iwch_cm.c

linux内核源码

 * Upcall from the adapter indicating data has been transmitted.
 * For us its just the single MPA request or reply.  We can now free
 * the skb holding the mpa message.
 */
static int tx_ack(struct t3cdev *tdev, struct sk_buff *skb, void *ctx)
{
	struct iwch_ep *ep = ctx;
	struct cpl_wr_ack *hdr = cplhdr(skb);
	unsigned int credits = ntohs(hdr->credits);

	PDBG("%s ep %p credits %u\n", __FUNCTION__, ep, credits);

	if (credits == 0)
		return CPL_RET_BUF_DONE;
	BUG_ON(credits != 1);
	BUG_ON(ep->mpa_skb == NULL);
	kfree_skb(ep->mpa_skb);
	ep->mpa_skb = NULL;
	dst_confirm(ep->dst);
	if (state_read(&ep->com) == MPA_REP_SENT) {
		ep->com.rpl_done = 1;
		PDBG("waking up ep %p\n", ep);
		wake_up(&ep->com.waitq);
	}
	return CPL_RET_BUF_DONE;
}

static int abort_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx)
{
	struct iwch_ep *ep = ctx;

	PDBG("%s ep %p\n", __FUNCTION__, ep);

	/*
	 * We get 2 abort replies from the HW.  The first one must
	 * be ignored except for scribbling that we need one more.
	 */
	if (!(ep->flags & ABORT_REQ_IN_PROGRESS)) {
		ep->flags |= ABORT_REQ_IN_PROGRESS;
		return CPL_RET_BUF_DONE;
	}

	close_complete_upcall(ep);
	state_set(&ep->com, DEAD);
	release_ep_resources(ep);
	return CPL_RET_BUF_DONE;
}

/*
 * Return whether a failed active open has allocated a TID
 */
static inline int act_open_has_tid(int status)
{
	return status != CPL_ERR_TCAM_FULL && status != CPL_ERR_CONN_EXIST &&
	       status != CPL_ERR_ARP_MISS;
}

static int act_open_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx)
{
	struct iwch_ep *ep = ctx;
	struct cpl_act_open_rpl *rpl = cplhdr(skb);

	PDBG("%s ep %p status %u errno %d\n", __FUNCTION__, ep, rpl->status,
	     status2errno(rpl->status));
	connect_reply_upcall(ep, status2errno(rpl->status));
	state_set(&ep->com, DEAD);
	if (ep->com.tdev->type == T3B && act_open_has_tid(rpl->status))
		release_tid(ep->com.tdev, GET_TID(rpl), NULL);
	cxgb3_free_atid(ep->com.tdev, ep->atid);
	dst_release(ep->dst);
	l2t_release(L2DATA(ep->com.tdev), ep->l2t);
	put_ep(&ep->com);
	return CPL_RET_BUF_DONE;
}

static int listen_start(struct iwch_listen_ep *ep)
{
	struct sk_buff *skb;
	struct cpl_pass_open_req *req;

	PDBG("%s ep %p\n", __FUNCTION__, ep);
	skb = get_skb(NULL, sizeof(*req), GFP_KERNEL);
	if (!skb) {
		printk(KERN_ERR MOD "t3c_listen_start failed to alloc skb!\n");
		return -ENOMEM;
	}

	req = (struct cpl_pass_open_req *) skb_put(skb, sizeof(*req));
	req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_PASS_OPEN_REQ, ep->stid));
	req->local_port = ep->com.local_addr.sin_port;
	req->local_ip = ep->com.local_addr.sin_addr.s_addr;
	req->peer_port = 0;
	req->peer_ip = 0;
	req->peer_netmask = 0;
	req->opt0h = htonl(F_DELACK | F_TCAM_BYPASS);
	req->opt0l = htonl(V_RCV_BUFSIZ(rcv_win>>10));
	req->opt1 = htonl(V_CONN_POLICY(CPL_CONN_POLICY_ASK));

	skb->priority = 1;
	cxgb3_ofld_send(ep->com.tdev, skb);
	return 0;
}

static int pass_open_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx)
{
	struct iwch_listen_ep *ep = ctx;
	struct cpl_pass_open_rpl *rpl = cplhdr(skb);

	PDBG("%s ep %p status %d error %d\n", __FUNCTION__, ep,
	     rpl->status, status2errno(rpl->status));
	ep->com.rpl_err = status2errno(rpl->status);
	ep->com.rpl_done = 1;
	wake_up(&ep->com.waitq);

	return CPL_RET_BUF_DONE;
}

static int listen_stop(struct iwch_listen_ep *ep)
{
	struct sk_buff *skb;
	struct cpl_close_listserv_req *req;

	PDBG("%s ep %p\n", __FUNCTION__, ep);
	skb = get_skb(NULL, sizeof(*req), GFP_KERNEL);
	if (!skb) {
		printk(KERN_ERR MOD "%s - failed to alloc skb\n", __FUNCTION__);
		return -ENOMEM;
	}
	req = (struct cpl_close_listserv_req *) skb_put(skb, sizeof(*req));
	req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
	req->cpu_idx = 0;
	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_CLOSE_LISTSRV_REQ, ep->stid));
	skb->priority = 1;
	cxgb3_ofld_send(ep->com.tdev, skb);
	return 0;
}

static int close_listsrv_rpl(struct t3cdev *tdev, struct sk_buff *skb,
			     void *ctx)
{
	struct iwch_listen_ep *ep = ctx;
	struct cpl_close_listserv_rpl *rpl = cplhdr(skb);

	PDBG("%s ep %p\n", __FUNCTION__, ep);
	ep->com.rpl_err = status2errno(rpl->status);
	ep->com.rpl_done = 1;
	wake_up(&ep->com.waitq);
	return CPL_RET_BUF_DONE;
}

static void accept_cr(struct iwch_ep *ep, __be32 peer_ip, struct sk_buff *skb)
{
	struct cpl_pass_accept_rpl *rpl;
	unsigned int mtu_idx;
	u32 opt0h, opt0l, opt2;
	int wscale;

	PDBG("%s ep %p\n", __FUNCTION__, ep);
	BUG_ON(skb_cloned(skb));
	skb_trim(skb, sizeof(*rpl));
	skb_get(skb);
	mtu_idx = find_best_mtu(T3C_DATA(ep->com.tdev), dst_mtu(ep->dst));
	wscale = compute_wscale(rcv_win);
	opt0h = V_NAGLE(0) |
	    V_NO_CONG(nocong) |
	    V_KEEP_ALIVE(1) |
	    F_TCAM_BYPASS |
	    V_WND_SCALE(wscale) |
	    V_MSS_IDX(mtu_idx) |
	    V_L2T_IDX(ep->l2t->idx) | V_TX_CHANNEL(ep->l2t->smt_idx);
	opt0l = V_TOS((ep->tos >> 2) & M_TOS) | V_RCV_BUFSIZ(rcv_win>>10);
	opt2 = V_FLAVORS_VALID(1) | V_CONG_CONTROL_FLAVOR(cong_flavor);

	rpl = cplhdr(skb);
	rpl->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
	OPCODE_TID(rpl) = htonl(MK_OPCODE_TID(CPL_PASS_ACCEPT_RPL, ep->hwtid));
	rpl->peer_ip = peer_ip;
	rpl->opt0h = htonl(opt0h);
	rpl->opt0l_status = htonl(opt0l | CPL_PASS_OPEN_ACCEPT);
	rpl->opt2 = htonl(opt2);
	rpl->rsvd = rpl->opt2;	/* workaround for HW bug */
	skb->priority = CPL_PRIORITY_SETUP;
	l2t_send(ep->com.tdev, skb, ep->l2t);

	return;
}

static void reject_cr(struct t3cdev *tdev, u32 hwtid, __be32 peer_ip,
		      struct sk_buff *skb)
{
	PDBG("%s t3cdev %p tid %u peer_ip %x\n", __FUNCTION__, tdev, hwtid,
	     peer_ip);
	BUG_ON(skb_cloned(skb));
	skb_trim(skb, sizeof(struct cpl_tid_release));
	skb_get(skb);

	if (tdev->type == T3B)
		release_tid(tdev, hwtid, skb);
	else {
		struct cpl_pass_accept_rpl *rpl;

		rpl = cplhdr(skb);
		skb->priority = CPL_PRIORITY_SETUP;
		rpl->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
		OPCODE_TID(rpl) = htonl(MK_OPCODE_TID(CPL_PASS_ACCEPT_RPL,
						      hwtid));
		rpl->peer_ip = peer_ip;
		rpl->opt0h = htonl(F_TCAM_BYPASS);
		rpl->opt0l_status = htonl(CPL_PASS_OPEN_REJECT);
		rpl->opt2 = 0;
		rpl->rsvd = rpl->opt2;
		cxgb3_ofld_send(tdev, skb);
	}
}

static int pass_accept_req(struct t3cdev *tdev, struct sk_buff *skb, void *ctx)
{
	struct iwch_ep *child_ep, *parent_ep = ctx;
	struct cpl_pass_accept_req *req = cplhdr(skb);
	unsigned int hwtid = GET_TID(req);
	struct dst_entry *dst;
	struct l2t_entry *l2t;
	struct rtable *rt;
	struct iff_mac tim;

	PDBG("%s parent ep %p tid %u\n", __FUNCTION__, parent_ep, hwtid);

	if (state_read(&parent_ep->com) != LISTEN) {
		printk(KERN_ERR "%s - listening ep not in LISTEN\n",
		       __FUNCTION__);
		goto reject;
	}

	/*
	 * Find the netdev for this connection request.
	 */
	tim.mac_addr = req->dst_mac;
	tim.vlan_tag = ntohs(req->vlan_tag);
	if (tdev->ctl(tdev, GET_IFF_FROM_MAC, &tim) < 0 || !tim.dev) {
		printk(KERN_ERR
			"%s bad dst mac %02x %02x %02x %02x %02x %02x\n",
			__FUNCTION__,
			req->dst_mac[0],
			req->dst_mac[1],
			req->dst_mac[2],
			req->dst_mac[3],
			req->dst_mac[4],
			req->dst_mac[5]);
		goto reject;
	}

	/* Find output route */
	rt = find_route(tdev,
			req->local_ip,
			req->peer_ip,
			req->local_port,
			req->peer_port, G_PASS_OPEN_TOS(ntohl(req->tos_tid)));
	if (!rt) {
		printk(KERN_ERR MOD "%s - failed to find dst entry!\n",
		       __FUNCTION__);
		goto reject;
	}
	dst = &rt->u.dst;
	l2t = t3_l2t_get(tdev, dst->neighbour, dst->neighbour->dev);
	if (!l2t) {
		printk(KERN_ERR MOD "%s - failed to allocate l2t entry!\n",
		       __FUNCTION__);
		dst_release(dst);
		goto reject;
	}
	child_ep = alloc_ep(sizeof(*child_ep), GFP_KERNEL);
	if (!child_ep) {
		printk(KERN_ERR MOD "%s - failed to allocate ep entry!\n",
		       __FUNCTION__);
		l2t_release(L2DATA(tdev), l2t);
		dst_release(dst);
		goto reject;
	}
	state_set(&child_ep->com, CONNECTING);
	child_ep->com.tdev = tdev;
	child_ep->com.cm_id = NULL;
	child_ep->com.local_addr.sin_family = PF_INET;
	child_ep->com.local_addr.sin_port = req->local_port;
	child_ep->com.local_addr.sin_addr.s_addr = req->local_ip;
	child_ep->com.remote_addr.sin_family = PF_INET;
	child_ep->com.remote_addr.sin_port = req->peer_port;
	child_ep->com.remote_addr.sin_addr.s_addr = req->peer_ip;
	get_ep(&parent_ep->com);
	child_ep->parent_ep = parent_ep;
	child_ep->tos = G_PASS_OPEN_TOS(ntohl(req->tos_tid));
	child_ep->l2t = l2t;
	child_ep->dst = dst;
	child_ep->hwtid = hwtid;
	init_timer(&child_ep->timer);
	cxgb3_insert_tid(tdev, &t3c_client, child_ep, hwtid);
	accept_cr(child_ep, req->peer_ip, skb);
	goto out;
reject:
	reject_cr(tdev, hwtid, req->peer_ip, skb);
out:
	return CPL_RET_BUF_DONE;
}

static int pass_establish(struct t3cdev *tdev, struct sk_buff *skb, void *ctx)
{
	struct iwch_ep *ep = ctx;
	struct cpl_pass_establish *req = cplhdr(skb);

	PDBG("%s ep %p\n", __FUNCTION__, ep);
	ep->snd_seq = ntohl(req->snd_isn);
	ep->rcv_seq = ntohl(req->rcv_isn);

	set_emss(ep, ntohs(req->tcp_opt));

	dst_confirm(ep->dst);
	state_set(&ep->com, MPA_REQ_WAIT);
	start_ep_timer(ep);

	return CPL_RET_BUF_DONE;
}

static int peer_close(struct t3cdev *tdev, struct sk_buff *skb, void *ctx)
{
	struct iwch_ep *ep = ctx;
	struct iwch_qp_attributes attrs;
	unsigned long flags;
	int disconnect = 1;
	int release = 0;

	PDBG("%s ep %p\n", __FUNCTION__, ep);
	dst_confirm(ep->dst);

	spin_lock_irqsave(&ep->com.lock, flags);
	switch (ep->com.state) {
	case MPA_REQ_WAIT:
		__state_set(&ep->com, CLOSING);
		break;
	case MPA_REQ_SENT:
		__state_set(&ep->com, CLOSING);
		connect_reply_upcall(ep, -ECONNRESET);
		break;
	case MPA_REQ_RCVD:

		/*
		 * We're gonna mark this puppy DEAD, but keep
		 * the reference on it until the ULP accepts or
		 * rejects the CR.
		 */
		__state_set(&ep->com, CLOSING);
		get_ep(&ep->com);
		break;
	case MPA_REP_SENT:
		__state_set(&ep->com, CLOSING);
		ep->com.rpl_done = 1;
		ep->com.rpl_err = -ECONNRESET;
		PDBG("waking up ep %p\n", ep);
		wake_up(&ep->com.waitq);
		break;
	case FPDU_MODE:
		start_ep_timer(ep);
		__state_set(&ep->com, CLOSING);
		attrs.next_state = IWCH_QP_STATE_CLOSING;
		iwch_modify_qp(ep->com.qp->rhp, ep->com.qp,
			       IWCH_QP_ATTR_NEXT_STATE, &attrs, 1);
		peer_close_upcall(ep);
		break;
	case ABORTING:
		disconnect = 0;
		break;
	case CLOSING:
		__state_set(&ep->com, MORIBUND);
		disconnect = 0;
		break;
	case MORIBUND:
		stop_ep_timer(ep);
		if (ep->com.cm_id && ep->com.qp) {
			attrs.next_state = IWCH_QP_STATE_IDLE;
			iwch_modify_qp(ep->com.qp->rhp, ep->com.qp,
				       IWCH_QP_ATTR_NEXT_STATE, &attrs, 1);
		}
		close_complete_upcall(ep);
		__state_set(&ep->com, DEAD);
		release = 1;
		disconnect = 0;
		break;
	case DEAD:
		disconnect = 0;
		break;
	default:
		BUG_ON(1);
	}
	spin_unlock_irqrestore(&ep->com.lock, flags);
	if (disconnect)
		iwch_ep_disconnect(ep, 0, GFP_KERNEL);
	if (release)
		release_ep_resources(ep);
	return CPL_RET_BUF_DONE;
}

/*
 * Returns whether an ABORT_REQ_RSS message is a negative advice.
 */
static int is_neg_adv_abort(unsigned int status)
{
	return status == CPL_ERR_RTX_NEG_ADVICE ||
	       status == CPL_ERR_PERSIST_NEG_ADVICE;
}

static int peer_abort(struct t3cdev *tdev, struct sk_buff *skb, void *ctx)
{
	struct cpl_abort_req_rss *req = cplhdr(skb);
	struct iwch_ep *ep = ctx;
	struct cpl_abort_rpl *rpl;
	struct sk_buff *rpl_skb;
	struct iwch_qp_attributes attrs;
	int ret;
	int state;

	if (is_neg_adv_abort(req->status)) {
		PDBG("%s neg_adv_abort ep %p tid %d\n", __FUNCTION__, ep,
		     ep->hwtid);
		t3_l2t_send_event(ep->com.tdev, ep->l2t);
		return CPL_RET_BUF_DONE;
	}

	/*
	 * We get 2 peer aborts from the HW.  The first one must
	 * be ignored except for scribbling that we need one more.
	 */
	if (!(ep->flags & PEER_ABORT_IN_PROGRESS)) {
		ep->flags |= PEER_ABORT_IN_PROGRESS;
		return CPL_RET_BUF_DONE;
	}

	state = state_read(&ep->com);
	PDBG("%s ep %p state %u\n", __FUNCTION__, ep, state);
	switch (state) {
	case CONNECTING:
		break;
	case MPA_REQ_WAIT:
		stop_ep_timer(ep);
		break;
	case MPA_REQ_SENT:
		stop_ep_timer(ep);
		connect_reply_upcall(ep, -ECONNRESET);
		break;
	case MPA_REP_SENT:
		ep->com.rpl_done = 1;
		ep->com.rpl_err = -ECONNRESET;
		PDBG("waking up ep %p\n", ep);
		wake_up(&ep->com.waitq);
		break;
	case MPA_REQ_RCVD:

		/*
		 * We're gonna mark this puppy DEAD, but keep
		 * the reference on it until the ULP accepts or
		 * rejects the CR.
		 */
		get_ep(&ep->com);
		break;
	case MORIBUND:
	case CLOSING:
		stop_ep_timer(ep);
		/*FALLTHROUGH*/
	case FPDU_MODE:
		if (ep->com.cm_id && ep->com.qp) {
			attrs.next_state = IWCH_QP_STATE_ERROR;
			ret = iwch_modify_qp(ep->com.qp->rhp,
				     ep->com.qp, IWCH_QP_ATTR_NEXT_STATE,
				     &attrs, 1);
			if (ret)
				printk(KERN_ERR MOD
				       "%s - qp <- error failed!\n",
				       __FUNCTION__);
		}
		peer_abort_upcall(ep);
		break;
	case ABORTING:
		break;
	case DEAD:
		PDBG("%s PEER_ABORT IN DEAD STATE!!!!\n", __FUNCTION__);
		return CPL_RET_BUF_DONE;
	default:
		BUG_ON(1);
		break;
	}
	dst_confirm(ep->dst);

	rpl_skb = get_skb(skb, sizeof(*rpl), GFP_KERNEL);
	if (!rpl_skb) {
		printk(KERN_ERR MOD "%s - cannot allocate skb!\n",
		       __FUNCTION__);
		dst_release(ep->dst);
		l2t_release(L2DATA(ep->com.tdev), ep->l2t);
		put_ep(&ep->com);
		return CPL_RET_BUF_DONE;
	}
	rpl_skb->priority = CPL_PRIORITY_DATA;
	rpl = (struct cpl_abort_rpl *) skb_put(rpl_skb, sizeof(*rpl));
	rpl->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_HOST_ABORT_CON_RPL));
	rpl->wr.wr_lo = htonl(V_WR_TID(ep->hwtid));
	OPCODE_TID(rpl) = htonl(MK_OPCODE_TID(CPL_ABORT_RPL, ep->hwtid));
	rpl->cmd = CPL_ABORT_NO_RST;
	cxgb3_ofld_send(ep->com.tdev, rpl_skb);
	if (state != ABORTING) {
		state_set(&ep->com, DEAD);
		release_ep_resources(ep);
	}
	return CPL_RET_BUF_DONE;
}

static int close_con_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx)
{
	struct iwch_ep *ep = ctx;
	struct iwch_qp_attributes attrs;
	unsigned long flags;
	int release = 0;

	PDBG("%s ep %p\n", __FUNCTION__, ep);
	BUG_ON(!ep);

	/* The cm_id may be null if we failed to connect */
	spin_lock_irqsave(&ep->com.lock, flags);
	switch (ep->com.state) {
	case CLOSING: