core.c 25.6 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
/*
 * Copyright (C) 2015 IT University of Copenhagen. All rights reserved.
 * Initial release: Matias Bjorling <m@bjorling.me>
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License version
 * 2 as published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful, but
 * WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; see the file COPYING.  If not, write to
 * the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139,
 * USA.
 *
 */

#include <linux/list.h>
#include <linux/types.h>
#include <linux/sem.h>
#include <linux/bitmap.h>
#include <linux/module.h>
#include <linux/miscdevice.h>
#include <linux/lightnvm.h>
28
#include <linux/sched/sysctl.h>
29

30
static LIST_HEAD(nvm_tgt_types);
31
static DECLARE_RWSEM(nvm_tgtt_lock);
32 33 34 35
static LIST_HEAD(nvm_mgrs);
static LIST_HEAD(nvm_devices);
static DECLARE_RWSEM(nvm_lock);

36
struct nvm_tgt_type *nvm_find_target_type(const char *name, int lock)
37
{
38
	struct nvm_tgt_type *tmp, *tt = NULL;
39

40
	if (lock)
41
		down_write(&nvm_tgtt_lock);
42

43 44 45 46 47
	list_for_each_entry(tmp, &nvm_tgt_types, list)
		if (!strcmp(name, tmp->name)) {
			tt = tmp;
			break;
		}
48

49
	if (lock)
50
		up_write(&nvm_tgtt_lock);
51
	return tt;
52
}
53
EXPORT_SYMBOL(nvm_find_target_type);
54

55
int nvm_register_tgt_type(struct nvm_tgt_type *tt)
56 57 58
{
	int ret = 0;

59
	down_write(&nvm_tgtt_lock);
60
	if (nvm_find_target_type(tt->name, 0))
61 62
		ret = -EEXIST;
	else
63
		list_add(&tt->list, &nvm_tgt_types);
64
	up_write(&nvm_tgtt_lock);
65 66 67

	return ret;
}
68
EXPORT_SYMBOL(nvm_register_tgt_type);
69

70
void nvm_unregister_tgt_type(struct nvm_tgt_type *tt)
71 72 73 74 75 76 77 78
{
	if (!tt)
		return;

	down_write(&nvm_lock);
	list_del(&tt->list);
	up_write(&nvm_lock);
}
79
EXPORT_SYMBOL(nvm_unregister_tgt_type);
80 81 82 83

void *nvm_dev_dma_alloc(struct nvm_dev *dev, gfp_t mem_flags,
							dma_addr_t *dma_handler)
{
84
	return dev->ops->dev_dma_alloc(dev, dev->dma_pool, mem_flags,
85 86 87 88
								dma_handler);
}
EXPORT_SYMBOL(nvm_dev_dma_alloc);

89
void nvm_dev_dma_free(struct nvm_dev *dev, void *addr,
90 91
							dma_addr_t dma_handler)
{
92
	dev->ops->dev_dma_free(dev->dma_pool, addr, dma_handler);
93 94 95 96 97 98 99 100 101 102 103 104 105 106
}
EXPORT_SYMBOL(nvm_dev_dma_free);

static struct nvmm_type *nvm_find_mgr_type(const char *name)
{
	struct nvmm_type *mt;

	list_for_each_entry(mt, &nvm_mgrs, list)
		if (!strcmp(name, mt->name))
			return mt;

	return NULL;
}

107
static struct nvmm_type *nvm_init_mgr(struct nvm_dev *dev)
108 109 110 111 112 113 114
{
	struct nvmm_type *mt;
	int ret;

	lockdep_assert_held(&nvm_lock);

	list_for_each_entry(mt, &nvm_mgrs, list) {
115 116 117
		if (strncmp(dev->sb.mmtype, mt->name, NVM_MMTYPE_LEN))
			continue;

118 119 120 121 122 123 124 125 126 127 128 129
		ret = mt->register_mgr(dev);
		if (ret < 0) {
			pr_err("nvm: media mgr failed to init (%d) on dev %s\n",
								ret, dev->name);
			return NULL; /* initialization failed */
		} else if (ret > 0)
			return mt;
	}

	return NULL;
}

130 131
int nvm_register_mgr(struct nvmm_type *mt)
{
132
	struct nvm_dev *dev;
133 134 135
	int ret = 0;

	down_write(&nvm_lock);
136
	if (nvm_find_mgr_type(mt->name)) {
137
		ret = -EEXIST;
138 139
		goto finish;
	} else {
140
		list_add(&mt->list, &nvm_mgrs);
141 142 143 144 145 146 147 148 149 150
	}

	/* try to register media mgr if any device have none configured */
	list_for_each_entry(dev, &nvm_devices, devices) {
		if (dev->mt)
			continue;

		dev->mt = nvm_init_mgr(dev);
	}
finish:
151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192
	up_write(&nvm_lock);

	return ret;
}
EXPORT_SYMBOL(nvm_register_mgr);

void nvm_unregister_mgr(struct nvmm_type *mt)
{
	if (!mt)
		return;

	down_write(&nvm_lock);
	list_del(&mt->list);
	up_write(&nvm_lock);
}
EXPORT_SYMBOL(nvm_unregister_mgr);

static struct nvm_dev *nvm_find_nvm_dev(const char *name)
{
	struct nvm_dev *dev;

	list_for_each_entry(dev, &nvm_devices, devices)
		if (!strcmp(name, dev->name))
			return dev;

	return NULL;
}

struct nvm_block *nvm_get_blk(struct nvm_dev *dev, struct nvm_lun *lun,
							unsigned long flags)
{
	return dev->mt->get_blk(dev, lun, flags);
}
EXPORT_SYMBOL(nvm_get_blk);

/* Assumes that all valid pages have already been moved on release to bm */
void nvm_put_blk(struct nvm_dev *dev, struct nvm_block *blk)
{
	return dev->mt->put_blk(dev, blk);
}
EXPORT_SYMBOL(nvm_put_blk);

193 194 195 196 197 198
void nvm_mark_blk(struct nvm_dev *dev, struct ppa_addr ppa, int type)
{
	return dev->mt->mark_blk(dev, ppa, type);
}
EXPORT_SYMBOL(nvm_mark_blk);

199 200 201 202 203 204 205 206 207 208 209 210
int nvm_submit_io(struct nvm_dev *dev, struct nvm_rq *rqd)
{
	return dev->mt->submit_io(dev, rqd);
}
EXPORT_SYMBOL(nvm_submit_io);

int nvm_erase_blk(struct nvm_dev *dev, struct nvm_block *blk)
{
	return dev->mt->erase_blk(dev, blk, 0);
}
EXPORT_SYMBOL(nvm_erase_blk);

211 212 213 214
void nvm_addr_to_generic_mode(struct nvm_dev *dev, struct nvm_rq *rqd)
{
	int i;

215 216
	if (rqd->nr_ppas > 1) {
		for (i = 0; i < rqd->nr_ppas; i++)
217 218 219 220 221 222 223 224 225 226 227 228
			rqd->ppa_list[i] = dev_to_generic_addr(dev,
							rqd->ppa_list[i]);
	} else {
		rqd->ppa_addr = dev_to_generic_addr(dev, rqd->ppa_addr);
	}
}
EXPORT_SYMBOL(nvm_addr_to_generic_mode);

void nvm_generic_to_addr_mode(struct nvm_dev *dev, struct nvm_rq *rqd)
{
	int i;

229 230
	if (rqd->nr_ppas > 1) {
		for (i = 0; i < rqd->nr_ppas; i++)
231 232 233 234 235 236 237 238
			rqd->ppa_list[i] = generic_to_dev_addr(dev,
							rqd->ppa_list[i]);
	} else {
		rqd->ppa_addr = generic_to_dev_addr(dev, rqd->ppa_addr);
	}
}
EXPORT_SYMBOL(nvm_generic_to_addr_mode);

239
int nvm_set_rqd_ppalist(struct nvm_dev *dev, struct nvm_rq *rqd,
240
			const struct ppa_addr *ppas, int nr_ppas, int vblk)
241
{
242
	int i, plane_cnt, pl_idx;
243
	struct ppa_addr ppa;
244

245
	if ((!vblk || dev->plane_mode == NVM_PLANE_SINGLE) && nr_ppas == 1) {
246
		rqd->nr_ppas = nr_ppas;
247
		rqd->ppa_addr = ppas[0];
248 249

		return 0;
250
	}
251

252
	rqd->nr_ppas = nr_ppas;
253 254 255 256 257 258
	rqd->ppa_list = nvm_dev_dma_alloc(dev, GFP_KERNEL, &rqd->dma_ppa_list);
	if (!rqd->ppa_list) {
		pr_err("nvm: failed to allocate dma memory\n");
		return -ENOMEM;
	}

259 260 261 262 263
	if (!vblk) {
		for (i = 0; i < nr_ppas; i++)
			rqd->ppa_list[i] = ppas[i];
	} else {
		plane_cnt = dev->plane_mode;
264
		rqd->nr_ppas *= plane_cnt;
265

266
		for (i = 0; i < nr_ppas; i++) {
267
			for (pl_idx = 0; pl_idx < plane_cnt; pl_idx++) {
268 269 270
				ppa = ppas[i];
				ppa.g.pl = pl_idx;
				rqd->ppa_list[(pl_idx * nr_ppas) + i] = ppa;
271
			}
272 273 274
		}
	}

275 276 277 278 279 280 281 282 283 284 285 286 287
	return 0;
}
EXPORT_SYMBOL(nvm_set_rqd_ppalist);

void nvm_free_rqd_ppalist(struct nvm_dev *dev, struct nvm_rq *rqd)
{
	if (!rqd->ppa_list)
		return;

	nvm_dev_dma_free(dev, rqd->ppa_list, rqd->dma_ppa_list);
}
EXPORT_SYMBOL(nvm_free_rqd_ppalist);

288
int nvm_erase_ppa(struct nvm_dev *dev, struct ppa_addr *ppas, int nr_ppas)
289 290 291 292 293 294 295 296 297
{
	struct nvm_rq rqd;
	int ret;

	if (!dev->ops->erase_block)
		return 0;

	memset(&rqd, 0, sizeof(struct nvm_rq));

298
	ret = nvm_set_rqd_ppalist(dev, &rqd, ppas, nr_ppas, 1);
299 300 301
	if (ret)
		return ret;

302 303 304 305
	nvm_generic_to_addr_mode(dev, &rqd);

	ret = dev->ops->erase_block(dev, &rqd);

306
	nvm_free_rqd_ppalist(dev, &rqd);
307 308 309 310 311

	return ret;
}
EXPORT_SYMBOL(nvm_erase_ppa);

312 313
void nvm_end_io(struct nvm_rq *rqd, int error)
{
314 315
	rqd->error = error;
	rqd->end_io(rqd);
316 317 318
}
EXPORT_SYMBOL(nvm_end_io);

319
static void nvm_end_io_sync(struct nvm_rq *rqd)
320 321 322 323 324 325 326 327
{
	struct completion *waiting = rqd->wait;

	rqd->wait = NULL;

	complete(waiting);
}

328
static int __nvm_submit_ppa(struct nvm_dev *dev, struct nvm_rq *rqd, int opcode,
329
						int flags, void *buf, int len)
330 331 332 333 334 335 336 337 338 339
{
	DECLARE_COMPLETION_ONSTACK(wait);
	struct bio *bio;
	int ret;
	unsigned long hang_check;

	bio = bio_map_kern(dev->q, buf, len, GFP_KERNEL);
	if (IS_ERR_OR_NULL(bio))
		return -ENOMEM;

340
	nvm_generic_to_addr_mode(dev, rqd);
341

342 343 344 345 346 347
	rqd->dev = dev;
	rqd->opcode = opcode;
	rqd->flags = flags;
	rqd->bio = bio;
	rqd->wait = &wait;
	rqd->end_io = nvm_end_io_sync;
348

349
	ret = dev->ops->submit_io(dev, rqd);
350 351 352 353
	if (ret) {
		bio_put(bio);
		return ret;
	}
354 355 356 357

	/* Prevent hang_check timer from firing at us during very long I/O */
	hang_check = sysctl_hung_task_timeout_secs;
	if (hang_check)
358 359 360
		while (!wait_for_completion_io_timeout(&wait,
							hang_check * (HZ/2)))
			;
361 362 363
	else
		wait_for_completion_io(&wait);

364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387
	return rqd->error;
}

/**
 * nvm_submit_ppa_list - submit user-defined ppa list to device. The user must
 *			 take to free ppa list if necessary.
 * @dev:	device
 * @ppa_list:	user created ppa_list
 * @nr_ppas:	length of ppa_list
 * @opcode:	device opcode
 * @flags:	device flags
 * @buf:	data buffer
 * @len:	data buffer length
 */
int nvm_submit_ppa_list(struct nvm_dev *dev, struct ppa_addr *ppa_list,
			int nr_ppas, int opcode, int flags, void *buf, int len)
{
	struct nvm_rq rqd;

	if (dev->ops->max_phys_sect < nr_ppas)
		return -EINVAL;

	memset(&rqd, 0, sizeof(struct nvm_rq));

388
	rqd.nr_ppas = nr_ppas;
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
	if (nr_ppas > 1)
		rqd.ppa_list = ppa_list;
	else
		rqd.ppa_addr = ppa_list[0];

	return __nvm_submit_ppa(dev, &rqd, opcode, flags, buf, len);
}
EXPORT_SYMBOL(nvm_submit_ppa_list);

/**
 * nvm_submit_ppa - submit PPAs to device. PPAs will automatically be unfolded
 *		    as single, dual, quad plane PPAs depending on device type.
 * @dev:	device
 * @ppa:	user created ppa_list
 * @nr_ppas:	length of ppa_list
 * @opcode:	device opcode
 * @flags:	device flags
 * @buf:	data buffer
 * @len:	data buffer length
 */
int nvm_submit_ppa(struct nvm_dev *dev, struct ppa_addr *ppa, int nr_ppas,
				int opcode, int flags, void *buf, int len)
{
	struct nvm_rq rqd;
	int ret;

	memset(&rqd, 0, sizeof(struct nvm_rq));
416
	ret = nvm_set_rqd_ppalist(dev, &rqd, ppa, nr_ppas, 1);
417 418 419 420 421
	if (ret)
		return ret;

	ret = __nvm_submit_ppa(dev, &rqd, opcode, flags, buf, len);

422 423
	nvm_free_rqd_ppalist(dev, &rqd);

424
	return ret;
425 426 427
}
EXPORT_SYMBOL(nvm_submit_ppa);

428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462
/*
 * folds a bad block list from its plane representation to its virtual
 * block representation. The fold is done in place and reduced size is
 * returned.
 *
 * If any of the planes status are bad or grown bad block, the virtual block
 * is marked bad. If not bad, the first plane state acts as the block state.
 */
int nvm_bb_tbl_fold(struct nvm_dev *dev, u8 *blks, int nr_blks)
{
	int blk, offset, pl, blktype;

	if (nr_blks != dev->blks_per_lun * dev->plane_mode)
		return -EINVAL;

	for (blk = 0; blk < dev->blks_per_lun; blk++) {
		offset = blk * dev->plane_mode;
		blktype = blks[offset];

		/* Bad blocks on any planes take precedence over other types */
		for (pl = 0; pl < dev->plane_mode; pl++) {
			if (blks[offset + pl] &
					(NVM_BLK_T_BAD|NVM_BLK_T_GRWN_BAD)) {
				blktype = blks[offset + pl];
				break;
			}
		}

		blks[blk] = blktype;
	}

	return dev->blks_per_lun;
}
EXPORT_SYMBOL(nvm_bb_tbl_fold);

463 464 465 466 467 468 469 470
int nvm_get_bb_tbl(struct nvm_dev *dev, struct ppa_addr ppa, u8 *blks)
{
	ppa = generic_to_dev_addr(dev, ppa);

	return dev->ops->get_bb_tbl(dev, ppa, blks);
}
EXPORT_SYMBOL(nvm_get_bb_tbl);

471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502
static int nvm_init_slc_tbl(struct nvm_dev *dev, struct nvm_id_group *grp)
{
	int i;

	dev->lps_per_blk = dev->pgs_per_blk;
	dev->lptbl = kcalloc(dev->lps_per_blk, sizeof(int), GFP_KERNEL);
	if (!dev->lptbl)
		return -ENOMEM;

	/* Just a linear array */
	for (i = 0; i < dev->lps_per_blk; i++)
		dev->lptbl[i] = i;

	return 0;
}

static int nvm_init_mlc_tbl(struct nvm_dev *dev, struct nvm_id_group *grp)
{
	int i, p;
	struct nvm_id_lp_mlc *mlc = &grp->lptbl.mlc;

	if (!mlc->num_pairs)
		return 0;

	dev->lps_per_blk = mlc->num_pairs;
	dev->lptbl = kcalloc(dev->lps_per_blk, sizeof(int), GFP_KERNEL);
	if (!dev->lptbl)
		return -ENOMEM;

	/* The lower page table encoding consists of a list of bytes, where each
	 * has a lower and an upper half. The first half byte maintains the
	 * increment value and every value after is an offset added to the
503 504
	 * previous incrementation value
	 */
505 506 507 508 509 510 511 512 513 514 515 516
	dev->lptbl[0] = mlc->pairs[0] & 0xF;
	for (i = 1; i < dev->lps_per_blk; i++) {
		p = mlc->pairs[i >> 1];
		if (i & 0x1) /* upper */
			dev->lptbl[i] = dev->lptbl[i - 1] + ((p & 0xF0) >> 4);
		else /* lower */
			dev->lptbl[i] = dev->lptbl[i - 1] + (p & 0xF);
	}

	return 0;
}

517 518 519 520
static int nvm_core_init(struct nvm_dev *dev)
{
	struct nvm_id *id = &dev->identity;
	struct nvm_id_group *grp = &id->groups[0];
521
	int ret;
522 523 524 525 526 527 528

	/* device values */
	dev->nr_chnls = grp->num_ch;
	dev->luns_per_chnl = grp->num_lun;
	dev->pgs_per_blk = grp->num_pg;
	dev->blks_per_lun = grp->num_blk;
	dev->nr_planes = grp->num_pln;
529 530
	dev->fpg_size = grp->fpg_sz;
	dev->pfpg_size = grp->fpg_sz * grp->num_pln;
531 532 533
	dev->sec_size = grp->csecs;
	dev->oob_size = grp->sos;
	dev->sec_per_pg = grp->fpg_sz / grp->csecs;
M
Matias Bjørling 已提交
534
	dev->mccap = grp->mccap;
535
	memcpy(&dev->ppaf, &id->ppaf, sizeof(struct nvm_addr_format));
536 537 538 539 540 541 542 543 544

	dev->plane_mode = NVM_PLANE_SINGLE;
	dev->max_rq_size = dev->ops->max_phys_sect * dev->sec_size;

	if (grp->mpos & 0x020202)
		dev->plane_mode = NVM_PLANE_DOUBLE;
	if (grp->mpos & 0x040404)
		dev->plane_mode = NVM_PLANE_QUAD;

545 546 547 548 549
	if (grp->mtype != 0) {
		pr_err("nvm: memory type not supported\n");
		return -EINVAL;
	}

550 551 552 553 554 555
	/* calculated values */
	dev->sec_per_pl = dev->sec_per_pg * dev->nr_planes;
	dev->sec_per_blk = dev->sec_per_pl * dev->pgs_per_blk;
	dev->sec_per_lun = dev->sec_per_blk * dev->blks_per_lun;
	dev->nr_luns = dev->luns_per_chnl * dev->nr_chnls;

556
	dev->total_secs = dev->nr_luns * dev->sec_per_lun;
W
Wenwei Tao 已提交
557 558 559 560
	dev->lun_map = kcalloc(BITS_TO_LONGS(dev->nr_luns),
					sizeof(unsigned long), GFP_KERNEL);
	if (!dev->lun_map)
		return -ENOMEM;
561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580

	switch (grp->fmtype) {
	case NVM_ID_FMTYPE_SLC:
		if (nvm_init_slc_tbl(dev, grp)) {
			ret = -ENOMEM;
			goto err_fmtype;
		}
		break;
	case NVM_ID_FMTYPE_MLC:
		if (nvm_init_mlc_tbl(dev, grp)) {
			ret = -ENOMEM;
			goto err_fmtype;
		}
		break;
	default:
		pr_err("nvm: flash type not supported\n");
		ret = -EINVAL;
		goto err_fmtype;
	}

581
	mutex_init(&dev->mlock);
582
	spin_lock_init(&dev->lock);
583 584

	return 0;
585 586 587
err_fmtype:
	kfree(dev->lun_map);
	return ret;
588 589
}

590 591 592 593 594 595 596 597 598
static void nvm_free_mgr(struct nvm_dev *dev)
{
	if (!dev->mt)
		return;

	dev->mt->unregister_mgr(dev);
	dev->mt = NULL;
}

599 600 601 602 603
static void nvm_free(struct nvm_dev *dev)
{
	if (!dev)
		return;

604
	nvm_free_mgr(dev);
605 606

	kfree(dev->lptbl);
607
	kfree(dev->lun_map);
608 609 610 611
}

static int nvm_init(struct nvm_dev *dev)
{
612
	int ret = -EINVAL;
613 614

	if (!dev->q || !dev->ops)
615
		return ret;
616

617
	if (dev->ops->identity(dev, &dev->identity)) {
618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653
		pr_err("nvm: device could not be identified\n");
		goto err;
	}

	pr_debug("nvm: ver:%x nvm_vendor:%x groups:%u\n",
			dev->identity.ver_id, dev->identity.vmnt,
							dev->identity.cgrps);

	if (dev->identity.ver_id != 1) {
		pr_err("nvm: device not supported by kernel.");
		goto err;
	}

	if (dev->identity.cgrps != 1) {
		pr_err("nvm: only one group configuration supported.");
		goto err;
	}

	ret = nvm_core_init(dev);
	if (ret) {
		pr_err("nvm: could not initialize core structures.\n");
		goto err;
	}

	pr_info("nvm: registered %s [%u/%u/%u/%u/%u/%u]\n",
			dev->name, dev->sec_per_pg, dev->nr_planes,
			dev->pgs_per_blk, dev->blks_per_lun, dev->nr_luns,
			dev->nr_chnls);
	return 0;
err:
	pr_err("nvm: failed to initialize nvm\n");
	return ret;
}

static void nvm_exit(struct nvm_dev *dev)
{
654 655
	if (dev->dma_pool)
		dev->ops->destroy_dma_pool(dev->dma_pool);
656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681
	nvm_free(dev);

	pr_info("nvm: successfully unloaded\n");
}

int nvm_register(struct request_queue *q, char *disk_name,
							struct nvm_dev_ops *ops)
{
	struct nvm_dev *dev;
	int ret;

	if (!ops->identity)
		return -EINVAL;

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

	dev->q = q;
	dev->ops = ops;
	strncpy(dev->name, disk_name, DISK_NAME_LEN);

	ret = nvm_init(dev);
	if (ret)
		goto err_init;

682 683 684 685 686 687
	if (dev->ops->max_phys_sect > 256) {
		pr_info("nvm: max sectors supported is 256.\n");
		ret = -EINVAL;
		goto err_init;
	}

688
	if (dev->ops->max_phys_sect > 1) {
689 690 691
		dev->dma_pool = dev->ops->create_dma_pool(dev, "ppalist");
		if (!dev->dma_pool) {
			pr_err("nvm: could not create dma pool\n");
692 693
			ret = -ENOMEM;
			goto err_init;
694 695 696
		}
	}

697 698 699 700 701 702 703
	if (dev->identity.cap & NVM_ID_DCAP_BBLKMGMT) {
		ret = nvm_get_sysblock(dev, &dev->sb);
		if (!ret)
			pr_err("nvm: device not initialized.\n");
		else if (ret < 0)
			pr_err("nvm: err (%d) on device initialization\n", ret);
	}
704

705
	/* register device with a supported media manager */
706
	down_write(&nvm_lock);
707 708
	if (ret > 0)
		dev->mt = nvm_init_mgr(dev);
709 710 711
	list_add(&dev->devices, &nvm_devices);
	up_write(&nvm_lock);

712 713
	return 0;
err_init:
W
Wenwei Tao 已提交
714
	kfree(dev->lun_map);
715 716 717 718 719 720 721
	kfree(dev);
	return ret;
}
EXPORT_SYMBOL(nvm_register);

void nvm_unregister(char *disk_name)
{
W
Wenwei Tao 已提交
722
	struct nvm_dev *dev;
723

W
Wenwei Tao 已提交
724 725
	down_write(&nvm_lock);
	dev = nvm_find_nvm_dev(disk_name);
726 727 728
	if (!dev) {
		pr_err("nvm: could not find device %s to unregister\n",
								disk_name);
W
Wenwei Tao 已提交
729
		up_write(&nvm_lock);
730 731 732 733 734
		return;
	}

	list_del(&dev->devices);
	up_write(&nvm_lock);
735 736 737

	nvm_exit(dev);
	kfree(dev);
738 739 740 741 742 743 744 745
}
EXPORT_SYMBOL(nvm_unregister);

static int __nvm_configure_create(struct nvm_ioctl_create *create)
{
	struct nvm_dev *dev;
	struct nvm_ioctl_create_simple *s;

W
Wenwei Tao 已提交
746
	down_write(&nvm_lock);
747
	dev = nvm_find_nvm_dev(create->dev);
W
Wenwei Tao 已提交
748
	up_write(&nvm_lock);
749

750 751 752 753 754
	if (!dev) {
		pr_err("nvm: device not found\n");
		return -EINVAL;
	}

755 756 757 758 759
	if (!dev->mt) {
		pr_info("nvm: device has no media manager registered.\n");
		return -ENODEV;
	}

760 761 762 763 764 765 766 767 768 769 770 771
	if (create->conf.type != NVM_CONFIG_TYPE_SIMPLE) {
		pr_err("nvm: config type not valid\n");
		return -EINVAL;
	}
	s = &create->conf.s;

	if (s->lun_begin > s->lun_end || s->lun_end > dev->nr_luns) {
		pr_err("nvm: lun out of bound (%u:%u > %u)\n",
			s->lun_begin, s->lun_end, dev->nr_luns);
		return -EINVAL;
	}

772
	return dev->mt->create_tgt(dev, create);
773 774 775 776 777 778 779 780 781 782 783 784 785 786 787
}

#ifdef CONFIG_NVM_DEBUG
static int nvm_configure_show(const char *val)
{
	struct nvm_dev *dev;
	char opcode, devname[DISK_NAME_LEN];
	int ret;

	ret = sscanf(val, "%c %32s", &opcode, devname);
	if (ret != 2) {
		pr_err("nvm: invalid command. Use \"opcode devicename\".\n");
		return -EINVAL;
	}

W
Wenwei Tao 已提交
788
	down_write(&nvm_lock);
789
	dev = nvm_find_nvm_dev(devname);
W
Wenwei Tao 已提交
790
	up_write(&nvm_lock);
791 792 793 794 795 796 797 798
	if (!dev) {
		pr_err("nvm: device not found\n");
		return -EINVAL;
	}

	if (!dev->mt)
		return 0;

799
	dev->mt->lun_info_print(dev);
800 801 802 803 804 805 806

	return 0;
}

static int nvm_configure_remove(const char *val)
{
	struct nvm_ioctl_remove remove;
807
	struct nvm_dev *dev;
808
	char opcode;
809
	int ret = 0;
810 811 812 813 814 815 816 817 818

	ret = sscanf(val, "%c %256s", &opcode, remove.tgtname);
	if (ret != 2) {
		pr_err("nvm: invalid command. Use \"d targetname\".\n");
		return -EINVAL;
	}

	remove.flags = 0;

819 820 821 822 823 824 825
	list_for_each_entry(dev, &nvm_devices, devices) {
		ret = dev->mt->remove_tgt(dev, &remove);
		if (!ret)
			break;
	}

	return ret;
826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880
}

static int nvm_configure_create(const char *val)
{
	struct nvm_ioctl_create create;
	char opcode;
	int lun_begin, lun_end, ret;

	ret = sscanf(val, "%c %256s %256s %48s %u:%u", &opcode, create.dev,
						create.tgtname, create.tgttype,
						&lun_begin, &lun_end);
	if (ret != 6) {
		pr_err("nvm: invalid command. Use \"opcode device name tgttype lun_begin:lun_end\".\n");
		return -EINVAL;
	}

	create.flags = 0;
	create.conf.type = NVM_CONFIG_TYPE_SIMPLE;
	create.conf.s.lun_begin = lun_begin;
	create.conf.s.lun_end = lun_end;

	return __nvm_configure_create(&create);
}


/* Exposes administrative interface through /sys/module/lnvm/configure_by_str */
static int nvm_configure_by_str_event(const char *val,
					const struct kernel_param *kp)
{
	char opcode;
	int ret;

	ret = sscanf(val, "%c", &opcode);
	if (ret != 1) {
		pr_err("nvm: string must have the format of \"cmd ...\"\n");
		return -EINVAL;
	}

	switch (opcode) {
	case 'a':
		return nvm_configure_create(val);
	case 'd':
		return nvm_configure_remove(val);
	case 's':
		return nvm_configure_show(val);
	default:
		pr_err("nvm: invalid command\n");
		return -EINVAL;
	}

	return 0;
}

static int nvm_configure_get(char *buf, const struct kernel_param *kp)
{
881
	int sz;
882 883
	struct nvm_dev *dev;

884
	sz = sprintf(buf, "available devices:\n");
885 886
	down_write(&nvm_lock);
	list_for_each_entry(dev, &nvm_devices, devices) {
887
		if (sz > 4095 - DISK_NAME_LEN - 2)
888
			break;
889
		sz += sprintf(buf + sz, " %32s\n", dev->name);
890 891 892
	}
	up_write(&nvm_lock);

893
	return sz;
894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926
}

static const struct kernel_param_ops nvm_configure_by_str_event_param_ops = {
	.set	= nvm_configure_by_str_event,
	.get	= nvm_configure_get,
};

#undef MODULE_PARAM_PREFIX
#define MODULE_PARAM_PREFIX	"lnvm."

module_param_cb(configure_debug, &nvm_configure_by_str_event_param_ops, NULL,
									0644);

#endif /* CONFIG_NVM_DEBUG */

static long nvm_ioctl_info(struct file *file, void __user *arg)
{
	struct nvm_ioctl_info *info;
	struct nvm_tgt_type *tt;
	int tgt_iter = 0;

	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

	info = memdup_user(arg, sizeof(struct nvm_ioctl_info));
	if (IS_ERR(info))
		return -EFAULT;

	info->version[0] = NVM_VERSION_MAJOR;
	info->version[1] = NVM_VERSION_MINOR;
	info->version[2] = NVM_VERSION_PATCH;

	down_write(&nvm_lock);
927
	list_for_each_entry(tt, &nvm_tgt_types, list) {
928 929 930 931 932 933 934 935 936 937 938 939 940
		struct nvm_ioctl_info_tgt *tgt = &info->tgts[tgt_iter];

		tgt->version[0] = tt->version[0];
		tgt->version[1] = tt->version[1];
		tgt->version[2] = tt->version[2];
		strncpy(tgt->tgtname, tt->name, NVM_TTYPE_NAME_MAX);

		tgt_iter++;
	}

	info->tgtsize = tgt_iter;
	up_write(&nvm_lock);

S
Sudip Mukherjee 已提交
941 942
	if (copy_to_user(arg, info, sizeof(struct nvm_ioctl_info))) {
		kfree(info);
943
		return -EFAULT;
S
Sudip Mukherjee 已提交
944
	}
945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986

	kfree(info);
	return 0;
}

static long nvm_ioctl_get_devices(struct file *file, void __user *arg)
{
	struct nvm_ioctl_get_devices *devices;
	struct nvm_dev *dev;
	int i = 0;

	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

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

	down_write(&nvm_lock);
	list_for_each_entry(dev, &nvm_devices, devices) {
		struct nvm_ioctl_device_info *info = &devices->info[i];

		sprintf(info->devname, "%s", dev->name);
		if (dev->mt) {
			info->bmversion[0] = dev->mt->version[0];
			info->bmversion[1] = dev->mt->version[1];
			info->bmversion[2] = dev->mt->version[2];
			sprintf(info->bmname, "%s", dev->mt->name);
		} else {
			sprintf(info->bmname, "none");
		}

		i++;
		if (i > 31) {
			pr_err("nvm: max 31 devices can be reported.\n");
			break;
		}
	}
	up_write(&nvm_lock);

	devices->nr_devices = i;

S
Sudip Mukherjee 已提交
987 988 989
	if (copy_to_user(arg, devices,
			 sizeof(struct nvm_ioctl_get_devices))) {
		kfree(devices);
990
		return -EFAULT;
S
Sudip Mukherjee 已提交
991
	}
992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021

	kfree(devices);
	return 0;
}

static long nvm_ioctl_dev_create(struct file *file, void __user *arg)
{
	struct nvm_ioctl_create create;

	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

	if (copy_from_user(&create, arg, sizeof(struct nvm_ioctl_create)))
		return -EFAULT;

	create.dev[DISK_NAME_LEN - 1] = '\0';
	create.tgttype[NVM_TTYPE_NAME_MAX - 1] = '\0';
	create.tgtname[DISK_NAME_LEN - 1] = '\0';

	if (create.flags != 0) {
		pr_err("nvm: no flags supported\n");
		return -EINVAL;
	}

	return __nvm_configure_create(&create);
}

static long nvm_ioctl_dev_remove(struct file *file, void __user *arg)
{
	struct nvm_ioctl_remove remove;
1022 1023
	struct nvm_dev *dev;
	int ret = 0;
1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037

	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

	if (copy_from_user(&remove, arg, sizeof(struct nvm_ioctl_remove)))
		return -EFAULT;

	remove.tgtname[DISK_NAME_LEN - 1] = '\0';

	if (remove.flags != 0) {
		pr_err("nvm: no flags supported\n");
		return -EINVAL;
	}

1038 1039 1040 1041 1042 1043 1044
	list_for_each_entry(dev, &nvm_devices, devices) {
		ret = dev->mt->remove_tgt(dev, &remove);
		if (!ret)
			break;
	}

	return ret;
1045 1046
}

1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057
static void nvm_setup_nvm_sb_info(struct nvm_sb_info *info)
{
	info->seqnr = 1;
	info->erase_cnt = 0;
	info->version = 1;
}

static long __nvm_ioctl_dev_init(struct nvm_ioctl_dev_init *init)
{
	struct nvm_dev *dev;
	struct nvm_sb_info info;
1058
	int ret;
1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072

	down_write(&nvm_lock);
	dev = nvm_find_nvm_dev(init->dev);
	up_write(&nvm_lock);
	if (!dev) {
		pr_err("nvm: device not found\n");
		return -EINVAL;
	}

	nvm_setup_nvm_sb_info(&info);

	strncpy(info.mmtype, init->mmtype, NVM_MMTYPE_LEN);
	info.fs_ppa.ppa = -1;

1073 1074 1075 1076 1077
	if (dev->identity.cap & NVM_ID_DCAP_BBLKMGMT) {
		ret = nvm_init_sysblock(dev, &info);
		if (ret)
			return ret;
	}
1078 1079 1080 1081 1082 1083 1084 1085

	memcpy(&dev->sb, &info, sizeof(struct nvm_sb_info));

	down_write(&nvm_lock);
	dev->mt = nvm_init_mgr(dev);
	up_write(&nvm_lock);

	return 0;
1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107
}

static long nvm_ioctl_dev_init(struct file *file, void __user *arg)
{
	struct nvm_ioctl_dev_init init;

	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

	if (copy_from_user(&init, arg, sizeof(struct nvm_ioctl_dev_init)))
		return -EFAULT;

	if (init.flags != 0) {
		pr_err("nvm: no flags supported\n");
		return -EINVAL;
	}

	init.dev[DISK_NAME_LEN - 1] = '\0';

	return __nvm_ioctl_dev_init(&init);
}

1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131
static long nvm_ioctl_dev_factory(struct file *file, void __user *arg)
{
	struct nvm_ioctl_dev_factory fact;
	struct nvm_dev *dev;

	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

	if (copy_from_user(&fact, arg, sizeof(struct nvm_ioctl_dev_factory)))
		return -EFAULT;

	fact.dev[DISK_NAME_LEN - 1] = '\0';

	if (fact.flags & ~(NVM_FACTORY_NR_BITS - 1))
		return -EINVAL;

	down_write(&nvm_lock);
	dev = nvm_find_nvm_dev(fact.dev);
	up_write(&nvm_lock);
	if (!dev) {
		pr_err("nvm: device not found\n");
		return -EINVAL;
	}

1132
	nvm_free_mgr(dev);
1133

1134 1135 1136 1137
	if (dev->identity.cap & NVM_ID_DCAP_BBLKMGMT)
		return nvm_dev_factory(dev, fact.flags);

	return 0;
1138 1139
}

1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152
static long nvm_ctl_ioctl(struct file *file, uint cmd, unsigned long arg)
{
	void __user *argp = (void __user *)arg;

	switch (cmd) {
	case NVM_INFO:
		return nvm_ioctl_info(file, argp);
	case NVM_GET_DEVICES:
		return nvm_ioctl_get_devices(file, argp);
	case NVM_DEV_CREATE:
		return nvm_ioctl_dev_create(file, argp);
	case NVM_DEV_REMOVE:
		return nvm_ioctl_dev_remove(file, argp);
1153 1154
	case NVM_DEV_INIT:
		return nvm_ioctl_dev_init(file, argp);
1155 1156
	case NVM_DEV_FACTORY:
		return nvm_ioctl_dev_factory(file, argp);
1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197
	}
	return 0;
}

static const struct file_operations _ctl_fops = {
	.open = nonseekable_open,
	.unlocked_ioctl = nvm_ctl_ioctl,
	.owner = THIS_MODULE,
	.llseek  = noop_llseek,
};

static struct miscdevice _nvm_misc = {
	.minor		= MISC_DYNAMIC_MINOR,
	.name		= "lightnvm",
	.nodename	= "lightnvm/control",
	.fops		= &_ctl_fops,
};

MODULE_ALIAS_MISCDEV(MISC_DYNAMIC_MINOR);

static int __init nvm_mod_init(void)
{
	int ret;

	ret = misc_register(&_nvm_misc);
	if (ret)
		pr_err("nvm: misc_register failed for control device");

	return ret;
}

static void __exit nvm_mod_exit(void)
{
	misc_deregister(&_nvm_misc);
}

MODULE_AUTHOR("Matias Bjorling <m@bjorling.me>");
MODULE_LICENSE("GPL v2");
MODULE_VERSION("0.1");
module_init(nvm_mod_init);
module_exit(nvm_mod_exit);