summaryrefslogtreecommitdiff
path: root/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c
blob: 88f7acce38dcb98858de96a753bda558ddf99af2 (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
// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
/* Copyright 2014-2016 Freescale Semiconductor Inc.
 * Copyright 2016-2017 NXP
 */
#include <linux/init.h>
#include <linux/module.h>
#include <linux/platform_device.h>
#include <linux/etherdevice.h>
#include <linux/of_net.h>
#include <linux/interrupt.h>
#include <linux/msi.h>
#include <linux/kthread.h>
#include <linux/iommu.h>
#include <linux/net_tstamp.h>
#include <linux/fsl/mc.h>

#include <net/sock.h>

#include "dpaa2-eth.h"

/* CREATE_TRACE_POINTS only needs to be defined once. Other dpa files
 * using trace events only need to #include <trace/events/sched.h>
 */
#define CREATE_TRACE_POINTS
#include "dpaa2-eth-trace.h"

MODULE_LICENSE("Dual BSD/GPL");
MODULE_AUTHOR("Freescale Semiconductor, Inc");
MODULE_DESCRIPTION("Freescale DPAA2 Ethernet Driver");

static void *dpaa2_iova_to_virt(struct iommu_domain *domain,
				dma_addr_t iova_addr)
{
	phys_addr_t phys_addr;

	phys_addr = domain ? iommu_iova_to_phys(domain, iova_addr) : iova_addr;

	return phys_to_virt(phys_addr);
}

static void validate_rx_csum(struct dpaa2_eth_priv *priv,
			     u32 fd_status,
			     struct sk_buff *skb)
{
	skb_checksum_none_assert(skb);

	/* HW checksum validation is disabled, nothing to do here */
	if (!(priv->net_dev->features & NETIF_F_RXCSUM))
		return;

	/* Read checksum validation bits */
	if (!((fd_status & DPAA2_FAS_L3CV) &&
	      (fd_status & DPAA2_FAS_L4CV)))
		return;

	/* Inform the stack there's no need to compute L3/L4 csum anymore */
	skb->ip_summed = CHECKSUM_UNNECESSARY;
}

/* Free a received FD.
 * Not to be used for Tx conf FDs or on any other paths.
 */
static void free_rx_fd(struct dpaa2_eth_priv *priv,
		       const struct dpaa2_fd *fd,
		       void *vaddr)
{
	struct device *dev = priv->net_dev->dev.parent;
	dma_addr_t addr = dpaa2_fd_get_addr(fd);
	u8 fd_format = dpaa2_fd_get_format(fd);
	struct dpaa2_sg_entry *sgt;
	void *sg_vaddr;
	int i;

	/* If single buffer frame, just free the data buffer */
	if (fd_format == dpaa2_fd_single)
		goto free_buf;
	else if (fd_format != dpaa2_fd_sg)
		/* We don't support any other format */
		return;

	/* For S/G frames, we first need to free all SG entries
	 * except the first one, which was taken care of already
	 */
	sgt = vaddr + dpaa2_fd_get_offset(fd);
	for (i = 1; i < DPAA2_ETH_MAX_SG_ENTRIES; i++) {
		addr = dpaa2_sg_get_addr(&sgt[i]);
		sg_vaddr = dpaa2_iova_to_virt(priv->iommu_domain, addr);
		dma_unmap_single(dev, addr, DPAA2_ETH_RX_BUF_SIZE,
				 DMA_FROM_DEVICE);

		skb_free_frag(sg_vaddr);
		if (dpaa2_sg_is_final(&sgt[i]))
			break;
	}

free_buf:
	skb_free_frag(vaddr);
}

/* Build a linear skb based on a single-buffer frame descriptor */
static struct sk_buff *build_linear_skb(struct dpaa2_eth_channel *ch,
					const struct dpaa2_fd *fd,
					void *fd_vaddr)
{
	struct sk_buff *skb = NULL;
	u16 fd_offset = dpaa2_fd_get_offset(fd);
	u32 fd_length = dpaa2_fd_get_len(fd);

	ch->buf_count--;

	skb = build_skb(fd_vaddr, DPAA2_ETH_SKB_SIZE);
	if (unlikely(!skb))
		return NULL;

	skb_reserve(skb, fd_offset);
	skb_put(skb, fd_length);

	return skb;
}

/* Build a non linear (fragmented) skb based on a S/G table */
static struct sk_buff *build_frag_skb(struct dpaa2_eth_priv *priv,
				      struct dpaa2_eth_channel *ch,
				      struct dpaa2_sg_entry *sgt)
{
	struct sk_buff *skb = NULL;
	struct device *dev = priv->net_dev->dev.parent;
	void *sg_vaddr;
	dma_addr_t sg_addr;
	u16 sg_offset;
	u32 sg_length;
	struct page *page, *head_page;
	int page_offset;
	int i;

	for (i = 0; i < DPAA2_ETH_MAX_SG_ENTRIES; i++) {
		struct dpaa2_sg_entry *sge = &sgt[i];

		/* NOTE: We only support SG entries in dpaa2_sg_single format,
		 * but this is the only format we may receive from HW anyway
		 */

		/* Get the address and length from the S/G entry */
		sg_addr = dpaa2_sg_get_addr(sge);
		sg_vaddr = dpaa2_iova_to_virt(priv->iommu_domain, sg_addr);
		dma_unmap_single(dev, sg_addr, DPAA2_ETH_RX_BUF_SIZE,
				 DMA_FROM_DEVICE);

		sg_length = dpaa2_sg_get_len(sge);

		if (i == 0) {
			/* We build the skb around the first data buffer */
			skb = build_skb(sg_vaddr, DPAA2_ETH_SKB_SIZE);
			if (unlikely(!skb)) {
				/* Free the first SG entry now, since we already
				 * unmapped it and obtained the virtual address
				 */
				skb_free_frag(sg_vaddr);

				/* We still need to subtract the buffers used
				 * by this FD from our software counter
				 */
				while (!dpaa2_sg_is_final(&sgt[i]) &&
				       i < DPAA2_ETH_MAX_SG_ENTRIES)
					i++;
				break;
			}

			sg_offset = dpaa2_sg_get_offset(sge);
			skb_reserve(skb, sg_offset);
			skb_put(skb, sg_length);
		} else {
			/* Rest of the data buffers are stored as skb frags */
			page = virt_to_page(sg_vaddr);
			head_page = virt_to_head_page(sg_vaddr);

			/* Offset in page (which may be compound).
			 * Data in subsequent SG entries is stored from the
			 * beginning of the buffer, so we don't need to add the
			 * sg_offset.
			 */
			page_offset = ((unsigned long)sg_vaddr &
				(PAGE_SIZE - 1)) +
				(page_address(page) - page_address(head_page));

			skb_add_rx_frag(skb, i - 1, head_page, page_offset,
					sg_length, DPAA2_ETH_RX_BUF_SIZE);
		}

		if (dpaa2_sg_is_final(sge))
			break;
	}

	WARN_ONCE(i == DPAA2_ETH_MAX_SG_ENTRIES, "Final bit not set in SGT");

	/* Count all data buffers + SG table buffer */
	ch->buf_count -= i + 2;

	return skb;
}

/* Main Rx frame processing routine */
static void dpaa2_eth_rx(struct dpaa2_eth_priv *priv,
			 struct dpaa2_eth_channel *ch,
			 const struct dpaa2_fd *fd,
			 struct napi_struct *napi,
			 u16 queue_id)
{
	dma_addr_t addr = dpaa2_fd_get_addr(fd);
	u8 fd_format = dpaa2_fd_get_format(fd);
	void *vaddr;
	struct sk_buff *skb;
	struct rtnl_link_stats64 *percpu_stats;
	struct dpaa2_eth_drv_stats *percpu_extras;
	struct device *dev = priv->net_dev->dev.parent;
	struct dpaa2_fas *fas;
	void *buf_data;
	u32 status = 0;

	/* Tracing point */
	trace_dpaa2_rx_fd(priv->net_dev, fd);

	vaddr = dpaa2_iova_to_virt(priv->iommu_domain, addr);
	dma_unmap_single(dev, addr, DPAA2_ETH_RX_BUF_SIZE, DMA_FROM_DEVICE);

	fas = dpaa2_get_fas(vaddr, false);
	prefetch(fas);
	buf_data = vaddr + dpaa2_fd_get_offset(fd);
	prefetch(buf_data);

	percpu_stats = this_cpu_ptr(priv->percpu_stats);
	percpu_extras = this_cpu_ptr(priv->percpu_extras);

	if (fd_format == dpaa2_fd_single) {
		skb = build_linear_skb(ch, fd, vaddr);
	} else if (fd_format == dpaa2_fd_sg) {
		skb = build_frag_skb(priv, ch, buf_data);
		skb_free_frag(vaddr);
		percpu_extras->rx_sg_frames++;
		percpu_extras->rx_sg_bytes += dpaa2_fd_get_len(fd);
	} else {
		/* We don't support any other format */
		goto err_frame_format;
	}

	if (unlikely(!skb))
		goto err_build_skb;

	prefetch(skb->data);

	/* Get the timestamp value */
	if (priv->rx_tstamp) {
		struct skb_shared_hwtstamps *shhwtstamps = skb_hwtstamps(skb);
		__le64 *ts = dpaa2_get_ts(vaddr, false);
		u64 ns;

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

		ns = DPAA2_PTP_CLK_PERIOD_NS * le64_to_cpup(ts);
		shhwtstamps->hwtstamp = ns_to_ktime(ns);
	}

	/* Check if we need to validate the L4 csum */
	if (likely(dpaa2_fd_get_frc(fd) & DPAA2_FD_FRC_FASV)) {
		status = le32_to_cpu(fas->status);
		validate_rx_csum(priv, status, skb);
	}

	skb->protocol = eth_type_trans(skb, priv->net_dev);
	skb_record_rx_queue(skb, queue_id);

	percpu_stats->rx_packets++;
	percpu_stats->rx_bytes += dpaa2_fd_get_len(fd);

	napi_gro_receive(napi, skb);

	return;

err_build_skb:
	free_rx_fd(priv, fd, vaddr);
err_frame_format:
	percpu_stats->rx_dropped++;
}

/* Consume all frames pull-dequeued into the store. This is the simplest way to
 * make sure we don't accidentally issue another volatile dequeue which would
 * overwrite (leak) frames already in the store.
 *
 * Observance of NAPI budget is not our concern, leaving that to the caller.
 */
static int consume_frames(struct dpaa2_eth_channel *ch,
			  enum dpaa2_eth_fq_type *type)
{
	struct dpaa2_eth_priv *priv = ch->priv;
	struct dpaa2_eth_fq *fq = NULL;
	struct dpaa2_dq *dq;
	const struct dpaa2_fd *fd;
	int cleaned = 0;
	int is_last;

	do {
		dq = dpaa2_io_store_next(ch->store, &is_last);
		if (unlikely(!dq)) {
			/* If we're here, we *must* have placed a
			 * volatile dequeue comnmand, so keep reading through
			 * the store until we get some sort of valid response
			 * token (either a valid frame or an "empty dequeue")
			 */
			continue;
		}

		fd = dpaa2_dq_fd(dq);
		fq = (struct dpaa2_eth_fq *)(uintptr_t)dpaa2_dq_fqd_ctx(dq);

		fq->consume(priv, ch, fd, &ch->napi, fq->flowid);
		cleaned++;
	} while (!is_last);

	if (!cleaned)
		return 0;

	fq->stats.frames += cleaned;
	ch->stats.frames += cleaned;

	/* A dequeue operation only pulls frames from a single queue
	 * into the store. Return the frame queue type as an out param.
	 */
	if (type)
		*type = fq->type;

	return cleaned;
}

/* Configure the egress frame annotation for timestamp update */
static void enable_tx_tstamp(struct dpaa2_fd *fd, void *buf_start)
{
	struct dpaa2_faead *faead;
	u32 ctrl, frc;

	/* Mark the egress frame annotation area as valid */
	frc = dpaa2_fd_get_frc(fd);
	dpaa2_fd_set_frc(fd, frc | DPAA2_FD_FRC_FAEADV);

	/* Set hardware annotation size */
	ctrl = dpaa2_fd_get_ctrl(fd);
	dpaa2_fd_set_ctrl(fd, ctrl | DPAA2_FD_CTRL_ASAL);

	/* enable UPD (update prepanded data) bit in FAEAD field of
	 * hardware frame annotation area
	 */
	ctrl = DPAA2_FAEAD_A2V | DPAA2_FAEAD_UPDV | DPAA2_FAEAD_UPD;
	faead = dpaa2_get_faead(buf_start, true);
	faead->ctrl = cpu_to_le32(ctrl);
}

/* Create a frame descriptor based on a fragmented skb */
static int build_sg_fd(struct dpaa2_eth_priv *priv,
		       struct sk_buff *skb,
		       struct dpaa2_fd *fd)
{
	struct device *dev = priv->net_dev->dev.parent;
	void *sgt_buf = NULL;
	dma_addr_t addr;
	int nr_frags = skb_shinfo(skb)->nr_frags;
	struct dpaa2_sg_entry *sgt;
	int i, err;
	int sgt_buf_size;
	struct scatterlist *scl, *crt_scl;
	int num_sg;
	int num_dma_bufs;
	struct dpaa2_eth_swa *swa;

	/* Create and map scatterlist.
	 * We don't advertise NETIF_F_FRAGLIST, so skb_to_sgvec() will not have
	 * to go beyond nr_frags+1.
	 * Note: We don't support chained scatterlists
	 */
	if (unlikely(PAGE_SIZE / sizeof(struct scatterlist) < nr_frags + 1))
		return -EINVAL;

	scl = kcalloc(nr_frags + 1, sizeof(struct scatterlist), GFP_ATOMIC);
	if (unlikely(!scl))
		return -ENOMEM;

	sg_init_table(scl, nr_frags + 1);
	num_sg = skb_to_sgvec(skb, scl, 0, skb->len);
	num_dma_bufs = dma_map_sg(dev, scl, num_sg, DMA_BIDIRECTIONAL);
	if (unlikely(!num_dma_bufs)) {
		err = -ENOMEM;
		goto dma_map_sg_failed;
	}

	/* Prepare the HW SGT structure */
	sgt_buf_size = priv->tx_data_offset +
		       sizeof(struct dpaa2_sg_entry) *  num_dma_bufs;
	sgt_buf = netdev_alloc_frag(sgt_buf_size + DPAA2_ETH_TX_BUF_ALIGN);
	if (unlikely(!sgt_buf)) {
		err = -ENOMEM;
		goto sgt_buf_alloc_failed;
	}
	sgt_buf = PTR_ALIGN(sgt_buf, DPAA2_ETH_TX_BUF_ALIGN);
	memset(sgt_buf, 0, sgt_buf_size);

	sgt = (struct dpaa2_sg_entry *)(sgt_buf + priv->tx_data_offset);

	/* Fill in the HW SGT structure.
	 *
	 * sgt_buf is zeroed out, so the following fields are implicit
	 * in all sgt entries:
	 *   - offset is 0
	 *   - format is 'dpaa2_sg_single'
	 */
	for_each_sg(scl, crt_scl, num_dma_bufs, i) {
		dpaa2_sg_set_addr(&sgt[i], sg_dma_address(crt_scl));
		dpaa2_sg_set_len(&sgt[i], sg_dma_len(crt_scl));
	}
	dpaa2_sg_set_final(&sgt[i - 1], true);

	/* Store the skb backpointer in the SGT buffer.
	 * Fit the scatterlist and the number of buffers alongside the
	 * skb backpointer in the software annotation area. We'll need
	 * all of them on Tx Conf.
	 */
	swa = (struct dpaa2_eth_swa *)sgt_buf;
	swa->skb = skb;
	swa->scl = scl;
	swa->num_sg = num_sg;
	swa->sgt_size = sgt_buf_size;

	/* Separately map the SGT buffer */
	addr = dma_map_single(dev, sgt_buf, sgt_buf_size, DMA_BIDIRECTIONAL);
	if (unlikely(dma_mapping_error(dev, addr))) {
		err = -ENOMEM;
		goto dma_map_single_failed;
	}
	dpaa2_fd_set_offset(fd, priv->tx_data_offset);
	dpaa2_fd_set_format(fd, dpaa2_fd_sg);
	dpaa2_fd_set_addr(fd, addr);
	dpaa2_fd_set_len(fd, skb->len);
	dpaa2_fd_set_ctrl(fd, FD_CTRL_PTA);

	if (priv->tx_tstamp && skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)
		enable_tx_tstamp(fd, sgt_buf);

	return 0;

dma_map_single_failed:
	skb_free_frag(sgt_buf);
sgt_buf_alloc_failed:
	dma_unmap_sg(dev, scl, num_sg, DMA_BIDIRECTIONAL);
dma_map_sg_failed:
	kfree(scl);
	return err;
}

/* Create a frame descriptor based on a linear skb */
static int build_single_fd(struct dpaa2_eth_priv *priv,
			   struct sk_buff *skb,
			   struct dpaa2_fd *fd)
{
	struct device *dev = priv->net_dev->dev.parent;
	u8 *buffer_start, *aligned_start;
	struct sk_buff **skbh;
	dma_addr_t addr;

	buffer_start = skb->data - dpaa2_eth_needed_headroom(priv, skb);

	/* If there's enough room to align the FD address, do it.
	 * It will help hardware optimize accesses.
	 */
	aligned_start = PTR_ALIGN(buffer_start - DPAA2_ETH_TX_BUF_ALIGN,
				  DPAA2_ETH_TX_BUF_ALIGN);
	if (aligned_start >= skb->head)
		buffer_start = aligned_start;

	/* Store a backpointer to the skb at the beginning of the buffer
	 * (in the private data area) such that we can release it
	 * on Tx confirm
	 */
	skbh = (struct sk_buff **)buffer_start;
	*skbh = skb;

	addr = dma_map_single(dev, buffer_start,
			      skb_tail_pointer(skb) - buffer_start,
			      DMA_BIDIRECTIONAL);
	if (unlikely(dma_mapping_error(dev, addr)))
		return -ENOMEM;

	dpaa2_fd_set_addr(fd, addr);
	dpaa2_fd_set_offset(fd, (u16)(skb->data - buffer_start));
	dpaa2_fd_set_len(fd, skb->len);
	dpaa2_fd_set_format(fd, dpaa2_fd_single);
	dpaa2_fd_set_ctrl(fd, FD_CTRL_PTA);

	if (priv->tx_tstamp && skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)
		enable_tx_tstamp(fd, buffer_start);

	return 0;
}

/* FD freeing routine on the Tx path
 *
 * DMA-unmap and free FD and possibly SGT buffer allocated on Tx. The skb
 * back-pointed to is also freed.
 * This can be called either from dpaa2_eth_tx_conf() or on the error path of
 * dpaa2_eth_tx().
 */
static void free_tx_fd(const struct dpaa2_eth_priv *priv,
		       const struct dpaa2_fd *fd)
{
	struct device *dev = priv->net_dev->dev.parent;
	dma_addr_t fd_addr;
	struct sk_buff **skbh, *skb;
	unsigned char *buffer_start;
	struct dpaa2_eth_swa *swa;
	u8 fd_format = dpaa2_fd_get_format(fd);

	fd_addr = dpaa2_fd_get_addr(fd);
	skbh = dpaa2_iova_to_virt(priv->iommu_domain, fd_addr);

	if (fd_format == dpaa2_fd_single) {
		skb = *skbh;
		buffer_start = (unsigned char *)skbh;
		/* Accessing the skb buffer is safe before dma unmap, because
		 * we didn't map the actual skb shell.
		 */
		dma_unmap_single(dev, fd_addr,
				 skb_tail_pointer(skb) - buffer_start,
				 DMA_BIDIRECTIONAL);
	} else if (fd_format == dpaa2_fd_sg) {
		swa = (struct dpaa2_eth_swa *)skbh;
		skb = swa->skb;

		/* Unmap the scatterlist */
		dma_unmap_sg(dev, swa->scl, swa->num_sg, DMA_BIDIRECTIONAL);
		kfree(swa->scl);

		/* Unmap the SGT buffer */
		dma_unmap_single(dev, fd_addr, swa->sgt_size,
				 DMA_BIDIRECTIONAL);
	} else {
		netdev_dbg(priv->net_dev, "Invalid FD format\n");
		return;
	}

	/* Get the timestamp value */
	if (priv->tx_tstamp && skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) {
		struct skb_shared_hwtstamps shhwtstamps;
		__le64 *ts = dpaa2_get_ts(skbh, true);
		u64 ns;

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

		ns = DPAA2_PTP_CLK_PERIOD_NS * le64_to_cpup(ts);
		shhwtstamps.hwtstamp = ns_to_ktime(ns);
		skb_tstamp_tx(skb, &shhwtstamps);
	}

	/* Free SGT buffer allocated on tx */
	if (fd_format != dpaa2_fd_single)
		skb_free_frag(skbh);

	/* Move on with skb release */
	dev_kfree_skb(skb);
}

static netdev_tx_t dpaa2_eth_tx(struct sk_buff *skb, struct net_device *net_dev)
{
	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
	struct dpaa2_fd fd;
	struct rtnl_link_stats64 *percpu_stats;
	struct dpaa2_eth_drv_stats *percpu_extras;
	struct dpaa2_eth_fq *fq;
	u16 queue_mapping;
	unsigned int needed_headroom;
	int err, i;

	percpu_stats = this_cpu_ptr(priv->percpu_stats);
	percpu_extras = this_cpu_ptr(priv->percpu_extras);

	needed_headroom = dpaa2_eth_needed_headroom(priv, skb);
	if (skb_headroom(skb) < needed_headroom) {
		struct sk_buff *ns;

		ns = skb_realloc_headroom(skb, needed_headroom);
		if (unlikely(!ns)) {
			percpu_stats->tx_dropped++;
			goto err_alloc_headroom;
		}
		percpu_extras->tx_reallocs++;

		if (skb->sk)
			skb_set_owner_w(ns, skb->sk);

		dev_kfree_skb(skb);
		skb = ns;
	}

	/* We'll be holding a back-reference to the skb until Tx Confirmation;
	 * we don't want that overwritten by a concurrent Tx with a cloned skb.
	 */
	skb = skb_unshare(skb, GFP_ATOMIC);
	if (unlikely(!skb)) {
		/* skb_unshare() has already freed the skb */
		percpu_stats->tx_dropped++;
		return NETDEV_TX_OK;
	}

	/* Setup the FD fields */
	memset(&fd, 0, sizeof(fd));

	if (skb_is_nonlinear(skb)) {
		err = build_sg_fd(priv, skb, &fd);
		percpu_extras->tx_sg_frames++;
		percpu_extras->tx_sg_bytes += skb->len;
	} else {
		err = build_single_fd(priv, skb, &fd);
	}

	if (unlikely(err)) {
		percpu_stats->tx_dropped++;
		goto err_build_fd;
	}

	/* Tracing point */
	trace_dpaa2_tx_fd(net_dev, &fd);

	/* TxConf FQ selection relies on queue id from the stack.
	 * In case of a forwarded frame from another DPNI interface, we choose
	 * a queue affined to the same core that processed the Rx frame
	 */
	queue_mapping = skb_get_queue_mapping(skb);
	fq = &priv->fq[queue_mapping];
	for (i = 0; i < DPAA2_ETH_ENQUEUE_RETRIES; i++) {
		err = dpaa2_io_service_enqueue_qd(fq->channel->dpio,
						  priv->tx_qdid, 0,
						  fq->tx_qdbin, &fd);
		if (err != -EBUSY)
			break;
	}
	percpu_extras->tx_portal_busy += i;
	if (unlikely(err < 0)) {
		percpu_stats->tx_errors++;
		/* Clean up everything, including freeing the skb */
		free_tx_fd(priv, &fd);
	} else {
		percpu_stats->tx_packets++;
		percpu_stats->tx_bytes += dpaa2_fd_get_len(&fd);
	}

	return NETDEV_TX_OK;

err_build_fd:
err_alloc_headroom:
	dev_kfree_skb(skb);

	return NETDEV_TX_OK;
}

/* Tx confirmation frame processing routine */
static void dpaa2_eth_tx_conf(struct dpaa2_eth_priv *priv,
			      struct dpaa2_eth_channel *ch __always_unused,
			      const struct dpaa2_fd *fd,
			      struct napi_struct *napi __always_unused,
			      u16 queue_id __always_unused)
{
	struct rtnl_link_stats64 *percpu_stats;
	struct dpaa2_eth_drv_stats *percpu_extras;
	u32 fd_errors;

	/* Tracing point */
	trace_dpaa2_tx_conf_fd(priv->net_dev, fd);

	percpu_extras = this_cpu_ptr(priv->percpu_extras);
	percpu_extras->tx_conf_frames++;
	percpu_extras->tx_conf_bytes += dpaa2_fd_get_len(fd);

	/* Check frame errors in the FD field */
	fd_errors = dpaa2_fd_get_ctrl(fd) & DPAA2_FD_TX_ERR_MASK;
	free_tx_fd(priv, fd);

	if (likely(!fd_errors))
		return;

	if (net_ratelimit())
		netdev_dbg(priv->net_dev, "TX frame FD error: 0x%08x\n",
			   fd_errors);

	percpu_stats = this_cpu_ptr(priv->percpu_stats);
	/* Tx-conf logically pertains to the egress path. */
	percpu_stats->tx_errors++;
}

static int set_rx_csum(struct dpaa2_eth_priv *priv, bool enable)
{
	int err;

	err = dpni_set_offload(priv->mc_io, 0, priv->mc_token,
			       DPNI_OFF_RX_L3_CSUM, enable);
	if (err) {
		netdev_err(priv->net_dev,
			   "dpni_set_offload(RX_L3_CSUM) failed\n");
		return err;
	}

	err = dpni_set_offload(priv->mc_io, 0, priv->mc_token,
			       DPNI_OFF_RX_L4_CSUM, enable);
	if (err) {
		netdev_err(priv->net_dev,
			   "dpni_set_offload(RX_L4_CSUM) failed\n");
		return err;
	}

	return 0;
}

static int set_tx_csum(struct dpaa2_eth_priv *priv, bool enable)
{
	int err;

	err = dpni_set_offload(priv->mc_io, 0, priv->mc_token,
			       DPNI_OFF_TX_L3_CSUM, enable);
	if (err) {
		netdev_err(priv->net_dev, "dpni_set_offload(TX_L3_CSUM) failed\n");
		return err;
	}

	err = dpni_set_offload(priv->mc_io, 0, priv->mc_token,
			       DPNI_OFF_TX_L4_CSUM, enable);
	if (err) {
		netdev_err(priv->net_dev, "dpni_set_offload(TX_L4_CSUM) failed\n");
		return err;
	}

	return 0;
}

/* Free buffers acquired from the buffer pool or which were meant to
 * be released in the pool
 */
static void free_bufs(struct dpaa2_eth_priv *priv, u64 *buf_array, int count)
{
	struct device *dev = priv->net_dev->dev.parent;
	void *vaddr;
	int i;

	for (i = 0; i < count; i++) {
		vaddr = dpaa2_iova_to_virt(priv->iommu_domain, buf_array[i]);
		dma_unmap_single(dev, buf_array[i], DPAA2_ETH_RX_BUF_SIZE,
				 DMA_FROM_DEVICE);
		skb_free_frag(vaddr);
	}
}

/* Perform a single release command to add buffers
 * to the specified buffer pool
 */
static int add_bufs(struct dpaa2_eth_priv *priv,
		    struct dpaa2_eth_channel *ch, u16 bpid)
{
	struct device *dev = priv->net_dev->dev.parent;
	u64 buf_array[DPAA2_ETH_BUFS_PER_CMD];
	void *buf;
	dma_addr_t addr;
	int i, err;

	for (i = 0; i < DPAA2_ETH_BUFS_PER_CMD; i++) {
		/* Allocate buffer visible to WRIOP + skb shared info +
		 * alignment padding
		 */
		buf = napi_alloc_frag(dpaa2_eth_buf_raw_size(priv));
		if (unlikely(!buf))
			goto err_alloc;

		buf = PTR_ALIGN(buf, priv->rx_buf_align);

		addr = dma_map_single(dev, buf, DPAA2_ETH_RX_BUF_SIZE,
				      DMA_FROM_DEVICE);
		if (unlikely(dma_mapping_error(dev, addr)))
			goto err_map;

		buf_array[i] = addr;

		/* tracing point */
		trace_dpaa2_eth_buf_seed(priv->net_dev,
					 buf, dpaa2_eth_buf_raw_size(priv),
					 addr, DPAA2_ETH_RX_BUF_SIZE,
					 bpid);
	}

release_bufs:
	/* In case the portal is busy, retry until successful */
	while ((err = dpaa2_io_service_release(ch->dpio, bpid,
					       buf_array, i)) == -EBUSY)
		cpu_relax();

	/* If release command failed, clean up and bail out;
	 * not much else we can do about it
	 */
	if (err) {
		free_bufs(priv, buf_array, i);
		return 0;
	}

	return i;

err_map:
	skb_free_frag(buf);
err_alloc:
	/* If we managed to allocate at least some buffers,
	 * release them to hardware
	 */
	if (i)
		goto release_bufs;

	return 0;
}

static int seed_pool(struct dpaa2_eth_priv *priv, u16 bpid)
{
	int i, j;
	int new_count;

	/* This is the lazy seeding of Rx buffer pools.
	 * dpaa2_add_bufs() is also used on the Rx hotpath and calls
	 * napi_alloc_frag(). The trouble with that is that it in turn ends up
	 * calling this_cpu_ptr(), which mandates execution in atomic context.
	 * Rather than splitting up the code, do a one-off preempt disable.
	 */
	preempt_disable();
	for (j = 0; j < priv->num_channels; j++) {
		for (i = 0; i < DPAA2_ETH_NUM_BUFS;
		     i += DPAA2_ETH_BUFS_PER_CMD) {
			new_count = add_bufs(priv, priv->channel[j], bpid);
			priv->channel[j]->buf_count += new_count;

			if (new_count < DPAA2_ETH_BUFS_PER_CMD) {
				preempt_enable();
				return -ENOMEM;
			}
		}
	}
	preempt_enable();

	return 0;
}

/**
 * Drain the specified number of buffers from the DPNI's private buffer pool.
 * @count must not exceeed DPAA2_ETH_BUFS_PER_CMD
 */
static void drain_bufs(struct dpaa2_eth_priv *priv, int count)
{
	u64 buf_array[DPAA2_ETH_BUFS_PER_CMD];
	int ret;

	do {
		ret = dpaa2_io_service_acquire(NULL, priv->bpid,
					       buf_array, count);
		if (ret < 0) {
			netdev_err(priv->net_dev, "dpaa2_io_service_acquire() failed\n");
			return;
		}
		free_bufs(priv, buf_array, ret);
	} while (ret);
}

static void drain_pool(struct dpaa2_eth_priv *priv)
{
	int i;

	drain_bufs(priv, DPAA2_ETH_BUFS_PER_CMD);
	drain_bufs(priv, 1);

	for (i = 0; i < priv->num_channels; i++)
		priv->channel[i]->buf_count = 0;
}

/* Function is called from softirq context only, so we don't need to guard
 * the access to percpu count
 */
static int refill_pool(struct dpaa2_eth_priv *priv,
		       struct dpaa2_eth_channel *ch,
		       u16 bpid)
{
	int new_count;

	if (likely(ch->buf_count >= DPAA2_ETH_REFILL_THRESH))
		return 0;

	do {
		new_count = add_bufs(priv, ch, bpid);
		if (unlikely(!new_count)) {
			/* Out of memory; abort for now, we'll try later on */
			break;
		}
		ch->buf_count += new_count;
	} while (ch->buf_count < DPAA2_ETH_NUM_BUFS);

	if (unlikely(ch->buf_count < DPAA2_ETH_NUM_BUFS))
		return -ENOMEM;

	return 0;
}

static int pull_channel(struct dpaa2_eth_channel *ch)
{
	int err;
	int dequeues = -1;

	/* Retry while portal is busy */
	do {
		err = dpaa2_io_service_pull_channel(ch->dpio, ch->ch_id,
						    ch->store);
		dequeues++;
		cpu_relax();
	} while (err == -EBUSY);

	ch->stats.dequeue_portal_busy += dequeues;
	if (unlikely(err))
		ch->stats.pull_err++;

	return err;
}

/* NAPI poll routine
 *
 * Frames are dequeued from the QMan channel associated with this NAPI context.
 * Rx, Tx confirmation and (if configured) Rx error frames all count
 * towards the NAPI budget.
 */
static int dpaa2_eth_poll(struct napi_struct *napi, int budget)
{
	struct dpaa2_eth_channel *ch;
	struct dpaa2_eth_priv *priv;
	int rx_cleaned = 0, txconf_cleaned = 0;
	enum dpaa2_eth_fq_type type = 0;
	int store_cleaned;
	int err;

	ch = container_of(napi, struct dpaa2_eth_channel, napi);
	priv = ch->priv;

	do {
		err = pull_channel(ch);
		if (unlikely(err))
			break;

		/* Refill pool if appropriate */
		refill_pool(priv, ch, priv->bpid);

		store_cleaned = consume_frames(ch, &type);
		if (type == DPAA2_RX_FQ)
			rx_cleaned += store_cleaned;
		else
			txconf_cleaned += store_cleaned;

		/* If we either consumed the whole NAPI budget with Rx frames
		 * or we reached the Tx confirmations threshold, we're done.
		 */
		if (rx_cleaned >= budget ||
		    txconf_cleaned >= DPAA2_ETH_TXCONF_PER_NAPI)
			return budget;
	} while (store_cleaned);

	/* We didn't consume the entire budget, so finish napi and
	 * re-enable data availability notifications
	 */
	napi_complete_done(napi, rx_cleaned);
	do {
		err = dpaa2_io_service_rearm(ch->dpio, &ch->nctx);
		cpu_relax();
	} while (err == -EBUSY);
	WARN_ONCE(err, "CDAN notifications rearm failed on core %d",
		  ch->nctx.desired_cpu);

	return max(rx_cleaned, 1);
}

static void enable_ch_napi(struct dpaa2_eth_priv *priv)
{
	struct dpaa2_eth_channel *ch;
	int i;

	for (i = 0; i < priv->num_channels; i++) {
		ch = priv->channel[i];
		napi_enable(&ch->napi);
	}
}

static void disable_ch_napi(struct dpaa2_eth_priv *priv)
{
	struct dpaa2_eth_channel *ch;
	int i;

	for (i = 0; i < priv->num_channels; i++) {
		ch = priv->channel[i];
		napi_disable(&ch->napi);
	}
}

static int link_state_update(struct dpaa2_eth_priv *priv)
{
	struct dpni_link_state state = {0};
	int err;

	err = dpni_get_link_state(priv->mc_io, 0, priv->mc_token, &state);
	if (unlikely(err)) {
		netdev_err(priv->net_dev,
			   "dpni_get_link_state() failed\n");
		return err;
	}

	/* Chech link state; speed / duplex changes are not treated yet */
	if (priv->link_state.up == state.up)
		return 0;

	priv->link_state = state;
	if (state.up) {
		netif_carrier_on(priv->net_dev);
		netif_tx_start_all_queues(priv->net_dev);
	} else {
		netif_tx_stop_all_queues(priv->net_dev);
		netif_carrier_off(priv->net_dev);
	}

	netdev_info(priv->net_dev, "Link Event: state %s\n",
		    state.up ? "up" : "down");

	return 0;
}

static int dpaa2_eth_open(struct net_device *net_dev)
{
	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
	int err;

	err = seed_pool(priv, priv->bpid);
	if (err) {
		/* Not much to do; the buffer pool, though not filled up,
		 * may still contain some buffers which would enable us
		 * to limp on.
		 */
		netdev_err(net_dev, "Buffer seeding failed for DPBP %d (bpid=%d)\n",
			   priv->dpbp_dev->obj_desc.id, priv->bpid);
	}

	/* We'll only start the txqs when the link is actually ready; make sure
	 * we don't race against the link up notification, which may come
	 * immediately after dpni_enable();
	 */
	netif_tx_stop_all_queues(net_dev);
	enable_ch_napi(priv);
	/* Also, explicitly set carrier off, otherwise netif_carrier_ok() will
	 * return true and cause 'ip link show' to report the LOWER_UP flag,
	 * even though the link notification wasn't even received.
	 */
	netif_carrier_off(net_dev);

	err = dpni_enable(priv->mc_io, 0, priv->mc_token);
	if (err < 0) {
		netdev_err(net_dev, "dpni_enable() failed\n");
		goto enable_err;
	}

	/* If the DPMAC object has already processed the link up interrupt,
	 * we have to learn the link state ourselves.
	 */
	err = link_state_update(priv);
	if (err < 0) {
		netdev_err(net_dev, "Can't update link state\n");
		goto link_state_err;
	}

	return 0;

link_state_err:
enable_err:
	disable_ch_napi(priv);
	drain_pool(priv);
	return err;
}

/* The DPIO store must be empty when we call this,
 * at the end of every NAPI cycle.
 */
static u32 drain_channel(struct dpaa2_eth_channel *ch)
{
	u32 drained = 0, total = 0;

	do {
		pull_channel(ch);
		drained = consume_frames(ch, NULL);
		total += drained;
	} while (drained);

	return total;
}

static u32 drain_ingress_frames(struct dpaa2_eth_priv *priv)
{
	struct dpaa2_eth_channel *ch;
	int i;
	u32 drained = 0;

	for (i = 0; i < priv->num_channels; i++) {
		ch = priv->channel[i];
		drained += drain_channel(ch);
	}

	return drained;
}

static int dpaa2_eth_stop(struct net_device *net_dev)
{
	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
	int dpni_enabled = 0;
	int retries = 10;
	u32 drained;

	netif_tx_stop_all_queues(net_dev);
	netif_carrier_off(net_dev);

	/* Loop while dpni_disable() attempts to drain the egress FQs
	 * and confirm them back to us.
	 */
	do {
		dpni_disable(priv->mc_io, 0, priv->mc_token);
		dpni_is_enabled(priv->mc_io, 0, priv->mc_token, &dpni_enabled);
		if (dpni_enabled)
			/* Allow the hardware some slack */
			msleep(100);
	} while (dpni_enabled && --retries);
	if (!retries) {
		netdev_warn(net_dev, "Retry count exceeded disabling DPNI\n");
		/* Must go on and disable NAPI nonetheless, so we don't crash at
		 * the next "ifconfig up"
		 */
	}

	/* Wait for NAPI to complete on every core and disable it.
	 * In particular, this will also prevent NAPI from being rescheduled if
	 * a new CDAN is serviced, effectively discarding the CDAN. We therefore
	 * don't even need to disarm the channels, except perhaps for the case
	 * of a huge coalescing value.
	 */
	disable_ch_napi(priv);

	 /* Manually drain the Rx and TxConf queues */
	drained = drain_ingress_frames(priv);
	if (drained)
		netdev_dbg(net_dev, "Drained %d frames.\n", drained);

	/* Empty the buffer pool */
	drain_pool(priv);

	return 0;
}

static int dpaa2_eth_set_addr(struct net_device *net_dev, void *addr)
{
	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
	struct device *dev = net_dev->dev.parent;
	int err;

	err = eth_mac_addr(net_dev, addr);
	if (err < 0) {
		dev_err(dev, "eth_mac_addr() failed (%d)\n", err);
		return err;
	}

	err = dpni_set_primary_mac_addr(priv->mc_io, 0, priv->mc_token,
					net_dev->dev_addr);
	if (err) {
		dev_err(dev, "dpni_set_primary_mac_addr() failed (%d)\n", err);
		return err;
	}

	return 0;
}

/** Fill in counters maintained by the GPP driver. These may be different from
 * the hardware counters obtained by ethtool.
 */
static void dpaa2_eth_get_stats(struct net_device *net_dev,
				struct rtnl_link_stats64 *stats)
{
	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
	struct rtnl_link_stats64 *percpu_stats;
	u64 *cpustats;
	u64 *netstats = (u64 *)stats;
	int i, j;
	int num = sizeof(struct rtnl_link_stats64) / sizeof(u64);

	for_each_possible_cpu(i) {
		percpu_stats = per_cpu_ptr(priv->percpu_stats, i);
		cpustats = (u64 *)percpu_stats;
		for (j = 0; j < num; j++)
			netstats[j] += cpustats[j];
	}
}

/* Copy mac unicast addresses from @net_dev to @priv.
 * Its sole purpose is to make dpaa2_eth_set_rx_mode() more readable.
 */
static void add_uc_hw_addr(const struct net_device *net_dev,
			   struct dpaa2_eth_priv *priv)
{
	struct netdev_hw_addr *ha;
	int err;

	netdev_for_each_uc_addr(ha, net_dev) {
		err = dpni_add_mac_addr(priv->mc_io, 0, priv->mc_token,
					ha->addr);
		if (err)
			netdev_warn(priv->net_dev,
				    "Could not add ucast MAC %pM to the filtering table (err %d)\n",
				    ha->addr, err);
	}
}

/* Copy mac multicast addresses from @net_dev to @priv
 * Its sole purpose is to make dpaa2_eth_set_rx_mode() more readable.
 */
static void add_mc_hw_addr(const struct net_device *net_dev,
			   struct dpaa2_eth_priv *priv)
{
	struct netdev_hw_addr *ha;
	int err;

	netdev_for_each_mc_addr(ha, net_dev) {
		err = dpni_add_mac_addr(priv->mc_io, 0, priv->mc_token,
					ha->addr);
		if (err)
			netdev_warn(priv->net_dev,
				    "Could not add mcast MAC %pM to the filtering table (err %d)\n",
				    ha->addr, err);
	}
}

static void dpaa2_eth_set_rx_mode(struct net_device *net_dev)
{
	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
	int uc_count = netdev_uc_count(net_dev);
	int mc_count = netdev_mc_count(net_dev);
	u8 max_mac = priv->dpni_attrs.mac_filter_entries;
	u32 options = priv->dpni_attrs.options;
	u16 mc_token = priv->mc_token;
	struct fsl_mc_io *mc_io = priv->mc_io;
	int err;

	/* Basic sanity checks; these probably indicate a misconfiguration */
	if (options & DPNI_OPT_NO_MAC_FILTER && max_mac != 0)
		netdev_info(net_dev,
			    "mac_filter_entries=%d, DPNI_OPT_NO_MAC_FILTER option must be disabled\n",
			    max_mac);

	/* Force promiscuous if the uc or mc counts exceed our capabilities. */
	if (uc_count > max_mac) {
		netdev_info(net_dev,
			    "Unicast addr count reached %d, max allowed is %d; forcing promisc\n",
			    uc_count, max_mac);
		goto force_promisc;
	}
	if (mc_count + uc_count > max_mac) {
		netdev_info(net_dev,
			    "Unicast + multicast addr count reached %d, max allowed is %d; forcing promisc\n",
			    uc_count + mc_count, max_mac);
		goto force_mc_promisc;
	}

	/* Adjust promisc settings due to flag combinations */
	if (net_dev->flags & IFF_PROMISC)
		goto force_promisc;
	if (net_dev->flags & IFF_ALLMULTI) {
		/* First, rebuild unicast filtering table. This should be done
		 * in promisc mode, in order to avoid frame loss while we
		 * progressively add entries to the table.
		 * We don't know whether we had been in promisc already, and
		 * making an MC call to find out is expensive; so set uc promisc
		 * nonetheless.
		 */
		err = dpni_set_unicast_promisc(mc_io, 0, mc_token, 1);
		if (err)
			netdev_warn(net_dev, "Can't set uc promisc\n");

		/* Actual uc table reconstruction. */
		err = dpni_clear_mac_filters(mc_io, 0, mc_token, 1, 0);
		if (err)
			netdev_warn(net_dev, "Can't clear uc filters\n");
		add_uc_hw_addr(net_dev, priv);

		/* Finally, clear uc promisc and set mc promisc as requested. */
		err = dpni_set_unicast_promisc(mc_io, 0, mc_token, 0);
		if (err)
			netdev_warn(net_dev, "Can't clear uc promisc\n");
		goto force_mc_promisc;
	}

	/* Neither unicast, nor multicast promisc will be on... eventually.
	 * For now, rebuild mac filtering tables while forcing both of them on.
	 */
	err = dpni_set_unicast_promisc(mc_io, 0, mc_token, 1);
	if (err)
		netdev_warn(net_dev, "Can't set uc promisc (%d)\n", err);
	err = dpni_set_multicast_promisc(mc_io, 0, mc_token, 1);
	if (err)
		netdev_warn(net_dev, "Can't set mc promisc (%d)\n", err);

	/* Actual mac filtering tables reconstruction */
	err = dpni_clear_mac_filters(mc_io, 0, mc_token, 1, 1);
	if (err)
		netdev_warn(net_dev, "Can't clear mac filters\n");
	add_mc_hw_addr(net_dev, priv);
	add_uc_hw_addr(net_dev, priv);

	/* Now we can clear both ucast and mcast promisc, without risking
	 * to drop legitimate frames anymore.
	 */
	err = dpni_set_unicast_promisc(mc_io, 0, mc_token, 0);
	if (err)
		netdev_warn(net_dev, "Can't clear ucast promisc\n");
	err = dpni_set_multicast_promisc(mc_io, 0, mc_token, 0);
	if (err)
		netdev_warn(net_dev, "Can't clear mcast promisc\n");

	return;

force_promisc:
	err = dpni_set_unicast_promisc(mc_io, 0, mc_token, 1);
	if (err)
		netdev_warn(net_dev, "Can't set ucast promisc\n");
force_mc_promisc:
	err = dpni_set_multicast_promisc(mc_io, 0, mc_token, 1);
	if (err)
		netdev_warn(net_dev, "Can't set mcast promisc\n");
}

static int dpaa2_eth_set_features(struct net_device *net_dev,
				  netdev_features_t features)
{
	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
	netdev_features_t changed = features ^ net_dev->features;
	bool enable;
	int err;

	if (changed & NETIF_F_RXCSUM) {
		enable = !!(features & NETIF_F_RXCSUM);
		err = set_rx_csum(priv, enable);
		if (err)
			return err;
	}

	if (changed & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)) {
		enable = !!(features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM));
		err = set_tx_csum(priv, enable);
		if (err)
			return err;
	}

	return 0;
}

static int dpaa2_eth_ts_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
{
	struct dpaa2_eth_priv *priv = netdev_priv(dev);
	struct hwtstamp_config config;

	if (copy_from_user(&config, rq->ifr_data, sizeof(config)))
		return -EFAULT;

	switch (config.tx_type) {
	case HWTSTAMP_TX_OFF:
		priv->tx_tstamp = false;
		break;
	case HWTSTAMP_TX_ON:
		priv->tx_tstamp = true;
		break;
	default:
		return -ERANGE;
	}

	if (config.rx_filter == HWTSTAMP_FILTER_NONE) {
		priv->rx_tstamp = false;
	} else {
		priv->rx_tstamp = true;
		/* TS is set for all frame types, not only those requested */
		config.rx_filter = HWTSTAMP_FILTER_ALL;
	}

	return copy_to_user(rq->ifr_data, &config, sizeof(config)) ?
			-EFAULT : 0;
}

static int dpaa2_eth_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
{
	if (cmd == SIOCSHWTSTAMP)
		return dpaa2_eth_ts_ioctl(dev, rq, cmd);

	return -EINVAL;
}

static const struct net_device_ops dpaa2_eth_ops = {
	.ndo_open = dpaa2_eth_open,
	.ndo_start_xmit = dpaa2_eth_tx,
	.ndo_stop = dpaa2_eth_stop,
	.ndo_set_mac_address = dpaa2_eth_set_addr,
	.ndo_get_stats64 = dpaa2_eth_get_stats,
	.ndo_set_rx_mode = dpaa2_eth_set_rx_mode,
	.ndo_set_features = dpaa2_eth_set_features,
	.ndo_do_ioctl = dpaa2_eth_ioctl,
};

static void cdan_cb(struct dpaa2_io_notification_ctx *ctx)
{
	struct dpaa2_eth_channel *ch;

	ch = container_of(ctx, struct dpaa2_eth_channel, nctx);

	/* Update NAPI statistics */
	ch->stats.cdan++;

	napi_schedule_irqoff(&ch->napi);
}

/* Allocate and configure a DPCON object */
static struct fsl_mc_device *setup_dpcon(struct dpaa2_eth_priv *priv)
{
	struct fsl_mc_device *dpcon;
	struct device *dev = priv->net_dev->dev.parent;
	struct dpcon_attr attrs;
	int err;

	err = fsl_mc_object_allocate(to_fsl_mc_device(dev),
				     FSL_MC_POOL_DPCON, &dpcon);
	if (err) {
		dev_info(dev, "Not enough DPCONs, will go on as-is\n");
		return NULL;
	}

	err = dpcon_open(priv->mc_io, 0, dpcon->obj_desc.id, &dpcon->mc_handle);
	if (err) {
		dev_err(dev, "dpcon_open() failed\n");
		goto free;
	}

	err = dpcon_reset(priv->mc_io, 0, dpcon->mc_handle);
	if (err) {
		dev_err(dev, "dpcon_reset() failed\n");
		goto close;
	}

	err = dpcon_get_attributes(priv->mc_io, 0, dpcon->mc_handle, &attrs);
	if (err) {
		dev_err(dev, "dpcon_get_attributes() failed\n");
		goto close;
	}

	err = dpcon_enable(priv->mc_io, 0, dpcon->mc_handle);
	if (err) {
		dev_err(dev, "dpcon_enable() failed\n");
		goto close;
	}

	return dpcon;

close:
	dpcon_close(priv->mc_io, 0, dpcon->mc_handle);
free:
	fsl_mc_object_free(dpcon);

	return NULL;
}

static void free_dpcon(struct dpaa2_eth_priv *priv,
		       struct fsl_mc_device *dpcon)
{
	dpcon_disable(priv->mc_io, 0, dpcon->mc_handle);
	dpcon_close(priv->mc_io, 0, dpcon->mc_handle);
	fsl_mc_object_free(dpcon);
}

static struct dpaa2_eth_channel *
alloc_channel(struct dpaa2_eth_priv *priv)
{
	struct dpaa2_eth_channel *channel;
	struct dpcon_attr attr;
	struct device *dev = priv->net_dev->dev.parent;
	int err;

	channel = kzalloc(sizeof(*channel), GFP_KERNEL);
	if (!channel)
		return NULL;

	channel->dpcon = setup_dpcon(priv);
	if (!channel->dpcon)
		goto err_setup;

	err = dpcon_get_attributes(priv->mc_io, 0, channel->dpcon->mc_handle,
				   &attr);
	if (err) {
		dev_err(dev, "dpcon_get_attributes() failed\n");
		goto err_get_attr;
	}

	channel->dpcon_id = attr.id;
	channel->ch_id = attr.qbman_ch_id;
	channel->priv = priv;

	return channel;

err_get_attr:
	free_dpcon(priv, channel->dpcon);
err_setup:
	kfree(channel);
	return NULL;
}

static void free_channel(struct dpaa2_eth_priv *priv,
			 struct dpaa2_eth_channel *channel)
{
	free_dpcon(priv, channel->dpcon);
	kfree(channel);
}

/* DPIO setup: allocate and configure QBMan channels, setup core affinity
 * and register data availability notifications
 */
static int setup_dpio(struct dpaa2_eth_priv *priv)
{
	struct dpaa2_io_notification_ctx *nctx;
	struct dpaa2_eth_channel *channel;
	struct dpcon_notification_cfg dpcon_notif_cfg;
	struct device *dev = priv->net_dev->dev.parent;
	int i, err;

	/* We want the ability to spread ingress traffic (RX, TX conf) to as
	 * many cores as possible, so we need one channel for each core
	 * (unless there's fewer queues than cores, in which case the extra
	 * channels would be wasted).
	 * Allocate one channel per core and register it to the core's
	 * affine DPIO. If not enough channels are available for all cores
	 * or if some cores don't have an affine DPIO, there will be no
	 * ingress frame processing on those cores.
	 */
	cpumask_clear(&priv->dpio_cpumask);
	for_each_online_cpu(i) {
		/* Try to allocate a channel */
		channel = alloc_channel(priv);
		if (!channel) {
			dev_info(dev,
				 "No affine channel for cpu %d and above\n", i);
			err = -ENODEV;
			goto err_alloc_ch;
		}

		priv->channel[priv->num_channels] = channel;

		nctx = &channel->nctx;
		nctx->is_cdan = 1;
		nctx->cb = cdan_cb;
		nctx->id = channel->ch_id;
		nctx->desired_cpu = i;

		/* Register the new context */
		channel->dpio = dpaa2_io_service_select(i);
		err = dpaa2_io_service_register(channel->dpio, nctx);
		if (err) {
			dev_dbg(dev, "No affine DPIO for cpu %d\n", i);
			/* If no affine DPIO for this core, there's probably
			 * none available for next cores either. Signal we want
			 * to retry later, in case the DPIO devices weren't
			 * probed yet.
			 */
			err = -EPROBE_DEFER;
			goto err_service_reg;
		}

		/* Register DPCON notification with MC */
		dpcon_notif_cfg.dpio_id = nctx->dpio_id;
		dpcon_notif_cfg.priority = 0;
		dpcon_notif_cfg.user_ctx = nctx->qman64;
		err = dpcon_set_notification(priv->mc_io, 0,
					     channel->dpcon->mc_handle,
					     &dpcon_notif_cfg);
		if (err) {
			dev_err(dev, "dpcon_set_notification failed()\n");
			goto err_set_cdan;
		}

		/* If we managed to allocate a channel and also found an affine
		 * DPIO for this core, add it to the final mask
		 */
		cpumask_set_cpu(i, &priv->dpio_cpumask);
		priv->num_channels++;

		/* Stop if we already have enough channels to accommodate all
		 * RX and TX conf queues
		 */
		if (priv->num_channels == dpaa2_eth_queue_count(priv))
			break;
	}

	return 0;

err_set_cdan:
	dpaa2_io_service_deregister(channel->dpio, nctx);
err_service_reg:
	free_channel(priv, channel);
err_alloc_ch:
	if (cpumask_empty(&priv->dpio_cpumask)) {
		dev_err(dev, "No cpu with an affine DPIO/DPCON\n");
		return err;
	}

	dev_info(dev, "Cores %*pbl available for processing ingress traffic\n",
		 cpumask_pr_args(&priv->dpio_cpumask));

	return 0;
}

static void free_dpio(struct dpaa2_eth_priv *priv)
{
	int i;
	struct dpaa2_eth_channel *ch;

	/* deregister CDAN notifications and free channels */
	for (i = 0; i < priv->num_channels; i++) {
		ch = priv->channel[i];
		dpaa2_io_service_deregister(ch->dpio, &ch->nctx);
		free_channel(priv, ch);
	}
}

static struct dpaa2_eth_channel *get_affine_channel(struct dpaa2_eth_priv *priv,
						    int cpu)
{
	struct device *dev = priv->net_dev->dev.parent;
	int i;

	for (i = 0; i < priv->num_channels; i++)
		if (priv->channel[i]->nctx.desired_cpu == cpu)
			return priv->channel[i];

	/* We should never get here. Issue a warning and return
	 * the first channel, because it's still better than nothing
	 */
	dev_warn(dev, "No affine channel found for cpu %d\n", cpu);

	return priv->channel[0];
}

static void set_fq_affinity(struct dpaa2_eth_priv *priv)
{
	struct device *dev = priv->net_dev->dev.parent;
	struct cpumask xps_mask;
	struct dpaa2_eth_fq *fq;
	int rx_cpu, txc_cpu;
	int i, err;

	/* For each FQ, pick one channel/CPU to deliver frames to.
	 * This may well change at runtime, either through irqbalance or
	 * through direct user intervention.
	 */
	rx_cpu = txc_cpu = cpumask_first(&priv->dpio_cpumask);

	for (i = 0; i < priv->num_fqs; i++) {
		fq = &priv->fq[i];
		switch (fq->type) {
		case DPAA2_RX_FQ:
			fq->target_cpu = rx_cpu;
			rx_cpu = cpumask_next(rx_cpu, &priv->dpio_cpumask);
			if (rx_cpu >= nr_cpu_ids)
				rx_cpu = cpumask_first(&priv->dpio_cpumask);
			break;
		case DPAA2_TX_CONF_FQ:
			fq->target_cpu = txc_cpu;

			/* Tell the stack to affine to txc_cpu the Tx queue
			 * associated with the confirmation one
			 */
			cpumask_clear(&xps_mask);
			cpumask_set_cpu(txc_cpu, &xps_mask);
			err = netif_set_xps_queue(priv->net_dev, &xps_mask,
						  fq->flowid);
			if (err)
				dev_err(dev, "Error setting XPS queue\n");

			txc_cpu = cpumask_next(txc_cpu, &priv->dpio_cpumask);
			if (txc_cpu >= nr_cpu_ids)
				txc_cpu = cpumask_first(&priv->dpio_cpumask);
			break;
		default:
			dev_err(dev, "Unknown FQ type: %d\n", fq->type);
		}
		fq->channel = get_affine_channel(priv, fq->target_cpu);
	}
}

static void setup_fqs(struct dpaa2_eth_priv *priv)
{
	int i;

	/* We have one TxConf FQ per Tx flow.
	 * The number of Tx and Rx queues is the same.
	 * Tx queues come first in the fq array.
	 */
	for (i = 0; i < dpaa2_eth_queue_count(priv); i++) {
		priv->fq[priv->num_fqs].type = DPAA2_TX_CONF_FQ;
		priv->fq[priv->num_fqs].consume = dpaa2_eth_tx_conf;
		priv->fq[priv->num_fqs++].flowid = (u16)i;
	}

	for (i = 0; i < dpaa2_eth_queue_count(priv); i++) {
		priv->fq[priv->num_fqs].type = DPAA2_RX_FQ;
		priv->fq[priv->num_fqs].consume = dpaa2_eth_rx;
		priv->fq[priv->num_fqs++].flowid = (u16)i;
	}

	/* For each FQ, decide on which core to process incoming frames */
	set_fq_affinity(priv);
}

/* Allocate and configure one buffer pool for each interface */
static int setup_dpbp(struct dpaa2_eth_priv *priv)
{
	int err;
	struct fsl_mc_device *dpbp_dev;
	struct device *dev = priv->net_dev->dev.parent;
	struct dpbp_attr dpbp_attrs;

	err = fsl_mc_object_allocate(to_fsl_mc_device(dev), FSL_MC_POOL_DPBP,
				     &dpbp_dev);
	if (err) {
		dev_err(dev, "DPBP device allocation failed\n");
		return err;
	}

	priv->dpbp_dev = dpbp_dev;

	err = dpbp_open(priv->mc_io, 0, priv->dpbp_dev->obj_desc.id,
			&dpbp_dev->mc_handle);
	if (err) {
		dev_err(dev, "dpbp_open() failed\n");
		goto err_open;
	}

	err = dpbp_reset(priv->mc_io, 0, dpbp_dev->mc_handle);
	if (err) {
		dev_err(dev, "dpbp_reset() failed\n");
		goto err_reset;
	}

	err = dpbp_enable(priv->mc_io, 0, dpbp_dev->mc_handle);
	if (err) {
		dev_err(dev, "dpbp_enable() failed\n");
		goto err_enable;
	}

	err = dpbp_get_attributes(priv->mc_io, 0, dpbp_dev->mc_handle,
				  &dpbp_attrs);
	if (err) {
		dev_err(dev, "dpbp_get_attributes() failed\n");
		goto err_get_attr;
	}
	priv->bpid = dpbp_attrs.bpid;

	return 0;

err_get_attr:
	dpbp_disable(priv->mc_io, 0, dpbp_dev->mc_handle);
err_enable:
err_reset:
	dpbp_close(priv->mc_io, 0, dpbp_dev->mc_handle);
err_open:
	fsl_mc_object_free(dpbp_dev);

	return err;
}

static void free_dpbp(struct dpaa2_eth_priv *priv)
{
	drain_pool(priv);
	dpbp_disable(priv->mc_io, 0, priv->dpbp_dev->mc_handle);
	dpbp_close(priv->mc_io, 0, priv->dpbp_dev->mc_handle);
	fsl_mc_object_free(priv->dpbp_dev);
}

static int set_buffer_layout(struct dpaa2_eth_priv *priv)
{
	struct device *dev = priv->net_dev->dev.parent;
	struct dpni_buffer_layout buf_layout = {0};
	int err;

	/* We need to check for WRIOP version 1.0.0, but depending on the MC
	 * version, this number is not always provided correctly on rev1.
	 * We need to check for both alternatives in this situation.
	 */
	if (priv->dpni_attrs.wriop_version == DPAA2_WRIOP_VERSION(0, 0, 0) ||
	    priv->dpni_attrs.wriop_version == DPAA2_WRIOP_VERSION(1, 0, 0))
		priv->rx_buf_align = DPAA2_ETH_RX_BUF_ALIGN_REV1;
	else
		priv->rx_buf_align = DPAA2_ETH_RX_BUF_ALIGN;

	/* tx buffer */
	buf_layout.private_data_size = DPAA2_ETH_SWA_SIZE;
	buf_layout.pass_timestamp = true;
	buf_layout.options = DPNI_BUF_LAYOUT_OPT_PRIVATE_DATA_SIZE |
			     DPNI_BUF_LAYOUT_OPT_TIMESTAMP;
	err = dpni_set_buffer_layout(priv->mc_io, 0, priv->mc_token,
				     DPNI_QUEUE_TX, &buf_layout);
	if (err) {
		dev_err(dev, "dpni_set_buffer_layout(TX) failed\n");
		return err;
	}

	/* tx-confirm buffer */
	buf_layout.options = DPNI_BUF_LAYOUT_OPT_TIMESTAMP;
	err = dpni_set_buffer_layout(priv->mc_io, 0, priv->mc_token,
				     DPNI_QUEUE_TX_CONFIRM, &buf_layout);
	if (err) {
		dev_err(dev, "dpni_set_buffer_layout(TX_CONF) failed\n");
		return err;
	}

	/* Now that we've set our tx buffer layout, retrieve the minimum
	 * required tx data offset.
	 */
	err = dpni_get_tx_data_offset(priv->mc_io, 0, priv->mc_token,
				      &priv->tx_data_offset);
	if (err) {
		dev_err(dev, "dpni_get_tx_data_offset() failed\n");
		return err;
	}

	if ((priv->tx_data_offset % 64) != 0)
		dev_warn(dev, "Tx data offset (%d) not a multiple of 64B\n",
			 priv->tx_data_offset);

	/* rx buffer */
	buf_layout.pass_frame_status = true;
	buf_layout.pass_parser_result = true;
	buf_layout.data_align = priv->rx_buf_align;
	buf_layout.data_head_room = dpaa2_eth_rx_head_room(priv);
	buf_layout.private_data_size = 0;
	buf_layout.options = DPNI_BUF_LAYOUT_OPT_PARSER_RESULT |
			     DPNI_BUF_LAYOUT_OPT_FRAME_STATUS |
			     DPNI_BUF_LAYOUT_OPT_DATA_ALIGN |
			     DPNI_BUF_LAYOUT_OPT_DATA_HEAD_ROOM |
			     DPNI_BUF_LAYOUT_OPT_TIMESTAMP;
	err = dpni_set_buffer_layout(priv->mc_io, 0, priv->mc_token,
				     DPNI_QUEUE_RX, &buf_layout);
	if (err) {
		dev_err(dev, "dpni_set_buffer_layout(RX) failed\n");
		return err;
	}

	return 0;
}

/* Configure the DPNI object this interface is associated with */
static int setup_dpni(struct fsl_mc_device *ls_dev)
{
	struct device *dev = &ls_dev->dev;
	struct dpaa2_eth_priv *priv;
	struct net_device *net_dev;
	int err;

	net_dev = dev_get_drvdata(dev);
	priv = netdev_priv(net_dev);

	/* get a handle for the DPNI object */
	err = dpni_open(priv->mc_io, 0, ls_dev->obj_desc.id, &priv->mc_token);
	if (err) {
		dev_err(dev, "dpni_open() failed\n");
		return err;
	}

	/* Check if we can work with this DPNI object */
	err = dpni_get_api_version(priv->mc_io, 0, &priv->dpni_ver_major,
				   &priv->dpni_ver_minor);
	if (err) {
		dev_err(dev, "dpni_get_api_version() failed\n");
		goto close;
	}
	if (dpaa2_eth_cmp_dpni_ver(priv, DPNI_VER_MAJOR, DPNI_VER_MINOR) < 0) {
		dev_err(dev, "DPNI version %u.%u not supported, need >= %u.%u\n",
			priv->dpni_ver_major, priv->dpni_ver_minor,
			DPNI_VER_MAJOR, DPNI_VER_MINOR);
		err = -ENOTSUPP;
		goto close;
	}

	ls_dev->mc_io = priv->mc_io;
	ls_dev->mc_handle = priv->mc_token;

	err = dpni_reset(priv->mc_io, 0, priv->mc_token);
	if (err) {
		dev_err(dev, "dpni_reset() failed\n");
		goto close;
	}

	err = dpni_get_attributes(priv->mc_io, 0, priv->mc_token,
				  &priv->dpni_attrs);
	if (err) {
		dev_err(dev, "dpni_get_attributes() failed (err=%d)\n", err);
		goto close;
	}

	err = set_buffer_layout(priv);
	if (err)
		goto close;

	priv->cls_rules = devm_kzalloc(dev, sizeof(struct dpaa2_eth_cls_rule) *
				       dpaa2_eth_fs_count(priv), GFP_KERNEL);
	if (!priv->cls_rules)
		goto close;

	return 0;

close:
	dpni_close(priv->mc_io, 0, priv->mc_token);

	return err;
}

static void free_dpni(struct dpaa2_eth_priv *priv)
{
	int err;

	err = dpni_reset(priv->mc_io, 0, priv->mc_token);
	if (err)
		netdev_warn(priv->net_dev, "dpni_reset() failed (err %d)\n",
			    err);

	dpni_close(priv->mc_io, 0, priv->mc_token);
}

static int setup_rx_flow(struct dpaa2_eth_priv *priv,
			 struct dpaa2_eth_fq *fq)
{
	struct device *dev = priv->net_dev->dev.parent;
	struct dpni_queue queue;
	struct dpni_queue_id qid;
	struct dpni_taildrop td;
	int err;

	err = dpni_get_queue(priv->mc_io, 0, priv->mc_token,
			     DPNI_QUEUE_RX, 0, fq->flowid, &queue, &qid);
	if (err) {
		dev_err(dev, "dpni_get_queue(RX) failed\n");
		return err;
	}

	fq->fqid = qid.fqid;

	queue.destination.id = fq->channel->dpcon_id;
	queue.destination.type = DPNI_DEST_DPCON;
	queue.destination.priority = 1;
	queue.user_context = (u64)(uintptr_t)fq;
	err = dpni_set_queue(priv->mc_io, 0, priv->mc_token,
			     DPNI_QUEUE_RX, 0, fq->flowid,
			     DPNI_QUEUE_OPT_USER_CTX | DPNI_QUEUE_OPT_DEST,
			     &queue);
	if (err) {
		dev_err(dev, "dpni_set_queue(RX) failed\n");
		return err;
	}

	td.enable = 1;
	td.threshold = DPAA2_ETH_TAILDROP_THRESH;
	err = dpni_set_taildrop(priv->mc_io, 0, priv->mc_token, DPNI_CP_QUEUE,
				DPNI_QUEUE_RX, 0, fq->flowid, &td);
	if (err) {
		dev_err(dev, "dpni_set_threshold() failed\n");
		return err;
	}

	return 0;
}

static int setup_tx_flow(struct dpaa2_eth_priv *priv,
			 struct dpaa2_eth_fq *fq)
{
	struct device *dev = priv->net_dev->dev.parent;
	struct dpni_queue queue;
	struct dpni_queue_id qid;
	int err;

	err = dpni_get_queue(priv->mc_io, 0, priv->mc_token,
			     DPNI_QUEUE_TX, 0, fq->flowid, &queue, &qid);
	if (err) {
		dev_err(dev, "dpni_get_queue(TX) failed\n");
		return err;
	}

	fq->tx_qdbin = qid.qdbin;

	err = dpni_get_queue(priv->mc_io, 0, priv->mc_token,
			     DPNI_QUEUE_TX_CONFIRM, 0, fq->flowid,
			     &queue, &qid);
	if (err) {
		dev_err(dev, "dpni_get_queue(TX_CONF) failed\n");
		return err;
	}

	fq->fqid = qid.fqid;

	queue.destination.id = fq->channel->dpcon_id;
	queue.destination.type = DPNI_DEST_DPCON;
	queue.destination.priority = 0;
	queue.user_context = (u64)(uintptr_t)fq;
	err = dpni_set_queue(priv->mc_io, 0, priv->mc_token,
			     DPNI_QUEUE_TX_CONFIRM, 0, fq->flowid,
			     DPNI_QUEUE_OPT_USER_CTX | DPNI_QUEUE_OPT_DEST,
			     &queue);
	if (err) {
		dev_err(dev, "dpni_set_queue(TX_CONF) failed\n");
		return err;
	}

	return 0;
}

/* Supported header fields for Rx hash distribution key */
static const struct dpaa2_eth_dist_fields dist_fields[] = {
	{
		/* L2 header */
		.rxnfc_field = RXH_L2DA,
		.cls_prot = NET_PROT_ETH,
		.cls_field = NH_FLD_ETH_DA,
		.size = 6,
	}, {
		.cls_prot = NET_PROT_ETH,
		.cls_field = NH_FLD_ETH_SA,
		.size = 6,
	}, {
		/* This is the last ethertype field parsed:
		 * depending on frame format, it can be the MAC ethertype
		 * or the VLAN etype.
		 */
		.cls_prot = NET_PROT_ETH,
		.cls_field = NH_FLD_ETH_TYPE,
		.size = 2,
	}, {
		/* VLAN header */
		.rxnfc_field = RXH_VLAN,
		.cls_prot = NET_PROT_VLAN,
		.cls_field = NH_FLD_VLAN_TCI,
		.size = 2,
	}, {
		/* IP header */
		.rxnfc_field = RXH_IP_SRC,
		.cls_prot = NET_PROT_IP,
		.cls_field = NH_FLD_IP_SRC,
		.size = 4,
	}, {
		.rxnfc_field = RXH_IP_DST,
		.cls_prot = NET_PROT_IP,
		.cls_field = NH_FLD_IP_DST,
		.size = 4,
	}, {
		.rxnfc_field = RXH_L3_PROTO,
		.cls_prot = NET_PROT_IP,
		.cls_field = NH_FLD_IP_PROTO,
		.size = 1,
	}, {
		/* Using UDP ports, this is functionally equivalent to raw
		 * byte pairs from L4 header.
		 */
		.rxnfc_field = RXH_L4_B_0_1,
		.cls_prot = NET_PROT_UDP,
		.cls_field = NH_FLD_UDP_PORT_SRC,
		.size = 2,
	}, {
		.rxnfc_field = RXH_L4_B_2_3,
		.cls_prot = NET_PROT_UDP,
		.cls_field = NH_FLD_UDP_PORT_DST,
		.size = 2,
	},
};

/* Configure the Rx hash key using the legacy API */
static int config_legacy_hash_key(struct dpaa2_eth_priv *priv, dma_addr_t key)
{
	struct device *dev = priv->net_dev->dev.parent;
	struct dpni_rx_tc_dist_cfg dist_cfg;
	int err;

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

	dist_cfg.key_cfg_iova = key;
	dist_cfg.dist_size = dpaa2_eth_queue_count(priv);
	dist_cfg.dist_mode = DPNI_DIST_MODE_HASH;

	err = dpni_set_rx_tc_dist(priv->mc_io, 0, priv->mc_token, 0, &dist_cfg);
	if (err)
		dev_err(dev, "dpni_set_rx_tc_dist failed\n");

	return err;
}

/* Configure the Rx hash key using the new API */
static int config_hash_key(struct dpaa2_eth_priv *priv, dma_addr_t key)
{
	struct device *dev = priv->net_dev->dev.parent;
	struct dpni_rx_dist_cfg dist_cfg;
	int err;

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

	dist_cfg.key_cfg_iova = key;
	dist_cfg.dist_size = dpaa2_eth_queue_count(priv);
	dist_cfg.enable = 1;

	err = dpni_set_rx_hash_dist(priv->mc_io, 0, priv->mc_token, &dist_cfg);
	if (err)
		dev_err(dev, "dpni_set_rx_hash_dist failed\n");

	return err;
}

/* Configure the Rx flow classification key */
static int config_cls_key(struct dpaa2_eth_priv *priv, dma_addr_t key)
{
	struct device *dev = priv->net_dev->dev.parent;
	struct dpni_rx_dist_cfg dist_cfg;
	int err;

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

	dist_cfg.key_cfg_iova = key;
	dist_cfg.dist_size = dpaa2_eth_queue_count(priv);
	dist_cfg.enable = 1;

	err = dpni_set_rx_fs_dist(priv->mc_io, 0, priv->mc_token, &dist_cfg);
	if (err)
		dev_err(dev, "dpni_set_rx_fs_dist failed\n");

	return err;
}

/* Size of the Rx flow classification key */
int dpaa2_eth_cls_key_size(void)
{
	int i, size = 0;

	for (i = 0; i < ARRAY_SIZE(dist_fields); i++)
		size += dist_fields[i].size;

	return size;
}

/* Offset of header field in Rx classification key */
int dpaa2_eth_cls_fld_off(int prot, int field)
{
	int i, off = 0;

	for (i = 0; i < ARRAY_SIZE(dist_fields); i++) {
		if (dist_fields[i].cls_prot == prot &&
		    dist_fields[i].cls_field == field)
			return off;
		off += dist_fields[i].size;
	}

	WARN_ONCE(1, "Unsupported header field used for Rx flow cls\n");
	return 0;
}

/* Set Rx distribution (hash or flow classification) key
 * flags is a combination of RXH_ bits
 */
static int dpaa2_eth_set_dist_key(struct net_device *net_dev,
				  enum dpaa2_eth_rx_dist type, u64 flags)
{
	struct device *dev = net_dev->dev.parent;
	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
	struct dpkg_profile_cfg cls_cfg;
	u32 rx_hash_fields = 0;
	dma_addr_t key_iova;
	u8 *dma_mem;
	int i;
	int err = 0;

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

	for (i = 0; i < ARRAY_SIZE(dist_fields); i++) {
		struct dpkg_extract *key =
			&cls_cfg.extracts[cls_cfg.num_extracts];

		/* For Rx hashing key we set only the selected fields.
		 * For Rx flow classification key we set all supported fields
		 */
		if (type == DPAA2_ETH_RX_DIST_HASH) {
			if (!(flags & dist_fields[i].rxnfc_field))
				continue;
			rx_hash_fields |= dist_fields[i].rxnfc_field;
		}

		if (cls_cfg.num_extracts >= DPKG_MAX_NUM_OF_EXTRACTS) {
			dev_err(dev, "error adding key extraction rule, too many rules?\n");
			return -E2BIG;
		}

		key->type = DPKG_EXTRACT_FROM_HDR;
		key->extract.from_hdr.prot = dist_fields[i].cls_prot;
		key->extract.from_hdr.type = DPKG_FULL_FIELD;
		key->extract.from_hdr.field = dist_fields[i].cls_field;
		cls_cfg.num_extracts++;
	}

	dma_mem = kzalloc(DPAA2_CLASSIFIER_DMA_SIZE, GFP_KERNEL);
	if (!dma_mem)
		return -ENOMEM;

	err = dpni_prepare_key_cfg(&cls_cfg, dma_mem);
	if (err) {
		dev_err(dev, "dpni_prepare_key_cfg error %d\n", err);
		goto free_key;
	}

	/* Prepare for setting the rx dist */
	key_iova = dma_map_single(dev, dma_mem, DPAA2_CLASSIFIER_DMA_SIZE,
				  DMA_TO_DEVICE);
	if (dma_mapping_error(dev, key_iova)) {
		dev_err(dev, "DMA mapping failed\n");
		err = -ENOMEM;
		goto free_key;
	}

	if (type == DPAA2_ETH_RX_DIST_HASH) {
		if (dpaa2_eth_has_legacy_dist(priv))
			err = config_legacy_hash_key(priv, key_iova);
		else
			err = config_hash_key(priv, key_iova);
	} else {
		err = config_cls_key(priv, key_iova);
	}

	dma_unmap_single(dev, key_iova, DPAA2_CLASSIFIER_DMA_SIZE,
			 DMA_TO_DEVICE);
	if (!err && type == DPAA2_ETH_RX_DIST_HASH)
		priv->rx_hash_fields = rx_hash_fields;

free_key:
	kfree(dma_mem);
	return err;
}

int dpaa2_eth_set_hash(struct net_device *net_dev, u64 flags)
{
	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);

	if (!dpaa2_eth_hash_enabled(priv))
		return -EOPNOTSUPP;

	return dpaa2_eth_set_dist_key(net_dev, DPAA2_ETH_RX_DIST_HASH, flags);
}

static int dpaa2_eth_set_cls(struct dpaa2_eth_priv *priv)
{
	struct device *dev = priv->net_dev->dev.parent;

	/* Check if we actually support Rx flow classification */
	if (dpaa2_eth_has_legacy_dist(priv)) {
		dev_dbg(dev, "Rx cls not supported by current MC version\n");
		return -EOPNOTSUPP;
	}

	if (priv->dpni_attrs.options & DPNI_OPT_NO_FS ||
	    !(priv->dpni_attrs.options & DPNI_OPT_HAS_KEY_MASKING)) {
		dev_dbg(dev, "Rx cls disabled in DPNI options\n");
		return -EOPNOTSUPP;
	}

	if (!dpaa2_eth_hash_enabled(priv)) {
		dev_dbg(dev, "Rx cls disabled for single queue DPNIs\n");
		return -EOPNOTSUPP;
	}

	priv->rx_cls_enabled = 1;

	return dpaa2_eth_set_dist_key(priv->net_dev, DPAA2_ETH_RX_DIST_CLS, 0);
}

/* Bind the DPNI to its needed objects and resources: buffer pool, DPIOs,
 * frame queues and channels
 */
static int bind_dpni(struct dpaa2_eth_priv *priv)
{
	struct net_device *net_dev = priv->net_dev;
	struct device *dev = net_dev->dev.parent;
	struct dpni_pools_cfg pools_params;
	struct dpni_error_cfg err_cfg;
	int err = 0;
	int i;

	pools_params.num_dpbp = 1;
	pools_params.pools[0].dpbp_id = priv->dpbp_dev->obj_desc.id;
	pools_params.pools[0].backup_pool = 0;
	pools_params.pools[0].buffer_size = DPAA2_ETH_RX_BUF_SIZE;
	err = dpni_set_pools(priv->mc_io, 0, priv->mc_token, &pools_params);
	if (err) {
		dev_err(dev, "dpni_set_pools() failed\n");
		return err;
	}

	/* have the interface implicitly distribute traffic based on
	 * the default hash key
	 */
	err = dpaa2_eth_set_hash(net_dev, DPAA2_RXH_DEFAULT);
	if (err && err != -EOPNOTSUPP)
		dev_err(dev, "Failed to configure hashing\n");

	/* Configure the flow classification key; it includes all
	 * supported header fields and cannot be modified at runtime
	 */
	err = dpaa2_eth_set_cls(priv);
	if (err && err != -EOPNOTSUPP)
		dev_err(dev, "Failed to configure Rx classification key\n");

	/* Configure handling of error frames */
	err_cfg.errors = DPAA2_FAS_RX_ERR_MASK;
	err_cfg.set_frame_annotation = 1;
	err_cfg.error_action = DPNI_ERROR_ACTION_DISCARD;
	err = dpni_set_errors_behavior(priv->mc_io, 0, priv->mc_token,
				       &err_cfg);
	if (err) {
		dev_err(dev, "dpni_set_errors_behavior failed\n");
		return err;
	}

	/* Configure Rx and Tx conf queues to generate CDANs */
	for (i = 0; i < priv->num_fqs; i++) {
		switch (priv->fq[i].type) {
		case DPAA2_RX_FQ:
			err = setup_rx_flow(priv, &priv->fq[i]);
			break;
		case DPAA2_TX_CONF_FQ:
			err = setup_tx_flow(priv, &priv->fq[i]);
			break;
		default:
			dev_err(dev, "Invalid FQ type %d\n", priv->fq[i].type);
			return -EINVAL;
		}
		if (err)
			return err;
	}

	err = dpni_get_qdid(priv->mc_io, 0, priv->mc_token,
			    DPNI_QUEUE_TX, &priv->tx_qdid);
	if (err) {
		dev_err(dev, "dpni_get_qdid() failed\n");
		return err;
	}

	return 0;
}

/* Allocate rings for storing incoming frame descriptors */
static int alloc_rings(struct dpaa2_eth_priv *priv)
{
	struct net_device *net_dev = priv->net_dev;
	struct device *dev = net_dev->dev.parent;
	int i;

	for (i = 0; i < priv->num_channels; i++) {
		priv->channel[i]->store =
			dpaa2_io_store_create(DPAA2_ETH_STORE_SIZE, dev);
		if (!priv->channel[i]->store) {
			netdev_err(net_dev, "dpaa2_io_store_create() failed\n");
			goto err_ring;
		}
	}

	return 0;

err_ring:
	for (i = 0; i < priv->num_channels; i++) {
		if (!priv->channel[i]->store)
			break;
		dpaa2_io_store_destroy(priv->channel[i]->store);
	}

	return -ENOMEM;
}

static void free_rings(struct dpaa2_eth_priv *priv)
{
	int i;

	for (i = 0; i < priv->num_channels; i++)
		dpaa2_io_store_destroy(priv->channel[i]->store);
}

static int set_mac_addr(struct dpaa2_eth_priv *priv)
{
	struct net_device *net_dev = priv->net_dev;
	struct device *dev = net_dev->dev.parent;
	u8 mac_addr[ETH_ALEN], dpni_mac_addr[ETH_ALEN];
	int err;

	/* Get firmware address, if any */
	err = dpni_get_port_mac_addr(priv->mc_io, 0, priv->mc_token, mac_addr);
	if (err) {
		dev_err(dev, "dpni_get_port_mac_addr() failed\n");
		return err;
	}

	/* Get DPNI attributes address, if any */
	err = dpni_get_primary_mac_addr(priv->mc_io, 0, priv->mc_token,
					dpni_mac_addr);
	if (err) {
		dev_err(dev, "dpni_get_primary_mac_addr() failed\n");
		return err;
	}

	/* First check if firmware has any address configured by bootloader */
	if (!is_zero_ether_addr(mac_addr)) {
		/* If the DPMAC addr != DPNI addr, update it */
		if (!ether_addr_equal(mac_addr, dpni_mac_addr)) {
			err = dpni_set_primary_mac_addr(priv->mc_io, 0,
							priv->mc_token,
							mac_addr);
			if (err) {
				dev_err(dev, "dpni_set_primary_mac_addr() failed\n");
				return err;
			}
		}
		memcpy(net_dev->dev_addr, mac_addr, net_dev->addr_len);
	} else if (is_zero_ether_addr(dpni_mac_addr)) {
		/* No MAC address configured, fill in net_dev->dev_addr
		 * with a random one
		 */
		eth_hw_addr_random(net_dev);
		dev_dbg_once(dev, "device(s) have all-zero hwaddr, replaced with random\n");

		err = dpni_set_primary_mac_addr(priv->mc_io, 0, priv->mc_token,
						net_dev->dev_addr);
		if (err) {
			dev_err(dev, "dpni_set_primary_mac_addr() failed\n");
			return err;
		}

		/* Override NET_ADDR_RANDOM set by eth_hw_addr_random(); for all
		 * practical purposes, this will be our "permanent" mac address,
		 * at least until the next reboot. This move will also permit
		 * register_netdevice() to properly fill up net_dev->perm_addr.
		 */
		net_dev->addr_assign_type = NET_ADDR_PERM;
	} else {
		/* NET_ADDR_PERM is default, all we have to do is
		 * fill in the device addr.
		 */
		memcpy(net_dev->dev_addr, dpni_mac_addr, net_dev->addr_len);
	}

	return 0;
}

static int netdev_init(struct net_device *net_dev)
{
	struct device *dev = net_dev->dev.parent;
	struct dpaa2_eth_priv *priv = netdev_priv(net_dev);
	u32 options = priv->dpni_attrs.options;
	u64 supported = 0, not_supported = 0;
	u8 bcast_addr[ETH_ALEN];
	u8 num_queues;
	int err;

	net_dev->netdev_ops = &dpaa2_eth_ops;
	net_dev->ethtool_ops = &dpaa2_ethtool_ops;

	err = set_mac_addr(priv);
	if (err)
		return err;

	/* Explicitly add the broadcast address to the MAC filtering table */
	eth_broadcast_addr(bcast_addr);
	err = dpni_add_mac_addr(priv->mc_io, 0, priv->mc_token, bcast_addr);
	if (err) {
		dev_err(dev, "dpni_add_mac_addr() failed\n");
		return err;
	}

	/* Set MTU upper limit; lower limit is 68B (default value) */
	net_dev->max_mtu = DPAA2_ETH_MAX_MTU;
	err = dpni_set_max_frame_length(priv->mc_io, 0, priv->mc_token,
					DPAA2_ETH_MFL);
	if (err) {
		dev_err(dev, "dpni_set_max_frame_length() failed\n");
		return err;
	}

	/* Set actual number of queues in the net device */
	num_queues = dpaa2_eth_queue_count(priv);
	err = netif_set_real_num_tx_queues(net_dev, num_queues);
	if (err) {
		dev_err(dev, "netif_set_real_num_tx_queues() failed\n");
		return err;
	}
	err = netif_set_real_num_rx_queues(net_dev, num_queues);
	if (err) {
		dev_err(dev, "netif_set_real_num_rx_queues() failed\n");
		return err;
	}

	/* Capabilities listing */
	supported |= IFF_LIVE_ADDR_CHANGE;

	if (options & DPNI_OPT_NO_MAC_FILTER)
		not_supported |= IFF_UNICAST_FLT;
	else
		supported |= IFF_UNICAST_FLT;

	net_dev->priv_flags |= supported;
	net_dev->priv_flags &= ~not_supported;

	/* Features */
	net_dev->features = NETIF_F_RXCSUM |
			    NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
			    NETIF_F_SG | NETIF_F_HIGHDMA |
			    NETIF_F_LLTX;
	net_dev->hw_features = net_dev->features;

	return 0;
}

static int poll_link_state(void *arg)
{
	struct dpaa2_eth_priv *priv = (struct dpaa2_eth_priv *)arg;
	int err;

	while (!kthread_should_stop()) {
		err = link_state_update(priv);
		if (unlikely(err))
			return err;

		msleep(DPAA2_ETH_LINK_STATE_REFRESH);
	}

	return 0;
}

static irqreturn_t dpni_irq0_handler_thread(int irq_num, void *arg)
{
	u32 status = ~0;
	struct device *dev = (struct device *)arg;
	struct fsl_mc_device *dpni_dev = to_fsl_mc_device(dev);
	struct net_device *net_dev = dev_get_drvdata(dev);
	int err;

	err = dpni_get_irq_status(dpni_dev->mc_io, 0, dpni_dev->mc_handle,
				  DPNI_IRQ_INDEX, &status);
	if (unlikely(err)) {
		netdev_err(net_dev, "Can't get irq status (err %d)\n", err);
		return IRQ_HANDLED;
	}

	if (status & DPNI_IRQ_EVENT_LINK_CHANGED)
		link_state_update(netdev_priv(net_dev));

	return IRQ_HANDLED;
}

static int setup_irqs(struct fsl_mc_device *ls_dev)
{
	int err = 0;
	struct fsl_mc_device_irq *irq;

	err = fsl_mc_allocate_irqs(ls_dev);
	if (err) {
		dev_err(&ls_dev->dev, "MC irqs allocation failed\n");
		return err;
	}

	irq = ls_dev->irqs[0];
	err = devm_request_threaded_irq(&ls_dev->dev, irq->msi_desc->irq,
					NULL, dpni_irq0_handler_thread,
					IRQF_NO_SUSPEND | IRQF_ONESHOT,
					dev_name(&ls_dev->dev), &ls_dev->dev);
	if (err < 0) {
		dev_err(&ls_dev->dev, "devm_request_threaded_irq(): %d\n", err);
		goto free_mc_irq;
	}

	err = dpni_set_irq_mask(ls_dev->mc_io, 0, ls_dev->mc_handle,
				DPNI_IRQ_INDEX, DPNI_IRQ_EVENT_LINK_CHANGED);
	if (err < 0) {
		dev_err(&ls_dev->dev, "dpni_set_irq_mask(): %d\n", err);
		goto free_irq;
	}

	err = dpni_set_irq_enable(ls_dev->mc_io, 0, ls_dev->mc_handle,
				  DPNI_IRQ_INDEX, 1);
	if (err < 0) {
		dev_err(&ls_dev->dev, "dpni_set_irq_enable(): %d\n", err);
		goto free_irq;
	}

	return 0;

free_irq:
	devm_free_irq(&ls_dev->dev, irq->msi_desc->irq, &ls_dev->dev);
free_mc_irq:
	fsl_mc_free_irqs(ls_dev);

	return err;
}

static void add_ch_napi(struct dpaa2_eth_priv *priv)
{
	int i;
	struct dpaa2_eth_channel *ch;

	for (i = 0; i < priv->num_channels; i++) {
		ch = priv->channel[i];
		/* NAPI weight *MUST* be a multiple of DPAA2_ETH_STORE_SIZE */
		netif_napi_add(priv->net_dev, &ch->napi, dpaa2_eth_poll,
			       NAPI_POLL_WEIGHT);
	}
}

static void del_ch_napi(struct dpaa2_eth_priv *priv)
{
	int i;
	struct dpaa2_eth_channel *ch;

	for (i = 0; i < priv->num_channels; i++) {
		ch = priv->channel[i];
		netif_napi_del(&ch->napi);
	}
}

static int dpaa2_eth_probe(struct fsl_mc_device *dpni_dev)
{
	struct device *dev;
	struct net_device *net_dev = NULL;
	struct dpaa2_eth_priv *priv = NULL;
	int err = 0;

	dev = &dpni_dev->dev;

	/* Net device */
	net_dev = alloc_etherdev_mq(sizeof(*priv), DPAA2_ETH_MAX_TX_QUEUES);
	if (!net_dev) {
		dev_err(dev, "alloc_etherdev_mq() failed\n");
		return -ENOMEM;
	}

	SET_NETDEV_DEV(net_dev, dev);
	dev_set_drvdata(dev, net_dev);

	priv = netdev_priv(net_dev);
	priv->net_dev = net_dev;

	priv->iommu_domain = iommu_get_domain_for_dev(dev);

	/* Obtain a MC portal */
	err = fsl_mc_portal_allocate(dpni_dev, FSL_MC_IO_ATOMIC_CONTEXT_PORTAL,
				     &priv->mc_io);
	if (err) {
		if (err == -ENXIO)
			err = -EPROBE_DEFER;
		else
			dev_err(dev, "MC portal allocation failed\n");
		goto err_portal_alloc;
	}

	/* MC objects initialization and configuration */
	err = setup_dpni(dpni_dev);
	if (err)
		goto err_dpni_setup;

	err = setup_dpio(priv);
	if (err)
		goto err_dpio_setup;

	setup_fqs(priv);

	err = setup_dpbp(priv);
	if (err)
		goto err_dpbp_setup;

	err = bind_dpni(priv);
	if (err)
		goto err_bind;

	/* Add a NAPI context for each channel */
	add_ch_napi(priv);

	/* Percpu statistics */
	priv->percpu_stats = alloc_percpu(*priv->percpu_stats);
	if (!priv->percpu_stats) {
		dev_err(dev, "alloc_percpu(percpu_stats) failed\n");
		err = -ENOMEM;
		goto err_alloc_percpu_stats;
	}
	priv->percpu_extras = alloc_percpu(*priv->percpu_extras);
	if (!priv->percpu_extras) {
		dev_err(dev, "alloc_percpu(percpu_extras) failed\n");
		err = -ENOMEM;
		goto err_alloc_percpu_extras;
	}

	err = netdev_init(net_dev);
	if (err)
		goto err_netdev_init;

	/* Configure checksum offload based on current interface flags */
	err = set_rx_csum(priv, !!(net_dev->features & NETIF_F_RXCSUM));
	if (err)
		goto err_csum;

	err = set_tx_csum(priv, !!(net_dev->features &
				   (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)));
	if (err)
		goto err_csum;

	err = alloc_rings(priv);
	if (err)
		goto err_alloc_rings;

	err = setup_irqs(dpni_dev);
	if (err) {
		netdev_warn(net_dev, "Failed to set link interrupt, fall back to polling\n");
		priv->poll_thread = kthread_run(poll_link_state, priv,
						"%s_poll_link", net_dev->name);
		if (IS_ERR(priv->poll_thread)) {
			dev_err(dev, "Error starting polling thread\n");
			goto err_poll_thread;
		}
		priv->do_link_poll = true;
	}

	err = register_netdev(net_dev);
	if (err < 0) {
		dev_err(dev, "register_netdev() failed\n");
		goto err_netdev_reg;
	}

	dev_info(dev, "Probed interface %s\n", net_dev->name);
	return 0;

err_netdev_reg:
	if (priv->do_link_poll)
		kthread_stop(priv->poll_thread);
	else
		fsl_mc_free_irqs(dpni_dev);
err_poll_thread:
	free_rings(priv);
err_alloc_rings:
err_csum:
err_netdev_init:
	free_percpu(priv->percpu_extras);
err_alloc_percpu_extras:
	free_percpu(priv->percpu_stats);
err_alloc_percpu_stats:
	del_ch_napi(priv);
err_bind:
	free_dpbp(priv);
err_dpbp_setup:
	free_dpio(priv);
err_dpio_setup:
	free_dpni(priv);
err_dpni_setup:
	fsl_mc_portal_free(priv->mc_io);
err_portal_alloc:
	dev_set_drvdata(dev, NULL);
	free_netdev(net_dev);

	return err;
}

static int dpaa2_eth_remove(struct fsl_mc_device *ls_dev)
{
	struct device *dev;
	struct net_device *net_dev;
	struct dpaa2_eth_priv *priv;

	dev = &ls_dev->dev;
	net_dev = dev_get_drvdata(dev);
	priv = netdev_priv(net_dev);

	unregister_netdev(net_dev);

	if (priv->do_link_poll)
		kthread_stop(priv->poll_thread);
	else
		fsl_mc_free_irqs(ls_dev);

	free_rings(priv);
	free_percpu(priv->percpu_stats);
	free_percpu(priv->percpu_extras);

	del_ch_napi(priv);
	free_dpbp(priv);
	free_dpio(priv);
	free_dpni(priv);

	fsl_mc_portal_free(priv->mc_io);

	free_netdev(net_dev);

	dev_dbg(net_dev->dev.parent, "Removed interface %s\n", net_dev->name);

	return 0;
}

static const struct fsl_mc_device_id dpaa2_eth_match_id_table[] = {
	{
		.vendor = FSL_MC_VENDOR_FREESCALE,
		.obj_type = "dpni",
	},
	{ .vendor = 0x0 }
};
MODULE_DEVICE_TABLE(fslmc, dpaa2_eth_match_id_table);

static struct fsl_mc_driver dpaa2_eth_driver = {
	.driver = {
		.name = KBUILD_MODNAME,
		.owner = THIS_MODULE,
	},
	.probe = dpaa2_eth_probe,
	.remove = dpaa2_eth_remove,
	.match_id_table = dpaa2_eth_match_id_table
};

module_fsl_mc_driver(dpaa2_eth_driver);