summaryrefslogtreecommitdiffstats
path: root/sys/dev/re/if_re.c
blob: 2b0ef3a46503fb10f5621178466d2885bafe06bd (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
/*-
 * Copyright (c) 1997, 1998-2003
 *	Bill Paul <wpaul@windriver.com>.  All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. 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.
 * 3. All advertising materials mentioning features or use of this software
 *    must display the following acknowledgement:
 *	This product includes software developed by Bill Paul.
 * 4. Neither the name of the author nor the names of any co-contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY Bill Paul AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL Bill Paul OR THE VOICES IN HIS HEAD
 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
 * THE POSSIBILITY OF SUCH DAMAGE.
 */

#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");

/*
 * RealTek 8139C+/8169/8169S/8110S/8168/8111/8101E PCI NIC driver
 *
 * Written by Bill Paul <wpaul@windriver.com>
 * Senior Networking Software Engineer
 * Wind River Systems
 */

/*
 * This driver is designed to support RealTek's next generation of
 * 10/100 and 10/100/1000 PCI ethernet controllers. There are currently
 * seven devices in this family: the RTL8139C+, the RTL8169, the RTL8169S,
 * RTL8110S, the RTL8168, the RTL8111 and the RTL8101E.
 *
 * The 8139C+ is a 10/100 ethernet chip. It is backwards compatible
 * with the older 8139 family, however it also supports a special
 * C+ mode of operation that provides several new performance enhancing
 * features. These include:
 *
 *	o Descriptor based DMA mechanism. Each descriptor represents
 *	  a single packet fragment. Data buffers may be aligned on
 *	  any byte boundary.
 *
 *	o 64-bit DMA
 *
 *	o TCP/IP checksum offload for both RX and TX
 *
 *	o High and normal priority transmit DMA rings
 *
 *	o VLAN tag insertion and extraction
 *
 *	o TCP large send (segmentation offload)
 *
 * Like the 8139, the 8139C+ also has a built-in 10/100 PHY. The C+
 * programming API is fairly straightforward. The RX filtering, EEPROM
 * access and PHY access is the same as it is on the older 8139 series
 * chips.
 *
 * The 8169 is a 64-bit 10/100/1000 gigabit ethernet MAC. It has almost the
 * same programming API and feature set as the 8139C+ with the following
 * differences and additions:
 *
 *	o 1000Mbps mode
 *
 *	o Jumbo frames
 *
 *	o GMII and TBI ports/registers for interfacing with copper
 *	  or fiber PHYs
 *
 *	o RX and TX DMA rings can have up to 1024 descriptors
 *	  (the 8139C+ allows a maximum of 64)
 *
 *	o Slight differences in register layout from the 8139C+
 *
 * The TX start and timer interrupt registers are at different locations
 * on the 8169 than they are on the 8139C+. Also, the status word in the
 * RX descriptor has a slightly different bit layout. The 8169 does not
 * have a built-in PHY. Most reference boards use a Marvell 88E1000 'Alaska'
 * copper gigE PHY.
 *
 * The 8169S/8110S 10/100/1000 devices have built-in copper gigE PHYs
 * (the 'S' stands for 'single-chip'). These devices have the same
 * programming API as the older 8169, but also have some vendor-specific
 * registers for the on-board PHY. The 8110S is a LAN-on-motherboard
 * part designed to be pin-compatible with the RealTek 8100 10/100 chip.
 *
 * This driver takes advantage of the RX and TX checksum offload and
 * VLAN tag insertion/extraction features. It also implements TX
 * interrupt moderation using the timer interrupt registers, which
 * significantly reduces TX interrupt load. There is also support
 * for jumbo frames, however the 8169/8169S/8110S can not transmit
 * jumbo frames larger than 7440, so the max MTU possible with this
 * driver is 7422 bytes.
 */

#ifdef HAVE_KERNEL_OPTION_HEADERS
#include "opt_device_polling.h"
#endif

#include <sys/param.h>
#include <sys/endian.h>
#include <sys/systm.h>
#include <sys/sockio.h>
#include <sys/mbuf.h>
#include <sys/malloc.h>
#include <sys/module.h>
#include <sys/kernel.h>
#include <sys/socket.h>
#include <sys/lock.h>
#include <sys/mutex.h>
#include <sys/taskqueue.h>

#include <net/if.h>
#include <net/if_arp.h>
#include <net/ethernet.h>
#include <net/if_dl.h>
#include <net/if_media.h>
#include <net/if_types.h>
#include <net/if_vlan_var.h>

#include <net/bpf.h>

#include <machine/bus.h>
#include <machine/resource.h>
#include <sys/bus.h>
#include <sys/rman.h>

#include <dev/mii/mii.h>
#include <dev/mii/miivar.h>

#include <dev/pci/pcireg.h>
#include <dev/pci/pcivar.h>

#include <pci/if_rlreg.h>

MODULE_DEPEND(re, pci, 1, 1, 1);
MODULE_DEPEND(re, ether, 1, 1, 1);
MODULE_DEPEND(re, miibus, 1, 1, 1);

/* "device miibus" required.  See GENERIC if you get errors here. */
#include "miibus_if.h"

/*
 * Default to using PIO access for this driver.
 */
#define RE_USEIOSPACE

/* Tunables. */
static int msi_disable = 1;
TUNABLE_INT("hw.re.msi_disable", &msi_disable);

#define RE_CSUM_FEATURES    (CSUM_IP | CSUM_TCP | CSUM_UDP)

/*
 * Various supported device vendors/types and their names.
 */
static struct rl_type re_devs[] = {
	{ DLINK_VENDORID, DLINK_DEVICEID_528T, RL_HWREV_8169S,
		"D-Link DGE-528(T) Gigabit Ethernet Adapter" },
	{ DLINK_VENDORID, DLINK_DEVICEID_528T, RL_HWREV_8169_8110SB,
		"D-Link DGE-528(T) Rev.B1 Gigabit Ethernet Adapter" },
	{ RT_VENDORID, RT_DEVICEID_8139, RL_HWREV_8139CPLUS,
		"RealTek 8139C+ 10/100BaseTX" },
	{ RT_VENDORID, RT_DEVICEID_8101E, RL_HWREV_8101E,
		"RealTek 8101E PCIe 10/100baseTX" },
	{ RT_VENDORID, RT_DEVICEID_8168, RL_HWREV_8168_SPIN1,
		"RealTek 8168/8111B PCIe Gigabit Ethernet" },
	{ RT_VENDORID, RT_DEVICEID_8168, RL_HWREV_8168_SPIN2,
		"RealTek 8168/8111B PCIe Gigabit Ethernet" },
	{ RT_VENDORID, RT_DEVICEID_8168, RL_HWREV_8168_SPIN3,
		"RealTek 8168/8111B PCIe Gigabit Ethernet" },
	{ RT_VENDORID, RT_DEVICEID_8169, RL_HWREV_8169,
		"RealTek 8169 Gigabit Ethernet" },
	{ RT_VENDORID, RT_DEVICEID_8169, RL_HWREV_8169S,
		"RealTek 8169S Single-chip Gigabit Ethernet" },
	{ RT_VENDORID, RT_DEVICEID_8169, RL_HWREV_8169_8110SB,
		"RealTek 8169SB/8110SB Single-chip Gigabit Ethernet" },
	{ RT_VENDORID, RT_DEVICEID_8169, RL_HWREV_8169_8110SC,
		"RealTek 8169SC/8110SC Single-chip Gigabit Ethernet" },
	{ RT_VENDORID, RT_DEVICEID_8169SC, RL_HWREV_8169_8110SC,
		"RealTek 8169SC/8110SC Single-chip Gigabit Ethernet" },
	{ RT_VENDORID, RT_DEVICEID_8169, RL_HWREV_8110S,
		"RealTek 8110S Single-chip Gigabit Ethernet" },
	{ COREGA_VENDORID, COREGA_DEVICEID_CGLAPCIGT, RL_HWREV_8169S,
		"Corega CG-LAPCIGT (RTL8169S) Gigabit Ethernet" },
	{ LINKSYS_VENDORID, LINKSYS_DEVICEID_EG1032, RL_HWREV_8169S,
		"Linksys EG1032 (RTL8169S) Gigabit Ethernet" },
	{ USR_VENDORID, USR_DEVICEID_997902, RL_HWREV_8169S,
		"US Robotics 997902 (RTL8169S) Gigabit Ethernet" }
};

static struct rl_hwrev re_hwrevs[] = {
	{ RL_HWREV_8139, RL_8139,  "" },
	{ RL_HWREV_8139A, RL_8139, "A" },
	{ RL_HWREV_8139AG, RL_8139, "A-G" },
	{ RL_HWREV_8139B, RL_8139, "B" },
	{ RL_HWREV_8130, RL_8139, "8130" },
	{ RL_HWREV_8139C, RL_8139, "C" },
	{ RL_HWREV_8139D, RL_8139, "8139D/8100B/8100C" },
	{ RL_HWREV_8139CPLUS, RL_8139CPLUS, "C+"},
	{ RL_HWREV_8168_SPIN1, RL_8169, "8168"},
	{ RL_HWREV_8169, RL_8169, "8169"},
	{ RL_HWREV_8169S, RL_8169, "8169S"},
	{ RL_HWREV_8110S, RL_8169, "8110S"},
	{ RL_HWREV_8169_8110SB, RL_8169, "8169SB"},
	{ RL_HWREV_8169_8110SC, RL_8169, "8169SC"},
	{ RL_HWREV_8100, RL_8139, "8100"},
	{ RL_HWREV_8101, RL_8139, "8101"},
	{ RL_HWREV_8100E, RL_8169, "8100E"},
	{ RL_HWREV_8101E, RL_8169, "8101E"},
	{ RL_HWREV_8168_SPIN2, RL_8169, "8168"},
	{ RL_HWREV_8168_SPIN3, RL_8169, "8168"},
	{ 0, 0, NULL }
};

static int re_probe		(device_t);
static int re_attach		(device_t);
static int re_detach		(device_t);

static int re_encap		(struct rl_softc *, struct mbuf **);

static void re_dma_map_addr	(void *, bus_dma_segment_t *, int, int);
static int re_allocmem		(device_t, struct rl_softc *);
static __inline void re_discard_rxbuf
				(struct rl_softc *, int);
static int re_newbuf		(struct rl_softc *, int);
static int re_rx_list_init	(struct rl_softc *);
static int re_tx_list_init	(struct rl_softc *);
#ifdef RE_FIXUP_RX
static __inline void re_fixup_rx
				(struct mbuf *);
#endif
static int re_rxeof		(struct rl_softc *);
static void re_txeof		(struct rl_softc *);
#ifdef DEVICE_POLLING
static void re_poll		(struct ifnet *, enum poll_cmd, int);
static void re_poll_locked	(struct ifnet *, enum poll_cmd, int);
#endif
static int re_intr		(void *);
static void re_tick		(void *);
static void re_tx_task		(void *, int);
static void re_int_task		(void *, int);
static void re_start		(struct ifnet *);
static int re_ioctl		(struct ifnet *, u_long, caddr_t);
static void re_init		(void *);
static void re_init_locked	(struct rl_softc *);
static void re_stop		(struct rl_softc *);
static void re_watchdog		(struct rl_softc *);
static int re_suspend		(device_t);
static int re_resume		(device_t);
static int re_shutdown		(device_t);
static int re_ifmedia_upd	(struct ifnet *);
static void re_ifmedia_sts	(struct ifnet *, struct ifmediareq *);

static void re_eeprom_putbyte	(struct rl_softc *, int);
static void re_eeprom_getword	(struct rl_softc *, int, u_int16_t *);
static void re_read_eeprom	(struct rl_softc *, caddr_t, int, int);
static int re_gmii_readreg	(device_t, int, int);
static int re_gmii_writereg	(device_t, int, int, int);

static int re_miibus_readreg	(device_t, int, int);
static int re_miibus_writereg	(device_t, int, int, int);
static void re_miibus_statchg	(device_t);

static void re_setmulti		(struct rl_softc *);
static void re_reset		(struct rl_softc *);
static void re_setwol		(struct rl_softc *);
static void re_clrwol		(struct rl_softc *);

#ifdef RE_DIAG
static int re_diag		(struct rl_softc *);
#endif

#ifdef RE_USEIOSPACE
#define RL_RES			SYS_RES_IOPORT
#define RL_RID			RL_PCI_LOIO
#else
#define RL_RES			SYS_RES_MEMORY
#define RL_RID			RL_PCI_LOMEM
#endif

static device_method_t re_methods[] = {
	/* Device interface */
	DEVMETHOD(device_probe,		re_probe),
	DEVMETHOD(device_attach,	re_attach),
	DEVMETHOD(device_detach,	re_detach),
	DEVMETHOD(device_suspend,	re_suspend),
	DEVMETHOD(device_resume,	re_resume),
	DEVMETHOD(device_shutdown,	re_shutdown),

	/* bus interface */
	DEVMETHOD(bus_print_child,	bus_generic_print_child),
	DEVMETHOD(bus_driver_added,	bus_generic_driver_added),

	/* MII interface */
	DEVMETHOD(miibus_readreg,	re_miibus_readreg),
	DEVMETHOD(miibus_writereg,	re_miibus_writereg),
	DEVMETHOD(miibus_statchg,	re_miibus_statchg),

	{ 0, 0 }
};

static driver_t re_driver = {
	"re",
	re_methods,
	sizeof(struct rl_softc)
};

static devclass_t re_devclass;

DRIVER_MODULE(re, pci, re_driver, re_devclass, 0, 0);
DRIVER_MODULE(re, cardbus, re_driver, re_devclass, 0, 0);
DRIVER_MODULE(miibus, re, miibus_driver, miibus_devclass, 0, 0);

#define EE_SET(x)					\
	CSR_WRITE_1(sc, RL_EECMD,			\
		CSR_READ_1(sc, RL_EECMD) | x)

#define EE_CLR(x)					\
	CSR_WRITE_1(sc, RL_EECMD,			\
		CSR_READ_1(sc, RL_EECMD) & ~x)

/*
 * Send a read command and address to the EEPROM, check for ACK.
 */
static void
re_eeprom_putbyte(sc, addr)
	struct rl_softc		*sc;
	int			addr;
{
	register int		d, i;

	d = addr | (RL_9346_READ << sc->rl_eewidth);

	/*
	 * Feed in each bit and strobe the clock.
	 */

	for (i = 1 << (sc->rl_eewidth + 3); i; i >>= 1) {
		if (d & i) {
			EE_SET(RL_EE_DATAIN);
		} else {
			EE_CLR(RL_EE_DATAIN);
		}
		DELAY(100);
		EE_SET(RL_EE_CLK);
		DELAY(150);
		EE_CLR(RL_EE_CLK);
		DELAY(100);
	}

	return;
}

/*
 * Read a word of data stored in the EEPROM at address 'addr.'
 */
static void
re_eeprom_getword(sc, addr, dest)
	struct rl_softc		*sc;
	int			addr;
	u_int16_t		*dest;
{
	register int		i;
	u_int16_t		word = 0;

	/*
	 * Send address of word we want to read.
	 */
	re_eeprom_putbyte(sc, addr);

	/*
	 * Start reading bits from EEPROM.
	 */
	for (i = 0x8000; i; i >>= 1) {
		EE_SET(RL_EE_CLK);
		DELAY(100);
		if (CSR_READ_1(sc, RL_EECMD) & RL_EE_DATAOUT)
			word |= i;
		EE_CLR(RL_EE_CLK);
		DELAY(100);
	}

	*dest = word;

	return;
}

/*
 * Read a sequence of words from the EEPROM.
 */
static void
re_read_eeprom(sc, dest, off, cnt)
	struct rl_softc		*sc;
	caddr_t			dest;
	int			off;
	int			cnt;
{
	int			i;
	u_int16_t		word = 0, *ptr;

	CSR_SETBIT_1(sc, RL_EECMD, RL_EEMODE_PROGRAM);

        DELAY(100);

	for (i = 0; i < cnt; i++) {
		CSR_SETBIT_1(sc, RL_EECMD, RL_EE_SEL);
		re_eeprom_getword(sc, off + i, &word);
		CSR_CLRBIT_1(sc, RL_EECMD, RL_EE_SEL);
		ptr = (u_int16_t *)(dest + (i * 2));
                *ptr = word;
	}

	CSR_CLRBIT_1(sc, RL_EECMD, RL_EEMODE_PROGRAM);

	return;
}

static int
re_gmii_readreg(dev, phy, reg)
	device_t		dev;
	int			phy, reg;
{
	struct rl_softc		*sc;
	u_int32_t		rval;
	int			i;

	if (phy != 1)
		return (0);

	sc = device_get_softc(dev);

	/* Let the rgephy driver read the GMEDIASTAT register */

	if (reg == RL_GMEDIASTAT) {
		rval = CSR_READ_1(sc, RL_GMEDIASTAT);
		return (rval);
	}

	CSR_WRITE_4(sc, RL_PHYAR, reg << 16);
	DELAY(1000);

	for (i = 0; i < RL_TIMEOUT; i++) {
		rval = CSR_READ_4(sc, RL_PHYAR);
		if (rval & RL_PHYAR_BUSY)
			break;
		DELAY(100);
	}

	if (i == RL_TIMEOUT) {
		device_printf(sc->rl_dev, "PHY read failed\n");
		return (0);
	}

	return (rval & RL_PHYAR_PHYDATA);
}

static int
re_gmii_writereg(dev, phy, reg, data)
	device_t		dev;
	int			phy, reg, data;
{
	struct rl_softc		*sc;
	u_int32_t		rval;
	int			i;

	sc = device_get_softc(dev);

	CSR_WRITE_4(sc, RL_PHYAR, (reg << 16) |
	    (data & RL_PHYAR_PHYDATA) | RL_PHYAR_BUSY);
	DELAY(1000);

	for (i = 0; i < RL_TIMEOUT; i++) {
		rval = CSR_READ_4(sc, RL_PHYAR);
		if (!(rval & RL_PHYAR_BUSY))
			break;
		DELAY(100);
	}

	if (i == RL_TIMEOUT) {
		device_printf(sc->rl_dev, "PHY write failed\n");
		return (0);
	}

	return (0);
}

static int
re_miibus_readreg(dev, phy, reg)
	device_t		dev;
	int			phy, reg;
{
	struct rl_softc		*sc;
	u_int16_t		rval = 0;
	u_int16_t		re8139_reg = 0;

	sc = device_get_softc(dev);

	if (sc->rl_type == RL_8169) {
		rval = re_gmii_readreg(dev, phy, reg);
		return (rval);
	}

	/* Pretend the internal PHY is only at address 0 */
	if (phy) {
		return (0);
	}
	switch (reg) {
	case MII_BMCR:
		re8139_reg = RL_BMCR;
		break;
	case MII_BMSR:
		re8139_reg = RL_BMSR;
		break;
	case MII_ANAR:
		re8139_reg = RL_ANAR;
		break;
	case MII_ANER:
		re8139_reg = RL_ANER;
		break;
	case MII_ANLPAR:
		re8139_reg = RL_LPAR;
		break;
	case MII_PHYIDR1:
	case MII_PHYIDR2:
		return (0);
	/*
	 * Allow the rlphy driver to read the media status
	 * register. If we have a link partner which does not
	 * support NWAY, this is the register which will tell
	 * us the results of parallel detection.
	 */
	case RL_MEDIASTAT:
		rval = CSR_READ_1(sc, RL_MEDIASTAT);
		return (rval);
	default:
		device_printf(sc->rl_dev, "bad phy register\n");
		return (0);
	}
	rval = CSR_READ_2(sc, re8139_reg);
	if (sc->rl_type == RL_8139CPLUS && re8139_reg == RL_BMCR) {
		/* 8139C+ has different bit layout. */
		rval &= ~(BMCR_LOOP | BMCR_ISO);
	}
	return (rval);
}

static int
re_miibus_writereg(dev, phy, reg, data)
	device_t		dev;
	int			phy, reg, data;
{
	struct rl_softc		*sc;
	u_int16_t		re8139_reg = 0;
	int			rval = 0;

	sc = device_get_softc(dev);

	if (sc->rl_type == RL_8169) {
		rval = re_gmii_writereg(dev, phy, reg, data);
		return (rval);
	}

	/* Pretend the internal PHY is only at address 0 */
	if (phy)
		return (0);

	switch (reg) {
	case MII_BMCR:
		re8139_reg = RL_BMCR;
		if (sc->rl_type == RL_8139CPLUS) {
			/* 8139C+ has different bit layout. */
			data &= ~(BMCR_LOOP | BMCR_ISO);
		}
		break;
	case MII_BMSR:
		re8139_reg = RL_BMSR;
		break;
	case MII_ANAR:
		re8139_reg = RL_ANAR;
		break;
	case MII_ANER:
		re8139_reg = RL_ANER;
		break;
	case MII_ANLPAR:
		re8139_reg = RL_LPAR;
		break;
	case MII_PHYIDR1:
	case MII_PHYIDR2:
		return (0);
		break;
	default:
		device_printf(sc->rl_dev, "bad phy register\n");
		return (0);
	}
	CSR_WRITE_2(sc, re8139_reg, data);
	return (0);
}

static void
re_miibus_statchg(dev)
	device_t		dev;
{

}

/*
 * Program the 64-bit multicast hash filter.
 */
static void
re_setmulti(sc)
	struct rl_softc		*sc;
{
	struct ifnet		*ifp;
	int			h = 0;
	u_int32_t		hashes[2] = { 0, 0 };
	struct ifmultiaddr	*ifma;
	u_int32_t		rxfilt;
	int			mcnt = 0;
	u_int32_t		hwrev;

	RL_LOCK_ASSERT(sc);

	ifp = sc->rl_ifp;


	rxfilt = CSR_READ_4(sc, RL_RXCFG);
	rxfilt &= ~(RL_RXCFG_RX_ALLPHYS | RL_RXCFG_RX_MULTI);
	if (ifp->if_flags & IFF_ALLMULTI || ifp->if_flags & IFF_PROMISC) {
		if (ifp->if_flags & IFF_PROMISC)
			rxfilt |= RL_RXCFG_RX_ALLPHYS;
		/*
		 * Unlike other hardwares, we have to explicitly set
		 * RL_RXCFG_RX_MULTI to receive multicast frames in
		 * promiscuous mode.
		 */
		rxfilt |= RL_RXCFG_RX_MULTI;
		CSR_WRITE_4(sc, RL_RXCFG, rxfilt);
		CSR_WRITE_4(sc, RL_MAR0, 0xFFFFFFFF);
		CSR_WRITE_4(sc, RL_MAR4, 0xFFFFFFFF);
		return;
	}

	/* first, zot all the existing hash bits */
	CSR_WRITE_4(sc, RL_MAR0, 0);
	CSR_WRITE_4(sc, RL_MAR4, 0);

	/* now program new ones */
	IF_ADDR_LOCK(ifp);
	TAILQ_FOREACH(ifma, &ifp->if_multiaddrs, ifma_link) {
		if (ifma->ifma_addr->sa_family != AF_LINK)
			continue;
		h = ether_crc32_be(LLADDR((struct sockaddr_dl *)
		    ifma->ifma_addr), ETHER_ADDR_LEN) >> 26;
		if (h < 32)
			hashes[0] |= (1 << h);
		else
			hashes[1] |= (1 << (h - 32));
		mcnt++;
	}
	IF_ADDR_UNLOCK(ifp);

	if (mcnt)
		rxfilt |= RL_RXCFG_RX_MULTI;
	else
		rxfilt &= ~RL_RXCFG_RX_MULTI;

	CSR_WRITE_4(sc, RL_RXCFG, rxfilt);

	/*
	 * For some unfathomable reason, RealTek decided to reverse
	 * the order of the multicast hash registers in the PCI Express
	 * parts. This means we have to write the hash pattern in reverse
	 * order for those devices.
	 */

	hwrev = CSR_READ_4(sc, RL_TXCFG) & RL_TXCFG_HWREV;

	switch (hwrev) {
	case RL_HWREV_8100E:
	case RL_HWREV_8101E:
	case RL_HWREV_8168_SPIN1:
	case RL_HWREV_8168_SPIN2:
	case RL_HWREV_8168_SPIN3:
		CSR_WRITE_4(sc, RL_MAR0, bswap32(hashes[1]));
		CSR_WRITE_4(sc, RL_MAR4, bswap32(hashes[0]));
		break;
	default:
		CSR_WRITE_4(sc, RL_MAR0, hashes[0]);
		CSR_WRITE_4(sc, RL_MAR4, hashes[1]);
		break;
	}
}

static void
re_reset(sc)
	struct rl_softc		*sc;
{
	register int		i;

	RL_LOCK_ASSERT(sc);

	CSR_WRITE_1(sc, RL_COMMAND, RL_CMD_RESET);

	for (i = 0; i < RL_TIMEOUT; i++) {
		DELAY(10);
		if (!(CSR_READ_1(sc, RL_COMMAND) & RL_CMD_RESET))
			break;
	}
	if (i == RL_TIMEOUT)
		device_printf(sc->rl_dev, "reset never completed!\n");

	CSR_WRITE_1(sc, 0x82, 1);
}

#ifdef RE_DIAG

/*
 * The following routine is designed to test for a defect on some
 * 32-bit 8169 cards. Some of these NICs have the REQ64# and ACK64#
 * lines connected to the bus, however for a 32-bit only card, they
 * should be pulled high. The result of this defect is that the
 * NIC will not work right if you plug it into a 64-bit slot: DMA
 * operations will be done with 64-bit transfers, which will fail
 * because the 64-bit data lines aren't connected.
 *
 * There's no way to work around this (short of talking a soldering
 * iron to the board), however we can detect it. The method we use
 * here is to put the NIC into digital loopback mode, set the receiver
 * to promiscuous mode, and then try to send a frame. We then compare
 * the frame data we sent to what was received. If the data matches,
 * then the NIC is working correctly, otherwise we know the user has
 * a defective NIC which has been mistakenly plugged into a 64-bit PCI
 * slot. In the latter case, there's no way the NIC can work correctly,
 * so we print out a message on the console and abort the device attach.
 */

static int
re_diag(sc)
	struct rl_softc		*sc;
{
	struct ifnet		*ifp = sc->rl_ifp;
	struct mbuf		*m0;
	struct ether_header	*eh;
	struct rl_desc		*cur_rx;
	u_int16_t		status;
	u_int32_t		rxstat;
	int			total_len, i, error = 0, phyaddr;
	u_int8_t		dst[] = { 0x00, 'h', 'e', 'l', 'l', 'o' };
	u_int8_t		src[] = { 0x00, 'w', 'o', 'r', 'l', 'd' };

	/* Allocate a single mbuf */
	MGETHDR(m0, M_DONTWAIT, MT_DATA);
	if (m0 == NULL)
		return (ENOBUFS);

	RL_LOCK(sc);

	/*
	 * Initialize the NIC in test mode. This sets the chip up
	 * so that it can send and receive frames, but performs the
	 * following special functions:
	 * - Puts receiver in promiscuous mode
	 * - Enables digital loopback mode
	 * - Leaves interrupts turned off
	 */

	ifp->if_flags |= IFF_PROMISC;
	sc->rl_testmode = 1;
	re_reset(sc);
	re_init_locked(sc);
	sc->rl_link = 1;
	if (sc->rl_type == RL_8169)
		phyaddr = 1;
	else
		phyaddr = 0;

	re_miibus_writereg(sc->rl_dev, phyaddr, MII_BMCR, BMCR_RESET);
	for (i = 0; i < RL_TIMEOUT; i++) {
		status = re_miibus_readreg(sc->rl_dev, phyaddr, MII_BMCR);
		if (!(status & BMCR_RESET))
			break;
	}

	re_miibus_writereg(sc->rl_dev, phyaddr, MII_BMCR, BMCR_LOOP);
	CSR_WRITE_2(sc, RL_ISR, RL_INTRS);

	DELAY(100000);

	/* Put some data in the mbuf */

	eh = mtod(m0, struct ether_header *);
	bcopy ((char *)&dst, eh->ether_dhost, ETHER_ADDR_LEN);
	bcopy ((char *)&src, eh->ether_shost, ETHER_ADDR_LEN);
	eh->ether_type = htons(ETHERTYPE_IP);
	m0->m_pkthdr.len = m0->m_len = ETHER_MIN_LEN - ETHER_CRC_LEN;

	/*
	 * Queue the packet, start transmission.
	 * Note: IF_HANDOFF() ultimately calls re_start() for us.
	 */

	CSR_WRITE_2(sc, RL_ISR, 0xFFFF);
	RL_UNLOCK(sc);
	/* XXX: re_diag must not be called when in ALTQ mode */
	IF_HANDOFF(&ifp->if_snd, m0, ifp);
	RL_LOCK(sc);
	m0 = NULL;

	/* Wait for it to propagate through the chip */

	DELAY(100000);
	for (i = 0; i < RL_TIMEOUT; i++) {
		status = CSR_READ_2(sc, RL_ISR);
		CSR_WRITE_2(sc, RL_ISR, status);
		if ((status & (RL_ISR_TIMEOUT_EXPIRED|RL_ISR_RX_OK)) ==
		    (RL_ISR_TIMEOUT_EXPIRED|RL_ISR_RX_OK))
			break;
		DELAY(10);
	}

	if (i == RL_TIMEOUT) {
		device_printf(sc->rl_dev,
		    "diagnostic failed, failed to receive packet in"
		    " loopback mode\n");
		error = EIO;
		goto done;
	}

	/*
	 * The packet should have been dumped into the first
	 * entry in the RX DMA ring. Grab it from there.
	 */

	bus_dmamap_sync(sc->rl_ldata.rl_rx_list_tag,
	    sc->rl_ldata.rl_rx_list_map,
	    BUS_DMASYNC_POSTREAD);
	bus_dmamap_sync(sc->rl_ldata.rl_rx_mtag,
	    sc->rl_ldata.rl_rx_desc[0].rx_dmamap,
	    BUS_DMASYNC_POSTREAD);
	bus_dmamap_unload(sc->rl_ldata.rl_rx_mtag,
	    sc->rl_ldata.rl_rx_desc[0].rx_dmamap);

	m0 = sc->rl_ldata.rl_rx_desc[0].rx_m;
	sc->rl_ldata.rl_rx_desc[0].rx_m = NULL;
	eh = mtod(m0, struct ether_header *);

	cur_rx = &sc->rl_ldata.rl_rx_list[0];
	total_len = RL_RXBYTES(cur_rx);
	rxstat = le32toh(cur_rx->rl_cmdstat);

	if (total_len != ETHER_MIN_LEN) {
		device_printf(sc->rl_dev,
		    "diagnostic failed, received short packet\n");
		error = EIO;
		goto done;
	}

	/* Test that the received packet data matches what we sent. */

	if (bcmp((char *)&eh->ether_dhost, (char *)&dst, ETHER_ADDR_LEN) ||
	    bcmp((char *)&eh->ether_shost, (char *)&src, ETHER_ADDR_LEN) ||
	    ntohs(eh->ether_type) != ETHERTYPE_IP) {
		device_printf(sc->rl_dev, "WARNING, DMA FAILURE!\n");
		device_printf(sc->rl_dev, "expected TX data: %6D/%6D/0x%x\n",
		    dst, ":", src, ":", ETHERTYPE_IP);
		device_printf(sc->rl_dev, "received RX data: %6D/%6D/0x%x\n",
		    eh->ether_dhost, ":",  eh->ether_shost, ":",
		    ntohs(eh->ether_type));
		device_printf(sc->rl_dev, "You may have a defective 32-bit "
		    "NIC plugged into a 64-bit PCI slot.\n");
		device_printf(sc->rl_dev, "Please re-install the NIC in a "
		    "32-bit slot for proper operation.\n");
		device_printf(sc->rl_dev, "Read the re(4) man page for more "
		    "details.\n");
		error = EIO;
	}

done:
	/* Turn interface off, release resources */

	sc->rl_testmode = 0;
	sc->rl_link = 0;
	ifp->if_flags &= ~IFF_PROMISC;
	re_stop(sc);
	if (m0 != NULL)
		m_freem(m0);

	RL_UNLOCK(sc);

	return (error);
}

#endif

/*
 * Probe for a RealTek 8139C+/8169/8110 chip. Check the PCI vendor and device
 * IDs against our list and return a device name if we find a match.
 */
static int
re_probe(dev)
	device_t		dev;
{
	struct rl_type		*t;
	uint16_t		devid, vendor;
	uint16_t		revid, sdevid;
	int			i;
	
	vendor = pci_get_vendor(dev);
	devid = pci_get_device(dev);
	revid = pci_get_revid(dev);
	sdevid = pci_get_subdevice(dev);

	if (vendor == LINKSYS_VENDORID && devid == LINKSYS_DEVICEID_EG1032) {
		if (sdevid != LINKSYS_SUBDEVICE_EG1032_REV3) {
			/*
			 * Only attach to rev. 3 of the Linksys EG1032 adapter.
			 * Rev. 2 is supported by sk(4).
			 */
			return (ENXIO);
		}
	}

	if (vendor == RT_VENDORID && devid == RT_DEVICEID_8139) {
		if (revid != 0x20) {
			/* 8139, let rl(4) take care of this device. */
			return (ENXIO);
		}
	}

	t = re_devs;
	for (i = 0; i < sizeof(re_devs) / sizeof(re_devs[0]); i++, t++) {
		if (vendor == t->rl_vid && devid == t->rl_did) {
			device_set_desc(dev, t->rl_name);
			return (BUS_PROBE_DEFAULT);
		}
	}

	return (ENXIO);
}

/*
 * Map a single buffer address.
 */

static void
re_dma_map_addr(arg, segs, nseg, error)
	void			*arg;
	bus_dma_segment_t	*segs;
	int			nseg;
	int			error;
{
	bus_addr_t		*addr;

	if (error)
		return;

	KASSERT(nseg == 1, ("too many DMA segments, %d should be 1", nseg));
	addr = arg;
	*addr = segs->ds_addr;
}

static int
re_allocmem(dev, sc)
	device_t		dev;
	struct rl_softc		*sc;
{
	bus_size_t		rx_list_size, tx_list_size;
	int			error;
	int			i;

	rx_list_size = sc->rl_ldata.rl_rx_desc_cnt * sizeof(struct rl_desc);
	tx_list_size = sc->rl_ldata.rl_tx_desc_cnt * sizeof(struct rl_desc);

	/*
	 * Allocate the parent bus DMA tag appropriate for PCI.
	 * In order to use DAC, RL_CPLUSCMD_PCI_DAC bit of RL_CPLUS_CMD
	 * register should be set. However some RealTek chips are known
	 * to be buggy on DAC handling, therefore disable DAC by limiting
	 * DMA address space to 32bit. PCIe variants of RealTek chips
	 * may not have the limitation but I took safer path.
	 */
	error = bus_dma_tag_create(bus_get_dma_tag(dev), 1, 0,
	    BUS_SPACE_MAXADDR_32BIT, BUS_SPACE_MAXADDR, NULL, NULL,
	    BUS_SPACE_MAXSIZE_32BIT, 0, BUS_SPACE_MAXSIZE_32BIT, 0,
	    NULL, NULL, &sc->rl_parent_tag);
	if (error) {
		device_printf(dev, "could not allocate parent DMA tag\n");
		return (error);
	}

	/*
	 * Allocate map for TX mbufs.
	 */
	error = bus_dma_tag_create(sc->rl_parent_tag, 1, 0,
	    BUS_SPACE_MAXADDR, BUS_SPACE_MAXADDR, NULL,
	    NULL, MCLBYTES * RL_NTXSEGS, RL_NTXSEGS, 4096, 0,
	    NULL, NULL, &sc->rl_ldata.rl_tx_mtag);
	if (error) {
		device_printf(dev, "could not allocate TX DMA tag\n");
		return (error);
	}

	/*
	 * Allocate map for RX mbufs.
	 */

	error = bus_dma_tag_create(sc->rl_parent_tag, sizeof(uint64_t), 0,
	    BUS_SPACE_MAXADDR, BUS_SPACE_MAXADDR, NULL, NULL,
	    MCLBYTES, 1, MCLBYTES, 0, NULL, NULL, &sc->rl_ldata.rl_rx_mtag);
	if (error) {
		device_printf(dev, "could not allocate RX DMA tag\n");
		return (error);
	}

	/*
	 * Allocate map for TX descriptor list.
	 */
	error = bus_dma_tag_create(sc->rl_parent_tag, RL_RING_ALIGN,
	    0, BUS_SPACE_MAXADDR_32BIT, BUS_SPACE_MAXADDR, NULL,
	    NULL, tx_list_size, 1, tx_list_size, 0,
	    NULL, NULL, &sc->rl_ldata.rl_tx_list_tag);
	if (error) {
		device_printf(dev, "could not allocate TX DMA ring tag\n");
		return (error);
	}

	/* Allocate DMA'able memory for the TX ring */

	error = bus_dmamem_alloc(sc->rl_ldata.rl_tx_list_tag,
	    (void **)&sc->rl_ldata.rl_tx_list,
	    BUS_DMA_WAITOK | BUS_DMA_COHERENT | BUS_DMA_ZERO,
	    &sc->rl_ldata.rl_tx_list_map);
	if (error) {
		device_printf(dev, "could not allocate TX DMA ring\n");
		return (error);
	}

	/* Load the map for the TX ring. */

	sc->rl_ldata.rl_tx_list_addr = 0;
	error = bus_dmamap_load(sc->rl_ldata.rl_tx_list_tag,
	     sc->rl_ldata.rl_tx_list_map, sc->rl_ldata.rl_tx_list,
	     tx_list_size, re_dma_map_addr,
	     &sc->rl_ldata.rl_tx_list_addr, BUS_DMA_NOWAIT);
	if (error != 0 || sc->rl_ldata.rl_tx_list_addr == 0) {
		device_printf(dev, "could not load TX DMA ring\n");
		return (ENOMEM);
	}

	/* Create DMA maps for TX buffers */

	for (i = 0; i < sc->rl_ldata.rl_tx_desc_cnt; i++) {
		error = bus_dmamap_create(sc->rl_ldata.rl_tx_mtag, 0,
		    &sc->rl_ldata.rl_tx_desc[i].tx_dmamap);
		if (error) {
			device_printf(dev, "could not create DMA map for TX\n");
			return (error);
		}
	}

	/*
	 * Allocate map for RX descriptor list.
	 */
	error = bus_dma_tag_create(sc->rl_parent_tag, RL_RING_ALIGN,
	    0, BUS_SPACE_MAXADDR_32BIT, BUS_SPACE_MAXADDR, NULL,
	    NULL, rx_list_size, 1, rx_list_size, 0,
	    NULL, NULL, &sc->rl_ldata.rl_rx_list_tag);
	if (error) {
		device_printf(dev, "could not create RX DMA ring tag\n");
		return (error);
	}

	/* Allocate DMA'able memory for the RX ring */

	error = bus_dmamem_alloc(sc->rl_ldata.rl_rx_list_tag,
	    (void **)&sc->rl_ldata.rl_rx_list,
	    BUS_DMA_WAITOK | BUS_DMA_COHERENT | BUS_DMA_ZERO,
	    &sc->rl_ldata.rl_rx_list_map);
	if (error) {
		device_printf(dev, "could not allocate RX DMA ring\n");
		return (error);
	}

	/* Load the map for the RX ring. */

	sc->rl_ldata.rl_rx_list_addr = 0;
	error = bus_dmamap_load(sc->rl_ldata.rl_rx_list_tag,
	     sc->rl_ldata.rl_rx_list_map, sc->rl_ldata.rl_rx_list,
	     rx_list_size, re_dma_map_addr,
	     &sc->rl_ldata.rl_rx_list_addr, BUS_DMA_NOWAIT);
	if (error != 0 || sc->rl_ldata.rl_rx_list_addr == 0) {
		device_printf(dev, "could not load RX DMA ring\n");
		return (ENOMEM);
	}

	/* Create DMA maps for RX buffers */

	error = bus_dmamap_create(sc->rl_ldata.rl_rx_mtag, 0,
	    &sc->rl_ldata.rl_rx_sparemap);
	if (error) {
		device_printf(dev, "could not create spare DMA map for RX\n");
		return (error);
	}
	for (i = 0; i < sc->rl_ldata.rl_rx_desc_cnt; i++) {
		error = bus_dmamap_create(sc->rl_ldata.rl_rx_mtag, 0,
		    &sc->rl_ldata.rl_rx_desc[i].rx_dmamap);
		if (error) {
			device_printf(dev, "could not create DMA map for RX\n");
			return (error);
		}
	}

	return (0);
}

/*
 * Attach the interface. Allocate softc structures, do ifmedia
 * setup and ethernet/BPF attach.
 */
static int
re_attach(dev)
	device_t		dev;
{
	u_char			eaddr[ETHER_ADDR_LEN];
	u_int16_t		as[ETHER_ADDR_LEN / 2];
	struct rl_softc		*sc;
	struct ifnet		*ifp;
	struct rl_hwrev		*hw_rev;
	int			hwrev;
	u_int16_t		re_did = 0;
	int			error = 0, rid, i;
	int			msic, reg;
	uint8_t			cfg;

	sc = device_get_softc(dev);
	sc->rl_dev = dev;

	mtx_init(&sc->rl_mtx, device_get_nameunit(dev), MTX_NETWORK_LOCK,
	    MTX_DEF);
	callout_init_mtx(&sc->rl_stat_callout, &sc->rl_mtx, 0);

	/*
	 * Map control/status registers.
	 */
	pci_enable_busmaster(dev);

	rid = RL_RID;
	sc->rl_res = bus_alloc_resource_any(dev, RL_RES, &rid,
	    RF_ACTIVE);

	if (sc->rl_res == NULL) {
		device_printf(dev, "couldn't map ports/memory\n");
		error = ENXIO;
		goto fail;
	}

	sc->rl_btag = rman_get_bustag(sc->rl_res);
	sc->rl_bhandle = rman_get_bushandle(sc->rl_res);

	msic = 0;
	if (pci_find_extcap(dev, PCIY_EXPRESS, &reg) == 0) {
		msic = pci_msi_count(dev);
		if (bootverbose)
			device_printf(dev, "MSI count : %d\n", msic);
	}
	if (msic == RL_MSI_MESSAGES  && msi_disable == 0) {
		if (pci_alloc_msi(dev, &msic) == 0) {
			if (msic == RL_MSI_MESSAGES) {
				device_printf(dev, "Using %d MSI messages\n",
				    msic);
				sc->rl_msi = 1;
				/* Explicitly set MSI enable bit. */
				CSR_WRITE_1(sc, RL_EECMD, RL_EE_MODE);
				cfg = CSR_READ_1(sc, RL_CFG2);
				cfg |= RL_CFG2_MSI;
				CSR_WRITE_1(sc, RL_CFG2, cfg);
				CSR_WRITE_1(sc, RL_EECMD, 0);
			} else
				pci_release_msi(dev);
		}
	}

	/* Allocate interrupt */
	if (sc->rl_msi == 0) {
		rid = 0;
		sc->rl_irq[0] = bus_alloc_resource_any(dev, SYS_RES_IRQ, &rid,
		    RF_SHAREABLE | RF_ACTIVE);
		if (sc->rl_irq[0] == NULL) {
			device_printf(dev, "couldn't allocate IRQ resources\n");
			error = ENXIO;
			goto fail;
		}
	} else {
		for (i = 0, rid = 1; i < RL_MSI_MESSAGES; i++, rid++) {
			sc->rl_irq[i] = bus_alloc_resource_any(dev,
			    SYS_RES_IRQ, &rid, RF_ACTIVE);
			if (sc->rl_irq[i] == NULL) {
				device_printf(dev,
				    "couldn't llocate IRQ resources for "
				    "message %d\n", rid);
				error = ENXIO;
				goto fail;
			}
		}
	}

	/* Reset the adapter. */
	RL_LOCK(sc);
	re_reset(sc);
	RL_UNLOCK(sc);

	hw_rev = re_hwrevs;
	hwrev = CSR_READ_4(sc, RL_TXCFG) & RL_TXCFG_HWREV;
	while (hw_rev->rl_desc != NULL) {
		if (hw_rev->rl_rev == hwrev) {
			sc->rl_type = hw_rev->rl_type;
			break;
		}
		hw_rev++;
	}
	if (hw_rev->rl_desc == NULL) {
		device_printf(dev, "Unknown H/W revision: %08x\n", hwrev);
		error = ENXIO;
		goto fail;
	}

	sc->rl_eewidth = RL_9356_ADDR_LEN;
	re_read_eeprom(sc, (caddr_t)&re_did, 0, 1);
	if (re_did != 0x8129)
	        sc->rl_eewidth = RL_9346_ADDR_LEN;

	/*
	 * Get station address from the EEPROM.
	 */
	re_read_eeprom(sc, (caddr_t)as, RL_EE_EADDR, 3);
	for (i = 0; i < ETHER_ADDR_LEN / 2; i++)
		as[i] = le16toh(as[i]);
	bcopy(as, eaddr, sizeof(eaddr));

	if (sc->rl_type == RL_8169) {
		/* Set RX length mask and number of descriptors. */
		sc->rl_rxlenmask = RL_RDESC_STAT_GFRAGLEN;
		sc->rl_txstart = RL_GTXSTART;
		sc->rl_ldata.rl_tx_desc_cnt = RL_8169_TX_DESC_CNT;
		sc->rl_ldata.rl_rx_desc_cnt = RL_8169_RX_DESC_CNT;
	} else {
		/* Set RX length mask and number of descriptors. */
		sc->rl_rxlenmask = RL_RDESC_STAT_FRAGLEN;
		sc->rl_txstart = RL_TXSTART;
		sc->rl_ldata.rl_tx_desc_cnt = RL_8139_TX_DESC_CNT;
		sc->rl_ldata.rl_rx_desc_cnt = RL_8139_RX_DESC_CNT;
	}
	if (hw_rev->rl_desc == NULL) {
		device_printf(dev, "Unsupported revision : 0x%08x\n", hwrev);
		error = ENXIO;
		goto fail;
	}

	error = re_allocmem(dev, sc);
	if (error)
		goto fail;

	ifp = sc->rl_ifp = if_alloc(IFT_ETHER);
	if (ifp == NULL) {
		device_printf(dev, "can not if_alloc()\n");
		error = ENOSPC;
		goto fail;
	}

	/* Do MII setup */
	if (mii_phy_probe(dev, &sc->rl_miibus,
	    re_ifmedia_upd, re_ifmedia_sts)) {
		device_printf(dev, "MII without any phy!\n");
		error = ENXIO;
		goto fail;
	}

	/* Take PHY out of power down mode. */
	if (sc->rl_type == RL_8169) {
		uint32_t rev;

		rev = CSR_READ_4(sc, RL_TXCFG);
		/* HWVERID 0, 1 and 2 :  bit26-30, bit23 */
		rev &= 0x7c800000;
		if (rev != 0) {
			/* RTL8169S single chip */
			switch (rev) {
			case RL_HWREV_8169_8110SB:
			case RL_HWREV_8169_8110SC:
			case RL_HWREV_8168_SPIN2:
			case RL_HWREV_8168_SPIN3:
				re_gmii_writereg(dev, 1, 0x1f, 0);
				re_gmii_writereg(dev, 1, 0x0e, 0);
				break;
			default:
				break;
			}
		}
	}

	ifp->if_softc = sc;
	if_initname(ifp, device_get_name(dev), device_get_unit(dev));
	ifp->if_flags = IFF_BROADCAST | IFF_SIMPLEX | IFF_MULTICAST;
	ifp->if_ioctl = re_ioctl;
	ifp->if_start = re_start;
	ifp->if_hwassist = RE_CSUM_FEATURES | CSUM_TSO;
	ifp->if_capabilities = IFCAP_HWCSUM | IFCAP_TSO4;
	ifp->if_capenable = ifp->if_capabilities;
	ifp->if_init = re_init;
	IFQ_SET_MAXLEN(&ifp->if_snd, RL_IFQ_MAXLEN);
	ifp->if_snd.ifq_drv_maxlen = RL_IFQ_MAXLEN;
	IFQ_SET_READY(&ifp->if_snd);

	TASK_INIT(&sc->rl_txtask, 1, re_tx_task, ifp);
	TASK_INIT(&sc->rl_inttask, 0, re_int_task, sc);

	/*
	 * Call MI attach routine.
	 */
	ether_ifattach(ifp, eaddr);

	/* VLAN capability setup */
	ifp->if_capabilities |= IFCAP_VLAN_MTU | IFCAP_VLAN_HWTAGGING;
	if (ifp->if_capabilities & IFCAP_HWCSUM)
		ifp->if_capabilities |= IFCAP_VLAN_HWCSUM;
	/* Enable WOL if PM is supported. */
	if (pci_find_extcap(sc->rl_dev, PCIY_PMG, &reg) == 0)
		ifp->if_capabilities |= IFCAP_WOL;
	ifp->if_capenable = ifp->if_capabilities;
#ifdef DEVICE_POLLING
	ifp->if_capabilities |= IFCAP_POLLING;
#endif
	/*
	 * Tell the upper layer(s) we support long frames.
	 * Must appear after the call to ether_ifattach() because
	 * ether_ifattach() sets ifi_hdrlen to the default value.
	 */
	ifp->if_data.ifi_hdrlen = sizeof(struct ether_vlan_header);

#ifdef RE_DIAG
	/*
	 * Perform hardware diagnostic on the original RTL8169.
	 * Some 32-bit cards were incorrectly wired and would
	 * malfunction if plugged into a 64-bit slot.
	 */

	if (hwrev == RL_HWREV_8169) {
		error = re_diag(sc);
		if (error) {
			device_printf(dev,
		    	"attach aborted due to hardware diag failure\n");
			ether_ifdetach(ifp);
			goto fail;
		}
	}
#endif

	/* Hook interrupt last to avoid having to lock softc */
	if (sc->rl_msi == 0)
		error = bus_setup_intr(dev, sc->rl_irq[0],
		    INTR_TYPE_NET | INTR_MPSAFE, re_intr, NULL, sc,
		    &sc->rl_intrhand[0]);
	else {
		for (i = 0; i < RL_MSI_MESSAGES; i++) {
			error = bus_setup_intr(dev, sc->rl_irq[i],
			    INTR_TYPE_NET | INTR_MPSAFE, re_intr, NULL, sc,
		    	    &sc->rl_intrhand[i]);
			if (error != 0)
				break;
		}
	}
	if (error) {
		device_printf(dev, "couldn't set up irq\n");
		ether_ifdetach(ifp);
	}

fail:

	if (error)
		re_detach(dev);

	return (error);
}

/*
 * Shutdown hardware and free up resources. This can be called any
 * time after the mutex has been initialized. It is called in both
 * the error case in attach and the normal detach case so it needs
 * to be careful about only freeing resources that have actually been
 * allocated.
 */
static int
re_detach(dev)
	device_t		dev;
{
	struct rl_softc		*sc;
	struct ifnet		*ifp;
	int			i, rid;

	sc = device_get_softc(dev);
	ifp = sc->rl_ifp;
	KASSERT(mtx_initialized(&sc->rl_mtx), ("re mutex not initialized"));

#ifdef DEVICE_POLLING
	if (ifp->if_capenable & IFCAP_POLLING)
		ether_poll_deregister(ifp);
#endif
	/* These should only be active if attach succeeded */
	if (device_is_attached(dev)) {
		RL_LOCK(sc);
#if 0
		sc->suspended = 1;
#endif
		re_stop(sc);
		RL_UNLOCK(sc);
		callout_drain(&sc->rl_stat_callout);
		taskqueue_drain(taskqueue_fast, &sc->rl_inttask);
		taskqueue_drain(taskqueue_fast, &sc->rl_txtask);
		/*
		 * Force off the IFF_UP flag here, in case someone
		 * still had a BPF descriptor attached to this
		 * interface. If they do, ether_ifdetach() will cause
		 * the BPF code to try and clear the promisc mode
		 * flag, which will bubble down to re_ioctl(),
		 * which will try to call re_init() again. This will
		 * turn the NIC back on and restart the MII ticker,
		 * which will panic the system when the kernel tries
		 * to invoke the re_tick() function that isn't there
		 * anymore.
		 */
		ifp->if_flags &= ~IFF_UP;
		ether_ifdetach(ifp);
	}
	if (sc->rl_miibus)
		device_delete_child(dev, sc->rl_miibus);
	bus_generic_detach(dev);

	/*
	 * The rest is resource deallocation, so we should already be
	 * stopped here.
	 */

	for (i = 0; i < RL_MSI_MESSAGES; i++) {
		if (sc->rl_intrhand[i] != NULL) {
			bus_teardown_intr(dev, sc->rl_irq[i],
			    sc->rl_intrhand[i]);
			sc->rl_intrhand[i] = NULL;
		}
	}
	if (ifp != NULL)
		if_free(ifp);
	if (sc->rl_msi == 0) {
		if (sc->rl_irq[0] != NULL) {
			bus_release_resource(dev, SYS_RES_IRQ, 0,
			    sc->rl_irq[0]);
			sc->rl_irq[0] = NULL;
		}
	} else {
		for (i = 0, rid = 1; i < RL_MSI_MESSAGES; i++, rid++) {
			if (sc->rl_irq[i] != NULL) {
				bus_release_resource(dev, SYS_RES_IRQ, rid,
				    sc->rl_irq[i]);
				sc->rl_irq[i] = NULL;
			}
		}
		pci_release_msi(dev);
	}
	if (sc->rl_res)
		bus_release_resource(dev, RL_RES, RL_RID, sc->rl_res);

	/* Unload and free the RX DMA ring memory and map */

	if (sc->rl_ldata.rl_rx_list_tag) {
		bus_dmamap_unload(sc->rl_ldata.rl_rx_list_tag,
		    sc->rl_ldata.rl_rx_list_map);
		bus_dmamem_free(sc->rl_ldata.rl_rx_list_tag,
		    sc->rl_ldata.rl_rx_list,
		    sc->rl_ldata.rl_rx_list_map);
		bus_dma_tag_destroy(sc->rl_ldata.rl_rx_list_tag);
	}

	/* Unload and free the TX DMA ring memory and map */

	if (sc->rl_ldata.rl_tx_list_tag) {
		bus_dmamap_unload(sc->rl_ldata.rl_tx_list_tag,
		    sc->rl_ldata.rl_tx_list_map);
		bus_dmamem_free(sc->rl_ldata.rl_tx_list_tag,
		    sc->rl_ldata.rl_tx_list,
		    sc->rl_ldata.rl_tx_list_map);
		bus_dma_tag_destroy(sc->rl_ldata.rl_tx_list_tag);
	}

	/* Destroy all the RX and TX buffer maps */

	if (sc->rl_ldata.rl_tx_mtag) {
		for (i = 0; i < sc->rl_ldata.rl_tx_desc_cnt; i++)
			bus_dmamap_destroy(sc->rl_ldata.rl_tx_mtag,
			    sc->rl_ldata.rl_tx_desc[i].tx_dmamap);
		bus_dma_tag_destroy(sc->rl_ldata.rl_tx_mtag);
	}
	if (sc->rl_ldata.rl_rx_mtag) {
		for (i = 0; i < sc->rl_ldata.rl_rx_desc_cnt; i++)
			bus_dmamap_destroy(sc->rl_ldata.rl_rx_mtag,
			    sc->rl_ldata.rl_rx_desc[i].rx_dmamap);
		if (sc->rl_ldata.rl_rx_sparemap)
			bus_dmamap_destroy(sc->rl_ldata.rl_rx_mtag,
			    sc->rl_ldata.rl_rx_sparemap);
		bus_dma_tag_destroy(sc->rl_ldata.rl_rx_mtag);
	}

	/* Unload and free the stats buffer and map */

	if (sc->rl_ldata.rl_stag) {
		bus_dmamap_unload(sc->rl_ldata.rl_stag,
		    sc->rl_ldata.rl_rx_list_map);
		bus_dmamem_free(sc->rl_ldata.rl_stag,
		    sc->rl_ldata.rl_stats,
		    sc->rl_ldata.rl_smap);
		bus_dma_tag_destroy(sc->rl_ldata.rl_stag);
	}

	if (sc->rl_parent_tag)
		bus_dma_tag_destroy(sc->rl_parent_tag);

	mtx_destroy(&sc->rl_mtx);

	return (0);
}

static __inline void
re_discard_rxbuf(sc, idx)
	struct rl_softc		*sc;
	int			idx;
{
	struct rl_desc		*desc;
	struct rl_rxdesc	*rxd;
	uint32_t		cmdstat;

	rxd = &sc->rl_ldata.rl_rx_desc[idx];
	desc = &sc->rl_ldata.rl_rx_list[idx];
	desc->rl_vlanctl = 0;
	cmdstat = rxd->rx_size;
	if (idx == sc->rl_ldata.rl_rx_desc_cnt - 1)
		cmdstat |= RL_RDESC_CMD_EOR;
	desc->rl_cmdstat = htole32(cmdstat | RL_RDESC_CMD_OWN);
}

static int
re_newbuf(sc, idx)
	struct rl_softc		*sc;
	int			idx;
{
	struct mbuf		*m;
	struct rl_rxdesc	*rxd;
	bus_dma_segment_t	segs[1];
	bus_dmamap_t		map;
	struct rl_desc		*desc;
	uint32_t		cmdstat;
	int			error, nsegs;

	m = m_getcl(M_DONTWAIT, MT_DATA, M_PKTHDR);
	if (m == NULL)
		return (ENOBUFS);

	m->m_len = m->m_pkthdr.len = MCLBYTES;
#ifdef RE_FIXUP_RX
	/*
	 * This is part of an evil trick to deal with non-x86 platforms.
	 * The RealTek chip requires RX buffers to be aligned on 64-bit
	 * boundaries, but that will hose non-x86 machines. To get around
	 * this, we leave some empty space at the start of each buffer
	 * and for non-x86 hosts, we copy the buffer back six bytes
	 * to achieve word alignment. This is slightly more efficient
	 * than allocating a new buffer, copying the contents, and
	 * discarding the old buffer.
	 */
	m_adj(m, RE_ETHER_ALIGN);
#endif
	error = bus_dmamap_load_mbuf_sg(sc->rl_ldata.rl_rx_mtag,
	    sc->rl_ldata.rl_rx_sparemap, m, segs, &nsegs, BUS_DMA_NOWAIT);
	if (error != 0) {
		m_freem(m);
		return (ENOBUFS);
	}
	KASSERT(nsegs == 1, ("%s: %d segment returned!", __func__, nsegs));

	rxd = &sc->rl_ldata.rl_rx_desc[idx];
	if (rxd->rx_m != NULL) {
		bus_dmamap_sync(sc->rl_ldata.rl_rx_mtag, rxd->rx_dmamap,
		    BUS_DMASYNC_POSTREAD);
		bus_dmamap_unload(sc->rl_ldata.rl_rx_mtag, rxd->rx_dmamap);
	}

	rxd->rx_m = m;
	map = rxd->rx_dmamap;
	rxd->rx_dmamap = sc->rl_ldata.rl_rx_sparemap;
	rxd->rx_size = segs[0].ds_len;
	sc->rl_ldata.rl_rx_sparemap = map;
	bus_dmamap_sync(sc->rl_ldata.rl_rx_mtag, rxd->rx_dmamap,
	    BUS_DMASYNC_PREREAD);

	desc = &sc->rl_ldata.rl_rx_list[idx];
	desc->rl_vlanctl = 0;
	desc->rl_bufaddr_lo = htole32(RL_ADDR_LO(segs[0].ds_addr));
	desc->rl_bufaddr_hi = htole32(RL_ADDR_HI(segs[0].ds_addr));
	cmdstat = segs[0].ds_len;
	if (idx == sc->rl_ldata.rl_rx_desc_cnt - 1)
		cmdstat |= RL_RDESC_CMD_EOR;
	desc->rl_cmdstat = htole32(cmdstat | RL_RDESC_CMD_OWN);

	return (0);
}

#ifdef RE_FIXUP_RX
static __inline void
re_fixup_rx(m)
	struct mbuf		*m;
{
	int                     i;
	uint16_t                *src, *dst;

	src = mtod(m, uint16_t *);
	dst = src - (RE_ETHER_ALIGN - ETHER_ALIGN) / sizeof *src;

	for (i = 0; i < (m->m_len / sizeof(uint16_t) + 1); i++)
		*dst++ = *src++;

	m->m_data -= RE_ETHER_ALIGN - ETHER_ALIGN;

	return;
}
#endif

static int
re_tx_list_init(sc)
	struct rl_softc		*sc;
{
	struct rl_desc		*desc;
	int			i;

	RL_LOCK_ASSERT(sc);

	bzero(sc->rl_ldata.rl_tx_list,
	    sc->rl_ldata.rl_tx_desc_cnt * sizeof(struct rl_desc));
	for (i = 0; i < sc->rl_ldata.rl_tx_desc_cnt; i++)
		sc->rl_ldata.rl_tx_desc[i].tx_m = NULL;
	/* Set EOR. */
	desc = &sc->rl_ldata.rl_tx_list[sc->rl_ldata.rl_tx_desc_cnt - 1];
	desc->rl_cmdstat |= htole32(RL_TDESC_CMD_EOR);

	bus_dmamap_sync(sc->rl_ldata.rl_tx_list_tag,
	    sc->rl_ldata.rl_tx_list_map,
	    BUS_DMASYNC_PREREAD | BUS_DMASYNC_PREWRITE);

	sc->rl_ldata.rl_tx_prodidx = 0;
	sc->rl_ldata.rl_tx_considx = 0;
	sc->rl_ldata.rl_tx_free = sc->rl_ldata.rl_tx_desc_cnt;

	return (0);
}

static int
re_rx_list_init(sc)
	struct rl_softc		*sc;
{
	int			error, i;

	bzero(sc->rl_ldata.rl_rx_list,
	    sc->rl_ldata.rl_rx_desc_cnt * sizeof(struct rl_desc));
	for (i = 0; i < sc->rl_ldata.rl_rx_desc_cnt; i++) {
		sc->rl_ldata.rl_rx_desc[i].rx_m = NULL;
		if ((error = re_newbuf(sc, i)) != 0)
			return (error);
	}

	/* Flush the RX descriptors */

	bus_dmamap_sync(sc->rl_ldata.rl_rx_list_tag,
	    sc->rl_ldata.rl_rx_list_map,
	    BUS_DMASYNC_PREWRITE|BUS_DMASYNC_PREREAD);

	sc->rl_ldata.rl_rx_prodidx = 0;
	sc->rl_head = sc->rl_tail = NULL;

	return (0);
}

/*
 * RX handler for C+ and 8169. For the gigE chips, we support
 * the reception of jumbo frames that have been fragmented
 * across multiple 2K mbuf cluster buffers.
 */
static int
re_rxeof(sc)
	struct rl_softc		*sc;
{
	struct mbuf		*m;
	struct ifnet		*ifp;
	int			i, total_len;
	struct rl_desc		*cur_rx;
	u_int32_t		rxstat, rxvlan;
	int			maxpkt = 16;

	RL_LOCK_ASSERT(sc);

	ifp = sc->rl_ifp;

	/* Invalidate the descriptor memory */

	bus_dmamap_sync(sc->rl_ldata.rl_rx_list_tag,
	    sc->rl_ldata.rl_rx_list_map,
	    BUS_DMASYNC_POSTREAD | BUS_DMASYNC_POSTWRITE);

	for (i = sc->rl_ldata.rl_rx_prodidx; maxpkt > 0;
	    i = RL_RX_DESC_NXT(sc, i)) {
		cur_rx = &sc->rl_ldata.rl_rx_list[i];
		rxstat = le32toh(cur_rx->rl_cmdstat);
		if ((rxstat & RL_RDESC_STAT_OWN) != 0)
			break;
		total_len = rxstat & sc->rl_rxlenmask;
		rxvlan = le32toh(cur_rx->rl_vlanctl);
		m = sc->rl_ldata.rl_rx_desc[i].rx_m;

		if (!(rxstat & RL_RDESC_STAT_EOF)) {
			if (re_newbuf(sc, i) != 0) {
				/*
				 * If this is part of a multi-fragment packet,
				 * discard all the pieces.
				 */
				if (sc->rl_head != NULL) {
					m_freem(sc->rl_head);
					sc->rl_head = sc->rl_tail = NULL;
				}
				re_discard_rxbuf(sc, i);
				continue;
			}
			m->m_len = RE_RX_DESC_BUFLEN;
			if (sc->rl_head == NULL)
				sc->rl_head = sc->rl_tail = m;
			else {
				m->m_flags &= ~M_PKTHDR;
				sc->rl_tail->m_next = m;
				sc->rl_tail = m;
			}
			continue;
		}

		/*
		 * NOTE: for the 8139C+, the frame length field
		 * is always 12 bits in size, but for the gigE chips,
		 * it is 13 bits (since the max RX frame length is 16K).
		 * Unfortunately, all 32 bits in the status word
		 * were already used, so to make room for the extra
		 * length bit, RealTek took out the 'frame alignment
		 * error' bit and shifted the other status bits
		 * over one slot. The OWN, EOR, FS and LS bits are
		 * still in the same places. We have already extracted
		 * the frame length and checked the OWN bit, so rather
		 * than using an alternate bit mapping, we shift the
		 * status bits one space to the right so we can evaluate
		 * them using the 8169 status as though it was in the
		 * same format as that of the 8139C+.
		 */
		if (sc->rl_type == RL_8169)
			rxstat >>= 1;

		/*
		 * if total_len > 2^13-1, both _RXERRSUM and _GIANT will be
		 * set, but if CRC is clear, it will still be a valid frame.
		 */
		if (rxstat & RL_RDESC_STAT_RXERRSUM && !(total_len > 8191 &&
		    (rxstat & RL_RDESC_STAT_ERRS) == RL_RDESC_STAT_GIANT)) {
			ifp->if_ierrors++;
			/*
			 * If this is part of a multi-fragment packet,
			 * discard all the pieces.
			 */
			if (sc->rl_head != NULL) {
				m_freem(sc->rl_head);
				sc->rl_head = sc->rl_tail = NULL;
			}
			re_discard_rxbuf(sc, i);
			continue;
		}

		/*
		 * If allocating a replacement mbuf fails,
		 * reload the current one.
		 */

		if (re_newbuf(sc, i) != 0) {
			ifp->if_iqdrops++;
			if (sc->rl_head != NULL) {
				m_freem(sc->rl_head);
				sc->rl_head = sc->rl_tail = NULL;
			}
			re_discard_rxbuf(sc, i);
			continue;
		}

		if (sc->rl_head != NULL) {
			m->m_len = total_len % RE_RX_DESC_BUFLEN;
			if (m->m_len == 0)
				m->m_len = RE_RX_DESC_BUFLEN;
			/*
			 * Special case: if there's 4 bytes or less
			 * in this buffer, the mbuf can be discarded:
			 * the last 4 bytes is the CRC, which we don't
			 * care about anyway.
			 */
			if (m->m_len <= ETHER_CRC_LEN) {
				sc->rl_tail->m_len -=
				    (ETHER_CRC_LEN - m->m_len);
				m_freem(m);
			} else {
				m->m_len -= ETHER_CRC_LEN;
				m->m_flags &= ~M_PKTHDR;
				sc->rl_tail->m_next = m;
			}
			m = sc->rl_head;
			sc->rl_head = sc->rl_tail = NULL;
			m->m_pkthdr.len = total_len - ETHER_CRC_LEN;
		} else
			m->m_pkthdr.len = m->m_len =
			    (total_len - ETHER_CRC_LEN);

#ifdef RE_FIXUP_RX
		re_fixup_rx(m);
#endif
		ifp->if_ipackets++;
		m->m_pkthdr.rcvif = ifp;

		/* Do RX checksumming if enabled */

		if (ifp->if_capenable & IFCAP_RXCSUM) {

			/* Check IP header checksum */
			if (rxstat & RL_RDESC_STAT_PROTOID)
				m->m_pkthdr.csum_flags |= CSUM_IP_CHECKED;
			if (!(rxstat & RL_RDESC_STAT_IPSUMBAD))
				m->m_pkthdr.csum_flags |= CSUM_IP_VALID;

			/* Check TCP/UDP checksum */
			if ((RL_TCPPKT(rxstat) &&
			    !(rxstat & RL_RDESC_STAT_TCPSUMBAD)) ||
			    (RL_UDPPKT(rxstat) &&
			    !(rxstat & RL_RDESC_STAT_UDPSUMBAD))) {
				m->m_pkthdr.csum_flags |=
				    CSUM_DATA_VALID|CSUM_PSEUDO_HDR;
				m->m_pkthdr.csum_data = 0xffff;
			}
		}
		maxpkt--;
		if (rxvlan & RL_RDESC_VLANCTL_TAG) {
			m->m_pkthdr.ether_vtag =
			    ntohs((rxvlan & RL_RDESC_VLANCTL_DATA));
			m->m_flags |= M_VLANTAG;
		}
		RL_UNLOCK(sc);
		(*ifp->if_input)(ifp, m);
		RL_LOCK(sc);
	}

	/* Flush the RX DMA ring */

	bus_dmamap_sync(sc->rl_ldata.rl_rx_list_tag,
	    sc->rl_ldata.rl_rx_list_map,
	    BUS_DMASYNC_PREWRITE|BUS_DMASYNC_PREREAD);

	sc->rl_ldata.rl_rx_prodidx = i;

	if (maxpkt)
		return(EAGAIN);

	return(0);
}

static void
re_txeof(sc)
	struct rl_softc		*sc;
{
	struct ifnet		*ifp;
	struct rl_txdesc	*txd;
	u_int32_t		txstat;
	int			cons;

	cons = sc->rl_ldata.rl_tx_considx;
	if (cons == sc->rl_ldata.rl_tx_prodidx)
		return;

	ifp = sc->rl_ifp;
	/* Invalidate the TX descriptor list */
	bus_dmamap_sync(sc->rl_ldata.rl_tx_list_tag,
	    sc->rl_ldata.rl_tx_list_map,
	    BUS_DMASYNC_POSTREAD | BUS_DMASYNC_POSTWRITE);

	for (; cons != sc->rl_ldata.rl_tx_prodidx;
	    cons = RL_TX_DESC_NXT(sc, cons)) {
		txstat = le32toh(sc->rl_ldata.rl_tx_list[cons].rl_cmdstat);
		if (txstat & RL_TDESC_STAT_OWN)
			break;
		/*
		 * We only stash mbufs in the last descriptor
		 * in a fragment chain, which also happens to
		 * be the only place where the TX status bits
		 * are valid.
		 */
		if (txstat & RL_TDESC_CMD_EOF) {
			txd = &sc->rl_ldata.rl_tx_desc[cons];
			bus_dmamap_sync(sc->rl_ldata.rl_tx_mtag,
			    txd->tx_dmamap, BUS_DMASYNC_POSTWRITE);
			bus_dmamap_unload(sc->rl_ldata.rl_tx_mtag,
			    txd->tx_dmamap);
			KASSERT(txd->tx_m != NULL,
			    ("%s: freeing NULL mbufs!", __func__));
			m_freem(txd->tx_m);
			txd->tx_m = NULL;
			if (txstat & (RL_TDESC_STAT_EXCESSCOL|
			    RL_TDESC_STAT_COLCNT))
				ifp->if_collisions++;
			if (txstat & RL_TDESC_STAT_TXERRSUM)
				ifp->if_oerrors++;
			else
				ifp->if_opackets++;
		}
		sc->rl_ldata.rl_tx_free++;
		ifp->if_drv_flags &= ~IFF_DRV_OACTIVE;
	}
	sc->rl_ldata.rl_tx_considx = cons;

	/* No changes made to the TX ring, so no flush needed */

	if (sc->rl_ldata.rl_tx_free != sc->rl_ldata.rl_tx_desc_cnt) {
		/*
		 * Some chips will ignore a second TX request issued
		 * while an existing transmission is in progress. If
		 * the transmitter goes idle but there are still
		 * packets waiting to be sent, we need to restart the
		 * channel here to flush them out. This only seems to
		 * be required with the PCIe devices.
		 */
		CSR_WRITE_1(sc, sc->rl_txstart, RL_TXSTART_START);

#ifdef RE_TX_MODERATION
		/*
		 * If not all descriptors have been reaped yet, reload
		 * the timer so that we will eventually get another
		 * interrupt that will cause us to re-enter this routine.
		 * This is done in case the transmitter has gone idle.
		 */
		CSR_WRITE_4(sc, RL_TIMERCNT, 1);
#endif
	} else
		sc->rl_watchdog_timer = 0;
}

static void
re_tick(xsc)
	void			*xsc;
{
	struct rl_softc		*sc;
	struct mii_data		*mii;
	struct ifnet		*ifp;

	sc = xsc;
	ifp = sc->rl_ifp;

	RL_LOCK_ASSERT(sc);

	re_watchdog(sc);

	mii = device_get_softc(sc->rl_miibus);
	mii_tick(mii);
	if (sc->rl_link) {
		if (!(mii->mii_media_status & IFM_ACTIVE))
			sc->rl_link = 0;
	} else {
		if (mii->mii_media_status & IFM_ACTIVE &&
		    IFM_SUBTYPE(mii->mii_media_active) != IFM_NONE) {
			sc->rl_link = 1;
			if (!IFQ_DRV_IS_EMPTY(&ifp->if_snd))
				taskqueue_enqueue_fast(taskqueue_fast,
				    &sc->rl_txtask);
		}
	}

	callout_reset(&sc->rl_stat_callout, hz, re_tick, sc);
}

#ifdef DEVICE_POLLING
static void
re_poll(struct ifnet *ifp, enum poll_cmd cmd, int count)
{
	struct rl_softc *sc = ifp->if_softc;

	RL_LOCK(sc);
	if (ifp->if_drv_flags & IFF_DRV_RUNNING)
		re_poll_locked(ifp, cmd, count);
	RL_UNLOCK(sc);
}

static void
re_poll_locked(struct ifnet *ifp, enum poll_cmd cmd, int count)
{
	struct rl_softc *sc = ifp->if_softc;

	RL_LOCK_ASSERT(sc);

	sc->rxcycles = count;
	re_rxeof(sc);
	re_txeof(sc);

	if (!IFQ_DRV_IS_EMPTY(&ifp->if_snd))
		taskqueue_enqueue_fast(taskqueue_fast, &sc->rl_txtask);

	if (cmd == POLL_AND_CHECK_STATUS) { /* also check status register */
		u_int16_t       status;

		status = CSR_READ_2(sc, RL_ISR);
		if (status == 0xffff)
			return;
		if (status)
			CSR_WRITE_2(sc, RL_ISR, status);

		/*
		 * XXX check behaviour on receiver stalls.
		 */

		if (status & RL_ISR_SYSTEM_ERR) {
			re_reset(sc);
			re_init_locked(sc);
		}
	}
}
#endif /* DEVICE_POLLING */

static int
re_intr(arg)
	void			*arg;
{
	struct rl_softc		*sc;
	uint16_t		status;

	sc = arg;

	status = CSR_READ_2(sc, RL_ISR);
	if (status == 0xFFFF || (status & RL_INTRS_CPLUS) == 0)
                return (FILTER_STRAY);
	CSR_WRITE_2(sc, RL_IMR, 0);

	taskqueue_enqueue_fast(taskqueue_fast, &sc->rl_inttask);

	return (FILTER_HANDLED);
}

static void
re_int_task(arg, npending)
	void			*arg;
	int			npending;
{
	struct rl_softc		*sc;
	struct ifnet		*ifp;
	u_int16_t		status;
	int			rval = 0;

	sc = arg;
	ifp = sc->rl_ifp;

	RL_LOCK(sc);

	status = CSR_READ_2(sc, RL_ISR);
        CSR_WRITE_2(sc, RL_ISR, status);

	if (sc->suspended ||
	    (ifp->if_drv_flags & IFF_DRV_RUNNING) == 0) {
		RL_UNLOCK(sc);
		return;
	}

#ifdef DEVICE_POLLING
	if  (ifp->if_capenable & IFCAP_POLLING) {
		RL_UNLOCK(sc);
		return;
	}
#endif

	if (status & (RL_ISR_RX_OK|RL_ISR_RX_ERR|RL_ISR_FIFO_OFLOW))
		rval = re_rxeof(sc);

#ifdef RE_TX_MODERATION
	if (status & (RL_ISR_TIMEOUT_EXPIRED|
#else
	if (status & (RL_ISR_TX_OK|
#endif
	    RL_ISR_TX_ERR|RL_ISR_TX_DESC_UNAVAIL))
		re_txeof(sc);

	if (status & RL_ISR_SYSTEM_ERR) {
		re_reset(sc);
		re_init_locked(sc);
	}

	if (status & RL_ISR_LINKCHG) {
		callout_stop(&sc->rl_stat_callout);
		re_tick(sc);
	}

	if (!IFQ_DRV_IS_EMPTY(&ifp->if_snd))
		taskqueue_enqueue_fast(taskqueue_fast, &sc->rl_txtask);

	RL_UNLOCK(sc);

        if ((CSR_READ_2(sc, RL_ISR) & RL_INTRS_CPLUS) || rval) {
		taskqueue_enqueue_fast(taskqueue_fast, &sc->rl_inttask);
		return;
	}

	CSR_WRITE_2(sc, RL_IMR, RL_INTRS_CPLUS);

	return;
}

static int
re_encap(sc, m_head)
	struct rl_softc		*sc;
	struct mbuf		**m_head;
{
	struct rl_txdesc	*txd, *txd_last;
	bus_dma_segment_t	segs[RL_NTXSEGS];
	bus_dmamap_t		map;
	struct mbuf		*m_new;
	struct rl_desc		*desc;
	int			nsegs, prod;
	int			i, error, ei, si;
	int			padlen;
	uint32_t		cmdstat, csum_flags, vlanctl;

	RL_LOCK_ASSERT(sc);
	M_ASSERTPKTHDR((*m_head));

	/*
	 * With some of the RealTek chips, using the checksum offload
	 * support in conjunction with the autopadding feature results
	 * in the transmission of corrupt frames. For example, if we
	 * need to send a really small IP fragment that's less than 60
	 * bytes in size, and IP header checksumming is enabled, the
	 * resulting ethernet frame that appears on the wire will
	 * have garbled payload. To work around this, if TX IP checksum
	 * offload is enabled, we always manually pad short frames out
	 * to the minimum ethernet frame size.
	 */
	if ((*m_head)->m_pkthdr.len < RL_IP4CSUMTX_PADLEN &&
	    ((*m_head)->m_pkthdr.csum_flags & CSUM_IP) != 0) {
		padlen = RL_MIN_FRAMELEN - (*m_head)->m_pkthdr.len;
		if (M_WRITABLE(*m_head) == 0) {
			/* Get a writable copy. */
			m_new = m_dup(*m_head, M_DONTWAIT);
			m_freem(*m_head);
			if (m_new == NULL) {
				*m_head = NULL;
				return (ENOBUFS);
			}
			*m_head = m_new;
		}
		if ((*m_head)->m_next != NULL ||
		    M_TRAILINGSPACE(*m_head) < padlen) {
			m_new = m_defrag(*m_head, M_DONTWAIT);
			if (m_new == NULL) {
				m_freem(*m_head);
				*m_head = NULL;
				return (ENOBUFS);
			}
		} else
			m_new = *m_head;

		/*
		 * Manually pad short frames, and zero the pad space
		 * to avoid leaking data.
		 */
		bzero(mtod(m_new, char *) + m_new->m_pkthdr.len, padlen);
		m_new->m_pkthdr.len += padlen;
		m_new->m_len = m_new->m_pkthdr.len;
		*m_head = m_new;
	}

	prod = sc->rl_ldata.rl_tx_prodidx;
	txd = &sc->rl_ldata.rl_tx_desc[prod];
	error = bus_dmamap_load_mbuf_sg(sc->rl_ldata.rl_tx_mtag, txd->tx_dmamap,
	    *m_head, segs, &nsegs, BUS_DMA_NOWAIT);
	if (error == EFBIG) {
		m_new = m_collapse(*m_head, M_DONTWAIT, RL_NTXSEGS);
		if (m_new == NULL) {
			m_freem(*m_head);
			*m_head = NULL;
			return (ENOBUFS);
		}
		*m_head = m_new;
		error = bus_dmamap_load_mbuf_sg(sc->rl_ldata.rl_tx_mtag,
		    txd->tx_dmamap, *m_head, segs, &nsegs, BUS_DMA_NOWAIT);
		if (error != 0) {
			m_freem(*m_head);
			*m_head = NULL;
			return (error);
		}
	} else if (error != 0)
		return (error);
	if (nsegs == 0) {
		m_freem(*m_head);
		*m_head = NULL;
		return (EIO);
	}

	/* Check for number of available descriptors. */
	if (sc->rl_ldata.rl_tx_free - nsegs <= 1) {
		bus_dmamap_unload(sc->rl_ldata.rl_tx_mtag, txd->tx_dmamap);
		return (ENOBUFS);
	}

	bus_dmamap_sync(sc->rl_ldata.rl_tx_mtag, txd->tx_dmamap,
	    BUS_DMASYNC_PREWRITE);

	/*
	 * Set up checksum offload. Note: checksum offload bits must
	 * appear in all descriptors of a multi-descriptor transmit
	 * attempt. This is according to testing done with an 8169
	 * chip. This is a requirement.
	 */
	csum_flags = 0;
	if (((*m_head)->m_pkthdr.csum_flags & CSUM_TSO) != 0)
		csum_flags = RL_TDESC_CMD_LGSEND |
		    ((uint32_t)(*m_head)->m_pkthdr.tso_segsz <<
		    RL_TDESC_CMD_MSSVAL_SHIFT);
	else {
		/*
		 * Unconditionally enable IP checksum if TCP or UDP
		 * checksum is required. Otherwise, TCP/UDP checksum
		 * does't make effects.
		 */
		if (((*m_head)->m_pkthdr.csum_flags & RE_CSUM_FEATURES) != 0) {
			csum_flags |= RL_TDESC_CMD_IPCSUM;
			if (((*m_head)->m_pkthdr.csum_flags & CSUM_TCP) != 0)
				csum_flags |= RL_TDESC_CMD_TCPCSUM;
			if (((*m_head)->m_pkthdr.csum_flags & CSUM_UDP) != 0)
				csum_flags |= RL_TDESC_CMD_UDPCSUM;
		}
	}

	/*
	 * Set up hardware VLAN tagging. Note: vlan tag info must
	 * appear in all descriptors of a multi-descriptor
	 * transmission attempt.
	 */
	vlanctl = 0;
	if ((*m_head)->m_flags & M_VLANTAG)
		vlanctl =
		    htole32(htons((*m_head)->m_pkthdr.ether_vtag) |
		    RL_TDESC_VLANCTL_TAG);

	si = prod;
	for (i = 0; i < nsegs; i++, prod = RL_TX_DESC_NXT(sc, prod)) {
		desc = &sc->rl_ldata.rl_tx_list[prod];
		desc->rl_vlanctl = vlanctl;
		desc->rl_bufaddr_lo = htole32(RL_ADDR_LO(segs[i].ds_addr));
		desc->rl_bufaddr_hi = htole32(RL_ADDR_HI(segs[i].ds_addr));
		cmdstat = segs[i].ds_len;
		if (i != 0)
			cmdstat |= RL_TDESC_CMD_OWN;
		if (prod == sc->rl_ldata.rl_tx_desc_cnt - 1)
			cmdstat |= RL_TDESC_CMD_EOR;
		desc->rl_cmdstat = htole32(cmdstat | csum_flags);
		sc->rl_ldata.rl_tx_free--;
	}
	/* Update producer index. */
	sc->rl_ldata.rl_tx_prodidx = prod;

	/* Set EOF on the last descriptor. */
	ei = RL_TX_DESC_PRV(sc, prod);
	desc = &sc->rl_ldata.rl_tx_list[ei];
	desc->rl_cmdstat |= htole32(RL_TDESC_CMD_EOF);

	desc = &sc->rl_ldata.rl_tx_list[si];
	/* Set SOF and transfer ownership of packet to the chip. */
	desc->rl_cmdstat |= htole32(RL_TDESC_CMD_OWN | RL_TDESC_CMD_SOF);

	/*
	 * Insure that the map for this transmission
	 * is placed at the array index of the last descriptor
	 * in this chain.  (Swap last and first dmamaps.)
	 */
	txd_last = &sc->rl_ldata.rl_tx_desc[ei];
	map = txd->tx_dmamap;
	txd->tx_dmamap = txd_last->tx_dmamap;
	txd_last->tx_dmamap = map;
	txd_last->tx_m = *m_head;

	return (0);
}

static void
re_tx_task(arg, npending)
	void			*arg;
	int			npending;
{
	struct ifnet		*ifp;

	ifp = arg;
	re_start(ifp);

	return;
}

/*
 * Main transmit routine for C+ and gigE NICs.
 */
static void
re_start(ifp)
	struct ifnet		*ifp;
{
	struct rl_softc		*sc;
	struct mbuf		*m_head;
	int			queued;

	sc = ifp->if_softc;

	RL_LOCK(sc);

	if ((ifp->if_drv_flags & (IFF_DRV_RUNNING | IFF_DRV_OACTIVE)) !=
	    IFF_DRV_RUNNING || sc->rl_link == 0) {
		RL_UNLOCK(sc);
		return;
	}

	for (queued = 0; !IFQ_DRV_IS_EMPTY(&ifp->if_snd) &&
	    sc->rl_ldata.rl_tx_free > 1;) {
		IFQ_DRV_DEQUEUE(&ifp->if_snd, m_head);
		if (m_head == NULL)
			break;

		if (re_encap(sc, &m_head) != 0) {
			if (m_head == NULL)
				break;
			IFQ_DRV_PREPEND(&ifp->if_snd, m_head);
			ifp->if_drv_flags |= IFF_DRV_OACTIVE;
			break;
		}

		/*
		 * If there's a BPF listener, bounce a copy of this frame
		 * to him.
		 */
		ETHER_BPF_MTAP(ifp, m_head);

		queued++;
	}

	if (queued == 0) {
#ifdef RE_TX_MODERATION
		if (sc->rl_ldata.rl_tx_free != sc->rl_ldata.rl_tx_desc_cnt)
			CSR_WRITE_4(sc, RL_TIMERCNT, 1);
#endif
		RL_UNLOCK(sc);
		return;
	}

	/* Flush the TX descriptors */

	bus_dmamap_sync(sc->rl_ldata.rl_tx_list_tag,
	    sc->rl_ldata.rl_tx_list_map,
	    BUS_DMASYNC_PREWRITE|BUS_DMASYNC_PREREAD);

	CSR_WRITE_1(sc, sc->rl_txstart, RL_TXSTART_START);

#ifdef RE_TX_MODERATION
	/*
	 * Use the countdown timer for interrupt moderation.
	 * 'TX done' interrupts are disabled. Instead, we reset the
	 * countdown timer, which will begin counting until it hits
	 * the value in the TIMERINT register, and then trigger an
	 * interrupt. Each time we write to the TIMERCNT register,
	 * the timer count is reset to 0.
	 */
	CSR_WRITE_4(sc, RL_TIMERCNT, 1);
#endif

	/*
	 * Set a timeout in case the chip goes out to lunch.
	 */
	sc->rl_watchdog_timer = 5;

	RL_UNLOCK(sc);

	return;
}

static void
re_init(xsc)
	void			*xsc;
{
	struct rl_softc		*sc = xsc;

	RL_LOCK(sc);
	re_init_locked(sc);
	RL_UNLOCK(sc);
}

static void
re_init_locked(sc)
	struct rl_softc		*sc;
{
	struct ifnet		*ifp = sc->rl_ifp;
	struct mii_data		*mii;
	u_int32_t		rxcfg = 0;
	uint16_t		cfg;
	union {
		uint32_t align_dummy;
		u_char eaddr[ETHER_ADDR_LEN];
        } eaddr;

	RL_LOCK_ASSERT(sc);

	mii = device_get_softc(sc->rl_miibus);

	/*
	 * Cancel pending I/O and free all RX/TX buffers.
	 */
	re_stop(sc);

	/*
	 * Enable C+ RX and TX mode, as well as VLAN stripping and
	 * RX checksum offload. We must configure the C+ register
	 * before all others.
	 */
	cfg = RL_CPLUSCMD_PCI_MRW;
	if ((ifp->if_capenable & IFCAP_RXCSUM) != 0)
		cfg |= RL_CPLUSCMD_RXCSUM_ENB;
	if ((ifp->if_capenable & IFCAP_VLAN_HWTAGGING) != 0)
		cfg |= RL_CPLUSCMD_VLANSTRIP;
	CSR_WRITE_2(sc, RL_CPLUS_CMD,
	    cfg | RL_CPLUSCMD_RXENB | RL_CPLUSCMD_TXENB);

	/*
	 * Init our MAC address.  Even though the chipset
	 * documentation doesn't mention it, we need to enter "Config
	 * register write enable" mode to modify the ID registers.
	 */
	/* Copy MAC address on stack to align. */
	bcopy(IF_LLADDR(ifp), eaddr.eaddr, ETHER_ADDR_LEN);
	CSR_WRITE_1(sc, RL_EECMD, RL_EEMODE_WRITECFG);
	CSR_WRITE_4(sc, RL_IDR0,
	    htole32(*(u_int32_t *)(&eaddr.eaddr[0])));
	CSR_WRITE_4(sc, RL_IDR4,
	    htole32(*(u_int32_t *)(&eaddr.eaddr[4])));
	CSR_WRITE_1(sc, RL_EECMD, RL_EEMODE_OFF);

	/*
	 * For C+ mode, initialize the RX descriptors and mbufs.
	 */
	re_rx_list_init(sc);
	re_tx_list_init(sc);

	/*
	 * Load the addresses of the RX and TX lists into the chip.
	 */

	CSR_WRITE_4(sc, RL_RXLIST_ADDR_HI,
	    RL_ADDR_HI(sc->rl_ldata.rl_rx_list_addr));
	CSR_WRITE_4(sc, RL_RXLIST_ADDR_LO,
	    RL_ADDR_LO(sc->rl_ldata.rl_rx_list_addr));

	CSR_WRITE_4(sc, RL_TXLIST_ADDR_HI,
	    RL_ADDR_HI(sc->rl_ldata.rl_tx_list_addr));
	CSR_WRITE_4(sc, RL_TXLIST_ADDR_LO,
	    RL_ADDR_LO(sc->rl_ldata.rl_tx_list_addr));

	/*
	 * Enable transmit and receive.
	 */
	CSR_WRITE_1(sc, RL_COMMAND, RL_CMD_TX_ENB|RL_CMD_RX_ENB);

	/*
	 * Set the initial TX and RX configuration.
	 */
	if (sc->rl_testmode) {
		if (sc->rl_type == RL_8169)
			CSR_WRITE_4(sc, RL_TXCFG,
			    RL_TXCFG_CONFIG|RL_LOOPTEST_ON);
		else
			CSR_WRITE_4(sc, RL_TXCFG,
			    RL_TXCFG_CONFIG|RL_LOOPTEST_ON_CPLUS);
	} else
		CSR_WRITE_4(sc, RL_TXCFG, RL_TXCFG_CONFIG);

	CSR_WRITE_1(sc, RL_EARLY_TX_THRESH, 16);

	CSR_WRITE_4(sc, RL_RXCFG, RL_RXCFG_CONFIG);

	/* Set the individual bit to receive frames for this host only. */
	rxcfg = CSR_READ_4(sc, RL_RXCFG);
	rxcfg |= RL_RXCFG_RX_INDIV;

	/* If we want promiscuous mode, set the allframes bit. */
	if (ifp->if_flags & IFF_PROMISC)
		rxcfg |= RL_RXCFG_RX_ALLPHYS;
	else
		rxcfg &= ~RL_RXCFG_RX_ALLPHYS;
	CSR_WRITE_4(sc, RL_RXCFG, rxcfg);

	/*
	 * Set capture broadcast bit to capture broadcast frames.
	 */
	if (ifp->if_flags & IFF_BROADCAST)
		rxcfg |= RL_RXCFG_RX_BROAD;
	else
		rxcfg &= ~RL_RXCFG_RX_BROAD;
	CSR_WRITE_4(sc, RL_RXCFG, rxcfg);

	/*
	 * Program the multicast filter, if necessary.
	 */
	re_setmulti(sc);

#ifdef DEVICE_POLLING
	/*
	 * Disable interrupts if we are polling.
	 */
	if (ifp->if_capenable & IFCAP_POLLING)
		CSR_WRITE_2(sc, RL_IMR, 0);
	else	/* otherwise ... */
#endif

	/*
	 * Enable interrupts.
	 */
	if (sc->rl_testmode)
		CSR_WRITE_2(sc, RL_IMR, 0);
	else
		CSR_WRITE_2(sc, RL_IMR, RL_INTRS_CPLUS);
	CSR_WRITE_2(sc, RL_ISR, RL_INTRS_CPLUS);

	/* Set initial TX threshold */
	sc->rl_txthresh = RL_TX_THRESH_INIT;

	/* Start RX/TX process. */
	CSR_WRITE_4(sc, RL_MISSEDPKT, 0);
#ifdef notdef
	/* Enable receiver and transmitter. */
	CSR_WRITE_1(sc, RL_COMMAND, RL_CMD_TX_ENB|RL_CMD_RX_ENB);
#endif

#ifdef RE_TX_MODERATION
	/*
	 * Initialize the timer interrupt register so that
	 * a timer interrupt will be generated once the timer
	 * reaches a certain number of ticks. The timer is
	 * reloaded on each transmit. This gives us TX interrupt
	 * moderation, which dramatically improves TX frame rate.
	 */
	if (sc->rl_type == RL_8169)
		CSR_WRITE_4(sc, RL_TIMERINT_8169, 0x800);
	else
		CSR_WRITE_4(sc, RL_TIMERINT, 0x400);
#endif

	/*
	 * For 8169 gigE NICs, set the max allowed RX packet
	 * size so we can receive jumbo frames.
	 */
	if (sc->rl_type == RL_8169)
		CSR_WRITE_2(sc, RL_MAXRXPKTLEN, 16383);

	if (sc->rl_testmode)
		return;

	mii_mediachg(mii);

	CSR_WRITE_1(sc, RL_CFG1, CSR_READ_1(sc, RL_CFG1) | RL_CFG1_DRVLOAD);

	ifp->if_drv_flags |= IFF_DRV_RUNNING;
	ifp->if_drv_flags &= ~IFF_DRV_OACTIVE;

	sc->rl_link = 0;
	sc->rl_watchdog_timer = 0;
	callout_reset(&sc->rl_stat_callout, hz, re_tick, sc);
}

/*
 * Set media options.
 */
static int
re_ifmedia_upd(ifp)
	struct ifnet		*ifp;
{
	struct rl_softc		*sc;
	struct mii_data		*mii;

	sc = ifp->if_softc;
	mii = device_get_softc(sc->rl_miibus);
	RL_LOCK(sc);
	mii_mediachg(mii);
	RL_UNLOCK(sc);

	return (0);
}

/*
 * Report current media status.
 */
static void
re_ifmedia_sts(ifp, ifmr)
	struct ifnet		*ifp;
	struct ifmediareq	*ifmr;
{
	struct rl_softc		*sc;
	struct mii_data		*mii;

	sc = ifp->if_softc;
	mii = device_get_softc(sc->rl_miibus);

	RL_LOCK(sc);
	mii_pollstat(mii);
	RL_UNLOCK(sc);
	ifmr->ifm_active = mii->mii_media_active;
	ifmr->ifm_status = mii->mii_media_status;
}

static int
re_ioctl(ifp, command, data)
	struct ifnet		*ifp;
	u_long			command;
	caddr_t			data;
{
	struct rl_softc		*sc = ifp->if_softc;
	struct ifreq		*ifr = (struct ifreq *) data;
	struct mii_data		*mii;
	int			error = 0;

	switch (command) {
	case SIOCSIFMTU:
		if (ifr->ifr_mtu < ETHERMIN || ifr->ifr_mtu > RL_JUMBO_MTU) {
			error = EINVAL;
			break;
		}
		if (sc->rl_type == RL_8139CPLUS &&
		    ifr->ifr_mtu > RL_MAX_FRAMELEN) {
			error = EINVAL;
			break;
		}
		RL_LOCK(sc);
		if (ifp->if_mtu != ifr->ifr_mtu)
			ifp->if_mtu = ifr->ifr_mtu;
		RL_UNLOCK(sc);
		break;
	case SIOCSIFFLAGS:
		RL_LOCK(sc);
		if ((ifp->if_flags & IFF_UP) != 0) {
			if ((ifp->if_drv_flags & IFF_DRV_RUNNING) != 0) {
				if (((ifp->if_flags ^ sc->rl_if_flags)
				    & IFF_PROMISC) != 0)
					re_setmulti(sc);
			} else
				re_init_locked(sc);
		} else {
			if ((ifp->if_drv_flags & IFF_DRV_RUNNING) != 0)
				re_stop(sc);
		}
		sc->rl_if_flags = ifp->if_flags;
		RL_UNLOCK(sc);
		break;
	case SIOCADDMULTI:
	case SIOCDELMULTI:
		RL_LOCK(sc);
		re_setmulti(sc);
		RL_UNLOCK(sc);
		break;
	case SIOCGIFMEDIA:
	case SIOCSIFMEDIA:
		mii = device_get_softc(sc->rl_miibus);
		error = ifmedia_ioctl(ifp, ifr, &mii->mii_media, command);
		break;
	case SIOCSIFCAP:
	    {
		int mask, reinit;

		mask = ifr->ifr_reqcap ^ ifp->if_capenable;
		reinit = 0;
#ifdef DEVICE_POLLING
		if (mask & IFCAP_POLLING) {
			if (ifr->ifr_reqcap & IFCAP_POLLING) {
				error = ether_poll_register(re_poll, ifp);
				if (error)
					return(error);
				RL_LOCK(sc);
				/* Disable interrupts */
				CSR_WRITE_2(sc, RL_IMR, 0x0000);
				ifp->if_capenable |= IFCAP_POLLING;
				RL_UNLOCK(sc);
			} else {
				error = ether_poll_deregister(ifp);
				/* Enable interrupts. */
				RL_LOCK(sc);
				CSR_WRITE_2(sc, RL_IMR, RL_INTRS_CPLUS);
				ifp->if_capenable &= ~IFCAP_POLLING;
				RL_UNLOCK(sc);
			}
		}
#endif /* DEVICE_POLLING */
		if (mask & IFCAP_HWCSUM) {
			ifp->if_capenable ^= IFCAP_HWCSUM;
			if (ifp->if_capenable & IFCAP_TXCSUM)
				ifp->if_hwassist |= RE_CSUM_FEATURES;
			else
				ifp->if_hwassist &= ~RE_CSUM_FEATURES;
			reinit = 1;
		}
		if (mask & IFCAP_VLAN_HWTAGGING) {
			ifp->if_capenable ^= IFCAP_VLAN_HWTAGGING;
			reinit = 1;
		}
		if (mask & IFCAP_TSO4) {
			ifp->if_capenable ^= IFCAP_TSO4;
			if ((IFCAP_TSO4 & ifp->if_capenable) &&
			    (IFCAP_TSO4 & ifp->if_capabilities))
				ifp->if_hwassist |= CSUM_TSO;
			else
				ifp->if_hwassist &= ~CSUM_TSO;
		}
		if ((mask & IFCAP_WOL) != 0 &&
		    (ifp->if_capabilities & IFCAP_WOL) != 0) {
			if ((mask & IFCAP_WOL_UCAST) != 0)
				ifp->if_capenable ^= IFCAP_WOL_UCAST;
			if ((mask & IFCAP_WOL_MCAST) != 0)
				ifp->if_capenable ^= IFCAP_WOL_MCAST;
			if ((mask & IFCAP_WOL_MAGIC) != 0)
				ifp->if_capenable ^= IFCAP_WOL_MAGIC;
		}
		if (reinit && ifp->if_drv_flags & IFF_DRV_RUNNING)
			re_init(sc);
		VLAN_CAPABILITIES(ifp);
	    }
		break;
	default:
		error = ether_ioctl(ifp, command, data);
		break;
	}

	return (error);
}

static void
re_watchdog(sc)
	struct rl_softc		*sc;
{

	RL_LOCK_ASSERT(sc);

	if (sc->rl_watchdog_timer == 0 || --sc->rl_watchdog_timer != 0)
		return;

	device_printf(sc->rl_dev, "watchdog timeout\n");
	sc->rl_ifp->if_oerrors++;

	re_txeof(sc);
	re_rxeof(sc);
	re_init_locked(sc);
}

/*
 * Stop the adapter and free any mbufs allocated to the
 * RX and TX lists.
 */
static void
re_stop(sc)
	struct rl_softc		*sc;
{
	register int		i;
	struct ifnet		*ifp;
	struct rl_txdesc	*txd;
	struct rl_rxdesc	*rxd;

	RL_LOCK_ASSERT(sc);

	ifp = sc->rl_ifp;

	sc->rl_watchdog_timer = 0;
	callout_stop(&sc->rl_stat_callout);
	ifp->if_drv_flags &= ~(IFF_DRV_RUNNING | IFF_DRV_OACTIVE);

	CSR_WRITE_1(sc, RL_COMMAND, 0x00);
	CSR_WRITE_2(sc, RL_IMR, 0x0000);
	CSR_WRITE_2(sc, RL_ISR, 0xFFFF);

	if (sc->rl_head != NULL) {
		m_freem(sc->rl_head);
		sc->rl_head = sc->rl_tail = NULL;
	}

	/* Free the TX list buffers. */

	for (i = 0; i < sc->rl_ldata.rl_tx_desc_cnt; i++) {
		txd = &sc->rl_ldata.rl_tx_desc[i];
		if (txd->tx_m != NULL) {
			bus_dmamap_sync(sc->rl_ldata.rl_tx_mtag,
			    txd->tx_dmamap, BUS_DMASYNC_POSTWRITE);
			bus_dmamap_unload(sc->rl_ldata.rl_tx_mtag,
			    txd->tx_dmamap);
			m_freem(txd->tx_m);
			txd->tx_m = NULL;
		}
	}

	/* Free the RX list buffers. */

	for (i = 0; i < sc->rl_ldata.rl_rx_desc_cnt; i++) {
		rxd = &sc->rl_ldata.rl_rx_desc[i];
		if (rxd->rx_m != NULL) {
			bus_dmamap_sync(sc->rl_ldata.rl_tx_mtag,
			    rxd->rx_dmamap, BUS_DMASYNC_POSTREAD);
			bus_dmamap_unload(sc->rl_ldata.rl_rx_mtag,
			    rxd->rx_dmamap);
			m_freem(rxd->rx_m);
			rxd->rx_m = NULL;
		}
	}
}

/*
 * Device suspend routine.  Stop the interface and save some PCI
 * settings in case the BIOS doesn't restore them properly on
 * resume.
 */
static int
re_suspend(dev)
	device_t		dev;
{
	struct rl_softc		*sc;

	sc = device_get_softc(dev);

	RL_LOCK(sc);
	re_stop(sc);
	re_setwol(sc);
	sc->suspended = 1;
	RL_UNLOCK(sc);

	return (0);
}

/*
 * Device resume routine.  Restore some PCI settings in case the BIOS
 * doesn't, re-enable busmastering, and restart the interface if
 * appropriate.
 */
static int
re_resume(dev)
	device_t		dev;
{
	struct rl_softc		*sc;
	struct ifnet		*ifp;

	sc = device_get_softc(dev);

	RL_LOCK(sc);

	ifp = sc->rl_ifp;

	/* reinitialize interface if necessary */
	if (ifp->if_flags & IFF_UP)
		re_init_locked(sc);

	/*
	 * Clear WOL matching such that normal Rx filtering
	 * wouldn't interfere with WOL patterns.
	 */
	re_clrwol(sc);
	sc->suspended = 0;
	RL_UNLOCK(sc);

	return (0);
}

/*
 * Stop all chip I/O so that the kernel's probe routines don't
 * get confused by errant DMAs when rebooting.
 */
static int
re_shutdown(dev)
	device_t		dev;
{
	struct rl_softc		*sc;

	sc = device_get_softc(dev);

	RL_LOCK(sc);
	re_stop(sc);
	/*
	 * Mark interface as down since otherwise we will panic if
	 * interrupt comes in later on, which can happen in some
	 * cases.
	 */
	sc->rl_ifp->if_flags &= ~IFF_UP;
	re_setwol(sc);
	RL_UNLOCK(sc);

	return (0);
}

static void
re_setwol(sc)
	struct rl_softc		*sc;
{
	struct ifnet		*ifp;
	int			pmc;
	uint16_t		pmstat;
	uint8_t			v;

	RL_LOCK_ASSERT(sc);

	if (pci_find_extcap(sc->rl_dev, PCIY_PMG, &pmc) != 0)
		return;

	ifp = sc->rl_ifp;
	/* Enable config register write. */
	CSR_WRITE_1(sc, RL_EECMD, RL_EE_MODE);

	/* Enable PME. */
	v = CSR_READ_1(sc, RL_CFG1);
	v &= ~RL_CFG1_PME;
	if ((ifp->if_capenable & IFCAP_WOL) != 0)
		v |= RL_CFG1_PME;
	CSR_WRITE_1(sc, RL_CFG1, v);

	v = CSR_READ_1(sc, RL_CFG3);
	v &= ~(RL_CFG3_WOL_LINK | RL_CFG3_WOL_MAGIC);
	if ((ifp->if_capenable & IFCAP_WOL_MAGIC) != 0)
		v |= RL_CFG3_WOL_MAGIC;
	CSR_WRITE_1(sc, RL_CFG3, v);

	/* Config register write done. */
	CSR_WRITE_1(sc, RL_EECMD, 0);

	v = CSR_READ_1(sc, RL_CFG5);
	v &= ~(RL_CFG5_WOL_BCAST | RL_CFG5_WOL_MCAST | RL_CFG5_WOL_UCAST);
	v &= ~RL_CFG5_WOL_LANWAKE;
	if ((ifp->if_capenable & IFCAP_WOL_UCAST) != 0)
		v |= RL_CFG5_WOL_UCAST;
	if ((ifp->if_capenable & IFCAP_WOL_MCAST) != 0)
		v |= RL_CFG5_WOL_MCAST | RL_CFG5_WOL_BCAST;
	if ((ifp->if_capenable & IFCAP_WOL) != 0)
		v |= RL_CFG5_WOL_LANWAKE;
	CSR_WRITE_1(sc, RL_CFG5, v);

	/*
	 * It seems that hardware resets its link speed to 100Mbps in
	 * power down mode so switching to 100Mbps in driver is not
	 * needed.
	 */

	/* Request PME if WOL is requested. */
	pmstat = pci_read_config(sc->rl_dev, pmc + PCIR_POWER_STATUS, 2);
	pmstat &= ~(PCIM_PSTAT_PME | PCIM_PSTAT_PMEENABLE);
	if ((ifp->if_capenable & IFCAP_WOL) != 0)
		pmstat |= PCIM_PSTAT_PME | PCIM_PSTAT_PMEENABLE;
	pci_write_config(sc->rl_dev, pmc + PCIR_POWER_STATUS, pmstat, 2);
}

static void
re_clrwol(sc)
	struct rl_softc		*sc;
{
	int			pmc;
	uint8_t			v;

	RL_LOCK_ASSERT(sc);

	if (pci_find_extcap(sc->rl_dev, PCIY_PMG, &pmc) != 0)
		return;

	/* Enable config register write. */
	CSR_WRITE_1(sc, RL_EECMD, RL_EE_MODE);

	v = CSR_READ_1(sc, RL_CFG3);
	v &= ~(RL_CFG3_WOL_LINK | RL_CFG3_WOL_MAGIC);
	CSR_WRITE_1(sc, RL_CFG3, v);

	/* Config register write done. */
	CSR_WRITE_1(sc, RL_EECMD, 0);

	v = CSR_READ_1(sc, RL_CFG5);
	v &= ~(RL_CFG5_WOL_BCAST | RL_CFG5_WOL_MCAST | RL_CFG5_WOL_UCAST);
	v &= ~RL_CFG5_WOL_LANWAKE;
	CSR_WRITE_1(sc, RL_CFG5, v);
}
OpenPOWER on IntegriCloud