1 files changed, 82 insertions, 105 deletions

diff --git a/fs/dlm/user.c b/fs/dlm/user.c
index e5cefa90b1ce..51daf4acbe31 100644
--- a/fs/dlm/user.c
+++ b/fs/dlm/user.c
@@ -16,6 +16,8 @@
 #include <linux/slab.h>
 #include <linux/sched/signal.h>
 
+#include <trace/events/dlm.h>
+
 #include "dlm_internal.h"
 #include "lockspace.h"
 #include "lock.h"
@@ -23,6 +25,7 @@
 #include "user.h"
 #include "ast.h"
 #include "config.h"
+#include "memory.h"
 
 static const char name_prefix[] = "dlm";
 static const struct file_operations device_fops;
@@ -108,11 +111,11 @@ static void compat_input(struct dlm_write_request *kb,
 		kb->i.lock.parent = kb32->i.lock.parent;
 		kb->i.lock.xid = kb32->i.lock.xid;
 		kb->i.lock.timeout = kb32->i.lock.timeout;
-		kb->i.lock.castparam = (void *)(long)kb32->i.lock.castparam;
-		kb->i.lock.castaddr = (void *)(long)kb32->i.lock.castaddr;
-		kb->i.lock.bastparam = (void *)(long)kb32->i.lock.bastparam;
-		kb->i.lock.bastaddr = (void *)(long)kb32->i.lock.bastaddr;
-		kb->i.lock.lksb = (void *)(long)kb32->i.lock.lksb;
+		kb->i.lock.castparam = (__user void *)(long)kb32->i.lock.castparam;
+		kb->i.lock.castaddr = (__user void *)(long)kb32->i.lock.castaddr;
+		kb->i.lock.bastparam = (__user void *)(long)kb32->i.lock.bastparam;
+		kb->i.lock.bastaddr = (__user void *)(long)kb32->i.lock.bastaddr;
+		kb->i.lock.lksb = (__user void *)(long)kb32->i.lock.lksb;
 		memcpy(kb->i.lock.lvb, kb32->i.lock.lvb, DLM_USER_LVB_LEN);
 		memcpy(kb->i.lock.name, kb32->i.lock.name, namelen);
 	}
@@ -127,9 +130,9 @@ static void compat_output(struct dlm_lock_result *res,
 	res32->version[1] = res->version[1];
 	res32->version[2] = res->version[2];
 
-	res32->user_astaddr = (__u32)(long)res->user_astaddr;
-	res32->user_astparam = (__u32)(long)res->user_astparam;
-	res32->user_lksb = (__u32)(long)res->user_lksb;
+	res32->user_astaddr = (__u32)(__force long)res->user_astaddr;
+	res32->user_astparam = (__u32)(__force long)res->user_astparam;
+	res32->user_lksb = (__u32)(__force long)res->user_lksb;
 	res32->bast_mode = res->bast_mode;
 
 	res32->lvb_offset = res->lvb_offset;
@@ -173,18 +176,20 @@ static int lkb_is_endoflife(int mode, int status)
    being removed and then remove that lkb from the orphans list and free it */
 
 void dlm_user_add_ast(struct dlm_lkb *lkb, uint32_t flags, int mode,
-		      int status, uint32_t sbflags, uint64_t seq)
+		      int status, uint32_t sbflags)
 {
 	struct dlm_ls *ls;
 	struct dlm_user_args *ua;
 	struct dlm_user_proc *proc;
-	int rv;
+	struct dlm_callback *cb;
+	int rv, copy_lvb;
 
-	if (lkb->lkb_flags & (DLM_IFL_ORPHAN | DLM_IFL_DEAD))
+	if (test_bit(DLM_DFL_ORPHAN_BIT, &lkb->lkb_dflags) ||
+	    test_bit(DLM_IFL_DEAD_BIT, &lkb->lkb_iflags))
 		return;
 
 	ls = lkb->lkb_resource->res_ls;
-	mutex_lock(&ls->ls_clear_proc_locks);
+	spin_lock_bh(&ls->ls_clear_proc_locks);
 
 	/* If ORPHAN/DEAD flag is set, it means the process is dead so an ast
 	   can't be delivered.  For ORPHAN's, dlm_clear_proc_locks() freed
@@ -192,7 +197,8 @@ void dlm_user_add_ast(struct dlm_lkb *lkb, uint32_t flags, int mode,
 	   for cases where a completion ast is received for an operation that
 	   began before clear_proc_locks did its cancel/unlock. */
 
-	if (lkb->lkb_flags & (DLM_IFL_ORPHAN | DLM_IFL_DEAD))
+	if (test_bit(DLM_DFL_ORPHAN_BIT, &lkb->lkb_dflags) ||
+	    test_bit(DLM_IFL_DEAD_BIT, &lkb->lkb_iflags))
 		goto out;
 
 	DLM_ASSERT(lkb->lkb_ua, dlm_print_lkb(lkb););
@@ -203,34 +209,40 @@ void dlm_user_add_ast(struct dlm_lkb *lkb, uint32_t flags, int mode,
 		goto out;
 
 	if ((flags & DLM_CB_CAST) && lkb_is_endoflife(mode, status))
-		lkb->lkb_flags |= DLM_IFL_ENDOFLIFE;
-
-	spin_lock(&proc->asts_spin);
-
-	rv = dlm_add_lkb_callback(lkb, flags, mode, status, sbflags, seq);
-	if (rv < 0) {
-		spin_unlock(&proc->asts_spin);
-		goto out;
-	}
-
-	if (list_empty(&lkb->lkb_cb_list)) {
-		kref_get(&lkb->lkb_ref);
-		list_add_tail(&lkb->lkb_cb_list, &proc->asts);
-		wake_up_interruptible(&proc->wait);
+		set_bit(DLM_IFL_ENDOFLIFE_BIT, &lkb->lkb_iflags);
+
+	spin_lock_bh(&proc->asts_spin);
+
+	if (!dlm_may_skip_callback(lkb, flags, mode, status, sbflags,
+				   &copy_lvb)) {
+		rv = dlm_get_cb(lkb, flags, mode, status, sbflags, &cb);
+		if (!rv) {
+			cb->copy_lvb = copy_lvb;
+			cb->ua = *ua;
+			cb->lkb_lksb = &cb->ua.lksb;
+			if (copy_lvb) {
+				memcpy(cb->lvbptr, ua->lksb.sb_lvbptr,
+				       DLM_USER_LVB_LEN);
+				cb->lkb_lksb->sb_lvbptr = cb->lvbptr;
+			}
+
+			list_add_tail(&cb->list, &proc->asts);
+			wake_up_interruptible(&proc->wait);
+		}
 	}
-	spin_unlock(&proc->asts_spin);
+	spin_unlock_bh(&proc->asts_spin);
 
-	if (lkb->lkb_flags & DLM_IFL_ENDOFLIFE) {
+	if (test_bit(DLM_IFL_ENDOFLIFE_BIT, &lkb->lkb_iflags)) {
 		/* N.B. spin_lock locks_spin, not asts_spin */
-		spin_lock(&proc->locks_spin);
+		spin_lock_bh(&proc->locks_spin);
 		if (!list_empty(&lkb->lkb_ownqueue)) {
 			list_del_init(&lkb->lkb_ownqueue);
 			dlm_put_lkb(lkb);
 		}
-		spin_unlock(&proc->locks_spin);
+		spin_unlock_bh(&proc->locks_spin);
 	}
  out:
-	mutex_unlock(&ls->ls_clear_proc_locks);
+	spin_unlock_bh(&ls->ls_clear_proc_locks);
 }
 
 static int device_user_lock(struct dlm_user_proc *proc,
@@ -263,22 +275,19 @@ static int device_user_lock(struct dlm_user_proc *proc,
 
 	if (params->flags & DLM_LKF_CONVERT) {
 		error = dlm_user_convert(ls, ua,
-				         params->mode, params->flags,
-				         params->lkid, params->lvb,
-					 (unsigned long) params->timeout);
+					 params->mode, params->flags,
+					 params->lkid, params->lvb);
 	} else if (params->flags & DLM_LKF_ORPHAN) {
 		error = dlm_user_adopt_orphan(ls, ua,
 					 params->mode, params->flags,
 					 params->name, params->namelen,
-					 (unsigned long) params->timeout,
 					 &lkid);
 		if (!error)
 			error = lkid;
 	} else {
 		error = dlm_user_request(ls, ua,
 					 params->mode, params->flags,
-					 params->name, params->namelen,
-					 (unsigned long) params->timeout);
+					 params->name, params->namelen);
 		if (!error)
 			error = ua->lksb.sb_lkid;
 	}
@@ -402,9 +411,9 @@ static int device_create_lockspace(struct dlm_lspace_params *params)
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
-	error = dlm_new_lockspace(params->name, dlm_config.ci_cluster_name, params->flags,
-				  DLM_USER_LVB_LEN, NULL, NULL, NULL,
-				  &lockspace);
+	error = dlm_new_user_lockspace(params->name, dlm_config.ci_cluster_name,
+				       params->flags, DLM_USER_LVB_LEN, NULL,
+				       NULL, NULL, &lockspace);
 	if (error)
 		return error;
 
@@ -416,7 +425,7 @@ static int device_create_lockspace(struct dlm_lspace_params *params)
 	dlm_put_lockspace(ls);
 
 	if (error)
-		dlm_release_lockspace(lockspace, 0);
+		dlm_release_lockspace(lockspace, DLM_RELEASE_NO_LOCKS);
 	else
 		error = ls->ls_device.minor;
 
@@ -427,7 +436,7 @@ static int device_remove_lockspace(struct dlm_lspace_params *params)
 {
 	dlm_lockspace_t *lockspace;
 	struct dlm_ls *ls;
-	int error, force = 0;
+	int error, force = DLM_RELEASE_NO_LOCKS;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
@@ -437,9 +446,9 @@ static int device_remove_lockspace(struct dlm_lspace_params *params)
 		return -ENOENT;
 
 	if (params->flags & DLM_USER_LSFLG_FORCEFREE)
-		force = 2;
+		force = DLM_RELEASE_NORMAL;
 
-	lockspace = ls->ls_local_handle;
+	lockspace = ls;
 	dlm_put_lockspace(ls);
 
 	/* The final dlm_release_lockspace waits for references to go to
@@ -642,7 +651,7 @@ static int device_open(struct inode *inode, struct file *file)
 		return -ENOMEM;
 	}
 
-	proc->lockspace = ls->ls_local_handle;
+	proc->lockspace = ls;
 	INIT_LIST_HEAD(&proc->asts);
 	INIT_LIST_HEAD(&proc->locks);
 	INIT_LIST_HEAD(&proc->unlocking);
@@ -777,11 +786,9 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count,
 			   loff_t *ppos)
 {
 	struct dlm_user_proc *proc = file->private_data;
-	struct dlm_lkb *lkb;
 	DECLARE_WAITQUEUE(wait, current);
-	struct dlm_callback cb;
-	int rv, resid, copy_lvb = 0;
-	int old_mode, new_mode;
+	struct dlm_callback *cb;
+	int rv, ret;
 
 	if (count == sizeof(struct dlm_device_version)) {
 		rv = copy_version_to_user(buf, count);
@@ -800,16 +807,14 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count,
 #endif
 		return -EINVAL;
 
- try_another:
-
 	/* do we really need this? can a read happen after a close? */
 	if (test_bit(DLM_PROC_FLAGS_CLOSING, &proc->flags))
 		return -EINVAL;
 
-	spin_lock(&proc->asts_spin);
+	spin_lock_bh(&proc->asts_spin);
 	if (list_empty(&proc->asts)) {
 		if (file->f_flags & O_NONBLOCK) {
-			spin_unlock(&proc->asts_spin);
+			spin_unlock_bh(&proc->asts_spin);
 			return -EAGAIN;
 		}
 
@@ -818,16 +823,16 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count,
 	repeat:
 		set_current_state(TASK_INTERRUPTIBLE);
 		if (list_empty(&proc->asts) && !signal_pending(current)) {
-			spin_unlock(&proc->asts_spin);
+			spin_unlock_bh(&proc->asts_spin);
 			schedule();
-			spin_lock(&proc->asts_spin);
+			spin_lock_bh(&proc->asts_spin);
 			goto repeat;
 		}
 		set_current_state(TASK_RUNNING);
 		remove_wait_queue(&proc->wait, &wait);
 
 		if (signal_pending(current)) {
-			spin_unlock(&proc->asts_spin);
+			spin_unlock_bh(&proc->asts_spin);
 			return -ERESTARTSYS;
 		}
 	}
@@ -836,53 +841,25 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count,
 	   without removing lkb_cb_list; so empty lkb_cb_list is always
 	   consistent with empty lkb_callbacks */
 
-	lkb = list_entry(proc->asts.next, struct dlm_lkb, lkb_cb_list);
-
-	/* rem_lkb_callback sets a new lkb_last_cast */
-	old_mode = lkb->lkb_last_cast.mode;
-
-	rv = dlm_rem_lkb_callback(lkb->lkb_resource->res_ls, lkb, &cb, &resid);
-	if (rv < 0) {
-		/* this shouldn't happen; lkb should have been removed from
-		   list when resid was zero */
-		log_print("dlm_rem_lkb_callback empty %x", lkb->lkb_id);
-		list_del_init(&lkb->lkb_cb_list);
-		spin_unlock(&proc->asts_spin);
-		/* removes ref for proc->asts, may cause lkb to be freed */
-		dlm_put_lkb(lkb);
-		goto try_another;
+	cb = list_first_entry(&proc->asts, struct dlm_callback, list);
+	list_del(&cb->list);
+	spin_unlock_bh(&proc->asts_spin);
+
+	if (cb->flags & DLM_CB_BAST) {
+		trace_dlm_bast(cb->ls_id, cb->lkb_id, cb->mode, cb->res_name,
+			       cb->res_length);
+	} else if (cb->flags & DLM_CB_CAST) {
+		cb->lkb_lksb->sb_status = cb->sb_status;
+		cb->lkb_lksb->sb_flags = cb->sb_flags;
+		trace_dlm_ast(cb->ls_id, cb->lkb_id, cb->sb_status,
+			      cb->sb_flags, cb->res_name, cb->res_length);
 	}
-	if (!resid)
-		list_del_init(&lkb->lkb_cb_list);
-	spin_unlock(&proc->asts_spin);
-
-	if (cb.flags & DLM_CB_SKIP) {
-		/* removes ref for proc->asts, may cause lkb to be freed */
-		if (!resid)
-			dlm_put_lkb(lkb);
-		goto try_another;
-	}
-
-	if (cb.flags & DLM_CB_CAST) {
-		new_mode = cb.mode;
-
-		if (!cb.sb_status && lkb->lkb_lksb->sb_lvbptr &&
-		    dlm_lvb_operations[old_mode + 1][new_mode + 1])
-			copy_lvb = 1;
-
-		lkb->lkb_lksb->sb_status = cb.sb_status;
-		lkb->lkb_lksb->sb_flags = cb.sb_flags;
-	}
-
-	rv = copy_result_to_user(lkb->lkb_ua,
-				 test_bit(DLM_PROC_FLAGS_COMPAT, &proc->flags),
-				 cb.flags, cb.mode, copy_lvb, buf, count);
-
-	/* removes ref for proc->asts, may cause lkb to be freed */
-	if (!resid)
-		dlm_put_lkb(lkb);
 
-	return rv;
+	ret = copy_result_to_user(&cb->ua,
+				  test_bit(DLM_PROC_FLAGS_COMPAT, &proc->flags),
+				  cb->flags, cb->mode, cb->copy_lvb, buf, count);
+	dlm_free_cb(cb);
+	return ret;
 }
 
 static __poll_t device_poll(struct file *file, poll_table *wait)
@@ -891,12 +868,12 @@ static __poll_t device_poll(struct file *file, poll_table *wait)
 
 	poll_wait(file, &proc->wait, wait);
 
-	spin_lock(&proc->asts_spin);
+	spin_lock_bh(&proc->asts_spin);
 	if (!list_empty(&proc->asts)) {
-		spin_unlock(&proc->asts_spin);
+		spin_unlock_bh(&proc->asts_spin);
 		return EPOLLIN | EPOLLRDNORM;
 	}
-	spin_unlock(&proc->asts_spin);
+	spin_unlock_bh(&proc->asts_spin);
 	return 0;
 }