forked from torvalds/linux
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathste_dma40.c
3722 lines (3103 loc) · 96.2 KB
/
ste_dma40.c
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
// SPDX-License-Identifier: GPL-2.0-only
/*
* Copyright (C) Ericsson AB 2007-2008
* Copyright (C) ST-Ericsson SA 2008-2010
* Author: Per Forlin <[email protected]> for ST-Ericsson
* Author: Jonas Aaberg <[email protected]> for ST-Ericsson
*/
#include <linux/dma-mapping.h>
#include <linux/kernel.h>
#include <linux/slab.h>
#include <linux/export.h>
#include <linux/dmaengine.h>
#include <linux/platform_device.h>
#include <linux/clk.h>
#include <linux/delay.h>
#include <linux/log2.h>
#include <linux/pm.h>
#include <linux/pm_runtime.h>
#include <linux/err.h>
#include <linux/of.h>
#include <linux/of_dma.h>
#include <linux/amba/bus.h>
#include <linux/regulator/consumer.h>
#include <linux/platform_data/dma-ste-dma40.h>
#include "dmaengine.h"
#include "ste_dma40_ll.h"
#define D40_NAME "dma40"
#define D40_PHY_CHAN -1
/* For masking out/in 2 bit channel positions */
#define D40_CHAN_POS(chan) (2 * (chan / 2))
#define D40_CHAN_POS_MASK(chan) (0x3 << D40_CHAN_POS(chan))
/* Maximum iterations taken before giving up suspending a channel */
#define D40_SUSPEND_MAX_IT 500
/* Milliseconds */
#define DMA40_AUTOSUSPEND_DELAY 100
/* Hardware requirement on LCLA alignment */
#define LCLA_ALIGNMENT 0x40000
/* Max number of links per event group */
#define D40_LCLA_LINK_PER_EVENT_GRP 128
#define D40_LCLA_END D40_LCLA_LINK_PER_EVENT_GRP
/* Max number of logical channels per physical channel */
#define D40_MAX_LOG_CHAN_PER_PHY 32
/* Attempts before giving up to trying to get pages that are aligned */
#define MAX_LCLA_ALLOC_ATTEMPTS 256
/* Bit markings for allocation map */
#define D40_ALLOC_FREE BIT(31)
#define D40_ALLOC_PHY BIT(30)
#define D40_ALLOC_LOG_FREE 0
#define D40_MEMCPY_MAX_CHANS 8
/* Reserved event lines for memcpy only. */
#define DB8500_DMA_MEMCPY_EV_0 51
#define DB8500_DMA_MEMCPY_EV_1 56
#define DB8500_DMA_MEMCPY_EV_2 57
#define DB8500_DMA_MEMCPY_EV_3 58
#define DB8500_DMA_MEMCPY_EV_4 59
#define DB8500_DMA_MEMCPY_EV_5 60
static int dma40_memcpy_channels[] = {
DB8500_DMA_MEMCPY_EV_0,
DB8500_DMA_MEMCPY_EV_1,
DB8500_DMA_MEMCPY_EV_2,
DB8500_DMA_MEMCPY_EV_3,
DB8500_DMA_MEMCPY_EV_4,
DB8500_DMA_MEMCPY_EV_5,
};
/* Default configuration for physcial memcpy */
static const struct stedma40_chan_cfg dma40_memcpy_conf_phy = {
.mode = STEDMA40_MODE_PHYSICAL,
.dir = DMA_MEM_TO_MEM,
.src_info.data_width = DMA_SLAVE_BUSWIDTH_1_BYTE,
.src_info.psize = STEDMA40_PSIZE_PHY_1,
.src_info.flow_ctrl = STEDMA40_NO_FLOW_CTRL,
.dst_info.data_width = DMA_SLAVE_BUSWIDTH_1_BYTE,
.dst_info.psize = STEDMA40_PSIZE_PHY_1,
.dst_info.flow_ctrl = STEDMA40_NO_FLOW_CTRL,
};
/* Default configuration for logical memcpy */
static const struct stedma40_chan_cfg dma40_memcpy_conf_log = {
.mode = STEDMA40_MODE_LOGICAL,
.dir = DMA_MEM_TO_MEM,
.src_info.data_width = DMA_SLAVE_BUSWIDTH_1_BYTE,
.src_info.psize = STEDMA40_PSIZE_LOG_1,
.src_info.flow_ctrl = STEDMA40_NO_FLOW_CTRL,
.dst_info.data_width = DMA_SLAVE_BUSWIDTH_1_BYTE,
.dst_info.psize = STEDMA40_PSIZE_LOG_1,
.dst_info.flow_ctrl = STEDMA40_NO_FLOW_CTRL,
};
/**
* enum 40_command - The different commands and/or statuses.
*
* @D40_DMA_STOP: DMA channel command STOP or status STOPPED,
* @D40_DMA_RUN: The DMA channel is RUNNING of the command RUN.
* @D40_DMA_SUSPEND_REQ: Request the DMA to SUSPEND as soon as possible.
* @D40_DMA_SUSPENDED: The DMA channel is SUSPENDED.
*/
enum d40_command {
D40_DMA_STOP = 0,
D40_DMA_RUN = 1,
D40_DMA_SUSPEND_REQ = 2,
D40_DMA_SUSPENDED = 3
};
/*
* enum d40_events - The different Event Enables for the event lines.
*
* @D40_DEACTIVATE_EVENTLINE: De-activate Event line, stopping the logical chan.
* @D40_ACTIVATE_EVENTLINE: Activate the Event line, to start a logical chan.
* @D40_SUSPEND_REQ_EVENTLINE: Requesting for suspending a event line.
* @D40_ROUND_EVENTLINE: Status check for event line.
*/
enum d40_events {
D40_DEACTIVATE_EVENTLINE = 0,
D40_ACTIVATE_EVENTLINE = 1,
D40_SUSPEND_REQ_EVENTLINE = 2,
D40_ROUND_EVENTLINE = 3
};
/*
* These are the registers that has to be saved and later restored
* when the DMA hw is powered off.
* TODO: Add save/restore of D40_DREG_GCC on dma40 v3 or later, if that works.
*/
static __maybe_unused u32 d40_backup_regs[] = {
D40_DREG_LCPA,
D40_DREG_LCLA,
D40_DREG_PRMSE,
D40_DREG_PRMSO,
D40_DREG_PRMOE,
D40_DREG_PRMOO,
};
#define BACKUP_REGS_SZ ARRAY_SIZE(d40_backup_regs)
/*
* since 9540 and 8540 has the same HW revision
* use v4a for 9540 or ealier
* use v4b for 8540 or later
* HW revision:
* DB8500ed has revision 0
* DB8500v1 has revision 2
* DB8500v2 has revision 3
* AP9540v1 has revision 4
* DB8540v1 has revision 4
* TODO: Check if all these registers have to be saved/restored on dma40 v4a
*/
static u32 d40_backup_regs_v4a[] = {
D40_DREG_PSEG1,
D40_DREG_PSEG2,
D40_DREG_PSEG3,
D40_DREG_PSEG4,
D40_DREG_PCEG1,
D40_DREG_PCEG2,
D40_DREG_PCEG3,
D40_DREG_PCEG4,
D40_DREG_RSEG1,
D40_DREG_RSEG2,
D40_DREG_RSEG3,
D40_DREG_RSEG4,
D40_DREG_RCEG1,
D40_DREG_RCEG2,
D40_DREG_RCEG3,
D40_DREG_RCEG4,
};
#define BACKUP_REGS_SZ_V4A ARRAY_SIZE(d40_backup_regs_v4a)
static u32 d40_backup_regs_v4b[] = {
D40_DREG_CPSEG1,
D40_DREG_CPSEG2,
D40_DREG_CPSEG3,
D40_DREG_CPSEG4,
D40_DREG_CPSEG5,
D40_DREG_CPCEG1,
D40_DREG_CPCEG2,
D40_DREG_CPCEG3,
D40_DREG_CPCEG4,
D40_DREG_CPCEG5,
D40_DREG_CRSEG1,
D40_DREG_CRSEG2,
D40_DREG_CRSEG3,
D40_DREG_CRSEG4,
D40_DREG_CRSEG5,
D40_DREG_CRCEG1,
D40_DREG_CRCEG2,
D40_DREG_CRCEG3,
D40_DREG_CRCEG4,
D40_DREG_CRCEG5,
};
#define BACKUP_REGS_SZ_V4B ARRAY_SIZE(d40_backup_regs_v4b)
static __maybe_unused u32 d40_backup_regs_chan[] = {
D40_CHAN_REG_SSCFG,
D40_CHAN_REG_SSELT,
D40_CHAN_REG_SSPTR,
D40_CHAN_REG_SSLNK,
D40_CHAN_REG_SDCFG,
D40_CHAN_REG_SDELT,
D40_CHAN_REG_SDPTR,
D40_CHAN_REG_SDLNK,
};
#define BACKUP_REGS_SZ_MAX ((BACKUP_REGS_SZ_V4A > BACKUP_REGS_SZ_V4B) ? \
BACKUP_REGS_SZ_V4A : BACKUP_REGS_SZ_V4B)
/**
* struct d40_interrupt_lookup - lookup table for interrupt handler
*
* @src: Interrupt mask register.
* @clr: Interrupt clear register.
* @is_error: true if this is an error interrupt.
* @offset: start delta in the lookup_log_chans in d40_base. If equals to
* D40_PHY_CHAN, the lookup_phy_chans shall be used instead.
*/
struct d40_interrupt_lookup {
u32 src;
u32 clr;
bool is_error;
int offset;
};
static struct d40_interrupt_lookup il_v4a[] = {
{D40_DREG_LCTIS0, D40_DREG_LCICR0, false, 0},
{D40_DREG_LCTIS1, D40_DREG_LCICR1, false, 32},
{D40_DREG_LCTIS2, D40_DREG_LCICR2, false, 64},
{D40_DREG_LCTIS3, D40_DREG_LCICR3, false, 96},
{D40_DREG_LCEIS0, D40_DREG_LCICR0, true, 0},
{D40_DREG_LCEIS1, D40_DREG_LCICR1, true, 32},
{D40_DREG_LCEIS2, D40_DREG_LCICR2, true, 64},
{D40_DREG_LCEIS3, D40_DREG_LCICR3, true, 96},
{D40_DREG_PCTIS, D40_DREG_PCICR, false, D40_PHY_CHAN},
{D40_DREG_PCEIS, D40_DREG_PCICR, true, D40_PHY_CHAN},
};
static struct d40_interrupt_lookup il_v4b[] = {
{D40_DREG_CLCTIS1, D40_DREG_CLCICR1, false, 0},
{D40_DREG_CLCTIS2, D40_DREG_CLCICR2, false, 32},
{D40_DREG_CLCTIS3, D40_DREG_CLCICR3, false, 64},
{D40_DREG_CLCTIS4, D40_DREG_CLCICR4, false, 96},
{D40_DREG_CLCTIS5, D40_DREG_CLCICR5, false, 128},
{D40_DREG_CLCEIS1, D40_DREG_CLCICR1, true, 0},
{D40_DREG_CLCEIS2, D40_DREG_CLCICR2, true, 32},
{D40_DREG_CLCEIS3, D40_DREG_CLCICR3, true, 64},
{D40_DREG_CLCEIS4, D40_DREG_CLCICR4, true, 96},
{D40_DREG_CLCEIS5, D40_DREG_CLCICR5, true, 128},
{D40_DREG_CPCTIS, D40_DREG_CPCICR, false, D40_PHY_CHAN},
{D40_DREG_CPCEIS, D40_DREG_CPCICR, true, D40_PHY_CHAN},
};
/**
* struct d40_reg_val - simple lookup struct
*
* @reg: The register.
* @val: The value that belongs to the register in reg.
*/
struct d40_reg_val {
unsigned int reg;
unsigned int val;
};
static __initdata struct d40_reg_val dma_init_reg_v4a[] = {
/* Clock every part of the DMA block from start */
{ .reg = D40_DREG_GCC, .val = D40_DREG_GCC_ENABLE_ALL},
/* Interrupts on all logical channels */
{ .reg = D40_DREG_LCMIS0, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_LCMIS1, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_LCMIS2, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_LCMIS3, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_LCICR0, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_LCICR1, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_LCICR2, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_LCICR3, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_LCTIS0, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_LCTIS1, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_LCTIS2, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_LCTIS3, .val = 0xFFFFFFFF}
};
static __initdata struct d40_reg_val dma_init_reg_v4b[] = {
/* Clock every part of the DMA block from start */
{ .reg = D40_DREG_GCC, .val = D40_DREG_GCC_ENABLE_ALL},
/* Interrupts on all logical channels */
{ .reg = D40_DREG_CLCMIS1, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_CLCMIS2, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_CLCMIS3, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_CLCMIS4, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_CLCMIS5, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_CLCICR1, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_CLCICR2, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_CLCICR3, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_CLCICR4, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_CLCICR5, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_CLCTIS1, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_CLCTIS2, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_CLCTIS3, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_CLCTIS4, .val = 0xFFFFFFFF},
{ .reg = D40_DREG_CLCTIS5, .val = 0xFFFFFFFF}
};
/**
* struct d40_lli_pool - Structure for keeping LLIs in memory
*
* @base: Pointer to memory area when the pre_alloc_lli's are not large
* enough, IE bigger than the most common case, 1 dst and 1 src. NULL if
* pre_alloc_lli is used.
* @dma_addr: DMA address, if mapped
* @size: The size in bytes of the memory at base or the size of pre_alloc_lli.
* @pre_alloc_lli: Pre allocated area for the most common case of transfers,
* one buffer to one buffer.
*/
struct d40_lli_pool {
void *base;
int size;
dma_addr_t dma_addr;
/* Space for dst and src, plus an extra for padding */
u8 pre_alloc_lli[3 * sizeof(struct d40_phy_lli)];
};
/**
* struct d40_desc - A descriptor is one DMA job.
*
* @lli_phy: LLI settings for physical channel. Both src and dst=
* points into the lli_pool, to base if lli_len > 1 or to pre_alloc_lli if
* lli_len equals one.
* @lli_log: Same as above but for logical channels.
* @lli_pool: The pool with two entries pre-allocated.
* @lli_len: Number of llis of current descriptor.
* @lli_current: Number of transferred llis.
* @lcla_alloc: Number of LCLA entries allocated.
* @txd: DMA engine struct. Used for among other things for communication
* during a transfer.
* @node: List entry.
* @is_in_client_list: true if the client owns this descriptor.
* @cyclic: true if this is a cyclic job
*
* This descriptor is used for both logical and physical transfers.
*/
struct d40_desc {
/* LLI physical */
struct d40_phy_lli_bidir lli_phy;
/* LLI logical */
struct d40_log_lli_bidir lli_log;
struct d40_lli_pool lli_pool;
int lli_len;
int lli_current;
int lcla_alloc;
struct dma_async_tx_descriptor txd;
struct list_head node;
bool is_in_client_list;
bool cyclic;
};
/**
* struct d40_lcla_pool - LCLA pool settings and data.
*
* @base: The virtual address of LCLA. 18 bit aligned.
* @dma_addr: DMA address, if mapped
* @base_unaligned: The orignal kmalloc pointer, if kmalloc is used.
* This pointer is only there for clean-up on error.
* @pages: The number of pages needed for all physical channels.
* Only used later for clean-up on error
* @lock: Lock to protect the content in this struct.
* @alloc_map: big map over which LCLA entry is own by which job.
*/
struct d40_lcla_pool {
void *base;
dma_addr_t dma_addr;
void *base_unaligned;
int pages;
spinlock_t lock;
struct d40_desc **alloc_map;
};
/**
* struct d40_phy_res - struct for handling eventlines mapped to physical
* channels.
*
* @lock: A lock protection this entity.
* @reserved: True if used by secure world or otherwise.
* @num: The physical channel number of this entity.
* @allocated_src: Bit mapped to show which src event line's are mapped to
* this physical channel. Can also be free or physically allocated.
* @allocated_dst: Same as for src but is dst.
* allocated_dst and allocated_src uses the D40_ALLOC* defines as well as
* event line number.
* @use_soft_lli: To mark if the linked lists of channel are managed by SW.
*/
struct d40_phy_res {
spinlock_t lock;
bool reserved;
int num;
u32 allocated_src;
u32 allocated_dst;
bool use_soft_lli;
};
struct d40_base;
/**
* struct d40_chan - Struct that describes a channel.
*
* @lock: A spinlock to protect this struct.
* @log_num: The logical number, if any of this channel.
* @pending_tx: The number of pending transfers. Used between interrupt handler
* and tasklet.
* @busy: Set to true when transfer is ongoing on this channel.
* @phy_chan: Pointer to physical channel which this instance runs on. If this
* point is NULL, then the channel is not allocated.
* @chan: DMA engine handle.
* @tasklet: Tasklet that gets scheduled from interrupt context to complete a
* transfer and call client callback.
* @client: Cliented owned descriptor list.
* @pending_queue: Submitted jobs, to be issued by issue_pending()
* @active: Active descriptor.
* @done: Completed jobs
* @queue: Queued jobs.
* @prepare_queue: Prepared jobs.
* @dma_cfg: The client configuration of this dma channel.
* @slave_config: DMA slave configuration.
* @configured: whether the dma_cfg configuration is valid
* @base: Pointer to the device instance struct.
* @src_def_cfg: Default cfg register setting for src.
* @dst_def_cfg: Default cfg register setting for dst.
* @log_def: Default logical channel settings.
* @lcpa: Pointer to dst and src lcpa settings.
* @runtime_addr: runtime configured address.
* @runtime_direction: runtime configured direction.
*
* This struct can either "be" a logical or a physical channel.
*/
struct d40_chan {
spinlock_t lock;
int log_num;
int pending_tx;
bool busy;
struct d40_phy_res *phy_chan;
struct dma_chan chan;
struct tasklet_struct tasklet;
struct list_head client;
struct list_head pending_queue;
struct list_head active;
struct list_head done;
struct list_head queue;
struct list_head prepare_queue;
struct stedma40_chan_cfg dma_cfg;
struct dma_slave_config slave_config;
bool configured;
struct d40_base *base;
/* Default register configurations */
u32 src_def_cfg;
u32 dst_def_cfg;
struct d40_def_lcsp log_def;
struct d40_log_lli_full *lcpa;
/* Runtime reconfiguration */
dma_addr_t runtime_addr;
enum dma_transfer_direction runtime_direction;
};
/**
* struct d40_gen_dmac - generic values to represent u8500/u8540 DMA
* controller
*
* @backup: the pointer to the registers address array for backup
* @backup_size: the size of the registers address array for backup
* @realtime_en: the realtime enable register
* @realtime_clear: the realtime clear register
* @high_prio_en: the high priority enable register
* @high_prio_clear: the high priority clear register
* @interrupt_en: the interrupt enable register
* @interrupt_clear: the interrupt clear register
* @il: the pointer to struct d40_interrupt_lookup
* @il_size: the size of d40_interrupt_lookup array
* @init_reg: the pointer to the struct d40_reg_val
* @init_reg_size: the size of d40_reg_val array
*/
struct d40_gen_dmac {
u32 *backup;
u32 backup_size;
u32 realtime_en;
u32 realtime_clear;
u32 high_prio_en;
u32 high_prio_clear;
u32 interrupt_en;
u32 interrupt_clear;
struct d40_interrupt_lookup *il;
u32 il_size;
struct d40_reg_val *init_reg;
u32 init_reg_size;
};
/**
* struct d40_base - The big global struct, one for each probe'd instance.
*
* @interrupt_lock: Lock used to make sure one interrupt is handle a time.
* @execmd_lock: Lock for execute command usage since several channels share
* the same physical register.
* @dev: The device structure.
* @virtbase: The virtual base address of the DMA's register.
* @rev: silicon revision detected.
* @clk: Pointer to the DMA clock structure.
* @phy_start: Physical memory start of the DMA registers.
* @phy_size: Size of the DMA register map.
* @irq: The IRQ number.
* @num_memcpy_chans: The number of channels used for memcpy (mem-to-mem
* transfers).
* @num_phy_chans: The number of physical channels. Read from HW. This
* is the number of available channels for this driver, not counting "Secure
* mode" allocated physical channels.
* @num_log_chans: The number of logical channels. Calculated from
* num_phy_chans.
* @dma_both: dma_device channels that can do both memcpy and slave transfers.
* @dma_slave: dma_device channels that can do only do slave transfers.
* @dma_memcpy: dma_device channels that can do only do memcpy transfers.
* @phy_chans: Room for all possible physical channels in system.
* @log_chans: Room for all possible logical channels in system.
* @lookup_log_chans: Used to map interrupt number to logical channel. Points
* to log_chans entries.
* @lookup_phy_chans: Used to map interrupt number to physical channel. Points
* to phy_chans entries.
* @plat_data: Pointer to provided platform_data which is the driver
* configuration.
* @lcpa_regulator: Pointer to hold the regulator for the esram bank for lcla.
* @phy_res: Vector containing all physical channels.
* @lcla_pool: lcla pool settings and data.
* @lcpa_base: The virtual mapped address of LCPA.
* @phy_lcpa: The physical address of the LCPA.
* @lcpa_size: The size of the LCPA area.
* @desc_slab: cache for descriptors.
* @reg_val_backup: Here the values of some hardware registers are stored
* before the DMA is powered off. They are restored when the power is back on.
* @reg_val_backup_v4: Backup of registers that only exits on dma40 v3 and
* later
* @reg_val_backup_chan: Backup data for standard channel parameter registers.
* @regs_interrupt: Scratch space for registers during interrupt.
* @gcc_pwr_off_mask: Mask to maintain the channels that can be turned off.
* @gen_dmac: the struct for generic registers values to represent u8500/8540
* DMA controller
*/
struct d40_base {
spinlock_t interrupt_lock;
spinlock_t execmd_lock;
struct device *dev;
void __iomem *virtbase;
u8 rev:4;
struct clk *clk;
phys_addr_t phy_start;
resource_size_t phy_size;
int irq;
int num_memcpy_chans;
int num_phy_chans;
int num_log_chans;
struct dma_device dma_both;
struct dma_device dma_slave;
struct dma_device dma_memcpy;
struct d40_chan *phy_chans;
struct d40_chan *log_chans;
struct d40_chan **lookup_log_chans;
struct d40_chan **lookup_phy_chans;
struct stedma40_platform_data *plat_data;
struct regulator *lcpa_regulator;
/* Physical half channels */
struct d40_phy_res *phy_res;
struct d40_lcla_pool lcla_pool;
void *lcpa_base;
dma_addr_t phy_lcpa;
resource_size_t lcpa_size;
struct kmem_cache *desc_slab;
u32 reg_val_backup[BACKUP_REGS_SZ];
u32 reg_val_backup_v4[BACKUP_REGS_SZ_MAX];
u32 *reg_val_backup_chan;
u32 *regs_interrupt;
u16 gcc_pwr_off_mask;
struct d40_gen_dmac gen_dmac;
};
static struct device *chan2dev(struct d40_chan *d40c)
{
return &d40c->chan.dev->device;
}
static bool chan_is_physical(struct d40_chan *chan)
{
return chan->log_num == D40_PHY_CHAN;
}
static bool chan_is_logical(struct d40_chan *chan)
{
return !chan_is_physical(chan);
}
static void __iomem *chan_base(struct d40_chan *chan)
{
return chan->base->virtbase + D40_DREG_PCBASE +
chan->phy_chan->num * D40_DREG_PCDELTA;
}
#define d40_err(dev, format, arg...) \
dev_err(dev, "[%s] " format, __func__, ## arg)
#define chan_err(d40c, format, arg...) \
d40_err(chan2dev(d40c), format, ## arg)
static int d40_set_runtime_config_write(struct dma_chan *chan,
struct dma_slave_config *config,
enum dma_transfer_direction direction);
static int d40_pool_lli_alloc(struct d40_chan *d40c, struct d40_desc *d40d,
int lli_len)
{
bool is_log = chan_is_logical(d40c);
u32 align;
void *base;
if (is_log)
align = sizeof(struct d40_log_lli);
else
align = sizeof(struct d40_phy_lli);
if (lli_len == 1) {
base = d40d->lli_pool.pre_alloc_lli;
d40d->lli_pool.size = sizeof(d40d->lli_pool.pre_alloc_lli);
d40d->lli_pool.base = NULL;
} else {
d40d->lli_pool.size = lli_len * 2 * align;
base = kmalloc(d40d->lli_pool.size + align, GFP_NOWAIT);
d40d->lli_pool.base = base;
if (d40d->lli_pool.base == NULL)
return -ENOMEM;
}
if (is_log) {
d40d->lli_log.src = PTR_ALIGN(base, align);
d40d->lli_log.dst = d40d->lli_log.src + lli_len;
d40d->lli_pool.dma_addr = 0;
} else {
d40d->lli_phy.src = PTR_ALIGN(base, align);
d40d->lli_phy.dst = d40d->lli_phy.src + lli_len;
d40d->lli_pool.dma_addr = dma_map_single(d40c->base->dev,
d40d->lli_phy.src,
d40d->lli_pool.size,
DMA_TO_DEVICE);
if (dma_mapping_error(d40c->base->dev,
d40d->lli_pool.dma_addr)) {
kfree(d40d->lli_pool.base);
d40d->lli_pool.base = NULL;
d40d->lli_pool.dma_addr = 0;
return -ENOMEM;
}
}
return 0;
}
static void d40_pool_lli_free(struct d40_chan *d40c, struct d40_desc *d40d)
{
if (d40d->lli_pool.dma_addr)
dma_unmap_single(d40c->base->dev, d40d->lli_pool.dma_addr,
d40d->lli_pool.size, DMA_TO_DEVICE);
kfree(d40d->lli_pool.base);
d40d->lli_pool.base = NULL;
d40d->lli_pool.size = 0;
d40d->lli_log.src = NULL;
d40d->lli_log.dst = NULL;
d40d->lli_phy.src = NULL;
d40d->lli_phy.dst = NULL;
}
static int d40_lcla_alloc_one(struct d40_chan *d40c,
struct d40_desc *d40d)
{
unsigned long flags;
int i;
int ret = -EINVAL;
spin_lock_irqsave(&d40c->base->lcla_pool.lock, flags);
/*
* Allocate both src and dst at the same time, therefore the half
* start on 1 since 0 can't be used since zero is used as end marker.
*/
for (i = 1 ; i < D40_LCLA_LINK_PER_EVENT_GRP / 2; i++) {
int idx = d40c->phy_chan->num * D40_LCLA_LINK_PER_EVENT_GRP + i;
if (!d40c->base->lcla_pool.alloc_map[idx]) {
d40c->base->lcla_pool.alloc_map[idx] = d40d;
d40d->lcla_alloc++;
ret = i;
break;
}
}
spin_unlock_irqrestore(&d40c->base->lcla_pool.lock, flags);
return ret;
}
static int d40_lcla_free_all(struct d40_chan *d40c,
struct d40_desc *d40d)
{
unsigned long flags;
int i;
int ret = -EINVAL;
if (chan_is_physical(d40c))
return 0;
spin_lock_irqsave(&d40c->base->lcla_pool.lock, flags);
for (i = 1 ; i < D40_LCLA_LINK_PER_EVENT_GRP / 2; i++) {
int idx = d40c->phy_chan->num * D40_LCLA_LINK_PER_EVENT_GRP + i;
if (d40c->base->lcla_pool.alloc_map[idx] == d40d) {
d40c->base->lcla_pool.alloc_map[idx] = NULL;
d40d->lcla_alloc--;
if (d40d->lcla_alloc == 0) {
ret = 0;
break;
}
}
}
spin_unlock_irqrestore(&d40c->base->lcla_pool.lock, flags);
return ret;
}
static void d40_desc_remove(struct d40_desc *d40d)
{
list_del(&d40d->node);
}
static struct d40_desc *d40_desc_get(struct d40_chan *d40c)
{
struct d40_desc *desc = NULL;
if (!list_empty(&d40c->client)) {
struct d40_desc *d;
struct d40_desc *_d;
list_for_each_entry_safe(d, _d, &d40c->client, node) {
if (async_tx_test_ack(&d->txd)) {
d40_desc_remove(d);
desc = d;
memset(desc, 0, sizeof(*desc));
break;
}
}
}
if (!desc)
desc = kmem_cache_zalloc(d40c->base->desc_slab, GFP_NOWAIT);
if (desc)
INIT_LIST_HEAD(&desc->node);
return desc;
}
static void d40_desc_free(struct d40_chan *d40c, struct d40_desc *d40d)
{
d40_pool_lli_free(d40c, d40d);
d40_lcla_free_all(d40c, d40d);
kmem_cache_free(d40c->base->desc_slab, d40d);
}
static void d40_desc_submit(struct d40_chan *d40c, struct d40_desc *desc)
{
list_add_tail(&desc->node, &d40c->active);
}
static void d40_phy_lli_load(struct d40_chan *chan, struct d40_desc *desc)
{
struct d40_phy_lli *lli_dst = desc->lli_phy.dst;
struct d40_phy_lli *lli_src = desc->lli_phy.src;
void __iomem *base = chan_base(chan);
writel(lli_src->reg_cfg, base + D40_CHAN_REG_SSCFG);
writel(lli_src->reg_elt, base + D40_CHAN_REG_SSELT);
writel(lli_src->reg_ptr, base + D40_CHAN_REG_SSPTR);
writel(lli_src->reg_lnk, base + D40_CHAN_REG_SSLNK);
writel(lli_dst->reg_cfg, base + D40_CHAN_REG_SDCFG);
writel(lli_dst->reg_elt, base + D40_CHAN_REG_SDELT);
writel(lli_dst->reg_ptr, base + D40_CHAN_REG_SDPTR);
writel(lli_dst->reg_lnk, base + D40_CHAN_REG_SDLNK);
}
static void d40_desc_done(struct d40_chan *d40c, struct d40_desc *desc)
{
list_add_tail(&desc->node, &d40c->done);
}
static void d40_log_lli_to_lcxa(struct d40_chan *chan, struct d40_desc *desc)
{
struct d40_lcla_pool *pool = &chan->base->lcla_pool;
struct d40_log_lli_bidir *lli = &desc->lli_log;
int lli_current = desc->lli_current;
int lli_len = desc->lli_len;
bool cyclic = desc->cyclic;
int curr_lcla = -EINVAL;
int first_lcla = 0;
bool use_esram_lcla = chan->base->plat_data->use_esram_lcla;
bool linkback;
/*
* We may have partially running cyclic transfers, in case we did't get
* enough LCLA entries.
*/
linkback = cyclic && lli_current == 0;
/*
* For linkback, we need one LCLA even with only one link, because we
* can't link back to the one in LCPA space
*/
if (linkback || (lli_len - lli_current > 1)) {
/*
* If the channel is expected to use only soft_lli don't
* allocate a lcla. This is to avoid a HW issue that exists
* in some controller during a peripheral to memory transfer
* that uses linked lists.
*/
if (!(chan->phy_chan->use_soft_lli &&
chan->dma_cfg.dir == DMA_DEV_TO_MEM))
curr_lcla = d40_lcla_alloc_one(chan, desc);
first_lcla = curr_lcla;
}
/*
* For linkback, we normally load the LCPA in the loop since we need to
* link it to the second LCLA and not the first. However, if we
* couldn't even get a first LCLA, then we have to run in LCPA and
* reload manually.
*/
if (!linkback || curr_lcla == -EINVAL) {
unsigned int flags = 0;
if (curr_lcla == -EINVAL)
flags |= LLI_TERM_INT;
d40_log_lli_lcpa_write(chan->lcpa,
&lli->dst[lli_current],
&lli->src[lli_current],
curr_lcla,
flags);
lli_current++;
}
if (curr_lcla < 0)
goto set_current;
for (; lli_current < lli_len; lli_current++) {
unsigned int lcla_offset = chan->phy_chan->num * 1024 +
8 * curr_lcla * 2;
struct d40_log_lli *lcla = pool->base + lcla_offset;
unsigned int flags = 0;
int next_lcla;
if (lli_current + 1 < lli_len)
next_lcla = d40_lcla_alloc_one(chan, desc);
else
next_lcla = linkback ? first_lcla : -EINVAL;
if (cyclic || next_lcla == -EINVAL)
flags |= LLI_TERM_INT;
if (linkback && curr_lcla == first_lcla) {
/* First link goes in both LCPA and LCLA */
d40_log_lli_lcpa_write(chan->lcpa,
&lli->dst[lli_current],
&lli->src[lli_current],
next_lcla, flags);
}
/*
* One unused LCLA in the cyclic case if the very first
* next_lcla fails...
*/
d40_log_lli_lcla_write(lcla,
&lli->dst[lli_current],
&lli->src[lli_current],
next_lcla, flags);
/*
* Cache maintenance is not needed if lcla is
* mapped in esram
*/
if (!use_esram_lcla) {
dma_sync_single_range_for_device(chan->base->dev,
pool->dma_addr, lcla_offset,
2 * sizeof(struct d40_log_lli),
DMA_TO_DEVICE);
}
curr_lcla = next_lcla;
if (curr_lcla == -EINVAL || curr_lcla == first_lcla) {
lli_current++;
break;
}
}
set_current:
desc->lli_current = lli_current;
}
static void d40_desc_load(struct d40_chan *d40c, struct d40_desc *d40d)
{
if (chan_is_physical(d40c)) {
d40_phy_lli_load(d40c, d40d);
d40d->lli_current = d40d->lli_len;
} else
d40_log_lli_to_lcxa(d40c, d40d);
}
static struct d40_desc *d40_first_active_get(struct d40_chan *d40c)
{
return list_first_entry_or_null(&d40c->active, struct d40_desc, node);
}
/* remove desc from current queue and add it to the pending_queue */
static void d40_desc_queue(struct d40_chan *d40c, struct d40_desc *desc)
{
d40_desc_remove(desc);
desc->is_in_client_list = false;
list_add_tail(&desc->node, &d40c->pending_queue);
}
static struct d40_desc *d40_first_pending(struct d40_chan *d40c)
{
return list_first_entry_or_null(&d40c->pending_queue, struct d40_desc,
node);
}
static struct d40_desc *d40_first_queued(struct d40_chan *d40c)
{
return list_first_entry_or_null(&d40c->queue, struct d40_desc, node);
}
static struct d40_desc *d40_first_done(struct d40_chan *d40c)
{
return list_first_entry_or_null(&d40c->done, struct d40_desc, node);
}
static int d40_psize_2_burst_size(bool is_log, int psize)
{
if (is_log) {
if (psize == STEDMA40_PSIZE_LOG_1)
return 1;
} else {
if (psize == STEDMA40_PSIZE_PHY_1)
return 1;
}
return 2 << psize;
}
/*
* The dma only supports transmitting packages up to
* STEDMA40_MAX_SEG_SIZE * data_width, where data_width is stored in Bytes.
*
* Calculate the total number of dma elements required to send the entire sg list.
*/
static int d40_size_2_dmalen(int size, u32 data_width1, u32 data_width2)
{
int dmalen;
u32 max_w = max(data_width1, data_width2);