Whamcloud - gitweb
LU-3544 nfs: writing to new files will return ENOENT
[fs/lustre-release.git] / lustre / llite / file.c
index c3dade8..87b30b4 100644 (file)
@@ -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/
@@ -54,8 +54,12 @@ struct ll_file_data *ll_file_data_get(void)
 {
        struct ll_file_data *fd;
 
-       OBD_SLAB_ALLOC_PTR_GFP(fd, ll_file_data_slab, CFS_ALLOC_IO);
+       OBD_SLAB_ALLOC_PTR_GFP(fd, ll_file_data_slab, __GFP_IO);
+       if (fd == NULL)
+               return NULL;
+
        fd->fd_write_failed = false;
+
        return fd;
 }
 
@@ -95,8 +99,9 @@ static void ll_prepare_close(struct inode *inode, struct md_op_data *op_data,
 {
         ENTRY;
 
-        op_data->op_attr.ia_valid = ATTR_MODE | ATTR_ATIME_SET |
-                                 ATTR_MTIME_SET | ATTR_CTIME_SET;
+       op_data->op_attr.ia_valid = ATTR_MODE | ATTR_ATIME | ATTR_ATIME_SET |
+                                       ATTR_MTIME | ATTR_MTIME_SET |
+                                       ATTR_CTIME | ATTR_CTIME_SET;
 
         if (!(och->och_flags & FMODE_WRITE))
                 goto out;
@@ -246,6 +251,24 @@ int ll_md_close(struct obd_export *md_exp, struct inode *inode,
         if (unlikely(fd->fd_flags & LL_FILE_GROUP_LOCKED))
                 ll_put_grouplock(inode, file, fd->fd_grouplock.cg_gid);
 
+       if (fd->fd_lease_och != NULL) {
+               bool lease_broken;
+
+               /* Usually the lease is not released when the
+                * application crashed, we need to release here. */
+               rc = ll_lease_close(fd->fd_lease_och, inode, &lease_broken);
+               CDEBUG(rc ? D_ERROR : D_INODE, "Clean up lease "DFID" %d/%d\n",
+                       PFID(&lli->lli_fid), rc, lease_broken);
+
+               fd->fd_lease_och = NULL;
+       }
+
+       if (fd->fd_och != NULL) {
+               rc = ll_close_inode_openhandle(md_exp, inode, fd->fd_och);
+               fd->fd_och = NULL;
+               GOTO(out, rc);
+       }
+
         /* Let's see if we have good enough OPEN lock on the file and if
            we can skip talking to MDS */
         if (file->f_dentry->d_inode) { /* Can this ever be false? */
@@ -282,11 +305,12 @@ int ll_md_close(struct obd_export *md_exp, struct inode *inode,
                        file, file->f_dentry, file->f_dentry->d_name.name);
         }
 
-        LUSTRE_FPRIVATE(file) = NULL;
-        ll_file_data_put(fd);
-        ll_capa_close(inode);
+out:
+       LUSTRE_FPRIVATE(file) = NULL;
+       ll_file_data_put(fd);
+       ll_capa_close(inode);
 
-        RETURN(rc);
+       RETURN(rc);
 }
 
 /* While this returns an error code, fput() the caller does not, so we need
@@ -355,8 +379,6 @@ static int ll_intent_file_open(struct file *file, void *lmm,
 {
         struct ll_sb_info *sbi = ll_i2sbi(file->f_dentry->d_inode);
         struct dentry *parent = file->f_dentry->d_parent;
-        const char *name = file->f_dentry->d_name.name;
-        const int len = file->f_dentry->d_name.len;
         struct md_op_data *op_data;
         struct ptlrpc_request *req;
         __u32 opc = LUSTRE_OPC_ANY;
@@ -381,9 +403,10 @@ static int ll_intent_file_open(struct file *file, void *lmm,
                         opc = LUSTRE_OPC_CREATE;
         }
 
-        op_data  = ll_prep_md_op_data(NULL, parent->d_inode,
-                                      file->f_dentry->d_inode, name, len,
-                                      O_RDWR, opc, NULL);
+       op_data  = ll_prep_md_op_data(NULL, parent->d_inode,
+                                     file->f_dentry->d_inode, NULL, 0,
+                                     O_RDWR, opc, NULL);
+
         if (IS_ERR(op_data))
                 RETURN(PTR_ERR(op_data));
 
@@ -438,24 +461,20 @@ void ll_ioepoch_open(struct ll_inode_info *lli, __u64 ioepoch)
         }
 }
 
-static int ll_och_fill(struct obd_export *md_exp, struct ll_inode_info *lli,
-                       struct lookup_intent *it, struct obd_client_handle *och)
+static int ll_och_fill(struct obd_export *md_exp, struct lookup_intent *it,
+                      struct obd_client_handle *och)
 {
-        struct ptlrpc_request *req = it->d.lustre.it_data;
-        struct mdt_body *body;
-
-        LASSERT(och);
-
-        body = req_capsule_server_get(&req->rq_pill, &RMF_MDT_BODY);
-        LASSERT(body != NULL);                      /* reply already checked out */
+       struct ptlrpc_request *req = it->d.lustre.it_data;
+       struct mdt_body *body;
 
-        memcpy(&och->och_fh, &body->handle, sizeof(body->handle));
-        och->och_magic = OBD_CLIENT_HANDLE_MAGIC;
-        och->och_fid = lli->lli_fid;
-        och->och_flags = it->it_flags;
-        ll_ioepoch_open(lli, body->ioepoch);
+       body = req_capsule_server_get(&req->rq_pill, &RMF_MDT_BODY);
+       och->och_fh = body->handle;
+       och->och_fid = body->fid1;
+       och->och_lease_handle.cookie = it->d.lustre.it_lock_handle;
+       och->och_magic = OBD_CLIENT_HANDLE_MAGIC;
+       och->och_flags = it->it_flags;
 
-        return md_set_open_replay_data(md_exp, och, req);
+       return md_set_open_replay_data(md_exp, och, req);
 }
 
 int ll_local_open(struct file *file, struct lookup_intent *it,
@@ -474,21 +493,19 @@ int ll_local_open(struct file *file, struct lookup_intent *it,
                 struct mdt_body *body;
                 int rc;
 
-                rc = ll_och_fill(ll_i2sbi(inode)->ll_md_exp, lli, it, och);
-                if (rc)
-                        RETURN(rc);
+               rc = ll_och_fill(ll_i2sbi(inode)->ll_md_exp, it, och);
+               if (rc != 0)
+                       RETURN(rc);
 
-                body = req_capsule_server_get(&req->rq_pill, &RMF_MDT_BODY);
-                if ((it->it_flags & FMODE_WRITE) &&
-                    (body->valid & OBD_MD_FLSIZE))
-                        CDEBUG(D_INODE, "Epoch "LPU64" opened on "DFID"\n",
-                               lli->lli_ioepoch, PFID(&lli->lli_fid));
-        }
+               body = req_capsule_server_get(&req->rq_pill, &RMF_MDT_BODY);
+               ll_ioepoch_open(lli, body->ioepoch);
+       }
 
-        LUSTRE_FPRIVATE(file) = fd;
-        ll_readahead_init(inode, &fd->fd_ras);
-        fd->fd_omode = it->it_flags;
-        RETURN(0);
+       LUSTRE_FPRIVATE(file) = fd;
+       ll_readahead_init(inode, &fd->fd_ras);
+       fd->fd_omode = it->it_flags & (FMODE_READ | FMODE_WRITE | FMODE_EXEC);
+
+       RETURN(0);
 }
 
 /* Open a file, and (for the very first open) create objects on the OSTs at
@@ -521,12 +538,12 @@ int ll_file_open(struct inode *inode, struct file *file)
         it = file->private_data; /* XXX: compat macro */
         file->private_data = NULL; /* prevent ll_local_open assertion */
 
-        fd = ll_file_data_get();
-        if (fd == NULL)
-                GOTO(out_och_free, rc = -ENOMEM);
+       fd = ll_file_data_get();
+       if (fd == NULL)
+               GOTO(out_openerr, rc = -ENOMEM);
 
-        fd->fd_file = file;
-        if (S_ISDIR(inode->i_mode)) {
+       fd->fd_file = file;
+       if (S_ISDIR(inode->i_mode)) {
                spin_lock(&lli->lli_sa_lock);
                if (lli->lli_opendir_key == NULL && lli->lli_sai == NULL &&
                    lli->lli_opendir_pid == 0) {
@@ -535,7 +552,7 @@ int ll_file_open(struct inode *inode, struct file *file)
                        opendir_set = 1;
                }
                spin_unlock(&lli->lli_sa_lock);
-        }
+       }
 
         if (inode->i_sb->s_root == file->f_dentry) {
                 LUSTRE_FPRIVATE(file) = fd;
@@ -690,6 +707,199 @@ out_openerr:
         return rc;
 }
 
+static int ll_md_blocking_lease_ast(struct ldlm_lock *lock,
+                       struct ldlm_lock_desc *desc, void *data, int flag)
+{
+       int rc;
+       struct lustre_handle lockh;
+       ENTRY;
+
+       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: %d\n", rc);
+                       RETURN(rc);
+               }
+               break;
+       case LDLM_CB_CANCELING:
+               /* do nothing */
+               break;
+       }
+       RETURN(0);
+}
+
+/**
+ * Acquire a lease and open the file.
+ */
+struct obd_client_handle *ll_lease_open(struct inode *inode, struct file *file,
+                                       fmode_t fmode)
+{
+       struct lookup_intent it = { .it_op = IT_OPEN };
+       struct ll_sb_info *sbi = ll_i2sbi(inode);
+       struct md_op_data *op_data;
+       struct ptlrpc_request *req;
+       struct lustre_handle old_handle = { 0 };
+       struct obd_client_handle *och = NULL;
+       int rc;
+       int rc2;
+       ENTRY;
+
+       if (fmode != FMODE_WRITE && fmode != FMODE_READ)
+               RETURN(ERR_PTR(-EINVAL));
+
+       if (file != NULL) {
+               struct ll_inode_info *lli = ll_i2info(inode);
+               struct ll_file_data *fd = LUSTRE_FPRIVATE(file);
+               struct obd_client_handle **och_p;
+               __u64 *och_usecount;
+
+               if (!(fmode & file->f_mode) || (file->f_mode & FMODE_EXEC))
+                       RETURN(ERR_PTR(-EPERM));
+
+               /* Get the openhandle of the file */
+               rc = -EBUSY;
+               mutex_lock(&lli->lli_och_mutex);
+               if (fd->fd_lease_och != NULL) {
+                       mutex_unlock(&lli->lli_och_mutex);
+                       RETURN(ERR_PTR(rc));
+               }
+
+               if (fd->fd_och == NULL) {
+                       if (file->f_mode & FMODE_WRITE) {
+                               LASSERT(lli->lli_mds_write_och != NULL);
+                               och_p = &lli->lli_mds_write_och;
+                               och_usecount = &lli->lli_open_fd_write_count;
+                       } else {
+                               LASSERT(lli->lli_mds_read_och != NULL);
+                               och_p = &lli->lli_mds_read_och;
+                               och_usecount = &lli->lli_open_fd_read_count;
+                       }
+                       if (*och_usecount == 1) {
+                               fd->fd_och = *och_p;
+                               *och_p = NULL;
+                               *och_usecount = 0;
+                               rc = 0;
+                       }
+               }
+               mutex_unlock(&lli->lli_och_mutex);
+               if (rc < 0) /* more than 1 opener */
+                       RETURN(ERR_PTR(rc));
+
+               LASSERT(fd->fd_och != NULL);
+               old_handle = fd->fd_och->och_fh;
+       }
+
+       OBD_ALLOC_PTR(och);
+       if (och == NULL)
+               RETURN(ERR_PTR(-ENOMEM));
+
+       op_data = ll_prep_md_op_data(NULL, inode, inode, NULL, 0, 0,
+                                       LUSTRE_OPC_ANY, NULL);
+       if (IS_ERR(op_data))
+               GOTO(out, rc = PTR_ERR(op_data));
+
+       /* To tell the MDT this openhandle is from the same owner */
+       op_data->op_handle = old_handle;
+
+       it.it_flags = fmode | MDS_OPEN_LOCK | MDS_OPEN_BY_FID | MDS_OPEN_LEASE;
+       rc = md_intent_lock(sbi->ll_md_exp, op_data, NULL, 0, &it, 0, &req,
+                               ll_md_blocking_lease_ast,
+       /* LDLM_FL_NO_LRU: To not put the lease lock into LRU list, otherwise
+        * it can be cancelled which may mislead applications that the lease is
+        * broken;
+        * LDLM_FL_EXCL: Set this flag so that it won't be matched by normal
+        * open in ll_md_blocking_ast(). Otherwise as ll_md_blocking_lease_ast
+        * doesn't deal with openhandle, so normal openhandle will be leaked. */
+                               LDLM_FL_NO_LRU | LDLM_FL_EXCL);
+       ll_finish_md_op_data(op_data);
+       if (req != NULL) {
+               ptlrpc_req_finished(req);
+               it_clear_disposition(&it, DISP_ENQ_COMPLETE);
+       }
+       if (rc < 0)
+               GOTO(out_release_it, rc);
+
+       if (it_disposition(&it, DISP_LOOKUP_NEG))
+               GOTO(out_release_it, rc = -ENOENT);
+
+       rc = it_open_error(DISP_OPEN_OPEN, &it);
+       if (rc)
+               GOTO(out_release_it, rc);
+
+       LASSERT(it_disposition(&it, DISP_ENQ_OPEN_REF));
+       ll_och_fill(sbi->ll_md_exp, &it, och);
+
+       if (!it_disposition(&it, DISP_OPEN_LEASE)) /* old server? */
+               GOTO(out_close, rc = -EOPNOTSUPP);
+
+       /* already get lease, handle lease lock */
+       ll_set_lock_data(sbi->ll_md_exp, inode, &it, NULL);
+       if (it.d.lustre.it_lock_mode == 0 ||
+           it.d.lustre.it_lock_bits != MDS_INODELOCK_OPEN) {
+               /* open lock must return for lease */
+               CERROR(DFID "lease granted but no open lock, %d/%Lu.\n",
+                       PFID(ll_inode2fid(inode)), it.d.lustre.it_lock_mode,
+                       it.d.lustre.it_lock_bits);
+               GOTO(out_close, rc = -EPROTO);
+       }
+
+       ll_intent_release(&it);
+       RETURN(och);
+
+out_close:
+       rc2 = ll_close_inode_openhandle(sbi->ll_md_exp, inode, och);
+       if (rc2)
+               CERROR("Close openhandle returned %d\n", rc2);
+
+       /* cancel open lock */
+       if (it.d.lustre.it_lock_mode != 0) {
+               ldlm_lock_decref_and_cancel(&och->och_lease_handle,
+                                               it.d.lustre.it_lock_mode);
+               it.d.lustre.it_lock_mode = 0;
+       }
+out_release_it:
+       ll_intent_release(&it);
+out:
+       OBD_FREE_PTR(och);
+       RETURN(ERR_PTR(rc));
+}
+EXPORT_SYMBOL(ll_lease_open);
+
+/**
+ * Release lease and close the file.
+ * It will check if the lease has ever broken.
+ */
+int ll_lease_close(struct obd_client_handle *och, struct inode *inode,
+                       bool *lease_broken)
+{
+       struct ldlm_lock *lock;
+       bool cancelled = true;
+       int rc;
+       ENTRY;
+
+       lock = ldlm_handle2lock(&och->och_lease_handle);
+       if (lock != NULL) {
+               lock_res_and_lock(lock);
+               cancelled = ldlm_is_cancel(lock);
+               unlock_res_and_lock(lock);
+               ldlm_lock_put(lock);
+       }
+
+       CDEBUG(D_INODE, "lease for "DFID" broken? %d\n",
+               PFID(&ll_i2info(inode)->lli_fid), cancelled);
+
+       if (!cancelled)
+               ldlm_cli_cancel(&och->och_lease_handle, 0);
+       if (lease_broken != NULL)
+               *lease_broken = cancelled;
+
+       rc = ll_close_inode_openhandle(ll_i2sbi(inode)->ll_md_exp, inode, och);
+       RETURN(rc);
+}
+EXPORT_SYMBOL(ll_lease_close);
+
 /* Fills the obdo with the attributes for the lsm */
 static int ll_lsm_getattr(struct lov_stripe_md *lsm, struct obd_export *exp,
                           struct obd_capa *capa, struct obdo *obdo,
@@ -922,7 +1132,7 @@ out:
         cl_io_fini(env, io);
        /* If any bit been read/written (result != 0), we just return
         * short read/write instead of restart io. */
-       if (result == 0 && io->ci_need_restart) {
+       if ((result == 0 || result == -ENODATA) && io->ci_need_restart) {
                CDEBUG(D_VFSTRACE, "Restart %s on %s from %lld, count:%zd\n",
                       iot == CIT_READ ? "read" : "write",
                       file->f_dentry->d_name.name, *ppos, count);
@@ -1201,35 +1411,6 @@ static ssize_t ll_file_write(struct file *file, const char *buf, size_t count,
 }
 #endif
 
-
-#ifdef HAVE_KERNEL_SENDFILE
-/*
- * Send file content (through pagecache) somewhere with helper
- */
-static ssize_t ll_file_sendfile(struct file *in_file, loff_t *ppos,size_t count,
-                                read_actor_t actor, void *target)
-{
-        struct lu_env      *env;
-        struct vvp_io_args *args;
-        ssize_t             result;
-        int                 refcheck;
-        ENTRY;
-
-        env = cl_env_get(&refcheck);
-        if (IS_ERR(env))
-                RETURN(PTR_ERR(env));
-
-        args = vvp_env_args(env, IO_SENDFILE);
-        args->u.sendfile.via_target = target;
-        args->u.sendfile.via_actor = actor;
-
-        result = ll_file_io_generic(env, args, in_file, CIT_READ, ppos, count);
-        cl_env_put(env, &refcheck);
-        RETURN(result);
-}
-#endif
-
-#ifdef HAVE_KERNEL_SPLICE_READ
 /*
  * Send file content (through pagecache) somewhere with helper
  */
@@ -1255,7 +1436,6 @@ static ssize_t ll_file_splice_read(struct file *in_file, loff_t *ppos,
         cl_env_put(env, &refcheck);
         RETURN(result);
 }
-#endif
 
 static int ll_lov_recreate(struct inode *inode, struct ost_id *oi,
                            obd_count ost_idx)
@@ -1273,7 +1453,7 @@ static int ll_lov_recreate(struct inode *inode, struct ost_id *oi,
                RETURN(-ENOMEM);
 
        lsm = ccc_inode_lsm_get(inode);
-        if (lsm == NULL)
+       if (!lsm_has_objects(lsm))
                 GOTO(out, rc = -ENOENT);
 
         lsm_size = sizeof(*lsm) + (sizeof(struct lov_oinfo) *
@@ -1429,6 +1609,12 @@ int ll_lov_getstripe_ea_info(struct inode *inode, const char *filename,
          * passing it to userspace.
          */
         if (LOV_MAGIC != cpu_to_le32(LOV_MAGIC)) {
+               int stripe_count;
+
+               stripe_count = le16_to_cpu(lmm->lmm_stripe_count);
+               if (le32_to_cpu(lmm->lmm_pattern) & LOV_PATTERN_F_RELEASED)
+                       stripe_count = 0;
+
                 /* if function called for directory - we should
                  * avoid swab not existent lsm objects */
                 if (lmm->lmm_magic == cpu_to_le32(LOV_MAGIC_V1)) {
@@ -1436,13 +1622,13 @@ int ll_lov_getstripe_ea_info(struct inode *inode, const char *filename,
                         if (S_ISREG(body->mode))
                                 lustre_swab_lov_user_md_objects(
                                  ((struct lov_user_md_v1 *)lmm)->lmm_objects,
-                                 ((struct lov_user_md_v1 *)lmm)->lmm_stripe_count);
+                                 stripe_count);
                 } else if (lmm->lmm_magic == cpu_to_le32(LOV_MAGIC_V3)) {
                         lustre_swab_lov_user_md_v3((struct lov_user_md_v3 *)lmm);
                         if (S_ISREG(body->mode))
                                 lustre_swab_lov_user_md_objects(
                                  ((struct lov_user_md_v3 *)lmm)->lmm_objects,
-                                 ((struct lov_user_md_v3 *)lmm)->lmm_stripe_count);
+                                 stripe_count);
                 }
         }
 
@@ -1639,8 +1825,7 @@ int ll_release_openhandle(struct dentry *dentry, struct lookup_intent *it)
         if (!och)
                 GOTO(out, rc = -ENOMEM);
 
-        ll_och_fill(ll_i2sbi(inode)->ll_md_exp,
-                    ll_i2info(inode), it, och);
+       ll_och_fill(ll_i2sbi(inode)->ll_md_exp, it, och);
 
         rc = ll_close_inode_openhandle(ll_i2sbi(inode)->ll_md_exp,
                                        inode, och);
@@ -1828,20 +2013,18 @@ int ll_data_version(struct inode *inode, __u64 *data_version,
 
        /* If no stripe, we consider version is 0. */
        lsm = ccc_inode_lsm_get(inode);
-       if (lsm == NULL) {
+       if (!lsm_has_objects(lsm)) {
                *data_version = 0;
                CDEBUG(D_INODE, "No object for inode\n");
-               RETURN(0);
+               GOTO(out, rc = 0);
        }
 
        OBD_ALLOC_PTR(obdo);
-       if (obdo == NULL) {
-               ccc_inode_lsm_put(inode, lsm);
-               RETURN(-ENOMEM);
-       }
+       if (obdo == NULL)
+               GOTO(out, rc = -ENOMEM);
 
        rc = ll_lsm_getattr(lsm, sbi->ll_dt_exp, NULL, obdo, 0, extent_lock);
-       if (!rc) {
+       if (rc == 0) {
                if (!(obdo->o_valid & OBD_MD_FLDATAVERSION))
                        rc = -EOPNOTSUPP;
                else
@@ -1849,8 +2032,9 @@ int ll_data_version(struct inode *inode, __u64 *data_version,
        }
 
        OBD_FREE_PTR(obdo);
+       EXIT;
+out:
        ccc_inode_lsm_put(inode, lsm);
-
        RETURN(rc);
 }
 
@@ -1881,8 +2065,8 @@ static int ll_swap_layouts(struct file *file1, struct file *file2,
        if (!S_ISREG(llss->inode2->i_mode))
                GOTO(free, rc = -EINVAL);
 
-       if (ll_permission(llss->inode1, MAY_WRITE, NULL) ||
-           ll_permission(llss->inode2, MAY_WRITE, NULL))
+       if (inode_permission(llss->inode1, MAY_WRITE) ||
+           inode_permission(llss->inode2, MAY_WRITE))
                GOTO(free, rc = -EPERM);
 
        if (llss->inode2->i_sb != llss->inode1->i_sb)
@@ -1962,12 +2146,12 @@ static int ll_swap_layouts(struct file *file1, struct file *file2,
        rc = -ENOMEM;
        op_data = ll_prep_md_op_data(NULL, llss->inode1, llss->inode2, NULL, 0,
                                     0, LUSTRE_OPC_ANY, &msl);
-       if (op_data != NULL) {
-               rc = obd_iocontrol(LL_IOC_LOV_SWAP_LAYOUTS,
-                                  ll_i2mdexp(llss->inode1),
-                                  sizeof(*op_data), op_data, NULL);
-               ll_finish_md_op_data(op_data);
-       }
+       if (IS_ERR(op_data))
+               GOTO(free, rc = PTR_ERR(op_data));
+
+       rc = obd_iocontrol(LL_IOC_LOV_SWAP_LAYOUTS, ll_i2mdexp(llss->inode1),
+                          sizeof(*op_data), op_data, NULL);
+       ll_finish_md_op_data(op_data);
 
 putgl:
        if (gid != 0) {
@@ -2064,7 +2248,7 @@ long ll_file_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
                struct file *file2;
                struct lustre_swap_layouts lsl;
 
-               if (cfs_copy_from_user(&lsl, (char *)arg,
+               if (copy_from_user(&lsl, (char *)arg,
                                       sizeof(struct lustre_swap_layouts)))
                        RETURN(-EFAULT);
 
@@ -2161,9 +2345,9 @@ long ll_file_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 
                op_data = ll_prep_md_op_data(NULL, inode, NULL, NULL, 0, 0,
                                             LUSTRE_OPC_ANY, hus);
-               if (op_data == NULL) {
+               if (IS_ERR(op_data)) {
                        OBD_FREE_PTR(hus);
-                       RETURN(-ENOMEM);
+                       RETURN(PTR_ERR(op_data));
                }
 
                rc = obd_iocontrol(cmd, ll_i2mdexp(inode), sizeof(*op_data),
@@ -2199,9 +2383,9 @@ long ll_file_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 
                op_data = ll_prep_md_op_data(NULL, inode, NULL, NULL, 0, 0,
                                             LUSTRE_OPC_ANY, hss);
-               if (op_data == NULL) {
+               if (IS_ERR(op_data)) {
                        OBD_FREE_PTR(hss);
-                       RETURN(-ENOMEM);
+                       RETURN(PTR_ERR(op_data));
                }
 
                rc = obd_iocontrol(cmd, ll_i2mdexp(inode), sizeof(*op_data),
@@ -2223,21 +2407,105 @@ long ll_file_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 
                op_data = ll_prep_md_op_data(NULL, inode, NULL, NULL, 0, 0,
                                             LUSTRE_OPC_ANY, hca);
-               if (op_data == NULL) {
+               if (IS_ERR(op_data)) {
                        OBD_FREE_PTR(hca);
-                       RETURN(-ENOMEM);
+                       RETURN(PTR_ERR(op_data));
                }
 
                rc = obd_iocontrol(cmd, ll_i2mdexp(inode), sizeof(*op_data),
                                   op_data, NULL);
 
-               if (cfs_copy_to_user((char *)arg, hca, sizeof(*hca)))
+               if (copy_to_user((char *)arg, hca, sizeof(*hca)))
                        rc = -EFAULT;
 
                ll_finish_md_op_data(op_data);
                OBD_FREE_PTR(hca);
                RETURN(rc);
        }
+       case LL_IOC_SET_LEASE: {
+               struct ll_inode_info *lli = ll_i2info(inode);
+               struct obd_client_handle *och = NULL;
+               bool lease_broken;
+               fmode_t mode = 0;
+
+               switch (arg) {
+               case F_WRLCK:
+                       if (!(file->f_mode & FMODE_WRITE))
+                               RETURN(-EPERM);
+                       mode = FMODE_WRITE;
+                       break;
+               case F_RDLCK:
+                       if (!(file->f_mode & FMODE_READ))
+                               RETURN(-EPERM);
+                       mode = FMODE_READ;
+                       break;
+               case F_UNLCK:
+                       mutex_lock(&lli->lli_och_mutex);
+                       if (fd->fd_lease_och != NULL) {
+                               och = fd->fd_lease_och;
+                               fd->fd_lease_och = NULL;
+                       }
+                       mutex_unlock(&lli->lli_och_mutex);
+
+                       if (och != NULL) {
+                               mode = och->och_flags &(FMODE_READ|FMODE_WRITE);
+                               rc = ll_lease_close(och, inode, &lease_broken);
+                               if (rc == 0 && lease_broken)
+                                       mode = 0;
+                       } else {
+                               rc = -ENOLCK;
+                       }
+
+                       /* return the type of lease or error */
+                       RETURN(rc < 0 ? rc : (int)mode);
+               default:
+                       RETURN(-EINVAL);
+               }
+
+               CDEBUG(D_INODE, "Set lease with mode %d\n", mode);
+
+               /* apply for lease */
+               och = ll_lease_open(inode, file, mode);
+               if (IS_ERR(och))
+                       RETURN(PTR_ERR(och));
+
+               rc = 0;
+               mutex_lock(&lli->lli_och_mutex);
+               if (fd->fd_lease_och == NULL) {
+                       fd->fd_lease_och = och;
+                       och = NULL;
+               }
+               mutex_unlock(&lli->lli_och_mutex);
+               if (och != NULL) {
+                       /* impossible now that only excl is supported for now */
+                       ll_lease_close(och, inode, &lease_broken);
+                       rc = -EBUSY;
+               }
+               RETURN(rc);
+       }
+       case LL_IOC_GET_LEASE: {
+               struct ll_inode_info *lli = ll_i2info(inode);
+               struct ldlm_lock *lock = NULL;
+
+               rc = 0;
+               mutex_lock(&lli->lli_och_mutex);
+               if (fd->fd_lease_och != NULL) {
+                       struct obd_client_handle *och = fd->fd_lease_och;
+
+                       lock = ldlm_handle2lock(&och->och_lease_handle);
+                       if (lock != NULL) {
+                               lock_res_and_lock(lock);
+                               if (!ldlm_is_cancel(lock))
+                                       rc = och->och_flags &
+                                               (FMODE_READ | FMODE_WRITE);
+                               unlock_res_and_lock(lock);
+                               ldlm_lock_put(lock);
+                       }
+               }
+               mutex_unlock(&lli->lli_och_mutex);
+
+               RETURN(rc);
+       }
        default: {
                int err;
 
@@ -2501,18 +2769,20 @@ int ll_fsync(struct file *file, struct dentry *dentry, int datasync)
 
 int ll_file_flock(struct file *file, int cmd, struct file_lock *file_lock)
 {
-        struct inode *inode = file->f_dentry->d_inode;
-        struct ll_sb_info *sbi = ll_i2sbi(inode);
-        struct ldlm_enqueue_info einfo = { .ei_type = LDLM_FLOCK,
-                                           .ei_cb_cp =ldlm_flock_completion_ast,
-                                           .ei_cbdata = file_lock };
-        struct md_op_data *op_data;
-        struct lustre_handle lockh = {0};
-        ldlm_policy_data_t flock = {{0}};
-        int flags = 0;
-        int rc;
+       struct inode *inode = file->f_dentry->d_inode;
+       struct ll_sb_info *sbi = ll_i2sbi(inode);
+       struct ldlm_enqueue_info einfo = {
+               .ei_type        = LDLM_FLOCK,
+               .ei_cb_cp       = ldlm_flock_completion_ast,
+               .ei_cbdata      = file_lock,
+       };
+       struct md_op_data *op_data;
+       struct lustre_handle lockh = {0};
+       ldlm_policy_data_t flock = {{0}};
+       int flags = 0;
+       int rc;
        int rc2 = 0;
-        ENTRY;
+       ENTRY;
 
         CDEBUG(D_VFSTRACE, "VFS Op:inode=%lu file_lock=%p\n",
                inode->i_ino, file_lock);
@@ -2662,11 +2932,11 @@ int ll_have_md_lock(struct inode *inode, __u64 *bits,  ldlm_mode_t l_req_mode)
         CDEBUG(D_INFO, "trying to match res "DFID" mode %s\n", PFID(fid),
                ldlm_lockname[mode]);
 
-        flags = LDLM_FL_BLOCK_GRANTED | LDLM_FL_CBPENDING | LDLM_FL_TEST_LOCK;
-        for (i = 0; i < MDS_INODELOCK_MAXSHIFT && *bits != 0; i++) {
-                policy.l_inodebits.bits = *bits & (1 << i);
-                if (policy.l_inodebits.bits == 0)
-                        continue;
+       flags = LDLM_FL_BLOCK_GRANTED | LDLM_FL_CBPENDING | LDLM_FL_TEST_LOCK;
+       for (i = 0; i <= MDS_INODELOCK_MAXSHIFT && *bits != 0; i++) {
+               policy.l_inodebits.bits = *bits & (1 << i);
+               if (policy.l_inodebits.bits == 0)
+                       continue;
 
                 if (md_lock_match(ll_i2mdexp(inode), flags, fid, LDLM_IBITS,
                                   &policy, mode, &lockh)) {
@@ -2778,7 +3048,7 @@ int __ll_inode_revalidate_it(struct dentry *dentry, struct lookup_intent *it,
                    here to preserve get_cwd functionality on 2.6.
                    Bug 10503 */
                if (!dentry->d_inode->i_nlink)
-                       d_lustre_invalidate(dentry);
+                       d_lustre_invalidate(dentry, 0);
 
                 ll_lookup_finish_locks(&oit, dentry);
         } else if (!ll_have_md_lock(dentry->d_inode, &ibits, LCK_MINMODE)) {
@@ -2819,13 +3089,13 @@ out:
 }
 
 int ll_inode_revalidate_it(struct dentry *dentry, struct lookup_intent *it,
-                           __u64 ibits)
+                          __u64 ibits)
 {
-        struct inode *inode = dentry->d_inode;
-        int rc;
-        ENTRY;
+       struct inode    *inode = dentry->d_inode;
+       int              rc;
+       ENTRY;
 
-        rc = __ll_inode_revalidate_it(dentry, it, ibits);
+       rc = __ll_inode_revalidate_it(dentry, it, ibits);
        if (rc != 0)
                RETURN(rc);
 
@@ -2835,9 +3105,17 @@ int ll_inode_revalidate_it(struct dentry *dentry, struct lookup_intent *it,
                LTIME_S(inode->i_mtime) = ll_i2info(inode)->lli_lvb.lvb_mtime;
                LTIME_S(inode->i_ctime) = ll_i2info(inode)->lli_lvb.lvb_ctime;
        } else {
-               rc = ll_glimpse_size(inode);
+               /* In case of restore, the MDT has the right size and has
+                * already send it back without granting the layout lock,
+                * inode is up-to-date so glimpse is useless.
+                * Also to glimpse we need the layout, in case of a running
+                * restore the MDT holds the layout lock so the glimpse will
+                * block up to the end of restore (getattr will block)
+                */
+               if (!(ll_i2info(inode)->lli_flags & LLIF_FILE_RESTORING))
+                       rc = ll_glimpse_size(inode);
        }
-        RETURN(rc);
+       RETURN(rc);
 }
 
 int ll_getattr_it(struct vfsmount *mnt, struct dentry *de,
@@ -3031,12 +3309,7 @@ struct file_operations ll_file_operations = {
         .release        = ll_file_release,
         .mmap           = ll_file_mmap,
         .llseek         = ll_file_seek,
-#ifdef HAVE_KERNEL_SENDFILE
-        .sendfile       = ll_file_sendfile,
-#endif
-#ifdef HAVE_KERNEL_SPLICE_READ
         .splice_read    = ll_file_splice_read,
-#endif
         .fsync          = ll_fsync,
         .flush          = ll_flush
 };
@@ -3051,12 +3324,7 @@ struct file_operations ll_file_operations_flock = {
         .release        = ll_file_release,
         .mmap           = ll_file_mmap,
         .llseek         = ll_file_seek,
-#ifdef HAVE_KERNEL_SENDFILE
-        .sendfile       = ll_file_sendfile,
-#endif
-#ifdef HAVE_KERNEL_SPLICE_READ
         .splice_read    = ll_file_splice_read,
-#endif
         .fsync          = ll_fsync,
         .flush          = ll_flush,
         .flock          = ll_file_flock,
@@ -3074,12 +3342,7 @@ struct file_operations ll_file_operations_noflock = {
         .release        = ll_file_release,
         .mmap           = ll_file_mmap,
         .llseek         = ll_file_seek,
-#ifdef HAVE_KERNEL_SENDFILE
-        .sendfile       = ll_file_sendfile,
-#endif
-#ifdef HAVE_KERNEL_SPLICE_READ
         .splice_read    = ll_file_splice_read,
-#endif
         .fsync          = ll_fsync,
         .flush          = ll_flush,
         .flock          = ll_file_noflock,
@@ -3250,7 +3513,11 @@ static int ll_layout_fetch(struct inode *inode, struct ldlm_lock *lock)
        int rc;
        ENTRY;
 
-       if (lock->l_lvb_data != NULL)
+       CDEBUG(D_INODE, DFID" LVB_READY=%d l_lvb_data=%p l_lvb_len=%d\n",
+              PFID(ll_inode2fid(inode)), !!(lock->l_flags & LDLM_FL_LVB_READY),
+              lock->l_lvb_data, lock->l_lvb_len);
+
+       if ((lock->l_lvb_data != NULL) && (lock->l_flags & LDLM_FL_LVB_READY))
                RETURN(0);
 
        /* if layout lock was granted right away, the layout is returned
@@ -3286,15 +3553,13 @@ static int ll_layout_fetch(struct inode *inode, struct ldlm_lock *lock)
 
        memcpy(lvbdata, lmm, lmmsize);
        lock_res_and_lock(lock);
-       if (lock->l_lvb_data == NULL) {
-               lock->l_lvb_data = lvbdata;
-               lock->l_lvb_len = lmmsize;
-               lvbdata = NULL;
-       }
+       if (lock->l_lvb_data != NULL)
+               OBD_FREE_LARGE(lock->l_lvb_data, lock->l_lvb_len);
+
+       lock->l_lvb_data = lvbdata;
+       lock->l_lvb_len = lmmsize;
        unlock_res_and_lock(lock);
 
-       if (lvbdata != NULL)
-               OBD_FREE_LARGE(lvbdata, lmmsize);
        EXIT;
 
 out:
@@ -3326,13 +3591,17 @@ static int ll_layout_lock_set(struct lustre_handle *lockh, ldlm_mode_t mode,
        LASSERT(ldlm_has_layout(lock));
 
        LDLM_DEBUG(lock, "File %p/"DFID" being reconfigured: %d.\n",
-               inode, PFID(&lli->lli_fid), reconf);
+                  inode, PFID(&lli->lli_fid), reconf);
+
+       /* in case this is a caching lock and reinstate with new inode */
+       md_set_lock_data(sbi->ll_md_exp, &lockh->cookie, inode, NULL);
 
        lock_res_and_lock(lock);
        lvb_ready = !!(lock->l_flags & LDLM_FL_LVB_READY);
        unlock_res_and_lock(lock);
        /* checking lvb_ready is racy but this is okay. The worst case is
         * that multi processes may configure the file on the same time. */
+
        if (lvb_ready || !reconf) {
                rc = -ENODATA;
                if (lvb_ready) {
@@ -3429,11 +3698,12 @@ int ll_layout_refresh(struct inode *inode, __u32 *gen)
        struct lookup_intent   it;
        struct lustre_handle   lockh;
        ldlm_mode_t            mode;
-       struct ldlm_enqueue_info einfo = { .ei_type = LDLM_IBITS,
-                                          .ei_mode = LCK_CR,
-                                          .ei_cb_bl = ll_md_blocking_ast,
-                                          .ei_cb_cp = ldlm_completion_ast,
-                                          .ei_cbdata = NULL };
+       struct ldlm_enqueue_info einfo = {
+               .ei_type = LDLM_IBITS,
+               .ei_mode = LCK_CR,
+               .ei_cb_bl = ll_md_blocking_ast,
+               .ei_cb_cp = ldlm_completion_ast,
+       };
        int rc;
        ENTRY;
 
@@ -3496,8 +3766,6 @@ again:
 
        ll_finish_md_op_data(op_data);
 
-       md_set_lock_data(sbi->ll_md_exp, &it.d.lustre.it_lock_handle, inode, NULL);
-
        mode = it.d.lustre.it_lock_mode;
        it.d.lustre.it_lock_mode = 0;
        ll_intent_drop_lock(&it);
@@ -3513,3 +3781,32 @@ again:
 
        RETURN(rc);
 }
+
+/**
+ *  This function send a restore request to the MDT
+ */
+int ll_layout_restore(struct inode *inode)
+{
+       struct hsm_user_request *hur;
+       int                      len, rc;
+       ENTRY;
+
+       len = sizeof(struct hsm_user_request) +
+             sizeof(struct hsm_user_item);
+       OBD_ALLOC(hur, len);
+       if (hur == NULL)
+               RETURN(-ENOMEM);
+
+       hur->hur_request.hr_action = HUA_RESTORE;
+       hur->hur_request.hr_archive_id = 0;
+       hur->hur_request.hr_flags = 0;
+       memcpy(&hur->hur_user_item[0].hui_fid, &ll_i2info(inode)->lli_fid,
+              sizeof(hur->hur_user_item[0].hui_fid));
+       hur->hur_user_item[0].hui_extent.length = -1;
+       hur->hur_request.hr_itemcount = 1;
+       rc = obd_iocontrol(LL_IOC_HSM_REQUEST, cl_i2sbi(inode)->ll_md_exp,
+                          len, hur, NULL);
+       OBD_FREE(hur, len);
+       RETURN(rc);
+}
+