aboutsummaryrefslogtreecommitdiffstats
path: root/target/linux/mediatek/patches-5.15/120-12-v5.19-spi-add-driver-for-MTK-SPI-NAND-Flash-Interface.patch
blob: b77b4ad4c473a32f0ef7c89e628eb3350af5a066 (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
From 8170bafa8936e9fbfdce992932a63bd20eca3bc3 Mon Sep 17 00:00:00 2001
From: Chuanhong Guo <gch981213@gmail.com>
Date: Sat, 2 Apr 2022 10:16:11 +0800
Subject: [PATCH v6 2/5] spi: add driver for MTK SPI NAND Flash Interface

This driver implements support for the SPI-NAND mode of MTK NAND Flash
Interface as a SPI-MEM controller with pipelined ECC capability.

Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
Tested-by: Daniel Golle <daniel@makrotopia.org>
---
Change since v1:
  fix CI warnings

Changes since v2:
 use streamed DMA api to avoid an extra memory copy during read
 make ECC engine config a per-nand context
 take user-requested ECC strength into account

Change since v3: none
Changes since v4:
 fix missing OOB write
 print page format with dev_dbg
 replace uint*_t copied from vendor driver with u*

Changes since v5:
 add missing nfi mode register configuration in probe
 fix an off-by-one bug in mtk_snand_mac_io

 drivers/spi/Kconfig        |   10 +
 drivers/spi/Makefile       |    1 +
 drivers/spi/spi-mtk-snfi.c | 1470 ++++++++++++++++++++++++++++++++++++
 3 files changed, 1481 insertions(+)
 create mode 100644 drivers/spi/spi-mtk-snfi.c

--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -530,6 +530,16 @@ config SPI_MTK_NOR
 	  SPI interface as well as several SPI NOR specific instructions
 	  via SPI MEM interface.
 
+config SPI_MTK_SNFI
+	tristate "MediaTek SPI NAND Flash Interface"
+	depends on ARCH_MEDIATEK || COMPILE_TEST
+	depends on MTD_NAND_ECC_MEDIATEK
+	help
+	  This enables support for SPI-NAND mode on the MediaTek NAND
+	  Flash Interface found on MediaTek ARM SoCs. This controller
+	  is implemented as a SPI-MEM controller with pipelined ECC
+	  capcability.
+
 config SPI_NPCM_FIU
 	tristate "Nuvoton NPCM FLASH Interface Unit"
 	depends on ARCH_NPCM || COMPILE_TEST
--- a/drivers/spi/Makefile
+++ b/drivers/spi/Makefile
@@ -71,6 +71,7 @@ obj-$(CONFIG_SPI_MPC52xx)		+= spi-mpc52x
 obj-$(CONFIG_SPI_MT65XX)                += spi-mt65xx.o
 obj-$(CONFIG_SPI_MT7621)		+= spi-mt7621.o
 obj-$(CONFIG_SPI_MTK_NOR)		+= spi-mtk-nor.o
+obj-$(CONFIG_SPI_MTK_SNFI)		+= spi-mtk-snfi.o
 obj-$(CONFIG_SPI_MXIC)			+= spi-mxic.o
 obj-$(CONFIG_SPI_MXS)			+= spi-mxs.o
 obj-$(CONFIG_SPI_NPCM_FIU)		+= spi-npcm-fiu.o
--- /dev/null
+++ b/drivers/spi/spi-mtk-snfi.c
@@ -0,0 +1,1470 @@
+// SPDX-License-Identifier: GPL-2.0
+//
+// Driver for the SPI-NAND mode of Mediatek NAND Flash Interface
+//
+// Copyright (c) 2022 Chuanhong Guo <gch981213@gmail.com>
+//
+// This driver is based on the SPI-NAND mtd driver from Mediatek SDK:
+//
+// Copyright (C) 2020 MediaTek Inc.
+// Author: Weijie Gao <weijie.gao@mediatek.com>
+//
+// This controller organize the page data as several interleaved sectors
+// like the following: (sizeof(FDM + ECC) = snf->nfi_cfg.spare_size)
+// +---------+------+------+---------+------+------+-----+
+// | Sector1 | FDM1 | ECC1 | Sector2 | FDM2 | ECC2 | ... |
+// +---------+------+------+---------+------+------+-----+
+// With auto-format turned on, DMA only returns this part:
+// +---------+---------+-----+
+// | Sector1 | Sector2 | ... |
+// +---------+---------+-----+
+// The FDM data will be filled to the registers, and ECC parity data isn't
+// accessible.
+// With auto-format off, all ((Sector+FDM+ECC)*nsectors) will be read over DMA
+// in it's original order shown in the first table. ECC can't be turned on when
+// auto-format is off.
+//
+// However, Linux SPI-NAND driver expects the data returned as:
+// +------+-----+
+// | Page | OOB |
+// +------+-----+
+// where the page data is continuously stored instead of interleaved.
+// So we assume all instructions matching the page_op template between ECC
+// prepare_io_req and finish_io_req are for page cache r/w.
+// Here's how this spi-mem driver operates when reading:
+//  1. Always set snf->autofmt = true in prepare_io_req (even when ECC is off).
+//  2. Perform page ops and let the controller fill the DMA bounce buffer with
+//     de-interleaved sector data and set FDM registers.
+//  3. Return the data as:
+//     +---------+---------+-----+------+------+-----+
+//     | Sector1 | Sector2 | ... | FDM1 | FDM2 | ... |
+//     +---------+---------+-----+------+------+-----+
+//  4. For other matching spi_mem ops outside a prepare/finish_io_req pair,
+//     read the data with auto-format off into the bounce buffer and copy
+//     needed data to the buffer specified in the request.
+//
+// Write requests operates in a similar manner.
+// As a limitation of this strategy, we won't be able to access any ECC parity
+// data at all in Linux.
+//
+// Here's the bad block mark situation on MTK chips:
+// In older chips like mt7622, MTK uses the first FDM byte in the first sector
+// as the bad block mark. After de-interleaving, this byte appears at [pagesize]
+// in the returned data, which is the BBM position expected by kernel. However,
+// the conventional bad block mark is the first byte of the OOB, which is part
+// of the last sector data in the interleaved layout. Instead of fixing their
+// hardware, MTK decided to address this inconsistency in software. On these
+// later chips, the BootROM expects the following:
+// 1. The [pagesize] byte on a nand page is used as BBM, which will appear at
+//    (page_size - (nsectors - 1) * spare_size) in the DMA buffer.
+// 2. The original byte stored at that position in the DMA buffer will be stored
+//    as the first byte of the FDM section in the last sector.
+// We can't disagree with the BootROM, so after de-interleaving, we need to
+// perform the following swaps in read:
+// 1. Store the BBM at [page_size - (nsectors - 1) * spare_size] to [page_size],
+//    which is the expected BBM position by kernel.
+// 2. Store the page data byte at [pagesize + (nsectors-1) * fdm] back to
+//    [page_size - (nsectors - 1) * spare_size]
+// Similarly, when writing, we need to perform swaps in the other direction.
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/device.h>
+#include <linux/mutex.h>
+#include <linux/clk.h>
+#include <linux/interrupt.h>
+#include <linux/dma-mapping.h>
+#include <linux/iopoll.h>
+#include <linux/of_platform.h>
+#include <linux/mtd/nand-ecc-mtk.h>
+#include <linux/spi/spi.h>
+#include <linux/spi/spi-mem.h>
+#include <linux/mtd/nand.h>
+
+// NFI registers
+#define NFI_CNFG 0x000
+#define CNFG_OP_MODE_S 12
+#define CNFG_OP_MODE_CUST 6
+#define CNFG_OP_MODE_PROGRAM 3
+#define CNFG_AUTO_FMT_EN BIT(9)
+#define CNFG_HW_ECC_EN BIT(8)
+#define CNFG_DMA_BURST_EN BIT(2)
+#define CNFG_READ_MODE BIT(1)
+#define CNFG_DMA_MODE BIT(0)
+
+#define NFI_PAGEFMT 0x0004
+#define NFI_SPARE_SIZE_LS_S 16
+#define NFI_FDM_ECC_NUM_S 12
+#define NFI_FDM_NUM_S 8
+#define NFI_SPARE_SIZE_S 4
+#define NFI_SEC_SEL_512 BIT(2)
+#define NFI_PAGE_SIZE_S 0
+#define NFI_PAGE_SIZE_512_2K 0
+#define NFI_PAGE_SIZE_2K_4K 1
+#define NFI_PAGE_SIZE_4K_8K 2
+#define NFI_PAGE_SIZE_8K_16K 3
+
+#define NFI_CON 0x008
+#define CON_SEC_NUM_S 12
+#define CON_BWR BIT(9)
+#define CON_BRD BIT(8)
+#define CON_NFI_RST BIT(1)
+#define CON_FIFO_FLUSH BIT(0)
+
+#define NFI_INTR_EN 0x010
+#define NFI_INTR_STA 0x014
+#define NFI_IRQ_INTR_EN BIT(31)
+#define NFI_IRQ_CUS_READ BIT(8)
+#define NFI_IRQ_CUS_PG BIT(7)
+
+#define NFI_CMD 0x020
+#define NFI_CMD_DUMMY_READ 0x00
+#define NFI_CMD_DUMMY_WRITE 0x80
+
+#define NFI_STRDATA 0x040
+#define STR_DATA BIT(0)
+
+#define NFI_STA 0x060
+#define NFI_NAND_FSM GENMASK(28, 24)
+#define NFI_FSM GENMASK(19, 16)
+#define READ_EMPTY BIT(12)
+
+#define NFI_FIFOSTA 0x064
+#define FIFO_WR_REMAIN_S 8
+#define FIFO_RD_REMAIN_S 0
+
+#define NFI_ADDRCNTR 0x070
+#define SEC_CNTR GENMASK(16, 12)
+#define SEC_CNTR_S 12
+#define NFI_SEC_CNTR(val) (((val)&SEC_CNTR) >> SEC_CNTR_S)
+
+#define NFI_STRADDR 0x080
+
+#define NFI_BYTELEN 0x084
+#define BUS_SEC_CNTR(val) (((val)&SEC_CNTR) >> SEC_CNTR_S)
+
+#define NFI_FDM0L 0x0a0
+#define NFI_FDM0M 0x0a4
+#define NFI_FDML(n) (NFI_FDM0L + (n)*8)
+#define NFI_FDMM(n) (NFI_FDM0M + (n)*8)
+
+#define NFI_DEBUG_CON1 0x220
+#define WBUF_EN BIT(2)
+
+#define NFI_MASTERSTA 0x224
+#define MAS_ADDR GENMASK(11, 9)
+#define MAS_RD GENMASK(8, 6)
+#define MAS_WR GENMASK(5, 3)
+#define MAS_RDDLY GENMASK(2, 0)
+#define NFI_MASTERSTA_MASK_7622 (MAS_ADDR | MAS_RD | MAS_WR | MAS_RDDLY)
+
+// SNFI registers
+#define SNF_MAC_CTL 0x500
+#define MAC_XIO_SEL BIT(4)
+#define SF_MAC_EN BIT(3)
+#define SF_TRIG BIT(2)
+#define WIP_READY BIT(1)
+#define WIP BIT(0)
+
+#define SNF_MAC_OUTL 0x504
+#define SNF_MAC_INL 0x508
+
+#define SNF_RD_CTL2 0x510
+#define DATA_READ_DUMMY_S 8
+#define DATA_READ_MAX_DUMMY 0xf
+#define DATA_READ_CMD_S 0
+
+#define SNF_RD_CTL3 0x514
+
+#define SNF_PG_CTL1 0x524
+#define PG_LOAD_CMD_S 8
+
+#define SNF_PG_CTL2 0x528
+
+#define SNF_MISC_CTL 0x538
+#define SW_RST BIT(28)
+#define FIFO_RD_LTC_S 25
+#define PG_LOAD_X4_EN BIT(20)
+#define DATA_READ_MODE_S 16
+#define DATA_READ_MODE GENMASK(18, 16)
+#define DATA_READ_MODE_X1 0
+#define DATA_READ_MODE_X2 1
+#define DATA_READ_MODE_X4 2
+#define DATA_READ_MODE_DUAL 5
+#define DATA_READ_MODE_QUAD 6
+#define PG_LOAD_CUSTOM_EN BIT(7)
+#define DATARD_CUSTOM_EN BIT(6)
+#define CS_DESELECT_CYC_S 0
+
+#define SNF_MISC_CTL2 0x53c
+#define PROGRAM_LOAD_BYTE_NUM_S 16
+#define READ_DATA_BYTE_NUM_S 11
+
+#define SNF_DLY_CTL3 0x548
+#define SFCK_SAM_DLY_S 0
+
+#define SNF_STA_CTL1 0x550
+#define CUS_PG_DONE BIT(28)
+#define CUS_READ_DONE BIT(27)
+#define SPI_STATE_S 0
+#define SPI_STATE GENMASK(3, 0)
+
+#define SNF_CFG 0x55c
+#define SPI_MODE BIT(0)
+
+#define SNF_GPRAM 0x800
+#define SNF_GPRAM_SIZE 0xa0
+
+#define SNFI_POLL_INTERVAL 1000000
+
+static const u8 mt7622_spare_sizes[] = { 16, 26, 27, 28 };
+
+struct mtk_snand_caps {
+	u16 sector_size;
+	u16 max_sectors;
+	u16 fdm_size;
+	u16 fdm_ecc_size;
+	u16 fifo_size;
+
+	bool bbm_swap;
+	bool empty_page_check;
+	u32 mastersta_mask;
+
+	const u8 *spare_sizes;
+	u32 num_spare_size;
+};
+
+static const struct mtk_snand_caps mt7622_snand_caps = {
+	.sector_size = 512,
+	.max_sectors = 8,
+	.fdm_size = 8,
+	.fdm_ecc_size = 1,
+	.fifo_size = 32,
+	.bbm_swap = false,
+	.empty_page_check = false,
+	.mastersta_mask = NFI_MASTERSTA_MASK_7622,
+	.spare_sizes = mt7622_spare_sizes,
+	.num_spare_size = ARRAY_SIZE(mt7622_spare_sizes)
+};
+
+static const struct mtk_snand_caps mt7629_snand_caps = {
+	.sector_size = 512,
+	.max_sectors = 8,
+	.fdm_size = 8,
+	.fdm_ecc_size = 1,
+	.fifo_size = 32,
+	.bbm_swap = true,
+	.empty_page_check = false,
+	.mastersta_mask = NFI_MASTERSTA_MASK_7622,
+	.spare_sizes = mt7622_spare_sizes,
+	.num_spare_size = ARRAY_SIZE(mt7622_spare_sizes)
+};
+
+struct mtk_snand_conf {
+	size_t page_size;
+	size_t oob_size;
+	u8 nsectors;
+	u8 spare_size;
+};
+
+struct mtk_snand {
+	struct spi_controller *ctlr;
+	struct device *dev;
+	struct clk *nfi_clk;
+	struct clk *pad_clk;
+	void __iomem *nfi_base;
+	int irq;
+	struct completion op_done;
+	const struct mtk_snand_caps *caps;
+	struct mtk_ecc_config *ecc_cfg;
+	struct mtk_ecc *ecc;
+	struct mtk_snand_conf nfi_cfg;
+	struct mtk_ecc_stats ecc_stats;
+	struct nand_ecc_engine ecc_eng;
+	bool autofmt;
+	u8 *buf;
+	size_t buf_len;
+};
+
+static struct mtk_snand *nand_to_mtk_snand(struct nand_device *nand)
+{
+	struct nand_ecc_engine *eng = nand->ecc.engine;
+
+	return container_of(eng, struct mtk_snand, ecc_eng);
+}
+
+static inline int snand_prepare_bouncebuf(struct mtk_snand *snf, size_t size)
+{
+	if (snf->buf_len >= size)
+		return 0;
+	kfree(snf->buf);
+	snf->buf = kmalloc(size, GFP_KERNEL);
+	if (!snf->buf)
+		return -ENOMEM;
+	snf->buf_len = size;
+	memset(snf->buf, 0xff, snf->buf_len);
+	return 0;
+}
+
+static inline u32 nfi_read32(struct mtk_snand *snf, u32 reg)
+{
+	return readl(snf->nfi_base + reg);
+}
+
+static inline void nfi_write32(struct mtk_snand *snf, u32 reg, u32 val)
+{
+	writel(val, snf->nfi_base + reg);
+}
+
+static inline void nfi_write16(struct mtk_snand *snf, u32 reg, u16 val)
+{
+	writew(val, snf->nfi_base + reg);
+}
+
+static inline void nfi_rmw32(struct mtk_snand *snf, u32 reg, u32 clr, u32 set)
+{
+	u32 val;
+
+	val = readl(snf->nfi_base + reg);
+	val &= ~clr;
+	val |= set;
+	writel(val, snf->nfi_base + reg);
+}
+
+static void nfi_read_data(struct mtk_snand *snf, u32 reg, u8 *data, u32 len)
+{
+	u32 i, val = 0, es = sizeof(u32);
+
+	for (i = reg; i < reg + len; i++) {
+		if (i == reg || i % es == 0)
+			val = nfi_read32(snf, i & ~(es - 1));
+
+		*data++ = (u8)(val >> (8 * (i % es)));
+	}
+}
+
+static int mtk_nfi_reset(struct mtk_snand *snf)
+{
+	u32 val, fifo_mask;
+	int ret;
+
+	nfi_write32(snf, NFI_CON, CON_FIFO_FLUSH | CON_NFI_RST);
+
+	ret = readw_poll_timeout(snf->nfi_base + NFI_MASTERSTA, val,
+				 !(val & snf->caps->mastersta_mask), 0,
+				 SNFI_POLL_INTERVAL);
+	if (ret) {
+		dev_err(snf->dev, "NFI master is still busy after reset\n");
+		return ret;
+	}
+
+	ret = readl_poll_timeout(snf->nfi_base + NFI_STA, val,
+				 !(val & (NFI_FSM | NFI_NAND_FSM)), 0,
+				 SNFI_POLL_INTERVAL);
+	if (ret) {
+		dev_err(snf->dev, "Failed to reset NFI\n");
+		return ret;
+	}
+
+	fifo_mask = ((snf->caps->fifo_size - 1) << FIFO_RD_REMAIN_S) |
+		    ((snf->caps->fifo_size - 1) << FIFO_WR_REMAIN_S);
+	ret = readw_poll_timeout(snf->nfi_base + NFI_FIFOSTA, val,
+				 !(val & fifo_mask), 0, SNFI_POLL_INTERVAL);
+	if (ret) {
+		dev_err(snf->dev, "NFI FIFOs are not empty\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int mtk_snand_mac_reset(struct mtk_snand *snf)
+{
+	int ret;
+	u32 val;
+
+	nfi_rmw32(snf, SNF_MISC_CTL, 0, SW_RST);
+
+	ret = readl_poll_timeout(snf->nfi_base + SNF_STA_CTL1, val,
+				 !(val & SPI_STATE), 0, SNFI_POLL_INTERVAL);
+	if (ret)
+		dev_err(snf->dev, "Failed to reset SNFI MAC\n");
+
+	nfi_write32(snf, SNF_MISC_CTL,
+		    (2 << FIFO_RD_LTC_S) | (10 << CS_DESELECT_CYC_S));
+
+	return ret;
+}
+
+static int mtk_snand_mac_trigger(struct mtk_snand *snf, u32 outlen, u32 inlen)
+{
+	int ret;
+	u32 val;
+
+	nfi_write32(snf, SNF_MAC_CTL, SF_MAC_EN);
+	nfi_write32(snf, SNF_MAC_OUTL, outlen);
+	nfi_write32(snf, SNF_MAC_INL, inlen);
+
+	nfi_write32(snf, SNF_MAC_CTL, SF_MAC_EN | SF_TRIG);
+
+	ret = readl_poll_timeout(snf->nfi_base + SNF_MAC_CTL, val,
+				 val & WIP_READY, 0, SNFI_POLL_INTERVAL);
+	if (ret) {
+		dev_err(snf->dev, "Timed out waiting for WIP_READY\n");
+		goto cleanup;
+	}
+
+	ret = readl_poll_timeout(snf->nfi_base + SNF_MAC_CTL, val, !(val & WIP),
+				 0, SNFI_POLL_INTERVAL);
+	if (ret)
+		dev_err(snf->dev, "Timed out waiting for WIP cleared\n");
+
+cleanup:
+	nfi_write32(snf, SNF_MAC_CTL, 0);
+
+	return ret;
+}
+
+static int mtk_snand_mac_io(struct mtk_snand *snf, const struct spi_mem_op *op)
+{
+	u32 rx_len = 0;
+	u32 reg_offs = 0;
+	u32 val = 0;
+	const u8 *tx_buf = NULL;
+	u8 *rx_buf = NULL;
+	int i, ret;
+	u8 b;
+
+	if (op->data.dir == SPI_MEM_DATA_IN) {
+		rx_len = op->data.nbytes;
+		rx_buf = op->data.buf.in;
+	} else {
+		tx_buf = op->data.buf.out;
+	}
+
+	mtk_snand_mac_reset(snf);
+
+	for (i = 0; i < op->cmd.nbytes; i++, reg_offs++) {
+		b = (op->cmd.opcode >> ((op->cmd.nbytes - i - 1) * 8)) & 0xff;
+		val |= b << (8 * (reg_offs % 4));
+		if (reg_offs % 4 == 3) {
+			nfi_write32(snf, SNF_GPRAM + reg_offs - 3, val);
+			val = 0;
+		}
+	}
+
+	for (i = 0; i < op->addr.nbytes; i++, reg_offs++) {
+		b = (op->addr.val >> ((op->addr.nbytes - i - 1) * 8)) & 0xff;
+		val |= b << (8 * (reg_offs % 4));
+		if (reg_offs % 4 == 3) {
+			nfi_write32(snf, SNF_GPRAM + reg_offs - 3, val);
+			val = 0;
+		}
+	}
+
+	for (i = 0; i < op->dummy.nbytes; i++, reg_offs++) {
+		if (reg_offs % 4 == 3) {
+			nfi_write32(snf, SNF_GPRAM + reg_offs - 3, val);
+			val = 0;
+		}
+	}
+
+	if (op->data.dir == SPI_MEM_DATA_OUT) {
+		for (i = 0; i < op->data.nbytes; i++, reg_offs++) {
+			val |= tx_buf[i] << (8 * (reg_offs % 4));
+			if (reg_offs % 4 == 3) {
+				nfi_write32(snf, SNF_GPRAM + reg_offs - 3, val);
+				val = 0;
+			}
+		}
+	}
+
+	if (reg_offs % 4)
+		nfi_write32(snf, SNF_GPRAM + (reg_offs & ~3), val);
+
+	for (i = 0; i < reg_offs; i += 4)
+		dev_dbg(snf->dev, "%d: %08X", i,
+			nfi_read32(snf, SNF_GPRAM + i));
+
+	dev_dbg(snf->dev, "SNF TX: %u RX: %u", reg_offs, rx_len);
+
+	ret = mtk_snand_mac_trigger(snf, reg_offs, rx_len);
+	if (ret)
+		return ret;
+
+	if (!rx_len)
+		return 0;
+
+	nfi_read_data(snf, SNF_GPRAM + reg_offs, rx_buf, rx_len);
+	return 0;
+}
+
+static int mtk_snand_setup_pagefmt(struct mtk_snand *snf, u32 page_size,
+				   u32 oob_size)
+{
+	int spare_idx = -1;
+	u32 spare_size, spare_size_shift, pagesize_idx;
+	u32 sector_size_512;
+	u8 nsectors;
+	int i;
+
+	// skip if it's already configured as required.
+	if (snf->nfi_cfg.page_size == page_size &&
+	    snf->nfi_cfg.oob_size == oob_size)
+		return 0;
+
+	nsectors = page_size / snf->caps->sector_size;
+	if (nsectors > snf->caps->max_sectors) {
+		dev_err(snf->dev, "too many sectors required.\n");
+		goto err;
+	}
+
+	if (snf->caps->sector_size == 512) {
+		sector_size_512 = NFI_SEC_SEL_512;
+		spare_size_shift = NFI_SPARE_SIZE_S;
+	} else {
+		sector_size_512 = 0;
+		spare_size_shift = NFI_SPARE_SIZE_LS_S;
+	}
+
+	switch (page_size) {
+	case SZ_512:
+		pagesize_idx = NFI_PAGE_SIZE_512_2K;
+		break;
+	case SZ_2K:
+		if (snf->caps->sector_size == 512)
+			pagesize_idx = NFI_PAGE_SIZE_2K_4K;
+		else
+			pagesize_idx = NFI_PAGE_SIZE_512_2K;
+		break;
+	case SZ_4K:
+		if (snf->caps->sector_size == 512)
+			pagesize_idx = NFI_PAGE_SIZE_4K_8K;
+		else
+			pagesize_idx = NFI_PAGE_SIZE_2K_4K;
+		break;
+	case SZ_8K:
+		if (snf->caps->sector_size == 512)
+			pagesize_idx = NFI_PAGE_SIZE_8K_16K;
+		else
+			pagesize_idx = NFI_PAGE_SIZE_4K_8K;
+		break;
+	case SZ_16K:
+		pagesize_idx = NFI_PAGE_SIZE_8K_16K;
+		break;
+	default:
+		dev_err(snf->dev, "unsupported page size.\n");
+		goto err;
+	}
+
+	spare_size = oob_size / nsectors;
+	// If we're using the 1KB sector size, HW will automatically double the
+	// spare size. We should only use half of the value in this case.
+	if (snf->caps->sector_size == 1024)
+		spare_size /= 2;
+
+	for (i = snf->caps->num_spare_size - 1; i >= 0; i--) {
+		if (snf->caps->spare_sizes[i] <= spare_size) {
+			spare_size = snf->caps->spare_sizes[i];
+			if (snf->caps->sector_size == 1024)
+				spare_size *= 2;
+			spare_idx = i;
+			break;
+		}
+	}
+
+	if (spare_idx < 0) {
+		dev_err(snf->dev, "unsupported spare size: %u\n", spare_size);
+		goto err;
+	}
+
+	nfi_write32(snf, NFI_PAGEFMT,
+		    (snf->caps->fdm_ecc_size << NFI_FDM_ECC_NUM_S) |
+			    (snf->caps->fdm_size << NFI_FDM_NUM_S) |
+			    (spare_idx << spare_size_shift) |
+			    (pagesize_idx << NFI_PAGE_SIZE_S) |
+			    sector_size_512);
+
+	snf->nfi_cfg.page_size = page_size;
+	snf->nfi_cfg.oob_size = oob_size;
+	snf->nfi_cfg.nsectors = nsectors;
+	snf->nfi_cfg.spare_size = spare_size;
+
+	dev_dbg(snf->dev, "page format: (%u + %u) * %u\n",
+		snf->caps->sector_size, spare_size, nsectors);
+	return snand_prepare_bouncebuf(snf, page_size + oob_size);
+err:
+	dev_err(snf->dev, "page size %u + %u is not supported\n", page_size,
+		oob_size);
+	return -EOPNOTSUPP;
+}
+
+static int mtk_snand_ooblayout_ecc(struct mtd_info *mtd, int section,
+				   struct mtd_oob_region *oobecc)
+{
+	// ECC area is not accessible
+	return -ERANGE;
+}
+
+static int mtk_snand_ooblayout_free(struct mtd_info *mtd, int section,
+				    struct mtd_oob_region *oobfree)
+{
+	struct nand_device *nand = mtd_to_nanddev(mtd);
+	struct mtk_snand *ms = nand_to_mtk_snand(nand);
+
+	if (section >= ms->nfi_cfg.nsectors)
+		return -ERANGE;
+
+	oobfree->length = ms->caps->fdm_size - 1;
+	oobfree->offset = section * ms->caps->fdm_size + 1;
+	return 0;
+}
+
+static const struct mtd_ooblayout_ops mtk_snand_ooblayout = {
+	.ecc = mtk_snand_ooblayout_ecc,
+	.free = mtk_snand_ooblayout_free,
+};
+
+static int mtk_snand_ecc_init_ctx(struct nand_device *nand)
+{
+	struct mtk_snand *snf = nand_to_mtk_snand(nand);
+	struct nand_ecc_props *conf = &nand->ecc.ctx.conf;
+	struct nand_ecc_props *reqs = &nand->ecc.requirements;
+	struct nand_ecc_props *user = &nand->ecc.user_conf;
+	struct mtd_info *mtd = nanddev_to_mtd(nand);
+	int step_size = 0, strength = 0, desired_correction = 0, steps;
+	bool ecc_user = false;
+	int ret;
+	u32 parity_bits, max_ecc_bytes;
+	struct mtk_ecc_config *ecc_cfg;
+
+	ret = mtk_snand_setup_pagefmt(snf, nand->memorg.pagesize,
+				      nand->memorg.oobsize);
+	if (ret)
+		return ret;
+
+	ecc_cfg = kzalloc(sizeof(*ecc_cfg), GFP_KERNEL);
+	if (!ecc_cfg)
+		return -ENOMEM;
+
+	nand->ecc.ctx.priv = ecc_cfg;
+
+	if (user->step_size && user->strength) {
+		step_size = user->step_size;
+		strength = user->strength;
+		ecc_user = true;
+	} else if (reqs->step_size && reqs->strength) {
+		step_size = reqs->step_size;
+		strength = reqs->strength;
+	}
+
+	if (step_size && strength) {
+		steps = mtd->writesize / step_size;
+		desired_correction = steps * strength;
+		strength = desired_correction / snf->nfi_cfg.nsectors;
+	}
+
+	ecc_cfg->mode = ECC_NFI_MODE;
+	ecc_cfg->sectors = snf->nfi_cfg.nsectors;
+	ecc_cfg->len = snf->caps->sector_size + snf->caps->fdm_ecc_size;
+
+	// calculate the max possible strength under current page format
+	parity_bits = mtk_ecc_get_parity_bits(snf->ecc);
+	max_ecc_bytes = snf->nfi_cfg.spare_size - snf->caps->fdm_size;
+	ecc_cfg->strength = max_ecc_bytes * 8 / parity_bits;
+	mtk_ecc_adjust_strength(snf->ecc, &ecc_cfg->strength);
+
+	// if there's a user requested strength, find the minimum strength that
+	// meets the requirement. Otherwise use the maximum strength which is
+	// expected by BootROM.
+	if (ecc_user && strength) {
+		u32 s_next = ecc_cfg->strength - 1;
+
+		while (1) {
+			mtk_ecc_adjust_strength(snf->ecc, &s_next);
+			if (s_next >= ecc_cfg->strength)
+				break;
+			if (s_next < strength)
+				break;
+			s_next = ecc_cfg->strength - 1;
+		}
+	}
+
+	mtd_set_ooblayout(mtd, &mtk_snand_ooblayout);
+
+	conf->step_size = snf->caps->sector_size;
+	conf->strength = ecc_cfg->strength;
+
+	if (ecc_cfg->strength < strength)
+		dev_warn(snf->dev, "unable to fulfill ECC of %u bits.\n",
+			 strength);
+	dev_info(snf->dev, "ECC strength: %u bits per %u bytes\n",
+		 ecc_cfg->strength, snf->caps->sector_size);
+
+	return 0;
+}
+
+static void mtk_snand_ecc_cleanup_ctx(struct nand_device *nand)
+{
+	struct mtk_ecc_config *ecc_cfg = nand_to_ecc_ctx(nand);
+
+	kfree(ecc_cfg);
+}
+
+static int mtk_snand_ecc_prepare_io_req(struct nand_device *nand,
+					struct nand_page_io_req *req)
+{
+	struct mtk_snand *snf = nand_to_mtk_snand(nand);
+	struct mtk_ecc_config *ecc_cfg = nand_to_ecc_ctx(nand);
+	int ret;
+
+	ret = mtk_snand_setup_pagefmt(snf, nand->memorg.pagesize,
+				      nand->memorg.oobsize);
+	if (ret)
+		return ret;
+	snf->autofmt = true;
+	snf->ecc_cfg = ecc_cfg;
+	return 0;
+}
+
+static int mtk_snand_ecc_finish_io_req(struct nand_device *nand,
+				       struct nand_page_io_req *req)
+{
+	struct mtk_snand *snf = nand_to_mtk_snand(nand);
+	struct mtd_info *mtd = nanddev_to_mtd(nand);
+
+	snf->ecc_cfg = NULL;
+	snf->autofmt = false;
+	if ((req->mode == MTD_OPS_RAW) || (req->type != NAND_PAGE_READ))
+		return 0;
+
+	if (snf->ecc_stats.failed)
+		mtd->ecc_stats.failed += snf->ecc_stats.failed;
+	mtd->ecc_stats.corrected += snf->ecc_stats.corrected;
+	return snf->ecc_stats.failed ? -EBADMSG : snf->ecc_stats.bitflips;
+}
+
+static struct nand_ecc_engine_ops mtk_snfi_ecc_engine_ops = {
+	.init_ctx = mtk_snand_ecc_init_ctx,
+	.cleanup_ctx = mtk_snand_ecc_cleanup_ctx,
+	.prepare_io_req = mtk_snand_ecc_prepare_io_req,
+	.finish_io_req = mtk_snand_ecc_finish_io_req,
+};
+
+static void mtk_snand_read_fdm(struct mtk_snand *snf, u8 *buf)
+{
+	u32 vall, valm;
+	u8 *oobptr = buf;
+	int i, j;
+
+	for (i = 0; i < snf->nfi_cfg.nsectors; i++) {
+		vall = nfi_read32(snf, NFI_FDML(i));
+		valm = nfi_read32(snf, NFI_FDMM(i));
+
+		for (j = 0; j < snf->caps->fdm_size; j++)
+			oobptr[j] = (j >= 4 ? valm : vall) >> ((j % 4) * 8);
+
+		oobptr += snf->caps->fdm_size;
+	}
+}
+
+static void mtk_snand_write_fdm(struct mtk_snand *snf, const u8 *buf)
+{
+	u32 fdm_size = snf->caps->fdm_size;
+	const u8 *oobptr = buf;
+	u32 vall, valm;
+	int i, j;
+
+	for (i = 0; i < snf->nfi_cfg.nsectors; i++) {
+		vall = 0;
+		valm = 0;
+
+		for (j = 0; j < 8; j++) {
+			if (j < 4)
+				vall |= (j < fdm_size ? oobptr[j] : 0xff)
+					<< (j * 8);
+			else
+				valm |= (j < fdm_size ? oobptr[j] : 0xff)
+					<< ((j - 4) * 8);
+		}
+
+		nfi_write32(snf, NFI_FDML(i), vall);
+		nfi_write32(snf, NFI_FDMM(i), valm);
+
+		oobptr += fdm_size;
+	}
+}
+
+static void mtk_snand_bm_swap(struct mtk_snand *snf, u8 *buf)
+{
+	u32 buf_bbm_pos, fdm_bbm_pos;
+
+	if (!snf->caps->bbm_swap || snf->nfi_cfg.nsectors == 1)
+		return;
+
+	// swap [pagesize] byte on nand with the first fdm byte
+	// in the last sector.
+	buf_bbm_pos = snf->nfi_cfg.page_size -
+		      (snf->nfi_cfg.nsectors - 1) * snf->nfi_cfg.spare_size;
+	fdm_bbm_pos = snf->nfi_cfg.page_size +
+		      (snf->nfi_cfg.nsectors - 1) * snf->caps->fdm_size;
+
+	swap(snf->buf[fdm_bbm_pos], buf[buf_bbm_pos]);
+}
+
+static void mtk_snand_fdm_bm_swap(struct mtk_snand *snf)
+{
+	u32 fdm_bbm_pos1, fdm_bbm_pos2;
+
+	if (!snf->caps->bbm_swap || snf->nfi_cfg.nsectors == 1)
+		return;
+
+	// swap the first fdm byte in the first and the last sector.
+	fdm_bbm_pos1 = snf->nfi_cfg.page_size;
+	fdm_bbm_pos2 = snf->nfi_cfg.page_size +
+		       (snf->nfi_cfg.nsectors - 1) * snf->caps->fdm_size;
+	swap(snf->buf[fdm_bbm_pos1], snf->buf[fdm_bbm_pos2]);
+}
+
+static int mtk_snand_read_page_cache(struct mtk_snand *snf,
+				     const struct spi_mem_op *op)
+{
+	u8 *buf = snf->buf;
+	u8 *buf_fdm = buf + snf->nfi_cfg.page_size;
+	// the address part to be sent by the controller
+	u32 op_addr = op->addr.val;
+	// where to start copying data from bounce buffer
+	u32 rd_offset = 0;
+	u32 dummy_clk = (op->dummy.nbytes * BITS_PER_BYTE / op->dummy.buswidth);
+	u32 op_mode = 0;
+	u32 dma_len = snf->buf_len;
+	int ret = 0;
+	u32 rd_mode, rd_bytes, val;
+	dma_addr_t buf_dma;
+
+	if (snf->autofmt) {
+		u32 last_bit;
+		u32 mask;
+
+		dma_len = snf->nfi_cfg.page_size;
+		op_mode = CNFG_AUTO_FMT_EN;
+		if (op->data.ecc)
+			op_mode |= CNFG_HW_ECC_EN;
+		// extract the plane bit:
+		// Find the highest bit set in (pagesize+oobsize).
+		// Bits higher than that in op->addr are kept and sent over SPI
+		// Lower bits are used as an offset for copying data from DMA
+		// bounce buffer.
+		last_bit = fls(snf->nfi_cfg.page_size + snf->nfi_cfg.oob_size);
+		mask = (1 << last_bit) - 1;
+		rd_offset = op_addr & mask;
+		op_addr &= ~mask;
+
+		// check if we can dma to the caller memory
+		if (rd_offset == 0 && op->data.nbytes >= snf->nfi_cfg.page_size)
+			buf = op->data.buf.in;
+	}
+	mtk_snand_mac_reset(snf);
+	mtk_nfi_reset(snf);
+
+	// command and dummy cycles
+	nfi_write32(snf, SNF_RD_CTL2,
+		    (dummy_clk << DATA_READ_DUMMY_S) |
+			    (op->cmd.opcode << DATA_READ_CMD_S));
+
+	// read address
+	nfi_write32(snf, SNF_RD_CTL3, op_addr);
+
+	// Set read op_mode
+	if (op->data.buswidth == 4)
+		rd_mode = op->addr.buswidth == 4 ? DATA_READ_MODE_QUAD :
+						   DATA_READ_MODE_X4;
+	else if (op->data.buswidth == 2)
+		rd_mode = op->addr.buswidth == 2 ? DATA_READ_MODE_DUAL :
+						   DATA_READ_MODE_X2;
+	else
+		rd_mode = DATA_READ_MODE_X1;
+	rd_mode <<= DATA_READ_MODE_S;
+	nfi_rmw32(snf, SNF_MISC_CTL, DATA_READ_MODE,
+		  rd_mode | DATARD_CUSTOM_EN);
+
+	// Set bytes to read
+	rd_bytes = (snf->nfi_cfg.spare_size + snf->caps->sector_size) *
+		   snf->nfi_cfg.nsectors;
+	nfi_write32(snf, SNF_MISC_CTL2,
+		    (rd_bytes << PROGRAM_LOAD_BYTE_NUM_S) | rd_bytes);
+
+	// NFI read prepare
+	nfi_write16(snf, NFI_CNFG,
+		    (CNFG_OP_MODE_CUST << CNFG_OP_MODE_S) | CNFG_DMA_BURST_EN |
+			    CNFG_READ_MODE | CNFG_DMA_MODE | op_mode);
+
+	nfi_write32(snf, NFI_CON, (snf->nfi_cfg.nsectors << CON_SEC_NUM_S));
+
+	buf_dma = dma_map_single(snf->dev, buf, dma_len, DMA_FROM_DEVICE);
+	if (dma_mapping_error(snf->dev, buf_dma)) {
+		dev_err(snf->dev, "DMA mapping failed.\n");
+		goto cleanup;
+	}
+	nfi_write32(snf, NFI_STRADDR, buf_dma);
+	if (op->data.ecc) {
+		snf->ecc_cfg->op = ECC_DECODE;
+		ret = mtk_ecc_enable(snf->ecc, snf->ecc_cfg);
+		if (ret)
+			goto cleanup_dma;
+	}
+	// Prepare for custom read interrupt
+	nfi_write32(snf, NFI_INTR_EN, NFI_IRQ_INTR_EN | NFI_IRQ_CUS_READ);
+	reinit_completion(&snf->op_done);
+
+	// Trigger NFI into custom mode
+	nfi_write16(snf, NFI_CMD, NFI_CMD_DUMMY_READ);
+
+	// Start DMA read
+	nfi_rmw32(snf, NFI_CON, 0, CON_BRD);
+	nfi_write16(snf, NFI_STRDATA, STR_DATA);
+
+	if (!wait_for_completion_timeout(
+		    &snf->op_done, usecs_to_jiffies(SNFI_POLL_INTERVAL))) {
+		dev_err(snf->dev, "DMA timed out for reading from cache.\n");
+		ret = -ETIMEDOUT;
+		goto cleanup;
+	}
+
+	// Wait for BUS_SEC_CNTR returning expected value
+	ret = readl_poll_timeout(snf->nfi_base + NFI_BYTELEN, val,
+				 BUS_SEC_CNTR(val) >= snf->nfi_cfg.nsectors, 0,
+				 SNFI_POLL_INTERVAL);
+	if (ret) {
+		dev_err(snf->dev, "Timed out waiting for BUS_SEC_CNTR\n");
+		goto cleanup2;
+	}
+
+	// Wait for bus becoming idle
+	ret = readl_poll_timeout(snf->nfi_base + NFI_MASTERSTA, val,
+				 !(val & snf->caps->mastersta_mask), 0,
+				 SNFI_POLL_INTERVAL);
+	if (ret) {
+		dev_err(snf->dev, "Timed out waiting for bus becoming idle\n");
+		goto cleanup2;
+	}
+
+	if (op->data.ecc) {
+		ret = mtk_ecc_wait_done(snf->ecc, ECC_DECODE);
+		if (ret) {
+			dev_err(snf->dev, "wait ecc done timeout\n");
+			goto cleanup2;
+		}
+		// save status before disabling ecc
+		mtk_ecc_get_stats(snf->ecc, &snf->ecc_stats,
+				  snf->nfi_cfg.nsectors);
+	}
+
+	dma_unmap_single(snf->dev, buf_dma, dma_len, DMA_FROM_DEVICE);
+
+	if (snf->autofmt) {
+		mtk_snand_read_fdm(snf, buf_fdm);
+		if (snf->caps->bbm_swap) {
+			mtk_snand_bm_swap(snf, buf);
+			mtk_snand_fdm_bm_swap(snf);
+		}
+	}
+
+	// copy data back
+	if (nfi_read32(snf, NFI_STA) & READ_EMPTY) {
+		memset(op->data.buf.in, 0xff, op->data.nbytes);
+		snf->ecc_stats.bitflips = 0;
+		snf->ecc_stats.failed = 0;
+		snf->ecc_stats.corrected = 0;
+	} else {
+		if (buf == op->data.buf.in) {
+			u32 cap_len = snf->buf_len - snf->nfi_cfg.page_size;
+			u32 req_left = op->data.nbytes - snf->nfi_cfg.page_size;
+
+			if (req_left)
+				memcpy(op->data.buf.in + snf->nfi_cfg.page_size,
+				       buf_fdm,
+				       cap_len < req_left ? cap_len : req_left);
+		} else if (rd_offset < snf->buf_len) {
+			u32 cap_len = snf->buf_len - rd_offset;
+
+			if (op->data.nbytes < cap_len)
+				cap_len = op->data.nbytes;
+			memcpy(op->data.buf.in, snf->buf + rd_offset, cap_len);
+		}
+	}
+cleanup2:
+	if (op->data.ecc)
+		mtk_ecc_disable(snf->ecc);
+cleanup_dma:
+	// unmap dma only if any error happens. (otherwise it's done before
+	// data copying)
+	if (ret)
+		dma_unmap_single(snf->dev, buf_dma, dma_len, DMA_FROM_DEVICE);
+cleanup:
+	// Stop read
+	nfi_write32(snf, NFI_CON, 0);
+	nfi_write16(snf, NFI_CNFG, 0);
+
+	// Clear SNF done flag
+	nfi_rmw32(snf, SNF_STA_CTL1, 0, CUS_READ_DONE);
+	nfi_write32(snf, SNF_STA_CTL1, 0);
+
+	// Disable interrupt
+	nfi_read32(snf, NFI_INTR_STA);
+	nfi_write32(snf, NFI_INTR_EN, 0);
+
+	nfi_rmw32(snf, SNF_MISC_CTL, DATARD_CUSTOM_EN, 0);
+	return ret;
+}
+
+static int mtk_snand_write_page_cache(struct mtk_snand *snf,
+				      const struct spi_mem_op *op)
+{
+	// the address part to be sent by the controller
+	u32 op_addr = op->addr.val;
+	// where to start copying data from bounce buffer
+	u32 wr_offset = 0;
+	u32 op_mode = 0;
+	int ret = 0;
+	u32 wr_mode = 0;
+	u32 dma_len = snf->buf_len;
+	u32 wr_bytes, val;
+	size_t cap_len;
+	dma_addr_t buf_dma;
+
+	if (snf->autofmt) {
+		u32 last_bit;
+		u32 mask;
+
+		dma_len = snf->nfi_cfg.page_size;
+		op_mode = CNFG_AUTO_FMT_EN;
+		if (op->data.ecc)
+			op_mode |= CNFG_HW_ECC_EN;
+
+		last_bit = fls(snf->nfi_cfg.page_size + snf->nfi_cfg.oob_size);
+		mask = (1 << last_bit) - 1;
+		wr_offset = op_addr & mask;
+		op_addr &= ~mask;
+	}
+	mtk_snand_mac_reset(snf);
+	mtk_nfi_reset(snf);
+
+	if (wr_offset)
+		memset(snf->buf, 0xff, wr_offset);
+
+	cap_len = snf->buf_len - wr_offset;
+	if (op->data.nbytes < cap_len)
+		cap_len = op->data.nbytes;
+	memcpy(snf->buf + wr_offset, op->data.buf.out, cap_len);
+	if (snf->autofmt) {
+		if (snf->caps->bbm_swap) {
+			mtk_snand_fdm_bm_swap(snf);
+			mtk_snand_bm_swap(snf, snf->buf);
+		}
+		mtk_snand_write_fdm(snf, snf->buf + snf->nfi_cfg.page_size);
+	}
+
+	// Command
+	nfi_write32(snf, SNF_PG_CTL1, (op->cmd.opcode << PG_LOAD_CMD_S));
+
+	// write address
+	nfi_write32(snf, SNF_PG_CTL2, op_addr);
+
+	// Set read op_mode
+	if (op->data.buswidth == 4)
+		wr_mode = PG_LOAD_X4_EN;
+
+	nfi_rmw32(snf, SNF_MISC_CTL, PG_LOAD_X4_EN,
+		  wr_mode | PG_LOAD_CUSTOM_EN);
+
+	// Set bytes to write
+	wr_bytes = (snf->nfi_cfg.spare_size + snf->caps->sector_size) *
+		   snf->nfi_cfg.nsectors;
+	nfi_write32(snf, SNF_MISC_CTL2,
+		    (wr_bytes << PROGRAM_LOAD_BYTE_NUM_S) | wr_bytes);
+
+	// NFI write prepare
+	nfi_write16(snf, NFI_CNFG,
+		    (CNFG_OP_MODE_PROGRAM << CNFG_OP_MODE_S) |
+			    CNFG_DMA_BURST_EN | CNFG_DMA_MODE | op_mode);
+
+	nfi_write32(snf, NFI_CON, (snf->nfi_cfg.nsectors << CON_SEC_NUM_S));
+	buf_dma = dma_map_single(snf->dev, snf->buf, dma_len, DMA_TO_DEVICE);
+	if (dma_mapping_error(snf->dev, buf_dma)) {
+		dev_err(snf->dev, "DMA mapping failed.\n");
+		goto cleanup;
+	}
+	nfi_write32(snf, NFI_STRADDR, buf_dma);
+	if (op->data.ecc) {
+		snf->ecc_cfg->op = ECC_ENCODE;
+		ret = mtk_ecc_enable(snf->ecc, snf->ecc_cfg);
+		if (ret)
+			goto cleanup_dma;
+	}
+	// Prepare for custom write interrupt
+	nfi_write32(snf, NFI_INTR_EN, NFI_IRQ_INTR_EN | NFI_IRQ_CUS_PG);
+	reinit_completion(&snf->op_done);
+	;
+
+	// Trigger NFI into custom mode
+	nfi_write16(snf, NFI_CMD, NFI_CMD_DUMMY_WRITE);
+
+	// Start DMA write
+	nfi_rmw32(snf, NFI_CON, 0, CON_BWR);
+	nfi_write16(snf, NFI_STRDATA, STR_DATA);
+
+	if (!wait_for_completion_timeout(
+		    &snf->op_done, usecs_to_jiffies(SNFI_POLL_INTERVAL))) {
+		dev_err(snf->dev, "DMA timed out for program load.\n");
+		ret = -ETIMEDOUT;
+		goto cleanup_ecc;
+	}
+
+	// Wait for NFI_SEC_CNTR returning expected value
+	ret = readl_poll_timeout(snf->nfi_base + NFI_ADDRCNTR, val,
+				 NFI_SEC_CNTR(val) >= snf->nfi_cfg.nsectors, 0,
+				 SNFI_POLL_INTERVAL);
+	if (ret)
+		dev_err(snf->dev, "Timed out waiting for NFI_SEC_CNTR\n");
+
+cleanup_ecc:
+	if (op->data.ecc)
+		mtk_ecc_disable(snf->ecc);
+cleanup_dma:
+	dma_unmap_single(snf->dev, buf_dma, dma_len, DMA_TO_DEVICE);
+cleanup:
+	// Stop write
+	nfi_write32(snf, NFI_CON, 0);
+	nfi_write16(snf, NFI_CNFG, 0);
+
+	// Clear SNF done flag
+	nfi_rmw32(snf, SNF_STA_CTL1, 0, CUS_PG_DONE);
+	nfi_write32(snf, SNF_STA_CTL1, 0);
+
+	// Disable interrupt
+	nfi_read32(snf, NFI_INTR_STA);
+	nfi_write32(snf, NFI_INTR_EN, 0);
+
+	nfi_rmw32(snf, SNF_MISC_CTL, PG_LOAD_CUSTOM_EN, 0);
+
+	return ret;
+}
+
+/**
+ * mtk_snand_is_page_ops() - check if the op is a controller supported page op.
+ * @op spi-mem op to check
+ *
+ * Check whether op can be executed with read_from_cache or program_load
+ * mode in the controller.
+ * This controller can execute typical Read From Cache and Program Load
+ * instructions found on SPI-NAND with 2-byte address.
+ * DTR and cmd buswidth & nbytes should be checked before calling this.
+ *
+ * Return: true if the op matches the instruction template
+ */
+static bool mtk_snand_is_page_ops(const struct spi_mem_op *op)
+{
+	if (op->addr.nbytes != 2)
+		return false;
+
+	if (op->addr.buswidth != 1 && op->addr.buswidth != 2 &&
+	    op->addr.buswidth != 4)
+		return false;
+
+	// match read from page instructions
+	if (op->data.dir == SPI_MEM_DATA_IN) {
+		// check dummy cycle first
+		if (op->dummy.nbytes * BITS_PER_BYTE / op->dummy.buswidth >
+		    DATA_READ_MAX_DUMMY)
+			return false;
+		// quad io / quad out
+		if ((op->addr.buswidth == 4 || op->addr.buswidth == 1) &&
+		    op->data.buswidth == 4)
+			return true;
+
+		// dual io / dual out
+		if ((op->addr.buswidth == 2 || op->addr.buswidth == 1) &&
+		    op->data.buswidth == 2)
+			return true;
+
+		// standard spi
+		if (op->addr.buswidth == 1 && op->data.buswidth == 1)
+			return true;
+	} else if (op->data.dir == SPI_MEM_DATA_OUT) {
+		// check dummy cycle first
+		if (op->dummy.nbytes)
+			return false;
+		// program load quad out
+		if (op->addr.buswidth == 1 && op->data.buswidth == 4)
+			return true;
+		// standard spi
+		if (op->addr.buswidth == 1 && op->data.buswidth == 1)
+			return true;
+	}
+	return false;
+}
+
+static bool mtk_snand_supports_op(struct spi_mem *mem,
+				  const struct spi_mem_op *op)
+{
+	if (!spi_mem_default_supports_op(mem, op))
+		return false;
+	if (op->cmd.nbytes != 1 || op->cmd.buswidth != 1)
+		return false;
+	if (mtk_snand_is_page_ops(op))
+		return true;
+	return ((op->addr.nbytes == 0 || op->addr.buswidth == 1) &&
+		(op->dummy.nbytes == 0 || op->dummy.buswidth == 1) &&
+		(op->data.nbytes == 0 || op->data.buswidth == 1));
+}
+
+static int mtk_snand_adjust_op_size(struct spi_mem *mem, struct spi_mem_op *op)
+{
+	struct mtk_snand *ms = spi_controller_get_devdata(mem->spi->master);
+	// page ops transfer size must be exactly ((sector_size + spare_size) *
+	// nsectors). Limit the op size if the caller requests more than that.
+	// exec_op will read more than needed and discard the leftover if the
+	// caller requests less data.
+	if (mtk_snand_is_page_ops(op)) {
+		size_t l;
+		// skip adjust_op_size for page ops
+		if (ms->autofmt)
+			return 0;
+		l = ms->caps->sector_size + ms->nfi_cfg.spare_size;
+		l *= ms->nfi_cfg.nsectors;
+		if (op->data.nbytes > l)
+			op->data.nbytes = l;
+	} else {
+		size_t hl = op->cmd.nbytes + op->addr.nbytes + op->dummy.nbytes;
+
+		if (hl >= SNF_GPRAM_SIZE)
+			return -EOPNOTSUPP;
+		if (op->data.nbytes > SNF_GPRAM_SIZE - hl)
+			op->data.nbytes = SNF_GPRAM_SIZE - hl;
+	}
+	return 0;
+}
+
+static int mtk_snand_exec_op(struct spi_mem *mem, const struct spi_mem_op *op)
+{
+	struct mtk_snand *ms = spi_controller_get_devdata(mem->spi->master);
+
+	dev_dbg(ms->dev, "OP %02x ADDR %08llX@%d:%u DATA %d:%u", op->cmd.opcode,
+		op->addr.val, op->addr.buswidth, op->addr.nbytes,
+		op->data.buswidth, op->data.nbytes);
+	if (mtk_snand_is_page_ops(op)) {
+		if (op->data.dir == SPI_MEM_DATA_IN)
+			return mtk_snand_read_page_cache(ms, op);
+		else
+			return mtk_snand_write_page_cache(ms, op);
+	} else {
+		return mtk_snand_mac_io(ms, op);
+	}
+}
+
+static const struct spi_controller_mem_ops mtk_snand_mem_ops = {
+	.adjust_op_size = mtk_snand_adjust_op_size,
+	.supports_op = mtk_snand_supports_op,
+	.exec_op = mtk_snand_exec_op,
+};
+
+static const struct spi_controller_mem_caps mtk_snand_mem_caps = {
+	.ecc = true,
+};
+
+static irqreturn_t mtk_snand_irq(int irq, void *id)
+{
+	struct mtk_snand *snf = id;
+	u32 sta, ien;
+
+	sta = nfi_read32(snf, NFI_INTR_STA);
+	ien = nfi_read32(snf, NFI_INTR_EN);
+
+	if (!(sta & ien))
+		return IRQ_NONE;
+
+	nfi_write32(snf, NFI_INTR_EN, 0);
+	complete(&snf->op_done);
+	return IRQ_HANDLED;
+}
+
+static const struct of_device_id mtk_snand_ids[] = {
+	{ .compatible = "mediatek,mt7622-snand", .data = &mt7622_snand_caps },
+	{ .compatible = "mediatek,mt7629-snand", .data = &mt7629_snand_caps },
+	{},
+};
+
+MODULE_DEVICE_TABLE(of, mtk_snand_ids);
+
+static int mtk_snand_enable_clk(struct mtk_snand *ms)
+{
+	int ret;
+
+	ret = clk_prepare_enable(ms->nfi_clk);
+	if (ret) {
+		dev_err(ms->dev, "unable to enable nfi clk\n");
+		return ret;
+	}
+	ret = clk_prepare_enable(ms->pad_clk);
+	if (ret) {
+		dev_err(ms->dev, "unable to enable pad clk\n");
+		goto err1;
+	}
+	return 0;
+err1:
+	clk_disable_unprepare(ms->nfi_clk);
+	return ret;
+}
+
+static void mtk_snand_disable_clk(struct mtk_snand *ms)
+{
+	clk_disable_unprepare(ms->pad_clk);
+	clk_disable_unprepare(ms->nfi_clk);
+}
+
+static int mtk_snand_probe(struct platform_device *pdev)
+{
+	struct device_node *np = pdev->dev.of_node;
+	const struct of_device_id *dev_id;
+	struct spi_controller *ctlr;
+	struct mtk_snand *ms;
+	int ret;
+
+	dev_id = of_match_node(mtk_snand_ids, np);
+	if (!dev_id)
+		return -EINVAL;
+
+	ctlr = devm_spi_alloc_master(&pdev->dev, sizeof(*ms));
+	if (!ctlr)
+		return -ENOMEM;
+	platform_set_drvdata(pdev, ctlr);
+
+	ms = spi_controller_get_devdata(ctlr);
+
+	ms->ctlr = ctlr;
+	ms->caps = dev_id->data;
+
+	ms->ecc = of_mtk_ecc_get(np);
+	if (IS_ERR(ms->ecc))
+		return PTR_ERR(ms->ecc);
+	else if (!ms->ecc)
+		return -ENODEV;
+
+	ms->nfi_base = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(ms->nfi_base)) {
+		ret = PTR_ERR(ms->nfi_base);
+		goto release_ecc;
+	}
+
+	ms->dev = &pdev->dev;
+
+	ms->nfi_clk = devm_clk_get(&pdev->dev, "nfi_clk");
+	if (IS_ERR(ms->nfi_clk)) {
+		ret = PTR_ERR(ms->nfi_clk);
+		dev_err(&pdev->dev, "unable to get nfi_clk, err = %d\n", ret);
+		goto release_ecc;
+	}
+
+	ms->pad_clk = devm_clk_get(&pdev->dev, "pad_clk");
+	if (IS_ERR(ms->pad_clk)) {
+		ret = PTR_ERR(ms->pad_clk);
+		dev_err(&pdev->dev, "unable to get pad_clk, err = %d\n", ret);
+		goto release_ecc;
+	}
+
+	ret = mtk_snand_enable_clk(ms);
+	if (ret)
+		goto release_ecc;
+
+	init_completion(&ms->op_done);
+
+	ms->irq = platform_get_irq(pdev, 0);
+	if (ms->irq < 0) {
+		ret = ms->irq;
+		goto disable_clk;
+	}
+	ret = devm_request_irq(ms->dev, ms->irq, mtk_snand_irq, 0x0,
+			       "mtk-snand", ms);
+	if (ret) {
+		dev_err(ms->dev, "failed to request snfi irq\n");
+		goto disable_clk;
+	}
+
+	ret = dma_set_mask(ms->dev, DMA_BIT_MASK(32));
+	if (ret) {
+		dev_err(ms->dev, "failed to set dma mask\n");
+		goto disable_clk;
+	}
+
+	// switch to SNFI mode
+	nfi_write32(ms, SNF_CFG, SPI_MODE);
+
+	// setup an initial page format for ops matching page_cache_op template
+	// before ECC is called.
+	ret = mtk_snand_setup_pagefmt(ms, ms->caps->sector_size,
+				      ms->caps->spare_sizes[0]);
+	if (ret) {
+		dev_err(ms->dev, "failed to set initial page format\n");
+		goto disable_clk;
+	}
+
+	// setup ECC engine
+	ms->ecc_eng.dev = &pdev->dev;
+	ms->ecc_eng.integration = NAND_ECC_ENGINE_INTEGRATION_PIPELINED;
+	ms->ecc_eng.ops = &mtk_snfi_ecc_engine_ops;
+	ms->ecc_eng.priv = ms;
+
+	ret = nand_ecc_register_on_host_hw_engine(&ms->ecc_eng);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to register ecc engine.\n");
+		goto disable_clk;
+	}
+
+	ctlr->num_chipselect = 1;
+	ctlr->mem_ops = &mtk_snand_mem_ops;
+	ctlr->mem_caps = &mtk_snand_mem_caps;
+	ctlr->bits_per_word_mask = SPI_BPW_MASK(8);
+	ctlr->mode_bits = SPI_RX_DUAL | SPI_RX_QUAD | SPI_TX_DUAL | SPI_TX_QUAD;
+	ctlr->dev.of_node = pdev->dev.of_node;
+	ret = spi_register_controller(ctlr);
+	if (ret) {
+		dev_err(&pdev->dev, "spi_register_controller failed.\n");
+		goto disable_clk;
+	}
+
+	return 0;
+disable_clk:
+	mtk_snand_disable_clk(ms);
+release_ecc:
+	mtk_ecc_release(ms->ecc);
+	return ret;
+}
+
+static int mtk_snand_remove(struct platform_device *pdev)
+{
+	struct spi_controller *ctlr = platform_get_drvdata(pdev);
+	struct mtk_snand *ms = spi_controller_get_devdata(ctlr);
+
+	spi_unregister_controller(ctlr);
+	mtk_snand_disable_clk(ms);
+	mtk_ecc_release(ms->ecc);
+	kfree(ms->buf);
+	return 0;
+}
+
+static struct platform_driver mtk_snand_driver = {
+	.probe = mtk_snand_probe,
+	.remove = mtk_snand_remove,
+	.driver = {
+		.name = "mtk-snand",
+		.of_match_table = mtk_snand_ids,
+	},
+};
+
+module_platform_driver(mtk_snand_driver);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Chuanhong Guo <gch981213@gmail.com>");
+MODULE_DESCRIPTION("MeidaTek SPI-NAND Flash Controller Driver");