dimm_devs.c 20.8 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13
/*
 * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of version 2 of the GNU General Public License 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.
 */
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
14
#include <linux/vmalloc.h>
15
#include <linux/device.h>
16
#include <linux/ndctl.h>
17 18 19 20 21
#include <linux/slab.h>
#include <linux/io.h>
#include <linux/fs.h>
#include <linux/mm.h>
#include "nd-core.h"
D
Dan Williams 已提交
22
#include "label.h"
23
#include "pmem.h"
24
#include "nd.h"
25 26 27

static DEFINE_IDA(dimm_ida);

28 29 30 31
/*
 * Retrieve bus and dimm handle and return if this bus supports
 * get_config_data commands
 */
32
int nvdimm_check_config_data(struct device *dev)
33
{
34
	struct nvdimm *nvdimm = to_nvdimm(dev);
35

36 37
	if (!nvdimm->cmd_mask ||
	    !test_bit(ND_CMD_GET_CONFIG_DATA, &nvdimm->cmd_mask)) {
38
		if (test_bit(NDD_ALIASING, &nvdimm->flags))
39 40 41 42
			return -ENXIO;
		else
			return -ENOTTY;
	}
43 44 45 46 47 48

	return 0;
}

static int validate_dimm(struct nvdimm_drvdata *ndd)
{
49
	int rc;
50

51 52 53 54 55
	if (!ndd)
		return -EINVAL;

	rc = nvdimm_check_config_data(ndd->dev);
	if (rc)
56 57 58 59 60 61 62 63 64 65 66 67 68 69 70
		dev_dbg(ndd->dev, "%pf: %s error: %d\n",
				__builtin_return_address(0), __func__, rc);
	return rc;
}

/**
 * nvdimm_init_nsarea - determine the geometry of a dimm's namespace area
 * @nvdimm: dimm to initialize
 */
int nvdimm_init_nsarea(struct nvdimm_drvdata *ndd)
{
	struct nd_cmd_get_config_size *cmd = &ndd->nsarea;
	struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(ndd->dev);
	struct nvdimm_bus_descriptor *nd_desc;
	int rc = validate_dimm(ndd);
71
	int cmd_rc = 0;
72 73 74 75 76 77 78 79 80

	if (rc)
		return rc;

	if (cmd->config_size)
		return 0; /* already valid */

	memset(cmd, 0, sizeof(*cmd));
	nd_desc = nvdimm_bus->nd_desc;
81 82 83 84 85
	rc = nd_desc->ndctl(nd_desc, to_nvdimm(ndd->dev),
			ND_CMD_GET_CONFIG_SIZE, cmd, sizeof(*cmd), &cmd_rc);
	if (rc < 0)
		return rc;
	return cmd_rc;
86 87 88 89 90
}

int nvdimm_init_config_data(struct nvdimm_drvdata *ndd)
{
	struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(ndd->dev);
91
	int rc = validate_dimm(ndd), cmd_rc = 0;
92 93 94 95 96 97 98 99 100 101 102
	struct nd_cmd_get_config_data_hdr *cmd;
	struct nvdimm_bus_descriptor *nd_desc;
	u32 max_cmd_size, config_size;
	size_t offset;

	if (rc)
		return rc;

	if (ndd->data)
		return 0;

103 104 105 106
	if (ndd->nsarea.status || ndd->nsarea.max_xfer == 0
			|| ndd->nsarea.config_size < ND_LABEL_MIN_SIZE) {
		dev_dbg(ndd->dev, "failed to init config data area: (%d:%d)\n",
				ndd->nsarea.max_xfer, ndd->nsarea.config_size);
107
		return -ENXIO;
108
	}
109

110
	ndd->data = kvmalloc(ndd->nsarea.config_size, GFP_KERNEL);
111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126
	if (!ndd->data)
		return -ENOMEM;

	max_cmd_size = min_t(u32, PAGE_SIZE, ndd->nsarea.max_xfer);
	cmd = kzalloc(max_cmd_size + sizeof(*cmd), GFP_KERNEL);
	if (!cmd)
		return -ENOMEM;

	nd_desc = nvdimm_bus->nd_desc;
	for (config_size = ndd->nsarea.config_size, offset = 0;
			config_size; config_size -= cmd->in_length,
			offset += cmd->in_length) {
		cmd->in_length = min(config_size, max_cmd_size);
		cmd->in_offset = offset;
		rc = nd_desc->ndctl(nd_desc, to_nvdimm(ndd->dev),
				ND_CMD_GET_CONFIG_DATA, cmd,
127 128 129 130 131
				cmd->in_length + sizeof(*cmd), &cmd_rc);
		if (rc < 0)
			break;
		if (cmd_rc < 0) {
			rc = cmd_rc;
132 133 134 135
			break;
		}
		memcpy(ndd->data + offset, cmd->out_buf, cmd->in_length);
	}
136
	dev_dbg(ndd->dev, "len: %zu rc: %d\n", offset, rc);
137 138 139 140 141
	kfree(cmd);

	return rc;
}

D
Dan Williams 已提交
142 143 144 145 146
int nvdimm_set_config_data(struct nvdimm_drvdata *ndd, size_t offset,
		void *buf, size_t len)
{
	size_t max_cmd_size, buf_offset;
	struct nd_cmd_set_config_hdr *cmd;
147
	int rc = validate_dimm(ndd), cmd_rc = 0;
D
Dan Williams 已提交
148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177
	struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(ndd->dev);
	struct nvdimm_bus_descriptor *nd_desc = nvdimm_bus->nd_desc;

	if (rc)
		return rc;

	if (!ndd->data)
		return -ENXIO;

	if (offset + len > ndd->nsarea.config_size)
		return -ENXIO;

	max_cmd_size = min_t(u32, PAGE_SIZE, len);
	max_cmd_size = min_t(u32, max_cmd_size, ndd->nsarea.max_xfer);
	cmd = kzalloc(max_cmd_size + sizeof(*cmd) + sizeof(u32), GFP_KERNEL);
	if (!cmd)
		return -ENOMEM;

	for (buf_offset = 0; len; len -= cmd->in_length,
			buf_offset += cmd->in_length) {
		size_t cmd_size;

		cmd->in_offset = offset + buf_offset;
		cmd->in_length = min(max_cmd_size, len);
		memcpy(cmd->in_buf, buf + buf_offset, cmd->in_length);

		/* status is output in the last 4-bytes of the command buffer */
		cmd_size = sizeof(*cmd) + cmd->in_length + sizeof(u32);

		rc = nd_desc->ndctl(nd_desc, to_nvdimm(ndd->dev),
178 179 180 181 182
				ND_CMD_SET_CONFIG_DATA, cmd, cmd_size, &cmd_rc);
		if (rc < 0)
			break;
		if (cmd_rc < 0) {
			rc = cmd_rc;
D
Dan Williams 已提交
183 184 185 186 187 188 189 190
			break;
		}
	}
	kfree(cmd);

	return rc;
}

191 192 193 194
void nvdimm_set_aliasing(struct device *dev)
{
	struct nvdimm *nvdimm = to_nvdimm(dev);

195 196 197 198 199 200 201 202
	set_bit(NDD_ALIASING, &nvdimm->flags);
}

void nvdimm_set_locked(struct device *dev)
{
	struct nvdimm *nvdimm = to_nvdimm(dev);

	set_bit(NDD_LOCKED, &nvdimm->flags);
203 204
}

205 206 207 208 209 210 211
void nvdimm_clear_locked(struct device *dev)
{
	struct nvdimm *nvdimm = to_nvdimm(dev);

	clear_bit(NDD_LOCKED, &nvdimm->flags);
}

212 213 214 215 216 217 218 219 220 221 222 223 224
static void nvdimm_release(struct device *dev)
{
	struct nvdimm *nvdimm = to_nvdimm(dev);

	ida_simple_remove(&dimm_ida, nvdimm->id);
	kfree(nvdimm);
}

static struct device_type nvdimm_device_type = {
	.name = "nvdimm",
	.release = nvdimm_release,
};

225
bool is_nvdimm(struct device *dev)
226 227 228 229 230 231 232 233 234 235 236 237 238
{
	return dev->type == &nvdimm_device_type;
}

struct nvdimm *to_nvdimm(struct device *dev)
{
	struct nvdimm *nvdimm = container_of(dev, struct nvdimm, dev);

	WARN_ON(!is_nvdimm(dev));
	return nvdimm;
}
EXPORT_SYMBOL_GPL(to_nvdimm);

239 240 241 242 243 244 245 246 247
struct nvdimm *nd_blk_region_to_dimm(struct nd_blk_region *ndbr)
{
	struct nd_region *nd_region = &ndbr->nd_region;
	struct nd_mapping *nd_mapping = &nd_region->mapping[0];

	return nd_mapping->nvdimm;
}
EXPORT_SYMBOL_GPL(nd_blk_region_to_dimm);

248 249 250 251 252 253 254
unsigned long nd_blk_memremap_flags(struct nd_blk_region *ndbr)
{
	/* pmem mapping properties are private to libnvdimm */
	return ARCH_MEMREMAP_PMEM;
}
EXPORT_SYMBOL_GPL(nd_blk_memremap_flags);

255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270
struct nvdimm_drvdata *to_ndd(struct nd_mapping *nd_mapping)
{
	struct nvdimm *nvdimm = nd_mapping->nvdimm;

	WARN_ON_ONCE(!is_nvdimm_bus_locked(&nvdimm->dev));

	return dev_get_drvdata(&nvdimm->dev);
}
EXPORT_SYMBOL(to_ndd);

void nvdimm_drvdata_release(struct kref *kref)
{
	struct nvdimm_drvdata *ndd = container_of(kref, typeof(*ndd), kref);
	struct device *dev = ndd->dev;
	struct resource *res, *_r;

271
	dev_dbg(dev, "trace\n");
272 273 274 275 276
	nvdimm_bus_lock(dev);
	for_each_dpa_resource_safe(ndd, res, _r)
		nvdimm_free_dpa(ndd, res);
	nvdimm_bus_unlock(dev);

277
	kvfree(ndd->data);
278 279 280 281 282 283 284 285 286 287 288 289 290 291 292
	kfree(ndd);
	put_device(dev);
}

void get_ndd(struct nvdimm_drvdata *ndd)
{
	kref_get(&ndd->kref);
}

void put_ndd(struct nvdimm_drvdata *ndd)
{
	if (ndd)
		kref_put(&ndd->kref, nvdimm_drvdata_release);
}

293 294 295 296 297 298
const char *nvdimm_name(struct nvdimm *nvdimm)
{
	return dev_name(&nvdimm->dev);
}
EXPORT_SYMBOL_GPL(nvdimm_name);

299 300 301 302 303 304
struct kobject *nvdimm_kobj(struct nvdimm *nvdimm)
{
	return &nvdimm->dev.kobj;
}
EXPORT_SYMBOL_GPL(nvdimm_kobj);

305 306 307 308 309 310
unsigned long nvdimm_cmd_mask(struct nvdimm *nvdimm)
{
	return nvdimm->cmd_mask;
}
EXPORT_SYMBOL_GPL(nvdimm_cmd_mask);

311 312
void *nvdimm_provider_data(struct nvdimm *nvdimm)
{
313 314 315
	if (nvdimm)
		return nvdimm->provider_data;
	return NULL;
316 317 318
}
EXPORT_SYMBOL_GPL(nvdimm_provider_data);

319 320 321 322 323 324
static ssize_t commands_show(struct device *dev,
		struct device_attribute *attr, char *buf)
{
	struct nvdimm *nvdimm = to_nvdimm(dev);
	int cmd, len = 0;

325
	if (!nvdimm->cmd_mask)
326 327
		return sprintf(buf, "\n");

328
	for_each_set_bit(cmd, &nvdimm->cmd_mask, BITS_PER_LONG)
329 330 331 332 333 334
		len += sprintf(buf + len, "%s ", nvdimm_cmd_name(cmd));
	len += sprintf(buf + len, "\n");
	return len;
}
static DEVICE_ATTR_RO(commands);

335 336 337 338 339 340 341 342 343 344 345
static ssize_t flags_show(struct device *dev,
		struct device_attribute *attr, char *buf)
{
	struct nvdimm *nvdimm = to_nvdimm(dev);

	return sprintf(buf, "%s%s\n",
			test_bit(NDD_ALIASING, &nvdimm->flags) ? "alias " : "",
			test_bit(NDD_LOCKED, &nvdimm->flags) ? "lock " : "");
}
static DEVICE_ATTR_RO(flags);

346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361
static ssize_t state_show(struct device *dev, struct device_attribute *attr,
		char *buf)
{
	struct nvdimm *nvdimm = to_nvdimm(dev);

	/*
	 * The state may be in the process of changing, userspace should
	 * quiesce probing if it wants a static answer
	 */
	nvdimm_bus_lock(dev);
	nvdimm_bus_unlock(dev);
	return sprintf(buf, "%s\n", atomic_read(&nvdimm->busy)
			? "active" : "idle");
}
static DEVICE_ATTR_RO(state);

D
Dan Williams 已提交
362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384
static ssize_t available_slots_show(struct device *dev,
		struct device_attribute *attr, char *buf)
{
	struct nvdimm_drvdata *ndd = dev_get_drvdata(dev);
	ssize_t rc;
	u32 nfree;

	if (!ndd)
		return -ENXIO;

	nvdimm_bus_lock(dev);
	nfree = nd_label_nfree(ndd);
	if (nfree - 1 > nfree) {
		dev_WARN_ONCE(dev, 1, "we ate our last label?\n");
		nfree = 0;
	} else
		nfree--;
	rc = sprintf(buf, "%d\n", nfree);
	nvdimm_bus_unlock(dev);
	return rc;
}
static DEVICE_ATTR_RO(available_slots);

385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404
static ssize_t security_show(struct device *dev,
		struct device_attribute *attr, char *buf)
{
	struct nvdimm *nvdimm = to_nvdimm(dev);

	switch (nvdimm->sec.state) {
	case NVDIMM_SECURITY_DISABLED:
		return sprintf(buf, "disabled\n");
	case NVDIMM_SECURITY_UNLOCKED:
		return sprintf(buf, "unlocked\n");
	case NVDIMM_SECURITY_LOCKED:
		return sprintf(buf, "locked\n");
	case NVDIMM_SECURITY_FROZEN:
		return sprintf(buf, "frozen\n");
	case NVDIMM_SECURITY_OVERWRITE:
		return sprintf(buf, "overwrite\n");
	}

	return -ENOTTY;
}
405

406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422
#define OPS						\
	C( OP_FREEZE,        "freeze",        1),	\
	C( OP_DISABLE,       "disable",       2)
#undef C
#define C(a, b, c) a
enum nvdimmsec_op_ids { OPS };
#undef C
#define C(a, b, c) { b, c }
static struct {
	const char *name;
	int args;
} ops[] = { OPS };
#undef C

#define SEC_CMD_SIZE 32
#define KEY_ID_SIZE 10

423 424 425 426
static ssize_t __security_store(struct device *dev, const char *buf, size_t len)
{
	struct nvdimm *nvdimm = to_nvdimm(dev);
	ssize_t rc;
427 428 429 430
	char cmd[SEC_CMD_SIZE+1], keystr[KEY_ID_SIZE+1],
		nkeystr[KEY_ID_SIZE+1];
	unsigned int key, newkey;
	int i;
431 432 433 434

	if (atomic_read(&nvdimm->busy))
		return -EBUSY;

435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453
	rc = sscanf(buf, "%"__stringify(SEC_CMD_SIZE)"s"
			" %"__stringify(KEY_ID_SIZE)"s"
			" %"__stringify(KEY_ID_SIZE)"s",
			cmd, keystr, nkeystr);
	if (rc < 1)
		return -EINVAL;
	for (i = 0; i < ARRAY_SIZE(ops); i++)
		if (sysfs_streq(cmd, ops[i].name))
			break;
	if (i >= ARRAY_SIZE(ops))
		return -EINVAL;
	if (ops[i].args > 1)
		rc = kstrtouint(keystr, 0, &key);
	if (rc >= 0 && ops[i].args > 2)
		rc = kstrtouint(nkeystr, 0, &newkey);
	if (rc < 0)
		return rc;

	if (i == OP_FREEZE) {
454 455
		dev_dbg(dev, "freeze\n");
		rc = nvdimm_security_freeze(nvdimm);
456 457 458
	} else if (i == OP_DISABLE) {
		dev_dbg(dev, "disable %u\n", key);
		rc = nvdimm_security_disable(nvdimm, key);
459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487
	} else
		return -EINVAL;

	if (rc == 0)
		rc = len;
	return rc;
}

static ssize_t security_store(struct device *dev,
		struct device_attribute *attr, const char *buf, size_t len)

{
	ssize_t rc;

	/*
	 * Require all userspace triggered security management to be
	 * done while probing is idle and the DIMM is not in active use
	 * in any region.
	 */
	device_lock(dev);
	nvdimm_bus_lock(dev);
	wait_nvdimm_bus_probe_idle(dev);
	rc = __security_store(dev, buf, len);
	nvdimm_bus_unlock(dev);
	device_unlock(dev);

	return rc;
}
static DEVICE_ATTR_RW(security);
488

489
static struct attribute *nvdimm_attributes[] = {
490
	&dev_attr_state.attr,
491
	&dev_attr_flags.attr,
492
	&dev_attr_commands.attr,
D
Dan Williams 已提交
493
	&dev_attr_available_slots.attr,
494
	&dev_attr_security.attr,
495 496 497
	NULL,
};

498 499 500 501 502 503 504 505 506
static umode_t nvdimm_visible(struct kobject *kobj, struct attribute *a, int n)
{
	struct device *dev = container_of(kobj, typeof(*dev), kobj);
	struct nvdimm *nvdimm = to_nvdimm(dev);

	if (a != &dev_attr_security.attr)
		return a->mode;
	if (nvdimm->sec.state < 0)
		return 0;
507
	/* Are there any state mutation ops? */
508
	if (nvdimm->sec.ops->freeze || nvdimm->sec.ops->disable)
509 510
		return a->mode;
	return 0444;
511 512
}

513 514
struct attribute_group nvdimm_attribute_group = {
	.attrs = nvdimm_attributes,
515
	.is_visible = nvdimm_visible,
516 517 518
};
EXPORT_SYMBOL_GPL(nvdimm_attribute_group);

519 520 521
struct nvdimm *__nvdimm_create(struct nvdimm_bus *nvdimm_bus,
		void *provider_data, const struct attribute_group **groups,
		unsigned long flags, unsigned long cmd_mask, int num_flush,
522 523
		struct resource *flush_wpq, const char *dimm_id,
		const struct nvdimm_security_ops *sec_ops)
524 525 526 527 528 529 530 531 532 533 534 535
{
	struct nvdimm *nvdimm = kzalloc(sizeof(*nvdimm), GFP_KERNEL);
	struct device *dev;

	if (!nvdimm)
		return NULL;

	nvdimm->id = ida_simple_get(&dimm_ida, 0, 0, GFP_KERNEL);
	if (nvdimm->id < 0) {
		kfree(nvdimm);
		return NULL;
	}
536 537

	nvdimm->dimm_id = dimm_id;
538 539
	nvdimm->provider_data = provider_data;
	nvdimm->flags = flags;
540
	nvdimm->cmd_mask = cmd_mask;
541 542
	nvdimm->num_flush = num_flush;
	nvdimm->flush_wpq = flush_wpq;
543
	atomic_set(&nvdimm->busy, 0);
544 545 546 547
	dev = &nvdimm->dev;
	dev_set_name(dev, "nmem%d", nvdimm->id);
	dev->parent = &nvdimm_bus->dev;
	dev->type = &nvdimm_device_type;
548
	dev->devt = MKDEV(nvdimm_major, nvdimm->id);
549
	dev->groups = groups;
550 551 552 553 554 555
	nvdimm->sec.ops = sec_ops;
	/*
	 * Security state must be initialized before device_add() for
	 * attribute visibility.
	 */
	nvdimm->sec.state = nvdimm_security_state(nvdimm);
556
	nd_device_register(dev);
557 558 559

	return nvdimm;
}
560
EXPORT_SYMBOL_GPL(__nvdimm_create);
561

562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579
int nvdimm_security_freeze(struct nvdimm *nvdimm)
{
	int rc;

	WARN_ON_ONCE(!is_nvdimm_bus_locked(&nvdimm->dev));

	if (!nvdimm->sec.ops || !nvdimm->sec.ops->freeze)
		return -EOPNOTSUPP;

	if (nvdimm->sec.state < 0)
		return -EIO;

	rc = nvdimm->sec.ops->freeze(nvdimm);
	nvdimm->sec.state = nvdimm_security_state(nvdimm);

	return rc;
}

580
int alias_dpa_busy(struct device *dev, void *data)
581
{
582
	resource_size_t map_end, blk_start, new;
583 584 585 586 587 588 589
	struct blk_alloc_info *info = data;
	struct nd_mapping *nd_mapping;
	struct nd_region *nd_region;
	struct nvdimm_drvdata *ndd;
	struct resource *res;
	int i;

590
	if (!is_memory(dev))
591 592 593 594 595 596 597 598 599 600 601 602 603 604 605
		return 0;

	nd_region = to_nd_region(dev);
	for (i = 0; i < nd_region->ndr_mappings; i++) {
		nd_mapping  = &nd_region->mapping[i];
		if (nd_mapping->nvdimm == info->nd_mapping->nvdimm)
			break;
	}

	if (i >= nd_region->ndr_mappings)
		return 0;

	ndd = to_ndd(nd_mapping);
	map_end = nd_mapping->start + nd_mapping->size - 1;
	blk_start = nd_mapping->start;
606 607 608 609 610 611 612 613 614 615 616 617 618 619

	/*
	 * In the allocation case ->res is set to free space that we are
	 * looking to validate against PMEM aliasing collision rules
	 * (i.e. BLK is allocated after all aliased PMEM).
	 */
	if (info->res) {
		if (info->res->start >= nd_mapping->start
				&& info->res->start < map_end)
			/* pass */;
		else
			return 0;
	}

620 621 622
 retry:
	/*
	 * Find the free dpa from the end of the last pmem allocation to
623
	 * the end of the interleave-set mapping.
624 625
	 */
	for_each_dpa_resource(ndd, res) {
626 627
		if (strncmp(res->name, "pmem", 4) != 0)
			continue;
628 629 630
		if ((res->start >= blk_start && res->start < map_end)
				|| (res->end >= blk_start
					&& res->end <= map_end)) {
631 632 633 634 635
			new = max(blk_start, min(map_end + 1, res->end + 1));
			if (new != blk_start) {
				blk_start = new;
				goto retry;
			}
636 637 638
		}
	}

639 640 641 642 643 644 645 646
	/* update the free space range with the probed blk_start */
	if (info->res && blk_start > info->res->start) {
		info->res->start = max(info->res->start, blk_start);
		if (info->res->start > info->res->end)
			info->res->end = info->res->start - 1;
		return 1;
	}

647
	info->available -= blk_start - nd_mapping->start;
648

649 650 651
	return 0;
}

652 653 654 655
/**
 * nd_blk_available_dpa - account the unused dpa of BLK region
 * @nd_mapping: container of dpa-resource-root + labels
 *
656 657 658
 * Unlike PMEM, BLK namespaces can occupy discontiguous DPA ranges, but
 * we arrange for them to never start at an lower dpa than the last
 * PMEM allocation in an aliased region.
659
 */
660
resource_size_t nd_blk_available_dpa(struct nd_region *nd_region)
661
{
662 663
	struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(&nd_region->dev);
	struct nd_mapping *nd_mapping = &nd_region->mapping[0];
664
	struct nvdimm_drvdata *ndd = to_ndd(nd_mapping);
665 666 667
	struct blk_alloc_info info = {
		.nd_mapping = nd_mapping,
		.available = nd_mapping->size,
668
		.res = NULL,
669
	};
670 671 672 673 674
	struct resource *res;

	if (!ndd)
		return 0;

675
	device_for_each_child(&nvdimm_bus->dev, &info, alias_dpa_busy);
676

677 678 679 680
	/* now account for busy blk allocations in unaliased dpa */
	for_each_dpa_resource(ndd, res) {
		if (strncmp(res->name, "blk", 3) != 0)
			continue;
681
		info.available -= resource_size(res);
682 683 684
	}

	return info.available;
685 686
}

687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717
/**
 * nd_pmem_max_contiguous_dpa - For the given dimm+region, return the max
 *			   contiguous unallocated dpa range.
 * @nd_region: constrain available space check to this reference region
 * @nd_mapping: container of dpa-resource-root + labels
 */
resource_size_t nd_pmem_max_contiguous_dpa(struct nd_region *nd_region,
					   struct nd_mapping *nd_mapping)
{
	struct nvdimm_drvdata *ndd = to_ndd(nd_mapping);
	struct nvdimm_bus *nvdimm_bus;
	resource_size_t max = 0;
	struct resource *res;

	/* if a dimm is disabled the available capacity is zero */
	if (!ndd)
		return 0;

	nvdimm_bus = walk_to_nvdimm_bus(ndd->dev);
	if (__reserve_free_pmem(&nd_region->dev, nd_mapping->nvdimm))
		return 0;
	for_each_dpa_resource(ndd, res) {
		if (strcmp(res->name, "pmem-reserve") != 0)
			continue;
		if (resource_size(res) > max)
			max = resource_size(res);
	}
	release_free_pmem(nvdimm_bus, nd_mapping);
	return max;
}

718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746
/**
 * nd_pmem_available_dpa - for the given dimm+region account unallocated dpa
 * @nd_mapping: container of dpa-resource-root + labels
 * @nd_region: constrain available space check to this reference region
 * @overlap: calculate available space assuming this level of overlap
 *
 * Validate that a PMEM label, if present, aligns with the start of an
 * interleave set and truncate the available size at the lowest BLK
 * overlap point.
 *
 * The expectation is that this routine is called multiple times as it
 * probes for the largest BLK encroachment for any single member DIMM of
 * the interleave set.  Once that value is determined the PMEM-limit for
 * the set can be established.
 */
resource_size_t nd_pmem_available_dpa(struct nd_region *nd_region,
		struct nd_mapping *nd_mapping, resource_size_t *overlap)
{
	resource_size_t map_start, map_end, busy = 0, available, blk_start;
	struct nvdimm_drvdata *ndd = to_ndd(nd_mapping);
	struct resource *res;
	const char *reason;

	if (!ndd)
		return 0;

	map_start = nd_mapping->start;
	map_end = map_start + nd_mapping->size - 1;
	blk_start = max(map_start, map_end + 1 - *overlap);
747
	for_each_dpa_resource(ndd, res) {
748 749
		if (res->start >= map_start && res->start < map_end) {
			if (strncmp(res->name, "blk", 3) == 0)
750 751 752
				blk_start = min(blk_start,
						max(map_start, res->start));
			else if (res->end > map_end) {
753 754
				reason = "misaligned to iset";
				goto err;
755
			} else
756 757 758 759 760 761 762 763 764
				busy += resource_size(res);
		} else if (res->end >= map_start && res->end <= map_end) {
			if (strncmp(res->name, "blk", 3) == 0) {
				/*
				 * If a BLK allocation overlaps the start of
				 * PMEM the entire interleave set may now only
				 * be used for BLK.
				 */
				blk_start = map_start;
765 766
			} else
				busy += resource_size(res);
767 768 769 770 771
		} else if (map_start > res->start && map_start < res->end) {
			/* total eclipse of the mapping */
			busy += nd_mapping->size;
			blk_start = map_start;
		}
772
	}
773 774 775 776 777 778 779 780 781 782 783 784

	*overlap = map_end + 1 - blk_start;
	available = blk_start - map_start;
	if (busy < available)
		return available - busy;
	return 0;

 err:
	nd_dbg_dpa(nd_region, ndd, res, "%s\n", reason);
	return 0;
}

785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808
void nvdimm_free_dpa(struct nvdimm_drvdata *ndd, struct resource *res)
{
	WARN_ON_ONCE(!is_nvdimm_bus_locked(ndd->dev));
	kfree(res->name);
	__release_region(&ndd->dpa, res->start, resource_size(res));
}

struct resource *nvdimm_allocate_dpa(struct nvdimm_drvdata *ndd,
		struct nd_label_id *label_id, resource_size_t start,
		resource_size_t n)
{
	char *name = kmemdup(label_id, sizeof(*label_id), GFP_KERNEL);
	struct resource *res;

	if (!name)
		return NULL;

	WARN_ON_ONCE(!is_nvdimm_bus_locked(ndd->dev));
	res = __request_region(&ndd->dpa, start, n, name, 0);
	if (!res)
		kfree(name);
	return res;
}

809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826
/**
 * nvdimm_allocated_dpa - sum up the dpa currently allocated to this label_id
 * @nvdimm: container of dpa-resource-root + labels
 * @label_id: dpa resource name of the form {pmem|blk}-<human readable uuid>
 */
resource_size_t nvdimm_allocated_dpa(struct nvdimm_drvdata *ndd,
		struct nd_label_id *label_id)
{
	resource_size_t allocated = 0;
	struct resource *res;

	for_each_dpa_resource(ndd, res)
		if (strcmp(res->name, label_id->id) == 0)
			allocated += resource_size(res);

	return allocated;
}

827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842
static int count_dimms(struct device *dev, void *c)
{
	int *count = c;

	if (is_nvdimm(dev))
		(*count)++;
	return 0;
}

int nvdimm_bus_check_dimm_count(struct nvdimm_bus *nvdimm_bus, int dimm_count)
{
	int count = 0;
	/* Flush any possible dimm registration failures */
	nd_synchronize();

	device_for_each_child(&nvdimm_bus->dev, &count, count_dimms);
843
	dev_dbg(&nvdimm_bus->dev, "count: %d\n", count);
844 845 846 847 848
	if (count != dimm_count)
		return -ENXIO;
	return 0;
}
EXPORT_SYMBOL_GPL(nvdimm_bus_check_dimm_count);
D
Dan Williams 已提交
849 850 851 852 853

void __exit nvdimm_devs_exit(void)
{
	ida_destroy(&dimm_ida);
}