Archive-Team
/
zfs
2
0
Fork 0
Code Issues Pull Requests Packages Projects Releases 103 Wiki Activity

Add a "try" operation for range locks 

zfs_rangelock_tryenter() bails immediately instead of waiting for the
lock to become available.  This will be used to resolve a deadlock in
the FreeBSD page-in code.  No functional change intended.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@ixsystems.com>
Signed-off-by: Mark Johnston <markj@FreeBSD.org>
Closes #10519
This commit is contained in:
Mark Johnston 2020-06-30 15:35:29 -04:00 committed by Brian Behlendorf
parent a4b0a74c7f
commit 6e00561712
2 changed files with 49 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
include/sys/zfs_rlock.h
View File

@ -71,6 +71,8 @@ void zfs_rangelock_fini(zfs_rangelock_t *);
zfs_locked_range_t *zfs_rangelock_enter(zfs_rangelock_t *,
uint64_t, uint64_t, zfs_rangelock_type_t);
zfs_locked_range_t *zfs_rangelock_tryenter(zfs_rangelock_t *,
uint64_t, uint64_t, zfs_rangelock_type_t);
void zfs_rangelock_exit(zfs_locked_range_t *);
void zfs_rangelock_reduce(zfs_locked_range_t *, uint64_t, uint64_t);

65
module/zfs/zfs_rlock.c
View File

@ -150,10 +150,12 @@ zfs_rangelock_fini(zfs_rangelock_t *rl)
}
/*
* Check if a write lock can be grabbed, or wait and recheck until available.
* 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.
*/
static void
zfs_rangelock_enter_writer(zfs_rangelock_t *rl, zfs_locked_range_t *new)
static boolean_t
zfs_rangelock_enter_writer(zfs_rangelock_t *rl, zfs_locked_range_t *new,
boolean_t nonblock)
{
avl_tree_t *tree = &rl->rl_tree;
zfs_locked_range_t *lr;
@ -183,7 +185,7 @@ zfs_rangelock_enter_writer(zfs_rangelock_t *rl, zfs_locked_range_t *new)
*/
if (avl_numnodes(tree) == 0) {
avl_add(tree, new);
return;
return (B_TRUE);
}
/*
@ -204,8 +206,10 @@ zfs_rangelock_enter_writer(zfs_rangelock_t *rl, zfs_locked_range_t *new)
goto wait;
avl_insert(tree, new, where);
return;
return (B_TRUE);
wait:
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;
@ -391,10 +395,12 @@ zfs_rangelock_add_reader(avl_tree_t *tree, zfs_locked_range_t *new,
}
/*
* Check if a reader lock can be grabbed, or wait and recheck until available.
* 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.
*/
static void
zfs_rangelock_enter_reader(zfs_rangelock_t *rl, zfs_locked_range_t *new)
static boolean_t
zfs_rangelock_enter_reader(zfs_rangelock_t *rl, zfs_locked_range_t *new,
boolean_t nonblock)
{
avl_tree_t *tree = &rl->rl_tree;
zfs_locked_range_t *prev, *next;
@ -415,6 +421,8 @@ retry:
*/
if (prev && (off < prev->lr_offset + prev->lr_length)) {
if ((prev->lr_type == RL_WRITER) || (prev->lr_write_wanted)) {
if (nonblock)
return (B_FALSE);
if (!prev->lr_read_wanted) {
cv_init(&prev->lr_read_cv,
NULL, CV_DEFAULT, NULL);
@ -439,6 +447,8 @@ retry:
if (off + len <= next->lr_offset)
goto got_lock;
if ((next->lr_type == RL_WRITER) || (next->lr_write_wanted)) {
if (nonblock)
return (B_FALSE);
if (!next->lr_read_wanted) {
cv_init(&next->lr_read_cv,
NULL, CV_DEFAULT, NULL);
@ -457,6 +467,7 @@ got_lock:
* locks and bumping ref counts (r_count).
*/
zfs_rangelock_add_reader(tree, new, prev, where);
return (B_TRUE);
}
/*
@ -464,11 +475,12 @@ got_lock:
* (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
* entire file is locked as RL_WRITER).
* entire file is locked as RL_WRITER), or NULL if nonblock is true and the
* lock could not be acquired immediately.
*/
zfs_locked_range_t *
zfs_rangelock_enter(zfs_rangelock_t *rl, uint64_t off, uint64_t len,
zfs_rangelock_type_t type)
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)
{
zfs_locked_range_t *new;
@ -491,18 +503,34 @@ zfs_rangelock_enter(zfs_rangelock_t *rl, uint64_t off, uint64_t len,
/*
* First check for the usual case of no locks
*/
if (avl_numnodes(&rl->rl_tree) == 0)
if (avl_numnodes(&rl->rl_tree) == 0) {
avl_add(&rl->rl_tree, new);
else
zfs_rangelock_enter_reader(rl, new);
} else {
/* RL_WRITER or RL_APPEND */
zfs_rangelock_enter_writer(rl, new);
} 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);
return (new);
}
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));
}
/*
* Safely free the zfs_locked_range_t.
*/
@ -657,6 +685,7 @@ zfs_rangelock_reduce(zfs_locked_range_t *lr, uint64_t off, uint64_t len)
EXPORT_SYMBOL(zfs_rangelock_init);
EXPORT_SYMBOL(zfs_rangelock_fini);
EXPORT_SYMBOL(zfs_rangelock_enter);
EXPORT_SYMBOL(zfs_rangelock_tryenter);
EXPORT_SYMBOL(zfs_rangelock_exit);
EXPORT_SYMBOL(zfs_rangelock_reduce);
#endif