diff options
Diffstat (limited to 'drivers/infiniband/core')
| -rw-r--r-- | drivers/infiniband/core/Makefile | 12 | ||||
| -rw-r--r-- | drivers/infiniband/core/agent.c | 373 | ||||
| -rw-r--r-- | drivers/infiniband/core/agent.h | 55 | ||||
| -rw-r--r-- | drivers/infiniband/core/agent_priv.h | 63 | ||||
| -rw-r--r-- | drivers/infiniband/core/cache.c | 365 | ||||
| -rw-r--r-- | drivers/infiniband/core/core_priv.h | 52 | ||||
| -rw-r--r-- | drivers/infiniband/core/device.c | 614 | ||||
| -rw-r--r-- | drivers/infiniband/core/fmr_pool.c | 507 | ||||
| -rw-r--r-- | drivers/infiniband/core/mad.c | 2714 | ||||
| -rw-r--r-- | drivers/infiniband/core/mad_priv.h | 199 | ||||
| -rw-r--r-- | drivers/infiniband/core/packer.c | 201 | ||||
| -rw-r--r-- | drivers/infiniband/core/sa_query.c | 866 | ||||
| -rw-r--r-- | drivers/infiniband/core/smi.c | 234 | ||||
| -rw-r--r-- | drivers/infiniband/core/smi.h | 67 | ||||
| -rw-r--r-- | drivers/infiniband/core/sysfs.c | 762 | ||||
| -rw-r--r-- | drivers/infiniband/core/ud_header.c | 365 | ||||
| -rw-r--r-- | drivers/infiniband/core/user_mad.c | 840 | ||||
| -rw-r--r-- | drivers/infiniband/core/verbs.c | 434 |
18 files changed, 8723 insertions, 0 deletions
diff --git a/drivers/infiniband/core/Makefile b/drivers/infiniband/core/Makefile new file mode 100644 index 000000000000..d2dbfb52c0a3 --- /dev/null +++ b/drivers/infiniband/core/Makefile @@ -0,0 +1,12 @@ +EXTRA_CFLAGS += -Idrivers/infiniband/include + +obj-$(CONFIG_INFINIBAND) += ib_core.o ib_mad.o ib_sa.o ib_umad.o + +ib_core-y := packer.o ud_header.o verbs.o sysfs.o \ + device.o fmr_pool.o cache.o + +ib_mad-y := mad.o smi.o agent.o + +ib_sa-y := sa_query.o + +ib_umad-y := user_mad.o diff --git a/drivers/infiniband/core/agent.c b/drivers/infiniband/core/agent.c new file mode 100644 index 000000000000..7aee5ebf3f01 --- /dev/null +++ b/drivers/infiniband/core/agent.c @@ -0,0 +1,373 @@ +/* + * Copyright (c) 2004 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2004 Infinicon Corporation. All rights reserved. + * Copyright (c) 2004 Intel Corporation. All rights reserved. + * Copyright (c) 2004 Topspin Corporation. All rights reserved. + * Copyright (c) 2004 Voltaire Corporation. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + * + * $Id: agent.c 1389 2004-12-27 22:56:47Z roland $ + */ + +#include <linux/dma-mapping.h> + +#include <asm/bug.h> + +#include <ib_smi.h> + +#include "smi.h" +#include "agent_priv.h" +#include "mad_priv.h" +#include "agent.h" + +spinlock_t ib_agent_port_list_lock; +static LIST_HEAD(ib_agent_port_list); + +/* + * Caller must hold ib_agent_port_list_lock + */ +static inline struct ib_agent_port_private * +__ib_get_agent_port(struct ib_device *device, int port_num, + struct ib_mad_agent *mad_agent) +{ + struct ib_agent_port_private *entry; + + BUG_ON(!(!!device ^ !!mad_agent)); /* Exactly one MUST be (!NULL) */ + + if (device) { + list_for_each_entry(entry, &ib_agent_port_list, port_list) { + if (entry->smp_agent->device == device && + entry->port_num == port_num) + return entry; + } + } else { + list_for_each_entry(entry, &ib_agent_port_list, port_list) { + if ((entry->smp_agent == mad_agent) || + (entry->perf_mgmt_agent == mad_agent)) + return entry; + } + } + return NULL; +} + +static inline struct ib_agent_port_private * +ib_get_agent_port(struct ib_device *device, int port_num, + struct ib_mad_agent *mad_agent) +{ + struct ib_agent_port_private *entry; + unsigned long flags; + + spin_lock_irqsave(&ib_agent_port_list_lock, flags); + entry = __ib_get_agent_port(device, port_num, mad_agent); + spin_unlock_irqrestore(&ib_agent_port_list_lock, flags); + + return entry; +} + +int smi_check_local_dr_smp(struct ib_smp *smp, + struct ib_device *device, + int port_num) +{ + struct ib_agent_port_private *port_priv; + + if (smp->mgmt_class != IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE) + return 1; + port_priv = ib_get_agent_port(device, port_num, NULL); + if (!port_priv) { + printk(KERN_DEBUG SPFX "smi_check_local_dr_smp %s port %d " + "not open\n", + device->name, port_num); + return 1; + } + + return smi_check_local_smp(port_priv->smp_agent, smp); +} + +static int agent_mad_send(struct ib_mad_agent *mad_agent, + struct ib_agent_port_private *port_priv, + struct ib_mad_private *mad_priv, + struct ib_grh *grh, + struct ib_wc *wc) +{ + struct ib_agent_send_wr *agent_send_wr; + struct ib_sge gather_list; + struct ib_send_wr send_wr; + struct ib_send_wr *bad_send_wr; + struct ib_ah_attr ah_attr; + unsigned long flags; + int ret = 1; + + agent_send_wr = kmalloc(sizeof(*agent_send_wr), GFP_KERNEL); + if (!agent_send_wr) + goto out; + agent_send_wr->mad = mad_priv; + + /* PCI mapping */ + gather_list.addr = dma_map_single(mad_agent->device->dma_device, + &mad_priv->mad, + sizeof(mad_priv->mad), + DMA_TO_DEVICE); + gather_list.length = sizeof(mad_priv->mad); + gather_list.lkey = (*port_priv->mr).lkey; + + send_wr.next = NULL; + send_wr.opcode = IB_WR_SEND; + send_wr.sg_list = &gather_list; + send_wr.num_sge = 1; + send_wr.wr.ud.remote_qpn = wc->src_qp; /* DQPN */ + send_wr.wr.ud.timeout_ms = 0; + send_wr.send_flags = IB_SEND_SIGNALED | IB_SEND_SOLICITED; + + ah_attr.dlid = wc->slid; + ah_attr.port_num = mad_agent->port_num; + ah_attr.src_path_bits = wc->dlid_path_bits; + ah_attr.sl = wc->sl; + ah_attr.static_rate = 0; + ah_attr.ah_flags = 0; /* No GRH */ + if (mad_priv->mad.mad.mad_hdr.mgmt_class == IB_MGMT_CLASS_PERF_MGMT) { + if (wc->wc_flags & IB_WC_GRH) { + ah_attr.ah_flags = IB_AH_GRH; + /* Should sgid be looked up ? */ + ah_attr.grh.sgid_index = 0; + ah_attr.grh.hop_limit = grh->hop_limit; + ah_attr.grh.flow_label = be32_to_cpup( + &grh->version_tclass_flow) & 0xfffff; + ah_attr.grh.traffic_class = (be32_to_cpup( + &grh->version_tclass_flow) >> 20) & 0xff; + memcpy(ah_attr.grh.dgid.raw, + grh->sgid.raw, + sizeof(ah_attr.grh.dgid)); + } + } + + agent_send_wr->ah = ib_create_ah(mad_agent->qp->pd, &ah_attr); + if (IS_ERR(agent_send_wr->ah)) { + printk(KERN_ERR SPFX "No memory for address handle\n"); + kfree(agent_send_wr); + goto out; + } + + send_wr.wr.ud.ah = agent_send_wr->ah; + if (mad_priv->mad.mad.mad_hdr.mgmt_class == IB_MGMT_CLASS_PERF_MGMT) { + send_wr.wr.ud.pkey_index = wc->pkey_index; + send_wr.wr.ud.remote_qkey = IB_QP1_QKEY; + } else { /* for SMPs */ + send_wr.wr.ud.pkey_index = 0; + send_wr.wr.ud.remote_qkey = 0; + } + send_wr.wr.ud.mad_hdr = &mad_priv->mad.mad.mad_hdr; + send_wr.wr_id = (unsigned long)agent_send_wr; + + pci_unmap_addr_set(agent_send_wr, mapping, gather_list.addr); + + /* Send */ + spin_lock_irqsave(&port_priv->send_list_lock, flags); + if (ib_post_send_mad(mad_agent, &send_wr, &bad_send_wr)) { + spin_unlock_irqrestore(&port_priv->send_list_lock, flags); + dma_unmap_single(mad_agent->device->dma_device, + pci_unmap_addr(agent_send_wr, mapping), + sizeof(mad_priv->mad), + DMA_TO_DEVICE); + ib_destroy_ah(agent_send_wr->ah); + kfree(agent_send_wr); + } else { + list_add_tail(&agent_send_wr->send_list, + &port_priv->send_posted_list); + spin_unlock_irqrestore(&port_priv->send_list_lock, flags); + ret = 0; + } + +out: + return ret; +} + +int agent_send(struct ib_mad_private *mad, + struct ib_grh *grh, + struct ib_wc *wc, + struct ib_device *device, + int port_num) +{ + struct ib_agent_port_private *port_priv; + struct ib_mad_agent *mad_agent; + + port_priv = ib_get_agent_port(device, port_num, NULL); + if (!port_priv) { + printk(KERN_DEBUG SPFX "agent_send %s port %d not open\n", + device->name, port_num); + return 1; + } + + /* Get mad agent based on mgmt_class in MAD */ + switch (mad->mad.mad.mad_hdr.mgmt_class) { + case IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE: + case IB_MGMT_CLASS_SUBN_LID_ROUTED: + mad_agent = port_priv->smp_agent; + break; + case IB_MGMT_CLASS_PERF_MGMT: + mad_agent = port_priv->perf_mgmt_agent; + break; + default: + return 1; + } + + return agent_mad_send(mad_agent, port_priv, mad, grh, wc); +} + +static void agent_send_handler(struct ib_mad_agent *mad_agent, + struct ib_mad_send_wc *mad_send_wc) +{ + struct ib_agent_port_private *port_priv; + struct ib_agent_send_wr *agent_send_wr; + unsigned long flags; + + /* Find matching MAD agent */ + port_priv = ib_get_agent_port(NULL, 0, mad_agent); + if (!port_priv) { + printk(KERN_ERR SPFX "agent_send_handler: no matching MAD " + "agent %p\n", mad_agent); + return; + } + + agent_send_wr = (struct ib_agent_send_wr *)(unsigned long)mad_send_wc->wr_id; + spin_lock_irqsave(&port_priv->send_list_lock, flags); + /* Remove completed send from posted send MAD list */ + list_del(&agent_send_wr->send_list); + spin_unlock_irqrestore(&port_priv->send_list_lock, flags); + + /* Unmap PCI */ + dma_unmap_single(mad_agent->device->dma_device, + pci_unmap_addr(agent_send_wr, mapping), + sizeof(agent_send_wr->mad->mad), + DMA_TO_DEVICE); + + ib_destroy_ah(agent_send_wr->ah); + + /* Release allocated memory */ + kmem_cache_free(ib_mad_cache, agent_send_wr->mad); + kfree(agent_send_wr); +} + +int ib_agent_port_open(struct ib_device *device, int port_num) +{ + int ret; + struct ib_agent_port_private *port_priv; + unsigned long flags; + + /* First, check if port already open for SMI */ + port_priv = ib_get_agent_port(device, port_num, NULL); + if (port_priv) { + printk(KERN_DEBUG SPFX "%s port %d already open\n", + device->name, port_num); + return 0; + } + + /* Create new device info */ + port_priv = kmalloc(sizeof *port_priv, GFP_KERNEL); + if (!port_priv) { + printk(KERN_ERR SPFX "No memory for ib_agent_port_private\n"); + ret = -ENOMEM; + goto error1; + } + + memset(port_priv, 0, sizeof *port_priv); + port_priv->port_num = port_num; + spin_lock_init(&port_priv->send_list_lock); + INIT_LIST_HEAD(&port_priv->send_posted_list); + + /* Obtain send only MAD agent for SM class (SMI QP) */ + port_priv->smp_agent = ib_register_mad_agent(device, port_num, + IB_QPT_SMI, + NULL, 0, + &agent_send_handler, + NULL, NULL); + + if (IS_ERR(port_priv->smp_agent)) { + ret = PTR_ERR(port_priv->smp_agent); + goto error2; + } + + /* Obtain send only MAD agent for PerfMgmt class (GSI QP) */ + port_priv->perf_mgmt_agent = ib_register_mad_agent(device, port_num, + IB_QPT_GSI, + NULL, 0, + &agent_send_handler, + NULL, NULL); + if (IS_ERR(port_priv->perf_mgmt_agent)) { + ret = PTR_ERR(port_priv->perf_mgmt_agent); + goto error3; + } + + port_priv->mr = ib_get_dma_mr(port_priv->smp_agent->qp->pd, + IB_ACCESS_LOCAL_WRITE); + if (IS_ERR(port_priv->mr)) { + printk(KERN_ERR SPFX "Couldn't get DMA MR\n"); + ret = PTR_ERR(port_priv->mr); + goto error4; + } + + spin_lock_irqsave(&ib_agent_port_list_lock, flags); + list_add_tail(&port_priv->port_list, &ib_agent_port_list); + spin_unlock_irqrestore(&ib_agent_port_list_lock, flags); + + return 0; + +error4: + ib_unregister_mad_agent(port_priv->perf_mgmt_agent); +error3: + ib_unregister_mad_agent(port_priv->smp_agent); +error2: + kfree(port_priv); +error1: + return ret; +} + +int ib_agent_port_close(struct ib_device *device, int port_num) +{ + struct ib_agent_port_private *port_priv; + unsigned long flags; + + spin_lock_irqsave(&ib_agent_port_list_lock, flags); + port_priv = __ib_get_agent_port(device, port_num, NULL); + if (port_priv == NULL) { + spin_unlock_irqrestore(&ib_agent_port_list_lock, flags); + printk(KERN_ERR SPFX "Port %d not found\n", port_num); + return -ENODEV; + } + list_del(&port_priv->port_list); + spin_unlock_irqrestore(&ib_agent_port_list_lock, flags); + + ib_dereg_mr(port_priv->mr); + + ib_unregister_mad_agent(port_priv->perf_mgmt_agent); + ib_unregister_mad_agent(port_priv->smp_agent); + kfree(port_priv); + + return 0; +} diff --git a/drivers/infiniband/core/agent.h b/drivers/infiniband/core/agent.h new file mode 100644 index 000000000000..d9426842254a --- /dev/null +++ b/drivers/infiniband/core/agent.h @@ -0,0 +1,55 @@ +/* + * Copyright (c) 2004 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2004 Infinicon Corporation. All rights reserved. + * Copyright (c) 2004 Intel Corporation. All rights reserved. + * Copyright (c) 2004 Topspin Corporation. All rights reserved. + * Copyright (c) 2004 Voltaire Corporation. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + * + * $Id: agent.h 1389 2004-12-27 22:56:47Z roland $ + */ + +#ifndef __AGENT_H_ +#define __AGENT_H_ + +extern spinlock_t ib_agent_port_list_lock; + +extern int ib_agent_port_open(struct ib_device *device, + int port_num); + +extern int ib_agent_port_close(struct ib_device *device, int port_num); + +extern int agent_send(struct ib_mad_private *mad, + struct ib_grh *grh, + struct ib_wc *wc, + struct ib_device *device, + int port_num); + +#endif /* __AGENT_H_ */ diff --git a/drivers/infiniband/core/agent_priv.h b/drivers/infiniband/core/agent_priv.h new file mode 100644 index 000000000000..17a0cce5813c --- /dev/null +++ b/drivers/infiniband/core/agent_priv.h @@ -0,0 +1,63 @@ +/* + * Copyright (c) 2004 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2004 Infinicon Corporation. All rights reserved. + * Copyright (c) 2004 Intel Corporation. All rights reserved. + * Copyright (c) 2004 Topspin Corporation. All rights reserved. + * Copyright (c) 2004 Voltaire Corporation. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + * + * $Id: agent_priv.h 1389 2004-12-27 22:56:47Z roland $ + */ + +#ifndef __IB_AGENT_PRIV_H__ +#define __IB_AGENT_PRIV_H__ + +#include <linux/pci.h> + +#define SPFX "ib_agent: " + +struct ib_agent_send_wr { + struct list_head send_list; + struct ib_ah *ah; + struct ib_mad_private *mad; + DECLARE_PCI_UNMAP_ADDR(mapping) +}; + +struct ib_agent_port_private { + struct list_head port_list; + struct list_head send_posted_list; + spinlock_t send_list_lock; + int port_num; + struct ib_mad_agent *smp_agent; /* SM class */ + struct ib_mad_agent *perf_mgmt_agent; /* PerfMgmt class */ + struct ib_mr *mr; +}; + +#endif /* __IB_AGENT_PRIV_H__ */ diff --git a/drivers/infiniband/core/cache.c b/drivers/infiniband/core/cache.c new file mode 100644 index 000000000000..3042360c97e1 --- /dev/null +++ b/drivers/infiniband/core/cache.c @@ -0,0 +1,365 @@ +/* + * Copyright (c) 2004 Topspin Communications. 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. + * + * $Id: cache.c 1349 2004-12-16 21:09:43Z roland $ + */ + +#include <linux/version.h> +#include <linux/module.h> +#include <linux/errno.h> +#include <linux/slab.h> + +#include <ib_cache.h> + +#include "core_priv.h" + +struct ib_pkey_cache { + int table_len; + u16 table[0]; +}; + +struct ib_gid_cache { + int table_len; + union ib_gid table[0]; +}; + +struct ib_update_work { + struct work_struct work; + struct ib_device *device; + u8 port_num; +}; + +static inline int start_port(struct ib_device *device) +{ + return device->node_type == IB_NODE_SWITCH ? 0 : 1; +} + +static inline int end_port(struct ib_device *device) +{ + return device->node_type == IB_NODE_SWITCH ? 0 : device->phys_port_cnt; +} + +int ib_get_cached_gid(struct ib_device *device, + u8 port_num, + int index, + union ib_gid *gid) +{ + struct ib_gid_cache *cache; + unsigned long flags; + int ret = 0; + + if (port_num < start_port(device) || port_num > end_port(device)) + return -EINVAL; + + read_lock_irqsave(&device->cache.lock, flags); + + cache = device->cache.gid_cache[port_num - start_port(device)]; + + if (index < 0 || index >= cache->table_len) + ret = -EINVAL; + else + *gid = cache->table[index]; + + read_unlock_irqrestore(&device->cache.lock, flags); + + return ret; +} +EXPORT_SYMBOL(ib_get_cached_gid); + +int ib_find_cached_gid(struct ib_device *device, + union ib_gid *gid, + u8 *port_num, + u16 *index) +{ + struct ib_gid_cache *cache; + unsigned long flags; + int p, i; + int ret = -ENOENT; + + *port_num = -1; + if (index) + *index = -1; + + read_lock_irqsave(&device->cache.lock, flags); + + for (p = 0; p <= end_port(device) - start_port(device); ++p) { + cache = device->cache.gid_cache[p]; + for (i = 0; i < cache->table_len; ++i) { + if (!memcmp(gid, &cache->table[i], sizeof *gid)) { + *port_num = p + start_port(device); + if (index) + *index = i; + ret = 0; + goto found; + } + } + } +found: + read_unlock_irqrestore(&device->cache.lock, flags); + + return ret; +} +EXPORT_SYMBOL(ib_find_cached_gid); + +int ib_get_cached_pkey(struct ib_device *device, + u8 port_num, + int index, + u16 *pkey) +{ + struct ib_pkey_cache *cache; + unsigned long flags; + int ret = 0; + + if (port_num < start_port(device) || port_num > end_port(device)) + return -EINVAL; + + read_lock_irqsave(&device->cache.lock, flags); + + cache = device->cache.pkey_cache[port_num - start_port(device)]; + + if (index < 0 || index >= cache->table_len) + ret = -EINVAL; + else + *pkey = cache->table[index]; + + read_unlock_irqrestore(&device->cache.lock, flags); + + return ret; +} +EXPORT_SYMBOL(ib_get_cached_pkey); + +int ib_find_cached_pkey(struct ib_device *device, + u8 port_num, + u16 pkey, + u16 *index) +{ + struct ib_pkey_cache *cache; + unsigned long flags; + int i; + int ret = -ENOENT; + + if (port_num < start_port(device) || port_num > end_port(device)) + return -EINVAL; + + read_lock_irqsave(&device->cache.lock, flags); + + cache = device->cache.pkey_cache[port_num - start_port(device)]; + + *index = -1; + + for (i = 0; i < cache->table_len; ++i) + if ((cache->table[i] & 0x7fff) == (pkey & 0x7fff)) { + *index = i; + ret = 0; + break; + } + + read_unlock_irqrestore(&device->cache.lock, flags); + + return ret; +} +EXPORT_SYMBOL(ib_find_cached_pkey); + +static void ib_cache_update(struct ib_device *device, + u8 port) +{ + struct ib_port_attr *tprops = NULL; + struct ib_pkey_cache *pkey_cache = NULL, *old_pkey_cache; + struct ib_gid_cache *gid_cache = NULL, *old_gid_cache; + int i; + int ret; + + tprops = kmalloc(sizeof *tprops, GFP_KERNEL); + if (!tprops) + return; + + ret = ib_query_port(device, port, tprops); + if (ret) { + printk(KERN_WARNING "ib_query_port failed (%d) for %s\n", + ret, device->name); + goto err; + } + + pkey_cache = kmalloc(sizeof *pkey_cache + tprops->pkey_tbl_len * + sizeof *pkey_cache->table, GFP_KERNEL); + if (!pkey_cache) + goto err; + + pkey_cache->table_len = tprops->pkey_tbl_len; + + gid_cache = kmalloc(sizeof *gid_cache + tprops->gid_tbl_len * + sizeof *gid_cache->table, GFP_KERNEL); + if (!gid_cache) + goto err; + + gid_cache->table_len = tprops->gid_tbl_len; + + for (i = 0; i < pkey_cache->table_len; ++i) { + ret = ib_query_pkey(device, port, i, pkey_cache->table + i); + if (ret) { + printk(KERN_WARNING "ib_query_pkey failed (%d) for %s (index %d)\n", + ret, device->name, i); + goto err; + } + } + + for (i = 0; i < gid_cache->table_len; ++i) { + ret = ib_query_gid(device, port, i, gid_cache->table + i); + if (ret) { + printk(KERN_WARNING "ib_query_gid failed (%d) for %s (index %d)\n", + ret, device->name, i); + goto err; + } + } + + write_lock_irq(&device->cache.lock); + + old_pkey_cache = device->cache.pkey_cache[port - start_port(device)]; + old_gid_cache = device->cache.gid_cache [port - start_port(device)]; + + device->cache.pkey_cache[port - start_port(device)] = pkey_cache; + device->cache.gid_cache [port - start_port(device)] = gid_cache; + + write_unlock_irq(&device->cache.lock); + + kfree(old_pkey_cache); + kfree(old_gid_cache); + kfree(tprops); + return; + +err: + kfree(pkey_cache); + kfree(gid_cache); + kfree(tprops); +} + +static void ib_cache_task(void *work_ptr) +{ + struct ib_update_work *work = work_ptr; + + ib_cache_update(work->device, work->port_num); + kfree(work); +} + +static void ib_cache_event(struct ib_event_handler *handler, + struct ib_event *event) +{ + struct ib_update_work *work; + + 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) { + work = kmalloc(sizeof *work, GFP_ATOMIC); + if (work) { + INIT_WORK(&work->work, ib_cache_task, work); + work->device = event->device; + work->port_num = event->element.port_num; + schedule_work(&work->work); + } + } +} + +static void ib_cache_setup_one(struct ib_device *device) +{ + int p; + + rwlock_init(&device->cache.lock); + + device->cache.pkey_cache = + kmalloc(sizeof *device->cache.pkey_cache * + (end_port(device) - start_port(device) + 1), GFP_KERNEL); + device->cache.gid_cache = + kmalloc(sizeof *device->cache.pkey_cache * + (end_port(device) - start_port(device) + 1), GFP_KERNEL); + + if (!device->cache.pkey_cache || !device->cache.gid_cache) { + printk(KERN_WARNING "Couldn't allocate cache " + "for %s\n", device->name); + goto err; + } + + for (p = 0; p <= end_port(device) - start_port(device); ++p) { + device->cache.pkey_cache[p] = NULL; + device->cache.gid_cache [p] = NULL; + ib_cache_update(device, p + start_port(device)); + } + + INIT_IB_EVENT_HANDLER(&device->cache.event_handler, + device, ib_cache_event); + if (ib_register_event_handler(&device->cache.event_handler)) + goto err_cache; + + return; + +err_cache: + for (p = 0; p <= end_port(device) - start_port(device); ++p) { + kfree(device->cache.pkey_cache[p]); + kfree(device->cache.gid_cache[p]); + } + +err: + kfree(device->cache.pkey_cache); + kfree(device->cache.gid_cache); +} + +static void ib_cache_cleanup_one(struct ib_device *device) +{ + int p; + + ib_unregister_event_handler(&device->cache.event_handler); + flush_scheduled_work(); + + for (p = 0; p <= end_port(device) - start_port(device); ++p) { + kfree(device->cache.pkey_cache[p]); + kfree(device->cache.gid_cache[p]); + } + + kfree(device->cache.pkey_cache); + kfree(device->cache.gid_cache); +} + +static struct ib_client cache_client = { + .name = "cache", + .add = ib_cache_setup_one, + .remove = ib_cache_cleanup_one +}; + +int __init ib_cache_setup(void) +{ + return ib_register_client(&cache_client); +} + +void __exit ib_cache_cleanup(void) +{ + ib_unregister_client(&cache_client); +} diff --git a/drivers/infiniband/core/core_priv.h b/drivers/infiniband/core/core_priv.h new file mode 100644 index 000000000000..797049626ff6 --- /dev/null +++ b/drivers/infiniband/core/core_priv.h @@ -0,0 +1,52 @@ +/* + * Copyright (c) 2004 Topspin Communications. 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. + * + * $Id: core_priv.h 1349 2004-12-16 21:09:43Z roland $ + */ + +#ifndef _CORE_PRIV_H +#define _CORE_PRIV_H + +#include <linux/list.h> +#include <linux/spinlock.h> + +#include <ib_verbs.h> + +int ib_device_register_sysfs(struct ib_device *device); +void ib_device_unregister_sysfs(struct ib_device *device); + +int ib_sysfs_setup(void); +void ib_sysfs_cleanup(void); + +int ib_cache_setup(void); +void ib_cache_cleanup(void); + +#endif /* _CORE_PRIV_H */ diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c new file mode 100644 index 000000000000..9197e92d708a --- /dev/null +++ b/drivers/infiniband/core/device.c @@ -0,0 +1,614 @@ +/* + * Copyright (c) 2004 Topspin Communications. 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. + * + * $Id: device.c 1349 2004-12-16 21:09:43Z roland $ + */ + +#include <linux/module.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/init.h> + +#include <asm/semaphore.h> + +#include "core_priv.h" + +MODULE_AUTHOR("Roland Dreier"); +MODULE_DESCRIPTION("core kernel InfiniBand API"); +MODULE_LICENSE("Dual BSD/GPL"); + +struct ib_client_data { + struct list_head list; + struct ib_client *client; + void * data; +}; + +static LIST_HEAD(device_list); +static LIST_HEAD(client_list); + +/* + * device_sem protects access to both device_list and client_list. + * There's no real point to using multiple locks or something fancier + * like an rwsem: we always access both lists, and we're always + * modifying one list or the other list. In any case this is not a + * hot path so there's no point in trying to optimize. + */ +static DECLARE_MUTEX(device_sem); + +static int ib_device_check_mandatory(struct ib_device *device) +{ +#define IB_MANDATORY_FUNC(x) { offsetof(struct ib_device, x), #x } + static const struct { + size_t offset; + char *name; + } mandatory_table[] = { + IB_MANDATORY_FUNC(query_device), + IB_MANDATORY_FUNC(query_port), + IB_MANDATORY_FUNC(query_pkey), + IB_MANDATORY_FUNC(query_gid), + IB_MANDATORY_FUNC(alloc_pd), + IB_MANDATORY_FUNC(dealloc_pd), + IB_MANDATORY_FUNC(create_ah), + IB_MANDATORY_FUNC(destroy_ah), + IB_MANDATORY_FUNC(create_qp), + IB_MANDATORY_FUNC(modify_qp), + IB_MANDATORY_FUNC(destroy_qp), + IB_MANDATORY_FUNC(post_send), + IB_MANDATORY_FUNC(post_recv), + IB_MANDATORY_FUNC(create_cq), + IB_MANDATORY_FUNC(destroy_cq), + IB_MANDATORY_FUNC(poll_cq), + IB_MANDATORY_FUNC(req_notify_cq), + IB_MANDATORY_FUNC(get_dma_mr), + IB_MANDATORY_FUNC(dereg_mr) + }; + int i; + + for (i = 0; i < sizeof mandatory_table / sizeof mandatory_table[0]; ++i) { + if (!*(void **) ((void *) device + mandatory_table[i].offset)) { + printk(KERN_WARNING "Device %s is missing mandatory function %s\n", + device->name, mandatory_table[i].name); + return -EINVAL; + } + } + + return 0; +} + +static struct ib_device *__ib_device_get_by_name(const char *name) +{ + struct ib_device *device; + + list_for_each_entry(device, &device_list, core_list) + if (!strncmp(name, device->name, IB_DEVICE_NAME_MAX)) + return device; + + return NULL; +} + + +static int alloc_name(char *name) +{ + long *inuse; + char buf[IB_DEVICE_NAME_MAX]; + struct ib_device *device; + int i; + + inuse = (long *) get_zeroed_page(GFP_KERNEL); + if (!inuse) + return -ENOMEM; + + list_for_each_entry(device, &device_list, core_list) { + if (!sscanf(device->name, name, &i)) + continue; + if (i < 0 || i >= PAGE_SIZE * 8) + continue; + snprintf(buf, sizeof buf, name, i); + if (!strncmp(buf, device->name, IB_DEVICE_NAME_MAX)) + set_bit(i, inuse); + } + + i = find_first_zero_bit(inuse, PAGE_SIZE * 8); + free_page((unsigned long) inuse); + snprintf(buf, sizeof buf, name, i); + + if (__ib_device_get_by_name(buf)) + return -ENFILE; + + strlcpy(name, buf, IB_DEVICE_NAME_MAX); + return 0; +} + +/** + * ib_alloc_device - allocate an IB device struct + * @size:size of structure to allocate + * + * Low-level drivers should use ib_alloc_device() to allocate &struct + * ib_device. @size is the size of the structure to be allocated, + * including any private data used by the low-level driver. + * ib_dealloc_device() must be used to free structures allocated with + * ib_alloc_device(). + */ +struct ib_device *ib_alloc_device(size_t size) +{ + void *dev; + + BUG_ON(size < sizeof (struct ib_device)); + + dev = kmalloc(size, GFP_KERNEL); + if (!dev) + return NULL; + + memset(dev, 0, size); + + return dev; +} +EXPORT_SYMBOL(ib_alloc_device); + +/** + * ib_dealloc_device - free an IB device struct + * @device:structure to free + * + * Free a structure allocated with ib_alloc_device(). + */ +void ib_dealloc_device(struct ib_device *device) +{ + if (device->reg_state == IB_DEV_UNINITIALIZED) { + kfree(device); + return; + } + + BUG_ON(device->reg_state != IB_DEV_UNREGISTERED); + + ib_device_unregister_sysfs(device); +} +EXPORT_SYMBOL(ib_dealloc_device); + +static int add_client_context(struct ib_device *device, struct ib_client *client) +{ + struct ib_client_data *context; + unsigned long flags; + + context = kmalloc(sizeof *context, GFP_KERNEL); + if (!context) { + printk(KERN_WARNING "Couldn't allocate client context for %s/%s\n", + device->name, client->name); + return -ENOMEM; + } + + context->client = client; + context->data = NULL; + + spin_lock_irqsave(&device->client_data_lock, flags); + list_add(&context->list, &device->client_data_list); + spin_unlock_irqrestore(&device->client_data_lock, flags); + + return 0; +} + +/** + * ib_register_device - Register an IB device with IB core + * @device:Device to register + * + * Low-level drivers use ib_register_device() to register their + * devices with the IB core. All registered clients will receive a + * callback for each device that is added. @device must be allocated + * with ib_alloc_device(). + */ +int ib_register_device(struct ib_device *device) +{ + int ret; + + down(&device_sem); + + if (strchr(device->name, '%')) { + ret = alloc_name(device->name); + if (ret) + goto out; + } + + if (ib_device_check_mandatory(device)) { + ret = -EINVAL; + goto out; + } + + INIT_LIST_HEAD(&device->event_handler_list); + INIT_LIST_HEAD(&device->client_data_list); + spin_lock_init(&device->event_handler_lock); + spin_lock_init(&device->client_data_lock); + + ret = ib_device_register_sysfs(device); + if (ret) { + printk(KERN_WARNING "Couldn't register device %s with driver model\n", + device->name); + goto out; + } + + list_add_tail(&device->core_list, &device_list); + + device->reg_state = IB_DEV_REGISTERED; + + { + struct ib_client *client; + + list_for_each_entry(client, &client_list, list) + if (client->add && !add_client_context(device, client)) + client->add(device); + } + + out: + up(&device_sem); + return ret; +} +EXPORT_SYMBOL(ib_register_device); + +/** + * ib_unregister_device - Unregister an IB device + * @device:Device to unregister + * + * Unregister an IB device. All clients will receive a remove callback. + */ +void ib_unregister_device(struct ib_device *device) +{ + struct ib_client *client; + struct ib_client_data *context, *tmp; + unsigned long flags; + + down(&device_sem); + + list_for_each_entry_reverse(client, &client_list, list) + if (client->remove) + client->remove(device); + + list_del(&device->core_list); + + up(&device_sem); + + spin_lock_irqsave(&device->client_data_lock, flags); + list_for_each_entry_safe(context, tmp, &device->client_data_list, list) + kfree(context); + spin_unlock_irqrestore(&device->client_data_lock, flags); + + device->reg_state = IB_DEV_UNREGISTERED; +} +EXPORT_SYMBOL(ib_unregister_device); + +/** + * ib_register_client - Register an IB client + * @client:Client to register + * + * Upper level users of the IB drivers can use ib_register_client() to + * register callbacks for IB device addition and removal. When an IB + * device is added, each registered client's add method will be called + * (in the order the clients were registered), and when a device is + * removed, each client's remove method will be called (in the reverse + * order that clients were registered). In addition, when + * ib_register_client() is called, the client will receive an add + * callback for all devices already registered. + */ +int ib_register_client(struct ib_client *client) +{ + struct ib_device *device; + + down(&device_sem); + + list_add_tail(&client->list, &client_list); + list_for_each_entry(device, &device_list, core_list) + if (client->add && !add_client_context(device, client)) + client->add(device); + + up(&device_sem); + + return 0; +} +EXPORT_SYMBOL(ib_register_client); + +/** + * ib_unregister_client - Unregister an IB client + * @client:Client to unregister + * + * Upper level users use ib_unregister_client() to remove their client + * registration. When ib_unregister_client() is called, the client + * will receive a remove callback for each IB device still registered. + */ +void ib_unregister_client(struct ib_client *client) +{ + struct ib_client_data *context, *tmp; + struct ib_device *device; + unsigned long flags; + + down(&device_sem); + + list_for_each_entry(device, &device_list, core_list) { + if (client->remove) + client->remove(device); + + spin_lock_irqsave(&device->client_data_lock, flags); + list_for_each_entry_safe(context, tmp, &device->client_data_list, list) + if (context->client == client) { + list_del(&context->list); + kfree(context); + } + spin_unlock_irqrestore(&device->client_data_lock, flags); + } + list_del(&client->list); + + up(&device_sem); +} +EXPORT_SYMBOL(ib_unregister_client); + +/** + * ib_get_client_data - Get IB client context + * @device:Device to get context for + * @client:Client to get context for + * + * ib_get_client_data() returns client context set with + * ib_set_client_data(). + */ +void *ib_get_client_data(struct ib_device *device, struct ib_client *client) +{ + struct ib_client_data *context; + void *ret = NULL; + unsigned long flags; + + spin_lock_irqsave(&device->client_data_lock, flags); + list_for_each_entry(context, &device->client_data_list, list) + if (context->client == client) { + ret = context->data; + break; + } + spin_unlock_irqrestore(&device->client_data_lock, flags); + + return ret; +} +EXPORT_SYMBOL(ib_get_client_data); + +/** + * ib_set_client_data - Get IB client context + * @device:Device to set context for + * @client:Client to set context for + * @data:Context to set + * + * ib_set_client_data() sets client context that can be retrieved with + * ib_get_client_data(). + */ +void ib_set_client_data(struct ib_device *device, struct ib_client *client, + void *data) +{ + struct ib_client_data *context; + unsigned long flags; + + spin_lock_irqsave(&device->client_data_lock, flags); + list_for_each_entry(context, &device->client_data_list, list) + if (context->client == client) { + context->data = data; + goto out; + } + + printk(KERN_WARNING "No client context found for %s/%s\n", + device->name, client->name); + +out: + spin_unlock_irqrestore(&device->client_data_lock, flags); +} +EXPORT_SYMBOL(ib_set_client_data); + +/** + * ib_register_event_handler - Register an IB event handler + * @event_handler:Handler to register + * + * ib_register_event_handler() registers an event handler that will be + * called back when asynchronous IB events occur (as defined in + * chapter 11 of the InfiniBand Architecture Specification). This + * callback may occur in interrupt context. + */ +int ib_register_event_handler (struct ib_event_handler *event_handler) +{ + unsigned long flags; + + spin_lock_irqsave(&event_handler->device->event_handler_lock, flags); + list_add_tail(&event_handler->list, + &event_handler->device->event_handler_list); + spin_unlock_irqrestore(&event_handler->device->event_handler_lock, flags); + + return 0; +} +EXPORT_SYMBOL(ib_register_event_handler); + +/** + * ib_unregister_event_handler - Unregister an event handler + * @event_handler:Handler to unregister + * + * Unregister an event handler registered with + * ib_register_event_handler(). + */ +int ib_unregister_event_handler(struct ib_event_handler *event_handler) +{ + unsigned long flags; + + spin_lock_irqsave(&event_handler->device->event_handler_lock, flags); + list_del(&event_handler->list); + spin_unlock_irqrestore(&event_handler->device->event_handler_lock, flags); + + return 0; +} +EXPORT_SYMBOL(ib_unregister_event_handler); + +/** + * ib_dispatch_event - Dispatch an asynchronous event + * @event:Event to dispatch + * + * Low-level drivers must call ib_dispatch_event() to dispatch the + * event to all registered event handlers when an asynchronous event + * occurs. + */ +void ib_dispatch_event(struct ib_event *event) +{ + unsigned long flags; + struct ib_event_handler *handler; + + spin_lock_irqsave(&event->device->event_handler_lock, flags); + + list_for_each_entry(handler, &event->device->event_handler_list, list) + handler->handler(handler, event); + + spin_unlock_irqrestore(&event->device->event_handler_lock, flags); +} +EXPORT_SYMBOL(ib_dispatch_event); + +/** + * ib_query_device - Query IB device attributes + * @device:Device to query + * @device_attr:Device attributes + * + * ib_query_device() returns the attributes of a device through the + * @device_attr pointer. + */ +int ib_query_device(struct ib_device *device, + struct ib_device_attr *device_attr) +{ + return device->query_device(device, device_attr); +} +EXPORT_SYMBOL(ib_query_device); + +/** + * ib_query_port - Query IB port attributes + * @device:Device to query + * @port_num:Port number to query + * @port_attr:Port attributes + * + * ib_query_port() returns the attributes of a port through the + * @port_attr pointer. + */ +int ib_query_port(struct ib_device *device, + u8 port_num, + struct ib_port_attr *port_attr) +{ + return device->query_port(device, port_num, port_attr); +} +EXPORT_SYMBOL(ib_query_port); + +/** + * ib_query_gid - Get GID table entry + * @device:Device to query + * @port_num:Port number to query + * @index:GID table index to query + * @gid:Returned GID + * + * ib_query_gid() fetches the specified GID table entry. + */ +int ib_query_gid(struct ib_device *device, + u8 port_num, int index, union ib_gid *gid) +{ + return device->query_gid(device, port_num, index, gid); +} +EXPORT_SYMBOL(ib_query_gid); + +/** + * ib_query_pkey - Get P_Key table entry + * @device:Device to query + * @port_num:Port number to query + * @index:P_Key table index to query + * @pkey:Returned P_Key + * + * ib_query_pkey() fetches the specified P_Key table entry. + */ +int ib_query_pkey(struct ib_device *device, + u8 port_num, u16 index, u16 *pkey) +{ + return device->query_pkey(device, port_num, index, pkey); +} +EXPORT_SYMBOL(ib_query_pkey); + +/** + * ib_modify_device - Change IB device attributes + * @device:Device to modify + * @device_modify_mask:Mask of attributes to change + * @device_modify:New attribute values + * + * ib_modify_device() changes a device's attributes as specified by + * the @device_modify_mask and @device_modify structure. + */ +int ib_modify_device(struct ib_device *device, + int device_modify_mask, + struct ib_device_modify *device_modify) +{ + return device->modify_device(device, device_modify_mask, + device_modify); +} +EXPORT_SYMBOL(ib_modify_device); + +/** + * ib_modify_port - Modifies the attributes for the specified port. + * @device: The device to modify. + * @port_num: The number of the port to modify. + * @port_modify_mask: Mask used to specify which attributes of the port + * to change. + * @port_modify: New attribute values for the port. + * + * ib_modify_port() changes a port's attributes as specified by the + * @port_modify_mask and @port_modify structure. + */ +int ib_modify_port(struct ib_device *device, + u8 port_num, int port_modify_mask, + struct ib_port_modify *port_modify) +{ + return device->modify_port(device, port_num, port_modify_mask, + port_modify); +} +EXPORT_SYMBOL(ib_modify_port); + +static int __init ib_core_init(void) +{ + int ret; + + ret = ib_sysfs_setup(); + if (ret) + printk(KERN_WARNING "Couldn't create InfiniBand device class\n"); + + ret = ib_cache_setup(); + if (ret) { + printk(KERN_WARNING "Couldn't set up InfiniBand P_Key/GID cache\n"); + ib_sysfs_cleanup(); + } + + return ret; +} + +static void __exit ib_core_cleanup(void) +{ + ib_cache_cleanup(); + ib_sysfs_cleanup(); +} + +module_init(ib_core_init); +module_exit(ib_core_cleanup); diff --git a/drivers/infiniband/core/fmr_pool.c b/drivers/infiniband/core/fmr_pool.c new file mode 100644 index 000000000000..2e9469f18926 --- /dev/null +++ b/drivers/infiniband/core/fmr_pool.c @@ -0,0 +1,507 @@ +/* + * Copyright (c) 2004 Topspin Communications. 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. + * + * $Id: fmr_pool.c 1349 2004-12-16 21:09:43Z roland $ + */ + +#include <linux/errno.h> +#include <linux/spinlock.h> +#include <linux/slab.h> +#include <linux/jhash.h> +#include <linux/kthread.h> + +#include <ib_fmr_pool.h> + +#include "core_priv.h" + +enum { + IB_FMR_MAX_REMAPS = 32, + + IB_FMR_HASH_BITS = 8, + IB_FMR_HASH_SIZE = 1 << IB_FMR_HASH_BITS, + IB_FMR_HASH_MASK = IB_FMR_HASH_SIZE - 1 +}; + +/* + * If an FMR is not in use, then the list member will point to either + * its pool's free_list (if the FMR can be mapped again; that is, + * remap_count < IB_FMR_MAX_REMAPS) or its pool's dirty_list (if the + * FMR needs to be unmapped before being remapped). In either of + * these cases it is a bug if the ref_count is not 0. In other words, + * if ref_count is > 0, then the list member must not be linked into + * either free_list or dirty_list. + * + * The cache_node member is used to link the FMR into a cache bucket + * (if caching is enabled). This is independent of the reference + * count of the FMR. When a valid FMR is released, its ref_count is + * decremented, and if ref_count reaches 0, the FMR is placed in + * either free_list or dirty_list as appropriate. However, it is not + * removed from the cache and may be "revived" if a call to + * ib_fmr_register_physical() occurs before the FMR is remapped. In + * this case we just increment the ref_count and remove the FMR from + * free_list/dirty_list. + * + * Before we remap an FMR from free_list, we remove it from the cache + * (to prevent another user from obtaining a stale FMR). When an FMR + * is released, we add it to the tail of the free list, so that our + * cache eviction policy is "least recently used." + * + * All manipulation of ref_count, list and cache_node is protected by + * pool_lock to maintain consistency. + */ + +struct ib_fmr_pool { + spinlock_t pool_lock; + + int pool_size; + int max_pages; + int dirty_watermark; + int dirty_len; + struct list_head free_list; + struct list_head dirty_list; + struct hlist_head *cache_bucket; + + void (*flush_function)(struct ib_fmr_pool *pool, + void * arg); + void *flush_arg; + + struct task_struct *thread; + + atomic_t req_ser; + atomic_t flush_ser; + + wait_queue_head_t force_wait; +}; + +static inline u32 ib_fmr_hash(u64 first_page) +{ + return jhash_2words((u32) first_page, + (u32) (first_page >> 32), + 0); +} + +/* Caller must hold pool_lock */ +static inline struct ib_pool_fmr *ib_fmr_cache_lookup(struct ib_fmr_pool *pool, + u64 *page_list, + int page_list_len, + u64 io_virtual_address) +{ + struct hlist_head *bucket; + struct ib_pool_fmr *fmr; + struct hlist_node *pos; + + if (!pool->cache_bucket) + return NULL; + + bucket = pool->cache_bucket + ib_fmr_hash(*page_list); + + hlist_for_each_entry(fmr, pos, bucket, cache_node) + if (io_virtual_address == fmr->io_virtual_address && + page_list_len == fmr->page_list_len && + !memcmp(page_list, fmr->page_list, + page_list_len * sizeof *page_list)) + return fmr; + + return NULL; +} + +static void ib_fmr_batch_release(struct ib_fmr_pool *pool) +{ + int ret; + struct ib_pool_fmr *fmr; + LIST_HEAD(unmap_list); + LIST_HEAD(fmr_list); + + spin_lock_irq(&pool->pool_lock); + + list_for_each_entry(fmr, &pool->dirty_list, list) { + hlist_del_init(&fmr->cache_node); + fmr->remap_count = 0; + list_add_tail(&fmr->fmr->list, &fmr_list); + +#ifdef DEBUG + if (fmr->ref_count !=0) { + printk(KERN_WARNING "Unmapping FMR 0x%08x with ref count %d", + fmr, fmr->ref_count); + } +#endif + } + + list_splice(&pool->dirty_list, &unmap_list); + INIT_LIST_HEAD(&pool->dirty_list); + pool->dirty_len = 0; + + spin_unlock_irq(&pool->pool_lock); + + if (list_empty(&unmap_list)) { + return; + } + + ret = ib_unmap_fmr(&fmr_list); + if (ret) + printk(KERN_WARNING "ib_unmap_fmr returned %d", ret); + + spin_lock_irq(&pool->pool_lock); + list_splice(&unmap_list, &pool->free_list); + spin_unlock_irq(&pool->pool_lock); +} + +static int ib_fmr_cleanup_thread(void *pool_ptr) +{ + struct ib_fmr_pool *pool = pool_ptr; + + do { + if (pool->dirty_len >= pool->dirty_watermark || + 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); + } + + set_current_state(TASK_INTERRUPTIBLE); + if (pool->dirty_len < pool->dirty_watermark && + atomic_read(&pool->flush_ser) - atomic_read(&pool->req_ser) >= 0 && + !kthread_should_stop()) + schedule(); + __set_current_state(TASK_RUNNING); + } while (!kthread_should_stop()); + + return 0; +} + +/** + * ib_create_fmr_pool - Create an FMR pool + * @pd:Protection domain for FMRs + * @params:FMR pool parameters + * + * Create a pool of FMRs. Return value is pointer to new pool or + * error code if creation failed. + */ +struct ib_fmr_pool *ib_create_fmr_pool(struct ib_pd *pd, + struct ib_fmr_pool_param *params) +{ + struct ib_device *device; + struct ib_fmr_pool *pool; + int i; + int ret; + + if (!params) + return ERR_PTR(-EINVAL); + + device = pd->device; + if (!device->alloc_fmr || !device->dealloc_fmr || + !device->map_phys_fmr || !device->unmap_fmr) { + printk(KERN_WARNING "Device %s does not support fast memory regions", + device->name); + return ERR_PTR(-ENOSYS); + } + + pool = kmalloc(sizeof *pool, GFP_KERNEL); + if (!pool) { + printk(KERN_WARNING "couldn't allocate pool struct"); + return ERR_PTR(-ENOMEM); + } + + pool->cache_bucket = NULL; + + pool->flush_function = params->flush_function; + pool->flush_arg = params->flush_arg; + + INIT_LIST_HEAD(&pool->free_list); + INIT_LIST_HEAD(&pool->dirty_list); + + if (params->cache) { + pool->cache_bucket = + kmalloc(IB_FMR_HASH_SIZE * sizeof *pool->cache_bucket, + GFP_KERNEL); + if (!pool->cache_bucket) { + printk(KERN_WARNING "Failed to allocate cache in pool"); + ret = -ENOMEM; + goto out_free_pool; + } + + for (i = 0; i < IB_FMR_HASH_SIZE; ++i) + INIT_HLIST_HEAD(pool->cache_bucket + i); + } + + pool->pool_size = 0; + pool->max_pages = params->max_pages_per_fmr; + pool->dirty_watermark = params->dirty_watermark; + pool->dirty_len = 0; + spin_lock_init(&pool->pool_lock); + atomic_set(&pool->req_ser, 0); + atomic_set(&pool->flush_ser, 0); + init_waitqueue_head(&pool->force_wait); + + pool->thread = kthread_create(ib_fmr_cleanup_thread, + pool, + "ib_fmr(%s)", + device->name); + if (IS_ERR(pool->thread)) { + printk(KERN_WARNING "couldn't start cleanup thread"); + ret = PTR_ERR(pool->thread); + goto out_free_pool; + } + + { + struct ib_pool_fmr *fmr; + struct ib_fmr_attr attr = { + .max_pages = params->max_pages_per_fmr, + .max_maps = IB_FMR_MAX_REMAPS, + .page_size = PAGE_SHIFT + }; + + for (i = 0; i < params->pool_size; ++i) { + fmr = kmalloc(sizeof *fmr + params->max_pages_per_fmr * sizeof (u64), + GFP_KERNEL); + if (!fmr) { + printk(KERN_WARNING "failed to allocate fmr struct " + "for FMR %d", i); + goto out_fail; + } + + fmr->pool = pool; + fmr->remap_count = 0; + fmr->ref_count = 0; + INIT_HLIST_NODE(&fmr->cache_node); + + fmr->fmr = ib_alloc_fmr(pd, params->access, &attr); + if (IS_ERR(fmr->fmr)) { + printk(KERN_WARNING "fmr_create failed for FMR %d", i); + kfree(fmr); + goto out_fail; + } + + list_add_tail(&fmr->list, &pool->free_list); + ++pool->pool_size; + } + } + + return pool; + + out_free_pool: + kfree(pool->cache_bucket); + kfree(pool); + + return ERR_PTR(ret); + + out_fail: + ib_destroy_fmr_pool(pool); + + return ERR_PTR(-ENOMEM); +} +EXPORT_SYMBOL(ib_create_fmr_pool); + +/** + * ib_destroy_fmr_pool - Free FMR pool + * @pool:FMR pool to free + * + * Destroy an FMR pool and free all associated resources. + */ +int ib_destroy_fmr_pool(struct ib_fmr_pool *pool) +{ + struct ib_pool_fmr *fmr; + struct ib_pool_fmr *tmp; + int i; + + kthread_stop(pool->thread); + ib_fmr_batch_release(pool); + + i = 0; + list_for_each_entry_safe(fmr, tmp, &pool->free_list, list) { + ib_dealloc_fmr(fmr->fmr); + list_del(&fmr->list); + kfree(fmr); + ++i; + } + + if (i < pool->pool_size) + printk(KERN_WARNING "pool still has %d regions registered", + pool->pool_size - i); + + kfree(pool->cache_bucket); + kfree(pool); + + return 0; +} +EXPORT_SYMBOL(ib_destroy_fmr_pool); + +/** + * ib_flush_fmr_pool - Invalidate all unmapped FMRs + * @pool:FMR pool to flush + * + * Ensure that all unmapped FMRs are fully invalidated. + */ +int ib_flush_fmr_pool(struct ib_fmr_pool *pool) +{ + int serial; + + atomic_inc(&pool->req_ser); + /* + * It's OK if someone else bumps req_ser again here -- we'll + * just wait a little longer. + */ + serial = atomic_read(&pool->req_ser); + + wake_up_process(pool->thread); + + if (wait_event_interruptible(pool->force_wait, + atomic_read(&pool->flush_ser) - + atomic_read(&pool->req_ser) >= 0)) + return -EINTR; + + return 0; +} +EXPORT_SYMBOL(ib_flush_fmr_pool); + +/** + * ib_fmr_pool_map_phys - + * @pool:FMR pool to allocate FMR from + * @page_list:List of pages to map + * @list_len:Number of pages in @page_list + * @io_virtual_address:I/O virtual address for new FMR + * + * Map an FMR from an FMR pool. + */ +struct ib_pool_fmr *ib_fmr_pool_map_phys(struct ib_fmr_pool *pool_handle, + u64 *page_list, + int list_len, + u64 *io_virtual_address) +{ + struct ib_fmr_pool *pool = pool_handle; + struct ib_pool_fmr *fmr; + unsigned long flags; + int result; + + if (list_len < 1 || list_len > pool->max_pages) + return ERR_PTR(-EINVAL); + + spin_lock_irqsave(&pool->pool_lock, flags); + fmr = ib_fmr_cache_lookup(pool, + page_list, + list_len, + *io_virtual_address); + if (fmr) { + /* found in cache */ + ++fmr->ref_count; + if (fmr->ref_count == 1) { + list_del(&fmr->list); + } + + spin_unlock_irqrestore(&pool->pool_lock, flags); + + return fmr; + } + + if (list_empty(&pool->free_list)) { + spin_unlock_irqrestore(&pool->pool_lock, flags); + return ERR_PTR(-EAGAIN); + } + + fmr = list_entry(pool->free_list.next, struct ib_pool_fmr, list); + list_del(&fmr->list); + hlist_del_init(&fmr->cache_node); + spin_unlock_irqrestore(&pool->pool_lock, flags); + + result = ib_map_phys_fmr(fmr->fmr, page_list, list_len, + *io_virtual_address); + + if (result) { + spin_lock_irqsave(&pool->pool_lock, flags); + list_add(&fmr->list, &pool->free_list); + spin_unlock_irqrestore(&pool->pool_lock, flags); + + printk(KERN_WARNING "fmr_map returns %d", + result); + + return ERR_PTR(result); + } + + ++fmr->remap_count; + fmr->ref_count = 1; + + if (pool->cache_bucket) { + fmr->io_virtual_address = *io_virtual_address; + fmr->page_list_len = list_len; + memcpy(fmr->page_list, page_list, list_len * sizeof(*page_list)); + + spin_lock_irqsave(&pool->pool_lock, flags); + hlist_add_head(&fmr->cache_node, + pool->cache_bucket + ib_fmr_hash(fmr->page_list[0])); + spin_unlock_irqrestore(&pool->pool_lock, flags); + } + + return fmr; +} +EXPORT_SYMBOL(ib_fmr_pool_map_phys); + +/** + * ib_fmr_pool_unmap - Unmap FMR + * @fmr:FMR to unmap + * + * Unmap an FMR. The FMR mapping may remain valid until the FMR is + * reused (or until ib_flush_fmr_pool() is called). + */ +int ib_fmr_pool_unmap(struct ib_pool_fmr *fmr) +{ + struct ib_fmr_pool *pool; + unsigned long flags; + + pool = fmr->pool; + + spin_lock_irqsave(&pool->pool_lock, flags); + + --fmr->ref_count; + if (!fmr->ref_count) { + if (fmr->remap_count < IB_FMR_MAX_REMAPS) { + list_add_tail(&fmr->list, &pool->free_list); + } else { + list_add_tail(&fmr->list, &pool->dirty_list); + ++pool->dirty_len; + wake_up_process(pool->thread); + } + } + +#ifdef DEBUG + if (fmr->ref_count < 0) + printk(KERN_WARNING "FMR %p has ref count %d < 0", + fmr, fmr->ref_count); +#endif + + spin_unlock_irqrestore(&pool->pool_lock, flags); + + return 0; +} +EXPORT_SYMBOL(ib_fmr_pool_unmap); diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c new file mode 100644 index 000000000000..4ec7fff29b5d --- /dev/null +++ b/drivers/infiniband/core/mad.c @@ -0,0 +1,2714 @@ +/* + * Copyright (c) 2004, 2005 Voltaire, 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. + * + * $Id: mad.c 1389 2004-12-27 22:56:47Z roland $ + */ + +#include <linux/dma-mapping.h> +#include <linux/interrupt.h> + +#include <ib_mad.h> + +#include "mad_priv.h" +#include "smi.h" +#include "agent.h" + +MODULE_LICENSE("Dual BSD/GPL"); +MODULE_DESCRIPTION("kernel IB MAD API"); +MODULE_AUTHOR("Hal Rosenstock"); +MODULE_AUTHOR("Sean Hefty"); + + +kmem_cache_t *ib_mad_cache; +static struct list_head ib_mad_port_list; +static u32 ib_mad_client_id = 0; + +/* Port list lock */ +static spinlock_t ib_mad_port_list_lock; + + +/* Forward declarations */ +static int method_in_use(struct ib_mad_mgmt_method_table **method, + struct ib_mad_reg_req *mad_reg_req); +static void remove_mad_reg_req(struct ib_mad_agent_private *priv); +static struct ib_mad_agent_private *find_mad_agent( + struct ib_mad_port_private *port_priv, + struct ib_mad *mad, int solicited); +static int ib_mad_post_receive_mads(struct ib_mad_qp_info *qp_info, + struct ib_mad_private *mad); +static void cancel_mads(struct ib_mad_agent_private *mad_agent_priv); +static void ib_mad_complete_send_wr(struct ib_mad_send_wr_private *mad_send_wr, + struct ib_mad_send_wc *mad_send_wc); +static void timeout_sends(void *data); +static void cancel_sends(void *data); +static void local_completions(void *data); +static int solicited_mad(struct ib_mad *mad); +static int add_nonoui_reg_req(struct ib_mad_reg_req *mad_reg_req, + struct ib_mad_agent_private *agent_priv, + u8 mgmt_class); +static int add_oui_reg_req(struct ib_mad_reg_req *mad_reg_req, + struct ib_mad_agent_private *agent_priv); + +/* + * Returns a ib_mad_port_private structure or NULL for a device/port + * Assumes ib_mad_port_list_lock is being held + */ +static inline struct ib_mad_port_private * +__ib_get_mad_port(struct ib_device *device, int port_num) +{ + struct ib_mad_port_private *entry; + + list_for_each_entry(entry, &ib_mad_port_list, port_list) { + if (entry->device == device && entry->port_num == port_num) + return entry; + } + return NULL; +} + +/* + * Wrapper function to return a ib_mad_port_private structure or NULL + * for a device/port + */ +static inline struct ib_mad_port_private * +ib_get_mad_port(struct ib_device *device, int port_num) +{ + struct ib_mad_port_private *entry; + unsigned long flags; + + spin_lock_irqsave(&ib_mad_port_list_lock, flags); + entry = __ib_get_mad_port(device, port_num); + spin_unlock_irqrestore(&ib_mad_port_list_lock, flags); + + return entry; +} + +static inline u8 convert_mgmt_class(u8 mgmt_class) +{ + /* Alias IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE to 0 */ + return mgmt_class == IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE ? + 0 : mgmt_class; +} + +static int get_spl_qp_index(enum ib_qp_type qp_type) +{ + switch (qp_type) + { + case IB_QPT_SMI: + return 0; + case IB_QPT_GSI: + return 1; + default: + return -1; + } +} + +static int vendor_class_index(u8 mgmt_class) +{ + return mgmt_class - IB_MGMT_CLASS_VENDOR_RANGE2_START; +} + +static int is_vendor_class(u8 mgmt_class) +{ + if ((mgmt_class < IB_MGMT_CLASS_VENDOR_RANGE2_START) || + (mgmt_class > IB_MGMT_CLASS_VENDOR_RANGE2_END)) + return 0; + return 1; +} + +static int is_vendor_oui(char *oui) +{ + if (oui[0] || oui[1] || oui[2]) + return 1; + return 0; +} + +static int is_vendor_method_in_use( + struct ib_mad_mgmt_vendor_class *vendor_class, + struct ib_mad_reg_req *mad_reg_req) +{ + struct ib_mad_mgmt_method_table *method; + int i; + + for (i = 0; i < MAX_MGMT_OUI; i++) { + if (!memcmp(vendor_class->oui[i], mad_reg_req->oui, 3)) { + method = vendor_class->method_table[i]; + if (method) { + if (method_in_use(&method, mad_reg_req)) + return 1; + else + break; + } + } + } + return 0; +} + +/* + * ib_register_mad_agent - Register to send/receive MADs + */ +struct ib_mad_agent *ib_register_mad_agent(struct ib_device *device, + u8 port_num, + enum ib_qp_type qp_type, + struct ib_mad_reg_req *mad_reg_req, + u8 rmpp_version, + ib_mad_send_handler send_handler, + ib_mad_recv_handler recv_handler, + void *context) +{ + struct ib_mad_port_private *port_priv; + struct ib_mad_agent *ret = ERR_PTR(-EINVAL); + struct ib_mad_agent_private *mad_agent_priv; + struct ib_mad_reg_req *reg_req = NULL; + struct ib_mad_mgmt_class_table *class; + struct ib_mad_mgmt_vendor_class_table *vendor; + struct ib_mad_mgmt_vendor_class *vendor_class; + struct ib_mad_mgmt_method_table *method; + int ret2, qpn; + unsigned long flags; + u8 mgmt_class, vclass; + + /* Validate parameters */ + qpn = get_spl_qp_index(qp_type); + if (qpn == -1) + goto error1; + + if (rmpp_version) + goto error1; /* XXX: until RMPP implemented */ + + /* Validate MAD registration request if supplied */ + if (mad_reg_req) { + if (mad_reg_req->mgmt_class_version >= MAX_MGMT_VERSION) + goto error1; + if (!recv_handler) + goto error1; + if (mad_reg_req->mgmt_class >= MAX_MGMT_CLASS) { + /* + * IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE is the only + * one in this range currently allowed + */ + if (mad_reg_req->mgmt_class != + IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE) + goto error1; + } else if (mad_reg_req->mgmt_class == 0) { + /* + * Class 0 is reserved in IBA and is used for + * aliasing of IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE + */ + goto error1; + } else if (is_vendor_class(mad_reg_req->mgmt_class)) { + /* + * If class is in "new" vendor range, + * ensure supplied OUI is not zero + */ + if (!is_vendor_oui(mad_reg_req->oui)) + goto error1; + } + /* Make sure class supplied is consistent with QP type */ + if (qp_type == IB_QPT_SMI) { + if ((mad_reg_req->mgmt_class != + IB_MGMT_CLASS_SUBN_LID_ROUTED) && + (mad_reg_req->mgmt_class != + IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE)) + goto error1; + } else { + if ((mad_reg_req->mgmt_class == + IB_MGMT_CLASS_SUBN_LID_ROUTED) || + (mad_reg_req->mgmt_class == + IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE)) + goto error1; + } + } else { + /* No registration request supplied */ + if (!send_handler) + goto error1; + } + + /* Validate device and port */ + port_priv = ib_get_mad_port(device, port_num); + if (!port_priv) { + ret = ERR_PTR(-ENODEV); + goto error1; + } + + /* Allocate structures */ + mad_agent_priv = kmalloc(sizeof *mad_agent_priv, GFP_KERNEL); + if (!mad_agent_priv) { + ret = ERR_PTR(-ENOMEM); + goto error1; + } + + if (mad_reg_req) { + reg_req = kmalloc(sizeof *reg_req, GFP_KERNEL); + if (!reg_req) { + ret = ERR_PTR(-ENOMEM); + goto error2; + } + /* Make a copy of the MAD registration request */ + memcpy(reg_req, mad_reg_req, sizeof *reg_req); + } + + /* Now, fill in the various structures */ + memset(mad_agent_priv, 0, sizeof *mad_agent_priv); + mad_agent_priv->qp_info = &port_priv->qp_info[qpn]; + mad_agent_priv->reg_req = reg_req; + mad_agent_priv->rmpp_version = rmpp_version; + mad_agent_priv->agent.device = device; + mad_agent_priv->agent.recv_handler = recv_handler; + mad_agent_priv->agent.send_handler = send_handler; + mad_agent_priv->agent.context = context; + mad_agent_priv->agent.qp = port_priv->qp_info[qpn].qp; + mad_agent_priv->agent.port_num = port_num; + + spin_lock_irqsave(&port_priv->reg_lock, flags); + mad_agent_priv->agent.hi_tid = ++ib_mad_client_id; + + /* + * Make sure MAD registration (if supplied) + * is non overlapping with any existing ones + */ + if (mad_reg_req) { + mgmt_class = convert_mgmt_class(mad_reg_req->mgmt_class); + if (!is_vendor_class(mgmt_class)) { + class = port_priv->version[mad_reg_req-> + mgmt_class_version].class; + if (class) { + method = class->method_table[mgmt_class]; + if (method) { + if (method_in_use(&method, + mad_reg_req)) + goto error3; + } + } + ret2 = add_nonoui_reg_req(mad_reg_req, mad_agent_priv, + mgmt_class); + } else { + /* "New" vendor class range */ + vendor = port_priv->version[mad_reg_req-> + mgmt_class_version].vendor; + if (vendor) { + vclass = vendor_class_index(mgmt_class); + vendor_class = vendor->vendor_class[vclass]; + if (vendor_class) { + if (is_vendor_method_in_use( + vendor_class, + mad_reg_req)) + goto error3; + } + } + ret2 = add_oui_reg_req(mad_reg_req, mad_agent_priv); + } + if (ret2) { + ret = ERR_PTR(ret2); + goto error3; + } + } + + /* Add mad agent into port's agent list */ + list_add_tail(&mad_agent_priv->agent_list, &port_priv->agent_list); + spin_unlock_irqrestore(&port_priv->reg_lock, flags); + + spin_lock_init(&mad_agent_priv->lock); + INIT_LIST_HEAD(&mad_agent_priv->send_list); + INIT_LIST_HEAD(&mad_agent_priv->wait_list); + INIT_WORK(&mad_agent_priv->timed_work, timeout_sends, mad_agent_priv); + INIT_LIST_HEAD(&mad_agent_priv->local_list); + INIT_WORK(&mad_agent_priv->local_work, local_completions, + mad_agent_priv); + INIT_LIST_HEAD(&mad_agent_priv->canceled_list); + INIT_WORK(&mad_agent_priv->canceled_work, cancel_sends, mad_agent_priv); + atomic_set(&mad_agent_priv->refcount, 1); + init_waitqueue_head(&mad_agent_priv->wait); + + return &mad_agent_priv->agent; + +error3: + spin_unlock_irqrestore(&port_priv->reg_lock, flags); + kfree(reg_req); +error2: + kfree(mad_agent_priv); +error1: + return ret; +} +EXPORT_SYMBOL(ib_register_mad_agent); + +static inline int is_snooping_sends(int mad_snoop_flags) +{ + return (mad_snoop_flags & + (/*IB_MAD_SNOOP_POSTED_SENDS | + IB_MAD_SNOOP_RMPP_SENDS |*/ + IB_MAD_SNOOP_SEND_COMPLETIONS /*| + IB_MAD_SNOOP_RMPP_SEND_COMPLETIONS*/)); +} + +static inline int is_snooping_recvs(int mad_snoop_flags) +{ + return (mad_snoop_flags & + (IB_MAD_SNOOP_RECVS /*| + IB_MAD_SNOOP_RMPP_RECVS*/)); +} + +static int register_snoop_agent(struct ib_mad_qp_info *qp_info, + struct ib_mad_snoop_private *mad_snoop_priv) +{ + struct ib_mad_snoop_private **new_snoop_table; + unsigned long flags; + int i; + + spin_lock_irqsave(&qp_info->snoop_lock, flags); + /* Check for empty slot in array. */ + for (i = 0; i < qp_info->snoop_table_size; i++) + if (!qp_info->snoop_table[i]) + break; + + if (i == qp_info->snoop_table_size) { + /* Grow table. */ + new_snoop_table = kmalloc(sizeof mad_snoop_priv * + qp_info->snoop_table_size + 1, + GFP_ATOMIC); + if (!new_snoop_table) { + i = -ENOMEM; + goto out; + } + if (qp_info->snoop_table) { + memcpy(new_snoop_table, qp_info->snoop_table, + sizeof mad_snoop_priv * + qp_info->snoop_table_size); + kfree(qp_info->snoop_table); + } + qp_info->snoop_table = new_snoop_table; + qp_info->snoop_table_size++; + } + qp_info->snoop_table[i] = mad_snoop_priv; + atomic_inc(&qp_info->snoop_count); +out: + spin_unlock_irqrestore(&qp_info->snoop_lock, flags); + return i; +} + +struct ib_mad_agent *ib_register_mad_snoop(struct ib_device *device, + u8 port_num, + enum ib_qp_type qp_type, + int mad_snoop_flags, + ib_mad_snoop_handler snoop_handler, + ib_mad_recv_handler recv_handler, + void *context) +{ + struct ib_mad_port_private *port_priv; + struct ib_mad_agent *ret; + struct ib_mad_snoop_private *mad_snoop_priv; + int qpn; + + /* Validate parameters */ + if ((is_snooping_sends(mad_snoop_flags) && !snoop_handler) || + (is_snooping_recvs(mad_snoop_flags) && !recv_handler)) { + ret = ERR_PTR(-EINVAL); + goto error1; + } + qpn = get_spl_qp_index(qp_type); + if (qpn == -1) { + ret = ERR_PTR(-EINVAL); + goto error1; + } + port_priv = ib_get_mad_port(device, port_num); + if (!port_priv) { + ret = ERR_PTR(-ENODEV); + goto error1; + } + /* Allocate structures */ + mad_snoop_priv = kmalloc(sizeof *mad_snoop_priv, GFP_KERNEL); + if (!mad_snoop_priv) { + ret = ERR_PTR(-ENOMEM); + goto error1; + } + + /* Now, fill in the various structures */ + memset(mad_snoop_priv, 0, sizeof *mad_snoop_priv); + mad_snoop_priv->qp_info = &port_priv->qp_info[qpn]; + mad_snoop_priv->agent.device = device; + mad_snoop_priv->agent.recv_handler = recv_handler; + mad_snoop_priv->agent.snoop_handler = snoop_handler; + mad_snoop_priv->agent.context = context; + mad_snoop_priv->agent.qp = port_priv->qp_info[qpn].qp; + mad_snoop_priv->agent.port_num = port_num; + mad_snoop_priv->mad_snoop_flags = mad_snoop_flags; + init_waitqueue_head(&mad_snoop_priv->wait); + mad_snoop_priv->snoop_index = register_snoop_agent( + &port_priv->qp_info[qpn], + mad_snoop_priv); + if (mad_snoop_priv->snoop_index < 0) { + ret = ERR_PTR(mad_snoop_priv->snoop_index); + goto error2; + } + + atomic_set(&mad_snoop_priv->refcount, 1); + return &mad_snoop_priv->agent; + +error2: + kfree(mad_snoop_priv); +error1: + return ret; +} +EXPORT_SYMBOL(ib_register_mad_snoop); + +static void unregister_mad_agent(struct ib_mad_agent_private *mad_agent_priv) +{ + struct ib_mad_port_private *port_priv; + unsigned long flags; + + /* Note that we could still be handling received MADs */ + + /* + * Canceling all sends results in dropping received response + * MADs, preventing us from queuing additional work + */ + cancel_mads(mad_agent_priv); + + port_priv = mad_agent_priv->qp_info->port_priv; + + cancel_delayed_work(&mad_agent_priv->timed_work); + flush_workqueue(port_priv->wq); + + spin_lock_irqsave(&port_priv->reg_lock, flags); + remove_mad_reg_req(mad_agent_priv); + list_del(&mad_agent_priv->agent_list); + spin_unlock_irqrestore(&port_priv->reg_lock, flags); + + /* XXX: Cleanup pending RMPP receives for this agent */ + + atomic_dec(&mad_agent_priv->refcount); + wait_event(mad_agent_priv->wait, + !atomic_read(&mad_agent_priv->refcount)); + + if (mad_agent_priv->reg_req) + kfree(mad_agent_priv->reg_req); + kfree(mad_agent_priv); +} + +static void unregister_mad_snoop(struct ib_mad_snoop_private *mad_snoop_priv) +{ + struct ib_mad_qp_info *qp_info; + unsigned long flags; + + qp_info = mad_snoop_priv->qp_info; + spin_lock_irqsave(&qp_info->snoop_lock, flags); + qp_info->snoop_table[mad_snoop_priv->snoop_index] = NULL; + atomic_dec(&qp_info->snoop_count); + spin_unlock_irqrestore(&qp_info->snoop_lock, flags); + + atomic_dec(&mad_snoop_priv->refcount); + wait_event(mad_snoop_priv->wait, + !atomic_read(&mad_snoop_priv->refcount)); + + kfree(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) +{ + struct ib_mad_agent_private *mad_agent_priv; + struct ib_mad_snoop_private *mad_snoop_priv; + + /* If the TID is zero, the agent can only snoop. */ + if (mad_agent->hi_tid) { + mad_agent_priv = container_of(mad_agent, + struct ib_mad_agent_private, + agent); + unregister_mad_agent(mad_agent_priv); + } else { + mad_snoop_priv = container_of(mad_agent, + struct ib_mad_snoop_private, + agent); + unregister_mad_snoop(mad_snoop_priv); + } + return 0; +} +EXPORT_SYMBOL(ib_unregister_mad_agent); + +static void dequeue_mad(struct ib_mad_list_head *mad_list) +{ + struct ib_mad_queue *mad_queue; + unsigned long flags; + + BUG_ON(!mad_list->mad_queue); + mad_queue = mad_list->mad_queue; + spin_lock_irqsave(&mad_queue->lock, flags); + list_del(&mad_list->list); + mad_queue->count--; + spin_unlock_irqrestore(&mad_queue->lock, flags); +} + +static void snoop_send(struct ib_mad_qp_info *qp_info, + struct ib_send_wr *send_wr, + struct ib_mad_send_wc *mad_send_wc, + int mad_snoop_flags) +{ + struct ib_mad_snoop_private *mad_snoop_priv; + unsigned long flags; + int i; + + spin_lock_irqsave(&qp_info->snoop_lock, flags); + for (i = 0; i < qp_info->snoop_table_size; i++) { + mad_snoop_priv = qp_info->snoop_table[i]; + if (!mad_snoop_priv || + !(mad_snoop_priv->mad_snoop_flags & mad_snoop_flags)) + continue; + + atomic_inc(&mad_snoop_priv->refcount); + spin_unlock_irqrestore(&qp_info->snoop_lock, flags); + mad_snoop_priv->agent.snoop_handler(&mad_snoop_priv->agent, + send_wr, mad_send_wc); + if (atomic_dec_and_test(&mad_snoop_priv->refcount)) + wake_up(&mad_snoop_priv->wait); + spin_lock_irqsave(&qp_info->snoop_lock, flags); + } + spin_unlock_irqrestore(&qp_info->snoop_lock, flags); +} + +static void snoop_recv(struct ib_mad_qp_info *qp_info, + struct ib_mad_recv_wc *mad_recv_wc, + int mad_snoop_flags) +{ + struct ib_mad_snoop_private *mad_snoop_priv; + unsigned long flags; + int i; + + spin_lock_irqsave(&qp_info->snoop_lock, flags); + for (i = 0; i < qp_info->snoop_table_size; i++) { + mad_snoop_priv = qp_info->snoop_table[i]; + if (!mad_snoop_priv || + !(mad_snoop_priv->mad_snoop_flags & mad_snoop_flags)) + continue; + + atomic_inc(&mad_snoop_priv->refcount); + spin_unlock_irqrestore(&qp_info->snoop_lock, flags); + mad_snoop_priv->agent.recv_handler(&mad_snoop_priv->agent, + mad_recv_wc); + if (atomic_dec_and_test(&mad_snoop_priv->refcount)) + wake_up(&mad_snoop_priv->wait); + spin_lock_irqsave(&qp_info->snoop_lock, flags); + } + spin_unlock_irqrestore(&qp_info->snoop_lock, flags); +} + +static void build_smp_wc(u64 wr_id, u16 slid, u16 pkey_index, u8 port_num, + struct ib_wc *wc) +{ + memset(wc, 0, sizeof *wc); + wc->wr_id = wr_id; + wc->status = IB_WC_SUCCESS; + wc->opcode = IB_WC_RECV; + wc->pkey_index = pkey_index; + wc->byte_len = sizeof(struct ib_mad) + sizeof(struct ib_grh); + wc->src_qp = IB_QP0; + wc->qp_num = IB_QP0; + wc->slid = slid; + wc->sl = 0; + wc->dlid_path_bits = 0; + wc->port_num = port_num; +} + +/* + * Return 0 if SMP is to be sent + * Return 1 if SMP was consumed locally (whether or not solicited) + * Return < 0 if error + */ +static int handle_outgoing_dr_smp(struct ib_mad_agent_private *mad_agent_priv, + struct ib_smp *smp, + struct ib_send_wr *send_wr) +{ + int ret, solicited; + unsigned long flags; + struct ib_mad_local_private *local; + struct ib_mad_private *mad_priv; + struct ib_mad_port_private *port_priv; + struct ib_mad_agent_private *recv_mad_agent = NULL; + struct ib_device *device = mad_agent_priv->agent.device; + u8 port_num = mad_agent_priv->agent.port_num; + struct ib_wc mad_wc; + + if (!smi_handle_dr_smp_send(smp, device->node_type, port_num)) { + ret = -EINVAL; + printk(KERN_ERR PFX "Invalid directed route\n"); + goto out; + } + /* Check to post send on QP or process locally */ + ret = smi_check_local_dr_smp(smp, device, port_num); + if (!ret || !device->process_mad) + goto out; + + local = kmalloc(sizeof *local, GFP_ATOMIC); + if (!local) { + ret = -ENOMEM; + printk(KERN_ERR PFX "No memory for ib_mad_local_private\n"); + goto out; + } + local->mad_priv = NULL; + local->recv_mad_agent = NULL; + mad_priv = kmem_cache_alloc(ib_mad_cache, GFP_ATOMIC); + if (!mad_priv) { + ret = -ENOMEM; + printk(KERN_ERR PFX "No memory for local response MAD\n"); + kfree(local); + goto out; + } + + build_smp_wc(send_wr->wr_id, smp->dr_slid, send_wr->wr.ud.pkey_index, + send_wr->wr.ud.port_num, &mad_wc); + + /* No GRH for DR SMP */ + ret = device->process_mad(device, 0, port_num, &mad_wc, NULL, + (struct ib_mad *)smp, + (struct ib_mad *)&mad_priv->mad); + switch (ret) + { + case IB_MAD_RESULT_SUCCESS | IB_MAD_RESULT_REPLY: + /* + * See if response is solicited and + * there is a recv handler + */ + if (solicited_mad(&mad_priv->mad.mad) && + mad_agent_priv->agent.recv_handler) { + local->mad_priv = mad_priv; + local->recv_mad_agent = mad_agent_priv; + /* + * Reference MAD agent until receive + * side of local completion handled + */ + atomic_inc(&mad_agent_priv->refcount); + } else + kmem_cache_free(ib_mad_cache, mad_priv); + break; + case IB_MAD_RESULT_SUCCESS | IB_MAD_RESULT_CONSUMED: + kmem_cache_free(ib_mad_cache, mad_priv); + break; + case IB_MAD_RESULT_SUCCESS: + /* Treat like an incoming receive MAD */ + solicited = solicited_mad(&mad_priv->mad.mad); + port_priv = ib_get_mad_port(mad_agent_priv->agent.device, + mad_agent_priv->agent.port_num); + if (port_priv) { + mad_priv->mad.mad.mad_hdr.tid = + ((struct ib_mad *)smp)->mad_hdr.tid; + recv_mad_agent = find_mad_agent(port_priv, + &mad_priv->mad.mad, + solicited); + } + if (!port_priv || !recv_mad_agent) { + kmem_cache_free(ib_mad_cache, mad_priv); + kfree(local); + ret = 0; + goto out; + } + local->mad_priv = mad_priv; + local->recv_mad_agent = recv_mad_agent; + break; + default: + kmem_cache_free(ib_mad_cache, mad_priv); + kfree(local); + ret = -EINVAL; + goto out; + } + + local->send_wr = *send_wr; + local->send_wr.sg_list = local->sg_list; + memcpy(local->sg_list, send_wr->sg_list, + sizeof *send_wr->sg_list * send_wr->num_sge); + local->send_wr.next = NULL; + local->tid = send_wr->wr.ud.mad_hdr->tid; + local->wr_id = send_wr->wr_id; + /* Reference MAD agent until send side of local completion handled */ + atomic_inc(&mad_agent_priv->refcount); + /* Queue local completion to local list */ + spin_lock_irqsave(&mad_agent_priv->lock, flags); + list_add_tail(&local->completion_list, &mad_agent_priv->local_list); + spin_unlock_irqrestore(&mad_agent_priv->lock, flags); + queue_work(mad_agent_priv->qp_info->port_priv->wq, + &mad_agent_priv->local_work); + ret = 1; +out: + return ret; +} + +static int ib_send_mad(struct ib_mad_agent_private *mad_agent_priv, + struct ib_mad_send_wr_private *mad_send_wr) +{ + struct ib_mad_qp_info *qp_info; + struct ib_send_wr *bad_send_wr; + unsigned long flags; + int ret; + + /* Replace user's WR ID with our own to find WR upon completion */ + qp_info = mad_agent_priv->qp_info; + mad_send_wr->wr_id = mad_send_wr->send_wr.wr_id; + mad_send_wr->send_wr.wr_id = (unsigned long)&mad_send_wr->mad_list; + mad_send_wr->mad_list.mad_queue = &qp_info->send_queue; + + spin_lock_irqsave(&qp_info->send_queue.lock, flags); + if (qp_info->send_queue.count++ < qp_info->send_queue.max_active) { + list_add_tail(&mad_send_wr->mad_list.list, + &qp_info->send_queue.list); + spin_unlock_irqrestore(&qp_info->send_queue.lock, flags); + ret = ib_post_send(mad_agent_priv->agent.qp, + &mad_send_wr->send_wr, &bad_send_wr); + if (ret) { + printk(KERN_ERR PFX "ib_post_send failed: %d\n", ret); + dequeue_mad(&mad_send_wr->mad_list); + } + } else { + list_add_tail(&mad_send_wr->mad_list.list, + &qp_info->overflow_list); + spin_unlock_irqrestore(&qp_info->send_queue.lock, flags); + ret = 0; + } + return ret; +} + +/* + * ib_post_send_mad - Posts MAD(s) to the send queue of the QP associated + * with the registered client + */ +int ib_post_send_mad(struct ib_mad_agent *mad_agent, + struct ib_send_wr *send_wr, + struct ib_send_wr **bad_send_wr) +{ + int ret = -EINVAL; + struct ib_mad_agent_private *mad_agent_priv; + + /* Validate supplied parameters */ + if (!bad_send_wr) + goto error1; + + if (!mad_agent || !send_wr) + goto error2; + + if (!mad_agent->send_handler) + goto error2; + + mad_agent_priv = container_of(mad_agent, + struct ib_mad_agent_private, + agent); + + /* Walk list of send WRs and post each on send list */ + while (send_wr) { + unsigned long flags; + struct ib_send_wr *next_send_wr; + struct ib_mad_send_wr_private *mad_send_wr; + struct ib_smp *smp; + + /* Validate more parameters */ + if (send_wr->num_sge > IB_MAD_SEND_REQ_MAX_SG) + goto error2; + + if (send_wr->wr.ud.timeout_ms && !mad_agent->recv_handler) + goto error2; + + if (!send_wr->wr.ud.mad_hdr) { + printk(KERN_ERR PFX "MAD header must be supplied " + "in WR %p\n", send_wr); + goto error2; + } + + /* + * Save pointer to next work request to post in case the + * current one completes, and the user modifies the work + * request associated with the completion + */ + next_send_wr = (struct ib_send_wr *)send_wr->next; + + smp = (struct ib_smp *)send_wr->wr.ud.mad_hdr; + if (smp->mgmt_class == IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE) { + ret = handle_outgoing_dr_smp(mad_agent_priv, smp, + send_wr); + if (ret < 0) /* error */ + goto error2; + else if (ret == 1) /* locally consumed */ + goto next; + } + + /* Allocate MAD send WR tracking structure */ + mad_send_wr = kmalloc(sizeof *mad_send_wr, GFP_ATOMIC); + if (!mad_send_wr) { + printk(KERN_ERR PFX "No memory for " + "ib_mad_send_wr_private\n"); + ret = -ENOMEM; + goto error2; + } + + mad_send_wr->send_wr = *send_wr; + mad_send_wr->send_wr.sg_list = mad_send_wr->sg_list; + memcpy(mad_send_wr->sg_list, send_wr->sg_list, + sizeof *send_wr->sg_list * send_wr->num_sge); + mad_send_wr->send_wr.next = NULL; + mad_send_wr->tid = send_wr->wr.ud.mad_hdr->tid; + mad_send_wr->agent = mad_agent; + /* Timeout will be updated after send completes */ + mad_send_wr->timeout = msecs_to_jiffies(send_wr->wr. + ud.timeout_ms); + mad_send_wr->retry = 0; + /* One reference for each work request to QP + response */ + mad_send_wr->refcount = 1 + (mad_send_wr->timeout > 0); + mad_send_wr->status = IB_WC_SUCCESS; + + /* Reference MAD agent until send completes */ + atomic_inc(&mad_agent_priv->refcount); + spin_lock_irqsave(&mad_agent_priv->lock, flags); + list_add_tail(&mad_send_wr->agent_list, + &mad_agent_priv->send_list); + spin_unlock_irqrestore(&mad_agent_priv->lock, flags); + + ret = ib_send_mad(mad_agent_priv, mad_send_wr); + if (ret) { + /* Fail send request */ + spin_lock_irqsave(&mad_agent_priv->lock, flags); + list_del(&mad_send_wr->agent_list); + spin_unlock_irqrestore(&mad_agent_priv->lock, flags); + atomic_dec(&mad_agent_priv->refcount); + goto error2; + } +next: + send_wr = next_send_wr; + } + return 0; + +error2: + *bad_send_wr = send_wr; +error1: + return ret; +} +EXPORT_SYMBOL(ib_post_send_mad); + +/* + * ib_free_recv_mad - Returns data buffers used to receive + * a MAD to the access layer + */ +void ib_free_recv_mad(struct ib_mad_recv_wc *mad_recv_wc) +{ + struct ib_mad_recv_buf *entry; + struct ib_mad_private_header *mad_priv_hdr; + struct ib_mad_private *priv; + + mad_priv_hdr = container_of(mad_recv_wc, + struct ib_mad_private_header, + recv_wc); + priv = container_of(mad_priv_hdr, struct ib_mad_private, header); + + /* + * Walk receive buffer list associated with this WC + * No need to remove them from list of receive buffers + */ + list_for_each_entry(entry, &mad_recv_wc->recv_buf.list, list) { + /* Free previous receive buffer */ + kmem_cache_free(ib_mad_cache, priv); + mad_priv_hdr = container_of(mad_recv_wc, + struct ib_mad_private_header, + recv_wc); + priv = container_of(mad_priv_hdr, struct ib_mad_private, + header); + } + + /* Free last buffer */ + kmem_cache_free(ib_mad_cache, priv); +} +EXPORT_SYMBOL(ib_free_recv_mad); + +void ib_coalesce_recv_mad(struct ib_mad_recv_wc *mad_recv_wc, + void *buf) +{ + printk(KERN_ERR PFX "ib_coalesce_recv_mad() not implemented yet\n"); +} +EXPORT_SYMBOL(ib_coalesce_recv_mad); + +struct ib_mad_agent *ib_redirect_mad_qp(struct ib_qp *qp, + u8 rmpp_version, + ib_mad_send_handler send_handler, + ib_mad_recv_handler recv_handler, + void *context) +{ + return ERR_PTR(-EINVAL); /* XXX: for now */ +} +EXPORT_SYMBOL(ib_redirect_mad_qp); + +int ib_process_mad_wc(struct ib_mad_agent *mad_agent, + struct ib_wc *wc) +{ + printk(KERN_ERR PFX "ib_process_mad_wc() not implemented yet\n"); + return 0; +} +EXPORT_SYMBOL(ib_process_mad_wc); + +static int method_in_use(struct ib_mad_mgmt_method_table **method, + struct ib_mad_reg_req *mad_reg_req) +{ + int i; + + for (i = find_first_bit(mad_reg_req->method_mask, IB_MGMT_MAX_METHODS); + i < IB_MGMT_MAX_METHODS; + i = find_next_bit(mad_reg_req->method_mask, IB_MGMT_MAX_METHODS, + 1+i)) { + if ((*method)->agent[i]) { + printk(KERN_ERR PFX "Method %d already in use\n", i); + return -EINVAL; + } + } + return 0; +} + +static int allocate_method_table(struct ib_mad_mgmt_method_table **method) +{ + /* Allocate management method table */ + *method = kmalloc(sizeof **method, GFP_ATOMIC); + if (!*method) { + printk(KERN_ERR PFX "No memory for " + "ib_mad_mgmt_method_table\n"); + return -ENOMEM; + } + /* Clear management method table */ + memset(*method, 0, sizeof **method); + + return 0; +} + +/* + * Check to see if there are any methods still in use + */ +static int check_method_table(struct ib_mad_mgmt_method_table *method) +{ + int i; + + for (i = 0; i < IB_MGMT_MAX_METHODS; i++) + if (method->agent[i]) + return 1; + return 0; +} + +/* + * Check to see if there are any method tables for this class still in use + */ +static int check_class_table(struct ib_mad_mgmt_class_table *class) +{ + int i; + + for (i = 0; i < MAX_MGMT_CLASS; i++) + if (class->method_table[i]) + return 1; + return 0; +} + +static int check_vendor_class(struct ib_mad_mgmt_vendor_class *vendor_class) +{ + int i; + + for (i = 0; i < MAX_MGMT_OUI; i++) + if (vendor_class->method_table[i]) + return 1; + return 0; +} + +static int find_vendor_oui(struct ib_mad_mgmt_vendor_class *vendor_class, + char *oui) +{ + int i; + + for (i = 0; i < MAX_MGMT_OUI; i++) + /* Is there matching OUI for this vendor class ? */ + if (!memcmp(vendor_class->oui[i], oui, 3)) + return i; + + return -1; +} + +static int check_vendor_table(struct ib_mad_mgmt_vendor_class_table *vendor) +{ + int i; + + for (i = 0; i < MAX_MGMT_VENDOR_RANGE2; i++) + if (vendor->vendor_class[i]) + return 1; + + return 0; +} + +static void remove_methods_mad_agent(struct ib_mad_mgmt_method_table *method, + struct ib_mad_agent_private *agent) +{ + int i; + + /* Remove any methods for this mad agent */ + for (i = 0; i < IB_MGMT_MAX_METHODS; i++) { + if (method->agent[i] == agent) { + method->agent[i] = NULL; + } + } +} + +static int add_nonoui_reg_req(struct ib_mad_reg_req *mad_reg_req, + struct ib_mad_agent_private *agent_priv, + u8 mgmt_class) +{ + struct ib_mad_port_private *port_priv; + struct ib_mad_mgmt_class_table **class; + struct ib_mad_mgmt_method_table **method; + int i, ret; + + port_priv = agent_priv->qp_info->port_priv; + class = &port_priv->version[mad_reg_req->mgmt_class_version].class; + if (!*class) { + /* Allocate management class table for "new" class version */ + *class = kmalloc(sizeof **class, GFP_ATOMIC); + if (!*class) { + printk(KERN_ERR PFX "No memory for " + "ib_mad_mgmt_class_table\n"); + ret = -ENOMEM; + goto error1; + } + /* Clear management class table */ + memset(*class, 0, sizeof(**class)); + /* Allocate method table for this management class */ + method = &(*class)->method_table[mgmt_class]; + if ((ret = allocate_method_table(method))) + goto error2; + } else { + method = &(*class)->method_table[mgmt_class]; + if (!*method) { + /* Allocate method table for this management class */ + if ((ret = allocate_method_table(method))) + goto error1; + } + } + + /* Now, make sure methods are not already in use */ + if (method_in_use(method, mad_reg_req)) + goto error3; + + /* Finally, add in methods being registered */ + for (i = find_first_bit(mad_reg_req->method_mask, + IB_MGMT_MAX_METHODS); + i < IB_MGMT_MAX_METHODS; + i = find_next_bit(mad_reg_req->method_mask, IB_MGMT_MAX_METHODS, + 1+i)) { + (*method)->agent[i] = agent_priv; + } + return 0; + +error3: + /* Remove any methods for this mad agent */ + remove_methods_mad_agent(*method, agent_priv); + /* Now, check to see if there are any methods in use */ + if (!check_method_table(*method)) { + /* If not, release management method table */ + kfree(*method); + *method = NULL; + } + ret = -EINVAL; + goto error1; +error2: + kfree(*class); + *class = NULL; +error1: + return ret; +} + +static int add_oui_reg_req(struct ib_mad_reg_req *mad_reg_req, + struct ib_mad_agent_private *agent_priv) +{ + struct ib_mad_port_private *port_priv; + struct ib_mad_mgmt_vendor_class_table **vendor_table; + struct ib_mad_mgmt_vendor_class_table *vendor = NULL; + struct ib_mad_mgmt_vendor_class *vendor_class = NULL; + struct ib_mad_mgmt_method_table **method; + int i, ret = -ENOMEM; + u8 vclass; + + /* "New" vendor (with OUI) class */ + vclass = vendor_class_index(mad_reg_req->mgmt_class); + port_priv = agent_priv->qp_info->port_priv; + vendor_table = &port_priv->version[ + mad_reg_req->mgmt_class_version].vendor; + if (!*vendor_table) { + /* Allocate mgmt vendor class table for "new" class version */ + vendor = kmalloc(sizeof *vendor, GFP_ATOMIC); + if (!vendor) { + printk(KERN_ERR PFX "No memory for " + "ib_mad_mgmt_vendor_class_table\n"); + goto error1; + } + /* Clear management vendor class table */ + memset(vendor, 0, sizeof(*vendor)); + *vendor_table = vendor; + } + if (!(*vendor_table)->vendor_class[vclass]) { + /* Allocate table for this management vendor class */ + vendor_class = kmalloc(sizeof *vendor_class, GFP_ATOMIC); + if (!vendor_class) { + printk(KERN_ERR PFX "No memory for " + "ib_mad_mgmt_vendor_class\n"); + goto error2; + } + memset(vendor_class, 0, sizeof(*vendor_class)); + (*vendor_table)->vendor_class[vclass] = vendor_class; + } + for (i = 0; i < MAX_MGMT_OUI; i++) { + /* Is there matching OUI for this vendor class ? */ + if (!memcmp((*vendor_table)->vendor_class[vclass]->oui[i], + mad_reg_req->oui, 3)) { + method = &(*vendor_table)->vendor_class[ + vclass]->method_table[i]; + BUG_ON(!*method); + goto check_in_use; + } + } + for (i = 0; i < MAX_MGMT_OUI; i++) { + /* OUI slot available ? */ + if (!is_vendor_oui((*vendor_table)->vendor_class[ + vclass]->oui[i])) { + method = &(*vendor_table)->vendor_class[ + vclass]->method_table[i]; + BUG_ON(*method); + /* Allocate method table for this OUI */ + if ((ret = allocate_method_table(method))) + goto error3; + memcpy((*vendor_table)->vendor_class[vclass]->oui[i], + mad_reg_req->oui, 3); + goto check_in_use; + } + } + printk(KERN_ERR PFX "All OUI slots in use\n"); + goto error3; + +check_in_use: + /* Now, make sure methods are not already in use */ + if (method_in_use(method, mad_reg_req)) + goto error4; + + /* Finally, add in methods being registered */ + for (i = find_first_bit(mad_reg_req->method_mask, + IB_MGMT_MAX_METHODS); + i < IB_MGMT_MAX_METHODS; + i = find_next_bit(mad_reg_req->method_mask, IB_MGMT_MAX_METHODS, + 1+i)) { + (*method)->agent[i] = agent_priv; + } + return 0; + +error4: + /* Remove any methods for this mad agent */ + remove_methods_mad_agent(*method, agent_priv); + /* Now, check to see if there are any methods in use */ + if (!check_method_table(*method)) { + /* If not, release management method table */ + kfree(*method); + *method = NULL; + } + ret = -EINVAL; +error3: + if (vendor_class) { + (*vendor_table)->vendor_class[vclass] = NULL; + kfree(vendor_class); + } +error2: + if (vendor) { + *vendor_table = NULL; + kfree(vendor); + } +error1: + return ret; +} + +static void remove_mad_reg_req(struct ib_mad_agent_private *agent_priv) +{ + struct ib_mad_port_private *port_priv; + struct ib_mad_mgmt_class_table *class; + struct ib_mad_mgmt_method_table *method; + struct ib_mad_mgmt_vendor_class_table *vendor; + struct ib_mad_mgmt_vendor_class *vendor_class; + int index; + u8 mgmt_class; + + /* + * Was MAD registration request supplied + * with original registration ? + */ + if (!agent_priv->reg_req) { + goto out; + } + + port_priv = agent_priv->qp_info->port_priv; + mgmt_class = convert_mgmt_class(agent_priv->reg_req->mgmt_class); + class = port_priv->version[ + agent_priv->reg_req->mgmt_class_version].class; + if (!class) + goto vendor_check; + + method = class->method_table[mgmt_class]; + if (method) { + /* Remove any methods for this mad agent */ + remove_methods_mad_agent(method, agent_priv); + /* Now, check to see if there are any methods still in use */ + if (!check_method_table(method)) { + /* If not, release management method table */ + kfree(method); + class->method_table[mgmt_class] = NULL; + /* Any management classes left ? */ + if (!check_class_table(class)) { + /* If not, release management class table */ + kfree(class); + port_priv->version[ + agent_priv->reg_req-> + mgmt_class_version].class = NULL; + } + } + } + +vendor_check: + if (!is_vendor_class(mgmt_class)) + goto out; + + /* normalize mgmt_class to vendor range 2 */ + mgmt_class = vendor_class_index(agent_priv->reg_req->mgmt_class); + vendor = port_priv->version[ + agent_priv->reg_req->mgmt_class_version].vendor; + + if (!vendor) + goto out; + + vendor_class = vendor->vendor_class[mgmt_class]; + if (vendor_class) { + index = find_vendor_oui(vendor_class, agent_priv->reg_req->oui); + if (index < 0) + goto out; + method = vendor_class->method_table[index]; + if (method) { + /* Remove any methods for this mad agent */ + remove_methods_mad_agent(method, agent_priv); + /* + * Now, check to see if there are + * any methods still in use + */ + if (!check_method_table(method)) { + /* If not, release management method table */ + kfree(method); + vendor_class->method_table[index] = NULL; + memset(vendor_class->oui[index], 0, 3); + /* Any OUIs left ? */ + if (!check_vendor_class(vendor_class)) { + /* If not, release vendor class table */ + kfree(vendor_class); + vendor->vendor_class[mgmt_class] = NULL; + /* Any other vendor classes left ? */ + if (!check_vendor_table(vendor)) { + kfree(vendor); + port_priv->version[ + agent_priv->reg_req-> + mgmt_class_version]. + vendor = NULL; + } + } + } + } + } + +out: + return; +} + +static int response_mad(struct ib_mad *mad) +{ + /* Trap represses are responses although response bit is reset */ + return ((mad->mad_hdr.method == IB_MGMT_METHOD_TRAP_REPRESS) || + (mad->mad_hdr.method & IB_MGMT_METHOD_RESP)); +} + +static int solicited_mad(struct ib_mad *mad) +{ + /* CM MADs are never solicited */ + if (mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_CM) { + return 0; + } + + /* XXX: Determine whether MAD is using RMPP */ + + /* Not using RMPP */ + /* Is this MAD a response to a previous MAD ? */ + return response_mad(mad); +} + +static struct ib_mad_agent_private * +find_mad_agent(struct ib_mad_port_private *port_priv, + struct ib_mad *mad, + int solicited) +{ + struct ib_mad_agent_private *mad_agent = NULL; + unsigned long flags; + + spin_lock_irqsave(&port_priv->reg_lock, flags); + + /* + * Whether MAD was solicited determines type of routing to + * MAD client. + */ + if (solicited) { + u32 hi_tid; + struct ib_mad_agent_private *entry; + + /* + * Routing is based on high 32 bits of transaction ID + * of MAD. + */ + hi_tid = be64_to_cpu(mad->mad_hdr.tid) >> 32; + list_for_each_entry(entry, &port_priv->agent_list, + agent_list) { + if (entry->agent.hi_tid == hi_tid) { + mad_agent = entry; + break; + } + } + } else { + struct ib_mad_mgmt_class_table *class; + struct ib_mad_mgmt_method_table *method; + struct ib_mad_mgmt_vendor_class_table *vendor; + struct ib_mad_mgmt_vendor_class *vendor_class; + struct ib_vendor_mad *vendor_mad; + int index; + + /* + * Routing is based on version, class, and method + * For "newer" vendor MADs, also based on OUI + */ + if (mad->mad_hdr.class_version >= MAX_MGMT_VERSION) + goto out; + if (!is_vendor_class(mad->mad_hdr.mgmt_class)) { + class = port_priv->version[ + mad->mad_hdr.class_version].class; + if (!class) + goto out; + method = class->method_table[convert_mgmt_class( + mad->mad_hdr.mgmt_class)]; + if (method) + mad_agent = method->agent[mad->mad_hdr.method & + ~IB_MGMT_METHOD_RESP]; + } else { + vendor = port_priv->version[ + mad->mad_hdr.class_version].vendor; + if (!vendor) + goto out; + vendor_class = vendor->vendor_class[vendor_class_index( + mad->mad_hdr.mgmt_class)]; + if (!vendor_class) + goto out; + /* Find matching OUI */ + vendor_mad = (struct ib_vendor_mad *)mad; + index = find_vendor_oui(vendor_class, vendor_mad->oui); + if (index == -1) + goto out; + method = vendor_class->method_table[index]; + if (method) { + mad_agent = method->agent[mad->mad_hdr.method & + ~IB_MGMT_METHOD_RESP]; + } + } + } + + if (mad_agent) { + if (mad_agent->agent.recv_handler) + atomic_inc(&mad_agent->refcount); + else { + printk(KERN_NOTICE PFX "No receive handler for client " + "%p on port %d\n", + &mad_agent->agent, port_priv->port_num); + mad_agent = NULL; + } + } +out: + spin_unlock_irqrestore(&port_priv->reg_lock, flags); + + return mad_agent; +} + +static int validate_mad(struct ib_mad *mad, u32 qp_num) +{ + int valid = 0; + + /* Make sure MAD base version is understood */ + if (mad->mad_hdr.base_version != IB_MGMT_BASE_VERSION) { + printk(KERN_ERR PFX "MAD received with unsupported base " + "version %d\n", mad->mad_hdr.base_version); + goto out; + } + + /* Filter SMI packets sent to other than QP0 */ + if ((mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_SUBN_LID_ROUTED) || + (mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE)) { + if (qp_num == 0) + valid = 1; + } else { + /* Filter GSI packets sent to QP0 */ + if (qp_num != 0) + valid = 1; + } + +out: + return valid; +} + +/* + * Return start of fully reassembled MAD, or NULL, if MAD isn't assembled yet + */ +static struct ib_mad_private * +reassemble_recv(struct ib_mad_agent_private *mad_agent_priv, + struct ib_mad_private *recv) +{ + /* Until we have RMPP, all receives are reassembled!... */ + INIT_LIST_HEAD(&recv->header.recv_wc.recv_buf.list); + return recv; +} + +static struct ib_mad_send_wr_private* +find_send_req(struct ib_mad_agent_private *mad_agent_priv, + u64 tid) +{ + struct ib_mad_send_wr_private *mad_send_wr; + + list_for_each_entry(mad_send_wr, &mad_agent_priv->wait_list, + agent_list) { + if (mad_send_wr->tid == tid) + return mad_send_wr; + } + + /* + * It's possible to receive the response before we've + * been notified that the send has completed + */ + list_for_each_entry(mad_send_wr, &mad_agent_priv->send_list, + agent_list) { + if (mad_send_wr->tid == tid && mad_send_wr->timeout) { + /* Verify request has not been canceled */ + return (mad_send_wr->status == IB_WC_SUCCESS) ? + mad_send_wr : NULL; + } + } + return NULL; +} + +static void ib_mad_complete_recv(struct ib_mad_agent_private *mad_agent_priv, + struct ib_mad_private *recv, + int solicited) +{ + struct ib_mad_send_wr_private *mad_send_wr; + struct ib_mad_send_wc mad_send_wc; + unsigned long flags; + + /* Fully reassemble receive before processing */ + recv = reassemble_recv(mad_agent_priv, recv); + if (!recv) { + if (atomic_dec_and_test(&mad_agent_priv->refcount)) + wake_up(&mad_agent_priv->wait); + return; + } + + /* Complete corresponding request */ + if (solicited) { + spin_lock_irqsave(&mad_agent_priv->lock, flags); + mad_send_wr = find_send_req(mad_agent_priv, + recv->mad.mad.mad_hdr.tid); + if (!mad_send_wr) { + spin_unlock_irqrestore(&mad_agent_priv->lock, flags); + ib_free_recv_mad(&recv->header.recv_wc); + if (atomic_dec_and_test(&mad_agent_priv->refcount)) + wake_up(&mad_agent_priv->wait); + return; + } + /* Timeout = 0 means that we won't wait for a response */ + mad_send_wr->timeout = 0; + spin_unlock_irqrestore(&mad_agent_priv->lock, flags); + + /* Defined behavior is to complete response before request */ + recv->header.recv_wc.wc->wr_id = mad_send_wr->wr_id; + mad_agent_priv->agent.recv_handler( + &mad_agent_priv->agent, + &recv->header.recv_wc); + atomic_dec(&mad_agent_priv->refcount); + + mad_send_wc.status = IB_WC_SUCCESS; + mad_send_wc.vendor_err = 0; + mad_send_wc.wr_id = mad_send_wr->wr_id; + ib_mad_complete_send_wr(mad_send_wr, &mad_send_wc); + } else { + mad_agent_priv->agent.recv_handler( + &mad_agent_priv->agent, + &recv->header.recv_wc); + if (atomic_dec_and_test(&mad_agent_priv->refcount)) + wake_up(&mad_agent_priv->wait); + } +} + +static void ib_mad_recv_done_handler(struct ib_mad_port_private *port_priv, + struct ib_wc *wc) +{ + struct ib_mad_qp_info *qp_info; + struct ib_mad_private_header *mad_priv_hdr; + struct ib_mad_private *recv, *response; + struct ib_mad_list_head *mad_list; + struct ib_mad_agent_private *mad_agent; + int solicited; + + response = kmem_cache_alloc(ib_mad_cache, GFP_KERNEL); + if (!response) + printk(KERN_ERR PFX "ib_mad_recv_done_handler no memory " + "for response buffer\n"); + + mad_list = (struct ib_mad_list_head *)(unsigned long)wc->wr_id; + qp_info = mad_list->mad_queue->qp_info; + dequeue_mad(mad_list); + + mad_priv_hdr = container_of(mad_list, struct ib_mad_private_header, + mad_list); + recv = container_of(mad_priv_hdr, struct ib_mad_private, header); + dma_unmap_single(port_priv->device->dma_device, + pci_unmap_addr(&recv->header, mapping), + sizeof(struct ib_mad_private) - + sizeof(struct ib_mad_private_header), + DMA_FROM_DEVICE); + + /* Setup MAD receive work completion from "normal" work completion */ + recv->header.recv_wc.wc = wc; + recv->header.recv_wc.mad_len = sizeof(struct ib_mad); + recv->header.recv_wc.recv_buf.mad = &recv->mad.mad; + recv->header.recv_wc.recv_buf.grh = &recv->grh; + + if (atomic_read(&qp_info->snoop_count)) + snoop_recv(qp_info, &recv->header.recv_wc, IB_MAD_SNOOP_RECVS); + + /* Validate MAD */ + if (!validate_mad(&recv->mad.mad, qp_info->qp->qp_num)) + goto out; + + if (recv->mad.mad.mad_hdr.mgmt_class == + IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE) { + if (!smi_handle_dr_smp_recv(&recv->mad.smp, + port_priv->device->node_type, + port_priv->port_num, + port_priv->device->phys_port_cnt)) + goto out; + if (!smi_check_forward_dr_smp(&recv->mad.smp)) + goto local; + if (!smi_handle_dr_smp_send(&recv->mad.smp, + port_priv->device->node_type, + port_priv->port_num)) + goto out; + if (!smi_check_local_dr_smp(&recv->mad.smp, + port_priv->device, + port_priv->port_num)) + goto out; + } + +local: + /* Give driver "right of first refusal" on incoming MAD */ + if (port_priv->device->process_mad) { + int ret; + + if (!response) { + printk(KERN_ERR PFX "No memory for response MAD\n"); + /* + * Is it better to assume that + * it wouldn't be processed ? + */ + goto out; + } + + ret = port_priv->device->process_mad(port_priv->device, 0, + port_priv->port_num, + wc, &recv->grh, + &recv->mad.mad, + &response->mad.mad); + if (ret & IB_MAD_RESULT_SUCCESS) { + if (ret & IB_MAD_RESULT_CONSUMED) + goto out; + if (ret & IB_MAD_RESULT_REPLY) { + /* Send response */ + if (!agent_send(response, &recv->grh, wc, + port_priv->device, + port_priv->port_num)) + response = NULL; + goto out; + } + } + } + + /* Determine corresponding MAD agent for incoming receive MAD */ + solicited = solicited_mad(&recv->mad.mad); + mad_agent = find_mad_agent(port_priv, &recv->mad.mad, solicited); + if (mad_agent) { + ib_mad_complete_recv(mad_agent, recv, solicited); + /* + * recv is freed up in error cases in ib_mad_complete_recv + * or via recv_handler in ib_mad_complete_recv() + */ + recv = NULL; + } + +out: + /* Post another receive request for this QP */ + if (response) { + ib_mad_post_receive_mads(qp_info, response); + if (recv) + kmem_cache_free(ib_mad_cache, recv); + } else + ib_mad_post_receive_mads(qp_info, recv); +} + +static void adjust_timeout(struct ib_mad_agent_private *mad_agent_priv) +{ + struct ib_mad_send_wr_private *mad_send_wr; + unsigned long delay; + + if (list_empty(&mad_agent_priv->wait_list)) { + cancel_delayed_work(&mad_agent_priv->timed_work); + } else { + mad_send_wr = list_entry(mad_agent_priv->wait_list.next, + struct ib_mad_send_wr_private, + agent_list); + + if (time_after(mad_agent_priv->timeout, + mad_send_wr->timeout)) { + mad_agent_priv->timeout = mad_send_wr->timeout; + cancel_delayed_work(&mad_agent_priv->timed_work); + delay = mad_send_wr->timeout - jiffies; + if ((long)delay <= 0) + delay = 1; + queue_delayed_work(mad_agent_priv->qp_info-> + port_priv->wq, + &mad_agent_priv->timed_work, delay); + } + } +} + +static void wait_for_response(struct ib_mad_agent_private *mad_agent_priv, + struct ib_mad_send_wr_private *mad_send_wr ) +{ + struct ib_mad_send_wr_private *temp_mad_send_wr; + struct list_head *list_item; + unsigned long delay; + + list_del(&mad_send_wr->agent_list); + + delay = mad_send_wr->timeout; + mad_send_wr->timeout += jiffies; + + list_for_each_prev(list_item, &mad_agent_priv->wait_list) { + temp_mad_send_wr = list_entry(list_item, + struct ib_mad_send_wr_private, + agent_list); + if (time_after(mad_send_wr->timeout, + temp_mad_send_wr->timeout)) + break; + } + list_add(&mad_send_wr->agent_list, list_item); + + /* Reschedule a work item if we have a shorter timeout */ + if (mad_agent_priv->wait_list.next == &mad_send_wr->agent_list) { + cancel_delayed_work(&mad_agent_priv->timed_work); + queue_delayed_work(mad_agent_priv->qp_info->port_priv->wq, + &mad_agent_priv->timed_work, delay); + } +} + +/* + * Process a send work completion + */ +static void ib_mad_complete_send_wr(struct ib_mad_send_wr_private *mad_send_wr, + struct ib_mad_send_wc *mad_send_wc) +{ + struct ib_mad_agent_private *mad_agent_priv; + unsigned long flags; + + mad_agent_priv = container_of(mad_send_wr->agent, + struct ib_mad_agent_private, agent); + + spin_lock_irqsave(&mad_agent_priv->lock, flags); + if (mad_send_wc->status != IB_WC_SUCCESS && + mad_send_wr->status == IB_WC_SUCCESS) { + mad_send_wr->status = mad_send_wc->status; + mad_send_wr->refcount -= (mad_send_wr->timeout > 0); + } + + if (--mad_send_wr->refcount > 0) { + if (mad_send_wr->refcount == 1 && mad_send_wr->timeout && + mad_send_wr->status == IB_WC_SUCCESS) { + wait_for_response(mad_agent_priv, mad_send_wr); + } + spin_unlock_irqrestore(&mad_agent_priv->lock, flags); + return; + } + + /* Remove send from MAD agent and notify client of completion */ + list_del(&mad_send_wr->agent_list); + adjust_timeout(mad_agent_priv); + spin_unlock_irqrestore(&mad_agent_priv->lock, flags); + + if (mad_send_wr->status != IB_WC_SUCCESS ) + mad_send_wc->status = mad_send_wr->status; + mad_agent_priv->agent.send_handler(&mad_agent_priv->agent, + mad_send_wc); + + /* Release reference on agent taken when sending */ + if (atomic_dec_and_test(&mad_agent_priv->refcount)) + wake_up(&mad_agent_priv->wait); + + kfree(mad_send_wr); +} + +static void ib_mad_send_done_handler(struct ib_mad_port_private *port_priv, + struct ib_wc *wc) +{ + struct ib_mad_send_wr_private *mad_send_wr, *queued_send_wr; + struct ib_mad_list_head *mad_list; + struct ib_mad_qp_info *qp_info; + struct ib_mad_queue *send_queue; + struct ib_send_wr *bad_send_wr; + unsigned long flags; + int ret; + + mad_list = (struct ib_mad_list_head *)(unsigned long)wc->wr_id; + mad_send_wr = container_of(mad_list, struct ib_mad_send_wr_private, + mad_list); + send_queue = mad_list->mad_queue; + qp_info = send_queue->qp_info; + +retry: + queued_send_wr = NULL; + spin_lock_irqsave(&send_queue->lock, flags); + list_del(&mad_list->list); + + /* Move queued send to the send queue */ + if (send_queue->count-- > send_queue->max_active) { + mad_list = container_of(qp_info->overflow_list.next, + struct ib_mad_list_head, list); + queued_send_wr = container_of(mad_list, + struct ib_mad_send_wr_private, + mad_list); + list_del(&mad_list->list); + list_add_tail(&mad_list->list, &send_queue->list); + } + spin_unlock_irqrestore(&send_queue->lock, flags); + + /* Restore client wr_id in WC and complete send */ + wc->wr_id = mad_send_wr->wr_id; + if (atomic_read(&qp_info->snoop_count)) + snoop_send(qp_info, &mad_send_wr->send_wr, + (struct ib_mad_send_wc *)wc, + IB_MAD_SNOOP_SEND_COMPLETIONS); + ib_mad_complete_send_wr(mad_send_wr, (struct ib_mad_send_wc *)wc); + + if (queued_send_wr) { + ret = ib_post_send(qp_info->qp, &queued_send_wr->send_wr, + &bad_send_wr); + if (ret) { + printk(KERN_ERR PFX "ib_post_send failed: %d\n", ret); + mad_send_wr = queued_send_wr; + wc->status = IB_WC_LOC_QP_OP_ERR; + goto retry; + } + } +} + +static void mark_sends_for_retry(struct ib_mad_qp_info *qp_info) +{ + struct ib_mad_send_wr_private *mad_send_wr; + struct ib_mad_list_head *mad_list; + unsigned long flags; + + spin_lock_irqsave(&qp_info->send_queue.lock, flags); + list_for_each_entry(mad_list, &qp_info->send_queue.list, list) { + mad_send_wr = container_of(mad_list, + struct ib_mad_send_wr_private, + mad_list); + mad_send_wr->retry = 1; + } + spin_unlock_irqrestore(&qp_info->send_queue.lock, flags); +} + +static void mad_error_handler(struct ib_mad_port_private *port_priv, + struct ib_wc *wc) +{ + struct ib_mad_list_head *mad_list; + struct ib_mad_qp_info *qp_info; + struct ib_mad_send_wr_private *mad_send_wr; + int ret; + + /* Determine if failure was a send or receive */ + mad_list = (struct ib_mad_list_head *)(unsigned long)wc->wr_id; + qp_info = mad_list->mad_queue->qp_info; + if (mad_list->mad_queue == &qp_info->recv_queue) + /* + * Receive errors indicate that the QP has entered the error + * state - error handling/shutdown code will cleanup + */ + return; + + /* + * Send errors will transition the QP to SQE - move + * QP to RTS and repost flushed work requests + */ + mad_send_wr = container_of(mad_list, struct ib_mad_send_wr_private, + mad_list); + if (wc->status == IB_WC_WR_FLUSH_ERR) { + if (mad_send_wr->retry) { + /* Repost send */ + struct ib_send_wr *bad_send_wr; + + mad_send_wr->retry = 0; + ret = ib_post_send(qp_info->qp, &mad_send_wr->send_wr, + &bad_send_wr); + if (ret) + ib_mad_send_done_handler(port_priv, wc); + } else + ib_mad_send_done_handler(port_priv, wc); + } else { + struct ib_qp_attr *attr; + + /* Transition QP to RTS and fail offending send */ + attr = kmalloc(sizeof *attr, GFP_KERNEL); + if (attr) { + attr->qp_state = IB_QPS_RTS; + attr->cur_qp_state = IB_QPS_SQE; + ret = ib_modify_qp(qp_info->qp, attr, + IB_QP_STATE | IB_QP_CUR_STATE); + kfree(attr); + if (ret) + printk(KERN_ERR PFX "mad_error_handler - " + "ib_modify_qp to RTS : %d\n", ret); + else + mark_sends_for_retry(qp_info); + } + ib_mad_send_done_handler(port_priv, wc); + } +} + +/* + * IB MAD completion callback + */ +static void ib_mad_completion_handler(void *data) +{ + struct ib_mad_port_private *port_priv; + struct ib_wc wc; + + port_priv = (struct ib_mad_port_private *)data; + ib_req_notify_cq(port_priv->cq, IB_CQ_NEXT_COMP); + + while (ib_poll_cq(port_priv->cq, 1, &wc) == 1) { + if (wc.status == IB_WC_SUCCESS) { + switch (wc.opcode) { + case IB_WC_SEND: + ib_mad_send_done_handler(port_priv, &wc); + break; + case IB_WC_RECV: + ib_mad_recv_done_handler(port_priv, &wc); + break; + default: + BUG_ON(1); + break; + } + } else + mad_error_handler(port_priv, &wc); + } +} + +static void cancel_mads(struct ib_mad_agent_private *mad_agent_priv) +{ + unsigned long flags; + struct ib_mad_send_wr_private *mad_send_wr, *temp_mad_send_wr; + struct ib_mad_send_wc mad_send_wc; + struct list_head cancel_list; + + INIT_LIST_HEAD(&cancel_list); + + spin_lock_irqsave(&mad_agent_priv->lock, flags); + list_for_each_entry_safe(mad_send_wr, temp_mad_send_wr, + &mad_agent_priv->send_list, agent_list) { + if (mad_send_wr->status == IB_WC_SUCCESS) { + mad_send_wr->status = IB_WC_WR_FLUSH_ERR; + mad_send_wr->refcount -= (mad_send_wr->timeout > 0); + } + } + + /* Empty wait list to prevent receives from finding a request */ + list_splice_init(&mad_agent_priv->wait_list, &cancel_list); + spin_unlock_irqrestore(&mad_agent_priv->lock, flags); + + /* Report all cancelled requests */ + mad_send_wc.status = IB_WC_WR_FLUSH_ERR; + mad_send_wc.vendor_err = 0; + + list_for_each_entry_safe(mad_send_wr, temp_mad_send_wr, + &cancel_list, agent_list) { + mad_send_wc.wr_id = mad_send_wr->wr_id; + mad_agent_priv->agent.send_handler(&mad_agent_priv->agent, + &mad_send_wc); + + list_del(&mad_send_wr->agent_list); + kfree(mad_send_wr); + atomic_dec(&mad_agent_priv->refcount); + } +} + +static struct ib_mad_send_wr_private* +find_send_by_wr_id(struct ib_mad_agent_private *mad_agent_priv, + u64 wr_id) +{ + struct ib_mad_send_wr_private *mad_send_wr; + + list_for_each_entry(mad_send_wr, &mad_agent_priv->wait_list, + agent_list) { + if (mad_send_wr->wr_id == wr_id) + return mad_send_wr; + } + + list_for_each_entry(mad_send_wr, &mad_agent_priv->send_list, + agent_list) { + if (mad_send_wr->wr_id == wr_id) + return mad_send_wr; + } + return NULL; +} + +void cancel_sends(void *data) +{ + struct ib_mad_agent_private *mad_agent_priv; + struct ib_mad_send_wr_private *mad_send_wr; + struct ib_mad_send_wc mad_send_wc; + unsigned long flags; + + mad_agent_priv = data; + + mad_send_wc.status = IB_WC_WR_FLUSH_ERR; + mad_send_wc.vendor_err = 0; + + spin_lock_irqsave(&mad_agent_priv->lock, flags); + while (!list_empty(&mad_agent_priv->canceled_list)) { + mad_send_wr = list_entry(mad_agent_priv->canceled_list.next, + struct ib_mad_send_wr_private, + agent_list); + + list_del(&mad_send_wr->agent_list); + spin_unlock_irqrestore(&mad_agent_priv->lock, flags); + + mad_send_wc.wr_id = mad_send_wr->wr_id; + mad_agent_priv->agent.send_handler(&mad_agent_priv->agent, + &mad_send_wc); + + kfree(mad_send_wr); + if (atomic_dec_and_test(&mad_agent_priv->refcount)) + wake_up(&mad_agent_priv->wait); + spin_lock_irqsave(&mad_agent_priv->lock, flags); + } + spin_unlock_irqrestore(&mad_agent_priv->lock, flags); +} + +void ib_cancel_mad(struct ib_mad_agent *mad_agent, + u64 wr_id) +{ + struct ib_mad_agent_private *mad_agent_priv; + struct ib_mad_send_wr_private *mad_send_wr; + unsigned long flags; + + mad_agent_priv = container_of(mad_agent, struct ib_mad_agent_private, + agent); + spin_lock_irqsave(&mad_agent_priv->lock, flags); + mad_send_wr = find_send_by_wr_id(mad_agent_priv, wr_id); + if (!mad_send_wr) { + spin_unlock_irqrestore(&mad_agent_priv->lock, flags); + goto out; + } + + if (mad_send_wr->status == IB_WC_SUCCESS) + mad_send_wr->refcount -= (mad_send_wr->timeout > 0); + + if (mad_send_wr->refcount != 0) { + mad_send_wr->status = IB_WC_WR_FLUSH_ERR; + spin_unlock_irqrestore(&mad_agent_priv->lock, flags); + goto out; + } + + list_del(&mad_send_wr->agent_list); + list_add_tail(&mad_send_wr->agent_list, &mad_agent_priv->canceled_list); + adjust_timeout(mad_agent_priv); + spin_unlock_irqrestore(&mad_agent_priv->lock, flags); + + queue_work(mad_agent_priv->qp_info->port_priv->wq, + &mad_agent_priv->canceled_work); +out: + return; +} +EXPORT_SYMBOL(ib_cancel_mad); + +static void local_completions(void *data) +{ + struct ib_mad_agent_private *mad_agent_priv; + struct ib_mad_local_private *local; + struct ib_mad_agent_private *recv_mad_agent; + unsigned long flags; + struct ib_wc wc; + struct ib_mad_send_wc mad_send_wc; + + mad_agent_priv = (struct ib_mad_agent_private *)data; + + spin_lock_irqsave(&mad_agent_priv->lock, flags); + while (!list_empty(&mad_agent_priv->local_list)) { + local = list_entry(mad_agent_priv->local_list.next, + struct ib_mad_local_private, + completion_list); + spin_unlock_irqrestore(&mad_agent_priv->lock, flags); + if (local->mad_priv) { + recv_mad_agent = local->recv_mad_agent; + if (!recv_mad_agent) { + printk(KERN_ERR PFX "No receive MAD agent for local completion\n"); + kmem_cache_free(ib_mad_cache, local->mad_priv); + goto local_send_completion; + } + + /* + * Defined behavior is to complete response + * before request + */ + build_smp_wc(local->wr_id, IB_LID_PERMISSIVE, + 0 /* pkey index */, + recv_mad_agent->agent.port_num, &wc); + + local->mad_priv->header.recv_wc.wc = &wc; + local->mad_priv->header.recv_wc.mad_len = + sizeof(struct ib_mad); + INIT_LIST_HEAD(&local->mad_priv->header.recv_wc.recv_buf.list); + local->mad_priv->header.recv_wc.recv_buf.grh = NULL; + local->mad_priv->header.recv_wc.recv_buf.mad = + &local->mad_priv->mad.mad; + if (atomic_read(&recv_mad_agent->qp_info->snoop_count)) + snoop_recv(recv_mad_agent->qp_info, + &local->mad_priv->header.recv_wc, + IB_MAD_SNOOP_RECVS); + recv_mad_agent->agent.recv_handler( + &recv_mad_agent->agent, + &local->mad_priv->header.recv_wc); + spin_lock_irqsave(&recv_mad_agent->lock, flags); + atomic_dec(&recv_mad_agent->refcount); + spin_unlock_irqrestore(&recv_mad_agent->lock, flags); + } + +local_send_completion: + /* Complete send */ + mad_send_wc.status = IB_WC_SUCCESS; + mad_send_wc.vendor_err = 0; + mad_send_wc.wr_id = local->wr_id; + if (atomic_read(&mad_agent_priv->qp_info->snoop_count)) + snoop_send(mad_agent_priv->qp_info, &local->send_wr, + &mad_send_wc, + IB_MAD_SNOOP_SEND_COMPLETIONS); + mad_agent_priv->agent.send_handler(&mad_agent_priv->agent, + &mad_send_wc); + + spin_lock_irqsave(&mad_agent_priv->lock, flags); + list_del(&local->completion_list); + atomic_dec(&mad_agent_priv->refcount); + kfree(local); + } + spin_unlock_irqrestore(&mad_agent_priv->lock, flags); +} + +static void timeout_sends(void *data) +{ + struct ib_mad_agent_private *mad_agent_priv; + struct ib_mad_send_wr_private *mad_send_wr; + struct ib_mad_send_wc mad_send_wc; + unsigned long flags, delay; + + mad_agent_priv = (struct ib_mad_agent_private *)data; + + mad_send_wc.status = IB_WC_RESP_TIMEOUT_ERR; + mad_send_wc.vendor_err = 0; + + spin_lock_irqsave(&mad_agent_priv->lock, flags); + while (!list_empty(&mad_agent_priv->wait_list)) { + mad_send_wr = list_entry(mad_agent_priv->wait_list.next, + struct ib_mad_send_wr_private, + agent_list); + + if (time_after(mad_send_wr->timeout, jiffies)) { + delay = mad_send_wr->timeout - jiffies; + if ((long)delay <= 0) + delay = 1; + queue_delayed_work(mad_agent_priv->qp_info-> + port_priv->wq, + &mad_agent_priv->timed_work, delay); + break; + } + + list_del(&mad_send_wr->agent_list); + spin_unlock_irqrestore(&mad_agent_priv->lock, flags); + + mad_send_wc.wr_id = mad_send_wr->wr_id; + mad_agent_priv->agent.send_handler(&mad_agent_priv->agent, + &mad_send_wc); + + kfree(mad_send_wr); + atomic_dec(&mad_agent_priv->refcount); + spin_lock_irqsave(&mad_agent_priv->lock, flags); + } + spin_unlock_irqrestore(&mad_agent_priv->lock, flags); +} + +static void ib_mad_thread_completion_handler(struct ib_cq *cq) +{ + struct ib_mad_port_private *port_priv = cq->cq_context; + + queue_work(port_priv->wq, &port_priv->work); +} + +/* + * Allocate receive MADs and post receive WRs for them + */ +static int ib_mad_post_receive_mads(struct ib_mad_qp_info *qp_info, + struct ib_mad_private *mad) +{ + unsigned long flags; + int post, ret; + struct ib_mad_private *mad_priv; + struct ib_sge sg_list; + struct ib_recv_wr recv_wr, *bad_recv_wr; + struct ib_mad_queue *recv_queue = &qp_info->recv_queue; + + /* Initialize common scatter list fields */ + sg_list.length = sizeof *mad_priv - sizeof mad_priv->header; + sg_list.lkey = (*qp_info->port_priv->mr).lkey; + + /* Initialize common receive WR fields */ + recv_wr.next = NULL; + recv_wr.sg_list = &sg_list; + recv_wr.num_sge = 1; + + do { + /* Allocate and map receive buffer */ + if (mad) { + mad_priv = mad; + mad = NULL; + } else { + mad_priv = kmem_cache_alloc(ib_mad_cache, GFP_KERNEL); + if (!mad_priv) { + printk(KERN_ERR PFX "No memory for receive buffer\n"); + ret = -ENOMEM; + break; + } + } + sg_list.addr = dma_map_single(qp_info->port_priv-> + device->dma_device, + &mad_priv->grh, + sizeof *mad_priv - + sizeof mad_priv->header, + DMA_FROM_DEVICE); + pci_unmap_addr_set(&mad_priv->header, mapping, sg_list.addr); + recv_wr.wr_id = (unsigned long)&mad_priv->header.mad_list; + mad_priv->header.mad_list.mad_queue = recv_queue; + + /* Post receive WR */ + spin_lock_irqsave(&recv_queue->lock, flags); + post = (++recv_queue->count < recv_queue->max_active); + list_add_tail(&mad_priv->header.mad_list.list, &recv_queue->list); + spin_unlock_irqrestore(&recv_queue->lock, flags); + ret = ib_post_recv(qp_info->qp, &recv_wr, &bad_recv_wr); + if (ret) { + spin_lock_irqsave(&recv_queue->lock, flags); + list_del(&mad_priv->header.mad_list.list); + recv_queue->count--; + spin_unlock_irqrestore(&recv_queue->lock, flags); + dma_unmap_single(qp_info->port_priv->device->dma_device, + pci_unmap_addr(&mad_priv->header, + mapping), + sizeof *mad_priv - + sizeof mad_priv->header, + DMA_FROM_DEVICE); + kmem_cache_free(ib_mad_cache, mad_priv); + printk(KERN_ERR PFX "ib_post_recv failed: %d\n", ret); + break; + } + } while (post); + + return ret; +} + +/* + * Return all the posted receive MADs + */ +static void cleanup_recv_queue(struct ib_mad_qp_info *qp_info) +{ + struct ib_mad_private_header *mad_priv_hdr; + struct ib_mad_private *recv; + struct ib_mad_list_head *mad_list; + + while (!list_empty(&qp_info->recv_queue.list)) { + + mad_list = list_entry(qp_info->recv_queue.list.next, + struct ib_mad_list_head, list); + mad_priv_hdr = container_of(mad_list, + struct ib_mad_private_header, + mad_list); + recv = container_of(mad_priv_hdr, struct ib_mad_private, + header); + + /* Remove from posted receive MAD list */ + list_del(&mad_list->list); + + /* Undo PCI mapping */ + dma_unmap_single(qp_info->port_priv->device->dma_device, + pci_unmap_addr(&recv->header, mapping), + sizeof(struct ib_mad_private) - + sizeof(struct ib_mad_private_header), + DMA_FROM_DEVICE); + kmem_cache_free(ib_mad_cache, recv); + } + + qp_info->recv_queue.count = 0; +} + +/* + * Start the port + */ +static int ib_mad_port_start(struct ib_mad_port_private *port_priv) +{ + int ret, i; + struct ib_qp_attr *attr; + struct ib_qp *qp; + + attr = kmalloc(sizeof *attr, GFP_KERNEL); + if (!attr) { + printk(KERN_ERR PFX "Couldn't kmalloc ib_qp_attr\n"); + return -ENOMEM; + } + + for (i = 0; i < IB_MAD_QPS_CORE; i++) { + qp = port_priv->qp_info[i].qp; + /* + * PKey index for QP1 is irrelevant but + * one is needed for the Reset to Init transition + */ + attr->qp_state = IB_QPS_INIT; + attr->pkey_index = 0; + attr->qkey = (qp->qp_num == 0) ? 0 : IB_QP1_QKEY; + ret = ib_modify_qp(qp, attr, IB_QP_STATE | + IB_QP_PKEY_INDEX | IB_QP_QKEY); + if (ret) { + printk(KERN_ERR PFX "Couldn't change QP%d state to " + "INIT: %d\n", i, ret); + goto out; + } + + attr->qp_state = IB_QPS_RTR; + ret = ib_modify_qp(qp, attr, IB_QP_STATE); + if (ret) { + printk(KERN_ERR PFX "Couldn't change QP%d state to " + "RTR: %d\n", i, ret); + goto out; + } + + attr->qp_state = IB_QPS_RTS; + attr->sq_psn = IB_MAD_SEND_Q_PSN; + ret = ib_modify_qp(qp, attr, IB_QP_STATE | IB_QP_SQ_PSN); + if (ret) { + printk(KERN_ERR PFX "Couldn't change QP%d state to " + "RTS: %d\n", i, ret); + goto out; + } + } + + ret = ib_req_notify_cq(port_priv->cq, IB_CQ_NEXT_COMP); + if (ret) { + printk(KERN_ERR PFX "Failed to request completion " + "notification: %d\n", ret); + goto out; + } + + for (i = 0; i < IB_MAD_QPS_CORE; i++) { + ret = ib_mad_post_receive_mads(&port_priv->qp_info[i], NULL); + if (ret) { + printk(KERN_ERR PFX "Couldn't post receive WRs\n"); + goto out; + } + } +out: + kfree(attr); + return ret; +} + +static void qp_event_handler(struct ib_event *event, void *qp_context) +{ + struct ib_mad_qp_info *qp_info = qp_context; + + /* It's worse than that! He's dead, Jim! */ + printk(KERN_ERR PFX "Fatal error (%d) on MAD QP (%d)\n", + event->event, qp_info->qp->qp_num); +} + +static void init_mad_queue(struct ib_mad_qp_info *qp_info, + struct ib_mad_queue *mad_queue) +{ + mad_queue->qp_info = qp_info; + mad_queue->count = 0; + spin_lock_init(&mad_queue->lock); + INIT_LIST_HEAD(&mad_queue->list); +} + +static void init_mad_qp(struct ib_mad_port_private *port_priv, + struct ib_mad_qp_info *qp_info) +{ + qp_info->port_priv = port_priv; + init_mad_queue(qp_info, &qp_info->send_queue); + init_mad_queue(qp_info, &qp_info->recv_queue); + INIT_LIST_HEAD(&qp_info->overflow_list); + spin_lock_init(&qp_info->snoop_lock); + qp_info->snoop_table = NULL; + qp_info->snoop_table_size = 0; + atomic_set(&qp_info->snoop_count, 0); +} + +static int create_mad_qp(struct ib_mad_qp_info *qp_info, + enum ib_qp_type qp_type) +{ + struct ib_qp_init_attr qp_init_attr; + int ret; + + memset(&qp_init_attr, 0, sizeof qp_init_attr); + qp_init_attr.send_cq = qp_info->port_priv->cq; + qp_init_attr.recv_cq = qp_info->port_priv->cq; + qp_init_attr.sq_sig_type = IB_SIGNAL_ALL_WR; + qp_init_attr.cap.max_send_wr = IB_MAD_QP_SEND_SIZE; + qp_init_attr.cap.max_recv_wr = IB_MAD_QP_RECV_SIZE; + qp_init_attr.cap.max_send_sge = IB_MAD_SEND_REQ_MAX_SG; + qp_init_attr.cap.max_recv_sge = IB_MAD_RECV_REQ_MAX_SG; + qp_init_attr.qp_type = qp_type; + qp_init_attr.port_num = qp_info->port_priv->port_num; + qp_init_attr.qp_context = qp_info; + qp_init_attr.event_handler = qp_event_handler; + qp_info->qp = ib_create_qp(qp_info->port_priv->pd, &qp_init_attr); + if (IS_ERR(qp_info->qp)) { + printk(KERN_ERR PFX "Couldn't create ib_mad QP%d\n", + get_spl_qp_index(qp_type)); + ret = PTR_ERR(qp_info->qp); + goto error; + } + /* Use minimum queue sizes unless the CQ is resized */ + qp_info->send_queue.max_active = IB_MAD_QP_SEND_SIZE; + qp_info->recv_queue.max_active = IB_MAD_QP_RECV_SIZE; + return 0; + +error: + return ret; +} + +static void destroy_mad_qp(struct ib_mad_qp_info *qp_info) +{ + ib_destroy_qp(qp_info->qp); + if (qp_info->snoop_table) + kfree(qp_info->snoop_table); +} + +/* + * Open the port + * Create the QP, PD, MR, and CQ if needed + */ +static int ib_mad_port_open(struct ib_device *device, + int port_num) +{ + int ret, cq_size; + struct ib_mad_port_private *port_priv; + unsigned long flags; + char name[sizeof "ib_mad123"]; + + /* First, check if port already open at MAD layer */ + port_priv = ib_get_mad_port(device, port_num); + if (port_priv) { + printk(KERN_DEBUG PFX "%s port %d already open\n", + device->name, port_num); + return 0; + } + + /* Create new device info */ + port_priv = kmalloc(sizeof *port_priv, GFP_KERNEL); + if (!port_priv) { + printk(KERN_ERR PFX "No memory for ib_mad_port_private\n"); + return -ENOMEM; + } + memset(port_priv, 0, sizeof *port_priv); + port_priv->device = device; + port_priv->port_num = port_num; + spin_lock_init(&port_priv->reg_lock); + INIT_LIST_HEAD(&port_priv->agent_list); + init_mad_qp(port_priv, &port_priv->qp_info[0]); + init_mad_qp(port_priv, &port_priv->qp_info[1]); + + cq_size = (IB_MAD_QP_SEND_SIZE + IB_MAD_QP_RECV_SIZE) * 2; + port_priv->cq = ib_create_cq(port_priv->device, + (ib_comp_handler) + ib_mad_thread_completion_handler, + NULL, port_priv, cq_size); + if (IS_ERR(port_priv->cq)) { + printk(KERN_ERR PFX "Couldn't create ib_mad CQ\n"); + ret = PTR_ERR(port_priv->cq); + goto error3; + } + + port_priv->pd = ib_alloc_pd(device); + if (IS_ERR(port_priv->pd)) { + printk(KERN_ERR PFX "Couldn't create ib_mad PD\n"); + ret = PTR_ERR(port_priv->pd); + goto error4; + } + + port_priv->mr = ib_get_dma_mr(port_priv->pd, IB_ACCESS_LOCAL_WRITE); + if (IS_ERR(port_priv->mr)) { + printk(KERN_ERR PFX "Couldn't get ib_mad DMA MR\n"); + ret = PTR_ERR(port_priv->mr); + goto error5; + } + + ret = create_mad_qp(&port_priv->qp_info[0], IB_QPT_SMI); + if (ret) + goto error6; + ret = create_mad_qp(&port_priv->qp_info[1], IB_QPT_GSI); + if (ret) + goto error7; + + snprintf(name, sizeof name, "ib_mad%d", port_num); + port_priv->wq = create_singlethread_workqueue(name); + if (!port_priv->wq) { + ret = -ENOMEM; + goto error8; + } + INIT_WORK(&port_priv->work, ib_mad_completion_handler, port_priv); + + ret = ib_mad_port_start(port_priv); + if (ret) { + printk(KERN_ERR PFX "Couldn't start port\n"); + goto error9; + } + + spin_lock_irqsave(&ib_mad_port_list_lock, flags); + list_add_tail(&port_priv->port_list, &ib_mad_port_list); + spin_unlock_irqrestore(&ib_mad_port_list_lock, flags); + return 0; + +error9: + destroy_workqueue(port_priv->wq); +error8: + destroy_mad_qp(&port_priv->qp_info[1]); +error7: + destroy_mad_qp(&port_priv->qp_info[0]); +error6: + ib_dereg_mr(port_priv->mr); +error5: + ib_dealloc_pd(port_priv->pd); +error4: + ib_destroy_cq(port_priv->cq); + cleanup_recv_queue(&port_priv->qp_info[1]); + cleanup_recv_queue(&port_priv->qp_info[0]); +error3: + kfree(port_priv); + + return ret; +} + +/* + * Close the port + * If there are no classes using the port, free the port + * resources (CQ, MR, PD, QP) and remove the port's info structure + */ +static int ib_mad_port_close(struct ib_device *device, int port_num) +{ + struct ib_mad_port_private *port_priv; + unsigned long flags; + + spin_lock_irqsave(&ib_mad_port_list_lock, flags); + port_priv = __ib_get_mad_port(device, port_num); + if (port_priv == NULL) { + spin_unlock_irqrestore(&ib_mad_port_list_lock, flags); + printk(KERN_ERR PFX "Port %d not found\n", port_num); + return -ENODEV; + } + list_del(&port_priv->port_list); + spin_unlock_irqrestore(&ib_mad_port_list_lock, flags); + + /* Stop processing completions. */ + flush_workqueue(port_priv->wq); + destroy_workqueue(port_priv->wq); + destroy_mad_qp(&port_priv->qp_info[1]); + destroy_mad_qp(&port_priv->qp_info[0]); + ib_dereg_mr(port_priv->mr); + ib_dealloc_pd(port_priv->pd); + ib_destroy_cq(port_priv->cq); + cleanup_recv_queue(&port_priv->qp_info[1]); + cleanup_recv_queue(&port_priv->qp_info[0]); + /* XXX: Handle deallocation of MAD registration tables */ + + kfree(port_priv); + + return 0; +} + +static void ib_mad_init_device(struct ib_device *device) +{ + int ret, num_ports, cur_port, i, ret2; + + if (device->node_type == IB_NODE_SWITCH) { + num_ports = 1; + cur_port = 0; + } else { + num_ports = device->phys_port_cnt; + cur_port = 1; + } + for (i = 0; i < num_ports; i++, cur_port++) { + ret = ib_mad_port_open(device, cur_port); + if (ret) { + printk(KERN_ERR PFX "Couldn't open %s port %d\n", + device->name, cur_port); + goto error_device_open; + } + ret = ib_agent_port_open(device, cur_port); + if (ret) { + printk(KERN_ERR PFX "Couldn't open %s port %d " + "for agents\n", + device->name, cur_port); + goto error_device_open; + } + } + + goto error_device_query; + +error_device_open: + while (i > 0) { + cur_port--; + ret2 = ib_agent_port_close(device, cur_port); + if (ret2) { + printk(KERN_ERR PFX "Couldn't close %s port %d " + "for agents\n", + device->name, cur_port); + } + ret2 = ib_mad_port_close(device, cur_port); + if (ret2) { + printk(KERN_ERR PFX "Couldn't close %s port %d\n", + device->name, cur_port); + } + i--; + } + +error_device_query: + return; +} + +static void ib_mad_remove_device(struct ib_device *device) +{ + int ret = 0, i, num_ports, cur_port, ret2; + + if (device->node_type == IB_NODE_SWITCH) { + num_ports = 1; + cur_port = 0; + } else { + num_ports = device->phys_port_cnt; + cur_port = 1; + } + for (i = 0; i < num_ports; i++, cur_port++) { + ret2 = ib_agent_port_close(device, cur_port); + if (ret2) { + printk(KERN_ERR PFX "Couldn't close %s port %d " + "for agents\n", + device->name, cur_port); + if (!ret) + ret = ret2; + } + ret2 = ib_mad_port_close(device, cur_port); + if (ret2) { + printk(KERN_ERR PFX "Couldn't close %s port %d\n", + device->name, cur_port); + if (!ret) + ret = ret2; + } + } +} + +static struct ib_client mad_client = { + .name = "mad", + .add = ib_mad_init_device, + .remove = ib_mad_remove_device +}; + +static int __init ib_mad_init_module(void) +{ + int ret; + + spin_lock_init(&ib_mad_port_list_lock); + spin_lock_init(&ib_agent_port_list_lock); + + ib_mad_cache = kmem_cache_create("ib_mad", + sizeof(struct ib_mad_private), + 0, + SLAB_HWCACHE_ALIGN, + NULL, + NULL); + if (!ib_mad_cache) { + printk(KERN_ERR PFX "Couldn't create ib_mad cache\n"); + ret = -ENOMEM; + goto error1; + } + + INIT_LIST_HEAD(&ib_mad_port_list); + + if (ib_register_client(&mad_client)) { + printk(KERN_ERR PFX "Couldn't register ib_mad client\n"); + ret = -EINVAL; + goto error2; + } + + return 0; + +error2: + kmem_cache_destroy(ib_mad_cache); +error1: + return ret; +} + +static void __exit ib_mad_cleanup_module(void) +{ + ib_unregister_client(&mad_client); + + if (kmem_cache_destroy(ib_mad_cache)) { + printk(KERN_DEBUG PFX "Failed to destroy ib_mad cache\n"); + } +} + +module_init(ib_mad_init_module); +module_exit(ib_mad_cleanup_module); diff --git a/drivers/infiniband/core/mad_priv.h b/drivers/infiniband/core/mad_priv.h new file mode 100644 index 000000000000..4ba9f726bf1d --- /dev/null +++ b/drivers/infiniband/core/mad_priv.h @@ -0,0 +1,199 @@ +/* + * Copyright (c) 2004, 2005, Voltaire, 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. + * + * $Id: mad_priv.h 1389 2004-12-27 22:56:47Z roland $ + */ + +#ifndef __IB_MAD_PRIV_H__ +#define __IB_MAD_PRIV_H__ + +#include <linux/pci.h> +#include <linux/kthread.h> +#include <linux/workqueue.h> +#include <ib_mad.h> +#include <ib_smi.h> + + +#define PFX "ib_mad: " + +#define IB_MAD_QPS_CORE 2 /* Always QP0 and QP1 as a minimum */ + +/* QP and CQ parameters */ +#define IB_MAD_QP_SEND_SIZE 128 +#define IB_MAD_QP_RECV_SIZE 512 +#define IB_MAD_SEND_REQ_MAX_SG 2 +#define IB_MAD_RECV_REQ_MAX_SG 1 + +#define IB_MAD_SEND_Q_PSN 0 + +/* Registration table sizes */ +#define MAX_MGMT_CLASS 80 +#define MAX_MGMT_VERSION 8 +#define MAX_MGMT_OUI 8 +#define MAX_MGMT_VENDOR_RANGE2 (IB_MGMT_CLASS_VENDOR_RANGE2_END - \ + IB_MGMT_CLASS_VENDOR_RANGE2_START + 1) + +struct ib_mad_list_head { + struct list_head list; + struct ib_mad_queue *mad_queue; +}; + +struct ib_mad_private_header { + struct ib_mad_list_head mad_list; + struct ib_mad_recv_wc recv_wc; + DECLARE_PCI_UNMAP_ADDR(mapping) +} __attribute__ ((packed)); + +struct ib_mad_private { + struct ib_mad_private_header header; + struct ib_grh grh; + union { + struct ib_mad mad; + struct ib_rmpp_mad rmpp_mad; + struct ib_smp smp; + } mad; +} __attribute__ ((packed)); + +struct ib_mad_agent_private { + struct list_head agent_list; + struct ib_mad_agent agent; + struct ib_mad_reg_req *reg_req; + struct ib_mad_qp_info *qp_info; + + spinlock_t lock; + struct list_head send_list; + struct list_head wait_list; + struct work_struct timed_work; + unsigned long timeout; + struct list_head local_list; + struct work_struct local_work; + struct list_head canceled_list; + struct work_struct canceled_work; + + atomic_t refcount; + wait_queue_head_t wait; + u8 rmpp_version; +}; + +struct ib_mad_snoop_private { + struct ib_mad_agent agent; + struct ib_mad_qp_info *qp_info; + int snoop_index; + int mad_snoop_flags; + atomic_t refcount; + wait_queue_head_t wait; +}; + +struct ib_mad_send_wr_private { + struct ib_mad_list_head mad_list; + struct list_head agent_list; + struct ib_mad_agent *agent; + struct ib_send_wr send_wr; + struct ib_sge sg_list[IB_MAD_SEND_REQ_MAX_SG]; + u64 wr_id; /* client WR ID */ + u64 tid; + unsigned long timeout; + int retry; + int refcount; + enum ib_wc_status status; +}; + +struct ib_mad_local_private { + struct list_head completion_list; + struct ib_mad_private *mad_priv; + struct ib_mad_agent_private *recv_mad_agent; + struct ib_send_wr send_wr; + struct ib_sge sg_list[IB_MAD_SEND_REQ_MAX_SG]; + u64 wr_id; /* client WR ID */ + u64 tid; +}; + +struct ib_mad_mgmt_method_table { + struct ib_mad_agent_private *agent[IB_MGMT_MAX_METHODS]; +}; + +struct ib_mad_mgmt_class_table { + struct ib_mad_mgmt_method_table *method_table[MAX_MGMT_CLASS]; +}; + +struct ib_mad_mgmt_vendor_class { + u8 oui[MAX_MGMT_OUI][3]; + struct ib_mad_mgmt_method_table *method_table[MAX_MGMT_OUI]; +}; + +struct ib_mad_mgmt_vendor_class_table { + struct ib_mad_mgmt_vendor_class *vendor_class[MAX_MGMT_VENDOR_RANGE2]; +}; + +struct ib_mad_mgmt_version_table { + struct ib_mad_mgmt_class_table *class; + struct ib_mad_mgmt_vendor_class_table *vendor; +}; + +struct ib_mad_queue { + spinlock_t lock; + struct list_head list; + int count; + int max_active; + struct ib_mad_qp_info *qp_info; +}; + +struct ib_mad_qp_info { + struct ib_mad_port_private *port_priv; + struct ib_qp *qp; + struct ib_mad_queue send_queue; + struct ib_mad_queue recv_queue; + struct list_head overflow_list; + spinlock_t snoop_lock; + struct ib_mad_snoop_private **snoop_table; + int snoop_table_size; + atomic_t snoop_count; +}; + +struct ib_mad_port_private { + struct list_head port_list; + struct ib_device *device; + int port_num; + struct ib_cq *cq; + struct ib_pd *pd; + struct ib_mr *mr; + + spinlock_t reg_lock; + struct ib_mad_mgmt_version_table version[MAX_MGMT_VERSION]; + struct list_head agent_list; + struct workqueue_struct *wq; + struct work_struct work; + struct ib_mad_qp_info qp_info[IB_MAD_QPS_CORE]; +}; + +extern kmem_cache_t *ib_mad_cache; + +#endif /* __IB_MAD_PRIV_H__ */ diff --git a/drivers/infiniband/core/packer.c b/drivers/infiniband/core/packer.c new file mode 100644 index 000000000000..5f15feffeae2 --- /dev/null +++ b/drivers/infiniband/core/packer.c @@ -0,0 +1,201 @@ +/* + * Copyright (c) 2004 Topspin Corporation. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + * + * $Id: packer.c 1349 2004-12-16 21:09:43Z roland $ + */ + +#include <ib_pack.h> + +static u64 value_read(int offset, int size, void *structure) +{ + switch (size) { + case 1: return *(u8 *) (structure + offset); + case 2: return be16_to_cpup((__be16 *) (structure + offset)); + case 4: return be32_to_cpup((__be32 *) (structure + offset)); + case 8: return be64_to_cpup((__be64 *) (structure + offset)); + default: + printk(KERN_WARNING "Field size %d bits not handled\n", size * 8); + return 0; + } +} + +/** + * ib_pack - Pack a structure into a buffer + * @desc:Array of structure field descriptions + * @desc_len:Number of entries in @desc + * @structure:Structure to pack from + * @buf:Buffer to pack into + * + * ib_pack() packs a list of structure fields into a buffer, + * controlled by the array of fields in @desc. + */ +void ib_pack(const struct ib_field *desc, + int desc_len, + void *structure, + void *buf) +{ + int i; + + for (i = 0; i < desc_len; ++i) { + if (desc[i].size_bits <= 32) { + int shift; + u32 val; + __be32 mask; + __be32 *addr; + + shift = 32 - desc[i].offset_bits - desc[i].size_bits; + if (desc[i].struct_size_bytes) + val = value_read(desc[i].struct_offset_bytes, + desc[i].struct_size_bytes, + structure) << shift; + else + val = 0; + + mask = cpu_to_be32(((1ull << desc[i].size_bits) - 1) << shift); + addr = (__be32 *) buf + desc[i].offset_words; + *addr = (*addr & ~mask) | (cpu_to_be32(val) & mask); + } else if (desc[i].size_bits <= 64) { + int shift; + u64 val; + __be64 mask; + __be64 *addr; + + shift = 64 - desc[i].offset_bits - desc[i].size_bits; + if (desc[i].struct_size_bytes) + val = value_read(desc[i].struct_offset_bytes, + desc[i].struct_size_bytes, + structure) << shift; + else + val = 0; + + mask = cpu_to_be64(((1ull << desc[i].size_bits) - 1) << shift); + addr = (__be64 *) ((__be32 *) buf + desc[i].offset_words); + *addr = (*addr & ~mask) | (cpu_to_be64(val) & mask); + } else { + if (desc[i].offset_bits % 8 || + desc[i].size_bits % 8) { + printk(KERN_WARNING "Structure field %s of size %d " + "bits is not byte-aligned\n", + desc[i].field_name, desc[i].size_bits); + } + + if (desc[i].struct_size_bytes) + memcpy(buf + desc[i].offset_words * 4 + + desc[i].offset_bits / 8, + structure + desc[i].struct_offset_bytes, + desc[i].size_bits / 8); + else + memset(buf + desc[i].offset_words * 4 + + desc[i].offset_bits / 8, + 0, + desc[i].size_bits / 8); + } + } +} +EXPORT_SYMBOL(ib_pack); + +static void value_write(int offset, int size, u64 val, void *structure) +{ + switch (size * 8) { + case 8: *( u8 *) (structure + offset) = val; break; + case 16: *(__be16 *) (structure + offset) = cpu_to_be16(val); break; + case 32: *(__be32 *) (structure + offset) = cpu_to_be32(val); break; + case 64: *(__be64 *) (structure + offset) = cpu_to_be64(val); break; + default: + printk(KERN_WARNING "Field size %d bits not handled\n", size * 8); + } +} + +/** + * ib_unpack - Unpack a buffer into a structure + * @desc:Array of structure field descriptions + * @desc_len:Number of entries in @desc + * @buf:Buffer to unpack from + * @structure:Structure to unpack into + * + * ib_pack() unpacks a list of structure fields from a buffer, + * controlled by the array of fields in @desc. + */ +void ib_unpack(const struct ib_field *desc, + int desc_len, + void *buf, + void *structure) +{ + int i; + + for (i = 0; i < desc_len; ++i) { + if (!desc[i].struct_size_bytes) + continue; + + if (desc[i].size_bits <= 32) { + int shift; + u32 val; + u32 mask; + __be32 *addr; + + shift = 32 - desc[i].offset_bits - desc[i].size_bits; + mask = ((1ull << desc[i].size_bits) - 1) << shift; + addr = (__be32 *) buf + desc[i].offset_words; + val = (be32_to_cpup(addr) & mask) >> shift; + value_write(desc[i].struct_offset_bytes, + desc[i].struct_size_bytes, + val, + structure); + } else if (desc[i].size_bits <= 64) { + int shift; + u64 val; + u64 mask; + __be64 *addr; + + shift = 64 - desc[i].offset_bits - desc[i].size_bits; + mask = ((1ull << desc[i].size_bits) - 1) << shift; + addr = (__be64 *) buf + desc[i].offset_words; + val = (be64_to_cpup(addr) & mask) >> shift; + value_write(desc[i].struct_offset_bytes, + desc[i].struct_size_bytes, + val, + structure); + } else { + if (desc[i].offset_bits % 8 || + desc[i].size_bits % 8) { + printk(KERN_WARNING "Structure field %s of size %d " + "bits is not byte-aligned\n", + desc[i].field_name, desc[i].size_bits); + } + + memcpy(structure + desc[i].struct_offset_bytes, + buf + desc[i].offset_words * 4 + + desc[i].offset_bits / 8, + desc[i].size_bits / 8); + } + } +} +EXPORT_SYMBOL(ib_unpack); diff --git a/drivers/infiniband/core/sa_query.c b/drivers/infiniband/core/sa_query.c new file mode 100644 index 000000000000..d4233ee61c35 --- /dev/null +++ b/drivers/infiniband/core/sa_query.c @@ -0,0 +1,866 @@ +/* + * Copyright (c) 2004 Topspin Communications. 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. + * + * $Id: sa_query.c 1389 2004-12-27 22:56:47Z roland $ + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/err.h> +#include <linux/random.h> +#include <linux/spinlock.h> +#include <linux/slab.h> +#include <linux/pci.h> +#include <linux/dma-mapping.h> +#include <linux/kref.h> +#include <linux/idr.h> + +#include <ib_pack.h> +#include <ib_sa.h> + +MODULE_AUTHOR("Roland Dreier"); +MODULE_DESCRIPTION("InfiniBand subnet administration query support"); +MODULE_LICENSE("Dual BSD/GPL"); + +/* + * These two structures must be packed because they have 64-bit fields + * that are only 32-bit aligned. 64-bit architectures will lay them + * out wrong otherwise. (And unfortunately they are sent on the wire + * so we can't change the layout) + */ +struct ib_sa_hdr { + u64 sm_key; + u16 attr_offset; + u16 reserved; + ib_sa_comp_mask comp_mask; +} __attribute__ ((packed)); + +struct ib_sa_mad { + struct ib_mad_hdr mad_hdr; + struct ib_rmpp_hdr rmpp_hdr; + struct ib_sa_hdr sa_hdr; + u8 data[200]; +} __attribute__ ((packed)); + +struct ib_sa_sm_ah { + struct ib_ah *ah; + struct kref ref; +}; + +struct ib_sa_port { + struct ib_mad_agent *agent; + struct ib_mr *mr; + struct ib_sa_sm_ah *sm_ah; + struct work_struct update_task; + spinlock_t ah_lock; + u8 port_num; +}; + +struct ib_sa_device { + int start_port, end_port; + struct ib_event_handler event_handler; + struct ib_sa_port port[0]; +}; + +struct ib_sa_query { + void (*callback)(struct ib_sa_query *, int, struct ib_sa_mad *); + void (*release)(struct ib_sa_query *); + struct ib_sa_port *port; + struct ib_sa_mad *mad; + struct ib_sa_sm_ah *sm_ah; + DECLARE_PCI_UNMAP_ADDR(mapping) + int id; +}; + +struct ib_sa_path_query { + void (*callback)(int, struct ib_sa_path_rec *, void *); + void *context; + struct ib_sa_query sa_query; +}; + +struct ib_sa_mcmember_query { + void (*callback)(int, struct ib_sa_mcmember_rec *, void *); + void *context; + struct ib_sa_query sa_query; +}; + +static void ib_sa_add_one(struct ib_device *device); +static void ib_sa_remove_one(struct ib_device *device); + +static struct ib_client sa_client = { + .name = "sa", + .add = ib_sa_add_one, + .remove = ib_sa_remove_one +}; + +static spinlock_t idr_lock; +static DEFINE_IDR(query_idr); + +static spinlock_t tid_lock; +static u32 tid; + +enum { + IB_SA_ATTR_CLASS_PORTINFO = 0x01, + IB_SA_ATTR_NOTICE = 0x02, + IB_SA_ATTR_INFORM_INFO = 0x03, + IB_SA_ATTR_NODE_REC = 0x11, + IB_SA_ATTR_PORT_INFO_REC = 0x12, + IB_SA_ATTR_SL2VL_REC = 0x13, + IB_SA_ATTR_SWITCH_REC = 0x14, + IB_SA_ATTR_LINEAR_FDB_REC = 0x15, + IB_SA_ATTR_RANDOM_FDB_REC = 0x16, + IB_SA_ATTR_MCAST_FDB_REC = 0x17, + IB_SA_ATTR_SM_INFO_REC = 0x18, + IB_SA_ATTR_LINK_REC = 0x20, + IB_SA_ATTR_GUID_INFO_REC = 0x30, + IB_SA_ATTR_SERVICE_REC = 0x31, + IB_SA_ATTR_PARTITION_REC = 0x33, + IB_SA_ATTR_RANGE_REC = 0x34, + IB_SA_ATTR_PATH_REC = 0x35, + IB_SA_ATTR_VL_ARB_REC = 0x36, + IB_SA_ATTR_MC_GROUP_REC = 0x37, + IB_SA_ATTR_MC_MEMBER_REC = 0x38, + IB_SA_ATTR_TRACE_REC = 0x39, + IB_SA_ATTR_MULTI_PATH_REC = 0x3a, + IB_SA_ATTR_SERVICE_ASSOC_REC = 0x3b +}; + +#define PATH_REC_FIELD(field) \ + .struct_offset_bytes = offsetof(struct ib_sa_path_rec, field), \ + .struct_size_bytes = sizeof ((struct ib_sa_path_rec *) 0)->field, \ + .field_name = "sa_path_rec:" #field + +static const struct ib_field path_rec_table[] = { + { RESERVED, + .offset_words = 0, + .offset_bits = 0, + .size_bits = 32 }, + { RESERVED, + .offset_words = 1, + .offset_bits = 0, + .size_bits = 32 }, + { PATH_REC_FIELD(dgid), + .offset_words = 2, + .offset_bits = 0, + .size_bits = 128 }, + { PATH_REC_FIELD(sgid), + .offset_words = 6, + .offset_bits = 0, + .size_bits = 128 }, + { PATH_REC_FIELD(dlid), + .offset_words = 10, + .offset_bits = 0, + .size_bits = 16 }, + { PATH_REC_FIELD(slid), + .offset_words = 10, + .offset_bits = 16, + .size_bits = 16 }, + { PATH_REC_FIELD(raw_traffic), + .offset_words = 11, + .offset_bits = 0, + .size_bits = 1 }, + { RESERVED, + .offset_words = 11, + .offset_bits = 1, + .size_bits = 3 }, + { PATH_REC_FIELD(flow_label), + .offset_words = 11, + .offset_bits = 4, + .size_bits = 20 }, + { PATH_REC_FIELD(hop_limit), + .offset_words = 11, + .offset_bits = 24, + .size_bits = 8 }, + { PATH_REC_FIELD(traffic_class), + .offset_words = 12, + .offset_bits = 0, + .size_bits = 8 }, + { PATH_REC_FIELD(reversible), + .offset_words = 12, + .offset_bits = 8, + .size_bits = 1 }, + { PATH_REC_FIELD(numb_path), + .offset_words = 12, + .offset_bits = 9, + .size_bits = 7 }, + { PATH_REC_FIELD(pkey), + .offset_words = 12, + .offset_bits = 16, + .size_bits = 16 }, + { RESERVED, + .offset_words = 13, + .offset_bits = 0, + .size_bits = 12 }, + { PATH_REC_FIELD(sl), + .offset_words = 13, + .offset_bits = 12, + .size_bits = 4 }, + { PATH_REC_FIELD(mtu_selector), + .offset_words = 13, + .offset_bits = 16, + .size_bits = 2 }, + { PATH_REC_FIELD(mtu), + .offset_words = 13, + .offset_bits = 18, + .size_bits = 6 }, + { PATH_REC_FIELD(rate_selector), + .offset_words = 13, + .offset_bits = 24, + .size_bits = 2 }, + { PATH_REC_FIELD(rate), + .offset_words = 13, + .offset_bits = 26, + .size_bits = 6 }, + { PATH_REC_FIELD(packet_life_time_selector), + .offset_words = 14, + .offset_bits = 0, + .size_bits = 2 }, + { PATH_REC_FIELD(packet_life_time), + .offset_words = 14, + .offset_bits = 2, + .size_bits = 6 }, + { PATH_REC_FIELD(preference), + .offset_words = 14, + .offset_bits = 8, + .size_bits = 8 }, + { RESERVED, + .offset_words = 14, + .offset_bits = 16, + .size_bits = 48 }, +}; + +#define MCMEMBER_REC_FIELD(field) \ + .struct_offset_bytes = offsetof(struct ib_sa_mcmember_rec, field), \ + .struct_size_bytes = sizeof ((struct ib_sa_mcmember_rec *) 0)->field, \ + .field_name = "sa_mcmember_rec:" #field + +static const struct ib_field mcmember_rec_table[] = { + { MCMEMBER_REC_FIELD(mgid), + .offset_words = 0, + .offset_bits = 0, + .size_bits = 128 }, + { MCMEMBER_REC_FIELD(port_gid), + .offset_words = 4, + .offset_bits = 0, + .size_bits = 128 }, + { MCMEMBER_REC_FIELD(qkey), + .offset_words = 8, + .offset_bits = 0, + .size_bits = 32 }, + { MCMEMBER_REC_FIELD(mlid), + .offset_words = 9, + .offset_bits = 0, + .size_bits = 16 }, + { MCMEMBER_REC_FIELD(mtu_selector), + .offset_words = 9, + .offset_bits = 16, + .size_bits = 2 }, + { MCMEMBER_REC_FIELD(mtu), + .offset_words = 9, + .offset_bits = 18, + .size_bits = 6 }, + { MCMEMBER_REC_FIELD(traffic_class), + .offset_words = 9, + .offset_bits = 24, + .size_bits = 8 }, + { MCMEMBER_REC_FIELD(pkey), + .offset_words = 10, + .offset_bits = 0, + .size_bits = 16 }, + { MCMEMBER_REC_FIELD(rate_selector), + .offset_words = 10, + .offset_bits = 16, + .size_bits = 2 }, + { MCMEMBER_REC_FIELD(rate), + .offset_words = 10, + .offset_bits = 18, + .size_bits = 6 }, + { MCMEMBER_REC_FIELD(packet_life_time_selector), + .offset_words = 10, + .offset_bits = 24, + .size_bits = 2 }, + { MCMEMBER_REC_FIELD(packet_life_time), + .offset_words = 10, + .offset_bits = 26, + .size_bits = 6 }, + { MCMEMBER_REC_FIELD(sl), + .offset_words = 11, + .offset_bits = 0, + .size_bits = 4 }, + { MCMEMBER_REC_FIELD(flow_label), + .offset_words = 11, + .offset_bits = 4, + .size_bits = 20 }, + { MCMEMBER_REC_FIELD(hop_limit), + .offset_words = 11, + .offset_bits = 24, + .size_bits = 8 }, + { MCMEMBER_REC_FIELD(scope), + .offset_words = 12, + .offset_bits = 0, + .size_bits = 4 }, + { MCMEMBER_REC_FIELD(join_state), + .offset_words = 12, + .offset_bits = 4, + .size_bits = 4 }, + { MCMEMBER_REC_FIELD(proxy_join), + .offset_words = 12, + .offset_bits = 8, + .size_bits = 1 }, + { RESERVED, + .offset_words = 12, + .offset_bits = 9, + .size_bits = 23 }, +}; + +static void free_sm_ah(struct kref *kref) +{ + struct ib_sa_sm_ah *sm_ah = container_of(kref, struct ib_sa_sm_ah, ref); + + ib_destroy_ah(sm_ah->ah); + kfree(sm_ah); +} + +static void update_sm_ah(void *port_ptr) +{ + struct ib_sa_port *port = port_ptr; + struct ib_sa_sm_ah *new_ah, *old_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)) { + printk(KERN_WARNING "Couldn't query port\n"); + return; + } + + new_ah = kmalloc(sizeof *new_ah, GFP_KERNEL); + if (!new_ah) { + printk(KERN_WARNING "Couldn't allocate new SM AH\n"); + return; + } + + kref_init(&new_ah->ref); + + 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; + + new_ah->ah = ib_create_ah(port->agent->qp->pd, &ah_attr); + if (IS_ERR(new_ah->ah)) { + printk(KERN_WARNING "Couldn't create new SM AH\n"); + kfree(new_ah); + return; + } + + spin_lock_irq(&port->ah_lock); + old_ah = port->sm_ah; + port->sm_ah = new_ah; + spin_unlock_irq(&port->ah_lock); + + if (old_ah) + kref_put(&old_ah->ref, free_sm_ah); +} + +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) { + struct ib_sa_device *sa_dev = + ib_get_client_data(event->device, &sa_client); + + schedule_work(&sa_dev->port[event->element.port_num - + sa_dev->start_port].update_task); + } +} + +/** + * ib_sa_cancel_query - try to cancel an SA query + * @id:ID of query to cancel + * @query:query pointer to cancel + * + * Try to cancel an SA query. If the id and query don't match up or + * the query has already completed, nothing is done. Otherwise the + * query is canceled and will complete with a status of -EINTR. + */ +void ib_sa_cancel_query(int id, struct ib_sa_query *query) +{ + unsigned long flags; + struct ib_mad_agent *agent; + + spin_lock_irqsave(&idr_lock, flags); + if (idr_find(&query_idr, id) != query) { + spin_unlock_irqrestore(&idr_lock, flags); + return; + } + agent = query->port->agent; + spin_unlock_irqrestore(&idr_lock, flags); + + ib_cancel_mad(agent, id); +} +EXPORT_SYMBOL(ib_sa_cancel_query); + +static void init_mad(struct ib_sa_mad *mad, struct ib_mad_agent *agent) +{ + unsigned long flags; + + memset(mad, 0, sizeof *mad); + + mad->mad_hdr.base_version = IB_MGMT_BASE_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++); + spin_unlock_irqrestore(&tid_lock, flags); +} + +static int send_mad(struct ib_sa_query *query, int timeout_ms) +{ + struct ib_sa_port *port = query->port; + unsigned long flags; + int ret; + struct ib_sge gather_list; + struct ib_send_wr *bad_wr, wr = { + .opcode = IB_WR_SEND, + .sg_list = &gather_list, + .num_sge = 1, + .send_flags = IB_SEND_SIGNALED, + .wr = { + .ud = { + .mad_hdr = &query->mad->mad_hdr, + .remote_qpn = 1, + .remote_qkey = IB_QP1_QKEY, + .timeout_ms = timeout_ms + } + } + }; + +retry: + if (!idr_pre_get(&query_idr, GFP_ATOMIC)) + return -ENOMEM; + spin_lock_irqsave(&idr_lock, flags); + ret = idr_get_new(&query_idr, query, &query->id); + spin_unlock_irqrestore(&idr_lock, flags); + if (ret == -EAGAIN) + goto retry; + if (ret) + return ret; + + wr.wr_id = query->id; + + spin_lock_irqsave(&port->ah_lock, flags); + kref_get(&port->sm_ah->ref); + query->sm_ah = port->sm_ah; + wr.wr.ud.ah = port->sm_ah->ah; + spin_unlock_irqrestore(&port->ah_lock, flags); + + gather_list.addr = dma_map_single(port->agent->device->dma_device, + query->mad, + sizeof (struct ib_sa_mad), + DMA_TO_DEVICE); + gather_list.length = sizeof (struct ib_sa_mad); + gather_list.lkey = port->mr->lkey; + pci_unmap_addr_set(query, mapping, gather_list.addr); + + ret = ib_post_send_mad(port->agent, &wr, &bad_wr); + if (ret) { + dma_unmap_single(port->agent->device->dma_device, + pci_unmap_addr(query, mapping), + sizeof (struct ib_sa_mad), + DMA_TO_DEVICE); + kref_put(&query->sm_ah->ref, free_sm_ah); + spin_lock_irqsave(&idr_lock, flags); + idr_remove(&query_idr, query->id); + spin_unlock_irqrestore(&idr_lock, flags); + } + + return ret; +} + +static void ib_sa_path_rec_callback(struct ib_sa_query *sa_query, + int status, + struct ib_sa_mad *mad) +{ + struct ib_sa_path_query *query = + container_of(sa_query, struct ib_sa_path_query, sa_query); + + if (mad) { + struct ib_sa_path_rec rec; + + ib_unpack(path_rec_table, ARRAY_SIZE(path_rec_table), + mad->data, &rec); + query->callback(status, &rec, query->context); + } else + query->callback(status, NULL, query->context); +} + +static void ib_sa_path_rec_release(struct ib_sa_query *sa_query) +{ + kfree(sa_query->mad); + kfree(container_of(sa_query, struct ib_sa_path_query, sa_query)); +} + +/** + * ib_sa_path_rec_get - Start a Path get query + * @device:device to send query on + * @port_num: port number to send query on + * @rec:Path Record to send in query + * @comp_mask:component mask to send in query + * @timeout_ms:time to wait for response + * @gfp_mask:GFP mask to use for internal allocations + * @callback:function called when query completes, times out or is + * canceled + * @context:opaque user context passed to callback + * @sa_query:query context, used to cancel query + * + * Send a Path Record Get query to the SA to look up a path. The + * callback function will be called when the query completes (or + * fails); status is 0 for a successful response, -EINTR if the query + * is canceled, -ETIMEDOUT is the query timed out, or -EIO if an error + * occurred sending the query. The resp parameter of the callback is + * only valid if status is 0. + * + * If the return value of ib_sa_path_rec_get() is negative, it is an + * error code. Otherwise it is a query ID that can be used to cancel + * the query. + */ +int ib_sa_path_rec_get(struct ib_device *device, u8 port_num, + struct ib_sa_path_rec *rec, + ib_sa_comp_mask comp_mask, + int timeout_ms, int gfp_mask, + void (*callback)(int status, + struct ib_sa_path_rec *resp, + void *context), + void *context, + struct ib_sa_query **sa_query) +{ + struct ib_sa_path_query *query; + struct ib_sa_device *sa_dev = ib_get_client_data(device, &sa_client); + struct ib_sa_port *port = &sa_dev->port[port_num - sa_dev->start_port]; + struct ib_mad_agent *agent = port->agent; + int ret; + + query = kmalloc(sizeof *query, gfp_mask); + if (!query) + return -ENOMEM; + query->sa_query.mad = kmalloc(sizeof *query->sa_query.mad, gfp_mask); + if (!query->sa_query.mad) { + kfree(query); + return -ENOMEM; + } + + query->callback = callback; + query->context = context; + + init_mad(query->sa_query.mad, agent); + + query->sa_query.callback = ib_sa_path_rec_callback; + query->sa_query.release = ib_sa_path_rec_release; + query->sa_query.port = port; + query->sa_query.mad->mad_hdr.method = IB_MGMT_METHOD_GET; + query->sa_query.mad->mad_hdr.attr_id = cpu_to_be16(IB_SA_ATTR_PATH_REC); + query->sa_query.mad->sa_hdr.comp_mask = comp_mask; + + ib_pack(path_rec_table, ARRAY_SIZE(path_rec_table), + rec, query->sa_query.mad->data); + + *sa_query = &query->sa_query; + ret = send_mad(&query->sa_query, timeout_ms); + if (ret) { + *sa_query = NULL; + kfree(query->sa_query.mad); + kfree(query); + } + + return ret ? ret : query->sa_query.id; +} +EXPORT_SYMBOL(ib_sa_path_rec_get); + +static void ib_sa_mcmember_rec_callback(struct ib_sa_query *sa_query, + int status, + struct ib_sa_mad *mad) +{ + struct ib_sa_mcmember_query *query = + container_of(sa_query, struct ib_sa_mcmember_query, sa_query); + + if (mad) { + struct ib_sa_mcmember_rec rec; + + ib_unpack(mcmember_rec_table, ARRAY_SIZE(mcmember_rec_table), + mad->data, &rec); + query->callback(status, &rec, query->context); + } else + query->callback(status, NULL, query->context); +} + +static void ib_sa_mcmember_rec_release(struct ib_sa_query *sa_query) +{ + kfree(sa_query->mad); + kfree(container_of(sa_query, struct ib_sa_mcmember_query, sa_query)); +} + +int ib_sa_mcmember_rec_query(struct ib_device *device, u8 port_num, + u8 method, + struct ib_sa_mcmember_rec *rec, + ib_sa_comp_mask comp_mask, + int timeout_ms, int gfp_mask, + void (*callback)(int status, + struct ib_sa_mcmember_rec *resp, + void *context), + void *context, + struct ib_sa_query **sa_query) +{ + struct ib_sa_mcmember_query *query; + struct ib_sa_device *sa_dev = ib_get_client_data(device, &sa_client); + struct ib_sa_port *port = &sa_dev->port[port_num - sa_dev->start_port]; + struct ib_mad_agent *agent = port->agent; + int ret; + + query = kmalloc(sizeof *query, gfp_mask); + if (!query) + return -ENOMEM; + query->sa_query.mad = kmalloc(sizeof *query->sa_query.mad, gfp_mask); + if (!query->sa_query.mad) { + kfree(query); + return -ENOMEM; + } + + query->callback = callback; + query->context = context; + + init_mad(query->sa_query.mad, agent); + + query->sa_query.callback = ib_sa_mcmember_rec_callback; + query->sa_query.release = ib_sa_mcmember_rec_release; + query->sa_query.port = port; + query->sa_query.mad->mad_hdr.method = method; + query->sa_query.mad->mad_hdr.attr_id = cpu_to_be16(IB_SA_ATTR_MC_MEMBER_REC); + query->sa_query.mad->sa_hdr.comp_mask = comp_mask; + + ib_pack(mcmember_rec_table, ARRAY_SIZE(mcmember_rec_table), + rec, query->sa_query.mad->data); + + *sa_query = &query->sa_query; + ret = send_mad(&query->sa_query, timeout_ms); + if (ret) { + *sa_query = NULL; + kfree(query->sa_query.mad); + kfree(query); + } + + return ret ? ret : query->sa_query.id; +} +EXPORT_SYMBOL(ib_sa_mcmember_rec_query); + +static void send_handler(struct ib_mad_agent *agent, + struct ib_mad_send_wc *mad_send_wc) +{ + struct ib_sa_query *query; + unsigned long flags; + + spin_lock_irqsave(&idr_lock, flags); + query = idr_find(&query_idr, mad_send_wc->wr_id); + spin_unlock_irqrestore(&idr_lock, flags); + + if (!query) + return; + + switch (mad_send_wc->status) { + case IB_WC_SUCCESS: + /* No callback -- already got recv */ + break; + case IB_WC_RESP_TIMEOUT_ERR: + query->callback(query, -ETIMEDOUT, NULL); + break; + case IB_WC_WR_FLUSH_ERR: + query->callback(query, -EINTR, NULL); + break; + default: + query->callback(query, -EIO, NULL); + break; + } + + dma_unmap_single(agent->device->dma_device, + pci_unmap_addr(query, mapping), + sizeof (struct ib_sa_mad), + DMA_TO_DEVICE); + kref_put(&query->sm_ah->ref, free_sm_ah); + + query->release(query); + + spin_lock_irqsave(&idr_lock, flags); + idr_remove(&query_idr, mad_send_wc->wr_id); + spin_unlock_irqrestore(&idr_lock, flags); +} + +static void recv_handler(struct ib_mad_agent *mad_agent, + struct ib_mad_recv_wc *mad_recv_wc) +{ + struct ib_sa_query *query; + unsigned long flags; + + spin_lock_irqsave(&idr_lock, flags); + query = idr_find(&query_idr, mad_recv_wc->wc->wr_id); + spin_unlock_irqrestore(&idr_lock, flags); + + if (query) { + if (mad_recv_wc->wc->status == IB_WC_SUCCESS) + query->callback(query, + mad_recv_wc->recv_buf.mad->mad_hdr.status ? + -EINVAL : 0, + (struct ib_sa_mad *) mad_recv_wc->recv_buf.mad); + else + query->callback(query, -EIO, NULL); + } + + ib_free_recv_mad(mad_recv_wc); +} + +static void ib_sa_add_one(struct ib_device *device) +{ + struct ib_sa_device *sa_dev; + int s, e, i; + + if (device->node_type == IB_NODE_SWITCH) + s = e = 0; + else { + s = 1; + e = device->phys_port_cnt; + } + + sa_dev = kmalloc(sizeof *sa_dev + + (e - s + 1) * sizeof (struct ib_sa_port), + GFP_KERNEL); + if (!sa_dev) + return; + + sa_dev->start_port = s; + sa_dev->end_port = e; + + for (i = 0; i <= e - s; ++i) { + sa_dev->port[i].mr = NULL; + sa_dev->port[i].sm_ah = NULL; + sa_dev->port[i].port_num = i + s; + spin_lock_init(&sa_dev->port[i].ah_lock); + + sa_dev->port[i].agent = + ib_register_mad_agent(device, i + s, IB_QPT_GSI, + NULL, 0, send_handler, + recv_handler, sa_dev); + if (IS_ERR(sa_dev->port[i].agent)) + goto err; + + sa_dev->port[i].mr = ib_get_dma_mr(sa_dev->port[i].agent->qp->pd, + IB_ACCESS_LOCAL_WRITE); + if (IS_ERR(sa_dev->port[i].mr)) { + ib_unregister_mad_agent(sa_dev->port[i].agent); + goto err; + } + + INIT_WORK(&sa_dev->port[i].update_task, + update_sm_ah, &sa_dev->port[i]); + } + + ib_set_client_data(device, &sa_client, sa_dev); + + /* + * We register our event handler after everything is set up, + * and then update our cached info after the event handler is + * registered to avoid any problems if a port changes state + * during our initialization. + */ + + INIT_IB_EVENT_HANDLER(&sa_dev->event_handler, device, ib_sa_event); + if (ib_register_event_handler(&sa_dev->event_handler)) + goto err; + + for (i = 0; i <= e - s; ++i) + update_sm_ah(&sa_dev->port[i]); + + return; + +err: + while (--i >= 0) { + ib_dereg_mr(sa_dev->port[i].mr); + ib_unregister_mad_agent(sa_dev->port[i].agent); + } + + kfree(sa_dev); + + return; +} + +static void ib_sa_remove_one(struct ib_device *device) +{ + struct ib_sa_device *sa_dev = ib_get_client_data(device, &sa_client); + int i; + + if (!sa_dev) + return; + + ib_unregister_event_handler(&sa_dev->event_handler); + + for (i = 0; i <= sa_dev->end_port - sa_dev->start_port; ++i) { + ib_unregister_mad_agent(sa_dev->port[i].agent); + kref_put(&sa_dev->port[i].sm_ah->ref, free_sm_ah); + } + + kfree(sa_dev); +} + +static int __init ib_sa_init(void) +{ + int ret; + + spin_lock_init(&idr_lock); + spin_lock_init(&tid_lock); + + get_random_bytes(&tid, sizeof tid); + + ret = ib_register_client(&sa_client); + if (ret) + printk(KERN_ERR "Couldn't register ib_sa client\n"); + + return ret; +} + +static void __exit ib_sa_cleanup(void) +{ + ib_unregister_client(&sa_client); +} + +module_init(ib_sa_init); +module_exit(ib_sa_cleanup); diff --git a/drivers/infiniband/core/smi.c b/drivers/infiniband/core/smi.c new file mode 100644 index 000000000000..b4b284324a33 --- /dev/null +++ b/drivers/infiniband/core/smi.c @@ -0,0 +1,234 @@ +/* + * Copyright (c) 2004 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2004 Infinicon Corporation. All rights reserved. + * Copyright (c) 2004 Intel Corporation. All rights reserved. + * Copyright (c) 2004 Topspin Corporation. All rights reserved. + * Copyright (c) 2004 Voltaire Corporation. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + * + * $Id: smi.c 1389 2004-12-27 22:56:47Z roland $ + */ + +#include <ib_smi.h> +#include "smi.h" + +/* + * Fixup a directed route SMP for sending + * Return 0 if the SMP should be discarded + */ +int smi_handle_dr_smp_send(struct ib_smp *smp, + u8 node_type, + int port_num) +{ + u8 hop_ptr, hop_cnt; + + hop_ptr = smp->hop_ptr; + hop_cnt = smp->hop_cnt; + + /* See section 14.2.2.2, Vol 1 IB spec */ + if (!ib_get_smp_direction(smp)) { + /* C14-9:1 */ + if (hop_cnt && hop_ptr == 0) { + smp->hop_ptr++; + return (smp->initial_path[smp->hop_ptr] == + port_num); + } + + /* C14-9:2 */ + if (hop_ptr && hop_ptr < hop_cnt) { + if (node_type != IB_NODE_SWITCH) + return 0; + + /* smp->return_path set when received */ + smp->hop_ptr++; + return (smp->initial_path[smp->hop_ptr] == + port_num); + } + + /* C14-9:3 -- We're at the end of the DR segment of path */ + if (hop_ptr == hop_cnt) { + /* smp->return_path set when received */ + smp->hop_ptr++; + return (node_type == IB_NODE_SWITCH || + smp->dr_dlid == IB_LID_PERMISSIVE); + } + + /* C14-9:4 -- hop_ptr = hop_cnt + 1 -> give to SMA/SM */ + /* C14-9:5 -- Fail unreasonable hop pointer */ + return (hop_ptr == hop_cnt + 1); + + } else { + /* C14-13:1 */ + if (hop_cnt && hop_ptr == hop_cnt + 1) { + smp->hop_ptr--; + return (smp->return_path[smp->hop_ptr] == + port_num); + } + + /* C14-13:2 */ + if (2 <= hop_ptr && hop_ptr <= hop_cnt) { + if (node_type != IB_NODE_SWITCH) + return 0; + + smp->hop_ptr--; + return (smp->return_path[smp->hop_ptr] == + port_num); + } + + /* C14-13:3 -- at the end of the DR segment of path */ + if (hop_ptr == 1) { + smp->hop_ptr--; + /* C14-13:3 -- SMPs destined for SM shouldn't be here */ + return (node_type == IB_NODE_SWITCH || + smp->dr_slid == IB_LID_PERMISSIVE); + } + + /* C14-13:4 -- hop_ptr = 0 -> should have gone to SM */ + if (hop_ptr == 0) + return 1; + + /* C14-13:5 -- Check for unreasonable hop pointer */ + return 0; + } +} + +/* + * Adjust information for a received SMP + * Return 0 if the SMP should be dropped + */ +int smi_handle_dr_smp_recv(struct ib_smp *smp, + u8 node_type, + int port_num, + int phys_port_cnt) +{ + u8 hop_ptr, hop_cnt; + + hop_ptr = smp->hop_ptr; + hop_cnt = smp->hop_cnt; + + /* See section 14.2.2.2, Vol 1 IB spec */ + if (!ib_get_smp_direction(smp)) { + /* C14-9:1 -- sender should have incremented hop_ptr */ + if (hop_cnt && hop_ptr == 0) + return 0; + + /* C14-9:2 -- intermediate hop */ + if (hop_ptr && hop_ptr < hop_cnt) { + if (node_type != IB_NODE_SWITCH) + return 0; + + smp->return_path[hop_ptr] = port_num; + /* smp->hop_ptr updated when sending */ + return (smp->initial_path[hop_ptr+1] <= phys_port_cnt); + } + + /* C14-9:3 -- We're at the end of the DR segment of path */ + if (hop_ptr == hop_cnt) { + if (hop_cnt) + smp->return_path[hop_ptr] = port_num; + /* smp->hop_ptr updated when sending */ + + return (node_type == IB_NODE_SWITCH || + smp->dr_dlid == IB_LID_PERMISSIVE); + } + + /* C14-9:4 -- hop_ptr = hop_cnt + 1 -> give to SMA/SM */ + /* C14-9:5 -- fail unreasonable hop pointer */ + return (hop_ptr == hop_cnt + 1); + + } else { + + /* C14-13:1 */ + if (hop_cnt && hop_ptr == hop_cnt + 1) { + smp->hop_ptr--; + return (smp->return_path[smp->hop_ptr] == + port_num); + } + + /* C14-13:2 */ + if (2 <= hop_ptr && hop_ptr <= hop_cnt) { + if (node_type != IB_NODE_SWITCH) + return 0; + + /* smp->hop_ptr updated when sending */ + return (smp->return_path[hop_ptr-1] <= phys_port_cnt); + } + + /* C14-13:3 -- We're at the end of the DR segment of path */ + if (hop_ptr == 1) { + if (smp->dr_slid == IB_LID_PERMISSIVE) { + /* giving SMP to SM - update hop_ptr */ + smp->hop_ptr--; + return 1; + } + /* smp->hop_ptr updated when sending */ + return (node_type == IB_NODE_SWITCH); + } + + /* C14-13:4 -- hop_ptr = 0 -> give to SM */ + /* C14-13:5 -- Check for unreasonable hop pointer */ + return (hop_ptr == 0); + } +} + +/* + * Return 1 if the received DR SMP should be forwarded to the send queue + * Return 0 if the SMP should be completed up the stack + */ +int smi_check_forward_dr_smp(struct ib_smp *smp) +{ + u8 hop_ptr, hop_cnt; + + hop_ptr = smp->hop_ptr; + hop_cnt = smp->hop_cnt; + + if (!ib_get_smp_direction(smp)) { + /* C14-9:2 -- intermediate hop */ + if (hop_ptr && hop_ptr < hop_cnt) + return 1; + + /* C14-9:3 -- at the end of the DR segment of path */ + if (hop_ptr == hop_cnt) + return (smp->dr_dlid == IB_LID_PERMISSIVE); + + /* C14-9:4 -- hop_ptr = hop_cnt + 1 -> give to SMA/SM */ + if (hop_ptr == hop_cnt + 1) + return 1; + } else { + /* C14-13:2 */ + if (2 <= hop_ptr && hop_ptr <= hop_cnt) + return 1; + + /* C14-13:3 -- at the end of the DR segment of path */ + if (hop_ptr == 1) + return (smp->dr_slid != IB_LID_PERMISSIVE); + } + return 0; +} diff --git a/drivers/infiniband/core/smi.h b/drivers/infiniband/core/smi.h new file mode 100644 index 000000000000..db25503a0736 --- /dev/null +++ b/drivers/infiniband/core/smi.h @@ -0,0 +1,67 @@ +/* + * Copyright (c) 2004 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2004 Infinicon Corporation. All rights reserved. + * Copyright (c) 2004 Intel Corporation. All rights reserved. + * Copyright (c) 2004 Topspin Corporation. All rights reserved. + * Copyright (c) 2004 Voltaire Corporation. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + * + * $Id: smi.h 1389 2004-12-27 22:56:47Z roland $ + */ + +#ifndef __SMI_H_ +#define __SMI_H_ + +int smi_handle_dr_smp_recv(struct ib_smp *smp, + u8 node_type, + int port_num, + int phys_port_cnt); +extern int smi_check_forward_dr_smp(struct ib_smp *smp); +extern int smi_handle_dr_smp_send(struct ib_smp *smp, + u8 node_type, + int port_num); +extern int smi_check_local_dr_smp(struct ib_smp *smp, + struct ib_device *device, + int port_num); + +/* + * Return 1 if the SMP should be handled by the local SMA/SM via process_mad + */ +static inline int smi_check_local_smp(struct ib_mad_agent *mad_agent, + struct ib_smp *smp) +{ + /* C14-9:3 -- We're at the end of the DR segment of path */ + /* C14-9:4 -- Hop Pointer = Hop Count + 1 -> give to SMA/SM */ + return ((mad_agent->device->process_mad && + !ib_get_smp_direction(smp) && + (smp->hop_ptr == smp->hop_cnt + 1))); +} + +#endif /* __SMI_H_ */ diff --git a/drivers/infiniband/core/sysfs.c b/drivers/infiniband/core/sysfs.c new file mode 100644 index 000000000000..3a413f72ff6d --- /dev/null +++ b/drivers/infiniband/core/sysfs.c @@ -0,0 +1,762 @@ +/* + * Copyright (c) 2004, 2005 Topspin Communications. 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. + * + * $Id: sysfs.c 1349 2004-12-16 21:09:43Z roland $ + */ + +#include "core_priv.h" + +#include <ib_mad.h> + +struct ib_port { + struct kobject kobj; + struct ib_device *ibdev; + struct attribute_group gid_group; + struct attribute **gid_attr; + struct attribute_group pkey_group; + struct attribute **pkey_attr; + u8 port_num; +}; + +struct port_attribute { + struct attribute attr; + ssize_t (*show)(struct ib_port *, struct port_attribute *, char *buf); + ssize_t (*store)(struct ib_port *, struct port_attribute *, + const char *buf, size_t count); +}; + +#define PORT_ATTR(_name, _mode, _show, _store) \ +struct port_attribute port_attr_##_name = __ATTR(_name, _mode, _show, _store) + +#define PORT_ATTR_RO(_name) \ +struct port_attribute port_attr_##_name = __ATTR_RO(_name) + +struct port_table_attribute { + struct port_attribute attr; + int index; +}; + +static ssize_t port_attr_show(struct kobject *kobj, + struct attribute *attr, char *buf) +{ + struct port_attribute *port_attr = + container_of(attr, struct port_attribute, attr); + struct ib_port *p = container_of(kobj, struct ib_port, kobj); + + if (!port_attr->show) + return 0; + + return port_attr->show(p, port_attr, buf); +} + +static struct sysfs_ops port_sysfs_ops = { + .show = port_attr_show +}; + +static ssize_t state_show(struct ib_port *p, struct port_attribute *unused, + char *buf) +{ + struct ib_port_attr attr; + ssize_t ret; + + static const char *state_name[] = { + [IB_PORT_NOP] = "NOP", + [IB_PORT_DOWN] = "DOWN", + [IB_PORT_INIT] = "INIT", + [IB_PORT_ARMED] = "ARMED", + [IB_PORT_ACTIVE] = "ACTIVE", + [IB_PORT_ACTIVE_DEFER] = "ACTIVE_DEFER" + }; + + ret = ib_query_port(p->ibdev, p->port_num, &attr); + if (ret) + return ret; + + return sprintf(buf, "%d: %s\n", attr.state, + attr.state >= 0 && attr.state <= ARRAY_SIZE(state_name) ? + state_name[attr.state] : "UNKNOWN"); +} + +static ssize_t lid_show(struct ib_port *p, struct port_attribute *unused, + char *buf) +{ + struct ib_port_attr attr; + ssize_t ret; + + ret = ib_query_port(p->ibdev, p->port_num, &attr); + if (ret) + return ret; + + return sprintf(buf, "0x%x\n", attr.lid); +} + +static ssize_t lid_mask_count_show(struct ib_port *p, + struct port_attribute *unused, + char *buf) +{ + struct ib_port_attr attr; + ssize_t ret; + + ret = ib_query_port(p->ibdev, p->port_num, &attr); + if (ret) + return ret; + + return sprintf(buf, "%d\n", attr.lmc); +} + +static ssize_t sm_lid_show(struct ib_port *p, struct port_attribute *unused, + char *buf) +{ + struct ib_port_attr attr; + ssize_t ret; + + ret = ib_query_port(p->ibdev, p->port_num, &attr); + if (ret) + return ret; + + return sprintf(buf, "0x%x\n", attr.sm_lid); +} + +static ssize_t sm_sl_show(struct ib_port *p, struct port_attribute *unused, + char *buf) +{ + struct ib_port_attr attr; + ssize_t ret; + + ret = ib_query_port(p->ibdev, p->port_num, &attr); + if (ret) + return ret; + + return sprintf(buf, "%d\n", attr.sm_sl); +} + +static ssize_t cap_mask_show(struct ib_port *p, struct port_attribute *unused, + char *buf) +{ + struct ib_port_attr attr; + ssize_t ret; + + ret = ib_query_port(p->ibdev, p->port_num, &attr); + if (ret) + return ret; + + return sprintf(buf, "0x%08x\n", attr.port_cap_flags); +} + +static ssize_t rate_show(struct ib_port *p, struct port_attribute *unused, + char *buf) +{ + struct ib_port_attr attr; + char *speed = ""; + int rate; + ssize_t ret; + + ret = ib_query_port(p->ibdev, p->port_num, &attr); + if (ret) + return ret; + + switch (attr.active_speed) { + case 2: speed = " DDR"; break; + case 4: speed = " QDR"; break; + } + + rate = 25 * ib_width_enum_to_int(attr.active_width) * attr.active_speed; + if (rate < 0) + return -EINVAL; + + return sprintf(buf, "%d%s Gb/sec (%dX%s)\n", + rate / 10, rate % 10 ? ".5" : "", + ib_width_enum_to_int(attr.active_width), speed); +} + +static ssize_t phys_state_show(struct ib_port *p, struct port_attribute *unused, + char *buf) +{ + struct ib_port_attr attr; + + ssize_t ret; + + ret = ib_query_port(p->ibdev, p->port_num, &attr); + if (ret) + return ret; + + switch (attr.phys_state) { + case 1: return sprintf(buf, "1: Sleep\n"); + case 2: return sprintf(buf, "2: Polling\n"); + case 3: return sprintf(buf, "3: Disabled\n"); + case 4: return sprintf(buf, "4: PortConfigurationTraining\n"); + case 5: return sprintf(buf, "5: LinkUp\n"); + case 6: return sprintf(buf, "6: LinkErrorRecovery\n"); + case 7: return sprintf(buf, "7: Phy Test\n"); + default: return sprintf(buf, "%d: <unknown>\n", attr.phys_state); + } +} + +static PORT_ATTR_RO(state); +static PORT_ATTR_RO(lid); +static PORT_ATTR_RO(lid_mask_count); +static PORT_ATTR_RO(sm_lid); +static PORT_ATTR_RO(sm_sl); +static PORT_ATTR_RO(cap_mask); +static PORT_ATTR_RO(rate); +static PORT_ATTR_RO(phys_state); + +static struct attribute *port_default_attrs[] = { + &port_attr_state.attr, + &port_attr_lid.attr, + &port_attr_lid_mask_count.attr, + &port_attr_sm_lid.attr, + &port_attr_sm_sl.attr, + &port_attr_cap_mask.attr, + &port_attr_rate.attr, + &port_attr_phys_state.attr, + NULL +}; + +static ssize_t show_port_gid(struct ib_port *p, struct port_attribute *attr, + char *buf) +{ + struct port_table_attribute *tab_attr = + container_of(attr, struct port_table_attribute, attr); + union ib_gid gid; + ssize_t ret; + + ret = ib_query_gid(p->ibdev, p->port_num, tab_attr->index, &gid); + if (ret) + return ret; + + return sprintf(buf, "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + be16_to_cpu(((u16 *) gid.raw)[0]), + be16_to_cpu(((u16 *) gid.raw)[1]), + be16_to_cpu(((u16 *) gid.raw)[2]), + be16_to_cpu(((u16 *) gid.raw)[3]), + be16_to_cpu(((u16 *) gid.raw)[4]), + be16_to_cpu(((u16 *) gid.raw)[5]), + be16_to_cpu(((u16 *) gid.raw)[6]), + be16_to_cpu(((u16 *) gid.raw)[7])); +} + +static ssize_t show_port_pkey(struct ib_port *p, struct port_attribute *attr, + char *buf) +{ + struct port_table_attribute *tab_attr = + container_of(attr, struct port_table_attribute, attr); + u16 pkey; + ssize_t ret; + + ret = ib_query_pkey(p->ibdev, p->port_num, tab_attr->index, &pkey); + if (ret) + return ret; + + return sprintf(buf, "0x%04x\n", pkey); +} + +#define PORT_PMA_ATTR(_name, _counter, _width, _offset) \ +struct port_table_attribute port_pma_attr_##_name = { \ + .attr = __ATTR(_name, S_IRUGO, show_pma_counter, NULL), \ + .index = (_offset) | ((_width) << 16) | ((_counter) << 24) \ +} + +static ssize_t show_pma_counter(struct ib_port *p, struct port_attribute *attr, + char *buf) +{ + struct port_table_attribute *tab_attr = + container_of(attr, struct port_table_attribute, attr); + int offset = tab_attr->index & 0xffff; + int width = (tab_attr->index >> 16) & 0xff; + struct ib_mad *in_mad = NULL; + struct ib_mad *out_mad = NULL; + ssize_t ret; + + if (!p->ibdev->process_mad) + return sprintf(buf, "N/A (no PMA)\n"); + + in_mad = kmalloc(sizeof *in_mad, GFP_KERNEL); + out_mad = kmalloc(sizeof *in_mad, GFP_KERNEL); + if (!in_mad || !out_mad) { + ret = -ENOMEM; + goto out; + } + + memset(in_mad, 0, sizeof *in_mad); + in_mad->mad_hdr.base_version = 1; + in_mad->mad_hdr.mgmt_class = IB_MGMT_CLASS_PERF_MGMT; + in_mad->mad_hdr.class_version = 1; + in_mad->mad_hdr.method = IB_MGMT_METHOD_GET; + in_mad->mad_hdr.attr_id = cpu_to_be16(0x12); /* PortCounters */ + + in_mad->data[41] = p->port_num; /* PortSelect field */ + + if ((p->ibdev->process_mad(p->ibdev, IB_MAD_IGNORE_MKEY, + p->port_num, NULL, NULL, in_mad, out_mad) & + (IB_MAD_RESULT_SUCCESS | IB_MAD_RESULT_REPLY)) != + (IB_MAD_RESULT_SUCCESS | IB_MAD_RESULT_REPLY)) { + ret = -EINVAL; + goto out; + } + + switch (width) { + case 4: + ret = sprintf(buf, "%u\n", (out_mad->data[40 + offset / 8] >> + (offset % 4)) & 0xf); + break; + case 8: + ret = sprintf(buf, "%u\n", out_mad->data[40 + offset / 8]); + break; + case 16: + ret = sprintf(buf, "%u\n", + be16_to_cpup((u16 *)(out_mad->data + 40 + offset / 8))); + break; + case 32: + ret = sprintf(buf, "%u\n", + be32_to_cpup((u32 *)(out_mad->data + 40 + offset / 8))); + break; + default: + ret = 0; + } + +out: + kfree(in_mad); + kfree(out_mad); + + return ret; +} + +static PORT_PMA_ATTR(symbol_error , 0, 16, 32); +static PORT_PMA_ATTR(link_error_recovery , 1, 8, 48); +static PORT_PMA_ATTR(link_downed , 2, 8, 56); +static PORT_PMA_ATTR(port_rcv_errors , 3, 16, 64); +static PORT_PMA_ATTR(port_rcv_remote_physical_errors, 4, 16, 80); +static PORT_PMA_ATTR(port_rcv_switch_relay_errors , 5, 16, 96); +static PORT_PMA_ATTR(port_xmit_discards , 6, 16, 112); +static PORT_PMA_ATTR(port_xmit_constraint_errors , 7, 8, 128); +static PORT_PMA_ATTR(port_rcv_constraint_errors , 8, 8, 136); +static PORT_PMA_ATTR(local_link_integrity_errors , 9, 4, 152); +static PORT_PMA_ATTR(excessive_buffer_overrun_errors, 10, 4, 156); +static PORT_PMA_ATTR(VL15_dropped , 11, 16, 176); +static PORT_PMA_ATTR(port_xmit_data , 12, 32, 192); +static PORT_PMA_ATTR(port_rcv_data , 13, 32, 224); +static PORT_PMA_ATTR(port_xmit_packets , 14, 32, 256); +static PORT_PMA_ATTR(port_rcv_packets , 15, 32, 288); + +static struct attribute *pma_attrs[] = { + &port_pma_attr_symbol_error.attr.attr, + &port_pma_attr_link_error_recovery.attr.attr, + &port_pma_attr_link_downed.attr.attr, + &port_pma_attr_port_rcv_errors.attr.attr, + &port_pma_attr_port_rcv_remote_physical_errors.attr.attr, + &port_pma_attr_port_rcv_switch_relay_errors.attr.attr, + &port_pma_attr_port_xmit_discards.attr.attr, + &port_pma_attr_port_xmit_constraint_errors.attr.attr, + &port_pma_attr_port_rcv_constraint_errors.attr.attr, + &port_pma_attr_local_link_integrity_errors.attr.attr, + &port_pma_attr_excessive_buffer_overrun_errors.attr.attr, + &port_pma_attr_VL15_dropped.attr.attr, + &port_pma_attr_port_xmit_data.attr.attr, + &port_pma_attr_port_rcv_data.attr.attr, + &port_pma_attr_port_xmit_packets.attr.attr, + &port_pma_attr_port_rcv_packets.attr.attr, + NULL +}; + +static struct attribute_group pma_group = { + .name = "counters", + .attrs = pma_attrs +}; + +static void ib_port_release(struct kobject *kobj) +{ + struct ib_port *p = container_of(kobj, struct ib_port, kobj); + struct attribute *a; + int i; + + for (i = 0; (a = p->gid_attr[i]); ++i) { + kfree(a->name); + kfree(a); + } + + for (i = 0; (a = p->pkey_attr[i]); ++i) { + kfree(a->name); + kfree(a); + } + + kfree(p->gid_attr); + kfree(p); +} + +static struct kobj_type port_type = { + .release = ib_port_release, + .sysfs_ops = &port_sysfs_ops, + .default_attrs = port_default_attrs +}; + +static void ib_device_release(struct class_device *cdev) +{ + struct ib_device *dev = container_of(cdev, struct ib_device, class_dev); + + kfree(dev); +} + +static int ib_device_hotplug(struct class_device *cdev, char **envp, + int num_envp, char *buf, int size) +{ + struct ib_device *dev = container_of(cdev, struct ib_device, class_dev); + int i = 0, len = 0; + + if (add_hotplug_env_var(envp, num_envp, &i, buf, size, &len, + "NAME=%s", dev->name)) + return -ENOMEM; + + /* + * It might be nice to pass the node GUID to hotplug, but + * right now the only way to get it is to query the device + * provider, and this can crash during device removal because + * we are will be running after driver removal has started. + * We could add a node_guid field to struct ib_device, or we + * could just let the hotplug script read the node GUID from + * sysfs when devices are added. + */ + + envp[i] = NULL; + return 0; +} + +static int alloc_group(struct attribute ***attr, + ssize_t (*show)(struct ib_port *, + struct port_attribute *, char *buf), + int len) +{ + struct port_table_attribute ***tab_attr = + (struct port_table_attribute ***) attr; + int i; + int ret; + + *tab_attr = kmalloc((1 + len) * sizeof *tab_attr, GFP_KERNEL); + if (!*tab_attr) + return -ENOMEM; + + memset(*tab_attr, 0, (1 + len) * sizeof *tab_attr); + + for (i = 0; i < len; ++i) { + (*tab_attr)[i] = kmalloc(sizeof *(*tab_attr)[i], GFP_KERNEL); + if (!(*tab_attr)[i]) { + ret = -ENOMEM; + goto err; + } + memset((*tab_attr)[i], 0, sizeof *(*tab_attr)[i]); + (*tab_attr)[i]->attr.attr.name = kmalloc(8, GFP_KERNEL); + if (!(*tab_attr)[i]->attr.attr.name) { + ret = -ENOMEM; + goto err; + } + + if (snprintf((*tab_attr)[i]->attr.attr.name, 8, "%d", i) >= 8) { + ret = -ENOMEM; + goto err; + } + + (*tab_attr)[i]->attr.attr.mode = S_IRUGO; + (*tab_attr)[i]->attr.attr.owner = THIS_MODULE; + (*tab_attr)[i]->attr.show = show; + (*tab_attr)[i]->index = i; + } + + return 0; + +err: + for (i = 0; i < len; ++i) { + if ((*tab_attr)[i]) + kfree((*tab_attr)[i]->attr.attr.name); + kfree((*tab_attr)[i]); + } + + kfree(*tab_attr); + + return ret; +} + +static int add_port(struct ib_device *device, int port_num) +{ + struct ib_port *p; + struct ib_port_attr attr; + int i; + int ret; + + ret = ib_query_port(device, port_num, &attr); + if (ret) + return ret; + + p = kmalloc(sizeof *p, GFP_KERNEL); + if (!p) + return -ENOMEM; + memset(p, 0, sizeof *p); + + p->ibdev = device; + p->port_num = port_num; + p->kobj.ktype = &port_type; + + p->kobj.parent = kobject_get(&device->ports_parent); + if (!p->kobj.parent) { + ret = -EBUSY; + goto err; + } + + ret = kobject_set_name(&p->kobj, "%d", port_num); + if (ret) + goto err_put; + + ret = kobject_register(&p->kobj); + if (ret) + goto err_put; + + ret = sysfs_create_group(&p->kobj, &pma_group); + if (ret) + goto err_put; + + ret = alloc_group(&p->gid_attr, show_port_gid, attr.gid_tbl_len); + if (ret) + goto err_remove_pma; + + p->gid_group.name = "gids"; + p->gid_group.attrs = p->gid_attr; + + ret = sysfs_create_group(&p->kobj, &p->gid_group); + if (ret) + goto err_free_gid; + + ret = alloc_group(&p->pkey_attr, show_port_pkey, attr.pkey_tbl_len); + if (ret) + goto err_remove_gid; + + p->pkey_group.name = "pkeys"; + p->pkey_group.attrs = p->pkey_attr; + + ret = sysfs_create_group(&p->kobj, &p->pkey_group); + if (ret) + goto err_free_pkey; + + list_add_tail(&p->kobj.entry, &device->port_list); + + return 0; + +err_free_pkey: + for (i = 0; i < attr.pkey_tbl_len; ++i) { + kfree(p->pkey_attr[i]->name); + kfree(p->pkey_attr[i]); + } + + kfree(p->pkey_attr); + +err_remove_gid: + sysfs_remove_group(&p->kobj, &p->gid_group); + +err_free_gid: + for (i = 0; i < attr.gid_tbl_len; ++i) { + kfree(p->gid_attr[i]->name); + kfree(p->gid_attr[i]); + } + + kfree(p->gid_attr); + +err_remove_pma: + sysfs_remove_group(&p->kobj, &pma_group); + +err_put: + kobject_put(&device->ports_parent); + +err: + kfree(p); + return ret; +} + +static ssize_t show_node_type(struct class_device *cdev, char *buf) +{ + struct ib_device *dev = container_of(cdev, struct ib_device, class_dev); + + switch (dev->node_type) { + case IB_NODE_CA: return sprintf(buf, "%d: CA\n", dev->node_type); + case IB_NODE_SWITCH: return sprintf(buf, "%d: switch\n", dev->node_type); + case IB_NODE_ROUTER: return sprintf(buf, "%d: router\n", dev->node_type); + default: return sprintf(buf, "%d: <unknown>\n", dev->node_type); + } +} + +static ssize_t show_sys_image_guid(struct class_device *cdev, char *buf) +{ + struct ib_device *dev = container_of(cdev, struct ib_device, class_dev); + struct ib_device_attr attr; + ssize_t ret; + + ret = ib_query_device(dev, &attr); + if (ret) + return ret; + + return sprintf(buf, "%04x:%04x:%04x:%04x\n", + be16_to_cpu(((u16 *) &attr.sys_image_guid)[0]), + be16_to_cpu(((u16 *) &attr.sys_image_guid)[1]), + be16_to_cpu(((u16 *) &attr.sys_image_guid)[2]), + be16_to_cpu(((u16 *) &attr.sys_image_guid)[3])); +} + +static ssize_t show_node_guid(struct class_device *cdev, char *buf) +{ + struct ib_device *dev = container_of(cdev, struct ib_device, class_dev); + struct ib_device_attr attr; + ssize_t ret; + + ret = ib_query_device(dev, &attr); + if (ret) + return ret; + + return sprintf(buf, "%04x:%04x:%04x:%04x\n", + be16_to_cpu(((u16 *) &attr.node_guid)[0]), + be16_to_cpu(((u16 *) &attr.node_guid)[1]), + be16_to_cpu(((u16 *) &attr.node_guid)[2]), + be16_to_cpu(((u16 *) &attr.node_guid)[3])); +} + +static CLASS_DEVICE_ATTR(node_type, S_IRUGO, show_node_type, NULL); +static CLASS_DEVICE_ATTR(sys_image_guid, S_IRUGO, show_sys_image_guid, NULL); +static CLASS_DEVICE_ATTR(node_guid, S_IRUGO, show_node_guid, NULL); + +static struct class_device_attribute *ib_class_attributes[] = { + &class_device_attr_node_type, + &class_device_attr_sys_image_guid, + &class_device_attr_node_guid +}; + +static struct class ib_class = { + .name = "infiniband", + .release = ib_device_release, + .hotplug = ib_device_hotplug, +}; + +int ib_device_register_sysfs(struct ib_device *device) +{ + struct class_device *class_dev = &device->class_dev; + int ret; + int i; + + class_dev->class = &ib_class; + class_dev->class_data = device; + strlcpy(class_dev->class_id, device->name, BUS_ID_SIZE); + + INIT_LIST_HEAD(&device->port_list); + + ret = class_device_register(class_dev); + if (ret) + goto err; + + for (i = 0; i < ARRAY_SIZE(ib_class_attributes); ++i) { + ret = class_device_create_file(class_dev, ib_class_attributes[i]); + if (ret) + goto err_unregister; + } + + device->ports_parent.parent = kobject_get(&class_dev->kobj); + if (!device->ports_parent.parent) { + ret = -EBUSY; + goto err_unregister; + } + ret = kobject_set_name(&device->ports_parent, "ports"); + if (ret) + goto err_put; + ret = kobject_register(&device->ports_parent); + if (ret) + goto err_put; + + if (device->node_type == IB_NODE_SWITCH) { + ret = add_port(device, 0); + if (ret) + goto err_put; + } else { + int i; + + for (i = 1; i <= device->phys_port_cnt; ++i) { + ret = add_port(device, i); + if (ret) + goto err_put; + } + } + + return 0; + +err_put: + { + struct kobject *p, *t; + struct ib_port *port; + + list_for_each_entry_safe(p, t, &device->port_list, entry) { + list_del(&p->entry); + port = container_of(p, struct ib_port, kobj); + sysfs_remove_group(p, &pma_group); + sysfs_remove_group(p, &port->pkey_group); + sysfs_remove_group(p, &port->gid_group); + kobject_unregister(p); + } + } + + kobject_put(&class_dev->kobj); + +err_unregister: + class_device_unregister(class_dev); + +err: + return ret; +} + +void ib_device_unregister_sysfs(struct ib_device *device) +{ + struct kobject *p, *t; + struct ib_port *port; + + list_for_each_entry_safe(p, t, &device->port_list, entry) { + list_del(&p->entry); + port = container_of(p, struct ib_port, kobj); + sysfs_remove_group(p, &pma_group); + sysfs_remove_group(p, &port->pkey_group); + sysfs_remove_group(p, &port->gid_group); + kobject_unregister(p); + } + + kobject_unregister(&device->ports_parent); + class_device_unregister(&device->class_dev); +} + +int ib_sysfs_setup(void) +{ + return class_register(&ib_class); +} + +void ib_sysfs_cleanup(void) +{ + class_unregister(&ib_class); +} diff --git a/drivers/infiniband/core/ud_header.c b/drivers/infiniband/core/ud_header.c new file mode 100644 index 000000000000..dc4eb1db5e96 --- /dev/null +++ b/drivers/infiniband/core/ud_header.c @@ -0,0 +1,365 @@ +/* + * Copyright (c) 2004 Topspin Corporation. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + * + * $Id: ud_header.c 1349 2004-12-16 21:09:43Z roland $ + */ + +#include <linux/errno.h> + +#include <ib_pack.h> + +#define STRUCT_FIELD(header, field) \ + .struct_offset_bytes = offsetof(struct ib_unpacked_ ## header, field), \ + .struct_size_bytes = sizeof ((struct ib_unpacked_ ## header *) 0)->field, \ + .field_name = #header ":" #field + +static const struct ib_field lrh_table[] = { + { STRUCT_FIELD(lrh, virtual_lane), + .offset_words = 0, + .offset_bits = 0, + .size_bits = 4 }, + { STRUCT_FIELD(lrh, link_version), + .offset_words = 0, + .offset_bits = 4, + .size_bits = 4 }, + { STRUCT_FIELD(lrh, service_level), + .offset_words = 0, + .offset_bits = 8, + .size_bits = 4 }, + { RESERVED, + .offset_words = 0, + .offset_bits = 12, + .size_bits = 2 }, + { STRUCT_FIELD(lrh, link_next_header), + .offset_words = 0, + .offset_bits = 14, + .size_bits = 2 }, + { STRUCT_FIELD(lrh, destination_lid), + .offset_words = 0, + .offset_bits = 16, + .size_bits = 16 }, + { RESERVED, + .offset_words = 1, + .offset_bits = 0, + .size_bits = 5 }, + { STRUCT_FIELD(lrh, packet_length), + .offset_words = 1, + .offset_bits = 5, + .size_bits = 11 }, + { STRUCT_FIELD(lrh, source_lid), + .offset_words = 1, + .offset_bits = 16, + .size_bits = 16 } +}; + +static const struct ib_field grh_table[] = { + { STRUCT_FIELD(grh, ip_version), + .offset_words = 0, + .offset_bits = 0, + .size_bits = 4 }, + { STRUCT_FIELD(grh, traffic_class), + .offset_words = 0, + .offset_bits = 4, + .size_bits = 8 }, + { STRUCT_FIELD(grh, flow_label), + .offset_words = 0, + .offset_bits = 12, + .size_bits = 20 }, + { STRUCT_FIELD(grh, payload_length), + .offset_words = 1, + .offset_bits = 0, + .size_bits = 16 }, + { STRUCT_FIELD(grh, next_header), + .offset_words = 1, + .offset_bits = 16, + .size_bits = 8 }, + { STRUCT_FIELD(grh, hop_limit), + .offset_words = 1, + .offset_bits = 24, + .size_bits = 8 }, + { STRUCT_FIELD(grh, source_gid), + .offset_words = 2, + .offset_bits = 0, + .size_bits = 128 }, + { STRUCT_FIELD(grh, destination_gid), + .offset_words = 6, + .offset_bits = 0, + .size_bits = 128 } +}; + +static const struct ib_field bth_table[] = { + { STRUCT_FIELD(bth, opcode), + .offset_words = 0, + .offset_bits = 0, + .size_bits = 8 }, + { STRUCT_FIELD(bth, solicited_event), + .offset_words = 0, + .offset_bits = 8, + .size_bits = 1 }, + { STRUCT_FIELD(bth, mig_req), + .offset_words = 0, + .offset_bits = 9, + .size_bits = 1 }, + { STRUCT_FIELD(bth, pad_count), + .offset_words = 0, + .offset_bits = 10, + .size_bits = 2 }, + { STRUCT_FIELD(bth, transport_header_version), + .offset_words = 0, + .offset_bits = 12, + .size_bits = 4 }, + { STRUCT_FIELD(bth, pkey), + .offset_words = 0, + .offset_bits = 16, + .size_bits = 16 }, + { RESERVED, + .offset_words = 1, + .offset_bits = 0, + .size_bits = 8 }, + { STRUCT_FIELD(bth, destination_qpn), + .offset_words = 1, + .offset_bits = 8, + .size_bits = 24 }, + { STRUCT_FIELD(bth, ack_req), + .offset_words = 2, + .offset_bits = 0, + .size_bits = 1 }, + { RESERVED, + .offset_words = 2, + .offset_bits = 1, + .size_bits = 7 }, + { STRUCT_FIELD(bth, psn), + .offset_words = 2, + .offset_bits = 8, + .size_bits = 24 } +}; + +static const struct ib_field deth_table[] = { + { STRUCT_FIELD(deth, qkey), + .offset_words = 0, + .offset_bits = 0, + .size_bits = 32 }, + { RESERVED, + .offset_words = 1, + .offset_bits = 0, + .size_bits = 8 }, + { STRUCT_FIELD(deth, source_qpn), + .offset_words = 1, + .offset_bits = 8, + .size_bits = 24 } +}; + +/** + * ib_ud_header_init - Initialize UD header structure + * @payload_bytes:Length of packet payload + * @grh_present:GRH flag (if non-zero, GRH will be included) + * @header:Structure to initialize + * + * ib_ud_header_init() initializes the lrh.link_version, lrh.link_next_header, + * lrh.packet_length, grh.ip_version, grh.payload_length, + * grh.next_header, bth.opcode, bth.pad_count and + * bth.transport_header_version fields of a &struct ib_ud_header given + * the payload length and whether a GRH will be included. + */ +void ib_ud_header_init(int payload_bytes, + int grh_present, + struct ib_ud_header *header) +{ + int header_len; + + memset(header, 0, sizeof *header); + + header_len = + IB_LRH_BYTES + + IB_BTH_BYTES + + IB_DETH_BYTES; + if (grh_present) { + header_len += IB_GRH_BYTES; + } + + header->lrh.link_version = 0; + header->lrh.link_next_header = + grh_present ? IB_LNH_IBA_GLOBAL : IB_LNH_IBA_LOCAL; + header->lrh.packet_length = (IB_LRH_BYTES + + IB_BTH_BYTES + + IB_DETH_BYTES + + payload_bytes + + 4 + /* ICRC */ + 3) / 4; /* round up */ + + header->grh_present = grh_present; + if (grh_present) { + header->lrh.packet_length += IB_GRH_BYTES / 4; + + header->grh.ip_version = 6; + header->grh.payload_length = + cpu_to_be16((IB_BTH_BYTES + + IB_DETH_BYTES + + payload_bytes + + 4 + /* ICRC */ + 3) & ~3); /* round up */ + header->grh.next_header = 0x1b; + } + + cpu_to_be16s(&header->lrh.packet_length); + + if (header->immediate_present) + header->bth.opcode = IB_OPCODE_UD_SEND_ONLY_WITH_IMMEDIATE; + else + header->bth.opcode = IB_OPCODE_UD_SEND_ONLY; + header->bth.pad_count = (4 - payload_bytes) & 3; + header->bth.transport_header_version = 0; +} +EXPORT_SYMBOL(ib_ud_header_init); + +/** + * ib_ud_header_pack - Pack UD header struct into wire format + * @header:UD header struct + * @buf:Buffer to pack into + * + * ib_ud_header_pack() packs the UD header structure @header into wire + * format in the buffer @buf. + */ +int ib_ud_header_pack(struct ib_ud_header *header, + void *buf) +{ + int len = 0; + + ib_pack(lrh_table, ARRAY_SIZE(lrh_table), + &header->lrh, buf); + len += IB_LRH_BYTES; + + if (header->grh_present) { + ib_pack(grh_table, ARRAY_SIZE(grh_table), + &header->grh, buf + len); + len += IB_GRH_BYTES; + } + + ib_pack(bth_table, ARRAY_SIZE(bth_table), + &header->bth, buf + len); + len += IB_BTH_BYTES; + + ib_pack(deth_table, ARRAY_SIZE(deth_table), + &header->deth, buf + len); + len += IB_DETH_BYTES; + + if (header->immediate_present) { + memcpy(buf + len, &header->immediate_data, sizeof header->immediate_data); + len += sizeof header->immediate_data; + } + + return len; +} +EXPORT_SYMBOL(ib_ud_header_pack); + +/** + * ib_ud_header_unpack - Unpack UD header struct from wire format + * @header:UD header struct + * @buf:Buffer to pack into + * + * ib_ud_header_pack() unpacks the UD header structure @header from wire + * format in the buffer @buf. + */ +int ib_ud_header_unpack(void *buf, + struct ib_ud_header *header) +{ + ib_unpack(lrh_table, ARRAY_SIZE(lrh_table), + buf, &header->lrh); + buf += IB_LRH_BYTES; + + if (header->lrh.link_version != 0) { + printk(KERN_WARNING "Invalid LRH.link_version %d\n", + header->lrh.link_version); + return -EINVAL; + } + + switch (header->lrh.link_next_header) { + case IB_LNH_IBA_LOCAL: + header->grh_present = 0; + break; + + case IB_LNH_IBA_GLOBAL: + header->grh_present = 1; + ib_unpack(grh_table, ARRAY_SIZE(grh_table), + buf, &header->grh); + buf += IB_GRH_BYTES; + + if (header->grh.ip_version != 6) { + printk(KERN_WARNING "Invalid GRH.ip_version %d\n", + header->grh.ip_version); + return -EINVAL; + } + if (header->grh.next_header != 0x1b) { + printk(KERN_WARNING "Invalid GRH.next_header 0x%02x\n", + header->grh.next_header); + return -EINVAL; + } + break; + + default: + printk(KERN_WARNING "Invalid LRH.link_next_header %d\n", + header->lrh.link_next_header); + return -EINVAL; + } + + ib_unpack(bth_table, ARRAY_SIZE(bth_table), + buf, &header->bth); + buf += IB_BTH_BYTES; + + switch (header->bth.opcode) { + case IB_OPCODE_UD_SEND_ONLY: + header->immediate_present = 0; + break; + case IB_OPCODE_UD_SEND_ONLY_WITH_IMMEDIATE: + header->immediate_present = 1; + break; + default: + printk(KERN_WARNING "Invalid BTH.opcode 0x%02x\n", + header->bth.opcode); + return -EINVAL; + } + + if (header->bth.transport_header_version != 0) { + printk(KERN_WARNING "Invalid BTH.transport_header_version %d\n", + header->bth.transport_header_version); + return -EINVAL; + } + + ib_unpack(deth_table, ARRAY_SIZE(deth_table), + buf, &header->deth); + buf += IB_DETH_BYTES; + + if (header->immediate_present) + memcpy(&header->immediate_data, buf, sizeof header->immediate_data); + + return 0; +} +EXPORT_SYMBOL(ib_ud_header_unpack); diff --git a/drivers/infiniband/core/user_mad.c b/drivers/infiniband/core/user_mad.c new file mode 100644 index 000000000000..54b4f33b0bf9 --- /dev/null +++ b/drivers/infiniband/core/user_mad.c @@ -0,0 +1,840 @@ +/* + * Copyright (c) 2004 Topspin Communications. 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. + * + * $Id: user_mad.c 1389 2004-12-27 22:56:47Z roland $ + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/device.h> +#include <linux/err.h> +#include <linux/fs.h> +#include <linux/cdev.h> +#include <linux/pci.h> +#include <linux/dma-mapping.h> +#include <linux/poll.h> +#include <linux/rwsem.h> +#include <linux/kref.h> + +#include <asm/uaccess.h> +#include <asm/semaphore.h> + +#include <ib_mad.h> +#include <ib_user_mad.h> + +MODULE_AUTHOR("Roland Dreier"); +MODULE_DESCRIPTION("InfiniBand userspace MAD packet access"); +MODULE_LICENSE("Dual BSD/GPL"); + +enum { + IB_UMAD_MAX_PORTS = 64, + IB_UMAD_MAX_AGENTS = 32, + + IB_UMAD_MAJOR = 231, + IB_UMAD_MINOR_BASE = 0 +}; + +struct ib_umad_port { + int devnum; + struct cdev dev; + struct class_device class_dev; + + int sm_devnum; + struct cdev sm_dev; + struct class_device sm_class_dev; + struct semaphore sm_sem; + + struct ib_device *ib_dev; + struct ib_umad_device *umad_dev; + u8 port_num; +}; + +struct ib_umad_device { + int start_port, end_port; + struct kref ref; + struct ib_umad_port port[0]; +}; + +struct ib_umad_file { + struct ib_umad_port *port; + spinlock_t recv_lock; + struct list_head recv_list; + wait_queue_head_t recv_wait; + struct rw_semaphore agent_mutex; + struct ib_mad_agent *agent[IB_UMAD_MAX_AGENTS]; + struct ib_mr *mr[IB_UMAD_MAX_AGENTS]; +}; + +struct ib_umad_packet { + struct ib_user_mad mad; + struct ib_ah *ah; + struct list_head list; + DECLARE_PCI_UNMAP_ADDR(mapping) +}; + +static const dev_t base_dev = MKDEV(IB_UMAD_MAJOR, IB_UMAD_MINOR_BASE); +static spinlock_t map_lock; +static DECLARE_BITMAP(dev_map, IB_UMAD_MAX_PORTS * 2); + +static void ib_umad_add_one(struct ib_device *device); +static void ib_umad_remove_one(struct ib_device *device); + +static int queue_packet(struct ib_umad_file *file, + struct ib_mad_agent *agent, + struct ib_umad_packet *packet) +{ + int ret = 1; + + down_read(&file->agent_mutex); + for (packet->mad.id = 0; + packet->mad.id < IB_UMAD_MAX_AGENTS; + packet->mad.id++) + if (agent == file->agent[packet->mad.id]) { + spin_lock_irq(&file->recv_lock); + list_add_tail(&packet->list, &file->recv_list); + spin_unlock_irq(&file->recv_lock); + wake_up_interruptible(&file->recv_wait); + ret = 0; + break; + } + + up_read(&file->agent_mutex); + + return ret; +} + +static void send_handler(struct ib_mad_agent *agent, + struct ib_mad_send_wc *send_wc) +{ + struct ib_umad_file *file = agent->context; + struct ib_umad_packet *packet = + (void *) (unsigned long) send_wc->wr_id; + + dma_unmap_single(agent->device->dma_device, + pci_unmap_addr(packet, mapping), + sizeof packet->mad.data, + DMA_TO_DEVICE); + ib_destroy_ah(packet->ah); + + if (send_wc->status == IB_WC_RESP_TIMEOUT_ERR) { + packet->mad.status = ETIMEDOUT; + + if (!queue_packet(file, agent, packet)) + return; + } + + kfree(packet); +} + +static void recv_handler(struct ib_mad_agent *agent, + struct ib_mad_recv_wc *mad_recv_wc) +{ + struct ib_umad_file *file = agent->context; + struct ib_umad_packet *packet; + + if (mad_recv_wc->wc->status != IB_WC_SUCCESS) + goto out; + + packet = kmalloc(sizeof *packet, GFP_KERNEL); + if (!packet) + goto out; + + memset(packet, 0, sizeof *packet); + + memcpy(packet->mad.data, mad_recv_wc->recv_buf.mad, sizeof packet->mad.data); + packet->mad.status = 0; + packet->mad.qpn = cpu_to_be32(mad_recv_wc->wc->src_qp); + packet->mad.lid = cpu_to_be16(mad_recv_wc->wc->slid); + packet->mad.sl = mad_recv_wc->wc->sl; + packet->mad.path_bits = mad_recv_wc->wc->dlid_path_bits; + packet->mad.grh_present = !!(mad_recv_wc->wc->wc_flags & IB_WC_GRH); + if (packet->mad.grh_present) { + /* XXX parse GRH */ + packet->mad.gid_index = 0; + packet->mad.hop_limit = 0; + packet->mad.traffic_class = 0; + memset(packet->mad.gid, 0, 16); + packet->mad.flow_label = 0; + } + + if (queue_packet(file, agent, packet)) + kfree(packet); + +out: + ib_free_recv_mad(mad_recv_wc); +} + +static ssize_t ib_umad_read(struct file *filp, char __user *buf, + size_t count, loff_t *pos) +{ + struct ib_umad_file *file = filp->private_data; + struct ib_umad_packet *packet; + ssize_t ret; + + if (count < sizeof (struct ib_user_mad)) + return -EINVAL; + + spin_lock_irq(&file->recv_lock); + + while (list_empty(&file->recv_list)) { + spin_unlock_irq(&file->recv_lock); + + if (filp->f_flags & O_NONBLOCK) + return -EAGAIN; + + if (wait_event_interruptible(file->recv_wait, + !list_empty(&file->recv_list))) + return -ERESTARTSYS; + + spin_lock_irq(&file->recv_lock); + } + + packet = list_entry(file->recv_list.next, struct ib_umad_packet, list); + list_del(&packet->list); + + spin_unlock_irq(&file->recv_lock); + + if (copy_to_user(buf, &packet->mad, sizeof packet->mad)) + ret = -EFAULT; + else + ret = sizeof packet->mad; + + kfree(packet); + return ret; +} + +static ssize_t ib_umad_write(struct file *filp, const char __user *buf, + size_t count, loff_t *pos) +{ + struct ib_umad_file *file = filp->private_data; + struct ib_umad_packet *packet; + struct ib_mad_agent *agent; + struct ib_ah_attr ah_attr; + struct ib_sge gather_list; + struct ib_send_wr *bad_wr, wr = { + .opcode = IB_WR_SEND, + .sg_list = &gather_list, + .num_sge = 1, + .send_flags = IB_SEND_SIGNALED, + }; + u8 method; + u64 *tid; + int ret; + + if (count < sizeof (struct ib_user_mad)) + return -EINVAL; + + packet = kmalloc(sizeof *packet, GFP_KERNEL); + if (!packet) + return -ENOMEM; + + if (copy_from_user(&packet->mad, buf, sizeof packet->mad)) { + kfree(packet); + return -EFAULT; + } + + if (packet->mad.id < 0 || packet->mad.id >= IB_UMAD_MAX_AGENTS) { + ret = -EINVAL; + goto err; + } + + down_read(&file->agent_mutex); + + agent = file->agent[packet->mad.id]; + if (!agent) { + ret = -EINVAL; + goto err_up; + } + + /* + * If userspace is generating a request that will generate a + * response, we need to make sure the high-order part of the + * transaction ID matches the agent being used to send the + * MAD. + */ + method = ((struct ib_mad_hdr *) packet->mad.data)->method; + + if (!(method & IB_MGMT_METHOD_RESP) && + method != IB_MGMT_METHOD_TRAP_REPRESS && + method != IB_MGMT_METHOD_SEND) { + tid = &((struct ib_mad_hdr *) packet->mad.data)->tid; + *tid = cpu_to_be64(((u64) agent->hi_tid) << 32 | + (be64_to_cpup(tid) & 0xffffffff)); + } + + memset(&ah_attr, 0, sizeof ah_attr); + ah_attr.dlid = be16_to_cpu(packet->mad.lid); + ah_attr.sl = packet->mad.sl; + ah_attr.src_path_bits = packet->mad.path_bits; + ah_attr.port_num = file->port->port_num; + if (packet->mad.grh_present) { + ah_attr.ah_flags = IB_AH_GRH; + memcpy(ah_attr.grh.dgid.raw, packet->mad.gid, 16); + ah_attr.grh.flow_label = packet->mad.flow_label; + ah_attr.grh.hop_limit = packet->mad.hop_limit; + ah_attr.grh.traffic_class = packet->mad.traffic_class; + } + + packet->ah = ib_create_ah(agent->qp->pd, &ah_attr); + if (IS_ERR(packet->ah)) { + ret = PTR_ERR(packet->ah); + goto err_up; + } + + gather_list.addr = dma_map_single(agent->device->dma_device, + packet->mad.data, + sizeof packet->mad.data, + DMA_TO_DEVICE); + gather_list.length = sizeof packet->mad.data; + gather_list.lkey = file->mr[packet->mad.id]->lkey; + pci_unmap_addr_set(packet, mapping, gather_list.addr); + + wr.wr.ud.mad_hdr = (struct ib_mad_hdr *) packet->mad.data; + wr.wr.ud.ah = packet->ah; + wr.wr.ud.remote_qpn = be32_to_cpu(packet->mad.qpn); + wr.wr.ud.remote_qkey = be32_to_cpu(packet->mad.qkey); + wr.wr.ud.timeout_ms = packet->mad.timeout_ms; + + wr.wr_id = (unsigned long) packet; + + ret = ib_post_send_mad(agent, &wr, &bad_wr); + if (ret) { + dma_unmap_single(agent->device->dma_device, + pci_unmap_addr(packet, mapping), + sizeof packet->mad.data, + DMA_TO_DEVICE); + goto err_up; + } + + up_read(&file->agent_mutex); + + return sizeof packet->mad; + +err_up: + up_read(&file->agent_mutex); + +err: + kfree(packet); + return ret; +} + +static unsigned int ib_umad_poll(struct file *filp, struct poll_table_struct *wait) +{ + struct ib_umad_file *file = filp->private_data; + + /* we will always be able to post a MAD send */ + unsigned int mask = POLLOUT | POLLWRNORM; + + poll_wait(filp, &file->recv_wait, wait); + + if (!list_empty(&file->recv_list)) + mask |= POLLIN | POLLRDNORM; + + return mask; +} + +static int ib_umad_reg_agent(struct ib_umad_file *file, unsigned long arg) +{ + struct ib_user_mad_reg_req ureq; + struct ib_mad_reg_req req; + struct ib_mad_agent *agent; + int agent_id; + int ret; + + down_write(&file->agent_mutex); + + if (copy_from_user(&ureq, (void __user *) arg, sizeof ureq)) { + ret = -EFAULT; + goto out; + } + + if (ureq.qpn != 0 && ureq.qpn != 1) { + ret = -EINVAL; + goto out; + } + + for (agent_id = 0; agent_id < IB_UMAD_MAX_AGENTS; ++agent_id) + if (!file->agent[agent_id]) + goto found; + + ret = -ENOMEM; + goto out; + +found: + req.mgmt_class = ureq.mgmt_class; + req.mgmt_class_version = ureq.mgmt_class_version; + memcpy(req.method_mask, ureq.method_mask, sizeof req.method_mask); + memcpy(req.oui, ureq.oui, sizeof req.oui); + + agent = ib_register_mad_agent(file->port->ib_dev, file->port->port_num, + ureq.qpn ? IB_QPT_GSI : IB_QPT_SMI, + &req, 0, send_handler, recv_handler, + file); + if (IS_ERR(agent)) { + ret = PTR_ERR(agent); + goto out; + } + + file->agent[agent_id] = agent; + + file->mr[agent_id] = ib_get_dma_mr(agent->qp->pd, IB_ACCESS_LOCAL_WRITE); + if (IS_ERR(file->mr[agent_id])) { + ret = -ENOMEM; + goto err; + } + + if (put_user(agent_id, + (u32 __user *) (arg + offsetof(struct ib_user_mad_reg_req, id)))) { + ret = -EFAULT; + goto err_mr; + } + + ret = 0; + goto out; + +err_mr: + ib_dereg_mr(file->mr[agent_id]); + +err: + file->agent[agent_id] = NULL; + ib_unregister_mad_agent(agent); + +out: + up_write(&file->agent_mutex); + return ret; +} + +static int ib_umad_unreg_agent(struct ib_umad_file *file, unsigned long arg) +{ + u32 id; + int ret = 0; + + down_write(&file->agent_mutex); + + if (get_user(id, (u32 __user *) arg)) { + ret = -EFAULT; + goto out; + } + + if (id < 0 || id >= IB_UMAD_MAX_AGENTS || !file->agent[id]) { + ret = -EINVAL; + goto out; + } + + ib_dereg_mr(file->mr[id]); + ib_unregister_mad_agent(file->agent[id]); + file->agent[id] = NULL; + +out: + up_write(&file->agent_mutex); + return ret; +} + +static long ib_umad_ioctl(struct file *filp, + unsigned int cmd, unsigned long arg) +{ + switch (cmd) { + case IB_USER_MAD_REGISTER_AGENT: + return ib_umad_reg_agent(filp->private_data, arg); + case IB_USER_MAD_UNREGISTER_AGENT: + return ib_umad_unreg_agent(filp->private_data, arg); + default: + return -ENOIOCTLCMD; + } +} + +static int ib_umad_open(struct inode *inode, struct file *filp) +{ + struct ib_umad_port *port = + container_of(inode->i_cdev, struct ib_umad_port, dev); + struct ib_umad_file *file; + + file = kmalloc(sizeof *file, GFP_KERNEL); + if (!file) + return -ENOMEM; + + memset(file, 0, sizeof *file); + + spin_lock_init(&file->recv_lock); + init_rwsem(&file->agent_mutex); + INIT_LIST_HEAD(&file->recv_list); + init_waitqueue_head(&file->recv_wait); + + file->port = port; + filp->private_data = file; + + return 0; +} + +static int ib_umad_close(struct inode *inode, struct file *filp) +{ + struct ib_umad_file *file = filp->private_data; + int i; + + for (i = 0; i < IB_UMAD_MAX_AGENTS; ++i) + if (file->agent[i]) { + ib_dereg_mr(file->mr[i]); + ib_unregister_mad_agent(file->agent[i]); + } + + kfree(file); + + return 0; +} + +static struct file_operations umad_fops = { + .owner = THIS_MODULE, + .read = ib_umad_read, + .write = ib_umad_write, + .poll = ib_umad_poll, + .unlocked_ioctl = ib_umad_ioctl, + .compat_ioctl = ib_umad_ioctl, + .open = ib_umad_open, + .release = ib_umad_close +}; + +static int ib_umad_sm_open(struct inode *inode, struct file *filp) +{ + struct ib_umad_port *port = + container_of(inode->i_cdev, struct ib_umad_port, sm_dev); + struct ib_port_modify props = { + .set_port_cap_mask = IB_PORT_SM + }; + int ret; + + if (filp->f_flags & O_NONBLOCK) { + if (down_trylock(&port->sm_sem)) + return -EAGAIN; + } else { + if (down_interruptible(&port->sm_sem)) + return -ERESTARTSYS; + } + + ret = ib_modify_port(port->ib_dev, port->port_num, 0, &props); + if (ret) { + up(&port->sm_sem); + return ret; + } + + filp->private_data = port; + + return 0; +} + +static int ib_umad_sm_close(struct inode *inode, struct file *filp) +{ + struct ib_umad_port *port = filp->private_data; + struct ib_port_modify props = { + .clr_port_cap_mask = IB_PORT_SM + }; + int ret; + + ret = ib_modify_port(port->ib_dev, port->port_num, 0, &props); + up(&port->sm_sem); + + return ret; +} + +static struct file_operations umad_sm_fops = { + .owner = THIS_MODULE, + .open = ib_umad_sm_open, + .release = ib_umad_sm_close +}; + +static struct ib_client umad_client = { + .name = "umad", + .add = ib_umad_add_one, + .remove = ib_umad_remove_one +}; + +static ssize_t show_dev(struct class_device *class_dev, char *buf) +{ + struct ib_umad_port *port = class_get_devdata(class_dev); + + if (class_dev == &port->class_dev) + return print_dev_t(buf, port->dev.dev); + else + return print_dev_t(buf, port->sm_dev.dev); +} +static CLASS_DEVICE_ATTR(dev, S_IRUGO, show_dev, NULL); + +static ssize_t show_ibdev(struct class_device *class_dev, char *buf) +{ + struct ib_umad_port *port = class_get_devdata(class_dev); + + return sprintf(buf, "%s\n", port->ib_dev->name); +} +static CLASS_DEVICE_ATTR(ibdev, S_IRUGO, show_ibdev, NULL); + +static ssize_t show_port(struct class_device *class_dev, char *buf) +{ + struct ib_umad_port *port = class_get_devdata(class_dev); + + return sprintf(buf, "%d\n", port->port_num); +} +static CLASS_DEVICE_ATTR(port, S_IRUGO, show_port, NULL); + +static void ib_umad_release_dev(struct kref *ref) +{ + struct ib_umad_device *dev = + container_of(ref, struct ib_umad_device, ref); + + kfree(dev); +} + +static void ib_umad_release_port(struct class_device *class_dev) +{ + struct ib_umad_port *port = class_get_devdata(class_dev); + + if (class_dev == &port->class_dev) { + cdev_del(&port->dev); + clear_bit(port->devnum, dev_map); + } else { + cdev_del(&port->sm_dev); + clear_bit(port->sm_devnum, dev_map); + } + + kref_put(&port->umad_dev->ref, ib_umad_release_dev); +} + +static struct class umad_class = { + .name = "infiniband_mad", + .release = ib_umad_release_port +}; + +static ssize_t show_abi_version(struct class *class, char *buf) +{ + return sprintf(buf, "%d\n", IB_USER_MAD_ABI_VERSION); +} +static CLASS_ATTR(abi_version, S_IRUGO, show_abi_version, NULL); + +static int ib_umad_init_port(struct ib_device *device, int port_num, + struct ib_umad_port *port) +{ + spin_lock(&map_lock); + port->devnum = find_first_zero_bit(dev_map, IB_UMAD_MAX_PORTS); + if (port->devnum >= IB_UMAD_MAX_PORTS) { + spin_unlock(&map_lock); + return -1; + } + port->sm_devnum = find_next_zero_bit(dev_map, IB_UMAD_MAX_PORTS * 2, IB_UMAD_MAX_PORTS); + if (port->sm_devnum >= IB_UMAD_MAX_PORTS * 2) { + spin_unlock(&map_lock); + return -1; + } + set_bit(port->devnum, dev_map); + set_bit(port->sm_devnum, dev_map); + spin_unlock(&map_lock); + + port->ib_dev = device; + port->port_num = port_num; + init_MUTEX(&port->sm_sem); + + cdev_init(&port->dev, &umad_fops); + port->dev.owner = THIS_MODULE; + kobject_set_name(&port->dev.kobj, "umad%d", port->devnum); + if (cdev_add(&port->dev, base_dev + port->devnum, 1)) + return -1; + + port->class_dev.class = &umad_class; + port->class_dev.dev = device->dma_device; + + snprintf(port->class_dev.class_id, BUS_ID_SIZE, "umad%d", port->devnum); + + if (class_device_register(&port->class_dev)) + goto err_cdev; + + class_set_devdata(&port->class_dev, port); + kref_get(&port->umad_dev->ref); + + if (class_device_create_file(&port->class_dev, &class_device_attr_dev)) + goto err_class; + if (class_device_create_file(&port->class_dev, &class_device_attr_ibdev)) + goto err_class; + if (class_device_create_file(&port->class_dev, &class_device_attr_port)) + goto err_class; + + cdev_init(&port->sm_dev, &umad_sm_fops); + port->sm_dev.owner = THIS_MODULE; + kobject_set_name(&port->dev.kobj, "issm%d", port->sm_devnum - IB_UMAD_MAX_PORTS); + if (cdev_add(&port->sm_dev, base_dev + port->sm_devnum, 1)) + return -1; + + port->sm_class_dev.class = &umad_class; + port->sm_class_dev.dev = device->dma_device; + + snprintf(port->sm_class_dev.class_id, BUS_ID_SIZE, "issm%d", port->sm_devnum - IB_UMAD_MAX_PORTS); + + if (class_device_register(&port->sm_class_dev)) + goto err_sm_cdev; + + class_set_devdata(&port->sm_class_dev, port); + kref_get(&port->umad_dev->ref); + + if (class_device_create_file(&port->sm_class_dev, &class_device_attr_dev)) + goto err_sm_class; + if (class_device_create_file(&port->sm_class_dev, &class_device_attr_ibdev)) + goto err_sm_class; + if (class_device_create_file(&port->sm_class_dev, &class_device_attr_port)) + goto err_sm_class; + + return 0; + +err_sm_class: + class_device_unregister(&port->sm_class_dev); + +err_sm_cdev: + cdev_del(&port->sm_dev); + +err_class: + class_device_unregister(&port->class_dev); + +err_cdev: + cdev_del(&port->dev); + clear_bit(port->devnum, dev_map); + + return -1; +} + +static void ib_umad_add_one(struct ib_device *device) +{ + struct ib_umad_device *umad_dev; + int s, e, i; + + if (device->node_type == IB_NODE_SWITCH) + s = e = 0; + else { + s = 1; + e = device->phys_port_cnt; + } + + umad_dev = kmalloc(sizeof *umad_dev + + (e - s + 1) * sizeof (struct ib_umad_port), + GFP_KERNEL); + if (!umad_dev) + return; + + memset(umad_dev, 0, sizeof *umad_dev + + (e - s + 1) * sizeof (struct ib_umad_port)); + + kref_init(&umad_dev->ref); + + umad_dev->start_port = s; + umad_dev->end_port = e; + + for (i = s; i <= e; ++i) { + umad_dev->port[i - s].umad_dev = umad_dev; + + if (ib_umad_init_port(device, i, &umad_dev->port[i - s])) + goto err; + } + + ib_set_client_data(device, &umad_client, umad_dev); + + return; + +err: + while (--i >= s) { + class_device_unregister(&umad_dev->port[i - s].class_dev); + class_device_unregister(&umad_dev->port[i - s].sm_class_dev); + } + + kref_put(&umad_dev->ref, ib_umad_release_dev); +} + +static void ib_umad_remove_one(struct ib_device *device) +{ + struct ib_umad_device *umad_dev = ib_get_client_data(device, &umad_client); + int i; + + if (!umad_dev) + return; + + for (i = 0; i <= umad_dev->end_port - umad_dev->start_port; ++i) { + class_device_unregister(&umad_dev->port[i].class_dev); + class_device_unregister(&umad_dev->port[i].sm_class_dev); + } + + kref_put(&umad_dev->ref, ib_umad_release_dev); +} + +static int __init ib_umad_init(void) +{ + int ret; + + spin_lock_init(&map_lock); + + ret = register_chrdev_region(base_dev, IB_UMAD_MAX_PORTS * 2, + "infiniband_mad"); + if (ret) { + printk(KERN_ERR "user_mad: couldn't register device number\n"); + goto out; + } + + ret = class_register(&umad_class); + if (ret) { + printk(KERN_ERR "user_mad: couldn't create class infiniband_mad\n"); + goto out_chrdev; + } + + ret = class_create_file(&umad_class, &class_attr_abi_version); + if (ret) { + printk(KERN_ERR "user_mad: couldn't create abi_version attribute\n"); + goto out_class; + } + + ret = ib_register_client(&umad_client); + if (ret) { + printk(KERN_ERR "user_mad: couldn't register ib_umad client\n"); + goto out_class; + } + + return 0; + +out_class: + class_unregister(&umad_class); + +out_chrdev: + unregister_chrdev_region(base_dev, IB_UMAD_MAX_PORTS * 2); + +out: + return ret; +} + +static void __exit ib_umad_cleanup(void) +{ + ib_unregister_client(&umad_client); + class_unregister(&umad_class); + unregister_chrdev_region(base_dev, IB_UMAD_MAX_PORTS * 2); +} + +module_init(ib_umad_init); +module_exit(ib_umad_cleanup); diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c new file mode 100644 index 000000000000..7c08ed0cd7dd --- /dev/null +++ b/drivers/infiniband/core/verbs.c @@ -0,0 +1,434 @@ +/* + * Copyright (c) 2004 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2004 Infinicon Corporation. All rights reserved. + * Copyright (c) 2004 Intel Corporation. All rights reserved. + * Copyright (c) 2004 Topspin Corporation. All rights reserved. + * Copyright (c) 2004 Voltaire Corporation. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + * + * $Id: verbs.c 1349 2004-12-16 21:09:43Z roland $ + */ + +#include <linux/errno.h> +#include <linux/err.h> + +#include <ib_verbs.h> + +/* Protection domains */ + +struct ib_pd *ib_alloc_pd(struct ib_device *device) +{ + struct ib_pd *pd; + + pd = device->alloc_pd(device); + + if (!IS_ERR(pd)) { + pd->device = device; + atomic_set(&pd->usecnt, 0); + } + + return pd; +} +EXPORT_SYMBOL(ib_alloc_pd); + +int ib_dealloc_pd(struct ib_pd *pd) +{ + if (atomic_read(&pd->usecnt)) + return -EBUSY; + + return pd->device->dealloc_pd(pd); +} +EXPORT_SYMBOL(ib_dealloc_pd); + +/* Address handles */ + +struct ib_ah *ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr) +{ + struct ib_ah *ah; + + ah = pd->device->create_ah(pd, ah_attr); + + if (!IS_ERR(ah)) { + ah->device = pd->device; + ah->pd = pd; + atomic_inc(&pd->usecnt); + } + + return ah; +} +EXPORT_SYMBOL(ib_create_ah); + +int ib_modify_ah(struct ib_ah *ah, struct ib_ah_attr *ah_attr) +{ + return ah->device->modify_ah ? + ah->device->modify_ah(ah, ah_attr) : + -ENOSYS; +} +EXPORT_SYMBOL(ib_modify_ah); + +int ib_query_ah(struct ib_ah *ah, struct ib_ah_attr *ah_attr) +{ + return ah->device->query_ah ? + ah->device->query_ah(ah, ah_attr) : + -ENOSYS; +} +EXPORT_SYMBOL(ib_query_ah); + +int ib_destroy_ah(struct ib_ah *ah) +{ + struct ib_pd *pd; + int ret; + + pd = ah->pd; + ret = ah->device->destroy_ah(ah); + if (!ret) + atomic_dec(&pd->usecnt); + + return ret; +} +EXPORT_SYMBOL(ib_destroy_ah); + +/* Queue pairs */ + +struct ib_qp *ib_create_qp(struct ib_pd *pd, + struct ib_qp_init_attr *qp_init_attr) +{ + struct ib_qp *qp; + + qp = pd->device->create_qp(pd, qp_init_attr); + + if (!IS_ERR(qp)) { + qp->device = pd->device; + qp->pd = pd; + qp->send_cq = qp_init_attr->send_cq; + qp->recv_cq = qp_init_attr->recv_cq; + qp->srq = qp_init_attr->srq; + qp->event_handler = qp_init_attr->event_handler; + qp->qp_context = qp_init_attr->qp_context; + qp->qp_type = qp_init_attr->qp_type; + atomic_inc(&pd->usecnt); + atomic_inc(&qp_init_attr->send_cq->usecnt); + atomic_inc(&qp_init_attr->recv_cq->usecnt); + if (qp_init_attr->srq) + atomic_inc(&qp_init_attr->srq->usecnt); + } + + return qp; +} +EXPORT_SYMBOL(ib_create_qp); + +int ib_modify_qp(struct ib_qp *qp, + struct ib_qp_attr *qp_attr, + int qp_attr_mask) +{ + return qp->device->modify_qp(qp, qp_attr, qp_attr_mask); +} +EXPORT_SYMBOL(ib_modify_qp); + +int ib_query_qp(struct ib_qp *qp, + struct ib_qp_attr *qp_attr, + int qp_attr_mask, + struct ib_qp_init_attr *qp_init_attr) +{ + return qp->device->query_qp ? + qp->device->query_qp(qp, qp_attr, qp_attr_mask, qp_init_attr) : + -ENOSYS; +} +EXPORT_SYMBOL(ib_query_qp); + +int ib_destroy_qp(struct ib_qp *qp) +{ + struct ib_pd *pd; + struct ib_cq *scq, *rcq; + struct ib_srq *srq; + int ret; + + pd = qp->pd; + scq = qp->send_cq; + rcq = qp->recv_cq; + srq = qp->srq; + + ret = qp->device->destroy_qp(qp); + if (!ret) { + atomic_dec(&pd->usecnt); + atomic_dec(&scq->usecnt); + atomic_dec(&rcq->usecnt); + if (srq) + atomic_dec(&srq->usecnt); + } + + return ret; +} +EXPORT_SYMBOL(ib_destroy_qp); + +/* Completion queues */ + +struct ib_cq *ib_create_cq(struct ib_device *device, + ib_comp_handler comp_handler, + void (*event_handler)(struct ib_event *, void *), + void *cq_context, int cqe) +{ + struct ib_cq *cq; + + cq = device->create_cq(device, cqe); + + if (!IS_ERR(cq)) { + cq->device = device; + cq->comp_handler = comp_handler; + cq->event_handler = event_handler; + cq->cq_context = cq_context; + atomic_set(&cq->usecnt, 0); + } + + return cq; +} +EXPORT_SYMBOL(ib_create_cq); + +int ib_destroy_cq(struct ib_cq *cq) +{ + if (atomic_read(&cq->usecnt)) + return -EBUSY; + + return cq->device->destroy_cq(cq); +} +EXPORT_SYMBOL(ib_destroy_cq); + +int ib_resize_cq(struct ib_cq *cq, + int cqe) +{ + int ret; + + if (!cq->device->resize_cq) + return -ENOSYS; + + ret = cq->device->resize_cq(cq, &cqe); + if (!ret) + cq->cqe = cqe; + + return ret; +} +EXPORT_SYMBOL(ib_resize_cq); + +/* Memory regions */ + +struct ib_mr *ib_get_dma_mr(struct ib_pd *pd, int mr_access_flags) +{ + struct ib_mr *mr; + + mr = pd->device->get_dma_mr(pd, mr_access_flags); + + if (!IS_ERR(mr)) { + mr->device = pd->device; + mr->pd = pd; + atomic_inc(&pd->usecnt); + atomic_set(&mr->usecnt, 0); + } + + return mr; +} +EXPORT_SYMBOL(ib_get_dma_mr); + +struct ib_mr *ib_reg_phys_mr(struct ib_pd *pd, + struct ib_phys_buf *phys_buf_array, + int num_phys_buf, + int mr_access_flags, + u64 *iova_start) +{ + struct ib_mr *mr; + + mr = pd->device->reg_phys_mr(pd, phys_buf_array, num_phys_buf, + mr_access_flags, iova_start); + + if (!IS_ERR(mr)) { + mr->device = pd->device; + mr->pd = pd; + atomic_inc(&pd->usecnt); + atomic_set(&mr->usecnt, 0); + } + + return mr; +} +EXPORT_SYMBOL(ib_reg_phys_mr); + +int ib_rereg_phys_mr(struct ib_mr *mr, + int mr_rereg_mask, + struct ib_pd *pd, + struct ib_phys_buf *phys_buf_array, + int num_phys_buf, + int mr_access_flags, + u64 *iova_start) +{ + struct ib_pd *old_pd; + int ret; + + if (!mr->device->rereg_phys_mr) + return -ENOSYS; + + if (atomic_read(&mr->usecnt)) + return -EBUSY; + + old_pd = mr->pd; + + ret = mr->device->rereg_phys_mr(mr, mr_rereg_mask, pd, + phys_buf_array, num_phys_buf, + mr_access_flags, iova_start); + + if (!ret && (mr_rereg_mask & IB_MR_REREG_PD)) { + atomic_dec(&old_pd->usecnt); + atomic_inc(&pd->usecnt); + } + + return ret; +} +EXPORT_SYMBOL(ib_rereg_phys_mr); + +int ib_query_mr(struct ib_mr *mr, struct ib_mr_attr *mr_attr) +{ + return mr->device->query_mr ? + mr->device->query_mr(mr, mr_attr) : -ENOSYS; +} +EXPORT_SYMBOL(ib_query_mr); + +int ib_dereg_mr(struct ib_mr *mr) +{ + struct ib_pd *pd; + int ret; + + if (atomic_read(&mr->usecnt)) + return -EBUSY; + + pd = mr->pd; + ret = mr->device->dereg_mr(mr); + if (!ret) + atomic_dec(&pd->usecnt); + + return ret; +} +EXPORT_SYMBOL(ib_dereg_mr); + +/* Memory windows */ + +struct ib_mw *ib_alloc_mw(struct ib_pd *pd) +{ + struct ib_mw *mw; + + if (!pd->device->alloc_mw) + return ERR_PTR(-ENOSYS); + + mw = pd->device->alloc_mw(pd); + if (!IS_ERR(mw)) { + mw->device = pd->device; + mw->pd = pd; + atomic_inc(&pd->usecnt); + } + + return mw; +} +EXPORT_SYMBOL(ib_alloc_mw); + +int ib_dealloc_mw(struct ib_mw *mw) +{ + struct ib_pd *pd; + int ret; + + pd = mw->pd; + ret = mw->device->dealloc_mw(mw); + if (!ret) + atomic_dec(&pd->usecnt); + + return ret; +} +EXPORT_SYMBOL(ib_dealloc_mw); + +/* "Fast" memory regions */ + +struct ib_fmr *ib_alloc_fmr(struct ib_pd *pd, + int mr_access_flags, + struct ib_fmr_attr *fmr_attr) +{ + struct ib_fmr *fmr; + + if (!pd->device->alloc_fmr) + return ERR_PTR(-ENOSYS); + + fmr = pd->device->alloc_fmr(pd, mr_access_flags, fmr_attr); + if (!IS_ERR(fmr)) { + fmr->device = pd->device; + fmr->pd = pd; + atomic_inc(&pd->usecnt); + } + + return fmr; +} +EXPORT_SYMBOL(ib_alloc_fmr); + +int ib_unmap_fmr(struct list_head *fmr_list) +{ + struct ib_fmr *fmr; + + if (list_empty(fmr_list)) + return 0; + + fmr = list_entry(fmr_list->next, struct ib_fmr, list); + return fmr->device->unmap_fmr(fmr_list); +} +EXPORT_SYMBOL(ib_unmap_fmr); + +int ib_dealloc_fmr(struct ib_fmr *fmr) +{ + struct ib_pd *pd; + int ret; + + pd = fmr->pd; + ret = fmr->device->dealloc_fmr(fmr); + if (!ret) + atomic_dec(&pd->usecnt); + + return ret; +} +EXPORT_SYMBOL(ib_dealloc_fmr); + +/* Multicast groups */ + +int ib_attach_mcast(struct ib_qp *qp, union ib_gid *gid, u16 lid) +{ + return qp->device->attach_mcast ? + qp->device->attach_mcast(qp, gid, lid) : + -ENOSYS; +} +EXPORT_SYMBOL(ib_attach_mcast); + +int ib_detach_mcast(struct ib_qp *qp, union ib_gid *gid, u16 lid) +{ + return qp->device->detach_mcast ? + qp->device->detach_mcast(qp, gid, lid) : + -ENOSYS; +} +EXPORT_SYMBOL(ib_detach_mcast); |