmacg_uncharge(&uobj->cg_obj, context->device, - RDMACG_RESOURCE_HCA_OBJECT); - ib_uverbs_release_ucq(file, ev_file, ucq); - kfree(ucq); - } - - list_for_each_entry_safe(uobj, tmp, &context->mr_list, list) { - struct ib_mr *mr = uobj->object; - - idr_remove_uobj(&ib_uverbs_mr_idr, uobj); - ib_dereg_mr(mr); - ib_rdmacg_uncharge(&uobj->cg_obj, context->device, - RDMACG_RESOURCE_HCA_OBJECT); - kfree(uobj); - } - - mutex_lock(&file->device->xrcd_tree_mutex); - list_for_each_entry_safe(uobj, tmp, &context->xrcd_list, list) { - struct ib_xrcd *xrcd = uobj->object; - struct ib_uxrcd_object *uxrcd = - container_of(uobj, struct ib_uxrcd_object, uobject); - - idr_remove_uobj(&ib_uverbs_xrcd_idr, uobj); - ib_uverbs_dealloc_xrcd(file->device, xrcd); - kfree(uxrcd); - } - mutex_unlock(&file->device->xrcd_tree_mutex); - - list_for_each_entry_safe(uobj, tmp, &context->pd_list, list) { - struct ib_pd *pd = uobj->object; - - idr_remove_uobj(&ib_uverbs_pd_idr, uobj); - ib_dealloc_pd(pd); - ib_rdmacg_uncharge(&uobj->cg_obj, context->device, - RDMACG_RESOURCE_HCA_OBJECT); - kfree(uobj); - } - + uverbs_cleanup_ucontext(context, device_removed); put_pid(context->tgid); ib_rdmacg_uncharge(&context->cg_obj, context->device, @@ -372,7 +233,7 @@ static void ib_uverbs_comp_dev(struct ib_uverbs_device *dev) complete(&dev->comp); } -static void ib_uverbs_release_file(struct kref *ref) +void ib_uverbs_release_file(struct kref *ref) { struct ib_uverbs_file *file = container_of(ref, struct ib_uverbs_file, ref); @@ -392,58 +253,54 @@ static void ib_uverbs_release_file(struct kref *ref) kfree(file); } -static ssize_t ib_uverbs_event_read(struct file *filp, char __user *buf, - size_t count, loff_t *pos) +static ssize_t ib_uverbs_event_read(struct ib_uverbs_event_queue *ev_queue, + struct ib_uverbs_file *uverbs_file, + struct file *filp, char __user *buf, + size_t count, loff_t *pos, + size_t eventsz) { - struct ib_uverbs_event_file *file = filp->private_data; struct ib_uverbs_event *event; - int eventsz; int ret = 0; - spin_lock_irq(&file->lock); + spin_lock_irq(&ev_queue->lock); - while (list_empty(&file->event_list)) { - spin_unlock_irq(&file->lock); + while (list_empty(&ev_queue->event_list)) { + spin_unlock_irq(&ev_queue->lock); if (filp->f_flags & O_NONBLOCK) return -EAGAIN; - if (wait_event_interruptible(file->poll_wait, - (!list_empty(&file->event_list) || + if (wait_event_interruptible(ev_queue->poll_wait, + (!list_empty(&ev_queue->event_list) || /* The barriers built into wait_event_interruptible() * and wake_up() guarentee this will see the null set * without using RCU */ - !file->uverbs_file->device->ib_dev))) + !uverbs_file->device->ib_dev))) return -ERESTARTSYS; /* If device was disassociated and no event exists set an error */ - if (list_empty(&file->event_list) && - !file->uverbs_file->device->ib_dev) + if (list_empty(&ev_queue->event_list) && + !uverbs_file->device->ib_dev) return -EIO; - spin_lock_irq(&file->lock); + spin_lock_irq(&ev_queue->lock); } - event = list_entry(file->event_list.next, struct ib_uverbs_event, list); - - if (file->is_async) - eventsz = sizeof (struct ib_uverbs_async_event_desc); - else - eventsz = sizeof (struct ib_uverbs_comp_event_desc); + event = list_entry(ev_queue->event_list.next, struct ib_uverbs_event, list); if (eventsz > count) { ret = -EINVAL; event = NULL; } else { - list_del(file->event_list.next); + list_del(ev_queue->event_list.next); if (event->counter) { ++(*event->counter); list_del(&event->obj_list); } } - spin_unlock_irq(&file->lock); + spin_unlock_irq(&ev_queue->lock); if (event) { if (copy_to_user(buf, event, eventsz)) @@ -457,87 +314,158 @@ static ssize_t ib_uverbs_event_read(struct file *filp, char __user *buf, return ret; } -static unsigned int ib_uverbs_event_poll(struct file *filp, +static ssize_t ib_uverbs_async_event_read(struct file *filp, char __user *buf, + size_t count, loff_t *pos) +{ + struct ib_uverbs_async_event_file *file = filp->private_data; + + return ib_uverbs_event_read(&file->ev_queue, file->uverbs_file, filp, + buf, count, pos, + sizeof(struct ib_uverbs_async_event_desc)); +} + +static ssize_t ib_uverbs_comp_event_read(struct file *filp, char __user *buf, + size_t count, loff_t *pos) +{ + struct ib_uverbs_completion_event_file *comp_ev_file = + filp->private_data; + + return ib_uverbs_event_read(&comp_ev_file->ev_queue, + comp_ev_file->uobj_file.ufile, filp, + buf, count, pos, + sizeof(struct ib_uverbs_comp_event_desc)); +} + +static unsigned int ib_uverbs_event_poll(struct ib_uverbs_event_queue *ev_queue, + struct file *filp, struct poll_table_struct *wait) { unsigned int pollflags = 0; - struct ib_uverbs_event_file *file = filp->private_data; - poll_wait(filp, &file->poll_wait, wait); + poll_wait(filp, &ev_queue->poll_wait, wait); - spin_lock_irq(&file->lock); - if (!list_empty(&file->event_list)) + spin_lock_irq(&ev_queue->lock); + if (!list_empty(&ev_queue->event_list)) pollflags = POLLIN | POLLRDNORM; - spin_unlock_irq(&file->lock); + spin_unlock_irq(&ev_queue->lock); return pollflags; } -static int ib_uverbs_event_fasync(int fd, struct file *filp, int on) +static unsigned int ib_uverbs_async_event_poll(struct file *filp, + struct poll_table_struct *wait) +{ + return ib_uverbs_event_poll(filp->private_data, filp, wait); +} + +static unsigned int ib_uverbs_comp_event_poll(struct file *filp, + struct poll_table_struct *wait) { - struct ib_uverbs_event_file *file = filp->private_data; + struct ib_uverbs_completion_event_file *comp_ev_file = + filp->private_data; - return fasync_helper(fd, filp, on, &file->async_queue); + return ib_uverbs_event_poll(&comp_ev_file->ev_queue, filp, wait); } -static int ib_uverbs_event_close(struct inode *inode, struct file *filp) +static int ib_uverbs_async_event_fasync(int fd, struct file *filp, int on) { - struct ib_uverbs_event_file *file = filp->private_data; + struct ib_uverbs_event_queue *ev_queue = filp->private_data; + + return fasync_helper(fd, filp, on, &ev_queue->async_queue); +} + +static int ib_uverbs_comp_event_fasync(int fd, struct file *filp, int on) +{ + struct ib_uverbs_completion_event_file *comp_ev_file = + filp->private_data; + + return fasync_helper(fd, filp, on, &comp_ev_file->ev_queue.async_queue); +} + +static int ib_uverbs_async_event_close(struct inode *inode, struct file *filp) +{ + struct ib_uverbs_async_event_file *file = filp->private_data; + struct ib_uverbs_file *uverbs_file = file->uverbs_file; struct ib_uverbs_event *entry, *tmp; int closed_already = 0; - mutex_lock(&file->uverbs_file->device->lists_mutex); - spin_lock_irq(&file->lock); - closed_already = file->is_closed; - file->is_closed = 1; - list_for_each_entry_safe(entry, tmp, &file->event_list, list) { + mutex_lock(&uverbs_file->device->lists_mutex); + spin_lock_irq(&file->ev_queue.lock); + closed_already = file->ev_queue.is_closed; + file->ev_queue.is_closed = 1; + list_for_each_entry_safe(entry, tmp, &file->ev_queue.event_list, list) { if (entry->counter) list_del(&entry->obj_list); kfree(entry); } - spin_unlock_irq(&file->lock); + spin_unlock_irq(&file->ev_queue.lock); if (!closed_already) { list_del(&file->list); - if (file->is_async) - ib_unregister_event_handler(&file->uverbs_file-> - event_handler); + ib_unregister_event_handler(&uverbs_file->event_handler); + } + mutex_unlock(&uverbs_file->device->lists_mutex); + + kref_put(&uverbs_file->ref, ib_uverbs_release_file); + kref_put(&file->ref, ib_uverbs_release_async_event_file); + + return 0; +} + +static int ib_uverbs_comp_event_close(struct inode *inode, struct file *filp) +{ + struct ib_uverbs_completion_event_file *file = filp->private_data; + struct ib_uverbs_event *entry, *tmp; + + spin_lock_irq(&file->ev_queue.lock); + list_for_each_entry_safe(entry, tmp, &file->ev_queue.event_list, list) { + if (entry->counter) + list_del(&entry->obj_list); + kfree(entry); } - mutex_unlock(&file->uverbs_file->device->lists_mutex); + spin_unlock_irq(&file->ev_queue.lock); - kref_put(&file->uverbs_file->ref, ib_uverbs_release_file); - kref_put(&file->ref, ib_uverbs_release_event_file); + uverbs_close_fd(filp); return 0; } -static const struct file_operations uverbs_event_fops = { +const struct file_operations uverbs_event_fops = { + .owner = THIS_MODULE, + .read = ib_uverbs_comp_event_read, + .poll = ib_uverbs_comp_event_poll, + .release = ib_uverbs_comp_event_close, + .fasync = ib_uverbs_comp_event_fasync, + .llseek = no_llseek, +}; + +static const struct file_operations uverbs_async_event_fops = { .owner = THIS_MODULE, - .read = ib_uverbs_event_read, - .poll = ib_uverbs_event_poll, - .release = ib_uverbs_event_close, - .fasync = ib_uverbs_event_fasync, + .read = ib_uverbs_async_event_read, + .poll = ib_uverbs_async_event_poll, + .release = ib_uverbs_async_event_close, + .fasync = ib_uverbs_async_event_fasync, .llseek = no_llseek, }; void ib_uverbs_comp_handler(struct ib_cq *cq, void *cq_context) { - struct ib_uverbs_event_file *file = cq_context; + struct ib_uverbs_event_queue *ev_queue = cq_context; struct ib_ucq_object *uobj; struct ib_uverbs_event *entry; unsigned long flags; - if (!file) + if (!ev_queue) return; - spin_lock_irqsave(&file->lock, flags); - if (file->is_closed) { - spin_unlock_irqrestore(&file->lock, flags); + spin_lock_irqsave(&ev_queue->lock, flags); + if (ev_queue->is_closed) { + spin_unlock_irqrestore(&ev_queue->lock, flags); return; } entry = kmalloc(sizeof *entry, GFP_ATOMIC); if (!entry) { - spin_unlock_irqrestore(&file->lock, flags); + spin_unlock_irqrestore(&ev_queue->lock, flags); return; } @@ -546,12 +474,12 @@ void ib_uverbs_comp_handler(struct ib_cq *cq, void *cq_context) entry->desc.comp.cq_handle = cq->uobject->user_handle; entry->counter = &uobj->comp_events_reported; - list_add_tail(&entry->list, &file->event_list); + list_add_tail(&entry->list, &ev_queue->event_list); list_add_tail(&entry->obj_list, &uobj->comp_list); - spin_unlock_irqrestore(&file->lock, flags); + spin_unlock_irqrestore(&ev_queue->lock, flags); - wake_up_interruptible(&file->poll_wait); - kill_fasync(&file->async_queue, SIGIO, POLL_IN); + wake_up_interruptible(&ev_queue->poll_wait); + kill_fasync(&ev_queue->async_queue, SIGIO, POLL_IN); } static void ib_uverbs_async_handler(struct ib_uverbs_file *file, @@ -562,15 +490,15 @@ static void ib_uverbs_async_handler(struct ib_uverbs_file *file, struct ib_uverbs_event *entry; unsigned long flags; - spin_lock_irqsave(&file->async_file->lock, flags); - if (file->async_file->is_closed) { - spin_unlock_irqrestore(&file->async_file->lock, flags); + spin_lock_irqsave(&file->async_file->ev_queue.lock, flags); + if (file->async_file->ev_queue.is_closed) { + spin_unlock_irqrestore(&file->async_file->ev_queue.lock, flags); return; } entry = kmalloc(sizeof *entry, GFP_ATOMIC); if (!entry) { - spin_unlock_irqrestore(&file->async_file->lock, flags); + spin_unlock_irqrestore(&file->async_file->ev_queue.lock, flags); return; } @@ -579,13 +507,13 @@ static void ib_uverbs_async_handler(struct ib_uverbs_file *file, entry->desc.async.reserved = 0; entry->counter = counter; - list_add_tail(&entry->list, &file->async_file->event_list); + list_add_tail(&entry->list, &file->async_file->ev_queue.event_list); if (obj_list) list_add_tail(&entry->obj_list, obj_list); - spin_unlock_irqrestore(&file->async_file->lock, flags); + spin_unlock_irqrestore(&file->async_file->ev_queue.lock, flags); - wake_up_interruptible(&file->async_file->poll_wait); - kill_fasync(&file->async_file->async_queue, SIGIO, POLL_IN); + wake_up_interruptible(&file->async_file->ev_queue.poll_wait); + kill_fasync(&file->async_file->ev_queue.async_queue, SIGIO, POLL_IN); } void ib_uverbs_cq_event_handler(struct ib_event *event, void *context_ptr) @@ -603,7 +531,7 @@ void ib_uverbs_qp_event_handler(struct ib_event *event, void *context_ptr) struct ib_uevent_object *uobj; /* for XRC target qp's, check that qp is live */ - if (!event->element.qp->uobject || !event->element.qp->uobject->live) + if (!event->element.qp->uobject) return; uobj = container_of(event->element.qp->uobject, @@ -648,15 +576,23 @@ void ib_uverbs_event_handler(struct ib_event_handler *handler, void ib_uverbs_free_async_event_file(struct ib_uverbs_file *file) { - kref_put(&file->async_file->ref, ib_uverbs_release_event_file); + kref_put(&file->async_file->ref, ib_uverbs_release_async_event_file); file->async_file = NULL; } -struct file *ib_uverbs_alloc_event_file(struct ib_uverbs_file *uverbs_file, - struct ib_device *ib_dev, - int is_async) +void ib_uverbs_init_event_queue(struct ib_uverbs_event_queue *ev_queue) { - struct ib_uverbs_event_file *ev_file; + spin_lock_init(&ev_queue->lock); + INIT_LIST_HEAD(&ev_queue->event_list); + init_waitqueue_head(&ev_queue->poll_wait); + ev_queue->is_closed = 0; + ev_queue->async_queue = NULL; +} + +struct file *ib_uverbs_alloc_async_event_file(struct ib_uverbs_file *uverbs_file, + struct ib_device *ib_dev) +{ + struct ib_uverbs_async_event_file *ev_file; struct file *filp; int ret; @@ -664,16 +600,11 @@ struct file *ib_uverbs_alloc_event_file(struct ib_uverbs_file *uverbs_file, if (!ev_file) return ERR_PTR(-ENOMEM); - kref_init(&ev_file->ref); - spin_lock_init(&ev_file->lock); - INIT_LIST_HEAD(&ev_file->event_list); - init_waitqueue_head(&ev_file->poll_wait); + ib_uverbs_init_event_queue(&ev_file->ev_queue); ev_file->uverbs_file = uverbs_file; kref_get(&ev_file->uverbs_file->ref); - ev_file->async_queue = NULL; - ev_file->is_closed = 0; - - filp = anon_inode_getfile("[infinibandevent]", &uverbs_event_fops, + kref_init(&ev_file->ref); + filp = anon_inode_getfile("[infinibandevent]", &uverbs_async_event_fops, ev_file, O_RDONLY); if (IS_ERR(filp)) goto err_put_refs; @@ -683,64 +614,33 @@ struct file *ib_uverbs_alloc_event_file(struct ib_uverbs_file *uverbs_file, &uverbs_file->device->uverbs_events_file_list); mutex_unlock(&uverbs_file->device->lists_mutex); - if (is_async) { - WARN_ON(uverbs_file->async_file); - uverbs_file->async_file = ev_file; - kref_get(&uverbs_file->async_file->ref); - INIT_IB_EVENT_HANDLER(&uverbs_file->event_handler, - ib_dev, - ib_uverbs_event_handler); - ret = ib_register_event_handler(&uverbs_file->event_handler); - if (ret) - goto err_put_file; - - /* At that point async file stuff was fully set */ - ev_file->is_async = 1; - } + WARN_ON(uverbs_file->async_file); + uverbs_file->async_file = ev_file; + kref_get(&uverbs_file->async_file->ref); + INIT_IB_EVENT_HANDLER(&uverbs_file->event_handler, + ib_dev, + ib_uverbs_event_handler); + ret = ib_register_event_handler(&uverbs_file->event_handler); + if (ret) + goto err_put_file; + + /* At that point async file stuff was fully set */ return filp; err_put_file: fput(filp); - kref_put(&uverbs_file->async_file->ref, ib_uverbs_release_event_file); + kref_put(&uverbs_file->async_file->ref, + ib_uverbs_release_async_event_file); uverbs_file->async_file = NULL; return ERR_PTR(ret); err_put_refs: kref_put(&ev_file->uverbs_file->ref, ib_uverbs_release_file); - kref_put(&ev_file->ref, ib_uverbs_release_event_file); + kref_put(&ev_file->ref, ib_uverbs_release_async_event_file); return filp; } -/* - * Look up a completion event file by FD. If lookup is successful, - * takes a ref to the event file struct that it returns; if - * unsuccessful, returns NULL. - */ -struct ib_uverbs_event_file *ib_uverbs_lookup_comp_file(int fd) -{ - struct ib_uverbs_event_file *ev_file = NULL; - struct fd f = fdget(fd); - - if (!f.file) - return NULL; - - if (f.file->f_op != &uverbs_event_fops) - goto out; - - ev_file = f.file->private_data; - if (ev_file->is_async) { - ev_file = NULL; - goto out; - } - - kref_get(&ev_file->ref); - -out: - fdput(f); - return ev_file; -} - static int verify_command_mask(struct ib_device *ib_dev, __u32 command) { u64 mask; @@ -986,6 +886,8 @@ static int ib_uverbs_open(struct inode *inode, struct file *filp) } file->device = dev; + spin_lock_init(&file->idr_lock); + idr_init(&file->idr); file->ucontext = NULL; file->async_file = NULL; kref_init(&file->ref); @@ -1019,10 +921,11 @@ static int ib_uverbs_close(struct inode *inode, struct file *filp) mutex_lock(&file->cleanup_mutex); if (file->ucontext) { - ib_uverbs_cleanup_ucontext(file, file->ucontext); + ib_uverbs_cleanup_ucontext(file, file->ucontext, false); file->ucontext = NULL; } mutex_unlock(&file->cleanup_mutex); + idr_destroy(&file->idr); mutex_lock(&file->device->lists_mutex); if (!file->is_closed) { @@ -1032,7 +935,8 @@ static int ib_uverbs_close(struct inode *inode, struct file *filp) mutex_unlock(&file->device->lists_mutex); if (file->async_file) - kref_put(&file->async_file->ref, ib_uverbs_release_event_file); + kref_put(&file->async_file->ref, + ib_uverbs_release_async_event_file); kref_put(&file->ref, ib_uverbs_release_file); kobject_put(&dev->kobj); @@ -1231,7 +1135,7 @@ static void ib_uverbs_free_hw_resources(struct ib_uverbs_device *uverbs_dev, struct ib_device *ib_dev) { struct ib_uverbs_file *file; - struct ib_uverbs_event_file *event_file; + struct ib_uverbs_async_event_file *event_file; struct ib_event event; /* Pending running commands to terminate */ @@ -1268,7 +1172,9 @@ static void ib_uverbs_free_hw_resources(struct ib_uverbs_device *uverbs_dev, * (e.g mmput). */ ib_dev->disassociate_ucontext(ucontext); - ib_uverbs_cleanup_ucontext(file, ucontext); + mutex_lock(&file->cleanup_mutex); + ib_uverbs_cleanup_ucontext(file, ucontext, true); + mutex_unlock(&file->cleanup_mutex); } mutex_lock(&uverbs_dev->lists_mutex); @@ -1278,21 +1184,20 @@ static void ib_uverbs_free_hw_resources(struct ib_uverbs_device *uverbs_dev, while (!list_empty(&uverbs_dev->uverbs_events_file_list)) { event_file = list_first_entry(&uverbs_dev-> uverbs_events_file_list, - struct ib_uverbs_event_file, + struct ib_uverbs_async_event_file, list); - spin_lock_irq(&event_file->lock); - event_file->is_closed = 1; - spin_unlock_irq(&event_file->lock); + spin_lock_irq(&event_file->ev_queue.lock); + event_file->ev_queue.is_closed = 1; + spin_unlock_irq(&event_file->ev_queue.lock); list_del(&event_file->list); - if (event_file->is_async) { - ib_unregister_event_handler(&event_file->uverbs_file-> - event_handler); - event_file->uverbs_file->event_handler.device = NULL; - } + ib_unregister_event_handler( + &event_file->uverbs_file->event_handler); + event_file->uverbs_file->event_handler.device = + NULL; - wake_up_interruptible(&event_file->poll_wait); - kill_fasync(&event_file->async_queue, SIGIO, POLL_IN); + wake_up_interruptible(&event_file->ev_queue.poll_wait); + kill_fasync(&event_file->ev_queue.async_queue, SIGIO, POLL_IN); } mutex_unlock(&uverbs_dev->lists_mutex); } @@ -1396,13 +1301,6 @@ static void __exit ib_uverbs_cleanup(void) unregister_chrdev_region(IB_UVERBS_BASE_DEV, IB_UVERBS_MAX_DEVICES); if (overflow_maj) unregister_chrdev_region(overflow_maj, IB_UVERBS_MAX_DEVICES); - idr_destroy(&ib_uverbs_pd_idr); - idr_destroy(&ib_uverbs_mr_idr); - idr_destroy(&ib_uverbs_mw_idr); - idr_destroy(&ib_uverbs_ah_idr); - idr_destroy(&ib_uverbs_cq_idr); - idr_destroy(&ib_uverbs_qp_idr); - idr_destroy(&ib_uverbs_srq_idr); } module_init(ib_uverbs_init); diff --git a/drivers/infiniband/core/uverbs_std_types.c b/drivers/infiniband/core/uverbs_std_types.c new file mode 100644 index 000000000000..ad3caad40945 --- /dev/null +++ b/drivers/infiniband/core/uverbs_std_types.c @@ -0,0 +1,275 @@ +/* + * Copyright (c) 2017, Mellanox Technologies inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include <rdma/uverbs_std_types.h> +#include <rdma/ib_user_verbs.h> +#include <rdma/ib_verbs.h> +#include <linux/bug.h> +#include <linux/file.h> +#include "rdma_core.h" +#include "uverbs.h" + +static int uverbs_free_ah(struct ib_uobject *uobject, + enum rdma_remove_reason why) +{ + return ib_destroy_ah((struct ib_ah *)uobject->object); +} + +static int uverbs_free_flow(struct ib_uobject *uobject, + enum rdma_remove_reason why) +{ + return ib_destroy_flow((struct ib_flow *)uobject->object); +} + +static int uverbs_free_mw(struct ib_uobject *uobject, + enum rdma_remove_reason why) +{ + return uverbs_dealloc_mw((struct ib_mw *)uobject->object); +} + +static int uverbs_free_qp(struct ib_uobject *uobject, + enum rdma_remove_reason why) +{ + struct ib_qp *qp = uobject->object; + struct ib_uqp_object *uqp = + container_of(uobject, struct ib_uqp_object, uevent.uobject); + int ret; + + if (why == RDMA_REMOVE_DESTROY) { + if (!list_empty(&uqp->mcast_list)) + return -EBUSY; + } else if (qp == qp->real_qp) { + ib_uverbs_detach_umcast(qp, uqp); + } + + ret = ib_destroy_qp(qp); + if (ret && why == RDMA_REMOVE_DESTROY) + return ret; + + if (uqp->uxrcd) + atomic_dec(&uqp->uxrcd->refcnt); + + ib_uverbs_release_uevent(uobject->context->ufile, &uqp->uevent); + return ret; +} + +static int uverbs_free_rwq_ind_tbl(struct ib_uobject *uobject, + enum rdma_remove_reason why) +{ + struct ib_rwq_ind_table *rwq_ind_tbl = uobject->object; + struct ib_wq **ind_tbl = rwq_ind_tbl->ind_tbl; + int ret; + + ret = ib_destroy_rwq_ind_table(rwq_ind_tbl); + if (!ret || why != RDMA_REMOVE_DESTROY) + kfree(ind_tbl); + return ret; +} + +static int uverbs_free_wq(struct ib_uobject *uobject, + enum rdma_remove_reason why) +{ + struct ib_wq *wq = uobject->object; + struct ib_uwq_object *uwq = + container_of(uobject, struct ib_uwq_object, uevent.uobject); + int ret; + + ret = ib_destroy_wq(wq); + if (!ret || why != RDMA_REMOVE_DESTROY) + ib_uverbs_release_uevent(uobject->context->ufile, &uwq->uevent); + return ret; +} + +static int uverbs_free_srq(struct ib_uobject *uobject, + enum rdma_remove_reason why) +{ + struct ib_srq *srq = uobject->object; + struct ib_uevent_object *uevent = + container_of(uobject, struct ib_uevent_object, uobject); + enum ib_srq_type srq_type = srq->srq_type; + int ret; + + ret = ib_destroy_srq(srq); + + if (ret && why == RDMA_REMOVE_DESTROY) + return ret; + + if (srq_type == IB_SRQT_XRC) { + struct ib_usrq_object *us = + container_of(uevent, struct ib_usrq_object, uevent); + + atomic_dec(&us->uxrcd->refcnt); + } + + ib_uverbs_release_uevent(uobject->context->ufile, uevent); + return ret; +} + +static int uverbs_free_cq(struct ib_uobject *uobject, + enum rdma_remove_reason why) +{ + struct ib_cq *cq = uobject->object; + struct ib_uverbs_event_queue *ev_queue = cq->cq_context; + struct ib_ucq_object *ucq = + container_of(uobject, struct ib_ucq_object, uobject); + int ret; + + ret = ib_destroy_cq(cq); + if (!ret || why != RDMA_REMOVE_DESTROY) + ib_uverbs_release_ucq(uobject->context->ufile, ev_queue ? + container_of(ev_queue, + struct ib_uverbs_completion_event_file, + ev_queue) : NULL, + ucq); + return ret; +} + +static int uverbs_free_mr(struct ib_uobject *uobject, + enum rdma_remove_reason why) +{ + return ib_dereg_mr((struct ib_mr *)uobject->object); +} + +static int uverbs_free_xrcd(struct ib_uobject *uobject, + enum rdma_remove_reason why) +{ + struct ib_xrcd *xrcd = uobject->object; + struct ib_uxrcd_object *uxrcd = + container_of(uobject, struct ib_uxrcd_object, uobject); + int ret; + + mutex_lock(&uobject->context->ufile->device->xrcd_tree_mutex); + if (why == RDMA_REMOVE_DESTROY && atomic_read(&uxrcd->refcnt)) + ret = -EBUSY; + else + ret = ib_uverbs_dealloc_xrcd(uobject->context->ufile->device, + xrcd, why); + mutex_unlock(&uobject->context->ufile->device->xrcd_tree_mutex); + + return ret; +} + +static int uverbs_free_pd(struct ib_uobject *uobject, + enum rdma_remove_reason why) +{ + struct ib_pd *pd = uobject->object; + + if (why == RDMA_REMOVE_DESTROY && atomic_read(&pd->usecnt)) + return -EBUSY; + + ib_dealloc_pd((struct ib_pd *)uobject->object); + return 0; +} + +static int uverbs_hot_unplug_completion_event_file(struct ib_uobject_file *uobj_file, + enum rdma_remove_reason why) +{ + struct ib_uverbs_completion_event_file *comp_event_file = + container_of(uobj_file, struct ib_uverbs_completion_event_file, + uobj_file); + struct ib_uverbs_event_queue *event_queue = &comp_event_file->ev_queue; + + spin_lock_irq(&event_queue->lock); + event_queue->is_closed = 1; + spin_unlock_irq(&event_queue->lock); + + if (why == RDMA_REMOVE_DRIVER_REMOVE) { + wake_up_interruptible(&event_queue->poll_wait); + kill_fasync(&event_queue->async_queue, SIGIO, POLL_IN); + } + return 0; +}; + +const struct uverbs_obj_fd_type uverbs_type_attrs_comp_channel = { + .type = UVERBS_TYPE_ALLOC_FD(sizeof(struct ib_uverbs_completion_event_file), 0), + .context_closed = uverbs_hot_unplug_completion_event_file, + .fops = &uverbs_event_fops, + .name = "[infinibandevent]", + .flags = O_RDONLY, +}; + +const struct uverbs_obj_idr_type uverbs_type_attrs_cq = { + .type = UVERBS_TYPE_ALLOC_IDR_SZ(sizeof(struct ib_ucq_object), 0), + .destroy_object = uverbs_free_cq, +}; + +const struct uverbs_obj_idr_type uverbs_type_attrs_qp = { + .type = UVERBS_TYPE_ALLOC_IDR_SZ(sizeof(struct ib_uqp_object), 0), + .destroy_object = uverbs_free_qp, +}; + +const struct uverbs_obj_idr_type uverbs_type_attrs_mw = { + .type = UVERBS_TYPE_ALLOC_IDR(0), + .destroy_object = uverbs_free_mw, +}; + +const struct uverbs_obj_idr_type uverbs_type_attrs_mr = { + /* 1 is used in order to free the MR after all the MWs */ + .type = UVERBS_TYPE_ALLOC_IDR(1), + .destroy_object = uverbs_free_mr, +}; + +const struct uverbs_obj_idr_type uverbs_type_attrs_srq = { + .type = UVERBS_TYPE_ALLOC_IDR_SZ(sizeof(struct ib_usrq_object), 0), + .destroy_object = uverbs_free_srq, +}; + +const struct uverbs_obj_idr_type uverbs_type_attrs_ah = { + .type = UVERBS_TYPE_ALLOC_IDR(0), + .destroy_object = uverbs_free_ah, +}; + +const struct uverbs_obj_idr_type uverbs_type_attrs_flow = { + .type = UVERBS_TYPE_ALLOC_IDR(0), + .destroy_object = uverbs_free_flow, +}; + +const struct uverbs_obj_idr_type uverbs_type_attrs_wq = { + .type = UVERBS_TYPE_ALLOC_IDR_SZ(sizeof(struct ib_uwq_object), 0), + .destroy_object = uverbs_free_wq, +}; + +const struct uverbs_obj_idr_type uverbs_type_attrs_rwq_ind_table = { + .type = UVERBS_TYPE_ALLOC_IDR(0), + .destroy_object = uverbs_free_rwq_ind_tbl, +}; + +const struct uverbs_obj_idr_type uverbs_type_attrs_xrcd = { + .type = UVERBS_TYPE_ALLOC_IDR_SZ(sizeof(struct ib_uxrcd_object), 0), + .destroy_object = uverbs_free_xrcd, +}; + +const struct uverbs_obj_idr_type uverbs_type_attrs_pd = { + /* 2 is used in order to free the PD after MRs */ + .type = UVERBS_TYPE_ALLOC_IDR(2), + .destroy_object = uverbs_free_pd, +}; diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c index 85ed5051fdfd..207e5c2457cc 100644 --- a/drivers/infiniband/core/verbs.c +++ b/drivers/infiniband/core/verbs.c @@ -1519,7 +1519,9 @@ int ib_attach_mcast(struct ib_qp *qp, union ib_gid *gid, u16 lid) if (!qp->device->attach_mcast) return -ENOSYS; - if (gid->raw[0] != 0xff || qp->qp_type != IB_QPT_UD) + if (gid->raw[0] != 0xff || qp->qp_type != IB_QPT_UD || + lid < be16_to_cpu(IB_MULTICAST_LID_BASE) || + lid == be16_to_cpu(IB_LID_PERMISSIVE)) return -EINVAL; ret = qp->device->attach_mcast(qp, gid, lid); @@ -1535,7 +1537,9 @@ int ib_detach_mcast(struct ib_qp *qp, union ib_gid *gid, u16 lid) if (!qp->device->detach_mcast) return -ENOSYS; - if (gid->raw[0] != 0xff || qp->qp_type != IB_QPT_UD) + if (gid->raw[0] != 0xff || qp->qp_type != IB_QPT_UD || + lid < be16_to_cpu(IB_MULTICAST_LID_BASE) || + lid == be16_to_cpu(IB_LID_PERMISSIVE)) return -EINVAL; ret = qp->device->detach_mcast(qp, gid, lid); diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.c b/drivers/infiniband/hw/bnxt_re/ib_verbs.c index 33af2e3de399..a4e8e0b075d2 100644 --- a/drivers/infiniband/hw/bnxt_re/ib_verbs.c +++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.c @@ -692,9 +692,9 @@ int bnxt_re_destroy_qp(struct ib_qp *ib_qp) kfree(rdev->qp1_sqp); } - if (qp->rumem && !IS_ERR(qp->rumem)) + if (!IS_ERR_OR_NULL(qp->rumem)) ib_umem_release(qp->rumem); - if (qp->sumem && !IS_ERR(qp->sumem)) + if (!IS_ERR_OR_NULL(qp->sumem)) ib_umem_release(qp->sumem); mutex_lock(&rdev->qp_lock); @@ -2116,7 +2116,7 @@ int bnxt_re_destroy_cq(struct ib_cq *ib_cq) dev_err(rdev_to_dev(rdev), "Failed to destroy HW CQ"); return rc; } - if (cq->umem && !IS_ERR(cq->umem)) + if (!IS_ERR_OR_NULL(cq->umem)) ib_umem_release(cq->umem); if (cq) { @@ -2818,7 +2818,7 @@ int bnxt_re_dereg_mr(struct ib_mr *ib_mr) { struct bnxt_re_mr *mr = container_of(ib_mr, struct bnxt_re_mr, ib_mr); struct bnxt_re_dev *rdev = mr->rdev; - int rc = 0; + int rc; if (mr->npages && mr->pages) { rc = bnxt_qplib_free_fast_reg_page_list(&rdev->qplib_res, @@ -2829,7 +2829,7 @@ int bnxt_re_dereg_mr(struct ib_mr *ib_mr) } rc = bnxt_qplib_free_mrw(&rdev->qplib_res, &mr->qplib_mr); - if (!IS_ERR(mr->ib_umem) && mr->ib_umem) + if (!IS_ERR_OR_NULL(mr->ib_umem)) ib_umem_release(mr->ib_umem); kfree(mr); @@ -3016,7 +3016,7 @@ struct ib_mr *bnxt_re_reg_user_mr(struct ib_pd *ib_pd, u64 start, u64 length, struct bnxt_re_mr *mr; struct ib_umem *umem; u64 *pbl_tbl, *pbl_tbl_orig; - int i, umem_pgs, pages, page_shift, rc; + int i, umem_pgs, pages, rc; struct scatterlist *sg; int entry; @@ -3062,22 +3062,22 @@ struct ib_mr *bnxt_re_reg_user_mr(struct ib_pd *ib_pd, u64 start, u64 length, } pbl_tbl_orig = pbl_tbl; - page_shift = ilog2(umem->page_size); if (umem->hugetlb) { dev_err(rdev_to_dev(rdev), "umem hugetlb not supported!"); rc = -EFAULT; goto fail; } - if (umem->page_size != PAGE_SIZE) { - dev_err(rdev_to_dev(rdev), "umem page size unsupported!"); + + if (umem->page_shift != PAGE_SHIFT) { + dev_err(rdev_to_dev(rdev), "umem page shift unsupported!"); rc = -EFAULT; goto fail; } /* Map umem buf ptrs to the PBL */ for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) { - pages = sg_dma_len(sg) >> page_shift; + pages = sg_dma_len(sg) >> umem->page_shift; for (i = 0; i < pages; i++, pbl_tbl++) - *pbl_tbl = sg_dma_address(sg) + (i << page_shift); + *pbl_tbl = sg_dma_address(sg) + (i << umem->page_shift); } rc = bnxt_qplib_reg_mr(&rdev->qplib_res, &mr->qplib_mr, pbl_tbl_orig, umem_pgs, false); diff --git a/drivers/infiniband/hw/cxgb3/cxio_dbg.c b/drivers/infiniband/hw/cxgb3/cxio_dbg.c index 445e89e5e7cf..97dbe728520a 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_dbg.c +++ b/drivers/infiniband/hw/cxgb3/cxio_dbg.c @@ -51,17 +51,18 @@ void cxio_dump_tpt(struct cxio_rdev *rdev, u32 stag) m->mem_id = MEM_PMRX; m->addr = (stag>>8) * 32 + rdev->rnic_info.tpt_base; m->len = size; - PDBG("%s TPT addr 0x%x len %d\n", __func__, m->addr, m->len); + pr_debug("%s TPT addr 0x%x len %d\n", __func__, m->addr, m->len); rc = rdev->t3cdev_p->ctl(rdev->t3cdev_p, RDMA_GET_MEM, m); if (rc) { - PDBG("%s toectl returned error %d\n", __func__, rc); + pr_debug("%s toectl returned error %d\n", __func__, rc); kfree(m); return; } data = (u64 *)m->buf; while (size > 0) { - PDBG("TPT %08x: %016llx\n", m->addr, (unsigned long long) *data); + pr_debug("TPT %08x: %016llx\n", + m->addr, (unsigned long long)*data); size -= 8; data++; m->addr += 8; @@ -87,18 +88,19 @@ void cxio_dump_pbl(struct cxio_rdev *rdev, u32 pbl_addr, uint len, u8 shift) m->mem_id = MEM_PMRX; m->addr = pbl_addr; m->len = size; - PDBG("%s PBL addr 0x%x len %d depth %d\n", - __func__, m->addr, m->len, npages); + pr_debug("%s PBL addr 0x%x len %d depth %d\n", + __func__, m->addr, m->len, npages); rc = rdev->t3cdev_p->ctl(rdev->t3cdev_p, RDMA_GET_MEM, m); if (rc) { - PDBG("%s toectl returned error %d\n", __func__, rc); + pr_debug("%s toectl returned error %d\n", __func__, rc); kfree(m); return; } data = (u64 *)m->buf; while (size > 0) { - PDBG("PBL %08x: %016llx\n", m->addr, (unsigned long long) *data); + pr_debug("PBL %08x: %016llx\n", + m->addr, (unsigned long long)*data); size -= 8; data++; m->addr += 8; @@ -114,8 +116,8 @@ void cxio_dump_wqe(union t3_wr *wqe) if (size == 0) size = 8; while (size > 0) { - PDBG("WQE %p: %016llx\n", data, - (unsigned long long) be64_to_cpu(*data)); + pr_debug("WQE %p: %016llx\n", + data, (unsigned long long)be64_to_cpu(*data)); size--; data++; } @@ -127,8 +129,8 @@ void cxio_dump_wce(struct t3_cqe *wce) int size = sizeof(*wce); while (size > 0) { - PDBG("WCE %p: %016llx\n", data, - (unsigned long long) be64_to_cpu(*data)); + pr_debug("WCE %p: %016llx\n", + data, (unsigned long long)be64_to_cpu(*data)); size -= 8; data++; } @@ -148,17 +150,18 @@ void cxio_dump_rqt(struct cxio_rdev *rdev, u32 hwtid, int nents) m->mem_id = MEM_PMRX; m->addr = ((hwtid)<<10) + rdev->rnic_info.rqt_base; m->len = size; - PDBG("%s RQT addr 0x%x len %d\n", __func__, m->addr, m->len); + pr_debug("%s RQT addr 0x%x len %d\n", __func__, m->addr, m->len); rc = rdev->t3cdev_p->ctl(rdev->t3cdev_p, RDMA_GET_MEM, m); if (rc) { - PDBG("%s toectl returned error %d\n", __func__, rc); + pr_debug("%s toectl returned error %d\n", __func__, rc); kfree(m); return; } data = (u64 *)m->buf; while (size > 0) { - PDBG("RQT %08x: %016llx\n", m->addr, (unsigned long long) *data); + pr_debug("RQT %08x: %016llx\n", + m->addr, (unsigned long long)*data); size -= 8; data++; m->addr += 8; @@ -180,10 +183,10 @@ void cxio_dump_tcb(struct cxio_rdev *rdev, u32 hwtid) m->mem_id = MEM_CM; m->addr = hwtid * size; m->len = size; - PDBG("%s TCB %d len %d\n", __func__, m->addr, m->len); + pr_debug("%s TCB %d len %d\n", __func__, m->addr, m->len); rc = rdev->t3cdev_p->ctl(rdev->t3cdev_p, RDMA_GET_MEM, m); if (rc) { - PDBG("%s toectl returned error %d\n", __func__, rc); + pr_debug("%s toectl returned error %d\n", __func__, rc); kfree(m); return; } diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.c b/drivers/infiniband/hw/cxgb3/cxio_hal.c index ada2e5009c86..558d6a03375d 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_hal.c +++ b/drivers/infiniband/hw/cxgb3/cxio_hal.c @@ -110,8 +110,7 @@ int cxio_hal_cq_op(struct cxio_rdev *rdev_p, struct t3_cq *cq, while (!CQ_VLD_ENTRY(rptr, cq->size_log2, cqe)) { udelay(1); if (i++ > 1000000) { - printk(KERN_ERR "%s: stalled rnic\n", - rdev_p->dev_name); + pr_err("%s: stalled rnic\n", rdev_p->dev_name); return -EIO; } } @@ -140,7 +139,7 @@ static int cxio_hal_clear_qp_ctx(struct cxio_rdev *rdev_p, u32 qpid) struct t3_modify_qp_wr *wqe; struct sk_buff *skb = alloc_skb(sizeof(*wqe), GFP_KERNEL); if (!skb) { - PDBG("%s alloc_skb failed\n", __func__); + pr_debug("%s alloc_skb failed\n", __func__); return -ENOMEM; } wqe = (struct t3_modify_qp_wr *) skb_put(skb, sizeof(*wqe)); @@ -230,7 +229,7 @@ static u32 get_qpid(struct cxio_rdev *rdev_p, struct cxio_ucontext *uctx) } out: mutex_unlock(&uctx->lock); - PDBG("%s qpid 0x%x\n", __func__, qpid); + pr_debug("%s qpid 0x%x\n", __func__, qpid); return qpid; } @@ -242,7 +241,7 @@ static void put_qpid(struct cxio_rdev *rdev_p, u32 qpid, entry = kmalloc(sizeof *entry, GFP_KERNEL); if (!entry) return; - PDBG("%s qpid 0x%x\n", __func__, qpid); + pr_debug("%s qpid 0x%x\n", __func__, qpid); entry->qpid = qpid; mutex_lock(&uctx->lock); list_add_tail(&entry->entry, &uctx->qpids); @@ -306,8 +305,8 @@ int cxio_create_qp(struct cxio_rdev *rdev_p, u32 kernel_domain, wq->udb = (u64)rdev_p->rnic_info.udbell_physbase + (wq->qpid << rdev_p->qpshift); wq->rdev = rdev_p; - PDBG("%s qpid 0x%x doorbell 0x%p udb 0x%llx\n", __func__, - wq->qpid, wq->doorbell, (unsigned long long) wq->udb); + pr_debug("%s qpid 0x%x doorbell 0x%p udb 0x%llx\n", + __func__, wq->qpid, wq->doorbell, (unsigned long long)wq->udb); return 0; err4: kfree(wq->sq); @@ -351,8 +350,8 @@ static void insert_recv_cqe(struct t3_wq *wq, struct t3_cq *cq) { struct t3_cqe cqe; - PDBG("%s wq %p cq %p sw_rptr 0x%x sw_wptr 0x%x\n", __func__, - wq, cq, cq->sw_rptr, cq->sw_wptr); + pr_debug("%s wq %p cq %p sw_rptr 0x%x sw_wptr 0x%x\n", __func__, + wq, cq, cq->sw_rptr, cq->sw_wptr); memset(&cqe, 0, sizeof(cqe)); cqe.header = cpu_to_be32(V_CQE_STATUS(TPT_ERR_SWFLUSH) | V_CQE_OPCODE(T3_SEND) | @@ -370,11 +369,11 @@ int cxio_flush_rq(struct t3_wq *wq, struct t3_cq *cq, int count) u32 ptr; int flushed = 0; - PDBG("%s wq %p cq %p\n", __func__, wq, cq); + pr_debug("%s wq %p cq %p\n", __func__, wq, cq); /* flush RQ */ - PDBG("%s rq_rptr %u rq_wptr %u skip count %u\n", __func__, - wq->rq_rptr, wq->rq_wptr, count); + pr_debug("%s rq_rptr %u rq_wptr %u skip count %u\n", __func__, + wq->rq_rptr, wq->rq_wptr, count); ptr = wq->rq_rptr + count; while (ptr++ != wq->rq_wptr) { insert_recv_cqe(wq, cq); @@ -388,8 +387,8 @@ static void insert_sq_cqe(struct t3_wq *wq, struct t3_cq *cq, { struct t3_cqe cqe; - PDBG("%s wq %p cq %p sw_rptr 0x%x sw_wptr 0x%x\n", __func__, - wq, cq, cq->sw_rptr, cq->sw_wptr); + pr_debug("%s wq %p cq %p sw_rptr 0x%x sw_wptr 0x%x\n", __func__, + wq, cq, cq->sw_rptr, cq->sw_wptr); memset(&cqe, 0, sizeof(cqe)); cqe.header = cpu_to_be32(V_CQE_STATUS(TPT_ERR_SWFLUSH) | V_CQE_OPCODE(sqp->opcode) | @@ -429,11 +428,11 @@ void cxio_flush_hw_cq(struct t3_cq *cq) { struct t3_cqe *cqe, *swcqe; - PDBG("%s cq %p cqid 0x%x\n", __func__, cq, cq->cqid); + pr_debug("%s cq %p cqid 0x%x\n", __func__, cq, cq->cqid); cqe = cxio_next_hw_cqe(cq); while (cqe) { - PDBG("%s flushing hwcq rptr 0x%x to swcq wptr 0x%x\n", - __func__, cq->rptr, cq->sw_wptr); + pr_debug("%s flushing hwcq rptr 0x%x to swcq wptr 0x%x\n", + __func__, cq->rptr, cq->sw_wptr); swcqe = cq->sw_queue + Q_PTR2IDX(cq->sw_wptr, cq->size_log2); *swcqe = *cqe; swcqe->header |= cpu_to_be32(V_CQE_SWCQE(1)); @@ -476,7 +475,7 @@ void cxio_count_scqes(struct t3_cq *cq, struct t3_wq *wq, int *count) (*count)++; ptr++; } - PDBG("%s cq %p count %d\n", __func__, cq, *count); + pr_debug("%s cq %p count %d\n", __func__, cq, *count); } void cxio_count_rcqes(struct t3_cq *cq, struct t3_wq *wq, int *count) @@ -485,7 +484,7 @@ void cxio_count_rcqes(struct t3_cq *cq, struct t3_wq *wq, int *count) u32 ptr; *count = 0; - PDBG("%s count zero %d\n", __func__, *count); + pr_debug("%s count zero %d\n", __func__, *count); ptr = cq->sw_rptr; while (!Q_EMPTY(ptr, cq->sw_wptr)) { cqe = cq->sw_queue + (Q_PTR2IDX(ptr, cq->size_log2)); @@ -494,7 +493,7 @@ void cxio_count_rcqes(struct t3_cq *cq, struct t3_wq *wq, int *count) (*count)++; ptr++; } - PDBG("%s cq %p count %d\n", __func__, cq, *count); + pr_debug("%s cq %p count %d\n", __func__, cq, *count); } static int cxio_hal_init_ctrl_cq(struct cxio_rdev *rdev_p) @@ -521,12 +520,12 @@ static int cxio_hal_init_ctrl_qp(struct cxio_rdev *rdev_p) skb = alloc_skb(sizeof(*wqe), GFP_KERNEL); if (!skb) { - PDBG("%s alloc_skb failed\n", __func__); + pr_debug("%s alloc_skb failed\n", __func__); return -ENOMEM; } err = cxio_hal_init_ctrl_cq(rdev_p); if (err) { - PDBG("%s err %d initializing ctrl_cq\n", __func__, err); + pr_debug("%s err %d initializing ctrl_cq\n", __func__, err); goto err; } rdev_p->ctrl_qp.workq = dma_alloc_coherent( @@ -536,7 +535,7 @@ static int cxio_hal_init_ctrl_qp(struct cxio_rdev *rdev_p) &(rdev_p->ctrl_qp.dma_addr), GFP_KERNEL); if (!rdev_p->ctrl_qp.workq) { - PDBG("%s dma_alloc_coherent failed\n", __func__); + pr_debug("%s dma_alloc_coherent failed\n", __func__); err = -ENOMEM; goto err; } @@ -571,9 +570,9 @@ static int cxio_hal_init_ctrl_qp(struct cxio_rdev *rdev_p) wqe->sge_cmd = cpu_to_be64(sge_cmd); wqe->ctx1 = cpu_to_be64(ctx1); wqe->ctx0 = cpu_to_be64(ctx0); - PDBG("CtrlQP dma_addr 0x%llx workq %p size %d\n", - (unsigned long long) rdev_p->ctrl_qp.dma_addr, - rdev_p->ctrl_qp.workq, 1 << T3_CTRL_QP_SIZE_LOG2); + pr_debug("CtrlQP dma_addr 0x%llx workq %p size %d\n", + (unsigned long long)rdev_p->ctrl_qp.dma_addr, + rdev_p->ctrl_qp.workq, 1 << T3_CTRL_QP_SIZE_LOG2); skb->priority = CPL_PRIORITY_CONTROL; return iwch_cxgb3_ofld_send(rdev_p->t3cdev_p, skb); err: @@ -605,26 +604,26 @@ static int cxio_hal_ctrl_qp_write_mem(struct cxio_rdev *rdev_p, u32 addr, u64 utx_cmd; addr &= 0x7FFFFFF; nr_wqe = len % 96 ? len / 96 + 1 : len / 96; /* 96B max per WQE */ - PDBG("%s wptr 0x%x rptr 0x%x len %d, nr_wqe %d data %p addr 0x%0x\n", - __func__, rdev_p->ctrl_qp.wptr, rdev_p->ctrl_qp.rptr, len, - nr_wqe, data, addr); + pr_debug("%s wptr 0x%x rptr 0x%x len %d, nr_wqe %d data %p addr 0x%0x\n", + __func__, rdev_p->ctrl_qp.wptr, rdev_p->ctrl_qp.rptr, len, + nr_wqe, data, addr); utx_len = 3; /* in 32B unit */ for (i = 0; i < nr_wqe; i++) { if (Q_FULL(rdev_p->ctrl_qp.rptr, rdev_p->ctrl_qp.wptr, T3_CTRL_QP_SIZE_LOG2)) { - PDBG("%s ctrl_qp full wtpr 0x%0x rptr 0x%0x, " - "wait for more space i %d\n", __func__, - rdev_p->ctrl_qp.wptr, rdev_p->ctrl_qp.rptr, i); + pr_debug("%s ctrl_qp full wtpr 0x%0x rptr 0x%0x, wait for more space i %d\n", + __func__, + rdev_p->ctrl_qp.wptr, rdev_p->ctrl_qp.rptr, i); if (wait_event_interruptible(rdev_p->ctrl_qp.waitq, !Q_FULL(rdev_p->ctrl_qp.rptr, rdev_p->ctrl_qp.wptr, T3_CTRL_QP_SIZE_LOG2))) { - PDBG("%s ctrl_qp workq interrupted\n", - __func__); + pr_debug("%s ctrl_qp workq interrupted\n", + __func__); return -ERESTARTSYS; } - PDBG("%s ctrl_qp wakeup, continue posting work request " - "i %d\n", __func__, i); + pr_debug("%s ctrl_qp wakeup, continue posting work request i %d\n", + __func__, i); } wqe = (__be64 *)(rdev_p->ctrl_qp.workq + (rdev_p->ctrl_qp.wptr % (1 << T3_CTRL_QP_SIZE_LOG2))); @@ -645,7 +644,7 @@ static int cxio_hal_ctrl_qp_write_mem(struct cxio_rdev *rdev_p, u32 addr, if ((i != 0) && (i % (((1 << T3_CTRL_QP_SIZE_LOG2)) >> 1) == 0)) { flag = T3_COMPLETION_FLAG; - PDBG("%s force completion at i %d\n", __func__, i); + pr_debug("%s force completion at i %d\n", __func__, i); } /* build the utx mem command */ @@ -717,8 +716,8 @@ static int __cxio_tpt_op(struct cxio_rdev *rdev_p, u32 reset_tpt_entry, return -ENOMEM; *stag = (stag_idx << 8) | ((*stag) & 0xFF); } - PDBG("%s stag_state 0x%0x type 0x%0x pdid 0x%0x, stag_idx 0x%x\n", - __func__, stag_state, type, pdid, stag_idx); + pr_debug("%s stag_state 0x%0x type 0x%0x pdid 0x%0x, stag_idx 0x%x\n", + __func__, stag_state, type, pdid, stag_idx); mutex_lock(&rdev_p->ctrl_qp.lock); @@ -767,9 +766,9 @@ int cxio_write_pbl(struct cxio_rdev *rdev_p, __be64 *pbl, u32 wptr; int err; - PDBG("%s *pdb_addr 0x%x, pbl_base 0x%x, pbl_size %d\n", - __func__, pbl_addr, rdev_p->rnic_info.pbl_base, - pbl_size); + pr_debug("%s *pdb_addr 0x%x, pbl_base 0x%x, pbl_size %d\n", + __func__, pbl_addr, rdev_p->rnic_info.pbl_base, + pbl_size); mutex_lock(&rdev_p->ctrl_qp.lock); err = cxio_hal_ctrl_qp_write_mem(rdev_p, pbl_addr >> 5, pbl_size << 3, @@ -837,7 +836,7 @@ int cxio_rdma_init(struct cxio_rdev *rdev_p, struct t3_rdma_init_attr *attr) struct sk_buff *skb = alloc_skb(sizeof(*wqe), GFP_ATOMIC); if (!skb) return -ENOMEM; - PDBG("%s rdev_p %p\n", __func__, rdev_p); + pr_debug("%s rdev_p %p\n", __func__, rdev_p); wqe = (struct t3_rdma_init_wr *) __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) | @@ -880,22 +879,20 @@ static int cxio_hal_ev_handler(struct t3cdev *t3cdev_p, struct sk_buff *skb) static int cnt; struct cxio_rdev *rdev_p = NULL; struct respQ_msg_t *rsp_msg = (struct respQ_msg_t *) skb->data; - PDBG("%d: %s cq_id 0x%x cq_ptr 0x%x genbit %0x overflow %0x an %0x" - " se %0x notify %0x cqbranch %0x creditth %0x\n", - cnt, __func__, RSPQ_CQID(rsp_msg), RSPQ_CQPTR(rsp_msg), - RSPQ_GENBIT(rsp_msg), RSPQ_OVERFLOW(rsp_msg), RSPQ_AN(rsp_msg), - RSPQ_SE(rsp_msg), RSPQ_NOTIFY(rsp_msg), RSPQ_CQBRANCH(rsp_msg), - RSPQ_CREDIT_THRESH(rsp_msg)); - PDBG("CQE: QPID 0x%0x genbit %0x type 0x%0x status 0x%0x opcode %d " - "len 0x%0x wrid_hi_stag 0x%x wrid_low_msn 0x%x\n", - CQE_QPID(rsp_msg->cqe), CQE_GENBIT(rsp_msg->cqe), - CQE_TYPE(rsp_msg->cqe), CQE_STATUS(rsp_msg->cqe), - CQE_OPCODE(rsp_msg->cqe), CQE_LEN(rsp_msg->cqe), - CQE_WRID_HI(rsp_msg->cqe), CQE_WRID_LOW(rsp_msg->cqe)); + pr_debug("%d: %s cq_id 0x%x cq_ptr 0x%x genbit %0x overflow %0x an %0x se %0x notify %0x cqbranch %0x creditth %0x\n", + cnt, __func__, RSPQ_CQID(rsp_msg), RSPQ_CQPTR(rsp_msg), + RSPQ_GENBIT(rsp_msg), RSPQ_OVERFLOW(rsp_msg), RSPQ_AN(rsp_msg), + RSPQ_SE(rsp_msg), RSPQ_NOTIFY(rsp_msg), RSPQ_CQBRANCH(rsp_msg), + RSPQ_CREDIT_THRESH(rsp_msg)); + pr_debug("CQE: QPID 0x%0x genbit %0x type 0x%0x status 0x%0x opcode %d len 0x%0x wrid_hi_stag 0x%x wrid_low_msn 0x%x\n", + CQE_QPID(rsp_msg->cqe), CQE_GENBIT(rsp_msg->cqe), + CQE_TYPE(rsp_msg->cqe), CQE_STATUS(rsp_msg->cqe), + CQE_OPCODE(rsp_msg->cqe), CQE_LEN(rsp_msg->cqe), + CQE_WRID_HI(rsp_msg->cqe), CQE_WRID_LOW(rsp_msg->cqe)); rdev_p = (struct cxio_rdev *)t3cdev_p->ulp; if (!rdev_p) { - PDBG("%s called by t3cdev %p with null ulp\n", __func__, - t3cdev_p); + pr_debug("%s called by t3cdev %p with null ulp\n", __func__, + t3cdev_p); return 0; } if (CQE_QPID(rsp_msg->cqe) == T3_CTRL_QP_ID) { @@ -934,13 +931,13 @@ int cxio_rdev_open(struct cxio_rdev *rdev_p) strncpy(rdev_p->dev_name, rdev_p->t3cdev_p->name, T3_MAX_DEV_NAME_LEN); } else { - PDBG("%s t3cdev_p or dev_name must be set\n", __func__); + pr_debug("%s t3cdev_p or dev_name must be set\n", __func__); return -EINVAL; } list_add_tail(&rdev_p->entry, &rdev_list); - PDBG("%s opening rnic dev %s\n", __func__, rdev_p->dev_name); + pr_debug("%s opening rnic dev %s\n", __func__, rdev_p->dev_name); memset(&rdev_p->ctrl_qp, 0, sizeof(rdev_p->ctrl_qp)); if (!rdev_p->t3cdev_p) rdev_p->t3cdev_p = dev2t3cdev(netdev_p); @@ -949,13 +946,12 @@ int cxio_rdev_open(struct cxio_rdev *rdev_p) err = rdev_p->t3cdev_p->ctl(rdev_p->t3cdev_p, GET_EMBEDDED_INFO, &(rdev_p->fw_info)); if (err) { - printk(KERN_ERR "%s t3cdev_p(%p)->ctl returned error %d.\n", - __func__, rdev_p->t3cdev_p, err); + pr_err("%s t3cdev_p(%p)->ctl returned error %d\n", + __func__, rdev_p->t3cdev_p, err); goto err1; } if (G_FW_VERSION_MAJOR(rdev_p->fw_info.fw_vers) != CXIO_FW_MAJ) { - printk(KERN_ERR MOD "fatal firmware version mismatch: " - "need version %u but adapter has version %u\n", + pr_err("fatal firmware version mismatch: need version %u but adapter has version %u\n", CXIO_FW_MAJ, G_FW_VERSION_MAJOR(rdev_p->fw_info.fw_vers)); err = -EINVAL; @@ -965,15 +961,15 @@ int cxio_rdev_open(struct cxio_rdev *rdev_p) err = rdev_p->t3cdev_p->ctl(rdev_p->t3cdev_p, RDMA_GET_PARAMS, &(rdev_p->rnic_info)); if (err) { - printk(KERN_ERR "%s t3cdev_p(%p)->ctl returned error %d.\n", - __func__, rdev_p->t3cdev_p, err); + pr_err("%s t3cdev_p(%p)->ctl returned error %d\n", + __func__, rdev_p->t3cdev_p, err); goto err1; } err = rdev_p->t3cdev_p->ctl(rdev_p->t3cdev_p, GET_PORTS, &(rdev_p->port_info)); if (err) { - printk(KERN_ERR "%s t3cdev_p(%p)->ctl returned error %d.\n", - __func__, rdev_p->t3cdev_p, err); + pr_err("%s t3cdev_p(%p)->ctl returned error %d\n", + __func__, rdev_p->t3cdev_p, err); goto err1; } @@ -988,42 +984,39 @@ int cxio_rdev_open(struct cxio_rdev *rdev_p) PAGE_SHIFT)); rdev_p->qpnr = rdev_p->rnic_info.udbell_len >> PAGE_SHIFT; rdev_p->qpmask = (65536 >> ilog2(rdev_p->qpnr)) - 1; - PDBG("%s rnic %s info: tpt_base 0x%0x tpt_top 0x%0x num stags %d " - "pbl_base 0x%0x pbl_top 0x%0x rqt_base 0x%0x, rqt_top 0x%0x\n", - __func__, rdev_p->dev_name, rdev_p->rnic_info.tpt_base, - rdev_p->rnic_info.tpt_top, cxio_num_stags(rdev_p), - rdev_p->rnic_info.pbl_base, - rdev_p->rnic_info.pbl_top, rdev_p->rnic_info.rqt_base, - rdev_p->rnic_info.rqt_top); - PDBG("udbell_len 0x%0x udbell_physbase 0x%lx kdb_addr %p qpshift %lu " - "qpnr %d qpmask 0x%x\n", - rdev_p->rnic_info.udbell_len, - rdev_p->rnic_info.udbell_physbase, rdev_p->rnic_info.kdb_addr, - rdev_p->qpshift, rdev_p->qpnr, rdev_p->qpmask); + pr_debug("%s rnic %s info: tpt_base 0x%0x tpt_top 0x%0x num stags %d pbl_base 0x%0x pbl_top 0x%0x rqt_base 0x%0x, rqt_top 0x%0x\n", + __func__, rdev_p->dev_name, rdev_p->rnic_info.tpt_base, + rdev_p->rnic_info.tpt_top, cxio_num_stags(rdev_p), + rdev_p->rnic_info.pbl_base, + rdev_p->rnic_info.pbl_top, rdev_p->rnic_info.rqt_base, + rdev_p->rnic_info.rqt_top); + pr_debug("udbell_len 0x%0x udbell_physbase 0x%lx kdb_addr %p qpshift %lu qpnr %d qpmask 0x%x\n", + rdev_p->rnic_info.udbell_len, + rdev_p->rnic_info.udbell_physbase, rdev_p->rnic_info.kdb_addr, + rdev_p->qpshift, rdev_p->qpnr, rdev_p->qpmask); err = cxio_hal_init_ctrl_qp(rdev_p); if (err) { - printk(KERN_ERR "%s error %d initializing ctrl_qp.\n", - __func__, err); + pr_err("%s error %d initializing ctrl_qp\n", __func__, err); goto err1; } err = cxio_hal_init_resource(rdev_p, cxio_num_stags(rdev_p), 0, 0, T3_MAX_NUM_QP, T3_MAX_NUM_CQ, T3_MAX_NUM_PD); if (err) { - printk(KERN_ERR "%s error %d initializing hal resources.\n", + pr_err("%s error %d initializing hal resources\n", __func__, err); goto err2; } err = cxio_hal_pblpool_create(rdev_p); if (err) { - printk(KERN_ERR "%s error %d initializing pbl mem pool.\n", + pr_err("%s error %d initializing pbl mem pool\n", __func__, err); goto err3; } err = cxio_hal_rqtpool_create(rdev_p); if (err) { - printk(KERN_ERR "%s error %d initializing rqt mem pool.\n", + pr_err("%s error %d initializing rqt mem pool\n", __func__, err); goto err4; } @@ -1086,9 +1079,9 @@ static void flush_completed_wrs(struct t3_wq *wq, struct t3_cq *cq) /* * Insert this completed cqe into the swcq. */ - PDBG("%s moving cqe into swcq sq idx %ld cq idx %ld\n", - __func__, Q_PTR2IDX(ptr, wq->sq_size_log2), - Q_PTR2IDX(cq->sw_wptr, cq->size_log2)); + pr_debug("%s moving cqe into swcq sq idx %ld cq idx %ld\n", + __func__, Q_PTR2IDX(ptr, wq->sq_size_log2), + Q_PTR2IDX(cq->sw_wptr, cq->size_log2)); sqp->cqe.header |= htonl(V_CQE_SWCQE(1)); *(cq->sw_queue + Q_PTR2IDX(cq->sw_wptr, cq->size_log2)) = sqp->cqe; @@ -1154,12 +1147,11 @@ int cxio_poll_cq(struct t3_wq *wq, struct t3_cq *cq, struct t3_cqe *cqe, *credit = 0; hw_cqe = cxio_next_cqe(cq); - PDBG("%s CQE OOO %d qpid 0x%0x genbit %d type %d status 0x%0x" - " opcode 0x%0x len 0x%0x wrid_hi_stag 0x%x wrid_low_msn 0x%x\n", - __func__, CQE_OOO(*hw_cqe), CQE_QPID(*hw_cqe), - CQE_GENBIT(*hw_cqe), CQE_TYPE(*hw_cqe), CQE_STATUS(*hw_cqe), - CQE_OPCODE(*hw_cqe), CQE_LEN(*hw_cqe), CQE_WRID_HI(*hw_cqe), - CQE_WRID_LOW(*hw_cqe)); + pr_debug("%s CQE OOO %d qpid 0x%0x genbit %d type %d status 0x%0x opcode 0x%0x len 0x%0x wrid_hi_stag 0x%x wrid_low_msn 0x%x\n", + __func__, CQE_OOO(*hw_cqe), CQE_QPID(*hw_cqe), + CQE_GENBIT(*hw_cqe), CQE_TYPE(*hw_cqe), CQE_STATUS(*hw_cqe), + CQE_OPCODE(*hw_cqe), CQE_LEN(*hw_cqe), CQE_WRID_HI(*hw_cqe), + CQE_WRID_LOW(*hw_cqe)); /* * skip cqe's not affiliated with a QP. @@ -1278,9 +1270,10 @@ int cxio_poll_cq(struct t3_wq *wq, struct t3_cq *cq, struct t3_cqe *cqe, if (!SW_CQE(*hw_cqe) && (CQE_WRID_SQ_WPTR(*hw_cqe) != wq->sq_rptr)) { struct t3_swsq *sqp; - PDBG("%s out of order completion going in swsq at idx %ld\n", - __func__, - Q_PTR2IDX(CQE_WRID_SQ_WPTR(*hw_cqe), wq->sq_size_log2)); + pr_debug("%s out of order completion going in swsq at idx %ld\n", + __func__, + Q_PTR2IDX(CQE_WRID_SQ_WPTR(*hw_cqe), + wq->sq_size_log2)); sqp = wq->sq + Q_PTR2IDX(CQE_WRID_SQ_WPTR(*hw_cqe), wq->sq_size_log2); sqp->cqe = *hw_cqe; @@ -1298,13 +1291,13 @@ proc_cqe: */ if (SQ_TYPE(*hw_cqe)) { wq->sq_rptr = CQE_WRID_SQ_WPTR(*hw_cqe); - PDBG("%s completing sq idx %ld\n", __func__, - Q_PTR2IDX(wq->sq_rptr, wq->sq_size_log2)); + pr_debug("%s completing sq idx %ld\n", __func__, + Q_PTR2IDX(wq->sq_rptr, wq->sq_size_log2)); *cookie = wq->sq[Q_PTR2IDX(wq->sq_rptr, wq->sq_size_log2)].wr_id; wq->sq_rptr++; } else { - PDBG("%s completing rq idx %ld\n", __func__, - Q_PTR2IDX(wq->rq_rptr, wq->rq_size_log2)); + pr_debug("%s completing rq idx %ld\n", __func__, + Q_PTR2IDX(wq->rq_rptr, wq->rq_size_log2)); *cookie = wq->rq[Q_PTR2IDX(wq->rq_rptr, wq->rq_size_log2)].wr_id; if (wq->rq[Q_PTR2IDX(wq->rq_rptr, wq->rq_size_log2)].pbl_addr) cxio_hal_pblpool_free(wq->rdev, @@ -1322,12 +1315,12 @@ flush_wq: skip_cqe: if (SW_CQE(*hw_cqe)) { - PDBG("%s cq %p cqid 0x%x skip sw cqe sw_rptr 0x%x\n", - __func__, cq, cq->cqid, cq->sw_rptr); + pr_debug("%s cq %p cqid 0x%x skip sw cqe sw_rptr 0x%x\n", + __func__, cq, cq->cqid, cq->sw_rptr); ++cq->sw_rptr; } else { - PDBG("%s cq %p cqid 0x%x skip hw cqe rptr 0x%x\n", - __func__, cq, cq->cqid, cq->rptr); + pr_debug("%s cq %p cqid 0x%x skip hw cqe rptr 0x%x\n", + __func__, cq, cq->cqid, cq->rptr); ++cq->rptr; /* diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.h b/drivers/infiniband/hw/cxgb3/cxio_hal.h index 78fbe9ffe7f0..7e70c5492262 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_hal.h +++ b/drivers/infiniband/hw/cxgb3/cxio_hal.h @@ -196,8 +196,11 @@ int cxio_poll_cq(struct t3_wq *wq, struct t3_cq *cq, struct t3_cqe *cqe, u8 *cqe_flushed, u64 *cookie, u32 *credit); int iwch_cxgb3_ofld_send(struct t3cdev *tdev, struct sk_buff *skb); -#define MOD "iw_cxgb3: " -#define PDBG(fmt, args...) pr_debug(MOD fmt, ## args) +#ifdef pr_fmt +#undef pr_fmt +#endif + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #ifdef DEBUG void cxio_dump_tpt(struct cxio_rdev *rev, u32 stag); diff --git a/drivers/infiniband/hw/cxgb3/cxio_resource.c b/drivers/infiniband/hw/cxgb3/cxio_resource.c index c40088ecf9f3..c6e7bc4420b6 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_resource.c +++ b/drivers/infiniband/hw/cxgb3/cxio_resource.c @@ -209,13 +209,13 @@ u32 cxio_hal_get_qpid(struct cxio_hal_resource *rscp) { u32 qpid = cxio_hal_get_resource(&rscp->qpid_fifo, &rscp->qpid_fifo_lock); - PDBG("%s qpid 0x%x\n", __func__, qpid); + pr_debug("%s qpid 0x%x\n", __func__, qpid); return qpid; } void cxio_hal_put_qpid(struct cxio_hal_resource *rscp, u32 qpid) { - PDBG("%s qpid 0x%x\n", __func__, qpid); + pr_debug("%s qpid 0x%x\n", __func__, qpid); cxio_hal_put_resource(&rscp->qpid_fifo, &rscp->qpid_fifo_lock, qpid); } @@ -257,13 +257,13 @@ void cxio_hal_destroy_resource(struct cxio_hal_resource *rscp) u32 cxio_hal_pblpool_alloc(struct cxio_rdev *rdev_p, int size) { unsigned long addr = gen_pool_alloc(rdev_p->pbl_pool, size); - PDBG("%s addr 0x%x size %d\n", __func__, (u32)addr, size); + pr_debug("%s addr 0x%x size %d\n", __func__, (u32)addr, size); return (u32)addr; } void cxio_hal_pblpool_free(struct cxio_rdev *rdev_p, u32 addr, int size) { - PDBG("%s addr 0x%x size %d\n", __func__, addr, size); + pr_debug("%s addr 0x%x size %d\n", __func__, addr, size); gen_pool_free(rdev_p->pbl_pool, (unsigned long)addr, size); } @@ -282,17 +282,18 @@ int cxio_hal_pblpool_create(struct cxio_rdev *rdev_p) pbl_chunk = min(rdev_p->rnic_info.pbl_top - pbl_start + 1, pbl_chunk); if (gen_pool_add(rdev_p->pbl_pool, pbl_start, pbl_chunk, -1)) { - PDBG("%s failed to add PBL chunk (%x/%x)\n", - __func__, pbl_start, pbl_chunk); + pr_debug("%s failed to add PBL chunk (%x/%x)\n", + __func__, pbl_start, pbl_chunk); if (pbl_chunk <= 1024 << MIN_PBL_SHIFT) { - printk(KERN_WARNING MOD "%s: Failed to add all PBL chunks (%x/%x)\n", - __func__, pbl_start, rdev_p->rnic_info.pbl_top - pbl_start); + pr_warn("%s: Failed to add all PBL chunks (%x/%x)\n", + __func__, pbl_start, + rdev_p->rnic_info.pbl_top - pbl_start); return 0; } pbl_chunk >>= 1; } else { - PDBG("%s added PBL chunk (%x/%x)\n", - __func__, pbl_start, pbl_chunk); + pr_debug("%s added PBL chunk (%x/%x)\n", + __func__, pbl_start, pbl_chunk); pbl_start += pbl_chunk; } } @@ -315,13 +316,13 @@ void cxio_hal_pblpool_destroy(struct cxio_rdev *rdev_p) u32 cxio_hal_rqtpool_alloc(struct cxio_rdev *rdev_p, int size) { unsigned long addr = gen_pool_alloc(rdev_p->rqt_pool, size << 6); - PDBG("%s addr 0x%x size %d\n", __func__, (u32)addr, size << 6); + pr_debug("%s addr 0x%x size %d\n", __func__, (u32)addr, size << 6); return (u32)addr; } void cxio_hal_rqtpool_free(struct cxio_rdev *rdev_p, u32 addr, int size) { - PDBG("%s addr 0x%x size %d\n", __func__, addr, size << 6); + pr_debug("%s addr 0x%x size %d\n", __func__, addr, size << 6); gen_pool_free(rdev_p->rqt_pool, (unsigned long)addr, size << 6); } diff --git a/drivers/infiniband/hw/cxgb3/iwch.c b/drivers/infiniband/hw/cxgb3/iwch.c index b3e11329801d..47b2ce2ef203 100644 --- a/drivers/infiniband/hw/cxgb3/iwch.c +++ b/drivers/infiniband/hw/cxgb3/iwch.c @@ -105,7 +105,7 @@ static void iwch_db_drop_task(struct work_struct *work) static void rnic_init(struct iwch_dev *rnicp) { - PDBG("%s iwch_dev %p\n", __func__, rnicp); + pr_debug("%s iwch_dev %p\n", __func__, rnicp); idr_init(&rnicp->cqidr); idr_init(&rnicp->qpidr); idr_init(&rnicp->mmidr); @@ -145,12 +145,11 @@ static void open_rnic_dev(struct t3cdev *tdev) { struct iwch_dev *rnicp; - PDBG("%s t3cdev %p\n", __func__, tdev); - printk_once(KERN_INFO MOD "Chelsio T3 RDMA Driver - version %s\n", - DRV_VERSION); + pr_debug("%s t3cdev %p\n", __func__, tdev); + pr_info_once("Chelsio T3 RDMA Driver - version %s\n", DRV_VERSION); rnicp = (struct iwch_dev *)ib_alloc_device(sizeof(*rnicp)); if (!rnicp) { - printk(KERN_ERR MOD "Cannot allocate ib device\n"); + pr_err("Cannot allocate ib device\n"); return; } rnicp->rdev.ulp = rnicp; @@ -160,7 +159,7 @@ static void open_rnic_dev(struct t3cdev *tdev) if (cxio_rdev_open(&rnicp->rdev)) { mutex_unlock(&dev_mutex); - printk(KERN_ERR MOD "Unable to open CXIO rdev\n"); + pr_err("Unable to open CXIO rdev\n"); ib_dealloc_device(&rnicp->ibdev); return; } @@ -171,18 +170,18 @@ static void open_rnic_dev(struct t3cdev *tdev) mutex_unlock(&dev_mutex); if (iwch_register_device(rnicp)) { - printk(KERN_ERR MOD "Unable to register device\n"); + pr_err("Unable to register device\n"); close_rnic_dev(tdev); } - printk(KERN_INFO MOD "Initialized device %s\n", - pci_name(rnicp->rdev.rnic_info.pdev)); + pr_info("Initialized device %s\n", + pci_name(rnicp->rdev.rnic_info.pdev)); return; } static void close_rnic_dev(struct t3cdev *tdev) { struct iwch_dev *dev, *tmp; - PDBG("%s t3cdev %p\n", __func__, tdev); + pr_debug("%s t3cdev %p\n", __func__, tdev); mutex_lock(&dev_mutex); list_for_each_entry_safe(dev, tmp, &dev_list, entry) { if (dev->rdev.t3cdev_p == tdev) { diff --git a/drivers/infiniband/hw/cxgb3/iwch_cm.c b/drivers/infiniband/hw/cxgb3/iwch_cm.c index 65ee64400deb..b61630eba912 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_cm.c +++ b/drivers/infiniband/hw/cxgb3/iwch_cm.c @@ -112,9 +112,9 @@ static void connect_reply_upcall(struct iwch_ep *ep, int status); static void start_ep_timer(struct iwch_ep *ep) { - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); if (timer_pending(&ep->timer)) { - PDBG("%s stopped / restarted timer ep %p\n", __func__, ep); + pr_debug("%s stopped / restarted timer ep %p\n", __func__, ep); del_timer_sync(&ep->timer); } else get_ep(&ep->com); @@ -126,7 +126,7 @@ static void start_ep_timer(struct iwch_ep *ep) static void stop_ep_timer(struct iwch_ep *ep) { - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); if (!timer_pending(&ep->timer)) { WARN(1, "%s timer stopped when its not running! ep %p state %u\n", __func__, ep, ep->com.state); @@ -227,13 +227,13 @@ int iwch_resume_tid(struct iwch_ep *ep) static void set_emss(struct iwch_ep *ep, u16 opt) { - PDBG("%s ep %p opt %u\n", __func__, ep, opt); + pr_debug("%s ep %p opt %u\n", __func__, ep, opt); ep->emss = T3C_DATA(ep->com.tdev)->mtus[G_TCPOPT_MSS(opt)] - 40; if (G_TCPOPT_TSTAMP(opt)) ep->emss -= 12; if (ep->emss < 128) ep->emss = 128; - PDBG("emss=%d\n", ep->emss); + pr_debug("emss=%d\n", ep->emss); } static enum iwch_ep_state state_read(struct iwch_ep_common *epc) @@ -257,7 +257,7 @@ static void state_set(struct iwch_ep_common *epc, enum iwch_ep_state new) unsigned long flags; spin_lock_irqsave(&epc->lock, flags); - PDBG("%s - %s -> %s\n", __func__, states[epc->state], states[new]); + pr_debug("%s - %s -> %s\n", __func__, states[epc->state], states[new]); __state_set(epc, new); spin_unlock_irqrestore(&epc->lock, flags); return; @@ -273,7 +273,7 @@ static void *alloc_ep(int size, gfp_t gfp) spin_lock_init(&epc->lock); init_waitqueue_head(&epc->waitq); } - PDBG("%s alloc ep %p\n", __func__, epc); + pr_debug("%s alloc ep %p\n", __func__, epc); return epc; } @@ -282,7 +282,8 @@ void __free_ep(struct kref *kref) struct iwch_ep *ep; ep = container_of(container_of(kref, struct iwch_ep_common, kref), struct iwch_ep, com); - PDBG("%s ep %p state %s\n", __func__, ep, states[state_read(&ep->com)]); + pr_debug("%s ep %p state %s\n", + __func__, ep, states[state_read(&ep->com)]); if (test_bit(RELEASE_RESOURCES, &ep->com.flags)) { cxgb3_remove_tid(ep->com.tdev, (void *)ep, ep->hwtid); dst_release(ep->dst); @@ -293,7 +294,7 @@ void __free_ep(struct kref *kref) static void release_ep_resources(struct iwch_ep *ep) { - PDBG("%s ep %p tid %d\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %d\n", __func__, ep, ep->hwtid); set_bit(RELEASE_RESOURCES, &ep->com.flags); put_ep(&ep->com); } @@ -358,7 +359,7 @@ static unsigned int find_best_mtu(const struct t3c_data *d, unsigned short mtu) static void arp_failure_discard(struct t3cdev *dev, struct sk_buff *skb) { - PDBG("%s t3cdev %p\n", __func__, dev); + pr_debug("%s t3cdev %p\n", __func__, dev); kfree_skb(skb); } @@ -367,7 +368,7 @@ static void arp_failure_discard(struct t3cdev *dev, struct sk_buff *skb) */ static void act_open_req_arp_failure(struct t3cdev *dev, struct sk_buff *skb) { - printk(KERN_ERR MOD "ARP failure during connect\n"); + pr_err("ARP failure during connect\n"); kfree_skb(skb); } @@ -379,7 +380,7 @@ static void abort_arp_failure(struct t3cdev *dev, struct sk_buff *skb) { struct cpl_abort_req *req = cplhdr(skb); - PDBG("%s t3cdev %p\n", __func__, dev); + pr_debug("%s t3cdev %p\n", __func__, dev); req->cmd = CPL_ABORT_NO_RST; iwch_cxgb3_ofld_send(dev, skb); } @@ -389,10 +390,10 @@ static int send_halfclose(struct iwch_ep *ep, gfp_t gfp) struct cpl_close_con_req *req; struct sk_buff *skb; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); skb = get_skb(NULL, sizeof(*req), gfp); if (!skb) { - printk(KERN_ERR MOD "%s - failed to alloc skb\n", __func__); + pr_err("%s - failed to alloc skb\n", __func__); return -ENOMEM; } skb->priority = CPL_PRIORITY_DATA; @@ -408,11 +409,10 @@ static int send_abort(struct iwch_ep *ep, struct sk_buff *skb, gfp_t gfp) { struct cpl_abort_req *req; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); skb = get_skb(skb, sizeof(*req), gfp); if (!skb) { - printk(KERN_ERR MOD "%s - failed to alloc skb.\n", - __func__); + pr_err("%s - failed to alloc skb\n", __func__); return -ENOMEM; } skb->priority = CPL_PRIORITY_DATA; @@ -434,12 +434,11 @@ static int send_connect(struct iwch_ep *ep) unsigned int mtu_idx; int wscale; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); skb = get_skb(NULL, sizeof(*req), GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "%s - failed to alloc skb.\n", - __func__); + pr_err("%s - failed to alloc skb\n", __func__); return -ENOMEM; } mtu_idx = find_best_mtu(T3C_DATA(ep->com.tdev), dst_mtu(ep->dst)); @@ -478,7 +477,7 @@ static void send_mpa_req(struct iwch_ep *ep, struct sk_buff *skb) struct mpa_message *mpa; int len; - PDBG("%s ep %p pd_len %d\n", __func__, ep, ep->plen); + pr_debug("%s ep %p pd_len %d\n", __func__, ep, ep->plen); BUG_ON(skb_cloned(skb)); @@ -538,13 +537,13 @@ static int send_mpa_reject(struct iwch_ep *ep, const void *pdata, u8 plen) struct mpa_message *mpa; struct sk_buff *skb; - PDBG("%s ep %p plen %d\n", __func__, ep, plen); + pr_debug("%s ep %p plen %d\n", __func__, ep, plen); mpalen = sizeof(*mpa) + plen; skb = get_skb(NULL, mpalen + sizeof(*req), GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "%s - cannot alloc skb!\n", __func__); + pr_err("%s - cannot alloc skb!\n", __func__); return -ENOMEM; } skb_reserve(skb, sizeof(*req)); @@ -587,13 +586,13 @@ static int send_mpa_reply(struct iwch_ep *ep, const void *pdata, u8 plen) int len; struct sk_buff *skb; - PDBG("%s ep %p plen %d\n", __func__, ep, plen); + pr_debug("%s ep %p plen %d\n", __func__, ep, plen); mpalen = sizeof(*mpa) + plen; skb = get_skb(NULL, mpalen + sizeof(*req), GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "%s - cannot alloc skb!\n", __func__); + pr_err("%s - cannot alloc skb!\n", __func__); return -ENOMEM; } skb->priority = CPL_PRIORITY_DATA; @@ -636,7 +635,7 @@ static int act_establish(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) struct cpl_act_establish *req = cplhdr(skb); unsigned int tid = GET_TID(req); - PDBG("%s ep %p tid %d\n", __func__, ep, tid); + pr_debug("%s ep %p tid %d\n", __func__, ep, tid); dst_confirm(ep->dst); @@ -660,7 +659,7 @@ static int act_establish(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) static void abort_connection(struct iwch_ep *ep, struct sk_buff *skb, gfp_t gfp) { - PDBG("%s ep %p\n", __FILE__, ep); + pr_debug("%s ep %p\n", __FILE__, ep); state_set(&ep->com, ABORTING); send_abort(ep, skb, gfp); } @@ -669,12 +668,12 @@ static void close_complete_upcall(struct iwch_ep *ep) { struct iw_cm_event event; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_CLOSE; if (ep->com.cm_id) { - PDBG("close complete delivered ep %p cm_id %p tid %d\n", - ep, ep->com.cm_id, ep->hwtid); + pr_debug("close complete delivered ep %p cm_id %p tid %d\n", + ep, ep->com.cm_id, ep->hwtid); ep->com.cm_id->event_handler(ep->com.cm_id, &event); ep->com.cm_id->rem_ref(ep->com.cm_id); ep->com.cm_id = NULL; @@ -686,12 +685,12 @@ static void peer_close_upcall(struct iwch_ep *ep) { struct iw_cm_event event; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_DISCONNECT; if (ep->com.cm_id) { - PDBG("peer close delivered ep %p cm_id %p tid %d\n", - ep, ep->com.cm_id, ep->hwtid); + pr_debug("peer close delivered ep %p cm_id %p tid %d\n", + ep, ep->com.cm_id, ep->hwtid); ep->com.cm_id->event_handler(ep->com.cm_id, &event); } } @@ -700,13 +699,13 @@ static void peer_abort_upcall(struct iwch_ep *ep) { struct iw_cm_event event; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_CLOSE; event.status = -ECONNRESET; if (ep->com.cm_id) { - PDBG("abort delivered ep %p cm_id %p tid %d\n", ep, - ep->com.cm_id, ep->hwtid); + pr_debug("abort delivered ep %p cm_id %p tid %d\n", ep, + ep->com.cm_id, ep->hwtid); ep->com.cm_id->event_handler(ep->com.cm_id, &event); ep->com.cm_id->rem_ref(ep->com.cm_id); ep->com.cm_id = NULL; @@ -718,7 +717,7 @@ static void connect_reply_upcall(struct iwch_ep *ep, int status) { struct iw_cm_event event; - PDBG("%s ep %p status %d\n", __func__, ep, status); + pr_debug("%s ep %p status %d\n", __func__, ep, status); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_CONNECT_REPLY; event.status = status; @@ -732,8 +731,8 @@ static void connect_reply_upcall(struct iwch_ep *ep, int status) event.private_data = ep->mpa_pkt + sizeof(struct mpa_message); } if (ep->com.cm_id) { - PDBG("%s ep %p tid %d status %d\n", __func__, ep, - ep->hwtid, status); + pr_debug("%s ep %p tid %d status %d\n", __func__, ep, + ep->hwtid, status); ep->com.cm_id->event_handler(ep->com.cm_id, &event); } if (status < 0) { @@ -747,7 +746,7 @@ static void connect_request_upcall(struct iwch_ep *ep) { struct iw_cm_event event; - PDBG("%s ep %p tid %d\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %d\n", __func__, ep, ep->hwtid); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_CONNECT_REQUEST; memcpy(&event.local_addr, &ep->com.local_addr, @@ -776,7 +775,7 @@ static void established_upcall(struct iwch_ep *ep) { struct iw_cm_event event; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_ESTABLISHED; /* @@ -785,7 +784,7 @@ static void established_upcall(struct iwch_ep *ep) */ event.ird = event.ord = 8; if (ep->com.cm_id) { - PDBG("%s ep %p tid %d\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %d\n", __func__, ep, ep->hwtid); ep->com.cm_id->event_handler(ep->com.cm_id, &event); } } @@ -795,10 +794,10 @@ static int update_rx_credits(struct iwch_ep *ep, u32 credits) struct cpl_rx_data_ack *req; struct sk_buff *skb; - PDBG("%s ep %p credits %u\n", __func__, ep, credits); + pr_debug("%s ep %p credits %u\n", __func__, ep, credits); skb = get_skb(NULL, sizeof(*req), GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "update_rx_credits - cannot alloc skb!\n"); + pr_err("update_rx_credits - cannot alloc skb!\n"); return 0; } @@ -819,7 +818,7 @@ static void process_mpa_reply(struct iwch_ep *ep, struct sk_buff *skb) enum iwch_qp_attr_mask mask; int err; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); /* * Stop mpa timer. If it expired, then the state has @@ -906,10 +905,10 @@ static void process_mpa_reply(struct iwch_ep *ep, struct sk_buff *skb) ep->mpa_attr.recv_marker_enabled = markers_enabled; ep->mpa_attr.xmit_marker_enabled = mpa->flags & MPA_MARKERS ? 1 : 0; ep->mpa_attr.version = mpa_rev; - PDBG("%s - crc_enabled=%d, recv_marker_enabled=%d, " - "xmit_marker_enabled=%d, version=%d\n", __func__, - ep->mpa_attr.crc_enabled, ep->mpa_attr.recv_marker_enabled, - ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version); + pr_debug("%s - crc_enabled=%d, recv_marker_enabled=%d, xmit_marker_enabled=%d, version=%d\n", + __func__, + ep->mpa_attr.crc_enabled, ep->mpa_attr.recv_marker_enabled, + ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version); attrs.mpa_attr = ep->mpa_attr; attrs.max_ird = ep->ird; @@ -944,7 +943,7 @@ static void process_mpa_request(struct iwch_ep *ep, struct sk_buff *skb) struct mpa_message *mpa; u16 plen; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); /* * Stop mpa timer. If it expired, then the state has @@ -964,7 +963,7 @@ static void process_mpa_request(struct iwch_ep *ep, struct sk_buff *skb) return; } - PDBG("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__); + pr_debug("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__); /* * Copy the new data into our accumulation buffer. @@ -979,7 +978,7 @@ static void process_mpa_request(struct iwch_ep *ep, struct sk_buff *skb) */ if (ep->mpa_pkt_len < sizeof(*mpa)) return; - PDBG("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__); + pr_debug("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__); mpa = (struct mpa_message *) ep->mpa_pkt; /* @@ -1029,10 +1028,10 @@ static void process_mpa_request(struct iwch_ep *ep, struct sk_buff *skb) ep->mpa_attr.recv_marker_enabled = markers_enabled; ep->mpa_attr.xmit_marker_enabled = mpa->flags & MPA_MARKERS ? 1 : 0; ep->mpa_attr.version = mpa_rev; - PDBG("%s - crc_enabled=%d, recv_marker_enabled=%d, " - "xmit_marker_enabled=%d, version=%d\n", __func__, - ep->mpa_attr.crc_enabled, ep->mpa_attr.recv_marker_enabled, - ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version); + pr_debug("%s - crc_enabled=%d, recv_marker_enabled=%d, xmit_marker_enabled=%d, version=%d\n", + __func__, + ep->mpa_attr.crc_enabled, ep->mpa_attr.recv_marker_enabled, + ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version); state_set(&ep->com, MPA_REQ_RCVD); @@ -1047,7 +1046,7 @@ static int rx_data(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) struct cpl_rx_data *hdr = cplhdr(skb); unsigned int dlen = ntohs(hdr->len); - PDBG("%s ep %p dlen %u\n", __func__, ep, dlen); + pr_debug("%s ep %p dlen %u\n", __func__, ep, dlen); skb_pull(skb, sizeof(*hdr)); skb_trim(skb, dlen); @@ -1065,8 +1064,7 @@ static int rx_data(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) case MPA_REP_SENT: break; default: - printk(KERN_ERR MOD "%s Unexpected streaming data." - " ep %p state %d tid %d\n", + pr_err("%s Unexpected streaming data. ep %p state %d tid %d\n", __func__, ep, state_read(&ep->com), ep->hwtid); /* @@ -1095,11 +1093,11 @@ static int tx_ack(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) unsigned long flags; int post_zb = 0; - PDBG("%s ep %p credits %u\n", __func__, ep, credits); + pr_debug("%s ep %p credits %u\n", __func__, ep, credits); if (credits == 0) { - PDBG("%s 0 credit ack ep %p state %u\n", - __func__, ep, state_read(&ep->com)); + pr_debug("%s 0 credit ack ep %p state %u\n", + __func__, ep, state_read(&ep->com)); return CPL_RET_BUF_DONE; } @@ -1107,24 +1105,24 @@ static int tx_ack(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) BUG_ON(credits != 1); dst_confirm(ep->dst); if (!ep->mpa_skb) { - PDBG("%s rdma_init wr_ack ep %p state %u\n", - __func__, ep, ep->com.state); + pr_debug("%s rdma_init wr_ack ep %p state %u\n", + __func__, ep, ep->com.state); if (ep->mpa_attr.initiator) { - PDBG("%s initiator ep %p state %u\n", - __func__, ep, ep->com.state); + pr_debug("%s initiator ep %p state %u\n", + __func__, ep, ep->com.state); if (peer2peer && ep->com.state == FPDU_MODE) post_zb = 1; } else { - PDBG("%s responder ep %p state %u\n", - __func__, ep, ep->com.state); + pr_debug("%s responder ep %p state %u\n", + __func__, ep, ep->com.state); if (ep->com.state == MPA_REQ_RCVD) { ep->com.rpl_done = 1; wake_up(&ep->com.waitq); } } } else { - PDBG("%s lsm ack ep %p state %u freeing skb\n", - __func__, ep, ep->com.state); + pr_debug("%s lsm ack ep %p state %u freeing skb\n", + __func__, ep, ep->com.state); kfree_skb(ep->mpa_skb); ep->mpa_skb = NULL; } @@ -1140,7 +1138,7 @@ static int abort_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) unsigned long flags; int release = 0; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); BUG_ON(!ep); /* @@ -1159,8 +1157,7 @@ static int abort_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) release = 1; break; default: - printk(KERN_ERR "%s ep %p state %d\n", - __func__, ep, ep->com.state); + pr_err("%s ep %p state %d\n", __func__, ep, ep->com.state); break; } spin_unlock_irqrestore(&ep->com.lock, flags); @@ -1184,8 +1181,8 @@ static int act_open_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) struct iwch_ep *ep = ctx; struct cpl_act_open_rpl *rpl = cplhdr(skb); - PDBG("%s ep %p status %u errno %d\n", __func__, ep, rpl->status, - status2errno(rpl->status)); + pr_debug("%s ep %p status %u errno %d\n", __func__, ep, rpl->status, + status2errno(rpl->status)); connect_reply_upcall(ep, status2errno(rpl->status)); state_set(&ep->com, DEAD); if (ep->com.tdev->type != T3A && act_open_has_tid(rpl->status)) @@ -1202,10 +1199,10 @@ static int listen_start(struct iwch_listen_ep *ep) struct sk_buff *skb; struct cpl_pass_open_req *req; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); skb = get_skb(NULL, sizeof(*req), GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "t3c_listen_start failed to alloc skb!\n"); + pr_err("t3c_listen_start failed to alloc skb!\n"); return -ENOMEM; } @@ -1230,8 +1227,8 @@ static int pass_open_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) struct iwch_listen_ep *ep = ctx; struct cpl_pass_open_rpl *rpl = cplhdr(skb); - PDBG("%s ep %p status %d error %d\n", __func__, ep, - rpl->status, status2errno(rpl->status)); + pr_debug("%s ep %p status %d error %d\n", __func__, ep, + rpl->status, status2errno(rpl->status)); ep->com.rpl_err = status2errno(rpl->status); ep->com.rpl_done = 1; wake_up(&ep->com.waitq); @@ -1244,10 +1241,10 @@ static int listen_stop(struct iwch_listen_ep *ep) struct sk_buff *skb; struct cpl_close_listserv_req *req; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); skb = get_skb(NULL, sizeof(*req), GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "%s - failed to alloc skb\n", __func__); + pr_err("%s - failed to alloc skb\n", __func__); return -ENOMEM; } req = (struct cpl_close_listserv_req *) skb_put(skb, sizeof(*req)); @@ -1264,7 +1261,7 @@ static int close_listsrv_rpl(struct t3cdev *tdev, struct sk_buff *skb, struct iwch_listen_ep *ep = ctx; struct cpl_close_listserv_rpl *rpl = cplhdr(skb); - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); ep->com.rpl_err = status2errno(rpl->status); ep->com.rpl_done = 1; wake_up(&ep->com.waitq); @@ -1278,7 +1275,7 @@ static void accept_cr(struct iwch_ep *ep, __be32 peer_ip, struct sk_buff *skb) u32 opt0h, opt0l, opt2; int wscale; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); BUG_ON(skb_cloned(skb)); skb_trim(skb, sizeof(*rpl)); skb_get(skb); @@ -1312,8 +1309,8 @@ static void accept_cr(struct iwch_ep *ep, __be32 peer_ip, struct sk_buff *skb) static void reject_cr(struct t3cdev *tdev, u32 hwtid, __be32 peer_ip, struct sk_buff *skb) { - PDBG("%s t3cdev %p tid %u peer_ip %x\n", __func__, tdev, hwtid, - peer_ip); + pr_debug("%s t3cdev %p tid %u peer_ip %x\n", __func__, tdev, hwtid, + peer_ip); BUG_ON(skb_cloned(skb)); skb_trim(skb, sizeof(struct cpl_tid_release)); skb_get(skb); @@ -1347,11 +1344,10 @@ static int pass_accept_req(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) struct rtable *rt; struct iff_mac tim; - PDBG("%s parent ep %p tid %u\n", __func__, parent_ep, hwtid); + pr_debug("%s parent ep %p tid %u\n", __func__, parent_ep, hwtid); if (state_read(&parent_ep->com) != LISTEN) { - printk(KERN_ERR "%s - listening ep not in LISTEN\n", - __func__); + pr_err("%s - listening ep not in LISTEN\n", __func__); goto reject; } @@ -1361,8 +1357,7 @@ static int pass_accept_req(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) tim.mac_addr = req->dst_mac; tim.vlan_tag = ntohs(req->vlan_tag); if (tdev->ctl(tdev, GET_IFF_FROM_MAC, &tim) < 0 || !tim.dev) { - printk(KERN_ERR "%s bad dst mac %pM\n", - __func__, req->dst_mac); + pr_err("%s bad dst mac %pM\n", __func__, req->dst_mac); goto reject; } @@ -1373,22 +1368,19 @@ static int pass_accept_req(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) req->local_port, req->peer_port, G_PASS_OPEN_TOS(ntohl(req->tos_tid))); if (!rt) { - printk(KERN_ERR MOD "%s - failed to find dst entry!\n", - __func__); + pr_err("%s - failed to find dst entry!\n", __func__); goto reject; } dst = &rt->dst; l2t = t3_l2t_get(tdev, dst, NULL, &req->peer_ip); if (!l2t) { - printk(KERN_ERR MOD "%s - failed to allocate l2t entry!\n", - __func__); + pr_err("%s - failed to allocate l2t entry!\n", __func__); dst_release(dst); goto reject; } child_ep = alloc_ep(sizeof(*child_ep), GFP_KERNEL); if (!child_ep) { - printk(KERN_ERR MOD "%s - failed to allocate ep entry!\n", - __func__); + pr_err("%s - failed to allocate ep entry!\n", __func__); l2t_release(tdev, l2t); dst_release(dst); goto reject; @@ -1423,7 +1415,7 @@ static int pass_establish(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) struct iwch_ep *ep = ctx; struct cpl_pass_establish *req = cplhdr(skb); - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); ep->snd_seq = ntohl(req->snd_isn); ep->rcv_seq = ntohl(req->rcv_isn); @@ -1444,7 +1436,7 @@ static int peer_close(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) int disconnect = 1; int release = 0; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); dst_confirm(ep->dst); spin_lock_irqsave(&ep->com.lock, flags); @@ -1467,14 +1459,14 @@ static int peer_close(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) __state_set(&ep->com, CLOSING); ep->com.rpl_done = 1; ep->com.rpl_err = -ECONNRESET; - PDBG("waking up ep %p\n", ep); + pr_debug("waking up ep %p\n", ep); wake_up(&ep->com.waitq); break; case MPA_REP_SENT: __state_set(&ep->com, CLOSING); ep->com.rpl_done = 1; ep->com.rpl_err = -ECONNRESET; - PDBG("waking up ep %p\n", ep); + pr_debug("waking up ep %p\n", ep); wake_up(&ep->com.waitq); break; case FPDU_MODE: @@ -1539,8 +1531,8 @@ static int peer_abort(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) unsigned long flags; if (is_neg_adv_abort(req->status)) { - PDBG("%s neg_adv_abort ep %p tid %d\n", __func__, ep, - ep->hwtid); + pr_debug("%s neg_adv_abort ep %p tid %d\n", __func__, ep, + ep->hwtid); t3_l2t_send_event(ep->com.tdev, ep->l2t); return CPL_RET_BUF_DONE; } @@ -1554,7 +1546,7 @@ static int peer_abort(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) } spin_lock_irqsave(&ep->com.lock, flags); - PDBG("%s ep %p state %u\n", __func__, ep, ep->com.state); + pr_debug("%s ep %p state %u\n", __func__, ep, ep->com.state); switch (ep->com.state) { case CONNECTING: break; @@ -1568,7 +1560,7 @@ static int peer_abort(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) case MPA_REP_SENT: ep->com.rpl_done = 1; ep->com.rpl_err = -ECONNRESET; - PDBG("waking up ep %p\n", ep); + pr_debug("waking up ep %p\n", ep); wake_up(&ep->com.waitq); break; case MPA_REQ_RCVD: @@ -1581,7 +1573,7 @@ static int peer_abort(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) */ ep->com.rpl_done = 1; ep->com.rpl_err = -ECONNRESET; - PDBG("waking up ep %p\n", ep); + pr_debug("waking up ep %p\n", ep); wake_up(&ep->com.waitq); break; case MORIBUND: @@ -1595,16 +1587,14 @@ static int peer_abort(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) ep->com.qp, IWCH_QP_ATTR_NEXT_STATE, &attrs, 1); if (ret) - printk(KERN_ERR MOD - "%s - qp <- error failed!\n", - __func__); + pr_err("%s - qp <- error failed!\n", __func__); } peer_abort_upcall(ep); break; case ABORTING: break; case DEAD: - PDBG("%s PEER_ABORT IN DEAD STATE!!!!\n", __func__); + pr_debug("%s PEER_ABORT IN DEAD STATE!!!!\n", __func__); spin_unlock_irqrestore(&ep->com.lock, flags); return CPL_RET_BUF_DONE; default: @@ -1620,8 +1610,7 @@ static int peer_abort(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) rpl_skb = get_skb(skb, sizeof(*rpl), GFP_KERNEL); if (!rpl_skb) { - printk(KERN_ERR MOD "%s - cannot allocate skb!\n", - __func__); + pr_err("%s - cannot allocate skb!\n", __func__); release = 1; goto out; } @@ -1645,7 +1634,7 @@ static int close_con_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) unsigned long flags; int release = 0; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); BUG_ON(!ep); /* The cm_id may be null if we failed to connect */ @@ -1699,9 +1688,9 @@ static int terminate(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) if (state_read(&ep->com) != FPDU_MODE) return CPL_RET_BUF_DONE; - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); skb_pull(skb, sizeof(struct cpl_rdma_terminate)); - PDBG("%s saving %d bytes of term msg\n", __func__, skb->len); + pr_debug("%s saving %d bytes of term msg\n", __func__, skb->len); skb_copy_from_linear_data(skb, ep->com.qp->attr.terminate_buffer, skb->len); ep->com.qp->attr.terminate_msg_len = skb->len; @@ -1714,12 +1703,12 @@ static int ec_status(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) struct cpl_rdma_ec_status *rep = cplhdr(skb); struct iwch_ep *ep = ctx; - PDBG("%s ep %p tid %u status %d\n", __func__, ep, ep->hwtid, - rep->status); + pr_debug("%s ep %p tid %u status %d\n", __func__, ep, ep->hwtid, + rep->status); if (rep->status) { struct iwch_qp_attributes attrs; - printk(KERN_ERR MOD "%s BAD CLOSE - Aborting tid %u\n", + pr_err("%s BAD CLOSE - Aborting tid %u\n", __func__, ep->hwtid); stop_ep_timer(ep); attrs.next_state = IWCH_QP_STATE_ERROR; @@ -1739,8 +1728,8 @@ static void ep_timeout(unsigned long arg) int abort = 1; spin_lock_irqsave(&ep->com.lock, flags); - PDBG("%s ep %p tid %u state %d\n", __func__, ep, ep->hwtid, - ep->com.state); + pr_debug("%s ep %p tid %u state %d\n", __func__, ep, ep->hwtid, + ep->com.state); switch (ep->com.state) { case MPA_REQ_SENT: __state_set(&ep->com, ABORTING); @@ -1774,7 +1763,7 @@ int iwch_reject_cr(struct iw_cm_id *cm_id, const void *pdata, u8 pdata_len) { int err; struct iwch_ep *ep = to_ep(cm_id); - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); if (state_read(&ep->com) == DEAD) { put_ep(&ep->com); @@ -1800,7 +1789,7 @@ int iwch_accept_cr(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) struct iwch_dev *h = to_iwch_dev(cm_id->device); struct iwch_qp *qp = get_qhp(h, conn_param->qpn); - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); if (state_read(&ep->com) == DEAD) { err = -ECONNRESET; goto err; @@ -1826,7 +1815,7 @@ int iwch_accept_cr(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) if (peer2peer && ep->ird == 0) ep->ird = 1; - PDBG("%s %d ird %d ord %d\n", __func__, __LINE__, ep->ird, ep->ord); + pr_debug("%s %d ird %d ord %d\n", __func__, __LINE__, ep->ird, ep->ord); /* bind QP to EP and move to RTS */ attrs.mpa_attr = ep->mpa_attr; @@ -1907,7 +1896,7 @@ int iwch_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) ep = alloc_ep(sizeof(*ep), GFP_KERNEL); if (!ep) { - printk(KERN_ERR MOD "%s - cannot alloc ep.\n", __func__); + pr_err("%s - cannot alloc ep\n", __func__); err = -ENOMEM; goto out; } @@ -1928,15 +1917,15 @@ int iwch_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) ep->com.cm_id = cm_id; ep->com.qp = get_qhp(h, conn_param->qpn); BUG_ON(!ep->com.qp); - PDBG("%s qpn 0x%x qp %p cm_id %p\n", __func__, conn_param->qpn, - ep->com.qp, cm_id); + pr_debug("%s qpn 0x%x qp %p cm_id %p\n", __func__, conn_param->qpn, + ep->com.qp, cm_id); /* * Allocate an active TID to initiate a TCP connection. */ ep->atid = cxgb3_alloc_atid(h->rdev.t3cdev_p, &t3c_client, ep); if (ep->atid == -1) { - printk(KERN_ERR MOD "%s - cannot alloc atid.\n", __func__); + pr_err("%s - cannot alloc atid\n", __func__); err = -ENOMEM; goto fail2; } @@ -1946,7 +1935,7 @@ int iwch_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) raddr->sin_addr.s_addr, laddr->sin_port, raddr->sin_port, IPTOS_LOWDELAY); if (!rt) { - printk(KERN_ERR MOD "%s - cannot find route.\n", __func__); + pr_err("%s - cannot find route\n", __func__); err = -EHOSTUNREACH; goto fail3; } @@ -1954,7 +1943,7 @@ int iwch_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) ep->l2t = t3_l2t_get(ep->com.tdev, ep->dst, NULL, &raddr->sin_addr.s_addr); if (!ep->l2t) { - printk(KERN_ERR MOD "%s - cannot alloc l2e.\n", __func__); + pr_err("%s - cannot alloc l2e\n", __func__); err = -ENOMEM; goto fail4; } @@ -1999,11 +1988,11 @@ int iwch_create_listen(struct iw_cm_id *cm_id, int backlog) ep = alloc_ep(sizeof(*ep), GFP_KERNEL); if (!ep) { - printk(KERN_ERR MOD "%s - cannot alloc ep.\n", __func__); + pr_err("%s - cannot alloc ep\n", __func__); err = -ENOMEM; goto fail1; } - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); ep->com.tdev = h->rdev.t3cdev_p; cm_id->add_ref(cm_id); ep->com.cm_id = cm_id; @@ -2016,7 +2005,7 @@ int iwch_create_listen(struct iw_cm_id *cm_id, int backlog) */ ep->stid = cxgb3_alloc_stid(h->rdev.t3cdev_p, &t3c_client, ep); if (ep->stid == -1) { - printk(KERN_ERR MOD "%s - cannot alloc atid.\n", __func__); + pr_err("%s - cannot alloc atid\n", __func__); err = -ENOMEM; goto fail2; } @@ -2048,7 +2037,7 @@ int iwch_destroy_listen(struct iw_cm_id *cm_id) int err; struct iwch_listen_ep *ep = to_listen_ep(cm_id); - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); might_sleep(); state_set(&ep->com, DEAD); @@ -2077,8 +2066,8 @@ int iwch_ep_disconnect(struct iwch_ep *ep, int abrupt, gfp_t gfp) spin_lock_irqsave(&ep->com.lock, flags); - PDBG("%s ep %p state %s, abrupt %d\n", __func__, ep, - states[ep->com.state], abrupt); + pr_debug("%s ep %p state %s, abrupt %d\n", __func__, ep, + states[ep->com.state], abrupt); tdev = (struct t3cdev *)ep->com.tdev; rdev = (struct cxio_rdev *)tdev->ulp; @@ -2115,8 +2104,8 @@ int iwch_ep_disconnect(struct iwch_ep *ep, int abrupt, gfp_t gfp) case MORIBUND: case ABORTING: case DEAD: - PDBG("%s ignoring disconnect ep %p state %u\n", - __func__, ep, ep->com.state); + pr_debug("%s ignoring disconnect ep %p state %u\n", + __func__, ep, ep->com.state); break; default: BUG(); @@ -2145,8 +2134,8 @@ int iwch_ep_redirect(void *ctx, struct dst_entry *old, struct dst_entry *new, if (ep->dst != old) return 0; - PDBG("%s ep %p redirect to dst %p l2t %p\n", __func__, ep, new, - l2t); + pr_debug("%s ep %p redirect to dst %p l2t %p\n", __func__, ep, new, + l2t); dst_hold(new); l2t_release(ep->com.tdev, ep->l2t); ep->l2t = l2t; @@ -2225,8 +2214,8 @@ static int set_tcb_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) struct cpl_set_tcb_rpl *rpl = cplhdr(skb); if (rpl->status != CPL_ERR_NONE) { - printk(KERN_ERR MOD "Unexpected SET_TCB_RPL status %u " - "for tid %u\n", rpl->status, GET_TID(rpl)); + pr_err("Unexpected SET_TCB_RPL status %u for tid %u\n", + rpl->status, GET_TID(rpl)); } return CPL_RET_BUF_DONE; } diff --git a/drivers/infiniband/hw/cxgb3/iwch_cm.h b/drivers/infiniband/hw/cxgb3/iwch_cm.h index e66e75921797..cc7fe644d260 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_cm.h +++ b/drivers/infiniband/hw/cxgb3/iwch_cm.h @@ -53,17 +53,17 @@ #define MPA_MARKERS 0x80 #define MPA_FLAGS_MASK 0xE0 -#define put_ep(ep) { \ - PDBG("put_ep (via %s:%u) ep %p refcnt %d\n", __func__, __LINE__, \ - ep, kref_read(&((ep)->kref))); \ - WARN_ON(kref_read(&((ep)->kref)) < 1); \ - kref_put(&((ep)->kref), __free_ep); \ +#define put_ep(ep) { \ + pr_debug("put_ep (via %s:%u) ep %p refcnt %d\n", \ + __func__, __LINE__, ep, kref_read(&((ep)->kref))); \ + WARN_ON(kref_read(&((ep)->kref)) < 1); \ + kref_put(&((ep)->kref), __free_ep); \ } -#define get_ep(ep) { \ - PDBG("get_ep (via %s:%u) ep %p, refcnt %d\n", __func__, __LINE__, \ - ep, kref_read(&((ep)->kref))); \ - kref_get(&((ep)->kref)); \ +#define get_ep(ep) { \ + pr_debug("get_ep (via %s:%u) ep %p, refcnt %d\n", \ + __func__, __LINE__, ep, kref_read(&((ep)->kref))); \ + kref_get(&((ep)->kref)); \ } struct mpa_message { diff --git a/drivers/infiniband/hw/cxgb3/iwch_cq.c b/drivers/infiniband/hw/cxgb3/iwch_cq.c index 97fbfd2c298e..dd5348e48806 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_cq.c +++ b/drivers/infiniband/hw/cxgb3/iwch_cq.c @@ -67,8 +67,8 @@ static int iwch_poll_cq_one(struct iwch_dev *rhp, struct iwch_cq *chp, ret = cxio_poll_cq(wq, &(chp->cq), &cqe, &cqe_flushed, &cookie, &credit); if (t3a_device(chp->rhp) && credit) { - PDBG("%s updating %d cq credits on id %d\n", __func__, - credit, chp->cq.cqid); + pr_debug("%s updating %d cq credits on id %d\n", __func__, + credit, chp->cq.cqid); cxio_hal_cq_op(&rhp->rdev, &chp->cq, CQ_CREDIT_UPDATE, credit); } @@ -83,11 +83,11 @@ static int iwch_poll_cq_one(struct iwch_dev *rhp, struct iwch_cq *chp, wc->vendor_err = CQE_STATUS(cqe); wc->wc_flags = 0; - PDBG("%s qpid 0x%x type %d opcode %d status 0x%x wrid hi 0x%x " - "lo 0x%x cookie 0x%llx\n", __func__, - CQE_QPID(cqe), CQE_TYPE(cqe), - CQE_OPCODE(cqe), CQE_STATUS(cqe), CQE_WRID_HI(cqe), - CQE_WRID_LOW(cqe), (unsigned long long) cookie); + pr_debug("%s qpid 0x%x type %d opcode %d status 0x%x wrid hi 0x%x lo 0x%x cookie 0x%llx\n", + __func__, + CQE_QPID(cqe), CQE_TYPE(cqe), + CQE_OPCODE(cqe), CQE_STATUS(cqe), CQE_WRID_HI(cqe), + CQE_WRID_LOW(cqe), (unsigned long long)cookie); if (CQE_TYPE(cqe) == 0) { if (!CQE_STATUS(cqe)) @@ -122,8 +122,7 @@ static int iwch_poll_cq_one(struct iwch_dev *rhp, struct iwch_cq *chp, wc->opcode = IB_WC_REG_MR; break; default: - printk(KERN_ERR MOD "Unexpected opcode %d " - "in the CQE received for QPID=0x%0x\n", + pr_err("Unexpected opcode %d in the CQE received for QPID=0x%0x\n", CQE_OPCODE(cqe), CQE_QPID(cqe)); ret = -EINVAL; goto out; @@ -177,8 +176,8 @@ static int iwch_poll_cq_one(struct iwch_dev *rhp, struct iwch_cq *chp, wc->status = IB_WC_WR_FLUSH_ERR; break; default: - printk(KERN_ERR MOD "Unexpected cqe_status 0x%x for " - "QPID=0x%0x\n", CQE_STATUS(cqe), CQE_QPID(cqe)); + pr_err("Unexpected cqe_status 0x%x for QPID=0x%0x\n", + CQE_STATUS(cqe), CQE_QPID(cqe)); ret = -EINVAL; } } diff --git a/drivers/infiniband/hw/cxgb3/iwch_ev.c b/drivers/infiniband/hw/cxgb3/iwch_ev.c index abcc9e76962b..4a0c82a8fb60 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_ev.c +++ b/drivers/infiniband/hw/cxgb3/iwch_ev.c @@ -52,7 +52,7 @@ static void post_qp_event(struct iwch_dev *rnicp, struct iwch_cq *chp, qhp = get_qhp(rnicp, CQE_QPID(rsp_msg->cqe)); if (!qhp) { - printk(KERN_ERR "%s unaffiliated error 0x%x qpid 0x%x\n", + pr_err("%s unaffiliated error 0x%x qpid 0x%x\n", __func__, CQE_STATUS(rsp_msg->cqe), CQE_QPID(rsp_msg->cqe)); spin_unlock(&rnicp->lock); @@ -61,15 +61,16 @@ static void post_qp_event(struct iwch_dev *rnicp, struct iwch_cq *chp, if ((qhp->attr.state == IWCH_QP_STATE_ERROR) || (qhp->attr.state == IWCH_QP_STATE_TERMINATE)) { - PDBG("%s AE received after RTS - " - "qp state %d qpid 0x%x status 0x%x\n", __func__, - qhp->attr.state, qhp->wq.qpid, CQE_STATUS(rsp_msg->cqe)); + pr_debug("%s AE received after RTS - qp state %d qpid 0x%x status 0x%x\n", + __func__, + qhp->attr.state, qhp->wq.qpid, + CQE_STATUS(rsp_msg->cqe)); spin_unlock(&rnicp->lock); return; } - printk(KERN_ERR "%s - AE qpid 0x%x opcode %d status 0x%x " - "type %d wrid.hi 0x%x wrid.lo 0x%x \n", __func__, + pr_err("%s - AE qpid 0x%x opcode %d status 0x%x type %d wrid.hi 0x%x wrid.lo 0x%x\n", + __func__, CQE_QPID(rsp_msg->cqe), CQE_OPCODE(rsp_msg->cqe), CQE_STATUS(rsp_msg->cqe), CQE_TYPE(rsp_msg->cqe), CQE_WRID_HI(rsp_msg->cqe), CQE_WRID_LOW(rsp_msg->cqe)); @@ -117,8 +118,7 @@ void iwch_ev_dispatch(struct cxio_rdev *rdev_p, struct sk_buff *skb) chp = get_chp(rnicp, cqid); qhp = get_qhp(rnicp, CQE_QPID(rsp_msg->cqe)); if (!chp || !qhp) { - printk(KERN_ERR MOD "BAD AE cqid 0x%x qpid 0x%x opcode %d " - "status 0x%x type %d wrid.hi 0x%x wrid.lo 0x%x \n", + pr_err("BAD AE cqid 0x%x qpid 0x%x opcode %d status 0x%x type %d wrid.hi 0x%x wrid.lo 0x%x\n", cqid, CQE_QPID(rsp_msg->cqe), CQE_OPCODE(rsp_msg->cqe), CQE_STATUS(rsp_msg->cqe), CQE_TYPE(rsp_msg->cqe), CQE_WRID_HI(rsp_msg->cqe), @@ -137,12 +137,12 @@ void iwch_ev_dispatch(struct cxio_rdev *rdev_p, struct sk_buff *skb) if ((CQE_OPCODE(rsp_msg->cqe) == T3_TERMINATE) && (CQE_STATUS(rsp_msg->cqe) == 0)) { if (SQ_TYPE(rsp_msg->cqe)) { - PDBG("%s QPID 0x%x ep %p disconnecting\n", - __func__, qhp->wq.qpid, qhp->ep); + pr_debug("%s QPID 0x%x ep %p disconnecting\n", + __func__, qhp->wq.qpid, qhp->ep); iwch_ep_disconnect(qhp->ep, 0, GFP_ATOMIC); } else { - PDBG("%s post REQ_ERR AE QPID 0x%x\n", __func__, - qhp->wq.qpid); + pr_debug("%s post REQ_ERR AE QPID 0x%x\n", __func__, + qhp->wq.qpid); post_qp_event(rnicp, chp, rsp_msg, IB_EVENT_QP_REQ_ERR, 0); iwch_ep_disconnect(qhp->ep, 0, GFP_ATOMIC); @@ -218,7 +218,7 @@ void iwch_ev_dispatch(struct cxio_rdev *rdev_p, struct sk_buff *skb) break; default: - printk(KERN_ERR MOD "Unknown T3 status 0x%x QPID 0x%x\n", + pr_err("Unknown T3 status 0x%x QPID 0x%x\n", CQE_STATUS(rsp_msg->cqe), qhp->wq.qpid); post_qp_event(rnicp, chp, rsp_msg, IB_EVENT_QP_FATAL, 1); break; diff --git a/drivers/infiniband/hw/cxgb3/iwch_mem.c b/drivers/infiniband/hw/cxgb3/iwch_mem.c index 1d04c872c9d5..12886b1b4b10 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_mem.c +++ b/drivers/infiniband/hw/cxgb3/iwch_mem.c @@ -48,7 +48,7 @@ static int iwch_finish_mem_reg(struct iwch_mr *mhp, u32 stag) mhp->attr.stag = stag; mmid = stag >> 8; mhp->ibmr.rkey = mhp->ibmr.lkey = stag; - PDBG("%s mmid 0x%x mhp %p\n", __func__, mmid, mhp); + pr_debug("%s mmid 0x%x mhp %p\n", __func__, mmid, mhp); return insert_handle(mhp->rhp, &mhp->rhp->mmidr, mhp, mmid); } diff --git a/drivers/infiniband/hw/cxgb3/iwch_provider.c b/drivers/infiniband/hw/cxgb3/iwch_provider.c index 86ecd3ea6a4b..790d7c79fe3e 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_provider.c +++ b/drivers/infiniband/hw/cxgb3/iwch_provider.c @@ -103,7 +103,7 @@ static int iwch_dealloc_ucontext(struct ib_ucontext *context) struct iwch_ucontext *ucontext = to_iwch_ucontext(context); struct iwch_mm_entry *mm, *tmp; - PDBG("%s context %p\n", __func__, context); + pr_debug("%s context %p\n", __func__, context); list_for_each_entry_safe(mm, tmp, &ucontext->mmaps, entry) kfree(mm); cxio_release_ucontext(&rhp->rdev, &ucontext->uctx); @@ -117,7 +117,7 @@ static struct ib_ucontext *iwch_alloc_ucontext(struct ib_device *ibdev, struct iwch_ucontext *context; struct iwch_dev *rhp = to_iwch_dev(ibdev); - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); context = kzalloc(sizeof(*context), GFP_KERNEL); if (!context) return ERR_PTR(-ENOMEM); @@ -131,7 +131,7 @@ static int iwch_destroy_cq(struct ib_cq *ib_cq) { struct iwch_cq *chp; - PDBG("%s ib_cq %p\n", __func__, ib_cq); + pr_debug("%s ib_cq %p\n", __func__, ib_cq); chp = to_iwch_cq(ib_cq); remove_handle(chp->rhp, &chp->rhp->cqidr, chp->cq.cqid); @@ -157,7 +157,7 @@ static struct ib_cq *iwch_create_cq(struct ib_device *ibdev, static int warned; size_t resplen; - PDBG("%s ib_dev %p entries %d\n", __func__, ibdev, entries); + pr_debug("%s ib_dev %p entries %d\n", __func__, ibdev, entries); if (attr->flags) return ERR_PTR(-EINVAL); @@ -227,8 +227,7 @@ static struct ib_cq *iwch_create_cq(struct ib_device *ibdev, mm->addr = virt_to_phys(chp->cq.queue); if (udata->outlen < sizeof uresp) { if (!warned++) - printk(KERN_WARNING MOD "Warning - " - "downlevel libcxgb3 (non-fatal).\n"); + pr_warn("Warning - downlevel libcxgb3 (non-fatal)\n"); mm->len = PAGE_ALIGN((1UL << uresp.size_log2) * sizeof(struct t3_cqe)); resplen = sizeof(struct iwch_create_cq_resp_v0); @@ -246,9 +245,9 @@ static struct ib_cq *iwch_create_cq(struct ib_device *ibdev, } insert_mmap(ucontext, mm); } - PDBG("created cqid 0x%0x chp %p size 0x%0x, dma_addr 0x%0llx\n", - chp->cq.cqid, chp, (1 << chp->cq.size_log2), - (unsigned long long) chp->cq.dma_addr); + pr_debug("created cqid 0x%0x chp %p size 0x%0x, dma_addr 0x%0llx\n", + chp->cq.cqid, chp, (1 << chp->cq.size_log2), + (unsigned long long)chp->cq.dma_addr); return &chp->ibcq; } @@ -259,7 +258,7 @@ static int iwch_resize_cq(struct ib_cq *cq, int cqe, struct ib_udata *udata) struct t3_cq oldcq, newcq; int ret; - PDBG("%s ib_cq %p cqe %d\n", __func__, cq, cqe); + pr_debug("%s ib_cq %p cqe %d\n", __func__, cq, cqe); /* We don't downsize... */ if (cqe <= cq->cqe) @@ -306,8 +305,7 @@ static int iwch_resize_cq(struct ib_cq *cq, int cqe, struct ib_udata *udata) oldcq.cqid = newcq.cqid; ret = cxio_destroy_cq(&chp->rhp->rdev, &oldcq); if (ret) { - printk(KERN_ERR MOD "%s - cxio_destroy_cq failed %d\n", - __func__, ret); + pr_err("%s - cxio_destroy_cq failed %d\n", __func__, ret); } /* add user hooks here */ @@ -342,12 +340,11 @@ static int iwch_arm_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags flags) chp->cq.rptr = rptr; } else spin_lock_irqsave(&chp->lock, flag); - PDBG("%s rptr 0x%x\n", __func__, chp->cq.rptr); + pr_debug("%s rptr 0x%x\n", __func__, chp->cq.rptr); err = cxio_hal_cq_op(&rhp->rdev, &chp->cq, cq_op, 0); spin_unlock_irqrestore(&chp->lock, flag); if (err < 0) - printk(KERN_ERR MOD "Error %d rearming CQID 0x%x\n", err, - chp->cq.cqid); + pr_err("Error %d rearming CQID 0x%x\n", err, chp->cq.cqid); if (err > 0 && !(flags & IB_CQ_REPORT_MISSED_EVENTS)) err = 0; return err; @@ -363,8 +360,8 @@ static int iwch_mmap(struct ib_ucontext *context, struct vm_area_struct *vma) struct iwch_ucontext *ucontext; u64 addr; - PDBG("%s pgoff 0x%lx key 0x%x len %d\n", __func__, vma->vm_pgoff, - key, len); + pr_debug("%s pgoff 0x%lx key 0x%x len %d\n", __func__, vma->vm_pgoff, + key, len); if (vma->vm_start & (PAGE_SIZE-1)) { return -EINVAL; @@ -416,7 +413,7 @@ static int iwch_deallocate_pd(struct ib_pd *pd) php = to_iwch_pd(pd); rhp = php->rhp; - PDBG("%s ibpd %p pdid 0x%x\n", __func__, pd, php->pdid); + pr_debug("%s ibpd %p pdid 0x%x\n", __func__, pd, php->pdid); cxio_hal_put_pdid(rhp->rdev.rscp, php->pdid); kfree(php); return 0; @@ -430,7 +427,7 @@ static struct ib_pd *iwch_allocate_pd(struct ib_device *ibdev, u32 pdid; struct iwch_dev *rhp; - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); rhp = (struct iwch_dev *) ibdev; pdid = cxio_hal_get_pdid(rhp->rdev.rscp); if (!pdid) @@ -448,7 +445,7 @@ static struct ib_pd *iwch_allocate_pd(struct ib_device *ibdev, return ERR_PTR(-EFAULT); } } - PDBG("%s pdid 0x%0x ptr 0x%p\n", __func__, pdid, php); + pr_debug("%s pdid 0x%0x ptr 0x%p\n", __func__, pdid, php); return &php->ibpd; } @@ -458,7 +455,7 @@ static int iwch_dereg_mr(struct ib_mr *ib_mr) struct iwch_mr *mhp; u32 mmid; - PDBG("%s ib_mr %p\n", __func__, ib_mr); + pr_debug("%s ib_mr %p\n", __func__, ib_mr); mhp = to_iwch_mr(ib_mr); kfree(mhp->pages); @@ -472,7 +469,7 @@ static int iwch_dereg_mr(struct ib_mr *ib_mr) kfree((void *) (unsigned long) mhp->kva); if (mhp->umem) ib_umem_release(mhp->umem); - PDBG("%s mmid 0x%x ptr %p\n", __func__, mmid, mhp); + pr_debug("%s mmid 0x%x ptr %p\n", __func__, mmid, mhp); kfree(mhp); return 0; } @@ -487,13 +484,13 @@ static struct ib_mr *iwch_get_dma_mr(struct ib_pd *pd, int acc) __be64 *page_list; int shift = 26, npages, ret, i; - PDBG("%s ib_pd %p\n", __func__, pd); + pr_debug("%s ib_pd %p\n", __func__, pd); /* * T3 only supports 32 bits of size. */ if (sizeof(phys_addr_t) > 4) { - pr_warn_once(MOD "Cannot support dma_mrs on this platform.\n"); + pr_warn_once("Cannot support dma_mrs on this platform\n"); return ERR_PTR(-ENOTSUPP); } @@ -518,8 +515,8 @@ static struct ib_mr *iwch_get_dma_mr(struct ib_pd *pd, int acc) for (i = 0; i < npages; i++) page_list[i] = cpu_to_be64((u64)i << shift); - PDBG("%s mask 0x%llx shift %d len %lld pbl_size %d\n", - __func__, mask, shift, total_size, npages); + pr_debug("%s mask 0x%llx shift %d len %lld pbl_size %d\n", + __func__, mask, shift, total_size, npages); ret = iwch_alloc_pbl(mhp, npages); if (ret) { @@ -567,7 +564,7 @@ static struct ib_mr *iwch_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, struct iwch_mr *mhp; struct iwch_reg_user_mr_resp uresp; struct scatterlist *sg; - PDBG("%s ib_pd %p\n", __func__, pd); + pr_debug("%s ib_pd %p\n", __func__, pd); php = to_iwch_pd(pd); rhp = php->rhp; @@ -584,7 +581,7 @@ static struct ib_mr *iwch_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, return ERR_PTR(err); } - shift = ffs(mhp->umem->page_size) - 1; + shift = mhp->umem->page_shift; n = mhp->umem->nmap; @@ -604,7 +601,7 @@ static struct ib_mr *iwch_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, len = sg_dma_len(sg) >> shift; for (k = 0; k < len; ++k) { pages[i++] = cpu_to_be64(sg_dma_address(sg) + - mhp->umem->page_size * k); + (k << shift)); if (i == PAGE_SIZE / sizeof *pages) { err = iwch_write_pbl(mhp, pages, i, n); if (err) @@ -637,8 +634,8 @@ pbl_done: if (udata && !t3a_device(rhp)) { uresp.pbl_addr = (mhp->attr.pbl_addr - rhp->rdev.rnic_info.pbl_base) >> 3; - PDBG("%s user resp pbl_addr 0x%x\n", __func__, - uresp.pbl_addr); + pr_debug("%s user resp pbl_addr 0x%x\n", __func__, + uresp.pbl_addr); if (ib_copy_to_udata(udata, &uresp, sizeof (uresp))) { iwch_dereg_mr(&mhp->ibmr); @@ -692,7 +689,7 @@ static struct ib_mw *iwch_alloc_mw(struct ib_pd *pd, enum ib_mw_type type, kfree(mhp); return ERR_PTR(-ENOMEM); } - PDBG("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); + pr_debug("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); return &(mhp->ibmw); } @@ -707,7 +704,7 @@ static int iwch_dealloc_mw(struct ib_mw *mw) mmid = (mw->rkey) >> 8; cxio_deallocate_window(&rhp->rdev, mhp->attr.stag); remove_handle(rhp, &rhp->mmidr, mmid); - PDBG("%s ib_mw %p mmid 0x%x ptr %p\n", __func__, mw, mmid, mhp); + pr_debug("%s ib_mw %p mmid 0x%x ptr %p\n", __func__, mw, mmid, mhp); kfree(mhp); return 0; } @@ -757,7 +754,7 @@ static struct ib_mr *iwch_alloc_mr(struct ib_pd *pd, if (insert_handle(rhp, &rhp->mmidr, mhp, mmid)) goto err3; - PDBG("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); + pr_debug("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); return &(mhp->ibmr); err3: cxio_dereg_mem(&rhp->rdev, stag, mhp->attr.pbl_size, @@ -818,8 +815,8 @@ static int iwch_destroy_qp(struct ib_qp *ib_qp) cxio_destroy_qp(&rhp->rdev, &qhp->wq, ucontext ? &ucontext->uctx : &rhp->rdev.uctx); - PDBG("%s ib_qp %p qpid 0x%0x qhp %p\n", __func__, - ib_qp, qhp->wq.qpid, qhp); + pr_debug("%s ib_qp %p qpid 0x%0x qhp %p\n", __func__, + ib_qp, qhp->wq.qpid, qhp); kfree(qhp); return 0; } @@ -837,7 +834,7 @@ static struct ib_qp *iwch_create_qp(struct ib_pd *pd, int wqsize, sqsize, rqsize; struct iwch_ucontext *ucontext; - PDBG("%s ib_pd %p\n", __func__, pd); + pr_debug("%s ib_pd %p\n", __func__, pd); if (attrs->qp_type != IB_QPT_RC) return ERR_PTR(-EINVAL); php = to_iwch_pd(pd); @@ -878,8 +875,8 @@ static struct ib_qp *iwch_create_qp(struct ib_pd *pd, if (!ucontext && wqsize < (rqsize + (2 * sqsize))) wqsize = roundup_pow_of_two(rqsize + roundup_pow_of_two(attrs->cap.max_send_wr * 2)); - PDBG("%s wqsize %d sqsize %d rqsize %d\n", __func__, - wqsize, sqsize, rqsize); + pr_debug("%s wqsize %d sqsize %d rqsize %d\n", __func__, + wqsize, sqsize, rqsize); qhp = kzalloc(sizeof(*qhp), GFP_KERNEL); if (!qhp) return ERR_PTR(-ENOMEM); @@ -974,11 +971,10 @@ static struct ib_qp *iwch_create_qp(struct ib_pd *pd, } qhp->ibqp.qp_num = qhp->wq.qpid; init_timer(&(qhp->timer)); - PDBG("%s sq_num_entries %d, rq_num_entries %d " - "qpid 0x%0x qhp %p dma_addr 0x%llx size %d rq_addr 0x%x\n", - __func__, qhp->attr.sq_num_entries, qhp->attr.rq_num_entries, - qhp->wq.qpid, qhp, (unsigned long long) qhp->wq.dma_addr, - 1 << qhp->wq.size_log2, qhp->wq.rq_addr); + pr_debug("%s sq_num_entries %d, rq_num_entries %d qpid 0x%0x qhp %p dma_addr 0x%llx size %d rq_addr 0x%x\n", + __func__, qhp->attr.sq_num_entries, qhp->attr.rq_num_entries, + qhp->wq.qpid, qhp, (unsigned long long)qhp->wq.dma_addr, + 1 << qhp->wq.size_log2, qhp->wq.rq_addr); return &qhp->ibqp; } @@ -990,7 +986,7 @@ static int iwch_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, enum iwch_qp_attr_mask mask = 0; struct iwch_qp_attributes attrs; - PDBG("%s ib_qp %p\n", __func__, ibqp); + pr_debug("%s ib_qp %p\n", __func__, ibqp); /* iwarp does not support the RTR state */ if ((attr_mask & IB_QP_STATE) && (attr->qp_state == IB_QPS_RTR)) @@ -1023,20 +1019,20 @@ static int iwch_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, void iwch_qp_add_ref(struct ib_qp *qp) { - PDBG("%s ib_qp %p\n", __func__, qp); + pr_debug("%s ib_qp %p\n", __func__, qp); atomic_inc(&(to_iwch_qp(qp)->refcnt)); } void iwch_qp_rem_ref(struct ib_qp *qp) { - PDBG("%s ib_qp %p\n", __func__, qp); + pr_debug("%s ib_qp %p\n", __func__, qp); if (atomic_dec_and_test(&(to_iwch_qp(qp)->refcnt))) wake_up(&(to_iwch_qp(qp)->wait)); } static struct ib_qp *iwch_get_qp(struct ib_device *dev, int qpn) { - PDBG("%s ib_dev %p qpn 0x%x\n", __func__, dev, qpn); + pr_debug("%s ib_dev %p qpn 0x%x\n", __func__, dev, qpn); return (struct ib_qp *)get_qhp(to_iwch_dev(dev), qpn); } @@ -1044,7 +1040,7 @@ static struct ib_qp *iwch_get_qp(struct ib_device *dev, int qpn) static int iwch_query_pkey(struct ib_device *ibdev, u8 port, u16 index, u16 * pkey) { - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); *pkey = 0; return 0; } @@ -1054,8 +1050,8 @@ static int iwch_query_gid(struct ib_device *ibdev, u8 port, { struct iwch_dev *dev; - PDBG("%s ibdev %p, port %d, index %d, gid %p\n", - __func__, ibdev, port, index, gid); + pr_debug("%s ibdev %p, port %d, index %d, gid %p\n", + __func__, ibdev, port, index, gid); dev = to_iwch_dev(ibdev); BUG_ON(port == 0 || port > 2); memset(&(gid->raw[0]), 0, sizeof(gid->raw)); @@ -1090,7 +1086,7 @@ static int iwch_query_device(struct ib_device *ibdev, struct ib_device_attr *pro struct iwch_dev *dev; - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); if (uhw->inlen || uhw->outlen) return -EINVAL; @@ -1128,7 +1124,7 @@ static int iwch_query_port(struct ib_device *ibdev, struct net_device *netdev; struct in_device *inetdev; - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); dev = to_iwch_dev(ibdev); netdev = dev->rdev.port_info.lldevs[port-1]; @@ -1171,7 +1167,7 @@ static ssize_t show_rev(struct device *dev, struct device_attribute *attr, { struct iwch_dev *iwch_dev = container_of(dev, struct iwch_dev, ibdev.dev); - PDBG("%s dev 0x%p\n", __func__, dev); + pr_debug("%s dev 0x%p\n", __func__, dev); return sprintf(buf, "%d\n", iwch_dev->rdev.t3cdev_p->type); } @@ -1183,7 +1179,7 @@ static ssize_t show_hca(struct device *dev, struct device_attribute *attr, struct ethtool_drvinfo info; struct net_device *lldev = iwch_dev->rdev.t3cdev_p->lldev; - PDBG("%s dev 0x%p\n", __func__, dev); + pr_debug("%s dev 0x%p\n", __func__, dev); lldev->ethtool_ops->get_drvinfo(lldev, &info); return sprintf(buf, "%s\n", info.driver); } @@ -1193,7 +1189,7 @@ static ssize_t show_board(struct device *dev, struct device_attribute *attr, { struct iwch_dev *iwch_dev = container_of(dev, struct iwch_dev, ibdev.dev); - PDBG("%s dev 0x%p\n", __func__, dev); + pr_debug("%s dev 0x%p\n", __func__, dev); return sprintf(buf, "%x.%x\n", iwch_dev->rdev.rnic_info.pdev->vendor, iwch_dev->rdev.rnic_info.pdev->device); } @@ -1278,7 +1274,7 @@ static int iwch_get_mib(struct ib_device *ibdev, struct rdma_hw_stats *stats, if (port != 0 || !stats) return -ENOSYS; - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); dev = to_iwch_dev(ibdev); ret = dev->rdev.t3cdev_p->ctl(dev->rdev.t3cdev_p, RDMA_GET_MIB, &m); if (ret) @@ -1348,7 +1344,7 @@ static void get_dev_fw_ver_str(struct ib_device *ibdev, char *str, struct ethtool_drvinfo info; struct net_device *lldev = iwch_dev->rdev.t3cdev_p->lldev; - PDBG("%s dev 0x%p\n", __func__, iwch_dev); + pr_debug("%s dev 0x%p\n", __func__, iwch_dev); lldev->ethtool_ops->get_drvinfo(lldev, &info); snprintf(str, str_len, "%s", info.fw_version); } @@ -1358,7 +1354,7 @@ int iwch_register_device(struct iwch_dev *dev) int ret; int i; - PDBG("%s iwch_dev %p\n", __func__, dev); + pr_debug("%s iwch_dev %p\n", __func__, dev); strlcpy(dev->ibdev.name, "cxgb3_%d", IB_DEVICE_NAME_MAX); memset(&dev->ibdev.node_guid, 0, sizeof(dev->ibdev.node_guid)); memcpy(&dev->ibdev.node_guid, dev->rdev.t3cdev_p->lldev->dev_addr, 6); @@ -1469,7 +1465,7 @@ void iwch_unregister_device(struct iwch_dev *dev) { int i; - PDBG("%s iwch_dev %p\n", __func__, dev); + pr_debug("%s iwch_dev %p\n", __func__, dev); for (i = 0; i < ARRAY_SIZE(iwch_class_attributes); ++i) device_remove_file(&dev->ibdev.dev, iwch_class_attributes[i]); diff --git a/drivers/infiniband/hw/cxgb3/iwch_provider.h b/drivers/infiniband/hw/cxgb3/iwch_provider.h index 252c464a09f6..9e216edec4c0 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_provider.h +++ b/drivers/infiniband/hw/cxgb3/iwch_provider.h @@ -217,8 +217,9 @@ static inline struct iwch_mm_entry *remove_mmap(struct iwch_ucontext *ucontext, if (mm->key == key && mm->len == len) { list_del_init(&mm->entry); spin_unlock(&ucontext->mmap_lock); - PDBG("%s key 0x%x addr 0x%llx len %d\n", __func__, - key, (unsigned long long) mm->addr, mm->len); + pr_debug("%s key 0x%x addr 0x%llx len %d\n", + __func__, key, + (unsigned long long)mm->addr, mm->len); return mm; } } @@ -230,8 +231,8 @@ static inline void insert_mmap(struct iwch_ucontext *ucontext, struct iwch_mm_entry *mm) { spin_lock(&ucontext->mmap_lock); - PDBG("%s key 0x%x addr 0x%llx len %d\n", __func__, - mm->key, (unsigned long long) mm->addr, mm->len); + pr_debug("%s key 0x%x addr 0x%llx len %d\n", + __func__, mm->key, (unsigned long long)mm->addr, mm->len); list_add_tail(&mm->entry, &ucontext->mmaps); spin_unlock(&ucontext->mmap_lock); } diff --git a/drivers/infiniband/hw/cxgb3/iwch_qp.c b/drivers/infiniband/hw/cxgb3/iwch_qp.c index a9194db7f9b8..ba6d5d281b03 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_qp.c +++ b/drivers/infiniband/hw/cxgb3/iwch_qp.c @@ -208,30 +208,30 @@ static int iwch_sgl2pbl_map(struct iwch_dev *rhp, struct ib_sge *sg_list, mhp = get_mhp(rhp, (sg_list[i].lkey) >> 8); if (!mhp) { - PDBG("%s %d\n", __func__, __LINE__); + pr_debug("%s %d\n", __func__, __LINE__); return -EIO; } if (!mhp->attr.state) { - PDBG("%s %d\n", __func__, __LINE__); + pr_debug("%s %d\n", __func__, __LINE__); return -EIO; } if (mhp->attr.zbva) { - PDBG("%s %d\n", __func__, __LINE__); + pr_debug("%s %d\n", __func__, __LINE__); return -EIO; } if (sg_list[i].addr < mhp->attr.va_fbo) { - PDBG("%s %d\n", __func__, __LINE__); + pr_debug("%s %d\n", __func__, __LINE__); return -EINVAL; } if (sg_list[i].addr + ((u64) sg_list[i].length) < sg_list[i].addr) { - PDBG("%s %d\n", __func__, __LINE__); + pr_debug("%s %d\n", __func__, __LINE__); return -EINVAL; } if (sg_list[i].addr + ((u64) sg_list[i].length) > mhp->attr.va_fbo + ((u64) mhp->attr.len)) { - PDBG("%s %d\n", __func__, __LINE__); + pr_debug("%s %d\n", __func__, __LINE__); return -EINVAL; } offset = sg_list[i].addr - mhp->attr.va_fbo; @@ -427,8 +427,8 @@ int iwch_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, err = build_inv_stag(wqe, wr, &t3_wr_flit_cnt); break; default: - PDBG("%s post of type=%d TBD!\n", __func__, - wr->opcode); + pr_debug("%s post of type=%d TBD!\n", __func__, + wr->opcode); err = -EINVAL; } if (err) @@ -444,10 +444,10 @@ int iwch_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, Q_GENBIT(qhp->wq.wptr, qhp->wq.size_log2), 0, t3_wr_flit_cnt, (wr_cnt == 1) ? T3_SOPEOP : T3_SOP); - PDBG("%s cookie 0x%llx wq idx 0x%x swsq idx %ld opcode %d\n", - __func__, (unsigned long long) wr->wr_id, idx, - Q_PTR2IDX(qhp->wq.sq_wptr, qhp->wq.sq_size_log2), - sqp->opcode); + pr_debug("%s cookie 0x%llx wq idx 0x%x swsq idx %ld opcode %d\n", + __func__, (unsigned long long)wr->wr_id, idx, + Q_PTR2IDX(qhp->wq.sq_wptr, qhp->wq.sq_size_log2), + sqp->opcode); wr = wr->next; num_wrs--; qhp->wq.wptr += wr_cnt; @@ -508,9 +508,9 @@ int iwch_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr, build_fw_riwrh((void *) wqe, T3_WR_RCV, T3_COMPLETION_FLAG, Q_GENBIT(qhp->wq.wptr, qhp->wq.size_log2), 0, sizeof(struct t3_receive_wr) >> 3, T3_SOPEOP); - PDBG("%s cookie 0x%llx idx 0x%x rq_wptr 0x%x rw_rptr 0x%x " - "wqe %p \n", __func__, (unsigned long long) wr->wr_id, - idx, qhp->wq.rq_wptr, qhp->wq.rq_rptr, wqe); + pr_debug("%s cookie 0x%llx idx 0x%x rq_wptr 0x%x rw_rptr 0x%x wqe %p\n", + __func__, (unsigned long long)wr->wr_id, + idx, qhp->wq.rq_wptr, qhp->wq.rq_rptr, wqe); ++(qhp->wq.rq_wptr); ++(qhp->wq.wptr); wr = wr->next; @@ -664,10 +664,10 @@ int iwch_post_zb_read(struct iwch_ep *ep) struct sk_buff *skb; u8 flit_cnt = sizeof(struct t3_rdma_read_wr) >> 3; - PDBG("%s enter\n", __func__); + pr_debug("%s enter\n", __func__); skb = alloc_skb(40, GFP_KERNEL); if (!skb) { - printk(KERN_ERR "%s cannot send zb_read!!\n", __func__); + pr_err("%s cannot send zb_read!!\n", __func__); return -ENOMEM; } wqe = (union t3_wr *)skb_put(skb, sizeof(struct t3_rdma_read_wr)); @@ -696,10 +696,10 @@ int iwch_post_terminate(struct iwch_qp *qhp, struct respQ_msg_t *rsp_msg) struct terminate_message *term; struct sk_buff *skb; - PDBG("%s %d\n", __func__, __LINE__); + pr_debug("%s %d\n", __func__, __LINE__); skb = alloc_skb(40, GFP_ATOMIC); if (!skb) { - printk(KERN_ERR "%s cannot send TERMINATE!\n", __func__); + pr_err("%s cannot send TERMINATE!\n", __func__); return -ENOMEM; } wqe = (union t3_wr *)skb_put(skb, 40); @@ -729,7 +729,7 @@ static void __flush_qp(struct iwch_qp *qhp, struct iwch_cq *rchp, int flushed; - PDBG("%s qhp %p rchp %p schp %p\n", __func__, qhp, rchp, schp); + pr_debug("%s qhp %p rchp %p schp %p\n", __func__, qhp, rchp, schp); /* take a ref on the qhp since we must release the lock */ atomic_inc(&qhp->refcnt); spin_unlock(&qhp->lock); @@ -807,7 +807,7 @@ u16 iwch_rqes_posted(struct iwch_qp *qhp) count++; wqe++; } - PDBG("%s qhp %p count %u\n", __func__, qhp, count); + pr_debug("%s qhp %p count %u\n", __func__, qhp, count); return count; } @@ -854,12 +854,12 @@ static int rdma_init(struct iwch_dev *rhp, struct iwch_qp *qhp, } else init_attr.rtr_type = 0; init_attr.irs = qhp->ep->rcv_seq; - PDBG("%s init_attr.rq_addr 0x%x init_attr.rq_size = %d " - "flags 0x%x qpcaps 0x%x\n", __func__, - init_attr.rq_addr, init_attr.rq_size, - init_attr.flags, init_attr.qpcaps); + pr_debug("%s init_attr.rq_addr 0x%x init_attr.rq_size = %d flags 0x%x qpcaps 0x%x\n", + __func__, + init_attr.rq_addr, init_attr.rq_size, + init_attr.flags, init_attr.qpcaps); ret = cxio_rdma_init(&rhp->rdev, &init_attr); - PDBG("%s ret %d\n", __func__, ret); + pr_debug("%s ret %d\n", __func__, ret); return ret; } @@ -877,9 +877,9 @@ int iwch_modify_qp(struct iwch_dev *rhp, struct iwch_qp *qhp, int free = 0; struct iwch_ep *ep = NULL; - PDBG("%s qhp %p qpid 0x%x ep %p state %d -> %d\n", __func__, - qhp, qhp->wq.qpid, qhp->ep, qhp->attr.state, - (mask & IWCH_QP_ATTR_NEXT_STATE) ? attrs->next_state : -1); + pr_debug("%s qhp %p qpid 0x%x ep %p state %d -> %d\n", __func__, + qhp, qhp->wq.qpid, qhp->ep, qhp->attr.state, + (mask & IWCH_QP_ATTR_NEXT_STATE) ? attrs->next_state : -1); spin_lock_irqsave(&qhp->lock, flag); @@ -1034,16 +1034,15 @@ int iwch_modify_qp(struct iwch_dev *rhp, struct iwch_qp *qhp, goto err; break; default: - printk(KERN_ERR "%s in a bad state %d\n", - __func__, qhp->attr.state); + pr_err("%s in a bad state %d\n", __func__, qhp->attr.state); ret = -EINVAL; goto err; break; } goto out; err: - PDBG("%s disassociating ep %p qpid 0x%x\n", __func__, qhp->ep, - qhp->wq.qpid); + pr_debug("%s disassociating ep %p qpid 0x%x\n", __func__, qhp->ep, + qhp->wq.qpid); /* disassociate the LLP connection */ qhp->attr.llp_stream_handle = NULL; @@ -1077,6 +1076,6 @@ out: if (free) put_ep(&ep->com); - PDBG("%s exit state %d\n", __func__, qhp->attr.state); + pr_debug("%s exit state %d\n", __func__, qhp->attr.state); return ret; } diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index 03a1b0e64fc3..b6fe45924c6e 100644 --- a/drivers/infiniband/hw/cxgb4/cm.c +++ b/drivers/infiniband/hw/cxgb4/cm.c @@ -101,7 +101,7 @@ MODULE_PARM_DESC(enable_tcp_window_scaling, int c4iw_debug; module_param(c4iw_debug, int, 0644); -MODULE_PARM_DESC(c4iw_debug, "Enable debug logging (default=0)"); +MODULE_PARM_DESC(c4iw_debug, "obsolete"); static int peer2peer = 1; module_param(peer2peer, int, 0644); @@ -180,7 +180,7 @@ static void ref_qp(struct c4iw_ep *ep) static void start_ep_timer(struct c4iw_ep *ep) { - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); if (timer_pending(&ep->timer)) { pr_err("%s timer already started! ep %p\n", __func__, ep); @@ -196,7 +196,7 @@ static void start_ep_timer(struct c4iw_ep *ep) static int stop_ep_timer(struct c4iw_ep *ep) { - PDBG("%s ep %p stopping\n", __func__, ep); + pr_debug("%s ep %p stopping\n", __func__, ep); del_timer_sync(&ep->timer); if (!test_and_set_bit(TIMEOUT, &ep->com.flags)) { c4iw_put_ep(&ep->com); @@ -212,7 +212,7 @@ static int c4iw_l2t_send(struct c4iw_rdev *rdev, struct sk_buff *skb, if (c4iw_fatal_error(rdev)) { kfree_skb(skb); - PDBG("%s - device in error state - dropping\n", __func__); + pr_debug("%s - device in error state - dropping\n", __func__); return -EIO; } error = cxgb4_l2t_send(rdev->lldi.ports[0], skb, l2e); @@ -229,7 +229,7 @@ int c4iw_ofld_send(struct c4iw_rdev *rdev, struct sk_buff *skb) if (c4iw_fatal_error(rdev)) { kfree_skb(skb); - PDBG("%s - device in error state - dropping\n", __func__); + pr_debug("%s - device in error state - dropping\n", __func__); return -EIO; } error = cxgb4_ofld_send(rdev->lldi.ports[0], skb); @@ -263,10 +263,10 @@ static void set_emss(struct c4iw_ep *ep, u16 opt) if (ep->emss < 128) ep->emss = 128; if (ep->emss & 7) - PDBG("Warning: misaligned mtu idx %u mss %u emss=%u\n", - TCPOPT_MSS_G(opt), ep->mss, ep->emss); - PDBG("%s mss_idx %u mss %u emss=%u\n", __func__, TCPOPT_MSS_G(opt), - ep->mss, ep->emss); + pr_debug("Warning: misaligned mtu idx %u mss %u emss=%u\n", + TCPOPT_MSS_G(opt), ep->mss, ep->emss); + pr_debug("%s mss_idx %u mss %u emss=%u\n", __func__, TCPOPT_MSS_G(opt), + ep->mss, ep->emss); } static enum c4iw_ep_state state_read(struct c4iw_ep_common *epc) @@ -287,7 +287,7 @@ static void __state_set(struct c4iw_ep_common *epc, enum c4iw_ep_state new) static void state_set(struct c4iw_ep_common *epc, enum c4iw_ep_state new) { mutex_lock(&epc->mutex); - PDBG("%s - %s -> %s\n", __func__, states[epc->state], states[new]); + pr_debug("%s - %s -> %s\n", __func__, states[epc->state], states[new]); __state_set(epc, new); mutex_unlock(&epc->mutex); return; @@ -322,7 +322,7 @@ static void *alloc_ep(int size, gfp_t gfp) mutex_init(&epc->mutex); c4iw_init_wr_wait(&epc->wr_wait); } - PDBG("%s alloc ep %p\n", __func__, epc); + pr_debug("%s alloc ep %p\n", __func__, epc); return epc; } @@ -384,7 +384,7 @@ void _c4iw_free_ep(struct kref *kref) struct c4iw_ep *ep; ep = container_of(kref, struct c4iw_ep, com.kref); - PDBG("%s ep %p state %s\n", __func__, ep, states[ep->com.state]); + pr_debug("%s ep %p state %s\n", __func__, ep, states[ep->com.state]); if (test_bit(QP_REFERENCED, &ep->com.flags)) deref_qp(ep); if (test_bit(RELEASE_RESOURCES, &ep->com.flags)) { @@ -467,7 +467,7 @@ static struct net_device *get_real_dev(struct net_device *egress_dev) static void arp_failure_discard(void *handle, struct sk_buff *skb) { - pr_err(MOD "ARP failure\n"); + pr_err("ARP failure\n"); kfree_skb(skb); } @@ -528,7 +528,7 @@ static void pass_accept_rpl_arp_failure(void *handle, struct sk_buff *skb) { struct c4iw_ep *ep = handle; - pr_err(MOD "ARP failure during accept - tid %u -dropping connection\n", + pr_err("ARP failure during accept - tid %u - dropping connection\n", ep->hwtid); __state_set(&ep->com, DEAD); @@ -542,7 +542,7 @@ static void act_open_req_arp_failure(void *handle, struct sk_buff *skb) { struct c4iw_ep *ep = handle; - printk(KERN_ERR MOD "ARP failure during connect\n"); + pr_err("ARP failure during connect\n"); connect_reply_upcall(ep, -EHOSTUNREACH); __state_set(&ep->com, DEAD); if (ep->com.remote_addr.ss_family == AF_INET6) { @@ -567,7 +567,7 @@ static void abort_arp_failure(void *handle, struct sk_buff *skb) struct c4iw_rdev *rdev = &ep->com.dev->rdev; struct cpl_abort_req *req = cplhdr(skb); - PDBG("%s rdev %p\n", __func__, rdev); + pr_debug("%s rdev %p\n", __func__, rdev); req->cmd = CPL_ABORT_NO_RST; ret = c4iw_ofld_send(rdev, skb); if (ret) { @@ -642,7 +642,7 @@ static int send_halfclose(struct c4iw_ep *ep) struct sk_buff *skb = skb_dequeue(&ep->com.ep_skb_list); u32 wrlen = roundup(sizeof(struct cpl_close_con_req), 16); - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); if (WARN_ON(!skb)) return -ENOMEM; @@ -657,7 +657,7 @@ static int send_abort(struct c4iw_ep *ep) u32 wrlen = roundup(sizeof(struct cpl_abort_req), 16); struct sk_buff *req_skb = skb_dequeue(&ep->com.ep_skb_list); - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); if (WARN_ON(!req_skb)) return -ENOMEM; @@ -720,12 +720,11 @@ static int send_connect(struct c4iw_ep *ep) roundup(sizev4, 16) : roundup(sizev6, 16); - PDBG("%s ep %p atid %u\n", __func__, ep, ep->atid); + pr_debug("%s ep %p atid %u\n", __func__, ep, ep->atid); skb = get_skb(NULL, wrlen, GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "%s - failed to alloc skb.\n", - __func__); + pr_err("%s - failed to alloc skb\n", __func__); return -ENOMEM; } set_wr_txq(skb, CPL_PRIORITY_SETUP, ep->ctrlq_idx); @@ -822,13 +821,13 @@ static int send_connect(struct c4iw_ep *ep) t5req->params = cpu_to_be64(FILTER_TUPLE_V(params)); t5req->rsvd = cpu_to_be32(isn); - PDBG("%s snd_isn %u\n", __func__, t5req->rsvd); + pr_debug("%s snd_isn %u\n", __func__, t5req->rsvd); t5req->opt2 = cpu_to_be32(opt2); } else { t6req->params = cpu_to_be64(FILTER_TUPLE_V(params)); t6req->rsvd = cpu_to_be32(isn); - PDBG("%s snd_isn %u\n", __func__, t6req->rsvd); + pr_debug("%s snd_isn %u\n", __func__, t6req->rsvd); t6req->opt2 = cpu_to_be32(opt2); } } @@ -877,13 +876,13 @@ static int send_connect(struct c4iw_ep *ep) t5req6->params = cpu_to_be64(FILTER_TUPLE_V(params)); t5req6->rsvd = cpu_to_be32(isn); - PDBG("%s snd_isn %u\n", __func__, t5req6->rsvd); + pr_debug("%s snd_isn %u\n", __func__, t5req6->rsvd); t5req6->opt2 = cpu_to_be32(opt2); } else { t6req6->params = cpu_to_be64(FILTER_TUPLE_V(params)); t6req6->rsvd = cpu_to_be32(isn); - PDBG("%s snd_isn %u\n", __func__, t6req6->rsvd); + pr_debug("%s snd_isn %u\n", __func__, t6req6->rsvd); t6req6->opt2 = cpu_to_be32(opt2); } @@ -907,7 +906,8 @@ static int send_mpa_req(struct c4iw_ep *ep, struct sk_buff *skb, struct mpa_message *mpa; struct mpa_v2_conn_params mpa_v2_params; - PDBG("%s ep %p tid %u pd_len %d\n", __func__, ep, ep->hwtid, ep->plen); + pr_debug("%s ep %p tid %u pd_len %d\n", + __func__, ep, ep->hwtid, ep->plen); BUG_ON(skb_cloned(skb)); @@ -961,8 +961,8 @@ static int send_mpa_req(struct c4iw_ep *ep, struct sk_buff *skb, if (mpa_rev_to_use == 2) { mpa->private_data_size = htons(ntohs(mpa->private_data_size) + sizeof (struct mpa_v2_conn_params)); - PDBG("%s initiator ird %u ord %u\n", __func__, ep->ird, - ep->ord); + pr_debug("%s initiator ird %u ord %u\n", __func__, ep->ird, + ep->ord); mpa_v2_params.ird = htons((u16)ep->ird); mpa_v2_params.ord = htons((u16)ep->ord); @@ -1014,7 +1014,8 @@ static int send_mpa_reject(struct c4iw_ep *ep, const void *pdata, u8 plen) struct sk_buff *skb; struct mpa_v2_conn_params mpa_v2_params; - PDBG("%s ep %p tid %u pd_len %d\n", __func__, ep, ep->hwtid, ep->plen); + pr_debug("%s ep %p tid %u pd_len %d\n", + __func__, ep, ep->hwtid, ep->plen); mpalen = sizeof(*mpa) + plen; if (ep->mpa_attr.version == 2 && ep->mpa_attr.enhanced_rdma_conn) @@ -1023,7 +1024,7 @@ static int send_mpa_reject(struct c4iw_ep *ep, const void *pdata, u8 plen) skb = get_skb(NULL, wrlen, GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "%s - cannot alloc skb!\n", __func__); + pr_err("%s - cannot alloc skb!\n", __func__); return -ENOMEM; } set_wr_txq(skb, CPL_PRIORITY_DATA, ep->txq_idx); @@ -1094,7 +1095,8 @@ static int send_mpa_reply(struct c4iw_ep *ep, const void *pdata, u8 plen) struct sk_buff *skb; struct mpa_v2_conn_params mpa_v2_params; - PDBG("%s ep %p tid %u pd_len %d\n", __func__, ep, ep->hwtid, ep->plen); + pr_debug("%s ep %p tid %u pd_len %d\n", + __func__, ep, ep->hwtid, ep->plen); mpalen = sizeof(*mpa) + plen; if (ep->mpa_attr.version == 2 && ep->mpa_attr.enhanced_rdma_conn) @@ -1103,7 +1105,7 @@ static int send_mpa_reply(struct c4iw_ep *ep, const void *pdata, u8 plen) skb = get_skb(NULL, wrlen, GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "%s - cannot alloc skb!\n", __func__); + pr_err("%s - cannot alloc skb!\n", __func__); return -ENOMEM; } set_wr_txq(skb, CPL_PRIORITY_DATA, ep->txq_idx); @@ -1185,8 +1187,8 @@ static int act_establish(struct c4iw_dev *dev, struct sk_buff *skb) ep = lookup_atid(t, atid); - PDBG("%s ep %p tid %u snd_isn %u rcv_isn %u\n", __func__, ep, tid, - be32_to_cpu(req->snd_isn), be32_to_cpu(req->rcv_isn)); + pr_debug("%s ep %p tid %u snd_isn %u rcv_isn %u\n", __func__, ep, tid, + be32_to_cpu(req->snd_isn), be32_to_cpu(req->rcv_isn)); mutex_lock(&ep->com.mutex); dst_confirm(ep->dst); @@ -1229,13 +1231,13 @@ static void close_complete_upcall(struct c4iw_ep *ep, int status) { struct iw_cm_event event; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_CLOSE; event.status = status; if (ep->com.cm_id) { - PDBG("close complete delivered ep %p cm_id %p tid %u\n", - ep, ep->com.cm_id, ep->hwtid); + pr_debug("close complete delivered ep %p cm_id %p tid %u\n", + ep, ep->com.cm_id, ep->hwtid); ep->com.cm_id->event_handler(ep->com.cm_id, &event); deref_cm_id(&ep->com); set_bit(CLOSE_UPCALL, &ep->com.history); @@ -1246,12 +1248,12 @@ static void peer_close_upcall(struct c4iw_ep *ep) { struct iw_cm_event event; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_DISCONNECT; if (ep->com.cm_id) { - PDBG("peer close delivered ep %p cm_id %p tid %u\n", - ep, ep->com.cm_id, ep->hwtid); + pr_debug("peer close delivered ep %p cm_id %p tid %u\n", + ep, ep->com.cm_id, ep->hwtid); ep->com.cm_id->event_handler(ep->com.cm_id, &event); set_bit(DISCONN_UPCALL, &ep->com.history); } @@ -1261,13 +1263,13 @@ static void peer_abort_upcall(struct c4iw_ep *ep) { struct iw_cm_event event; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_CLOSE; event.status = -ECONNRESET; if (ep->com.cm_id) { - PDBG("abort delivered ep %p cm_id %p tid %u\n", ep, - ep->com.cm_id, ep->hwtid); + pr_debug("abort delivered ep %p cm_id %p tid %u\n", ep, + ep->com.cm_id, ep->hwtid); ep->com.cm_id->event_handler(ep->com.cm_id, &event); deref_cm_id(&ep->com); set_bit(ABORT_UPCALL, &ep->com.history); @@ -1278,7 +1280,8 @@ static void connect_reply_upcall(struct c4iw_ep *ep, int status) { struct iw_cm_event event; - PDBG("%s ep %p tid %u status %d\n", __func__, ep, ep->hwtid, status); + pr_debug("%s ep %p tid %u status %d\n", + __func__, ep, ep->hwtid, status); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_CONNECT_REPLY; event.status = status; @@ -1307,8 +1310,8 @@ static void connect_reply_upcall(struct c4iw_ep *ep, int status) } } - PDBG("%s ep %p tid %u status %d\n", __func__, ep, - ep->hwtid, status); + pr_debug("%s ep %p tid %u status %d\n", __func__, ep, + ep->hwtid, status); set_bit(CONN_RPL_UPCALL, &ep->com.history); ep->com.cm_id->event_handler(ep->com.cm_id, &event); @@ -1321,7 +1324,7 @@ static int connect_request_upcall(struct c4iw_ep *ep) struct iw_cm_event event; int ret; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_CONNECT_REQUEST; memcpy(&event.local_addr, &ep->com.local_addr, @@ -1358,13 +1361,13 @@ static void established_upcall(struct c4iw_ep *ep) { struct iw_cm_event event; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); memset(&event, 0, sizeof(event)); event.event = IW_CM_EVENT_ESTABLISHED; event.ird = ep->ord; event.ord = ep->ird; if (ep->com.cm_id) { - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); ep->com.cm_id->event_handler(ep->com.cm_id, &event); set_bit(ESTAB_UPCALL, &ep->com.history); } @@ -1376,10 +1379,11 @@ static int update_rx_credits(struct c4iw_ep *ep, u32 credits) u32 wrlen = roundup(sizeof(struct cpl_rx_data_ack), 16); u32 credit_dack; - PDBG("%s ep %p tid %u credits %u\n", __func__, ep, ep->hwtid, credits); + pr_debug("%s ep %p tid %u credits %u\n", + __func__, ep, ep->hwtid, credits); skb = get_skb(NULL, wrlen, GFP_KERNEL); if (!skb) { - printk(KERN_ERR MOD "update_rx_credits - cannot alloc skb!\n"); + pr_err("update_rx_credits - cannot alloc skb!\n"); return 0; } @@ -1427,7 +1431,7 @@ static int process_mpa_reply(struct c4iw_ep *ep, struct sk_buff *skb) int err; int disconnect = 0; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); /* * If we get more than the supported amount of private data @@ -1454,8 +1458,8 @@ static int process_mpa_reply(struct c4iw_ep *ep, struct sk_buff *skb) /* Validate MPA header. */ if (mpa->revision > mpa_rev) { - printk(KERN_ERR MOD "%s MPA version mismatch. Local = %d," - " Received = %d\n", __func__, mpa_rev, mpa->revision); + pr_err("%s MPA version mismatch. Local = %d, Received = %d\n", + __func__, mpa_rev, mpa->revision); err = -EPROTO; goto err_stop_timer; } @@ -1525,8 +1529,9 @@ static int process_mpa_reply(struct c4iw_ep *ep, struct sk_buff *skb) MPA_V2_IRD_ORD_MASK; resp_ord = ntohs(mpa_v2_params->ord) & MPA_V2_IRD_ORD_MASK; - PDBG("%s responder ird %u ord %u ep ird %u ord %u\n", - __func__, resp_ird, resp_ord, ep->ird, ep->ord); + pr_debug("%s responder ird %u ord %u ep ird %u ord %u\n", + __func__, + resp_ird, resp_ord, ep->ird, ep->ord); /* * This is a double-check. Ideally, below checks are @@ -1570,12 +1575,11 @@ static int process_mpa_reply(struct c4iw_ep *ep, struct sk_buff *skb) if (peer2peer) ep->mpa_attr.p2p_type = p2p_type; - PDBG("%s - crc_enabled=%d, recv_marker_enabled=%d, " - "xmit_marker_enabled=%d, version=%d p2p_type=%d local-p2p_type = " - "%d\n", __func__, ep->mpa_attr.crc_enabled, - ep->mpa_attr.recv_marker_enabled, - ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version, - ep->mpa_attr.p2p_type, p2p_type); + pr_debug("%s - crc_enabled=%d, recv_marker_enabled=%d, xmit_marker_enabled=%d, version=%d p2p_type=%d local-p2p_type = %d\n", + __func__, ep->mpa_attr.crc_enabled, + ep->mpa_attr.recv_marker_enabled, + ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version, + ep->mpa_attr.p2p_type, p2p_type); /* * If responder's RTR does not match with that of initiator, assign @@ -1610,7 +1614,7 @@ static int process_mpa_reply(struct c4iw_ep *ep, struct sk_buff *skb) * supports, generate TERM message */ if (rtr_mismatch) { - printk(KERN_ERR "%s: RTR mismatch, sending TERM\n", __func__); + pr_err("%s: RTR mismatch, sending TERM\n", __func__); attrs.layer_etype = LAYER_MPA | DDP_LLP; attrs.ecode = MPA_NOMATCH_RTR; attrs.next_state = C4IW_QP_STATE_TERMINATE; @@ -1629,8 +1633,7 @@ static int process_mpa_reply(struct c4iw_ep *ep, struct sk_buff *skb) * initiator ORD. */ if (insuff_ird) { - printk(KERN_ERR "%s: Insufficient IRD, sending TERM\n", - __func__); + pr_err("%s: Insufficient IRD, sending TERM\n", __func__); attrs.layer_etype = LAYER_MPA | DDP_LLP; attrs.ecode = MPA_INSUFF_IRD; attrs.next_state = C4IW_QP_STATE_TERMINATE; @@ -1669,7 +1672,7 @@ static int process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb) struct mpa_v2_conn_params *mpa_v2_params; u16 plen; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); /* * If we get more than the supported amount of private data @@ -1678,7 +1681,7 @@ static int process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb) if (ep->mpa_pkt_len + skb->len > sizeof(ep->mpa_pkt)) goto err_stop_timer; - PDBG("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__); + pr_debug("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__); /* * Copy the new data into our accumulation buffer. @@ -1694,15 +1697,15 @@ static int process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb) if (ep->mpa_pkt_len < sizeof(*mpa)) return 0; - PDBG("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__); + pr_debug("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__); mpa = (struct mpa_message *) ep->mpa_pkt; /* * Validate MPA Header. */ if (mpa->revision > mpa_rev) { - printk(KERN_ERR MOD "%s MPA version mismatch. Local = %d," - " Received = %d\n", __func__, mpa_rev, mpa->revision); + pr_err("%s MPA version mismatch. Local = %d, Received = %d\n", + __func__, mpa_rev, mpa->revision); goto err_stop_timer; } @@ -1757,8 +1760,8 @@ static int process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb) MPA_V2_IRD_ORD_MASK; ep->ord = min_t(u32, ep->ord, cur_max_read_depth(ep->com.dev)); - PDBG("%s initiator ird %u ord %u\n", __func__, ep->ird, - ep->ord); + pr_debug("%s initiator ird %u ord %u\n", + __func__, ep->ird, ep->ord); if (ntohs(mpa_v2_params->ird) & MPA_V2_PEER2PEER_MODEL) if (peer2peer) { if (ntohs(mpa_v2_params->ord) & @@ -1775,11 +1778,11 @@ static int process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb) if (peer2peer) ep->mpa_attr.p2p_type = p2p_type; - PDBG("%s - crc_enabled=%d, recv_marker_enabled=%d, " - "xmit_marker_enabled=%d, version=%d p2p_type=%d\n", __func__, - ep->mpa_attr.crc_enabled, ep->mpa_attr.recv_marker_enabled, - ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version, - ep->mpa_attr.p2p_type); + pr_debug("%s - crc_enabled=%d, recv_marker_enabled=%d, xmit_marker_enabled=%d, version=%d p2p_type=%d\n", + __func__, + ep->mpa_attr.crc_enabled, ep->mpa_attr.recv_marker_enabled, + ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version, + ep->mpa_attr.p2p_type); __state_set(&ep->com, MPA_REQ_RCVD); @@ -1815,7 +1818,7 @@ static int rx_data(struct c4iw_dev *dev, struct sk_buff *skb) ep = get_ep_from_tid(dev, tid); if (!ep) return 0; - PDBG("%s ep %p tid %u dlen %u\n", __func__, ep, ep->hwtid, dlen); + pr_debug("%s ep %p tid %u dlen %u\n", __func__, ep, ep->hwtid, dlen); skb_pull(skb, sizeof(*hdr)); skb_trim(skb, dlen); mutex_lock(&ep->com.mutex); @@ -1866,10 +1869,10 @@ static int abort_rpl(struct c4iw_dev *dev, struct sk_buff *skb) ep = get_ep_from_tid(dev, tid); if (!ep) { - printk(KERN_WARNING MOD "Abort rpl to freed endpoint\n"); + pr_warn("Abort rpl to freed endpoint\n"); return 0; } - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); mutex_lock(&ep->com.mutex); switch (ep->com.state) { case ABORTING: @@ -1878,8 +1881,7 @@ static int abort_rpl(struct c4iw_dev *dev, struct sk_buff *skb) release = 1; break; default: - printk(KERN_ERR "%s ep %p state %d\n", - __func__, ep, ep->com.state); + pr_err("%s ep %p state %d\n", __func__, ep, ep->com.state); break; } mutex_unlock(&ep->com.mutex); @@ -1995,7 +1997,8 @@ static void set_tcp_window(struct c4iw_ep *ep, struct port_info *pi) { ep->snd_win = snd_win; ep->rcv_win = rcv_win; - PDBG("%s snd_win %d rcv_win %d\n", __func__, ep->snd_win, ep->rcv_win); + pr_debug("%s snd_win %d rcv_win %d\n", + __func__, ep->snd_win, ep->rcv_win); } #define ACT_OPEN_RETRY_COUNT 2 @@ -2100,7 +2103,7 @@ static int c4iw_reconnect(struct c4iw_ep *ep) int iptype; __u8 *ra; - PDBG("%s qp %p cm_id %p\n", __func__, ep->com.qp, ep->com.cm_id); + pr_debug("%s qp %p cm_id %p\n", __func__, ep->com.qp, ep->com.cm_id); init_timer(&ep->timer); c4iw_init_wr_wait(&ep->com.wr_wait); @@ -2124,7 +2127,7 @@ static int c4iw_reconnect(struct c4iw_ep *ep) */ ep->atid = cxgb4_alloc_atid(ep->com.dev->rdev.lldi.tids, ep); if (ep->atid == -1) { - pr_err("%s - cannot alloc atid.\n", __func__); + pr_err("%s - cannot alloc atid\n", __func__); err = -ENOMEM; goto fail2; } @@ -2151,7 +2154,7 @@ static int c4iw_reconnect(struct c4iw_ep *ep) ra = (__u8 *)&raddr6->sin6_addr; } if (!ep->dst) { - pr_err("%s - cannot find route.\n", __func__); + pr_err("%s - cannot find route\n", __func__); err = -EHOSTUNREACH; goto fail3; } @@ -2159,13 +2162,13 @@ static int c4iw_reconnect(struct c4iw_ep *ep) ep->com.dev->rdev.lldi.adapter_type, ep->com.cm_id->tos); if (err) { - pr_err("%s - cannot alloc l2e.\n", __func__); + pr_err("%s - cannot alloc l2e\n", __func__); goto fail4; } - PDBG("%s txq_idx %u tx_chan %u smac_idx %u rss_qid %u l2t_idx %u\n", - __func__, ep->txq_idx, ep->tx_chan, ep->smac_idx, ep->rss_qid, - ep->l2t->idx); + pr_debug("%s txq_idx %u tx_chan %u smac_idx %u rss_qid %u l2t_idx %u\n", + __func__, ep->txq_idx, ep->tx_chan, ep->smac_idx, ep->rss_qid, + ep->l2t->idx); state_set(&ep->com, CONNECTING); ep->tos = ep->com.cm_id->tos; @@ -2215,12 +2218,12 @@ static int act_open_rpl(struct c4iw_dev *dev, struct sk_buff *skb) la6 = (struct sockaddr_in6 *)&ep->com.local_addr; ra6 = (struct sockaddr_in6 *)&ep->com.remote_addr; - PDBG("%s ep %p atid %u status %u errno %d\n", __func__, ep, atid, - status, status2errno(status)); + pr_debug("%s ep %p atid %u status %u errno %d\n", __func__, ep, atid, + status, status2errno(status)); if (cxgb_is_neg_adv(status)) { - PDBG("%s Connection problems for atid %u status %u (%s)\n", - __func__, atid, status, neg_adv_str(status)); + pr_debug("%s Connection problems for atid %u status %u (%s)\n", + __func__, atid, status, neg_adv_str(status)); ep->stats.connect_neg_adv++; mutex_lock(&dev->rdev.stats.lock); dev->rdev.stats.neg_adv++; @@ -2315,11 +2318,11 @@ static int pass_open_rpl(struct c4iw_dev *dev, struct sk_buff *skb) struct c4iw_listen_ep *ep = get_ep_from_stid(dev, stid); if (!ep) { - PDBG("%s stid %d lookup failure!\n", __func__, stid); + pr_debug("%s stid %d lookup failure!\n", __func__, stid); goto out; } - PDBG("%s ep %p status %d error %d\n", __func__, ep, - rpl->status, status2errno(rpl->status)); + pr_debug("%s ep %p status %d error %d\n", __func__, ep, + rpl->status, status2errno(rpl->status)); c4iw_wake_up(&ep->com.wr_wait, status2errno(rpl->status)); c4iw_put_ep(&ep->com); out: @@ -2332,7 +2335,7 @@ static int close_listsrv_rpl(struct c4iw_dev *dev, struct sk_buff *skb) unsigned int stid = GET_TID(rpl); struct c4iw_listen_ep *ep = get_ep_from_stid(dev, stid); - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); c4iw_wake_up(&ep->com.wr_wait, status2errno(rpl->status)); c4iw_put_ep(&ep->com); return 0; @@ -2350,7 +2353,7 @@ static int accept_cr(struct c4iw_ep *ep, struct sk_buff *skb, int win; enum chip_type adapter_type = ep->com.dev->rdev.lldi.adapter_type; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); BUG_ON(skb_cloned(skb)); skb_get(skb); @@ -2421,7 +2424,7 @@ static int accept_cr(struct c4iw_ep *ep, struct sk_buff *skb, if (peer2peer) isn += 4; rpl5->iss = cpu_to_be32(isn); - PDBG("%s iss %u\n", __func__, be32_to_cpu(rpl5->iss)); + pr_debug("%s iss %u\n", __func__, be32_to_cpu(rpl5->iss)); } rpl->opt0 = cpu_to_be64(opt0); @@ -2434,7 +2437,7 @@ static int accept_cr(struct c4iw_ep *ep, struct sk_buff *skb, static void reject_cr(struct c4iw_dev *dev, u32 hwtid, struct sk_buff *skb) { - PDBG("%s c4iw_dev %p tid %u\n", __func__, dev, hwtid); + pr_debug("%s c4iw_dev %p tid %u\n", __func__, dev, hwtid); BUG_ON(skb_cloned(skb)); skb_trim(skb, sizeof(struct cpl_tid_release)); release_tid(&dev->rdev, hwtid, skb); @@ -2460,12 +2463,13 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb) parent_ep = (struct c4iw_ep *)get_ep_from_stid(dev, stid); if (!parent_ep) { - PDBG("%s connect request on invalid stid %d\n", __func__, stid); + pr_debug("%s connect request on invalid stid %d\n", + __func__, stid); goto reject; } if (state_read(&parent_ep->com) != LISTEN) { - PDBG("%s - listening ep not in LISTEN\n", __func__); + pr_debug("%s - listening ep not in LISTEN\n", __func__); goto reject; } @@ -2474,18 +2478,18 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb) /* Find output route */ if (iptype == 4) { - PDBG("%s parent ep %p hwtid %u laddr %pI4 raddr %pI4 lport %d rport %d peer_mss %d\n" - , __func__, parent_ep, hwtid, - local_ip, peer_ip, ntohs(local_port), - ntohs(peer_port), peer_mss); + pr_debug("%s parent ep %p hwtid %u laddr %pI4 raddr %pI4 lport %d rport %d peer_mss %d\n" + , __func__, parent_ep, hwtid, + local_ip, peer_ip, ntohs(local_port), + ntohs(peer_port), peer_mss); dst = cxgb_find_route(&dev->rdev.lldi, get_real_dev, *(__be32 *)local_ip, *(__be32 *)peer_ip, local_port, peer_port, tos); } else { - PDBG("%s parent ep %p hwtid %u laddr %pI6 raddr %pI6 lport %d rport %d peer_mss %d\n" - , __func__, parent_ep, hwtid, - local_ip, peer_ip, ntohs(local_port), - ntohs(peer_port), peer_mss); + pr_debug("%s parent ep %p hwtid %u laddr %pI6 raddr %pI6 lport %d rport %d peer_mss %d\n" + , __func__, parent_ep, hwtid, + local_ip, peer_ip, ntohs(local_port), + ntohs(peer_port), peer_mss); dst = cxgb_find_route6(&dev->rdev.lldi, get_real_dev, local_ip, peer_ip, local_port, peer_port, PASS_OPEN_TOS_G(ntohl(req->tos_stid)), @@ -2493,15 +2497,13 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb) &parent_ep->com.local_addr)->sin6_scope_id); } if (!dst) { - printk(KERN_ERR MOD "%s - failed to find dst entry!\n", - __func__); + pr_err("%s - failed to find dst entry!\n", __func__); goto reject; } child_ep = alloc_ep(sizeof(*child_ep), GFP_KERNEL); if (!child_ep) { - printk(KERN_ERR MOD "%s - failed to allocate ep entry!\n", - __func__); + pr_err("%s - failed to allocate ep entry!\n", __func__); dst_release(dst); goto reject; } @@ -2509,8 +2511,7 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb) err = import_ep(child_ep, iptype, peer_ip, dst, dev, false, parent_ep->com.dev->rdev.lldi.adapter_type, tos); if (err) { - printk(KERN_ERR MOD "%s - failed to allocate l2t entry!\n", - __func__); + pr_err("%s - failed to allocate l2t entry!\n", __func__); dst_release(dst); kfree(child_ep); goto reject; @@ -2571,8 +2572,8 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb) child_ep->dst = dst; child_ep->hwtid = hwtid; - PDBG("%s tx_chan %u smac_idx %u rss_qid %u\n", __func__, - child_ep->tx_chan, child_ep->smac_idx, child_ep->rss_qid); + pr_debug("%s tx_chan %u smac_idx %u rss_qid %u\n", __func__, + child_ep->tx_chan, child_ep->smac_idx, child_ep->rss_qid); init_timer(&child_ep->timer); cxgb4_insert_tid(t, child_ep, hwtid); @@ -2607,12 +2608,12 @@ static int pass_establish(struct c4iw_dev *dev, struct sk_buff *skb) int ret; ep = get_ep_from_tid(dev, tid); - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); ep->snd_seq = be32_to_cpu(req->snd_isn); ep->rcv_seq = be32_to_cpu(req->rcv_isn); - PDBG("%s ep %p hwtid %u tcp_opt 0x%02x\n", __func__, ep, tid, - ntohs(req->tcp_opt)); + pr_debug("%s ep %p hwtid %u tcp_opt 0x%02x\n", __func__, ep, tid, + ntohs(req->tcp_opt)); set_emss(ep, ntohs(req->tcp_opt)); @@ -2644,7 +2645,7 @@ static int peer_close(struct c4iw_dev *dev, struct sk_buff *skb) if (!ep) return 0; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); dst_confirm(ep->dst); set_bit(PEER_CLOSE, &ep->com.history); @@ -2666,12 +2667,12 @@ static int peer_close(struct c4iw_dev *dev, struct sk_buff *skb) * in rdma connection migration (see c4iw_accept_cr()). */ __state_set(&ep->com, CLOSING); - PDBG("waking up ep %p tid %u\n", ep, ep->hwtid); + pr_debug("waking up ep %p tid %u\n", ep, ep->hwtid); c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET); break; case MPA_REP_SENT: __state_set(&ep->com, CLOSING); - PDBG("waking up ep %p tid %u\n", ep, ep->hwtid); + pr_debug("waking up ep %p tid %u\n", ep, ep->hwtid); c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET); break; case FPDU_MODE: @@ -2735,17 +2736,17 @@ static int peer_abort(struct c4iw_dev *dev, struct sk_buff *skb) return 0; if (cxgb_is_neg_adv(req->status)) { - PDBG("%s Negative advice on abort- tid %u status %d (%s)\n", - __func__, ep->hwtid, req->status, - neg_adv_str(req->status)); + pr_debug("%s Negative advice on abort- tid %u status %d (%s)\n", + __func__, ep->hwtid, req->status, + neg_adv_str(req->status)); ep->stats.abort_neg_adv++; mutex_lock(&dev->rdev.stats.lock); dev->rdev.stats.neg_adv++; mutex_unlock(&dev->rdev.stats.lock); goto deref_ep; } - PDBG("%s ep %p tid %u state %u\n", __func__, ep, ep->hwtid, - ep->com.state); + pr_debug("%s ep %p tid %u state %u\n", __func__, ep, ep->hwtid, + ep->com.state); set_bit(PEER_ABORT, &ep->com.history); /* @@ -2777,8 +2778,8 @@ static int peer_abort(struct c4iw_dev *dev, struct sk_buff *skb) * do some housekeeping so as to re-initiate the * connection */ - PDBG("%s: mpa_rev=%d. Retrying with mpav1\n", __func__, - mpa_rev); + pr_debug("%s: mpa_rev=%d. Retrying with mpav1\n", + __func__, mpa_rev); ep->retry_with_mpa_v1 = 1; } break; @@ -2797,16 +2798,14 @@ static int peer_abort(struct c4iw_dev *dev, struct sk_buff *skb) ep->com.qp, C4IW_QP_ATTR_NEXT_STATE, &attrs, 1); if (ret) - printk(KERN_ERR MOD - "%s - qp <- error failed!\n", - __func__); + pr_err("%s - qp <- error failed!\n", __func__); } peer_abort_upcall(ep); break; case ABORTING: break; case DEAD: - PDBG("%s PEER_ABORT IN DEAD STATE!!!!\n", __func__); + pr_debug("%s PEER_ABORT IN DEAD STATE!!!!\n", __func__); mutex_unlock(&ep->com.mutex); goto deref_ep; default: @@ -2870,7 +2869,7 @@ static int close_con_rpl(struct c4iw_dev *dev, struct sk_buff *skb) if (!ep) return 0; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); BUG_ON(!ep); /* The cm_id may be null if we failed to connect */ @@ -2918,13 +2917,13 @@ static int terminate(struct c4iw_dev *dev, struct sk_buff *skb) BUG_ON(!ep); if (ep && ep->com.qp) { - printk(KERN_WARNING MOD "TERM received tid %u qpid %u\n", tid, - ep->com.qp->wq.sq.qid); + pr_warn("TERM received tid %u qpid %u\n", + tid, ep->com.qp->wq.sq.qid); attrs.next_state = C4IW_QP_STATE_TERMINATE; c4iw_modify_qp(ep->com.qp->rhp, ep->com.qp, C4IW_QP_ATTR_NEXT_STATE, &attrs, 1); } else - printk(KERN_WARNING MOD "TERM received tid %u no ep/qp\n", tid); + pr_warn("TERM received tid %u no ep/qp\n", tid); c4iw_put_ep(&ep->com); return 0; @@ -2946,18 +2945,19 @@ static int fw4_ack(struct c4iw_dev *dev, struct sk_buff *skb) ep = get_ep_from_tid(dev, tid); if (!ep) return 0; - PDBG("%s ep %p tid %u credits %u\n", __func__, ep, ep->hwtid, credits); + pr_debug("%s ep %p tid %u credits %u\n", + __func__, ep, ep->hwtid, credits); if (credits == 0) { - PDBG("%s 0 credit ack ep %p tid %u state %u\n", - __func__, ep, ep->hwtid, state_read(&ep->com)); + pr_debug("%s 0 credit ack ep %p tid %u state %u\n", + __func__, ep, ep->hwtid, state_read(&ep->com)); goto out; } dst_confirm(ep->dst); if (ep->mpa_skb) { - PDBG("%s last streaming msg ack ep %p tid %u state %u " - "initiator %u freeing skb\n", __func__, ep, ep->hwtid, - state_read(&ep->com), ep->mpa_attr.initiator ? 1 : 0); + pr_debug("%s last streaming msg ack ep %p tid %u state %u initiator %u freeing skb\n", + __func__, ep, ep->hwtid, + state_read(&ep->com), ep->mpa_attr.initiator ? 1 : 0); mutex_lock(&ep->com.mutex); kfree_skb(ep->mpa_skb); ep->mpa_skb = NULL; @@ -2975,7 +2975,7 @@ int c4iw_reject_cr(struct iw_cm_id *cm_id, const void *pdata, u8 pdata_len) int abort; struct c4iw_ep *ep = to_ep(cm_id); - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); mutex_lock(&ep->com.mutex); if (ep->com.state != MPA_REQ_RCVD) { @@ -3006,7 +3006,7 @@ int c4iw_accept_cr(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) struct c4iw_qp *qp = get_qhp(h, conn_param->qpn); int abort = 0; - PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); + pr_debug("%s ep %p tid %u\n", __func__, ep, ep->hwtid); mutex_lock(&ep->com.mutex); if (ep->com.state != MPA_REQ_RCVD) { @@ -3059,7 +3059,7 @@ int c4iw_accept_cr(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) ep->ird = 1; } - PDBG("%s %d ird %d ord %d\n", __func__, __LINE__, ep->ird, ep->ord); + pr_debug("%s %d ird %d ord %d\n", __func__, __LINE__, ep->ird, ep->ord); ep->com.cm_id = cm_id; ref_cm_id(&ep->com); @@ -3188,7 +3188,7 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) } ep = alloc_ep(sizeof(*ep), GFP_KERNEL); if (!ep) { - printk(KERN_ERR MOD "%s - cannot alloc ep.\n", __func__); + pr_err("%s - cannot alloc ep\n", __func__); err = -ENOMEM; goto out; } @@ -3215,20 +3215,20 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) ep->com.dev = dev; ep->com.qp = get_qhp(dev, conn_param->qpn); if (!ep->com.qp) { - PDBG("%s qpn 0x%x not found!\n", __func__, conn_param->qpn); + pr_debug("%s qpn 0x%x not found!\n", __func__, conn_param->qpn); err = -EINVAL; goto fail2; } ref_qp(ep); - PDBG("%s qpn 0x%x qp %p cm_id %p\n", __func__, conn_param->qpn, - ep->com.qp, cm_id); + pr_debug("%s qpn 0x%x qp %p cm_id %p\n", __func__, conn_param->qpn, + ep->com.qp, cm_id); /* * Allocate an active TID to initiate a TCP connection. */ ep->atid = cxgb4_alloc_atid(dev->rdev.lldi.tids, ep); if (ep->atid == -1) { - printk(KERN_ERR MOD "%s - cannot alloc atid.\n", __func__); + pr_err("%s - cannot alloc atid\n", __func__); err = -ENOMEM; goto fail2; } @@ -3258,9 +3258,9 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) } /* find a route */ - PDBG("%s saddr %pI4 sport 0x%x raddr %pI4 rport 0x%x\n", - __func__, &laddr->sin_addr, ntohs(laddr->sin_port), - ra, ntohs(raddr->sin_port)); + pr_debug("%s saddr %pI4 sport 0x%x raddr %pI4 rport 0x%x\n", + __func__, &laddr->sin_addr, ntohs(laddr->sin_port), + ra, ntohs(raddr->sin_port)); ep->dst = cxgb_find_route(&dev->rdev.lldi, get_real_dev, laddr->sin_addr.s_addr, raddr->sin_addr.s_addr, @@ -3280,10 +3280,10 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) } /* find a route */ - PDBG("%s saddr %pI6 sport 0x%x raddr %pI6 rport 0x%x\n", - __func__, laddr6->sin6_addr.s6_addr, - ntohs(laddr6->sin6_port), - raddr6->sin6_addr.s6_addr, ntohs(raddr6->sin6_port)); + pr_debug("%s saddr %pI6 sport 0x%x raddr %pI6 rport 0x%x\n", + __func__, laddr6->sin6_addr.s6_addr, + ntohs(laddr6->sin6_port), + raddr6->sin6_addr.s6_addr, ntohs(raddr6->sin6_port)); ep->dst = cxgb_find_route6(&dev->rdev.lldi, get_real_dev, laddr6->sin6_addr.s6_addr, raddr6->sin6_addr.s6_addr, @@ -3292,7 +3292,7 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) raddr6->sin6_scope_id); } if (!ep->dst) { - printk(KERN_ERR MOD "%s - cannot find route.\n", __func__); + pr_err("%s - cannot find route\n", __func__); err = -EHOSTUNREACH; goto fail3; } @@ -3300,13 +3300,13 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) err = import_ep(ep, iptype, ra, ep->dst, ep->com.dev, true, ep->com.dev->rdev.lldi.adapter_type, cm_id->tos); if (err) { - printk(KERN_ERR MOD "%s - cannot alloc l2e.\n", __func__); + pr_err("%s - cannot alloc l2e\n", __func__); goto fail4; } - PDBG("%s txq_idx %u tx_chan %u smac_idx %u rss_qid %u l2t_idx %u\n", - __func__, ep->txq_idx, ep->tx_chan, ep->smac_idx, ep->rss_qid, - ep->l2t->idx); + pr_debug("%s txq_idx %u tx_chan %u smac_idx %u rss_qid %u l2t_idx %u\n", + __func__, ep->txq_idx, ep->tx_chan, ep->smac_idx, ep->rss_qid, + ep->l2t->idx); state_set(&ep->com, CONNECTING); ep->tos = cm_id->tos; @@ -3414,12 +3414,12 @@ int c4iw_create_listen(struct iw_cm_id *cm_id, int backlog) ep = alloc_ep(sizeof(*ep), GFP_KERNEL); if (!ep) { - printk(KERN_ERR MOD "%s - cannot alloc ep.\n", __func__); + pr_err("%s - cannot alloc ep\n", __func__); err = -ENOMEM; goto fail1; } skb_queue_head_init(&ep->com.ep_skb_list); - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); ep->com.cm_id = cm_id; ref_cm_id(&ep->com); ep->com.dev = dev; @@ -3439,7 +3439,7 @@ int c4iw_create_listen(struct iw_cm_id *cm_id, int backlog) cm_id->m_local_addr.ss_family, ep); if (ep->stid == -1) { - printk(KERN_ERR MOD "%s - cannot alloc stid.\n", __func__); + pr_err("%s - cannot alloc stid\n", __func__); err = -ENOMEM; goto fail2; } @@ -3473,7 +3473,7 @@ int c4iw_destroy_listen(struct iw_cm_id *cm_id) int err; struct c4iw_listen_ep *ep = to_listen_ep(cm_id); - PDBG("%s ep %p\n", __func__, ep); + pr_debug("%s ep %p\n", __func__, ep); might_sleep(); state_set(&ep->com, DEAD); @@ -3514,8 +3514,8 @@ int c4iw_ep_disconnect(struct c4iw_ep *ep, int abrupt, gfp_t gfp) mutex_lock(&ep->com.mutex); - PDBG("%s ep %p state %s, abrupt %d\n", __func__, ep, - states[ep->com.state], abrupt); + pr_debug("%s ep %p state %s, abrupt %d\n", __func__, ep, + states[ep->com.state], abrupt); /* * Ref the ep here in case we have fatal errors causing the @@ -3568,8 +3568,8 @@ int c4iw_ep_disconnect(struct c4iw_ep *ep, int abrupt, gfp_t gfp) case MORIBUND: case ABORTING: case DEAD: - PDBG("%s ignoring disconnect ep %p state %u\n", - __func__, ep, ep->com.state); + pr_debug("%s ignoring disconnect ep %p state %u\n", + __func__, ep, ep->com.state); break; default: BUG(); @@ -3600,8 +3600,7 @@ int c4iw_ep_disconnect(struct c4iw_ep *ep, int abrupt, gfp_t gfp) C4IW_QP_ATTR_NEXT_STATE, &attrs, 1); if (ret) - pr_err(MOD - "%s - qp <- error failed!\n", + pr_err("%s - qp <- error failed!\n", __func__); } fatal = 1; @@ -3674,7 +3673,7 @@ static void passive_ofld_conn_reply(struct c4iw_dev *dev, struct sk_buff *skb, rpl_skb = (struct sk_buff *)(unsigned long)req->cookie; BUG_ON(!rpl_skb); if (req->retval) { - PDBG("%s passive open failure %d\n", __func__, req->retval); + pr_debug("%s passive open failure %d\n", __func__, req->retval); mutex_lock(&dev->rdev.stats.lock); dev->rdev.stats.pas_ofld_conn_fails++; mutex_unlock(&dev->rdev.stats.lock); @@ -3800,6 +3799,8 @@ static void send_fw_pass_open_req(struct c4iw_dev *dev, struct sk_buff *skb, int ret; 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->op_compl = htonl(WR_OP_V(FW_OFLD_CONNECTION_WR) | FW_WR_COMPL_F); @@ -3890,7 +3891,8 @@ static int rx_pkt(struct c4iw_dev *dev, struct sk_buff *skb) lep = (struct c4iw_ep *)get_ep_from_stid(dev, stid); if (!lep) { - PDBG("%s connect request on invalid stid %d\n", __func__, stid); + pr_debug("%s connect request on invalid stid %d\n", + __func__, stid); goto reject; } @@ -3927,9 +3929,9 @@ static int rx_pkt(struct c4iw_dev *dev, struct sk_buff *skb) skb_set_transport_header(skb, (void *)tcph - (void *)rss); skb_get(skb); - PDBG("%s lip 0x%x lport %u pip 0x%x pport %u tos %d\n", __func__, - ntohl(iph->daddr), ntohs(tcph->dest), ntohl(iph->saddr), - ntohs(tcph->source), iph->tos); + pr_debug("%s lip 0x%x lport %u pip 0x%x pport %u tos %d\n", __func__, + ntohl(iph->daddr), ntohs(tcph->dest), ntohl(iph->saddr), + ntohs(tcph->source), iph->tos); dst = cxgb_find_route(&dev->rdev.lldi, get_real_dev, iph->daddr, iph->saddr, tcph->dest, @@ -4026,8 +4028,8 @@ static void process_timeout(struct c4iw_ep *ep) int abort = 1; mutex_lock(&ep->com.mutex); - PDBG("%s ep %p tid %u state %d\n", __func__, ep, ep->hwtid, - ep->com.state); + pr_debug("%s ep %p tid %u state %d\n", __func__, ep, ep->hwtid, + ep->com.state); set_bit(TIMEDOUT, &ep->com.history); switch (ep->com.state) { case MPA_REQ_SENT: @@ -4157,8 +4159,8 @@ static int set_tcb_rpl(struct c4iw_dev *dev, struct sk_buff *skb) struct cpl_set_tcb_rpl *rpl = cplhdr(skb); if (rpl->status != CPL_ERR_NONE) { - printk(KERN_ERR MOD "Unexpected SET_TCB_RPL status %u " - "for tid %u\n", rpl->status, GET_TID(rpl)); + pr_err("Unexpected SET_TCB_RPL status %u for tid %u\n", + rpl->status, GET_TID(rpl)); } kfree_skb(skb); return 0; @@ -4170,13 +4172,13 @@ static int fw6_msg(struct c4iw_dev *dev, struct sk_buff *skb) struct c4iw_wr_wait *wr_waitp; int ret; - PDBG("%s type %u\n", __func__, rpl->type); + pr_debug("%s type %u\n", __func__, rpl->type); switch (rpl->type) { case FW6_TYPE_WR_RPL: ret = (int)((be64_to_cpu(rpl->data[0]) >> 8) & 0xff); wr_waitp = (struct c4iw_wr_wait *)(__force unsigned long) rpl->data[1]; - PDBG("%s wr_waitp %p ret %u\n", __func__, wr_waitp, ret); + pr_debug("%s wr_waitp %p ret %u\n", __func__, wr_waitp, ret); if (wr_waitp) c4iw_wake_up(wr_waitp, ret ? -ret : 0); kfree_skb(skb); @@ -4186,8 +4188,8 @@ static int fw6_msg(struct c4iw_dev *dev, struct sk_buff *skb) sched(dev, skb); break; default: - printk(KERN_ERR MOD "%s unexpected fw6 msg type %u\n", __func__, - rpl->type); + pr_err("%s unexpected fw6 msg type %u\n", + __func__, rpl->type); kfree_skb(skb); break; } @@ -4203,19 +4205,18 @@ static int peer_abort_intr(struct c4iw_dev *dev, struct sk_buff *skb) ep = get_ep_from_tid(dev, tid); /* This EP will be dereferenced in peer_abort() */ if (!ep) { - printk(KERN_WARNING MOD - "Abort on non-existent endpoint, tid %d\n", tid); + pr_warn("Abort on non-existent endpoint, tid %d\n", tid); kfree_skb(skb); return 0; } if (cxgb_is_neg_adv(req->status)) { - PDBG("%s Negative advice on abort- tid %u status %d (%s)\n", - __func__, ep->hwtid, req->status, - neg_adv_str(req->status)); + pr_debug("%s Negative advice on abort- tid %u status %d (%s)\n", + __func__, ep->hwtid, req->status, + neg_adv_str(req->status)); goto out; } - PDBG("%s ep %p tid %u state %u\n", __func__, ep, ep->hwtid, - ep->com.state); + pr_debug("%s ep %p tid %u state %u\n", __func__, ep, ep->hwtid, + ep->com.state); c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET); out: diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c index bec82a600d77..14de5bde1b63 100644 --- a/drivers/infiniband/hw/cxgb4/cq.c +++ b/drivers/infiniband/hw/cxgb4/cq.c @@ -146,7 +146,7 @@ static int create_cq(struct c4iw_rdev *rdev, struct t4_cq *cq, ret = c4iw_ofld_send(rdev, skb); if (ret) goto err4; - PDBG("%s wait_event wr_wait %p\n", __func__, &wr_wait); + pr_debug("%s wait_event wr_wait %p\n", __func__, &wr_wait); ret = c4iw_wait_for_reply(rdev, &wr_wait, 0, 0, __func__); if (ret) goto err4; @@ -159,7 +159,7 @@ static int create_cq(struct c4iw_rdev *rdev, struct t4_cq *cq, &cq->bar2_qid, user ? &cq->bar2_pa : NULL); if (user && !cq->bar2_pa) { - pr_warn(MOD "%s: cqid %u not in BAR2 range.\n", + pr_warn("%s: cqid %u not in BAR2 range\n", pci_name(rdev->lldi.pdev), cq->cqid); ret = -EINVAL; goto err4; @@ -180,8 +180,8 @@ static void insert_recv_cqe(struct t4_wq *wq, struct t4_cq *cq) { struct t4_cqe cqe; - PDBG("%s wq %p cq %p sw_cidx %u sw_pidx %u\n", __func__, - wq, cq, cq->sw_cidx, cq->sw_pidx); + pr_debug("%s wq %p cq %p sw_cidx %u sw_pidx %u\n", __func__, + wq, cq, cq->sw_cidx, cq->sw_pidx); memset(&cqe, 0, sizeof(cqe)); cqe.header = cpu_to_be32(CQE_STATUS_V(T4_ERR_SWFLUSH) | CQE_OPCODE_V(FW_RI_SEND) | @@ -199,8 +199,8 @@ int c4iw_flush_rq(struct t4_wq *wq, struct t4_cq *cq, int count) int in_use = wq->rq.in_use - count; BUG_ON(in_use < 0); - PDBG("%s wq %p cq %p rq.in_use %u skip count %u\n", __func__, - wq, cq, wq->rq.in_use, count); + pr_debug("%s wq %p cq %p rq.in_use %u skip count %u\n", __func__, + wq, cq, wq->rq.in_use, count); while (in_use--) { insert_recv_cqe(wq, cq); flushed++; @@ -213,8 +213,8 @@ static void insert_sq_cqe(struct t4_wq *wq, struct t4_cq *cq, { struct t4_cqe cqe; - PDBG("%s wq %p cq %p sw_cidx %u sw_pidx %u\n", __func__, - wq, cq, cq->sw_cidx, cq->sw_pidx); + pr_debug("%s wq %p cq %p sw_cidx %u sw_pidx %u\n", __func__, + wq, cq, cq->sw_cidx, cq->sw_pidx); memset(&cqe, 0, sizeof(cqe)); cqe.header = cpu_to_be32(CQE_STATUS_V(T4_ERR_SWFLUSH) | CQE_OPCODE_V(swcqe->opcode) | @@ -283,8 +283,8 @@ static void flush_completed_wrs(struct t4_wq *wq, struct t4_cq *cq) /* * Insert this completed cqe into the swcq. */ - PDBG("%s moving cqe into swcq sq idx %u cq idx %u\n", - __func__, cidx, cq->sw_pidx); + pr_debug("%s moving cqe into swcq sq idx %u cq idx %u\n", + __func__, cidx, cq->sw_pidx); swsqe->cqe.header |= htonl(CQE_SWCQE_V(1)); cq->sw_queue[cq->sw_pidx] = swsqe->cqe; t4_swcq_produce(cq); @@ -339,7 +339,7 @@ void c4iw_flush_hw_cq(struct c4iw_cq *chp) struct t4_swsqe *swsqe; int ret; - PDBG("%s cqid 0x%x\n", __func__, chp->cq.cqid); + pr_debug("%s cqid 0x%x\n", __func__, chp->cq.cqid); ret = t4_next_hw_cqe(&chp->cq, &hw_cqe); /* @@ -432,7 +432,7 @@ void c4iw_count_rcqes(struct t4_cq *cq, struct t4_wq *wq, int *count) u32 ptr; *count = 0; - PDBG("%s count zero %d\n", __func__, *count); + pr_debug("%s count zero %d\n", __func__, *count); ptr = cq->sw_cidx; while (ptr != cq->sw_pidx) { cqe = &cq->sw_queue[ptr]; @@ -442,7 +442,7 @@ void c4iw_count_rcqes(struct t4_cq *cq, struct t4_wq *wq, int *count) if (++ptr == cq->size) ptr = 0; } - PDBG("%s cq %p count %d\n", __func__, cq, *count); + pr_debug("%s cq %p count %d\n", __func__, cq, *count); } /* @@ -473,12 +473,11 @@ static int poll_cq(struct t4_wq *wq, struct t4_cq *cq, struct t4_cqe *cqe, if (ret) return ret; - PDBG("%s CQE OVF %u qpid 0x%0x genbit %u type %u status 0x%0x" - " opcode 0x%0x len 0x%0x wrid_hi_stag 0x%x wrid_low_msn 0x%x\n", - __func__, CQE_OVFBIT(hw_cqe), CQE_QPID(hw_cqe), - CQE_GENBIT(hw_cqe), CQE_TYPE(hw_cqe), CQE_STATUS(hw_cqe), - CQE_OPCODE(hw_cqe), CQE_LEN(hw_cqe), CQE_WRID_HI(hw_cqe), - CQE_WRID_LOW(hw_cqe)); + pr_debug("%s CQE OVF %u qpid 0x%0x genbit %u type %u status 0x%0x opcode 0x%0x len 0x%0x wrid_hi_stag 0x%x wrid_low_msn 0x%x\n", + __func__, CQE_OVFBIT(hw_cqe), CQE_QPID(hw_cqe), + CQE_GENBIT(hw_cqe), CQE_TYPE(hw_cqe), CQE_STATUS(hw_cqe), + CQE_OPCODE(hw_cqe), CQE_LEN(hw_cqe), CQE_WRID_HI(hw_cqe), + CQE_WRID_LOW(hw_cqe)); /* * skip cqe's not affiliated with a QP. @@ -606,8 +605,8 @@ static int poll_cq(struct t4_wq *wq, struct t4_cq *cq, struct t4_cqe *cqe, if (!SW_CQE(hw_cqe) && (CQE_WRID_SQ_IDX(hw_cqe) != wq->sq.cidx)) { struct t4_swsqe *swsqe; - PDBG("%s out of order completion going in sw_sq at idx %u\n", - __func__, CQE_WRID_SQ_IDX(hw_cqe)); + pr_debug("%s out of order completion going in sw_sq at idx %u\n", + __func__, CQE_WRID_SQ_IDX(hw_cqe)); swsqe = &wq->sq.sw_sq[CQE_WRID_SQ_IDX(hw_cqe)]; swsqe->cqe = *hw_cqe; swsqe->complete = 1; @@ -641,13 +640,13 @@ proc_cqe: BUG_ON(wq->sq.in_use <= 0 && wq->sq.in_use >= wq->sq.size); wq->sq.cidx = (uint16_t)idx; - PDBG("%s completing sq idx %u\n", __func__, wq->sq.cidx); + pr_debug("%s completing sq idx %u\n", __func__, wq->sq.cidx); *cookie = wq->sq.sw_sq[wq->sq.cidx].wr_id; if (c4iw_wr_log) c4iw_log_wr_stats(wq, hw_cqe); t4_sq_consume(wq); } else { - PDBG("%s completing rq idx %u\n", __func__, wq->rq.cidx); + pr_debug("%s completing rq idx %u\n", __func__, wq->rq.cidx); *cookie = wq->rq.sw_rq[wq->rq.cidx].wr_id; BUG_ON(t4_rq_empty(wq)); if (c4iw_wr_log) @@ -664,12 +663,12 @@ flush_wq: skip_cqe: if (SW_CQE(hw_cqe)) { - PDBG("%s cq %p cqid 0x%x skip sw cqe cidx %u\n", - __func__, cq, cq->cqid, cq->sw_cidx); + pr_debug("%s cq %p cqid 0x%x skip sw cqe cidx %u\n", + __func__, cq, cq->cqid, cq->sw_cidx); t4_swcq_consume(cq); } else { - PDBG("%s cq %p cqid 0x%x skip hw cqe cidx %u\n", - __func__, cq, cq->cqid, cq->cidx); + pr_debug("%s cq %p cqid 0x%x skip hw cqe cidx %u\n", + __func__, cq, cq->cqid, cq->cidx); t4_hwcq_consume(cq); } return ret; @@ -715,10 +714,12 @@ static int c4iw_poll_cq_one(struct c4iw_cq *chp, struct ib_wc *wc) wc->vendor_err = CQE_STATUS(&cqe); wc->wc_flags = 0; - PDBG("%s qpid 0x%x type %d opcode %d status 0x%x len %u wrid hi 0x%x " - "lo 0x%x cookie 0x%llx\n", __func__, CQE_QPID(&cqe), - CQE_TYPE(&cqe), CQE_OPCODE(&cqe), CQE_STATUS(&cqe), CQE_LEN(&cqe), - CQE_WRID_HI(&cqe), CQE_WRID_LOW(&cqe), (unsigned long long)cookie); + pr_debug("%s qpid 0x%x type %d opcode %d status 0x%x len %u wrid hi 0x%x lo 0x%x cookie 0x%llx\n", + __func__, CQE_QPID(&cqe), + CQE_TYPE(&cqe), CQE_OPCODE(&cqe), + CQE_STATUS(&cqe), CQE_LEN(&cqe), + CQE_WRID_HI(&cqe), CQE_WRID_LOW(&cqe), + (unsigned long long)cookie); if (CQE_TYPE(&cqe) == 0) { if (!CQE_STATUS(&cqe)) @@ -766,8 +767,7 @@ static int c4iw_poll_cq_one(struct c4iw_cq *chp, struct ib_wc *wc) wc->opcode = IB_WC_SEND; break; default: - printk(KERN_ERR MOD "Unexpected opcode %d " - "in the CQE received for QPID=0x%0x\n", + pr_err("Unexpected opcode %d in the CQE received for QPID=0x%0x\n", CQE_OPCODE(&cqe), CQE_QPID(&cqe)); ret = -EINVAL; goto out; @@ -822,8 +822,7 @@ static int c4iw_poll_cq_one(struct c4iw_cq *chp, struct ib_wc *wc) wc->status = IB_WC_WR_FLUSH_ERR; break; default: - printk(KERN_ERR MOD - "Unexpected cqe_status 0x%x for QPID=0x%0x\n", + pr_err("Unexpected cqe_status 0x%x for QPID=0x%0x\n", CQE_STATUS(&cqe), CQE_QPID(&cqe)); wc->status = IB_WC_FATAL_ERR; } @@ -860,7 +859,7 @@ int c4iw_destroy_cq(struct ib_cq *ib_cq) struct c4iw_cq *chp; struct c4iw_ucontext *ucontext; - PDBG("%s ib_cq %p\n", __func__, ib_cq); + pr_debug("%s ib_cq %p\n", __func__, ib_cq); chp = to_c4iw_cq(ib_cq); remove_handle(chp->rhp, &chp->rhp->cqidr, chp->cq.cqid); @@ -892,7 +891,7 @@ struct ib_cq *c4iw_create_cq(struct ib_device *ibdev, size_t memsize, hwentries; struct c4iw_mm_entry *mm, *mm2; - PDBG("%s ib_dev %p entries %d\n", __func__, ibdev, entries); + pr_debug("%s ib_dev %p entries %d\n", __func__, ibdev, entries); if (attr->flags) return ERR_PTR(-EINVAL); @@ -998,9 +997,9 @@ struct ib_cq *c4iw_create_cq(struct ib_device *ibdev, mm2->len = PAGE_SIZE; insert_mmap(ucontext, mm2); } - PDBG("%s cqid 0x%0x chp %p size %u memsize %zu, dma_addr 0x%0llx\n", - __func__, chp->cq.cqid, chp, chp->cq.size, - chp->cq.memsize, (unsigned long long) chp->cq.dma_addr); + pr_debug("%s cqid 0x%0x chp %p size %u memsize %zu, dma_addr 0x%0llx\n", + __func__, chp->cq.cqid, chp, chp->cq.size, + chp->cq.memsize, (unsigned long long)chp->cq.dma_addr); return &chp->ibcq; err6: kfree(mm2); diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c index 4e4f1a732b01..329fb65e8fb0 100644 --- a/drivers/infiniband/hw/cxgb4/device.c +++ b/drivers/infiniband/hw/cxgb4/device.c @@ -334,7 +334,7 @@ static int qp_release(struct inode *inode, struct file *file) { struct c4iw_debugfs_data *qpd = file->private_data; if (!qpd) { - printk(KERN_INFO "%s null qpd?\n", __func__); + pr_info("%s null qpd?\n", __func__); return 0; } vfree(qpd->buf); @@ -422,7 +422,7 @@ static int stag_release(struct inode *inode, struct file *file) { struct c4iw_debugfs_data *stagd = file->private_data; if (!stagd) { - printk(KERN_INFO "%s null stagd?\n", __func__); + pr_info("%s null stagd?\n", __func__); return 0; } vfree(stagd->buf); @@ -796,15 +796,14 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev) * cqid and qpid range must match for now. */ if (rdev->lldi.udb_density != rdev->lldi.ucq_density) { - pr_err(MOD "%s: unsupported udb/ucq densities %u/%u\n", + pr_err("%s: unsupported udb/ucq densities %u/%u\n", pci_name(rdev->lldi.pdev), rdev->lldi.udb_density, rdev->lldi.ucq_density); return -EINVAL; } if (rdev->lldi.vr->qp.start != rdev->lldi.vr->cq.start || rdev->lldi.vr->qp.size != rdev->lldi.vr->cq.size) { - pr_err(MOD "%s: unsupported qp and cq id ranges " - "qp start %u size %u cq start %u size %u\n", + pr_err("%s: unsupported qp and cq id ranges qp start %u size %u cq start %u size %u\n", pci_name(rdev->lldi.pdev), rdev->lldi.vr->qp.start, rdev->lldi.vr->qp.size, rdev->lldi.vr->cq.size, rdev->lldi.vr->cq.size); @@ -813,23 +812,20 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev) rdev->qpmask = rdev->lldi.udb_density - 1; rdev->cqmask = rdev->lldi.ucq_density - 1; - PDBG("%s dev %s stag start 0x%0x size 0x%0x num stags %d " - "pbl start 0x%0x size 0x%0x rq start 0x%0x size 0x%0x " - "qp qid start %u size %u cq qid start %u size %u\n", - __func__, pci_name(rdev->lldi.pdev), rdev->lldi.vr->stag.start, - rdev->lldi.vr->stag.size, c4iw_num_stags(rdev), - rdev->lldi.vr->pbl.start, - rdev->lldi.vr->pbl.size, rdev->lldi.vr->rq.start, - rdev->lldi.vr->rq.size, - rdev->lldi.vr->qp.start, - rdev->lldi.vr->qp.size, - rdev->lldi.vr->cq.start, - rdev->lldi.vr->cq.size); - PDBG("udb %pR db_reg %p gts_reg %p " - "qpmask 0x%x cqmask 0x%x\n", - &rdev->lldi.pdev->resource[2], - rdev->lldi.db_reg, rdev->lldi.gts_reg, - rdev->qpmask, rdev->cqmask); + pr_debug("%s dev %s stag start 0x%0x size 0x%0x num stags %d pbl start 0x%0x size 0x%0x rq start 0x%0x size 0x%0x qp qid start %u size %u cq qid start %u size %u\n", + __func__, pci_name(rdev->lldi.pdev), rdev->lldi.vr->stag.start, + rdev->lldi.vr->stag.size, c4iw_num_stags(rdev), + rdev->lldi.vr->pbl.start, + rdev->lldi.vr->pbl.size, rdev->lldi.vr->rq.start, + rdev->lldi.vr->rq.size, + rdev->lldi.vr->qp.start, + rdev->lldi.vr->qp.size, + rdev->lldi.vr->cq.start, + rdev->lldi.vr->cq.size); + pr_debug("udb %pR db_reg %p gts_reg %p qpmask 0x%x cqmask 0x%x\n", + &rdev->lldi.pdev->resource[2], + rdev->lldi.db_reg, rdev->lldi.gts_reg, + rdev->qpmask, rdev->cqmask); if (c4iw_num_stags(rdev) == 0) return -EINVAL; @@ -843,22 +839,22 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev) err = c4iw_init_resource(rdev, c4iw_num_stags(rdev), T4_MAX_NUM_PD); if (err) { - printk(KERN_ERR MOD "error %d initializing resources\n", err); + pr_err("error %d initializing resources\n", err); return err; } err = c4iw_pblpool_create(rdev); if (err) { - printk(KERN_ERR MOD "error %d initializing pbl pool\n", err); + pr_err("error %d initializing pbl pool\n", err); goto destroy_resource; } err = c4iw_rqtpool_create(rdev); if (err) { - printk(KERN_ERR MOD "error %d initializing rqt pool\n", err); + pr_err("error %d initializing rqt pool\n", err); goto destroy_pblpool; } err = c4iw_ocqp_pool_create(rdev); if (err) { - printk(KERN_ERR MOD "error %d initializing ocqp pool\n", err); + pr_err("error %d initializing ocqp pool\n", err); goto destroy_rqtpool; } rdev->status_page = (struct t4_dev_status_page *) @@ -936,7 +932,7 @@ static void c4iw_dealloc(struct uld_ctx *ctx) static void c4iw_remove(struct uld_ctx *ctx) { - PDBG("%s c4iw_dev %p\n", __func__, ctx->dev); + pr_debug("%s c4iw_dev %p\n", __func__, ctx->dev); c4iw_unregister_device(ctx->dev); c4iw_dealloc(ctx); } @@ -954,25 +950,25 @@ static struct c4iw_dev *c4iw_alloc(const struct cxgb4_lld_info *infop) int ret; if (!rdma_supported(infop)) { - printk(KERN_INFO MOD "%s: RDMA not supported on this device.\n", - pci_name(infop->pdev)); + pr_info("%s: RDMA not supported on this device\n", + pci_name(infop->pdev)); return ERR_PTR(-ENOSYS); } if (!ocqp_supported(infop)) - pr_info("%s: On-Chip Queues not supported on this device.\n", + pr_info("%s: On-Chip Queues not supported on this device\n", pci_name(infop->pdev)); devp = (struct c4iw_dev *)ib_alloc_device(sizeof(*devp)); if (!devp) { - printk(KERN_ERR MOD "Cannot allocate ib device\n"); + pr_err("Cannot allocate ib device\n"); return ERR_PTR(-ENOMEM); } devp->rdev.lldi = *infop; /* init various hw-queue params based on lld info */ - PDBG("%s: Ing. padding boundary is %d, egrsstatuspagesize = %d\n", - __func__, devp->rdev.lldi.sge_ingpadboundary, - devp->rdev.lldi.sge_egrstatuspagesize); + pr_debug("%s: Ing. padding boundary is %d, egrsstatuspagesize = %d\n", + __func__, devp->rdev.lldi.sge_ingpadboundary, + devp->rdev.lldi.sge_egrstatuspagesize); devp->rdev.hw_queue.t4_eq_status_entries = devp->rdev.lldi.sge_ingpadboundary > 64 ? 2 : 1; @@ -1000,7 +996,7 @@ static struct c4iw_dev *c4iw_alloc(const struct cxgb4_lld_info *infop) devp->rdev.bar2_kva = ioremap_wc(devp->rdev.bar2_pa, pci_resource_len(devp->rdev.lldi.pdev, 2)); if (!devp->rdev.bar2_kva) { - pr_err(MOD "Unable to ioremap BAR2\n"); + pr_err("Unable to ioremap BAR2\n"); ib_dealloc_device(&devp->ibdev); return ERR_PTR(-EINVAL); } @@ -1012,20 +1008,19 @@ static struct c4iw_dev *c4iw_alloc(const struct cxgb4_lld_info *infop) devp->rdev.oc_mw_kva = ioremap_wc(devp->rdev.oc_mw_pa, devp->rdev.lldi.vr->ocq.size); if (!devp->rdev.oc_mw_kva) { - pr_err(MOD "Unable to ioremap onchip mem\n"); + pr_err("Unable to ioremap onchip mem\n"); ib_dealloc_device(&devp->ibdev); return ERR_PTR(-EINVAL); } } - PDBG(KERN_INFO MOD "ocq memory: " - "hw_start 0x%x size %u mw_pa 0x%lx mw_kva %p\n", - devp->rdev.lldi.vr->ocq.start, devp->rdev.lldi.vr->ocq.size, - devp->rdev.oc_mw_pa, devp->rdev.oc_mw_kva); + pr_debug("ocq memory: hw_start 0x%x size %u mw_pa 0x%lx mw_kva %p\n", + devp->rdev.lldi.vr->ocq.start, devp->rdev.lldi.vr->ocq.size, + devp->rdev.oc_mw_pa, devp->rdev.oc_mw_kva); ret = c4iw_rdev_open(&devp->rdev); if (ret) { - printk(KERN_ERR MOD "Unable to open CXIO rdev err %d\n", ret); + pr_err("Unable to open CXIO rdev err %d\n", ret); ib_dealloc_device(&devp->ibdev); return ERR_PTR(ret); } @@ -1071,17 +1066,17 @@ static void *c4iw_uld_add(const struct cxgb4_lld_info *infop) } ctx->lldi = *infop; - PDBG("%s found device %s nchan %u nrxq %u ntxq %u nports %u\n", - __func__, pci_name(ctx->lldi.pdev), - ctx->lldi.nchan, ctx->lldi.nrxq, - ctx->lldi.ntxq, ctx->lldi.nports); + pr_debug("%s found device %s nchan %u nrxq %u ntxq %u nports %u\n", + __func__, pci_name(ctx->lldi.pdev), + ctx->lldi.nchan, ctx->lldi.nrxq, + ctx->lldi.ntxq, ctx->lldi.nports); mutex_lock(&dev_mutex); list_add_tail(&ctx->entry, &uld_ctx_list); mutex_unlock(&dev_mutex); for (i = 0; i < ctx->lldi.nrxq; i++) - PDBG("rxqid[%u] %u\n", i, ctx->lldi.rxq_ids[i]); + pr_debug("rxqid[%u] %u\n", i, ctx->lldi.rxq_ids[i]); out: return ctx; } @@ -1138,8 +1133,7 @@ static inline int recv_rx_pkt(struct c4iw_dev *dev, const struct pkt_gl *gl, goto out; if (c4iw_handlers[opcode] == NULL) { - pr_info("%s no handler opcode 0x%x...\n", __func__, - opcode); + pr_info("%s no handler opcode 0x%x...\n", __func__, opcode); kfree_skb(skb); goto out; } @@ -1176,13 +1170,11 @@ static int c4iw_uld_rx_handler(void *handle, const __be64 *rsp, if (recv_rx_pkt(dev, gl, rsp)) return 0; - pr_info("%s: unexpected FL contents at %p, " \ - "RSS %#llx, FL %#llx, len %u\n", - pci_name(ctx->lldi.pdev), gl->va, - (unsigned long long)be64_to_cpu(*rsp), - (unsigned long long)be64_to_cpu( - *(__force __be64 *)gl->va), - gl->tot_len); + pr_info("%s: unexpected FL contents at %p, RSS %#llx, FL %#llx, len %u\n", + pci_name(ctx->lldi.pdev), gl->va, + be64_to_cpu(*rsp), + be64_to_cpu(*(__force __be64 *)gl->va), + gl->tot_len); return 0; } else { @@ -1195,8 +1187,7 @@ static int c4iw_uld_rx_handler(void *handle, const __be64 *rsp, if (c4iw_handlers[opcode]) { c4iw_handlers[opcode](dev, skb); } else { - pr_info("%s no handler opcode 0x%x...\n", __func__, - opcode); + pr_info("%s no handler opcode 0x%x...\n", __func__, opcode); kfree_skb(skb); } @@ -1209,17 +1200,16 @@ static int c4iw_uld_state_change(void *handle, enum cxgb4_state new_state) { struct uld_ctx *ctx = handle; - PDBG("%s new_state %u\n", __func__, new_state); + pr_debug("%s new_state %u\n", __func__, new_state); switch (new_state) { case CXGB4_STATE_UP: - printk(KERN_INFO MOD "%s: Up\n", pci_name(ctx->lldi.pdev)); + pr_info("%s: Up\n", pci_name(ctx->lldi.pdev)); if (!ctx->dev) { int ret; ctx->dev = c4iw_alloc(&ctx->lldi); if (IS_ERR(ctx->dev)) { - printk(KERN_ERR MOD - "%s: initialization failed: %ld\n", + pr_err("%s: initialization failed: %ld\n", pci_name(ctx->lldi.pdev), PTR_ERR(ctx->dev)); ctx->dev = NULL; @@ -1227,22 +1217,19 @@ static int c4iw_uld_state_change(void *handle, enum cxgb4_state new_state) } ret = c4iw_register_device(ctx->dev); if (ret) { - printk(KERN_ERR MOD - "%s: RDMA registration failed: %d\n", + pr_err("%s: RDMA registration failed: %d\n", pci_name(ctx->lldi.pdev), ret); c4iw_dealloc(ctx); } } break; case CXGB4_STATE_DOWN: - printk(KERN_INFO MOD "%s: Down\n", - pci_name(ctx->lldi.pdev)); + pr_info("%s: Down\n", pci_name(ctx->lldi.pdev)); if (ctx->dev) c4iw_remove(ctx); break; case CXGB4_STATE_START_RECOVERY: - printk(KERN_INFO MOD "%s: Fatal Error\n", - pci_name(ctx->lldi.pdev)); + pr_info("%s: Fatal Error\n", pci_name(ctx->lldi.pdev)); if (ctx->dev) { struct ib_event event; @@ -1255,8 +1242,7 @@ static int c4iw_uld_state_change(void *handle, enum cxgb4_state new_state) } break; case CXGB4_STATE_DETACH: - printk(KERN_INFO MOD "%s: Detach\n", - pci_name(ctx->lldi.pdev)); + pr_info("%s: Detach\n", pci_name(ctx->lldi.pdev)); if (ctx->dev) c4iw_remove(ctx); break; @@ -1406,9 +1392,7 @@ static void recover_lost_dbs(struct uld_ctx *ctx, struct qp_list *qp_list) t4_sq_host_wq_pidx(&qp->wq), t4_sq_wq_size(&qp->wq)); if (ret) { - pr_err(MOD "%s: Fatal error - " - "DB overflow recovery failed - " - "error syncing SQ qid %u\n", + pr_err("%s: Fatal error - DB overflow recovery failed - error syncing SQ qid %u\n", pci_name(ctx->lldi.pdev), qp->wq.sq.qid); spin_unlock(&qp->lock); spin_unlock_irq(&qp->rhp->lock); @@ -1422,9 +1406,7 @@ static void recover_lost_dbs(struct uld_ctx *ctx, struct qp_list *qp_list) t4_rq_wq_size(&qp->wq)); if (ret) { - pr_err(MOD "%s: Fatal error - " - "DB overflow recovery failed - " - "error syncing RQ qid %u\n", + pr_err("%s: Fatal error - DB overflow recovery failed - error syncing RQ qid %u\n", pci_name(ctx->lldi.pdev), qp->wq.rq.qid); spin_unlock(&qp->lock); spin_unlock_irq(&qp->rhp->lock); @@ -1455,7 +1437,7 @@ static void recover_queues(struct uld_ctx *ctx) /* flush the SGE contexts */ ret = cxgb4_flush_eq_cache(ctx->dev->rdev.lldi.ports[0]); if (ret) { - printk(KERN_ERR MOD "%s: Fatal error - DB overflow recovery failed\n", + pr_err("%s: Fatal error - DB overflow recovery failed\n", pci_name(ctx->lldi.pdev)); return; } @@ -1513,8 +1495,8 @@ static int c4iw_uld_control(void *handle, enum cxgb4_control control, ...) mutex_unlock(&ctx->dev->rdev.stats.lock); break; default: - printk(KERN_WARNING MOD "%s: unknown control cmd %u\n", - pci_name(ctx->lldi.pdev), control); + pr_warn("%s: unknown control cmd %u\n", + pci_name(ctx->lldi.pdev), control); break; } return 0; @@ -1543,8 +1525,7 @@ static int __init c4iw_init_module(void) c4iw_debugfs_root = debugfs_create_dir(DRV_NAME, NULL); if (!c4iw_debugfs_root) - printk(KERN_WARNING MOD - "could not create debugfs entry, continuing\n"); + pr_warn("could not create debugfs entry, continuing\n"); cxgb4_register_uld(CXGB4_ULD_RDMA, &c4iw_uld_info); diff --git a/drivers/infiniband/hw/cxgb4/ev.c b/drivers/infiniband/hw/cxgb4/ev.c index bdfac2ccb704..8f963df0bffc 100644 --- a/drivers/infiniband/hw/cxgb4/ev.c +++ b/drivers/infiniband/hw/cxgb4/ev.c @@ -47,17 +47,16 @@ static void print_tpte(struct c4iw_dev *dev, u32 stag) "%s cxgb4_read_tpte err %d\n", __func__, ret); return; } - PDBG("stag idx 0x%x valid %d key 0x%x state %d pdid %d " - "perm 0x%x ps %d len 0x%llx va 0x%llx\n", - stag & 0xffffff00, - FW_RI_TPTE_VALID_G(ntohl(tpte.valid_to_pdid)), - FW_RI_TPTE_STAGKEY_G(ntohl(tpte.valid_to_pdid)), - FW_RI_TPTE_STAGSTATE_G(ntohl(tpte.valid_to_pdid)), - FW_RI_TPTE_PDID_G(ntohl(tpte.valid_to_pdid)), - FW_RI_TPTE_PERM_G(ntohl(tpte.locread_to_qpid)), - FW_RI_TPTE_PS_G(ntohl(tpte.locread_to_qpid)), - ((u64)ntohl(tpte.len_hi) << 32) | ntohl(tpte.len_lo), - ((u64)ntohl(tpte.va_hi) << 32) | ntohl(tpte.va_lo_fbo)); + pr_debug("stag idx 0x%x valid %d key 0x%x state %d pdid %d perm 0x%x ps %d len 0x%llx va 0x%llx\n", + stag & 0xffffff00, + FW_RI_TPTE_VALID_G(ntohl(tpte.valid_to_pdid)), + FW_RI_TPTE_STAGKEY_G(ntohl(tpte.valid_to_pdid)), + FW_RI_TPTE_STAGSTATE_G(ntohl(tpte.valid_to_pdid)), + FW_RI_TPTE_PDID_G(ntohl(tpte.valid_to_pdid)), + FW_RI_TPTE_PERM_G(ntohl(tpte.locread_to_qpid)), + FW_RI_TPTE_PS_G(ntohl(tpte.locread_to_qpid)), + ((u64)ntohl(tpte.len_hi) << 32) | ntohl(tpte.len_lo), + ((u64)ntohl(tpte.va_hi) << 32) | ntohl(tpte.va_lo_fbo)); } static void dump_err_cqe(struct c4iw_dev *dev, struct t4_cqe *err_cqe) @@ -71,9 +70,9 @@ static void dump_err_cqe(struct c4iw_dev *dev, struct t4_cqe *err_cqe) CQE_STATUS(err_cqe), CQE_TYPE(err_cqe), ntohl(err_cqe->len), CQE_WRID_HI(err_cqe), CQE_WRID_LOW(err_cqe)); - PDBG("%016llx %016llx %016llx %016llx\n", - be64_to_cpu(p[0]), be64_to_cpu(p[1]), be64_to_cpu(p[2]), - be64_to_cpu(p[3])); + pr_debug("%016llx %016llx %016llx %016llx\n", + be64_to_cpu(p[0]), be64_to_cpu(p[1]), be64_to_cpu(p[2]), + be64_to_cpu(p[3])); /* * Ingress WRITE and READ_RESP errors provide @@ -124,8 +123,7 @@ void c4iw_ev_dispatch(struct c4iw_dev *dev, struct t4_cqe *err_cqe) spin_lock_irq(&dev->lock); qhp = get_qhp(dev, CQE_QPID(err_cqe)); if (!qhp) { - printk(KERN_ERR MOD "BAD AE qpid 0x%x opcode %d " - "status 0x%x type %d wrid.hi 0x%x wrid.lo 0x%x\n", + pr_err("BAD AE qpid 0x%x opcode %d status 0x%x type %d wrid.hi 0x%x wrid.lo 0x%x\n", CQE_QPID(err_cqe), CQE_OPCODE(err_cqe), CQE_STATUS(err_cqe), CQE_TYPE(err_cqe), CQE_WRID_HI(err_cqe), @@ -140,8 +138,7 @@ void c4iw_ev_dispatch(struct c4iw_dev *dev, struct t4_cqe *err_cqe) cqid = qhp->attr.rcq; chp = get_chp(dev, cqid); if (!chp) { - printk(KERN_ERR MOD "BAD AE cqid 0x%x qpid 0x%x opcode %d " - "status 0x%x type %d wrid.hi 0x%x wrid.lo 0x%x\n", + pr_err("BAD AE cqid 0x%x qpid 0x%x opcode %d status 0x%x type %d wrid.hi 0x%x wrid.lo 0x%x\n", cqid, CQE_QPID(err_cqe), CQE_OPCODE(err_cqe), CQE_STATUS(err_cqe), CQE_TYPE(err_cqe), CQE_WRID_HI(err_cqe), @@ -165,7 +162,7 @@ void c4iw_ev_dispatch(struct c4iw_dev *dev, struct t4_cqe *err_cqe) /* Completion Events */ case T4_ERR_SUCCESS: - printk(KERN_ERR MOD "AE with status 0!\n"); + pr_err("AE with status 0!\n"); break; case T4_ERR_STAG: @@ -207,7 +204,7 @@ void c4iw_ev_dispatch(struct c4iw_dev *dev, struct t4_cqe *err_cqe) break; default: - printk(KERN_ERR MOD "Unknown T4 status 0x%x QPID 0x%x\n", + pr_err("Unknown T4 status 0x%x QPID 0x%x\n", CQE_STATUS(err_cqe), qhp->wq.sq.qid); post_qp_event(dev, chp, qhp, err_cqe, IB_EVENT_QP_FATAL); break; @@ -237,7 +234,7 @@ int c4iw_ev_handler(struct c4iw_dev *dev, u32 qid) if (atomic_dec_and_test(&chp->refcnt)) wake_up(&chp->wait); } else { - PDBG("%s unknown cqid 0x%x\n", __func__, qid); + pr_debug("%s unknown cqid 0x%x\n", __func__, qid); spin_unlock_irqrestore(&dev->lock, flag); } return 0; diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h index 5846c47c8d55..819a30635d53 100644 --- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h +++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h @@ -64,12 +64,11 @@ #define DRV_NAME "iw_cxgb4" #define MOD DRV_NAME ":" -extern int c4iw_debug; -#define PDBG(fmt, args...) \ -do { \ - if (c4iw_debug) \ - printk(MOD fmt, ## args); \ -} while (0) +#ifdef pr_fmt +#undef pr_fmt +#endif + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include "t4.h" @@ -231,15 +230,15 @@ static inline int c4iw_wait_for_reply(struct c4iw_rdev *rdev, ret = wait_for_completion_timeout(&wr_waitp->completion, C4IW_WR_TO); if (!ret) { - PDBG("%s - Device %s not responding (disabling device) - tid %u qpid %u\n", - func, pci_name(rdev->lldi.pdev), hwtid, qpid); + pr_debug("%s - Device %s not responding (disabling device) - tid %u qpid %u\n", + func, pci_name(rdev->lldi.pdev), hwtid, qpid); rdev->flags |= T4_FATAL_ERROR; wr_waitp->ret = -EIO; } out: if (wr_waitp->ret) - PDBG("%s: FW reply %d tid %u qpid %u\n", - pci_name(rdev->lldi.pdev), wr_waitp->ret, hwtid, qpid); + pr_debug("%s: FW reply %d tid %u qpid %u\n", + pci_name(rdev->lldi.pdev), wr_waitp->ret, hwtid, qpid); return wr_waitp->ret; } @@ -538,8 +537,9 @@ static inline struct c4iw_mm_entry *remove_mmap(struct c4iw_ucontext *ucontext, if (mm->key == key && mm->len == len) { list_del_init(&mm->entry); spin_unlock(&ucontext->mmap_lock); - PDBG("%s key 0x%x addr 0x%llx len %d\n", __func__, - key, (unsigned long long) mm->addr, mm->len); + pr_debug("%s key 0x%x addr 0x%llx len %d\n", + __func__, key, + (unsigned long long)mm->addr, mm->len); return mm; } } @@ -551,8 +551,8 @@ static inline void insert_mmap(struct c4iw_ucontext *ucontext, struct c4iw_mm_entry *mm) { spin_lock(&ucontext->mmap_lock); - PDBG("%s key 0x%x addr 0x%llx len %d\n", __func__, - mm->key, (unsigned long long) mm->addr, mm->len); + pr_debug("%s key 0x%x addr 0x%llx len %d\n", + __func__, mm->key, (unsigned long long)mm->addr, mm->len); list_add_tail(&mm->entry, &ucontext->mmaps); spin_unlock(&ucontext->mmap_lock); } @@ -670,17 +670,19 @@ enum c4iw_mmid_state { #define MPA_V2_RDMA_READ_RTR 0x4000 #define MPA_V2_IRD_ORD_MASK 0x3FFF -#define c4iw_put_ep(ep) { \ - PDBG("put_ep (via %s:%u) ep %p refcnt %d\n", __func__, __LINE__, \ - ep, kref_read(&((ep)->kref))); \ - WARN_ON(kref_read(&((ep)->kref)) < 1); \ - kref_put(&((ep)->kref), _c4iw_free_ep); \ +#define c4iw_put_ep(ep) { \ + pr_debug("put_ep (via %s:%u) ep %p refcnt %d\n", \ + __func__, __LINE__, \ + ep, kref_read(&((ep)->kref))); \ + WARN_ON(kref_read(&((ep)->kref)) < 1); \ + kref_put(&((ep)->kref), _c4iw_free_ep); \ } -#define c4iw_get_ep(ep) { \ - PDBG("get_ep (via %s:%u) ep %p, refcnt %d\n", __func__, __LINE__, \ - ep, kref_read(&((ep)->kref))); \ - kref_get(&((ep)->kref)); \ +#define c4iw_get_ep(ep) { \ + pr_debug("get_ep (via %s:%u) ep %p, refcnt %d\n", \ + __func__, __LINE__, \ + ep, kref_read(&((ep)->kref))); \ + kref_get(&((ep)->kref)); \ } void _c4iw_free_ep(struct kref *kref); diff --git a/drivers/infiniband/hw/cxgb4/mem.c b/drivers/infiniband/hw/cxgb4/mem.c index 410408f886c1..3ee7f43e419a 100644 --- a/drivers/infiniband/hw/cxgb4/mem.c +++ b/drivers/infiniband/hw/cxgb4/mem.c @@ -38,9 +38,9 @@ #include "iw_cxgb4.h" -int use_dsgl = 0; +int use_dsgl = 1; module_param(use_dsgl, int, 0644); -MODULE_PARM_DESC(use_dsgl, "Use DSGL for PBL/FastReg (default=0)"); +MODULE_PARM_DESC(use_dsgl, "Use DSGL for PBL/FastReg (default=1) (DEPRECATED)"); #define T4_ULPTX_MIN_IO 32 #define C4IW_MAX_INLINE_SIZE 96 @@ -125,7 +125,7 @@ static int _c4iw_write_mem_inline(struct c4iw_rdev *rdev, u32 addr, u32 len, cmd |= cpu_to_be32(T5_ULP_MEMIO_IMM_F); addr &= 0x7FFFFFF; - PDBG("%s addr 0x%x len %u\n", __func__, addr, len); + pr_debug("%s addr 0x%x len %u\n", __func__, addr, len); num_wqe = DIV_ROUND_UP(len, C4IW_MAX_INLINE_SIZE); c4iw_init_wr_wait(&wr_wait); for (i = 0; i < num_wqe; i++) { @@ -231,13 +231,11 @@ out: static int write_adapter_mem(struct c4iw_rdev *rdev, u32 addr, u32 len, void *data, struct sk_buff *skb) { - if (is_t5(rdev->lldi.adapter_type) && use_dsgl) { + if (rdev->lldi.ulptx_memwrite_dsgl && use_dsgl) { if (len > inline_threshold) { if (_c4iw_write_mem_dma(rdev, addr, len, data, skb)) { - printk_ratelimited(KERN_WARNING - "%s: dma map" - " failure (non fatal)\n", - pci_name(rdev->lldi.pdev)); + pr_warn_ratelimited("%s: dma map failure (non fatal)\n", + pci_name(rdev->lldi.pdev)); return _c4iw_write_mem_inline(rdev, addr, len, data, skb); } else { @@ -289,8 +287,8 @@ static int write_tpt_entry(struct c4iw_rdev *rdev, u32 reset_tpt_entry, mutex_unlock(&rdev->stats.lock); *stag = (stag_idx << 8) | (atomic_inc_return(&key) & 0xff); } - PDBG("%s stag_state 0x%0x type 0x%0x pdid 0x%0x, stag_idx 0x%x\n", - __func__, stag_state, type, pdid, stag_idx); + pr_debug("%s stag_state 0x%0x type 0x%0x pdid 0x%0x, stag_idx 0x%x\n", + __func__, stag_state, type, pdid, stag_idx); /* write TPT entry */ if (reset_tpt_entry) @@ -331,9 +329,9 @@ static int write_pbl(struct c4iw_rdev *rdev, __be64 *pbl, { int err; - PDBG("%s *pdb_addr 0x%x, pbl_base 0x%x, pbl_size %d\n", - __func__, pbl_addr, rdev->lldi.vr->pbl.start, - pbl_size); + pr_debug("%s *pdb_addr 0x%x, pbl_base 0x%x, pbl_size %d\n", + __func__, pbl_addr, rdev->lldi.vr->pbl.start, + pbl_size); err = write_adapter_mem(rdev, pbl_addr >> 5, pbl_size << 3, pbl, NULL); return err; @@ -376,7 +374,7 @@ static int finish_mem_reg(struct c4iw_mr *mhp, u32 stag) mhp->attr.stag = stag; mmid = stag >> 8; mhp->ibmr.rkey = mhp->ibmr.lkey = stag; - PDBG("%s mmid 0x%x mhp %p\n", __func__, mmid, mhp); + pr_debug("%s mmid 0x%x mhp %p\n", __func__, mmid, mhp); return insert_handle(mhp->rhp, &mhp->rhp->mmidr, mhp, mmid); } @@ -426,7 +424,7 @@ struct ib_mr *c4iw_get_dma_mr(struct ib_pd *pd, int acc) int ret; u32 stag = T4_STAG_UNSET; - PDBG("%s ib_pd %p\n", __func__, pd); + pr_debug("%s ib_pd %p\n", __func__, pd); php = to_c4iw_pd(pd); rhp = php->rhp; @@ -483,7 +481,7 @@ struct ib_mr *c4iw_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, struct c4iw_pd *php; struct c4iw_mr *mhp; - PDBG("%s ib_pd %p\n", __func__, pd); + pr_debug("%s ib_pd %p\n", __func__, pd); if (length == ~0ULL) return ERR_PTR(-EINVAL); @@ -517,7 +515,7 @@ struct ib_mr *c4iw_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, return ERR_PTR(err); } - shift = ffs(mhp->umem->page_size) - 1; + shift = mhp->umem->page_shift; n = mhp->umem->nmap; err = alloc_pbl(mhp, n); @@ -536,7 +534,7 @@ struct ib_mr *c4iw_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, len = sg_dma_len(sg) >> shift; for (k = 0; k < len; ++k) { pages[i++] = cpu_to_be64(sg_dma_address(sg) + - mhp->umem->page_size * k); + (k << shift)); if (i == PAGE_SIZE / sizeof *pages) { err = write_pbl(&mhp->rhp->rdev, pages, @@ -620,7 +618,7 @@ struct ib_mw *c4iw_alloc_mw(struct ib_pd *pd, enum ib_mw_type type, ret = -ENOMEM; goto dealloc_win; } - PDBG("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); + pr_debug("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); return &(mhp->ibmw); dealloc_win: @@ -645,7 +643,7 @@ int c4iw_dealloc_mw(struct ib_mw *mw) deallocate_window(&rhp->rdev, mhp->attr.stag, mhp->dereg_skb); kfree_skb(mhp->dereg_skb); kfree(mhp); - PDBG("%s ib_mw %p mmid 0x%x ptr %p\n", __func__, mw, mmid, mhp); + pr_debug("%s ib_mw %p mmid 0x%x ptr %p\n", __func__, mw, mmid, mhp); return 0; } @@ -703,7 +701,7 @@ struct ib_mr *c4iw_alloc_mr(struct ib_pd *pd, goto err3; } - PDBG("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); + pr_debug("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); return &(mhp->ibmr); err3: dereg_mem(&rhp->rdev, stag, mhp->attr.pbl_size, @@ -748,7 +746,7 @@ int c4iw_dereg_mr(struct ib_mr *ib_mr) struct c4iw_mr *mhp; u32 mmid; - PDBG("%s ib_mr %p\n", __func__, ib_mr); + pr_debug("%s ib_mr %p\n", __func__, ib_mr); mhp = to_c4iw_mr(ib_mr); rhp = mhp->rhp; @@ -766,7 +764,7 @@ int c4iw_dereg_mr(struct ib_mr *ib_mr) kfree((void *) (unsigned long) mhp->kva); if (mhp->umem) ib_umem_release(mhp->umem); - PDBG("%s mmid 0x%x ptr %p\n", __func__, mmid, mhp); + pr_debug("%s mmid 0x%x ptr %p\n", __func__, mmid, mhp); kfree(mhp); return 0; } diff --git a/drivers/infiniband/hw/cxgb4/provider.c b/drivers/infiniband/hw/cxgb4/provider.c index df64417ab6f2..6770a314b086 100644 --- a/drivers/infiniband/hw/cxgb4/provider.c +++ b/drivers/infiniband/hw/cxgb4/provider.c @@ -102,7 +102,7 @@ void _c4iw_free_ucontext(struct kref *kref) ucontext = container_of(kref, struct c4iw_ucontext, kref); rhp = to_c4iw_dev(ucontext->ibucontext.device); - PDBG("%s ucontext %p\n", __func__, ucontext); + pr_debug("%s ucontext %p\n", __func__, ucontext); list_for_each_entry_safe(mm, tmp, &ucontext->mmaps, entry) kfree(mm); c4iw_release_dev_ucontext(&rhp->rdev, &ucontext->uctx); @@ -113,7 +113,7 @@ static int c4iw_dealloc_ucontext(struct ib_ucontext *context) { struct c4iw_ucontext *ucontext = to_c4iw_ucontext(context); - PDBG("%s context %p\n", __func__, context); + pr_debug("%s context %p\n", __func__, context); c4iw_put_ucontext(ucontext); return 0; } @@ -123,12 +123,11 @@ static struct ib_ucontext *c4iw_alloc_ucontext(struct ib_device *ibdev, { struct c4iw_ucontext *context; struct c4iw_dev *rhp = to_c4iw_dev(ibdev); - static int warned; struct c4iw_alloc_ucontext_resp uresp; int ret = 0; struct c4iw_mm_entry *mm = NULL; - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); context = kzalloc(sizeof(*context), GFP_KERNEL); if (!context) { ret = -ENOMEM; @@ -141,8 +140,7 @@ static struct ib_ucontext *c4iw_alloc_ucontext(struct ib_device *ibdev, kref_init(&context->kref); if (udata->outlen < sizeof(uresp) - sizeof(uresp.reserved)) { - if (!warned++) - pr_err(MOD "Warning - downlevel libcxgb4 (non-fatal), device status page disabled."); + pr_err_once("Warning - downlevel libcxgb4 (non-fatal), device status page disabled\n"); rhp->rdev.flags |= T4_STATUS_PAGE_DISABLED; } else { mm = kmalloc(sizeof(*mm), GFP_KERNEL); @@ -187,8 +185,8 @@ static int c4iw_mmap(struct ib_ucontext *context, struct vm_area_struct *vma) struct c4iw_ucontext *ucontext; u64 addr; - PDBG("%s pgoff 0x%lx key 0x%x len %d\n", __func__, vma->vm_pgoff, - key, len); + pr_debug("%s pgoff 0x%lx key 0x%x len %d\n", __func__, vma->vm_pgoff, + key, len); if (vma->vm_start & (PAGE_SIZE-1)) return -EINVAL; @@ -253,7 +251,7 @@ static int c4iw_deallocate_pd(struct ib_pd *pd) php = to_c4iw_pd(pd); rhp = php->rhp; - PDBG("%s ibpd %p pdid 0x%x\n", __func__, pd, php->pdid); + pr_debug("%s ibpd %p pdid 0x%x\n", __func__, pd, php->pdid); c4iw_put_resource(&rhp->rdev.resource.pdid_table, php->pdid); mutex_lock(&rhp->rdev.stats.lock); rhp->rdev.stats.pd.cur--; @@ -270,7 +268,7 @@ static struct ib_pd *c4iw_allocate_pd(struct ib_device *ibdev, u32 pdid; struct c4iw_dev *rhp; - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); rhp = (struct c4iw_dev *) ibdev; pdid = c4iw_get_resource(&rhp->rdev.resource.pdid_table); if (!pdid) @@ -293,14 +291,14 @@ static struct ib_pd *c4iw_allocate_pd(struct ib_device *ibdev, if (rhp->rdev.stats.pd.cur > rhp->rdev.stats.pd.max) rhp->rdev.stats.pd.max = rhp->rdev.stats.pd.cur; mutex_unlock(&rhp->rdev.stats.lock); - PDBG("%s pdid 0x%0x ptr 0x%p\n", __func__, pdid, php); + pr_debug("%s pdid 0x%0x ptr 0x%p\n", __func__, pdid, php); return &php->ibpd; } static int c4iw_query_pkey(struct ib_device *ibdev, u8 port, u16 index, u16 *pkey) { - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); *pkey = 0; return 0; } @@ -310,8 +308,8 @@ static int c4iw_query_gid(struct ib_device *ibdev, u8 port, int index, { struct c4iw_dev *dev; - PDBG("%s ibdev %p, port %d, index %d, gid %p\n", - __func__, ibdev, port, index, gid); + pr_debug("%s ibdev %p, port %d, index %d, gid %p\n", + __func__, ibdev, port, index, gid); dev = to_c4iw_dev(ibdev); BUG_ON(port == 0); memset(&(gid->raw[0]), 0, sizeof(gid->raw)); @@ -325,7 +323,7 @@ static int c4iw_query_device(struct ib_device *ibdev, struct ib_device_attr *pro struct c4iw_dev *dev; - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); if (uhw->inlen || uhw->outlen) return -EINVAL; @@ -366,7 +364,7 @@ static int c4iw_query_port(struct ib_device *ibdev, u8 port, struct net_device *netdev; struct in_device *inetdev; - PDBG("%s ibdev %p\n", __func__, ibdev); + pr_debug("%s ibdev %p\n", __func__, ibdev); dev = to_c4iw_dev(ibdev); netdev = dev->rdev.lldi.ports[port-1]; @@ -408,7 +406,7 @@ static ssize_t show_rev(struct device *dev, struct device_attribute *attr, { struct c4iw_dev *c4iw_dev = container_of(dev, struct c4iw_dev, ibdev.dev); - PDBG("%s dev 0x%p\n", __func__, dev); + pr_debug("%s dev 0x%p\n", __func__, dev); return sprintf(buf, "%d\n", CHELSIO_CHIP_RELEASE(c4iw_dev->rdev.lldi.adapter_type)); } @@ -421,7 +419,7 @@ static ssize_t show_hca(struct device *dev, struct device_attribute *attr, struct ethtool_drvinfo info; struct net_device *lldev = c4iw_dev->rdev.lldi.ports[0]; - PDBG("%s dev 0x%p\n", __func__, dev); + pr_debug("%s dev 0x%p\n", __func__, dev); lldev->ethtool_ops->get_drvinfo(lldev, &info); return sprintf(buf, "%s\n", info.driver); } @@ -431,7 +429,7 @@ static ssize_t show_board(struct device *dev, struct device_attribute *attr, { struct c4iw_dev *c4iw_dev = container_of(dev, struct c4iw_dev, ibdev.dev); - PDBG("%s dev 0x%p\n", __func__, dev); + pr_debug("%s dev 0x%p\n", __func__, dev); return sprintf(buf, "%x.%x\n", c4iw_dev->rdev.lldi.pdev->vendor, c4iw_dev->rdev.lldi.pdev->device); } @@ -524,7 +522,7 @@ static void get_dev_fw_str(struct ib_device *dev, char *str, { struct c4iw_dev *c4iw_dev = container_of(dev, struct c4iw_dev, ibdev); - PDBG("%s dev 0x%p\n", __func__, dev); + pr_debug("%s dev 0x%p\n", __func__, dev); snprintf(str, str_len, "%u.%u.%u.%u", FW_HDR_FW_VER_MAJOR_G(c4iw_dev->rdev.lldi.fw_vers), @@ -538,7 +536,7 @@ int c4iw_register_device(struct c4iw_dev *dev) int ret; int i; - PDBG("%s c4iw_dev %p\n", __func__, dev); + pr_debug("%s c4iw_dev %p\n", __func__, dev); BUG_ON(!dev->rdev.lldi.ports[0]); strlcpy(dev->ibdev.name, "cxgb4_%d", IB_DEVICE_NAME_MAX); memset(&dev->ibdev.node_guid, 0, sizeof(dev->ibdev.node_guid)); @@ -648,7 +646,7 @@ void c4iw_unregister_device(struct c4iw_dev *dev) { int i; - PDBG("%s c4iw_dev %p\n", __func__, dev); + pr_debug("%s c4iw_dev %p\n", __func__, dev); for (i = 0; i < ARRAY_SIZE(c4iw_class_attributes); ++i) device_remove_file(&dev->ibdev.dev, c4iw_class_attributes[i]); diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c index d4fd2f5c8326..8e4154b4253e 100644 --- a/drivers/infiniband/hw/cxgb4/qp.c +++ b/drivers/infiniband/hw/cxgb4/qp.c @@ -254,11 +254,11 @@ static int create_qp(struct c4iw_rdev *rdev, struct t4_wq *wq, ret = -ENOMEM; goto free_sq; } - PDBG("%s sq base va 0x%p pa 0x%llx rq base va 0x%p pa 0x%llx\n", - __func__, wq->sq.queue, - (unsigned long long)virt_to_phys(wq->sq.queue), - wq->rq.queue, - (unsigned long long)virt_to_phys(wq->rq.queue)); + pr_debug("%s sq base va 0x%p pa 0x%llx rq base va 0x%p pa 0x%llx\n", + __func__, wq->sq.queue, + (unsigned long long)virt_to_phys(wq->sq.queue), + wq->rq.queue, + (unsigned long long)virt_to_phys(wq->rq.queue)); memset(wq->rq.queue, 0, wq->rq.memsize); dma_unmap_addr_set(&wq->rq, mapping, wq->rq.dma_addr); @@ -275,7 +275,7 @@ static int create_qp(struct c4iw_rdev *rdev, struct t4_wq *wq, * User mode must have bar2 access. */ if (user && (!wq->sq.bar2_pa || !wq->rq.bar2_pa)) { - pr_warn(MOD "%s: sqid %u or rqid %u not in BAR2 range.\n", + pr_warn("%s: sqid %u or rqid %u not in BAR2 range\n", pci_name(rdev->lldi.pdev), wq->sq.qid, wq->rq.qid); goto free_dma; } @@ -362,9 +362,9 @@ static int create_qp(struct c4iw_rdev *rdev, struct t4_wq *wq, if (ret) goto free_dma; - PDBG("%s sqid 0x%x rqid 0x%x kdb 0x%p sq_bar2_addr %p rq_bar2_addr %p\n", - __func__, wq->sq.qid, wq->rq.qid, wq->db, - wq->sq.bar2_va, wq->rq.bar2_va); + pr_debug("%s sqid 0x%x rqid 0x%x kdb 0x%p sq_bar2_addr %p rq_bar2_addr %p\n", + __func__, wq->sq.qid, wq->rq.qid, wq->db, + wq->sq.bar2_va, wq->rq.bar2_va); return 0; free_dma: @@ -725,7 +725,7 @@ static void free_qp_work(struct work_struct *work) ucontext = qhp->ucontext; rhp = qhp->rhp; - PDBG("%s qhp %p ucontext %p\n", __func__, qhp, ucontext); + pr_debug("%s qhp %p ucontext %p\n", __func__, qhp, ucontext); destroy_qp(&rhp->rdev, &qhp->wq, ucontext ? &ucontext->uctx : &rhp->rdev.uctx); @@ -739,19 +739,19 @@ static void queue_qp_free(struct kref *kref) struct c4iw_qp *qhp; qhp = container_of(kref, struct c4iw_qp, kref); - PDBG("%s qhp %p\n", __func__, qhp); + pr_debug("%s qhp %p\n", __func__, qhp); queue_work(qhp->rhp->rdev.free_workq, &qhp->free_work); } void c4iw_qp_add_ref(struct ib_qp *qp) { - PDBG("%s ib_qp %p\n", __func__, qp); + pr_debug("%s ib_qp %p\n", __func__, qp); kref_get(&to_c4iw_qp(qp)->kref); } void c4iw_qp_rem_ref(struct ib_qp *qp) { - PDBG("%s ib_qp %p\n", __func__, qp); + pr_debug("%s ib_qp %p\n", __func__, qp); kref_put(&to_c4iw_qp(qp)->kref, queue_qp_free); } @@ -959,8 +959,8 @@ int c4iw_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, c4iw_invalidate_mr(qhp->rhp, wr->ex.invalidate_rkey); break; default: - PDBG("%s post of type=%d TBD!\n", __func__, - wr->opcode); + pr_debug("%s post of type=%d TBD!\n", __func__, + wr->opcode); err = -EINVAL; } if (err) { @@ -981,9 +981,10 @@ int c4iw_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, init_wr_hdr(wqe, qhp->wq.sq.pidx, fw_opcode, fw_flags, len16); - PDBG("%s cookie 0x%llx pidx 0x%x opcode 0x%x read_len %u\n", - __func__, (unsigned long long)wr->wr_id, qhp->wq.sq.pidx, - swsqe->opcode, swsqe->read_len); + pr_debug("%s cookie 0x%llx pidx 0x%x opcode 0x%x read_len %u\n", + __func__, + (unsigned long long)wr->wr_id, qhp->wq.sq.pidx, + swsqe->opcode, swsqe->read_len); wr = wr->next; num_wrs--; t4_sq_produce(&qhp->wq, len16); @@ -1057,8 +1058,9 @@ int c4iw_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr, wqe->recv.r2[1] = 0; wqe->recv.r2[2] = 0; wqe->recv.len16 = len16; - PDBG("%s cookie 0x%llx pidx %u\n", __func__, - (unsigned long long) wr->wr_id, qhp->wq.rq.pidx); + pr_debug("%s cookie 0x%llx pidx %u\n", + __func__, + (unsigned long long)wr->wr_id, qhp->wq.rq.pidx); t4_rq_produce(&qhp->wq, len16); idx += DIV_ROUND_UP(len16*16, T4_EQ_ENTRY_SIZE); wr = wr->next; @@ -1217,8 +1219,8 @@ static void post_terminate(struct c4iw_qp *qhp, struct t4_cqe *err_cqe, struct sk_buff *skb; struct terminate_message *term; - PDBG("%s qhp %p qid 0x%x tid %u\n", __func__, qhp, qhp->wq.sq.qid, - qhp->ep->hwtid); + pr_debug("%s qhp %p qid 0x%x tid %u\n", __func__, qhp, qhp->wq.sq.qid, + qhp->ep->hwtid); skb = skb_dequeue(&qhp->ep->com.ep_skb_list); if (WARN_ON(!skb)) @@ -1254,7 +1256,7 @@ static void __flush_qp(struct c4iw_qp *qhp, struct c4iw_cq *rchp, int rq_flushed, sq_flushed; unsigned long flag; - PDBG("%s qhp %p rchp %p schp %p\n", __func__, qhp, rchp, schp); + pr_debug("%s qhp %p rchp %p schp %p\n", __func__, qhp, rchp, schp); /* locking hierarchy: cq lock first, then qp lock. */ spin_lock_irqsave(&rchp->lock, flag); @@ -1339,8 +1341,8 @@ static int rdma_fini(struct c4iw_dev *rhp, struct c4iw_qp *qhp, int ret; struct sk_buff *skb; - PDBG("%s qhp %p qid 0x%x tid %u\n", __func__, qhp, qhp->wq.sq.qid, - ep->hwtid); + pr_debug("%s qhp %p qid 0x%x tid %u\n", __func__, qhp, qhp->wq.sq.qid, + ep->hwtid); skb = skb_dequeue(&ep->com.ep_skb_list); if (WARN_ON(!skb)) @@ -1366,13 +1368,13 @@ static int rdma_fini(struct c4iw_dev *rhp, struct c4iw_qp *qhp, ret = c4iw_wait_for_reply(&rhp->rdev, &ep->com.wr_wait, qhp->ep->hwtid, qhp->wq.sq.qid, __func__); out: - PDBG("%s ret %d\n", __func__, ret); + pr_debug("%s ret %d\n", __func__, ret); return ret; } static void build_rtr_msg(u8 p2p_type, struct fw_ri_init *init) { - PDBG("%s p2p_type = %d\n", __func__, p2p_type); + pr_debug("%s p2p_type = %d\n", __func__, p2p_type); memset(&init->u, 0, sizeof init->u); switch (p2p_type) { case FW_RI_INIT_P2PTYPE_RDMA_WRITE: @@ -1401,8 +1403,8 @@ static int rdma_init(struct c4iw_dev *rhp, struct c4iw_qp *qhp) int ret; struct sk_buff *skb; - PDBG("%s qhp %p qid 0x%x tid %u ird %u ord %u\n", __func__, qhp, - qhp->wq.sq.qid, qhp->ep->hwtid, qhp->ep->ird, qhp->ep->ord); + pr_debug("%s qhp %p qid 0x%x tid %u ird %u ord %u\n", __func__, qhp, + qhp->wq.sq.qid, qhp->ep->hwtid, qhp->ep->ird, qhp->ep->ord); skb = alloc_skb(sizeof *wqe, GFP_KERNEL); if (!skb) { @@ -1474,7 +1476,7 @@ static int rdma_init(struct c4iw_dev *rhp, struct c4iw_qp *qhp) err1: free_ird(rhp, qhp->attr.max_ird); out: - PDBG("%s ret %d\n", __func__, ret); + pr_debug("%s ret %d\n", __func__, ret); return ret; } @@ -1491,9 +1493,10 @@ int c4iw_modify_qp(struct c4iw_dev *rhp, struct c4iw_qp *qhp, int free = 0; struct c4iw_ep *ep = NULL; - PDBG("%s qhp %p sqid 0x%x rqid 0x%x ep %p state %d -> %d\n", __func__, - qhp, qhp->wq.sq.qid, qhp->wq.rq.qid, qhp->ep, qhp->attr.state, - (mask & C4IW_QP_ATTR_NEXT_STATE) ? attrs->next_state : -1); + pr_debug("%s qhp %p sqid 0x%x rqid 0x%x ep %p state %d -> %d\n", + __func__, + qhp, qhp->wq.sq.qid, qhp->wq.rq.qid, qhp->ep, qhp->attr.state, + (mask & C4IW_QP_ATTR_NEXT_STATE) ? attrs->next_state : -1); mutex_lock(&qhp->mutex); @@ -1671,16 +1674,15 @@ int c4iw_modify_qp(struct c4iw_dev *rhp, struct c4iw_qp *qhp, goto err; break; default: - printk(KERN_ERR "%s in a bad state %d\n", - __func__, qhp->attr.state); + pr_err("%s in a bad state %d\n", __func__, qhp->attr.state); ret = -EINVAL; goto err; break; } goto out; err: - PDBG("%s disassociating ep %p qpid 0x%x\n", __func__, qhp->ep, - qhp->wq.sq.qid); + pr_debug("%s disassociating ep %p qpid 0x%x\n", __func__, qhp->ep, + qhp->wq.sq.qid); /* disassociate the LLP connection */ qhp->attr.llp_stream_handle = NULL; @@ -1716,7 +1718,7 @@ out: */ if (free) c4iw_put_ep(&ep->com); - PDBG("%s exit state %d\n", __func__, qhp->attr.state); + pr_debug("%s exit state %d\n", __func__, qhp->attr.state); return ret; } @@ -1746,7 +1748,7 @@ int c4iw_destroy_qp(struct ib_qp *ib_qp) c4iw_qp_rem_ref(ib_qp); - PDBG("%s ib_qp %p qpid 0x%0x\n", __func__, ib_qp, qhp->wq.sq.qid); + pr_debug("%s ib_qp %p qpid 0x%0x\n", __func__, ib_qp, qhp->wq.sq.qid); return 0; } @@ -1765,7 +1767,7 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs, struct c4iw_mm_entry *sq_key_mm, *rq_key_mm = NULL, *sq_db_key_mm; struct c4iw_mm_entry *rq_db_key_mm = NULL, *ma_sync_key_mm = NULL; - PDBG("%s ib_pd %p\n", __func__, pd); + pr_debug("%s ib_pd %p\n", __func__, pd); if (attrs->qp_type != IB_QPT_RC) return ERR_PTR(-EINVAL); @@ -1936,11 +1938,11 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs, qhp->ibqp.qp_num = qhp->wq.sq.qid; init_timer(&(qhp->timer)); INIT_LIST_HEAD(&qhp->db_fc_entry); - PDBG("%s sq id %u size %u memsize %zu num_entries %u " - "rq id %u size %u memsize %zu num_entries %u\n", __func__, - qhp->wq.sq.qid, qhp->wq.sq.size, qhp->wq.sq.memsize, - attrs->cap.max_send_wr, qhp->wq.rq.qid, qhp->wq.rq.size, - qhp->wq.rq.memsize, attrs->cap.max_recv_wr); + pr_debug("%s sq id %u size %u memsize %zu num_entries %u rq id %u size %u memsize %zu num_entries %u\n", + __func__, + qhp->wq.sq.qid, qhp->wq.sq.size, qhp->wq.sq.memsize, + attrs->cap.max_send_wr, qhp->wq.rq.qid, qhp->wq.rq.size, + qhp->wq.rq.memsize, attrs->cap.max_recv_wr); return &qhp->ibqp; err8: kfree(ma_sync_key_mm); @@ -1970,7 +1972,7 @@ int c4iw_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, enum c4iw_qp_attr_mask mask = 0; struct c4iw_qp_attributes attrs; - PDBG("%s ib_qp %p\n", __func__, ibqp); + pr_debug("%s ib_qp %p\n", __func__, ibqp); /* iwarp does not support the RTR state */ if ((attr_mask & IB_QP_STATE) && (attr->qp_state == IB_QPS_RTR)) @@ -2016,7 +2018,7 @@ int c4iw_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, struct ib_qp *c4iw_get_qp(struct ib_device *dev, int qpn) { - PDBG("%s ib_dev %p qpn 0x%x\n", __func__, dev, qpn); + pr_debug("%s ib_dev %p qpn 0x%x\n", __func__, dev, qpn); return (struct ib_qp *)get_qhp(to_c4iw_dev(dev), qpn); } diff --git a/drivers/infiniband/hw/cxgb4/resource.c b/drivers/infiniband/hw/cxgb4/resource.c index 67df71a7012e..8ff0cbe5cb16 100644 --- a/drivers/infiniband/hw/cxgb4/resource.c +++ b/drivers/infiniband/hw/cxgb4/resource.c @@ -90,7 +90,7 @@ u32 c4iw_get_resource(struct c4iw_id_table *id_table) void c4iw_put_resource(struct c4iw_id_table *id_table, u32 entry) { - PDBG("%s entry 0x%x\n", __func__, entry); + pr_debug("%s entry 0x%x\n", __func__, entry); c4iw_id_free(id_table, entry); } @@ -141,7 +141,7 @@ u32 c4iw_get_cqid(struct c4iw_rdev *rdev, struct c4iw_dev_ucontext *uctx) } out: mutex_unlock(&uctx->lock); - PDBG("%s qid 0x%x\n", __func__, qid); + pr_debug("%s qid 0x%x\n", __func__, qid); mutex_lock(&rdev->stats.lock); if (rdev->stats.qid.cur > rdev->stats.qid.max) rdev->stats.qid.max = rdev->stats.qid.cur; @@ -157,7 +157,7 @@ void c4iw_put_cqid(struct c4iw_rdev *rdev, u32 qid, entry = kmalloc(sizeof *entry, GFP_KERNEL); if (!entry) return; - PDBG("%s qid 0x%x\n", __func__, qid); + pr_debug("%s qid 0x%x\n", __func__, qid); entry->qid = qid; mutex_lock(&uctx->lock); list_add_tail(&entry->entry, &uctx->cqids); @@ -215,7 +215,7 @@ u32 c4iw_get_qpid(struct c4iw_rdev *rdev, struct c4iw_dev_ucontext *uctx) } out: mutex_unlock(&uctx->lock); - PDBG("%s qid 0x%x\n", __func__, qid); + pr_debug("%s qid 0x%x\n", __func__, qid); mutex_lock(&rdev->stats.lock); if (rdev->stats.qid.cur > rdev->stats.qid.max) rdev->stats.qid.max = rdev->stats.qid.cur; @@ -231,7 +231,7 @@ void c4iw_put_qpid(struct c4iw_rdev *rdev, u32 qid, entry = kmalloc(sizeof *entry, GFP_KERNEL); if (!entry) return; - PDBG("%s qid 0x%x\n", __func__, qid); + pr_debug("%s qid 0x%x\n", __func__, qid); entry->qid = qid; mutex_lock(&uctx->lock); list_add_tail(&entry->entry, &uctx->qpids); @@ -254,7 +254,7 @@ void c4iw_destroy_resource(struct c4iw_resource *rscp) u32 c4iw_pblpool_alloc(struct c4iw_rdev *rdev, int size) { unsigned long addr = gen_pool_alloc(rdev->pbl_pool, size); - PDBG("%s addr 0x%x size %d\n", __func__, (u32)addr, size); + pr_debug("%s addr 0x%x size %d\n", __func__, (u32)addr, size); mutex_lock(&rdev->stats.lock); if (addr) { rdev->stats.pbl.cur += roundup(size, 1 << MIN_PBL_SHIFT); @@ -268,7 +268,7 @@ u32 c4iw_pblpool_alloc(struct c4iw_rdev *rdev, int size) void c4iw_pblpool_free(struct c4iw_rdev *rdev, u32 addr, int size) { - PDBG("%s addr 0x%x size %d\n", __func__, addr, size); + pr_debug("%s addr 0x%x size %d\n", __func__, addr, size); mutex_lock(&rdev->stats.lock); rdev->stats.pbl.cur -= roundup(size, 1 << MIN_PBL_SHIFT); mutex_unlock(&rdev->stats.lock); @@ -290,19 +290,17 @@ int c4iw_pblpool_create(struct c4iw_rdev *rdev) while (pbl_start < pbl_top) { pbl_chunk = min(pbl_top - pbl_start + 1, pbl_chunk); if (gen_pool_add(rdev->pbl_pool, pbl_start, pbl_chunk, -1)) { - PDBG("%s failed to add PBL chunk (%x/%x)\n", - __func__, pbl_start, pbl_chunk); + pr_debug("%s failed to add PBL chunk (%x/%x)\n", + __func__, pbl_start, pbl_chunk); if (pbl_chunk <= 1024 << MIN_PBL_SHIFT) { - printk(KERN_WARNING MOD - "Failed to add all PBL chunks (%x/%x)\n", - pbl_start, - pbl_top - pbl_start); + pr_warn("Failed to add all PBL chunks (%x/%x)\n", + pbl_start, pbl_top - pbl_start); return 0; } pbl_chunk >>= 1; } else { - PDBG("%s added PBL chunk (%x/%x)\n", - __func__, pbl_start, pbl_chunk); + pr_debug("%s added PBL chunk (%x/%x)\n", + __func__, pbl_start, pbl_chunk); pbl_start += pbl_chunk; } } @@ -324,9 +322,9 @@ void c4iw_pblpool_destroy(struct c4iw_rdev *rdev) u32 c4iw_rqtpool_alloc(struct c4iw_rdev *rdev, int size) { unsigned long addr = gen_pool_alloc(rdev->rqt_pool, size << 6); - PDBG("%s addr 0x%x size %d\n", __func__, (u32)addr, size << 6); + pr_debug("%s addr 0x%x size %d\n", __func__, (u32)addr, size << 6); if (!addr) - pr_warn_ratelimited(MOD "%s: Out of RQT memory\n", + pr_warn_ratelimited("%s: Out of RQT memory\n", pci_name(rdev->lldi.pdev)); mutex_lock(&rdev->stats.lock); if (addr) { @@ -341,7 +339,7 @@ u32 c4iw_rqtpool_alloc(struct c4iw_rdev *rdev, int size) void c4iw_rqtpool_free(struct c4iw_rdev *rdev, u32 addr, int size) { - PDBG("%s addr 0x%x size %d\n", __func__, addr, size << 6); + pr_debug("%s addr 0x%x size %d\n", __func__, addr, size << 6); mutex_lock(&rdev->stats.lock); rdev->stats.rqt.cur -= roundup(size << 6, 1 << MIN_RQT_SHIFT); mutex_unlock(&rdev->stats.lock); @@ -363,18 +361,17 @@ int c4iw_rqtpool_create(struct c4iw_rdev *rdev) while (rqt_start < rqt_top) { rqt_chunk = min(rqt_top - rqt_start + 1, rqt_chunk); if (gen_pool_add(rdev->rqt_pool, rqt_start, rqt_chunk, -1)) { - PDBG("%s failed to add RQT chunk (%x/%x)\n", - __func__, rqt_start, rqt_chunk); + pr_debug("%s failed to add RQT chunk (%x/%x)\n", + __func__, rqt_start, rqt_chunk); if (rqt_chunk <= 1024 << MIN_RQT_SHIFT) { - printk(KERN_WARNING MOD - "Failed to add all RQT chunks (%x/%x)\n", - rqt_start, rqt_top - rqt_start); + pr_warn("Failed to add all RQT chunks (%x/%x)\n", + rqt_start, rqt_top - rqt_start); return 0; } rqt_chunk >>= 1; } else { - PDBG("%s added RQT chunk (%x/%x)\n", - __func__, rqt_start, rqt_chunk); + pr_debug("%s added RQT chunk (%x/%x)\n", + __func__, rqt_start, rqt_chunk); rqt_start += rqt_chunk; } } @@ -394,7 +391,7 @@ void c4iw_rqtpool_destroy(struct c4iw_rdev *rdev) u32 c4iw_ocqp_pool_alloc(struct c4iw_rdev *rdev, int size) { unsigned long addr = gen_pool_alloc(rdev->ocqp_pool, size); - PDBG("%s addr 0x%x size %d\n", __func__, (u32)addr, size); + pr_debug("%s addr 0x%x size %d\n", __func__, (u32)addr, size); if (addr) { mutex_lock(&rdev->stats.lock); rdev->stats.ocqp.cur += roundup(size, 1 << MIN_OCQP_SHIFT); @@ -407,7 +404,7 @@ u32 c4iw_ocqp_pool_alloc(struct c4iw_rdev *rdev, int size) void c4iw_ocqp_pool_free(struct c4iw_rdev *rdev, u32 addr, int size) { - PDBG("%s addr 0x%x size %d\n", __func__, addr, size); + pr_debug("%s addr 0x%x size %d\n", __func__, addr, size); mutex_lock(&rdev->stats.lock); rdev->stats.ocqp.cur -= roundup(size, 1 << MIN_OCQP_SHIFT); mutex_unlock(&rdev->stats.lock); @@ -429,18 +426,17 @@ int c4iw_ocqp_pool_create(struct c4iw_rdev *rdev) while (start < top) { chunk = min(top - start + 1, chunk); if (gen_pool_add(rdev->ocqp_pool, start, chunk, -1)) { - PDBG("%s failed to add OCQP chunk (%x/%x)\n", - __func__, start, chunk); + pr_debug("%s failed to add OCQP chunk (%x/%x)\n", + __func__, start, chunk); if (chunk <= 1024 << MIN_OCQP_SHIFT) { - printk(KERN_WARNING MOD - "Failed to add all OCQP chunks (%x/%x)\n", - start, top - start); + pr_warn("Failed to add all OCQP chunks (%x/%x)\n", + start, top - start); return 0; } chunk >>= 1; } else { - PDBG("%s added OCQP chunk (%x/%x)\n", - __func__, start, chunk); + pr_debug("%s added OCQP chunk (%x/%x)\n", + __func__, start, chunk); start += chunk; } } diff --git a/drivers/infiniband/hw/cxgb4/t4.h b/drivers/infiniband/hw/cxgb4/t4.h index 640d22148a3e..e765c00303cd 100644 --- a/drivers/infiniband/hw/cxgb4/t4.h +++ b/drivers/infiniband/hw/cxgb4/t4.h @@ -466,14 +466,14 @@ static inline void t4_ring_sq_db(struct t4_wq *wq, u16 inc, union t4_wr *wqe) wmb(); if (wq->sq.bar2_va) { if (inc == 1 && wq->sq.bar2_qid == 0 && wqe) { - PDBG("%s: WC wq->sq.pidx = %d\n", - __func__, wq->sq.pidx); + pr_debug("%s: WC wq->sq.pidx = %d\n", + __func__, wq->sq.pidx); pio_copy((u64 __iomem *) (wq->sq.bar2_va + SGE_UDB_WCDOORBELL), (u64 *)wqe); } else { - PDBG("%s: DB wq->sq.pidx = %d\n", - __func__, wq->sq.pidx); + pr_debug("%s: DB wq->sq.pidx = %d\n", + __func__, wq->sq.pidx); writel(PIDX_T5_V(inc) | QID_V(wq->sq.bar2_qid), wq->sq.bar2_va + SGE_UDB_KDOORBELL); } @@ -493,14 +493,14 @@ static inline void t4_ring_rq_db(struct t4_wq *wq, u16 inc, wmb(); if (wq->rq.bar2_va) { if (inc == 1 && wq->rq.bar2_qid == 0 && wqe) { - PDBG("%s: WC wq->rq.pidx = %d\n", - __func__, wq->rq.pidx); + pr_debug("%s: WC wq->rq.pidx = %d\n", + __func__, wq->rq.pidx); pio_copy((u64 __iomem *) (wq->rq.bar2_va + SGE_UDB_WCDOORBELL), (void *)wqe); } else { - PDBG("%s: DB wq->rq.pidx = %d\n", - __func__, wq->rq.pidx); + pr_debug("%s: DB wq->rq.pidx = %d\n", + __func__, wq->rq.pidx); writel(PIDX_T5_V(inc) | QID_V(wq->rq.bar2_qid), wq->rq.bar2_va + SGE_UDB_KDOORBELL); } @@ -601,7 +601,8 @@ static inline void t4_swcq_produce(struct t4_cq *cq) { cq->sw_in_use++; if (cq->sw_in_use == cq->size) { - PDBG("%s cxgb4 sw cq overflow cqid %u\n", __func__, cq->cqid); + pr_debug("%s cxgb4 sw cq overflow cqid %u\n", + __func__, cq->cqid); cq->error = 1; BUG_ON(1); } @@ -656,7 +657,7 @@ static inline int t4_next_hw_cqe(struct t4_cq *cq, struct t4_cqe **cqe) if (cq->queue[prev_cidx].bits_type_ts != cq->bits_type_ts) { ret = -EOVERFLOW; cq->error = 1; - printk(KERN_ERR MOD "cq overflow cqid %u\n", cq->cqid); + pr_err("cq overflow cqid %u\n", cq->cqid); BUG_ON(1); } else if (t4_valid_cqe(cq, &cq->queue[cq->cidx])) { @@ -672,7 +673,8 @@ static inline int t4_next_hw_cqe(struct t4_cq *cq, struct t4_cqe **cqe) static inline struct t4_cqe *t4_next_sw_cqe(struct t4_cq *cq) { if (cq->sw_in_use == cq->size) { - PDBG("%s cxgb4 sw cq overflow cqid %u\n", __func__, cq->cqid); + pr_debug("%s cxgb4 sw cq overflow cqid %u\n", + __func__, cq->cqid); cq->error = 1; BUG_ON(1); return NULL; diff --git a/drivers/infiniband/hw/hfi1/Makefile b/drivers/infiniband/hw/hfi1/Makefile index 0cf97a09b64b..88085f65432e 100644 --- a/drivers/infiniband/hw/hfi1/Makefile +++ b/drivers/infiniband/hw/hfi1/Makefile @@ -12,7 +12,7 @@ hfi1-y := affinity.o chip.o device.o driver.o efivar.o \ init.o intr.o mad.o mmu_rb.o pcie.o pio.o pio_copy.o platform.o \ qp.o qsfp.o rc.o ruc.o sdma.o sysfs.o trace.o \ uc.o ud.o user_exp_rcv.o user_pages.o user_sdma.o verbs.o \ - verbs_txreq.o + verbs_txreq.o vnic_main.o vnic_sdma.o hfi1-$(CONFIG_DEBUG_FS) += debugfs.o CFLAGS_trace.o = -I$(src) diff --git a/drivers/infiniband/hw/hfi1/aspm.h b/drivers/infiniband/hw/hfi1/aspm.h index 0d58fe3b49b5..794e6814a531 100644 --- a/drivers/infiniband/hw/hfi1/aspm.h +++ b/drivers/infiniband/hw/hfi1/aspm.h @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -229,14 +229,17 @@ static inline void aspm_ctx_timer_function(unsigned long data) spin_unlock_irqrestore(&rcd->aspm_lock, flags); } -/* Disable interrupt processing for verbs contexts when PSM contexts are open */ +/* + * Disable interrupt processing for verbs contexts when PSM or VNIC contexts + * are open. + */ static inline void aspm_disable_all(struct hfi1_devdata *dd) { struct hfi1_ctxtdata *rcd; unsigned long flags; unsigned i; - for (i = 0; i < dd->first_user_ctxt; i++) { + for (i = 0; i < dd->first_dyn_alloc_ctxt; i++) { rcd = dd->rcd[i]; del_timer_sync(&rcd->aspm_timer); spin_lock_irqsave(&rcd->aspm_lock, flags); @@ -260,7 +263,7 @@ static inline void aspm_enable_all(struct hfi1_devdata *dd) if (aspm_mode != ASPM_MODE_DYNAMIC) return; - for (i = 0; i < dd->first_user_ctxt; i++) { + for (i = 0; i < dd->first_dyn_alloc_ctxt; i++) { rcd = dd->rcd[i]; spin_lock_irqsave(&rcd->aspm_lock, flags); rcd->aspm_intr_enable = true; @@ -276,7 +279,7 @@ static inline void aspm_ctx_init(struct hfi1_ctxtdata *rcd) (unsigned long)rcd); rcd->aspm_intr_supported = rcd->dd->aspm_supported && aspm_mode == ASPM_MODE_DYNAMIC && - rcd->ctxt < rcd->dd->first_user_ctxt; + rcd->ctxt < rcd->dd->first_dyn_alloc_ctxt; } static inline void aspm_init(struct hfi1_devdata *dd) @@ -286,7 +289,7 @@ static inline void aspm_init(struct hfi1_devdata *dd) spin_lock_init(&dd->aspm_lock); dd->aspm_supported = aspm_hw_l1_supported(dd); - for (i = 0; i < dd->first_user_ctxt; i++) + for (i = 0; i < dd->first_dyn_alloc_ctxt; i++) aspm_ctx_init(dd->rcd[i]); /* Start with ASPM disabled */ diff --git a/drivers/infiniband/hw/hfi1/chip.c b/drivers/infiniband/hw/hfi1/chip.c index d037f72e4d96..39279fd630bc 100644 --- a/drivers/infiniband/hw/hfi1/chip.c +++ b/drivers/infiniband/hw/hfi1/chip.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015 - 2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -64,6 +64,7 @@ #include "platform.h" #include "aspm.h" #include "affinity.h" +#include "debugfs.h" #define NUM_IB_PORTS 1 @@ -125,9 +126,16 @@ struct flag_table { #define DEFAULT_KRCVQS 2 #define MIN_KERNEL_KCTXTS 2 #define FIRST_KERNEL_KCTXT 1 -/* sizes for both the QP and RSM map tables */ -#define NUM_MAP_ENTRIES 256 -#define NUM_MAP_REGS 32 + +/* + * RSM instance allocation + * 0 - Verbs + * 1 - User Fecn Handling + * 2 - Vnic + */ +#define RSM_INS_VERBS 0 +#define RSM_INS_FECN 1 +#define RSM_INS_VNIC 2 /* Bit offset into the GUID which carries HFI id information */ #define GUID_HFI_INDEX_SHIFT 39 @@ -138,8 +146,7 @@ struct flag_table { #define is_emulator_p(dd) ((((dd)->irev) & 0xf) == 3) #define is_emulator_s(dd) ((((dd)->irev) & 0xf) == 4) -/* RSM fields */ - +/* RSM fields for Verbs */ /* packet type */ #define IB_PACKET_TYPE 2ull #define QW_SHIFT 6ull @@ -169,6 +176,28 @@ struct flag_table { /* QPN[m+n:1] QW 1, OFFSET 1 */ #define QPN_SELECT_OFFSET ((1ull << QW_SHIFT) | (1ull)) +/* RSM fields for Vnic */ +/* L2_TYPE: QW 0, OFFSET 61 - for match */ +#define L2_TYPE_QW 0ull +#define L2_TYPE_BIT_OFFSET 61ull +#define L2_TYPE_OFFSET(off) ((L2_TYPE_QW << QW_SHIFT) | (off)) +#define L2_TYPE_MATCH_OFFSET L2_TYPE_OFFSET(L2_TYPE_BIT_OFFSET) +#define L2_TYPE_MASK 3ull +#define L2_16B_VALUE 2ull + +/* L4_TYPE QW 1, OFFSET 0 - for match */ +#define L4_TYPE_QW 1ull +#define L4_TYPE_BIT_OFFSET 0ull +#define L4_TYPE_OFFSET(off) ((L4_TYPE_QW << QW_SHIFT) | (off)) +#define L4_TYPE_MATCH_OFFSET L4_TYPE_OFFSET(L4_TYPE_BIT_OFFSET) +#define L4_16B_TYPE_MASK 0xFFull +#define L4_16B_ETH_VALUE 0x78ull + +/* 16B VESWID - for select */ +#define L4_16B_HDR_VESWID_OFFSET ((2 << QW_SHIFT) | (16ull)) +/* 16B ENTROPY - for select */ +#define L2_16B_ENTROPY_OFFSET ((1 << QW_SHIFT) | (32ull)) + /* defines to build power on SC2VL table */ #define SC2VL_VAL( \ num, \ @@ -1045,6 +1074,8 @@ static void dc_start(struct hfi1_devdata *); static int qos_rmt_entries(struct hfi1_devdata *dd, unsigned int *mp, unsigned int *np); static void clear_full_mgmt_pkey(struct hfi1_pportdata *ppd); +static int wait_link_transfer_active(struct hfi1_devdata *dd, int wait_ms); +static void clear_rsm_rule(struct hfi1_devdata *dd, u8 rule_index); /* * Error interrupt table entry. This is used as input to the interrupt @@ -6379,18 +6410,17 @@ static void lcb_shutdown(struct hfi1_devdata *dd, int abort) * * The expectation is that the caller of this routine would have taken * care of properly transitioning the link into the correct state. + * NOTE: the caller needs to acquire the dd->dc8051_lock lock + * before calling this function. */ -static void dc_shutdown(struct hfi1_devdata *dd) +static void _dc_shutdown(struct hfi1_devdata *dd) { - unsigned long flags; + lockdep_assert_held(&dd->dc8051_lock); - spin_lock_irqsave(&dd->dc8051_lock, flags); - if (dd->dc_shutdown) { - spin_unlock_irqrestore(&dd->dc8051_lock, flags); + if (dd->dc_shutdown) return; - } + dd->dc_shutdown = 1; - spin_unlock_irqrestore(&dd->dc8051_lock, flags); /* Shutdown the LCB */ lcb_shutdown(dd, 1); /* @@ -6401,35 +6431,45 @@ static void dc_shutdown(struct hfi1_devdata *dd) write_csr(dd, DC_DC8051_CFG_RST, 0x1); } +static void dc_shutdown(struct hfi1_devdata *dd) +{ + mutex_lock(&dd->dc8051_lock); + _dc_shutdown(dd); + mutex_unlock(&dd->dc8051_lock); +} + /* * Calling this after the DC has been brought out of reset should not * do any damage. + * NOTE: the caller needs to acquire the dd->dc8051_lock lock + * before calling this function. */ -static void dc_start(struct hfi1_devdata *dd) +static void _dc_start(struct hfi1_devdata *dd) { - unsigned long flags; - int ret; + lockdep_assert_held(&dd->dc8051_lock); - spin_lock_irqsave(&dd->dc8051_lock, flags); if (!dd->dc_shutdown) - goto done; - spin_unlock_irqrestore(&dd->dc8051_lock, flags); + return; + /* Take the 8051 out of reset */ write_csr(dd, DC_DC8051_CFG_RST, 0ull); /* Wait until 8051 is ready */ - ret = wait_fm_ready(dd, TIMEOUT_8051_START); - if (ret) { + if (wait_fm_ready(dd, TIMEOUT_8051_START)) dd_dev_err(dd, "%s: timeout starting 8051 firmware\n", __func__); - } + /* Take away reset for LCB and RX FPE (set in lcb_shutdown). */ write_csr(dd, DCC_CFG_RESET, 0x10); /* lcb_shutdown() with abort=1 does not restore these */ write_csr(dd, DC_LCB_ERR_EN, dd->lcb_err_en); - spin_lock_irqsave(&dd->dc8051_lock, flags); dd->dc_shutdown = 0; -done: - spin_unlock_irqrestore(&dd->dc8051_lock, flags); +} + +static void dc_start(struct hfi1_devdata *dd) +{ + mutex_lock(&dd->dc8051_lock); + _dc_start(dd); + mutex_unlock(&dd->dc8051_lock); } /* @@ -6701,7 +6741,13 @@ static void rxe_kernel_unfreeze(struct hfi1_devdata *dd) int i; /* enable all kernel contexts */ - for (i = 0; i < dd->n_krcv_queues; i++) { + for (i = 0; i < dd->num_rcv_contexts; i++) { + struct hfi1_ctxtdata *rcd = dd->rcd[i]; + + /* Ensure all non-user contexts(including vnic) are enabled */ + if (!rcd || !rcd->sc || (rcd->sc->type == SC_USER)) + continue; + rcvmask = HFI1_RCVCTRL_CTXT_ENB; /* HFI1_RCVCTRL_TAILUPD_[ENB|DIS] needs to be set explicitly */ rcvmask |= HFI1_CAP_KGET_MASK(dd->rcd[i]->flags, DMA_RTAIL) ? @@ -7077,7 +7123,7 @@ static void add_full_mgmt_pkey(struct hfi1_pportdata *ppd) { struct hfi1_devdata *dd = ppd->dd; - /* Sanity check - ppd->pkeys[2] should be 0, or already initalized */ + /* Sanity check - ppd->pkeys[2] should be 0, or already initialized */ if (!((ppd->pkeys[2] == 0) || (ppd->pkeys[2] == FULL_MGMT_P_KEY))) dd_dev_warn(dd, "%s pkey[2] already set to 0x%x, resetting it to 0x%x\n", __func__, ppd->pkeys[2], FULL_MGMT_P_KEY); @@ -7165,7 +7211,7 @@ static void get_link_widths(struct hfi1_devdata *dd, u16 *tx_width, * set the max_rate field in handle_verify_cap until v0.19. */ if ((dd->icode == ICODE_RTL_SILICON) && - (dd->dc8051_ver < dc8051_ver(0, 19))) { + (dd->dc8051_ver < dc8051_ver(0, 19, 0))) { /* max_rate: 0 = 12.5G, 1 = 25G */ switch (max_rate) { case 0: @@ -7277,15 +7323,6 @@ void handle_verify_cap(struct work_struct *work) lcb_shutdown(dd, 0); adjust_lcb_for_fpga_serdes(dd); - /* - * These are now valid: - * remote VerifyCap fields in the general LNI config - * CSR DC8051_STS_REMOTE_GUID - * CSR DC8051_STS_REMOTE_NODE_TYPE - * CSR DC8051_STS_REMOTE_FM_SECURITY - * CSR DC8051_STS_REMOTE_PORT_NO - */ - read_vc_remote_phy(dd, &power_management, &continious); read_vc_remote_fabric(dd, &vau, &z, &vcu, &vl15buf, &partner_supported_crc); @@ -7350,7 +7387,7 @@ void handle_verify_cap(struct work_struct *work) } ppd->link_speed_active = 0; /* invalid value */ - if (dd->dc8051_ver < dc8051_ver(0, 20)) { + if (dd->dc8051_ver < dc8051_ver(0, 20, 0)) { /* remote_tx_rate: 0 = 12.5G, 1 = 25G */ switch (remote_tx_rate) { case 0: @@ -7416,20 +7453,6 @@ void handle_verify_cap(struct work_struct *work) write_csr(dd, DC_LCB_ERR_EN, 0); /* mask LCB errors */ set_8051_lcb_access(dd); - ppd->neighbor_guid = - read_csr(dd, DC_DC8051_STS_REMOTE_GUID); - ppd->neighbor_port_number = read_csr(dd, DC_DC8051_STS_REMOTE_PORT_NO) & - DC_DC8051_STS_REMOTE_PORT_NO_VAL_SMASK; - ppd->neighbor_type = - read_csr(dd, DC_DC8051_STS_REMOTE_NODE_TYPE) & - DC_DC8051_STS_REMOTE_NODE_TYPE_VAL_MASK; - ppd->neighbor_fm_security = - read_csr(dd, DC_DC8051_STS_REMOTE_FM_SECURITY) & - DC_DC8051_STS_LOCAL_FM_SECURITY_DISABLED_MASK; - dd_dev_info(dd, - "Neighbor Guid: %llx Neighbor type %d MgmtAllowed %d FM security bypass %d\n", - ppd->neighbor_guid, ppd->neighbor_type, - ppd->mgmt_allowed, ppd->neighbor_fm_security); if (ppd->mgmt_allowed) add_full_mgmt_pkey(ppd); @@ -7897,6 +7920,9 @@ static void handle_dcc_err(struct hfi1_devdata *dd, u32 unused, u64 reg) reg &= ~DCC_ERR_FLG_EN_CSR_ACCESS_BLOCKED_HOST_SMASK; } + if (unlikely(hfi1_dbg_fault_suppress_err(&dd->verbs_dev))) + reg &= ~DCC_ERR_FLG_LATE_EBP_ERR_SMASK; + /* report any remaining errors */ if (reg) dd_dev_info_ratelimited(dd, "DCC Error: %s\n", @@ -7995,7 +8021,9 @@ static void is_rcv_avail_int(struct hfi1_devdata *dd, unsigned int source) if (likely(source < dd->num_rcv_contexts)) { rcd = dd->rcd[source]; if (rcd) { - if (source < dd->first_user_ctxt) + /* Check for non-user contexts, including vnic */ + if ((source < dd->first_dyn_alloc_ctxt) || + (rcd->sc && (rcd->sc->type == SC_KERNEL))) rcd->do_interrupt(rcd, 0); else handle_user_interrupt(rcd); @@ -8023,7 +8051,8 @@ static void is_rcv_urgent_int(struct hfi1_devdata *dd, unsigned int source) rcd = dd->rcd[source]; if (rcd) { /* only pay attention to user urgent interrupts */ - if (source >= dd->first_user_ctxt) + if ((source >= dd->first_dyn_alloc_ctxt) && + (!rcd->sc || (rcd->sc->type == SC_USER))) handle_user_interrupt(rcd); return; /* OK */ } @@ -8156,10 +8185,10 @@ static irqreturn_t sdma_interrupt(int irq, void *data) /* handle the interrupt(s) */ sdma_engine_interrupt(sde, status); - } else + } else { dd_dev_err(dd, "SDMA engine %u interrupt, but no status bits set\n", sde->this_idx); - + } return IRQ_HANDLED; } @@ -8344,6 +8373,52 @@ static int read_lcb_via_8051(struct hfi1_devdata *dd, u32 addr, u64 *data) } /* + * Provide a cache for some of the LCB registers in case the LCB is + * unavailable. + * (The LCB is unavailable in certain link states, for example.) + */ +struct lcb_datum { + u32 off; + u64 val; +}; + +static struct lcb_datum lcb_cache[] = { + { DC_LCB_ERR_INFO_RX_REPLAY_CNT, 0}, + { DC_LCB_ERR_INFO_SEQ_CRC_CNT, 0 }, + { DC_LCB_ERR_INFO_REINIT_FROM_PEER_CNT, 0 }, +}; + +static void update_lcb_cache(struct hfi1_devdata *dd) +{ + int i; + int ret; + u64 val; + + for (i = 0; i < ARRAY_SIZE(lcb_cache); i++) { + ret = read_lcb_csr(dd, lcb_cache[i].off, &val); + + /* Update if we get good data */ + if (likely(ret != -EBUSY)) + lcb_cache[i].val = val; + } +} + +static int read_lcb_cache(u32 off, u64 *val) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(lcb_cache); i++) { + if (lcb_cache[i].off == off) { + *val = lcb_cache[i].val; + return 0; + } + } + + pr_warn("%s bad offset 0x%x\n", __func__, off); + return -1; +} + +/* * Read an LCB CSR. Access may not be in host control, so check. * Return 0 on success, -EBUSY on failure. */ @@ -8354,9 +8429,13 @@ int read_lcb_csr(struct hfi1_devdata *dd, u32 addr, u64 *data) /* if up, go through the 8051 for the value */ if (ppd->host_link_state & HLS_UP) return read_lcb_via_8051(dd, addr, data); - /* if going up or down, no access */ - if (ppd->host_link_state & (HLS_GOING_UP | HLS_GOING_OFFLINE)) - return -EBUSY; + /* if going up or down, check the cache, otherwise, no access */ + if (ppd->host_link_state & (HLS_GOING_UP | HLS_GOING_OFFLINE)) { + if (read_lcb_cache(addr, data)) + return -EBUSY; + return 0; + } + /* otherwise, host has access */ *data = read_csr(dd, addr); return 0; @@ -8371,7 +8450,7 @@ static int write_lcb_via_8051(struct hfi1_devdata *dd, u32 addr, u64 data) int ret; if (dd->icode == ICODE_FUNCTIONAL_SIMULATOR || - (dd->dc8051_ver < dc8051_ver(0, 20))) { + (dd->dc8051_ver < dc8051_ver(0, 20, 0))) { if (acquire_lcb_access(dd, 0) == 0) { write_csr(dd, addr, data); release_lcb_access(dd, 0); @@ -8420,16 +8499,11 @@ static int do_8051_command( { u64 reg, completed; int return_code; - unsigned long flags; unsigned long timeout; hfi1_cdbg(DC8051, "type %d, data 0x%012llx", type, in_data); - /* - * Alternative to holding the lock for a long time: - * - keep busy wait - have other users bounce off - */ - spin_lock_irqsave(&dd->dc8051_lock, flags); + mutex_lock(&dd->dc8051_lock); /* We can't send any commands to the 8051 if it's in reset */ if (dd->dc_shutdown) { @@ -8455,10 +8529,8 @@ static int do_8051_command( return_code = -ENXIO; goto fail; } - spin_unlock_irqrestore(&dd->dc8051_lock, flags); - dc_shutdown(dd); - dc_start(dd); - spin_lock_irqsave(&dd->dc8051_lock, flags); + _dc_shutdown(dd); + _dc_start(dd); } /* @@ -8539,8 +8611,7 @@ static int do_8051_command( write_csr(dd, DC_DC8051_CFG_HOST_CMD_0, 0); fail: - spin_unlock_irqrestore(&dd->dc8051_lock, flags); - + mutex_unlock(&dd->dc8051_lock); return return_code; } @@ -8677,13 +8748,20 @@ static void read_remote_device_id(struct hfi1_devdata *dd, u16 *device_id, & REMOTE_DEVICE_REV_MASK; } -void read_misc_status(struct hfi1_devdata *dd, u8 *ver_a, u8 *ver_b) +void read_misc_status(struct hfi1_devdata *dd, u8 *ver_major, u8 *ver_minor, + u8 *ver_patch) { u32 frame; read_8051_config(dd, MISC_STATUS, GENERAL_CONFIG, &frame); - *ver_a = (frame >> STS_FM_VERSION_A_SHIFT) & STS_FM_VERSION_A_MASK; - *ver_b = (frame >> STS_FM_VERSION_B_SHIFT) & STS_FM_VERSION_B_MASK; + *ver_major = (frame >> STS_FM_VERSION_MAJOR_SHIFT) & + STS_FM_VERSION_MAJOR_MASK; + *ver_minor = (frame >> STS_FM_VERSION_MINOR_SHIFT) & + STS_FM_VERSION_MINOR_MASK; + + read_8051_config(dd, VERSION_PATCH, GENERAL_CONFIG, &frame); + *ver_patch = (frame >> STS_FM_VERSION_PATCH_SHIFT) & + STS_FM_VERSION_PATCH_MASK; } static void read_vc_remote_phy(struct hfi1_devdata *dd, u8 *power_management, @@ -8891,8 +8969,6 @@ int send_idle_sma(struct hfi1_devdata *dd, u64 message) */ static int do_quick_linkup(struct hfi1_devdata *dd) { - u64 reg; - unsigned long timeout; int ret; lcb_shutdown(dd, 0); @@ -8915,19 +8991,9 @@ static int do_quick_linkup(struct hfi1_devdata *dd) write_csr(dd, DC_LCB_CFG_RUN, 1ull << DC_LCB_CFG_RUN_EN_SHIFT); - /* watch LCB_STS_LINK_TRANSFER_ACTIVE */ - timeout = jiffies + msecs_to_jiffies(10); - while (1) { - reg = read_csr(dd, DC_LCB_STS_LINK_TRANSFER_ACTIVE); - if (reg) - break; - if (time_after(jiffies, timeout)) { - dd_dev_err(dd, - "timeout waiting for LINK_TRANSFER_ACTIVE\n"); - return -ETIMEDOUT; - } - udelay(2); - } + ret = wait_link_transfer_active(dd, 10); + if (ret) + return ret; write_csr(dd, DC_LCB_CFG_ALLOW_LINK_UP, 1ull << DC_LCB_CFG_ALLOW_LINK_UP_VAL_SHIFT); @@ -9091,7 +9157,7 @@ static int set_local_link_attributes(struct hfi1_pportdata *ppd) if (ret) goto set_local_link_attributes_fail; - if (dd->dc8051_ver < dc8051_ver(0, 20)) { + if (dd->dc8051_ver < dc8051_ver(0, 20, 0)) { /* set the tx rate to the fastest enabled */ if (ppd->link_speed_enabled & OPA_LINK_SPEED_25G) ppd->local_tx_rate = 1; @@ -9274,7 +9340,7 @@ static int handle_qsfp_error_conditions(struct hfi1_pportdata *ppd, if ((qsfp_interrupt_status[0] & QSFP_HIGH_TEMP_ALARM) || (qsfp_interrupt_status[0] & QSFP_HIGH_TEMP_WARNING)) - dd_dev_info(dd, "%s: QSFP cable on fire\n", + dd_dev_info(dd, "%s: QSFP cable temperature too high\n", __func__); if ((qsfp_interrupt_status[0] & QSFP_LOW_TEMP_ALARM) || @@ -9494,8 +9560,11 @@ static int test_qsfp_read(struct hfi1_pportdata *ppd) int ret; u8 status; - /* report success if not a QSFP */ - if (ppd->port_type != PORT_TYPE_QSFP) + /* + * Report success if not a QSFP or, if it is a QSFP, but the cable is + * not present + */ + if (ppd->port_type != PORT_TYPE_QSFP || !qsfp_mod_present(ppd)) return 0; /* read byte 2, the status byte */ @@ -10082,6 +10151,64 @@ static void check_lni_states(struct hfi1_pportdata *ppd) decode_state_complete(ppd, last_remote_state, "received"); } +/* wait for wait_ms for LINK_TRANSFER_ACTIVE to go to 1 */ +static int wait_link_transfer_active(struct hfi1_devdata *dd, int wait_ms) +{ + u64 reg; + unsigned long timeout; + + /* watch LCB_STS_LINK_TRANSFER_ACTIVE */ + timeout = jiffies + msecs_to_jiffies(wait_ms); + while (1) { + reg = read_csr(dd, DC_LCB_STS_LINK_TRANSFER_ACTIVE); + if (reg) + break; + if (time_after(jiffies, timeout)) { + dd_dev_err(dd, + "timeout waiting for LINK_TRANSFER_ACTIVE\n"); + return -ETIMEDOUT; + } + udelay(2); + } + return 0; +} + +/* called when the logical link state is not down as it should be */ +static void force_logical_link_state_down(struct hfi1_pportdata *ppd) +{ + struct hfi1_devdata *dd = ppd->dd; + + /* + * Bring link up in LCB loopback + */ + write_csr(dd, DC_LCB_CFG_TX_FIFOS_RESET, 1); + write_csr(dd, DC_LCB_CFG_IGNORE_LOST_RCLK, + DC_LCB_CFG_IGNORE_LOST_RCLK_EN_SMASK); + + write_csr(dd, DC_LCB_CFG_LANE_WIDTH, 0); + write_csr(dd, DC_LCB_CFG_REINIT_AS_SLAVE, 0); + write_csr(dd, DC_LCB_CFG_CNT_FOR_SKIP_STALL, 0x110); + write_csr(dd, DC_LCB_CFG_LOOPBACK, 0x2); + + write_csr(dd, DC_LCB_CFG_TX_FIFOS_RESET, 0); + (void)read_csr(dd, DC_LCB_CFG_TX_FIFOS_RESET); + udelay(3); + write_csr(dd, DC_LCB_CFG_ALLOW_LINK_UP, 1); + write_csr(dd, DC_LCB_CFG_RUN, 1ull << DC_LCB_CFG_RUN_EN_SHIFT); + + wait_link_transfer_active(dd, 100); + + /* + * Bring the link down again. + */ + write_csr(dd, DC_LCB_CFG_TX_FIFOS_RESET, 1); + write_csr(dd, DC_LCB_CFG_ALLOW_LINK_UP, 0); + write_csr(dd, DC_LCB_CFG_IGNORE_LOST_RCLK, 0); + + /* call again to adjust ppd->statusp, if needed */ + get_logical_state(ppd); +} + /* * Helper for set_link_state(). Do not call except from that routine. * Expects ppd->hls_mutex to be held. @@ -10098,6 +10225,8 @@ static int goto_offline(struct hfi1_pportdata *ppd, u8 rem_reason) int do_transition; int do_wait; + update_lcb_cache(dd); + previous_state = ppd->host_link_state; ppd->host_link_state = HLS_GOING_OFFLINE; pstate = read_physical_state(dd); @@ -10135,15 +10264,18 @@ static int goto_offline(struct hfi1_pportdata *ppd, u8 rem_reason) return ret; } - /* make sure the logical state is also down */ - wait_logical_linkstate(ppd, IB_PORT_DOWN, 1000); - /* * Now in charge of LCB - must be after the physical state is * offline.quiet and before host_link_state is changed. */ set_host_lcb_access(dd); write_csr(dd, DC_LCB_ERR_EN, ~0ull); /* watch LCB errors */ + + /* make sure the logical state is also down */ + ret = wait_logical_linkstate(ppd, IB_PORT_DOWN, 1000); + if (ret) + force_logical_link_state_down(ppd); + ppd->host_link_state = HLS_LINK_COOLDOWN; /* LCB access allowed */ if (ppd->port_type == PORT_TYPE_QSFP && @@ -10380,11 +10512,8 @@ int set_link_state(struct hfi1_pportdata *ppd, u32 state) goto unexpected; } - ppd->host_link_state = HLS_UP_INIT; ret = wait_logical_linkstate(ppd, IB_PORT_INIT, 1000); if (ret) { - /* logical state didn't change, stay at going_up */ - ppd->host_link_state = HLS_GOING_UP; dd_dev_err(dd, "%s: logical state did not change to INIT\n", __func__); @@ -10398,6 +10527,7 @@ int set_link_state(struct hfi1_pportdata *ppd, u32 state) add_rcvctrl(dd, RCV_CTRL_RCV_PORT_ENABLE_SMASK); handle_linkup_change(dd, 1); + ppd->host_link_state = HLS_UP_INIT; } break; case HLS_UP_ARMED: @@ -11853,6 +11983,10 @@ static void free_cntrs(struct hfi1_devdata *dd) dd->scntrs = NULL; kfree(dd->cntrnames); dd->cntrnames = NULL; + if (dd->update_cntr_wq) { + destroy_workqueue(dd->update_cntr_wq); + dd->update_cntr_wq = NULL; + } } static u64 read_dev_port_cntr(struct hfi1_devdata *dd, struct cntr_entry *entry, @@ -12008,7 +12142,7 @@ u64 write_port_cntr(struct hfi1_pportdata *ppd, int index, int vl, u64 data) return write_dev_port_cntr(ppd->dd, entry, sval, ppd, vl, data); } -static void update_synth_timer(unsigned long opaque) +static void do_update_synth_timer(struct work_struct *work) { u64 cur_tx; u64 cur_rx; @@ -12017,8 +12151,8 @@ static void update_synth_timer(unsigned long opaque) int i, j, vl; struct hfi1_pportdata *ppd; struct cntr_entry *entry; - - struct hfi1_devdata *dd = (struct hfi1_devdata *)opaque; + struct hfi1_devdata *dd = container_of(work, struct hfi1_devdata, + update_cntr_work); /* * Rather than keep beating on the CSRs pick a minimal set that we can @@ -12101,7 +12235,13 @@ static void update_synth_timer(unsigned long opaque) } else { hfi1_cdbg(CNTR, "[%d] No update necessary", dd->unit); } +} +static void update_synth_timer(unsigned long opaque) +{ + struct hfi1_devdata *dd = (struct hfi1_devdata *)opaque; + + queue_work(dd->update_cntr_wq, &dd->update_cntr_work); mod_timer(&dd->synth_stats_timer, jiffies + HZ * SYNTH_CNT_TIME); } @@ -12337,6 +12477,13 @@ static int init_cntrs(struct hfi1_devdata *dd) if (init_cpu_counters(dd)) goto bail; + dd->update_cntr_wq = alloc_ordered_workqueue("hfi1_update_cntr_%d", + WQ_MEM_RECLAIM, dd->unit); + if (!dd->update_cntr_wq) + goto bail; + + INIT_WORK(&dd->update_cntr_work, do_update_synth_timer); + mod_timer(&dd->synth_stats_timer, jiffies + HZ * SYNTH_CNT_TIME); return 0; bail: @@ -12726,7 +12873,10 @@ static int request_msix_irqs(struct hfi1_devdata *dd) first_sdma = last_general; last_sdma = first_sdma + dd->num_sdma; first_rx = last_sdma; - last_rx = first_rx + dd->n_krcv_queues; + last_rx = first_rx + dd->n_krcv_queues + HFI1_NUM_VNIC_CTXT; + + /* VNIC MSIx interrupts get mapped when VNIC contexts are created */ + dd->first_dyn_msix_idx = first_rx + dd->n_krcv_queues; /* * Sanity check - the code expects all SDMA chip source @@ -12740,7 +12890,7 @@ static int request_msix_irqs(struct hfi1_devdata *dd) const char *err_info; irq_handler_t handler; irq_handler_t thread = NULL; - void *arg; + void *arg = NULL; int idx; struct hfi1_ctxtdata *rcd = NULL; struct sdma_engine *sde = NULL; @@ -12767,24 +12917,25 @@ static int request_msix_irqs(struct hfi1_devdata *dd) } else if (first_rx <= i && i < last_rx) { idx = i - first_rx; rcd = dd->rcd[idx]; - /* no interrupt if no rcd */ - if (!rcd) - continue; - /* - * Set the interrupt register and mask for this - * context's interrupt. - */ - rcd->ireg = (IS_RCVAVAIL_START + idx) / 64; - rcd->imask = ((u64)1) << - ((IS_RCVAVAIL_START + idx) % 64); - handler = receive_context_interrupt; - thread = receive_context_thread; - arg = rcd; - snprintf(me->name, sizeof(me->name), - DRIVER_NAME "_%d kctxt%d", dd->unit, idx); - err_info = "receive context"; - remap_intr(dd, IS_RCVAVAIL_START + idx, i); - me->type = IRQ_RCVCTXT; + if (rcd) { + /* + * Set the interrupt register and mask for this + * context's interrupt. + */ + rcd->ireg = (IS_RCVAVAIL_START + idx) / 64; + rcd->imask = ((u64)1) << + ((IS_RCVAVAIL_START + idx) % 64); + handler = receive_context_interrupt; + thread = receive_context_thread; + arg = rcd; + snprintf(me->name, sizeof(me->name), + DRIVER_NAME "_%d kctxt%d", + dd->unit, idx); + err_info = "receive context"; + remap_intr(dd, IS_RCVAVAIL_START + idx, i); + me->type = IRQ_RCVCTXT; + rcd->msix_intr = i; + } } else { /* not in our expected range - complain, then * ignore it @@ -12822,6 +12973,84 @@ static int request_msix_irqs(struct hfi1_devdata *dd) return ret; } +void hfi1_vnic_synchronize_irq(struct hfi1_devdata *dd) +{ + int i; + + if (!dd->num_msix_entries) { + synchronize_irq(dd->pcidev->irq); + return; + } + + for (i = 0; i < dd->vnic.num_ctxt; i++) { + struct hfi1_ctxtdata *rcd = dd->vnic.ctxt[i]; + struct hfi1_msix_entry *me = &dd->msix_entries[rcd->msix_intr]; + + synchronize_irq(me->msix.vector); + } +} + +void hfi1_reset_vnic_msix_info(struct hfi1_ctxtdata *rcd) +{ + struct hfi1_devdata *dd = rcd->dd; + struct hfi1_msix_entry *me = &dd->msix_entries[rcd->msix_intr]; + + if (!me->arg) /* => no irq, no affinity */ + return; + + hfi1_put_irq_affinity(dd, me); + free_irq(me->msix.vector, me->arg); + + me->arg = NULL; +} + +void hfi1_set_vnic_msix_info(struct hfi1_ctxtdata *rcd) +{ + struct hfi1_devdata *dd = rcd->dd; + struct hfi1_msix_entry *me; + int idx = rcd->ctxt; + void *arg = rcd; + int ret; + + rcd->msix_intr = dd->vnic.msix_idx++; + me = &dd->msix_entries[rcd->msix_intr]; + + /* + * Set the interrupt register and mask for this + * context's interrupt. + */ + rcd->ireg = (IS_RCVAVAIL_START + idx) / 64; + rcd->imask = ((u64)1) << + ((IS_RCVAVAIL_START + idx) % 64); + + snprintf(me->name, sizeof(me->name), + DRIVER_NAME "_%d kctxt%d", dd->unit, idx); + me->name[sizeof(me->name) - 1] = 0; + me->type = IRQ_RCVCTXT; + + remap_intr(dd, IS_RCVAVAIL_START + idx, rcd->msix_intr); + + ret = request_threaded_irq(me->msix.vector, receive_context_interrupt, + receive_context_thread, 0, me->name, arg); + if (ret) { + dd_dev_err(dd, "vnic irq request (vector %d, idx %d) fail %d\n", + me->msix.vector, idx, ret); + return; + } + /* + * assign arg after request_irq call, so it will be + * cleaned up + */ + me->arg = arg; + + ret = hfi1_get_irq_affinity(dd, me); + if (ret) { + dd_dev_err(dd, + "unable to pin IRQ %d\n", ret); + free_irq(me->msix.vector, me->arg); + } +} + /* * Set the general handler to accept all interrupts, remap all * chip interrupts back to MSI-X 0. @@ -12853,7 +13082,7 @@ static int set_up_interrupts(struct hfi1_devdata *dd) * N interrupts - one per used SDMA engine * M interrupt - one per kernel receive context */ - total = 1 + dd->num_sdma + dd->n_krcv_queues; + total = 1 + dd->num_sdma + dd->n_krcv_queues + HFI1_NUM_VNIC_CTXT; entries = kcalloc(total, sizeof(*entries), GFP_KERNEL); if (!entries) { @@ -12918,7 +13147,8 @@ fail: * * num_rcv_contexts - number of contexts being used * n_krcv_queues - number of kernel contexts - * first_user_ctxt - first non-kernel context in array of contexts + * first_dyn_alloc_ctxt - first dynamically allocated context + * in array of contexts * freectxts - number of free user contexts * num_send_contexts - number of PIO send contexts being used */ @@ -12995,10 +13225,14 @@ static int set_up_context_variables(struct hfi1_devdata *dd) total_contexts = num_kernel_contexts + num_user_contexts; } - /* the first N are kernel contexts, the rest are user contexts */ + /* Accommodate VNIC contexts */ + if ((total_contexts + HFI1_NUM_VNIC_CTXT) <= dd->chip_rcv_contexts) + total_contexts += HFI1_NUM_VNIC_CTXT; + + /* the first N are kernel contexts, the rest are user/vnic contexts */ dd->num_rcv_contexts = total_contexts; dd->n_krcv_queues = num_kernel_contexts; - dd->first_user_ctxt = num_kernel_contexts; + dd->first_dyn_alloc_ctxt = num_kernel_contexts; dd->num_user_contexts = num_user_contexts; dd->freectxts = num_user_contexts; dd_dev_info(dd, @@ -13454,11 +13688,8 @@ static void reset_rxe_csrs(struct hfi1_devdata *dd) write_csr(dd, RCV_COUNTER_ARRAY32 + (8 * i), 0); for (i = 0; i < RXE_NUM_64_BIT_COUNTERS; i++) write_csr(dd, RCV_COUNTER_ARRAY64 + (8 * i), 0); - for (i = 0; i < RXE_NUM_RSM_INSTANCES; i++) { - write_csr(dd, RCV_RSM_CFG + (8 * i), 0); - write_csr(dd, RCV_RSM_SELECT + (8 * i), 0); - write_csr(dd, RCV_RSM_MATCH + (8 * i), 0); - } + for (i = 0; i < RXE_NUM_RSM_INSTANCES; i++) + clear_rsm_rule(dd, i); for (i = 0; i < 32; i++) write_csr(dd, RCV_RSM_MAP_TABLE + (8 * i), 0); @@ -13817,6 +14048,16 @@ static void add_rsm_rule(struct hfi1_devdata *dd, u8 rule_index, (u64)rrd->value2 << RCV_RSM_MATCH_VALUE2_SHIFT); } +/* + * Clear a receive side mapping rule. + */ +static void clear_rsm_rule(struct hfi1_devdata *dd, u8 rule_index) +{ + write_csr(dd, RCV_RSM_CFG + (8 * rule_index), 0); + write_csr(dd, RCV_RSM_SELECT + (8 * rule_index), 0); + write_csr(dd, RCV_RSM_MATCH + (8 * rule_index), 0); +} + /* return the number of RSM map table entries that will be used for QOS */ static int qos_rmt_entries(struct hfi1_devdata *dd, unsigned int *mp, unsigned int *np) @@ -13932,7 +14173,7 @@ static void init_qos(struct hfi1_devdata *dd, struct rsm_map_table *rmt) rrd.value2 = LRH_SC_VALUE; /* add rule 0 */ - add_rsm_rule(dd, 0, &rrd); + add_rsm_rule(dd, RSM_INS_VERBS, &rrd); /* mark RSM map entries as used */ rmt->used += rmt_entries; @@ -13962,7 +14203,7 @@ static void init_user_fecn_handling(struct hfi1_devdata *dd, /* * RSM will extract the destination context as an index into the * map table. The destination contexts are a sequential block - * in the range first_user_ctxt...num_rcv_contexts-1 (inclusive). + * in the range first_dyn_alloc_ctxt...num_rcv_contexts-1 (inclusive). * Map entries are accessed as offset + extracted value. Adjust * the added offset so this sequence can be placed anywhere in * the table - as long as the entries themselves do not wrap. @@ -13970,9 +14211,9 @@ static void init_user_fecn_handling(struct hfi1_devdata *dd, * start with that to allow for a "negative" offset. */ offset = (u8)(NUM_MAP_ENTRIES + (int)rmt->used - - (int)dd->first_user_ctxt); + (int)dd->first_dyn_alloc_ctxt); - for (i = dd->first_user_ctxt, idx = rmt->used; + for (i = dd->first_dyn_alloc_ctxt, idx = rmt->used; i < dd->num_rcv_contexts; i++, idx++) { /* replace with identity mapping */ regoff = (idx % 8) * 8; @@ -14006,11 +14247,84 @@ static void init_user_fecn_handling(struct hfi1_devdata *dd, rrd.value2 = 1; /* add rule 1 */ - add_rsm_rule(dd, 1, &rrd); + add_rsm_rule(dd, RSM_INS_FECN, &rrd); rmt->used += dd->num_user_contexts; } +/* Initialize RSM for VNIC */ +void hfi1_init_vnic_rsm(struct hfi1_devdata *dd) +{ + u8 i, j; + u8 ctx_id = 0; + u64 reg; + u32 regoff; + struct rsm_rule_data rrd; + + if (hfi1_vnic_is_rsm_full(dd, NUM_VNIC_MAP_ENTRIES)) { + dd_dev_err(dd, "Vnic RSM disabled, rmt entries used = %d\n", + dd->vnic.rmt_start); + return; + } + + dev_dbg(&(dd)->pcidev->dev, "Vnic rsm start = %d, end %d\n", + dd->vnic.rmt_start, + dd->vnic.rmt_start + NUM_VNIC_MAP_ENTRIES); + + /* Update RSM mapping table, 32 regs, 256 entries - 1 ctx per byte */ + regoff = RCV_RSM_MAP_TABLE + (dd->vnic.rmt_start / 8) * 8; + reg = read_csr(dd, regoff); + for (i = 0; i < NUM_VNIC_MAP_ENTRIES; i++) { + /* Update map register with vnic context */ + j = (dd->vnic.rmt_start + i) % 8; + reg &= ~(0xffllu << (j * 8)); + reg |= (u64)dd->vnic.ctxt[ctx_id++]->ctxt << (j * 8); + /* Wrap up vnic ctx index */ + ctx_id %= dd->vnic.num_ctxt; + /* Write back map register */ + if (j == 7 || ((i + 1) == NUM_VNIC_MAP_ENTRIES)) { + dev_dbg(&(dd)->pcidev->dev, + "Vnic rsm map reg[%d] =0x%llx\n", + regoff - RCV_RSM_MAP_TABLE, reg); + + write_csr(dd, regoff, reg); + regoff += 8; + if (i < (NUM_VNIC_MAP_ENTRIES - 1)) + reg = read_csr(dd, regoff); + } + } + + /* Add rule for vnic */ + rrd.offset = dd->vnic.rmt_start; + rrd.pkt_type = 4; + /* Match 16B packets */ + rrd.field1_off = L2_TYPE_MATCH_OFFSET; + rrd.mask1 = L2_TYPE_MASK; + rrd.value1 = L2_16B_VALUE; + /* Match ETH L4 packets */ + rrd.field2_off = L4_TYPE_MATCH_OFFSET; + rrd.mask2 = L4_16B_TYPE_MASK; + rrd.value2 = L4_16B_ETH_VALUE; + /* Calc context from veswid and entropy */ + rrd.index1_off = L4_16B_HDR_VESWID_OFFSET; + rrd.index1_width = ilog2(NUM_VNIC_MAP_ENTRIES); + rrd.index2_off = L2_16B_ENTROPY_OFFSET; + rrd.index2_width = ilog2(NUM_VNIC_MAP_ENTRIES); + add_rsm_rule(dd, RSM_INS_VNIC, &rrd); + + /* Enable RSM if not already enabled */ + add_rcvctrl(dd, RCV_CTRL_RCV_RSM_ENABLE_SMASK); +} + +void hfi1_deinit_vnic_rsm(struct hfi1_devdata *dd) +{ + clear_rsm_rule(dd, RSM_INS_VNIC); + + /* Disable RSM if used only by vnic */ + if (dd->vnic.rmt_start == 0) + clear_rcvctrl(dd, RCV_CTRL_RCV_RSM_ENABLE_SMASK); +} + static void init_rxe(struct hfi1_devdata *dd) { struct rsm_map_table *rmt; @@ -14023,6 +14337,8 @@ static void init_rxe(struct hfi1_devdata *dd) init_qos(dd, rmt); init_user_fecn_handling(dd, rmt); complete_rsm_map_table(dd, rmt); + /* record number of used rsm map entries for vnic */ + dd->vnic.rmt_start = rmt->used; kfree(rmt); /* diff --git a/drivers/infiniband/hw/hfi1/chip.h b/drivers/infiniband/hw/hfi1/chip.h index 043fd21dc5f3..b9dbf16d7703 100644 --- a/drivers/infiniband/hw/hfi1/chip.h +++ b/drivers/infiniband/hw/hfi1/chip.h @@ -1,7 +1,7 @@ #ifndef _CHIP_H #define _CHIP_H /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015 - 2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -394,7 +394,8 @@ #define LAST_REMOTE_STATE_COMPLETE 0x13 #define LINK_QUALITY_INFO 0x14 #define REMOTE_DEVICE_ID 0x15 -#define LINK_DOWN_REASON 0x16 +#define LINK_DOWN_REASON 0x16 /* first byte of offset 0x16 */ +#define VERSION_PATCH 0x16 /* last byte of offset 0x16 */ /* 8051 lane specific register field IDs */ #define TX_EQ_SETTINGS 0x00 @@ -524,10 +525,12 @@ enum { #define SUPPORTED_CRCS (CAP_CRC_14B | CAP_CRC_48B) /* misc status version fields */ -#define STS_FM_VERSION_A_SHIFT 16 -#define STS_FM_VERSION_A_MASK 0xff -#define STS_FM_VERSION_B_SHIFT 24 -#define STS_FM_VERSION_B_MASK 0xff +#define STS_FM_VERSION_MINOR_SHIFT 16 +#define STS_FM_VERSION_MINOR_MASK 0xff +#define STS_FM_VERSION_MAJOR_SHIFT 24 +#define STS_FM_VERSION_MAJOR_MASK 0xff +#define STS_FM_VERSION_PATCH_SHIFT 24 +#define STS_FM_VERSION_PATCH_MASK 0xff /* LCB_CFG_CRC_MODE TX_VAL and RX_VAL CRC mode values */ #define LCB_CRC_16B 0x0 /* 16b CRC */ @@ -698,7 +701,8 @@ void fabric_serdes_reset(struct hfi1_devdata *dd); int read_8051_data(struct hfi1_devdata *dd, u32 addr, u32 len, u64 *result); /* chip.c */ -void read_misc_status(struct hfi1_devdata *dd, u8 *ver_a, u8 *ver_b); +void read_misc_status(struct hfi1_devdata *dd, u8 *ver_major, u8 *ver_minor, + u8 *ver_patch); void read_guid(struct hfi1_devdata *dd); int wait_fm_ready(struct hfi1_devdata *dd, u32 mstimeout); void set_link_down_reason(struct hfi1_pportdata *ppd, u8 lcl_reason, @@ -1358,6 +1362,8 @@ int hfi1_clear_ctxt_jkey(struct hfi1_devdata *dd, unsigned ctxt); int hfi1_set_ctxt_pkey(struct hfi1_devdata *dd, unsigned ctxt, u16 pkey); int hfi1_clear_ctxt_pkey(struct hfi1_devdata *dd, unsigned ctxt); void hfi1_read_link_quality(struct hfi1_devdata *dd, u8 *link_quality); +void hfi1_init_vnic_rsm(struct hfi1_devdata *dd); +void hfi1_deinit_vnic_rsm(struct hfi1_devdata *dd); /* * Interrupt source table. diff --git a/drivers/infiniband/hw/hfi1/common.h b/drivers/infiniband/hw/hfi1/common.h index 1b783bbee4bb..995d62c7f9a7 100644 --- a/drivers/infiniband/hw/hfi1/common.h +++ b/drivers/infiniband/hw/hfi1/common.h @@ -331,12 +331,15 @@ struct diag_pkt { #define FULL_MGMT_P_KEY 0xFFFF #define DEFAULT_P_KEY LIM_MGMT_P_KEY -#define HFI1_FECN_SHIFT 31 -#define HFI1_FECN_MASK 1 -#define HFI1_FECN_SMASK BIT(HFI1_FECN_SHIFT) -#define HFI1_BECN_SHIFT 30 -#define HFI1_BECN_MASK 1 -#define HFI1_BECN_SMASK BIT(HFI1_BECN_SHIFT) + +/** + * 0xF8 - 4 bits of multicast range and 1 bit for collective range + * Example: For 24 bit LID space, + * Multicast range: 0xF00000 to 0xF7FFFF + * Collective range: 0xF80000 to 0xFFFFFE + */ +#define HFI1_MCAST_NR 0x4 /* Number of top bits set */ +#define HFI1_COLLECTIVE_NR 0x1 /* Number of bits after MCAST_NR */ #define HFI1_PSM_IOC_BASE_SEQ 0x0 diff --git a/drivers/infiniband/hw/hfi1/debugfs.c b/drivers/infiniband/hw/hfi1/debugfs.c index 7fe9dd885746..e9fa3c293e42 100644 --- a/drivers/infiniband/hw/hfi1/debugfs.c +++ b/drivers/infiniband/hw/hfi1/debugfs.c @@ -1,6 +1,6 @@ #ifdef CONFIG_DEBUG_FS /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -51,8 +51,12 @@ #include <linux/export.h> #include <linux/module.h> #include <linux/string.h> +#include <linux/types.h> +#include <linux/ratelimit.h> +#include <linux/fault-inject.h> #include "hfi.h" +#include "trace.h" #include "debugfs.h" #include "device.h" #include "qp.h" @@ -170,7 +174,7 @@ static int _opcode_stats_seq_show(struct seq_file *s, void *v) struct hfi1_ibdev *ibd = (struct hfi1_ibdev *)s->private; struct hfi1_devdata *dd = dd_from_dev(ibd); - for (j = 0; j < dd->first_user_ctxt; j++) { + for (j = 0; j < dd->first_dyn_alloc_ctxt; j++) { if (!dd->rcd[j]) continue; n_packets += dd->rcd[j]->opstats->stats[i].n_packets; @@ -196,7 +200,7 @@ static void *_ctx_stats_seq_start(struct seq_file *s, loff_t *pos) if (!*pos) return SEQ_START_TOKEN; - if (*pos >= dd->first_user_ctxt) + if (*pos >= dd->first_dyn_alloc_ctxt) return NULL; return pos; } @@ -210,7 +214,7 @@ static void *_ctx_stats_seq_next(struct seq_file *s, void *v, loff_t *pos) return pos; ++*pos; - if (*pos >= dd->first_user_ctxt) + if (*pos >= dd->first_dyn_alloc_ctxt) return NULL; return pos; } @@ -1063,6 +1067,222 @@ DEBUGFS_SEQ_FILE_OPS(sdma_cpu_list); DEBUGFS_SEQ_FILE_OPEN(sdma_cpu_list) DEBUGFS_FILE_OPS(sdma_cpu_list); +#ifdef CONFIG_FAULT_INJECTION +static void *_fault_stats_seq_start(struct seq_file *s, loff_t *pos) +{ + struct hfi1_opcode_stats_perctx *opstats; + + if (*pos >= ARRAY_SIZE(opstats->stats)) + return NULL; + return pos; +} + +static void *_fault_stats_seq_next(struct seq_file *s, void *v, loff_t *pos) +{ + struct hfi1_opcode_stats_perctx *opstats; + + ++*pos; + if (*pos >= ARRAY_SIZE(opstats->stats)) + return NULL; + return pos; +} + +static void _fault_stats_seq_stop(struct seq_file *s, void *v) +{ +} + +static int _fault_stats_seq_show(struct seq_file *s, void *v) +{ + loff_t *spos = v; + loff_t i = *spos, j; + u64 n_packets = 0, n_bytes = 0; + struct hfi1_ibdev *ibd = (struct hfi1_ibdev *)s->private; + struct hfi1_devdata *dd = dd_from_dev(ibd); + + for (j = 0; j < dd->first_dyn_alloc_ctxt; j++) { + if (!dd->rcd[j]) + continue; + n_packets += dd->rcd[j]->opstats->stats[i].n_packets; + n_bytes += dd->rcd[j]->opstats->stats[i].n_bytes; + } + if (!n_packets && !n_bytes) + return SEQ_SKIP; + if (!ibd->fault_opcode->n_rxfaults[i] && + !ibd->fault_opcode->n_txfaults[i]) + return SEQ_SKIP; + seq_printf(s, "%02llx %llu/%llu (faults rx:%llu faults: tx:%llu)\n", i, + (unsigned long long)n_packets, + (unsigned long long)n_bytes, + (unsigned long long)ibd->fault_opcode->n_rxfaults[i], + (unsigned long long)ibd->fault_opcode->n_txfaults[i]); + return 0; +} + +DEBUGFS_SEQ_FILE_OPS(fault_stats); +DEBUGFS_SEQ_FILE_OPEN(fault_stats); +DEBUGFS_FILE_OPS(fault_stats); + +static void fault_exit_opcode_debugfs(struct hfi1_ibdev *ibd) +{ + debugfs_remove_recursive(ibd->fault_opcode->dir); + kfree(ibd->fault_opcode); + ibd->fault_opcode = NULL; +} + +static int fault_init_opcode_debugfs(struct hfi1_ibdev *ibd) +{ + struct dentry *parent = ibd->hfi1_ibdev_dbg; + + ibd->fault_opcode = kzalloc(sizeof(*ibd->fault_opcode), GFP_KERNEL); + if (!ibd->fault_opcode) + return -ENOMEM; + + ibd->fault_opcode->attr.interval = 1; + ibd->fault_opcode->attr.require_end = ULONG_MAX; + ibd->fault_opcode->attr.stacktrace_depth = 32; + ibd->fault_opcode->attr.dname = NULL; + ibd->fault_opcode->attr.verbose = 0; + ibd->fault_opcode->fault_by_opcode = false; + ibd->fault_opcode->opcode = 0; + ibd->fault_opcode->mask = 0xff; + + ibd->fault_opcode->dir = + fault_create_debugfs_attr("fault_opcode", + parent, + &ibd->fault_opcode->attr); + if (IS_ERR(ibd->fault_opcode->dir)) { + kfree(ibd->fault_opcode); + return -ENOENT; + } + + DEBUGFS_SEQ_FILE_CREATE(fault_stats, ibd->fault_opcode->dir, ibd); + if (!debugfs_create_bool("fault_by_opcode", 0600, + ibd->fault_opcode->dir, + &ibd->fault_opcode->fault_by_opcode)) + goto fail; + if (!debugfs_create_x8("opcode", 0600, ibd->fault_opcode->dir, + &ibd->fault_opcode->opcode)) + goto fail; + if (!debugfs_create_x8("mask", 0600, ibd->fault_opcode->dir, + &ibd->fault_opcode->mask)) + goto fail; + + return 0; +fail: + fault_exit_opcode_debugfs(ibd); + return -ENOMEM; +} + +static void fault_exit_packet_debugfs(struct hfi1_ibdev *ibd) +{ + debugfs_remove_recursive(ibd->fault_packet->dir); + kfree(ibd->fault_packet); + ibd->fault_packet = NULL; +} + +static int fault_init_packet_debugfs(struct hfi1_ibdev *ibd) +{ + struct dentry *parent = ibd->hfi1_ibdev_dbg; + + ibd->fault_packet = kzalloc(sizeof(*ibd->fault_packet), GFP_KERNEL); + if (!ibd->fault_packet) + return -ENOMEM; + + ibd->fault_packet->attr.interval = 1; + ibd->fault_packet->attr.require_end = ULONG_MAX; + ibd->fault_packet->attr.stacktrace_depth = 32; + ibd->fault_packet->attr.dname = NULL; + ibd->fault_packet->attr.verbose = 0; + ibd->fault_packet->fault_by_packet = false; + + ibd->fault_packet->dir = + fault_create_debugfs_attr("fault_packet", + parent, + &ibd->fault_opcode->attr); + if (IS_ERR(ibd->fault_packet->dir)) { + kfree(ibd->fault_packet); + return -ENOENT; + } + + if (!debugfs_create_bool("fault_by_packet", 0600, + ibd->fault_packet->dir, + &ibd->fault_packet->fault_by_packet)) + goto fail; + if (!debugfs_create_u64("fault_stats", 0400, + ibd->fault_packet->dir, + &ibd->fault_packet->n_faults)) + goto fail; + + return 0; +fail: + fault_exit_packet_debugfs(ibd); + return -ENOMEM; +} + +static void fault_exit_debugfs(struct hfi1_ibdev *ibd) +{ + fault_exit_opcode_debugfs(ibd); + fault_exit_packet_debugfs(ibd); +} + +static int fault_init_debugfs(struct hfi1_ibdev *ibd) +{ + int ret = 0; + + ret = fault_init_opcode_debugfs(ibd); + if (ret) + return ret; + + ret = fault_init_packet_debugfs(ibd); + if (ret) + fault_exit_opcode_debugfs(ibd); + + return ret; +} + +bool hfi1_dbg_fault_suppress_err(struct hfi1_ibdev *ibd) +{ + return ibd->fault_suppress_err; +} + +bool hfi1_dbg_fault_opcode(struct rvt_qp *qp, u32 opcode, bool rx) +{ + bool ret = false; + struct hfi1_ibdev *ibd = to_idev(qp->ibqp.device); + + if (!ibd->fault_opcode || !ibd->fault_opcode->fault_by_opcode) + return false; + if (ibd->fault_opcode->opcode != (opcode & ibd->fault_opcode->mask)) + return false; + ret = should_fail(&ibd->fault_opcode->attr, 1); + if (ret) { + trace_hfi1_fault_opcode(qp, opcode); + if (rx) + ibd->fault_opcode->n_rxfaults[opcode]++; + else + ibd->fault_opcode->n_txfaults[opcode]++; + } + return ret; +} + +bool hfi1_dbg_fault_packet(struct hfi1_packet *packet) +{ + struct rvt_dev_info *rdi = &packet->rcd->ppd->dd->verbs_dev.rdi; + struct hfi1_ibdev *ibd = dev_from_rdi(rdi); + bool ret = false; + + if (!ibd->fault_packet || !ibd->fault_packet->fault_by_packet) + return false; + + ret = should_fail(&ibd->fault_packet->attr, 1); + if (ret) { + ++ibd->fault_packet->n_faults; + trace_hfi1_fault_packet(packet); + } + return ret; +} +#endif + void hfi1_dbg_ibdev_init(struct hfi1_ibdev *ibd) { char name[sizeof("port0counters") + 1]; @@ -1112,12 +1332,22 @@ void hfi1_dbg_ibdev_init(struct hfi1_ibdev *ibd) !port_cntr_ops[i].ops.write ? S_IRUGO : S_IRUGO | S_IWUSR); } + +#ifdef CONFIG_FAULT_INJECTION + debugfs_create_bool("fault_suppress_err", 0600, + ibd->hfi1_ibdev_dbg, + &ibd->fault_suppress_err); + fault_init_debugfs(ibd); +#endif } void hfi1_dbg_ibdev_exit(struct hfi1_ibdev *ibd) { if (!hfi1_dbg_root) goto out; +#ifdef CONFIG_FAULT_INJECTION + fault_exit_debugfs(ibd); +#endif debugfs_remove(ibd->hfi1_ibdev_link); debugfs_remove_recursive(ibd->hfi1_ibdev_dbg); out: diff --git a/drivers/infiniband/hw/hfi1/debugfs.h b/drivers/infiniband/hw/hfi1/debugfs.h index b6fb6814f1b8..38c38a98156d 100644 --- a/drivers/infiniband/hw/hfi1/debugfs.h +++ b/drivers/infiniband/hw/hfi1/debugfs.h @@ -53,23 +53,79 @@ void hfi1_dbg_ibdev_init(struct hfi1_ibdev *ibd); void hfi1_dbg_ibdev_exit(struct hfi1_ibdev *ibd); void hfi1_dbg_init(void); void hfi1_dbg_exit(void); + +#ifdef CONFIG_FAULT_INJECTION +#include <linux/fault-inject.h> +struct fault_opcode { + struct fault_attr attr; + struct dentry *dir; + bool fault_by_opcode; + u64 n_rxfaults[256]; + u64 n_txfaults[256]; + u8 opcode; + u8 mask; +}; + +struct fault_packet { + struct fault_attr attr; + struct dentry *dir; + bool fault_by_packet; + u64 n_faults; +}; + +bool hfi1_dbg_fault_opcode(struct rvt_qp *qp, u32 opcode, bool rx); +bool hfi1_dbg_fault_packet(struct hfi1_packet *packet); +bool hfi1_dbg_fault_suppress_err(struct hfi1_ibdev *ibd); +#else +static inline bool hfi1_dbg_fault_packet(struct hfi1_packet *packet) +{ + return false; +} + +static inline bool hfi1_dbg_fault_opcode(struct rvt_qp *qp, + u32 opcode, bool rx) +{ + return false; +} + +static inline bool hfi1_dbg_fault_suppress_err(struct hfi1_ibdev *ibd) +{ + return false; +} +#endif + #else static inline void hfi1_dbg_ibdev_init(struct hfi1_ibdev *ibd) { } -void hfi1_dbg_ibdev_exit(struct hfi1_ibdev *ibd) +static inline void hfi1_dbg_ibdev_exit(struct hfi1_ibdev *ibd) +{ +} + +static inline void hfi1_dbg_init(void) { } -void hfi1_dbg_init(void) +static inline void hfi1_dbg_exit(void) { } -void hfi1_dbg_exit(void) +static inline bool hfi1_dbg_fault_packet(struct hfi1_packet *packet) { + return false; } +static inline bool hfi1_dbg_fault_opcode(struct rvt_qp *qp, + u32 opcode, bool rx) +{ + return false; +} + +static inline bool hfi1_dbg_fault_suppress_err(struct hfi1_ibdev *ibd) +{ + return false; +} #endif #endif /* _HFI1_DEBUGFS_H */ diff --git a/drivers/infiniband/hw/hfi1/driver.c b/drivers/infiniband/hw/hfi1/driver.c index 3881c951f6af..500b129ed565 100644 --- a/drivers/infiniband/hw/hfi1/driver.c +++ b/drivers/infiniband/hw/hfi1/driver.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -59,6 +59,8 @@ #include "trace.h" #include "qp.h" #include "sdma.h" +#include "debugfs.h" +#include "vnic.h" #undef pr_fmt #define pr_fmt(fmt) DRIVER_NAME ": " fmt @@ -283,7 +285,7 @@ static void rcv_hdrerr(struct hfi1_ctxtdata *rcd, struct hfi1_pportdata *ppd, { struct ib_header *rhdr = packet->hdr; u32 rte = rhf_rcv_type_err(packet->rhf); - int lnh = be16_to_cpu(rhdr->lrh[0]) & 3; + int lnh = ib_get_lnh(rhdr); struct hfi1_ibport *ibp = rcd_to_iport(rcd); struct hfi1_devdata *dd = ppd->dd; struct rvt_dev_info *rdi = &dd->verbs_dev.rdi; @@ -295,7 +297,7 @@ static void rcv_hdrerr(struct hfi1_ctxtdata *rcd, struct hfi1_pportdata *ppd, /* For TIDERR and RC QPs preemptively schedule a NAK */ struct ib_other_headers *ohdr = NULL; u32 tlen = rhf_pkt_len(packet->rhf); /* in bytes */ - u16 lid = be16_to_cpu(rhdr->lrh[1]); + u16 lid = ib_get_dlid(rhdr); u32 qp_num; u32 rcv_flags = 0; @@ -396,7 +398,7 @@ static void rcv_hdrerr(struct hfi1_ctxtdata *rcd, struct hfi1_pportdata *ppd, u16 rlid; u8 svc_type, sl, sc5; - sc5 = hdr2sc(rhdr, packet->rhf); + sc5 = hfi1_9B_get_sc5(rhdr, packet->rhf); sl = ibp->sc_to_sl[sc5]; lqpn = be32_to_cpu(bth[1]) & RVT_QPN_MASK; @@ -414,7 +416,7 @@ static void rcv_hdrerr(struct hfi1_ctxtdata *rcd, struct hfi1_pportdata *ppd, svc_type = IB_CC_SVCTYPE_UD; break; case IB_QPT_UC: - rlid = be16_to_cpu(rhdr->lrh[3]); + rlid = ib_get_slid(rhdr); rqpn = qp->remote_qpn; svc_type = IB_CC_SVCTYPE_UC; break; @@ -460,7 +462,7 @@ void hfi1_process_ecn_slowpath(struct rvt_qp *qp, struct hfi1_packet *pkt, struct ib_other_headers *ohdr = pkt->ohdr; struct ib_grh *grh = NULL; u32 rqpn = 0, bth1; - u16 rlid, dlid = be16_to_cpu(hdr->lrh[1]); + u16 rlid, dlid = ib_get_dlid(hdr); u8 sc, svc_type; bool is_mcast = false; @@ -471,7 +473,7 @@ void hfi1_process_ecn_slowpath(struct rvt_qp *qp, struct hfi1_packet *pkt, case IB_QPT_SMI: case IB_QPT_GSI: case IB_QPT_UD: - rlid = be16_to_cpu(hdr->lrh[3]); + rlid = ib_get_slid(hdr); rqpn = be32_to_cpu(ohdr->u.ud.deth[1]) & RVT_QPN_MASK; svc_type = IB_CC_SVCTYPE_UD; is_mcast = (dlid > be16_to_cpu(IB_MULTICAST_LID_BASE)) && @@ -491,16 +493,16 @@ void hfi1_process_ecn_slowpath(struct rvt_qp *qp, struct hfi1_packet *pkt, return; } - sc = hdr2sc(hdr, pkt->rhf); + sc = hfi1_9B_get_sc5(hdr, pkt->rhf); bth1 = be32_to_cpu(ohdr->bth[1]); - if (do_cnp && (bth1 & HFI1_FECN_SMASK)) { + if (do_cnp && (bth1 & IB_FECN_SMASK)) { u16 pkey = (u16)be32_to_cpu(ohdr->bth[0]); return_cnp(ibp, qp, rqpn, pkey, dlid, rlid, sc, grh); } - if (!is_mcast && (bth1 & HFI1_BECN_SMASK)) { + if (!is_mcast && (bth1 & IB_BECN_SMASK)) { struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); u32 lqpn = bth1 & RVT_QPN_MASK; u8 sl = ibp->sc_to_sl[sc]; @@ -621,8 +623,7 @@ static void __prescan_rxq(struct hfi1_packet *packet) packet->hdr = hfi1_get_msgheader(dd, rhf_addr); hdr = packet->hdr; - - lnh = be16_to_cpu(hdr->lrh[0]) & 3; + lnh = ib_get_lnh(hdr); if (lnh == HFI1_LRH_BTH) { packet->ohdr = &hdr->u.oth; @@ -634,7 +635,7 @@ static void __prescan_rxq(struct hfi1_packet *packet) } bth1 = be32_to_cpu(packet->ohdr->bth[1]); - is_ecn = !!(bth1 & (HFI1_FECN_SMASK | HFI1_BECN_SMASK)); + is_ecn = !!(bth1 & (IB_FECN_SMASK | IB_BECN_SMASK)); if (!is_ecn) goto next; @@ -652,7 +653,7 @@ static void __prescan_rxq(struct hfi1_packet *packet) rcu_read_unlock(); /* turn off BECN, FECN */ - bth1 &= ~(HFI1_FECN_SMASK | HFI1_BECN_SMASK); + bth1 &= ~(IB_FECN_SMASK | IB_BECN_SMASK); packet->ohdr->bth[1] = cpu_to_be32(bth1); next: update_ps_mdata(&mdata, rcd); @@ -872,20 +873,42 @@ bail: return last; } -static inline void set_all_nodma_rtail(struct hfi1_devdata *dd) +static inline void set_nodma_rtail(struct hfi1_devdata *dd, u8 ctxt) { int i; - for (i = HFI1_CTRL_CTXT + 1; i < dd->first_user_ctxt; i++) + /* + * For dynamically allocated kernel contexts (like vnic) switch + * interrupt handler only for that context. Otherwise, switch + * interrupt handler for all statically allocated kernel contexts. + */ + if (ctxt >= dd->first_dyn_alloc_ctxt) { + dd->rcd[ctxt]->do_interrupt = + &handle_receive_interrupt_nodma_rtail; + return; + } + + for (i = HFI1_CTRL_CTXT + 1; i < dd->first_dyn_alloc_ctxt; i++) dd->rcd[i]->do_interrupt = &handle_receive_interrupt_nodma_rtail; } -static inline void set_all_dma_rtail(struct hfi1_devdata *dd) +static inline void set_dma_rtail(struct hfi1_devdata *dd, u8 ctxt) { int i; - for (i = HFI1_CTRL_CTXT + 1; i < dd->first_user_ctxt; i++) + /* + * For dynamically allocated kernel contexts (like vnic) switch + * interrupt handler only for that context. Otherwise, switch + * interrupt handler for all statically allocated kernel contexts. + */ + if (ctxt >= dd->first_dyn_alloc_ctxt) { + dd->rcd[ctxt]->do_interrupt = + &handle_receive_interrupt_dma_rtail; + return; + } + + for (i = HFI1_CTRL_CTXT + 1; i < dd->first_dyn_alloc_ctxt; i++) dd->rcd[i]->do_interrupt = &handle_receive_interrupt_dma_rtail; } @@ -895,8 +918,13 @@ void set_all_slowpath(struct hfi1_devdata *dd) int i; /* HFI1_CTRL_CTXT must always use the slow path interrupt handler */ - for (i = HFI1_CTRL_CTXT + 1; i < dd->first_user_ctxt; i++) - dd->rcd[i]->do_interrupt = &handle_receive_interrupt; + for (i = HFI1_CTRL_CTXT + 1; i < dd->num_rcv_contexts; i++) { + struct hfi1_ctxtdata *rcd = dd->rcd[i]; + + if ((i < dd->first_dyn_alloc_ctxt) || + (rcd && rcd->sc && (rcd->sc->type == SC_KERNEL))) + rcd->do_interrupt = &handle_receive_interrupt; + } } static inline int set_armed_to_active(struct hfi1_ctxtdata *rcd, @@ -908,7 +936,8 @@ static inline int set_armed_to_active(struct hfi1_ctxtdata *rcd, packet->rhf_addr); u8 etype = rhf_rcv_type(packet->rhf); - if (etype == RHF_RCV_TYPE_IB && hdr2sc(hdr, packet->rhf) != 0xf) { + if (etype == RHF_RCV_TYPE_IB && + hfi1_9B_get_sc5(hdr, packet->rhf) != 0xf) { int hwstate = read_logical_state(dd); if (hwstate != LSTATE_ACTIVE) { @@ -1006,7 +1035,7 @@ int handle_receive_interrupt(struct hfi1_ctxtdata *rcd, int thread) last = RCV_PKT_DONE; if (needset) { dd_dev_info(dd, "Switching to NO_DMA_RTAIL\n"); - set_all_nodma_rtail(dd); + set_nodma_rtail(dd, rcd->ctxt); needset = 0; } } else { @@ -1028,7 +1057,7 @@ int handle_receive_interrupt(struct hfi1_ctxtdata *rcd, int thread) if (needset) { dd_dev_info(dd, "Switching to DMA_RTAIL\n"); - set_all_dma_rtail(dd); + set_dma_rtail(dd, rcd->ctxt); needset = 0; } } @@ -1077,10 +1106,10 @@ void receive_interrupt_work(struct work_struct *work) set_link_state(ppd, HLS_UP_ACTIVE); /* - * Interrupt all kernel contexts that could have had an - * interrupt during auto activation. + * Interrupt all statically allocated kernel contexts that could + * have had an interrupt during auto activation. */ - for (i = HFI1_CTRL_CTXT; i < dd->first_user_ctxt; i++) + for (i = HFI1_CTRL_CTXT; i < dd->first_dyn_alloc_ctxt; i++) force_recv_intr(dd->rcd[i]); } @@ -1294,7 +1323,8 @@ int hfi1_reset_device(int unit) spin_lock_irqsave(&dd->uctxt_lock, flags); if (dd->rcd) - for (i = dd->first_user_ctxt; i < dd->num_rcv_contexts; i++) { + for (i = dd->first_dyn_alloc_ctxt; + i < dd->num_rcv_contexts; i++) { if (!dd->rcd[i] || !dd->rcd[i]->cnt) continue; spin_unlock_irqrestore(&dd->uctxt_lock, flags); @@ -1354,6 +1384,9 @@ void handle_eflags(struct hfi1_packet *packet) */ int process_receive_ib(struct hfi1_packet *packet) { + if (unlikely(hfi1_dbg_fault_packet(packet))) + return RHF_RCV_CONTINUE; + trace_hfi1_rcvhdr(packet->rcd->ppd->dd, packet->rcd->ctxt, rhf_err_flags(packet->rhf), @@ -1363,6 +1396,11 @@ int process_receive_ib(struct hfi1_packet *packet) packet->updegr, rhf_egr_index(packet->rhf)); + if (unlikely( + (hfi1_dbg_fault_suppress_err(&packet->rcd->dd->verbs_dev) && + (packet->rhf & RHF_DC_ERR)))) + return RHF_RCV_CONTINUE; + if (unlikely(rhf_err_flags(packet->rhf))) { handle_eflags(packet); return RHF_RCV_CONTINUE; @@ -1372,15 +1410,31 @@ int process_receive_ib(struct hfi1_packet *packet) return RHF_RCV_CONTINUE; } +static inline bool hfi1_is_vnic_packet(struct hfi1_packet *packet) +{ + /* Packet received in VNIC context via RSM */ + if (packet->rcd->is_vnic) + return true; + + if ((HFI1_GET_L2_TYPE(packet->ebuf) == OPA_VNIC_L2_TYPE) && + (HFI1_GET_L4_TYPE(packet->ebuf) == OPA_VNIC_L4_ETHR)) + return true; + + return false; +} + int process_receive_bypass(struct hfi1_packet *packet) { struct hfi1_devdata *dd = packet->rcd->dd; - if (unlikely(rhf_err_flags(packet->rhf))) + if (unlikely(rhf_err_flags(packet->rhf))) { handle_eflags(packet); + } else if (hfi1_is_vnic_packet(packet)) { + hfi1_vnic_bypass_rcv(packet); + return RHF_RCV_CONTINUE; + } - dd_dev_err(dd, - "Bypass packets are not supported in normal operation. Dropping\n"); + dd_dev_err(dd, "Unsupported bypass packet. Dropping\n"); incr_cntr64(&dd->sw_rcv_bypass_packet_errors); if (!(dd->err_info_rcvport.status_and_code & OPA_EI_STATUS_SMASK)) { u64 *flits = packet->ebuf; @@ -1398,6 +1452,12 @@ int process_receive_bypass(struct hfi1_packet *packet) int process_receive_error(struct hfi1_packet *packet) { + /* KHdrHCRCErr -- KDETH packet with a bad HCRC */ + if (unlikely( + hfi1_dbg_fault_suppress_err(&packet->rcd->dd->verbs_dev) && + rhf_rcv_type_err(packet->rhf) == 3)) + return RHF_RCV_CONTINUE; + handle_eflags(packet); if (unlikely(rhf_err_flags(packet->rhf))) @@ -1409,6 +1469,8 @@ int process_receive_error(struct hfi1_packet *packet) int kdeth_process_expected(struct hfi1_packet *packet) { + if (unlikely(hfi1_dbg_fault_packet(packet))) + return RHF_RCV_CONTINUE; if (unlikely(rhf_err_flags(packet->rhf))) handle_eflags(packet); @@ -1421,6 +1483,8 @@ int kdeth_process_eager(struct hfi1_packet *packet) { if (unlikely(rhf_err_flags(packet->rhf))) handle_eflags(packet); + if (unlikely(hfi1_dbg_fault_packet(packet))) + return RHF_RCV_CONTINUE; dd_dev_err(packet->rcd->dd, "Unhandled eager packet received. Dropping.\n"); diff --git a/drivers/infiniband/hw/hfi1/file_ops.c b/drivers/infiniband/hw/hfi1/file_ops.c index f78c739b330a..3d9bce4bfcc7 100644 --- a/drivers/infiniband/hw/hfi1/file_ops.c +++ b/drivers/infiniband/hw/hfi1/file_ops.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -586,8 +586,8 @@ static int hfi1_file_mmap(struct file *fp, struct vm_area_struct *vma) * knows where it's own bitmap is within the page. */ memaddr = (unsigned long)(dd->events + - ((uctxt->ctxt - dd->first_user_ctxt) * - HFI1_MAX_SHARED_CTXTS)) & PAGE_MASK; + ((uctxt->ctxt - dd->first_dyn_alloc_ctxt) * + HFI1_MAX_SHARED_CTXTS)) & PAGE_MASK; memlen = PAGE_SIZE; /* * v3.7 removes VM_RESERVED but the effect is kept by @@ -597,6 +597,10 @@ static int hfi1_file_mmap(struct file *fp, struct vm_area_struct *vma) vmf = 1; break; case STATUS: + if (flags & (unsigned long)(VM_WRITE | VM_EXEC)) { + ret = -EPERM; + goto done; + } memaddr = kvirt_to_phys((void *)dd->status); memlen = PAGE_SIZE; flags |= VM_IO | VM_DONTEXPAND; @@ -756,7 +760,7 @@ static int hfi1_file_close(struct inode *inode, struct file *fp) * Clear any left over, unhandled events so the next process that * gets this context doesn't get confused. */ - ev = dd->events + ((uctxt->ctxt - dd->first_user_ctxt) * + ev = dd->events + ((uctxt->ctxt - dd->first_dyn_alloc_ctxt) * HFI1_MAX_SHARED_CTXTS) + fdata->subctxt; *ev = 0; @@ -909,12 +913,18 @@ static int find_shared_ctxt(struct file *fp, if (!(dd && (dd->flags & HFI1_PRESENT) && dd->kregbase)) continue; - for (i = dd->first_user_ctxt; i < dd->num_rcv_contexts; i++) { + for (i = dd->first_dyn_alloc_ctxt; + i < dd->num_rcv_contexts; i++) { struct hfi1_ctxtdata *uctxt = dd->rcd[i]; /* Skip ctxts which are not yet open */ if (!uctxt || !uctxt->cnt) continue; + + /* Skip dynamically allocted kernel contexts */ + if (uctxt->sc && (uctxt->sc->type == SC_KERNEL)) + continue; + /* Skip ctxt if it doesn't match the requested one */ if (memcmp(uctxt->uuid, uinfo->uuid, sizeof(uctxt->uuid)) || @@ -960,7 +970,8 @@ static int allocate_ctxt(struct file *fp, struct hfi1_devdata *dd, return -EIO; } - for (ctxt = dd->first_user_ctxt; ctxt < dd->num_rcv_contexts; ctxt++) + for (ctxt = dd->first_dyn_alloc_ctxt; + ctxt < dd->num_rcv_contexts; ctxt++) if (!dd->rcd[ctxt]) break; @@ -1306,7 +1317,7 @@ static int get_base_info(struct file *fp, void __user *ubase, __u32 len) */ binfo.user_regbase = HFI1_MMAP_TOKEN(UREGS, uctxt->ctxt, fd->subctxt, 0); - offset = offset_in_page((((uctxt->ctxt - dd->first_user_ctxt) * + offset = offset_in_page((((uctxt->ctxt - dd->first_dyn_alloc_ctxt) * HFI1_MAX_SHARED_CTXTS) + fd->subctxt) * sizeof(*dd->events)); binfo.events_bufbase = HFI1_MMAP_TOKEN(EVENTS, uctxt->ctxt, @@ -1400,12 +1411,12 @@ int hfi1_set_uevent_bits(struct hfi1_pportdata *ppd, const int evtbit) } spin_lock_irqsave(&dd->uctxt_lock, flags); - for (ctxt = dd->first_user_ctxt; ctxt < dd->num_rcv_contexts; + for (ctxt = dd->first_dyn_alloc_ctxt; ctxt < dd->num_rcv_contexts; ctxt++) { uctxt = dd->rcd[ctxt]; if (uctxt) { unsigned long *evs = dd->events + - (uctxt->ctxt - dd->first_user_ctxt) * + (uctxt->ctxt - dd->first_dyn_alloc_ctxt) * HFI1_MAX_SHARED_CTXTS; int i; /* @@ -1477,7 +1488,7 @@ static int user_event_ack(struct hfi1_ctxtdata *uctxt, int subctxt, if (!dd->events) return 0; - evs = dd->events + ((uctxt->ctxt - dd->first_user_ctxt) * + evs = dd->events + ((uctxt->ctxt - dd->first_dyn_alloc_ctxt) * HFI1_MAX_SHARED_CTXTS) + subctxt; for (i = 0; i <= _HFI1_MAX_EVENT_BIT; i++) { diff --git a/drivers/infiniband/hw/hfi1/firmware.c b/drivers/infiniband/hw/hfi1/firmware.c index 0dd50cdb039a..4042c11b2742 100644 --- a/drivers/infiniband/hw/hfi1/firmware.c +++ b/drivers/infiniband/hw/hfi1/firmware.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015 - 2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -1004,7 +1004,9 @@ static int load_8051_firmware(struct hfi1_devdata *dd, { u64 reg; int ret; - u8 ver_a, ver_b; + u8 ver_major; + u8 ver_minor; + u8 ver_patch; /* * DC Reset sequence @@ -1073,10 +1075,10 @@ static int load_8051_firmware(struct hfi1_devdata *dd, return -ETIMEDOUT; } - read_misc_status(dd, &ver_a, &ver_b); - dd_dev_info(dd, "8051 firmware version %d.%d\n", - (int)ver_b, (int)ver_a); - dd->dc8051_ver = dc8051_ver(ver_b, ver_a); + read_misc_status(dd, &ver_major, &ver_minor, &ver_patch); + dd_dev_info(dd, "8051 firmware version %d.%d.%d\n", + (int)ver_major, (int)ver_minor, (int)ver_patch); + dd->dc8051_ver = dc8051_ver(ver_major, ver_minor, ver_patch); return 0; } diff --git a/drivers/infiniband/hw/hfi1/hfi.h b/drivers/infiniband/hw/hfi1/hfi.h index 40d7559fa723..14063bd30c2a 100644 --- a/drivers/infiniband/hw/hfi1/hfi.h +++ b/drivers/infiniband/hw/hfi1/hfi.h @@ -1,7 +1,7 @@ #ifndef _HFI1_KERNEL_H #define _HFI1_KERNEL_H /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -54,6 +54,7 @@ #include <linux/list.h> #include <linux/scatterlist.h> #include <linux/slab.h> +#include <linux/idr.h> #include <linux/io.h> #include <linux/fs.h> #include <linux/completion.h> @@ -66,6 +67,7 @@ #include <linux/i2c-algo-bit.h> #include <rdma/ib_hdrs.h> #include <linux/rhashtable.h> +#include <linux/netdevice.h> #include <rdma/rdma_vt.h> #include "chip_registers.h" @@ -278,6 +280,8 @@ struct hfi1_ctxtdata { struct hfi1_devdata *dd; /* so functions that need physical port can get it easily */ struct hfi1_pportdata *ppd; + /* associated msix interrupt */ + u32 msix_intr; /* A page of memory for rcvhdrhead, rcvegrhead, rcvegrtail * N */ void *subctxt_uregbase; /* An array of pages for the eager receive buffers * N */ @@ -337,6 +341,12 @@ struct hfi1_ctxtdata { * packets with the wrong interrupt handler. */ int (*do_interrupt)(struct hfi1_ctxtdata *rcd, int threaded); + + /* Indicates that this is vnic context */ + bool is_vnic; + + /* vnic queue index this context is mapped to */ + u8 vnic_q_idx; }; /* @@ -474,7 +484,7 @@ struct rvt_sge_state; #define HFI1_PART_ENFORCE_OUT 0x2 /* how often we check for synthetic counter wrap around */ -#define SYNTH_CNT_TIME 2 +#define SYNTH_CNT_TIME 3 /* Counter flags */ #define CNTR_NORMAL 0x0 /* Normal counters, just read register */ @@ -808,6 +818,32 @@ struct hfi1_asic_data { struct hfi1_i2c_bus *i2c_bus1; }; +/* sizes for both the QP and RSM map tables */ +#define NUM_MAP_ENTRIES 256 +#define NUM_MAP_REGS 32 + +/* + * Number of VNIC contexts used. Ensure it is less than or equal to + * max queues supported by VNIC (HFI1_VNIC_MAX_QUEUE). + */ +#define HFI1_NUM_VNIC_CTXT 8 + +/* Number of VNIC RSM entries */ +#define NUM_VNIC_MAP_ENTRIES 8 + +/* Virtual NIC information */ +struct hfi1_vnic_data { + struct hfi1_ctxtdata *ctxt[HFI1_NUM_VNIC_CTXT]; + struct kmem_cache *txreq_cache; + u8 num_vports; + struct idr vesw_idr; + u8 rmt_start; + u8 num_ctxt; + u32 msix_idx; +}; + +struct hfi1_vnic_vport_info; + /* device data struct now contains only "general per-device" info. * fields related to a physical IB port are in a hfi1_pportdata struct. */ @@ -926,8 +962,9 @@ struct hfi1_devdata { spinlock_t rcvctrl_lock; /* protect changes to RcvCtrl */ /* around rcd and (user ctxts) ctxt_cnt use (intr vs free) */ spinlock_t uctxt_lock; /* rcd and user context changes */ - /* exclusive access to 8051 */ - spinlock_t dc8051_lock; + struct mutex dc8051_lock; /* exclusive access to 8051 */ + struct workqueue_struct *update_cntr_wq; + struct work_struct update_cntr_work; /* exclusive access to 8051 memory */ spinlock_t dc8051_memlock; int dc8051_timed_out; /* remember if the 8051 timed out */ @@ -1020,7 +1057,7 @@ struct hfi1_devdata { u8 qos_shift; u16 irev; /* implementation revision */ - u16 dc8051_ver; /* 8051 firmware version */ + u32 dc8051_ver; /* 8051 firmware version */ spinlock_t hfi1_diag_trans_lock; /* protect diag observer ops */ struct platform_config platform_config; @@ -1031,6 +1068,7 @@ struct hfi1_devdata { /* MSI-X information */ struct hfi1_msix_entry *msix_entries; u32 num_msix_entries; + u32 first_dyn_msix_idx; /* INTx information */ u32 requested_intx_irq; /* did we request one? */ @@ -1115,6 +1153,9 @@ struct hfi1_devdata { send_routine process_dma_send; void (*pio_inline_send)(struct hfi1_devdata *dd, struct pio_buf *pbuf, u64 pbc, const void *from, size_t count); + int (*process_vnic_dma_send)(struct hfi1_devdata *dd, u8 q_idx, + struct hfi1_vnic_vport_info *vinfo, + struct sk_buff *skb, u64 pbc, u8 plen); /* hfi1_pportdata, points to array of (physical) port-specific * data structs, indexed by pidx (0..n-1) */ @@ -1126,8 +1167,8 @@ struct hfi1_devdata { u16 flags; /* Number of physical ports available */ u8 num_pports; - /* Lowest context number which can be used by user processes */ - u8 first_user_ctxt; + /* Lowest context number which can be used by user processes or VNIC */ + u8 first_dyn_alloc_ctxt; /* adding a new field here would make it part of this cacheline */ /* seqlock for sc2vl */ @@ -1167,15 +1208,24 @@ struct hfi1_devdata { bool eprom_available; /* true if EPROM is available for this device */ bool aspm_supported; /* Does HW support ASPM */ bool aspm_enabled; /* ASPM state: enabled/disabled */ - struct rhashtable sdma_rht; + struct rhashtable *sdma_rht; struct kobject kobj; + + /* vnic data */ + struct hfi1_vnic_data vnic; }; +static inline bool hfi1_vnic_is_rsm_full(struct hfi1_devdata *dd, int spare) +{ + return (dd->vnic.rmt_start + spare) > NUM_MAP_ENTRIES; +} + /* 8051 firmware version helper */ -#define dc8051_ver(a, b) ((a) << 8 | (b)) -#define dc8051_ver_maj(a) ((a & 0xff00) >> 8) -#define dc8051_ver_min(a) (a & 0x00ff) +#define dc8051_ver(a, b, c) ((a) << 16 | (b) << 8 | (c)) +#define dc8051_ver_maj(a) (((a) & 0xff0000) >> 16) +#define dc8051_ver_min(a) (((a) & 0x00ff00) >> 8) +#define dc8051_ver_patch(a) ((a) & 0x0000ff) /* f_put_tid types */ #define PT_EXPECTED 0 @@ -1235,6 +1285,9 @@ int handle_receive_interrupt(struct hfi1_ctxtdata *, int); int handle_receive_interrupt_nodma_rtail(struct hfi1_ctxtdata *, int); int handle_receive_interrupt_dma_rtail(struct hfi1_ctxtdata *, int); void set_all_slowpath(struct hfi1_devdata *dd); +void hfi1_vnic_synchronize_irq(struct hfi1_devdata *dd); +void hfi1_set_vnic_msix_info(struct hfi1_ctxtdata *rcd); +void hfi1_reset_vnic_msix_info(struct hfi1_ctxtdata *rcd); extern const struct pci_device_id hfi1_pci_tbl[]; @@ -1254,16 +1307,24 @@ int hfi1_reset_device(int); /* return the driver's idea of the logical OPA port state */ static inline u32 driver_lstate(struct hfi1_pportdata *ppd) { - return ppd->lstate; /* use the cached value */ + /* + * The driver does some processing from the time the logical + * link state is at INIT to the time the SM can be notified + * as such. Return IB_PORT_DOWN until the software state + * is ready. + */ + if (ppd->lstate == IB_PORT_INIT && !(ppd->host_link_state & HLS_UP)) + return IB_PORT_DOWN; + else + return ppd->lstate; } void receive_interrupt_work(struct work_struct *work); /* extract service channel from header and rhf */ -static inline int hdr2sc(struct ib_header *hdr, u64 rhf) +static inline int hfi1_9B_get_sc5(struct ib_header *hdr, u64 rhf) { - return ((be16_to_cpu(hdr->lrh[0]) >> 12) & 0xf) | - ((!!(rhf_dc_info(rhf))) << 4); + return ib_get_sc(hdr) | ((!!(rhf_dc_info(rhf))) << 4); } #define HFI1_JKEY_WIDTH 16 @@ -1597,9 +1658,9 @@ static inline bool process_ecn(struct rvt_qp *qp, struct hfi1_packet *pkt, u32 bth1; bth1 = be32_to_cpu(ohdr->bth[1]); - if (unlikely(bth1 & (HFI1_BECN_SMASK | HFI1_FECN_SMASK))) { + if (unlikely(bth1 & (IB_BECN_SMASK | IB_FECN_SMASK))) { hfi1_process_ecn_slowpath(qp, pkt, do_cnp); - return bth1 & HFI1_FECN_SMASK; + return !!(bth1 & IB_FECN_SMASK); } return false; } diff --git a/drivers/infiniband/hw/hfi1/init.c b/drivers/infiniband/hw/hfi1/init.c index f40864e9a3b2..4d6b9f82efa3 100644 --- a/drivers/infiniband/hw/hfi1/init.c +++ b/drivers/infiniband/hw/hfi1/init.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -65,6 +65,7 @@ #include "verbs.h" #include "aspm.h" #include "affinity.h" +#include "vnic.h" #undef pr_fmt #define pr_fmt(fmt) DRIVER_NAME ": " fmt @@ -139,7 +140,7 @@ int hfi1_create_ctxts(struct hfi1_devdata *dd) goto nomem; /* create one or more kernel contexts */ - for (i = 0; i < dd->first_user_ctxt; ++i) { + for (i = 0; i < dd->first_dyn_alloc_ctxt; ++i) { struct hfi1_pportdata *ppd; struct hfi1_ctxtdata *rcd; @@ -214,9 +215,9 @@ struct hfi1_ctxtdata *hfi1_create_ctxtdata(struct hfi1_pportdata *ppd, u32 ctxt, u32 base; if (dd->rcv_entries.nctxt_extra > - dd->num_rcv_contexts - dd->first_user_ctxt) + dd->num_rcv_contexts - dd->first_dyn_alloc_ctxt) kctxt_ngroups = (dd->rcv_entries.nctxt_extra - - (dd->num_rcv_contexts - dd->first_user_ctxt)); + (dd->num_rcv_contexts - dd->first_dyn_alloc_ctxt)); rcd = kzalloc_node(sizeof(*rcd), GFP_KERNEL, numa); if (rcd) { u32 rcvtids, max_entries; @@ -238,27 +239,29 @@ struct hfi1_ctxtdata *hfi1_create_ctxtdata(struct hfi1_pportdata *ppd, u32 ctxt, * Calculate the context's RcvArray entry starting point. * We do this here because we have to take into account all * the RcvArray entries that previous context would have - * taken and we have to account for any extra groups - * assigned to the kernel or user contexts. + * taken and we have to account for any extra groups assigned + * to the static (kernel) or dynamic (vnic/user) contexts. */ - if (ctxt < dd->first_user_ctxt) { + if (ctxt < dd->first_dyn_alloc_ctxt) { if (ctxt < kctxt_ngroups) { base = ctxt * (dd->rcv_entries.ngroups + 1); rcd->rcv_array_groups++; - } else + } else { base = kctxt_ngroups + (ctxt * dd->rcv_entries.ngroups); + } } else { - u16 ct = ctxt - dd->first_user_ctxt; + u16 ct = ctxt - dd->first_dyn_alloc_ctxt; base = ((dd->n_krcv_queues * dd->rcv_entries.ngroups) + kctxt_ngroups); if (ct < dd->rcv_entries.nctxt_extra) { base += ct * (dd->rcv_entries.ngroups + 1); rcd->rcv_array_groups++; - } else + } else { base += dd->rcv_entries.nctxt_extra + (ct * dd->rcv_entries.ngroups); + } } rcd->eager_base = base * dd->rcv_entries.group_size; @@ -322,7 +325,8 @@ struct hfi1_ctxtdata *hfi1_create_ctxtdata(struct hfi1_pportdata *ppd, u32 ctxt, } rcd->egrbufs.rcvtid_size = HFI1_MAX_EAGER_BUFFER_SIZE; - if (ctxt < dd->first_user_ctxt) { /* N/A for PSM contexts */ + /* Applicable only for statically created kernel contexts */ + if (ctxt < dd->first_dyn_alloc_ctxt) { rcd->opstats = kzalloc_node(sizeof(*rcd->opstats), GFP_KERNEL, numa); if (!rcd->opstats) @@ -482,6 +486,9 @@ void hfi1_init_pportdata(struct pci_dev *pdev, struct hfi1_pportdata *ppd, default_pkey_idx = 1; ppd->pkeys[default_pkey_idx] = DEFAULT_P_KEY; + ppd->part_enforce |= HFI1_PART_ENFORCE_IN; + ppd->part_enforce |= HFI1_PART_ENFORCE_OUT; + if (loopback) { hfi1_early_err(&pdev->dev, "Faking data partition 0x8001 in idx %u\n", @@ -585,7 +592,7 @@ static void enable_chip(struct hfi1_devdata *dd) * Enable kernel ctxts' receive and receive interrupt. * Other ctxts done as user opens and initializes them. */ - for (i = 0; i < dd->first_user_ctxt; ++i) { + for (i = 0; i < dd->first_dyn_alloc_ctxt; ++i) { rcvmask = HFI1_RCVCTRL_CTXT_ENB | HFI1_RCVCTRL_INTRAVAIL_ENB; rcvmask |= HFI1_CAP_KGET_MASK(dd->rcd[i]->flags, DMA_RTAIL) ? HFI1_RCVCTRL_TAILUPD_ENB : HFI1_RCVCTRL_TAILUPD_DIS; @@ -679,6 +686,7 @@ int hfi1_init(struct hfi1_devdata *dd, int reinit) dd->process_pio_send = hfi1_verbs_send_pio; dd->process_dma_send = hfi1_verbs_send_dma; dd->pio_inline_send = pio_copy; + dd->process_vnic_dma_send = hfi1_vnic_send_dma; if (is_ax(dd)) { atomic_set(&dd->drop_packet, DROP_PACKET_ON); @@ -714,7 +722,7 @@ int hfi1_init(struct hfi1_devdata *dd, int reinit) } /* dd->rcd can be NULL if early initialization failed */ - for (i = 0; dd->rcd && i < dd->first_user_ctxt; ++i) { + for (i = 0; dd->rcd && i < dd->first_dyn_alloc_ctxt; ++i) { /* * Set up the (kernel) rcvhdr queue and egr TIDs. If doing * re-init, the simplest way to handle this is to free @@ -1078,11 +1086,11 @@ struct hfi1_devdata *hfi1_alloc_devdata(struct pci_dev *pdev, size_t extra) spin_lock_init(&dd->uctxt_lock); spin_lock_init(&dd->hfi1_diag_trans_lock); spin_lock_init(&dd->sc_init_lock); - spin_lock_init(&dd->dc8051_lock); spin_lock_init(&dd->dc8051_memlock); seqlock_init(&dd->sc2vl_lock); spin_lock_init(&dd->sde_map_lock); spin_lock_init(&dd->pio_map_lock); + mutex_init(&dd->dc8051_lock); init_waitqueue_head(&dd->event_queue); dd->int_counter = alloc_percpu(u64); @@ -1425,6 +1433,16 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) /* First, lock the non-writable module parameters */ HFI1_CAP_LOCK(); + /* Validate dev ids */ + if (!(ent->device == PCI_DEVICE_ID_INTEL0 || + ent->device == PCI_DEVICE_ID_INTEL1)) { + hfi1_early_err(&pdev->dev, + "Failing on unknown Intel deviceid 0x%x\n", + ent->device); + ret = -ENODEV; + goto bail; + } + /* Validate some global module parameters */ ret = init_validate_rcvhdrcnt(&pdev->dev, rcvhdrcnt); if (ret) @@ -1470,15 +1488,6 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (ret) goto bail; - if (!(ent->device == PCI_DEVICE_ID_INTEL0 || - ent->device == PCI_DEVICE_ID_INTEL1)) { - hfi1_early_err(&pdev->dev, - "Failing on unknown Intel deviceid 0x%x\n", - ent->device); - ret = -ENODEV; - goto clean_bail; - } - /* * Do device-specific initialization, function table setup, dd * allocation, etc. @@ -1497,6 +1506,9 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) /* do the generic initialization */ initfail = hfi1_init(dd, 0); + /* setup vnic */ + hfi1_vnic_setup(dd); + ret = hfi1_register_ib_device(dd); /* @@ -1530,6 +1542,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) hfi1_device_remove(dd); if (!ret) hfi1_unregister_ib_device(dd); + hfi1_vnic_cleanup(dd); postinit_cleanup(dd); if (initfail) ret = initfail; @@ -1574,6 +1587,9 @@ static void remove_one(struct pci_dev *pdev) /* unregister from IB core */ hfi1_unregister_ib_device(dd); + /* cleanup vnic */ + hfi1_vnic_cleanup(dd); + /* * Disable the IB link, disable interrupts on the device, * clear dma engines, etc. @@ -1613,8 +1629,11 @@ int hfi1_create_rcvhdrq(struct hfi1_devdata *dd, struct hfi1_ctxtdata *rcd) amt = PAGE_ALIGN(rcd->rcvhdrq_cnt * rcd->rcvhdrqentsize * sizeof(u32)); - gfp_flags = (rcd->ctxt >= dd->first_user_ctxt) ? - GFP_USER : GFP_KERNEL; + if ((rcd->ctxt < dd->first_dyn_alloc_ctxt) || + (rcd->sc && (rcd->sc->type == SC_KERNEL))) + gfp_flags = GFP_KERNEL; + else + gfp_flags = GFP_USER; rcd->rcvhdrq = dma_zalloc_coherent( &dd->pcidev->dev, amt, &rcd->rcvhdrq_dma, gfp_flags | __GFP_COMP); diff --git a/drivers/infiniband/hw/hfi1/intr.c b/drivers/infiniband/hw/hfi1/intr.c index 65348d16ab2f..232014d46f79 100644 --- a/drivers/infiniband/hw/hfi1/intr.c +++ b/drivers/infiniband/hw/hfi1/intr.c @@ -131,19 +131,24 @@ void handle_linkup_change(struct hfi1_devdata *dd, u32 linkup) if (quick_linkup || dd->icode == ICODE_FUNCTIONAL_SIMULATOR) { set_up_vl15(dd, dd->vau, dd->vl15_init); assign_remote_cm_au_table(dd, dd->vcu); - ppd->neighbor_guid = - read_csr(dd, DC_DC8051_STS_REMOTE_GUID); - ppd->neighbor_type = - read_csr(dd, DC_DC8051_STS_REMOTE_NODE_TYPE) & - DC_DC8051_STS_REMOTE_NODE_TYPE_VAL_MASK; - ppd->neighbor_port_number = - read_csr(dd, DC_DC8051_STS_REMOTE_PORT_NO) & - DC_DC8051_STS_REMOTE_PORT_NO_VAL_SMASK; - dd_dev_info(dd, "Neighbor GUID: %llx Neighbor type %d\n", - ppd->neighbor_guid, - ppd->neighbor_type); } + ppd->neighbor_guid = + read_csr(dd, DC_DC8051_STS_REMOTE_GUID); + ppd->neighbor_type = + read_csr(dd, DC_DC8051_STS_REMOTE_NODE_TYPE) & + DC_DC8051_STS_REMOTE_NODE_TYPE_VAL_MASK; + ppd->neighbor_port_number = + read_csr(dd, DC_DC8051_STS_REMOTE_PORT_NO) & + DC_DC8051_STS_REMOTE_PORT_NO_VAL_SMASK; + ppd->neighbor_fm_security = + read_csr(dd, DC_DC8051_STS_REMOTE_FM_SECURITY) & + DC_DC8051_STS_LOCAL_FM_SECURITY_DISABLED_MASK; + dd_dev_info(dd, + "Neighbor Guid %llx, Type %d, Port Num %d\n", + ppd->neighbor_guid, ppd->neighbor_type, + ppd->neighbor_port_number); + /* physical link went up */ ppd->linkup = 1; ppd->offline_disabled_reason = diff --git a/drivers/infiniband/hw/hfi1/mad.c b/drivers/infiniband/hw/hfi1/mad.c index 09cda3c35e82..836d00b04547 100644 --- a/drivers/infiniband/hw/hfi1/mad.c +++ b/drivers/infiniband/hw/hfi1/mad.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -53,6 +53,7 @@ #include "mad.h" #include "trace.h" #include "qp.h" +#include "vnic.h" /* the reset value from the FM is supposed to be 0xffff, handle both */ #define OPA_LINK_WIDTH_RESET_OLD 0x0fff @@ -650,9 +651,11 @@ static int __subn_get_opa_portinfo(struct opa_smp *smp, u32 am, u8 *data, OPA_PI_MASK_PORT_ACTIVE_OPTOMIZE : 0); pi->port_packet_format.supported = - cpu_to_be16(OPA_PORT_PACKET_FORMAT_9B); + cpu_to_be16(OPA_PORT_PACKET_FORMAT_9B | + OPA_PORT_PACKET_FORMAT_16B); pi->port_packet_format.enabled = - cpu_to_be16(OPA_PORT_PACKET_FORMAT_9B); + cpu_to_be16(OPA_PORT_PACKET_FORMAT_9B | + OPA_PORT_PACKET_FORMAT_16B); /* flit_control.interleave is (OPA V1, version .76): * bits use @@ -701,7 +704,13 @@ static int __subn_get_opa_portinfo(struct opa_smp *smp, u32 am, u8 *data, buffer_units |= (dd->vl15_init << 11) & OPA_PI_MASK_BUF_UNIT_VL15_INIT; pi->buffer_units = cpu_to_be32(buffer_units); - pi->opa_cap_mask = cpu_to_be16(OPA_CAP_MASK3_IsSharedSpaceSupported); + pi->opa_cap_mask = cpu_to_be16(OPA_CAP_MASK3_IsSharedSpaceSupported | + OPA_CAP_MASK3_IsEthOnFabricSupported); + /* Driver does not support mcast/collective configuration */ + pi->opa_cap_mask &= + cpu_to_be16(~OPA_CAP_MASK3_IsAddrRangeConfigSupported); + pi->collectivemask_multicastmask = ((HFI1_COLLECTIVE_NR & 0x7) + << 3 | (HFI1_MCAST_NR & 0x7)); /* HFI supports a replay buffer 128 LTPs in size */ pi->replay_depth.buffer = 0x80; @@ -1146,16 +1155,6 @@ static int __subn_set_opa_portinfo(struct opa_smp *smp, u32 am, u8 *data, ppd->linkinit_reason = (pi->partenforce_filterraw & OPA_PI_MASK_LINKINIT_REASON); - /* enable/disable SW pkey checking as per FM control */ - if (pi->partenforce_filterraw & OPA_PI_MASK_PARTITION_ENFORCE_IN) - ppd->part_enforce |= HFI1_PART_ENFORCE_IN; - else - ppd->part_enforce &= ~HFI1_PART_ENFORCE_IN; - - if (pi->partenforce_filterraw & OPA_PI_MASK_PARTITION_ENFORCE_OUT) - ppd->part_enforce |= HFI1_PART_ENFORCE_OUT; - else - ppd->part_enforce &= ~HFI1_PART_ENFORCE_OUT; /* Must be a valid unicast LID address. */ if ((smlid == 0 && ls_old > IB_PORT_INIT) || @@ -1465,25 +1464,15 @@ static int __subn_set_opa_pkeytable(struct opa_smp *smp, u32 am, u8 *data, return __subn_get_opa_pkeytable(smp, am, data, ibdev, port, resp_len); } -static int get_sc2vlt_tables(struct hfi1_devdata *dd, void *data) -{ - u64 *val = data; - - *val++ = read_csr(dd, SEND_SC2VLT0); - *val++ = read_csr(dd, SEND_SC2VLT1); - *val++ = read_csr(dd, SEND_SC2VLT2); - *val++ = read_csr(dd, SEND_SC2VLT3); - return 0; -} - #define ILLEGAL_VL 12 /* * filter_sc2vlt changes mappings to VL15 to ILLEGAL_VL (except * for SC15, which must map to VL15). If we don't remap things this * way it is possible for VL15 counters to increment when we try to * send on a SC which is mapped to an invalid VL. + * When getting the table convert ILLEGAL_VL back to VL15. */ -static void filter_sc2vlt(void *data) +static void filter_sc2vlt(void *data, bool set) { int i; u8 *pd = data; @@ -1491,8 +1480,14 @@ static void filter_sc2vlt(void *data) for (i = 0; i < OPA_MAX_SCS; i++) { if (i == 15) continue; - if ((pd[i] & 0x1f) == 0xf) - pd[i] = ILLEGAL_VL; + + if (set) { + if ((pd[i] & 0x1f) == 0xf) + pd[i] = ILLEGAL_VL; + } else { + if ((pd[i] & 0x1f) == ILLEGAL_VL) + pd[i] = 0xf; + } } } @@ -1500,7 +1495,7 @@ static int set_sc2vlt_tables(struct hfi1_devdata *dd, void *data) { u64 *val = data; - filter_sc2vlt(data); + filter_sc2vlt(data, true); write_csr(dd, SEND_SC2VLT0, *val++); write_csr(dd, SEND_SC2VLT1, *val++); @@ -1512,6 +1507,19 @@ static int set_sc2vlt_tables(struct hfi1_devdata *dd, void *data) return 0; } +static int get_sc2vlt_tables(struct hfi1_devdata *dd, void *data) +{ + u64 *val = (u64 *)data; + + *val++ = read_csr(dd, SEND_SC2VLT0); + *val++ = read_csr(dd, SEND_SC2VLT1); + *val++ = read_csr(dd, SEND_SC2VLT2); + *val++ = read_csr(dd, SEND_SC2VLT3); + + filter_sc2vlt((u64 *)data, false); + return 0; +} + static int __subn_get_opa_sl_to_sc(struct opa_smp *smp, u32 am, u8 *data, struct ib_device *ibdev, u8 port, u32 *resp_len) @@ -1986,31 +1994,6 @@ struct opa_pma_mad { u8 data[2024]; } __packed; -struct opa_class_port_info { - u8 base_version; - u8 class_version; - __be16 cap_mask; - __be32 cap_mask2_resp_time; - - u8 redirect_gid[16]; - __be32 redirect_tc_fl; - __be32 redirect_lid; - __be32 redirect_sl_qp; - __be32 redirect_qkey; - - u8 trap_gid[16]; - __be32 trap_tc_fl; - __be32 trap_lid; - __be32 trap_hl_qp; - __be32 trap_qkey; - - __be16 trap_pkey; - __be16 redirect_pkey; - - u8 trap_sl_rsvd; - u8 reserved[3]; -} __packed; - struct opa_port_status_req { __u8 port_num; __u8 reserved[3]; diff --git a/drivers/infiniband/hw/hfi1/pcie.c b/drivers/infiniband/hw/hfi1/pcie.c index c81556e84831..93faf86d54b6 100644 --- a/drivers/infiniband/hw/hfi1/pcie.c +++ b/drivers/infiniband/hw/hfi1/pcie.c @@ -553,7 +553,7 @@ pci_mmio_enabled(struct pci_dev *pdev) if (words == ~0ULL) ret = PCI_ERS_RESULT_NEED_RESET; dd_dev_info(dd, - "HFI1 mmio_enabled function called, read wordscntr %Lx, returning %d\n", + "HFI1 mmio_enabled function called, read wordscntr %llx, returning %d\n", words, ret); } return ret; diff --git a/drivers/infiniband/hw/hfi1/pio.c b/drivers/infiniband/hw/hfi1/pio.c index 615be68e40b3..ed72b5aca139 100644 --- a/drivers/infiniband/hw/hfi1/pio.c +++ b/drivers/infiniband/hw/hfi1/pio.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -703,6 +703,7 @@ struct send_context *sc_alloc(struct hfi1_devdata *dd, int type, { struct send_context_info *sci; struct send_context *sc = NULL; + int req_type = type; dma_addr_t dma; unsigned long flags; u64 reg; @@ -729,6 +730,13 @@ struct send_context *sc_alloc(struct hfi1_devdata *dd, int type, return NULL; } + /* + * VNIC contexts are dynamically allocated. + * Hence, pick a user context for VNIC. + */ + if (type == SC_VNIC) + type = SC_USER; + spin_lock_irqsave(&dd->sc_lock, flags); ret = sc_hw_alloc(dd, type, &sw_index, &hw_context); if (ret) { @@ -738,6 +746,15 @@ struct send_context *sc_alloc(struct hfi1_devdata *dd, int type, return NULL; } + /* + * VNIC contexts are used by kernel driver. + * Hence, mark them as kernel contexts. + */ + if (req_type == SC_VNIC) { + dd->send_contexts[sw_index].type = SC_KERNEL; + type = SC_KERNEL; + } + sci = &dd->send_contexts[sw_index]; sci->sc = sc; diff --git a/drivers/infiniband/hw/hfi1/pio.h b/drivers/infiniband/hw/hfi1/pio.h index 867e5ffc3595..99ca5edb0b43 100644 --- a/drivers/infiniband/hw/hfi1/pio.h +++ b/drivers/infiniband/hw/hfi1/pio.h @@ -1,7 +1,7 @@ #ifndef _PIO_H #define _PIO_H /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -54,6 +54,12 @@ #define SC_USER 3 /* must be the last one: it may take all left */ #define SC_MAX 4 /* count of send context types */ +/* + * SC_VNIC types are allocated (dynamically) from the user context pool, + * (SC_USER) and used by kernel driver as kernel contexts (SC_KERNEL). + */ +#define SC_VNIC SC_MAX + /* invalid send context index */ #define INVALID_SCI 0xff @@ -195,7 +201,7 @@ struct sc_config_sizes { * | mask | --/ |--------------------| * |--------------------------| -/ | * | * | actual_vls (max 8) | -/ |--------------------| - * |--------------------------| --/ | ksc[n] -> sc n | + * |--------------------------| --/ | ksc[n-1] -> sc n | * | vls (max 8) | -/ +--------------------+ * |--------------------------| --/ * | map[0] |-/ @@ -208,21 +214,21 @@ struct sc_config_sizes { * |--------------------------| |--------------------| * | map[vls - 1] |- | * | * +--------------------------+ \- |--------------------| - * \- | ksc[m] -> sc m+n | + * \- | ksc[m-1] -> sc m+n | * \ +--------------------+ * \- * \ - * \- +--------------------+ - * \- | mask | - * \ |--------------------| - * \- | ksc[0] -> sc 1+m+n | - * \- |--------------------| - * >| ksc[1] -> sc 2+m+n | - * |--------------------| - * | * | - * |--------------------| - * | ksc[o] -> sc o+m+n | - * +--------------------+ + * \- +----------------------+ + * \- | mask | + * \ |----------------------| + * \- | ksc[0] -> sc 1+m+n | + * \- |----------------------| + * >| ksc[1] -> sc 2+m+n | + * |----------------------| + * | * | + * |----------------------| + * | ksc[o-1] -> sc o+m+n | + * +----------------------+ * */ diff --git a/drivers/infiniband/hw/hfi1/rc.c b/drivers/infiniband/hw/hfi1/rc.c index 7382be11afca..9b3333fd9dc0 100644 --- a/drivers/infiniband/hw/hfi1/rc.c +++ b/drivers/infiniband/hw/hfi1/rc.c @@ -773,7 +773,7 @@ void hfi1_send_rc_ack(struct hfi1_ctxtdata *rcd, struct rvt_qp *qp, hdr.lrh[3] = cpu_to_be16(ppd->lid | qp->remote_ah_attr.src_path_bits); ohdr->bth[0] = cpu_to_be32(bth0); ohdr->bth[1] = cpu_to_be32(qp->remote_qpn); - ohdr->bth[1] |= cpu_to_be32((!!is_fecn) << HFI1_BECN_SHIFT); + ohdr->bth[1] |= cpu_to_be32((!!is_fecn) << IB_BECN_SHIFT); ohdr->bth[2] = cpu_to_be32(mask_psn(qp->r_ack_psn)); /* Don't try to send ACKs if the link isn't ACTIVE */ @@ -994,12 +994,12 @@ void hfi1_rc_send_complete(struct rvt_qp *qp, struct ib_header *hdr) return; /* Find out where the BTH is */ - if ((be16_to_cpu(hdr->lrh[0]) & 3) == HFI1_LRH_BTH) + if (ib_get_lnh(hdr) == HFI1_LRH_BTH) ohdr = &hdr->u.oth; else ohdr = &hdr->u.l.oth; - opcode = be32_to_cpu(ohdr->bth[0]) >> 24; + opcode = ib_bth_get_opcode(ohdr); if (opcode >= OP(RDMA_READ_RESPONSE_FIRST) && opcode <= OP(ATOMIC_ACKNOWLEDGE)) { WARN_ON(!qp->s_rdma_ack_cnt); @@ -1028,13 +1028,17 @@ void hfi1_rc_send_complete(struct rvt_qp *qp, struct ib_header *hdr) cmp_psn(qp->s_sending_psn, qp->s_sending_hpsn) <= 0) break; s_last = qp->s_last; + trace_hfi1_qp_send_completion(qp, wqe, s_last); if (++s_last >= qp->s_size) s_last = 0; qp->s_last = s_last; /* see post_send() */ barrier(); rvt_put_swqe(wqe); - rvt_qp_swqe_complete(qp, wqe, IB_WC_SUCCESS); + rvt_qp_swqe_complete(qp, + wqe, + ib_hfi1_wc_opcode[wqe->wr.opcode], + IB_WC_SUCCESS); } /* * If we were waiting for sends to complete before re-sending, @@ -1076,12 +1080,16 @@ static struct rvt_swqe *do_rc_completion(struct rvt_qp *qp, rvt_put_swqe(wqe); s_last = qp->s_last; + trace_hfi1_qp_send_completion(qp, wqe, s_last); if (++s_last >= qp->s_size) s_last = 0; qp->s_last = s_last; /* see post_send() */ barrier(); - rvt_qp_swqe_complete(qp, wqe, IB_WC_SUCCESS); + rvt_qp_swqe_complete(qp, + wqe, + ib_hfi1_wc_opcode[wqe->wr.opcode], + IB_WC_SUCCESS); } else { struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); @@ -1516,7 +1524,7 @@ read_middle: if (!do_rc_ack(qp, aeth, psn, opcode, 0, rcd)) goto ack_done; /* Get the number of bytes the message was padded by. */ - pad = (be32_to_cpu(ohdr->bth[0]) >> 20) & 3; + pad = ib_bth_get_pad(ohdr); /* * Check that the data size is >= 0 && <= pmtu. * Remember to account for ICRC (4). @@ -1540,7 +1548,7 @@ read_middle: if (unlikely(wqe->wr.opcode != IB_WR_RDMA_READ)) goto ack_op_err; /* Get the number of bytes the message was padded by. */ - pad = (be32_to_cpu(ohdr->bth[0]) >> 20) & 3; + pad = ib_bth_get_pad(ohdr); /* * Check that the data size is >= 1 && <= pmtu. * Remember to account for ICRC (4). @@ -1922,7 +1930,8 @@ void hfi1_rc_rcv(struct hfi1_packet *packet) int diff; struct ib_reth *reth; unsigned long flags; - int ret, is_fecn = 0; + int ret; + bool is_fecn = false; bool copy_last = false; u32 rkey; @@ -1934,7 +1943,7 @@ void hfi1_rc_rcv(struct hfi1_packet *packet) is_fecn = process_ecn(qp, packet, false); psn = be32_to_cpu(ohdr->bth[2]); - opcode = (bth0 >> 24) & 0xff; + opcode = ib_bth_get_opcode(ohdr); /* * Process responses (ACKs) before anything else. Note that the @@ -2065,7 +2074,7 @@ no_immediate_data: wc.ex.imm_data = 0; send_last: /* Get the number of bytes the message was padded by. */ - pad = (bth0 >> 20) & 3; + pad = ib_bth_get_pad(ohdr); /* Check for invalid length. */ /* LAST len should be >= 1 */ if (unlikely(tlen < (hdrsize + pad + 4))) @@ -2378,7 +2387,7 @@ void hfi1_rc_hdrerr( return; psn = be32_to_cpu(ohdr->bth[2]); - opcode = (bth0 >> 24) & 0xff; + opcode = ib_bth_get_opcode(ohdr); /* Only deal with RDMA Writes for now */ if (opcode < IB_OPCODE_RC_RDMA_READ_RESPONSE_FIRST) { diff --git a/drivers/infiniband/hw/hfi1/ruc.c b/drivers/infiniband/hw/hfi1/ruc.c index aa15bcbfb079..ccf8d8037355 100644 --- a/drivers/infiniband/hw/hfi1/ruc.c +++ b/drivers/infiniband/hw/hfi1/ruc.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015 - 2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -238,18 +238,18 @@ int hfi1_ruc_check_hdr(struct hfi1_ibport *ibp, struct ib_header *hdr, qp->alt_ah_attr.grh.dgid.global.interface_id)) goto err; } - if (unlikely(rcv_pkey_check(ppd_from_ibp(ibp), (u16)bth0, - sc5, be16_to_cpu(hdr->lrh[3])))) { + if (unlikely(rcv_pkey_check(ppd_from_ibp(ibp), (u16)bth0, sc5, + ib_get_slid(hdr)))) { hfi1_bad_pqkey(ibp, OPA_TRAP_BAD_P_KEY, (u16)bth0, - (be16_to_cpu(hdr->lrh[0]) >> 4) & 0xF, + ib_get_sl(hdr), 0, qp->ibqp.qp_num, - be16_to_cpu(hdr->lrh[3]), - be16_to_cpu(hdr->lrh[1])); + ib_get_slid(hdr), + ib_get_dlid(hdr)); goto err; } /* Validate the SLID. See Ch. 9.6.1.5 and 17.2.8 */ - if (be16_to_cpu(hdr->lrh[3]) != qp->alt_ah_attr.dlid || + if (ib_get_slid(hdr) != qp->alt_ah_attr.dlid || ppd_from_ibp(ibp)->port != qp->alt_ah_attr.port_num) goto err; spin_lock_irqsave(&qp->s_lock, flags); @@ -273,18 +273,18 @@ int hfi1_ruc_check_hdr(struct hfi1_ibport *ibp, struct ib_header *hdr, qp->remote_ah_attr.grh.dgid.global.interface_id)) goto err; } - if (unlikely(rcv_pkey_check(ppd_from_ibp(ibp), (u16)bth0, - sc5, be16_to_cpu(hdr->lrh[3])))) { + if (unlikely(rcv_pkey_check(ppd_from_ibp(ibp), (u16)bth0, sc5, + ib_get_slid(hdr)))) { hfi1_bad_pqkey(ibp, OPA_TRAP_BAD_P_KEY, (u16)bth0, - (be16_to_cpu(hdr->lrh[0]) >> 4) & 0xF, + ib_get_sl(hdr), 0, qp->ibqp.qp_num, - be16_to_cpu(hdr->lrh[3]), - be16_to_cpu(hdr->lrh[1])); + ib_get_slid(hdr), + ib_get_dlid(hdr)); goto err; } /* Validate the SLID. See Ch. 9.6.1.5 */ - if (be16_to_cpu(hdr->lrh[3]) != qp->remote_ah_attr.dlid || + if (ib_get_slid(hdr) != qp->remote_ah_attr.dlid || ppd_from_ibp(ibp)->port != qp->port_num) goto err; if (qp->s_mig_state == IB_MIG_REARM && @@ -775,7 +775,7 @@ void hfi1_make_ruc_header(struct rvt_qp *qp, struct ib_other_headers *ohdr, if (qp->s_flags & RVT_S_ECN) { qp->s_flags &= ~RVT_S_ECN; /* we recently received a FECN, so return a BECN */ - bth1 |= (HFI1_BECN_MASK << HFI1_BECN_SHIFT); + bth1 |= (IB_BECN_MASK << IB_BECN_SHIFT); } ohdr->bth[1] = cpu_to_be32(bth1); ohdr->bth[2] = cpu_to_be32(bth2); @@ -784,23 +784,29 @@ void hfi1_make_ruc_header(struct rvt_qp *qp, struct ib_other_headers *ohdr, /* when sending, force a reschedule every one of these periods */ #define SEND_RESCHED_TIMEOUT (5 * HZ) /* 5s in jiffies */ +void hfi1_do_send_from_rvt(struct rvt_qp *qp) +{ + hfi1_do_send(qp, false); +} + void _hfi1_do_send(struct work_struct *work) { struct iowait *wait = container_of(work, struct iowait, iowork); struct rvt_qp *qp = iowait_to_qp(wait); - hfi1_do_send(qp); + hfi1_do_send(qp, true); } /** * hfi1_do_send - perform a send on a QP * @work: contains a pointer to the QP + * @in_thread: true if in a workqueue thread * * Process entries in the send work queue until credit or queue is * exhausted. Only allow one CPU to send a packet per QP. * Otherwise, two threads could send packets out of order. */ -void hfi1_do_send(struct rvt_qp *qp) +void hfi1_do_send(struct rvt_qp *qp, bool in_thread) { struct hfi1_pkt_state ps; struct hfi1_qp_priv *priv = qp->priv; @@ -868,8 +874,10 @@ void hfi1_do_send(struct rvt_qp *qp) qp->s_hdrwords = 0; /* allow other tasks to run */ if (unlikely(time_after(jiffies, timeout))) { - if (workqueue_congested(cpu, - ps.ppd->hfi1_wq)) { + if (!in_thread || + workqueue_congested( + cpu, + ps.ppd->hfi1_wq)) { spin_lock_irqsave( &qp->s_lock, ps.flags); @@ -882,11 +890,9 @@ void hfi1_do_send(struct rvt_qp *qp) *ps.ppd->dd->send_schedule); return; } - if (!irqs_disabled()) { - cond_resched(); - this_cpu_inc( - *ps.ppd->dd->send_schedule); - } + cond_resched(); + this_cpu_inc( + *ps.ppd->dd->send_schedule); timeout = jiffies + (timeout_int) / 8; } spin_lock_irqsave(&qp->s_lock, ps.flags); @@ -909,8 +915,10 @@ void hfi1_send_complete(struct rvt_qp *qp, struct rvt_swqe *wqe, last = qp->s_last; old_last = last; + trace_hfi1_qp_send_completion(qp, wqe, last); if (++last >= qp->s_size) last = 0; + trace_hfi1_qp_send_completion(qp, wqe, last); qp->s_last = last; /* See post_send() */ barrier(); @@ -920,7 +928,10 @@ void hfi1_send_complete(struct rvt_qp *qp, struct rvt_swqe *wqe, qp->ibqp.qp_type == IB_QPT_GSI) atomic_dec(&ibah_to_rvtah(wqe->ud_wr.ah)->refcount); - rvt_qp_swqe_complete(qp, wqe, status); + rvt_qp_swqe_complete(qp, + wqe, + ib_hfi1_wc_opcode[wqe->wr.opcode], + status); if (qp->s_acked == old_last) qp->s_acked = last; diff --git a/drivers/infiniband/hw/hfi1/sdma.c b/drivers/infiniband/hw/hfi1/sdma.c index 5cde1ecda0fe..bfd0d5187e9b 100644 --- a/drivers/infiniband/hw/hfi1/sdma.c +++ b/drivers/infiniband/hw/hfi1/sdma.c @@ -868,7 +868,7 @@ struct sdma_engine *sdma_select_user_engine(struct hfi1_devdata *dd, cpu_id = smp_processor_id(); rcu_read_lock(); - rht_node = rhashtable_lookup_fast(&dd->sdma_rht, &cpu_id, + rht_node = rhashtable_lookup_fast(dd->sdma_rht, &cpu_id, sdma_rht_params); if (rht_node && rht_node->map[vl]) { @@ -962,7 +962,12 @@ ssize_t sdma_set_cpu_to_sde_map(struct sdma_engine *sde, const char *buf, continue; } - rht_node = rhashtable_lookup_fast(&dd->sdma_rht, &cpu, + if (vl >= ARRAY_SIZE(rht_node->map)) { + ret = -EINVAL; + goto out; + } + + rht_node = rhashtable_lookup_fast(dd->sdma_rht, &cpu, sdma_rht_params); if (!rht_node) { rht_node = kzalloc(sizeof(*rht_node), GFP_KERNEL); @@ -982,7 +987,7 @@ ssize_t sdma_set_cpu_to_sde_map(struct sdma_engine *sde, const char *buf, rht_node->map[vl]->ctr = 1; rht_node->map[vl]->sde[0] = sde; - ret = rhashtable_insert_fast(&dd->sdma_rht, + ret = rhashtable_insert_fast(dd->sdma_rht, &rht_node->node, sdma_rht_params); if (ret) { @@ -1025,7 +1030,7 @@ ssize_t sdma_set_cpu_to_sde_map(struct sdma_engine *sde, const char *buf, if (cpumask_test_cpu(cpu, mask)) continue; - rht_node = rhashtable_lookup_fast(&dd->sdma_rht, &cpu, + rht_node = rhashtable_lookup_fast(dd->sdma_rht, &cpu, sdma_rht_params); if (rht_node) { bool empty = true; @@ -1049,7 +1054,7 @@ ssize_t sdma_set_cpu_to_sde_map(struct sdma_engine *sde, const char *buf, } if (empty) { - ret = rhashtable_remove_fast(&dd->sdma_rht, + ret = rhashtable_remove_fast(dd->sdma_rht, &rht_node->node, sdma_rht_params); WARN_ON(ret); @@ -1108,7 +1113,7 @@ void sdma_seqfile_dump_cpu_list(struct seq_file *s, struct sdma_rht_node *rht_node; int i, j; - rht_node = rhashtable_lookup_fast(&dd->sdma_rht, &cpuid, + rht_node = rhashtable_lookup_fast(dd->sdma_rht, &cpuid, sdma_rht_params); if (!rht_node) return; @@ -1322,6 +1327,12 @@ static void sdma_clean(struct hfi1_devdata *dd, size_t num_engines) synchronize_rcu(); kfree(dd->per_sdma); dd->per_sdma = NULL; + + if (dd->sdma_rht) { + rhashtable_free_and_destroy(dd->sdma_rht, sdma_rht_free, NULL); + kfree(dd->sdma_rht); + dd->sdma_rht = NULL; + } } /** @@ -1341,12 +1352,14 @@ int sdma_init(struct hfi1_devdata *dd, u8 port) { unsigned this_idx; struct sdma_engine *sde; + struct rhashtable *tmp_sdma_rht; u16 descq_cnt; void *curr_head; struct hfi1_pportdata *ppd = dd->pport + port; u32 per_sdma_credits; uint idle_cnt = sdma_idle_cnt; size_t num_engines = dd->chip_sdma_engines; + int ret = -ENOMEM; if (!HFI1_CAP_IS_KSET(SDMA)) { HFI1_CAP_CLEAR(SDMA_AHG); @@ -1378,7 +1391,7 @@ int sdma_init(struct hfi1_devdata *dd, u8 port) /* alloc memory for array of send engines */ dd->per_sdma = kcalloc(num_engines, sizeof(*dd->per_sdma), GFP_KERNEL); if (!dd->per_sdma) - return -ENOMEM; + return ret; idle_cnt = ns_to_cclock(dd, idle_cnt); if (!sdma_desct_intr) @@ -1507,18 +1520,27 @@ int sdma_init(struct hfi1_devdata *dd, u8 port) dd->flags |= HFI1_HAS_SEND_DMA; dd->flags |= idle_cnt ? HFI1_HAS_SDMA_TIMEOUT : 0; dd->num_sdma = num_engines; - if (sdma_map_init(dd, port, ppd->vls_operational, NULL)) + ret = sdma_map_init(dd, port, ppd->vls_operational, NULL); + if (ret < 0) goto bail; - if (rhashtable_init(&dd->sdma_rht, &sdma_rht_params)) + tmp_sdma_rht = kzalloc(sizeof(*tmp_sdma_rht), GFP_KERNEL); + if (!tmp_sdma_rht) { + ret = -ENOMEM; goto bail; + } + + ret = rhashtable_init(tmp_sdma_rht, &sdma_rht_params); + if (ret < 0) + goto bail; + dd->sdma_rht = tmp_sdma_rht; dd_dev_info(dd, "SDMA num_sdma: %u\n", dd->num_sdma); return 0; bail: sdma_clean(dd, num_engines); - return -ENOMEM; + return ret; } /** @@ -1604,7 +1626,6 @@ void sdma_exit(struct hfi1_devdata *dd) sdma_finalput(&sde->state); } sdma_clean(dd, dd->num_sdma); - rhashtable_free_and_destroy(&dd->sdma_rht, sdma_rht_free, NULL); } /* diff --git a/drivers/infiniband/hw/hfi1/sdma.h b/drivers/infiniband/hw/hfi1/sdma.h index 21f1e2834f37..64f10b8b5db8 100644 --- a/drivers/infiniband/hw/hfi1/sdma.h +++ b/drivers/infiniband/hw/hfi1/sdma.h @@ -966,34 +966,34 @@ void sdma_engine_interrupt(struct sdma_engine *sde, u64 status); * | mask | --/ |--------------------| * |--------------------------| -/ | * | * | actual_vls (max 8) | -/ |--------------------| - * |--------------------------| --/ | sde[n] -> eng n | + * |--------------------------| --/ | sde[n-1] -> eng n | * | vls (max 8) | -/ +--------------------+ * |--------------------------| --/ * | map[0] |-/ - * |--------------------------| +--------------------+ - * | map[1] |--- | mask | - * |--------------------------| \---- |--------------------| - * | * | \-- | sde[0] -> eng 1+n | - * | * | \---- |--------------------| - * | * | \->| sde[1] -> eng 2+n | - * |--------------------------| |--------------------| - * | map[vls - 1] |- | * | - * +--------------------------+ \- |--------------------| - * \- | sde[m] -> eng m+n | - * \ +--------------------+ + * |--------------------------| +---------------------+ + * | map[1] |--- | mask | + * |--------------------------| \---- |---------------------| + * | * | \-- | sde[0] -> eng 1+n | + * | * | \---- |---------------------| + * | * | \->| sde[1] -> eng 2+n | + * |--------------------------| |---------------------| + * | map[vls - 1] |- | * | + * +--------------------------+ \- |---------------------| + * \- | sde[m-1] -> eng m+n | + * \ +---------------------+ * \- * \ - * \- +--------------------+ - * \- | mask | - * \ |--------------------| - * \- | sde[0] -> eng 1+m+n| - * \- |--------------------| - * >| sde[1] -> eng 2+m+n| - * |--------------------| - * | * | - * |--------------------| - * | sde[o] -> eng o+m+n| - * +--------------------+ + * \- +----------------------+ + * \- | mask | + * \ |----------------------| + * \- | sde[0] -> eng 1+m+n | + * \- |----------------------| + * >| sde[1] -> eng 2+m+n | + * |----------------------| + * | * | + * |----------------------| + * | sde[o-1] -> eng o+m+n| + * +----------------------+ * */ diff --git a/drivers/infiniband/hw/hfi1/sysfs.c b/drivers/infiniband/hw/hfi1/sysfs.c index 919a5474e651..50d140d25e38 100644 --- a/drivers/infiniband/hw/hfi1/sysfs.c +++ b/drivers/infiniband/hw/hfi1/sysfs.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -542,7 +542,7 @@ static ssize_t show_nctxts(struct device *device, * give a more accurate picture of total contexts available. */ return scnprintf(buf, PAGE_SIZE, "%u\n", - min(dd->num_rcv_contexts - dd->first_user_ctxt, + min(dd->num_rcv_contexts - dd->first_dyn_alloc_ctxt, (u32)dd->sc_sizes[SC_USER].count)); } diff --git a/drivers/infiniband/hw/hfi1/trace.c b/drivers/infiniband/hw/hfi1/trace.c index e86798af6903..eafae487face 100644 --- a/drivers/infiniband/hw/hfi1/trace.c +++ b/drivers/infiniband/hw/hfi1/trace.c @@ -51,13 +51,12 @@ u8 ibhdr_exhdr_len(struct ib_header *hdr) { struct ib_other_headers *ohdr; u8 opcode; - u8 lnh = (u8)(be16_to_cpu(hdr->lrh[0]) & 3); - if (lnh == HFI1_LRH_BTH) + if (ib_get_lnh(hdr) == HFI1_LRH_BTH) ohdr = &hdr->u.oth; else ohdr = &hdr->u.l.oth; - opcode = be32_to_cpu(ohdr->bth[0]) >> 24; + opcode = ib_bth_get_opcode(ohdr); return hdr_len_by_opcode[opcode] == 0 ? 0 : hdr_len_by_opcode[opcode] - (12 + 8); } diff --git a/drivers/infiniband/hw/hfi1/trace_ibhdrs.h b/drivers/infiniband/hw/hfi1/trace_ibhdrs.h index 382fcda3a5f6..090f6b506953 100644 --- a/drivers/infiniband/hw/hfi1/trace_ibhdrs.h +++ b/drivers/infiniband/hw/hfi1/trace_ibhdrs.h @@ -139,11 +139,11 @@ DECLARE_EVENT_CLASS(hfi1_ibhdr_template, __entry->pkey = be32_to_cpu(ohdr->bth[0]) & 0xffff; __entry->f = - (be32_to_cpu(ohdr->bth[1]) >> HFI1_FECN_SHIFT) & - HFI1_FECN_MASK; + (be32_to_cpu(ohdr->bth[1]) >> IB_FECN_SHIFT) & + IB_FECN_MASK; __entry->b = - (be32_to_cpu(ohdr->bth[1]) >> HFI1_BECN_SHIFT) & - HFI1_BECN_MASK; + (be32_to_cpu(ohdr->bth[1]) >> IB_BECN_SHIFT) & + IB_BECN_MASK; __entry->qpn = be32_to_cpu(ohdr->bth[1]) & RVT_QPN_MASK; __entry->a = diff --git a/drivers/infiniband/hw/hfi1/trace_misc.h b/drivers/infiniband/hw/hfi1/trace_misc.h index d308454af7fd..deac77ddaeab 100644 --- a/drivers/infiniband/hw/hfi1/trace_misc.h +++ b/drivers/infiniband/hw/hfi1/trace_misc.h @@ -72,6 +72,54 @@ TRACE_EVENT(hfi1_interrupt, __entry->src) ); +#ifdef CONFIG_FAULT_INJECTION +TRACE_EVENT(hfi1_fault_opcode, + TP_PROTO(struct rvt_qp *qp, u8 opcode), + TP_ARGS(qp, opcode), + TP_STRUCT__entry(DD_DEV_ENTRY(dd_from_ibdev(qp->ibqp.device)) + __field(u32, qpn) + __field(u8, opcode) + ), + TP_fast_assign(DD_DEV_ASSIGN(dd_from_ibdev(qp->ibqp.device)) + __entry->qpn = qp->ibqp.qp_num; + __entry->opcode = opcode; + ), + TP_printk("[%s] qpn 0x%x opcode 0x%x", + __get_str(dev), __entry->qpn, __entry->opcode) +); + +TRACE_EVENT(hfi1_fault_packet, + TP_PROTO(struct hfi1_packet *packet), + TP_ARGS(packet), + TP_STRUCT__entry(DD_DEV_ENTRY(packet->rcd->ppd->dd) + __field(u64, eflags) + __field(u32, ctxt) + __field(u32, hlen) + __field(u32, tlen) + __field(u32, updegr) + __field(u32, etail) + ), + TP_fast_assign(DD_DEV_ASSIGN(packet->rcd->ppd->dd); + __entry->eflags = rhf_err_flags(packet->rhf); + __entry->ctxt = packet->rcd->ctxt; + __entry->hlen = packet->hlen; + __entry->tlen = packet->tlen; + __entry->updegr = packet->updegr; + __entry->etail = rhf_egr_index(packet->rhf); + ), + TP_printk( + "[%s] ctxt %d eflags 0x%llx hlen %d tlen %d updegr %d etail %d", + __get_str(dev), + __entry->ctxt, + __entry->eflags, + __entry->hlen, + __entry->tlen, + __entry->updegr, + __entry->etail + ) +); +#endif + #endif /* __HFI1_TRACE_MISC_H */ #undef TRACE_INCLUDE_PATH diff --git a/drivers/infiniband/hw/hfi1/trace_rc.h b/drivers/infiniband/hw/hfi1/trace_rc.h index 5ea5005f9f41..8ce476570462 100644 --- a/drivers/infiniband/hw/hfi1/trace_rc.h +++ b/drivers/infiniband/hw/hfi1/trace_rc.h @@ -1,5 +1,5 @@ /* -* Copyright(c) 2015, 2016 Intel Corporation. +* Copyright(c) 2015, 2016, 2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -104,11 +104,6 @@ DEFINE_EVENT(hfi1_rc_template, hfi1_ack, TP_ARGS(qp, psn) ); -DEFINE_EVENT(hfi1_rc_template, hfi1_timeout, - TP_PROTO(struct rvt_qp *qp, u32 psn), - TP_ARGS(qp, psn) -); - DEFINE_EVENT(hfi1_rc_template, hfi1_rcv_error, TP_PROTO(struct rvt_qp *qp, u32 psn), TP_ARGS(qp, psn) diff --git a/drivers/infiniband/hw/hfi1/trace_tx.h b/drivers/infiniband/hw/hfi1/trace_tx.h index 415d6be42c5d..2c9ac57657d3 100644 --- a/drivers/infiniband/hw/hfi1/trace_tx.h +++ b/drivers/infiniband/hw/hfi1/trace_tx.h @@ -633,6 +633,49 @@ DEFINE_EVENT(hfi1_bct_template, bct_get, TP_PROTO(struct hfi1_devdata *dd, struct buffer_control *bc), TP_ARGS(dd, bc)); +TRACE_EVENT( + hfi1_qp_send_completion, + TP_PROTO(struct rvt_qp *qp, struct rvt_swqe *wqe, u32 idx), + TP_ARGS(qp, wqe, idx), + TP_STRUCT__entry( + DD_DEV_ENTRY(dd_from_ibdev(qp->ibqp.device)) + __field(struct rvt_swqe *, wqe) + __field(u64, wr_id) + __field(u32, qpn) + __field(u32, qpt) + __field(u32, length) + __field(u32, idx) + __field(u32, ssn) + __field(enum ib_wr_opcode, opcode) + __field(int, send_flags) + ), + TP_fast_assign( + DD_DEV_ASSIGN(dd_from_ibdev(qp->ibqp.device)) + __entry->wqe = wqe; + __entry->wr_id = wqe->wr.wr_id; + __entry->qpn = qp->ibqp.qp_num; + __entry->qpt = qp->ibqp.qp_type; + __entry->length = wqe->length; + __entry->idx = idx; + __entry->ssn = wqe->ssn; + __entry->opcode = wqe->wr.opcode; + __entry->send_flags = wqe->wr.send_flags; + ), + TP_printk( + "[%s] qpn 0x%x qpt %u wqe %p idx %u wr_id %llx length %u ssn %u opcode %x send_flags %x", + __get_str(dev), + __entry->qpn, + __entry->qpt, + __entry->wqe, + __entry->idx, + __entry->wr_id, + __entry->length, + __entry->ssn, + __entry->opcode, + __entry->send_flags + ) +); + #endif /* __HFI1_TRACE_TX_H */ #undef TRACE_INCLUDE_PATH diff --git a/drivers/infiniband/hw/hfi1/uc.c b/drivers/infiniband/hw/hfi1/uc.c index 4b2a8400c823..f0bdb100e005 100644 --- a/drivers/infiniband/hw/hfi1/uc.c +++ b/drivers/infiniband/hw/hfi1/uc.c @@ -320,7 +320,7 @@ void hfi1_uc_rcv(struct hfi1_packet *packet) process_ecn(qp, packet, true); psn = be32_to_cpu(ohdr->bth[2]); - opcode = (bth0 >> 24) & 0xff; + opcode = ib_bth_get_opcode(ohdr); /* Compare the PSN verses the expected PSN. */ if (unlikely(cmp_psn(psn, qp->r_psn) != 0)) { @@ -433,7 +433,7 @@ no_immediate_data: wc.wc_flags = 0; send_last: /* Get the number of bytes the message was padded by. */ - pad = (be32_to_cpu(ohdr->bth[0]) >> 20) & 3; + pad = ib_bth_get_pad(ohdr); /* Check for invalid length. */ /* LAST len should be >= 1 */ if (unlikely(tlen < (hdrsize + pad + 4))) @@ -528,7 +528,7 @@ rdma_last_imm: wc.wc_flags = IB_WC_WITH_IMM; /* Get the number of bytes the message was padded by. */ - pad = (be32_to_cpu(ohdr->bth[0]) >> 20) & 3; + pad = ib_bth_get_pad(ohdr); /* Check for invalid length. */ /* LAST len should be >= 1 */ if (unlikely(tlen < (hdrsize + pad + 4))) @@ -555,7 +555,7 @@ rdma_last_imm: case OP(RDMA_WRITE_LAST): rdma_last: /* Get the number of bytes the message was padded by. */ - pad = (be32_to_cpu(ohdr->bth[0]) >> 20) & 3; + pad = ib_bth_get_pad(ohdr); /* Check for invalid length. */ /* LAST len should be >= 1 */ if (unlikely(tlen < (hdrsize + pad + 4))) diff --git a/drivers/infiniband/hw/hfi1/ud.c b/drivers/infiniband/hw/hfi1/ud.c index 13ea4eb6ef3d..45bc3f04793e 100644 --- a/drivers/infiniband/hw/hfi1/ud.c +++ b/drivers/infiniband/hw/hfi1/ud.c @@ -537,7 +537,7 @@ void return_cnp(struct hfi1_ibport *ibp, struct rvt_qp *qp, u32 remote_qpn, bth0 = pkey | (IB_OPCODE_CNP << 24); ohdr->bth[0] = cpu_to_be32(bth0); - ohdr->bth[1] = cpu_to_be32(remote_qpn | (1 << HFI1_BECN_SHIFT)); + ohdr->bth[1] = cpu_to_be32(remote_qpn | (1 << IB_BECN_SHIFT)); ohdr->bth[2] = 0; /* PSN 0 */ hdr.lrh[0] = cpu_to_be16(lrh0); @@ -680,7 +680,7 @@ void hfi1_ud_rcv(struct hfi1_packet *packet) u32 tlen = packet->tlen; struct rvt_qp *qp = packet->qp; bool has_grh = rcv_flags & HFI1_HAS_GRH; - u8 sc5 = hdr2sc(hdr, packet->rhf); + u8 sc5 = hfi1_9B_get_sc5(hdr, packet->rhf); u32 bth1; u8 sl_from_sc, sl; u16 slid; @@ -688,18 +688,16 @@ void hfi1_ud_rcv(struct hfi1_packet *packet) qkey = be32_to_cpu(ohdr->u.ud.deth[0]); src_qp = be32_to_cpu(ohdr->u.ud.deth[1]) & RVT_QPN_MASK; - dlid = be16_to_cpu(hdr->lrh[1]); + dlid = ib_get_dlid(hdr); bth1 = be32_to_cpu(ohdr->bth[1]); - slid = be16_to_cpu(hdr->lrh[3]); - pkey = (u16)be32_to_cpu(ohdr->bth[0]); - sl = (be16_to_cpu(hdr->lrh[0]) >> 4) & 0xf; - extra_bytes = (be32_to_cpu(ohdr->bth[0]) >> 20) & 3; + slid = ib_get_slid(hdr); + pkey = ib_bth_get_pkey(ohdr); + opcode = ib_bth_get_opcode(ohdr); + sl = ib_get_sl(hdr); + extra_bytes = ib_bth_get_pad(ohdr); extra_bytes += (SIZE_OF_CRC << 2); sl_from_sc = ibp->sc_to_sl[sc5]; - opcode = be32_to_cpu(ohdr->bth[0]) >> 24; - opcode &= 0xff; - process_ecn(qp, packet, (opcode != IB_OPCODE_CNP)); /* * Get the number of bytes the message was padded by diff --git a/drivers/infiniband/hw/hfi1/user_exp_rcv.c b/drivers/infiniband/hw/hfi1/user_exp_rcv.c index 4a8295399e71..35c6e7ec8ad6 100644 --- a/drivers/infiniband/hw/hfi1/user_exp_rcv.c +++ b/drivers/infiniband/hw/hfi1/user_exp_rcv.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -200,8 +200,9 @@ int hfi1_user_exp_rcv_init(struct file *fp) if (!HFI1_CAP_UGET_MASK(uctxt->flags, TID_UNMAP)) { fd->invalid_tid_idx = 0; - fd->invalid_tids = kzalloc(uctxt->expected_count * - sizeof(u32), GFP_KERNEL); + fd->invalid_tids = kcalloc(uctxt->expected_count, + sizeof(*fd->invalid_tids), + GFP_KERNEL); if (!fd->invalid_tids) { ret = -ENOMEM; goto done; @@ -578,6 +579,9 @@ int hfi1_user_exp_rcv_clear(struct file *fp, struct hfi1_tid_info *tinfo) u32 *tidinfo; unsigned tididx; + if (unlikely(tinfo->tidcnt > fd->tid_used)) + return -EINVAL; + tidinfo = memdup_user((void __user *)(unsigned long)tinfo->tidlist, sizeof(tidinfo[0]) * tinfo->tidcnt); if (IS_ERR(tidinfo)) @@ -607,7 +611,7 @@ int hfi1_user_exp_rcv_invalid(struct file *fp, struct hfi1_tid_info *tinfo) struct hfi1_filedata *fd = fp->private_data; struct hfi1_ctxtdata *uctxt = fd->uctxt; unsigned long *ev = uctxt->dd->events + - (((uctxt->ctxt - uctxt->dd->first_user_ctxt) * + (((uctxt->ctxt - uctxt->dd->first_dyn_alloc_ctxt) * HFI1_MAX_SHARED_CTXTS) + fd->subctxt); u32 *array; int ret = 0; @@ -1011,8 +1015,8 @@ static int tid_rb_invalidate(void *arg, struct mmu_rb_node *mnode) * process in question. */ ev = uctxt->dd->events + - (((uctxt->ctxt - uctxt->dd->first_user_ctxt) * - HFI1_MAX_SHARED_CTXTS) + fdata->subctxt); + (((uctxt->ctxt - uctxt->dd->first_dyn_alloc_ctxt) * + HFI1_MAX_SHARED_CTXTS) + fdata->subctxt); set_bit(_HFI1_EVENT_TID_MMU_NOTIFY_BIT, ev); } fdata->invalid_tid_idx++; diff --git a/drivers/infiniband/hw/hfi1/user_pages.c b/drivers/infiniband/hw/hfi1/user_pages.c index 68295a12b771..e341e6dcc388 100644 --- a/drivers/infiniband/hw/hfi1/user_pages.c +++ b/drivers/infiniband/hw/hfi1/user_pages.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015-2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -73,7 +73,8 @@ bool hfi1_can_pin_pages(struct hfi1_devdata *dd, struct mm_struct *mm, { unsigned long ulimit = rlimit(RLIMIT_MEMLOCK), pinned, cache_limit, size = (cache_size * (1UL << 20)); /* convert to bytes */ - unsigned usr_ctxts = dd->num_rcv_contexts - dd->first_user_ctxt; + unsigned int usr_ctxts = + dd->num_rcv_contexts - dd->first_dyn_alloc_ctxt; bool can_lock = capable(CAP_IPC_LOCK); /* diff --git a/drivers/infiniband/hw/hfi1/user_sdma.c b/drivers/infiniband/hw/hfi1/user_sdma.c index e6811c4edc73..0749689d7643 100644 --- a/drivers/infiniband/hw/hfi1/user_sdma.c +++ b/drivers/infiniband/hw/hfi1/user_sdma.c @@ -376,7 +376,6 @@ int hfi1_user_sdma_alloc_queues(struct hfi1_ctxtdata *uctxt, struct file *fp) { struct hfi1_filedata *fd; int ret = 0; - unsigned memsize; char buf[64]; struct hfi1_devdata *dd; struct hfi1_user_sdma_comp_q *cq; @@ -401,13 +400,15 @@ int hfi1_user_sdma_alloc_queues(struct hfi1_ctxtdata *uctxt, struct file *fp) if (!pq) goto pq_nomem; - memsize = sizeof(*pq->reqs) * hfi1_sdma_comp_ring_size; - pq->reqs = kzalloc(memsize, GFP_KERNEL); + pq->reqs = kcalloc(hfi1_sdma_comp_ring_size, + sizeof(*pq->reqs), + GFP_KERNEL); if (!pq->reqs) goto pq_reqs_nomem; - memsize = BITS_TO_LONGS(hfi1_sdma_comp_ring_size) * sizeof(long); - pq->req_in_use = kzalloc(memsize, GFP_KERNEL); + pq->req_in_use = kcalloc(BITS_TO_LONGS(hfi1_sdma_comp_ring_size), + sizeof(*pq->req_in_use), + GFP_KERNEL); if (!pq->req_in_use) goto pq_reqs_no_in_use; @@ -442,8 +443,8 @@ int hfi1_user_sdma_alloc_queues(struct hfi1_ctxtdata *uctxt, struct file *fp) if (!cq) goto cq_nomem; - memsize = PAGE_ALIGN(sizeof(*cq->comps) * hfi1_sdma_comp_ring_size); - cq->comps = vmalloc_user(memsize); + cq->comps = vmalloc_user(PAGE_ALIGN(sizeof(*cq->comps) + * hfi1_sdma_comp_ring_size)); if (!cq->comps) goto cq_comps_nomem; @@ -704,7 +705,9 @@ int hfi1_user_sdma_process_request(struct file *fp, struct iovec *iovec, /* Save all the IO vector structures */ for (i = 0; i < req->data_iovs; i++) { INIT_LIST_HEAD(&req->iovs[i].list); - memcpy(&req->iovs[i].iov, iovec + idx++, sizeof(struct iovec)); + memcpy(&req->iovs[i].iov, + iovec + idx++, + sizeof(req->iovs[i].iov)); ret = pin_vector_pages(req, &req->iovs[i]); if (ret) { req->status = ret; @@ -1615,9 +1618,10 @@ static inline void set_comp_state(struct hfi1_user_sdma_pkt_q *pq, { hfi1_cdbg(SDMA, "[%u:%u:%u:%u] Setting completion status %u %d", pq->dd->unit, pq->ctxt, pq->subctxt, idx, state, ret); - cq->comps[idx].status = state; if (state == ERROR) cq->comps[idx].errcode = -ret; + smp_wmb(); /* make sure errcode is visible first */ + cq->comps[idx].status = state; trace_hfi1_sdma_user_completion(pq->dd, pq->ctxt, pq->subctxt, idx, state, ret); } diff --git a/drivers/infiniband/hw/hfi1/verbs.c b/drivers/infiniband/hw/hfi1/verbs.c index 222315fadab1..7174a18ebaac 100644 --- a/drivers/infiniband/hw/hfi1/verbs.c +++ b/drivers/infiniband/hw/hfi1/verbs.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015 - 2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -60,6 +60,8 @@ #include "trace.h" #include "qp.h" #include "verbs_txreq.h" +#include "debugfs.h" +#include "vnic.h" static unsigned int hfi1_lkey_table_size = 16; module_param_named(lkey_table_size, hfi1_lkey_table_size, uint, @@ -297,6 +299,22 @@ static inline bool wss_exceeds_threshold(void) } /* + * Translate ib_wr_opcode into ib_wc_opcode. + */ +const enum ib_wc_opcode ib_hfi1_wc_opcode[] = { + [IB_WR_RDMA_WRITE] = IB_WC_RDMA_WRITE, + [IB_WR_RDMA_WRITE_WITH_IMM] = IB_WC_RDMA_WRITE, + [IB_WR_SEND] = IB_WC_SEND, + [IB_WR_SEND_WITH_IMM] = IB_WC_SEND, + [IB_WR_RDMA_READ] = IB_WC_RDMA_READ, + [IB_WR_ATOMIC_CMP_AND_SWP] = IB_WC_COMP_SWAP, + [IB_WR_ATOMIC_FETCH_AND_ADD] = IB_WC_FETCH_ADD, + [IB_WR_SEND_WITH_INV] = IB_WC_SEND, + [IB_WR_LOCAL_INV] = IB_WC_LOCAL_INV, + [IB_WR_REG_MR] = IB_WC_REG_MR +}; + +/* * Length of header by opcode, 0 --> not supported */ const u8 hdr_len_by_opcode[256] = { @@ -501,6 +519,35 @@ static inline opcode_handler qp_ok(int opcode, struct hfi1_packet *packet) return NULL; } +static u64 hfi1_fault_tx(struct rvt_qp *qp, u8 opcode, u64 pbc) +{ +#ifdef CONFIG_FAULT_INJECTION + if ((opcode & IB_OPCODE_MSP) == IB_OPCODE_MSP) + /* + * In order to drop non-IB traffic we + * set PbcInsertHrc to NONE (0x2). + * The packet will still be delivered + * to the receiving node but a + * KHdrHCRCErr (KDETH packet with a bad + * HCRC) will be triggered and the + * packet will not be delivered to the + * correct context. + */ + pbc |= (u64)PBC_IHCRC_NONE << PBC_INSERT_HCRC_SHIFT; + else + /* + * In order to drop regular verbs + * traffic we set the PbcTestEbp + * flag. The packet will still be + * delivered to the receiving node but + * a 'late ebp error' will be + * triggered and will be dropped. + */ + pbc |= PBC_TEST_EBP; +#endif + return pbc; +} + /** * hfi1_ib_rcv - process an incoming packet * @packet: data packet information @@ -525,7 +572,7 @@ void hfi1_ib_rcv(struct hfi1_packet *packet) u16 lid; /* Check for GRH */ - lnh = be16_to_cpu(hdr->lrh[0]) & 3; + lnh = ib_get_lnh(hdr); if (lnh == HFI1_LRH_BTH) { packet->ohdr = &hdr->u.oth; } else if (lnh == HFI1_LRH_GRH) { @@ -544,12 +591,12 @@ void hfi1_ib_rcv(struct hfi1_packet *packet) trace_input_ibhdr(rcd->dd, hdr); - opcode = (be32_to_cpu(packet->ohdr->bth[0]) >> 24); + opcode = ib_bth_get_opcode(packet->ohdr); inc_opstats(tlen, &rcd->opstats->stats[opcode]); /* Get the destination QP number. */ qp_num = be32_to_cpu(packet->ohdr->bth[1]) & RVT_QPN_MASK; - lid = be16_to_cpu(hdr->lrh[1]); + lid = ib_get_dlid(hdr); if (unlikely((lid >= be16_to_cpu(IB_MULTICAST_LID_BASE)) && (lid != be16_to_cpu(IB_LID_PERMISSIVE)))) { struct rvt_mcast *mcast; @@ -557,7 +604,7 @@ void hfi1_ib_rcv(struct hfi1_packet *packet) if (lnh != HFI1_LRH_GRH) goto drop; - mcast = rvt_mcast_find(&ibp->rvp, &hdr->u.l.grh.dgid); + mcast = rvt_mcast_find(&ibp->rvp, &hdr->u.l.grh.dgid, lid); if (!mcast) goto drop; list_for_each_entry_rcu(p, &mcast->qp_list, list) { @@ -583,6 +630,11 @@ void hfi1_ib_rcv(struct hfi1_packet *packet) rcu_read_unlock(); goto drop; } + if (unlikely(hfi1_dbg_fault_opcode(packet->qp, opcode, + true))) { + rcu_read_unlock(); + goto drop; + } spin_lock_irqsave(&packet->qp->r_lock, flags); packet_handler = qp_ok(opcode, packet); if (likely(packet_handler)) @@ -781,7 +833,6 @@ static int build_verbs_tx_desc( if (ret) goto bail_txadd; } - /* add the ulp payload - if any. tx->ss can be NULL for acks */ if (tx->ss) ret = build_verbs_ulp_payload(sde, length, tx); @@ -800,7 +851,6 @@ int hfi1_verbs_send_dma(struct rvt_qp *qp, struct hfi1_pkt_state *ps, struct hfi1_ibdev *dev = ps->dev; struct hfi1_pportdata *ppd = ps->ppd; struct verbs_txreq *tx; - u64 pbc_flags = 0; u8 sc5 = priv->s_sc; int ret; @@ -809,12 +859,16 @@ int hfi1_verbs_send_dma(struct rvt_qp *qp, struct hfi1_pkt_state *ps, if (!sdma_txreq_built(&tx->txreq)) { if (likely(pbc == 0)) { u32 vl = sc_to_vlt(dd_from_ibdev(qp->ibqp.device), sc5); + u8 opcode = get_opcode(&tx->phdr.hdr); + /* No vl15 here */ /* set PBC_DC_INFO bit (aka SC[4]) in pbc_flags */ - pbc_flags |= (!!(sc5 & 0x10)) << PBC_DC_INFO_SHIFT; + pbc |= (!!(sc5 & 0x10)) << PBC_DC_INFO_SHIFT; + if (unlikely(hfi1_dbg_fault_opcode(qp, opcode, false))) + pbc = hfi1_fault_tx(qp, opcode, pbc); pbc = create_pbc(ppd, - pbc_flags, + pbc, qp->srate_mbps, vl, plen); @@ -917,7 +971,6 @@ int hfi1_verbs_send_pio(struct rvt_qp *qp, struct hfi1_pkt_state *ps, u32 plen = hdrwords + dwords + 2; /* includes pbc */ struct hfi1_pportdata *ppd = ps->ppd; u32 *hdr = (u32 *)&ps->s_txreq->phdr.hdr; - u64 pbc_flags = 0; u8 sc5; unsigned long flags = 0; struct send_context *sc; @@ -942,9 +995,14 @@ int hfi1_verbs_send_pio(struct rvt_qp *qp, struct hfi1_pkt_state *ps, if (likely(pbc == 0)) { u8 vl = sc_to_vlt(dd_from_ibdev(qp->ibqp.device), sc5); + struct verbs_txreq *tx = ps->s_txreq; + u8 opcode = get_opcode(&tx->phdr.hdr); + /* set PBC_DC_INFO bit (aka SC[4]) in pbc_flags */ - pbc_flags |= (!!(sc5 & 0x10)) << PBC_DC_INFO_SHIFT; - pbc = create_pbc(ppd, pbc_flags, qp->srate_mbps, vl, plen); + pbc |= (!!(sc5 & 0x10)) << PBC_DC_INFO_SHIFT; + if (unlikely(hfi1_dbg_fault_opcode(qp, opcode, false))) + pbc = hfi1_fault_tx(qp, opcode, pbc); + pbc = create_pbc(ppd, pbc, qp->srate_mbps, vl, plen); } if (cb) iowait_pio_inc(&priv->s_iowait); @@ -1173,7 +1231,7 @@ int hfi1_verbs_send(struct rvt_qp *qp, struct hfi1_pkt_state *ps) hdr = &ps->s_txreq->phdr.hdr; /* locate the pkey within the headers */ - lnh = be16_to_cpu(hdr->lrh[0]) & 3; + lnh = ib_get_lnh(hdr); if (lnh == HFI1_LRH_GRH) ohdr = &hdr->u.l.oth; else @@ -1220,17 +1278,20 @@ int hfi1_verbs_send(struct rvt_qp *qp, struct hfi1_pkt_state *ps) static void hfi1_fill_device_attr(struct hfi1_devdata *dd) { struct rvt_dev_info *rdi = &dd->verbs_dev.rdi; - u16 ver = dd->dc8051_ver; + u32 ver = dd->dc8051_ver; memset(&rdi->dparms.props, 0, sizeof(rdi->dparms.props)); - rdi->dparms.props.fw_ver = ((u64)(dc8051_ver_maj(ver)) << 16) | - (u64)dc8051_ver_min(ver); + rdi->dparms.props.fw_ver = ((u64)(dc8051_ver_maj(ver)) << 32) | + ((u64)(dc8051_ver_min(ver)) << 16) | + (u64)dc8051_ver_patch(ver); + rdi->dparms.props.device_cap_flags = IB_DEVICE_BAD_PKEY_CNTR | IB_DEVICE_BAD_QKEY_CNTR | IB_DEVICE_SHUTDOWN_PORT | IB_DEVICE_SYS_IMAGE_GUID | IB_DEVICE_RC_RNR_NAK_GEN | IB_DEVICE_PORT_ACTIVE_EVENT | IB_DEVICE_SRQ_RESIZE | - IB_DEVICE_MEM_MGT_EXTENSIONS; + IB_DEVICE_MEM_MGT_EXTENSIONS | + IB_DEVICE_RDMA_NETDEV_OPA_VNIC; rdi->dparms.props.page_size_cap = PAGE_SIZE; rdi->dparms.props.vendor_id = dd->oui1 << 16 | dd->oui2 << 8 | dd->oui3; rdi->dparms.props.vendor_part_id = dd->pcidev->device; @@ -1504,10 +1565,10 @@ static void hfi1_get_dev_fw_str(struct ib_device *ibdev, char *str, { struct rvt_dev_info *rdi = ib_to_rvt(ibdev); struct hfi1_ibdev *dev = dev_from_rdi(rdi); - u16 ver = dd_from_dev(dev)->dc8051_ver; + u32 ver = dd_from_dev(dev)->dc8051_ver; - snprintf(str, str_len, "%u.%u", dc8051_ver_maj(ver), - dc8051_ver_min(ver)); + snprintf(str, str_len, "%u.%u.%u", dc8051_ver_maj(ver), + dc8051_ver_min(ver), dc8051_ver_patch(ver)); } static const char * const driver_cntr_names[] = { @@ -1524,6 +1585,7 @@ static const char * const driver_cntr_names[] = { "DRIVER_EgrHdrFull" }; +static DEFINE_MUTEX(cntr_names_lock); /* protects the *_cntr_names bufers */ static const char **dev_cntr_names; static const char **port_cntr_names; static int num_driver_cntrs = ARRAY_SIZE(driver_cntr_names); @@ -1578,6 +1640,7 @@ static struct rdma_hw_stats *alloc_hw_stats(struct ib_device *ibdev, { int i, err; + mutex_lock(&cntr_names_lock); if (!cntr_names_initialized) { struct hfi1_devdata *dd = dd_from_ibdev(ibdev); @@ -1586,8 +1649,10 @@ static struct rdma_hw_stats *alloc_hw_stats(struct ib_device *ibdev, num_driver_cntrs, &num_dev_cntrs, &dev_cntr_names); - if (err) + if (err) { + mutex_unlock(&cntr_names_lock); return NULL; + } for (i = 0; i < num_driver_cntrs; i++) dev_cntr_names[num_dev_cntrs + i] = @@ -1601,10 +1666,12 @@ static struct rdma_hw_stats *alloc_hw_stats(struct ib_device *ibdev, if (err) { kfree(dev_cntr_names); dev_cntr_names = NULL; + mutex_unlock(&cntr_names_lock); return NULL; } cntr_names_initialized = 1; } + mutex_unlock(&cntr_names_lock); if (!port_num) return rdma_alloc_hw_stats_struct( @@ -1707,6 +1774,8 @@ int hfi1_register_ib_device(struct hfi1_devdata *dd) ibdev->modify_device = modify_device; 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; @@ -1751,7 +1820,7 @@ int hfi1_register_ib_device(struct hfi1_devdata *dd) dd->verbs_dev.rdi.driver_f.qp_priv_free = qp_priv_free; dd->verbs_dev.rdi.driver_f.free_all_qps = free_all_qps; dd->verbs_dev.rdi.driver_f.notify_qp_reset = notify_qp_reset; - dd->verbs_dev.rdi.driver_f.do_send = hfi1_do_send; + dd->verbs_dev.rdi.driver_f.do_send = hfi1_do_send_from_rvt; dd->verbs_dev.rdi.driver_f.schedule_send = hfi1_schedule_send; dd->verbs_dev.rdi.driver_f.schedule_send_no_lock = _hfi1_schedule_send; dd->verbs_dev.rdi.driver_f.get_pmtu_from_attr = get_pmtu_from_attr; @@ -1823,9 +1892,13 @@ void hfi1_unregister_ib_device(struct hfi1_devdata *dd) del_timer_sync(&dev->mem_timer); verbs_txreq_exit(dev); + mutex_lock(&cntr_names_lock); kfree(dev_cntr_names); kfree(port_cntr_names); + dev_cntr_names = NULL; + port_cntr_names = NULL; cntr_names_initialized = 0; + mutex_unlock(&cntr_names_lock); } void hfi1_cnp_rcv(struct hfi1_packet *packet) @@ -1859,7 +1932,7 @@ void hfi1_cnp_rcv(struct hfi1_packet *packet) return; } - sc5 = hdr2sc(hdr, packet->rhf); + sc5 = hfi1_9B_get_sc5(hdr, packet->rhf); sl = ibp->sc_to_sl[sc5]; lqpn = qp->ibqp.qp_num; diff --git a/drivers/infiniband/hw/hfi1/verbs.h b/drivers/infiniband/hw/hfi1/verbs.h index 3a0b589e41c2..46b00ed9f2dc 100644 --- a/drivers/infiniband/hw/hfi1/verbs.h +++ b/drivers/infiniband/hw/hfi1/verbs.h @@ -1,5 +1,5 @@ /* - * Copyright(c) 2015, 2016 Intel Corporation. + * Copyright(c) 2015 - 2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -195,6 +195,11 @@ struct hfi1_ibdev { struct dentry *hfi1_ibdev_dbg; /* per HFI symlinks to above */ struct dentry *hfi1_ibdev_link; +#ifdef CONFIG_FAULT_INJECTION + struct fault_opcode *fault_opcode; + struct fault_packet *fault_packet; + bool fault_suppress_err; +#endif #endif }; @@ -350,7 +355,9 @@ void hfi1_make_ruc_header(struct rvt_qp *qp, struct ib_other_headers *ohdr, void _hfi1_do_send(struct work_struct *work); -void hfi1_do_send(struct rvt_qp *qp); +void hfi1_do_send_from_rvt(struct rvt_qp *qp); + +void hfi1_do_send(struct rvt_qp *qp, bool in_thread); void hfi1_send_complete(struct rvt_qp *qp, struct rvt_swqe *wqe, enum ib_wc_status status); diff --git a/drivers/infiniband/hw/hfi1/vnic.h b/drivers/infiniband/hw/hfi1/vnic.h new file mode 100644 index 000000000000..e2c455299b53 --- /dev/null +++ b/drivers/infiniband/hw/hfi1/vnic.h @@ -0,0 +1,184 @@ +#ifndef _HFI1_VNIC_H +#define _HFI1_VNIC_H +/* + * Copyright(c) 2017 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ + +#include <rdma/opa_vnic.h> +#include "hfi.h" +#include "sdma.h" + +#define HFI1_VNIC_MAX_TXQ 16 +#define HFI1_VNIC_MAX_PAD 12 + +/* L2 header definitions */ +#define HFI1_L2_TYPE_OFFSET 0x7 +#define HFI1_L2_TYPE_SHFT 0x5 +#define HFI1_L2_TYPE_MASK 0x3 + +#define HFI1_GET_L2_TYPE(hdr) \ + ((*((u8 *)(hdr) + HFI1_L2_TYPE_OFFSET) >> HFI1_L2_TYPE_SHFT) & \ + HFI1_L2_TYPE_MASK) + +/* L4 type definitions */ +#define HFI1_L4_TYPE_OFFSET 8 + +#define HFI1_GET_L4_TYPE(data) \ + (*((u8 *)(data) + HFI1_L4_TYPE_OFFSET)) + +/* L4 header definitions */ +#define HFI1_VNIC_L4_HDR_OFFSET OPA_VNIC_L2_HDR_LEN + +#define HFI1_VNIC_GET_L4_HDR(data) \ + (*((u16 *)((u8 *)(data) + HFI1_VNIC_L4_HDR_OFFSET))) + +#define HFI1_VNIC_GET_VESWID(data) \ + (HFI1_VNIC_GET_L4_HDR(data) & 0xFFF) + +/* Service class */ +#define HFI1_VNIC_SC_OFFSET_LOW 6 +#define HFI1_VNIC_SC_OFFSET_HI 7 +#define HFI1_VNIC_SC_SHIFT 4 + +#define HFI1_VNIC_MAX_QUEUE 16 + +/** + * struct hfi1_vnic_sdma - VNIC per Tx ring SDMA information + * @dd - device data pointer + * @sde - sdma engine + * @vinfo - vnic info pointer + * @wait - iowait structure + * @stx - sdma tx request + * @state - vnic Tx ring SDMA state + * @q_idx - vnic Tx queue index + */ +struct hfi1_vnic_sdma { + struct hfi1_devdata *dd; + struct sdma_engine *sde; + struct hfi1_vnic_vport_info *vinfo; + struct iowait wait; + struct sdma_txreq stx; + unsigned int state; + u8 q_idx; +}; + +/** + * struct hfi1_vnic_rx_queue - HFI1 VNIC receive queue + * @idx: queue index + * @vinfo: pointer to vport information + * @netdev: network device + * @napi: netdev napi structure + * @skbq: queue of received socket buffers + */ +struct hfi1_vnic_rx_queue { + u8 idx; + struct hfi1_vnic_vport_info *vinfo; + struct net_device *netdev; + struct napi_struct napi; + struct sk_buff_head skbq; +}; + +/** + * struct hfi1_vnic_vport_info - HFI1 VNIC virtual port information + * @dd: device data pointer + * @netdev: net device pointer + * @flags: state flags + * @lock: vport lock + * @num_tx_q: number of transmit queues + * @num_rx_q: number of receive queues + * @vesw_id: virtual switch id + * @rxq: Array of receive queues + * @stats: per queue stats + * @sdma: VNIC SDMA structure per TXQ + */ +struct hfi1_vnic_vport_info { + struct hfi1_devdata *dd; + struct net_device *netdev; + unsigned long flags; + + /* Lock used around state updates */ + struct mutex lock; + + u8 num_tx_q; + u8 num_rx_q; + u16 vesw_id; + struct hfi1_vnic_rx_queue rxq[HFI1_NUM_VNIC_CTXT]; + + struct opa_vnic_stats stats[HFI1_VNIC_MAX_QUEUE]; + struct hfi1_vnic_sdma sdma[HFI1_VNIC_MAX_TXQ]; +}; + +#define v_dbg(format, arg...) \ + netdev_dbg(vinfo->netdev, format, ## arg) +#define v_err(format, arg...) \ + netdev_err(vinfo->netdev, format, ## arg) +#define v_info(format, arg...) \ + netdev_info(vinfo->netdev, format, ## arg) + +/* vnic hfi1 internal functions */ +void hfi1_vnic_setup(struct hfi1_devdata *dd); +void hfi1_vnic_cleanup(struct hfi1_devdata *dd); +int hfi1_vnic_txreq_init(struct hfi1_devdata *dd); +void hfi1_vnic_txreq_deinit(struct hfi1_devdata *dd); + +void hfi1_vnic_bypass_rcv(struct hfi1_packet *packet); +void hfi1_vnic_sdma_init(struct hfi1_vnic_vport_info *vinfo); +bool hfi1_vnic_sdma_write_avail(struct hfi1_vnic_vport_info *vinfo, + u8 q_idx); + +/* vnic rdma netdev operations */ +struct net_device *hfi1_vnic_alloc_rn(struct ib_device *device, + u8 port_num, + enum rdma_netdev_t type, + 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); + +#endif /* _HFI1_VNIC_H */ diff --git a/drivers/infiniband/hw/hfi1/vnic_main.c b/drivers/infiniband/hw/hfi1/vnic_main.c new file mode 100644 index 000000000000..392f4d57f3e3 --- /dev/null +++ b/drivers/infiniband/hw/hfi1/vnic_main.c @@ -0,0 +1,907 @@ +/* + * Copyright(c) 2017 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ + +/* + * This file contains HFI1 support for VNIC functionality + */ + +#include <linux/io.h> +#include <linux/if_vlan.h> + +#include "vnic.h" + +#define HFI_TX_TIMEOUT_MS 1000 + +#define HFI1_VNIC_RCV_Q_SIZE 1024 + +#define HFI1_VNIC_UP 0 + +static DEFINE_SPINLOCK(vport_cntr_lock); + +static int setup_vnic_ctxt(struct hfi1_devdata *dd, struct hfi1_ctxtdata *uctxt) +{ + unsigned int rcvctrl_ops = 0; + int ret; + + ret = hfi1_init_ctxt(uctxt->sc); + if (ret) + goto done; + + uctxt->do_interrupt = &handle_receive_interrupt; + + /* Now allocate the RcvHdr queue and eager buffers. */ + ret = hfi1_create_rcvhdrq(dd, uctxt); + if (ret) + goto done; + + ret = hfi1_setup_eagerbufs(uctxt); + if (ret) + goto done; + + set_bit(HFI1_CTXT_SETUP_DONE, &uctxt->event_flags); + + if (uctxt->rcvhdrtail_kvaddr) + clear_rcvhdrtail(uctxt); + + rcvctrl_ops = HFI1_RCVCTRL_CTXT_ENB; + rcvctrl_ops |= HFI1_RCVCTRL_INTRAVAIL_ENB; + + if (!HFI1_CAP_KGET_MASK(uctxt->flags, MULTI_PKT_EGR)) + rcvctrl_ops |= HFI1_RCVCTRL_ONE_PKT_EGR_ENB; + if (HFI1_CAP_KGET_MASK(uctxt->flags, NODROP_EGR_FULL)) + rcvctrl_ops |= HFI1_RCVCTRL_NO_EGR_DROP_ENB; + if (HFI1_CAP_KGET_MASK(uctxt->flags, NODROP_RHQ_FULL)) + rcvctrl_ops |= HFI1_RCVCTRL_NO_RHQ_DROP_ENB; + if (HFI1_CAP_KGET_MASK(uctxt->flags, DMA_RTAIL)) + rcvctrl_ops |= HFI1_RCVCTRL_TAILUPD_ENB; + + hfi1_rcvctrl(uctxt->dd, rcvctrl_ops, uctxt->ctxt); + + uctxt->is_vnic = true; +done: + return ret; +} + +static int allocate_vnic_ctxt(struct hfi1_devdata *dd, + struct hfi1_ctxtdata **vnic_ctxt) +{ + struct hfi1_ctxtdata *uctxt; + unsigned int ctxt; + int ret; + + if (dd->flags & HFI1_FROZEN) + return -EIO; + + for (ctxt = dd->first_dyn_alloc_ctxt; + ctxt < dd->num_rcv_contexts; ctxt++) + if (!dd->rcd[ctxt]) + break; + + if (ctxt == dd->num_rcv_contexts) + return -EBUSY; + + uctxt = hfi1_create_ctxtdata(dd->pport, ctxt, dd->node); + if (!uctxt) { + dd_dev_err(dd, "Unable to create ctxtdata, failing open\n"); + return -ENOMEM; + } + + uctxt->flags = HFI1_CAP_KGET(MULTI_PKT_EGR) | + HFI1_CAP_KGET(NODROP_RHQ_FULL) | + HFI1_CAP_KGET(NODROP_EGR_FULL) | + HFI1_CAP_KGET(DMA_RTAIL); + uctxt->seq_cnt = 1; + + /* Allocate and enable a PIO send context */ + uctxt->sc = sc_alloc(dd, SC_VNIC, uctxt->rcvhdrqentsize, + uctxt->numa_id); + + ret = uctxt->sc ? 0 : -ENOMEM; + if (ret) + goto bail; + + dd_dev_dbg(dd, "allocated vnic send context %u(%u)\n", + uctxt->sc->sw_index, uctxt->sc->hw_context); + ret = sc_enable(uctxt->sc); + if (ret) + goto bail; + + if (dd->num_msix_entries) + hfi1_set_vnic_msix_info(uctxt); + + hfi1_stats.sps_ctxts++; + dd_dev_dbg(dd, "created vnic context %d\n", uctxt->ctxt); + *vnic_ctxt = uctxt; + + return ret; +bail: + /* + * hfi1_free_ctxtdata() also releases send_context + * structure if uctxt->sc is not null + */ + dd->rcd[uctxt->ctxt] = NULL; + hfi1_free_ctxtdata(dd, uctxt); + dd_dev_dbg(dd, "vnic allocation failed. rc %d\n", ret); + return ret; +} + +static void deallocate_vnic_ctxt(struct hfi1_devdata *dd, + struct hfi1_ctxtdata *uctxt) +{ + unsigned long flags; + + dd_dev_dbg(dd, "closing vnic context %d\n", uctxt->ctxt); + flush_wc(); + + if (dd->num_msix_entries) + hfi1_reset_vnic_msix_info(uctxt); + + spin_lock_irqsave(&dd->uctxt_lock, flags); + /* + * Disable receive context and interrupt available, reset all + * RcvCtxtCtrl bits to default values. + */ + hfi1_rcvctrl(dd, HFI1_RCVCTRL_CTXT_DIS | + HFI1_RCVCTRL_TIDFLOW_DIS | + HFI1_RCVCTRL_INTRAVAIL_DIS | + HFI1_RCVCTRL_ONE_PKT_EGR_DIS | + HFI1_RCVCTRL_NO_RHQ_DROP_DIS | + HFI1_RCVCTRL_NO_EGR_DROP_DIS, uctxt->ctxt); + /* + * VNIC contexts are allocated from user context pool. + * Release them back to user context pool. + * + * Reset context integrity checks to default. + * (writes to CSRs probably belong in chip.c) + */ + write_kctxt_csr(dd, uctxt->sc->hw_context, SEND_CTXT_CHECK_ENABLE, + hfi1_pkt_default_send_ctxt_mask(dd, SC_USER)); + sc_disable(uctxt->sc); + + dd->send_contexts[uctxt->sc->sw_index].type = SC_USER; + spin_unlock_irqrestore(&dd->uctxt_lock, flags); + + dd->rcd[uctxt->ctxt] = NULL; + uctxt->event_flags = 0; + + hfi1_clear_tids(uctxt); + hfi1_clear_ctxt_pkey(dd, uctxt->ctxt); + + hfi1_stats.sps_ctxts--; + hfi1_free_ctxtdata(dd, uctxt); +} + +void hfi1_vnic_setup(struct hfi1_devdata *dd) +{ + idr_init(&dd->vnic.vesw_idr); +} + +void hfi1_vnic_cleanup(struct hfi1_devdata *dd) +{ + idr_destroy(&dd->vnic.vesw_idr); +} + +#define SUM_GRP_COUNTERS(stats, qstats, x_grp) do { \ + u64 *src64, *dst64; \ + for (src64 = &qstats->x_grp.unicast, \ + dst64 = &stats->x_grp.unicast; \ + dst64 <= &stats->x_grp.s_1519_max;) { \ + *dst64++ += *src64++; \ + } \ + } while (0) + +/* hfi1_vnic_update_stats - update statistics */ +static void hfi1_vnic_update_stats(struct hfi1_vnic_vport_info *vinfo, + struct opa_vnic_stats *stats) +{ + struct net_device *netdev = vinfo->netdev; + u8 i; + + /* add tx counters on different queues */ + for (i = 0; i < vinfo->num_tx_q; i++) { + struct opa_vnic_stats *qstats = &vinfo->stats[i]; + struct rtnl_link_stats64 *qnstats = &vinfo->stats[i].netstats; + + stats->netstats.tx_fifo_errors += qnstats->tx_fifo_errors; + stats->netstats.tx_carrier_errors += qnstats->tx_carrier_errors; + stats->tx_drop_state += qstats->tx_drop_state; + stats->tx_dlid_zero += qstats->tx_dlid_zero; + + SUM_GRP_COUNTERS(stats, qstats, tx_grp); + stats->netstats.tx_packets += qnstats->tx_packets; + stats->netstats.tx_bytes += qnstats->tx_bytes; + } + + /* add rx counters on different queues */ + for (i = 0; i < vinfo->num_rx_q; i++) { + struct opa_vnic_stats *qstats = &vinfo->stats[i]; + struct rtnl_link_stats64 *qnstats = &vinfo->stats[i].netstats; + + stats->netstats.rx_fifo_errors += qnstats->rx_fifo_errors; + stats->netstats.rx_nohandler += qnstats->rx_nohandler; + stats->rx_drop_state += qstats->rx_drop_state; + stats->rx_oversize += qstats->rx_oversize; + stats->rx_runt += qstats->rx_runt; + + SUM_GRP_COUNTERS(stats, qstats, rx_grp); + stats->netstats.rx_packets += qnstats->rx_packets; + stats->netstats.rx_bytes += qnstats->rx_bytes; + } + + stats->netstats.tx_errors = stats->netstats.tx_fifo_errors + + stats->netstats.tx_carrier_errors + + stats->tx_drop_state + stats->tx_dlid_zero; + stats->netstats.tx_dropped = stats->netstats.tx_errors; + + stats->netstats.rx_errors = stats->netstats.rx_fifo_errors + + stats->netstats.rx_nohandler + + stats->rx_drop_state + stats->rx_oversize + + stats->rx_runt; + stats->netstats.rx_dropped = stats->netstats.rx_errors; + + netdev->stats.tx_packets = stats->netstats.tx_packets; + netdev->stats.tx_bytes = stats->netstats.tx_bytes; + netdev->stats.tx_fifo_errors = stats->netstats.tx_fifo_errors; + netdev->stats.tx_carrier_errors = stats->netstats.tx_carrier_errors; + netdev->stats.tx_errors = stats->netstats.tx_errors; + netdev->stats.tx_dropped = stats->netstats.tx_dropped; + + netdev->stats.rx_packets = stats->netstats.rx_packets; + netdev->stats.rx_bytes = stats->netstats.rx_bytes; + netdev->stats.rx_fifo_errors = stats->netstats.rx_fifo_errors; + netdev->stats.multicast = stats->rx_grp.mcastbcast; + netdev->stats.rx_length_errors = stats->rx_oversize + stats->rx_runt; + netdev->stats.rx_errors = stats->netstats.rx_errors; + netdev->stats.rx_dropped = stats->netstats.rx_dropped; +} + +/* update_len_counters - update pkt's len histogram counters */ +static inline void update_len_counters(struct opa_vnic_grp_stats *grp, + int len) +{ + /* account for 4 byte FCS */ + if (len >= 1515) + grp->s_1519_max++; + else if (len >= 1020) + grp->s_1024_1518++; + else if (len >= 508) + grp->s_512_1023++; + else if (len >= 252) + grp->s_256_511++; + else if (len >= 124) + grp->s_128_255++; + else if (len >= 61) + grp->s_65_127++; + else + grp->s_64++; +} + +/* hfi1_vnic_update_tx_counters - update transmit counters */ +static void hfi1_vnic_update_tx_counters(struct hfi1_vnic_vport_info *vinfo, + u8 q_idx, struct sk_buff *skb, int err) +{ + struct ethhdr *mac_hdr = (struct ethhdr *)skb_mac_header(skb); + struct opa_vnic_stats *stats = &vinfo->stats[q_idx]; + struct opa_vnic_grp_stats *tx_grp = &stats->tx_grp; + u16 vlan_tci; + + stats->netstats.tx_packets++; + stats->netstats.tx_bytes += skb->len + ETH_FCS_LEN; + + update_len_counters(tx_grp, skb->len); + + /* rest of the counts are for good packets only */ + if (unlikely(err)) + return; + + if (is_multicast_ether_addr(mac_hdr->h_dest)) + tx_grp->mcastbcast++; + else + tx_grp->unicast++; + + if (!__vlan_get_tag(skb, &vlan_tci)) + tx_grp->vlan++; + else + tx_grp->untagged++; +} + +/* hfi1_vnic_update_rx_counters - update receive counters */ +static void hfi1_vnic_update_rx_counters(struct hfi1_vnic_vport_info *vinfo, + u8 q_idx, struct sk_buff *skb, int err) +{ + struct ethhdr *mac_hdr = (struct ethhdr *)skb->data; + struct opa_vnic_stats *stats = &vinfo->stats[q_idx]; + struct opa_vnic_grp_stats *rx_grp = &stats->rx_grp; + u16 vlan_tci; + + stats->netstats.rx_packets++; + stats->netstats.rx_bytes += skb->len + ETH_FCS_LEN; + + update_len_counters(rx_grp, skb->len); + + /* rest of the counts are for good packets only */ + if (unlikely(err)) + return; + + if (is_multicast_ether_addr(mac_hdr->h_dest)) + rx_grp->mcastbcast++; + else + rx_grp->unicast++; + + if (!__vlan_get_tag(skb, &vlan_tci)) + rx_grp->vlan++; + else + rx_grp->untagged++; +} + +/* This function is overloaded for opa_vnic specific implementation */ +static void hfi1_vnic_get_stats64(struct net_device *netdev, + struct rtnl_link_stats64 *stats) +{ + struct opa_vnic_stats *vstats = (struct opa_vnic_stats *)stats; + struct hfi1_vnic_vport_info *vinfo = opa_vnic_dev_priv(netdev); + + hfi1_vnic_update_stats(vinfo, vstats); +} + +static u64 create_bypass_pbc(u32 vl, u32 dw_len) +{ + u64 pbc; + + pbc = ((u64)PBC_IHCRC_NONE << PBC_INSERT_HCRC_SHIFT) + | PBC_INSERT_BYPASS_ICRC | PBC_CREDIT_RETURN + | PBC_PACKET_BYPASS + | ((vl & PBC_VL_MASK) << PBC_VL_SHIFT) + | (dw_len & PBC_LENGTH_DWS_MASK) << PBC_LENGTH_DWS_SHIFT; + + return pbc; +} + +/* hfi1_vnic_maybe_stop_tx - stop tx queue if required */ +static void hfi1_vnic_maybe_stop_tx(struct hfi1_vnic_vport_info *vinfo, + u8 q_idx) +{ + netif_stop_subqueue(vinfo->netdev, q_idx); + if (!hfi1_vnic_sdma_write_avail(vinfo, q_idx)) + return; + + netif_start_subqueue(vinfo->netdev, q_idx); +} + +static netdev_tx_t hfi1_netdev_start_xmit(struct sk_buff *skb, + struct net_device *netdev) +{ + struct hfi1_vnic_vport_info *vinfo = opa_vnic_dev_priv(netdev); + u8 pad_len, q_idx = skb->queue_mapping; + struct hfi1_devdata *dd = vinfo->dd; + struct opa_vnic_skb_mdata *mdata; + u32 pkt_len, total_len; + int err = -EINVAL; + u64 pbc; + + v_dbg("xmit: queue %d skb len %d\n", q_idx, skb->len); + if (unlikely(!netif_oper_up(netdev))) { + vinfo->stats[q_idx].tx_drop_state++; + goto tx_finish; + } + + /* take out meta data */ + mdata = (struct opa_vnic_skb_mdata *)skb->data; + skb_pull(skb, sizeof(*mdata)); + if (unlikely(mdata->flags & OPA_VNIC_SKB_MDATA_ENCAP_ERR)) { + vinfo->stats[q_idx].tx_dlid_zero++; + goto tx_finish; + } + + /* add tail padding (for 8 bytes size alignment) and icrc */ + pad_len = -(skb->len + OPA_VNIC_ICRC_TAIL_LEN) & 0x7; + pad_len += OPA_VNIC_ICRC_TAIL_LEN; + + /* + * pkt_len is how much data we have to write, includes header and data. + * total_len is length of the packet in Dwords plus the PBC should not + * include the CRC. + */ + pkt_len = (skb->len + pad_len) >> 2; + total_len = pkt_len + 2; /* PBC + packet */ + + pbc = create_bypass_pbc(mdata->vl, total_len); + + skb_get(skb); + v_dbg("pbc 0x%016llX len %d pad_len %d\n", pbc, skb->len, pad_len); + err = dd->process_vnic_dma_send(dd, q_idx, vinfo, skb, pbc, pad_len); + if (unlikely(err)) { + if (err == -ENOMEM) + vinfo->stats[q_idx].netstats.tx_fifo_errors++; + else if (err != -EBUSY) + vinfo->stats[q_idx].netstats.tx_carrier_errors++; + } + /* remove the header before updating tx counters */ + skb_pull(skb, OPA_VNIC_HDR_LEN); + + if (unlikely(err == -EBUSY)) { + hfi1_vnic_maybe_stop_tx(vinfo, q_idx); + dev_kfree_skb_any(skb); + return NETDEV_TX_BUSY; + } + +tx_finish: + /* update tx counters */ + hfi1_vnic_update_tx_counters(vinfo, q_idx, skb, err); + dev_kfree_skb_any(skb); + return NETDEV_TX_OK; +} + +static u16 hfi1_vnic_select_queue(struct net_device *netdev, + struct sk_buff *skb, + void *accel_priv, + select_queue_fallback_t fallback) +{ + struct hfi1_vnic_vport_info *vinfo = opa_vnic_dev_priv(netdev); + struct opa_vnic_skb_mdata *mdata; + struct sdma_engine *sde; + + mdata = (struct opa_vnic_skb_mdata *)skb->data; + sde = sdma_select_engine_vl(vinfo->dd, mdata->entropy, mdata->vl); + return sde->this_idx; +} + +/* hfi1_vnic_decap_skb - strip OPA header from the skb (ethernet) packet */ +static inline int hfi1_vnic_decap_skb(struct hfi1_vnic_rx_queue *rxq, + struct sk_buff *skb) +{ + struct hfi1_vnic_vport_info *vinfo = rxq->vinfo; + int max_len = vinfo->netdev->mtu + VLAN_ETH_HLEN; + int rc = -EFAULT; + + skb_pull(skb, OPA_VNIC_HDR_LEN); + + /* Validate Packet length */ + if (unlikely(skb->len > max_len)) + vinfo->stats[rxq->idx].rx_oversize++; + else if (unlikely(skb->len < ETH_ZLEN)) + vinfo->stats[rxq->idx].rx_runt++; + else + rc = 0; + return rc; +} + +static inline struct sk_buff *hfi1_vnic_get_skb(struct hfi1_vnic_rx_queue *rxq) +{ + unsigned char *pad_info; + struct sk_buff *skb; + + skb = skb_dequeue(&rxq->skbq); + if (unlikely(!skb)) + return NULL; + + /* remove tail padding and icrc */ + pad_info = skb->data + skb->len - 1; + skb_trim(skb, (skb->len - OPA_VNIC_ICRC_TAIL_LEN - + ((*pad_info) & 0x7))); + + return skb; +} + +/* hfi1_vnic_handle_rx - handle skb receive */ +static void hfi1_vnic_handle_rx(struct hfi1_vnic_rx_queue *rxq, + int *work_done, int work_to_do) +{ + struct hfi1_vnic_vport_info *vinfo = rxq->vinfo; + struct sk_buff *skb; + int rc; + + while (1) { + if (*work_done >= work_to_do) + break; + + skb = hfi1_vnic_get_skb(rxq); + if (unlikely(!skb)) + break; + + rc = hfi1_vnic_decap_skb(rxq, skb); + /* update rx counters */ + hfi1_vnic_update_rx_counters(vinfo, rxq->idx, skb, rc); + if (unlikely(rc)) { + dev_kfree_skb_any(skb); + continue; + } + + skb_checksum_none_assert(skb); + skb->protocol = eth_type_trans(skb, rxq->netdev); + + napi_gro_receive(&rxq->napi, skb); + (*work_done)++; + } +} + +/* hfi1_vnic_napi - napi receive polling callback function */ +static int hfi1_vnic_napi(struct napi_struct *napi, int budget) +{ + struct hfi1_vnic_rx_queue *rxq = container_of(napi, + struct hfi1_vnic_rx_queue, napi); + struct hfi1_vnic_vport_info *vinfo = rxq->vinfo; + int work_done = 0; + + v_dbg("napi %d budget %d\n", rxq->idx, budget); + hfi1_vnic_handle_rx(rxq, &work_done, budget); + + v_dbg("napi %d work_done %d\n", rxq->idx, work_done); + if (work_done < budget) + napi_complete(napi); + + return work_done; +} + +void hfi1_vnic_bypass_rcv(struct hfi1_packet *packet) +{ + struct hfi1_devdata *dd = packet->rcd->dd; + struct hfi1_vnic_vport_info *vinfo = NULL; + struct hfi1_vnic_rx_queue *rxq; + struct sk_buff *skb; + int l4_type, vesw_id = -1; + u8 q_idx; + + l4_type = HFI1_GET_L4_TYPE(packet->ebuf); + if (likely(l4_type == OPA_VNIC_L4_ETHR)) { + vesw_id = HFI1_VNIC_GET_VESWID(packet->ebuf); + vinfo = idr_find(&dd->vnic.vesw_idr, vesw_id); + + /* + * In case of invalid vesw id, count the error on + * the first available vport. + */ + if (unlikely(!vinfo)) { + struct hfi1_vnic_vport_info *vinfo_tmp; + int id_tmp = 0; + + vinfo_tmp = idr_get_next(&dd->vnic.vesw_idr, &id_tmp); + if (vinfo_tmp) { + spin_lock(&vport_cntr_lock); + vinfo_tmp->stats[0].netstats.rx_nohandler++; + spin_unlock(&vport_cntr_lock); + } + } + } + + if (unlikely(!vinfo)) { + dd_dev_warn(dd, "vnic rcv err: l4 %d vesw id %d ctx %d\n", + l4_type, vesw_id, packet->rcd->ctxt); + return; + } + + q_idx = packet->rcd->vnic_q_idx; + rxq = &vinfo->rxq[q_idx]; + if (unlikely(!netif_oper_up(vinfo->netdev))) { + vinfo->stats[q_idx].rx_drop_state++; + skb_queue_purge(&rxq->skbq); + return; + } + + if (unlikely(skb_queue_len(&rxq->skbq) > HFI1_VNIC_RCV_Q_SIZE)) { + vinfo->stats[q_idx].netstats.rx_fifo_errors++; + return; + } + + skb = netdev_alloc_skb(vinfo->netdev, packet->tlen); + if (unlikely(!skb)) { + vinfo->stats[q_idx].netstats.rx_fifo_errors++; + return; + } + + memcpy(skb->data, packet->ebuf, packet->tlen); + skb_put(skb, packet->tlen); + skb_queue_tail(&rxq->skbq, skb); + + if (napi_schedule_prep(&rxq->napi)) { + v_dbg("napi %d scheduling\n", q_idx); + __napi_schedule(&rxq->napi); + } +} + +static int hfi1_vnic_up(struct hfi1_vnic_vport_info *vinfo) +{ + struct hfi1_devdata *dd = vinfo->dd; + struct net_device *netdev = vinfo->netdev; + int i, rc; + + /* ensure virtual eth switch id is valid */ + if (!vinfo->vesw_id) + return -EINVAL; + + rc = idr_alloc(&dd->vnic.vesw_idr, vinfo, vinfo->vesw_id, + vinfo->vesw_id + 1, GFP_NOWAIT); + if (rc < 0) + return rc; + + for (i = 0; i < vinfo->num_rx_q; i++) { + struct hfi1_vnic_rx_queue *rxq = &vinfo->rxq[i]; + + skb_queue_head_init(&rxq->skbq); + napi_enable(&rxq->napi); + } + + netif_carrier_on(netdev); + netif_tx_start_all_queues(netdev); + set_bit(HFI1_VNIC_UP, &vinfo->flags); + + return 0; +} + +static void hfi1_vnic_down(struct hfi1_vnic_vport_info *vinfo) +{ + struct hfi1_devdata *dd = vinfo->dd; + u8 i; + + clear_bit(HFI1_VNIC_UP, &vinfo->flags); + netif_carrier_off(vinfo->netdev); + netif_tx_disable(vinfo->netdev); + idr_remove(&dd->vnic.vesw_idr, vinfo->vesw_id); + + /* ensure irqs see the change */ + hfi1_vnic_synchronize_irq(dd); + + /* remove unread skbs */ + for (i = 0; i < vinfo->num_rx_q; i++) { + struct hfi1_vnic_rx_queue *rxq = &vinfo->rxq[i]; + + napi_disable(&rxq->napi); + skb_queue_purge(&rxq->skbq); + } +} + +static int hfi1_netdev_open(struct net_device *netdev) +{ + struct hfi1_vnic_vport_info *vinfo = opa_vnic_dev_priv(netdev); + int rc; + + mutex_lock(&vinfo->lock); + rc = hfi1_vnic_up(vinfo); + mutex_unlock(&vinfo->lock); + return rc; +} + +static int hfi1_netdev_close(struct net_device *netdev) +{ + struct hfi1_vnic_vport_info *vinfo = opa_vnic_dev_priv(netdev); + + mutex_lock(&vinfo->lock); + if (test_bit(HFI1_VNIC_UP, &vinfo->flags)) + hfi1_vnic_down(vinfo); + mutex_unlock(&vinfo->lock); + return 0; +} + +static int hfi1_vnic_allot_ctxt(struct hfi1_devdata *dd, + struct hfi1_ctxtdata **vnic_ctxt) +{ + int rc; + + rc = allocate_vnic_ctxt(dd, vnic_ctxt); + if (rc) { + dd_dev_err(dd, "vnic ctxt alloc failed %d\n", rc); + return rc; + } + + rc = setup_vnic_ctxt(dd, *vnic_ctxt); + if (rc) { + dd_dev_err(dd, "vnic ctxt setup failed %d\n", rc); + deallocate_vnic_ctxt(dd, *vnic_ctxt); + *vnic_ctxt = NULL; + } + + return rc; +} + +static int hfi1_vnic_init(struct hfi1_vnic_vport_info *vinfo) +{ + struct hfi1_devdata *dd = vinfo->dd; + int i, rc = 0; + + mutex_lock(&hfi1_mutex); + if (!dd->vnic.num_vports) { + rc = hfi1_vnic_txreq_init(dd); + if (rc) + goto txreq_fail; + + dd->vnic.msix_idx = dd->first_dyn_msix_idx; + } + + for (i = dd->vnic.num_ctxt; i < vinfo->num_rx_q; i++) { + rc = hfi1_vnic_allot_ctxt(dd, &dd->vnic.ctxt[i]); + if (rc) + break; + dd->vnic.ctxt[i]->vnic_q_idx = i; + } + + if (i < vinfo->num_rx_q) { + /* + * If required amount of contexts is not + * allocated successfully then remaining contexts + * are released. + */ + while (i-- > dd->vnic.num_ctxt) { + deallocate_vnic_ctxt(dd, dd->vnic.ctxt[i]); + dd->vnic.ctxt[i] = NULL; + } + goto alloc_fail; + } + + if (dd->vnic.num_ctxt != i) { + dd->vnic.num_ctxt = i; + hfi1_init_vnic_rsm(dd); + } + + dd->vnic.num_vports++; + hfi1_vnic_sdma_init(vinfo); +alloc_fail: + if (!dd->vnic.num_vports) + hfi1_vnic_txreq_deinit(dd); +txreq_fail: + mutex_unlock(&hfi1_mutex); + return rc; +} + +static void hfi1_vnic_deinit(struct hfi1_vnic_vport_info *vinfo) +{ + struct hfi1_devdata *dd = vinfo->dd; + int i; + + mutex_lock(&hfi1_mutex); + if (--dd->vnic.num_vports == 0) { + for (i = 0; i < dd->vnic.num_ctxt; i++) { + deallocate_vnic_ctxt(dd, dd->vnic.ctxt[i]); + dd->vnic.ctxt[i] = NULL; + } + hfi1_deinit_vnic_rsm(dd); + dd->vnic.num_ctxt = 0; + hfi1_vnic_txreq_deinit(dd); + } + mutex_unlock(&hfi1_mutex); +} + +static void hfi1_vnic_set_vesw_id(struct net_device *netdev, int id) +{ + struct hfi1_vnic_vport_info *vinfo = opa_vnic_dev_priv(netdev); + bool reopen = false; + + /* + * If vesw_id is being changed, and if the vnic port is up, + * reset the vnic port to ensure new vesw_id gets picked up + */ + if (id != vinfo->vesw_id) { + mutex_lock(&vinfo->lock); + if (test_bit(HFI1_VNIC_UP, &vinfo->flags)) { + hfi1_vnic_down(vinfo); + reopen = true; + } + + vinfo->vesw_id = id; + if (reopen) + hfi1_vnic_up(vinfo); + + mutex_unlock(&vinfo->lock); + } +} + +/* netdev ops */ +static const struct net_device_ops hfi1_netdev_ops = { + .ndo_open = hfi1_netdev_open, + .ndo_stop = hfi1_netdev_close, + .ndo_start_xmit = hfi1_netdev_start_xmit, + .ndo_select_queue = hfi1_vnic_select_queue, + .ndo_get_stats64 = hfi1_vnic_get_stats64, +}; + +struct net_device *hfi1_vnic_alloc_rn(struct ib_device *device, + u8 port_num, + enum rdma_netdev_t type, + const char *name, + unsigned char name_assign_type, + void (*setup)(struct net_device *)) +{ + struct hfi1_devdata *dd = dd_from_ibdev(device); + struct hfi1_vnic_vport_info *vinfo; + struct net_device *netdev; + struct rdma_netdev *rn; + int i, size, rc; + + if (!port_num || (port_num > dd->num_pports)) + return ERR_PTR(-EINVAL); + + if (type != RDMA_NETDEV_OPA_VNIC) + return ERR_PTR(-EOPNOTSUPP); + + size = sizeof(struct opa_vnic_rdma_netdev) + sizeof(*vinfo); + netdev = alloc_netdev_mqs(size, name, name_assign_type, setup, + dd->chip_sdma_engines, HFI1_NUM_VNIC_CTXT); + if (!netdev) + return ERR_PTR(-ENOMEM); + + rn = netdev_priv(netdev); + vinfo = opa_vnic_dev_priv(netdev); + vinfo->dd = dd; + vinfo->num_tx_q = dd->chip_sdma_engines; + vinfo->num_rx_q = HFI1_NUM_VNIC_CTXT; + vinfo->netdev = netdev; + rn->set_id = hfi1_vnic_set_vesw_id; + + netdev->features = NETIF_F_HIGHDMA | NETIF_F_SG; + netdev->hw_features = netdev->features; + netdev->vlan_features = netdev->features; + netdev->watchdog_timeo = msecs_to_jiffies(HFI_TX_TIMEOUT_MS); + netdev->netdev_ops = &hfi1_netdev_ops; + mutex_init(&vinfo->lock); + + for (i = 0; i < vinfo->num_rx_q; i++) { + struct hfi1_vnic_rx_queue *rxq = &vinfo->rxq[i]; + + rxq->idx = i; + rxq->vinfo = vinfo; + rxq->netdev = netdev; + netif_napi_add(netdev, &rxq->napi, hfi1_vnic_napi, 64); + } + + rc = hfi1_vnic_init(vinfo); + if (rc) + goto init_fail; + + return netdev; +init_fail: + mutex_destroy(&vinfo->lock); + 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/hfi1/vnic_sdma.c b/drivers/infiniband/hw/hfi1/vnic_sdma.c new file mode 100644 index 000000000000..51a817d3aa14 --- /dev/null +++ b/drivers/infiniband/hw/hfi1/vnic_sdma.c @@ -0,0 +1,323 @@ +/* + * Copyright(c) 2017 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ + +/* + * This file contains HFI1 support for VNIC SDMA functionality + */ + +#include "sdma.h" +#include "vnic.h" + +#define HFI1_VNIC_SDMA_Q_ACTIVE BIT(0) +#define HFI1_VNIC_SDMA_Q_DEFERRED BIT(1) + +#define HFI1_VNIC_TXREQ_NAME_LEN 32 +#define HFI1_VNIC_SDMA_DESC_WTRMRK 64 +#define HFI1_VNIC_SDMA_RETRY_COUNT 1 + +/* + * struct vnic_txreq - VNIC transmit descriptor + * @txreq: sdma transmit request + * @sdma: vnic sdma pointer + * @skb: skb to send + * @pad: pad buffer + * @plen: pad length + * @pbc_val: pbc value + * @retry_count: tx retry count + */ +struct vnic_txreq { + struct sdma_txreq txreq; + struct hfi1_vnic_sdma *sdma; + + struct sk_buff *skb; + unsigned char pad[HFI1_VNIC_MAX_PAD]; + u16 plen; + __le64 pbc_val; + + u32 retry_count; +}; + +static void vnic_sdma_complete(struct sdma_txreq *txreq, + int status) +{ + struct vnic_txreq *tx = container_of(txreq, struct vnic_txreq, txreq); + struct hfi1_vnic_sdma *vnic_sdma = tx->sdma; + + sdma_txclean(vnic_sdma->dd, txreq); + dev_kfree_skb_any(tx->skb); + kmem_cache_free(vnic_sdma->dd->vnic.txreq_cache, tx); +} + +static noinline int build_vnic_ulp_payload(struct sdma_engine *sde, + struct vnic_txreq *tx) +{ + int i, ret = 0; + + ret = sdma_txadd_kvaddr( + sde->dd, + &tx->txreq, + tx->skb->data, + skb_headlen(tx->skb)); + if (unlikely(ret)) + goto bail_txadd; + + for (i = 0; i < skb_shinfo(tx->skb)->nr_frags; i++) { + struct skb_frag_struct *frag = &skb_shinfo(tx->skb)->frags[i]; + + /* combine physically continuous fragments later? */ + ret = sdma_txadd_page(sde->dd, + &tx->txreq, + skb_frag_page(frag), + frag->page_offset, + skb_frag_size(frag)); + if (unlikely(ret)) + goto bail_txadd; + } + + if (tx->plen) + ret = sdma_txadd_kvaddr(sde->dd, &tx->txreq, + tx->pad + HFI1_VNIC_MAX_PAD - tx->plen, + tx->plen); + +bail_txadd: + return ret; +} + +static int build_vnic_tx_desc(struct sdma_engine *sde, + struct vnic_txreq *tx, + u64 pbc) +{ + int ret = 0; + u16 hdrbytes = 2 << 2; /* PBC */ + + ret = sdma_txinit_ahg( + &tx->txreq, + 0, + hdrbytes + tx->skb->len + tx->plen, + 0, + 0, + NULL, + 0, + vnic_sdma_complete); + if (unlikely(ret)) + goto bail_txadd; + + /* add pbc */ + tx->pbc_val = cpu_to_le64(pbc); + ret = sdma_txadd_kvaddr( + sde->dd, + &tx->txreq, + &tx->pbc_val, + hdrbytes); + if (unlikely(ret)) + goto bail_txadd; + + /* add the ulp payload */ + ret = build_vnic_ulp_payload(sde, tx); +bail_txadd: + return ret; +} + +/* setup the last plen bypes of pad */ +static inline void hfi1_vnic_update_pad(unsigned char *pad, u8 plen) +{ + pad[HFI1_VNIC_MAX_PAD - 1] = plen - OPA_VNIC_ICRC_TAIL_LEN; +} + +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) +{ + struct hfi1_vnic_sdma *vnic_sdma = &vinfo->sdma[q_idx]; + struct sdma_engine *sde = vnic_sdma->sde; + struct vnic_txreq *tx; + int ret = -ECOMM; + + if (unlikely(READ_ONCE(vnic_sdma->state) != HFI1_VNIC_SDMA_Q_ACTIVE)) + goto tx_err; + + if (unlikely(!sde || !sdma_running(sde))) + goto tx_err; + + tx = kmem_cache_alloc(dd->vnic.txreq_cache, GFP_ATOMIC); + if (unlikely(!tx)) { + ret = -ENOMEM; + goto tx_err; + } + + tx->sdma = vnic_sdma; + tx->skb = skb; + hfi1_vnic_update_pad(tx->pad, plen); + tx->plen = plen; + ret = build_vnic_tx_desc(sde, tx, pbc); + if (unlikely(ret)) + goto free_desc; + tx->retry_count = 0; + + ret = sdma_send_txreq(sde, &vnic_sdma->wait, &tx->txreq); + /* When -ECOMM, sdma callback will be called with ABORT status */ + if (unlikely(ret && unlikely(ret != -ECOMM))) + goto free_desc; + + return ret; + +free_desc: + sdma_txclean(dd, &tx->txreq); + kmem_cache_free(dd->vnic.txreq_cache, tx); +tx_err: + if (ret != -EBUSY) + dev_kfree_skb_any(skb); + return ret; +} + +/* + * hfi1_vnic_sdma_sleep - vnic sdma sleep function + * + * This function gets called from sdma_send_txreq() when there are not enough + * sdma descriptors available to send the packet. It adds Tx queue's wait + * structure to sdma engine's dmawait list to be woken up when descriptors + * become available. + */ +static int hfi1_vnic_sdma_sleep(struct sdma_engine *sde, + struct iowait *wait, + struct sdma_txreq *txreq, + unsigned int seq) +{ + struct hfi1_vnic_sdma *vnic_sdma = + container_of(wait, struct hfi1_vnic_sdma, wait); + struct hfi1_ibdev *dev = &vnic_sdma->dd->verbs_dev; + struct vnic_txreq *tx = container_of(txreq, struct vnic_txreq, txreq); + + if (sdma_progress(sde, seq, txreq)) + if (tx->retry_count++ < HFI1_VNIC_SDMA_RETRY_COUNT) + return -EAGAIN; + + vnic_sdma->state = HFI1_VNIC_SDMA_Q_DEFERRED; + write_seqlock(&dev->iowait_lock); + if (list_empty(&vnic_sdma->wait.list)) + list_add_tail(&vnic_sdma->wait.list, &sde->dmawait); + write_sequnlock(&dev->iowait_lock); + return -EBUSY; +} + +/* + * hfi1_vnic_sdma_wakeup - vnic sdma wakeup function + * + * This function gets called when SDMA descriptors becomes available and Tx + * queue's wait structure was previously added to sdma engine's dmawait list. + * It notifies the upper driver about Tx queue wakeup. + */ +static void hfi1_vnic_sdma_wakeup(struct iowait *wait, int reason) +{ + struct hfi1_vnic_sdma *vnic_sdma = + container_of(wait, struct hfi1_vnic_sdma, wait); + struct hfi1_vnic_vport_info *vinfo = vnic_sdma->vinfo; + + vnic_sdma->state = HFI1_VNIC_SDMA_Q_ACTIVE; + if (__netif_subqueue_stopped(vinfo->netdev, vnic_sdma->q_idx)) + netif_wake_subqueue(vinfo->netdev, vnic_sdma->q_idx); +}; + +inline bool hfi1_vnic_sdma_write_avail(struct hfi1_vnic_vport_info *vinfo, + u8 q_idx) +{ + struct hfi1_vnic_sdma *vnic_sdma = &vinfo->sdma[q_idx]; + + return (READ_ONCE(vnic_sdma->state) == HFI1_VNIC_SDMA_Q_ACTIVE); +} + +void hfi1_vnic_sdma_init(struct hfi1_vnic_vport_info *vinfo) +{ + int i; + + for (i = 0; i < vinfo->num_tx_q; i++) { + struct hfi1_vnic_sdma *vnic_sdma = &vinfo->sdma[i]; + + iowait_init(&vnic_sdma->wait, 0, NULL, hfi1_vnic_sdma_sleep, + hfi1_vnic_sdma_wakeup, NULL); + vnic_sdma->sde = &vinfo->dd->per_sdma[i]; + vnic_sdma->dd = vinfo->dd; + vnic_sdma->vinfo = vinfo; + vnic_sdma->q_idx = i; + vnic_sdma->state = HFI1_VNIC_SDMA_Q_ACTIVE; + + /* Add a free descriptor watermark for wakeups */ + if (vnic_sdma->sde->descq_cnt > HFI1_VNIC_SDMA_DESC_WTRMRK) { + INIT_LIST_HEAD(&vnic_sdma->stx.list); + vnic_sdma->stx.num_desc = HFI1_VNIC_SDMA_DESC_WTRMRK; + list_add_tail(&vnic_sdma->stx.list, + &vnic_sdma->wait.tx_head); + } + } +} + +static void hfi1_vnic_txreq_kmem_cache_ctor(void *obj) +{ + struct vnic_txreq *tx = (struct vnic_txreq *)obj; + + memset(tx, 0, sizeof(*tx)); +} + +int hfi1_vnic_txreq_init(struct hfi1_devdata *dd) +{ + char buf[HFI1_VNIC_TXREQ_NAME_LEN]; + + snprintf(buf, sizeof(buf), "hfi1_%u_vnic_txreq_cache", dd->unit); + dd->vnic.txreq_cache = kmem_cache_create(buf, + sizeof(struct vnic_txreq), + 0, SLAB_HWCACHE_ALIGN, + hfi1_vnic_txreq_kmem_cache_ctor); + if (!dd->vnic.txreq_cache) + return -ENOMEM; + return 0; +} + +void hfi1_vnic_txreq_deinit(struct hfi1_devdata *dd) +{ + kmem_cache_destroy(dd->vnic.txreq_cache); + dd->vnic.txreq_cache = NULL; +} diff --git a/drivers/infiniband/hw/hns/hns_roce_cmd.c b/drivers/infiniband/hw/hns/hns_roce_cmd.c index 8c1f7a6f84d2..b94dcd823ad1 100644 --- a/drivers/infiniband/hw/hns/hns_roce_cmd.c +++ b/drivers/infiniband/hw/hns/hns_roce_cmd.c @@ -299,9 +299,9 @@ int hns_roce_cmd_use_events(struct hns_roce_dev *hr_dev) struct hns_roce_cmdq *hr_cmd = &hr_dev->cmd; int i; - hr_cmd->context = kmalloc(hr_cmd->max_cmds * - sizeof(struct hns_roce_cmd_context), - GFP_KERNEL); + hr_cmd->context = kmalloc_array(hr_cmd->max_cmds, + sizeof(*hr_cmd->context), + GFP_KERNEL); if (!hr_cmd->context) return -ENOMEM; diff --git a/drivers/infiniband/hw/hns/hns_roce_cq.c b/drivers/infiniband/hw/hns/hns_roce_cq.c index 589496c8fb9e..b89fd711019e 100644 --- a/drivers/infiniband/hw/hns/hns_roce_cq.c +++ b/drivers/infiniband/hw/hns/hns_roce_cq.c @@ -219,8 +219,7 @@ static int hns_roce_ib_get_cq_umem(struct hns_roce_dev *hr_dev, return PTR_ERR(*umem); ret = hns_roce_mtt_init(hr_dev, ib_umem_page_count(*umem), - ilog2((unsigned int)(*umem)->page_size), - &buf->hr_mtt); + (*umem)->page_shift, &buf->hr_mtt); if (ret) goto err_buf; diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c index b8111b0c8877..e637beb209f7 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c @@ -33,6 +33,7 @@ #include <linux/platform_device.h> #include <linux/acpi.h> #include <linux/etherdevice.h> +#include <linux/of.h> #include <rdma/ib_umem.h> #include "hns_roce_common.h" #include "hns_roce_device.h" @@ -1851,6 +1852,7 @@ void hns_roce_v1_cq_set_ci(struct hns_roce_cq *hr_cq, u32 cons_index) u32 doorbell[2]; doorbell[0] = cons_index & ((hr_cq->cq_depth << 1) - 1); + doorbell[1] = 0; roce_set_bit(doorbell[1], ROCEE_DB_OTHERS_H_ROCEE_DB_OTH_HW_SYNS_S, 1); roce_set_field(doorbell[1], ROCEE_DB_OTHERS_H_ROCEE_DB_OTH_CMD_M, ROCEE_DB_OTHERS_H_ROCEE_DB_OTH_CMD_S, 3); diff --git a/drivers/infiniband/hw/hns/hns_roce_mr.c b/drivers/infiniband/hw/hns/hns_roce_mr.c index 4139abee3b54..dc5c97c8f070 100644 --- a/drivers/infiniband/hw/hns/hns_roce_mr.c +++ b/drivers/infiniband/hw/hns/hns_roce_mr.c @@ -127,11 +127,12 @@ static int hns_roce_buddy_init(struct hns_roce_buddy *buddy, int max_order) buddy->max_order = max_order; spin_lock_init(&buddy->lock); - - buddy->bits = kzalloc((buddy->max_order + 1) * sizeof(long *), - GFP_KERNEL); - buddy->num_free = kzalloc((buddy->max_order + 1) * sizeof(int *), - GFP_KERNEL); + buddy->bits = kcalloc(buddy->max_order + 1, + sizeof(*buddy->bits), + GFP_KERNEL); + buddy->num_free = kcalloc(buddy->max_order + 1, + sizeof(*buddy->num_free), + GFP_KERNEL); if (!buddy->bits || !buddy->num_free) goto err_out; @@ -503,7 +504,8 @@ int hns_roce_ib_umem_write_mtt(struct hns_roce_dev *hr_dev, for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) { len = sg_dma_len(sg) >> mtt->page_shift; for (k = 0; k < len; ++k) { - pages[i++] = sg_dma_address(sg) + umem->page_size * k; + pages[i++] = sg_dma_address(sg) + + (k << umem->page_shift); if (i == PAGE_SIZE / sizeof(u64)) { ret = hns_roce_write_mtt(hr_dev, mtt, n, i, pages); @@ -563,9 +565,9 @@ struct ib_mr *hns_roce_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, } n = ib_umem_page_count(mr->umem); - if (mr->umem->page_size != HNS_ROCE_HEM_PAGE_SIZE) { - dev_err(dev, "Just support 4K page size but is 0x%x now!\n", - mr->umem->page_size); + if (mr->umem->page_shift != HNS_ROCE_HEM_PAGE_SHIFT) { + dev_err(dev, "Just support 4K page size but is 0x%lx now!\n", + BIT(mr->umem->page_shift)); ret = -EINVAL; goto err_umem; } diff --git a/drivers/infiniband/hw/hns/hns_roce_qp.c b/drivers/infiniband/hw/hns/hns_roce_qp.c index 3f44f2f91f03..054c52699090 100644 --- a/drivers/infiniband/hw/hns/hns_roce_qp.c +++ b/drivers/infiniband/hw/hns/hns_roce_qp.c @@ -437,8 +437,7 @@ static int hns_roce_create_qp_common(struct hns_roce_dev *hr_dev, } ret = hns_roce_mtt_init(hr_dev, ib_umem_page_count(hr_qp->umem), - ilog2((unsigned int)hr_qp->umem->page_size), - &hr_qp->mtt); + hr_qp->umem->page_shift, &hr_qp->mtt); if (ret) { dev_err(dev, "hns_roce_mtt_init error for create qp\n"); goto err_buf; diff --git a/drivers/infiniband/hw/i40iw/i40iw_cm.c b/drivers/infiniband/hw/i40iw/i40iw_cm.c index 95a0586a4da8..f3bc01bce483 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_cm.c +++ b/drivers/infiniband/hw/i40iw/i40iw_cm.c @@ -3184,9 +3184,8 @@ void i40iw_setup_cm_core(struct i40iw_device *iwdev) INIT_LIST_HEAD(&cm_core->connected_nodes); INIT_LIST_HEAD(&cm_core->listen_nodes); - init_timer(&cm_core->tcp_timer); - cm_core->tcp_timer.function = i40iw_cm_timer_tick; - cm_core->tcp_timer.data = (unsigned long)cm_core; + setup_timer(&cm_core->tcp_timer, i40iw_cm_timer_tick, + (unsigned long)cm_core); spin_lock_init(&cm_core->ht_lock); spin_lock_init(&cm_core->listen_list_lock); diff --git a/drivers/infiniband/hw/i40iw/i40iw_utils.c b/drivers/infiniband/hw/i40iw/i40iw_utils.c index 70c3e9e79508..409a3781e735 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_utils.c +++ b/drivers/infiniband/hw/i40iw/i40iw_utils.c @@ -844,10 +844,9 @@ void i40iw_terminate_start_timer(struct i40iw_sc_qp *qp) iwqp = (struct i40iw_qp *)qp->back_qp; i40iw_add_ref(&iwqp->ibqp); - init_timer(&iwqp->terminate_timer); - iwqp->terminate_timer.function = i40iw_terminate_timeout; + setup_timer(&iwqp->terminate_timer, i40iw_terminate_timeout, + (unsigned long)iwqp); iwqp->terminate_timer.expires = jiffies + HZ; - iwqp->terminate_timer.data = (unsigned long)iwqp; add_timer(&iwqp->terminate_timer); } @@ -1436,9 +1435,8 @@ void i40iw_hw_stats_start_timer(struct i40iw_sc_vsi *vsi) { struct i40iw_vsi_pestat *devstat = vsi->pestat; - init_timer(&devstat->stats_timer); - devstat->stats_timer.function = i40iw_hw_stats_timeout; - devstat->stats_timer.data = (unsigned long)vsi; + setup_timer(&devstat->stats_timer, i40iw_hw_stats_timeout, + (unsigned long)vsi); mod_timer(&devstat->stats_timer, jiffies + msecs_to_jiffies(STATS_TIMER_DELAY)); } diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.c b/drivers/infiniband/hw/i40iw/i40iw_verbs.c index 9b2849979756..378c75759be4 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_verbs.c +++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.c @@ -1345,7 +1345,7 @@ static void i40iw_copy_user_pgaddrs(struct i40iw_mr *iwmr, { struct ib_umem *region = iwmr->region; struct i40iw_pbl *iwpbl = &iwmr->iwpbl; - int chunk_pages, entry, pg_shift, i; + int chunk_pages, entry, i; struct i40iw_pble_alloc *palloc = &iwpbl->pble_alloc; struct i40iw_pble_info *pinfo; struct scatterlist *sg; @@ -1354,14 +1354,14 @@ static void i40iw_copy_user_pgaddrs(struct i40iw_mr *iwmr, pinfo = (level == I40IW_LEVEL_1) ? NULL : palloc->level2.leaf; - pg_shift = ffs(region->page_size) - 1; for_each_sg(region->sg_head.sgl, sg, region->nmap, entry) { - chunk_pages = sg_dma_len(sg) >> pg_shift; + chunk_pages = sg_dma_len(sg) >> region->page_shift; if ((iwmr->type == IW_MEMREG_TYPE_QP) && !iwpbl->qp_mr.sq_page) iwpbl->qp_mr.sq_page = sg_page(sg); for (i = 0; i < chunk_pages; i++) { - pg_addr = sg_dma_address(sg) + region->page_size * i; + pg_addr = sg_dma_address(sg) + + (i << region->page_shift); if ((entry + i) == 0) *pbl = cpu_to_le64(pg_addr & iwmr->page_msk); @@ -1847,7 +1847,7 @@ static struct ib_mr *i40iw_reg_user_mr(struct ib_pd *pd, iwmr->ibmr.device = pd->device; ucontext = to_ucontext(pd->uobject->context); - iwmr->page_size = region->page_size; + iwmr->page_size = PAGE_SIZE; iwmr->page_msk = PAGE_MASK; if (region->hugetlb && (req.reg_type == IW_MEMREG_TYPE_MEM)) diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c index 6a0fec357dae..4f5a143fc0a7 100644 --- a/drivers/infiniband/hw/mlx4/cq.c +++ b/drivers/infiniband/hw/mlx4/cq.c @@ -147,7 +147,7 @@ static int mlx4_ib_get_cq_umem(struct mlx4_ib_dev *dev, struct ib_ucontext *cont return PTR_ERR(*umem); err = mlx4_mtt_init(dev->dev, ib_umem_page_count(*umem), - ilog2((*umem)->page_size), &buf->mtt); + (*umem)->page_shift, &buf->mtt); if (err) goto err_buf; diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index fba94df28cf1..521d0def2d9e 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -1173,7 +1173,7 @@ static void mlx4_ib_disassociate_ucontext(struct ib_ucontext *ibcontext) /* need to protect from a race on closing the vma as part of * mlx4_ib_vma_close(). */ - down_read(&owning_mm->mmap_sem); + down_write(&owning_mm->mmap_sem); for (i = 0; i < HW_BAR_COUNT; i++) { vma = context->hw_bar_info[i].vma; if (!vma) @@ -1187,11 +1187,13 @@ static void mlx4_ib_disassociate_ucontext(struct ib_ucontext *ibcontext) BUG_ON(1); } + context->hw_bar_info[i].vma->vm_flags &= + ~(VM_SHARED | VM_MAYSHARE); /* context going to be destroyed, should not access ops any more */ context->hw_bar_info[i].vma->vm_ops = NULL; } - up_read(&owning_mm->mmap_sem); + up_write(&owning_mm->mmap_sem); mmput(owning_mm); put_task_struct(owning_process); } @@ -2867,23 +2869,19 @@ static void *mlx4_ib_add(struct mlx4_dev *dev) if (mlx4_ib_init_sriov(ibdev)) goto err_mad; - if (dev->caps.flags & MLX4_DEV_CAP_FLAG_IBOE || - dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ROCE_V1_V2) { - if (!iboe->nb.notifier_call) { - iboe->nb.notifier_call = mlx4_ib_netdev_event; - err = register_netdevice_notifier(&iboe->nb); - if (err) { - iboe->nb.notifier_call = NULL; - goto err_notif; - } - } - if (dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ROCE_V1_V2) { - err = mlx4_config_roce_v2_port(dev, ROCE_V2_UDP_DPORT); - if (err) { - goto err_notif; - } + if (!iboe->nb.notifier_call) { + iboe->nb.notifier_call = mlx4_ib_netdev_event; + err = register_netdevice_notifier(&iboe->nb); + if (err) { + iboe->nb.notifier_call = NULL; + goto err_notif; } } + if (dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ROCE_V1_V2) { + err = mlx4_config_roce_v2_port(dev, ROCE_V2_UDP_DPORT); + if (err) + goto err_notif; + } for (j = 0; j < ARRAY_SIZE(mlx4_class_attributes); ++j) { if (device_create_file(&ibdev->ib_dev.dev, @@ -2941,6 +2939,7 @@ err_counter: mlx4_ib_delete_counters_table(ibdev, &ibdev->counters_table[i]); err_map: + mlx4_ib_free_eqs(dev, ibdev); iounmap(ibdev->uar_map); err_uar: diff --git a/drivers/infiniband/hw/mlx4/mcg.c b/drivers/infiniband/hw/mlx4/mcg.c index e010fe459e67..8772d88d324d 100644 --- a/drivers/infiniband/hw/mlx4/mcg.c +++ b/drivers/infiniband/hw/mlx4/mcg.c @@ -1102,7 +1102,8 @@ static void _mlx4_ib_mcg_port_cleanup(struct mlx4_ib_demux_ctx *ctx, int destroy while ((p = rb_first(&ctx->mcg_table)) != NULL) { group = rb_entry(p, struct mcast_group, node); if (atomic_read(&group->refcount)) - mcg_warn_group(group, "group refcount %d!!! (pointer %p)\n", atomic_read(&group->refcount), group); + mcg_debug_group(group, "group refcount %d!!! (pointer %p)\n", + atomic_read(&group->refcount), group); force_clean_group(group); } diff --git a/drivers/infiniband/hw/mlx4/mr.c b/drivers/infiniband/hw/mlx4/mr.c index 433bcdbdd680..e6f77f63da75 100644 --- a/drivers/infiniband/hw/mlx4/mr.c +++ b/drivers/infiniband/hw/mlx4/mr.c @@ -107,7 +107,7 @@ int mlx4_ib_umem_write_mtt(struct mlx4_ib_dev *dev, struct mlx4_mtt *mtt, len = sg_dma_len(sg) >> mtt->page_shift; for (k = 0; k < len; ++k) { pages[i++] = sg_dma_address(sg) + - umem->page_size * k; + (k << umem->page_shift); /* * Be friendly to mlx4_write_mtt() and * pass it chunks of appropriate size. @@ -155,7 +155,7 @@ struct ib_mr *mlx4_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, } n = ib_umem_page_count(mr->umem); - shift = ilog2(mr->umem->page_size); + shift = mr->umem->page_shift; err = mlx4_mr_alloc(dev->dev, to_mpd(pd)->pdn, virt_addr, length, convert_access(access_flags), n, shift, &mr->mmr); @@ -239,7 +239,7 @@ int mlx4_ib_rereg_user_mr(struct ib_mr *mr, int flags, goto release_mpt_entry; } n = ib_umem_page_count(mmr->umem); - shift = ilog2(mmr->umem->page_size); + shift = mmr->umem->page_shift; err = mlx4_mr_rereg_mem_write(dev->dev, &mmr->mmr, virt_addr, length, n, shift, diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c index c34eebc7db65..8f382318f888 100644 --- a/drivers/infiniband/hw/mlx4/qp.c +++ b/drivers/infiniband/hw/mlx4/qp.c @@ -745,7 +745,7 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, } err = mlx4_mtt_init(dev->dev, ib_umem_page_count(qp->umem), - ilog2(qp->umem->page_size), &qp->mtt); + qp->umem->page_shift, &qp->mtt); if (err) goto err_buf; diff --git a/drivers/infiniband/hw/mlx4/srq.c b/drivers/infiniband/hw/mlx4/srq.c index 7dd3f267f06b..e32dd58937a8 100644 --- a/drivers/infiniband/hw/mlx4/srq.c +++ b/drivers/infiniband/hw/mlx4/srq.c @@ -122,7 +122,7 @@ struct ib_srq *mlx4_ib_create_srq(struct ib_pd *pd, } err = mlx4_mtt_init(dev->dev, ib_umem_page_count(srq->umem), - ilog2(srq->umem->page_size), &srq->mtt); + srq->umem->page_shift, &srq->mtt); if (err) goto err_buf; diff --git a/drivers/infiniband/hw/mlx5/cmd.c b/drivers/infiniband/hw/mlx5/cmd.c index cdc2d3017da7..18d5e1db93ed 100644 --- a/drivers/infiniband/hw/mlx5/cmd.c +++ b/drivers/infiniband/hw/mlx5/cmd.c @@ -46,3 +46,14 @@ int mlx5_cmd_null_mkey(struct mlx5_core_dev *dev, u32 *null_mkey) null_mkey); return err; } + +int mlx5_cmd_query_cong_counter(struct mlx5_core_dev *dev, + bool reset, void *out, int out_size) +{ + u32 in[MLX5_ST_SZ_DW(query_cong_statistics_in)] = { }; + + MLX5_SET(query_cong_statistics_in, in, opcode, + MLX5_CMD_OP_QUERY_CONG_STATISTICS); + MLX5_SET(query_cong_statistics_in, in, clear, reset); + return mlx5_cmd_exec(dev, in, sizeof(in), out, out_size); +} diff --git a/drivers/infiniband/hw/mlx5/cmd.h b/drivers/infiniband/hw/mlx5/cmd.h index 7ca8a7b6434d..fa09228193a6 100644 --- a/drivers/infiniband/hw/mlx5/cmd.h +++ b/drivers/infiniband/hw/mlx5/cmd.h @@ -37,4 +37,6 @@ #include <linux/mlx5/driver.h> int mlx5_cmd_null_mkey(struct mlx5_core_dev *dev, u32 *null_mkey); +int mlx5_cmd_query_cong_counter(struct mlx5_core_dev *dev, + bool reset, void *out, int out_size); #endif /* MLX5_IB_CMD_H */ diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c index 31803b367104..94c049b62c2f 100644 --- a/drivers/infiniband/hw/mlx5/cq.c +++ b/drivers/infiniband/hw/mlx5/cq.c @@ -172,6 +172,8 @@ static void handle_responder(struct ib_wc *wc, struct mlx5_cqe64 *cqe, struct mlx5_ib_srq *srq; struct mlx5_ib_wq *wq; u16 wqe_ctr; + u8 roce_packet_type; + bool vlan_present; u8 g; if (qp->ibqp.srq || qp->ibqp.xrcd) { @@ -223,7 +225,6 @@ static void handle_responder(struct ib_wc *wc, struct mlx5_cqe64 *cqe, break; } wc->slid = be16_to_cpu(cqe->slid); - wc->sl = (be32_to_cpu(cqe->flags_rqpn) >> 24) & 0xf; wc->src_qp = be32_to_cpu(cqe->flags_rqpn) & 0xffffff; wc->dlid_path_bits = cqe->ml_path; g = (be32_to_cpu(cqe->flags_rqpn) >> 28) & 3; @@ -237,10 +238,22 @@ static void handle_responder(struct ib_wc *wc, struct mlx5_cqe64 *cqe, wc->pkey_index = 0; } - if (ll != IB_LINK_LAYER_ETHERNET) + if (ll != IB_LINK_LAYER_ETHERNET) { + wc->sl = (be32_to_cpu(cqe->flags_rqpn) >> 24) & 0xf; return; + } + + vlan_present = cqe->l4_l3_hdr_type & 0x1; + roce_packet_type = (be32_to_cpu(cqe->flags_rqpn) >> 24) & 0x3; + if (vlan_present) { + wc->vlan_id = (be16_to_cpu(cqe->vlan_info)) & 0xfff; + wc->sl = (be16_to_cpu(cqe->vlan_info) >> 13) & 0x7; + wc->wc_flags |= IB_WC_WITH_VLAN; + } else { + wc->sl = 0; + } - switch (wc->sl & 0x3) { + switch (roce_packet_type) { case MLX5_CQE_ROCE_L3_HEADER_TYPE_GRH: wc->network_hdr_type = RDMA_NETWORK_IB; break; @@ -818,7 +831,7 @@ static int create_cq_user(struct mlx5_ib_dev *dev, struct ib_udata *udata, return 0; err_cqb: - kfree(cqb); + kfree(*cqb); err_db: mlx5_ib_db_unmap_user(to_mucontext(context), &cq->db); diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 4dc0a8785fe0..9f3ba320ce70 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -57,6 +57,7 @@ #include <linux/mlx5/fs.h> #include <linux/mlx5/vport.h> #include "mlx5_ib.h" +#include "cmd.h" #define DRIVER_NAME "mlx5_ib" #define DRIVER_VERSION "2.2-1" @@ -165,15 +166,82 @@ static struct net_device *mlx5_ib_get_netdev(struct ib_device *device, return ndev; } -static int mlx5_query_port_roce(struct ib_device *device, u8 port_num, - struct ib_port_attr *props) +static int translate_eth_proto_oper(u32 eth_proto_oper, u8 *active_speed, + u8 *active_width) +{ + switch (eth_proto_oper) { + case MLX5E_PROT_MASK(MLX5E_1000BASE_CX_SGMII): + case MLX5E_PROT_MASK(MLX5E_1000BASE_KX): + case MLX5E_PROT_MASK(MLX5E_100BASE_TX): + case MLX5E_PROT_MASK(MLX5E_1000BASE_T): + *active_width = IB_WIDTH_1X; + *active_speed = IB_SPEED_SDR; + break; + case MLX5E_PROT_MASK(MLX5E_10GBASE_T): + case MLX5E_PROT_MASK(MLX5E_10GBASE_CX4): + case MLX5E_PROT_MASK(MLX5E_10GBASE_KX4): + case MLX5E_PROT_MASK(MLX5E_10GBASE_KR): + case MLX5E_PROT_MASK(MLX5E_10GBASE_CR): + case MLX5E_PROT_MASK(MLX5E_10GBASE_SR): + case MLX5E_PROT_MASK(MLX5E_10GBASE_ER): + *active_width = IB_WIDTH_1X; + *active_speed = IB_SPEED_QDR; + break; + case MLX5E_PROT_MASK(MLX5E_25GBASE_CR): + case MLX5E_PROT_MASK(MLX5E_25GBASE_KR): + case MLX5E_PROT_MASK(MLX5E_25GBASE_SR): + *active_width = IB_WIDTH_1X; + *active_speed = IB_SPEED_EDR; + break; + case MLX5E_PROT_MASK(MLX5E_40GBASE_CR4): + case MLX5E_PROT_MASK(MLX5E_40GBASE_KR4): + case MLX5E_PROT_MASK(MLX5E_40GBASE_SR4): + case MLX5E_PROT_MASK(MLX5E_40GBASE_LR4): + *active_width = IB_WIDTH_4X; + *active_speed = IB_SPEED_QDR; + break; + case MLX5E_PROT_MASK(MLX5E_50GBASE_CR2): + case MLX5E_PROT_MASK(MLX5E_50GBASE_KR2): + case MLX5E_PROT_MASK(MLX5E_50GBASE_SR2): + *active_width = IB_WIDTH_1X; + *active_speed = IB_SPEED_HDR; + break; + case MLX5E_PROT_MASK(MLX5E_56GBASE_R4): + *active_width = IB_WIDTH_4X; + *active_speed = IB_SPEED_FDR; + break; + case MLX5E_PROT_MASK(MLX5E_100GBASE_CR4): + case MLX5E_PROT_MASK(MLX5E_100GBASE_SR4): + case MLX5E_PROT_MASK(MLX5E_100GBASE_KR4): + case MLX5E_PROT_MASK(MLX5E_100GBASE_LR4): + *active_width = IB_WIDTH_4X; + *active_speed = IB_SPEED_EDR; + break; + default: + return -EINVAL; + } + + return 0; +} + +static void 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; struct net_device *ndev, *upper; enum ib_mtu ndev_ib_mtu; u16 qkey_viol_cntr; + u32 eth_prot_oper; - /* props being zeroed by the caller, avoid zeroing it here */ + /* 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; + + translate_eth_proto_oper(eth_prot_oper, &props->active_speed, + &props->active_width); props->port_cap_flags |= IB_PORT_CM_SUP; props->port_cap_flags |= IB_PORT_IP_BASED_GIDS; @@ -191,7 +259,7 @@ static int mlx5_query_port_roce(struct ib_device *device, u8 port_num, ndev = mlx5_ib_get_netdev(device, port_num); if (!ndev) - return 0; + return; if (mlx5_lag_is_active(dev->mdev)) { rcu_read_lock(); @@ -214,11 +282,6 @@ static int mlx5_query_port_roce(struct ib_device *device, u8 port_num, dev_put(ndev); props->active_mtu = min(props->max_mtu, ndev_ib_mtu); - - props->active_width = IB_WIDTH_4X; /* TODO */ - props->active_speed = IB_SPEED_QDR; /* TODO */ - - return 0; } static void ib_gid_to_mlx5_roce_addr(const union ib_gid *gid, @@ -924,7 +987,8 @@ int mlx5_ib_query_port(struct ib_device *ibdev, u8 port, return mlx5_query_hca_port(ibdev, port, props); case MLX5_VPORT_ACCESS_METHOD_NIC: - return mlx5_query_port_roce(ibdev, port, props); + mlx5_query_port_roce(ibdev, port, props); + return 0; default: return -EINVAL; @@ -1478,7 +1542,7 @@ static void mlx5_ib_disassociate_ucontext(struct ib_ucontext *ibcontext) /* need to protect from a race on closing the vma as part of * mlx5_ib_vma_close. */ - down_read(&owning_mm->mmap_sem); + down_write(&owning_mm->mmap_sem); list_for_each_entry_safe(vma_private, n, &context->vma_private_list, list) { vma = vma_private->vma; @@ -1488,11 +1552,12 @@ static void mlx5_ib_disassociate_ucontext(struct ib_ucontext *ibcontext) /* context going to be destroyed, should * not access ops any more. */ + vma->vm_flags &= ~(VM_SHARED | VM_MAYSHARE); vma->vm_ops = NULL; list_del(&vma_private->list); kfree(vma_private); } - up_read(&owning_mm->mmap_sem); + up_write(&owning_mm->mmap_sem); mmput(owning_mm); put_task_struct(owning_process); } @@ -1726,6 +1791,7 @@ static void set_tos(void *outer_c, void *outer_v, u8 mask, u8 val) #define LAST_TCP_UDP_FIELD src_port #define LAST_TUNNEL_FIELD tunnel_id #define LAST_FLOW_TAG_FIELD tag_id +#define LAST_DROP_FIELD size /* Field is the last supported field */ #define FIELDS_NOT_SUPPORTED(filter, field)\ @@ -1735,8 +1801,11 @@ static void set_tos(void *outer_c, void *outer_v, u8 mask, u8 val) offsetof(typeof(filter), field) -\ sizeof(filter.field)) -static int parse_flow_attr(u32 *match_c, u32 *match_v, - const union ib_flow_spec *ib_spec, u32 *tag_id) +#define IPV4_VERSION 4 +#define IPV6_VERSION 6 +static int parse_flow_attr(struct mlx5_core_dev *mdev, u32 *match_c, + u32 *match_v, const union ib_flow_spec *ib_spec, + u32 *tag_id, bool *is_drop) { void *misc_params_c = MLX5_ADDR_OF(fte_match_param, match_c, misc_parameters); @@ -1744,17 +1813,22 @@ static int parse_flow_attr(u32 *match_c, u32 *match_v, misc_parameters); void *headers_c; void *headers_v; + int match_ipv; if (ib_spec->type & IB_FLOW_SPEC_INNER) { headers_c = MLX5_ADDR_OF(fte_match_param, match_c, inner_headers); headers_v = MLX5_ADDR_OF(fte_match_param, match_v, inner_headers); + match_ipv = MLX5_CAP_FLOWTABLE_NIC_RX(mdev, + ft_field_support.inner_ip_version); } else { headers_c = MLX5_ADDR_OF(fte_match_param, match_c, outer_headers); headers_v = MLX5_ADDR_OF(fte_match_param, match_v, outer_headers); + match_ipv = MLX5_CAP_FLOWTABLE_NIC_RX(mdev, + ft_field_support.outer_ip_version); } switch (ib_spec->type & ~IB_FLOW_SPEC_INNER) { @@ -1810,10 +1884,17 @@ static int parse_flow_attr(u32 *match_c, u32 *match_v, if (FIELDS_NOT_SUPPORTED(ib_spec->ipv4.mask, LAST_IPV4_FIELD)) return -EOPNOTSUPP; - MLX5_SET(fte_match_set_lyr_2_4, headers_c, - ethertype, 0xffff); - MLX5_SET(fte_match_set_lyr_2_4, headers_v, - ethertype, ETH_P_IP); + if (match_ipv) { + MLX5_SET(fte_match_set_lyr_2_4, headers_c, + ip_version, 0xf); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, + ip_version, IPV4_VERSION); + } else { + MLX5_SET(fte_match_set_lyr_2_4, headers_c, + ethertype, 0xffff); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, + ethertype, ETH_P_IP); + } memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, src_ipv4_src_ipv6.ipv4_layout.ipv4), @@ -1842,10 +1923,17 @@ static int parse_flow_attr(u32 *match_c, u32 *match_v, if (FIELDS_NOT_SUPPORTED(ib_spec->ipv6.mask, LAST_IPV6_FIELD)) return -EOPNOTSUPP; - MLX5_SET(fte_match_set_lyr_2_4, headers_c, - ethertype, 0xffff); - MLX5_SET(fte_match_set_lyr_2_4, headers_v, - ethertype, ETH_P_IPV6); + if (match_ipv) { + MLX5_SET(fte_match_set_lyr_2_4, headers_c, + ip_version, 0xf); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, + ip_version, IPV6_VERSION); + } else { + MLX5_SET(fte_match_set_lyr_2_4, headers_c, + ethertype, 0xffff); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, + ethertype, ETH_P_IPV6); + } memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c, src_ipv4_src_ipv6.ipv6_layout.ipv6), @@ -1937,6 +2025,12 @@ static int parse_flow_attr(u32 *match_c, u32 *match_v, *tag_id = ib_spec->flow_tag.tag_id; break; + case IB_FLOW_SPEC_ACTION_DROP: + if (FIELDS_NOT_SUPPORTED(ib_spec->drop, + LAST_DROP_FIELD)) + return -EOPNOTSUPP; + *is_drop = true; + break; default: return -EINVAL; } @@ -1967,26 +2061,60 @@ static bool flow_is_multicast_only(struct ib_flow_attr *ib_attr) is_multicast_ether_addr(eth_spec->val.dst_mac); } -static bool is_valid_attr(const struct ib_flow_attr *flow_attr) +static bool is_valid_ethertype(struct mlx5_core_dev *mdev, + const struct ib_flow_attr *flow_attr, + bool check_inner) { union ib_flow_spec *ib_spec = (union ib_flow_spec *)(flow_attr + 1); - bool has_ipv4_spec = false; - bool eth_type_ipv4 = true; + int match_ipv = check_inner ? + MLX5_CAP_FLOWTABLE_NIC_RX(mdev, + ft_field_support.inner_ip_version) : + MLX5_CAP_FLOWTABLE_NIC_RX(mdev, + ft_field_support.outer_ip_version); + int inner_bit = check_inner ? IB_FLOW_SPEC_INNER : 0; + bool ipv4_spec_valid, ipv6_spec_valid; + unsigned int ip_spec_type = 0; + bool has_ethertype = false; unsigned int spec_index; + bool mask_valid = true; + u16 eth_type = 0; + bool type_valid; /* Validate that ethertype is correct */ for (spec_index = 0; spec_index < flow_attr->num_of_specs; spec_index++) { - if (ib_spec->type == IB_FLOW_SPEC_ETH && + if ((ib_spec->type == (IB_FLOW_SPEC_ETH | inner_bit)) && ib_spec->eth.mask.ether_type) { - if (!((ib_spec->eth.mask.ether_type == htons(0xffff)) && - ib_spec->eth.val.ether_type == htons(ETH_P_IP))) - eth_type_ipv4 = false; - } else if (ib_spec->type == IB_FLOW_SPEC_IPV4) { - has_ipv4_spec = true; + mask_valid = (ib_spec->eth.mask.ether_type == + htons(0xffff)); + has_ethertype = true; + eth_type = ntohs(ib_spec->eth.val.ether_type); + } else if ((ib_spec->type == (IB_FLOW_SPEC_IPV4 | inner_bit)) || + (ib_spec->type == (IB_FLOW_SPEC_IPV6 | inner_bit))) { + ip_spec_type = ib_spec->type; } ib_spec = (void *)ib_spec + ib_spec->size; } - return !has_ipv4_spec || eth_type_ipv4; + + type_valid = (!has_ethertype) || (!ip_spec_type); + if (!type_valid && mask_valid) { + ipv4_spec_valid = (eth_type == ETH_P_IP) && + (ip_spec_type == (IB_FLOW_SPEC_IPV4 | inner_bit)); + ipv6_spec_valid = (eth_type == ETH_P_IPV6) && + (ip_spec_type == (IB_FLOW_SPEC_IPV6 | inner_bit)); + + type_valid = (ipv4_spec_valid) || (ipv6_spec_valid) || + (((eth_type == ETH_P_MPLS_UC) || + (eth_type == ETH_P_MPLS_MC)) && match_ipv); + } + + return type_valid; +} + +static bool is_valid_attr(struct mlx5_core_dev *mdev, + const struct ib_flow_attr *flow_attr) +{ + return is_valid_ethertype(mdev, flow_attr, false) && + is_valid_ethertype(mdev, flow_attr, true); } static void put_flow_table(struct mlx5_ib_dev *dev, @@ -2038,8 +2166,8 @@ enum flow_table_type { MLX5_IB_FT_TX }; -#define MLX5_FS_MAX_TYPES 10 -#define MLX5_FS_MAX_ENTRIES 32000UL +#define MLX5_FS_MAX_TYPES 6 +#define MLX5_FS_MAX_ENTRIES BIT(16) static struct mlx5_ib_flow_prio *get_flow_table(struct mlx5_ib_dev *dev, struct ib_flow_attr *flow_attr, enum flow_table_type ft_type) @@ -2048,11 +2176,14 @@ static struct mlx5_ib_flow_prio *get_flow_table(struct mlx5_ib_dev *dev, struct mlx5_flow_namespace *ns = NULL; struct mlx5_ib_flow_prio *prio; struct mlx5_flow_table *ft; + int max_table_size; int num_entries; int num_groups; int priority; int err = 0; + max_table_size = BIT(MLX5_CAP_FLOWTABLE_NIC_RX(dev->mdev, + log_max_ft_size)); if (flow_attr->type == IB_FLOW_ATTR_NORMAL) { if (flow_is_multicast_only(flow_attr) && !dont_trap) @@ -2091,6 +2222,9 @@ static struct mlx5_ib_flow_prio *get_flow_table(struct mlx5_ib_dev *dev, if (!ns) return ERR_PTR(-ENOTSUPP); + if (num_entries > max_table_size) + return ERR_PTR(-ENOMEM); + ft = prio->flow_table; if (!ft) { ft = mlx5_create_auto_grouped_flow_table(ns, priority, @@ -2118,12 +2252,15 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, struct mlx5_ib_flow_handler *handler; struct mlx5_flow_act flow_act = {0}; struct mlx5_flow_spec *spec; + struct mlx5_flow_destination *rule_dst = dst; const void *ib_flow = (const void *)flow_attr + sizeof(*flow_attr); unsigned int spec_index; u32 flow_tag = MLX5_FS_DEFAULT_FLOW_TAG; + bool is_drop = false; int err = 0; + int dest_num = 1; - if (!is_valid_attr(flow_attr)) + if (!is_valid_attr(dev->mdev, flow_attr)) return ERR_PTR(-EINVAL); spec = mlx5_vzalloc(sizeof(*spec)); @@ -2136,8 +2273,9 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, INIT_LIST_HEAD(&handler->list); for (spec_index = 0; spec_index < flow_attr->num_of_specs; spec_index++) { - err = parse_flow_attr(spec->match_criteria, - spec->match_value, ib_flow, &flow_tag); + err = parse_flow_attr(dev->mdev, spec->match_criteria, + spec->match_value, + ib_flow, &flow_tag, &is_drop); if (err < 0) goto free; @@ -2145,8 +2283,14 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, } spec->match_criteria_enable = get_match_criteria_enable(spec->match_criteria); - flow_act.action = dst ? MLX5_FLOW_CONTEXT_ACTION_FWD_DEST : - MLX5_FLOW_CONTEXT_ACTION_FWD_NEXT_PRIO; + if (is_drop) { + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP; + rule_dst = NULL; + dest_num = 0; + } else { + flow_act.action = dst ? MLX5_FLOW_CONTEXT_ACTION_FWD_DEST : + MLX5_FLOW_CONTEXT_ACTION_FWD_NEXT_PRIO; + } if (flow_tag != MLX5_FS_DEFAULT_FLOW_TAG && (flow_attr->type == IB_FLOW_ATTR_ALL_DEFAULT || @@ -2159,7 +2303,7 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, flow_act.flow_tag = flow_tag; handler->rule = mlx5_add_flow_rules(ft, spec, &flow_act, - dst, 1); + rule_dst, dest_num); if (IS_ERR(handler->rule)) { err = PTR_ERR(handler->rule); @@ -2315,7 +2459,7 @@ static struct ib_flow *mlx5_ib_create_flow(struct ib_qp *qp, int err; if (flow_attr->priority > MLX5_IB_FLOW_LAST_PRIO) - return ERR_PTR(-ENOSPC); + return ERR_PTR(-ENOMEM); if (domain != IB_FLOW_DOMAIN_USER || flow_attr->port > MLX5_CAP_GEN(dev->mdev, num_ports) || @@ -3133,7 +3277,7 @@ static void mlx5_disable_eth(struct mlx5_ib_dev *dev) mlx5_nic_vport_disable_roce(dev->mdev); } -struct mlx5_ib_q_counter { +struct mlx5_ib_counter { const char *name; size_t offset; }; @@ -3141,18 +3285,18 @@ struct mlx5_ib_q_counter { #define INIT_Q_COUNTER(_name) \ { .name = #_name, .offset = MLX5_BYTE_OFF(query_q_counter_out, _name)} -static const struct mlx5_ib_q_counter basic_q_cnts[] = { +static const struct mlx5_ib_counter basic_q_cnts[] = { INIT_Q_COUNTER(rx_write_requests), INIT_Q_COUNTER(rx_read_requests), INIT_Q_COUNTER(rx_atomic_requests), INIT_Q_COUNTER(out_of_buffer), }; -static const struct mlx5_ib_q_counter out_of_seq_q_cnts[] = { +static const struct mlx5_ib_counter out_of_seq_q_cnts[] = { INIT_Q_COUNTER(out_of_sequence), }; -static const struct mlx5_ib_q_counter retrans_q_cnts[] = { +static const struct mlx5_ib_counter retrans_q_cnts[] = { INIT_Q_COUNTER(duplicate_request), INIT_Q_COUNTER(rnr_nak_retry_err), INIT_Q_COUNTER(packet_seq_err), @@ -3160,22 +3304,31 @@ static const struct mlx5_ib_q_counter retrans_q_cnts[] = { INIT_Q_COUNTER(local_ack_timeout_err), }; -static void mlx5_ib_dealloc_q_counters(struct mlx5_ib_dev *dev) +#define INIT_CONG_COUNTER(_name) \ + { .name = #_name, .offset = \ + MLX5_BYTE_OFF(query_cong_statistics_out, _name ## _high)} + +static const struct mlx5_ib_counter cong_cnts[] = { + INIT_CONG_COUNTER(rp_cnp_ignored), + INIT_CONG_COUNTER(rp_cnp_handled), + INIT_CONG_COUNTER(np_ecn_marked_roce_packets), + INIT_CONG_COUNTER(np_cnp_sent), +}; + +static void mlx5_ib_dealloc_counters(struct mlx5_ib_dev *dev) { unsigned int i; for (i = 0; i < dev->num_ports; i++) { mlx5_core_dealloc_q_counter(dev->mdev, - dev->port[i].q_cnts.set_id); - kfree(dev->port[i].q_cnts.names); - kfree(dev->port[i].q_cnts.offsets); + dev->port[i].cnts.set_id); + kfree(dev->port[i].cnts.names); + kfree(dev->port[i].cnts.offsets); } } -static int __mlx5_ib_alloc_q_counters(struct mlx5_ib_dev *dev, - const char ***names, - size_t **offsets, - u32 *num) +static int __mlx5_ib_alloc_counters(struct mlx5_ib_dev *dev, + struct mlx5_ib_counters *cnts) { u32 num_counters; @@ -3186,27 +3339,32 @@ static int __mlx5_ib_alloc_q_counters(struct mlx5_ib_dev *dev, if (MLX5_CAP_GEN(dev->mdev, retransmission_q_counters)) num_counters += ARRAY_SIZE(retrans_q_cnts); + cnts->num_q_counters = num_counters; - *names = kcalloc(num_counters, sizeof(**names), GFP_KERNEL); - if (!*names) + if (MLX5_CAP_GEN(dev->mdev, cc_query_allowed)) { + cnts->num_cong_counters = ARRAY_SIZE(cong_cnts); + num_counters += ARRAY_SIZE(cong_cnts); + } + + cnts->names = kcalloc(num_counters, sizeof(cnts->names), GFP_KERNEL); + if (!cnts->names) return -ENOMEM; - *offsets = kcalloc(num_counters, sizeof(**offsets), GFP_KERNEL); - if (!*offsets) + cnts->offsets = kcalloc(num_counters, + sizeof(cnts->offsets), GFP_KERNEL); + if (!cnts->offsets) goto err_names; - *num = num_counters; - return 0; err_names: - kfree(*names); + kfree(cnts->names); return -ENOMEM; } -static void mlx5_ib_fill_q_counters(struct mlx5_ib_dev *dev, - const char **names, - size_t *offsets) +static void mlx5_ib_fill_counters(struct mlx5_ib_dev *dev, + const char **names, + size_t *offsets) { int i; int j = 0; @@ -3229,9 +3387,16 @@ static void mlx5_ib_fill_q_counters(struct mlx5_ib_dev *dev, offsets[j] = retrans_q_cnts[i].offset; } } + + if (MLX5_CAP_GEN(dev->mdev, cc_query_allowed)) { + for (i = 0; i < ARRAY_SIZE(cong_cnts); i++, j++) { + names[j] = cong_cnts[i].name; + offsets[j] = cong_cnts[i].offset; + } + } } -static int mlx5_ib_alloc_q_counters(struct mlx5_ib_dev *dev) +static int mlx5_ib_alloc_counters(struct mlx5_ib_dev *dev) { int i; int ret; @@ -3240,7 +3405,7 @@ static int mlx5_ib_alloc_q_counters(struct mlx5_ib_dev *dev) struct mlx5_ib_port *port = &dev->port[i]; ret = mlx5_core_alloc_q_counter(dev->mdev, - &port->q_cnts.set_id); + &port->cnts.set_id); if (ret) { mlx5_ib_warn(dev, "couldn't allocate queue counter for port %d, err %d\n", @@ -3248,15 +3413,12 @@ static int mlx5_ib_alloc_q_counters(struct mlx5_ib_dev *dev) goto dealloc_counters; } - ret = __mlx5_ib_alloc_q_counters(dev, - &port->q_cnts.names, - &port->q_cnts.offsets, - &port->q_cnts.num_counters); + ret = __mlx5_ib_alloc_counters(dev, &port->cnts); if (ret) goto dealloc_counters; - mlx5_ib_fill_q_counters(dev, port->q_cnts.names, - port->q_cnts.offsets); + mlx5_ib_fill_counters(dev, port->cnts.names, + port->cnts.offsets); } return 0; @@ -3264,7 +3426,7 @@ static int mlx5_ib_alloc_q_counters(struct mlx5_ib_dev *dev) dealloc_counters: while (--i >= 0) mlx5_core_dealloc_q_counter(dev->mdev, - dev->port[i].q_cnts.set_id); + dev->port[i].cnts.set_id); return ret; } @@ -3279,44 +3441,93 @@ static struct rdma_hw_stats *mlx5_ib_alloc_hw_stats(struct ib_device *ibdev, if (port_num == 0) return NULL; - return rdma_alloc_hw_stats_struct(port->q_cnts.names, - port->q_cnts.num_counters, + return rdma_alloc_hw_stats_struct(port->cnts.names, + port->cnts.num_q_counters + + port->cnts.num_cong_counters, RDMA_HW_STATS_DEFAULT_LIFESPAN); } -static int mlx5_ib_get_hw_stats(struct ib_device *ibdev, - struct rdma_hw_stats *stats, - u8 port_num, int index) +static int mlx5_ib_query_q_counters(struct mlx5_ib_dev *dev, + struct mlx5_ib_port *port, + struct rdma_hw_stats *stats) { - struct mlx5_ib_dev *dev = to_mdev(ibdev); - struct mlx5_ib_port *port = &dev->port[port_num - 1]; int outlen = MLX5_ST_SZ_BYTES(query_q_counter_out); void *out; __be32 val; - int ret; - int i; - - if (!stats) - return -ENOSYS; + int ret, i; out = mlx5_vzalloc(outlen); if (!out) return -ENOMEM; ret = mlx5_core_query_q_counter(dev->mdev, - port->q_cnts.set_id, 0, + port->cnts.set_id, 0, out, outlen); if (ret) goto free; - for (i = 0; i < port->q_cnts.num_counters; i++) { - val = *(__be32 *)(out + port->q_cnts.offsets[i]); + for (i = 0; i < port->cnts.num_q_counters; i++) { + val = *(__be32 *)(out + port->cnts.offsets[i]); stats->value[i] = (u64)be32_to_cpu(val); } free: kvfree(out); - return port->q_cnts.num_counters; + return ret; +} + +static int mlx5_ib_query_cong_counters(struct mlx5_ib_dev *dev, + struct mlx5_ib_port *port, + struct rdma_hw_stats *stats) +{ + int outlen = MLX5_ST_SZ_BYTES(query_cong_statistics_out); + void *out; + int ret, i; + int offset = port->cnts.num_q_counters; + + out = mlx5_vzalloc(outlen); + if (!out) + return -ENOMEM; + + ret = mlx5_cmd_query_cong_counter(dev->mdev, false, out, outlen); + if (ret) + goto free; + + for (i = 0; i < port->cnts.num_cong_counters; i++) { + stats->value[i + offset] = + be64_to_cpup((__be64 *)(out + + port->cnts.offsets[i + offset])); + } + +free: + kvfree(out); + return ret; +} + +static int mlx5_ib_get_hw_stats(struct ib_device *ibdev, + struct rdma_hw_stats *stats, + u8 port_num, int index) +{ + struct mlx5_ib_dev *dev = to_mdev(ibdev); + struct mlx5_ib_port *port = &dev->port[port_num - 1]; + int ret, num_counters; + + if (!stats) + return -EINVAL; + + ret = mlx5_ib_query_q_counters(dev, port, stats); + if (ret) + return ret; + num_counters = port->cnts.num_q_counters; + + if (MLX5_CAP_GEN(dev->mdev, cc_query_allowed)) { + ret = mlx5_ib_query_cong_counters(dev, port, stats); + if (ret) + return ret; + num_counters += port->cnts.num_cong_counters; + } + + return num_counters; } static void *mlx5_ib_add(struct mlx5_core_dev *mdev) @@ -3523,14 +3734,14 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev) goto err_rsrc; if (MLX5_CAP_GEN(dev->mdev, max_qp_cnt)) { - err = mlx5_ib_alloc_q_counters(dev); + err = mlx5_ib_alloc_counters(dev); if (err) goto err_odp; } dev->mdev->priv.uar = mlx5_get_uars_page(dev->mdev); if (!dev->mdev->priv.uar) - goto err_q_cnt; + goto err_cnt; err = mlx5_alloc_bfreg(dev->mdev, &dev->bfreg, false, false); if (err) @@ -3574,9 +3785,9 @@ err_bfreg: err_uar_page: mlx5_put_uars_page(dev->mdev, dev->mdev->priv.uar); -err_q_cnt: +err_cnt: if (MLX5_CAP_GEN(dev->mdev, max_qp_cnt)) - mlx5_ib_dealloc_q_counters(dev); + mlx5_ib_dealloc_counters(dev); err_odp: mlx5_ib_odp_remove_one(dev); @@ -3610,7 +3821,7 @@ static void mlx5_ib_remove(struct mlx5_core_dev *mdev, void *context) mlx5_free_bfreg(dev->mdev, &dev->bfreg); mlx5_put_uars_page(dev->mdev, mdev->priv.uar); if (MLX5_CAP_GEN(dev->mdev, max_qp_cnt)) - mlx5_ib_dealloc_q_counters(dev); + mlx5_ib_dealloc_counters(dev); destroy_umrc_res(dev); mlx5_ib_odp_remove_one(dev); destroy_dev_resources(&dev->devr); diff --git a/drivers/infiniband/hw/mlx5/mem.c b/drivers/infiniband/hw/mlx5/mem.c index 778d8a18925f..914f212e7ef6 100644 --- a/drivers/infiniband/hw/mlx5/mem.c +++ b/drivers/infiniband/hw/mlx5/mem.c @@ -59,15 +59,14 @@ void mlx5_ib_cont_pages(struct ib_umem *umem, u64 addr, u64 pfn; struct scatterlist *sg; int entry; - unsigned long page_shift = ilog2(umem->page_size); + unsigned long page_shift = umem->page_shift; - /* With ODP we must always match OS page size. */ if (umem->odp_data) { - *count = ib_umem_page_count(umem); - *shift = PAGE_SHIFT; - *ncont = *count; + *ncont = ib_umem_page_count(umem); + *count = *ncont << (page_shift - PAGE_SHIFT); + *shift = page_shift; if (order) - *order = ilog2(roundup_pow_of_two(*count)); + *order = ilog2(roundup_pow_of_two(*ncont)); return; } @@ -156,7 +155,7 @@ void __mlx5_ib_populate_pas(struct mlx5_ib_dev *dev, struct ib_umem *umem, int page_shift, size_t offset, size_t num_pages, __be64 *pas, int access_flags) { - unsigned long umem_page_shift = ilog2(umem->page_size); + unsigned long umem_page_shift = umem->page_shift; int shift = page_shift - umem_page_shift; int mask = (1 << shift) - 1; int i, k, idx; diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h index ce8ba617d46e..93c646691208 100644 --- a/drivers/infiniband/hw/mlx5/mlx5_ib.h +++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h @@ -513,6 +513,7 @@ struct mlx5_ib_mr { struct mlx5_ib_mw { struct ib_mw ibmw; struct mlx5_core_mkey mmkey; + int ndescs; }; struct mlx5_ib_umr_context { @@ -595,15 +596,16 @@ struct mlx5_ib_resources { struct mutex mutex; }; -struct mlx5_ib_q_counters { +struct mlx5_ib_counters { const char **names; size_t *offsets; - u32 num_counters; + u32 num_q_counters; + u32 num_cong_counters; u16 set_id; }; struct mlx5_ib_port { - struct mlx5_ib_q_counters q_cnts; + struct mlx5_ib_counters cnts; }; struct mlx5_roce { diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c index b8f9382a8b7d..93c0e82aa491 100644 --- a/drivers/infiniband/hw/mlx5/mr.c +++ b/drivers/infiniband/hw/mlx5/mr.c @@ -1009,7 +1009,7 @@ int mlx5_ib_update_xlt(struct mlx5_ib_mr *mr, u64 idx, int npages, } if (!xlt) { - uctx = to_mucontext(mr->ibmr.uobject->context); + uctx = to_mucontext(mr->ibmr.pd->uobject->context); mlx5_ib_warn(dev, "Using XLT emergency buffer\n"); size = PAGE_SIZE; xlt = (void *)uctx->upd_xlt_page; @@ -1045,8 +1045,9 @@ int mlx5_ib_update_xlt(struct mlx5_ib_mr *mr, u64 idx, int npages, for (pages_mapped = 0; pages_mapped < pages_to_map && !err; pages_mapped += pages_iter, idx += pages_iter) { + npages = min_t(int, pages_iter, pages_to_map - pages_mapped); dma_sync_single_for_cpu(ddev, dma, size, DMA_TO_DEVICE); - npages = populate_xlt(mr, idx, pages_iter, xlt, + npages = populate_xlt(mr, idx, npages, xlt, page_shift, size, flags); dma_sync_single_for_device(ddev, dma, size, DMA_TO_DEVICE); @@ -1687,6 +1688,7 @@ struct ib_mw *mlx5_ib_alloc_mw(struct ib_pd *pd, enum ib_mw_type type, mw->mmkey.type = MLX5_MKEY_MW; mw->ibmw.rkey = mw->mmkey.key; + mw->ndescs = ndescs; resp.response_length = min(offsetof(typeof(resp), response_length) + sizeof(resp.response_length), udata->outlen); diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c index d7b12f0750e2..ae0746754008 100644 --- a/drivers/infiniband/hw/mlx5/odp.c +++ b/drivers/infiniband/hw/mlx5/odp.c @@ -57,7 +57,7 @@ static int check_parent(struct ib_umem_odp *odp, { struct mlx5_ib_mr *mr = odp->private; - return mr && mr->parent == parent; + return mr && mr->parent == parent && !odp->dying; } static struct ib_umem_odp *odp_next(struct ib_umem_odp *odp) @@ -158,13 +158,6 @@ static void mr_leaf_free_action(struct work_struct *work) mr->parent = NULL; synchronize_srcu(&mr->dev->mr_srcu); - if (!READ_ONCE(odp->dying)) { - mr->parent = imr; - if (atomic_dec_and_test(&imr->num_leaf_free)) - wake_up(&imr->q_leaf_free); - return; - } - ib_umem_release(odp->umem); if (imr->live) mlx5_ib_update_xlt(imr, idx, 1, 0, @@ -206,8 +199,8 @@ void mlx5_ib_invalidate_range(struct ib_umem *umem, unsigned long start, * but they will write 0s as well, so no difference in the end result. */ - for (addr = start; addr < end; addr += (u64)umem->page_size) { - idx = (addr - ib_umem_start(umem)) / PAGE_SIZE; + for (addr = start; addr < end; addr += BIT(umem->page_shift)) { + idx = (addr - ib_umem_start(umem)) >> umem->page_shift; /* * Strive to write the MTTs in chunks, but avoid overwriting * non-existing MTTs. The huristic here can be improved to @@ -225,8 +218,7 @@ void mlx5_ib_invalidate_range(struct ib_umem *umem, unsigned long start, if (in_block && umr_offset == 0) { mlx5_ib_update_xlt(mr, blk_start_idx, - idx - blk_start_idx, - PAGE_SHIFT, + idx - blk_start_idx, 0, MLX5_IB_UPD_XLT_ZAP | MLX5_IB_UPD_XLT_ATOMIC); in_block = 0; @@ -235,8 +227,7 @@ void mlx5_ib_invalidate_range(struct ib_umem *umem, unsigned long start, } if (in_block) mlx5_ib_update_xlt(mr, blk_start_idx, - idx - blk_start_idx + 1, - PAGE_SHIFT, + idx - blk_start_idx + 1, 0, MLX5_IB_UPD_XLT_ZAP | MLX5_IB_UPD_XLT_ATOMIC); /* @@ -297,24 +288,6 @@ void mlx5_ib_internal_fill_odp_caps(struct mlx5_ib_dev *dev) return; } -static struct mlx5_ib_mr *mlx5_ib_odp_find_mr_lkey(struct mlx5_ib_dev *dev, - u32 key) -{ - u32 base_key = mlx5_base_mkey(key); - struct mlx5_core_mkey *mmkey = __mlx5_mr_lookup(dev->mdev, base_key); - struct mlx5_ib_mr *mr; - - if (!mmkey || mmkey->key != key || mmkey->type != MLX5_MKEY_MR) - return NULL; - - mr = container_of(mmkey, struct mlx5_ib_mr, mmkey); - - if (!mr->live) - return NULL; - - return container_of(mmkey, struct mlx5_ib_mr, mmkey); -} - static void mlx5_ib_page_fault_resume(struct mlx5_ib_dev *dev, struct mlx5_pagefault *pfault, int error) @@ -436,8 +409,6 @@ next_mr: nentries++; } - odp->dying = 0; - /* Return first odp if region not covered by single one */ if (likely(!result)) result = odp; @@ -522,77 +493,38 @@ void mlx5_ib_free_implicit_mr(struct mlx5_ib_mr *imr) wait_event(imr->q_leaf_free, !atomic_read(&imr->num_leaf_free)); } -/* - * Handle a single data segment in a page-fault WQE or RDMA region. - * - * Returns number of pages retrieved on success. The caller may continue to - * the next data segment. - * Can return the following error codes: - * -EAGAIN to designate a temporary error. The caller will abort handling the - * page fault and resolve it. - * -EFAULT when there's an error mapping the requested pages. The caller will - * abort the page fault handling. - */ -static int pagefault_single_data_segment(struct mlx5_ib_dev *dev, - u32 key, u64 io_virt, size_t bcnt, - u32 *bytes_committed, - u32 *bytes_mapped) +static int pagefault_mr(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr, + u64 io_virt, size_t bcnt, u32 *bytes_mapped) { - int srcu_key; - unsigned int current_seq = 0; - u64 start_idx; - int npages = 0, ret = 0; - struct mlx5_ib_mr *mr; u64 access_mask = ODP_READ_ALLOWED_BIT; + int npages = 0, page_shift, np; + u64 start_idx, page_mask; struct ib_umem_odp *odp; - int implicit = 0; + int current_seq; size_t size; - - srcu_key = srcu_read_lock(&dev->mr_srcu); - mr = mlx5_ib_odp_find_mr_lkey(dev, key); - /* - * If we didn't find the MR, it means the MR was closed while we were - * handling the ODP event. In this case we return -EFAULT so that the - * QP will be closed. - */ - if (!mr || !mr->ibmr.pd) { - mlx5_ib_dbg(dev, "Failed to find relevant mr for lkey=0x%06x, probably the MR was destroyed\n", - key); - ret = -EFAULT; - goto srcu_unlock; - } - if (!mr->umem->odp_data) { - mlx5_ib_dbg(dev, "skipping non ODP MR (lkey=0x%06x) in page fault handler.\n", - key); - if (bytes_mapped) - *bytes_mapped += - (bcnt - *bytes_committed); - goto srcu_unlock; - } - - /* - * Avoid branches - this code will perform correctly - * in all iterations (in iteration 2 and above, - * bytes_committed == 0). - */ - io_virt += *bytes_committed; - bcnt -= *bytes_committed; + int ret; if (!mr->umem->odp_data->page_list) { odp = implicit_mr_get_data(mr, io_virt, bcnt); - if (IS_ERR(odp)) { - ret = PTR_ERR(odp); - goto srcu_unlock; - } + if (IS_ERR(odp)) + return PTR_ERR(odp); mr = odp->private; - implicit = 1; } else { odp = mr->umem->odp_data; } next_mr: + size = min_t(size_t, bcnt, ib_umem_end(odp->umem) - io_virt); + + page_shift = mr->umem->page_shift; + page_mask = ~(BIT(page_shift) - 1); + start_idx = (io_virt - (mr->mmkey.iova & page_mask)) >> page_shift; + + if (mr->umem->writable) + access_mask |= ODP_WRITE_ALLOWED_BIT; + current_seq = READ_ONCE(odp->notifiers_seq); /* * Ensure the sequence number is valid for some time before we call @@ -600,51 +532,43 @@ next_mr: */ smp_rmb(); - size = min_t(size_t, bcnt, ib_umem_end(odp->umem) - io_virt); - start_idx = (io_virt - (mr->mmkey.iova & PAGE_MASK)) >> PAGE_SHIFT; - - if (mr->umem->writable) - access_mask |= ODP_WRITE_ALLOWED_BIT; - ret = ib_umem_odp_map_dma_pages(mr->umem, io_virt, size, access_mask, current_seq); if (ret < 0) - goto srcu_unlock; + goto out; - if (ret > 0) { - int np = ret; - - mutex_lock(&odp->umem_mutex); - if (!ib_umem_mmu_notifier_retry(mr->umem, current_seq)) { - /* - * No need to check whether the MTTs really belong to - * this MR, since ib_umem_odp_map_dma_pages already - * checks this. - */ - ret = mlx5_ib_update_xlt(mr, start_idx, np, - PAGE_SHIFT, - MLX5_IB_UPD_XLT_ATOMIC); - } else { - ret = -EAGAIN; - } - mutex_unlock(&odp->umem_mutex); - if (ret < 0) { - if (ret != -EAGAIN) - mlx5_ib_err(dev, "Failed to update mkey page tables\n"); - goto srcu_unlock; - } + np = ret; - if (bytes_mapped) { - u32 new_mappings = np * PAGE_SIZE - - (io_virt - round_down(io_virt, PAGE_SIZE)); - *bytes_mapped += min_t(u32, new_mappings, size); - } + mutex_lock(&odp->umem_mutex); + if (!ib_umem_mmu_notifier_retry(mr->umem, current_seq)) { + /* + * No need to check whether the MTTs really belong to + * this MR, since ib_umem_odp_map_dma_pages already + * checks this. + */ + ret = mlx5_ib_update_xlt(mr, start_idx, np, + page_shift, MLX5_IB_UPD_XLT_ATOMIC); + } else { + ret = -EAGAIN; + } + mutex_unlock(&odp->umem_mutex); - npages += np; + if (ret < 0) { + if (ret != -EAGAIN) + mlx5_ib_err(dev, "Failed to update mkey page tables\n"); + goto out; + } + + if (bytes_mapped) { + u32 new_mappings = (np << page_shift) - + (io_virt - round_down(io_virt, 1 << page_shift)); + *bytes_mapped += min_t(u32, new_mappings, size); } + npages += np << (page_shift - PAGE_SHIFT); bcnt -= size; + if (unlikely(bcnt)) { struct ib_umem_odp *next; @@ -653,17 +577,18 @@ next_mr: if (unlikely(!next || next->umem->address != io_virt)) { mlx5_ib_dbg(dev, "next implicit leaf removed at 0x%llx. got %p\n", io_virt, next); - ret = -EAGAIN; - goto srcu_unlock_no_wait; + return -EAGAIN; } odp = next; mr = odp->private; goto next_mr; } -srcu_unlock: + return npages; + +out: if (ret == -EAGAIN) { - if (implicit || !odp->dying) { + if (mr->parent || !odp->dying) { unsigned long timeout = msecs_to_jiffies(MMU_NOTIFIER_TIMEOUT); @@ -679,7 +604,155 @@ srcu_unlock: } } -srcu_unlock_no_wait: + return ret; +} + +struct pf_frame { + struct pf_frame *next; + u32 key; + u64 io_virt; + size_t bcnt; + int depth; +}; + +/* + * Handle a single data segment in a page-fault WQE or RDMA region. + * + * Returns number of OS pages retrieved on success. The caller may continue to + * the next data segment. + * Can return the following error codes: + * -EAGAIN to designate a temporary error. The caller will abort handling the + * page fault and resolve it. + * -EFAULT when there's an error mapping the requested pages. The caller will + * abort the page fault handling. + */ +static int pagefault_single_data_segment(struct mlx5_ib_dev *dev, + u32 key, u64 io_virt, size_t bcnt, + u32 *bytes_committed, + u32 *bytes_mapped) +{ + int npages = 0, srcu_key, ret, i, outlen, cur_outlen = 0, depth = 0; + struct pf_frame *head = NULL, *frame; + struct mlx5_core_mkey *mmkey; + struct mlx5_ib_mw *mw; + struct mlx5_ib_mr *mr; + struct mlx5_klm *pklm; + u32 *out = NULL; + size_t offset; + + srcu_key = srcu_read_lock(&dev->mr_srcu); + + io_virt += *bytes_committed; + bcnt -= *bytes_committed; + +next_mr: + mmkey = __mlx5_mr_lookup(dev->mdev, mlx5_base_mkey(key)); + if (!mmkey || mmkey->key != key) { + mlx5_ib_dbg(dev, "failed to find mkey %x\n", key); + ret = -EFAULT; + goto srcu_unlock; + } + + switch (mmkey->type) { + case MLX5_MKEY_MR: + mr = container_of(mmkey, struct mlx5_ib_mr, mmkey); + if (!mr->live || !mr->ibmr.pd) { + mlx5_ib_dbg(dev, "got dead MR\n"); + ret = -EFAULT; + goto srcu_unlock; + } + + ret = pagefault_mr(dev, mr, io_virt, bcnt, bytes_mapped); + if (ret < 0) + goto srcu_unlock; + + npages += ret; + ret = 0; + break; + + case MLX5_MKEY_MW: + mw = container_of(mmkey, struct mlx5_ib_mw, mmkey); + + if (depth >= MLX5_CAP_GEN(dev->mdev, max_indirection)) { + mlx5_ib_dbg(dev, "indirection level exceeded\n"); + ret = -EFAULT; + goto srcu_unlock; + } + + outlen = MLX5_ST_SZ_BYTES(query_mkey_out) + + sizeof(*pklm) * (mw->ndescs - 2); + + if (outlen > cur_outlen) { + kfree(out); + out = kzalloc(outlen, GFP_KERNEL); + if (!out) { + ret = -ENOMEM; + goto srcu_unlock; + } + cur_outlen = outlen; + } + + pklm = (struct mlx5_klm *)MLX5_ADDR_OF(query_mkey_out, out, + bsf0_klm0_pas_mtt0_1); + + ret = mlx5_core_query_mkey(dev->mdev, &mw->mmkey, out, outlen); + if (ret) + goto srcu_unlock; + + offset = io_virt - MLX5_GET64(query_mkey_out, out, + memory_key_mkey_entry.start_addr); + + for (i = 0; bcnt && i < mw->ndescs; i++, pklm++) { + if (offset >= be32_to_cpu(pklm->bcount)) { + offset -= be32_to_cpu(pklm->bcount); + continue; + } + + frame = kzalloc(sizeof(*frame), GFP_KERNEL); + if (!frame) { + ret = -ENOMEM; + goto srcu_unlock; + } + + frame->key = be32_to_cpu(pklm->key); + frame->io_virt = be64_to_cpu(pklm->va) + offset; + frame->bcnt = min_t(size_t, bcnt, + be32_to_cpu(pklm->bcount) - offset); + frame->depth = depth + 1; + frame->next = head; + head = frame; + + bcnt -= frame->bcnt; + } + break; + + default: + mlx5_ib_dbg(dev, "wrong mkey type %d\n", mmkey->type); + ret = -EFAULT; + goto srcu_unlock; + } + + if (head) { + frame = head; + head = frame->next; + + key = frame->key; + io_virt = frame->io_virt; + bcnt = frame->bcnt; + depth = frame->depth; + kfree(frame); + + goto next_mr; + } + +srcu_unlock: + while (head) { + frame = head; + head = frame->next; + kfree(frame); + } + kfree(out); + srcu_read_unlock(&dev->mr_srcu, srcu_key); *bytes_committed = 0; return ret ? ret : npages; @@ -997,9 +1070,6 @@ static void mlx5_ib_mr_wqe_pfault_handler(struct mlx5_ib_dev *dev, resume_with_error = 0; goto resolve_page_fault; } else if (ret < 0 || total_wqe_bytes > bytes_mapped) { - if (ret != -ENOENT) - mlx5_ib_err(dev, "PAGE FAULT error: %d. QP 0x%x. type: 0x%x\n", - ret, pfault->wqe.wq_num, pfault->type); goto resolve_page_fault; } @@ -1059,8 +1129,8 @@ static void mlx5_ib_mr_rdma_pfault_handler(struct mlx5_ib_dev *dev, } else if (ret < 0 || pages_in_range(address, length) > ret) { mlx5_ib_page_fault_resume(dev, pfault, 1); if (ret != -ENOENT) - mlx5_ib_warn(dev, "PAGE FAULT error %d. QP 0x%x, type: 0x%x\n", - ret, pfault->token, pfault->type); + mlx5_ib_dbg(dev, "PAGE FAULT error %d. QP 0x%x, type: 0x%x\n", + ret, pfault->token, pfault->type); return; } @@ -1081,8 +1151,8 @@ static void mlx5_ib_mr_rdma_pfault_handler(struct mlx5_ib_dev *dev, prefetch_len, &bytes_committed, NULL); if (ret < 0 && ret != -EAGAIN) { - mlx5_ib_warn(dev, "Prefetch failed. ret: %d, QP 0x%x, address: 0x%.16llx, length = 0x%.16x\n", - ret, pfault->token, address, prefetch_len); + mlx5_ib_dbg(dev, "Prefetch failed. ret: %d, QP 0x%x, address: 0x%.16llx, length = 0x%.16x\n", + ret, pfault->token, address, prefetch_len); } } } diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index ed6320186f89..4e5a811d33c7 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -2799,7 +2799,7 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp, qp->port) - 1; mibport = &dev->port[port_num]; context->qp_counter_set_usr_page |= - cpu_to_be32((u32)(mibport->q_cnts.set_id) << 24); + cpu_to_be32((u32)(mibport->cnts.set_id) << 24); } if (!ibqp->uobject && cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) @@ -2827,7 +2827,7 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp, raw_qp_param.operation = op; if (cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) { - raw_qp_param.rq_q_ctr_id = mibport->q_cnts.set_id; + raw_qp_param.rq_q_ctr_id = mibport->cnts.set_id; raw_qp_param.set_mask |= MLX5_RAW_QP_MOD_SET_RQ_Q_CTR_ID; } @@ -4965,7 +4965,8 @@ int mlx5_ib_modify_wq(struct ib_wq *wq, struct ib_wq_attr *wq_attr, if (MLX5_CAP_GEN(dev->mdev, modify_rq_counter_set_id)) { MLX5_SET64(modify_rq_in, in, modify_bitmask, MLX5_MODIFY_RQ_IN_MODIFY_BITMASK_RQ_COUNTER_SET_ID); - MLX5_SET(rqc, rqc, counter_set_id, dev->port->q_cnts.set_id); + MLX5_SET(rqc, rqc, counter_set_id, + dev->port->cnts.set_id); } else pr_info_once("%s: Receive WQ counters are not supported on current FW\n", dev->ib_dev.name); diff --git a/drivers/infiniband/hw/mthca/mthca_cmd.c b/drivers/infiniband/hw/mthca/mthca_cmd.c index c7f49bbb0c72..9d83a53c0c67 100644 --- a/drivers/infiniband/hw/mthca/mthca_cmd.c +++ b/drivers/infiniband/hw/mthca/mthca_cmd.c @@ -367,12 +367,16 @@ static int mthca_cmd_poll(struct mthca_dev *dev, goto out; } - if (out_is_imm) + if (out_is_imm && out_param) { *out_param = (u64) be32_to_cpu((__force __be32) __raw_readl(dev->hcr + HCR_OUT_PARAM_OFFSET)) << 32 | (u64) be32_to_cpu((__force __be32) __raw_readl(dev->hcr + HCR_OUT_PARAM_OFFSET + 4)); + } else if (out_is_imm) { + err = -EINVAL; + goto out; + } status = be32_to_cpu((__force __be32) __raw_readl(dev->hcr + HCR_STATUS_OFFSET)) >> 24; if (status) { @@ -450,8 +454,12 @@ static int mthca_cmd_wait(struct mthca_dev *dev, err = mthca_status_to_errno(context->status); } - if (out_is_imm) + if (out_is_imm && out_param) { *out_param = context->out_param; + } else if (out_is_imm) { + err = -EINVAL; + goto out; + } out: spin_lock(&dev->cmd.context_lock); diff --git a/drivers/infiniband/hw/mthca/mthca_provider.c b/drivers/infiniband/hw/mthca/mthca_provider.c index 22d0e6ee5af6..e1b8940558d2 100644 --- a/drivers/infiniband/hw/mthca/mthca_provider.c +++ b/drivers/infiniband/hw/mthca/mthca_provider.c @@ -937,7 +937,7 @@ static struct ib_mr *mthca_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, goto err; } - shift = ffs(mr->umem->page_size) - 1; + shift = mr->umem->page_shift; n = mr->umem->nmap; mr->mtt = mthca_alloc_mtt(dev, n); @@ -959,8 +959,7 @@ static struct ib_mr *mthca_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, for_each_sg(mr->umem->sg_head.sgl, sg, mr->umem->nmap, entry) { len = sg_dma_len(sg) >> shift; for (k = 0; k < len; ++k) { - pages[i++] = sg_dma_address(sg) + - mr->umem->page_size * k; + pages[i++] = sg_dma_address(sg) + (k << shift); /* * Be friendly to write_mtt and pass it chunks * of appropriate size. diff --git a/drivers/infiniband/hw/nes/nes_hw.c b/drivers/infiniband/hw/nes/nes_hw.c index 19acd13c6cb1..8f9d8b4ad583 100644 --- a/drivers/infiniband/hw/nes/nes_hw.c +++ b/drivers/infiniband/hw/nes/nes_hw.c @@ -1849,9 +1849,8 @@ int nes_init_nic_qp(struct nes_device *nesdev, struct net_device *netdev) wqe_count -= counter; nes_write32(nesdev->regs+NES_WQE_ALLOC, (counter << 24) | nesvnic->nic.qp_id); } while (wqe_count); - init_timer(&nesvnic->rq_wqes_timer); - nesvnic->rq_wqes_timer.function = nes_rq_wqes_timeout; - nesvnic->rq_wqes_timer.data = (unsigned long)nesvnic; + setup_timer(&nesvnic->rq_wqes_timer, nes_rq_wqes_timeout, + (unsigned long)nesvnic); nes_debug(NES_DBG_INIT, "NAPI support Enabled\n"); if (nesdev->nesadapter->et_use_adaptive_rx_coalesce) { @@ -3055,7 +3054,7 @@ static void nes_cqp_ce_handler(struct nes_device *nesdev, struct nes_hw_cq *cq) memcpy(cqp_wqe, &cqp_request->cqp_wqe, sizeof(*cqp_wqe)); barrier(); - opcode = cqp_wqe->wqe_words[NES_CQP_WQE_OPCODE_IDX]; + opcode = le32_to_cpu(cqp_wqe->wqe_words[NES_CQP_WQE_OPCODE_IDX]); if ((opcode & NES_CQP_OPCODE_MASK) == NES_CQP_DOWNLOAD_SEGMENT) ctx_index = NES_CQP_WQE_DL_COMP_CTX_LOW_IDX; else diff --git a/drivers/infiniband/hw/nes/nes_mgt.c b/drivers/infiniband/hw/nes/nes_mgt.c index 33624f17c347..77226cf4ea02 100644 --- a/drivers/infiniband/hw/nes/nes_mgt.c +++ b/drivers/infiniband/hw/nes/nes_mgt.c @@ -1040,9 +1040,8 @@ int nes_init_mgt_qp(struct nes_device *nesdev, struct net_device *netdev, struct mgtvnic->mgt.rx_skb[counter] = skb; } - init_timer(&mgtvnic->rq_wqes_timer); - mgtvnic->rq_wqes_timer.function = nes_mgt_rq_wqes_timeout; - mgtvnic->rq_wqes_timer.data = (unsigned long)mgtvnic; + setup_timer(&mgtvnic->rq_wqes_timer, nes_mgt_rq_wqes_timeout, + (unsigned long)mgtvnic); wqe_count = NES_MGT_WQ_COUNT - 1; mgtvnic->mgt.rq_head = wqe_count; diff --git a/drivers/infiniband/hw/nes/nes_verbs.c b/drivers/infiniband/hw/nes/nes_verbs.c index ccf0a4cffe9c..c5eb78f4ca8b 100644 --- a/drivers/infiniband/hw/nes/nes_verbs.c +++ b/drivers/infiniband/hw/nes/nes_verbs.c @@ -1308,9 +1308,8 @@ static struct ib_qp *nes_create_qp(struct ib_pd *ibpd, init_completion(&nesqp->rq_drained); nesqp->sig_all = (init_attr->sq_sig_type == IB_SIGNAL_ALL_WR); - init_timer(&nesqp->terminate_timer); - nesqp->terminate_timer.function = nes_terminate_timeout; - nesqp->terminate_timer.data = (unsigned long)nesqp; + setup_timer(&nesqp->terminate_timer, nes_terminate_timeout, + (unsigned long)nesqp); /* update the QP table */ nesdev->nesadapter->qp_table[nesqp->hwqp.qp_id-NES_FIRST_QPN] = nesqp; @@ -2165,9 +2164,9 @@ static struct ib_mr *nes_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, } nes_debug(NES_DBG_MR, "User base = 0x%lX, Virt base = 0x%lX, length = %u," - " offset = %u, page size = %u.\n", + " offset = %u, page size = %lu.\n", (unsigned long int)start, (unsigned long int)virt, (u32)length, - ib_umem_offset(region), region->page_size); + ib_umem_offset(region), BIT(region->page_shift)); skip_pages = ((u32)ib_umem_offset(region)) >> 12; diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c index c52edeafd616..c57e387b55a2 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c @@ -914,21 +914,18 @@ static void build_user_pbes(struct ocrdma_dev *dev, struct ocrdma_mr *mr, pbe = (struct ocrdma_pbe *)pbl_tbl->va; pbe_cnt = 0; - shift = ilog2(umem->page_size); + shift = umem->page_shift; for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) { pages = sg_dma_len(sg) >> shift; for (pg_cnt = 0; pg_cnt < pages; pg_cnt++) { /* store the page address in pbe */ pbe->pa_lo = - cpu_to_le32(sg_dma_address - (sg) + - (umem->page_size * pg_cnt)); + cpu_to_le32(sg_dma_address(sg) + + (pg_cnt << shift)); pbe->pa_hi = - cpu_to_le32(upper_32_bits - ((sg_dma_address - (sg) + - umem->page_size * pg_cnt))); + cpu_to_le32(upper_32_bits(sg_dma_address(sg) + + (pg_cnt << shift))); pbe_cnt += 1; total_num_pbes += 1; pbe++; @@ -978,7 +975,7 @@ struct ib_mr *ocrdma_reg_user_mr(struct ib_pd *ibpd, u64 start, u64 len, if (status) goto umem_err; - mr->hwmr.pbe_size = mr->umem->page_size; + mr->hwmr.pbe_size = BIT(mr->umem->page_shift); mr->hwmr.fbo = ib_umem_offset(mr->umem); mr->hwmr.va = usr_addr; mr->hwmr.len = len; diff --git a/drivers/infiniband/hw/qedr/main.c b/drivers/infiniband/hw/qedr/main.c index ced0461d6e9f..ef11e770f822 100644 --- a/drivers/infiniband/hw/qedr/main.c +++ b/drivers/infiniband/hw/qedr/main.c @@ -340,43 +340,58 @@ static void qedr_remove_sysfiles(struct qedr_dev *dev) static void qedr_pci_set_atomic(struct qedr_dev *dev, struct pci_dev *pdev) { struct pci_dev *bridge; - u32 val; - - dev->atomic_cap = IB_ATOMIC_NONE; + u32 ctl2, cap2; + u16 flags; + int rc; bridge = pdev->bus->self; if (!bridge) - return; - - /* Check whether we are connected directly or via a switch */ - while (bridge && bridge->bus->parent) { - DP_DEBUG(dev, QEDR_MSG_INIT, - "Device is not connected directly to root. bridge->bus->number=%d primary=%d\n", - bridge->bus->number, bridge->bus->primary); - /* Need to check Atomic Op Routing Supported all the way to - * root complex. - */ - pcie_capability_read_dword(bridge, PCI_EXP_DEVCAP2, &val); - if (!(val & PCI_EXP_DEVCAP2_ATOMIC_ROUTE)) { - pcie_capability_clear_word(pdev, - PCI_EXP_DEVCTL2, - PCI_EXP_DEVCTL2_ATOMIC_REQ); - return; - } + goto disable; + + /* Check atomic routing support all the way to root complex */ + while (bridge->bus->parent) { + rc = pcie_capability_read_word(bridge, PCI_EXP_FLAGS, &flags); + if (rc || ((flags & PCI_EXP_FLAGS_VERS) < 2)) + goto disable; + + rc = pcie_capability_read_dword(bridge, PCI_EXP_DEVCAP2, &cap2); + if (rc) + goto disable; + + rc = pcie_capability_read_dword(bridge, PCI_EXP_DEVCTL2, &ctl2); + if (rc) + goto disable; + + if (!(cap2 & PCI_EXP_DEVCAP2_ATOMIC_ROUTE) || + (ctl2 & PCI_EXP_DEVCTL2_ATOMIC_EGRESS_BLOCK)) + goto disable; bridge = bridge->bus->parent->self; } - bridge = pdev->bus->self; - /* according to bridge capability */ - pcie_capability_read_dword(bridge, PCI_EXP_DEVCAP2, &val); - if (val & PCI_EXP_DEVCAP2_ATOMIC_COMP64) { - pcie_capability_set_word(pdev, PCI_EXP_DEVCTL2, - PCI_EXP_DEVCTL2_ATOMIC_REQ); - dev->atomic_cap = IB_ATOMIC_GLOB; - } else { - pcie_capability_clear_word(pdev, PCI_EXP_DEVCTL2, - PCI_EXP_DEVCTL2_ATOMIC_REQ); - } + rc = pcie_capability_read_word(bridge, PCI_EXP_FLAGS, &flags); + if (rc || ((flags & PCI_EXP_FLAGS_VERS) < 2)) + goto disable; + + rc = pcie_capability_read_dword(bridge, PCI_EXP_DEVCAP2, &cap2); + if (rc || !(cap2 & PCI_EXP_DEVCAP2_ATOMIC_COMP64)) + goto disable; + + /* Set atomic operations */ + pcie_capability_set_word(pdev, PCI_EXP_DEVCTL2, + PCI_EXP_DEVCTL2_ATOMIC_REQ); + dev->atomic_cap = IB_ATOMIC_GLOB; + + DP_DEBUG(dev, QEDR_MSG_INIT, "Atomic capability enabled\n"); + + return; + +disable: + pcie_capability_clear_word(pdev, PCI_EXP_DEVCTL2, + PCI_EXP_DEVCTL2_ATOMIC_REQ); + dev->atomic_cap = IB_ATOMIC_NONE; + + DP_DEBUG(dev, QEDR_MSG_INIT, "Atomic capability disabled\n"); + } static const struct qed_rdma_ops *qed_ops; @@ -423,14 +438,21 @@ static irqreturn_t qedr_irq_handler(int irq, void *handle) cq->arm_flags = 0; - if (cq->ibcq.comp_handler) + if (!cq->destroyed && cq->ibcq.comp_handler) (*cq->ibcq.comp_handler) (&cq->ibcq, cq->ibcq.cq_context); + /* The CQ's CNQ notification counter is checked before + * destroying the CQ in a busy-wait loop that waits for all of + * the CQ's CNQ interrupts to be processed. It is increased + * here, only after the completion handler, to ensure that the + * the handler is not running when the CQ is destroyed. + */ + cq->cnq_notif++; + sw_comp_cons = qed_chain_get_cons_idx(&cnq->pbl); cnq->n_comp++; - } qed_ops->rdma_cnq_prod_update(cnq->dev->rdma_ctx, cnq->index, diff --git a/drivers/infiniband/hw/qedr/qedr.h b/drivers/infiniband/hw/qedr/qedr.h index 5cb9195513bd..bf02ae4c8891 100644 --- a/drivers/infiniband/hw/qedr/qedr.h +++ b/drivers/infiniband/hw/qedr/qedr.h @@ -272,6 +272,8 @@ struct qedr_cq { u32 cq_cons; struct qedr_userq q; + u8 destroyed; + u16 cnq_notif; }; struct qedr_pd { @@ -429,7 +431,8 @@ struct qedr_mr { RDMA_CQE_RESPONDER_IMM_FLG_SHIFT) #define QEDR_RESP_RDMA (RDMA_CQE_RESPONDER_RDMA_FLG_MASK << \ RDMA_CQE_RESPONDER_RDMA_FLG_SHIFT) -#define QEDR_RESP_RDMA_IMM (QEDR_RESP_IMM | QEDR_RESP_RDMA) +#define QEDR_RESP_INV (RDMA_CQE_RESPONDER_INV_FLG_MASK << \ + RDMA_CQE_RESPONDER_INV_FLG_SHIFT) static inline void qedr_inc_sw_cons(struct qedr_qp_hwq_info *info) { diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c index 2091902848e6..a3afd2b21cc5 100644 --- a/drivers/infiniband/hw/qedr/verbs.c +++ b/drivers/infiniband/hw/qedr/verbs.c @@ -681,16 +681,16 @@ static void qedr_populate_pbls(struct qedr_dev *dev, struct ib_umem *umem, pbe_cnt = 0; - shift = ilog2(umem->page_size); + shift = umem->page_shift; for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) { pages = sg_dma_len(sg) >> shift; for (pg_cnt = 0; pg_cnt < pages; pg_cnt++) { /* store the page address in pbe */ pbe->lo = cpu_to_le32(sg_dma_address(sg) + - umem->page_size * pg_cnt); + (pg_cnt << shift)); addr = upper_32_bits(sg_dma_address(sg) + - umem->page_size * pg_cnt); + (pg_cnt << shift)); pbe->hi = cpu_to_le32(addr); pbe_cnt++; total_num_pbes++; @@ -822,6 +822,17 @@ int qedr_arm_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags flags) { struct qedr_cq *cq = get_qedr_cq(ibcq); unsigned long sflags; + struct qedr_dev *dev; + + dev = get_qedr_dev(ibcq->device); + + if (cq->destroyed) { + DP_ERR(dev, + "warning: arm was invoked after destroy for cq %p (icid=%d)\n", + cq, cq->icid); + return -EINVAL; + } + if (cq->cq_type == QEDR_CQ_TYPE_GSI) return 0; @@ -987,35 +998,82 @@ int qedr_resize_cq(struct ib_cq *ibcq, int new_cnt, struct ib_udata *udata) return 0; } +#define QEDR_DESTROY_CQ_MAX_ITERATIONS (10) +#define QEDR_DESTROY_CQ_ITER_DURATION (10) + int qedr_destroy_cq(struct ib_cq *ibcq) { struct qedr_dev *dev = get_qedr_dev(ibcq->device); struct qed_rdma_destroy_cq_out_params oparams; struct qed_rdma_destroy_cq_in_params iparams; struct qedr_cq *cq = get_qedr_cq(ibcq); + int iter; + int rc; + + DP_DEBUG(dev, QEDR_MSG_CQ, "destroy cq %p (icid=%d)\n", cq, cq->icid); - DP_DEBUG(dev, QEDR_MSG_CQ, "destroy cq: cq_id %d", cq->icid); + cq->destroyed = 1; /* GSIs CQs are handled by driver, so they don't exist in the FW */ - if (cq->cq_type != QEDR_CQ_TYPE_GSI) { - int rc; + if (cq->cq_type == QEDR_CQ_TYPE_GSI) + goto done; - iparams.icid = cq->icid; - rc = dev->ops->rdma_destroy_cq(dev->rdma_ctx, &iparams, - &oparams); - if (rc) - return rc; - dev->ops->common->chain_free(dev->cdev, &cq->pbl); - } + iparams.icid = cq->icid; + rc = dev->ops->rdma_destroy_cq(dev->rdma_ctx, &iparams, &oparams); + if (rc) + return rc; + + dev->ops->common->chain_free(dev->cdev, &cq->pbl); if (ibcq->uobject && ibcq->uobject->context) { qedr_free_pbl(dev, &cq->q.pbl_info, cq->q.pbl_tbl); ib_umem_release(cq->q.umem); } + /* We don't want the IRQ handler to handle a non-existing CQ so we + * wait until all CNQ interrupts, if any, are received. This will always + * happen and will always happen very fast. If not, then a serious error + * has occured. That is why we can use a long delay. + * We spin for a short time so we don’t lose time on context switching + * in case all the completions are handled in that span. Otherwise + * we sleep for a while and check again. Since the CNQ may be + * associated with (only) the current CPU we use msleep to allow the + * current CPU to be freed. + * The CNQ notification is increased in qedr_irq_handler(). + */ + iter = QEDR_DESTROY_CQ_MAX_ITERATIONS; + while (oparams.num_cq_notif != READ_ONCE(cq->cnq_notif) && iter) { + udelay(QEDR_DESTROY_CQ_ITER_DURATION); + iter--; + } + + iter = QEDR_DESTROY_CQ_MAX_ITERATIONS; + while (oparams.num_cq_notif != READ_ONCE(cq->cnq_notif) && iter) { + msleep(QEDR_DESTROY_CQ_ITER_DURATION); + iter--; + } + + if (oparams.num_cq_notif != cq->cnq_notif) + goto err; + + /* Note that we don't need to have explicit code to wait for the + * completion of the event handler because it is invoked from the EQ. + * Since the destroy CQ ramrod has also been received on the EQ we can + * be certain that there's no event handler in process. + */ +done: + cq->sig = ~cq->sig; + kfree(cq); return 0; + +err: + DP_ERR(dev, + "CQ %p (icid=%d) not freed, expecting %d ints but got %d ints\n", + cq, cq->icid, oparams.num_cq_notif, cq->cnq_notif); + + return -EINVAL; } static inline int get_gid_info_from_table(struct ib_qp *ibqp, @@ -2190,7 +2248,7 @@ struct ib_mr *qedr_reg_user_mr(struct ib_pd *ibpd, u64 start, u64 len, mr->hw_mr.pbl_ptr = mr->info.pbl_table[0].pa; mr->hw_mr.pbl_two_level = mr->info.pbl_info.two_layered; mr->hw_mr.pbl_page_size_log = ilog2(mr->info.pbl_info.pbl_size); - mr->hw_mr.page_size_log = ilog2(mr->umem->page_size); + mr->hw_mr.page_size_log = mr->umem->page_shift; mr->hw_mr.fbo = ib_umem_offset(mr->umem); mr->hw_mr.length = len; mr->hw_mr.vaddr = usr_addr; @@ -2625,6 +2683,8 @@ static int qedr_prepare_reg(struct qedr_qp *qp, fwqe1->addr.lo = lower_32_bits(mr->ibmr.iova); fwqe1->l_key = wr->key; + fwqe2->access_ctrl = 0; + SET_FIELD2(fwqe2->access_ctrl, RDMA_SQ_FMR_WQE_2ND_REMOTE_READ, !!(wr->access & IB_ACCESS_REMOTE_READ)); SET_FIELD2(fwqe2->access_ctrl, RDMA_SQ_FMR_WQE_2ND_REMOTE_WRITE, @@ -3271,57 +3331,81 @@ static int qedr_poll_cq_req(struct qedr_dev *dev, return cnt; } -static void __process_resp_one(struct qedr_dev *dev, struct qedr_qp *qp, - struct qedr_cq *cq, struct ib_wc *wc, - struct rdma_cqe_responder *resp, u64 wr_id) +static inline int qedr_cqe_resp_status_to_ib(u8 status) { - enum ib_wc_status wc_status = IB_WC_SUCCESS; - u8 flags; - - wc->opcode = IB_WC_RECV; - wc->wc_flags = 0; - - switch (resp->status) { + switch (status) { case RDMA_CQE_RESP_STS_LOCAL_ACCESS_ERR: - wc_status = IB_WC_LOC_ACCESS_ERR; - break; + return IB_WC_LOC_ACCESS_ERR; case RDMA_CQE_RESP_STS_LOCAL_LENGTH_ERR: - wc_status = IB_WC_LOC_LEN_ERR; - break; + return IB_WC_LOC_LEN_ERR; case RDMA_CQE_RESP_STS_LOCAL_QP_OPERATION_ERR: - wc_status = IB_WC_LOC_QP_OP_ERR; - break; + return IB_WC_LOC_QP_OP_ERR; case RDMA_CQE_RESP_STS_LOCAL_PROTECTION_ERR: - wc_status = IB_WC_LOC_PROT_ERR; - break; + return IB_WC_LOC_PROT_ERR; case RDMA_CQE_RESP_STS_MEMORY_MGT_OPERATION_ERR: - wc_status = IB_WC_MW_BIND_ERR; - break; + return IB_WC_MW_BIND_ERR; case RDMA_CQE_RESP_STS_REMOTE_INVALID_REQUEST_ERR: - wc_status = IB_WC_REM_INV_RD_REQ_ERR; - break; + return IB_WC_REM_INV_RD_REQ_ERR; case RDMA_CQE_RESP_STS_OK: - wc_status = IB_WC_SUCCESS; - wc->byte_len = le32_to_cpu(resp->length); + return IB_WC_SUCCESS; + default: + return IB_WC_GENERAL_ERR; + } +} + +static inline int qedr_set_ok_cqe_resp_wc(struct rdma_cqe_responder *resp, + struct ib_wc *wc) +{ + wc->status = IB_WC_SUCCESS; + wc->byte_len = le32_to_cpu(resp->length); - flags = resp->flags & QEDR_RESP_RDMA_IMM; + if (resp->flags & QEDR_RESP_IMM) { + wc->ex.imm_data = le32_to_cpu(resp->imm_data_or_inv_r_Key); + wc->wc_flags |= IB_WC_WITH_IMM; - if (flags == QEDR_RESP_RDMA_IMM) + if (resp->flags & QEDR_RESP_RDMA) wc->opcode = IB_WC_RECV_RDMA_WITH_IMM; - if (flags == QEDR_RESP_RDMA_IMM || flags == QEDR_RESP_IMM) { - wc->ex.imm_data = - le32_to_cpu(resp->imm_data_or_inv_r_Key); - wc->wc_flags |= IB_WC_WITH_IMM; - } - break; - default: - wc->status = IB_WC_GENERAL_ERR; - DP_ERR(dev, "Invalid CQE status detected\n"); + if (resp->flags & QEDR_RESP_INV) + return -EINVAL; + + } else if (resp->flags & QEDR_RESP_INV) { + wc->ex.imm_data = le32_to_cpu(resp->imm_data_or_inv_r_Key); + wc->wc_flags |= IB_WC_WITH_INVALIDATE; + + if (resp->flags & QEDR_RESP_RDMA) + return -EINVAL; + + } else if (resp->flags & QEDR_RESP_RDMA) { + return -EINVAL; } - /* fill WC */ - wc->status = wc_status; + return 0; +} + +static void __process_resp_one(struct qedr_dev *dev, struct qedr_qp *qp, + struct qedr_cq *cq, struct ib_wc *wc, + struct rdma_cqe_responder *resp, u64 wr_id) +{ + /* Must fill fields before qedr_set_ok_cqe_resp_wc() */ + wc->opcode = IB_WC_RECV; + wc->wc_flags = 0; + + if (likely(resp->status == RDMA_CQE_RESP_STS_OK)) { + if (qedr_set_ok_cqe_resp_wc(resp, wc)) + DP_ERR(dev, + "CQ %p (icid=%d) has invalid CQE responder flags=0x%x\n", + cq, cq->icid, resp->flags); + + } else { + wc->status = qedr_cqe_resp_status_to_ib(resp->status); + if (wc->status == IB_WC_GENERAL_ERR) + DP_ERR(dev, + "CQ %p (icid=%d) contains an invalid CQE status %d\n", + cq, cq->icid, resp->status); + } + + /* Fill the rest of the WC */ wc->vendor_err = 0; wc->src_qp = qp->id; wc->qp = &qp->ibqp; @@ -3416,6 +3500,13 @@ int qedr_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *wc) int update = 0; int done = 0; + if (cq->destroyed) { + DP_ERR(dev, + "warning: poll was invoked after destroy for cq %p (icid=%d)\n", + cq, cq->icid); + return 0; + } + if (cq->cq_type == QEDR_CQ_TYPE_GSI) return qedr_gsi_poll_cq(ibcq, num_entries, wc); diff --git a/drivers/infiniband/hw/qib/qib_iba6120.c b/drivers/infiniband/hw/qib/qib_iba6120.c index 06de1cbcf67d..e423b71e6ea0 100644 --- a/drivers/infiniband/hw/qib/qib_iba6120.c +++ b/drivers/infiniband/hw/qib/qib_iba6120.c @@ -3295,13 +3295,11 @@ static int init_6120_variables(struct qib_devdata *dd) dd->rhdrhead_intr_off = 1ULL << 32; /* setup the stats timer; the add_timer is done at end of init */ - init_timer(&dd->stats_timer); - dd->stats_timer.function = qib_get_6120_faststats; - dd->stats_timer.data = (unsigned long) dd; + setup_timer(&dd->stats_timer, qib_get_6120_faststats, + (unsigned long)dd); - init_timer(&dd->cspec->pma_timer); - dd->cspec->pma_timer.function = pma_6120_timer; - dd->cspec->pma_timer.data = (unsigned long) ppd; + setup_timer(&dd->cspec->pma_timer, pma_6120_timer, + (unsigned long)ppd); dd->ureg_align = qib_read_kreg32(dd, kr_palign); diff --git a/drivers/infiniband/hw/qib/qib_iba7220.c b/drivers/infiniband/hw/qib/qib_iba7220.c index 55a18384c22d..c3679c48e61c 100644 --- a/drivers/infiniband/hw/qib/qib_iba7220.c +++ b/drivers/infiniband/hw/qib/qib_iba7220.c @@ -4074,9 +4074,8 @@ static int qib_init_7220_variables(struct qib_devdata *dd) if (!qib_mini_init) qib_write_kreg(dd, kr_rcvbthqp, QIB_KD_QP); - init_timer(&ppd->cpspec->chase_timer); - ppd->cpspec->chase_timer.function = reenable_7220_chase; - ppd->cpspec->chase_timer.data = (unsigned long)ppd; + setup_timer(&ppd->cpspec->chase_timer, reenable_7220_chase, + (unsigned long)ppd); qib_num_cfg_vls = 1; /* if any 7220's, only one VL */ diff --git a/drivers/infiniband/hw/qib/qib_iba7322.c b/drivers/infiniband/hw/qib/qib_iba7322.c index af9f596bb68b..bb2439fff8fa 100644 --- a/drivers/infiniband/hw/qib/qib_iba7322.c +++ b/drivers/infiniband/hw/qib/qib_iba7322.c @@ -6611,9 +6611,8 @@ static int qib_init_7322_variables(struct qib_devdata *dd) if (!qib_mini_init) write_7322_init_portregs(ppd); - init_timer(&cp->chase_timer); - cp->chase_timer.function = reenable_chase; - cp->chase_timer.data = (unsigned long)ppd; + setup_timer(&cp->chase_timer, reenable_chase, + (unsigned long)ppd); ppd++; } @@ -6639,9 +6638,8 @@ static int qib_init_7322_variables(struct qib_devdata *dd) (u64) rcv_int_count << IBA7322_HDRHEAD_PKTINT_SHIFT; /* setup the stats timer; the add_timer is done at end of init */ - init_timer(&dd->stats_timer); - dd->stats_timer.function = qib_get_7322_faststats; - dd->stats_timer.data = (unsigned long) dd; + setup_timer(&dd->stats_timer, qib_get_7322_faststats, + (unsigned long)dd); dd->ureg_align = 0x10000; /* 64KB alignment */ diff --git a/drivers/infiniband/hw/qib/qib_init.c b/drivers/infiniband/hw/qib/qib_init.c index b50240b1d5a4..6c16ba1107ba 100644 --- a/drivers/infiniband/hw/qib/qib_init.c +++ b/drivers/infiniband/hw/qib/qib_init.c @@ -233,9 +233,8 @@ int qib_init_pportdata(struct qib_pportdata *ppd, struct qib_devdata *dd, spin_lock_init(&ppd->cc_shadow_lock); init_waitqueue_head(&ppd->state_wait); - init_timer(&ppd->symerr_clear_timer); - ppd->symerr_clear_timer.function = qib_clear_symerror_on_linkup; - ppd->symerr_clear_timer.data = (unsigned long)ppd; + setup_timer(&ppd->symerr_clear_timer, qib_clear_symerror_on_linkup, + (unsigned long)ppd); ppd->qib_wq = NULL; ppd->ibport_data.pmastats = @@ -429,9 +428,8 @@ static int loadtime_init(struct qib_devdata *dd) qib_get_eeprom_info(dd); /* setup time (don't start yet) to verify we got interrupt */ - init_timer(&dd->intrchk_timer); - dd->intrchk_timer.function = verify_interrupt; - dd->intrchk_timer.data = (unsigned long) dd; + setup_timer(&dd->intrchk_timer, verify_interrupt, + (unsigned long)dd); done: return ret; } @@ -755,9 +753,8 @@ done: continue; if (dd->flags & QIB_HAS_SEND_DMA) ret = qib_setup_sdma(ppd); - init_timer(&ppd->hol_timer); - ppd->hol_timer.function = qib_hol_event; - ppd->hol_timer.data = (unsigned long)ppd; + setup_timer(&ppd->hol_timer, qib_hol_event, + (unsigned long)ppd); ppd->hol_state = QIB_HOL_UP; } diff --git a/drivers/infiniband/hw/qib/qib_rc.c b/drivers/infiniband/hw/qib/qib_rc.c index 12658e3fe154..023498745b8a 100644 --- a/drivers/infiniband/hw/qib/qib_rc.c +++ b/drivers/infiniband/hw/qib/qib_rc.c @@ -938,7 +938,10 @@ void qib_rc_send_complete(struct rvt_qp *qp, struct ib_header *hdr) /* see post_send() */ barrier(); rvt_put_swqe(wqe); - rvt_qp_swqe_complete(qp, wqe, IB_WC_SUCCESS); + rvt_qp_swqe_complete(qp, + wqe, + ib_qib_wc_opcode[wqe->wr.opcode], + IB_WC_SUCCESS); } /* * If we were waiting for sends to complete before resending, @@ -983,7 +986,10 @@ static struct rvt_swqe *do_rc_completion(struct rvt_qp *qp, qp->s_last = s_last; /* see post_send() */ barrier(); - rvt_qp_swqe_complete(qp, wqe, IB_WC_SUCCESS); + rvt_qp_swqe_complete(qp, + wqe, + ib_qib_wc_opcode[wqe->wr.opcode], + IB_WC_SUCCESS); } else this_cpu_inc(*ibp->rvp.rc_delayed_comp); diff --git a/drivers/infiniband/hw/qib/qib_ruc.c b/drivers/infiniband/hw/qib/qib_ruc.c index 17655cc3e6fe..6e1adf709483 100644 --- a/drivers/infiniband/hw/qib/qib_ruc.c +++ b/drivers/infiniband/hw/qib/qib_ruc.c @@ -769,7 +769,10 @@ void qib_send_complete(struct rvt_qp *qp, struct rvt_swqe *wqe, qp->ibqp.qp_type == IB_QPT_GSI) atomic_dec(&ibah_to_rvtah(wqe->ud_wr.ah)->refcount); - rvt_qp_swqe_complete(qp, wqe, status); + rvt_qp_swqe_complete(qp, + wqe, + ib_qib_wc_opcode[wqe->wr.opcode], + status); if (qp->s_acked == old_last) qp->s_acked = last; diff --git a/drivers/infiniband/hw/qib/qib_verbs.c b/drivers/infiniband/hw/qib/qib_verbs.c index 83f8b5f24381..68d449cdb794 100644 --- a/drivers/infiniband/hw/qib/qib_verbs.c +++ b/drivers/infiniband/hw/qib/qib_verbs.c @@ -114,6 +114,19 @@ module_param_named(disable_sma, ib_qib_disable_sma, uint, S_IWUSR | S_IRUGO); MODULE_PARM_DESC(disable_sma, "Disable the SMA"); /* + * Translate ib_wr_opcode into ib_wc_opcode. + */ +const enum ib_wc_opcode ib_qib_wc_opcode[] = { + [IB_WR_RDMA_WRITE] = IB_WC_RDMA_WRITE, + [IB_WR_RDMA_WRITE_WITH_IMM] = IB_WC_RDMA_WRITE, + [IB_WR_SEND] = IB_WC_SEND, + [IB_WR_SEND_WITH_IMM] = IB_WC_SEND, + [IB_WR_RDMA_READ] = IB_WC_RDMA_READ, + [IB_WR_ATOMIC_CMP_AND_SWP] = IB_WC_COMP_SWAP, + [IB_WR_ATOMIC_FETCH_AND_ADD] = IB_WC_FETCH_ADD +}; + +/* * System image GUID. */ __be64 ib_qib_sys_image_guid; @@ -343,7 +356,7 @@ void qib_ib_rcv(struct qib_ctxtdata *rcd, void *rhdr, void *data, u32 tlen) if (lnh != QIB_LRH_GRH) goto drop; - mcast = rvt_mcast_find(&ibp->rvp, &hdr->u.l.grh.dgid); + mcast = rvt_mcast_find(&ibp->rvp, &hdr->u.l.grh.dgid, lid); if (mcast == NULL) goto drop; this_cpu_inc(ibp->pmastats->n_multicast_rcv); diff --git a/drivers/infiniband/hw/usnic/usnic_common_util.h b/drivers/infiniband/hw/usnic/usnic_common_util.h index b54986de5f0c..ddd81294fa46 100644 --- a/drivers/infiniband/hw/usnic/usnic_common_util.h +++ b/drivers/infiniband/hw/usnic/usnic_common_util.h @@ -34,21 +34,7 @@ #ifndef USNIC_CMN_UTIL_H #define USNIC_CMN_UTIL_H -static inline void -usnic_mac_to_gid(const char *const mac, char *raw_gid) -{ - raw_gid[0] = 0xfe; - raw_gid[1] = 0x80; - memset(&raw_gid[2], 0, 6); - raw_gid[8] = mac[0]^2; - raw_gid[9] = mac[1]; - raw_gid[10] = mac[2]; - raw_gid[11] = 0xff; - raw_gid[12] = 0xfe; - raw_gid[13] = mac[3]; - raw_gid[14] = mac[4]; - raw_gid[15] = mac[5]; -} +#include <net/addrconf.h> static inline void usnic_mac_ip_to_gid(const char *const mac, const __be32 inaddr, char *raw_gid) @@ -57,27 +43,7 @@ usnic_mac_ip_to_gid(const char *const mac, const __be32 inaddr, char *raw_gid) raw_gid[1] = 0x80; memset(&raw_gid[2], 0, 2); memcpy(&raw_gid[4], &inaddr, 4); - raw_gid[8] = mac[0]^2; - raw_gid[9] = mac[1]; - raw_gid[10] = mac[2]; - raw_gid[11] = 0xff; - raw_gid[12] = 0xfe; - raw_gid[13] = mac[3]; - raw_gid[14] = mac[4]; - raw_gid[15] = mac[5]; -} - -static inline void -usnic_write_gid_if_id_from_mac(char *mac, char *raw_gid) -{ - raw_gid[8] = mac[0]^2; - raw_gid[9] = mac[1]; - raw_gid[10] = mac[2]; - raw_gid[11] = 0xff; - raw_gid[12] = 0xfe; - raw_gid[13] = mac[3]; - raw_gid[14] = mac[4]; - raw_gid[15] = mac[5]; + addrconf_addr_eui48(&raw_gid[8], mac); } #endif /* USNIC_COMMON_UTIL_H */ diff --git a/drivers/infiniband/hw/usnic/usnic_ib_sysfs.c b/drivers/infiniband/hw/usnic/usnic_ib_sysfs.c index 04443242e258..32956f9f5715 100644 --- a/drivers/infiniband/hw/usnic/usnic_ib_sysfs.c +++ b/drivers/infiniband/hw/usnic/usnic_ib_sysfs.c @@ -44,6 +44,7 @@ #include "usnic_vnic.h" #include "usnic_ib_verbs.h" #include "usnic_log.h" +#include "usnic_ib_sysfs.h" static ssize_t usnic_ib_show_board(struct device *device, struct device_attribute *attr, diff --git a/drivers/infiniband/hw/usnic/usnic_ib_verbs.c b/drivers/infiniband/hw/usnic/usnic_ib_verbs.c index 3284730d3c09..44a93326f7b5 100644 --- a/drivers/infiniband/hw/usnic/usnic_ib_verbs.c +++ b/drivers/infiniband/hw/usnic/usnic_ib_verbs.c @@ -46,6 +46,7 @@ #include "usnic_log.h" #include "usnic_uiom.h" #include "usnic_transport.h" +#include "usnic_ib_verbs.h" #define USNIC_DEFAULT_TRANSPORT USNIC_TRANSPORT_ROCE_CUSTOM @@ -151,7 +152,7 @@ find_free_vf_and_create_qp_grp(struct usnic_ib_dev *us_ibdev, struct usnic_vnic *vnic; struct usnic_ib_qp_grp *qp_grp; struct device *dev, **dev_list; - int i, found = 0; + int i; BUG_ON(!mutex_is_locked(&us_ibdev->usdev_lock)); @@ -173,8 +174,13 @@ find_free_vf_and_create_qp_grp(struct usnic_ib_dev *us_ibdev, us_ibdev->ib_dev.name, pci_name(usnic_vnic_get_pdev( vnic))); - found = 1; - break; + qp_grp = usnic_ib_qp_grp_create(us_ibdev->ufdev, + vf, pd, + res_spec, + trans_spec); + + spin_unlock(&vf->lock); + goto qp_grp_check; } spin_unlock(&vf->lock); @@ -182,34 +188,30 @@ find_free_vf_and_create_qp_grp(struct usnic_ib_dev *us_ibdev, usnic_uiom_free_dev_list(dev_list); } - if (!found) { - /* Try to find resources on an unused vf */ - list_for_each_entry(vf, &us_ibdev->vf_dev_list, link) { - spin_lock(&vf->lock); - vnic = vf->vnic; - if (vf->qp_grp_ref_cnt == 0 && - usnic_vnic_check_room(vnic, res_spec) == 0) { - found = 1; - break; - } + /* Try to find resources on an unused vf */ + list_for_each_entry(vf, &us_ibdev->vf_dev_list, link) { + spin_lock(&vf->lock); + vnic = vf->vnic; + if (vf->qp_grp_ref_cnt == 0 && + usnic_vnic_check_room(vnic, res_spec) == 0) { + qp_grp = usnic_ib_qp_grp_create(us_ibdev->ufdev, vf, + pd, res_spec, + trans_spec); + spin_unlock(&vf->lock); + goto qp_grp_check; } + spin_unlock(&vf->lock); } - if (!found) { - usnic_info("No free qp grp found on %s\n", - us_ibdev->ib_dev.name); - return ERR_PTR(-ENOMEM); - } + usnic_info("No free qp grp found on %s\n", us_ibdev->ib_dev.name); + return ERR_PTR(-ENOMEM); - qp_grp = usnic_ib_qp_grp_create(us_ibdev->ufdev, vf, pd, res_spec, - trans_spec); - spin_unlock(&vf->lock); +qp_grp_check: if (IS_ERR_OR_NULL(qp_grp)) { usnic_err("Failed to allocate qp_grp\n"); return ERR_PTR(qp_grp ? PTR_ERR(qp_grp) : -ENOMEM); } - return qp_grp; } diff --git a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_misc.c b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_misc.c index 948b5ccd2a70..6ef4df6c8c4a 100644 --- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_misc.c +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_misc.c @@ -194,7 +194,7 @@ int pvrdma_page_dir_insert_umem(struct pvrdma_page_dir *pdir, len = sg_dma_len(sg) >> PAGE_SHIFT; for (j = 0; j < len; j++) { dma_addr_t addr = sg_dma_address(sg) + - umem->page_size * j; + (j << umem->page_shift); ret = pvrdma_page_dir_insert_dma(pdir, i, addr); if (ret) diff --git a/drivers/infiniband/sw/rdmavt/ah.c b/drivers/infiniband/sw/rdmavt/ah.c index 16c446142c2a..b0f09fb45c72 100644 --- a/drivers/infiniband/sw/rdmavt/ah.c +++ b/drivers/infiniband/sw/rdmavt/ah.c @@ -119,7 +119,7 @@ struct ib_ah *rvt_create_ah(struct ib_pd *pd, spin_lock_irqsave(&dev->n_ahs_lock, flags); if (dev->n_ahs_allocated == dev->dparms.props.max_ah) { - spin_unlock(&dev->n_ahs_lock); + spin_unlock_irqrestore(&dev->n_ahs_lock, flags); kfree(ah); return ERR_PTR(-ENOMEM); } diff --git a/drivers/infiniband/sw/rdmavt/cq.c b/drivers/infiniband/sw/rdmavt/cq.c index 7aa7a4e312f1..0ae2ff8cf81e 100644 --- a/drivers/infiniband/sw/rdmavt/cq.c +++ b/drivers/infiniband/sw/rdmavt/cq.c @@ -50,6 +50,7 @@ #include <linux/kthread.h> #include "cq.h" #include "vt.h" +#include "trace.h" /** * rvt_cq_enter - add a new entry to the completion queue @@ -93,6 +94,7 @@ void rvt_cq_enter(struct rvt_cq *cq, struct ib_wc *entry, bool solicited) } return; } + trace_rvt_cq_enter(cq, entry, head); if (cq->ip) { wc->uqueue[head].wr_id = entry->wr_id; wc->uqueue[head].status = entry->status; @@ -482,6 +484,7 @@ int rvt_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *entry) if (tail == wc->head) break; /* The kernel doesn't need a RMB since it has the lock. */ + trace_rvt_cq_poll(cq, &wc->kqueue[tail], npolled); *entry = wc->kqueue[tail]; if (tail >= cq->ibcq.cqe) tail = 0; diff --git a/drivers/infiniband/sw/rdmavt/mcast.c b/drivers/infiniband/sw/rdmavt/mcast.c index 05c8c2afb0e3..1f12b69a0d07 100644 --- a/drivers/infiniband/sw/rdmavt/mcast.c +++ b/drivers/infiniband/sw/rdmavt/mcast.c @@ -100,10 +100,11 @@ static void rvt_mcast_qp_free(struct rvt_mcast_qp *mqp) /** * mcast_alloc - allocate the multicast GID structure * @mgid: the multicast GID + * @lid: the muilticast LID (host order) * * A list of QPs will be attached to this structure. */ -static struct rvt_mcast *rvt_mcast_alloc(union ib_gid *mgid) +static struct rvt_mcast *rvt_mcast_alloc(union ib_gid *mgid, u16 lid) { struct rvt_mcast *mcast; @@ -111,7 +112,9 @@ static struct rvt_mcast *rvt_mcast_alloc(union ib_gid *mgid) if (!mcast) goto bail; - mcast->mgid = *mgid; + mcast->mcast_addr.mgid = *mgid; + mcast->mcast_addr.lid = lid; + INIT_LIST_HEAD(&mcast->qp_list); init_waitqueue_head(&mcast->wait); atomic_set(&mcast->refcount, 0); @@ -131,15 +134,19 @@ static void rvt_mcast_free(struct rvt_mcast *mcast) } /** - * rvt_mcast_find - search the global table for the given multicast GID + * rvt_mcast_find - search the global table for the given multicast GID/LID + * NOTE: It is valid to have 1 MLID with multiple MGIDs. It is not valid + * to have 1 MGID with multiple MLIDs. * @ibp: the IB port structure * @mgid: the multicast GID to search for + * @lid: the multicast LID portion of the multicast address (host order) * * The caller is responsible for decrementing the reference count if found. * * Return: NULL if not found. */ -struct rvt_mcast *rvt_mcast_find(struct rvt_ibport *ibp, union ib_gid *mgid) +struct rvt_mcast *rvt_mcast_find(struct rvt_ibport *ibp, union ib_gid *mgid, + u16 lid) { struct rb_node *n; unsigned long flags; @@ -153,15 +160,18 @@ struct rvt_mcast *rvt_mcast_find(struct rvt_ibport *ibp, union ib_gid *mgid) mcast = rb_entry(n, struct rvt_mcast, rb_node); - ret = memcmp(mgid->raw, mcast->mgid.raw, - sizeof(union ib_gid)); + ret = memcmp(mgid->raw, mcast->mcast_addr.mgid.raw, + sizeof(*mgid)); if (ret < 0) { n = n->rb_left; } else if (ret > 0) { n = n->rb_right; } else { - atomic_inc(&mcast->refcount); - found = mcast; + /* MGID/MLID must match */ + if (mcast->mcast_addr.lid == lid) { + atomic_inc(&mcast->refcount); + found = mcast; + } break; } } @@ -177,7 +187,8 @@ EXPORT_SYMBOL(rvt_mcast_find); * * Return: zero if both were added. Return EEXIST if the GID was already in * the table but the QP was added. Return ESRCH if the QP was already - * attached and neither structure was added. + * attached and neither structure was added. Return EINVAL if the MGID was + * found, but the MLID did NOT match. */ static int rvt_mcast_add(struct rvt_dev_info *rdi, struct rvt_ibport *ibp, struct rvt_mcast *mcast, struct rvt_mcast_qp *mqp) @@ -195,8 +206,9 @@ static int rvt_mcast_add(struct rvt_dev_info *rdi, struct rvt_ibport *ibp, pn = *n; tmcast = rb_entry(pn, struct rvt_mcast, rb_node); - ret = memcmp(mcast->mgid.raw, tmcast->mgid.raw, - sizeof(union ib_gid)); + ret = memcmp(mcast->mcast_addr.mgid.raw, + tmcast->mcast_addr.mgid.raw, + sizeof(mcast->mcast_addr.mgid)); if (ret < 0) { n = &pn->rb_left; continue; @@ -206,6 +218,11 @@ static int rvt_mcast_add(struct rvt_dev_info *rdi, struct rvt_ibport *ibp, continue; } + if (tmcast->mcast_addr.lid != mcast->mcast_addr.lid) { + ret = EINVAL; + goto bail; + } + /* Search the QP list to see if this is already there. */ list_for_each_entry_rcu(p, &tmcast->qp_list, list) { if (p->qp == mqp->qp) { @@ -276,7 +293,7 @@ int rvt_attach_mcast(struct ib_qp *ibqp, union ib_gid *gid, u16 lid) * Allocate data structures since its better to do this outside of * spin locks and it will most likely be needed. */ - mcast = rvt_mcast_alloc(gid); + mcast = rvt_mcast_alloc(gid, lid); if (!mcast) return -ENOMEM; @@ -296,6 +313,10 @@ int rvt_attach_mcast(struct ib_qp *ibqp, union ib_gid *gid, u16 lid) /* Exceeded the maximum number of mcast groups. */ ret = -ENOMEM; goto bail_mqp; + case EINVAL: + /* Invalid MGID/MLID pair */ + ret = -EINVAL; + goto bail_mqp; default: break; } @@ -344,14 +365,20 @@ int rvt_detach_mcast(struct ib_qp *ibqp, union ib_gid *gid, u16 lid) } mcast = rb_entry(n, struct rvt_mcast, rb_node); - ret = memcmp(gid->raw, mcast->mgid.raw, - sizeof(union ib_gid)); - if (ret < 0) + ret = memcmp(gid->raw, mcast->mcast_addr.mgid.raw, + sizeof(*gid)); + if (ret < 0) { n = n->rb_left; - else if (ret > 0) + } else if (ret > 0) { n = n->rb_right; - else + } else { + /* MGID/MLID must match */ + if (mcast->mcast_addr.lid != lid) { + spin_unlock_irq(&ibp->lock); + return -EINVAL; + } break; + } } /* Search the QP list. */ diff --git a/drivers/infiniband/sw/rdmavt/mr.c b/drivers/infiniband/sw/rdmavt/mr.c index ae30b6838d79..aa5f9ea318e4 100644 --- a/drivers/infiniband/sw/rdmavt/mr.c +++ b/drivers/infiniband/sw/rdmavt/mr.c @@ -191,8 +191,9 @@ static int rvt_alloc_lkey(struct rvt_mregion *mr, int dma_region) tmr = rcu_access_pointer(dev->dma_mr); if (!tmr) { - rcu_assign_pointer(dev->dma_mr, mr); mr->lkey_published = 1; + /* Insure published written first */ + rcu_assign_pointer(dev->dma_mr, mr); rvt_get_mr(mr); } goto success; @@ -224,8 +225,9 @@ static int rvt_alloc_lkey(struct rvt_mregion *mr, int dma_region) mr->lkey |= 1 << 8; rkt->gen++; } - rcu_assign_pointer(rkt->table[r], mr); mr->lkey_published = 1; + /* Insure published written first */ + rcu_assign_pointer(rkt->table[r], mr); success: spin_unlock_irqrestore(&rkt->lock, flags); out: @@ -253,23 +255,24 @@ static void rvt_free_lkey(struct rvt_mregion *mr) spin_lock_irqsave(&rkt->lock, flags); if (!lkey) { if (mr->lkey_published) { - RCU_INIT_POINTER(dev->dma_mr, NULL); + mr->lkey_published = 0; + /* insure published is written before pointer */ + rcu_assign_pointer(dev->dma_mr, NULL); rvt_put_mr(mr); } } else { if (!mr->lkey_published) goto out; r = lkey >> (32 - dev->dparms.lkey_table_size); - RCU_INIT_POINTER(rkt->table[r], NULL); + mr->lkey_published = 0; + /* insure published is written before pointer */ + rcu_assign_pointer(rkt->table[r], NULL); } - mr->lkey_published = 0; freed++; out: spin_unlock_irqrestore(&rkt->lock, flags); - if (freed) { - synchronize_rcu(); + if (freed) percpu_ref_kill(&mr->refcount); - } } static struct rvt_mr *__rvt_alloc_mr(int count, struct ib_pd *pd) @@ -405,8 +408,7 @@ struct ib_mr *rvt_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, mr->mr.access_flags = mr_access_flags; mr->umem = umem; - if (is_power_of_2(umem->page_size)) - mr->mr.page_shift = ilog2(umem->page_size); + mr->mr.page_shift = umem->page_shift; m = 0; n = 0; for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) { @@ -418,8 +420,9 @@ struct ib_mr *rvt_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, goto bail_inval; } mr->mr.map[m]->segs[n].vaddr = vaddr; - mr->mr.map[m]->segs[n].length = umem->page_size; - trace_rvt_mr_user_seg(&mr->mr, m, n, vaddr, umem->page_size); + mr->mr.map[m]->segs[n].length = BIT(umem->page_shift); + trace_rvt_mr_user_seg(&mr->mr, m, n, vaddr, + BIT(umem->page_shift)); n++; if (n == RVT_SEGSZ) { m++; @@ -822,16 +825,21 @@ int rvt_lkey_ok(struct rvt_lkey_table *rkt, struct rvt_pd *pd, goto ok; } mr = rcu_dereference(rkt->table[sge->lkey >> rkt->shift]); - if (unlikely(!mr || atomic_read(&mr->lkey_invalid) || - mr->lkey != sge->lkey || mr->pd != &pd->ibpd)) + if (!mr) goto bail; + rvt_get_mr(mr); + if (!READ_ONCE(mr->lkey_published)) + goto bail_unref; + + if (unlikely(atomic_read(&mr->lkey_invalid) || + mr->lkey != sge->lkey || mr->pd != &pd->ibpd)) + goto bail_unref; off = sge->addr - mr->user_base; if (unlikely(sge->addr < mr->user_base || off + sge->length > mr->length || (mr->access_flags & acc) != acc)) - goto bail; - rvt_get_mr(mr); + goto bail_unref; rcu_read_unlock(); off += mr->offset; @@ -867,6 +875,8 @@ int rvt_lkey_ok(struct rvt_lkey_table *rkt, struct rvt_pd *pd, isge->n = n; ok: return 1; +bail_unref: + rvt_put_mr(mr); bail: rcu_read_unlock(); return 0; @@ -922,15 +932,20 @@ int rvt_rkey_ok(struct rvt_qp *qp, struct rvt_sge *sge, } mr = rcu_dereference(rkt->table[rkey >> rkt->shift]); - if (unlikely(!mr || atomic_read(&mr->lkey_invalid) || - mr->lkey != rkey || qp->ibqp.pd != mr->pd)) + if (!mr) goto bail; + rvt_get_mr(mr); + /* insure mr read is before test */ + if (!READ_ONCE(mr->lkey_published)) + goto bail_unref; + if (unlikely(atomic_read(&mr->lkey_invalid) || + mr->lkey != rkey || qp->ibqp.pd != mr->pd)) + goto bail_unref; off = vaddr - mr->iova; if (unlikely(vaddr < mr->iova || off + len > mr->length || (mr->access_flags & acc) == 0)) - goto bail; - rvt_get_mr(mr); + goto bail_unref; rcu_read_unlock(); off += mr->offset; @@ -966,6 +981,8 @@ int rvt_rkey_ok(struct rvt_qp *qp, struct rvt_sge *sge, sge->n = n; ok: return 1; +bail_unref: + rvt_put_mr(mr); bail: rcu_read_unlock(); return 0; diff --git a/drivers/infiniband/sw/rdmavt/qp.c b/drivers/infiniband/sw/rdmavt/qp.c index f5ad8d4bfb39..728f5f1218c8 100644 --- a/drivers/infiniband/sw/rdmavt/qp.c +++ b/drivers/infiniband/sw/rdmavt/qp.c @@ -1,5 +1,5 @@ /* - * Copyright(c) 2016 Intel Corporation. + * Copyright(c) 2016, 2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -117,23 +117,6 @@ const int ib_rvt_state_ops[IB_QPS_ERR + 1] = { }; EXPORT_SYMBOL(ib_rvt_state_ops); -/* - * Translate ib_wr_opcode into ib_wc_opcode. - */ -const enum ib_wc_opcode ib_rvt_wc_opcode[] = { - [IB_WR_RDMA_WRITE] = IB_WC_RDMA_WRITE, - [IB_WR_RDMA_WRITE_WITH_IMM] = IB_WC_RDMA_WRITE, - [IB_WR_SEND] = IB_WC_SEND, - [IB_WR_SEND_WITH_IMM] = IB_WC_SEND, - [IB_WR_RDMA_READ] = IB_WC_RDMA_READ, - [IB_WR_ATOMIC_CMP_AND_SWP] = IB_WC_COMP_SWAP, - [IB_WR_ATOMIC_FETCH_AND_ADD] = IB_WC_FETCH_ADD, - [IB_WR_SEND_WITH_INV] = IB_WC_SEND, - [IB_WR_LOCAL_INV] = IB_WC_LOCAL_INV, - [IB_WR_REG_MR] = IB_WC_REG_MR -}; -EXPORT_SYMBOL(ib_rvt_wc_opcode); - static void get_map_page(struct rvt_qpn_table *qpt, struct rvt_qpn_map *map, gfp_t gfp) @@ -1789,11 +1772,14 @@ static int rvt_post_one_wr(struct rvt_qp *qp, 0); qp->s_next_psn = wqe->lpsn + 1; } - trace_rvt_post_one_wr(qp, wqe); - if (unlikely(reserved_op)) + if (unlikely(reserved_op)) { + wqe->wr.send_flags |= RVT_SEND_RESERVE_USED; rvt_qp_wqe_reserve(qp, wqe); - else + } else { + wqe->wr.send_flags &= ~RVT_SEND_RESERVE_USED; qp->s_avail--; + } + trace_rvt_post_one_wr(qp, wqe); smp_wmb(); /* see request builders */ qp->s_head = next; @@ -2069,8 +2055,12 @@ static void rvt_rc_timeout(unsigned long arg) spin_lock_irqsave(&qp->r_lock, flags); spin_lock(&qp->s_lock); if (qp->s_flags & RVT_S_TIMER) { + struct rvt_ibport *rvp = rdi->ports[qp->port_num - 1]; + qp->s_flags &= ~RVT_S_TIMER; + rvp->n_rc_timeouts++; del_timer(&qp->s_timer); + trace_rvt_rc_timeout(qp, qp->s_last_psn + 1); if (rdi->driver_f.notify_restart_rc) rdi->driver_f.notify_restart_rc(qp, qp->s_last_psn + 1, diff --git a/drivers/infiniband/sw/rdmavt/trace.h b/drivers/infiniband/sw/rdmavt/trace.h index e2d23acb6a7d..bb4b1e710f22 100644 --- a/drivers/infiniband/sw/rdmavt/trace.h +++ b/drivers/infiniband/sw/rdmavt/trace.h @@ -1,5 +1,5 @@ /* - * Copyright(c) 2016 Intel Corporation. + * Copyright(c) 2016, 2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -52,3 +52,5 @@ #include "trace_qp.h" #include "trace_tx.h" #include "trace_mr.h" +#include "trace_cq.h" +#include "trace_rc.h" diff --git a/drivers/infiniband/sw/rdmavt/trace_cq.h b/drivers/infiniband/sw/rdmavt/trace_cq.h new file mode 100644 index 000000000000..a315850aa9bb --- /dev/null +++ b/drivers/infiniband/sw/rdmavt/trace_cq.h @@ -0,0 +1,127 @@ +/* + * Copyright(c) 2016 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ +#if !defined(__RVT_TRACE_CQ_H) || defined(TRACE_HEADER_MULTI_READ) +#define __RVT_TRACE_CQ_H + +#include <linux/tracepoint.h> +#include <linux/trace_seq.h> + +#include <rdma/ib_verbs.h> +#include <rdma/rdmavt_cq.h> + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM rvt_cq + +#define wc_opcode_name(opcode) { IB_WC_##opcode, #opcode } +#define show_wc_opcode(opcode) \ +__print_symbolic(opcode, \ + wc_opcode_name(SEND), \ + wc_opcode_name(RDMA_WRITE), \ + wc_opcode_name(RDMA_READ), \ + wc_opcode_name(COMP_SWAP), \ + wc_opcode_name(FETCH_ADD), \ + wc_opcode_name(LSO), \ + wc_opcode_name(LOCAL_INV), \ + wc_opcode_name(REG_MR), \ + wc_opcode_name(MASKED_COMP_SWAP), \ + wc_opcode_name(RECV), \ + wc_opcode_name(RECV_RDMA_WITH_IMM)) + +#define CQ_PRN \ +"[%s] idx %u wr_id %llx status %u opcode %u,%s length %u qpn %x" + +DECLARE_EVENT_CLASS( + rvt_cq_entry_template, + TP_PROTO(struct rvt_cq *cq, struct ib_wc *wc, u32 idx), + TP_ARGS(cq, wc, idx), + TP_STRUCT__entry( + RDI_DEV_ENTRY(cq->rdi) + __field(u64, wr_id) + __field(u32, status) + __field(u32, opcode) + __field(u32, qpn) + __field(u32, length) + __field(u32, idx) + ), + TP_fast_assign( + RDI_DEV_ASSIGN(cq->rdi) + __entry->wr_id = wc->wr_id; + __entry->status = wc->status; + __entry->opcode = wc->opcode; + __entry->length = wc->byte_len; + __entry->qpn = wc->qp->qp_num; + __entry->idx = idx; + ), + TP_printk( + CQ_PRN, + __get_str(dev), + __entry->idx, + __entry->wr_id, + __entry->status, + __entry->opcode, show_wc_opcode(__entry->opcode), + __entry->length, + __entry->qpn + ) +); + +DEFINE_EVENT( + rvt_cq_entry_template, rvt_cq_enter, + TP_PROTO(struct rvt_cq *cq, struct ib_wc *wc, u32 idx), + TP_ARGS(cq, wc, idx)); + +DEFINE_EVENT( + rvt_cq_entry_template, rvt_cq_poll, + TP_PROTO(struct rvt_cq *cq, struct ib_wc *wc, u32 idx), + TP_ARGS(cq, wc, idx)); + +#endif /* __RVT_TRACE_CQ_H */ + +#undef TRACE_INCLUDE_PATH +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_PATH . +#define TRACE_INCLUDE_FILE trace_cq +#include <trace/define_trace.h> diff --git a/drivers/infiniband/sw/rdmavt/trace_rc.h b/drivers/infiniband/sw/rdmavt/trace_rc.h new file mode 100644 index 000000000000..995276933a55 --- /dev/null +++ b/drivers/infiniband/sw/rdmavt/trace_rc.h @@ -0,0 +1,109 @@ +/* + * Copyright(c) 2017 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ +#if !defined(__RVT_TRACE_RC_H) || defined(TRACE_HEADER_MULTI_READ) +#define __RVT_TRACE_RC_H + +#include <linux/tracepoint.h> +#include <linux/trace_seq.h> + +#include <rdma/ib_verbs.h> +#include <rdma/rdma_vt.h> + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM rvt_rc + +DECLARE_EVENT_CLASS(rvt_rc_template, + TP_PROTO(struct rvt_qp *qp, u32 psn), + TP_ARGS(qp, psn), + TP_STRUCT__entry( + RDI_DEV_ENTRY(ib_to_rvt(qp->ibqp.device)) + __field(u32, qpn) + __field(u32, s_flags) + __field(u32, psn) + __field(u32, s_psn) + __field(u32, s_next_psn) + __field(u32, s_sending_psn) + __field(u32, s_sending_hpsn) + __field(u32, r_psn) + ), + TP_fast_assign( + RDI_DEV_ASSIGN(ib_to_rvt(qp->ibqp.device)) + __entry->qpn = qp->ibqp.qp_num; + __entry->s_flags = qp->s_flags; + __entry->psn = psn; + __entry->s_psn = qp->s_psn; + __entry->s_next_psn = qp->s_next_psn; + __entry->s_sending_psn = qp->s_sending_psn; + __entry->s_sending_hpsn = qp->s_sending_hpsn; + __entry->r_psn = qp->r_psn; + ), + TP_printk( + "[%s] qpn 0x%x s_flags 0x%x psn 0x%x s_psn 0x%x s_next_psn 0x%x s_sending_psn 0x%x sending_hpsn 0x%x r_psn 0x%x", + __get_str(dev), + __entry->qpn, + __entry->s_flags, + __entry->psn, + __entry->s_psn, + __entry->s_next_psn, + __entry->s_sending_psn, + __entry->s_sending_hpsn, + __entry->r_psn + ) +); + +DEFINE_EVENT(rvt_rc_template, rvt_rc_timeout, + TP_PROTO(struct rvt_qp *qp, u32 psn), + TP_ARGS(qp, psn) +); + +#endif /* __RVT_TRACE_RC_H */ + +#undef TRACE_INCLUDE_PATH +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_PATH . +#define TRACE_INCLUDE_FILE trace_rc +#include <trace/define_trace.h> diff --git a/drivers/infiniband/sw/rdmavt/trace_tx.h b/drivers/infiniband/sw/rdmavt/trace_tx.h index 0e03173662d8..a613a2223751 100644 --- a/drivers/infiniband/sw/rdmavt/trace_tx.h +++ b/drivers/infiniband/sw/rdmavt/trace_tx.h @@ -71,10 +71,20 @@ __print_symbolic(opcode, \ wr_opcode_name(RDMA_READ_WITH_INV), \ wr_opcode_name(LOCAL_INV), \ wr_opcode_name(MASKED_ATOMIC_CMP_AND_SWP), \ - wr_opcode_name(MASKED_ATOMIC_FETCH_AND_ADD)) + wr_opcode_name(MASKED_ATOMIC_FETCH_AND_ADD), \ + wr_opcode_name(RESERVED1), \ + wr_opcode_name(RESERVED2), \ + wr_opcode_name(RESERVED3), \ + wr_opcode_name(RESERVED4), \ + wr_opcode_name(RESERVED5), \ + wr_opcode_name(RESERVED6), \ + wr_opcode_name(RESERVED7), \ + wr_opcode_name(RESERVED8), \ + wr_opcode_name(RESERVED9), \ + wr_opcode_name(RESERVED10)) #define POS_PRN \ -"[%s] wr_id %llx qpn %x psn 0x%x lpsn 0x%x length %u opcode 0x%.2x,%s size %u avail %u head %u last %u" +"[%s] wqe %p wr_id %llx send_flags %x qpn %x qpt %u psn %x lpsn %x ssn %x length %u opcode 0x%.2x,%s size %u avail %u head %u last %u pid %u num_sge %u" TRACE_EVENT( rvt_post_one_wr, @@ -83,7 +93,9 @@ TRACE_EVENT( TP_STRUCT__entry( RDI_DEV_ENTRY(ib_to_rvt(qp->ibqp.device)) __field(u64, wr_id) + __field(struct rvt_swqe *, wqe) __field(u32, qpn) + __field(u32, qpt) __field(u32, psn) __field(u32, lpsn) __field(u32, length) @@ -92,11 +104,17 @@ TRACE_EVENT( __field(u32, avail) __field(u32, head) __field(u32, last) + __field(u32, ssn) + __field(int, send_flags) + __field(pid_t, pid) + __field(int, num_sge) ), TP_fast_assign( RDI_DEV_ASSIGN(ib_to_rvt(qp->ibqp.device)) + __entry->wqe = wqe; __entry->wr_id = wqe->wr.wr_id; __entry->qpn = qp->ibqp.qp_num; + __entry->qpt = qp->ibqp.qp_type; __entry->psn = wqe->psn; __entry->lpsn = wqe->lpsn; __entry->length = wqe->length; @@ -105,20 +123,30 @@ TRACE_EVENT( __entry->avail = qp->s_avail; __entry->head = qp->s_head; __entry->last = qp->s_last; + __entry->pid = qp->pid; + __entry->ssn = wqe->ssn; + __entry->send_flags = wqe->wr.send_flags; + __entry->num_sge = wqe->wr.num_sge; ), TP_printk( POS_PRN, __get_str(dev), + __entry->wqe, __entry->wr_id, + __entry->send_flags, __entry->qpn, + __entry->qpt, __entry->psn, __entry->lpsn, + __entry->ssn, __entry->length, __entry->opcode, show_wr_opcode(__entry->opcode), __entry->size, __entry->avail, __entry->head, - __entry->last + __entry->last, + __entry->pid, + __entry->num_sge ) ); diff --git a/drivers/infiniband/sw/rxe/Kconfig b/drivers/infiniband/sw/rxe/Kconfig index 6332dedc11e8..320bffc980d8 100644 --- a/drivers/infiniband/sw/rxe/Kconfig +++ b/drivers/infiniband/sw/rxe/Kconfig @@ -2,6 +2,7 @@ config RDMA_RXE tristate "Software RDMA over Ethernet (RoCE) driver" depends on INET && PCI && INFINIBAND depends on NET_UDP_TUNNEL + depends on CRYPTO_CRC32 select DMA_VIRT_OPS ---help--- This driver implements the InfiniBand RDMA transport over diff --git a/drivers/infiniband/sw/rxe/Makefile b/drivers/infiniband/sw/rxe/Makefile index ec35ff022a42..3f12beb7076f 100644 --- a/drivers/infiniband/sw/rxe/Makefile +++ b/drivers/infiniband/sw/rxe/Makefile @@ -20,4 +20,5 @@ rdma_rxe-y := \ rxe_mcast.o \ rxe_task.o \ rxe_net.o \ - rxe_sysfs.o + rxe_sysfs.o \ + rxe_hw_counters.o diff --git a/drivers/infiniband/sw/rxe/rxe.c b/drivers/infiniband/sw/rxe/rxe.c index b12dd9b5a89d..c21c913f911a 100644 --- a/drivers/infiniband/sw/rxe/rxe.c +++ b/drivers/infiniband/sw/rxe/rxe.c @@ -31,6 +31,7 @@ * SOFTWARE. */ +#include <net/addrconf.h> #include "rxe.h" #include "rxe_loc.h" @@ -64,6 +65,8 @@ static void rxe_cleanup(struct rxe_dev *rxe) rxe_pool_cleanup(&rxe->mc_elem_pool); rxe_cleanup_ports(rxe); + + crypto_free_shash(rxe->tfm); } /* called when all references have been dropped */ @@ -178,7 +181,8 @@ static int rxe_init_ports(struct rxe_dev *rxe) return -ENOMEM; port->pkey_tbl[0] = 0xffff; - port->port_guid = rxe_port_guid(rxe); + addrconf_addr_eui48((unsigned char *)&port->port_guid, + rxe->ndev->dev_addr); spin_lock_init(&port->port_lock); diff --git a/drivers/infiniband/sw/rxe/rxe.h b/drivers/infiniband/sw/rxe/rxe.h index a696af81e4a5..ecdba2fce083 100644 --- a/drivers/infiniband/sw/rxe/rxe.h +++ b/drivers/infiniband/sw/rxe/rxe.h @@ -50,6 +50,7 @@ #include <rdma/ib_umem.h> #include <rdma/ib_cache.h> #include <rdma/ib_addr.h> +#include <crypto/hash.h> #include "rxe_net.h" #include "rxe_opcode.h" @@ -64,6 +65,25 @@ #define RXE_ROCE_V2_SPORT (0xc000) +static inline u32 rxe_crc32(struct rxe_dev *rxe, + u32 crc, void *next, size_t len) +{ + int err; + + SHASH_DESC_ON_STACK(shash, rxe->tfm); + + shash->tfm = rxe->tfm; + shash->flags = 0; + *(u32 *)shash_desc_ctx(shash) = crc; + err = crypto_shash_update(shash, next, len); + if (unlikely(err)) { + pr_warn_ratelimited("failed crc calculation, err: %d\n", err); + return crc32_le(crc, next, len); + } + + return *(u32 *)shash_desc_ctx(shash); +} + int rxe_set_mtu(struct rxe_dev *rxe, unsigned int dev_mtu); int rxe_add(struct rxe_dev *rxe, unsigned int mtu); diff --git a/drivers/infiniband/sw/rxe/rxe_comp.c b/drivers/infiniband/sw/rxe/rxe_comp.c index 4cd55d5617f7..9eb12c2e3c74 100644 --- a/drivers/infiniband/sw/rxe/rxe_comp.c +++ b/drivers/infiniband/sw/rxe/rxe_comp.c @@ -154,6 +154,8 @@ void rxe_comp_queue_pkt(struct rxe_dev *rxe, struct rxe_qp *qp, skb_queue_tail(&qp->resp_pkts, skb); must_sched = skb_queue_len(&qp->resp_pkts) > 1; + if (must_sched != 0) + rxe_counter_inc(rxe, RXE_CNT_COMPLETER_SCHED); rxe_run_task(&qp->comp.task, must_sched); } @@ -236,6 +238,7 @@ static inline enum comp_state check_ack(struct rxe_qp *qp, { unsigned int mask = pkt->mask; u8 syn; + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); /* Check the sequence only */ switch (qp->comp.opcode) { @@ -298,6 +301,7 @@ static inline enum comp_state check_ack(struct rxe_qp *qp, return COMPST_WRITE_SEND; case AETH_RNR_NAK: + rxe_counter_inc(rxe, RXE_CNT_RCV_RNR); return COMPST_RNR_RETRY; case AETH_NAK: @@ -307,6 +311,8 @@ static inline enum comp_state check_ack(struct rxe_qp *qp, * before */ if (psn_compare(pkt->psn, qp->comp.psn) > 0) { + rxe_counter_inc(rxe, + RXE_CNT_RCV_SEQ_ERR); qp->comp.psn = pkt->psn; if (qp->req.wait_psn) { qp->req.wait_psn = 0; @@ -534,6 +540,7 @@ static void rxe_drain_resp_pkts(struct rxe_qp *qp, bool notify) int rxe_completer(void *arg) { struct rxe_qp *qp = (struct rxe_qp *)arg; + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); struct rxe_send_wqe *wqe = wqe; struct sk_buff *skb = NULL; struct rxe_pkt_info *pkt = NULL; @@ -683,8 +690,10 @@ int rxe_completer(void *arg) if (psn_compare(qp->req.psn, qp->comp.psn) > 0) { /* tell the requester to retry the - * send send queue next time around + * send queue next time around */ + rxe_counter_inc(rxe, + RXE_CNT_COMP_RETRY); qp->req.need_retry = 1; rxe_run_task(&qp->req.task, 1); } @@ -699,6 +708,7 @@ int rxe_completer(void *arg) goto exit; } else { + rxe_counter_inc(rxe, RXE_CNT_RETRY_EXCEEDED); wqe->status = IB_WC_RETRY_EXC_ERR; state = COMPST_ERROR; } @@ -720,6 +730,8 @@ int rxe_completer(void *arg) skb = NULL; goto exit; } else { + rxe_counter_inc(rxe, + RXE_CNT_RNR_RETRY_EXCEEDED); wqe->status = IB_WC_RNR_RETRY_EXC_ERR; state = COMPST_ERROR; } diff --git a/drivers/infiniband/sw/rxe/rxe_hw_counters.c b/drivers/infiniband/sw/rxe/rxe_hw_counters.c new file mode 100644 index 000000000000..7ef90aad7dfd --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_hw_counters.c @@ -0,0 +1,78 @@ +/* + * Copyright (c) 2017 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. + */ + +#include "rxe.h" +#include "rxe_hw_counters.h" + +const char * const rxe_counter_name[] = { + [RXE_CNT_SENT_PKTS] = "sent_pkts", + [RXE_CNT_RCVD_PKTS] = "rcvd_pkts", + [RXE_CNT_DUP_REQ] = "duplicate_request", + [RXE_CNT_OUT_OF_SEQ_REQ] = "out_of_sequence", + [RXE_CNT_RCV_RNR] = "rcvd_rnr_err", + [RXE_CNT_SND_RNR] = "send_rnr_err", + [RXE_CNT_RCV_SEQ_ERR] = "rcvd_seq_err", + [RXE_CNT_COMPLETER_SCHED] = "ack_deffered", + [RXE_CNT_RETRY_EXCEEDED] = "retry_exceeded_err", + [RXE_CNT_RNR_RETRY_EXCEEDED] = "retry_rnr_exceeded_err", + [RXE_CNT_COMP_RETRY] = "completer_retry_err", + [RXE_CNT_SEND_ERR] = "send_err", +}; + +int rxe_ib_get_hw_stats(struct ib_device *ibdev, + struct rdma_hw_stats *stats, + u8 port, int index) +{ + struct rxe_dev *dev = to_rdev(ibdev); + unsigned int cnt; + + if (!port || !stats) + return -EINVAL; + + for (cnt = 0; cnt < ARRAY_SIZE(rxe_counter_name); cnt++) + stats->value[cnt] = dev->stats_counters[cnt]; + + return ARRAY_SIZE(rxe_counter_name); +} + +struct rdma_hw_stats *rxe_ib_alloc_hw_stats(struct ib_device *ibdev, + u8 port_num) +{ + BUILD_BUG_ON(ARRAY_SIZE(rxe_counter_name) != RXE_NUM_OF_COUNTERS); + /* We support only per port stats */ + if (!port_num) + return NULL; + + return rdma_alloc_hw_stats_struct(rxe_counter_name, + ARRAY_SIZE(rxe_counter_name), + RDMA_HW_STATS_DEFAULT_LIFESPAN); +} diff --git a/drivers/infiniband/sw/rxe/rxe_hw_counters.h b/drivers/infiniband/sw/rxe/rxe_hw_counters.h new file mode 100644 index 000000000000..f44df1b76742 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_hw_counters.h @@ -0,0 +1,61 @@ +/* + * Copyright (c) 2017 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. + */ + +#ifndef RXE_HW_COUNTERS_H +#define RXE_HW_COUNTERS_H + +/* + * when adding counters to enum also add + * them to rxe_counter_name[] vector. + */ +enum rxe_counters { + RXE_CNT_SENT_PKTS, + RXE_CNT_RCVD_PKTS, + RXE_CNT_DUP_REQ, + RXE_CNT_OUT_OF_SEQ_REQ, + RXE_CNT_RCV_RNR, + RXE_CNT_SND_RNR, + RXE_CNT_RCV_SEQ_ERR, + RXE_CNT_COMPLETER_SCHED, + RXE_CNT_RETRY_EXCEEDED, + RXE_CNT_RNR_RETRY_EXCEEDED, + RXE_CNT_COMP_RETRY, + RXE_CNT_SEND_ERR, + RXE_NUM_OF_COUNTERS +}; + +struct rdma_hw_stats *rxe_ib_alloc_hw_stats(struct ib_device *ibdev, + u8 port_num); +int rxe_ib_get_hw_stats(struct ib_device *ibdev, + struct rdma_hw_stats *stats, + u8 port, int index); +#endif /* RXE_HW_COUNTERS_H */ diff --git a/drivers/infiniband/sw/rxe/rxe_icrc.c b/drivers/infiniband/sw/rxe/rxe_icrc.c index 413b56b23a06..39e0be31aab1 100644 --- a/drivers/infiniband/sw/rxe/rxe_icrc.c +++ b/drivers/infiniband/sw/rxe/rxe_icrc.c @@ -87,10 +87,10 @@ u32 rxe_icrc_hdr(struct rxe_pkt_info *pkt, struct sk_buff *skb) bth->qpn |= cpu_to_be32(~BTH_QPN_MASK); length = hdr_size + RXE_BTH_BYTES; - crc = crc32_le(crc, pshdr, length); + crc = rxe_crc32(pkt->rxe, crc, pshdr, length); /* And finish to compute the CRC on the remainder of the headers. */ - crc = crc32_le(crc, pkt->hdr + RXE_BTH_BYTES, - rxe_opcode[pkt->opcode].length - RXE_BTH_BYTES); + crc = rxe_crc32(pkt->rxe, crc, pkt->hdr + RXE_BTH_BYTES, + rxe_opcode[pkt->opcode].length - RXE_BTH_BYTES); return crc; } diff --git a/drivers/infiniband/sw/rxe/rxe_loc.h b/drivers/infiniband/sw/rxe/rxe_loc.h index 183a9d379b41..ed809138fd55 100644 --- a/drivers/infiniband/sw/rxe/rxe_loc.h +++ b/drivers/infiniband/sw/rxe/rxe_loc.h @@ -145,7 +145,6 @@ int advance_dma_data(struct rxe_dma_info *dma, unsigned int length); int rxe_loopback(struct sk_buff *skb); int rxe_send(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, struct sk_buff *skb); -__be64 rxe_port_guid(struct rxe_dev *rxe); struct sk_buff *rxe_init_packet(struct rxe_dev *rxe, struct rxe_av *av, int paylen, struct rxe_pkt_info *pkt); int rxe_prepare(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, @@ -153,7 +152,6 @@ int rxe_prepare(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, enum rdma_link_layer rxe_link_layer(struct rxe_dev *rxe, unsigned int port_num); const char *rxe_parent_name(struct rxe_dev *rxe, unsigned int port_num); struct device *rxe_dma_device(struct rxe_dev *rxe); -__be64 rxe_node_guid(struct rxe_dev *rxe); int rxe_mcast_add(struct rxe_dev *rxe, union ib_gid *mgid); int rxe_mcast_delete(struct rxe_dev *rxe, union ib_gid *mgid); @@ -278,6 +276,7 @@ static inline int rxe_xmit_packet(struct rxe_dev *rxe, struct rxe_qp *qp, if (err) { rxe->xmit_errors++; + rxe_counter_inc(rxe, RXE_CNT_SEND_ERR); return err; } @@ -287,6 +286,7 @@ static inline int rxe_xmit_packet(struct rxe_dev *rxe, struct rxe_qp *qp, rxe_run_task(&qp->comp.task, 1); } + rxe_counter_inc(rxe, RXE_CNT_SENT_PKTS); goto done; drop: diff --git a/drivers/infiniband/sw/rxe/rxe_mr.c b/drivers/infiniband/sw/rxe/rxe_mr.c index 37eea7441ca4..ced15c4446bd 100644 --- a/drivers/infiniband/sw/rxe/rxe_mr.c +++ b/drivers/infiniband/sw/rxe/rxe_mr.c @@ -191,10 +191,8 @@ int rxe_mem_init_user(struct rxe_dev *rxe, struct rxe_pd *pd, u64 start, goto err1; } - WARN_ON_ONCE(!is_power_of_2(umem->page_size)); - - mem->page_shift = ilog2(umem->page_size); - mem->page_mask = umem->page_size - 1; + mem->page_shift = umem->page_shift; + mem->page_mask = BIT(umem->page_shift) - 1; num_buf = 0; map = mem->map; @@ -210,7 +208,7 @@ int rxe_mem_init_user(struct rxe_dev *rxe, struct rxe_pd *pd, u64 start, } buf->addr = (uintptr_t)vaddr; - buf->size = umem->page_size; + buf->size = BIT(umem->page_shift); num_buf++; buf++; @@ -370,7 +368,8 @@ int rxe_mem_copy(struct rxe_mem *mem, u64 iova, void *addr, int length, ((void *)(uintptr_t)iova) : addr; if (crcp) - *crcp = crc32_le(*crcp, src, length); + crc = rxe_crc32(to_rdev(mem->pd->ibpd.device), + *crcp, src, length); memcpy(dest, src, length); @@ -403,7 +402,8 @@ int rxe_mem_copy(struct rxe_mem *mem, u64 iova, void *addr, int length, bytes = length; if (crcp) - crc = crc32_le(crc, src, bytes); + crc = rxe_crc32(to_rdev(mem->pd->ibpd.device), + crc, src, bytes); memcpy(dest, src, bytes); diff --git a/drivers/infiniband/sw/rxe/rxe_net.c b/drivers/infiniband/sw/rxe/rxe_net.c index d8610960630a..c3a140ed4df2 100644 --- a/drivers/infiniband/sw/rxe/rxe_net.c +++ b/drivers/infiniband/sw/rxe/rxe_net.c @@ -84,34 +84,6 @@ struct rxe_dev *get_rxe_by_name(const char *name) struct rxe_recv_sockets recv_sockets; -static __be64 rxe_mac_to_eui64(struct net_device *ndev) -{ - unsigned char *mac_addr = ndev->dev_addr; - __be64 eui64; - unsigned char *dst = (unsigned char *)&eui64; - - dst[0] = mac_addr[0] ^ 2; - dst[1] = mac_addr[1]; - dst[2] = mac_addr[2]; - dst[3] = 0xff; - dst[4] = 0xfe; - dst[5] = mac_addr[3]; - dst[6] = mac_addr[4]; - dst[7] = mac_addr[5]; - - return eui64; -} - -__be64 rxe_node_guid(struct rxe_dev *rxe) -{ - return rxe_mac_to_eui64(rxe->ndev); -} - -__be64 rxe_port_guid(struct rxe_dev *rxe) -{ - return rxe_mac_to_eui64(rxe->ndev); -} - struct device *rxe_dma_device(struct rxe_dev *rxe) { struct net_device *ndev; @@ -210,6 +182,39 @@ static struct dst_entry *rxe_find_route6(struct net_device *ndev, #endif +static struct dst_entry *rxe_find_route(struct rxe_dev *rxe, + struct rxe_qp *qp, + struct rxe_av *av) +{ + struct dst_entry *dst = NULL; + + if (qp_type(qp) == IB_QPT_RC) + dst = sk_dst_get(qp->sk->sk); + + if (!dst || !(dst->obsolete && dst->ops->check(dst, 0))) { + if (dst) + dst_release(dst); + + if (av->network_type == RDMA_NETWORK_IPV4) { + struct in_addr *saddr; + struct in_addr *daddr; + + saddr = &av->sgid_addr._sockaddr_in.sin_addr; + daddr = &av->dgid_addr._sockaddr_in.sin_addr; + dst = rxe_find_route4(rxe->ndev, saddr, daddr); + } else if (av->network_type == RDMA_NETWORK_IPV6) { + struct in6_addr *saddr6; + struct in6_addr *daddr6; + + saddr6 = &av->sgid_addr._sockaddr_in6.sin6_addr; + daddr6 = &av->dgid_addr._sockaddr_in6.sin6_addr; + dst = rxe_find_route6(rxe->ndev, saddr6, daddr6); + } + } + + return dst; +} + static int rxe_udp_encap_recv(struct sock *sk, struct sk_buff *skb) { struct udphdr *udph; @@ -301,7 +306,7 @@ static void prepare_ipv4_hdr(struct dst_entry *dst, struct sk_buff *skb, skb_scrub_packet(skb, xnet); skb_clear_hash(skb); - skb_dst_set(skb, dst); + skb_dst_set(skb, dst_clone(dst)); memset(IPCB(skb), 0, sizeof(*IPCB(skb))); skb_push(skb, sizeof(struct iphdr)); @@ -349,13 +354,14 @@ static void prepare_ipv6_hdr(struct dst_entry *dst, struct sk_buff *skb, static int prepare4(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, struct sk_buff *skb, struct rxe_av *av) { + struct rxe_qp *qp = pkt->qp; struct dst_entry *dst; bool xnet = false; __be16 df = htons(IP_DF); struct in_addr *saddr = &av->sgid_addr._sockaddr_in.sin_addr; struct in_addr *daddr = &av->dgid_addr._sockaddr_in.sin_addr; - dst = rxe_find_route4(rxe->ndev, saddr, daddr); + dst = rxe_find_route(rxe, qp, av); if (!dst) { pr_err("Host not reachable\n"); return -EHOSTUNREACH; @@ -369,17 +375,24 @@ static int prepare4(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, prepare_ipv4_hdr(dst, skb, saddr->s_addr, daddr->s_addr, IPPROTO_UDP, av->grh.traffic_class, av->grh.hop_limit, df, xnet); + + if (qp_type(qp) == IB_QPT_RC) + sk_dst_set(qp->sk->sk, dst); + else + dst_release(dst); + return 0; } static int prepare6(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, struct sk_buff *skb, struct rxe_av *av) { - struct dst_entry *dst; + struct rxe_qp *qp = pkt->qp; + struct dst_entry *dst = NULL; struct in6_addr *saddr = &av->sgid_addr._sockaddr_in6.sin6_addr; struct in6_addr *daddr = &av->dgid_addr._sockaddr_in6.sin6_addr; - dst = rxe_find_route6(rxe->ndev, saddr, daddr); + dst = rxe_find_route(rxe, qp, av); if (!dst) { pr_err("Host not reachable\n"); return -EHOSTUNREACH; @@ -394,6 +407,12 @@ static int prepare6(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, prepare_ipv6_hdr(dst, skb, saddr, daddr, IPPROTO_UDP, av->grh.traffic_class, av->grh.hop_limit); + + if (qp_type(qp) == IB_QPT_RC) + sk_dst_set(qp->sk->sk, dst); + else + dst_release(dst); + return 0; } diff --git a/drivers/infiniband/sw/rxe/rxe_qp.c b/drivers/infiniband/sw/rxe/rxe_qp.c index f98a19e61a3d..1a123edf555e 100644 --- a/drivers/infiniband/sw/rxe/rxe_qp.c +++ b/drivers/infiniband/sw/rxe/rxe_qp.c @@ -273,10 +273,11 @@ static int rxe_qp_init_req(struct rxe_dev *rxe, struct rxe_qp *qp, rxe_init_task(rxe, &qp->comp.task, qp, rxe_completer, "comp"); - setup_timer(&qp->rnr_nak_timer, rnr_nak_timer, (unsigned long)qp); - setup_timer(&qp->retrans_timer, retransmit_timer, (unsigned long)qp); qp->qp_timeout_jiffies = 0; /* Can't be set for UD/UC in modify_qp */ - + if (init->qp_type == IB_QPT_RC) { + setup_timer(&qp->rnr_nak_timer, rnr_nak_timer, (unsigned long)qp); + setup_timer(&qp->retrans_timer, retransmit_timer, (unsigned long)qp); + } return 0; } @@ -804,8 +805,10 @@ void rxe_qp_destroy(struct rxe_qp *qp) qp->qp_timeout_jiffies = 0; rxe_cleanup_task(&qp->resp.task); - del_timer_sync(&qp->retrans_timer); - del_timer_sync(&qp->rnr_nak_timer); + if (qp_type(qp) == IB_QPT_RC) { + del_timer_sync(&qp->retrans_timer); + del_timer_sync(&qp->rnr_nak_timer); + } rxe_cleanup_task(&qp->req.task); rxe_cleanup_task(&qp->comp.task); @@ -846,6 +849,14 @@ void rxe_qp_cleanup(struct rxe_pool_entry *arg) qp->resp.mr = NULL; } + if (qp_type(qp) == IB_QPT_RC) { + struct dst_entry *dst = NULL; + + dst = sk_dst_get(qp->sk->sk); + if (dst) + dst_release(dst); + } + free_rd_atomic_resources(qp); kernel_sock_shutdown(qp->sk, SHUT_RDWR); diff --git a/drivers/infiniband/sw/rxe/rxe_recv.c b/drivers/infiniband/sw/rxe/rxe_recv.c index 50886031096f..fb8c83e055e1 100644 --- a/drivers/infiniband/sw/rxe/rxe_recv.c +++ b/drivers/infiniband/sw/rxe/rxe_recv.c @@ -387,8 +387,8 @@ int rxe_rcv(struct sk_buff *skb) pack_icrc = be32_to_cpu(*icrcp); calc_icrc = rxe_icrc_hdr(pkt, skb); - calc_icrc = crc32_le(calc_icrc, (u8 *)payload_addr(pkt), - payload_size(pkt)); + calc_icrc = rxe_crc32(rxe, calc_icrc, (u8 *)payload_addr(pkt), + payload_size(pkt)); calc_icrc = (__force u32)cpu_to_be32(~calc_icrc); if (unlikely(calc_icrc != pack_icrc)) { if (skb->protocol == htons(ETH_P_IPV6)) @@ -403,6 +403,8 @@ int rxe_rcv(struct sk_buff *skb) goto drop; } + rxe_counter_inc(rxe, RXE_CNT_RCVD_PKTS); + if (unlikely(bth_qpn(pkt) == IB_MULTICAST_QPN)) rxe_rcv_mcast_pkt(rxe, skb); else @@ -417,4 +419,3 @@ drop: kfree_skb(skb); return 0; } -EXPORT_SYMBOL(rxe_rcv); diff --git a/drivers/infiniband/sw/rxe/rxe_req.c b/drivers/infiniband/sw/rxe/rxe_req.c index 9f95f50b2909..7ee465d1a1e1 100644 --- a/drivers/infiniband/sw/rxe/rxe_req.c +++ b/drivers/infiniband/sw/rxe/rxe_req.c @@ -32,6 +32,7 @@ */ #include <linux/skbuff.h> +#include <crypto/hash.h> #include "rxe.h" #include "rxe_loc.h" @@ -483,8 +484,7 @@ static int fill_packet(struct rxe_qp *qp, struct rxe_send_wqe *wqe, if (wqe->wr.send_flags & IB_SEND_INLINE) { u8 *tmp = &wqe->dma.inline_data[wqe->dma.sge_offset]; - crc = crc32_le(crc, tmp, paylen); - + crc = rxe_crc32(rxe, crc, tmp, paylen); memcpy(payload_addr(pkt), tmp, paylen); wqe->dma.resid -= paylen; diff --git a/drivers/infiniband/sw/rxe/rxe_resp.c b/drivers/infiniband/sw/rxe/rxe_resp.c index c9dd385ce62e..ec11a9c25f23 100644 --- a/drivers/infiniband/sw/rxe/rxe_resp.c +++ b/drivers/infiniband/sw/rxe/rxe_resp.c @@ -149,6 +149,7 @@ static enum resp_states check_psn(struct rxe_qp *qp, struct rxe_pkt_info *pkt) { int diff = psn_compare(pkt->psn, qp->resp.psn); + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); switch (qp_type(qp)) { case IB_QPT_RC: @@ -157,9 +158,11 @@ static enum resp_states check_psn(struct rxe_qp *qp, return RESPST_CLEANUP; qp->resp.sent_psn_nak = 1; + rxe_counter_inc(rxe, RXE_CNT_OUT_OF_SEQ_REQ); return RESPST_ERR_PSN_OUT_OF_SEQ; } else if (diff < 0) { + rxe_counter_inc(rxe, RXE_CNT_DUP_REQ); return RESPST_DUPLICATE_REQUEST; } @@ -1223,6 +1226,7 @@ void rxe_drain_req_pkts(struct rxe_qp *qp, bool notify) int rxe_responder(void *arg) { struct rxe_qp *qp = (struct rxe_qp *)arg; + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); enum resp_states state; struct rxe_pkt_info *pkt = NULL; int ret = 0; @@ -1311,6 +1315,7 @@ int rxe_responder(void *arg) break; case RESPST_ERR_RNR: if (qp_type(qp) == IB_QPT_RC) { + rxe_counter_inc(rxe, RXE_CNT_SND_RNR); /* RC - class B */ send_ack(qp, pkt, AETH_RNR_NAK | (~AETH_TYPE_MASK & diff --git a/drivers/infiniband/sw/rxe/rxe_verbs.c b/drivers/infiniband/sw/rxe/rxe_verbs.c index 5113e502f6f9..dbede3026138 100644 --- a/drivers/infiniband/sw/rxe/rxe_verbs.c +++ b/drivers/infiniband/sw/rxe/rxe_verbs.c @@ -32,9 +32,11 @@ */ #include <linux/dma-mapping.h> +#include <net/addrconf.h> #include "rxe.h" #include "rxe_loc.h" #include "rxe_queue.h" +#include "rxe_hw_counters.h" static int rxe_query_device(struct ib_device *dev, struct ib_device_attr *attr, @@ -1237,7 +1239,8 @@ int rxe_register_device(struct rxe_dev *rxe) dev->num_comp_vectors = RXE_NUM_COMP_VECTORS; dev->dev.parent = rxe_dma_device(rxe); dev->local_dma_lkey = 0; - dev->node_guid = rxe_node_guid(rxe); + addrconf_addr_eui48((unsigned char *)&dev->node_guid, + rxe->ndev->dev_addr); dev->dev.dma_ops = &dma_virt_ops; dev->uverbs_abi_ver = RXE_UVERBS_ABI_VERSION; @@ -1318,6 +1321,15 @@ int rxe_register_device(struct rxe_dev *rxe) dev->map_mr_sg = rxe_map_mr_sg; dev->attach_mcast = rxe_attach_mcast; dev->detach_mcast = rxe_detach_mcast; + dev->get_hw_stats = rxe_ib_get_hw_stats; + dev->alloc_hw_stats = rxe_ib_alloc_hw_stats; + + rxe->tfm = crypto_alloc_shash("crc32", 0, 0); + if (IS_ERR(rxe->tfm)) { + pr_err("failed to allocate crc algorithm err:%ld\n", + PTR_ERR(rxe->tfm)); + return PTR_ERR(rxe->tfm); + } err = ib_register_device(dev, NULL); if (err) { @@ -1339,6 +1351,8 @@ int rxe_register_device(struct rxe_dev *rxe) err2: ib_unregister_device(dev); err1: + crypto_free_shash(rxe->tfm); + return err; } diff --git a/drivers/infiniband/sw/rxe/rxe_verbs.h b/drivers/infiniband/sw/rxe/rxe_verbs.h index e100c500ae85..5a180fbe40d9 100644 --- a/drivers/infiniband/sw/rxe/rxe_verbs.h +++ b/drivers/infiniband/sw/rxe/rxe_verbs.h @@ -38,6 +38,7 @@ #include <rdma/rdma_user_rxe.h> #include "rxe_pool.h" #include "rxe_task.h" +#include "rxe_hw_counters.h" static inline int pkey_match(u16 key1, u16 key2) { @@ -401,10 +402,18 @@ struct rxe_dev { spinlock_t mmap_offset_lock; /* guard mmap_offset */ int mmap_offset; + u64 stats_counters[RXE_NUM_OF_COUNTERS]; + struct rxe_port port; struct list_head list; + struct crypto_shash *tfm; }; +static inline void rxe_counter_inc(struct rxe_dev *rxe, enum rxe_counters cnt) +{ + rxe->stats_counters[cnt]++; +} + static inline struct rxe_dev *to_rdev(struct ib_device *dev) { return dev ? container_of(dev, struct rxe_dev, ib_dev) : NULL; diff --git a/drivers/infiniband/ulp/Makefile b/drivers/infiniband/ulp/Makefile index f3c7dcf03098..c28af1823a2d 100644 --- a/drivers/infiniband/ulp/Makefile +++ b/drivers/infiniband/ulp/Makefile @@ -3,3 +3,4 @@ obj-$(CONFIG_INFINIBAND_SRP) += srp/ obj-$(CONFIG_INFINIBAND_SRPT) += srpt/ obj-$(CONFIG_INFINIBAND_ISER) += iser/ obj-$(CONFIG_INFINIBAND_ISERT) += isert/ +obj-$(CONFIG_INFINIBAND_OPA_VNIC) += opa_vnic/ diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h index bed233bf45c3..7c8de226e7f7 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib.h +++ b/drivers/infiniband/ulp/ipoib/ipoib.h @@ -52,7 +52,6 @@ #include <rdma/ib_pack.h> #include <rdma/ib_sa.h> #include <linux/sched.h> - /* constants */ enum ipoib_flush_level { @@ -153,6 +152,13 @@ static inline void skb_add_pseudo_hdr(struct sk_buff *skb) skb_pull(skb, IPOIB_HARD_LEN); } +static inline struct ipoib_dev_priv *ipoib_priv(const struct net_device *dev) +{ + struct rdma_netdev *rn = netdev_priv(dev); + + return rn->clnt_priv; +} + /* Used for all multicast joins (broadcast, IPv4 mcast and IPv6 mcast) */ struct ipoib_mcast { struct ib_sa_mcmember_rec mcmember; @@ -404,6 +410,7 @@ struct ipoib_dev_priv { struct timer_list poll_timer; unsigned max_send_sge; bool sm_fullmember_sendonly_support; + const struct net_device_ops *rn_ops; }; struct ipoib_ah { @@ -482,27 +489,28 @@ int ipoib_open(struct net_device *dev); int ipoib_add_pkey_attr(struct net_device *dev); int ipoib_add_umcast_attr(struct net_device *dev); -void ipoib_send(struct net_device *dev, struct sk_buff *skb, - struct ipoib_ah *address, u32 qpn); +int ipoib_send(struct net_device *dev, struct sk_buff *skb, + struct ib_ah *address, u32 dqpn); void ipoib_reap_ah(struct work_struct *work); struct ipoib_path *__path_find(struct net_device *dev, void *gid); void ipoib_mark_paths_invalid(struct net_device *dev); void ipoib_flush_paths(struct net_device *dev); -int ipoib_check_sm_sendonly_fullmember_support(struct ipoib_dev_priv *priv); -struct ipoib_dev_priv *ipoib_intf_alloc(const char *format); - -int ipoib_ib_dev_init(struct net_device *dev, struct ib_device *ca, int port); +struct ipoib_dev_priv *ipoib_intf_alloc(struct ib_device *hca, u8 port, + const char *format); +void ipoib_ib_tx_timer_func(unsigned long ctx); void ipoib_ib_dev_flush_light(struct work_struct *work); void ipoib_ib_dev_flush_normal(struct work_struct *work); void ipoib_ib_dev_flush_heavy(struct work_struct *work); void ipoib_pkey_event(struct work_struct *work); void ipoib_ib_dev_cleanup(struct net_device *dev); +int ipoib_ib_dev_open_default(struct net_device *dev); int ipoib_ib_dev_open(struct net_device *dev); +int ipoib_ib_dev_stop(struct net_device *dev); void ipoib_ib_dev_up(struct net_device *dev); void ipoib_ib_dev_down(struct net_device *dev); -void ipoib_ib_dev_stop(struct net_device *dev); +int ipoib_ib_dev_stop_default(struct net_device *dev); void ipoib_pkey_dev_check_presence(struct net_device *dev); int ipoib_dev_init(struct net_device *dev, struct ib_device *ca, int port); @@ -562,8 +570,10 @@ void ipoib_path_iter_read(struct ipoib_path_iter *iter, struct ipoib_path *path); #endif -int ipoib_mcast_attach(struct net_device *dev, u16 mlid, - union ib_gid *mgid, int set_qkey); +int ipoib_mcast_attach(struct net_device *dev, struct ib_device *hca, + union ib_gid *mgid, u16 mlid, int set_qkey, u32 qkey); +int ipoib_mcast_detach(struct net_device *dev, struct ib_device *hca, + union ib_gid *mgid, u16 mlid); void ipoib_mcast_remove_list(struct list_head *remove_list); void ipoib_check_and_add_mcast_sendonly(struct ipoib_dev_priv *priv, u8 *mgid, struct list_head *remove_list); @@ -587,7 +597,7 @@ void __exit ipoib_netlink_fini(void); void ipoib_set_umcast(struct net_device *ndev, int umcast_val); int ipoib_set_mode(struct net_device *dev, const char *buf); -void ipoib_setup(struct net_device *dev); +void ipoib_setup_common(struct net_device *dev); void ipoib_pkey_open(struct ipoib_dev_priv *priv); void ipoib_drain_cq(struct net_device *dev); @@ -607,14 +617,14 @@ extern int ipoib_max_conn_qp; static inline int ipoib_cm_admin_enabled(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); return IPOIB_CM_SUPPORTED(dev->dev_addr) && test_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags); } static inline int ipoib_cm_enabled(struct net_device *dev, u8 *hwaddr) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); return IPOIB_CM_SUPPORTED(hwaddr) && test_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags); } @@ -637,13 +647,13 @@ static inline void ipoib_cm_set(struct ipoib_neigh *neigh, struct ipoib_cm_tx *t static inline int ipoib_cm_has_srq(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); return !!priv->cm.srq; } static inline unsigned int ipoib_cm_max_mtu(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); return priv->cm.max_cm_mtu; } diff --git a/drivers/infiniband/ulp/ipoib/ipoib_cm.c b/drivers/infiniband/ulp/ipoib/ipoib_cm.c index 0cdf2b7f272f..21303c07ca55 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_cm.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_cm.c @@ -92,7 +92,7 @@ static void ipoib_cm_dma_unmap_rx(struct ipoib_dev_priv *priv, int frags, static int ipoib_cm_post_receive_srq(struct net_device *dev, int id) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ib_recv_wr *bad_wr; int i, ret; @@ -118,7 +118,7 @@ static int ipoib_cm_post_receive_nonsrq(struct net_device *dev, struct ib_recv_wr *wr, struct ib_sge *sge, int id) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ib_recv_wr *bad_wr; int i, ret; @@ -145,7 +145,7 @@ static struct sk_buff *ipoib_cm_alloc_rx_skb(struct net_device *dev, u64 mapping[IPOIB_CM_RX_SG], gfp_t gfp) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct sk_buff *skb; int i; @@ -196,7 +196,7 @@ partial_error: static void ipoib_cm_free_rx_ring(struct net_device *dev, struct ipoib_cm_rx_buf *rx_ring) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); int i; for (i = 0; i < ipoib_recvq_size; ++i) @@ -235,7 +235,7 @@ static void ipoib_cm_start_rx_drain(struct ipoib_dev_priv *priv) static void ipoib_cm_rx_event_handler(struct ib_event *event, void *ctx) { struct ipoib_cm_rx *p = ctx; - struct ipoib_dev_priv *priv = netdev_priv(p->dev); + struct ipoib_dev_priv *priv = ipoib_priv(p->dev); unsigned long flags; if (event->event != IB_EVENT_QP_LAST_WQE_REACHED) @@ -251,7 +251,7 @@ static void ipoib_cm_rx_event_handler(struct ib_event *event, void *ctx) static struct ib_qp *ipoib_cm_create_rx_qp(struct net_device *dev, struct ipoib_cm_rx *p) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ib_qp_init_attr attr = { .event_handler = ipoib_cm_rx_event_handler, .send_cq = priv->recv_cq, /* For drain WR */ @@ -276,7 +276,7 @@ static int ipoib_cm_modify_rx_qp(struct net_device *dev, struct ib_cm_id *cm_id, struct ib_qp *qp, unsigned psn) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ib_qp_attr qp_attr; int qp_attr_mask, ret; @@ -331,7 +331,7 @@ static void ipoib_cm_init_rx_wr(struct net_device *dev, struct ib_recv_wr *wr, struct ib_sge *sge) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); int i; for (i = 0; i < priv->cm.num_frags; ++i) @@ -349,7 +349,7 @@ static void ipoib_cm_init_rx_wr(struct net_device *dev, static int ipoib_cm_nonsrq_init_rx(struct net_device *dev, struct ib_cm_id *cm_id, struct ipoib_cm_rx *rx) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct { struct ib_recv_wr wr; struct ib_sge sge[IPOIB_CM_RX_SG]; @@ -422,7 +422,7 @@ static int ipoib_cm_send_rep(struct net_device *dev, struct ib_cm_id *cm_id, struct ib_qp *qp, struct ib_cm_req_event_param *req, unsigned psn) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_cm_data data = {}; struct ib_cm_rep_param rep = {}; @@ -442,7 +442,7 @@ static int ipoib_cm_send_rep(struct net_device *dev, struct ib_cm_id *cm_id, static int ipoib_cm_req_handler(struct ib_cm_id *cm_id, struct ib_cm_event *event) { struct net_device *dev = cm_id->context; - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_cm_rx *p; unsigned psn; int ret; @@ -515,7 +515,7 @@ static int ipoib_cm_rx_handler(struct ib_cm_id *cm_id, /* Fall through */ case IB_CM_REJ_RECEIVED: p = cm_id->context; - priv = netdev_priv(p->dev); + priv = ipoib_priv(p->dev); if (ib_modify_qp(p->qp, &ipoib_cm_err_attr, IB_QP_STATE)) ipoib_warn(priv, "unable to move qp to error state\n"); /* Fall through */ @@ -559,7 +559,7 @@ static void skb_put_frags(struct sk_buff *skb, unsigned int hdr_space, void ipoib_cm_handle_rx_wc(struct net_device *dev, struct ib_wc *wc) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_cm_rx_buf *rx_ring; unsigned int wr_id = wc->wr_id & ~(IPOIB_OP_CM | IPOIB_OP_RECV); struct sk_buff *skb, *newskb; @@ -708,7 +708,7 @@ static inline int post_send(struct ipoib_dev_priv *priv, void ipoib_cm_send(struct net_device *dev, struct sk_buff *skb, struct ipoib_cm_tx *tx) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_tx_buf *tx_req; int rc; unsigned usable_sge = tx->max_send_sge - !!skb_headlen(skb); @@ -786,7 +786,7 @@ void ipoib_cm_send(struct net_device *dev, struct sk_buff *skb, struct ipoib_cm_ void ipoib_cm_handle_tx_wc(struct net_device *dev, struct ib_wc *wc) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_cm_tx *tx = wc->qp->qp_context; unsigned int wr_id = wc->wr_id & ~IPOIB_OP_CM; struct ipoib_tx_buf *tx_req; @@ -855,7 +855,7 @@ void ipoib_cm_handle_tx_wc(struct net_device *dev, struct ib_wc *wc) int ipoib_cm_dev_open(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); int ret; if (!IPOIB_CM_SUPPORTED(dev->dev_addr)) @@ -887,7 +887,7 @@ err_cm: static void ipoib_cm_free_rx_reap_list(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_cm_rx *rx, *n; LIST_HEAD(list); @@ -910,7 +910,7 @@ static void ipoib_cm_free_rx_reap_list(struct net_device *dev) void ipoib_cm_dev_stop(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_cm_rx *p; unsigned long begin; int ret; @@ -969,7 +969,7 @@ void ipoib_cm_dev_stop(struct net_device *dev) static int ipoib_cm_rep_handler(struct ib_cm_id *cm_id, struct ib_cm_event *event) { struct ipoib_cm_tx *p = cm_id->context; - struct ipoib_dev_priv *priv = netdev_priv(p->dev); + struct ipoib_dev_priv *priv = ipoib_priv(p->dev); struct ipoib_cm_data *data = event->private_data; struct sk_buff_head skqueue; struct ib_qp_attr qp_attr; @@ -1037,7 +1037,7 @@ static int ipoib_cm_rep_handler(struct ib_cm_id *cm_id, struct ib_cm_event *even static struct ib_qp *ipoib_cm_create_tx_qp(struct net_device *dev, struct ipoib_cm_tx *tx) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ib_qp_init_attr attr = { .send_cq = priv->recv_cq, .recv_cq = priv->recv_cq, @@ -1070,7 +1070,7 @@ static int ipoib_cm_send_req(struct net_device *dev, u32 qpn, struct ib_sa_path_rec *pathrec) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_cm_data data = {}; struct ib_cm_req_param req = {}; @@ -1105,7 +1105,7 @@ static int ipoib_cm_send_req(struct net_device *dev, static int ipoib_cm_modify_tx_init(struct net_device *dev, struct ib_cm_id *cm_id, struct ib_qp *qp) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ib_qp_attr qp_attr; int qp_attr_mask, ret; ret = ib_find_pkey(priv->ca, priv->port, priv->pkey, &qp_attr.pkey_index); @@ -1130,7 +1130,7 @@ static int ipoib_cm_modify_tx_init(struct net_device *dev, static int ipoib_cm_tx_init(struct ipoib_cm_tx *p, u32 qpn, struct ib_sa_path_rec *pathrec) { - struct ipoib_dev_priv *priv = netdev_priv(p->dev); + struct ipoib_dev_priv *priv = ipoib_priv(p->dev); int ret; p->tx_ring = __vmalloc(ipoib_sendq_size * sizeof *p->tx_ring, @@ -1186,7 +1186,7 @@ err_tx: static void ipoib_cm_tx_destroy(struct ipoib_cm_tx *p) { - struct ipoib_dev_priv *priv = netdev_priv(p->dev); + struct ipoib_dev_priv *priv = ipoib_priv(p->dev); struct ipoib_tx_buf *tx_req; unsigned long begin; @@ -1236,7 +1236,7 @@ static int ipoib_cm_tx_handler(struct ib_cm_id *cm_id, struct ib_cm_event *event) { struct ipoib_cm_tx *tx = cm_id->context; - struct ipoib_dev_priv *priv = netdev_priv(tx->dev); + struct ipoib_dev_priv *priv = ipoib_priv(tx->dev); struct net_device *dev = priv->dev; struct ipoib_neigh *neigh; unsigned long flags; @@ -1287,7 +1287,7 @@ static int ipoib_cm_tx_handler(struct ib_cm_id *cm_id, struct ipoib_cm_tx *ipoib_cm_create_tx(struct net_device *dev, struct ipoib_path *path, struct ipoib_neigh *neigh) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_cm_tx *tx; tx = kzalloc(sizeof *tx, GFP_ATOMIC); @@ -1306,7 +1306,7 @@ struct ipoib_cm_tx *ipoib_cm_create_tx(struct net_device *dev, struct ipoib_path void ipoib_cm_destroy_tx(struct ipoib_cm_tx *tx) { - struct ipoib_dev_priv *priv = netdev_priv(tx->dev); + struct ipoib_dev_priv *priv = ipoib_priv(tx->dev); unsigned long flags; if (test_and_clear_bit(IPOIB_FLAG_INITIALIZED, &tx->flags)) { spin_lock_irqsave(&priv->lock, flags); @@ -1441,7 +1441,7 @@ static void ipoib_cm_skb_reap(struct work_struct *work) void ipoib_cm_skb_too_long(struct net_device *dev, struct sk_buff *skb, unsigned int mtu) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); int e = skb_queue_empty(&priv->cm.skb_queue); if (skb_dst(skb)) @@ -1490,7 +1490,8 @@ static void ipoib_cm_stale_task(struct work_struct *work) static ssize_t show_mode(struct device *d, struct device_attribute *attr, char *buf) { - struct ipoib_dev_priv *priv = netdev_priv(to_net_dev(d)); + struct net_device *dev = to_net_dev(d); + struct ipoib_dev_priv *priv = ipoib_priv(dev); if (test_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags)) return sprintf(buf, "connected\n"); @@ -1503,7 +1504,7 @@ static ssize_t set_mode(struct device *d, struct device_attribute *attr, { struct net_device *dev = to_net_dev(d); int ret; - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); if (test_bit(IPOIB_FLAG_GOING_DOWN, &priv->flags)) return -EPERM; @@ -1532,7 +1533,7 @@ int ipoib_cm_add_mode_attr(struct net_device *dev) static void ipoib_cm_create_srq(struct net_device *dev, int max_sge) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ib_srq_init_attr srq_init_attr = { .srq_type = IB_SRQT_BASIC, .attr = { @@ -1561,7 +1562,7 @@ static void ipoib_cm_create_srq(struct net_device *dev, int max_sge) int ipoib_cm_dev_init(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); int max_srq_sge, i; INIT_LIST_HEAD(&priv->cm.passive_ids); @@ -1622,7 +1623,7 @@ int ipoib_cm_dev_init(struct net_device *dev) void ipoib_cm_dev_cleanup(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); int ret; if (!priv->cm.srq) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c b/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c index bac455a1942d..379c02fb4181 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c @@ -60,7 +60,7 @@ static const struct ipoib_stats ipoib_gstrings_stats[] = { static void ipoib_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo) { - struct ipoib_dev_priv *priv = netdev_priv(netdev); + struct ipoib_dev_priv *priv = ipoib_priv(netdev); ib_get_device_fw_str(priv->ca, drvinfo->fw_version, sizeof(drvinfo->fw_version)); @@ -77,7 +77,7 @@ static void ipoib_get_drvinfo(struct net_device *netdev, static int ipoib_get_coalesce(struct net_device *dev, struct ethtool_coalesce *coal) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); coal->rx_coalesce_usecs = priv->ethtool.coalesce_usecs; coal->rx_max_coalesced_frames = priv->ethtool.max_coalesced_frames; @@ -88,7 +88,7 @@ static int ipoib_get_coalesce(struct net_device *dev, static int ipoib_set_coalesce(struct net_device *dev, struct ethtool_coalesce *coal) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); int ret; /* diff --git a/drivers/infiniband/ulp/ipoib/ipoib_fs.c b/drivers/infiniband/ulp/ipoib/ipoib_fs.c index 6bd5740e2691..a15664b880ed 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_fs.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_fs.c @@ -261,7 +261,7 @@ static const struct file_operations ipoib_path_fops = { void ipoib_create_debug_files(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); char name[IFNAMSIZ + sizeof "_path"]; snprintf(name, sizeof name, "%s_mcg", dev->name); @@ -279,10 +279,13 @@ void ipoib_create_debug_files(struct net_device *dev) void ipoib_delete_debug_files(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); + WARN_ONCE(!priv->mcg_dentry, "null mcg debug file\n"); + WARN_ONCE(!priv->path_dentry, "null path debug file\n"); debugfs_remove(priv->mcg_dentry); debugfs_remove(priv->path_dentry); + priv->mcg_dentry = priv->path_dentry = NULL; } int ipoib_register_debugfs(void) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c index 12c4f84a6639..2bfcfb1554d8 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c @@ -71,7 +71,7 @@ struct ipoib_ah *ipoib_create_ah(struct net_device *dev, ah = (struct ipoib_ah *)vah; } else { ah->ah = vah; - ipoib_dbg(netdev_priv(dev), "Created ah %p\n", ah->ah); + ipoib_dbg(ipoib_priv(dev), "Created ah %p\n", ah->ah); } return ah; @@ -80,7 +80,7 @@ struct ipoib_ah *ipoib_create_ah(struct net_device *dev, void ipoib_free_ah(struct kref *kref) { struct ipoib_ah *ah = container_of(kref, struct ipoib_ah, ref); - struct ipoib_dev_priv *priv = netdev_priv(ah->dev); + struct ipoib_dev_priv *priv = ipoib_priv(ah->dev); unsigned long flags; @@ -99,7 +99,7 @@ static void ipoib_ud_dma_unmap_rx(struct ipoib_dev_priv *priv, static int ipoib_ib_post_receive(struct net_device *dev, int id) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ib_recv_wr *bad_wr; int ret; @@ -121,7 +121,7 @@ static int ipoib_ib_post_receive(struct net_device *dev, int id) static struct sk_buff *ipoib_alloc_rx_skb(struct net_device *dev, int id) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct sk_buff *skb; int buf_size; u64 *mapping; @@ -153,7 +153,7 @@ error: static int ipoib_ib_post_receives(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); int i; for (i = 0; i < ipoib_recvq_size; ++i) { @@ -172,7 +172,7 @@ static int ipoib_ib_post_receives(struct net_device *dev) static void ipoib_ib_handle_rx_wc(struct net_device *dev, struct ib_wc *wc) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); unsigned int wr_id = wc->wr_id & ~IPOIB_OP_RECV; struct sk_buff *skb; u64 mapping[IPOIB_UD_RX_SG]; @@ -381,7 +381,7 @@ free_res: static void ipoib_ib_handle_tx_wc(struct net_device *dev, struct ib_wc *wc) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); unsigned int wr_id = wc->wr_id; struct ipoib_tx_buf *tx_req; @@ -485,14 +485,14 @@ poll_more: void ipoib_ib_completion(struct ib_cq *cq, void *dev_ptr) { struct net_device *dev = dev_ptr; - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); napi_schedule(&priv->napi); } static void drain_tx_cq(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); netif_tx_lock(dev); while (poll_tx(priv)) @@ -506,14 +506,14 @@ static void drain_tx_cq(struct net_device *dev) void ipoib_send_comp_handler(struct ib_cq *cq, void *dev_ptr) { - struct ipoib_dev_priv *priv = netdev_priv(dev_ptr); + struct ipoib_dev_priv *priv = ipoib_priv(dev_ptr); mod_timer(&priv->poll_timer, jiffies); } static inline int post_send(struct ipoib_dev_priv *priv, unsigned int wr_id, - struct ib_ah *address, u32 qpn, + struct ib_ah *address, u32 dqpn, struct ipoib_tx_buf *tx_req, void *head, int hlen) { @@ -523,7 +523,7 @@ static inline int post_send(struct ipoib_dev_priv *priv, ipoib_build_sge(priv, tx_req); priv->tx_wr.wr.wr_id = wr_id; - priv->tx_wr.remote_qpn = qpn; + priv->tx_wr.remote_qpn = dqpn; priv->tx_wr.ah = address; if (head) { @@ -537,10 +537,10 @@ static inline int post_send(struct ipoib_dev_priv *priv, return ib_post_send(priv->qp, &priv->tx_wr.wr, &bad_wr); } -void ipoib_send(struct net_device *dev, struct sk_buff *skb, - struct ipoib_ah *address, u32 qpn) +int ipoib_send(struct net_device *dev, struct sk_buff *skb, + struct ib_ah *address, u32 dqpn) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_tx_buf *tx_req; int hlen, rc; void *phead; @@ -554,7 +554,7 @@ void ipoib_send(struct net_device *dev, struct sk_buff *skb, ++dev->stats.tx_dropped; ++dev->stats.tx_errors; dev_kfree_skb_any(skb); - return; + return -1; } } else { if (unlikely(skb->len > priv->mcast_mtu + IPOIB_ENCAP_LEN)) { @@ -563,7 +563,7 @@ void ipoib_send(struct net_device *dev, struct sk_buff *skb, ++dev->stats.tx_dropped; ++dev->stats.tx_errors; ipoib_cm_skb_too_long(dev, skb, priv->mcast_mtu); - return; + return -1; } phead = NULL; hlen = 0; @@ -574,7 +574,7 @@ void ipoib_send(struct net_device *dev, struct sk_buff *skb, ++dev->stats.tx_dropped; ++dev->stats.tx_errors; dev_kfree_skb_any(skb); - return; + return -1; } /* Does skb_linearize return ok without reducing nr_frags? */ if (skb_shinfo(skb)->nr_frags > usable_sge) { @@ -582,12 +582,13 @@ void ipoib_send(struct net_device *dev, struct sk_buff *skb, ++dev->stats.tx_dropped; ++dev->stats.tx_errors; dev_kfree_skb_any(skb); - return; + return -1; } } - ipoib_dbg_data(priv, "sending packet, length=%d address=%p qpn=0x%06x\n", - skb->len, address, qpn); + ipoib_dbg_data(priv, + "sending packet, length=%d address=%p dqpn=0x%06x\n", + skb->len, address, dqpn); /* * We put the skb into the tx_ring _before_ we call post_send() @@ -601,7 +602,7 @@ void ipoib_send(struct net_device *dev, struct sk_buff *skb, if (unlikely(ipoib_dma_map_tx(priv->ca, tx_req))) { ++dev->stats.tx_errors; dev_kfree_skb_any(skb); - return; + return -1; } if (skb->ip_summed == CHECKSUM_PARTIAL) @@ -620,7 +621,7 @@ void ipoib_send(struct net_device *dev, struct sk_buff *skb, skb_dst_drop(skb); rc = post_send(priv, priv->tx_head & (ipoib_sendq_size - 1), - address->ah, qpn, tx_req, phead, hlen); + address, dqpn, tx_req, phead, hlen); if (unlikely(rc)) { ipoib_warn(priv, "post_send failed, error %d\n", rc); ++dev->stats.tx_errors; @@ -629,21 +630,24 @@ void ipoib_send(struct net_device *dev, struct sk_buff *skb, dev_kfree_skb_any(skb); if (netif_queue_stopped(dev)) netif_wake_queue(dev); + rc = 0; } else { netif_trans_update(dev); - address->last_send = priv->tx_head; + rc = priv->tx_head; ++priv->tx_head; } if (unlikely(priv->tx_outstanding > MAX_SEND_CQE)) while (poll_tx(priv)) ; /* nothing */ + + return rc; } static void __ipoib_reap_ah(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_ah *ah, *tah; LIST_HEAD(remove_list); unsigned long flags; @@ -677,7 +681,7 @@ void ipoib_reap_ah(struct work_struct *work) static void ipoib_flush_ah(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); cancel_delayed_work(&priv->ah_reap_task); flush_workqueue(priv->wq); @@ -686,30 +690,124 @@ static void ipoib_flush_ah(struct net_device *dev) static void ipoib_stop_ah(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); set_bit(IPOIB_STOP_REAPER, &priv->flags); ipoib_flush_ah(dev); } -static void ipoib_ib_tx_timer_func(unsigned long ctx) +static int recvs_pending(struct net_device *dev) { - drain_tx_cq((struct net_device *)ctx); + struct ipoib_dev_priv *priv = ipoib_priv(dev); + int pending = 0; + int i; + + for (i = 0; i < ipoib_recvq_size; ++i) + if (priv->rx_ring[i].skb) + ++pending; + + return pending; } -int ipoib_ib_dev_open(struct net_device *dev) +int ipoib_ib_dev_stop_default(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); - int ret; + struct ipoib_dev_priv *priv = ipoib_priv(dev); + struct ib_qp_attr qp_attr; + unsigned long begin; + struct ipoib_tx_buf *tx_req; + int i; - ipoib_pkey_dev_check_presence(dev); + if (test_bit(IPOIB_FLAG_INITIALIZED, &priv->flags)) + napi_disable(&priv->napi); - if (!test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags)) { - ipoib_warn(priv, "P_Key 0x%04x is %s\n", priv->pkey, - (!(priv->pkey & 0x7fff) ? "Invalid" : "not found")); - return -1; + ipoib_cm_dev_stop(dev); + + /* + * Move our QP to the error state and then reinitialize in + * when all work requests have completed or have been flushed. + */ + qp_attr.qp_state = IB_QPS_ERR; + if (ib_modify_qp(priv->qp, &qp_attr, IB_QP_STATE)) + ipoib_warn(priv, "Failed to modify QP to ERROR state\n"); + + /* Wait for all sends and receives to complete */ + begin = jiffies; + + while (priv->tx_head != priv->tx_tail || recvs_pending(dev)) { + if (time_after(jiffies, begin + 5 * HZ)) { + ipoib_warn(priv, + "timing out; %d sends %d receives not completed\n", + priv->tx_head - priv->tx_tail, + recvs_pending(dev)); + + /* + * assume the HW is wedged and just free up + * all our pending work requests. + */ + while ((int)priv->tx_tail - (int)priv->tx_head < 0) { + tx_req = &priv->tx_ring[priv->tx_tail & + (ipoib_sendq_size - 1)]; + ipoib_dma_unmap_tx(priv, tx_req); + dev_kfree_skb_any(tx_req->skb); + ++priv->tx_tail; + --priv->tx_outstanding; + } + + for (i = 0; i < ipoib_recvq_size; ++i) { + struct ipoib_rx_buf *rx_req; + + rx_req = &priv->rx_ring[i]; + if (!rx_req->skb) + continue; + ipoib_ud_dma_unmap_rx(priv, + priv->rx_ring[i].mapping); + dev_kfree_skb_any(rx_req->skb); + rx_req->skb = NULL; + } + + goto timeout; + } + + ipoib_drain_cq(dev); + + msleep(1); } + ipoib_dbg(priv, "All sends and receives done.\n"); + +timeout: + del_timer_sync(&priv->poll_timer); + qp_attr.qp_state = IB_QPS_RESET; + if (ib_modify_qp(priv->qp, &qp_attr, IB_QP_STATE)) + ipoib_warn(priv, "Failed to modify QP to RESET state\n"); + + ib_req_notify_cq(priv->recv_cq, IB_CQ_NEXT_COMP); + + return 0; +} + +int ipoib_ib_dev_stop(struct net_device *dev) +{ + struct ipoib_dev_priv *priv = ipoib_priv(dev); + + priv->rn_ops->ndo_stop(dev); + + clear_bit(IPOIB_FLAG_INITIALIZED, &priv->flags); + ipoib_flush_ah(dev); + + return 0; +} + +void ipoib_ib_tx_timer_func(unsigned long ctx) +{ + drain_tx_cq((struct net_device *)ctx); +} + +int ipoib_ib_dev_open_default(struct net_device *dev) +{ + struct ipoib_dev_priv *priv = ipoib_priv(dev); + int ret; + ret = ipoib_init_qp(dev); if (ret) { ipoib_warn(priv, "ipoib_init_qp returned %d\n", ret); @@ -719,33 +817,60 @@ int ipoib_ib_dev_open(struct net_device *dev) ret = ipoib_ib_post_receives(dev); if (ret) { ipoib_warn(priv, "ipoib_ib_post_receives returned %d\n", ret); - goto dev_stop; + goto out; } ret = ipoib_cm_dev_open(dev); if (ret) { ipoib_warn(priv, "ipoib_cm_dev_open returned %d\n", ret); - goto dev_stop; + goto out; + } + + if (!test_bit(IPOIB_FLAG_INITIALIZED, &priv->flags)) + napi_enable(&priv->napi); + + return 0; +out: + return -1; +} + +int ipoib_ib_dev_open(struct net_device *dev) +{ + struct ipoib_dev_priv *priv = ipoib_priv(dev); + + ipoib_pkey_dev_check_presence(dev); + + if (!test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags)) { + ipoib_warn(priv, "P_Key 0x%04x is %s\n", priv->pkey, + (!(priv->pkey & 0x7fff) ? "Invalid" : "not found")); + return -1; } clear_bit(IPOIB_STOP_REAPER, &priv->flags); queue_delayed_work(priv->wq, &priv->ah_reap_task, round_jiffies_relative(HZ)); - if (!test_and_set_bit(IPOIB_FLAG_INITIALIZED, &priv->flags)) - napi_enable(&priv->napi); + if (priv->rn_ops->ndo_open(dev)) { + pr_warn("%s: Failed to open dev\n", dev->name); + goto dev_stop; + } + + set_bit(IPOIB_FLAG_INITIALIZED, &priv->flags); return 0; + dev_stop: - if (!test_and_set_bit(IPOIB_FLAG_INITIALIZED, &priv->flags)) - napi_enable(&priv->napi); + set_bit(IPOIB_STOP_REAPER, &priv->flags); + cancel_delayed_work(&priv->ah_reap_task); + set_bit(IPOIB_FLAG_INITIALIZED, &priv->flags); + napi_enable(&priv->napi); ipoib_ib_dev_stop(dev); return -1; } void ipoib_pkey_dev_check_presence(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); if (!(priv->pkey & 0x7fff) || ib_find_pkey(priv->ca, priv->port, priv->pkey, @@ -757,7 +882,7 @@ void ipoib_pkey_dev_check_presence(struct net_device *dev) void ipoib_ib_dev_up(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); ipoib_pkey_dev_check_presence(dev); @@ -773,7 +898,7 @@ void ipoib_ib_dev_up(struct net_device *dev) void ipoib_ib_dev_down(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); ipoib_dbg(priv, "downing ib_dev\n"); @@ -786,22 +911,9 @@ void ipoib_ib_dev_down(struct net_device *dev) ipoib_flush_paths(dev); } -static int recvs_pending(struct net_device *dev) -{ - struct ipoib_dev_priv *priv = netdev_priv(dev); - int pending = 0; - int i; - - for (i = 0; i < ipoib_recvq_size; ++i) - if (priv->rx_ring[i].skb) - ++pending; - - return pending; -} - void ipoib_drain_cq(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); int i, n; /* @@ -838,107 +950,6 @@ void ipoib_drain_cq(struct net_device *dev) local_bh_enable(); } -void ipoib_ib_dev_stop(struct net_device *dev) -{ - struct ipoib_dev_priv *priv = netdev_priv(dev); - struct ib_qp_attr qp_attr; - unsigned long begin; - struct ipoib_tx_buf *tx_req; - int i; - - if (test_and_clear_bit(IPOIB_FLAG_INITIALIZED, &priv->flags)) - napi_disable(&priv->napi); - - ipoib_cm_dev_stop(dev); - - /* - * Move our QP to the error state and then reinitialize in - * when all work requests have completed or have been flushed. - */ - qp_attr.qp_state = IB_QPS_ERR; - if (ib_modify_qp(priv->qp, &qp_attr, IB_QP_STATE)) - ipoib_warn(priv, "Failed to modify QP to ERROR state\n"); - - /* Wait for all sends and receives to complete */ - begin = jiffies; - - while (priv->tx_head != priv->tx_tail || recvs_pending(dev)) { - if (time_after(jiffies, begin + 5 * HZ)) { - ipoib_warn(priv, "timing out; %d sends %d receives not completed\n", - priv->tx_head - priv->tx_tail, recvs_pending(dev)); - - /* - * assume the HW is wedged and just free up - * all our pending work requests. - */ - while ((int) priv->tx_tail - (int) priv->tx_head < 0) { - tx_req = &priv->tx_ring[priv->tx_tail & - (ipoib_sendq_size - 1)]; - ipoib_dma_unmap_tx(priv, tx_req); - dev_kfree_skb_any(tx_req->skb); - ++priv->tx_tail; - --priv->tx_outstanding; - } - - for (i = 0; i < ipoib_recvq_size; ++i) { - struct ipoib_rx_buf *rx_req; - - rx_req = &priv->rx_ring[i]; - if (!rx_req->skb) - continue; - ipoib_ud_dma_unmap_rx(priv, - priv->rx_ring[i].mapping); - dev_kfree_skb_any(rx_req->skb); - rx_req->skb = NULL; - } - - goto timeout; - } - - ipoib_drain_cq(dev); - - msleep(1); - } - - ipoib_dbg(priv, "All sends and receives done.\n"); - -timeout: - del_timer_sync(&priv->poll_timer); - qp_attr.qp_state = IB_QPS_RESET; - if (ib_modify_qp(priv->qp, &qp_attr, IB_QP_STATE)) - ipoib_warn(priv, "Failed to modify QP to RESET state\n"); - - ipoib_flush_ah(dev); - - ib_req_notify_cq(priv->recv_cq, IB_CQ_NEXT_COMP); -} - -int ipoib_ib_dev_init(struct net_device *dev, struct ib_device *ca, int port) -{ - struct ipoib_dev_priv *priv = netdev_priv(dev); - - priv->ca = ca; - priv->port = port; - priv->qp = NULL; - - if (ipoib_transport_dev_init(dev, ca)) { - printk(KERN_WARNING "%s: ipoib_transport_dev_init failed\n", ca->name); - return -ENODEV; - } - - setup_timer(&priv->poll_timer, ipoib_ib_tx_timer_func, - (unsigned long) dev); - - if (dev->flags & IFF_UP) { - if (ipoib_ib_dev_open(dev)) { - ipoib_transport_dev_cleanup(dev); - return -ENODEV; - } - } - - return 0; -} - /* * Takes whatever value which is in pkey index 0 and updates priv->pkey * returns 0 if the pkey value was changed. @@ -967,6 +978,19 @@ static inline int update_parent_pkey(struct ipoib_dev_priv *priv) */ priv->dev->broadcast[8] = priv->pkey >> 8; priv->dev->broadcast[9] = priv->pkey & 0xff; + + /* + * Update the broadcast address in the priv->broadcast object, + * in case it already exists, otherwise no one will do that. + */ + if (priv->broadcast) { + spin_lock_irq(&priv->lock); + memcpy(priv->broadcast->mcmember.mgid.raw, + priv->dev->broadcast + 4, + sizeof(union ib_gid)); + spin_unlock_irq(&priv->lock); + } + return 0; } @@ -1216,7 +1240,7 @@ void ipoib_ib_dev_flush_heavy(struct work_struct *work) void ipoib_ib_dev_cleanup(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); ipoib_dbg(priv, "cleaning up ib_dev\n"); /* @@ -1236,7 +1260,13 @@ void ipoib_ib_dev_cleanup(struct net_device *dev) */ ipoib_stop_ah(dev); - ipoib_transport_dev_cleanup(dev); -} + clear_bit(IPOIB_PKEY_ASSIGNED, &priv->flags); + priv->rn_ops->ndo_uninit(dev); + + if (priv->pd) { + ib_dealloc_pd(priv->pd); + priv->pd = NULL; + } +} diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index d1d3fb7a6127..05a07224fe7a 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -108,9 +108,36 @@ static struct ib_client ipoib_client = { .get_net_dev_by_params = ipoib_get_net_dev_by_params, }; +#ifdef CONFIG_INFINIBAND_IPOIB_DEBUG +static int ipoib_netdev_event(struct notifier_block *this, + unsigned long event, void *ptr) +{ + struct netdev_notifier_info *ni = ptr; + struct net_device *dev = ni->dev; + + if (dev->netdev_ops->ndo_open != ipoib_open) + return NOTIFY_DONE; + + switch (event) { + case NETDEV_REGISTER: + ipoib_create_debug_files(dev); + break; + case NETDEV_CHANGENAME: + ipoib_delete_debug_files(dev); + ipoib_create_debug_files(dev); + break; + case NETDEV_UNREGISTER: + ipoib_delete_debug_files(dev); + break; + } + + return NOTIFY_DONE; +} +#endif + int ipoib_open(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); ipoib_dbg(priv, "bringing up interface\n"); @@ -157,7 +184,7 @@ err_disable: static int ipoib_stop(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); ipoib_dbg(priv, "stopping interface\n"); @@ -195,7 +222,7 @@ static void ipoib_uninit(struct net_device *dev) static netdev_features_t ipoib_fix_features(struct net_device *dev, netdev_features_t features) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); if (test_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags)) features &= ~(NETIF_F_IP_CSUM | NETIF_F_TSO); @@ -205,7 +232,7 @@ 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 = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); /* dev->mtu > 2K ==> connected mode */ if (ipoib_cm_admin_enabled(dev)) { @@ -468,7 +495,7 @@ static struct net_device *ipoib_get_net_dev_by_params( int ipoib_set_mode(struct net_device *dev, const char *buf) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); if ((test_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags) && !strcmp(buf, "connected\n")) || @@ -505,7 +532,7 @@ int ipoib_set_mode(struct net_device *dev, const char *buf) struct ipoib_path *__path_find(struct net_device *dev, void *gid) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct rb_node *n = priv->path_tree.rb_node; struct ipoib_path *path; int ret; @@ -529,7 +556,7 @@ struct ipoib_path *__path_find(struct net_device *dev, void *gid) static int __path_add(struct net_device *dev, struct ipoib_path *path) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct rb_node **n = &priv->path_tree.rb_node; struct rb_node *pn = NULL; struct ipoib_path *tpath; @@ -564,7 +591,7 @@ static void path_free(struct net_device *dev, struct ipoib_path *path) while ((skb = __skb_dequeue(&path->queue))) dev_kfree_skb_irq(skb); - ipoib_dbg(netdev_priv(dev), "path_free\n"); + ipoib_dbg(ipoib_priv(dev), "path_free\n"); /* remove all neigh connected to this path */ ipoib_del_neighs_by_gid(dev, path->pathrec.dgid.raw); @@ -598,7 +625,7 @@ struct ipoib_path_iter *ipoib_path_iter_init(struct net_device *dev) int ipoib_path_iter_next(struct ipoib_path_iter *iter) { - struct ipoib_dev_priv *priv = netdev_priv(iter->dev); + struct ipoib_dev_priv *priv = ipoib_priv(iter->dev); struct rb_node *n; struct ipoib_path *path; int ret = 1; @@ -635,7 +662,7 @@ void ipoib_path_iter_read(struct ipoib_path_iter *iter, void ipoib_mark_paths_invalid(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_path *path, *tp; spin_lock_irq(&priv->lock); @@ -650,77 +677,6 @@ void ipoib_mark_paths_invalid(struct net_device *dev) spin_unlock_irq(&priv->lock); } -struct classport_info_context { - struct ipoib_dev_priv *priv; - struct completion done; - struct ib_sa_query *sa_query; -}; - -static void classport_info_query_cb(int status, struct ib_class_port_info *rec, - void *context) -{ - struct classport_info_context *cb_ctx = context; - struct ipoib_dev_priv *priv; - - WARN_ON(!context); - - priv = cb_ctx->priv; - - if (status || !rec) { - pr_debug("device: %s failed query classport_info status: %d\n", - priv->dev->name, status); - /* keeps the default, will try next mcast_restart */ - priv->sm_fullmember_sendonly_support = false; - goto out; - } - - if (ib_get_cpi_capmask2(rec) & - IB_SA_CAP_MASK2_SENDONLY_FULL_MEM_SUPPORT) { - pr_debug("device: %s enabled fullmember-sendonly for sendonly MCG\n", - priv->dev->name); - priv->sm_fullmember_sendonly_support = true; - } else { - pr_debug("device: %s disabled fullmember-sendonly for sendonly MCG\n", - priv->dev->name); - priv->sm_fullmember_sendonly_support = false; - } - -out: - complete(&cb_ctx->done); -} - -int ipoib_check_sm_sendonly_fullmember_support(struct ipoib_dev_priv *priv) -{ - struct classport_info_context *callback_context; - int ret; - - callback_context = kmalloc(sizeof(*callback_context), GFP_KERNEL); - if (!callback_context) - return -ENOMEM; - - callback_context->priv = priv; - init_completion(&callback_context->done); - - ret = ib_sa_classport_info_rec_query(&ipoib_sa_client, - priv->ca, priv->port, 3000, - GFP_KERNEL, - classport_info_query_cb, - callback_context, - &callback_context->sa_query); - if (ret < 0) { - pr_info("%s failed to send ib_sa_classport_info query, ret: %d\n", - priv->dev->name, ret); - kfree(callback_context); - return ret; - } - - /* waiting for the callback to finish before returnning */ - wait_for_completion(&callback_context->done); - kfree(callback_context); - - return ret; -} - static void push_pseudo_header(struct sk_buff *skb, const char *daddr) { struct ipoib_pseudo_header *phdr; @@ -731,7 +687,7 @@ static void push_pseudo_header(struct sk_buff *skb, const char *daddr) void ipoib_flush_paths(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_path *path, *tp; LIST_HEAD(remove_list); unsigned long flags; @@ -765,7 +721,7 @@ static void path_rec_completion(int status, { struct ipoib_path *path = path_ptr; struct net_device *dev = path->dev; - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_ah *ah = NULL; struct ipoib_ah *old_ah = NULL; struct ipoib_neigh *neigh, *tn; @@ -858,7 +814,7 @@ static void path_rec_completion(int status, static struct ipoib_path *path_rec_create(struct net_device *dev, void *gid) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_path *path; if (!priv->broadcast) @@ -886,7 +842,7 @@ static struct ipoib_path *path_rec_create(struct net_device *dev, void *gid) static int path_rec_start(struct net_device *dev, struct ipoib_path *path) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); ipoib_dbg(priv, "Start path record lookup for %pI6\n", path->pathrec.dgid.raw); @@ -917,7 +873,8 @@ static int path_rec_start(struct net_device *dev, static void neigh_add_path(struct sk_buff *skb, u8 *daddr, struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); + struct rdma_netdev *rn = netdev_priv(dev); struct ipoib_path *path; struct ipoib_neigh *neigh; unsigned long flags; @@ -964,7 +921,8 @@ static void neigh_add_path(struct sk_buff *skb, u8 *daddr, } } else { spin_unlock_irqrestore(&priv->lock, flags); - ipoib_send(dev, skb, path->ah, IPOIB_QPN(daddr)); + path->ah->last_send = rn->send(dev, skb, path->ah->ah, + IPOIB_QPN(daddr)); ipoib_neigh_put(neigh); return; } @@ -998,7 +956,8 @@ err_drop: static void unicast_arp_send(struct sk_buff *skb, struct net_device *dev, struct ipoib_pseudo_header *phdr) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); + struct rdma_netdev *rn = netdev_priv(dev); struct ipoib_path *path; unsigned long flags; @@ -1042,7 +1001,8 @@ static void unicast_arp_send(struct sk_buff *skb, struct net_device *dev, be16_to_cpu(path->pathrec.dlid)); spin_unlock_irqrestore(&priv->lock, flags); - ipoib_send(dev, skb, path->ah, IPOIB_QPN(phdr->hwaddr)); + path->ah->last_send = rn->send(dev, skb, path->ah->ah, + IPOIB_QPN(phdr->hwaddr)); return; } else if ((path->query || !path_rec_start(dev, path)) && skb_queue_len(&path->queue) < IPOIB_MAX_PATH_REC_QUEUE) { @@ -1058,7 +1018,8 @@ static void unicast_arp_send(struct sk_buff *skb, struct net_device *dev, static int ipoib_start_xmit(struct sk_buff *skb, struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); + struct rdma_netdev *rn = netdev_priv(dev); struct ipoib_neigh *neigh; struct ipoib_pseudo_header *phdr; struct ipoib_header *header; @@ -1122,7 +1083,8 @@ send_using_neigh: goto unref; } } else if (neigh->ah) { - ipoib_send(dev, skb, neigh->ah, IPOIB_QPN(phdr->hwaddr)); + neigh->ah->last_send = rn->send(dev, skb, neigh->ah->ah, + IPOIB_QPN(phdr->hwaddr)); goto unref; } @@ -1144,7 +1106,7 @@ unref: static void ipoib_timeout(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); ipoib_warn(priv, "transmit timeout: latency %d msecs\n", jiffies_to_msecs(jiffies - dev_trans_start(dev))); @@ -1178,7 +1140,7 @@ static int ipoib_hard_header(struct sk_buff *skb, static void ipoib_set_mcast_list(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); if (!test_bit(IPOIB_FLAG_OPER_UP, &priv->flags)) { ipoib_dbg(priv, "IPOIB_FLAG_OPER_UP not set"); @@ -1190,7 +1152,7 @@ static void ipoib_set_mcast_list(struct net_device *dev) static int ipoib_get_iflink(const struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); /* parent interface */ if (!test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags)) @@ -1218,7 +1180,7 @@ static u32 ipoib_addr_hash(struct ipoib_neigh_hash *htbl, u8 *daddr) struct ipoib_neigh *ipoib_neigh_get(struct net_device *dev, u8 *daddr) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_neigh_table *ntbl = &priv->ntbl; struct ipoib_neigh_hash *htbl; struct ipoib_neigh *neigh = NULL; @@ -1347,7 +1309,7 @@ static struct ipoib_neigh *ipoib_neigh_ctor(u8 *daddr, struct ipoib_neigh *ipoib_neigh_alloc(u8 *daddr, struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_neigh_table *ntbl = &priv->ntbl; struct ipoib_neigh_hash *htbl; struct ipoib_neigh *neigh; @@ -1404,7 +1366,7 @@ void ipoib_neigh_dtor(struct ipoib_neigh *neigh) { /* neigh reference count was dropprd to zero */ struct net_device *dev = neigh->dev; - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct sk_buff *skb; if (neigh->ah) ipoib_put_ah(neigh->ah); @@ -1414,7 +1376,7 @@ void ipoib_neigh_dtor(struct ipoib_neigh *neigh) } if (ipoib_cm_get(neigh)) ipoib_cm_destroy_tx(ipoib_cm_get(neigh)); - ipoib_dbg(netdev_priv(dev), + ipoib_dbg(ipoib_priv(dev), "neigh free for %06x %pI6\n", IPOIB_QPN(neigh->daddr), neigh->daddr + 4); @@ -1436,7 +1398,7 @@ static void ipoib_neigh_reclaim(struct rcu_head *rp) void ipoib_neigh_free(struct ipoib_neigh *neigh) { struct net_device *dev = neigh->dev; - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_neigh_table *ntbl = &priv->ntbl; struct ipoib_neigh_hash *htbl; struct ipoib_neigh __rcu **np; @@ -1519,7 +1481,7 @@ static void neigh_hash_free_rcu(struct rcu_head *head) void ipoib_del_neighs_by_gid(struct net_device *dev, u8 *gid) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ipoib_neigh_table *ntbl = &priv->ntbl; struct ipoib_neigh_hash *htbl; unsigned long flags; @@ -1605,7 +1567,7 @@ out_unlock: static void ipoib_neigh_hash_uninit(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); int stopped; ipoib_dbg(priv, "ipoib_neigh_hash_uninit\n"); @@ -1622,10 +1584,26 @@ static void ipoib_neigh_hash_uninit(struct net_device *dev) wait_for_completion(&priv->ntbl.deleted); } +void ipoib_dev_uninit_default(struct net_device *dev) +{ + struct ipoib_dev_priv *priv = ipoib_priv(dev); -int ipoib_dev_init(struct net_device *dev, struct ib_device *ca, int port) + ipoib_transport_dev_cleanup(dev); + + ipoib_cm_dev_cleanup(dev); + + kfree(priv->rx_ring); + vfree(priv->tx_ring); + + priv->rx_ring = NULL; + priv->tx_ring = NULL; +} + +static int ipoib_dev_init_default(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); + + netif_napi_add(dev, &priv->napi, ipoib_poll, NAPI_POLL_WEIGHT); /* Allocate RX/TX "rings" to hold queued skbs */ priv->rx_ring = kzalloc(ipoib_recvq_size * sizeof *priv->rx_ring, @@ -1636,46 +1614,111 @@ int ipoib_dev_init(struct net_device *dev, struct ib_device *ca, int port) priv->tx_ring = vzalloc(ipoib_sendq_size * sizeof *priv->tx_ring); if (!priv->tx_ring) { printk(KERN_WARNING "%s: failed to allocate TX ring (%d entries)\n", - ca->name, ipoib_sendq_size); + priv->ca->name, ipoib_sendq_size); goto out_rx_ring_cleanup; } /* priv->tx_head, tx_tail & tx_outstanding are already 0 */ - if (ipoib_ib_dev_init(dev, ca, port)) + if (ipoib_transport_dev_init(dev, priv->ca)) { + pr_warn("%s: ipoib_transport_dev_init failed\n", + priv->ca->name); goto out_tx_ring_cleanup; + } + + /* after qp created set dev address */ + priv->dev->dev_addr[1] = (priv->qp->qp_num >> 16) & 0xff; + priv->dev->dev_addr[2] = (priv->qp->qp_num >> 8) & 0xff; + priv->dev->dev_addr[3] = (priv->qp->qp_num) & 0xff; + + setup_timer(&priv->poll_timer, ipoib_ib_tx_timer_func, + (unsigned long)dev); + + return 0; + +out_tx_ring_cleanup: + vfree(priv->tx_ring); + +out_rx_ring_cleanup: + kfree(priv->rx_ring); + +out: + return -ENOMEM; +} + +int ipoib_dev_init(struct net_device *dev, struct ib_device *ca, int port) +{ + struct ipoib_dev_priv *priv = ipoib_priv(dev); + int ret = -ENOMEM; + + priv->ca = ca; + priv->port = port; + priv->qp = NULL; /* - * Must be after ipoib_ib_dev_init so we can allocate a per - * device wq there and use it here + * the various IPoIB tasks assume they will never race against + * themselves, so always use a single thread workqueue */ - if (ipoib_neigh_hash_init(priv) < 0) + priv->wq = alloc_ordered_workqueue("ipoib_wq", WQ_MEM_RECLAIM); + if (!priv->wq) { + pr_warn("%s: failed to allocate device WQ\n", dev->name); + goto out; + } + + /* create pd, which used both for control and datapath*/ + priv->pd = ib_alloc_pd(priv->ca, 0); + if (IS_ERR(priv->pd)) { + pr_warn("%s: failed to allocate PD\n", ca->name); + goto clean_wq; + } + + ret = priv->rn_ops->ndo_init(dev); + if (ret) { + pr_warn("%s failed to init HW resource\n", dev->name); + goto out_free_pd; + } + + if (ipoib_neigh_hash_init(priv) < 0) { + pr_warn("%s failed to init neigh hash\n", dev->name); goto out_dev_uninit; + } + + if (dev->flags & IFF_UP) { + if (ipoib_ib_dev_open(dev)) { + pr_warn("%s failed to open device\n", dev->name); + ret = -ENODEV; + goto out_dev_uninit; + } + } return 0; out_dev_uninit: ipoib_ib_dev_cleanup(dev); -out_tx_ring_cleanup: - vfree(priv->tx_ring); +out_free_pd: + if (priv->pd) { + ib_dealloc_pd(priv->pd); + priv->pd = NULL; + } -out_rx_ring_cleanup: - kfree(priv->rx_ring); +clean_wq: + if (priv->wq) { + destroy_workqueue(priv->wq); + priv->wq = NULL; + } out: - return -ENOMEM; + return ret; } void ipoib_dev_cleanup(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev), *cpriv, *tcpriv; + struct ipoib_dev_priv *priv = ipoib_priv(dev), *cpriv, *tcpriv; LIST_HEAD(head); ASSERT_RTNL(); - ipoib_delete_debug_files(dev); - /* Delete any child interfaces first */ list_for_each_entry_safe(cpriv, tcpriv, &priv->child_intfs, list) { /* Stop GC on child */ @@ -1685,24 +1728,21 @@ void ipoib_dev_cleanup(struct net_device *dev) } unregister_netdevice_many(&head); - /* - * Must be before ipoib_ib_dev_cleanup or we delete an in use - * work queue - */ ipoib_neigh_hash_uninit(dev); ipoib_ib_dev_cleanup(dev); - kfree(priv->rx_ring); - vfree(priv->tx_ring); - - priv->rx_ring = NULL; - priv->tx_ring = NULL; + /* no more works over the priv->wq */ + if (priv->wq) { + flush_workqueue(priv->wq); + destroy_workqueue(priv->wq); + priv->wq = NULL; + } } static int ipoib_set_vf_link_state(struct net_device *dev, int vf, int link_state) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); return ib_set_vf_link_state(priv->ca, vf, priv->port, link_state); } @@ -1710,7 +1750,7 @@ static int ipoib_set_vf_link_state(struct net_device *dev, int vf, int link_stat static int ipoib_get_vf_config(struct net_device *dev, int vf, struct ifla_vf_info *ivf) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); int err; err = ib_get_vf_config(priv->ca, vf, priv->port, ivf); @@ -1724,7 +1764,7 @@ static int ipoib_get_vf_config(struct net_device *dev, int vf, static int ipoib_set_vf_guid(struct net_device *dev, int vf, u64 guid, int type) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); if (type != IFLA_VF_IB_NODE_GUID && type != IFLA_VF_IB_PORT_GUID) return -EINVAL; @@ -1735,7 +1775,7 @@ static int ipoib_set_vf_guid(struct net_device *dev, int vf, u64 guid, int type) static int ipoib_get_vf_stats(struct net_device *dev, int vf, struct ifla_vf_stats *vf_stats) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); return ib_get_vf_stats(priv->ca, vf, priv->port, vf_stats); } @@ -1773,21 +1813,12 @@ static const struct net_device_ops ipoib_netdev_ops_vf = { .ndo_get_iflink = ipoib_get_iflink, }; -void ipoib_setup(struct net_device *dev) +void ipoib_setup_common(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); - - if (priv->hca_caps & IB_DEVICE_VIRTUAL_FUNCTION) - dev->netdev_ops = &ipoib_netdev_ops_vf; - else - dev->netdev_ops = &ipoib_netdev_ops_pf; - dev->header_ops = &ipoib_header_ops; ipoib_set_ethtool_ops(dev); - netif_napi_add(dev, &priv->napi, ipoib_poll, NAPI_POLL_WEIGHT); - dev->watchdog_timeo = HZ; dev->flags |= IFF_BROADCAST | IFF_MULTICAST; @@ -1801,11 +1832,14 @@ void ipoib_setup(struct net_device *dev) netif_keep_dst(dev); memcpy(dev->broadcast, ipv4_bcast_addr, INFINIBAND_ALEN); +} - priv->dev = dev; +static void ipoib_build_priv(struct net_device *dev) +{ + struct ipoib_dev_priv *priv = ipoib_priv(dev); + priv->dev = dev; spin_lock_init(&priv->lock); - init_rwsem(&priv->vlan_rwsem); INIT_LIST_HEAD(&priv->path_list); @@ -1823,22 +1857,99 @@ void ipoib_setup(struct net_device *dev) INIT_DELAYED_WORK(&priv->neigh_reap_task, ipoib_reap_neigh); } -struct ipoib_dev_priv *ipoib_intf_alloc(const char *name) +static const struct net_device_ops ipoib_netdev_default_pf = { + .ndo_init = ipoib_dev_init_default, + .ndo_uninit = ipoib_dev_uninit_default, + .ndo_open = ipoib_ib_dev_open_default, + .ndo_stop = ipoib_ib_dev_stop_default, +}; + +static struct net_device +*ipoib_create_netdev_default(struct ib_device *hca, + const char *name, + unsigned char name_assign_type, + void (*setup)(struct net_device *)) { struct net_device *dev; + struct rdma_netdev *rn; - dev = alloc_netdev((int)sizeof(struct ipoib_dev_priv), name, - NET_NAME_UNKNOWN, ipoib_setup); + dev = alloc_netdev((int)sizeof(struct rdma_netdev), + name, + name_assign_type, setup); if (!dev) return NULL; - return netdev_priv(dev); + rn = netdev_priv(dev); + + rn->send = ipoib_send; + rn->attach_mcast = ipoib_mcast_attach; + rn->detach_mcast = ipoib_mcast_detach; + rn->hca = hca; + + dev->netdev_ops = &ipoib_netdev_default_pf; + + return dev; +} + +static struct net_device *ipoib_get_netdev(struct ib_device *hca, u8 port, + const char *name) +{ + struct net_device *dev; + + if (hca->alloc_rdma_netdev) { + dev = hca->alloc_rdma_netdev(hca, port, + RDMA_NETDEV_IPOIB, name, + NET_NAME_UNKNOWN, + ipoib_setup_common); + if (IS_ERR_OR_NULL(dev) && PTR_ERR(dev) != -EOPNOTSUPP) + return NULL; + } + + if (!hca->alloc_rdma_netdev || PTR_ERR(dev) == -EOPNOTSUPP) + dev = ipoib_create_netdev_default(hca, name, NET_NAME_UNKNOWN, + ipoib_setup_common); + + return dev; +} + +struct ipoib_dev_priv *ipoib_intf_alloc(struct ib_device *hca, u8 port, + const char *name) +{ + struct net_device *dev; + struct ipoib_dev_priv *priv; + struct rdma_netdev *rn; + + priv = kzalloc(sizeof(*priv), GFP_KERNEL); + if (!priv) + return NULL; + + dev = ipoib_get_netdev(hca, port, name); + if (!dev) + goto free_priv; + + priv->rn_ops = dev->netdev_ops; + + /* fixme : should be after the query_cap */ + if (priv->hca_caps & IB_DEVICE_VIRTUAL_FUNCTION) + dev->netdev_ops = &ipoib_netdev_ops_vf; + else + dev->netdev_ops = &ipoib_netdev_ops_pf; + + rn = netdev_priv(dev); + rn->clnt_priv = priv; + ipoib_build_priv(dev); + + return priv; +free_priv: + kfree(priv); + return NULL; } static ssize_t show_pkey(struct device *dev, struct device_attribute *attr, char *buf) { - struct ipoib_dev_priv *priv = netdev_priv(to_net_dev(dev)); + struct net_device *ndev = to_net_dev(dev); + struct ipoib_dev_priv *priv = ipoib_priv(ndev); return sprintf(buf, "0x%04x\n", priv->pkey); } @@ -1847,14 +1958,15 @@ static DEVICE_ATTR(pkey, S_IRUGO, show_pkey, NULL); static ssize_t show_umcast(struct device *dev, struct device_attribute *attr, char *buf) { - struct ipoib_dev_priv *priv = netdev_priv(to_net_dev(dev)); + struct net_device *ndev = to_net_dev(dev); + struct ipoib_dev_priv *priv = ipoib_priv(ndev); return sprintf(buf, "%d\n", test_bit(IPOIB_FLAG_UMCAST, &priv->flags)); } void ipoib_set_umcast(struct net_device *ndev, int umcast_val) { - struct ipoib_dev_priv *priv = netdev_priv(ndev); + struct ipoib_dev_priv *priv = ipoib_priv(ndev); if (umcast_val > 0) { set_bit(IPOIB_FLAG_UMCAST, &priv->flags); @@ -1927,7 +2039,7 @@ static int ipoib_check_lladdr(struct net_device *dev, static int ipoib_set_mac(struct net_device *dev, void *addr) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct sockaddr_storage *ss = addr; int ret; @@ -2000,7 +2112,7 @@ void ipoib_set_dev_features(struct ipoib_dev_priv *priv, struct ib_device *hca) priv->hca_caps = hca->attrs.device_cap_flags; if (priv->hca_caps & IB_DEVICE_UD_IP_CSUM) { - priv->dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_RXCSUM; + priv->dev->hw_features |= NETIF_F_IP_CSUM | NETIF_F_RXCSUM; if (priv->hca_caps & IB_DEVICE_UD_TSO) priv->dev->hw_features |= NETIF_F_TSO; @@ -2016,7 +2128,7 @@ static struct net_device *ipoib_add_port(const char *format, struct ib_port_attr attr; int result = -ENOMEM; - priv = ipoib_intf_alloc(format); + priv = ipoib_intf_alloc(hca, port, format); if (!priv) goto alloc_mem_failed; @@ -2090,8 +2202,6 @@ static struct net_device *ipoib_add_port(const char *format, goto register_failed; } - ipoib_create_debug_files(priv->dev); - if (ipoib_cm_add_mode_attr(priv->dev)) goto sysfs_failed; if (ipoib_add_pkey_attr(priv->dev)) @@ -2106,7 +2216,6 @@ static struct net_device *ipoib_add_port(const char *format, return priv->dev; sysfs_failed: - ipoib_delete_debug_files(priv->dev); unregister_netdev(priv->dev); register_failed: @@ -2146,7 +2255,7 @@ static void ipoib_add_one(struct ib_device *device) continue; dev = ipoib_add_port("ib%d", device, p); if (!IS_ERR(dev)) { - priv = netdev_priv(dev); + priv = ipoib_priv(dev); list_add_tail(&priv->list, dev_list); count++; } @@ -2186,11 +2295,18 @@ static void ipoib_remove_one(struct ib_device *device, void *client_data) unregister_netdev(priv->dev); free_netdev(priv->dev); + kfree(priv); } kfree(dev_list); } +#ifdef CONFIG_INFINIBAND_IPOIB_DEBUG +static struct notifier_block ipoib_netdev_notifier = { + .notifier_call = ipoib_netdev_event, +}; +#endif + static int __init ipoib_init_module(void) { int ret; @@ -2243,6 +2359,9 @@ static int __init ipoib_init_module(void) if (ret) goto err_client; +#ifdef CONFIG_INFINIBAND_IPOIB_DEBUG + register_netdevice_notifier(&ipoib_netdev_notifier); +#endif return 0; err_client: @@ -2260,6 +2379,9 @@ err_fs: static void __exit ipoib_cleanup_module(void) { +#ifdef CONFIG_INFINIBAND_IPOIB_DEBUG + unregister_netdevice_notifier(&ipoib_netdev_notifier); +#endif ipoib_netlink_fini(); ib_unregister_client(&ipoib_client); ib_sa_unregister_client(&ipoib_sa_client); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c index 69e146cdc306..3385869443ec 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c @@ -114,7 +114,7 @@ static void ipoib_mcast_free(struct ipoib_mcast *mcast) struct net_device *dev = mcast->dev; int tx_dropped = 0; - ipoib_dbg_mcast(netdev_priv(dev), "deleting multicast group %pI6\n", + ipoib_dbg_mcast(ipoib_priv(dev), "deleting multicast group %pI6\n", mcast->mcmember.mgid.raw); /* remove all neigh connected to this mcast */ @@ -158,7 +158,7 @@ static struct ipoib_mcast *ipoib_mcast_alloc(struct net_device *dev, static struct ipoib_mcast *__ipoib_mcast_find(struct net_device *dev, void *mgid) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct rb_node *n = priv->multicast_tree.rb_node; while (n) { @@ -182,7 +182,7 @@ static struct ipoib_mcast *__ipoib_mcast_find(struct net_device *dev, void *mgid static int __ipoib_mcast_add(struct net_device *dev, struct ipoib_mcast *mcast) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct rb_node **n = &priv->multicast_tree.rb_node, *pn = NULL; while (*n) { @@ -212,7 +212,8 @@ static int ipoib_mcast_join_finish(struct ipoib_mcast *mcast, struct ib_sa_mcmember_rec *mcmember) { struct net_device *dev = mcast->dev; - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); + struct rdma_netdev *rn = netdev_priv(dev); struct ipoib_ah *ah; int ret; int set_qkey = 0; @@ -260,8 +261,9 @@ static int ipoib_mcast_join_finish(struct ipoib_mcast *mcast, return 0; } - ret = ipoib_mcast_attach(dev, be16_to_cpu(mcast->mcmember.mlid), - &mcast->mcmember.mgid, set_qkey); + ret = rn->attach_mcast(dev, priv->ca, &mcast->mcmember.mgid, + be16_to_cpu(mcast->mcmember.mlid), + set_qkey, priv->qkey); if (ret < 0) { ipoib_warn(priv, "couldn't attach QP to multicast group %pI6\n", mcast->mcmember.mgid.raw); @@ -331,7 +333,6 @@ void ipoib_mcast_carrier_on_task(struct work_struct *work) struct ipoib_dev_priv *priv = container_of(work, struct ipoib_dev_priv, carrier_on_task); struct ib_port_attr attr; - int ret; if (ib_query_port(priv->ca, priv->port, &attr) || attr.state != IB_PORT_ACTIVE) { @@ -344,11 +345,9 @@ void ipoib_mcast_carrier_on_task(struct work_struct *work) * because the broadcast group must always be joined first and is always * re-joined if the SM changes substantially. */ - ret = ipoib_check_sm_sendonly_fullmember_support(priv); - if (ret < 0) - pr_debug("%s failed query sm support for sendonly-fullmember (ret: %d)\n", - priv->dev->name, ret); - + priv->sm_fullmember_sendonly_support = + ib_sa_sendonly_fullmem_support(&ipoib_sa_client, + priv->ca, priv->port); /* * Take rtnl_lock to avoid racing with ipoib_stop() and * turning the carrier back on while a device is being @@ -375,7 +374,7 @@ static int ipoib_mcast_join_complete(int status, { struct ipoib_mcast *mcast = multicast->context; struct net_device *dev = mcast->dev; - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); ipoib_dbg_mcast(priv, "%sjoin completion for %pI6 (status %d)\n", test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags) ? @@ -477,7 +476,7 @@ out_locked: */ static int ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ib_sa_multicast *multicast; struct ib_sa_mcmember_rec rec = { .join_state = 1 @@ -489,6 +488,9 @@ static int ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast) !test_bit(IPOIB_FLAG_OPER_UP, &priv->flags)) return -EINVAL; + init_completion(&mcast->done); + set_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags); + ipoib_dbg_mcast(priv, "joining MGID %pI6\n", mcast->mcmember.mgid.raw); rec.mgid = mcast->mcmember.mgid; @@ -647,8 +649,6 @@ void ipoib_mcast_join_task(struct work_struct *work) if (mcast->backoff == 1 || time_after_eq(jiffies, mcast->delay_until)) { /* Found the next unjoined group */ - init_completion(&mcast->done); - set_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags); if (ipoib_mcast_join(dev, mcast)) { spin_unlock_irq(&priv->lock); return; @@ -668,17 +668,15 @@ out: queue_delayed_work(priv->wq, &priv->mcast_task, delay_until - jiffies); } - if (mcast) { - init_completion(&mcast->done); - set_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags); + if (mcast) ipoib_mcast_join(dev, mcast); - } + spin_unlock_irq(&priv->lock); } void ipoib_mcast_start_thread(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); unsigned long flags; ipoib_dbg_mcast(priv, "starting multicast thread\n"); @@ -690,7 +688,7 @@ void ipoib_mcast_start_thread(struct net_device *dev) int ipoib_mcast_stop_thread(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); unsigned long flags; ipoib_dbg_mcast(priv, "stopping multicast thread\n"); @@ -706,7 +704,8 @@ int ipoib_mcast_stop_thread(struct net_device *dev) static int ipoib_mcast_leave(struct net_device *dev, struct ipoib_mcast *mcast) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); + struct rdma_netdev *rn = netdev_priv(dev); int ret = 0; if (test_and_clear_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags)) @@ -720,8 +719,8 @@ static int ipoib_mcast_leave(struct net_device *dev, struct ipoib_mcast *mcast) mcast->mcmember.mgid.raw); /* Remove ourselves from the multicast group */ - ret = ib_detach_mcast(priv->qp, &mcast->mcmember.mgid, - be16_to_cpu(mcast->mcmember.mlid)); + ret = rn->detach_mcast(dev, priv->ca, &mcast->mcmember.mgid, + be16_to_cpu(mcast->mcmember.mlid)); if (ret) ipoib_warn(priv, "ib_detach_mcast failed (result = %d)\n", ret); } else if (!test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags)) @@ -762,7 +761,8 @@ void ipoib_mcast_remove_list(struct list_head *remove_list) void ipoib_mcast_send(struct net_device *dev, u8 *daddr, struct sk_buff *skb) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); + struct rdma_netdev *rn = netdev_priv(dev); struct ipoib_mcast *mcast; unsigned long flags; void *mgid = daddr + 4; @@ -825,7 +825,8 @@ void ipoib_mcast_send(struct net_device *dev, u8 *daddr, struct sk_buff *skb) } } spin_unlock_irqrestore(&priv->lock, flags); - ipoib_send(dev, skb, mcast->ah, IB_MULTICAST_QPN); + mcast->ah->last_send = rn->send(dev, skb, mcast->ah->ah, + IB_MULTICAST_QPN); if (neigh) ipoib_neigh_put(neigh); return; @@ -837,7 +838,7 @@ unlock: void ipoib_mcast_dev_flush(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); LIST_HEAD(remove_list); struct ipoib_mcast *mcast, *tmcast; unsigned long flags; @@ -1029,7 +1030,7 @@ struct ipoib_mcast_iter *ipoib_mcast_iter_init(struct net_device *dev) int ipoib_mcast_iter_next(struct ipoib_mcast_iter *iter) { - struct ipoib_dev_priv *priv = netdev_priv(iter->dev); + struct ipoib_dev_priv *priv = ipoib_priv(iter->dev); struct rb_node *n; struct ipoib_mcast *mcast; int ret = 1; diff --git a/drivers/infiniband/ulp/ipoib/ipoib_netlink.c b/drivers/infiniband/ulp/ipoib/ipoib_netlink.c index cdc7df4fdb8a..28884781311b 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_netlink.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_netlink.c @@ -44,7 +44,7 @@ static const struct nla_policy ipoib_policy[IFLA_IPOIB_MAX + 1] = { static int ipoib_fill_info(struct sk_buff *skb, const struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); u16 val; if (nla_put_u16(skb, IFLA_IPOIB_PKEY, priv->pkey)) @@ -107,7 +107,7 @@ static int ipoib_new_child_link(struct net *src_net, struct net_device *dev, if (!pdev || pdev->type != ARPHRD_INFINIBAND) return -ENODEV; - ppriv = netdev_priv(pdev); + ppriv = ipoib_priv(pdev); if (test_bit(IPOIB_FLAG_SUBINTERFACE, &ppriv->flags)) { ipoib_warn(ppriv, "child creation disallowed for child devices\n"); @@ -129,7 +129,8 @@ static int ipoib_new_child_link(struct net *src_net, struct net_device *dev, */ child_pkey |= 0x8000; - err = __ipoib_vlan_add(ppriv, netdev_priv(dev), child_pkey, IPOIB_RTNL_CHILD); + err = __ipoib_vlan_add(ppriv, ipoib_priv(dev), + child_pkey, IPOIB_RTNL_CHILD); if (!err && data) err = ipoib_changelink(dev, tb, data); @@ -140,8 +141,8 @@ static void ipoib_unregister_child_dev(struct net_device *dev, struct list_head { struct ipoib_dev_priv *priv, *ppriv; - priv = netdev_priv(dev); - ppriv = netdev_priv(priv->parent); + priv = ipoib_priv(dev); + ppriv = ipoib_priv(priv->parent); down_write(&ppriv->vlan_rwsem); unregister_netdevice_queue(dev, head); @@ -161,7 +162,7 @@ static struct rtnl_link_ops ipoib_link_ops __read_mostly = { .maxtype = IFLA_IPOIB_MAX, .policy = ipoib_policy, .priv_size = sizeof(struct ipoib_dev_priv), - .setup = ipoib_setup, + .setup = ipoib_setup_common, .newlink = ipoib_new_child_link, .changelink = ipoib_changelink, .dellink = ipoib_unregister_child_dev, diff --git a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c index 189dcd1709d2..bb64baf25309 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c @@ -35,9 +35,10 @@ #include "ipoib.h" -int ipoib_mcast_attach(struct net_device *dev, u16 mlid, union ib_gid *mgid, int set_qkey) +int ipoib_mcast_attach(struct net_device *dev, struct ib_device *hca, + union ib_gid *mgid, u16 mlid, int set_qkey, u32 qkey) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ib_qp_attr *qp_attr = NULL; int ret; u16 pkey_index; @@ -56,7 +57,7 @@ int ipoib_mcast_attach(struct net_device *dev, u16 mlid, union ib_gid *mgid, int goto out; /* set correct QKey for QP */ - qp_attr->qkey = priv->qkey; + qp_attr->qkey = qkey; ret = ib_modify_qp(priv->qp, qp_attr, IB_QP_QKEY); if (ret) { ipoib_warn(priv, "failed to modify QP, ret = %d\n", ret); @@ -74,9 +75,20 @@ out: return ret; } +int ipoib_mcast_detach(struct net_device *dev, struct ib_device *hca, + union ib_gid *mgid, u16 mlid) +{ + struct ipoib_dev_priv *priv = ipoib_priv(dev); + int ret; + + ret = ib_detach_mcast(priv->qp, mgid, mlid); + + return ret; +} + int ipoib_init_qp(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); int ret; struct ib_qp_attr qp_attr; int attr_mask; @@ -130,7 +142,7 @@ out_fail: int ipoib_transport_dev_init(struct net_device *dev, struct ib_device *ca) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); struct ib_qp_init_attr init_attr = { .cap = { .max_send_wr = ipoib_sendq_size, @@ -147,22 +159,6 @@ int ipoib_transport_dev_init(struct net_device *dev, struct ib_device *ca) int ret, size; int i; - priv->pd = ib_alloc_pd(priv->ca, 0); - if (IS_ERR(priv->pd)) { - printk(KERN_WARNING "%s: failed to allocate PD\n", ca->name); - return -ENODEV; - } - - /* - * the various IPoIB tasks assume they will never race against - * themselves, so always use a single thread workqueue - */ - priv->wq = alloc_ordered_workqueue("ipoib_wq", WQ_MEM_RECLAIM); - if (!priv->wq) { - printk(KERN_WARNING "ipoib: failed to allocate device WQ\n"); - goto out_free_pd; - } - size = ipoib_recvq_size + 1; ret = ipoib_cm_dev_init(dev); if (!ret) { @@ -173,7 +169,7 @@ int ipoib_transport_dev_init(struct net_device *dev, struct ib_device *ca) size += ipoib_recvq_size * ipoib_max_conn_qp; } else if (ret != -ENOSYS) - goto out_free_wq; + return -ENODEV; cq_attr.cqe = size; priv->recv_cq = ib_create_cq(priv->ca, ipoib_ib_completion, NULL, @@ -212,10 +208,6 @@ int ipoib_transport_dev_init(struct net_device *dev, struct ib_device *ca) goto out_free_send_cq; } - priv->dev->dev_addr[1] = (priv->qp->qp_num >> 16) & 0xff; - priv->dev->dev_addr[2] = (priv->qp->qp_num >> 8) & 0xff; - priv->dev->dev_addr[3] = (priv->qp->qp_num ) & 0xff; - for (i = 0; i < MAX_SKB_FRAGS + 1; ++i) priv->tx_sge[i].lkey = priv->pd->local_dma_lkey; @@ -247,26 +239,18 @@ out_free_recv_cq: out_cm_dev_cleanup: ipoib_cm_dev_cleanup(dev); -out_free_wq: - destroy_workqueue(priv->wq); - priv->wq = NULL; - -out_free_pd: - ib_dealloc_pd(priv->pd); - return -ENODEV; } void ipoib_transport_dev_cleanup(struct net_device *dev) { - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); if (priv->qp) { if (ib_destroy_qp(priv->qp)) ipoib_warn(priv, "ib_qp_destroy failed\n"); priv->qp = NULL; - clear_bit(IPOIB_PKEY_ASSIGNED, &priv->flags); } if (ib_destroy_cq(priv->send_cq)) @@ -274,16 +258,6 @@ void ipoib_transport_dev_cleanup(struct net_device *dev) if (ib_destroy_cq(priv->recv_cq)) ipoib_warn(priv, "ib_cq_destroy (recv) failed\n"); - - ipoib_cm_dev_cleanup(dev); - - if (priv->wq) { - flush_workqueue(priv->wq); - destroy_workqueue(priv->wq); - priv->wq = NULL; - } - - ib_dealloc_pd(priv->pd); } void ipoib_event(struct ib_event_handler *handler, diff --git a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c index 3e10e3dac2e7..36dc4fcaa3cd 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c @@ -44,7 +44,7 @@ static ssize_t show_parent(struct device *d, struct device_attribute *attr, char *buf) { struct net_device *dev = to_net_dev(d); - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = ipoib_priv(dev); return sprintf(buf, "%s\n", priv->parent->name); } @@ -86,8 +86,6 @@ int __ipoib_vlan_add(struct ipoib_dev_priv *ppriv, struct ipoib_dev_priv *priv, goto register_failed; } - ipoib_create_debug_files(priv->dev); - /* RTNL childs don't need proprietary sysfs entries */ if (type == IPOIB_LEGACY_CHILD) { if (ipoib_cm_add_mode_attr(priv->dev)) @@ -108,7 +106,6 @@ int __ipoib_vlan_add(struct ipoib_dev_priv *ppriv, struct ipoib_dev_priv *priv, sysfs_failed: result = -ENOMEM; - ipoib_delete_debug_files(priv->dev); unregister_netdevice(priv->dev); register_failed: @@ -128,14 +125,15 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey) if (!capable(CAP_NET_ADMIN)) return -EPERM; - ppriv = netdev_priv(pdev); + ppriv = ipoib_priv(pdev); if (test_bit(IPOIB_FLAG_GOING_DOWN, &ppriv->flags)) return -EPERM; snprintf(intf_name, sizeof intf_name, "%s.%04x", ppriv->dev->name, pkey); - priv = ipoib_intf_alloc(intf_name); + + priv = ipoib_intf_alloc(ppriv->ca, ppriv->port, intf_name); if (!priv) return -ENOMEM; @@ -183,7 +181,7 @@ int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey) if (!capable(CAP_NET_ADMIN)) return -EPERM; - ppriv = netdev_priv(pdev); + ppriv = ipoib_priv(pdev); if (test_bit(IPOIB_FLAG_GOING_DOWN, &ppriv->flags)) return -EPERM; diff --git a/drivers/infiniband/ulp/iser/iser_initiator.c b/drivers/infiniband/ulp/iser/iser_initiator.c index 81ae2e30dd12..12ed62ce9ff7 100644 --- a/drivers/infiniband/ulp/iser/iser_initiator.c +++ b/drivers/infiniband/ulp/iser/iser_initiator.c @@ -612,7 +612,7 @@ iser_check_remote_inv(struct iser_conn *iser_conn, iser_conn, rkey); if (unlikely(!iser_conn->snd_w_inv)) { - iser_err("conn %p: unexepected remote invalidation, " + iser_err("conn %p: unexpected remote invalidation, " "terminating connection\n", iser_conn); return -EPROTO; } diff --git a/drivers/infiniband/ulp/opa_vnic/Kconfig b/drivers/infiniband/ulp/opa_vnic/Kconfig new file mode 100644 index 000000000000..48132ab5e6b9 --- /dev/null +++ b/drivers/infiniband/ulp/opa_vnic/Kconfig @@ -0,0 +1,8 @@ +config INFINIBAND_OPA_VNIC + tristate "Intel OPA VNIC support" + depends on X86_64 && INFINIBAND + ---help--- + This is Omni-Path (OPA) Virtual Network Interface Controller (VNIC) + driver for Ethernet over Omni-Path feature. It implements the HW + independent VNIC functionality. It interfaces with Linux stack for + data path and IB MAD for the control path. diff --git a/drivers/infiniband/ulp/opa_vnic/Makefile b/drivers/infiniband/ulp/opa_vnic/Makefile new file mode 100644 index 000000000000..8061b287cfe4 --- /dev/null +++ b/drivers/infiniband/ulp/opa_vnic/Makefile @@ -0,0 +1,7 @@ +# Makefile - Intel Omni-Path Virtual Network Controller driver +# Copyright(c) 2017, Intel Corporation. +# +obj-$(CONFIG_INFINIBAND_OPA_VNIC) += opa_vnic.o + +opa_vnic-y := opa_vnic_netdev.o opa_vnic_encap.o opa_vnic_ethtool.o \ + opa_vnic_vema.o opa_vnic_vema_iface.o diff --git a/drivers/infiniband/ulp/opa_vnic/opa_vnic_encap.c b/drivers/infiniband/ulp/opa_vnic/opa_vnic_encap.c new file mode 100644 index 000000000000..2e8fee982436 --- /dev/null +++ b/drivers/infiniband/ulp/opa_vnic/opa_vnic_encap.c @@ -0,0 +1,475 @@ +/* + * Copyright(c) 2017 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ + +/* + * This file contains OPA VNIC encapsulation/decapsulation function. + */ + +#include <linux/if_ether.h> +#include <linux/if_vlan.h> + +#include "opa_vnic_internal.h" + +/* OPA 16B Header fields */ +#define OPA_16B_LID_MASK 0xFFFFFull +#define OPA_16B_SLID_HIGH_SHFT 8 +#define OPA_16B_SLID_MASK 0xF00ull +#define OPA_16B_DLID_MASK 0xF000ull +#define OPA_16B_DLID_HIGH_SHFT 12 +#define OPA_16B_LEN_SHFT 20 +#define OPA_16B_SC_SHFT 20 +#define OPA_16B_RC_SHFT 25 +#define OPA_16B_PKEY_SHFT 16 + +#define OPA_VNIC_L4_HDR_SHFT 16 + +/* L2+L4 hdr len is 20 bytes (5 quad words) */ +#define OPA_VNIC_HDR_QW_LEN 5 + +static inline void opa_vnic_make_header(u8 *hdr, u32 slid, u32 dlid, u16 len, + u16 pkey, u16 entropy, u8 sc, u8 rc, + u8 l4_type, u16 l4_hdr) +{ + /* h[1]: LT=1, 16B L2=10 */ + u32 h[OPA_VNIC_HDR_QW_LEN] = {0, 0xc0000000, 0, 0, 0}; + + h[2] = l4_type; + h[3] = entropy; + h[4] = l4_hdr << OPA_VNIC_L4_HDR_SHFT; + + /* Extract and set 4 upper bits and 20 lower bits of the lids */ + h[0] |= (slid & OPA_16B_LID_MASK); + h[2] |= ((slid >> (20 - OPA_16B_SLID_HIGH_SHFT)) & OPA_16B_SLID_MASK); + + h[1] |= (dlid & OPA_16B_LID_MASK); + h[2] |= ((dlid >> (20 - OPA_16B_DLID_HIGH_SHFT)) & OPA_16B_DLID_MASK); + + h[0] |= (len << OPA_16B_LEN_SHFT); + h[1] |= (rc << OPA_16B_RC_SHFT); + h[1] |= (sc << OPA_16B_SC_SHFT); + h[2] |= ((u32)pkey << OPA_16B_PKEY_SHFT); + + memcpy(hdr, h, OPA_VNIC_HDR_LEN); +} + +/* + * Using a simple hash table for mac table implementation with the last octet + * of mac address as a key. + */ +static void opa_vnic_free_mac_tbl(struct hlist_head *mactbl) +{ + struct opa_vnic_mac_tbl_node *node; + struct hlist_node *tmp; + int bkt; + + if (!mactbl) + return; + + vnic_hash_for_each_safe(mactbl, bkt, tmp, node, hlist) { + hash_del(&node->hlist); + kfree(node); + } + kfree(mactbl); +} + +static struct hlist_head *opa_vnic_alloc_mac_tbl(void) +{ + u32 size = sizeof(struct hlist_head) * OPA_VNIC_MAC_TBL_SIZE; + struct hlist_head *mactbl; + + mactbl = kzalloc(size, GFP_KERNEL); + if (!mactbl) + return ERR_PTR(-ENOMEM); + + vnic_hash_init(mactbl); + return mactbl; +} + +/* opa_vnic_release_mac_tbl - empty and free the mac table */ +void opa_vnic_release_mac_tbl(struct opa_vnic_adapter *adapter) +{ + struct hlist_head *mactbl; + + mutex_lock(&adapter->mactbl_lock); + mactbl = rcu_access_pointer(adapter->mactbl); + rcu_assign_pointer(adapter->mactbl, NULL); + synchronize_rcu(); + opa_vnic_free_mac_tbl(mactbl); + mutex_unlock(&adapter->mactbl_lock); +} + +/* + * opa_vnic_query_mac_tbl - query the mac table for a section + * + * This function implements query of specific function of the mac table. + * The function also expects the requested range to be valid. + */ +void opa_vnic_query_mac_tbl(struct opa_vnic_adapter *adapter, + struct opa_veswport_mactable *tbl) +{ + struct opa_vnic_mac_tbl_node *node; + struct hlist_head *mactbl; + int bkt; + u16 loffset, lnum_entries; + + rcu_read_lock(); + mactbl = rcu_dereference(adapter->mactbl); + if (!mactbl) + goto get_mac_done; + + loffset = be16_to_cpu(tbl->offset); + lnum_entries = be16_to_cpu(tbl->num_entries); + + vnic_hash_for_each(mactbl, bkt, node, hlist) { + struct __opa_vnic_mactable_entry *nentry = &node->entry; + struct opa_veswport_mactable_entry *entry; + + if ((node->index < loffset) || + (node->index >= (loffset + lnum_entries))) + continue; + + /* populate entry in the tbl corresponding to the index */ + entry = &tbl->tbl_entries[node->index - loffset]; + memcpy(entry->mac_addr, nentry->mac_addr, + ARRAY_SIZE(entry->mac_addr)); + memcpy(entry->mac_addr_mask, nentry->mac_addr_mask, + ARRAY_SIZE(entry->mac_addr_mask)); + entry->dlid_sd = cpu_to_be32(nentry->dlid_sd); + } + tbl->mac_tbl_digest = cpu_to_be32(adapter->info.vport.mac_tbl_digest); +get_mac_done: + rcu_read_unlock(); +} + +/* + * opa_vnic_update_mac_tbl - update mac table section + * + * This function updates the specified section of the mac table. + * The procedure includes following steps. + * - Allocate a new mac (hash) table. + * - Add the specified entries to the new table. + * (except the ones that are requested to be deleted). + * - Add all the other entries from the old mac table. + * - If there is a failure, free the new table and return. + * - Switch to the new table. + * - Free the old table and return. + * + * The function also expects the requested range to be valid. + */ +int opa_vnic_update_mac_tbl(struct opa_vnic_adapter *adapter, + struct opa_veswport_mactable *tbl) +{ + struct opa_vnic_mac_tbl_node *node, *new_node; + struct hlist_head *new_mactbl, *old_mactbl; + int i, bkt, rc = 0; + u8 key; + u16 loffset, lnum_entries; + + mutex_lock(&adapter->mactbl_lock); + /* allocate new mac table */ + new_mactbl = opa_vnic_alloc_mac_tbl(); + if (IS_ERR(new_mactbl)) { + mutex_unlock(&adapter->mactbl_lock); + return PTR_ERR(new_mactbl); + } + + loffset = be16_to_cpu(tbl->offset); + lnum_entries = be16_to_cpu(tbl->num_entries); + + /* add updated entries to the new mac table */ + for (i = 0; i < lnum_entries; i++) { + struct __opa_vnic_mactable_entry *nentry; + struct opa_veswport_mactable_entry *entry = + &tbl->tbl_entries[i]; + u8 *mac_addr = entry->mac_addr; + u8 empty_mac[ETH_ALEN] = { 0 }; + + v_dbg("new mac entry %4d: %02x:%02x:%02x:%02x:%02x:%02x %x\n", + loffset + i, mac_addr[0], mac_addr[1], mac_addr[2], + mac_addr[3], mac_addr[4], mac_addr[5], + entry->dlid_sd); + + /* if the entry is being removed, do not add it */ + if (!memcmp(mac_addr, empty_mac, ARRAY_SIZE(empty_mac))) + continue; + + node = kzalloc(sizeof(*node), GFP_KERNEL); + if (!node) { + rc = -ENOMEM; + goto updt_done; + } + + node->index = loffset + i; + nentry = &node->entry; + memcpy(nentry->mac_addr, entry->mac_addr, + ARRAY_SIZE(nentry->mac_addr)); + memcpy(nentry->mac_addr_mask, entry->mac_addr_mask, + ARRAY_SIZE(nentry->mac_addr_mask)); + nentry->dlid_sd = be32_to_cpu(entry->dlid_sd); + key = node->entry.mac_addr[OPA_VNIC_MAC_HASH_IDX]; + vnic_hash_add(new_mactbl, &node->hlist, key); + } + + /* add other entries from current mac table to new mac table */ + old_mactbl = rcu_access_pointer(adapter->mactbl); + if (!old_mactbl) + goto switch_tbl; + + vnic_hash_for_each(old_mactbl, bkt, node, hlist) { + if ((node->index >= loffset) && + (node->index < (loffset + lnum_entries))) + continue; + + new_node = kzalloc(sizeof(*new_node), GFP_KERNEL); + if (!new_node) { + rc = -ENOMEM; + goto updt_done; + } + + new_node->index = node->index; + memcpy(&new_node->entry, &node->entry, sizeof(node->entry)); + key = new_node->entry.mac_addr[OPA_VNIC_MAC_HASH_IDX]; + vnic_hash_add(new_mactbl, &new_node->hlist, key); + } + +switch_tbl: + /* switch to new table */ + rcu_assign_pointer(adapter->mactbl, new_mactbl); + synchronize_rcu(); + + adapter->info.vport.mac_tbl_digest = be32_to_cpu(tbl->mac_tbl_digest); +updt_done: + /* upon failure, free the new table; otherwise, free the old table */ + if (rc) + opa_vnic_free_mac_tbl(new_mactbl); + else + opa_vnic_free_mac_tbl(old_mactbl); + + mutex_unlock(&adapter->mactbl_lock); + return rc; +} + +/* opa_vnic_chk_mac_tbl - check mac table for dlid */ +static uint32_t opa_vnic_chk_mac_tbl(struct opa_vnic_adapter *adapter, + struct ethhdr *mac_hdr) +{ + struct opa_vnic_mac_tbl_node *node; + struct hlist_head *mactbl; + u32 dlid = 0; + u8 key; + + rcu_read_lock(); + mactbl = rcu_dereference(adapter->mactbl); + if (unlikely(!mactbl)) + goto chk_done; + + key = mac_hdr->h_dest[OPA_VNIC_MAC_HASH_IDX]; + vnic_hash_for_each_possible(mactbl, node, hlist, key) { + struct __opa_vnic_mactable_entry *entry = &node->entry; + + /* if related to source mac, skip */ + if (unlikely(OPA_VNIC_DLID_SD_IS_SRC_MAC(entry->dlid_sd))) + continue; + + if (!memcmp(node->entry.mac_addr, mac_hdr->h_dest, + ARRAY_SIZE(node->entry.mac_addr))) { + /* mac address found */ + dlid = OPA_VNIC_DLID_SD_GET_DLID(node->entry.dlid_sd); + break; + } + } + +chk_done: + rcu_read_unlock(); + return dlid; +} + +/* opa_vnic_get_dlid - find and return the DLID */ +static uint32_t opa_vnic_get_dlid(struct opa_vnic_adapter *adapter, + struct sk_buff *skb, u8 def_port) +{ + struct __opa_veswport_info *info = &adapter->info; + struct ethhdr *mac_hdr = (struct ethhdr *)skb_mac_header(skb); + u32 dlid; + + dlid = opa_vnic_chk_mac_tbl(adapter, mac_hdr); + if (dlid) + return dlid; + + if (is_multicast_ether_addr(mac_hdr->h_dest)) { + dlid = info->vesw.u_mcast_dlid; + } else { + if (is_local_ether_addr(mac_hdr->h_dest)) { + dlid = ((uint32_t)mac_hdr->h_dest[5] << 16) | + ((uint32_t)mac_hdr->h_dest[4] << 8) | + mac_hdr->h_dest[3]; + if (unlikely(!dlid)) + v_warn("Null dlid in MAC address\n"); + } else if (def_port != OPA_VNIC_INVALID_PORT) { + dlid = info->vesw.u_ucast_dlid[def_port]; + } + } + + return dlid; +} + +/* opa_vnic_get_sc - return the service class */ +static u8 opa_vnic_get_sc(struct __opa_veswport_info *info, + struct sk_buff *skb) +{ + struct ethhdr *mac_hdr = (struct ethhdr *)skb_mac_header(skb); + u16 vlan_tci; + u8 sc; + + if (!__vlan_get_tag(skb, &vlan_tci)) { + u8 pcp = OPA_VNIC_VLAN_PCP(vlan_tci); + + if (is_multicast_ether_addr(mac_hdr->h_dest)) + sc = info->vport.pcp_to_sc_mc[pcp]; + else + sc = info->vport.pcp_to_sc_uc[pcp]; + } else { + if (is_multicast_ether_addr(mac_hdr->h_dest)) + sc = info->vport.non_vlan_sc_mc; + else + sc = info->vport.non_vlan_sc_uc; + } + + return sc; +} + +u8 opa_vnic_get_vl(struct opa_vnic_adapter *adapter, struct sk_buff *skb) +{ + struct ethhdr *mac_hdr = (struct ethhdr *)skb_mac_header(skb); + struct __opa_veswport_info *info = &adapter->info; + u8 vl; + + if (skb_vlan_tag_present(skb)) { + u8 pcp = skb_vlan_tag_get(skb) >> VLAN_PRIO_SHIFT; + + if (is_multicast_ether_addr(mac_hdr->h_dest)) + vl = info->vport.pcp_to_vl_mc[pcp]; + else + vl = info->vport.pcp_to_vl_uc[pcp]; + } else { + if (is_multicast_ether_addr(mac_hdr->h_dest)) + vl = info->vport.non_vlan_vl_mc; + else + vl = info->vport.non_vlan_vl_uc; + } + + return vl; +} + +/* opa_vnic_calc_entropy - calculate the packet entropy */ +u8 opa_vnic_calc_entropy(struct opa_vnic_adapter *adapter, struct sk_buff *skb) +{ + u16 hash16; + + /* + * Get flow based 16-bit hash and then XOR the upper and lower bytes + * to get the entropy. + * __skb_tx_hash limits qcount to 16 bits. Hence, get 15-bit hash. + */ + hash16 = __skb_tx_hash(adapter->netdev, skb, BIT(15)); + return (u8)((hash16 >> 8) ^ (hash16 & 0xff)); +} + +/* opa_vnic_get_def_port - get default port based on entropy */ +static inline u8 opa_vnic_get_def_port(struct opa_vnic_adapter *adapter, + u8 entropy) +{ + u8 flow_id; + + /* Add the upper and lower 4-bits of entropy to get the flow id */ + flow_id = ((entropy & 0xf) + (entropy >> 4)); + return adapter->flow_tbl[flow_id & (OPA_VNIC_FLOW_TBL_SIZE - 1)]; +} + +/* Calculate packet length including OPA header, crc and padding */ +static inline int opa_vnic_wire_length(struct sk_buff *skb) +{ + u32 pad_len; + + /* padding for 8 bytes size alignment */ + pad_len = -(skb->len + OPA_VNIC_ICRC_TAIL_LEN) & 0x7; + pad_len += OPA_VNIC_ICRC_TAIL_LEN; + + return (skb->len + pad_len) >> 3; +} + +/* opa_vnic_encap_skb - encapsulate skb packet with OPA header and meta data */ +void opa_vnic_encap_skb(struct opa_vnic_adapter *adapter, struct sk_buff *skb) +{ + struct __opa_veswport_info *info = &adapter->info; + struct opa_vnic_skb_mdata *mdata; + u8 def_port, sc, entropy, *hdr; + u16 len, l4_hdr; + u32 dlid; + + hdr = skb_push(skb, OPA_VNIC_HDR_LEN); + + entropy = opa_vnic_calc_entropy(adapter, skb); + def_port = opa_vnic_get_def_port(adapter, entropy); + len = opa_vnic_wire_length(skb); + dlid = opa_vnic_get_dlid(adapter, skb, def_port); + 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->vl = opa_vnic_get_vl(adapter, skb); + mdata->entropy = entropy; + mdata->flags = 0; + if (unlikely(!dlid)) { + mdata->flags = OPA_VNIC_SKB_MDATA_ENCAP_ERR; + return; + } + + opa_vnic_make_header(hdr, info->vport.encap_slid, dlid, len, + info->vesw.pkey, entropy, sc, 0, + OPA_VNIC_L4_ETHR, l4_hdr); +} diff --git a/drivers/infiniband/ulp/opa_vnic/opa_vnic_encap.h b/drivers/infiniband/ulp/opa_vnic/opa_vnic_encap.h new file mode 100644 index 000000000000..4c434b9dd84c --- /dev/null +++ b/drivers/infiniband/ulp/opa_vnic/opa_vnic_encap.h @@ -0,0 +1,489 @@ +#ifndef _OPA_VNIC_ENCAP_H +#define _OPA_VNIC_ENCAP_H +/* + * Copyright(c) 2017 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ + +/* + * This file contains all OPA VNIC declaration required for encapsulation + * and decapsulation of Ethernet packets + */ + +#include <linux/types.h> +#include <rdma/ib_mad.h> + +/* EMA class version */ +#define OPA_EMA_CLASS_VERSION 0x80 + +/* + * Define the Intel vendor management class for OPA + * ETHERNET MANAGEMENT + */ +#define OPA_MGMT_CLASS_INTEL_EMA 0x34 + +/* EM attribute IDs */ +#define OPA_EM_ATTR_CLASS_PORT_INFO 0x0001 +#define OPA_EM_ATTR_VESWPORT_INFO 0x0011 +#define OPA_EM_ATTR_VESWPORT_MAC_ENTRIES 0x0012 +#define OPA_EM_ATTR_IFACE_UCAST_MACS 0x0013 +#define OPA_EM_ATTR_IFACE_MCAST_MACS 0x0014 +#define OPA_EM_ATTR_DELETE_VESW 0x0015 +#define OPA_EM_ATTR_VESWPORT_SUMMARY_COUNTERS 0x0020 +#define OPA_EM_ATTR_VESWPORT_ERROR_COUNTERS 0x0022 + +/* VNIC configured and operational state values */ +#define OPA_VNIC_STATE_DROP_ALL 0x1 +#define OPA_VNIC_STATE_FORWARDING 0x3 + +#define OPA_VESW_MAX_NUM_DEF_PORT 16 +#define OPA_VNIC_MAX_NUM_PCP 8 + +#define OPA_VNIC_EMA_DATA (OPA_MGMT_MAD_SIZE - IB_MGMT_VENDOR_HDR) + +/* Defines for vendor specific notice(trap) attributes */ +#define OPA_INTEL_EMA_NOTICE_TYPE_INFO 0x04 + +/* INTEL OUI */ +#define INTEL_OUI_1 0x00 +#define INTEL_OUI_2 0x06 +#define INTEL_OUI_3 0x6a + +/* Trap opcodes sent from VNIC */ +#define OPA_VESWPORT_TRAP_IFACE_UCAST_MAC_CHANGE 0x1 +#define OPA_VESWPORT_TRAP_IFACE_MCAST_MAC_CHANGE 0x2 +#define OPA_VESWPORT_TRAP_ETH_LINK_STATUS_CHANGE 0x3 + +#define OPA_VNIC_DLID_SD_IS_SRC_MAC(dlid_sd) (!!((dlid_sd) & 0x20)) +#define OPA_VNIC_DLID_SD_GET_DLID(dlid_sd) ((dlid_sd) >> 8) + +/* VNIC Ethernet link status */ +#define OPA_VNIC_ETH_LINK_UP 1 +#define OPA_VNIC_ETH_LINK_DOWN 2 + +/** + * struct opa_vesw_info - OPA vnic switch information + * @fabric_id: 10-bit fabric id + * @vesw_id: 12-bit virtual ethernet switch id + * @def_port_mask: bitmask of default ports + * @pkey: partition key + * @u_mcast_dlid: unknown multicast dlid + * @u_ucast_dlid: array of unknown unicast dlids + * @eth_mtu: MTUs for each vlan PCP + * @eth_mtu_non_vlan: MTU for non vlan packets + */ +struct opa_vesw_info { + __be16 fabric_id; + __be16 vesw_id; + + u8 rsvd0[6]; + __be16 def_port_mask; + + u8 rsvd1[2]; + __be16 pkey; + + u8 rsvd2[4]; + __be32 u_mcast_dlid; + __be32 u_ucast_dlid[OPA_VESW_MAX_NUM_DEF_PORT]; + + u8 rsvd3[44]; + __be16 eth_mtu[OPA_VNIC_MAX_NUM_PCP]; + __be16 eth_mtu_non_vlan; + u8 rsvd4[2]; +} __packed; + +/** + * struct opa_per_veswport_info - OPA vnic per port information + * @port_num: port number + * @eth_link_status: current ethernet link state + * @base_mac_addr: base mac address + * @config_state: configured port state + * @oper_state: operational port state + * @max_mac_tbl_ent: max number of mac table entries + * @max_smac_ent: max smac entries in mac table + * @mac_tbl_digest: mac table digest + * @encap_slid: base slid for the port + * @pcp_to_sc_uc: sc by pcp index for unicast ethernet packets + * @pcp_to_vl_uc: vl by pcp index for unicast ethernet packets + * @pcp_to_sc_mc: sc by pcp index for multicast ethernet packets + * @pcp_to_vl_mc: vl by pcp index for multicast ethernet packets + * @non_vlan_sc_uc: sc for non-vlan unicast ethernet packets + * @non_vlan_vl_uc: vl for non-vlan unicast ethernet packets + * @non_vlan_sc_mc: sc for non-vlan multicast ethernet packets + * @non_vlan_vl_mc: vl for non-vlan multicast ethernet packets + * @uc_macs_gen_count: generation count for unicast macs list + * @mc_macs_gen_count: generation count for multicast macs list + */ +struct opa_per_veswport_info { + __be32 port_num; + + u8 eth_link_status; + u8 rsvd0[3]; + + u8 base_mac_addr[ETH_ALEN]; + u8 config_state; + u8 oper_state; + + __be16 max_mac_tbl_ent; + __be16 max_smac_ent; + __be32 mac_tbl_digest; + u8 rsvd1[4]; + + __be32 encap_slid; + + u8 pcp_to_sc_uc[OPA_VNIC_MAX_NUM_PCP]; + u8 pcp_to_vl_uc[OPA_VNIC_MAX_NUM_PCP]; + u8 pcp_to_sc_mc[OPA_VNIC_MAX_NUM_PCP]; + u8 pcp_to_vl_mc[OPA_VNIC_MAX_NUM_PCP]; + + u8 non_vlan_sc_uc; + u8 non_vlan_vl_uc; + u8 non_vlan_sc_mc; + u8 non_vlan_vl_mc; + + u8 rsvd2[48]; + + __be16 uc_macs_gen_count; + __be16 mc_macs_gen_count; + + u8 rsvd3[8]; +} __packed; + +/** + * struct opa_veswport_info - OPA vnic port information + * @vesw: OPA vnic switch information + * @vport: OPA vnic per port information + * + * On host, each of the virtual ethernet ports belongs + * to a different virtual ethernet switches. + */ +struct opa_veswport_info { + struct opa_vesw_info vesw; + struct opa_per_veswport_info vport; +}; + +/** + * struct opa_veswport_mactable_entry - single entry in the forwarding table + * @mac_addr: MAC address + * @mac_addr_mask: MAC address bit mask + * @dlid_sd: Matching DLID and side data + * + * On the host each virtual ethernet port will have + * a forwarding table. These tables are used to + * map a MAC to a LID and other data. For more + * details see struct opa_veswport_mactable_entries. + * This is the structure of a single mactable entry + */ +struct opa_veswport_mactable_entry { + u8 mac_addr[ETH_ALEN]; + u8 mac_addr_mask[ETH_ALEN]; + __be32 dlid_sd; +} __packed; + +/** + * struct opa_veswport_mactable - Forwarding table array + * @offset: mac table starting offset + * @num_entries: Number of entries to get or set + * @mac_tbl_digest: mac table digest + * @tbl_entries[]: Array of table entries + * + * The EM sends down this structure in a MAD indicating + * the starting offset in the forwarding table that this + * entry is to be loaded into and the number of entries + * that that this MAD instance contains + * The mac_tbl_digest has been added to this MAD structure. It will be set by + * the EM and it will be used by the EM to check if there are any + * discrepancies with this value and the value + * maintained by the EM in the case of VNIC port being deleted or unloaded + * A new instantiation of a VNIC will always have a value of zero. + * This value is stored as part of the vnic adapter structure and will be + * accessed by the GET and SET routines for both the mactable entries and the + * veswport info. + */ +struct opa_veswport_mactable { + __be16 offset; + __be16 num_entries; + __be32 mac_tbl_digest; + struct opa_veswport_mactable_entry tbl_entries[0]; +} __packed; + +/** + * struct opa_veswport_summary_counters - summary counters + * @vp_instance: vport instance on the OPA port + * @vesw_id: virtual ethernet switch id + * @veswport_num: virtual ethernet switch port number + * @tx_errors: transmit errors + * @rx_errors: receive errors + * @tx_packets: transmit packets + * @rx_packets: receive packets + * @tx_bytes: transmit bytes + * @rx_bytes: receive bytes + * @tx_unicast: unicast packets transmitted + * @tx_mcastbcast: multicast/broadcast packets transmitted + * @tx_untagged: non-vlan packets transmitted + * @tx_vlan: vlan packets transmitted + * @tx_64_size: transmit packet length is 64 bytes + * @tx_65_127: transmit packet length is >=65 and < 127 bytes + * @tx_128_255: transmit packet length is >=128 and < 255 bytes + * @tx_256_511: transmit packet length is >=256 and < 511 bytes + * @tx_512_1023: transmit packet length is >=512 and < 1023 bytes + * @tx_1024_1518: transmit packet length is >=1024 and < 1518 bytes + * @tx_1519_max: transmit packet length >= 1519 bytes + * @rx_unicast: unicast packets received + * @rx_mcastbcast: multicast/broadcast packets received + * @rx_untagged: non-vlan packets received + * @rx_vlan: vlan packets received + * @rx_64_size: received packet length is 64 bytes + * @rx_65_127: received packet length is >=65 and < 127 bytes + * @rx_128_255: received packet length is >=128 and < 255 bytes + * @rx_256_511: received packet length is >=256 and < 511 bytes + * @rx_512_1023: received packet length is >=512 and < 1023 bytes + * @rx_1024_1518: received packet length is >=1024 and < 1518 bytes + * @rx_1519_max: received packet length >= 1519 bytes + * + * All the above are counters of corresponding conditions. + */ +struct opa_veswport_summary_counters { + __be16 vp_instance; + __be16 vesw_id; + __be32 veswport_num; + + __be64 tx_errors; + __be64 rx_errors; + __be64 tx_packets; + __be64 rx_packets; + __be64 tx_bytes; + __be64 rx_bytes; + + __be64 tx_unicast; + __be64 tx_mcastbcast; + + __be64 tx_untagged; + __be64 tx_vlan; + + __be64 tx_64_size; + __be64 tx_65_127; + __be64 tx_128_255; + __be64 tx_256_511; + __be64 tx_512_1023; + __be64 tx_1024_1518; + __be64 tx_1519_max; + + __be64 rx_unicast; + __be64 rx_mcastbcast; + + __be64 rx_untagged; + __be64 rx_vlan; + + __be64 rx_64_size; + __be64 rx_65_127; + __be64 rx_128_255; + __be64 rx_256_511; + __be64 rx_512_1023; + __be64 rx_1024_1518; + __be64 rx_1519_max; + + __be64 reserved[16]; +} __packed; + +/** + * struct opa_veswport_error_counters - error counters + * @vp_instance: vport instance on the OPA port + * @vesw_id: virtual ethernet switch id + * @veswport_num: virtual ethernet switch port number + * @tx_errors: transmit errors + * @rx_errors: receive errors + * @tx_smac_filt: smac filter errors + * @tx_dlid_zero: transmit packets with invalid dlid + * @tx_logic: other transmit errors + * @tx_drop_state: packet tansmission in non-forward port state + * @rx_bad_veswid: received packet with invalid vesw id + * @rx_runt: received ethernet packet with length < 64 bytes + * @rx_oversize: received ethernet packet with length > MTU size + * @rx_eth_down: received packets when interface is down + * @rx_drop_state: received packets in non-forwarding port state + * @rx_logic: other receive errors + * + * All the above are counters of corresponding erorr conditions. + */ +struct opa_veswport_error_counters { + __be16 vp_instance; + __be16 vesw_id; + __be32 veswport_num; + + __be64 tx_errors; + __be64 rx_errors; + + __be64 rsvd0; + __be64 tx_smac_filt; + __be64 rsvd1; + __be64 rsvd2; + __be64 rsvd3; + __be64 tx_dlid_zero; + __be64 rsvd4; + __be64 tx_logic; + __be64 rsvd5; + __be64 tx_drop_state; + + __be64 rx_bad_veswid; + __be64 rsvd6; + __be64 rx_runt; + __be64 rx_oversize; + __be64 rsvd7; + __be64 rx_eth_down; + __be64 rx_drop_state; + __be64 rx_logic; + __be64 rsvd8; + + __be64 rsvd9[16]; +} __packed; + +/** + * struct opa_veswport_trap - Trap message sent to EM by VNIC + * @fabric_id: 10 bit fabric id + * @veswid: 12 bit virtual ethernet switch id + * @veswportnum: logical port number on the Virtual switch + * @opaportnum: physical port num (redundant on host) + * @veswportindex: switch port index on opa port 0 based + * @opcode: operation + * @reserved: 32 bit for alignment + * + * The VNIC will send trap messages to the Ethernet manager to + * inform it about changes to the VNIC config, behaviour etc. + * This is the format of the trap payload. + */ +struct opa_veswport_trap { + __be16 fabric_id; + __be16 veswid; + __be32 veswportnum; + __be16 opaportnum; + u8 veswportindex; + u8 opcode; + __be32 reserved; +} __packed; + +/** + * struct opa_vnic_iface_macs_entry - single entry in the mac list + * @mac_addr: MAC address + */ +struct opa_vnic_iface_mac_entry { + u8 mac_addr[ETH_ALEN]; +}; + +/** + * struct opa_veswport_iface_macs - Msg to set globally administered MAC + * @start_idx: position of first entry (0 based) + * @num_macs_in_msg: number of MACs in this message + * @tot_macs_in_lst: The total number of MACs the agent has + * @gen_count: gen_count to indicate change + * @entry: The mac list entry + * + * Same attribute IDS and attribute modifiers as in locally administered + * addresses used to set globally administered addresses + */ +struct opa_veswport_iface_macs { + __be16 start_idx; + __be16 num_macs_in_msg; + __be16 tot_macs_in_lst; + __be16 gen_count; + struct opa_vnic_iface_mac_entry entry[0]; +} __packed; + +/** + * struct opa_vnic_vema_mad - Generic VEMA MAD + * @mad_hdr: Generic MAD header + * @rmpp_hdr: RMPP header for vendor specific MADs + * @oui: Unique org identifier + * @data: MAD data + */ +struct opa_vnic_vema_mad { + struct ib_mad_hdr mad_hdr; + struct ib_rmpp_hdr rmpp_hdr; + u8 reserved; + u8 oui[3]; + u8 data[OPA_VNIC_EMA_DATA]; +}; + +/** + * struct opa_vnic_notice_attr - Generic Notice MAD + * @gen_type: Generic/Specific bit and type of notice + * @oui_1: Vendor ID byte 1 + * @oui_2: Vendor ID byte 2 + * @oui_3: Vendor ID byte 3 + * @trap_num: Trap number + * @toggle_count: Notice toggle bit and count value + * @issuer_lid: Trap issuer's lid + * @issuer_gid: Issuer GID (only if Report method) + * @raw_data: Trap message body + */ +struct opa_vnic_notice_attr { + u8 gen_type; + u8 oui_1; + u8 oui_2; + u8 oui_3; + __be16 trap_num; + __be16 toggle_count; + __be32 issuer_lid; + __be32 reserved; + u8 issuer_gid[16]; + u8 raw_data[64]; +} __packed; + +/** + * struct opa_vnic_vema_mad_trap - Generic VEMA MAD Trap + * @mad_hdr: Generic MAD header + * @rmpp_hdr: RMPP header for vendor specific MADs + * @oui: Unique org identifier + * @notice: Notice structure + */ +struct opa_vnic_vema_mad_trap { + struct ib_mad_hdr mad_hdr; + struct ib_rmpp_hdr rmpp_hdr; + u8 reserved; + u8 oui[3]; + struct opa_vnic_notice_attr notice; +}; + +#endif /* _OPA_VNIC_ENCAP_H */ diff --git a/drivers/infiniband/ulp/opa_vnic/opa_vnic_ethtool.c b/drivers/infiniband/ulp/opa_vnic/opa_vnic_ethtool.c new file mode 100644 index 000000000000..d66540e24885 --- /dev/null +++ b/drivers/infiniband/ulp/opa_vnic/opa_vnic_ethtool.c @@ -0,0 +1,187 @@ +/* + * Copyright(c) 2017 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ + +/* + * This file contains OPA VNIC ethtool functions + */ + +#include <linux/ethtool.h> + +#include "opa_vnic_internal.h" + +enum {NETDEV_STATS, VNIC_STATS}; + +struct vnic_stats { + char stat_string[ETH_GSTRING_LEN]; + struct { + int sizeof_stat; + int stat_offset; + }; +}; + +#define VNIC_STAT(m) { FIELD_SIZEOF(struct opa_vnic_stats, m), \ + offsetof(struct opa_vnic_stats, m) } + +static struct vnic_stats vnic_gstrings_stats[] = { + /* NETDEV stats */ + {"rx_packets", VNIC_STAT(netstats.rx_packets)}, + {"tx_packets", VNIC_STAT(netstats.tx_packets)}, + {"rx_bytes", VNIC_STAT(netstats.rx_bytes)}, + {"tx_bytes", VNIC_STAT(netstats.tx_bytes)}, + {"rx_errors", VNIC_STAT(netstats.rx_errors)}, + {"tx_errors", VNIC_STAT(netstats.tx_errors)}, + {"rx_dropped", VNIC_STAT(netstats.rx_dropped)}, + {"tx_dropped", VNIC_STAT(netstats.tx_dropped)}, + + /* SUMMARY counters */ + {"tx_unicast", VNIC_STAT(tx_grp.unicast)}, + {"tx_mcastbcast", VNIC_STAT(tx_grp.mcastbcast)}, + {"tx_untagged", VNIC_STAT(tx_grp.untagged)}, + {"tx_vlan", VNIC_STAT(tx_grp.vlan)}, + + {"tx_64_size", VNIC_STAT(tx_grp.s_64)}, + {"tx_65_127", VNIC_STAT(tx_grp.s_65_127)}, + {"tx_128_255", VNIC_STAT(tx_grp.s_128_255)}, + {"tx_256_511", VNIC_STAT(tx_grp.s_256_511)}, + {"tx_512_1023", VNIC_STAT(tx_grp.s_512_1023)}, + {"tx_1024_1518", VNIC_STAT(tx_grp.s_1024_1518)}, + {"tx_1519_max", VNIC_STAT(tx_grp.s_1519_max)}, + + {"rx_unicast", VNIC_STAT(rx_grp.unicast)}, + {"rx_mcastbcast", VNIC_STAT(rx_grp.mcastbcast)}, + {"rx_untagged", VNIC_STAT(rx_grp.untagged)}, + {"rx_vlan", VNIC_STAT(rx_grp.vlan)}, + + {"rx_64_size", VNIC_STAT(rx_grp.s_64)}, + {"rx_65_127", VNIC_STAT(rx_grp.s_65_127)}, + {"rx_128_255", VNIC_STAT(rx_grp.s_128_255)}, + {"rx_256_511", VNIC_STAT(rx_grp.s_256_511)}, + {"rx_512_1023", VNIC_STAT(rx_grp.s_512_1023)}, + {"rx_1024_1518", VNIC_STAT(rx_grp.s_1024_1518)}, + {"rx_1519_max", VNIC_STAT(rx_grp.s_1519_max)}, + + /* ERROR counters */ + {"rx_fifo_errors", VNIC_STAT(netstats.rx_fifo_errors)}, + {"rx_length_errors", VNIC_STAT(netstats.rx_length_errors)}, + + {"tx_fifo_errors", VNIC_STAT(netstats.tx_fifo_errors)}, + {"tx_carrier_errors", VNIC_STAT(netstats.tx_carrier_errors)}, + + {"tx_dlid_zero", VNIC_STAT(tx_dlid_zero)}, + {"tx_drop_state", VNIC_STAT(tx_drop_state)}, + {"rx_drop_state", VNIC_STAT(rx_drop_state)}, + {"rx_oversize", VNIC_STAT(rx_oversize)}, + {"rx_runt", VNIC_STAT(rx_runt)}, +}; + +#define VNIC_STATS_LEN ARRAY_SIZE(vnic_gstrings_stats) + +/* vnic_get_drvinfo - get driver info */ +static void vnic_get_drvinfo(struct net_device *netdev, + struct ethtool_drvinfo *drvinfo) +{ + strlcpy(drvinfo->driver, opa_vnic_driver_name, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, opa_vnic_driver_version, + sizeof(drvinfo->version)); + strlcpy(drvinfo->bus_info, dev_name(netdev->dev.parent), + sizeof(drvinfo->bus_info)); +} + +/* vnic_get_sset_count - get string set count */ +static int vnic_get_sset_count(struct net_device *netdev, int sset) +{ + return (sset == ETH_SS_STATS) ? VNIC_STATS_LEN : -EOPNOTSUPP; +} + +/* vnic_get_ethtool_stats - get statistics */ +static void vnic_get_ethtool_stats(struct net_device *netdev, + struct ethtool_stats *stats, u64 *data) +{ + struct opa_vnic_adapter *adapter = opa_vnic_priv(netdev); + struct opa_vnic_stats vstats; + int i; + + memset(&vstats, 0, sizeof(vstats)); + mutex_lock(&adapter->stats_lock); + adapter->rn_ops->ndo_get_stats64(netdev, &vstats.netstats); + 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 */ +static void vnic_get_strings(struct net_device *netdev, u32 stringset, u8 *data) +{ + int i; + + if (stringset != ETH_SS_STATS) + return; + + for (i = 0; i < VNIC_STATS_LEN; i++) + memcpy(data + i * ETH_GSTRING_LEN, + vnic_gstrings_stats[i].stat_string, + ETH_GSTRING_LEN); +} + +/* ethtool ops */ +static const struct ethtool_ops opa_vnic_ethtool_ops = { + .get_drvinfo = vnic_get_drvinfo, + .get_link = ethtool_op_get_link, + .get_strings = vnic_get_strings, + .get_sset_count = vnic_get_sset_count, + .get_ethtool_stats = vnic_get_ethtool_stats, +}; + +/* opa_vnic_set_ethtool_ops - set ethtool ops */ +void opa_vnic_set_ethtool_ops(struct net_device *netdev) +{ + netdev->ethtool_ops = &opa_vnic_ethtool_ops; +} diff --git a/drivers/infiniband/ulp/opa_vnic/opa_vnic_internal.h b/drivers/infiniband/ulp/opa_vnic/opa_vnic_internal.h new file mode 100644 index 000000000000..6bba886bec1f --- /dev/null +++ b/drivers/infiniband/ulp/opa_vnic/opa_vnic_internal.h @@ -0,0 +1,329 @@ +#ifndef _OPA_VNIC_INTERNAL_H +#define _OPA_VNIC_INTERNAL_H +/* + * Copyright(c) 2017 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ + +/* + * This file contains OPA VNIC driver internal declarations + */ + +#include <linux/bitops.h> +#include <linux/etherdevice.h> +#include <linux/hashtable.h> +#include <linux/sizes.h> +#include <rdma/opa_vnic.h> + +#include "opa_vnic_encap.h" + +#define OPA_VNIC_VLAN_PCP(vlan_tci) \ + (((vlan_tci) & VLAN_PRIO_MASK) >> VLAN_PRIO_SHIFT) + +/* Flow to default port redirection table size */ +#define OPA_VNIC_FLOW_TBL_SIZE 32 + +/* Invalid port number */ +#define OPA_VNIC_INVALID_PORT 0xff + +struct opa_vnic_adapter; + +/** + * struct __opa_vesw_info - OPA vnic virtual switch info + * + * Same as opa_vesw_info without bitwise attribute. + */ +struct __opa_vesw_info { + u16 fabric_id; + u16 vesw_id; + + u8 rsvd0[6]; + u16 def_port_mask; + + u8 rsvd1[2]; + u16 pkey; + + u8 rsvd2[4]; + u32 u_mcast_dlid; + u32 u_ucast_dlid[OPA_VESW_MAX_NUM_DEF_PORT]; + + u8 rsvd3[44]; + u16 eth_mtu[OPA_VNIC_MAX_NUM_PCP]; + u16 eth_mtu_non_vlan; + u8 rsvd4[2]; +} __packed; + +/** + * struct __opa_per_veswport_info - OPA vnic per port info + * + * Same as opa_per_veswport_info without bitwise attribute. + */ +struct __opa_per_veswport_info { + u32 port_num; + + u8 eth_link_status; + u8 rsvd0[3]; + + u8 base_mac_addr[ETH_ALEN]; + u8 config_state; + u8 oper_state; + + u16 max_mac_tbl_ent; + u16 max_smac_ent; + u32 mac_tbl_digest; + u8 rsvd1[4]; + + u32 encap_slid; + + u8 pcp_to_sc_uc[OPA_VNIC_MAX_NUM_PCP]; + u8 pcp_to_vl_uc[OPA_VNIC_MAX_NUM_PCP]; + u8 pcp_to_sc_mc[OPA_VNIC_MAX_NUM_PCP]; + u8 pcp_to_vl_mc[OPA_VNIC_MAX_NUM_PCP]; + + u8 non_vlan_sc_uc; + u8 non_vlan_vl_uc; + u8 non_vlan_sc_mc; + u8 non_vlan_vl_mc; + + u8 rsvd2[48]; + + u16 uc_macs_gen_count; + u16 mc_macs_gen_count; + + u8 rsvd3[8]; +} __packed; + +/** + * struct __opa_veswport_info - OPA vnic port info + * + * Same as opa_veswport_info without bitwise attribute. + */ +struct __opa_veswport_info { + struct __opa_vesw_info vesw; + struct __opa_per_veswport_info vport; +}; + +/** + * struct __opa_veswport_trap - OPA vnic trap info + * + * Same as opa_veswport_trap without bitwise attribute. + */ +struct __opa_veswport_trap { + u16 fabric_id; + u16 veswid; + u32 veswportnum; + u16 opaportnum; + u8 veswportindex; + u8 opcode; + u32 reserved; +} __packed; + +/** + * struct opa_vnic_ctrl_port - OPA virtual NIC control port + * @ibdev: pointer to ib device + * @ops: opa vnic control operations + * @num_ports: number of opa ports + */ +struct opa_vnic_ctrl_port { + struct ib_device *ibdev; + struct opa_vnic_ctrl_ops *ops; + u8 num_ports; +}; + +/** + * struct opa_vnic_adapter - OPA VNIC netdev private data structure + * @netdev: pointer to associated netdev + * @ibdev: ib device + * @cport: pointer to opa vnic control port + * @rn_ops: rdma netdev's net_device_ops + * @port_num: OPA port number + * @vport_num: vesw port number + * @lock: adapter lock + * @info: virtual ethernet switch port information + * @vema_mac_addr: mac address configured by vema + * @umac_hash: unicast maclist hash + * @mmac_hash: multicast maclist hash + * @mactbl: hash table of MAC entries + * @mactbl_lock: mac table lock + * @stats_lock: statistics lock + * @flow_tbl: flow to default port redirection table + * @trap_timeout: trap timeout + * @trap_count: no. of traps allowed within timeout period + */ +struct opa_vnic_adapter { + struct net_device *netdev; + struct ib_device *ibdev; + struct opa_vnic_ctrl_port *cport; + const struct net_device_ops *rn_ops; + + u8 port_num; + u8 vport_num; + + /* Lock used around concurrent updates to netdev */ + struct mutex lock; + + struct __opa_veswport_info info; + u8 vema_mac_addr[ETH_ALEN]; + u32 umac_hash; + u32 mmac_hash; + struct hlist_head __rcu *mactbl; + + /* Lock used to protect updates to mac table */ + struct mutex mactbl_lock; + + /* Lock used to protect access to vnic counters */ + struct mutex stats_lock; + + u8 flow_tbl[OPA_VNIC_FLOW_TBL_SIZE]; + + unsigned long trap_timeout; + u8 trap_count; +}; + +/* Same as opa_veswport_mactable_entry, but without bitwise attribute */ +struct __opa_vnic_mactable_entry { + u8 mac_addr[ETH_ALEN]; + u8 mac_addr_mask[ETH_ALEN]; + u32 dlid_sd; +} __packed; + +/** + * struct opa_vnic_mac_tbl_node - OPA VNIC mac table node + * @hlist: hash list handle + * @index: index of entry in the mac table + * @entry: entry in the table + */ +struct opa_vnic_mac_tbl_node { + struct hlist_node hlist; + u16 index; + struct __opa_vnic_mactable_entry entry; +}; + +#define v_dbg(format, arg...) \ + netdev_dbg(adapter->netdev, format, ## arg) +#define v_err(format, arg...) \ + netdev_err(adapter->netdev, format, ## arg) +#define v_info(format, arg...) \ + netdev_info(adapter->netdev, format, ## arg) +#define v_warn(format, arg...) \ + netdev_warn(adapter->netdev, format, ## arg) + +#define c_err(format, arg...) \ + dev_err(&cport->ibdev->dev, format, ## arg) +#define c_info(format, arg...) \ + dev_info(&cport->ibdev->dev, format, ## arg) +#define c_dbg(format, arg...) \ + dev_dbg(&cport->ibdev->dev, format, ## arg) + +/* The maximum allowed entries in the mac table */ +#define OPA_VNIC_MAC_TBL_MAX_ENTRIES 2048 +/* Limit of smac entries in mac table */ +#define OPA_VNIC_MAX_SMAC_LIMIT 256 + +/* The last octet of the MAC address is used as the key to the hash table */ +#define OPA_VNIC_MAC_HASH_IDX 5 + +/* The VNIC MAC hash table is of size 2^8 */ +#define OPA_VNIC_MAC_TBL_HASH_BITS 8 +#define OPA_VNIC_MAC_TBL_SIZE BIT(OPA_VNIC_MAC_TBL_HASH_BITS) + +/* VNIC HASH MACROS */ +#define vnic_hash_init(hashtable) __hash_init(hashtable, OPA_VNIC_MAC_TBL_SIZE) + +#define vnic_hash_add(hashtable, node, key) \ + hlist_add_head(node, \ + &hashtable[hash_min(key, ilog2(OPA_VNIC_MAC_TBL_SIZE))]) + +#define vnic_hash_for_each_safe(name, bkt, tmp, obj, member) \ + for ((bkt) = 0, obj = NULL; \ + !obj && (bkt) < OPA_VNIC_MAC_TBL_SIZE; (bkt)++) \ + hlist_for_each_entry_safe(obj, tmp, &name[bkt], member) + +#define vnic_hash_for_each_possible(name, obj, member, key) \ + hlist_for_each_entry(obj, \ + &name[hash_min(key, ilog2(OPA_VNIC_MAC_TBL_SIZE))], member) + +#define vnic_hash_for_each(name, bkt, obj, member) \ + for ((bkt) = 0, obj = NULL; \ + !obj && (bkt) < OPA_VNIC_MAC_TBL_SIZE; (bkt)++) \ + hlist_for_each_entry(obj, &name[bkt], member) + +extern char opa_vnic_driver_name[]; +extern const char opa_vnic_driver_version[]; + +struct opa_vnic_adapter *opa_vnic_add_netdev(struct ib_device *ibdev, + u8 port_num, u8 vport_num); +void opa_vnic_rem_netdev(struct opa_vnic_adapter *adapter); +void opa_vnic_encap_skb(struct opa_vnic_adapter *adapter, struct sk_buff *skb); +u8 opa_vnic_get_vl(struct opa_vnic_adapter *adapter, struct sk_buff *skb); +u8 opa_vnic_calc_entropy(struct opa_vnic_adapter *adapter, struct sk_buff *skb); +void opa_vnic_process_vema_config(struct opa_vnic_adapter *adapter); +void opa_vnic_release_mac_tbl(struct opa_vnic_adapter *adapter); +void opa_vnic_query_mac_tbl(struct opa_vnic_adapter *adapter, + struct opa_veswport_mactable *tbl); +int opa_vnic_update_mac_tbl(struct opa_vnic_adapter *adapter, + struct opa_veswport_mactable *tbl); +void opa_vnic_query_ucast_macs(struct opa_vnic_adapter *adapter, + struct opa_veswport_iface_macs *macs); +void opa_vnic_query_mcast_macs(struct opa_vnic_adapter *adapter, + struct opa_veswport_iface_macs *macs); +void opa_vnic_get_summary_counters(struct opa_vnic_adapter *adapter, + struct opa_veswport_summary_counters *cntrs); +void opa_vnic_get_error_counters(struct opa_vnic_adapter *adapter, + struct opa_veswport_error_counters *cntrs); +void opa_vnic_get_vesw_info(struct opa_vnic_adapter *adapter, + struct opa_vesw_info *info); +void opa_vnic_set_vesw_info(struct opa_vnic_adapter *adapter, + struct opa_vesw_info *info); +void opa_vnic_get_per_veswport_info(struct opa_vnic_adapter *adapter, + struct opa_per_veswport_info *info); +void opa_vnic_set_per_veswport_info(struct opa_vnic_adapter *adapter, + struct opa_per_veswport_info *info); +void opa_vnic_vema_report_event(struct opa_vnic_adapter *adapter, u8 event); +void opa_vnic_set_ethtool_ops(struct net_device *netdev); +void opa_vnic_vema_send_trap(struct opa_vnic_adapter *adapter, + struct __opa_veswport_trap *data, u32 lid); + +#endif /* _OPA_VNIC_INTERNAL_H */ diff --git a/drivers/infiniband/ulp/opa_vnic/opa_vnic_netdev.c b/drivers/infiniband/ulp/opa_vnic/opa_vnic_netdev.c new file mode 100644 index 000000000000..905f39dda5aa --- /dev/null +++ b/drivers/infiniband/ulp/opa_vnic/opa_vnic_netdev.c @@ -0,0 +1,389 @@ +/* + * Copyright(c) 2017 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ + +/* + * This file contains OPA Virtual Network Interface Controller (VNIC) driver + * netdev functionality. + */ + +#include <linux/module.h> +#include <linux/if_vlan.h> +#include <linux/crc32.h> + +#include "opa_vnic_internal.h" + +#define OPA_TX_TIMEOUT_MS 1000 + +#define OPA_VNIC_SKB_HEADROOM \ + ALIGN((OPA_VNIC_HDR_LEN + OPA_VNIC_SKB_MDATA_LEN), 8) + +/* This function is overloaded for opa_vnic specific implementation */ +static void opa_vnic_get_stats64(struct net_device *netdev, + struct rtnl_link_stats64 *stats) +{ + struct opa_vnic_adapter *adapter = opa_vnic_priv(netdev); + struct opa_vnic_stats vstats; + + memset(&vstats, 0, sizeof(vstats)); + mutex_lock(&adapter->stats_lock); + adapter->rn_ops->ndo_get_stats64(netdev, &vstats.netstats); + mutex_unlock(&adapter->stats_lock); + memcpy(stats, &vstats.netstats, sizeof(*stats)); +} + +/* opa_netdev_start_xmit - transmit function */ +static netdev_tx_t opa_netdev_start_xmit(struct sk_buff *skb, + struct net_device *netdev) +{ + struct opa_vnic_adapter *adapter = opa_vnic_priv(netdev); + + v_dbg("xmit: queue %d skb len %d\n", skb->queue_mapping, skb->len); + /* pad to ensure mininum ethernet packet length */ + if (unlikely(skb->len < ETH_ZLEN)) { + if (skb_padto(skb, ETH_ZLEN)) + return NETDEV_TX_OK; + + skb_put(skb, ETH_ZLEN - skb->len); + } + + opa_vnic_encap_skb(adapter, skb); + return adapter->rn_ops->ndo_start_xmit(skb, netdev); +} + +static u16 opa_vnic_select_queue(struct net_device *netdev, struct sk_buff *skb, + void *accel_priv, + select_queue_fallback_t fallback) +{ + struct opa_vnic_adapter *adapter = opa_vnic_priv(netdev); + struct opa_vnic_skb_mdata *mdata; + int rc; + + /* pass entropy and vl as metadata in skb */ + mdata = (struct opa_vnic_skb_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, + accel_priv, fallback); + skb_pull(skb, sizeof(*mdata)); + return rc; +} + +/* opa_vnic_process_vema_config - process vema configuration updates */ +void opa_vnic_process_vema_config(struct opa_vnic_adapter *adapter) +{ + struct __opa_veswport_info *info = &adapter->info; + struct rdma_netdev *rn = netdev_priv(adapter->netdev); + u8 port_num[OPA_VESW_MAX_NUM_DEF_PORT] = { 0 }; + struct net_device *netdev = adapter->netdev; + u8 i, port_count = 0; + u16 port_mask; + + /* If the base_mac_addr is changed, update the interface mac address */ + if (memcmp(info->vport.base_mac_addr, adapter->vema_mac_addr, + ARRAY_SIZE(info->vport.base_mac_addr))) { + struct sockaddr saddr; + + memcpy(saddr.sa_data, info->vport.base_mac_addr, + ARRAY_SIZE(info->vport.base_mac_addr)); + mutex_lock(&adapter->lock); + eth_mac_addr(netdev, &saddr); + memcpy(adapter->vema_mac_addr, + info->vport.base_mac_addr, ETH_ALEN); + mutex_unlock(&adapter->lock); + } + + rn->set_id(netdev, info->vesw.vesw_id); + + /* Handle MTU limit change */ + rtnl_lock(); + netdev->max_mtu = max_t(unsigned int, info->vesw.eth_mtu_non_vlan, + netdev->min_mtu); + if (netdev->mtu > netdev->max_mtu) + dev_set_mtu(netdev, netdev->max_mtu); + rtnl_unlock(); + + /* Update flow to default port redirection table */ + port_mask = info->vesw.def_port_mask; + for (i = 0; i < OPA_VESW_MAX_NUM_DEF_PORT; i++) { + if (port_mask & 1) + port_num[port_count++] = i; + port_mask >>= 1; + } + + /* + * Build the flow table. Flow table is required when destination LID + * is not available. Up to OPA_VNIC_FLOW_TBL_SIZE flows supported. + * Each flow need a default port number to get its dlid from the + * u_ucast_dlid array. + */ + for (i = 0; i < OPA_VNIC_FLOW_TBL_SIZE; i++) + adapter->flow_tbl[i] = port_count ? port_num[i % port_count] : + OPA_VNIC_INVALID_PORT; + + /* Operational state can only be DROP_ALL or FORWARDING */ + if (info->vport.config_state == OPA_VNIC_STATE_FORWARDING) { + info->vport.oper_state = OPA_VNIC_STATE_FORWARDING; + netif_dormant_off(netdev); + } else { + info->vport.oper_state = OPA_VNIC_STATE_DROP_ALL; + netif_dormant_on(netdev); + } +} + +/* + * Set the power on default values in adapter's vema interface structure. + */ +static inline void opa_vnic_set_pod_values(struct opa_vnic_adapter *adapter) +{ + adapter->info.vport.max_mac_tbl_ent = OPA_VNIC_MAC_TBL_MAX_ENTRIES; + adapter->info.vport.max_smac_ent = OPA_VNIC_MAX_SMAC_LIMIT; + adapter->info.vport.config_state = OPA_VNIC_STATE_DROP_ALL; + adapter->info.vport.eth_link_status = OPA_VNIC_ETH_LINK_DOWN; +} + +/* opa_vnic_set_mac_addr - change mac address */ +static int opa_vnic_set_mac_addr(struct net_device *netdev, void *addr) +{ + struct opa_vnic_adapter *adapter = opa_vnic_priv(netdev); + struct sockaddr *sa = addr; + int rc; + + if (!memcmp(netdev->dev_addr, sa->sa_data, ETH_ALEN)) + return 0; + + mutex_lock(&adapter->lock); + rc = eth_mac_addr(netdev, addr); + mutex_unlock(&adapter->lock); + if (rc) + return rc; + + adapter->info.vport.uc_macs_gen_count++; + opa_vnic_vema_report_event(adapter, + OPA_VESWPORT_TRAP_IFACE_UCAST_MAC_CHANGE); + return 0; +} + +/* + * opa_vnic_mac_send_event - post event on possible mac list exchange + * Send trap when digest from uc/mc mac list differs from previous run. + * Digest is evaluated similar to how cksum does. + */ +static void opa_vnic_mac_send_event(struct net_device *netdev, u8 event) +{ + struct opa_vnic_adapter *adapter = opa_vnic_priv(netdev); + struct netdev_hw_addr *ha; + struct netdev_hw_addr_list *hw_list; + u32 *ref_crc; + u32 l, crc = 0; + + switch (event) { + case OPA_VESWPORT_TRAP_IFACE_UCAST_MAC_CHANGE: + hw_list = &netdev->uc; + adapter->info.vport.uc_macs_gen_count++; + ref_crc = &adapter->umac_hash; + break; + case OPA_VESWPORT_TRAP_IFACE_MCAST_MAC_CHANGE: + hw_list = &netdev->mc; + adapter->info.vport.mc_macs_gen_count++; + ref_crc = &adapter->mmac_hash; + break; + default: + return; + } + netdev_hw_addr_list_for_each(ha, hw_list) { + crc = crc32_le(crc, ha->addr, ETH_ALEN); + } + l = netdev_hw_addr_list_count(hw_list) * ETH_ALEN; + crc = ~crc32_le(crc, (void *)&l, sizeof(l)); + + if (crc != *ref_crc) { + *ref_crc = crc; + opa_vnic_vema_report_event(adapter, event); + } +} + +/* opa_vnic_set_rx_mode - handle uc/mc mac list change */ +static void opa_vnic_set_rx_mode(struct net_device *netdev) +{ + opa_vnic_mac_send_event(netdev, + OPA_VESWPORT_TRAP_IFACE_UCAST_MAC_CHANGE); + + opa_vnic_mac_send_event(netdev, + OPA_VESWPORT_TRAP_IFACE_MCAST_MAC_CHANGE); +} + +/* opa_netdev_open - activate network interface */ +static int opa_netdev_open(struct net_device *netdev) +{ + struct opa_vnic_adapter *adapter = opa_vnic_priv(netdev); + int rc; + + rc = adapter->rn_ops->ndo_open(adapter->netdev); + if (rc) { + v_dbg("open failed %d\n", rc); + return rc; + } + + /* Update eth link status and send trap */ + adapter->info.vport.eth_link_status = OPA_VNIC_ETH_LINK_UP; + opa_vnic_vema_report_event(adapter, + OPA_VESWPORT_TRAP_ETH_LINK_STATUS_CHANGE); + return 0; +} + +/* opa_netdev_close - disable network interface */ +static int opa_netdev_close(struct net_device *netdev) +{ + struct opa_vnic_adapter *adapter = opa_vnic_priv(netdev); + int rc; + + rc = adapter->rn_ops->ndo_stop(adapter->netdev); + if (rc) { + v_dbg("close failed %d\n", rc); + return rc; + } + + /* Update eth link status and send trap */ + adapter->info.vport.eth_link_status = OPA_VNIC_ETH_LINK_DOWN; + opa_vnic_vema_report_event(adapter, + OPA_VESWPORT_TRAP_ETH_LINK_STATUS_CHANGE); + return 0; +} + +/* netdev ops */ +static const struct net_device_ops opa_netdev_ops = { + .ndo_open = opa_netdev_open, + .ndo_stop = opa_netdev_close, + .ndo_start_xmit = opa_netdev_start_xmit, + .ndo_get_stats64 = opa_vnic_get_stats64, + .ndo_set_rx_mode = opa_vnic_set_rx_mode, + .ndo_select_queue = opa_vnic_select_queue, + .ndo_set_mac_address = opa_vnic_set_mac_addr, +}; + +/* opa_vnic_add_netdev - create vnic netdev interface */ +struct opa_vnic_adapter *opa_vnic_add_netdev(struct ib_device *ibdev, + u8 port_num, u8 vport_num) +{ + struct opa_vnic_adapter *adapter; + struct net_device *netdev; + struct rdma_netdev *rn; + int rc; + + netdev = ibdev->alloc_rdma_netdev(ibdev, port_num, + RDMA_NETDEV_OPA_VNIC, + "veth%d", NET_NAME_UNKNOWN, + ether_setup); + if (!netdev) + return ERR_PTR(-ENOMEM); + else if (IS_ERR(netdev)) + return ERR_CAST(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; + adapter->netdev = netdev; + adapter->ibdev = ibdev; + adapter->port_num = port_num; + adapter->vport_num = vport_num; + adapter->rn_ops = netdev->netdev_ops; + + netdev->netdev_ops = &opa_netdev_ops; + netdev->priv_flags |= IFF_LIVE_ADDR_CHANGE; + netdev->hard_header_len += OPA_VNIC_SKB_HEADROOM; + mutex_init(&adapter->lock); + mutex_init(&adapter->mactbl_lock); + mutex_init(&adapter->stats_lock); + + SET_NETDEV_DEV(netdev, ibdev->dev.parent); + + opa_vnic_set_ethtool_ops(netdev); + + opa_vnic_set_pod_values(adapter); + + rc = register_netdev(netdev); + if (rc) + goto netdev_err; + + netif_carrier_off(netdev); + netif_dormant_on(netdev); + v_info("initialized\n"); + + return adapter; +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); + + return ERR_PTR(rc); +} + +/* opa_vnic_rem_netdev - remove vnic netdev interface */ +void opa_vnic_rem_netdev(struct opa_vnic_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + struct ib_device *ibdev = adapter->ibdev; + + 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); +} diff --git a/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c b/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c new file mode 100644 index 000000000000..cee49aa6676c --- /dev/null +++ b/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c @@ -0,0 +1,1053 @@ +/* + * Copyright(c) 2017 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ + +/* + * This file contains OPA Virtual Network Interface Controller (VNIC) + * Ethernet Management Agent (EMA) driver + */ + +#include <linux/module.h> +#include <rdma/ib_addr.h> +#include <rdma/ib_smi.h> + +#include "opa_vnic_internal.h" + +#define DRV_VERSION "1.0" +char opa_vnic_driver_name[] = "opa_vnic"; +const char opa_vnic_driver_version[] = DRV_VERSION; + +/* + * The trap service level is kept in bits 3 to 7 in the trap_sl_rsvd + * field in the class port info MAD. + */ +#define GET_TRAP_SL_FROM_CLASS_PORT_INFO(x) (((x) >> 3) & 0x1f) + +/* Cap trap bursts to a reasonable limit good for normal cases */ +#define OPA_VNIC_TRAP_BURST_LIMIT 4 + +/* + * VNIC trap limit timeout. + * Inverse of cap2_mask response time out (1.0737 secs) = 0.9 + * secs approx IB spec 13.4.6.2.1 PortInfoSubnetTimeout and + * 13.4.9 Traps. + */ +#define OPA_VNIC_TRAP_TIMEOUT ((4096 * (1UL << 18)) / 1000) + +#define OPA_VNIC_UNSUP_ATTR \ + cpu_to_be16(IB_MGMT_MAD_STATUS_UNSUPPORTED_METHOD_ATTRIB) + +#define OPA_VNIC_INVAL_ATTR \ + cpu_to_be16(IB_MGMT_MAD_STATUS_INVALID_ATTRIB_VALUE) + +#define OPA_VNIC_CLASS_CAP_TRAP 0x1 + +/* Maximum number of VNIC ports supported */ +#define OPA_VNIC_MAX_NUM_VPORT 255 + +/** + * struct opa_vnic_vema_port -- VNIC VEMA port details + * @cport: pointer to port + * @mad_agent: pointer to mad agent for port + * @class_port_info: Class port info information. + * @tid: Transaction id + * @port_num: OPA port number + * @vport_idr: vnic ports idr + * @event_handler: ib event handler + * @lock: adapter interface lock + */ +struct opa_vnic_vema_port { + struct opa_vnic_ctrl_port *cport; + struct ib_mad_agent *mad_agent; + struct opa_class_port_info class_port_info; + u64 tid; + u8 port_num; + struct idr vport_idr; + struct ib_event_handler event_handler; + + /* Lock to query/update network adapter */ + struct mutex lock; +}; + +static void opa_vnic_vema_add_one(struct ib_device *device); +static void opa_vnic_vema_rem_one(struct ib_device *device, + void *client_data); + +static struct ib_client opa_vnic_client = { + .name = opa_vnic_driver_name, + .add = opa_vnic_vema_add_one, + .remove = opa_vnic_vema_rem_one, +}; + +/** + * vema_get_vport_num -- Get the vnic from the mad + * @recvd_mad: Received mad + * + * Return: returns value of the vnic port number + */ +static inline u8 vema_get_vport_num(struct opa_vnic_vema_mad *recvd_mad) +{ + return be32_to_cpu(recvd_mad->mad_hdr.attr_mod) & 0xff; +} + +/** + * vema_get_vport_adapter -- Get vnic port adapter from recvd mad + * @recvd_mad: received mad + * @port: ptr to port struct on which MAD was recvd + * + * Return: vnic adapter + */ +static inline struct opa_vnic_adapter * +vema_get_vport_adapter(struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_port *port) +{ + u8 vport_num = vema_get_vport_num(recvd_mad); + + return idr_find(&port->vport_idr, vport_num); +} + +/** + * vema_mac_tbl_req_ok -- Check if mac request has correct values + * @mac_tbl: mac table + * + * This function checks for the validity of the offset and number of + * entries required. + * + * Return: true if offset and num_entries are valid + */ +static inline bool vema_mac_tbl_req_ok(struct opa_veswport_mactable *mac_tbl) +{ + u16 offset, num_entries; + u16 req_entries = ((OPA_VNIC_EMA_DATA - sizeof(*mac_tbl)) / + sizeof(mac_tbl->tbl_entries[0])); + + offset = be16_to_cpu(mac_tbl->offset); + num_entries = be16_to_cpu(mac_tbl->num_entries); + + return ((num_entries <= req_entries) && + (offset + num_entries <= OPA_VNIC_MAC_TBL_MAX_ENTRIES)); +} + +/* + * Return the power on default values in the port info structure + * in big endian format as required by MAD. + */ +static inline void vema_get_pod_values(struct opa_veswport_info *port_info) +{ + memset(port_info, 0, sizeof(*port_info)); + port_info->vport.max_mac_tbl_ent = + cpu_to_be16(OPA_VNIC_MAC_TBL_MAX_ENTRIES); + port_info->vport.max_smac_ent = + cpu_to_be16(OPA_VNIC_MAX_SMAC_LIMIT); + port_info->vport.oper_state = OPA_VNIC_STATE_DROP_ALL; + port_info->vport.config_state = OPA_VNIC_STATE_DROP_ALL; +} + +/** + * vema_add_vport -- Add a new vnic port + * @port: ptr to opa_vnic_vema_port struct + * @vport_num: vnic port number (to be added) + * + * Return a pointer to the vnic adapter structure + */ +static struct opa_vnic_adapter *vema_add_vport(struct opa_vnic_vema_port *port, + u8 vport_num) +{ + struct opa_vnic_ctrl_port *cport = port->cport; + struct opa_vnic_adapter *adapter; + + adapter = opa_vnic_add_netdev(cport->ibdev, port->port_num, vport_num); + if (!IS_ERR(adapter)) { + int rc; + + adapter->cport = cport; + rc = idr_alloc(&port->vport_idr, adapter, vport_num, + vport_num + 1, GFP_NOWAIT); + if (rc < 0) { + opa_vnic_rem_netdev(adapter); + adapter = ERR_PTR(rc); + } + } + + return adapter; +} + +/** + * vema_get_class_port_info -- Get class info for port + * @port: Port on whic MAD was received + * @recvd_mad: pointer to the received mad + * @rsp_mad: pointer to respose mad + * + * This function copies the latest class port info value set for the + * port and stores it for generating traps + */ +static void vema_get_class_port_info(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + struct opa_class_port_info *port_info; + + port_info = (struct opa_class_port_info *)rsp_mad->data; + memcpy(port_info, &port->class_port_info, sizeof(*port_info)); + port_info->base_version = OPA_MGMT_BASE_VERSION, + port_info->class_version = OPA_EMA_CLASS_VERSION; + + /* + * Set capability mask bit indicating agent generates traps, + * and set the maximum number of VNIC ports supported. + */ + port_info->cap_mask = cpu_to_be16((OPA_VNIC_CLASS_CAP_TRAP | + (OPA_VNIC_MAX_NUM_VPORT << 8))); + + /* + * Since a get routine is always sent by the EM first we + * set the expected response time to + * 4.096 usec * 2^18 == 1.0737 sec here. + */ + port_info->cap_mask2_resp_time = cpu_to_be32(18); +} + +/** + * vema_set_class_port_info -- Get class info for port + * @port: Port on whic MAD was received + * @recvd_mad: pointer to the received mad + * @rsp_mad: pointer to respose mad + * + * This function updates the port class info for the specific vnic + * and sets up the response mad data + */ +static void vema_set_class_port_info(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + memcpy(&port->class_port_info, recvd_mad->data, + sizeof(port->class_port_info)); + + vema_get_class_port_info(port, recvd_mad, rsp_mad); +} + +/** + * vema_get_veswport_info -- Get veswport info + * @port: source port on which MAD was received + * @recvd_mad: pointer to the received mad + * @rsp_mad: pointer to respose mad + */ +static void vema_get_veswport_info(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + struct opa_veswport_info *port_info = + (struct opa_veswport_info *)rsp_mad->data; + struct opa_vnic_adapter *adapter; + + adapter = vema_get_vport_adapter(recvd_mad, port); + if (adapter) { + memset(port_info, 0, sizeof(*port_info)); + opa_vnic_get_vesw_info(adapter, &port_info->vesw); + opa_vnic_get_per_veswport_info(adapter, + &port_info->vport); + } else { + vema_get_pod_values(port_info); + } +} + +/** + * vema_set_veswport_info -- Set veswport info + * @port: source port on which MAD was received + * @recvd_mad: pointer to the received mad + * @rsp_mad: pointer to respose mad + * + * This function gets the port class infor for vnic + */ +static void vema_set_veswport_info(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + struct opa_vnic_ctrl_port *cport = port->cport; + struct opa_veswport_info *port_info; + struct opa_vnic_adapter *adapter; + u8 vport_num; + + vport_num = vema_get_vport_num(recvd_mad); + + adapter = vema_get_vport_adapter(recvd_mad, port); + if (!adapter) { + adapter = vema_add_vport(port, vport_num); + if (IS_ERR(adapter)) { + c_err("failed to add vport %d: %ld\n", + vport_num, PTR_ERR(adapter)); + goto err_exit; + } + } + + port_info = (struct opa_veswport_info *)recvd_mad->data; + opa_vnic_set_vesw_info(adapter, &port_info->vesw); + opa_vnic_set_per_veswport_info(adapter, &port_info->vport); + + /* Process the new config settings */ + opa_vnic_process_vema_config(adapter); + + vema_get_veswport_info(port, recvd_mad, rsp_mad); + return; + +err_exit: + rsp_mad->mad_hdr.status = OPA_VNIC_INVAL_ATTR; +} + +/** + * vema_get_mac_entries -- Get MAC entries in VNIC MAC table + * @port: source port on which MAD was received + * @recvd_mad: pointer to the received mad + * @rsp_mad: pointer to respose mad + * + * This function gets the MAC entries that are programmed into + * the VNIC MAC forwarding table. It checks for the validity of + * the index into the MAC table and the number of entries that + * are to be retrieved. + */ +static void vema_get_mac_entries(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + struct opa_veswport_mactable *mac_tbl_in, *mac_tbl_out; + struct opa_vnic_adapter *adapter; + + adapter = vema_get_vport_adapter(recvd_mad, port); + if (!adapter) { + rsp_mad->mad_hdr.status = OPA_VNIC_INVAL_ATTR; + return; + } + + mac_tbl_in = (struct opa_veswport_mactable *)recvd_mad->data; + mac_tbl_out = (struct opa_veswport_mactable *)rsp_mad->data; + + if (vema_mac_tbl_req_ok(mac_tbl_in)) { + mac_tbl_out->offset = mac_tbl_in->offset; + mac_tbl_out->num_entries = mac_tbl_in->num_entries; + opa_vnic_query_mac_tbl(adapter, mac_tbl_out); + } else { + rsp_mad->mad_hdr.status = OPA_VNIC_INVAL_ATTR; + } +} + +/** + * vema_set_mac_entries -- Set MAC entries in VNIC MAC table + * @port: source port on which MAD was received + * @recvd_mad: pointer to the received mad + * @rsp_mad: pointer to respose mad + * + * This function sets the MAC entries in the VNIC forwarding table + * It checks for the validity of the index and the number of forwarding + * table entries to be programmed. + */ +static void vema_set_mac_entries(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + struct opa_veswport_mactable *mac_tbl; + struct opa_vnic_adapter *adapter; + + adapter = vema_get_vport_adapter(recvd_mad, port); + if (!adapter) { + rsp_mad->mad_hdr.status = OPA_VNIC_INVAL_ATTR; + return; + } + + mac_tbl = (struct opa_veswport_mactable *)recvd_mad->data; + if (vema_mac_tbl_req_ok(mac_tbl)) { + if (opa_vnic_update_mac_tbl(adapter, mac_tbl)) + rsp_mad->mad_hdr.status = OPA_VNIC_UNSUP_ATTR; + } else { + rsp_mad->mad_hdr.status = OPA_VNIC_UNSUP_ATTR; + } + vema_get_mac_entries(port, recvd_mad, rsp_mad); +} + +/** + * vema_set_delete_vesw -- Reset VESW info to POD values + * @port: source port on which MAD was received + * @recvd_mad: pointer to the received mad + * @rsp_mad: pointer to respose mad + * + * This function clears all the fields of veswport info for the requested vesw + * and sets them back to the power-on default values. It does not delete the + * vesw. + */ +static void vema_set_delete_vesw(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + struct opa_veswport_info *port_info = + (struct opa_veswport_info *)rsp_mad->data; + struct opa_vnic_adapter *adapter; + + adapter = vema_get_vport_adapter(recvd_mad, port); + if (!adapter) { + rsp_mad->mad_hdr.status = OPA_VNIC_INVAL_ATTR; + return; + } + + vema_get_pod_values(port_info); + opa_vnic_set_vesw_info(adapter, &port_info->vesw); + opa_vnic_set_per_veswport_info(adapter, &port_info->vport); + + /* Process the new config settings */ + opa_vnic_process_vema_config(adapter); + + opa_vnic_release_mac_tbl(adapter); + + vema_get_veswport_info(port, recvd_mad, rsp_mad); +} + +/** + * vema_get_mac_list -- Get the unicast/multicast macs. + * @port: source port on which MAD was received + * @recvd_mad: Received mad contains fields to set vnic parameters + * @rsp_mad: Response mad to be built + * @attr_id: Attribute ID indicating multicast or unicast mac list + */ +static void vema_get_mac_list(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad, + u16 attr_id) +{ + struct opa_veswport_iface_macs *macs_in, *macs_out; + int max_entries = (OPA_VNIC_EMA_DATA - sizeof(*macs_out)) / ETH_ALEN; + struct opa_vnic_adapter *adapter; + + adapter = vema_get_vport_adapter(recvd_mad, port); + if (!adapter) { + rsp_mad->mad_hdr.status = OPA_VNIC_INVAL_ATTR; + return; + } + + macs_in = (struct opa_veswport_iface_macs *)recvd_mad->data; + macs_out = (struct opa_veswport_iface_macs *)rsp_mad->data; + + macs_out->start_idx = macs_in->start_idx; + if (macs_in->num_macs_in_msg) + macs_out->num_macs_in_msg = macs_in->num_macs_in_msg; + else + macs_out->num_macs_in_msg = cpu_to_be16(max_entries); + + if (attr_id == OPA_EM_ATTR_IFACE_MCAST_MACS) + opa_vnic_query_mcast_macs(adapter, macs_out); + else + opa_vnic_query_ucast_macs(adapter, macs_out); +} + +/** + * vema_get_summary_counters -- Gets summary counters. + * @port: source port on which MAD was received + * @recvd_mad: Received mad contains fields to set vnic parameters + * @rsp_mad: Response mad to be built + */ +static void vema_get_summary_counters(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + struct opa_veswport_summary_counters *cntrs; + struct opa_vnic_adapter *adapter; + + adapter = vema_get_vport_adapter(recvd_mad, port); + if (adapter) { + cntrs = (struct opa_veswport_summary_counters *)rsp_mad->data; + opa_vnic_get_summary_counters(adapter, cntrs); + } else { + rsp_mad->mad_hdr.status = OPA_VNIC_INVAL_ATTR; + } +} + +/** + * vema_get_error_counters -- Gets summary counters. + * @port: source port on which MAD was received + * @recvd_mad: Received mad contains fields to set vnic parameters + * @rsp_mad: Response mad to be built + */ +static void vema_get_error_counters(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + struct opa_veswport_error_counters *cntrs; + struct opa_vnic_adapter *adapter; + + adapter = vema_get_vport_adapter(recvd_mad, port); + if (adapter) { + cntrs = (struct opa_veswport_error_counters *)rsp_mad->data; + opa_vnic_get_error_counters(adapter, cntrs); + } else { + rsp_mad->mad_hdr.status = OPA_VNIC_INVAL_ATTR; + } +} + +/** + * vema_get -- Process received get MAD + * @port: source port on which MAD was received + * @recvd_mad: Received mad + * @rsp_mad: Response mad to be built + */ +static void vema_get(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + u16 attr_id = be16_to_cpu(recvd_mad->mad_hdr.attr_id); + + switch (attr_id) { + case OPA_EM_ATTR_CLASS_PORT_INFO: + vema_get_class_port_info(port, recvd_mad, rsp_mad); + break; + case OPA_EM_ATTR_VESWPORT_INFO: + vema_get_veswport_info(port, recvd_mad, rsp_mad); + break; + case OPA_EM_ATTR_VESWPORT_MAC_ENTRIES: + vema_get_mac_entries(port, recvd_mad, rsp_mad); + break; + case OPA_EM_ATTR_IFACE_UCAST_MACS: + /* fall through */ + case OPA_EM_ATTR_IFACE_MCAST_MACS: + vema_get_mac_list(port, recvd_mad, rsp_mad, attr_id); + break; + case OPA_EM_ATTR_VESWPORT_SUMMARY_COUNTERS: + vema_get_summary_counters(port, recvd_mad, rsp_mad); + break; + case OPA_EM_ATTR_VESWPORT_ERROR_COUNTERS: + vema_get_error_counters(port, recvd_mad, rsp_mad); + break; + default: + rsp_mad->mad_hdr.status = OPA_VNIC_UNSUP_ATTR; + break; + } +} + +/** + * vema_set -- Process received set MAD + * @port: source port on which MAD was received + * @recvd_mad: Received mad contains fields to set vnic parameters + * @rsp_mad: Response mad to be built + */ +static void vema_set(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + u16 attr_id = be16_to_cpu(recvd_mad->mad_hdr.attr_id); + + switch (attr_id) { + case OPA_EM_ATTR_CLASS_PORT_INFO: + vema_set_class_port_info(port, recvd_mad, rsp_mad); + break; + case OPA_EM_ATTR_VESWPORT_INFO: + vema_set_veswport_info(port, recvd_mad, rsp_mad); + break; + case OPA_EM_ATTR_VESWPORT_MAC_ENTRIES: + vema_set_mac_entries(port, recvd_mad, rsp_mad); + break; + case OPA_EM_ATTR_DELETE_VESW: + vema_set_delete_vesw(port, recvd_mad, rsp_mad); + break; + default: + rsp_mad->mad_hdr.status = OPA_VNIC_UNSUP_ATTR; + break; + } +} + +/** + * vema_send -- Send handler for VEMA MAD agent + * @mad_agent: pointer to the mad agent + * @mad_wc: pointer to mad send work completion information + * + * Free all the data structures associated with the sent MAD + */ +static void vema_send(struct ib_mad_agent *mad_agent, + struct ib_mad_send_wc *mad_wc) +{ + ib_destroy_ah(mad_wc->send_buf->ah); + ib_free_send_mad(mad_wc->send_buf); +} + +/** + * vema_recv -- Recv handler for VEMA MAD agent + * @mad_agent: pointer to the mad agent + * @send_buf: Send buffer if found, else NULL + * @mad_wc: pointer to mad send work completion information + * + * Handle only set and get methods and respond to other methods + * as unsupported. Allocate response buffer and address handle + * for the response MAD. + */ +static void vema_recv(struct ib_mad_agent *mad_agent, + struct ib_mad_send_buf *send_buf, + struct ib_mad_recv_wc *mad_wc) +{ + struct opa_vnic_vema_port *port; + struct ib_ah *ah; + struct ib_mad_send_buf *rsp; + struct opa_vnic_vema_mad *vema_mad; + + if (!mad_wc || !mad_wc->recv_buf.mad) + return; + + port = mad_agent->context; + ah = ib_create_ah_from_wc(mad_agent->qp->pd, mad_wc->wc, + mad_wc->recv_buf.grh, mad_agent->port_num); + if (IS_ERR(ah)) + goto free_recv_mad; + + rsp = ib_create_send_mad(mad_agent, mad_wc->wc->src_qp, + mad_wc->wc->pkey_index, 0, + IB_MGMT_VENDOR_HDR, OPA_VNIC_EMA_DATA, + GFP_KERNEL, OPA_MGMT_BASE_VERSION); + if (IS_ERR(rsp)) + goto err_rsp; + + rsp->ah = ah; + vema_mad = rsp->mad; + memcpy(vema_mad, mad_wc->recv_buf.mad, IB_MGMT_VENDOR_HDR); + vema_mad->mad_hdr.method = IB_MGMT_METHOD_GET_RESP; + vema_mad->mad_hdr.status = 0; + + /* Lock ensures network adapter is not removed */ + mutex_lock(&port->lock); + + switch (mad_wc->recv_buf.mad->mad_hdr.method) { + case IB_MGMT_METHOD_GET: + vema_get(port, (struct opa_vnic_vema_mad *)mad_wc->recv_buf.mad, + vema_mad); + break; + case IB_MGMT_METHOD_SET: + vema_set(port, (struct opa_vnic_vema_mad *)mad_wc->recv_buf.mad, + vema_mad); + break; + default: + vema_mad->mad_hdr.status = OPA_VNIC_UNSUP_ATTR; + break; + } + mutex_unlock(&port->lock); + + if (!ib_post_send_mad(rsp, NULL)) { + /* + * with post send successful ah and send mad + * will be destroyed in send handler + */ + goto free_recv_mad; + } + + ib_free_send_mad(rsp); + +err_rsp: + ib_destroy_ah(ah); +free_recv_mad: + ib_free_recv_mad(mad_wc); +} + +/** + * vema_get_port -- Gets the opa_vnic_vema_port + * @cport: pointer to control dev + * @port_num: Port number + * + * This function loops through the ports and returns + * the opa_vnic_vema port structure that is associated + * with the OPA port number + * + * Return: ptr to requested opa_vnic_vema_port strucure + * if success, NULL if not + */ +static struct opa_vnic_vema_port * +vema_get_port(struct opa_vnic_ctrl_port *cport, u8 port_num) +{ + struct opa_vnic_vema_port *port = (void *)cport + sizeof(*cport); + + if (port_num > cport->num_ports) + return NULL; + + return port + (port_num - 1); +} + +/** + * opa_vnic_vema_send_trap -- This function sends a trap to the EM + * @cport: pointer to vnic control port + * @data: pointer to trap data filled by calling function + * @lid: issuers lid (encap_slid from vesw_port_info) + * + * This function is called from the VNIC driver to send a trap if there + * is somethng the EM should be notified about. These events currently + * are + * 1) UNICAST INTERFACE MACADDRESS changes + * 2) MULTICAST INTERFACE MACADDRESS changes + * 3) ETHERNET LINK STATUS changes + * While allocating the send mad the remote site qpn used is 1 + * as this is the well known QP. + * + */ +void opa_vnic_vema_send_trap(struct opa_vnic_adapter *adapter, + struct __opa_veswport_trap *data, u32 lid) +{ + struct opa_vnic_ctrl_port *cport = adapter->cport; + struct ib_mad_send_buf *send_buf; + struct opa_vnic_vema_port *port; + struct ib_device *ibp; + struct opa_vnic_vema_mad_trap *trap_mad; + struct opa_class_port_info *class; + struct ib_ah_attr ah_attr; + struct ib_ah *ah; + struct opa_veswport_trap *trap; + u32 trap_lid; + u16 pkey_idx; + + if (!cport) + goto err_exit; + ibp = cport->ibdev; + port = vema_get_port(cport, data->opaportnum); + if (!port || !port->mad_agent) + goto err_exit; + + if (time_before(jiffies, adapter->trap_timeout)) { + if (adapter->trap_count == OPA_VNIC_TRAP_BURST_LIMIT) { + v_warn("Trap rate exceeded\n"); + goto err_exit; + } else { + adapter->trap_count++; + } + } else { + adapter->trap_count = 0; + } + + class = &port->class_port_info; + /* Set up address handle */ + memset(&ah_attr, 0, sizeof(ah_attr)); + ah_attr.sl = GET_TRAP_SL_FROM_CLASS_PORT_INFO(class->trap_sl_rsvd); + ah_attr.port_num = port->port_num; + trap_lid = be32_to_cpu(class->trap_lid); + /* + * check for trap lid validity, must not be zero + * The trap sink could change after we fashion the MAD but since traps + * are not guaranteed we won't use a lock as anyway the change will take + * place even with locking. + */ + if (!trap_lid) { + c_err("%s: Invalid dlid\n", __func__); + goto err_exit; + } + + ah_attr.dlid = trap_lid; + ah = ib_create_ah(port->mad_agent->qp->pd, &ah_attr); + if (IS_ERR(ah)) { + c_err("%s:Couldn't create new AH = %p\n", __func__, ah); + c_err("%s:dlid = %d, sl = %d, port = %d\n", __func__, + ah_attr.dlid, ah_attr.sl, ah_attr.port_num); + goto err_exit; + } + + if (ib_find_pkey(ibp, data->opaportnum, IB_DEFAULT_PKEY_FULL, + &pkey_idx) < 0) { + c_err("%s:full key not found, defaulting to partial\n", + __func__); + if (ib_find_pkey(ibp, data->opaportnum, IB_DEFAULT_PKEY_PARTIAL, + &pkey_idx) < 0) + pkey_idx = 1; + } + + 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); + if (IS_ERR(send_buf)) { + c_err("%s:Couldn't allocate send buf\n", __func__); + goto err_sndbuf; + } + + send_buf->ah = ah; + + /* Set up common MAD hdr */ + trap_mad = send_buf->mad; + trap_mad->mad_hdr.base_version = OPA_MGMT_BASE_VERSION; + trap_mad->mad_hdr.mgmt_class = OPA_MGMT_CLASS_INTEL_EMA; + trap_mad->mad_hdr.class_version = OPA_EMA_CLASS_VERSION; + trap_mad->mad_hdr.method = IB_MGMT_METHOD_TRAP; + port->tid++; + trap_mad->mad_hdr.tid = cpu_to_be64(port->tid); + trap_mad->mad_hdr.attr_id = IB_SMP_ATTR_NOTICE; + + /* Set up vendor OUI */ + trap_mad->oui[0] = INTEL_OUI_1; + trap_mad->oui[1] = INTEL_OUI_2; + trap_mad->oui[2] = INTEL_OUI_3; + + /* Setup notice attribute portion */ + trap_mad->notice.gen_type = OPA_INTEL_EMA_NOTICE_TYPE_INFO << 1; + trap_mad->notice.oui_1 = INTEL_OUI_1; + trap_mad->notice.oui_2 = INTEL_OUI_2; + trap_mad->notice.oui_3 = INTEL_OUI_3; + trap_mad->notice.issuer_lid = cpu_to_be32(lid); + + /* copy the actual trap data */ + trap = (struct opa_veswport_trap *)trap_mad->notice.raw_data; + trap->fabric_id = cpu_to_be16(data->fabric_id); + trap->veswid = cpu_to_be16(data->veswid); + trap->veswportnum = cpu_to_be32(data->veswportnum); + trap->opaportnum = cpu_to_be16(data->opaportnum); + trap->veswportindex = data->veswportindex; + trap->opcode = data->opcode; + + /* If successful send set up rate limit timeout else bail */ + if (ib_post_send_mad(send_buf, NULL)) { + ib_free_send_mad(send_buf); + } else { + if (adapter->trap_count) + return; + adapter->trap_timeout = jiffies + + usecs_to_jiffies(OPA_VNIC_TRAP_TIMEOUT); + return; + } + +err_sndbuf: + ib_destroy_ah(ah); +err_exit: + v_err("Aborting trap\n"); +} + +static int vema_rem_vport(int id, void *p, void *data) +{ + struct opa_vnic_adapter *adapter = p; + + opa_vnic_rem_netdev(adapter); + return 0; +} + +static int vema_enable_vport(int id, void *p, void *data) +{ + struct opa_vnic_adapter *adapter = p; + + netif_carrier_on(adapter->netdev); + return 0; +} + +static int vema_disable_vport(int id, void *p, void *data) +{ + struct opa_vnic_adapter *adapter = p; + + netif_carrier_off(adapter->netdev); + return 0; +} + +static void opa_vnic_event(struct ib_event_handler *handler, + struct ib_event *record) +{ + struct opa_vnic_vema_port *port = + container_of(handler, struct opa_vnic_vema_port, event_handler); + struct opa_vnic_ctrl_port *cport = port->cport; + + if (record->element.port_num != port->port_num) + return; + + c_dbg("OPA_VNIC received event %d on device %s port %d\n", + record->event, record->device->name, record->element.port_num); + + if (record->event == IB_EVENT_PORT_ERR) + idr_for_each(&port->vport_idr, vema_disable_vport, NULL); + if (record->event == IB_EVENT_PORT_ACTIVE) + idr_for_each(&port->vport_idr, vema_enable_vport, NULL); +} + +/** + * vema_unregister -- Unregisters agent + * @cport: pointer to control port + * + * This deletes the registration by VEMA for MADs + */ +static void vema_unregister(struct opa_vnic_ctrl_port *cport) +{ + int i; + + for (i = 1; i <= cport->num_ports; i++) { + struct opa_vnic_vema_port *port = vema_get_port(cport, i); + + if (!port->mad_agent) + continue; + + /* Lock ensures no MAD is being processed */ + mutex_lock(&port->lock); + idr_for_each(&port->vport_idr, vema_rem_vport, NULL); + mutex_unlock(&port->lock); + + ib_unregister_mad_agent(port->mad_agent); + port->mad_agent = NULL; + mutex_destroy(&port->lock); + idr_destroy(&port->vport_idr); + ib_unregister_event_handler(&port->event_handler); + } +} + +/** + * vema_register -- Registers agent + * @cport: pointer to control port + * + * This function registers the handlers for the VEMA MADs + * + * Return: returns 0 on success. non zero otherwise + */ +static int vema_register(struct opa_vnic_ctrl_port *cport) +{ + struct ib_mad_reg_req reg_req = { + .mgmt_class = OPA_MGMT_CLASS_INTEL_EMA, + .mgmt_class_version = OPA_MGMT_BASE_VERSION, + .oui = { INTEL_OUI_1, INTEL_OUI_2, INTEL_OUI_3 } + }; + int i; + + set_bit(IB_MGMT_METHOD_GET, reg_req.method_mask); + set_bit(IB_MGMT_METHOD_SET, reg_req.method_mask); + + /* register ib event handler and mad agent for each port on dev */ + for (i = 1; i <= cport->num_ports; i++) { + struct opa_vnic_vema_port *port = vema_get_port(cport, i); + int ret; + + port->cport = cport; + port->port_num = i; + + INIT_IB_EVENT_HANDLER(&port->event_handler, + cport->ibdev, opa_vnic_event); + ret = ib_register_event_handler(&port->event_handler); + if (ret) { + c_err("port %d: event handler register failed\n", i); + vema_unregister(cport); + return ret; + } + + idr_init(&port->vport_idr); + mutex_init(&port->lock); + port->mad_agent = ib_register_mad_agent(cport->ibdev, i, + IB_QPT_GSI, ®_req, + IB_MGMT_RMPP_VERSION, + vema_send, vema_recv, + port, 0); + if (IS_ERR(port->mad_agent)) { + ret = PTR_ERR(port->mad_agent); + port->mad_agent = NULL; + mutex_destroy(&port->lock); + idr_destroy(&port->vport_idr); + vema_unregister(cport); + return ret; + } + } + + return 0; +} + +/** + * opa_vnic_vema_add_one -- Handle new ib device + * @device: ib device pointer + * + * Allocate the vnic control port and initialize it. + */ +static void opa_vnic_vema_add_one(struct ib_device *device) +{ + struct opa_vnic_ctrl_port *cport; + int rc, size = sizeof(*cport); + + if (!rdma_cap_opa_vnic(device)) + return; + + size += device->phys_port_cnt * sizeof(struct opa_vnic_vema_port); + cport = kzalloc(size, GFP_KERNEL); + if (!cport) + return; + + cport->num_ports = device->phys_port_cnt; + cport->ibdev = device; + + /* Initialize opa vnic management agent (vema) */ + rc = vema_register(cport); + if (!rc) + c_info("VNIC client initialized\n"); + + ib_set_client_data(device, &opa_vnic_client, cport); +} + +/** + * opa_vnic_vema_rem_one -- Handle ib device removal + * @device: ib device pointer + * @client_data: ib client data + * + * Uninitialize and free the vnic control port. + */ +static void opa_vnic_vema_rem_one(struct ib_device *device, + void *client_data) +{ + struct opa_vnic_ctrl_port *cport = client_data; + + if (!cport) + return; + + c_info("removing VNIC client\n"); + vema_unregister(cport); + kfree(cport); +} + +static int __init opa_vnic_init(void) +{ + int rc; + + pr_info("OPA Virtual Network Driver - v%s\n", + opa_vnic_driver_version); + + rc = ib_register_client(&opa_vnic_client); + if (rc) + pr_err("VNIC driver register failed %d\n", rc); + + return rc; +} +module_init(opa_vnic_init); + +static void opa_vnic_deinit(void) +{ + ib_unregister_client(&opa_vnic_client); +} +module_exit(opa_vnic_deinit); + +MODULE_LICENSE("Dual BSD/GPL"); +MODULE_AUTHOR("Intel Corporation"); +MODULE_DESCRIPTION("Intel OPA Virtual Network driver"); +MODULE_VERSION(DRV_VERSION); diff --git a/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema_iface.c b/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema_iface.c new file mode 100644 index 000000000000..a51bf977f4d6 --- /dev/null +++ b/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema_iface.c @@ -0,0 +1,390 @@ +/* + * Copyright(c) 2017 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ + +/* + * This file contains OPA VNIC EMA Interface functions. + */ + +#include "opa_vnic_internal.h" + +/** + * opa_vnic_vema_report_event - sent trap to report the specified event + * @adapter: vnic port adapter + * @event: event to be reported + * + * This function calls vema api to sent a trap for the given event. + */ +void opa_vnic_vema_report_event(struct opa_vnic_adapter *adapter, u8 event) +{ + struct __opa_veswport_info *info = &adapter->info; + struct __opa_veswport_trap trap_data; + + trap_data.fabric_id = info->vesw.fabric_id; + trap_data.veswid = info->vesw.vesw_id; + trap_data.veswportnum = info->vport.port_num; + trap_data.opaportnum = adapter->port_num; + trap_data.veswportindex = adapter->vport_num; + trap_data.opcode = event; + + opa_vnic_vema_send_trap(adapter, &trap_data, info->vport.encap_slid); +} + +/** + * opa_vnic_get_error_counters - get summary counters + * @adapter: vnic port adapter + * @cntrs: pointer to destination summary counters structure + * + * This function populates the summary counters that is maintained by the + * given adapter to destination address provided. + */ +void opa_vnic_get_summary_counters(struct opa_vnic_adapter *adapter, + struct opa_veswport_summary_counters *cntrs) +{ + struct opa_vnic_stats vstats; + __be64 *dst; + u64 *src; + + memset(&vstats, 0, sizeof(vstats)); + mutex_lock(&adapter->stats_lock); + adapter->rn_ops->ndo_get_stats64(adapter->netdev, &vstats.netstats); + mutex_unlock(&adapter->stats_lock); + + cntrs->vp_instance = cpu_to_be16(adapter->vport_num); + cntrs->vesw_id = cpu_to_be16(adapter->info.vesw.vesw_id); + cntrs->veswport_num = cpu_to_be32(adapter->port_num); + + cntrs->tx_errors = cpu_to_be64(vstats.netstats.tx_errors); + cntrs->rx_errors = cpu_to_be64(vstats.netstats.rx_errors); + cntrs->tx_packets = cpu_to_be64(vstats.netstats.tx_packets); + cntrs->rx_packets = cpu_to_be64(vstats.netstats.rx_packets); + cntrs->tx_bytes = cpu_to_be64(vstats.netstats.tx_bytes); + cntrs->rx_bytes = cpu_to_be64(vstats.netstats.rx_bytes); + + /* + * This loop depends on layout of + * opa_veswport_summary_counters opa_vnic_stats structures. + */ + for (dst = &cntrs->tx_unicast, src = &vstats.tx_grp.unicast; + dst < &cntrs->reserved[0]; dst++, src++) { + *dst = cpu_to_be64(*src); + } +} + +/** + * opa_vnic_get_error_counters - get error counters + * @adapter: vnic port adapter + * @cntrs: pointer to destination error counters structure + * + * This function populates the error counters that is maintained by the + * given adapter to destination address provided. + */ +void opa_vnic_get_error_counters(struct opa_vnic_adapter *adapter, + struct opa_veswport_error_counters *cntrs) +{ + struct opa_vnic_stats vstats; + + memset(&vstats, 0, sizeof(vstats)); + mutex_lock(&adapter->stats_lock); + adapter->rn_ops->ndo_get_stats64(adapter->netdev, &vstats.netstats); + mutex_unlock(&adapter->stats_lock); + + cntrs->vp_instance = cpu_to_be16(adapter->vport_num); + cntrs->vesw_id = cpu_to_be16(adapter->info.vesw.vesw_id); + cntrs->veswport_num = cpu_to_be32(adapter->port_num); + + cntrs->tx_errors = cpu_to_be64(vstats.netstats.tx_errors); + cntrs->rx_errors = cpu_to_be64(vstats.netstats.rx_errors); + cntrs->tx_dlid_zero = cpu_to_be64(vstats.tx_dlid_zero); + cntrs->tx_drop_state = cpu_to_be64(vstats.tx_drop_state); + cntrs->tx_logic = cpu_to_be64(vstats.netstats.tx_fifo_errors + + vstats.netstats.tx_carrier_errors); + + cntrs->rx_bad_veswid = cpu_to_be64(vstats.netstats.rx_nohandler); + cntrs->rx_runt = cpu_to_be64(vstats.rx_runt); + cntrs->rx_oversize = cpu_to_be64(vstats.rx_oversize); + cntrs->rx_drop_state = cpu_to_be64(vstats.rx_drop_state); + cntrs->rx_logic = cpu_to_be64(vstats.netstats.rx_fifo_errors); +} + +/** + * opa_vnic_get_vesw_info -- Get the vesw information + * @adapter: vnic port adapter + * @info: pointer to destination vesw info structure + * + * This function copies the vesw info that is maintained by the + * given adapter to destination address provided. + */ +void opa_vnic_get_vesw_info(struct opa_vnic_adapter *adapter, + struct opa_vesw_info *info) +{ + struct __opa_vesw_info *src = &adapter->info.vesw; + int i; + + info->fabric_id = cpu_to_be16(src->fabric_id); + info->vesw_id = cpu_to_be16(src->vesw_id); + memcpy(info->rsvd0, src->rsvd0, ARRAY_SIZE(src->rsvd0)); + info->def_port_mask = cpu_to_be16(src->def_port_mask); + memcpy(info->rsvd1, src->rsvd1, ARRAY_SIZE(src->rsvd1)); + info->pkey = cpu_to_be16(src->pkey); + + memcpy(info->rsvd2, src->rsvd2, ARRAY_SIZE(src->rsvd2)); + info->u_mcast_dlid = cpu_to_be32(src->u_mcast_dlid); + for (i = 0; i < OPA_VESW_MAX_NUM_DEF_PORT; i++) + info->u_ucast_dlid[i] = cpu_to_be32(src->u_ucast_dlid[i]); + + memcpy(info->rsvd3, src->rsvd3, ARRAY_SIZE(src->rsvd3)); + for (i = 0; i < OPA_VNIC_MAX_NUM_PCP; i++) + info->eth_mtu[i] = cpu_to_be16(src->eth_mtu[i]); + + info->eth_mtu_non_vlan = cpu_to_be16(src->eth_mtu_non_vlan); + memcpy(info->rsvd4, src->rsvd4, ARRAY_SIZE(src->rsvd4)); +} + +/** + * opa_vnic_set_vesw_info -- Set the vesw information + * @adapter: vnic port adapter + * @info: pointer to vesw info structure + * + * This function updates the vesw info that is maintained by the + * given adapter with vesw info provided. Reserved fields are stored + * and returned back to EM as is. + */ +void opa_vnic_set_vesw_info(struct opa_vnic_adapter *adapter, + struct opa_vesw_info *info) +{ + struct __opa_vesw_info *dst = &adapter->info.vesw; + int i; + + dst->fabric_id = be16_to_cpu(info->fabric_id); + dst->vesw_id = be16_to_cpu(info->vesw_id); + memcpy(dst->rsvd0, info->rsvd0, ARRAY_SIZE(info->rsvd0)); + dst->def_port_mask = be16_to_cpu(info->def_port_mask); + memcpy(dst->rsvd1, info->rsvd1, ARRAY_SIZE(info->rsvd1)); + dst->pkey = be16_to_cpu(info->pkey); + + memcpy(dst->rsvd2, info->rsvd2, ARRAY_SIZE(info->rsvd2)); + dst->u_mcast_dlid = be32_to_cpu(info->u_mcast_dlid); + for (i = 0; i < OPA_VESW_MAX_NUM_DEF_PORT; i++) + dst->u_ucast_dlid[i] = be32_to_cpu(info->u_ucast_dlid[i]); + + memcpy(dst->rsvd3, info->rsvd3, ARRAY_SIZE(info->rsvd3)); + for (i = 0; i < OPA_VNIC_MAX_NUM_PCP; i++) + dst->eth_mtu[i] = be16_to_cpu(info->eth_mtu[i]); + + dst->eth_mtu_non_vlan = be16_to_cpu(info->eth_mtu_non_vlan); + memcpy(dst->rsvd4, info->rsvd4, ARRAY_SIZE(info->rsvd4)); +} + +/** + * opa_vnic_get_per_veswport_info -- Get the vesw per port information + * @adapter: vnic port adapter + * @info: pointer to destination vport info structure + * + * This function copies the vesw per port info that is maintained by the + * given adapter to destination address provided. + * Note that the read only fields are not copied. + */ +void opa_vnic_get_per_veswport_info(struct opa_vnic_adapter *adapter, + struct opa_per_veswport_info *info) +{ + struct __opa_per_veswport_info *src = &adapter->info.vport; + + info->port_num = cpu_to_be32(src->port_num); + info->eth_link_status = src->eth_link_status; + memcpy(info->rsvd0, src->rsvd0, ARRAY_SIZE(src->rsvd0)); + + memcpy(info->base_mac_addr, src->base_mac_addr, + ARRAY_SIZE(info->base_mac_addr)); + info->config_state = src->config_state; + info->oper_state = src->oper_state; + info->max_mac_tbl_ent = cpu_to_be16(src->max_mac_tbl_ent); + info->max_smac_ent = cpu_to_be16(src->max_smac_ent); + info->mac_tbl_digest = cpu_to_be32(src->mac_tbl_digest); + memcpy(info->rsvd1, src->rsvd1, ARRAY_SIZE(src->rsvd1)); + + info->encap_slid = cpu_to_be32(src->encap_slid); + memcpy(info->pcp_to_sc_uc, src->pcp_to_sc_uc, + ARRAY_SIZE(info->pcp_to_sc_uc)); + memcpy(info->pcp_to_vl_uc, src->pcp_to_vl_uc, + ARRAY_SIZE(info->pcp_to_vl_uc)); + memcpy(info->pcp_to_sc_mc, src->pcp_to_sc_mc, + ARRAY_SIZE(info->pcp_to_sc_mc)); + memcpy(info->pcp_to_vl_mc, src->pcp_to_vl_mc, + ARRAY_SIZE(info->pcp_to_vl_mc)); + info->non_vlan_sc_uc = src->non_vlan_sc_uc; + info->non_vlan_vl_uc = src->non_vlan_vl_uc; + info->non_vlan_sc_mc = src->non_vlan_sc_mc; + info->non_vlan_vl_mc = src->non_vlan_vl_mc; + memcpy(info->rsvd2, src->rsvd2, ARRAY_SIZE(src->rsvd2)); + + info->uc_macs_gen_count = cpu_to_be16(src->uc_macs_gen_count); + info->mc_macs_gen_count = cpu_to_be16(src->mc_macs_gen_count); + memcpy(info->rsvd3, src->rsvd3, ARRAY_SIZE(src->rsvd3)); +} + +/** + * opa_vnic_set_per_veswport_info -- Set vesw per port information + * @adapter: vnic port adapter + * @info: pointer to vport info structure + * + * This function updates the vesw per port info that is maintained by the + * given adapter with vesw per port info provided. Reserved fields are + * stored and returned back to EM as is. + */ +void opa_vnic_set_per_veswport_info(struct opa_vnic_adapter *adapter, + struct opa_per_veswport_info *info) +{ + struct __opa_per_veswport_info *dst = &adapter->info.vport; + + dst->port_num = be32_to_cpu(info->port_num); + memcpy(dst->rsvd0, info->rsvd0, ARRAY_SIZE(info->rsvd0)); + + memcpy(dst->base_mac_addr, info->base_mac_addr, + ARRAY_SIZE(dst->base_mac_addr)); + dst->config_state = info->config_state; + memcpy(dst->rsvd1, info->rsvd1, ARRAY_SIZE(info->rsvd1)); + + dst->encap_slid = be32_to_cpu(info->encap_slid); + memcpy(dst->pcp_to_sc_uc, info->pcp_to_sc_uc, + ARRAY_SIZE(dst->pcp_to_sc_uc)); + memcpy(dst->pcp_to_vl_uc, info->pcp_to_vl_uc, + ARRAY_SIZE(dst->pcp_to_vl_uc)); + memcpy(dst->pcp_to_sc_mc, info->pcp_to_sc_mc, + ARRAY_SIZE(dst->pcp_to_sc_mc)); + memcpy(dst->pcp_to_vl_mc, info->pcp_to_vl_mc, + ARRAY_SIZE(dst->pcp_to_vl_mc)); + dst->non_vlan_sc_uc = info->non_vlan_sc_uc; + dst->non_vlan_vl_uc = info->non_vlan_vl_uc; + dst->non_vlan_sc_mc = info->non_vlan_sc_mc; + dst->non_vlan_vl_mc = info->non_vlan_vl_mc; + memcpy(dst->rsvd2, info->rsvd2, ARRAY_SIZE(info->rsvd2)); + memcpy(dst->rsvd3, info->rsvd3, ARRAY_SIZE(info->rsvd3)); +} + +/** + * opa_vnic_query_mcast_macs - query multicast mac list + * @adapter: vnic port adapter + * @macs: pointer mac list + * + * This function populates the provided mac list with the configured + * multicast addresses in the adapter. + */ +void opa_vnic_query_mcast_macs(struct opa_vnic_adapter *adapter, + struct opa_veswport_iface_macs *macs) +{ + u16 start_idx, num_macs, idx = 0, count = 0; + struct netdev_hw_addr *ha; + + start_idx = be16_to_cpu(macs->start_idx); + num_macs = be16_to_cpu(macs->num_macs_in_msg); + netdev_for_each_mc_addr(ha, adapter->netdev) { + struct opa_vnic_iface_mac_entry *entry = &macs->entry[count]; + + if (start_idx > idx++) + continue; + else if (num_macs == count) + break; + memcpy(entry, ha->addr, sizeof(*entry)); + count++; + } + + macs->tot_macs_in_lst = cpu_to_be16(netdev_mc_count(adapter->netdev)); + macs->num_macs_in_msg = cpu_to_be16(count); + macs->gen_count = cpu_to_be16(adapter->info.vport.mc_macs_gen_count); +} + +/** + * opa_vnic_query_ucast_macs - query unicast mac list + * @adapter: vnic port adapter + * @macs: pointer mac list + * + * This function populates the provided mac list with the configured + * unicast addresses in the adapter. + */ +void opa_vnic_query_ucast_macs(struct opa_vnic_adapter *adapter, + struct opa_veswport_iface_macs *macs) +{ + u16 start_idx, tot_macs, num_macs, idx = 0, count = 0; + struct netdev_hw_addr *ha; + + start_idx = be16_to_cpu(macs->start_idx); + num_macs = be16_to_cpu(macs->num_macs_in_msg); + /* loop through dev_addrs list first */ + for_each_dev_addr(adapter->netdev, ha) { + struct opa_vnic_iface_mac_entry *entry = &macs->entry[count]; + + /* Do not include EM specified MAC address */ + if (!memcmp(adapter->info.vport.base_mac_addr, ha->addr, + ARRAY_SIZE(adapter->info.vport.base_mac_addr))) + continue; + + if (start_idx > idx++) + continue; + else if (num_macs == count) + break; + memcpy(entry, ha->addr, sizeof(*entry)); + count++; + } + + /* loop through uc list */ + netdev_for_each_uc_addr(ha, adapter->netdev) { + struct opa_vnic_iface_mac_entry *entry = &macs->entry[count]; + + if (start_idx > idx++) + continue; + else if (num_macs == count) + break; + memcpy(entry, ha->addr, sizeof(*entry)); + count++; + } + + tot_macs = netdev_hw_addr_list_count(&adapter->netdev->dev_addrs) + + netdev_uc_count(adapter->netdev); + macs->tot_macs_in_lst = cpu_to_be16(tot_macs); + macs->num_macs_in_msg = cpu_to_be16(count); + macs->gen_count = cpu_to_be16(adapter->info.vport.uc_macs_gen_count); +} diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h index 1beb1ec2fbdf..74b765ce48ab 100644 --- a/include/linux/mlx4/device.h +++ b/include/linux/mlx4/device.h @@ -1011,8 +1011,7 @@ struct mlx4_mad_ifc { #define mlx4_foreach_ib_transport_port(port, dev) \ for ((port) = 1; (port) <= (dev)->caps.num_ports; (port)++) \ if (((dev)->caps.port_mask[port] == MLX4_PORT_TYPE_IB) || \ - ((dev)->caps.flags & MLX4_DEV_CAP_FLAG_IBOE) || \ - ((dev)->caps.flags2 & MLX4_DEV_CAP_FLAG2_ROCE_V1_V2)) + ((dev)->caps.port_mask[port] == MLX4_PORT_TYPE_ETH)) #define MLX4_INVALID_SLAVE_ID 0xFF #define MLX4_SINK_COUNTER_INDEX(dev) (dev->caps.max_counters - 1) diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 7c50bd39b297..954f42c268a4 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -236,7 +236,7 @@ struct mlx5_ifc_flow_table_fields_supported_bits { u8 outer_dmac[0x1]; u8 outer_smac[0x1]; u8 outer_ether_type[0x1]; - u8 reserved_at_3[0x1]; + u8 outer_ip_version[0x1]; u8 outer_first_prio[0x1]; u8 outer_first_cfi[0x1]; u8 outer_first_vid[0x1]; @@ -265,7 +265,7 @@ struct mlx5_ifc_flow_table_fields_supported_bits { u8 inner_dmac[0x1]; u8 inner_smac[0x1]; u8 inner_ether_type[0x1]; - u8 reserved_at_23[0x1]; + u8 inner_ip_version[0x1]; u8 inner_first_prio[0x1]; u8 inner_first_cfi[0x1]; u8 inner_first_vid[0x1]; @@ -371,7 +371,7 @@ struct mlx5_ifc_fte_match_set_lyr_2_4_bits { u8 cvlan_tag[0x1]; u8 svlan_tag[0x1]; u8 frag[0x1]; - u8 reserved_at_93[0x4]; + u8 ip_version[0x4]; u8 tcp_flags[0x9]; u8 tcp_sport[0x10]; @@ -4735,17 +4735,17 @@ struct mlx5_ifc_query_cong_statistics_out_bits { u8 reserved_at_40[0x40]; - u8 cur_flows[0x20]; + u8 rp_cur_flows[0x20]; u8 sum_flows[0x20]; - u8 cnp_ignored_high[0x20]; + u8 rp_cnp_ignored_high[0x20]; - u8 cnp_ignored_low[0x20]; + u8 rp_cnp_ignored_low[0x20]; - u8 cnp_handled_high[0x20]; + u8 rp_cnp_handled_high[0x20]; - u8 cnp_handled_low[0x20]; + u8 rp_cnp_handled_low[0x20]; u8 reserved_at_140[0x100]; @@ -4755,13 +4755,13 @@ struct mlx5_ifc_query_cong_statistics_out_bits { u8 accumulators_period[0x20]; - u8 ecn_marked_roce_packets_high[0x20]; + u8 np_ecn_marked_roce_packets_high[0x20]; - u8 ecn_marked_roce_packets_low[0x20]; + u8 np_ecn_marked_roce_packets_low[0x20]; - u8 cnps_sent_high[0x20]; + u8 np_cnp_sent_high[0x20]; - u8 cnps_sent_low[0x20]; + u8 np_cnp_sent_low[0x20]; u8 reserved_at_320[0x560]; }; diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 1aeb25dd42a7..2452e6449532 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -103,12 +103,24 @@ int addrconf_prefix_rcv_add_addr(struct net *net, struct net_device *dev, u32 addr_flags, bool sllao, bool tokenized, __u32 valid_lft, u32 prefered_lft); +static inline void addrconf_addr_eui48_base(u8 *eui, const char *const addr) +{ + memcpy(eui, addr, 3); + eui[3] = 0xFF; + eui[4] = 0xFE; + memcpy(eui + 5, addr + 3, 3); +} + +static inline void addrconf_addr_eui48(u8 *eui, const char *const addr) +{ + addrconf_addr_eui48_base(eui, addr); + eui[0] ^= 2; +} + static inline int addrconf_ifid_eui48(u8 *eui, struct net_device *dev) { if (dev->addr_len != ETH_ALEN) return -1; - memcpy(eui, dev->dev_addr, 3); - memcpy(eui + 5, dev->dev_addr + 3, 3); /* * The zSeries OSA network cards can be shared among various @@ -123,14 +135,16 @@ static inline int addrconf_ifid_eui48(u8 *eui, struct net_device *dev) * case. Hence the resulting interface identifier has local * scope according to RFC2373. */ + + addrconf_addr_eui48_base(eui, dev->dev_addr); + if (dev->dev_id) { eui[3] = (dev->dev_id >> 8) & 0xFF; eui[4] = dev->dev_id & 0xFF; } else { - eui[3] = 0xFF; - eui[4] = 0xFE; eui[0] ^= 2; } + return 0; } diff --git a/include/rdma/ib_hdrs.h b/include/rdma/ib_hdrs.h index c755325f0831..5519f31f043a 100644 --- a/include/rdma/ib_hdrs.h +++ b/include/rdma/ib_hdrs.h @@ -74,6 +74,12 @@ #define IB_GRH_FLOW_MASK 0xFFFFF #define IB_GRH_FLOW_SHIFT 0 #define IB_GRH_NEXT_HDR 0x1B +#define IB_FECN_SHIFT 31 +#define IB_FECN_MASK 1 +#define IB_FECN_SMASK BIT(IB_FECN_SHIFT) +#define IB_BECN_SHIFT 30 +#define IB_BECN_MASK 1 +#define IB_BECN_SMASK BIT(IB_BECN_SHIFT) #define IB_AETH_CREDIT_SHIFT 24 #define IB_AETH_CREDIT_MASK 0x1F @@ -181,4 +187,64 @@ static inline void put_ib_ateth_compare(u64 val, struct ib_atomic_eth *ateth) ib_u64_put(val, &ateth->compare_data); } +/* + * 9B/IB Packet Format + */ +#define IB_LNH_MASK 3 +#define IB_SC_MASK 0xf +#define IB_SC_SHIFT 12 +#define IB_SL_MASK 0xf +#define IB_SL_SHIFT 4 + +static inline u8 ib_get_lnh(struct ib_header *hdr) +{ + return (be16_to_cpu(hdr->lrh[0]) & IB_LNH_MASK); +} + +static inline u8 ib_get_sc(struct ib_header *hdr) +{ + return ((be16_to_cpu(hdr->lrh[0]) >> IB_SC_SHIFT) & IB_SC_MASK); +} + +static inline u8 ib_get_sl(struct ib_header *hdr) +{ + return ((be16_to_cpu(hdr->lrh[0]) >> IB_SL_SHIFT) & IB_SL_MASK); +} + +static inline u16 ib_get_dlid(struct ib_header *hdr) +{ + return (be16_to_cpu(hdr->lrh[1])); +} + +static inline u16 ib_get_slid(struct ib_header *hdr) +{ + return (be16_to_cpu(hdr->lrh[3])); +} + +/* + * BTH + */ +#define IB_BTH_OPCODE_MASK 0xff +#define IB_BTH_OPCODE_SHIFT 24 +#define IB_BTH_PAD_MASK 3 +#define IB_BTH_PKEY_MASK 0xffff +#define IB_BTH_PAD_SHIFT 20 + +static inline u8 ib_bth_get_pad(struct ib_other_headers *ohdr) +{ + return ((be32_to_cpu(ohdr->bth[0]) >> IB_BTH_PAD_SHIFT) & + IB_BTH_PAD_MASK); +} + +static inline u16 ib_bth_get_pkey(struct ib_other_headers *ohdr) +{ + return (be32_to_cpu(ohdr->bth[0]) & IB_BTH_PKEY_MASK); +} + +static inline u8 ib_bth_get_opcode(struct ib_other_headers *ohdr) +{ + return ((be32_to_cpu(ohdr->bth[0]) >> IB_BTH_OPCODE_SHIFT) & + IB_BTH_OPCODE_MASK); +} + #endif /* IB_HDRS_H */ diff --git a/include/rdma/ib_mad.h b/include/rdma/ib_mad.h index 981214b3790c..3d5f5d6031ec 100644 --- a/include/rdma/ib_mad.h +++ b/include/rdma/ib_mad.h @@ -262,6 +262,31 @@ struct ib_class_port_info { __be32 trap_qkey; }; +struct opa_class_port_info { + u8 base_version; + u8 class_version; + __be16 cap_mask; + __be32 cap_mask2_resp_time; + + u8 redirect_gid[16]; + __be32 redirect_tc_fl; + __be32 redirect_lid; + __be32 redirect_sl_qp; + __be32 redirect_qkey; + + u8 trap_gid[16]; + __be32 trap_tc_fl; + __be32 trap_lid; + __be32 trap_hl_qp; + __be32 trap_qkey; + + __be16 trap_pkey; + __be16 redirect_pkey; + + u8 trap_sl_rsvd; + u8 reserved[3]; +} __packed; + /** * ib_get_cpi_resp_time - Returns the resp_time value from * cap_mask2_resp_time in ib_class_port_info. @@ -673,7 +698,7 @@ struct ib_mad_agent *ib_register_mad_snoop(struct ib_device *device, * After invoking this routine, MAD services are no longer usable by the * client on the associated QP. */ -int ib_unregister_mad_agent(struct ib_mad_agent *mad_agent); +void ib_unregister_mad_agent(struct ib_mad_agent *mad_agent); /** * ib_post_send_mad - Posts MAD(s) to the send queue of the QP associated diff --git a/include/rdma/ib_pack.h b/include/rdma/ib_pack.h index b13419ce99ff..36655899ee02 100644 --- a/include/rdma/ib_pack.h +++ b/include/rdma/ib_pack.h @@ -80,6 +80,8 @@ enum { IB_OPCODE_UD = 0x60, /* per IBTA 1.3 vol 1 Table 38, A10.3.2 */ IB_OPCODE_CNP = 0x80, + /* Manufacturer specific */ + IB_OPCODE_MSP = 0xe0, /* operations -- just used to define real constants */ IB_OPCODE_SEND_FIRST = 0x00, diff --git a/include/rdma/ib_sa.h b/include/rdma/ib_sa.h index fd0e53219f93..843b56288ecf 100644 --- a/include/rdma/ib_sa.h +++ b/include/rdma/ib_sa.h @@ -56,6 +56,7 @@ enum { IB_SA_METHOD_GET_TRACE_TBL = 0x13 }; +#define OPA_SA_CLASS_VERSION 0x80 enum { IB_SA_ATTR_CLASS_PORTINFO = 0x01, IB_SA_ATTR_NOTICE = 0x02, @@ -454,14 +455,8 @@ int ib_sa_guid_info_rec_query(struct ib_sa_client *client, void *context, struct ib_sa_query **sa_query); -/* Support get SA ClassPortInfo */ -int ib_sa_classport_info_rec_query(struct ib_sa_client *client, - struct ib_device *device, u8 port_num, - int timeout_ms, gfp_t gfp_mask, - void (*callback)(int status, - struct ib_class_port_info *resp, - void *context), - void *context, - struct ib_sa_query **sa_query); +bool ib_sa_sendonly_fullmem_support(struct ib_sa_client *client, + struct ib_device *device, + u8 port_num); #endif /* IB_SA_H */ diff --git a/include/rdma/ib_umem.h b/include/rdma/ib_umem.h index 2d83cfd7e6ce..23159dd5be18 100644 --- a/include/rdma/ib_umem.h +++ b/include/rdma/ib_umem.h @@ -44,7 +44,7 @@ struct ib_umem { struct ib_ucontext *context; size_t length; unsigned long address; - int page_size; + int page_shift; int writable; int hugetlb; struct work_struct work; @@ -60,7 +60,7 @@ struct ib_umem { /* Returns the offset of the umem start relative to the first page. */ static inline int ib_umem_offset(struct ib_umem *umem) { - return umem->address & ((unsigned long)umem->page_size - 1); + return umem->address & (BIT(umem->page_shift) - 1); } /* Returns the first page of an ODP umem. */ @@ -72,12 +72,12 @@ static inline unsigned long ib_umem_start(struct ib_umem *umem) /* Returns the address of the page after the last one of an ODP umem. */ static inline unsigned long ib_umem_end(struct ib_umem *umem) { - return PAGE_ALIGN(umem->address + umem->length); + return ALIGN(umem->address + umem->length, BIT(umem->page_shift)); } static inline size_t ib_umem_num_pages(struct ib_umem *umem) { - return (ib_umem_end(umem) - ib_umem_start(umem)) >> PAGE_SHIFT; + return (ib_umem_end(umem) - ib_umem_start(umem)) >> umem->page_shift; } #ifdef CONFIG_INFINIBAND_USER_MEM diff --git a/include/rdma/ib_umem_odp.h b/include/rdma/ib_umem_odp.h index 542cd8b3414c..fb67554aabd6 100644 --- a/include/rdma/ib_umem_odp.h +++ b/include/rdma/ib_umem_odp.h @@ -84,7 +84,8 @@ struct ib_umem_odp { #ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING -int ib_umem_odp_get(struct ib_ucontext *context, struct ib_umem *umem); +int ib_umem_odp_get(struct ib_ucontext *context, struct ib_umem *umem, + int access); struct ib_umem *ib_alloc_odp_umem(struct ib_ucontext *context, unsigned long addr, size_t size); @@ -154,7 +155,8 @@ static inline int ib_umem_mmu_notifier_retry(struct ib_umem *item, #else /* CONFIG_INFINIBAND_ON_DEMAND_PAGING */ static inline int ib_umem_odp_get(struct ib_ucontext *context, - struct ib_umem *umem) + struct ib_umem *umem, + int access) { return -EINVAL; } diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 43ac7ba75685..9d0d55d844d4 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -55,6 +55,7 @@ #include <net/ip.h> #include <linux/string.h> #include <linux/slab.h> +#include <linux/netdevice.h> #include <linux/if_link.h> #include <linux/atomic.h> @@ -225,6 +226,7 @@ enum ib_device_cap_flags { IB_DEVICE_VIRTUAL_FUNCTION = (1ULL << 33), /* Deprecated. Please use IB_RAW_PACKET_CAP_SCATTER_FCS. */ IB_DEVICE_RAW_SCATTER_FCS = (1ULL << 34), + IB_DEVICE_RDMA_NETDEV_OPA_VNIC = (1ULL << 35), }; enum ib_signature_prot_cap { @@ -432,7 +434,8 @@ enum ib_port_speed { IB_SPEED_QDR = 4, IB_SPEED_FDR10 = 8, IB_SPEED_FDR = 16, - IB_SPEED_EDR = 32 + IB_SPEED_EDR = 32, + IB_SPEED_HDR = 64 }; /** @@ -499,6 +502,7 @@ static inline struct rdma_hw_stats *rdma_alloc_hw_stats_struct( /* Address format 0x000FF000 */ #define RDMA_CORE_CAP_AF_IB 0x00001000 #define RDMA_CORE_CAP_ETH_AH 0x00002000 +#define RDMA_CORE_CAP_OPA_AH 0x00004000 /* Protocol 0xFFF00000 */ #define RDMA_CORE_CAP_PROT_IB 0x00100000 @@ -1337,6 +1341,7 @@ enum ib_access_flags { IB_ACCESS_MW_BIND = (1<<4), IB_ZERO_BASED = (1<<5), IB_ACCESS_ON_DEMAND = (1<<6), + IB_ACCESS_HUGETLB = (1<<7), }; /* @@ -1358,6 +1363,17 @@ struct ib_fmr_attr { struct ib_umem; +enum rdma_remove_reason { + /* Userspace requested uobject deletion. Call could fail */ + RDMA_REMOVE_DESTROY, + /* Context deletion. This call should delete the actual object itself */ + RDMA_REMOVE_CLOSE, + /* Driver is being hot-unplugged. This call should delete the actual object itself */ + RDMA_REMOVE_DRIVER_REMOVE, + /* Context is being cleaned-up, but commit was just completed */ + RDMA_REMOVE_DURING_CLEANUP, +}; + struct ib_rdmacg_object { #ifdef CONFIG_CGROUP_RDMA struct rdma_cgroup *cg; /* owner rdma cgroup */ @@ -1366,19 +1382,16 @@ struct ib_rdmacg_object { struct ib_ucontext { struct ib_device *device; - struct list_head pd_list; - struct list_head mr_list; - struct list_head mw_list; - struct list_head cq_list; - struct list_head qp_list; - struct list_head srq_list; - struct list_head ah_list; - struct list_head xrcd_list; - struct list_head rule_list; - struct list_head wq_list; - struct list_head rwq_ind_tbl_list; + struct ib_uverbs_file *ufile; int closing; + /* locking the uobjects_list */ + struct mutex uobjects_lock; + struct list_head uobjects; + /* protects cleanup process from other actions */ + struct rw_semaphore cleanup_rwsem; + enum rdma_remove_reason cleanup_reason; + struct pid *tgid; #ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING struct rb_root umem_tree; @@ -1408,9 +1421,16 @@ struct ib_uobject { struct ib_rdmacg_object cg_obj; /* rdmacg object */ int id; /* index into kernel idr */ struct kref ref; - struct rw_semaphore mutex; /* protects .live */ + atomic_t usecnt; /* protects exclusive access */ struct rcu_head rcu; /* kfree_rcu() overhead */ - int live; + + const struct uverbs_obj_type *type; +}; + +struct ib_uobject_file { + struct ib_uobject uobj; + /* ufile contains the lock between context release and file close */ + struct ib_uverbs_file *ufile; }; struct ib_udata { @@ -1663,6 +1683,7 @@ enum ib_flow_spec_type { IB_FLOW_SPEC_INNER = 0x100, /* Actions */ IB_FLOW_SPEC_ACTION_TAG = 0x1000, + IB_FLOW_SPEC_ACTION_DROP = 0x1001, }; #define IB_FLOW_SPEC_LAYER_MASK 0xF0 #define IB_FLOW_SPEC_SUPPORT_LAYERS 8 @@ -1791,6 +1812,11 @@ struct ib_flow_spec_action_tag { u32 tag_id; }; +struct ib_flow_spec_action_drop { + enum ib_flow_spec_type type; + u16 size; +}; + union ib_flow_spec { struct { u32 type; @@ -1803,6 +1829,7 @@ union ib_flow_spec { struct ib_flow_spec_ipv6 ipv6; struct ib_flow_spec_tunnel tunnel; struct ib_flow_spec_action_tag flow_tag; + struct ib_flow_spec_action_drop drop; }; struct ib_flow_attr { @@ -1861,6 +1888,34 @@ struct ib_port_immutable { u32 max_mad_size; }; +/* rdma netdev type - specifies protocol type */ +enum rdma_netdev_t { + RDMA_NETDEV_OPA_VNIC, + RDMA_NETDEV_IPOIB, +}; + +/** + * struct rdma_netdev - rdma netdev + * For cases where netstack interfacing is required. + */ +struct rdma_netdev { + void *clnt_priv; + struct ib_device *hca; + u8 port_num; + + /* control functions */ + void (*set_id)(struct net_device *netdev, int id); + /* send packet */ + int (*send)(struct net_device *dev, struct sk_buff *skb, + struct ib_ah *address, u32 dqpn); + /* multicast */ + int (*attach_mcast)(struct net_device *dev, struct ib_device *hca, + union ib_gid *gid, u16 mlid, + int set_qkey, u32 qkey); + int (*detach_mcast)(struct net_device *dev, struct ib_device *hca, + union ib_gid *gid, u16 mlid); +}; + struct ib_device { /* Do not access @dma_device directly from ULP nor from HW drivers. */ struct device *dma_device; @@ -2114,6 +2169,20 @@ struct ib_device { struct ib_rwq_ind_table_init_attr *init_attr, struct ib_udata *udata); int (*destroy_rwq_ind_table)(struct ib_rwq_ind_table *wq_ind_table); + /** + * rdma netdev operations + * + * Driver implementing alloc_rdma_netdev must return -EOPNOTSUPP if it + * doesn't support the specified rdma netdev type. + */ + struct net_device *(*alloc_rdma_netdev)( + struct ib_device *device, + u8 port_num, + enum rdma_netdev_t type, + const char *name, + unsigned char name_assign_type, + void (*setup)(struct net_device *)); + void (*free_rdma_netdev)(struct net_device *netdev); struct module *owner; struct device dev; @@ -2536,6 +2605,21 @@ static inline bool rdma_cap_eth_ah(const struct ib_device *device, u8 port_num) } /** + * rdma_cap_opa_ah - Check if the port of device supports + * OPA Address handles + * @device: Device to check + * @port_num: Port number to check + * + * Return: true if we are running on an OPA device which supports + * the extended OPA addressing. + */ +static inline bool rdma_cap_opa_ah(struct ib_device *device, u8 port_num) +{ + return (device->port_immutable[port_num].core_cap_flags & + RDMA_CORE_CAP_OPA_AH) == RDMA_CORE_CAP_OPA_AH; +} + +/** * rdma_max_mad_size - Return the max MAD size required by this RDMA Port. * * @device: Device diff --git a/include/rdma/opa_port_info.h b/include/rdma/opa_port_info.h index 9303e0e4f508..b4f0ac02f283 100644 --- a/include/rdma/opa_port_info.h +++ b/include/rdma/opa_port_info.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2014 Intel Corporation. All rights reserved. + * Copyright (c) 2014-2017 Intel Corporation. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU @@ -127,6 +127,7 @@ #define OPA_LINK_WIDTH_3X 0x0004 #define OPA_LINK_WIDTH_4X 0x0008 +#define OPA_CAP_MASK3_IsEthOnFabricSupported (1 << 13) #define OPA_CAP_MASK3_IsSnoopSupported (1 << 7) #define OPA_CAP_MASK3_IsAsyncSC2VLSupported (1 << 6) #define OPA_CAP_MASK3_IsAddrRangeConfigSupported (1 << 5) diff --git a/include/rdma/opa_vnic.h b/include/rdma/opa_vnic.h new file mode 100644 index 000000000000..39d6890616a6 --- /dev/null +++ b/include/rdma/opa_vnic.h @@ -0,0 +1,141 @@ +#ifndef _OPA_VNIC_H +#define _OPA_VNIC_H +/* + * Copyright(c) 2017 Intel Corporation. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * - Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * - Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ + +/* + * This file contains Intel Omni-Path (OPA) Virtual Network Interface + * Controller (VNIC) specific declarations. + */ + +#include <rdma/ib_verbs.h> + +/* VNIC uses 16B header format */ +#define OPA_VNIC_L2_TYPE 0x2 + +/* 16 header bytes + 2 reserved bytes */ +#define OPA_VNIC_L2_HDR_LEN (16 + 2) + +#define OPA_VNIC_L4_HDR_LEN 2 + +#define OPA_VNIC_HDR_LEN (OPA_VNIC_L2_HDR_LEN + \ + OPA_VNIC_L4_HDR_LEN) + +#define OPA_VNIC_L4_ETHR 0x78 + +#define OPA_VNIC_ICRC_LEN 4 +#define OPA_VNIC_TAIL_LEN 1 +#define OPA_VNIC_ICRC_TAIL_LEN (OPA_VNIC_ICRC_LEN + OPA_VNIC_TAIL_LEN) + +#define OPA_VNIC_SKB_MDATA_LEN 4 +#define OPA_VNIC_SKB_MDATA_ENCAP_ERR 0x1 + +/* opa vnic rdma netdev's private data structure */ +struct opa_vnic_rdma_netdev { + struct rdma_netdev rn; /* keep this first */ + /* followed by device private data */ + char *dev_priv[0]; +}; + +static inline void *opa_vnic_priv(const struct net_device *dev) +{ + struct rdma_netdev *rn = netdev_priv(dev); + + return rn->clnt_priv; +} + +static inline void *opa_vnic_dev_priv(const struct net_device *dev) +{ + struct opa_vnic_rdma_netdev *oparn = netdev_priv(dev); + + return oparn->dev_priv; +} + +/* opa_vnic skb meta data structrue */ +struct opa_vnic_skb_mdata { + u8 vl; + u8 entropy; + u8 flags; + u8 rsvd; +} __packed; + +/* OPA VNIC group statistics */ +struct opa_vnic_grp_stats { + u64 unicast; + u64 mcastbcast; + u64 untagged; + u64 vlan; + u64 s_64; + u64 s_65_127; + u64 s_128_255; + u64 s_256_511; + u64 s_512_1023; + u64 s_1024_1518; + u64 s_1519_max; +}; + +struct opa_vnic_stats { + /* standard netdev statistics */ + struct rtnl_link_stats64 netstats; + + /* OPA VNIC statistics */ + struct opa_vnic_grp_stats tx_grp; + struct opa_vnic_grp_stats rx_grp; + u64 tx_dlid_zero; + u64 tx_drop_state; + u64 rx_drop_state; + u64 rx_runt; + u64 rx_oversize; +}; + +static inline bool rdma_cap_opa_vnic(struct ib_device *device) +{ + return !!(device->attrs.device_cap_flags & + IB_DEVICE_RDMA_NETDEV_OPA_VNIC); +} + +#endif /* _OPA_VNIC_H */ diff --git a/include/rdma/rdma_vt.h b/include/rdma/rdma_vt.h index 8fc1ca7b6f23..7f8956893526 100644 --- a/include/rdma/rdma_vt.h +++ b/include/rdma/rdma_vt.h @@ -516,6 +516,7 @@ int rvt_rkey_ok(struct rvt_qp *qp, struct rvt_sge *sge, u32 len, u64 vaddr, u32 rkey, int acc); int rvt_lkey_ok(struct rvt_lkey_table *rkt, struct rvt_pd *pd, struct rvt_sge *isge, struct ib_sge *sge, int acc); -struct rvt_mcast *rvt_mcast_find(struct rvt_ibport *ibp, union ib_gid *mgid); +struct rvt_mcast *rvt_mcast_find(struct rvt_ibport *ibp, union ib_gid *mgid, + u16 lid); #endif /* DEF_RDMA_VT_H */ diff --git a/include/rdma/rdmavt_qp.h b/include/rdma/rdmavt_qp.h index f3816396c76a..75a077c59742 100644 --- a/include/rdma/rdmavt_qp.h +++ b/include/rdma/rdmavt_qp.h @@ -2,7 +2,7 @@ #define DEF_RDMAVT_INCQP_H /* - * Copyright(c) 2016 Intel Corporation. + * Copyright(c) 2016, 2017 Intel Corporation. * * This file is provided under a dual BSD/GPLv2 license. When using or * redistributing this file, you may do so under either license. @@ -435,9 +435,14 @@ struct rvt_mcast_qp { struct rvt_qp *qp; }; +struct rvt_mcast_addr { + union ib_gid mgid; + u16 lid; +}; + struct rvt_mcast { struct rb_node rb_node; - union ib_gid mgid; + struct rvt_mcast_addr mcast_addr; struct list_head qp_list; wait_queue_head_t wait; atomic_t refcount; @@ -526,7 +531,6 @@ static inline void rvt_qp_wqe_reserve( struct rvt_qp *qp, struct rvt_swqe *wqe) { - wqe->wr.send_flags |= RVT_SEND_RESERVE_USED; atomic_inc(&qp->s_reserved_used); } @@ -550,7 +554,6 @@ static inline void rvt_qp_wqe_unreserve( struct rvt_swqe *wqe) { if (unlikely(wqe->wr.send_flags & RVT_SEND_RESERVE_USED)) { - wqe->wr.send_flags &= ~RVT_SEND_RESERVE_USED; atomic_dec(&qp->s_reserved_used); /* insure no compiler re-order up to s_last change */ smp_mb__after_atomic(); @@ -574,6 +577,7 @@ extern const enum ib_wc_opcode ib_rvt_wc_opcode[]; static inline void rvt_qp_swqe_complete( struct rvt_qp *qp, struct rvt_swqe *wqe, + enum ib_wc_opcode opcode, enum ib_wc_status status) { if (unlikely(wqe->wr.send_flags & RVT_SEND_RESERVE_USED)) @@ -586,7 +590,7 @@ static inline void rvt_qp_swqe_complete( memset(&wc, 0, sizeof(wc)); wc.wr_id = wqe->wr.wr_id; wc.status = status; - wc.opcode = ib_rvt_wc_opcode[wqe->wr.opcode]; + wc.opcode = opcode; wc.qp = &qp->ibqp; wc.byte_len = wqe->length; rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.send_cq), &wc, diff --git a/include/rdma/uverbs_std_types.h b/include/rdma/uverbs_std_types.h new file mode 100644 index 000000000000..7771ce966952 --- /dev/null +++ b/include/rdma/uverbs_std_types.h @@ -0,0 +1,114 @@ +/* + * Copyright (c) 2017, Mellanox Technologies inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef _UVERBS_STD_TYPES__ +#define _UVERBS_STD_TYPES__ + +#include <rdma/uverbs_types.h> + +extern const struct uverbs_obj_fd_type uverbs_type_attrs_comp_channel; +extern const struct uverbs_obj_idr_type uverbs_type_attrs_cq; +extern const struct uverbs_obj_idr_type uverbs_type_attrs_qp; +extern const struct uverbs_obj_idr_type uverbs_type_attrs_rwq_ind_table; +extern const struct uverbs_obj_idr_type uverbs_type_attrs_wq; +extern const struct uverbs_obj_idr_type uverbs_type_attrs_srq; +extern const struct uverbs_obj_idr_type uverbs_type_attrs_ah; +extern const struct uverbs_obj_idr_type uverbs_type_attrs_flow; +extern const struct uverbs_obj_idr_type uverbs_type_attrs_mr; +extern const struct uverbs_obj_idr_type uverbs_type_attrs_mw; +extern const struct uverbs_obj_idr_type uverbs_type_attrs_pd; +extern const struct uverbs_obj_idr_type uverbs_type_attrs_xrcd; + +static inline struct ib_uobject *__uobj_get(const struct uverbs_obj_type *type, + bool write, + struct ib_ucontext *ucontext, + int id) +{ + return rdma_lookup_get_uobject(type, ucontext, id, write); +} + +#define uobj_get_type(_type) uverbs_type_attrs_##_type.type + +#define uobj_get_read(_type, _id, _ucontext) \ + __uobj_get(&(_type), false, _ucontext, _id) + +#define uobj_get_obj_read(_type, _id, _ucontext) \ +({ \ + struct ib_uobject *uobj = \ + __uobj_get(&uobj_get_type(_type), \ + false, _ucontext, _id); \ + \ + (struct ib_##_type *)(IS_ERR(uobj) ? NULL : uobj->object); \ +}) + +#define uobj_get_write(_type, _id, _ucontext) \ + __uobj_get(&(_type), true, _ucontext, _id) + +static inline void uobj_put_read(struct ib_uobject *uobj) +{ + rdma_lookup_put_uobject(uobj, false); +} + +#define uobj_put_obj_read(_obj) \ + uobj_put_read((_obj)->uobject) + +static inline void uobj_put_write(struct ib_uobject *uobj) +{ + rdma_lookup_put_uobject(uobj, true); +} + +static inline int __must_check uobj_remove_commit(struct ib_uobject *uobj) +{ + return rdma_remove_commit_uobject(uobj); +} + +static inline void uobj_alloc_commit(struct ib_uobject *uobj) +{ + rdma_alloc_commit_uobject(uobj); +} + +static inline void uobj_alloc_abort(struct ib_uobject *uobj) +{ + rdma_alloc_abort_uobject(uobj); +} + +static inline struct ib_uobject *__uobj_alloc(const struct uverbs_obj_type *type, + struct ib_ucontext *ucontext) +{ + return rdma_alloc_begin_uobject(type, ucontext); +} + +#define uobj_alloc(_type, ucontext) \ + __uobj_alloc(&(_type), ucontext) + +#endif + diff --git a/include/rdma/uverbs_types.h b/include/rdma/uverbs_types.h new file mode 100644 index 000000000000..351ea185df44 --- /dev/null +++ b/include/rdma/uverbs_types.h @@ -0,0 +1,172 @@ +/* + * Copyright (c) 2017, Mellanox Technologies inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef _UVERBS_TYPES_ +#define _UVERBS_TYPES_ + +#include <linux/kernel.h> +#include <rdma/ib_verbs.h> + +struct uverbs_obj_type; + +struct uverbs_obj_type_class { + /* + * Get an ib_uobject that corresponds to the given id from ucontext, + * These functions could create or destroy objects if required. + * The action will be finalized only when commit, abort or put fops are + * called. + * The flow of the different actions is: + * [alloc]: Starts with alloc_begin. The handlers logic is than + * executed. If the handler is successful, alloc_commit + * is called and the object is inserted to the repository. + * Once alloc_commit completes the object is visible to + * other threads and userspace. + e Otherwise, alloc_abort is called and the object is + * destroyed. + * [lookup]: Starts with lookup_get which fetches and locks the + * object. After the handler finished using the object, it + * needs to call lookup_put to unlock it. The exclusive + * flag indicates if the object is locked for exclusive + * access. + * [remove]: Starts with lookup_get with exclusive flag set. This + * locks the object for exclusive access. If the handler + * code completed successfully, remove_commit is called + * and the ib_uobject is removed from the context's + * uobjects repository and put. The object itself is + * destroyed as well. Once remove succeeds new krefs to + * the object cannot be acquired by other threads or + * userspace and the hardware driver is removed from the + * object. Other krefs on the object may still exist. + * If the handler code failed, lookup_put should be + * called. This callback is used when the context + * is destroyed as well (process termination, + * reset flow). + */ + struct ib_uobject *(*alloc_begin)(const struct uverbs_obj_type *type, + struct ib_ucontext *ucontext); + void (*alloc_commit)(struct ib_uobject *uobj); + void (*alloc_abort)(struct ib_uobject *uobj); + + struct ib_uobject *(*lookup_get)(const struct uverbs_obj_type *type, + struct ib_ucontext *ucontext, int id, + bool exclusive); + void (*lookup_put)(struct ib_uobject *uobj, bool exclusive); + /* + * Must be called with the exclusive lock held. If successful uobj is + * invalid on return. On failure uobject is left completely + * unchanged + */ + int __must_check (*remove_commit)(struct ib_uobject *uobj, + enum rdma_remove_reason why); + u8 needs_kfree_rcu; +}; + +struct uverbs_obj_type { + const struct uverbs_obj_type_class * const type_class; + size_t obj_size; + unsigned int destroy_order; +}; + +/* + * Objects type classes which support a detach state (object is still alive but + * it's not attached to any context need to make sure: + * (a) no call through to a driver after a detach is called + * (b) detach isn't called concurrently with context_cleanup + */ + +struct uverbs_obj_idr_type { + /* + * In idr based objects, uverbs_obj_type_class points to a generic + * idr operations. In order to specialize the underlying types (e.g. CQ, + * QPs, etc.), we add destroy_object specific callbacks. + */ + struct uverbs_obj_type type; + + /* Free driver resources from the uobject, make the driver uncallable, + * and move the uobject to the detached state. If the object was + * destroyed by the user's request, a failure should leave the uobject + * completely unchanged. + */ + int __must_check (*destroy_object)(struct ib_uobject *uobj, + enum rdma_remove_reason why); +}; + +struct ib_uobject *rdma_lookup_get_uobject(const struct uverbs_obj_type *type, + struct ib_ucontext *ucontext, + int id, bool exclusive); +void rdma_lookup_put_uobject(struct ib_uobject *uobj, bool exclusive); +struct ib_uobject *rdma_alloc_begin_uobject(const struct uverbs_obj_type *type, + struct ib_ucontext *ucontext); +void rdma_alloc_abort_uobject(struct ib_uobject *uobj); +int __must_check rdma_remove_commit_uobject(struct ib_uobject *uobj); +int rdma_alloc_commit_uobject(struct ib_uobject *uobj); + +struct uverbs_obj_fd_type { + /* + * In fd based objects, uverbs_obj_type_ops points to generic + * fd operations. In order to specialize the underlying types (e.g. + * completion_channel), we use fops, name and flags for fd creation. + * context_closed is called when the context is closed either when + * the driver is removed or the process terminated. + */ + struct uverbs_obj_type type; + int (*context_closed)(struct ib_uobject_file *uobj_file, + enum rdma_remove_reason why); + const struct file_operations *fops; + const char *name; + int flags; +}; + +extern const struct uverbs_obj_type_class uverbs_idr_class; +extern const struct uverbs_obj_type_class uverbs_fd_class; + +#define UVERBS_BUILD_BUG_ON(cond) (sizeof(char[1 - 2 * !!(cond)]) - \ + sizeof(char)) +#define UVERBS_TYPE_ALLOC_FD(_size, _order) \ + { \ + .destroy_order = _order, \ + .type_class = &uverbs_fd_class, \ + .obj_size = (_size) + \ + UVERBS_BUILD_BUG_ON((_size) < \ + sizeof(struct ib_uobject_file)),\ + } +#define UVERBS_TYPE_ALLOC_IDR_SZ(_size, _order) \ + { \ + .destroy_order = _order, \ + .type_class = &uverbs_idr_class, \ + .obj_size = (_size) + \ + UVERBS_BUILD_BUG_ON((_size) < \ + sizeof(struct ib_uobject)), \ + } +#define UVERBS_TYPE_ALLOC_IDR(_order) \ + UVERBS_TYPE_ALLOC_IDR_SZ(sizeof(struct ib_uobject), _order) +#endif diff --git a/include/uapi/linux/pci_regs.h b/include/uapi/linux/pci_regs.h index fff521c9458c..d56bb0051009 100644 --- a/include/uapi/linux/pci_regs.h +++ b/include/uapi/linux/pci_regs.h @@ -630,6 +630,7 @@ #define PCI_EXP_DEVCTL2_COMP_TIMEOUT 0x000f /* Completion Timeout Value */ #define PCI_EXP_DEVCTL2_ARI 0x0020 /* Alternative Routing-ID */ #define PCI_EXP_DEVCTL2_ATOMIC_REQ 0x0040 /* Set Atomic requests */ +#define PCI_EXP_DEVCTL2_ATOMIC_EGRESS_BLOCK 0x0080 /* Block atomic egress */ #define PCI_EXP_DEVCTL2_IDO_REQ_EN 0x0100 /* Allow IDO for requests */ #define PCI_EXP_DEVCTL2_IDO_CMP_EN 0x0200 /* Allow IDO for completions */ #define PCI_EXP_DEVCTL2_LTR_EN 0x0400 /* Enable LTR mechanism */ diff --git a/include/uapi/rdma/ib_user_verbs.h b/include/uapi/rdma/ib_user_verbs.h index 8edce2b65903..270c350bedc6 100644 --- a/include/uapi/rdma/ib_user_verbs.h +++ b/include/uapi/rdma/ib_user_verbs.h @@ -947,6 +947,17 @@ struct ib_uverbs_flow_spec_action_tag { __u32 reserved1; }; +struct ib_uverbs_flow_spec_action_drop { + union { + struct ib_uverbs_flow_spec_hdr hdr; + struct { + __u32 type; + __u16 size; + __u16 reserved; + }; + }; +}; + struct ib_uverbs_flow_tunnel_filter { __be32 tunnel_id; }; diff --git a/include/uapi/rdma/vmw_pvrdma-abi.h b/include/uapi/rdma/vmw_pvrdma-abi.h index 5016abc9ee97..c8c1d2d6df4d 100644 --- a/include/uapi/rdma/vmw_pvrdma-abi.h +++ b/include/uapi/rdma/vmw_pvrdma-abi.h @@ -222,7 +222,7 @@ struct pvrdma_sq_wqe_hdr { __u32 opcode; /* operation type */ __u32 send_flags; /* wr flags */ union { - __u32 imm_data; + __be32 imm_data; __u32 invalidate_rkey; } ex; __u32 reserved; @@ -273,7 +273,7 @@ struct pvrdma_cqe { __u32 opcode; __u32 status; __u32 byte_len; - __u32 imm_data; + __be32 imm_data; __u32 src_qp; __u32 wc_flags; __u32 vendor_err; |