[mirror_ubuntu-bionic-kernel.git] / block / partition-generic.c

// SPDX-License-Identifier: GPL-2.0
/*
 *  Code extracted from drivers/block/genhd.c
 *  Copyright (C) 1991-1998  Linus Torvalds
 *  Re-organised Feb 1998 Russell King
 *
 *  We now have independent partition support from the
 *  block drivers, which allows all the partition code to
 *  be grouped in one location, and it to be mostly self
 *  contained.
 */

#include <linux/init.h>
#include <linux/module.h>
#include <linux/fs.h>
#include <linux/slab.h>
#include <linux/kmod.h>
#include <linux/ctype.h>
#include <linux/genhd.h>
#include <linux/blktrace_api.h>

#include "partitions/check.h"

#ifdef CONFIG_BLK_DEV_MD
extern void md_autodetect_dev(dev_t dev);
#endif
 
/*
 * disk_name() is used by partition check code and the genhd driver.
 * It formats the devicename of the indicated disk into
 * the supplied buffer (of size at least 32), and returns
 * a pointer to that same buffer (for convenience).
 */

char *disk_name(struct gendisk *hd, int partno, char *buf)
{
	if (!partno)
		snprintf(buf, BDEVNAME_SIZE, "%s", hd->disk_name);
	else if (isdigit(hd->disk_name[strlen(hd->disk_name)-1]))
		snprintf(buf, BDEVNAME_SIZE, "%sp%d", hd->disk_name, partno);
	else
		snprintf(buf, BDEVNAME_SIZE, "%s%d", hd->disk_name, partno);

	return buf;
}

const char *bdevname(struct block_device *bdev, char *buf)
{
	return disk_name(bdev->bd_disk, bdev->bd_part->partno, buf);
}

EXPORT_SYMBOL(bdevname);

const char *bio_devname(struct bio *bio, char *buf)
{
	return disk_name(bio->bi_disk, bio->bi_partno, buf);
}
EXPORT_SYMBOL(bio_devname);

/*
 * There's very little reason to use this, you should really
 * have a struct block_device just about everywhere and use
 * bdevname() instead.
 */
const char *__bdevname(dev_t dev, char *buffer)
{
	scnprintf(buffer, BDEVNAME_SIZE, "unknown-block(%u,%u)",
				MAJOR(dev), MINOR(dev));
	return buffer;
}

EXPORT_SYMBOL(__bdevname);

static ssize_t part_partition_show(struct device *dev,
				   struct device_attribute *attr, char *buf)
{
	struct hd_struct *p = dev_to_part(dev);

	return sprintf(buf, "%d\n", p->partno);
}

static ssize_t part_start_show(struct device *dev,
			       struct device_attribute *attr, char *buf)
{
	struct hd_struct *p = dev_to_part(dev);

	return sprintf(buf, "%llu\n",(unsigned long long)p->start_sect);
}

ssize_t part_size_show(struct device *dev,
		       struct device_attribute *attr, char *buf)
{
	struct hd_struct *p = dev_to_part(dev);
	return sprintf(buf, "%llu\n",(unsigned long long)part_nr_sects_read(p));
}

static ssize_t part_ro_show(struct device *dev,
			    struct device_attribute *attr, char *buf)
{
	struct hd_struct *p = dev_to_part(dev);
	return sprintf(buf, "%d\n", p->policy ? 1 : 0);
}

static ssize_t part_alignment_offset_show(struct device *dev,
					  struct device_attribute *attr, char *buf)
{
	struct hd_struct *p = dev_to_part(dev);
	return sprintf(buf, "%llu\n", (unsigned long long)p->alignment_offset);
}

static ssize_t part_discard_alignment_show(struct device *dev,
					   struct device_attribute *attr, char *buf)
{
	struct hd_struct *p = dev_to_part(dev);
	return sprintf(buf, "%u\n", p->discard_alignment);
}

ssize_t part_stat_show(struct device *dev,
		       struct device_attribute *attr, char *buf)
{
	struct hd_struct *p = dev_to_part(dev);
	struct request_queue *q = part_to_disk(p)->queue;
	unsigned int inflight[2];
	int cpu;

	cpu = part_stat_lock();
	part_round_stats(q, cpu, p);
	part_stat_unlock();
	part_in_flight(q, p, inflight);
	return sprintf(buf,
		"%8lu %8lu %8llu %8u "
		"%8lu %8lu %8llu %8u "
		"%8u %8u %8u"
		"\n",
		part_stat_read(p, ios[READ]),
		part_stat_read(p, merges[READ]),
		(unsigned long long)part_stat_read(p, sectors[READ]),
		jiffies_to_msecs(part_stat_read(p, ticks[READ])),
		part_stat_read(p, ios[WRITE]),
		part_stat_read(p, merges[WRITE]),
		(unsigned long long)part_stat_read(p, sectors[WRITE]),
		jiffies_to_msecs(part_stat_read(p, ticks[WRITE])),
		inflight[0],
		jiffies_to_msecs(part_stat_read(p, io_ticks)),
		jiffies_to_msecs(part_stat_read(p, time_in_queue)));
}

ssize_t part_inflight_show(struct device *dev, struct device_attribute *attr,
			   char *buf)
{
	struct hd_struct *p = dev_to_part(dev);
	struct request_queue *q = part_to_disk(p)->queue;
	unsigned int inflight[2];

	part_in_flight_rw(q, p, inflight);
	return sprintf(buf, "%8u %8u\n", inflight[0], inflight[1]);
}

#ifdef CONFIG_FAIL_MAKE_REQUEST
ssize_t part_fail_show(struct device *dev,
		       struct device_attribute *attr, char *buf)
{
	struct hd_struct *p = dev_to_part(dev);

	return sprintf(buf, "%d\n", p->make_it_fail);
}

ssize_t part_fail_store(struct device *dev,
			struct device_attribute *attr,
			const char *buf, size_t count)
{
	struct hd_struct *p = dev_to_part(dev);
	int i;

	if (count > 0 && sscanf(buf, "%d", &i) > 0)
		p->make_it_fail = (i == 0) ? 0 : 1;

	return count;
}
#endif

static DEVICE_ATTR(partition, S_IRUGO, part_partition_show, NULL);
static DEVICE_ATTR(start, S_IRUGO, part_start_show, NULL);
static DEVICE_ATTR(size, S_IRUGO, part_size_show, NULL);
static DEVICE_ATTR(ro, S_IRUGO, part_ro_show, NULL);
static DEVICE_ATTR(alignment_offset, S_IRUGO, part_alignment_offset_show, NULL);
static DEVICE_ATTR(discard_alignment, S_IRUGO, part_discard_alignment_show,
		   NULL);
static DEVICE_ATTR(stat, S_IRUGO, part_stat_show, NULL);
static DEVICE_ATTR(inflight, S_IRUGO, part_inflight_show, NULL);
#ifdef CONFIG_FAIL_MAKE_REQUEST
static struct device_attribute dev_attr_fail =
	__ATTR(make-it-fail, S_IRUGO|S_IWUSR, part_fail_show, part_fail_store);
#endif

static struct attribute *part_attrs[] = {
	&dev_attr_partition.attr,
	&dev_attr_start.attr,
	&dev_attr_size.attr,
	&dev_attr_ro.attr,
	&dev_attr_alignment_offset.attr,
	&dev_attr_discard_alignment.attr,
	&dev_attr_stat.attr,
	&dev_attr_inflight.attr,
#ifdef CONFIG_FAIL_MAKE_REQUEST
	&dev_attr_fail.attr,
#endif
	NULL
};

static struct attribute_group part_attr_group = {
	.attrs = part_attrs,
};

static const struct attribute_group *part_attr_groups[] = {
	&part_attr_group,
#ifdef CONFIG_BLK_DEV_IO_TRACE
	&blk_trace_attr_group,
#endif
	NULL
};

static void part_release(struct device *dev)
{
	struct hd_struct *p = dev_to_part(dev);
	blk_free_devt(dev->devt);
	hd_free_part(p);
	kfree(p);
}

static int part_uevent(struct device *dev, struct kobj_uevent_env *env)
{
	struct hd_struct *part = dev_to_part(dev);

	add_uevent_var(env, "PARTN=%u", part->partno);
	if (part->info && part->info->volname[0])
		add_uevent_var(env, "PARTNAME=%s", part->info->volname);
	return 0;
}

struct device_type part_type = {
	.name		= "partition",
	.groups		= part_attr_groups,
	.release	= part_release,
	.uevent		= part_uevent,
};

static void delete_partition_work_fn(struct work_struct *work)
{
	struct hd_struct *part = container_of(to_rcu_work(work), struct hd_struct,
					rcu_work);

	part->start_sect = 0;
	part->nr_sects = 0;
	part_stat_set_all(part, 0);
	put_device(part_to_dev(part));
}

void __delete_partition(struct percpu_ref *ref)
{
	struct hd_struct *part = container_of(ref, struct hd_struct, ref);
	INIT_RCU_WORK(&part->rcu_work, delete_partition_work_fn);
	queue_rcu_work(system_wq, &part->rcu_work);
}

/*
 * Must be called either with bd_mutex held, before a disk can be opened or
 * after all disk users are gone.
 */
void delete_partition(struct gendisk *disk, int partno)
{
	struct disk_part_tbl *ptbl =
		rcu_dereference_protected(disk->part_tbl, 1);
	struct hd_struct *part;

	if (partno >= ptbl->len)
		return;

	part = rcu_dereference_protected(ptbl->part[partno], 1);
	if (!part)
		return;

	rcu_assign_pointer(ptbl->part[partno], NULL);
	rcu_assign_pointer(ptbl->last_lookup, NULL);
	kobject_put(part->holder_dir);
	device_del(part_to_dev(part));

	hd_struct_kill(part);
}

static ssize_t whole_disk_show(struct device *dev,
			       struct device_attribute *attr, char *buf)
{
	return 0;
}
static DEVICE_ATTR(whole_disk, S_IRUSR | S_IRGRP | S_IROTH,
		   whole_disk_show, NULL);

/*
 * Must be called either with bd_mutex held, before a disk can be opened or
 * after all disk users are gone.
 */
struct hd_struct *add_partition(struct gendisk *disk, int partno,
				sector_t start, sector_t len, int flags,
				struct partition_meta_info *info)
{
	struct hd_struct *p;
	dev_t devt = MKDEV(0, 0);
	struct device *ddev = disk_to_dev(disk);
	struct device *pdev;
	struct disk_part_tbl *ptbl;
	const char *dname;
	int err;

	err = disk_expand_part_tbl(disk, partno);
	if (err)
		return ERR_PTR(err);
	ptbl = rcu_dereference_protected(disk->part_tbl, 1);

	if (ptbl->part[partno])
		return ERR_PTR(-EBUSY);

	p = kzalloc(sizeof(*p), GFP_KERNEL);
	if (!p)
		return ERR_PTR(-EBUSY);

	if (!init_part_stats(p)) {
		err = -ENOMEM;
		goto out_free;
	}

	seqcount_init(&p->nr_sects_seq);
	pdev = part_to_dev(p);

	p->start_sect = start;
	p->alignment_offset =
		queue_limit_alignment_offset(&disk->queue->limits, start);
	p->discard_alignment =
		queue_limit_discard_alignment(&disk->queue->limits, start);
	p->nr_sects = len;
	p->partno = partno;
	p->policy = get_disk_ro(disk);

	if (info) {
		struct partition_meta_info *pinfo = alloc_part_info(disk);
		if (!pinfo) {
			err = -ENOMEM;
			goto out_free_stats;
		}
		memcpy(pinfo, info, sizeof(*info));
		p->info = pinfo;
	}

	dname = dev_name(ddev);
	if (isdigit(dname[strlen(dname) - 1]))
		dev_set_name(pdev, "%sp%d", dname, partno);
	else
		dev_set_name(pdev, "%s%d", dname, partno);

	device_initialize(pdev);
	pdev->class = &block_class;
	pdev->type = &part_type;
	pdev->parent = ddev;

	err = blk_alloc_devt(p, &devt);
	if (err)
		goto out_free_info;
	pdev->devt = devt;

	/* delay uevent until 'holders' subdir is created */
	dev_set_uevent_suppress(pdev, 1);
	err = device_add(pdev);
	if (err)
		goto out_put;

	err = -ENOMEM;
	p->holder_dir = kobject_create_and_add("holders", &pdev->kobj);
	if (!p->holder_dir)
		goto out_del;

	dev_set_uevent_suppress(pdev, 0);
	if (flags & ADDPART_FLAG_WHOLEDISK) {
		err = device_create_file(pdev, &dev_attr_whole_disk);
		if (err)
			goto out_del;
	}

	err = hd_ref_init(p);
	if (err) {
		if (flags & ADDPART_FLAG_WHOLEDISK)
			goto out_remove_file;
		goto out_del;
	}

	/* everything is up and running, commence */
	rcu_assign_pointer(ptbl->part[partno], p);

	/* suppress uevent if the disk suppresses it */
	if (!dev_get_uevent_suppress(ddev))
		kobject_uevent(&pdev->kobj, KOBJ_ADD);
	return p;

out_free_info:
	free_part_info(p);
out_free_stats:
	free_part_stats(p);
out_free:
	kfree(p);
	return ERR_PTR(err);
out_remove_file:
	device_remove_file(pdev, &dev_attr_whole_disk);
out_del:
	kobject_put(p->holder_dir);
	device_del(pdev);
out_put:
	put_device(pdev);
	return ERR_PTR(err);
}

static bool disk_unlock_native_capacity(struct gendisk *disk)
{
	const struct block_device_operations *bdops = disk->fops;

	if (bdops->unlock_native_capacity &&
	    !(disk->flags & GENHD_FL_NATIVE_CAPACITY)) {
		printk(KERN_CONT "enabling native capacity\n");
		bdops->unlock_native_capacity(disk);
		disk->flags |= GENHD_FL_NATIVE_CAPACITY;
		return true;
	} else {
		printk(KERN_CONT "truncated\n");
		return false;
	}
}

static int drop_partitions(struct gendisk *disk, struct block_device *bdev)
{
	struct disk_part_iter piter;
	struct hd_struct *part;
	int res;

	if (bdev->bd_part_count || bdev->bd_super)
		return -EBUSY;
	res = invalidate_partition(disk, 0);
	if (res)
		return res;

	disk_part_iter_init(&piter, disk, DISK_PITER_INCL_EMPTY);
	while ((part = disk_part_iter_next(&piter)))
		delete_partition(disk, part->partno);
	disk_part_iter_exit(&piter);

	return 0;
}

static bool part_zone_aligned(struct gendisk *disk,
			      struct block_device *bdev,
			      sector_t from, sector_t size)
{
	unsigned int zone_sectors = bdev_zone_sectors(bdev);

	/*
	 * If this function is called, then the disk is a zoned block device
	 * (host-aware or host-managed). This can be detected even if the
	 * zoned block device support is disabled (CONFIG_BLK_DEV_ZONED not
	 * set). In this case, however, only host-aware devices will be seen
	 * as a block device is not created for host-managed devices. Without
	 * zoned block device support, host-aware drives can still be used as
	 * regular block devices (no zone operation) and their zone size will
	 * be reported as 0. Allow this case.
	 */
	if (!zone_sectors)
		return true;

	/*
	 * Check partition start and size alignement. If the drive has a
	 * smaller last runt zone, ignore it and allow the partition to
	 * use it. Check the zone size too: it should be a power of 2 number
	 * of sectors.
	 */
	if (WARN_ON_ONCE(!is_power_of_2(zone_sectors))) {
		u32 rem;

		div_u64_rem(from, zone_sectors, &rem);
		if (rem)
			return false;
		if ((from + size) < get_capacity(disk)) {
			div_u64_rem(size, zone_sectors, &rem);
			if (rem)
				return false;
		}

	} else {

		if (from & (zone_sectors - 1))
			return false;
		if ((from + size) < get_capacity(disk) &&
		    (size & (zone_sectors - 1)))
			return false;

	}

	return true;
}

int rescan_partitions(struct gendisk *disk, struct block_device *bdev)
{
	struct parsed_partitions *state = NULL;
	struct hd_struct *part;
	int p, highest, res;
rescan:
	if (state && !IS_ERR(state)) {
		free_partitions(state);
		state = NULL;
	}

	res = drop_partitions(disk, bdev);
	if (res)
		return res;

	if (disk->fops->revalidate_disk)
		disk->fops->revalidate_disk(disk);
	check_disk_size_change(disk, bdev);
	bdev->bd_invalidated = 0;
	if (!get_capacity(disk) || !(state = check_partition(disk, bdev)))
		return 0;
	if (IS_ERR(state)) {
		/*
		 * I/O error reading the partition table.  If any
		 * partition code tried to read beyond EOD, retry
		 * after unlocking native capacity.
		 */
		if (PTR_ERR(state) == -ENOSPC) {
			printk(KERN_WARNING "%s: partition table beyond EOD, ",
			       disk->disk_name);
			if (disk_unlock_native_capacity(disk))
				goto rescan;
		}
		return -EIO;
	}
	/*
	 * If any partition code tried to read beyond EOD, try
	 * unlocking native capacity even if partition table is
	 * successfully read as we could be missing some partitions.
	 */
	if (state->access_beyond_eod) {
		printk(KERN_WARNING
		       "%s: partition table partially beyond EOD, ",
		       disk->disk_name);
		if (disk_unlock_native_capacity(disk))
			goto rescan;
	}

	/* tell userspace that the media / partition table may have changed */
	kobject_uevent(&disk_to_dev(disk)->kobj, KOBJ_CHANGE);

	/* Detect the highest partition number and preallocate
	 * disk->part_tbl.  This is an optimization and not strictly
	 * necessary.
	 */
	for (p = 1, highest = 0; p < state->limit; p++)
		if (state->parts[p].size)
			highest = p;

	disk_expand_part_tbl(disk, highest);

	/* add partitions */
	for (p = 1; p < state->limit; p++) {
		sector_t size, from;

		size = state->parts[p].size;
		if (!size)
			continue;

		from = state->parts[p].from;
		if (from >= get_capacity(disk)) {
			printk(KERN_WARNING
			       "%s: p%d start %llu is beyond EOD, ",
			       disk->disk_name, p, (unsigned long long) from);
			if (disk_unlock_native_capacity(disk))
				goto rescan;
			continue;
		}

		if (from + size > get_capacity(disk)) {
			printk(KERN_WARNING
			       "%s: p%d size %llu extends beyond EOD, ",
			       disk->disk_name, p, (unsigned long long) size);

			if (disk_unlock_native_capacity(disk)) {
				/* free state and restart */
				goto rescan;
			} else {
				/*
				 * we can not ignore partitions of broken tables
				 * created by for example camera firmware, but
				 * we limit them to the end of the disk to avoid
				 * creating invalid block devices
				 */
				size = get_capacity(disk) - from;
			}
		}

		/*
		 * On a zoned block device, partitions should be aligned on the
		 * device zone size (i.e. zone boundary crossing not allowed).
		 * Otherwise, resetting the write pointer of the last zone of
		 * one partition may impact the following partition.
		 */
		if (bdev_is_zoned(bdev) &&
		    !part_zone_aligned(disk, bdev, from, size)) {
			printk(KERN_WARNING
			       "%s: p%d start %llu+%llu is not zone aligned\n",
			       disk->disk_name, p, (unsigned long long) from,
			       (unsigned long long) size);
			continue;
		}

		part = add_partition(disk, p, from, size,
				     state->parts[p].flags,
				     &state->parts[p].info);
		if (IS_ERR(part)) {
			printk(KERN_ERR " %s: p%d could not be added: %ld\n",
			       disk->disk_name, p, -PTR_ERR(part));
			continue;
		}
#ifdef CONFIG_BLK_DEV_MD
		if (state->parts[p].flags & ADDPART_FLAG_RAID)
			md_autodetect_dev(part_to_dev(part)->devt);
#endif
	}
	free_partitions(state);
	return 0;
}

int invalidate_partitions(struct gendisk *disk, struct block_device *bdev)
{
	int res;

	if (!bdev->bd_invalidated)
		return 0;

	res = drop_partitions(disk, bdev);
	if (res)
		return res;

	set_capacity(disk, 0);
	check_disk_size_change(disk, bdev);
	bdev->bd_invalidated = 0;
	/* tell userspace that the media / partition table may have changed */
	kobject_uevent(&disk_to_dev(disk)->kobj, KOBJ_CHANGE);

	return 0;
}

unsigned char *read_dev_sector(struct block_device *bdev, sector_t n, Sector *p)
{
	struct address_space *mapping = bdev->bd_inode->i_mapping;
	struct page *page;

	page = read_mapping_page(mapping, (pgoff_t)(n >> (PAGE_SHIFT-9)), NULL);
	if (!IS_ERR(page)) {
		if (PageError(page))
			goto fail;
		p->v = page;
		return (unsigned char *)page_address(page) +  ((n & ((1 << (PAGE_SHIFT - 9)) - 1)) << 9);
fail:
		put_page(page);
	}
	p->v = NULL;
	return NULL;
}

EXPORT_SYMBOL(read_dev_sector);
Commit	Line	Data
b2441318	1	// SPDX-License-Identifier: GPL-2.0
94ea4158 AV	2	/*
	3	* Code extracted from drivers/block/genhd.c
	4	* Copyright (C) 1991-1998 Linus Torvalds
	5	* Re-organised Feb 1998 Russell King
	6	*
	7	* We now have independent partition support from the
	8	* block drivers, which allows all the partition code to
	9	* be grouped in one location, and it to be mostly self
	10	* contained.
	11	*/
	12
	13	#include <linux/init.h>
	14	#include <linux/module.h>
	15	#include <linux/fs.h>
	16	#include <linux/slab.h>
	17	#include <linux/kmod.h>
	18	#include <linux/ctype.h>
	19	#include <linux/genhd.h>
	20	#include <linux/blktrace_api.h>
	21
	22	#include "partitions/check.h"
	23
	24	#ifdef CONFIG_BLK_DEV_MD
	25	extern void md_autodetect_dev(dev_t dev);
	26	#endif
	27
	28	/*
	29	* disk_name() is used by partition check code and the genhd driver.
	30	* It formats the devicename of the indicated disk into
	31	* the supplied buffer (of size at least 32), and returns
	32	* a pointer to that same buffer (for convenience).
	33	*/
	34
	35	char disk_name(struct gendisk hd, int partno, char *buf)
	36	{
	37	if (!partno)
	38	snprintf(buf, BDEVNAME_SIZE, "%s", hd->disk_name);
	39	else if (isdigit(hd->disk_name[strlen(hd->disk_name)-1]))
	40	snprintf(buf, BDEVNAME_SIZE, "%sp%d", hd->disk_name, partno);
	41	else
	42	snprintf(buf, BDEVNAME_SIZE, "%s%d", hd->disk_name, partno);
	43
	44	return buf;
	45	}
	46
	47	const char bdevname(struct block_device bdev, char *buf)
	48	{
	49	return disk_name(bdev->bd_disk, bdev->bd_part->partno, buf);
	50	}
	51
	52	EXPORT_SYMBOL(bdevname);
	53
68060090 JX	54	const char bio_devname(struct bio bio, char *buf)
	55	{
	56	return disk_name(bio->bi_disk, bio->bi_partno, buf);
	57	}
	58	EXPORT_SYMBOL(bio_devname);
	59
94ea4158 AV	60	/*
	61	* There's very little reason to use this, you should really
	62	* have a struct block_device just about everywhere and use
	63	* bdevname() instead.
	64	*/
	65	const char __bdevname(dev_t dev, char buffer)
	66	{
	67	scnprintf(buffer, BDEVNAME_SIZE, "unknown-block(%u,%u)",
	68	MAJOR(dev), MINOR(dev));
	69	return buffer;
	70	}
	71
	72	EXPORT_SYMBOL(__bdevname);
	73
	74	static ssize_t part_partition_show(struct device *dev,
	75	struct device_attribute attr, char buf)
	76	{
	77	struct hd_struct *p = dev_to_part(dev);
	78
	79	return sprintf(buf, "%d\n", p->partno);
	80	}
	81
	82	static ssize_t part_start_show(struct device *dev,
	83	struct device_attribute attr, char buf)
	84	{
	85	struct hd_struct *p = dev_to_part(dev);
	86
	87	return sprintf(buf, "%llu\n",(unsigned long long)p->start_sect);
	88	}
	89
	90	ssize_t part_size_show(struct device *dev,
	91	struct device_attribute attr, char buf)
	92	{
	93	struct hd_struct *p = dev_to_part(dev);
c83f6bf9	94	return sprintf(buf, "%llu\n",(unsigned long long)part_nr_sects_read(p));
94ea4158 AV	95	}
	96
	97	static ssize_t part_ro_show(struct device *dev,
	98	struct device_attribute attr, char buf)
	99	{
	100	struct hd_struct *p = dev_to_part(dev);
	101	return sprintf(buf, "%d\n", p->policy ? 1 : 0);
	102	}
	103
	104	static ssize_t part_alignment_offset_show(struct device *dev,
	105	struct device_attribute attr, char buf)
	106	{
	107	struct hd_struct *p = dev_to_part(dev);
	108	return sprintf(buf, "%llu\n", (unsigned long long)p->alignment_offset);
	109	}
	110
	111	static ssize_t part_discard_alignment_show(struct device *dev,
	112	struct device_attribute attr, char buf)
	113	{
	114	struct hd_struct *p = dev_to_part(dev);
	115	return sprintf(buf, "%u\n", p->discard_alignment);
	116	}
	117
	118	ssize_t part_stat_show(struct device *dev,
	119	struct device_attribute attr, char buf)
	120	{
	121	struct hd_struct *p = dev_to_part(dev);
f5c156c4	122	struct request_queue *q = part_to_disk(p)->queue;
0609e0ef	123	unsigned int inflight[2];
94ea4158 AV	124	int cpu;
	125
	126	cpu = part_stat_lock();
d62e26b3	127	part_round_stats(q, cpu, p);
94ea4158	128	part_stat_unlock();
0609e0ef	129	part_in_flight(q, p, inflight);
94ea4158 AV	130	return sprintf(buf,
	131	"%8lu %8lu %8llu %8u "
	132	"%8lu %8lu %8llu %8u "
	133	"%8u %8u %8u"
	134	"\n",
	135	part_stat_read(p, ios[READ]),
	136	part_stat_read(p, merges[READ]),
	137	(unsigned long long)part_stat_read(p, sectors[READ]),
	138	jiffies_to_msecs(part_stat_read(p, ticks[READ])),
	139	part_stat_read(p, ios[WRITE]),
	140	part_stat_read(p, merges[WRITE]),
	141	(unsigned long long)part_stat_read(p, sectors[WRITE]),
	142	jiffies_to_msecs(part_stat_read(p, ticks[WRITE])),
0609e0ef	143	inflight[0],
94ea4158 AV	144	jiffies_to_msecs(part_stat_read(p, io_ticks)),
	145	jiffies_to_msecs(part_stat_read(p, time_in_queue)));
	146	}
	147
416d3aa7 OS	148	ssize_t part_inflight_show(struct device dev, struct device_attribute attr,
416d3aa7 OS	149	char *buf)
94ea4158 AV	150	{
94ea4158 AV	151	struct hd_struct *p = dev_to_part(dev);
416d3aa7 OS	152	struct request_queue *q = part_to_disk(p)->queue;
416d3aa7 OS	153	unsigned int inflight[2];
94ea4158	154
416d3aa7 OS	155	part_in_flight_rw(q, p, inflight);
416d3aa7 OS	156	return sprintf(buf, "%8u %8u\n", inflight[0], inflight[1]);
94ea4158 AV	157	}
	158
	159	#ifdef CONFIG_FAIL_MAKE_REQUEST
	160	ssize_t part_fail_show(struct device *dev,
	161	struct device_attribute attr, char buf)
	162	{
	163	struct hd_struct *p = dev_to_part(dev);
	164
	165	return sprintf(buf, "%d\n", p->make_it_fail);
	166	}
	167
	168	ssize_t part_fail_store(struct device *dev,
	169	struct device_attribute *attr,
	170	const char *buf, size_t count)
	171	{
	172	struct hd_struct *p = dev_to_part(dev);
	173	int i;
	174
	175	if (count > 0 && sscanf(buf, "%d", &i) > 0)
	176	p->make_it_fail = (i == 0) ? 0 : 1;
	177
	178	return count;
	179	}
	180	#endif
	181
	182	static DEVICE_ATTR(partition, S_IRUGO, part_partition_show, NULL);
	183	static DEVICE_ATTR(start, S_IRUGO, part_start_show, NULL);
	184	static DEVICE_ATTR(size, S_IRUGO, part_size_show, NULL);
	185	static DEVICE_ATTR(ro, S_IRUGO, part_ro_show, NULL);
	186	static DEVICE_ATTR(alignment_offset, S_IRUGO, part_alignment_offset_show, NULL);
	187	static DEVICE_ATTR(discard_alignment, S_IRUGO, part_discard_alignment_show,
	188	NULL);
	189	static DEVICE_ATTR(stat, S_IRUGO, part_stat_show, NULL);
	190	static DEVICE_ATTR(inflight, S_IRUGO, part_inflight_show, NULL);
	191	#ifdef CONFIG_FAIL_MAKE_REQUEST
	192	static struct device_attribute dev_attr_fail =
	193	__ATTR(make-it-fail, S_IRUGO\|S_IWUSR, part_fail_show, part_fail_store);
	194	#endif
	195
	196	static struct attribute *part_attrs[] = {
	197	&dev_attr_partition.attr,
	198	&dev_attr_start.attr,
	199	&dev_attr_size.attr,
	200	&dev_attr_ro.attr,
	201	&dev_attr_alignment_offset.attr,
	202	&dev_attr_discard_alignment.attr,
	203	&dev_attr_stat.attr,
	204	&dev_attr_inflight.attr,
	205	#ifdef CONFIG_FAIL_MAKE_REQUEST
	206	&dev_attr_fail.attr,
	207	#endif
	208	NULL
	209	};
	210
	211	static struct attribute_group part_attr_group = {
	212	.attrs = part_attrs,
	213	};
	214
	215	static const struct attribute_group *part_attr_groups[] = {
	216	&part_attr_group,
	217	#ifdef CONFIG_BLK_DEV_IO_TRACE
	218	&blk_trace_attr_group,
	219	#endif
	220	NULL
221	};
222
223	static void part_release(struct device *dev)
224	{
225	struct hd_struct *p = dev_to_part(dev);
2da78092	226	blk_free_devt(dev->devt);
b54e5ed8	227	hd_free_part(p);
94ea4158 AV	228	kfree(p);
	229	}
	230
0d9c51a6 SM	231	static int part_uevent(struct device dev, struct kobj_uevent_env env)
	232	{
	233	struct hd_struct *part = dev_to_part(dev);
	234
	235	add_uevent_var(env, "PARTN=%u", part->partno);
	236	if (part->info && part->info->volname[0])
	237	add_uevent_var(env, "PARTNAME=%s", part->info->volname);
	238	return 0;
	239	}
	240
94ea4158 AV	241	struct device_type part_type = {
	242	.name = "partition",
	243	.groups = part_attr_groups,
	244	.release = part_release,
0d9c51a6	245	.uevent = part_uevent,
94ea4158 AV	246	};
94ea4158 AV	247
51f2e869	248	static void delete_partition_work_fn(struct work_struct *work)
94ea4158	249	{
51f2e869 YY	250	struct hd_struct *part = container_of(to_rcu_work(work), struct hd_struct,
51f2e869 YY	251	rcu_work);
94ea4158 AV	252
	253	part->start_sect = 0;
	254	part->nr_sects = 0;
	255	part_stat_set_all(part, 0);
	256	put_device(part_to_dev(part));
	257	}
	258
6c71013e	259	void __delete_partition(struct percpu_ref *ref)
94ea4158	260	{
6c71013e	261	struct hd_struct *part = container_of(ref, struct hd_struct, ref);
51f2e869 YY	262	INIT_RCU_WORK(&part->rcu_work, delete_partition_work_fn);
51f2e869 YY	263	queue_rcu_work(system_wq, &part->rcu_work);
94ea4158 AV	264	}
94ea4158 AV	265
6d2cf6f2 BVA	266	/*
	267	* Must be called either with bd_mutex held, before a disk can be opened or
	268	* after all disk users are gone.
	269	*/
94ea4158 AV	270	void delete_partition(struct gendisk *disk, int partno)
94ea4158 AV	271	{
6d2cf6f2 BVA	272	struct disk_part_tbl *ptbl =
6d2cf6f2 BVA	273	rcu_dereference_protected(disk->part_tbl, 1);
94ea4158 AV	274	struct hd_struct *part;
	275
	276	if (partno >= ptbl->len)
	277	return;
	278
6d2cf6f2	279	part = rcu_dereference_protected(ptbl->part[partno], 1);
94ea4158 AV	280	if (!part)
	281	return;
	282
94ea4158 AV	283	rcu_assign_pointer(ptbl->part[partno], NULL);
	284	rcu_assign_pointer(ptbl->last_lookup, NULL);
	285	kobject_put(part->holder_dir);
	286	device_del(part_to_dev(part));
	287
6c71013e	288	hd_struct_kill(part);
94ea4158 AV	289	}
	290
	291	static ssize_t whole_disk_show(struct device *dev,
	292	struct device_attribute attr, char buf)
	293	{
	294	return 0;
	295	}
	296	static DEVICE_ATTR(whole_disk, S_IRUSR \| S_IRGRP \| S_IROTH,
	297	whole_disk_show, NULL);
	298
6d2cf6f2 BVA	299	/*
	300	* Must be called either with bd_mutex held, before a disk can be opened or
	301	* after all disk users are gone.
	302	*/
94ea4158 AV	303	struct hd_struct add_partition(struct gendisk disk, int partno,
	304	sector_t start, sector_t len, int flags,
	305	struct partition_meta_info *info)
	306	{
	307	struct hd_struct *p;
	308	dev_t devt = MKDEV(0, 0);
	309	struct device *ddev = disk_to_dev(disk);
	310	struct device *pdev;
	311	struct disk_part_tbl *ptbl;
	312	const char *dname;
	313	int err;
	314
	315	err = disk_expand_part_tbl(disk, partno);
	316	if (err)
	317	return ERR_PTR(err);
6d2cf6f2	318	ptbl = rcu_dereference_protected(disk->part_tbl, 1);
94ea4158 AV	319
	320	if (ptbl->part[partno])
	321	return ERR_PTR(-EBUSY);
	322
	323	p = kzalloc(sizeof(*p), GFP_KERNEL);
	324	if (!p)
	325	return ERR_PTR(-EBUSY);
	326
	327	if (!init_part_stats(p)) {
	328	err = -ENOMEM;
	329	goto out_free;
	330	}
c83f6bf9 VG	331
c83f6bf9 VG	332	seqcount_init(&p->nr_sects_seq);
94ea4158 AV	333	pdev = part_to_dev(p);
	334
	335	p->start_sect = start;
	336	p->alignment_offset =
	337	queue_limit_alignment_offset(&disk->queue->limits, start);
	338	p->discard_alignment =
	339	queue_limit_discard_alignment(&disk->queue->limits, start);
	340	p->nr_sects = len;
	341	p->partno = partno;
	342	p->policy = get_disk_ro(disk);
	343
	344	if (info) {
	345	struct partition_meta_info *pinfo = alloc_part_info(disk);
7bd897cf DC	346	if (!pinfo) {
7bd897cf DC	347	err = -ENOMEM;
94ea4158	348	goto out_free_stats;
7bd897cf	349	}
94ea4158 AV	350	memcpy(pinfo, info, sizeof(*info));
	351	p->info = pinfo;
	352	}
	353
	354	dname = dev_name(ddev);
	355	if (isdigit(dname[strlen(dname) - 1]))
	356	dev_set_name(pdev, "%sp%d", dname, partno);
	357	else
	358	dev_set_name(pdev, "%s%d", dname, partno);
	359
	360	device_initialize(pdev);
	361	pdev->class = &block_class;
	362	pdev->type = &part_type;
	363	pdev->parent = ddev;
	364
	365	err = blk_alloc_devt(p, &devt);
	366	if (err)
	367	goto out_free_info;
	368	pdev->devt = devt;
	369
	370	/* delay uevent until 'holders' subdir is created */
	371	dev_set_uevent_suppress(pdev, 1);
	372	err = device_add(pdev);
	373	if (err)
	374	goto out_put;
	375
	376	err = -ENOMEM;
	377	p->holder_dir = kobject_create_and_add("holders", &pdev->kobj);
	378	if (!p->holder_dir)
	379	goto out_del;
	380
	381	dev_set_uevent_suppress(pdev, 0);
	382	if (flags & ADDPART_FLAG_WHOLEDISK) {
	383	err = device_create_file(pdev, &dev_attr_whole_disk);
	384	if (err)
	385	goto out_del;
	386	}
	387
b30a337c ML	388	err = hd_ref_init(p);
	389	if (err) {
	390	if (flags & ADDPART_FLAG_WHOLEDISK)
	391	goto out_remove_file;
	392	goto out_del;
	393	}
	394
94ea4158 AV	395	/* everything is up and running, commence */
	396	rcu_assign_pointer(ptbl->part[partno], p);
	397
	398	/* suppress uevent if the disk suppresses it */
	399	if (!dev_get_uevent_suppress(ddev))
	400	kobject_uevent(&pdev->kobj, KOBJ_ADD);
b30a337c	401	return p;
94ea4158 AV	402
	403	out_free_info:
	404	free_part_info(p);
	405	out_free_stats:
	406	free_part_stats(p);
	407	out_free:
	408	kfree(p);
	409	return ERR_PTR(err);
b30a337c ML	410	out_remove_file:
b30a337c ML	411	device_remove_file(pdev, &dev_attr_whole_disk);
94ea4158 AV	412	out_del:
	413	kobject_put(p->holder_dir);
	414	device_del(pdev);
	415	out_put:
	416	put_device(pdev);
94ea4158 AV	417	return ERR_PTR(err);
	418	}
	419
	420	static bool disk_unlock_native_capacity(struct gendisk *disk)
	421	{
	422	const struct block_device_operations *bdops = disk->fops;
	423
	424	if (bdops->unlock_native_capacity &&
	425	!(disk->flags & GENHD_FL_NATIVE_CAPACITY)) {
	426	printk(KERN_CONT "enabling native capacity\n");
	427	bdops->unlock_native_capacity(disk);
	428	disk->flags \|= GENHD_FL_NATIVE_CAPACITY;
	429	return true;
	430	} else {
	431	printk(KERN_CONT "truncated\n");
	432	return false;
	433	}
	434	}
	435
fe316bf2	436	static int drop_partitions(struct gendisk disk, struct block_device bdev)
94ea4158	437	{
94ea4158 AV	438	struct disk_part_iter piter;
94ea4158 AV	439	struct hd_struct *part;
fe316bf2	440	int res;
94ea4158	441
77032ca6	442	if (bdev->bd_part_count \|\| bdev->bd_super)
94ea4158 AV	443	return -EBUSY;
	444	res = invalidate_partition(disk, 0);
	445	if (res)
	446	return res;
	447
	448	disk_part_iter_init(&piter, disk, DISK_PITER_INCL_EMPTY);
	449	while ((part = disk_part_iter_next(&piter)))
	450	delete_partition(disk, part->partno);
	451	disk_part_iter_exit(&piter);
	452
fe316bf2 JN	453	return 0;
	454	}
	455
b02d8aae DLM	456	static bool part_zone_aligned(struct gendisk *disk,
	457	struct block_device *bdev,
	458	sector_t from, sector_t size)
	459	{
f99e8648	460	unsigned int zone_sectors = bdev_zone_sectors(bdev);
b02d8aae DLM	461
	462	/*
	463	* If this function is called, then the disk is a zoned block device
	464	* (host-aware or host-managed). This can be detected even if the
	465	* zoned block device support is disabled (CONFIG_BLK_DEV_ZONED not
	466	* set). In this case, however, only host-aware devices will be seen
	467	* as a block device is not created for host-managed devices. Without
	468	* zoned block device support, host-aware drives can still be used as
	469	* regular block devices (no zone operation) and their zone size will
	470	* be reported as 0. Allow this case.
	471	*/
f99e8648	472	if (!zone_sectors)
b02d8aae DLM	473	return true;
	474
	475	/*
	476	* Check partition start and size alignement. If the drive has a
	477	* smaller last runt zone, ignore it and allow the partition to
	478	* use it. Check the zone size too: it should be a power of 2 number
	479	* of sectors.
	480	*/
f99e8648	481	if (WARN_ON_ONCE(!is_power_of_2(zone_sectors))) {
b02d8aae DLM	482	u32 rem;
b02d8aae DLM	483
f99e8648	484	div_u64_rem(from, zone_sectors, &rem);
b02d8aae DLM	485	if (rem)
	486	return false;
	487	if ((from + size) < get_capacity(disk)) {
f99e8648	488	div_u64_rem(size, zone_sectors, &rem);
b02d8aae DLM	489	if (rem)
	490	return false;
	491	}
	492
	493	} else {
	494
f99e8648	495	if (from & (zone_sectors - 1))
b02d8aae DLM	496	return false;
b02d8aae DLM	497	if ((from + size) < get_capacity(disk) &&
f99e8648	498	(size & (zone_sectors - 1)))
b02d8aae DLM	499	return false;
	500
	501	}
	502
	503	return true;
	504	}
	505
fe316bf2 JN	506	int rescan_partitions(struct gendisk disk, struct block_device bdev)
	507	{
	508	struct parsed_partitions *state = NULL;
	509	struct hd_struct *part;
	510	int p, highest, res;
	511	rescan:
	512	if (state && !IS_ERR(state)) {
ac2e5327	513	free_partitions(state);
fe316bf2 JN	514	state = NULL;
	515	}
	516
	517	res = drop_partitions(disk, bdev);
	518	if (res)
	519	return res;
	520
94ea4158 AV	521	if (disk->fops->revalidate_disk)
	522	disk->fops->revalidate_disk(disk);
	523	check_disk_size_change(disk, bdev);
	524	bdev->bd_invalidated = 0;
	525	if (!get_capacity(disk) \|\| !(state = check_partition(disk, bdev)))
	526	return 0;
	527	if (IS_ERR(state)) {
	528	/*
	529	* I/O error reading the partition table. If any
	530	* partition code tried to read beyond EOD, retry
	531	* after unlocking native capacity.
	532	*/
	533	if (PTR_ERR(state) == -ENOSPC) {
	534	printk(KERN_WARNING "%s: partition table beyond EOD, ",
	535	disk->disk_name);
	536	if (disk_unlock_native_capacity(disk))
	537	goto rescan;
	538	}
	539	return -EIO;
	540	}
	541	/*
	542	* If any partition code tried to read beyond EOD, try
	543	* unlocking native capacity even if partition table is
	544	* successfully read as we could be missing some partitions.
	545	*/
	546	if (state->access_beyond_eod) {
	547	printk(KERN_WARNING
	548	"%s: partition table partially beyond EOD, ",
	549	disk->disk_name);
	550	if (disk_unlock_native_capacity(disk))
	551	goto rescan;
	552	}
	553
	554	/* tell userspace that the media / partition table may have changed */
	555	kobject_uevent(&disk_to_dev(disk)->kobj, KOBJ_CHANGE);
	556
	557	/* Detect the highest partition number and preallocate
	558	* disk->part_tbl. This is an optimization and not strictly
	559	* necessary.
	560	*/
	561	for (p = 1, highest = 0; p < state->limit; p++)
	562	if (state->parts[p].size)
	563	highest = p;
	564
	565	disk_expand_part_tbl(disk, highest);
	566
	567	/* add partitions */
	568	for (p = 1; p < state->limit; p++) {
	569	sector_t size, from;
94ea4158 AV	570
	571	size = state->parts[p].size;
	572	if (!size)
	573	continue;
	574
	575	from = state->parts[p].from;
	576	if (from >= get_capacity(disk)) {
	577	printk(KERN_WARNING
	578	"%s: p%d start %llu is beyond EOD, ",
	579	disk->disk_name, p, (unsigned long long) from);
	580	if (disk_unlock_native_capacity(disk))
	581	goto rescan;
	582	continue;
	583	}
	584
	585	if (from + size > get_capacity(disk)) {
	586	printk(KERN_WARNING
	587	"%s: p%d size %llu extends beyond EOD, ",
	588	disk->disk_name, p, (unsigned long long) size);
	589
	590	if (disk_unlock_native_capacity(disk)) {
	591	/* free state and restart */
	592	goto rescan;
	593	} else {
	594	/*
	595	* we can not ignore partitions of broken tables
	596	* created by for example camera firmware, but
	597	* we limit them to the end of the disk to avoid
	598	* creating invalid block devices
	599	*/
	600	size = get_capacity(disk) - from;
	601	}
	602	}
	603
b02d8aae DLM	604	/*
	605	* On a zoned block device, partitions should be aligned on the
	606	* device zone size (i.e. zone boundary crossing not allowed).
	607	* Otherwise, resetting the write pointer of the last zone of
	608	* one partition may impact the following partition.
	609	*/
	610	if (bdev_is_zoned(bdev) &&
	611	!part_zone_aligned(disk, bdev, from, size)) {
	612	printk(KERN_WARNING
	613	"%s: p%d start %llu+%llu is not zone aligned\n",
	614	disk->disk_name, p, (unsigned long long) from,
	615	(unsigned long long) size);
	616	continue;
	617	}
	618
94ea4158 AV	619	part = add_partition(disk, p, from, size,
	620	state->parts[p].flags,
	621	&state->parts[p].info);
	622	if (IS_ERR(part)) {
	623	printk(KERN_ERR " %s: p%d could not be added: %ld\n",
	624	disk->disk_name, p, -PTR_ERR(part));
	625	continue;
	626	}
	627	#ifdef CONFIG_BLK_DEV_MD
	628	if (state->parts[p].flags & ADDPART_FLAG_RAID)
	629	md_autodetect_dev(part_to_dev(part)->devt);
	630	#endif
	631	}
ac2e5327	632	free_partitions(state);
94ea4158 AV	633	return 0;
	634	}
	635
fe316bf2 JN	636	int invalidate_partitions(struct gendisk disk, struct block_device bdev)
	637	{
	638	int res;
	639
	640	if (!bdev->bd_invalidated)
	641	return 0;
	642
	643	res = drop_partitions(disk, bdev);
	644	if (res)
	645	return res;
	646
	647	set_capacity(disk, 0);
	648	check_disk_size_change(disk, bdev);
	649	bdev->bd_invalidated = 0;
	650	/* tell userspace that the media / partition table may have changed */
	651	kobject_uevent(&disk_to_dev(disk)->kobj, KOBJ_CHANGE);
	652
	653	return 0;
	654	}
	655
d1a5f2b4 DW	656	unsigned char read_dev_sector(struct block_device bdev, sector_t n, Sector *p)
d1a5f2b4 DW	657	{
a41fe02b	658	struct address_space *mapping = bdev->bd_inode->i_mapping;
94ea4158 AV	659	struct page *page;
94ea4158 AV	660
a41fe02b	661	page = read_mapping_page(mapping, (pgoff_t)(n >> (PAGE_SHIFT-9)), NULL);
94ea4158 AV	662	if (!IS_ERR(page)) {
	663	if (PageError(page))
	664	goto fail;
	665	p->v = page;
09cbfeaf	666	return (unsigned char *)page_address(page) + ((n & ((1 << (PAGE_SHIFT - 9)) - 1)) << 9);
94ea4158	667	fail:
09cbfeaf	668	put_page(page);
94ea4158 AV	669	}
	670	p->v = NULL;
	671	return NULL;
	672	}
	673
	674	EXPORT_SYMBOL(read_dev_sector);