Free Electrons

Embedded Linux Experts

   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
/* Copyright (c) 2012 - 2015 UNISYS CORPORATION
 * All rights reserved.
 *
 * This program is free software; you can redistribute it and/or modify it
 * under the terms and conditions of the GNU General Public License,
 * version 2, as published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful, but
 * WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
 * NON INFRINGEMENT.  See the GNU General Public License for more
 * details.
 */

/* This driver lives in a spar partition, and registers to ethernet io
 * channels from the visorbus driver. It creates netdev devices and
 * forwards transmit to the IO channel and accepts rcvs from the IO
 * Partition via the IO channel.
 */

#include <linux/debugfs.h>
#include <linux/etherdevice.h>
#include <linux/netdevice.h>
#include <linux/kthread.h>
#include <linux/skbuff.h>
#include <linux/rtnetlink.h>

#include "visorbus.h"
#include "iochannel.h"

#define VISORNIC_INFINITE_RSP_WAIT 0

/* MAX_BUF = 64 lines x 32 MAXVNIC x 80 characters
 *         = 163840 bytes
 */
#define MAX_BUF 163840
#define NAPI_WEIGHT 64

/* GUIDS for director channel type supported by this driver.  */
static struct visor_channeltype_descriptor visornic_channel_types[] = {
	/* Note that the only channel type we expect to be reported by the
	 * bus driver is the VISOR_VNIC channel.
	 */
	{ VISOR_VNIC_CHANNEL_UUID, "ultravnic" },
	{ NULL_UUID_LE, NULL }
};
MODULE_DEVICE_TABLE(visorbus, visornic_channel_types);
/*
 * FIXME XXX: This next line of code must be fixed and removed before
 * acceptance into the 'normal' part of the kernel.  It is only here as a place
 * holder to get module autoloading functionality working for visorbus.  Code
 * must be added to scripts/mode/file2alias.c, etc., to get this working
 * properly.
 */
MODULE_ALIAS("visorbus:" VISOR_VNIC_CHANNEL_UUID_STR);

struct chanstat {
	unsigned long got_rcv;
	unsigned long got_enbdisack;
	unsigned long got_xmit_done;
	unsigned long xmit_fail;
	unsigned long sent_enbdis;
	unsigned long sent_promisc;
	unsigned long sent_post;
	unsigned long sent_post_failed;
	unsigned long sent_xmit;
	unsigned long reject_count;
	unsigned long extra_rcvbufs_sent;
};

struct visornic_devdata {
	/* 0 disabled 1 enabled to receive */
	unsigned short enabled;
	/* NET_RCV_ENABLE/DISABLE acked by IOPART */
	unsigned short enab_dis_acked;

	struct visor_device *dev;
	struct net_device *netdev;
	struct net_device_stats net_stats;
	atomic_t interrupt_rcvd;
	wait_queue_head_t rsp_queue;
	struct sk_buff **rcvbuf;
	/* incarnation_id lets IOPART know about re-birth */
	u64 incarnation_id;
	/* flags as they were prior to set_multicast_list */
	unsigned short old_flags;
	atomic_t usage;	/* count of users */

	/* number of rcv buffers the vnic will post */
	int num_rcv_bufs;
	int num_rcv_bufs_could_not_alloc;
	atomic_t num_rcvbuf_in_iovm;
	unsigned long alloc_failed_in_if_needed_cnt;
	unsigned long alloc_failed_in_repost_rtn_cnt;

	/* absolute max number of outstanding xmits - should never hit this */
	unsigned long max_outstanding_net_xmits;
	/* high water mark for calling netif_stop_queue() */
	unsigned long upper_threshold_net_xmits;
	/* high water mark for calling netif_wake_queue() */
	unsigned long lower_threshold_net_xmits;
	/* xmitbufhead - head of the xmit buffer list sent to the IOPART end */
	struct sk_buff_head xmitbufhead;

	visorbus_state_complete_func server_down_complete_func;
	struct work_struct timeout_reset;
	/* cmdrsp_rcv is used for posting/unposting rcv buffers  */
	struct uiscmdrsp *cmdrsp_rcv;
	/* xmit_cmdrsp - issues NET_XMIT - only one active xmit at a time */
	struct uiscmdrsp *xmit_cmdrsp;

	bool server_down;		 /* IOPART is down */
	bool server_change_state;	 /* Processing SERVER_CHANGESTATE msg */
	bool going_away;		 /* device is being torn down */
	struct dentry *eth_debugfs_dir;
	u64 interrupts_rcvd;
	u64 interrupts_notme;
	u64 interrupts_disabled;
	u64 busy_cnt;
	spinlock_t priv_lock;  /* spinlock to access devdata structures */

	/* flow control counter */
	u64 flow_control_upper_hits;
	u64 flow_control_lower_hits;

	/* debug counters */
	unsigned long n_rcv0;			/* # rcvs of 0 buffers */
	unsigned long n_rcv1;			/* # rcvs of 1 buffers */
	unsigned long n_rcv2;			/* # rcvs of 2 buffers */
	unsigned long n_rcvx;			/* # rcvs of >2 buffers */
	unsigned long found_repost_rcvbuf_cnt;	/* # repost_rcvbuf_cnt */
	unsigned long repost_found_skb_cnt;	/* # of found the skb */
	unsigned long n_repost_deficit;		/* # of lost rcv buffers */
	unsigned long bad_rcv_buf; /* # of unknown rcv skb  not freed */
	unsigned long n_rcv_packets_not_accepted;/* # bogs rcv packets */

	int queuefullmsg_logged;
	struct chanstat chstat;
	struct timer_list irq_poll_timer;
	struct napi_struct napi;
	struct uiscmdrsp cmdrsp[SIZEOF_CMDRSP];
};

/* Returns next non-zero index on success or 0 on failure (i.e. out of room). */
static u16
add_physinfo_entries(u64 inp_pfn, u16 inp_off, u32 inp_len, u16 index,
		     u16 max_pi_arr_entries, struct phys_info pi_arr[])
{
	u32 len;
	u16 i, firstlen;

	firstlen = PI_PAGE_SIZE - inp_off;
	if (inp_len <= firstlen) {
		/* The input entry spans only one page - add as is. */
		if (index >= max_pi_arr_entries)
			return 0;
		pi_arr[index].pi_pfn = inp_pfn;
		pi_arr[index].pi_off = (u16)inp_off;
		pi_arr[index].pi_len = (u16)inp_len;
		return index + 1;
	}

	/* This entry spans multiple pages. */
	for (len = inp_len, i = 0; len;
		len -= pi_arr[index + i].pi_len, i++) {
		if (index + i >= max_pi_arr_entries)
			return 0;
		pi_arr[index + i].pi_pfn = inp_pfn + i;
		if (i == 0) {
			pi_arr[index].pi_off = inp_off;
			pi_arr[index].pi_len = firstlen;
		} else {
			pi_arr[index + i].pi_off = 0;
			pi_arr[index + i].pi_len =
			    (u16)MINNUM(len, (u32)PI_PAGE_SIZE);
		}
	}
	return index + i;
}

/*
 *	visor_copy_fragsinfo_from_skb(
 *	@skb_in: skbuff that we are pulling the frags from
 *	@firstfraglen: length of first fragment in skb
 *	@frags_max: max len of frags array
 *	@frags: frags array filled in on output
 *
 *	Copy the fragment list in the SKB to a phys_info
 *	array that the IOPART understands.
 *	Return value indicates number of entries filled in frags
 *	Negative values indicate an error.
 */
static int
visor_copy_fragsinfo_from_skb(struct sk_buff *skb, unsigned int firstfraglen,
			      unsigned int frags_max,
			      struct phys_info frags[])
{
	unsigned int count = 0, frag, size, offset = 0, numfrags;
	unsigned int total_count;

	numfrags = skb_shinfo(skb)->nr_frags;

	/* Compute the number of fragments this skb has, and if its more than
	 * frag array can hold, linearize the skb
	 */
	total_count = numfrags + (firstfraglen / PI_PAGE_SIZE);
	if (firstfraglen % PI_PAGE_SIZE)
		total_count++;

	if (total_count > frags_max) {
		if (skb_linearize(skb))
			return -EINVAL;
		numfrags = skb_shinfo(skb)->nr_frags;
		firstfraglen = 0;
	}

	while (firstfraglen) {
		if (count == frags_max)
			return -EINVAL;

		frags[count].pi_pfn =
			page_to_pfn(virt_to_page(skb->data + offset));
		frags[count].pi_off =
			(unsigned long)(skb->data + offset) & PI_PAGE_MASK;
		size = min_t(unsigned int, firstfraglen,
			     PI_PAGE_SIZE - frags[count].pi_off);

		/* can take smallest of firstfraglen (what's left) OR
		 * bytes left in the page
		 */
		frags[count].pi_len = size;
		firstfraglen -= size;
		offset += size;
		count++;
	}
	if (numfrags) {
		if ((count + numfrags) > frags_max)
			return -EINVAL;

		for (frag = 0; frag < numfrags; frag++) {
			count = add_physinfo_entries(page_to_pfn(
				skb_frag_page(&skb_shinfo(skb)->frags[frag])),
					      skb_shinfo(skb)->frags[frag].
					      page_offset,
					      skb_shinfo(skb)->frags[frag].
					      size, count, frags_max, frags);
			/* add_physinfo_entries only returns
			 * zero if the frags array is out of room
			 * That should never happen because we
			 * fail above, if count+numfrags > frags_max.
			 */
			if (!count)
				return -EINVAL;
		}
	}
	if (skb_shinfo(skb)->frag_list) {
		struct sk_buff *skbinlist;
		int c;

		for (skbinlist = skb_shinfo(skb)->frag_list; skbinlist;
		     skbinlist = skbinlist->next) {
			c = visor_copy_fragsinfo_from_skb(skbinlist,
							  skbinlist->len -
							  skbinlist->data_len,
							  frags_max - count,
							  &frags[count]);
			if (c < 0)
				return c;
			count += c;
		}
	}
	return count;
}

static ssize_t enable_ints_write(struct file *file,
				 const char __user *buffer,
				 size_t count, loff_t *ppos)
{
	/* Don't want to break ABI here by having a debugfs
	 * file that no longer exists or is writable, so
	 * lets just make this a vestigual function
	 */
	return count;
}

static const struct file_operations debugfs_enable_ints_fops = {
	.write = enable_ints_write,
};

/*
 *	visornic_serverdown_complete - IOPART went down, pause device
 *	@work: Work queue it was scheduled on
 *
 *	The IO partition has gone down and we need to do some cleanup
 *	for when it comes back. Treat the IO partition as the link
 *	being down.
 *	Returns void.
 */
static void
visornic_serverdown_complete(struct visornic_devdata *devdata)
{
	struct net_device *netdev;

	netdev = devdata->netdev;

	/* Stop polling for interrupts */
	del_timer_sync(&devdata->irq_poll_timer);

	rtnl_lock();
	dev_close(netdev);
	rtnl_unlock();

	atomic_set(&devdata->num_rcvbuf_in_iovm, 0);
	devdata->chstat.sent_xmit = 0;
	devdata->chstat.got_xmit_done = 0;

	if (devdata->server_down_complete_func)
		(*devdata->server_down_complete_func)(devdata->dev, 0);

	devdata->server_down = true;
	devdata->server_change_state = false;
	devdata->server_down_complete_func = NULL;
}

/*
 *	visornic_serverdown - Command has notified us that IOPART is down
 *	@devdata: device that is being managed by IOPART
 *
 *	Schedule the work needed to handle the server down request. Make
 *	sure we haven't already handled the server change state event.
 *	Returns 0 if we scheduled the work, -EINVAL on error.
 */
static int
visornic_serverdown(struct visornic_devdata *devdata,
		    visorbus_state_complete_func complete_func)
{
	unsigned long flags;
	int err;

	spin_lock_irqsave(&devdata->priv_lock, flags);
	if (devdata->server_change_state) {
		dev_dbg(&devdata->dev->device, "%s changing state\n",
			__func__);
		err = -EINVAL;
		goto err_unlock;
	}
	if (devdata->server_down) {
		dev_dbg(&devdata->dev->device, "%s already down\n",
			__func__);
		err = -EINVAL;
		goto err_unlock;
	}
	if (devdata->going_away) {
		dev_dbg(&devdata->dev->device,
			"%s aborting because device removal pending\n",
			__func__);
		err = -ENODEV;
		goto err_unlock;
	}
	devdata->server_change_state = true;
	devdata->server_down_complete_func = complete_func;
	spin_unlock_irqrestore(&devdata->priv_lock, flags);

	visornic_serverdown_complete(devdata);
	return 0;

err_unlock:
	spin_unlock_irqrestore(&devdata->priv_lock, flags);
	return err;
}

/*
 *	alloc_rcv_buf	- alloc rcv buffer to be given to the IO Partition.
 *	@netdev: network adapter the rcv bufs are attached too.
 *
 *	Create an sk_buff (rcv_buf) that will be passed to the IO Partition
 *	so that it can write rcv data into our memory space.
 *	Return pointer to sk_buff
 */
static struct sk_buff *
alloc_rcv_buf(struct net_device *netdev)
{
	struct sk_buff *skb;

	/* NOTE: the first fragment in each rcv buffer is pointed to by
	 * rcvskb->data. For now all rcv buffers will be RCVPOST_BUF_SIZE
	 * in length, so the first frag is large enough to hold 1514.
	 */
	skb = alloc_skb(RCVPOST_BUF_SIZE, GFP_ATOMIC);
	if (!skb)
		return NULL;
	skb->dev = netdev;
	/* current value of mtu doesn't come into play here; large
	 * packets will just end up using multiple rcv buffers all of
	 * same size.
	 */
	skb->len = RCVPOST_BUF_SIZE;
	/* alloc_skb already zeroes it out for clarification. */
	skb->data_len = 0;
	return skb;
}

/*
 *	post_skb	- post a skb to the IO Partition.
 *	@cmdrsp: cmdrsp packet to be send to the IO Partition
 *	@devdata: visornic_devdata to post the skb too
 *	@skb: skb to give to the IO partition
 *
 *	Send the skb to the IO Partition.
 *	Returns 0 or error
 */
static int
post_skb(struct uiscmdrsp *cmdrsp,
	 struct visornic_devdata *devdata, struct sk_buff *skb)
{
	int err;

	cmdrsp->net.buf = skb;
	cmdrsp->net.rcvpost.frag.pi_pfn = page_to_pfn(virt_to_page(skb->data));
	cmdrsp->net.rcvpost.frag.pi_off =
		(unsigned long)skb->data & PI_PAGE_MASK;
	cmdrsp->net.rcvpost.frag.pi_len = skb->len;
	cmdrsp->net.rcvpost.unique_num = devdata->incarnation_id;

	if ((cmdrsp->net.rcvpost.frag.pi_off + skb->len) > PI_PAGE_SIZE)
		return -EINVAL;

	cmdrsp->net.type = NET_RCV_POST;
	cmdrsp->cmdtype = CMD_NET_TYPE;
	err = visorchannel_signalinsert(devdata->dev->visorchannel,
					IOCHAN_TO_IOPART,
					cmdrsp);
	if (err) {
		devdata->chstat.sent_post_failed++;
		return err;
	}

	atomic_inc(&devdata->num_rcvbuf_in_iovm);
	devdata->chstat.sent_post++;

	return 0;
}

/*
 *	send_enbdis	- send NET_RCV_ENBDIS to IO Partition
 *	@netdev: netdevice we are enable/disable, used as context
 *		 return value
 *	@state: enable = 1/disable = 0
 *	@devdata: visornic device we are enabling/disabling
 *
 *	Send the enable/disable message to the IO Partition.
 *	Returns 0 or error
 */
static int
send_enbdis(struct net_device *netdev, int state,
	    struct visornic_devdata *devdata)
{
	int err;

	devdata->cmdrsp_rcv->net.enbdis.enable = state;
	devdata->cmdrsp_rcv->net.enbdis.context = netdev;
	devdata->cmdrsp_rcv->net.type = NET_RCV_ENBDIS;
	devdata->cmdrsp_rcv->cmdtype = CMD_NET_TYPE;
	err = visorchannel_signalinsert(devdata->dev->visorchannel,
					IOCHAN_TO_IOPART,
					devdata->cmdrsp_rcv);
	if (err)
		return err;
	devdata->chstat.sent_enbdis++;
	return 0;
}

/*
 *	visornic_disable_with_timeout - Disable network adapter
 *	@netdev: netdevice to disable
 *	@timeout: timeout to wait for disable
 *
 *	Disable the network adapter and inform the IO Partition that we
 *	are disabled, reclaim memory from rcv bufs.
 *	Returns 0 on success, negative for failure of IO Partition
 *	responding.
 *
 */
static int
visornic_disable_with_timeout(struct net_device *netdev, const int timeout)
{
	struct visornic_devdata *devdata = netdev_priv(netdev);
	int i;
	unsigned long flags;
	int wait = 0;
	int err;

	/* send a msg telling the other end we are stopping incoming pkts */
	spin_lock_irqsave(&devdata->priv_lock, flags);
	devdata->enabled = 0;
	devdata->enab_dis_acked = 0; /* must wait for ack */
	spin_unlock_irqrestore(&devdata->priv_lock, flags);

	/* send disable and wait for ack -- don't hold lock when sending
	 * disable because if the queue is full, insert might sleep.
	 * If an error occurs, don't wait for the timeout.
	 */
	err = send_enbdis(netdev, 0, devdata);
	if (err)
		return err;

	/* wait for ack to arrive before we try to free rcv buffers
	 * NOTE: the other end automatically unposts the rcv buffers when
	 * when it gets a disable.
	 */
	spin_lock_irqsave(&devdata->priv_lock, flags);
	while ((timeout == VISORNIC_INFINITE_RSP_WAIT) ||
	       (wait < timeout)) {
		if (devdata->enab_dis_acked)
			break;
		if (devdata->server_down || devdata->server_change_state) {
			dev_dbg(&netdev->dev, "%s server went away\n",
				__func__);
			break;
		}
		set_current_state(TASK_INTERRUPTIBLE);
		spin_unlock_irqrestore(&devdata->priv_lock, flags);
		wait += schedule_timeout(msecs_to_jiffies(10));
		spin_lock_irqsave(&devdata->priv_lock, flags);
	}

	/* Wait for usage to go to 1 (no other users) before freeing
	 * rcv buffers
	 */
	if (atomic_read(&devdata->usage) > 1) {
		while (1) {
			set_current_state(TASK_INTERRUPTIBLE);
			spin_unlock_irqrestore(&devdata->priv_lock, flags);
			schedule_timeout(msecs_to_jiffies(10));
			spin_lock_irqsave(&devdata->priv_lock, flags);
			if (atomic_read(&devdata->usage))
				break;
		}
	}
	/* we've set enabled to 0, so we can give up the lock. */
	spin_unlock_irqrestore(&devdata->priv_lock, flags);

	/* stop the transmit queue so nothing more can be transmitted */
	netif_stop_queue(netdev);

	napi_disable(&devdata->napi);

	skb_queue_purge(&devdata->xmitbufhead);

	/* Free rcv buffers - other end has automatically unposed them on
	 * disable
	 */
	for (i = 0; i < devdata->num_rcv_bufs; i++) {
		if (devdata->rcvbuf[i]) {
			kfree_skb(devdata->rcvbuf[i]);
			devdata->rcvbuf[i] = NULL;
		}
	}

	return 0;
}

/*
 *	init_rcv_bufs  -- initialize receive bufs and send them to the IO Part
 *	@netdev: struct netdevice
 *	@devdata: visornic_devdata
 *
 *	Allocate rcv buffers and post them to the IO Partition.
 *	Return 0 for success, and negative for failure.
 */
static int
init_rcv_bufs(struct net_device *netdev, struct visornic_devdata *devdata)
{
	int i, j, count, err;

	/* allocate fixed number of receive buffers to post to uisnic
	 * post receive buffers after we've allocated a required amount
	 */
	for (i = 0; i < devdata->num_rcv_bufs; i++) {
		devdata->rcvbuf[i] = alloc_rcv_buf(netdev);
		if (!devdata->rcvbuf[i])
			break; /* if we failed to allocate one let us stop */
	}
	if (i == 0) /* couldn't even allocate one -- bail out */
		return -ENOMEM;
	count = i;

	/* Ensure we can alloc 2/3rd of the requested number of buffers.
	 * 2/3 is an arbitrary choice; used also in ndis init.c
	 */
	if (count < ((2 * devdata->num_rcv_bufs) / 3)) {
		/* free receive buffers we did alloc and then bail out */
		for (i = 0; i < count; i++) {
			kfree_skb(devdata->rcvbuf[i]);
			devdata->rcvbuf[i] = NULL;
		}
		return -ENOMEM;
	}

	/* post receive buffers to receive incoming input - without holding
	 * lock - we've not enabled nor started the queue so there shouldn't
	 * be any rcv or xmit activity
	 */
	for (i = 0; i < count; i++) {
		err = post_skb(devdata->cmdrsp_rcv, devdata,
			       devdata->rcvbuf[i]);
		if (!err)
			continue;

		/* Error handling -
		 * If we posted at least one skb, we should return success,
		 * but need to free the resources that we have not successfully
		 * posted.
		 */
		for (j = i; j < count; j++) {
			kfree_skb(devdata->rcvbuf[j]);
			devdata->rcvbuf[j] = NULL;
		}
		if (i == 0)
			return err;
		break;
	}

	return 0;
}

/*
 *	visornic_enable_with_timeout	- send enable to IO Part
 *	@netdev: struct net_device
 *	@timeout: Time to wait for the ACK from the enable
 *
 *	Sends enable to IOVM, inits, and posts receive buffers to IOVM
 *	timeout is defined in msecs (timeout of 0 specifies infinite wait)
 *	Return 0 for success, negative for failure.
 */
static int
visornic_enable_with_timeout(struct net_device *netdev, const int timeout)
{
	int err = 0;
	struct visornic_devdata *devdata = netdev_priv(netdev);
	unsigned long flags;
	int wait = 0;

	napi_enable(&devdata->napi);

	/* NOTE: the other end automatically unposts the rcv buffers when it
	 * gets a disable.
	 */
	err = init_rcv_bufs(netdev, devdata);
	if (err < 0) {
		dev_err(&netdev->dev,
			"%s failed to init rcv bufs\n", __func__);
		return err;
	}

	spin_lock_irqsave(&devdata->priv_lock, flags);
	devdata->enabled = 1;
	devdata->enab_dis_acked = 0;

	/* now we're ready, let's send an ENB to uisnic but until we get
	 * an ACK back from uisnic, we'll drop the packets
	 */
	devdata->n_rcv_packets_not_accepted = 0;
	spin_unlock_irqrestore(&devdata->priv_lock, flags);

	/* send enable and wait for ack -- don't hold lock when sending enable
	 * because if the queue is full, insert might sleep. If an error
	 * occurs error out.
	 */
	err = send_enbdis(netdev, 1, devdata);
	if (err)
		return err;

	spin_lock_irqsave(&devdata->priv_lock, flags);
	while ((timeout == VISORNIC_INFINITE_RSP_WAIT) ||
	       (wait < timeout)) {
		if (devdata->enab_dis_acked)
			break;
		if (devdata->server_down || devdata->server_change_state) {
			dev_dbg(&netdev->dev, "%s server went away\n",
				__func__);
			break;
		}
		set_current_state(TASK_INTERRUPTIBLE);
		spin_unlock_irqrestore(&devdata->priv_lock, flags);
		wait += schedule_timeout(msecs_to_jiffies(10));
		spin_lock_irqsave(&devdata->priv_lock, flags);
	}

	spin_unlock_irqrestore(&devdata->priv_lock, flags);

	if (!devdata->enab_dis_acked) {
		dev_err(&netdev->dev, "%s missing ACK\n", __func__);
		return -EIO;
	}

	netif_start_queue(netdev);

	return 0;
}

/*
 *	visornic_timeout_reset	- handle xmit timeout resets
 *	@work	work item that scheduled the work
 *
 *	Transmit Timeouts are typically handled by resetting the
 *	device for our virtual NIC we will send a Disable and Enable
 *	to the IOVM. If it doesn't respond we will trigger a serverdown.
 */
static void
visornic_timeout_reset(struct work_struct *work)
{
	struct visornic_devdata *devdata;
	struct net_device *netdev;
	int response = 0;

	devdata = container_of(work, struct visornic_devdata, timeout_reset);
	netdev = devdata->netdev;

	rtnl_lock();
	if (!netif_running(netdev)) {
		rtnl_unlock();
		return;
	}

	response = visornic_disable_with_timeout(netdev,
						 VISORNIC_INFINITE_RSP_WAIT);
	if (response)
		goto call_serverdown;

	response = visornic_enable_with_timeout(netdev,
						VISORNIC_INFINITE_RSP_WAIT);
	if (response)
		goto call_serverdown;

	rtnl_unlock();

	return;

call_serverdown:
	visornic_serverdown(devdata, NULL);
	rtnl_unlock();
}

/*
 *	visornic_open - Enable the visornic device and mark the queue started
 *	@netdev: netdevice to start
 *
 *      Enable the device and start the transmit queue.
 *      Return 0 for success
 */
static int
visornic_open(struct net_device *netdev)
{
	visornic_enable_with_timeout(netdev, VISORNIC_INFINITE_RSP_WAIT);

	return 0;
}

/*
 *	visornic_close - Disables the visornic device and stops the queues
 *	@netdev: netdevice to start
 *
 *      Disable the device and stop the transmit queue.
 *      Return 0 for success
 */
static int
visornic_close(struct net_device *netdev)
{
	visornic_disable_with_timeout(netdev, VISORNIC_INFINITE_RSP_WAIT);

	return 0;
}

/*
 *	devdata_xmits_outstanding - compute outstanding xmits
 *	@devdata: visornic_devdata for device
 *
 *	Return value is the number of outstanding xmits.
 */
static unsigned long devdata_xmits_outstanding(struct visornic_devdata *devdata)
{
	if (devdata->chstat.sent_xmit >= devdata->chstat.got_xmit_done)
		return devdata->chstat.sent_xmit -
			devdata->chstat.got_xmit_done;
	return (ULONG_MAX - devdata->chstat.got_xmit_done
		+ devdata->chstat.sent_xmit + 1);
}

/*
 *	vnic_hit_high_watermark
 *	@devdata: indicates visornic device we are checking
 *	@high_watermark: max num of unacked xmits we will tolerate,
 *                       before we will start throttling
 *
 *      Returns true iff the number of unacked xmits sent to
 *      the IO partition is >= high_watermark.
 */
static bool vnic_hit_high_watermark(struct visornic_devdata *devdata,
				    ulong high_watermark)
{
	return (devdata_xmits_outstanding(devdata) >= high_watermark);
}

/*
 *	vnic_hit_low_watermark
 *	@devdata: indicates visornic device we are checking
 *	@low_watermark: we will wait until the num of unacked xmits
 *                      drops to this value or lower before we start
 *                      transmitting again
 *
 *      Returns true iff the number of unacked xmits sent to
 *      the IO partition is <= low_watermark.
 */
static bool vnic_hit_low_watermark(struct visornic_devdata *devdata,
				   ulong low_watermark)
{
	return (devdata_xmits_outstanding(devdata) <= low_watermark);
}

/*
 *	visornic_xmit - send a packet to the IO Partition
 *	@skb: Packet to be sent
 *	@netdev: net device the packet is being sent from
 *
 *	Convert the skb to a cmdrsp so the IO Partition can understand it.
 *	Send the XMIT command to the IO Partition for processing. This
 *	function is protected from concurrent calls by a spinlock xmit_lock
 *	in the net_device struct, but as soon as the function returns it
 *	can be called again.
 *	Returns NETDEV_TX_OK.
 */
static int
visornic_xmit(struct sk_buff *skb, struct net_device *netdev)
{
	struct visornic_devdata *devdata;
	int len, firstfraglen, padlen;
	struct uiscmdrsp *cmdrsp = NULL;
	unsigned long flags;
	int err;

	devdata = netdev_priv(netdev);
	spin_lock_irqsave(&devdata->priv_lock, flags);

	if (netif_queue_stopped(netdev) || devdata->server_down ||
	    devdata->server_change_state) {
		spin_unlock_irqrestore(&devdata->priv_lock, flags);
		devdata->busy_cnt++;
		dev_dbg(&netdev->dev,
			"%s busy - queue stopped\n", __func__);
		kfree_skb(skb);
		return NETDEV_TX_OK;
	}

	/* sk_buff struct is used to host network data throughout all the
	 * linux network subsystems
	 */
	len = skb->len;

	/* skb->len is the FULL length of data (including fragmentary portion)
	 * skb->data_len is the length of the fragment portion in frags
	 * skb->len - skb->data_len is size of the 1st fragment in skb->data
	 * calculate the length of the first fragment that skb->data is
	 * pointing to
	 */
	firstfraglen = skb->len - skb->data_len;
	if (firstfraglen < ETH_HLEN) {
		spin_unlock_irqrestore(&devdata->priv_lock, flags);
		devdata->busy_cnt++;
		dev_err(&netdev->dev,
			"%s busy - first frag too small (%d)\n",
			__func__, firstfraglen);
		kfree_skb(skb);
		return NETDEV_TX_OK;
	}

	if ((len < ETH_MIN_PACKET_SIZE) &&
	    ((skb_end_pointer(skb) - skb->data) >= ETH_MIN_PACKET_SIZE)) {
		/* pad the packet out to minimum size */
		padlen = ETH_MIN_PACKET_SIZE - len;
		memset(&skb->data[len], 0, padlen);
		skb->tail += padlen;
		skb->len += padlen;
		len += padlen;
		firstfraglen += padlen;
	}

	cmdrsp = devdata->xmit_cmdrsp;
	/* clear cmdrsp */
	memset(cmdrsp, 0, SIZEOF_CMDRSP);
	cmdrsp->net.type = NET_XMIT;
	cmdrsp->cmdtype = CMD_NET_TYPE;

	/* save the pointer to skb -- we'll need it for completion */
	cmdrsp->net.buf = skb;

	if (vnic_hit_high_watermark(devdata,
				    devdata->max_outstanding_net_xmits)) {
		/* extra NET_XMITs queued over to IOVM - need to wait */
		devdata->chstat.reject_count++;
		if (!devdata->queuefullmsg_logged &&
		    ((devdata->chstat.reject_count & 0x3ff) == 1))
			devdata->queuefullmsg_logged = 1;
		netif_stop_queue(netdev);
		spin_unlock_irqrestore(&devdata->priv_lock, flags);
		devdata->busy_cnt++;
		dev_dbg(&netdev->dev,
			"%s busy - waiting for iovm to catch up\n",
			__func__);
		kfree_skb(skb);
		return NETDEV_TX_OK;
	}
	if (devdata->queuefullmsg_logged)
		devdata->queuefullmsg_logged = 0;

	if (skb->ip_summed == CHECKSUM_UNNECESSARY) {
		cmdrsp->net.xmt.lincsum.valid = 1;
		cmdrsp->net.xmt.lincsum.protocol = skb->protocol;
		if (skb_transport_header(skb) > skb->data) {
			cmdrsp->net.xmt.lincsum.hrawoff =
				skb_transport_header(skb) - skb->data;
			cmdrsp->net.xmt.lincsum.hrawoff = 1;
		}
		if (skb_network_header(skb) > skb->data) {
			cmdrsp->net.xmt.lincsum.nhrawoff =
				skb_network_header(skb) - skb->data;
			cmdrsp->net.xmt.lincsum.nhrawoffv = 1;
		}
		cmdrsp->net.xmt.lincsum.csum = skb->csum;
	} else {
		cmdrsp->net.xmt.lincsum.valid = 0;
	}

	/* save off the length of the entire data packet */
	cmdrsp->net.xmt.len = len;

	/* copy ethernet header from first frag into ocmdrsp
	 * - everything else will be pass in frags & DMA'ed
	 */
	memcpy(cmdrsp->net.xmt.ethhdr, skb->data, ETH_HLEN);
	/* copy frags info - from skb->data we need to only provide access
	 * beyond eth header
	 */
	cmdrsp->net.xmt.num_frags =
		visor_copy_fragsinfo_from_skb(skb, firstfraglen,
					      MAX_PHYS_INFO,
					      cmdrsp->net.xmt.frags);
	if (cmdrsp->net.xmt.num_frags < 0) {
		spin_unlock_irqrestore(&devdata->priv_lock, flags);
		devdata->busy_cnt++;
		dev_err(&netdev->dev,
			"%s busy - copy frags failed\n", __func__);
		kfree_skb(skb);
		return NETDEV_TX_OK;
	}

	err = visorchannel_signalinsert(devdata->dev->visorchannel,
					IOCHAN_TO_IOPART, cmdrsp);
	if (err) {
		netif_stop_queue(netdev);
		spin_unlock_irqrestore(&devdata->priv_lock, flags);
		devdata->busy_cnt++;
		dev_dbg(&netdev->dev,
			"%s busy - signalinsert failed\n", __func__);
		kfree_skb(skb);
		return NETDEV_TX_OK;
	}

	/* Track the skbs that have been sent to the IOVM for XMIT */
	skb_queue_head(&devdata->xmitbufhead, skb);

	/* update xmt stats */
	devdata->net_stats.tx_packets++;
	devdata->net_stats.tx_bytes += skb->len;
	devdata->chstat.sent_xmit++;

	/* check if we have hit the high watermark for netif_stop_queue() */
	if (vnic_hit_high_watermark(devdata,
				    devdata->upper_threshold_net_xmits)) {
		/* extra NET_XMITs queued over to IOVM - need to wait */
		/* stop queue - call netif_wake_queue() after lower threshold */
		netif_stop_queue(netdev);
		dev_dbg(&netdev->dev,
			"%s busy - invoking iovm flow control\n",
			__func__);
		devdata->flow_control_upper_hits++;
	}
	spin_unlock_irqrestore(&devdata->priv_lock, flags);

	/* skb will be freed when we get back NET_XMIT_DONE */
	return NETDEV_TX_OK;
}

/*
 *	visornic_get_stats - returns net_stats of the visornic device
 *	@netdev: netdevice
 *
 *	Returns the net_device_stats for the device
 */
static struct net_device_stats *
visornic_get_stats(struct net_device *netdev)
{
	struct visornic_devdata *devdata = netdev_priv(netdev);

	return &devdata->net_stats;
}

/*
 *	visornic_change_mtu - changes mtu of device.
 *	@netdev: netdevice
 *	@new_mtu: value of new mtu
 *
 *	MTU cannot be changed by system, must be changed via
 *	CONTROLVM message. All vnics and pnics in a switch have
 *	to have the same MTU for everything to work.
 *	Currently not supported.
 *	Returns EINVAL
 */
static int
visornic_change_mtu(struct net_device *netdev, int new_mtu)
{
	return -EINVAL;
}

/*
 *	visornic_set_multi - changes mtu of device.
 *	@netdev: netdevice
 *
 *	Only flag we support currently is IFF_PROMISC
 *	Returns void
 */
static void
visornic_set_multi(struct net_device *netdev)
{
	struct uiscmdrsp *cmdrsp;
	struct visornic_devdata *devdata = netdev_priv(netdev);
	int err = 0;

	if (devdata->old_flags == netdev->flags)
		return;

	if ((netdev->flags & IFF_PROMISC) ==
	    (devdata->old_flags & IFF_PROMISC))
		goto out_save_flags;

	cmdrsp = kmalloc(SIZEOF_CMDRSP, GFP_ATOMIC);
	if (!cmdrsp)
		return;
	cmdrsp->cmdtype = CMD_NET_TYPE;
	cmdrsp->net.type = NET_RCV_PROMISC;
	cmdrsp->net.enbdis.context = netdev;
	cmdrsp->net.enbdis.enable =
		netdev->flags & IFF_PROMISC;
	err = visorchannel_signalinsert(devdata->dev->visorchannel,
					IOCHAN_TO_IOPART,
					cmdrsp);
	kfree(cmdrsp);
	if (err)
		return;

out_save_flags:
	devdata->old_flags = netdev->flags;
}

/*
 *	visornic_xmit_timeout - request to timeout the xmit
 *	@netdev
 *
 *	Queue the work and return. Make sure we have not already
 *	been informed the IO Partition is gone, if it is gone
 *	we will already timeout the xmits.
 */
static void
visornic_xmit_timeout(struct net_device *netdev)
{
	struct visornic_devdata *devdata = netdev_priv(netdev);
	unsigned long flags;

	spin_lock_irqsave(&devdata->priv_lock, flags);
	if (devdata->going_away) {
		spin_unlock_irqrestore(&devdata->priv_lock, flags);
		dev_dbg(&devdata->dev->device,
			"%s aborting because device removal pending\n",
			__func__);
		return;
	}

	/* Ensure that a ServerDown message hasn't been received */
	if (!devdata->enabled ||
	    (devdata->server_down && !devdata->server_change_state)) {
		dev_dbg(&netdev->dev, "%s no processing\n",
			__func__);
		spin_unlock_irqrestore(&devdata->priv_lock, flags);
		return;
	}
	schedule_work(&devdata->timeout_reset);
	spin_unlock_irqrestore(&devdata->priv_lock, flags);
}

/*
 *	repost_return	- repost rcv bufs that have come back
 *	@cmdrsp: io channel command struct to post
 *	@devdata: visornic devdata for the device
 *	@skb: skb
 *	@netdev: netdevice
 *
 *	Repost rcv buffers that have been returned to us when
 *	we are finished with them.
 *	Returns 0 for success, -1 for error.
 */
static int
repost_return(struct uiscmdrsp *cmdrsp, struct visornic_devdata *devdata,
	      struct sk_buff *skb, struct net_device *netdev)
{
	struct net_pkt_rcv copy;
	int i = 0, cc, numreposted;
	int found_skb = 0;
	int status = 0;

	copy = cmdrsp->net.rcv;
	switch (copy.numrcvbufs) {
	case 0:
		devdata->n_rcv0++;
		break;
	case 1:
		devdata->n_rcv1++;
		break;
	case 2:
		devdata->n_rcv2++;
		break;
	default:
		devdata->n_rcvx++;
		break;
	}
	for (cc = 0, numreposted = 0; cc < copy.numrcvbufs; cc++) {
		for (i = 0; i < devdata->num_rcv_bufs; i++) {
			if (devdata->rcvbuf[i] != copy.rcvbuf[cc])
				continue;

			if ((skb) && devdata->rcvbuf[i] == skb) {
				devdata->found_repost_rcvbuf_cnt++;
				found_skb = 1;
				devdata->repost_found_skb_cnt++;
			}
			devdata->rcvbuf[i] = alloc_rcv_buf(netdev);
			if (!devdata->rcvbuf[i]) {
				devdata->num_rcv_bufs_could_not_alloc++;
				devdata->alloc_failed_in_repost_rtn_cnt++;
				status = -ENOMEM;
				break;
			}
			status = post_skb(cmdrsp, devdata, devdata->rcvbuf[i]);
			if (status) {
				kfree_skb(devdata->rcvbuf[i]);
				devdata->rcvbuf[i] = NULL;
				break;
			}
			numreposted++;
			break;
		}
	}
	if (numreposted != copy.numrcvbufs) {
		devdata->n_repost_deficit++;
		status = -EINVAL;
	}
	if (skb) {
		if (found_skb) {
			kfree_skb(skb);
		} else {
			status = -EINVAL;
			devdata->bad_rcv_buf++;
		}
	}
	return status;
}

/*
 *	visornic_rx - Handle receive packets coming back from IO Part
 *	@cmdrsp: Receive packet returned from IO Part
 *
 *	Got a receive packet back from the IO Part, handle it and send
 *	it up the stack.
 *	Returns 1 iff an skb was received, otherwise 0
 */
static int
visornic_rx(struct uiscmdrsp *cmdrsp)
{
	struct visornic_devdata *devdata;
	struct sk_buff *skb, *prev, *curr;
	struct net_device *netdev;
	int cc, currsize, off;
	struct ethhdr *eth;
	unsigned long flags;

	/* post new rcv buf to the other end using the cmdrsp we have at hand
	 * post it without holding lock - but we'll use the signal lock to
	 * synchronize the queue insert the cmdrsp that contains the net.rcv
	 * is the one we are using to repost, so copy the info we need from it.
	 */
	skb = cmdrsp->net.buf;
	netdev = skb->dev;

	devdata = netdev_priv(netdev);

	spin_lock_irqsave(&devdata->priv_lock, flags);
	atomic_dec(&devdata->num_rcvbuf_in_iovm);

	/* set length to how much was ACTUALLY received -
	 * NOTE: rcv_done_len includes actual length of data rcvd
	 * including ethhdr
	 */
	skb->len = cmdrsp->net.rcv.rcv_done_len;

	/* update rcv stats - call it with priv_lock held */
	devdata->net_stats.rx_packets++;
	devdata->net_stats.rx_bytes += skb->len;

	/* test enabled while holding lock */
	if (!(devdata->enabled && devdata->enab_dis_acked)) {
		/* don't process it unless we're in enable mode and until
		 * we've gotten an ACK saying the other end got our RCV enable
		 */
		spin_unlock_irqrestore(&devdata->priv_lock, flags);
		repost_return(cmdrsp, devdata, skb, netdev);
		return 0;
	}

	spin_unlock_irqrestore(&devdata->priv_lock, flags);

	/* when skb was allocated, skb->dev, skb->data, skb->len and
	 * skb->data_len were setup. AND, data has already put into the
	 * skb (both first frag and in frags pages)
	 * NOTE: firstfragslen is the amount of data in skb->data and that
	 * which is not in nr_frags or frag_list. This is now simply
	 * RCVPOST_BUF_SIZE. bump tail to show how much data is in
	 * firstfrag & set data_len to show rest see if we have to chain
	 * frag_list.
	 */
	if (skb->len > RCVPOST_BUF_SIZE) {	/* do PRECAUTIONARY check */
		if (cmdrsp->net.rcv.numrcvbufs < 2) {
			if (repost_return(cmdrsp, devdata, skb, netdev) < 0)
				dev_err(&devdata->netdev->dev,
					"repost_return failed");
			return 0;
		}
		/* length rcvd is greater than firstfrag in this skb rcv buf  */
		skb->tail += RCVPOST_BUF_SIZE;	/* amount in skb->data */
		skb->data_len = skb->len - RCVPOST_BUF_SIZE;	/* amount that
								 *  will be in
								 * frag_list
								 */
	} else {
		/* data fits in this skb - no chaining - do
		 * PRECAUTIONARY check
		 */
		if (cmdrsp->net.rcv.numrcvbufs != 1) {	/* should be 1 */
			if (repost_return(cmdrsp, devdata, skb, netdev) < 0)
				dev_err(&devdata->netdev->dev,
					"repost_return failed");
			return 0;
		}
		skb->tail += skb->len;
		skb->data_len = 0;	/* nothing rcvd in frag_list */
	}
	off = skb_tail_pointer(skb) - skb->data;

	/* amount we bumped tail by in the head skb
	 * it is used to calculate the size of each chained skb below
	 * it is also used to index into bufline to continue the copy
	 * (for chansocktwopc)
	 * if necessary chain the rcv skbs together.
	 * NOTE: index 0 has the same as cmdrsp->net.rcv.skb; we need to
	 * chain the rest to that one.
	 * - do PRECAUTIONARY check
	 */
	if (cmdrsp->net.rcv.rcvbuf[0] != skb) {
		if (repost_return(cmdrsp, devdata, skb, netdev) < 0)
			dev_err(&devdata->netdev->dev, "repost_return failed");
		return 0;
	}

	if (cmdrsp->net.rcv.numrcvbufs > 1) {
		/* chain the various rcv buffers into the skb's frag_list. */
		/* Note: off was initialized above  */
		for (cc = 1, prev = NULL;
		     cc < cmdrsp->net.rcv.numrcvbufs; cc++) {
			curr = (struct sk_buff *)cmdrsp->net.rcv.rcvbuf[cc];
			curr->next = NULL;
			if (!prev)	/* start of list- set head */
				skb_shinfo(skb)->frag_list = curr;
			else
				prev->next = curr;
			prev = curr;

			/* should we set skb->len and skb->data_len for each
			 * buffer being chained??? can't hurt!
			 */
			currsize = min(skb->len - off,
				       (unsigned int)RCVPOST_BUF_SIZE);
			curr->len = currsize;
			curr->tail += currsize;
			curr->data_len = 0;
			off += currsize;
		}
		/* assert skb->len == off */
		if (skb->len != off) {
			netdev_err(devdata->netdev,
				   "something wrong; skb->len:%d != off:%d\n",
				   skb->len, off);
		}
	}

	/* set up packet's protocol type using ethernet header - this
	 * sets up skb->pkt_type & it also PULLS out the eth header
	 */
	skb->protocol = eth_type_trans(skb, netdev);

	eth = eth_hdr(skb);

	skb->csum = 0;
	skb->ip_summed = CHECKSUM_NONE;

	do {
		if (netdev->flags & IFF_PROMISC)
			break;	/* accept all packets */
		if (skb->pkt_type == PACKET_BROADCAST) {
			if (netdev->flags & IFF_BROADCAST)
				break;	/* accept all broadcast packets */
		} else if (skb->pkt_type == PACKET_MULTICAST) {
			if ((netdev->flags & IFF_MULTICAST) &&
			    (netdev_mc_count(netdev))) {
				struct netdev_hw_addr *ha;
				int found_mc = 0;

				/* only accept multicast packets that we can
				 * find in our multicast address list
				 */
				netdev_for_each_mc_addr(ha, netdev) {
					if (ether_addr_equal(eth->h_dest,
							     ha->addr)) {
						found_mc = 1;
						break;
					}
				}
				/* accept pkt, dest matches a multicast addr */
				if (found_mc)
					break;
			}
		/* accept packet, h_dest must match vnic  mac address */
		} else if (skb->pkt_type == PACKET_HOST) {
			break;
		} else if (skb->pkt_type == PACKET_OTHERHOST) {
			/* something is not right */
			dev_err(&devdata->netdev->dev,
				"**** FAILED to deliver rcv packet to OS; name:%s Dest:%pM VNIC:%pM\n",
				netdev->name, eth->h_dest, netdev->dev_addr);
		}
		/* drop packet - don't forward it up to OS */
		devdata->n_rcv_packets_not_accepted++;
		repost_return(cmdrsp, devdata, skb, netdev);
		return 0;
	} while (0);

	netif_receive_skb(skb);
	/* netif_rx returns various values, but "in practice most drivers
	 * ignore the return value
	 */

	skb = NULL;
	/*
	 * whether the packet got dropped or handled, the skb is freed by
	 * kernel code, so we shouldn't free it. but we should repost a
	 * new rcv buffer.
	 */
	repost_return(cmdrsp, devdata, skb, netdev);
	return 1;
}

/*
 *	devdata_initialize	- Initialize devdata structure
 *	@devdata: visornic_devdata structure to initialize
 *	#dev: visorbus_deviced it belongs to
 *
 *	Setup initial values for the visornic based on channel and default
 *	values.
 *	Returns a pointer to the devdata structure
 */
static struct visornic_devdata *
devdata_initialize(struct visornic_devdata *devdata, struct visor_device *dev)
{
	devdata->dev = dev;
	devdata->incarnation_id = get_jiffies_64();
	return devdata;
}

/*
 *	devdata_release	- Frees up references in devdata
 *	@devdata: struct to clean up
 *
 *	Frees up references in devdata.
 *	Returns void
 */
static void devdata_release(struct visornic_devdata *devdata)
{
	kfree(devdata->rcvbuf);
	kfree(devdata->cmdrsp_rcv);
	kfree(devdata->xmit_cmdrsp);
}

static const struct net_device_ops visornic_dev_ops = {
	.ndo_open = visornic_open,
	.ndo_stop = visornic_close,
	.ndo_start_xmit = visornic_xmit,
	.ndo_get_stats = visornic_get_stats,
	.ndo_change_mtu = visornic_change_mtu,
	.ndo_tx_timeout = visornic_xmit_timeout,
	.ndo_set_rx_mode = visornic_set_multi,
};

/* DebugFS code */
static ssize_t info_debugfs_read(struct file *file, char __user *buf,
				 size_t len, loff_t *offset)
{
	ssize_t bytes_read = 0;
	int str_pos = 0;
	struct visornic_devdata *devdata;
	struct net_device *dev;
	char *vbuf;

	if (len > MAX_BUF)
		len = MAX_BUF;
	vbuf = kzalloc(len, GFP_KERNEL);
	if (!vbuf)
		return -ENOMEM;

	/* for each vnic channel dump out channel specific data */
	rcu_read_lock();
	for_each_netdev_rcu(current->nsproxy->net_ns, dev) {
		/* Only consider netdevs that are visornic, and are open */
		if ((dev->netdev_ops != &visornic_dev_ops) ||
		    (!netif_queue_stopped(dev)))
			continue;

		devdata = netdev_priv(dev);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     "netdev = %s (0x%p), MAC Addr %pM\n",
				     dev->name,
				     dev,
				     dev->dev_addr);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     "VisorNic Dev Info = 0x%p\n", devdata);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " num_rcv_bufs = %d\n",
				     devdata->num_rcv_bufs);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " max_outstanding_next_xmits = %lu\n",
				    devdata->max_outstanding_net_xmits);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " upper_threshold_net_xmits = %lu\n",
				     devdata->upper_threshold_net_xmits);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " lower_threshold_net_xmits = %lu\n",
				     devdata->lower_threshold_net_xmits);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " queuefullmsg_logged = %d\n",
				     devdata->queuefullmsg_logged);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " chstat.got_rcv = %lu\n",
				     devdata->chstat.got_rcv);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " chstat.got_enbdisack = %lu\n",
				     devdata->chstat.got_enbdisack);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " chstat.got_xmit_done = %lu\n",
				     devdata->chstat.got_xmit_done);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " chstat.xmit_fail = %lu\n",
				     devdata->chstat.xmit_fail);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " chstat.sent_enbdis = %lu\n",
				     devdata->chstat.sent_enbdis);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " chstat.sent_promisc = %lu\n",
				     devdata->chstat.sent_promisc);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " chstat.sent_post = %lu\n",
				     devdata->chstat.sent_post);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " chstat.sent_post_failed = %lu\n",
				     devdata->chstat.sent_post_failed);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " chstat.sent_xmit = %lu\n",
				     devdata->chstat.sent_xmit);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " chstat.reject_count = %lu\n",
				     devdata->chstat.reject_count);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " chstat.extra_rcvbufs_sent = %lu\n",
				     devdata->chstat.extra_rcvbufs_sent);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " n_rcv0 = %lu\n", devdata->n_rcv0);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " n_rcv1 = %lu\n", devdata->n_rcv1);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " n_rcv2 = %lu\n", devdata->n_rcv2);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " n_rcvx = %lu\n", devdata->n_rcvx);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " num_rcvbuf_in_iovm = %d\n",
				     atomic_read(&devdata->num_rcvbuf_in_iovm));
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " alloc_failed_in_if_needed_cnt = %lu\n",
				     devdata->alloc_failed_in_if_needed_cnt);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " alloc_failed_in_repost_rtn_cnt = %lu\n",
				     devdata->alloc_failed_in_repost_rtn_cnt);
		/* str_pos += scnprintf(vbuf + str_pos, len - str_pos,
		 *		     " inner_loop_limit_reached_cnt = %lu\n",
		 *		     devdata->inner_loop_limit_reached_cnt);
		 */
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " found_repost_rcvbuf_cnt = %lu\n",
				     devdata->found_repost_rcvbuf_cnt);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " repost_found_skb_cnt = %lu\n",
				     devdata->repost_found_skb_cnt);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " n_repost_deficit = %lu\n",
				     devdata->n_repost_deficit);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " bad_rcv_buf = %lu\n",
				     devdata->bad_rcv_buf);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " n_rcv_packets_not_accepted = %lu\n",
				     devdata->n_rcv_packets_not_accepted);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " interrupts_rcvd = %llu\n",
				     devdata->interrupts_rcvd);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " interrupts_notme = %llu\n",
				     devdata->interrupts_notme);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " interrupts_disabled = %llu\n",
				     devdata->interrupts_disabled);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " busy_cnt = %llu\n",
				     devdata->busy_cnt);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " flow_control_upper_hits = %llu\n",
				     devdata->flow_control_upper_hits);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " flow_control_lower_hits = %llu\n",
				     devdata->flow_control_lower_hits);
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " netif_queue = %s\n",
				     netif_queue_stopped(devdata->netdev) ?
				     "stopped" : "running");
		str_pos += scnprintf(vbuf + str_pos, len - str_pos,
				     " xmits_outstanding = %lu\n",
				     devdata_xmits_outstanding(devdata));
	}
	rcu_read_unlock();
	bytes_read = simple_read_from_buffer(buf, len, offset, vbuf, str_pos);
	kfree(vbuf);
	return bytes_read;
}

static struct dentry *visornic_debugfs_dir;
static const struct file_operations debugfs_info_fops = {
	.read = info_debugfs_read,
};

/*
 *	send_rcv_posts_if_needed
 *	@devdata: visornic device
 *
 *	Send receive buffers to the IO Partition.
 *	Returns void
 */
static int
send_rcv_posts_if_needed(struct visornic_devdata *devdata)
{
	int i;
	struct net_device *netdev;
	struct uiscmdrsp *cmdrsp = devdata->cmdrsp_rcv;
	int cur_num_rcv_bufs_to_alloc, rcv_bufs_allocated;
	int err;

	/* don't do this until vnic is marked ready */
	if (!(devdata->enabled && devdata->enab_dis_acked))
		return 0;

	netdev = devdata->netdev;
	rcv_bufs_allocated = 0;
	/* this code is trying to prevent getting stuck here forever,
	 * but still retry it if you cant allocate them all this time.
	 */
	cur_num_rcv_bufs_to_alloc = devdata->num_rcv_bufs_could_not_alloc;
	while (cur_num_rcv_bufs_to_alloc > 0) {
		cur_num_rcv_bufs_to_alloc--;
		for (i = 0; i < devdata->num_rcv_bufs; i++) {
			if (devdata->rcvbuf[i])
				continue;
			devdata->rcvbuf[i] = alloc_rcv_buf(netdev);
			if (!devdata->rcvbuf[i]) {
				devdata->alloc_failed_in_if_needed_cnt++;
				break;
			}
			rcv_bufs_allocated++;
			err = post_skb(cmdrsp, devdata, devdata->rcvbuf[i]);
			if (err) {
				kfree_skb(devdata->rcvbuf[i]);
				devdata->rcvbuf[i] = NULL;
				break;
			}
			devdata->chstat.extra_rcvbufs_sent++;
		}
	}
	devdata->num_rcv_bufs_could_not_alloc -= rcv_bufs_allocated;
	return 0;
}

/*
 *	drain_resp_queue  - drains and ignores all messages from the resp queue
 *	@cmdrsp: io channel command response message
 *	@devdata: visornic device to drain
 */
static void
drain_resp_queue(struct uiscmdrsp *cmdrsp, struct visornic_devdata *devdata)
{
	while (!visorchannel_signalremove(devdata->dev->visorchannel,
					  IOCHAN_FROM_IOPART,
					  cmdrsp))
		;
}

/*
 *	service_resp_queue	- drains the response queue
 *	@cmdrsp: io channel command response message
 *	@devdata: visornic device to drain
 *
 *	Drain the response queue of any responses from the IO partition.
 *	Process the responses as we get them.
 *	Returns when response queue is empty or when the thread stops.
 */
static void
service_resp_queue(struct uiscmdrsp *cmdrsp, struct visornic_devdata *devdata,
		   int *rx_work_done, int budget)
{
	unsigned long flags;
	struct net_device *netdev;

	while (*rx_work_done < budget) {
	/* TODO: CLIENT ACQUIRE -- Don't really need this at the
	 * moment
	 */
		if (visorchannel_signalremove(devdata->dev->visorchannel,
					      IOCHAN_FROM_IOPART,
					      cmdrsp))
			break; /* queue empty */

		switch (cmdrsp->net.type) {
		case NET_RCV:
			devdata->chstat.got_rcv++;
			/* process incoming packet */
			*rx_work_done += visornic_rx(cmdrsp);
			break;
		case NET_XMIT_DONE:
			spin_lock_irqsave(&devdata->priv_lock, flags);
			devdata->chstat.got_xmit_done++;
			if (cmdrsp->net.xmtdone.xmt_done_result)
				devdata->chstat.xmit_fail++;
			/* only call queue wake if we stopped it */
			netdev = ((struct sk_buff *)cmdrsp->net.buf)->dev;
			/* ASSERT netdev == vnicinfo->netdev; */
			if ((netdev == devdata->netdev) &&
			    netif_queue_stopped(netdev)) {
				/* check if we have crossed the lower watermark
				 * for netif_wake_queue()
				 */
				if (vnic_hit_low_watermark
				    (devdata,
				     devdata->lower_threshold_net_xmits)) {
					/* enough NET_XMITs completed
					 * so can restart netif queue
					 */
					netif_wake_queue(netdev);
					devdata->flow_control_lower_hits++;
				}
			}
			skb_unlink(cmdrsp->net.buf, &devdata->xmitbufhead);
			spin_unlock_irqrestore(&devdata->priv_lock, flags);
			kfree_skb(cmdrsp->net.buf);
			break;
		case NET_RCV_ENBDIS_ACK:
			devdata->chstat.got_enbdisack++;
			netdev = (struct net_device *)
			cmdrsp->net.enbdis.context;
			spin_lock_irqsave(&devdata->priv_lock, flags);
			devdata->enab_dis_acked = 1;
			spin_unlock_irqrestore(&devdata->priv_lock, flags);

			if (devdata->server_down &&
			    devdata->server_change_state) {
				/* Inform Linux that the link is up */
				devdata->server_down = false;
				devdata->server_change_state = false;
				netif_wake_queue(netdev);
				netif_carrier_on(netdev);
			}
			break;
		case NET_CONNECT_STATUS:
			netdev = devdata->netdev;
			if (cmdrsp->net.enbdis.enable == 1) {
				spin_lock_irqsave(&devdata->priv_lock, flags);
				devdata->enabled = cmdrsp->net.enbdis.enable;
				spin_unlock_irqrestore(&devdata->priv_lock,
						       flags);
				netif_wake_queue(netdev);
				netif_carrier_on(netdev);
			} else {
				netif_stop_queue(netdev);
				netif_carrier_off(netdev);
				spin_lock_irqsave(&devdata->priv_lock, flags);
				devdata->enabled = cmdrsp->net.enbdis.enable;
				spin_unlock_irqrestore(&devdata->priv_lock,
						       flags);
			}
			break;
		default:
			break;
		}
		/* cmdrsp is now available for reuse  */
	}
}

static int visornic_poll(struct napi_struct *napi, int budget)
{
	struct visornic_devdata *devdata = container_of(napi,
							struct visornic_devdata,
							napi);
	int rx_count = 0;
	int err;

	err = send_rcv_posts_if_needed(devdata);
	if (err)
		return err;

	service_resp_queue(devdata->cmdrsp, devdata, &rx_count, budget);

	/* If there aren't any more packets to receive stop the poll */
	if (rx_count < budget)
		napi_complete_done(napi, rx_count);

	return rx_count;
}

/*
 *	poll_for_irq	- Checks the status of the response queue.
 *	@v: void pointer to the visronic devdata
 *
 *	Main function of the vnic_incoming thread. Periodically check the
 *	response queue and drain it if needed.
 *	Returns when thread has stopped.
 */
static void
poll_for_irq(unsigned long v)
{
	struct visornic_devdata *devdata = (struct visornic_devdata *)v;

	if (!visorchannel_signalempty(
				   devdata->dev->visorchannel,
				   IOCHAN_FROM_IOPART))
		napi_schedule(&devdata->napi);

	atomic_set(&devdata->interrupt_rcvd, 0);

	mod_timer(&devdata->irq_poll_timer, msecs_to_jiffies(2));
}

/*
 *	visornic_probe	- probe function for visornic devices
 *	@dev: The visor device discovered
 *
 *	Called when visorbus discovers a visornic device on its
 *	bus. It creates a new visornic ethernet adapter.
 *	Returns 0 or negative for error.
 */
static int visornic_probe(struct visor_device *dev)
{
	struct visornic_devdata *devdata = NULL;
	struct net_device *netdev = NULL;
	int err;
	int channel_offset = 0;
	u64 features;

	netdev = alloc_etherdev(sizeof(struct visornic_devdata));
	if (!netdev) {
		dev_err(&dev->device,
			"%s alloc_etherdev failed\n", __func__);
		return -ENOMEM;
	}

	netdev->netdev_ops = &visornic_dev_ops;
	netdev->watchdog_timeo = 5 * HZ;
	SET_NETDEV_DEV(netdev, &dev->device);

	/* Get MAC address from channel and read it into the device. */
	netdev->addr_len = ETH_ALEN;
	channel_offset = offsetof(struct visor_io_channel, vnic.macaddr);
	err = visorbus_read_channel(dev, channel_offset, netdev->dev_addr,
				    ETH_ALEN);
	if (err < 0) {
		dev_err(&dev->device,
			"%s failed to get mac addr from chan (%d)\n",
			__func__, err);
		goto cleanup_netdev;
	}

	devdata = devdata_initialize(netdev_priv(netdev), dev);
	if (!devdata) {
		dev_err(&dev->device,
			"%s devdata_initialize failed\n", __func__);
		err = -ENOMEM;
		goto cleanup_netdev;
	}
	/* don't trust messages laying around in the channel */
	drain_resp_queue(devdata->cmdrsp, devdata);

	devdata->netdev = netdev;
	dev_set_drvdata(&dev->device, devdata);
	init_waitqueue_head(&devdata->rsp_queue);
	spin_lock_init(&devdata->priv_lock);
	devdata->enabled = 0; /* not yet */
	atomic_set(&devdata->usage, 1);

	/* Setup rcv bufs */
	channel_offset = offsetof(struct visor_io_channel, vnic.num_rcv_bufs);
	err = visorbus_read_channel(dev, channel_offset,
				    &devdata->num_rcv_bufs, 4);
	if (err) {
		dev_err(&dev->device,
			"%s failed to get #rcv bufs from chan (%d)\n",
			__func__, err);
		goto cleanup_netdev;
	}

	devdata->rcvbuf = kcalloc(devdata->num_rcv_bufs,
				  sizeof(struct sk_buff *), GFP_KERNEL);
	if (!devdata->rcvbuf) {
		err = -ENOMEM;
		goto cleanup_netdev;
	}

	/* set the net_xmit outstanding threshold */
	/* always leave two slots open but you should have 3 at a minimum */
	/* note that max_outstanding_net_xmits must be > 0 */
	devdata->max_outstanding_net_xmits =
		max_t(unsigned long, 3, ((devdata->num_rcv_bufs / 3) - 2));
	devdata->upper_threshold_net_xmits =
		max_t(unsigned long,
		      2, (devdata->max_outstanding_net_xmits - 1));
	devdata->lower_threshold_net_xmits =
		max_t(unsigned long,
		      1, (devdata->max_outstanding_net_xmits / 2));

	skb_queue_head_init(&devdata->xmitbufhead);

	/* create a cmdrsp we can use to post and unpost rcv buffers */
	devdata->cmdrsp_rcv = kmalloc(SIZEOF_CMDRSP, GFP_ATOMIC);
	if (!devdata->cmdrsp_rcv) {
		err = -ENOMEM;
		goto cleanup_rcvbuf;
	}
	devdata->xmit_cmdrsp = kmalloc(SIZEOF_CMDRSP, GFP_ATOMIC);
	if (!devdata->xmit_cmdrsp) {
		err = -ENOMEM;
		goto cleanup_cmdrsp_rcv;
	}
	INIT_WORK(&devdata->timeout_reset, visornic_timeout_reset);
	devdata->server_down = false;
	devdata->server_change_state = false;

	/*set the default mtu */
	channel_offset = offsetof(struct visor_io_channel, vnic.mtu);
	err = visorbus_read_channel(dev, channel_offset, &netdev->mtu, 4);
	if (err) {
		dev_err(&dev->device,
			"%s failed to get mtu from chan (%d)\n",
			__func__, err);
		goto cleanup_xmit_cmdrsp;
	}

	/* TODO: Setup Interrupt information */
	/* Let's start our threads to get responses */
	netif_napi_add(netdev, &devdata->napi, visornic_poll, NAPI_WEIGHT);

	setup_timer(&devdata->irq_poll_timer, poll_for_irq,
		    (unsigned long)devdata);
	/* Note: This time has to start running before the while
	 * loop below because the napi routine is responsible for
	 * setting enab_dis_acked
	 */
	mod_timer(&devdata->irq_poll_timer, msecs_to_jiffies(2));

	channel_offset = offsetof(struct visor_io_channel,
				  channel_header.features);
	err = visorbus_read_channel(dev, channel_offset, &features, 8);
	if (err) {
		dev_err(&dev->device,
			"%s failed to get features from chan (%d)\n",
			__func__, err);
		goto cleanup_napi_add;
	}

	features |= VISOR_CHANNEL_IS_POLLING;
	features |= VISOR_DRIVER_ENHANCED_RCVBUF_CHECKING;
	err = visorbus_write_channel(dev, channel_offset, &features, 8);
	if (err) {
		dev_err(&dev->device,
			"%s failed to set features in chan (%d)\n",
			__func__, err);
		goto cleanup_napi_add;
	}

	/* Note: Interrupts have to be enable before the while
	 * loop below because the napi routine is responsible for
	 * setting enab_dis_acked
	 */
	visorbus_enable_channel_interrupts(dev);

	err = register_netdev(netdev);
	if (err) {
		dev_err(&dev->device,
			"%s register_netdev failed (%d)\n", __func__, err);
		goto cleanup_napi_add;
	}

	/* create debug/sysfs directories */
	devdata->eth_debugfs_dir = debugfs_create_dir(netdev->name,
						      visornic_debugfs_dir);
	if (!devdata->eth_debugfs_dir) {
		dev_err(&dev->device,
			"%s debugfs_create_dir %s failed\n",
			__func__, netdev->name);
		err = -ENOMEM;
		goto cleanup_register_netdev;
	}

	dev_info(&dev->device, "%s success netdev=%s\n",
		 __func__, netdev->name);
	return 0;

cleanup_register_netdev:
	unregister_netdev(netdev);

cleanup_napi_add:
	del_timer_sync(&devdata->irq_poll_timer);
	netif_napi_del(&devdata->napi);

cleanup_xmit_cmdrsp:
	kfree(devdata->xmit_cmdrsp);

cleanup_cmdrsp_rcv:
	kfree(devdata->cmdrsp_rcv);

cleanup_rcvbuf:
	kfree(devdata->rcvbuf);

cleanup_netdev:
	free_netdev(netdev);
	return err;
}

/*
 *	host_side_disappeared	- IO part is gone.
 *	@devdata: device object
 *
 *	IO partition servicing this device is gone, do cleanup
 *	Returns void.
 */
static void host_side_disappeared(struct visornic_devdata *devdata)
{
	unsigned long flags;

	spin_lock_irqsave(&devdata->priv_lock, flags);
	devdata->dev = NULL;   /* indicate device destroyed */
	spin_unlock_irqrestore(&devdata->priv_lock, flags);
}

/*
 *	visornic_remove		- Called when visornic dev goes away
 *	@dev: visornic device that is being removed
 *
 *	Called when DEVICE_DESTROY gets called to remove device.
 *	Returns void
 */
static void visornic_remove(struct visor_device *dev)
{
	struct visornic_devdata *devdata = dev_get_drvdata(&dev->device);
	struct net_device *netdev;
	unsigned long flags;

	if (!devdata) {
		dev_err(&dev->device, "%s no devdata\n", __func__);
		return;
	}
	spin_lock_irqsave(&devdata->priv_lock, flags);
	if (devdata->going_away) {
		spin_unlock_irqrestore(&devdata->priv_lock, flags);
		dev_err(&dev->device, "%s already being removed\n", __func__);
		return;
	}
	devdata->going_away = true;
	spin_unlock_irqrestore(&devdata->priv_lock, flags);
	netdev = devdata->netdev;
	if (!netdev) {
		dev_err(&dev->device, "%s not net device\n", __func__);
		return;
	}

	/* going_away prevents new items being added to the workqueues */
	cancel_work_sync(&devdata->timeout_reset);

	debugfs_remove_recursive(devdata->eth_debugfs_dir);

	unregister_netdev(netdev);  /* this will call visornic_close() */

	del_timer_sync(&devdata->irq_poll_timer);
	netif_napi_del(&devdata->napi);

	dev_set_drvdata(&dev->device, NULL);
	host_side_disappeared(devdata);
	devdata_release(devdata);
	free_netdev(netdev);
}

/*
 *	visornic_pause		- Called when IO Part disappears
 *	@dev: visornic device that is being serviced
 *	@complete_func: call when finished.
 *
 *	Called when the IO Partition has gone down. Need to free
 *	up resources and wait for IO partition to come back. Mark
 *	link as down and don't attempt any DMA. When we have freed
 *	memory call the complete_func so that Command knows we are
 *	done. If we don't call complete_func, IO part will never
 *	come back.
 *	Returns 0 for success.
 */
static int visornic_pause(struct visor_device *dev,
			  visorbus_state_complete_func complete_func)
{
	struct visornic_devdata *devdata = dev_get_drvdata(&dev->device);

	visornic_serverdown(devdata, complete_func);
	return 0;
}

/*
 *	visornic_resume		- Called when IO part has recovered
 *	@dev: visornic device that is being serviced
 *	@compelte_func: call when finished
 *
 *	Called when the IO partition has recovered. Reestablish
 *	connection to the IO part and set the link up. Okay to do
 *	DMA again.
 *	Returns 0 for success.
 */
static int visornic_resume(struct visor_device *dev,
			   visorbus_state_complete_func complete_func)
{
	struct visornic_devdata *devdata;
	struct net_device *netdev;
	unsigned long flags;

	devdata = dev_get_drvdata(&dev->device);
	if (!devdata) {
		dev_err(&dev->device, "%s no devdata\n", __func__);
		return -EINVAL;
	}

	netdev = devdata->netdev;

	spin_lock_irqsave(&devdata->priv_lock, flags);
	if (devdata->server_change_state) {
		spin_unlock_irqrestore(&devdata->priv_lock, flags);
		dev_err(&dev->device, "%s server already changing state\n",
			__func__);
		return -EINVAL;
	}
	if (!devdata->server_down) {
		spin_unlock_irqrestore(&devdata->priv_lock, flags);
		dev_err(&dev->device, "%s server not down\n", __func__);
		complete_func(dev, 0);
		return 0;
	}
	devdata->server_change_state = true;
	spin_unlock_irqrestore(&devdata->priv_lock, flags);

	/* Must transition channel to ATTACHED state BEFORE
	 * we can start using the device again.
	 * TODO: State transitions
	 */
	mod_timer(&devdata->irq_poll_timer, msecs_to_jiffies(2));

	rtnl_lock();
	dev_open(netdev);
	rtnl_unlock();

	complete_func(dev, 0);
	return 0;
}

/* This is used to tell the visorbus driver which types of visor devices
 * we support, and what functions to call when a visor device that we support
 * is attached or removed.
 */
static struct visor_driver visornic_driver = {
	.name = "visornic",
	.owner = THIS_MODULE,
	.channel_types = visornic_channel_types,
	.probe = visornic_probe,
	.remove = visornic_remove,
	.pause = visornic_pause,
	.resume = visornic_resume,
	.channel_interrupt = NULL,
};

/*
 *	visornic_init	- Init function
 *
 *	Init function for the visornic driver. Do initial driver setup
 *	and wait for devices.
 *	Returns 0 for success, negative for error.
 */
static int visornic_init(void)
{
	struct dentry *ret;
	int err = -ENOMEM;

	visornic_debugfs_dir = debugfs_create_dir("visornic", NULL);
	if (!visornic_debugfs_dir)
		return err;

	ret = debugfs_create_file("info", 0400, visornic_debugfs_dir, NULL,
				  &debugfs_info_fops);
	if (!ret)
		goto cleanup_debugfs;
	ret = debugfs_create_file("enable_ints", 0200, visornic_debugfs_dir,
				  NULL, &debugfs_enable_ints_fops);
	if (!ret)
		goto cleanup_debugfs;

	err = visorbus_register_visor_driver(&visornic_driver);
	if (err)
		goto cleanup_debugfs;

	return 0;

cleanup_debugfs:
	debugfs_remove_recursive(visornic_debugfs_dir);

	return err;
}

/*
 *	visornic_cleanup	- driver exit routine
 *
 *	Unregister driver from the bus and free up memory.
 */
static void visornic_cleanup(void)
{
	visorbus_unregister_visor_driver(&visornic_driver);

	debugfs_remove_recursive(visornic_debugfs_dir);
}

module_init(visornic_init);
module_exit(visornic_cleanup);

MODULE_AUTHOR("Unisys");
MODULE_LICENSE("GPL");
MODULE_DESCRIPTION("s-Par NIC driver for virtual network devices");