ib_srp.c

linux-2.6.15.6

/*
 * Copyright (c) 2005 Cisco Systems.  All rights reserved.
 *
 * This software is available to you under a choice of one of two
 * licenses.  You may choose to be licensed under the terms of the GNU
 * General Public License (GPL) Version 2, available from the file
 * COPYING in the main directory of this source tree, or the
 * OpenIB.org BSD license below:
 *
 *     Redistribution and use in source and binary forms, with or
 *     without modification, are permitted provided that the following
 *     conditions are met:
 *
 *      - Redistributions of source code must retain the above
 *        copyright notice, this list of conditions and the following
 *        disclaimer.
 *
 *      - Redistributions in binary form must reproduce the above
 *        copyright notice, this list of conditions and the following
 *        disclaimer in the documentation and/or other materials
 *        provided with the distribution.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
 * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
 * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
 * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
 * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 * SOFTWARE.
 *
 * $Id: ib_srp.c 3932 2005-11-01 17:19:29Z roland $
 */

#include <linux/module.h>
#include <linux/init.h>
#include <linux/slab.h>
#include <linux/err.h>
#include <linux/string.h>
#include <linux/parser.h>
#include <linux/random.h>

#include <asm/atomic.h>

#include <scsi/scsi.h>
#include <scsi/scsi_device.h>
#include <scsi/scsi_dbg.h>
#include <scsi/srp.h>

#include <rdma/ib_cache.h>

#include "ib_srp.h"

#define DRV_NAME	"ib_srp"
#define PFX		DRV_NAME ": "
#define DRV_VERSION	"0.2"
#define DRV_RELDATE	"November 1, 2005"

MODULE_AUTHOR("Roland Dreier");
MODULE_DESCRIPTION("InfiniBand SCSI RDMA Protocol initiator "
		   "v" DRV_VERSION " (" DRV_RELDATE ")");
MODULE_LICENSE("Dual BSD/GPL");

static int topspin_workarounds = 1;

module_param(topspin_workarounds, int, 0444);
MODULE_PARM_DESC(topspin_workarounds,
		 "Enable workarounds for Topspin/Cisco SRP target bugs if != 0");

static const u8 topspin_oui[3] = { 0x00, 0x05, 0xad };

static void srp_add_one(struct ib_device *device);
static void srp_remove_one(struct ib_device *device);
static void srp_completion(struct ib_cq *cq, void *target_ptr);
static int srp_cm_handler(struct ib_cm_id *cm_id, struct ib_cm_event *event);

static struct ib_client srp_client = {
	.name   = "srp",
	.add    = srp_add_one,
	.remove = srp_remove_one
};

static inline struct srp_target_port *host_to_target(struct Scsi_Host *host)
{
	return (struct srp_target_port *) host->hostdata;
}

static const char *srp_target_info(struct Scsi_Host *host)
{
	return host_to_target(host)->target_name;
}

static struct srp_iu *srp_alloc_iu(struct srp_host *host, size_t size,
				   gfp_t gfp_mask,
				   enum dma_data_direction direction)
{
	struct srp_iu *iu;

	iu = kmalloc(sizeof *iu, gfp_mask);
	if (!iu)
		goto out;

	iu->buf = kzalloc(size, gfp_mask);
	if (!iu->buf)
		goto out_free_iu;

	iu->dma = dma_map_single(host->dev->dma_device, iu->buf, size, direction);
	if (dma_mapping_error(iu->dma))
		goto out_free_buf;

	iu->size      = size;
	iu->direction = direction;

	return iu;

out_free_buf:
	kfree(iu->buf);
out_free_iu:
	kfree(iu);
out:
	return NULL;
}

static void srp_free_iu(struct srp_host *host, struct srp_iu *iu)
{
	if (!iu)
		return;

	dma_unmap_single(host->dev->dma_device, iu->dma, iu->size, iu->direction);
	kfree(iu->buf);
	kfree(iu);
}

static void srp_qp_event(struct ib_event *event, void *context)
{
	printk(KERN_ERR PFX "QP event %d\n", event->event);
}

static int srp_init_qp(struct srp_target_port *target,
		       struct ib_qp *qp)
{
	struct ib_qp_attr *attr;
	int ret;

	attr = kmalloc(sizeof *attr, GFP_KERNEL);
	if (!attr)
		return -ENOMEM;

	ret = ib_find_cached_pkey(target->srp_host->dev,
				  target->srp_host->port,
				  be16_to_cpu(target->path.pkey),
				  &attr->pkey_index);
	if (ret)
		goto out;

	attr->qp_state        = IB_QPS_INIT;
	attr->qp_access_flags = (IB_ACCESS_REMOTE_READ |
				    IB_ACCESS_REMOTE_WRITE);
	attr->port_num        = target->srp_host->port;

	ret = ib_modify_qp(qp, attr,
			   IB_QP_STATE		|
			   IB_QP_PKEY_INDEX	|
			   IB_QP_ACCESS_FLAGS	|
			   IB_QP_PORT);

out:
	kfree(attr);
	return ret;
}

static int srp_create_target_ib(struct srp_target_port *target)
{
	struct ib_qp_init_attr *init_attr;
	int ret;

	init_attr = kzalloc(sizeof *init_attr, GFP_KERNEL);
	if (!init_attr)
		return -ENOMEM;

	target->cq = ib_create_cq(target->srp_host->dev, srp_completion,
				  NULL, target, SRP_CQ_SIZE);
	if (IS_ERR(target->cq)) {
		ret = PTR_ERR(target->cq);
		goto out;
	}

	ib_req_notify_cq(target->cq, IB_CQ_NEXT_COMP);

	init_attr->event_handler       = srp_qp_event;
	init_attr->cap.max_send_wr     = SRP_SQ_SIZE;
	init_attr->cap.max_recv_wr     = SRP_RQ_SIZE;
	init_attr->cap.max_recv_sge    = 1;
	init_attr->cap.max_send_sge    = 1;
	init_attr->sq_sig_type         = IB_SIGNAL_ALL_WR;
	init_attr->qp_type             = IB_QPT_RC;
	init_attr->send_cq             = target->cq;
	init_attr->recv_cq             = target->cq;

	target->qp = ib_create_qp(target->srp_host->pd, init_attr);
	if (IS_ERR(target->qp)) {
		ret = PTR_ERR(target->qp);
		ib_destroy_cq(target->cq);
		goto out;
	}

	ret = srp_init_qp(target, target->qp);
	if (ret) {
		ib_destroy_qp(target->qp);
		ib_destroy_cq(target->cq);
		goto out;
	}

out:
	kfree(init_attr);
	return ret;
}

static void srp_free_target_ib(struct srp_target_port *target)
{
	int i;

	ib_destroy_qp(target->qp);
	ib_destroy_cq(target->cq);

	for (i = 0; i < SRP_RQ_SIZE; ++i)
		srp_free_iu(target->srp_host, target->rx_ring[i]);
	for (i = 0; i < SRP_SQ_SIZE + 1; ++i)
		srp_free_iu(target->srp_host, target->tx_ring[i]);
}

static void srp_path_rec_completion(int status,
				    struct ib_sa_path_rec *pathrec,
				    void *target_ptr)
{
	struct srp_target_port *target = target_ptr;

	target->status = status;
	if (status)
		printk(KERN_ERR PFX "Got failed path rec status %d\n", status);
	else
		target->path = *pathrec;
	complete(&target->done);
}

static int srp_lookup_path(struct srp_target_port *target)
{
	target->path.numb_path = 1;

	init_completion(&target->done);

	target->path_query_id = ib_sa_path_rec_get(target->srp_host->dev,
						   target->srp_host->port,
						   &target->path,
						   IB_SA_PATH_REC_DGID		|
						   IB_SA_PATH_REC_SGID		|
						   IB_SA_PATH_REC_NUMB_PATH	|
						   IB_SA_PATH_REC_PKEY,
						   SRP_PATH_REC_TIMEOUT_MS,
						   GFP_KERNEL,
						   srp_path_rec_completion,
						   target, &target->path_query);
	if (target->path_query_id < 0)
		return target->path_query_id;

	wait_for_completion(&target->done);

	if (target->status < 0)
		printk(KERN_WARNING PFX "Path record query failed\n");

	return target->status;
}

static int srp_send_req(struct srp_target_port *target)
{
	struct {
		struct ib_cm_req_param param;
		struct srp_login_req   priv;
	} *req = NULL;
	int status;

	req = kzalloc(sizeof *req, GFP_KERNEL);
	if (!req)
		return -ENOMEM;

	req->param.primary_path 	      = &target->path;
	req->param.alternate_path 	      = NULL;
	req->param.service_id 		      = target->service_id;
	req->param.qp_num 		      = target->qp->qp_num;
	req->param.qp_type 		      = target->qp->qp_type;
	req->param.private_data 	      = &req->priv;
	req->param.private_data_len 	      = sizeof req->priv;
	req->param.flow_control 	      = 1;

	get_random_bytes(&req->param.starting_psn, 4);
	req->param.starting_psn 	     &= 0xffffff;

	/*
	 * Pick some arbitrary defaults here; we could make these
	 * module parameters if anyone cared about setting them.
	 */
	req->param.responder_resources	      = 4;
	req->param.remote_cm_response_timeout = 20;
	req->param.local_cm_response_timeout  = 20;
	req->param.retry_count 		      = 7;
	req->param.rnr_retry_count 	      = 7;
	req->param.max_cm_retries 	      = 15;

	req->priv.opcode     	= SRP_LOGIN_REQ;
	req->priv.tag        	= 0;
	req->priv.req_it_iu_len = cpu_to_be32(SRP_MAX_IU_LEN);
	req->priv.req_buf_fmt 	= cpu_to_be16(SRP_BUF_FORMAT_DIRECT |
					      SRP_BUF_FORMAT_INDIRECT);
	memcpy(req->priv.initiator_port_id, target->srp_host->initiator_port_id, 16);
	/*
	 * Topspin/Cisco SRP targets will reject our login unless we
	 * zero out the first 8 bytes of our initiator port ID.  The
	 * second 8 bytes must be our local node GUID, but we always
	 * use that anyway.
	 */
	if (topspin_workarounds && !memcmp(&target->ioc_guid, topspin_oui, 3)) {
		printk(KERN_DEBUG PFX "Topspin/Cisco initiator port ID workaround "
		       "activated for target GUID %016llx\n",
		       (unsigned long long) be64_to_cpu(target->ioc_guid));
		memset(req->priv.initiator_port_id, 0, 8);
	}
	memcpy(req->priv.target_port_id,     &target->id_ext, 8);
	memcpy(req->priv.target_port_id + 8, &target->ioc_guid, 8);

	status = ib_send_cm_req(target->cm_id, &req->param);

	kfree(req);

	return status;
}

static void srp_disconnect_target(struct srp_target_port *target)
{
	/* XXX should send SRP_I_LOGOUT request */

	init_completion(&target->done);
	ib_send_cm_dreq(target->cm_id, NULL, 0);
	wait_for_completion(&target->done);
}

static void srp_remove_work(void *target_ptr)
{
	struct srp_target_port *target = target_ptr;

	spin_lock_irq(target->scsi_host->host_lock);
	if (target->state != SRP_TARGET_DEAD) {
		spin_unlock_irq(target->scsi_host->host_lock);
		scsi_host_put(target->scsi_host);
		return;
	}
	target->state = SRP_TARGET_REMOVED;
	spin_unlock_irq(target->scsi_host->host_lock);

	down(&target->srp_host->target_mutex);
	list_del(&target->list);
	up(&target->srp_host->target_mutex);

	scsi_remove_host(target->scsi_host);
	ib_destroy_cm_id(target->cm_id);
	srp_free_target_ib(target);
	scsi_host_put(target->scsi_host);
	/* And another put to really free the target port... */
	scsi_host_put(target->scsi_host);
}

static int srp_connect_target(struct srp_target_port *target)
{
	int ret;

	ret = srp_lookup_path(target);
	if (ret)
		return ret;

	while (1) {
		init_completion(&target->done);
		ret = srp_send_req(target);
		if (ret)
			return ret;
		wait_for_completion(&target->done);

		/*
		 * The CM event handling code will set status to
		 * SRP_PORT_REDIRECT if we get a port redirect REJ
		 * back, or SRP_DLID_REDIRECT if we get a lid/qp
		 * redirect REJ back.
		 */
		switch (target->status) {
		case 0:
			return 0;

		case SRP_PORT_REDIRECT:
			ret = srp_lookup_path(target);
			if (ret)
				return ret;
			break;

		case SRP_DLID_REDIRECT:
			break;

		default:
			return target->status;
		}
	}
}

static int srp_reconnect_target(struct srp_target_port *target)
{
	struct ib_cm_id *new_cm_id;
	struct ib_qp_attr qp_attr;
	struct srp_request *req;
	struct ib_wc wc;
	int ret;
	int i;

	spin_lock_irq(target->scsi_host->host_lock);
	if (target->state != SRP_TARGET_LIVE) {
		spin_unlock_irq(target->scsi_host->host_lock);
		return -EAGAIN;
	}
	target->state = SRP_TARGET_CONNECTING;
	spin_unlock_irq(target->scsi_host->host_lock);

	srp_disconnect_target(target);
	/*
	 * Now get a new local CM ID so that we avoid confusing the
	 * target in case things are really fouled up.
	 */
	new_cm_id = ib_create_cm_id(target->srp_host->dev,
				    srp_cm_handler, target);
	if (IS_ERR(new_cm_id)) {
		ret = PTR_ERR(new_cm_id);
		goto err;
	}
	ib_destroy_cm_id(target->cm_id);
	target->cm_id = new_cm_id;

	qp_attr.qp_state = IB_QPS_RESET;
	ret = ib_modify_qp(target->qp, &qp_attr, IB_QP_STATE);
	if (ret)
		goto err;

	ret = srp_init_qp(target, target->qp);
	if (ret)
		goto err;

	while (ib_poll_cq(target->cq, 1, &wc) > 0)
		; /* nothing */

	list_for_each_entry(req, &target->req_queue, list) {
		req->scmnd->result = DID_RESET << 16;
		req->scmnd->scsi_done(req->scmnd);
	}

	target->rx_head	 = 0;
	target->tx_head	 = 0;
	target->tx_tail  = 0;
	target->req_head = 0;
	for (i = 0; i < SRP_SQ_SIZE - 1; ++i)
		target->req_ring[i].next = i + 1;
	target->req_ring[SRP_SQ_SIZE - 1].next = -1;
	INIT_LIST_HEAD(&target->req_queue);

	ret = srp_connect_target(target);
	if (ret)
		goto err;

	spin_lock_irq(target->scsi_host->host_lock);
	if (target->state == SRP_TARGET_CONNECTING) {
		ret = 0;
		target->state = SRP_TARGET_LIVE;
	} else
		ret = -EAGAIN;
	spin_unlock_irq(target->scsi_host->host_lock);

	return ret;

err:
	printk(KERN_ERR PFX "reconnect failed (%d), removing target port.\n", ret);

	/*
	 * We couldn't reconnect, so kill our target port off.
	 * However, we have to defer the real removal because we might
	 * be in the context of the SCSI error handler now, which
	 * would deadlock if we call scsi_remove_host().
	 */
	spin_lock_irq(target->scsi_host->host_lock);
	if (target->state == SRP_TARGET_CONNECTING) {
		target->state = SRP_TARGET_DEAD;
		INIT_WORK(&target->work, srp_remove_work, target);
		schedule_work(&target->work);
	}
	spin_unlock_irq(target->scsi_host->host_lock);

	return ret;
}

static int srp_map_data(struct scsi_cmnd *scmnd, struct srp_target_port *target,
			struct srp_request *req)
{
	struct srp_cmd *cmd = req->cmd->buf;
	int len;
	u8 fmt;

	if (!scmnd->request_buffer || scmnd->sc_data_direction == DMA_NONE)
		return sizeof (struct srp_cmd);

	if (scmnd->sc_data_direction != DMA_FROM_DEVICE &&
	    scmnd->sc_data_direction != DMA_TO_DEVICE) {
		printk(KERN_WARNING PFX "Unhandled data direction %d\n",
		       scmnd->sc_data_direction);
		return -EINVAL;
	}

	if (scmnd->use_sg) {
		struct scatterlist *scat = scmnd->request_buffer;
		int n;
		int i;

		n = dma_map_sg(target->srp_host->dev->dma_device,
			       scat, scmnd->use_sg, scmnd->sc_data_direction);

		if (n == 1) {
			struct srp_direct_buf *buf = (void *) cmd->add_data;

			fmt = SRP_DATA_DESC_DIRECT;

			buf->va  = cpu_to_be64(sg_dma_address(scat));
			buf->key = cpu_to_be32(target->srp_host->mr->rkey);
			buf->len = cpu_to_be32(sg_dma_len(scat));

			len = sizeof (struct srp_cmd) +
				sizeof (struct srp_direct_buf);
		} else {
			struct srp_indirect_buf *buf = (void *) cmd->add_data;
			u32 datalen = 0;

			fmt = SRP_DATA_DESC_INDIRECT;

			if (scmnd->sc_data_direction == DMA_TO_DEVICE)
				cmd->data_out_desc_cnt = n;
			else
				cmd->data_in_desc_cnt = n;

			buf->table_desc.va  = cpu_to_be64(req->cmd->dma +
							  sizeof *cmd +
							  sizeof *buf);
			buf->table_desc.key =
				cpu_to_be32(target->srp_host->mr->rkey);
			buf->table_desc.len =
				cpu_to_be32(n * sizeof (struct srp_direct_buf));

			for (i = 0; i < n; ++i) {
				buf->desc_list[i].va  = cpu_to_be64(sg_dma_address(&scat[i]));
				buf->desc_list[i].key =
					cpu_to_be32(target->srp_host->mr->rkey);
				buf->desc_list[i].len = cpu_to_be32(sg_dma_len(&scat[i]));

				datalen += sg_dma_len(&scat[i]);
			}

			buf->len = cpu_to_be32(datalen);

			len = sizeof (struct srp_cmd) +
				sizeof (struct srp_indirect_buf) +