diff --git a/drivers/scsi/cxgbi/cxgb3i/Kbuild b/drivers/scsi/cxgbi/cxgb3i/Kbuild
new file mode 100644
index 0000000..09dbf9e
--- /dev/null
+++ b/drivers/scsi/cxgbi/cxgb3i/Kbuild
@@ -0,0 +1,3 @@
+EXTRA_CFLAGS += -I$(srctree)/drivers/net/cxgb3
+
+obj-$(CONFIG_SCSI_CXGB3_ISCSI) += cxgb3i.o
diff --git a/drivers/scsi/cxgbi/cxgb3i/Kconfig b/drivers/scsi/cxgbi/cxgb3i/Kconfig
new file mode 100644
index 0000000..5cf4e98
--- /dev/null
+++ b/drivers/scsi/cxgbi/cxgb3i/Kconfig
@@ -0,0 +1,7 @@
+config SCSI_CXGB3_ISCSI
+	tristate "Chelsio T3 iSCSI support"
+	depends on CHELSIO_T3_DEPENDS
+	select CHELSIO_T3
+	select SCSI_ISCSI_ATTRS
+	---help---
+	  This driver supports iSCSI offload for the Chelsio T3 devices.
diff --git a/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c
new file mode 100644
index 0000000..a01c1e2
--- /dev/null
+++ b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c
@@ -0,0 +1,1432 @@
+/*
+ * cxgb3i_offload.c: Chelsio S3xx iscsi offloaded tcp connection management
+ *
+ * Copyright (C) 2003-2008 Chelsio Communications.  All rights reserved.
+ *
+ * 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 LICENSE file included in this
+ * release for licensing terms and conditions.
+ *
+ * Written by:	Dimitris Michailidis (dm@chelsio.com)
+ *		Karen Xie (kxie@chelsio.com)
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ":%s: " fmt, __func__
+
+#include <linux/version.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <scsi/scsi_host.h>
+
+#include "common.h"
+#include "t3_cpl.h"
+#include "t3cdev.h"
+#include "cxgb3_defs.h"
+#include "cxgb3_ctl_defs.h"
+#include "cxgb3_offload.h"
+#include "firmware_exports.h"
+#include "cxgb3i.h"
+
+static unsigned int dbg_level;
+#include "../libcxgbi.h"
+
+#define DRV_MODULE_NAME         "cxgb3i"
+#define DRV_MODULE_DESC         "Chelsio T3 iSCSI Driver"
+#define DRV_MODULE_VERSION	"2.0.0"
+#define DRV_MODULE_RELDATE	"Jun. 2010"
+
+static char version[] =
+	DRV_MODULE_DESC " " DRV_MODULE_NAME
+	" v" DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n";
+
+MODULE_AUTHOR("Chelsio Communications, Inc.");
+MODULE_DESCRIPTION(DRV_MODULE_DESC);
+MODULE_VERSION(DRV_MODULE_VERSION);
+MODULE_LICENSE("GPL");
+
+module_param(dbg_level, uint, 0644);
+MODULE_PARM_DESC(dbg_level, "debug flag (default=0)");
+
+static int cxgb3i_rcv_win = 256 * 1024;
+module_param(cxgb3i_rcv_win, int, 0644);
+MODULE_PARM_DESC(cxgb3i_rcv_win, "TCP receive window in bytes (default=256KB)");
+
+static int cxgb3i_snd_win = 128 * 1024;
+module_param(cxgb3i_snd_win, int, 0644);
+MODULE_PARM_DESC(cxgb3i_snd_win, "TCP send window in bytes (default=128KB)");
+
+static int cxgb3i_rx_credit_thres = 10 * 1024;
+module_param(cxgb3i_rx_credit_thres, int, 0644);
+MODULE_PARM_DESC(rx_credit_thres,
+		 "RX credits return threshold in bytes (default=10KB)");
+
+static unsigned int cxgb3i_max_connect = 8 * 1024;
+module_param(cxgb3i_max_connect, uint, 0644);
+MODULE_PARM_DESC(cxgb3i_max_connect, "Max. # of connections (default=8092)");
+
+static unsigned int cxgb3i_sport_base = 20000;
+module_param(cxgb3i_sport_base, uint, 0644);
+MODULE_PARM_DESC(cxgb3i_sport_base, "starting port number (default=20000)");
+
+static void cxgb3i_dev_open(struct t3cdev *);
+static void cxgb3i_dev_close(struct t3cdev *);
+static void cxgb3i_dev_event_handler(struct t3cdev *, u32, u32);
+
+static struct cxgb3_client t3_client = {
+	.name = DRV_MODULE_NAME,
+	.handlers = cxgb3i_cpl_handlers,
+	.add = cxgb3i_dev_open,
+	.remove = cxgb3i_dev_close,
+	.event_handler = cxgb3i_dev_event_handler,
+};
+
+static struct scsi_host_template cxgb3i_host_template = {
+	.module		= THIS_MODULE,
+	.name		= DRV_MODULE_NAME,
+	.proc_name	= DRV_MODULE_NAME,
+	.can_queue	= CXGB3I_SCSI_HOST_QDEPTH,
+	.queuecommand	= iscsi_queuecommand,
+	.change_queue_depth = iscsi_change_queue_depth,
+	.sg_tablesize	= SG_ALL,
+	.max_sectors	= 0xFFFF,
+	.cmd_per_lun	= ISCSI_DEF_CMD_PER_LUN,
+	.eh_abort_handler = iscsi_eh_abort,
+	.eh_device_reset_handler = iscsi_eh_device_reset,
+	.eh_target_reset_handler = iscsi_eh_recover_target,
+	.target_alloc	= iscsi_target_alloc,
+	.use_clustering	= DISABLE_CLUSTERING,
+	.this_id	= -1,
+};
+
+static struct iscsi_transport cxgb3i_iscsi_transport = {
+	.owner		= THIS_MODULE,
+	.name		= DRV_MODULE_NAME,
+	/* owner and name should be set already */
+	.caps		= CAP_RECOVERY_L0 | CAP_MULTI_R2T | CAP_HDRDGST
+				| CAP_DATADGST | CAP_DIGEST_OFFLOAD |
+				CAP_PADDING_OFFLOAD,
+	.param_mask	= ISCSI_MAX_RECV_DLENGTH | ISCSI_MAX_XMIT_DLENGTH |
+				ISCSI_HDRDGST_EN | ISCSI_DATADGST_EN |
+				ISCSI_INITIAL_R2T_EN | ISCSI_MAX_R2T |
+				ISCSI_IMM_DATA_EN | ISCSI_FIRST_BURST |
+				ISCSI_MAX_BURST | ISCSI_PDU_INORDER_EN |
+				ISCSI_DATASEQ_INORDER_EN | ISCSI_ERL |
+				ISCSI_CONN_PORT | ISCSI_CONN_ADDRESS |
+				ISCSI_EXP_STATSN | ISCSI_PERSISTENT_PORT |
+				ISCSI_PERSISTENT_ADDRESS |
+				ISCSI_TARGET_NAME | ISCSI_TPGT |
+				ISCSI_USERNAME | ISCSI_PASSWORD |
+				ISCSI_USERNAME_IN | ISCSI_PASSWORD_IN |
+				ISCSI_FAST_ABORT | ISCSI_ABORT_TMO |
+				ISCSI_LU_RESET_TMO | ISCSI_TGT_RESET_TMO |
+				ISCSI_PING_TMO | ISCSI_RECV_TMO |
+				ISCSI_IFACE_NAME | ISCSI_INITIATOR_NAME,
+	.host_param_mask	= ISCSI_HOST_HWADDRESS | ISCSI_HOST_IPADDRESS |
+				ISCSI_HOST_INITIATOR_NAME |
+				ISCSI_HOST_NETDEV_NAME,
+	.get_host_param	= cxgbi_get_host_param,
+	.set_host_param	= cxgbi_set_host_param,
+	/* session management */
+	.create_session	= cxgbi_create_session,
+	.destroy_session	= cxgbi_destroy_session,
+	.get_session_param = iscsi_session_get_param,
+	/* connection management */
+	.create_conn	= cxgbi_create_conn,
+	.bind_conn	= cxgbi_bind_conn,
+	.destroy_conn	= iscsi_tcp_conn_teardown,
+	.start_conn	= iscsi_conn_start,
+	.stop_conn	= iscsi_conn_stop,
+	.get_conn_param	= cxgbi_get_conn_param,
+	.set_param	= cxgbi_set_conn_param,
+	.get_stats	= cxgbi_get_conn_stats,
+	/* pdu xmit req from user space */
+	.send_pdu	= iscsi_conn_send_pdu,
+	/* task */
+	.init_task	= iscsi_tcp_task_init,
+	.xmit_task	= iscsi_tcp_task_xmit,
+	.cleanup_task	= cxgbi_cleanup_task,
+	/* pdu */
+	.alloc_pdu	= cxgbi_conn_alloc_pdu,
+	.init_pdu	= cxgbi_conn_init_pdu,
+	.xmit_pdu	= cxgbi_conn_xmit_pdu,
+	.parse_pdu_itt	= cxgbi_parse_pdu_itt,
+	/* TCP connect/disconnect */
+	.ep_connect	= cxgbi_ep_connect,
+	.ep_poll	= cxgbi_ep_poll,
+	.ep_disconnect	= cxgbi_ep_disconnect,
+	/* Error recovery timeout call */
+	.session_recovery_timedout = iscsi_session_recovery_timedout,
+};
+
+static struct scsi_transport_template *cxgb3i_stt;
+
+/*
+ * CPL (Chelsio Protocol Language) defines a message passing interface between
+ * the host driver and Chelsio asic.
+ * The section below implments CPLs that related to iscsi tcp connection
+ * open/close/abort and data send/receive.
+ */
+
+static int push_tx_frames(struct cxgbi_sock *csk, int req_completion);
+
+static void send_act_open_req(struct cxgbi_sock *csk, struct sk_buff *skb,
+			      const struct l2t_entry *e)
+{
+	unsigned int wscale = cxgbi_sock_compute_wscale(cxgb3i_rcv_win);
+	struct cpl_act_open_req *req = (struct cpl_act_open_req *)skb->head;
+
+	skb->priority = CPL_PRIORITY_SETUP;
+
+	req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
+	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_ACT_OPEN_REQ, csk->atid));
+	req->local_port = csk->saddr.sin_port;
+	req->peer_port = csk->daddr.sin_port;
+	req->local_ip = csk->saddr.sin_addr.s_addr;
+	req->peer_ip = csk->daddr.sin_addr.s_addr;
+
+	req->opt0h = htonl(V_KEEP_ALIVE(1) | F_TCAM_BYPASS |
+			V_WND_SCALE(wscale) | V_MSS_IDX(csk->mss_idx) |
+			V_L2T_IDX(e->idx) | V_TX_CHANNEL(e->smt_idx));
+	req->opt0l = htonl(V_ULP_MODE(ULP2_MODE_ISCSI) |
+			V_RCV_BUFSIZ(cxgb3i_rcv_win>>10));
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_SOCK,
+		"csk 0x%p,%u,0x%lx,%u, %pI4:%u-%pI4:%u, %u,%u,%u.\n",
+		csk, csk->state, csk->flags, csk->atid,
+		&req->local_ip, ntohs(req->local_port),
+		&req->peer_ip, ntohs(req->peer_port),
+		csk->mss_idx, e->idx, e->smt_idx);
+
+	l2t_send(csk->cdev->lldev, skb, csk->l2t);
+}
+
+static inline void act_open_arp_failure(struct t3cdev *dev, struct sk_buff *skb)
+{
+	cxgbi_sock_act_open_req_arp_failure(NULL, skb);
+}
+
+/*
+ * CPL connection close request: host ->
+ *
+ * Close a connection by sending a CPL_CLOSE_CON_REQ message and queue it to
+ * the write queue (i.e., after any unsent txt data).
+ */
+static void send_close_req(struct cxgbi_sock *csk)
+{
+	struct sk_buff *skb = csk->cpl_close;
+	struct cpl_close_con_req *req = (struct cpl_close_con_req *)skb->head;
+	unsigned int tid = csk->tid;
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_SOCK,
+		"csk 0x%p,%u,0x%lx,%u.\n",
+		csk, csk->state, csk->flags, csk->tid);
+
+	csk->cpl_close = NULL;
+	req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_CLOSE_CON));
+	req->wr.wr_lo = htonl(V_WR_TID(tid));
+	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_CLOSE_CON_REQ, tid));
+	req->rsvd = htonl(csk->write_seq);
+
+	cxgbi_sock_skb_entail(csk, skb);
+	if (csk->state >= CTP_ESTABLISHED)
+		push_tx_frames(csk, 1);
+}
+
+/*
+ * CPL connection abort request: host ->
+ *
+ * Send an ABORT_REQ message. Makes sure we do not send multiple ABORT_REQs
+ * for the same connection and also that we do not try to send a message
+ * after the connection has closed.
+ */
+static void abort_arp_failure(struct t3cdev *tdev, struct sk_buff *skb)
+{
+	struct cpl_abort_req *req = cplhdr(skb);
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_SOCK,
+		"t3dev 0x%p, tid %u, skb 0x%p.\n",
+		tdev, GET_TID(req), skb);
+	req->cmd = CPL_ABORT_NO_RST;
+	cxgb3_ofld_send(tdev, skb);
+}
+
+static void send_abort_req(struct cxgbi_sock *csk)
+{
+	struct sk_buff *skb = csk->cpl_abort_req;
+	struct cpl_abort_req *req;
+
+	if (unlikely(csk->state == CTP_ABORTING || !skb))
+		return;
+	cxgbi_sock_set_state(csk, CTP_ABORTING);
+	cxgbi_sock_set_flag(csk, CTPF_ABORT_RPL_PENDING);
+	/* Purge the send queue so we don't send anything after an abort. */
+	cxgbi_sock_purge_write_queue(csk);
+
+	csk->cpl_abort_req = NULL;
+	req = (struct cpl_abort_req *)skb->head;
+	skb->priority = CPL_PRIORITY_DATA;
+	set_arp_failure_handler(skb, abort_arp_failure);
+	req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_HOST_ABORT_CON_REQ));
+	req->wr.wr_lo = htonl(V_WR_TID(csk->tid));
+	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_ABORT_REQ, csk->tid));
+	req->rsvd0 = htonl(csk->snd_nxt);
+	req->rsvd1 = !cxgbi_sock_flag(csk, CTPF_TX_DATA_SENT);
+	req->cmd = CPL_ABORT_SEND_RST;
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_SOCK,
+		"csk 0x%p,%u,0x%lx,%u, snd_nxt %u, 0x%x.\n",
+		csk, csk->state, csk->flags, csk->tid, csk->snd_nxt,
+		req->rsvd1);
+
+	l2t_send(csk->cdev->lldev, skb, csk->l2t);
+}
+
+/*
+ * CPL connection abort reply: host ->
+ *
+ * Send an ABORT_RPL message in response of the ABORT_REQ received.
+ */
+static void send_abort_rpl(struct cxgbi_sock *csk, int rst_status)
+{
+	struct sk_buff *skb = csk->cpl_abort_rpl;
+	struct cpl_abort_rpl *rpl = (struct cpl_abort_rpl *)skb->head;
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_SOCK,
+		"csk 0x%p,%u,0x%lx,%u, status %d.\n",
+		csk, csk->state, csk->flags, csk->tid, rst_status);
+
+	csk->cpl_abort_rpl = NULL;
+	skb->priority = CPL_PRIORITY_DATA;
+	rpl->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_HOST_ABORT_CON_RPL));
+	rpl->wr.wr_lo = htonl(V_WR_TID(csk->tid));
+	OPCODE_TID(rpl) = htonl(MK_OPCODE_TID(CPL_ABORT_RPL, csk->tid));
+	rpl->cmd = rst_status;
+	cxgb3_ofld_send(csk->cdev->lldev, skb);
+}
+
+/*
+ * CPL connection rx data ack: host ->
+ * Send RX credits through an RX_DATA_ACK CPL message. Returns the number of
+ * credits sent.
+ */
+static u32 send_rx_credits(struct cxgbi_sock *csk, u32 credits)
+{
+	struct sk_buff *skb;
+	struct cpl_rx_data_ack *req;
+	u32 dack = F_RX_DACK_CHANGE | V_RX_DACK_MODE(1);
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_PDU_RX,
+		"csk 0x%p,%u,0x%lx,%u, credit %u, dack %u.\n",
+		csk, csk->state, csk->flags, csk->tid, credits, dack);
+
+	skb = alloc_cpl(sizeof(*req), 0, GFP_ATOMIC);
+	if (!skb) {
+		pr_info("csk 0x%p, credit %u, OOM.\n", csk, credits);
+		return 0;
+	}
+	req = (struct cpl_rx_data_ack *)skb->head;
+	req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
+	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_RX_DATA_ACK, csk->tid));
+	req->credit_dack = htonl(F_RX_DACK_CHANGE | V_RX_DACK_MODE(1) |
+				V_RX_CREDITS(credits));
+	skb->priority = CPL_PRIORITY_ACK;
+	cxgb3_ofld_send(csk->cdev->lldev, skb);
+	return credits;
+}
+
+/*
+ * CPL connection tx data: host ->
+ *
+ * Send iscsi PDU via TX_DATA CPL message. Returns the number of
+ * credits sent.
+ * Each TX_DATA consumes work request credit (wrs), so we need to keep track of
+ * how many we've used so far and how many are pending (i.e., yet ack'ed by T3).
+ */
+
+static unsigned int wrlen __read_mostly;
+static unsigned int skb_wrs[SKB_WR_LIST_SIZE] __read_mostly;
+
+static void init_wr_tab(unsigned int wr_len)
+{
+	int i;
+
+	if (skb_wrs[1])		/* already initialized */
+		return;
+	for (i = 1; i < SKB_WR_LIST_SIZE; i++) {
+		int sgl_len = (3 * i) / 2 + (i & 1);
+
+		sgl_len += 3;
+		skb_wrs[i] = (sgl_len <= wr_len
+			      ? 1 : 1 + (sgl_len - 2) / (wr_len - 1));
+	}
+	wrlen = wr_len * 8;
+}
+
+static inline void make_tx_data_wr(struct cxgbi_sock *csk, struct sk_buff *skb,
+				   int len, int req_completion)
+{
+	struct tx_data_wr *req;
+	struct l2t_entry *l2t = csk->l2t;
+
+	skb_reset_transport_header(skb);
+	req = (struct tx_data_wr *)__skb_push(skb, sizeof(*req));
+	req->wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_TX_DATA) |
+			(req_completion ? F_WR_COMPL : 0));
+	req->wr_lo = htonl(V_WR_TID(csk->tid));
+	/* len includes the length of any HW ULP additions */
+	req->len = htonl(len);
+	/* V_TX_ULP_SUBMODE sets both the mode and submode */
+	req->flags = htonl(V_TX_ULP_SUBMODE(cxgbi_skcb_ulp_mode(skb)) |
+			   V_TX_SHOVE((skb_peek(&csk->write_queue) ? 0 : 1)));
+	req->sndseq = htonl(csk->snd_nxt);
+	req->param = htonl(V_TX_PORT(l2t->smt_idx));
+
+	if (!cxgbi_sock_flag(csk, CTPF_TX_DATA_SENT)) {
+		req->flags |= htonl(V_TX_ACK_PAGES(2) | F_TX_INIT |
+				    V_TX_CPU_IDX(csk->rss_qid));
+		/* sendbuffer is in units of 32KB. */
+		req->param |= htonl(V_TX_SNDBUF(cxgb3i_snd_win >> 15));
+		cxgbi_sock_set_flag(csk, CTPF_TX_DATA_SENT);
+	}
+}
+
+/**
+ * push_tx_frames -- start transmit
+ * @c3cn: the offloaded connection
+ * @req_completion: request wr_ack or not
+ *
+ * Prepends TX_DATA_WR or CPL_CLOSE_CON_REQ headers to buffers waiting in a
+ * connection's send queue and sends them on to T3.  Must be called with the
+ * connection's lock held.  Returns the amount of send buffer space that was
+ * freed as a result of sending queued data to T3.
+ */
+
+static void arp_failure_skb_discard(struct t3cdev *dev, struct sk_buff *skb)
+{
+	kfree_skb(skb);
+}
+
+static int push_tx_frames(struct cxgbi_sock *csk, int req_completion)
+{
+	int total_size = 0;
+	struct sk_buff *skb;
+
+	if (unlikely(csk->state < CTP_ESTABLISHED ||
+		csk->state == CTP_CLOSE_WAIT_1 || csk->state >= CTP_ABORTING)) {
+			log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_PDU_TX,
+				"csk 0x%p,%u,0x%lx,%u, in closing state.\n",
+				csk, csk->state, csk->flags, csk->tid);
+		return 0;
+	}
+
+	while (csk->wr_cred && (skb = skb_peek(&csk->write_queue)) != NULL) {
+		int len = skb->len;	/* length before skb_push */
+		int frags = skb_shinfo(skb)->nr_frags + (len != skb->data_len);
+		int wrs_needed = skb_wrs[frags];
+
+		if (wrs_needed > 1 && len + sizeof(struct tx_data_wr) <= wrlen)
+			wrs_needed = 1;
+
+		WARN_ON(frags >= SKB_WR_LIST_SIZE || wrs_needed < 1);
+
+		if (csk->wr_cred < wrs_needed) {
+			log_debug(1 << CXGBI_DBG_PDU_TX,
+				"csk 0x%p, skb len %u/%u, frag %u, wr %d<%u.\n",
+				csk, skb->len, skb->data_len, frags,
+				wrs_needed, csk->wr_cred);
+			break;
+		}
+
+		__skb_unlink(skb, &csk->write_queue);
+		skb->priority = CPL_PRIORITY_DATA;
+		skb->csum = wrs_needed;	/* remember this until the WR_ACK */
+		csk->wr_cred -= wrs_needed;
+		csk->wr_una_cred += wrs_needed;
+		cxgbi_sock_enqueue_wr(csk, skb);
+
+		log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_PDU_TX,
+			"csk 0x%p, enqueue, skb len %u/%u, frag %u, wr %d, "
+			"left %u, unack %u.\n",
+			csk, skb->len, skb->data_len, frags, skb->csum,
+			csk->wr_cred, csk->wr_una_cred);
+
+		if (likely(cxgbi_skcb_test_flag(skb, SKCBF_TX_NEED_HDR))) {
+			if ((req_completion &&
+				csk->wr_una_cred == wrs_needed) ||
+			     csk->wr_una_cred >= csk->wr_max_cred / 2) {
+				req_completion = 1;
+				csk->wr_una_cred = 0;
+			}
+			len += cxgbi_ulp_extra_len(cxgbi_skcb_ulp_mode(skb));
+			make_tx_data_wr(csk, skb, len, req_completion);
+			csk->snd_nxt += len;
+			cxgbi_skcb_clear_flag(skb, SKCBF_TX_NEED_HDR);
+		}
+		total_size += skb->truesize;
+		log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_PDU_TX,
+			"csk 0x%p, tid 0x%x, send skb 0x%p.\n",
+			csk, csk->tid, skb);
+		set_arp_failure_handler(skb, arp_failure_skb_discard);
+		l2t_send(csk->cdev->lldev, skb, csk->l2t);
+	}
+	return total_size;
+}
+
+/*
+ * Process a CPL_ACT_ESTABLISH message: -> host
+ * Updates connection state from an active establish CPL message.  Runs with
+ * the connection lock held.
+ */
+
+static inline void free_atid(struct cxgbi_sock *csk)
+{
+	if (cxgbi_sock_flag(csk, CTPF_HAS_ATID)) {
+		cxgb3_free_atid(csk->cdev->lldev, csk->atid);
+		cxgbi_sock_clear_flag(csk, CTPF_HAS_ATID);
+		cxgbi_sock_put(csk);
+	}
+}
+
+static int do_act_establish(struct t3cdev *tdev, struct sk_buff *skb, void *ctx)
+{
+	struct cxgbi_sock *csk = ctx;
+	struct cpl_act_establish *req = cplhdr(skb);
+	unsigned int tid = GET_TID(req);
+	unsigned int atid = G_PASS_OPEN_TID(ntohl(req->tos_tid));
+	u32 rcv_isn = ntohl(req->rcv_isn);	/* real RCV_ISN + 1 */
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_SOCK,
+		"atid 0x%x,tid 0x%x, csk 0x%p,%u,0x%lx, isn %u.\n",
+		atid, atid, csk, csk->state, csk->flags, rcv_isn);
+
+	cxgbi_sock_get(csk);
+	cxgbi_sock_set_flag(csk, CTPF_HAS_TID);
+	csk->tid = tid;
+	cxgb3_insert_tid(csk->cdev->lldev, &t3_client, csk, tid);
+
+	free_atid(csk);
+
+	csk->rss_qid = G_QNUM(ntohs(skb->csum));
+
+	spin_lock_bh(&csk->lock);
+	if (csk->retry_timer.function) {
+		del_timer(&csk->retry_timer);
+		csk->retry_timer.function = NULL;
+	}
+
+	if (unlikely(csk->state != CTP_ACTIVE_OPEN))
+		pr_info("csk 0x%p,%u,0x%lx,%u, got EST.\n",
+			csk, csk->state, csk->flags, csk->tid);
+
+	csk->copied_seq = csk->rcv_wup = csk->rcv_nxt = rcv_isn;
+	if (cxgb3i_rcv_win > (M_RCV_BUFSIZ << 10))
+		csk->rcv_wup -= cxgb3i_rcv_win - (M_RCV_BUFSIZ << 10);
+
+	cxgbi_sock_established(csk, ntohl(req->snd_isn), ntohs(req->tcp_opt));
+
+	if (unlikely(cxgbi_sock_flag(csk, CTPF_ACTIVE_CLOSE_NEEDED)))
+		/* upper layer has requested closing */
+		send_abort_req(csk);
+	else {
+		if (skb_queue_len(&csk->write_queue))
+			push_tx_frames(csk, 1);
+		cxgbi_conn_tx_open(csk);
+	}
+
+	spin_unlock_bh(&csk->lock);
+	__kfree_skb(skb);
+	return 0;
+}
+
+/*
+ * Process a CPL_ACT_OPEN_RPL message: -> host
+ * Handle active open failures.
+ */
+static int act_open_rpl_status_to_errno(int status)
+{
+	switch (status) {
+	case CPL_ERR_CONN_RESET:
+		return -ECONNREFUSED;
+	case CPL_ERR_ARP_MISS:
+		return -EHOSTUNREACH;
+	case CPL_ERR_CONN_TIMEDOUT:
+		return -ETIMEDOUT;
+	case CPL_ERR_TCAM_FULL:
+		return -ENOMEM;
+	case CPL_ERR_CONN_EXIST:
+		return -EADDRINUSE;
+	default:
+		return -EIO;
+	}
+}
+
+static void act_open_retry_timer(unsigned long data)
+{
+	struct sk_buff *skb;
+	struct cxgbi_sock *csk = (struct cxgbi_sock *)data;
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_SOCK,
+		"csk 0x%p,%u,0x%lx,%u.\n",
+		csk, csk->state, csk->flags, csk->tid);
+
+	cxgbi_sock_get(csk);
+	spin_lock_bh(&csk->lock);
+	skb = alloc_cpl(sizeof(struct cpl_act_open_req), 0, GFP_ATOMIC);
+	if (!skb)
+		cxgbi_sock_fail_act_open(csk, -ENOMEM);
+	else {
+		skb->sk = (struct sock *)csk;
+		set_arp_failure_handler(skb, act_open_arp_failure);
+		send_act_open_req(csk, skb, csk->l2t);
+	}
+	spin_unlock_bh(&csk->lock);
+	cxgbi_sock_put(csk);
+}
+
+static int do_act_open_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx)
+{
+	struct cxgbi_sock *csk = ctx;
+	struct cpl_act_open_rpl *rpl = cplhdr(skb);
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_SOCK,
+		"csk 0x%p,%u,0x%lx,%u, status %u.\n",
+		csk, csk->state, csk->flags, csk->atid, rpl->status);
+
+	if (rpl->status != CPL_ERR_TCAM_FULL &&
+	    rpl->status != CPL_ERR_CONN_EXIST &&
+	    rpl->status != CPL_ERR_ARP_MISS)
+		cxgb3_queue_tid_release(tdev, GET_TID(rpl));
+
+	cxgbi_sock_get(csk);
+	spin_lock_bh(&csk->lock);
+	if (rpl->status == CPL_ERR_CONN_EXIST &&
+	    csk->retry_timer.function != act_open_retry_timer) {
+		csk->retry_timer.function = act_open_retry_timer;
+		mod_timer(&csk->retry_timer, jiffies + HZ / 2);
+	} else
+		cxgbi_sock_fail_act_open(csk,
+				act_open_rpl_status_to_errno(rpl->status));
+
+	spin_unlock_bh(&csk->lock);
+	cxgbi_sock_put(csk);
+	__kfree_skb(skb);
+	return 0;
+}
+
+/*
+ * Process PEER_CLOSE CPL messages: -> host
+ * Handle peer FIN.
+ */
+static int do_peer_close(struct t3cdev *cdev, struct sk_buff *skb, void *ctx)
+{
+	struct cxgbi_sock *csk = ctx;
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_SOCK,
+		"csk 0x%p,%u,0x%lx,%u.\n",
+		csk, csk->state, csk->flags, csk->tid);
+
+	cxgbi_sock_rcv_peer_close(csk);
+	__kfree_skb(skb);
+	return 0;
+}
+
+/*
+ * Process CLOSE_CONN_RPL CPL message: -> host
+ * Process a peer ACK to our FIN.
+ */
+static int do_close_con_rpl(struct t3cdev *cdev, struct sk_buff *skb,
+			    void *ctx)
+{
+	struct cxgbi_sock *csk = ctx;
+	struct cpl_close_con_rpl *rpl = cplhdr(skb);
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_SOCK,
+		"csk 0x%p,%u,0x%lx,%u, snxt %u.\n",
+		csk, csk->state, csk->flags, csk->tid, ntohl(rpl->snd_nxt));
+
+	cxgbi_sock_rcv_close_conn_rpl(csk, ntohl(rpl->snd_nxt));
+	__kfree_skb(skb);
+	return 0;
+}
+
+/*
+ * Process ABORT_REQ_RSS CPL message: -> host
+ * Process abort requests.  If we are waiting for an ABORT_RPL we ignore this
+ * request except that we need to reply to it.
+ */
+
+static int abort_status_to_errno(struct cxgbi_sock *csk, int abort_reason,
+				 int *need_rst)
+{
+	switch (abort_reason) {
+	case CPL_ERR_BAD_SYN: /* fall through */
+	case CPL_ERR_CONN_RESET:
+		return csk->state > CTP_ESTABLISHED ?
+			-EPIPE : -ECONNRESET;
+	case CPL_ERR_XMIT_TIMEDOUT:
+	case CPL_ERR_PERSIST_TIMEDOUT:
+	case CPL_ERR_FINWAIT2_TIMEDOUT:
+	case CPL_ERR_KEEPALIVE_TIMEDOUT:
+		return -ETIMEDOUT;
+	default:
+		return -EIO;
+	}
+}
+
+static int do_abort_req(struct t3cdev *cdev, struct sk_buff *skb, void *ctx)
+{
+	const struct cpl_abort_req_rss *req = cplhdr(skb);
+	struct cxgbi_sock *csk = ctx;
+	int rst_status = CPL_ABORT_NO_RST;
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_SOCK,
+		"csk 0x%p,%u,0x%lx,%u.\n",
+		csk, csk->state, csk->flags, csk->tid);
+
+	if (req->status == CPL_ERR_RTX_NEG_ADVICE ||
+	    req->status == CPL_ERR_PERSIST_NEG_ADVICE) {
+		goto done;
+	}
+
+	cxgbi_sock_get(csk);
+	spin_lock_bh(&csk->lock);
+
+	if (!cxgbi_sock_flag(csk, CTPF_ABORT_REQ_RCVD)) {
+		cxgbi_sock_set_flag(csk, CTPF_ABORT_REQ_RCVD);
+		cxgbi_sock_set_state(csk, CTP_ABORTING);
+		goto out;
+	}
+
+	cxgbi_sock_clear_flag(csk, CTPF_ABORT_REQ_RCVD);
+	send_abort_rpl(csk, rst_status);
+
+	if (!cxgbi_sock_flag(csk, CTPF_ABORT_RPL_PENDING)) {
+		csk->err = abort_status_to_errno(csk, req->status, &rst_status);
+		cxgbi_sock_closed(csk);
+	}
+
+out:
+	spin_unlock_bh(&csk->lock);
+	cxgbi_sock_put(csk);
+done:
+	__kfree_skb(skb);
+	return 0;
+}
+
+/*
+ * Process ABORT_RPL_RSS CPL message: -> host
+ * Process abort replies.  We only process these messages if we anticipate
+ * them as the coordination between SW and HW in this area is somewhat lacking
+ * and sometimes we get ABORT_RPLs after we are done with the connection that
+ * originated the ABORT_REQ.
+ */
+static int do_abort_rpl(struct t3cdev *cdev, struct sk_buff *skb, void *ctx)
+{
+	struct cpl_abort_rpl_rss *rpl = cplhdr(skb);
+	struct cxgbi_sock *csk = ctx;
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_SOCK,
+		"status 0x%x, csk 0x%p, s %u, 0x%lx.\n",
+		rpl->status, csk, csk ? csk->state : 0,
+		csk ? csk->flags : 0UL);
+	/*
+	 * Ignore replies to post-close aborts indicating that the abort was
+	 * requested too late.  These connections are terminated when we get
+	 * PEER_CLOSE or CLOSE_CON_RPL and by the time the abort_rpl_rss
+	 * arrives the TID is either no longer used or it has been recycled.
+	 */
+	if (rpl->status == CPL_ERR_ABORT_FAILED)
+		goto rel_skb;
+	/*
+	 * Sometimes we've already closed the connection, e.g., a post-close
+	 * abort races with ABORT_REQ_RSS, the latter frees the connection
+	 * expecting the ABORT_REQ will fail with CPL_ERR_ABORT_FAILED,
+	 * but FW turns the ABORT_REQ into a regular one and so we get
+	 * ABORT_RPL_RSS with status 0 and no connection.
+	 */
+	if (csk)
+		cxgbi_sock_rcv_abort_rpl(csk);
+rel_skb:
+	__kfree_skb(skb);
+	return 0;
+}
+
+/*
+ * Process RX_ISCSI_HDR CPL message: -> host
+ * Handle received PDUs, the payload could be DDP'ed. If not, the payload
+ * follow after the bhs.
+ */
+static int do_iscsi_hdr(struct t3cdev *t3dev, struct sk_buff *skb, void *ctx)
+{
+	struct cxgbi_sock *csk = ctx;
+	struct cpl_iscsi_hdr *hdr_cpl = cplhdr(skb);
+	struct cpl_iscsi_hdr_norss data_cpl;
+	struct cpl_rx_data_ddp_norss ddp_cpl;
+	unsigned int hdr_len, data_len, status;
+	unsigned int len;
+	int err;
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_PDU_RX,
+		"csk 0x%p,%u,0x%lx,%u, skb 0x%p,%u.\n",
+		csk, csk->state, csk->flags, csk->tid, skb, skb->len);
+
+	spin_lock_bh(&csk->lock);
+
+	if (unlikely(csk->state >= CTP_PASSIVE_CLOSE)) {
+		log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_SOCK,
+			"csk 0x%p,%u,0x%lx,%u, bad state.\n",
+			csk, csk->state, csk->flags, csk->tid);
+		if (csk->state != CTP_ABORTING)
+			goto abort_conn;
+		else
+			goto discard;
+	}
+
+	cxgbi_skcb_tcp_seq(skb) = ntohl(hdr_cpl->seq);
+	cxgbi_skcb_flags(skb) = 0;
+
+	skb_reset_transport_header(skb);
+	__skb_pull(skb, sizeof(struct cpl_iscsi_hdr));
+
+	len = hdr_len = ntohs(hdr_cpl->len);
+	/* msg coalesce is off or not enough data received */
+	if (skb->len <= hdr_len) {
+		pr_err("%s: tid %u, CPL_ISCSI_HDR, skb len %u < %u.\n",
+			csk->cdev->ports[csk->port_id]->name, csk->tid,
+			skb->len, hdr_len);
+		goto abort_conn;
+	}
+	cxgbi_skcb_set_flag(skb, SKCBF_RX_COALESCED);
+
+	err = skb_copy_bits(skb, skb->len - sizeof(ddp_cpl), &ddp_cpl,
+			    sizeof(ddp_cpl));
+	if (err < 0) {
+		pr_err("%s: tid %u, copy cpl_ddp %u-%zu failed %d.\n",
+			csk->cdev->ports[csk->port_id]->name, csk->tid,
+			skb->len, sizeof(ddp_cpl), err);
+		goto abort_conn;
+	}
+
+	cxgbi_skcb_set_flag(skb, SKCBF_RX_STATUS);
+	cxgbi_skcb_rx_pdulen(skb) = ntohs(ddp_cpl.len);
+	cxgbi_skcb_rx_ddigest(skb) = ntohl(ddp_cpl.ulp_crc);
+	status = ntohl(ddp_cpl.ddp_status);
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_PDU_RX,
+		"csk 0x%p, skb 0x%p,%u, pdulen %u, status 0x%x.\n",
+		csk, skb, skb->len, cxgbi_skcb_rx_pdulen(skb), status);
+
+	if (status & (1 << CPL_RX_DDP_STATUS_HCRC_SHIFT))
+		cxgbi_skcb_set_flag(skb, SKCBF_RX_HCRC_ERR);
+	if (status & (1 << CPL_RX_DDP_STATUS_DCRC_SHIFT))
+		cxgbi_skcb_set_flag(skb, SKCBF_RX_DCRC_ERR);
+	if (status & (1 << CPL_RX_DDP_STATUS_PAD_SHIFT))
+		cxgbi_skcb_set_flag(skb, SKCBF_RX_PAD_ERR);
+
+	if (skb->len > (hdr_len + sizeof(ddp_cpl))) {
+		err = skb_copy_bits(skb, hdr_len, &data_cpl, sizeof(data_cpl));
+		if (err < 0) {
+			pr_err("%s: tid %u, cp %zu/%u failed %d.\n",
+				csk->cdev->ports[csk->port_id]->name,
+				csk->tid, sizeof(data_cpl), skb->len, err);
+			goto abort_conn;
+		}
+		data_len = ntohs(data_cpl.len);
+		log_debug(1 << CXGBI_DBG_DDP | 1 << CXGBI_DBG_PDU_RX,
+			"skb 0x%p, pdu not ddp'ed %u/%u, status 0x%x.\n",
+			skb, data_len, cxgbi_skcb_rx_pdulen(skb), status);
+		len += sizeof(data_cpl) + data_len;
+	} else if (status & (1 << CPL_RX_DDP_STATUS_DDP_SHIFT))
+		cxgbi_skcb_set_flag(skb, SKCBF_RX_DATA_DDPD);
+
+	csk->rcv_nxt = ntohl(ddp_cpl.seq) + cxgbi_skcb_rx_pdulen(skb);
+	__pskb_trim(skb, len);
+	__skb_queue_tail(&csk->receive_queue, skb);
+	cxgbi_conn_pdu_ready(csk);
+
+	spin_unlock_bh(&csk->lock);
+	return 0;
+
+abort_conn:
+	send_abort_req(csk);
+discard:
+	spin_unlock_bh(&csk->lock);
+	__kfree_skb(skb);
+	return 0;
+}
+
+/*
+ * Process TX_DATA_ACK CPL messages: -> host
+ * Process an acknowledgment of WR completion.  Advance snd_una and send the
+ * next batch of work requests from the write queue.
+ */
+static int do_wr_ack(struct t3cdev *cdev, struct sk_buff *skb, void *ctx)
+{
+	struct cxgbi_sock *csk = ctx;
+	struct cpl_wr_ack *hdr = cplhdr(skb);
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_PDU_RX,
+		"csk 0x%p,%u,0x%lx,%u, cr %u.\n",
+		csk, csk->state, csk->flags, csk->tid, ntohs(hdr->credits));
+
+	cxgbi_sock_rcv_wr_ack(csk, ntohs(hdr->credits), ntohl(hdr->snd_una), 1);
+	__kfree_skb(skb);
+	return 0;
+}
+
+/*
+ * for each connection, pre-allocate skbs needed for close/abort requests. So
+ * that we can service the request right away.
+ */
+static int alloc_cpls(struct cxgbi_sock *csk)
+{
+	csk->cpl_close = alloc_cpl(sizeof(struct cpl_close_con_req), 0,
+					GFP_KERNEL);
+	if (!csk->cpl_close)
+		return -ENOMEM;
+	csk->cpl_abort_req = alloc_cpl(sizeof(struct cpl_abort_req), 0,
+					GFP_KERNEL);
+	if (!csk->cpl_abort_req)
+		goto free_cpl_skbs;
+
+	csk->cpl_abort_rpl = alloc_cpl(sizeof(struct cpl_abort_rpl), 0,
+					GFP_KERNEL);
+	if (!csk->cpl_abort_rpl)
+		goto free_cpl_skbs;
+
+	return 0;
+
+free_cpl_skbs:
+	cxgbi_sock_free_cpl_skbs(csk);
+	return -ENOMEM;
+}
+
+/**
+ * release_offload_resources - release offload resource
+ * @c3cn: the offloaded iscsi tcp connection.
+ * Release resources held by an offload connection (TID, L2T entry, etc.)
+ */
+static void l2t_put(struct cxgbi_sock *csk)
+{
+	struct t3cdev *t3dev = (struct t3cdev *)csk->cdev->lldev;
+
+	if (csk->l2t) {
+		l2t_release(L2DATA(t3dev), csk->l2t);
+		csk->l2t = NULL;
+		cxgbi_sock_put(csk);
+	}
+}
+
+static void release_offload_resources(struct cxgbi_sock *csk)
+{
+	struct t3cdev *t3dev = (struct t3cdev *)csk->cdev->lldev;
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_SOCK,
+		"csk 0x%p,%u,0x%lx,%u.\n",
+		csk, csk->state, csk->flags, csk->tid);
+
+	csk->rss_qid = 0;
+	cxgbi_sock_free_cpl_skbs(csk);
+
+	if (csk->wr_cred != csk->wr_max_cred) {
+		cxgbi_sock_purge_wr_queue(csk);
+		cxgbi_sock_reset_wr_list(csk);
+	}
+	l2t_put(csk);
+	if (cxgbi_sock_flag(csk, CTPF_HAS_ATID))
+		free_atid(csk);
+	else if (cxgbi_sock_flag(csk, CTPF_HAS_TID)) {
+		cxgb3_remove_tid(t3dev, (void *)csk, csk->tid);
+		cxgbi_sock_clear_flag(csk, CTPF_HAS_TID);
+		cxgbi_sock_put(csk);
+	}
+	csk->dst = NULL;
+	csk->cdev = NULL;
+}
+
+static int init_act_open(struct cxgbi_sock *csk)
+{
+	struct dst_entry *dst = csk->dst;
+	struct cxgbi_device *cdev = csk->cdev;
+	struct t3cdev *t3dev = (struct t3cdev *)cdev->lldev;
+	struct net_device *ndev = cdev->ports[csk->port_id];
+	struct sk_buff *skb = NULL;
+
+	log_debug(1 << CXGBI_DBG_TOE | 1 << CXGBI_DBG_SOCK,
+		"csk 0x%p,%u,0x%lx.\n", csk, csk->state, csk->flags);
+
+	csk->rss_qid = 0;
+	csk->l2t = t3_l2t_get(t3dev, dst->neighbour, ndev);
+	if (!csk->l2t) {
+		pr_err("NO l2t available.\n");
+		return -EINVAL;
+	}
+	cxgbi_sock_get(csk);
+
+	csk->atid = cxgb3_alloc_atid(t3dev, &t3_client, csk);
+	if (csk->atid < 0) {
+		pr_err("NO atid available.\n");
+		goto rel_resource;
+	}
+	cxgbi_sock_set_flag(csk, CTPF_HAS_ATID);
+	cxgbi_sock_get(csk);
+
+	skb = alloc_cpl(sizeof(struct cpl_act_open_req), 0, GFP_KERNEL);
+	if (!skb)
+		goto rel_resource;
+	skb->sk = (struct sock *)csk;
+	set_arp_failure_handler(skb, act_open_arp_failure);
+
+	csk->wr_max_cred = csk->wr_cred = T3C_DATA(t3dev)->max_wrs - 1;
+	csk->wr_una_cred = 0;
+	csk->mss_idx = cxgbi_sock_select_mss(csk, dst_mtu(dst));
+	cxgbi_sock_reset_wr_list(csk);
+	csk->err = 0;
+
+	cxgbi_sock_set_state(csk, CTP_ACTIVE_OPEN);
+	send_act_open_req(csk, skb, csk->l2t);
+	return 0;
+
+rel_resource:
+	if (skb)
+		__kfree_skb(skb);
+	return -EINVAL;
+}
+
+cxgb3_cpl_handler_func cxgb3i_cpl_handlers[NUM_CPL_CMDS] = {
+	[CPL_ACT_ESTABLISH] = do_act_establish,
+	[CPL_ACT_OPEN_RPL] = do_act_open_rpl,
+	[CPL_PEER_CLOSE] = do_peer_close,
+	[CPL_ABORT_REQ_RSS] = do_abort_req,
+	[CPL_ABORT_RPL_RSS] = do_abort_rpl,
+	[CPL_CLOSE_CON_RPL] = do_close_con_rpl,
+	[CPL_TX_DMA_ACK] = do_wr_ack,
+	[CPL_ISCSI_HDR] = do_iscsi_hdr,
+};
+
+/**
+ * cxgb3i_ofld_init - allocate and initialize resources for each adapter found
+ * @cdev:	cxgbi adapter
+ */
+int cxgb3i_ofld_init(struct cxgbi_device *cdev)
+{
+	struct t3cdev *t3dev = (struct t3cdev *)cdev->lldev;
+	struct adap_ports port;
+	struct ofld_page_info rx_page_info;
+	unsigned int wr_len;
+	int rc;
+
+	if (t3dev->ctl(t3dev, GET_WR_LEN, &wr_len) < 0 ||
+	    t3dev->ctl(t3dev, GET_PORTS, &port) < 0 ||
+	    t3dev->ctl(t3dev, GET_RX_PAGE_INFO, &rx_page_info) < 0) {
+		pr_warn("t3 0x%p, offload up, ioctl failed.\n", t3dev);
+		return -EINVAL;
+	}
+
+	if (cxgb3i_max_connect > CXGBI_MAX_CONN)
+		cxgb3i_max_connect = CXGBI_MAX_CONN;
+
+	rc = cxgbi_device_portmap_create(cdev, cxgb3i_sport_base,
+					cxgb3i_max_connect);
+	if (rc < 0)
+		return rc;
+
+	init_wr_tab(wr_len);
+	cdev->csk_release_offload_resources = release_offload_resources;
+	cdev->csk_push_tx_frames = push_tx_frames;
+	cdev->csk_send_abort_req = send_abort_req;
+	cdev->csk_send_close_req = send_close_req;
+	cdev->csk_send_rx_credits = send_rx_credits;
+	cdev->csk_alloc_cpls = alloc_cpls;
+	cdev->csk_init_act_open = init_act_open;
+
+	pr_info("cdev 0x%p, offload up, added.\n", cdev);
+	return 0;
+}
+
+/*
+ * functions to program the pagepod in h/w
+ */
+static inline void ulp_mem_io_set_hdr(struct sk_buff *skb, unsigned int addr)
+{
+	struct ulp_mem_io *req = (struct ulp_mem_io *)skb->head;
+
+	memset(req, 0, sizeof(*req));
+
+	req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_BYPASS));
+	req->cmd_lock_addr = htonl(V_ULP_MEMIO_ADDR(addr >> 5) |
+				   V_ULPTX_CMD(ULP_MEM_WRITE));
+	req->len = htonl(V_ULP_MEMIO_DATA_LEN(PPOD_SIZE >> 5) |
+			 V_ULPTX_NFLITS((PPOD_SIZE >> 3) + 1));
+}
+
+static int ddp_set_map(struct cxgbi_sock *csk, struct cxgbi_pagepod_hdr *hdr,
+			unsigned int idx, unsigned int npods,
+				struct cxgbi_gather_list *gl)
+{
+	struct cxgbi_device *cdev = csk->cdev;
+	struct cxgbi_ddp_info *ddp = cdev->ddp;
+	unsigned int pm_addr = (idx << PPOD_SIZE_SHIFT) + ddp->llimit;
+	int i;
+
+	log_debug(1 << CXGBI_DBG_DDP,
+		"csk 0x%p, idx %u, npods %u, gl 0x%p.\n",
+		csk, idx, npods, gl);
+
+	for (i = 0; i < npods; i++, idx++, pm_addr += PPOD_SIZE) {
+		struct sk_buff *skb = ddp->gl_skb[idx];
+
+		/* hold on to the skb until we clear the ddp mapping */
+		skb_get(skb);
+
+		ulp_mem_io_set_hdr(skb, pm_addr);
+		cxgbi_ddp_ppod_set((struct cxgbi_pagepod *)(skb->head +
+					sizeof(struct ulp_mem_io)),
+				   hdr, gl, i * PPOD_PAGES_MAX);
+		skb->priority = CPL_PRIORITY_CONTROL;
+		cxgb3_ofld_send(cdev->lldev, skb);
+	}
+	return 0;
+}
+
+static void ddp_clear_map(struct cxgbi_hba *chba, unsigned int tag,
+			  unsigned int idx, unsigned int npods)
+{
+	struct cxgbi_device *cdev = chba->cdev;
+	struct cxgbi_ddp_info *ddp = cdev->ddp;
+	unsigned int pm_addr = (idx << PPOD_SIZE_SHIFT) + ddp->llimit;
+	int i;
+
+	log_debug(1 << CXGBI_DBG_DDP,
+		"cdev 0x%p, idx %u, npods %u, tag 0x%x.\n",
+		cdev, idx, npods, tag);
+
+	for (i = 0; i < npods; i++, idx++, pm_addr += PPOD_SIZE) {
+		struct sk_buff *skb = ddp->gl_skb[idx];
+
+		if (!skb) {
+			pr_err("tag 0x%x, 0x%x, %d/%u, skb NULL.\n",
+				tag, idx, i, npods);
+			continue;
+		}
+		ddp->gl_skb[idx] = NULL;
+		memset(skb->head + sizeof(struct ulp_mem_io), 0, PPOD_SIZE);
+		ulp_mem_io_set_hdr(skb, pm_addr);
+		skb->priority = CPL_PRIORITY_CONTROL;
+		cxgb3_ofld_send(cdev->lldev, skb);
+	}
+}
+
+static void ddp_free_gl_skb(struct cxgbi_ddp_info *ddp, int idx, int cnt)
+{
+	int i;
+
+	log_debug(1 << CXGBI_DBG_DDP,
+		"ddp 0x%p, idx %d, cnt %d.\n", ddp, idx, cnt);
+
+	for (i = 0; i < cnt; i++, idx++)
+		if (ddp->gl_skb[idx]) {
+			kfree_skb(ddp->gl_skb[idx]);
+			ddp->gl_skb[idx] = NULL;
+		}
+}
+
+static int ddp_alloc_gl_skb(struct cxgbi_ddp_info *ddp, int idx,
+				   int cnt, gfp_t gfp)
+{
+	int i;
+
+	log_debug(1 << CXGBI_DBG_DDP,
+		"ddp 0x%p, idx %d, cnt %d.\n", ddp, idx, cnt);
+
+	for (i = 0; i < cnt; i++) {
+		struct sk_buff *skb = alloc_cpl(sizeof(struct ulp_mem_io) +
+						PPOD_SIZE, 0, gfp);
+		if (skb) {
+			ddp->gl_skb[idx + i] = skb;
+		} else {
+			ddp_free_gl_skb(ddp, idx, i);
+			return -ENOMEM;
+		}
+	}
+	return 0;
+}
+
+static int ddp_setup_conn_pgidx(struct cxgbi_sock *csk,
+				       unsigned int tid, int pg_idx, bool reply)
+{
+	struct sk_buff *skb = alloc_cpl(sizeof(struct cpl_set_tcb_field), 0,
+					GFP_KERNEL);
+	struct cpl_set_tcb_field *req;
+	u64 val = pg_idx < DDP_PGIDX_MAX ? pg_idx : 0;
+
+	log_debug(1 << CXGBI_DBG_DDP,
+		"csk 0x%p, tid %u, pg_idx %d.\n", csk, tid, pg_idx);
+	if (!skb)
+		return -ENOMEM;
+
+	/* set up ulp submode and page size */
+	req = (struct cpl_set_tcb_field *)skb->head;
+	req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
+	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_SET_TCB_FIELD, tid));
+	req->reply = V_NO_REPLY(reply ? 0 : 1);
+	req->cpu_idx = 0;
+	req->word = htons(31);
+	req->mask = cpu_to_be64(0xF0000000);
+	req->val = cpu_to_be64(val << 28);
+	skb->priority = CPL_PRIORITY_CONTROL;
+
+	cxgb3_ofld_send(csk->cdev->lldev, skb);
+	return 0;
+}
+
+/**
+ * cxgb3i_setup_conn_digest - setup conn. digest setting
+ * @csk: cxgb tcp socket
+ * @tid: connection id
+ * @hcrc: header digest enabled
+ * @dcrc: data digest enabled
+ * @reply: request reply from h/w
+ * set up the iscsi digest settings for a connection identified by tid
+ */
+static int ddp_setup_conn_digest(struct cxgbi_sock *csk, unsigned int tid,
+			     int hcrc, int dcrc, int reply)
+{
+	struct sk_buff *skb = alloc_cpl(sizeof(struct cpl_set_tcb_field), 0,
+					GFP_KERNEL);
+	struct cpl_set_tcb_field *req;
+	u64 val = (hcrc ? 1 : 0) | (dcrc ? 2 : 0);
+
+	log_debug(1 << CXGBI_DBG_DDP,
+		"csk 0x%p, tid %u, crc %d,%d.\n", csk, tid, hcrc, dcrc);
+	if (!skb)
+		return -ENOMEM;
+
+	/* set up ulp submode and page size */
+	req = (struct cpl_set_tcb_field *)skb->head;
+	req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
+	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_SET_TCB_FIELD, tid));
+	req->reply = V_NO_REPLY(reply ? 0 : 1);
+	req->cpu_idx = 0;
+	req->word = htons(31);
+	req->mask = cpu_to_be64(0x0F000000);
+	req->val = cpu_to_be64(val << 24);
+	skb->priority = CPL_PRIORITY_CONTROL;
+
+	cxgb3_ofld_send(csk->cdev->lldev, skb);
+	return 0;
+}
+
+/**
+ * t3_ddp_cleanup - release the cxgb3 adapter's ddp resource
+ * @cdev: cxgb3i adapter
+ * release all the resource held by the ddp pagepod manager for a given
+ * adapter if needed
+ */
+
+static void t3_ddp_cleanup(struct cxgbi_device *cdev)
+{
+	struct t3cdev *tdev = (struct t3cdev *)cdev->lldev;
+
+	if (cxgbi_ddp_cleanup(cdev)) {
+		pr_info("t3dev 0x%p, ulp_iscsi no more user.\n", tdev);
+		tdev->ulp_iscsi = NULL;
+	}
+}
+
+/**
+ * ddp_init - initialize the cxgb3 adapter's ddp resource
+ * @cdev: cxgb3i adapter
+ * initialize the ddp pagepod manager for a given adapter
+ */
+static int cxgb3i_ddp_init(struct cxgbi_device *cdev)
+{
+	struct t3cdev *tdev = (struct t3cdev *)cdev->lldev;
+	struct cxgbi_ddp_info *ddp = tdev->ulp_iscsi;
+	struct ulp_iscsi_info uinfo;
+	unsigned int pgsz_factor[4];
+	int err;
+
+	if (ddp) {
+		kref_get(&ddp->refcnt);
+		pr_warn("t3dev 0x%p, ddp 0x%p already set up.\n",
+			tdev, tdev->ulp_iscsi);
+		cdev->ddp = ddp;
+		return -EALREADY;
+	}
+
+	err = tdev->ctl(tdev, ULP_ISCSI_GET_PARAMS, &uinfo);
+	if (err < 0) {
+		pr_err("%s, failed to get iscsi param err=%d.\n",
+			 tdev->name, err);
+		return err;
+	}
+
+	err = cxgbi_ddp_init(cdev, uinfo.llimit, uinfo.ulimit,
+			uinfo.max_txsz, uinfo.max_rxsz);
+	if (err < 0)
+		return err;
+
+	ddp = cdev->ddp;
+
+	uinfo.tagmask = ddp->idx_mask << PPOD_IDX_SHIFT;
+	cxgbi_ddp_page_size_factor(pgsz_factor);
+	uinfo.ulimit = uinfo.llimit + (ddp->nppods << PPOD_SIZE_SHIFT);
+
+	err = tdev->ctl(tdev, ULP_ISCSI_SET_PARAMS, &uinfo);
+	if (err < 0) {
+		pr_warn("%s unable to set iscsi param err=%d, ddp disabled.\n",
+			tdev->name, err);
+		cxgbi_ddp_cleanup(cdev);
+		return err;
+	}
+	tdev->ulp_iscsi = ddp;
+
+	cdev->csk_ddp_free_gl_skb = ddp_free_gl_skb;
+	cdev->csk_ddp_alloc_gl_skb = ddp_alloc_gl_skb;
+	cdev->csk_ddp_setup_digest = ddp_setup_conn_digest;
+	cdev->csk_ddp_setup_pgidx = ddp_setup_conn_pgidx;
+	cdev->csk_ddp_set = ddp_set_map;
+	cdev->csk_ddp_clear = ddp_clear_map;
+
+	pr_info("tdev 0x%p, nppods %u, bits %u, mask 0x%x,0x%x pkt %u/%u, "
+		"%u/%u.\n",
+		tdev, ddp->nppods, ddp->idx_bits, ddp->idx_mask,
+		ddp->rsvd_tag_mask, ddp->max_txsz, uinfo.max_txsz,
+		ddp->max_rxsz, uinfo.max_rxsz);
+	return 0;
+}
+
+static void cxgb3i_dev_close(struct t3cdev *t3dev)
+{
+	struct cxgbi_device *cdev = cxgbi_device_find_by_lldev(t3dev);
+
+	if (!cdev || cdev->flags & CXGBI_FLAG_ADAPTER_RESET) {
+		pr_info("0x%p close, f 0x%x.\n", cdev, cdev ? cdev->flags : 0);
+		return;
+	}
+
+	cxgbi_device_unregister(cdev);
+}
+
+/**
+ * cxgb3i_dev_open - init a t3 adapter structure and any h/w settings
+ * @t3dev: t3cdev adapter
+ */
+static void cxgb3i_dev_open(struct t3cdev *t3dev)
+{
+	struct cxgbi_device *cdev = cxgbi_device_find_by_lldev(t3dev);
+	struct adapter *adapter = tdev2adap(t3dev);
+	int i, err;
+
+	if (cdev) {
+		pr_info("0x%p, updating.\n", cdev);
+		return;
+	}
+
+	cdev = cxgbi_device_register(0, adapter->params.nports);
+	if (!cdev) {
+		pr_warn("device 0x%p register failed.\n", t3dev);
+		return;
+	}
+
+	cdev->flags = CXGBI_FLAG_DEV_T3 | CXGBI_FLAG_IPV4_SET;
+	cdev->lldev = t3dev;
+	cdev->pdev = adapter->pdev;
+	cdev->ports = adapter->port;
+	cdev->nports = adapter->params.nports;
+	cdev->mtus = adapter->params.mtus;
+	cdev->nmtus = NMTUS;
+	cdev->snd_win = cxgb3i_snd_win;
+	cdev->rcv_win = cxgb3i_rcv_win;
+	cdev->rx_credit_thres = cxgb3i_rx_credit_thres;
+	cdev->skb_tx_rsvd = CXGB3I_TX_HEADER_LEN;
+	cdev->skb_rx_extra = sizeof(struct cpl_iscsi_hdr_norss);
+	cdev->dev_ddp_cleanup = t3_ddp_cleanup;
+	cdev->itp = &cxgb3i_iscsi_transport;
+
+	err = cxgb3i_ddp_init(cdev);
+	if (err) {
+		pr_info("0x%p ddp init failed\n", cdev);
+		goto err_out;
+	}
+
+	err = cxgb3i_ofld_init(cdev);
+	if (err) {
+		pr_info("0x%p offload init failed\n", cdev);
+		goto err_out;
+	}
+
+	err = cxgbi_hbas_add(cdev, CXGB3I_MAX_LUN, CXGBI_MAX_CONN,
+				&cxgb3i_host_template, cxgb3i_stt);
+	if (err)
+		goto err_out;
+
+	for (i = 0; i < cdev->nports; i++)
+		cdev->hbas[i]->ipv4addr =
+			cxgb3i_get_private_ipv4addr(cdev->ports[i]);
+
+	pr_info("cdev 0x%p, f 0x%x, t3dev 0x%p open, err %d.\n",
+		cdev, cdev ? cdev->flags : 0, t3dev, err);
+	return;
+
+err_out:
+	cxgbi_device_unregister(cdev);
+}
+
+static void cxgb3i_dev_event_handler(struct t3cdev *t3dev, u32 event, u32 port)
+{
+	struct cxgbi_device *cdev = cxgbi_device_find_by_lldev(t3dev);
+
+	log_debug(1 << CXGBI_DBG_TOE,
+		"0x%p, cdev 0x%p, event 0x%x, port 0x%x.\n",
+		t3dev, cdev, event, port);
+	if (!cdev)
+		return;
+
+	switch (event) {
+	case OFFLOAD_STATUS_DOWN:
+		cdev->flags |= CXGBI_FLAG_ADAPTER_RESET;
+		break;
+	case OFFLOAD_STATUS_UP:
+		cdev->flags &= ~CXGBI_FLAG_ADAPTER_RESET;
+		break;
+	}
+}
+
+/**
+ * cxgb3i_init_module - module init entry point
+ *
+ * initialize any driver wide global data structures and register itself
+ *	with the cxgb3 module
+ */
+static int __init cxgb3i_init_module(void)
+{
+	int rc;
+
+	printk(KERN_INFO "%s", version);
+
+	rc = cxgbi_iscsi_init(&cxgb3i_iscsi_transport, &cxgb3i_stt);
+	if (rc < 0)
+		return rc;
+
+	cxgb3_register_client(&t3_client);
+	return 0;
+}
+
+/**
+ * cxgb3i_exit_module - module cleanup/exit entry point
+ *
+ * go through the driver hba list and for each hba, release any resource held.
+ *	and unregisters iscsi transport and the cxgb3 module
+ */
+static void __exit cxgb3i_exit_module(void)
+{
+	cxgb3_unregister_client(&t3_client);
+	cxgbi_device_unregister_all(CXGBI_FLAG_DEV_T3);
+	cxgbi_iscsi_cleanup(&cxgb3i_iscsi_transport, &cxgb3i_stt);
+}
+
+module_init(cxgb3i_init_module);
+module_exit(cxgb3i_exit_module);
diff --git a/drivers/scsi/cxgbi/cxgb3i/cxgb3i.h b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.h
new file mode 100644
index 0000000..5f5e339
--- /dev/null
+++ b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.h
@@ -0,0 +1,51 @@
+/*
+ * cxgb3i.h: Chelsio S3xx iSCSI driver.
+ *
+ * Copyright (c) 2008 Chelsio Communications, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ *
+ * Written by: Karen Xie (kxie@chelsio.com)
+ */
+
+#ifndef __CXGB3I_H__
+#define __CXGB3I_H__
+
+#define CXGB3I_SCSI_HOST_QDEPTH 1024
+#define CXGB3I_MAX_LUN		512
+#define ISCSI_PDU_NONPAYLOAD_MAX \
+	(sizeof(struct iscsi_hdr) + ISCSI_MAX_AHS_SIZE + 2*ISCSI_DIGEST_SIZE)
+
+/*for TX: a skb must have a headroom of at least TX_HEADER_LEN bytes */
+#define CXGB3I_TX_HEADER_LEN \
+	(sizeof(struct tx_data_wr) + sizeof(struct sge_opaque_hdr))
+
+extern cxgb3_cpl_handler_func cxgb3i_cpl_handlers[NUM_CPL_CMDS];
+
+#define cxgb3i_get_private_ipv4addr(ndev) \
+	(((struct port_info *)(netdev_priv(ndev)))->iscsi_ipv4addr)
+#define cxgb3i_set_private_ipv4addr(ndev, addr) \
+	(((struct port_info *)(netdev_priv(ndev)))->iscsi_ipv4addr) = addr
+
+struct cpl_iscsi_hdr_norss {
+	union opcode_tid ot;
+	u16 pdu_len_ddp;
+	u16 len;
+	u32 seq;
+	u16 urg;
+	u8 rsvd;
+	u8 status;
+};
+
+struct cpl_rx_data_ddp_norss {
+	union opcode_tid ot;
+	u16 urg;
+	u16 len;
+	u32 seq;
+	u32 nxt_seq;
+	u32 ulp_crc;
+	u32 ddp_status;
+};
+#endif