X-Git-Url: https://git.whamcloud.com/?p=fs%2Flustre-release.git;a=blobdiff_plain;f=lustre%2Fllite%2Fnamei.c;h=69da17eaa74b574d28e346286ed7ed63609503bf;hp=1d39dfddd91365f6a19177b288b00f142a7e4d3c;hb=8d161d44214f907a9f2d9cf5a79cd2c83de995c3;hpb=08aa217ce49aba1ded52e0f7adb8a607035123fd

diff --git a/lustre/llite/namei.c b/lustre/llite/namei.c
index 1d39dfd..69da17e 100644
--- a/lustre/llite/namei.c
+++ b/lustre/llite/namei.c
@@ -27,7 +27,7 @@
  * Copyright (c) 2002, 2010, Oracle and/or its affiliates. All rights reserved.
  * Use is subject to license terms.
  *
- * Copyright (c) 2011, 2012, Intel Corporation.
+ * Copyright (c) 2011, 2013, Intel Corporation.
  */
 /*
  * This file is part of Lustre, http://www.lustre.org/
@@ -59,7 +59,7 @@ static int ll_create_it(struct inode *, struct dentry *,
  * In such a case there would always be dentry present.
  */
 static int ll_d_mountpoint(struct dentry *dparent, struct dentry *dchild,
-                           struct qstr *name)
+			   struct qstr *name)
 {
         int mounted = 0;
 
@@ -75,16 +75,6 @@ static int ll_d_mountpoint(struct dentry *dparent, struct dentry *dchild,
         return mounted;
 }
 
-int ll_unlock(__u32 mode, struct lustre_handle *lockh)
-{
-        ENTRY;
-
-        ldlm_lock_decref(lockh, mode);
-
-        RETURN(0);
-}
-
-
 /* called from iget5_locked->find_inode() under inode_lock spinlock */
 static int ll_test_inode(struct inode *inode, void *opaque)
 {
@@ -131,47 +121,58 @@ static int ll_set_inode(struct inode *inode, void *opaque)
 }
 
 
-/*
- * Get an inode by inode number (already instantiated by the intent lookup).
- * Returns inode or NULL
+/**
+ * Get an inode by inode number(@hash), which is already instantiated by
+ * the intent lookup).
  */
 struct inode *ll_iget(struct super_block *sb, ino_t hash,
                       struct lustre_md *md)
 {
-        struct inode         *inode;
-        ENTRY;
+	struct inode	*inode;
+	int		rc = 0;
+
+	ENTRY;
 
         LASSERT(hash != 0);
         inode = iget5_locked(sb, hash, ll_test_inode, ll_set_inode, md);
+	if (inode == NULL)
+		RETURN(ERR_PTR(-ENOMEM));
+
+	if (inode->i_state & I_NEW) {
+		rc = ll_read_inode2(inode, md);
+		if (rc == 0 && S_ISREG(inode->i_mode) &&
+		    ll_i2info(inode)->lli_clob == NULL) {
+			CDEBUG(D_INODE, "%s: apply lsm %p to inode "DFID"\n",
+				ll_get_fsname(sb, NULL, 0), md->lsm,
+				PFID(ll_inode2fid(inode)));
+			rc = cl_file_inode_init(inode, md);
+		}
+		if (rc != 0) {
+			make_bad_inode(inode);
+			unlock_new_inode(inode);
+			iput(inode);
+			inode = ERR_PTR(rc);
+		} else {
+			unlock_new_inode(inode);
+		}
+	} else if (!(inode->i_state & (I_FREEING | I_CLEAR))) {
+		rc = ll_update_inode(inode, md);
+		CDEBUG(D_VFSTRACE, "got inode: "DFID"(%p): rc = %d\n",
+		       PFID(&md->body->fid1), inode, rc);
+		if (rc != 0) {
+			make_bad_inode(inode);
+			iput(inode);
+			inode = ERR_PTR(rc);
+		}
+	}
 
-        if (inode) {
-                if (inode->i_state & I_NEW) {
-                        int rc = 0;
-
-                        ll_read_inode2(inode, md);
-                        if (S_ISREG(inode->i_mode) &&
-                            ll_i2info(inode)->lli_clob == NULL)
-                                rc = cl_file_inode_init(inode, md);
-                        if (rc != 0) {
-                                md->lsm = NULL;
-                                make_bad_inode(inode);
-                                unlock_new_inode(inode);
-                                iput(inode);
-                                inode = ERR_PTR(rc);
-                        } else
-                                unlock_new_inode(inode);
-                } else if (!(inode->i_state & (I_FREEING | I_CLEAR)))
-                        ll_update_inode(inode, md);
-                CDEBUG(D_VFSTRACE, "got inode: %p for "DFID"\n",
-                       inode, PFID(&md->body->fid1));
-        }
         RETURN(inode);
 }
 
 static void ll_invalidate_negative_children(struct inode *dir)
 {
 	struct dentry *dentry, *tmp_subdir;
-	struct ll_d_hlist_node *p;
+	DECLARE_LL_D_HLIST_NODE_PTR(p);
 
 	ll_lock_dcache(dir);
 	ll_d_hlist_for_each_entry(dentry, p, &dir->i_dentry, d_alias) {
@@ -183,7 +184,7 @@ static void ll_invalidate_negative_children(struct inode *dir)
 						 &dentry->d_subdirs,
 						 d_u.d_child) {
 				if (child->d_inode == NULL)
-					d_lustre_invalidate(child);
+					d_lustre_invalidate(child, 1);
 			}
 		}
 		spin_unlock(&dentry->d_lock);
@@ -191,111 +192,164 @@ static void ll_invalidate_negative_children(struct inode *dir)
 	ll_unlock_dcache(dir);
 }
 
+int ll_test_inode_by_fid(struct inode *inode, void *opaque)
+{
+	return lu_fid_eq(&ll_i2info(inode)->lli_fid, opaque);
+}
+
 int ll_md_blocking_ast(struct ldlm_lock *lock, struct ldlm_lock_desc *desc,
-                       void *data, int flag)
+		       void *data, int flag)
 {
-        int rc;
-        struct lustre_handle lockh;
-        ENTRY;
+	struct lustre_handle lockh;
+	int rc;
+	ENTRY;
 
-        switch (flag) {
-        case LDLM_CB_BLOCKING:
-                ldlm_lock2handle(lock, &lockh);
-                rc = ldlm_cli_cancel(&lockh);
-                if (rc < 0) {
-                        CDEBUG(D_INODE, "ldlm_cli_cancel: %d\n", rc);
-                        RETURN(rc);
-                }
-                break;
-        case LDLM_CB_CANCELING: {
-                struct inode *inode = ll_inode_from_lock(lock);
-                struct ll_inode_info *lli;
-                __u64 bits = lock->l_policy_data.l_inodebits.bits;
-                struct lu_fid *fid;
-                ldlm_mode_t mode = lock->l_req_mode;
-
-                /* Invalidate all dentries associated with this inode */
-                if (inode == NULL)
-                        break;
-
-                LASSERT(lock->l_flags & LDLM_FL_CANCELING);
-                /* For OPEN locks we differentiate between lock modes
+	switch (flag) {
+	case LDLM_CB_BLOCKING:
+		ldlm_lock2handle(lock, &lockh);
+		rc = ldlm_cli_cancel(&lockh, LCF_ASYNC);
+		if (rc < 0) {
+			CDEBUG(D_INODE, "ldlm_cli_cancel: rc = %d\n", rc);
+			RETURN(rc);
+		}
+		break;
+	case LDLM_CB_CANCELING: {
+		struct inode *inode = ll_inode_from_resource_lock(lock);
+		__u64 bits = lock->l_policy_data.l_inodebits.bits;
+
+		/* Inode is set to lock->l_resource->lr_lvb_inode
+		 * for mdc - bug 24555 */
+		LASSERT(lock->l_ast_data == NULL);
+
+		if (inode == NULL)
+			break;
+
+		/* Invalidate all dentries associated with this inode */
+		LASSERT(ldlm_is_canceling(lock));
+
+		if (!fid_res_name_eq(ll_inode2fid(inode),
+				     &lock->l_resource->lr_name)) {
+			LDLM_ERROR(lock, "data mismatch with object "DFID"(%p)",
+				   PFID(ll_inode2fid(inode)), inode);
+			LBUG();
+		}
+
+		if (bits & MDS_INODELOCK_XATTR) {
+			ll_xattr_cache_destroy(inode);
+			bits &= ~MDS_INODELOCK_XATTR;
+		}
+
+		/* For OPEN locks we differentiate between lock modes
 		 * LCK_CR, LCK_CW, LCK_PR - bug 22891 */
-		if (bits & (MDS_INODELOCK_LOOKUP | MDS_INODELOCK_UPDATE |
-			    MDS_INODELOCK_LAYOUT))
-                        ll_have_md_lock(inode, &bits, LCK_MINMODE);
-
-                if (bits & MDS_INODELOCK_OPEN)
-                        ll_have_md_lock(inode, &bits, mode);
-
-                fid = ll_inode2fid(inode);
-                if (lock->l_resource->lr_name.name[0] != fid_seq(fid) ||
-                    lock->l_resource->lr_name.name[1] != fid_oid(fid) ||
-                    lock->l_resource->lr_name.name[2] != fid_ver(fid)) {
-                        LDLM_ERROR(lock, "data mismatch with object "
-                                   DFID" (%p)", PFID(fid), inode);
-                }
+		if (bits & MDS_INODELOCK_OPEN)
+			ll_have_md_lock(inode, &bits, lock->l_req_mode);
+
+		if (bits & MDS_INODELOCK_OPEN) {
+			fmode_t fmode;
+
+			switch (lock->l_req_mode) {
+			case LCK_CW:
+				fmode = FMODE_WRITE;
+				break;
+			case LCK_PR:
+				fmode = FMODE_EXEC;
+				break;
+			case LCK_CR:
+				fmode = FMODE_READ;
+				break;
+			default:
+				LDLM_ERROR(lock, "bad lock mode for OPEN lock");
+				LBUG();
+			}
 
-                if (bits & MDS_INODELOCK_OPEN) {
-                        int flags = 0;
-                        switch (lock->l_req_mode) {
-                        case LCK_CW:
-                                flags = FMODE_WRITE;
-                                break;
-                        case LCK_PR:
-                                flags = FMODE_EXEC;
-                                break;
-                        case LCK_CR:
-                                flags = FMODE_READ;
-                                break;
-                        default:
-                                CERROR("Unexpected lock mode for OPEN lock "
-                                       "%d, inode %ld\n", lock->l_req_mode,
-                                       inode->i_ino);
-                        }
-                        ll_md_real_close(inode, flags);
-                }
+			ll_md_real_close(inode, fmode);
+
+			bits &= ~MDS_INODELOCK_OPEN;
+		}
+
+		if (bits & (MDS_INODELOCK_LOOKUP | MDS_INODELOCK_UPDATE |
+			    MDS_INODELOCK_LAYOUT | MDS_INODELOCK_PERM))
+			ll_have_md_lock(inode, &bits, LCK_MINMODE);
 
-		lli = ll_i2info(inode);
 		if (bits & MDS_INODELOCK_LAYOUT) {
-			struct cl_object_conf conf = { .coc_inode = inode,
-						       .coc_invalidate = true };
+			struct cl_object_conf conf = {
+				.coc_opc = OBJECT_CONF_INVALIDATE,
+				.coc_inode = inode,
+			};
+
 			rc = ll_layout_conf(inode, &conf);
-			if (rc)
-				CDEBUG(D_INODE, "invaliding layout %d.\n", rc);
+			if (rc < 0)
+				CDEBUG(D_INODE, "cannot invalidate layout of "
+				       DFID": rc = %d\n",
+				       PFID(ll_inode2fid(inode)), rc);
 		}
 
-                if (bits & MDS_INODELOCK_UPDATE)
-                        lli->lli_flags &= ~LLIF_MDS_SIZE_LOCK;
+		if (bits & MDS_INODELOCK_UPDATE) {
+			struct ll_inode_info *lli = ll_i2info(inode);
 
-                if (S_ISDIR(inode->i_mode) &&
-                     (bits & MDS_INODELOCK_UPDATE)) {
-                        CDEBUG(D_INODE, "invalidating inode %lu\n",
-                               inode->i_ino);
-                        truncate_inode_pages(inode->i_mapping, 0);
-			ll_invalidate_negative_children(inode);
+			spin_lock(&lli->lli_lock);
+			lli->lli_flags &= ~LLIF_MDS_SIZE_LOCK;
+			spin_unlock(&lli->lli_lock);
+		}
+
+		if ((bits & MDS_INODELOCK_UPDATE) && S_ISDIR(inode->i_mode)) {
+			struct ll_inode_info *lli = ll_i2info(inode);
+
+			CDEBUG(D_INODE, "invalidating inode "DFID" lli = %p, "
+			       "pfid  = "DFID"\n", PFID(ll_inode2fid(inode)),
+			       lli, PFID(&lli->lli_pfid));
+			truncate_inode_pages(inode->i_mapping, 0);
+
+			if (unlikely(!fid_is_zero(&lli->lli_pfid))) {
+				struct inode *master_inode = NULL;
+				unsigned long hash;
+
+				/* This is slave inode, since all of the child
+				 * dentry is connected on the master inode, so
+				 * we have to invalidate the negative children
+				 * on master inode */
+				CDEBUG(D_INODE, "Invalidate s"DFID" m"DFID"\n",
+				       PFID(ll_inode2fid(inode)),
+				       PFID(&lli->lli_pfid));
+
+				hash = cl_fid_build_ino(&lli->lli_pfid,
+					ll_need_32bit_api(ll_i2sbi(inode)));
+
+				master_inode = ilookup5(inode->i_sb, hash,
+							ll_test_inode_by_fid,
+							(void *)&lli->lli_pfid);
+				if (master_inode != NULL &&
+					!IS_ERR(master_inode)) {
+					ll_invalidate_negative_children(
+								master_inode);
+					iput(master_inode);
+				}
+			} else {
+				ll_invalidate_negative_children(inode);
+			}
 		}
 
-		if (inode->i_sb->s_root &&
-		    inode != inode->i_sb->s_root->d_inode &&
-		    (bits & MDS_INODELOCK_LOOKUP))
+		if ((bits & (MDS_INODELOCK_LOOKUP | MDS_INODELOCK_PERM)) &&
+		    inode->i_sb->s_root != NULL &&
+		    inode != inode->i_sb->s_root->d_inode)
 			ll_invalidate_aliases(inode);
-                iput(inode);
-                break;
-        }
-        default:
-                LBUG();
-        }
 
-        RETURN(0);
+		iput(inode);
+		break;
+	}
+	default:
+		LBUG();
+	}
+
+	RETURN(0);
 }
 
 __u32 ll_i2suppgid(struct inode *i)
 {
-        if (cfs_curproc_is_in_groups(i->i_gid))
-                return (__u32)i->i_gid;
-        else
-                return (__u32)(-1);
+	if (in_group_p(i->i_gid))
+		return (__u32)from_kgid(&init_user_ns, i->i_gid);
+	else
+		return (__u32) __kgid_val(INVALID_GID);
 }
 
 /* Pack the required supplementary groups into the supplied groups array.
@@ -304,35 +358,15 @@ __u32 ll_i2suppgid(struct inode *i)
  * array in case it might be useful.  Not needed if doing an MDS-side upcall. */
 void ll_i2gids(__u32 *suppgids, struct inode *i1, struct inode *i2)
 {
-#if 0
-        int i;
-#endif
+	LASSERT(i1 != NULL);
+	LASSERT(suppgids != NULL);
 
-        LASSERT(i1 != NULL);
-        LASSERT(suppgids != NULL);
+	suppgids[0] = ll_i2suppgid(i1);
 
-        suppgids[0] = ll_i2suppgid(i1);
-
-        if (i2)
-                suppgids[1] = ll_i2suppgid(i2);
-                else
-                        suppgids[1] = -1;
-
-#if 0
-        for (i = 0; i < current_ngroups; i++) {
-                if (suppgids[0] == -1) {
-                        if (current_groups[i] != suppgids[1])
-                                suppgids[0] = current_groups[i];
-                        continue;
-                }
-                if (suppgids[1] == -1) {
-                        if (current_groups[i] != suppgids[0])
-                                suppgids[1] = current_groups[i];
-                        continue;
-                }
-                break;
-        }
-#endif
+	if (i2)
+		suppgids[1] = ll_i2suppgid(i2);
+	else
+		suppgids[1] = -1;
 }
 
 /*
@@ -346,7 +380,7 @@ void ll_i2gids(__u32 *suppgids, struct inode *i1, struct inode *i2)
 static struct dentry *ll_find_alias(struct inode *inode, struct dentry *dentry)
 {
 	struct dentry *alias, *discon_alias, *invalid_alias;
-	struct ll_d_hlist_node *p;
+	DECLARE_LL_D_HLIST_NODE_PTR(p);
 
 	if (ll_d_hlist_empty(&inode->i_dentry))
 		return NULL;
@@ -390,36 +424,40 @@ static struct dentry *ll_find_alias(struct inode *inode, struct dentry *dentry)
 struct dentry *ll_splice_alias(struct inode *inode, struct dentry *de)
 {
 	struct dentry *new;
+	int rc;
 
 	if (inode) {
 		new = ll_find_alias(inode, de);
 		if (new) {
-			ll_dops_init(new, 1, 1);
+			rc = ll_d_init(new);
+			if (rc < 0) {
+				dput(new);
+				return ERR_PTR(rc);
+			}
 			d_move(new, de);
 			iput(inode);
 			CDEBUG(D_DENTRY,
 			       "Reuse dentry %p inode %p refc %d flags %#x\n",
-			      new, new->d_inode, d_refcount(new), new->d_flags);
+			      new, new->d_inode, d_count(new), new->d_flags);
 			return new;
 		}
 	}
-	ll_dops_init(de, 1, 1);
-	__d_lustre_invalidate(de);
+	rc = ll_d_init(de);
+	if (rc < 0)
+		return ERR_PTR(rc);
 	d_add(de, inode);
 	CDEBUG(D_DENTRY, "Add dentry %p inode %p refc %d flags %#x\n",
-	       de, de->d_inode, d_refcount(de), de->d_flags);
+	       de, de->d_inode, d_count(de), de->d_flags);
         return de;
 }
 
-int ll_lookup_it_finish(struct ptlrpc_request *request,
-                        struct lookup_intent *it, void *data)
+static int ll_lookup_it_finish(struct ptlrpc_request *request,
+			       struct lookup_intent *it,
+			       struct inode *parent, struct dentry **de)
 {
-        struct it_cb_data *icbd = data;
-        struct dentry **de = icbd->icbd_childp;
-        struct inode *parent = icbd->icbd_parent;
-        struct inode *inode = NULL;
-	__u64 bits = 0;
-	int rc;
+	struct inode		 *inode = NULL;
+	__u64			  bits = 0;
+	int			  rc;
 	ENTRY;
 
 	/* NB 1 request reference will be taken away by ll_intent_lock()
@@ -427,7 +465,7 @@ int ll_lookup_it_finish(struct ptlrpc_request *request,
 	CDEBUG(D_DENTRY, "it %p it_disposition %x\n", it,
 	       it->d.lustre.it_disposition);
 	if (!it_disposition(it, DISP_LOOKUP_NEG)) {
-                rc = ll_prep_inode(&inode, request, (*de)->d_sb);
+                rc = ll_prep_inode(&inode, request, (*de)->d_sb, it);
                 if (rc)
                         RETURN(rc);
 
@@ -446,8 +484,20 @@ int ll_lookup_it_finish(struct ptlrpc_request *request,
 	/* Only hash *de if it is unhashed (new dentry).
 	 * Atoimc_open may passin hashed dentries for open.
 	 */
-	if (d_unhashed(*de))
-		*de = ll_splice_alias(inode, *de);
+	if (d_unhashed(*de)) {
+		struct dentry *alias;
+
+		alias = ll_splice_alias(inode, *de);
+		if (IS_ERR(alias))
+			RETURN(PTR_ERR(alias));
+		*de = alias;
+	} else if (!it_disposition(it, DISP_LOOKUP_NEG)  &&
+		   !it_disposition(it, DISP_OPEN_CREATE)) {
+		/* With DISP_OPEN_CREATE dentry will
+		   instantiated in ll_create_it. */
+		LASSERT((*de)->d_inode == NULL);
+		d_instantiate(*de, inode);
+	}
 
 	if (!it_disposition(it, DISP_LOOKUP_NEG)) {
 		/* we have lookup look - unhide dentry */
@@ -480,7 +530,6 @@ static struct dentry *ll_lookup_it(struct inode *parent, struct dentry *dentry,
         struct dentry *save = dentry, *retval;
         struct ptlrpc_request *req = NULL;
         struct md_op_data *op_data;
-        struct it_cb_data icbd;
         __u32 opc;
         int rc;
         ENTRY;
@@ -488,24 +537,15 @@ static struct dentry *ll_lookup_it(struct inode *parent, struct dentry *dentry,
         if (dentry->d_name.len > ll_i2sbi(parent)->ll_namelen)
                 RETURN(ERR_PTR(-ENAMETOOLONG));
 
-        CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s,dir=%lu/%u(%p),intent=%s\n",
-               dentry->d_name.len, dentry->d_name.name, parent->i_ino,
-               parent->i_generation, parent, LL_IT2STR(it));
+	CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p), intent=%s\n",
+	       dentry->d_name.len, dentry->d_name.name,
+	       PFID(ll_inode2fid(parent)), parent, LL_IT2STR(it));
 
         if (d_mountpoint(dentry))
                 CERROR("Tell Peter, lookup on mtpt, it %s\n", LL_IT2STR(it));
 
-        ll_frob_intent(&it, &lookup_it);
-
-        /* As do_lookup is called before follow_mount, root dentry may be left
-         * not valid, revalidate it here. */
-        if (parent->i_sb->s_root && (parent->i_sb->s_root->d_inode == parent) &&
-            (it->it_op & (IT_OPEN | IT_CREAT))) {
-                rc = ll_inode_revalidate_it(parent->i_sb->s_root, it,
-                                            MDS_INODELOCK_LOOKUP);
-                if (rc)
-                        RETURN(ERR_PTR(rc));
-        }
+	if (it == NULL || it->it_op == IT_GETXATTR)
+		it = &lookup_it;
 
         if (it->it_op == IT_GETATTR) {
                 rc = ll_statahead_enter(parent, &dentry, 0);
@@ -516,14 +556,10 @@ static struct dentry *ll_lookup_it(struct inode *parent, struct dentry *dentry,
                 }
         }
 
-        icbd.icbd_childp = &dentry;
-        icbd.icbd_parent = parent;
-
-        if (it->it_op & IT_CREAT ||
-            (it->it_op & IT_OPEN && it->it_create_mode & O_CREAT))
-                opc = LUSTRE_OPC_CREATE;
-        else
-                opc = LUSTRE_OPC_ANY;
+	if (it->it_op & IT_CREAT)
+		opc = LUSTRE_OPC_CREATE;
+	else
+		opc = LUSTRE_OPC_ANY;
 
         op_data = ll_prep_md_op_data(NULL, parent, NULL, dentry->d_name.name,
                                      dentry->d_name.len, lookup_flags, opc,
@@ -533,15 +569,15 @@ static struct dentry *ll_lookup_it(struct inode *parent, struct dentry *dentry,
 
         /* enforce umask if acl disabled or MDS doesn't support umask */
         if (!IS_POSIXACL(parent) || !exp_connect_umask(ll_i2mdexp(parent)))
-                it->it_create_mode &= ~cfs_curproc_umask();
+		it->it_create_mode &= ~current_umask();
 
-        rc = md_intent_lock(ll_i2mdexp(parent), op_data, NULL, 0, it,
-                            lookup_flags, &req, ll_md_blocking_ast, 0);
+	rc = md_intent_lock(ll_i2mdexp(parent), op_data, it, &req,
+			    &ll_md_blocking_ast, 0);
         ll_finish_md_op_data(op_data);
         if (rc < 0)
                 GOTO(out, retval = ERR_PTR(rc));
 
-        rc = ll_lookup_it_finish(req, it, &icbd);
+	rc = ll_lookup_it_finish(req, it, parent, &dentry);
         if (rc != 0) {
                 ll_intent_release(it);
                 GOTO(out, retval = ERR_PTR(rc));
@@ -573,17 +609,13 @@ static struct dentry *ll_lookup_nd(struct inode *parent, struct dentry *dentry,
 	struct lookup_intent *itp, it = { .it_op = IT_GETATTR };
 	struct dentry *de;
 
-	CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s,dir=%lu/%u(%p),flags=%u\n",
-	       dentry->d_name.len, dentry->d_name.name, parent->i_ino,
-	       parent->i_generation, parent, flags);
+	CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p), flags=%u\n",
+	       dentry->d_name.len, dentry->d_name.name,
+	       PFID(ll_inode2fid(parent)), parent, flags);
 
 	/* Optimize away (CREATE && !OPEN). Let .create handle the race. */
-	if ((flags & LOOKUP_CREATE ) && !(flags & LOOKUP_OPEN)) {
-		ll_dops_init(dentry, 1, 1);
-		__d_lustre_invalidate(dentry);
-		d_add(dentry, NULL);
+	if ((flags & LOOKUP_CREATE) && !(flags & LOOKUP_OPEN))
 		return NULL;
-	}
 
 	if (flags & (LOOKUP_PARENT|LOOKUP_OPEN|LOOKUP_CREATE))
 		itp = NULL;
@@ -611,17 +643,17 @@ static int ll_atomic_open(struct inode *dir, struct dentry *dentry,
 	int rc = 0;
 	ENTRY;
 
-	CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s,dir=%lu/%u(%p),file %p,"
-			   "open_flags %x,mode %x opened %d\n",
-	       dentry->d_name.len, dentry->d_name.name, dir->i_ino,
-	       dir->i_generation, dir, file, open_flags, mode, *opened);
+	CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p), file %p,"
+			   "open_flags %x, mode %x opened %d\n",
+	       dentry->d_name.len, dentry->d_name.name,
+	       PFID(ll_inode2fid(dir)), dir, file, open_flags, mode, *opened);
 
 	OBD_ALLOC(it, sizeof(*it));
 	if (!it)
 		RETURN(-ENOMEM);
 
 	it->it_op = IT_OPEN;
-	if (mode) {
+	if (open_flags & O_CREAT) {
 		it->it_op |= IT_CREAT;
 		lookup_flags |= LOOKUP_CREATE;
 	}
@@ -681,8 +713,8 @@ out_release:
 }
 
 #else /* !HAVE_IOP_ATOMIC_OPEN */
-struct lookup_intent *ll_convert_intent(struct open_intent *oit,
-                                        int lookup_flags)
+static struct lookup_intent *
+ll_convert_intent(struct open_intent *oit, int lookup_flags)
 {
 	struct lookup_intent *it;
 
@@ -716,14 +748,10 @@ static struct dentry *ll_lookup_nd(struct inode *parent, struct dentry *dentry,
                         it = ll_d2d(dentry)->lld_it;
                         ll_d2d(dentry)->lld_it = NULL;
                 } else {
-                        if ((nd->flags & LOOKUP_CREATE ) && !(nd->flags & LOOKUP_OPEN)) {
-                                /* We are sure this is new dentry, so we need to create
-                                   our private data and set the dentry ops */ 
-                                ll_dops_init(dentry, 1, 1);
-				__d_lustre_invalidate(dentry);
-				d_add(dentry, NULL);
+			if ((nd->flags & LOOKUP_CREATE) &&
+			    !(nd->flags & LOOKUP_OPEN))
                                 RETURN(NULL);
-                        }
+
                         it = ll_convert_intent(&nd->intent.open, nd->flags);
                         if (IS_ERR(it))
                                 RETURN((struct dentry *)it);
@@ -778,10 +806,7 @@ static struct dentry *ll_lookup_nd(struct inode *parent, struct dentry *dentry,
 #endif /* HAVE_IOP_ATOMIC_OPEN */
 
 /* We depend on "mode" being set with the proper file type/umask by now */
-static struct inode *ll_create_node(struct inode *dir, const char *name,
-                                    int namelen, const void *data, int datalen,
-                                    int mode, __u64 extra,
-                                    struct lookup_intent *it)
+static struct inode *ll_create_node(struct inode *dir, struct lookup_intent *it)
 {
         struct inode *inode = NULL;
         struct ptlrpc_request *request = NULL;
@@ -794,7 +819,7 @@ static struct inode *ll_create_node(struct inode *dir, const char *name,
         LASSERT(it_disposition(it, DISP_ENQ_CREATE_REF));
         request = it->d.lustre.it_data;
         it_clear_disposition(it, DISP_ENQ_CREATE_REF);
-        rc = ll_prep_inode(&inode, request, dir->i_sb);
+        rc = ll_prep_inode(&inode, request, dir->i_sb, it);
         if (rc)
                 GOTO(out, inode = ERR_PTR(rc));
 
@@ -803,8 +828,8 @@ static struct inode *ll_create_node(struct inode *dir, const char *name,
         /* We asked for a lock on the directory, but were granted a
          * lock on the inode.  Since we finally have an inode pointer,
          * stuff it in the lock. */
-        CDEBUG(D_DLMTRACE, "setting l_ast_data to inode %p (%lu/%u)\n",
-               inode, inode->i_ino, inode->i_generation);
+	CDEBUG(D_DLMTRACE, "setting l_ast_data to inode "DFID"(%p)\n",
+	       PFID(ll_inode2fid(inode)), inode);
         ll_set_lock_data(sbi->ll_md_exp, inode, it, NULL);
         EXIT;
  out:
@@ -827,31 +852,29 @@ static struct inode *ll_create_node(struct inode *dir, const char *name,
  * with d_instantiate().
  */
 static int ll_create_it(struct inode *dir, struct dentry *dentry, int mode,
-                        struct lookup_intent *it)
+			struct lookup_intent *it)
 {
-        struct inode *inode;
-        int rc = 0;
-        ENTRY;
+	struct inode *inode;
+	int rc = 0;
+	ENTRY;
 
-        CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s,dir=%lu/%u(%p),intent=%s\n",
-               dentry->d_name.len, dentry->d_name.name, dir->i_ino,
-               dir->i_generation, dir, LL_IT2STR(it));
+	CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p), intent=%s\n",
+	       dentry->d_name.len, dentry->d_name.name,
+	       PFID(ll_inode2fid(dir)), dir, LL_IT2STR(it));
 
-        rc = it_open_error(DISP_OPEN_CREATE, it);
-        if (rc)
-                RETURN(rc);
+	rc = it_open_error(DISP_OPEN_CREATE, it);
+	if (rc)
+		RETURN(rc);
 
-        inode = ll_create_node(dir, dentry->d_name.name, dentry->d_name.len,
-                               NULL, 0, mode, 0, it);
-        if (IS_ERR(inode))
-                RETURN(PTR_ERR(inode));
+	inode = ll_create_node(dir, it);
+	if (IS_ERR(inode))
+		RETURN(PTR_ERR(inode));
 
-        d_instantiate(dentry, inode);
-        RETURN(0);
+	d_instantiate(dentry, inode);
+	RETURN(0);
 }
 
-static void ll_update_times(struct ptlrpc_request *request,
-                            struct inode *inode)
+void ll_update_times(struct ptlrpc_request *request, struct inode *inode)
 {
         struct mdt_body *body = req_capsule_server_get(&request->rq_pill,
                                                        &RMF_MDT_BODY);
@@ -859,8 +882,9 @@ static void ll_update_times(struct ptlrpc_request *request,
         LASSERT(body);
         if (body->valid & OBD_MD_FLMTIME &&
             body->mtime > LTIME_S(inode->i_mtime)) {
-                CDEBUG(D_INODE, "setting ino %lu mtime from %lu to "LPU64"\n",
-                       inode->i_ino, LTIME_S(inode->i_mtime), body->mtime);
+		CDEBUG(D_INODE, "setting fid "DFID" mtime from %lu to "LPU64
+				"\n", PFID(ll_inode2fid(inode)),
+				LTIME_S(inode->i_mtime), body->mtime);
                 LTIME_S(inode->i_mtime) = body->mtime;
         }
         if (body->valid & OBD_MD_FLCTIME &&
@@ -888,17 +912,18 @@ static int ll_new_node(struct inode *dir, struct qstr *name,
         if (IS_ERR(op_data))
                 GOTO(err_exit, err = PTR_ERR(op_data));
 
-        err = md_create(sbi->ll_md_exp, op_data, tgt, tgt_len, mode,
-                        cfs_curproc_fsuid(), cfs_curproc_fsgid(),
-                        cfs_curproc_cap_pack(), rdev, &request);
-        ll_finish_md_op_data(op_data);
+	err = md_create(sbi->ll_md_exp, op_data, tgt, tgt_len, mode,
+			from_kuid(&init_user_ns, current_fsuid()),
+			from_kgid(&init_user_ns, current_fsgid()),
+			cfs_curproc_cap_pack(), rdev, &request);
+	ll_finish_md_op_data(op_data);
         if (err)
                 GOTO(err_exit, err);
 
         ll_update_times(request, dir);
 
         if (dchild) {
-                err = ll_prep_inode(&inode, request, dchild->d_sb);
+                err = ll_prep_inode(&inode, request, dchild->d_sb, NULL);
                 if (err)
                      GOTO(err_exit, err);
 
@@ -917,12 +942,12 @@ static int ll_mknod_generic(struct inode *dir, struct qstr *name, int mode,
         int err;
         ENTRY;
 
-        CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s,dir=%lu/%u(%p) mode %o dev %x\n",
-               name->len, name->name, dir->i_ino, dir->i_generation, dir,
+	CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p) mode %o dev %x\n",
+	       name->len, name->name, PFID(ll_inode2fid(dir)), dir,
                mode, rdev);
 
-        if (!IS_POSIXACL(dir) || !exp_connect_umask(ll_i2mdexp(dir)))
-                mode &= ~cfs_curproc_umask();
+	if (!IS_POSIXACL(dir) || !exp_connect_umask(ll_i2mdexp(dir)))
+		mode &= ~current_umask();
 
         switch (mode & S_IFMT) {
         case 0:
@@ -957,10 +982,10 @@ static int ll_create_nd(struct inode *dir, struct dentry *dentry,
 {
 	int rc;
 
-	CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s,dir=%lu/%u(%p),"
-			   "flags=%u, excl=%d\n",
-	       dentry->d_name.len, dentry->d_name.name, dir->i_ino,
-	       dir->i_generation, dir, mode, want_excl);
+	CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p), "
+			   "flags=%u, excl=%d\n", dentry->d_name.len,
+	       dentry->d_name.name, PFID(ll_inode2fid(dir)),
+	       dir, mode, want_excl);
 
 	rc = ll_mknod_generic(dir, &dentry->d_name, mode, 0, dentry);
 
@@ -975,13 +1000,17 @@ static int ll_create_nd(struct inode *dir, struct dentry *dentry,
 static int ll_create_nd(struct inode *dir, struct dentry *dentry,
 			ll_umode_t mode, struct nameidata *nd)
 {
-        struct lookup_intent *it = ll_d2d(dentry)->lld_it;
+	struct ll_dentry_data *lld = ll_d2d(dentry);
+	struct lookup_intent *it = NULL;
         int rc;
 
+	if (lld != NULL)
+		it = lld->lld_it;
+
         if (!it)
                 return ll_mknod_generic(dir, &dentry->d_name, mode, 0, dentry);
 
-        ll_d2d(dentry)->lld_it = NULL;
+	lld->lld_it = NULL;
 
         /* Was there an error? Propagate it! */
         if (it->d.lustre.it_status) {
@@ -1016,9 +1045,9 @@ static int ll_symlink_generic(struct inode *dir, struct qstr *name,
         int err;
         ENTRY;
 
-        CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s,dir=%lu/%u(%p),target=%.*s\n",
-               name->len, name->name, dir->i_ino, dir->i_generation,
-               dir, 3000, tgt);
+	CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p), target=%.*s\n",
+	       name->len, name->name, PFID(ll_inode2fid(dir)),
+	       dir, 3000, tgt);
 
         err = ll_new_node(dir, name, (char *)tgt, S_IFLNK | S_IRWXUGO,
                           0, dchild, LUSTRE_OPC_SYMLINK);
@@ -1038,10 +1067,9 @@ static int ll_link_generic(struct inode *src,  struct inode *dir,
         int err;
 
         ENTRY;
-        CDEBUG(D_VFSTRACE,
-               "VFS Op: inode=%lu/%u(%p), dir=%lu/%u(%p), target=%.*s\n",
-               src->i_ino, src->i_generation, src, dir->i_ino,
-               dir->i_generation, dir, name->len, name->name);
+	CDEBUG(D_VFSTRACE, "VFS Op: inode="DFID"(%p), dir="DFID
+	       "(%p), target=%.*s\n", PFID(ll_inode2fid(src)), src,
+	       PFID(ll_inode2fid(dir)), dir, name->len, name->name);
 
         op_data = ll_prep_md_op_data(NULL, src, dir, name->name, name->len,
                                      0, LUSTRE_OPC_ANY, NULL);
@@ -1068,13 +1096,13 @@ static int ll_mkdir_generic(struct inode *dir, struct qstr *name,
         int err;
         ENTRY;
 
-        CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s,dir=%lu/%u(%p)\n",
-               name->len, name->name, dir->i_ino, dir->i_generation, dir);
+	CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p)\n",
+	       name->len, name->name, PFID(ll_inode2fid(dir)), dir);
 
-        if (!IS_POSIXACL(dir) || !exp_connect_umask(ll_i2mdexp(dir)))
-                mode &= ~cfs_curproc_umask();
-        mode = (mode & (S_IRWXUGO|S_ISVTX)) | S_IFDIR;
-        err = ll_new_node(dir, name, NULL, mode, 0, dchild, LUSTRE_OPC_MKDIR);
+	if (!IS_POSIXACL(dir) || !exp_connect_umask(ll_i2mdexp(dir)))
+		mode &= ~current_umask();
+	mode = (mode & (S_IRWXUGO|S_ISVTX)) | S_IFDIR;
+	err = ll_new_node(dir, name, NULL, mode, 0, dchild, LUSTRE_OPC_MKDIR);
 
         if (!err)
                 ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_MKDIR, 1);
@@ -1082,22 +1110,6 @@ static int ll_mkdir_generic(struct inode *dir, struct qstr *name,
         RETURN(err);
 }
 
-/* Try to find the child dentry by its name.
-   If found, put the result fid into @fid. */
-static void ll_get_child_fid(struct inode * dir, struct qstr *name,
-                             struct lu_fid *fid)
-{
-	struct dentry *parent, *child;
-
-	parent = ll_d_hlist_entry(dir->i_dentry, struct dentry, d_alias);
-	child = d_lookup(parent, name);
-        if (child) {
-                if (child->d_inode)
-                        *fid = *ll_inode2fid(child->d_inode);
-                dput(child);
-        }
-}
-
 static int ll_rmdir_generic(struct inode *dir, struct dentry *dparent,
                             struct dentry *dchild, struct qstr *name)
 {
@@ -1106,8 +1118,8 @@ static int ll_rmdir_generic(struct inode *dir, struct dentry *dparent,
         int rc;
         ENTRY;
 
-        CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s,dir=%lu/%u(%p)\n",
-               name->len, name->name, dir->i_ino, dir->i_generation, dir);
+	CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p)\n",
+	       name->len, name->name, PFID(ll_inode2fid(dir)), dir);
 
         if (unlikely(ll_d_mountpoint(dparent, dchild, name)))
                 RETURN(-EBUSY);
@@ -1117,7 +1129,9 @@ static int ll_rmdir_generic(struct inode *dir, struct dentry *dparent,
         if (IS_ERR(op_data))
                 RETURN(PTR_ERR(op_data));
 
-        ll_get_child_fid(dir, name, &op_data->op_fid3);
+	if (dchild != NULL && dchild->d_inode != NULL)
+		op_data->op_fid3 = *ll_inode2fid(dchild->d_inode);
+	op_data->op_fid2 = op_data->op_fid3;
         rc = md_unlink(ll_i2sbi(dir)->ll_md_exp, op_data, &request);
         ll_finish_md_op_data(op_data);
         if (rc == 0) {
@@ -1129,6 +1143,35 @@ static int ll_rmdir_generic(struct inode *dir, struct dentry *dparent,
         RETURN(rc);
 }
 
+/**
+ * Remove dir entry
+ **/
+int ll_rmdir_entry(struct inode *dir, char *name, int namelen)
+{
+	struct ptlrpc_request *request = NULL;
+	struct md_op_data *op_data;
+	int rc;
+	ENTRY;
+
+	CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p)\n",
+	       namelen, name, PFID(ll_inode2fid(dir)), dir);
+
+	op_data = ll_prep_md_op_data(NULL, dir, NULL, name, strlen(name),
+				     S_IFDIR, LUSTRE_OPC_ANY, NULL);
+	if (IS_ERR(op_data))
+		RETURN(PTR_ERR(op_data));
+	op_data->op_cli_flags |= CLI_RM_ENTRY;
+	rc = md_unlink(ll_i2sbi(dir)->ll_md_exp, op_data, &request);
+	ll_finish_md_op_data(op_data);
+	if (rc == 0) {
+		ll_update_times(request, dir);
+		ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_RMDIR, 1);
+	}
+
+	ptlrpc_req_finished(request);
+	RETURN(rc);
+}
+
 int ll_objects_destroy(struct ptlrpc_request *request, struct inode *dir)
 {
         struct mdt_body *body;
@@ -1169,8 +1212,7 @@ int ll_objects_destroy(struct ptlrpc_request *request, struct inode *dir)
         if (oa == NULL)
                 GOTO(out_free_memmd, rc = -ENOMEM);
 
-        oa->o_id = lsm->lsm_object_id;
-        oa->o_seq = lsm->lsm_object_seq;
+	oa->o_oi = lsm->lsm_oi;
         oa->o_mode = body->mode & S_IFMT;
         oa->o_valid = OBD_MD_FLID | OBD_MD_FLTYPE | OBD_MD_FLGROUP;
 
@@ -1193,17 +1235,17 @@ int ll_objects_destroy(struct ptlrpc_request *request, struct inode *dir)
                         GOTO(out_free_memmd, rc);
         }
 
-        rc = obd_destroy(NULL, ll_i2dtexp(dir), oa, lsm, &oti,
-                         ll_i2mdexp(dir), oc);
-        capa_put(oc);
-        OBDO_FREE(oa);
-        if (rc)
-                CERROR("obd destroy objid "LPX64" error %d\n",
-                       lsm->lsm_object_id, rc);
- out_free_memmd:
-        obd_free_memmd(ll_i2dtexp(dir), &lsm);
- out:
-        return rc;
+	rc = obd_destroy(NULL, ll_i2dtexp(dir), oa, lsm, &oti,
+			 ll_i2mdexp(dir), oc);
+	capa_put(oc);
+	if (rc)
+		CERROR("obd destroy objid "DOSTID" error %d\n",
+		       POSTID(&lsm->lsm_oi), rc);
+out_free_memmd:
+	obd_free_memmd(ll_i2dtexp(dir), &lsm);
+	OBDO_FREE(oa);
+out:
+	return rc;
 }
 
 /* ll_unlink_generic() doesn't update the inode with the new link count.
@@ -1217,8 +1259,8 @@ static int ll_unlink_generic(struct inode *dir, struct dentry *dparent,
         struct md_op_data *op_data;
         int rc;
         ENTRY;
-        CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s,dir=%lu/%u(%p)\n",
-               name->len, name->name, dir->i_ino, dir->i_generation, dir);
+	CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p)\n",
+	       name->len, name->name, PFID(ll_inode2fid(dir)), dir);
 
         /*
          * XXX: unlink bind mountpoint maybe call to here,
@@ -1232,11 +1274,14 @@ static int ll_unlink_generic(struct inode *dir, struct dentry *dparent,
         if (IS_ERR(op_data))
                 RETURN(PTR_ERR(op_data));
 
-        ll_get_child_fid(dir, name, &op_data->op_fid3);
-        rc = md_unlink(ll_i2sbi(dir)->ll_md_exp, op_data, &request);
-        ll_finish_md_op_data(op_data);
-        if (rc)
-                GOTO(out, rc);
+	if (dchild != NULL && dchild->d_inode != NULL)
+		op_data->op_fid3 = *ll_inode2fid(dchild->d_inode);
+
+	op_data->op_fid2 = op_data->op_fid3;
+	rc = md_unlink(ll_i2sbi(dir)->ll_md_exp, op_data, &request);
+	ll_finish_md_op_data(op_data);
+	if (rc)
+		GOTO(out, rc);
 
         ll_update_times(request, dir);
         ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_UNLINK, 1);
@@ -1257,10 +1302,11 @@ static int ll_rename_generic(struct inode *src, struct dentry *src_dparent,
         struct md_op_data *op_data;
         int err;
         ENTRY;
-        CDEBUG(D_VFSTRACE,"VFS Op:oldname=%.*s,src_dir=%lu/%u(%p),newname=%.*s,"
-               "tgt_dir=%lu/%u(%p)\n", src_name->len, src_name->name,
-               src->i_ino, src->i_generation, src, tgt_name->len,
-               tgt_name->name, tgt->i_ino, tgt->i_generation, tgt);
+	CDEBUG(D_VFSTRACE, "VFS Op:oldname=%.*s, src_dir="DFID
+	       "(%p), newname=%.*s, tgt_dir="DFID"(%p)\n",
+	       src_name->len, src_name->name,
+	       PFID(ll_inode2fid(src)), src, tgt_name->len,
+	       tgt_name->name, PFID(ll_inode2fid(tgt)), tgt);
 
         if (unlikely(ll_d_mountpoint(src_dparent, src_dchild, src_name) ||
             ll_d_mountpoint(tgt_dparent, tgt_dchild, tgt_name)))
@@ -1271,8 +1317,11 @@ static int ll_rename_generic(struct inode *src, struct dentry *src_dparent,
         if (IS_ERR(op_data))
                 RETURN(PTR_ERR(op_data));
 
-        ll_get_child_fid(src, src_name, &op_data->op_fid3);
-        ll_get_child_fid(tgt, tgt_name, &op_data->op_fid4);
+	if (src_dchild != NULL && src_dchild->d_inode != NULL)
+		op_data->op_fid3 = *ll_inode2fid(src_dchild->d_inode);
+	if (tgt_dchild != NULL && tgt_dchild->d_inode != NULL)
+		op_data->op_fid4 = *ll_inode2fid(tgt_dchild->d_inode);
+
         err = md_rename(sbi->ll_md_exp, op_data,
                         src_name->name, src_name->len,
                         tgt_name->name, tgt_name->len, &request);
@@ -1332,16 +1381,12 @@ static int ll_rename(struct inode *old_dir, struct dentry *old_dentry,
                                  old_dentry, &old_dentry->d_name,
                                  new_dir, NULL, new_dentry,
                                  &new_dentry->d_name);
-        if (!err) {
-#ifndef HAVE_FS_RENAME_DOES_D_MOVE
-                if (!S_ISDIR(old_dentry->d_inode->i_mode))
-#endif
+	if (!err)
                         d_move(old_dentry, new_dentry);
-        }
         return err;
 }
 
-struct inode_operations ll_dir_inode_operations = {
+const struct inode_operations ll_dir_inode_operations = {
 	.mknod              = ll_mknod,
 #ifdef HAVE_IOP_ATOMIC_OPEN
 	.atomic_open	    = ll_atomic_open,
@@ -1367,7 +1412,7 @@ struct inode_operations ll_dir_inode_operations = {
 #endif
 };
 
-struct inode_operations ll_special_inode_operations = {
+const struct inode_operations ll_special_inode_operations = {
 	.setattr        = ll_setattr,
 	.getattr        = ll_getattr,
 	.permission     = ll_inode_permission,