~hc/RK356X_SDK_RELEASE.git

..	..	@@ -51,24 +51,48 @@
51	51
52	52	#include "hfi.h"
53	53	#include "qp.h"
	54	+#include "rc.h"
54	55	#include "verbs_txreq.h"
55	56	#include "trace.h"
56	57
57		-/* cut down ridiculously long IB macro names */
58		-#define OP(x) RC_OP(x)
59		-
60		-static u32 restart_sge(struct rvt_sge_state ss, struct rvt_swqe wqe,
61		- u32 psn, u32 pmtu)
	58	+struct rvt_ack_entry find_prev_entry(struct rvt_qp qp, u32 psn, u8 *prev,
	59	+ u8 prev_ack, bool scheduled)
	60	+ __must_hold(&qp->s_lock)
62	61	{
63		- u32 len;
	62	+ struct rvt_ack_entry *e = NULL;
	63	+ u8 i, p;
	64	+ bool s = true;
64	65
65		- len = delta_psn(psn, wqe->psn) * pmtu;
66		- ss->sge = wqe->sg_list[0];
67		- ss->sg_list = wqe->sg_list + 1;
68		- ss->num_sge = wqe->wr.num_sge;
69		- ss->total_len = wqe->length;
70		- rvt_skip_sge(ss, len, false);
71		- return wqe->length - len;
	66	+ for (i = qp->r_head_ack_queue; ; i = p) {
	67	+ if (i == qp->s_tail_ack_queue)
	68	+ s = false;
	69	+ if (i)
	70	+ p = i - 1;
	71	+ else
	72	+ p = rvt_size_atomic(ib_to_rvt(qp->ibqp.device));
	73	+ if (p == qp->r_head_ack_queue) {
	74	+ e = NULL;
	75	+ break;
	76	+ }
	77	+ e = &qp->s_ack_queue[p];
	78	+ if (!e->opcode) {
	79	+ e = NULL;
	80	+ break;
	81	+ }
	82	+ if (cmp_psn(psn, e->psn) >= 0) {
	83	+ if (p == qp->s_tail_ack_queue &&
	84	+ cmp_psn(psn, e->lpsn) <= 0)
	85	+ s = false;
	86	+ break;
	87	+ }
	88	+ }
	89	+ if (prev)
	90	+ *prev = p;
	91	+ if (prev_ack)
	92	+ *prev_ack = i;
	93	+ if (scheduled)
	94	+ *scheduled = s;
	95	+ return e;
72	96	}
73	97
74	98	/**
..	..	@@ -87,20 +111,25 @@
87	111	struct hfi1_pkt_state *ps)
88	112	{
89	113	struct rvt_ack_entry *e;
90		- u32 hwords;
91		- u32 len;
92		- u32 bth0;
93		- u32 bth2;
	114	+ u32 hwords, hdrlen;
	115	+ u32 len = 0;
	116	+ u32 bth0 = 0, bth2 = 0;
	117	+ u32 bth1 = qp->remote_qpn \| (HFI1_CAP_IS_KSET(OPFN) << IB_BTHE_E_SHIFT);
94	118	int middle = 0;
95	119	u32 pmtu = qp->pmtu;
96		- struct hfi1_qp_priv *priv = qp->priv;
	120	+ struct hfi1_qp_priv *qpriv = qp->priv;
	121	+ bool last_pkt;
	122	+ u32 delta;
	123	+ u8 next = qp->s_tail_ack_queue;
	124	+ struct tid_rdma_request *req;
97	125
	126	+ trace_hfi1_rsp_make_rc_ack(qp, 0);
98	127	lockdep_assert_held(&qp->s_lock);
99	128	/* Don't send an ACK if we aren't supposed to. */
100	129	if (!(ib_rvt_state_ops[qp->state] & RVT_PROCESS_RECV_OK))
101	130	goto bail;
102	131
103		- if (priv->hdr_type == HFI1_PKT_TYPE_9B)
	132	+ if (qpriv->hdr_type == HFI1_PKT_TYPE_9B)
104	133	/* header size in 32-bit words LRH+BTH = (8+12)/4. */
105	134	hwords = 5;
106	135	else
..	..	@@ -111,20 +140,27 @@
111	140	case OP(RDMA_READ_RESPONSE_LAST):
112	141	case OP(RDMA_READ_RESPONSE_ONLY):
113	142	e = &qp->s_ack_queue[qp->s_tail_ack_queue];
114		- if (e->rdma_sge.mr) {
115		- rvt_put_mr(e->rdma_sge.mr);
116		- e->rdma_sge.mr = NULL;
117		- }
118		- /* FALLTHROUGH */
	143	+ release_rdma_sge_mr(e);
	144	+ fallthrough;
119	145	case OP(ATOMIC_ACKNOWLEDGE):
120	146	/*
121	147	* We can increment the tail pointer now that the last
122	148	* response has been sent instead of only being
123	149	* constructed.
124	150	*/
125		- if (++qp->s_tail_ack_queue > HFI1_MAX_RDMA_ATOMIC)
126		- qp->s_tail_ack_queue = 0;
127		- /* FALLTHROUGH */
	151	+ if (++next > rvt_size_atomic(&dev->rdi))
	152	+ next = 0;
	153	+ /*
	154	+ * Only advance the s_acked_ack_queue pointer if there
	155	+ * have been no TID RDMA requests.
	156	+ */
	157	+ e = &qp->s_ack_queue[qp->s_tail_ack_queue];
	158	+ if (e->opcode != TID_OP(WRITE_REQ) &&
	159	+ qp->s_acked_ack_queue == qp->s_tail_ack_queue)
	160	+ qp->s_acked_ack_queue = next;
	161	+ qp->s_tail_ack_queue = next;
	162	+ trace_hfi1_rsp_make_rc_ack(qp, e->psn);
	163	+ fallthrough;
128	164	case OP(SEND_ONLY):
129	165	case OP(ACKNOWLEDGE):
130	166	/* Check for no next entry in the queue. */
..	..	@@ -135,6 +171,12 @@
135	171	}
136	172
137	173	e = &qp->s_ack_queue[qp->s_tail_ack_queue];
	174	+ /* Check for tid write fence */
	175	+ if ((qpriv->s_flags & HFI1_R_TID_WAIT_INTERLCK) \|\|
	176	+ hfi1_tid_rdma_ack_interlock(qp, e)) {
	177	+ iowait_set_flag(&qpriv->s_iowait, IOWAIT_PENDING_IB);
	178	+ goto bail;
	179	+ }
138	180	if (e->opcode == OP(RDMA_READ_REQUEST)) {
139	181	/*
140	182	* If a RDMA read response is being resent and
..	..	@@ -144,6 +186,10 @@
144	186	*/
145	187	len = e->rdma_sge.sge_length;
146	188	if (len && !e->rdma_sge.mr) {
	189	+ if (qp->s_acked_ack_queue ==
	190	+ qp->s_tail_ack_queue)
	191	+ qp->s_acked_ack_queue =
	192	+ qp->r_head_ack_queue;
147	193	qp->s_tail_ack_queue = qp->r_head_ack_queue;
148	194	goto bail;
149	195	}
..	..	@@ -165,6 +211,45 @@
165	211	hwords++;
166	212	qp->s_ack_rdma_psn = e->psn;
167	213	bth2 = mask_psn(qp->s_ack_rdma_psn++);
	214	+ } else if (e->opcode == TID_OP(WRITE_REQ)) {
	215	+ /*
	216	+ * If a TID RDMA WRITE RESP is being resent, we have to
	217	+ * wait for the actual request. All requests that are to
	218	+ * be resent will have their state set to
	219	+ * TID_REQUEST_RESEND. When the new request arrives, the
	220	+ * state will be changed to TID_REQUEST_RESEND_ACTIVE.
	221	+ */
	222	+ req = ack_to_tid_req(e);
	223	+ if (req->state == TID_REQUEST_RESEND \|\|
	224	+ req->state == TID_REQUEST_INIT_RESEND)
	225	+ goto bail;
	226	+ qp->s_ack_state = TID_OP(WRITE_RESP);
	227	+ qp->s_ack_rdma_psn = mask_psn(e->psn + req->cur_seg);
	228	+ goto write_resp;
	229	+ } else if (e->opcode == TID_OP(READ_REQ)) {
	230	+ /*
	231	+ * If a TID RDMA read response is being resent and
	232	+ * we haven't seen the duplicate request yet,
	233	+ * then stop sending the remaining responses the
	234	+ * responder has seen until the requester re-sends it.
	235	+ */
	236	+ len = e->rdma_sge.sge_length;
	237	+ if (len && !e->rdma_sge.mr) {
	238	+ if (qp->s_acked_ack_queue ==
	239	+ qp->s_tail_ack_queue)
	240	+ qp->s_acked_ack_queue =
	241	+ qp->r_head_ack_queue;
	242	+ qp->s_tail_ack_queue = qp->r_head_ack_queue;
	243	+ goto bail;
	244	+ }
	245	+ /* Copy SGE state in case we need to resend */
	246	+ ps->s_txreq->mr = e->rdma_sge.mr;
	247	+ if (ps->s_txreq->mr)
	248	+ rvt_get_mr(ps->s_txreq->mr);
	249	+ qp->s_ack_rdma_sge.sge = e->rdma_sge;
	250	+ qp->s_ack_rdma_sge.num_sge = 1;
	251	+ qp->s_ack_state = TID_OP(READ_RESP);
	252	+ goto read_resp;
168	253	} else {
169	254	/* COMPARE_SWAP or FETCH_ADD */
170	255	ps->s_txreq->ss = NULL;
..	..	@@ -176,12 +261,13 @@
176	261	bth2 = mask_psn(e->psn);
177	262	e->sent = 1;
178	263	}
	264	+ trace_hfi1_tid_write_rsp_make_rc_ack(qp);
179	265	bth0 = qp->s_ack_state << 24;
180	266	break;
181	267
182	268	case OP(RDMA_READ_RESPONSE_FIRST):
183	269	qp->s_ack_state = OP(RDMA_READ_RESPONSE_MIDDLE);
184		- /* FALLTHROUGH */
	270	+ fallthrough;
185	271	case OP(RDMA_READ_RESPONSE_MIDDLE):
186	272	ps->s_txreq->ss = &qp->s_ack_rdma_sge;
187	273	ps->s_txreq->mr = qp->s_ack_rdma_sge.sge.mr;
..	..	@@ -202,6 +288,84 @@
202	288	bth2 = mask_psn(qp->s_ack_rdma_psn++);
203	289	break;
204	290
	291	+ case TID_OP(WRITE_RESP):
	292	+write_resp:
	293	+ /*
	294	+ * 1. Check if RVT_S_ACK_PENDING is set. If yes,
	295	+ * goto normal.
	296	+ * 2. Attempt to allocate TID resources.
	297	+ * 3. Remove RVT_S_RESP_PENDING flags from s_flags
	298	+ * 4. If resources not available:
	299	+ * 4.1 Set RVT_S_WAIT_TID_SPACE
	300	+ * 4.2 Queue QP on RCD TID queue
	301	+ * 4.3 Put QP on iowait list.
	302	+ * 4.4 Build IB RNR NAK with appropriate timeout value
	303	+ * 4.5 Return indication progress made.
	304	+ * 5. If resources are available:
	305	+ * 5.1 Program HW flow CSRs
	306	+ * 5.2 Build TID RDMA WRITE RESP packet
	307	+ * 5.3 If more resources needed, do 2.1 - 2.3.
	308	+ * 5.4 Wake up next QP on RCD TID queue.
	309	+ * 5.5 Return indication progress made.
	310	+ */
	311	+
	312	+ e = &qp->s_ack_queue[qp->s_tail_ack_queue];
	313	+ req = ack_to_tid_req(e);
	314	+
	315	+ /*
	316	+ * Send scheduled RNR NAK's. RNR NAK's need to be sent at
	317	+ * segment boundaries, not at request boundaries. Don't change
	318	+ * s_ack_state because we are still in the middle of a request
	319	+ */
	320	+ if (qpriv->rnr_nak_state == TID_RNR_NAK_SEND &&
	321	+ qp->s_tail_ack_queue == qpriv->r_tid_alloc &&
	322	+ req->cur_seg == req->alloc_seg) {
	323	+ qpriv->rnr_nak_state = TID_RNR_NAK_SENT;
	324	+ goto normal_no_state;
	325	+ }
	326	+
	327	+ bth2 = mask_psn(qp->s_ack_rdma_psn);
	328	+ hdrlen = hfi1_build_tid_rdma_write_resp(qp, e, ohdr, &bth1,
	329	+ bth2, &len,
	330	+ &ps->s_txreq->ss);
	331	+ if (!hdrlen)
	332	+ return 0;
	333	+
	334	+ hwords += hdrlen;
	335	+ bth0 = qp->s_ack_state << 24;
	336	+ qp->s_ack_rdma_psn++;
	337	+ trace_hfi1_tid_req_make_rc_ack_write(qp, 0, e->opcode, e->psn,
	338	+ e->lpsn, req);
	339	+ if (req->cur_seg != req->total_segs)
	340	+ break;
	341	+
	342	+ e->sent = 1;
	343	+ /* Do not free e->rdma_sge until all data are received */
	344	+ qp->s_ack_state = OP(ATOMIC_ACKNOWLEDGE);
	345	+ break;
	346	+
	347	+ case TID_OP(READ_RESP):
	348	+read_resp:
	349	+ e = &qp->s_ack_queue[qp->s_tail_ack_queue];
	350	+ ps->s_txreq->ss = &qp->s_ack_rdma_sge;
	351	+ delta = hfi1_build_tid_rdma_read_resp(qp, e, ohdr, &bth0,
	352	+ &bth1, &bth2, &len,
	353	+ &last_pkt);
	354	+ if (delta == 0)
	355	+ goto error_qp;
	356	+ hwords += delta;
	357	+ if (last_pkt) {
	358	+ e->sent = 1;
	359	+ /*
	360	+ * Increment qp->s_tail_ack_queue through s_ack_state
	361	+ * transition.
	362	+ */
	363	+ qp->s_ack_state = OP(RDMA_READ_RESPONSE_LAST);
	364	+ }
	365	+ break;
	366	+ case TID_OP(READ_REQ):
	367	+ goto bail;
	368	+
205	369	default:
206	370	normal:
207	371	/*
..	..	@@ -211,8 +375,7 @@
211	375	* (see above).
212	376	*/
213	377	qp->s_ack_state = OP(SEND_ONLY);
214		- qp->s_flags &= ~RVT_S_ACK_PENDING;
215		- ps->s_txreq->ss = NULL;
	378	+normal_no_state:
216	379	if (qp->s_nak_state)
217	380	ohdr->u.aeth =
218	381	cpu_to_be32((qp->r_msn & IB_MSN_MASK) \|
..	..	@@ -224,14 +387,24 @@
224	387	len = 0;
225	388	bth0 = OP(ACKNOWLEDGE) << 24;
226	389	bth2 = mask_psn(qp->s_ack_psn);
	390	+ qp->s_flags &= ~RVT_S_ACK_PENDING;
	391	+ ps->s_txreq->txreq.flags \|= SDMA_TXREQ_F_VIP;
	392	+ ps->s_txreq->ss = NULL;
227	393	}
228	394	qp->s_rdma_ack_cnt++;
229		- ps->s_txreq->sde = priv->s_sde;
	395	+ ps->s_txreq->sde = qpriv->s_sde;
230	396	ps->s_txreq->s_cur_size = len;
231	397	ps->s_txreq->hdr_dwords = hwords;
232		- hfi1_make_ruc_header(qp, ohdr, bth0, bth2, middle, ps);
	398	+ hfi1_make_ruc_header(qp, ohdr, bth0, bth1, bth2, middle, ps);
233	399	return 1;
234		-
	400	+error_qp:
	401	+ spin_unlock_irqrestore(&qp->s_lock, ps->flags);
	402	+ spin_lock_irqsave(&qp->r_lock, ps->flags);
	403	+ spin_lock(&qp->s_lock);
	404	+ rvt_error_qp(qp, IB_WC_WR_FLUSH_ERR);
	405	+ spin_unlock(&qp->s_lock);
	406	+ spin_unlock_irqrestore(&qp->r_lock, ps->flags);
	407	+ spin_lock_irqsave(&qp->s_lock, ps->flags);
235	408	bail:
236	409	qp->s_ack_state = OP(ACKNOWLEDGE);
237	410	/*
..	..	@@ -258,17 +431,23 @@
258	431	struct hfi1_qp_priv *priv = qp->priv;
259	432	struct hfi1_ibdev *dev = to_idev(qp->ibqp.device);
260	433	struct ib_other_headers *ohdr;
261		- struct rvt_sge_state *ss;
	434	+ struct rvt_sge_state *ss = NULL;
262	435	struct rvt_swqe *wqe;
263		- u32 hwords;
264		- u32 len;
265		- u32 bth0 = 0;
266		- u32 bth2;
	436	+ struct hfi1_swqe_priv *wpriv;
	437	+ struct tid_rdma_request *req = NULL;
	438	+ /* header size in 32-bit words LRH+BTH = (8+12)/4. */
	439	+ u32 hwords = 5;
	440	+ u32 len = 0;
	441	+ u32 bth0 = 0, bth2 = 0;
	442	+ u32 bth1 = qp->remote_qpn \| (HFI1_CAP_IS_KSET(OPFN) << IB_BTHE_E_SHIFT);
267	443	u32 pmtu = qp->pmtu;
268	444	char newreq;
269	445	int middle = 0;
270	446	int delta;
	447	+ struct tid_rdma_flow *flow = NULL;
	448	+ struct tid_rdma_params *remote;
271	449
	450	+ trace_hfi1_sender_make_rc_req(qp);
272	451	lockdep_assert_held(&qp->s_lock);
273	452	ps->s_txreq = get_txreq(ps->dev, qp);
274	453	if (!ps->s_txreq)
..	..	@@ -309,13 +488,13 @@
309	488	}
310	489	clear_ahg(qp);
311	490	wqe = rvt_get_swqe_ptr(qp, qp->s_last);
312		- hfi1_send_complete(qp, wqe, qp->s_last != qp->s_acked ?
313		- IB_WC_SUCCESS : IB_WC_WR_FLUSH_ERR);
	491	+ hfi1_trdma_send_complete(qp, wqe, qp->s_last != qp->s_acked ?
	492	+ IB_WC_SUCCESS : IB_WC_WR_FLUSH_ERR);
314	493	/* will get called again */
315	494	goto done_free_tx;
316	495	}
317	496
318		- if (qp->s_flags & (RVT_S_WAIT_RNR \| RVT_S_WAIT_ACK))
	497	+ if (qp->s_flags & (RVT_S_WAIT_RNR \| RVT_S_WAIT_ACK \| HFI1_S_WAIT_HALT))
319	498	goto bail;
320	499
321	500	if (cmp_psn(qp->s_psn, qp->s_sending_hpsn) <= 0) {
..	..	@@ -329,6 +508,7 @@
329	508
330	509	/* Send a request. */
331	510	wqe = rvt_get_swqe_ptr(qp, qp->s_cur);
	511	+check_s_state:
332	512	switch (qp->s_state) {
333	513	default:
334	514	if (!(ib_rvt_state_ops[qp->state] & RVT_PROCESS_NEXT_SEND_OK))
..	..	@@ -350,9 +530,13 @@
350	530	/*
351	531	* If a fence is requested, wait for previous
352	532	* RDMA read and atomic operations to finish.
	533	+ * However, there is no need to guard against
	534	+ * TID RDMA READ after TID RDMA READ.
353	535	*/
354	536	if ((wqe->wr.send_flags & IB_SEND_FENCE) &&
355		- qp->s_num_rd_atomic) {
	537	+ qp->s_num_rd_atomic &&
	538	+ (wqe->wr.opcode != IB_WR_TID_RDMA_READ \|\|
	539	+ priv->pending_tid_r_segs < qp->s_num_rd_atomic)) {
356	540	qp->s_flags \|= RVT_S_WAIT_FENCE;
357	541	goto bail;
358	542	}
..	..	@@ -378,9 +562,9 @@
378	562	wqe->wr.ex.invalidate_rkey);
379	563	local_ops = 1;
380	564	}
381		- hfi1_send_complete(qp, wqe,
382		- err ? IB_WC_LOC_PROT_ERR
383		- : IB_WC_SUCCESS);
	565	+ rvt_send_complete(qp, wqe,
	566	+ err ? IB_WC_LOC_PROT_ERR
	567	+ : IB_WC_SUCCESS);
384	568	if (local_ops)
385	569	atomic_dec(&qp->local_ops_pending);
386	570	goto done_free_tx;
..	..	@@ -397,16 +581,22 @@
397	581	len = wqe->length;
398	582	ss = &qp->s_sge;
399	583	bth2 = mask_psn(qp->s_psn);
	584	+
	585	+ /*
	586	+ * Interlock between various IB requests and TID RDMA
	587	+ * if necessary.
	588	+ */
	589	+ if ((priv->s_flags & HFI1_S_TID_WAIT_INTERLCK) \|\|
	590	+ hfi1_tid_rdma_wqe_interlock(qp, wqe))
	591	+ goto bail;
	592	+
400	593	switch (wqe->wr.opcode) {
401	594	case IB_WR_SEND:
402	595	case IB_WR_SEND_WITH_IMM:
403	596	case IB_WR_SEND_WITH_INV:
404	597	/* If no credit, return. */
405		- if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT) &&
406		- rvt_cmp_msn(wqe->ssn, qp->s_lsn + 1) > 0) {
407		- qp->s_flags \|= RVT_S_WAIT_SSN_CREDIT;
	598	+ if (!rvt_rc_credit_avail(qp, wqe))
408	599	goto bail;
409		- }
410	600	if (len > pmtu) {
411	601	qp->s_state = OP(SEND_FIRST);
412	602	len = pmtu;
..	..	@@ -439,11 +629,8 @@
439	629	goto no_flow_control;
440	630	case IB_WR_RDMA_WRITE_WITH_IMM:
441	631	/* If no credit, return. */
442		- if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT) &&
443		- rvt_cmp_msn(wqe->ssn, qp->s_lsn + 1) > 0) {
444		- qp->s_flags \|= RVT_S_WAIT_SSN_CREDIT;
	632	+ if (!rvt_rc_credit_avail(qp, wqe))
445	633	goto bail;
446		- }
447	634	no_flow_control:
448	635	put_ib_reth_vaddr(
449	636	wqe->rdma_wr.remote_addr,
..	..	@@ -473,21 +660,126 @@
473	660	qp->s_cur = 0;
474	661	break;
475	662
	663	+ case IB_WR_TID_RDMA_WRITE:
	664	+ if (newreq) {
	665	+ /*
	666	+ * Limit the number of TID RDMA WRITE requests.
	667	+ */
	668	+ if (atomic_read(&priv->n_tid_requests) >=
	669	+ HFI1_TID_RDMA_WRITE_CNT)
	670	+ goto bail;
	671	+
	672	+ if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT))
	673	+ qp->s_lsn++;
	674	+ }
	675	+
	676	+ hwords += hfi1_build_tid_rdma_write_req(qp, wqe, ohdr,
	677	+ &bth1, &bth2,
	678	+ &len);
	679	+ ss = NULL;
	680	+ if (priv->s_tid_cur == HFI1_QP_WQE_INVALID) {
	681	+ priv->s_tid_cur = qp->s_cur;
	682	+ if (priv->s_tid_tail == HFI1_QP_WQE_INVALID) {
	683	+ priv->s_tid_tail = qp->s_cur;
	684	+ priv->s_state = TID_OP(WRITE_RESP);
	685	+ }
	686	+ } else if (priv->s_tid_cur == priv->s_tid_head) {
	687	+ struct rvt_swqe *__w;
	688	+ struct tid_rdma_request *__r;
	689	+
	690	+ __w = rvt_get_swqe_ptr(qp, priv->s_tid_cur);
	691	+ __r = wqe_to_tid_req(__w);
	692	+
	693	+ /*
	694	+ * The s_tid_cur pointer is advanced to s_cur if
	695	+ * any of the following conditions about the WQE
	696	+ * to which s_ti_cur currently points to are
	697	+ * satisfied:
	698	+ * 1. The request is not a TID RDMA WRITE
	699	+ * request,
	700	+ * 2. The request is in the INACTIVE or
	701	+ * COMPLETE states (TID RDMA READ requests
	702	+ * stay at INACTIVE and TID RDMA WRITE
	703	+ * transition to COMPLETE when done),
	704	+ * 3. The request is in the ACTIVE or SYNC
	705	+ * state and the number of completed
	706	+ * segments is equal to the total segment
	707	+ * count.
	708	+ * (If ACTIVE, the request is waiting for
	709	+ * ACKs. If SYNC, the request has not
	710	+ * received any responses because it's
	711	+ * waiting on a sync point.)
	712	+ */
	713	+ if (__w->wr.opcode != IB_WR_TID_RDMA_WRITE \|\|
	714	+ __r->state == TID_REQUEST_INACTIVE \|\|
	715	+ __r->state == TID_REQUEST_COMPLETE \|\|
	716	+ ((__r->state == TID_REQUEST_ACTIVE \|\|
	717	+ __r->state == TID_REQUEST_SYNC) &&
	718	+ __r->comp_seg == __r->total_segs)) {
	719	+ if (priv->s_tid_tail ==
	720	+ priv->s_tid_cur &&
	721	+ priv->s_state ==
	722	+ TID_OP(WRITE_DATA_LAST)) {
	723	+ priv->s_tid_tail = qp->s_cur;
	724	+ priv->s_state =
	725	+ TID_OP(WRITE_RESP);
	726	+ }
	727	+ priv->s_tid_cur = qp->s_cur;
	728	+ }
	729	+ /*
	730	+ * A corner case: when the last TID RDMA WRITE
	731	+ * request was completed, s_tid_head,
	732	+ * s_tid_cur, and s_tid_tail all point to the
	733	+ * same location. Other requests are posted and
	734	+ * s_cur wraps around to the same location,
	735	+ * where a new TID RDMA WRITE is posted. In
	736	+ * this case, none of the indices need to be
	737	+ * updated. However, the priv->s_state should.
	738	+ */
	739	+ if (priv->s_tid_tail == qp->s_cur &&
	740	+ priv->s_state == TID_OP(WRITE_DATA_LAST))
	741	+ priv->s_state = TID_OP(WRITE_RESP);
	742	+ }
	743	+ req = wqe_to_tid_req(wqe);
	744	+ if (newreq) {
	745	+ priv->s_tid_head = qp->s_cur;
	746	+ priv->pending_tid_w_resp += req->total_segs;
	747	+ atomic_inc(&priv->n_tid_requests);
	748	+ atomic_dec(&priv->n_requests);
	749	+ } else {
	750	+ req->state = TID_REQUEST_RESEND;
	751	+ req->comp_seg = delta_psn(bth2, wqe->psn);
	752	+ /*
	753	+ * Pull back any segments since we are going
	754	+ * to re-receive them.
	755	+ */
	756	+ req->setup_head = req->clear_tail;
	757	+ priv->pending_tid_w_resp +=
	758	+ delta_psn(wqe->lpsn, bth2) + 1;
	759	+ }
	760	+
	761	+ trace_hfi1_tid_write_sender_make_req(qp, newreq);
	762	+ trace_hfi1_tid_req_make_req_write(qp, newreq,
	763	+ wqe->wr.opcode,
	764	+ wqe->psn, wqe->lpsn,
	765	+ req);
	766	+ if (++qp->s_cur == qp->s_size)
	767	+ qp->s_cur = 0;
	768	+ break;
	769	+
476	770	case IB_WR_RDMA_READ:
477	771	/*
478	772	* Don't allow more operations to be started
479	773	* than the QP limits allow.
480	774	*/
481		- if (newreq) {
482		- if (qp->s_num_rd_atomic >=
483		- qp->s_max_rd_atomic) {
484		- qp->s_flags \|= RVT_S_WAIT_RDMAR;
485		- goto bail;
486		- }
487		- qp->s_num_rd_atomic++;
488		- if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT))
489		- qp->s_lsn++;
	775	+ if (qp->s_num_rd_atomic >=
	776	+ qp->s_max_rd_atomic) {
	777	+ qp->s_flags \|= RVT_S_WAIT_RDMAR;
	778	+ goto bail;
490	779	}
	780	+ qp->s_num_rd_atomic++;
	781	+ if (newreq && !(qp->s_flags & RVT_S_UNLIMITED_CREDIT))
	782	+ qp->s_lsn++;
491	783	put_ib_reth_vaddr(
492	784	wqe->rdma_wr.remote_addr,
493	785	&ohdr->u.rc.reth);
..	..	@@ -503,23 +795,98 @@
503	795	qp->s_cur = 0;
504	796	break;
505	797
	798	+ case IB_WR_TID_RDMA_READ:
	799	+ trace_hfi1_tid_read_sender_make_req(qp, newreq);
	800	+ wpriv = wqe->priv;
	801	+ req = wqe_to_tid_req(wqe);
	802	+ trace_hfi1_tid_req_make_req_read(qp, newreq,
	803	+ wqe->wr.opcode,
	804	+ wqe->psn, wqe->lpsn,
	805	+ req);
	806	+ delta = cmp_psn(qp->s_psn, wqe->psn);
	807	+
	808	+ /*
	809	+ * Don't allow more operations to be started
	810	+ * than the QP limits allow. We could get here under
	811	+ * three conditions; (1) It's a new request; (2) We are
	812	+ * sending the second or later segment of a request,
	813	+ * but the qp->s_state is set to OP(RDMA_READ_REQUEST)
	814	+ * when the last segment of a previous request is
	815	+ * received just before this; (3) We are re-sending a
	816	+ * request.
	817	+ */
	818	+ if (qp->s_num_rd_atomic >= qp->s_max_rd_atomic) {
	819	+ qp->s_flags \|= RVT_S_WAIT_RDMAR;
	820	+ goto bail;
	821	+ }
	822	+ if (newreq) {
	823	+ struct tid_rdma_flow *flow =
	824	+ &req->flows[req->setup_head];
	825	+
	826	+ /*
	827	+ * Set up s_sge as it is needed for TID
	828	+ * allocation. However, if the pages have been
	829	+ * walked and mapped, skip it. An earlier try
	830	+ * has failed to allocate the TID entries.
	831	+ */
	832	+ if (!flow->npagesets) {
	833	+ qp->s_sge.sge = wqe->sg_list[0];
	834	+ qp->s_sge.sg_list = wqe->sg_list + 1;
	835	+ qp->s_sge.num_sge = wqe->wr.num_sge;
	836	+ qp->s_sge.total_len = wqe->length;
	837	+ qp->s_len = wqe->length;
	838	+ req->isge = 0;
	839	+ req->clear_tail = req->setup_head;
	840	+ req->flow_idx = req->setup_head;
	841	+ req->state = TID_REQUEST_ACTIVE;
	842	+ }
	843	+ } else if (delta == 0) {
	844	+ /* Re-send a request */
	845	+ req->cur_seg = 0;
	846	+ req->comp_seg = 0;
	847	+ req->ack_pending = 0;
	848	+ req->flow_idx = req->clear_tail;
	849	+ req->state = TID_REQUEST_RESEND;
	850	+ }
	851	+ req->s_next_psn = qp->s_psn;
	852	+ /* Read one segment at a time */
	853	+ len = min_t(u32, req->seg_len,
	854	+ wqe->length - req->seg_len * req->cur_seg);
	855	+ delta = hfi1_build_tid_rdma_read_req(qp, wqe, ohdr,
	856	+ &bth1, &bth2,
	857	+ &len);
	858	+ if (delta <= 0) {
	859	+ /* Wait for TID space */
	860	+ goto bail;
	861	+ }
	862	+ if (newreq && !(qp->s_flags & RVT_S_UNLIMITED_CREDIT))
	863	+ qp->s_lsn++;
	864	+ hwords += delta;
	865	+ ss = &wpriv->ss;
	866	+ /* Check if this is the last segment */
	867	+ if (req->cur_seg >= req->total_segs &&
	868	+ ++qp->s_cur == qp->s_size)
	869	+ qp->s_cur = 0;
	870	+ break;
	871	+
506	872	case IB_WR_ATOMIC_CMP_AND_SWP:
507	873	case IB_WR_ATOMIC_FETCH_AND_ADD:
508	874	/*
509	875	* Don't allow more operations to be started
510	876	* than the QP limits allow.
511	877	*/
512		- if (newreq) {
513		- if (qp->s_num_rd_atomic >=
514		- qp->s_max_rd_atomic) {
515		- qp->s_flags \|= RVT_S_WAIT_RDMAR;
516		- goto bail;
517		- }
518		- qp->s_num_rd_atomic++;
519		- if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT))
520		- qp->s_lsn++;
	878	+ if (qp->s_num_rd_atomic >=
	879	+ qp->s_max_rd_atomic) {
	880	+ qp->s_flags \|= RVT_S_WAIT_RDMAR;
	881	+ goto bail;
521	882	}
522		- if (wqe->wr.opcode == IB_WR_ATOMIC_CMP_AND_SWP) {
	883	+ qp->s_num_rd_atomic++;
	884	+ fallthrough;
	885	+ case IB_WR_OPFN:
	886	+ if (newreq && !(qp->s_flags & RVT_S_UNLIMITED_CREDIT))
	887	+ qp->s_lsn++;
	888	+ if (wqe->wr.opcode == IB_WR_ATOMIC_CMP_AND_SWP \|\|
	889	+ wqe->wr.opcode == IB_WR_OPFN) {
523	890	qp->s_state = OP(COMPARE_SWAP);
524	891	put_ib_ateth_swap(wqe->atomic_wr.swap,
525	892	&ohdr->u.atomic_eth);
..	..	@@ -546,18 +913,23 @@
546	913	default:
547	914	goto bail;
548	915	}
549		- qp->s_sge.sge = wqe->sg_list[0];
550		- qp->s_sge.sg_list = wqe->sg_list + 1;
551		- qp->s_sge.num_sge = wqe->wr.num_sge;
552		- qp->s_sge.total_len = wqe->length;
553		- qp->s_len = wqe->length;
	916	+ if (wqe->wr.opcode != IB_WR_TID_RDMA_READ) {
	917	+ qp->s_sge.sge = wqe->sg_list[0];
	918	+ qp->s_sge.sg_list = wqe->sg_list + 1;
	919	+ qp->s_sge.num_sge = wqe->wr.num_sge;
	920	+ qp->s_sge.total_len = wqe->length;
	921	+ qp->s_len = wqe->length;
	922	+ }
554	923	if (newreq) {
555	924	qp->s_tail++;
556	925	if (qp->s_tail >= qp->s_size)
557	926	qp->s_tail = 0;
558	927	}
559		- if (wqe->wr.opcode == IB_WR_RDMA_READ)
	928	+ if (wqe->wr.opcode == IB_WR_RDMA_READ \|\|
	929	+ wqe->wr.opcode == IB_WR_TID_RDMA_WRITE)
560	930	qp->s_psn = wqe->lpsn + 1;
	931	+ else if (wqe->wr.opcode == IB_WR_TID_RDMA_READ)
	932	+ qp->s_psn = req->s_next_psn;
561	933	else
562	934	qp->s_psn++;
563	935	break;
..	..	@@ -573,10 +945,10 @@
573	945	* See restart_rc().
574	946	*/
575	947	qp->s_len = restart_sge(&qp->s_sge, wqe, qp->s_psn, pmtu);
576		- /* FALLTHROUGH */
	948	+ fallthrough;
577	949	case OP(SEND_FIRST):
578	950	qp->s_state = OP(SEND_MIDDLE);
579		- /* FALLTHROUGH */
	951	+ fallthrough;
580	952	case OP(SEND_MIDDLE):
581	953	bth2 = mask_psn(qp->s_psn++);
582	954	ss = &qp->s_sge;
..	..	@@ -618,10 +990,10 @@
618	990	* See restart_rc().
619	991	*/
620	992	qp->s_len = restart_sge(&qp->s_sge, wqe, qp->s_psn, pmtu);
621		- /* FALLTHROUGH */
	993	+ fallthrough;
622	994	case OP(RDMA_WRITE_FIRST):
623	995	qp->s_state = OP(RDMA_WRITE_MIDDLE);
624		- /* FALLTHROUGH */
	996	+ fallthrough;
625	997	case OP(RDMA_WRITE_MIDDLE):
626	998	bth2 = mask_psn(qp->s_psn++);
627	999	ss = &qp->s_sge;
..	..	@@ -674,10 +1046,137 @@
674	1046	if (qp->s_cur == qp->s_size)
675	1047	qp->s_cur = 0;
676	1048	break;
	1049	+
	1050	+ case TID_OP(WRITE_RESP):
	1051	+ /*
	1052	+ * This value for s_state is used for restarting a TID RDMA
	1053	+ * WRITE request. See comment in OP(RDMA_READ_RESPONSE_MIDDLE
	1054	+ * for more).
	1055	+ */
	1056	+ req = wqe_to_tid_req(wqe);
	1057	+ req->state = TID_REQUEST_RESEND;
	1058	+ rcu_read_lock();
	1059	+ remote = rcu_dereference(priv->tid_rdma.remote);
	1060	+ req->comp_seg = delta_psn(qp->s_psn, wqe->psn);
	1061	+ len = wqe->length - (req->comp_seg * remote->max_len);
	1062	+ rcu_read_unlock();
	1063	+
	1064	+ bth2 = mask_psn(qp->s_psn);
	1065	+ hwords += hfi1_build_tid_rdma_write_req(qp, wqe, ohdr, &bth1,
	1066	+ &bth2, &len);
	1067	+ qp->s_psn = wqe->lpsn + 1;
	1068	+ ss = NULL;
	1069	+ qp->s_state = TID_OP(WRITE_REQ);
	1070	+ priv->pending_tid_w_resp += delta_psn(wqe->lpsn, bth2) + 1;
	1071	+ priv->s_tid_cur = qp->s_cur;
	1072	+ if (++qp->s_cur == qp->s_size)
	1073	+ qp->s_cur = 0;
	1074	+ trace_hfi1_tid_req_make_req_write(qp, 0, wqe->wr.opcode,
	1075	+ wqe->psn, wqe->lpsn, req);
	1076	+ break;
	1077	+
	1078	+ case TID_OP(READ_RESP):
	1079	+ if (wqe->wr.opcode != IB_WR_TID_RDMA_READ)
	1080	+ goto bail;
	1081	+ /* This is used to restart a TID read request */
	1082	+ req = wqe_to_tid_req(wqe);
	1083	+ wpriv = wqe->priv;
	1084	+ /*
	1085	+ * Back down. The field qp->s_psn has been set to the psn with
	1086	+ * which the request should be restart. It's OK to use division
	1087	+ * as this is on the retry path.
	1088	+ */
	1089	+ req->cur_seg = delta_psn(qp->s_psn, wqe->psn) / priv->pkts_ps;
	1090	+
	1091	+ /*
	1092	+ * The following function need to be redefined to return the
	1093	+ * status to make sure that we find the flow. At the same
	1094	+ * time, we can use the req->state change to check if the
	1095	+ * call succeeds or not.
	1096	+ */
	1097	+ req->state = TID_REQUEST_RESEND;
	1098	+ hfi1_tid_rdma_restart_req(qp, wqe, &bth2);
	1099	+ if (req->state != TID_REQUEST_ACTIVE) {
	1100	+ /*
	1101	+ * Failed to find the flow. Release all allocated tid
	1102	+ * resources.
	1103	+ */
	1104	+ hfi1_kern_exp_rcv_clear_all(req);
	1105	+ hfi1_kern_clear_hw_flow(priv->rcd, qp);
	1106	+
	1107	+ hfi1_trdma_send_complete(qp, wqe, IB_WC_LOC_QP_OP_ERR);
	1108	+ goto bail;
	1109	+ }
	1110	+ req->state = TID_REQUEST_RESEND;
	1111	+ len = min_t(u32, req->seg_len,
	1112	+ wqe->length - req->seg_len * req->cur_seg);
	1113	+ flow = &req->flows[req->flow_idx];
	1114	+ len -= flow->sent;
	1115	+ req->s_next_psn = flow->flow_state.ib_lpsn + 1;
	1116	+ delta = hfi1_build_tid_rdma_read_packet(wqe, ohdr, &bth1,
	1117	+ &bth2, &len);
	1118	+ if (delta <= 0) {
	1119	+ /* Wait for TID space */
	1120	+ goto bail;
	1121	+ }
	1122	+ hwords += delta;
	1123	+ ss = &wpriv->ss;
	1124	+ /* Check if this is the last segment */
	1125	+ if (req->cur_seg >= req->total_segs &&
	1126	+ ++qp->s_cur == qp->s_size)
	1127	+ qp->s_cur = 0;
	1128	+ qp->s_psn = req->s_next_psn;
	1129	+ trace_hfi1_tid_req_make_req_read(qp, 0, wqe->wr.opcode,
	1130	+ wqe->psn, wqe->lpsn, req);
	1131	+ break;
	1132	+ case TID_OP(READ_REQ):
	1133	+ req = wqe_to_tid_req(wqe);
	1134	+ delta = cmp_psn(qp->s_psn, wqe->psn);
	1135	+ /*
	1136	+ * If the current WR is not TID RDMA READ, or this is the start
	1137	+ * of a new request, we need to change the qp->s_state so that
	1138	+ * the request can be set up properly.
	1139	+ */
	1140	+ if (wqe->wr.opcode != IB_WR_TID_RDMA_READ \|\| delta == 0 \|\|
	1141	+ qp->s_cur == qp->s_tail) {
	1142	+ qp->s_state = OP(RDMA_READ_REQUEST);
	1143	+ if (delta == 0 \|\| qp->s_cur == qp->s_tail)
	1144	+ goto check_s_state;
	1145	+ else
	1146	+ goto bail;
	1147	+ }
	1148	+
	1149	+ /* Rate limiting */
	1150	+ if (qp->s_num_rd_atomic >= qp->s_max_rd_atomic) {
	1151	+ qp->s_flags \|= RVT_S_WAIT_RDMAR;
	1152	+ goto bail;
	1153	+ }
	1154	+
	1155	+ wpriv = wqe->priv;
	1156	+ /* Read one segment at a time */
	1157	+ len = min_t(u32, req->seg_len,
	1158	+ wqe->length - req->seg_len * req->cur_seg);
	1159	+ delta = hfi1_build_tid_rdma_read_req(qp, wqe, ohdr, &bth1,
	1160	+ &bth2, &len);
	1161	+ if (delta <= 0) {
	1162	+ /* Wait for TID space */
	1163	+ goto bail;
	1164	+ }
	1165	+ hwords += delta;
	1166	+ ss = &wpriv->ss;
	1167	+ /* Check if this is the last segment */
	1168	+ if (req->cur_seg >= req->total_segs &&
	1169	+ ++qp->s_cur == qp->s_size)
	1170	+ qp->s_cur = 0;
	1171	+ qp->s_psn = req->s_next_psn;
	1172	+ trace_hfi1_tid_req_make_req_read(qp, 0, wqe->wr.opcode,
	1173	+ wqe->psn, wqe->lpsn, req);
	1174	+ break;
677	1175	}
678	1176	qp->s_sending_hpsn = bth2;
679	1177	delta = delta_psn(bth2, wqe->psn);
680		- if (delta && delta % HFI1_PSN_CREDIT == 0)
	1178	+ if (delta && delta % HFI1_PSN_CREDIT == 0 &&
	1179	+ wqe->wr.opcode != IB_WR_TID_RDMA_WRITE)
681	1180	bth2 \|= IB_BTH_REQ_ACK;
682	1181	if (qp->s_flags & RVT_S_SEND_ONE) {
683	1182	qp->s_flags &= ~RVT_S_SEND_ONE;
..	..	@@ -693,6 +1192,7 @@
693	1192	qp,
694	1193	ohdr,
695	1194	bth0 \| (qp->s_state << 24),
	1195	+ bth1,
696	1196	bth2,
697	1197	middle,
698	1198	ps);
..	..	@@ -709,6 +1209,12 @@
709	1209	bail_no_tx:
710	1210	ps->s_txreq = NULL;
711	1211	qp->s_flags &= ~RVT_S_BUSY;
	1212	+ /*
	1213	+ * If we didn't get a txreq, the QP will be woken up later to try
	1214	+ * again. Set the flags to indicate which work item to wake
	1215	+ * up.
	1216	+ */
	1217	+ iowait_set_flag(&priv->s_iowait, IOWAIT_PENDING_IB);
712	1218	return 0;
713	1219	}
714	1220
..	..	@@ -796,6 +1302,11 @@
796	1302	if (qp->s_mig_state == IB_MIG_MIGRATED)
797	1303	bth0 \|= IB_BTH_MIG_REQ;
798	1304	bth1 = (!!is_fecn) << IB_BECN_SHIFT;
	1305	+ /*
	1306	+ * Inline ACKs go out without the use of the Verbs send engine, so
	1307	+ * we need to set the STL Verbs Extended bit here
	1308	+ */
	1309	+ bth1 \|= HFI1_CAP_IS_KSET(OPFN) << IB_BTHE_E_SHIFT;
799	1310	hfi1_make_bth_aeth(qp, ohdr, bth0, bth1);
800	1311	}
801	1312
..	..	@@ -936,6 +1447,48 @@
936	1447	}
937	1448
938	1449	/**
	1450	+ * update_num_rd_atomic - update the qp->s_num_rd_atomic
	1451	+ * @qp: the QP
	1452	+ * @psn: the packet sequence number to restart at
	1453	+ * @wqe: the wqe
	1454	+ *
	1455	+ * This is called from reset_psn() to update qp->s_num_rd_atomic
	1456	+ * for the current wqe.
	1457	+ * Called at interrupt level with the QP s_lock held.
	1458	+ */
	1459	+static void update_num_rd_atomic(struct rvt_qp *qp, u32 psn,
	1460	+ struct rvt_swqe *wqe)
	1461	+{
	1462	+ u32 opcode = wqe->wr.opcode;
	1463	+
	1464	+ if (opcode == IB_WR_RDMA_READ \|\|
	1465	+ opcode == IB_WR_ATOMIC_CMP_AND_SWP \|\|
	1466	+ opcode == IB_WR_ATOMIC_FETCH_AND_ADD) {
	1467	+ qp->s_num_rd_atomic++;
	1468	+ } else if (opcode == IB_WR_TID_RDMA_READ) {
	1469	+ struct tid_rdma_request *req = wqe_to_tid_req(wqe);
	1470	+ struct hfi1_qp_priv *priv = qp->priv;
	1471	+
	1472	+ if (cmp_psn(psn, wqe->lpsn) <= 0) {
	1473	+ u32 cur_seg;
	1474	+
	1475	+ cur_seg = (psn - wqe->psn) / priv->pkts_ps;
	1476	+ req->ack_pending = cur_seg - req->comp_seg;
	1477	+ priv->pending_tid_r_segs += req->ack_pending;
	1478	+ qp->s_num_rd_atomic += req->ack_pending;
	1479	+ trace_hfi1_tid_req_update_num_rd_atomic(qp, 0,
	1480	+ wqe->wr.opcode,
	1481	+ wqe->psn,
	1482	+ wqe->lpsn,
	1483	+ req);
	1484	+ } else {
	1485	+ priv->pending_tid_r_segs += req->total_segs;
	1486	+ qp->s_num_rd_atomic += req->total_segs;
	1487	+ }
	1488	+ }
	1489	+}
	1490	+
	1491	+/**
939	1492	* reset_psn - reset the QP state to send starting from PSN
940	1493	* @qp: the QP
941	1494	* @psn: the packet sequence number to restart at
..	..	@@ -949,9 +1502,13 @@
949	1502	u32 n = qp->s_acked;
950	1503	struct rvt_swqe *wqe = rvt_get_swqe_ptr(qp, n);
951	1504	u32 opcode;
	1505	+ struct hfi1_qp_priv *priv = qp->priv;
952	1506
953	1507	lockdep_assert_held(&qp->s_lock);
954	1508	qp->s_cur = n;
	1509	+ priv->pending_tid_r_segs = 0;
	1510	+ priv->pending_tid_w_resp = 0;
	1511	+ qp->s_num_rd_atomic = 0;
955	1512
956	1513	/*
957	1514	* If we are starting the request from the beginning,
..	..	@@ -961,9 +1518,9 @@
961	1518	qp->s_state = OP(SEND_LAST);
962	1519	goto done;
963	1520	}
	1521	+ update_num_rd_atomic(qp, psn, wqe);
964	1522
965	1523	/* Find the work request opcode corresponding to the given PSN. */
966		- opcode = wqe->wr.opcode;
967	1524	for (;;) {
968	1525	int diff;
969	1526
..	..	@@ -973,8 +1530,11 @@
973	1530	break;
974	1531	wqe = rvt_get_swqe_ptr(qp, n);
975	1532	diff = cmp_psn(psn, wqe->psn);
976		- if (diff < 0)
	1533	+ if (diff < 0) {
	1534	+ /* Point wqe back to the previous one*/
	1535	+ wqe = rvt_get_swqe_ptr(qp, qp->s_cur);
977	1536	break;
	1537	+ }
978	1538	qp->s_cur = n;
979	1539	/*
980	1540	* If we are starting the request from the beginning,
..	..	@@ -984,8 +1544,10 @@
984	1544	qp->s_state = OP(SEND_LAST);
985	1545	goto done;
986	1546	}
987		- opcode = wqe->wr.opcode;
	1547	+
	1548	+ update_num_rd_atomic(qp, psn, wqe);
988	1549	}
	1550	+ opcode = wqe->wr.opcode;
989	1551
990	1552	/*
991	1553	* Set the state to restart in the middle of a request.
..	..	@@ -1003,8 +1565,16 @@
1003	1565	qp->s_state = OP(RDMA_READ_RESPONSE_LAST);
1004	1566	break;
1005	1567
	1568	+ case IB_WR_TID_RDMA_WRITE:
	1569	+ qp->s_state = TID_OP(WRITE_RESP);
	1570	+ break;
	1571	+
1006	1572	case IB_WR_RDMA_READ:
1007	1573	qp->s_state = OP(RDMA_READ_RESPONSE_MIDDLE);
	1574	+ break;
	1575	+
	1576	+ case IB_WR_TID_RDMA_READ:
	1577	+ qp->s_state = TID_OP(READ_RESP);
1008	1578	break;
1009	1579
1010	1580	default:
..	..	@@ -1015,6 +1585,7 @@
1015	1585	qp->s_state = OP(SEND_LAST);
1016	1586	}
1017	1587	done:
	1588	+ priv->s_flags &= ~HFI1_S_TID_WAIT_INTERLCK;
1018	1589	qp->s_psn = psn;
1019	1590	/*
1020	1591	* Set RVT_S_WAIT_PSN as rc_complete() may start the timer
..	..	@@ -1025,6 +1596,7 @@
1025	1596	(cmp_psn(qp->s_sending_psn, qp->s_sending_hpsn) <= 0))
1026	1597	qp->s_flags \|= RVT_S_WAIT_PSN;
1027	1598	qp->s_flags &= ~HFI1_S_AHG_VALID;
	1599	+ trace_hfi1_sender_reset_psn(qp);
1028	1600	}
1029	1601
1030	1602	/*
..	..	@@ -1033,18 +1605,47 @@
1033	1605	*/
1034	1606	void hfi1_restart_rc(struct rvt_qp *qp, u32 psn, int wait)
1035	1607	{
	1608	+ struct hfi1_qp_priv *priv = qp->priv;
1036	1609	struct rvt_swqe *wqe = rvt_get_swqe_ptr(qp, qp->s_acked);
1037	1610	struct hfi1_ibport *ibp;
1038	1611
1039	1612	lockdep_assert_held(&qp->r_lock);
1040	1613	lockdep_assert_held(&qp->s_lock);
	1614	+ trace_hfi1_sender_restart_rc(qp);
1041	1615	if (qp->s_retry == 0) {
1042	1616	if (qp->s_mig_state == IB_MIG_ARMED) {
1043	1617	hfi1_migrate_qp(qp);
1044	1618	qp->s_retry = qp->s_retry_cnt;
1045	1619	} else if (qp->s_last == qp->s_acked) {
1046		- hfi1_send_complete(qp, wqe, IB_WC_RETRY_EXC_ERR);
1047		- rvt_error_qp(qp, IB_WC_WR_FLUSH_ERR);
	1620	+ /*
	1621	+ * We need special handling for the OPFN request WQEs as
	1622	+ * they are not allowed to generate real user errors
	1623	+ */
	1624	+ if (wqe->wr.opcode == IB_WR_OPFN) {
	1625	+ struct hfi1_ibport *ibp =
	1626	+ to_iport(qp->ibqp.device, qp->port_num);
	1627	+ /*
	1628	+ * Call opfn_conn_reply() with capcode and
	1629	+ * remaining data as 0 to close out the
	1630	+ * current request
	1631	+ */
	1632	+ opfn_conn_reply(qp, priv->opfn.curr);
	1633	+ wqe = do_rc_completion(qp, wqe, ibp);
	1634	+ qp->s_flags &= ~RVT_S_WAIT_ACK;
	1635	+ } else {
	1636	+ trace_hfi1_tid_write_sender_restart_rc(qp, 0);
	1637	+ if (wqe->wr.opcode == IB_WR_TID_RDMA_READ) {
	1638	+ struct tid_rdma_request *req;
	1639	+
	1640	+ req = wqe_to_tid_req(wqe);
	1641	+ hfi1_kern_exp_rcv_clear_all(req);
	1642	+ hfi1_kern_clear_hw_flow(priv->rcd, qp);
	1643	+ }
	1644	+
	1645	+ hfi1_trdma_send_complete(qp, wqe,
	1646	+ IB_WC_RETRY_EXC_ERR);
	1647	+ rvt_error_qp(qp, IB_WC_WR_FLUSH_ERR);
	1648	+ }
1048	1649	return;
1049	1650	} else { /* need to handle delayed completion */
1050	1651	return;
..	..	@@ -1054,14 +1655,15 @@
1054	1655	}
1055	1656
1056	1657	ibp = to_iport(qp->ibqp.device, qp->port_num);
1057		- if (wqe->wr.opcode == IB_WR_RDMA_READ)
	1658	+ if (wqe->wr.opcode == IB_WR_RDMA_READ \|\|
	1659	+ wqe->wr.opcode == IB_WR_TID_RDMA_READ)
1058	1660	ibp->rvp.n_rc_resends++;
1059	1661	else
1060	1662	ibp->rvp.n_rc_resends += delta_psn(qp->s_psn, psn);
1061	1663
1062	1664	qp->s_flags &= ~(RVT_S_WAIT_FENCE \| RVT_S_WAIT_RDMAR \|
1063	1665	RVT_S_WAIT_SSN_CREDIT \| RVT_S_WAIT_PSN \|
1064		- RVT_S_WAIT_ACK);
	1666	+ RVT_S_WAIT_ACK \| HFI1_S_WAIT_TID_RESP);
1065	1667	if (wait)
1066	1668	qp->s_flags \|= RVT_S_SEND_ONE;
1067	1669	reset_psn(qp, psn);
..	..	@@ -1069,7 +1671,8 @@
1069	1671
1070	1672	/*
1071	1673	* Set qp->s_sending_psn to the next PSN after the given one.
1072		- * This would be psn+1 except when RDMA reads are present.
	1674	+ * This would be psn+1 except when RDMA reads or TID RDMA ops
	1675	+ * are present.
1073	1676	*/
1074	1677	static void reset_sending_psn(struct rvt_qp *qp, u32 psn)
1075	1678	{
..	..	@@ -1081,7 +1684,9 @@
1081	1684	for (;;) {
1082	1685	wqe = rvt_get_swqe_ptr(qp, n);
1083	1686	if (cmp_psn(psn, wqe->lpsn) <= 0) {
1084		- if (wqe->wr.opcode == IB_WR_RDMA_READ)
	1687	+ if (wqe->wr.opcode == IB_WR_RDMA_READ \|\|
	1688	+ wqe->wr.opcode == IB_WR_TID_RDMA_READ \|\|
	1689	+ wqe->wr.opcode == IB_WR_TID_RDMA_WRITE)
1085	1690	qp->s_sending_psn = wqe->lpsn + 1;
1086	1691	else
1087	1692	qp->s_sending_psn = psn + 1;
..	..	@@ -1094,6 +1699,36 @@
1094	1699	}
1095	1700	}
1096	1701
	1702	+/**
	1703	+ * hfi1_rc_verbs_aborted - handle abort status
	1704	+ * @qp: the QP
	1705	+ * @opah: the opa header
	1706	+ *
	1707	+ * This code modifies both ACK bit in BTH[2]
	1708	+ * and the s_flags to go into send one mode.
	1709	+ *
	1710	+ * This serves to throttle the send engine to only
	1711	+ * send a single packet in the likely case the
	1712	+ * a link has gone down.
	1713	+ */
	1714	+void hfi1_rc_verbs_aborted(struct rvt_qp qp, struct hfi1_opa_header opah)
	1715	+{
	1716	+ struct ib_other_headers *ohdr = hfi1_get_rc_ohdr(opah);
	1717	+ u8 opcode = ib_bth_get_opcode(ohdr);
	1718	+ u32 psn;
	1719	+
	1720	+ /* ignore responses */
	1721	+ if ((opcode >= OP(RDMA_READ_RESPONSE_FIRST) &&
	1722	+ opcode <= OP(ATOMIC_ACKNOWLEDGE)) \|\|
	1723	+ opcode == TID_OP(READ_RESP) \|\|
	1724	+ opcode == TID_OP(WRITE_RESP))
	1725	+ return;
	1726	+
	1727	+ psn = ib_bth_get_psn(ohdr) \| IB_BTH_REQ_ACK;
	1728	+ ohdr->bth[2] = cpu_to_be32(psn);
	1729	+ qp->s_flags \|= RVT_S_SEND_ONE;
	1730	+}
	1731	+
1097	1732	/*
1098	1733	* This should be called with the QP s_lock held and interrupts disabled.
1099	1734	*/
..	..	@@ -1102,71 +1737,104 @@
1102	1737	struct ib_other_headers *ohdr;
1103	1738	struct hfi1_qp_priv *priv = qp->priv;
1104	1739	struct rvt_swqe *wqe;
1105		- struct ib_header *hdr = NULL;
1106		- struct hfi1_16b_header *hdr_16b = NULL;
1107		- u32 opcode;
	1740	+ u32 opcode, head, tail;
1108	1741	u32 psn;
	1742	+ struct tid_rdma_request *req;
1109	1743
1110	1744	lockdep_assert_held(&qp->s_lock);
1111	1745	if (!(ib_rvt_state_ops[qp->state] & RVT_SEND_OR_FLUSH_OR_RECV_OK))
1112	1746	return;
1113	1747
1114		- /* Find out where the BTH is */
1115		- if (priv->hdr_type == HFI1_PKT_TYPE_9B) {
1116		- hdr = &opah->ibh;
1117		- if (ib_get_lnh(hdr) == HFI1_LRH_BTH)
1118		- ohdr = &hdr->u.oth;
1119		- else
1120		- ohdr = &hdr->u.l.oth;
1121		- } else {
1122		- u8 l4;
1123		-
1124		- hdr_16b = &opah->opah;
1125		- l4 = hfi1_16B_get_l4(hdr_16b);
1126		- if (l4 == OPA_16B_L4_IB_LOCAL)
1127		- ohdr = &hdr_16b->u.oth;
1128		- else
1129		- ohdr = &hdr_16b->u.l.oth;
1130		- }
1131		-
	1748	+ ohdr = hfi1_get_rc_ohdr(opah);
1132	1749	opcode = ib_bth_get_opcode(ohdr);
1133		- if (opcode >= OP(RDMA_READ_RESPONSE_FIRST) &&
1134		- opcode <= OP(ATOMIC_ACKNOWLEDGE)) {
	1750	+ if ((opcode >= OP(RDMA_READ_RESPONSE_FIRST) &&
	1751	+ opcode <= OP(ATOMIC_ACKNOWLEDGE)) \|\|
	1752	+ opcode == TID_OP(READ_RESP) \|\|
	1753	+ opcode == TID_OP(WRITE_RESP)) {
1135	1754	WARN_ON(!qp->s_rdma_ack_cnt);
1136	1755	qp->s_rdma_ack_cnt--;
1137	1756	return;
1138	1757	}
1139	1758
1140	1759	psn = ib_bth_get_psn(ohdr);
1141		- reset_sending_psn(qp, psn);
	1760	+ /*
	1761	+ * Don't attempt to reset the sending PSN for packets in the
	1762	+ * KDETH PSN space since the PSN does not match anything.
	1763	+ */
	1764	+ if (opcode != TID_OP(WRITE_DATA) &&
	1765	+ opcode != TID_OP(WRITE_DATA_LAST) &&
	1766	+ opcode != TID_OP(ACK) && opcode != TID_OP(RESYNC))
	1767	+ reset_sending_psn(qp, psn);
	1768	+
	1769	+ /* Handle TID RDMA WRITE packets differently */
	1770	+ if (opcode >= TID_OP(WRITE_REQ) &&
	1771	+ opcode <= TID_OP(WRITE_DATA_LAST)) {
	1772	+ head = priv->s_tid_head;
	1773	+ tail = priv->s_tid_cur;
	1774	+ /*
	1775	+ * s_tid_cur is set to s_tid_head in the case, where
	1776	+ * a new TID RDMA request is being started and all
	1777	+ * previous ones have been completed.
	1778	+ * Therefore, we need to do a secondary check in order
	1779	+ * to properly determine whether we should start the
	1780	+ * RC timer.
	1781	+ */
	1782	+ wqe = rvt_get_swqe_ptr(qp, tail);
	1783	+ req = wqe_to_tid_req(wqe);
	1784	+ if (head == tail && req->comp_seg < req->total_segs) {
	1785	+ if (tail == 0)
	1786	+ tail = qp->s_size - 1;
	1787	+ else
	1788	+ tail -= 1;
	1789	+ }
	1790	+ } else {
	1791	+ head = qp->s_tail;
	1792	+ tail = qp->s_acked;
	1793	+ }
1142	1794
1143	1795	/*
1144	1796	* Start timer after a packet requesting an ACK has been sent and
1145	1797	* there are still requests that haven't been acked.
1146	1798	*/
1147		- if ((psn & IB_BTH_REQ_ACK) && qp->s_acked != qp->s_tail &&
	1799	+ if ((psn & IB_BTH_REQ_ACK) && tail != head &&
	1800	+ opcode != TID_OP(WRITE_DATA) && opcode != TID_OP(WRITE_DATA_LAST) &&
	1801	+ opcode != TID_OP(RESYNC) &&
1148	1802	!(qp->s_flags &
1149		- (RVT_S_TIMER \| RVT_S_WAIT_RNR \| RVT_S_WAIT_PSN)) &&
1150		- (ib_rvt_state_ops[qp->state] & RVT_PROCESS_RECV_OK))
1151		- rvt_add_retry_timer(qp);
	1803	+ (RVT_S_TIMER \| RVT_S_WAIT_RNR \| RVT_S_WAIT_PSN)) &&
	1804	+ (ib_rvt_state_ops[qp->state] & RVT_PROCESS_RECV_OK)) {
	1805	+ if (opcode == TID_OP(READ_REQ))
	1806	+ rvt_add_retry_timer_ext(qp, priv->timeout_shift);
	1807	+ else
	1808	+ rvt_add_retry_timer(qp);
	1809	+ }
	1810	+
	1811	+ /* Start TID RDMA ACK timer */
	1812	+ if ((opcode == TID_OP(WRITE_DATA) \|\|
	1813	+ opcode == TID_OP(WRITE_DATA_LAST) \|\|
	1814	+ opcode == TID_OP(RESYNC)) &&
	1815	+ (psn & IB_BTH_REQ_ACK) &&
	1816	+ !(priv->s_flags & HFI1_S_TID_RETRY_TIMER) &&
	1817	+ (ib_rvt_state_ops[qp->state] & RVT_PROCESS_RECV_OK)) {
	1818	+ /*
	1819	+ * The TID RDMA ACK packet could be received before this
	1820	+ * function is called. Therefore, add the timer only if TID
	1821	+ * RDMA ACK packets are actually pending.
	1822	+ */
	1823	+ wqe = rvt_get_swqe_ptr(qp, qp->s_acked);
	1824	+ req = wqe_to_tid_req(wqe);
	1825	+ if (wqe->wr.opcode == IB_WR_TID_RDMA_WRITE &&
	1826	+ req->ack_seg < req->cur_seg)
	1827	+ hfi1_add_tid_retry_timer(qp);
	1828	+ }
1152	1829
1153	1830	while (qp->s_last != qp->s_acked) {
1154		- u32 s_last;
1155		-
1156	1831	wqe = rvt_get_swqe_ptr(qp, qp->s_last);
1157	1832	if (cmp_psn(wqe->lpsn, qp->s_sending_psn) >= 0 &&
1158	1833	cmp_psn(qp->s_sending_psn, qp->s_sending_hpsn) <= 0)
1159	1834	break;
1160		- rvt_qp_wqe_unreserve(qp, wqe);
1161		- s_last = qp->s_last;
1162		- trace_hfi1_qp_send_completion(qp, wqe, s_last);
1163		- if (++s_last >= qp->s_size)
1164		- s_last = 0;
1165		- qp->s_last = s_last;
1166		- /* see post_send() */
1167		- barrier();
1168		- rvt_put_swqe(wqe);
1169		- rvt_qp_swqe_complete(qp,
	1835	+ trdma_clean_swqe(qp, wqe);
	1836	+ trace_hfi1_qp_send_completion(qp, wqe, qp->s_last);
	1837	+ rvt_qp_complete_swqe(qp,
1170	1838	wqe,
1171	1839	ib_hfi1_wc_opcode[wqe->wr.opcode],
1172	1840	IB_WC_SUCCESS);
..	..	@@ -1195,30 +1863,24 @@
1195	1863	* This is similar to hfi1_send_complete but has to check to be sure
1196	1864	* that the SGEs are not being referenced if the SWQE is being resent.
1197	1865	*/
1198		-static struct rvt_swqe do_rc_completion(struct rvt_qp qp,
1199		- struct rvt_swqe *wqe,
1200		- struct hfi1_ibport *ibp)
	1866	+struct rvt_swqe do_rc_completion(struct rvt_qp qp,
	1867	+ struct rvt_swqe *wqe,
	1868	+ struct hfi1_ibport *ibp)
1201	1869	{
	1870	+ struct hfi1_qp_priv *priv = qp->priv;
	1871	+
1202	1872	lockdep_assert_held(&qp->s_lock);
1203	1873	/*
1204	1874	* Don't decrement refcount and don't generate a
1205	1875	* completion if the SWQE is being resent until the send
1206	1876	* is finished.
1207	1877	*/
	1878	+ trace_hfi1_rc_completion(qp, wqe->lpsn);
1208	1879	if (cmp_psn(wqe->lpsn, qp->s_sending_psn) < 0 \|\|
1209	1880	cmp_psn(qp->s_sending_psn, qp->s_sending_hpsn) > 0) {
1210		- u32 s_last;
1211		-
1212		- rvt_put_swqe(wqe);
1213		- rvt_qp_wqe_unreserve(qp, wqe);
1214		- s_last = qp->s_last;
1215		- trace_hfi1_qp_send_completion(qp, wqe, s_last);
1216		- if (++s_last >= qp->s_size)
1217		- s_last = 0;
1218		- qp->s_last = s_last;
1219		- /* see post_send() */
1220		- barrier();
1221		- rvt_qp_swqe_complete(qp,
	1881	+ trdma_clean_swqe(qp, wqe);
	1882	+ trace_hfi1_qp_send_completion(qp, wqe, qp->s_last);
	1883	+ rvt_qp_complete_swqe(qp,
1222	1884	wqe,
1223	1885	ib_hfi1_wc_opcode[wqe->wr.opcode],
1224	1886	IB_WC_SUCCESS);
..	..	@@ -1243,7 +1905,16 @@
1243	1905	}
1244	1906
1245	1907	qp->s_retry = qp->s_retry_cnt;
1246		- update_last_psn(qp, wqe->lpsn);
	1908	+ /*
	1909	+ * Don't update the last PSN if the request being completed is
	1910	+ * a TID RDMA WRITE request.
	1911	+ * Completion of the TID RDMA WRITE requests are done by the
	1912	+ * TID RDMA ACKs and as such could be for a request that has
	1913	+ * already been ACKed as far as the IB state machine is
	1914	+ * concerned.
	1915	+ */
	1916	+ if (wqe->wr.opcode != IB_WR_TID_RDMA_WRITE)
	1917	+ update_last_psn(qp, wqe->lpsn);
1247	1918
1248	1919	/*
1249	1920	* If we are completing a request which is in the process of
..	..	@@ -1266,7 +1937,59 @@
1266	1937	qp->s_draining = 0;
1267	1938	wqe = rvt_get_swqe_ptr(qp, qp->s_acked);
1268	1939	}
	1940	+ if (priv->s_flags & HFI1_S_TID_WAIT_INTERLCK) {
	1941	+ priv->s_flags &= ~HFI1_S_TID_WAIT_INTERLCK;
	1942	+ hfi1_schedule_send(qp);
	1943	+ }
1269	1944	return wqe;
	1945	+}
	1946	+
	1947	+static void set_restart_qp(struct rvt_qp qp, struct hfi1_ctxtdata rcd)
	1948	+{
	1949	+ /* Retry this request. */
	1950	+ if (!(qp->r_flags & RVT_R_RDMAR_SEQ)) {
	1951	+ qp->r_flags \|= RVT_R_RDMAR_SEQ;
	1952	+ hfi1_restart_rc(qp, qp->s_last_psn + 1, 0);
	1953	+ if (list_empty(&qp->rspwait)) {
	1954	+ qp->r_flags \|= RVT_R_RSP_SEND;
	1955	+ rvt_get_qp(qp);
	1956	+ list_add_tail(&qp->rspwait, &rcd->qp_wait_list);
	1957	+ }
	1958	+ }
	1959	+}
	1960	+
	1961	+/**
	1962	+ * update_qp_retry_state - Update qp retry state.
	1963	+ * @qp: the QP
	1964	+ * @psn: the packet sequence number of the TID RDMA WRITE RESP.
	1965	+ * @spsn: The start psn for the given TID RDMA WRITE swqe.
	1966	+ * @lpsn: The last psn for the given TID RDMA WRITE swqe.
	1967	+ *
	1968	+ * This function is called to update the qp retry state upon
	1969	+ * receiving a TID WRITE RESP after the qp is scheduled to retry
	1970	+ * a request.
	1971	+ */
	1972	+static void update_qp_retry_state(struct rvt_qp *qp, u32 psn, u32 spsn,
	1973	+ u32 lpsn)
	1974	+{
	1975	+ struct hfi1_qp_priv *qpriv = qp->priv;
	1976	+
	1977	+ qp->s_psn = psn + 1;
	1978	+ /*
	1979	+ * If this is the first TID RDMA WRITE RESP packet for the current
	1980	+ * request, change the s_state so that the retry will be processed
	1981	+ * correctly. Similarly, if this is the last TID RDMA WRITE RESP
	1982	+ * packet, change the s_state and advance the s_cur.
	1983	+ */
	1984	+ if (cmp_psn(psn, lpsn) >= 0) {
	1985	+ qp->s_cur = qpriv->s_tid_cur + 1;
	1986	+ if (qp->s_cur >= qp->s_size)
	1987	+ qp->s_cur = 0;
	1988	+ qp->s_state = TID_OP(WRITE_REQ);
	1989	+ } else if (!cmp_psn(psn, spsn)) {
	1990	+ qp->s_cur = qpriv->s_tid_cur;
	1991	+ qp->s_state = TID_OP(WRITE_RESP);
	1992	+ }
1270	1993	}
1271	1994
1272	1995	/**
..	..	@@ -1280,15 +2003,17 @@
1280	2003	* May be called at interrupt level, with the QP s_lock held.
1281	2004	* Returns 1 if OK, 0 if current operation should be aborted (NAK).
1282	2005	*/
1283		-static int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode,
1284		- u64 val, struct hfi1_ctxtdata *rcd)
	2006	+int do_rc_ack(struct rvt_qp *qp, u32 aeth, u32 psn, int opcode,
	2007	+ u64 val, struct hfi1_ctxtdata *rcd)
1285	2008	{
1286	2009	struct hfi1_ibport *ibp;
1287	2010	enum ib_wc_status status;
	2011	+ struct hfi1_qp_priv *qpriv = qp->priv;
1288	2012	struct rvt_swqe *wqe;
1289	2013	int ret = 0;
1290	2014	u32 ack_psn;
1291	2015	int diff;
	2016	+ struct rvt_dev_info *rdi;
1292	2017
1293	2018	lockdep_assert_held(&qp->s_lock);
1294	2019	/*
..	..	@@ -1331,20 +2056,14 @@
1331	2056	*/
1332	2057	if ((wqe->wr.opcode == IB_WR_RDMA_READ &&
1333	2058	(opcode != OP(RDMA_READ_RESPONSE_LAST) \|\| diff != 0)) \|\|
	2059	+ (wqe->wr.opcode == IB_WR_TID_RDMA_READ &&
	2060	+ (opcode != TID_OP(READ_RESP) \|\| diff != 0)) \|\|
1334	2061	((wqe->wr.opcode == IB_WR_ATOMIC_CMP_AND_SWP \|\|
1335	2062	wqe->wr.opcode == IB_WR_ATOMIC_FETCH_AND_ADD) &&
1336		- (opcode != OP(ATOMIC_ACKNOWLEDGE) \|\| diff != 0))) {
1337		- /* Retry this request. */
1338		- if (!(qp->r_flags & RVT_R_RDMAR_SEQ)) {
1339		- qp->r_flags \|= RVT_R_RDMAR_SEQ;
1340		- hfi1_restart_rc(qp, qp->s_last_psn + 1, 0);
1341		- if (list_empty(&qp->rspwait)) {
1342		- qp->r_flags \|= RVT_R_RSP_SEND;
1343		- rvt_get_qp(qp);
1344		- list_add_tail(&qp->rspwait,
1345		- &rcd->qp_wait_list);
1346		- }
1347		- }
	2063	+ (opcode != OP(ATOMIC_ACKNOWLEDGE) \|\| diff != 0)) \|\|
	2064	+ (wqe->wr.opcode == IB_WR_TID_RDMA_WRITE &&
	2065	+ (delta_psn(psn, qp->s_last_psn) != 1))) {
	2066	+ set_restart_qp(qp, rcd);
1348	2067	/*
1349	2068	* No need to process the ACK/NAK since we are
1350	2069	* restarting an earlier request.
..	..	@@ -1356,6 +2075,9 @@
1356	2075	u64 *vaddr = wqe->sg_list[0].vaddr;
1357	2076	*vaddr = val;
1358	2077	}
	2078	+ if (wqe->wr.opcode == IB_WR_OPFN)
	2079	+ opfn_conn_reply(qp, val);
	2080	+
1359	2081	if (qp->s_num_rd_atomic &&
1360	2082	(wqe->wr.opcode == IB_WR_RDMA_READ \|\|
1361	2083	wqe->wr.opcode == IB_WR_ATOMIC_CMP_AND_SWP \|\|
..	..	@@ -1373,26 +2095,85 @@
1373	2095	hfi1_schedule_send(qp);
1374	2096	}
1375	2097	}
	2098	+
	2099	+ /*
	2100	+ * TID RDMA WRITE requests will be completed by the TID RDMA
	2101	+ * ACK packet handler (see tid_rdma.c).
	2102	+ */
	2103	+ if (wqe->wr.opcode == IB_WR_TID_RDMA_WRITE)
	2104	+ break;
	2105	+
1376	2106	wqe = do_rc_completion(qp, wqe, ibp);
1377	2107	if (qp->s_acked == qp->s_tail)
1378	2108	break;
1379	2109	}
1380	2110
	2111	+ trace_hfi1_rc_ack_do(qp, aeth, psn, wqe);
	2112	+ trace_hfi1_sender_do_rc_ack(qp);
1381	2113	switch (aeth >> IB_AETH_NAK_SHIFT) {
1382	2114	case 0: /* ACK */
1383	2115	this_cpu_inc(*ibp->rvp.rc_acks);
1384		- if (qp->s_acked != qp->s_tail) {
	2116	+ if (wqe->wr.opcode == IB_WR_TID_RDMA_READ) {
	2117	+ if (wqe_to_tid_req(wqe)->ack_pending)
	2118	+ rvt_mod_retry_timer_ext(qp,
	2119	+ qpriv->timeout_shift);
	2120	+ else
	2121	+ rvt_stop_rc_timers(qp);
	2122	+ } else if (qp->s_acked != qp->s_tail) {
	2123	+ struct rvt_swqe *__w = NULL;
	2124	+
	2125	+ if (qpriv->s_tid_cur != HFI1_QP_WQE_INVALID)
	2126	+ __w = rvt_get_swqe_ptr(qp, qpriv->s_tid_cur);
	2127	+
1385	2128	/*
1386		- * We are expecting more ACKs so
1387		- * mod the retry timer.
	2129	+ * Stop timers if we've received all of the TID RDMA
	2130	+ * WRITE * responses.
1388	2131	*/
1389		- rvt_mod_retry_timer(qp);
1390		- /*
1391		- * We can stop re-sending the earlier packets and
1392		- * continue with the next packet the receiver wants.
1393		- */
1394		- if (cmp_psn(qp->s_psn, psn) <= 0)
1395		- reset_psn(qp, psn + 1);
	2132	+ if (__w && __w->wr.opcode == IB_WR_TID_RDMA_WRITE &&
	2133	+ opcode == TID_OP(WRITE_RESP)) {
	2134	+ /*
	2135	+ * Normally, the loop above would correctly
	2136	+ * process all WQEs from s_acked onward and
	2137	+ * either complete them or check for correct
	2138	+ * PSN sequencing.
	2139	+ * However, for TID RDMA, due to pipelining,
	2140	+ * the response may not be for the request at
	2141	+ * s_acked so the above look would just be
	2142	+ * skipped. This does not allow for checking
	2143	+ * the PSN sequencing. It has to be done
	2144	+ * separately.
	2145	+ */
	2146	+ if (cmp_psn(psn, qp->s_last_psn + 1)) {
	2147	+ set_restart_qp(qp, rcd);
	2148	+ goto bail_stop;
	2149	+ }
	2150	+ /*
	2151	+ * If the psn is being resent, stop the
	2152	+ * resending.
	2153	+ */
	2154	+ if (qp->s_cur != qp->s_tail &&
	2155	+ cmp_psn(qp->s_psn, psn) <= 0)
	2156	+ update_qp_retry_state(qp, psn,
	2157	+ __w->psn,
	2158	+ __w->lpsn);
	2159	+ else if (--qpriv->pending_tid_w_resp)
	2160	+ rvt_mod_retry_timer(qp);
	2161	+ else
	2162	+ rvt_stop_rc_timers(qp);
	2163	+ } else {
	2164	+ /*
	2165	+ * We are expecting more ACKs so
	2166	+ * mod the retry timer.
	2167	+ */
	2168	+ rvt_mod_retry_timer(qp);
	2169	+ /*
	2170	+ * We can stop re-sending the earlier packets
	2171	+ * and continue with the next packet the
	2172	+ * receiver wants.
	2173	+ */
	2174	+ if (cmp_psn(qp->s_psn, psn) <= 0)
	2175	+ reset_psn(qp, psn + 1);
	2176	+ }
1396	2177	} else {
1397	2178	/* No more acks - kill all timers */
1398	2179	rvt_stop_rc_timers(qp);
..	..	@@ -1408,6 +2189,15 @@
1408	2189	rvt_get_credit(qp, aeth);
1409	2190	qp->s_rnr_retry = qp->s_rnr_retry_cnt;
1410	2191	qp->s_retry = qp->s_retry_cnt;
	2192	+ /*
	2193	+ * If the current request is a TID RDMA WRITE request and the
	2194	+ * response is not a TID RDMA WRITE RESP packet, s_last_psn
	2195	+ * can't be advanced.
	2196	+ */
	2197	+ if (wqe->wr.opcode == IB_WR_TID_RDMA_WRITE &&
	2198	+ opcode != TID_OP(WRITE_RESP) &&
	2199	+ cmp_psn(psn, wqe->psn) >= 0)
	2200	+ return 1;
1411	2201	update_last_psn(qp, psn);
1412	2202	return 1;
1413	2203
..	..	@@ -1417,20 +2207,31 @@
1417	2207	goto bail_stop;
1418	2208	if (qp->s_flags & RVT_S_WAIT_RNR)
1419	2209	goto bail_stop;
1420		- if (qp->s_rnr_retry == 0) {
1421		- status = IB_WC_RNR_RETRY_EXC_ERR;
1422		- goto class_b;
	2210	+ rdi = ib_to_rvt(qp->ibqp.device);
	2211	+ if (!(rdi->post_parms[wqe->wr.opcode].flags &
	2212	+ RVT_OPERATION_IGN_RNR_CNT)) {
	2213	+ if (qp->s_rnr_retry == 0) {
	2214	+ status = IB_WC_RNR_RETRY_EXC_ERR;
	2215	+ goto class_b;
	2216	+ }
	2217	+ if (qp->s_rnr_retry_cnt < 7 && qp->s_rnr_retry_cnt > 0)
	2218	+ qp->s_rnr_retry--;
1423	2219	}
1424		- if (qp->s_rnr_retry_cnt < 7)
1425		- qp->s_rnr_retry--;
1426	2220
1427		- /* The last valid PSN is the previous PSN. */
1428		- update_last_psn(qp, psn - 1);
	2221	+ /*
	2222	+ * The last valid PSN is the previous PSN. For TID RDMA WRITE
	2223	+ * request, s_last_psn should be incremented only when a TID
	2224	+ * RDMA WRITE RESP is received to avoid skipping lost TID RDMA
	2225	+ * WRITE RESP packets.
	2226	+ */
	2227	+ if (wqe->wr.opcode == IB_WR_TID_RDMA_WRITE) {
	2228	+ reset_psn(qp, qp->s_last_psn + 1);
	2229	+ } else {
	2230	+ update_last_psn(qp, psn - 1);
	2231	+ reset_psn(qp, psn);
	2232	+ }
1429	2233
1430	2234	ibp->rvp.n_rc_resends += delta_psn(qp->s_psn, psn);
1431		-
1432		- reset_psn(qp, psn);
1433		-
1434	2235	qp->s_flags &= ~(RVT_S_WAIT_SSN_CREDIT \| RVT_S_WAIT_ACK);
1435	2236	rvt_stop_rc_timers(qp);
1436	2237	rvt_add_rnr_timer(qp, aeth);
..	..	@@ -1470,7 +2271,10 @@
1470	2271	ibp->rvp.n_other_naks++;
1471	2272	class_b:
1472	2273	if (qp->s_last == qp->s_acked) {
1473		- hfi1_send_complete(qp, wqe, status);
	2274	+ if (wqe->wr.opcode == IB_WR_TID_RDMA_READ)
	2275	+ hfi1_kern_read_tid_flow_free(qp);
	2276	+
	2277	+ hfi1_trdma_send_complete(qp, wqe, status);
1474	2278	rvt_error_qp(qp, IB_WC_WR_FLUSH_ERR);
1475	2279	}
1476	2280	break;
..	..	@@ -1511,6 +2315,8 @@
1511	2315
1512	2316	while (cmp_psn(psn, wqe->lpsn) > 0) {
1513	2317	if (wqe->wr.opcode == IB_WR_RDMA_READ \|\|
	2318	+ wqe->wr.opcode == IB_WR_TID_RDMA_READ \|\|
	2319	+ wqe->wr.opcode == IB_WR_TID_RDMA_WRITE \|\|
1514	2320	wqe->wr.opcode == IB_WR_ATOMIC_CMP_AND_SWP \|\|
1515	2321	wqe->wr.opcode == IB_WR_ATOMIC_FETCH_AND_ADD)
1516	2322	break;
..	..	@@ -1646,7 +2452,8 @@
1646	2452	qp->s_rdma_read_len -= pmtu;
1647	2453	update_last_psn(qp, psn);
1648	2454	spin_unlock_irqrestore(&qp->s_lock, flags);
1649		- hfi1_copy_sge(&qp->s_rdma_read_sge, data, pmtu, false, false);
	2455	+ rvt_copy_sge(qp, &qp->s_rdma_read_sge,
	2456	+ data, pmtu, false, false);
1650	2457	goto bail;
1651	2458
1652	2459	case OP(RDMA_READ_RESPONSE_ONLY):
..	..	@@ -1686,7 +2493,8 @@
1686	2493	if (unlikely(tlen != qp->s_rdma_read_len))
1687	2494	goto ack_len_err;
1688	2495	aeth = be32_to_cpu(ohdr->u.aeth);
1689		- hfi1_copy_sge(&qp->s_rdma_read_sge, data, tlen, false, false);
	2496	+ rvt_copy_sge(qp, &qp->s_rdma_read_sge,
	2497	+ data, tlen, false, false);
1690	2498	WARN_ON(qp->s_rdma_read_sge.num_sge);
1691	2499	(void)do_rc_ack(qp, aeth, psn,
1692	2500	OP(RDMA_READ_RESPONSE_LAST), 0, rcd);
..	..	@@ -1706,23 +2514,13 @@
1706	2514	status = IB_WC_LOC_LEN_ERR;
1707	2515	ack_err:
1708	2516	if (qp->s_last == qp->s_acked) {
1709		- hfi1_send_complete(qp, wqe, status);
	2517	+ rvt_send_complete(qp, wqe, status);
1710	2518	rvt_error_qp(qp, IB_WC_WR_FLUSH_ERR);
1711	2519	}
1712	2520	ack_done:
1713	2521	spin_unlock_irqrestore(&qp->s_lock, flags);
1714	2522	bail:
1715	2523	return;
1716		-}
1717		-
1718		-static inline void rc_defered_ack(struct hfi1_ctxtdata *rcd,
1719		- struct rvt_qp *qp)
1720		-{
1721		- if (list_empty(&qp->rspwait)) {
1722		- qp->r_flags \|= RVT_R_RSP_NAK;
1723		- rvt_get_qp(qp);
1724		- list_add_tail(&qp->rspwait, &rcd->qp_wait_list);
1725		- }
1726	2524	}
1727	2525
1728	2526	static inline void rc_cancel_ack(struct rvt_qp *qp)
..	..	@@ -1757,8 +2555,9 @@
1757	2555	struct hfi1_ibport *ibp = rcd_to_iport(rcd);
1758	2556	struct rvt_ack_entry *e;
1759	2557	unsigned long flags;
1760		- u8 i, prev;
1761		- int old_req;
	2558	+ u8 prev;
	2559	+ u8 mra; /* most recent ACK */
	2560	+ bool old_req;
1762	2561
1763	2562	trace_hfi1_rcv_error(qp, psn);
1764	2563	if (diff > 0) {
..	..	@@ -1799,34 +2598,13 @@
1799	2598	* to be sent before sending this one.
1800	2599	*/
1801	2600	e = NULL;
1802		- old_req = 1;
	2601	+ old_req = true;
1803	2602	ibp->rvp.n_rc_dupreq++;
1804	2603
1805	2604	spin_lock_irqsave(&qp->s_lock, flags);
1806	2605
1807		- for (i = qp->r_head_ack_queue; ; i = prev) {
1808		- if (i == qp->s_tail_ack_queue)
1809		- old_req = 0;
1810		- if (i)
1811		- prev = i - 1;
1812		- else
1813		- prev = HFI1_MAX_RDMA_ATOMIC;
1814		- if (prev == qp->r_head_ack_queue) {
1815		- e = NULL;
1816		- break;
1817		- }
1818		- e = &qp->s_ack_queue[prev];
1819		- if (!e->opcode) {
1820		- e = NULL;
1821		- break;
1822		- }
1823		- if (cmp_psn(psn, e->psn) >= 0) {
1824		- if (prev == qp->s_tail_ack_queue &&
1825		- cmp_psn(psn, e->lpsn) <= 0)
1826		- old_req = 0;
1827		- break;
1828		- }
1829		- }
	2606	+ e = find_prev_entry(qp, psn, &prev, &mra, &old_req);
	2607	+
1830	2608	switch (opcode) {
1831	2609	case OP(RDMA_READ_REQUEST): {
1832	2610	struct ib_reth *reth;
..	..	@@ -1852,10 +2630,7 @@
1852	2630	len = be32_to_cpu(reth->length);
1853	2631	if (unlikely(offset + len != e->rdma_sge.sge_length))
1854	2632	goto unlock_done;
1855		- if (e->rdma_sge.mr) {
1856		- rvt_put_mr(e->rdma_sge.mr);
1857		- e->rdma_sge.mr = NULL;
1858		- }
	2633	+ release_rdma_sge_mr(e);
1859	2634	if (len != 0) {
1860	2635	u32 rkey = be32_to_cpu(reth->rkey);
1861	2636	u64 vaddr = get_ib_reth_vaddr(reth);
..	..	@@ -1873,6 +2648,8 @@
1873	2648	e->psn = psn;
1874	2649	if (old_req)
1875	2650	goto unlock_done;
	2651	+ if (qp->s_acked_ack_queue == qp->s_tail_ack_queue)
	2652	+ qp->s_acked_ack_queue = prev;
1876	2653	qp->s_tail_ack_queue = prev;
1877	2654	break;
1878	2655	}
..	..	@@ -1886,6 +2663,8 @@
1886	2663	*/
1887	2664	if (!e \|\| e->opcode != (u8)opcode \|\| old_req)
1888	2665	goto unlock_done;
	2666	+ if (qp->s_tail_ack_queue == qp->s_acked_ack_queue)
	2667	+ qp->s_acked_ack_queue = prev;
1889	2668	qp->s_tail_ack_queue = prev;
1890	2669	break;
1891	2670	}
..	..	@@ -1901,7 +2680,7 @@
1901	2680	* Resend the most recent ACK if this request is
1902	2681	* after all the previous RDMA reads and atomics.
1903	2682	*/
1904		- if (i == qp->r_head_ack_queue) {
	2683	+ if (mra == qp->r_head_ack_queue) {
1905	2684	spin_unlock_irqrestore(&qp->s_lock, flags);
1906	2685	qp->r_nak_state = 0;
1907	2686	qp->r_ack_psn = qp->r_psn - 1;
..	..	@@ -1912,7 +2691,9 @@
1912	2691	* Resend the RDMA read or atomic op which
1913	2692	* ACKs this duplicate request.
1914	2693	*/
1915		- qp->s_tail_ack_queue = i;
	2694	+ if (qp->s_tail_ack_queue == qp->s_acked_ack_queue)
	2695	+ qp->s_acked_ack_queue = mra;
	2696	+ qp->s_tail_ack_queue = mra;
1916	2697	break;
1917	2698	}
1918	2699	qp->s_ack_state = OP(ACKNOWLEDGE);
..	..	@@ -1927,17 +2708,6 @@
1927	2708
1928	2709	send_ack:
1929	2710	return 0;
1930		-}
1931		-
1932		-static inline void update_ack_queue(struct rvt_qp *qp, unsigned n)
1933		-{
1934		- unsigned next;
1935		-
1936		- next = n + 1;
1937		- if (next > HFI1_MAX_RDMA_ATOMIC)
1938		- next = 0;
1939		- qp->s_tail_ack_queue = next;
1940		- qp->s_ack_state = OP(ACKNOWLEDGE);
1941	2711	}
1942	2712
1943	2713	static void log_cca_event(struct hfi1_pportdata *ppd, u8 sl, u32 rlid,
..	..	@@ -2037,6 +2807,7 @@
2037	2807	void *data = packet->payload;
2038	2808	u32 tlen = packet->tlen;
2039	2809	struct rvt_qp *qp = packet->qp;
	2810	+ struct hfi1_qp_priv *qpriv = qp->priv;
2040	2811	struct hfi1_ibport *ibp = rcd_to_iport(rcd);
2041	2812	struct ib_other_headers *ohdr = packet->ohdr;
2042	2813	u32 opcode = packet->opcode;
..	..	@@ -2059,6 +2830,7 @@
2059	2830	return;
2060	2831
2061	2832	fecn = process_ecn(qp, packet);
	2833	+ opfn_trigger_conn_request(qp, be32_to_cpu(ohdr->bth[1]));
2062	2834
2063	2835	/*
2064	2836	* Process responses (ACKs) before anything else. Note that the
..	..	@@ -2128,7 +2900,7 @@
2128	2900	if (!ret)
2129	2901	goto rnr_nak;
2130	2902	qp->r_rcv_len = 0;
2131		- /* FALLTHROUGH */
	2903	+ fallthrough;
2132	2904	case OP(SEND_MIDDLE):
2133	2905	case OP(RDMA_WRITE_MIDDLE):
2134	2906	send_middle:
..	..	@@ -2143,7 +2915,7 @@
2143	2915	qp->r_rcv_len += pmtu;
2144	2916	if (unlikely(qp->r_rcv_len > qp->r_len))
2145	2917	goto nack_inv;
2146		- hfi1_copy_sge(&qp->r_sge, data, pmtu, true, false);
	2918	+ rvt_copy_sge(qp, &qp->r_sge, data, pmtu, true, false);
2147	2919	break;
2148	2920
2149	2921	case OP(RDMA_WRITE_LAST_WITH_IMMEDIATE):
..	..	@@ -2168,7 +2940,7 @@
2168	2940	goto no_immediate_data;
2169	2941	if (opcode == OP(SEND_ONLY_WITH_INVALIDATE))
2170	2942	goto send_last_inv;
2171		- /* FALLTHROUGH -- for SEND_ONLY_WITH_IMMEDIATE */
	2943	+ fallthrough; /* for SEND_ONLY_WITH_IMMEDIATE */
2172	2944	case OP(SEND_LAST_WITH_IMMEDIATE):
2173	2945	send_last_imm:
2174	2946	wc.ex.imm_data = ohdr->u.imm_data;
..	..	@@ -2184,7 +2956,7 @@
2184	2956	goto send_last;
2185	2957	case OP(RDMA_WRITE_LAST):
2186	2958	copy_last = rvt_is_user_qp(qp);
2187		- /* fall through */
	2959	+ fallthrough;
2188	2960	case OP(SEND_LAST):
2189	2961	no_immediate_data:
2190	2962	wc.wc_flags = 0;
..	..	@@ -2199,7 +2971,7 @@
2199	2971	wc.byte_len = tlen + qp->r_rcv_len;
2200	2972	if (unlikely(wc.byte_len > qp->r_len))
2201	2973	goto nack_inv;
2202		- hfi1_copy_sge(&qp->r_sge, data, tlen, true, copy_last);
	2974	+ rvt_copy_sge(qp, &qp->r_sge, data, tlen, true, copy_last);
2203	2975	rvt_put_ss(&qp->r_sge);
2204	2976	qp->r_msn++;
2205	2977	if (!__test_and_clear_bit(RVT_R_WRID_VALID, &qp->r_aflags))
..	..	@@ -2232,13 +3004,12 @@
2232	3004	wc.dlid_path_bits = 0;
2233	3005	wc.port_num = 0;
2234	3006	/* Signal completion event if the solicited bit is set. */
2235		- rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.recv_cq), &wc,
2236		- ib_bth_is_solicited(ohdr));
	3007	+ rvt_recv_cq(qp, &wc, ib_bth_is_solicited(ohdr));
2237	3008	break;
2238	3009
2239	3010	case OP(RDMA_WRITE_ONLY):
2240	3011	copy_last = rvt_is_user_qp(qp);
2241		- /* fall through */
	3012	+ fallthrough;
2242	3013	case OP(RDMA_WRITE_FIRST):
2243	3014	case OP(RDMA_WRITE_ONLY_WITH_IMMEDIATE):
2244	3015	if (unlikely(!(qp->qp_access_flags & IB_ACCESS_REMOTE_WRITE)))
..	..	@@ -2290,20 +3061,17 @@
2290	3061	if (unlikely(!(qp->qp_access_flags & IB_ACCESS_REMOTE_READ)))
2291	3062	goto nack_inv;
2292	3063	next = qp->r_head_ack_queue + 1;
2293		- /* s_ack_queue is size HFI1_MAX_RDMA_ATOMIC+1 so use > not >= */
2294		- if (next > HFI1_MAX_RDMA_ATOMIC)
	3064	+ /* s_ack_queue is size rvt_size_atomic()+1 so use > not >= */
	3065	+ if (next > rvt_size_atomic(ib_to_rvt(qp->ibqp.device)))
2295	3066	next = 0;
2296	3067	spin_lock_irqsave(&qp->s_lock, flags);
2297		- if (unlikely(next == qp->s_tail_ack_queue)) {
	3068	+ if (unlikely(next == qp->s_acked_ack_queue)) {
2298	3069	if (!qp->s_ack_queue[next].sent)
2299	3070	goto nack_inv_unlck;
2300	3071	update_ack_queue(qp, next);
2301	3072	}
2302	3073	e = &qp->s_ack_queue[qp->r_head_ack_queue];
2303		- if (e->rdma_sge.mr) {
2304		- rvt_put_mr(e->rdma_sge.mr);
2305		- e->rdma_sge.mr = NULL;
2306		- }
	3074	+ release_rdma_sge_mr(e);
2307	3075	reth = &ohdr->u.rc.reth;
2308	3076	len = be32_to_cpu(reth->length);
2309	3077	if (len) {
..	..	@@ -2341,6 +3109,7 @@
2341	3109	qp->r_state = opcode;
2342	3110	qp->r_nak_state = 0;
2343	3111	qp->r_head_ack_queue = next;
	3112	+ qpriv->r_tid_alloc = qp->r_head_ack_queue;
2344	3113
2345	3114	/* Schedule the send engine. */
2346	3115	qp->s_flags \|= RVT_S_RESP_PENDING;
..	..	@@ -2354,32 +3123,35 @@
2354	3123
2355	3124	case OP(COMPARE_SWAP):
2356	3125	case OP(FETCH_ADD): {
2357		- struct ib_atomic_eth *ateth;
	3126	+ struct ib_atomic_eth *ateth = &ohdr->u.atomic_eth;
	3127	+ u64 vaddr = get_ib_ateth_vaddr(ateth);
	3128	+ bool opfn = opcode == OP(COMPARE_SWAP) &&
	3129	+ vaddr == HFI1_VERBS_E_ATOMIC_VADDR;
2358	3130	struct rvt_ack_entry *e;
2359		- u64 vaddr;
2360	3131	atomic64_t *maddr;
2361	3132	u64 sdata;
2362	3133	u32 rkey;
2363	3134	u8 next;
2364	3135
2365		- if (unlikely(!(qp->qp_access_flags & IB_ACCESS_REMOTE_ATOMIC)))
	3136	+ if (unlikely(!(qp->qp_access_flags & IB_ACCESS_REMOTE_ATOMIC) &&
	3137	+ !opfn))
2366	3138	goto nack_inv;
2367	3139	next = qp->r_head_ack_queue + 1;
2368		- if (next > HFI1_MAX_RDMA_ATOMIC)
	3140	+ if (next > rvt_size_atomic(ib_to_rvt(qp->ibqp.device)))
2369	3141	next = 0;
2370	3142	spin_lock_irqsave(&qp->s_lock, flags);
2371		- if (unlikely(next == qp->s_tail_ack_queue)) {
	3143	+ if (unlikely(next == qp->s_acked_ack_queue)) {
2372	3144	if (!qp->s_ack_queue[next].sent)
2373	3145	goto nack_inv_unlck;
2374	3146	update_ack_queue(qp, next);
2375	3147	}
2376	3148	e = &qp->s_ack_queue[qp->r_head_ack_queue];
2377		- if (e->rdma_sge.mr) {
2378		- rvt_put_mr(e->rdma_sge.mr);
2379		- e->rdma_sge.mr = NULL;
	3149	+ release_rdma_sge_mr(e);
	3150	+ /* Process OPFN special virtual address */
	3151	+ if (opfn) {
	3152	+ opfn_conn_response(qp, e, ateth);
	3153	+ goto ack;
2380	3154	}
2381		- ateth = &ohdr->u.atomic_eth;
2382		- vaddr = get_ib_ateth_vaddr(ateth);
2383	3155	if (unlikely(vaddr & (sizeof(u64) - 1)))
2384	3156	goto nack_inv_unlck;
2385	3157	rkey = be32_to_cpu(ateth->rkey);
..	..	@@ -2398,6 +3170,7 @@
2398	3170	sdata);
2399	3171	rvt_put_mr(qp->r_sge.sge.mr);
2400	3172	qp->r_sge.num_sge = 0;
	3173	+ack:
2401	3174	e->opcode = opcode;
2402	3175	e->sent = 0;
2403	3176	e->psn = psn;
..	..	@@ -2407,6 +3180,7 @@
2407	3180	qp->r_state = opcode;
2408	3181	qp->r_nak_state = 0;
2409	3182	qp->r_head_ack_queue = next;
	3183	+ qpriv->r_tid_alloc = qp->r_head_ack_queue;
2410	3184
2411	3185	/* Schedule the send engine. */
2412	3186	qp->s_flags \|= RVT_S_RESP_PENDING;