core.c 72.6 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 39
unsigned char admin_timeout = 60;
module_param(admin_timeout, byte, 0644);
MODULE_PARM_DESC(admin_timeout, "timeout in seconds for admin commands");
40
EXPORT_SYMBOL_GPL(admin_timeout);
41 42 43 44

unsigned char nvme_io_timeout = 30;
module_param_named(io_timeout, nvme_io_timeout, byte, 0644);
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 struct class *nvme_class;

79 80 81 82 83
static __le32 nvme_get_log_dw10(u8 lid, size_t size)
{
	return cpu_to_le32((((size / 4) - 1) << 16) | lid);
}

84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103
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;
}

104
static blk_status_t nvme_error_status(struct request *req)
105 106 107
{
	switch (nvme_req(req)->status & 0x7ff) {
	case NVME_SC_SUCCESS:
108
		return BLK_STS_OK;
109
	case NVME_SC_CAP_EXCEEDED:
110
		return BLK_STS_NOSPC;
111
	case NVME_SC_ONCS_NOT_SUPPORTED:
112
		return BLK_STS_NOTSUPP;
113 114 115
	case NVME_SC_WRITE_FAULT:
	case NVME_SC_READ_ERROR:
	case NVME_SC_UNWRITTEN_BLOCK:
116 117 118
		return BLK_STS_MEDIUM;
	default:
		return BLK_STS_IOERR;
119 120 121
	}
}

122
static inline bool nvme_req_needs_retry(struct request *req)
123
{
124 125
	if (blk_noretry_request(req))
		return false;
126
	if (nvme_req(req)->status & NVME_SC_DNR)
127 128 129
		return false;
	if (jiffies - req->start_time >= req->timeout)
		return false;
130
	if (nvme_req(req)->retries >= nvme_max_retries)
131 132
		return false;
	return true;
133 134 135 136
}

void nvme_complete_rq(struct request *req)
{
137 138
	if (unlikely(nvme_req(req)->status && nvme_req_needs_retry(req))) {
		nvme_req(req)->retries++;
139
		blk_mq_requeue_request(req, true);
140
		return;
141 142
	}

143
	blk_mq_end_request(req, nvme_error_status(req));
144 145 146
}
EXPORT_SYMBOL_GPL(nvme_complete_rq);

147 148 149 150 151 152 153 154 155 156 157 158 159
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;
160
	nvme_req(req)->status = status;
161
	blk_mq_complete_request(req);
162

163 164 165
}
EXPORT_SYMBOL_GPL(nvme_cancel_request);

166 167 168
bool nvme_change_ctrl_state(struct nvme_ctrl *ctrl,
		enum nvme_ctrl_state new_state)
{
169
	enum nvme_ctrl_state old_state;
170 171 172
	bool changed = false;

	spin_lock_irq(&ctrl->lock);
173 174

	old_state = ctrl->state;
175 176 177
	switch (new_state) {
	case NVME_CTRL_LIVE:
		switch (old_state) {
178
		case NVME_CTRL_NEW:
179
		case NVME_CTRL_RESETTING:
180
		case NVME_CTRL_RECONNECTING:
181 182 183 184 185 186 187 188 189
			changed = true;
			/* FALLTHRU */
		default:
			break;
		}
		break;
	case NVME_CTRL_RESETTING:
		switch (old_state) {
		case NVME_CTRL_NEW:
190 191 192 193 194 195 196 197 198
		case NVME_CTRL_LIVE:
			changed = true;
			/* FALLTHRU */
		default:
			break;
		}
		break;
	case NVME_CTRL_RECONNECTING:
		switch (old_state) {
199 200 201 202 203 204 205 206 207 208 209
		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:
210
		case NVME_CTRL_RECONNECTING:
211 212 213 214 215 216
			changed = true;
			/* FALLTHRU */
		default:
			break;
		}
		break;
217 218 219 220 221 222 223 224 225
	case NVME_CTRL_DEAD:
		switch (old_state) {
		case NVME_CTRL_DELETING:
			changed = true;
			/* FALLTHRU */
		default:
			break;
		}
		break;
226 227 228 229 230 231 232
	default:
		break;
	}

	if (changed)
		ctrl->state = new_state;

233 234
	spin_unlock_irq(&ctrl->lock);

235 236 237 238
	return changed;
}
EXPORT_SYMBOL_GPL(nvme_change_ctrl_state);

239 240 241 242
static void nvme_free_ns(struct kref *kref)
{
	struct nvme_ns *ns = container_of(kref, struct nvme_ns, kref);

243 244
	if (ns->ndev)
		nvme_nvm_unregister(ns);
245

246 247 248 249 250
	if (ns->disk) {
		spin_lock(&dev_list_lock);
		ns->disk->private_data = NULL;
		spin_unlock(&dev_list_lock);
	}
251 252

	put_disk(ns->disk);
253 254
	ida_simple_remove(&ns->ctrl->ns_ida, ns->instance);
	nvme_put_ctrl(ns->ctrl);
255 256 257
	kfree(ns);
}

258
static void nvme_put_ns(struct nvme_ns *ns)
259 260 261 262 263 264 265 266 267 268
{
	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;
269 270 271 272 273 274
	if (ns) {
		if (!kref_get_unless_zero(&ns->kref))
			goto fail;
		if (!try_module_get(ns->ctrl->ops->module))
			goto fail_put_ns;
	}
275 276 277
	spin_unlock(&dev_list_lock);

	return ns;
278 279 280 281 282 283

fail_put_ns:
	kref_put(&ns->kref, nvme_free_ns);
fail:
	spin_unlock(&dev_list_lock);
	return NULL;
284 285
}

286
struct request *nvme_alloc_request(struct request_queue *q,
287
		struct nvme_command *cmd, unsigned int flags, int qid)
288
{
289
	unsigned op = nvme_is_write(cmd) ? REQ_OP_DRV_OUT : REQ_OP_DRV_IN;
290 291
	struct request *req;

292
	if (qid == NVME_QID_ANY) {
293
		req = blk_mq_alloc_request(q, op, flags);
294
	} else {
295
		req = blk_mq_alloc_request_hctx(q, op, flags,
296 297
				qid ? qid - 1 : 0);
	}
298
	if (IS_ERR(req))
299
		return req;
300 301

	req->cmd_flags |= REQ_FAILFAST_DRIVER;
302
	nvme_req(req)->cmd = cmd;
303

304 305
	return req;
}
306
EXPORT_SYMBOL_GPL(nvme_alloc_request);
307

308 309 310 311 312 313 314
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
315
	c.directive.nsid = cpu_to_le32(NVME_NSID_ALL);
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
	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);
344
	c.directive.numd = cpu_to_le32((sizeof(*s) >> 2) - 1);
345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364
	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
365
	ret = nvme_get_stream_params(ctrl, &s, NVME_NSID_ALL);
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
	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
407 408 409 410 411 412 413 414
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);
}

415
static blk_status_t nvme_setup_discard(struct nvme_ns *ns, struct request *req,
M
Ming Lin 已提交
416 417
		struct nvme_command *cmnd)
{
418
	unsigned short segments = blk_rq_nr_discard_segments(req), n = 0;
M
Ming Lin 已提交
419
	struct nvme_dsm_range *range;
420
	struct bio *bio;
M
Ming Lin 已提交
421

422
	range = kmalloc_array(segments, sizeof(*range), GFP_ATOMIC);
M
Ming Lin 已提交
423
	if (!range)
424
		return BLK_STS_RESOURCE;
M
Ming Lin 已提交
425

426 427 428 429 430 431 432 433 434 435 436 437
	__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);
438
		return BLK_STS_IOERR;
439
	}
M
Ming Lin 已提交
440 441 442 443

	memset(cmnd, 0, sizeof(*cmnd));
	cmnd->dsm.opcode = nvme_cmd_dsm;
	cmnd->dsm.nsid = cpu_to_le32(ns->ns_id);
444
	cmnd->dsm.nr = cpu_to_le32(segments - 1);
M
Ming Lin 已提交
445 446
	cmnd->dsm.attributes = cpu_to_le32(NVME_DSMGMT_AD);

447 448
	req->special_vec.bv_page = virt_to_page(range);
	req->special_vec.bv_offset = offset_in_page(range);
449
	req->special_vec.bv_len = sizeof(*range) * segments;
450
	req->rq_flags |= RQF_SPECIAL_PAYLOAD;
M
Ming Lin 已提交
451

452
	return BLK_STS_OK;
M
Ming Lin 已提交
453 454
}

455 456
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
457
{
458
	struct nvme_ctrl *ctrl = ns->ctrl;
M
Ming Lin 已提交
459 460 461
	u16 control = 0;
	u32 dsmgmt = 0;

462 463 464 465 466
	/*
	 * 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.
	 */
467 468
	if (ns && ns->ms &&
	    (!ns->pi_type || ns->ms != sizeof(struct t10_pi_tuple)) &&
469 470 471
	    !blk_integrity_rq(req) && !blk_rq_is_passthrough(req))
		return BLK_STS_NOTSUPP;

M
Ming Lin 已提交
472 473 474 475 476 477 478 479 480 481 482 483 484 485
	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);

486 487 488
	if (req_op(req) == REQ_OP_WRITE && ctrl->nr_streams)
		nvme_assign_write_stream(ctrl, req, &control, &dsmgmt);

M
Ming Lin 已提交
489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507
	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);
508
	return 0;
M
Ming Lin 已提交
509 510
}

511
blk_status_t nvme_setup_cmd(struct nvme_ns *ns, struct request *req,
M
Ming Lin 已提交
512 513
		struct nvme_command *cmd)
{
514
	blk_status_t ret = BLK_STS_OK;
M
Ming Lin 已提交
515

516
	if (!(req->rq_flags & RQF_DONTPREP)) {
517
		nvme_req(req)->retries = 0;
518
		nvme_req(req)->flags = 0;
519 520 521
		req->rq_flags |= RQF_DONTPREP;
	}

522 523 524
	switch (req_op(req)) {
	case REQ_OP_DRV_IN:
	case REQ_OP_DRV_OUT:
525
		memcpy(cmd, nvme_req(req)->cmd, sizeof(*cmd));
526 527
		break;
	case REQ_OP_FLUSH:
M
Ming Lin 已提交
528
		nvme_setup_flush(ns, cmd);
529
		break;
530 531
	case REQ_OP_WRITE_ZEROES:
		/* currently only aliased to deallocate for a few ctrls: */
532
	case REQ_OP_DISCARD:
M
Ming Lin 已提交
533
		ret = nvme_setup_discard(ns, req, cmd);
534 535 536
		break;
	case REQ_OP_READ:
	case REQ_OP_WRITE:
537
		ret = nvme_setup_rw(ns, req, cmd);
538 539 540
		break;
	default:
		WARN_ON_ONCE(1);
541
		return BLK_STS_IOERR;
542
	}
M
Ming Lin 已提交
543

544
	cmd->common.command_id = req->tag;
M
Ming Lin 已提交
545 546 547 548
	return ret;
}
EXPORT_SYMBOL_GPL(nvme_setup_cmd);

549 550 551 552 553
/*
 * 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,
554
		union nvme_result *result, void *buffer, unsigned bufflen,
555
		unsigned timeout, int qid, int at_head, int flags)
556 557 558 559
{
	struct request *req;
	int ret;

560
	req = nvme_alloc_request(q, cmd, flags, qid);
561 562 563 564 565
	if (IS_ERR(req))
		return PTR_ERR(req);

	req->timeout = timeout ? timeout : ADMIN_TIMEOUT;

566 567 568 569
	if (buffer && bufflen) {
		ret = blk_rq_map_kern(q, req, buffer, bufflen, GFP_KERNEL);
		if (ret)
			goto out;
570 571
	}

572
	blk_execute_rq(req->q, NULL, req, at_head);
573 574
	if (result)
		*result = nvme_req(req)->result;
575 576 577 578
	if (nvme_req(req)->flags & NVME_REQ_CANCELLED)
		ret = -EINTR;
	else
		ret = nvme_req(req)->status;
579 580 581 582
 out:
	blk_mq_free_request(req);
	return ret;
}
583
EXPORT_SYMBOL_GPL(__nvme_submit_sync_cmd);
584 585 586 587

int nvme_submit_sync_cmd(struct request_queue *q, struct nvme_command *cmd,
		void *buffer, unsigned bufflen)
{
588 589
	return __nvme_submit_sync_cmd(q, cmd, NULL, buffer, bufflen, 0,
			NVME_QID_ANY, 0, 0);
590
}
591
EXPORT_SYMBOL_GPL(nvme_submit_sync_cmd);
592

593 594 595 596
int __nvme_submit_user_cmd(struct request_queue *q, struct nvme_command *cmd,
		void __user *ubuffer, unsigned bufflen,
		void __user *meta_buffer, unsigned meta_len, u32 meta_seed,
		u32 *result, unsigned timeout)
597
{
598
	bool write = nvme_is_write(cmd);
599 600
	struct nvme_ns *ns = q->queuedata;
	struct gendisk *disk = ns ? ns->disk : NULL;
601
	struct request *req;
602 603
	struct bio *bio = NULL;
	void *meta = NULL;
604 605
	int ret;

606
	req = nvme_alloc_request(q, cmd, 0, NVME_QID_ANY);
607 608 609 610 611 612
	if (IS_ERR(req))
		return PTR_ERR(req);

	req->timeout = timeout ? timeout : ADMIN_TIMEOUT;

	if (ubuffer && bufflen) {
613 614 615 616 617 618
		ret = blk_rq_map_user(q, req, NULL, ubuffer, bufflen,
				GFP_KERNEL);
		if (ret)
			goto out;
		bio = req->bio;

619 620
		if (!disk)
			goto submit;
621
		bio->bi_disk = disk;
622

623
		if (meta_buffer && meta_len) {
624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640
			struct bio_integrity_payload *bip;

			meta = kmalloc(meta_len, GFP_KERNEL);
			if (!meta) {
				ret = -ENOMEM;
				goto out_unmap;
			}

			if (write) {
				if (copy_from_user(meta, meta_buffer,
						meta_len)) {
					ret = -EFAULT;
					goto out_free_meta;
				}
			}

			bip = bio_integrity_alloc(bio, GFP_KERNEL, 1);
641 642
			if (IS_ERR(bip)) {
				ret = PTR_ERR(bip);
643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658
				goto out_free_meta;
			}

			bip->bip_iter.bi_size = meta_len;
			bip->bip_iter.bi_sector = meta_seed;

			ret = bio_integrity_add_page(bio, virt_to_page(meta),
					meta_len, offset_in_page(meta));
			if (ret != meta_len) {
				ret = -ENOMEM;
				goto out_free_meta;
			}
		}
	}
 submit:
	blk_execute_rq(req->q, disk, req, 0);
659 660 661 662
	if (nvme_req(req)->flags & NVME_REQ_CANCELLED)
		ret = -EINTR;
	else
		ret = nvme_req(req)->status;
663
	if (result)
664
		*result = le32_to_cpu(nvme_req(req)->result.u32);
665 666 667 668 669 670 671
	if (meta && !ret && !write) {
		if (copy_to_user(meta_buffer, meta, meta_len))
			ret = -EFAULT;
	}
 out_free_meta:
	kfree(meta);
 out_unmap:
672
	if (bio)
673
		blk_rq_unmap_user(bio);
674 675 676 677 678
 out:
	blk_mq_free_request(req);
	return ret;
}

679 680 681 682 683 684 685 686
int nvme_submit_user_cmd(struct request_queue *q, struct nvme_command *cmd,
		void __user *ubuffer, unsigned bufflen, u32 *result,
		unsigned timeout)
{
	return __nvme_submit_user_cmd(q, cmd, ubuffer, bufflen, NULL, 0, 0,
			result, timeout);
}

687
static void nvme_keep_alive_end_io(struct request *rq, blk_status_t status)
S
Sagi Grimberg 已提交
688 689 690 691 692
{
	struct nvme_ctrl *ctrl = rq->end_io_data;

	blk_mq_free_request(rq);

693
	if (status) {
S
Sagi Grimberg 已提交
694
		dev_err(ctrl->device,
695 696
			"failed nvme_keep_alive_end_io error=%d\n",
				status);
S
Sagi Grimberg 已提交
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
		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");
732
		nvme_reset_ctrl(ctrl);
S
Sagi Grimberg 已提交
733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755
		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
756
static int nvme_identify_ctrl(struct nvme_ctrl *dev, struct nvme_id_ctrl **id)
757 758 759 760 761 762
{
	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;
763
	c.identify.cns = NVME_ID_CNS_CTRL;
764 765 766 767 768 769 770 771 772 773 774 775

	*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;
}

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
static int nvme_identify_ns_descs(struct nvme_ns *ns, unsigned nsid)
{
	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;

	status = nvme_submit_sync_cmd(ns->ctrl->admin_q, &c, data,
				      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) {
				dev_warn(ns->ctrl->device,
					 "ctrl returned bogus length: %d for NVME_NIDT_EUI64\n",
					 cur->nidl);
				goto free_data;
			}
			len = NVME_NIDT_EUI64_LEN;
			memcpy(ns->eui, data + pos + sizeof(*cur), len);
			break;
		case NVME_NIDT_NGUID:
			if (cur->nidl != NVME_NIDT_NGUID_LEN) {
				dev_warn(ns->ctrl->device,
					 "ctrl returned bogus length: %d for NVME_NIDT_NGUID\n",
					 cur->nidl);
				goto free_data;
			}
			len = NVME_NIDT_NGUID_LEN;
			memcpy(ns->nguid, data + pos + sizeof(*cur), len);
			break;
		case NVME_NIDT_UUID:
			if (cur->nidl != NVME_NIDT_UUID_LEN) {
				dev_warn(ns->ctrl->device,
					 "ctrl returned bogus length: %d for NVME_NIDT_UUID\n",
					 cur->nidl);
				goto free_data;
			}
			len = NVME_NIDT_UUID_LEN;
			uuid_copy(&ns->uuid, data + pos + sizeof(*cur));
			break;
		default:
			/* Skip unnkown types */
			len = cur->nidl;
			break;
		}

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

847 848 849 850 851
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;
852
	c.identify.cns = NVME_ID_CNS_NS_ACTIVE_LIST;
853 854 855 856
	c.identify.nsid = cpu_to_le32(nsid);
	return nvme_submit_sync_cmd(dev->admin_q, &c, ns_list, 0x1000);
}

K
Keith Busch 已提交
857
static int nvme_identify_ns(struct nvme_ctrl *dev, unsigned nsid,
858 859 860 861 862 863
		struct nvme_id_ns **id)
{
	struct nvme_command c = { };
	int error;

	/* gcc-4.4.4 (at least) has issues with initializers and anon unions */
864 865
	c.identify.opcode = nvme_admin_identify;
	c.identify.nsid = cpu_to_le32(nsid);
866
	c.identify.cns = NVME_ID_CNS_NS;
867 868 869 870 871 872 873 874 875 876 877 878

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

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

K
Keith Busch 已提交
879
static int nvme_set_features(struct nvme_ctrl *dev, unsigned fid, unsigned dword11,
880
		      void *buffer, size_t buflen, u32 *result)
881 882
{
	struct nvme_command c;
883
	union nvme_result res;
884
	int ret;
885 886 887 888 889 890

	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);

891
	ret = __nvme_submit_sync_cmd(dev->admin_q, &c, &res,
892
			buffer, buflen, 0, NVME_QID_ANY, 0, 0);
893
	if (ret >= 0 && result)
894
		*result = le32_to_cpu(res.u32);
895
	return ret;
896 897
}

C
Christoph Hellwig 已提交
898 899 900 901 902 903
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;

904
	status = nvme_set_features(ctrl, NVME_FEAT_NUM_QUEUES, q_count, NULL, 0,
C
Christoph Hellwig 已提交
905
			&result);
906
	if (status < 0)
C
Christoph Hellwig 已提交
907 908
		return status;

909 910 911 912 913 914
	/*
	 * 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) {
915
		dev_err(ctrl->device, "Could not set queue count (%d)\n", status);
916 917 918 919 920 921
		*count = 0;
	} else {
		nr_io_queues = min(result & 0xffff, result >> 16) + 1;
		*count = min(*count, nr_io_queues);
	}

C
Christoph Hellwig 已提交
922 923
	return 0;
}
924
EXPORT_SYMBOL_GPL(nvme_set_queue_count);
C
Christoph Hellwig 已提交
925

926 927 928 929 930 931 932 933 934
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;
935 936
	if (io.flags)
		return -EINVAL;
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

	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);

	return __nvme_submit_user_cmd(ns->queue, &c,
			(void __user *)(uintptr_t)io.addr, length,
			metadata, meta_len, io.slba, NULL, 0);
}

976
static int nvme_user_cmd(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
977 978 979 980 981 982 983 984 985 986 987
			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;
988 989
	if (cmd.flags)
		return -EINVAL;
990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007

	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,
1008
			(void __user *)(uintptr_t)cmd.addr, cmd.data_len,
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
			&cmd.result, timeout);
	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:
1034 1035 1036 1037
#ifdef CONFIG_NVM
		if (ns->ndev)
			return nvme_nvm_ioctl(ns, cmd, arg);
#endif
1038
		if (is_sed_ioctl(cmd))
1039
			return sed_ioctl(ns->ctrl->opal_dev, cmd,
1040
					 (void __user *) arg);
1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061
		return -ENOTTY;
	}
}

#ifdef CONFIG_COMPAT
static int nvme_compat_ioctl(struct block_device *bdev, fmode_t mode,
			unsigned int cmd, unsigned long arg)
{
	return nvme_ioctl(bdev, mode, cmd, arg);
}
#else
#define nvme_compat_ioctl	NULL
#endif

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)
{
1062 1063 1064 1065
	struct nvme_ns *ns = disk->private_data;

	module_put(ns->ctrl->ops->module);
	nvme_put_ns(ns);
1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077
}

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
1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100
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;
}

1101 1102 1103 1104
static void nvme_init_integrity(struct nvme_ns *ns)
{
	struct blk_integrity integrity;

1105
	memset(&integrity, 0, sizeof(integrity));
1106 1107 1108
	switch (ns->pi_type) {
	case NVME_NS_DPS_PI_TYPE3:
		integrity.profile = &t10_pi_type3_crc;
1109 1110
		integrity.tag_size = sizeof(u16) + sizeof(u32);
		integrity.flags |= BLK_INTEGRITY_DEVICE_CAPABLE;
1111 1112 1113 1114
		break;
	case NVME_NS_DPS_PI_TYPE1:
	case NVME_NS_DPS_PI_TYPE2:
		integrity.profile = &t10_pi_type1_crc;
1115 1116
		integrity.tag_size = sizeof(u16);
		integrity.flags |= BLK_INTEGRITY_DEVICE_CAPABLE;
1117 1118 1119 1120 1121 1122 1123 1124 1125 1126
		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
1127 1128 1129 1130
static void nvme_prep_integrity(struct gendisk *disk, struct nvme_id_ns *id,
		u16 bs)
{
}
1131 1132 1133 1134 1135
static void nvme_init_integrity(struct nvme_ns *ns)
{
}
#endif /* CONFIG_BLK_DEV_INTEGRITY */

1136 1137 1138 1139 1140 1141
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));
}

1142 1143
static void nvme_config_discard(struct nvme_ns *ns)
{
1144
	struct nvme_ctrl *ctrl = ns->ctrl;
1145
	u32 logical_block_size = queue_logical_block_size(ns->queue);
1146

1147 1148 1149
	BUILD_BUG_ON(PAGE_SIZE / sizeof(struct nvme_dsm_range) <
			NVME_DSM_MAX_RANGES);

1150 1151 1152 1153 1154 1155 1156 1157 1158
	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;
	}
1159
	blk_queue_max_discard_sectors(ns->queue, UINT_MAX);
1160
	blk_queue_max_discard_segments(ns->queue, NVME_DSM_MAX_RANGES);
1161
	queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, ns->queue);
1162 1163 1164

	if (ctrl->quirks & NVME_QUIRK_DEALLOCATE_ZEROES)
		blk_queue_max_write_zeroes_sectors(ns->queue, UINT_MAX);
1165 1166
}

1167
static int nvme_revalidate_ns(struct nvme_ns *ns, struct nvme_id_ns **id)
1168
{
1169
	if (nvme_identify_ns(ns->ctrl, ns->ns_id, id)) {
1170
		dev_warn(ns->ctrl->device, "Identify namespace failed\n");
1171 1172 1173
		return -ENODEV;
	}

1174 1175 1176
	if ((*id)->ncap == 0) {
		kfree(*id);
		return -ENODEV;
1177 1178
	}

1179
	if (ns->ctrl->vs >= NVME_VS(1, 1, 0))
1180
		memcpy(ns->eui, (*id)->eui64, sizeof(ns->eui));
1181
	if (ns->ctrl->vs >= NVME_VS(1, 2, 0))
1182
		memcpy(ns->nguid, (*id)->nguid, sizeof(ns->nguid));
1183 1184 1185 1186 1187 1188 1189 1190
	if (ns->ctrl->vs >= NVME_VS(1, 3, 0)) {
		 /* Don't treat error as fatal we potentially
		  * already have a NGUID or EUI-64
		  */
		if (nvme_identify_ns_descs(ns, ns->ns_id))
			dev_warn(ns->ctrl->device,
				 "%s: Identify Descriptors failed\n", __func__);
	}
1191 1192 1193 1194 1195 1196 1197

	return 0;
}

static void __nvme_revalidate_disk(struct gendisk *disk, struct nvme_id_ns *id)
{
	struct nvme_ns *ns = disk->private_data;
1198
	struct nvme_ctrl *ctrl = ns->ctrl;
1199
	u16 bs;
1200 1201 1202 1203 1204

	/*
	 * If identify namespace failed, use default 512 byte block size so
	 * block layer can use before failing read/write for 0 capacity.
	 */
1205
	ns->lba_shift = id->lbaf[id->flbas & NVME_NS_FLBAS_LBA_MASK].ds;
1206 1207 1208
	if (ns->lba_shift == 0)
		ns->lba_shift = 9;
	bs = 1 << ns->lba_shift;
1209
	ns->noiob = le16_to_cpu(id->noiob);
1210 1211 1212

	blk_mq_freeze_queue(disk->queue);

1213
	if (ctrl->ops->flags & NVME_F_METADATA_SUPPORTED)
1214
		nvme_prep_integrity(disk, id, bs);
1215
	blk_queue_logical_block_size(ns->queue, bs);
1216 1217
	if (ns->noiob)
		nvme_set_chunk_size(ns);
K
Keith Busch 已提交
1218
	if (ns->ms && !blk_get_integrity(disk) && !ns->ext)
1219 1220 1221 1222 1223 1224
		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));

1225
	if (ctrl->oncs & NVME_CTRL_ONCS_DSM)
1226 1227
		nvme_config_discard(ns);
	blk_mq_unfreeze_queue(disk->queue);
1228
}
1229

1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245
static int nvme_revalidate_disk(struct gendisk *disk)
{
	struct nvme_ns *ns = disk->private_data;
	struct nvme_id_ns *id = NULL;
	int ret;

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

	ret = nvme_revalidate_ns(ns, &id);
	if (ret)
		return ret;

	__nvme_revalidate_disk(disk, id);
1246
	kfree(id);
1247

1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324
	return 0;
}

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)
{
1325
	u32 cdw10 = 1 | (key ? 1 << 3 : 0);
1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342
	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,
};

1343
#ifdef CONFIG_BLK_SED_OPAL
1344 1345
int nvme_sec_submit(void *data, u16 spsp, u8 secp, void *buffer, size_t len,
		bool send)
1346
{
1347
	struct nvme_ctrl *ctrl = data;
1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364
	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 */

1365
static const struct block_device_operations nvme_fops = {
1366 1367 1368 1369 1370 1371 1372 1373 1374 1375
	.owner		= THIS_MODULE,
	.ioctl		= nvme_ioctl,
	.compat_ioctl	= nvme_compat_ioctl,
	.open		= nvme_open,
	.release	= nvme_release,
	.getgeo		= nvme_getgeo,
	.revalidate_disk= nvme_revalidate_disk,
	.pr_ops		= &nvme_pr_ops,
};

1376 1377 1378 1379 1380 1381 1382 1383
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
1384 1385
		if (csts == ~0)
			return -ENODEV;
1386 1387 1388 1389 1390 1391 1392
		if ((csts & NVME_CSTS_RDY) == bit)
			break;

		msleep(100);
		if (fatal_signal_pending(current))
			return -EINTR;
		if (time_after(jiffies, timeout)) {
1393
			dev_err(ctrl->device,
1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418
				"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;
1419

1420
	if (ctrl->quirks & NVME_QUIRK_DELAY_BEFORE_CHK_RDY)
1421 1422
		msleep(NVME_QUIRK_DELAY_AMOUNT);

1423 1424
	return nvme_wait_ready(ctrl, cap, false);
}
1425
EXPORT_SYMBOL_GPL(nvme_disable_ctrl);
1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437

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) {
1438
		dev_err(ctrl->device,
1439 1440 1441 1442 1443 1444 1445 1446 1447
			"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;
1448
	ctrl->ctrl_config |= NVME_CC_AMS_RR | NVME_CC_SHN_NONE;
1449 1450 1451 1452 1453 1454 1455 1456
	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);
}
1457
EXPORT_SYMBOL_GPL(nvme_enable_ctrl);
1458 1459 1460

int nvme_shutdown_ctrl(struct nvme_ctrl *ctrl)
{
1461
	unsigned long timeout = jiffies + (shutdown_timeout * HZ);
1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479
	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)) {
1480
			dev_err(ctrl->device,
1481 1482 1483 1484 1485 1486 1487
				"Device shutdown incomplete; abort shutdown\n");
			return -ENODEV;
		}
	}

	return ret;
}
1488
EXPORT_SYMBOL_GPL(nvme_shutdown_ctrl);
1489

1490 1491 1492
static void nvme_set_queue_limits(struct nvme_ctrl *ctrl,
		struct request_queue *q)
{
1493 1494
	bool vwc = false;

1495
	if (ctrl->max_hw_sectors) {
1496 1497 1498
		u32 max_segments =
			(ctrl->max_hw_sectors / (ctrl->page_size >> 9)) + 1;

1499
		blk_queue_max_hw_sectors(q, ctrl->max_hw_sectors);
1500
		blk_queue_max_segments(q, min_t(u32, max_segments, USHRT_MAX));
1501
	}
K
Keith Busch 已提交
1502 1503
	if (ctrl->quirks & NVME_QUIRK_STRIPE_SIZE)
		blk_queue_chunk_sectors(q, ctrl->max_hw_sectors);
1504
	blk_queue_virt_boundary(q, ctrl->page_size - 1);
1505 1506 1507
	if (ctrl->vwc & NVME_CTRL_VWC_PRESENT)
		vwc = true;
	blk_queue_write_cache(q, vwc, vwc);
1508 1509
}

1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526
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;
}

1527
static int nvme_configure_apst(struct nvme_ctrl *ctrl)
1528 1529 1530 1531 1532 1533 1534 1535
{
	/*
	 * 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
1536
	 * non-operational state after waiting 50 * (enlat + exlat)
1537
	 * microseconds, as long as that state's exit latency is under
1538 1539 1540 1541 1542 1543 1544 1545 1546
	 * 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;
1547 1548
	u64 max_lat_us = 0;
	int max_ps = -1;
1549 1550 1551 1552 1553 1554 1555
	int ret;

	/*
	 * If APST isn't supported or if we haven't been initialized yet,
	 * then don't do anything.
	 */
	if (!ctrl->apsta)
1556
		return 0;
1557 1558 1559

	if (ctrl->npss > 31) {
		dev_warn(ctrl->device, "NPSS is invalid; not using APST\n");
1560
		return 0;
1561 1562 1563 1564
	}

	table = kzalloc(sizeof(*table), GFP_KERNEL);
	if (!table)
1565
		return 0;
1566

1567
	if (!ctrl->apst_enabled || ctrl->ps_max_latency_us == 0) {
1568 1569
		/* Turn off APST. */
		apste = 0;
1570
		dev_dbg(ctrl->device, "APST disabled\n");
1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581
	} 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--) {
1582
			u64 total_latency_us, exit_latency_us, transition_ms;
1583 1584 1585 1586

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

1587 1588 1589 1590 1591 1592 1593 1594
			/*
			 * 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;

1595 1596 1597 1598 1599 1600 1601 1602
			/*
			 * 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;

1603 1604 1605
			exit_latency_us =
				(u64)le32_to_cpu(ctrl->psd[state].exit_lat);
			if (exit_latency_us > ctrl->ps_max_latency_us)
1606 1607
				continue;

1608 1609 1610 1611
			total_latency_us =
				exit_latency_us +
				le32_to_cpu(ctrl->psd[state].entry_lat);

1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622
			/*
			 * 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));
1623 1624 1625 1626 1627 1628

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

			if (total_latency_us > max_lat_us)
				max_lat_us = total_latency_us;
1629 1630 1631
		}

		apste = 1;
1632 1633 1634 1635 1636 1637 1638

		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);
		}
1639 1640 1641 1642 1643 1644 1645 1646
	}

	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);
1647
	return ret;
1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670
}

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);
	}
}

1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683
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[] = {
1684
	{
1685 1686 1687 1688 1689 1690
		/*
		 * 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",
1691
		.quirks = NVME_QUIRK_NO_APST,
1692
	}
1693 1694 1695 1696 1697 1698 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
};

/* 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));
}

1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748
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);
}

1749 1750 1751 1752 1753 1754 1755 1756 1757 1758
/*
 * 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;
1759
	u32 max_hw_sectors;
1760
	bool prev_apst_enabled;
1761

1762 1763
	ret = ctrl->ops->reg_read32(ctrl, NVME_REG_VS, &ctrl->vs);
	if (ret) {
1764
		dev_err(ctrl->device, "Reading VS failed (%d)\n", ret);
1765 1766 1767
		return ret;
	}

1768 1769
	ret = ctrl->ops->reg_read64(ctrl, NVME_REG_CAP, &cap);
	if (ret) {
1770
		dev_err(ctrl->device, "Reading CAP failed (%d)\n", ret);
1771 1772 1773 1774
		return ret;
	}
	page_shift = NVME_CAP_MPSMIN(cap) + 12;

1775
	if (ctrl->vs >= NVME_VS(1, 1, 0))
1776 1777
		ctrl->subsystem = NVME_CAP_NSSRC(cap);

1778 1779
	ret = nvme_identify_ctrl(ctrl, &id);
	if (ret) {
1780
		dev_err(ctrl->device, "Identify Controller failed (%d)\n", ret);
1781 1782 1783
		return -EIO;
	}

1784 1785
	nvme_init_subnqn(ctrl, id);

1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803
	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;
		}
	}

1804
	if (force_apst && (ctrl->quirks & NVME_QUIRK_NO_DEEPEST_PS)) {
1805
		dev_warn(ctrl->device, "forcibly allowing all power states due to nvme_core.force_apst -- use at your own risk\n");
1806 1807 1808
		ctrl->quirks &= ~NVME_QUIRK_NO_DEEPEST_PS;
	}

1809
	ctrl->oacs = le16_to_cpu(id->oacs);
1810
	ctrl->vid = le16_to_cpu(id->vid);
1811
	ctrl->oncs = le16_to_cpup(&id->oncs);
1812
	atomic_set(&ctrl->abort_limit, id->acl + 1);
1813
	ctrl->vwc = id->vwc;
M
Ming Lin 已提交
1814
	ctrl->cntlid = le16_to_cpup(&id->cntlid);
1815 1816 1817 1818
	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)
1819
		max_hw_sectors = 1 << (id->mdts + page_shift - 9);
1820
	else
1821 1822 1823
		max_hw_sectors = UINT_MAX;
	ctrl->max_hw_sectors =
		min_not_zero(ctrl->max_hw_sectors, max_hw_sectors);
1824

1825
	nvme_set_queue_limits(ctrl, ctrl->admin_q);
1826
	ctrl->sgls = le32_to_cpu(id->sgls);
S
Sagi Grimberg 已提交
1827
	ctrl->kas = le16_to_cpu(id->kas);
1828

1829
	ctrl->npss = id->npss;
1830 1831
	ctrl->apsta = id->apsta;
	prev_apst_enabled = ctrl->apst_enabled;
1832 1833
	if (ctrl->quirks & NVME_QUIRK_NO_APST) {
		if (force_apst && id->apsta) {
1834
			dev_warn(ctrl->device, "forcibly allowing APST due to nvme_core.force_apst -- use at your own risk\n");
1835
			ctrl->apst_enabled = true;
1836
		} else {
1837
			ctrl->apst_enabled = false;
1838 1839
		}
	} else {
1840
		ctrl->apst_enabled = id->apsta;
1841
	}
1842 1843
	memcpy(ctrl->psd, id->psd, sizeof(ctrl->psd));

1844
	if (ctrl->ops->flags & NVME_F_FABRICS) {
1845 1846 1847 1848 1849 1850 1851 1852 1853
		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
		 */
1854
		if (ctrl->cntlid != le16_to_cpu(id->cntlid)) {
1855
			ret = -EINVAL;
1856 1857
			goto out_free;
		}
S
Sagi Grimberg 已提交
1858 1859

		if (!ctrl->opts->discovery_nqn && !ctrl->kas) {
1860
			dev_err(ctrl->device,
S
Sagi Grimberg 已提交
1861 1862
				"keep-alive support is mandatory for fabrics\n");
			ret = -EINVAL;
1863
			goto out_free;
S
Sagi Grimberg 已提交
1864
		}
1865 1866
	} else {
		ctrl->cntlid = le16_to_cpu(id->cntlid);
1867 1868
		ctrl->hmpre = le32_to_cpu(id->hmpre);
		ctrl->hmmin = le32_to_cpu(id->hmmin);
1869
	}
1870

1871
	kfree(id);
1872

1873
	if (ctrl->apst_enabled && !prev_apst_enabled)
1874
		dev_pm_qos_expose_latency_tolerance(ctrl->device);
1875
	else if (!ctrl->apst_enabled && prev_apst_enabled)
1876 1877
		dev_pm_qos_hide_latency_tolerance(ctrl->device);

1878 1879 1880
	ret = nvme_configure_apst(ctrl);
	if (ret < 0)
		return ret;
1881 1882 1883 1884
	
	ret = nvme_configure_timestamp(ctrl);
	if (ret < 0)
		return ret;
1885 1886 1887 1888

	ret = nvme_configure_directives(ctrl);
	if (ret < 0)
		return ret;
1889

1890
	ctrl->identified = true;
1891

1892 1893 1894 1895
	return 0;

out_free:
	kfree(id);
1896
	return ret;
1897
}
1898
EXPORT_SYMBOL_GPL(nvme_init_identify);
1899

1900
static int nvme_dev_open(struct inode *inode, struct file *file)
1901
{
1902 1903 1904
	struct nvme_ctrl *ctrl;
	int instance = iminor(inode);
	int ret = -ENODEV;
1905

1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923
	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;
1924 1925
}

1926
static int nvme_dev_release(struct inode *inode, struct file *file)
1927
{
1928 1929 1930 1931
	nvme_put_ctrl(file->private_data);
	return 0;
}

1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944
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)) {
1945
		dev_warn(ctrl->device,
1946 1947 1948 1949 1950
			"NVME_IOCTL_IO_CMD not supported when multiple namespaces present!\n");
		ret = -EINVAL;
		goto out_unlock;
	}

1951
	dev_warn(ctrl->device,
1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964
		"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;
}

1965 1966 1967 1968 1969 1970 1971 1972 1973 1974
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:
1975
		return nvme_dev_user_cmd(ctrl, argp);
1976
	case NVME_IOCTL_RESET:
1977
		dev_warn(ctrl->device, "resetting controller\n");
1978
		return nvme_reset_ctrl_sync(ctrl);
1979 1980
	case NVME_IOCTL_SUBSYS_RESET:
		return nvme_reset_subsystem(ctrl);
K
Keith Busch 已提交
1981 1982 1983
	case NVME_IOCTL_RESCAN:
		nvme_queue_scan(ctrl);
		return 0;
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
	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;

2004
	ret = nvme_reset_ctrl_sync(ctrl);
2005 2006 2007
	if (ret < 0)
		return ret;
	return count;
2008
}
2009
static DEVICE_ATTR(reset_controller, S_IWUSR, NULL, nvme_sysfs_reset);
2010

K
Keith Busch 已提交
2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
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);

2022 2023 2024
static ssize_t wwid_show(struct device *dev, struct device_attribute *attr,
								char *buf)
{
2025
	struct nvme_ns *ns = nvme_get_ns_from_dev(dev);
2026 2027 2028 2029
	struct nvme_ctrl *ctrl = ns->ctrl;
	int serial_len = sizeof(ctrl->serial);
	int model_len = sizeof(ctrl->model);

2030 2031 2032
	if (!uuid_is_null(&ns->uuid))
		return sprintf(buf, "uuid.%pU\n", &ns->uuid);

2033 2034
	if (memchr_inv(ns->nguid, 0, sizeof(ns->nguid)))
		return sprintf(buf, "eui.%16phN\n", ns->nguid);
2035 2036 2037 2038

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

2039 2040
	while (serial_len > 0 && (ctrl->serial[serial_len - 1] == ' ' ||
				  ctrl->serial[serial_len - 1] == '\0'))
2041
		serial_len--;
2042 2043
	while (model_len > 0 && (ctrl->model[model_len - 1] == ' ' ||
				 ctrl->model[model_len - 1] == '\0'))
2044 2045 2046 2047 2048 2049 2050
		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);

2051 2052 2053 2054 2055 2056 2057 2058
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);

2059 2060 2061
static ssize_t uuid_show(struct device *dev, struct device_attribute *attr,
								char *buf)
{
2062
	struct nvme_ns *ns = nvme_get_ns_from_dev(dev);
2063 2064 2065 2066 2067 2068 2069 2070 2071 2072

	/* 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);
2073 2074 2075 2076 2077 2078
}
static DEVICE_ATTR(uuid, S_IRUGO, uuid_show, NULL);

static ssize_t eui_show(struct device *dev, struct device_attribute *attr,
								char *buf)
{
2079
	struct nvme_ns *ns = nvme_get_ns_from_dev(dev);
2080 2081 2082 2083 2084 2085 2086
	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)
{
2087
	struct nvme_ns *ns = nvme_get_ns_from_dev(dev);
2088 2089 2090 2091 2092
	return sprintf(buf, "%d\n", ns->ns_id);
}
static DEVICE_ATTR(nsid, S_IRUGO, nsid_show, NULL);

static struct attribute *nvme_ns_attrs[] = {
2093
	&dev_attr_wwid.attr,
2094
	&dev_attr_uuid.attr,
2095
	&dev_attr_nguid.attr,
2096 2097 2098 2099 2100
	&dev_attr_eui.attr,
	&dev_attr_nsid.attr,
	NULL,
};

M
Ming Lin 已提交
2101
static umode_t nvme_ns_attrs_are_visible(struct kobject *kobj,
2102 2103 2104
		struct attribute *a, int n)
{
	struct device *dev = container_of(kobj, struct device, kobj);
2105
	struct nvme_ns *ns = nvme_get_ns_from_dev(dev);
2106 2107

	if (a == &dev_attr_uuid.attr) {
2108 2109 2110 2111 2112
		if (uuid_is_null(&ns->uuid) ||
		    !memchr_inv(ns->nguid, 0, sizeof(ns->nguid)))
			return 0;
	}
	if (a == &dev_attr_nguid.attr) {
2113
		if (!memchr_inv(ns->nguid, 0, sizeof(ns->nguid)))
2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124
			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
2125
	.is_visible	= nvme_ns_attrs_are_visible,
2126 2127
};

M
Ming Lin 已提交
2128
#define nvme_show_str_function(field)						\
2129 2130 2131 2132 2133 2134 2135 2136
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
2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149
#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);
2150

M
Ming Lin 已提交
2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172
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);

2173 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_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
2196 2197 2198 2199 2200 2201
static ssize_t nvme_sysfs_show_subsysnqn(struct device *dev,
					 struct device_attribute *attr,
					 char *buf)
{
	struct nvme_ctrl *ctrl = dev_get_drvdata(dev);

2202
	return snprintf(buf, PAGE_SIZE, "%s\n", ctrl->subnqn);
M
Ming Lin 已提交
2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215
}
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);

2216 2217
static struct attribute *nvme_dev_attrs[] = {
	&dev_attr_reset_controller.attr,
K
Keith Busch 已提交
2218
	&dev_attr_rescan_controller.attr,
2219 2220 2221
	&dev_attr_model.attr,
	&dev_attr_serial.attr,
	&dev_attr_firmware_rev.attr,
M
Ming Lin 已提交
2222
	&dev_attr_cntlid.attr,
M
Ming Lin 已提交
2223 2224 2225 2226
	&dev_attr_delete_controller.attr,
	&dev_attr_transport.attr,
	&dev_attr_subsysnqn.attr,
	&dev_attr_address.attr,
2227
	&dev_attr_state.attr,
2228 2229 2230
	NULL
};

M
Ming Lin 已提交
2231 2232 2233 2234 2235 2236
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);

2237 2238 2239 2240
	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
2241 2242 2243 2244

	return a->mode;
}

2245
static struct attribute_group nvme_dev_attrs_group = {
M
Ming Lin 已提交
2246 2247
	.attrs		= nvme_dev_attrs,
	.is_visible	= nvme_dev_attrs_are_visible,
2248 2249 2250 2251 2252 2253 2254
};

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

2255 2256 2257 2258 2259 2260 2261 2262
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;
}

2263
static struct nvme_ns *nvme_find_get_ns(struct nvme_ctrl *ctrl, unsigned nsid)
2264
{
2265
	struct nvme_ns *ns, *ret = NULL;
2266

2267
	mutex_lock(&ctrl->namespaces_mutex);
2268
	list_for_each_entry(ns, &ctrl->namespaces, list) {
2269 2270 2271 2272 2273
		if (ns->ns_id == nsid) {
			kref_get(&ns->kref);
			ret = ns;
			break;
		}
2274 2275 2276
		if (ns->ns_id > nsid)
			break;
	}
2277 2278
	mutex_unlock(&ctrl->namespaces_mutex);
	return ret;
2279 2280
}

2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306
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;
}

2307 2308 2309 2310
static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
{
	struct nvme_ns *ns;
	struct gendisk *disk;
2311 2312
	struct nvme_id_ns *id;
	char disk_name[DISK_NAME_LEN];
2313 2314 2315 2316 2317 2318
	int node = dev_to_node(ctrl->dev);

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

2319 2320 2321 2322
	ns->instance = ida_simple_get(&ctrl->ns_ida, 1, 0, GFP_KERNEL);
	if (ns->instance < 0)
		goto out_free_ns;

2323 2324
	ns->queue = blk_mq_init_queue(ctrl->tagset);
	if (IS_ERR(ns->queue))
2325
		goto out_release_instance;
2326 2327 2328 2329 2330 2331 2332 2333 2334
	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);
2335
	nvme_set_queue_limits(ctrl, ns->queue);
2336
	nvme_setup_streams_ns(ctrl, ns);
2337

2338
	sprintf(disk_name, "nvme%dn%d", ctrl->instance, ns->instance);
2339

2340 2341 2342
	if (nvme_revalidate_ns(ns, &id))
		goto out_free_queue;

2343 2344
	if (nvme_nvm_ns_supported(ns, id) &&
				nvme_nvm_register(ns, disk_name, node)) {
2345
		dev_warn(ctrl->device, "%s: LightNVM init failure\n", __func__);
2346 2347
		goto out_free_id;
	}
2348

2349 2350 2351
	disk = alloc_disk_node(0, node);
	if (!disk)
		goto out_free_id;
2352

2353 2354 2355 2356 2357 2358 2359 2360
	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);
2361

2362 2363 2364 2365
	mutex_lock(&ctrl->namespaces_mutex);
	list_add_tail(&ns->list, &ctrl->namespaces);
	mutex_unlock(&ctrl->namespaces_mutex);

2366
	kref_get(&ctrl->kref);
2367 2368 2369

	kfree(id);

2370
	device_add_disk(ctrl->device, ns->disk);
2371 2372 2373 2374
	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);
2375 2376 2377
	if (ns->ndev && nvme_nvm_register_sysfs(ns))
		pr_warn("%s: failed to register lightnvm sysfs group for identification\n",
			ns->disk->disk_name);
2378
	return;
2379 2380
 out_free_id:
	kfree(id);
2381 2382
 out_free_queue:
	blk_cleanup_queue(ns->queue);
2383 2384
 out_release_instance:
	ida_simple_remove(&ctrl->ns_ida, ns->instance);
2385 2386 2387 2388 2389 2390
 out_free_ns:
	kfree(ns);
}

static void nvme_ns_remove(struct nvme_ns *ns)
{
2391 2392
	if (test_and_set_bit(NVME_NS_REMOVING, &ns->flags))
		return;
2393

2394
	if (ns->disk && ns->disk->flags & GENHD_FL_UP) {
2395 2396
		if (blk_get_integrity(ns->disk))
			blk_integrity_unregister(ns->disk);
2397 2398
		sysfs_remove_group(&disk_to_dev(ns->disk)->kobj,
					&nvme_ns_attr_group);
2399 2400
		if (ns->ndev)
			nvme_nvm_unregister_sysfs(ns);
2401 2402 2403
		del_gendisk(ns->disk);
		blk_cleanup_queue(ns->queue);
	}
2404 2405

	mutex_lock(&ns->ctrl->namespaces_mutex);
2406
	list_del_init(&ns->list);
2407 2408
	mutex_unlock(&ns->ctrl->namespaces_mutex);

2409 2410 2411
	nvme_put_ns(ns);
}

2412 2413 2414 2415
static void nvme_validate_ns(struct nvme_ctrl *ctrl, unsigned nsid)
{
	struct nvme_ns *ns;

2416
	ns = nvme_find_get_ns(ctrl, nsid);
2417
	if (ns) {
2418
		if (ns->disk && revalidate_disk(ns->disk))
2419
			nvme_ns_remove(ns);
2420
		nvme_put_ns(ns);
2421 2422 2423 2424
	} else
		nvme_alloc_ns(ctrl, nsid);
}

2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435
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);
	}
}

2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449
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)
2450
			goto free;
2451 2452 2453 2454 2455 2456 2457 2458 2459

		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) {
2460 2461
				ns = nvme_find_get_ns(ctrl, prev);
				if (ns) {
2462
					nvme_ns_remove(ns);
2463 2464
					nvme_put_ns(ns);
				}
2465 2466 2467 2468 2469
			}
		}
		nn -= j;
	}
 out:
2470 2471
	nvme_remove_invalid_namespaces(ctrl, prev);
 free:
2472 2473 2474 2475
	kfree(ns_list);
	return ret;
}

2476
static void nvme_scan_ns_sequential(struct nvme_ctrl *ctrl, unsigned nn)
2477 2478 2479
{
	unsigned i;

2480 2481 2482
	for (i = 1; i <= nn; i++)
		nvme_validate_ns(ctrl, i);

2483
	nvme_remove_invalid_namespaces(ctrl, nn);
2484 2485
}

2486
static void nvme_scan_work(struct work_struct *work)
2487
{
2488 2489
	struct nvme_ctrl *ctrl =
		container_of(work, struct nvme_ctrl, scan_work);
2490
	struct nvme_id_ctrl *id;
2491
	unsigned nn;
2492

2493 2494 2495
	if (ctrl->state != NVME_CTRL_LIVE)
		return;

2496 2497
	if (nvme_identify_ctrl(ctrl, &id))
		return;
2498 2499

	nn = le32_to_cpu(id->nn);
2500
	if (ctrl->vs >= NVME_VS(1, 1, 0) &&
2501 2502 2503 2504
	    !(ctrl->quirks & NVME_QUIRK_IDENTIFY_CNS)) {
		if (!nvme_scan_ns_list(ctrl, nn))
			goto done;
	}
2505
	nvme_scan_ns_sequential(ctrl, nn);
2506
 done:
2507
	mutex_lock(&ctrl->namespaces_mutex);
2508
	list_sort(NULL, &ctrl->namespaces, ns_cmp);
2509
	mutex_unlock(&ctrl->namespaces_mutex);
2510 2511
	kfree(id);
}
2512 2513 2514 2515 2516 2517 2518 2519

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)
2520
		queue_work(nvme_wq, &ctrl->scan_work);
2521 2522
}
EXPORT_SYMBOL_GPL(nvme_queue_scan);
2523

2524 2525 2526 2527 2528
/*
 * 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.
 */
2529 2530 2531 2532
void nvme_remove_namespaces(struct nvme_ctrl *ctrl)
{
	struct nvme_ns *ns, *next;

2533 2534 2535 2536 2537 2538 2539 2540 2541
	/*
	 * 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);

2542 2543 2544
	list_for_each_entry_safe(ns, next, &ctrl->namespaces, list)
		nvme_ns_remove(ns);
}
2545
EXPORT_SYMBOL_GPL(nvme_remove_namespaces);
2546

2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562
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);
	while (ctrl->event_limit > 0) {
		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);
}

2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586
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
2587
	c.common.nsid = cpu_to_le32(NVME_NSID_ALL);
2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627
	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);
}

2628 2629
void nvme_complete_async_event(struct nvme_ctrl *ctrl, __le16 status,
		union nvme_result *res)
2630
{
2631 2632
	u32 result = le32_to_cpu(res->u32);
	bool done = true;
2633

2634 2635 2636 2637 2638
	switch (le16_to_cpu(status) >> 1) {
	case NVME_SC_SUCCESS:
		done = false;
		/*FALLTHRU*/
	case NVME_SC_ABORT_REQ:
2639
		++ctrl->event_limit;
2640
		queue_work(nvme_wq, &ctrl->async_event_work);
2641 2642 2643
		break;
	default:
		break;
2644 2645
	}

2646
	if (done)
2647 2648 2649 2650 2651 2652 2653
		return;

	switch (result & 0xff07) {
	case NVME_AER_NOTICE_NS_CHANGED:
		dev_info(ctrl->device, "rescanning\n");
		nvme_queue_scan(ctrl);
		break;
2654 2655 2656
	case NVME_AER_NOTICE_FW_ACT_STARTING:
		schedule_work(&ctrl->fw_act_work);
		break;
2657 2658 2659 2660 2661 2662 2663 2664 2665
	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;
2666
	queue_work(nvme_wq, &ctrl->async_event_work);
2667 2668 2669
}
EXPORT_SYMBOL_GPL(nvme_queue_async_events);

2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698
static DEFINE_IDA(nvme_instance_ida);

static int nvme_set_instance(struct nvme_ctrl *ctrl)
{
	int instance, error;

	do {
		if (!ida_pre_get(&nvme_instance_ida, GFP_KERNEL))
			return -ENODEV;

		spin_lock(&dev_list_lock);
		error = ida_get_new(&nvme_instance_ida, &instance);
		spin_unlock(&dev_list_lock);
	} while (error == -EAGAIN);

	if (error)
		return -ENODEV;

	ctrl->instance = instance;
	return 0;
}

static void nvme_release_instance(struct nvme_ctrl *ctrl)
{
	spin_lock(&dev_list_lock);
	ida_remove(&nvme_instance_ida, ctrl->instance);
	spin_unlock(&dev_list_lock);
}

2699
void nvme_stop_ctrl(struct nvme_ctrl *ctrl)
2700
{
2701
	nvme_stop_keep_alive(ctrl);
2702
	flush_work(&ctrl->async_event_work);
2703
	flush_work(&ctrl->scan_work);
2704
	cancel_work_sync(&ctrl->fw_act_work);
2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719
}
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);
2720

2721 2722
void nvme_uninit_ctrl(struct nvme_ctrl *ctrl)
{
2723
	device_destroy(nvme_class, MKDEV(nvme_char_major, ctrl->instance));
2724 2725 2726 2727

	spin_lock(&dev_list_lock);
	list_del(&ctrl->node);
	spin_unlock(&dev_list_lock);
2728
}
2729
EXPORT_SYMBOL_GPL(nvme_uninit_ctrl);
2730 2731 2732 2733

static void nvme_free_ctrl(struct kref *kref)
{
	struct nvme_ctrl *ctrl = container_of(kref, struct nvme_ctrl, kref);
2734 2735 2736

	put_device(ctrl->device);
	nvme_release_instance(ctrl);
2737
	ida_destroy(&ctrl->ns_ida);
2738 2739 2740 2741 2742 2743 2744 2745

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

void nvme_put_ctrl(struct nvme_ctrl *ctrl)
{
	kref_put(&ctrl->kref, nvme_free_ctrl);
}
2746
EXPORT_SYMBOL_GPL(nvme_put_ctrl);
2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757

/*
 * 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;

2758 2759
	ctrl->state = NVME_CTRL_NEW;
	spin_lock_init(&ctrl->lock);
2760
	INIT_LIST_HEAD(&ctrl->namespaces);
2761
	mutex_init(&ctrl->namespaces_mutex);
2762 2763 2764 2765
	kref_init(&ctrl->kref);
	ctrl->dev = dev;
	ctrl->ops = ops;
	ctrl->quirks = quirks;
2766
	INIT_WORK(&ctrl->scan_work, nvme_scan_work);
2767
	INIT_WORK(&ctrl->async_event_work, nvme_async_event_work);
2768
	INIT_WORK(&ctrl->fw_act_work, nvme_fw_act_work);
2769 2770 2771 2772 2773

	ret = nvme_set_instance(ctrl);
	if (ret)
		goto out;

2774
	ctrl->device = device_create_with_groups(nvme_class, ctrl->dev,
2775
				MKDEV(nvme_char_major, ctrl->instance),
2776
				ctrl, nvme_dev_attr_groups,
2777
				"nvme%d", ctrl->instance);
2778 2779 2780 2781 2782
	if (IS_ERR(ctrl->device)) {
		ret = PTR_ERR(ctrl->device);
		goto out_release_instance;
	}
	get_device(ctrl->device);
2783
	ida_init(&ctrl->ns_ida);
2784 2785 2786 2787 2788

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

2789 2790 2791 2792 2793 2794 2795 2796
	/*
	 * 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));

2797 2798 2799 2800 2801 2802
	return 0;
out_release_instance:
	nvme_release_instance(ctrl);
out:
	return ret;
}
2803
EXPORT_SYMBOL_GPL(nvme_init_ctrl);
2804

2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815
/**
 * 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;

2816
	mutex_lock(&ctrl->namespaces_mutex);
M
Ming Lei 已提交
2817

2818
	/* Forcibly unquiesce queues to avoid blocking dispatch */
2819 2820
	if (ctrl->admin_q)
		blk_mq_unquiesce_queue(ctrl->admin_q);
2821

2822
	list_for_each_entry(ns, &ctrl->namespaces, list) {
2823 2824 2825 2826
		/*
		 * Revalidating a dead namespace sets capacity to 0. This will
		 * end buffered writers dirtying pages that can't be synced.
		 */
2827 2828 2829
		if (!ns->disk || test_and_set_bit(NVME_NS_DEAD, &ns->flags))
			continue;
		revalidate_disk(ns->disk);
2830
		blk_set_queue_dying(ns->queue);
2831

2832 2833
		/* Forcibly unquiesce queues to avoid blocking dispatch */
		blk_mq_unquiesce_queue(ns->queue);
2834
	}
2835
	mutex_unlock(&ctrl->namespaces_mutex);
2836
}
2837
EXPORT_SYMBOL_GPL(nvme_kill_queues);
2838

K
Keith Busch 已提交
2839 2840 2841 2842 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
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)
2881
		blk_freeze_queue_start(ns->queue);
K
Keith Busch 已提交
2882 2883 2884 2885
	mutex_unlock(&ctrl->namespaces_mutex);
}
EXPORT_SYMBOL_GPL(nvme_start_freeze);

2886
void nvme_stop_queues(struct nvme_ctrl *ctrl)
2887 2888 2889
{
	struct nvme_ns *ns;

2890
	mutex_lock(&ctrl->namespaces_mutex);
2891
	list_for_each_entry(ns, &ctrl->namespaces, list)
2892
		blk_mq_quiesce_queue(ns->queue);
2893
	mutex_unlock(&ctrl->namespaces_mutex);
2894
}
2895
EXPORT_SYMBOL_GPL(nvme_stop_queues);
2896

2897
void nvme_start_queues(struct nvme_ctrl *ctrl)
2898 2899 2900
{
	struct nvme_ns *ns;

2901
	mutex_lock(&ctrl->namespaces_mutex);
2902
	list_for_each_entry(ns, &ctrl->namespaces, list)
2903
		blk_mq_unquiesce_queue(ns->queue);
2904
	mutex_unlock(&ctrl->namespaces_mutex);
2905
}
2906
EXPORT_SYMBOL_GPL(nvme_start_queues);
2907

2908 2909 2910 2911
int __init nvme_core_init(void)
{
	int result;

2912 2913 2914 2915 2916
	nvme_wq = alloc_workqueue("nvme-wq",
			WQ_UNBOUND | WQ_MEM_RECLAIM | WQ_SYSFS, 0);
	if (!nvme_wq)
		return -ENOMEM;

2917 2918 2919
	result = __register_chrdev(nvme_char_major, 0, NVME_MINORS, "nvme",
							&nvme_dev_fops);
	if (result < 0)
2920
		goto destroy_wq;
2921 2922 2923 2924 2925 2926 2927 2928 2929
	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;
	}

2930
	return 0;
2931

2932
unregister_chrdev:
2933
	__unregister_chrdev(nvme_char_major, 0, NVME_MINORS, "nvme");
2934 2935
destroy_wq:
	destroy_workqueue(nvme_wq);
2936
	return result;
2937 2938 2939 2940
}

void nvme_core_exit(void)
{
2941 2942
	class_destroy(nvme_class);
	__unregister_chrdev(nvme_char_major, 0, NVME_MINORS, "nvme");
2943
	destroy_workqueue(nvme_wq);
2944
}
2945 2946 2947 2948 2949

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