bus.c 28.1 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/uaccess.h>
16
#include <linux/module.h>
17
#include <linux/blkdev.h>
18
#include <linux/fcntl.h>
19
#include <linux/async.h>
20
#include <linux/genhd.h>
21
#include <linux/ndctl.h>
22
#include <linux/sched.h>
23 24 25
#include <linux/slab.h>
#include <linux/fs.h>
#include <linux/io.h>
26
#include <linux/mm.h>
27
#include <linux/nd.h>
28
#include "nd-core.h"
29
#include "nd.h"
30
#include "pfn.h"
31

32
int nvdimm_major;
33 34
static int nvdimm_bus_major;
static struct class *nd_class;
35
static DEFINE_IDA(nd_ida);
36

37 38 39 40
static int to_nd_device_type(struct device *dev)
{
	if (is_nvdimm(dev))
		return ND_DEVICE_DIMM;
41 42 43 44
	else if (is_nd_pmem(dev))
		return ND_DEVICE_REGION_PMEM;
	else if (is_nd_blk(dev))
		return ND_DEVICE_REGION_BLK;
45 46
	else if (is_nd_dax(dev))
		return ND_DEVICE_DAX_PMEM;
47 48
	else if (is_nd_pmem(dev->parent) || is_nd_blk(dev->parent))
		return nd_region_to_nstype(to_nd_region(dev->parent));
49 50 51 52 53 54

	return 0;
}

static int nvdimm_bus_uevent(struct device *dev, struct kobj_uevent_env *env)
{
55 56 57 58 59 60
	/*
	 * Ensure that region devices always have their numa node set as
	 * early as possible.
	 */
	if (is_nd_pmem(dev) || is_nd_blk(dev))
		set_dev_node(dev, to_nd_region(dev)->numa_node);
61 62 63 64
	return add_uevent_var(env, "MODALIAS=" ND_DEVICE_MODALIAS_FMT,
			to_nd_device_type(dev));
}

65 66 67 68 69 70
static struct module *to_bus_provider(struct device *dev)
{
	/* pin bus providers while regions are enabled */
	if (is_nd_pmem(dev) || is_nd_blk(dev)) {
		struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(dev);

71
		return nvdimm_bus->nd_desc->module;
72 73 74 75
	}
	return NULL;
}

76 77 78 79 80 81 82 83 84 85 86 87 88 89 90
static void nvdimm_bus_probe_start(struct nvdimm_bus *nvdimm_bus)
{
	nvdimm_bus_lock(&nvdimm_bus->dev);
	nvdimm_bus->probe_active++;
	nvdimm_bus_unlock(&nvdimm_bus->dev);
}

static void nvdimm_bus_probe_end(struct nvdimm_bus *nvdimm_bus)
{
	nvdimm_bus_lock(&nvdimm_bus->dev);
	if (--nvdimm_bus->probe_active == 0)
		wake_up(&nvdimm_bus->probe_wait);
	nvdimm_bus_unlock(&nvdimm_bus->dev);
}

91 92 93
static int nvdimm_bus_probe(struct device *dev)
{
	struct nd_device_driver *nd_drv = to_nd_device_driver(dev->driver);
94
	struct module *provider = to_bus_provider(dev);
95 96 97
	struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(dev);
	int rc;

98 99 100
	if (!try_module_get(provider))
		return -ENXIO;

101
	nvdimm_bus_probe_start(nvdimm_bus);
102
	rc = nd_drv->probe(dev);
103 104
	if (rc == 0)
		nd_region_probe_success(nvdimm_bus, dev);
105 106
	else
		nd_region_disable(nvdimm_bus, dev);
107 108
	nvdimm_bus_probe_end(nvdimm_bus);

109 110
	dev_dbg(&nvdimm_bus->dev, "%s.probe(%s) = %d\n", dev->driver->name,
			dev_name(dev), rc);
111

112 113
	if (rc != 0)
		module_put(provider);
114 115 116 117 118 119
	return rc;
}

static int nvdimm_bus_remove(struct device *dev)
{
	struct nd_device_driver *nd_drv = to_nd_device_driver(dev->driver);
120
	struct module *provider = to_bus_provider(dev);
121
	struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(dev);
122
	int rc = 0;
123

124 125
	if (nd_drv->remove)
		rc = nd_drv->remove(dev);
126 127
	nd_region_disable(nvdimm_bus, dev);

128 129
	dev_dbg(&nvdimm_bus->dev, "%s.remove(%s) = %d\n", dev->driver->name,
			dev_name(dev), rc);
130
	module_put(provider);
131 132 133
	return rc;
}

134 135 136 137 138 139 140 141 142 143 144 145 146 147 148
static void nvdimm_bus_shutdown(struct device *dev)
{
	struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(dev);
	struct nd_device_driver *nd_drv = NULL;

	if (dev->driver)
		nd_drv = to_nd_device_driver(dev->driver);

	if (nd_drv && nd_drv->shutdown) {
		nd_drv->shutdown(dev);
		dev_dbg(&nvdimm_bus->dev, "%s.shutdown(%s)\n",
				dev->driver->name, dev_name(dev));
	}
}

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
void nd_device_notify(struct device *dev, enum nvdimm_event event)
{
	device_lock(dev);
	if (dev->driver) {
		struct nd_device_driver *nd_drv;

		nd_drv = to_nd_device_driver(dev->driver);
		if (nd_drv->notify)
			nd_drv->notify(dev, event);
	}
	device_unlock(dev);
}
EXPORT_SYMBOL(nd_device_notify);

void nvdimm_region_notify(struct nd_region *nd_region, enum nvdimm_event event)
{
	struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(&nd_region->dev);

	if (!nvdimm_bus)
		return;

	/* caller is responsible for holding a reference on the device */
	nd_device_notify(&nd_region->dev, event);
}
EXPORT_SYMBOL_GPL(nvdimm_region_notify);

175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
struct clear_badblocks_context {
	resource_size_t phys, cleared;
};

static int nvdimm_clear_badblocks_region(struct device *dev, void *data)
{
	struct clear_badblocks_context *ctx = data;
	struct nd_region *nd_region;
	resource_size_t ndr_end;
	sector_t sector;

	/* make sure device is a region */
	if (!is_nd_pmem(dev))
		return 0;

	nd_region = to_nd_region(dev);
	ndr_end = nd_region->ndr_start + nd_region->ndr_size - 1;

	/* make sure we are in the region */
	if (ctx->phys < nd_region->ndr_start
			|| (ctx->phys + ctx->cleared) > ndr_end)
		return 0;

	sector = (ctx->phys - nd_region->ndr_start) / 512;
	badblocks_clear(&nd_region->bb, sector, ctx->cleared / 512);

201 202 203
	if (nd_region->bb_state)
		sysfs_notify_dirent(nd_region->bb_state);

204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228
	return 0;
}

static void nvdimm_clear_badblocks_regions(struct nvdimm_bus *nvdimm_bus,
		phys_addr_t phys, u64 cleared)
{
	struct clear_badblocks_context ctx = {
		.phys = phys,
		.cleared = cleared,
	};

	device_for_each_child(&nvdimm_bus->dev, &ctx,
			nvdimm_clear_badblocks_region);
}

static void nvdimm_account_cleared_poison(struct nvdimm_bus *nvdimm_bus,
		phys_addr_t phys, u64 cleared)
{
	if (cleared > 0)
		nvdimm_forget_poison(nvdimm_bus, phys, cleared);

	if (cleared > 0 && cleared / 512)
		nvdimm_clear_badblocks_regions(nvdimm_bus, phys, cleared);
}

229 230 231 232 233 234 235 236 237 238 239 240 241 242
long nvdimm_clear_poison(struct device *dev, phys_addr_t phys,
		unsigned int len)
{
	struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(dev);
	struct nvdimm_bus_descriptor *nd_desc;
	struct nd_cmd_clear_error clear_err;
	struct nd_cmd_ars_cap ars_cap;
	u32 clear_err_unit, mask;
	int cmd_rc, rc;

	if (!nvdimm_bus)
		return -ENXIO;

	nd_desc = nvdimm_bus->nd_desc;
243 244 245 246
	/*
	 * if ndctl does not exist, it's PMEM_LEGACY and
	 * we want to just pretend everything is handled.
	 */
247
	if (!nd_desc->ndctl)
248
		return len;
249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274

	memset(&ars_cap, 0, sizeof(ars_cap));
	ars_cap.address = phys;
	ars_cap.length = len;
	rc = nd_desc->ndctl(nd_desc, NULL, ND_CMD_ARS_CAP, &ars_cap,
			sizeof(ars_cap), &cmd_rc);
	if (rc < 0)
		return rc;
	if (cmd_rc < 0)
		return cmd_rc;
	clear_err_unit = ars_cap.clear_err_unit;
	if (!clear_err_unit || !is_power_of_2(clear_err_unit))
		return -ENXIO;

	mask = clear_err_unit - 1;
	if ((phys | len) & mask)
		return -ENXIO;
	memset(&clear_err, 0, sizeof(clear_err));
	clear_err.address = phys;
	clear_err.length = len;
	rc = nd_desc->ndctl(nd_desc, NULL, ND_CMD_CLEAR_ERROR, &clear_err,
			sizeof(clear_err), &cmd_rc);
	if (rc < 0)
		return rc;
	if (cmd_rc < 0)
		return cmd_rc;
275

276
	nvdimm_account_cleared_poison(nvdimm_bus, phys, clear_err.cleared);
277

278 279 280 281
	return clear_err.cleared;
}
EXPORT_SYMBOL_GPL(nvdimm_clear_poison);

282 283
static int nvdimm_bus_match(struct device *dev, struct device_driver *drv);

284
static struct bus_type nvdimm_bus_type = {
285
	.name = "nd",
286 287 288 289
	.uevent = nvdimm_bus_uevent,
	.match = nvdimm_bus_match,
	.probe = nvdimm_bus_probe,
	.remove = nvdimm_bus_remove,
290
	.shutdown = nvdimm_bus_shutdown,
291 292
};

293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344
static void nvdimm_bus_release(struct device *dev)
{
	struct nvdimm_bus *nvdimm_bus;

	nvdimm_bus = container_of(dev, struct nvdimm_bus, dev);
	ida_simple_remove(&nd_ida, nvdimm_bus->id);
	kfree(nvdimm_bus);
}

static bool is_nvdimm_bus(struct device *dev)
{
	return dev->release == nvdimm_bus_release;
}

struct nvdimm_bus *walk_to_nvdimm_bus(struct device *nd_dev)
{
	struct device *dev;

	for (dev = nd_dev; dev; dev = dev->parent)
		if (is_nvdimm_bus(dev))
			break;
	dev_WARN_ONCE(nd_dev, !dev, "invalid dev, not on nd bus\n");
	if (dev)
		return to_nvdimm_bus(dev);
	return NULL;
}

struct nvdimm_bus *to_nvdimm_bus(struct device *dev)
{
	struct nvdimm_bus *nvdimm_bus;

	nvdimm_bus = container_of(dev, struct nvdimm_bus, dev);
	WARN_ON(!is_nvdimm_bus(dev));
	return nvdimm_bus;
}
EXPORT_SYMBOL_GPL(to_nvdimm_bus);

struct nvdimm_bus *nvdimm_bus_register(struct device *parent,
		struct nvdimm_bus_descriptor *nd_desc)
{
	struct nvdimm_bus *nvdimm_bus;
	int rc;

	nvdimm_bus = kzalloc(sizeof(*nvdimm_bus), GFP_KERNEL);
	if (!nvdimm_bus)
		return NULL;
	INIT_LIST_HEAD(&nvdimm_bus->list);
	INIT_LIST_HEAD(&nvdimm_bus->mapping_list);
	INIT_LIST_HEAD(&nvdimm_bus->poison_list);
	init_waitqueue_head(&nvdimm_bus->probe_wait);
	nvdimm_bus->id = ida_simple_get(&nd_ida, 0, 0, GFP_KERNEL);
	mutex_init(&nvdimm_bus->reconfig_mutex);
345
	spin_lock_init(&nvdimm_bus->poison_lock);
346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413
	if (nvdimm_bus->id < 0) {
		kfree(nvdimm_bus);
		return NULL;
	}
	nvdimm_bus->nd_desc = nd_desc;
	nvdimm_bus->dev.parent = parent;
	nvdimm_bus->dev.release = nvdimm_bus_release;
	nvdimm_bus->dev.groups = nd_desc->attr_groups;
	nvdimm_bus->dev.bus = &nvdimm_bus_type;
	dev_set_name(&nvdimm_bus->dev, "ndbus%d", nvdimm_bus->id);
	rc = device_register(&nvdimm_bus->dev);
	if (rc) {
		dev_dbg(&nvdimm_bus->dev, "registration failed: %d\n", rc);
		goto err;
	}

	return nvdimm_bus;
 err:
	put_device(&nvdimm_bus->dev);
	return NULL;
}
EXPORT_SYMBOL_GPL(nvdimm_bus_register);

void nvdimm_bus_unregister(struct nvdimm_bus *nvdimm_bus)
{
	if (!nvdimm_bus)
		return;
	device_unregister(&nvdimm_bus->dev);
}
EXPORT_SYMBOL_GPL(nvdimm_bus_unregister);

static int child_unregister(struct device *dev, void *data)
{
	/*
	 * the singular ndctl class device per bus needs to be
	 * "device_destroy"ed, so skip it here
	 *
	 * i.e. remove classless children
	 */
	if (dev->class)
		/* pass */;
	else
		nd_device_unregister(dev, ND_SYNC);
	return 0;
}

static void free_poison_list(struct list_head *poison_list)
{
	struct nd_poison *pl, *next;

	list_for_each_entry_safe(pl, next, poison_list, list) {
		list_del(&pl->list);
		kfree(pl);
	}
	list_del_init(poison_list);
}

static int nd_bus_remove(struct device *dev)
{
	struct nvdimm_bus *nvdimm_bus = to_nvdimm_bus(dev);

	mutex_lock(&nvdimm_bus_list_mutex);
	list_del_init(&nvdimm_bus->list);
	mutex_unlock(&nvdimm_bus_list_mutex);

	nd_synchronize();
	device_for_each_child(&nvdimm_bus->dev, NULL, child_unregister);

414
	spin_lock(&nvdimm_bus->poison_lock);
415
	free_poison_list(&nvdimm_bus->poison_list);
416
	spin_unlock(&nvdimm_bus->poison_lock);
417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451

	nvdimm_bus_destroy_ndctl(nvdimm_bus);

	return 0;
}

static int nd_bus_probe(struct device *dev)
{
	struct nvdimm_bus *nvdimm_bus = to_nvdimm_bus(dev);
	int rc;

	rc = nvdimm_bus_create_ndctl(nvdimm_bus);
	if (rc)
		return rc;

	mutex_lock(&nvdimm_bus_list_mutex);
	list_add_tail(&nvdimm_bus->list, &nvdimm_bus_list);
	mutex_unlock(&nvdimm_bus_list_mutex);

	/* enable bus provider attributes to look up their local context */
	dev_set_drvdata(dev, nvdimm_bus->nd_desc);

	return 0;
}

static struct nd_device_driver nd_bus_driver = {
	.probe = nd_bus_probe,
	.remove = nd_bus_remove,
	.drv = {
		.name = "nd_bus",
		.suppress_bind_attrs = true,
		.bus = &nvdimm_bus_type,
		.owner = THIS_MODULE,
		.mod_name = KBUILD_MODNAME,
	},
452 453
};

454 455 456 457 458 459 460 461 462 463
static int nvdimm_bus_match(struct device *dev, struct device_driver *drv)
{
	struct nd_device_driver *nd_drv = to_nd_device_driver(drv);

	if (is_nvdimm_bus(dev) && nd_drv == &nd_bus_driver)
		return true;

	return !!test_bit(to_nd_device_type(dev), &nd_drv->type);
}

464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486
static ASYNC_DOMAIN_EXCLUSIVE(nd_async_domain);

void nd_synchronize(void)
{
	async_synchronize_full_domain(&nd_async_domain);
}
EXPORT_SYMBOL_GPL(nd_synchronize);

static void nd_async_device_register(void *d, async_cookie_t cookie)
{
	struct device *dev = d;

	if (device_add(dev) != 0) {
		dev_err(dev, "%s: failed\n", __func__);
		put_device(dev);
	}
	put_device(dev);
}

static void nd_async_device_unregister(void *d, async_cookie_t cookie)
{
	struct device *dev = d;

D
Dan Williams 已提交
487 488 489 490
	/* flush bus operations before delete */
	nvdimm_bus_lock(dev);
	nvdimm_bus_unlock(dev);

491 492 493 494
	device_unregister(dev);
	put_device(dev);
}

495
void __nd_device_register(struct device *dev)
496
{
497 498
	if (!dev)
		return;
499 500 501 502 503
	dev->bus = &nvdimm_bus_type;
	get_device(dev);
	async_schedule_domain(nd_async_device_register, dev,
			&nd_async_domain);
}
504 505 506 507 508 509

void nd_device_register(struct device *dev)
{
	device_initialize(dev);
	__nd_device_register(dev);
}
510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544
EXPORT_SYMBOL(nd_device_register);

void nd_device_unregister(struct device *dev, enum nd_async_mode mode)
{
	switch (mode) {
	case ND_ASYNC:
		get_device(dev);
		async_schedule_domain(nd_async_device_unregister, dev,
				&nd_async_domain);
		break;
	case ND_SYNC:
		nd_synchronize();
		device_unregister(dev);
		break;
	}
}
EXPORT_SYMBOL(nd_device_unregister);

/**
 * __nd_driver_register() - register a region or a namespace driver
 * @nd_drv: driver to register
 * @owner: automatically set by nd_driver_register() macro
 * @mod_name: automatically set by nd_driver_register() macro
 */
int __nd_driver_register(struct nd_device_driver *nd_drv, struct module *owner,
		const char *mod_name)
{
	struct device_driver *drv = &nd_drv->drv;

	if (!nd_drv->type) {
		pr_debug("driver type bitmask not set (%pf)\n",
				__builtin_return_address(0));
		return -EINVAL;
	}

545 546
	if (!nd_drv->probe) {
		pr_debug("%s ->probe() must be specified\n", mod_name);
547 548 549 550 551 552 553 554 555 556 557
		return -EINVAL;
	}

	drv->bus = &nvdimm_bus_type;
	drv->owner = owner;
	drv->mod_name = mod_name;

	return driver_register(drv);
}
EXPORT_SYMBOL(__nd_driver_register);

558 559
int nvdimm_revalidate_disk(struct gendisk *disk)
{
D
Dan Williams 已提交
560
	struct device *dev = disk_to_dev(disk)->parent;
561 562 563 564 565 566 567 568 569 570 571 572 573 574 575
	struct nd_region *nd_region = to_nd_region(dev->parent);
	const char *pol = nd_region->ro ? "only" : "write";

	if (nd_region->ro == get_disk_ro(disk))
		return 0;

	dev_info(dev, "%s read-%s, marking %s read-%s\n",
			dev_name(&nd_region->dev), pol, disk->disk_name, pol);
	set_disk_ro(disk, nd_region->ro);

	return 0;

}
EXPORT_SYMBOL(nvdimm_revalidate_disk);

576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601
static ssize_t modalias_show(struct device *dev, struct device_attribute *attr,
		char *buf)
{
	return sprintf(buf, ND_DEVICE_MODALIAS_FMT "\n",
			to_nd_device_type(dev));
}
static DEVICE_ATTR_RO(modalias);

static ssize_t devtype_show(struct device *dev, struct device_attribute *attr,
		char *buf)
{
	return sprintf(buf, "%s\n", dev->type->name);
}
static DEVICE_ATTR_RO(devtype);

static struct attribute *nd_device_attributes[] = {
	&dev_attr_modalias.attr,
	&dev_attr_devtype.attr,
	NULL,
};

/**
 * nd_device_attribute_group - generic attributes for all devices on an nd bus
 */
struct attribute_group nd_device_attribute_group = {
	.attrs = nd_device_attributes,
602
};
603
EXPORT_SYMBOL_GPL(nd_device_attribute_group);
604

605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634
static ssize_t numa_node_show(struct device *dev,
		struct device_attribute *attr, char *buf)
{
	return sprintf(buf, "%d\n", dev_to_node(dev));
}
static DEVICE_ATTR_RO(numa_node);

static struct attribute *nd_numa_attributes[] = {
	&dev_attr_numa_node.attr,
	NULL,
};

static umode_t nd_numa_attr_visible(struct kobject *kobj, struct attribute *a,
		int n)
{
	if (!IS_ENABLED(CONFIG_NUMA))
		return 0;

	return a->mode;
}

/**
 * nd_numa_attribute_group - NUMA attributes for all devices on an nd bus
 */
struct attribute_group nd_numa_attribute_group = {
	.attrs = nd_numa_attributes,
	.is_visible = nd_numa_attr_visible,
};
EXPORT_SYMBOL_GPL(nd_numa_attribute_group);

635 636 637 638 639 640 641 642
int nvdimm_bus_create_ndctl(struct nvdimm_bus *nvdimm_bus)
{
	dev_t devt = MKDEV(nvdimm_bus_major, nvdimm_bus->id);
	struct device *dev;

	dev = device_create(nd_class, &nvdimm_bus->dev, devt, nvdimm_bus,
			"ndctl%d", nvdimm_bus->id);

D
Dan Williams 已提交
643
	if (IS_ERR(dev))
644 645
		dev_dbg(&nvdimm_bus->dev, "failed to register ndctl%d: %ld\n",
				nvdimm_bus->id, PTR_ERR(dev));
D
Dan Williams 已提交
646
	return PTR_ERR_OR_ZERO(dev);
647 648 649 650 651 652 653
}

void nvdimm_bus_destroy_ndctl(struct nvdimm_bus *nvdimm_bus)
{
	device_destroy(nd_class, MKDEV(nvdimm_bus_major, nvdimm_bus->id));
}

654 655 656 657
static const struct nd_cmd_desc __nd_cmd_dimm_descs[] = {
	[ND_CMD_IMPLEMENTED] = { },
	[ND_CMD_SMART] = {
		.out_num = 2,
658
		.out_sizes = { 4, 128, },
659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689
	},
	[ND_CMD_SMART_THRESHOLD] = {
		.out_num = 2,
		.out_sizes = { 4, 8, },
	},
	[ND_CMD_DIMM_FLAGS] = {
		.out_num = 2,
		.out_sizes = { 4, 4 },
	},
	[ND_CMD_GET_CONFIG_SIZE] = {
		.out_num = 3,
		.out_sizes = { 4, 4, 4, },
	},
	[ND_CMD_GET_CONFIG_DATA] = {
		.in_num = 2,
		.in_sizes = { 4, 4, },
		.out_num = 2,
		.out_sizes = { 4, UINT_MAX, },
	},
	[ND_CMD_SET_CONFIG_DATA] = {
		.in_num = 3,
		.in_sizes = { 4, 4, UINT_MAX, },
		.out_num = 1,
		.out_sizes = { 4, },
	},
	[ND_CMD_VENDOR] = {
		.in_num = 3,
		.in_sizes = { 4, 4, UINT_MAX, },
		.out_num = 3,
		.out_sizes = { 4, 4, UINT_MAX, },
	},
690 691 692 693 694 695
	[ND_CMD_CALL] = {
		.in_num = 2,
		.in_sizes = { sizeof(struct nd_cmd_pkg), UINT_MAX, },
		.out_num = 1,
		.out_sizes = { UINT_MAX, },
	},
696 697 698 699 700 701 702 703 704 705 706 707 708 709 710
};

const struct nd_cmd_desc *nd_cmd_dimm_desc(int cmd)
{
	if (cmd < ARRAY_SIZE(__nd_cmd_dimm_descs))
		return &__nd_cmd_dimm_descs[cmd];
	return NULL;
}
EXPORT_SYMBOL_GPL(nd_cmd_dimm_desc);

static const struct nd_cmd_desc __nd_cmd_bus_descs[] = {
	[ND_CMD_IMPLEMENTED] = { },
	[ND_CMD_ARS_CAP] = {
		.in_num = 2,
		.in_sizes = { 8, 8, },
711 712
		.out_num = 4,
		.out_sizes = { 4, 4, 4, 4, },
713 714
	},
	[ND_CMD_ARS_START] = {
715 716 717 718
		.in_num = 5,
		.in_sizes = { 8, 8, 2, 1, 5, },
		.out_num = 2,
		.out_sizes = { 4, 4, },
719 720
	},
	[ND_CMD_ARS_STATUS] = {
721 722
		.out_num = 3,
		.out_sizes = { 4, 4, UINT_MAX, },
723
	},
724 725 726 727 728 729
	[ND_CMD_CLEAR_ERROR] = {
		.in_num = 2,
		.in_sizes = { 8, 8, },
		.out_num = 3,
		.out_sizes = { 4, 4, 8, },
	},
730 731 732 733 734 735
	[ND_CMD_CALL] = {
		.in_num = 2,
		.in_sizes = { sizeof(struct nd_cmd_pkg), UINT_MAX, },
		.out_num = 1,
		.out_sizes = { UINT_MAX, },
	},
736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762
};

const struct nd_cmd_desc *nd_cmd_bus_desc(int cmd)
{
	if (cmd < ARRAY_SIZE(__nd_cmd_bus_descs))
		return &__nd_cmd_bus_descs[cmd];
	return NULL;
}
EXPORT_SYMBOL_GPL(nd_cmd_bus_desc);

u32 nd_cmd_in_size(struct nvdimm *nvdimm, int cmd,
		const struct nd_cmd_desc *desc, int idx, void *buf)
{
	if (idx >= desc->in_num)
		return UINT_MAX;

	if (desc->in_sizes[idx] < UINT_MAX)
		return desc->in_sizes[idx];

	if (nvdimm && cmd == ND_CMD_SET_CONFIG_DATA && idx == 2) {
		struct nd_cmd_set_config_hdr *hdr = buf;

		return hdr->in_length;
	} else if (nvdimm && cmd == ND_CMD_VENDOR && idx == 2) {
		struct nd_cmd_vendor_hdr *hdr = buf;

		return hdr->in_length;
763 764 765 766
	} else if (cmd == ND_CMD_CALL) {
		struct nd_cmd_pkg *pkg = buf;

		return pkg->nd_size_in;
767 768 769 770 771 772 773 774
	}

	return UINT_MAX;
}
EXPORT_SYMBOL_GPL(nd_cmd_in_size);

u32 nd_cmd_out_size(struct nvdimm *nvdimm, int cmd,
		const struct nd_cmd_desc *desc, int idx, const u32 *in_field,
775
		const u32 *out_field, unsigned long remainder)
776 777 778 779 780 781 782 783 784 785 786
{
	if (idx >= desc->out_num)
		return UINT_MAX;

	if (desc->out_sizes[idx] < UINT_MAX)
		return desc->out_sizes[idx];

	if (nvdimm && cmd == ND_CMD_GET_CONFIG_DATA && idx == 1)
		return in_field[1];
	else if (nvdimm && cmd == ND_CMD_VENDOR && idx == 2)
		return out_field[1];
787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804
	else if (!nvdimm && cmd == ND_CMD_ARS_STATUS && idx == 2) {
		/*
		 * Per table 9-276 ARS Data in ACPI 6.1, out_field[1] is
		 * "Size of Output Buffer in bytes, including this
		 * field."
		 */
		if (out_field[1] < 4)
			return 0;
		/*
		 * ACPI 6.1 is ambiguous if 'status' is included in the
		 * output size. If we encounter an output size that
		 * overshoots the remainder by 4 bytes, assume it was
		 * including 'status'.
		 */
		if (out_field[1] - 8 == remainder)
			return remainder;
		return out_field[1] - 4;
	} else if (cmd == ND_CMD_CALL) {
805 806 807 808 809
		struct nd_cmd_pkg *pkg = (struct nd_cmd_pkg *) in_field;

		return pkg->nd_size_out;
	}

810 811 812 813 814

	return UINT_MAX;
}
EXPORT_SYMBOL_GPL(nd_cmd_out_size);

815
void wait_nvdimm_bus_probe_idle(struct device *dev)
816
{
817 818
	struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(dev);

819 820 821 822 823 824 825 826 827 828
	do {
		if (nvdimm_bus->probe_active == 0)
			break;
		nvdimm_bus_unlock(&nvdimm_bus->dev);
		wait_event(nvdimm_bus->probe_wait,
				nvdimm_bus->probe_active == 0);
		nvdimm_bus_lock(&nvdimm_bus->dev);
	} while (true);
}

829
static int nd_pmem_forget_poison_check(struct device *dev, void *data)
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
	struct nd_cmd_clear_error *clear_err =
		(struct nd_cmd_clear_error *)data;
	struct nd_btt *nd_btt = is_nd_btt(dev) ? to_nd_btt(dev) : NULL;
	struct nd_pfn *nd_pfn = is_nd_pfn(dev) ? to_nd_pfn(dev) : NULL;
	struct nd_dax *nd_dax = is_nd_dax(dev) ? to_nd_dax(dev) : NULL;
	struct nd_namespace_common *ndns = NULL;
	struct nd_namespace_io *nsio;
	resource_size_t offset = 0, end_trunc = 0, start, end, pstart, pend;

	if (nd_dax || !dev->driver)
		return 0;

	start = clear_err->address;
	end = clear_err->address + clear_err->cleared - 1;

	if (nd_btt || nd_pfn || nd_dax) {
		if (nd_btt)
			ndns = nd_btt->ndns;
		else if (nd_pfn)
			ndns = nd_pfn->ndns;
		else if (nd_dax)
			ndns = nd_dax->nd_pfn.ndns;

		if (!ndns)
			return 0;
	} else
		ndns = to_ndns(dev);

	nsio = to_nd_namespace_io(&ndns->dev);
	pstart = nsio->res.start + offset;
	pend = nsio->res.end - end_trunc;

	if ((pstart >= start) && (pend <= end))
864
		return -EBUSY;
865

866
	return 0;
867 868 869 870 871 872

}

static int nd_ns_forget_poison_check(struct device *dev, void *data)
{
	return device_for_each_child(dev, data, nd_pmem_forget_poison_check);
873 874
}

875
/* set_config requires an idle interleave set */
876
static int nd_cmd_clear_to_send(struct nvdimm_bus *nvdimm_bus,
877
		struct nvdimm *nvdimm, unsigned int cmd, void *data)
878
{
879 880 881 882 883 884 885 886 887
	struct nvdimm_bus_descriptor *nd_desc = nvdimm_bus->nd_desc;

	/* ask the bus provider if it would like to block this request */
	if (nd_desc->clear_to_send) {
		int rc = nd_desc->clear_to_send(nd_desc, nvdimm, cmd);

		if (rc)
			return rc;
	}
888

889 890
	/* require clear error to go through the pmem driver */
	if (!nvdimm && cmd == ND_CMD_CLEAR_ERROR)
891 892
		return device_for_each_child(&nvdimm_bus->dev, data,
				nd_ns_forget_poison_check);
893

894 895 896
	if (!nvdimm || cmd != ND_CMD_SET_CONFIG_DATA)
		return 0;

897
	/* prevent label manipulation while the kernel owns label updates */
898
	wait_nvdimm_bus_probe_idle(&nvdimm_bus->dev);
899 900 901 902 903
	if (atomic_read(&nvdimm->busy))
		return -EBUSY;
	return 0;
}

904 905 906 907 908 909 910 911 912
static int __nd_ioctl(struct nvdimm_bus *nvdimm_bus, struct nvdimm *nvdimm,
		int read_only, unsigned int ioctl_cmd, unsigned long arg)
{
	struct nvdimm_bus_descriptor *nd_desc = nvdimm_bus->nd_desc;
	size_t buf_len = 0, in_len = 0, out_len = 0;
	static char out_env[ND_CMD_MAX_ENVELOPE];
	static char in_env[ND_CMD_MAX_ENVELOPE];
	const struct nd_cmd_desc *desc = NULL;
	unsigned int cmd = _IOC_NR(ioctl_cmd);
913
	unsigned int func = cmd;
914 915
	void __user *p = (void __user *) arg;
	struct device *dev = &nvdimm_bus->dev;
916
	struct nd_cmd_pkg pkg;
917
	const char *cmd_name, *dimm_name;
918
	unsigned long cmd_mask;
919
	void *buf;
920
	int rc, i, cmd_rc;
921 922 923 924

	if (nvdimm) {
		desc = nd_cmd_dimm_desc(cmd);
		cmd_name = nvdimm_cmd_name(cmd);
925
		cmd_mask = nvdimm->cmd_mask;
926 927 928 929
		dimm_name = dev_name(&nvdimm->dev);
	} else {
		desc = nd_cmd_bus_desc(cmd);
		cmd_name = nvdimm_bus_cmd_name(cmd);
930
		cmd_mask = nd_desc->cmd_mask;
931 932 933
		dimm_name = "bus";
	}

934 935 936 937 938
	if (cmd == ND_CMD_CALL) {
		if (copy_from_user(&pkg, p, sizeof(pkg)))
			return -EFAULT;
	}

939
	if (!desc || (desc->out_num + desc->in_num == 0) ||
940
			!test_bit(cmd, &cmd_mask))
941 942 943 944
		return -ENOTTY;

	/* fail write commands (when read-only) */
	if (read_only)
945 946 947 948
		switch (cmd) {
		case ND_CMD_VENDOR:
		case ND_CMD_SET_CONFIG_DATA:
		case ND_CMD_ARS_START:
949
		case ND_CMD_CLEAR_ERROR:
950
		case ND_CMD_CALL:
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
			dev_dbg(&nvdimm_bus->dev, "'%s' command while read-only.\n",
					nvdimm ? nvdimm_cmd_name(cmd)
					: nvdimm_bus_cmd_name(cmd));
			return -EPERM;
		default:
			break;
		}

	/* process an input envelope */
	for (i = 0; i < desc->in_num; i++) {
		u32 in_size, copy;

		in_size = nd_cmd_in_size(nvdimm, cmd, desc, i, in_env);
		if (in_size == UINT_MAX) {
			dev_err(dev, "%s:%s unknown input size cmd: %s field: %d\n",
					__func__, dimm_name, cmd_name, i);
			return -ENXIO;
		}
		if (in_len < sizeof(in_env))
			copy = min_t(u32, sizeof(in_env) - in_len, in_size);
		else
			copy = 0;
		if (copy && copy_from_user(&in_env[in_len], p + in_len, copy))
			return -EFAULT;
		in_len += in_size;
	}

978
	if (cmd == ND_CMD_CALL) {
979
		func = pkg.nd_command;
980 981 982 983 984 985 986 987 988
		dev_dbg(dev, "%s:%s, idx: %llu, in: %zu, out: %zu, len %zu\n",
				__func__, dimm_name, pkg.nd_command,
				in_len, out_len, buf_len);

		for (i = 0; i < ARRAY_SIZE(pkg.nd_reserved2); i++)
			if (pkg.nd_reserved2[i])
				return -EINVAL;
	}

989 990 991
	/* process an output envelope */
	for (i = 0; i < desc->out_num; i++) {
		u32 out_size = nd_cmd_out_size(nvdimm, cmd, desc, i,
992
				(u32 *) in_env, (u32 *) out_env, 0);
993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026
		u32 copy;

		if (out_size == UINT_MAX) {
			dev_dbg(dev, "%s:%s unknown output size cmd: %s field: %d\n",
					__func__, dimm_name, cmd_name, i);
			return -EFAULT;
		}
		if (out_len < sizeof(out_env))
			copy = min_t(u32, sizeof(out_env) - out_len, out_size);
		else
			copy = 0;
		if (copy && copy_from_user(&out_env[out_len],
					p + in_len + out_len, copy))
			return -EFAULT;
		out_len += out_size;
	}

	buf_len = out_len + in_len;
	if (buf_len > ND_IOCTL_MAX_BUFLEN) {
		dev_dbg(dev, "%s:%s cmd: %s buf_len: %zu > %d\n", __func__,
				dimm_name, cmd_name, buf_len,
				ND_IOCTL_MAX_BUFLEN);
		return -EINVAL;
	}

	buf = vmalloc(buf_len);
	if (!buf)
		return -ENOMEM;

	if (copy_from_user(buf, p, buf_len)) {
		rc = -EFAULT;
		goto out;
	}

1027
	nvdimm_bus_lock(&nvdimm_bus->dev);
1028
	rc = nd_cmd_clear_to_send(nvdimm_bus, nvdimm, func, buf);
1029 1030 1031
	if (rc)
		goto out_unlock;

1032
	rc = nd_desc->ndctl(nd_desc, nvdimm, cmd, buf, buf_len, &cmd_rc);
1033
	if (rc < 0)
1034
		goto out_unlock;
1035 1036 1037 1038

	if (!nvdimm && cmd == ND_CMD_CLEAR_ERROR && cmd_rc >= 0) {
		struct nd_cmd_clear_error *clear_err = buf;

1039 1040
		nvdimm_account_cleared_poison(nvdimm_bus, clear_err->address,
				clear_err->cleared);
1041
	}
1042 1043
	nvdimm_bus_unlock(&nvdimm_bus->dev);

1044 1045
	if (copy_to_user(p, buf, buf_len))
		rc = -EFAULT;
1046 1047 1048 1049

	vfree(buf);
	return rc;

1050 1051
 out_unlock:
	nvdimm_bus_unlock(&nvdimm_bus->dev);
1052 1053 1054 1055 1056
 out:
	vfree(buf);
	return rc;
}

1057 1058
static long nd_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
1059
	long id = (long) file->private_data;
1060
	int rc = -ENXIO, ro;
1061 1062
	struct nvdimm_bus *nvdimm_bus;

1063
	ro = ((file->f_flags & O_ACCMODE) == O_RDONLY);
1064 1065 1066
	mutex_lock(&nvdimm_bus_list_mutex);
	list_for_each_entry(nvdimm_bus, &nvdimm_bus_list, list) {
		if (nvdimm_bus->id == id) {
1067
			rc = __nd_ioctl(nvdimm_bus, NULL, ro, cmd, arg);
1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090
			break;
		}
	}
	mutex_unlock(&nvdimm_bus_list_mutex);

	return rc;
}

static int match_dimm(struct device *dev, void *data)
{
	long id = (long) data;

	if (is_nvdimm(dev)) {
		struct nvdimm *nvdimm = to_nvdimm(dev);

		return nvdimm->id == id;
	}

	return 0;
}

static long nvdimm_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
1091
	int rc = -ENXIO, ro;
1092 1093
	struct nvdimm_bus *nvdimm_bus;

1094
	ro = ((file->f_flags & O_ACCMODE) == O_RDONLY);
1095 1096 1097 1098 1099 1100 1101 1102 1103 1104
	mutex_lock(&nvdimm_bus_list_mutex);
	list_for_each_entry(nvdimm_bus, &nvdimm_bus_list, list) {
		struct device *dev = device_find_child(&nvdimm_bus->dev,
				file->private_data, match_dimm);
		struct nvdimm *nvdimm;

		if (!dev)
			continue;

		nvdimm = to_nvdimm(dev);
1105
		rc = __nd_ioctl(nvdimm_bus, nvdimm, ro, cmd, arg);
1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119
		put_device(dev);
		break;
	}
	mutex_unlock(&nvdimm_bus_list_mutex);

	return rc;
}

static int nd_open(struct inode *inode, struct file *file)
{
	long minor = iminor(inode);

	file->private_data = (void *) minor;
	return 0;
1120 1121 1122 1123
}

static const struct file_operations nvdimm_bus_fops = {
	.owner = THIS_MODULE,
1124
	.open = nd_open,
1125 1126 1127 1128 1129
	.unlocked_ioctl = nd_ioctl,
	.compat_ioctl = nd_ioctl,
	.llseek = noop_llseek,
};

1130 1131 1132 1133 1134 1135 1136 1137
static const struct file_operations nvdimm_fops = {
	.owner = THIS_MODULE,
	.open = nd_open,
	.unlocked_ioctl = nvdimm_ioctl,
	.compat_ioctl = nvdimm_ioctl,
	.llseek = noop_llseek,
};

1138 1139 1140 1141
int __init nvdimm_bus_init(void)
{
	int rc;

1142 1143 1144
	BUILD_BUG_ON(sizeof(struct nd_smart_payload) != 128);
	BUILD_BUG_ON(sizeof(struct nd_smart_threshold_payload) != 8);

1145 1146 1147 1148
	rc = bus_register(&nvdimm_bus_type);
	if (rc)
		return rc;

1149 1150
	rc = register_chrdev(0, "ndctl", &nvdimm_bus_fops);
	if (rc < 0)
1151
		goto err_bus_chrdev;
1152 1153
	nvdimm_bus_major = rc;

1154 1155 1156 1157 1158
	rc = register_chrdev(0, "dimmctl", &nvdimm_fops);
	if (rc < 0)
		goto err_dimm_chrdev;
	nvdimm_major = rc;

1159
	nd_class = class_create(THIS_MODULE, "nd");
1160 1161
	if (IS_ERR(nd_class)) {
		rc = PTR_ERR(nd_class);
1162
		goto err_class;
1163
	}
1164

1165 1166 1167 1168
	rc = driver_register(&nd_bus_driver.drv);
	if (rc)
		goto err_nd_bus;

1169 1170
	return 0;

1171 1172
 err_nd_bus:
	class_destroy(nd_class);
1173
 err_class:
1174 1175
	unregister_chrdev(nvdimm_major, "dimmctl");
 err_dimm_chrdev:
1176
	unregister_chrdev(nvdimm_bus_major, "ndctl");
1177
 err_bus_chrdev:
1178
	bus_unregister(&nvdimm_bus_type);
1179 1180 1181 1182

	return rc;
}

1183
void nvdimm_bus_exit(void)
1184
{
1185
	driver_unregister(&nd_bus_driver.drv);
1186 1187
	class_destroy(nd_class);
	unregister_chrdev(nvdimm_bus_major, "ndctl");
1188
	unregister_chrdev(nvdimm_major, "dimmctl");
1189
	bus_unregister(&nvdimm_bus_type);
1190
	ida_destroy(&nd_ida);
1191
}