Revision 8e7f82deb0c0386a03b62e30082574347f8b57d5 authored by Filipe Manana on 12 September 2023, 10:45:39 UTC, committed by David Sterba on 14 September 2023, 21:24:42 UTC
When opening a directory (opendir(3)) or rewinding it (rewinddir(3)), we
are not holding the directory's inode locked, and this can result in later
attempting to add two entries to the directory with the same index number,
resulting in a transaction abort, with -EEXIST (-17), when inserting the
second delayed dir index. This results in a trace like the following:

  Sep 11 22:34:59 myhostname kernel: BTRFS error (device dm-3): err add delayed dir index item(name: cockroach-stderr.log) into the insertion tree of the delayed node(root id: 5, inode id: 4539217, errno: -17)
  Sep 11 22:34:59 myhostname kernel: ------------[ cut here ]------------
  Sep 11 22:34:59 myhostname kernel: kernel BUG at fs/btrfs/delayed-inode.c:1504!
  Sep 11 22:34:59 myhostname kernel: invalid opcode: 0000 [#1] PREEMPT SMP NOPTI
  Sep 11 22:34:59 myhostname kernel: CPU: 0 PID: 7159 Comm: cockroach Not tainted 6.4.15-200.fc38.x86_64 #1
  Sep 11 22:34:59 myhostname kernel: Hardware name: ASUS ESC500 G3/P9D WS, BIOS 2402 06/27/2018
  Sep 11 22:34:59 myhostname kernel: RIP: 0010:btrfs_insert_delayed_dir_index+0x1da/0x260
  Sep 11 22:34:59 myhostname kernel: Code: eb dd 48 (...)
  Sep 11 22:34:59 myhostname kernel: RSP: 0000:ffffa9980e0fbb28 EFLAGS: 00010282
  Sep 11 22:34:59 myhostname kernel: RAX: 0000000000000000 RBX: ffff8b10b8f4a3c0 RCX: 0000000000000000
  Sep 11 22:34:59 myhostname kernel: RDX: 0000000000000000 RSI: ffff8b177ec21540 RDI: ffff8b177ec21540
  Sep 11 22:34:59 myhostname kernel: RBP: ffff8b110cf80888 R08: 0000000000000000 R09: ffffa9980e0fb938
  Sep 11 22:34:59 myhostname kernel: R10: 0000000000000003 R11: ffffffff86146508 R12: 0000000000000014
  Sep 11 22:34:59 myhostname kernel: R13: ffff8b1131ae5b40 R14: ffff8b10b8f4a418 R15: 00000000ffffffef
  Sep 11 22:34:59 myhostname kernel: FS:  00007fb14a7fe6c0(0000) GS:ffff8b177ec00000(0000) knlGS:0000000000000000
  Sep 11 22:34:59 myhostname kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  Sep 11 22:34:59 myhostname kernel: CR2: 000000c00143d000 CR3: 00000001b3b4e002 CR4: 00000000001706f0
  Sep 11 22:34:59 myhostname kernel: Call Trace:
  Sep 11 22:34:59 myhostname kernel:  <TASK>
  Sep 11 22:34:59 myhostname kernel:  ? die+0x36/0x90
  Sep 11 22:34:59 myhostname kernel:  ? do_trap+0xda/0x100
  Sep 11 22:34:59 myhostname kernel:  ? btrfs_insert_delayed_dir_index+0x1da/0x260
  Sep 11 22:34:59 myhostname kernel:  ? do_error_trap+0x6a/0x90
  Sep 11 22:34:59 myhostname kernel:  ? btrfs_insert_delayed_dir_index+0x1da/0x260
  Sep 11 22:34:59 myhostname kernel:  ? exc_invalid_op+0x50/0x70
  Sep 11 22:34:59 myhostname kernel:  ? btrfs_insert_delayed_dir_index+0x1da/0x260
  Sep 11 22:34:59 myhostname kernel:  ? asm_exc_invalid_op+0x1a/0x20
  Sep 11 22:34:59 myhostname kernel:  ? btrfs_insert_delayed_dir_index+0x1da/0x260
  Sep 11 22:34:59 myhostname kernel:  ? btrfs_insert_delayed_dir_index+0x1da/0x260
  Sep 11 22:34:59 myhostname kernel:  btrfs_insert_dir_item+0x200/0x280
  Sep 11 22:34:59 myhostname kernel:  btrfs_add_link+0xab/0x4f0
  Sep 11 22:34:59 myhostname kernel:  ? ktime_get_real_ts64+0x47/0xe0
  Sep 11 22:34:59 myhostname kernel:  btrfs_create_new_inode+0x7cd/0xa80
  Sep 11 22:34:59 myhostname kernel:  btrfs_symlink+0x190/0x4d0
  Sep 11 22:34:59 myhostname kernel:  ? schedule+0x5e/0xd0
  Sep 11 22:34:59 myhostname kernel:  ? __d_lookup+0x7e/0xc0
  Sep 11 22:34:59 myhostname kernel:  vfs_symlink+0x148/0x1e0
  Sep 11 22:34:59 myhostname kernel:  do_symlinkat+0x130/0x140
  Sep 11 22:34:59 myhostname kernel:  __x64_sys_symlinkat+0x3d/0x50
  Sep 11 22:34:59 myhostname kernel:  do_syscall_64+0x5d/0x90
  Sep 11 22:34:59 myhostname kernel:  ? syscall_exit_to_user_mode+0x2b/0x40
  Sep 11 22:34:59 myhostname kernel:  ? do_syscall_64+0x6c/0x90
  Sep 11 22:34:59 myhostname kernel:  entry_SYSCALL_64_after_hwframe+0x72/0xdc

The race leading to the problem happens like this:

1) Directory inode X is loaded into memory, its ->index_cnt field is
   initialized to (u64)-1 (at btrfs_alloc_inode());

2) Task A is adding a new file to directory X, holding its vfs inode lock,
   and calls btrfs_set_inode_index() to get an index number for the entry.

   Because the inode's index_cnt field is set to (u64)-1 it calls
   btrfs_inode_delayed_dir_index_count() which fails because no dir index
   entries were added yet to the delayed inode and then it calls
   btrfs_set_inode_index_count(). This functions finds the last dir index
   key and then sets index_cnt to that index value + 1. It found that the
   last index key has an offset of 100. However before it assigns a value
   of 101 to index_cnt...

3) Task B calls opendir(3), ending up at btrfs_opendir(), where the VFS
   lock for inode X is not taken, so it calls btrfs_get_dir_last_index()
   and sees index_cnt still with a value of (u64)-1. Because of that it
   calls btrfs_inode_delayed_dir_index_count() which fails since no dir
   index entries were added to the delayed inode yet, and then it also
   calls btrfs_set_inode_index_count(). This also finds that the last
   index key has an offset of 100, and before it assigns the value 101
   to the index_cnt field of inode X...

4) Task A assigns a value of 101 to index_cnt. And then the code flow
   goes to btrfs_set_inode_index() where it increments index_cnt from
   101 to 102. Task A then creates a delayed dir index entry with a
   sequence number of 101 and adds it to the delayed inode;

5) Task B assigns 101 to the index_cnt field of inode X;

6) At some later point when someone tries to add a new entry to the
   directory, btrfs_set_inode_index() will return 101 again and shortly
   after an attempt to add another delayed dir index key with index
   number 101 will fail with -EEXIST resulting in a transaction abort.

Fix this by locking the inode at btrfs_get_dir_last_index(), which is only
only used when opening a directory or attempting to lseek on it.

Reported-by: ken <ken@bllue.org>
Link: https://lore.kernel.org/linux-btrfs/CAE6xmH+Lp=Q=E61bU+v9eWX8gYfLvu6jLYxjxjFpo3zHVPR0EQ@mail.gmail.com/
Reported-by: syzbot+d13490c82ad5353c779d@syzkaller.appspotmail.com
Link: https://lore.kernel.org/linux-btrfs/00000000000036e1290603e097e0@google.com/
Fixes: 9b378f6ad48c ("btrfs: fix infinite directory reads")
CC: stable@vger.kernel.org # 6.5+
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
1 parent e60aa5d
Raw File
idr.c
// SPDX-License-Identifier: GPL-2.0-only
#include <linux/bitmap.h>
#include <linux/bug.h>
#include <linux/export.h>
#include <linux/idr.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/xarray.h>

/**
 * idr_alloc_u32() - Allocate an ID.
 * @idr: IDR handle.
 * @ptr: Pointer to be associated with the new ID.
 * @nextid: Pointer to an ID.
 * @max: The maximum ID to allocate (inclusive).
 * @gfp: Memory allocation flags.
 *
 * Allocates an unused ID in the range specified by @nextid and @max.
 * Note that @max is inclusive whereas the @end parameter to idr_alloc()
 * is exclusive.  The new ID is assigned to @nextid before the pointer
 * is inserted into the IDR, so if @nextid points into the object pointed
 * to by @ptr, a concurrent lookup will not find an uninitialised ID.
 *
 * The caller should provide their own locking to ensure that two
 * concurrent modifications to the IDR are not possible.  Read-only
 * accesses to the IDR may be done under the RCU read lock or may
 * exclude simultaneous writers.
 *
 * Return: 0 if an ID was allocated, -ENOMEM if memory allocation failed,
 * or -ENOSPC if no free IDs could be found.  If an error occurred,
 * @nextid is unchanged.
 */
int idr_alloc_u32(struct idr *idr, void *ptr, u32 *nextid,
			unsigned long max, gfp_t gfp)
{
	struct radix_tree_iter iter;
	void __rcu **slot;
	unsigned int base = idr->idr_base;
	unsigned int id = *nextid;

	if (WARN_ON_ONCE(!(idr->idr_rt.xa_flags & ROOT_IS_IDR)))
		idr->idr_rt.xa_flags |= IDR_RT_MARKER;

	id = (id < base) ? 0 : id - base;
	radix_tree_iter_init(&iter, id);
	slot = idr_get_free(&idr->idr_rt, &iter, gfp, max - base);
	if (IS_ERR(slot))
		return PTR_ERR(slot);

	*nextid = iter.index + base;
	/* there is a memory barrier inside radix_tree_iter_replace() */
	radix_tree_iter_replace(&idr->idr_rt, &iter, slot, ptr);
	radix_tree_iter_tag_clear(&idr->idr_rt, &iter, IDR_FREE);

	return 0;
}
EXPORT_SYMBOL_GPL(idr_alloc_u32);

/**
 * idr_alloc() - Allocate an ID.
 * @idr: IDR handle.
 * @ptr: Pointer to be associated with the new ID.
 * @start: The minimum ID (inclusive).
 * @end: The maximum ID (exclusive).
 * @gfp: Memory allocation flags.
 *
 * Allocates an unused ID in the range specified by @start and @end.  If
 * @end is <= 0, it is treated as one larger than %INT_MAX.  This allows
 * callers to use @start + N as @end as long as N is within integer range.
 *
 * The caller should provide their own locking to ensure that two
 * concurrent modifications to the IDR are not possible.  Read-only
 * accesses to the IDR may be done under the RCU read lock or may
 * exclude simultaneous writers.
 *
 * Return: The newly allocated ID, -ENOMEM if memory allocation failed,
 * or -ENOSPC if no free IDs could be found.
 */
int idr_alloc(struct idr *idr, void *ptr, int start, int end, gfp_t gfp)
{
	u32 id = start;
	int ret;

	if (WARN_ON_ONCE(start < 0))
		return -EINVAL;

	ret = idr_alloc_u32(idr, ptr, &id, end > 0 ? end - 1 : INT_MAX, gfp);
	if (ret)
		return ret;

	return id;
}
EXPORT_SYMBOL_GPL(idr_alloc);

/**
 * idr_alloc_cyclic() - Allocate an ID cyclically.
 * @idr: IDR handle.
 * @ptr: Pointer to be associated with the new ID.
 * @start: The minimum ID (inclusive).
 * @end: The maximum ID (exclusive).
 * @gfp: Memory allocation flags.
 *
 * Allocates an unused ID in the range specified by @nextid and @end.  If
 * @end is <= 0, it is treated as one larger than %INT_MAX.  This allows
 * callers to use @start + N as @end as long as N is within integer range.
 * The search for an unused ID will start at the last ID allocated and will
 * wrap around to @start if no free IDs are found before reaching @end.
 *
 * The caller should provide their own locking to ensure that two
 * concurrent modifications to the IDR are not possible.  Read-only
 * accesses to the IDR may be done under the RCU read lock or may
 * exclude simultaneous writers.
 *
 * Return: The newly allocated ID, -ENOMEM if memory allocation failed,
 * or -ENOSPC if no free IDs could be found.
 */
int idr_alloc_cyclic(struct idr *idr, void *ptr, int start, int end, gfp_t gfp)
{
	u32 id = idr->idr_next;
	int err, max = end > 0 ? end - 1 : INT_MAX;

	if ((int)id < start)
		id = start;

	err = idr_alloc_u32(idr, ptr, &id, max, gfp);
	if ((err == -ENOSPC) && (id > start)) {
		id = start;
		err = idr_alloc_u32(idr, ptr, &id, max, gfp);
	}
	if (err)
		return err;

	idr->idr_next = id + 1;
	return id;
}
EXPORT_SYMBOL(idr_alloc_cyclic);

/**
 * idr_remove() - Remove an ID from the IDR.
 * @idr: IDR handle.
 * @id: Pointer ID.
 *
 * Removes this ID from the IDR.  If the ID was not previously in the IDR,
 * this function returns %NULL.
 *
 * Since this function modifies the IDR, the caller should provide their
 * own locking to ensure that concurrent modification of the same IDR is
 * not possible.
 *
 * Return: The pointer formerly associated with this ID.
 */
void *idr_remove(struct idr *idr, unsigned long id)
{
	return radix_tree_delete_item(&idr->idr_rt, id - idr->idr_base, NULL);
}
EXPORT_SYMBOL_GPL(idr_remove);

/**
 * idr_find() - Return pointer for given ID.
 * @idr: IDR handle.
 * @id: Pointer ID.
 *
 * Looks up the pointer associated with this ID.  A %NULL pointer may
 * indicate that @id is not allocated or that the %NULL pointer was
 * associated with this ID.
 *
 * This function can be called under rcu_read_lock(), given that the leaf
 * pointers lifetimes are correctly managed.
 *
 * Return: The pointer associated with this ID.
 */
void *idr_find(const struct idr *idr, unsigned long id)
{
	return radix_tree_lookup(&idr->idr_rt, id - idr->idr_base);
}
EXPORT_SYMBOL_GPL(idr_find);

/**
 * idr_for_each() - Iterate through all stored pointers.
 * @idr: IDR handle.
 * @fn: Function to be called for each pointer.
 * @data: Data passed to callback function.
 *
 * The callback function will be called for each entry in @idr, passing
 * the ID, the entry and @data.
 *
 * If @fn returns anything other than %0, the iteration stops and that
 * value is returned from this function.
 *
 * idr_for_each() can be called concurrently with idr_alloc() and
 * idr_remove() if protected by RCU.  Newly added entries may not be
 * seen and deleted entries may be seen, but adding and removing entries
 * will not cause other entries to be skipped, nor spurious ones to be seen.
 */
int idr_for_each(const struct idr *idr,
		int (*fn)(int id, void *p, void *data), void *data)
{
	struct radix_tree_iter iter;
	void __rcu **slot;
	int base = idr->idr_base;

	radix_tree_for_each_slot(slot, &idr->idr_rt, &iter, 0) {
		int ret;
		unsigned long id = iter.index + base;

		if (WARN_ON_ONCE(id > INT_MAX))
			break;
		ret = fn(id, rcu_dereference_raw(*slot), data);
		if (ret)
			return ret;
	}

	return 0;
}
EXPORT_SYMBOL(idr_for_each);

/**
 * idr_get_next_ul() - Find next populated entry.
 * @idr: IDR handle.
 * @nextid: Pointer to an ID.
 *
 * Returns the next populated entry in the tree with an ID greater than
 * or equal to the value pointed to by @nextid.  On exit, @nextid is updated
 * to the ID of the found value.  To use in a loop, the value pointed to by
 * nextid must be incremented by the user.
 */
void *idr_get_next_ul(struct idr *idr, unsigned long *nextid)
{
	struct radix_tree_iter iter;
	void __rcu **slot;
	void *entry = NULL;
	unsigned long base = idr->idr_base;
	unsigned long id = *nextid;

	id = (id < base) ? 0 : id - base;
	radix_tree_for_each_slot(slot, &idr->idr_rt, &iter, id) {
		entry = rcu_dereference_raw(*slot);
		if (!entry)
			continue;
		if (!xa_is_internal(entry))
			break;
		if (slot != &idr->idr_rt.xa_head && !xa_is_retry(entry))
			break;
		slot = radix_tree_iter_retry(&iter);
	}
	if (!slot)
		return NULL;

	*nextid = iter.index + base;
	return entry;
}
EXPORT_SYMBOL(idr_get_next_ul);

/**
 * idr_get_next() - Find next populated entry.
 * @idr: IDR handle.
 * @nextid: Pointer to an ID.
 *
 * Returns the next populated entry in the tree with an ID greater than
 * or equal to the value pointed to by @nextid.  On exit, @nextid is updated
 * to the ID of the found value.  To use in a loop, the value pointed to by
 * nextid must be incremented by the user.
 */
void *idr_get_next(struct idr *idr, int *nextid)
{
	unsigned long id = *nextid;
	void *entry = idr_get_next_ul(idr, &id);

	if (WARN_ON_ONCE(id > INT_MAX))
		return NULL;
	*nextid = id;
	return entry;
}
EXPORT_SYMBOL(idr_get_next);

/**
 * idr_replace() - replace pointer for given ID.
 * @idr: IDR handle.
 * @ptr: New pointer to associate with the ID.
 * @id: ID to change.
 *
 * Replace the pointer registered with an ID and return the old value.
 * This function can be called under the RCU read lock concurrently with
 * idr_alloc() and idr_remove() (as long as the ID being removed is not
 * the one being replaced!).
 *
 * Returns: the old value on success.  %-ENOENT indicates that @id was not
 * found.  %-EINVAL indicates that @ptr was not valid.
 */
void *idr_replace(struct idr *idr, void *ptr, unsigned long id)
{
	struct radix_tree_node *node;
	void __rcu **slot = NULL;
	void *entry;

	id -= idr->idr_base;

	entry = __radix_tree_lookup(&idr->idr_rt, id, &node, &slot);
	if (!slot || radix_tree_tag_get(&idr->idr_rt, id, IDR_FREE))
		return ERR_PTR(-ENOENT);

	__radix_tree_replace(&idr->idr_rt, node, slot, ptr);

	return entry;
}
EXPORT_SYMBOL(idr_replace);

/**
 * DOC: IDA description
 *
 * The IDA is an ID allocator which does not provide the ability to
 * associate an ID with a pointer.  As such, it only needs to store one
 * bit per ID, and so is more space efficient than an IDR.  To use an IDA,
 * define it using DEFINE_IDA() (or embed a &struct ida in a data structure,
 * then initialise it using ida_init()).  To allocate a new ID, call
 * ida_alloc(), ida_alloc_min(), ida_alloc_max() or ida_alloc_range().
 * To free an ID, call ida_free().
 *
 * ida_destroy() can be used to dispose of an IDA without needing to
 * free the individual IDs in it.  You can use ida_is_empty() to find
 * out whether the IDA has any IDs currently allocated.
 *
 * The IDA handles its own locking.  It is safe to call any of the IDA
 * functions without synchronisation in your code.
 *
 * IDs are currently limited to the range [0-INT_MAX].  If this is an awkward
 * limitation, it should be quite straightforward to raise the maximum.
 */

/*
 * Developer's notes:
 *
 * The IDA uses the functionality provided by the XArray to store bitmaps in
 * each entry.  The XA_FREE_MARK is only cleared when all bits in the bitmap
 * have been set.
 *
 * I considered telling the XArray that each slot is an order-10 node
 * and indexing by bit number, but the XArray can't allow a single multi-index
 * entry in the head, which would significantly increase memory consumption
 * for the IDA.  So instead we divide the index by the number of bits in the
 * leaf bitmap before doing a radix tree lookup.
 *
 * As an optimisation, if there are only a few low bits set in any given
 * leaf, instead of allocating a 128-byte bitmap, we store the bits
 * as a value entry.  Value entries never have the XA_FREE_MARK cleared
 * because we can always convert them into a bitmap entry.
 *
 * It would be possible to optimise further; once we've run out of a
 * single 128-byte bitmap, we currently switch to a 576-byte node, put
 * the 128-byte bitmap in the first entry and then start allocating extra
 * 128-byte entries.  We could instead use the 512 bytes of the node's
 * data as a bitmap before moving to that scheme.  I do not believe this
 * is a worthwhile optimisation; Rasmus Villemoes surveyed the current
 * users of the IDA and almost none of them use more than 1024 entries.
 * Those that do use more than the 8192 IDs that the 512 bytes would
 * provide.
 *
 * The IDA always uses a lock to alloc/free.  If we add a 'test_bit'
 * equivalent, it will still need locking.  Going to RCU lookup would require
 * using RCU to free bitmaps, and that's not trivial without embedding an
 * RCU head in the bitmap, which adds a 2-pointer overhead to each 128-byte
 * bitmap, which is excessive.
 */

/**
 * ida_alloc_range() - Allocate an unused ID.
 * @ida: IDA handle.
 * @min: Lowest ID to allocate.
 * @max: Highest ID to allocate.
 * @gfp: Memory allocation flags.
 *
 * Allocate an ID between @min and @max, inclusive.  The allocated ID will
 * not exceed %INT_MAX, even if @max is larger.
 *
 * Context: Any context. It is safe to call this function without
 * locking in your code.
 * Return: The allocated ID, or %-ENOMEM if memory could not be allocated,
 * or %-ENOSPC if there are no free IDs.
 */
int ida_alloc_range(struct ida *ida, unsigned int min, unsigned int max,
			gfp_t gfp)
{
	XA_STATE(xas, &ida->xa, min / IDA_BITMAP_BITS);
	unsigned bit = min % IDA_BITMAP_BITS;
	unsigned long flags;
	struct ida_bitmap *bitmap, *alloc = NULL;

	if ((int)min < 0)
		return -ENOSPC;

	if ((int)max < 0)
		max = INT_MAX;

retry:
	xas_lock_irqsave(&xas, flags);
next:
	bitmap = xas_find_marked(&xas, max / IDA_BITMAP_BITS, XA_FREE_MARK);
	if (xas.xa_index > min / IDA_BITMAP_BITS)
		bit = 0;
	if (xas.xa_index * IDA_BITMAP_BITS + bit > max)
		goto nospc;

	if (xa_is_value(bitmap)) {
		unsigned long tmp = xa_to_value(bitmap);

		if (bit < BITS_PER_XA_VALUE) {
			bit = find_next_zero_bit(&tmp, BITS_PER_XA_VALUE, bit);
			if (xas.xa_index * IDA_BITMAP_BITS + bit > max)
				goto nospc;
			if (bit < BITS_PER_XA_VALUE) {
				tmp |= 1UL << bit;
				xas_store(&xas, xa_mk_value(tmp));
				goto out;
			}
		}
		bitmap = alloc;
		if (!bitmap)
			bitmap = kzalloc(sizeof(*bitmap), GFP_NOWAIT);
		if (!bitmap)
			goto alloc;
		bitmap->bitmap[0] = tmp;
		xas_store(&xas, bitmap);
		if (xas_error(&xas)) {
			bitmap->bitmap[0] = 0;
			goto out;
		}
	}

	if (bitmap) {
		bit = find_next_zero_bit(bitmap->bitmap, IDA_BITMAP_BITS, bit);
		if (xas.xa_index * IDA_BITMAP_BITS + bit > max)
			goto nospc;
		if (bit == IDA_BITMAP_BITS)
			goto next;

		__set_bit(bit, bitmap->bitmap);
		if (bitmap_full(bitmap->bitmap, IDA_BITMAP_BITS))
			xas_clear_mark(&xas, XA_FREE_MARK);
	} else {
		if (bit < BITS_PER_XA_VALUE) {
			bitmap = xa_mk_value(1UL << bit);
		} else {
			bitmap = alloc;
			if (!bitmap)
				bitmap = kzalloc(sizeof(*bitmap), GFP_NOWAIT);
			if (!bitmap)
				goto alloc;
			__set_bit(bit, bitmap->bitmap);
		}
		xas_store(&xas, bitmap);
	}
out:
	xas_unlock_irqrestore(&xas, flags);
	if (xas_nomem(&xas, gfp)) {
		xas.xa_index = min / IDA_BITMAP_BITS;
		bit = min % IDA_BITMAP_BITS;
		goto retry;
	}
	if (bitmap != alloc)
		kfree(alloc);
	if (xas_error(&xas))
		return xas_error(&xas);
	return xas.xa_index * IDA_BITMAP_BITS + bit;
alloc:
	xas_unlock_irqrestore(&xas, flags);
	alloc = kzalloc(sizeof(*bitmap), gfp);
	if (!alloc)
		return -ENOMEM;
	xas_set(&xas, min / IDA_BITMAP_BITS);
	bit = min % IDA_BITMAP_BITS;
	goto retry;
nospc:
	xas_unlock_irqrestore(&xas, flags);
	kfree(alloc);
	return -ENOSPC;
}
EXPORT_SYMBOL(ida_alloc_range);

/**
 * ida_free() - Release an allocated ID.
 * @ida: IDA handle.
 * @id: Previously allocated ID.
 *
 * Context: Any context. It is safe to call this function without
 * locking in your code.
 */
void ida_free(struct ida *ida, unsigned int id)
{
	XA_STATE(xas, &ida->xa, id / IDA_BITMAP_BITS);
	unsigned bit = id % IDA_BITMAP_BITS;
	struct ida_bitmap *bitmap;
	unsigned long flags;

	if ((int)id < 0)
		return;

	xas_lock_irqsave(&xas, flags);
	bitmap = xas_load(&xas);

	if (xa_is_value(bitmap)) {
		unsigned long v = xa_to_value(bitmap);
		if (bit >= BITS_PER_XA_VALUE)
			goto err;
		if (!(v & (1UL << bit)))
			goto err;
		v &= ~(1UL << bit);
		if (!v)
			goto delete;
		xas_store(&xas, xa_mk_value(v));
	} else {
		if (!test_bit(bit, bitmap->bitmap))
			goto err;
		__clear_bit(bit, bitmap->bitmap);
		xas_set_mark(&xas, XA_FREE_MARK);
		if (bitmap_empty(bitmap->bitmap, IDA_BITMAP_BITS)) {
			kfree(bitmap);
delete:
			xas_store(&xas, NULL);
		}
	}
	xas_unlock_irqrestore(&xas, flags);
	return;
 err:
	xas_unlock_irqrestore(&xas, flags);
	WARN(1, "ida_free called for id=%d which is not allocated.\n", id);
}
EXPORT_SYMBOL(ida_free);

/**
 * ida_destroy() - Free all IDs.
 * @ida: IDA handle.
 *
 * Calling this function frees all IDs and releases all resources used
 * by an IDA.  When this call returns, the IDA is empty and can be reused
 * or freed.  If the IDA is already empty, there is no need to call this
 * function.
 *
 * Context: Any context. It is safe to call this function without
 * locking in your code.
 */
void ida_destroy(struct ida *ida)
{
	XA_STATE(xas, &ida->xa, 0);
	struct ida_bitmap *bitmap;
	unsigned long flags;

	xas_lock_irqsave(&xas, flags);
	xas_for_each(&xas, bitmap, ULONG_MAX) {
		if (!xa_is_value(bitmap))
			kfree(bitmap);
		xas_store(&xas, NULL);
	}
	xas_unlock_irqrestore(&xas, flags);
}
EXPORT_SYMBOL(ida_destroy);

#ifndef __KERNEL__
extern void xa_dump_index(unsigned long index, unsigned int shift);
#define IDA_CHUNK_SHIFT		ilog2(IDA_BITMAP_BITS)

static void ida_dump_entry(void *entry, unsigned long index)
{
	unsigned long i;

	if (!entry)
		return;

	if (xa_is_node(entry)) {
		struct xa_node *node = xa_to_node(entry);
		unsigned int shift = node->shift + IDA_CHUNK_SHIFT +
			XA_CHUNK_SHIFT;

		xa_dump_index(index * IDA_BITMAP_BITS, shift);
		xa_dump_node(node);
		for (i = 0; i < XA_CHUNK_SIZE; i++)
			ida_dump_entry(node->slots[i],
					index | (i << node->shift));
	} else if (xa_is_value(entry)) {
		xa_dump_index(index * IDA_BITMAP_BITS, ilog2(BITS_PER_LONG));
		pr_cont("value: data %lx [%px]\n", xa_to_value(entry), entry);
	} else {
		struct ida_bitmap *bitmap = entry;

		xa_dump_index(index * IDA_BITMAP_BITS, IDA_CHUNK_SHIFT);
		pr_cont("bitmap: %p data", bitmap);
		for (i = 0; i < IDA_BITMAP_LONGS; i++)
			pr_cont(" %lx", bitmap->bitmap[i]);
		pr_cont("\n");
	}
}

static void ida_dump(struct ida *ida)
{
	struct xarray *xa = &ida->xa;
	pr_debug("ida: %p node %p free %d\n", ida, xa->xa_head,
				xa->xa_flags >> ROOT_TAG_SHIFT);
	ida_dump_entry(xa->xa_head, 0);
}
#endif
back to top