summaryrefslogtreecommitdiff
path: root/drivers/block/sunvdc.c
blob: c6d43a2a807dd4c2a032cf2c520db923dc7230b0 (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
/* sunvdc.c: Sun LDOM Virtual Disk Client.
 *
 * Copyright (C) 2007, 2008 David S. Miller <davem@davemloft.net>
 */

#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/types.h>
#include <linux/blkdev.h>
#include <linux/hdreg.h>
#include <linux/genhd.h>
#include <linux/cdrom.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/completion.h>
#include <linux/delay.h>
#include <linux/init.h>
#include <linux/list.h>
#include <linux/scatterlist.h>

#include <asm/vio.h>
#include <asm/ldc.h>

#define DRV_MODULE_NAME		"sunvdc"
#define PFX DRV_MODULE_NAME	": "
#define DRV_MODULE_VERSION	"1.2"
#define DRV_MODULE_RELDATE	"November 24, 2014"

static char version[] =
	DRV_MODULE_NAME ".c:v" DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n";
MODULE_AUTHOR("David S. Miller (davem@davemloft.net)");
MODULE_DESCRIPTION("Sun LDOM virtual disk client driver");
MODULE_LICENSE("GPL");
MODULE_VERSION(DRV_MODULE_VERSION);

#define VDC_TX_RING_SIZE	512

#define WAITING_FOR_LINK_UP	0x01
#define WAITING_FOR_TX_SPACE	0x02
#define WAITING_FOR_GEN_CMD	0x04
#define WAITING_FOR_ANY		-1

#define	VDC_MAX_RETRIES	10

static struct workqueue_struct *sunvdc_wq;

struct vdc_req_entry {
	struct request		*req;
};

struct vdc_port {
	struct vio_driver_state	vio;

	struct gendisk		*disk;

	struct vdc_completion	*cmp;

	u64			req_id;
	u64			seq;
	struct vdc_req_entry	rq_arr[VDC_TX_RING_SIZE];

	unsigned long		ring_cookies;

	u64			max_xfer_size;
	u32			vdisk_block_size;

	u64			ldc_timeout;
	struct timer_list	ldc_reset_timer;
	struct work_struct	ldc_reset_work;

	/* The server fills these in for us in the disk attribute
	 * ACK packet.
	 */
	u64			operations;
	u32			vdisk_size;
	u8			vdisk_type;
	u8			vdisk_mtype;

	char			disk_name[32];
};

static void vdc_ldc_reset(struct vdc_port *port);
static void vdc_ldc_reset_work(struct work_struct *work);
static void vdc_ldc_reset_timer(unsigned long _arg);

static inline struct vdc_port *to_vdc_port(struct vio_driver_state *vio)
{
	return container_of(vio, struct vdc_port, vio);
}

/* Ordered from largest major to lowest */
static struct vio_version vdc_versions[] = {
	{ .major = 1, .minor = 1 },
	{ .major = 1, .minor = 0 },
};

static inline int vdc_version_supported(struct vdc_port *port,
					u16 major, u16 minor)
{
	return port->vio.ver.major == major && port->vio.ver.minor >= minor;
}

#define VDCBLK_NAME	"vdisk"
static int vdc_major;
#define PARTITION_SHIFT	3

static inline u32 vdc_tx_dring_avail(struct vio_dring_state *dr)
{
	return vio_dring_avail(dr, VDC_TX_RING_SIZE);
}

static int vdc_getgeo(struct block_device *bdev, struct hd_geometry *geo)
{
	struct gendisk *disk = bdev->bd_disk;
	sector_t nsect = get_capacity(disk);
	sector_t cylinders = nsect;

	geo->heads = 0xff;
	geo->sectors = 0x3f;
	sector_div(cylinders, geo->heads * geo->sectors);
	geo->cylinders = cylinders;
	if ((sector_t)(geo->cylinders + 1) * geo->heads * geo->sectors < nsect)
		geo->cylinders = 0xffff;

	return 0;
}

/* Add ioctl/CDROM_GET_CAPABILITY to support cdrom_id in udev
 * when vdisk_mtype is VD_MEDIA_TYPE_CD or VD_MEDIA_TYPE_DVD.
 * Needed to be able to install inside an ldom from an iso image.
 */
static int vdc_ioctl(struct block_device *bdev, fmode_t mode,
		     unsigned command, unsigned long argument)
{
	int i;
	struct gendisk *disk;

	switch (command) {
	case CDROMMULTISESSION:
		pr_debug(PFX "Multisession CDs not supported\n");
		for (i = 0; i < sizeof(struct cdrom_multisession); i++)
			if (put_user(0, (char __user *)(argument + i)))
				return -EFAULT;
		return 0;

	case CDROM_GET_CAPABILITY:
		disk = bdev->bd_disk;

		if (bdev->bd_disk && (disk->flags & GENHD_FL_CD))
			return 0;
		return -EINVAL;

	default:
		pr_debug(PFX "ioctl %08x not supported\n", command);
		return -EINVAL;
	}
}

static const struct block_device_operations vdc_fops = {
	.owner		= THIS_MODULE,
	.getgeo		= vdc_getgeo,
	.ioctl		= vdc_ioctl,
};

static void vdc_blk_queue_start(struct vdc_port *port)
{
	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];

	/* restart blk queue when ring is half emptied. also called after
	 * handshake completes, so check for initial handshake before we've
	 * allocated a disk.
	 */
	if (port->disk && blk_queue_stopped(port->disk->queue) &&
	    vdc_tx_dring_avail(dr) * 100 / VDC_TX_RING_SIZE >= 50) {
		blk_start_queue(port->disk->queue);
	}

}

static void vdc_finish(struct vio_driver_state *vio, int err, int waiting_for)
{
	if (vio->cmp &&
	    (waiting_for == -1 ||
	     vio->cmp->waiting_for == waiting_for)) {
		vio->cmp->err = err;
		complete(&vio->cmp->com);
		vio->cmp = NULL;
	}
}

static void vdc_handshake_complete(struct vio_driver_state *vio)
{
	struct vdc_port *port = to_vdc_port(vio);

	del_timer(&port->ldc_reset_timer);
	vdc_finish(vio, 0, WAITING_FOR_LINK_UP);
	vdc_blk_queue_start(port);
}

static int vdc_handle_unknown(struct vdc_port *port, void *arg)
{
	struct vio_msg_tag *pkt = arg;

	printk(KERN_ERR PFX "Received unknown msg [%02x:%02x:%04x:%08x]\n",
	       pkt->type, pkt->stype, pkt->stype_env, pkt->sid);
	printk(KERN_ERR PFX "Resetting connection.\n");

	ldc_disconnect(port->vio.lp);

	return -ECONNRESET;
}

static int vdc_send_attr(struct vio_driver_state *vio)
{
	struct vdc_port *port = to_vdc_port(vio);
	struct vio_disk_attr_info pkt;

	memset(&pkt, 0, sizeof(pkt));

	pkt.tag.type = VIO_TYPE_CTRL;
	pkt.tag.stype = VIO_SUBTYPE_INFO;
	pkt.tag.stype_env = VIO_ATTR_INFO;
	pkt.tag.sid = vio_send_sid(vio);

	pkt.xfer_mode = VIO_DRING_MODE;
	pkt.vdisk_block_size = port->vdisk_block_size;
	pkt.max_xfer_size = port->max_xfer_size;

	viodbg(HS, "SEND ATTR xfer_mode[0x%x] blksz[%u] max_xfer[%llu]\n",
	       pkt.xfer_mode, pkt.vdisk_block_size, pkt.max_xfer_size);

	return vio_ldc_send(&port->vio, &pkt, sizeof(pkt));
}

static int vdc_handle_attr(struct vio_driver_state *vio, void *arg)
{
	struct vdc_port *port = to_vdc_port(vio);
	struct vio_disk_attr_info *pkt = arg;

	viodbg(HS, "GOT ATTR stype[0x%x] ops[%llx] disk_size[%llu] disk_type[%x] "
	       "mtype[0x%x] xfer_mode[0x%x] blksz[%u] max_xfer[%llu]\n",
	       pkt->tag.stype, pkt->operations,
	       pkt->vdisk_size, pkt->vdisk_type, pkt->vdisk_mtype,
	       pkt->xfer_mode, pkt->vdisk_block_size,
	       pkt->max_xfer_size);

	if (pkt->tag.stype == VIO_SUBTYPE_ACK) {
		switch (pkt->vdisk_type) {
		case VD_DISK_TYPE_DISK:
		case VD_DISK_TYPE_SLICE:
			break;

		default:
			printk(KERN_ERR PFX "%s: Bogus vdisk_type 0x%x\n",
			       vio->name, pkt->vdisk_type);
			return -ECONNRESET;
		}

		if (pkt->vdisk_block_size > port->vdisk_block_size) {
			printk(KERN_ERR PFX "%s: BLOCK size increased "
			       "%u --> %u\n",
			       vio->name,
			       port->vdisk_block_size, pkt->vdisk_block_size);
			return -ECONNRESET;
		}

		port->operations = pkt->operations;
		port->vdisk_type = pkt->vdisk_type;
		if (vdc_version_supported(port, 1, 1)) {
			port->vdisk_size = pkt->vdisk_size;
			port->vdisk_mtype = pkt->vdisk_mtype;
		}
		if (pkt->max_xfer_size < port->max_xfer_size)
			port->max_xfer_size = pkt->max_xfer_size;
		port->vdisk_block_size = pkt->vdisk_block_size;
		return 0;
	} else {
		printk(KERN_ERR PFX "%s: Attribute NACK\n", vio->name);

		return -ECONNRESET;
	}
}

static void vdc_end_special(struct vdc_port *port, struct vio_disk_desc *desc)
{
	int err = desc->status;

	vdc_finish(&port->vio, -err, WAITING_FOR_GEN_CMD);
}

static void vdc_end_one(struct vdc_port *port, struct vio_dring_state *dr,
			unsigned int index)
{
	struct vio_disk_desc *desc = vio_dring_entry(dr, index);
	struct vdc_req_entry *rqe = &port->rq_arr[index];
	struct request *req;

	if (unlikely(desc->hdr.state != VIO_DESC_DONE))
		return;

	ldc_unmap(port->vio.lp, desc->cookies, desc->ncookies);
	desc->hdr.state = VIO_DESC_FREE;
	dr->cons = vio_dring_next(dr, index);

	req = rqe->req;
	if (req == NULL) {
		vdc_end_special(port, desc);
		return;
	}

	rqe->req = NULL;

	__blk_end_request(req, (desc->status ? -EIO : 0), desc->size);

	vdc_blk_queue_start(port);
}

static int vdc_ack(struct vdc_port *port, void *msgbuf)
{
	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];
	struct vio_dring_data *pkt = msgbuf;

	if (unlikely(pkt->dring_ident != dr->ident ||
		     pkt->start_idx != pkt->end_idx ||
		     pkt->start_idx >= VDC_TX_RING_SIZE))
		return 0;

	vdc_end_one(port, dr, pkt->start_idx);

	return 0;
}

static int vdc_nack(struct vdc_port *port, void *msgbuf)
{
	/* XXX Implement me XXX */
	return 0;
}

static void vdc_event(void *arg, int event)
{
	struct vdc_port *port = arg;
	struct vio_driver_state *vio = &port->vio;
	unsigned long flags;
	int err;

	spin_lock_irqsave(&vio->lock, flags);

	if (unlikely(event == LDC_EVENT_RESET)) {
		vio_link_state_change(vio, event);
		queue_work(sunvdc_wq, &port->ldc_reset_work);
		goto out;
	}

	if (unlikely(event == LDC_EVENT_UP)) {
		vio_link_state_change(vio, event);
		goto out;
	}

	if (unlikely(event != LDC_EVENT_DATA_READY)) {
		pr_warn(PFX "Unexpected LDC event %d\n", event);
		goto out;
	}

	err = 0;
	while (1) {
		union {
			struct vio_msg_tag tag;
			u64 raw[8];
		} msgbuf;

		err = ldc_read(vio->lp, &msgbuf, sizeof(msgbuf));
		if (unlikely(err < 0)) {
			if (err == -ECONNRESET)
				vio_conn_reset(vio);
			break;
		}
		if (err == 0)
			break;
		viodbg(DATA, "TAG [%02x:%02x:%04x:%08x]\n",
		       msgbuf.tag.type,
		       msgbuf.tag.stype,
		       msgbuf.tag.stype_env,
		       msgbuf.tag.sid);
		err = vio_validate_sid(vio, &msgbuf.tag);
		if (err < 0)
			break;

		if (likely(msgbuf.tag.type == VIO_TYPE_DATA)) {
			if (msgbuf.tag.stype == VIO_SUBTYPE_ACK)
				err = vdc_ack(port, &msgbuf);
			else if (msgbuf.tag.stype == VIO_SUBTYPE_NACK)
				err = vdc_nack(port, &msgbuf);
			else
				err = vdc_handle_unknown(port, &msgbuf);
		} else if (msgbuf.tag.type == VIO_TYPE_CTRL) {
			err = vio_control_pkt_engine(vio, &msgbuf);
		} else {
			err = vdc_handle_unknown(port, &msgbuf);
		}
		if (err < 0)
			break;
	}
	if (err < 0)
		vdc_finish(&port->vio, err, WAITING_FOR_ANY);
out:
	spin_unlock_irqrestore(&vio->lock, flags);
}

static int __vdc_tx_trigger(struct vdc_port *port)
{
	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];
	struct vio_dring_data hdr = {
		.tag = {
			.type		= VIO_TYPE_DATA,
			.stype		= VIO_SUBTYPE_INFO,
			.stype_env	= VIO_DRING_DATA,
			.sid		= vio_send_sid(&port->vio),
		},
		.dring_ident		= dr->ident,
		.start_idx		= dr->prod,
		.end_idx		= dr->prod,
	};
	int err, delay;
	int retries = 0;

	hdr.seq = dr->snd_nxt;
	delay = 1;
	do {
		err = vio_ldc_send(&port->vio, &hdr, sizeof(hdr));
		if (err > 0) {
			dr->snd_nxt++;
			break;
		}
		udelay(delay);
		if ((delay <<= 1) > 128)
			delay = 128;
		if (retries++ > VDC_MAX_RETRIES)
			break;
	} while (err == -EAGAIN);

	if (err == -ENOTCONN)
		vdc_ldc_reset(port);
	return err;
}

static int __send_request(struct request *req)
{
	struct vdc_port *port = req->rq_disk->private_data;
	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];
	struct scatterlist sg[port->ring_cookies];
	struct vdc_req_entry *rqe;
	struct vio_disk_desc *desc;
	unsigned int map_perm;
	int nsg, err, i;
	u64 len;
	u8 op;

	map_perm = LDC_MAP_SHADOW | LDC_MAP_DIRECT | LDC_MAP_IO;

	if (rq_data_dir(req) == READ) {
		map_perm |= LDC_MAP_W;
		op = VD_OP_BREAD;
	} else {
		map_perm |= LDC_MAP_R;
		op = VD_OP_BWRITE;
	}

	sg_init_table(sg, port->ring_cookies);
	nsg = blk_rq_map_sg(req->q, req, sg);

	len = 0;
	for (i = 0; i < nsg; i++)
		len += sg[i].length;

	desc = vio_dring_cur(dr);

	err = ldc_map_sg(port->vio.lp, sg, nsg,
			 desc->cookies, port->ring_cookies,
			 map_perm);
	if (err < 0) {
		printk(KERN_ERR PFX "ldc_map_sg() failure, err=%d.\n", err);
		return err;
	}

	rqe = &port->rq_arr[dr->prod];
	rqe->req = req;

	desc->hdr.ack = VIO_ACK_ENABLE;
	desc->req_id = port->req_id;
	desc->operation = op;
	if (port->vdisk_type == VD_DISK_TYPE_DISK) {
		desc->slice = 0xff;
	} else {
		desc->slice = 0;
	}
	desc->status = ~0;
	desc->offset = (blk_rq_pos(req) << 9) / port->vdisk_block_size;
	desc->size = len;
	desc->ncookies = err;

	/* This has to be a non-SMP write barrier because we are writing
	 * to memory which is shared with the peer LDOM.
	 */
	wmb();
	desc->hdr.state = VIO_DESC_READY;

	err = __vdc_tx_trigger(port);
	if (err < 0) {
		printk(KERN_ERR PFX "vdc_tx_trigger() failure, err=%d\n", err);
	} else {
		port->req_id++;
		dr->prod = vio_dring_next(dr, dr->prod);
	}

	return err;
}

static void do_vdc_request(struct request_queue *rq)
{
	struct request *req;

	while ((req = blk_peek_request(rq)) != NULL) {
		struct vdc_port *port;
		struct vio_dring_state *dr;

		port = req->rq_disk->private_data;
		dr = &port->vio.drings[VIO_DRIVER_TX_RING];
		if (unlikely(vdc_tx_dring_avail(dr) < 1))
			goto wait;

		blk_start_request(req);

		if (__send_request(req) < 0) {
			blk_requeue_request(rq, req);
wait:
			/* Avoid pointless unplugs. */
			blk_stop_queue(rq);
			break;
		}
	}
}

static int generic_request(struct vdc_port *port, u8 op, void *buf, int len)
{
	struct vio_dring_state *dr;
	struct vio_completion comp;
	struct vio_disk_desc *desc;
	unsigned int map_perm;
	unsigned long flags;
	int op_len, err;
	void *req_buf;

	if (!(((u64)1 << (u64)op) & port->operations))
		return -EOPNOTSUPP;

	switch (op) {
	case VD_OP_BREAD:
	case VD_OP_BWRITE:
	default:
		return -EINVAL;

	case VD_OP_FLUSH:
		op_len = 0;
		map_perm = 0;
		break;

	case VD_OP_GET_WCE:
		op_len = sizeof(u32);
		map_perm = LDC_MAP_W;
		break;

	case VD_OP_SET_WCE:
		op_len = sizeof(u32);
		map_perm = LDC_MAP_R;
		break;

	case VD_OP_GET_VTOC:
		op_len = sizeof(struct vio_disk_vtoc);
		map_perm = LDC_MAP_W;
		break;

	case VD_OP_SET_VTOC:
		op_len = sizeof(struct vio_disk_vtoc);
		map_perm = LDC_MAP_R;
		break;

	case VD_OP_GET_DISKGEOM:
		op_len = sizeof(struct vio_disk_geom);
		map_perm = LDC_MAP_W;
		break;

	case VD_OP_SET_DISKGEOM:
		op_len = sizeof(struct vio_disk_geom);
		map_perm = LDC_MAP_R;
		break;

	case VD_OP_SCSICMD:
		op_len = 16;
		map_perm = LDC_MAP_RW;
		break;

	case VD_OP_GET_DEVID:
		op_len = sizeof(struct vio_disk_devid);
		map_perm = LDC_MAP_W;
		break;

	case VD_OP_GET_EFI:
	case VD_OP_SET_EFI:
		return -EOPNOTSUPP;
		break;
	};

	map_perm |= LDC_MAP_SHADOW | LDC_MAP_DIRECT | LDC_MAP_IO;

	op_len = (op_len + 7) & ~7;
	req_buf = kzalloc(op_len, GFP_KERNEL);
	if (!req_buf)
		return -ENOMEM;

	if (len > op_len)
		len = op_len;

	if (map_perm & LDC_MAP_R)
		memcpy(req_buf, buf, len);

	spin_lock_irqsave(&port->vio.lock, flags);

	dr = &port->vio.drings[VIO_DRIVER_TX_RING];

	/* XXX If we want to use this code generically we have to
	 * XXX handle TX ring exhaustion etc.
	 */
	desc = vio_dring_cur(dr);

	err = ldc_map_single(port->vio.lp, req_buf, op_len,
			     desc->cookies, port->ring_cookies,
			     map_perm);
	if (err < 0) {
		spin_unlock_irqrestore(&port->vio.lock, flags);
		kfree(req_buf);
		return err;
	}

	init_completion(&comp.com);
	comp.waiting_for = WAITING_FOR_GEN_CMD;
	port->vio.cmp = &comp;

	desc->hdr.ack = VIO_ACK_ENABLE;
	desc->req_id = port->req_id;
	desc->operation = op;
	desc->slice = 0;
	desc->status = ~0;
	desc->offset = 0;
	desc->size = op_len;
	desc->ncookies = err;

	/* This has to be a non-SMP write barrier because we are writing
	 * to memory which is shared with the peer LDOM.
	 */
	wmb();
	desc->hdr.state = VIO_DESC_READY;

	err = __vdc_tx_trigger(port);
	if (err >= 0) {
		port->req_id++;
		dr->prod = vio_dring_next(dr, dr->prod);
		spin_unlock_irqrestore(&port->vio.lock, flags);

		wait_for_completion(&comp.com);
		err = comp.err;
	} else {
		port->vio.cmp = NULL;
		spin_unlock_irqrestore(&port->vio.lock, flags);
	}

	if (map_perm & LDC_MAP_W)
		memcpy(buf, req_buf, len);

	kfree(req_buf);

	return err;
}

static int vdc_alloc_tx_ring(struct vdc_port *port)
{
	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];
	unsigned long len, entry_size;
	int ncookies;
	void *dring;

	entry_size = sizeof(struct vio_disk_desc) +
		(sizeof(struct ldc_trans_cookie) * port->ring_cookies);
	len = (VDC_TX_RING_SIZE * entry_size);

	ncookies = VIO_MAX_RING_COOKIES;
	dring = ldc_alloc_exp_dring(port->vio.lp, len,
				    dr->cookies, &ncookies,
				    (LDC_MAP_SHADOW |
				     LDC_MAP_DIRECT |
				     LDC_MAP_RW));
	if (IS_ERR(dring))
		return PTR_ERR(dring);

	dr->base = dring;
	dr->entry_size = entry_size;
	dr->num_entries = VDC_TX_RING_SIZE;
	dr->prod = dr->cons = 0;
	dr->pending = VDC_TX_RING_SIZE;
	dr->ncookies = ncookies;

	return 0;
}

static void vdc_free_tx_ring(struct vdc_port *port)
{
	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];

	if (dr->base) {
		ldc_free_exp_dring(port->vio.lp, dr->base,
				   (dr->entry_size * dr->num_entries),
				   dr->cookies, dr->ncookies);
		dr->base = NULL;
		dr->entry_size = 0;
		dr->num_entries = 0;
		dr->pending = 0;
		dr->ncookies = 0;
	}
}

static int vdc_port_up(struct vdc_port *port)
{
	struct vio_completion comp;

	init_completion(&comp.com);
	comp.err = 0;
	comp.waiting_for = WAITING_FOR_LINK_UP;
	port->vio.cmp = &comp;

	vio_port_up(&port->vio);
	wait_for_completion(&comp.com);
	return comp.err;
}

static void vdc_port_down(struct vdc_port *port)
{
	ldc_disconnect(port->vio.lp);
	ldc_unbind(port->vio.lp);
	vdc_free_tx_ring(port);
	vio_ldc_free(&port->vio);
}

static int probe_disk(struct vdc_port *port)
{
	struct request_queue *q;
	struct gendisk *g;
	int err;

	err = vdc_port_up(port);
	if (err)
		return err;

	if (vdc_version_supported(port, 1, 1)) {
		/* vdisk_size should be set during the handshake, if it wasn't
		 * then the underlying disk is reserved by another system
		 */
		if (port->vdisk_size == -1)
			return -ENODEV;
	} else {
		struct vio_disk_geom geom;

		err = generic_request(port, VD_OP_GET_DISKGEOM,
				      &geom, sizeof(geom));
		if (err < 0) {
			printk(KERN_ERR PFX "VD_OP_GET_DISKGEOM returns "
			       "error %d\n", err);
			return err;
		}
		port->vdisk_size = ((u64)geom.num_cyl *
				    (u64)geom.num_hd *
				    (u64)geom.num_sec);
	}

	q = blk_init_queue(do_vdc_request, &port->vio.lock);
	if (!q) {
		printk(KERN_ERR PFX "%s: Could not allocate queue.\n",
		       port->vio.name);
		return -ENOMEM;
	}
	g = alloc_disk(1 << PARTITION_SHIFT);
	if (!g) {
		printk(KERN_ERR PFX "%s: Could not allocate gendisk.\n",
		       port->vio.name);
		blk_cleanup_queue(q);
		return -ENOMEM;
	}

	port->disk = g;

	/* Each segment in a request is up to an aligned page in size. */
	blk_queue_segment_boundary(q, PAGE_SIZE - 1);
	blk_queue_max_segment_size(q, PAGE_SIZE);

	blk_queue_max_segments(q, port->ring_cookies);
	blk_queue_max_hw_sectors(q, port->max_xfer_size);
	g->major = vdc_major;
	g->first_minor = port->vio.vdev->dev_no << PARTITION_SHIFT;
	strcpy(g->disk_name, port->disk_name);

	g->fops = &vdc_fops;
	g->queue = q;
	g->private_data = port;

	set_capacity(g, port->vdisk_size);

	if (vdc_version_supported(port, 1, 1)) {
		switch (port->vdisk_mtype) {
		case VD_MEDIA_TYPE_CD:
			pr_info(PFX "Virtual CDROM %s\n", port->disk_name);
			g->flags |= GENHD_FL_CD;
			g->flags |= GENHD_FL_REMOVABLE;
			set_disk_ro(g, 1);
			break;

		case VD_MEDIA_TYPE_DVD:
			pr_info(PFX "Virtual DVD %s\n", port->disk_name);
			g->flags |= GENHD_FL_CD;
			g->flags |= GENHD_FL_REMOVABLE;
			set_disk_ro(g, 1);
			break;

		case VD_MEDIA_TYPE_FIXED:
			pr_info(PFX "Virtual Hard disk %s\n", port->disk_name);
			break;
		}
	}

	pr_info(PFX "%s: %u sectors (%u MB) protocol %d.%d\n",
	       g->disk_name,
	       port->vdisk_size, (port->vdisk_size >> (20 - 9)),
	       port->vio.ver.major, port->vio.ver.minor);

	device_add_disk(&port->vio.vdev->dev, g);

	return 0;
}

static struct ldc_channel_config vdc_ldc_cfg = {
	.event		= vdc_event,
	.mtu		= 64,
	.mode		= LDC_MODE_UNRELIABLE,
};

static struct vio_driver_ops vdc_vio_ops = {
	.send_attr		= vdc_send_attr,
	.handle_attr		= vdc_handle_attr,
	.handshake_complete	= vdc_handshake_complete,
};

static void print_version(void)
{
	static int version_printed;

	if (version_printed++ == 0)
		printk(KERN_INFO "%s", version);
}

static int vdc_port_probe(struct vio_dev *vdev, const struct vio_device_id *id)
{
	struct mdesc_handle *hp;
	struct vdc_port *port;
	int err;
	const u64 *ldc_timeout;

	print_version();

	hp = mdesc_grab();

	err = -ENODEV;
	if ((vdev->dev_no << PARTITION_SHIFT) & ~(u64)MINORMASK) {
		printk(KERN_ERR PFX "Port id [%llu] too large.\n",
		       vdev->dev_no);
		goto err_out_release_mdesc;
	}

	port = kzalloc(sizeof(*port), GFP_KERNEL);
	err = -ENOMEM;
	if (!port) {
		printk(KERN_ERR PFX "Cannot allocate vdc_port.\n");
		goto err_out_release_mdesc;
	}

	if (vdev->dev_no >= 26)
		snprintf(port->disk_name, sizeof(port->disk_name),
			 VDCBLK_NAME "%c%c",
			 'a' + ((int)vdev->dev_no / 26) - 1,
			 'a' + ((int)vdev->dev_no % 26));
	else
		snprintf(port->disk_name, sizeof(port->disk_name),
			 VDCBLK_NAME "%c", 'a' + ((int)vdev->dev_no % 26));
	port->vdisk_size = -1;

	/* Actual wall time may be double due to do_generic_file_read() doing
	 * a readahead I/O first, and once that fails it will try to read a
	 * single page.
	 */
	ldc_timeout = mdesc_get_property(hp, vdev->mp, "vdc-timeout", NULL);
	port->ldc_timeout = ldc_timeout ? *ldc_timeout : 0;
	setup_timer(&port->ldc_reset_timer, vdc_ldc_reset_timer,
		    (unsigned long)port);
	INIT_WORK(&port->ldc_reset_work, vdc_ldc_reset_work);

	err = vio_driver_init(&port->vio, vdev, VDEV_DISK,
			      vdc_versions, ARRAY_SIZE(vdc_versions),
			      &vdc_vio_ops, port->disk_name);
	if (err)
		goto err_out_free_port;

	port->vdisk_block_size = 512;
	port->max_xfer_size = ((128 * 1024) / port->vdisk_block_size);
	port->ring_cookies = ((port->max_xfer_size *
			       port->vdisk_block_size) / PAGE_SIZE) + 2;

	err = vio_ldc_alloc(&port->vio, &vdc_ldc_cfg, port);
	if (err)
		goto err_out_free_port;

	err = vdc_alloc_tx_ring(port);
	if (err)
		goto err_out_free_ldc;

	err = probe_disk(port);
	if (err)
		goto err_out_free_tx_ring;

	dev_set_drvdata(&vdev->dev, port);

	mdesc_release(hp);

	return 0;

err_out_free_tx_ring:
	vdc_free_tx_ring(port);

err_out_free_ldc:
	vio_ldc_free(&port->vio);

err_out_free_port:
	kfree(port);

err_out_release_mdesc:
	mdesc_release(hp);
	return err;
}

static int vdc_port_remove(struct vio_dev *vdev)
{
	struct vdc_port *port = dev_get_drvdata(&vdev->dev);

	if (port) {
		unsigned long flags;

		spin_lock_irqsave(&port->vio.lock, flags);
		blk_stop_queue(port->disk->queue);
		spin_unlock_irqrestore(&port->vio.lock, flags);

		flush_work(&port->ldc_reset_work);
		del_timer_sync(&port->ldc_reset_timer);
		del_timer_sync(&port->vio.timer);

		del_gendisk(port->disk);
		blk_cleanup_queue(port->disk->queue);
		put_disk(port->disk);
		port->disk = NULL;

		vdc_free_tx_ring(port);
		vio_ldc_free(&port->vio);

		dev_set_drvdata(&vdev->dev, NULL);

		kfree(port);
	}
	return 0;
}

static void vdc_requeue_inflight(struct vdc_port *port)
{
	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];
	u32 idx;

	for (idx = dr->cons; idx != dr->prod; idx = vio_dring_next(dr, idx)) {
		struct vio_disk_desc *desc = vio_dring_entry(dr, idx);
		struct vdc_req_entry *rqe = &port->rq_arr[idx];
		struct request *req;

		ldc_unmap(port->vio.lp, desc->cookies, desc->ncookies);
		desc->hdr.state = VIO_DESC_FREE;
		dr->cons = vio_dring_next(dr, idx);

		req = rqe->req;
		if (req == NULL) {
			vdc_end_special(port, desc);
			continue;
		}

		rqe->req = NULL;
		blk_requeue_request(port->disk->queue, req);
	}
}

static void vdc_queue_drain(struct vdc_port *port)
{
	struct request *req;

	while ((req = blk_fetch_request(port->disk->queue)) != NULL)
		__blk_end_request_all(req, -EIO);
}

static void vdc_ldc_reset_timer(unsigned long _arg)
{
	struct vdc_port *port = (struct vdc_port *) _arg;
	struct vio_driver_state *vio = &port->vio;
	unsigned long flags;

	spin_lock_irqsave(&vio->lock, flags);
	if (!(port->vio.hs_state & VIO_HS_COMPLETE)) {
		pr_warn(PFX "%s ldc down %llu seconds, draining queue\n",
			port->disk_name, port->ldc_timeout);
		vdc_queue_drain(port);
		vdc_blk_queue_start(port);
	}
	spin_unlock_irqrestore(&vio->lock, flags);
}

static void vdc_ldc_reset_work(struct work_struct *work)
{
	struct vdc_port *port;
	struct vio_driver_state *vio;
	unsigned long flags;

	port = container_of(work, struct vdc_port, ldc_reset_work);
	vio = &port->vio;

	spin_lock_irqsave(&vio->lock, flags);
	vdc_ldc_reset(port);
	spin_unlock_irqrestore(&vio->lock, flags);
}

static void vdc_ldc_reset(struct vdc_port *port)
{
	int err;

	assert_spin_locked(&port->vio.lock);

	pr_warn(PFX "%s ldc link reset\n", port->disk_name);
	blk_stop_queue(port->disk->queue);
	vdc_requeue_inflight(port);
	vdc_port_down(port);

	err = vio_ldc_alloc(&port->vio, &vdc_ldc_cfg, port);
	if (err) {
		pr_err(PFX "%s vio_ldc_alloc:%d\n", port->disk_name, err);
		return;
	}

	err = vdc_alloc_tx_ring(port);
	if (err) {
		pr_err(PFX "%s vio_alloc_tx_ring:%d\n", port->disk_name, err);
		goto err_free_ldc;
	}

	if (port->ldc_timeout)
		mod_timer(&port->ldc_reset_timer,
			  round_jiffies(jiffies + HZ * port->ldc_timeout));
	mod_timer(&port->vio.timer, round_jiffies(jiffies + HZ));
	return;

err_free_ldc:
	vio_ldc_free(&port->vio);
}

static const struct vio_device_id vdc_port_match[] = {
	{
		.type = "vdc-port",
	},
	{},
};
MODULE_DEVICE_TABLE(vio, vdc_port_match);

static struct vio_driver vdc_port_driver = {
	.id_table	= vdc_port_match,
	.probe		= vdc_port_probe,
	.remove		= vdc_port_remove,
	.name		= "vdc_port",
};

static int __init vdc_init(void)
{
	int err;

	sunvdc_wq = alloc_workqueue("sunvdc", 0, 0);
	if (!sunvdc_wq)
		return -ENOMEM;

	err = register_blkdev(0, VDCBLK_NAME);
	if (err < 0)
		goto out_free_wq;

	vdc_major = err;

	err = vio_register_driver(&vdc_port_driver);
	if (err)
		goto out_unregister_blkdev;

	return 0;

out_unregister_blkdev:
	unregister_blkdev(vdc_major, VDCBLK_NAME);
	vdc_major = 0;

out_free_wq:
	destroy_workqueue(sunvdc_wq);
	return err;
}

static void __exit vdc_exit(void)
{
	vio_unregister_driver(&vdc_port_driver);
	unregister_blkdev(vdc_major, VDCBLK_NAME);
	destroy_workqueue(sunvdc_wq);
}

module_init(vdc_init);
module_exit(vdc_exit);