summaryrefslogtreecommitdiff
path: root/drivers/infiniband/hw/qedr/verbs.c
blob: 769ac07c3c8eb72d3bd2bf95446236f9c07dc34b (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
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
/* QLogic qedr NIC Driver
 * Copyright (c) 2015-2016  QLogic Corporation
 *
 * 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/dma-mapping.h>
#include <linux/crc32.h>
#include <net/ip.h>
#include <net/ipv6.h>
#include <net/udp.h>
#include <linux/iommu.h>

#include <rdma/ib_verbs.h>
#include <rdma/ib_user_verbs.h>
#include <rdma/iw_cm.h>
#include <rdma/ib_umem.h>
#include <rdma/ib_addr.h>
#include <rdma/ib_cache.h>

#include <linux/qed/common_hsi.h>
#include "qedr_hsi_rdma.h"
#include <linux/qed/qed_if.h>
#include "qedr.h"
#include "verbs.h"
#include <rdma/qedr-abi.h>
#include "qedr_cm.h"

#define DB_ADDR_SHIFT(addr)		((addr) << DB_PWM_ADDR_OFFSET_SHIFT)

static inline int qedr_ib_copy_to_udata(struct ib_udata *udata, void *src,
					size_t len)
{
	size_t min_len = min_t(size_t, len, udata->outlen);

	return ib_copy_to_udata(udata, src, min_len);
}

int qedr_query_pkey(struct ib_device *ibdev, u8 port, u16 index, u16 *pkey)
{
	if (index > QEDR_ROCE_PKEY_TABLE_LEN)
		return -EINVAL;

	*pkey = QEDR_ROCE_PKEY_DEFAULT;
	return 0;
}

int qedr_query_gid(struct ib_device *ibdev, u8 port, int index,
		   union ib_gid *sgid)
{
	struct qedr_dev *dev = get_qedr_dev(ibdev);
	int rc = 0;

	if (!rdma_cap_roce_gid_table(ibdev, port))
		return -ENODEV;

	rc = ib_get_cached_gid(ibdev, port, index, sgid, NULL);
	if (rc == -EAGAIN) {
		memcpy(sgid, &zgid, sizeof(*sgid));
		return 0;
	}

	DP_DEBUG(dev, QEDR_MSG_INIT, "query gid: index=%d %llx:%llx\n", index,
		 sgid->global.interface_id, sgid->global.subnet_prefix);

	return rc;
}

int qedr_add_gid(struct ib_device *device, u8 port_num,
		 unsigned int index, const union ib_gid *gid,
		 const struct ib_gid_attr *attr, void **context)
{
	if (!rdma_cap_roce_gid_table(device, port_num))
		return -EINVAL;

	if (port_num > QEDR_MAX_PORT)
		return -EINVAL;

	if (!context)
		return -EINVAL;

	return 0;
}

int qedr_del_gid(struct ib_device *device, u8 port_num,
		 unsigned int index, void **context)
{
	if (!rdma_cap_roce_gid_table(device, port_num))
		return -EINVAL;

	if (port_num > QEDR_MAX_PORT)
		return -EINVAL;

	if (!context)
		return -EINVAL;

	return 0;
}

int qedr_query_device(struct ib_device *ibdev,
		      struct ib_device_attr *attr, struct ib_udata *udata)
{
	struct qedr_dev *dev = get_qedr_dev(ibdev);
	struct qedr_device_attr *qattr = &dev->attr;

	if (!dev->rdma_ctx) {
		DP_ERR(dev,
		       "qedr_query_device called with invalid params rdma_ctx=%p\n",
		       dev->rdma_ctx);
		return -EINVAL;
	}

	memset(attr, 0, sizeof(*attr));

	attr->fw_ver = qattr->fw_ver;
	attr->sys_image_guid = qattr->sys_image_guid;
	attr->max_mr_size = qattr->max_mr_size;
	attr->page_size_cap = qattr->page_size_caps;
	attr->vendor_id = qattr->vendor_id;
	attr->vendor_part_id = qattr->vendor_part_id;
	attr->hw_ver = qattr->hw_ver;
	attr->max_qp = qattr->max_qp;
	attr->max_qp_wr = max_t(u32, qattr->max_sqe, qattr->max_rqe);
	attr->device_cap_flags = IB_DEVICE_CURR_QP_STATE_MOD |
	    IB_DEVICE_RC_RNR_NAK_GEN |
	    IB_DEVICE_LOCAL_DMA_LKEY | IB_DEVICE_MEM_MGT_EXTENSIONS;

	attr->max_sge = qattr->max_sge;
	attr->max_sge_rd = qattr->max_sge;
	attr->max_cq = qattr->max_cq;
	attr->max_cqe = qattr->max_cqe;
	attr->max_mr = qattr->max_mr;
	attr->max_mw = qattr->max_mw;
	attr->max_pd = qattr->max_pd;
	attr->atomic_cap = dev->atomic_cap;
	attr->max_fmr = qattr->max_fmr;
	attr->max_map_per_fmr = 16;
	attr->max_qp_init_rd_atom =
	    1 << (fls(qattr->max_qp_req_rd_atomic_resc) - 1);
	attr->max_qp_rd_atom =
	    min(1 << (fls(qattr->max_qp_resp_rd_atomic_resc) - 1),
		attr->max_qp_init_rd_atom);

	attr->max_srq = qattr->max_srq;
	attr->max_srq_sge = qattr->max_srq_sge;
	attr->max_srq_wr = qattr->max_srq_wr;

	attr->local_ca_ack_delay = qattr->dev_ack_delay;
	attr->max_fast_reg_page_list_len = qattr->max_mr / 8;
	attr->max_pkeys = QEDR_ROCE_PKEY_MAX;
	attr->max_ah = qattr->max_ah;

	return 0;
}

#define QEDR_SPEED_SDR		(1)
#define QEDR_SPEED_DDR		(2)
#define QEDR_SPEED_QDR		(4)
#define QEDR_SPEED_FDR10	(8)
#define QEDR_SPEED_FDR		(16)
#define QEDR_SPEED_EDR		(32)

static inline void get_link_speed_and_width(int speed, u8 *ib_speed,
					    u8 *ib_width)
{
	switch (speed) {
	case 1000:
		*ib_speed = QEDR_SPEED_SDR;
		*ib_width = IB_WIDTH_1X;
		break;
	case 10000:
		*ib_speed = QEDR_SPEED_QDR;
		*ib_width = IB_WIDTH_1X;
		break;

	case 20000:
		*ib_speed = QEDR_SPEED_DDR;
		*ib_width = IB_WIDTH_4X;
		break;

	case 25000:
		*ib_speed = QEDR_SPEED_EDR;
		*ib_width = IB_WIDTH_1X;
		break;

	case 40000:
		*ib_speed = QEDR_SPEED_QDR;
		*ib_width = IB_WIDTH_4X;
		break;

	case 50000:
		*ib_speed = QEDR_SPEED_QDR;
		*ib_width = IB_WIDTH_4X;
		break;

	case 100000:
		*ib_speed = QEDR_SPEED_EDR;
		*ib_width = IB_WIDTH_4X;
		break;

	default:
		/* Unsupported */
		*ib_speed = QEDR_SPEED_SDR;
		*ib_width = IB_WIDTH_1X;
	}
}

int qedr_query_port(struct ib_device *ibdev, u8 port, struct ib_port_attr *attr)
{
	struct qedr_dev *dev;
	struct qed_rdma_port *rdma_port;

	dev = get_qedr_dev(ibdev);
	if (port > 1) {
		DP_ERR(dev, "invalid_port=0x%x\n", port);
		return -EINVAL;
	}

	if (!dev->rdma_ctx) {
		DP_ERR(dev, "rdma_ctx is NULL\n");
		return -EINVAL;
	}

	rdma_port = dev->ops->rdma_query_port(dev->rdma_ctx);

	/* *attr being zeroed by the caller, avoid zeroing it here */
	if (rdma_port->port_state == QED_RDMA_PORT_UP) {
		attr->state = IB_PORT_ACTIVE;
		attr->phys_state = 5;
	} else {
		attr->state = IB_PORT_DOWN;
		attr->phys_state = 3;
	}
	attr->max_mtu = IB_MTU_4096;
	attr->active_mtu = iboe_get_mtu(dev->ndev->mtu);
	attr->lid = 0;
	attr->lmc = 0;
	attr->sm_lid = 0;
	attr->sm_sl = 0;
	attr->port_cap_flags = IB_PORT_IP_BASED_GIDS;
	attr->gid_tbl_len = QEDR_MAX_SGID;
	attr->pkey_tbl_len = QEDR_ROCE_PKEY_TABLE_LEN;
	attr->bad_pkey_cntr = rdma_port->pkey_bad_counter;
	attr->qkey_viol_cntr = 0;
	get_link_speed_and_width(rdma_port->link_speed,
				 &attr->active_speed, &attr->active_width);
	attr->max_msg_sz = rdma_port->max_msg_size;
	attr->max_vl_num = 4;

	return 0;
}

int qedr_modify_port(struct ib_device *ibdev, u8 port, int mask,
		     struct ib_port_modify *props)
{
	struct qedr_dev *dev;

	dev = get_qedr_dev(ibdev);
	if (port > 1) {
		DP_ERR(dev, "invalid_port=0x%x\n", port);
		return -EINVAL;
	}

	return 0;
}

static int qedr_add_mmap(struct qedr_ucontext *uctx, u64 phy_addr,
			 unsigned long len)
{
	struct qedr_mm *mm;

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

	mm->key.phy_addr = phy_addr;
	/* This function might be called with a length which is not a multiple
	 * of PAGE_SIZE, while the mapping is PAGE_SIZE grained and the kernel
	 * forces this granularity by increasing the requested size if needed.
	 * When qedr_mmap is called, it will search the list with the updated
	 * length as a key. To prevent search failures, the length is rounded up
	 * in advance to PAGE_SIZE.
	 */
	mm->key.len = roundup(len, PAGE_SIZE);
	INIT_LIST_HEAD(&mm->entry);

	mutex_lock(&uctx->mm_list_lock);
	list_add(&mm->entry, &uctx->mm_head);
	mutex_unlock(&uctx->mm_list_lock);

	DP_DEBUG(uctx->dev, QEDR_MSG_MISC,
		 "added (addr=0x%llx,len=0x%lx) for ctx=%p\n",
		 (unsigned long long)mm->key.phy_addr,
		 (unsigned long)mm->key.len, uctx);

	return 0;
}

static bool qedr_search_mmap(struct qedr_ucontext *uctx, u64 phy_addr,
			     unsigned long len)
{
	bool found = false;
	struct qedr_mm *mm;

	mutex_lock(&uctx->mm_list_lock);
	list_for_each_entry(mm, &uctx->mm_head, entry) {
		if (len != mm->key.len || phy_addr != mm->key.phy_addr)
			continue;

		found = true;
		break;
	}
	mutex_unlock(&uctx->mm_list_lock);
	DP_DEBUG(uctx->dev, QEDR_MSG_MISC,
		 "searched for (addr=0x%llx,len=0x%lx) for ctx=%p, result=%d\n",
		 mm->key.phy_addr, mm->key.len, uctx, found);

	return found;
}

struct ib_ucontext *qedr_alloc_ucontext(struct ib_device *ibdev,
					struct ib_udata *udata)
{
	int rc;
	struct qedr_ucontext *ctx;
	struct qedr_alloc_ucontext_resp uresp;
	struct qedr_dev *dev = get_qedr_dev(ibdev);
	struct qed_rdma_add_user_out_params oparams;

	if (!udata)
		return ERR_PTR(-EFAULT);

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

	rc = dev->ops->rdma_add_user(dev->rdma_ctx, &oparams);
	if (rc) {
		DP_ERR(dev,
		       "failed to allocate a DPI for a new RoCE application, rc=%d. To overcome this consider to increase the number of DPIs, increase the doorbell BAR size or just close unnecessary RoCE applications. In order to increase the number of DPIs consult the qedr readme\n",
		       rc);
		goto err;
	}

	ctx->dpi = oparams.dpi;
	ctx->dpi_addr = oparams.dpi_addr;
	ctx->dpi_phys_addr = oparams.dpi_phys_addr;
	ctx->dpi_size = oparams.dpi_size;
	INIT_LIST_HEAD(&ctx->mm_head);
	mutex_init(&ctx->mm_list_lock);

	memset(&uresp, 0, sizeof(uresp));

	uresp.dpm_enabled = dev->user_dpm_enabled;
	uresp.wids_enabled = 1;
	uresp.wid_count = oparams.wid_count;
	uresp.db_pa = ctx->dpi_phys_addr;
	uresp.db_size = ctx->dpi_size;
	uresp.max_send_wr = dev->attr.max_sqe;
	uresp.max_recv_wr = dev->attr.max_rqe;
	uresp.max_srq_wr = dev->attr.max_srq_wr;
	uresp.sges_per_send_wr = QEDR_MAX_SQE_ELEMENTS_PER_SQE;
	uresp.sges_per_recv_wr = QEDR_MAX_RQE_ELEMENTS_PER_RQE;
	uresp.sges_per_srq_wr = dev->attr.max_srq_sge;
	uresp.max_cqes = QEDR_MAX_CQES;

	rc = qedr_ib_copy_to_udata(udata, &uresp, sizeof(uresp));
	if (rc)
		goto err;

	ctx->dev = dev;

	rc = qedr_add_mmap(ctx, ctx->dpi_phys_addr, ctx->dpi_size);
	if (rc)
		goto err;

	DP_DEBUG(dev, QEDR_MSG_INIT, "Allocating user context %p\n",
		 &ctx->ibucontext);
	return &ctx->ibucontext;

err:
	kfree(ctx);
	return ERR_PTR(rc);
}

int qedr_dealloc_ucontext(struct ib_ucontext *ibctx)
{
	struct qedr_ucontext *uctx = get_qedr_ucontext(ibctx);
	struct qedr_mm *mm, *tmp;
	int status = 0;

	DP_DEBUG(uctx->dev, QEDR_MSG_INIT, "Deallocating user context %p\n",
		 uctx);
	uctx->dev->ops->rdma_remove_user(uctx->dev->rdma_ctx, uctx->dpi);

	list_for_each_entry_safe(mm, tmp, &uctx->mm_head, entry) {
		DP_DEBUG(uctx->dev, QEDR_MSG_MISC,
			 "deleted (addr=0x%llx,len=0x%lx) for ctx=%p\n",
			 mm->key.phy_addr, mm->key.len, uctx);
		list_del(&mm->entry);
		kfree(mm);
	}

	kfree(uctx);
	return status;
}

int qedr_mmap(struct ib_ucontext *context, struct vm_area_struct *vma)
{
	struct qedr_ucontext *ucontext = get_qedr_ucontext(context);
	struct qedr_dev *dev = get_qedr_dev(context->device);
	unsigned long vm_page = vma->vm_pgoff << PAGE_SHIFT;
	u64 unmapped_db = dev->db_phys_addr;
	unsigned long len = (vma->vm_end - vma->vm_start);
	int rc = 0;
	bool found;

	DP_DEBUG(dev, QEDR_MSG_INIT,
		 "qedr_mmap called vm_page=0x%lx vm_pgoff=0x%lx unmapped_db=0x%llx db_size=%x, len=%lx\n",
		 vm_page, vma->vm_pgoff, unmapped_db, dev->db_size, len);
	if (vma->vm_start & (PAGE_SIZE - 1)) {
		DP_ERR(dev, "Vma_start not page aligned = %ld\n",
		       vma->vm_start);
		return -EINVAL;
	}

	found = qedr_search_mmap(ucontext, vm_page, len);
	if (!found) {
		DP_ERR(dev, "Vma_pgoff not found in mapped array = %ld\n",
		       vma->vm_pgoff);
		return -EINVAL;
	}

	DP_DEBUG(dev, QEDR_MSG_INIT, "Mapping doorbell bar\n");

	if ((vm_page >= unmapped_db) && (vm_page <= (unmapped_db +
						     dev->db_size))) {
		DP_DEBUG(dev, QEDR_MSG_INIT, "Mapping doorbell bar\n");
		if (vma->vm_flags & VM_READ) {
			DP_ERR(dev, "Trying to map doorbell bar for read\n");
			return -EPERM;
		}

		vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot);

		rc = io_remap_pfn_range(vma, vma->vm_start, vma->vm_pgoff,
					PAGE_SIZE, vma->vm_page_prot);
	} else {
		DP_DEBUG(dev, QEDR_MSG_INIT, "Mapping chains\n");
		rc = remap_pfn_range(vma, vma->vm_start,
				     vma->vm_pgoff, len, vma->vm_page_prot);
	}
	DP_DEBUG(dev, QEDR_MSG_INIT, "qedr_mmap return code: %d\n", rc);
	return rc;
}

struct ib_pd *qedr_alloc_pd(struct ib_device *ibdev,
			    struct ib_ucontext *context, struct ib_udata *udata)
{
	struct qedr_dev *dev = get_qedr_dev(ibdev);
	struct qedr_pd *pd;
	u16 pd_id;
	int rc;

	DP_DEBUG(dev, QEDR_MSG_INIT, "Function called from: %s\n",
		 (udata && context) ? "User Lib" : "Kernel");

	if (!dev->rdma_ctx) {
		DP_ERR(dev, "invalid RDMA context\n");
		return ERR_PTR(-EINVAL);
	}

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

	rc = dev->ops->rdma_alloc_pd(dev->rdma_ctx, &pd_id);
	if (rc)
		goto err;

	pd->pd_id = pd_id;

	if (udata && context) {
		struct qedr_alloc_pd_uresp uresp;

		uresp.pd_id = pd_id;

		rc = qedr_ib_copy_to_udata(udata, &uresp, sizeof(uresp));
		if (rc) {
			DP_ERR(dev, "copy error pd_id=0x%x.\n", pd_id);
			dev->ops->rdma_dealloc_pd(dev->rdma_ctx, pd_id);
			goto err;
		}

		pd->uctx = get_qedr_ucontext(context);
		pd->uctx->pd = pd;
	}

	return &pd->ibpd;

err:
	kfree(pd);
	return ERR_PTR(rc);
}

int qedr_dealloc_pd(struct ib_pd *ibpd)
{
	struct qedr_dev *dev = get_qedr_dev(ibpd->device);
	struct qedr_pd *pd = get_qedr_pd(ibpd);

	if (!pd) {
		pr_err("Invalid PD received in dealloc_pd\n");
		return -EINVAL;
	}

	DP_DEBUG(dev, QEDR_MSG_INIT, "Deallocating PD %d\n", pd->pd_id);
	dev->ops->rdma_dealloc_pd(dev->rdma_ctx, pd->pd_id);

	kfree(pd);

	return 0;
}

static void qedr_free_pbl(struct qedr_dev *dev,
			  struct qedr_pbl_info *pbl_info, struct qedr_pbl *pbl)
{
	struct pci_dev *pdev = dev->pdev;
	int i;

	for (i = 0; i < pbl_info->num_pbls; i++) {
		if (!pbl[i].va)
			continue;
		dma_free_coherent(&pdev->dev, pbl_info->pbl_size,
				  pbl[i].va, pbl[i].pa);
	}

	kfree(pbl);
}

#define MIN_FW_PBL_PAGE_SIZE (4 * 1024)
#define MAX_FW_PBL_PAGE_SIZE (64 * 1024)

#define NUM_PBES_ON_PAGE(_page_size) (_page_size / sizeof(u64))
#define MAX_PBES_ON_PAGE NUM_PBES_ON_PAGE(MAX_FW_PBL_PAGE_SIZE)
#define MAX_PBES_TWO_LAYER (MAX_PBES_ON_PAGE * MAX_PBES_ON_PAGE)

static struct qedr_pbl *qedr_alloc_pbl_tbl(struct qedr_dev *dev,
					   struct qedr_pbl_info *pbl_info,
					   gfp_t flags)
{
	struct pci_dev *pdev = dev->pdev;
	struct qedr_pbl *pbl_table;
	dma_addr_t *pbl_main_tbl;
	dma_addr_t pa;
	void *va;
	int i;

	pbl_table = kcalloc(pbl_info->num_pbls, sizeof(*pbl_table), flags);
	if (!pbl_table)
		return ERR_PTR(-ENOMEM);

	for (i = 0; i < pbl_info->num_pbls; i++) {
		va = dma_alloc_coherent(&pdev->dev, pbl_info->pbl_size,
					&pa, flags);
		if (!va)
			goto err;

		memset(va, 0, pbl_info->pbl_size);
		pbl_table[i].va = va;
		pbl_table[i].pa = pa;
	}

	/* Two-Layer PBLs, if we have more than one pbl we need to initialize
	 * the first one with physical pointers to all of the rest
	 */
	pbl_main_tbl = (dma_addr_t *)pbl_table[0].va;
	for (i = 0; i < pbl_info->num_pbls - 1; i++)
		pbl_main_tbl[i] = pbl_table[i + 1].pa;

	return pbl_table;

err:
	for (i--; i >= 0; i--)
		dma_free_coherent(&pdev->dev, pbl_info->pbl_size,
				  pbl_table[i].va, pbl_table[i].pa);

	qedr_free_pbl(dev, pbl_info, pbl_table);

	return ERR_PTR(-ENOMEM);
}

static int qedr_prepare_pbl_tbl(struct qedr_dev *dev,
				struct qedr_pbl_info *pbl_info,
				u32 num_pbes, int two_layer_capable)
{
	u32 pbl_capacity;
	u32 pbl_size;
	u32 num_pbls;

	if ((num_pbes > MAX_PBES_ON_PAGE) && two_layer_capable) {
		if (num_pbes > MAX_PBES_TWO_LAYER) {
			DP_ERR(dev, "prepare pbl table: too many pages %d\n",
			       num_pbes);
			return -EINVAL;
		}

		/* calculate required pbl page size */
		pbl_size = MIN_FW_PBL_PAGE_SIZE;
		pbl_capacity = NUM_PBES_ON_PAGE(pbl_size) *
			       NUM_PBES_ON_PAGE(pbl_size);

		while (pbl_capacity < num_pbes) {
			pbl_size *= 2;
			pbl_capacity = pbl_size / sizeof(u64);
			pbl_capacity = pbl_capacity * pbl_capacity;
		}

		num_pbls = DIV_ROUND_UP(num_pbes, NUM_PBES_ON_PAGE(pbl_size));
		num_pbls++;	/* One for the layer0 ( points to the pbls) */
		pbl_info->two_layered = true;
	} else {
		/* One layered PBL */
		num_pbls = 1;
		pbl_size = max_t(u32, MIN_FW_PBL_PAGE_SIZE,
				 roundup_pow_of_two((num_pbes * sizeof(u64))));
		pbl_info->two_layered = false;
	}

	pbl_info->num_pbls = num_pbls;
	pbl_info->pbl_size = pbl_size;
	pbl_info->num_pbes = num_pbes;

	DP_DEBUG(dev, QEDR_MSG_MR,
		 "prepare pbl table: num_pbes=%d, num_pbls=%d, pbl_size=%d\n",
		 pbl_info->num_pbes, pbl_info->num_pbls, pbl_info->pbl_size);

	return 0;
}

static void qedr_populate_pbls(struct qedr_dev *dev, struct ib_umem *umem,
			       struct qedr_pbl *pbl,
			       struct qedr_pbl_info *pbl_info, u32 pg_shift)
{
	int shift, pg_cnt, pages, pbe_cnt, total_num_pbes = 0;
	u32 fw_pg_cnt, fw_pg_per_umem_pg;
	struct qedr_pbl *pbl_tbl;
	struct scatterlist *sg;
	struct regpair *pbe;
	u64 pg_addr;
	int entry;

	if (!pbl_info->num_pbes)
		return;

	/* If we have a two layered pbl, the first pbl points to the rest
	 * of the pbls and the first entry lays on the second pbl in the table
	 */
	if (pbl_info->two_layered)
		pbl_tbl = &pbl[1];
	else
		pbl_tbl = pbl;

	pbe = (struct regpair *)pbl_tbl->va;
	if (!pbe) {
		DP_ERR(dev, "cannot populate PBL due to a NULL PBE\n");
		return;
	}

	pbe_cnt = 0;

	shift = umem->page_shift;

	fw_pg_per_umem_pg = BIT(umem->page_shift - pg_shift);

	for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) {
		pages = sg_dma_len(sg) >> shift;
		pg_addr = sg_dma_address(sg);
		for (pg_cnt = 0; pg_cnt < pages; pg_cnt++) {
			for (fw_pg_cnt = 0; fw_pg_cnt < fw_pg_per_umem_pg;) {
				pbe->lo = cpu_to_le32(pg_addr);
				pbe->hi = cpu_to_le32(upper_32_bits(pg_addr));

				pg_addr += BIT(pg_shift);
				pbe_cnt++;
				total_num_pbes++;
				pbe++;

				if (total_num_pbes == pbl_info->num_pbes)
					return;

				/* If the given pbl is full storing the pbes,
				 * move to next pbl.
				 */
				if (pbe_cnt ==
				    (pbl_info->pbl_size / sizeof(u64))) {
					pbl_tbl++;
					pbe = (struct regpair *)pbl_tbl->va;
					pbe_cnt = 0;
				}

				fw_pg_cnt++;
			}
		}
	}
}

static int qedr_copy_cq_uresp(struct qedr_dev *dev,
			      struct qedr_cq *cq, struct ib_udata *udata)
{
	struct qedr_create_cq_uresp uresp;
	int rc;

	memset(&uresp, 0, sizeof(uresp));

	uresp.db_offset = DB_ADDR_SHIFT(DQ_PWM_OFFSET_UCM_RDMA_CQ_CONS_32BIT);
	uresp.icid = cq->icid;

	rc = qedr_ib_copy_to_udata(udata, &uresp, sizeof(uresp));
	if (rc)
		DP_ERR(dev, "copy error cqid=0x%x.\n", cq->icid);

	return rc;
}

static void consume_cqe(struct qedr_cq *cq)
{
	if (cq->latest_cqe == cq->toggle_cqe)
		cq->pbl_toggle ^= RDMA_CQE_REQUESTER_TOGGLE_BIT_MASK;

	cq->latest_cqe = qed_chain_consume(&cq->pbl);
}

static inline int qedr_align_cq_entries(int entries)
{
	u64 size, aligned_size;

	/* We allocate an extra entry that we don't report to the FW. */
	size = (entries + 1) * QEDR_CQE_SIZE;
	aligned_size = ALIGN(size, PAGE_SIZE);

	return aligned_size / QEDR_CQE_SIZE;
}

static inline int qedr_init_user_queue(struct ib_ucontext *ib_ctx,
				       struct qedr_dev *dev,
				       struct qedr_userq *q,
				       u64 buf_addr, size_t buf_len,
				       int access, int dmasync)
{
	u32 fw_pages;
	int rc;

	q->buf_addr = buf_addr;
	q->buf_len = buf_len;
	q->umem = ib_umem_get(ib_ctx, q->buf_addr, q->buf_len, access, dmasync);
	if (IS_ERR(q->umem)) {
		DP_ERR(dev, "create user queue: failed ib_umem_get, got %ld\n",
		       PTR_ERR(q->umem));
		return PTR_ERR(q->umem);
	}

	fw_pages = ib_umem_page_count(q->umem) <<
	    (q->umem->page_shift - FW_PAGE_SHIFT);

	rc = qedr_prepare_pbl_tbl(dev, &q->pbl_info, fw_pages, 0);
	if (rc)
		goto err0;

	q->pbl_tbl = qedr_alloc_pbl_tbl(dev, &q->pbl_info, GFP_KERNEL);
	if (IS_ERR(q->pbl_tbl)) {
		rc = PTR_ERR(q->pbl_tbl);
		goto err0;
	}

		qedr_populate_pbls(dev, q->umem, q->pbl_tbl, &q->pbl_info,
				   FW_PAGE_SHIFT);

	return 0;

err0:
	ib_umem_release(q->umem);

	return rc;
}

static inline void qedr_init_cq_params(struct qedr_cq *cq,
				       struct qedr_ucontext *ctx,
				       struct qedr_dev *dev, int vector,
				       int chain_entries, int page_cnt,
				       u64 pbl_ptr,
				       struct qed_rdma_create_cq_in_params
				       *params)
{
	memset(params, 0, sizeof(*params));
	params->cq_handle_hi = upper_32_bits((uintptr_t)cq);
	params->cq_handle_lo = lower_32_bits((uintptr_t)cq);
	params->cnq_id = vector;
	params->cq_size = chain_entries - 1;
	params->dpi = (ctx) ? ctx->dpi : dev->dpi;
	params->pbl_num_pages = page_cnt;
	params->pbl_ptr = pbl_ptr;
	params->pbl_two_level = 0;
}

static void doorbell_cq(struct qedr_cq *cq, u32 cons, u8 flags)
{
	/* Flush data before signalling doorbell */
	wmb();
	cq->db.data.agg_flags = flags;
	cq->db.data.value = cpu_to_le32(cons);
	writeq(cq->db.raw, cq->db_addr);

	/* Make sure write would stick */
	mmiowb();
}

int qedr_arm_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags flags)
{
	struct qedr_cq *cq = get_qedr_cq(ibcq);
	unsigned long sflags;
	struct qedr_dev *dev;

	dev = get_qedr_dev(ibcq->device);

	if (cq->destroyed) {
		DP_ERR(dev,
		       "warning: arm was invoked after destroy for cq %p (icid=%d)\n",
		       cq, cq->icid);
		return -EINVAL;
	}


	if (cq->cq_type == QEDR_CQ_TYPE_GSI)
		return 0;

	spin_lock_irqsave(&cq->cq_lock, sflags);

	cq->arm_flags = 0;

	if (flags & IB_CQ_SOLICITED)
		cq->arm_flags |= DQ_UCM_ROCE_CQ_ARM_SE_CF_CMD;

	if (flags & IB_CQ_NEXT_COMP)
		cq->arm_flags |= DQ_UCM_ROCE_CQ_ARM_CF_CMD;

	doorbell_cq(cq, cq->cq_cons - 1, cq->arm_flags);

	spin_unlock_irqrestore(&cq->cq_lock, sflags);

	return 0;
}

struct ib_cq *qedr_create_cq(struct ib_device *ibdev,
			     const struct ib_cq_init_attr *attr,
			     struct ib_ucontext *ib_ctx, struct ib_udata *udata)
{
	struct qedr_ucontext *ctx = get_qedr_ucontext(ib_ctx);
	struct qed_rdma_destroy_cq_out_params destroy_oparams;
	struct qed_rdma_destroy_cq_in_params destroy_iparams;
	struct qedr_dev *dev = get_qedr_dev(ibdev);
	struct qed_rdma_create_cq_in_params params;
	struct qedr_create_cq_ureq ureq;
	int vector = attr->comp_vector;
	int entries = attr->cqe;
	struct qedr_cq *cq;
	int chain_entries;
	int page_cnt;
	u64 pbl_ptr;
	u16 icid;
	int rc;

	DP_DEBUG(dev, QEDR_MSG_INIT,
		 "create_cq: called from %s. entries=%d, vector=%d\n",
		 udata ? "User Lib" : "Kernel", entries, vector);

	if (entries > QEDR_MAX_CQES) {
		DP_ERR(dev,
		       "create cq: the number of entries %d is too high. Must be equal or below %d.\n",
		       entries, QEDR_MAX_CQES);
		return ERR_PTR(-EINVAL);
	}

	chain_entries = qedr_align_cq_entries(entries);
	chain_entries = min_t(int, chain_entries, QEDR_MAX_CQES);

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

	if (udata) {
		memset(&ureq, 0, sizeof(ureq));
		if (ib_copy_from_udata(&ureq, udata, sizeof(ureq))) {
			DP_ERR(dev,
			       "create cq: problem copying data from user space\n");
			goto err0;
		}

		if (!ureq.len) {
			DP_ERR(dev,
			       "create cq: cannot create a cq with 0 entries\n");
			goto err0;
		}

		cq->cq_type = QEDR_CQ_TYPE_USER;

		rc = qedr_init_user_queue(ib_ctx, dev, &cq->q, ureq.addr,
					  ureq.len, IB_ACCESS_LOCAL_WRITE, 1);
		if (rc)
			goto err0;

		pbl_ptr = cq->q.pbl_tbl->pa;
		page_cnt = cq->q.pbl_info.num_pbes;

		cq->ibcq.cqe = chain_entries;
	} else {
		cq->cq_type = QEDR_CQ_TYPE_KERNEL;

		rc = dev->ops->common->chain_alloc(dev->cdev,
						   QED_CHAIN_USE_TO_CONSUME,
						   QED_CHAIN_MODE_PBL,
						   QED_CHAIN_CNT_TYPE_U32,
						   chain_entries,
						   sizeof(union rdma_cqe),
						   &cq->pbl, NULL);
		if (rc)
			goto err1;

		page_cnt = qed_chain_get_page_cnt(&cq->pbl);
		pbl_ptr = qed_chain_get_pbl_phys(&cq->pbl);
		cq->ibcq.cqe = cq->pbl.capacity;
	}

	qedr_init_cq_params(cq, ctx, dev, vector, chain_entries, page_cnt,
			    pbl_ptr, &params);

	rc = dev->ops->rdma_create_cq(dev->rdma_ctx, &params, &icid);
	if (rc)
		goto err2;

	cq->icid = icid;
	cq->sig = QEDR_CQ_MAGIC_NUMBER;
	spin_lock_init(&cq->cq_lock);

	if (ib_ctx) {
		rc = qedr_copy_cq_uresp(dev, cq, udata);
		if (rc)
			goto err3;
	} else {
		/* Generate doorbell address. */
		cq->db_addr = dev->db_addr +
		    DB_ADDR_SHIFT(DQ_PWM_OFFSET_UCM_RDMA_CQ_CONS_32BIT);
		cq->db.data.icid = cq->icid;
		cq->db.data.params = DB_AGG_CMD_SET <<
		    RDMA_PWM_VAL32_DATA_AGG_CMD_SHIFT;

		/* point to the very last element, passing it we will toggle */
		cq->toggle_cqe = qed_chain_get_last_elem(&cq->pbl);
		cq->pbl_toggle = RDMA_CQE_REQUESTER_TOGGLE_BIT_MASK;
		cq->latest_cqe = NULL;
		consume_cqe(cq);
		cq->cq_cons = qed_chain_get_cons_idx_u32(&cq->pbl);
	}

	DP_DEBUG(dev, QEDR_MSG_CQ,
		 "create cq: icid=0x%0x, addr=%p, size(entries)=0x%0x\n",
		 cq->icid, cq, params.cq_size);

	return &cq->ibcq;

err3:
	destroy_iparams.icid = cq->icid;
	dev->ops->rdma_destroy_cq(dev->rdma_ctx, &destroy_iparams,
				  &destroy_oparams);
err2:
	if (udata)
		qedr_free_pbl(dev, &cq->q.pbl_info, cq->q.pbl_tbl);
	else
		dev->ops->common->chain_free(dev->cdev, &cq->pbl);
err1:
	if (udata)
		ib_umem_release(cq->q.umem);
err0:
	kfree(cq);
	return ERR_PTR(-EINVAL);
}

int qedr_resize_cq(struct ib_cq *ibcq, int new_cnt, struct ib_udata *udata)
{
	struct qedr_dev *dev = get_qedr_dev(ibcq->device);
	struct qedr_cq *cq = get_qedr_cq(ibcq);

	DP_ERR(dev, "cq %p RESIZE NOT SUPPORTED\n", cq);

	return 0;
}

#define QEDR_DESTROY_CQ_MAX_ITERATIONS		(10)
#define QEDR_DESTROY_CQ_ITER_DURATION		(10)

int qedr_destroy_cq(struct ib_cq *ibcq)
{
	struct qedr_dev *dev = get_qedr_dev(ibcq->device);
	struct qed_rdma_destroy_cq_out_params oparams;
	struct qed_rdma_destroy_cq_in_params iparams;
	struct qedr_cq *cq = get_qedr_cq(ibcq);
	int iter;
	int rc;

	DP_DEBUG(dev, QEDR_MSG_CQ, "destroy cq %p (icid=%d)\n", cq, cq->icid);

	cq->destroyed = 1;

	/* GSIs CQs are handled by driver, so they don't exist in the FW */
	if (cq->cq_type == QEDR_CQ_TYPE_GSI)
		goto done;

	iparams.icid = cq->icid;
	rc = dev->ops->rdma_destroy_cq(dev->rdma_ctx, &iparams, &oparams);
	if (rc)
		return rc;

	dev->ops->common->chain_free(dev->cdev, &cq->pbl);

	if (ibcq->uobject && ibcq->uobject->context) {
		qedr_free_pbl(dev, &cq->q.pbl_info, cq->q.pbl_tbl);
		ib_umem_release(cq->q.umem);
	}

	/* We don't want the IRQ handler to handle a non-existing CQ so we
	 * wait until all CNQ interrupts, if any, are received. This will always
	 * happen and will always happen very fast. If not, then a serious error
	 * has occured. That is why we can use a long delay.
	 * We spin for a short time so we don’t lose time on context switching
	 * in case all the completions are handled in that span. Otherwise
	 * we sleep for a while and check again. Since the CNQ may be
	 * associated with (only) the current CPU we use msleep to allow the
	 * current CPU to be freed.
	 * The CNQ notification is increased in qedr_irq_handler().
	 */
	iter = QEDR_DESTROY_CQ_MAX_ITERATIONS;
	while (oparams.num_cq_notif != READ_ONCE(cq->cnq_notif) && iter) {
		udelay(QEDR_DESTROY_CQ_ITER_DURATION);
		iter--;
	}

	iter = QEDR_DESTROY_CQ_MAX_ITERATIONS;
	while (oparams.num_cq_notif != READ_ONCE(cq->cnq_notif) && iter) {
		msleep(QEDR_DESTROY_CQ_ITER_DURATION);
		iter--;
	}

	if (oparams.num_cq_notif != cq->cnq_notif)
		goto err;

	/* Note that we don't need to have explicit code to wait for the
	 * completion of the event handler because it is invoked from the EQ.
	 * Since the destroy CQ ramrod has also been received on the EQ we can
	 * be certain that there's no event handler in process.
	 */
done:
	cq->sig = ~cq->sig;

	kfree(cq);

	return 0;

err:
	DP_ERR(dev,
	       "CQ %p (icid=%d) not freed, expecting %d ints but got %d ints\n",
	       cq, cq->icid, oparams.num_cq_notif, cq->cnq_notif);

	return -EINVAL;
}

static inline int get_gid_info_from_table(struct ib_qp *ibqp,
					  struct ib_qp_attr *attr,
					  int attr_mask,
					  struct qed_rdma_modify_qp_in_params
					  *qp_params)
{
	enum rdma_network_type nw_type;
	struct ib_gid_attr gid_attr;
	const struct ib_global_route *grh = rdma_ah_read_grh(&attr->ah_attr);
	union ib_gid gid;
	u32 ipv4_addr;
	int rc = 0;
	int i;

	rc = ib_get_cached_gid(ibqp->device,
			       rdma_ah_get_port_num(&attr->ah_attr),
			       grh->sgid_index, &gid, &gid_attr);
	if (rc)
		return rc;

	if (!memcmp(&gid, &zgid, sizeof(gid)))
		return -ENOENT;

	if (gid_attr.ndev) {
		qp_params->vlan_id = rdma_vlan_dev_vlan_id(gid_attr.ndev);

		dev_put(gid_attr.ndev);
		nw_type = ib_gid_to_network_type(gid_attr.gid_type, &gid);
		switch (nw_type) {
		case RDMA_NETWORK_IPV6:
			memcpy(&qp_params->sgid.bytes[0], &gid.raw[0],
			       sizeof(qp_params->sgid));
			memcpy(&qp_params->dgid.bytes[0],
			       &grh->dgid,
			       sizeof(qp_params->dgid));
			qp_params->roce_mode = ROCE_V2_IPV6;
			SET_FIELD(qp_params->modify_flags,
				  QED_ROCE_MODIFY_QP_VALID_ROCE_MODE, 1);
			break;
		case RDMA_NETWORK_IB:
			memcpy(&qp_params->sgid.bytes[0], &gid.raw[0],
			       sizeof(qp_params->sgid));
			memcpy(&qp_params->dgid.bytes[0],
			       &grh->dgid,
			       sizeof(qp_params->dgid));
			qp_params->roce_mode = ROCE_V1;
			break;
		case RDMA_NETWORK_IPV4:
			memset(&qp_params->sgid, 0, sizeof(qp_params->sgid));
			memset(&qp_params->dgid, 0, sizeof(qp_params->dgid));
			ipv4_addr = qedr_get_ipv4_from_gid(gid.raw);
			qp_params->sgid.ipv4_addr = ipv4_addr;
			ipv4_addr =
			    qedr_get_ipv4_from_gid(grh->dgid.raw);
			qp_params->dgid.ipv4_addr = ipv4_addr;
			SET_FIELD(qp_params->modify_flags,
				  QED_ROCE_MODIFY_QP_VALID_ROCE_MODE, 1);
			qp_params->roce_mode = ROCE_V2_IPV4;
			break;
		}
	}

	for (i = 0; i < 4; i++) {
		qp_params->sgid.dwords[i] = ntohl(qp_params->sgid.dwords[i]);
		qp_params->dgid.dwords[i] = ntohl(qp_params->dgid.dwords[i]);
	}

	if (qp_params->vlan_id >= VLAN_CFI_MASK)
		qp_params->vlan_id = 0;

	return 0;
}

static int qedr_check_qp_attrs(struct ib_pd *ibpd, struct qedr_dev *dev,
			       struct ib_qp_init_attr *attrs)
{
	struct qedr_device_attr *qattr = &dev->attr;

	/* QP0... attrs->qp_type == IB_QPT_GSI */
	if (attrs->qp_type != IB_QPT_RC && attrs->qp_type != IB_QPT_GSI) {
		DP_DEBUG(dev, QEDR_MSG_QP,
			 "create qp: unsupported qp type=0x%x requested\n",
			 attrs->qp_type);
		return -EINVAL;
	}

	if (attrs->cap.max_send_wr > qattr->max_sqe) {
		DP_ERR(dev,
		       "create qp: cannot create a SQ with %d elements (max_send_wr=0x%x)\n",
		       attrs->cap.max_send_wr, qattr->max_sqe);
		return -EINVAL;
	}

	if (attrs->cap.max_inline_data > qattr->max_inline) {
		DP_ERR(dev,
		       "create qp: unsupported inline data size=0x%x requested (max_inline=0x%x)\n",
		       attrs->cap.max_inline_data, qattr->max_inline);
		return -EINVAL;
	}

	if (attrs->cap.max_send_sge > qattr->max_sge) {
		DP_ERR(dev,
		       "create qp: unsupported send_sge=0x%x requested (max_send_sge=0x%x)\n",
		       attrs->cap.max_send_sge, qattr->max_sge);
		return -EINVAL;
	}

	if (attrs->cap.max_recv_sge > qattr->max_sge) {
		DP_ERR(dev,
		       "create qp: unsupported recv_sge=0x%x requested (max_recv_sge=0x%x)\n",
		       attrs->cap.max_recv_sge, qattr->max_sge);
		return -EINVAL;
	}

	/* Unprivileged user space cannot create special QP */
	if (ibpd->uobject && attrs->qp_type == IB_QPT_GSI) {
		DP_ERR(dev,
		       "create qp: userspace can't create special QPs of type=0x%x\n",
		       attrs->qp_type);
		return -EINVAL;
	}

	return 0;
}

static void qedr_copy_rq_uresp(struct qedr_create_qp_uresp *uresp,
			       struct qedr_qp *qp)
{
	uresp->rq_db_offset = DB_ADDR_SHIFT(DQ_PWM_OFFSET_TCM_ROCE_RQ_PROD);
	uresp->rq_icid = qp->icid;
}

static void qedr_copy_sq_uresp(struct qedr_create_qp_uresp *uresp,
			       struct qedr_qp *qp)
{
	uresp->sq_db_offset = DB_ADDR_SHIFT(DQ_PWM_OFFSET_XCM_RDMA_SQ_PROD);
	uresp->sq_icid = qp->icid + 1;
}

static int qedr_copy_qp_uresp(struct qedr_dev *dev,
			      struct qedr_qp *qp, struct ib_udata *udata)
{
	struct qedr_create_qp_uresp uresp;
	int rc;

	memset(&uresp, 0, sizeof(uresp));
	qedr_copy_sq_uresp(&uresp, qp);
	qedr_copy_rq_uresp(&uresp, qp);

	uresp.atomic_supported = dev->atomic_cap != IB_ATOMIC_NONE;
	uresp.qp_id = qp->qp_id;

	rc = qedr_ib_copy_to_udata(udata, &uresp, sizeof(uresp));
	if (rc)
		DP_ERR(dev,
		       "create qp: failed a copy to user space with qp icid=0x%x.\n",
		       qp->icid);

	return rc;
}

static void qedr_set_common_qp_params(struct qedr_dev *dev,
				      struct qedr_qp *qp,
				      struct qedr_pd *pd,
				      struct ib_qp_init_attr *attrs)
{
	spin_lock_init(&qp->q_lock);
	qp->pd = pd;
	qp->qp_type = attrs->qp_type;
	qp->max_inline_data = attrs->cap.max_inline_data;
	qp->sq.max_sges = attrs->cap.max_send_sge;
	qp->state = QED_ROCE_QP_STATE_RESET;
	qp->signaled = (attrs->sq_sig_type == IB_SIGNAL_ALL_WR) ? true : false;
	qp->sq_cq = get_qedr_cq(attrs->send_cq);
	qp->rq_cq = get_qedr_cq(attrs->recv_cq);
	qp->dev = dev;
	qp->rq.max_sges = attrs->cap.max_recv_sge;

	DP_DEBUG(dev, QEDR_MSG_QP,
		 "RQ params:\trq_max_sges = %d, rq_cq_id = %d\n",
		 qp->rq.max_sges, qp->rq_cq->icid);
	DP_DEBUG(dev, QEDR_MSG_QP,
		 "QP params:\tpd = %d, qp_type = %d, max_inline_data = %d, state = %d, signaled = %d, use_srq=%d\n",
		 pd->pd_id, qp->qp_type, qp->max_inline_data,
		 qp->state, qp->signaled, (attrs->srq) ? 1 : 0);
	DP_DEBUG(dev, QEDR_MSG_QP,
		 "SQ params:\tsq_max_sges = %d, sq_cq_id = %d\n",
		 qp->sq.max_sges, qp->sq_cq->icid);
}

static void qedr_set_roce_db_info(struct qedr_dev *dev, struct qedr_qp *qp)
{
	qp->sq.db = dev->db_addr +
		    DB_ADDR_SHIFT(DQ_PWM_OFFSET_XCM_RDMA_SQ_PROD);
	qp->sq.db_data.data.icid = qp->icid + 1;
	qp->rq.db = dev->db_addr +
		    DB_ADDR_SHIFT(DQ_PWM_OFFSET_TCM_ROCE_RQ_PROD);
	qp->rq.db_data.data.icid = qp->icid;
}

static inline void
qedr_init_common_qp_in_params(struct qedr_dev *dev,
			      struct qedr_pd *pd,
			      struct qedr_qp *qp,
			      struct ib_qp_init_attr *attrs,
			      bool fmr_and_reserved_lkey,
			      struct qed_rdma_create_qp_in_params *params)
{
	/* QP handle to be written in an async event */
	params->qp_handle_async_lo = lower_32_bits((uintptr_t) qp);
	params->qp_handle_async_hi = upper_32_bits((uintptr_t) qp);

	params->signal_all = (attrs->sq_sig_type == IB_SIGNAL_ALL_WR);
	params->fmr_and_reserved_lkey = fmr_and_reserved_lkey;
	params->pd = pd->pd_id;
	params->dpi = pd->uctx ? pd->uctx->dpi : dev->dpi;
	params->sq_cq_id = get_qedr_cq(attrs->send_cq)->icid;
	params->stats_queue = 0;
	params->rq_cq_id = get_qedr_cq(attrs->recv_cq)->icid;
	params->srq_id = 0;
	params->use_srq = false;
}

static inline void qedr_qp_user_print(struct qedr_dev *dev, struct qedr_qp *qp)
{
	DP_DEBUG(dev, QEDR_MSG_QP, "create qp: successfully created user QP. "
		 "qp=%p. "
		 "sq_addr=0x%llx, "
		 "sq_len=%zd, "
		 "rq_addr=0x%llx, "
		 "rq_len=%zd"
		 "\n",
		 qp,
		 qp->usq.buf_addr,
		 qp->usq.buf_len, qp->urq.buf_addr, qp->urq.buf_len);
}

static void qedr_cleanup_user(struct qedr_dev *dev, struct qedr_qp *qp)
{
	if (qp->usq.umem)
		ib_umem_release(qp->usq.umem);
	qp->usq.umem = NULL;

	if (qp->urq.umem)
		ib_umem_release(qp->urq.umem);
	qp->urq.umem = NULL;
}

static int qedr_create_user_qp(struct qedr_dev *dev,
			       struct qedr_qp *qp,
			       struct ib_pd *ibpd,
			       struct ib_udata *udata,
			       struct ib_qp_init_attr *attrs)
{
	struct qed_rdma_create_qp_in_params in_params;
	struct qed_rdma_create_qp_out_params out_params;
	struct qedr_pd *pd = get_qedr_pd(ibpd);
	struct ib_ucontext *ib_ctx = NULL;
	struct qedr_ucontext *ctx = NULL;
	struct qedr_create_qp_ureq ureq;
	int rc = -EINVAL;

	ib_ctx = ibpd->uobject->context;
	ctx = get_qedr_ucontext(ib_ctx);

	memset(&ureq, 0, sizeof(ureq));
	rc = ib_copy_from_udata(&ureq, udata, sizeof(ureq));
	if (rc) {
		DP_ERR(dev, "Problem copying data from user space\n");
		return rc;
	}

	/* SQ - read access only (0), dma sync not required (0) */
	rc = qedr_init_user_queue(ib_ctx, dev, &qp->usq, ureq.sq_addr,
				  ureq.sq_len, 0, 0);
	if (rc)
		return rc;

	/* RQ - read access only (0), dma sync not required (0) */
	rc = qedr_init_user_queue(ib_ctx, dev, &qp->urq, ureq.rq_addr,
				  ureq.rq_len, 0, 0);

	if (rc)
		return rc;

	memset(&in_params, 0, sizeof(in_params));
	qedr_init_common_qp_in_params(dev, pd, qp, attrs, false, &in_params);
	in_params.qp_handle_lo = ureq.qp_handle_lo;
	in_params.qp_handle_hi = ureq.qp_handle_hi;
	in_params.sq_num_pages = qp->usq.pbl_info.num_pbes;
	in_params.sq_pbl_ptr = qp->usq.pbl_tbl->pa;
	in_params.rq_num_pages = qp->urq.pbl_info.num_pbes;
	in_params.rq_pbl_ptr = qp->urq.pbl_tbl->pa;

	qp->qed_qp = dev->ops->rdma_create_qp(dev->rdma_ctx,
					      &in_params, &out_params);

	if (!qp->qed_qp) {
		rc = -ENOMEM;
		goto err1;
	}

	qp->qp_id = out_params.qp_id;
	qp->icid = out_params.icid;

	rc = qedr_copy_qp_uresp(dev, qp, udata);
	if (rc)
		goto err;

	qedr_qp_user_print(dev, qp);

	return 0;
err:
	rc = dev->ops->rdma_destroy_qp(dev->rdma_ctx, qp->qed_qp);
	if (rc)
		DP_ERR(dev, "create qp: fatal fault. rc=%d", rc);

err1:
	qedr_cleanup_user(dev, qp);
	return rc;
}

static int
qedr_roce_create_kernel_qp(struct qedr_dev *dev,
			   struct qedr_qp *qp,
			   struct qed_rdma_create_qp_in_params *in_params,
			   u32 n_sq_elems, u32 n_rq_elems)
{
	struct qed_rdma_create_qp_out_params out_params;
	int rc;

	rc = dev->ops->common->chain_alloc(dev->cdev,
					   QED_CHAIN_USE_TO_PRODUCE,
					   QED_CHAIN_MODE_PBL,
					   QED_CHAIN_CNT_TYPE_U32,
					   n_sq_elems,
					   QEDR_SQE_ELEMENT_SIZE,
					   &qp->sq.pbl, NULL);

	if (rc)
		return rc;

	in_params->sq_num_pages = qed_chain_get_page_cnt(&qp->sq.pbl);
	in_params->sq_pbl_ptr = qed_chain_get_pbl_phys(&qp->sq.pbl);

	rc = dev->ops->common->chain_alloc(dev->cdev,
					   QED_CHAIN_USE_TO_CONSUME_PRODUCE,
					   QED_CHAIN_MODE_PBL,
					   QED_CHAIN_CNT_TYPE_U32,
					   n_rq_elems,
					   QEDR_RQE_ELEMENT_SIZE,
					   &qp->rq.pbl, NULL);
	if (rc)
		return rc;

	in_params->rq_num_pages = qed_chain_get_page_cnt(&qp->rq.pbl);
	in_params->rq_pbl_ptr = qed_chain_get_pbl_phys(&qp->rq.pbl);

	qp->qed_qp = dev->ops->rdma_create_qp(dev->rdma_ctx,
					      in_params, &out_params);

	if (!qp->qed_qp)
		return -EINVAL;

	qp->qp_id = out_params.qp_id;
	qp->icid = out_params.icid;

	qedr_set_roce_db_info(dev, qp);

	return 0;
}

static void qedr_cleanup_kernel(struct qedr_dev *dev, struct qedr_qp *qp)
{
	dev->ops->common->chain_free(dev->cdev, &qp->sq.pbl);
	kfree(qp->wqe_wr_id);

	dev->ops->common->chain_free(dev->cdev, &qp->rq.pbl);
	kfree(qp->rqe_wr_id);
}

static int qedr_create_kernel_qp(struct qedr_dev *dev,
				 struct qedr_qp *qp,
				 struct ib_pd *ibpd,
				 struct ib_qp_init_attr *attrs)
{
	struct qed_rdma_create_qp_in_params in_params;
	struct qedr_pd *pd = get_qedr_pd(ibpd);
	int rc = -EINVAL;
	u32 n_rq_elems;
	u32 n_sq_elems;
	u32 n_sq_entries;

	memset(&in_params, 0, sizeof(in_params));

	/* A single work request may take up to QEDR_MAX_SQ_WQE_SIZE elements in
	 * the ring. The ring should allow at least a single WR, even if the
	 * user requested none, due to allocation issues.
	 * We should add an extra WR since the prod and cons indices of
	 * wqe_wr_id are managed in such a way that the WQ is considered full
	 * when (prod+1)%max_wr==cons. We currently don't do that because we
	 * double the number of entries due an iSER issue that pushes far more
	 * WRs than indicated. If we decline its ib_post_send() then we get
	 * error prints in the dmesg we'd like to avoid.
	 */
	qp->sq.max_wr = min_t(u32, attrs->cap.max_send_wr * dev->wq_multiplier,
			      dev->attr.max_sqe);

	qp->wqe_wr_id = kzalloc(qp->sq.max_wr * sizeof(*qp->wqe_wr_id),
				GFP_KERNEL);
	if (!qp->wqe_wr_id) {
		DP_ERR(dev, "create qp: failed SQ shadow memory allocation\n");
		return -ENOMEM;
	}

	/* QP handle to be written in CQE */
	in_params.qp_handle_lo = lower_32_bits((uintptr_t) qp);
	in_params.qp_handle_hi = upper_32_bits((uintptr_t) qp);

	/* A single work request may take up to QEDR_MAX_RQ_WQE_SIZE elements in
	 * the ring. There ring should allow at least a single WR, even if the
	 * user requested none, due to allocation issues.
	 */
	qp->rq.max_wr = (u16) max_t(u32, attrs->cap.max_recv_wr, 1);

	/* Allocate driver internal RQ array */
	qp->rqe_wr_id = kzalloc(qp->rq.max_wr * sizeof(*qp->rqe_wr_id),
				GFP_KERNEL);
	if (!qp->rqe_wr_id) {
		DP_ERR(dev,
		       "create qp: failed RQ shadow memory allocation\n");
		kfree(qp->wqe_wr_id);
		return -ENOMEM;
	}

	qedr_init_common_qp_in_params(dev, pd, qp, attrs, true, &in_params);

	n_sq_entries = attrs->cap.max_send_wr;
	n_sq_entries = min_t(u32, n_sq_entries, dev->attr.max_sqe);
	n_sq_entries = max_t(u32, n_sq_entries, 1);
	n_sq_elems = n_sq_entries * QEDR_MAX_SQE_ELEMENTS_PER_SQE;

	n_rq_elems = qp->rq.max_wr * QEDR_MAX_RQE_ELEMENTS_PER_RQE;

	rc = qedr_roce_create_kernel_qp(dev, qp, &in_params,
					n_sq_elems, n_rq_elems);
	if (rc)
		qedr_cleanup_kernel(dev, qp);

	return rc;
}

struct ib_qp *qedr_create_qp(struct ib_pd *ibpd,
			     struct ib_qp_init_attr *attrs,
			     struct ib_udata *udata)
{
	struct qedr_dev *dev = get_qedr_dev(ibpd->device);
	struct qedr_pd *pd = get_qedr_pd(ibpd);
	struct qedr_qp *qp;
	struct ib_qp *ibqp;
	int rc = 0;

	DP_DEBUG(dev, QEDR_MSG_QP, "create qp: called from %s, pd=%p\n",
		 udata ? "user library" : "kernel", pd);

	rc = qedr_check_qp_attrs(ibpd, dev, attrs);
	if (rc)
		return ERR_PTR(rc);

	if (attrs->srq)
		return ERR_PTR(-EINVAL);

	DP_DEBUG(dev, QEDR_MSG_QP,
		 "create qp: called from %s, event_handler=%p, eepd=%p sq_cq=%p, sq_icid=%d, rq_cq=%p, rq_icid=%d\n",
		 udata ? "user library" : "kernel", attrs->event_handler, pd,
		 get_qedr_cq(attrs->send_cq),
		 get_qedr_cq(attrs->send_cq)->icid,
		 get_qedr_cq(attrs->recv_cq),
		 get_qedr_cq(attrs->recv_cq)->icid);

	qp = kzalloc(sizeof(*qp), GFP_KERNEL);
	if (!qp) {
		DP_ERR(dev, "create qp: failed allocating memory\n");
		return ERR_PTR(-ENOMEM);
	}

	qedr_set_common_qp_params(dev, qp, pd, attrs);

	if (attrs->qp_type == IB_QPT_GSI) {
		ibqp = qedr_create_gsi_qp(dev, attrs, qp);
		if (IS_ERR(ibqp))
			kfree(qp);
		return ibqp;
	}

	if (udata)
		rc = qedr_create_user_qp(dev, qp, ibpd, udata, attrs);
	else
		rc = qedr_create_kernel_qp(dev, qp, ibpd, attrs);

	if (rc)
		goto err;

	qp->ibqp.qp_num = qp->qp_id;

	return &qp->ibqp;

err:
	kfree(qp);

	return ERR_PTR(-EFAULT);
}

static enum ib_qp_state qedr_get_ibqp_state(enum qed_roce_qp_state qp_state)
{
	switch (qp_state) {
	case QED_ROCE_QP_STATE_RESET:
		return IB_QPS_RESET;
	case QED_ROCE_QP_STATE_INIT:
		return IB_QPS_INIT;
	case QED_ROCE_QP_STATE_RTR:
		return IB_QPS_RTR;
	case QED_ROCE_QP_STATE_RTS:
		return IB_QPS_RTS;
	case QED_ROCE_QP_STATE_SQD:
		return IB_QPS_SQD;
	case QED_ROCE_QP_STATE_ERR:
		return IB_QPS_ERR;
	case QED_ROCE_QP_STATE_SQE:
		return IB_QPS_SQE;
	}
	return IB_QPS_ERR;
}

static enum qed_roce_qp_state qedr_get_state_from_ibqp(
					enum ib_qp_state qp_state)
{
	switch (qp_state) {
	case IB_QPS_RESET:
		return QED_ROCE_QP_STATE_RESET;
	case IB_QPS_INIT:
		return QED_ROCE_QP_STATE_INIT;
	case IB_QPS_RTR:
		return QED_ROCE_QP_STATE_RTR;
	case IB_QPS_RTS:
		return QED_ROCE_QP_STATE_RTS;
	case IB_QPS_SQD:
		return QED_ROCE_QP_STATE_SQD;
	case IB_QPS_ERR:
		return QED_ROCE_QP_STATE_ERR;
	default:
		return QED_ROCE_QP_STATE_ERR;
	}
}

static void qedr_reset_qp_hwq_info(struct qedr_qp_hwq_info *qph)
{
	qed_chain_reset(&qph->pbl);
	qph->prod = 0;
	qph->cons = 0;
	qph->wqe_cons = 0;
	qph->db_data.data.value = cpu_to_le16(0);
}

static int qedr_update_qp_state(struct qedr_dev *dev,
				struct qedr_qp *qp,
				enum qed_roce_qp_state new_state)
{
	int status = 0;

	if (new_state == qp->state)
		return 0;

	switch (qp->state) {
	case QED_ROCE_QP_STATE_RESET:
		switch (new_state) {
		case QED_ROCE_QP_STATE_INIT:
			qp->prev_wqe_size = 0;
			qedr_reset_qp_hwq_info(&qp->sq);
			qedr_reset_qp_hwq_info(&qp->rq);
			break;
		default:
			status = -EINVAL;
			break;
		};
		break;
	case QED_ROCE_QP_STATE_INIT:
		switch (new_state) {
		case QED_ROCE_QP_STATE_RTR:
			/* Update doorbell (in case post_recv was
			 * done before move to RTR)
			 */
			wmb();
			writel(qp->rq.db_data.raw, qp->rq.db);
			/* Make sure write takes effect */
			mmiowb();
			break;
		case QED_ROCE_QP_STATE_ERR:
			break;
		default:
			/* Invalid state change. */
			status = -EINVAL;
			break;
		};
		break;
	case QED_ROCE_QP_STATE_RTR:
		/* RTR->XXX */
		switch (new_state) {
		case QED_ROCE_QP_STATE_RTS:
			break;
		case QED_ROCE_QP_STATE_ERR:
			break;
		default:
			/* Invalid state change. */
			status = -EINVAL;
			break;
		};
		break;
	case QED_ROCE_QP_STATE_RTS:
		/* RTS->XXX */
		switch (new_state) {
		case QED_ROCE_QP_STATE_SQD:
			break;
		case QED_ROCE_QP_STATE_ERR:
			break;
		default:
			/* Invalid state change. */
			status = -EINVAL;
			break;
		};
		break;
	case QED_ROCE_QP_STATE_SQD:
		/* SQD->XXX */
		switch (new_state) {
		case QED_ROCE_QP_STATE_RTS:
		case QED_ROCE_QP_STATE_ERR:
			break;
		default:
			/* Invalid state change. */
			status = -EINVAL;
			break;
		};
		break;
	case QED_ROCE_QP_STATE_ERR:
		/* ERR->XXX */
		switch (new_state) {
		case QED_ROCE_QP_STATE_RESET:
			if ((qp->rq.prod != qp->rq.cons) ||
			    (qp->sq.prod != qp->sq.cons)) {
				DP_NOTICE(dev,
					  "Error->Reset with rq/sq not empty rq.prod=%x rq.cons=%x sq.prod=%x sq.cons=%x\n",
					  qp->rq.prod, qp->rq.cons, qp->sq.prod,
					  qp->sq.cons);
				status = -EINVAL;
			}
			break;
		default:
			status = -EINVAL;
			break;
		};
		break;
	default:
		status = -EINVAL;
		break;
	};

	return status;
}

int qedr_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
		   int attr_mask, struct ib_udata *udata)
{
	struct qedr_qp *qp = get_qedr_qp(ibqp);
	struct qed_rdma_modify_qp_in_params qp_params = { 0 };
	struct qedr_dev *dev = get_qedr_dev(&qp->dev->ibdev);
	const struct ib_global_route *grh = rdma_ah_read_grh(&attr->ah_attr);
	enum ib_qp_state old_qp_state, new_qp_state;
	int rc = 0;

	DP_DEBUG(dev, QEDR_MSG_QP,
		 "modify qp: qp %p attr_mask=0x%x, state=%d", qp, attr_mask,
		 attr->qp_state);

	old_qp_state = qedr_get_ibqp_state(qp->state);
	if (attr_mask & IB_QP_STATE)
		new_qp_state = attr->qp_state;
	else
		new_qp_state = old_qp_state;

	if (!ib_modify_qp_is_ok
	    (old_qp_state, new_qp_state, ibqp->qp_type, attr_mask,
	     IB_LINK_LAYER_ETHERNET)) {
		DP_ERR(dev,
		       "modify qp: invalid attribute mask=0x%x specified for\n"
		       "qpn=0x%x of type=0x%x old_qp_state=0x%x, new_qp_state=0x%x\n",
		       attr_mask, qp->qp_id, ibqp->qp_type, old_qp_state,
		       new_qp_state);
		rc = -EINVAL;
		goto err;
	}

	/* Translate the masks... */
	if (attr_mask & IB_QP_STATE) {
		SET_FIELD(qp_params.modify_flags,
			  QED_RDMA_MODIFY_QP_VALID_NEW_STATE, 1);
		qp_params.new_state = qedr_get_state_from_ibqp(attr->qp_state);
	}

	if (attr_mask & IB_QP_EN_SQD_ASYNC_NOTIFY)
		qp_params.sqd_async = true;

	if (attr_mask & IB_QP_PKEY_INDEX) {
		SET_FIELD(qp_params.modify_flags,
			  QED_ROCE_MODIFY_QP_VALID_PKEY, 1);
		if (attr->pkey_index >= QEDR_ROCE_PKEY_TABLE_LEN) {
			rc = -EINVAL;
			goto err;
		}

		qp_params.pkey = QEDR_ROCE_PKEY_DEFAULT;
	}

	if (attr_mask & IB_QP_QKEY)
		qp->qkey = attr->qkey;

	if (attr_mask & IB_QP_ACCESS_FLAGS) {
		SET_FIELD(qp_params.modify_flags,
			  QED_RDMA_MODIFY_QP_VALID_RDMA_OPS_EN, 1);
		qp_params.incoming_rdma_read_en = attr->qp_access_flags &
						  IB_ACCESS_REMOTE_READ;
		qp_params.incoming_rdma_write_en = attr->qp_access_flags &
						   IB_ACCESS_REMOTE_WRITE;
		qp_params.incoming_atomic_en = attr->qp_access_flags &
					       IB_ACCESS_REMOTE_ATOMIC;
	}

	if (attr_mask & (IB_QP_AV | IB_QP_PATH_MTU)) {
		if (attr_mask & IB_QP_PATH_MTU) {
			if (attr->path_mtu < IB_MTU_256 ||
			    attr->path_mtu > IB_MTU_4096) {
				pr_err("error: Only MTU sizes of 256, 512, 1024, 2048 and 4096 are supported by RoCE\n");
				rc = -EINVAL;
				goto err;
			}
			qp->mtu = min(ib_mtu_enum_to_int(attr->path_mtu),
				      ib_mtu_enum_to_int(iboe_get_mtu
							 (dev->ndev->mtu)));
		}

		if (!qp->mtu) {
			qp->mtu =
			ib_mtu_enum_to_int(iboe_get_mtu(dev->ndev->mtu));
			pr_err("Fixing zeroed MTU to qp->mtu = %d\n", qp->mtu);
		}

		SET_FIELD(qp_params.modify_flags,
			  QED_ROCE_MODIFY_QP_VALID_ADDRESS_VECTOR, 1);

		qp_params.traffic_class_tos = grh->traffic_class;
		qp_params.flow_label = grh->flow_label;
		qp_params.hop_limit_ttl = grh->hop_limit;

		qp->sgid_idx = grh->sgid_index;

		rc = get_gid_info_from_table(ibqp, attr, attr_mask, &qp_params);
		if (rc) {
			DP_ERR(dev,
			       "modify qp: problems with GID index %d (rc=%d)\n",
			       grh->sgid_index, rc);
			return rc;
		}

		rc = qedr_get_dmac(dev, &attr->ah_attr,
				   qp_params.remote_mac_addr);
		if (rc)
			return rc;

		qp_params.use_local_mac = true;
		ether_addr_copy(qp_params.local_mac_addr, dev->ndev->dev_addr);

		DP_DEBUG(dev, QEDR_MSG_QP, "dgid=%x:%x:%x:%x\n",
			 qp_params.dgid.dwords[0], qp_params.dgid.dwords[1],
			 qp_params.dgid.dwords[2], qp_params.dgid.dwords[3]);
		DP_DEBUG(dev, QEDR_MSG_QP, "sgid=%x:%x:%x:%x\n",
			 qp_params.sgid.dwords[0], qp_params.sgid.dwords[1],
			 qp_params.sgid.dwords[2], qp_params.sgid.dwords[3]);
		DP_DEBUG(dev, QEDR_MSG_QP, "remote_mac=[%pM]\n",
			 qp_params.remote_mac_addr);

		qp_params.mtu = qp->mtu;
		qp_params.lb_indication = false;
	}

	if (!qp_params.mtu) {
		/* Stay with current MTU */
		if (qp->mtu)
			qp_params.mtu = qp->mtu;
		else
			qp_params.mtu =
			    ib_mtu_enum_to_int(iboe_get_mtu(dev->ndev->mtu));
	}

	if (attr_mask & IB_QP_TIMEOUT) {
		SET_FIELD(qp_params.modify_flags,
			  QED_ROCE_MODIFY_QP_VALID_ACK_TIMEOUT, 1);

		qp_params.ack_timeout = attr->timeout;
		if (attr->timeout) {
			u32 temp;

			temp = 4096 * (1UL << attr->timeout) / 1000 / 1000;
			/* FW requires [msec] */
			qp_params.ack_timeout = temp;
		} else {
			/* Infinite */
			qp_params.ack_timeout = 0;
		}
	}
	if (attr_mask & IB_QP_RETRY_CNT) {
		SET_FIELD(qp_params.modify_flags,
			  QED_ROCE_MODIFY_QP_VALID_RETRY_CNT, 1);
		qp_params.retry_cnt = attr->retry_cnt;
	}

	if (attr_mask & IB_QP_RNR_RETRY) {
		SET_FIELD(qp_params.modify_flags,
			  QED_ROCE_MODIFY_QP_VALID_RNR_RETRY_CNT, 1);
		qp_params.rnr_retry_cnt = attr->rnr_retry;
	}

	if (attr_mask & IB_QP_RQ_PSN) {
		SET_FIELD(qp_params.modify_flags,
			  QED_ROCE_MODIFY_QP_VALID_RQ_PSN, 1);
		qp_params.rq_psn = attr->rq_psn;
		qp->rq_psn = attr->rq_psn;
	}

	if (attr_mask & IB_QP_MAX_QP_RD_ATOMIC) {
		if (attr->max_rd_atomic > dev->attr.max_qp_req_rd_atomic_resc) {
			rc = -EINVAL;
			DP_ERR(dev,
			       "unsupported max_rd_atomic=%d, supported=%d\n",
			       attr->max_rd_atomic,
			       dev->attr.max_qp_req_rd_atomic_resc);
			goto err;
		}

		SET_FIELD(qp_params.modify_flags,
			  QED_RDMA_MODIFY_QP_VALID_MAX_RD_ATOMIC_REQ, 1);
		qp_params.max_rd_atomic_req = attr->max_rd_atomic;
	}

	if (attr_mask & IB_QP_MIN_RNR_TIMER) {
		SET_FIELD(qp_params.modify_flags,
			  QED_ROCE_MODIFY_QP_VALID_MIN_RNR_NAK_TIMER, 1);
		qp_params.min_rnr_nak_timer = attr->min_rnr_timer;
	}

	if (attr_mask & IB_QP_SQ_PSN) {
		SET_FIELD(qp_params.modify_flags,
			  QED_ROCE_MODIFY_QP_VALID_SQ_PSN, 1);
		qp_params.sq_psn = attr->sq_psn;
		qp->sq_psn = attr->sq_psn;
	}

	if (attr_mask & IB_QP_MAX_DEST_RD_ATOMIC) {
		if (attr->max_dest_rd_atomic >
		    dev->attr.max_qp_resp_rd_atomic_resc) {
			DP_ERR(dev,
			       "unsupported max_dest_rd_atomic=%d, supported=%d\n",
			       attr->max_dest_rd_atomic,
			       dev->attr.max_qp_resp_rd_atomic_resc);

			rc = -EINVAL;
			goto err;
		}

		SET_FIELD(qp_params.modify_flags,
			  QED_RDMA_MODIFY_QP_VALID_MAX_RD_ATOMIC_RESP, 1);
		qp_params.max_rd_atomic_resp = attr->max_dest_rd_atomic;
	}

	if (attr_mask & IB_QP_DEST_QPN) {
		SET_FIELD(qp_params.modify_flags,
			  QED_ROCE_MODIFY_QP_VALID_DEST_QP, 1);

		qp_params.dest_qp = attr->dest_qp_num;
		qp->dest_qp_num = attr->dest_qp_num;
	}

	if (qp->qp_type != IB_QPT_GSI)
		rc = dev->ops->rdma_modify_qp(dev->rdma_ctx,
					      qp->qed_qp, &qp_params);

	if (attr_mask & IB_QP_STATE) {
		if ((qp->qp_type != IB_QPT_GSI) && (!udata))
			rc = qedr_update_qp_state(dev, qp, qp_params.new_state);
		qp->state = qp_params.new_state;
	}

err:
	return rc;
}

static int qedr_to_ib_qp_acc_flags(struct qed_rdma_query_qp_out_params *params)
{
	int ib_qp_acc_flags = 0;

	if (params->incoming_rdma_write_en)
		ib_qp_acc_flags |= IB_ACCESS_REMOTE_WRITE;
	if (params->incoming_rdma_read_en)
		ib_qp_acc_flags |= IB_ACCESS_REMOTE_READ;
	if (params->incoming_atomic_en)
		ib_qp_acc_flags |= IB_ACCESS_REMOTE_ATOMIC;
	ib_qp_acc_flags |= IB_ACCESS_LOCAL_WRITE;
	return ib_qp_acc_flags;
}

int qedr_query_qp(struct ib_qp *ibqp,
		  struct ib_qp_attr *qp_attr,
		  int attr_mask, struct ib_qp_init_attr *qp_init_attr)
{
	struct qed_rdma_query_qp_out_params params;
	struct qedr_qp *qp = get_qedr_qp(ibqp);
	struct qedr_dev *dev = qp->dev;
	int rc = 0;

	memset(&params, 0, sizeof(params));

	rc = dev->ops->rdma_query_qp(dev->rdma_ctx, qp->qed_qp, &params);
	if (rc)
		goto err;

	memset(qp_attr, 0, sizeof(*qp_attr));
	memset(qp_init_attr, 0, sizeof(*qp_init_attr));

	qp_attr->qp_state = qedr_get_ibqp_state(params.state);
	qp_attr->cur_qp_state = qedr_get_ibqp_state(params.state);
	qp_attr->path_mtu = ib_mtu_int_to_enum(params.mtu);
	qp_attr->path_mig_state = IB_MIG_MIGRATED;
	qp_attr->rq_psn = params.rq_psn;
	qp_attr->sq_psn = params.sq_psn;
	qp_attr->dest_qp_num = params.dest_qp;

	qp_attr->qp_access_flags = qedr_to_ib_qp_acc_flags(&params);

	qp_attr->cap.max_send_wr = qp->sq.max_wr;
	qp_attr->cap.max_recv_wr = qp->rq.max_wr;
	qp_attr->cap.max_send_sge = qp->sq.max_sges;
	qp_attr->cap.max_recv_sge = qp->rq.max_sges;
	qp_attr->cap.max_inline_data = ROCE_REQ_MAX_INLINE_DATA_SIZE;
	qp_init_attr->cap = qp_attr->cap;

	qp_attr->ah_attr.type = RDMA_AH_ATTR_TYPE_ROCE;
	rdma_ah_set_grh(&qp_attr->ah_attr, NULL,
			params.flow_label, qp->sgid_idx,
			params.hop_limit_ttl, params.traffic_class_tos);
	rdma_ah_set_dgid_raw(&qp_attr->ah_attr, &params.dgid.bytes[0]);
	rdma_ah_set_port_num(&qp_attr->ah_attr, 1);
	rdma_ah_set_sl(&qp_attr->ah_attr, 0);
	qp_attr->timeout = params.timeout;
	qp_attr->rnr_retry = params.rnr_retry;
	qp_attr->retry_cnt = params.retry_cnt;
	qp_attr->min_rnr_timer = params.min_rnr_nak_timer;
	qp_attr->pkey_index = params.pkey_index;
	qp_attr->port_num = 1;
	rdma_ah_set_path_bits(&qp_attr->ah_attr, 0);
	rdma_ah_set_static_rate(&qp_attr->ah_attr, 0);
	qp_attr->alt_pkey_index = 0;
	qp_attr->alt_port_num = 0;
	qp_attr->alt_timeout = 0;
	memset(&qp_attr->alt_ah_attr, 0, sizeof(qp_attr->alt_ah_attr));

	qp_attr->sq_draining = (params.state == QED_ROCE_QP_STATE_SQD) ? 1 : 0;
	qp_attr->max_dest_rd_atomic = params.max_dest_rd_atomic;
	qp_attr->max_rd_atomic = params.max_rd_atomic;
	qp_attr->en_sqd_async_notify = (params.sqd_async) ? 1 : 0;

	DP_DEBUG(dev, QEDR_MSG_QP, "QEDR_QUERY_QP: max_inline_data=%d\n",
		 qp_attr->cap.max_inline_data);

err:
	return rc;
}

int qedr_free_qp_resources(struct qedr_dev *dev, struct qedr_qp *qp)
{
	int rc = 0;

	if (qp->qp_type != IB_QPT_GSI) {
		rc = dev->ops->rdma_destroy_qp(dev->rdma_ctx, qp->qed_qp);
		if (rc)
			return rc;
	}

	if (qp->ibqp.uobject && qp->ibqp.uobject->context)
		qedr_cleanup_user(dev, qp);
	else
		qedr_cleanup_kernel(dev, qp);

	return 0;
}

int qedr_destroy_qp(struct ib_qp *ibqp)
{
	struct qedr_qp *qp = get_qedr_qp(ibqp);
	struct qedr_dev *dev = qp->dev;
	struct ib_qp_attr attr;
	int attr_mask = 0;
	int rc = 0;

	DP_DEBUG(dev, QEDR_MSG_QP, "destroy qp: destroying %p, qp type=%d\n",
		 qp, qp->qp_type);

	if ((qp->state != QED_ROCE_QP_STATE_RESET) &&
	    (qp->state != QED_ROCE_QP_STATE_ERR) &&
	    (qp->state != QED_ROCE_QP_STATE_INIT)) {

		attr.qp_state = IB_QPS_ERR;
		attr_mask |= IB_QP_STATE;

		/* Change the QP state to ERROR */
		qedr_modify_qp(ibqp, &attr, attr_mask, NULL);
	}

	if (qp->qp_type == IB_QPT_GSI)
		qedr_destroy_gsi_qp(dev);

	qedr_free_qp_resources(dev, qp);

	kfree(qp);

	return rc;
}

struct ib_ah *qedr_create_ah(struct ib_pd *ibpd, struct rdma_ah_attr *attr,
			     struct ib_udata *udata)
{
	struct qedr_ah *ah;

	ah = kzalloc(sizeof(*ah), GFP_ATOMIC);
	if (!ah)
		return ERR_PTR(-ENOMEM);

	ah->attr = *attr;

	return &ah->ibah;
}

int qedr_destroy_ah(struct ib_ah *ibah)
{
	struct qedr_ah *ah = get_qedr_ah(ibah);

	kfree(ah);
	return 0;
}

static void free_mr_info(struct qedr_dev *dev, struct mr_info *info)
{
	struct qedr_pbl *pbl, *tmp;

	if (info->pbl_table)
		list_add_tail(&info->pbl_table->list_entry,
			      &info->free_pbl_list);

	if (!list_empty(&info->inuse_pbl_list))
		list_splice(&info->inuse_pbl_list, &info->free_pbl_list);

	list_for_each_entry_safe(pbl, tmp, &info->free_pbl_list, list_entry) {
		list_del(&pbl->list_entry);
		qedr_free_pbl(dev, &info->pbl_info, pbl);
	}
}

static int init_mr_info(struct qedr_dev *dev, struct mr_info *info,
			size_t page_list_len, bool two_layered)
{
	struct qedr_pbl *tmp;
	int rc;

	INIT_LIST_HEAD(&info->free_pbl_list);
	INIT_LIST_HEAD(&info->inuse_pbl_list);

	rc = qedr_prepare_pbl_tbl(dev, &info->pbl_info,
				  page_list_len, two_layered);
	if (rc)
		goto done;

	info->pbl_table = qedr_alloc_pbl_tbl(dev, &info->pbl_info, GFP_KERNEL);
	if (IS_ERR(info->pbl_table)) {
		rc = PTR_ERR(info->pbl_table);
		goto done;
	}

	DP_DEBUG(dev, QEDR_MSG_MR, "pbl_table_pa = %pa\n",
		 &info->pbl_table->pa);

	/* in usual case we use 2 PBLs, so we add one to free
	 * list and allocating another one
	 */
	tmp = qedr_alloc_pbl_tbl(dev, &info->pbl_info, GFP_KERNEL);
	if (IS_ERR(tmp)) {
		DP_DEBUG(dev, QEDR_MSG_MR, "Extra PBL is not allocated\n");
		goto done;
	}

	list_add_tail(&tmp->list_entry, &info->free_pbl_list);

	DP_DEBUG(dev, QEDR_MSG_MR, "extra pbl_table_pa = %pa\n", &tmp->pa);

done:
	if (rc)
		free_mr_info(dev, info);

	return rc;
}

struct ib_mr *qedr_reg_user_mr(struct ib_pd *ibpd, u64 start, u64 len,
			       u64 usr_addr, int acc, struct ib_udata *udata)
{
	struct qedr_dev *dev = get_qedr_dev(ibpd->device);
	struct qedr_mr *mr;
	struct qedr_pd *pd;
	int rc = -ENOMEM;

	pd = get_qedr_pd(ibpd);
	DP_DEBUG(dev, QEDR_MSG_MR,
		 "qedr_register user mr pd = %d start = %lld, len = %lld, usr_addr = %lld, acc = %d\n",
		 pd->pd_id, start, len, usr_addr, acc);

	if (acc & IB_ACCESS_REMOTE_WRITE && !(acc & IB_ACCESS_LOCAL_WRITE))
		return ERR_PTR(-EINVAL);

	mr = kzalloc(sizeof(*mr), GFP_KERNEL);
	if (!mr)
		return ERR_PTR(rc);

	mr->type = QEDR_MR_USER;

	mr->umem = ib_umem_get(ibpd->uobject->context, start, len, acc, 0);
	if (IS_ERR(mr->umem)) {
		rc = -EFAULT;
		goto err0;
	}

	rc = init_mr_info(dev, &mr->info, ib_umem_page_count(mr->umem), 1);
	if (rc)
		goto err1;

	qedr_populate_pbls(dev, mr->umem, mr->info.pbl_table,
			   &mr->info.pbl_info, mr->umem->page_shift);

	rc = dev->ops->rdma_alloc_tid(dev->rdma_ctx, &mr->hw_mr.itid);
	if (rc) {
		DP_ERR(dev, "roce alloc tid returned an error %d\n", rc);
		goto err1;
	}

	/* Index only, 18 bit long, lkey = itid << 8 | key */
	mr->hw_mr.tid_type = QED_RDMA_TID_REGISTERED_MR;
	mr->hw_mr.key = 0;
	mr->hw_mr.pd = pd->pd_id;
	mr->hw_mr.local_read = 1;
	mr->hw_mr.local_write = (acc & IB_ACCESS_LOCAL_WRITE) ? 1 : 0;
	mr->hw_mr.remote_read = (acc & IB_ACCESS_REMOTE_READ) ? 1 : 0;
	mr->hw_mr.remote_write = (acc & IB_ACCESS_REMOTE_WRITE) ? 1 : 0;
	mr->hw_mr.remote_atomic = (acc & IB_ACCESS_REMOTE_ATOMIC) ? 1 : 0;
	mr->hw_mr.mw_bind = false;
	mr->hw_mr.pbl_ptr = mr->info.pbl_table[0].pa;
	mr->hw_mr.pbl_two_level = mr->info.pbl_info.two_layered;
	mr->hw_mr.pbl_page_size_log = ilog2(mr->info.pbl_info.pbl_size);
	mr->hw_mr.page_size_log = mr->umem->page_shift;
	mr->hw_mr.fbo = ib_umem_offset(mr->umem);
	mr->hw_mr.length = len;
	mr->hw_mr.vaddr = usr_addr;
	mr->hw_mr.zbva = false;
	mr->hw_mr.phy_mr = false;
	mr->hw_mr.dma_mr = false;

	rc = dev->ops->rdma_register_tid(dev->rdma_ctx, &mr->hw_mr);
	if (rc) {
		DP_ERR(dev, "roce register tid returned an error %d\n", rc);
		goto err2;
	}

	mr->ibmr.lkey = mr->hw_mr.itid << 8 | mr->hw_mr.key;
	if (mr->hw_mr.remote_write || mr->hw_mr.remote_read ||
	    mr->hw_mr.remote_atomic)
		mr->ibmr.rkey = mr->hw_mr.itid << 8 | mr->hw_mr.key;

	DP_DEBUG(dev, QEDR_MSG_MR, "register user mr lkey: %x\n",
		 mr->ibmr.lkey);
	return &mr->ibmr;

err2:
	dev->ops->rdma_free_tid(dev->rdma_ctx, mr->hw_mr.itid);
err1:
	qedr_free_pbl(dev, &mr->info.pbl_info, mr->info.pbl_table);
err0:
	kfree(mr);
	return ERR_PTR(rc);
}

int qedr_dereg_mr(struct ib_mr *ib_mr)
{
	struct qedr_mr *mr = get_qedr_mr(ib_mr);
	struct qedr_dev *dev = get_qedr_dev(ib_mr->device);
	int rc = 0;

	rc = dev->ops->rdma_deregister_tid(dev->rdma_ctx, mr->hw_mr.itid);
	if (rc)
		return rc;

	dev->ops->rdma_free_tid(dev->rdma_ctx, mr->hw_mr.itid);

	if ((mr->type != QEDR_MR_DMA) && (mr->type != QEDR_MR_FRMR))
		qedr_free_pbl(dev, &mr->info.pbl_info, mr->info.pbl_table);

	/* it could be user registered memory. */
	if (mr->umem)
		ib_umem_release(mr->umem);

	kfree(mr);

	return rc;
}

static struct qedr_mr *__qedr_alloc_mr(struct ib_pd *ibpd,
				       int max_page_list_len)
{
	struct qedr_pd *pd = get_qedr_pd(ibpd);
	struct qedr_dev *dev = get_qedr_dev(ibpd->device);
	struct qedr_mr *mr;
	int rc = -ENOMEM;

	DP_DEBUG(dev, QEDR_MSG_MR,
		 "qedr_alloc_frmr pd = %d max_page_list_len= %d\n", pd->pd_id,
		 max_page_list_len);

	mr = kzalloc(sizeof(*mr), GFP_KERNEL);
	if (!mr)
		return ERR_PTR(rc);

	mr->dev = dev;
	mr->type = QEDR_MR_FRMR;

	rc = init_mr_info(dev, &mr->info, max_page_list_len, 1);
	if (rc)
		goto err0;

	rc = dev->ops->rdma_alloc_tid(dev->rdma_ctx, &mr->hw_mr.itid);
	if (rc) {
		DP_ERR(dev, "roce alloc tid returned an error %d\n", rc);
		goto err0;
	}

	/* Index only, 18 bit long, lkey = itid << 8 | key */
	mr->hw_mr.tid_type = QED_RDMA_TID_FMR;
	mr->hw_mr.key = 0;
	mr->hw_mr.pd = pd->pd_id;
	mr->hw_mr.local_read = 1;
	mr->hw_mr.local_write = 0;
	mr->hw_mr.remote_read = 0;
	mr->hw_mr.remote_write = 0;
	mr->hw_mr.remote_atomic = 0;
	mr->hw_mr.mw_bind = false;
	mr->hw_mr.pbl_ptr = 0;
	mr->hw_mr.pbl_two_level = mr->info.pbl_info.two_layered;
	mr->hw_mr.pbl_page_size_log = ilog2(mr->info.pbl_info.pbl_size);
	mr->hw_mr.fbo = 0;
	mr->hw_mr.length = 0;
	mr->hw_mr.vaddr = 0;
	mr->hw_mr.zbva = false;
	mr->hw_mr.phy_mr = true;
	mr->hw_mr.dma_mr = false;

	rc = dev->ops->rdma_register_tid(dev->rdma_ctx, &mr->hw_mr);
	if (rc) {
		DP_ERR(dev, "roce register tid returned an error %d\n", rc);
		goto err1;
	}

	mr->ibmr.lkey = mr->hw_mr.itid << 8 | mr->hw_mr.key;
	mr->ibmr.rkey = mr->ibmr.lkey;

	DP_DEBUG(dev, QEDR_MSG_MR, "alloc frmr: %x\n", mr->ibmr.lkey);
	return mr;

err1:
	dev->ops->rdma_free_tid(dev->rdma_ctx, mr->hw_mr.itid);
err0:
	kfree(mr);
	return ERR_PTR(rc);
}

struct ib_mr *qedr_alloc_mr(struct ib_pd *ibpd,
			    enum ib_mr_type mr_type, u32 max_num_sg)
{
	struct qedr_dev *dev;
	struct qedr_mr *mr;

	if (mr_type != IB_MR_TYPE_MEM_REG)
		return ERR_PTR(-EINVAL);

	mr = __qedr_alloc_mr(ibpd, max_num_sg);

	if (IS_ERR(mr))
		return ERR_PTR(-EINVAL);

	dev = mr->dev;

	return &mr->ibmr;
}

static int qedr_set_page(struct ib_mr *ibmr, u64 addr)
{
	struct qedr_mr *mr = get_qedr_mr(ibmr);
	struct qedr_pbl *pbl_table;
	struct regpair *pbe;
	u32 pbes_in_page;

	if (unlikely(mr->npages == mr->info.pbl_info.num_pbes)) {
		DP_ERR(mr->dev, "qedr_set_page failes when %d\n", mr->npages);
		return -ENOMEM;
	}

	DP_DEBUG(mr->dev, QEDR_MSG_MR, "qedr_set_page pages[%d] = 0x%llx\n",
		 mr->npages, addr);

	pbes_in_page = mr->info.pbl_info.pbl_size / sizeof(u64);
	pbl_table = mr->info.pbl_table + (mr->npages / pbes_in_page);
	pbe = (struct regpair *)pbl_table->va;
	pbe +=  mr->npages % pbes_in_page;
	pbe->lo = cpu_to_le32((u32)addr);
	pbe->hi = cpu_to_le32((u32)upper_32_bits(addr));

	mr->npages++;

	return 0;
}

static void handle_completed_mrs(struct qedr_dev *dev, struct mr_info *info)
{
	int work = info->completed - info->completed_handled - 1;

	DP_DEBUG(dev, QEDR_MSG_MR, "Special FMR work = %d\n", work);
	while (work-- > 0 && !list_empty(&info->inuse_pbl_list)) {
		struct qedr_pbl *pbl;

		/* Free all the page list that are possible to be freed
		 * (all the ones that were invalidated), under the assumption
		 * that if an FMR was completed successfully that means that
		 * if there was an invalidate operation before it also ended
		 */
		pbl = list_first_entry(&info->inuse_pbl_list,
				       struct qedr_pbl, list_entry);
		list_move_tail(&pbl->list_entry, &info->free_pbl_list);
		info->completed_handled++;
	}
}

int qedr_map_mr_sg(struct ib_mr *ibmr, struct scatterlist *sg,
		   int sg_nents, unsigned int *sg_offset)
{
	struct qedr_mr *mr = get_qedr_mr(ibmr);

	mr->npages = 0;

	handle_completed_mrs(mr->dev, &mr->info);
	return ib_sg_to_pages(ibmr, sg, sg_nents, NULL, qedr_set_page);
}

struct ib_mr *qedr_get_dma_mr(struct ib_pd *ibpd, int acc)
{
	struct qedr_dev *dev = get_qedr_dev(ibpd->device);
	struct qedr_pd *pd = get_qedr_pd(ibpd);
	struct qedr_mr *mr;
	int rc;

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

	mr->type = QEDR_MR_DMA;

	rc = dev->ops->rdma_alloc_tid(dev->rdma_ctx, &mr->hw_mr.itid);
	if (rc) {
		DP_ERR(dev, "roce alloc tid returned an error %d\n", rc);
		goto err1;
	}

	/* index only, 18 bit long, lkey = itid << 8 | key */
	mr->hw_mr.tid_type = QED_RDMA_TID_REGISTERED_MR;
	mr->hw_mr.pd = pd->pd_id;
	mr->hw_mr.local_read = 1;
	mr->hw_mr.local_write = (acc & IB_ACCESS_LOCAL_WRITE) ? 1 : 0;
	mr->hw_mr.remote_read = (acc & IB_ACCESS_REMOTE_READ) ? 1 : 0;
	mr->hw_mr.remote_write = (acc & IB_ACCESS_REMOTE_WRITE) ? 1 : 0;
	mr->hw_mr.remote_atomic = (acc & IB_ACCESS_REMOTE_ATOMIC) ? 1 : 0;
	mr->hw_mr.dma_mr = true;

	rc = dev->ops->rdma_register_tid(dev->rdma_ctx, &mr->hw_mr);
	if (rc) {
		DP_ERR(dev, "roce register tid returned an error %d\n", rc);
		goto err2;
	}

	mr->ibmr.lkey = mr->hw_mr.itid << 8 | mr->hw_mr.key;
	if (mr->hw_mr.remote_write || mr->hw_mr.remote_read ||
	    mr->hw_mr.remote_atomic)
		mr->ibmr.rkey = mr->hw_mr.itid << 8 | mr->hw_mr.key;

	DP_DEBUG(dev, QEDR_MSG_MR, "get dma mr: lkey = %x\n", mr->ibmr.lkey);
	return &mr->ibmr;

err2:
	dev->ops->rdma_free_tid(dev->rdma_ctx, mr->hw_mr.itid);
err1:
	kfree(mr);
	return ERR_PTR(rc);
}

static inline int qedr_wq_is_full(struct qedr_qp_hwq_info *wq)
{
	return (((wq->prod + 1) % wq->max_wr) == wq->cons);
}

static int sge_data_len(struct ib_sge *sg_list, int num_sge)
{
	int i, len = 0;

	for (i = 0; i < num_sge; i++)
		len += sg_list[i].length;

	return len;
}

static void swap_wqe_data64(u64 *p)
{
	int i;

	for (i = 0; i < QEDR_SQE_ELEMENT_SIZE / sizeof(u64); i++, p++)
		*p = cpu_to_be64(cpu_to_le64(*p));
}

static u32 qedr_prepare_sq_inline_data(struct qedr_dev *dev,
				       struct qedr_qp *qp, u8 *wqe_size,
				       struct ib_send_wr *wr,
				       struct ib_send_wr **bad_wr, u8 *bits,
				       u8 bit)
{
	u32 data_size = sge_data_len(wr->sg_list, wr->num_sge);
	char *seg_prt, *wqe;
	int i, seg_siz;

	if (data_size > ROCE_REQ_MAX_INLINE_DATA_SIZE) {
		DP_ERR(dev, "Too much inline data in WR: %d\n", data_size);
		*bad_wr = wr;
		return 0;
	}

	if (!data_size)
		return data_size;

	*bits |= bit;

	seg_prt = NULL;
	wqe = NULL;
	seg_siz = 0;

	/* Copy data inline */
	for (i = 0; i < wr->num_sge; i++) {
		u32 len = wr->sg_list[i].length;
		void *src = (void *)(uintptr_t)wr->sg_list[i].addr;

		while (len > 0) {
			u32 cur;

			/* New segment required */
			if (!seg_siz) {
				wqe = (char *)qed_chain_produce(&qp->sq.pbl);
				seg_prt = wqe;
				seg_siz = sizeof(struct rdma_sq_common_wqe);
				(*wqe_size)++;
			}

			/* Calculate currently allowed length */
			cur = min_t(u32, len, seg_siz);
			memcpy(seg_prt, src, cur);

			/* Update segment variables */
			seg_prt += cur;
			seg_siz -= cur;

			/* Update sge variables */
			src += cur;
			len -= cur;

			/* Swap fully-completed segments */
			if (!seg_siz)
				swap_wqe_data64((u64 *)wqe);
		}
	}

	/* swap last not completed segment */
	if (seg_siz)
		swap_wqe_data64((u64 *)wqe);

	return data_size;
}

#define RQ_SGE_SET(sge, vaddr, vlength, vflags)			\
	do {							\
		DMA_REGPAIR_LE(sge->addr, vaddr);		\
		(sge)->length = cpu_to_le32(vlength);		\
		(sge)->flags = cpu_to_le32(vflags);		\
	} while (0)

#define SRQ_HDR_SET(hdr, vwr_id, num_sge)			\
	do {							\
		DMA_REGPAIR_LE(hdr->wr_id, vwr_id);		\
		(hdr)->num_sges = num_sge;			\
	} while (0)

#define SRQ_SGE_SET(sge, vaddr, vlength, vlkey)			\
	do {							\
		DMA_REGPAIR_LE(sge->addr, vaddr);		\
		(sge)->length = cpu_to_le32(vlength);		\
		(sge)->l_key = cpu_to_le32(vlkey);		\
	} while (0)

static u32 qedr_prepare_sq_sges(struct qedr_qp *qp, u8 *wqe_size,
				struct ib_send_wr *wr)
{
	u32 data_size = 0;
	int i;

	for (i = 0; i < wr->num_sge; i++) {
		struct rdma_sq_sge *sge = qed_chain_produce(&qp->sq.pbl);

		DMA_REGPAIR_LE(sge->addr, wr->sg_list[i].addr);
		sge->l_key = cpu_to_le32(wr->sg_list[i].lkey);
		sge->length = cpu_to_le32(wr->sg_list[i].length);
		data_size += wr->sg_list[i].length;
	}

	if (wqe_size)
		*wqe_size += wr->num_sge;

	return data_size;
}

static u32 qedr_prepare_sq_rdma_data(struct qedr_dev *dev,
				     struct qedr_qp *qp,
				     struct rdma_sq_rdma_wqe_1st *rwqe,
				     struct rdma_sq_rdma_wqe_2nd *rwqe2,
				     struct ib_send_wr *wr,
				     struct ib_send_wr **bad_wr)
{
	rwqe2->r_key = cpu_to_le32(rdma_wr(wr)->rkey);
	DMA_REGPAIR_LE(rwqe2->remote_va, rdma_wr(wr)->remote_addr);

	if (wr->send_flags & IB_SEND_INLINE &&
	    (wr->opcode == IB_WR_RDMA_WRITE_WITH_IMM ||
	     wr->opcode == IB_WR_RDMA_WRITE)) {
		u8 flags = 0;

		SET_FIELD2(flags, RDMA_SQ_RDMA_WQE_1ST_INLINE_FLG, 1);
		return qedr_prepare_sq_inline_data(dev, qp, &rwqe->wqe_size, wr,
						   bad_wr, &rwqe->flags, flags);
	}

	return qedr_prepare_sq_sges(qp, &rwqe->wqe_size, wr);
}

static u32 qedr_prepare_sq_send_data(struct qedr_dev *dev,
				     struct qedr_qp *qp,
				     struct rdma_sq_send_wqe_1st *swqe,
				     struct rdma_sq_send_wqe_2st *swqe2,
				     struct ib_send_wr *wr,
				     struct ib_send_wr **bad_wr)
{
	memset(swqe2, 0, sizeof(*swqe2));
	if (wr->send_flags & IB_SEND_INLINE) {
		u8 flags = 0;

		SET_FIELD2(flags, RDMA_SQ_SEND_WQE_INLINE_FLG, 1);
		return qedr_prepare_sq_inline_data(dev, qp, &swqe->wqe_size, wr,
						   bad_wr, &swqe->flags, flags);
	}

	return qedr_prepare_sq_sges(qp, &swqe->wqe_size, wr);
}

static int qedr_prepare_reg(struct qedr_qp *qp,
			    struct rdma_sq_fmr_wqe_1st *fwqe1,
			    struct ib_reg_wr *wr)
{
	struct qedr_mr *mr = get_qedr_mr(wr->mr);
	struct rdma_sq_fmr_wqe_2nd *fwqe2;

	fwqe2 = (struct rdma_sq_fmr_wqe_2nd *)qed_chain_produce(&qp->sq.pbl);
	fwqe1->addr.hi = upper_32_bits(mr->ibmr.iova);
	fwqe1->addr.lo = lower_32_bits(mr->ibmr.iova);
	fwqe1->l_key = wr->key;

	fwqe2->access_ctrl = 0;

	SET_FIELD2(fwqe2->access_ctrl, RDMA_SQ_FMR_WQE_2ND_REMOTE_READ,
		   !!(wr->access & IB_ACCESS_REMOTE_READ));
	SET_FIELD2(fwqe2->access_ctrl, RDMA_SQ_FMR_WQE_2ND_REMOTE_WRITE,
		   !!(wr->access & IB_ACCESS_REMOTE_WRITE));
	SET_FIELD2(fwqe2->access_ctrl, RDMA_SQ_FMR_WQE_2ND_ENABLE_ATOMIC,
		   !!(wr->access & IB_ACCESS_REMOTE_ATOMIC));
	SET_FIELD2(fwqe2->access_ctrl, RDMA_SQ_FMR_WQE_2ND_LOCAL_READ, 1);
	SET_FIELD2(fwqe2->access_ctrl, RDMA_SQ_FMR_WQE_2ND_LOCAL_WRITE,
		   !!(wr->access & IB_ACCESS_LOCAL_WRITE));
	fwqe2->fmr_ctrl = 0;

	SET_FIELD2(fwqe2->fmr_ctrl, RDMA_SQ_FMR_WQE_2ND_PAGE_SIZE_LOG,
		   ilog2(mr->ibmr.page_size) - 12);

	fwqe2->length_hi = 0;
	fwqe2->length_lo = mr->ibmr.length;
	fwqe2->pbl_addr.hi = upper_32_bits(mr->info.pbl_table->pa);
	fwqe2->pbl_addr.lo = lower_32_bits(mr->info.pbl_table->pa);

	qp->wqe_wr_id[qp->sq.prod].mr = mr;

	return 0;
}

static enum ib_wc_opcode qedr_ib_to_wc_opcode(enum ib_wr_opcode opcode)
{
	switch (opcode) {
	case IB_WR_RDMA_WRITE:
	case IB_WR_RDMA_WRITE_WITH_IMM:
		return IB_WC_RDMA_WRITE;
	case IB_WR_SEND_WITH_IMM:
	case IB_WR_SEND:
	case IB_WR_SEND_WITH_INV:
		return IB_WC_SEND;
	case IB_WR_RDMA_READ:
		return IB_WC_RDMA_READ;
	case IB_WR_ATOMIC_CMP_AND_SWP:
		return IB_WC_COMP_SWAP;
	case IB_WR_ATOMIC_FETCH_AND_ADD:
		return IB_WC_FETCH_ADD;
	case IB_WR_REG_MR:
		return IB_WC_REG_MR;
	case IB_WR_LOCAL_INV:
		return IB_WC_LOCAL_INV;
	default:
		return IB_WC_SEND;
	}
}

static inline bool qedr_can_post_send(struct qedr_qp *qp, struct ib_send_wr *wr)
{
	int wq_is_full, err_wr, pbl_is_full;
	struct qedr_dev *dev = qp->dev;

	/* prevent SQ overflow and/or processing of a bad WR */
	err_wr = wr->num_sge > qp->sq.max_sges;
	wq_is_full = qedr_wq_is_full(&qp->sq);
	pbl_is_full = qed_chain_get_elem_left_u32(&qp->sq.pbl) <
		      QEDR_MAX_SQE_ELEMENTS_PER_SQE;
	if (wq_is_full || err_wr || pbl_is_full) {
		if (wq_is_full && !(qp->err_bitmap & QEDR_QP_ERR_SQ_FULL)) {
			DP_ERR(dev,
			       "error: WQ is full. Post send on QP %p failed (this error appears only once)\n",
			       qp);
			qp->err_bitmap |= QEDR_QP_ERR_SQ_FULL;
		}

		if (err_wr && !(qp->err_bitmap & QEDR_QP_ERR_BAD_SR)) {
			DP_ERR(dev,
			       "error: WR is bad. Post send on QP %p failed (this error appears only once)\n",
			       qp);
			qp->err_bitmap |= QEDR_QP_ERR_BAD_SR;
		}

		if (pbl_is_full &&
		    !(qp->err_bitmap & QEDR_QP_ERR_SQ_PBL_FULL)) {
			DP_ERR(dev,
			       "error: WQ PBL is full. Post send on QP %p failed (this error appears only once)\n",
			       qp);
			qp->err_bitmap |= QEDR_QP_ERR_SQ_PBL_FULL;
		}
		return false;
	}
	return true;
}

static int __qedr_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
		     struct ib_send_wr **bad_wr)
{
	struct qedr_dev *dev = get_qedr_dev(ibqp->device);
	struct qedr_qp *qp = get_qedr_qp(ibqp);
	struct rdma_sq_atomic_wqe_1st *awqe1;
	struct rdma_sq_atomic_wqe_2nd *awqe2;
	struct rdma_sq_atomic_wqe_3rd *awqe3;
	struct rdma_sq_send_wqe_2st *swqe2;
	struct rdma_sq_local_inv_wqe *iwqe;
	struct rdma_sq_rdma_wqe_2nd *rwqe2;
	struct rdma_sq_send_wqe_1st *swqe;
	struct rdma_sq_rdma_wqe_1st *rwqe;
	struct rdma_sq_fmr_wqe_1st *fwqe1;
	struct rdma_sq_common_wqe *wqe;
	u32 length;
	int rc = 0;
	bool comp;

	if (!qedr_can_post_send(qp, wr)) {
		*bad_wr = wr;
		return -ENOMEM;
	}

	wqe = qed_chain_produce(&qp->sq.pbl);
	qp->wqe_wr_id[qp->sq.prod].signaled =
		!!(wr->send_flags & IB_SEND_SIGNALED) || qp->signaled;

	wqe->flags = 0;
	SET_FIELD2(wqe->flags, RDMA_SQ_SEND_WQE_SE_FLG,
		   !!(wr->send_flags & IB_SEND_SOLICITED));
	comp = (!!(wr->send_flags & IB_SEND_SIGNALED)) || qp->signaled;
	SET_FIELD2(wqe->flags, RDMA_SQ_SEND_WQE_COMP_FLG, comp);
	SET_FIELD2(wqe->flags, RDMA_SQ_SEND_WQE_RD_FENCE_FLG,
		   !!(wr->send_flags & IB_SEND_FENCE));
	wqe->prev_wqe_size = qp->prev_wqe_size;

	qp->wqe_wr_id[qp->sq.prod].opcode = qedr_ib_to_wc_opcode(wr->opcode);

	switch (wr->opcode) {
	case IB_WR_SEND_WITH_IMM:
		wqe->req_type = RDMA_SQ_REQ_TYPE_SEND_WITH_IMM;
		swqe = (struct rdma_sq_send_wqe_1st *)wqe;
		swqe->wqe_size = 2;
		swqe2 = qed_chain_produce(&qp->sq.pbl);

		swqe->inv_key_or_imm_data = cpu_to_le32(wr->ex.imm_data);
		length = qedr_prepare_sq_send_data(dev, qp, swqe, swqe2,
						   wr, bad_wr);
		swqe->length = cpu_to_le32(length);
		qp->wqe_wr_id[qp->sq.prod].wqe_size = swqe->wqe_size;
		qp->prev_wqe_size = swqe->wqe_size;
		qp->wqe_wr_id[qp->sq.prod].bytes_len = swqe->length;
		break;
	case IB_WR_SEND:
		wqe->req_type = RDMA_SQ_REQ_TYPE_SEND;
		swqe = (struct rdma_sq_send_wqe_1st *)wqe;

		swqe->wqe_size = 2;
		swqe2 = qed_chain_produce(&qp->sq.pbl);
		length = qedr_prepare_sq_send_data(dev, qp, swqe, swqe2,
						   wr, bad_wr);
		swqe->length = cpu_to_le32(length);
		qp->wqe_wr_id[qp->sq.prod].wqe_size = swqe->wqe_size;
		qp->prev_wqe_size = swqe->wqe_size;
		qp->wqe_wr_id[qp->sq.prod].bytes_len = swqe->length;
		break;
	case IB_WR_SEND_WITH_INV:
		wqe->req_type = RDMA_SQ_REQ_TYPE_SEND_WITH_INVALIDATE;
		swqe = (struct rdma_sq_send_wqe_1st *)wqe;
		swqe2 = qed_chain_produce(&qp->sq.pbl);
		swqe->wqe_size = 2;
		swqe->inv_key_or_imm_data = cpu_to_le32(wr->ex.invalidate_rkey);
		length = qedr_prepare_sq_send_data(dev, qp, swqe, swqe2,
						   wr, bad_wr);
		swqe->length = cpu_to_le32(length);
		qp->wqe_wr_id[qp->sq.prod].wqe_size = swqe->wqe_size;
		qp->prev_wqe_size = swqe->wqe_size;
		qp->wqe_wr_id[qp->sq.prod].bytes_len = swqe->length;
		break;

	case IB_WR_RDMA_WRITE_WITH_IMM:
		wqe->req_type = RDMA_SQ_REQ_TYPE_RDMA_WR_WITH_IMM;
		rwqe = (struct rdma_sq_rdma_wqe_1st *)wqe;

		rwqe->wqe_size = 2;
		rwqe->imm_data = htonl(cpu_to_le32(wr->ex.imm_data));
		rwqe2 = qed_chain_produce(&qp->sq.pbl);
		length = qedr_prepare_sq_rdma_data(dev, qp, rwqe, rwqe2,
						   wr, bad_wr);
		rwqe->length = cpu_to_le32(length);
		qp->wqe_wr_id[qp->sq.prod].wqe_size = rwqe->wqe_size;
		qp->prev_wqe_size = rwqe->wqe_size;
		qp->wqe_wr_id[qp->sq.prod].bytes_len = rwqe->length;
		break;
	case IB_WR_RDMA_WRITE:
		wqe->req_type = RDMA_SQ_REQ_TYPE_RDMA_WR;
		rwqe = (struct rdma_sq_rdma_wqe_1st *)wqe;

		rwqe->wqe_size = 2;
		rwqe2 = qed_chain_produce(&qp->sq.pbl);
		length = qedr_prepare_sq_rdma_data(dev, qp, rwqe, rwqe2,
						   wr, bad_wr);
		rwqe->length = cpu_to_le32(length);
		qp->wqe_wr_id[qp->sq.prod].wqe_size = rwqe->wqe_size;
		qp->prev_wqe_size = rwqe->wqe_size;
		qp->wqe_wr_id[qp->sq.prod].bytes_len = rwqe->length;
		break;
	case IB_WR_RDMA_READ_WITH_INV:
		DP_ERR(dev,
		       "RDMA READ WITH INVALIDATE not supported\n");
		*bad_wr = wr;
		rc = -EINVAL;
		break;

	case IB_WR_RDMA_READ:
		wqe->req_type = RDMA_SQ_REQ_TYPE_RDMA_RD;
		rwqe = (struct rdma_sq_rdma_wqe_1st *)wqe;

		rwqe->wqe_size = 2;
		rwqe2 = qed_chain_produce(&qp->sq.pbl);
		length = qedr_prepare_sq_rdma_data(dev, qp, rwqe, rwqe2,
						   wr, bad_wr);
		rwqe->length = cpu_to_le32(length);
		qp->wqe_wr_id[qp->sq.prod].wqe_size = rwqe->wqe_size;
		qp->prev_wqe_size = rwqe->wqe_size;
		qp->wqe_wr_id[qp->sq.prod].bytes_len = rwqe->length;
		break;

	case IB_WR_ATOMIC_CMP_AND_SWP:
	case IB_WR_ATOMIC_FETCH_AND_ADD:
		awqe1 = (struct rdma_sq_atomic_wqe_1st *)wqe;
		awqe1->wqe_size = 4;

		awqe2 = qed_chain_produce(&qp->sq.pbl);
		DMA_REGPAIR_LE(awqe2->remote_va, atomic_wr(wr)->remote_addr);
		awqe2->r_key = cpu_to_le32(atomic_wr(wr)->rkey);

		awqe3 = qed_chain_produce(&qp->sq.pbl);

		if (wr->opcode == IB_WR_ATOMIC_FETCH_AND_ADD) {
			wqe->req_type = RDMA_SQ_REQ_TYPE_ATOMIC_ADD;
			DMA_REGPAIR_LE(awqe3->swap_data,
				       atomic_wr(wr)->compare_add);
		} else {
			wqe->req_type = RDMA_SQ_REQ_TYPE_ATOMIC_CMP_AND_SWAP;
			DMA_REGPAIR_LE(awqe3->swap_data,
				       atomic_wr(wr)->swap);
			DMA_REGPAIR_LE(awqe3->cmp_data,
				       atomic_wr(wr)->compare_add);
		}

		qedr_prepare_sq_sges(qp, NULL, wr);

		qp->wqe_wr_id[qp->sq.prod].wqe_size = awqe1->wqe_size;
		qp->prev_wqe_size = awqe1->wqe_size;
		break;

	case IB_WR_LOCAL_INV:
		iwqe = (struct rdma_sq_local_inv_wqe *)wqe;
		iwqe->wqe_size = 1;

		iwqe->req_type = RDMA_SQ_REQ_TYPE_LOCAL_INVALIDATE;
		iwqe->inv_l_key = wr->ex.invalidate_rkey;
		qp->wqe_wr_id[qp->sq.prod].wqe_size = iwqe->wqe_size;
		qp->prev_wqe_size = iwqe->wqe_size;
		break;
	case IB_WR_REG_MR:
		DP_DEBUG(dev, QEDR_MSG_CQ, "REG_MR\n");
		wqe->req_type = RDMA_SQ_REQ_TYPE_FAST_MR;
		fwqe1 = (struct rdma_sq_fmr_wqe_1st *)wqe;
		fwqe1->wqe_size = 2;

		rc = qedr_prepare_reg(qp, fwqe1, reg_wr(wr));
		if (rc) {
			DP_ERR(dev, "IB_REG_MR failed rc=%d\n", rc);
			*bad_wr = wr;
			break;
		}

		qp->wqe_wr_id[qp->sq.prod].wqe_size = fwqe1->wqe_size;
		qp->prev_wqe_size = fwqe1->wqe_size;
		break;
	default:
		DP_ERR(dev, "invalid opcode 0x%x!\n", wr->opcode);
		rc = -EINVAL;
		*bad_wr = wr;
		break;
	}

	if (*bad_wr) {
		u16 value;

		/* Restore prod to its position before
		 * this WR was processed
		 */
		value = le16_to_cpu(qp->sq.db_data.data.value);
		qed_chain_set_prod(&qp->sq.pbl, value, wqe);

		/* Restore prev_wqe_size */
		qp->prev_wqe_size = wqe->prev_wqe_size;
		rc = -EINVAL;
		DP_ERR(dev, "POST SEND FAILED\n");
	}

	return rc;
}

int qedr_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
		   struct ib_send_wr **bad_wr)
{
	struct qedr_dev *dev = get_qedr_dev(ibqp->device);
	struct qedr_qp *qp = get_qedr_qp(ibqp);
	unsigned long flags;
	int rc = 0;

	*bad_wr = NULL;

	if (qp->qp_type == IB_QPT_GSI)
		return qedr_gsi_post_send(ibqp, wr, bad_wr);

	spin_lock_irqsave(&qp->q_lock, flags);

	if ((qp->state != QED_ROCE_QP_STATE_RTS) &&
	    (qp->state != QED_ROCE_QP_STATE_ERR) &&
	    (qp->state != QED_ROCE_QP_STATE_SQD)) {
		spin_unlock_irqrestore(&qp->q_lock, flags);
		*bad_wr = wr;
		DP_DEBUG(dev, QEDR_MSG_CQ,
			 "QP in wrong state! QP icid=0x%x state %d\n",
			 qp->icid, qp->state);
		return -EINVAL;
	}

	while (wr) {
		rc = __qedr_post_send(ibqp, wr, bad_wr);
		if (rc)
			break;

		qp->wqe_wr_id[qp->sq.prod].wr_id = wr->wr_id;

		qedr_inc_sw_prod(&qp->sq);

		qp->sq.db_data.data.value++;

		wr = wr->next;
	}

	/* Trigger doorbell
	 * If there was a failure in the first WR then it will be triggered in
	 * vane. However this is not harmful (as long as the producer value is
	 * unchanged). For performance reasons we avoid checking for this
	 * redundant doorbell.
	 */
	wmb();
	writel(qp->sq.db_data.raw, qp->sq.db);

	/* Make sure write sticks */
	mmiowb();

	spin_unlock_irqrestore(&qp->q_lock, flags);

	return rc;
}

int qedr_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr,
		   struct ib_recv_wr **bad_wr)
{
	struct qedr_qp *qp = get_qedr_qp(ibqp);
	struct qedr_dev *dev = qp->dev;
	unsigned long flags;
	int status = 0;

	if (qp->qp_type == IB_QPT_GSI)
		return qedr_gsi_post_recv(ibqp, wr, bad_wr);

	spin_lock_irqsave(&qp->q_lock, flags);

	if (qp->state == QED_ROCE_QP_STATE_RESET) {
		spin_unlock_irqrestore(&qp->q_lock, flags);
		*bad_wr = wr;
		return -EINVAL;
	}

	while (wr) {
		int i;

		if (qed_chain_get_elem_left_u32(&qp->rq.pbl) <
		    QEDR_MAX_RQE_ELEMENTS_PER_RQE ||
		    wr->num_sge > qp->rq.max_sges) {
			DP_ERR(dev, "Can't post WR  (%d < %d) || (%d > %d)\n",
			       qed_chain_get_elem_left_u32(&qp->rq.pbl),
			       QEDR_MAX_RQE_ELEMENTS_PER_RQE, wr->num_sge,
			       qp->rq.max_sges);
			status = -ENOMEM;
			*bad_wr = wr;
			break;
		}
		for (i = 0; i < wr->num_sge; i++) {
			u32 flags = 0;
			struct rdma_rq_sge *rqe =
			    qed_chain_produce(&qp->rq.pbl);

			/* First one must include the number
			 * of SGE in the list
			 */
			if (!i)
				SET_FIELD(flags, RDMA_RQ_SGE_NUM_SGES,
					  wr->num_sge);

			SET_FIELD(flags, RDMA_RQ_SGE_L_KEY,
				  wr->sg_list[i].lkey);

			RQ_SGE_SET(rqe, wr->sg_list[i].addr,
				   wr->sg_list[i].length, flags);
		}

		/* Special case of no sges. FW requires between 1-4 sges...
		 * in this case we need to post 1 sge with length zero. this is
		 * because rdma write with immediate consumes an RQ.
		 */
		if (!wr->num_sge) {
			u32 flags = 0;
			struct rdma_rq_sge *rqe =
			    qed_chain_produce(&qp->rq.pbl);

			/* First one must include the number
			 * of SGE in the list
			 */
			SET_FIELD(flags, RDMA_RQ_SGE_L_KEY, 0);
			SET_FIELD(flags, RDMA_RQ_SGE_NUM_SGES, 1);

			RQ_SGE_SET(rqe, 0, 0, flags);
			i = 1;
		}

		qp->rqe_wr_id[qp->rq.prod].wr_id = wr->wr_id;
		qp->rqe_wr_id[qp->rq.prod].wqe_size = i;

		qedr_inc_sw_prod(&qp->rq);

		/* Flush all the writes before signalling doorbell */
		wmb();

		qp->rq.db_data.data.value++;

		writel(qp->rq.db_data.raw, qp->rq.db);

		/* Make sure write sticks */
		mmiowb();

		wr = wr->next;
	}

	spin_unlock_irqrestore(&qp->q_lock, flags);

	return status;
}

static int is_valid_cqe(struct qedr_cq *cq, union rdma_cqe *cqe)
{
	struct rdma_cqe_requester *resp_cqe = &cqe->req;

	return (resp_cqe->flags & RDMA_CQE_REQUESTER_TOGGLE_BIT_MASK) ==
		cq->pbl_toggle;
}

static struct qedr_qp *cqe_get_qp(union rdma_cqe *cqe)
{
	struct rdma_cqe_requester *resp_cqe = &cqe->req;
	struct qedr_qp *qp;

	qp = (struct qedr_qp *)(uintptr_t)HILO_GEN(resp_cqe->qp_handle.hi,
						   resp_cqe->qp_handle.lo,
						   u64);
	return qp;
}

static enum rdma_cqe_type cqe_get_type(union rdma_cqe *cqe)
{
	struct rdma_cqe_requester *resp_cqe = &cqe->req;

	return GET_FIELD(resp_cqe->flags, RDMA_CQE_REQUESTER_TYPE);
}

/* Return latest CQE (needs processing) */
static union rdma_cqe *get_cqe(struct qedr_cq *cq)
{
	return cq->latest_cqe;
}

/* In fmr we need to increase the number of fmr completed counter for the fmr
 * algorithm determining whether we can free a pbl or not.
 * we need to perform this whether the work request was signaled or not. for
 * this purpose we call this function from the condition that checks if a wr
 * should be skipped, to make sure we don't miss it ( possibly this fmr
 * operation was not signalted)
 */
static inline void qedr_chk_if_fmr(struct qedr_qp *qp)
{
	if (qp->wqe_wr_id[qp->sq.cons].opcode == IB_WC_REG_MR)
		qp->wqe_wr_id[qp->sq.cons].mr->info.completed++;
}

static int process_req(struct qedr_dev *dev, struct qedr_qp *qp,
		       struct qedr_cq *cq, int num_entries,
		       struct ib_wc *wc, u16 hw_cons, enum ib_wc_status status,
		       int force)
{
	u16 cnt = 0;

	while (num_entries && qp->sq.wqe_cons != hw_cons) {
		if (!qp->wqe_wr_id[qp->sq.cons].signaled && !force) {
			qedr_chk_if_fmr(qp);
			/* skip WC */
			goto next_cqe;
		}

		/* fill WC */
		wc->status = status;
		wc->vendor_err = 0;
		wc->wc_flags = 0;
		wc->src_qp = qp->id;
		wc->qp = &qp->ibqp;

		wc->wr_id = qp->wqe_wr_id[qp->sq.cons].wr_id;
		wc->opcode = qp->wqe_wr_id[qp->sq.cons].opcode;

		switch (wc->opcode) {
		case IB_WC_RDMA_WRITE:
			wc->byte_len = qp->wqe_wr_id[qp->sq.cons].bytes_len;
			break;
		case IB_WC_COMP_SWAP:
		case IB_WC_FETCH_ADD:
			wc->byte_len = 8;
			break;
		case IB_WC_REG_MR:
			qp->wqe_wr_id[qp->sq.cons].mr->info.completed++;
			break;
		case IB_WC_RDMA_READ:
		case IB_WC_SEND:
			wc->byte_len = qp->wqe_wr_id[qp->sq.cons].bytes_len;
			break;
		default:
			break;
		}

		num_entries--;
		wc++;
		cnt++;
next_cqe:
		while (qp->wqe_wr_id[qp->sq.cons].wqe_size--)
			qed_chain_consume(&qp->sq.pbl);
		qedr_inc_sw_cons(&qp->sq);
	}

	return cnt;
}

static int qedr_poll_cq_req(struct qedr_dev *dev,
			    struct qedr_qp *qp, struct qedr_cq *cq,
			    int num_entries, struct ib_wc *wc,
			    struct rdma_cqe_requester *req)
{
	int cnt = 0;

	switch (req->status) {
	case RDMA_CQE_REQ_STS_OK:
		cnt = process_req(dev, qp, cq, num_entries, wc, req->sq_cons,
				  IB_WC_SUCCESS, 0);
		break;
	case RDMA_CQE_REQ_STS_WORK_REQUEST_FLUSHED_ERR:
		if (qp->state != QED_ROCE_QP_STATE_ERR)
			DP_ERR(dev,
			       "Error: POLL CQ with RDMA_CQE_REQ_STS_WORK_REQUEST_FLUSHED_ERR. CQ icid=0x%x, QP icid=0x%x\n",
			       cq->icid, qp->icid);
		cnt = process_req(dev, qp, cq, num_entries, wc, req->sq_cons,
				  IB_WC_WR_FLUSH_ERR, 1);
		break;
	default:
		/* process all WQE before the cosumer */
		qp->state = QED_ROCE_QP_STATE_ERR;
		cnt = process_req(dev, qp, cq, num_entries, wc,
				  req->sq_cons - 1, IB_WC_SUCCESS, 0);
		wc += cnt;
		/* if we have extra WC fill it with actual error info */
		if (cnt < num_entries) {
			enum ib_wc_status wc_status;

			switch (req->status) {
			case RDMA_CQE_REQ_STS_BAD_RESPONSE_ERR:
				DP_ERR(dev,
				       "Error: POLL CQ with RDMA_CQE_REQ_STS_BAD_RESPONSE_ERR. CQ icid=0x%x, QP icid=0x%x\n",
				       cq->icid, qp->icid);
				wc_status = IB_WC_BAD_RESP_ERR;
				break;
			case RDMA_CQE_REQ_STS_LOCAL_LENGTH_ERR:
				DP_ERR(dev,
				       "Error: POLL CQ with RDMA_CQE_REQ_STS_LOCAL_LENGTH_ERR. CQ icid=0x%x, QP icid=0x%x\n",
				       cq->icid, qp->icid);
				wc_status = IB_WC_LOC_LEN_ERR;
				break;
			case RDMA_CQE_REQ_STS_LOCAL_QP_OPERATION_ERR:
				DP_ERR(dev,
				       "Error: POLL CQ with RDMA_CQE_REQ_STS_LOCAL_QP_OPERATION_ERR. CQ icid=0x%x, QP icid=0x%x\n",
				       cq->icid, qp->icid);
				wc_status = IB_WC_LOC_QP_OP_ERR;
				break;
			case RDMA_CQE_REQ_STS_LOCAL_PROTECTION_ERR:
				DP_ERR(dev,
				       "Error: POLL CQ with RDMA_CQE_REQ_STS_LOCAL_PROTECTION_ERR. CQ icid=0x%x, QP icid=0x%x\n",
				       cq->icid, qp->icid);
				wc_status = IB_WC_LOC_PROT_ERR;
				break;
			case RDMA_CQE_REQ_STS_MEMORY_MGT_OPERATION_ERR:
				DP_ERR(dev,
				       "Error: POLL CQ with RDMA_CQE_REQ_STS_MEMORY_MGT_OPERATION_ERR. CQ icid=0x%x, QP icid=0x%x\n",
				       cq->icid, qp->icid);
				wc_status = IB_WC_MW_BIND_ERR;
				break;
			case RDMA_CQE_REQ_STS_REMOTE_INVALID_REQUEST_ERR:
				DP_ERR(dev,
				       "Error: POLL CQ with RDMA_CQE_REQ_STS_REMOTE_INVALID_REQUEST_ERR. CQ icid=0x%x, QP icid=0x%x\n",
				       cq->icid, qp->icid);
				wc_status = IB_WC_REM_INV_REQ_ERR;
				break;
			case RDMA_CQE_REQ_STS_REMOTE_ACCESS_ERR:
				DP_ERR(dev,
				       "Error: POLL CQ with RDMA_CQE_REQ_STS_REMOTE_ACCESS_ERR. CQ icid=0x%x, QP icid=0x%x\n",
				       cq->icid, qp->icid);
				wc_status = IB_WC_REM_ACCESS_ERR;
				break;
			case RDMA_CQE_REQ_STS_REMOTE_OPERATION_ERR:
				DP_ERR(dev,
				       "Error: POLL CQ with RDMA_CQE_REQ_STS_REMOTE_OPERATION_ERR. CQ icid=0x%x, QP icid=0x%x\n",
				       cq->icid, qp->icid);
				wc_status = IB_WC_REM_OP_ERR;
				break;
			case RDMA_CQE_REQ_STS_RNR_NAK_RETRY_CNT_ERR:
				DP_ERR(dev,
				       "Error: POLL CQ with RDMA_CQE_REQ_STS_RNR_NAK_RETRY_CNT_ERR. CQ icid=0x%x, QP icid=0x%x\n",
				       cq->icid, qp->icid);
				wc_status = IB_WC_RNR_RETRY_EXC_ERR;
				break;
			case RDMA_CQE_REQ_STS_TRANSPORT_RETRY_CNT_ERR:
				DP_ERR(dev,
				       "Error: POLL CQ with ROCE_CQE_REQ_STS_TRANSPORT_RETRY_CNT_ERR. CQ icid=0x%x, QP icid=0x%x\n",
				       cq->icid, qp->icid);
				wc_status = IB_WC_RETRY_EXC_ERR;
				break;
			default:
				DP_ERR(dev,
				       "Error: POLL CQ with IB_WC_GENERAL_ERR. CQ icid=0x%x, QP icid=0x%x\n",
				       cq->icid, qp->icid);
				wc_status = IB_WC_GENERAL_ERR;
			}
			cnt += process_req(dev, qp, cq, 1, wc, req->sq_cons,
					   wc_status, 1);
		}
	}

	return cnt;
}

static inline int qedr_cqe_resp_status_to_ib(u8 status)
{
	switch (status) {
	case RDMA_CQE_RESP_STS_LOCAL_ACCESS_ERR:
		return IB_WC_LOC_ACCESS_ERR;
	case RDMA_CQE_RESP_STS_LOCAL_LENGTH_ERR:
		return IB_WC_LOC_LEN_ERR;
	case RDMA_CQE_RESP_STS_LOCAL_QP_OPERATION_ERR:
		return IB_WC_LOC_QP_OP_ERR;
	case RDMA_CQE_RESP_STS_LOCAL_PROTECTION_ERR:
		return IB_WC_LOC_PROT_ERR;
	case RDMA_CQE_RESP_STS_MEMORY_MGT_OPERATION_ERR:
		return IB_WC_MW_BIND_ERR;
	case RDMA_CQE_RESP_STS_REMOTE_INVALID_REQUEST_ERR:
		return IB_WC_REM_INV_RD_REQ_ERR;
	case RDMA_CQE_RESP_STS_OK:
		return IB_WC_SUCCESS;
	default:
		return IB_WC_GENERAL_ERR;
	}
}

static inline int qedr_set_ok_cqe_resp_wc(struct rdma_cqe_responder *resp,
					  struct ib_wc *wc)
{
	wc->status = IB_WC_SUCCESS;
	wc->byte_len = le32_to_cpu(resp->length);

	if (resp->flags & QEDR_RESP_IMM) {
		wc->ex.imm_data = le32_to_cpu(resp->imm_data_or_inv_r_Key);
		wc->wc_flags |= IB_WC_WITH_IMM;

		if (resp->flags & QEDR_RESP_RDMA)
			wc->opcode = IB_WC_RECV_RDMA_WITH_IMM;

		if (resp->flags & QEDR_RESP_INV)
			return -EINVAL;

	} else if (resp->flags & QEDR_RESP_INV) {
		wc->ex.imm_data = le32_to_cpu(resp->imm_data_or_inv_r_Key);
		wc->wc_flags |= IB_WC_WITH_INVALIDATE;

		if (resp->flags & QEDR_RESP_RDMA)
			return -EINVAL;

	} else if (resp->flags & QEDR_RESP_RDMA) {
		return -EINVAL;
	}

	return 0;
}

static void __process_resp_one(struct qedr_dev *dev, struct qedr_qp *qp,
			       struct qedr_cq *cq, struct ib_wc *wc,
			       struct rdma_cqe_responder *resp, u64 wr_id)
{
	/* Must fill fields before qedr_set_ok_cqe_resp_wc() */
	wc->opcode = IB_WC_RECV;
	wc->wc_flags = 0;

	if (likely(resp->status == RDMA_CQE_RESP_STS_OK)) {
		if (qedr_set_ok_cqe_resp_wc(resp, wc))
			DP_ERR(dev,
			       "CQ %p (icid=%d) has invalid CQE responder flags=0x%x\n",
			       cq, cq->icid, resp->flags);

	} else {
		wc->status = qedr_cqe_resp_status_to_ib(resp->status);
		if (wc->status == IB_WC_GENERAL_ERR)
			DP_ERR(dev,
			       "CQ %p (icid=%d) contains an invalid CQE status %d\n",
			       cq, cq->icid, resp->status);
	}

	/* Fill the rest of the WC */
	wc->vendor_err = 0;
	wc->src_qp = qp->id;
	wc->qp = &qp->ibqp;
	wc->wr_id = wr_id;
}

static int process_resp_one(struct qedr_dev *dev, struct qedr_qp *qp,
			    struct qedr_cq *cq, struct ib_wc *wc,
			    struct rdma_cqe_responder *resp)
{
	u64 wr_id = qp->rqe_wr_id[qp->rq.cons].wr_id;

	__process_resp_one(dev, qp, cq, wc, resp, wr_id);

	while (qp->rqe_wr_id[qp->rq.cons].wqe_size--)
		qed_chain_consume(&qp->rq.pbl);
	qedr_inc_sw_cons(&qp->rq);

	return 1;
}

static int process_resp_flush(struct qedr_qp *qp, struct qedr_cq *cq,
			      int num_entries, struct ib_wc *wc, u16 hw_cons)
{
	u16 cnt = 0;

	while (num_entries && qp->rq.wqe_cons != hw_cons) {
		/* fill WC */
		wc->status = IB_WC_WR_FLUSH_ERR;
		wc->vendor_err = 0;
		wc->wc_flags = 0;
		wc->src_qp = qp->id;
		wc->byte_len = 0;
		wc->wr_id = qp->rqe_wr_id[qp->rq.cons].wr_id;
		wc->qp = &qp->ibqp;
		num_entries--;
		wc++;
		cnt++;
		while (qp->rqe_wr_id[qp->rq.cons].wqe_size--)
			qed_chain_consume(&qp->rq.pbl);
		qedr_inc_sw_cons(&qp->rq);
	}

	return cnt;
}

static void try_consume_resp_cqe(struct qedr_cq *cq, struct qedr_qp *qp,
				 struct rdma_cqe_responder *resp, int *update)
{
	if (le16_to_cpu(resp->rq_cons) == qp->rq.wqe_cons) {
		consume_cqe(cq);
		*update |= 1;
	}
}

static int qedr_poll_cq_resp(struct qedr_dev *dev, struct qedr_qp *qp,
			     struct qedr_cq *cq, int num_entries,
			     struct ib_wc *wc, struct rdma_cqe_responder *resp,
			     int *update)
{
	int cnt;

	if (resp->status == RDMA_CQE_RESP_STS_WORK_REQUEST_FLUSHED_ERR) {
		cnt = process_resp_flush(qp, cq, num_entries, wc,
					 resp->rq_cons);
		try_consume_resp_cqe(cq, qp, resp, update);
	} else {
		cnt = process_resp_one(dev, qp, cq, wc, resp);
		consume_cqe(cq);
		*update |= 1;
	}

	return cnt;
}

static void try_consume_req_cqe(struct qedr_cq *cq, struct qedr_qp *qp,
				struct rdma_cqe_requester *req, int *update)
{
	if (le16_to_cpu(req->sq_cons) == qp->sq.wqe_cons) {
		consume_cqe(cq);
		*update |= 1;
	}
}

int qedr_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *wc)
{
	struct qedr_dev *dev = get_qedr_dev(ibcq->device);
	struct qedr_cq *cq = get_qedr_cq(ibcq);
	union rdma_cqe *cqe = cq->latest_cqe;
	u32 old_cons, new_cons;
	unsigned long flags;
	int update = 0;
	int done = 0;

	if (cq->destroyed) {
		DP_ERR(dev,
		       "warning: poll was invoked after destroy for cq %p (icid=%d)\n",
		       cq, cq->icid);
		return 0;
	}

	if (cq->cq_type == QEDR_CQ_TYPE_GSI)
		return qedr_gsi_poll_cq(ibcq, num_entries, wc);

	spin_lock_irqsave(&cq->cq_lock, flags);
	old_cons = qed_chain_get_cons_idx_u32(&cq->pbl);
	while (num_entries && is_valid_cqe(cq, cqe)) {
		struct qedr_qp *qp;
		int cnt = 0;

		/* prevent speculative reads of any field of CQE */
		rmb();

		qp = cqe_get_qp(cqe);
		if (!qp) {
			WARN(1, "Error: CQE QP pointer is NULL. CQE=%p\n", cqe);
			break;
		}

		wc->qp = &qp->ibqp;

		switch (cqe_get_type(cqe)) {
		case RDMA_CQE_TYPE_REQUESTER:
			cnt = qedr_poll_cq_req(dev, qp, cq, num_entries, wc,
					       &cqe->req);
			try_consume_req_cqe(cq, qp, &cqe->req, &update);
			break;
		case RDMA_CQE_TYPE_RESPONDER_RQ:
			cnt = qedr_poll_cq_resp(dev, qp, cq, num_entries, wc,
						&cqe->resp, &update);
			break;
		case RDMA_CQE_TYPE_INVALID:
		default:
			DP_ERR(dev, "Error: invalid CQE type = %d\n",
			       cqe_get_type(cqe));
		}
		num_entries -= cnt;
		wc += cnt;
		done += cnt;

		cqe = get_cqe(cq);
	}
	new_cons = qed_chain_get_cons_idx_u32(&cq->pbl);

	cq->cq_cons += new_cons - old_cons;

	if (update)
		/* doorbell notifies abount latest VALID entry,
		 * but chain already point to the next INVALID one
		 */
		doorbell_cq(cq, cq->cq_cons - 1, cq->arm_flags);

	spin_unlock_irqrestore(&cq->cq_lock, flags);
	return done;
}

int qedr_process_mad(struct ib_device *ibdev, int process_mad_flags,
		     u8 port_num,
		     const struct ib_wc *in_wc,
		     const struct ib_grh *in_grh,
		     const struct ib_mad_hdr *mad_hdr,
		     size_t in_mad_size, struct ib_mad_hdr *out_mad,
		     size_t *out_mad_size, u16 *out_mad_pkey_index)
{
	struct qedr_dev *dev = get_qedr_dev(ibdev);

	DP_DEBUG(dev, QEDR_MSG_GSI,
		 "QEDR_PROCESS_MAD in_mad %x %x %x %x %x %x %x %x\n",
		 mad_hdr->attr_id, mad_hdr->base_version, mad_hdr->attr_mod,
		 mad_hdr->class_specific, mad_hdr->class_version,
		 mad_hdr->method, mad_hdr->mgmt_class, mad_hdr->status);
	return IB_MAD_RESULT_SUCCESS;
}

int qedr_port_immutable(struct ib_device *ibdev, u8 port_num,
			struct ib_port_immutable *immutable)
{
	struct ib_port_attr attr;
	int err;

	immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE |
				    RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP;

	err = ib_query_port(ibdev, port_num, &attr);
	if (err)
		return err;

	immutable->pkey_tbl_len = attr.pkey_tbl_len;
	immutable->gid_tbl_len = attr.gid_tbl_len;
	immutable->max_mad_size = IB_MGMT_MAD_SIZE;

	return 0;
}