#include <linux/list_sort.h>
#include <linux/lockref.h>
#include <linux/rhashtable.h>
+ #include <linux/pid_namespace.h>
+ #include <linux/fdtable.h>
+ #include <linux/file.h>
#include "gfs2.h"
#include "incore.h"
static void do_xmote(struct gfs2_glock *gl, struct gfs2_holder *gh, unsigned int target);
static void __gfs2_glock_dq(struct gfs2_holder *gh);
+static void handle_callback(struct gfs2_glock *gl, unsigned int state,
+ unsigned long delay, bool remote);
static struct dentry *gfs2_root;
static struct workqueue_struct *glock_workqueue;
/**
* demote_incompat_holders - demote incompatible demoteable holders
* @gl: the glock we want to promote
- * @new_gh: the new holder to be promoted
+ * @current_gh: the newly promoted holder
+ *
+ * We're passing the newly promoted holder in @current_gh, but actually, any of
+ * the strong holders would do.
*/
static void demote_incompat_holders(struct gfs2_glock *gl,
- struct gfs2_holder *new_gh)
+ struct gfs2_holder *current_gh)
{
struct gfs2_holder *gh, *tmp;
*/
if (!test_bit(HIF_HOLDER, &gh->gh_iflags))
return;
+ if (gh == current_gh)
+ continue;
if (test_bit(HIF_MAY_DEMOTE, &gh->gh_iflags) &&
- !may_grant(gl, new_gh, gh)) {
+ !may_grant(gl, current_gh, gh)) {
/*
* We should not recurse into do_promote because
* __gfs2_glock_dq only calls handle_callback,
* gfs2_instantiate - Call the glops instantiate function
* @gh: The glock holder
*
- * Returns: 0 if instantiate was successful, 2 if type specific operation is
- * underway, or error.
+ * Returns: 0 if instantiate was successful, or error.
*/
int gfs2_instantiate(struct gfs2_holder *gh)
{
again:
if (!test_bit(GLF_INSTANTIATE_NEEDED, &gl->gl_flags))
- return 0;
+ goto done;
/*
* Since we unlock the lockref lock, we set a flag to indicate
goto again;
}
- ret = glops->go_instantiate(gh);
+ ret = glops->go_instantiate(gl);
if (!ret)
clear_bit(GLF_INSTANTIATE_NEEDED, &gl->gl_flags);
clear_and_wake_up_bit(GLF_INSTANTIATE_IN_PROG, &gl->gl_flags);
- return ret;
+ if (ret)
+ return ret;
+
+done:
+ if (glops->go_held)
+ return glops->go_held(gh);
+ return 0;
}
/**
* do_promote - promote as many requests as possible on the current queue
* @gl: The glock
*
- * Returns: 1 if there is a blocked holder at the head of the list, or 2
- * if a type specific operation is underway.
+ * Returns: 1 if there is a blocked holder at the head of the list
*/
static int do_promote(struct gfs2_glock *gl)
-__releases(&gl->gl_lockref.lock)
-__acquires(&gl->gl_lockref.lock)
{
- struct gfs2_holder *gh, *tmp, *first_gh;
+ struct gfs2_holder *gh, *current_gh;
bool incompat_holders_demoted = false;
- bool lock_released;
- int ret;
-restart:
- first_gh = find_first_strong_holder(gl);
- list_for_each_entry_safe(gh, tmp, &gl->gl_holders, gh_list) {
- lock_released = false;
+ current_gh = find_first_strong_holder(gl);
+ list_for_each_entry(gh, &gl->gl_holders, gh_list) {
if (test_bit(HIF_HOLDER, &gh->gh_iflags))
continue;
- if (!may_grant(gl, first_gh, gh)) {
+ if (!may_grant(gl, current_gh, gh)) {
/*
- * If we get here, it means we may not grant this holder for
- * some reason. If this holder is the head of the list, it
- * means we have a blocked holder at the head, so return 1.
+ * If we get here, it means we may not grant this
+ * holder for some reason. If this holder is at the
+ * head of the list, it means we have a blocked holder
+ * at the head, so return 1.
*/
if (list_is_first(&gh->gh_list, &gl->gl_holders))
return 1;
do_error(gl, 0);
break;
}
- if (!incompat_holders_demoted) {
- demote_incompat_holders(gl, first_gh);
- incompat_holders_demoted = true;
- first_gh = gh;
- }
- if (test_bit(GLF_INSTANTIATE_NEEDED, &gl->gl_flags) &&
- !(gh->gh_flags & GL_SKIP) && gl->gl_ops->go_instantiate) {
- lock_released = true;
- spin_unlock(&gl->gl_lockref.lock);
- ret = gfs2_instantiate(gh);
- spin_lock(&gl->gl_lockref.lock);
- if (ret) {
- if (ret == 1)
- return 2;
- gh->gh_error = ret;
- list_del_init(&gh->gh_list);
- trace_gfs2_glock_queue(gh, 0);
- gfs2_holder_wake(gh);
- goto restart;
- }
- }
set_bit(HIF_HOLDER, &gh->gh_iflags);
trace_gfs2_promote(gh);
gfs2_holder_wake(gh);
- /*
- * If we released the gl_lockref.lock the holders list may have
- * changed. For that reason, we start again at the start of
- * the holders queue.
- */
- if (lock_released)
- goto restart;
+ if (!incompat_holders_demoted) {
+ current_gh = gh;
+ demote_incompat_holders(gl, current_gh);
+ incompat_holders_demoted = true;
+ }
}
return 0;
}
const struct gfs2_glock_operations *glops = gl->gl_ops;
struct gfs2_holder *gh;
unsigned state = ret & LM_OUT_ST_MASK;
- int rv;
spin_lock(&gl->gl_lockref.lock);
trace_gfs2_glock_state_change(gl, state);
gfs2_demote_wake(gl);
if (state != LM_ST_UNLOCKED) {
if (glops->go_xmote_bh) {
+ int rv;
+
spin_unlock(&gl->gl_lockref.lock);
rv = glops->go_xmote_bh(gl);
spin_lock(&gl->gl_lockref.lock);
goto out;
}
}
- rv = do_promote(gl);
- if (rv == 2)
- goto out_locked;
+ do_promote(gl);
}
out:
clear_bit(GLF_LOCK, &gl->gl_flags);
-out_locked:
spin_unlock(&gl->gl_lockref.lock);
}
*
*/
-static void do_xmote(struct gfs2_glock *gl, struct gfs2_holder *gh, unsigned int target)
+static void do_xmote(struct gfs2_glock *gl, struct gfs2_holder *gh,
+ unsigned int target)
__releases(&gl->gl_lockref.lock)
__acquires(&gl->gl_lockref.lock)
{
if (target != LM_ST_UNLOCKED && glock_blocked_by_withdraw(gl) &&
gh && !(gh->gh_flags & LM_FLAG_NOEXP))
- return;
+ goto skip_inval;
+
lck_flags &= (LM_FLAG_TRY | LM_FLAG_TRY_1CB | LM_FLAG_NOEXP |
LM_FLAG_PRIORITY);
GLOCK_BUG_ON(gl, gl->gl_state == target);
(target != LM_ST_UNLOCKED ||
test_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags))) {
if (!is_system_glock(gl)) {
+ handle_callback(gl, LM_ST_UNLOCKED, 0, false); /* sets demote */
+ /*
+ * Ordinarily, we would call dlm and its callback would call
+ * finish_xmote, which would call state_change() to the new state.
+ * Since we withdrew, we won't call dlm, so call state_change
+ * manually, but to the UNLOCKED state we desire.
+ */
+ state_change(gl, LM_ST_UNLOCKED);
+ /*
+ * We skip telling dlm to do the locking, so we won't get a
+ * reply that would otherwise clear GLF_LOCK. So we clear it here.
+ */
+ clear_bit(GLF_LOCK, &gl->gl_flags);
+ clear_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags);
gfs2_glock_queue_work(gl, GL_GLOCK_DFT_HOLD);
goto out;
} else {
__acquires(&gl->gl_lockref.lock)
{
struct gfs2_holder *gh = NULL;
- int ret;
if (test_and_set_bit(GLF_LOCK, &gl->gl_flags))
return;
} else {
if (test_bit(GLF_DEMOTE, &gl->gl_flags))
gfs2_demote_wake(gl);
- ret = do_promote(gl);
- if (ret == 0)
+ if (do_promote(gl) == 0)
goto out_unlock;
- if (ret == 2)
- goto out;
gh = find_first_waiter(gl);
gl->gl_target = gh->gh_state;
if (!(gh->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB)))
do_error(gl, 0); /* Fail queued try locks */
}
do_xmote(gl, gh, gl->gl_target);
-out:
return;
out_sched:
if (gfs2_queue_delete_work(gl, 5 * HZ))
return;
}
- goto out;
}
inode = gfs2_lookup_by_inum(sdp, no_addr, gl->gl_no_formal_ino,
GFS2_BLKST_UNLINKED);
- if (!IS_ERR_OR_NULL(inode)) {
+ if (IS_ERR(inode)) {
+ if (PTR_ERR(inode) == -EAGAIN &&
+ (gfs2_queue_delete_work(gl, 5 * HZ)))
+ return;
+ } else {
d_prune_aliases(inode);
iput(inode);
}
-out:
gfs2_glock_put(gl);
}
}
}
+/**
+ * gfs2_glock_holder_ready - holder is ready and its error code can be collected
+ * @gh: the glock holder
+ *
+ * Called when a glock holder no longer needs to be waited for because it is
+ * now either held (HIF_HOLDER set; gh_error == 0), or acquiring the lock has
+ * failed (gh_error != 0).
+ */
+
+int gfs2_glock_holder_ready(struct gfs2_holder *gh)
+{
+ if (gh->gh_error || (gh->gh_flags & GL_SKIP))
+ return gh->gh_error;
+ gh->gh_error = gfs2_instantiate(gh);
+ if (gh->gh_error)
+ gfs2_glock_dq(gh);
+ return gh->gh_error;
+}
+
/**
* gfs2_glock_wait - wait on a glock acquisition
* @gh: the glock holder
might_sleep();
wait_on_bit(&gh->gh_iflags, HIF_WAIT, TASK_UNINTERRUPTIBLE);
gfs2_glock_update_hold_time(gh->gh_gl, start_time);
- return gh->gh_error;
+ return gfs2_glock_holder_ready(gh);
}
static int glocks_pending(unsigned int num_gh, struct gfs2_holder *ghs)
struct gfs2_sbd *sdp = ghs[0].gh_gl->gl_name.ln_sbd;
int i, ret = 0, timeout = 0;
unsigned long start_time = jiffies;
- bool keep_waiting;
might_sleep();
/*
for (i = 0; i < num_gh; i++)
timeout += ghs[i].gh_gl->gl_hold_time << 1;
-wait_for_dlm:
if (!wait_event_timeout(sdp->sd_async_glock_wait,
- !glocks_pending(num_gh, ghs), timeout))
+ !glocks_pending(num_gh, ghs), timeout)) {
ret = -ESTALE; /* request timed out. */
+ goto out;
+ }
- /*
- * If dlm granted all our requests, we need to adjust the glock
- * minimum hold time values according to how long we waited.
- *
- * If our request timed out, we need to repeatedly release any held
- * glocks we acquired thus far to allow dlm to acquire the remaining
- * glocks without deadlocking. We cannot currently cancel outstanding
- * glock acquisitions.
- *
- * The HIF_WAIT bit tells us which requests still need a response from
- * dlm.
- *
- * If dlm sent us any errors, we return the first error we find.
- */
- keep_waiting = false;
for (i = 0; i < num_gh; i++) {
- /* Skip holders we have already dequeued below. */
- if (!gfs2_holder_queued(&ghs[i]))
- continue;
- /* Skip holders with a pending DLM response. */
- if (test_bit(HIF_WAIT, &ghs[i].gh_iflags)) {
- keep_waiting = true;
- continue;
- }
+ struct gfs2_holder *gh = &ghs[i];
+ int ret2;
- if (test_bit(HIF_HOLDER, &ghs[i].gh_iflags)) {
- if (ret == -ESTALE)
- gfs2_glock_dq(&ghs[i]);
- else
- gfs2_glock_update_hold_time(ghs[i].gh_gl,
- start_time);
+ if (test_bit(HIF_HOLDER, &gh->gh_iflags)) {
+ gfs2_glock_update_hold_time(gh->gh_gl,
+ start_time);
}
+ ret2 = gfs2_glock_holder_ready(gh);
if (!ret)
- ret = ghs[i].gh_error;
+ ret = ret2;
}
- if (keep_waiting)
- goto wait_for_dlm;
+out:
+ if (ret) {
+ for (i = 0; i < num_gh; i++) {
+ struct gfs2_holder *gh = &ghs[i];
- /*
- * At this point, we've either acquired all locks or released them all.
- */
+ gfs2_glock_dq(gh);
+ }
+ }
return ret;
}
va_end(args);
}
+ static inline bool pid_is_meaningful(const struct gfs2_holder *gh)
+ {
+ if (!(gh->gh_flags & GL_NOPID))
+ return true;
+ if (gh->gh_state == LM_ST_UNLOCKED)
+ return true;
+ return false;
+ }
+
/**
* add_to_queue - Add a holder to the wait queue (but look for recursion)
* @gh: the holder structure to add
if (gh->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB)) {
if (test_bit(GLF_LOCK, &gl->gl_flags)) {
- struct gfs2_holder *first_gh;
+ struct gfs2_holder *current_gh;
- first_gh = find_first_strong_holder(gl);
- try_futile = !may_grant(gl, first_gh, gh);
+ current_gh = find_first_strong_holder(gl);
+ try_futile = !may_grant(gl, current_gh, gh);
}
if (test_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags))
goto fail;
}
list_for_each_entry(gh2, &gl->gl_holders, gh_list) {
- if (unlikely(gh2->gh_owner_pid == gh->gh_owner_pid &&
- (gh->gh_gl->gl_ops->go_type != LM_TYPE_FLOCK) &&
- !test_bit(HIF_MAY_DEMOTE, &gh2->gh_iflags)))
- goto trap_recursive;
+ if (likely(gh2->gh_owner_pid != gh->gh_owner_pid))
+ continue;
+ if (gh->gh_gl->gl_ops->go_type == LM_TYPE_FLOCK)
+ continue;
+ if (test_bit(HIF_MAY_DEMOTE, &gh2->gh_iflags))
+ continue;
+ if (!pid_is_meaningful(gh2))
+ continue;
+ goto trap_recursive;
+ }
+ list_for_each_entry(gh2, &gl->gl_holders, gh_list) {
if (try_futile &&
!(gh2->gh_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB))) {
fail:
}
/**
- * nq_m_sync - synchonously acquire more than one glock in deadlock free order
+ * nq_m_sync - synchronously acquire more than one glock in deadlock free order
* @num_gh: the number of structures
* @ghs: an array of struct gfs2_holder structures
* @p: placeholder for the holder structure to pass back
sort(p, num_gh, sizeof(struct gfs2_holder *), glock_compare, NULL);
for (x = 0; x < num_gh; x++) {
- p[x]->gh_flags &= ~(LM_FLAG_TRY | GL_ASYNC);
-
error = gfs2_glock_nq(p[x]);
if (error) {
while (x--)
* @num_gh: the number of structures
* @ghs: an array of struct gfs2_holder structures
*
- *
* Returns: 0 on success (all glocks acquired),
* errno on failure (no glocks acquired)
*/
case 0:
return 0;
case 1:
- ghs->gh_flags &= ~(LM_FLAG_TRY | GL_ASYNC);
return gfs2_glock_nq(ghs);
default:
if (num_gh <= 4)
dump_glock(NULL, gl, true);
}
+static void withdraw_dq(struct gfs2_glock *gl)
+{
+ spin_lock(&gl->gl_lockref.lock);
+ if (!__lockref_is_dead(&gl->gl_lockref) &&
+ glock_blocked_by_withdraw(gl))
+ do_error(gl, LM_OUT_ERROR); /* remove pending waiters */
+ spin_unlock(&gl->gl_lockref.lock);
+}
+
+void gfs2_gl_dq_holders(struct gfs2_sbd *sdp)
+{
+ glock_hash_walk(withdraw_dq, sdp);
+}
+
/**
* gfs2_gl_hash_clear - Empty out the glock hash table
* @sdp: the filesystem
glock_hash_walk(dump_glock_func, sdp);
}
-void gfs2_glock_finish_truncate(struct gfs2_inode *ip)
-{
- struct gfs2_glock *gl = ip->i_gl;
- int ret;
-
- ret = gfs2_truncatei_resume(ip);
- gfs2_glock_assert_withdraw(gl, ret == 0);
-
- spin_lock(&gl->gl_lockref.lock);
- clear_bit(GLF_LOCK, &gl->gl_flags);
- run_queue(gl, 1);
- spin_unlock(&gl->gl_lockref.lock);
-}
-
static const char *state2str(unsigned state)
{
switch(state) {
static void dump_holder(struct seq_file *seq, const struct gfs2_holder *gh,
const char *fs_id_buf)
{
- struct task_struct *gh_owner = NULL;
+ const char *comm = "(none)";
+ pid_t owner_pid = 0;
char flags_buf[32];
rcu_read_lock();
- if (gh->gh_owner_pid)
+ if (pid_is_meaningful(gh)) {
+ struct task_struct *gh_owner;
+
+ comm = "(ended)";
+ owner_pid = pid_nr(gh->gh_owner_pid);
gh_owner = pid_task(gh->gh_owner_pid, PIDTYPE_PID);
+ if (gh_owner)
+ comm = gh_owner->comm;
+ }
gfs2_print_dbg(seq, "%s H: s:%s f:%s e:%d p:%ld [%s] %pS\n",
fs_id_buf, state2str(gh->gh_state),
hflags2str(flags_buf, gh->gh_flags, gh->gh_iflags),
- gh->gh_error,
- gh->gh_owner_pid ? (long)pid_nr(gh->gh_owner_pid) : -1,
- gh_owner ? gh_owner->comm : "(ended)",
- (void *)gh->gh_ip);
+ gh->gh_error, (long)owner_pid, comm, (void *)gh->gh_ip);
rcu_read_unlock();
}
return -ENOMEM;
}
- ret = register_shrinker(&glock_shrinker);
+ ret = register_shrinker(&glock_shrinker, "gfs2-glock");
if (ret) {
destroy_workqueue(gfs2_delete_workqueue);
destroy_workqueue(glock_workqueue);
.release = gfs2_glocks_release,
};
+ struct gfs2_glockfd_iter {
+ struct super_block *sb;
+ unsigned int tgid;
+ struct task_struct *task;
+ unsigned int fd;
+ struct file *file;
+ };
+
+ static struct task_struct *gfs2_glockfd_next_task(struct gfs2_glockfd_iter *i)
+ {
+ struct pid_namespace *ns = task_active_pid_ns(current);
+ struct pid *pid;
+
+ if (i->task)
+ put_task_struct(i->task);
+
+ rcu_read_lock();
+ retry:
+ i->task = NULL;
+ pid = find_ge_pid(i->tgid, ns);
+ if (pid) {
+ i->tgid = pid_nr_ns(pid, ns);
+ i->task = pid_task(pid, PIDTYPE_TGID);
+ if (!i->task) {
+ i->tgid++;
+ goto retry;
+ }
+ get_task_struct(i->task);
+ }
+ rcu_read_unlock();
+ return i->task;
+ }
+
+ static struct file *gfs2_glockfd_next_file(struct gfs2_glockfd_iter *i)
+ {
+ if (i->file) {
+ fput(i->file);
+ i->file = NULL;
+ }
+
+ rcu_read_lock();
+ for(;; i->fd++) {
+ struct inode *inode;
+
+ i->file = task_lookup_next_fd_rcu(i->task, &i->fd);
+ if (!i->file) {
+ i->fd = 0;
+ break;
+ }
+ inode = file_inode(i->file);
+ if (inode->i_sb != i->sb)
+ continue;
+ if (get_file_rcu(i->file))
+ break;
+ }
+ rcu_read_unlock();
+ return i->file;
+ }
+
+ static void *gfs2_glockfd_seq_start(struct seq_file *seq, loff_t *pos)
+ {
+ struct gfs2_glockfd_iter *i = seq->private;
+
+ if (*pos)
+ return NULL;
+ while (gfs2_glockfd_next_task(i)) {
+ if (gfs2_glockfd_next_file(i))
+ return i;
+ i->tgid++;
+ }
+ return NULL;
+ }
+
+ static void *gfs2_glockfd_seq_next(struct seq_file *seq, void *iter_ptr,
+ loff_t *pos)
+ {
+ struct gfs2_glockfd_iter *i = seq->private;
+
+ (*pos)++;
+ i->fd++;
+ do {
+ if (gfs2_glockfd_next_file(i))
+ return i;
+ i->tgid++;
+ } while (gfs2_glockfd_next_task(i));
+ return NULL;
+ }
+
+ static void gfs2_glockfd_seq_stop(struct seq_file *seq, void *iter_ptr)
+ {
+ struct gfs2_glockfd_iter *i = seq->private;
+
+ if (i->file)
+ fput(i->file);
+ if (i->task)
+ put_task_struct(i->task);
+ }
+
+ static void gfs2_glockfd_seq_show_flock(struct seq_file *seq,
+ struct gfs2_glockfd_iter *i)
+ {
+ struct gfs2_file *fp = i->file->private_data;
+ struct gfs2_holder *fl_gh = &fp->f_fl_gh;
+ struct lm_lockname gl_name = { .ln_type = LM_TYPE_RESERVED };
+
+ if (!READ_ONCE(fl_gh->gh_gl))
+ return;
+
+ spin_lock(&i->file->f_lock);
+ if (gfs2_holder_initialized(fl_gh))
+ gl_name = fl_gh->gh_gl->gl_name;
+ spin_unlock(&i->file->f_lock);
+
+ if (gl_name.ln_type != LM_TYPE_RESERVED) {
+ seq_printf(seq, "%d %u %u/%llx\n",
+ i->tgid, i->fd, gl_name.ln_type,
+ (unsigned long long)gl_name.ln_number);
+ }
+ }
+
+ static int gfs2_glockfd_seq_show(struct seq_file *seq, void *iter_ptr)
+ {
+ struct gfs2_glockfd_iter *i = seq->private;
+ struct inode *inode = file_inode(i->file);
+ struct gfs2_glock *gl;
+
+ inode_lock_shared(inode);
+ gl = GFS2_I(inode)->i_iopen_gh.gh_gl;
+ if (gl) {
+ seq_printf(seq, "%d %u %u/%llx\n",
+ i->tgid, i->fd, gl->gl_name.ln_type,
+ (unsigned long long)gl->gl_name.ln_number);
+ }
+ gfs2_glockfd_seq_show_flock(seq, i);
+ inode_unlock_shared(inode);
+ return 0;
+ }
+
+ static const struct seq_operations gfs2_glockfd_seq_ops = {
+ .start = gfs2_glockfd_seq_start,
+ .next = gfs2_glockfd_seq_next,
+ .stop = gfs2_glockfd_seq_stop,
+ .show = gfs2_glockfd_seq_show,
+ };
+
+ static int gfs2_glockfd_open(struct inode *inode, struct file *file)
+ {
+ struct gfs2_glockfd_iter *i;
+ struct gfs2_sbd *sdp = inode->i_private;
+
+ i = __seq_open_private(file, &gfs2_glockfd_seq_ops,
+ sizeof(struct gfs2_glockfd_iter));
+ if (!i)
+ return -ENOMEM;
+ i->sb = sdp->sd_vfs;
+ return 0;
+ }
+
+ static const struct file_operations gfs2_glockfd_fops = {
+ .owner = THIS_MODULE,
+ .open = gfs2_glockfd_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = seq_release_private,
+ };
+
DEFINE_SEQ_ATTRIBUTE(gfs2_sbstats);
void gfs2_create_debugfs_file(struct gfs2_sbd *sdp)
debugfs_create_file("glocks", S_IFREG | S_IRUGO, sdp->debugfs_dir, sdp,
&gfs2_glocks_fops);
+ debugfs_create_file("glockfd", S_IFREG | S_IRUGO, sdp->debugfs_dir, sdp,
+ &gfs2_glockfd_fops);
+
debugfs_create_file("glstats", S_IFREG | S_IRUGO, sdp->debugfs_dir, sdp,
&gfs2_glstats_fops);