summaryrefslogtreecommitdiffstats
path: root/drivers/infiniband/sw
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/infiniband/sw')
-rw-r--r--drivers/infiniband/sw/Makefile1
-rw-r--r--drivers/infiniband/sw/rdmavt/Kconfig6
-rw-r--r--drivers/infiniband/sw/rdmavt/Makefile13
-rw-r--r--drivers/infiniband/sw/rdmavt/ah.c196
-rw-r--r--drivers/infiniband/sw/rdmavt/ah.h59
-rw-r--r--drivers/infiniband/sw/rdmavt/cq.c551
-rw-r--r--drivers/infiniband/sw/rdmavt/cq.h64
-rw-r--r--drivers/infiniband/sw/rdmavt/dma.c184
-rw-r--r--drivers/infiniband/sw/rdmavt/dma.h53
-rw-r--r--drivers/infiniband/sw/rdmavt/mad.c171
-rw-r--r--drivers/infiniband/sw/rdmavt/mad.h60
-rw-r--r--drivers/infiniband/sw/rdmavt/mcast.c417
-rw-r--r--drivers/infiniband/sw/rdmavt/mcast.h58
-rw-r--r--drivers/infiniband/sw/rdmavt/mmap.c208
-rw-r--r--drivers/infiniband/sw/rdmavt/mmap.h63
-rw-r--r--drivers/infiniband/sw/rdmavt/mr.c830
-rw-r--r--drivers/infiniband/sw/rdmavt/mr.h92
-rw-r--r--drivers/infiniband/sw/rdmavt/pd.c119
-rw-r--r--drivers/infiniband/sw/rdmavt/pd.h58
-rw-r--r--drivers/infiniband/sw/rdmavt/qp.c1696
-rw-r--r--drivers/infiniband/sw/rdmavt/qp.h69
-rw-r--r--drivers/infiniband/sw/rdmavt/srq.c353
-rw-r--r--drivers/infiniband/sw/rdmavt/srq.h62
-rw-r--r--drivers/infiniband/sw/rdmavt/trace.c49
-rw-r--r--drivers/infiniband/sw/rdmavt/trace.h187
-rw-r--r--drivers/infiniband/sw/rdmavt/vt.c873
-rw-r--r--drivers/infiniband/sw/rdmavt/vt.h104
27 files changed, 6596 insertions, 0 deletions
diff --git a/drivers/infiniband/sw/Makefile b/drivers/infiniband/sw/Makefile
new file mode 100644
index 000000000000..988b6a0101a4
--- /dev/null
+++ b/drivers/infiniband/sw/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_INFINIBAND_RDMAVT) += rdmavt/
diff --git a/drivers/infiniband/sw/rdmavt/Kconfig b/drivers/infiniband/sw/rdmavt/Kconfig
new file mode 100644
index 000000000000..11aa6a34bd71
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/Kconfig
@@ -0,0 +1,6 @@
+config INFINIBAND_RDMAVT
+ tristate "RDMA verbs transport library"
+ depends on 64BIT
+ default m
+ ---help---
+ This is a common software verbs provider for RDMA networks.
diff --git a/drivers/infiniband/sw/rdmavt/Makefile b/drivers/infiniband/sw/rdmavt/Makefile
new file mode 100644
index 000000000000..ccaa7992ac97
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/Makefile
@@ -0,0 +1,13 @@
+#
+# rdmavt driver
+#
+#
+#
+# Called from the kernel module build system.
+#
+obj-$(CONFIG_INFINIBAND_RDMAVT) += rdmavt.o
+
+rdmavt-y := vt.o ah.o cq.o dma.o mad.o mcast.o mmap.o mr.o pd.o qp.o srq.o \
+ trace.o
+
+CFLAGS_trace.o = -I$(src)
diff --git a/drivers/infiniband/sw/rdmavt/ah.c b/drivers/infiniband/sw/rdmavt/ah.c
new file mode 100644
index 000000000000..16c446142c2a
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/ah.c
@@ -0,0 +1,196 @@
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <linux/slab.h>
+#include "ah.h"
+#include "vt.h" /* for prints */
+
+/**
+ * rvt_check_ah - validate the attributes of AH
+ * @ibdev: the ib device
+ * @ah_attr: the attributes of the AH
+ *
+ * If driver supports a more detailed check_ah function call back to it
+ * otherwise just check the basics.
+ *
+ * Return: 0 on success
+ */
+int rvt_check_ah(struct ib_device *ibdev,
+ struct ib_ah_attr *ah_attr)
+{
+ int err;
+ struct ib_port_attr port_attr;
+ struct rvt_dev_info *rdi = ib_to_rvt(ibdev);
+ enum rdma_link_layer link = rdma_port_get_link_layer(ibdev,
+ ah_attr->port_num);
+
+ err = ib_query_port(ibdev, ah_attr->port_num, &port_attr);
+ if (err)
+ return -EINVAL;
+ if (ah_attr->port_num < 1 ||
+ ah_attr->port_num > ibdev->phys_port_cnt)
+ return -EINVAL;
+ if (ah_attr->static_rate != IB_RATE_PORT_CURRENT &&
+ ib_rate_to_mbps(ah_attr->static_rate) < 0)
+ return -EINVAL;
+ if ((ah_attr->ah_flags & IB_AH_GRH) &&
+ ah_attr->grh.sgid_index >= port_attr.gid_tbl_len)
+ return -EINVAL;
+ if (link != IB_LINK_LAYER_ETHERNET) {
+ if (ah_attr->dlid == 0)
+ return -EINVAL;
+ if (ah_attr->dlid >= be16_to_cpu(IB_MULTICAST_LID_BASE) &&
+ ah_attr->dlid != be16_to_cpu(IB_LID_PERMISSIVE) &&
+ !(ah_attr->ah_flags & IB_AH_GRH))
+ return -EINVAL;
+ }
+ if (rdi->driver_f.check_ah)
+ return rdi->driver_f.check_ah(ibdev, ah_attr);
+ return 0;
+}
+EXPORT_SYMBOL(rvt_check_ah);
+
+/**
+ * rvt_create_ah - create an address handle
+ * @pd: the protection domain
+ * @ah_attr: the attributes of the AH
+ *
+ * This may be called from interrupt context.
+ *
+ * Return: newly allocated ah
+ */
+struct ib_ah *rvt_create_ah(struct ib_pd *pd,
+ struct ib_ah_attr *ah_attr)
+{
+ struct rvt_ah *ah;
+ struct rvt_dev_info *dev = ib_to_rvt(pd->device);
+ unsigned long flags;
+
+ if (rvt_check_ah(pd->device, ah_attr))
+ return ERR_PTR(-EINVAL);
+
+ ah = kmalloc(sizeof(*ah), GFP_ATOMIC);
+ if (!ah)
+ return ERR_PTR(-ENOMEM);
+
+ spin_lock_irqsave(&dev->n_ahs_lock, flags);
+ if (dev->n_ahs_allocated == dev->dparms.props.max_ah) {
+ spin_unlock(&dev->n_ahs_lock);
+ kfree(ah);
+ return ERR_PTR(-ENOMEM);
+ }
+
+ dev->n_ahs_allocated++;
+ spin_unlock_irqrestore(&dev->n_ahs_lock, flags);
+
+ ah->attr = *ah_attr;
+ atomic_set(&ah->refcount, 0);
+
+ if (dev->driver_f.notify_new_ah)
+ dev->driver_f.notify_new_ah(pd->device, ah_attr, ah);
+
+ return &ah->ibah;
+}
+
+/**
+ * rvt_destory_ah - Destory an address handle
+ * @ibah: address handle
+ *
+ * Return: 0 on success
+ */
+int rvt_destroy_ah(struct ib_ah *ibah)
+{
+ struct rvt_dev_info *dev = ib_to_rvt(ibah->device);
+ struct rvt_ah *ah = ibah_to_rvtah(ibah);
+ unsigned long flags;
+
+ if (atomic_read(&ah->refcount) != 0)
+ return -EBUSY;
+
+ spin_lock_irqsave(&dev->n_ahs_lock, flags);
+ dev->n_ahs_allocated--;
+ spin_unlock_irqrestore(&dev->n_ahs_lock, flags);
+
+ kfree(ah);
+
+ return 0;
+}
+
+/**
+ * rvt_modify_ah - modify an ah with given attrs
+ * @ibah: address handle to modify
+ * @ah_attr: attrs to apply
+ *
+ * Return: 0 on success
+ */
+int rvt_modify_ah(struct ib_ah *ibah, struct ib_ah_attr *ah_attr)
+{
+ struct rvt_ah *ah = ibah_to_rvtah(ibah);
+
+ if (rvt_check_ah(ibah->device, ah_attr))
+ return -EINVAL;
+
+ ah->attr = *ah_attr;
+
+ return 0;
+}
+
+/**
+ * rvt_query_ah - return attrs for ah
+ * @ibah: address handle to query
+ * @ah_attr: return info in this
+ *
+ * Return: always 0
+ */
+int rvt_query_ah(struct ib_ah *ibah, struct ib_ah_attr *ah_attr)
+{
+ struct rvt_ah *ah = ibah_to_rvtah(ibah);
+
+ *ah_attr = ah->attr;
+
+ return 0;
+}
diff --git a/drivers/infiniband/sw/rdmavt/ah.h b/drivers/infiniband/sw/rdmavt/ah.h
new file mode 100644
index 000000000000..e9c36be87d79
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/ah.h
@@ -0,0 +1,59 @@
+#ifndef DEF_RVTAH_H
+#define DEF_RVTAH_H
+
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <rdma/rdma_vt.h>
+
+struct ib_ah *rvt_create_ah(struct ib_pd *pd,
+ struct ib_ah_attr *ah_attr);
+int rvt_destroy_ah(struct ib_ah *ibah);
+int rvt_modify_ah(struct ib_ah *ibah, struct ib_ah_attr *ah_attr);
+int rvt_query_ah(struct ib_ah *ibah, struct ib_ah_attr *ah_attr);
+
+#endif /* DEF_RVTAH_H */
diff --git a/drivers/infiniband/sw/rdmavt/cq.c b/drivers/infiniband/sw/rdmavt/cq.c
new file mode 100644
index 000000000000..b1ffc8b4a6c0
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/cq.c
@@ -0,0 +1,551 @@
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/kthread.h>
+#include "cq.h"
+#include "vt.h"
+
+/**
+ * rvt_cq_enter - add a new entry to the completion queue
+ * @cq: completion queue
+ * @entry: work completion entry to add
+ * @sig: true if @entry is solicited
+ *
+ * This may be called with qp->s_lock held.
+ */
+void rvt_cq_enter(struct rvt_cq *cq, struct ib_wc *entry, bool solicited)
+{
+ struct rvt_cq_wc *wc;
+ unsigned long flags;
+ u32 head;
+ u32 next;
+
+ spin_lock_irqsave(&cq->lock, flags);
+
+ /*
+ * Note that the head pointer might be writable by user processes.
+ * Take care to verify it is a sane value.
+ */
+ wc = cq->queue;
+ head = wc->head;
+ if (head >= (unsigned)cq->ibcq.cqe) {
+ head = cq->ibcq.cqe;
+ next = 0;
+ } else {
+ next = head + 1;
+ }
+
+ if (unlikely(next == wc->tail)) {
+ spin_unlock_irqrestore(&cq->lock, flags);
+ if (cq->ibcq.event_handler) {
+ struct ib_event ev;
+
+ ev.device = cq->ibcq.device;
+ ev.element.cq = &cq->ibcq;
+ ev.event = IB_EVENT_CQ_ERR;
+ cq->ibcq.event_handler(&ev, cq->ibcq.cq_context);
+ }
+ return;
+ }
+ if (cq->ip) {
+ wc->uqueue[head].wr_id = entry->wr_id;
+ wc->uqueue[head].status = entry->status;
+ wc->uqueue[head].opcode = entry->opcode;
+ wc->uqueue[head].vendor_err = entry->vendor_err;
+ wc->uqueue[head].byte_len = entry->byte_len;
+ wc->uqueue[head].ex.imm_data =
+ (__u32 __force)entry->ex.imm_data;
+ wc->uqueue[head].qp_num = entry->qp->qp_num;
+ wc->uqueue[head].src_qp = entry->src_qp;
+ wc->uqueue[head].wc_flags = entry->wc_flags;
+ wc->uqueue[head].pkey_index = entry->pkey_index;
+ wc->uqueue[head].slid = entry->slid;
+ wc->uqueue[head].sl = entry->sl;
+ wc->uqueue[head].dlid_path_bits = entry->dlid_path_bits;
+ wc->uqueue[head].port_num = entry->port_num;
+ /* Make sure entry is written before the head index. */
+ smp_wmb();
+ } else {
+ wc->kqueue[head] = *entry;
+ }
+ wc->head = next;
+
+ if (cq->notify == IB_CQ_NEXT_COMP ||
+ (cq->notify == IB_CQ_SOLICITED &&
+ (solicited || entry->status != IB_WC_SUCCESS))) {
+ struct kthread_worker *worker;
+ /*
+ * This will cause send_complete() to be called in
+ * another thread.
+ */
+ smp_read_barrier_depends(); /* see rvt_cq_exit */
+ worker = cq->rdi->worker;
+ if (likely(worker)) {
+ cq->notify = RVT_CQ_NONE;
+ cq->triggered++;
+ queue_kthread_work(worker, &cq->comptask);
+ }
+ }
+
+ spin_unlock_irqrestore(&cq->lock, flags);
+}
+EXPORT_SYMBOL(rvt_cq_enter);
+
+static void send_complete(struct kthread_work *work)
+{
+ struct rvt_cq *cq = container_of(work, struct rvt_cq, comptask);
+
+ /*
+ * The completion handler will most likely rearm the notification
+ * and poll for all pending entries. If a new completion entry
+ * is added while we are in this routine, queue_work()
+ * won't call us again until we return so we check triggered to
+ * see if we need to call the handler again.
+ */
+ for (;;) {
+ u8 triggered = cq->triggered;
+
+ /*
+ * IPoIB connected mode assumes the callback is from a
+ * soft IRQ. We simulate this by blocking "bottom halves".
+ * See the implementation for ipoib_cm_handle_tx_wc(),
+ * netif_tx_lock_bh() and netif_tx_lock().
+ */
+ local_bh_disable();
+ cq->ibcq.comp_handler(&cq->ibcq, cq->ibcq.cq_context);
+ local_bh_enable();
+
+ if (cq->triggered == triggered)
+ return;
+ }
+}
+
+/**
+ * rvt_create_cq - create a completion queue
+ * @ibdev: the device this completion queue is attached to
+ * @attr: creation attributes
+ * @context: unused by the QLogic_IB driver
+ * @udata: user data for libibverbs.so
+ *
+ * Called by ib_create_cq() in the generic verbs code.
+ *
+ * Return: pointer to the completion queue or negative errno values
+ * for failure.
+ */
+struct ib_cq *rvt_create_cq(struct ib_device *ibdev,
+ const struct ib_cq_init_attr *attr,
+ struct ib_ucontext *context,
+ struct ib_udata *udata)
+{
+ struct rvt_dev_info *rdi = ib_to_rvt(ibdev);
+ struct rvt_cq *cq;
+ struct rvt_cq_wc *wc;
+ struct ib_cq *ret;
+ u32 sz;
+ unsigned int entries = attr->cqe;
+
+ if (attr->flags)
+ return ERR_PTR(-EINVAL);
+
+ if (entries < 1 || entries > rdi->dparms.props.max_cqe)
+ return ERR_PTR(-EINVAL);
+
+ /* Allocate the completion queue structure. */
+ cq = kzalloc(sizeof(*cq), GFP_KERNEL);
+ if (!cq)
+ return ERR_PTR(-ENOMEM);
+
+ /*
+ * Allocate the completion queue entries and head/tail pointers.
+ * This is allocated separately so that it can be resized and
+ * also mapped into user space.
+ * We need to use vmalloc() in order to support mmap and large
+ * numbers of entries.
+ */
+ sz = sizeof(*wc);
+ if (udata && udata->outlen >= sizeof(__u64))
+ sz += sizeof(struct ib_uverbs_wc) * (entries + 1);
+ else
+ sz += sizeof(struct ib_wc) * (entries + 1);
+ wc = vmalloc_user(sz);
+ if (!wc) {
+ ret = ERR_PTR(-ENOMEM);
+ goto bail_cq;
+ }
+
+ /*
+ * Return the address of the WC as the offset to mmap.
+ * See rvt_mmap() for details.
+ */
+ if (udata && udata->outlen >= sizeof(__u64)) {
+ int err;
+
+ cq->ip = rvt_create_mmap_info(rdi, sz, context, wc);
+ if (!cq->ip) {
+ ret = ERR_PTR(-ENOMEM);
+ goto bail_wc;
+ }
+
+ err = ib_copy_to_udata(udata, &cq->ip->offset,
+ sizeof(cq->ip->offset));
+ if (err) {
+ ret = ERR_PTR(err);
+ goto bail_ip;
+ }
+ }
+
+ spin_lock(&rdi->n_cqs_lock);
+ if (rdi->n_cqs_allocated == rdi->dparms.props.max_cq) {
+ spin_unlock(&rdi->n_cqs_lock);
+ ret = ERR_PTR(-ENOMEM);
+ goto bail_ip;
+ }
+
+ rdi->n_cqs_allocated++;
+ spin_unlock(&rdi->n_cqs_lock);
+
+ if (cq->ip) {
+ spin_lock_irq(&rdi->pending_lock);
+ list_add(&cq->ip->pending_mmaps, &rdi->pending_mmaps);
+ spin_unlock_irq(&rdi->pending_lock);
+ }
+
+ /*
+ * ib_create_cq() will initialize cq->ibcq except for cq->ibcq.cqe.
+ * The number of entries should be >= the number requested or return
+ * an error.
+ */
+ cq->rdi = rdi;
+ cq->ibcq.cqe = entries;
+ cq->notify = RVT_CQ_NONE;
+ spin_lock_init(&cq->lock);
+ init_kthread_work(&cq->comptask, send_complete);
+ cq->queue = wc;
+
+ ret = &cq->ibcq;
+
+ goto done;
+
+bail_ip:
+ kfree(cq->ip);
+bail_wc:
+ vfree(wc);
+bail_cq:
+ kfree(cq);
+done:
+ return ret;
+}
+
+/**
+ * rvt_destroy_cq - destroy a completion queue
+ * @ibcq: the completion queue to destroy.
+ *
+ * Called by ib_destroy_cq() in the generic verbs code.
+ *
+ * Return: always 0
+ */
+int rvt_destroy_cq(struct ib_cq *ibcq)
+{
+ struct rvt_cq *cq = ibcq_to_rvtcq(ibcq);
+ struct rvt_dev_info *rdi = cq->rdi;
+
+ flush_kthread_work(&cq->comptask);
+ spin_lock(&rdi->n_cqs_lock);
+ rdi->n_cqs_allocated--;
+ spin_unlock(&rdi->n_cqs_lock);
+ if (cq->ip)
+ kref_put(&cq->ip->ref, rvt_release_mmap_info);
+ else
+ vfree(cq->queue);
+ kfree(cq);
+
+ return 0;
+}
+
+/**
+ * rvt_req_notify_cq - change the notification type for a completion queue
+ * @ibcq: the completion queue
+ * @notify_flags: the type of notification to request
+ *
+ * This may be called from interrupt context. Also called by
+ * ib_req_notify_cq() in the generic verbs code.
+ *
+ * Return: 0 for success.
+ */
+int rvt_req_notify_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags notify_flags)
+{
+ struct rvt_cq *cq = ibcq_to_rvtcq(ibcq);
+ unsigned long flags;
+ int ret = 0;
+
+ spin_lock_irqsave(&cq->lock, flags);
+ /*
+ * Don't change IB_CQ_NEXT_COMP to IB_CQ_SOLICITED but allow
+ * any other transitions (see C11-31 and C11-32 in ch. 11.4.2.2).
+ */
+ if (cq->notify != IB_CQ_NEXT_COMP)
+ cq->notify = notify_flags & IB_CQ_SOLICITED_MASK;
+
+ if ((notify_flags & IB_CQ_REPORT_MISSED_EVENTS) &&
+ cq->queue->head != cq->queue->tail)
+ ret = 1;
+
+ spin_unlock_irqrestore(&cq->lock, flags);
+
+ return ret;
+}
+
+/**
+ * rvt_resize_cq - change the size of the CQ
+ * @ibcq: the completion queue
+ *
+ * Return: 0 for success.
+ */
+int rvt_resize_cq(struct ib_cq *ibcq, int cqe, struct ib_udata *udata)
+{
+ struct rvt_cq *cq = ibcq_to_rvtcq(ibcq);
+ struct rvt_cq_wc *old_wc;
+ struct rvt_cq_wc *wc;
+ u32 head, tail, n;
+ int ret;
+ u32 sz;
+ struct rvt_dev_info *rdi = cq->rdi;
+
+ if (cqe < 1 || cqe > rdi->dparms.props.max_cqe)
+ return -EINVAL;
+
+ /*
+ * Need to use vmalloc() if we want to support large #s of entries.
+ */
+ sz = sizeof(*wc);
+ if (udata && udata->outlen >= sizeof(__u64))
+ sz += sizeof(struct ib_uverbs_wc) * (cqe + 1);
+ else
+ sz += sizeof(struct ib_wc) * (cqe + 1);
+ wc = vmalloc_user(sz);
+ if (!wc)
+ return -ENOMEM;
+
+ /* Check that we can write the offset to mmap. */
+ if (udata && udata->outlen >= sizeof(__u64)) {
+ __u64 offset = 0;
+
+ ret = ib_copy_to_udata(udata, &offset, sizeof(offset));
+ if (ret)
+ goto bail_free;
+ }
+
+ spin_lock_irq(&cq->lock);
+ /*
+ * Make sure head and tail are sane since they
+ * might be user writable.
+ */
+ old_wc = cq->queue;
+ head = old_wc->head;
+ if (head > (u32)cq->ibcq.cqe)
+ head = (u32)cq->ibcq.cqe;
+ tail = old_wc->tail;
+ if (tail > (u32)cq->ibcq.cqe)
+ tail = (u32)cq->ibcq.cqe;
+ if (head < tail)
+ n = cq->ibcq.cqe + 1 + head - tail;
+ else
+ n = head - tail;
+ if (unlikely((u32)cqe < n)) {
+ ret = -EINVAL;
+ goto bail_unlock;
+ }
+ for (n = 0; tail != head; n++) {
+ if (cq->ip)
+ wc->uqueue[n] = old_wc->uqueue[tail];
+ else
+ wc->kqueue[n] = old_wc->kqueue[tail];
+ if (tail == (u32)cq->ibcq.cqe)
+ tail = 0;
+ else
+ tail++;
+ }
+ cq->ibcq.cqe = cqe;
+ wc->head = n;
+ wc->tail = 0;
+ cq->queue = wc;
+ spin_unlock_irq(&cq->lock);
+
+ vfree(old_wc);
+
+ if (cq->ip) {
+ struct rvt_mmap_info *ip = cq->ip;
+
+ rvt_update_mmap_info(rdi, ip, sz, wc);
+
+ /*
+ * Return the offset to mmap.
+ * See rvt_mmap() for details.
+ */
+ if (udata && udata->outlen >= sizeof(__u64)) {
+ ret = ib_copy_to_udata(udata, &ip->offset,
+ sizeof(ip->offset));
+ if (ret)
+ return ret;
+ }
+
+ spin_lock_irq(&rdi->pending_lock);
+ if (list_empty(&ip->pending_mmaps))
+ list_add(&ip->pending_mmaps, &rdi->pending_mmaps);
+ spin_unlock_irq(&rdi->pending_lock);
+ }
+
+ return 0;
+
+bail_unlock:
+ spin_unlock_irq(&cq->lock);
+bail_free:
+ vfree(wc);
+ return ret;
+}
+
+/**
+ * rvt_poll_cq - poll for work completion entries
+ * @ibcq: the completion queue to poll
+ * @num_entries: the maximum number of entries to return
+ * @entry: pointer to array where work completions are placed
+ *
+ * This may be called from interrupt context. Also called by ib_poll_cq()
+ * in the generic verbs code.
+ *
+ * Return: the number of completion entries polled.
+ */
+int rvt_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *entry)
+{
+ struct rvt_cq *cq = ibcq_to_rvtcq(ibcq);
+ struct rvt_cq_wc *wc;
+ unsigned long flags;
+ int npolled;
+ u32 tail;
+
+ /* The kernel can only poll a kernel completion queue */
+ if (cq->ip)
+ return -EINVAL;
+
+ spin_lock_irqsave(&cq->lock, flags);
+
+ wc = cq->queue;
+ tail = wc->tail;
+ if (tail > (u32)cq->ibcq.cqe)
+ tail = (u32)cq->ibcq.cqe;
+ for (npolled = 0; npolled < num_entries; ++npolled, ++entry) {
+ if (tail == wc->head)
+ break;
+ /* The kernel doesn't need a RMB since it has the lock. */
+ *entry = wc->kqueue[tail];
+ if (tail >= cq->ibcq.cqe)
+ tail = 0;
+ else
+ tail++;
+ }
+ wc->tail = tail;
+
+ spin_unlock_irqrestore(&cq->lock, flags);
+
+ return npolled;
+}
+
+/**
+ * rvt_driver_cq_init - Init cq resources on behalf of driver
+ * @rdi: rvt dev structure
+ *
+ * Return: 0 on success
+ */
+int rvt_driver_cq_init(struct rvt_dev_info *rdi)
+{
+ int ret = 0;
+ int cpu;
+ struct task_struct *task;
+
+ if (rdi->worker)
+ return 0;
+ rdi->worker = kzalloc(sizeof(*rdi->worker), GFP_KERNEL);
+ if (!rdi->worker)
+ return -ENOMEM;
+ init_kthread_worker(rdi->worker);
+ task = kthread_create_on_node(
+ kthread_worker_fn,
+ rdi->worker,
+ rdi->dparms.node,
+ "%s", rdi->dparms.cq_name);
+ if (IS_ERR(task)) {
+ kfree(rdi->worker);
+ rdi->worker = NULL;
+ return PTR_ERR(task);
+ }
+
+ cpu = cpumask_first(cpumask_of_node(rdi->dparms.node));
+ kthread_bind(task, cpu);
+ wake_up_process(task);
+ return ret;
+}
+
+/**
+ * rvt_cq_exit - tear down cq reources
+ * @rdi: rvt dev structure
+ */
+void rvt_cq_exit(struct rvt_dev_info *rdi)
+{
+ struct kthread_worker *worker;
+
+ worker = rdi->worker;
+ if (!worker)
+ return;
+ /* blocks future queuing from send_complete() */
+ rdi->worker = NULL;
+ smp_wmb(); /* See rdi_cq_enter */
+ flush_kthread_worker(worker);
+ kthread_stop(worker->task);
+ kfree(worker);
+}
diff --git a/drivers/infiniband/sw/rdmavt/cq.h b/drivers/infiniband/sw/rdmavt/cq.h
new file mode 100644
index 000000000000..6182c29eff66
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/cq.h
@@ -0,0 +1,64 @@
+#ifndef DEF_RVTCQ_H
+#define DEF_RVTCQ_H
+
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <rdma/rdma_vt.h>
+#include <rdma/rdmavt_cq.h>
+
+struct ib_cq *rvt_create_cq(struct ib_device *ibdev,
+ const struct ib_cq_init_attr *attr,
+ struct ib_ucontext *context,
+ struct ib_udata *udata);
+int rvt_destroy_cq(struct ib_cq *ibcq);
+int rvt_req_notify_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags notify_flags);
+int rvt_resize_cq(struct ib_cq *ibcq, int cqe, struct ib_udata *udata);
+int rvt_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *entry);
+int rvt_driver_cq_init(struct rvt_dev_info *rdi);
+void rvt_cq_exit(struct rvt_dev_info *rdi);
+#endif /* DEF_RVTCQ_H */
diff --git a/drivers/infiniband/sw/rdmavt/dma.c b/drivers/infiniband/sw/rdmavt/dma.c
new file mode 100644
index 000000000000..33076a5eee2f
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/dma.c
@@ -0,0 +1,184 @@
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+#include <linux/types.h>
+#include <linux/scatterlist.h>
+#include <rdma/ib_verbs.h>
+
+#include "dma.h"
+
+#define BAD_DMA_ADDRESS ((u64)0)
+
+/*
+ * The following functions implement driver specific replacements
+ * for the ib_dma_*() functions.
+ *
+ * These functions return kernel virtual addresses instead of
+ * device bus addresses since the driver uses the CPU to copy
+ * data instead of using hardware DMA.
+ */
+
+static int rvt_mapping_error(struct ib_device *dev, u64 dma_addr)
+{
+ return dma_addr == BAD_DMA_ADDRESS;
+}
+
+static u64 rvt_dma_map_single(struct ib_device *dev, void *cpu_addr,
+ size_t size, enum dma_data_direction direction)
+{
+ if (WARN_ON(!valid_dma_direction(direction)))
+ return BAD_DMA_ADDRESS;
+
+ return (u64)cpu_addr;
+}
+
+static void rvt_dma_unmap_single(struct ib_device *dev, u64 addr, size_t size,
+ enum dma_data_direction direction)
+{
+ /* This is a stub, nothing to be done here */
+}
+
+static u64 rvt_dma_map_page(struct ib_device *dev, struct page *page,
+ unsigned long offset, size_t size,
+ enum dma_data_direction direction)
+{
+ u64 addr;
+
+ if (WARN_ON(!valid_dma_direction(direction)))
+ return BAD_DMA_ADDRESS;
+
+ if (offset + size > PAGE_SIZE)
+ return BAD_DMA_ADDRESS;
+
+ addr = (u64)page_address(page);
+ if (addr)
+ addr += offset;
+
+ return addr;
+}
+
+static void rvt_dma_unmap_page(struct ib_device *dev, u64 addr, size_t size,
+ enum dma_data_direction direction)
+{
+ /* This is a stub, nothing to be done here */
+}
+
+static int rvt_map_sg(struct ib_device *dev, struct scatterlist *sgl,
+ int nents, enum dma_data_direction direction)
+{
+ struct scatterlist *sg;
+ u64 addr;
+ int i;
+ int ret = nents;
+
+ if (WARN_ON(!valid_dma_direction(direction)))
+ return 0;
+
+ for_each_sg(sgl, sg, nents, i) {
+ addr = (u64)page_address(sg_page(sg));
+ if (!addr) {
+ ret = 0;
+ break;
+ }
+ sg->dma_address = addr + sg->offset;
+#ifdef CONFIG_NEED_SG_DMA_LENGTH
+ sg->dma_length = sg->length;
+#endif
+ }
+ return ret;
+}
+
+static void rvt_unmap_sg(struct ib_device *dev,
+ struct scatterlist *sg, int nents,
+ enum dma_data_direction direction)
+{
+ /* This is a stub, nothing to be done here */
+}
+
+static void rvt_sync_single_for_cpu(struct ib_device *dev, u64 addr,
+ size_t size, enum dma_data_direction dir)
+{
+}
+
+static void rvt_sync_single_for_device(struct ib_device *dev, u64 addr,
+ size_t size,
+ enum dma_data_direction dir)
+{
+}
+
+static void *rvt_dma_alloc_coherent(struct ib_device *dev, size_t size,
+ u64 *dma_handle, gfp_t flag)
+{
+ struct page *p;
+ void *addr = NULL;
+
+ p = alloc_pages(flag, get_order(size));
+ if (p)
+ addr = page_address(p);
+ if (dma_handle)
+ *dma_handle = (u64)addr;
+ return addr;
+}
+
+static void rvt_dma_free_coherent(struct ib_device *dev, size_t size,
+ void *cpu_addr, u64 dma_handle)
+{
+ free_pages((unsigned long)cpu_addr, get_order(size));
+}
+
+struct ib_dma_mapping_ops rvt_default_dma_mapping_ops = {
+ .mapping_error = rvt_mapping_error,
+ .map_single = rvt_dma_map_single,
+ .unmap_single = rvt_dma_unmap_single,
+ .map_page = rvt_dma_map_page,
+ .unmap_page = rvt_dma_unmap_page,
+ .map_sg = rvt_map_sg,
+ .unmap_sg = rvt_unmap_sg,
+ .sync_single_for_cpu = rvt_sync_single_for_cpu,
+ .sync_single_for_device = rvt_sync_single_for_device,
+ .alloc_coherent = rvt_dma_alloc_coherent,
+ .free_coherent = rvt_dma_free_coherent
+};
diff --git a/drivers/infiniband/sw/rdmavt/dma.h b/drivers/infiniband/sw/rdmavt/dma.h
new file mode 100644
index 000000000000..979f07e09195
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/dma.h
@@ -0,0 +1,53 @@
+#ifndef DEF_RDMAVTDMA_H
+#define DEF_RDMAVTDMA_H
+
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+extern struct ib_dma_mapping_ops rvt_default_dma_mapping_ops;
+
+#endif /* DEF_RDMAVTDMA_H */
diff --git a/drivers/infiniband/sw/rdmavt/mad.c b/drivers/infiniband/sw/rdmavt/mad.c
new file mode 100644
index 000000000000..f6e99778d7ca
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/mad.c
@@ -0,0 +1,171 @@
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <rdma/ib_mad.h>
+#include "mad.h"
+#include "vt.h"
+
+/**
+ * rvt_process_mad - process an incoming MAD packet
+ * @ibdev: the infiniband device this packet came in on
+ * @mad_flags: MAD flags
+ * @port_num: the port number this packet came in on, 1 based from ib core
+ * @in_wc: the work completion entry for this packet
+ * @in_grh: the global route header for this packet
+ * @in_mad: the incoming MAD
+ * @out_mad: any outgoing MAD reply
+ *
+ * Note that the verbs framework has already done the MAD sanity checks,
+ * and hop count/pointer updating for IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE
+ * MADs.
+ *
+ * This is called by the ib_mad module.
+ *
+ * Return: IB_MAD_RESULT_SUCCESS or error
+ */
+int rvt_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
+ const struct ib_wc *in_wc, const struct ib_grh *in_grh,
+ const struct ib_mad_hdr *in, size_t in_mad_size,
+ struct ib_mad_hdr *out, size_t *out_mad_size,
+ u16 *out_mad_pkey_index)
+{
+ /*
+ * MAD processing is quite different between hfi1 and qib. Therfore this
+ * is expected to be provided by the driver. Other drivers in the future
+ * may chose to implement this but it should not be made into a
+ * requirement.
+ */
+ if (ibport_num_to_idx(ibdev, port_num) < 0)
+ return -EINVAL;
+
+ return IB_MAD_RESULT_FAILURE;
+}
+
+static void rvt_send_mad_handler(struct ib_mad_agent *agent,
+ struct ib_mad_send_wc *mad_send_wc)
+{
+ ib_free_send_mad(mad_send_wc->send_buf);
+}
+
+/**
+ * rvt_create_mad_agents - create mad agents
+ * @rdi: rvt dev struct
+ *
+ * If driver needs to be notified of mad agent creation then call back
+ *
+ * Return 0 on success
+ */
+int rvt_create_mad_agents(struct rvt_dev_info *rdi)
+{
+ struct ib_mad_agent *agent;
+ struct rvt_ibport *rvp;
+ int p;
+ int ret;
+
+ for (p = 0; p < rdi->dparms.nports; p++) {
+ rvp = rdi->ports[p];
+ agent = ib_register_mad_agent(&rdi->ibdev, p + 1,
+ IB_QPT_SMI,
+ NULL, 0, rvt_send_mad_handler,
+ NULL, NULL, 0);
+ if (IS_ERR(agent)) {
+ ret = PTR_ERR(agent);
+ goto err;
+ }
+
+ rvp->send_agent = agent;
+
+ if (rdi->driver_f.notify_create_mad_agent)
+ rdi->driver_f.notify_create_mad_agent(rdi, p);
+ }
+
+ return 0;
+
+err:
+ for (p = 0; p < rdi->dparms.nports; p++) {
+ rvp = rdi->ports[p];
+ if (rvp->send_agent) {
+ agent = rvp->send_agent;
+ rvp->send_agent = NULL;
+ ib_unregister_mad_agent(agent);
+ if (rdi->driver_f.notify_free_mad_agent)
+ rdi->driver_f.notify_free_mad_agent(rdi, p);
+ }
+ }
+
+ return ret;
+}
+
+/**
+ * rvt_free_mad_agents - free up mad agents
+ * @rdi: rvt dev struct
+ *
+ * If driver needs notification of mad agent removal make the call back
+ */
+void rvt_free_mad_agents(struct rvt_dev_info *rdi)
+{
+ struct ib_mad_agent *agent;
+ struct rvt_ibport *rvp;
+ int p;
+
+ for (p = 0; p < rdi->dparms.nports; p++) {
+ rvp = rdi->ports[p];
+ if (rvp->send_agent) {
+ agent = rvp->send_agent;
+ rvp->send_agent = NULL;
+ ib_unregister_mad_agent(agent);
+ }
+ if (rvp->sm_ah) {
+ ib_destroy_ah(&rvp->sm_ah->ibah);
+ rvp->sm_ah = NULL;
+ }
+
+ if (rdi->driver_f.notify_free_mad_agent)
+ rdi->driver_f.notify_free_mad_agent(rdi, p);
+ }
+}
+
diff --git a/drivers/infiniband/sw/rdmavt/mad.h b/drivers/infiniband/sw/rdmavt/mad.h
new file mode 100644
index 000000000000..a9d6eecc3723
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/mad.h
@@ -0,0 +1,60 @@
+#ifndef DEF_RVTMAD_H
+#define DEF_RVTMAD_H
+
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <rdma/rdma_vt.h>
+
+int rvt_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
+ const struct ib_wc *in_wc, const struct ib_grh *in_grh,
+ const struct ib_mad_hdr *in, size_t in_mad_size,
+ struct ib_mad_hdr *out, size_t *out_mad_size,
+ u16 *out_mad_pkey_index);
+int rvt_create_mad_agents(struct rvt_dev_info *rdi);
+void rvt_free_mad_agents(struct rvt_dev_info *rdi);
+#endif /* DEF_RVTMAD_H */
diff --git a/drivers/infiniband/sw/rdmavt/mcast.c b/drivers/infiniband/sw/rdmavt/mcast.c
new file mode 100644
index 000000000000..983d319ac976
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/mcast.c
@@ -0,0 +1,417 @@
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <linux/slab.h>
+#include <linux/sched.h>
+#include <linux/rculist.h>
+#include <rdma/rdma_vt.h>
+#include <rdma/rdmavt_qp.h>
+
+#include "mcast.h"
+
+/**
+ * rvt_driver_mcast - init resources for multicast
+ * @rdi: rvt dev struct
+ *
+ * This is per device that registers with rdmavt
+ */
+void rvt_driver_mcast_init(struct rvt_dev_info *rdi)
+{
+ /*
+ * Anything that needs setup for multicast on a per driver or per rdi
+ * basis should be done in here.
+ */
+ spin_lock_init(&rdi->n_mcast_grps_lock);
+}
+
+/**
+ * mcast_qp_alloc - alloc a struct to link a QP to mcast GID struct
+ * @qp: the QP to link
+ */
+static struct rvt_mcast_qp *rvt_mcast_qp_alloc(struct rvt_qp *qp)
+{
+ struct rvt_mcast_qp *mqp;
+
+ mqp = kmalloc(sizeof(*mqp), GFP_KERNEL);
+ if (!mqp)
+ goto bail;
+
+ mqp->qp = qp;
+ atomic_inc(&qp->refcount);
+
+bail:
+ return mqp;
+}
+
+static void rvt_mcast_qp_free(struct rvt_mcast_qp *mqp)
+{
+ struct rvt_qp *qp = mqp->qp;
+
+ /* Notify hfi1_destroy_qp() if it is waiting. */
+ if (atomic_dec_and_test(&qp->refcount))
+ wake_up(&qp->wait);
+
+ kfree(mqp);
+}
+
+/**
+ * mcast_alloc - allocate the multicast GID structure
+ * @mgid: the multicast GID
+ *
+ * A list of QPs will be attached to this structure.
+ */
+static struct rvt_mcast *rvt_mcast_alloc(union ib_gid *mgid)
+{
+ struct rvt_mcast *mcast;
+
+ mcast = kzalloc(sizeof(*mcast), GFP_KERNEL);
+ if (!mcast)
+ goto bail;
+
+ mcast->mgid = *mgid;
+ INIT_LIST_HEAD(&mcast->qp_list);
+ init_waitqueue_head(&mcast->wait);
+ atomic_set(&mcast->refcount, 0);
+
+bail:
+ return mcast;
+}
+
+static void rvt_mcast_free(struct rvt_mcast *mcast)
+{
+ struct rvt_mcast_qp *p, *tmp;
+
+ list_for_each_entry_safe(p, tmp, &mcast->qp_list, list)
+ rvt_mcast_qp_free(p);
+
+ kfree(mcast);
+}
+
+/**
+ * rvt_mcast_find - search the global table for the given multicast GID
+ * @ibp: the IB port structure
+ * @mgid: the multicast GID to search for
+ *
+ * The caller is responsible for decrementing the reference count if found.
+ *
+ * Return: NULL if not found.
+ */
+struct rvt_mcast *rvt_mcast_find(struct rvt_ibport *ibp, union ib_gid *mgid)
+{
+ struct rb_node *n;
+ unsigned long flags;
+ struct rvt_mcast *found = NULL;
+
+ spin_lock_irqsave(&ibp->lock, flags);
+ n = ibp->mcast_tree.rb_node;
+ while (n) {
+ int ret;
+ struct rvt_mcast *mcast;
+
+ mcast = rb_entry(n, struct rvt_mcast, rb_node);
+
+ ret = memcmp(mgid->raw, mcast->mgid.raw,
+ sizeof(union ib_gid));
+ if (ret < 0) {
+ n = n->rb_left;
+ } else if (ret > 0) {
+ n = n->rb_right;
+ } else {
+ atomic_inc(&mcast->refcount);
+ found = mcast;
+ break;
+ }
+ }
+ spin_unlock_irqrestore(&ibp->lock, flags);
+ return found;
+}
+EXPORT_SYMBOL(rvt_mcast_find);
+
+/**
+ * mcast_add - insert mcast GID into table and attach QP struct
+ * @mcast: the mcast GID table
+ * @mqp: the QP to attach
+ *
+ * Return: zero if both were added. Return EEXIST if the GID was already in
+ * the table but the QP was added. Return ESRCH if the QP was already
+ * attached and neither structure was added.
+ */
+static int rvt_mcast_add(struct rvt_dev_info *rdi, struct rvt_ibport *ibp,
+ struct rvt_mcast *mcast, struct rvt_mcast_qp *mqp)
+{
+ struct rb_node **n = &ibp->mcast_tree.rb_node;
+ struct rb_node *pn = NULL;
+ int ret;
+
+ spin_lock_irq(&ibp->lock);
+
+ while (*n) {
+ struct rvt_mcast *tmcast;
+ struct rvt_mcast_qp *p;
+
+ pn = *n;
+ tmcast = rb_entry(pn, struct rvt_mcast, rb_node);
+
+ ret = memcmp(mcast->mgid.raw, tmcast->mgid.raw,
+ sizeof(union ib_gid));
+ if (ret < 0) {
+ n = &pn->rb_left;
+ continue;
+ }
+ if (ret > 0) {
+ n = &pn->rb_right;
+ continue;
+ }
+
+ /* Search the QP list to see if this is already there. */
+ list_for_each_entry_rcu(p, &tmcast->qp_list, list) {
+ if (p->qp == mqp->qp) {
+ ret = ESRCH;
+ goto bail;
+ }
+ }
+ if (tmcast->n_attached ==
+ rdi->dparms.props.max_mcast_qp_attach) {
+ ret = ENOMEM;
+ goto bail;
+ }
+
+ tmcast->n_attached++;
+
+ list_add_tail_rcu(&mqp->list, &tmcast->qp_list);
+ ret = EEXIST;
+ goto bail;
+ }
+
+ spin_lock(&rdi->n_mcast_grps_lock);
+ if (rdi->n_mcast_grps_allocated == rdi->dparms.props.max_mcast_grp) {
+ spin_unlock(&rdi->n_mcast_grps_lock);
+ ret = ENOMEM;
+ goto bail;
+ }
+
+ rdi->n_mcast_grps_allocated++;
+ spin_unlock(&rdi->n_mcast_grps_lock);
+
+ mcast->n_attached++;
+
+ list_add_tail_rcu(&mqp->list, &mcast->qp_list);
+
+ atomic_inc(&mcast->refcount);
+ rb_link_node(&mcast->rb_node, pn, n);
+ rb_insert_color(&mcast->rb_node, &ibp->mcast_tree);
+
+ ret = 0;
+
+bail:
+ spin_unlock_irq(&ibp->lock);
+
+ return ret;
+}
+
+/**
+ * rvt_attach_mcast - attach a qp to a multicast group
+ * @ibqp: Infiniband qp
+ * @igd: multicast guid
+ * @lid: multicast lid
+ *
+ * Return: 0 on success
+ */
+int rvt_attach_mcast(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
+{
+ struct rvt_qp *qp = ibqp_to_rvtqp(ibqp);
+ struct rvt_dev_info *rdi = ib_to_rvt(ibqp->device);
+ struct rvt_ibport *ibp = rdi->ports[qp->port_num - 1];
+ struct rvt_mcast *mcast;
+ struct rvt_mcast_qp *mqp;
+ int ret = -ENOMEM;
+
+ if (ibqp->qp_num <= 1 || qp->state == IB_QPS_RESET)
+ return -EINVAL;
+
+ /*
+ * Allocate data structures since its better to do this outside of
+ * spin locks and it will most likely be needed.
+ */
+ mcast = rvt_mcast_alloc(gid);
+ if (!mcast)
+ return -ENOMEM;
+
+ mqp = rvt_mcast_qp_alloc(qp);
+ if (!mqp)
+ goto bail_mcast;
+
+ switch (rvt_mcast_add(rdi, ibp, mcast, mqp)) {
+ case ESRCH:
+ /* Neither was used: OK to attach the same QP twice. */
+ ret = 0;
+ goto bail_mqp;
+ case EEXIST: /* The mcast wasn't used */
+ ret = 0;
+ goto bail_mcast;
+ case ENOMEM:
+ /* Exceeded the maximum number of mcast groups. */
+ ret = -ENOMEM;
+ goto bail_mqp;
+ default:
+ break;
+ }
+
+ return 0;
+
+bail_mqp:
+ rvt_mcast_qp_free(mqp);
+
+bail_mcast:
+ rvt_mcast_free(mcast);
+
+ return ret;
+}
+
+/**
+ * rvt_detach_mcast - remove a qp from a multicast group
+ * @ibqp: Infiniband qp
+ * @igd: multicast guid
+ * @lid: multicast lid
+ *
+ * Return: 0 on success
+ */
+int rvt_detach_mcast(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
+{
+ struct rvt_qp *qp = ibqp_to_rvtqp(ibqp);
+ struct rvt_dev_info *rdi = ib_to_rvt(ibqp->device);
+ struct rvt_ibport *ibp = rdi->ports[qp->port_num - 1];
+ struct rvt_mcast *mcast = NULL;
+ struct rvt_mcast_qp *p, *tmp, *delp = NULL;
+ struct rb_node *n;
+ int last = 0;
+ int ret = 0;
+
+ if (ibqp->qp_num <= 1 || qp->state == IB_QPS_RESET)
+ return -EINVAL;
+
+ spin_lock_irq(&ibp->lock);
+
+ /* Find the GID in the mcast table. */
+ n = ibp->mcast_tree.rb_node;
+ while (1) {
+ if (!n) {
+ spin_unlock_irq(&ibp->lock);
+ return -EINVAL;
+ }
+
+ mcast = rb_entry(n, struct rvt_mcast, rb_node);
+ ret = memcmp(gid->raw, mcast->mgid.raw,
+ sizeof(union ib_gid));
+ if (ret < 0)
+ n = n->rb_left;
+ else if (ret > 0)
+ n = n->rb_right;
+ else
+ break;
+ }
+
+ /* Search the QP list. */
+ list_for_each_entry_safe(p, tmp, &mcast->qp_list, list) {
+ if (p->qp != qp)
+ continue;
+ /*
+ * We found it, so remove it, but don't poison the forward
+ * link until we are sure there are no list walkers.
+ */
+ list_del_rcu(&p->list);
+ mcast->n_attached--;
+ delp = p;
+
+ /* If this was the last attached QP, remove the GID too. */
+ if (list_empty(&mcast->qp_list)) {
+ rb_erase(&mcast->rb_node, &ibp->mcast_tree);
+ last = 1;
+ }
+ break;
+ }
+
+ spin_unlock_irq(&ibp->lock);
+ /* QP not attached */
+ if (!delp)
+ return -EINVAL;
+
+ /*
+ * Wait for any list walkers to finish before freeing the
+ * list element.
+ */
+ wait_event(mcast->wait, atomic_read(&mcast->refcount) <= 1);
+ rvt_mcast_qp_free(delp);
+
+ if (last) {
+ atomic_dec(&mcast->refcount);
+ wait_event(mcast->wait, !atomic_read(&mcast->refcount));
+ rvt_mcast_free(mcast);
+ spin_lock_irq(&rdi->n_mcast_grps_lock);
+ rdi->n_mcast_grps_allocated--;
+ spin_unlock_irq(&rdi->n_mcast_grps_lock);
+ }
+
+ return 0;
+}
+
+/**
+ *rvt_mast_tree_empty - determine if any qps are attached to any mcast group
+ *@rdi: rvt dev struct
+ *
+ * Return: in use count
+ */
+int rvt_mcast_tree_empty(struct rvt_dev_info *rdi)
+{
+ int i;
+ int in_use = 0;
+
+ for (i = 0; i < rdi->dparms.nports; i++)
+ if (rdi->ports[i]->mcast_tree.rb_node)
+ in_use++;
+ return in_use;
+}
diff --git a/drivers/infiniband/sw/rdmavt/mcast.h b/drivers/infiniband/sw/rdmavt/mcast.h
new file mode 100644
index 000000000000..29f579267608
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/mcast.h
@@ -0,0 +1,58 @@
+#ifndef DEF_RVTMCAST_H
+#define DEF_RVTMCAST_H
+
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <rdma/rdma_vt.h>
+
+void rvt_driver_mcast_init(struct rvt_dev_info *rdi);
+int rvt_attach_mcast(struct ib_qp *ibqp, union ib_gid *gid, u16 lid);
+int rvt_detach_mcast(struct ib_qp *ibqp, union ib_gid *gid, u16 lid);
+int rvt_mcast_tree_empty(struct rvt_dev_info *rdi);
+
+#endif /* DEF_RVTMCAST_H */
diff --git a/drivers/infiniband/sw/rdmavt/mmap.c b/drivers/infiniband/sw/rdmavt/mmap.c
new file mode 100644
index 000000000000..e202b8142759
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/mmap.c
@@ -0,0 +1,208 @@
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/mm.h>
+#include <asm/pgtable.h>
+#include "mmap.h"
+
+/**
+ * rvt_mmap_init - init link list and lock for mem map
+ * @rdi: rvt dev struct
+ */
+void rvt_mmap_init(struct rvt_dev_info *rdi)
+{
+ INIT_LIST_HEAD(&rdi->pending_mmaps);
+ spin_lock_init(&rdi->pending_lock);
+ rdi->mmap_offset = PAGE_SIZE;
+ spin_lock_init(&rdi->mmap_offset_lock);
+}
+
+/**
+ * rvt_release_mmap_info - free mmap info structure
+ * @ref: a pointer to the kref within struct rvt_mmap_info
+ */
+void rvt_release_mmap_info(struct kref *ref)
+{
+ struct rvt_mmap_info *ip =
+ container_of(ref, struct rvt_mmap_info, ref);
+ struct rvt_dev_info *rdi = ib_to_rvt(ip->context->device);
+
+ spin_lock_irq(&rdi->pending_lock);
+ list_del(&ip->pending_mmaps);
+ spin_unlock_irq(&rdi->pending_lock);
+
+ vfree(ip->obj);
+ kfree(ip);
+}
+
+static void rvt_vma_open(struct vm_area_struct *vma)
+{
+ struct rvt_mmap_info *ip = vma->vm_private_data;
+
+ kref_get(&ip->ref);
+}
+
+static void rvt_vma_close(struct vm_area_struct *vma)
+{
+ struct rvt_mmap_info *ip = vma->vm_private_data;
+
+ kref_put(&ip->ref, rvt_release_mmap_info);
+}
+
+static const struct vm_operations_struct rvt_vm_ops = {
+ .open = rvt_vma_open,
+ .close = rvt_vma_close,
+};
+
+/**
+ * rvt_mmap - create a new mmap region
+ * @context: the IB user context of the process making the mmap() call
+ * @vma: the VMA to be initialized
+ *
+ * Return: zero if the mmap is OK. Otherwise, return an errno.
+ */
+int rvt_mmap(struct ib_ucontext *context, struct vm_area_struct *vma)
+{
+ struct rvt_dev_info *rdi = ib_to_rvt(context->device);
+ unsigned long offset = vma->vm_pgoff << PAGE_SHIFT;
+ unsigned long size = vma->vm_end - vma->vm_start;
+ struct rvt_mmap_info *ip, *pp;
+ int ret = -EINVAL;
+
+ /*
+ * Search the device's list of objects waiting for a mmap call.
+ * Normally, this list is very short since a call to create a
+ * CQ, QP, or SRQ is soon followed by a call to mmap().
+ */
+ spin_lock_irq(&rdi->pending_lock);
+ list_for_each_entry_safe(ip, pp, &rdi->pending_mmaps,
+ pending_mmaps) {
+ /* Only the creator is allowed to mmap the object */
+ if (context != ip->context || (__u64)offset != ip->offset)
+ continue;
+ /* Don't allow a mmap larger than the object. */
+ if (size > ip->size)
+ break;
+
+ list_del_init(&ip->pending_mmaps);
+ spin_unlock_irq(&rdi->pending_lock);
+
+ ret = remap_vmalloc_range(vma, ip->obj, 0);
+ if (ret)
+ goto done;
+ vma->vm_ops = &rvt_vm_ops;
+ vma->vm_private_data = ip;
+ rvt_vma_open(vma);
+ goto done;
+ }
+ spin_unlock_irq(&rdi->pending_lock);
+done:
+ return ret;
+}
+
+/**
+ * rvt_create_mmap_info - allocate information for hfi1_mmap
+ * @rdi: rvt dev struct
+ * @size: size in bytes to map
+ * @context: user context
+ * @obj: opaque pointer to a cq, wq etc
+ *
+ * Return: rvt_mmap struct on success
+ */
+struct rvt_mmap_info *rvt_create_mmap_info(struct rvt_dev_info *rdi,
+ u32 size,
+ struct ib_ucontext *context,
+ void *obj)
+{
+ struct rvt_mmap_info *ip;
+
+ ip = kmalloc_node(sizeof(*ip), GFP_KERNEL, rdi->dparms.node);
+ if (!ip)
+ return ip;
+
+ size = PAGE_ALIGN(size);
+
+ spin_lock_irq(&rdi->mmap_offset_lock);
+ if (rdi->mmap_offset == 0)
+ rdi->mmap_offset = PAGE_SIZE;
+ ip->offset = rdi->mmap_offset;
+ rdi->mmap_offset += size;
+ spin_unlock_irq(&rdi->mmap_offset_lock);
+
+ INIT_LIST_HEAD(&ip->pending_mmaps);
+ ip->size = size;
+ ip->context = context;
+ ip->obj = obj;
+ kref_init(&ip->ref);
+
+ return ip;
+}
+
+/**
+ * rvt_update_mmap_info - update a mem map
+ * @rdi: rvt dev struct
+ * @ip: mmap info pointer
+ * @size: size to grow by
+ * @obj: opaque pointer to cq, wq, etc.
+ */
+void rvt_update_mmap_info(struct rvt_dev_info *rdi, struct rvt_mmap_info *ip,
+ u32 size, void *obj)
+{
+ size = PAGE_ALIGN(size);
+
+ spin_lock_irq(&rdi->mmap_offset_lock);
+ if (rdi->mmap_offset == 0)
+ rdi->mmap_offset = PAGE_SIZE;
+ ip->offset = rdi->mmap_offset;
+ rdi->mmap_offset += size;
+ spin_unlock_irq(&rdi->mmap_offset_lock);
+
+ ip->size = size;
+ ip->obj = obj;
+}
diff --git a/drivers/infiniband/sw/rdmavt/mmap.h b/drivers/infiniband/sw/rdmavt/mmap.h
new file mode 100644
index 000000000000..fab0e7b1daf9
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/mmap.h
@@ -0,0 +1,63 @@
+#ifndef DEF_RDMAVTMMAP_H
+#define DEF_RDMAVTMMAP_H
+
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <rdma/rdma_vt.h>
+
+void rvt_mmap_init(struct rvt_dev_info *rdi);
+void rvt_release_mmap_info(struct kref *ref);
+int rvt_mmap(struct ib_ucontext *context, struct vm_area_struct *vma);
+struct rvt_mmap_info *rvt_create_mmap_info(struct rvt_dev_info *rdi,
+ u32 size,
+ struct ib_ucontext *context,
+ void *obj);
+void rvt_update_mmap_info(struct rvt_dev_info *rdi, struct rvt_mmap_info *ip,
+ u32 size, void *obj);
+
+#endif /* DEF_RDMAVTMMAP_H */
diff --git a/drivers/infiniband/sw/rdmavt/mr.c b/drivers/infiniband/sw/rdmavt/mr.c
new file mode 100644
index 000000000000..0ff765bfd619
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/mr.c
@@ -0,0 +1,830 @@
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <rdma/ib_umem.h>
+#include <rdma/rdma_vt.h>
+#include "vt.h"
+#include "mr.h"
+
+/**
+ * rvt_driver_mr_init - Init MR resources per driver
+ * @rdi: rvt dev struct
+ *
+ * Do any intilization needed when a driver registers with rdmavt.
+ *
+ * Return: 0 on success or errno on failure
+ */
+int rvt_driver_mr_init(struct rvt_dev_info *rdi)
+{
+ unsigned int lkey_table_size = rdi->dparms.lkey_table_size;
+ unsigned lk_tab_size;
+ int i;
+
+ /*
+ * The top hfi1_lkey_table_size bits are used to index the
+ * table. The lower 8 bits can be owned by the user (copied from
+ * the LKEY). The remaining bits act as a generation number or tag.
+ */
+ if (!lkey_table_size)
+ return -EINVAL;
+
+ spin_lock_init(&rdi->lkey_table.lock);
+
+ /* ensure generation is at least 4 bits */
+ if (lkey_table_size > RVT_MAX_LKEY_TABLE_BITS) {
+ rvt_pr_warn(rdi, "lkey bits %u too large, reduced to %u\n",
+ lkey_table_size, RVT_MAX_LKEY_TABLE_BITS);
+ rdi->dparms.lkey_table_size = RVT_MAX_LKEY_TABLE_BITS;
+ lkey_table_size = rdi->dparms.lkey_table_size;
+ }
+ rdi->lkey_table.max = 1 << lkey_table_size;
+ lk_tab_size = rdi->lkey_table.max * sizeof(*rdi->lkey_table.table);
+ rdi->lkey_table.table = (struct rvt_mregion __rcu **)
+ vmalloc_node(lk_tab_size, rdi->dparms.node);
+ if (!rdi->lkey_table.table)
+ return -ENOMEM;
+
+ RCU_INIT_POINTER(rdi->dma_mr, NULL);
+ for (i = 0; i < rdi->lkey_table.max; i++)
+ RCU_INIT_POINTER(rdi->lkey_table.table[i], NULL);
+
+ return 0;
+}
+
+/**
+ *rvt_mr_exit: clean up MR
+ *@rdi: rvt dev structure
+ *
+ * called when drivers have unregistered or perhaps failed to register with us
+ */
+void rvt_mr_exit(struct rvt_dev_info *rdi)
+{
+ if (rdi->dma_mr)
+ rvt_pr_err(rdi, "DMA MR not null!\n");
+
+ vfree(rdi->lkey_table.table);
+}
+
+static void rvt_deinit_mregion(struct rvt_mregion *mr)
+{
+ int i = mr->mapsz;
+
+ mr->mapsz = 0;
+ while (i)
+ kfree(mr->map[--i]);
+}
+
+static int rvt_init_mregion(struct rvt_mregion *mr, struct ib_pd *pd,
+ int count)
+{
+ int m, i = 0;
+
+ mr->mapsz = 0;
+ m = (count + RVT_SEGSZ - 1) / RVT_SEGSZ;
+ for (; i < m; i++) {
+ mr->map[i] = kzalloc(sizeof(*mr->map[0]), GFP_KERNEL);
+ if (!mr->map[i]) {
+ rvt_deinit_mregion(mr);
+ return -ENOMEM;
+ }
+ mr->mapsz++;
+ }
+ init_completion(&mr->comp);
+ /* count returning the ptr to user */
+ atomic_set(&mr->refcount, 1);
+ mr->pd = pd;
+ mr->max_segs = count;
+ return 0;
+}
+
+/**
+ * rvt_alloc_lkey - allocate an lkey
+ * @mr: memory region that this lkey protects
+ * @dma_region: 0->normal key, 1->restricted DMA key
+ *
+ * Returns 0 if successful, otherwise returns -errno.
+ *
+ * Increments mr reference count as required.
+ *
+ * Sets the lkey field mr for non-dma regions.
+ *
+ */
+static int rvt_alloc_lkey(struct rvt_mregion *mr, int dma_region)
+{
+ unsigned long flags;
+ u32 r;
+ u32 n;
+ int ret = 0;
+ struct rvt_dev_info *dev = ib_to_rvt(mr->pd->device);
+ struct rvt_lkey_table *rkt = &dev->lkey_table;
+
+ rvt_get_mr(mr);
+ spin_lock_irqsave(&rkt->lock, flags);
+
+ /* special case for dma_mr lkey == 0 */
+ if (dma_region) {
+ struct rvt_mregion *tmr;
+
+ tmr = rcu_access_pointer(dev->dma_mr);
+ if (!tmr) {
+ rcu_assign_pointer(dev->dma_mr, mr);
+ mr->lkey_published = 1;
+ } else {
+ rvt_put_mr(mr);
+ }
+ goto success;
+ }
+
+ /* Find the next available LKEY */
+ r = rkt->next;
+ n = r;
+ for (;;) {
+ if (!rcu_access_pointer(rkt->table[r]))
+ break;
+ r = (r + 1) & (rkt->max - 1);
+ if (r == n)
+ goto bail;
+ }
+ rkt->next = (r + 1) & (rkt->max - 1);
+ /*
+ * Make sure lkey is never zero which is reserved to indicate an
+ * unrestricted LKEY.
+ */
+ rkt->gen++;
+ /*
+ * bits are capped to ensure enough bits for generation number
+ */
+ mr->lkey = (r << (32 - dev->dparms.lkey_table_size)) |
+ ((((1 << (24 - dev->dparms.lkey_table_size)) - 1) & rkt->gen)
+ << 8);
+ if (mr->lkey == 0) {
+ mr->lkey |= 1 << 8;
+ rkt->gen++;
+ }
+ rcu_assign_pointer(rkt->table[r], mr);
+ mr->lkey_published = 1;
+success:
+ spin_unlock_irqrestore(&rkt->lock, flags);
+out:
+ return ret;
+bail:
+ rvt_put_mr(mr);
+ spin_unlock_irqrestore(&rkt->lock, flags);
+ ret = -ENOMEM;
+ goto out;
+}
+
+/**
+ * rvt_free_lkey - free an lkey
+ * @mr: mr to free from tables
+ */
+static void rvt_free_lkey(struct rvt_mregion *mr)
+{
+ unsigned long flags;
+ u32 lkey = mr->lkey;
+ u32 r;
+ struct rvt_dev_info *dev = ib_to_rvt(mr->pd->device);
+ struct rvt_lkey_table *rkt = &dev->lkey_table;
+ int freed = 0;
+
+ spin_lock_irqsave(&rkt->lock, flags);
+ if (!mr->lkey_published)
+ goto out;
+ if (lkey == 0) {
+ RCU_INIT_POINTER(dev->dma_mr, NULL);
+ } else {
+ r = lkey >> (32 - dev->dparms.lkey_table_size);
+ RCU_INIT_POINTER(rkt->table[r], NULL);
+ }
+ mr->lkey_published = 0;
+ freed++;
+out:
+ spin_unlock_irqrestore(&rkt->lock, flags);
+ if (freed) {
+ synchronize_rcu();
+ rvt_put_mr(mr);
+ }
+}
+
+static struct rvt_mr *__rvt_alloc_mr(int count, struct ib_pd *pd)
+{
+ struct rvt_mr *mr;
+ int rval = -ENOMEM;
+ int m;
+
+ /* Allocate struct plus pointers to first level page tables. */
+ m = (count + RVT_SEGSZ - 1) / RVT_SEGSZ;
+ mr = kzalloc(sizeof(*mr) + m * sizeof(mr->mr.map[0]), GFP_KERNEL);
+ if (!mr)
+ goto bail;
+
+ rval = rvt_init_mregion(&mr->mr, pd, count);
+ if (rval)
+ goto bail;
+ /*
+ * ib_reg_phys_mr() will initialize mr->ibmr except for
+ * lkey and rkey.
+ */
+ rval = rvt_alloc_lkey(&mr->mr, 0);
+ if (rval)
+ goto bail_mregion;
+ mr->ibmr.lkey = mr->mr.lkey;
+ mr->ibmr.rkey = mr->mr.lkey;
+done:
+ return mr;
+
+bail_mregion:
+ rvt_deinit_mregion(&mr->mr);
+bail:
+ kfree(mr);
+ mr = ERR_PTR(rval);
+ goto done;
+}
+
+static void __rvt_free_mr(struct rvt_mr *mr)
+{
+ rvt_deinit_mregion(&mr->mr);
+ rvt_free_lkey(&mr->mr);
+ vfree(mr);
+}
+
+/**
+ * rvt_get_dma_mr - get a DMA memory region
+ * @pd: protection domain for this memory region
+ * @acc: access flags
+ *
+ * Return: the memory region on success, otherwise returns an errno.
+ * Note that all DMA addresses should be created via the
+ * struct ib_dma_mapping_ops functions (see dma.c).
+ */
+struct ib_mr *rvt_get_dma_mr(struct ib_pd *pd, int acc)
+{
+ struct rvt_mr *mr;
+ struct ib_mr *ret;
+ int rval;
+
+ if (ibpd_to_rvtpd(pd)->user)
+ return ERR_PTR(-EPERM);
+
+ mr = kzalloc(sizeof(*mr), GFP_KERNEL);
+ if (!mr) {
+ ret = ERR_PTR(-ENOMEM);
+ goto bail;
+ }
+
+ rval = rvt_init_mregion(&mr->mr, pd, 0);
+ if (rval) {
+ ret = ERR_PTR(rval);
+ goto bail;
+ }
+
+ rval = rvt_alloc_lkey(&mr->mr, 1);
+ if (rval) {
+ ret = ERR_PTR(rval);
+ goto bail_mregion;
+ }
+
+ mr->mr.access_flags = acc;
+ ret = &mr->ibmr;
+done:
+ return ret;
+
+bail_mregion:
+ rvt_deinit_mregion(&mr->mr);
+bail:
+ kfree(mr);
+ goto done;
+}
+
+/**
+ * rvt_reg_user_mr - register a userspace memory region
+ * @pd: protection domain for this memory region
+ * @start: starting userspace address
+ * @length: length of region to register
+ * @mr_access_flags: access flags for this memory region
+ * @udata: unused by the driver
+ *
+ * Return: the memory region on success, otherwise returns an errno.
+ */
+struct ib_mr *rvt_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
+ u64 virt_addr, int mr_access_flags,
+ struct ib_udata *udata)
+{
+ struct rvt_mr *mr;
+ struct ib_umem *umem;
+ struct scatterlist *sg;
+ int n, m, entry;
+ struct ib_mr *ret;
+
+ if (length == 0)
+ return ERR_PTR(-EINVAL);
+
+ umem = ib_umem_get(pd->uobject->context, start, length,
+ mr_access_flags, 0);
+ if (IS_ERR(umem))
+ return (void *)umem;
+
+ n = umem->nmap;
+
+ mr = __rvt_alloc_mr(n, pd);
+ if (IS_ERR(mr)) {
+ ret = (struct ib_mr *)mr;
+ goto bail_umem;
+ }
+
+ mr->mr.user_base = start;
+ mr->mr.iova = virt_addr;
+ mr->mr.length = length;
+ mr->mr.offset = ib_umem_offset(umem);
+ mr->mr.access_flags = mr_access_flags;
+ mr->umem = umem;
+
+ if (is_power_of_2(umem->page_size))
+ mr->mr.page_shift = ilog2(umem->page_size);
+ m = 0;
+ n = 0;
+ for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) {
+ void *vaddr;
+
+ vaddr = page_address(sg_page(sg));
+ if (!vaddr) {
+ ret = ERR_PTR(-EINVAL);
+ goto bail_inval;
+ }
+ mr->mr.map[m]->segs[n].vaddr = vaddr;
+ mr->mr.map[m]->segs[n].length = umem->page_size;
+ n++;
+ if (n == RVT_SEGSZ) {
+ m++;
+ n = 0;
+ }
+ }
+ return &mr->ibmr;
+
+bail_inval:
+ __rvt_free_mr(mr);
+
+bail_umem:
+ ib_umem_release(umem);
+
+ return ret;
+}
+
+/**
+ * rvt_dereg_mr - unregister and free a memory region
+ * @ibmr: the memory region to free
+ *
+ *
+ * Note that this is called to free MRs created by rvt_get_dma_mr()
+ * or rvt_reg_user_mr().
+ *
+ * Returns 0 on success.
+ */
+int rvt_dereg_mr(struct ib_mr *ibmr)
+{
+ struct rvt_mr *mr = to_imr(ibmr);
+ struct rvt_dev_info *rdi = ib_to_rvt(ibmr->pd->device);
+ int ret = 0;
+ unsigned long timeout;
+
+ rvt_free_lkey(&mr->mr);
+
+ rvt_put_mr(&mr->mr); /* will set completion if last */
+ timeout = wait_for_completion_timeout(&mr->mr.comp, 5 * HZ);
+ if (!timeout) {
+ rvt_pr_err(rdi,
+ "rvt_dereg_mr timeout mr %p pd %p refcount %u\n",
+ mr, mr->mr.pd, atomic_read(&mr->mr.refcount));
+ rvt_get_mr(&mr->mr);
+ ret = -EBUSY;
+ goto out;
+ }
+ rvt_deinit_mregion(&mr->mr);
+ if (mr->umem)
+ ib_umem_release(mr->umem);
+ kfree(mr);
+out:
+ return ret;
+}
+
+/**
+ * rvt_alloc_mr - Allocate a memory region usable with the
+ * @pd: protection domain for this memory region
+ * @mr_type: mem region type
+ * @max_num_sg: Max number of segments allowed
+ *
+ * Return: the memory region on success, otherwise return an errno.
+ */
+struct ib_mr *rvt_alloc_mr(struct ib_pd *pd,
+ enum ib_mr_type mr_type,
+ u32 max_num_sg)
+{
+ struct rvt_mr *mr;
+
+ if (mr_type != IB_MR_TYPE_MEM_REG)
+ return ERR_PTR(-EINVAL);
+
+ mr = __rvt_alloc_mr(max_num_sg, pd);
+ if (IS_ERR(mr))
+ return (struct ib_mr *)mr;
+
+ return &mr->ibmr;
+}
+
+/**
+ * rvt_alloc_fmr - allocate a fast memory region
+ * @pd: the protection domain for this memory region
+ * @mr_access_flags: access flags for this memory region
+ * @fmr_attr: fast memory region attributes
+ *
+ * Return: the memory region on success, otherwise returns an errno.
+ */
+struct ib_fmr *rvt_alloc_fmr(struct ib_pd *pd, int mr_access_flags,
+ struct ib_fmr_attr *fmr_attr)
+{
+ struct rvt_fmr *fmr;
+ int m;
+ struct ib_fmr *ret;
+ int rval = -ENOMEM;
+
+ /* Allocate struct plus pointers to first level page tables. */
+ m = (fmr_attr->max_pages + RVT_SEGSZ - 1) / RVT_SEGSZ;
+ fmr = kzalloc(sizeof(*fmr) + m * sizeof(fmr->mr.map[0]), GFP_KERNEL);
+ if (!fmr)
+ goto bail;
+
+ rval = rvt_init_mregion(&fmr->mr, pd, fmr_attr->max_pages);
+ if (rval)
+ goto bail;
+
+ /*
+ * ib_alloc_fmr() will initialize fmr->ibfmr except for lkey &
+ * rkey.
+ */
+ rval = rvt_alloc_lkey(&fmr->mr, 0);
+ if (rval)
+ goto bail_mregion;
+ fmr->ibfmr.rkey = fmr->mr.lkey;
+ fmr->ibfmr.lkey = fmr->mr.lkey;
+ /*
+ * Resources are allocated but no valid mapping (RKEY can't be
+ * used).
+ */
+ fmr->mr.access_flags = mr_access_flags;
+ fmr->mr.max_segs = fmr_attr->max_pages;
+ fmr->mr.page_shift = fmr_attr->page_shift;
+
+ ret = &fmr->ibfmr;
+done:
+ return ret;
+
+bail_mregion:
+ rvt_deinit_mregion(&fmr->mr);
+bail:
+ kfree(fmr);
+ ret = ERR_PTR(rval);
+ goto done;
+}
+
+/**
+ * rvt_map_phys_fmr - set up a fast memory region
+ * @ibmfr: the fast memory region to set up
+ * @page_list: the list of pages to associate with the fast memory region
+ * @list_len: the number of pages to associate with the fast memory region
+ * @iova: the virtual address of the start of the fast memory region
+ *
+ * This may be called from interrupt context.
+ *
+ * Return: 0 on success
+ */
+
+int rvt_map_phys_fmr(struct ib_fmr *ibfmr, u64 *page_list,
+ int list_len, u64 iova)
+{
+ struct rvt_fmr *fmr = to_ifmr(ibfmr);
+ struct rvt_lkey_table *rkt;
+ unsigned long flags;
+ int m, n, i;
+ u32 ps;
+ struct rvt_dev_info *rdi = ib_to_rvt(ibfmr->device);
+
+ i = atomic_read(&fmr->mr.refcount);
+ if (i > 2)
+ return -EBUSY;
+
+ if (list_len > fmr->mr.max_segs)
+ return -EINVAL;
+
+ rkt = &rdi->lkey_table;
+ spin_lock_irqsave(&rkt->lock, flags);
+ fmr->mr.user_base = iova;
+ fmr->mr.iova = iova;
+ ps = 1 << fmr->mr.page_shift;
+ fmr->mr.length = list_len * ps;
+ m = 0;
+ n = 0;
+ for (i = 0; i < list_len; i++) {
+ fmr->mr.map[m]->segs[n].vaddr = (void *)page_list[i];
+ fmr->mr.map[m]->segs[n].length = ps;
+ if (++n == RVT_SEGSZ) {
+ m++;
+ n = 0;
+ }
+ }
+ spin_unlock_irqrestore(&rkt->lock, flags);
+ return 0;
+}
+
+/**
+ * rvt_unmap_fmr - unmap fast memory regions
+ * @fmr_list: the list of fast memory regions to unmap
+ *
+ * Return: 0 on success.
+ */
+int rvt_unmap_fmr(struct list_head *fmr_list)
+{
+ struct rvt_fmr *fmr;
+ struct rvt_lkey_table *rkt;
+ unsigned long flags;
+ struct rvt_dev_info *rdi;
+
+ list_for_each_entry(fmr, fmr_list, ibfmr.list) {
+ rdi = ib_to_rvt(fmr->ibfmr.device);
+ rkt = &rdi->lkey_table;
+ spin_lock_irqsave(&rkt->lock, flags);
+ fmr->mr.user_base = 0;
+ fmr->mr.iova = 0;
+ fmr->mr.length = 0;
+ spin_unlock_irqrestore(&rkt->lock, flags);
+ }
+ return 0;
+}
+
+/**
+ * rvt_dealloc_fmr - deallocate a fast memory region
+ * @ibfmr: the fast memory region to deallocate
+ *
+ * Return: 0 on success.
+ */
+int rvt_dealloc_fmr(struct ib_fmr *ibfmr)
+{
+ struct rvt_fmr *fmr = to_ifmr(ibfmr);
+ int ret = 0;
+ unsigned long timeout;
+
+ rvt_free_lkey(&fmr->mr);
+ rvt_put_mr(&fmr->mr); /* will set completion if last */
+ timeout = wait_for_completion_timeout(&fmr->mr.comp, 5 * HZ);
+ if (!timeout) {
+ rvt_get_mr(&fmr->mr);
+ ret = -EBUSY;
+ goto out;
+ }
+ rvt_deinit_mregion(&fmr->mr);
+ kfree(fmr);
+out:
+ return ret;
+}
+
+/**
+ * rvt_lkey_ok - check IB SGE for validity and initialize
+ * @rkt: table containing lkey to check SGE against
+ * @pd: protection domain
+ * @isge: outgoing internal SGE
+ * @sge: SGE to check
+ * @acc: access flags
+ *
+ * Check the IB SGE for validity and initialize our internal version
+ * of it.
+ *
+ * Return: 1 if valid and successful, otherwise returns 0.
+ *
+ * increments the reference count upon success
+ *
+ */
+int rvt_lkey_ok(struct rvt_lkey_table *rkt, struct rvt_pd *pd,
+ struct rvt_sge *isge, struct ib_sge *sge, int acc)
+{
+ struct rvt_mregion *mr;
+ unsigned n, m;
+ size_t off;
+ struct rvt_dev_info *dev = ib_to_rvt(pd->ibpd.device);
+
+ /*
+ * We use LKEY == zero for kernel virtual addresses
+ * (see rvt_get_dma_mr and dma.c).
+ */
+ rcu_read_lock();
+ if (sge->lkey == 0) {
+ if (pd->user)
+ goto bail;
+ mr = rcu_dereference(dev->dma_mr);
+ if (!mr)
+ goto bail;
+ atomic_inc(&mr->refcount);
+ rcu_read_unlock();
+
+ isge->mr = mr;
+ isge->vaddr = (void *)sge->addr;
+ isge->length = sge->length;
+ isge->sge_length = sge->length;
+ isge->m = 0;
+ isge->n = 0;
+ goto ok;
+ }
+ mr = rcu_dereference(
+ rkt->table[(sge->lkey >> (32 - dev->dparms.lkey_table_size))]);
+ if (unlikely(!mr || mr->lkey != sge->lkey || mr->pd != &pd->ibpd))
+ goto bail;
+
+ off = sge->addr - mr->user_base;
+ if (unlikely(sge->addr < mr->user_base ||
+ off + sge->length > mr->length ||
+ (mr->access_flags & acc) != acc))
+ goto bail;
+ atomic_inc(&mr->refcount);
+ rcu_read_unlock();
+
+ off += mr->offset;
+ if (mr->page_shift) {
+ /*
+ * page sizes are uniform power of 2 so no loop is necessary
+ * entries_spanned_by_off is the number of times the loop below
+ * would have executed.
+ */
+ size_t entries_spanned_by_off;
+
+ entries_spanned_by_off = off >> mr->page_shift;
+ off -= (entries_spanned_by_off << mr->page_shift);
+ m = entries_spanned_by_off / RVT_SEGSZ;
+ n = entries_spanned_by_off % RVT_SEGSZ;
+ } else {
+ m = 0;
+ n = 0;
+ while (off >= mr->map[m]->segs[n].length) {
+ off -= mr->map[m]->segs[n].length;
+ n++;
+ if (n >= RVT_SEGSZ) {
+ m++;
+ n = 0;
+ }
+ }
+ }
+ isge->mr = mr;
+ isge->vaddr = mr->map[m]->segs[n].vaddr + off;
+ isge->length = mr->map[m]->segs[n].length - off;
+ isge->sge_length = sge->length;
+ isge->m = m;
+ isge->n = n;
+ok:
+ return 1;
+bail:
+ rcu_read_unlock();
+ return 0;
+}
+EXPORT_SYMBOL(rvt_lkey_ok);
+
+/**
+ * rvt_rkey_ok - check the IB virtual address, length, and RKEY
+ * @qp: qp for validation
+ * @sge: SGE state
+ * @len: length of data
+ * @vaddr: virtual address to place data
+ * @rkey: rkey to check
+ * @acc: access flags
+ *
+ * Return: 1 if successful, otherwise 0.
+ *
+ * increments the reference count upon success
+ */
+int rvt_rkey_ok(struct rvt_qp *qp, struct rvt_sge *sge,
+ u32 len, u64 vaddr, u32 rkey, int acc)
+{
+ struct rvt_dev_info *dev = ib_to_rvt(qp->ibqp.device);
+ struct rvt_lkey_table *rkt = &dev->lkey_table;
+ struct rvt_mregion *mr;
+ unsigned n, m;
+ size_t off;
+
+ /*
+ * We use RKEY == zero for kernel virtual addresses
+ * (see rvt_get_dma_mr and dma.c).
+ */
+ rcu_read_lock();
+ if (rkey == 0) {
+ struct rvt_pd *pd = ibpd_to_rvtpd(qp->ibqp.pd);
+ struct rvt_dev_info *rdi = ib_to_rvt(pd->ibpd.device);
+
+ if (pd->user)
+ goto bail;
+ mr = rcu_dereference(rdi->dma_mr);
+ if (!mr)
+ goto bail;
+ atomic_inc(&mr->refcount);
+ rcu_read_unlock();
+
+ sge->mr = mr;
+ sge->vaddr = (void *)vaddr;
+ sge->length = len;
+ sge->sge_length = len;
+ sge->m = 0;
+ sge->n = 0;
+ goto ok;
+ }
+
+ mr = rcu_dereference(
+ rkt->table[(rkey >> (32 - dev->dparms.lkey_table_size))]);
+ if (unlikely(!mr || mr->lkey != rkey || qp->ibqp.pd != mr->pd))
+ goto bail;
+
+ off = vaddr - mr->iova;
+ if (unlikely(vaddr < mr->iova || off + len > mr->length ||
+ (mr->access_flags & acc) == 0))
+ goto bail;
+ atomic_inc(&mr->refcount);
+ rcu_read_unlock();
+
+ off += mr->offset;
+ if (mr->page_shift) {
+ /*
+ * page sizes are uniform power of 2 so no loop is necessary
+ * entries_spanned_by_off is the number of times the loop below
+ * would have executed.
+ */
+ size_t entries_spanned_by_off;
+
+ entries_spanned_by_off = off >> mr->page_shift;
+ off -= (entries_spanned_by_off << mr->page_shift);
+ m = entries_spanned_by_off / RVT_SEGSZ;
+ n = entries_spanned_by_off % RVT_SEGSZ;
+ } else {
+ m = 0;
+ n = 0;
+ while (off >= mr->map[m]->segs[n].length) {
+ off -= mr->map[m]->segs[n].length;
+ n++;
+ if (n >= RVT_SEGSZ) {
+ m++;
+ n = 0;
+ }
+ }
+ }
+ sge->mr = mr;
+ sge->vaddr = mr->map[m]->segs[n].vaddr + off;
+ sge->length = mr->map[m]->segs[n].length - off;
+ sge->sge_length = len;
+ sge->m = m;
+ sge->n = n;
+ok:
+ return 1;
+bail:
+ rcu_read_unlock();
+ return 0;
+}
+EXPORT_SYMBOL(rvt_rkey_ok);
diff --git a/drivers/infiniband/sw/rdmavt/mr.h b/drivers/infiniband/sw/rdmavt/mr.h
new file mode 100644
index 000000000000..69380512c6d1
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/mr.h
@@ -0,0 +1,92 @@
+#ifndef DEF_RVTMR_H
+#define DEF_RVTMR_H
+
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <rdma/rdma_vt.h>
+struct rvt_fmr {
+ struct ib_fmr ibfmr;
+ struct rvt_mregion mr; /* must be last */
+};
+
+struct rvt_mr {
+ struct ib_mr ibmr;
+ struct ib_umem *umem;
+ struct rvt_mregion mr; /* must be last */
+};
+
+static inline struct rvt_fmr *to_ifmr(struct ib_fmr *ibfmr)
+{
+ return container_of(ibfmr, struct rvt_fmr, ibfmr);
+}
+
+static inline struct rvt_mr *to_imr(struct ib_mr *ibmr)
+{
+ return container_of(ibmr, struct rvt_mr, ibmr);
+}
+
+int rvt_driver_mr_init(struct rvt_dev_info *rdi);
+void rvt_mr_exit(struct rvt_dev_info *rdi);
+
+/* Mem Regions */
+struct ib_mr *rvt_get_dma_mr(struct ib_pd *pd, int acc);
+struct ib_mr *rvt_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
+ u64 virt_addr, int mr_access_flags,
+ struct ib_udata *udata);
+int rvt_dereg_mr(struct ib_mr *ibmr);
+struct ib_mr *rvt_alloc_mr(struct ib_pd *pd,
+ enum ib_mr_type mr_type,
+ u32 max_num_sg);
+struct ib_fmr *rvt_alloc_fmr(struct ib_pd *pd, int mr_access_flags,
+ struct ib_fmr_attr *fmr_attr);
+int rvt_map_phys_fmr(struct ib_fmr *ibfmr, u64 *page_list,
+ int list_len, u64 iova);
+int rvt_unmap_fmr(struct list_head *fmr_list);
+int rvt_dealloc_fmr(struct ib_fmr *ibfmr);
+
+#endif /* DEF_RVTMR_H */
diff --git a/drivers/infiniband/sw/rdmavt/pd.c b/drivers/infiniband/sw/rdmavt/pd.c
new file mode 100644
index 000000000000..d1292f324c67
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/pd.c
@@ -0,0 +1,119 @@
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <linux/slab.h>
+#include "pd.h"
+
+/**
+ * rvt_alloc_pd - allocate a protection domain
+ * @ibdev: ib device
+ * @context: optional user context
+ * @udata: optional user data
+ *
+ * Allocate and keep track of a PD.
+ *
+ * Return: 0 on success
+ */
+struct ib_pd *rvt_alloc_pd(struct ib_device *ibdev,
+ struct ib_ucontext *context,
+ struct ib_udata *udata)
+{
+ struct rvt_dev_info *dev = ib_to_rvt(ibdev);
+ struct rvt_pd *pd;
+ struct ib_pd *ret;
+
+ pd = kmalloc(sizeof(*pd), GFP_KERNEL);
+ if (!pd) {
+ ret = ERR_PTR(-ENOMEM);
+ goto bail;
+ }
+ /*
+ * While we could continue allocating protecetion domains, being
+ * constrained only by system resources. The IBTA spec defines that
+ * there is a max_pd limit that can be set and we need to check for
+ * that.
+ */
+
+ spin_lock(&dev->n_pds_lock);
+ if (dev->n_pds_allocated == dev->dparms.props.max_pd) {
+ spin_unlock(&dev->n_pds_lock);
+ kfree(pd);
+ ret = ERR_PTR(-ENOMEM);
+ goto bail;
+ }
+
+ dev->n_pds_allocated++;
+ spin_unlock(&dev->n_pds_lock);
+
+ /* ib_alloc_pd() will initialize pd->ibpd. */
+ pd->user = udata ? 1 : 0;
+
+ ret = &pd->ibpd;
+
+bail:
+ return ret;
+}
+
+/**
+ * rvt_dealloc_pd - Free PD
+ * @ibpd: Free up PD
+ *
+ * Return: always 0
+ */
+int rvt_dealloc_pd(struct ib_pd *ibpd)
+{
+ struct rvt_pd *pd = ibpd_to_rvtpd(ibpd);
+ struct rvt_dev_info *dev = ib_to_rvt(ibpd->device);
+
+ spin_lock(&dev->n_pds_lock);
+ dev->n_pds_allocated--;
+ spin_unlock(&dev->n_pds_lock);
+
+ kfree(pd);
+
+ return 0;
+}
diff --git a/drivers/infiniband/sw/rdmavt/pd.h b/drivers/infiniband/sw/rdmavt/pd.h
new file mode 100644
index 000000000000..1892ca4a9746
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/pd.h
@@ -0,0 +1,58 @@
+#ifndef DEF_RDMAVTPD_H
+#define DEF_RDMAVTPD_H
+
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <rdma/rdma_vt.h>
+
+struct ib_pd *rvt_alloc_pd(struct ib_device *ibdev,
+ struct ib_ucontext *context,
+ struct ib_udata *udata);
+int rvt_dealloc_pd(struct ib_pd *ibpd);
+
+#endif /* DEF_RDMAVTPD_H */
diff --git a/drivers/infiniband/sw/rdmavt/qp.c b/drivers/infiniband/sw/rdmavt/qp.c
new file mode 100644
index 000000000000..bd82a6948dc8
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/qp.c
@@ -0,0 +1,1696 @@
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <linux/hash.h>
+#include <linux/bitops.h>
+#include <linux/lockdep.h>
+#include <linux/vmalloc.h>
+#include <linux/slab.h>
+#include <rdma/ib_verbs.h>
+#include "qp.h"
+#include "vt.h"
+#include "trace.h"
+
+/*
+ * Note that it is OK to post send work requests in the SQE and ERR
+ * states; rvt_do_send() will process them and generate error
+ * completions as per IB 1.2 C10-96.
+ */
+const int ib_rvt_state_ops[IB_QPS_ERR + 1] = {
+ [IB_QPS_RESET] = 0,
+ [IB_QPS_INIT] = RVT_POST_RECV_OK,
+ [IB_QPS_RTR] = RVT_POST_RECV_OK | RVT_PROCESS_RECV_OK,
+ [IB_QPS_RTS] = RVT_POST_RECV_OK | RVT_PROCESS_RECV_OK |
+ RVT_POST_SEND_OK | RVT_PROCESS_SEND_OK |
+ RVT_PROCESS_NEXT_SEND_OK,
+ [IB_QPS_SQD] = RVT_POST_RECV_OK | RVT_PROCESS_RECV_OK |
+ RVT_POST_SEND_OK | RVT_PROCESS_SEND_OK,
+ [IB_QPS_SQE] = RVT_POST_RECV_OK | RVT_PROCESS_RECV_OK |
+ RVT_POST_SEND_OK | RVT_FLUSH_SEND,
+ [IB_QPS_ERR] = RVT_POST_RECV_OK | RVT_FLUSH_RECV |
+ RVT_POST_SEND_OK | RVT_FLUSH_SEND,
+};
+EXPORT_SYMBOL(ib_rvt_state_ops);
+
+static void get_map_page(struct rvt_qpn_table *qpt,
+ struct rvt_qpn_map *map,
+ gfp_t gfp)
+{
+ unsigned long page = get_zeroed_page(gfp);
+
+ /*
+ * Free the page if someone raced with us installing it.
+ */
+
+ spin_lock(&qpt->lock);
+ if (map->page)
+ free_page(page);
+ else
+ map->page = (void *)page;
+ spin_unlock(&qpt->lock);
+}
+
+/**
+ * init_qpn_table - initialize the QP number table for a device
+ * @qpt: the QPN table
+ */
+static int init_qpn_table(struct rvt_dev_info *rdi, struct rvt_qpn_table *qpt)
+{
+ u32 offset, i;
+ struct rvt_qpn_map *map;
+ int ret = 0;
+
+ if (!(rdi->dparms.qpn_res_end >= rdi->dparms.qpn_res_start))
+ return -EINVAL;
+
+ spin_lock_init(&qpt->lock);
+
+ qpt->last = rdi->dparms.qpn_start;
+ qpt->incr = rdi->dparms.qpn_inc << rdi->dparms.qos_shift;
+
+ /*
+ * Drivers may want some QPs beyond what we need for verbs let them use
+ * our qpn table. No need for two. Lets go ahead and mark the bitmaps
+ * for those. The reserved range must be *after* the range which verbs
+ * will pick from.
+ */
+
+ /* Figure out number of bit maps needed before reserved range */
+ qpt->nmaps = rdi->dparms.qpn_res_start / RVT_BITS_PER_PAGE;
+
+ /* This should always be zero */
+ offset = rdi->dparms.qpn_res_start & RVT_BITS_PER_PAGE_MASK;
+
+ /* Starting with the first reserved bit map */
+ map = &qpt->map[qpt->nmaps];
+
+ rvt_pr_info(rdi, "Reserving QPNs from 0x%x to 0x%x for non-verbs use\n",
+ rdi->dparms.qpn_res_start, rdi->dparms.qpn_res_end);
+ for (i = rdi->dparms.qpn_res_start; i <= rdi->dparms.qpn_res_end; i++) {
+ if (!map->page) {
+ get_map_page(qpt, map, GFP_KERNEL);
+ if (!map->page) {
+ ret = -ENOMEM;
+ break;
+ }
+ }
+ set_bit(offset, map->page);
+ offset++;
+ if (offset == RVT_BITS_PER_PAGE) {
+ /* next page */
+ qpt->nmaps++;
+ map++;
+ offset = 0;
+ }
+ }
+ return ret;
+}
+
+/**
+ * free_qpn_table - free the QP number table for a device
+ * @qpt: the QPN table
+ */
+static void free_qpn_table(struct rvt_qpn_table *qpt)
+{
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(qpt->map); i++)
+ free_page((unsigned long)qpt->map[i].page);
+}
+
+/**
+ * rvt_driver_qp_init - Init driver qp resources
+ * @rdi: rvt dev strucutre
+ *
+ * Return: 0 on success
+ */
+int rvt_driver_qp_init(struct rvt_dev_info *rdi)
+{
+ int i;
+ int ret = -ENOMEM;
+
+ if (!rdi->dparms.qp_table_size)
+ return -EINVAL;
+
+ /*
+ * If driver is not doing any QP allocation then make sure it is
+ * providing the necessary QP functions.
+ */
+ if (!rdi->driver_f.free_all_qps ||
+ !rdi->driver_f.qp_priv_alloc ||
+ !rdi->driver_f.qp_priv_free ||
+ !rdi->driver_f.notify_qp_reset)
+ return -EINVAL;
+
+ /* allocate parent object */
+ rdi->qp_dev = kzalloc_node(sizeof(*rdi->qp_dev), GFP_KERNEL,
+ rdi->dparms.node);
+ if (!rdi->qp_dev)
+ return -ENOMEM;
+
+ /* allocate hash table */
+ rdi->qp_dev->qp_table_size = rdi->dparms.qp_table_size;
+ rdi->qp_dev->qp_table_bits = ilog2(rdi->dparms.qp_table_size);
+ rdi->qp_dev->qp_table =
+ kmalloc_node(rdi->qp_dev->qp_table_size *
+ sizeof(*rdi->qp_dev->qp_table),
+ GFP_KERNEL, rdi->dparms.node);
+ if (!rdi->qp_dev->qp_table)
+ goto no_qp_table;
+
+ for (i = 0; i < rdi->qp_dev->qp_table_size; i++)
+ RCU_INIT_POINTER(rdi->qp_dev->qp_table[i], NULL);
+
+ spin_lock_init(&rdi->qp_dev->qpt_lock);
+
+ /* initialize qpn map */
+ if (init_qpn_table(rdi, &rdi->qp_dev->qpn_table))
+ goto fail_table;
+
+ spin_lock_init(&rdi->n_qps_lock);
+
+ return 0;
+
+fail_table:
+ kfree(rdi->qp_dev->qp_table);
+ free_qpn_table(&rdi->qp_dev->qpn_table);
+
+no_qp_table:
+ kfree(rdi->qp_dev);
+
+ return ret;
+}
+
+/**
+ * free_all_qps - check for QPs still in use
+ * @qpt: the QP table to empty
+ *
+ * There should not be any QPs still in use.
+ * Free memory for table.
+ */
+static unsigned rvt_free_all_qps(struct rvt_dev_info *rdi)
+{
+ unsigned long flags;
+ struct rvt_qp *qp;
+ unsigned n, qp_inuse = 0;
+ spinlock_t *ql; /* work around too long line below */
+
+ if (rdi->driver_f.free_all_qps)
+ qp_inuse = rdi->driver_f.free_all_qps(rdi);
+
+ qp_inuse += rvt_mcast_tree_empty(rdi);
+
+ if (!rdi->qp_dev)
+ return qp_inuse;
+
+ ql = &rdi->qp_dev->qpt_lock;
+ spin_lock_irqsave(ql, flags);
+ for (n = 0; n < rdi->qp_dev->qp_table_size; n++) {
+ qp = rcu_dereference_protected(rdi->qp_dev->qp_table[n],
+ lockdep_is_held(ql));
+ RCU_INIT_POINTER(rdi->qp_dev->qp_table[n], NULL);
+
+ for (; qp; qp = rcu_dereference_protected(qp->next,
+ lockdep_is_held(ql)))
+ qp_inuse++;
+ }
+ spin_unlock_irqrestore(ql, flags);
+ synchronize_rcu();
+ return qp_inuse;
+}
+
+/**
+ * rvt_qp_exit - clean up qps on device exit
+ * @rdi: rvt dev structure
+ *
+ * Check for qp leaks and free resources.
+ */
+void rvt_qp_exit(struct rvt_dev_info *rdi)
+{
+ u32 qps_inuse = rvt_free_all_qps(rdi);
+
+ if (qps_inuse)
+ rvt_pr_err(rdi, "QP memory leak! %u still in use\n",
+ qps_inuse);
+ if (!rdi->qp_dev)
+ return;
+
+ kfree(rdi->qp_dev->qp_table);
+ free_qpn_table(&rdi->qp_dev->qpn_table);
+ kfree(rdi->qp_dev);
+}
+
+static inline unsigned mk_qpn(struct rvt_qpn_table *qpt,
+ struct rvt_qpn_map *map, unsigned off)
+{
+ return (map - qpt->map) * RVT_BITS_PER_PAGE + off;
+}
+
+/**
+ * alloc_qpn - Allocate the next available qpn or zero/one for QP type
+ * IB_QPT_SMI/IB_QPT_GSI
+ *@rdi: rvt device info structure
+ *@qpt: queue pair number table pointer
+ *@port_num: IB port number, 1 based, comes from core
+ *
+ * Return: The queue pair number
+ */
+static int alloc_qpn(struct rvt_dev_info *rdi, struct rvt_qpn_table *qpt,
+ enum ib_qp_type type, u8 port_num, gfp_t gfp)
+{
+ u32 i, offset, max_scan, qpn;
+ struct rvt_qpn_map *map;
+ u32 ret;
+
+ if (rdi->driver_f.alloc_qpn)
+ return rdi->driver_f.alloc_qpn(rdi, qpt, type, port_num, gfp);
+
+ if (type == IB_QPT_SMI || type == IB_QPT_GSI) {
+ unsigned n;
+
+ ret = type == IB_QPT_GSI;
+ n = 1 << (ret + 2 * (port_num - 1));
+ spin_lock(&qpt->lock);
+ if (qpt->flags & n)
+ ret = -EINVAL;
+ else
+ qpt->flags |= n;
+ spin_unlock(&qpt->lock);
+ goto bail;
+ }
+
+ qpn = qpt->last + qpt->incr;
+ if (qpn >= RVT_QPN_MAX)
+ qpn = qpt->incr | ((qpt->last & 1) ^ 1);
+ /* offset carries bit 0 */
+ offset = qpn & RVT_BITS_PER_PAGE_MASK;
+ map = &qpt->map[qpn / RVT_BITS_PER_PAGE];
+ max_scan = qpt->nmaps - !offset;
+ for (i = 0;;) {
+ if (unlikely(!map->page)) {
+ get_map_page(qpt, map, gfp);
+ if (unlikely(!map->page))
+ break;
+ }
+ do {
+ if (!test_and_set_bit(offset, map->page)) {
+ qpt->last = qpn;
+ ret = qpn;
+ goto bail;
+ }
+ offset += qpt->incr;
+ /*
+ * This qpn might be bogus if offset >= BITS_PER_PAGE.
+ * That is OK. It gets re-assigned below
+ */
+ qpn = mk_qpn(qpt, map, offset);
+ } while (offset < RVT_BITS_PER_PAGE && qpn < RVT_QPN_MAX);
+ /*
+ * In order to keep the number of pages allocated to a
+ * minimum, we scan the all existing pages before increasing
+ * the size of the bitmap table.
+ */
+ if (++i > max_scan) {
+ if (qpt->nmaps == RVT_QPNMAP_ENTRIES)
+ break;
+ map = &qpt->map[qpt->nmaps++];
+ /* start at incr with current bit 0 */
+ offset = qpt->incr | (offset & 1);
+ } else if (map < &qpt->map[qpt->nmaps]) {
+ ++map;
+ /* start at incr with current bit 0 */
+ offset = qpt->incr | (offset & 1);
+ } else {
+ map = &qpt->map[0];
+ /* wrap to first map page, invert bit 0 */
+ offset = qpt->incr | ((offset & 1) ^ 1);
+ }
+ /* there can be no bits at shift and below */
+ WARN_ON(offset & (rdi->dparms.qos_shift - 1));
+ qpn = mk_qpn(qpt, map, offset);
+ }
+
+ ret = -ENOMEM;
+
+bail:
+ return ret;
+}
+
+static void free_qpn(struct rvt_qpn_table *qpt, u32 qpn)
+{
+ struct rvt_qpn_map *map;
+
+ map = qpt->map + qpn / RVT_BITS_PER_PAGE;
+ if (map->page)
+ clear_bit(qpn & RVT_BITS_PER_PAGE_MASK, map->page);
+}
+
+/**
+ * rvt_clear_mr_refs - Drop help mr refs
+ * @qp: rvt qp data structure
+ * @clr_sends: If shoudl clear send side or not
+ */
+static void rvt_clear_mr_refs(struct rvt_qp *qp, int clr_sends)
+{
+ unsigned n;
+
+ if (test_and_clear_bit(RVT_R_REWIND_SGE, &qp->r_aflags))
+ rvt_put_ss(&qp->s_rdma_read_sge);
+
+ rvt_put_ss(&qp->r_sge);
+
+ if (clr_sends) {
+ while (qp->s_last != qp->s_head) {
+ struct rvt_swqe *wqe = rvt_get_swqe_ptr(qp, qp->s_last);
+ unsigned i;
+
+ for (i = 0; i < wqe->wr.num_sge; i++) {
+ struct rvt_sge *sge = &wqe->sg_list[i];
+
+ rvt_put_mr(sge->mr);
+ }
+ if (qp->ibqp.qp_type == IB_QPT_UD ||
+ qp->ibqp.qp_type == IB_QPT_SMI ||
+ qp->ibqp.qp_type == IB_QPT_GSI)
+ atomic_dec(&ibah_to_rvtah(
+ wqe->ud_wr.ah)->refcount);
+ if (++qp->s_last >= qp->s_size)
+ qp->s_last = 0;
+ smp_wmb(); /* see qp_set_savail */
+ }
+ if (qp->s_rdma_mr) {
+ rvt_put_mr(qp->s_rdma_mr);
+ qp->s_rdma_mr = NULL;
+ }
+ }
+
+ if (qp->ibqp.qp_type != IB_QPT_RC)
+ return;
+
+ for (n = 0; n < ARRAY_SIZE(qp->s_ack_queue); n++) {
+ struct rvt_ack_entry *e = &qp->s_ack_queue[n];
+
+ if (e->opcode == IB_OPCODE_RC_RDMA_READ_REQUEST &&
+ e->rdma_sge.mr) {
+ rvt_put_mr(e->rdma_sge.mr);
+ e->rdma_sge.mr = NULL;
+ }
+ }
+}
+
+/**
+ * rvt_remove_qp - remove qp form table
+ * @rdi: rvt dev struct
+ * @qp: qp to remove
+ *
+ * Remove the QP from the table so it can't be found asynchronously by
+ * the receive routine.
+ */
+static void rvt_remove_qp(struct rvt_dev_info *rdi, struct rvt_qp *qp)
+{
+ struct rvt_ibport *rvp = rdi->ports[qp->port_num - 1];
+ u32 n = hash_32(qp->ibqp.qp_num, rdi->qp_dev->qp_table_bits);
+ unsigned long flags;
+ int removed = 1;
+
+ spin_lock_irqsave(&rdi->qp_dev->qpt_lock, flags);
+
+ if (rcu_dereference_protected(rvp->qp[0],
+ lockdep_is_held(&rdi->qp_dev->qpt_lock)) == qp) {
+ RCU_INIT_POINTER(rvp->qp[0], NULL);
+ } else if (rcu_dereference_protected(rvp->qp[1],
+ lockdep_is_held(&rdi->qp_dev->qpt_lock)) == qp) {
+ RCU_INIT_POINTER(rvp->qp[1], NULL);
+ } else {
+ struct rvt_qp *q;
+ struct rvt_qp __rcu **qpp;
+
+ removed = 0;
+ qpp = &rdi->qp_dev->qp_table[n];
+ for (; (q = rcu_dereference_protected(*qpp,
+ lockdep_is_held(&rdi->qp_dev->qpt_lock))) != NULL;
+ qpp = &q->next) {
+ if (q == qp) {
+ RCU_INIT_POINTER(*qpp,
+ rcu_dereference_protected(qp->next,
+ lockdep_is_held(&rdi->qp_dev->qpt_lock)));
+ removed = 1;
+ trace_rvt_qpremove(qp, n);
+ break;
+ }
+ }
+ }
+
+ spin_unlock_irqrestore(&rdi->qp_dev->qpt_lock, flags);
+ if (removed) {
+ synchronize_rcu();
+ if (atomic_dec_and_test(&qp->refcount))
+ wake_up(&qp->wait);
+ }
+}
+
+/**
+ * reset_qp - initialize the QP state to the reset state
+ * @qp: the QP to reset
+ * @type: the QP type
+ * r and s lock are required to be held by the caller
+ */
+static void rvt_reset_qp(struct rvt_dev_info *rdi, struct rvt_qp *qp,
+ enum ib_qp_type type)
+{
+ if (qp->state != IB_QPS_RESET) {
+ qp->state = IB_QPS_RESET;
+
+ /* Let drivers flush their waitlist */
+ rdi->driver_f.flush_qp_waiters(qp);
+ qp->s_flags &= ~(RVT_S_TIMER | RVT_S_ANY_WAIT);
+ spin_unlock(&qp->s_lock);
+ spin_unlock(&qp->s_hlock);
+ spin_unlock_irq(&qp->r_lock);
+
+ /* Stop the send queue and the retry timer */
+ rdi->driver_f.stop_send_queue(qp);
+
+ /* Wait for things to stop */
+ rdi->driver_f.quiesce_qp(qp);
+
+ /* take qp out the hash and wait for it to be unused */
+ rvt_remove_qp(rdi, qp);
+ wait_event(qp->wait, !atomic_read(&qp->refcount));
+
+ /* grab the lock b/c it was locked at call time */
+ spin_lock_irq(&qp->r_lock);
+ spin_lock(&qp->s_hlock);
+ spin_lock(&qp->s_lock);
+
+ rvt_clear_mr_refs(qp, 1);
+ }
+
+ /*
+ * Let the driver do any tear down it needs to for a qp
+ * that has been reset
+ */
+ rdi->driver_f.notify_qp_reset(qp);
+
+ qp->remote_qpn = 0;
+ qp->qkey = 0;
+ qp->qp_access_flags = 0;
+ qp->s_flags &= RVT_S_SIGNAL_REQ_WR;
+ qp->s_hdrwords = 0;
+ qp->s_wqe = NULL;
+ qp->s_draining = 0;
+ qp->s_next_psn = 0;
+ qp->s_last_psn = 0;
+ qp->s_sending_psn = 0;
+ qp->s_sending_hpsn = 0;
+ qp->s_psn = 0;
+ qp->r_psn = 0;
+ qp->r_msn = 0;
+ if (type == IB_QPT_RC) {
+ qp->s_state = IB_OPCODE_RC_SEND_LAST;
+ qp->r_state = IB_OPCODE_RC_SEND_LAST;
+ } else {
+ qp->s_state = IB_OPCODE_UC_SEND_LAST;
+ qp->r_state = IB_OPCODE_UC_SEND_LAST;
+ }
+ qp->s_ack_state = IB_OPCODE_RC_ACKNOWLEDGE;
+ qp->r_nak_state = 0;
+ qp->r_aflags = 0;
+ qp->r_flags = 0;
+ qp->s_head = 0;
+ qp->s_tail = 0;
+ qp->s_cur = 0;
+ qp->s_acked = 0;
+ qp->s_last = 0;
+ qp->s_ssn = 1;
+ qp->s_lsn = 0;
+ qp->s_mig_state = IB_MIG_MIGRATED;
+ memset(qp->s_ack_queue, 0, sizeof(qp->s_ack_queue));
+ qp->r_head_ack_queue = 0;
+ qp->s_tail_ack_queue = 0;
+ qp->s_num_rd_atomic = 0;
+ if (qp->r_rq.wq) {
+ qp->r_rq.wq->head = 0;
+ qp->r_rq.wq->tail = 0;
+ }
+ qp->r_sge.num_sge = 0;
+}
+
+/**
+ * rvt_create_qp - create a queue pair for a device
+ * @ibpd: the protection domain who's device we create the queue pair for
+ * @init_attr: the attributes of the queue pair
+ * @udata: user data for libibverbs.so
+ *
+ * Queue pair creation is mostly an rvt issue. However, drivers have their own
+ * unique idea of what queue pair numbers mean. For instance there is a reserved
+ * range for PSM.
+ *
+ * Return: the queue pair on success, otherwise returns an errno.
+ *
+ * Called by the ib_create_qp() core verbs function.
+ */
+struct ib_qp *rvt_create_qp(struct ib_pd *ibpd,
+ struct ib_qp_init_attr *init_attr,
+ struct ib_udata *udata)
+{
+ struct rvt_qp *qp;
+ int err;
+ struct rvt_swqe *swq = NULL;
+ size_t sz;
+ size_t sg_list_sz;
+ struct ib_qp *ret = ERR_PTR(-ENOMEM);
+ struct rvt_dev_info *rdi = ib_to_rvt(ibpd->device);
+ void *priv = NULL;
+ gfp_t gfp;
+
+ if (!rdi)
+ return ERR_PTR(-EINVAL);
+
+ if (init_attr->cap.max_send_sge > rdi->dparms.props.max_sge ||
+ init_attr->cap.max_send_wr > rdi->dparms.props.max_qp_wr ||
+ init_attr->create_flags & ~(IB_QP_CREATE_USE_GFP_NOIO))
+ return ERR_PTR(-EINVAL);
+
+ /* GFP_NOIO is applicable to RC QP's only */
+
+ if (init_attr->create_flags & IB_QP_CREATE_USE_GFP_NOIO &&
+ init_attr->qp_type != IB_QPT_RC)
+ return ERR_PTR(-EINVAL);
+
+ gfp = init_attr->create_flags & IB_QP_CREATE_USE_GFP_NOIO ?
+ GFP_NOIO : GFP_KERNEL;
+
+ /* Check receive queue parameters if no SRQ is specified. */
+ if (!init_attr->srq) {
+ if (init_attr->cap.max_recv_sge > rdi->dparms.props.max_sge ||
+ init_attr->cap.max_recv_wr > rdi->dparms.props.max_qp_wr)
+ return ERR_PTR(-EINVAL);
+
+ if (init_attr->cap.max_send_sge +
+ init_attr->cap.max_send_wr +
+ init_attr->cap.max_recv_sge +
+ init_attr->cap.max_recv_wr == 0)
+ return ERR_PTR(-EINVAL);
+ }
+
+ switch (init_attr->qp_type) {
+ case IB_QPT_SMI:
+ case IB_QPT_GSI:
+ if (init_attr->port_num == 0 ||
+ init_attr->port_num > ibpd->device->phys_port_cnt)
+ return ERR_PTR(-EINVAL);
+ case IB_QPT_UC:
+ case IB_QPT_RC:
+ case IB_QPT_UD:
+ sz = sizeof(struct rvt_sge) *
+ init_attr->cap.max_send_sge +
+ sizeof(struct rvt_swqe);
+ if (gfp == GFP_NOIO)
+ swq = __vmalloc(
+ (init_attr->cap.max_send_wr + 1) * sz,
+ gfp, PAGE_KERNEL);
+ else
+ swq = vmalloc_node(
+ (init_attr->cap.max_send_wr + 1) * sz,
+ rdi->dparms.node);
+ if (!swq)
+ return ERR_PTR(-ENOMEM);
+
+ sz = sizeof(*qp);
+ sg_list_sz = 0;
+ if (init_attr->srq) {
+ struct rvt_srq *srq = ibsrq_to_rvtsrq(init_attr->srq);
+
+ if (srq->rq.max_sge > 1)
+ sg_list_sz = sizeof(*qp->r_sg_list) *
+ (srq->rq.max_sge - 1);
+ } else if (init_attr->cap.max_recv_sge > 1)
+ sg_list_sz = sizeof(*qp->r_sg_list) *
+ (init_attr->cap.max_recv_sge - 1);
+ qp = kzalloc_node(sz + sg_list_sz, gfp, rdi->dparms.node);
+ if (!qp)
+ goto bail_swq;
+
+ RCU_INIT_POINTER(qp->next, NULL);
+
+ /*
+ * Driver needs to set up it's private QP structure and do any
+ * initialization that is needed.
+ */
+ priv = rdi->driver_f.qp_priv_alloc(rdi, qp, gfp);
+ if (!priv)
+ goto bail_qp;
+ qp->priv = priv;
+ qp->timeout_jiffies =
+ usecs_to_jiffies((4096UL * (1UL << qp->timeout)) /
+ 1000UL);
+ if (init_attr->srq) {
+ sz = 0;
+ } else {
+ qp->r_rq.size = init_attr->cap.max_recv_wr + 1;
+ qp->r_rq.max_sge = init_attr->cap.max_recv_sge;
+ sz = (sizeof(struct ib_sge) * qp->r_rq.max_sge) +
+ sizeof(struct rvt_rwqe);
+ if (udata)
+ qp->r_rq.wq = vmalloc_user(
+ sizeof(struct rvt_rwq) +
+ qp->r_rq.size * sz);
+ else if (gfp == GFP_NOIO)
+ qp->r_rq.wq = __vmalloc(
+ sizeof(struct rvt_rwq) +
+ qp->r_rq.size * sz,
+ gfp, PAGE_KERNEL);
+ else
+ qp->r_rq.wq = vmalloc_node(
+ sizeof(struct rvt_rwq) +
+ qp->r_rq.size * sz,
+ rdi->dparms.node);
+ if (!qp->r_rq.wq)
+ goto bail_driver_priv;
+ }
+
+ /*
+ * ib_create_qp() will initialize qp->ibqp
+ * except for qp->ibqp.qp_num.
+ */
+ spin_lock_init(&qp->r_lock);
+ spin_lock_init(&qp->s_hlock);
+ spin_lock_init(&qp->s_lock);
+ spin_lock_init(&qp->r_rq.lock);
+ atomic_set(&qp->refcount, 0);
+ init_waitqueue_head(&qp->wait);
+ init_timer(&qp->s_timer);
+ qp->s_timer.data = (unsigned long)qp;
+ INIT_LIST_HEAD(&qp->rspwait);
+ qp->state = IB_QPS_RESET;
+ qp->s_wq = swq;
+ qp->s_size = init_attr->cap.max_send_wr + 1;
+ qp->s_avail = init_attr->cap.max_send_wr;
+ qp->s_max_sge = init_attr->cap.max_send_sge;
+ if (init_attr->sq_sig_type == IB_SIGNAL_REQ_WR)
+ qp->s_flags = RVT_S_SIGNAL_REQ_WR;
+
+ err = alloc_qpn(rdi, &rdi->qp_dev->qpn_table,
+ init_attr->qp_type,
+ init_attr->port_num, gfp);
+ if (err < 0) {
+ ret = ERR_PTR(err);
+ goto bail_rq_wq;
+ }
+ qp->ibqp.qp_num = err;
+ qp->port_num = init_attr->port_num;
+ rvt_reset_qp(rdi, qp, init_attr->qp_type);
+ break;
+
+ default:
+ /* Don't support raw QPs */
+ return ERR_PTR(-EINVAL);
+ }
+
+ init_attr->cap.max_inline_data = 0;
+
+ /*
+ * Return the address of the RWQ as the offset to mmap.
+ * See rvt_mmap() for details.
+ */
+ if (udata && udata->outlen >= sizeof(__u64)) {
+ if (!qp->r_rq.wq) {
+ __u64 offset = 0;
+
+ err = ib_copy_to_udata(udata, &offset,
+ sizeof(offset));
+ if (err) {
+ ret = ERR_PTR(err);
+ goto bail_qpn;
+ }
+ } else {
+ u32 s = sizeof(struct rvt_rwq) + qp->r_rq.size * sz;
+
+ qp->ip = rvt_create_mmap_info(rdi, s,
+ ibpd->uobject->context,
+ qp->r_rq.wq);
+ if (!qp->ip) {
+ ret = ERR_PTR(-ENOMEM);
+ goto bail_qpn;
+ }
+
+ err = ib_copy_to_udata(udata, &qp->ip->offset,
+ sizeof(qp->ip->offset));
+ if (err) {
+ ret = ERR_PTR(err);
+ goto bail_ip;
+ }
+ }
+ qp->pid = current->pid;
+ }
+
+ spin_lock(&rdi->n_qps_lock);
+ if (rdi->n_qps_allocated == rdi->dparms.props.max_qp) {
+ spin_unlock(&rdi->n_qps_lock);
+ ret = ERR_PTR(-ENOMEM);
+ goto bail_ip;
+ }
+
+ rdi->n_qps_allocated++;
+ /*
+ * Maintain a busy_jiffies variable that will be added to the timeout
+ * period in mod_retry_timer and add_retry_timer. This busy jiffies
+ * is scaled by the number of rc qps created for the device to reduce
+ * the number of timeouts occurring when there is a large number of
+ * qps. busy_jiffies is incremented every rc qp scaling interval.
+ * The scaling interval is selected based on extensive performance
+ * evaluation of targeted workloads.
+ */
+ if (init_attr->qp_type == IB_QPT_RC) {
+ rdi->n_rc_qps++;
+ rdi->busy_jiffies = rdi->n_rc_qps / RC_QP_SCALING_INTERVAL;
+ }
+ spin_unlock(&rdi->n_qps_lock);
+
+ if (qp->ip) {
+ spin_lock_irq(&rdi->pending_lock);
+ list_add(&qp->ip->pending_mmaps, &rdi->pending_mmaps);
+ spin_unlock_irq(&rdi->pending_lock);
+ }
+
+ ret = &qp->ibqp;
+
+ /*
+ * We have our QP and its good, now keep track of what types of opcodes
+ * can be processed on this QP. We do this by keeping track of what the
+ * 3 high order bits of the opcode are.
+ */
+ switch (init_attr->qp_type) {
+ case IB_QPT_SMI:
+ case IB_QPT_GSI:
+ case IB_QPT_UD:
+ qp->allowed_ops = IB_OPCODE_UD_SEND_ONLY & RVT_OPCODE_QP_MASK;
+ break;
+ case IB_QPT_RC:
+ qp->allowed_ops = IB_OPCODE_RC_SEND_ONLY & RVT_OPCODE_QP_MASK;
+ break;
+ case IB_QPT_UC:
+ qp->allowed_ops = IB_OPCODE_UC_SEND_ONLY & RVT_OPCODE_QP_MASK;
+ break;
+ default:
+ ret = ERR_PTR(-EINVAL);
+ goto bail_ip;
+ }
+
+ return ret;
+
+bail_ip:
+ kref_put(&qp->ip->ref, rvt_release_mmap_info);
+
+bail_qpn:
+ free_qpn(&rdi->qp_dev->qpn_table, qp->ibqp.qp_num);
+
+bail_rq_wq:
+ vfree(qp->r_rq.wq);
+
+bail_driver_priv:
+ rdi->driver_f.qp_priv_free(rdi, qp);
+
+bail_qp:
+ kfree(qp);
+
+bail_swq:
+ vfree(swq);
+
+ return ret;
+}
+
+/**
+ * rvt_error_qp - put a QP into the error state
+ * @qp: the QP to put into the error state
+ * @err: the receive completion error to signal if a RWQE is active
+ *
+ * Flushes both send and receive work queues.
+ *
+ * Return: true if last WQE event should be generated.
+ * The QP r_lock and s_lock should be held and interrupts disabled.
+ * If we are already in error state, just return.
+ */
+int rvt_error_qp(struct rvt_qp *qp, enum ib_wc_status err)
+{
+ struct ib_wc wc;
+ int ret = 0;
+ struct rvt_dev_info *rdi = ib_to_rvt(qp->ibqp.device);
+
+ if (qp->state == IB_QPS_ERR || qp->state == IB_QPS_RESET)
+ goto bail;
+
+ qp->state = IB_QPS_ERR;
+
+ if (qp->s_flags & (RVT_S_TIMER | RVT_S_WAIT_RNR)) {
+ qp->s_flags &= ~(RVT_S_TIMER | RVT_S_WAIT_RNR);
+ del_timer(&qp->s_timer);
+ }
+
+ if (qp->s_flags & RVT_S_ANY_WAIT_SEND)
+ qp->s_flags &= ~RVT_S_ANY_WAIT_SEND;
+
+ rdi->driver_f.notify_error_qp(qp);
+
+ /* Schedule the sending tasklet to drain the send work queue. */
+ if (ACCESS_ONCE(qp->s_last) != qp->s_head)
+ rdi->driver_f.schedule_send(qp);
+
+ rvt_clear_mr_refs(qp, 0);
+
+ memset(&wc, 0, sizeof(wc));
+ wc.qp = &qp->ibqp;
+ wc.opcode = IB_WC_RECV;
+
+ if (test_and_clear_bit(RVT_R_WRID_VALID, &qp->r_aflags)) {
+ wc.wr_id = qp->r_wr_id;
+ wc.status = err;
+ rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.recv_cq), &wc, 1);
+ }
+ wc.status = IB_WC_WR_FLUSH_ERR;
+
+ if (qp->r_rq.wq) {
+ struct rvt_rwq *wq;
+ u32 head;
+ u32 tail;
+
+ spin_lock(&qp->r_rq.lock);
+
+ /* sanity check pointers before trusting them */
+ wq = qp->r_rq.wq;
+ head = wq->head;
+ if (head >= qp->r_rq.size)
+ head = 0;
+ tail = wq->tail;
+ if (tail >= qp->r_rq.size)
+ tail = 0;
+ while (tail != head) {
+ wc.wr_id = rvt_get_rwqe_ptr(&qp->r_rq, tail)->wr_id;
+ if (++tail >= qp->r_rq.size)
+ tail = 0;
+ rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.recv_cq), &wc, 1);
+ }
+ wq->tail = tail;
+
+ spin_unlock(&qp->r_rq.lock);
+ } else if (qp->ibqp.event_handler) {
+ ret = 1;
+ }
+
+bail:
+ return ret;
+}
+EXPORT_SYMBOL(rvt_error_qp);
+
+/*
+ * Put the QP into the hash table.
+ * The hash table holds a reference to the QP.
+ */
+static void rvt_insert_qp(struct rvt_dev_info *rdi, struct rvt_qp *qp)
+{
+ struct rvt_ibport *rvp = rdi->ports[qp->port_num - 1];
+ unsigned long flags;
+
+ atomic_inc(&qp->refcount);
+ spin_lock_irqsave(&rdi->qp_dev->qpt_lock, flags);
+
+ if (qp->ibqp.qp_num <= 1) {
+ rcu_assign_pointer(rvp->qp[qp->ibqp.qp_num], qp);
+ } else {
+ u32 n = hash_32(qp->ibqp.qp_num, rdi->qp_dev->qp_table_bits);
+
+ qp->next = rdi->qp_dev->qp_table[n];
+ rcu_assign_pointer(rdi->qp_dev->qp_table[n], qp);
+ trace_rvt_qpinsert(qp, n);
+ }
+
+ spin_unlock_irqrestore(&rdi->qp_dev->qpt_lock, flags);
+}
+
+/**
+ * qib_modify_qp - modify the attributes of a queue pair
+ * @ibqp: the queue pair who's attributes we're modifying
+ * @attr: the new attributes
+ * @attr_mask: the mask of attributes to modify
+ * @udata: user data for libibverbs.so
+ *
+ * Return: 0 on success, otherwise returns an errno.
+ */
+int rvt_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
+ int attr_mask, struct ib_udata *udata)
+{
+ struct rvt_dev_info *rdi = ib_to_rvt(ibqp->device);
+ struct rvt_qp *qp = ibqp_to_rvtqp(ibqp);
+ enum ib_qp_state cur_state, new_state;
+ struct ib_event ev;
+ int lastwqe = 0;
+ int mig = 0;
+ int pmtu = 0; /* for gcc warning only */
+ enum rdma_link_layer link;
+
+ link = rdma_port_get_link_layer(ibqp->device, qp->port_num);
+
+ spin_lock_irq(&qp->r_lock);
+ spin_lock(&qp->s_hlock);
+ spin_lock(&qp->s_lock);
+
+ cur_state = attr_mask & IB_QP_CUR_STATE ?
+ attr->cur_qp_state : qp->state;
+ new_state = attr_mask & IB_QP_STATE ? attr->qp_state : cur_state;
+
+ if (!ib_modify_qp_is_ok(cur_state, new_state, ibqp->qp_type,
+ attr_mask, link))
+ goto inval;
+
+ if (rdi->driver_f.check_modify_qp &&
+ rdi->driver_f.check_modify_qp(qp, attr, attr_mask, udata))
+ goto inval;
+
+ if (attr_mask & IB_QP_AV) {
+ if (attr->ah_attr.dlid >= be16_to_cpu(IB_MULTICAST_LID_BASE))
+ goto inval;
+ if (rvt_check_ah(qp->ibqp.device, &attr->ah_attr))
+ goto inval;
+ }
+
+ if (attr_mask & IB_QP_ALT_PATH) {
+ if (attr->alt_ah_attr.dlid >=
+ be16_to_cpu(IB_MULTICAST_LID_BASE))
+ goto inval;
+ if (rvt_check_ah(qp->ibqp.device, &attr->alt_ah_attr))
+ goto inval;
+ if (attr->alt_pkey_index >= rvt_get_npkeys(rdi))
+ goto inval;
+ }
+
+ if (attr_mask & IB_QP_PKEY_INDEX)
+ if (attr->pkey_index >= rvt_get_npkeys(rdi))
+ goto inval;
+
+ if (attr_mask & IB_QP_MIN_RNR_TIMER)
+ if (attr->min_rnr_timer > 31)
+ goto inval;
+
+ if (attr_mask & IB_QP_PORT)
+ if (qp->ibqp.qp_type == IB_QPT_SMI ||
+ qp->ibqp.qp_type == IB_QPT_GSI ||
+ attr->port_num == 0 ||
+ attr->port_num > ibqp->device->phys_port_cnt)
+ goto inval;
+
+ if (attr_mask & IB_QP_DEST_QPN)
+ if (attr->dest_qp_num > RVT_QPN_MASK)
+ goto inval;
+
+ if (attr_mask & IB_QP_RETRY_CNT)
+ if (attr->retry_cnt > 7)
+ goto inval;
+
+ if (attr_mask & IB_QP_RNR_RETRY)
+ if (attr->rnr_retry > 7)
+ goto inval;
+
+ /*
+ * Don't allow invalid path_mtu values. OK to set greater
+ * than the active mtu (or even the max_cap, if we have tuned
+ * that to a small mtu. We'll set qp->path_mtu
+ * to the lesser of requested attribute mtu and active,
+ * for packetizing messages.
+ * Note that the QP port has to be set in INIT and MTU in RTR.
+ */
+ if (attr_mask & IB_QP_PATH_MTU) {
+ pmtu = rdi->driver_f.get_pmtu_from_attr(rdi, qp, attr);
+ if (pmtu < 0)
+ goto inval;
+ }
+
+ if (attr_mask & IB_QP_PATH_MIG_STATE) {
+ if (attr->path_mig_state == IB_MIG_REARM) {
+ if (qp->s_mig_state == IB_MIG_ARMED)
+ goto inval;
+ if (new_state != IB_QPS_RTS)
+ goto inval;
+ } else if (attr->path_mig_state == IB_MIG_MIGRATED) {
+ if (qp->s_mig_state == IB_MIG_REARM)
+ goto inval;
+ if (new_state != IB_QPS_RTS && new_state != IB_QPS_SQD)
+ goto inval;
+ if (qp->s_mig_state == IB_MIG_ARMED)
+ mig = 1;
+ } else {
+ goto inval;
+ }
+ }
+
+ if (attr_mask & IB_QP_MAX_DEST_RD_ATOMIC)
+ if (attr->max_dest_rd_atomic > rdi->dparms.max_rdma_atomic)
+ goto inval;
+
+ switch (new_state) {
+ case IB_QPS_RESET:
+ if (qp->state != IB_QPS_RESET)
+ rvt_reset_qp(rdi, qp, ibqp->qp_type);
+ break;
+
+ case IB_QPS_RTR:
+ /* Allow event to re-trigger if QP set to RTR more than once */
+ qp->r_flags &= ~RVT_R_COMM_EST;
+ qp->state = new_state;
+ break;
+
+ case IB_QPS_SQD:
+ qp->s_draining = qp->s_last != qp->s_cur;
+ qp->state = new_state;
+ break;
+
+ case IB_QPS_SQE:
+ if (qp->ibqp.qp_type == IB_QPT_RC)
+ goto inval;
+ qp->state = new_state;
+ break;
+
+ case IB_QPS_ERR:
+ lastwqe = rvt_error_qp(qp, IB_WC_WR_FLUSH_ERR);
+ break;
+
+ default:
+ qp->state = new_state;
+ break;
+ }
+
+ if (attr_mask & IB_QP_PKEY_INDEX)
+ qp->s_pkey_index = attr->pkey_index;
+
+ if (attr_mask & IB_QP_PORT)
+ qp->port_num = attr->port_num;
+
+ if (attr_mask & IB_QP_DEST_QPN)
+ qp->remote_qpn = attr->dest_qp_num;
+
+ if (attr_mask & IB_QP_SQ_PSN) {
+ qp->s_next_psn = attr->sq_psn & rdi->dparms.psn_modify_mask;
+ qp->s_psn = qp->s_next_psn;
+ qp->s_sending_psn = qp->s_next_psn;
+ qp->s_last_psn = qp->s_next_psn - 1;
+ qp->s_sending_hpsn = qp->s_last_psn;
+ }
+
+ if (attr_mask & IB_QP_RQ_PSN)
+ qp->r_psn = attr->rq_psn & rdi->dparms.psn_modify_mask;
+
+ if (attr_mask & IB_QP_ACCESS_FLAGS)
+ qp->qp_access_flags = attr->qp_access_flags;
+
+ if (attr_mask & IB_QP_AV) {
+ qp->remote_ah_attr = attr->ah_attr;
+ qp->s_srate = attr->ah_attr.static_rate;
+ qp->srate_mbps = ib_rate_to_mbps(qp->s_srate);
+ }
+
+ if (attr_mask & IB_QP_ALT_PATH) {
+ qp->alt_ah_attr = attr->alt_ah_attr;
+ qp->s_alt_pkey_index = attr->alt_pkey_index;
+ }
+
+ if (attr_mask & IB_QP_PATH_MIG_STATE) {
+ qp->s_mig_state = attr->path_mig_state;
+ if (mig) {
+ qp->remote_ah_attr = qp->alt_ah_attr;
+ qp->port_num = qp->alt_ah_attr.port_num;
+ qp->s_pkey_index = qp->s_alt_pkey_index;
+ }
+ }
+
+ if (attr_mask & IB_QP_PATH_MTU) {
+ qp->pmtu = rdi->driver_f.mtu_from_qp(rdi, qp, pmtu);
+ qp->path_mtu = rdi->driver_f.mtu_to_path_mtu(qp->pmtu);
+ qp->log_pmtu = ilog2(qp->pmtu);
+ }
+
+ if (attr_mask & IB_QP_RETRY_CNT) {
+ qp->s_retry_cnt = attr->retry_cnt;
+ qp->s_retry = attr->retry_cnt;
+ }
+
+ if (attr_mask & IB_QP_RNR_RETRY) {
+ qp->s_rnr_retry_cnt = attr->rnr_retry;
+ qp->s_rnr_retry = attr->rnr_retry;
+ }
+
+ if (attr_mask & IB_QP_MIN_RNR_TIMER)
+ qp->r_min_rnr_timer = attr->min_rnr_timer;
+
+ if (attr_mask & IB_QP_TIMEOUT) {
+ qp->timeout = attr->timeout;
+ qp->timeout_jiffies =
+ usecs_to_jiffies((4096UL * (1UL << qp->timeout)) /
+ 1000UL);
+ }
+
+ if (attr_mask & IB_QP_QKEY)
+ qp->qkey = attr->qkey;
+
+ if (attr_mask & IB_QP_MAX_DEST_RD_ATOMIC)
+ qp->r_max_rd_atomic = attr->max_dest_rd_atomic;
+
+ if (attr_mask & IB_QP_MAX_QP_RD_ATOMIC)
+ qp->s_max_rd_atomic = attr->max_rd_atomic;
+
+ if (rdi->driver_f.modify_qp)
+ rdi->driver_f.modify_qp(qp, attr, attr_mask, udata);
+
+ spin_unlock(&qp->s_lock);
+ spin_unlock(&qp->s_hlock);
+ spin_unlock_irq(&qp->r_lock);
+
+ if (cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT)
+ rvt_insert_qp(rdi, qp);
+
+ if (lastwqe) {
+ ev.device = qp->ibqp.device;
+ ev.element.qp = &qp->ibqp;
+ ev.event = IB_EVENT_QP_LAST_WQE_REACHED;
+ qp->ibqp.event_handler(&ev, qp->ibqp.qp_context);
+ }
+ if (mig) {
+ ev.device = qp->ibqp.device;
+ ev.element.qp = &qp->ibqp;
+ ev.event = IB_EVENT_PATH_MIG;
+ qp->ibqp.event_handler(&ev, qp->ibqp.qp_context);
+ }
+ return 0;
+
+inval:
+ spin_unlock(&qp->s_lock);
+ spin_unlock(&qp->s_hlock);
+ spin_unlock_irq(&qp->r_lock);
+ return -EINVAL;
+}
+
+/** rvt_free_qpn - Free a qpn from the bit map
+ * @qpt: QP table
+ * @qpn: queue pair number to free
+ */
+static void rvt_free_qpn(struct rvt_qpn_table *qpt, u32 qpn)
+{
+ struct rvt_qpn_map *map;
+
+ map = qpt->map + qpn / RVT_BITS_PER_PAGE;
+ if (map->page)
+ clear_bit(qpn & RVT_BITS_PER_PAGE_MASK, map->page);
+}
+
+/**
+ * rvt_destroy_qp - destroy a queue pair
+ * @ibqp: the queue pair to destroy
+ *
+ * Note that this can be called while the QP is actively sending or
+ * receiving!
+ *
+ * Return: 0 on success.
+ */
+int rvt_destroy_qp(struct ib_qp *ibqp)
+{
+ struct rvt_qp *qp = ibqp_to_rvtqp(ibqp);
+ struct rvt_dev_info *rdi = ib_to_rvt(ibqp->device);
+
+ spin_lock_irq(&qp->r_lock);
+ spin_lock(&qp->s_hlock);
+ spin_lock(&qp->s_lock);
+ rvt_reset_qp(rdi, qp, ibqp->qp_type);
+ spin_unlock(&qp->s_lock);
+ spin_unlock(&qp->s_hlock);
+ spin_unlock_irq(&qp->r_lock);
+
+ /* qpn is now available for use again */
+ rvt_free_qpn(&rdi->qp_dev->qpn_table, qp->ibqp.qp_num);
+
+ spin_lock(&rdi->n_qps_lock);
+ rdi->n_qps_allocated--;
+ if (qp->ibqp.qp_type == IB_QPT_RC) {
+ rdi->n_rc_qps--;
+ rdi->busy_jiffies = rdi->n_rc_qps / RC_QP_SCALING_INTERVAL;
+ }
+ spin_unlock(&rdi->n_qps_lock);
+
+ if (qp->ip)
+ kref_put(&qp->ip->ref, rvt_release_mmap_info);
+ else
+ vfree(qp->r_rq.wq);
+ vfree(qp->s_wq);
+ rdi->driver_f.qp_priv_free(rdi, qp);
+ kfree(qp);
+ return 0;
+}
+
+/**
+ * rvt_query_qp - query an ipbq
+ * @ibqp: IB qp to query
+ * @attr: attr struct to fill in
+ * @attr_mask: attr mask ignored
+ * @init_attr: struct to fill in
+ *
+ * Return: always 0
+ */
+int rvt_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
+ int attr_mask, struct ib_qp_init_attr *init_attr)
+{
+ struct rvt_qp *qp = ibqp_to_rvtqp(ibqp);
+ struct rvt_dev_info *rdi = ib_to_rvt(ibqp->device);
+
+ attr->qp_state = qp->state;
+ attr->cur_qp_state = attr->qp_state;
+ attr->path_mtu = qp->path_mtu;
+ attr->path_mig_state = qp->s_mig_state;
+ attr->qkey = qp->qkey;
+ attr->rq_psn = qp->r_psn & rdi->dparms.psn_mask;
+ attr->sq_psn = qp->s_next_psn & rdi->dparms.psn_mask;
+ attr->dest_qp_num = qp->remote_qpn;
+ attr->qp_access_flags = qp->qp_access_flags;
+ attr->cap.max_send_wr = qp->s_size - 1;
+ attr->cap.max_recv_wr = qp->ibqp.srq ? 0 : qp->r_rq.size - 1;
+ attr->cap.max_send_sge = qp->s_max_sge;
+ attr->cap.max_recv_sge = qp->r_rq.max_sge;
+ attr->cap.max_inline_data = 0;
+ attr->ah_attr = qp->remote_ah_attr;
+ attr->alt_ah_attr = qp->alt_ah_attr;
+ attr->pkey_index = qp->s_pkey_index;
+ attr->alt_pkey_index = qp->s_alt_pkey_index;
+ attr->en_sqd_async_notify = 0;
+ attr->sq_draining = qp->s_draining;
+ attr->max_rd_atomic = qp->s_max_rd_atomic;
+ attr->max_dest_rd_atomic = qp->r_max_rd_atomic;
+ attr->min_rnr_timer = qp->r_min_rnr_timer;
+ attr->port_num = qp->port_num;
+ attr->timeout = qp->timeout;
+ attr->retry_cnt = qp->s_retry_cnt;
+ attr->rnr_retry = qp->s_rnr_retry_cnt;
+ attr->alt_port_num = qp->alt_ah_attr.port_num;
+ attr->alt_timeout = qp->alt_timeout;
+
+ init_attr->event_handler = qp->ibqp.event_handler;
+ init_attr->qp_context = qp->ibqp.qp_context;
+ init_attr->send_cq = qp->ibqp.send_cq;
+ init_attr->recv_cq = qp->ibqp.recv_cq;
+ init_attr->srq = qp->ibqp.srq;
+ init_attr->cap = attr->cap;
+ if (qp->s_flags & RVT_S_SIGNAL_REQ_WR)
+ init_attr->sq_sig_type = IB_SIGNAL_REQ_WR;
+ else
+ init_attr->sq_sig_type = IB_SIGNAL_ALL_WR;
+ init_attr->qp_type = qp->ibqp.qp_type;
+ init_attr->port_num = qp->port_num;
+ return 0;
+}
+
+/**
+ * rvt_post_receive - post a receive on a QP
+ * @ibqp: the QP to post the receive on
+ * @wr: the WR to post
+ * @bad_wr: the first bad WR is put here
+ *
+ * This may be called from interrupt context.
+ *
+ * Return: 0 on success otherwise errno
+ */
+int rvt_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr,
+ struct ib_recv_wr **bad_wr)
+{
+ struct rvt_qp *qp = ibqp_to_rvtqp(ibqp);
+ struct rvt_rwq *wq = qp->r_rq.wq;
+ unsigned long flags;
+ int qp_err_flush = (ib_rvt_state_ops[qp->state] & RVT_FLUSH_RECV) &&
+ !qp->ibqp.srq;
+
+ /* Check that state is OK to post receive. */
+ if (!(ib_rvt_state_ops[qp->state] & RVT_POST_RECV_OK) || !wq) {
+ *bad_wr = wr;
+ return -EINVAL;
+ }
+
+ for (; wr; wr = wr->next) {
+ struct rvt_rwqe *wqe;
+ u32 next;
+ int i;
+
+ if ((unsigned)wr->num_sge > qp->r_rq.max_sge) {
+ *bad_wr = wr;
+ return -EINVAL;
+ }
+
+ spin_lock_irqsave(&qp->r_rq.lock, flags);
+ next = wq->head + 1;
+ if (next >= qp->r_rq.size)
+ next = 0;
+ if (next == wq->tail) {
+ spin_unlock_irqrestore(&qp->r_rq.lock, flags);
+ *bad_wr = wr;
+ return -ENOMEM;
+ }
+ if (unlikely(qp_err_flush)) {
+ struct ib_wc wc;
+
+ memset(&wc, 0, sizeof(wc));
+ wc.qp = &qp->ibqp;
+ wc.opcode = IB_WC_RECV;
+ wc.wr_id = wr->wr_id;
+ wc.status = IB_WC_WR_FLUSH_ERR;
+ rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.recv_cq), &wc, 1);
+ } else {
+ wqe = rvt_get_rwqe_ptr(&qp->r_rq, wq->head);
+ wqe->wr_id = wr->wr_id;
+ wqe->num_sge = wr->num_sge;
+ for (i = 0; i < wr->num_sge; i++)
+ wqe->sg_list[i] = wr->sg_list[i];
+ /*
+ * Make sure queue entry is written
+ * before the head index.
+ */
+ smp_wmb();
+ wq->head = next;
+ }
+ spin_unlock_irqrestore(&qp->r_rq.lock, flags);
+ }
+ return 0;
+}
+
+/**
+ * qp_get_savail - return number of avail send entries
+ *
+ * @qp - the qp
+ *
+ * This assumes the s_hlock is held but the s_last
+ * qp variable is uncontrolled.
+ */
+static inline u32 qp_get_savail(struct rvt_qp *qp)
+{
+ u32 slast;
+ u32 ret;
+
+ smp_read_barrier_depends(); /* see rc.c */
+ slast = ACCESS_ONCE(qp->s_last);
+ if (qp->s_head >= slast)
+ ret = qp->s_size - (qp->s_head - slast);
+ else
+ ret = slast - qp->s_head;
+ return ret - 1;
+}
+
+/**
+ * rvt_post_one_wr - post one RC, UC, or UD send work request
+ * @qp: the QP to post on
+ * @wr: the work request to send
+ */
+static int rvt_post_one_wr(struct rvt_qp *qp,
+ struct ib_send_wr *wr,
+ int *call_send)
+{
+ struct rvt_swqe *wqe;
+ u32 next;
+ int i;
+ int j;
+ int acc;
+ struct rvt_lkey_table *rkt;
+ struct rvt_pd *pd;
+ struct rvt_dev_info *rdi = ib_to_rvt(qp->ibqp.device);
+ u8 log_pmtu;
+ int ret;
+
+ /* IB spec says that num_sge == 0 is OK. */
+ if (unlikely(wr->num_sge > qp->s_max_sge))
+ return -EINVAL;
+
+ /*
+ * Don't allow RDMA reads or atomic operations on UC or
+ * undefined operations.
+ * Make sure buffer is large enough to hold the result for atomics.
+ */
+ if (qp->ibqp.qp_type == IB_QPT_UC) {
+ if ((unsigned)wr->opcode >= IB_WR_RDMA_READ)
+ return -EINVAL;
+ } else if (qp->ibqp.qp_type != IB_QPT_RC) {
+ /* Check IB_QPT_SMI, IB_QPT_GSI, IB_QPT_UD opcode */
+ if (wr->opcode != IB_WR_SEND &&
+ wr->opcode != IB_WR_SEND_WITH_IMM)
+ return -EINVAL;
+ /* Check UD destination address PD */
+ if (qp->ibqp.pd != ud_wr(wr)->ah->pd)
+ return -EINVAL;
+ } else if ((unsigned)wr->opcode > IB_WR_ATOMIC_FETCH_AND_ADD) {
+ return -EINVAL;
+ } else if (wr->opcode >= IB_WR_ATOMIC_CMP_AND_SWP &&
+ (wr->num_sge == 0 ||
+ wr->sg_list[0].length < sizeof(u64) ||
+ wr->sg_list[0].addr & (sizeof(u64) - 1))) {
+ return -EINVAL;
+ } else if (wr->opcode >= IB_WR_RDMA_READ && !qp->s_max_rd_atomic) {
+ return -EINVAL;
+ }
+ /* check for avail */
+ if (unlikely(!qp->s_avail)) {
+ qp->s_avail = qp_get_savail(qp);
+ if (WARN_ON(qp->s_avail > (qp->s_size - 1)))
+ rvt_pr_err(rdi,
+ "More avail entries than QP RB size.\nQP: %u, size: %u, avail: %u\nhead: %u, tail: %u, cur: %u, acked: %u, last: %u",
+ qp->ibqp.qp_num, qp->s_size, qp->s_avail,
+ qp->s_head, qp->s_tail, qp->s_cur,
+ qp->s_acked, qp->s_last);
+ if (!qp->s_avail)
+ return -ENOMEM;
+ }
+ next = qp->s_head + 1;
+ if (next >= qp->s_size)
+ next = 0;
+
+ rkt = &rdi->lkey_table;
+ pd = ibpd_to_rvtpd(qp->ibqp.pd);
+ wqe = rvt_get_swqe_ptr(qp, qp->s_head);
+
+ if (qp->ibqp.qp_type != IB_QPT_UC &&
+ qp->ibqp.qp_type != IB_QPT_RC)
+ memcpy(&wqe->ud_wr, ud_wr(wr), sizeof(wqe->ud_wr));
+ else if (wr->opcode == IB_WR_RDMA_WRITE_WITH_IMM ||
+ wr->opcode == IB_WR_RDMA_WRITE ||
+ wr->opcode == IB_WR_RDMA_READ)
+ memcpy(&wqe->rdma_wr, rdma_wr(wr), sizeof(wqe->rdma_wr));
+ else if (wr->opcode == IB_WR_ATOMIC_CMP_AND_SWP ||
+ wr->opcode == IB_WR_ATOMIC_FETCH_AND_ADD)
+ memcpy(&wqe->atomic_wr, atomic_wr(wr), sizeof(wqe->atomic_wr));
+ else
+ memcpy(&wqe->wr, wr, sizeof(wqe->wr));
+
+ wqe->length = 0;
+ j = 0;
+ if (wr->num_sge) {
+ acc = wr->opcode >= IB_WR_RDMA_READ ?
+ IB_ACCESS_LOCAL_WRITE : 0;
+ for (i = 0; i < wr->num_sge; i++) {
+ u32 length = wr->sg_list[i].length;
+ int ok;
+
+ if (length == 0)
+ continue;
+ ok = rvt_lkey_ok(rkt, pd, &wqe->sg_list[j],
+ &wr->sg_list[i], acc);
+ if (!ok) {
+ ret = -EINVAL;
+ goto bail_inval_free;
+ }
+ wqe->length += length;
+ j++;
+ }
+ wqe->wr.num_sge = j;
+ }
+
+ /* general part of wqe valid - allow for driver checks */
+ if (rdi->driver_f.check_send_wqe) {
+ ret = rdi->driver_f.check_send_wqe(qp, wqe);
+ if (ret < 0)
+ goto bail_inval_free;
+ if (ret)
+ *call_send = ret;
+ }
+
+ log_pmtu = qp->log_pmtu;
+ if (qp->ibqp.qp_type != IB_QPT_UC &&
+ qp->ibqp.qp_type != IB_QPT_RC) {
+ struct rvt_ah *ah = ibah_to_rvtah(wqe->ud_wr.ah);
+
+ log_pmtu = ah->log_pmtu;
+ atomic_inc(&ibah_to_rvtah(ud_wr(wr)->ah)->refcount);
+ }
+
+ wqe->ssn = qp->s_ssn++;
+ wqe->psn = qp->s_next_psn;
+ wqe->lpsn = wqe->psn +
+ (wqe->length ? ((wqe->length - 1) >> log_pmtu) : 0);
+ qp->s_next_psn = wqe->lpsn + 1;
+ trace_rvt_post_one_wr(qp, wqe);
+ smp_wmb(); /* see request builders */
+ qp->s_avail--;
+ qp->s_head = next;
+
+ return 0;
+
+bail_inval_free:
+ /* release mr holds */
+ while (j) {
+ struct rvt_sge *sge = &wqe->sg_list[--j];
+
+ rvt_put_mr(sge->mr);
+ }
+ return ret;
+}
+
+/**
+ * rvt_post_send - post a send on a QP
+ * @ibqp: the QP to post the send on
+ * @wr: the list of work requests to post
+ * @bad_wr: the first bad WR is put here
+ *
+ * This may be called from interrupt context.
+ *
+ * Return: 0 on success else errno
+ */
+int rvt_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
+ struct ib_send_wr **bad_wr)
+{
+ struct rvt_qp *qp = ibqp_to_rvtqp(ibqp);
+ struct rvt_dev_info *rdi = ib_to_rvt(ibqp->device);
+ unsigned long flags = 0;
+ int call_send;
+ unsigned nreq = 0;
+ int err = 0;
+
+ spin_lock_irqsave(&qp->s_hlock, flags);
+
+ /*
+ * Ensure QP state is such that we can send. If not bail out early,
+ * there is no need to do this every time we post a send.
+ */
+ if (unlikely(!(ib_rvt_state_ops[qp->state] & RVT_POST_SEND_OK))) {
+ spin_unlock_irqrestore(&qp->s_hlock, flags);
+ return -EINVAL;
+ }
+
+ /*
+ * If the send queue is empty, and we only have a single WR then just go
+ * ahead and kick the send engine into gear. Otherwise we will always
+ * just schedule the send to happen later.
+ */
+ call_send = qp->s_head == ACCESS_ONCE(qp->s_last) && !wr->next;
+
+ for (; wr; wr = wr->next) {
+ err = rvt_post_one_wr(qp, wr, &call_send);
+ if (unlikely(err)) {
+ *bad_wr = wr;
+ goto bail;
+ }
+ nreq++;
+ }
+bail:
+ spin_unlock_irqrestore(&qp->s_hlock, flags);
+ if (nreq) {
+ if (call_send)
+ rdi->driver_f.schedule_send_no_lock(qp);
+ else
+ rdi->driver_f.do_send(qp);
+ }
+ return err;
+}
+
+/**
+ * rvt_post_srq_receive - post a receive on a shared receive queue
+ * @ibsrq: the SRQ to post the receive on
+ * @wr: the list of work requests to post
+ * @bad_wr: A pointer to the first WR to cause a problem is put here
+ *
+ * This may be called from interrupt context.
+ *
+ * Return: 0 on success else errno
+ */
+int rvt_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr,
+ struct ib_recv_wr **bad_wr)
+{
+ struct rvt_srq *srq = ibsrq_to_rvtsrq(ibsrq);
+ struct rvt_rwq *wq;
+ unsigned long flags;
+
+ for (; wr; wr = wr->next) {
+ struct rvt_rwqe *wqe;
+ u32 next;
+ int i;
+
+ if ((unsigned)wr->num_sge > srq->rq.max_sge) {
+ *bad_wr = wr;
+ return -EINVAL;
+ }
+
+ spin_lock_irqsave(&srq->rq.lock, flags);
+ wq = srq->rq.wq;
+ next = wq->head + 1;
+ if (next >= srq->rq.size)
+ next = 0;
+ if (next == wq->tail) {
+ spin_unlock_irqrestore(&srq->rq.lock, flags);
+ *bad_wr = wr;
+ return -ENOMEM;
+ }
+
+ wqe = rvt_get_rwqe_ptr(&srq->rq, wq->head);
+ wqe->wr_id = wr->wr_id;
+ wqe->num_sge = wr->num_sge;
+ for (i = 0; i < wr->num_sge; i++)
+ wqe->sg_list[i] = wr->sg_list[i];
+ /* Make sure queue entry is written before the head index. */
+ smp_wmb();
+ wq->head = next;
+ spin_unlock_irqrestore(&srq->rq.lock, flags);
+ }
+ return 0;
+}
diff --git a/drivers/infiniband/sw/rdmavt/qp.h b/drivers/infiniband/sw/rdmavt/qp.h
new file mode 100644
index 000000000000..8409f80d5f25
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/qp.h
@@ -0,0 +1,69 @@
+#ifndef DEF_RVTQP_H
+#define DEF_RVTQP_H
+
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <rdma/rdma_vt.h>
+
+int rvt_driver_qp_init(struct rvt_dev_info *rdi);
+void rvt_qp_exit(struct rvt_dev_info *rdi);
+struct ib_qp *rvt_create_qp(struct ib_pd *ibpd,
+ struct ib_qp_init_attr *init_attr,
+ struct ib_udata *udata);
+int rvt_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
+ int attr_mask, struct ib_udata *udata);
+int rvt_destroy_qp(struct ib_qp *ibqp);
+int rvt_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
+ int attr_mask, struct ib_qp_init_attr *init_attr);
+int rvt_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr,
+ struct ib_recv_wr **bad_wr);
+int rvt_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
+ struct ib_send_wr **bad_wr);
+int rvt_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr,
+ struct ib_recv_wr **bad_wr);
+#endif /* DEF_RVTQP_H */
diff --git a/drivers/infiniband/sw/rdmavt/srq.c b/drivers/infiniband/sw/rdmavt/srq.c
new file mode 100644
index 000000000000..f7c48e9023de
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/srq.c
@@ -0,0 +1,353 @@
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <linux/err.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+
+#include "srq.h"
+#include "vt.h"
+
+/**
+ * rvt_driver_srq_init - init srq resources on a per driver basis
+ * @rdi: rvt dev structure
+ *
+ * Do any initialization needed when a driver registers with rdmavt.
+ */
+void rvt_driver_srq_init(struct rvt_dev_info *rdi)
+{
+ spin_lock_init(&rdi->n_srqs_lock);
+ rdi->n_srqs_allocated = 0;
+}
+
+/**
+ * rvt_create_srq - create a shared receive queue
+ * @ibpd: the protection domain of the SRQ to create
+ * @srq_init_attr: the attributes of the SRQ
+ * @udata: data from libibverbs when creating a user SRQ
+ *
+ * Return: Allocated srq object
+ */
+struct ib_srq *rvt_create_srq(struct ib_pd *ibpd,
+ struct ib_srq_init_attr *srq_init_attr,
+ struct ib_udata *udata)
+{
+ struct rvt_dev_info *dev = ib_to_rvt(ibpd->device);
+ struct rvt_srq *srq;
+ u32 sz;
+ struct ib_srq *ret;
+
+ if (srq_init_attr->srq_type != IB_SRQT_BASIC)
+ return ERR_PTR(-ENOSYS);
+
+ if (srq_init_attr->attr.max_sge == 0 ||
+ srq_init_attr->attr.max_sge > dev->dparms.props.max_srq_sge ||
+ srq_init_attr->attr.max_wr == 0 ||
+ srq_init_attr->attr.max_wr > dev->dparms.props.max_srq_wr)
+ return ERR_PTR(-EINVAL);
+
+ srq = kmalloc(sizeof(*srq), GFP_KERNEL);
+ if (!srq)
+ return ERR_PTR(-ENOMEM);
+
+ /*
+ * Need to use vmalloc() if we want to support large #s of entries.
+ */
+ srq->rq.size = srq_init_attr->attr.max_wr + 1;
+ srq->rq.max_sge = srq_init_attr->attr.max_sge;
+ sz = sizeof(struct ib_sge) * srq->rq.max_sge +
+ sizeof(struct rvt_rwqe);
+ srq->rq.wq = vmalloc_user(sizeof(struct rvt_rwq) + srq->rq.size * sz);
+ if (!srq->rq.wq) {
+ ret = ERR_PTR(-ENOMEM);
+ goto bail_srq;
+ }
+
+ /*
+ * Return the address of the RWQ as the offset to mmap.
+ * See rvt_mmap() for details.
+ */
+ if (udata && udata->outlen >= sizeof(__u64)) {
+ int err;
+ u32 s = sizeof(struct rvt_rwq) + srq->rq.size * sz;
+
+ srq->ip =
+ rvt_create_mmap_info(dev, s, ibpd->uobject->context,
+ srq->rq.wq);
+ if (!srq->ip) {
+ ret = ERR_PTR(-ENOMEM);
+ goto bail_wq;
+ }
+
+ err = ib_copy_to_udata(udata, &srq->ip->offset,
+ sizeof(srq->ip->offset));
+ if (err) {
+ ret = ERR_PTR(err);
+ goto bail_ip;
+ }
+ } else {
+ srq->ip = NULL;
+ }
+
+ /*
+ * ib_create_srq() will initialize srq->ibsrq.
+ */
+ spin_lock_init(&srq->rq.lock);
+ srq->rq.wq->head = 0;
+ srq->rq.wq->tail = 0;
+ srq->limit = srq_init_attr->attr.srq_limit;
+
+ spin_lock(&dev->n_srqs_lock);
+ if (dev->n_srqs_allocated == dev->dparms.props.max_srq) {
+ spin_unlock(&dev->n_srqs_lock);
+ ret = ERR_PTR(-ENOMEM);
+ goto bail_ip;
+ }
+
+ dev->n_srqs_allocated++;
+ spin_unlock(&dev->n_srqs_lock);
+
+ if (srq->ip) {
+ spin_lock_irq(&dev->pending_lock);
+ list_add(&srq->ip->pending_mmaps, &dev->pending_mmaps);
+ spin_unlock_irq(&dev->pending_lock);
+ }
+
+ return &srq->ibsrq;
+
+bail_ip:
+ kfree(srq->ip);
+bail_wq:
+ vfree(srq->rq.wq);
+bail_srq:
+ kfree(srq);
+ return ret;
+}
+
+/**
+ * rvt_modify_srq - modify a shared receive queue
+ * @ibsrq: the SRQ to modify
+ * @attr: the new attributes of the SRQ
+ * @attr_mask: indicates which attributes to modify
+ * @udata: user data for libibverbs.so
+ *
+ * Return: 0 on success
+ */
+int rvt_modify_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr,
+ enum ib_srq_attr_mask attr_mask,
+ struct ib_udata *udata)
+{
+ struct rvt_srq *srq = ibsrq_to_rvtsrq(ibsrq);
+ struct rvt_dev_info *dev = ib_to_rvt(ibsrq->device);
+ struct rvt_rwq *wq;
+ int ret = 0;
+
+ if (attr_mask & IB_SRQ_MAX_WR) {
+ struct rvt_rwq *owq;
+ struct rvt_rwqe *p;
+ u32 sz, size, n, head, tail;
+
+ /* Check that the requested sizes are below the limits. */
+ if ((attr->max_wr > dev->dparms.props.max_srq_wr) ||
+ ((attr_mask & IB_SRQ_LIMIT) ?
+ attr->srq_limit : srq->limit) > attr->max_wr)
+ return -EINVAL;
+
+ sz = sizeof(struct rvt_rwqe) +
+ srq->rq.max_sge * sizeof(struct ib_sge);
+ size = attr->max_wr + 1;
+ wq = vmalloc_user(sizeof(struct rvt_rwq) + size * sz);
+ if (!wq)
+ return -ENOMEM;
+
+ /* Check that we can write the offset to mmap. */
+ if (udata && udata->inlen >= sizeof(__u64)) {
+ __u64 offset_addr;
+ __u64 offset = 0;
+
+ ret = ib_copy_from_udata(&offset_addr, udata,
+ sizeof(offset_addr));
+ if (ret)
+ goto bail_free;
+ udata->outbuf = (void __user *)
+ (unsigned long)offset_addr;
+ ret = ib_copy_to_udata(udata, &offset,
+ sizeof(offset));
+ if (ret)
+ goto bail_free;
+ }
+
+ spin_lock_irq(&srq->rq.lock);
+ /*
+ * validate head and tail pointer values and compute
+ * the number of remaining WQEs.
+ */
+ owq = srq->rq.wq;
+ head = owq->head;
+ tail = owq->tail;
+ if (head >= srq->rq.size || tail >= srq->rq.size) {
+ ret = -EINVAL;
+ goto bail_unlock;
+ }
+ n = head;
+ if (n < tail)
+ n += srq->rq.size - tail;
+ else
+ n -= tail;
+ if (size <= n) {
+ ret = -EINVAL;
+ goto bail_unlock;
+ }
+ n = 0;
+ p = wq->wq;
+ while (tail != head) {
+ struct rvt_rwqe *wqe;
+ int i;
+
+ wqe = rvt_get_rwqe_ptr(&srq->rq, tail);
+ p->wr_id = wqe->wr_id;
+ p->num_sge = wqe->num_sge;
+ for (i = 0; i < wqe->num_sge; i++)
+ p->sg_list[i] = wqe->sg_list[i];
+ n++;
+ p = (struct rvt_rwqe *)((char *)p + sz);
+ if (++tail >= srq->rq.size)
+ tail = 0;
+ }
+ srq->rq.wq = wq;
+ srq->rq.size = size;
+ wq->head = n;
+ wq->tail = 0;
+ if (attr_mask & IB_SRQ_LIMIT)
+ srq->limit = attr->srq_limit;
+ spin_unlock_irq(&srq->rq.lock);
+
+ vfree(owq);
+
+ if (srq->ip) {
+ struct rvt_mmap_info *ip = srq->ip;
+ struct rvt_dev_info *dev = ib_to_rvt(srq->ibsrq.device);
+ u32 s = sizeof(struct rvt_rwq) + size * sz;
+
+ rvt_update_mmap_info(dev, ip, s, wq);
+
+ /*
+ * Return the offset to mmap.
+ * See rvt_mmap() for details.
+ */
+ if (udata && udata->inlen >= sizeof(__u64)) {
+ ret = ib_copy_to_udata(udata, &ip->offset,
+ sizeof(ip->offset));
+ if (ret)
+ return ret;
+ }
+
+ /*
+ * Put user mapping info onto the pending list
+ * unless it already is on the list.
+ */
+ spin_lock_irq(&dev->pending_lock);
+ if (list_empty(&ip->pending_mmaps))
+ list_add(&ip->pending_mmaps,
+ &dev->pending_mmaps);
+ spin_unlock_irq(&dev->pending_lock);
+ }
+ } else if (attr_mask & IB_SRQ_LIMIT) {
+ spin_lock_irq(&srq->rq.lock);
+ if (attr->srq_limit >= srq->rq.size)
+ ret = -EINVAL;
+ else
+ srq->limit = attr->srq_limit;
+ spin_unlock_irq(&srq->rq.lock);
+ }
+ return ret;
+
+bail_unlock:
+ spin_unlock_irq(&srq->rq.lock);
+bail_free:
+ vfree(wq);
+ return ret;
+}
+
+/** rvt_query_srq - query srq data
+ * @ibsrq: srq to query
+ * @attr: return info in attr
+ *
+ * Return: always 0
+ */
+int rvt_query_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr)
+{
+ struct rvt_srq *srq = ibsrq_to_rvtsrq(ibsrq);
+
+ attr->max_wr = srq->rq.size - 1;
+ attr->max_sge = srq->rq.max_sge;
+ attr->srq_limit = srq->limit;
+ return 0;
+}
+
+/**
+ * rvt_destroy_srq - destory an srq
+ * @ibsrq: srq object to destroy
+ *
+ * Return always 0
+ */
+int rvt_destroy_srq(struct ib_srq *ibsrq)
+{
+ struct rvt_srq *srq = ibsrq_to_rvtsrq(ibsrq);
+ struct rvt_dev_info *dev = ib_to_rvt(ibsrq->device);
+
+ spin_lock(&dev->n_srqs_lock);
+ dev->n_srqs_allocated--;
+ spin_unlock(&dev->n_srqs_lock);
+ if (srq->ip)
+ kref_put(&srq->ip->ref, rvt_release_mmap_info);
+ else
+ vfree(srq->rq.wq);
+ kfree(srq);
+
+ return 0;
+}
diff --git a/drivers/infiniband/sw/rdmavt/srq.h b/drivers/infiniband/sw/rdmavt/srq.h
new file mode 100644
index 000000000000..bf0eaaf56465
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/srq.h
@@ -0,0 +1,62 @@
+#ifndef DEF_RVTSRQ_H
+#define DEF_RVTSRQ_H
+
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <rdma/rdma_vt.h>
+void rvt_driver_srq_init(struct rvt_dev_info *rdi);
+struct ib_srq *rvt_create_srq(struct ib_pd *ibpd,
+ struct ib_srq_init_attr *srq_init_attr,
+ struct ib_udata *udata);
+int rvt_modify_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr,
+ enum ib_srq_attr_mask attr_mask,
+ struct ib_udata *udata);
+int rvt_query_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr);
+int rvt_destroy_srq(struct ib_srq *ibsrq);
+
+#endif /* DEF_RVTSRQ_H */
diff --git a/drivers/infiniband/sw/rdmavt/trace.c b/drivers/infiniband/sw/rdmavt/trace.c
new file mode 100644
index 000000000000..d593285a349c
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/trace.c
@@ -0,0 +1,49 @@
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#define CREATE_TRACE_POINTS
+#include "trace.h"
diff --git a/drivers/infiniband/sw/rdmavt/trace.h b/drivers/infiniband/sw/rdmavt/trace.h
new file mode 100644
index 000000000000..6c0457db5499
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/trace.h
@@ -0,0 +1,187 @@
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#undef TRACE_SYSTEM_VAR
+#define TRACE_SYSTEM_VAR rdmavt
+
+#if !defined(__RDMAVT_TRACE_H) || defined(TRACE_HEADER_MULTI_READ)
+#define __RDMAVT_TRACE_H
+
+#include <linux/tracepoint.h>
+#include <linux/trace_seq.h>
+
+#include <rdma/ib_verbs.h>
+#include <rdma/rdma_vt.h>
+
+#define RDI_DEV_ENTRY(rdi) __string(dev, rdi->driver_f.get_card_name(rdi))
+#define RDI_DEV_ASSIGN(rdi) __assign_str(dev, rdi->driver_f.get_card_name(rdi))
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM rdmavt
+
+TRACE_EVENT(rvt_dbg,
+ TP_PROTO(struct rvt_dev_info *rdi,
+ const char *msg),
+ TP_ARGS(rdi, msg),
+ TP_STRUCT__entry(
+ RDI_DEV_ENTRY(rdi)
+ __string(msg, msg)
+ ),
+ TP_fast_assign(
+ RDI_DEV_ASSIGN(rdi);
+ __assign_str(msg, msg);
+ ),
+ TP_printk("[%s]: %s", __get_str(dev), __get_str(msg))
+);
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM rvt_qphash
+DECLARE_EVENT_CLASS(rvt_qphash_template,
+ TP_PROTO(struct rvt_qp *qp, u32 bucket),
+ TP_ARGS(qp, bucket),
+ TP_STRUCT__entry(
+ RDI_DEV_ENTRY(ib_to_rvt(qp->ibqp.device))
+ __field(u32, qpn)
+ __field(u32, bucket)
+ ),
+ TP_fast_assign(
+ RDI_DEV_ASSIGN(ib_to_rvt(qp->ibqp.device))
+ __entry->qpn = qp->ibqp.qp_num;
+ __entry->bucket = bucket;
+ ),
+ TP_printk(
+ "[%s] qpn 0x%x bucket %u",
+ __get_str(dev),
+ __entry->qpn,
+ __entry->bucket
+ )
+);
+
+DEFINE_EVENT(rvt_qphash_template, rvt_qpinsert,
+ TP_PROTO(struct rvt_qp *qp, u32 bucket),
+ TP_ARGS(qp, bucket));
+
+DEFINE_EVENT(rvt_qphash_template, rvt_qpremove,
+ TP_PROTO(struct rvt_qp *qp, u32 bucket),
+ TP_ARGS(qp, bucket));
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM rvt_tx
+
+#define wr_opcode_name(opcode) { IB_WR_##opcode, #opcode }
+#define show_wr_opcode(opcode) \
+__print_symbolic(opcode, \
+ wr_opcode_name(RDMA_WRITE), \
+ wr_opcode_name(RDMA_WRITE_WITH_IMM), \
+ wr_opcode_name(SEND), \
+ wr_opcode_name(SEND_WITH_IMM), \
+ wr_opcode_name(RDMA_READ), \
+ wr_opcode_name(ATOMIC_CMP_AND_SWP), \
+ wr_opcode_name(ATOMIC_FETCH_AND_ADD), \
+ wr_opcode_name(LSO), \
+ wr_opcode_name(SEND_WITH_INV), \
+ wr_opcode_name(RDMA_READ_WITH_INV), \
+ wr_opcode_name(LOCAL_INV), \
+ wr_opcode_name(MASKED_ATOMIC_CMP_AND_SWP), \
+ wr_opcode_name(MASKED_ATOMIC_FETCH_AND_ADD))
+
+#define POS_PRN \
+"[%s] wr_id %llx qpn %x psn 0x%x lpsn 0x%x length %u opcode 0x%.2x,%s size %u avail %u head %u last %u"
+
+TRACE_EVENT(
+ rvt_post_one_wr,
+ TP_PROTO(struct rvt_qp *qp, struct rvt_swqe *wqe),
+ TP_ARGS(qp, wqe),
+ TP_STRUCT__entry(
+ RDI_DEV_ENTRY(ib_to_rvt(qp->ibqp.device))
+ __field(u64, wr_id)
+ __field(u32, qpn)
+ __field(u32, psn)
+ __field(u32, lpsn)
+ __field(u32, length)
+ __field(u32, opcode)
+ __field(u32, size)
+ __field(u32, avail)
+ __field(u32, head)
+ __field(u32, last)
+ ),
+ TP_fast_assign(
+ RDI_DEV_ASSIGN(ib_to_rvt(qp->ibqp.device))
+ __entry->wr_id = wqe->wr.wr_id;
+ __entry->qpn = qp->ibqp.qp_num;
+ __entry->psn = wqe->psn;
+ __entry->lpsn = wqe->lpsn;
+ __entry->length = wqe->length;
+ __entry->opcode = wqe->wr.opcode;
+ __entry->size = qp->s_size;
+ __entry->avail = qp->s_avail;
+ __entry->head = qp->s_head;
+ __entry->last = qp->s_last;
+ ),
+ TP_printk(
+ POS_PRN,
+ __get_str(dev),
+ __entry->wr_id,
+ __entry->qpn,
+ __entry->psn,
+ __entry->lpsn,
+ __entry->length,
+ __entry->opcode, show_wr_opcode(__entry->opcode),
+ __entry->size,
+ __entry->avail,
+ __entry->head,
+ __entry->last
+ )
+);
+
+#endif /* __RDMAVT_TRACE_H */
+
+#undef TRACE_INCLUDE_PATH
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_PATH .
+#define TRACE_INCLUDE_FILE trace
+#include <trace/define_trace.h>
diff --git a/drivers/infiniband/sw/rdmavt/vt.c b/drivers/infiniband/sw/rdmavt/vt.c
new file mode 100644
index 000000000000..6caf5272ba1f
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/vt.c
@@ -0,0 +1,873 @@
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include "vt.h"
+#include "trace.h"
+
+#define RVT_UVERBS_ABI_VERSION 2
+
+MODULE_LICENSE("Dual BSD/GPL");
+MODULE_DESCRIPTION("RDMA Verbs Transport Library");
+
+static int rvt_init(void)
+{
+ /*
+ * rdmavt does not need to do anything special when it starts up. All it
+ * needs to do is sit and wait until a driver attempts registration.
+ */
+ return 0;
+}
+module_init(rvt_init);
+
+static void rvt_cleanup(void)
+{
+ /*
+ * Nothing to do at exit time either. The module won't be able to be
+ * removed until all drivers are gone which means all the dev structs
+ * are gone so there is really nothing to do.
+ */
+}
+module_exit(rvt_cleanup);
+
+/**
+ * rvt_alloc_device - allocate rdi
+ * @size: how big of a structure to allocate
+ * @nports: number of ports to allocate array slots for
+ *
+ * Use IB core device alloc to allocate space for the rdi which is assumed to be
+ * inside of the ib_device. Any extra space that drivers require should be
+ * included in size.
+ *
+ * We also allocate a port array based on the number of ports.
+ *
+ * Return: pointer to allocated rdi
+ */
+struct rvt_dev_info *rvt_alloc_device(size_t size, int nports)
+{
+ struct rvt_dev_info *rdi = ERR_PTR(-ENOMEM);
+
+ rdi = (struct rvt_dev_info *)ib_alloc_device(size);
+ if (!rdi)
+ return rdi;
+
+ rdi->ports = kcalloc(nports,
+ sizeof(struct rvt_ibport **),
+ GFP_KERNEL);
+ if (!rdi->ports)
+ ib_dealloc_device(&rdi->ibdev);
+
+ return rdi;
+}
+EXPORT_SYMBOL(rvt_alloc_device);
+
+static int rvt_query_device(struct ib_device *ibdev,
+ struct ib_device_attr *props,
+ struct ib_udata *uhw)
+{
+ struct rvt_dev_info *rdi = ib_to_rvt(ibdev);
+
+ if (uhw->inlen || uhw->outlen)
+ return -EINVAL;
+ /*
+ * Return rvt_dev_info.dparms.props contents
+ */
+ *props = rdi->dparms.props;
+ return 0;
+}
+
+static int rvt_modify_device(struct ib_device *device,
+ int device_modify_mask,
+ struct ib_device_modify *device_modify)
+{
+ /*
+ * There is currently no need to supply this based on qib and hfi1.
+ * Future drivers may need to implement this though.
+ */
+
+ return -EOPNOTSUPP;
+}
+
+/**
+ * rvt_query_port: Passes the query port call to the driver
+ * @ibdev: Verbs IB dev
+ * @port_num: port number, 1 based from ib core
+ * @props: structure to hold returned properties
+ *
+ * Return: 0 on success
+ */
+static int rvt_query_port(struct ib_device *ibdev, u8 port_num,
+ struct ib_port_attr *props)
+{
+ struct rvt_dev_info *rdi = ib_to_rvt(ibdev);
+ struct rvt_ibport *rvp;
+ int port_index = ibport_num_to_idx(ibdev, port_num);
+
+ if (port_index < 0)
+ return -EINVAL;
+
+ rvp = rdi->ports[port_index];
+ memset(props, 0, sizeof(*props));
+ props->sm_lid = rvp->sm_lid;
+ props->sm_sl = rvp->sm_sl;
+ props->port_cap_flags = rvp->port_cap_flags;
+ props->max_msg_sz = 0x80000000;
+ props->pkey_tbl_len = rvt_get_npkeys(rdi);
+ props->bad_pkey_cntr = rvp->pkey_violations;
+ props->qkey_viol_cntr = rvp->qkey_violations;
+ props->subnet_timeout = rvp->subnet_timeout;
+ props->init_type_reply = 0;
+
+ /* Populate the remaining ib_port_attr elements */
+ return rdi->driver_f.query_port_state(rdi, port_num, props);
+}
+
+/**
+ * rvt_modify_port
+ * @ibdev: Verbs IB dev
+ * @port_num: Port number, 1 based from ib core
+ * @port_modify_mask: How to change the port
+ * @props: Structure to fill in
+ *
+ * Return: 0 on success
+ */
+static int rvt_modify_port(struct ib_device *ibdev, u8 port_num,
+ int port_modify_mask, struct ib_port_modify *props)
+{
+ struct rvt_dev_info *rdi = ib_to_rvt(ibdev);
+ struct rvt_ibport *rvp;
+ int ret = 0;
+ int port_index = ibport_num_to_idx(ibdev, port_num);
+
+ if (port_index < 0)
+ return -EINVAL;
+
+ rvp = rdi->ports[port_index];
+ rvp->port_cap_flags |= props->set_port_cap_mask;
+ rvp->port_cap_flags &= ~props->clr_port_cap_mask;
+
+ if (props->set_port_cap_mask || props->clr_port_cap_mask)
+ rdi->driver_f.cap_mask_chg(rdi, port_num);
+ if (port_modify_mask & IB_PORT_SHUTDOWN)
+ ret = rdi->driver_f.shut_down_port(rdi, port_num);
+ if (port_modify_mask & IB_PORT_RESET_QKEY_CNTR)
+ rvp->qkey_violations = 0;
+
+ return ret;
+}
+
+/**
+ * rvt_query_pkey - Return a pkey from the table at a given index
+ * @ibdev: Verbs IB dev
+ * @port_num: Port number, 1 based from ib core
+ * @intex: Index into pkey table
+ *
+ * Return: 0 on failure pkey otherwise
+ */
+static int rvt_query_pkey(struct ib_device *ibdev, u8 port_num, u16 index,
+ u16 *pkey)
+{
+ /*
+ * Driver will be responsible for keeping rvt_dev_info.pkey_table up to
+ * date. This function will just return that value. There is no need to
+ * lock, if a stale value is read and sent to the user so be it there is
+ * no way to protect against that anyway.
+ */
+ struct rvt_dev_info *rdi = ib_to_rvt(ibdev);
+ int port_index;
+
+ port_index = ibport_num_to_idx(ibdev, port_num);
+ if (port_index < 0)
+ return -EINVAL;
+
+ if (index >= rvt_get_npkeys(rdi))
+ return -EINVAL;
+
+ *pkey = rvt_get_pkey(rdi, port_index, index);
+ return 0;
+}
+
+/**
+ * rvt_query_gid - Return a gid from the table
+ * @ibdev: Verbs IB dev
+ * @port_num: Port number, 1 based from ib core
+ * @index: = Index in table
+ * @gid: Gid to return
+ *
+ * Return: 0 on success
+ */
+static int rvt_query_gid(struct ib_device *ibdev, u8 port_num,
+ int guid_index, union ib_gid *gid)
+{
+ struct rvt_dev_info *rdi;
+ struct rvt_ibport *rvp;
+ int port_index;
+
+ /*
+ * Driver is responsible for updating the guid table. Which will be used
+ * to craft the return value. This will work similar to how query_pkey()
+ * is being done.
+ */
+ port_index = ibport_num_to_idx(ibdev, port_num);
+ if (port_index < 0)
+ return -EINVAL;
+
+ rdi = ib_to_rvt(ibdev);
+ rvp = rdi->ports[port_index];
+
+ gid->global.subnet_prefix = rvp->gid_prefix;
+
+ return rdi->driver_f.get_guid_be(rdi, rvp, guid_index,
+ &gid->global.interface_id);
+}
+
+struct rvt_ucontext {
+ struct ib_ucontext ibucontext;
+};
+
+static inline struct rvt_ucontext *to_iucontext(struct ib_ucontext
+ *ibucontext)
+{
+ return container_of(ibucontext, struct rvt_ucontext, ibucontext);
+}
+
+/**
+ * rvt_alloc_ucontext - Allocate a user context
+ * @ibdev: Vers IB dev
+ * @data: User data allocated
+ */
+static struct ib_ucontext *rvt_alloc_ucontext(struct ib_device *ibdev,
+ struct ib_udata *udata)
+{
+ struct rvt_ucontext *context;
+
+ context = kmalloc(sizeof(*context), GFP_KERNEL);
+ if (!context)
+ return ERR_PTR(-ENOMEM);
+ return &context->ibucontext;
+}
+
+/**
+ *rvt_dealloc_ucontext - Free a user context
+ *@context - Free this
+ */
+static int rvt_dealloc_ucontext(struct ib_ucontext *context)
+{
+ kfree(to_iucontext(context));
+ return 0;
+}
+
+static int rvt_get_port_immutable(struct ib_device *ibdev, u8 port_num,
+ struct ib_port_immutable *immutable)
+{
+ struct rvt_dev_info *rdi = ib_to_rvt(ibdev);
+ struct ib_port_attr attr;
+ int err, port_index;
+
+ port_index = ibport_num_to_idx(ibdev, port_num);
+ if (port_index < 0)
+ return -EINVAL;
+
+ err = rvt_query_port(ibdev, port_num, &attr);
+ if (err)
+ return err;
+
+ immutable->pkey_tbl_len = attr.pkey_tbl_len;
+ immutable->gid_tbl_len = attr.gid_tbl_len;
+ immutable->core_cap_flags = rdi->dparms.core_cap_flags;
+ immutable->max_mad_size = rdi->dparms.max_mad_size;
+
+ return 0;
+}
+
+enum {
+ MISC,
+ QUERY_DEVICE,
+ MODIFY_DEVICE,
+ QUERY_PORT,
+ MODIFY_PORT,
+ QUERY_PKEY,
+ QUERY_GID,
+ ALLOC_UCONTEXT,
+ DEALLOC_UCONTEXT,
+ GET_PORT_IMMUTABLE,
+ CREATE_QP,
+ MODIFY_QP,
+ DESTROY_QP,
+ QUERY_QP,
+ POST_SEND,
+ POST_RECV,
+ POST_SRQ_RECV,
+ CREATE_AH,
+ DESTROY_AH,
+ MODIFY_AH,
+ QUERY_AH,
+ CREATE_SRQ,
+ MODIFY_SRQ,
+ DESTROY_SRQ,
+ QUERY_SRQ,
+ ATTACH_MCAST,
+ DETACH_MCAST,
+ GET_DMA_MR,
+ REG_USER_MR,
+ DEREG_MR,
+ ALLOC_MR,
+ ALLOC_FMR,
+ MAP_PHYS_FMR,
+ UNMAP_FMR,
+ DEALLOC_FMR,
+ MMAP,
+ CREATE_CQ,
+ DESTROY_CQ,
+ POLL_CQ,
+ REQ_NOTFIY_CQ,
+ RESIZE_CQ,
+ ALLOC_PD,
+ DEALLOC_PD,
+ _VERB_IDX_MAX /* Must always be last! */
+};
+
+static inline int check_driver_override(struct rvt_dev_info *rdi,
+ size_t offset, void *func)
+{
+ if (!*(void **)((void *)&rdi->ibdev + offset)) {
+ *(void **)((void *)&rdi->ibdev + offset) = func;
+ return 0;
+ }
+
+ return 1;
+}
+
+static noinline int check_support(struct rvt_dev_info *rdi, int verb)
+{
+ switch (verb) {
+ case MISC:
+ /*
+ * These functions are not part of verbs specifically but are
+ * required for rdmavt to function.
+ */
+ if ((!rdi->driver_f.port_callback) ||
+ (!rdi->driver_f.get_card_name) ||
+ (!rdi->driver_f.get_pci_dev))
+ return -EINVAL;
+ break;
+
+ case QUERY_DEVICE:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ query_device),
+ rvt_query_device);
+ break;
+
+ case MODIFY_DEVICE:
+ /*
+ * rdmavt does not support modify device currently drivers must
+ * provide.
+ */
+ if (!check_driver_override(rdi, offsetof(struct ib_device,
+ modify_device),
+ rvt_modify_device))
+ return -EOPNOTSUPP;
+ break;
+
+ case QUERY_PORT:
+ if (!check_driver_override(rdi, offsetof(struct ib_device,
+ query_port),
+ rvt_query_port))
+ if (!rdi->driver_f.query_port_state)
+ return -EINVAL;
+ break;
+
+ case MODIFY_PORT:
+ if (!check_driver_override(rdi, offsetof(struct ib_device,
+ modify_port),
+ rvt_modify_port))
+ if (!rdi->driver_f.cap_mask_chg ||
+ !rdi->driver_f.shut_down_port)
+ return -EINVAL;
+ break;
+
+ case QUERY_PKEY:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ query_pkey),
+ rvt_query_pkey);
+ break;
+
+ case QUERY_GID:
+ if (!check_driver_override(rdi, offsetof(struct ib_device,
+ query_gid),
+ rvt_query_gid))
+ if (!rdi->driver_f.get_guid_be)
+ return -EINVAL;
+ break;
+
+ case ALLOC_UCONTEXT:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ alloc_ucontext),
+ rvt_alloc_ucontext);
+ break;
+
+ case DEALLOC_UCONTEXT:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ dealloc_ucontext),
+ rvt_dealloc_ucontext);
+ break;
+
+ case GET_PORT_IMMUTABLE:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ get_port_immutable),
+ rvt_get_port_immutable);
+ break;
+
+ case CREATE_QP:
+ if (!check_driver_override(rdi, offsetof(struct ib_device,
+ create_qp),
+ rvt_create_qp))
+ if (!rdi->driver_f.qp_priv_alloc ||
+ !rdi->driver_f.qp_priv_free ||
+ !rdi->driver_f.notify_qp_reset ||
+ !rdi->driver_f.flush_qp_waiters ||
+ !rdi->driver_f.stop_send_queue ||
+ !rdi->driver_f.quiesce_qp)
+ return -EINVAL;
+ break;
+
+ case MODIFY_QP:
+ if (!check_driver_override(rdi, offsetof(struct ib_device,
+ modify_qp),
+ rvt_modify_qp))
+ if (!rdi->driver_f.notify_qp_reset ||
+ !rdi->driver_f.schedule_send ||
+ !rdi->driver_f.get_pmtu_from_attr ||
+ !rdi->driver_f.flush_qp_waiters ||
+ !rdi->driver_f.stop_send_queue ||
+ !rdi->driver_f.quiesce_qp ||
+ !rdi->driver_f.notify_error_qp ||
+ !rdi->driver_f.mtu_from_qp ||
+ !rdi->driver_f.mtu_to_path_mtu ||
+ !rdi->driver_f.shut_down_port ||
+ !rdi->driver_f.cap_mask_chg)
+ return -EINVAL;
+ break;
+
+ case DESTROY_QP:
+ if (!check_driver_override(rdi, offsetof(struct ib_device,
+ destroy_qp),
+ rvt_destroy_qp))
+ if (!rdi->driver_f.qp_priv_free ||
+ !rdi->driver_f.notify_qp_reset ||
+ !rdi->driver_f.flush_qp_waiters ||
+ !rdi->driver_f.stop_send_queue ||
+ !rdi->driver_f.quiesce_qp)
+ return -EINVAL;
+ break;
+
+ case QUERY_QP:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ query_qp),
+ rvt_query_qp);
+ break;
+
+ case POST_SEND:
+ if (!check_driver_override(rdi, offsetof(struct ib_device,
+ post_send),
+ rvt_post_send))
+ if (!rdi->driver_f.schedule_send ||
+ !rdi->driver_f.do_send)
+ return -EINVAL;
+ break;
+
+ case POST_RECV:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ post_recv),
+ rvt_post_recv);
+ break;
+ case POST_SRQ_RECV:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ post_srq_recv),
+ rvt_post_srq_recv);
+ break;
+
+ case CREATE_AH:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ create_ah),
+ rvt_create_ah);
+ break;
+
+ case DESTROY_AH:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ destroy_ah),
+ rvt_destroy_ah);
+ break;
+
+ case MODIFY_AH:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ modify_ah),
+ rvt_modify_ah);
+ break;
+
+ case QUERY_AH:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ query_ah),
+ rvt_query_ah);
+ break;
+
+ case CREATE_SRQ:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ create_srq),
+ rvt_create_srq);
+ break;
+
+ case MODIFY_SRQ:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ modify_srq),
+ rvt_modify_srq);
+ break;
+
+ case DESTROY_SRQ:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ destroy_srq),
+ rvt_destroy_srq);
+ break;
+
+ case QUERY_SRQ:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ query_srq),
+ rvt_query_srq);
+ break;
+
+ case ATTACH_MCAST:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ attach_mcast),
+ rvt_attach_mcast);
+ break;
+
+ case DETACH_MCAST:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ detach_mcast),
+ rvt_detach_mcast);
+ break;
+
+ case GET_DMA_MR:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ get_dma_mr),
+ rvt_get_dma_mr);
+ break;
+
+ case REG_USER_MR:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ reg_user_mr),
+ rvt_reg_user_mr);
+ break;
+
+ case DEREG_MR:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ dereg_mr),
+ rvt_dereg_mr);
+ break;
+
+ case ALLOC_FMR:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ alloc_fmr),
+ rvt_alloc_fmr);
+ break;
+
+ case ALLOC_MR:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ alloc_mr),
+ rvt_alloc_mr);
+ break;
+
+ case MAP_PHYS_FMR:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ map_phys_fmr),
+ rvt_map_phys_fmr);
+ break;
+
+ case UNMAP_FMR:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ unmap_fmr),
+ rvt_unmap_fmr);
+ break;
+
+ case DEALLOC_FMR:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ dealloc_fmr),
+ rvt_dealloc_fmr);
+ break;
+
+ case MMAP:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ mmap),
+ rvt_mmap);
+ break;
+
+ case CREATE_CQ:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ create_cq),
+ rvt_create_cq);
+ break;
+
+ case DESTROY_CQ:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ destroy_cq),
+ rvt_destroy_cq);
+ break;
+
+ case POLL_CQ:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ poll_cq),
+ rvt_poll_cq);
+ break;
+
+ case REQ_NOTFIY_CQ:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ req_notify_cq),
+ rvt_req_notify_cq);
+ break;
+
+ case RESIZE_CQ:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ resize_cq),
+ rvt_resize_cq);
+ break;
+
+ case ALLOC_PD:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ alloc_pd),
+ rvt_alloc_pd);
+ break;
+
+ case DEALLOC_PD:
+ check_driver_override(rdi, offsetof(struct ib_device,
+ dealloc_pd),
+ rvt_dealloc_pd);
+ break;
+
+ default:
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+/**
+ * rvt_register_device - register a driver
+ * @rdi: main dev structure for all of rdmavt operations
+ *
+ * It is up to drivers to allocate the rdi and fill in the appropriate
+ * information.
+ *
+ * Return: 0 on success otherwise an errno.
+ */
+int rvt_register_device(struct rvt_dev_info *rdi)
+{
+ int ret = 0, i;
+
+ if (!rdi)
+ return -EINVAL;
+
+ /*
+ * Check to ensure drivers have setup the required helpers for the verbs
+ * they want rdmavt to handle
+ */
+ for (i = 0; i < _VERB_IDX_MAX; i++)
+ if (check_support(rdi, i)) {
+ pr_err("Driver support req not met at %d\n", i);
+ return -EINVAL;
+ }
+
+
+ /* Once we get past here we can use rvt_pr macros and tracepoints */
+ trace_rvt_dbg(rdi, "Driver attempting registration");
+ rvt_mmap_init(rdi);
+
+ /* Queue Pairs */
+ ret = rvt_driver_qp_init(rdi);
+ if (ret) {
+ pr_err("Error in driver QP init.\n");
+ return -EINVAL;
+ }
+
+ /* Address Handle */
+ spin_lock_init(&rdi->n_ahs_lock);
+ rdi->n_ahs_allocated = 0;
+
+ /* Shared Receive Queue */
+ rvt_driver_srq_init(rdi);
+
+ /* Multicast */
+ rvt_driver_mcast_init(rdi);
+
+ /* Mem Region */
+ ret = rvt_driver_mr_init(rdi);
+ if (ret) {
+ pr_err("Error in driver MR init.\n");
+ goto bail_no_mr;
+ }
+
+ /* Completion queues */
+ ret = rvt_driver_cq_init(rdi);
+ if (ret) {
+ pr_err("Error in driver CQ init.\n");
+ goto bail_mr;
+ }
+
+ /* DMA Operations */
+ rdi->ibdev.dma_ops =
+ rdi->ibdev.dma_ops ? : &rvt_default_dma_mapping_ops;
+
+ /* Protection Domain */
+ spin_lock_init(&rdi->n_pds_lock);
+ rdi->n_pds_allocated = 0;
+
+ /*
+ * There are some things which could be set by underlying drivers but
+ * really should be up to rdmavt to set. For instance drivers can't know
+ * exactly which functions rdmavt supports, nor do they know the ABI
+ * version, so we do all of this sort of stuff here.
+ */
+ rdi->ibdev.uverbs_abi_ver = RVT_UVERBS_ABI_VERSION;
+ rdi->ibdev.uverbs_cmd_mask =
+ (1ull << IB_USER_VERBS_CMD_GET_CONTEXT) |
+ (1ull << IB_USER_VERBS_CMD_QUERY_DEVICE) |
+ (1ull << IB_USER_VERBS_CMD_QUERY_PORT) |
+ (1ull << IB_USER_VERBS_CMD_ALLOC_PD) |
+ (1ull << IB_USER_VERBS_CMD_DEALLOC_PD) |
+ (1ull << IB_USER_VERBS_CMD_CREATE_AH) |
+ (1ull << IB_USER_VERBS_CMD_MODIFY_AH) |
+ (1ull << IB_USER_VERBS_CMD_QUERY_AH) |
+ (1ull << IB_USER_VERBS_CMD_DESTROY_AH) |
+ (1ull << IB_USER_VERBS_CMD_REG_MR) |
+ (1ull << IB_USER_VERBS_CMD_DEREG_MR) |
+ (1ull << IB_USER_VERBS_CMD_CREATE_COMP_CHANNEL) |
+ (1ull << IB_USER_VERBS_CMD_CREATE_CQ) |
+ (1ull << IB_USER_VERBS_CMD_RESIZE_CQ) |
+ (1ull << IB_USER_VERBS_CMD_DESTROY_CQ) |
+ (1ull << IB_USER_VERBS_CMD_POLL_CQ) |
+ (1ull << IB_USER_VERBS_CMD_REQ_NOTIFY_CQ) |
+ (1ull << IB_USER_VERBS_CMD_CREATE_QP) |
+ (1ull << IB_USER_VERBS_CMD_QUERY_QP) |
+ (1ull << IB_USER_VERBS_CMD_MODIFY_QP) |
+ (1ull << IB_USER_VERBS_CMD_DESTROY_QP) |
+ (1ull << IB_USER_VERBS_CMD_POST_SEND) |
+ (1ull << IB_USER_VERBS_CMD_POST_RECV) |
+ (1ull << IB_USER_VERBS_CMD_ATTACH_MCAST) |
+ (1ull << IB_USER_VERBS_CMD_DETACH_MCAST) |
+ (1ull << IB_USER_VERBS_CMD_CREATE_SRQ) |
+ (1ull << IB_USER_VERBS_CMD_MODIFY_SRQ) |
+ (1ull << IB_USER_VERBS_CMD_QUERY_SRQ) |
+ (1ull << IB_USER_VERBS_CMD_DESTROY_SRQ) |
+ (1ull << IB_USER_VERBS_CMD_POST_SRQ_RECV);
+ rdi->ibdev.node_type = RDMA_NODE_IB_CA;
+ rdi->ibdev.num_comp_vectors = 1;
+
+ /* We are now good to announce we exist */
+ ret = ib_register_device(&rdi->ibdev, rdi->driver_f.port_callback);
+ if (ret) {
+ rvt_pr_err(rdi, "Failed to register driver with ib core.\n");
+ goto bail_cq;
+ }
+
+ rvt_create_mad_agents(rdi);
+
+ rvt_pr_info(rdi, "Registration with rdmavt done.\n");
+ return ret;
+
+bail_cq:
+ rvt_cq_exit(rdi);
+
+bail_mr:
+ rvt_mr_exit(rdi);
+
+bail_no_mr:
+ rvt_qp_exit(rdi);
+
+ return ret;
+}
+EXPORT_SYMBOL(rvt_register_device);
+
+/**
+ * rvt_unregister_device - remove a driver
+ * @rdi: rvt dev struct
+ */
+void rvt_unregister_device(struct rvt_dev_info *rdi)
+{
+ trace_rvt_dbg(rdi, "Driver is unregistering.");
+ if (!rdi)
+ return;
+
+ rvt_free_mad_agents(rdi);
+
+ ib_unregister_device(&rdi->ibdev);
+ rvt_cq_exit(rdi);
+ rvt_mr_exit(rdi);
+ rvt_qp_exit(rdi);
+}
+EXPORT_SYMBOL(rvt_unregister_device);
+
+/**
+ * rvt_init_port - init internal data for driver port
+ * @rdi: rvt dev strut
+ * @port: rvt port
+ * @port_index: 0 based index of ports, different from IB core port num
+ *
+ * Keep track of a list of ports. No need to have a detach port.
+ * They persist until the driver goes away.
+ *
+ * Return: always 0
+ */
+int rvt_init_port(struct rvt_dev_info *rdi, struct rvt_ibport *port,
+ int port_index, u16 *pkey_table)
+{
+
+ rdi->ports[port_index] = port;
+ rdi->ports[port_index]->pkey_table = pkey_table;
+
+ return 0;
+}
+EXPORT_SYMBOL(rvt_init_port);
diff --git a/drivers/infiniband/sw/rdmavt/vt.h b/drivers/infiniband/sw/rdmavt/vt.h
new file mode 100644
index 000000000000..6b01eaa4461b
--- /dev/null
+++ b/drivers/infiniband/sw/rdmavt/vt.h
@@ -0,0 +1,104 @@
+#ifndef DEF_RDMAVT_H
+#define DEF_RDMAVT_H
+
+/*
+ * Copyright(c) 2016 Intel Corporation.
+ *
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * GPL LICENSE SUMMARY
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * BSD LICENSE
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * - Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * - Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * - Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#include <rdma/rdma_vt.h>
+#include <linux/pci.h>
+#include "dma.h"
+#include "pd.h"
+#include "qp.h"
+#include "ah.h"
+#include "mr.h"
+#include "srq.h"
+#include "mcast.h"
+#include "mmap.h"
+#include "cq.h"
+#include "mad.h"
+#include "mmap.h"
+
+#define rvt_pr_info(rdi, fmt, ...) \
+ __rvt_pr_info(rdi->driver_f.get_pci_dev(rdi), \
+ rdi->driver_f.get_card_name(rdi), \
+ fmt, \
+ ##__VA_ARGS__)
+
+#define rvt_pr_warn(rdi, fmt, ...) \
+ __rvt_pr_warn(rdi->driver_f.get_pci_dev(rdi), \
+ rdi->driver_f.get_card_name(rdi), \
+ fmt, \
+ ##__VA_ARGS__)
+
+#define rvt_pr_err(rdi, fmt, ...) \
+ __rvt_pr_err(rdi->driver_f.get_pci_dev(rdi), \
+ rdi->driver_f.get_card_name(rdi), \
+ fmt, \
+ ##__VA_ARGS__)
+
+#define __rvt_pr_info(pdev, name, fmt, ...) \
+ dev_info(&pdev->dev, "%s: " fmt, name, ##__VA_ARGS__)
+
+#define __rvt_pr_warn(pdev, name, fmt, ...) \
+ dev_warn(&pdev->dev, "%s: " fmt, name, ##__VA_ARGS__)
+
+#define __rvt_pr_err(pdev, name, fmt, ...) \
+ dev_err(&pdev->dev, "%s: " fmt, name, ##__VA_ARGS__)
+
+static inline int ibport_num_to_idx(struct ib_device *ibdev, u8 port_num)
+{
+ struct rvt_dev_info *rdi = ib_to_rvt(ibdev);
+ int port_index;
+
+ port_index = port_num - 1; /* IB ports start at 1 our arrays at 0 */
+ if ((port_index < 0) || (port_index >= rdi->dparms.nports))
+ return -EINVAL;
+
+ return port_index;
+}
+
+#endif /* DEF_RDMAVT_H */