[linux.git] / block / blk-ia-ranges.c

// SPDX-License-Identifier: GPL-2.0
/*
 *  Block device concurrent positioning ranges.
 *
 *  Copyright (C) 2021 Western Digital Corporation or its Affiliates.
 */
#include <linux/kernel.h>
#include <linux/blkdev.h>
#include <linux/slab.h>
#include <linux/init.h>

#include "blk.h"

static ssize_t
blk_ia_range_sector_show(struct blk_independent_access_range *iar,
			 char *buf)
{
	return sprintf(buf, "%llu\n", iar->sector);
}

static ssize_t
blk_ia_range_nr_sectors_show(struct blk_independent_access_range *iar,
			     char *buf)
{
	return sprintf(buf, "%llu\n", iar->nr_sectors);
}

struct blk_ia_range_sysfs_entry {
	struct attribute attr;
	ssize_t (*show)(struct blk_independent_access_range *iar, char *buf);
};

static struct blk_ia_range_sysfs_entry blk_ia_range_sector_entry = {
	.attr = { .name = "sector", .mode = 0444 },
	.show = blk_ia_range_sector_show,
};

static struct blk_ia_range_sysfs_entry blk_ia_range_nr_sectors_entry = {
	.attr = { .name = "nr_sectors", .mode = 0444 },
	.show = blk_ia_range_nr_sectors_show,
};

static struct attribute *blk_ia_range_attrs[] = {
	&blk_ia_range_sector_entry.attr,
	&blk_ia_range_nr_sectors_entry.attr,
	NULL,
};
ATTRIBUTE_GROUPS(blk_ia_range);

static ssize_t blk_ia_range_sysfs_show(struct kobject *kobj,
				      struct attribute *attr, char *buf)
{
	struct blk_ia_range_sysfs_entry *entry =
		container_of(attr, struct blk_ia_range_sysfs_entry, attr);
	struct blk_independent_access_range *iar =
		container_of(kobj, struct blk_independent_access_range, kobj);
	ssize_t ret;

	mutex_lock(&iar->queue->sysfs_lock);
	ret = entry->show(iar, buf);
	mutex_unlock(&iar->queue->sysfs_lock);

	return ret;
}

static const struct sysfs_ops blk_ia_range_sysfs_ops = {
	.show	= blk_ia_range_sysfs_show,
};

/*
 * Independent access range entries are not freed individually, but alltogether
 * with struct blk_independent_access_ranges and its array of ranges. Since
 * kobject_add() takes a reference on the parent kobject contained in
 * struct blk_independent_access_ranges, the array of independent access range
 * entries cannot be freed until kobject_del() is called for all entries.
 * So we do not need to do anything here, but still need this no-op release
 * operation to avoid complaints from the kobject code.
 */
static void blk_ia_range_sysfs_nop_release(struct kobject *kobj)
{
}

static struct kobj_type blk_ia_range_ktype = {
	.sysfs_ops	= &blk_ia_range_sysfs_ops,
	.default_groups	= blk_ia_range_groups,
	.release	= blk_ia_range_sysfs_nop_release,
};

/*
 * This will be executed only after all independent access range entries are
 * removed with kobject_del(), at which point, it is safe to free everything,
 * including the array of ranges.
 */
static void blk_ia_ranges_sysfs_release(struct kobject *kobj)
{
	struct blk_independent_access_ranges *iars =
		container_of(kobj, struct blk_independent_access_ranges, kobj);

	kfree(iars);
}

static struct kobj_type blk_ia_ranges_ktype = {
	.release	= blk_ia_ranges_sysfs_release,
};

/**
 * disk_register_independent_access_ranges - register with sysfs a set of
 *		independent access ranges
 * @disk:	Target disk
 * @new_iars:	New set of independent access ranges
 *
 * Register with sysfs a set of independent access ranges for @disk.
 * If @new_iars is not NULL, this set of ranges is registered and the old set
 * specified by q->ia_ranges is unregistered. Otherwise, q->ia_ranges is
 * registered if it is not already.
 */
int disk_register_independent_access_ranges(struct gendisk *disk,
				struct blk_independent_access_ranges *new_iars)
{
	struct request_queue *q = disk->queue;
	struct blk_independent_access_ranges *iars;
	int i, ret;

	lockdep_assert_held(&q->sysfs_dir_lock);
	lockdep_assert_held(&q->sysfs_lock);

	/* If a new range set is specified, unregister the old one */
	if (new_iars) {
		if (q->ia_ranges)
			disk_unregister_independent_access_ranges(disk);
		q->ia_ranges = new_iars;
	}

	iars = q->ia_ranges;
	if (!iars)
		return 0;

	/*
	 * At this point, iars is the new set of sector access ranges that needs
	 * to be registered with sysfs.
	 */
	WARN_ON(iars->sysfs_registered);
	ret = kobject_init_and_add(&iars->kobj, &blk_ia_ranges_ktype,
				   &q->kobj, "%s", "independent_access_ranges");
	if (ret) {
		q->ia_ranges = NULL;
		kobject_put(&iars->kobj);
		return ret;
	}

	for (i = 0; i < iars->nr_ia_ranges; i++) {
		iars->ia_range[i].queue = q;
		ret = kobject_init_and_add(&iars->ia_range[i].kobj,
					   &blk_ia_range_ktype, &iars->kobj,
					   "%d", i);
		if (ret) {
			while (--i >= 0)
				kobject_del(&iars->ia_range[i].kobj);
			kobject_del(&iars->kobj);
			kobject_put(&iars->kobj);
			return ret;
		}
	}

	iars->sysfs_registered = true;

	return 0;
}

void disk_unregister_independent_access_ranges(struct gendisk *disk)
{
	struct request_queue *q = disk->queue;
	struct blk_independent_access_ranges *iars = q->ia_ranges;
	int i;

	lockdep_assert_held(&q->sysfs_dir_lock);
	lockdep_assert_held(&q->sysfs_lock);

	if (!iars)
		return;

	if (iars->sysfs_registered) {
		for (i = 0; i < iars->nr_ia_ranges; i++)
			kobject_del(&iars->ia_range[i].kobj);
		kobject_del(&iars->kobj);
		kobject_put(&iars->kobj);
	} else {
		kfree(iars);
	}

	q->ia_ranges = NULL;
}

static struct blk_independent_access_range *
disk_find_ia_range(struct blk_independent_access_ranges *iars,
		  sector_t sector)
{
	struct blk_independent_access_range *iar;
	int i;

	for (i = 0; i < iars->nr_ia_ranges; i++) {
		iar = &iars->ia_range[i];
		if (sector >= iar->sector &&
		    sector < iar->sector + iar->nr_sectors)
			return iar;
	}

	return NULL;
}

static bool disk_check_ia_ranges(struct gendisk *disk,
				struct blk_independent_access_ranges *iars)
{
	struct blk_independent_access_range *iar, *tmp;
	sector_t capacity = get_capacity(disk);
	sector_t sector = 0;
	int i;

	/*
	 * While sorting the ranges in increasing LBA order, check that the
	 * ranges do not overlap, that there are no sector holes and that all
	 * sectors belong to one range.
	 */
	for (i = 0; i < iars->nr_ia_ranges; i++) {
		tmp = disk_find_ia_range(iars, sector);
		if (!tmp || tmp->sector != sector) {
			pr_warn("Invalid non-contiguous independent access ranges\n");
			return false;
		}

		iar = &iars->ia_range[i];
		if (tmp != iar) {
			swap(iar->sector, tmp->sector);
			swap(iar->nr_sectors, tmp->nr_sectors);
		}

		sector += iar->nr_sectors;
	}

	if (sector != capacity) {
		pr_warn("Independent access ranges do not match disk capacity\n");
		return false;
	}

	return true;
}

static bool disk_ia_ranges_changed(struct gendisk *disk,
				   struct blk_independent_access_ranges *new)
{
	struct blk_independent_access_ranges *old = disk->queue->ia_ranges;
	int i;

	if (!old)
		return true;

	if (old->nr_ia_ranges != new->nr_ia_ranges)
		return true;

	for (i = 0; i < old->nr_ia_ranges; i++) {
		if (new->ia_range[i].sector != old->ia_range[i].sector ||
		    new->ia_range[i].nr_sectors != old->ia_range[i].nr_sectors)
			return true;
	}

	return false;
}

/**
 * disk_alloc_independent_access_ranges - Allocate an independent access ranges
 *                                        data structure
 * @disk:		target disk
 * @nr_ia_ranges:	Number of independent access ranges
 *
 * Allocate a struct blk_independent_access_ranges structure with @nr_ia_ranges
 * access range descriptors.
 */
struct blk_independent_access_ranges *
disk_alloc_independent_access_ranges(struct gendisk *disk, int nr_ia_ranges)
{
	struct blk_independent_access_ranges *iars;

	iars = kzalloc_node(struct_size(iars, ia_range, nr_ia_ranges),
			    GFP_KERNEL, disk->queue->node);
	if (iars)
		iars->nr_ia_ranges = nr_ia_ranges;
	return iars;
}
EXPORT_SYMBOL_GPL(disk_alloc_independent_access_ranges);

/**
 * disk_set_independent_access_ranges - Set a disk independent access ranges
 * @disk:	target disk
 * @iars:	independent access ranges structure
 *
 * Set the independent access ranges information of the request queue
 * of @disk to @iars. If @iars is NULL and the independent access ranges
 * structure already set is cleared. If there are no differences between
 * @iars and the independent access ranges structure already set, @iars
 * is freed.
 */
void disk_set_independent_access_ranges(struct gendisk *disk,
				struct blk_independent_access_ranges *iars)
{
	struct request_queue *q = disk->queue;

	if (WARN_ON_ONCE(iars && !iars->nr_ia_ranges)) {
		kfree(iars);
		iars = NULL;
	}

	mutex_lock(&q->sysfs_dir_lock);
	mutex_lock(&q->sysfs_lock);

	if (iars) {
		if (!disk_check_ia_ranges(disk, iars)) {
			kfree(iars);
			iars = NULL;
			goto reg;
		}

		if (!disk_ia_ranges_changed(disk, iars)) {
			kfree(iars);
			goto unlock;
		}
	}

	/*
	 * This may be called for a registered queue. E.g. during a device
	 * revalidation. If that is the case, we need to unregister the old
	 * set of independent access ranges and register the new set. If the
	 * queue is not registered, registration of the device request queue
	 * will register the independent access ranges, so only swap in the
	 * new set and free the old one.
	 */
reg:
	if (blk_queue_registered(q)) {
		disk_register_independent_access_ranges(disk, iars);
	} else {
		swap(q->ia_ranges, iars);
		kfree(iars);
	}

unlock:
	mutex_unlock(&q->sysfs_lock);
	mutex_unlock(&q->sysfs_dir_lock);
}
EXPORT_SYMBOL_GPL(disk_set_independent_access_ranges);
Commit	Line	Data
a2247f19 DLM	1	// SPDX-License-Identifier: GPL-2.0
	2	/*
	3	* Block device concurrent positioning ranges.
	4	*
	5	* Copyright (C) 2021 Western Digital Corporation or its Affiliates.
	6	*/
	7	#include <linux/kernel.h>
	8	#include <linux/blkdev.h>
	9	#include <linux/slab.h>
	10	#include <linux/init.h>
	11
	12	#include "blk.h"
	13
	14	static ssize_t
	15	blk_ia_range_sector_show(struct blk_independent_access_range *iar,
	16	char *buf)
	17	{
	18	return sprintf(buf, "%llu\n", iar->sector);
	19	}
	20
	21	static ssize_t
	22	blk_ia_range_nr_sectors_show(struct blk_independent_access_range *iar,
	23	char *buf)
	24	{
	25	return sprintf(buf, "%llu\n", iar->nr_sectors);
	26	}
	27
	28	struct blk_ia_range_sysfs_entry {
	29	struct attribute attr;
	30	ssize_t (show)(struct blk_independent_access_range iar, char *buf);
	31	};
	32
	33	static struct blk_ia_range_sysfs_entry blk_ia_range_sector_entry = {
	34	.attr = { .name = "sector", .mode = 0444 },
	35	.show = blk_ia_range_sector_show,
	36	};
	37
	38	static struct blk_ia_range_sysfs_entry blk_ia_range_nr_sectors_entry = {
	39	.attr = { .name = "nr_sectors", .mode = 0444 },
	40	.show = blk_ia_range_nr_sectors_show,
	41	};
	42
	43	static struct attribute *blk_ia_range_attrs[] = {
	44	&blk_ia_range_sector_entry.attr,
	45	&blk_ia_range_nr_sectors_entry.attr,
	46	NULL,
	47	};
	48	ATTRIBUTE_GROUPS(blk_ia_range);
	49
	50	static ssize_t blk_ia_range_sysfs_show(struct kobject *kobj,
	51	struct attribute attr, char buf)
	52	{
	53	struct blk_ia_range_sysfs_entry *entry =
	54	container_of(attr, struct blk_ia_range_sysfs_entry, attr);
	55	struct blk_independent_access_range *iar =
	56	container_of(kobj, struct blk_independent_access_range, kobj);
	57	ssize_t ret;
	58
	59	mutex_lock(&iar->queue->sysfs_lock);
	60	ret = entry->show(iar, buf);
	61	mutex_unlock(&iar->queue->sysfs_lock);
	62
	63	return ret;
	64	}
65
66	static const struct sysfs_ops blk_ia_range_sysfs_ops = {
67	.show = blk_ia_range_sysfs_show,
68	};
69
70	/*
71	* Independent access range entries are not freed individually, but alltogether
72	* with struct blk_independent_access_ranges and its array of ranges. Since
73	* kobject_add() takes a reference on the parent kobject contained in
74	* struct blk_independent_access_ranges, the array of independent access range
75	* entries cannot be freed until kobject_del() is called for all entries.
76	* So we do not need to do anything here, but still need this no-op release
77	* operation to avoid complaints from the kobject code.
78	*/
79	static void blk_ia_range_sysfs_nop_release(struct kobject *kobj)
80	{
81	}
82
83	static struct kobj_type blk_ia_range_ktype = {
84	.sysfs_ops = &blk_ia_range_sysfs_ops,
85	.default_groups = blk_ia_range_groups,
86	.release = blk_ia_range_sysfs_nop_release,
87	};
88
89	/*
90	* This will be executed only after all independent access range entries are
91	* removed with kobject_del(), at which point, it is safe to free everything,
92	* including the array of ranges.
93	*/
94	static void blk_ia_ranges_sysfs_release(struct kobject *kobj)
95	{
96	struct blk_independent_access_ranges *iars =
97	container_of(kobj, struct blk_independent_access_ranges, kobj);
98
99	kfree(iars);
100	}
101
102	static struct kobj_type blk_ia_ranges_ktype = {
103	.release = blk_ia_ranges_sysfs_release,
104	};
105
106	/**
438cd742 JA	107	* disk_register_independent_access_ranges - register with sysfs a set of
438cd742 JA	108	* independent access ranges
a2247f19 DLM	109	* @disk: Target disk
	110	* @new_iars: New set of independent access ranges
	111	*
	112	* Register with sysfs a set of independent access ranges for @disk.
	113	* If @new_iars is not NULL, this set of ranges is registered and the old set
	114	* specified by q->ia_ranges is unregistered. Otherwise, q->ia_ranges is
	115	* registered if it is not already.
	116	*/
	117	int disk_register_independent_access_ranges(struct gendisk *disk,
	118	struct blk_independent_access_ranges *new_iars)
	119	{
	120	struct request_queue *q = disk->queue;
	121	struct blk_independent_access_ranges *iars;
	122	int i, ret;
	123
	124	lockdep_assert_held(&q->sysfs_dir_lock);
	125	lockdep_assert_held(&q->sysfs_lock);
	126
	127	/* If a new range set is specified, unregister the old one */
	128	if (new_iars) {
	129	if (q->ia_ranges)
	130	disk_unregister_independent_access_ranges(disk);
	131	q->ia_ranges = new_iars;
	132	}
	133
	134	iars = q->ia_ranges;
	135	if (!iars)
	136	return 0;
	137
	138	/*
	139	* At this point, iars is the new set of sector access ranges that needs
	140	* to be registered with sysfs.
	141	*/
	142	WARN_ON(iars->sysfs_registered);
	143	ret = kobject_init_and_add(&iars->kobj, &blk_ia_ranges_ktype,
	144	&q->kobj, "%s", "independent_access_ranges");
	145	if (ret) {
	146	q->ia_ranges = NULL;
83114df3	147	kobject_put(&iars->kobj);
a2247f19 DLM	148	return ret;
	149	}
	150
	151	for (i = 0; i < iars->nr_ia_ranges; i++) {
	152	iars->ia_range[i].queue = q;
	153	ret = kobject_init_and_add(&iars->ia_range[i].kobj,
	154	&blk_ia_range_ktype, &iars->kobj,
	155	"%d", i);
	156	if (ret) {
	157	while (--i >= 0)
	158	kobject_del(&iars->ia_range[i].kobj);
	159	kobject_del(&iars->kobj);
	160	kobject_put(&iars->kobj);
	161	return ret;
	162	}
	163	}
	164
	165	iars->sysfs_registered = true;
	166
	167	return 0;
	168	}
	169
	170	void disk_unregister_independent_access_ranges(struct gendisk *disk)
	171	{
	172	struct request_queue *q = disk->queue;
	173	struct blk_independent_access_ranges *iars = q->ia_ranges;
	174	int i;
	175
	176	lockdep_assert_held(&q->sysfs_dir_lock);
	177	lockdep_assert_held(&q->sysfs_lock);
	178
	179	if (!iars)
	180	return;
	181
	182	if (iars->sysfs_registered) {
	183	for (i = 0; i < iars->nr_ia_ranges; i++)
	184	kobject_del(&iars->ia_range[i].kobj);
	185	kobject_del(&iars->kobj);
	186	kobject_put(&iars->kobj);
	187	} else {
	188	kfree(iars);
	189	}
	190
	191	q->ia_ranges = NULL;
	192	}
	193
	194	static struct blk_independent_access_range *
	195	disk_find_ia_range(struct blk_independent_access_ranges *iars,
	196	sector_t sector)
	197	{
	198	struct blk_independent_access_range *iar;
	199	int i;
	200
	201	for (i = 0; i < iars->nr_ia_ranges; i++) {
	202	iar = &iars->ia_range[i];
	203	if (sector >= iar->sector &&
	204	sector < iar->sector + iar->nr_sectors)
	205	return iar;
	206	}
	207
	208	return NULL;
	209	}
	210
	211	static bool disk_check_ia_ranges(struct gendisk *disk,
212	struct blk_independent_access_ranges *iars)
213	{
214	struct blk_independent_access_range iar, tmp;
215	sector_t capacity = get_capacity(disk);
216	sector_t sector = 0;
217	int i;
218
219	/*
220	* While sorting the ranges in increasing LBA order, check that the
221	* ranges do not overlap, that there are no sector holes and that all
222	* sectors belong to one range.
223	*/
224	for (i = 0; i < iars->nr_ia_ranges; i++) {
225	tmp = disk_find_ia_range(iars, sector);
226	if (!tmp \|\| tmp->sector != sector) {
227	pr_warn("Invalid non-contiguous independent access ranges\n");
228	return false;
229	}
230
231	iar = &iars->ia_range[i];
232	if (tmp != iar) {
233	swap(iar->sector, tmp->sector);
234	swap(iar->nr_sectors, tmp->nr_sectors);
235	}
236
237	sector += iar->nr_sectors;
238	}
239
240	if (sector != capacity) {
241	pr_warn("Independent access ranges do not match disk capacity\n");
242	return false;
243	}
244
245	return true;
246	}
247
248	static bool disk_ia_ranges_changed(struct gendisk *disk,
249	struct blk_independent_access_ranges *new)
250	{
251	struct blk_independent_access_ranges *old = disk->queue->ia_ranges;
252	int i;
253
254	if (!old)
255	return true;
256
257	if (old->nr_ia_ranges != new->nr_ia_ranges)
258	return true;
259
260	for (i = 0; i < old->nr_ia_ranges; i++) {
261	if (new->ia_range[i].sector != old->ia_range[i].sector \|\|
262	new->ia_range[i].nr_sectors != old->ia_range[i].nr_sectors)
263	return true;
264	}
265
266	return false;
267	}
268
269	/**
270	* disk_alloc_independent_access_ranges - Allocate an independent access ranges
271	* data structure
272	* @disk: target disk
273	* @nr_ia_ranges: Number of independent access ranges
274	*
275	* Allocate a struct blk_independent_access_ranges structure with @nr_ia_ranges
276	* access range descriptors.
277	*/
278	struct blk_independent_access_ranges *
279	disk_alloc_independent_access_ranges(struct gendisk *disk, int nr_ia_ranges)
280	{
281	struct blk_independent_access_ranges *iars;
282
283	iars = kzalloc_node(struct_size(iars, ia_range, nr_ia_ranges),
284	GFP_KERNEL, disk->queue->node);
285	if (iars)
286	iars->nr_ia_ranges = nr_ia_ranges;
287	return iars;
288	}
289	EXPORT_SYMBOL_GPL(disk_alloc_independent_access_ranges);
290
291	/**
292	* disk_set_independent_access_ranges - Set a disk independent access ranges
293	* @disk: target disk
294	* @iars: independent access ranges structure
295	*
296	* Set the independent access ranges information of the request queue
297	* of @disk to @iars. If @iars is NULL and the independent access ranges
298	* structure already set is cleared. If there are no differences between
299	* @iars and the independent access ranges structure already set, @iars
300	* is freed.
301	*/
302	void disk_set_independent_access_ranges(struct gendisk *disk,
303	struct blk_independent_access_ranges *iars)
304	{
305	struct request_queue *q = disk->queue;
306
307	if (WARN_ON_ONCE(iars && !iars->nr_ia_ranges)) {
308	kfree(iars);
309	iars = NULL;
310	}
311
312	mutex_lock(&q->sysfs_dir_lock);
313	mutex_lock(&q->sysfs_lock);
314
315	if (iars) {
316	if (!disk_check_ia_ranges(disk, iars)) {
317	kfree(iars);
318	iars = NULL;
319	goto reg;
320	}
321
322	if (!disk_ia_ranges_changed(disk, iars)) {
323	kfree(iars);
324	goto unlock;
325	}
326	}
327
328	/*
329	* This may be called for a registered queue. E.g. during a device
330	* revalidation. If that is the case, we need to unregister the old
331	* set of independent access ranges and register the new set. If the
332	* queue is not registered, registration of the device request queue
333	* will register the independent access ranges, so only swap in the
334	* new set and free the old one.
335	*/
336	reg:
337	if (blk_queue_registered(q)) {
338	disk_register_independent_access_ranges(disk, iars);
339	} else {
340	swap(q->ia_ranges, iars);
341	kfree(iars);
342	}
343
344	unlock:
345	mutex_unlock(&q->sysfs_lock);
346	mutex_unlock(&q->sysfs_dir_lock);
347	}
348	EXPORT_SYMBOL_GPL(disk_set_independent_access_ranges);