summaryrefslogtreecommitdiffstats
path: root/drivers/infiniband/hw/hns/hns_roce_qp.c
blob: d1c07f1f8fe984f43fe6f2ebcb942763f9991596 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
/*
 * Copyright (c) 2016 Hisilicon Limited.
 * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved.
 *
 * This software is available to you under a choice of one of two
 * licenses.  You may choose to be licensed under the terms of the GNU
 * General Public License (GPL) Version 2, available from the file
 * COPYING in the main directory of this source tree, or the
 * OpenIB.org BSD license below:
 *
 *     Redistribution and use in source and binary forms, with or
 *     without modification, are permitted provided that the following
 *     conditions are met:
 *
 *      - Redistributions of source code must retain the above
 *        copyright notice, this list of conditions and the following
 *        disclaimer.
 *
 *      - Redistributions in binary form must reproduce the above
 *        copyright notice, this list of conditions and the following
 *        disclaimer in the documentation and/or other materials
 *        provided with the distribution.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
 * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
 * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
 * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
 * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 * SOFTWARE.
 */

#include <linux/pci.h>
#include <linux/platform_device.h>
#include <rdma/ib_addr.h>
#include <rdma/ib_umem.h>
#include <rdma/uverbs_ioctl.h>
#include "hns_roce_common.h"
#include "hns_roce_device.h"
#include "hns_roce_hem.h"
#include <rdma/hns-abi.h>

static void flush_work_handle(struct work_struct *work)
{
	struct hns_roce_work *flush_work = container_of(work,
					struct hns_roce_work, work);
	struct hns_roce_qp *hr_qp = container_of(flush_work,
					struct hns_roce_qp, flush_work);
	struct device *dev = flush_work->hr_dev->dev;
	struct ib_qp_attr attr;
	int attr_mask;
	int ret;

	attr_mask = IB_QP_STATE;
	attr.qp_state = IB_QPS_ERR;

	if (test_and_clear_bit(HNS_ROCE_FLUSH_FLAG, &hr_qp->flush_flag)) {
		ret = hns_roce_modify_qp(&hr_qp->ibqp, &attr, attr_mask, NULL);
		if (ret)
			dev_err(dev, "Modify QP to error state failed(%d) during CQE flush\n",
				ret);
	}

	/*
	 * make sure we signal QP destroy leg that flush QP was completed
	 * so that it can safely proceed ahead now and destroy QP
	 */
	if (atomic_dec_and_test(&hr_qp->refcount))
		complete(&hr_qp->free);
}

void init_flush_work(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp)
{
	struct hns_roce_work *flush_work = &hr_qp->flush_work;

	flush_work->hr_dev = hr_dev;
	INIT_WORK(&flush_work->work, flush_work_handle);
	atomic_inc(&hr_qp->refcount);
	queue_work(hr_dev->irq_workq, &flush_work->work);
}

void hns_roce_qp_event(struct hns_roce_dev *hr_dev, u32 qpn, int event_type)
{
	struct device *dev = hr_dev->dev;
	struct hns_roce_qp *qp;

	xa_lock(&hr_dev->qp_table_xa);
	qp = __hns_roce_qp_lookup(hr_dev, qpn);
	if (qp)
		atomic_inc(&qp->refcount);
	xa_unlock(&hr_dev->qp_table_xa);

	if (!qp) {
		dev_warn(dev, "Async event for bogus QP %08x\n", qpn);
		return;
	}

	if (hr_dev->hw_rev != HNS_ROCE_HW_VER1 &&
	    (event_type == HNS_ROCE_EVENT_TYPE_WQ_CATAS_ERROR ||
	     event_type == HNS_ROCE_EVENT_TYPE_INV_REQ_LOCAL_WQ_ERROR ||
	     event_type == HNS_ROCE_EVENT_TYPE_LOCAL_WQ_ACCESS_ERROR)) {
		qp->state = IB_QPS_ERR;
		if (!test_and_set_bit(HNS_ROCE_FLUSH_FLAG, &qp->flush_flag))
			init_flush_work(hr_dev, qp);
	}

	qp->event(qp, (enum hns_roce_event)event_type);

	if (atomic_dec_and_test(&qp->refcount))
		complete(&qp->free);
}

static void hns_roce_ib_qp_event(struct hns_roce_qp *hr_qp,
				 enum hns_roce_event type)
{
	struct ib_qp *ibqp = &hr_qp->ibqp;
	struct ib_event event;

	if (ibqp->event_handler) {
		event.device = ibqp->device;
		event.element.qp = ibqp;
		switch (type) {
		case HNS_ROCE_EVENT_TYPE_PATH_MIG:
			event.event = IB_EVENT_PATH_MIG;
			break;
		case HNS_ROCE_EVENT_TYPE_COMM_EST:
			event.event = IB_EVENT_COMM_EST;
			break;
		case HNS_ROCE_EVENT_TYPE_SQ_DRAINED:
			event.event = IB_EVENT_SQ_DRAINED;
			break;
		case HNS_ROCE_EVENT_TYPE_SRQ_LAST_WQE_REACH:
			event.event = IB_EVENT_QP_LAST_WQE_REACHED;
			break;
		case HNS_ROCE_EVENT_TYPE_WQ_CATAS_ERROR:
			event.event = IB_EVENT_QP_FATAL;
			break;
		case HNS_ROCE_EVENT_TYPE_PATH_MIG_FAILED:
			event.event = IB_EVENT_PATH_MIG_ERR;
			break;
		case HNS_ROCE_EVENT_TYPE_INV_REQ_LOCAL_WQ_ERROR:
			event.event = IB_EVENT_QP_REQ_ERR;
			break;
		case HNS_ROCE_EVENT_TYPE_LOCAL_WQ_ACCESS_ERROR:
			event.event = IB_EVENT_QP_ACCESS_ERR;
			break;
		default:
			dev_dbg(ibqp->device->dev.parent, "roce_ib: Unexpected event type %d on QP %06lx\n",
				type, hr_qp->qpn);
			return;
		}
		ibqp->event_handler(&event, ibqp->qp_context);
	}
}

static int alloc_qpn(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp)
{
	unsigned long num = 0;
	int ret;

	if (hr_qp->ibqp.qp_type == IB_QPT_GSI) {
		/* when hw version is v1, the sqpn is allocated */
		if (hr_dev->hw_rev == HNS_ROCE_HW_VER1)
			num = HNS_ROCE_MAX_PORTS +
			      hr_dev->iboe.phy_port[hr_qp->port];
		else
			num = 1;

		hr_qp->doorbell_qpn = 1;
	} else {
		ret = hns_roce_bitmap_alloc_range(&hr_dev->qp_table.bitmap,
						  1, 1, &num);
		if (ret) {
			ibdev_err(&hr_dev->ib_dev, "Failed to alloc bitmap\n");
			return -ENOMEM;
		}

		hr_qp->doorbell_qpn = (u32)num;
	}

	hr_qp->qpn = num;

	return 0;
}

enum hns_roce_qp_state to_hns_roce_state(enum ib_qp_state state)
{
	switch (state) {
	case IB_QPS_RESET:
		return HNS_ROCE_QP_STATE_RST;
	case IB_QPS_INIT:
		return HNS_ROCE_QP_STATE_INIT;
	case IB_QPS_RTR:
		return HNS_ROCE_QP_STATE_RTR;
	case IB_QPS_RTS:
		return HNS_ROCE_QP_STATE_RTS;
	case IB_QPS_SQD:
		return HNS_ROCE_QP_STATE_SQD;
	case IB_QPS_ERR:
		return HNS_ROCE_QP_STATE_ERR;
	default:
		return HNS_ROCE_QP_NUM_STATE;
	}
}

static void add_qp_to_list(struct hns_roce_dev *hr_dev,
			   struct hns_roce_qp *hr_qp,
			   struct ib_cq *send_cq, struct ib_cq *recv_cq)
{
	struct hns_roce_cq *hr_send_cq, *hr_recv_cq;
	unsigned long flags;

	hr_send_cq = send_cq ? to_hr_cq(send_cq) : NULL;
	hr_recv_cq = recv_cq ? to_hr_cq(recv_cq) : NULL;

	spin_lock_irqsave(&hr_dev->qp_list_lock, flags);
	hns_roce_lock_cqs(hr_send_cq, hr_recv_cq);

	list_add_tail(&hr_qp->node, &hr_dev->qp_list);
	if (hr_send_cq)
		list_add_tail(&hr_qp->sq_node, &hr_send_cq->sq_list);
	if (hr_recv_cq)
		list_add_tail(&hr_qp->rq_node, &hr_recv_cq->rq_list);

	hns_roce_unlock_cqs(hr_send_cq, hr_recv_cq);
	spin_unlock_irqrestore(&hr_dev->qp_list_lock, flags);
}

static int hns_roce_qp_store(struct hns_roce_dev *hr_dev,
			     struct hns_roce_qp *hr_qp,
			     struct ib_qp_init_attr *init_attr)
{
	struct xarray *xa = &hr_dev->qp_table_xa;
	int ret;

	if (!hr_qp->qpn)
		return -EINVAL;

	ret = xa_err(xa_store_irq(xa, hr_qp->qpn, hr_qp, GFP_KERNEL));
	if (ret)
		dev_err(hr_dev->dev, "Failed to xa store for QPC\n");
	else
		/* add QP to device's QP list for softwc */
		add_qp_to_list(hr_dev, hr_qp, init_attr->send_cq,
			       init_attr->recv_cq);

	return ret;
}

static int alloc_qpc(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp)
{
	struct hns_roce_qp_table *qp_table = &hr_dev->qp_table;
	struct device *dev = hr_dev->dev;
	int ret;

	if (!hr_qp->qpn)
		return -EINVAL;

	/* In v1 engine, GSI QP context is saved in the RoCE hw's register */
	if (hr_qp->ibqp.qp_type == IB_QPT_GSI &&
	    hr_dev->hw_rev == HNS_ROCE_HW_VER1)
		return 0;

	/* Alloc memory for QPC */
	ret = hns_roce_table_get(hr_dev, &qp_table->qp_table, hr_qp->qpn);
	if (ret) {
		dev_err(dev, "Failed to get QPC table\n");
		goto err_out;
	}

	/* Alloc memory for IRRL */
	ret = hns_roce_table_get(hr_dev, &qp_table->irrl_table, hr_qp->qpn);
	if (ret) {
		dev_err(dev, "Failed to get IRRL table\n");
		goto err_put_qp;
	}

	if (hr_dev->caps.trrl_entry_sz) {
		/* Alloc memory for TRRL */
		ret = hns_roce_table_get(hr_dev, &qp_table->trrl_table,
					 hr_qp->qpn);
		if (ret) {
			dev_err(dev, "Failed to get TRRL table\n");
			goto err_put_irrl;
		}
	}

	if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_QP_FLOW_CTRL) {
		/* Alloc memory for SCC CTX */
		ret = hns_roce_table_get(hr_dev, &qp_table->sccc_table,
					 hr_qp->qpn);
		if (ret) {
			dev_err(dev, "Failed to get SCC CTX table\n");
			goto err_put_trrl;
		}
	}

	return 0;

err_put_trrl:
	if (hr_dev->caps.trrl_entry_sz)
		hns_roce_table_put(hr_dev, &qp_table->trrl_table, hr_qp->qpn);

err_put_irrl:
	hns_roce_table_put(hr_dev, &qp_table->irrl_table, hr_qp->qpn);

err_put_qp:
	hns_roce_table_put(hr_dev, &qp_table->qp_table, hr_qp->qpn);

err_out:
	return ret;
}

void hns_roce_qp_remove(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp)
{
	struct xarray *xa = &hr_dev->qp_table_xa;
	unsigned long flags;

	list_del(&hr_qp->node);
	list_del(&hr_qp->sq_node);
	list_del(&hr_qp->rq_node);

	xa_lock_irqsave(xa, flags);
	__xa_erase(xa, hr_qp->qpn & (hr_dev->caps.num_qps - 1));
	xa_unlock_irqrestore(xa, flags);
}

static void free_qpc(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp)
{
	struct hns_roce_qp_table *qp_table = &hr_dev->qp_table;

	/* In v1 engine, GSI QP context is saved in the RoCE hw's register */
	if (hr_qp->ibqp.qp_type == IB_QPT_GSI &&
	    hr_dev->hw_rev == HNS_ROCE_HW_VER1)
		return;

	if (hr_dev->caps.trrl_entry_sz)
		hns_roce_table_put(hr_dev, &qp_table->trrl_table, hr_qp->qpn);
	hns_roce_table_put(hr_dev, &qp_table->irrl_table, hr_qp->qpn);
}

static void free_qpn(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp)
{
	struct hns_roce_qp_table *qp_table = &hr_dev->qp_table;

	if (hr_qp->ibqp.qp_type == IB_QPT_GSI)
		return;

	if (hr_qp->qpn < hr_dev->caps.reserved_qps)
		return;

	hns_roce_bitmap_free_range(&qp_table->bitmap, hr_qp->qpn, 1, BITMAP_RR);
}

static int set_rq_size(struct hns_roce_dev *hr_dev, struct ib_qp_cap *cap,
		       struct hns_roce_qp *hr_qp, int has_rq)
{
	u32 cnt;

	/* If srq exist, set zero for relative number of rq */
	if (!has_rq) {
		hr_qp->rq.wqe_cnt = 0;
		hr_qp->rq.max_gs = 0;
		hr_qp->rq_inl_buf.wqe_cnt = 0;
		cap->max_recv_wr = 0;
		cap->max_recv_sge = 0;

		return 0;
	}

	/* Check the validity of QP support capacity */
	if (!cap->max_recv_wr || cap->max_recv_wr > hr_dev->caps.max_wqes ||
	    cap->max_recv_sge > hr_dev->caps.max_rq_sg) {
		ibdev_err(&hr_dev->ib_dev, "RQ config error, depth=%u, sge=%d\n",
			  cap->max_recv_wr, cap->max_recv_sge);
		return -EINVAL;
	}

	cnt = roundup_pow_of_two(max(cap->max_recv_wr, hr_dev->caps.min_wqes));
	if (cnt > hr_dev->caps.max_wqes) {
		ibdev_err(&hr_dev->ib_dev, "rq depth %u too large\n",
			  cap->max_recv_wr);
		return -EINVAL;
	}

	hr_qp->rq.max_gs = roundup_pow_of_two(max(1U, cap->max_recv_sge));

	hr_qp->rq.wqe_shift = ilog2(hr_dev->caps.max_rq_desc_sz *
				    hr_qp->rq.max_gs);

	hr_qp->rq.wqe_cnt = cnt;
	if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RQ_INLINE)
		hr_qp->rq_inl_buf.wqe_cnt = cnt;
	else
		hr_qp->rq_inl_buf.wqe_cnt = 0;

	cap->max_recv_wr = cnt;
	cap->max_recv_sge = hr_qp->rq.max_gs;

	return 0;
}

static int set_extend_sge_param(struct hns_roce_dev *hr_dev, u32 sq_wqe_cnt,
				struct hns_roce_qp *hr_qp,
				struct ib_qp_cap *cap)
{
	u32 cnt;

	cnt = max(1U, cap->max_send_sge);
	if (hr_dev->hw_rev == HNS_ROCE_HW_VER1) {
		hr_qp->sq.max_gs = roundup_pow_of_two(cnt);
		hr_qp->sge.sge_cnt = 0;

		return 0;
	}

	hr_qp->sq.max_gs = cnt;

	/* UD sqwqe's sge use extend sge */
	if (hr_qp->ibqp.qp_type == IB_QPT_GSI ||
	    hr_qp->ibqp.qp_type == IB_QPT_UD) {
		cnt = roundup_pow_of_two(sq_wqe_cnt * hr_qp->sq.max_gs);
	} else if (hr_qp->sq.max_gs > HNS_ROCE_SGE_IN_WQE) {
		cnt = roundup_pow_of_two(sq_wqe_cnt *
				     (hr_qp->sq.max_gs - HNS_ROCE_SGE_IN_WQE));
	} else {
		cnt = 0;
	}

	hr_qp->sge.sge_shift = HNS_ROCE_SGE_SHIFT;

	/* If the number of extended sge is not zero, they MUST use the
	 * space of HNS_HW_PAGE_SIZE at least.
	 */
	hr_qp->sge.sge_cnt = cnt ?
			max(cnt, (u32)HNS_HW_PAGE_SIZE / HNS_ROCE_SGE_SIZE) : 0;

	return 0;
}

static int check_sq_size_with_integrity(struct hns_roce_dev *hr_dev,
					struct ib_qp_cap *cap,
					struct hns_roce_ib_create_qp *ucmd)
{
	u32 roundup_sq_stride = roundup_pow_of_two(hr_dev->caps.max_sq_desc_sz);
	u8 max_sq_stride = ilog2(roundup_sq_stride);

	/* Sanity check SQ size before proceeding */
	if (ucmd->log_sq_stride > max_sq_stride ||
	    ucmd->log_sq_stride < HNS_ROCE_IB_MIN_SQ_STRIDE) {
		ibdev_err(&hr_dev->ib_dev, "failed to check SQ stride size.\n");
		return -EINVAL;
	}

	if (cap->max_send_sge > hr_dev->caps.max_sq_sg) {
		ibdev_err(&hr_dev->ib_dev, "failed to check SQ SGE size %u.\n",
			  cap->max_send_sge);
		return -EINVAL;
	}

	return 0;
}

static int set_user_sq_size(struct hns_roce_dev *hr_dev,
			    struct ib_qp_cap *cap, struct hns_roce_qp *hr_qp,
			    struct hns_roce_ib_create_qp *ucmd)
{
	struct ib_device *ibdev = &hr_dev->ib_dev;
	u32 cnt = 0;
	int ret;

	if (check_shl_overflow(1, ucmd->log_sq_bb_count, &cnt) ||
	    cnt > hr_dev->caps.max_wqes)
		return -EINVAL;

	ret = check_sq_size_with_integrity(hr_dev, cap, ucmd);
	if (ret) {
		ibdev_err(ibdev, "failed to check user SQ size, ret = %d.\n",
			  ret);
		return ret;
	}

	ret = set_extend_sge_param(hr_dev, cnt, hr_qp, cap);
	if (ret)
		return ret;

	hr_qp->sq.wqe_shift = ucmd->log_sq_stride;
	hr_qp->sq.wqe_cnt = cnt;

	return 0;
}

static int set_wqe_buf_attr(struct hns_roce_dev *hr_dev,
			    struct hns_roce_qp *hr_qp,
			    struct hns_roce_buf_attr *buf_attr)
{
	int buf_size;
	int idx = 0;

	hr_qp->buff_size = 0;

	/* SQ WQE */
	hr_qp->sq.offset = 0;
	buf_size = to_hr_hem_entries_size(hr_qp->sq.wqe_cnt,
					  hr_qp->sq.wqe_shift);
	if (buf_size > 0 && idx < ARRAY_SIZE(buf_attr->region)) {
		buf_attr->region[idx].size = buf_size;
		buf_attr->region[idx].hopnum = hr_dev->caps.wqe_sq_hop_num;
		idx++;
		hr_qp->buff_size += buf_size;
	}

	/* extend SGE WQE in SQ */
	hr_qp->sge.offset = hr_qp->buff_size;
	buf_size = to_hr_hem_entries_size(hr_qp->sge.sge_cnt,
					  hr_qp->sge.sge_shift);
	if (buf_size > 0 && idx < ARRAY_SIZE(buf_attr->region)) {
		buf_attr->region[idx].size = buf_size;
		buf_attr->region[idx].hopnum = hr_dev->caps.wqe_sge_hop_num;
		idx++;
		hr_qp->buff_size += buf_size;
	}

	/* RQ WQE */
	hr_qp->rq.offset = hr_qp->buff_size;
	buf_size = to_hr_hem_entries_size(hr_qp->rq.wqe_cnt,
					  hr_qp->rq.wqe_shift);
	if (buf_size > 0 && idx < ARRAY_SIZE(buf_attr->region)) {
		buf_attr->region[idx].size = buf_size;
		buf_attr->region[idx].hopnum = hr_dev->caps.wqe_rq_hop_num;
		idx++;
		hr_qp->buff_size += buf_size;
	}

	if (hr_qp->buff_size < 1)
		return -EINVAL;

	buf_attr->page_shift = HNS_HW_PAGE_SHIFT + hr_dev->caps.mtt_buf_pg_sz;
	buf_attr->fixed_page = true;
	buf_attr->region_count = idx;

	return 0;
}

static int set_kernel_sq_size(struct hns_roce_dev *hr_dev,
			      struct ib_qp_cap *cap, struct hns_roce_qp *hr_qp)
{
	struct ib_device *ibdev = &hr_dev->ib_dev;
	u32 cnt;
	int ret;

	if (!cap->max_send_wr || cap->max_send_wr > hr_dev->caps.max_wqes ||
	    cap->max_send_sge > hr_dev->caps.max_sq_sg) {
		ibdev_err(ibdev,
			  "failed to check SQ WR or SGE num, ret = %d.\n",
			  -EINVAL);
		return -EINVAL;
	}

	cnt = roundup_pow_of_two(max(cap->max_send_wr, hr_dev->caps.min_wqes));
	if (cnt > hr_dev->caps.max_wqes) {
		ibdev_err(ibdev, "failed to check WQE num, WQE num = %u.\n",
			  cnt);
		return -EINVAL;
	}

	hr_qp->sq.wqe_shift = ilog2(hr_dev->caps.max_sq_desc_sz);
	hr_qp->sq.wqe_cnt = cnt;

	ret = set_extend_sge_param(hr_dev, cnt, hr_qp, cap);
	if (ret)
		return ret;

	/* sync the parameters of kernel QP to user's configuration */
	cap->max_send_wr = cnt;
	cap->max_send_sge = hr_qp->sq.max_gs;

	return 0;
}

static int hns_roce_qp_has_sq(struct ib_qp_init_attr *attr)
{
	if (attr->qp_type == IB_QPT_XRC_TGT || !attr->cap.max_send_wr)
		return 0;

	return 1;
}

static int hns_roce_qp_has_rq(struct ib_qp_init_attr *attr)
{
	if (attr->qp_type == IB_QPT_XRC_INI ||
	    attr->qp_type == IB_QPT_XRC_TGT || attr->srq ||
	    !attr->cap.max_recv_wr)
		return 0;

	return 1;
}

static int alloc_rq_inline_buf(struct hns_roce_qp *hr_qp,
			       struct ib_qp_init_attr *init_attr)
{
	u32 max_recv_sge = init_attr->cap.max_recv_sge;
	u32 wqe_cnt = hr_qp->rq_inl_buf.wqe_cnt;
	struct hns_roce_rinl_wqe *wqe_list;
	int i;

	/* allocate recv inline buf */
	wqe_list = kcalloc(wqe_cnt, sizeof(struct hns_roce_rinl_wqe),
			   GFP_KERNEL);

	if (!wqe_list)
		goto err;

	/* Allocate a continuous buffer for all inline sge we need */
	wqe_list[0].sg_list = kcalloc(wqe_cnt, (max_recv_sge *
				      sizeof(struct hns_roce_rinl_sge)),
				      GFP_KERNEL);
	if (!wqe_list[0].sg_list)
		goto err_wqe_list;

	/* Assign buffers of sg_list to each inline wqe */
	for (i = 1; i < wqe_cnt; i++)
		wqe_list[i].sg_list = &wqe_list[0].sg_list[i * max_recv_sge];

	hr_qp->rq_inl_buf.wqe_list = wqe_list;

	return 0;

err_wqe_list:
	kfree(wqe_list);

err:
	return -ENOMEM;
}

static void free_rq_inline_buf(struct hns_roce_qp *hr_qp)
{
	if (hr_qp->rq_inl_buf.wqe_list)
		kfree(hr_qp->rq_inl_buf.wqe_list[0].sg_list);
	kfree(hr_qp->rq_inl_buf.wqe_list);
}

static int alloc_qp_buf(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp,
			struct ib_qp_init_attr *init_attr,
			struct ib_udata *udata, unsigned long addr)
{
	struct ib_device *ibdev = &hr_dev->ib_dev;
	struct hns_roce_buf_attr buf_attr = {};
	int ret;

	if (!udata && hr_qp->rq_inl_buf.wqe_cnt) {
		ret = alloc_rq_inline_buf(hr_qp, init_attr);
		if (ret) {
			ibdev_err(ibdev,
				  "failed to alloc inline buf, ret = %d.\n",
				  ret);
			return ret;
		}
	} else {
		hr_qp->rq_inl_buf.wqe_list = NULL;
	}

	ret = set_wqe_buf_attr(hr_dev, hr_qp, &buf_attr);
	if (ret) {
		ibdev_err(ibdev, "failed to split WQE buf, ret = %d.\n", ret);
		goto err_inline;
	}
	ret = hns_roce_mtr_create(hr_dev, &hr_qp->mtr, &buf_attr,
				  HNS_HW_PAGE_SHIFT + hr_dev->caps.mtt_ba_pg_sz,
				  udata, addr);
	if (ret) {
		ibdev_err(ibdev, "failed to create WQE mtr, ret = %d.\n", ret);
		goto err_inline;
	}

	return 0;
err_inline:
	free_rq_inline_buf(hr_qp);

	return ret;
}

static void free_qp_buf(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp)
{
	hns_roce_mtr_destroy(hr_dev, &hr_qp->mtr);
	free_rq_inline_buf(hr_qp);
}

static inline bool user_qp_has_sdb(struct hns_roce_dev *hr_dev,
				   struct ib_qp_init_attr *init_attr,
				   struct ib_udata *udata,
				   struct hns_roce_ib_create_qp_resp *resp,
				   struct hns_roce_ib_create_qp *ucmd)
{
	return ((hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_SQ_RECORD_DB) &&
		udata->outlen >= offsetofend(typeof(*resp), cap_flags) &&
		hns_roce_qp_has_sq(init_attr) &&
		udata->inlen >= offsetofend(typeof(*ucmd), sdb_addr));
}

static inline bool user_qp_has_rdb(struct hns_roce_dev *hr_dev,
				   struct ib_qp_init_attr *init_attr,
				   struct ib_udata *udata,
				   struct hns_roce_ib_create_qp_resp *resp)
{
	return ((hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RECORD_DB) &&
		udata->outlen >= offsetofend(typeof(*resp), cap_flags) &&
		hns_roce_qp_has_rq(init_attr));
}

static inline bool kernel_qp_has_rdb(struct hns_roce_dev *hr_dev,
				     struct ib_qp_init_attr *init_attr)
{
	return ((hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RECORD_DB) &&
		hns_roce_qp_has_rq(init_attr));
}

static int alloc_qp_db(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp,
		       struct ib_qp_init_attr *init_attr,
		       struct ib_udata *udata,
		       struct hns_roce_ib_create_qp *ucmd,
		       struct hns_roce_ib_create_qp_resp *resp)
{
	struct hns_roce_ucontext *uctx = rdma_udata_to_drv_context(
		udata, struct hns_roce_ucontext, ibucontext);
	struct ib_device *ibdev = &hr_dev->ib_dev;
	int ret;

	if (udata) {
		if (user_qp_has_sdb(hr_dev, init_attr, udata, resp, ucmd)) {
			ret = hns_roce_db_map_user(uctx, udata, ucmd->sdb_addr,
						   &hr_qp->sdb);
			if (ret) {
				ibdev_err(ibdev,
					  "failed to map user SQ doorbell, ret = %d.\n",
					  ret);
				goto err_out;
			}
			hr_qp->en_flags |= HNS_ROCE_QP_CAP_SQ_RECORD_DB;
		}

		if (user_qp_has_rdb(hr_dev, init_attr, udata, resp)) {
			ret = hns_roce_db_map_user(uctx, udata, ucmd->db_addr,
						   &hr_qp->rdb);
			if (ret) {
				ibdev_err(ibdev,
					  "failed to map user RQ doorbell, ret = %d.\n",
					  ret);
				goto err_sdb;
			}
			hr_qp->en_flags |= HNS_ROCE_QP_CAP_RQ_RECORD_DB;
		}
	} else {
		/* QP doorbell register address */
		hr_qp->sq.db_reg_l = hr_dev->reg_base + hr_dev->sdb_offset +
				     DB_REG_OFFSET * hr_dev->priv_uar.index;
		hr_qp->rq.db_reg_l = hr_dev->reg_base + hr_dev->odb_offset +
				     DB_REG_OFFSET * hr_dev->priv_uar.index;

		if (kernel_qp_has_rdb(hr_dev, init_attr)) {
			ret = hns_roce_alloc_db(hr_dev, &hr_qp->rdb, 0);
			if (ret) {
				ibdev_err(ibdev,
					  "failed to alloc kernel RQ doorbell, ret = %d.\n",
					  ret);
				goto err_out;
			}
			*hr_qp->rdb.db_record = 0;
			hr_qp->en_flags |= HNS_ROCE_QP_CAP_RQ_RECORD_DB;
		}
	}

	return 0;
err_sdb:
	if (udata && hr_qp->en_flags & HNS_ROCE_QP_CAP_SQ_RECORD_DB)
		hns_roce_db_unmap_user(uctx, &hr_qp->sdb);
err_out:
	return ret;
}

static void free_qp_db(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp,
		       struct ib_udata *udata)
{
	struct hns_roce_ucontext *uctx = rdma_udata_to_drv_context(
		udata, struct hns_roce_ucontext, ibucontext);

	if (udata) {
		if (hr_qp->en_flags & HNS_ROCE_QP_CAP_RQ_RECORD_DB)
			hns_roce_db_unmap_user(uctx, &hr_qp->rdb);
		if (hr_qp->en_flags & HNS_ROCE_QP_CAP_SQ_RECORD_DB)
			hns_roce_db_unmap_user(uctx, &hr_qp->sdb);
	} else {
		if (hr_qp->en_flags & HNS_ROCE_QP_CAP_RQ_RECORD_DB)
			hns_roce_free_db(hr_dev, &hr_qp->rdb);
	}
}

static int alloc_kernel_wrid(struct hns_roce_dev *hr_dev,
			     struct hns_roce_qp *hr_qp)
{
	struct ib_device *ibdev = &hr_dev->ib_dev;
	u64 *sq_wrid = NULL;
	u64 *rq_wrid = NULL;
	int ret;

	sq_wrid = kcalloc(hr_qp->sq.wqe_cnt, sizeof(u64), GFP_KERNEL);
	if (ZERO_OR_NULL_PTR(sq_wrid)) {
		ibdev_err(ibdev, "failed to alloc SQ wrid.\n");
		return -ENOMEM;
	}

	if (hr_qp->rq.wqe_cnt) {
		rq_wrid = kcalloc(hr_qp->rq.wqe_cnt, sizeof(u64), GFP_KERNEL);
		if (ZERO_OR_NULL_PTR(rq_wrid)) {
			ibdev_err(ibdev, "failed to alloc RQ wrid.\n");
			ret = -ENOMEM;
			goto err_sq;
		}
	}

	hr_qp->sq.wrid = sq_wrid;
	hr_qp->rq.wrid = rq_wrid;
	return 0;
err_sq:
	kfree(sq_wrid);

	return ret;
}

static void free_kernel_wrid(struct hns_roce_qp *hr_qp)
{
	kfree(hr_qp->rq.wrid);
	kfree(hr_qp->sq.wrid);
}

static int set_qp_param(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp,
			struct ib_qp_init_attr *init_attr,
			struct ib_udata *udata,
			struct hns_roce_ib_create_qp *ucmd)
{
	struct ib_device *ibdev = &hr_dev->ib_dev;
	int ret;

	hr_qp->ibqp.qp_type = init_attr->qp_type;

	if (init_attr->cap.max_inline_data > hr_dev->caps.max_sq_inline)
		init_attr->cap.max_inline_data = hr_dev->caps.max_sq_inline;

	hr_qp->max_inline_data = init_attr->cap.max_inline_data;

	if (init_attr->sq_sig_type == IB_SIGNAL_ALL_WR)
		hr_qp->sq_signal_bits = IB_SIGNAL_ALL_WR;
	else
		hr_qp->sq_signal_bits = IB_SIGNAL_REQ_WR;

	ret = set_rq_size(hr_dev, &init_attr->cap, hr_qp,
			  hns_roce_qp_has_rq(init_attr));
	if (ret) {
		ibdev_err(ibdev, "failed to set user RQ size, ret = %d.\n",
			  ret);
		return ret;
	}

	if (udata) {
		ret = ib_copy_from_udata(ucmd, udata,
					 min(udata->inlen, sizeof(*ucmd)));
		if (ret) {
			ibdev_err(ibdev,
				  "failed to copy QP ucmd, ret = %d\n", ret);
			return ret;
		}

		ret = set_user_sq_size(hr_dev, &init_attr->cap, hr_qp, ucmd);
		if (ret)
			ibdev_err(ibdev,
				  "failed to set user SQ size, ret = %d.\n",
				  ret);
	} else {
		if (init_attr->create_flags &
		    IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK) {
			ibdev_err(ibdev, "Failed to check multicast loopback\n");
			return -EINVAL;
		}

		if (init_attr->create_flags & IB_QP_CREATE_IPOIB_UD_LSO) {
			ibdev_err(ibdev, "Failed to check ipoib ud lso\n");
			return -EINVAL;
		}

		ret = set_kernel_sq_size(hr_dev, &init_attr->cap, hr_qp);
		if (ret)
			ibdev_err(ibdev,
				  "failed to set kernel SQ size, ret = %d.\n",
				  ret);
	}

	return ret;
}

static int hns_roce_create_qp_common(struct hns_roce_dev *hr_dev,
				     struct ib_pd *ib_pd,
				     struct ib_qp_init_attr *init_attr,
				     struct ib_udata *udata,
				     struct hns_roce_qp *hr_qp)
{
	struct hns_roce_ib_create_qp_resp resp = {};
	struct ib_device *ibdev = &hr_dev->ib_dev;
	struct hns_roce_ib_create_qp ucmd = {};
	int ret;

	mutex_init(&hr_qp->mutex);
	spin_lock_init(&hr_qp->sq.lock);
	spin_lock_init(&hr_qp->rq.lock);

	hr_qp->state = IB_QPS_RESET;
	hr_qp->flush_flag = 0;

	ret = set_qp_param(hr_dev, hr_qp, init_attr, udata, &ucmd);
	if (ret) {
		ibdev_err(ibdev, "failed to set QP param, ret = %d.\n", ret);
		return ret;
	}

	if (!udata) {
		ret = alloc_kernel_wrid(hr_dev, hr_qp);
		if (ret) {
			ibdev_err(ibdev, "failed to alloc wrid, ret = %d.\n",
				  ret);
			return ret;
		}
	}

	ret = alloc_qp_db(hr_dev, hr_qp, init_attr, udata, &ucmd, &resp);
	if (ret) {
		ibdev_err(ibdev, "failed to alloc QP doorbell, ret = %d.\n",
			  ret);
		goto err_wrid;
	}

	ret = alloc_qp_buf(hr_dev, hr_qp, init_attr, udata, ucmd.buf_addr);
	if (ret) {
		ibdev_err(ibdev, "failed to alloc QP buffer, ret = %d.\n", ret);
		goto err_db;
	}

	ret = alloc_qpn(hr_dev, hr_qp);
	if (ret) {
		ibdev_err(ibdev, "failed to alloc QPN, ret = %d.\n", ret);
		goto err_buf;
	}

	ret = alloc_qpc(hr_dev, hr_qp);
	if (ret) {
		ibdev_err(ibdev, "failed to alloc QP context, ret = %d.\n",
			  ret);
		goto err_qpn;
	}

	ret = hns_roce_qp_store(hr_dev, hr_qp, init_attr);
	if (ret) {
		ibdev_err(ibdev, "failed to store QP, ret = %d.\n", ret);
		goto err_qpc;
	}

	if (udata) {
		resp.cap_flags = hr_qp->en_flags;
		ret = ib_copy_to_udata(udata, &resp,
				       min(udata->outlen, sizeof(resp)));
		if (ret) {
			ibdev_err(ibdev, "copy qp resp failed!\n");
			goto err_store;
		}
	}

	if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_QP_FLOW_CTRL) {
		ret = hr_dev->hw->qp_flow_control_init(hr_dev, hr_qp);
		if (ret)
			goto err_store;
	}

	hr_qp->ibqp.qp_num = hr_qp->qpn;
	hr_qp->event = hns_roce_ib_qp_event;
	atomic_set(&hr_qp->refcount, 1);
	init_completion(&hr_qp->free);

	return 0;

err_store:
	hns_roce_qp_remove(hr_dev, hr_qp);
err_qpc:
	free_qpc(hr_dev, hr_qp);
err_qpn:
	free_qpn(hr_dev, hr_qp);
err_buf:
	free_qp_buf(hr_dev, hr_qp);
err_db:
	free_qp_db(hr_dev, hr_qp, udata);
err_wrid:
	free_kernel_wrid(hr_qp);
	return ret;
}

void hns_roce_qp_destroy(struct hns_roce_dev *hr_dev, struct hns_roce_qp *hr_qp,
			 struct ib_udata *udata)
{
	if (atomic_dec_and_test(&hr_qp->refcount))
		complete(&hr_qp->free);
	wait_for_completion(&hr_qp->free);

	free_qpc(hr_dev, hr_qp);
	free_qpn(hr_dev, hr_qp);
	free_qp_buf(hr_dev, hr_qp);
	free_kernel_wrid(hr_qp);
	free_qp_db(hr_dev, hr_qp, udata);

	kfree(hr_qp);
}

struct ib_qp *hns_roce_create_qp(struct ib_pd *pd,
				 struct ib_qp_init_attr *init_attr,
				 struct ib_udata *udata)
{
	struct hns_roce_dev *hr_dev = to_hr_dev(pd->device);
	struct ib_device *ibdev = &hr_dev->ib_dev;
	struct hns_roce_qp *hr_qp;
	int ret;

	switch (init_attr->qp_type) {
	case IB_QPT_RC:
	case IB_QPT_GSI:
		break;
	default:
		ibdev_err(ibdev, "not support QP type %d\n",
			  init_attr->qp_type);
		return ERR_PTR(-EOPNOTSUPP);
	}

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

	if (init_attr->qp_type == IB_QPT_GSI) {
		hr_qp->port = init_attr->port_num - 1;
		hr_qp->phy_port = hr_dev->iboe.phy_port[hr_qp->port];
	}

	ret = hns_roce_create_qp_common(hr_dev, pd, init_attr, udata, hr_qp);
	if (ret) {
		ibdev_err(ibdev, "Create QP type 0x%x failed(%d)\n",
			  init_attr->qp_type, ret);
		ibdev_err(ibdev, "Create GSI QP failed!\n");
		kfree(hr_qp);
		return ERR_PTR(ret);
	}
	return &hr_qp->ibqp;
}

int to_hr_qp_type(int qp_type)
{
	int transport_type;

	if (qp_type == IB_QPT_RC)
		transport_type = SERV_TYPE_RC;
	else if (qp_type == IB_QPT_UC)
		transport_type = SERV_TYPE_UC;
	else if (qp_type == IB_QPT_UD)
		transport_type = SERV_TYPE_UD;
	else if (qp_type == IB_QPT_GSI)
		transport_type = SERV_TYPE_UD;
	else
		transport_type = -1;

	return transport_type;
}

static int check_mtu_validate(struct hns_roce_dev *hr_dev,
			      struct hns_roce_qp *hr_qp,
			      struct ib_qp_attr *attr, int attr_mask)
{
	enum ib_mtu active_mtu;
	int p;

	p = attr_mask & IB_QP_PORT ? (attr->port_num - 1) : hr_qp->port;
	active_mtu = iboe_get_mtu(hr_dev->iboe.netdevs[p]->mtu);

	if ((hr_dev->caps.max_mtu >= IB_MTU_2048 &&
	    attr->path_mtu > hr_dev->caps.max_mtu) ||
	    attr->path_mtu < IB_MTU_256 || attr->path_mtu > active_mtu) {
		ibdev_err(&hr_dev->ib_dev,
			"attr path_mtu(%d)invalid while modify qp",
			attr->path_mtu);
		return -EINVAL;
	}

	return 0;
}

static int hns_roce_check_qp_attr(struct ib_qp *ibqp, struct ib_qp_attr *attr,
				  int attr_mask)
{
	struct hns_roce_dev *hr_dev = to_hr_dev(ibqp->device);
	struct hns_roce_qp *hr_qp = to_hr_qp(ibqp);
	int p;

	if ((attr_mask & IB_QP_PORT) &&
	    (attr->port_num == 0 || attr->port_num > hr_dev->caps.num_ports)) {
		ibdev_err(&hr_dev->ib_dev, "invalid attr, port_num = %u.\n",
			  attr->port_num);
		return -EINVAL;
	}

	if (attr_mask & IB_QP_PKEY_INDEX) {
		p = attr_mask & IB_QP_PORT ? (attr->port_num - 1) : hr_qp->port;
		if (attr->pkey_index >= hr_dev->caps.pkey_table_len[p]) {
			ibdev_err(&hr_dev->ib_dev,
				  "invalid attr, pkey_index = %u.\n",
				  attr->pkey_index);
			return -EINVAL;
		}
	}

	if (attr_mask & IB_QP_MAX_QP_RD_ATOMIC &&
	    attr->max_rd_atomic > hr_dev->caps.max_qp_init_rdma) {
		ibdev_err(&hr_dev->ib_dev,
			  "invalid attr, max_rd_atomic = %u.\n",
			  attr->max_rd_atomic);
		return -EINVAL;
	}

	if (attr_mask & IB_QP_MAX_DEST_RD_ATOMIC &&
	    attr->max_dest_rd_atomic > hr_dev->caps.max_qp_dest_rdma) {
		ibdev_err(&hr_dev->ib_dev,
			  "invalid attr, max_dest_rd_atomic = %u.\n",
			  attr->max_dest_rd_atomic);
		return -EINVAL;
	}

	if (attr_mask & IB_QP_PATH_MTU)
		return check_mtu_validate(hr_dev, hr_qp, attr, attr_mask);

	return 0;
}

int hns_roce_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
		       int attr_mask, struct ib_udata *udata)
{
	struct hns_roce_dev *hr_dev = to_hr_dev(ibqp->device);
	struct hns_roce_qp *hr_qp = to_hr_qp(ibqp);
	enum ib_qp_state cur_state, new_state;
	int ret = -EINVAL;

	mutex_lock(&hr_qp->mutex);

	if (attr_mask & IB_QP_CUR_STATE && attr->cur_qp_state != hr_qp->state)
		goto out;

	cur_state = hr_qp->state;
	new_state = attr_mask & IB_QP_STATE ? attr->qp_state : cur_state;

	if (ibqp->uobject &&
	    (attr_mask & IB_QP_STATE) && new_state == IB_QPS_ERR) {
		if (hr_qp->en_flags & HNS_ROCE_QP_CAP_SQ_RECORD_DB) {
			hr_qp->sq.head = *(int *)(hr_qp->sdb.virt_addr);

			if (hr_qp->en_flags & HNS_ROCE_QP_CAP_RQ_RECORD_DB)
				hr_qp->rq.head = *(int *)(hr_qp->rdb.virt_addr);
		} else {
			ibdev_warn(&hr_dev->ib_dev,
				  "flush cqe is not supported in userspace!\n");
			goto out;
		}
	}

	if (!ib_modify_qp_is_ok(cur_state, new_state, ibqp->qp_type,
				attr_mask)) {
		ibdev_err(&hr_dev->ib_dev, "ib_modify_qp_is_ok failed\n");
		goto out;
	}

	ret = hns_roce_check_qp_attr(ibqp, attr, attr_mask);
	if (ret)
		goto out;

	if (cur_state == new_state && cur_state == IB_QPS_RESET) {
		if (hr_dev->hw_rev == HNS_ROCE_HW_VER1) {
			ret = -EPERM;
			ibdev_err(&hr_dev->ib_dev,
				  "RST2RST state is not supported\n");
		} else {
			ret = 0;
		}

		goto out;
	}

	ret = hr_dev->hw->modify_qp(ibqp, attr, attr_mask, cur_state,
				    new_state);

out:
	mutex_unlock(&hr_qp->mutex);

	return ret;
}

void hns_roce_lock_cqs(struct hns_roce_cq *send_cq, struct hns_roce_cq *recv_cq)
		       __acquires(&send_cq->lock) __acquires(&recv_cq->lock)
{
	if (unlikely(send_cq == NULL && recv_cq == NULL)) {
		__acquire(&send_cq->lock);
		__acquire(&recv_cq->lock);
	} else if (unlikely(send_cq != NULL && recv_cq == NULL)) {
		spin_lock_irq(&send_cq->lock);
		__acquire(&recv_cq->lock);
	} else if (unlikely(send_cq == NULL && recv_cq != NULL)) {
		spin_lock_irq(&recv_cq->lock);
		__acquire(&send_cq->lock);
	} else if (send_cq == recv_cq) {
		spin_lock_irq(&send_cq->lock);
		__acquire(&recv_cq->lock);
	} else if (send_cq->cqn < recv_cq->cqn) {
		spin_lock_irq(&send_cq->lock);
		spin_lock_nested(&recv_cq->lock, SINGLE_DEPTH_NESTING);
	} else {
		spin_lock_irq(&recv_cq->lock);
		spin_lock_nested(&send_cq->lock, SINGLE_DEPTH_NESTING);
	}
}

void hns_roce_unlock_cqs(struct hns_roce_cq *send_cq,
			 struct hns_roce_cq *recv_cq) __releases(&send_cq->lock)
			 __releases(&recv_cq->lock)
{
	if (unlikely(send_cq == NULL && recv_cq == NULL)) {
		__release(&recv_cq->lock);
		__release(&send_cq->lock);
	} else if (unlikely(send_cq != NULL && recv_cq == NULL)) {
		__release(&recv_cq->lock);
		spin_unlock(&send_cq->lock);
	} else if (unlikely(send_cq == NULL && recv_cq != NULL)) {
		__release(&send_cq->lock);
		spin_unlock(&recv_cq->lock);
	} else if (send_cq == recv_cq) {
		__release(&recv_cq->lock);
		spin_unlock_irq(&send_cq->lock);
	} else if (send_cq->cqn < recv_cq->cqn) {
		spin_unlock(&recv_cq->lock);
		spin_unlock_irq(&send_cq->lock);
	} else {
		spin_unlock(&send_cq->lock);
		spin_unlock_irq(&recv_cq->lock);
	}
}

static inline void *get_wqe(struct hns_roce_qp *hr_qp, int offset)
{
	return hns_roce_buf_offset(hr_qp->mtr.kmem, offset);
}

void *hns_roce_get_recv_wqe(struct hns_roce_qp *hr_qp, int n)
{
	return get_wqe(hr_qp, hr_qp->rq.offset + (n << hr_qp->rq.wqe_shift));
}

void *hns_roce_get_send_wqe(struct hns_roce_qp *hr_qp, int n)
{
	return get_wqe(hr_qp, hr_qp->sq.offset + (n << hr_qp->sq.wqe_shift));
}

void *hns_roce_get_extend_sge(struct hns_roce_qp *hr_qp, int n)
{
	return get_wqe(hr_qp, hr_qp->sge.offset + (n << hr_qp->sge.sge_shift));
}

bool hns_roce_wq_overflow(struct hns_roce_wq *hr_wq, int nreq,
			  struct ib_cq *ib_cq)
{
	struct hns_roce_cq *hr_cq;
	u32 cur;

	cur = hr_wq->head - hr_wq->tail;
	if (likely(cur + nreq < hr_wq->wqe_cnt))
		return false;

	hr_cq = to_hr_cq(ib_cq);
	spin_lock(&hr_cq->lock);
	cur = hr_wq->head - hr_wq->tail;
	spin_unlock(&hr_cq->lock);

	return cur + nreq >= hr_wq->wqe_cnt;
}

int hns_roce_init_qp_table(struct hns_roce_dev *hr_dev)
{
	struct hns_roce_qp_table *qp_table = &hr_dev->qp_table;
	int reserved_from_top = 0;
	int reserved_from_bot;
	int ret;

	mutex_init(&qp_table->scc_mutex);
	xa_init(&hr_dev->qp_table_xa);

	reserved_from_bot = hr_dev->caps.reserved_qps;

	ret = hns_roce_bitmap_init(&qp_table->bitmap, hr_dev->caps.num_qps,
				   hr_dev->caps.num_qps - 1, reserved_from_bot,
				   reserved_from_top);
	if (ret) {
		dev_err(hr_dev->dev, "qp bitmap init failed!error=%d\n",
			ret);
		return ret;
	}

	return 0;
}

void hns_roce_cleanup_qp_table(struct hns_roce_dev *hr_dev)
{
	hns_roce_bitmap_cleanup(&hr_dev->qp_table.bitmap);
}