core.c 73.5 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
/*
 * NVM Express device driver
 * Copyright (c) 2011-2014, Intel Corporation.
 *
 * 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.
 */

#include <linux/blkdev.h>
#include <linux/blk-mq.h>
17
#include <linux/delay.h>
18
#include <linux/errno.h>
19
#include <linux/hdreg.h>
20
#include <linux/kernel.h>
21 22
#include <linux/module.h>
#include <linux/list_sort.h>
23 24
#include <linux/slab.h>
#include <linux/types.h>
25 26 27 28
#include <linux/pr.h>
#include <linux/ptrace.h>
#include <linux/nvme_ioctl.h>
#include <linux/t10-pi.h>
29
#include <linux/pm_qos.h>
30
#include <asm/unaligned.h>
31 32

#include "nvme.h"
S
Sagi Grimberg 已提交
33
#include "fabrics.h"
34

35 36
#define NVME_MINORS		(1U << MINORBITS)

37 38
unsigned int admin_timeout = 60;
module_param(admin_timeout, uint, 0644);
39
MODULE_PARM_DESC(admin_timeout, "timeout in seconds for admin commands");
40
EXPORT_SYMBOL_GPL(admin_timeout);
41

42 43
unsigned int nvme_io_timeout = 30;
module_param_named(io_timeout, nvme_io_timeout, uint, 0644);
44
MODULE_PARM_DESC(io_timeout, "timeout in seconds for I/O");
45
EXPORT_SYMBOL_GPL(nvme_io_timeout);
46

47
static unsigned char shutdown_timeout = 5;
48 49 50
module_param(shutdown_timeout, byte, 0644);
MODULE_PARM_DESC(shutdown_timeout, "timeout in seconds for controller shutdown");

51 52
static u8 nvme_max_retries = 5;
module_param_named(max_retries, nvme_max_retries, byte, 0644);
K
Keith Busch 已提交
53
MODULE_PARM_DESC(max_retries, "max number of retries a command may have");
54

55 56 57
static int nvme_char_major;
module_param(nvme_char_major, int, 0);

58
static unsigned long default_ps_max_latency_us = 100000;
59 60 61 62
module_param(default_ps_max_latency_us, ulong, 0644);
MODULE_PARM_DESC(default_ps_max_latency_us,
		 "max power saving latency for new devices; use PM QOS to change per device");

63 64 65 66
static bool force_apst;
module_param(force_apst, bool, 0644);
MODULE_PARM_DESC(force_apst, "allow APST for newly enumerated devices even if quirked off");

67 68 69 70
static bool streams;
module_param(streams, bool, 0644);
MODULE_PARM_DESC(streams, "turn on support for Streams write directives");

71 72 73
struct workqueue_struct *nvme_wq;
EXPORT_SYMBOL_GPL(nvme_wq);

74
static LIST_HEAD(nvme_ctrl_list);
M
Ming Lin 已提交
75
static DEFINE_SPINLOCK(dev_list_lock);
76

77 78
static DEFINE_IDA(nvme_instance_ida);

79 80
static struct class *nvme_class;

81 82 83 84 85
static __le32 nvme_get_log_dw10(u8 lid, size_t size)
{
	return cpu_to_le32((((size / 4) - 1) << 16) | lid);
}

86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105
int nvme_reset_ctrl(struct nvme_ctrl *ctrl)
{
	if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_RESETTING))
		return -EBUSY;
	if (!queue_work(nvme_wq, &ctrl->reset_work))
		return -EBUSY;
	return 0;
}
EXPORT_SYMBOL_GPL(nvme_reset_ctrl);

static int nvme_reset_ctrl_sync(struct nvme_ctrl *ctrl)
{
	int ret;

	ret = nvme_reset_ctrl(ctrl);
	if (!ret)
		flush_work(&ctrl->reset_work);
	return ret;
}

106
static blk_status_t nvme_error_status(struct request *req)
107 108 109
{
	switch (nvme_req(req)->status & 0x7ff) {
	case NVME_SC_SUCCESS:
110
		return BLK_STS_OK;
111
	case NVME_SC_CAP_EXCEEDED:
112
		return BLK_STS_NOSPC;
113
	case NVME_SC_ONCS_NOT_SUPPORTED:
114
		return BLK_STS_NOTSUPP;
115 116 117
	case NVME_SC_WRITE_FAULT:
	case NVME_SC_READ_ERROR:
	case NVME_SC_UNWRITTEN_BLOCK:
118 119
	case NVME_SC_ACCESS_DENIED:
	case NVME_SC_READ_ONLY:
120
		return BLK_STS_MEDIUM;
121 122 123 124 125 126 127
	case NVME_SC_GUARD_CHECK:
	case NVME_SC_APPTAG_CHECK:
	case NVME_SC_REFTAG_CHECK:
	case NVME_SC_INVALID_PI:
		return BLK_STS_PROTECTION;
	case NVME_SC_RESERVATION_CONFLICT:
		return BLK_STS_NEXUS;
128 129
	default:
		return BLK_STS_IOERR;
130 131 132
	}
}

133
static inline bool nvme_req_needs_retry(struct request *req)
134
{
135 136
	if (blk_noretry_request(req))
		return false;
137
	if (nvme_req(req)->status & NVME_SC_DNR)
138
		return false;
139
	if (nvme_req(req)->retries >= nvme_max_retries)
140 141
		return false;
	return true;
142 143 144 145
}

void nvme_complete_rq(struct request *req)
{
146 147
	if (unlikely(nvme_req(req)->status && nvme_req_needs_retry(req))) {
		nvme_req(req)->retries++;
148
		blk_mq_requeue_request(req, true);
149
		return;
150 151
	}

152
	blk_mq_end_request(req, nvme_error_status(req));
153 154 155
}
EXPORT_SYMBOL_GPL(nvme_complete_rq);

156 157 158 159 160 161 162 163 164 165 166 167 168
void nvme_cancel_request(struct request *req, void *data, bool reserved)
{
	int status;

	if (!blk_mq_request_started(req))
		return;

	dev_dbg_ratelimited(((struct nvme_ctrl *) data)->device,
				"Cancelling I/O %d", req->tag);

	status = NVME_SC_ABORT_REQ;
	if (blk_queue_dying(req->q))
		status |= NVME_SC_DNR;
169
	nvme_req(req)->status = status;
170
	blk_mq_complete_request(req);
171

172 173 174
}
EXPORT_SYMBOL_GPL(nvme_cancel_request);

175 176 177
bool nvme_change_ctrl_state(struct nvme_ctrl *ctrl,
		enum nvme_ctrl_state new_state)
{
178
	enum nvme_ctrl_state old_state;
179
	unsigned long flags;
180 181
	bool changed = false;

182
	spin_lock_irqsave(&ctrl->lock, flags);
183 184

	old_state = ctrl->state;
185 186 187
	switch (new_state) {
	case NVME_CTRL_LIVE:
		switch (old_state) {
188
		case NVME_CTRL_NEW:
189
		case NVME_CTRL_RESETTING:
190
		case NVME_CTRL_RECONNECTING:
191 192 193 194 195 196 197 198 199
			changed = true;
			/* FALLTHRU */
		default:
			break;
		}
		break;
	case NVME_CTRL_RESETTING:
		switch (old_state) {
		case NVME_CTRL_NEW:
200 201 202 203 204 205 206 207 208
		case NVME_CTRL_LIVE:
			changed = true;
			/* FALLTHRU */
		default:
			break;
		}
		break;
	case NVME_CTRL_RECONNECTING:
		switch (old_state) {
209 210 211 212 213 214 215 216 217 218 219
		case NVME_CTRL_LIVE:
			changed = true;
			/* FALLTHRU */
		default:
			break;
		}
		break;
	case NVME_CTRL_DELETING:
		switch (old_state) {
		case NVME_CTRL_LIVE:
		case NVME_CTRL_RESETTING:
220
		case NVME_CTRL_RECONNECTING:
221 222 223 224 225 226
			changed = true;
			/* FALLTHRU */
		default:
			break;
		}
		break;
227 228 229 230 231 232 233 234 235
	case NVME_CTRL_DEAD:
		switch (old_state) {
		case NVME_CTRL_DELETING:
			changed = true;
			/* FALLTHRU */
		default:
			break;
		}
		break;
236 237 238 239 240 241 242
	default:
		break;
	}

	if (changed)
		ctrl->state = new_state;

243
	spin_unlock_irqrestore(&ctrl->lock, flags);
244

245 246 247 248
	return changed;
}
EXPORT_SYMBOL_GPL(nvme_change_ctrl_state);

249 250 251 252
static void nvme_free_ns(struct kref *kref)
{
	struct nvme_ns *ns = container_of(kref, struct nvme_ns, kref);

253 254
	if (ns->ndev)
		nvme_nvm_unregister(ns);
255

256 257 258 259 260
	if (ns->disk) {
		spin_lock(&dev_list_lock);
		ns->disk->private_data = NULL;
		spin_unlock(&dev_list_lock);
	}
261 262

	put_disk(ns->disk);
263 264
	ida_simple_remove(&ns->ctrl->ns_ida, ns->instance);
	nvme_put_ctrl(ns->ctrl);
265 266 267
	kfree(ns);
}

268
static void nvme_put_ns(struct nvme_ns *ns)
269 270 271 272 273 274 275 276 277 278
{
	kref_put(&ns->kref, nvme_free_ns);
}

static struct nvme_ns *nvme_get_ns_from_disk(struct gendisk *disk)
{
	struct nvme_ns *ns;

	spin_lock(&dev_list_lock);
	ns = disk->private_data;
279 280 281 282 283 284
	if (ns) {
		if (!kref_get_unless_zero(&ns->kref))
			goto fail;
		if (!try_module_get(ns->ctrl->ops->module))
			goto fail_put_ns;
	}
285 286 287
	spin_unlock(&dev_list_lock);

	return ns;
288 289 290 291 292 293

fail_put_ns:
	kref_put(&ns->kref, nvme_free_ns);
fail:
	spin_unlock(&dev_list_lock);
	return NULL;
294 295
}

296
struct request *nvme_alloc_request(struct request_queue *q,
297
		struct nvme_command *cmd, unsigned int flags, int qid)
298
{
299
	unsigned op = nvme_is_write(cmd) ? REQ_OP_DRV_OUT : REQ_OP_DRV_IN;
300 301
	struct request *req;

302
	if (qid == NVME_QID_ANY) {
303
		req = blk_mq_alloc_request(q, op, flags);
304
	} else {
305
		req = blk_mq_alloc_request_hctx(q, op, flags,
306 307
				qid ? qid - 1 : 0);
	}
308
	if (IS_ERR(req))
309
		return req;
310 311

	req->cmd_flags |= REQ_FAILFAST_DRIVER;
312
	nvme_req(req)->cmd = cmd;
313

314 315
	return req;
}
316
EXPORT_SYMBOL_GPL(nvme_alloc_request);
317

318 319 320 321 322 323 324
static int nvme_toggle_streams(struct nvme_ctrl *ctrl, bool enable)
{
	struct nvme_command c;

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

	c.directive.opcode = nvme_admin_directive_send;
A
Arnav Dawn 已提交
325
	c.directive.nsid = cpu_to_le32(NVME_NSID_ALL);
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
	c.directive.doper = NVME_DIR_SND_ID_OP_ENABLE;
	c.directive.dtype = NVME_DIR_IDENTIFY;
	c.directive.tdtype = NVME_DIR_STREAMS;
	c.directive.endir = enable ? NVME_DIR_ENDIR : 0;

	return nvme_submit_sync_cmd(ctrl->admin_q, &c, NULL, 0);
}

static int nvme_disable_streams(struct nvme_ctrl *ctrl)
{
	return nvme_toggle_streams(ctrl, false);
}

static int nvme_enable_streams(struct nvme_ctrl *ctrl)
{
	return nvme_toggle_streams(ctrl, true);
}

static int nvme_get_stream_params(struct nvme_ctrl *ctrl,
				  struct streams_directive_params *s, u32 nsid)
{
	struct nvme_command c;

	memset(&c, 0, sizeof(c));
	memset(s, 0, sizeof(*s));

	c.directive.opcode = nvme_admin_directive_recv;
	c.directive.nsid = cpu_to_le32(nsid);
354
	c.directive.numd = cpu_to_le32((sizeof(*s) >> 2) - 1);
355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374
	c.directive.doper = NVME_DIR_RCV_ST_OP_PARAM;
	c.directive.dtype = NVME_DIR_STREAMS;

	return nvme_submit_sync_cmd(ctrl->admin_q, &c, s, sizeof(*s));
}

static int nvme_configure_directives(struct nvme_ctrl *ctrl)
{
	struct streams_directive_params s;
	int ret;

	if (!(ctrl->oacs & NVME_CTRL_OACS_DIRECTIVES))
		return 0;
	if (!streams)
		return 0;

	ret = nvme_enable_streams(ctrl);
	if (ret)
		return ret;

A
Arnav Dawn 已提交
375
	ret = nvme_get_stream_params(ctrl, &s, NVME_NSID_ALL);
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
	if (ret)
		return ret;

	ctrl->nssa = le16_to_cpu(s.nssa);
	if (ctrl->nssa < BLK_MAX_WRITE_HINTS - 1) {
		dev_info(ctrl->device, "too few streams (%u) available\n",
					ctrl->nssa);
		nvme_disable_streams(ctrl);
		return 0;
	}

	ctrl->nr_streams = min_t(unsigned, ctrl->nssa, BLK_MAX_WRITE_HINTS - 1);
	dev_info(ctrl->device, "Using %u streams\n", ctrl->nr_streams);
	return 0;
}

/*
 * Check if 'req' has a write hint associated with it. If it does, assign
 * a valid namespace stream to the write.
 */
static void nvme_assign_write_stream(struct nvme_ctrl *ctrl,
				     struct request *req, u16 *control,
				     u32 *dsmgmt)
{
	enum rw_hint streamid = req->write_hint;

	if (streamid == WRITE_LIFE_NOT_SET || streamid == WRITE_LIFE_NONE)
		streamid = 0;
	else {
		streamid--;
		if (WARN_ON_ONCE(streamid > ctrl->nr_streams))
			return;

		*control |= NVME_RW_DTYPE_STREAMS;
		*dsmgmt |= streamid << 16;
	}

	if (streamid < ARRAY_SIZE(req->q->write_hints))
		req->q->write_hints[streamid] += blk_rq_bytes(req) >> 9;
}

M
Ming Lin 已提交
417 418 419 420 421 422 423 424
static inline void nvme_setup_flush(struct nvme_ns *ns,
		struct nvme_command *cmnd)
{
	memset(cmnd, 0, sizeof(*cmnd));
	cmnd->common.opcode = nvme_cmd_flush;
	cmnd->common.nsid = cpu_to_le32(ns->ns_id);
}

425
static blk_status_t nvme_setup_discard(struct nvme_ns *ns, struct request *req,
M
Ming Lin 已提交
426 427
		struct nvme_command *cmnd)
{
428
	unsigned short segments = blk_rq_nr_discard_segments(req), n = 0;
M
Ming Lin 已提交
429
	struct nvme_dsm_range *range;
430
	struct bio *bio;
M
Ming Lin 已提交
431

432
	range = kmalloc_array(segments, sizeof(*range), GFP_ATOMIC);
M
Ming Lin 已提交
433
	if (!range)
434
		return BLK_STS_RESOURCE;
M
Ming Lin 已提交
435

436 437 438 439 440 441 442 443 444 445 446 447
	__rq_for_each_bio(bio, req) {
		u64 slba = nvme_block_nr(ns, bio->bi_iter.bi_sector);
		u32 nlb = bio->bi_iter.bi_size >> ns->lba_shift;

		range[n].cattr = cpu_to_le32(0);
		range[n].nlb = cpu_to_le32(nlb);
		range[n].slba = cpu_to_le64(slba);
		n++;
	}

	if (WARN_ON_ONCE(n != segments)) {
		kfree(range);
448
		return BLK_STS_IOERR;
449
	}
M
Ming Lin 已提交
450 451 452 453

	memset(cmnd, 0, sizeof(*cmnd));
	cmnd->dsm.opcode = nvme_cmd_dsm;
	cmnd->dsm.nsid = cpu_to_le32(ns->ns_id);
454
	cmnd->dsm.nr = cpu_to_le32(segments - 1);
M
Ming Lin 已提交
455 456
	cmnd->dsm.attributes = cpu_to_le32(NVME_DSMGMT_AD);

457 458
	req->special_vec.bv_page = virt_to_page(range);
	req->special_vec.bv_offset = offset_in_page(range);
459
	req->special_vec.bv_len = sizeof(*range) * segments;
460
	req->rq_flags |= RQF_SPECIAL_PAYLOAD;
M
Ming Lin 已提交
461

462
	return BLK_STS_OK;
M
Ming Lin 已提交
463 464
}

465 466
static inline blk_status_t nvme_setup_rw(struct nvme_ns *ns,
		struct request *req, struct nvme_command *cmnd)
M
Ming Lin 已提交
467
{
468
	struct nvme_ctrl *ctrl = ns->ctrl;
M
Ming Lin 已提交
469 470 471
	u16 control = 0;
	u32 dsmgmt = 0;

472 473 474 475 476
	/*
	 * If formated with metadata, require the block layer provide a buffer
	 * unless this namespace is formated such that the metadata can be
	 * stripped/generated by the controller with PRACT=1.
	 */
477 478
	if (ns && ns->ms &&
	    (!ns->pi_type || ns->ms != sizeof(struct t10_pi_tuple)) &&
479 480 481
	    !blk_integrity_rq(req) && !blk_rq_is_passthrough(req))
		return BLK_STS_NOTSUPP;

M
Ming Lin 已提交
482 483 484 485 486 487 488 489 490 491 492 493 494 495
	if (req->cmd_flags & REQ_FUA)
		control |= NVME_RW_FUA;
	if (req->cmd_flags & (REQ_FAILFAST_DEV | REQ_RAHEAD))
		control |= NVME_RW_LR;

	if (req->cmd_flags & REQ_RAHEAD)
		dsmgmt |= NVME_RW_DSM_FREQ_PREFETCH;

	memset(cmnd, 0, sizeof(*cmnd));
	cmnd->rw.opcode = (rq_data_dir(req) ? nvme_cmd_write : nvme_cmd_read);
	cmnd->rw.nsid = cpu_to_le32(ns->ns_id);
	cmnd->rw.slba = cpu_to_le64(nvme_block_nr(ns, blk_rq_pos(req)));
	cmnd->rw.length = cpu_to_le16((blk_rq_bytes(req) >> ns->lba_shift) - 1);

496 497 498
	if (req_op(req) == REQ_OP_WRITE && ctrl->nr_streams)
		nvme_assign_write_stream(ctrl, req, &control, &dsmgmt);

M
Ming Lin 已提交
499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517
	if (ns->ms) {
		switch (ns->pi_type) {
		case NVME_NS_DPS_PI_TYPE3:
			control |= NVME_RW_PRINFO_PRCHK_GUARD;
			break;
		case NVME_NS_DPS_PI_TYPE1:
		case NVME_NS_DPS_PI_TYPE2:
			control |= NVME_RW_PRINFO_PRCHK_GUARD |
					NVME_RW_PRINFO_PRCHK_REF;
			cmnd->rw.reftag = cpu_to_le32(
					nvme_block_nr(ns, blk_rq_pos(req)));
			break;
		}
		if (!blk_integrity_rq(req))
			control |= NVME_RW_PRINFO_PRACT;
	}

	cmnd->rw.control = cpu_to_le16(control);
	cmnd->rw.dsmgmt = cpu_to_le32(dsmgmt);
518
	return 0;
M
Ming Lin 已提交
519 520
}

521
blk_status_t nvme_setup_cmd(struct nvme_ns *ns, struct request *req,
M
Ming Lin 已提交
522 523
		struct nvme_command *cmd)
{
524
	blk_status_t ret = BLK_STS_OK;
M
Ming Lin 已提交
525

526
	if (!(req->rq_flags & RQF_DONTPREP)) {
527
		nvme_req(req)->retries = 0;
528
		nvme_req(req)->flags = 0;
529 530 531
		req->rq_flags |= RQF_DONTPREP;
	}

532 533 534
	switch (req_op(req)) {
	case REQ_OP_DRV_IN:
	case REQ_OP_DRV_OUT:
535
		memcpy(cmd, nvme_req(req)->cmd, sizeof(*cmd));
536 537
		break;
	case REQ_OP_FLUSH:
M
Ming Lin 已提交
538
		nvme_setup_flush(ns, cmd);
539
		break;
540 541
	case REQ_OP_WRITE_ZEROES:
		/* currently only aliased to deallocate for a few ctrls: */
542
	case REQ_OP_DISCARD:
M
Ming Lin 已提交
543
		ret = nvme_setup_discard(ns, req, cmd);
544 545 546
		break;
	case REQ_OP_READ:
	case REQ_OP_WRITE:
547
		ret = nvme_setup_rw(ns, req, cmd);
548 549 550
		break;
	default:
		WARN_ON_ONCE(1);
551
		return BLK_STS_IOERR;
552
	}
M
Ming Lin 已提交
553

554
	cmd->common.command_id = req->tag;
M
Ming Lin 已提交
555 556 557 558
	return ret;
}
EXPORT_SYMBOL_GPL(nvme_setup_cmd);

559 560 561 562 563
/*
 * Returns 0 on success.  If the result is negative, it's a Linux error code;
 * if the result is positive, it's an NVM Express status code
 */
int __nvme_submit_sync_cmd(struct request_queue *q, struct nvme_command *cmd,
564
		union nvme_result *result, void *buffer, unsigned bufflen,
565
		unsigned timeout, int qid, int at_head, int flags)
566 567 568 569
{
	struct request *req;
	int ret;

570
	req = nvme_alloc_request(q, cmd, flags, qid);
571 572 573 574 575
	if (IS_ERR(req))
		return PTR_ERR(req);

	req->timeout = timeout ? timeout : ADMIN_TIMEOUT;

576 577 578 579
	if (buffer && bufflen) {
		ret = blk_rq_map_kern(q, req, buffer, bufflen, GFP_KERNEL);
		if (ret)
			goto out;
580 581
	}

582
	blk_execute_rq(req->q, NULL, req, at_head);
583 584
	if (result)
		*result = nvme_req(req)->result;
585 586 587 588
	if (nvme_req(req)->flags & NVME_REQ_CANCELLED)
		ret = -EINTR;
	else
		ret = nvme_req(req)->status;
589 590 591 592
 out:
	blk_mq_free_request(req);
	return ret;
}
593
EXPORT_SYMBOL_GPL(__nvme_submit_sync_cmd);
594 595 596 597

int nvme_submit_sync_cmd(struct request_queue *q, struct nvme_command *cmd,
		void *buffer, unsigned bufflen)
{
598 599
	return __nvme_submit_sync_cmd(q, cmd, NULL, buffer, bufflen, 0,
			NVME_QID_ANY, 0, 0);
600
}
601
EXPORT_SYMBOL_GPL(nvme_submit_sync_cmd);
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
static void *nvme_add_user_metadata(struct bio *bio, void __user *ubuf,
		unsigned len, u32 seed, bool write)
{
	struct bio_integrity_payload *bip;
	int ret = -ENOMEM;
	void *buf;

	buf = kmalloc(len, GFP_KERNEL);
	if (!buf)
		goto out;

	ret = -EFAULT;
	if (write && copy_from_user(buf, ubuf, len))
		goto out_free_meta;

	bip = bio_integrity_alloc(bio, GFP_KERNEL, 1);
	if (IS_ERR(bip)) {
		ret = PTR_ERR(bip);
		goto out_free_meta;
	}

	bip->bip_iter.bi_size = len;
	bip->bip_iter.bi_sector = seed;
	ret = bio_integrity_add_page(bio, virt_to_page(buf), len,
			offset_in_page(buf));
	if (ret == len)
		return buf;
	ret = -ENOMEM;
out_free_meta:
	kfree(buf);
out:
	return ERR_PTR(ret);
}

637
static int nvme_submit_user_cmd(struct request_queue *q,
638 639 640
		struct nvme_command *cmd, void __user *ubuffer,
		unsigned bufflen, void __user *meta_buffer, unsigned meta_len,
		u32 meta_seed, u32 *result, unsigned timeout)
641
{
642
	bool write = nvme_is_write(cmd);
643 644
	struct nvme_ns *ns = q->queuedata;
	struct gendisk *disk = ns ? ns->disk : NULL;
645
	struct request *req;
646 647
	struct bio *bio = NULL;
	void *meta = NULL;
648 649
	int ret;

650
	req = nvme_alloc_request(q, cmd, 0, NVME_QID_ANY);
651 652 653 654 655 656
	if (IS_ERR(req))
		return PTR_ERR(req);

	req->timeout = timeout ? timeout : ADMIN_TIMEOUT;

	if (ubuffer && bufflen) {
657 658 659 660 661
		ret = blk_rq_map_user(q, req, NULL, ubuffer, bufflen,
				GFP_KERNEL);
		if (ret)
			goto out;
		bio = req->bio;
662
		bio->bi_disk = disk;
663 664 665 666 667
		if (disk && meta_buffer && meta_len) {
			meta = nvme_add_user_metadata(bio, meta_buffer, meta_len,
					meta_seed, write);
			if (IS_ERR(meta)) {
				ret = PTR_ERR(meta);
668 669 670 671
				goto out_unmap;
			}
		}
	}
672

673
	blk_execute_rq(req->q, disk, req, 0);
674 675 676 677
	if (nvme_req(req)->flags & NVME_REQ_CANCELLED)
		ret = -EINTR;
	else
		ret = nvme_req(req)->status;
678
	if (result)
679
		*result = le32_to_cpu(nvme_req(req)->result.u32);
680 681 682 683 684 685
	if (meta && !ret && !write) {
		if (copy_to_user(meta_buffer, meta, meta_len))
			ret = -EFAULT;
	}
	kfree(meta);
 out_unmap:
686
	if (bio)
687
		blk_rq_unmap_user(bio);
688 689 690 691 692
 out:
	blk_mq_free_request(req);
	return ret;
}

693
static void nvme_keep_alive_end_io(struct request *rq, blk_status_t status)
S
Sagi Grimberg 已提交
694 695 696 697 698
{
	struct nvme_ctrl *ctrl = rq->end_io_data;

	blk_mq_free_request(rq);

699
	if (status) {
S
Sagi Grimberg 已提交
700
		dev_err(ctrl->device,
701 702
			"failed nvme_keep_alive_end_io error=%d\n",
				status);
S
Sagi Grimberg 已提交
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
		return;
	}

	schedule_delayed_work(&ctrl->ka_work, ctrl->kato * HZ);
}

static int nvme_keep_alive(struct nvme_ctrl *ctrl)
{
	struct nvme_command c;
	struct request *rq;

	memset(&c, 0, sizeof(c));
	c.common.opcode = nvme_admin_keep_alive;

	rq = nvme_alloc_request(ctrl->admin_q, &c, BLK_MQ_REQ_RESERVED,
			NVME_QID_ANY);
	if (IS_ERR(rq))
		return PTR_ERR(rq);

	rq->timeout = ctrl->kato * HZ;
	rq->end_io_data = ctrl;

	blk_execute_rq_nowait(rq->q, NULL, rq, 0, nvme_keep_alive_end_io);

	return 0;
}

static void nvme_keep_alive_work(struct work_struct *work)
{
	struct nvme_ctrl *ctrl = container_of(to_delayed_work(work),
			struct nvme_ctrl, ka_work);

	if (nvme_keep_alive(ctrl)) {
		/* allocation failure, reset the controller */
		dev_err(ctrl->device, "keep-alive failed\n");
738
		nvme_reset_ctrl(ctrl);
S
Sagi Grimberg 已提交
739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761
		return;
	}
}

void nvme_start_keep_alive(struct nvme_ctrl *ctrl)
{
	if (unlikely(ctrl->kato == 0))
		return;

	INIT_DELAYED_WORK(&ctrl->ka_work, nvme_keep_alive_work);
	schedule_delayed_work(&ctrl->ka_work, ctrl->kato * HZ);
}
EXPORT_SYMBOL_GPL(nvme_start_keep_alive);

void nvme_stop_keep_alive(struct nvme_ctrl *ctrl)
{
	if (unlikely(ctrl->kato == 0))
		return;

	cancel_delayed_work_sync(&ctrl->ka_work);
}
EXPORT_SYMBOL_GPL(nvme_stop_keep_alive);

K
Keith Busch 已提交
762
static int nvme_identify_ctrl(struct nvme_ctrl *dev, struct nvme_id_ctrl **id)
763 764 765 766 767 768
{
	struct nvme_command c = { };
	int error;

	/* gcc-4.4.4 (at least) has issues with initializers and anon unions */
	c.identify.opcode = nvme_admin_identify;
769
	c.identify.cns = NVME_ID_CNS_CTRL;
770 771 772 773 774 775 776 777 778 779 780 781

	*id = kmalloc(sizeof(struct nvme_id_ctrl), GFP_KERNEL);
	if (!*id)
		return -ENOMEM;

	error = nvme_submit_sync_cmd(dev->admin_q, &c, *id,
			sizeof(struct nvme_id_ctrl));
	if (error)
		kfree(*id);
	return error;
}

782 783
static int nvme_identify_ns_descs(struct nvme_ctrl *ctrl, unsigned nsid,
		u8 *eui64, u8 *nguid, uuid_t *uuid)
784 785 786 787 788 789 790 791 792 793 794 795 796 797 798
{
	struct nvme_command c = { };
	int status;
	void *data;
	int pos;
	int len;

	c.identify.opcode = nvme_admin_identify;
	c.identify.nsid = cpu_to_le32(nsid);
	c.identify.cns = NVME_ID_CNS_NS_DESC_LIST;

	data = kzalloc(NVME_IDENTIFY_DATA_SIZE, GFP_KERNEL);
	if (!data)
		return -ENOMEM;

799
	status = nvme_submit_sync_cmd(ctrl->admin_q, &c, data,
800 801 802 803 804 805 806 807 808 809 810 811 812
				      NVME_IDENTIFY_DATA_SIZE);
	if (status)
		goto free_data;

	for (pos = 0; pos < NVME_IDENTIFY_DATA_SIZE; pos += len) {
		struct nvme_ns_id_desc *cur = data + pos;

		if (cur->nidl == 0)
			break;

		switch (cur->nidt) {
		case NVME_NIDT_EUI64:
			if (cur->nidl != NVME_NIDT_EUI64_LEN) {
813
				dev_warn(ctrl->device,
814 815 816 817 818
					 "ctrl returned bogus length: %d for NVME_NIDT_EUI64\n",
					 cur->nidl);
				goto free_data;
			}
			len = NVME_NIDT_EUI64_LEN;
819
			memcpy(eui64, data + pos + sizeof(*cur), len);
820 821 822
			break;
		case NVME_NIDT_NGUID:
			if (cur->nidl != NVME_NIDT_NGUID_LEN) {
823
				dev_warn(ctrl->device,
824 825 826 827 828
					 "ctrl returned bogus length: %d for NVME_NIDT_NGUID\n",
					 cur->nidl);
				goto free_data;
			}
			len = NVME_NIDT_NGUID_LEN;
829
			memcpy(nguid, data + pos + sizeof(*cur), len);
830 831 832
			break;
		case NVME_NIDT_UUID:
			if (cur->nidl != NVME_NIDT_UUID_LEN) {
833
				dev_warn(ctrl->device,
834 835 836 837 838
					 "ctrl returned bogus length: %d for NVME_NIDT_UUID\n",
					 cur->nidl);
				goto free_data;
			}
			len = NVME_NIDT_UUID_LEN;
839
			uuid_copy(uuid, data + pos + sizeof(*cur));
840 841 842 843 844 845 846 847 848 849 850 851 852 853
			break;
		default:
			/* Skip unnkown types */
			len = cur->nidl;
			break;
		}

		len += sizeof(*cur);
	}
free_data:
	kfree(data);
	return status;
}

854 855 856 857 858
static int nvme_identify_ns_list(struct nvme_ctrl *dev, unsigned nsid, __le32 *ns_list)
{
	struct nvme_command c = { };

	c.identify.opcode = nvme_admin_identify;
859
	c.identify.cns = NVME_ID_CNS_NS_ACTIVE_LIST;
860 861 862 863
	c.identify.nsid = cpu_to_le32(nsid);
	return nvme_submit_sync_cmd(dev->admin_q, &c, ns_list, 0x1000);
}

864 865
static struct nvme_id_ns *nvme_identify_ns(struct nvme_ctrl *ctrl,
		unsigned nsid)
866
{
867
	struct nvme_id_ns *id;
868 869 870 871
	struct nvme_command c = { };
	int error;

	/* gcc-4.4.4 (at least) has issues with initializers and anon unions */
872 873
	c.identify.opcode = nvme_admin_identify;
	c.identify.nsid = cpu_to_le32(nsid);
874
	c.identify.cns = NVME_ID_CNS_NS;
875

876 877 878
	id = kmalloc(sizeof(*id), GFP_KERNEL);
	if (!id)
		return NULL;
879

880 881 882 883 884 885 886 887
	error = nvme_submit_sync_cmd(ctrl->admin_q, &c, id, sizeof(*id));
	if (error) {
		dev_warn(ctrl->device, "Identify namespace failed\n");
		kfree(id);
		return NULL;
	}

	return id;
888 889
}

K
Keith Busch 已提交
890
static int nvme_set_features(struct nvme_ctrl *dev, unsigned fid, unsigned dword11,
891
		      void *buffer, size_t buflen, u32 *result)
892 893
{
	struct nvme_command c;
894
	union nvme_result res;
895
	int ret;
896 897 898 899 900 901

	memset(&c, 0, sizeof(c));
	c.features.opcode = nvme_admin_set_features;
	c.features.fid = cpu_to_le32(fid);
	c.features.dword11 = cpu_to_le32(dword11);

902
	ret = __nvme_submit_sync_cmd(dev->admin_q, &c, &res,
903
			buffer, buflen, 0, NVME_QID_ANY, 0, 0);
904
	if (ret >= 0 && result)
905
		*result = le32_to_cpu(res.u32);
906
	return ret;
907 908
}

C
Christoph Hellwig 已提交
909 910 911 912 913 914
int nvme_set_queue_count(struct nvme_ctrl *ctrl, int *count)
{
	u32 q_count = (*count - 1) | ((*count - 1) << 16);
	u32 result;
	int status, nr_io_queues;

915
	status = nvme_set_features(ctrl, NVME_FEAT_NUM_QUEUES, q_count, NULL, 0,
C
Christoph Hellwig 已提交
916
			&result);
917
	if (status < 0)
C
Christoph Hellwig 已提交
918 919
		return status;

920 921 922 923 924 925
	/*
	 * Degraded controllers might return an error when setting the queue
	 * count.  We still want to be able to bring them online and offer
	 * access to the admin queue, as that might be only way to fix them up.
	 */
	if (status > 0) {
926
		dev_err(ctrl->device, "Could not set queue count (%d)\n", status);
927 928 929 930 931 932
		*count = 0;
	} else {
		nr_io_queues = min(result & 0xffff, result >> 16) + 1;
		*count = min(*count, nr_io_queues);
	}

C
Christoph Hellwig 已提交
933 934
	return 0;
}
935
EXPORT_SYMBOL_GPL(nvme_set_queue_count);
C
Christoph Hellwig 已提交
936

937 938 939 940 941 942 943 944 945
static int nvme_submit_io(struct nvme_ns *ns, struct nvme_user_io __user *uio)
{
	struct nvme_user_io io;
	struct nvme_command c;
	unsigned length, meta_len;
	void __user *metadata;

	if (copy_from_user(&io, uio, sizeof(io)))
		return -EFAULT;
946 947
	if (io.flags)
		return -EINVAL;
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

	switch (io.opcode) {
	case nvme_cmd_write:
	case nvme_cmd_read:
	case nvme_cmd_compare:
		break;
	default:
		return -EINVAL;
	}

	length = (io.nblocks + 1) << ns->lba_shift;
	meta_len = (io.nblocks + 1) * ns->ms;
	metadata = (void __user *)(uintptr_t)io.metadata;

	if (ns->ext) {
		length += meta_len;
		meta_len = 0;
	} else if (meta_len) {
		if ((io.metadata & 3) || !io.metadata)
			return -EINVAL;
	}

	memset(&c, 0, sizeof(c));
	c.rw.opcode = io.opcode;
	c.rw.flags = io.flags;
	c.rw.nsid = cpu_to_le32(ns->ns_id);
	c.rw.slba = cpu_to_le64(io.slba);
	c.rw.length = cpu_to_le16(io.nblocks);
	c.rw.control = cpu_to_le16(io.control);
	c.rw.dsmgmt = cpu_to_le32(io.dsmgmt);
	c.rw.reftag = cpu_to_le32(io.reftag);
	c.rw.apptag = cpu_to_le16(io.apptag);
	c.rw.appmask = cpu_to_le16(io.appmask);

982
	return nvme_submit_user_cmd(ns->queue, &c,
983 984 985 986
			(void __user *)(uintptr_t)io.addr, length,
			metadata, meta_len, io.slba, NULL, 0);
}

987
static int nvme_user_cmd(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
988 989 990 991 992 993 994 995 996 997 998
			struct nvme_passthru_cmd __user *ucmd)
{
	struct nvme_passthru_cmd cmd;
	struct nvme_command c;
	unsigned timeout = 0;
	int status;

	if (!capable(CAP_SYS_ADMIN))
		return -EACCES;
	if (copy_from_user(&cmd, ucmd, sizeof(cmd)))
		return -EFAULT;
999 1000
	if (cmd.flags)
		return -EINVAL;
1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018

	memset(&c, 0, sizeof(c));
	c.common.opcode = cmd.opcode;
	c.common.flags = cmd.flags;
	c.common.nsid = cpu_to_le32(cmd.nsid);
	c.common.cdw2[0] = cpu_to_le32(cmd.cdw2);
	c.common.cdw2[1] = cpu_to_le32(cmd.cdw3);
	c.common.cdw10[0] = cpu_to_le32(cmd.cdw10);
	c.common.cdw10[1] = cpu_to_le32(cmd.cdw11);
	c.common.cdw10[2] = cpu_to_le32(cmd.cdw12);
	c.common.cdw10[3] = cpu_to_le32(cmd.cdw13);
	c.common.cdw10[4] = cpu_to_le32(cmd.cdw14);
	c.common.cdw10[5] = cpu_to_le32(cmd.cdw15);

	if (cmd.timeout_ms)
		timeout = msecs_to_jiffies(cmd.timeout_ms);

	status = nvme_submit_user_cmd(ns ? ns->queue : ctrl->admin_q, &c,
1019
			(void __user *)(uintptr_t)cmd.addr, cmd.data_len,
1020 1021
			(void __user *)(uintptr_t)cmd.metadata, cmd.metadata,
			0, &cmd.result, timeout);
1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045
	if (status >= 0) {
		if (put_user(cmd.result, &ucmd->result))
			return -EFAULT;
	}

	return status;
}

static int nvme_ioctl(struct block_device *bdev, fmode_t mode,
		unsigned int cmd, unsigned long arg)
{
	struct nvme_ns *ns = bdev->bd_disk->private_data;

	switch (cmd) {
	case NVME_IOCTL_ID:
		force_successful_syscall_return();
		return ns->ns_id;
	case NVME_IOCTL_ADMIN_CMD:
		return nvme_user_cmd(ns->ctrl, NULL, (void __user *)arg);
	case NVME_IOCTL_IO_CMD:
		return nvme_user_cmd(ns->ctrl, ns, (void __user *)arg);
	case NVME_IOCTL_SUBMIT_IO:
		return nvme_submit_io(ns, (void __user *)arg);
	default:
1046 1047 1048 1049
#ifdef CONFIG_NVM
		if (ns->ndev)
			return nvme_nvm_ioctl(ns, cmd, arg);
#endif
1050
		if (is_sed_ioctl(cmd))
1051
			return sed_ioctl(ns->ctrl->opal_dev, cmd,
1052
					 (void __user *) arg);
1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063
		return -ENOTTY;
	}
}

static int nvme_open(struct block_device *bdev, fmode_t mode)
{
	return nvme_get_ns_from_disk(bdev->bd_disk) ? 0 : -ENXIO;
}

static void nvme_release(struct gendisk *disk, fmode_t mode)
{
1064 1065 1066 1067
	struct nvme_ns *ns = disk->private_data;

	module_put(ns->ctrl->ops->module);
	nvme_put_ns(ns);
1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079
}

static int nvme_getgeo(struct block_device *bdev, struct hd_geometry *geo)
{
	/* some standard values */
	geo->heads = 1 << 6;
	geo->sectors = 1 << 5;
	geo->cylinders = get_capacity(bdev->bd_disk) >> 11;
	return 0;
}

#ifdef CONFIG_BLK_DEV_INTEGRITY
1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102
static void nvme_prep_integrity(struct gendisk *disk, struct nvme_id_ns *id,
		u16 bs)
{
	struct nvme_ns *ns = disk->private_data;
	u16 old_ms = ns->ms;
	u8 pi_type = 0;

	ns->ms = le16_to_cpu(id->lbaf[id->flbas & NVME_NS_FLBAS_LBA_MASK].ms);
	ns->ext = ns->ms && (id->flbas & NVME_NS_FLBAS_META_EXT);

	/* PI implementation requires metadata equal t10 pi tuple size */
	if (ns->ms == sizeof(struct t10_pi_tuple))
		pi_type = id->dps & NVME_NS_DPS_PI_MASK;

	if (blk_get_integrity(disk) &&
	    (ns->pi_type != pi_type || ns->ms != old_ms ||
	     bs != queue_logical_block_size(disk->queue) ||
	     (ns->ms && ns->ext)))
		blk_integrity_unregister(disk);

	ns->pi_type = pi_type;
}

1103 1104 1105 1106
static void nvme_init_integrity(struct nvme_ns *ns)
{
	struct blk_integrity integrity;

1107
	memset(&integrity, 0, sizeof(integrity));
1108 1109 1110
	switch (ns->pi_type) {
	case NVME_NS_DPS_PI_TYPE3:
		integrity.profile = &t10_pi_type3_crc;
1111 1112
		integrity.tag_size = sizeof(u16) + sizeof(u32);
		integrity.flags |= BLK_INTEGRITY_DEVICE_CAPABLE;
1113 1114 1115 1116
		break;
	case NVME_NS_DPS_PI_TYPE1:
	case NVME_NS_DPS_PI_TYPE2:
		integrity.profile = &t10_pi_type1_crc;
1117 1118
		integrity.tag_size = sizeof(u16);
		integrity.flags |= BLK_INTEGRITY_DEVICE_CAPABLE;
1119 1120 1121 1122 1123 1124 1125 1126 1127 1128
		break;
	default:
		integrity.profile = NULL;
		break;
	}
	integrity.tuple_size = ns->ms;
	blk_integrity_register(ns->disk, &integrity);
	blk_queue_max_integrity_segments(ns->queue, 1);
}
#else
1129 1130 1131 1132
static void nvme_prep_integrity(struct gendisk *disk, struct nvme_id_ns *id,
		u16 bs)
{
}
1133 1134 1135 1136 1137
static void nvme_init_integrity(struct nvme_ns *ns)
{
}
#endif /* CONFIG_BLK_DEV_INTEGRITY */

1138 1139 1140 1141 1142 1143
static void nvme_set_chunk_size(struct nvme_ns *ns)
{
	u32 chunk_size = (((u32)ns->noiob) << (ns->lba_shift - 9));
	blk_queue_chunk_sectors(ns->queue, rounddown_pow_of_two(chunk_size));
}

1144 1145
static void nvme_config_discard(struct nvme_ns *ns)
{
1146
	struct nvme_ctrl *ctrl = ns->ctrl;
1147
	u32 logical_block_size = queue_logical_block_size(ns->queue);
1148

1149 1150 1151
	BUILD_BUG_ON(PAGE_SIZE / sizeof(struct nvme_dsm_range) <
			NVME_DSM_MAX_RANGES);

1152 1153 1154 1155 1156 1157 1158 1159 1160
	if (ctrl->nr_streams && ns->sws && ns->sgs) {
		unsigned int sz = logical_block_size * ns->sws * ns->sgs;

		ns->queue->limits.discard_alignment = sz;
		ns->queue->limits.discard_granularity = sz;
	} else {
		ns->queue->limits.discard_alignment = logical_block_size;
		ns->queue->limits.discard_granularity = logical_block_size;
	}
1161
	blk_queue_max_discard_sectors(ns->queue, UINT_MAX);
1162
	blk_queue_max_discard_segments(ns->queue, NVME_DSM_MAX_RANGES);
1163
	queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, ns->queue);
1164 1165 1166

	if (ctrl->quirks & NVME_QUIRK_DEALLOCATE_ZEROES)
		blk_queue_max_write_zeroes_sectors(ns->queue, UINT_MAX);
1167 1168
}

1169 1170
static void nvme_report_ns_ids(struct nvme_ctrl *ctrl, unsigned int nsid,
		struct nvme_id_ns *id, u8 *eui64, u8 *nguid, uuid_t *uuid)
1171
{
1172 1173 1174 1175 1176
	if (ctrl->vs >= NVME_VS(1, 1, 0))
		memcpy(eui64, id->eui64, sizeof(id->eui64));
	if (ctrl->vs >= NVME_VS(1, 2, 0))
		memcpy(nguid, id->nguid, sizeof(id->nguid));
	if (ctrl->vs >= NVME_VS(1, 3, 0)) {
1177 1178 1179
		 /* Don't treat error as fatal we potentially
		  * already have a NGUID or EUI-64
		  */
1180 1181
		if (nvme_identify_ns_descs(ctrl, nsid, eui64, nguid, uuid))
			dev_warn(ctrl->device,
1182 1183
				 "%s: Identify Descriptors failed\n", __func__);
	}
1184 1185 1186 1187 1188
}

static void __nvme_revalidate_disk(struct gendisk *disk, struct nvme_id_ns *id)
{
	struct nvme_ns *ns = disk->private_data;
1189
	struct nvme_ctrl *ctrl = ns->ctrl;
1190
	u16 bs;
1191 1192 1193 1194 1195

	/*
	 * If identify namespace failed, use default 512 byte block size so
	 * block layer can use before failing read/write for 0 capacity.
	 */
1196
	ns->lba_shift = id->lbaf[id->flbas & NVME_NS_FLBAS_LBA_MASK].ds;
1197 1198 1199
	if (ns->lba_shift == 0)
		ns->lba_shift = 9;
	bs = 1 << ns->lba_shift;
1200
	ns->noiob = le16_to_cpu(id->noiob);
1201 1202 1203

	blk_mq_freeze_queue(disk->queue);

1204
	if (ctrl->ops->flags & NVME_F_METADATA_SUPPORTED)
1205
		nvme_prep_integrity(disk, id, bs);
1206
	blk_queue_logical_block_size(ns->queue, bs);
1207 1208
	if (ns->noiob)
		nvme_set_chunk_size(ns);
K
Keith Busch 已提交
1209
	if (ns->ms && !blk_get_integrity(disk) && !ns->ext)
1210 1211 1212 1213 1214 1215
		nvme_init_integrity(ns);
	if (ns->ms && !(ns->ms == 8 && ns->pi_type) && !blk_get_integrity(disk))
		set_capacity(disk, 0);
	else
		set_capacity(disk, le64_to_cpup(&id->nsze) << (ns->lba_shift - 9));

1216
	if (ctrl->oncs & NVME_CTRL_ONCS_DSM)
1217 1218
		nvme_config_discard(ns);
	blk_mq_unfreeze_queue(disk->queue);
1219
}
1220

1221 1222 1223
static int nvme_revalidate_disk(struct gendisk *disk)
{
	struct nvme_ns *ns = disk->private_data;
1224 1225
	struct nvme_ctrl *ctrl = ns->ctrl;
	struct nvme_id_ns *id;
1226 1227
	u8 eui64[8] = { 0 }, nguid[16] = { 0 };
	uuid_t uuid = uuid_null;
1228
	int ret = 0;
1229 1230 1231 1232 1233 1234

	if (test_bit(NVME_NS_DEAD, &ns->flags)) {
		set_capacity(disk, 0);
		return -ENODEV;
	}

1235 1236 1237
	id = nvme_identify_ns(ctrl, ns->ns_id);
	if (!id)
		return -ENODEV;
1238

1239 1240 1241 1242
	if (id->ncap == 0) {
		ret = -ENODEV;
		goto out;
	}
1243

1244 1245 1246 1247 1248 1249 1250 1251 1252
	nvme_report_ns_ids(ctrl, ns->ns_id, id, eui64, nguid, &uuid);
	if (!uuid_equal(&ns->uuid, &uuid) ||
	    memcmp(&ns->nguid, &nguid, sizeof(ns->nguid)) ||
	    memcmp(&ns->eui, &eui64, sizeof(ns->eui))) {
		dev_err(ctrl->device,
			"identifiers changed for nsid %d\n", ns->ns_id);
		ret = -ENODEV;
	}

1253 1254 1255
out:
	kfree(id);
	return ret;
1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331
}

static char nvme_pr_type(enum pr_type type)
{
	switch (type) {
	case PR_WRITE_EXCLUSIVE:
		return 1;
	case PR_EXCLUSIVE_ACCESS:
		return 2;
	case PR_WRITE_EXCLUSIVE_REG_ONLY:
		return 3;
	case PR_EXCLUSIVE_ACCESS_REG_ONLY:
		return 4;
	case PR_WRITE_EXCLUSIVE_ALL_REGS:
		return 5;
	case PR_EXCLUSIVE_ACCESS_ALL_REGS:
		return 6;
	default:
		return 0;
	}
};

static int nvme_pr_command(struct block_device *bdev, u32 cdw10,
				u64 key, u64 sa_key, u8 op)
{
	struct nvme_ns *ns = bdev->bd_disk->private_data;
	struct nvme_command c;
	u8 data[16] = { 0, };

	put_unaligned_le64(key, &data[0]);
	put_unaligned_le64(sa_key, &data[8]);

	memset(&c, 0, sizeof(c));
	c.common.opcode = op;
	c.common.nsid = cpu_to_le32(ns->ns_id);
	c.common.cdw10[0] = cpu_to_le32(cdw10);

	return nvme_submit_sync_cmd(ns->queue, &c, data, 16);
}

static int nvme_pr_register(struct block_device *bdev, u64 old,
		u64 new, unsigned flags)
{
	u32 cdw10;

	if (flags & ~PR_FL_IGNORE_KEY)
		return -EOPNOTSUPP;

	cdw10 = old ? 2 : 0;
	cdw10 |= (flags & PR_FL_IGNORE_KEY) ? 1 << 3 : 0;
	cdw10 |= (1 << 30) | (1 << 31); /* PTPL=1 */
	return nvme_pr_command(bdev, cdw10, old, new, nvme_cmd_resv_register);
}

static int nvme_pr_reserve(struct block_device *bdev, u64 key,
		enum pr_type type, unsigned flags)
{
	u32 cdw10;

	if (flags & ~PR_FL_IGNORE_KEY)
		return -EOPNOTSUPP;

	cdw10 = nvme_pr_type(type) << 8;
	cdw10 |= ((flags & PR_FL_IGNORE_KEY) ? 1 << 3 : 0);
	return nvme_pr_command(bdev, cdw10, key, 0, nvme_cmd_resv_acquire);
}

static int nvme_pr_preempt(struct block_device *bdev, u64 old, u64 new,
		enum pr_type type, bool abort)
{
	u32 cdw10 = nvme_pr_type(type) << 8 | abort ? 2 : 1;
	return nvme_pr_command(bdev, cdw10, old, new, nvme_cmd_resv_acquire);
}

static int nvme_pr_clear(struct block_device *bdev, u64 key)
{
1332
	u32 cdw10 = 1 | (key ? 1 << 3 : 0);
1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349
	return nvme_pr_command(bdev, cdw10, key, 0, nvme_cmd_resv_register);
}

static int nvme_pr_release(struct block_device *bdev, u64 key, enum pr_type type)
{
	u32 cdw10 = nvme_pr_type(type) << 8 | key ? 1 << 3 : 0;
	return nvme_pr_command(bdev, cdw10, key, 0, nvme_cmd_resv_release);
}

static const struct pr_ops nvme_pr_ops = {
	.pr_register	= nvme_pr_register,
	.pr_reserve	= nvme_pr_reserve,
	.pr_release	= nvme_pr_release,
	.pr_preempt	= nvme_pr_preempt,
	.pr_clear	= nvme_pr_clear,
};

1350
#ifdef CONFIG_BLK_SED_OPAL
1351 1352
int nvme_sec_submit(void *data, u16 spsp, u8 secp, void *buffer, size_t len,
		bool send)
1353
{
1354
	struct nvme_ctrl *ctrl = data;
1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371
	struct nvme_command cmd;

	memset(&cmd, 0, sizeof(cmd));
	if (send)
		cmd.common.opcode = nvme_admin_security_send;
	else
		cmd.common.opcode = nvme_admin_security_recv;
	cmd.common.nsid = 0;
	cmd.common.cdw10[0] = cpu_to_le32(((u32)secp) << 24 | ((u32)spsp) << 8);
	cmd.common.cdw10[1] = cpu_to_le32(len);

	return __nvme_submit_sync_cmd(ctrl->admin_q, &cmd, NULL, buffer, len,
				      ADMIN_TIMEOUT, NVME_QID_ANY, 1, 0);
}
EXPORT_SYMBOL_GPL(nvme_sec_submit);
#endif /* CONFIG_BLK_SED_OPAL */

1372
static const struct block_device_operations nvme_fops = {
1373 1374
	.owner		= THIS_MODULE,
	.ioctl		= nvme_ioctl,
1375
	.compat_ioctl	= nvme_ioctl,
1376 1377 1378 1379 1380 1381 1382
	.open		= nvme_open,
	.release	= nvme_release,
	.getgeo		= nvme_getgeo,
	.revalidate_disk= nvme_revalidate_disk,
	.pr_ops		= &nvme_pr_ops,
};

1383 1384 1385 1386 1387 1388 1389 1390
static int nvme_wait_ready(struct nvme_ctrl *ctrl, u64 cap, bool enabled)
{
	unsigned long timeout =
		((NVME_CAP_TIMEOUT(cap) + 1) * HZ / 2) + jiffies;
	u32 csts, bit = enabled ? NVME_CSTS_RDY : 0;
	int ret;

	while ((ret = ctrl->ops->reg_read32(ctrl, NVME_REG_CSTS, &csts)) == 0) {
K
Keith Busch 已提交
1391 1392
		if (csts == ~0)
			return -ENODEV;
1393 1394 1395 1396 1397 1398 1399
		if ((csts & NVME_CSTS_RDY) == bit)
			break;

		msleep(100);
		if (fatal_signal_pending(current))
			return -EINTR;
		if (time_after(jiffies, timeout)) {
1400
			dev_err(ctrl->device,
1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425
				"Device not ready; aborting %s\n", enabled ?
						"initialisation" : "reset");
			return -ENODEV;
		}
	}

	return ret;
}

/*
 * If the device has been passed off to us in an enabled state, just clear
 * the enabled bit.  The spec says we should set the 'shutdown notification
 * bits', but doing so may cause the device to complete commands to the
 * admin queue ... and we don't know what memory that might be pointing at!
 */
int nvme_disable_ctrl(struct nvme_ctrl *ctrl, u64 cap)
{
	int ret;

	ctrl->ctrl_config &= ~NVME_CC_SHN_MASK;
	ctrl->ctrl_config &= ~NVME_CC_ENABLE;

	ret = ctrl->ops->reg_write32(ctrl, NVME_REG_CC, ctrl->ctrl_config);
	if (ret)
		return ret;
1426

1427
	if (ctrl->quirks & NVME_QUIRK_DELAY_BEFORE_CHK_RDY)
1428 1429
		msleep(NVME_QUIRK_DELAY_AMOUNT);

1430 1431
	return nvme_wait_ready(ctrl, cap, false);
}
1432
EXPORT_SYMBOL_GPL(nvme_disable_ctrl);
1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444

int nvme_enable_ctrl(struct nvme_ctrl *ctrl, u64 cap)
{
	/*
	 * Default to a 4K page size, with the intention to update this
	 * path in the future to accomodate architectures with differing
	 * kernel and IO page sizes.
	 */
	unsigned dev_page_min = NVME_CAP_MPSMIN(cap) + 12, page_shift = 12;
	int ret;

	if (page_shift < dev_page_min) {
1445
		dev_err(ctrl->device,
1446 1447 1448 1449 1450 1451 1452 1453 1454
			"Minimum device page size %u too large for host (%u)\n",
			1 << dev_page_min, 1 << page_shift);
		return -ENODEV;
	}

	ctrl->page_size = 1 << page_shift;

	ctrl->ctrl_config = NVME_CC_CSS_NVM;
	ctrl->ctrl_config |= (page_shift - 12) << NVME_CC_MPS_SHIFT;
1455
	ctrl->ctrl_config |= NVME_CC_AMS_RR | NVME_CC_SHN_NONE;
1456 1457 1458 1459 1460 1461 1462 1463
	ctrl->ctrl_config |= NVME_CC_IOSQES | NVME_CC_IOCQES;
	ctrl->ctrl_config |= NVME_CC_ENABLE;

	ret = ctrl->ops->reg_write32(ctrl, NVME_REG_CC, ctrl->ctrl_config);
	if (ret)
		return ret;
	return nvme_wait_ready(ctrl, cap, true);
}
1464
EXPORT_SYMBOL_GPL(nvme_enable_ctrl);
1465 1466 1467

int nvme_shutdown_ctrl(struct nvme_ctrl *ctrl)
{
1468
	unsigned long timeout = jiffies + (ctrl->shutdown_timeout * HZ);
1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486
	u32 csts;
	int ret;

	ctrl->ctrl_config &= ~NVME_CC_SHN_MASK;
	ctrl->ctrl_config |= NVME_CC_SHN_NORMAL;

	ret = ctrl->ops->reg_write32(ctrl, NVME_REG_CC, ctrl->ctrl_config);
	if (ret)
		return ret;

	while ((ret = ctrl->ops->reg_read32(ctrl, NVME_REG_CSTS, &csts)) == 0) {
		if ((csts & NVME_CSTS_SHST_MASK) == NVME_CSTS_SHST_CMPLT)
			break;

		msleep(100);
		if (fatal_signal_pending(current))
			return -EINTR;
		if (time_after(jiffies, timeout)) {
1487
			dev_err(ctrl->device,
1488 1489 1490 1491 1492 1493 1494
				"Device shutdown incomplete; abort shutdown\n");
			return -ENODEV;
		}
	}

	return ret;
}
1495
EXPORT_SYMBOL_GPL(nvme_shutdown_ctrl);
1496

1497 1498 1499
static void nvme_set_queue_limits(struct nvme_ctrl *ctrl,
		struct request_queue *q)
{
1500 1501
	bool vwc = false;

1502
	if (ctrl->max_hw_sectors) {
1503 1504 1505
		u32 max_segments =
			(ctrl->max_hw_sectors / (ctrl->page_size >> 9)) + 1;

1506
		blk_queue_max_hw_sectors(q, ctrl->max_hw_sectors);
1507
		blk_queue_max_segments(q, min_t(u32, max_segments, USHRT_MAX));
1508
	}
K
Keith Busch 已提交
1509 1510
	if (ctrl->quirks & NVME_QUIRK_STRIPE_SIZE)
		blk_queue_chunk_sectors(q, ctrl->max_hw_sectors);
1511
	blk_queue_virt_boundary(q, ctrl->page_size - 1);
1512 1513 1514
	if (ctrl->vwc & NVME_CTRL_VWC_PRESENT)
		vwc = true;
	blk_queue_write_cache(q, vwc, vwc);
1515 1516
}

1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533
static int nvme_configure_timestamp(struct nvme_ctrl *ctrl)
{
	__le64 ts;
	int ret;

	if (!(ctrl->oncs & NVME_CTRL_ONCS_TIMESTAMP))
		return 0;

	ts = cpu_to_le64(ktime_to_ms(ktime_get_real()));
	ret = nvme_set_features(ctrl, NVME_FEAT_TIMESTAMP, 0, &ts, sizeof(ts),
			NULL);
	if (ret)
		dev_warn_once(ctrl->device,
			"could not set timestamp (%d)\n", ret);
	return ret;
}

1534
static int nvme_configure_apst(struct nvme_ctrl *ctrl)
1535 1536 1537 1538 1539 1540 1541 1542
{
	/*
	 * APST (Autonomous Power State Transition) lets us program a
	 * table of power state transitions that the controller will
	 * perform automatically.  We configure it with a simple
	 * heuristic: we are willing to spend at most 2% of the time
	 * transitioning between power states.  Therefore, when running
	 * in any given state, we will enter the next lower-power
A
Andy Lutomirski 已提交
1543
	 * non-operational state after waiting 50 * (enlat + exlat)
1544
	 * microseconds, as long as that state's exit latency is under
1545 1546 1547 1548 1549 1550 1551 1552 1553
	 * the requested maximum latency.
	 *
	 * We will not autonomously enter any non-operational state for
	 * which the total latency exceeds ps_max_latency_us.  Users
	 * can set ps_max_latency_us to zero to turn off APST.
	 */

	unsigned apste;
	struct nvme_feat_auto_pst *table;
1554 1555
	u64 max_lat_us = 0;
	int max_ps = -1;
1556 1557 1558 1559 1560 1561 1562
	int ret;

	/*
	 * If APST isn't supported or if we haven't been initialized yet,
	 * then don't do anything.
	 */
	if (!ctrl->apsta)
1563
		return 0;
1564 1565 1566

	if (ctrl->npss > 31) {
		dev_warn(ctrl->device, "NPSS is invalid; not using APST\n");
1567
		return 0;
1568 1569 1570 1571
	}

	table = kzalloc(sizeof(*table), GFP_KERNEL);
	if (!table)
1572
		return 0;
1573

1574
	if (!ctrl->apst_enabled || ctrl->ps_max_latency_us == 0) {
1575 1576
		/* Turn off APST. */
		apste = 0;
1577
		dev_dbg(ctrl->device, "APST disabled\n");
1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588
	} else {
		__le64 target = cpu_to_le64(0);
		int state;

		/*
		 * Walk through all states from lowest- to highest-power.
		 * According to the spec, lower-numbered states use more
		 * power.  NPSS, despite the name, is the index of the
		 * lowest-power state, not the number of states.
		 */
		for (state = (int)ctrl->npss; state >= 0; state--) {
1589
			u64 total_latency_us, exit_latency_us, transition_ms;
1590 1591 1592 1593

			if (target)
				table->entries[state] = target;

1594 1595 1596 1597 1598 1599 1600 1601
			/*
			 * Don't allow transitions to the deepest state
			 * if it's quirked off.
			 */
			if (state == ctrl->npss &&
			    (ctrl->quirks & NVME_QUIRK_NO_DEEPEST_PS))
				continue;

1602 1603 1604 1605 1606 1607 1608 1609
			/*
			 * Is this state a useful non-operational state for
			 * higher-power states to autonomously transition to?
			 */
			if (!(ctrl->psd[state].flags &
			      NVME_PS_FLAGS_NON_OP_STATE))
				continue;

1610 1611 1612
			exit_latency_us =
				(u64)le32_to_cpu(ctrl->psd[state].exit_lat);
			if (exit_latency_us > ctrl->ps_max_latency_us)
1613 1614
				continue;

1615 1616 1617 1618
			total_latency_us =
				exit_latency_us +
				le32_to_cpu(ctrl->psd[state].entry_lat);

1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629
			/*
			 * This state is good.  Use it as the APST idle
			 * target for higher power states.
			 */
			transition_ms = total_latency_us + 19;
			do_div(transition_ms, 20);
			if (transition_ms > (1 << 24) - 1)
				transition_ms = (1 << 24) - 1;

			target = cpu_to_le64((state << 3) |
					     (transition_ms << 8));
1630 1631 1632 1633 1634 1635

			if (max_ps == -1)
				max_ps = state;

			if (total_latency_us > max_lat_us)
				max_lat_us = total_latency_us;
1636 1637 1638
		}

		apste = 1;
1639 1640 1641 1642 1643 1644 1645

		if (max_ps == -1) {
			dev_dbg(ctrl->device, "APST enabled but no non-operational states are available\n");
		} else {
			dev_dbg(ctrl->device, "APST enabled: max PS = %d, max round-trip latency = %lluus, table = %*phN\n",
				max_ps, max_lat_us, (int)sizeof(*table), table);
		}
1646 1647 1648 1649 1650 1651 1652 1653
	}

	ret = nvme_set_features(ctrl, NVME_FEAT_AUTO_PST, apste,
				table, sizeof(*table), NULL);
	if (ret)
		dev_err(ctrl->device, "failed to set APST feature (%d)\n", ret);

	kfree(table);
1654
	return ret;
1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677
}

static void nvme_set_latency_tolerance(struct device *dev, s32 val)
{
	struct nvme_ctrl *ctrl = dev_get_drvdata(dev);
	u64 latency;

	switch (val) {
	case PM_QOS_LATENCY_TOLERANCE_NO_CONSTRAINT:
	case PM_QOS_LATENCY_ANY:
		latency = U64_MAX;
		break;

	default:
		latency = val;
	}

	if (ctrl->ps_max_latency_us != latency) {
		ctrl->ps_max_latency_us = latency;
		nvme_configure_apst(ctrl);
	}
}

1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690
struct nvme_core_quirk_entry {
	/*
	 * NVMe model and firmware strings are padded with spaces.  For
	 * simplicity, strings in the quirk table are padded with NULLs
	 * instead.
	 */
	u16 vid;
	const char *mn;
	const char *fr;
	unsigned long quirks;
};

static const struct nvme_core_quirk_entry core_quirks[] = {
1691
	{
1692 1693 1694 1695 1696 1697
		/*
		 * This Toshiba device seems to die using any APST states.  See:
		 * https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1678184/comments/11
		 */
		.vid = 0x1179,
		.mn = "THNSF5256GPUK TOSHIBA",
1698
		.quirks = NVME_QUIRK_NO_APST,
1699
	}
1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730
};

/* match is null-terminated but idstr is space-padded. */
static bool string_matches(const char *idstr, const char *match, size_t len)
{
	size_t matchlen;

	if (!match)
		return true;

	matchlen = strlen(match);
	WARN_ON_ONCE(matchlen > len);

	if (memcmp(idstr, match, matchlen))
		return false;

	for (; matchlen < len; matchlen++)
		if (idstr[matchlen] != ' ')
			return false;

	return true;
}

static bool quirk_matches(const struct nvme_id_ctrl *id,
			  const struct nvme_core_quirk_entry *q)
{
	return q->vid == le16_to_cpu(id->vid) &&
		string_matches(id->mn, q->mn, sizeof(id->mn)) &&
		string_matches(id->fr, q->fr, sizeof(id->fr));
}

1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755
static void nvme_init_subnqn(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id)
{
	size_t nqnlen;
	int off;

	nqnlen = strnlen(id->subnqn, NVMF_NQN_SIZE);
	if (nqnlen > 0 && nqnlen < NVMF_NQN_SIZE) {
		strcpy(ctrl->subnqn, id->subnqn);
		return;
	}

	if (ctrl->vs >= NVME_VS(1, 2, 1))
		dev_warn(ctrl->device, "missing or invalid SUBNQN field.\n");

	/* Generate a "fake" NQN per Figure 254 in NVMe 1.3 + ECN 001 */
	off = snprintf(ctrl->subnqn, NVMF_NQN_SIZE,
			"nqn.2014.08.org.nvmexpress:%4x%4x",
			le16_to_cpu(id->vid), le16_to_cpu(id->ssvid));
	memcpy(ctrl->subnqn + off, id->sn, sizeof(id->sn));
	off += sizeof(id->sn);
	memcpy(ctrl->subnqn + off, id->mn, sizeof(id->mn));
	off += sizeof(id->mn);
	memset(ctrl->subnqn + off, 0, sizeof(ctrl->subnqn) - off);
}

1756 1757 1758 1759 1760 1761 1762 1763 1764 1765
/*
 * Initialize the cached copies of the Identify data and various controller
 * register in our nvme_ctrl structure.  This should be called as soon as
 * the admin queue is fully up and running.
 */
int nvme_init_identify(struct nvme_ctrl *ctrl)
{
	struct nvme_id_ctrl *id;
	u64 cap;
	int ret, page_shift;
1766
	u32 max_hw_sectors;
1767
	bool prev_apst_enabled;
1768

1769 1770
	ret = ctrl->ops->reg_read32(ctrl, NVME_REG_VS, &ctrl->vs);
	if (ret) {
1771
		dev_err(ctrl->device, "Reading VS failed (%d)\n", ret);
1772 1773 1774
		return ret;
	}

1775 1776
	ret = ctrl->ops->reg_read64(ctrl, NVME_REG_CAP, &cap);
	if (ret) {
1777
		dev_err(ctrl->device, "Reading CAP failed (%d)\n", ret);
1778 1779 1780 1781
		return ret;
	}
	page_shift = NVME_CAP_MPSMIN(cap) + 12;

1782
	if (ctrl->vs >= NVME_VS(1, 1, 0))
1783 1784
		ctrl->subsystem = NVME_CAP_NSSRC(cap);

1785 1786
	ret = nvme_identify_ctrl(ctrl, &id);
	if (ret) {
1787
		dev_err(ctrl->device, "Identify Controller failed (%d)\n", ret);
1788 1789 1790
		return -EIO;
	}

1791 1792
	nvme_init_subnqn(ctrl, id);

1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810
	if (!ctrl->identified) {
		/*
		 * Check for quirks.  Quirk can depend on firmware version,
		 * so, in principle, the set of quirks present can change
		 * across a reset.  As a possible future enhancement, we
		 * could re-scan for quirks every time we reinitialize
		 * the device, but we'd have to make sure that the driver
		 * behaves intelligently if the quirks change.
		 */

		int i;

		for (i = 0; i < ARRAY_SIZE(core_quirks); i++) {
			if (quirk_matches(id, &core_quirks[i]))
				ctrl->quirks |= core_quirks[i].quirks;
		}
	}

1811
	if (force_apst && (ctrl->quirks & NVME_QUIRK_NO_DEEPEST_PS)) {
1812
		dev_warn(ctrl->device, "forcibly allowing all power states due to nvme_core.force_apst -- use at your own risk\n");
1813 1814 1815
		ctrl->quirks &= ~NVME_QUIRK_NO_DEEPEST_PS;
	}

1816
	ctrl->oacs = le16_to_cpu(id->oacs);
1817
	ctrl->vid = le16_to_cpu(id->vid);
1818
	ctrl->oncs = le16_to_cpup(&id->oncs);
1819
	atomic_set(&ctrl->abort_limit, id->acl + 1);
1820
	ctrl->vwc = id->vwc;
M
Ming Lin 已提交
1821
	ctrl->cntlid = le16_to_cpup(&id->cntlid);
1822 1823 1824 1825
	memcpy(ctrl->serial, id->sn, sizeof(id->sn));
	memcpy(ctrl->model, id->mn, sizeof(id->mn));
	memcpy(ctrl->firmware_rev, id->fr, sizeof(id->fr));
	if (id->mdts)
1826
		max_hw_sectors = 1 << (id->mdts + page_shift - 9);
1827
	else
1828 1829 1830
		max_hw_sectors = UINT_MAX;
	ctrl->max_hw_sectors =
		min_not_zero(ctrl->max_hw_sectors, max_hw_sectors);
1831

1832
	nvme_set_queue_limits(ctrl, ctrl->admin_q);
1833
	ctrl->sgls = le32_to_cpu(id->sgls);
S
Sagi Grimberg 已提交
1834
	ctrl->kas = le16_to_cpu(id->kas);
1835

1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849
	if (id->rtd3e) {
		/* us -> s */
		u32 transition_time = le32_to_cpu(id->rtd3e) / 1000000;

		ctrl->shutdown_timeout = clamp_t(unsigned int, transition_time,
						 shutdown_timeout, 60);

		if (ctrl->shutdown_timeout != shutdown_timeout)
			dev_warn(ctrl->device,
				 "Shutdown timeout set to %u seconds\n",
				 ctrl->shutdown_timeout);
	} else
		ctrl->shutdown_timeout = shutdown_timeout;

1850
	ctrl->npss = id->npss;
1851 1852
	ctrl->apsta = id->apsta;
	prev_apst_enabled = ctrl->apst_enabled;
1853 1854
	if (ctrl->quirks & NVME_QUIRK_NO_APST) {
		if (force_apst && id->apsta) {
1855
			dev_warn(ctrl->device, "forcibly allowing APST due to nvme_core.force_apst -- use at your own risk\n");
1856
			ctrl->apst_enabled = true;
1857
		} else {
1858
			ctrl->apst_enabled = false;
1859 1860
		}
	} else {
1861
		ctrl->apst_enabled = id->apsta;
1862
	}
1863 1864
	memcpy(ctrl->psd, id->psd, sizeof(ctrl->psd));

1865
	if (ctrl->ops->flags & NVME_F_FABRICS) {
1866 1867 1868 1869 1870 1871 1872 1873 1874
		ctrl->icdoff = le16_to_cpu(id->icdoff);
		ctrl->ioccsz = le32_to_cpu(id->ioccsz);
		ctrl->iorcsz = le32_to_cpu(id->iorcsz);
		ctrl->maxcmd = le16_to_cpu(id->maxcmd);

		/*
		 * In fabrics we need to verify the cntlid matches the
		 * admin connect
		 */
1875
		if (ctrl->cntlid != le16_to_cpu(id->cntlid)) {
1876
			ret = -EINVAL;
1877 1878
			goto out_free;
		}
S
Sagi Grimberg 已提交
1879 1880

		if (!ctrl->opts->discovery_nqn && !ctrl->kas) {
1881
			dev_err(ctrl->device,
S
Sagi Grimberg 已提交
1882 1883
				"keep-alive support is mandatory for fabrics\n");
			ret = -EINVAL;
1884
			goto out_free;
S
Sagi Grimberg 已提交
1885
		}
1886 1887
	} else {
		ctrl->cntlid = le16_to_cpu(id->cntlid);
1888 1889
		ctrl->hmpre = le32_to_cpu(id->hmpre);
		ctrl->hmmin = le32_to_cpu(id->hmmin);
1890 1891
		ctrl->hmminds = le32_to_cpu(id->hmminds);
		ctrl->hmmaxd = le16_to_cpu(id->hmmaxd);
1892
	}
1893

1894
	kfree(id);
1895

1896
	if (ctrl->apst_enabled && !prev_apst_enabled)
1897
		dev_pm_qos_expose_latency_tolerance(ctrl->device);
1898
	else if (!ctrl->apst_enabled && prev_apst_enabled)
1899 1900
		dev_pm_qos_hide_latency_tolerance(ctrl->device);

1901 1902 1903
	ret = nvme_configure_apst(ctrl);
	if (ret < 0)
		return ret;
1904 1905 1906 1907
	
	ret = nvme_configure_timestamp(ctrl);
	if (ret < 0)
		return ret;
1908 1909 1910 1911

	ret = nvme_configure_directives(ctrl);
	if (ret < 0)
		return ret;
1912

1913
	ctrl->identified = true;
1914

1915 1916 1917 1918
	return 0;

out_free:
	kfree(id);
1919
	return ret;
1920
}
1921
EXPORT_SYMBOL_GPL(nvme_init_identify);
1922

1923
static int nvme_dev_open(struct inode *inode, struct file *file)
1924
{
1925 1926 1927
	struct nvme_ctrl *ctrl;
	int instance = iminor(inode);
	int ret = -ENODEV;
1928

1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946
	spin_lock(&dev_list_lock);
	list_for_each_entry(ctrl, &nvme_ctrl_list, node) {
		if (ctrl->instance != instance)
			continue;

		if (!ctrl->admin_q) {
			ret = -EWOULDBLOCK;
			break;
		}
		if (!kref_get_unless_zero(&ctrl->kref))
			break;
		file->private_data = ctrl;
		ret = 0;
		break;
	}
	spin_unlock(&dev_list_lock);

	return ret;
1947 1948
}

1949
static int nvme_dev_release(struct inode *inode, struct file *file)
1950
{
1951 1952 1953 1954
	nvme_put_ctrl(file->private_data);
	return 0;
}

1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967
static int nvme_dev_user_cmd(struct nvme_ctrl *ctrl, void __user *argp)
{
	struct nvme_ns *ns;
	int ret;

	mutex_lock(&ctrl->namespaces_mutex);
	if (list_empty(&ctrl->namespaces)) {
		ret = -ENOTTY;
		goto out_unlock;
	}

	ns = list_first_entry(&ctrl->namespaces, struct nvme_ns, list);
	if (ns != list_last_entry(&ctrl->namespaces, struct nvme_ns, list)) {
1968
		dev_warn(ctrl->device,
1969 1970 1971 1972 1973
			"NVME_IOCTL_IO_CMD not supported when multiple namespaces present!\n");
		ret = -EINVAL;
		goto out_unlock;
	}

1974
	dev_warn(ctrl->device,
1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987
		"using deprecated NVME_IOCTL_IO_CMD ioctl on the char device!\n");
	kref_get(&ns->kref);
	mutex_unlock(&ctrl->namespaces_mutex);

	ret = nvme_user_cmd(ctrl, ns, argp);
	nvme_put_ns(ns);
	return ret;

out_unlock:
	mutex_unlock(&ctrl->namespaces_mutex);
	return ret;
}

1988 1989 1990 1991 1992 1993 1994 1995 1996 1997
static long nvme_dev_ioctl(struct file *file, unsigned int cmd,
		unsigned long arg)
{
	struct nvme_ctrl *ctrl = file->private_data;
	void __user *argp = (void __user *)arg;

	switch (cmd) {
	case NVME_IOCTL_ADMIN_CMD:
		return nvme_user_cmd(ctrl, NULL, argp);
	case NVME_IOCTL_IO_CMD:
1998
		return nvme_dev_user_cmd(ctrl, argp);
1999
	case NVME_IOCTL_RESET:
2000
		dev_warn(ctrl->device, "resetting controller\n");
2001
		return nvme_reset_ctrl_sync(ctrl);
2002 2003
	case NVME_IOCTL_SUBSYS_RESET:
		return nvme_reset_subsystem(ctrl);
K
Keith Busch 已提交
2004 2005 2006
	case NVME_IOCTL_RESCAN:
		nvme_queue_scan(ctrl);
		return 0;
2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026
	default:
		return -ENOTTY;
	}
}

static const struct file_operations nvme_dev_fops = {
	.owner		= THIS_MODULE,
	.open		= nvme_dev_open,
	.release	= nvme_dev_release,
	.unlocked_ioctl	= nvme_dev_ioctl,
	.compat_ioctl	= nvme_dev_ioctl,
};

static ssize_t nvme_sysfs_reset(struct device *dev,
				struct device_attribute *attr, const char *buf,
				size_t count)
{
	struct nvme_ctrl *ctrl = dev_get_drvdata(dev);
	int ret;

2027
	ret = nvme_reset_ctrl_sync(ctrl);
2028 2029 2030
	if (ret < 0)
		return ret;
	return count;
2031
}
2032
static DEVICE_ATTR(reset_controller, S_IWUSR, NULL, nvme_sysfs_reset);
2033

K
Keith Busch 已提交
2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044
static ssize_t nvme_sysfs_rescan(struct device *dev,
				struct device_attribute *attr, const char *buf,
				size_t count)
{
	struct nvme_ctrl *ctrl = dev_get_drvdata(dev);

	nvme_queue_scan(ctrl);
	return count;
}
static DEVICE_ATTR(rescan_controller, S_IWUSR, NULL, nvme_sysfs_rescan);

2045 2046 2047
static ssize_t wwid_show(struct device *dev, struct device_attribute *attr,
								char *buf)
{
2048
	struct nvme_ns *ns = nvme_get_ns_from_dev(dev);
2049 2050 2051 2052
	struct nvme_ctrl *ctrl = ns->ctrl;
	int serial_len = sizeof(ctrl->serial);
	int model_len = sizeof(ctrl->model);

2053 2054 2055
	if (!uuid_is_null(&ns->uuid))
		return sprintf(buf, "uuid.%pU\n", &ns->uuid);

2056 2057
	if (memchr_inv(ns->nguid, 0, sizeof(ns->nguid)))
		return sprintf(buf, "eui.%16phN\n", ns->nguid);
2058 2059 2060 2061

	if (memchr_inv(ns->eui, 0, sizeof(ns->eui)))
		return sprintf(buf, "eui.%8phN\n", ns->eui);

2062 2063
	while (serial_len > 0 && (ctrl->serial[serial_len - 1] == ' ' ||
				  ctrl->serial[serial_len - 1] == '\0'))
2064
		serial_len--;
2065 2066
	while (model_len > 0 && (ctrl->model[model_len - 1] == ' ' ||
				 ctrl->model[model_len - 1] == '\0'))
2067 2068 2069 2070 2071 2072 2073
		model_len--;

	return sprintf(buf, "nvme.%04x-%*phN-%*phN-%08x\n", ctrl->vid,
		serial_len, ctrl->serial, model_len, ctrl->model, ns->ns_id);
}
static DEVICE_ATTR(wwid, S_IRUGO, wwid_show, NULL);

2074 2075 2076 2077 2078 2079 2080 2081
static ssize_t nguid_show(struct device *dev, struct device_attribute *attr,
			  char *buf)
{
	struct nvme_ns *ns = nvme_get_ns_from_dev(dev);
	return sprintf(buf, "%pU\n", ns->nguid);
}
static DEVICE_ATTR(nguid, S_IRUGO, nguid_show, NULL);

2082 2083 2084
static ssize_t uuid_show(struct device *dev, struct device_attribute *attr,
								char *buf)
{
2085
	struct nvme_ns *ns = nvme_get_ns_from_dev(dev);
2086 2087 2088 2089 2090 2091 2092 2093 2094 2095

	/* For backward compatibility expose the NGUID to userspace if
	 * we have no UUID set
	 */
	if (uuid_is_null(&ns->uuid)) {
		printk_ratelimited(KERN_WARNING
				   "No UUID available providing old NGUID\n");
		return sprintf(buf, "%pU\n", ns->nguid);
	}
	return sprintf(buf, "%pU\n", &ns->uuid);
2096 2097 2098 2099 2100 2101
}
static DEVICE_ATTR(uuid, S_IRUGO, uuid_show, NULL);

static ssize_t eui_show(struct device *dev, struct device_attribute *attr,
								char *buf)
{
2102
	struct nvme_ns *ns = nvme_get_ns_from_dev(dev);
2103 2104 2105 2106 2107 2108 2109
	return sprintf(buf, "%8phd\n", ns->eui);
}
static DEVICE_ATTR(eui, S_IRUGO, eui_show, NULL);

static ssize_t nsid_show(struct device *dev, struct device_attribute *attr,
								char *buf)
{
2110
	struct nvme_ns *ns = nvme_get_ns_from_dev(dev);
2111 2112 2113 2114 2115
	return sprintf(buf, "%d\n", ns->ns_id);
}
static DEVICE_ATTR(nsid, S_IRUGO, nsid_show, NULL);

static struct attribute *nvme_ns_attrs[] = {
2116
	&dev_attr_wwid.attr,
2117
	&dev_attr_uuid.attr,
2118
	&dev_attr_nguid.attr,
2119 2120 2121 2122 2123
	&dev_attr_eui.attr,
	&dev_attr_nsid.attr,
	NULL,
};

M
Ming Lin 已提交
2124
static umode_t nvme_ns_attrs_are_visible(struct kobject *kobj,
2125 2126 2127
		struct attribute *a, int n)
{
	struct device *dev = container_of(kobj, struct device, kobj);
2128
	struct nvme_ns *ns = nvme_get_ns_from_dev(dev);
2129 2130

	if (a == &dev_attr_uuid.attr) {
2131 2132 2133 2134 2135
		if (uuid_is_null(&ns->uuid) ||
		    !memchr_inv(ns->nguid, 0, sizeof(ns->nguid)))
			return 0;
	}
	if (a == &dev_attr_nguid.attr) {
2136
		if (!memchr_inv(ns->nguid, 0, sizeof(ns->nguid)))
2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147
			return 0;
	}
	if (a == &dev_attr_eui.attr) {
		if (!memchr_inv(ns->eui, 0, sizeof(ns->eui)))
			return 0;
	}
	return a->mode;
}

static const struct attribute_group nvme_ns_attr_group = {
	.attrs		= nvme_ns_attrs,
M
Ming Lin 已提交
2148
	.is_visible	= nvme_ns_attrs_are_visible,
2149 2150
};

M
Ming Lin 已提交
2151
#define nvme_show_str_function(field)						\
2152 2153 2154 2155 2156 2157 2158 2159
static ssize_t  field##_show(struct device *dev,				\
			    struct device_attribute *attr, char *buf)		\
{										\
        struct nvme_ctrl *ctrl = dev_get_drvdata(dev);				\
        return sprintf(buf, "%.*s\n", (int)sizeof(ctrl->field), ctrl->field);	\
}										\
static DEVICE_ATTR(field, S_IRUGO, field##_show, NULL);

M
Ming Lin 已提交
2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172
#define nvme_show_int_function(field)						\
static ssize_t  field##_show(struct device *dev,				\
			    struct device_attribute *attr, char *buf)		\
{										\
        struct nvme_ctrl *ctrl = dev_get_drvdata(dev);				\
        return sprintf(buf, "%d\n", ctrl->field);	\
}										\
static DEVICE_ATTR(field, S_IRUGO, field##_show, NULL);

nvme_show_str_function(model);
nvme_show_str_function(serial);
nvme_show_str_function(firmware_rev);
nvme_show_int_function(cntlid);
2173

M
Ming Lin 已提交
2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195
static ssize_t nvme_sysfs_delete(struct device *dev,
				struct device_attribute *attr, const char *buf,
				size_t count)
{
	struct nvme_ctrl *ctrl = dev_get_drvdata(dev);

	if (device_remove_file_self(dev, attr))
		ctrl->ops->delete_ctrl(ctrl);
	return count;
}
static DEVICE_ATTR(delete_controller, S_IWUSR, NULL, nvme_sysfs_delete);

static ssize_t nvme_sysfs_show_transport(struct device *dev,
					 struct device_attribute *attr,
					 char *buf)
{
	struct nvme_ctrl *ctrl = dev_get_drvdata(dev);

	return snprintf(buf, PAGE_SIZE, "%s\n", ctrl->ops->name);
}
static DEVICE_ATTR(transport, S_IRUGO, nvme_sysfs_show_transport, NULL);

2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218
static ssize_t nvme_sysfs_show_state(struct device *dev,
				     struct device_attribute *attr,
				     char *buf)
{
	struct nvme_ctrl *ctrl = dev_get_drvdata(dev);
	static const char *const state_name[] = {
		[NVME_CTRL_NEW]		= "new",
		[NVME_CTRL_LIVE]	= "live",
		[NVME_CTRL_RESETTING]	= "resetting",
		[NVME_CTRL_RECONNECTING]= "reconnecting",
		[NVME_CTRL_DELETING]	= "deleting",
		[NVME_CTRL_DEAD]	= "dead",
	};

	if ((unsigned)ctrl->state < ARRAY_SIZE(state_name) &&
	    state_name[ctrl->state])
		return sprintf(buf, "%s\n", state_name[ctrl->state]);

	return sprintf(buf, "unknown state\n");
}

static DEVICE_ATTR(state, S_IRUGO, nvme_sysfs_show_state, NULL);

M
Ming Lin 已提交
2219 2220 2221 2222 2223 2224
static ssize_t nvme_sysfs_show_subsysnqn(struct device *dev,
					 struct device_attribute *attr,
					 char *buf)
{
	struct nvme_ctrl *ctrl = dev_get_drvdata(dev);

2225
	return snprintf(buf, PAGE_SIZE, "%s\n", ctrl->subnqn);
M
Ming Lin 已提交
2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238
}
static DEVICE_ATTR(subsysnqn, S_IRUGO, nvme_sysfs_show_subsysnqn, NULL);

static ssize_t nvme_sysfs_show_address(struct device *dev,
					 struct device_attribute *attr,
					 char *buf)
{
	struct nvme_ctrl *ctrl = dev_get_drvdata(dev);

	return ctrl->ops->get_address(ctrl, buf, PAGE_SIZE);
}
static DEVICE_ATTR(address, S_IRUGO, nvme_sysfs_show_address, NULL);

2239 2240
static struct attribute *nvme_dev_attrs[] = {
	&dev_attr_reset_controller.attr,
K
Keith Busch 已提交
2241
	&dev_attr_rescan_controller.attr,
2242 2243 2244
	&dev_attr_model.attr,
	&dev_attr_serial.attr,
	&dev_attr_firmware_rev.attr,
M
Ming Lin 已提交
2245
	&dev_attr_cntlid.attr,
M
Ming Lin 已提交
2246 2247 2248 2249
	&dev_attr_delete_controller.attr,
	&dev_attr_transport.attr,
	&dev_attr_subsysnqn.attr,
	&dev_attr_address.attr,
2250
	&dev_attr_state.attr,
2251 2252 2253
	NULL
};

M
Ming Lin 已提交
2254 2255 2256 2257 2258 2259
static umode_t nvme_dev_attrs_are_visible(struct kobject *kobj,
		struct attribute *a, int n)
{
	struct device *dev = container_of(kobj, struct device, kobj);
	struct nvme_ctrl *ctrl = dev_get_drvdata(dev);

2260 2261 2262 2263
	if (a == &dev_attr_delete_controller.attr && !ctrl->ops->delete_ctrl)
		return 0;
	if (a == &dev_attr_address.attr && !ctrl->ops->get_address)
		return 0;
M
Ming Lin 已提交
2264 2265 2266 2267

	return a->mode;
}

2268
static struct attribute_group nvme_dev_attrs_group = {
M
Ming Lin 已提交
2269 2270
	.attrs		= nvme_dev_attrs,
	.is_visible	= nvme_dev_attrs_are_visible,
2271 2272 2273 2274 2275 2276 2277
};

static const struct attribute_group *nvme_dev_attr_groups[] = {
	&nvme_dev_attrs_group,
	NULL,
};

2278 2279 2280 2281 2282 2283 2284 2285
static int ns_cmp(void *priv, struct list_head *a, struct list_head *b)
{
	struct nvme_ns *nsa = container_of(a, struct nvme_ns, list);
	struct nvme_ns *nsb = container_of(b, struct nvme_ns, list);

	return nsa->ns_id - nsb->ns_id;
}

2286
static struct nvme_ns *nvme_find_get_ns(struct nvme_ctrl *ctrl, unsigned nsid)
2287
{
2288
	struct nvme_ns *ns, *ret = NULL;
2289

2290
	mutex_lock(&ctrl->namespaces_mutex);
2291
	list_for_each_entry(ns, &ctrl->namespaces, list) {
2292
		if (ns->ns_id == nsid) {
2293 2294
			if (!kref_get_unless_zero(&ns->kref))
				continue;
2295 2296 2297
			ret = ns;
			break;
		}
2298 2299 2300
		if (ns->ns_id > nsid)
			break;
	}
2301 2302
	mutex_unlock(&ctrl->namespaces_mutex);
	return ret;
2303 2304
}

2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330
static int nvme_setup_streams_ns(struct nvme_ctrl *ctrl, struct nvme_ns *ns)
{
	struct streams_directive_params s;
	int ret;

	if (!ctrl->nr_streams)
		return 0;

	ret = nvme_get_stream_params(ctrl, &s, ns->ns_id);
	if (ret)
		return ret;

	ns->sws = le32_to_cpu(s.sws);
	ns->sgs = le16_to_cpu(s.sgs);

	if (ns->sws) {
		unsigned int bs = 1 << ns->lba_shift;

		blk_queue_io_min(ns->queue, bs * ns->sws);
		if (ns->sgs)
			blk_queue_io_opt(ns->queue, bs * ns->sws * ns->sgs);
	}

	return 0;
}

2331 2332 2333 2334
static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
{
	struct nvme_ns *ns;
	struct gendisk *disk;
2335 2336
	struct nvme_id_ns *id;
	char disk_name[DISK_NAME_LEN];
2337 2338 2339 2340 2341 2342
	int node = dev_to_node(ctrl->dev);

	ns = kzalloc_node(sizeof(*ns), GFP_KERNEL, node);
	if (!ns)
		return;

2343 2344 2345 2346
	ns->instance = ida_simple_get(&ctrl->ns_ida, 1, 0, GFP_KERNEL);
	if (ns->instance < 0)
		goto out_free_ns;

2347 2348
	ns->queue = blk_mq_init_queue(ctrl->tagset);
	if (IS_ERR(ns->queue))
2349
		goto out_release_instance;
2350 2351 2352 2353 2354 2355 2356 2357 2358
	queue_flag_set_unlocked(QUEUE_FLAG_NONROT, ns->queue);
	ns->queue->queuedata = ns;
	ns->ctrl = ctrl;

	kref_init(&ns->kref);
	ns->ns_id = nsid;
	ns->lba_shift = 9; /* set to a default value for 512 until disk is validated */

	blk_queue_logical_block_size(ns->queue, 1 << ns->lba_shift);
2359
	nvme_set_queue_limits(ctrl, ns->queue);
2360
	nvme_setup_streams_ns(ctrl, ns);
2361

2362
	sprintf(disk_name, "nvme%dn%d", ctrl->instance, ns->instance);
2363

2364 2365
	id = nvme_identify_ns(ctrl, nsid);
	if (!id)
2366 2367
		goto out_free_queue;

2368 2369 2370 2371 2372
	if (id->ncap == 0)
		goto out_free_id;

	nvme_report_ns_ids(ctrl, ns->ns_id, id, ns->eui, ns->nguid, &ns->uuid);

C
Christoph Hellwig 已提交
2373 2374 2375 2376 2377
	if ((ctrl->quirks & NVME_QUIRK_LIGHTNVM) && id->vs[0] == 0x1) {
		if (nvme_nvm_register(ns, disk_name, node)) {
			dev_warn(ctrl->device, "LightNVM init failure\n");
			goto out_free_id;
		}
2378
	}
2379

2380 2381 2382
	disk = alloc_disk_node(0, node);
	if (!disk)
		goto out_free_id;
2383

2384 2385 2386 2387 2388 2389 2390 2391
	disk->fops = &nvme_fops;
	disk->private_data = ns;
	disk->queue = ns->queue;
	disk->flags = GENHD_FL_EXT_DEVT;
	memcpy(disk->disk_name, disk_name, DISK_NAME_LEN);
	ns->disk = disk;

	__nvme_revalidate_disk(disk, id);
2392

2393 2394 2395 2396
	mutex_lock(&ctrl->namespaces_mutex);
	list_add_tail(&ns->list, &ctrl->namespaces);
	mutex_unlock(&ctrl->namespaces_mutex);

2397
	kref_get(&ctrl->kref);
2398 2399 2400

	kfree(id);

2401
	device_add_disk(ctrl->device, ns->disk);
2402 2403 2404 2405
	if (sysfs_create_group(&disk_to_dev(ns->disk)->kobj,
					&nvme_ns_attr_group))
		pr_warn("%s: failed to create sysfs group for identification\n",
			ns->disk->disk_name);
2406 2407 2408
	if (ns->ndev && nvme_nvm_register_sysfs(ns))
		pr_warn("%s: failed to register lightnvm sysfs group for identification\n",
			ns->disk->disk_name);
2409
	return;
2410 2411
 out_free_id:
	kfree(id);
2412 2413
 out_free_queue:
	blk_cleanup_queue(ns->queue);
2414 2415
 out_release_instance:
	ida_simple_remove(&ctrl->ns_ida, ns->instance);
2416 2417 2418 2419 2420 2421
 out_free_ns:
	kfree(ns);
}

static void nvme_ns_remove(struct nvme_ns *ns)
{
2422 2423
	if (test_and_set_bit(NVME_NS_REMOVING, &ns->flags))
		return;
2424

2425
	if (ns->disk && ns->disk->flags & GENHD_FL_UP) {
2426 2427
		if (blk_get_integrity(ns->disk))
			blk_integrity_unregister(ns->disk);
2428 2429
		sysfs_remove_group(&disk_to_dev(ns->disk)->kobj,
					&nvme_ns_attr_group);
2430 2431
		if (ns->ndev)
			nvme_nvm_unregister_sysfs(ns);
2432 2433 2434
		del_gendisk(ns->disk);
		blk_cleanup_queue(ns->queue);
	}
2435 2436

	mutex_lock(&ns->ctrl->namespaces_mutex);
2437
	list_del_init(&ns->list);
2438 2439
	mutex_unlock(&ns->ctrl->namespaces_mutex);

2440 2441 2442
	nvme_put_ns(ns);
}

2443 2444 2445 2446
static void nvme_validate_ns(struct nvme_ctrl *ctrl, unsigned nsid)
{
	struct nvme_ns *ns;

2447
	ns = nvme_find_get_ns(ctrl, nsid);
2448
	if (ns) {
2449
		if (ns->disk && revalidate_disk(ns->disk))
2450
			nvme_ns_remove(ns);
2451
		nvme_put_ns(ns);
2452 2453 2454 2455
	} else
		nvme_alloc_ns(ctrl, nsid);
}

2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466
static void nvme_remove_invalid_namespaces(struct nvme_ctrl *ctrl,
					unsigned nsid)
{
	struct nvme_ns *ns, *next;

	list_for_each_entry_safe(ns, next, &ctrl->namespaces, list) {
		if (ns->ns_id > nsid)
			nvme_ns_remove(ns);
	}
}

2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480
static int nvme_scan_ns_list(struct nvme_ctrl *ctrl, unsigned nn)
{
	struct nvme_ns *ns;
	__le32 *ns_list;
	unsigned i, j, nsid, prev = 0, num_lists = DIV_ROUND_UP(nn, 1024);
	int ret = 0;

	ns_list = kzalloc(0x1000, GFP_KERNEL);
	if (!ns_list)
		return -ENOMEM;

	for (i = 0; i < num_lists; i++) {
		ret = nvme_identify_ns_list(ctrl, prev, ns_list);
		if (ret)
2481
			goto free;
2482 2483 2484 2485 2486 2487 2488 2489 2490

		for (j = 0; j < min(nn, 1024U); j++) {
			nsid = le32_to_cpu(ns_list[j]);
			if (!nsid)
				goto out;

			nvme_validate_ns(ctrl, nsid);

			while (++prev < nsid) {
2491 2492
				ns = nvme_find_get_ns(ctrl, prev);
				if (ns) {
2493
					nvme_ns_remove(ns);
2494 2495
					nvme_put_ns(ns);
				}
2496 2497 2498 2499 2500
			}
		}
		nn -= j;
	}
 out:
2501 2502
	nvme_remove_invalid_namespaces(ctrl, prev);
 free:
2503 2504 2505 2506
	kfree(ns_list);
	return ret;
}

2507
static void nvme_scan_ns_sequential(struct nvme_ctrl *ctrl, unsigned nn)
2508 2509 2510
{
	unsigned i;

2511 2512 2513
	for (i = 1; i <= nn; i++)
		nvme_validate_ns(ctrl, i);

2514
	nvme_remove_invalid_namespaces(ctrl, nn);
2515 2516
}

2517
static void nvme_scan_work(struct work_struct *work)
2518
{
2519 2520
	struct nvme_ctrl *ctrl =
		container_of(work, struct nvme_ctrl, scan_work);
2521
	struct nvme_id_ctrl *id;
2522
	unsigned nn;
2523

2524 2525 2526
	if (ctrl->state != NVME_CTRL_LIVE)
		return;

2527 2528
	if (nvme_identify_ctrl(ctrl, &id))
		return;
2529 2530

	nn = le32_to_cpu(id->nn);
2531
	if (ctrl->vs >= NVME_VS(1, 1, 0) &&
2532 2533 2534 2535
	    !(ctrl->quirks & NVME_QUIRK_IDENTIFY_CNS)) {
		if (!nvme_scan_ns_list(ctrl, nn))
			goto done;
	}
2536
	nvme_scan_ns_sequential(ctrl, nn);
2537
 done:
2538
	mutex_lock(&ctrl->namespaces_mutex);
2539
	list_sort(NULL, &ctrl->namespaces, ns_cmp);
2540
	mutex_unlock(&ctrl->namespaces_mutex);
2541 2542
	kfree(id);
}
2543 2544 2545 2546 2547 2548 2549 2550

void nvme_queue_scan(struct nvme_ctrl *ctrl)
{
	/*
	 * Do not queue new scan work when a controller is reset during
	 * removal.
	 */
	if (ctrl->state == NVME_CTRL_LIVE)
2551
		queue_work(nvme_wq, &ctrl->scan_work);
2552 2553
}
EXPORT_SYMBOL_GPL(nvme_queue_scan);
2554

2555 2556 2557 2558 2559
/*
 * This function iterates the namespace list unlocked to allow recovery from
 * controller failure. It is up to the caller to ensure the namespace list is
 * not modified by scan work while this function is executing.
 */
2560 2561 2562 2563
void nvme_remove_namespaces(struct nvme_ctrl *ctrl)
{
	struct nvme_ns *ns, *next;

2564 2565 2566 2567 2568 2569 2570 2571 2572
	/*
	 * The dead states indicates the controller was not gracefully
	 * disconnected. In that case, we won't be able to flush any data while
	 * removing the namespaces' disks; fail all the queues now to avoid
	 * potentially having to clean up the failed sync later.
	 */
	if (ctrl->state == NVME_CTRL_DEAD)
		nvme_kill_queues(ctrl);

2573 2574 2575
	list_for_each_entry_safe(ns, next, &ctrl->namespaces, list)
		nvme_ns_remove(ns);
}
2576
EXPORT_SYMBOL_GPL(nvme_remove_namespaces);
2577

2578 2579 2580 2581 2582 2583
static void nvme_async_event_work(struct work_struct *work)
{
	struct nvme_ctrl *ctrl =
		container_of(work, struct nvme_ctrl, async_event_work);

	spin_lock_irq(&ctrl->lock);
2584
	while (ctrl->state == NVME_CTRL_LIVE && ctrl->event_limit > 0) {
2585 2586 2587 2588 2589 2590 2591 2592 2593
		int aer_idx = --ctrl->event_limit;

		spin_unlock_irq(&ctrl->lock);
		ctrl->ops->submit_async_event(ctrl, aer_idx);
		spin_lock_irq(&ctrl->lock);
	}
	spin_unlock_irq(&ctrl->lock);
}

2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617
static bool nvme_ctrl_pp_status(struct nvme_ctrl *ctrl)
{

	u32 csts;

	if (ctrl->ops->reg_read32(ctrl, NVME_REG_CSTS, &csts))
		return false;

	if (csts == ~0)
		return false;

	return ((ctrl->ctrl_config & NVME_CC_ENABLE) && (csts & NVME_CSTS_PP));
}

static void nvme_get_fw_slot_info(struct nvme_ctrl *ctrl)
{
	struct nvme_command c = { };
	struct nvme_fw_slot_info_log *log;

	log = kmalloc(sizeof(*log), GFP_KERNEL);
	if (!log)
		return;

	c.common.opcode = nvme_admin_get_log_page;
A
Arnav Dawn 已提交
2618
	c.common.nsid = cpu_to_le32(NVME_NSID_ALL);
2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658
	c.common.cdw10[0] = nvme_get_log_dw10(NVME_LOG_FW_SLOT, sizeof(*log));

	if (!nvme_submit_sync_cmd(ctrl->admin_q, &c, log, sizeof(*log)))
		dev_warn(ctrl->device,
				"Get FW SLOT INFO log error\n");
	kfree(log);
}

static void nvme_fw_act_work(struct work_struct *work)
{
	struct nvme_ctrl *ctrl = container_of(work,
				struct nvme_ctrl, fw_act_work);
	unsigned long fw_act_timeout;

	if (ctrl->mtfa)
		fw_act_timeout = jiffies +
				msecs_to_jiffies(ctrl->mtfa * 100);
	else
		fw_act_timeout = jiffies +
				msecs_to_jiffies(admin_timeout * 1000);

	nvme_stop_queues(ctrl);
	while (nvme_ctrl_pp_status(ctrl)) {
		if (time_after(jiffies, fw_act_timeout)) {
			dev_warn(ctrl->device,
				"Fw activation timeout, reset controller\n");
			nvme_reset_ctrl(ctrl);
			break;
		}
		msleep(100);
	}

	if (ctrl->state != NVME_CTRL_LIVE)
		return;

	nvme_start_queues(ctrl);
	/* read FW slot informationi to clear the AER*/
	nvme_get_fw_slot_info(ctrl);
}

2659 2660
void nvme_complete_async_event(struct nvme_ctrl *ctrl, __le16 status,
		union nvme_result *res)
2661
{
2662 2663
	u32 result = le32_to_cpu(res->u32);
	bool done = true;
2664

2665 2666 2667 2668 2669
	switch (le16_to_cpu(status) >> 1) {
	case NVME_SC_SUCCESS:
		done = false;
		/*FALLTHRU*/
	case NVME_SC_ABORT_REQ:
2670
		++ctrl->event_limit;
2671
		if (ctrl->state == NVME_CTRL_LIVE)
2672
			queue_work(nvme_wq, &ctrl->async_event_work);
2673 2674 2675
		break;
	default:
		break;
2676 2677
	}

2678
	if (done)
2679 2680 2681 2682 2683 2684 2685
		return;

	switch (result & 0xff07) {
	case NVME_AER_NOTICE_NS_CHANGED:
		dev_info(ctrl->device, "rescanning\n");
		nvme_queue_scan(ctrl);
		break;
2686
	case NVME_AER_NOTICE_FW_ACT_STARTING:
2687
		queue_work(nvme_wq, &ctrl->fw_act_work);
2688
		break;
2689 2690 2691 2692 2693 2694 2695 2696 2697
	default:
		dev_warn(ctrl->device, "async event result %08x\n", result);
	}
}
EXPORT_SYMBOL_GPL(nvme_complete_async_event);

void nvme_queue_async_events(struct nvme_ctrl *ctrl)
{
	ctrl->event_limit = NVME_NR_AERS;
2698
	queue_work(nvme_wq, &ctrl->async_event_work);
2699 2700 2701
}
EXPORT_SYMBOL_GPL(nvme_queue_async_events);

2702
void nvme_stop_ctrl(struct nvme_ctrl *ctrl)
2703
{
2704
	nvme_stop_keep_alive(ctrl);
2705
	flush_work(&ctrl->async_event_work);
2706
	flush_work(&ctrl->scan_work);
2707
	cancel_work_sync(&ctrl->fw_act_work);
2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722
}
EXPORT_SYMBOL_GPL(nvme_stop_ctrl);

void nvme_start_ctrl(struct nvme_ctrl *ctrl)
{
	if (ctrl->kato)
		nvme_start_keep_alive(ctrl);

	if (ctrl->queue_count > 1) {
		nvme_queue_scan(ctrl);
		nvme_queue_async_events(ctrl);
		nvme_start_queues(ctrl);
	}
}
EXPORT_SYMBOL_GPL(nvme_start_ctrl);
2723

2724 2725
void nvme_uninit_ctrl(struct nvme_ctrl *ctrl)
{
2726
	device_destroy(nvme_class, MKDEV(nvme_char_major, ctrl->instance));
2727 2728 2729 2730

	spin_lock(&dev_list_lock);
	list_del(&ctrl->node);
	spin_unlock(&dev_list_lock);
2731
}
2732
EXPORT_SYMBOL_GPL(nvme_uninit_ctrl);
2733 2734 2735 2736

static void nvme_free_ctrl(struct kref *kref)
{
	struct nvme_ctrl *ctrl = container_of(kref, struct nvme_ctrl, kref);
2737 2738

	put_device(ctrl->device);
2739
	ida_simple_remove(&nvme_instance_ida, ctrl->instance);
2740
	ida_destroy(&ctrl->ns_ida);
2741 2742 2743 2744 2745 2746 2747 2748

	ctrl->ops->free_ctrl(ctrl);
}

void nvme_put_ctrl(struct nvme_ctrl *ctrl)
{
	kref_put(&ctrl->kref, nvme_free_ctrl);
}
2749
EXPORT_SYMBOL_GPL(nvme_put_ctrl);
2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760

/*
 * Initialize a NVMe controller structures.  This needs to be called during
 * earliest initialization so that we have the initialized structured around
 * during probing.
 */
int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev,
		const struct nvme_ctrl_ops *ops, unsigned long quirks)
{
	int ret;

2761 2762
	ctrl->state = NVME_CTRL_NEW;
	spin_lock_init(&ctrl->lock);
2763
	INIT_LIST_HEAD(&ctrl->namespaces);
2764
	mutex_init(&ctrl->namespaces_mutex);
2765 2766 2767 2768
	kref_init(&ctrl->kref);
	ctrl->dev = dev;
	ctrl->ops = ops;
	ctrl->quirks = quirks;
2769
	INIT_WORK(&ctrl->scan_work, nvme_scan_work);
2770
	INIT_WORK(&ctrl->async_event_work, nvme_async_event_work);
2771
	INIT_WORK(&ctrl->fw_act_work, nvme_fw_act_work);
2772

2773 2774
	ret = ida_simple_get(&nvme_instance_ida, 0, 0, GFP_KERNEL);
	if (ret < 0)
2775
		goto out;
2776
	ctrl->instance = ret;
2777

2778
	ctrl->device = device_create_with_groups(nvme_class, ctrl->dev,
2779
				MKDEV(nvme_char_major, ctrl->instance),
2780
				ctrl, nvme_dev_attr_groups,
2781
				"nvme%d", ctrl->instance);
2782 2783 2784 2785 2786
	if (IS_ERR(ctrl->device)) {
		ret = PTR_ERR(ctrl->device);
		goto out_release_instance;
	}
	get_device(ctrl->device);
2787
	ida_init(&ctrl->ns_ida);
2788 2789 2790 2791 2792

	spin_lock(&dev_list_lock);
	list_add_tail(&ctrl->node, &nvme_ctrl_list);
	spin_unlock(&dev_list_lock);

2793 2794 2795 2796 2797 2798 2799 2800
	/*
	 * Initialize latency tolerance controls.  The sysfs files won't
	 * be visible to userspace unless the device actually supports APST.
	 */
	ctrl->device->power.set_latency_tolerance = nvme_set_latency_tolerance;
	dev_pm_qos_update_user_latency_tolerance(ctrl->device,
		min(default_ps_max_latency_us, (unsigned long)S32_MAX));

2801 2802
	return 0;
out_release_instance:
2803
	ida_simple_remove(&nvme_instance_ida, ctrl->instance);
2804 2805 2806
out:
	return ret;
}
2807
EXPORT_SYMBOL_GPL(nvme_init_ctrl);
2808

2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819
/**
 * nvme_kill_queues(): Ends all namespace queues
 * @ctrl: the dead controller that needs to end
 *
 * Call this function when the driver determines it is unable to get the
 * controller in a state capable of servicing IO.
 */
void nvme_kill_queues(struct nvme_ctrl *ctrl)
{
	struct nvme_ns *ns;

2820
	mutex_lock(&ctrl->namespaces_mutex);
M
Ming Lei 已提交
2821

2822
	/* Forcibly unquiesce queues to avoid blocking dispatch */
2823 2824
	if (ctrl->admin_q)
		blk_mq_unquiesce_queue(ctrl->admin_q);
2825

2826
	list_for_each_entry(ns, &ctrl->namespaces, list) {
2827 2828 2829 2830
		/*
		 * Revalidating a dead namespace sets capacity to 0. This will
		 * end buffered writers dirtying pages that can't be synced.
		 */
2831 2832 2833
		if (!ns->disk || test_and_set_bit(NVME_NS_DEAD, &ns->flags))
			continue;
		revalidate_disk(ns->disk);
2834
		blk_set_queue_dying(ns->queue);
2835

2836 2837
		/* Forcibly unquiesce queues to avoid blocking dispatch */
		blk_mq_unquiesce_queue(ns->queue);
2838
	}
2839
	mutex_unlock(&ctrl->namespaces_mutex);
2840
}
2841
EXPORT_SYMBOL_GPL(nvme_kill_queues);
2842

K
Keith Busch 已提交
2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884
void nvme_unfreeze(struct nvme_ctrl *ctrl)
{
	struct nvme_ns *ns;

	mutex_lock(&ctrl->namespaces_mutex);
	list_for_each_entry(ns, &ctrl->namespaces, list)
		blk_mq_unfreeze_queue(ns->queue);
	mutex_unlock(&ctrl->namespaces_mutex);
}
EXPORT_SYMBOL_GPL(nvme_unfreeze);

void nvme_wait_freeze_timeout(struct nvme_ctrl *ctrl, long timeout)
{
	struct nvme_ns *ns;

	mutex_lock(&ctrl->namespaces_mutex);
	list_for_each_entry(ns, &ctrl->namespaces, list) {
		timeout = blk_mq_freeze_queue_wait_timeout(ns->queue, timeout);
		if (timeout <= 0)
			break;
	}
	mutex_unlock(&ctrl->namespaces_mutex);
}
EXPORT_SYMBOL_GPL(nvme_wait_freeze_timeout);

void nvme_wait_freeze(struct nvme_ctrl *ctrl)
{
	struct nvme_ns *ns;

	mutex_lock(&ctrl->namespaces_mutex);
	list_for_each_entry(ns, &ctrl->namespaces, list)
		blk_mq_freeze_queue_wait(ns->queue);
	mutex_unlock(&ctrl->namespaces_mutex);
}
EXPORT_SYMBOL_GPL(nvme_wait_freeze);

void nvme_start_freeze(struct nvme_ctrl *ctrl)
{
	struct nvme_ns *ns;

	mutex_lock(&ctrl->namespaces_mutex);
	list_for_each_entry(ns, &ctrl->namespaces, list)
2885
		blk_freeze_queue_start(ns->queue);
K
Keith Busch 已提交
2886 2887 2888 2889
	mutex_unlock(&ctrl->namespaces_mutex);
}
EXPORT_SYMBOL_GPL(nvme_start_freeze);

2890
void nvme_stop_queues(struct nvme_ctrl *ctrl)
2891 2892 2893
{
	struct nvme_ns *ns;

2894
	mutex_lock(&ctrl->namespaces_mutex);
2895
	list_for_each_entry(ns, &ctrl->namespaces, list)
2896
		blk_mq_quiesce_queue(ns->queue);
2897
	mutex_unlock(&ctrl->namespaces_mutex);
2898
}
2899
EXPORT_SYMBOL_GPL(nvme_stop_queues);
2900

2901
void nvme_start_queues(struct nvme_ctrl *ctrl)
2902 2903 2904
{
	struct nvme_ns *ns;

2905
	mutex_lock(&ctrl->namespaces_mutex);
2906
	list_for_each_entry(ns, &ctrl->namespaces, list)
2907
		blk_mq_unquiesce_queue(ns->queue);
2908
	mutex_unlock(&ctrl->namespaces_mutex);
2909
}
2910
EXPORT_SYMBOL_GPL(nvme_start_queues);
2911

S
Sagi Grimberg 已提交
2912 2913 2914 2915 2916 2917 2918 2919 2920 2921
int nvme_reinit_tagset(struct nvme_ctrl *ctrl, struct blk_mq_tag_set *set)
{
	if (!ctrl->ops->reinit_request)
		return 0;

	return blk_mq_tagset_iter(set, set->driver_data,
			ctrl->ops->reinit_request);
}
EXPORT_SYMBOL_GPL(nvme_reinit_tagset);

2922 2923 2924 2925
int __init nvme_core_init(void)
{
	int result;

2926 2927 2928 2929 2930
	nvme_wq = alloc_workqueue("nvme-wq",
			WQ_UNBOUND | WQ_MEM_RECLAIM | WQ_SYSFS, 0);
	if (!nvme_wq)
		return -ENOMEM;

2931 2932 2933
	result = __register_chrdev(nvme_char_major, 0, NVME_MINORS, "nvme",
							&nvme_dev_fops);
	if (result < 0)
2934
		goto destroy_wq;
2935 2936 2937 2938 2939 2940 2941 2942 2943
	else if (result > 0)
		nvme_char_major = result;

	nvme_class = class_create(THIS_MODULE, "nvme");
	if (IS_ERR(nvme_class)) {
		result = PTR_ERR(nvme_class);
		goto unregister_chrdev;
	}

2944
	return 0;
2945

2946
unregister_chrdev:
2947
	__unregister_chrdev(nvme_char_major, 0, NVME_MINORS, "nvme");
2948 2949
destroy_wq:
	destroy_workqueue(nvme_wq);
2950
	return result;
2951 2952 2953 2954
}

void nvme_core_exit(void)
{
2955 2956
	class_destroy(nvme_class);
	__unregister_chrdev(nvme_char_major, 0, NVME_MINORS, "nvme");
2957
	destroy_workqueue(nvme_wq);
2958
}
2959 2960 2961 2962 2963

MODULE_LICENSE("GPL");
MODULE_VERSION("1.0");
module_init(nvme_core_init);
module_exit(nvme_core_exit);