summaryrefslogtreecommitdiffstats
path: root/drivers/target/target_core_user.c
blob: 9a1b314f64824e4d6d0dcd3f1dcb56968832de2e (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
/*
 * Copyright (C) 2013 Shaohua Li <shli@kernel.org>
 * Copyright (C) 2014 Red Hat, Inc.
 *
 * 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 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.
 */

#include <linux/spinlock.h>
#include <linux/module.h>
#include <linux/idr.h>
#include <linux/timer.h>
#include <linux/parser.h>
#include <scsi/scsi.h>
#include <scsi/scsi_host.h>
#include <linux/uio_driver.h>
#include <net/genetlink.h>
#include <target/target_core_base.h>
#include <target/target_core_fabric.h>
#include <target/target_core_backend.h>
#include <linux/target_core_user.h>

/*
 * Define a shared-memory interface for LIO to pass SCSI commands and
 * data to userspace for processing. This is to allow backends that
 * are too complex for in-kernel support to be possible.
 *
 * It uses the UIO framework to do a lot of the device-creation and
 * introspection work for us.
 *
 * See the .h file for how the ring is laid out. Note that while the
 * command ring is defined, the particulars of the data area are
 * not. Offset values in the command entry point to other locations
 * internal to the mmap()ed area. There is separate space outside the
 * command ring for data buffers. This leaves maximum flexibility for
 * moving buffer allocations, or even page flipping or other
 * allocation techniques, without altering the command ring layout.
 *
 * SECURITY:
 * The user process must be assumed to be malicious. There's no way to
 * prevent it breaking the command ring protocol if it wants, but in
 * order to prevent other issues we must only ever read *data* from
 * the shared memory area, not offsets or sizes. This applies to
 * command ring entries as well as the mailbox. Extra code needed for
 * this may have a 'UAM' comment.
 */


#define TCMU_TIME_OUT (30 * MSEC_PER_SEC)

#define CMDR_SIZE (16 * 4096)
#define DATA_SIZE (257 * 4096)

#define TCMU_RING_SIZE (CMDR_SIZE + DATA_SIZE)

static struct device *tcmu_root_device;

struct tcmu_hba {
	u32 host_id;
};

/* User wants all cmds or just some */
enum passthru_level {
	TCMU_PASS_ALL = 0,
	TCMU_PASS_IO,
	TCMU_PASS_INVALID,
};

#define TCMU_CONFIG_LEN 256

struct tcmu_dev {
	struct se_device se_dev;

	char *name;
	struct se_hba *hba;

#define TCMU_DEV_BIT_OPEN 0
#define TCMU_DEV_BIT_BROKEN 1
	unsigned long flags;
	enum passthru_level pass_level;

	struct uio_info uio_info;

	struct tcmu_mailbox *mb_addr;
	size_t dev_size;
	u32 cmdr_size;
	u32 cmdr_last_cleaned;
	/* Offset of data ring from start of mb */
	size_t data_off;
	size_t data_size;
	/* Ring head + tail values. */
	/* Must add data_off and mb_addr to get the address */
	size_t data_head;
	size_t data_tail;

	wait_queue_head_t wait_cmdr;
	/* TODO should this be a mutex? */
	spinlock_t cmdr_lock;

	struct idr commands;
	spinlock_t commands_lock;

	struct timer_list timeout;

	char dev_config[TCMU_CONFIG_LEN];
};

#define TCMU_DEV(_se_dev) container_of(_se_dev, struct tcmu_dev, se_dev)

#define CMDR_OFF sizeof(struct tcmu_mailbox)

struct tcmu_cmd {
	struct se_cmd *se_cmd;
	struct tcmu_dev *tcmu_dev;

	uint16_t cmd_id;

	/* Can't use se_cmd->data_length when cleaning up expired cmds, because if
	   cmd has been completed then accessing se_cmd is off limits */
	size_t data_length;

	unsigned long deadline;

#define TCMU_CMD_BIT_EXPIRED 0
	unsigned long flags;
};

static struct kmem_cache *tcmu_cmd_cache;

/* multicast group */
enum tcmu_multicast_groups {
	TCMU_MCGRP_CONFIG,
};

static const struct genl_multicast_group tcmu_mcgrps[] = {
	[TCMU_MCGRP_CONFIG] = { .name = "config", },
};

/* Our generic netlink family */
static struct genl_family tcmu_genl_family = {
	.id = GENL_ID_GENERATE,
	.hdrsize = 0,
	.name = "TCM-USER",
	.version = 1,
	.maxattr = TCMU_ATTR_MAX,
	.mcgrps = tcmu_mcgrps,
	.n_mcgrps = ARRAY_SIZE(tcmu_mcgrps),
};

static struct tcmu_cmd *tcmu_alloc_cmd(struct se_cmd *se_cmd)
{
	struct se_device *se_dev = se_cmd->se_dev;
	struct tcmu_dev *udev = TCMU_DEV(se_dev);
	struct tcmu_cmd *tcmu_cmd;
	int cmd_id;

	tcmu_cmd = kmem_cache_zalloc(tcmu_cmd_cache, GFP_KERNEL);
	if (!tcmu_cmd)
		return NULL;

	tcmu_cmd->se_cmd = se_cmd;
	tcmu_cmd->tcmu_dev = udev;
	tcmu_cmd->data_length = se_cmd->data_length;

	tcmu_cmd->deadline = jiffies + msecs_to_jiffies(TCMU_TIME_OUT);

	idr_preload(GFP_KERNEL);
	spin_lock_irq(&udev->commands_lock);
	cmd_id = idr_alloc(&udev->commands, tcmu_cmd, 0,
		USHRT_MAX, GFP_NOWAIT);
	spin_unlock_irq(&udev->commands_lock);
	idr_preload_end();

	if (cmd_id < 0) {
		kmem_cache_free(tcmu_cmd_cache, tcmu_cmd);
		return NULL;
	}
	tcmu_cmd->cmd_id = cmd_id;

	return tcmu_cmd;
}

static inline void tcmu_flush_dcache_range(void *vaddr, size_t size)
{
	unsigned long offset = (unsigned long) vaddr & ~PAGE_MASK;

	size = round_up(size+offset, PAGE_SIZE);
	vaddr -= offset;

	while (size) {
		flush_dcache_page(virt_to_page(vaddr));
		size -= PAGE_SIZE;
	}
}

/*
 * Some ring helper functions. We don't assume size is a power of 2 so
 * we can't use circ_buf.h.
 */
static inline size_t spc_used(size_t head, size_t tail, size_t size)
{
	int diff = head - tail;

	if (diff >= 0)
		return diff;
	else
		return size + diff;
}

static inline size_t spc_free(size_t head, size_t tail, size_t size)
{
	/* Keep 1 byte unused or we can't tell full from empty */
	return (size - spc_used(head, tail, size) - 1);
}

static inline size_t head_to_end(size_t head, size_t size)
{
	return size - head;
}

#define UPDATE_HEAD(head, used, size) smp_store_release(&head, ((head % size) + used) % size)

/*
 * We can't queue a command until we have space available on the cmd ring *and* space
 * space avail on the data ring.
 *
 * Called with ring lock held.
 */
static bool is_ring_space_avail(struct tcmu_dev *udev, size_t cmd_size, size_t data_needed)
{
	struct tcmu_mailbox *mb = udev->mb_addr;
	size_t space;
	u32 cmd_head;
	size_t cmd_needed;

	tcmu_flush_dcache_range(mb, sizeof(*mb));

	cmd_head = mb->cmd_head % udev->cmdr_size; /* UAM */

	/*
	 * If cmd end-of-ring space is too small then we need space for a NOP plus
	 * original cmd - cmds are internally contiguous.
	 */
	if (head_to_end(cmd_head, udev->cmdr_size) >= cmd_size)
		cmd_needed = cmd_size;
	else
		cmd_needed = cmd_size + head_to_end(cmd_head, udev->cmdr_size);

	space = spc_free(cmd_head, udev->cmdr_last_cleaned, udev->cmdr_size);
	if (space < cmd_needed) {
		pr_debug("no cmd space: %u %u %u\n", cmd_head,
		       udev->cmdr_last_cleaned, udev->cmdr_size);
		return false;
	}

	space = spc_free(udev->data_head, udev->data_tail, udev->data_size);
	if (space < data_needed) {
		pr_debug("no data space: %zu %zu %zu\n", udev->data_head,
		       udev->data_tail, udev->data_size);
		return false;
	}

	return true;
}

static int tcmu_queue_cmd_ring(struct tcmu_cmd *tcmu_cmd)
{
	struct tcmu_dev *udev = tcmu_cmd->tcmu_dev;
	struct se_cmd *se_cmd = tcmu_cmd->se_cmd;
	size_t base_command_size, command_size;
	struct tcmu_mailbox *mb;
	struct tcmu_cmd_entry *entry;
	int i;
	struct scatterlist *sg;
	struct iovec *iov;
	int iov_cnt = 0;
	uint32_t cmd_head;
	uint64_t cdb_off;

	if (test_bit(TCMU_DEV_BIT_BROKEN, &udev->flags))
		return -EINVAL;

	/*
	 * Must be a certain minimum size for response sense info, but
	 * also may be larger if the iov array is large.
	 *
	 * iovs = sgl_nents+1, for end-of-ring case, plus another 1
	 * b/c size == offsetof one-past-element.
	*/
	base_command_size = max(offsetof(struct tcmu_cmd_entry,
					 req.iov[se_cmd->t_data_nents + 2]),
				sizeof(struct tcmu_cmd_entry));
	command_size = base_command_size
		+ round_up(scsi_command_size(se_cmd->t_task_cdb), TCMU_OP_ALIGN_SIZE);

	WARN_ON(command_size & (TCMU_OP_ALIGN_SIZE-1));

	spin_lock_irq(&udev->cmdr_lock);

	mb = udev->mb_addr;
	cmd_head = mb->cmd_head % udev->cmdr_size; /* UAM */
	if ((command_size > (udev->cmdr_size / 2))
	    || tcmu_cmd->data_length > (udev->data_size - 1))
		pr_warn("TCMU: Request of size %zu/%zu may be too big for %u/%zu "
			"cmd/data ring buffers\n", command_size, tcmu_cmd->data_length,
			udev->cmdr_size, udev->data_size);

	while (!is_ring_space_avail(udev, command_size, tcmu_cmd->data_length)) {
		int ret;
		DEFINE_WAIT(__wait);

		prepare_to_wait(&udev->wait_cmdr, &__wait, TASK_INTERRUPTIBLE);

		pr_debug("sleeping for ring space\n");
		spin_unlock_irq(&udev->cmdr_lock);
		ret = schedule_timeout(msecs_to_jiffies(TCMU_TIME_OUT));
		finish_wait(&udev->wait_cmdr, &__wait);
		if (!ret) {
			pr_warn("tcmu: command timed out\n");
			return -ETIMEDOUT;
		}

		spin_lock_irq(&udev->cmdr_lock);

		/* We dropped cmdr_lock, cmd_head is stale */
		cmd_head = mb->cmd_head % udev->cmdr_size; /* UAM */
	}

	/* Insert a PAD if end-of-ring space is too small */
	if (head_to_end(cmd_head, udev->cmdr_size) < command_size) {
		size_t pad_size = head_to_end(cmd_head, udev->cmdr_size);

		entry = (void *) mb + CMDR_OFF + cmd_head;
		tcmu_flush_dcache_range(entry, sizeof(*entry));
		tcmu_hdr_set_op(&entry->hdr, TCMU_OP_PAD);
		tcmu_hdr_set_len(&entry->hdr, pad_size);

		UPDATE_HEAD(mb->cmd_head, pad_size, udev->cmdr_size);

		cmd_head = mb->cmd_head % udev->cmdr_size; /* UAM */
		WARN_ON(cmd_head != 0);
	}

	entry = (void *) mb + CMDR_OFF + cmd_head;
	tcmu_flush_dcache_range(entry, sizeof(*entry));
	tcmu_hdr_set_op(&entry->hdr, TCMU_OP_CMD);
	tcmu_hdr_set_len(&entry->hdr, command_size);
	entry->cmd_id = tcmu_cmd->cmd_id;

	/*
	 * Fix up iovecs, and handle if allocation in data ring wrapped.
	 */
	iov = &entry->req.iov[0];
	for_each_sg(se_cmd->t_data_sg, sg, se_cmd->t_data_nents, i) {
		size_t copy_bytes = min((size_t)sg->length,
				     head_to_end(udev->data_head, udev->data_size));
		void *from = kmap_atomic(sg_page(sg)) + sg->offset;
		void *to = (void *) mb + udev->data_off + udev->data_head;

		if (tcmu_cmd->se_cmd->data_direction == DMA_TO_DEVICE) {
			memcpy(to, from, copy_bytes);
			tcmu_flush_dcache_range(to, copy_bytes);
		}

		/* Even iov_base is relative to mb_addr */
		iov->iov_len = copy_bytes;
		iov->iov_base = (void *) udev->data_off + udev->data_head;
		iov_cnt++;
		iov++;

		UPDATE_HEAD(udev->data_head, copy_bytes, udev->data_size);

		/* Uh oh, we wrapped the buffer. Must split sg across 2 iovs. */
		if (sg->length != copy_bytes) {
			from += copy_bytes;
			copy_bytes = sg->length - copy_bytes;

			iov->iov_len = copy_bytes;
			iov->iov_base = (void *) udev->data_off + udev->data_head;

			if (se_cmd->data_direction == DMA_TO_DEVICE) {
				to = (void *) mb + udev->data_off + udev->data_head;
				memcpy(to, from, copy_bytes);
				tcmu_flush_dcache_range(to, copy_bytes);
			}

			iov_cnt++;
			iov++;

			UPDATE_HEAD(udev->data_head, copy_bytes, udev->data_size);
		}

		kunmap_atomic(from);
	}
	entry->req.iov_cnt = iov_cnt;

	/* All offsets relative to mb_addr, not start of entry! */
	cdb_off = CMDR_OFF + cmd_head + base_command_size;
	memcpy((void *) mb + cdb_off, se_cmd->t_task_cdb, scsi_command_size(se_cmd->t_task_cdb));
	entry->req.cdb_off = cdb_off;
	tcmu_flush_dcache_range(entry, sizeof(*entry));

	UPDATE_HEAD(mb->cmd_head, command_size, udev->cmdr_size);
	tcmu_flush_dcache_range(mb, sizeof(*mb));

	spin_unlock_irq(&udev->cmdr_lock);

	/* TODO: only if FLUSH and FUA? */
	uio_event_notify(&udev->uio_info);

	mod_timer(&udev->timeout,
		round_jiffies_up(jiffies + msecs_to_jiffies(TCMU_TIME_OUT)));

	return 0;
}

static int tcmu_queue_cmd(struct se_cmd *se_cmd)
{
	struct se_device *se_dev = se_cmd->se_dev;
	struct tcmu_dev *udev = TCMU_DEV(se_dev);
	struct tcmu_cmd *tcmu_cmd;
	int ret;

	tcmu_cmd = tcmu_alloc_cmd(se_cmd);
	if (!tcmu_cmd)
		return -ENOMEM;

	ret = tcmu_queue_cmd_ring(tcmu_cmd);
	if (ret < 0) {
		pr_err("TCMU: Could not queue command\n");
		spin_lock_irq(&udev->commands_lock);
		idr_remove(&udev->commands, tcmu_cmd->cmd_id);
		spin_unlock_irq(&udev->commands_lock);

		kmem_cache_free(tcmu_cmd_cache, tcmu_cmd);
	}

	return ret;
}

static void tcmu_handle_completion(struct tcmu_cmd *cmd, struct tcmu_cmd_entry *entry)
{
	struct se_cmd *se_cmd = cmd->se_cmd;
	struct tcmu_dev *udev = cmd->tcmu_dev;

	if (test_bit(TCMU_CMD_BIT_EXPIRED, &cmd->flags)) {
		/* cmd has been completed already from timeout, just reclaim data
		   ring space */
		UPDATE_HEAD(udev->data_tail, cmd->data_length, udev->data_size);
		return;
	}

	if (entry->rsp.scsi_status == SAM_STAT_CHECK_CONDITION) {
		memcpy(se_cmd->sense_buffer, entry->rsp.sense_buffer,
			       se_cmd->scsi_sense_length);

		UPDATE_HEAD(udev->data_tail, cmd->data_length, udev->data_size);
	}
	else if (se_cmd->data_direction == DMA_FROM_DEVICE) {
		struct scatterlist *sg;
		int i;

		/* It'd be easier to look at entry's iovec again, but UAM */
		for_each_sg(se_cmd->t_data_sg, sg, se_cmd->t_data_nents, i) {
			size_t copy_bytes;
			void *to;
			void *from;

			copy_bytes = min((size_t)sg->length,
					 head_to_end(udev->data_tail, udev->data_size));

			to = kmap_atomic(sg_page(sg)) + sg->offset;
			WARN_ON(sg->length + sg->offset > PAGE_SIZE);
			from = (void *) udev->mb_addr + udev->data_off + udev->data_tail;
			tcmu_flush_dcache_range(from, copy_bytes);
			memcpy(to, from, copy_bytes);

			UPDATE_HEAD(udev->data_tail, copy_bytes, udev->data_size);

			/* Uh oh, wrapped the data buffer for this sg's data */
			if (sg->length != copy_bytes) {
				from = (void *) udev->mb_addr + udev->data_off + udev->data_tail;
				WARN_ON(udev->data_tail);
				to += copy_bytes;
				copy_bytes = sg->length - copy_bytes;
				tcmu_flush_dcache_range(from, copy_bytes);
				memcpy(to, from, copy_bytes);

				UPDATE_HEAD(udev->data_tail, copy_bytes, udev->data_size);
			}

			kunmap_atomic(to);
		}

	} else if (se_cmd->data_direction == DMA_TO_DEVICE) {
		UPDATE_HEAD(udev->data_tail, cmd->data_length, udev->data_size);
	} else {
		pr_warn("TCMU: data direction was %d!\n", se_cmd->data_direction);
	}

	target_complete_cmd(cmd->se_cmd, entry->rsp.scsi_status);
	cmd->se_cmd = NULL;

	kmem_cache_free(tcmu_cmd_cache, cmd);
}

static unsigned int tcmu_handle_completions(struct tcmu_dev *udev)
{
	struct tcmu_mailbox *mb;
	LIST_HEAD(cpl_cmds);
	unsigned long flags;
	int handled = 0;

	if (test_bit(TCMU_DEV_BIT_BROKEN, &udev->flags)) {
		pr_err("ring broken, not handling completions\n");
		return 0;
	}

	spin_lock_irqsave(&udev->cmdr_lock, flags);

	mb = udev->mb_addr;
	tcmu_flush_dcache_range(mb, sizeof(*mb));

	while (udev->cmdr_last_cleaned != ACCESS_ONCE(mb->cmd_tail)) {

		struct tcmu_cmd_entry *entry = (void *) mb + CMDR_OFF + udev->cmdr_last_cleaned;
		struct tcmu_cmd *cmd;

		tcmu_flush_dcache_range(entry, sizeof(*entry));

		if (tcmu_hdr_get_op(&entry->hdr) == TCMU_OP_PAD) {
			UPDATE_HEAD(udev->cmdr_last_cleaned, tcmu_hdr_get_len(&entry->hdr), udev->cmdr_size);
			continue;
		}
		WARN_ON(tcmu_hdr_get_op(&entry->hdr) != TCMU_OP_CMD);

		spin_lock(&udev->commands_lock);
		cmd = idr_find(&udev->commands, entry->cmd_id);
		if (cmd)
			idr_remove(&udev->commands, cmd->cmd_id);
		spin_unlock(&udev->commands_lock);

		if (!cmd) {
			pr_err("cmd_id not found, ring is broken\n");
			set_bit(TCMU_DEV_BIT_BROKEN, &udev->flags);
			break;
		}

		tcmu_handle_completion(cmd, entry);

		UPDATE_HEAD(udev->cmdr_last_cleaned, tcmu_hdr_get_len(&entry->hdr), udev->cmdr_size);

		handled++;
	}

	if (mb->cmd_tail == mb->cmd_head)
		del_timer(&udev->timeout); /* no more pending cmds */

	spin_unlock_irqrestore(&udev->cmdr_lock, flags);

	wake_up(&udev->wait_cmdr);

	return handled;
}

static int tcmu_check_expired_cmd(int id, void *p, void *data)
{
	struct tcmu_cmd *cmd = p;

	if (test_bit(TCMU_CMD_BIT_EXPIRED, &cmd->flags))
		return 0;

	if (!time_after(cmd->deadline, jiffies))
		return 0;

	set_bit(TCMU_CMD_BIT_EXPIRED, &cmd->flags);
	target_complete_cmd(cmd->se_cmd, SAM_STAT_CHECK_CONDITION);
	cmd->se_cmd = NULL;

	kmem_cache_free(tcmu_cmd_cache, cmd);

	return 0;
}

static void tcmu_device_timedout(unsigned long data)
{
	struct tcmu_dev *udev = (struct tcmu_dev *)data;
	unsigned long flags;
	int handled;

	handled = tcmu_handle_completions(udev);

	pr_warn("%d completions handled from timeout\n", handled);

	spin_lock_irqsave(&udev->commands_lock, flags);
	idr_for_each(&udev->commands, tcmu_check_expired_cmd, NULL);
	spin_unlock_irqrestore(&udev->commands_lock, flags);

	/*
	 * We don't need to wakeup threads on wait_cmdr since they have their
	 * own timeout.
	 */
}

static int tcmu_attach_hba(struct se_hba *hba, u32 host_id)
{
	struct tcmu_hba *tcmu_hba;

	tcmu_hba = kzalloc(sizeof(struct tcmu_hba), GFP_KERNEL);
	if (!tcmu_hba)
		return -ENOMEM;

	tcmu_hba->host_id = host_id;
	hba->hba_ptr = tcmu_hba;

	return 0;
}

static void tcmu_detach_hba(struct se_hba *hba)
{
	kfree(hba->hba_ptr);
	hba->hba_ptr = NULL;
}

static struct se_device *tcmu_alloc_device(struct se_hba *hba, const char *name)
{
	struct tcmu_dev *udev;

	udev = kzalloc(sizeof(struct tcmu_dev), GFP_KERNEL);
	if (!udev)
		return NULL;

	udev->name = kstrdup(name, GFP_KERNEL);
	if (!udev->name) {
		kfree(udev);
		return NULL;
	}

	udev->hba = hba;

	init_waitqueue_head(&udev->wait_cmdr);
	spin_lock_init(&udev->cmdr_lock);

	idr_init(&udev->commands);
	spin_lock_init(&udev->commands_lock);

	setup_timer(&udev->timeout, tcmu_device_timedout,
		(unsigned long)udev);

	udev->pass_level = TCMU_PASS_ALL;

	return &udev->se_dev;
}

static int tcmu_irqcontrol(struct uio_info *info, s32 irq_on)
{
	struct tcmu_dev *tcmu_dev = container_of(info, struct tcmu_dev, uio_info);

	tcmu_handle_completions(tcmu_dev);

	return 0;
}

/*
 * mmap code from uio.c. Copied here because we want to hook mmap()
 * and this stuff must come along.
 */
static int tcmu_find_mem_index(struct vm_area_struct *vma)
{
	struct tcmu_dev *udev = vma->vm_private_data;
	struct uio_info *info = &udev->uio_info;

	if (vma->vm_pgoff < MAX_UIO_MAPS) {
		if (info->mem[vma->vm_pgoff].size == 0)
			return -1;
		return (int)vma->vm_pgoff;
	}
	return -1;
}

static int tcmu_vma_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
{
	struct tcmu_dev *udev = vma->vm_private_data;
	struct uio_info *info = &udev->uio_info;
	struct page *page;
	unsigned long offset;
	void *addr;

	int mi = tcmu_find_mem_index(vma);
	if (mi < 0)
		return VM_FAULT_SIGBUS;

	/*
	 * We need to subtract mi because userspace uses offset = N*PAGE_SIZE
	 * to use mem[N].
	 */
	offset = (vmf->pgoff - mi) << PAGE_SHIFT;

	addr = (void *)(unsigned long)info->mem[mi].addr + offset;
	if (info->mem[mi].memtype == UIO_MEM_LOGICAL)
		page = virt_to_page(addr);
	else
		page = vmalloc_to_page(addr);
	get_page(page);
	vmf->page = page;
	return 0;
}

static const struct vm_operations_struct tcmu_vm_ops = {
	.fault = tcmu_vma_fault,
};

static int tcmu_mmap(struct uio_info *info, struct vm_area_struct *vma)
{
	struct tcmu_dev *udev = container_of(info, struct tcmu_dev, uio_info);

	vma->vm_flags |= VM_DONTEXPAND | VM_DONTDUMP;
	vma->vm_ops = &tcmu_vm_ops;

	vma->vm_private_data = udev;

	/* Ensure the mmap is exactly the right size */
	if (vma_pages(vma) != (TCMU_RING_SIZE >> PAGE_SHIFT))
		return -EINVAL;

	return 0;
}

static int tcmu_open(struct uio_info *info, struct inode *inode)
{
	struct tcmu_dev *udev = container_of(info, struct tcmu_dev, uio_info);

	/* O_EXCL not supported for char devs, so fake it? */
	if (test_and_set_bit(TCMU_DEV_BIT_OPEN, &udev->flags))
		return -EBUSY;

	pr_debug("open\n");

	return 0;
}

static int tcmu_release(struct uio_info *info, struct inode *inode)
{
	struct tcmu_dev *udev = container_of(info, struct tcmu_dev, uio_info);

	clear_bit(TCMU_DEV_BIT_OPEN, &udev->flags);

	pr_debug("close\n");

	return 0;
}

static int tcmu_netlink_event(enum tcmu_genl_cmd cmd, const char *name, int minor)
{
	struct sk_buff *skb;
	void *msg_header;
	int ret = -ENOMEM;

	skb = genlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
	if (!skb)
		return ret;

	msg_header = genlmsg_put(skb, 0, 0, &tcmu_genl_family, 0, cmd);
	if (!msg_header)
		goto free_skb;

	ret = nla_put_string(skb, TCMU_ATTR_DEVICE, name);
	if (ret < 0)
		goto free_skb;

	ret = nla_put_u32(skb, TCMU_ATTR_MINOR, minor);
	if (ret < 0)
		goto free_skb;

	ret = genlmsg_end(skb, msg_header);
	if (ret < 0)
		goto free_skb;

	ret = genlmsg_multicast(&tcmu_genl_family, skb, 0,
				TCMU_MCGRP_CONFIG, GFP_KERNEL);

	/* We don't care if no one is listening */
	if (ret == -ESRCH)
		ret = 0;

	return ret;
free_skb:
	nlmsg_free(skb);
	return ret;
}

static int tcmu_configure_device(struct se_device *dev)
{
	struct tcmu_dev *udev = TCMU_DEV(dev);
	struct tcmu_hba *hba = udev->hba->hba_ptr;
	struct uio_info *info;
	struct tcmu_mailbox *mb;
	size_t size;
	size_t used;
	int ret = 0;
	char *str;

	info = &udev->uio_info;

	size = snprintf(NULL, 0, "tcm-user/%u/%s/%s", hba->host_id, udev->name,
			udev->dev_config);
	size += 1; /* for \0 */
	str = kmalloc(size, GFP_KERNEL);
	if (!str)
		return -ENOMEM;

	used = snprintf(str, size, "tcm-user/%u/%s", hba->host_id, udev->name);

	if (udev->dev_config[0])
		snprintf(str + used, size - used, "/%s", udev->dev_config);

	info->name = str;

	udev->mb_addr = vzalloc(TCMU_RING_SIZE);
	if (!udev->mb_addr) {
		ret = -ENOMEM;
		goto err_vzalloc;
	}

	/* mailbox fits in first part of CMDR space */
	udev->cmdr_size = CMDR_SIZE - CMDR_OFF;
	udev->data_off = CMDR_SIZE;
	udev->data_size = TCMU_RING_SIZE - CMDR_SIZE;

	mb = udev->mb_addr;
	mb->version = 1;
	mb->cmdr_off = CMDR_OFF;
	mb->cmdr_size = udev->cmdr_size;

	WARN_ON(!PAGE_ALIGNED(udev->data_off));
	WARN_ON(udev->data_size % PAGE_SIZE);

	info->version = "1";

	info->mem[0].name = "tcm-user command & data buffer";
	info->mem[0].addr = (phys_addr_t) udev->mb_addr;
	info->mem[0].size = TCMU_RING_SIZE;
	info->mem[0].memtype = UIO_MEM_VIRTUAL;

	info->irqcontrol = tcmu_irqcontrol;
	info->irq = UIO_IRQ_CUSTOM;

	info->mmap = tcmu_mmap;
	info->open = tcmu_open;
	info->release = tcmu_release;

	ret = uio_register_device(tcmu_root_device, info);
	if (ret)
		goto err_register;

	/* Other attributes can be configured in userspace */
	dev->dev_attrib.hw_block_size = 512;
	dev->dev_attrib.hw_max_sectors = 128;
	dev->dev_attrib.hw_queue_depth = 128;

	ret = tcmu_netlink_event(TCMU_CMD_ADDED_DEVICE, udev->uio_info.name,
				 udev->uio_info.uio_dev->minor);
	if (ret)
		goto err_netlink;

	return 0;

err_netlink:
	uio_unregister_device(&udev->uio_info);
err_register:
	vfree(udev->mb_addr);
err_vzalloc:
	kfree(info->name);

	return ret;
}

static int tcmu_check_pending_cmd(int id, void *p, void *data)
{
	struct tcmu_cmd *cmd = p;

	if (test_bit(TCMU_CMD_BIT_EXPIRED, &cmd->flags))
		return 0;
	return -EINVAL;
}

static void tcmu_free_device(struct se_device *dev)
{
	struct tcmu_dev *udev = TCMU_DEV(dev);
	int i;

	del_timer_sync(&udev->timeout);

	vfree(udev->mb_addr);

	/* Upper layer should drain all requests before calling this */
	spin_lock_irq(&udev->commands_lock);
	i = idr_for_each(&udev->commands, tcmu_check_pending_cmd, NULL);
	idr_destroy(&udev->commands);
	spin_unlock_irq(&udev->commands_lock);
	WARN_ON(i);

	/* Device was configured */
	if (udev->uio_info.uio_dev) {
		tcmu_netlink_event(TCMU_CMD_REMOVED_DEVICE, udev->uio_info.name,
				   udev->uio_info.uio_dev->minor);

		uio_unregister_device(&udev->uio_info);
		kfree(udev->uio_info.name);
		kfree(udev->name);
	}

	kfree(udev);
}

enum {
	Opt_dev_config, Opt_dev_size, Opt_err, Opt_pass_level,
};

static match_table_t tokens = {
	{Opt_dev_config, "dev_config=%s"},
	{Opt_dev_size, "dev_size=%u"},
	{Opt_pass_level, "pass_level=%u"},
	{Opt_err, NULL}
};

static ssize_t tcmu_set_configfs_dev_params(struct se_device *dev,
		const char *page, ssize_t count)
{
	struct tcmu_dev *udev = TCMU_DEV(dev);
	char *orig, *ptr, *opts, *arg_p;
	substring_t args[MAX_OPT_ARGS];
	int ret = 0, token;
	int arg;

	opts = kstrdup(page, GFP_KERNEL);
	if (!opts)
		return -ENOMEM;

	orig = opts;

	while ((ptr = strsep(&opts, ",\n")) != NULL) {
		if (!*ptr)
			continue;

		token = match_token(ptr, tokens, args);
		switch (token) {
		case Opt_dev_config:
			if (match_strlcpy(udev->dev_config, &args[0],
					  TCMU_CONFIG_LEN) == 0) {
				ret = -EINVAL;
				break;
			}
			pr_debug("TCMU: Referencing Path: %s\n", udev->dev_config);
			break;
		case Opt_dev_size:
			arg_p = match_strdup(&args[0]);
			if (!arg_p) {
				ret = -ENOMEM;
				break;
			}
			ret = kstrtoul(arg_p, 0, (unsigned long *) &udev->dev_size);
			kfree(arg_p);
			if (ret < 0)
				pr_err("kstrtoul() failed for dev_size=\n");
			break;
		case Opt_pass_level:
			match_int(args, &arg);
			if (arg >= TCMU_PASS_INVALID) {
				pr_warn("TCMU: Invalid pass_level: %d\n", arg);
				break;
			}

			pr_debug("TCMU: Setting pass_level to %d\n", arg);
			udev->pass_level = arg;
			break;
		default:
			break;
		}
	}

	kfree(orig);
	return (!ret) ? count : ret;
}

static ssize_t tcmu_show_configfs_dev_params(struct se_device *dev, char *b)
{
	struct tcmu_dev *udev = TCMU_DEV(dev);
	ssize_t bl = 0;

	bl = sprintf(b + bl, "Config: %s ",
		     udev->dev_config[0] ? udev->dev_config : "NULL");
	bl += sprintf(b + bl, "Size: %zu PassLevel: %u\n",
		      udev->dev_size, udev->pass_level);

	return bl;
}

static sector_t tcmu_get_blocks(struct se_device *dev)
{
	struct tcmu_dev *udev = TCMU_DEV(dev);

	return div_u64(udev->dev_size - dev->dev_attrib.block_size,
		       dev->dev_attrib.block_size);
}

static sense_reason_t
tcmu_execute_rw(struct se_cmd *se_cmd, struct scatterlist *sgl, u32 sgl_nents,
		enum dma_data_direction data_direction)
{
	int ret;

	ret = tcmu_queue_cmd(se_cmd);

	if (ret != 0)
		return TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
	else
		return TCM_NO_SENSE;
}

static sense_reason_t
tcmu_pass_op(struct se_cmd *se_cmd)
{
	int ret = tcmu_queue_cmd(se_cmd);

	if (ret != 0)
		return TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
	else
		return TCM_NO_SENSE;
}

static struct sbc_ops tcmu_sbc_ops = {
	.execute_rw = tcmu_execute_rw,
	.execute_sync_cache	= tcmu_pass_op,
	.execute_write_same	= tcmu_pass_op,
	.execute_write_same_unmap = tcmu_pass_op,
	.execute_unmap		= tcmu_pass_op,
};

static sense_reason_t
tcmu_parse_cdb(struct se_cmd *cmd)
{
	unsigned char *cdb = cmd->t_task_cdb;
	struct tcmu_dev *udev = TCMU_DEV(cmd->se_dev);
	sense_reason_t ret;

	switch (udev->pass_level) {
	case TCMU_PASS_ALL:
		/* We're just like pscsi, then */
		/*
		 * For REPORT LUNS we always need to emulate the response, for everything
		 * else, pass it up.
		 */
		switch (cdb[0]) {
		case REPORT_LUNS:
			cmd->execute_cmd = spc_emulate_report_luns;
			break;
		case READ_6:
		case READ_10:
		case READ_12:
		case READ_16:
		case WRITE_6:
		case WRITE_10:
		case WRITE_12:
		case WRITE_16:
		case WRITE_VERIFY:
			cmd->se_cmd_flags |= SCF_SCSI_DATA_CDB;
			/* FALLTHROUGH */
		default:
			cmd->execute_cmd = tcmu_pass_op;
		}
		ret = TCM_NO_SENSE;
		break;
	case TCMU_PASS_IO:
		ret = sbc_parse_cdb(cmd, &tcmu_sbc_ops);
		break;
	default:
		pr_err("Unknown tcm-user pass level %d\n", udev->pass_level);
		ret = TCM_CHECK_CONDITION_ABORT_CMD;
	}

	return ret;
}

static struct se_subsystem_api tcmu_template = {
	.name			= "user",
	.inquiry_prod		= "USER",
	.inquiry_rev		= TCMU_VERSION,
	.owner			= THIS_MODULE,
	.transport_type		= TRANSPORT_PLUGIN_VHBA_PDEV,
	.attach_hba		= tcmu_attach_hba,
	.detach_hba		= tcmu_detach_hba,
	.alloc_device		= tcmu_alloc_device,
	.configure_device	= tcmu_configure_device,
	.free_device		= tcmu_free_device,
	.parse_cdb		= tcmu_parse_cdb,
	.set_configfs_dev_params = tcmu_set_configfs_dev_params,
	.show_configfs_dev_params = tcmu_show_configfs_dev_params,
	.get_device_type	= sbc_get_device_type,
	.get_blocks		= tcmu_get_blocks,
};

static int __init tcmu_module_init(void)
{
	int ret;

	BUILD_BUG_ON((sizeof(struct tcmu_cmd_entry) % TCMU_OP_ALIGN_SIZE) != 0);

	tcmu_cmd_cache = kmem_cache_create("tcmu_cmd_cache",
				sizeof(struct tcmu_cmd),
				__alignof__(struct tcmu_cmd),
				0, NULL);
	if (!tcmu_cmd_cache)
		return -ENOMEM;

	tcmu_root_device = root_device_register("tcm_user");
	if (IS_ERR(tcmu_root_device)) {
		ret = PTR_ERR(tcmu_root_device);
		goto out_free_cache;
	}

	ret = genl_register_family(&tcmu_genl_family);
	if (ret < 0) {
		goto out_unreg_device;
	}

	ret = transport_subsystem_register(&tcmu_template);
	if (ret)
		goto out_unreg_genl;

	return 0;

out_unreg_genl:
	genl_unregister_family(&tcmu_genl_family);
out_unreg_device:
	root_device_unregister(tcmu_root_device);
out_free_cache:
	kmem_cache_destroy(tcmu_cmd_cache);

	return ret;
}

static void __exit tcmu_module_exit(void)
{
	transport_subsystem_release(&tcmu_template);
	genl_unregister_family(&tcmu_genl_family);
	root_device_unregister(tcmu_root_device);
	kmem_cache_destroy(tcmu_cmd_cache);
}

MODULE_DESCRIPTION("TCM USER subsystem plugin");
MODULE_AUTHOR("Shaohua Li <shli@kernel.org>");
MODULE_AUTHOR("Andy Grover <agrover@redhat.com>");
MODULE_LICENSE("GPL");

module_init(tcmu_module_init);
module_exit(tcmu_module_exit);