Files
mirror_zfs/module/zfs/zfs_rlock.c
T

693 lines
20 KiB
C
Raw Normal View History

2025-01-04 11:04:27 +11:00
// SPDX-License-Identifier: CDDL-1.0
2008-11-20 12:01:55 -08:00
/*
* CDDL HEADER START
*
* The contents of this file are subject to the terms of the
* Common Development and Distribution License (the "License").
* You may not use this file except in compliance with the License.
*
* You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
2022-07-11 23:16:13 +02:00
* or https://opensource.org/licenses/CDDL-1.0.
2008-11-20 12:01:55 -08:00
* See the License for the specific language governing permissions
* and limitations under the License.
*
* When distributing Covered Code, include this CDDL HEADER in each
* file and include the License file at usr/src/OPENSOLARIS.LICENSE.
* If applicable, add the following below this CDDL HEADER, with the
* fields enclosed by brackets "[]" replaced with your own identifying
* information: Portions Copyright [yyyy] [name of copyright owner]
*
* CDDL HEADER END
*/
/*
* Copyright 2010 Sun Microsystems, Inc. All rights reserved.
2008-11-20 12:01:55 -08:00
* Use is subject to license terms.
*/
2013-05-10 14:17:03 -07:00
/*
* Copyright (c) 2012, 2018 by Delphix. All rights reserved.
2013-05-10 14:17:03 -07:00
*/
2008-11-20 12:01:55 -08:00
/*
* This file contains the code to implement file range locking in
2013-06-11 09:12:34 -08:00
* ZFS, although there isn't much specific to ZFS (all that comes to mind is
2008-11-20 12:01:55 -08:00
* support for growing the blocksize).
*
* Interface
* ---------
* Defined in zfs_rlock.h but essentially:
* lr = rangelock_enter(zp, off, len, lock_type);
* rangelock_reduce(lr, off, len); // optional
* rangelock_exit(lr);
2008-11-20 12:01:55 -08:00
*
* Range locking rules
* --------------------
* 1. When truncating a file (zfs_create, zfs_setattr, zfs_space) the whole
* file range needs to be locked as RL_WRITER. Only then can the pages be
* freed etc and zp_size reset. zp_size must be set within range lock.
* 2. For writes and punching holes (zfs_write & zfs_space) just the range
* being written or freed needs to be locked as RL_WRITER.
* Multiple writes at the end of the file must coordinate zp_size updates
* to ensure data isn't lost. A compare and swap loop is currently used
* to ensure the file size is at least the offset last written.
* 3. For reads (zfs_read, zfs_get_data & zfs_putapage) just the range being
* read needs to be locked as RL_READER. A check against zp_size can then
* be made for reading beyond end of file.
*
2008-11-20 12:01:55 -08:00
* AVL tree
* --------
* An AVL tree is used to maintain the state of the existing ranges
* that are locked for exclusive (writer) or shared (reader) use.
* The starting range offset is used for searching and sorting the tree.
*
* Common case
* -----------
* The (hopefully) usual case is of no overlaps or contention for locks. On
* entry to rangelock_enter(), a locked_range_t is allocated; the tree
* searched that finds no overlap, and *this* locked_range_t is placed in the
* tree.
2008-11-20 12:01:55 -08:00
*
* Overlaps/Reference counting/Proxy locks
* ---------------------------------------
* The avl code only allows one node at a particular offset. Also it's very
* inefficient to search through all previous entries looking for overlaps
* (because the very 1st in the ordered list might be at offset 0 but
* cover the whole file).
* So this implementation uses reference counts and proxy range locks.
* Firstly, only reader locks use reference counts and proxy locks,
* because writer locks are exclusive.
* When a reader lock overlaps with another then a proxy lock is created
* for that range and replaces the original lock. If the overlap
* is exact then the reference count of the proxy is simply incremented.
* Otherwise, the proxy lock is split into smaller lock ranges and
* new proxy locks created for non overlapping ranges.
* The reference counts are adjusted accordingly.
2017-01-03 18:31:18 +01:00
* Meanwhile, the original lock is kept around (this is the callers handle)
2008-11-20 12:01:55 -08:00
* and its offset and length are used when releasing the lock.
*
* Thread coordination
* -------------------
* In order to make wakeups efficient and to ensure multiple continuous
* readers on a range don't starve a writer for the same range lock,
* two condition variables are allocated in each rl_t.
* If a writer (or reader) can't get a range it initialises the writer
* (or reader) cv; sets a flag saying there's a writer (or reader) waiting;
* and waits on that cv. When a thread unlocks that range it wakes up all
* writers then all readers before destroying the lock.
*
* Append mode writes
* ------------------
* Append mode writes need to lock a range at the end of a file.
* The offset of the end of the file is determined under the
* range locking mutex, and the lock type converted from RL_APPEND to
* RL_WRITER and the range locked.
*
* Grow block handling
* -------------------
* ZFS supports multiple block sizes, up to 16MB. The smallest
2008-11-20 12:01:55 -08:00
* block size is used for the file which is grown as needed. During this
* growth all other writers and readers must be excluded.
* So if the block size needs to be grown then the whole file is
* exclusively locked, then later the caller will reduce the lock
* range to just the range to be written using rangelock_reduce().
2008-11-20 12:01:55 -08:00
*/
#include <sys/zfs_context.h>
2008-11-20 12:01:55 -08:00
#include <sys/zfs_rlock.h>
/*
* AVL comparison function used to order range locks
* Locks are ordered on the start offset of the range.
*/
static int
zfs_rangelock_compare(const void *arg1, const void *arg2)
{
2019-11-01 10:37:33 -07:00
const zfs_locked_range_t *rl1 = (const zfs_locked_range_t *)arg1;
const zfs_locked_range_t *rl2 = (const zfs_locked_range_t *)arg2;
2019-10-09 10:36:03 -07:00
return (TREE_CMP(rl1->lr_offset, rl2->lr_offset));
}
/*
* The callback is invoked when acquiring a RL_WRITER or RL_APPEND lock.
* It must convert RL_APPEND to RL_WRITER (starting at the end of the file),
* and may increase the range that's locked for RL_WRITER.
*/
void
2019-11-01 10:37:33 -07:00
zfs_rangelock_init(zfs_rangelock_t *rl, zfs_rangelock_cb_t *cb, void *arg)
{
mutex_init(&rl->rl_lock, NULL, MUTEX_DEFAULT, NULL);
avl_create(&rl->rl_tree, zfs_rangelock_compare,
2019-11-01 10:37:33 -07:00
sizeof (zfs_locked_range_t), offsetof(zfs_locked_range_t, lr_node));
rl->rl_cb = cb;
rl->rl_arg = arg;
}
void
2019-11-01 10:37:33 -07:00
zfs_rangelock_fini(zfs_rangelock_t *rl)
{
mutex_destroy(&rl->rl_lock);
avl_destroy(&rl->rl_tree);
}
2008-11-20 12:01:55 -08:00
/*
2020-06-30 15:35:29 -04:00
* Check if a write lock can be grabbed. If not, fail immediately or sleep and
* recheck until available, depending on the value of the "nonblock" parameter.
2008-11-20 12:01:55 -08:00
*/
2020-06-30 15:35:29 -04:00
static boolean_t
zfs_rangelock_enter_writer(zfs_rangelock_t *rl, zfs_locked_range_t *new,
boolean_t nonblock)
2008-11-20 12:01:55 -08:00
{
avl_tree_t *tree = &rl->rl_tree;
2019-11-01 10:37:33 -07:00
zfs_locked_range_t *lr;
2008-11-20 12:01:55 -08:00
avl_index_t where;
uint64_t orig_off = new->lr_offset;
uint64_t orig_len = new->lr_length;
2019-11-01 10:37:33 -07:00
zfs_rangelock_type_t orig_type = new->lr_type;
2008-11-20 12:01:55 -08:00
for (;;) {
/*
* Call callback which can modify new->r_off,len,type.
* Note, the callback is used by the ZPL to handle appending
* and changing blocksizes. It isn't needed for zvols.
2008-11-20 12:01:55 -08:00
*/
if (rl->rl_cb != NULL) {
rl->rl_cb(new, rl->rl_arg);
2008-11-20 12:01:55 -08:00
}
/*
* If the type was APPEND, the callback must convert it to
* WRITER.
*/
ASSERT3U(new->lr_type, ==, RL_WRITER);
2008-11-20 12:01:55 -08:00
/*
* First check for the usual case of no locks
*/
if (avl_numnodes(tree) == 0) {
avl_add(tree, new);
2020-06-30 15:35:29 -04:00
return (B_TRUE);
2008-11-20 12:01:55 -08:00
}
/*
* Look for any locks in the range.
*/
lr = avl_find(tree, new, &where);
if (lr != NULL)
2008-11-20 12:01:55 -08:00
goto wait; /* already locked at same offset */
2019-11-01 10:37:33 -07:00
lr = avl_nearest(tree, where, AVL_AFTER);
if (lr != NULL &&
lr->lr_offset < new->lr_offset + new->lr_length)
2008-11-20 12:01:55 -08:00
goto wait;
2019-11-01 10:37:33 -07:00
lr = avl_nearest(tree, where, AVL_BEFORE);
if (lr != NULL &&
lr->lr_offset + lr->lr_length > new->lr_offset)
2008-11-20 12:01:55 -08:00
goto wait;
avl_insert(tree, new, where);
2020-06-30 15:35:29 -04:00
return (B_TRUE);
2008-11-20 12:01:55 -08:00
wait:
2020-06-30 15:35:29 -04:00
if (nonblock)
return (B_FALSE);
if (!lr->lr_write_wanted) {
cv_init(&lr->lr_write_cv, NULL, CV_DEFAULT, NULL);
lr->lr_write_wanted = B_TRUE;
2008-11-20 12:01:55 -08:00
}
cv_wait(&lr->lr_write_cv, &rl->rl_lock);
2008-11-20 12:01:55 -08:00
/* reset to original */
new->lr_offset = orig_off;
new->lr_length = orig_len;
new->lr_type = orig_type;
2008-11-20 12:01:55 -08:00
}
}
/*
* If this is an original (non-proxy) lock then replace it by
* a proxy and return the proxy.
*/
2019-11-01 10:37:33 -07:00
static zfs_locked_range_t *
zfs_rangelock_proxify(avl_tree_t *tree, zfs_locked_range_t *lr)
2008-11-20 12:01:55 -08:00
{
2019-11-01 10:37:33 -07:00
zfs_locked_range_t *proxy;
2008-11-20 12:01:55 -08:00
if (lr->lr_proxy)
return (lr); /* already a proxy */
2008-11-20 12:01:55 -08:00
ASSERT3U(lr->lr_count, ==, 1);
ASSERT(lr->lr_write_wanted == B_FALSE);
ASSERT(lr->lr_read_wanted == B_FALSE);
avl_remove(tree, lr);
lr->lr_count = 0;
2008-11-20 12:01:55 -08:00
/* create a proxy range lock */
2019-11-01 10:37:33 -07:00
proxy = kmem_alloc(sizeof (zfs_locked_range_t), KM_SLEEP);
proxy->lr_offset = lr->lr_offset;
proxy->lr_length = lr->lr_length;
proxy->lr_count = 1;
proxy->lr_type = RL_READER;
proxy->lr_proxy = B_TRUE;
proxy->lr_write_wanted = B_FALSE;
proxy->lr_read_wanted = B_FALSE;
2008-11-20 12:01:55 -08:00
avl_add(tree, proxy);
return (proxy);
}
/*
* Split the range lock at the supplied offset
* returning the *front* proxy.
*/
2019-11-01 10:37:33 -07:00
static zfs_locked_range_t *
zfs_rangelock_split(avl_tree_t *tree, zfs_locked_range_t *lr, uint64_t off)
2008-11-20 12:01:55 -08:00
{
2019-11-01 10:37:33 -07:00
zfs_locked_range_t *rear;
ASSERT3U(lr->lr_length, >, 1);
ASSERT3U(off, >, lr->lr_offset);
ASSERT3U(off, <, lr->lr_offset + lr->lr_length);
ASSERT(lr->lr_write_wanted == B_FALSE);
ASSERT(lr->lr_read_wanted == B_FALSE);
2008-11-20 12:01:55 -08:00
/* create the rear proxy range lock */
2019-11-01 10:37:33 -07:00
rear = kmem_alloc(sizeof (zfs_locked_range_t), KM_SLEEP);
rear->lr_offset = off;
rear->lr_length = lr->lr_offset + lr->lr_length - off;
rear->lr_count = lr->lr_count;
rear->lr_type = RL_READER;
rear->lr_proxy = B_TRUE;
rear->lr_write_wanted = B_FALSE;
rear->lr_read_wanted = B_FALSE;
2019-11-01 10:37:33 -07:00
zfs_locked_range_t *front = zfs_rangelock_proxify(tree, lr);
front->lr_length = off - lr->lr_offset;
2008-11-20 12:01:55 -08:00
avl_insert_here(tree, rear, front, AVL_AFTER);
return (front);
}
/*
* Create and add a new proxy range lock for the supplied range.
*/
static void
zfs_rangelock_new_proxy(avl_tree_t *tree, uint64_t off, uint64_t len)
2008-11-20 12:01:55 -08:00
{
2019-11-01 10:37:33 -07:00
zfs_locked_range_t *lr;
ASSERT(len != 0);
2019-11-01 10:37:33 -07:00
lr = kmem_alloc(sizeof (zfs_locked_range_t), KM_SLEEP);
lr->lr_offset = off;
lr->lr_length = len;
lr->lr_count = 1;
lr->lr_type = RL_READER;
lr->lr_proxy = B_TRUE;
lr->lr_write_wanted = B_FALSE;
lr->lr_read_wanted = B_FALSE;
avl_add(tree, lr);
2008-11-20 12:01:55 -08:00
}
static void
2019-11-01 10:37:33 -07:00
zfs_rangelock_add_reader(avl_tree_t *tree, zfs_locked_range_t *new,
zfs_locked_range_t *prev, avl_index_t where)
2008-11-20 12:01:55 -08:00
{
2019-11-01 10:37:33 -07:00
zfs_locked_range_t *next;
uint64_t off = new->lr_offset;
uint64_t len = new->lr_length;
2008-11-20 12:01:55 -08:00
/*
* prev arrives either:
* - pointing to an entry at the same offset
* - pointing to the entry with the closest previous offset whose
* range may overlap with the new range
* - null, if there were no ranges starting before the new one
*/
if (prev != NULL) {
if (prev->lr_offset + prev->lr_length <= off) {
2008-11-20 12:01:55 -08:00
prev = NULL;
} else if (prev->lr_offset != off) {
2008-11-20 12:01:55 -08:00
/*
* convert to proxy if needed then
* split this entry and bump ref count
*/
prev = zfs_rangelock_split(tree, prev, off);
2008-11-20 12:01:55 -08:00
prev = AVL_NEXT(tree, prev); /* move to rear range */
}
}
ASSERT((prev == NULL) || (prev->lr_offset == off));
2008-11-20 12:01:55 -08:00
if (prev != NULL)
2008-11-20 12:01:55 -08:00
next = prev;
else
next = avl_nearest(tree, where, AVL_AFTER);
2008-11-20 12:01:55 -08:00
if (next == NULL || off + len <= next->lr_offset) {
2008-11-20 12:01:55 -08:00
/* no overlaps, use the original new rl_t in the tree */
avl_insert(tree, new, where);
return;
}
if (off < next->lr_offset) {
2008-11-20 12:01:55 -08:00
/* Add a proxy for initial range before the overlap */
zfs_rangelock_new_proxy(tree, off, next->lr_offset - off);
2008-11-20 12:01:55 -08:00
}
new->lr_count = 0; /* will use proxies in tree */
2008-11-20 12:01:55 -08:00
/*
* We now search forward through the ranges, until we go past the end
* of the new range. For each entry we make it a proxy if it
* isn't already, then bump its reference count. If there's any
* gaps between the ranges then we create a new proxy range.
*/
for (prev = NULL; next; prev = next, next = AVL_NEXT(tree, next)) {
if (off + len <= next->lr_offset)
2008-11-20 12:01:55 -08:00
break;
if (prev != NULL && prev->lr_offset + prev->lr_length <
next->lr_offset) {
2008-11-20 12:01:55 -08:00
/* there's a gap */
ASSERT3U(next->lr_offset, >,
prev->lr_offset + prev->lr_length);
zfs_rangelock_new_proxy(tree,
prev->lr_offset + prev->lr_length,
next->lr_offset -
(prev->lr_offset + prev->lr_length));
2008-11-20 12:01:55 -08:00
}
if (off + len == next->lr_offset + next->lr_length) {
2008-11-20 12:01:55 -08:00
/* exact overlap with end */
next = zfs_rangelock_proxify(tree, next);
next->lr_count++;
2008-11-20 12:01:55 -08:00
return;
}
if (off + len < next->lr_offset + next->lr_length) {
2008-11-20 12:01:55 -08:00
/* new range ends in the middle of this block */
next = zfs_rangelock_split(tree, next, off + len);
next->lr_count++;
2008-11-20 12:01:55 -08:00
return;
}
ASSERT3U(off + len, >, next->lr_offset + next->lr_length);
next = zfs_rangelock_proxify(tree, next);
next->lr_count++;
2008-11-20 12:01:55 -08:00
}
/* Add the remaining end range. */
zfs_rangelock_new_proxy(tree, prev->lr_offset + prev->lr_length,
(off + len) - (prev->lr_offset + prev->lr_length));
2008-11-20 12:01:55 -08:00
}
/*
2020-06-30 15:35:29 -04:00
* Check if a reader lock can be grabbed. If not, fail immediately or sleep and
* recheck until available, depending on the value of the "nonblock" parameter.
2008-11-20 12:01:55 -08:00
*/
2020-06-30 15:35:29 -04:00
static boolean_t
zfs_rangelock_enter_reader(zfs_rangelock_t *rl, zfs_locked_range_t *new,
boolean_t nonblock)
2008-11-20 12:01:55 -08:00
{
avl_tree_t *tree = &rl->rl_tree;
2019-11-01 10:37:33 -07:00
zfs_locked_range_t *prev, *next;
2008-11-20 12:01:55 -08:00
avl_index_t where;
uint64_t off = new->lr_offset;
uint64_t len = new->lr_length;
2008-11-20 12:01:55 -08:00
/*
* Look for any writer locks in the range.
*/
retry:
prev = avl_find(tree, new, &where);
if (prev == NULL)
2019-11-01 10:37:33 -07:00
prev = avl_nearest(tree, where, AVL_BEFORE);
2008-11-20 12:01:55 -08:00
/*
* Check the previous range for a writer lock overlap.
*/
if (prev && (off < prev->lr_offset + prev->lr_length)) {
if ((prev->lr_type == RL_WRITER) || (prev->lr_write_wanted)) {
2020-06-30 15:35:29 -04:00
if (nonblock)
return (B_FALSE);
if (!prev->lr_read_wanted) {
cv_init(&prev->lr_read_cv,
NULL, CV_DEFAULT, NULL);
prev->lr_read_wanted = B_TRUE;
2008-11-20 12:01:55 -08:00
}
cv_wait(&prev->lr_read_cv, &rl->rl_lock);
2008-11-20 12:01:55 -08:00
goto retry;
}
if (off + len < prev->lr_offset + prev->lr_length)
2008-11-20 12:01:55 -08:00
goto got_lock;
}
/*
* Search through the following ranges to see if there's
* write lock any overlap.
*/
if (prev != NULL)
2008-11-20 12:01:55 -08:00
next = AVL_NEXT(tree, prev);
else
2019-11-01 10:37:33 -07:00
next = avl_nearest(tree, where, AVL_AFTER);
for (; next != NULL; next = AVL_NEXT(tree, next)) {
if (off + len <= next->lr_offset)
2008-11-20 12:01:55 -08:00
goto got_lock;
if ((next->lr_type == RL_WRITER) || (next->lr_write_wanted)) {
2020-06-30 15:35:29 -04:00
if (nonblock)
return (B_FALSE);
if (!next->lr_read_wanted) {
cv_init(&next->lr_read_cv,
NULL, CV_DEFAULT, NULL);
next->lr_read_wanted = B_TRUE;
2008-11-20 12:01:55 -08:00
}
cv_wait(&next->lr_read_cv, &rl->rl_lock);
2008-11-20 12:01:55 -08:00
goto retry;
}
if (off + len <= next->lr_offset + next->lr_length)
2008-11-20 12:01:55 -08:00
goto got_lock;
}
got_lock:
/*
* Add the read lock, which may involve splitting existing
* locks and bumping ref counts (r_count).
2008-11-20 12:01:55 -08:00
*/
zfs_rangelock_add_reader(tree, new, prev, where);
2020-06-30 15:35:29 -04:00
return (B_TRUE);
2008-11-20 12:01:55 -08:00
}
/*
* Lock a range (offset, length) as either shared (RL_READER) or exclusive
* (RL_WRITER or RL_APPEND). If RL_APPEND is specified, rl_cb() will convert
* it to a RL_WRITER lock (with the offset at the end of the file). Returns
* the range lock structure for later unlocking (or reduce range if the
2020-06-30 15:35:29 -04:00
* entire file is locked as RL_WRITER), or NULL if nonblock is true and the
* lock could not be acquired immediately.
2008-11-20 12:01:55 -08:00
*/
2020-06-30 15:35:29 -04:00
static zfs_locked_range_t *
zfs_rangelock_enter_impl(zfs_rangelock_t *rl, uint64_t off, uint64_t len,
zfs_rangelock_type_t type, boolean_t nonblock)
2008-11-20 12:01:55 -08:00
{
2019-11-01 10:37:33 -07:00
zfs_locked_range_t *new;
2008-11-20 12:01:55 -08:00
ASSERT(type == RL_READER || type == RL_WRITER || type == RL_APPEND);
2019-11-01 10:37:33 -07:00
new = kmem_alloc(sizeof (zfs_locked_range_t), KM_SLEEP);
new->lr_rangelock = rl;
new->lr_offset = off;
2009-02-18 12:51:31 -08:00
if (len + off < off) /* overflow */
len = UINT64_MAX - off;
new->lr_length = len;
new->lr_count = 1; /* assume it's going to be in the tree */
new->lr_type = type;
new->lr_proxy = B_FALSE;
new->lr_write_wanted = B_FALSE;
new->lr_read_wanted = B_FALSE;
mutex_enter(&rl->rl_lock);
2008-11-20 12:01:55 -08:00
if (type == RL_READER) {
/*
* First check for the usual case of no locks
*/
2020-06-30 15:35:29 -04:00
if (avl_numnodes(&rl->rl_tree) == 0) {
avl_add(&rl->rl_tree, new);
2020-06-30 15:35:29 -04:00
} else if (!zfs_rangelock_enter_reader(rl, new, nonblock)) {
kmem_free(new, sizeof (*new));
new = NULL;
}
} else if (!zfs_rangelock_enter_writer(rl, new, nonblock)) {
kmem_free(new, sizeof (*new));
new = NULL;
}
mutex_exit(&rl->rl_lock);
2008-11-20 12:01:55 -08:00
return (new);
}
2020-06-30 15:35:29 -04:00
zfs_locked_range_t *
zfs_rangelock_enter(zfs_rangelock_t *rl, uint64_t off, uint64_t len,
zfs_rangelock_type_t type)
{
return (zfs_rangelock_enter_impl(rl, off, len, type, B_FALSE));
}
zfs_locked_range_t *
zfs_rangelock_tryenter(zfs_rangelock_t *rl, uint64_t off, uint64_t len,
zfs_rangelock_type_t type)
{
return (zfs_rangelock_enter_impl(rl, off, len, type, B_TRUE));
}
/*
2019-11-01 10:37:33 -07:00
* Safely free the zfs_locked_range_t.
*/
static void
2019-11-01 10:37:33 -07:00
zfs_rangelock_free(zfs_locked_range_t *lr)
{
if (lr->lr_write_wanted)
cv_destroy(&lr->lr_write_cv);
if (lr->lr_read_wanted)
cv_destroy(&lr->lr_read_cv);
2019-11-01 10:37:33 -07:00
kmem_free(lr, sizeof (zfs_locked_range_t));
}
2008-11-20 12:01:55 -08:00
/*
* Unlock a reader lock
*/
static void
2019-11-01 10:37:33 -07:00
zfs_rangelock_exit_reader(zfs_rangelock_t *rl, zfs_locked_range_t *remove,
list_t *free_list)
2008-11-20 12:01:55 -08:00
{
avl_tree_t *tree = &rl->rl_tree;
2008-11-20 12:01:55 -08:00
uint64_t len;
/*
* The common case is when the remove entry is in the tree
* (cnt == 1) meaning there's been no other reader locks overlapping
* with this one. Otherwise the remove entry will have been
* removed from the tree and replaced by proxies (one or
* more ranges mapping to the entire range).
*/
if (remove->lr_count == 1) {
2008-11-20 12:01:55 -08:00
avl_remove(tree, remove);
if (remove->lr_write_wanted)
cv_broadcast(&remove->lr_write_cv);
if (remove->lr_read_wanted)
cv_broadcast(&remove->lr_read_cv);
2011-03-08 12:17:35 -08:00
list_insert_tail(free_list, remove);
2008-11-20 12:01:55 -08:00
} else {
ASSERT0(remove->lr_count);
ASSERT0(remove->lr_write_wanted);
ASSERT0(remove->lr_read_wanted);
2008-11-20 12:01:55 -08:00
/*
* Find start proxy representing this reader lock,
* then decrement ref count on all proxies
* that make up this range, freeing them as needed.
*/
2019-11-01 10:37:33 -07:00
zfs_locked_range_t *lr = avl_find(tree, remove, NULL);
ASSERT3P(lr, !=, NULL);
ASSERT3U(lr->lr_count, !=, 0);
ASSERT3U(lr->lr_type, ==, RL_READER);
2019-11-01 10:37:33 -07:00
zfs_locked_range_t *next = NULL;
for (len = remove->lr_length; len != 0; lr = next) {
len -= lr->lr_length;
if (len != 0) {
next = AVL_NEXT(tree, lr);
ASSERT3P(next, !=, NULL);
ASSERT3U(lr->lr_offset + lr->lr_length, ==,
next->lr_offset);
ASSERT3U(next->lr_count, !=, 0);
ASSERT3U(next->lr_type, ==, RL_READER);
2008-11-20 12:01:55 -08:00
}
lr->lr_count--;
if (lr->lr_count == 0) {
avl_remove(tree, lr);
if (lr->lr_write_wanted)
cv_broadcast(&lr->lr_write_cv);
if (lr->lr_read_wanted)
cv_broadcast(&lr->lr_read_cv);
list_insert_tail(free_list, lr);
2008-11-20 12:01:55 -08:00
}
}
2019-11-01 10:37:33 -07:00
kmem_free(remove, sizeof (zfs_locked_range_t));
2008-11-20 12:01:55 -08:00
}
}
/*
* Unlock range and destroy range lock structure.
*/
void
2019-11-01 10:37:33 -07:00
zfs_rangelock_exit(zfs_locked_range_t *lr)
2008-11-20 12:01:55 -08:00
{
2019-11-01 10:37:33 -07:00
zfs_rangelock_t *rl = lr->lr_rangelock;
2011-03-08 12:17:35 -08:00
list_t free_list;
2019-11-01 10:37:33 -07:00
zfs_locked_range_t *free_lr;
2008-11-20 12:01:55 -08:00
ASSERT(lr->lr_type == RL_WRITER || lr->lr_type == RL_READER);
ASSERT(lr->lr_count == 1 || lr->lr_count == 0);
ASSERT(!lr->lr_proxy);
/*
* The free list is used to defer the cv_destroy() and
* subsequent kmem_free until after the mutex is dropped.
*/
2019-11-01 10:37:33 -07:00
list_create(&free_list, sizeof (zfs_locked_range_t),
offsetof(zfs_locked_range_t, lr_node));
mutex_enter(&rl->rl_lock);
if (lr->lr_type == RL_WRITER) {
/* writer locks can't be shared or split */
avl_remove(&rl->rl_tree, lr);
if (lr->lr_write_wanted)
cv_broadcast(&lr->lr_write_cv);
if (lr->lr_read_wanted)
cv_broadcast(&lr->lr_read_cv);
list_insert_tail(&free_list, lr);
2008-11-20 12:01:55 -08:00
} else {
/*
* lock may be shared, let rangelock_exit_reader()
2019-11-01 10:37:33 -07:00
* release the lock and free the zfs_locked_range_t.
2008-11-20 12:01:55 -08:00
*/
zfs_rangelock_exit_reader(rl, lr, &free_list);
2008-11-20 12:01:55 -08:00
}
mutex_exit(&rl->rl_lock);
2011-03-08 12:17:35 -08:00
while ((free_lr = list_remove_head(&free_list)) != NULL)
zfs_rangelock_free(free_lr);
2011-03-08 12:17:35 -08:00
list_destroy(&free_list);
2008-11-20 12:01:55 -08:00
}
/*
* Reduce range locked as RL_WRITER from whole file to specified range.
* Asserts the whole file is exclusively locked and so there's only one
2008-11-20 12:01:55 -08:00
* entry in the tree.
*/
void
2019-11-01 10:37:33 -07:00
zfs_rangelock_reduce(zfs_locked_range_t *lr, uint64_t off, uint64_t len)
2008-11-20 12:01:55 -08:00
{
2019-11-01 10:37:33 -07:00
zfs_rangelock_t *rl = lr->lr_rangelock;
2008-11-20 12:01:55 -08:00
/* Ensure there are no other locks */
ASSERT3U(avl_numnodes(&rl->rl_tree), ==, 1);
2025-08-04 20:22:42 +10:00
ASSERT0(lr->lr_offset);
ASSERT3U(lr->lr_type, ==, RL_WRITER);
ASSERT(!lr->lr_proxy);
ASSERT3U(lr->lr_length, ==, UINT64_MAX);
ASSERT3U(lr->lr_count, ==, 1);
mutex_enter(&rl->rl_lock);
lr->lr_offset = off;
lr->lr_length = len;
mutex_exit(&rl->rl_lock);
if (lr->lr_write_wanted)
cv_broadcast(&lr->lr_write_cv);
if (lr->lr_read_wanted)
cv_broadcast(&lr->lr_read_cv);
2008-11-20 12:01:55 -08:00
}
2016-04-11 14:53:48 -07:00
#if defined(_KERNEL)
EXPORT_SYMBOL(zfs_rangelock_init);
EXPORT_SYMBOL(zfs_rangelock_fini);
EXPORT_SYMBOL(zfs_rangelock_enter);
2020-06-30 15:35:29 -04:00
EXPORT_SYMBOL(zfs_rangelock_tryenter);
EXPORT_SYMBOL(zfs_rangelock_exit);
EXPORT_SYMBOL(zfs_rangelock_reduce);
2016-04-11 14:53:48 -07:00
#endif