* Copyright 2010 Sun Microsystems, Inc. All rights reserved.
* Use is subject to license terms.
*/
+/*
+ * Copyright (c) 2012 by Delphix. All rights reserved.
+ */
/*
* This file contains the code to implement file range locking in
* Range locking is also used by zvol and uses a
* dummied up znode. However, for zvol, we don't need to
* append or grow blocksize, and besides we don't have
- * a "sa" data or z_zfsvfs - so skip that processing.
+ * a "sa" data or zfs_sb_t - so skip that processing.
*
* Yes, this is ugly, and would be solved by not handling
* grow or append in range lock code. If that was done then
* we could make the range locking code generically available
* to other non-zfs consumers.
*/
- if (zp->z_vnode) { /* caller is ZPL */
+ if (!zp->z_is_zvol) { /* caller is ZPL */
/*
* If in append mode pick up the current end of file.
* This is done under z_range_lock to avoid races.
*/
end_size = MAX(zp->z_size, new->r_off + len);
if (end_size > zp->z_blksz && (!ISP2(zp->z_blksz) ||
- zp->z_blksz < zp->z_zfsvfs->z_max_blksz)) {
+ zp->z_blksz < ZTOZSB(zp)->z_max_blksz)) {
new->r_off = 0;
new->r_len = UINT64_MAX;
}
rl->r_cnt = 0;
/* create a proxy range lock */
- proxy = kmem_alloc(sizeof (rl_t), KM_SLEEP);
+ proxy = kmem_alloc(sizeof (rl_t), KM_PUSHPAGE);
proxy->r_off = rl->r_off;
proxy->r_len = rl->r_len;
proxy->r_cnt = 1;
ASSERT(rl->r_read_wanted == B_FALSE);
/* create the rear proxy range lock */
- rear = kmem_alloc(sizeof (rl_t), KM_SLEEP);
+ rear = kmem_alloc(sizeof (rl_t), KM_PUSHPAGE);
rear->r_off = off;
rear->r_len = rl->r_off + rl->r_len - off;
rear->r_cnt = rl->r_cnt;
rl_t *rl;
ASSERT(len);
- rl = kmem_alloc(sizeof (rl_t), KM_SLEEP);
+ rl = kmem_alloc(sizeof (rl_t), KM_PUSHPAGE);
rl->r_off = off;
rl->r_len = len;
rl->r_cnt = 1;
ASSERT(type == RL_READER || type == RL_WRITER || type == RL_APPEND);
- new = kmem_alloc(sizeof (rl_t), KM_SLEEP);
+ new = kmem_alloc(sizeof (rl_t), KM_PUSHPAGE);
new->r_zp = zp;
new->r_off = off;
if (len + off < off) /* overflow */
return (new);
}
+static void
+zfs_range_free(void *arg)
+{
+ rl_t *rl = arg;
+
+ if (rl->r_write_wanted)
+ cv_destroy(&rl->r_wr_cv);
+
+ if (rl->r_read_wanted)
+ cv_destroy(&rl->r_rd_cv);
+
+ kmem_free(rl, sizeof (rl_t));
+}
+
/*
* Unlock a reader lock
*/
static void
-zfs_range_unlock_reader(znode_t *zp, rl_t *remove)
+zfs_range_unlock_reader(znode_t *zp, rl_t *remove, list_t *free_list)
{
avl_tree_t *tree = &zp->z_range_avl;
- rl_t *rl, *next;
+ rl_t *rl, *next = NULL;
uint64_t len;
/*
*/
if (remove->r_cnt == 1) {
avl_remove(tree, remove);
- if (remove->r_write_wanted) {
+
+ if (remove->r_write_wanted)
cv_broadcast(&remove->r_wr_cv);
- cv_destroy(&remove->r_wr_cv);
- }
- if (remove->r_read_wanted) {
+
+ if (remove->r_read_wanted)
cv_broadcast(&remove->r_rd_cv);
- cv_destroy(&remove->r_rd_cv);
- }
+
+ list_insert_tail(free_list, remove);
} else {
- ASSERT3U(remove->r_cnt, ==, 0);
- ASSERT3U(remove->r_write_wanted, ==, 0);
- ASSERT3U(remove->r_read_wanted, ==, 0);
+ ASSERT0(remove->r_cnt);
+ ASSERT0(remove->r_write_wanted);
+ ASSERT0(remove->r_read_wanted);
/*
* Find start proxy representing this reader lock,
* then decrement ref count on all proxies
rl->r_cnt--;
if (rl->r_cnt == 0) {
avl_remove(tree, rl);
- if (rl->r_write_wanted) {
+
+ if (rl->r_write_wanted)
cv_broadcast(&rl->r_wr_cv);
- cv_destroy(&rl->r_wr_cv);
- }
- if (rl->r_read_wanted) {
+
+ if (rl->r_read_wanted)
cv_broadcast(&rl->r_rd_cv);
- cv_destroy(&rl->r_rd_cv);
- }
- kmem_free(rl, sizeof (rl_t));
+
+ list_insert_tail(free_list, rl);
}
}
+
+ kmem_free(remove, sizeof (rl_t));
}
- kmem_free(remove, sizeof (rl_t));
}
/*
zfs_range_unlock(rl_t *rl)
{
znode_t *zp = rl->r_zp;
+ list_t free_list;
+ rl_t *free_rl;
ASSERT(rl->r_type == RL_WRITER || rl->r_type == RL_READER);
ASSERT(rl->r_cnt == 1 || rl->r_cnt == 0);
ASSERT(!rl->r_proxy);
+ list_create(&free_list, sizeof(rl_t), offsetof(rl_t, rl_node));
mutex_enter(&zp->z_range_lock);
if (rl->r_type == RL_WRITER) {
/* writer locks can't be shared or split */
avl_remove(&zp->z_range_avl, rl);
- mutex_exit(&zp->z_range_lock);
- if (rl->r_write_wanted) {
+ if (rl->r_write_wanted)
cv_broadcast(&rl->r_wr_cv);
- cv_destroy(&rl->r_wr_cv);
- }
- if (rl->r_read_wanted) {
+
+ if (rl->r_read_wanted)
cv_broadcast(&rl->r_rd_cv);
- cv_destroy(&rl->r_rd_cv);
- }
- kmem_free(rl, sizeof (rl_t));
+
+ list_insert_tail(&free_list, rl);
} else {
/*
* lock may be shared, let zfs_range_unlock_reader()
- * release the lock and free the rl_t
+ * release the zp->z_range_lock lock and free the rl_t
*/
- zfs_range_unlock_reader(zp, rl);
- mutex_exit(&zp->z_range_lock);
+ zfs_range_unlock_reader(zp, rl, &free_list);
}
+ mutex_exit(&zp->z_range_lock);
+
+ while ((free_rl = list_head(&free_list)) != NULL) {
+ list_remove(&free_list, free_rl);
+ zfs_range_free(free_rl);
+ }
+
+ list_destroy(&free_list);
}
/*
mutex_enter(&zp->z_range_lock);
rl->r_off = off;
rl->r_len = len;
- mutex_exit(&zp->z_range_lock);
+
if (rl->r_write_wanted)
cv_broadcast(&rl->r_wr_cv);
if (rl->r_read_wanted)
cv_broadcast(&rl->r_rd_cv);
+
+ mutex_exit(&zp->z_range_lock);
}
/*
git://git.camperquake.de / zfs.git / blobdiff