summaryrefslogblamecommitdiff
path: root/drivers/scsi/scsi_lib.c
blob: 9f996499fa9d2b44dbdf9431040a47bce15855c5 (plain) (tree)
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
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
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





























































































                                                                                        
                                          


                 

                                                    






















                                                                          

                                                        




                                                                     
                                                            




























                                                                          

                                                                      



                                                                          
                                                                  

                                                                             






                                                     






























































































































                                                                             




















                                                                           






































                                                                     
                                                   
                            
                                                                       



























































































































                                                                                

                            
                                             



                                                     





































































































































































































































































































































































































































































                                                                                     
                             
                                     





























































































































































































































































































































































































































































































                                                                                                  
                                                  


















































































































































































































































































































































































































































































































































































































                                                                               
/*
 *  scsi_lib.c Copyright (C) 1999 Eric Youngdale
 *
 *  SCSI queueing library.
 *      Initial versions: Eric Youngdale (eric@andante.org).
 *                        Based upon conversations with large numbers
 *                        of people at Linux Expo.
 */

#include <linux/bio.h>
#include <linux/blkdev.h>
#include <linux/completion.h>
#include <linux/kernel.h>
#include <linux/mempool.h>
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/pci.h>
#include <linux/delay.h>

#include <scsi/scsi.h>
#include <scsi/scsi_dbg.h>
#include <scsi/scsi_device.h>
#include <scsi/scsi_driver.h>
#include <scsi/scsi_eh.h>
#include <scsi/scsi_host.h>
#include <scsi/scsi_request.h>

#include "scsi_priv.h"
#include "scsi_logging.h"


#define SG_MEMPOOL_NR		(sizeof(scsi_sg_pools)/sizeof(struct scsi_host_sg_pool))
#define SG_MEMPOOL_SIZE		32

struct scsi_host_sg_pool {
	size_t		size;
	char		*name; 
	kmem_cache_t	*slab;
	mempool_t	*pool;
};

#if (SCSI_MAX_PHYS_SEGMENTS < 32)
#error SCSI_MAX_PHYS_SEGMENTS is too small
#endif

#define SP(x) { x, "sgpool-" #x } 
struct scsi_host_sg_pool scsi_sg_pools[] = { 
	SP(8),
	SP(16),
	SP(32),
#if (SCSI_MAX_PHYS_SEGMENTS > 32)
	SP(64),
#if (SCSI_MAX_PHYS_SEGMENTS > 64)
	SP(128),
#if (SCSI_MAX_PHYS_SEGMENTS > 128)
	SP(256),
#if (SCSI_MAX_PHYS_SEGMENTS > 256)
#error SCSI_MAX_PHYS_SEGMENTS is too large
#endif
#endif
#endif
#endif
}; 	
#undef SP


/*
 * Function:    scsi_insert_special_req()
 *
 * Purpose:     Insert pre-formed request into request queue.
 *
 * Arguments:   sreq	- request that is ready to be queued.
 *              at_head	- boolean.  True if we should insert at head
 *                        of queue, false if we should insert at tail.
 *
 * Lock status: Assumed that lock is not held upon entry.
 *
 * Returns:     Nothing
 *
 * Notes:       This function is called from character device and from
 *              ioctl types of functions where the caller knows exactly
 *              what SCSI command needs to be issued.   The idea is that
 *              we merely inject the command into the queue (at the head
 *              for now), and then call the queue request function to actually
 *              process it.
 */
int scsi_insert_special_req(struct scsi_request *sreq, int at_head)
{
	/*
	 * Because users of this function are apt to reuse requests with no
	 * modification, we have to sanitise the request flags here
	 */
	sreq->sr_request->flags &= ~REQ_DONTPREP;
	blk_insert_request(sreq->sr_device->request_queue, sreq->sr_request,
		       	   at_head, sreq);
	return 0;
}

static void scsi_run_queue(struct request_queue *q);

/*
 * Function:    scsi_queue_insert()
 *
 * Purpose:     Insert a command in the midlevel queue.
 *
 * Arguments:   cmd    - command that we are adding to queue.
 *              reason - why we are inserting command to queue.
 *
 * Lock status: Assumed that lock is not held upon entry.
 *
 * Returns:     Nothing.
 *
 * Notes:       We do this for one of two cases.  Either the host is busy
 *              and it cannot accept any more commands for the time being,
 *              or the device returned QUEUE_FULL and can accept no more
 *              commands.
 * Notes:       This could be called either from an interrupt context or a
 *              normal process context.
 */
int scsi_queue_insert(struct scsi_cmnd *cmd, int reason)
{
	struct Scsi_Host *host = cmd->device->host;
	struct scsi_device *device = cmd->device;
	struct request_queue *q = device->request_queue;
	unsigned long flags;

	SCSI_LOG_MLQUEUE(1,
		 printk("Inserting command %p into mlqueue\n", cmd));

	/*
	 * Set the appropriate busy bit for the device/host.
	 *
	 * If the host/device isn't busy, assume that something actually
	 * completed, and that we should be able to queue a command now.
	 *
	 * Note that the prior mid-layer assumption that any host could
	 * always queue at least one command is now broken.  The mid-layer
	 * will implement a user specifiable stall (see
	 * scsi_host.max_host_blocked and scsi_device.max_device_blocked)
	 * if a command is requeued with no other commands outstanding
	 * either for the device or for the host.
	 */
	if (reason == SCSI_MLQUEUE_HOST_BUSY)
		host->host_blocked = host->max_host_blocked;
	else if (reason == SCSI_MLQUEUE_DEVICE_BUSY)
		device->device_blocked = device->max_device_blocked;

	/*
	 * Register the fact that we own the thing for now.
	 */
	cmd->state = SCSI_STATE_MLQUEUE;
	cmd->owner = SCSI_OWNER_MIDLEVEL;

	/*
	 * Decrement the counters, since these commands are no longer
	 * active on the host/device.
	 */
	scsi_device_unbusy(device);

	/*
	 * Requeue this command.  It will go before all other commands
	 * that are already in the queue.
	 *
	 * NOTE: there is magic here about the way the queue is plugged if
	 * we have no outstanding commands.
	 * 
	 * Although we *don't* plug the queue, we call the request
	 * function.  The SCSI request function detects the blocked condition
	 * and plugs the queue appropriately.
         */
	spin_lock_irqsave(q->queue_lock, flags);
	blk_requeue_request(q, cmd->request);
	spin_unlock_irqrestore(q->queue_lock, flags);

	scsi_run_queue(q);

	return 0;
}

/*
 * Function:    scsi_do_req
 *
 * Purpose:     Queue a SCSI request
 *
 * Arguments:   sreq	  - command descriptor.
 *              cmnd      - actual SCSI command to be performed.
 *              buffer    - data buffer.
 *              bufflen   - size of data buffer.
 *              done      - completion function to be run.
 *              timeout   - how long to let it run before timeout.
 *              retries   - number of retries we allow.
 *
 * Lock status: No locks held upon entry.
 *
 * Returns:     Nothing.
 *
 * Notes:	This function is only used for queueing requests for things
 *		like ioctls and character device requests - this is because
 *		we essentially just inject a request into the queue for the
 *		device.
 *
 *		In order to support the scsi_device_quiesce function, we
 *		now inject requests on the *head* of the device queue
 *		rather than the tail.
 */
void scsi_do_req(struct scsi_request *sreq, const void *cmnd,
		 void *buffer, unsigned bufflen,
		 void (*done)(struct scsi_cmnd *),
		 int timeout, int retries)
{
	/*
	 * If the upper level driver is reusing these things, then
	 * we should release the low-level block now.  Another one will
	 * be allocated later when this request is getting queued.
	 */
	__scsi_release_request(sreq);

	/*
	 * Our own function scsi_done (which marks the host as not busy,
	 * disables the timeout counter, etc) will be called by us or by the
	 * scsi_hosts[host].queuecommand() function needs to also call
	 * the completion function for the high level driver.
	 */
	memcpy(sreq->sr_cmnd, cmnd, sizeof(sreq->sr_cmnd));
	sreq->sr_bufflen = bufflen;
	sreq->sr_buffer = buffer;
	sreq->sr_allowed = retries;
	sreq->sr_done = done;
	sreq->sr_timeout_per_command = timeout;

	if (sreq->sr_cmd_len == 0)
		sreq->sr_cmd_len = COMMAND_SIZE(sreq->sr_cmnd[0]);

	/*
	 * head injection *required* here otherwise quiesce won't work
	 */
	scsi_insert_special_req(sreq, 1);
}
EXPORT_SYMBOL(scsi_do_req);

static void scsi_wait_done(struct scsi_cmnd *cmd)
{
	struct request *req = cmd->request;
	struct request_queue *q = cmd->device->request_queue;
	unsigned long flags;

	req->rq_status = RQ_SCSI_DONE;	/* Busy, but indicate request done */

	spin_lock_irqsave(q->queue_lock, flags);
	if (blk_rq_tagged(req))
		blk_queue_end_tag(q, req);
	spin_unlock_irqrestore(q->queue_lock, flags);

	if (req->waiting)
		complete(req->waiting);
}

/* This is the end routine we get to if a command was never attached
 * to the request.  Simply complete the request without changing
 * rq_status; this will cause a DRIVER_ERROR. */
static void scsi_wait_req_end_io(struct request *req)
{
	BUG_ON(!req->waiting);

	complete(req->waiting);
}

void scsi_wait_req(struct scsi_request *sreq, const void *cmnd, void *buffer,
		   unsigned bufflen, int timeout, int retries)
{
	DECLARE_COMPLETION(wait);
	
	sreq->sr_request->waiting = &wait;
	sreq->sr_request->rq_status = RQ_SCSI_BUSY;
	sreq->sr_request->end_io = scsi_wait_req_end_io;
	scsi_do_req(sreq, cmnd, buffer, bufflen, scsi_wait_done,
			timeout, retries);
	wait_for_completion(&wait);
	sreq->sr_request->waiting = NULL;
	if (sreq->sr_request->rq_status != RQ_SCSI_DONE)
		sreq->sr_result |= (DRIVER_ERROR << 24);

	__scsi_release_request(sreq);
}
EXPORT_SYMBOL(scsi_wait_req);

/*
 * Function:    scsi_init_cmd_errh()
 *
 * Purpose:     Initialize cmd fields related to error handling.
 *
 * Arguments:   cmd	- command that is ready to be queued.
 *
 * Returns:     Nothing
 *
 * Notes:       This function has the job of initializing a number of
 *              fields related to error handling.   Typically this will
 *              be called once for each command, as required.
 */
static int scsi_init_cmd_errh(struct scsi_cmnd *cmd)
{
	cmd->owner = SCSI_OWNER_MIDLEVEL;
	cmd->serial_number = 0;
	cmd->abort_reason = 0;

	memset(cmd->sense_buffer, 0, sizeof cmd->sense_buffer);

	if (cmd->cmd_len == 0)
		cmd->cmd_len = COMMAND_SIZE(cmd->cmnd[0]);

	/*
	 * We need saved copies of a number of fields - this is because
	 * error handling may need to overwrite these with different values
	 * to run different commands, and once error handling is complete,
	 * we will need to restore these values prior to running the actual
	 * command.
	 */
	cmd->old_use_sg = cmd->use_sg;
	cmd->old_cmd_len = cmd->cmd_len;
	cmd->sc_old_data_direction = cmd->sc_data_direction;
	cmd->old_underflow = cmd->underflow;
	memcpy(cmd->data_cmnd, cmd->cmnd, sizeof(cmd->cmnd));
	cmd->buffer = cmd->request_buffer;
	cmd->bufflen = cmd->request_bufflen;
	cmd->abort_reason = 0;

	return 1;
}

/*
 * Function:   scsi_setup_cmd_retry()
 *
 * Purpose:    Restore the command state for a retry
 *
 * Arguments:  cmd	- command to be restored
 *
 * Returns:    Nothing
 *
 * Notes:      Immediately prior to retrying a command, we need
 *             to restore certain fields that we saved above.
 */
void scsi_setup_cmd_retry(struct scsi_cmnd *cmd)
{
	memcpy(cmd->cmnd, cmd->data_cmnd, sizeof(cmd->data_cmnd));
	cmd->request_buffer = cmd->buffer;
	cmd->request_bufflen = cmd->bufflen;
	cmd->use_sg = cmd->old_use_sg;
	cmd->cmd_len = cmd->old_cmd_len;
	cmd->sc_data_direction = cmd->sc_old_data_direction;
	cmd->underflow = cmd->old_underflow;
}

void scsi_device_unbusy(struct scsi_device *sdev)
{
	struct Scsi_Host *shost = sdev->host;
	unsigned long flags;

	spin_lock_irqsave(shost->host_lock, flags);
	shost->host_busy--;
	if (unlikely(test_bit(SHOST_RECOVERY, &shost->shost_state) &&
		     shost->host_failed))
		scsi_eh_wakeup(shost);
	spin_unlock(shost->host_lock);
	spin_lock(sdev->request_queue->queue_lock);
	sdev->device_busy--;
	spin_unlock_irqrestore(sdev->request_queue->queue_lock, flags);
}

/*
 * Called for single_lun devices on IO completion. Clear starget_sdev_user,
 * and call blk_run_queue for all the scsi_devices on the target -
 * including current_sdev first.
 *
 * Called with *no* scsi locks held.
 */
static void scsi_single_lun_run(struct scsi_device *current_sdev)
{
	struct Scsi_Host *shost = current_sdev->host;
	struct scsi_device *sdev, *tmp;
	struct scsi_target *starget = scsi_target(current_sdev);
	unsigned long flags;

	spin_lock_irqsave(shost->host_lock, flags);
	starget->starget_sdev_user = NULL;
	spin_unlock_irqrestore(shost->host_lock, flags);

	/*
	 * Call blk_run_queue for all LUNs on the target, starting with
	 * current_sdev. We race with others (to set starget_sdev_user),
	 * but in most cases, we will be first. Ideally, each LU on the
	 * target would get some limited time or requests on the target.
	 */
	blk_run_queue(current_sdev->request_queue);

	spin_lock_irqsave(shost->host_lock, flags);
	if (starget->starget_sdev_user)
		goto out;
	list_for_each_entry_safe(sdev, tmp, &starget->devices,
			same_target_siblings) {
		if (sdev == current_sdev)
			continue;
		if (scsi_device_get(sdev))
			continue;

		spin_unlock_irqrestore(shost->host_lock, flags);
		blk_run_queue(sdev->request_queue);
		spin_lock_irqsave(shost->host_lock, flags);
	
		scsi_device_put(sdev);
	}
 out:
	spin_unlock_irqrestore(shost->host_lock, flags);
}

/*
 * Function:	scsi_run_queue()
 *
 * Purpose:	Select a proper request queue to serve next
 *
 * Arguments:	q	- last request's queue
 *
 * Returns:     Nothing
 *
 * Notes:	The previous command was completely finished, start
 *		a new one if possible.
 */
static void scsi_run_queue(struct request_queue *q)
{
	struct scsi_device *sdev = q->queuedata;
	struct Scsi_Host *shost = sdev->host;
	unsigned long flags;

	if (sdev->single_lun)
		scsi_single_lun_run(sdev);

	spin_lock_irqsave(shost->host_lock, flags);
	while (!list_empty(&shost->starved_list) &&
	       !shost->host_blocked && !shost->host_self_blocked &&
		!((shost->can_queue > 0) &&
		  (shost->host_busy >= shost->can_queue))) {
		/*
		 * As long as shost is accepting commands and we have
		 * starved queues, call blk_run_queue. scsi_request_fn
		 * drops the queue_lock and can add us back to the
		 * starved_list.
		 *
		 * host_lock protects the starved_list and starved_entry.
		 * scsi_request_fn must get the host_lock before checking
		 * or modifying starved_list or starved_entry.
		 */
		sdev = list_entry(shost->starved_list.next,
					  struct scsi_device, starved_entry);
		list_del_init(&sdev->starved_entry);
		spin_unlock_irqrestore(shost->host_lock, flags);

		blk_run_queue(sdev->request_queue);

		spin_lock_irqsave(shost->host_lock, flags);
		if (unlikely(!list_empty(&sdev->starved_entry)))
			/*
			 * sdev lost a race, and was put back on the
			 * starved list. This is unlikely but without this
			 * in theory we could loop forever.
			 */
			break;
	}
	spin_unlock_irqrestore(shost->host_lock, flags);

	blk_run_queue(q);
}

/*
 * Function:	scsi_requeue_command()
 *
 * Purpose:	Handle post-processing of completed commands.
 *
 * Arguments:	q	- queue to operate on
 *		cmd	- command that may need to be requeued.
 *
 * Returns:	Nothing
 *
 * Notes:	After command completion, there may be blocks left
 *		over which weren't finished by the previous command
 *		this can be for a number of reasons - the main one is
 *		I/O errors in the middle of the request, in which case
 *		we need to request the blocks that come after the bad
 *		sector.
 */
static void scsi_requeue_command(struct request_queue *q, struct scsi_cmnd *cmd)
{
	unsigned long flags;

	cmd->request->flags &= ~REQ_DONTPREP;

	spin_lock_irqsave(q->queue_lock, flags);
	blk_requeue_request(q, cmd->request);
	spin_unlock_irqrestore(q->queue_lock, flags);

	scsi_run_queue(q);
}

void scsi_next_command(struct scsi_cmnd *cmd)
{
	struct request_queue *q = cmd->device->request_queue;

	scsi_put_command(cmd);
	scsi_run_queue(q);
}

void scsi_run_host_queues(struct Scsi_Host *shost)
{
	struct scsi_device *sdev;

	shost_for_each_device(sdev, shost)
		scsi_run_queue(sdev->request_queue);
}

/*
 * Function:    scsi_end_request()
 *
 * Purpose:     Post-processing of completed commands (usually invoked at end
 *		of upper level post-processing and scsi_io_completion).
 *
 * Arguments:   cmd	 - command that is complete.
 *              uptodate - 1 if I/O indicates success, <= 0 for I/O error.
 *              bytes    - number of bytes of completed I/O
 *		requeue  - indicates whether we should requeue leftovers.
 *
 * Lock status: Assumed that lock is not held upon entry.
 *
 * Returns:     cmd if requeue done or required, NULL otherwise
 *
 * Notes:       This is called for block device requests in order to
 *              mark some number of sectors as complete.
 * 
 *		We are guaranteeing that the request queue will be goosed
 *		at some point during this call.
 */
static struct scsi_cmnd *scsi_end_request(struct scsi_cmnd *cmd, int uptodate,
					  int bytes, int requeue)
{
	request_queue_t *q = cmd->device->request_queue;
	struct request *req = cmd->request;
	unsigned long flags;

	/*
	 * If there are blocks left over at the end, set up the command
	 * to queue the remainder of them.
	 */
	if (end_that_request_chunk(req, uptodate, bytes)) {
		int leftover = (req->hard_nr_sectors << 9);

		if (blk_pc_request(req))
			leftover = req->data_len;

		/* kill remainder if no retrys */
		if (!uptodate && blk_noretry_request(req))
			end_that_request_chunk(req, 0, leftover);
		else {
			if (requeue)
				/*
				 * Bleah.  Leftovers again.  Stick the
				 * leftovers in the front of the
				 * queue, and goose the queue again.
				 */
				scsi_requeue_command(q, cmd);

			return cmd;
		}
	}

	add_disk_randomness(req->rq_disk);

	spin_lock_irqsave(q->queue_lock, flags);
	if (blk_rq_tagged(req))
		blk_queue_end_tag(q, req);
	end_that_request_last(req);
	spin_unlock_irqrestore(q->queue_lock, flags);

	/*
	 * This will goose the queue request function at the end, so we don't
	 * need to worry about launching another command.
	 */
	scsi_next_command(cmd);
	return NULL;
}

static struct scatterlist *scsi_alloc_sgtable(struct scsi_cmnd *cmd, int gfp_mask)
{
	struct scsi_host_sg_pool *sgp;
	struct scatterlist *sgl;

	BUG_ON(!cmd->use_sg);

	switch (cmd->use_sg) {
	case 1 ... 8:
		cmd->sglist_len = 0;
		break;
	case 9 ... 16:
		cmd->sglist_len = 1;
		break;
	case 17 ... 32:
		cmd->sglist_len = 2;
		break;
#if (SCSI_MAX_PHYS_SEGMENTS > 32)
	case 33 ... 64:
		cmd->sglist_len = 3;
		break;
#if (SCSI_MAX_PHYS_SEGMENTS > 64)
	case 65 ... 128:
		cmd->sglist_len = 4;
		break;
#if (SCSI_MAX_PHYS_SEGMENTS  > 128)
	case 129 ... 256:
		cmd->sglist_len = 5;
		break;
#endif
#endif
#endif
	default:
		return NULL;
	}

	sgp = scsi_sg_pools + cmd->sglist_len;
	sgl = mempool_alloc(sgp->pool, gfp_mask);
	if (sgl)
		memset(sgl, 0, sgp->size);
	return sgl;
}

static void scsi_free_sgtable(struct scatterlist *sgl, int index)
{
	struct scsi_host_sg_pool *sgp;

	BUG_ON(index > SG_MEMPOOL_NR);

	sgp = scsi_sg_pools + index;
	mempool_free(sgl, sgp->pool);
}

/*
 * Function:    scsi_release_buffers()
 *
 * Purpose:     Completion processing for block device I/O requests.
 *
 * Arguments:   cmd	- command that we are bailing.
 *
 * Lock status: Assumed that no lock is held upon entry.
 *
 * Returns:     Nothing
 *
 * Notes:       In the event that an upper level driver rejects a
 *		command, we must release resources allocated during
 *		the __init_io() function.  Primarily this would involve
 *		the scatter-gather table, and potentially any bounce
 *		buffers.
 */
static void scsi_release_buffers(struct scsi_cmnd *cmd)
{
	struct request *req = cmd->request;

	/*
	 * Free up any indirection buffers we allocated for DMA purposes. 
	 */
	if (cmd->use_sg)
		scsi_free_sgtable(cmd->request_buffer, cmd->sglist_len);
	else if (cmd->request_buffer != req->buffer)
		kfree(cmd->request_buffer);

	/*
	 * Zero these out.  They now point to freed memory, and it is
	 * dangerous to hang onto the pointers.
	 */
	cmd->buffer  = NULL;
	cmd->bufflen = 0;
	cmd->request_buffer = NULL;
	cmd->request_bufflen = 0;
}

/*
 * Function:    scsi_io_completion()
 *
 * Purpose:     Completion processing for block device I/O requests.
 *
 * Arguments:   cmd   - command that is finished.
 *
 * Lock status: Assumed that no lock is held upon entry.
 *
 * Returns:     Nothing
 *
 * Notes:       This function is matched in terms of capabilities to
 *              the function that created the scatter-gather list.
 *              In other words, if there are no bounce buffers
 *              (the normal case for most drivers), we don't need
 *              the logic to deal with cleaning up afterwards.
 *
 *		We must do one of several things here:
 *
 *		a) Call scsi_end_request.  This will finish off the
 *		   specified number of sectors.  If we are done, the
 *		   command block will be released, and the queue
 *		   function will be goosed.  If we are not done, then
 *		   scsi_end_request will directly goose the queue.
 *
 *		b) We can just use scsi_requeue_command() here.  This would
 *		   be used if we just wanted to retry, for example.
 */
void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes,
			unsigned int block_bytes)
{
	int result = cmd->result;
	int this_count = cmd->bufflen;
	request_queue_t *q = cmd->device->request_queue;
	struct request *req = cmd->request;
	int clear_errors = 1;
	struct scsi_sense_hdr sshdr;
	int sense_valid = 0;
	int sense_deferred = 0;

	if (blk_complete_barrier_rq(q, req, good_bytes >> 9))
		return;

	/*
	 * Free up any indirection buffers we allocated for DMA purposes. 
	 * For the case of a READ, we need to copy the data out of the
	 * bounce buffer and into the real buffer.
	 */
	if (cmd->use_sg)
		scsi_free_sgtable(cmd->buffer, cmd->sglist_len);
	else if (cmd->buffer != req->buffer) {
		if (rq_data_dir(req) == READ) {
			unsigned long flags;
			char *to = bio_kmap_irq(req->bio, &flags);
			memcpy(to, cmd->buffer, cmd->bufflen);
			bio_kunmap_irq(to, &flags);
		}
		kfree(cmd->buffer);
	}

	if (result) {
		sense_valid = scsi_command_normalize_sense(cmd, &sshdr);
		if (sense_valid)
			sense_deferred = scsi_sense_is_deferred(&sshdr);
	}
	if (blk_pc_request(req)) { /* SG_IO ioctl from block level */
		req->errors = result;
		if (result) {
			clear_errors = 0;
			if (sense_valid && req->sense) {
				/*
				 * SG_IO wants current and deferred errors
				 */
				int len = 8 + cmd->sense_buffer[7];

				if (len > SCSI_SENSE_BUFFERSIZE)
					len = SCSI_SENSE_BUFFERSIZE;
				memcpy(req->sense, cmd->sense_buffer,  len);
				req->sense_len = len;
			}
		} else
			req->data_len = cmd->resid;
	}

	/*
	 * Zero these out.  They now point to freed memory, and it is
	 * dangerous to hang onto the pointers.
	 */
	cmd->buffer  = NULL;
	cmd->bufflen = 0;
	cmd->request_buffer = NULL;
	cmd->request_bufflen = 0;

	/*
	 * Next deal with any sectors which we were able to correctly
	 * handle.
	 */
	if (good_bytes >= 0) {
		SCSI_LOG_HLCOMPLETE(1, printk("%ld sectors total, %d bytes done.\n",
					      req->nr_sectors, good_bytes));
		SCSI_LOG_HLCOMPLETE(1, printk("use_sg is %d\n", cmd->use_sg));

		if (clear_errors)
			req->errors = 0;
		/*
		 * If multiple sectors are requested in one buffer, then
		 * they will have been finished off by the first command.
		 * If not, then we have a multi-buffer command.
		 *
		 * If block_bytes != 0, it means we had a medium error
		 * of some sort, and that we want to mark some number of
		 * sectors as not uptodate.  Thus we want to inhibit
		 * requeueing right here - we will requeue down below
		 * when we handle the bad sectors.
		 */
		cmd = scsi_end_request(cmd, 1, good_bytes, result == 0);

		/*
		 * If the command completed without error, then either finish off the
		 * rest of the command, or start a new one.
		 */
		if (result == 0 || cmd == NULL ) {
			return;
		}
	}
	/*
	 * Now, if we were good little boys and girls, Santa left us a request
	 * sense buffer.  We can extract information from this, so we
	 * can choose a block to remap, etc.
	 */
	if (sense_valid && !sense_deferred) {
		switch (sshdr.sense_key) {
		case UNIT_ATTENTION:
			if (cmd->device->removable) {
				/* detected disc change.  set a bit 
				 * and quietly refuse further access.
				 */
				cmd->device->changed = 1;
				cmd = scsi_end_request(cmd, 0,
						this_count, 1);
				return;
			} else {
				/*
				* Must have been a power glitch, or a
				* bus reset.  Could not have been a
				* media change, so we just retry the
				* request and see what happens.  
				*/
				scsi_requeue_command(q, cmd);
				return;
			}
			break;
		case ILLEGAL_REQUEST:
			/*
		 	* If we had an ILLEGAL REQUEST returned, then we may
		 	* have performed an unsupported command.  The only
		 	* thing this should be would be a ten byte read where
			* only a six byte read was supported.  Also, on a
			* system where READ CAPACITY failed, we may have read
			* past the end of the disk.
		 	*/
			if (cmd->device->use_10_for_rw &&
			    (cmd->cmnd[0] == READ_10 ||
			     cmd->cmnd[0] == WRITE_10)) {
				cmd->device->use_10_for_rw = 0;
				/*
				 * This will cause a retry with a 6-byte
				 * command.
				 */
				scsi_requeue_command(q, cmd);
				result = 0;
			} else {
				cmd = scsi_end_request(cmd, 0, this_count, 1);
				return;
			}
			break;
		case NOT_READY:
			/*
			 * If the device is in the process of becoming ready,
			 * retry.
			 */
			if (sshdr.asc == 0x04 && sshdr.ascq == 0x01) {
				scsi_requeue_command(q, cmd);
				return;
			}
			printk(KERN_INFO "Device %s not ready.\n",
			       req->rq_disk ? req->rq_disk->disk_name : "");
			cmd = scsi_end_request(cmd, 0, this_count, 1);
			return;
		case VOLUME_OVERFLOW:
			printk(KERN_INFO "Volume overflow <%d %d %d %d> CDB: ",
			       cmd->device->host->host_no,
			       (int)cmd->device->channel,
			       (int)cmd->device->id, (int)cmd->device->lun);
			__scsi_print_command(cmd->data_cmnd);
			scsi_print_sense("", cmd);
			cmd = scsi_end_request(cmd, 0, block_bytes, 1);
			return;
		default:
			break;
		}
	}			/* driver byte != 0 */
	if (host_byte(result) == DID_RESET) {
		/*
		 * Third party bus reset or reset for error
		 * recovery reasons.  Just retry the request
		 * and see what happens.  
		 */
		scsi_requeue_command(q, cmd);
		return;
	}
	if (result) {
		printk(KERN_INFO "SCSI error : <%d %d %d %d> return code "
		       "= 0x%x\n", cmd->device->host->host_no,
		       cmd->device->channel,
		       cmd->device->id,
		       cmd->device->lun, result);

		if (driver_byte(result) & DRIVER_SENSE)
			scsi_print_sense("", cmd);
		/*
		 * Mark a single buffer as not uptodate.  Queue the remainder.
		 * We sometimes get this cruft in the event that a medium error
		 * isn't properly reported.
		 */
		block_bytes = req->hard_cur_sectors << 9;
		if (!block_bytes)
			block_bytes = req->data_len;
		cmd = scsi_end_request(cmd, 0, block_bytes, 1);
	}
}
EXPORT_SYMBOL(scsi_io_completion);

/*
 * Function:    scsi_init_io()
 *
 * Purpose:     SCSI I/O initialize function.
 *
 * Arguments:   cmd   - Command descriptor we wish to initialize
 *
 * Returns:     0 on success
 *		BLKPREP_DEFER if the failure is retryable
 *		BLKPREP_KILL if the failure is fatal
 */
static int scsi_init_io(struct scsi_cmnd *cmd)
{
	struct request     *req = cmd->request;
	struct scatterlist *sgpnt;
	int		   count;

	/*
	 * if this is a rq->data based REQ_BLOCK_PC, setup for a non-sg xfer
	 */
	if ((req->flags & REQ_BLOCK_PC) && !req->bio) {
		cmd->request_bufflen = req->data_len;
		cmd->request_buffer = req->data;
		req->buffer = req->data;
		cmd->use_sg = 0;
		return 0;
	}

	/*
	 * we used to not use scatter-gather for single segment request,
	 * but now we do (it makes highmem I/O easier to support without
	 * kmapping pages)
	 */
	cmd->use_sg = req->nr_phys_segments;

	/*
	 * if sg table allocation fails, requeue request later.
	 */
	sgpnt = scsi_alloc_sgtable(cmd, GFP_ATOMIC);
	if (unlikely(!sgpnt))
		return BLKPREP_DEFER;

	cmd->request_buffer = (char *) sgpnt;
	cmd->request_bufflen = req->nr_sectors << 9;
	if (blk_pc_request(req))
		cmd->request_bufflen = req->data_len;
	req->buffer = NULL;

	/* 
	 * Next, walk the list, and fill in the addresses and sizes of
	 * each segment.
	 */
	count = blk_rq_map_sg(req->q, req, cmd->request_buffer);

	/*
	 * mapped well, send it off
	 */
	if (likely(count <= cmd->use_sg)) {
		cmd->use_sg = count;
		return 0;
	}

	printk(KERN_ERR "Incorrect number of segments after building list\n");
	printk(KERN_ERR "counted %d, received %d\n", count, cmd->use_sg);
	printk(KERN_ERR "req nr_sec %lu, cur_nr_sec %u\n", req->nr_sectors,
			req->current_nr_sectors);

	/* release the command and kill it */
	scsi_release_buffers(cmd);
	scsi_put_command(cmd);
	return BLKPREP_KILL;
}

static int scsi_prepare_flush_fn(request_queue_t *q, struct request *rq)
{
	struct scsi_device *sdev = q->queuedata;
	struct scsi_driver *drv;

	if (sdev->sdev_state == SDEV_RUNNING) {
		drv = *(struct scsi_driver **) rq->rq_disk->private_data;

		if (drv->prepare_flush)
			return drv->prepare_flush(q, rq);
	}

	return 0;
}

static void scsi_end_flush_fn(request_queue_t *q, struct request *rq)
{
	struct scsi_device *sdev = q->queuedata;
	struct request *flush_rq = rq->end_io_data;
	struct scsi_driver *drv;

	if (flush_rq->errors) {
		printk("scsi: barrier error, disabling flush support\n");
		blk_queue_ordered(q, QUEUE_ORDERED_NONE);
	}

	if (sdev->sdev_state == SDEV_RUNNING) {
		drv = *(struct scsi_driver **) rq->rq_disk->private_data;
		drv->end_flush(q, rq);
	}
}

static int scsi_issue_flush_fn(request_queue_t *q, struct gendisk *disk,
			       sector_t *error_sector)
{
	struct scsi_device *sdev = q->queuedata;
	struct scsi_driver *drv;

	if (sdev->sdev_state != SDEV_RUNNING)
		return -ENXIO;

	drv = *(struct scsi_driver **) disk->private_data;
	if (drv->issue_flush)
		return drv->issue_flush(&sdev->sdev_gendev, error_sector);

	return -EOPNOTSUPP;
}

static int scsi_prep_fn(struct request_queue *q, struct request *req)
{
	struct scsi_device *sdev = q->queuedata;
	struct scsi_cmnd *cmd;
	int specials_only = 0;

	/*
	 * Just check to see if the device is online.  If it isn't, we
	 * refuse to process any commands.  The device must be brought
	 * online before trying any recovery commands
	 */
	if (unlikely(!scsi_device_online(sdev))) {
		printk(KERN_ERR "scsi%d (%d:%d): rejecting I/O to offline device\n",
		       sdev->host->host_no, sdev->id, sdev->lun);
		return BLKPREP_KILL;
	}
	if (unlikely(sdev->sdev_state != SDEV_RUNNING)) {
		/* OK, we're not in a running state don't prep
		 * user commands */
		if (sdev->sdev_state == SDEV_DEL) {
			/* Device is fully deleted, no commands
			 * at all allowed down */
			printk(KERN_ERR "scsi%d (%d:%d): rejecting I/O to dead device\n",
			       sdev->host->host_no, sdev->id, sdev->lun);
			return BLKPREP_KILL;
		}
		/* OK, we only allow special commands (i.e. not
		 * user initiated ones */
		specials_only = sdev->sdev_state;
	}

	/*
	 * Find the actual device driver associated with this command.
	 * The SPECIAL requests are things like character device or
	 * ioctls, which did not originate from ll_rw_blk.  Note that
	 * the special field is also used to indicate the cmd for
	 * the remainder of a partially fulfilled request that can 
	 * come up when there is a medium error.  We have to treat
	 * these two cases differently.  We differentiate by looking
	 * at request->cmd, as this tells us the real story.
	 */
	if (req->flags & REQ_SPECIAL) {
		struct scsi_request *sreq = req->special;

		if (sreq->sr_magic == SCSI_REQ_MAGIC) {
			cmd = scsi_get_command(sreq->sr_device, GFP_ATOMIC);
			if (unlikely(!cmd))
				goto defer;
			scsi_init_cmd_from_req(cmd, sreq);
		} else
			cmd = req->special;
	} else if (req->flags & (REQ_CMD | REQ_BLOCK_PC)) {

		if(unlikely(specials_only)) {
			if(specials_only == SDEV_QUIESCE ||
					specials_only == SDEV_BLOCK)
				return BLKPREP_DEFER;
			
			printk(KERN_ERR "scsi%d (%d:%d): rejecting I/O to device being removed\n",
			       sdev->host->host_no, sdev->id, sdev->lun);
			return BLKPREP_KILL;
		}
			
			
		/*
		 * Now try and find a command block that we can use.
		 */
		if (!req->special) {
			cmd = scsi_get_command(sdev, GFP_ATOMIC);
			if (unlikely(!cmd))
				goto defer;
		} else
			cmd = req->special;
		
		/* pull a tag out of the request if we have one */
		cmd->tag = req->tag;
	} else {
		blk_dump_rq_flags(req, "SCSI bad req");
		return BLKPREP_KILL;
	}
	
	/* note the overloading of req->special.  When the tag
	 * is active it always means cmd.  If the tag goes
	 * back for re-queueing, it may be reset */
	req->special = cmd;
	cmd->request = req;
	
	/*
	 * FIXME: drop the lock here because the functions below
	 * expect to be called without the queue lock held.  Also,
	 * previously, we dequeued the request before dropping the
	 * lock.  We hope REQ_STARTED prevents anything untoward from
	 * happening now.
	 */
	if (req->flags & (REQ_CMD | REQ_BLOCK_PC)) {
		struct scsi_driver *drv;
		int ret;

		/*
		 * This will do a couple of things:
		 *  1) Fill in the actual SCSI command.
		 *  2) Fill in any other upper-level specific fields
		 * (timeout).
		 *
		 * If this returns 0, it means that the request failed
		 * (reading past end of disk, reading offline device,
		 * etc).   This won't actually talk to the device, but
		 * some kinds of consistency checking may cause the	
		 * request to be rejected immediately.
		 */

		/* 
		 * This sets up the scatter-gather table (allocating if
		 * required).
		 */
		ret = scsi_init_io(cmd);
		if (ret)	/* BLKPREP_KILL return also releases the command */
			return ret;
		
		/*
		 * Initialize the actual SCSI command for this request.
		 */
		drv = *(struct scsi_driver **)req->rq_disk->private_data;
		if (unlikely(!drv->init_command(cmd))) {
			scsi_release_buffers(cmd);
			scsi_put_command(cmd);
			return BLKPREP_KILL;
		}
	}

	/*
	 * The request is now prepped, no need to come back here
	 */
	req->flags |= REQ_DONTPREP;
	return BLKPREP_OK;

 defer:
	/* If we defer, the elv_next_request() returns NULL, but the
	 * queue must be restarted, so we plug here if no returning
	 * command will automatically do that. */
	if (sdev->device_busy == 0)
		blk_plug_device(q);
	return BLKPREP_DEFER;
}

/*
 * scsi_dev_queue_ready: if we can send requests to sdev, return 1 else
 * return 0.
 *
 * Called with the queue_lock held.
 */
static inline int scsi_dev_queue_ready(struct request_queue *q,
				  struct scsi_device *sdev)
{
	if (sdev->device_busy >= sdev->queue_depth)
		return 0;
	if (sdev->device_busy == 0 && sdev->device_blocked) {
		/*
		 * unblock after device_blocked iterates to zero
		 */
		if (--sdev->device_blocked == 0) {
			SCSI_LOG_MLQUEUE(3,
				printk("scsi%d (%d:%d) unblocking device at"
				       " zero depth\n", sdev->host->host_no,
				       sdev->id, sdev->lun));
		} else {
			blk_plug_device(q);
			return 0;
		}
	}
	if (sdev->device_blocked)
		return 0;

	return 1;
}

/*
 * scsi_host_queue_ready: if we can send requests to shost, return 1 else
 * return 0. We must end up running the queue again whenever 0 is
 * returned, else IO can hang.
 *
 * Called with host_lock held.
 */
static inline int scsi_host_queue_ready(struct request_queue *q,
				   struct Scsi_Host *shost,
				   struct scsi_device *sdev)
{
	if (test_bit(SHOST_RECOVERY, &shost->shost_state))
		return 0;
	if (shost->host_busy == 0 && shost->host_blocked) {
		/*
		 * unblock after host_blocked iterates to zero
		 */
		if (--shost->host_blocked == 0) {
			SCSI_LOG_MLQUEUE(3,
				printk("scsi%d unblocking host at zero depth\n",
					shost->host_no));
		} else {
			blk_plug_device(q);
			return 0;
		}
	}
	if ((shost->can_queue > 0 && shost->host_busy >= shost->can_queue) ||
	    shost->host_blocked || shost->host_self_blocked) {
		if (list_empty(&sdev->starved_entry))
			list_add_tail(&sdev->starved_entry, &shost->starved_list);
		return 0;
	}

	/* We're OK to process the command, so we can't be starved */
	if (!list_empty(&sdev->starved_entry))
		list_del_init(&sdev->starved_entry);

	return 1;
}

/*
 * Kill requests for a dead device
 */
static void scsi_kill_requests(request_queue_t *q)
{
	struct request *req;

	while ((req = elv_next_request(q)) != NULL) {
		blkdev_dequeue_request(req);
		req->flags |= REQ_QUIET;
		while (end_that_request_first(req, 0, req->nr_sectors))
			;
		end_that_request_last(req);
	}
}

/*
 * Function:    scsi_request_fn()
 *
 * Purpose:     Main strategy routine for SCSI.
 *
 * Arguments:   q       - Pointer to actual queue.
 *
 * Returns:     Nothing
 *
 * Lock status: IO request lock assumed to be held when called.
 */
static void scsi_request_fn(struct request_queue *q)
{
	struct scsi_device *sdev = q->queuedata;
	struct Scsi_Host *shost;
	struct scsi_cmnd *cmd;
	struct request *req;

	if (!sdev) {
		printk("scsi: killing requests for dead queue\n");
		scsi_kill_requests(q);
		return;
	}

	if(!get_device(&sdev->sdev_gendev))
		/* We must be tearing the block queue down already */
		return;

	/*
	 * To start with, we keep looping until the queue is empty, or until
	 * the host is no longer able to accept any more requests.
	 */
	shost = sdev->host;
	while (!blk_queue_plugged(q)) {
		int rtn;
		/*
		 * get next queueable request.  We do this early to make sure
		 * that the request is fully prepared even if we cannot 
		 * accept it.
		 */
		req = elv_next_request(q);
		if (!req || !scsi_dev_queue_ready(q, sdev))
			break;

		if (unlikely(!scsi_device_online(sdev))) {
			printk(KERN_ERR "scsi%d (%d:%d): rejecting I/O to offline device\n",
			       sdev->host->host_no, sdev->id, sdev->lun);
			blkdev_dequeue_request(req);
			req->flags |= REQ_QUIET;
			while (end_that_request_first(req, 0, req->nr_sectors))
				;
			end_that_request_last(req);
			continue;
		}


		/*
		 * Remove the request from the request list.
		 */
		if (!(blk_queue_tagged(q) && !blk_queue_start_tag(q, req)))
			blkdev_dequeue_request(req);
		sdev->device_busy++;

		spin_unlock(q->queue_lock);
		spin_lock(shost->host_lock);

		if (!scsi_host_queue_ready(q, shost, sdev))
			goto not_ready;
		if (sdev->single_lun) {
			if (scsi_target(sdev)->starget_sdev_user &&
			    scsi_target(sdev)->starget_sdev_user != sdev)
				goto not_ready;
			scsi_target(sdev)->starget_sdev_user = sdev;
		}
		shost->host_busy++;

		/*
		 * XXX(hch): This is rather suboptimal, scsi_dispatch_cmd will
		 *		take the lock again.
		 */
		spin_unlock_irq(shost->host_lock);

		cmd = req->special;
		if (unlikely(cmd == NULL)) {
			printk(KERN_CRIT "impossible request in %s.\n"
					 "please mail a stack trace to "
					 "linux-scsi@vger.kernel.org",
					 __FUNCTION__);
			BUG();
		}

		/*
		 * Finally, initialize any error handling parameters, and set up
		 * the timers for timeouts.
		 */
		scsi_init_cmd_errh(cmd);

		/*
		 * Dispatch the command to the low-level driver.
		 */
		rtn = scsi_dispatch_cmd(cmd);
		spin_lock_irq(q->queue_lock);
		if(rtn) {
			/* we're refusing the command; because of
			 * the way locks get dropped, we need to 
			 * check here if plugging is required */
			if(sdev->device_busy == 0)
				blk_plug_device(q);

			break;
		}
	}

	goto out;

 not_ready:
	spin_unlock_irq(shost->host_lock);

	/*
	 * lock q, handle tag, requeue req, and decrement device_busy. We
	 * must return with queue_lock held.
	 *
	 * Decrementing device_busy without checking it is OK, as all such
	 * cases (host limits or settings) should run the queue at some
	 * later time.
	 */
	spin_lock_irq(q->queue_lock);
	blk_requeue_request(q, req);
	sdev->device_busy--;
	if(sdev->device_busy == 0)
		blk_plug_device(q);
 out:
	/* must be careful here...if we trigger the ->remove() function
	 * we cannot be holding the q lock */
	spin_unlock_irq(q->queue_lock);
	put_device(&sdev->sdev_gendev);
	spin_lock_irq(q->queue_lock);
}

u64 scsi_calculate_bounce_limit(struct Scsi_Host *shost)
{
	struct device *host_dev;
	u64 bounce_limit = 0xffffffff;

	if (shost->unchecked_isa_dma)
		return BLK_BOUNCE_ISA;
	/*
	 * Platforms with virtual-DMA translation
	 * hardware have no practical limit.
	 */
	if (!PCI_DMA_BUS_IS_PHYS)
		return BLK_BOUNCE_ANY;

	host_dev = scsi_get_device(shost);
	if (host_dev && host_dev->dma_mask)
		bounce_limit = *host_dev->dma_mask;

	return bounce_limit;
}
EXPORT_SYMBOL(scsi_calculate_bounce_limit);

struct request_queue *scsi_alloc_queue(struct scsi_device *sdev)
{
	struct Scsi_Host *shost = sdev->host;
	struct request_queue *q;

	q = blk_init_queue(scsi_request_fn, NULL);
	if (!q)
		return NULL;

	blk_queue_prep_rq(q, scsi_prep_fn);

	blk_queue_max_hw_segments(q, shost->sg_tablesize);
	blk_queue_max_phys_segments(q, SCSI_MAX_PHYS_SEGMENTS);
	blk_queue_max_sectors(q, shost->max_sectors);
	blk_queue_bounce_limit(q, scsi_calculate_bounce_limit(shost));
	blk_queue_segment_boundary(q, shost->dma_boundary);
	blk_queue_issue_flush_fn(q, scsi_issue_flush_fn);

	/*
	 * ordered tags are superior to flush ordering
	 */
	if (shost->ordered_tag)
		blk_queue_ordered(q, QUEUE_ORDERED_TAG);
	else if (shost->ordered_flush) {
		blk_queue_ordered(q, QUEUE_ORDERED_FLUSH);
		q->prepare_flush_fn = scsi_prepare_flush_fn;
		q->end_flush_fn = scsi_end_flush_fn;
	}

	if (!shost->use_clustering)
		clear_bit(QUEUE_FLAG_CLUSTER, &q->queue_flags);
	return q;
}

void scsi_free_queue(struct request_queue *q)
{
	blk_cleanup_queue(q);
}

/*
 * Function:    scsi_block_requests()
 *
 * Purpose:     Utility function used by low-level drivers to prevent further
 *		commands from being queued to the device.
 *
 * Arguments:   shost       - Host in question
 *
 * Returns:     Nothing
 *
 * Lock status: No locks are assumed held.
 *
 * Notes:       There is no timer nor any other means by which the requests
 *		get unblocked other than the low-level driver calling
 *		scsi_unblock_requests().
 */
void scsi_block_requests(struct Scsi_Host *shost)
{
	shost->host_self_blocked = 1;
}
EXPORT_SYMBOL(scsi_block_requests);

/*
 * Function:    scsi_unblock_requests()
 *
 * Purpose:     Utility function used by low-level drivers to allow further
 *		commands from being queued to the device.
 *
 * Arguments:   shost       - Host in question
 *
 * Returns:     Nothing
 *
 * Lock status: No locks are assumed held.
 *
 * Notes:       There is no timer nor any other means by which the requests
 *		get unblocked other than the low-level driver calling
 *		scsi_unblock_requests().
 *
 *		This is done as an API function so that changes to the
 *		internals of the scsi mid-layer won't require wholesale
 *		changes to drivers that use this feature.
 */
void scsi_unblock_requests(struct Scsi_Host *shost)
{
	shost->host_self_blocked = 0;
	scsi_run_host_queues(shost);
}
EXPORT_SYMBOL(scsi_unblock_requests);

int __init scsi_init_queue(void)
{
	int i;

	for (i = 0; i < SG_MEMPOOL_NR; i++) {
		struct scsi_host_sg_pool *sgp = scsi_sg_pools + i;
		int size = sgp->size * sizeof(struct scatterlist);

		sgp->slab = kmem_cache_create(sgp->name, size, 0,
				SLAB_HWCACHE_ALIGN, NULL, NULL);
		if (!sgp->slab) {
			printk(KERN_ERR "SCSI: can't init sg slab %s\n",
					sgp->name);
		}

		sgp->pool = mempool_create(SG_MEMPOOL_SIZE,
				mempool_alloc_slab, mempool_free_slab,
				sgp->slab);
		if (!sgp->pool) {
			printk(KERN_ERR "SCSI: can't init sg mempool %s\n",
					sgp->name);
		}
	}

	return 0;
}

void scsi_exit_queue(void)
{
	int i;

	for (i = 0; i < SG_MEMPOOL_NR; i++) {
		struct scsi_host_sg_pool *sgp = scsi_sg_pools + i;
		mempool_destroy(sgp->pool);
		kmem_cache_destroy(sgp->slab);
	}
}
/**
 *	__scsi_mode_sense - issue a mode sense, falling back from 10 to 
 *		six bytes if necessary.
 *	@sreq:	SCSI request to fill in with the MODE_SENSE
 *	@dbd:	set if mode sense will allow block descriptors to be returned
 *	@modepage: mode page being requested
 *	@buffer: request buffer (may not be smaller than eight bytes)
 *	@len:	length of request buffer.
 *	@timeout: command timeout
 *	@retries: number of retries before failing
 *	@data: returns a structure abstracting the mode header data
 *
 *	Returns zero if unsuccessful, or the header offset (either 4
 *	or 8 depending on whether a six or ten byte command was
 *	issued) if successful.
 **/
int
__scsi_mode_sense(struct scsi_request *sreq, int dbd, int modepage,
		  unsigned char *buffer, int len, int timeout, int retries,
		  struct scsi_mode_data *data) {
	unsigned char cmd[12];
	int use_10_for_ms;
	int header_length;

	memset(data, 0, sizeof(*data));
	memset(&cmd[0], 0, 12);
	cmd[1] = dbd & 0x18;	/* allows DBD and LLBA bits */
	cmd[2] = modepage;

 retry:
	use_10_for_ms = sreq->sr_device->use_10_for_ms;

	if (use_10_for_ms) {
		if (len < 8)
			len = 8;

		cmd[0] = MODE_SENSE_10;
		cmd[8] = len;
		header_length = 8;
	} else {
		if (len < 4)
			len = 4;

		cmd[0] = MODE_SENSE;
		cmd[4] = len;
		header_length = 4;
	}

	sreq->sr_cmd_len = 0;
	memset(sreq->sr_sense_buffer, 0, sizeof(sreq->sr_sense_buffer));
	sreq->sr_data_direction = DMA_FROM_DEVICE;

	memset(buffer, 0, len);

	scsi_wait_req(sreq, cmd, buffer, len, timeout, retries);

	/* This code looks awful: what it's doing is making sure an
	 * ILLEGAL REQUEST sense return identifies the actual command
	 * byte as the problem.  MODE_SENSE commands can return
	 * ILLEGAL REQUEST if the code page isn't supported */

	if (use_10_for_ms && !scsi_status_is_good(sreq->sr_result) &&
	    (driver_byte(sreq->sr_result) & DRIVER_SENSE)) {
		struct scsi_sense_hdr sshdr;

		if (scsi_request_normalize_sense(sreq, &sshdr)) {
			if ((sshdr.sense_key == ILLEGAL_REQUEST) &&
			    (sshdr.asc == 0x20) && (sshdr.ascq == 0)) {
				/* 
				 * Invalid command operation code
				 */
				sreq->sr_device->use_10_for_ms = 0;
				goto retry;
			}
		}
	}

	if(scsi_status_is_good(sreq->sr_result)) {
		data->header_length = header_length;
		if(use_10_for_ms) {
			data->length = buffer[0]*256 + buffer[1] + 2;
			data->medium_type = buffer[2];
			data->device_specific = buffer[3];
			data->longlba = buffer[4] & 0x01;
			data->block_descriptor_length = buffer[6]*256
				+ buffer[7];
		} else {
			data->length = buffer[0] + 1;
			data->medium_type = buffer[1];
			data->device_specific = buffer[2];
			data->block_descriptor_length = buffer[3];
		}
	}

	return sreq->sr_result;
}
EXPORT_SYMBOL(__scsi_mode_sense);

/**
 *	scsi_mode_sense - issue a mode sense, falling back from 10 to 
 *		six bytes if necessary.
 *	@sdev:	scsi device to send command to.
 *	@dbd:	set if mode sense will disable block descriptors in the return
 *	@modepage: mode page being requested
 *	@buffer: request buffer (may not be smaller than eight bytes)
 *	@len:	length of request buffer.
 *	@timeout: command timeout
 *	@retries: number of retries before failing
 *
 *	Returns zero if unsuccessful, or the header offset (either 4
 *	or 8 depending on whether a six or ten byte command was
 *	issued) if successful.
 **/
int
scsi_mode_sense(struct scsi_device *sdev, int dbd, int modepage,
		unsigned char *buffer, int len, int timeout, int retries,
		struct scsi_mode_data *data)
{
	struct scsi_request *sreq = scsi_allocate_request(sdev, GFP_KERNEL);
	int ret;

	if (!sreq)
		return -1;

	ret = __scsi_mode_sense(sreq, dbd, modepage, buffer, len,
				timeout, retries, data);

	scsi_release_request(sreq);

	return ret;
}
EXPORT_SYMBOL(scsi_mode_sense);

int
scsi_test_unit_ready(struct scsi_device *sdev, int timeout, int retries)
{
	struct scsi_request *sreq;
	char cmd[] = {
		TEST_UNIT_READY, 0, 0, 0, 0, 0,
	};
	int result;
	
	sreq = scsi_allocate_request(sdev, GFP_KERNEL);
	if (!sreq)
		return -ENOMEM;

	sreq->sr_data_direction = DMA_NONE;
	scsi_wait_req(sreq, cmd, NULL, 0, timeout, retries);

	if ((driver_byte(sreq->sr_result) & DRIVER_SENSE) && sdev->removable) {
		struct scsi_sense_hdr sshdr;

		if ((scsi_request_normalize_sense(sreq, &sshdr)) &&
		    ((sshdr.sense_key == UNIT_ATTENTION) ||
		     (sshdr.sense_key == NOT_READY))) {
			sdev->changed = 1;
			sreq->sr_result = 0;
		}
	}
	result = sreq->sr_result;
	scsi_release_request(sreq);
	return result;
}
EXPORT_SYMBOL(scsi_test_unit_ready);

/**
 *	scsi_device_set_state - Take the given device through the device
 *		state model.
 *	@sdev:	scsi device to change the state of.
 *	@state:	state to change to.
 *
 *	Returns zero if unsuccessful or an error if the requested 
 *	transition is illegal.
 **/
int
scsi_device_set_state(struct scsi_device *sdev, enum scsi_device_state state)
{
	enum scsi_device_state oldstate = sdev->sdev_state;

	if (state == oldstate)
		return 0;

	switch (state) {
	case SDEV_CREATED:
		/* There are no legal states that come back to
		 * created.  This is the manually initialised start
		 * state */
		goto illegal;
			
	case SDEV_RUNNING:
		switch (oldstate) {
		case SDEV_CREATED:
		case SDEV_OFFLINE:
		case SDEV_QUIESCE:
		case SDEV_BLOCK:
			break;
		default:
			goto illegal;
		}
		break;

	case SDEV_QUIESCE:
		switch (oldstate) {
		case SDEV_RUNNING:
		case SDEV_OFFLINE:
			break;
		default:
			goto illegal;
		}
		break;

	case SDEV_OFFLINE:
		switch (oldstate) {
		case SDEV_CREATED:
		case SDEV_RUNNING:
		case SDEV_QUIESCE:
		case SDEV_BLOCK:
			break;
		default:
			goto illegal;
		}
		break;

	case SDEV_BLOCK:
		switch (oldstate) {
		case SDEV_CREATED:
		case SDEV_RUNNING:
			break;
		default:
			goto illegal;
		}
		break;

	case SDEV_CANCEL:
		switch (oldstate) {
		case SDEV_CREATED:
		case SDEV_RUNNING:
		case SDEV_OFFLINE:
		case SDEV_BLOCK:
			break;
		default:
			goto illegal;
		}
		break;

	case SDEV_DEL:
		switch (oldstate) {
		case SDEV_CANCEL:
			break;
		default:
			goto illegal;
		}
		break;

	}
	sdev->sdev_state = state;
	return 0;

 illegal:
	SCSI_LOG_ERROR_RECOVERY(1, 
				dev_printk(KERN_ERR, &sdev->sdev_gendev,
					   "Illegal state transition %s->%s\n",
					   scsi_device_state_name(oldstate),
					   scsi_device_state_name(state))
				);
	return -EINVAL;
}
EXPORT_SYMBOL(scsi_device_set_state);

/**
 *	scsi_device_quiesce - Block user issued commands.
 *	@sdev:	scsi device to quiesce.
 *
 *	This works by trying to transition to the SDEV_QUIESCE state
 *	(which must be a legal transition).  When the device is in this
 *	state, only special requests will be accepted, all others will
 *	be deferred.  Since special requests may also be requeued requests,
 *	a successful return doesn't guarantee the device will be 
 *	totally quiescent.
 *
 *	Must be called with user context, may sleep.
 *
 *	Returns zero if unsuccessful or an error if not.
 **/
int
scsi_device_quiesce(struct scsi_device *sdev)
{
	int err = scsi_device_set_state(sdev, SDEV_QUIESCE);
	if (err)
		return err;

	scsi_run_queue(sdev->request_queue);
	while (sdev->device_busy) {
		msleep_interruptible(200);
		scsi_run_queue(sdev->request_queue);
	}
	return 0;
}
EXPORT_SYMBOL(scsi_device_quiesce);

/**
 *	scsi_device_resume - Restart user issued commands to a quiesced device.
 *	@sdev:	scsi device to resume.
 *
 *	Moves the device from quiesced back to running and restarts the
 *	queues.
 *
 *	Must be called with user context, may sleep.
 **/
void
scsi_device_resume(struct scsi_device *sdev)
{
	if(scsi_device_set_state(sdev, SDEV_RUNNING))
		return;
	scsi_run_queue(sdev->request_queue);
}
EXPORT_SYMBOL(scsi_device_resume);

static void
device_quiesce_fn(struct scsi_device *sdev, void *data)
{
	scsi_device_quiesce(sdev);
}

void
scsi_target_quiesce(struct scsi_target *starget)
{
	starget_for_each_device(starget, NULL, device_quiesce_fn);
}
EXPORT_SYMBOL(scsi_target_quiesce);

static void
device_resume_fn(struct scsi_device *sdev, void *data)
{
	scsi_device_resume(sdev);
}

void
scsi_target_resume(struct scsi_target *starget)
{
	starget_for_each_device(starget, NULL, device_resume_fn);
}
EXPORT_SYMBOL(scsi_target_resume);

/**
 * scsi_internal_device_block - internal function to put a device
 *				temporarily into the SDEV_BLOCK state
 * @sdev:	device to block
 *
 * Block request made by scsi lld's to temporarily stop all
 * scsi commands on the specified device.  Called from interrupt
 * or normal process context.
 *
 * Returns zero if successful or error if not
 *
 * Notes:       
 *	This routine transitions the device to the SDEV_BLOCK state
 *	(which must be a legal transition).  When the device is in this
 *	state, all commands are deferred until the scsi lld reenables
 *	the device with scsi_device_unblock or device_block_tmo fires.
 *	This routine assumes the host_lock is held on entry.
 **/
int
scsi_internal_device_block(struct scsi_device *sdev)
{
	request_queue_t *q = sdev->request_queue;
	unsigned long flags;
	int err = 0;

	err = scsi_device_set_state(sdev, SDEV_BLOCK);
	if (err)
		return err;

	/* 
	 * The device has transitioned to SDEV_BLOCK.  Stop the
	 * block layer from calling the midlayer with this device's
	 * request queue. 
	 */
	spin_lock_irqsave(q->queue_lock, flags);
	blk_stop_queue(q);
	spin_unlock_irqrestore(q->queue_lock, flags);

	return 0;
}
EXPORT_SYMBOL_GPL(scsi_internal_device_block);
 
/**
 * scsi_internal_device_unblock - resume a device after a block request
 * @sdev:	device to resume
 *
 * Called by scsi lld's or the midlayer to restart the device queue
 * for the previously suspended scsi device.  Called from interrupt or
 * normal process context.
 *
 * Returns zero if successful or error if not.
 *
 * Notes:       
 *	This routine transitions the device to the SDEV_RUNNING state
 *	(which must be a legal transition) allowing the midlayer to
 *	goose the queue for this device.  This routine assumes the 
 *	host_lock is held upon entry.
 **/
int
scsi_internal_device_unblock(struct scsi_device *sdev)
{
	request_queue_t *q = sdev->request_queue; 
	int err;
	unsigned long flags;
	
	/* 
	 * Try to transition the scsi device to SDEV_RUNNING
	 * and goose the device queue if successful.  
	 */
	err = scsi_device_set_state(sdev, SDEV_RUNNING);
	if (err)
		return err;

	spin_lock_irqsave(q->queue_lock, flags);
	blk_start_queue(q);
	spin_unlock_irqrestore(q->queue_lock, flags);

	return 0;
}
EXPORT_SYMBOL_GPL(scsi_internal_device_unblock);

static void
device_block(struct scsi_device *sdev, void *data)
{
	scsi_internal_device_block(sdev);
}

static int
target_block(struct device *dev, void *data)
{
	if (scsi_is_target_device(dev))
		starget_for_each_device(to_scsi_target(dev), NULL,
					device_block);
	return 0;
}

void
scsi_target_block(struct device *dev)
{
	if (scsi_is_target_device(dev))
		starget_for_each_device(to_scsi_target(dev), NULL,
					device_block);
	else
		device_for_each_child(dev, NULL, target_block);
}
EXPORT_SYMBOL_GPL(scsi_target_block);

static void
device_unblock(struct scsi_device *sdev, void *data)
{
	scsi_internal_device_unblock(sdev);
}

static int
target_unblock(struct device *dev, void *data)
{
	if (scsi_is_target_device(dev))
		starget_for_each_device(to_scsi_target(dev), NULL,
					device_unblock);
	return 0;
}

void
scsi_target_unblock(struct device *dev)
{
	if (scsi_is_target_device(dev))
		starget_for_each_device(to_scsi_target(dev), NULL,
					device_unblock);
	else
		device_for_each_child(dev, NULL, target_unblock);
}
EXPORT_SYMBOL_GPL(scsi_target_unblock);