summaryrefslogtreecommitdiff
path: root/drivers/dma/ioat/dma_v3.c
blob: 895f869d6c2c71dd1bb55de32ff4c11dbcb990ef (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
/*
 * This file is provided under a dual BSD/GPLv2 license.  When using or
 * redistributing this file, you may do so under either license.
 *
 * GPL LICENSE SUMMARY
 *
 * Copyright(c) 2004 - 2009 Intel 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.  See the GNU General Public License for
 * more details.
 *
 * You should have received a copy of the GNU General Public License along with
 * this program; if not, write to the Free Software Foundation, Inc.,
 * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
 *
 * The full GNU General Public License is included in this distribution in
 * the file called "COPYING".
 *
 * BSD LICENSE
 *
 * Copyright(c) 2004-2009 Intel Corporation. All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 *
 *   * Redistributions of source code must retain the above copyright
 *     notice, this list of conditions and the following disclaimer.
 *   * Redistributions in binary form must reproduce the above copyright
 *     notice, this list of conditions and the following disclaimer in
 *     the documentation and/or other materials provided with the
 *     distribution.
 *   * Neither the name of Intel Corporation nor the names of its
 *     contributors may be used to endorse or promote products derived
 *     from this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
 * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
 * POSSIBILITY OF SUCH DAMAGE.
 */

/*
 * Support routines for v3+ hardware
 */
#include <linux/module.h>
#include <linux/pci.h>
#include <linux/gfp.h>
#include <linux/dmaengine.h>
#include <linux/dma-mapping.h>
#include <linux/prefetch.h>
#include "../dmaengine.h"
#include "registers.h"
#include "hw.h"
#include "dma.h"
#include "dma_v2.h"

extern struct kmem_cache *ioat3_sed_cache;

/* ioat hardware assumes at least two sources for raid operations */
#define src_cnt_to_sw(x) ((x) + 2)
#define src_cnt_to_hw(x) ((x) - 2)
#define ndest_to_sw(x) ((x) + 1)
#define ndest_to_hw(x) ((x) - 1)
#define src16_cnt_to_sw(x) ((x) + 9)
#define src16_cnt_to_hw(x) ((x) - 9)

/* provide a lookup table for setting the source address in the base or
 * extended descriptor of an xor or pq descriptor
 */
static const u8 xor_idx_to_desc = 0xe0;
static const u8 xor_idx_to_field[] = { 1, 4, 5, 6, 7, 0, 1, 2 };
static const u8 pq_idx_to_desc = 0xf8;
static const u8 pq16_idx_to_desc[] = { 0, 0, 1, 1, 1, 1, 1, 1, 1,
				       2, 2, 2, 2, 2, 2, 2 };
static const u8 pq_idx_to_field[] = { 1, 4, 5, 0, 1, 2, 4, 5 };
static const u8 pq16_idx_to_field[] = { 1, 4, 1, 2, 3, 4, 5, 6, 7,
					0, 1, 2, 3, 4, 5, 6 };

static void ioat3_eh(struct ioat2_dma_chan *ioat);

static void xor_set_src(struct ioat_raw_descriptor *descs[2],
			dma_addr_t addr, u32 offset, int idx)
{
	struct ioat_raw_descriptor *raw = descs[xor_idx_to_desc >> idx & 1];

	raw->field[xor_idx_to_field[idx]] = addr + offset;
}

static dma_addr_t pq_get_src(struct ioat_raw_descriptor *descs[2], int idx)
{
	struct ioat_raw_descriptor *raw = descs[pq_idx_to_desc >> idx & 1];

	return raw->field[pq_idx_to_field[idx]];
}

static dma_addr_t pq16_get_src(struct ioat_raw_descriptor *desc[3], int idx)
{
	struct ioat_raw_descriptor *raw = desc[pq16_idx_to_desc[idx]];

	return raw->field[pq16_idx_to_field[idx]];
}

static void pq_set_src(struct ioat_raw_descriptor *descs[2],
		       dma_addr_t addr, u32 offset, u8 coef, int idx)
{
	struct ioat_pq_descriptor *pq = (struct ioat_pq_descriptor *) descs[0];
	struct ioat_raw_descriptor *raw = descs[pq_idx_to_desc >> idx & 1];

	raw->field[pq_idx_to_field[idx]] = addr + offset;
	pq->coef[idx] = coef;
}

static bool is_jf_ioat(struct pci_dev *pdev)
{
	switch (pdev->device) {
	case PCI_DEVICE_ID_INTEL_IOAT_JSF0:
	case PCI_DEVICE_ID_INTEL_IOAT_JSF1:
	case PCI_DEVICE_ID_INTEL_IOAT_JSF2:
	case PCI_DEVICE_ID_INTEL_IOAT_JSF3:
	case PCI_DEVICE_ID_INTEL_IOAT_JSF4:
	case PCI_DEVICE_ID_INTEL_IOAT_JSF5:
	case PCI_DEVICE_ID_INTEL_IOAT_JSF6:
	case PCI_DEVICE_ID_INTEL_IOAT_JSF7:
	case PCI_DEVICE_ID_INTEL_IOAT_JSF8:
	case PCI_DEVICE_ID_INTEL_IOAT_JSF9:
		return true;
	default:
		return false;
	}
}

static bool is_snb_ioat(struct pci_dev *pdev)
{
	switch (pdev->device) {
	case PCI_DEVICE_ID_INTEL_IOAT_SNB0:
	case PCI_DEVICE_ID_INTEL_IOAT_SNB1:
	case PCI_DEVICE_ID_INTEL_IOAT_SNB2:
	case PCI_DEVICE_ID_INTEL_IOAT_SNB3:
	case PCI_DEVICE_ID_INTEL_IOAT_SNB4:
	case PCI_DEVICE_ID_INTEL_IOAT_SNB5:
	case PCI_DEVICE_ID_INTEL_IOAT_SNB6:
	case PCI_DEVICE_ID_INTEL_IOAT_SNB7:
	case PCI_DEVICE_ID_INTEL_IOAT_SNB8:
	case PCI_DEVICE_ID_INTEL_IOAT_SNB9:
		return true;
	default:
		return false;
	}
}

static bool is_ivb_ioat(struct pci_dev *pdev)
{
	switch (pdev->device) {
	case PCI_DEVICE_ID_INTEL_IOAT_IVB0:
	case PCI_DEVICE_ID_INTEL_IOAT_IVB1:
	case PCI_DEVICE_ID_INTEL_IOAT_IVB2:
	case PCI_DEVICE_ID_INTEL_IOAT_IVB3:
	case PCI_DEVICE_ID_INTEL_IOAT_IVB4:
	case PCI_DEVICE_ID_INTEL_IOAT_IVB5:
	case PCI_DEVICE_ID_INTEL_IOAT_IVB6:
	case PCI_DEVICE_ID_INTEL_IOAT_IVB7:
	case PCI_DEVICE_ID_INTEL_IOAT_IVB8:
	case PCI_DEVICE_ID_INTEL_IOAT_IVB9:
		return true;
	default:
		return false;
	}

}

static bool is_hsw_ioat(struct pci_dev *pdev)
{
	switch (pdev->device) {
	case PCI_DEVICE_ID_INTEL_IOAT_HSW0:
	case PCI_DEVICE_ID_INTEL_IOAT_HSW1:
	case PCI_DEVICE_ID_INTEL_IOAT_HSW2:
	case PCI_DEVICE_ID_INTEL_IOAT_HSW3:
	case PCI_DEVICE_ID_INTEL_IOAT_HSW4:
	case PCI_DEVICE_ID_INTEL_IOAT_HSW5:
	case PCI_DEVICE_ID_INTEL_IOAT_HSW6:
	case PCI_DEVICE_ID_INTEL_IOAT_HSW7:
	case PCI_DEVICE_ID_INTEL_IOAT_HSW8:
	case PCI_DEVICE_ID_INTEL_IOAT_HSW9:
		return true;
	default:
		return false;
	}

}

static bool is_xeon_cb32(struct pci_dev *pdev)
{
	return is_jf_ioat(pdev) || is_snb_ioat(pdev) || is_ivb_ioat(pdev) ||
		is_hsw_ioat(pdev);
}

static bool is_bwd_ioat(struct pci_dev *pdev)
{
	switch (pdev->device) {
	case PCI_DEVICE_ID_INTEL_IOAT_BWD0:
	case PCI_DEVICE_ID_INTEL_IOAT_BWD1:
	case PCI_DEVICE_ID_INTEL_IOAT_BWD2:
	case PCI_DEVICE_ID_INTEL_IOAT_BWD3:
		return true;
	default:
		return false;
	}
}

static bool is_bwd_noraid(struct pci_dev *pdev)
{
	switch (pdev->device) {
	case PCI_DEVICE_ID_INTEL_IOAT_BWD2:
	case PCI_DEVICE_ID_INTEL_IOAT_BWD3:
		return true;
	default:
		return false;
	}

}

static void pq16_set_src(struct ioat_raw_descriptor *desc[3],
			dma_addr_t addr, u32 offset, u8 coef, unsigned idx)
{
	struct ioat_pq_descriptor *pq = (struct ioat_pq_descriptor *)desc[0];
	struct ioat_pq16a_descriptor *pq16 =
		(struct ioat_pq16a_descriptor *)desc[1];
	struct ioat_raw_descriptor *raw = desc[pq16_idx_to_desc[idx]];

	raw->field[pq16_idx_to_field[idx]] = addr + offset;

	if (idx < 8)
		pq->coef[idx] = coef;
	else
		pq16->coef[idx - 8] = coef;
}

static struct ioat_sed_ent *
ioat3_alloc_sed(struct ioatdma_device *device, unsigned int hw_pool)
{
	struct ioat_sed_ent *sed;
	gfp_t flags = __GFP_ZERO | GFP_ATOMIC;

	sed = kmem_cache_alloc(ioat3_sed_cache, flags);
	if (!sed)
		return NULL;

	sed->hw_pool = hw_pool;
	sed->hw = dma_pool_alloc(device->sed_hw_pool[hw_pool],
				 flags, &sed->dma);
	if (!sed->hw) {
		kmem_cache_free(ioat3_sed_cache, sed);
		return NULL;
	}

	return sed;
}

static void ioat3_free_sed(struct ioatdma_device *device, struct ioat_sed_ent *sed)
{
	if (!sed)
		return;

	dma_pool_free(device->sed_hw_pool[sed->hw_pool], sed->hw, sed->dma);
	kmem_cache_free(ioat3_sed_cache, sed);
}

static bool desc_has_ext(struct ioat_ring_ent *desc)
{
	struct ioat_dma_descriptor *hw = desc->hw;

	if (hw->ctl_f.op == IOAT_OP_XOR ||
	    hw->ctl_f.op == IOAT_OP_XOR_VAL) {
		struct ioat_xor_descriptor *xor = desc->xor;

		if (src_cnt_to_sw(xor->ctl_f.src_cnt) > 5)
			return true;
	} else if (hw->ctl_f.op == IOAT_OP_PQ ||
		   hw->ctl_f.op == IOAT_OP_PQ_VAL) {
		struct ioat_pq_descriptor *pq = desc->pq;

		if (src_cnt_to_sw(pq->ctl_f.src_cnt) > 3)
			return true;
	}

	return false;
}

static u64 ioat3_get_current_completion(struct ioat_chan_common *chan)
{
	u64 phys_complete;
	u64 completion;

	completion = *chan->completion;
	phys_complete = ioat_chansts_to_addr(completion);

	dev_dbg(to_dev(chan), "%s: phys_complete: %#llx\n", __func__,
		(unsigned long long) phys_complete);

	return phys_complete;
}

static bool ioat3_cleanup_preamble(struct ioat_chan_common *chan,
				   u64 *phys_complete)
{
	*phys_complete = ioat3_get_current_completion(chan);
	if (*phys_complete == chan->last_completion)
		return false;

	clear_bit(IOAT_COMPLETION_ACK, &chan->state);
	mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);

	return true;
}

static void
desc_get_errstat(struct ioat2_dma_chan *ioat, struct ioat_ring_ent *desc)
{
	struct ioat_dma_descriptor *hw = desc->hw;

	switch (hw->ctl_f.op) {
	case IOAT_OP_PQ_VAL:
	case IOAT_OP_PQ_VAL_16S:
	{
		struct ioat_pq_descriptor *pq = desc->pq;

		/* check if there's error written */
		if (!pq->dwbes_f.wbes)
			return;

		/* need to set a chanerr var for checking to clear later */

		if (pq->dwbes_f.p_val_err)
			*desc->result |= SUM_CHECK_P_RESULT;

		if (pq->dwbes_f.q_val_err)
			*desc->result |= SUM_CHECK_Q_RESULT;

		return;
	}
	default:
		return;
	}
}

/**
 * __cleanup - reclaim used descriptors
 * @ioat: channel (ring) to clean
 *
 * The difference from the dma_v2.c __cleanup() is that this routine
 * handles extended descriptors and dma-unmapping raid operations.
 */
static void __cleanup(struct ioat2_dma_chan *ioat, dma_addr_t phys_complete)
{
	struct ioat_chan_common *chan = &ioat->base;
	struct ioatdma_device *device = chan->device;
	struct ioat_ring_ent *desc;
	bool seen_current = false;
	int idx = ioat->tail, i;
	u16 active;

	dev_dbg(to_dev(chan), "%s: head: %#x tail: %#x issued: %#x\n",
		__func__, ioat->head, ioat->tail, ioat->issued);

	/*
	 * At restart of the channel, the completion address and the
	 * channel status will be 0 due to starting a new chain. Since
	 * it's new chain and the first descriptor "fails", there is
	 * nothing to clean up. We do not want to reap the entire submitted
	 * chain due to this 0 address value and then BUG.
	 */
	if (!phys_complete)
		return;

	active = ioat2_ring_active(ioat);
	for (i = 0; i < active && !seen_current; i++) {
		struct dma_async_tx_descriptor *tx;

		smp_read_barrier_depends();
		prefetch(ioat2_get_ring_ent(ioat, idx + i + 1));
		desc = ioat2_get_ring_ent(ioat, idx + i);
		dump_desc_dbg(ioat, desc);

		/* set err stat if we are using dwbes */
		if (device->cap & IOAT_CAP_DWBES)
			desc_get_errstat(ioat, desc);

		tx = &desc->txd;
		if (tx->cookie) {
			dma_cookie_complete(tx);
			dma_descriptor_unmap(tx);
			if (tx->callback) {
				tx->callback(tx->callback_param);
				tx->callback = NULL;
			}
		}

		if (tx->phys == phys_complete)
			seen_current = true;

		/* skip extended descriptors */
		if (desc_has_ext(desc)) {
			BUG_ON(i + 1 >= active);
			i++;
		}

		/* cleanup super extended descriptors */
		if (desc->sed) {
			ioat3_free_sed(device, desc->sed);
			desc->sed = NULL;
		}
	}
	smp_mb(); /* finish all descriptor reads before incrementing tail */
	ioat->tail = idx + i;
	BUG_ON(active && !seen_current); /* no active descs have written a completion? */
	chan->last_completion = phys_complete;

	if (active - i == 0) {
		dev_dbg(to_dev(chan), "%s: cancel completion timeout\n",
			__func__);
		clear_bit(IOAT_COMPLETION_PENDING, &chan->state);
		mod_timer(&chan->timer, jiffies + IDLE_TIMEOUT);
	}
	/* 5 microsecond delay per pending descriptor */
	writew(min((5 * (active - i)), IOAT_INTRDELAY_MASK),
	       chan->device->reg_base + IOAT_INTRDELAY_OFFSET);
}

static void ioat3_cleanup(struct ioat2_dma_chan *ioat)
{
	struct ioat_chan_common *chan = &ioat->base;
	u64 phys_complete;

	spin_lock_bh(&chan->cleanup_lock);

	if (ioat3_cleanup_preamble(chan, &phys_complete))
		__cleanup(ioat, phys_complete);

	if (is_ioat_halted(*chan->completion)) {
		u32 chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);

		if (chanerr & IOAT_CHANERR_HANDLE_MASK) {
			mod_timer(&chan->timer, jiffies + IDLE_TIMEOUT);
			ioat3_eh(ioat);
		}
	}

	spin_unlock_bh(&chan->cleanup_lock);
}

static void ioat3_cleanup_event(unsigned long data)
{
	struct ioat2_dma_chan *ioat = to_ioat2_chan((void *) data);
	struct ioat_chan_common *chan = &ioat->base;

	ioat3_cleanup(ioat);
	if (!test_bit(IOAT_RUN, &chan->state))
		return;
	writew(IOAT_CHANCTRL_RUN, ioat->base.reg_base + IOAT_CHANCTRL_OFFSET);
}

static void ioat3_restart_channel(struct ioat2_dma_chan *ioat)
{
	struct ioat_chan_common *chan = &ioat->base;
	u64 phys_complete;

	ioat2_quiesce(chan, 0);
	if (ioat3_cleanup_preamble(chan, &phys_complete))
		__cleanup(ioat, phys_complete);

	__ioat2_restart_chan(ioat);
}

static void ioat3_eh(struct ioat2_dma_chan *ioat)
{
	struct ioat_chan_common *chan = &ioat->base;
	struct pci_dev *pdev = to_pdev(chan);
	struct ioat_dma_descriptor *hw;
	u64 phys_complete;
	struct ioat_ring_ent *desc;
	u32 err_handled = 0;
	u32 chanerr_int;
	u32 chanerr;

	/* cleanup so tail points to descriptor that caused the error */
	if (ioat3_cleanup_preamble(chan, &phys_complete))
		__cleanup(ioat, phys_complete);

	chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
	pci_read_config_dword(pdev, IOAT_PCI_CHANERR_INT_OFFSET, &chanerr_int);

	dev_dbg(to_dev(chan), "%s: error = %x:%x\n",
		__func__, chanerr, chanerr_int);

	desc = ioat2_get_ring_ent(ioat, ioat->tail);
	hw = desc->hw;
	dump_desc_dbg(ioat, desc);

	switch (hw->ctl_f.op) {
	case IOAT_OP_XOR_VAL:
		if (chanerr & IOAT_CHANERR_XOR_P_OR_CRC_ERR) {
			*desc->result |= SUM_CHECK_P_RESULT;
			err_handled |= IOAT_CHANERR_XOR_P_OR_CRC_ERR;
		}
		break;
	case IOAT_OP_PQ_VAL:
	case IOAT_OP_PQ_VAL_16S:
		if (chanerr & IOAT_CHANERR_XOR_P_OR_CRC_ERR) {
			*desc->result |= SUM_CHECK_P_RESULT;
			err_handled |= IOAT_CHANERR_XOR_P_OR_CRC_ERR;
		}
		if (chanerr & IOAT_CHANERR_XOR_Q_ERR) {
			*desc->result |= SUM_CHECK_Q_RESULT;
			err_handled |= IOAT_CHANERR_XOR_Q_ERR;
		}
		break;
	}

	/* fault on unhandled error or spurious halt */
	if (chanerr ^ err_handled || chanerr == 0) {
		dev_err(to_dev(chan), "%s: fatal error (%x:%x)\n",
			__func__, chanerr, err_handled);
		BUG();
	}

	writel(chanerr, chan->reg_base + IOAT_CHANERR_OFFSET);
	pci_write_config_dword(pdev, IOAT_PCI_CHANERR_INT_OFFSET, chanerr_int);

	/* mark faulting descriptor as complete */
	*chan->completion = desc->txd.phys;

	spin_lock_bh(&ioat->prep_lock);
	ioat3_restart_channel(ioat);
	spin_unlock_bh(&ioat->prep_lock);
}

static void check_active(struct ioat2_dma_chan *ioat)
{
	struct ioat_chan_common *chan = &ioat->base;

	if (ioat2_ring_active(ioat)) {
		mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
		return;
	}

	if (test_and_clear_bit(IOAT_CHAN_ACTIVE, &chan->state))
		mod_timer(&chan->timer, jiffies + IDLE_TIMEOUT);
	else if (ioat->alloc_order > ioat_get_alloc_order()) {
		/* if the ring is idle, empty, and oversized try to step
		 * down the size
		 */
		reshape_ring(ioat, ioat->alloc_order - 1);

		/* keep shrinking until we get back to our minimum
		 * default size
		 */
		if (ioat->alloc_order > ioat_get_alloc_order())
			mod_timer(&chan->timer, jiffies + IDLE_TIMEOUT);
	}

}

static void ioat3_timer_event(unsigned long data)
{
	struct ioat2_dma_chan *ioat = to_ioat2_chan((void *) data);
	struct ioat_chan_common *chan = &ioat->base;
	dma_addr_t phys_complete;
	u64 status;

	status = ioat_chansts(chan);

	/* when halted due to errors check for channel
	 * programming errors before advancing the completion state
	 */
	if (is_ioat_halted(status)) {
		u32 chanerr;

		chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
		dev_err(to_dev(chan), "%s: Channel halted (%x)\n",
			__func__, chanerr);
		if (test_bit(IOAT_RUN, &chan->state))
			BUG_ON(is_ioat_bug(chanerr));
		else /* we never got off the ground */
			return;
	}

	/* if we haven't made progress and we have already
	 * acknowledged a pending completion once, then be more
	 * forceful with a restart
	 */
	spin_lock_bh(&chan->cleanup_lock);
	if (ioat_cleanup_preamble(chan, &phys_complete))
		__cleanup(ioat, phys_complete);
	else if (test_bit(IOAT_COMPLETION_ACK, &chan->state)) {
		spin_lock_bh(&ioat->prep_lock);
		ioat3_restart_channel(ioat);
		spin_unlock_bh(&ioat->prep_lock);
		spin_unlock_bh(&chan->cleanup_lock);
		return;
	} else {
		set_bit(IOAT_COMPLETION_ACK, &chan->state);
		mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
	}


	if (ioat2_ring_active(ioat))
		mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
	else {
		spin_lock_bh(&ioat->prep_lock);
		check_active(ioat);
		spin_unlock_bh(&ioat->prep_lock);
	}
	spin_unlock_bh(&chan->cleanup_lock);
}

static enum dma_status
ioat3_tx_status(struct dma_chan *c, dma_cookie_t cookie,
		struct dma_tx_state *txstate)
{
	struct ioat2_dma_chan *ioat = to_ioat2_chan(c);
	enum dma_status ret;

	ret = dma_cookie_status(c, cookie, txstate);
	if (ret == DMA_COMPLETE)
		return ret;

	ioat3_cleanup(ioat);

	return dma_cookie_status(c, cookie, txstate);
}

static struct dma_async_tx_descriptor *
__ioat3_prep_xor_lock(struct dma_chan *c, enum sum_check_flags *result,
		      dma_addr_t dest, dma_addr_t *src, unsigned int src_cnt,
		      size_t len, unsigned long flags)
{
	struct ioat2_dma_chan *ioat = to_ioat2_chan(c);
	struct ioat_ring_ent *compl_desc;
	struct ioat_ring_ent *desc;
	struct ioat_ring_ent *ext;
	size_t total_len = len;
	struct ioat_xor_descriptor *xor;
	struct ioat_xor_ext_descriptor *xor_ex = NULL;
	struct ioat_dma_descriptor *hw;
	int num_descs, with_ext, idx, i;
	u32 offset = 0;
	u8 op = result ? IOAT_OP_XOR_VAL : IOAT_OP_XOR;

	BUG_ON(src_cnt < 2);

	num_descs = ioat2_xferlen_to_descs(ioat, len);
	/* we need 2x the number of descriptors to cover greater than 5
	 * sources
	 */
	if (src_cnt > 5) {
		with_ext = 1;
		num_descs *= 2;
	} else
		with_ext = 0;

	/* completion writes from the raid engine may pass completion
	 * writes from the legacy engine, so we need one extra null
	 * (legacy) descriptor to ensure all completion writes arrive in
	 * order.
	 */
	if (likely(num_descs) && ioat2_check_space_lock(ioat, num_descs+1) == 0)
		idx = ioat->head;
	else
		return NULL;
	i = 0;
	do {
		struct ioat_raw_descriptor *descs[2];
		size_t xfer_size = min_t(size_t, len, 1 << ioat->xfercap_log);
		int s;

		desc = ioat2_get_ring_ent(ioat, idx + i);
		xor = desc->xor;

		/* save a branch by unconditionally retrieving the
		 * extended descriptor xor_set_src() knows to not write
		 * to it in the single descriptor case
		 */
		ext = ioat2_get_ring_ent(ioat, idx + i + 1);
		xor_ex = ext->xor_ex;

		descs[0] = (struct ioat_raw_descriptor *) xor;
		descs[1] = (struct ioat_raw_descriptor *) xor_ex;
		for (s = 0; s < src_cnt; s++)
			xor_set_src(descs, src[s], offset, s);
		xor->size = xfer_size;
		xor->dst_addr = dest + offset;
		xor->ctl = 0;
		xor->ctl_f.op = op;
		xor->ctl_f.src_cnt = src_cnt_to_hw(src_cnt);

		len -= xfer_size;
		offset += xfer_size;
		dump_desc_dbg(ioat, desc);
	} while ((i += 1 + with_ext) < num_descs);

	/* last xor descriptor carries the unmap parameters and fence bit */
	desc->txd.flags = flags;
	desc->len = total_len;
	if (result)
		desc->result = result;
	xor->ctl_f.fence = !!(flags & DMA_PREP_FENCE);

	/* completion descriptor carries interrupt bit */
	compl_desc = ioat2_get_ring_ent(ioat, idx + i);
	compl_desc->txd.flags = flags & DMA_PREP_INTERRUPT;
	hw = compl_desc->hw;
	hw->ctl = 0;
	hw->ctl_f.null = 1;
	hw->ctl_f.int_en = !!(flags & DMA_PREP_INTERRUPT);
	hw->ctl_f.compl_write = 1;
	hw->size = NULL_DESC_BUFFER_SIZE;
	dump_desc_dbg(ioat, compl_desc);

	/* we leave the channel locked to ensure in order submission */
	return &compl_desc->txd;
}

static struct dma_async_tx_descriptor *
ioat3_prep_xor(struct dma_chan *chan, dma_addr_t dest, dma_addr_t *src,
	       unsigned int src_cnt, size_t len, unsigned long flags)
{
	return __ioat3_prep_xor_lock(chan, NULL, dest, src, src_cnt, len, flags);
}

static struct dma_async_tx_descriptor *
ioat3_prep_xor_val(struct dma_chan *chan, dma_addr_t *src,
		    unsigned int src_cnt, size_t len,
		    enum sum_check_flags *result, unsigned long flags)
{
	/* the cleanup routine only sets bits on validate failure, it
	 * does not clear bits on validate success... so clear it here
	 */
	*result = 0;

	return __ioat3_prep_xor_lock(chan, result, src[0], &src[1],
				     src_cnt - 1, len, flags);
}

static void
dump_pq_desc_dbg(struct ioat2_dma_chan *ioat, struct ioat_ring_ent *desc, struct ioat_ring_ent *ext)
{
	struct device *dev = to_dev(&ioat->base);
	struct ioat_pq_descriptor *pq = desc->pq;
	struct ioat_pq_ext_descriptor *pq_ex = ext ? ext->pq_ex : NULL;
	struct ioat_raw_descriptor *descs[] = { (void *) pq, (void *) pq_ex };
	int src_cnt = src_cnt_to_sw(pq->ctl_f.src_cnt);
	int i;

	dev_dbg(dev, "desc[%d]: (%#llx->%#llx) flags: %#x"
		" sz: %#10.8x ctl: %#x (op: %#x int: %d compl: %d pq: '%s%s'"
		" src_cnt: %d)\n",
		desc_id(desc), (unsigned long long) desc->txd.phys,
		(unsigned long long) (pq_ex ? pq_ex->next : pq->next),
		desc->txd.flags, pq->size, pq->ctl, pq->ctl_f.op, pq->ctl_f.int_en,
		pq->ctl_f.compl_write,
		pq->ctl_f.p_disable ? "" : "p", pq->ctl_f.q_disable ? "" : "q",
		pq->ctl_f.src_cnt);
	for (i = 0; i < src_cnt; i++)
		dev_dbg(dev, "\tsrc[%d]: %#llx coef: %#x\n", i,
			(unsigned long long) pq_get_src(descs, i), pq->coef[i]);
	dev_dbg(dev, "\tP: %#llx\n", pq->p_addr);
	dev_dbg(dev, "\tQ: %#llx\n", pq->q_addr);
	dev_dbg(dev, "\tNEXT: %#llx\n", pq->next);
}

static void dump_pq16_desc_dbg(struct ioat2_dma_chan *ioat,
			       struct ioat_ring_ent *desc)
{
	struct device *dev = to_dev(&ioat->base);
	struct ioat_pq_descriptor *pq = desc->pq;
	struct ioat_raw_descriptor *descs[] = { (void *)pq,
						(void *)pq,
						(void *)pq };
	int src_cnt = src16_cnt_to_sw(pq->ctl_f.src_cnt);
	int i;

	if (desc->sed) {
		descs[1] = (void *)desc->sed->hw;
		descs[2] = (void *)desc->sed->hw + 64;
	}

	dev_dbg(dev, "desc[%d]: (%#llx->%#llx) flags: %#x"
		" sz: %#x ctl: %#x (op: %#x int: %d compl: %d pq: '%s%s'"
		" src_cnt: %d)\n",
		desc_id(desc), (unsigned long long) desc->txd.phys,
		(unsigned long long) pq->next,
		desc->txd.flags, pq->size, pq->ctl,
		pq->ctl_f.op, pq->ctl_f.int_en,
		pq->ctl_f.compl_write,
		pq->ctl_f.p_disable ? "" : "p", pq->ctl_f.q_disable ? "" : "q",
		pq->ctl_f.src_cnt);
	for (i = 0; i < src_cnt; i++) {
		dev_dbg(dev, "\tsrc[%d]: %#llx coef: %#x\n", i,
			(unsigned long long) pq16_get_src(descs, i),
			pq->coef[i]);
	}
	dev_dbg(dev, "\tP: %#llx\n", pq->p_addr);
	dev_dbg(dev, "\tQ: %#llx\n", pq->q_addr);
}

static struct dma_async_tx_descriptor *
__ioat3_prep_pq_lock(struct dma_chan *c, enum sum_check_flags *result,
		     const dma_addr_t *dst, const dma_addr_t *src,
		     unsigned int src_cnt, const unsigned char *scf,
		     size_t len, unsigned long flags)
{
	struct ioat2_dma_chan *ioat = to_ioat2_chan(c);
	struct ioat_chan_common *chan = &ioat->base;
	struct ioatdma_device *device = chan->device;
	struct ioat_ring_ent *compl_desc;
	struct ioat_ring_ent *desc;
	struct ioat_ring_ent *ext;
	size_t total_len = len;
	struct ioat_pq_descriptor *pq;
	struct ioat_pq_ext_descriptor *pq_ex = NULL;
	struct ioat_dma_descriptor *hw;
	u32 offset = 0;
	u8 op = result ? IOAT_OP_PQ_VAL : IOAT_OP_PQ;
	int i, s, idx, with_ext, num_descs;
	int cb32 = (device->version < IOAT_VER_3_3) ? 1 : 0;

	dev_dbg(to_dev(chan), "%s\n", __func__);
	/* the engine requires at least two sources (we provide
	 * at least 1 implied source in the DMA_PREP_CONTINUE case)
	 */
	BUG_ON(src_cnt + dmaf_continue(flags) < 2);

	num_descs = ioat2_xferlen_to_descs(ioat, len);
	/* we need 2x the number of descriptors to cover greater than 3
	 * sources (we need 1 extra source in the q-only continuation
	 * case and 3 extra sources in the p+q continuation case.
	 */
	if (src_cnt + dmaf_p_disabled_continue(flags) > 3 ||
	    (dmaf_continue(flags) && !dmaf_p_disabled_continue(flags))) {
		with_ext = 1;
		num_descs *= 2;
	} else
		with_ext = 0;

	/* completion writes from the raid engine may pass completion
	 * writes from the legacy engine, so we need one extra null
	 * (legacy) descriptor to ensure all completion writes arrive in
	 * order.
	 */
	if (likely(num_descs) &&
	    ioat2_check_space_lock(ioat, num_descs + cb32) == 0)
		idx = ioat->head;
	else
		return NULL;
	i = 0;
	do {
		struct ioat_raw_descriptor *descs[2];
		size_t xfer_size = min_t(size_t, len, 1 << ioat->xfercap_log);

		desc = ioat2_get_ring_ent(ioat, idx + i);
		pq = desc->pq;

		/* save a branch by unconditionally retrieving the
		 * extended descriptor pq_set_src() knows to not write
		 * to it in the single descriptor case
		 */
		ext = ioat2_get_ring_ent(ioat, idx + i + with_ext);
		pq_ex = ext->pq_ex;

		descs[0] = (struct ioat_raw_descriptor *) pq;
		descs[1] = (struct ioat_raw_descriptor *) pq_ex;

		for (s = 0; s < src_cnt; s++)
			pq_set_src(descs, src[s], offset, scf[s], s);

		/* see the comment for dma_maxpq in include/linux/dmaengine.h */
		if (dmaf_p_disabled_continue(flags))
			pq_set_src(descs, dst[1], offset, 1, s++);
		else if (dmaf_continue(flags)) {
			pq_set_src(descs, dst[0], offset, 0, s++);
			pq_set_src(descs, dst[1], offset, 1, s++);
			pq_set_src(descs, dst[1], offset, 0, s++);
		}
		pq->size = xfer_size;
		pq->p_addr = dst[0] + offset;
		pq->q_addr = dst[1] + offset;
		pq->ctl = 0;
		pq->ctl_f.op = op;
		/* we turn on descriptor write back error status */
		if (device->cap & IOAT_CAP_DWBES)
			pq->ctl_f.wb_en = result ? 1 : 0;
		pq->ctl_f.src_cnt = src_cnt_to_hw(s);
		pq->ctl_f.p_disable = !!(flags & DMA_PREP_PQ_DISABLE_P);
		pq->ctl_f.q_disable = !!(flags & DMA_PREP_PQ_DISABLE_Q);

		len -= xfer_size;
		offset += xfer_size;
	} while ((i += 1 + with_ext) < num_descs);

	/* last pq descriptor carries the unmap parameters and fence bit */
	desc->txd.flags = flags;
	desc->len = total_len;
	if (result)
		desc->result = result;
	pq->ctl_f.fence = !!(flags & DMA_PREP_FENCE);
	dump_pq_desc_dbg(ioat, desc, ext);

	if (!cb32) {
		pq->ctl_f.int_en = !!(flags & DMA_PREP_INTERRUPT);
		pq->ctl_f.compl_write = 1;
		compl_desc = desc;
	} else {
		/* completion descriptor carries interrupt bit */
		compl_desc = ioat2_get_ring_ent(ioat, idx + i);
		compl_desc->txd.flags = flags & DMA_PREP_INTERRUPT;
		hw = compl_desc->hw;
		hw->ctl = 0;
		hw->ctl_f.null = 1;
		hw->ctl_f.int_en = !!(flags & DMA_PREP_INTERRUPT);
		hw->ctl_f.compl_write = 1;
		hw->size = NULL_DESC_BUFFER_SIZE;
		dump_desc_dbg(ioat, compl_desc);
	}


	/* we leave the channel locked to ensure in order submission */
	return &compl_desc->txd;
}

static struct dma_async_tx_descriptor *
__ioat3_prep_pq16_lock(struct dma_chan *c, enum sum_check_flags *result,
		       const dma_addr_t *dst, const dma_addr_t *src,
		       unsigned int src_cnt, const unsigned char *scf,
		       size_t len, unsigned long flags)
{
	struct ioat2_dma_chan *ioat = to_ioat2_chan(c);
	struct ioat_chan_common *chan = &ioat->base;
	struct ioatdma_device *device = chan->device;
	struct ioat_ring_ent *desc;
	size_t total_len = len;
	struct ioat_pq_descriptor *pq;
	u32 offset = 0;
	u8 op;
	int i, s, idx, num_descs;

	/* this function is only called with 9-16 sources */
	op = result ? IOAT_OP_PQ_VAL_16S : IOAT_OP_PQ_16S;

	dev_dbg(to_dev(chan), "%s\n", __func__);

	num_descs = ioat2_xferlen_to_descs(ioat, len);

	/*
	 * 16 source pq is only available on cb3.3 and has no completion
	 * write hw bug.
	 */
	if (num_descs && ioat2_check_space_lock(ioat, num_descs) == 0)
		idx = ioat->head;
	else
		return NULL;

	i = 0;

	do {
		struct ioat_raw_descriptor *descs[4];
		size_t xfer_size = min_t(size_t, len, 1 << ioat->xfercap_log);

		desc = ioat2_get_ring_ent(ioat, idx + i);
		pq = desc->pq;

		descs[0] = (struct ioat_raw_descriptor *) pq;

		desc->sed = ioat3_alloc_sed(device, (src_cnt-2) >> 3);
		if (!desc->sed) {
			dev_err(to_dev(chan),
				"%s: no free sed entries\n", __func__);
			return NULL;
		}

		pq->sed_addr = desc->sed->dma;
		desc->sed->parent = desc;

		descs[1] = (struct ioat_raw_descriptor *)desc->sed->hw;
		descs[2] = (void *)descs[1] + 64;

		for (s = 0; s < src_cnt; s++)
			pq16_set_src(descs, src[s], offset, scf[s], s);

		/* see the comment for dma_maxpq in include/linux/dmaengine.h */
		if (dmaf_p_disabled_continue(flags))
			pq16_set_src(descs, dst[1], offset, 1, s++);
		else if (dmaf_continue(flags)) {
			pq16_set_src(descs, dst[0], offset, 0, s++);
			pq16_set_src(descs, dst[1], offset, 1, s++);
			pq16_set_src(descs, dst[1], offset, 0, s++);
		}

		pq->size = xfer_size;
		pq->p_addr = dst[0] + offset;
		pq->q_addr = dst[1] + offset;
		pq->ctl = 0;
		pq->ctl_f.op = op;
		pq->ctl_f.src_cnt = src16_cnt_to_hw(s);
		/* we turn on descriptor write back error status */
		if (device->cap & IOAT_CAP_DWBES)
			pq->ctl_f.wb_en = result ? 1 : 0;
		pq->ctl_f.p_disable = !!(flags & DMA_PREP_PQ_DISABLE_P);
		pq->ctl_f.q_disable = !!(flags & DMA_PREP_PQ_DISABLE_Q);

		len -= xfer_size;
		offset += xfer_size;
	} while (++i < num_descs);

	/* last pq descriptor carries the unmap parameters and fence bit */
	desc->txd.flags = flags;
	desc->len = total_len;
	if (result)
		desc->result = result;
	pq->ctl_f.fence = !!(flags & DMA_PREP_FENCE);

	/* with cb3.3 we should be able to do completion w/o a null desc */
	pq->ctl_f.int_en = !!(flags & DMA_PREP_INTERRUPT);
	pq->ctl_f.compl_write = 1;

	dump_pq16_desc_dbg(ioat, desc);

	/* we leave the channel locked to ensure in order submission */
	return &desc->txd;
}

static int src_cnt_flags(unsigned int src_cnt, unsigned long flags)
{
	if (dmaf_p_disabled_continue(flags))
		return src_cnt + 1;
	else if (dmaf_continue(flags))
		return src_cnt + 3;
	else
		return src_cnt;
}

static struct dma_async_tx_descriptor *
ioat3_prep_pq(struct dma_chan *chan, dma_addr_t *dst, dma_addr_t *src,
	      unsigned int src_cnt, const unsigned char *scf, size_t len,
	      unsigned long flags)
{
	/* specify valid address for disabled result */
	if (flags & DMA_PREP_PQ_DISABLE_P)
		dst[0] = dst[1];
	if (flags & DMA_PREP_PQ_DISABLE_Q)
		dst[1] = dst[0];

	/* handle the single source multiply case from the raid6
	 * recovery path
	 */
	if ((flags & DMA_PREP_PQ_DISABLE_P) && src_cnt == 1) {
		dma_addr_t single_source[2];
		unsigned char single_source_coef[2];

		BUG_ON(flags & DMA_PREP_PQ_DISABLE_Q);
		single_source[0] = src[0];
		single_source[1] = src[0];
		single_source_coef[0] = scf[0];
		single_source_coef[1] = 0;

		return src_cnt_flags(src_cnt, flags) > 8 ?
			__ioat3_prep_pq16_lock(chan, NULL, dst, single_source,
					       2, single_source_coef, len,
					       flags) :
			__ioat3_prep_pq_lock(chan, NULL, dst, single_source, 2,
					     single_source_coef, len, flags);

	} else {
		return src_cnt_flags(src_cnt, flags) > 8 ?
			__ioat3_prep_pq16_lock(chan, NULL, dst, src, src_cnt,
					       scf, len, flags) :
			__ioat3_prep_pq_lock(chan, NULL, dst, src, src_cnt,
					     scf, len, flags);
	}
}

static struct dma_async_tx_descriptor *
ioat3_prep_pq_val(struct dma_chan *chan, dma_addr_t *pq, dma_addr_t *src,
		  unsigned int src_cnt, const unsigned char *scf, size_t len,
		  enum sum_check_flags *pqres, unsigned long flags)
{
	/* specify valid address for disabled result */
	if (flags & DMA_PREP_PQ_DISABLE_P)
		pq[0] = pq[1];
	if (flags & DMA_PREP_PQ_DISABLE_Q)
		pq[1] = pq[0];

	/* the cleanup routine only sets bits on validate failure, it
	 * does not clear bits on validate success... so clear it here
	 */
	*pqres = 0;

	return src_cnt_flags(src_cnt, flags) > 8 ?
		__ioat3_prep_pq16_lock(chan, pqres, pq, src, src_cnt, scf, len,
				       flags) :
		__ioat3_prep_pq_lock(chan, pqres, pq, src, src_cnt, scf, len,
				     flags);
}

static struct dma_async_tx_descriptor *
ioat3_prep_pqxor(struct dma_chan *chan, dma_addr_t dst, dma_addr_t *src,
		 unsigned int src_cnt, size_t len, unsigned long flags)
{
	unsigned char scf[src_cnt];
	dma_addr_t pq[2];

	memset(scf, 0, src_cnt);
	pq[0] = dst;
	flags |= DMA_PREP_PQ_DISABLE_Q;
	pq[1] = dst; /* specify valid address for disabled result */

	return src_cnt_flags(src_cnt, flags) > 8 ?
		__ioat3_prep_pq16_lock(chan, NULL, pq, src, src_cnt, scf, len,
				       flags) :
		__ioat3_prep_pq_lock(chan, NULL, pq, src, src_cnt, scf, len,
				     flags);
}

static struct dma_async_tx_descriptor *
ioat3_prep_pqxor_val(struct dma_chan *chan, dma_addr_t *src,
		     unsigned int src_cnt, size_t len,
		     enum sum_check_flags *result, unsigned long flags)
{
	unsigned char scf[src_cnt];
	dma_addr_t pq[2];

	/* the cleanup routine only sets bits on validate failure, it
	 * does not clear bits on validate success... so clear it here
	 */
	*result = 0;

	memset(scf, 0, src_cnt);
	pq[0] = src[0];
	flags |= DMA_PREP_PQ_DISABLE_Q;
	pq[1] = pq[0]; /* specify valid address for disabled result */

	return src_cnt_flags(src_cnt, flags) > 8 ?
		__ioat3_prep_pq16_lock(chan, result, pq, &src[1], src_cnt - 1,
				       scf, len, flags) :
		__ioat3_prep_pq_lock(chan, result, pq, &src[1], src_cnt - 1,
				     scf, len, flags);
}

static struct dma_async_tx_descriptor *
ioat3_prep_interrupt_lock(struct dma_chan *c, unsigned long flags)
{
	struct ioat2_dma_chan *ioat = to_ioat2_chan(c);
	struct ioat_ring_ent *desc;
	struct ioat_dma_descriptor *hw;

	if (ioat2_check_space_lock(ioat, 1) == 0)
		desc = ioat2_get_ring_ent(ioat, ioat->head);
	else
		return NULL;

	hw = desc->hw;
	hw->ctl = 0;
	hw->ctl_f.null = 1;
	hw->ctl_f.int_en = 1;
	hw->ctl_f.fence = !!(flags & DMA_PREP_FENCE);
	hw->ctl_f.compl_write = 1;
	hw->size = NULL_DESC_BUFFER_SIZE;
	hw->src_addr = 0;
	hw->dst_addr = 0;

	desc->txd.flags = flags;
	desc->len = 1;

	dump_desc_dbg(ioat, desc);

	/* we leave the channel locked to ensure in order submission */
	return &desc->txd;
}

static void ioat3_dma_test_callback(void *dma_async_param)
{
	struct completion *cmp = dma_async_param;

	complete(cmp);
}

#define IOAT_NUM_SRC_TEST 6 /* must be <= 8 */
static int ioat_xor_val_self_test(struct ioatdma_device *device)
{
	int i, src_idx;
	struct page *dest;
	struct page *xor_srcs[IOAT_NUM_SRC_TEST];
	struct page *xor_val_srcs[IOAT_NUM_SRC_TEST + 1];
	dma_addr_t dma_srcs[IOAT_NUM_SRC_TEST + 1];
	dma_addr_t dest_dma;
	struct dma_async_tx_descriptor *tx;
	struct dma_chan *dma_chan;
	dma_cookie_t cookie;
	u8 cmp_byte = 0;
	u32 cmp_word;
	u32 xor_val_result;
	int err = 0;
	struct completion cmp;
	unsigned long tmo;
	struct device *dev = &device->pdev->dev;
	struct dma_device *dma = &device->common;
	u8 op = 0;

	dev_dbg(dev, "%s\n", __func__);

	if (!dma_has_cap(DMA_XOR, dma->cap_mask))
		return 0;

	for (src_idx = 0; src_idx < IOAT_NUM_SRC_TEST; src_idx++) {
		xor_srcs[src_idx] = alloc_page(GFP_KERNEL);
		if (!xor_srcs[src_idx]) {
			while (src_idx--)
				__free_page(xor_srcs[src_idx]);
			return -ENOMEM;
		}
	}

	dest = alloc_page(GFP_KERNEL);
	if (!dest) {
		while (src_idx--)
			__free_page(xor_srcs[src_idx]);
		return -ENOMEM;
	}

	/* Fill in src buffers */
	for (src_idx = 0; src_idx < IOAT_NUM_SRC_TEST; src_idx++) {
		u8 *ptr = page_address(xor_srcs[src_idx]);
		for (i = 0; i < PAGE_SIZE; i++)
			ptr[i] = (1 << src_idx);
	}

	for (src_idx = 0; src_idx < IOAT_NUM_SRC_TEST; src_idx++)
		cmp_byte ^= (u8) (1 << src_idx);

	cmp_word = (cmp_byte << 24) | (cmp_byte << 16) |
			(cmp_byte << 8) | cmp_byte;

	memset(page_address(dest), 0, PAGE_SIZE);

	dma_chan = container_of(dma->channels.next, struct dma_chan,
				device_node);
	if (dma->device_alloc_chan_resources(dma_chan) < 1) {
		err = -ENODEV;
		goto out;
	}

	/* test xor */
	op = IOAT_OP_XOR;

	dest_dma = dma_map_page(dev, dest, 0, PAGE_SIZE, DMA_FROM_DEVICE);
	for (i = 0; i < IOAT_NUM_SRC_TEST; i++)
		dma_srcs[i] = dma_map_page(dev, xor_srcs[i], 0, PAGE_SIZE,
					   DMA_TO_DEVICE);
	tx = dma->device_prep_dma_xor(dma_chan, dest_dma, dma_srcs,
				      IOAT_NUM_SRC_TEST, PAGE_SIZE,
				      DMA_PREP_INTERRUPT);

	if (!tx) {
		dev_err(dev, "Self-test xor prep failed\n");
		err = -ENODEV;
		goto dma_unmap;
	}

	async_tx_ack(tx);
	init_completion(&cmp);
	tx->callback = ioat3_dma_test_callback;
	tx->callback_param = &cmp;
	cookie = tx->tx_submit(tx);
	if (cookie < 0) {
		dev_err(dev, "Self-test xor setup failed\n");
		err = -ENODEV;
		goto dma_unmap;
	}
	dma->device_issue_pending(dma_chan);

	tmo = wait_for_completion_timeout(&cmp, msecs_to_jiffies(3000));

	if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_COMPLETE) {
		dev_err(dev, "Self-test xor timed out\n");
		err = -ENODEV;
		goto dma_unmap;
	}

	dma_unmap_page(dev, dest_dma, PAGE_SIZE, DMA_FROM_DEVICE);
	for (i = 0; i < IOAT_NUM_SRC_TEST; i++)
		dma_unmap_page(dev, dma_srcs[i], PAGE_SIZE, DMA_TO_DEVICE);

	dma_sync_single_for_cpu(dev, dest_dma, PAGE_SIZE, DMA_FROM_DEVICE);
	for (i = 0; i < (PAGE_SIZE / sizeof(u32)); i++) {
		u32 *ptr = page_address(dest);
		if (ptr[i] != cmp_word) {
			dev_err(dev, "Self-test xor failed compare\n");
			err = -ENODEV;
			goto free_resources;
		}
	}
	dma_sync_single_for_device(dev, dest_dma, PAGE_SIZE, DMA_FROM_DEVICE);

	/* skip validate if the capability is not present */
	if (!dma_has_cap(DMA_XOR_VAL, dma_chan->device->cap_mask))
		goto free_resources;

	op = IOAT_OP_XOR_VAL;

	/* validate the sources with the destintation page */
	for (i = 0; i < IOAT_NUM_SRC_TEST; i++)
		xor_val_srcs[i] = xor_srcs[i];
	xor_val_srcs[i] = dest;

	xor_val_result = 1;

	for (i = 0; i < IOAT_NUM_SRC_TEST + 1; i++)
		dma_srcs[i] = dma_map_page(dev, xor_val_srcs[i], 0, PAGE_SIZE,
					   DMA_TO_DEVICE);
	tx = dma->device_prep_dma_xor_val(dma_chan, dma_srcs,
					  IOAT_NUM_SRC_TEST + 1, PAGE_SIZE,
					  &xor_val_result, DMA_PREP_INTERRUPT);
	if (!tx) {
		dev_err(dev, "Self-test zero prep failed\n");
		err = -ENODEV;
		goto dma_unmap;
	}

	async_tx_ack(tx);
	init_completion(&cmp);
	tx->callback = ioat3_dma_test_callback;
	tx->callback_param = &cmp;
	cookie = tx->tx_submit(tx);
	if (cookie < 0) {
		dev_err(dev, "Self-test zero setup failed\n");
		err = -ENODEV;
		goto dma_unmap;
	}
	dma->device_issue_pending(dma_chan);

	tmo = wait_for_completion_timeout(&cmp, msecs_to_jiffies(3000));

	if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_COMPLETE) {
		dev_err(dev, "Self-test validate timed out\n");
		err = -ENODEV;
		goto dma_unmap;
	}

	for (i = 0; i < IOAT_NUM_SRC_TEST + 1; i++)
		dma_unmap_page(dev, dma_srcs[i], PAGE_SIZE, DMA_TO_DEVICE);

	if (xor_val_result != 0) {
		dev_err(dev, "Self-test validate failed compare\n");
		err = -ENODEV;
		goto free_resources;
	}

	memset(page_address(dest), 0, PAGE_SIZE);

	/* test for non-zero parity sum */
	op = IOAT_OP_XOR_VAL;

	xor_val_result = 0;
	for (i = 0; i < IOAT_NUM_SRC_TEST + 1; i++)
		dma_srcs[i] = dma_map_page(dev, xor_val_srcs[i], 0, PAGE_SIZE,
					   DMA_TO_DEVICE);
	tx = dma->device_prep_dma_xor_val(dma_chan, dma_srcs,
					  IOAT_NUM_SRC_TEST + 1, PAGE_SIZE,
					  &xor_val_result, DMA_PREP_INTERRUPT);
	if (!tx) {
		dev_err(dev, "Self-test 2nd zero prep failed\n");
		err = -ENODEV;
		goto dma_unmap;
	}

	async_tx_ack(tx);
	init_completion(&cmp);
	tx->callback = ioat3_dma_test_callback;
	tx->callback_param = &cmp;
	cookie = tx->tx_submit(tx);
	if (cookie < 0) {
		dev_err(dev, "Self-test  2nd zero setup failed\n");
		err = -ENODEV;
		goto dma_unmap;
	}
	dma->device_issue_pending(dma_chan);

	tmo = wait_for_completion_timeout(&cmp, msecs_to_jiffies(3000));

	if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_COMPLETE) {
		dev_err(dev, "Self-test 2nd validate timed out\n");
		err = -ENODEV;
		goto dma_unmap;
	}

	if (xor_val_result != SUM_CHECK_P_RESULT) {
		dev_err(dev, "Self-test validate failed compare\n");
		err = -ENODEV;
		goto dma_unmap;
	}

	for (i = 0; i < IOAT_NUM_SRC_TEST + 1; i++)
		dma_unmap_page(dev, dma_srcs[i], PAGE_SIZE, DMA_TO_DEVICE);

	goto free_resources;
dma_unmap:
	if (op == IOAT_OP_XOR) {
		dma_unmap_page(dev, dest_dma, PAGE_SIZE, DMA_FROM_DEVICE);
		for (i = 0; i < IOAT_NUM_SRC_TEST; i++)
			dma_unmap_page(dev, dma_srcs[i], PAGE_SIZE,
				       DMA_TO_DEVICE);
	} else if (op == IOAT_OP_XOR_VAL) {
		for (i = 0; i < IOAT_NUM_SRC_TEST + 1; i++)
			dma_unmap_page(dev, dma_srcs[i], PAGE_SIZE,
				       DMA_TO_DEVICE);
	}
free_resources:
	dma->device_free_chan_resources(dma_chan);
out:
	src_idx = IOAT_NUM_SRC_TEST;
	while (src_idx--)
		__free_page(xor_srcs[src_idx]);
	__free_page(dest);
	return err;
}

static int ioat3_dma_self_test(struct ioatdma_device *device)
{
	int rc = ioat_dma_self_test(device);

	if (rc)
		return rc;

	rc = ioat_xor_val_self_test(device);
	if (rc)
		return rc;

	return 0;
}

static int ioat3_irq_reinit(struct ioatdma_device *device)
{
	struct pci_dev *pdev = device->pdev;
	int irq = pdev->irq, i;

	if (!is_bwd_ioat(pdev))
		return 0;

	switch (device->irq_mode) {
	case IOAT_MSIX:
		for (i = 0; i < device->common.chancnt; i++) {
			struct msix_entry *msix = &device->msix_entries[i];
			struct ioat_chan_common *chan;

			chan = ioat_chan_by_index(device, i);
			devm_free_irq(&pdev->dev, msix->vector, chan);
		}

		pci_disable_msix(pdev);
		break;
	case IOAT_MSI:
		pci_disable_msi(pdev);
		/* fall through */
	case IOAT_INTX:
		devm_free_irq(&pdev->dev, irq, device);
		break;
	default:
		return 0;
	}
	device->irq_mode = IOAT_NOIRQ;

	return ioat_dma_setup_interrupts(device);
}

static int ioat3_reset_hw(struct ioat_chan_common *chan)
{
	/* throw away whatever the channel was doing and get it
	 * initialized, with ioat3 specific workarounds
	 */
	struct ioatdma_device *device = chan->device;
	struct pci_dev *pdev = device->pdev;
	u32 chanerr;
	u16 dev_id;
	int err;

	ioat2_quiesce(chan, msecs_to_jiffies(100));

	chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
	writel(chanerr, chan->reg_base + IOAT_CHANERR_OFFSET);

	if (device->version < IOAT_VER_3_3) {
		/* clear any pending errors */
		err = pci_read_config_dword(pdev,
				IOAT_PCI_CHANERR_INT_OFFSET, &chanerr);
		if (err) {
			dev_err(&pdev->dev,
				"channel error register unreachable\n");
			return err;
		}
		pci_write_config_dword(pdev,
				IOAT_PCI_CHANERR_INT_OFFSET, chanerr);

		/* Clear DMAUNCERRSTS Cfg-Reg Parity Error status bit
		 * (workaround for spurious config parity error after restart)
		 */
		pci_read_config_word(pdev, IOAT_PCI_DEVICE_ID_OFFSET, &dev_id);
		if (dev_id == PCI_DEVICE_ID_INTEL_IOAT_TBG0) {
			pci_write_config_dword(pdev,
					       IOAT_PCI_DMAUNCERRSTS_OFFSET,
					       0x10);
		}
	}

	err = ioat2_reset_sync(chan, msecs_to_jiffies(200));
	if (!err)
		err = ioat3_irq_reinit(device);

	if (err)
		dev_err(&pdev->dev, "Failed to reset: %d\n", err);

	return err;
}

static void ioat3_intr_quirk(struct ioatdma_device *device)
{
	struct dma_device *dma;
	struct dma_chan *c;
	struct ioat_chan_common *chan;
	u32 errmask;

	dma = &device->common;

	/*
	 * if we have descriptor write back error status, we mask the
	 * error interrupts
	 */
	if (device->cap & IOAT_CAP_DWBES) {
		list_for_each_entry(c, &dma->channels, device_node) {
			chan = to_chan_common(c);
			errmask = readl(chan->reg_base +
					IOAT_CHANERR_MASK_OFFSET);
			errmask |= IOAT_CHANERR_XOR_P_OR_CRC_ERR |
				   IOAT_CHANERR_XOR_Q_ERR;
			writel(errmask, chan->reg_base +
					IOAT_CHANERR_MASK_OFFSET);
		}
	}
}

int ioat3_dma_probe(struct ioatdma_device *device, int dca)
{
	struct pci_dev *pdev = device->pdev;
	int dca_en = system_has_dca_enabled(pdev);
	struct dma_device *dma;
	struct dma_chan *c;
	struct ioat_chan_common *chan;
	bool is_raid_device = false;
	int err;

	device->enumerate_channels = ioat2_enumerate_channels;
	device->reset_hw = ioat3_reset_hw;
	device->self_test = ioat3_dma_self_test;
	device->intr_quirk = ioat3_intr_quirk;
	dma = &device->common;
	dma->device_prep_dma_memcpy = ioat2_dma_prep_memcpy_lock;
	dma->device_issue_pending = ioat2_issue_pending;
	dma->device_alloc_chan_resources = ioat2_alloc_chan_resources;
	dma->device_free_chan_resources = ioat2_free_chan_resources;

	dma_cap_set(DMA_INTERRUPT, dma->cap_mask);
	dma->device_prep_dma_interrupt = ioat3_prep_interrupt_lock;

	device->cap = readl(device->reg_base + IOAT_DMA_CAP_OFFSET);

	if (is_xeon_cb32(pdev) || is_bwd_noraid(pdev))
		device->cap &= ~(IOAT_CAP_XOR | IOAT_CAP_PQ | IOAT_CAP_RAID16SS);

	/* dca is incompatible with raid operations */
	if (dca_en && (device->cap & (IOAT_CAP_XOR|IOAT_CAP_PQ)))
		device->cap &= ~(IOAT_CAP_XOR|IOAT_CAP_PQ);

	if (device->cap & IOAT_CAP_XOR) {
		is_raid_device = true;
		dma->max_xor = 8;

		dma_cap_set(DMA_XOR, dma->cap_mask);
		dma->device_prep_dma_xor = ioat3_prep_xor;

		dma_cap_set(DMA_XOR_VAL, dma->cap_mask);
		dma->device_prep_dma_xor_val = ioat3_prep_xor_val;
	}

	if (device->cap & IOAT_CAP_PQ) {
		is_raid_device = true;

		dma->device_prep_dma_pq = ioat3_prep_pq;
		dma->device_prep_dma_pq_val = ioat3_prep_pq_val;
		dma_cap_set(DMA_PQ, dma->cap_mask);
		dma_cap_set(DMA_PQ_VAL, dma->cap_mask);

		if (device->cap & IOAT_CAP_RAID16SS) {
			dma_set_maxpq(dma, 16, 0);
		} else {
			dma_set_maxpq(dma, 8, 0);
		}

		if (!(device->cap & IOAT_CAP_XOR)) {
			dma->device_prep_dma_xor = ioat3_prep_pqxor;
			dma->device_prep_dma_xor_val = ioat3_prep_pqxor_val;
			dma_cap_set(DMA_XOR, dma->cap_mask);
			dma_cap_set(DMA_XOR_VAL, dma->cap_mask);

			if (device->cap & IOAT_CAP_RAID16SS) {
				dma->max_xor = 16;
			} else {
				dma->max_xor = 8;
			}
		}
	}

	dma->device_tx_status = ioat3_tx_status;
	device->cleanup_fn = ioat3_cleanup_event;
	device->timer_fn = ioat3_timer_event;

	/* starting with CB3.3 super extended descriptors are supported */
	if (device->cap & IOAT_CAP_RAID16SS) {
		char pool_name[14];
		int i;

		for (i = 0; i < MAX_SED_POOLS; i++) {
			snprintf(pool_name, 14, "ioat_hw%d_sed", i);

			/* allocate SED DMA pool */
			device->sed_hw_pool[i] = dmam_pool_create(pool_name,
					&pdev->dev,
					SED_SIZE * (i + 1), 64, 0);
			if (!device->sed_hw_pool[i])
				return -ENOMEM;

		}
	}

	err = ioat_probe(device);
	if (err)
		return err;

	list_for_each_entry(c, &dma->channels, device_node) {
		chan = to_chan_common(c);
		writel(IOAT_DMA_DCA_ANY_CPU,
		       chan->reg_base + IOAT_DCACTRL_OFFSET);
	}

	err = ioat_register(device);
	if (err)
		return err;

	ioat_kobject_add(device, &ioat2_ktype);

	if (dca)
		device->dca = ioat3_dca_init(pdev, device->reg_base);

	return 0;
}