X-Git-Url: https://git.whamcloud.com/?p=fs%2Flustre-release.git;a=blobdiff_plain;f=lustre%2Flov%2Flov_object.c;h=cc041c031b821c96c5a68bfc2b345cd52761c3e8;hp=61b5a9f07e60231485e34c0086f02a98e7f42714;hb=707bab62f5d6c704b30e4ee9e769b5c9f026e1e7;hpb=3f3a24dc5d7d421e1514dc49cc7c2eb5cb762b26;ds=sidebyside

diff --git a/lustre/lov/lov_object.c b/lustre/lov/lov_object.c
index 61b5a9f..cc041c0 100644
--- a/lustre/lov/lov_object.c
+++ b/lustre/lov/lov_object.c
@@ -15,11 +15,7 @@
  *
  * You should have received a copy of the GNU General Public License
  * version 2 along with this program; If not, see
- * http://www.sun.com/software/products/lustre/docs/GPLv2.pdf
- *
- * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara,
- * CA 95054 USA or visit www.sun.com if you need additional information or
- * have any questions.
+ * http://www.gnu.org/licenses/gpl-2.0.html
  *
  * GPL HEADER END
  */
@@ -27,7 +23,7 @@
  * Copyright (c) 2008, 2010, Oracle and/or its affiliates. All rights reserved.
  * Use is subject to license terms.
  *
- * Copyright (c) 2011, 2013, Intel Corporation.
+ * Copyright (c) 2011, 2017, Intel Corporation.
  */
 /*
  * This file is part of Lustre, http://www.lustre.org/
@@ -41,8 +37,15 @@
 
 #define DEBUG_SUBSYSTEM S_LOV
 
+#include <linux/random.h>
+
 #include "lov_cl_internal.h"
 
+static inline struct lov_device *lov_object_dev(struct lov_object *obj)
+{
+	return lu2lov_dev(obj->lo_cl.co_lu.lo_dev);
+}
+
 /** \addtogroup lov
  *  @{
  */
@@ -54,16 +57,14 @@
  */
 
 struct lov_layout_operations {
-        int (*llo_init)(const struct lu_env *env, struct lov_device *dev,
-                        struct lov_object *lov,
-                        const struct cl_object_conf *conf,
-                        union lov_layout_state *state);
+	int (*llo_init)(const struct lu_env *env, struct lov_device *dev,
+			struct lov_object *lov, struct lov_stripe_md *lsm,
+			const struct cl_object_conf *conf,
+			union lov_layout_state *state);
 	int (*llo_delete)(const struct lu_env *env, struct lov_object *lov,
                            union lov_layout_state *state);
         void (*llo_fini)(const struct lu_env *env, struct lov_object *lov,
                          union lov_layout_state *state);
-        void (*llo_install)(const struct lu_env *env, struct lov_object *lov,
-                            union lov_layout_state *state);
         int  (*llo_print)(const struct lu_env *env, void *cookie,
                           lu_printer_t p, const struct lu_object *o);
         int  (*llo_page_init)(const struct lu_env *env, struct cl_object *obj,
@@ -75,60 +76,63 @@ struct lov_layout_operations {
                             struct cl_object *obj, struct cl_io *io);
         int  (*llo_getattr)(const struct lu_env *env, struct cl_object *obj,
                             struct cl_attr *attr);
+	int  (*llo_flush)(const struct lu_env *env, struct cl_object *obj,
+			  struct ldlm_lock *lock);
 };
 
 static int lov_layout_wait(const struct lu_env *env, struct lov_object *lov);
 
+static void lov_lsm_put(struct lov_stripe_md *lsm)
+{
+	if (lsm != NULL)
+		lov_free_memmd(&lsm);
+}
+
 /*****************************************************************************
  *
  * Lov object layout operations.
  *
  */
 
-static void lov_install_empty(const struct lu_env *env,
-                              struct lov_object *lov,
-                              union  lov_layout_state *state)
+static struct cl_object *lov_sub_find(const struct lu_env *env,
+				      struct cl_device *dev,
+				      const struct lu_fid *fid,
+				      const struct cl_object_conf *conf)
 {
-        /*
-         * File without objects.
-         */
-}
+	struct lu_object *o;
 
-static int lov_init_empty(const struct lu_env *env,
-                          struct lov_device *dev, struct lov_object *lov,
-                          const struct cl_object_conf *conf,
-                          union  lov_layout_state *state)
-{
-        return 0;
-}
+	ENTRY;
 
-static void lov_install_raid0(const struct lu_env *env,
-                              struct lov_object *lov,
-                              union  lov_layout_state *state)
-{
+	o = lu_object_find_at(env, cl2lu_dev(dev), fid, &conf->coc_lu);
+	LASSERT(ergo(!IS_ERR(o), o->lo_dev->ld_type == &lovsub_device_type));
+	RETURN(lu2cl(o));
 }
 
-static struct cl_object *lov_sub_find(const struct lu_env *env,
-                                      struct cl_device *dev,
-                                      const struct lu_fid *fid,
-                                      const struct cl_object_conf *conf)
+static int lov_page_slice_fixup(struct lov_object *lov,
+				struct cl_object *stripe)
 {
-        struct lu_object *o;
+	struct cl_object_header *hdr = cl_object_header(&lov->lo_cl);
+	struct cl_object *o;
 
-        ENTRY;
-        o = lu_object_find_at(env, cl2lu_dev(dev), fid, &conf->coc_lu);
-        LASSERT(ergo(!IS_ERR(o), o->lo_dev->ld_type == &lovsub_device_type));
-        RETURN(lu2cl(o));
+	if (stripe == NULL)
+		return hdr->coh_page_bufsize - lov->lo_cl.co_slice_off -
+		       cfs_size_round(sizeof(struct lov_page));
+
+	cl_object_for_each(o, stripe)
+		o->co_slice_off += hdr->coh_page_bufsize;
+
+	return cl_object_header(stripe)->coh_page_bufsize;
 }
 
 static int lov_init_sub(const struct lu_env *env, struct lov_object *lov,
-			struct cl_object *stripe, struct lov_layout_raid0 *r0,
+			struct cl_object *subobj, struct lov_oinfo *oinfo,
 			int idx)
 {
 	struct cl_object_header *hdr;
 	struct cl_object_header *subhdr;
 	struct cl_object_header *parent;
-	struct lov_oinfo        *oinfo;
+	int entry = lov_comp_entry(idx);
+	int stripe = lov_comp_stripe(idx);
 	int result;
 
 	if (OBD_FAIL_CHECK(OBD_FAIL_LOV_INIT)) {
@@ -137,32 +141,32 @@ static int lov_init_sub(const struct lu_env *env, struct lov_object *lov,
 		 * freed memory. This is because osc_object is referring to
 		 * lov_oinfo of lsm_stripe_data which will be freed due to
 		 * this failure. */
-		cl_object_kill(env, stripe);
-		cl_object_put(env, stripe);
+		cl_object_kill(env, subobj);
+		cl_object_put(env, subobj);
 		return -EIO;
 	}
 
-	hdr    = cl_object_header(lov2cl(lov));
-	subhdr = cl_object_header(stripe);
+	hdr = cl_object_header(lov2cl(lov));
+	subhdr = cl_object_header(subobj);
 
-	oinfo = lov->lo_lsm->lsm_oinfo[idx];
-	CDEBUG(D_INODE, DFID"@%p[%d] -> "DFID"@%p: ostid: "DOSTID
-	       " idx: %d gen: %d\n",
-	       PFID(&subhdr->coh_lu.loh_fid), subhdr, idx,
-	       PFID(&hdr->coh_lu.loh_fid), hdr, POSTID(&oinfo->loi_oi),
+	CDEBUG(D_INODE, DFID"@%p[%d:%d] -> "DFID"@%p: ostid: "DOSTID
+	       " ost idx: %d gen: %d\n",
+	       PFID(lu_object_fid(&subobj->co_lu)), subhdr, entry, stripe,
+	       PFID(lu_object_fid(lov2lu(lov))), hdr, POSTID(&oinfo->loi_oi),
 	       oinfo->loi_ost_idx, oinfo->loi_ost_gen);
 
 	/* reuse ->coh_attr_guard to protect coh_parent change */
 	spin_lock(&subhdr->coh_attr_guard);
 	parent = subhdr->coh_parent;
 	if (parent == NULL) {
+		struct lovsub_object *lso = cl2lovsub(subobj);
+
 		subhdr->coh_parent = hdr;
 		spin_unlock(&subhdr->coh_attr_guard);
 		subhdr->coh_nesting = hdr->coh_nesting + 1;
-		lu_object_ref_add(&stripe->co_lu, "lov-parent", lov);
-		r0->lo_sub[idx] = cl2lovsub(stripe);
-		r0->lo_sub[idx]->lso_super = lov;
-		r0->lo_sub[idx]->lso_index = idx;
+		lu_object_ref_add(&subobj->co_lu, "lov-parent", lov);
+		lso->lso_super = lov;
+		lso->lso_index = idx;
 		result = 0;
 	} else {
 		struct lu_object  *old_obj;
@@ -176,174 +180,129 @@ static int lov_init_sub(const struct lu_env *env, struct lov_object *lov,
 		if (old_lov->lo_layout_invalid) {
 			/* the object's layout has already changed but isn't
 			 * refreshed */
-			lu_object_unhash(env, &stripe->co_lu);
+			lu_object_unhash(env, &subobj->co_lu);
 			result = -EAGAIN;
 		} else {
 			mask = D_ERROR;
 			result = -EIO;
 		}
 
-		LU_OBJECT_DEBUG(mask, env, &stripe->co_lu,
-				"stripe %d is already owned.\n", idx);
-		LU_OBJECT_DEBUG(mask, env, old_obj, "owned.\n");
+		LU_OBJECT_DEBUG(mask, env, &subobj->co_lu,
+				"stripe %d is already owned.", idx);
+		LU_OBJECT_DEBUG(mask, env, old_obj, "owned.");
 		LU_OBJECT_HEADER(mask, env, lov2lu(lov), "try to own.\n");
-		cl_object_put(env, stripe);
+		cl_object_put(env, subobj);
 	}
 	return result;
 }
 
-static int lov_page_slice_fixup(struct lov_object *lov,
-				struct cl_object *stripe)
+static int lov_init_raid0(const struct lu_env *env, struct lov_device *dev,
+			  struct lov_object *lov, unsigned int index,
+			  const struct cl_object_conf *conf,
+			  struct lov_layout_entry *lle)
 {
-	struct cl_object_header *hdr = cl_object_header(&lov->lo_cl);
-	struct cl_object *o;
-
-	if (stripe == NULL)
-		return hdr->coh_page_bufsize - lov->lo_cl.co_slice_off -
-		       cfs_size_round(sizeof(struct lov_page));
-
-	cl_object_for_each(o, stripe)
-		o->co_slice_off += hdr->coh_page_bufsize;
+	struct lov_layout_raid0 *r0 = &lle->lle_raid0;
+	struct lov_thread_info *lti = lov_env_info(env);
+	struct cl_object_conf *subconf = &lti->lti_stripe_conf;
+	struct lu_fid *ofid = &lti->lti_fid;
+	struct cl_object *stripe;
+	struct lov_stripe_md_entry *lse  = lov_lse(lov, index);
+	int result;
+	int psz, sz;
+	int i;
 
-	return cl_object_header(stripe)->coh_page_bufsize;
-}
+	ENTRY;
 
-static int lov_init_raid0(const struct lu_env *env,
-                          struct lov_device *dev, struct lov_object *lov,
-                          const struct cl_object_conf *conf,
-                          union  lov_layout_state *state)
-{
-        int result;
-        int i;
+	spin_lock_init(&r0->lo_sub_lock);
+	r0->lo_nr = lse->lsme_stripe_count;
 
-        struct cl_object        *stripe;
-        struct lov_thread_info  *lti     = lov_env_info(env);
-        struct cl_object_conf   *subconf = &lti->lti_stripe_conf;
-        struct lov_stripe_md    *lsm     = conf->u.coc_md->lsm;
-        struct lu_fid           *ofid    = &lti->lti_fid;
-        struct lov_layout_raid0 *r0      = &state->raid0;
+	OBD_ALLOC_LARGE(r0->lo_sub, r0->lo_nr * sizeof(r0->lo_sub[0]));
+	if (r0->lo_sub == NULL)
+		GOTO(out, result = -ENOMEM);
 
-        ENTRY;
+	psz = 0;
+	result = 0;
+	memset(subconf, 0, sizeof(*subconf));
 
-	if (lsm->lsm_magic != LOV_MAGIC_V1 && lsm->lsm_magic != LOV_MAGIC_V3) {
-		dump_lsm(D_ERROR, lsm);
-		LASSERTF(0, "magic mismatch, expected %d/%d, actual %d.\n",
-			 LOV_MAGIC_V1, LOV_MAGIC_V3, lsm->lsm_magic);
-	}
+	/*
+	 * Create stripe cl_objects.
+	 */
+	for (i = 0; i < r0->lo_nr; ++i) {
+		struct cl_device *subdev;
+		struct lov_oinfo *oinfo = lse->lsme_oinfo[i];
+		int ost_idx = oinfo->loi_ost_idx;
 
-	LASSERT(lov->lo_lsm == NULL);
-	lov->lo_lsm = lsm_addref(lsm);
-	r0->lo_nr  = lsm->lsm_stripe_count;
-	LASSERT(r0->lo_nr <= lov_targets_nr(dev));
+		if (lov_oinfo_is_dummy(oinfo))
+			continue;
 
-	lov->lo_layout_invalid = true;
+		result = ostid_to_fid(ofid, &oinfo->loi_oi, oinfo->loi_ost_idx);
+		if (result != 0)
+			GOTO(out, result);
 
-	OBD_ALLOC_LARGE(r0->lo_sub, r0->lo_nr * sizeof r0->lo_sub[0]);
-	if (r0->lo_sub != NULL) {
-		int psz = 0;
+		if (dev->ld_target[ost_idx] == NULL) {
+			CERROR("%s: OST %04x is not initialized\n",
+			       lov2obd(dev->ld_lov)->obd_name, ost_idx);
+			GOTO(out, result = -EIO);
+		}
 
-		result = 0;
-		subconf->coc_inode = conf->coc_inode;
-		spin_lock_init(&r0->lo_sub_lock);
-                /*
-                 * Create stripe cl_objects.
-                 */
-                for (i = 0; i < r0->lo_nr && result == 0; ++i) {
-                        struct cl_device *subdev;
-                        struct lov_oinfo *oinfo = lsm->lsm_oinfo[i];
-                        int ost_idx = oinfo->loi_ost_idx;
-
-			if (lov_oinfo_is_dummy(oinfo))
-				continue;
+		subdev = lovsub2cl_dev(dev->ld_target[ost_idx]);
+		subconf->u.coc_oinfo = oinfo;
+		LASSERTF(subdev != NULL, "not init ost %d\n", ost_idx);
+		/* In the function below, .hs_keycmp resolves to
+		 * lu_obj_hop_keycmp() */
+		/* coverity[overrun-buffer-val] */
+		stripe = lov_sub_find(env, subdev, ofid, subconf);
+		if (IS_ERR(stripe))
+			GOTO(out, result = PTR_ERR(stripe));
+
+		result = lov_init_sub(env, lov, stripe, oinfo,
+				      lov_comp_index(index, i));
+		if (result == -EAGAIN) { /* try again */
+			--i;
+			result = 0;
+			continue;
+		}
 
-			result = ostid_to_fid(ofid, &oinfo->loi_oi,
-					      oinfo->loi_ost_idx);
-			if (result != 0)
-				GOTO(out, result);
-
-			subdev = lovsub2cl_dev(dev->ld_target[ost_idx]);
-			subconf->u.coc_oinfo = oinfo;
-			LASSERTF(subdev != NULL, "not init ost %d\n", ost_idx);
-			/* In the function below, .hs_keycmp resolves to
-			 * lu_obj_hop_keycmp() */
-			/* coverity[overrun-buffer-val] */
-			stripe = lov_sub_find(env, subdev, ofid, subconf);
-			if (!IS_ERR(stripe)) {
-				result = lov_init_sub(env, lov, stripe, r0, i);
-				if (result == -EAGAIN) { /* try again */
-					--i;
-					result = 0;
-					continue;
-				}
-			} else {
-				result = PTR_ERR(stripe);
-			}
+		if (result == 0) {
+			r0->lo_sub[i] = cl2lovsub(stripe);
 
-			if (result == 0) {
-				int sz = lov_page_slice_fixup(lov, stripe);
-				LASSERT(ergo(psz > 0, psz == sz));
-				psz = sz;
-			}
-                }
-		if (result == 0)
-			cl_object_header(&lov->lo_cl)->coh_page_bufsize += psz;
-	} else
-		result = -ENOMEM;
+			sz = lov_page_slice_fixup(lov, stripe);
+			LASSERT(ergo(psz > 0, psz == sz));
+			psz = sz;
+		}
+	}
+	if (result == 0)
+		result = psz;
 out:
 	RETURN(result);
 }
 
-static int lov_init_released(const struct lu_env *env,
-			struct lov_device *dev, struct lov_object *lov,
-			const struct cl_object_conf *conf,
-			union  lov_layout_state *state)
-{
-	struct lov_stripe_md *lsm = conf->u.coc_md->lsm;
-
-	LASSERT(lsm != NULL);
-	LASSERT(lsm_is_released(lsm));
-	LASSERT(lov->lo_lsm == NULL);
-
-	lov->lo_lsm = lsm_addref(lsm);
-	return 0;
-}
-
-static int lov_delete_empty(const struct lu_env *env, struct lov_object *lov,
-			    union lov_layout_state *state)
-{
-	LASSERT(lov->lo_type == LLT_EMPTY || lov->lo_type == LLT_RELEASED);
-
-	lov_layout_wait(env, lov);
-	return 0;
-}
-
 static void lov_subobject_kill(const struct lu_env *env, struct lov_object *lov,
+			       struct lov_layout_raid0 *r0,
 			       struct lovsub_object *los, int idx)
 {
 	struct cl_object        *sub;
-	struct lov_layout_raid0 *r0;
 	struct lu_site          *site;
-	struct lu_site_bkt_data *bkt;
-	wait_queue_t          *waiter;
+	wait_queue_head_t *wq;
+	wait_queue_entry_t *waiter;
 
-        r0  = &lov->u.raid0;
         LASSERT(r0->lo_sub[idx] == los);
 
-        sub  = lovsub2cl(los);
-        site = sub->co_lu.lo_dev->ld_site;
-        bkt  = lu_site_bkt_from_fid(site, &sub->co_lu.lo_header->loh_fid);
+	sub = lovsub2cl(los);
+	site = sub->co_lu.lo_dev->ld_site;
+	wq = lu_site_wq_from_fid(site, &sub->co_lu.lo_header->loh_fid);
 
         cl_object_kill(env, sub);
         /* release a reference to the sub-object and ... */
         lu_object_ref_del(&sub->co_lu, "lov-parent", lov);
         cl_object_put(env, sub);
 
-        /* ... wait until it is actually destroyed---sub-object clears its
-         * ->lo_sub[] slot in lovsub_object_fini() */
+	/* ... wait until it is actually destroyed---sub-object clears its
+	 * ->lo_sub[] slot in lovsub_object_free() */
 	if (r0->lo_sub[idx] == los) {
 		waiter = &lov_env_info(env)->lti_waiter;
-		init_waitqueue_entry_current(waiter);
-		add_wait_queue(&bkt->lsb_marche_funebre, waiter);
+		init_waitqueue_entry(waiter, current);
+		add_wait_queue(wq, waiter);
 		set_current_state(TASK_UNINTERRUPTIBLE);
 		while (1) {
 			/* this wait-queue is signaled at the end of
@@ -352,44 +311,548 @@ static void lov_subobject_kill(const struct lu_env *env, struct lov_object *lov,
 			spin_lock(&r0->lo_sub_lock);
 			if (r0->lo_sub[idx] == los) {
 				spin_unlock(&r0->lo_sub_lock);
-				waitq_wait(waiter, TASK_UNINTERRUPTIBLE);
+				schedule();
 			} else {
 				spin_unlock(&r0->lo_sub_lock);
 				set_current_state(TASK_RUNNING);
 				break;
 			}
 		}
-		remove_wait_queue(&bkt->lsb_marche_funebre, waiter);
+		remove_wait_queue(wq, waiter);
 	}
 	LASSERT(r0->lo_sub[idx] == NULL);
 }
 
-static int lov_delete_raid0(const struct lu_env *env, struct lov_object *lov,
-			    union lov_layout_state *state)
+static void lov_delete_raid0(const struct lu_env *env, struct lov_object *lov,
+			     struct lov_layout_entry *lle)
 {
-	struct lov_layout_raid0 *r0 = &state->raid0;
-	struct lov_stripe_md    *lsm = lov->lo_lsm;
+	struct lov_layout_raid0 *r0 = &lle->lle_raid0;
+
+	ENTRY;
+
+        if (r0->lo_sub != NULL) {
+		int i;
+
+		for (i = 0; i < r0->lo_nr; ++i) {
+			struct lovsub_object *los = r0->lo_sub[i];
+
+			if (los != NULL) {
+				cl_object_prune(env, &los->lso_cl);
+				/*
+				 * If top-level object is to be evicted from
+				 * the cache, so are its sub-objects.
+				 */
+				lov_subobject_kill(env, lov, r0, los, i);
+			}
+		}
+	}
+
+	EXIT;
+}
+
+static void lov_fini_raid0(const struct lu_env *env,
+			   struct lov_layout_entry *lle)
+{
+	struct lov_layout_raid0 *r0 = &lle->lle_raid0;
+
+	if (r0->lo_sub != NULL) {
+		OBD_FREE_LARGE(r0->lo_sub, r0->lo_nr * sizeof r0->lo_sub[0]);
+		r0->lo_sub = NULL;
+	}
+}
+
+static int lov_print_raid0(const struct lu_env *env, void *cookie,
+			   lu_printer_t p, const struct lov_layout_entry *lle)
+{
+	const struct lov_layout_raid0 *r0 = &lle->lle_raid0;
 	int i;
 
+	for (i = 0; i < r0->lo_nr; ++i) {
+		struct lu_object *sub;
+
+		if (r0->lo_sub[i] != NULL) {
+			sub = lovsub2lu(r0->lo_sub[i]);
+			lu_object_print(env, cookie, p, sub);
+		} else {
+			(*p)(env, cookie, "sub %d absent\n", i);
+		}
+	}
+	return 0;
+}
+
+static int lov_attr_get_raid0(const struct lu_env *env, struct lov_object *lov,
+			      unsigned int index, struct lov_layout_entry *lle,
+			      struct cl_attr **lov_attr)
+{
+	struct lov_layout_raid0 *r0 = &lle->lle_raid0;
+	struct lov_stripe_md *lsm = lov->lo_lsm;
+	struct ost_lvb *lvb = &lov_env_info(env)->lti_lvb;
+	struct cl_attr *attr = &r0->lo_attr;
+	__u64 kms = 0;
+	int result = 0;
+
+	if (r0->lo_attr_valid) {
+		*lov_attr = attr;
+		return 0;
+	}
+
+	memset(lvb, 0, sizeof(*lvb));
+
+	/* XXX: timestamps can be negative by sanity:test_39m,
+	 * how can it be? */
+	lvb->lvb_atime = LLONG_MIN;
+	lvb->lvb_ctime = LLONG_MIN;
+	lvb->lvb_mtime = LLONG_MIN;
+
+	/*
+	 * XXX that should be replaced with a loop over sub-objects,
+	 * doing cl_object_attr_get() on them. But for now, let's
+	 * reuse old lov code.
+	 */
+
+	/*
+	 * XXX take lsm spin-lock to keep lov_merge_lvb_kms()
+	 * happy. It's not needed, because new code uses
+	 * ->coh_attr_guard spin-lock to protect consistency of
+	 * sub-object attributes.
+	 */
+	lov_stripe_lock(lsm);
+	result = lov_merge_lvb_kms(lsm, index, lvb, &kms);
+	lov_stripe_unlock(lsm);
+	if (result == 0) {
+		cl_lvb2attr(attr, lvb);
+		attr->cat_kms = kms;
+		r0->lo_attr_valid = 1;
+		*lov_attr = attr;
+	}
+
+	return result;
+}
+
+static struct lov_comp_layout_entry_ops raid0_ops = {
+	.lco_init      = lov_init_raid0,
+	.lco_fini      = lov_fini_raid0,
+	.lco_getattr   = lov_attr_get_raid0,
+};
+
+static int lov_attr_get_dom(const struct lu_env *env, struct lov_object *lov,
+			    unsigned int index, struct lov_layout_entry *lle,
+			    struct cl_attr **lov_attr)
+{
+	struct lov_layout_dom *dom = &lle->lle_dom;
+	struct lov_oinfo *loi = dom->lo_loi;
+	struct cl_attr *attr = &dom->lo_dom_r0.lo_attr;
+
+	if (dom->lo_dom_r0.lo_attr_valid) {
+		*lov_attr = attr;
+		return 0;
+	}
+
+	if (OST_LVB_IS_ERR(loi->loi_lvb.lvb_blocks))
+		return OST_LVB_GET_ERR(loi->loi_lvb.lvb_blocks);
+
+	cl_lvb2attr(attr, &loi->loi_lvb);
+
+	/* DoM component size can be bigger than stripe size after
+	 * client's setattr RPC, so do not count anything beyond
+	 * component end. Alternatively, check that limit on server
+	 * and do not allow size overflow there. */
+	if (attr->cat_size > lle->lle_extent->e_end)
+		attr->cat_size = lle->lle_extent->e_end;
+
+	attr->cat_kms = attr->cat_size;
+
+	dom->lo_dom_r0.lo_attr_valid = 1;
+	*lov_attr = attr;
+
+	return 0;
+}
+
+/**
+ * Lookup FLD to get MDS index of the given DOM object FID.
+ *
+ * \param[in]  ld	LOV device
+ * \param[in]  fid	FID to lookup
+ * \param[out] nr	index in MDC array to return back
+ *
+ * \retval		0 and \a mds filled with MDS index if successful
+ * \retval		negative value on error
+ */
+static int lov_fld_lookup(struct lov_device *ld, const struct lu_fid *fid,
+			  __u32 *nr)
+{
+	__u32 mds_idx;
+	int i, rc;
+
 	ENTRY;
 
+	rc = fld_client_lookup(&ld->ld_lmv->u.lmv.lmv_fld, fid_seq(fid),
+			       &mds_idx, LU_SEQ_RANGE_MDT, NULL);
+	if (rc) {
+		CERROR("%s: error while looking for mds number. Seq %#llx"
+		       ", err = %d\n", lu_dev_name(cl2lu_dev(&ld->ld_cl)),
+		       fid_seq(fid), rc);
+		RETURN(rc);
+	}
+
+	CDEBUG(D_INODE, "FLD lookup got mds #%x for fid="DFID"\n",
+	       mds_idx, PFID(fid));
+
+	/* find proper MDC device in the array */
+	for (i = 0; i < ld->ld_md_tgts_nr; i++) {
+		if (ld->ld_md_tgts[i].ldm_mdc != NULL &&
+		    ld->ld_md_tgts[i].ldm_idx == mds_idx)
+			break;
+	}
+
+	if (i == ld->ld_md_tgts_nr) {
+		CERROR("%s: cannot find corresponding MDC device for mds #%x "
+		       "for fid="DFID"\n", lu_dev_name(cl2lu_dev(&ld->ld_cl)),
+		       mds_idx, PFID(fid));
+		rc = -EINVAL;
+	} else {
+		*nr = i;
+	}
+	RETURN(rc);
+}
+
+/**
+ * Implementation of lov_comp_layout_entry_ops::lco_init for DOM object.
+ *
+ * Init the DOM object for the first time. It prepares also RAID0 entry
+ * for it to use in common methods with ordinary RAID0 layout entries.
+ *
+ * \param[in] env	execution environment
+ * \param[in] dev	LOV device
+ * \param[in] lov	LOV object
+ * \param[in] index	Composite layout entry index in LSM
+ * \param[in] lle	Composite LOV layout entry
+ */
+static int lov_init_dom(const struct lu_env *env, struct lov_device *dev,
+			struct lov_object *lov, unsigned int index,
+			const struct cl_object_conf *conf,
+			struct lov_layout_entry *lle)
+{
+	struct lov_thread_info *lti = lov_env_info(env);
+	struct lov_stripe_md_entry *lsme = lov_lse(lov, index);
+	struct cl_object *clo;
+	struct lu_object *o = lov2lu(lov);
+	const struct lu_fid *fid = lu_object_fid(o);
+	struct cl_device *mdcdev;
+	struct lov_oinfo *loi = NULL;
+	struct cl_object_conf *sconf = &lti->lti_stripe_conf;
+
+	int rc;
+	__u32 idx = 0;
+
+	ENTRY;
+
+	LASSERT(index == 0);
+
+	/* find proper MDS device */
+	rc = lov_fld_lookup(dev, fid, &idx);
+	if (rc)
+		RETURN(rc);
+
+	LASSERTF(dev->ld_md_tgts[idx].ldm_mdc != NULL,
+		 "LOV md target[%u] is NULL\n", idx);
+
+	/* check lsm is DOM, more checks are needed */
+	LASSERT(lsme->lsme_stripe_count == 0);
+
+	/*
+	 * Create lower cl_objects.
+	 */
+	mdcdev = dev->ld_md_tgts[idx].ldm_mdc;
+
+	LASSERTF(mdcdev != NULL, "non-initialized mdc subdev\n");
+
+	/* DoM object has no oinfo in LSM entry, create it exclusively */
+	OBD_SLAB_ALLOC_PTR_GFP(loi, lov_oinfo_slab, GFP_NOFS);
+	if (loi == NULL)
+		RETURN(-ENOMEM);
+
+	fid_to_ostid(lu_object_fid(lov2lu(lov)), &loi->loi_oi);
+
+	sconf->u.coc_oinfo = loi;
+again:
+	clo = lov_sub_find(env, mdcdev, fid, sconf);
+	if (IS_ERR(clo))
+		GOTO(out, rc = PTR_ERR(clo));
+
+	rc = lov_init_sub(env, lov, clo, loi, lov_comp_index(index, 0));
+	if (rc == -EAGAIN) /* try again */
+		goto again;
+	else if (rc != 0)
+		GOTO(out, rc);
+
+	lle->lle_dom.lo_dom = cl2lovsub(clo);
+	spin_lock_init(&lle->lle_dom.lo_dom_r0.lo_sub_lock);
+	lle->lle_dom.lo_dom_r0.lo_nr = 1;
+	lle->lle_dom.lo_dom_r0.lo_sub = &lle->lle_dom.lo_dom;
+	lle->lle_dom.lo_loi = loi;
+
+	rc = lov_page_slice_fixup(lov, clo);
+	RETURN(rc);
+
+out:
+	if (loi != NULL)
+		OBD_SLAB_FREE_PTR(loi, lov_oinfo_slab);
+	return rc;
+}
+
+/**
+ * Implementation of lov_layout_operations::llo_fini for DOM object.
+ *
+ * Finish the DOM object and free related memory.
+ *
+ * \param[in] env	execution environment
+ * \param[in] lov	LOV object
+ * \param[in] state	LOV layout state
+ */
+static void lov_fini_dom(const struct lu_env *env,
+			 struct lov_layout_entry *lle)
+{
+	if (lle->lle_dom.lo_dom != NULL)
+		lle->lle_dom.lo_dom = NULL;
+	if (lle->lle_dom.lo_loi != NULL)
+		OBD_SLAB_FREE_PTR(lle->lle_dom.lo_loi, lov_oinfo_slab);
+}
+
+static struct lov_comp_layout_entry_ops dom_ops = {
+	.lco_init = lov_init_dom,
+	.lco_fini = lov_fini_dom,
+	.lco_getattr = lov_attr_get_dom,
+};
+
+static int lov_init_composite(const struct lu_env *env, struct lov_device *dev,
+			      struct lov_object *lov, struct lov_stripe_md *lsm,
+			      const struct cl_object_conf *conf,
+			      union lov_layout_state *state)
+{
+	struct lov_layout_composite *comp = &state->composite;
+	struct lov_layout_entry *lle;
+	struct lov_mirror_entry *lre;
+	unsigned int entry_count;
+	unsigned int psz = 0;
+	unsigned int mirror_count;
+	int flr_state = lsm->lsm_flags & LCM_FL_FLR_MASK;
+	int result = 0;
+	unsigned int seq;
+	int i, j;
+
+	ENTRY;
+
+	LASSERT(lsm->lsm_entry_count > 0);
+	LASSERT(lov->lo_lsm == NULL);
+	lov->lo_lsm = lsm_addref(lsm);
+	lov->lo_layout_invalid = true;
+
 	dump_lsm(D_INODE, lsm);
 
-	lov_layout_wait(env, lov);
-        if (r0->lo_sub != NULL) {
-                for (i = 0; i < r0->lo_nr; ++i) {
-                        struct lovsub_object *los = r0->lo_sub[i];
+	entry_count = lsm->lsm_entry_count;
 
-                        if (los != NULL) {
-				cl_object_prune(env, &los->lso_cl);
-                                /*
-                                 * If top-level object is to be evicted from
-                                 * the cache, so are its sub-objects.
-                                 */
-                                lov_subobject_kill(env, lov, los, i);
+	spin_lock_init(&comp->lo_write_lock);
+	comp->lo_flags = lsm->lsm_flags;
+	comp->lo_mirror_count = lsm->lsm_mirror_count + 1;
+	comp->lo_entry_count = lsm->lsm_entry_count;
+	comp->lo_preferred_mirror = -1;
+
+	if (equi(flr_state == LCM_FL_NONE, comp->lo_mirror_count > 1))
+		RETURN(-EINVAL);
+
+	OBD_ALLOC(comp->lo_mirrors,
+		  comp->lo_mirror_count * sizeof(*comp->lo_mirrors));
+	if (comp->lo_mirrors == NULL)
+		RETURN(-ENOMEM);
+
+	OBD_ALLOC(comp->lo_entries, entry_count * sizeof(*comp->lo_entries));
+	if (comp->lo_entries == NULL)
+		RETURN(-ENOMEM);
+
+	/* Initiate all entry types and extents data at first */
+	for (i = 0, j = 0, mirror_count = 1; i < entry_count; i++) {
+		int mirror_id = 0;
+
+		lle = &comp->lo_entries[i];
+
+		lle->lle_lsme = lsm->lsm_entries[i];
+		lle->lle_type = lov_entry_type(lle->lle_lsme);
+		switch (lle->lle_type) {
+		case LOV_PATTERN_RAID0:
+			lle->lle_comp_ops = &raid0_ops;
+			break;
+		case LOV_PATTERN_MDT:
+			lle->lle_comp_ops = &dom_ops;
+			break;
+		default:
+			CERROR("%s: unknown composite layout entry type %i\n",
+			       lov2obd(dev->ld_lov)->obd_name,
+			       lsm->lsm_entries[i]->lsme_pattern);
+			dump_lsm(D_ERROR, lsm);
+			RETURN(-EIO);
+		}
+
+		lle->lle_extent = &lle->lle_lsme->lsme_extent;
+		lle->lle_valid = !(lle->lle_lsme->lsme_flags & LCME_FL_STALE);
+
+		if (flr_state != LCM_FL_NONE)
+			mirror_id = mirror_id_of(lle->lle_lsme->lsme_id);
+
+		lre = &comp->lo_mirrors[j];
+		if (i > 0) {
+			if (mirror_id == lre->lre_mirror_id) {
+				lre->lre_valid |= lle->lle_valid;
+				lre->lre_stale |= !lle->lle_valid;
+				lre->lre_end = i;
+				continue;
 			}
+
+			/* new mirror detected, assume that the mirrors
+			 * are shorted in layout */
+			++mirror_count;
+			++j;
+			if (j >= comp->lo_mirror_count)
+				break;
+
+			lre = &comp->lo_mirrors[j];
 		}
+
+		/* entries must be sorted by mirrors */
+		lre->lre_mirror_id = mirror_id;
+		lre->lre_start = lre->lre_end = i;
+		lre->lre_preferred = !!(lle->lle_lsme->lsme_flags &
+					LCME_FL_PREF_RD);
+		lre->lre_valid = lle->lle_valid;
+		lre->lre_stale = !lle->lle_valid;
+	}
+
+	/* sanity check for FLR */
+	if (mirror_count != comp->lo_mirror_count) {
+		CDEBUG(D_INODE, DFID
+		       " doesn't have the # of mirrors it claims, %u/%u\n",
+		       PFID(lu_object_fid(lov2lu(lov))), mirror_count,
+		       comp->lo_mirror_count + 1);
+
+		GOTO(out, result = -EINVAL);
+	}
+
+	lov_foreach_layout_entry(lov, lle) {
+		int index = lov_layout_entry_index(lov, lle);
+
+		/**
+		 * If the component has not been init-ed on MDS side, for
+		 * PFL layout, we'd know that the components beyond this one
+		 * will be dynamically init-ed later on file write/trunc ops.
+		 */
+		if (!lsme_inited(lle->lle_lsme))
+			continue;
+
+		result = lle->lle_comp_ops->lco_init(env, dev, lov, index,
+						     conf, lle);
+		if (result < 0)
+			break;
+
+		LASSERT(ergo(psz > 0, psz == result));
+		psz = result;
+	}
+
+	if (psz > 0)
+		cl_object_header(&lov->lo_cl)->coh_page_bufsize += psz;
+
+	/* decide the preferred mirror. It uses the hash value of lov_object
+	 * so that different clients would use different mirrors for read. */
+	mirror_count = 0;
+	seq = hash_long((unsigned long)lov, 8);
+	for (i = 0; i < comp->lo_mirror_count; i++) {
+		unsigned int idx = (i + seq) % comp->lo_mirror_count;
+
+		lre = lov_mirror_entry(lov, idx);
+		if (lre->lre_stale)
+			continue;
+
+		mirror_count++; /* valid mirror */
+
+		if (lre->lre_preferred || comp->lo_preferred_mirror < 0)
+			comp->lo_preferred_mirror = idx;
 	}
+	if (!mirror_count) {
+		CDEBUG(D_INODE, DFID
+		       " doesn't have any valid mirrors\n",
+		       PFID(lu_object_fid(lov2lu(lov))));
+
+		comp->lo_preferred_mirror = 0;
+	}
+
+	LASSERT(comp->lo_preferred_mirror >= 0);
+
+	EXIT;
+out:
+	return result > 0 ? 0 : result;
+}
+
+static int lov_init_empty(const struct lu_env *env, struct lov_device *dev,
+			  struct lov_object *lov, struct lov_stripe_md *lsm,
+			  const struct cl_object_conf *conf,
+			  union lov_layout_state *state)
+{
+	return 0;
+}
+
+static int lov_init_released(const struct lu_env *env,
+			     struct lov_device *dev, struct lov_object *lov,
+			     struct lov_stripe_md *lsm,
+			     const struct cl_object_conf *conf,
+			     union lov_layout_state *state)
+{
+	LASSERT(lsm != NULL);
+	LASSERT(lsm->lsm_is_released);
+	LASSERT(lov->lo_lsm == NULL);
+
+	lov->lo_lsm = lsm_addref(lsm);
+	return 0;
+}
+
+static int lov_init_foreign(const struct lu_env *env,
+			    struct lov_device *dev, struct lov_object *lov,
+			    struct lov_stripe_md *lsm,
+			    const struct cl_object_conf *conf,
+			    union lov_layout_state *state)
+{
+	LASSERT(lsm != NULL);
+	LASSERT(lov->lo_type == LLT_FOREIGN);
+	LASSERT(lov->lo_lsm == NULL);
+
+	lov->lo_lsm = lsm_addref(lsm);
+	return 0;
+}
+
+static int lov_delete_empty(const struct lu_env *env, struct lov_object *lov,
+			    union lov_layout_state *state)
+{
+	LASSERT(lov->lo_type == LLT_EMPTY || lov->lo_type == LLT_RELEASED ||
+		lov->lo_type == LLT_FOREIGN);
+
+	lov_layout_wait(env, lov);
+	return 0;
+}
+
+static int lov_delete_composite(const struct lu_env *env,
+				struct lov_object *lov,
+				union lov_layout_state *state)
+{
+	struct lov_layout_entry *entry;
+	struct lov_layout_composite *comp = &state->composite;
+
+	ENTRY;
+
+	dump_lsm(D_INODE, lov->lo_lsm);
+
+	lov_layout_wait(env, lov);
+	if (comp->lo_entries)
+		lov_foreach_layout_entry(lov, entry)
+			lov_delete_raid0(env, lov, entry);
+
 	RETURN(0);
 }
 
@@ -399,17 +862,32 @@ static void lov_fini_empty(const struct lu_env *env, struct lov_object *lov,
 	LASSERT(lov->lo_type == LLT_EMPTY || lov->lo_type == LLT_RELEASED);
 }
 
-static void lov_fini_raid0(const struct lu_env *env, struct lov_object *lov,
-                           union lov_layout_state *state)
+static void lov_fini_composite(const struct lu_env *env,
+			       struct lov_object *lov,
+			       union lov_layout_state *state)
 {
-	struct lov_layout_raid0 *r0 = &state->raid0;
+	struct lov_layout_composite *comp = &state->composite;
 	ENTRY;
 
-	if (r0->lo_sub != NULL) {
-		OBD_FREE_LARGE(r0->lo_sub, r0->lo_nr * sizeof r0->lo_sub[0]);
-		r0->lo_sub = NULL;
+	if (comp->lo_entries != NULL) {
+		struct lov_layout_entry *entry;
+
+		lov_foreach_layout_entry(lov, entry)
+			entry->lle_comp_ops->lco_fini(env, entry);
+
+		OBD_FREE(comp->lo_entries,
+			 comp->lo_entry_count * sizeof(*comp->lo_entries));
+		comp->lo_entries = NULL;
 	}
 
+	if (comp->lo_mirrors != NULL) {
+		OBD_FREE(comp->lo_mirrors,
+			 comp->lo_mirror_count * sizeof(*comp->lo_mirrors));
+		comp->lo_mirrors = NULL;
+	}
+
+	memset(comp, 0, sizeof(*comp));
+
 	dump_lsm(D_INODE, lov->lo_lsm);
 	lov_free_memmd(&lov->lo_lsm);
 
@@ -432,42 +910,63 @@ static int lov_print_empty(const struct lu_env *env, void *cookie,
         return 0;
 }
 
-static int lov_print_raid0(const struct lu_env *env, void *cookie,
-			   lu_printer_t p, const struct lu_object *o)
+static int lov_print_composite(const struct lu_env *env, void *cookie,
+			       lu_printer_t p, const struct lu_object *o)
+{
+	struct lov_object *lov = lu2lov(o);
+	struct lov_stripe_md *lsm = lov->lo_lsm;
+	int i;
+
+	(*p)(env, cookie, "entries: %d, %s, lsm{%p 0x%08X %d %u}:\n",
+	     lsm->lsm_entry_count,
+	     lov->lo_layout_invalid ? "invalid" : "valid", lsm,
+	     lsm->lsm_magic, atomic_read(&lsm->lsm_refc),
+	     lsm->lsm_layout_gen);
+
+	for (i = 0; i < lsm->lsm_entry_count; i++) {
+		struct lov_stripe_md_entry *lse = lsm->lsm_entries[i];
+		struct lov_layout_entry *lle = lov_entry(lov, i);
+
+		(*p)(env, cookie,
+		     DEXT ": { 0x%08X, %u, %#x, %u, %#x, %u, %u }\n",
+		     PEXT(&lse->lsme_extent), lse->lsme_magic,
+		     lse->lsme_id, lse->lsme_pattern, lse->lsme_layout_gen,
+		     lse->lsme_flags, lse->lsme_stripe_count,
+		     lse->lsme_stripe_size);
+		lov_print_raid0(env, cookie, p, lle);
+	}
+
+	return 0;
+}
+
+static int lov_print_released(const struct lu_env *env, void *cookie,
+				lu_printer_t p, const struct lu_object *o)
 {
 	struct lov_object	*lov = lu2lov(o);
-	struct lov_layout_raid0	*r0  = lov_r0(lov);
 	struct lov_stripe_md	*lsm = lov->lo_lsm;
-	int			 i;
 
-	(*p)(env, cookie, "stripes: %d, %s, lsm{%p 0x%08X %d %u %u}:\n",
-		r0->lo_nr, lov->lo_layout_invalid ? "invalid" : "valid", lsm,
+	(*p)(env, cookie,
+		"released: %s, lsm{%p 0x%08X %d %u}:\n",
+		lov->lo_layout_invalid ? "invalid" : "valid", lsm,
 		lsm->lsm_magic, atomic_read(&lsm->lsm_refc),
-		lsm->lsm_stripe_count, lsm->lsm_layout_gen);
-	for (i = 0; i < r0->lo_nr; ++i) {
-		struct lu_object *sub;
-
-		if (r0->lo_sub[i] != NULL) {
-			sub = lovsub2lu(r0->lo_sub[i]);
-			lu_object_print(env, cookie, p, sub);
-		} else {
-			(*p)(env, cookie, "sub %d absent\n", i);
-		}
-	}
+		lsm->lsm_layout_gen);
 	return 0;
 }
 
-static int lov_print_released(const struct lu_env *env, void *cookie,
+static int lov_print_foreign(const struct lu_env *env, void *cookie,
 				lu_printer_t p, const struct lu_object *o)
 {
 	struct lov_object	*lov = lu2lov(o);
 	struct lov_stripe_md	*lsm = lov->lo_lsm;
 
 	(*p)(env, cookie,
-		"released: %s, lsm{%p 0x%08X %d %u %u}:\n",
+		"foreign: %s, lsm{%p 0x%08X %d %u}:\n",
 		lov->lo_layout_invalid ? "invalid" : "valid", lsm,
 		lsm->lsm_magic, atomic_read(&lsm->lsm_refc),
-		lsm->lsm_stripe_count, lsm->lsm_layout_gen);
+		lsm->lsm_layout_gen);
+	(*p)(env, cookie,
+		"raw_ea_content '%.*s'\n",
+		(int)lsm->lsm_foreign_size, (char *)lsm_foreign(lsm));
 	return 0;
 }
 
@@ -485,63 +984,48 @@ static int lov_attr_get_empty(const struct lu_env *env, struct cl_object *obj,
         return 0;
 }
 
-static int lov_attr_get_raid0(const struct lu_env *env, struct cl_object *obj,
-                              struct cl_attr *attr)
+static int lov_attr_get_composite(const struct lu_env *env,
+				  struct cl_object *obj,
+				  struct cl_attr *attr)
 {
 	struct lov_object	*lov = cl2lov(obj);
-	struct lov_layout_raid0 *r0 = lov_r0(lov);
-	struct cl_attr		*lov_attr = &r0->lo_attr;
+	struct lov_layout_entry *entry;
 	int			 result = 0;
 
-        ENTRY;
-
-	/* this is called w/o holding type guard mutex, so it must be inside
-	 * an on going IO otherwise lsm may be replaced.
-	 * LU-2117: it turns out there exists one exception. For mmaped files,
-	 * the lock of those files may be requested in the other file's IO
-	 * context, and this function is called in ccc_lock_state(), it will
-	 * hit this assertion.
-	 * Anyway, it's still okay to call attr_get w/o type guard as layout
-	 * can't go if locks exist. */
-	/* LASSERT(atomic_read(&lsm->lsm_refc) > 1); */
-
-	if (!r0->lo_attr_valid) {
-		struct lov_stripe_md    *lsm = lov->lo_lsm;
-		struct ost_lvb          *lvb = &lov_env_info(env)->lti_lvb;
-		__u64                    kms = 0;
-
-		memset(lvb, 0, sizeof(*lvb));
-		/* XXX: timestamps can be negative by sanity:test_39m,
-		 * how can it be? */
-		lvb->lvb_atime = LLONG_MIN;
-		lvb->lvb_ctime = LLONG_MIN;
-		lvb->lvb_mtime = LLONG_MIN;
-
-		/*
-		 * XXX that should be replaced with a loop over sub-objects,
-		 * doing cl_object_attr_get() on them. But for now, let's
-		 * reuse old lov code.
-		 */
+	ENTRY;
 
-		/*
-		 * XXX take lsm spin-lock to keep lov_merge_lvb_kms()
-		 * happy. It's not needed, because new code uses
-		 * ->coh_attr_guard spin-lock to protect consistency of
-		 * sub-object attributes.
-		 */
-		lov_stripe_lock(lsm);
-		result = lov_merge_lvb_kms(lsm, lvb, &kms);
-		lov_stripe_unlock(lsm);
-		if (result == 0) {
-			cl_lvb2attr(lov_attr, lvb);
-			lov_attr->cat_kms = kms;
-			r0->lo_attr_valid = 1;
-		}
-	}
-	if (result == 0) { /* merge results */
-		attr->cat_blocks = lov_attr->cat_blocks;
-		attr->cat_size = lov_attr->cat_size;
-		attr->cat_kms = lov_attr->cat_kms;
+	attr->cat_size = 0;
+	attr->cat_blocks = 0;
+	lov_foreach_layout_entry(lov, entry) {
+		struct cl_attr *lov_attr = NULL;
+		int index = lov_layout_entry_index(lov, entry);
+
+		if (!entry->lle_valid)
+			continue;
+
+		/* PFL: This component has not been init-ed. */
+		if (!lsm_entry_inited(lov->lo_lsm, index))
+			continue;
+
+		result = entry->lle_comp_ops->lco_getattr(env, lov, index,
+							  entry, &lov_attr);
+		if (result < 0)
+			RETURN(result);
+
+		if (lov_attr == NULL)
+			continue;
+
+		CDEBUG(D_INODE, "COMP ID #%i: s=%llu m=%llu a=%llu c=%llu "
+		       "b=%llu\n", index - 1, lov_attr->cat_size,
+		       lov_attr->cat_mtime, lov_attr->cat_atime,
+		       lov_attr->cat_ctime, lov_attr->cat_blocks);
+
+		/* merge results */
+		attr->cat_blocks += lov_attr->cat_blocks;
+		if (attr->cat_size < lov_attr->cat_size)
+			attr->cat_size = lov_attr->cat_size;
+		if (attr->cat_kms < lov_attr->cat_kms)
+			attr->cat_kms = lov_attr->cat_kms;
 		if (attr->cat_atime < lov_attr->cat_atime)
 			attr->cat_atime = lov_attr->cat_atime;
 		if (attr->cat_ctime < lov_attr->cat_ctime)
@@ -549,78 +1033,117 @@ static int lov_attr_get_raid0(const struct lu_env *env, struct cl_object *obj,
 		if (attr->cat_mtime < lov_attr->cat_mtime)
 			attr->cat_mtime = lov_attr->cat_mtime;
 	}
-	RETURN(result);
+
+	RETURN(0);
+}
+
+static int lov_flush_composite(const struct lu_env *env,
+			       struct cl_object *obj,
+			       struct ldlm_lock *lock)
+{
+	struct lov_object *lov = cl2lov(obj);
+	struct lovsub_object *lovsub;
+
+	ENTRY;
+
+	if (!lsme_is_dom(lov->lo_lsm->lsm_entries[0]))
+		RETURN(-EINVAL);
+
+	lovsub = lov->u.composite.lo_entries[0].lle_dom.lo_dom;
+	RETURN(cl_object_flush(env, lovsub2cl(lovsub), lock));
 }
 
 const static struct lov_layout_operations lov_dispatch[] = {
-        [LLT_EMPTY] = {
-                .llo_init      = lov_init_empty,
-                .llo_delete    = lov_delete_empty,
-                .llo_fini      = lov_fini_empty,
-                .llo_install   = lov_install_empty,
-                .llo_print     = lov_print_empty,
-                .llo_page_init = lov_page_init_empty,
-                .llo_lock_init = lov_lock_init_empty,
-                .llo_io_init   = lov_io_init_empty,
-                .llo_getattr   = lov_attr_get_empty
-        },
-        [LLT_RAID0] = {
-                .llo_init      = lov_init_raid0,
-                .llo_delete    = lov_delete_raid0,
-                .llo_fini      = lov_fini_raid0,
-                .llo_install   = lov_install_raid0,
-                .llo_print     = lov_print_raid0,
-                .llo_page_init = lov_page_init_raid0,
-                .llo_lock_init = lov_lock_init_raid0,
-                .llo_io_init   = lov_io_init_raid0,
-                .llo_getattr   = lov_attr_get_raid0
+	[LLT_EMPTY] = {
+		.llo_init      = lov_init_empty,
+		.llo_delete    = lov_delete_empty,
+		.llo_fini      = lov_fini_empty,
+		.llo_print     = lov_print_empty,
+		.llo_page_init = lov_page_init_empty,
+		.llo_lock_init = lov_lock_init_empty,
+		.llo_io_init   = lov_io_init_empty,
+		.llo_getattr   = lov_attr_get_empty,
+	},
+	[LLT_RELEASED] = {
+		.llo_init      = lov_init_released,
+		.llo_delete    = lov_delete_empty,
+		.llo_fini      = lov_fini_released,
+		.llo_print     = lov_print_released,
+		.llo_page_init = lov_page_init_empty,
+		.llo_lock_init = lov_lock_init_empty,
+		.llo_io_init   = lov_io_init_released,
+		.llo_getattr   = lov_attr_get_empty,
+	},
+	[LLT_COMP] = {
+		.llo_init      = lov_init_composite,
+		.llo_delete    = lov_delete_composite,
+		.llo_fini      = lov_fini_composite,
+		.llo_print     = lov_print_composite,
+		.llo_page_init = lov_page_init_composite,
+		.llo_lock_init = lov_lock_init_composite,
+		.llo_io_init   = lov_io_init_composite,
+		.llo_getattr   = lov_attr_get_composite,
+		.llo_flush     = lov_flush_composite,
+	},
+	[LLT_FOREIGN] = {
+		.llo_init      = lov_init_foreign,
+		.llo_delete    = lov_delete_empty,
+		.llo_fini      = lov_fini_released,
+		.llo_print     = lov_print_foreign,
+		.llo_page_init = lov_page_init_foreign,
+		.llo_lock_init = lov_lock_init_empty,
+		.llo_io_init   = lov_io_init_empty,
+		.llo_getattr   = lov_attr_get_empty,
 	},
-        [LLT_RELEASED] = {
-                .llo_init      = lov_init_released,
-                .llo_delete    = lov_delete_empty,
-                .llo_fini      = lov_fini_released,
-                .llo_install   = lov_install_empty,
-                .llo_print     = lov_print_released,
-                .llo_page_init = lov_page_init_empty,
-                .llo_lock_init = lov_lock_init_empty,
-                .llo_io_init   = lov_io_init_released,
-                .llo_getattr   = lov_attr_get_empty
-        }
 };
 
 /**
  * Performs a double-dispatch based on the layout type of an object.
  */
-#define LOV_2DISPATCH_NOLOCK(obj, op, ...)                              \
-({                                                                      \
-        struct lov_object                      *__obj = (obj);          \
-        enum lov_layout_type                    __llt;                  \
-                                                                        \
-        __llt = __obj->lo_type;                                         \
-        LASSERT(0 <= __llt && __llt < ARRAY_SIZE(lov_dispatch));        \
-        lov_dispatch[__llt].op(__VA_ARGS__);                            \
+#define LOV_2DISPATCH_NOLOCK(obj, op, ...)		\
+({							\
+	struct lov_object *__obj = (obj);		\
+	enum lov_layout_type __llt;			\
+							\
+	__llt = __obj->lo_type;				\
+	LASSERT(__llt < ARRAY_SIZE(lov_dispatch));	\
+	lov_dispatch[__llt].op(__VA_ARGS__);		\
 })
 
 /**
  * Return lov_layout_type associated with a given lsm
  */
-enum lov_layout_type lov_type(struct lov_stripe_md *lsm)
+static enum lov_layout_type lov_type(struct lov_stripe_md *lsm)
 {
 	if (lsm == NULL)
 		return LLT_EMPTY;
-	if (lsm_is_released(lsm))
+
+	if (lsm->lsm_is_released)
 		return LLT_RELEASED;
-	return LLT_RAID0;
+
+	if (lsm->lsm_magic == LOV_MAGIC_V1 ||
+	    lsm->lsm_magic == LOV_MAGIC_V3 ||
+	    lsm->lsm_magic == LOV_MAGIC_COMP_V1)
+		return LLT_COMP;
+
+	if (lsm->lsm_magic == LOV_MAGIC_FOREIGN)
+		return LLT_FOREIGN;
+
+	return LLT_EMPTY;
 }
 
 static inline void lov_conf_freeze(struct lov_object *lov)
 {
+	CDEBUG(D_INODE, "To take share lov(%p) owner %p/%p\n",
+		lov, lov->lo_owner, current);
 	if (lov->lo_owner != current)
 		down_read(&lov->lo_type_guard);
 }
 
 static inline void lov_conf_thaw(struct lov_object *lov)
 {
+	CDEBUG(D_INODE, "To release share lov(%p) owner %p/%p\n",
+		lov, lov->lo_owner, current);
 	if (lov->lo_owner != current)
 		up_read(&lov->lo_type_guard);
 }
@@ -652,7 +1175,7 @@ do {                                                                    \
                                                                         \
 	lov_conf_freeze(__obj);						\
         __llt = __obj->lo_type;                                         \
-        LASSERT(0 <= __llt && __llt < ARRAY_SIZE(lov_dispatch));        \
+	LASSERT(__llt < ARRAY_SIZE(lov_dispatch));			\
         lov_dispatch[__llt].op(__VA_ARGS__);                            \
 	lov_conf_thaw(__obj);						\
 } while (0)
@@ -663,10 +1186,14 @@ static void lov_conf_lock(struct lov_object *lov)
 	down_write(&lov->lo_type_guard);
 	LASSERT(lov->lo_owner == NULL);
 	lov->lo_owner = current;
+	CDEBUG(D_INODE, "Took exclusive lov(%p) owner %p\n",
+		lov, lov->lo_owner);
 }
 
 static void lov_conf_unlock(struct lov_object *lov)
 {
+	CDEBUG(D_INODE, "To release exclusive lov(%p) owner %p\n",
+		lov, lov->lo_owner);
 	lov->lo_owner = NULL;
 	up_write(&lov->lo_type_guard);
 }
@@ -688,32 +1215,26 @@ static int lov_layout_wait(const struct lu_env *env, struct lov_object *lov)
 }
 
 static int lov_layout_change(const struct lu_env *unused,
-			     struct lov_object *lov,
+			     struct lov_object *lov, struct lov_stripe_md *lsm,
 			     const struct cl_object_conf *conf)
 {
-	int result;
-	enum lov_layout_type llt = LLT_EMPTY;
+	enum lov_layout_type llt = lov_type(lsm);
 	union lov_layout_state *state = &lov->u;
 	const struct lov_layout_operations *old_ops;
 	const struct lov_layout_operations *new_ops;
-
-	void *cookie;
+	struct lov_device *lov_dev = lov_object_dev(lov);
 	struct lu_env *env;
-	int refcheck;
+	__u16 refcheck;
+	int rc;
 	ENTRY;
 
-	LASSERT(0 <= lov->lo_type && lov->lo_type < ARRAY_SIZE(lov_dispatch));
-
-	if (conf->u.coc_md != NULL)
-		llt = lov_type(conf->u.coc_md->lsm);
-	LASSERT(0 <= llt && llt < ARRAY_SIZE(lov_dispatch));
+	LASSERT(lov->lo_type < ARRAY_SIZE(lov_dispatch));
 
-	cookie = cl_env_reenter();
 	env = cl_env_get(&refcheck);
-	if (IS_ERR(env)) {
-		cl_env_reexit(cookie);
+	if (IS_ERR(env))
 		RETURN(PTR_ERR(env));
-	}
+
+	LASSERT(llt < ARRAY_SIZE(lov_dispatch));
 
 	CDEBUG(D_INODE, DFID" from %s to %s\n",
 	       PFID(lu_object_fid(lov2lu(lov))),
@@ -722,38 +1243,42 @@ static int lov_layout_change(const struct lu_env *unused,
 	old_ops = &lov_dispatch[lov->lo_type];
 	new_ops = &lov_dispatch[llt];
 
-	result = cl_object_prune(env, &lov->lo_cl);
-	if (result != 0)
-		GOTO(out, result);
+	rc = cl_object_prune(env, &lov->lo_cl);
+	if (rc != 0)
+		GOTO(out, rc);
 
-	result = old_ops->llo_delete(env, lov, &lov->u);
-	if (result == 0) {
-		old_ops->llo_fini(env, lov, &lov->u);
+	rc = old_ops->llo_delete(env, lov, &lov->u);
+	if (rc != 0)
+		GOTO(out, rc);
 
-		LASSERT(atomic_read(&lov->lo_active_ios) == 0);
+	old_ops->llo_fini(env, lov, &lov->u);
 
-		lov->lo_type = LLT_EMPTY;
-		/* page bufsize fixup */
-		cl_object_header(&lov->lo_cl)->coh_page_bufsize -=
-			lov_page_slice_fixup(lov, NULL);
+	LASSERT(atomic_read(&lov->lo_active_ios) == 0);
 
-		result = new_ops->llo_init(env,
-					lu2lov_dev(lov->lo_cl.co_lu.lo_dev),
-					lov, conf, state);
-		if (result == 0) {
-			new_ops->llo_install(env, lov, state);
-			lov->lo_type = llt;
-		} else {
-			new_ops->llo_delete(env, lov, state);
-			new_ops->llo_fini(env, lov, state);
-			/* this file becomes an EMPTY file. */
-		}
+	CDEBUG(D_INODE, DFID "Apply new layout lov %p, type %d\n",
+	       PFID(lu_object_fid(lov2lu(lov))), lov, llt);
+
+	/* page bufsize fixup */
+	cl_object_header(&lov->lo_cl)->coh_page_bufsize -=
+		lov_page_slice_fixup(lov, NULL);
+
+	lov->lo_type = llt;
+	rc = new_ops->llo_init(env, lov_dev, lov, lsm, conf, state);
+	if (rc != 0) {
+		struct obd_device *obd = lov2obd(lov_dev->ld_lov);
+
+		CERROR("%s: cannot apply new layout on "DFID" : rc = %d\n",
+		       obd->obd_name, PFID(lu_object_fid(lov2lu(lov))), rc);
+		new_ops->llo_delete(env, lov, state);
+		new_ops->llo_fini(env, lov, state);
+		/* this file becomes an EMPTY file. */
+		lov->lo_type = LLT_EMPTY;
+		GOTO(out, rc);
 	}
 
 out:
 	cl_env_put(env, &refcheck);
-	cl_env_reexit(cookie);
-	RETURN(result);
+	RETURN(rc);
 }
 
 /*****************************************************************************
@@ -762,29 +1287,44 @@ out:
  *
  */
 int lov_object_init(const struct lu_env *env, struct lu_object *obj,
-                    const struct lu_object_conf *conf)
+		    const struct lu_object_conf *conf)
 {
-        struct lov_device            *dev   = lu2lov_dev(obj->lo_dev);
-        struct lov_object            *lov   = lu2lov(obj);
-        const struct cl_object_conf  *cconf = lu2cl_conf(conf);
-        union  lov_layout_state      *set   = &lov->u;
-        const struct lov_layout_operations *ops;
-        int result;
+	struct lov_object            *lov   = lu2lov(obj);
+	struct lov_device            *dev   = lov_object_dev(lov);
+	const struct cl_object_conf  *cconf = lu2cl_conf(conf);
+	union lov_layout_state	     *set   = &lov->u;
+	const struct lov_layout_operations *ops;
+	struct lov_stripe_md *lsm = NULL;
+	int rc;
+	ENTRY;
 
-        ENTRY;
 	init_rwsem(&lov->lo_type_guard);
 	atomic_set(&lov->lo_active_ios, 0);
 	init_waitqueue_head(&lov->lo_waitq);
-
 	cl_object_page_init(lu2cl(obj), sizeof(struct lov_page));
 
-        /* no locking is necessary, as object is being created */
-	lov->lo_type = lov_type(cconf->u.coc_md->lsm);
-        ops = &lov_dispatch[lov->lo_type];
-        result = ops->llo_init(env, dev, lov, cconf, set);
-        if (result == 0)
-                ops->llo_install(env, lov, set);
-        RETURN(result);
+	lov->lo_type = LLT_EMPTY;
+	if (cconf->u.coc_layout.lb_buf != NULL) {
+		lsm = lov_unpackmd(dev->ld_lov,
+				   cconf->u.coc_layout.lb_buf,
+				   cconf->u.coc_layout.lb_len);
+		if (IS_ERR(lsm))
+			RETURN(PTR_ERR(lsm));
+
+		dump_lsm(D_INODE, lsm);
+	}
+
+	/* no locking is necessary, as object is being created */
+	lov->lo_type = lov_type(lsm);
+	ops = &lov_dispatch[lov->lo_type];
+	rc = ops->llo_init(env, dev, lov, lsm, cconf, set);
+	if (rc != 0)
+		GOTO(out_lsm, rc);
+
+out_lsm:
+	lov_lsm_put(lsm);
+
+	RETURN(rc);
 }
 
 static int lov_conf_set(const struct lu_env *env, struct cl_object *obj,
@@ -795,6 +1335,16 @@ static int lov_conf_set(const struct lu_env *env, struct cl_object *obj,
 	int			 result = 0;
 	ENTRY;
 
+	if (conf->coc_opc == OBJECT_CONF_SET &&
+	    conf->u.coc_layout.lb_buf != NULL) {
+		lsm = lov_unpackmd(lov_object_dev(lov)->ld_lov,
+				   conf->u.coc_layout.lb_buf,
+				   conf->u.coc_layout.lb_len);
+		if (IS_ERR(lsm))
+			RETURN(PTR_ERR(lsm));
+		dump_lsm(D_INODE, lsm);
+	}
+
 	lov_conf_lock(lov);
 	if (conf->coc_opc == OBJECT_CONF_INVALIDATE) {
 		lov->lo_layout_invalid = true;
@@ -813,12 +1363,11 @@ static int lov_conf_set(const struct lu_env *env, struct cl_object *obj,
 
 	LASSERT(conf->coc_opc == OBJECT_CONF_SET);
 
-	if (conf->u.coc_md != NULL)
-		lsm = conf->u.coc_md->lsm;
 	if ((lsm == NULL && lov->lo_lsm == NULL) ||
 	    ((lsm != NULL && lov->lo_lsm != NULL) &&
 	     (lov->lo_lsm->lsm_layout_gen == lsm->lsm_layout_gen) &&
-	     (lov->lo_lsm->lsm_pattern == lsm->lsm_pattern))) {
+	     (lov->lo_lsm->lsm_entries[0]->lsme_pattern ==
+	      lsm->lsm_entries[0]->lsme_pattern))) {
 		/* same version of layout */
 		lov->lo_layout_invalid = false;
 		GOTO(out, result = 0);
@@ -830,12 +1379,13 @@ static int lov_conf_set(const struct lu_env *env, struct cl_object *obj,
 		GOTO(out, result = -EBUSY);
 	}
 
-	result = lov_layout_change(env, lov, conf);
+	result = lov_layout_change(env, lov, lsm, conf);
 	lov->lo_layout_invalid = result != 0;
 	EXIT;
 
 out:
 	lov_conf_unlock(lov);
+	lov_lsm_put(lsm);
 	CDEBUG(D_INODE, DFID" lo_layout_invalid=%d\n",
 	       PFID(lu_object_fid(lov2lu(lov))), lov->lo_layout_invalid);
 	RETURN(result);
@@ -881,9 +1431,19 @@ int lov_page_init(const struct lu_env *env, struct cl_object *obj,
 int lov_io_init(const struct lu_env *env, struct cl_object *obj,
 		struct cl_io *io)
 {
-	CL_IO_SLICE_CLEAN(lov_env_io(env), lis_cl);
+	CL_IO_SLICE_CLEAN(lov_env_io(env), lis_preserved);
+
+	CDEBUG(D_INODE, DFID "io %p type %d ignore/verify layout %d/%d\n",
+	       PFID(lu_object_fid(&obj->co_lu)), io, io->ci_type,
+	       io->ci_ignore_layout, io->ci_verify_layout);
+
+	/* IO type CIT_MISC with ci_ignore_layout set are usually invoked from
+	 * the OSC layer. It shouldn't take lov layout conf lock in that case,
+	 * because as long as the OSC object exists, the layout can't be
+	 * reconfigured. */
 	return LOV_2DISPATCH_MAYLOCK(cl2lov(obj), llo_io_init,
-				     !io->ci_ignore_layout, env, obj, io);
+			!(io->ci_ignore_layout && io->ci_type == CIT_MISC),
+			env, obj, io);
 }
 
 /**
@@ -899,13 +1459,13 @@ static int lov_attr_get(const struct lu_env *env, struct cl_object *obj,
         return LOV_2DISPATCH_NOLOCK(cl2lov(obj), llo_getattr, env, obj, attr);
 }
 
-static int lov_attr_set(const struct lu_env *env, struct cl_object *obj,
-                        const struct cl_attr *attr, unsigned valid)
+static int lov_attr_update(const struct lu_env *env, struct cl_object *obj,
+			   const struct cl_attr *attr, unsigned valid)
 {
-        /*
-         * No dispatch is required here, as no layout implements this.
-         */
-        return 0;
+	/*
+	 * No dispatch is required here, as no layout implements this.
+	 */
+	return 0;
 }
 
 int lov_lock_init(const struct lu_env *env, struct cl_object *obj,
@@ -916,13 +1476,645 @@ int lov_lock_init(const struct lu_env *env, struct cl_object *obj,
 				    io);
 }
 
+/**
+ * We calculate on which OST the mapping will end. If the length of mapping
+ * is greater than (stripe_size * stripe_count) then the last_stripe will
+ * will be one just before start_stripe. Else we check if the mapping
+ * intersects each OST and find last_stripe.
+ * This function returns the last_stripe and also sets the stripe_count
+ * over which the mapping is spread
+ *
+ * \param lsm [in]		striping information for the file
+ * \param index [in]		stripe component index
+ * \param ext [in]		logical extent of mapping
+ * \param start_stripe [in]	starting stripe of the mapping
+ * \param stripe_count [out]	the number of stripes across which to map is
+ *				returned
+ *
+ * \retval last_stripe		return the last stripe of the mapping
+ */
+static int fiemap_calc_last_stripe(struct lov_stripe_md *lsm, int index,
+				   struct lu_extent *ext,
+				   int start_stripe, int *stripe_count)
+{
+	struct lov_stripe_md_entry *lsme = lsm->lsm_entries[index];
+	int last_stripe;
+	u64 obd_start;
+	u64 obd_end;
+	int i, j;
+
+	if (ext->e_end - ext->e_start >
+	    lsme->lsme_stripe_size * lsme->lsme_stripe_count) {
+		last_stripe = (start_stripe < 1 ? lsme->lsme_stripe_count - 1 :
+						  start_stripe - 1);
+		*stripe_count = lsme->lsme_stripe_count;
+	} else {
+		for (j = 0, i = start_stripe; j < lsme->lsme_stripe_count;
+		     i = (i + 1) % lsme->lsme_stripe_count, j++) {
+			if ((lov_stripe_intersects(lsm, index,  i, ext,
+						   &obd_start, &obd_end)) == 0)
+				break;
+		}
+		*stripe_count = j;
+		last_stripe = (start_stripe + j - 1) % lsme->lsme_stripe_count;
+	}
+
+	return last_stripe;
+}
+
+/**
+ * Set fe_device and copy extents from local buffer into main return buffer.
+ *
+ * \param fiemap [out]		fiemap to hold all extents
+ * \param lcl_fm_ext [in]	array of fiemap extents get from OSC layer
+ * \param ost_index [in]	OST index to be written into the fm_device
+ *				field for each extent
+ * \param ext_count [in]	number of extents to be copied
+ * \param current_extent [in]	where to start copying in the extent array
+ */
+static void fiemap_prepare_and_copy_exts(struct fiemap *fiemap,
+					 struct fiemap_extent *lcl_fm_ext,
+					 int ost_index, unsigned int ext_count,
+					 int current_extent)
+{
+	char		*to;
+	unsigned int	ext;
+
+	for (ext = 0; ext < ext_count; ext++) {
+		lcl_fm_ext[ext].fe_device = ost_index;
+		lcl_fm_ext[ext].fe_flags |= FIEMAP_EXTENT_NET;
+	}
+
+	/* Copy fm_extent's from fm_local to return buffer */
+	to = (char *)fiemap + fiemap_count_to_size(current_extent);
+	memcpy(to, lcl_fm_ext, ext_count * sizeof(struct fiemap_extent));
+}
+
+#define FIEMAP_BUFFER_SIZE 4096
+
+/**
+ * Non-zero fe_logical indicates that this is a continuation FIEMAP
+ * call. The local end offset and the device are sent in the first
+ * fm_extent. This function calculates the stripe number from the index.
+ * This function returns a stripe_no on which mapping is to be restarted.
+ *
+ * This function returns fm_end_offset which is the in-OST offset at which
+ * mapping should be restarted. If fm_end_offset=0 is returned then caller
+ * will re-calculate proper offset in next stripe.
+ * Note that the first extent is passed to lov_get_info via the value field.
+ *
+ * \param fiemap [in]		fiemap request header
+ * \param lsm [in]		striping information for the file
+ * \param index [in]		stripe component index
+ * \param ext [in]		logical extent of mapping
+ * \param start_stripe [out]	starting stripe will be returned in this
+ */
+static u64 fiemap_calc_fm_end_offset(struct fiemap *fiemap,
+				     struct lov_stripe_md *lsm,
+				     int index, struct lu_extent *ext,
+				     int *start_stripe)
+{
+	struct lov_stripe_md_entry *lsme = lsm->lsm_entries[index];
+	u64 local_end = fiemap->fm_extents[0].fe_logical;
+	u64 lun_start;
+	u64 lun_end;
+	u64 fm_end_offset;
+	int stripe_no = -1;
+	int i;
+
+	if (fiemap->fm_extent_count == 0 ||
+	    fiemap->fm_extents[0].fe_logical == 0)
+		return 0;
+
+	/* Find out stripe_no from ost_index saved in the fe_device */
+	for (i = 0; i < lsme->lsme_stripe_count; i++) {
+		struct lov_oinfo *oinfo = lsme->lsme_oinfo[i];
+
+		if (lov_oinfo_is_dummy(oinfo))
+			continue;
+
+		if (oinfo->loi_ost_idx == fiemap->fm_extents[0].fe_device) {
+			stripe_no = i;
+			break;
+		}
+	}
+
+	if (stripe_no == -1)
+		return -EINVAL;
+
+	/* If we have finished mapping on previous device, shift logical
+	 * offset to start of next device */
+	if (lov_stripe_intersects(lsm, index, stripe_no, ext,
+				   &lun_start, &lun_end) != 0 &&
+	    local_end < lun_end) {
+		fm_end_offset = local_end;
+		*start_stripe = stripe_no;
+	} else {
+		/* This is a special value to indicate that caller should
+		 * calculate offset in next stripe. */
+		fm_end_offset = 0;
+		*start_stripe = (stripe_no + 1) % lsme->lsme_stripe_count;
+	}
+
+	return fm_end_offset;
+}
+
+struct fiemap_state {
+	struct fiemap		*fs_fm;
+	struct lu_extent	fs_ext;
+	u64			fs_length;
+	u64			fs_end_offset;
+	int			fs_cur_extent;
+	int			fs_cnt_need;
+	int			fs_start_stripe;
+	int			fs_last_stripe;
+	bool			fs_device_done;
+	bool			fs_finish_stripe;
+	bool			fs_enough;
+};
+
+static struct cl_object *lov_find_subobj(const struct lu_env *env,
+					 struct lov_object *lov,
+					 struct lov_stripe_md *lsm,
+					 int index)
+{
+	struct lov_device	*dev = lu2lov_dev(lov2lu(lov)->lo_dev);
+	struct lov_thread_info  *lti = lov_env_info(env);
+	struct lu_fid		*ofid = &lti->lti_fid;
+	struct lov_oinfo	*oinfo;
+	struct cl_device	*subdev;
+	int			entry = lov_comp_entry(index);
+	int			stripe = lov_comp_stripe(index);
+	int			ost_idx;
+	int			rc;
+	struct cl_object	*result;
+
+	if (lov->lo_type != LLT_COMP)
+		GOTO(out, result = NULL);
+
+	if (entry >= lsm->lsm_entry_count ||
+	    stripe >= lsm->lsm_entries[entry]->lsme_stripe_count)
+		GOTO(out, result = NULL);
+
+	oinfo = lsm->lsm_entries[entry]->lsme_oinfo[stripe];
+	ost_idx = oinfo->loi_ost_idx;
+	rc = ostid_to_fid(ofid, &oinfo->loi_oi, ost_idx);
+	if (rc != 0)
+		GOTO(out, result = NULL);
+
+	subdev = lovsub2cl_dev(dev->ld_target[ost_idx]);
+	result = lov_sub_find(env, subdev, ofid, NULL);
+out:
+	if (result == NULL)
+		result = ERR_PTR(-EINVAL);
+	return result;
+}
+
+int fiemap_for_stripe(const struct lu_env *env, struct cl_object *obj,
+		      struct lov_stripe_md *lsm, struct fiemap *fiemap,
+		      size_t *buflen, struct ll_fiemap_info_key *fmkey,
+		      int index, int stripeno, struct fiemap_state *fs)
+{
+	struct lov_stripe_md_entry *lsme = lsm->lsm_entries[index];
+	struct cl_object *subobj;
+	struct lov_obd *lov = lu2lov_dev(obj->co_lu.lo_dev)->ld_lov;
+	struct fiemap_extent *fm_ext = &fs->fs_fm->fm_extents[0];
+	u64 req_fm_len; /* Stores length of required mapping */
+	u64 len_mapped_single_call;
+	u64 lun_start;
+	u64 lun_end;
+	u64 obd_object_end;
+	unsigned int ext_count;
+	/* EOF for object */
+	bool ost_eof = false;
+	/* done with required mapping for this OST? */
+	bool ost_done = false;
+	int ost_index;
+	int rc = 0;
+
+	fs->fs_device_done = false;
+	/* Find out range of mapping on this stripe */
+	if ((lov_stripe_intersects(lsm, index, stripeno, &fs->fs_ext,
+				   &lun_start, &obd_object_end)) == 0)
+		return 0;
+
+	if (lov_oinfo_is_dummy(lsme->lsme_oinfo[stripeno]))
+		return -EIO;
+
+	/* If this is a continuation FIEMAP call and we are on
+	 * starting stripe then lun_start needs to be set to
+	 * end_offset */
+	if (fs->fs_end_offset != 0 && stripeno == fs->fs_start_stripe)
+		lun_start = fs->fs_end_offset;
+	lun_end = lov_size_to_stripe(lsm, index, fs->fs_ext.e_end, stripeno);
+	if (lun_start == lun_end)
+		return 0;
+
+	req_fm_len = obd_object_end - lun_start + 1;
+	fs->fs_fm->fm_length = 0;
+	len_mapped_single_call = 0;
+
+	/* find lobsub object */
+	subobj = lov_find_subobj(env, cl2lov(obj), lsm,
+				 lov_comp_index(index, stripeno));
+	if (IS_ERR(subobj))
+		return PTR_ERR(subobj);
+	/* If the output buffer is very large and the objects have many
+	 * extents we may need to loop on a single OST repeatedly */
+	do {
+		if (fiemap->fm_extent_count > 0) {
+			/* Don't get too many extents. */
+			if (fs->fs_cur_extent + fs->fs_cnt_need >
+			    fiemap->fm_extent_count)
+				fs->fs_cnt_need = fiemap->fm_extent_count -
+						  fs->fs_cur_extent;
+		}
+
+		lun_start += len_mapped_single_call;
+		fs->fs_fm->fm_length = req_fm_len - len_mapped_single_call;
+		req_fm_len = fs->fs_fm->fm_length;
+		/**
+		 * If we've collected enough extent map, we'd request 1 more,
+		 * to see whether we coincidentally finished all available
+		 * extent map, so that FIEMAP_EXTENT_LAST would be set.
+		 */
+		fs->fs_fm->fm_extent_count = fs->fs_enough ?
+					     1 : fs->fs_cnt_need;
+		fs->fs_fm->fm_mapped_extents = 0;
+		fs->fs_fm->fm_flags = fiemap->fm_flags;
+
+		ost_index = lsme->lsme_oinfo[stripeno]->loi_ost_idx;
+
+		if (ost_index < 0 || ost_index >= lov->desc.ld_tgt_count)
+			GOTO(obj_put, rc = -EINVAL);
+		/* If OST is inactive, return extent with UNKNOWN flag. */
+		if (!lov->lov_tgts[ost_index]->ltd_active) {
+			fs->fs_fm->fm_flags |= FIEMAP_EXTENT_LAST;
+			fs->fs_fm->fm_mapped_extents = 1;
+
+			fm_ext[0].fe_logical = lun_start;
+			fm_ext[0].fe_length = obd_object_end - lun_start + 1;
+			fm_ext[0].fe_flags |= FIEMAP_EXTENT_UNKNOWN;
+
+			goto inactive_tgt;
+		}
+
+		fs->fs_fm->fm_start = lun_start;
+		fs->fs_fm->fm_flags &= ~FIEMAP_FLAG_DEVICE_ORDER;
+		memcpy(&fmkey->lfik_fiemap, fs->fs_fm, sizeof(*fs->fs_fm));
+		*buflen = fiemap_count_to_size(fs->fs_fm->fm_extent_count);
+
+		rc = cl_object_fiemap(env, subobj, fmkey, fs->fs_fm, buflen);
+		if (rc != 0)
+			GOTO(obj_put, rc);
+inactive_tgt:
+		ext_count = fs->fs_fm->fm_mapped_extents;
+		if (ext_count == 0) {
+			ost_done = true;
+			fs->fs_device_done = true;
+			/* If last stripe has hold at the end,
+			 * we need to return */
+			if (stripeno == fs->fs_last_stripe) {
+				fiemap->fm_mapped_extents = 0;
+				fs->fs_finish_stripe = true;
+				GOTO(obj_put, rc);
+			}
+			break;
+		} else if (fs->fs_enough) {
+			/*
+			 * We've collected enough extents and there are
+			 * more extents after it.
+			 */
+			GOTO(obj_put, rc);
+		}
+
+		/* If we just need num of extents, got to next device */
+		if (fiemap->fm_extent_count == 0) {
+			fs->fs_cur_extent += ext_count;
+			break;
+		}
+
+		/* prepare to copy retrived map extents */
+		len_mapped_single_call = fm_ext[ext_count - 1].fe_logical +
+					 fm_ext[ext_count - 1].fe_length -
+					 lun_start;
+
+		/* Have we finished mapping on this device? */
+		if (req_fm_len <= len_mapped_single_call) {
+			ost_done = true;
+			fs->fs_device_done = true;
+		}
+
+		/* Clear the EXTENT_LAST flag which can be present on
+		 * the last extent */
+		if (fm_ext[ext_count - 1].fe_flags & FIEMAP_EXTENT_LAST)
+			fm_ext[ext_count - 1].fe_flags &= ~FIEMAP_EXTENT_LAST;
+		if (lov_stripe_size(lsm, index,
+				    fm_ext[ext_count - 1].fe_logical +
+				    fm_ext[ext_count - 1].fe_length,
+				    stripeno) >= fmkey->lfik_oa.o_size) {
+			ost_eof = true;
+			fs->fs_device_done = true;
+		}
+
+		fiemap_prepare_and_copy_exts(fiemap, fm_ext, ost_index,
+					     ext_count, fs->fs_cur_extent);
+		fs->fs_cur_extent += ext_count;
+
+		/* Ran out of available extents? */
+		if (fs->fs_cur_extent >= fiemap->fm_extent_count)
+			fs->fs_enough = true;
+	} while (!ost_done && !ost_eof);
+
+	if (stripeno == fs->fs_last_stripe)
+		fs->fs_finish_stripe = true;
+obj_put:
+	cl_object_put(env, subobj);
+
+	return rc;
+}
+
+/**
+ * Break down the FIEMAP request and send appropriate calls to individual OSTs.
+ * This also handles the restarting of FIEMAP calls in case mapping overflows
+ * the available number of extents in single call.
+ *
+ * \param env [in]		lustre environment
+ * \param obj [in]		file object
+ * \param fmkey [in]		fiemap request header and other info
+ * \param fiemap [out]		fiemap buffer holding retrived map extents
+ * \param buflen [in/out]	max buffer length of @fiemap, when iterate
+ *				each OST, it is used to limit max map needed
+ * \retval 0	success
+ * \retval < 0	error
+ */
+static int lov_object_fiemap(const struct lu_env *env, struct cl_object *obj,
+			     struct ll_fiemap_info_key *fmkey,
+			     struct fiemap *fiemap, size_t *buflen)
+{
+	struct lov_stripe_md_entry *lsme;
+	struct lov_stripe_md *lsm;
+	struct fiemap *fm_local = NULL;
+	loff_t whole_start;
+	loff_t whole_end;
+	int entry;
+	int start_entry;
+	int end_entry;
+	int cur_stripe = 0;
+	int stripe_count;
+	unsigned int buffer_size = FIEMAP_BUFFER_SIZE;
+	int rc = 0;
+	struct fiemap_state fs = { 0 };
+	ENTRY;
+
+	lsm = lov_lsm_addref(cl2lov(obj));
+	if (lsm == NULL) {
+		/* no extent: there is no object for mapping */
+		fiemap->fm_mapped_extents = 0;
+		return 0;
+	}
+
+	if (!(fiemap->fm_flags & FIEMAP_FLAG_DEVICE_ORDER)) {
+		/**
+		 * If the entry count > 1 or stripe_count > 1 and the
+		 * application does not understand DEVICE_ORDER flag,
+		 * it cannot interpret the extents correctly.
+		 */
+		if (lsm->lsm_entry_count > 1 ||
+		    (lsm->lsm_entry_count == 1 &&
+		     lsm->lsm_entries[0]->lsme_stripe_count > 1))
+			GOTO(out_lsm, rc = -ENOTSUPP);
+	}
+
+	/* No support for DOM layout yet. */
+	if (lsme_is_dom(lsm->lsm_entries[0]))
+		GOTO(out_lsm, rc = -ENOTSUPP);
+
+	if (lsm->lsm_is_released) {
+		if (fiemap->fm_start < fmkey->lfik_oa.o_size) {
+			/**
+			 * released file, return a minimal FIEMAP if
+			 * request fits in file-size.
+			 */
+			fiemap->fm_mapped_extents = 1;
+			fiemap->fm_extents[0].fe_logical = fiemap->fm_start;
+			if (fiemap->fm_start + fiemap->fm_length <
+			    fmkey->lfik_oa.o_size)
+				fiemap->fm_extents[0].fe_length =
+					fiemap->fm_length;
+			else
+				fiemap->fm_extents[0].fe_length =
+					fmkey->lfik_oa.o_size -
+					fiemap->fm_start;
+			fiemap->fm_extents[0].fe_flags |=
+				FIEMAP_EXTENT_UNKNOWN | FIEMAP_EXTENT_LAST;
+		}
+		GOTO(out_lsm, rc = 0);
+	}
+
+	/* buffer_size is small to hold fm_extent_count of extents. */
+	if (fiemap_count_to_size(fiemap->fm_extent_count) < buffer_size)
+		buffer_size = fiemap_count_to_size(fiemap->fm_extent_count);
+
+	OBD_ALLOC_LARGE(fm_local, buffer_size);
+	if (fm_local == NULL)
+		GOTO(out_lsm, rc = -ENOMEM);
+
+	/**
+	 * Requested extent count exceeds the fiemap buffer size, shrink our
+	 * ambition.
+	 */
+	if (fiemap_count_to_size(fiemap->fm_extent_count) > *buflen)
+		fiemap->fm_extent_count = fiemap_size_to_count(*buflen);
+	if (fiemap->fm_extent_count == 0)
+		fs.fs_cnt_need = 0;
+
+	fs.fs_enough = false;
+	fs.fs_cur_extent = 0;
+	fs.fs_fm = fm_local;
+	fs.fs_cnt_need = fiemap_size_to_count(buffer_size);
+
+	whole_start = fiemap->fm_start;
+	/* whole_start is beyond the end of the file */
+	if (whole_start > fmkey->lfik_oa.o_size)
+		GOTO(out_fm_local, rc = -EINVAL);
+	whole_end = (fiemap->fm_length == OBD_OBJECT_EOF) ?
+					fmkey->lfik_oa.o_size :
+					whole_start + fiemap->fm_length - 1;
+	/**
+	 * If fiemap->fm_length != OBD_OBJECT_EOF but whole_end exceeds file
+	 * size
+	 */
+	if (whole_end > fmkey->lfik_oa.o_size)
+		whole_end = fmkey->lfik_oa.o_size;
+
+	start_entry = lov_lsm_entry(lsm, whole_start);
+	end_entry = lov_lsm_entry(lsm, whole_end);
+	if (end_entry == -1)
+		end_entry = lsm->lsm_entry_count - 1;
+
+	if (start_entry == -1 || end_entry == -1)
+		GOTO(out_fm_local, rc = -EINVAL);
+
+	/* TODO: rewrite it with lov_foreach_io_layout() */
+	for (entry = start_entry; entry <= end_entry; entry++) {
+		lsme = lsm->lsm_entries[entry];
+
+		if (!lsme_inited(lsme))
+			break;
+
+		if (entry == start_entry)
+			fs.fs_ext.e_start = whole_start;
+		else
+			fs.fs_ext.e_start = lsme->lsme_extent.e_start;
+		if (entry == end_entry)
+			fs.fs_ext.e_end = whole_end;
+		else
+			fs.fs_ext.e_end = lsme->lsme_extent.e_end - 1;
+		fs.fs_length = fs.fs_ext.e_end - fs.fs_ext.e_start + 1;
+
+		/* Calculate start stripe, last stripe and length of mapping */
+		fs.fs_start_stripe = lov_stripe_number(lsm, entry,
+						       fs.fs_ext.e_start);
+		fs.fs_last_stripe = fiemap_calc_last_stripe(lsm, entry,
+					&fs.fs_ext, fs.fs_start_stripe,
+					&stripe_count);
+		fs.fs_end_offset = fiemap_calc_fm_end_offset(fiemap, lsm, entry,
+					&fs.fs_ext, &fs.fs_start_stripe);
+		/* Check each stripe */
+		for (cur_stripe = fs.fs_start_stripe; stripe_count > 0;
+		     --stripe_count,
+		     cur_stripe = (cur_stripe + 1) % lsme->lsme_stripe_count) {
+			rc = fiemap_for_stripe(env, obj, lsm, fiemap, buflen,
+					       fmkey, entry, cur_stripe, &fs);
+			if (rc < 0)
+				GOTO(out_fm_local, rc);
+			if (fs.fs_enough)
+				GOTO(finish, rc);
+			if (fs.fs_finish_stripe)
+				break;
+		} /* for each stripe */
+	} /* for covering layout component */
+	/*
+	 * We've traversed all components, set @entry to the last component
+	 * entry, it's for the last stripe check.
+	 */
+	entry--;
+finish:
+	/* Indicate that we are returning device offsets unless file just has
+	 * single stripe */
+	if (lsm->lsm_entry_count > 1 ||
+	    (lsm->lsm_entry_count == 1 &&
+	     lsm->lsm_entries[0]->lsme_stripe_count > 1))
+		fiemap->fm_flags |= FIEMAP_FLAG_DEVICE_ORDER;
+
+	if (fiemap->fm_extent_count == 0)
+		goto skip_last_device_calc;
+
+	/* Check if we have reached the last stripe and whether mapping for that
+	 * stripe is done. */
+	if ((cur_stripe == fs.fs_last_stripe) && fs.fs_device_done)
+		fiemap->fm_extents[fs.fs_cur_extent - 1].fe_flags |=
+							     FIEMAP_EXTENT_LAST;
+skip_last_device_calc:
+	fiemap->fm_mapped_extents = fs.fs_cur_extent;
+out_fm_local:
+	OBD_FREE_LARGE(fm_local, buffer_size);
+
+out_lsm:
+	lov_lsm_put(lsm);
+	return rc;
+}
+
+static int lov_object_getstripe(const struct lu_env *env, struct cl_object *obj,
+				struct lov_user_md __user *lum, size_t size)
+{
+	struct lov_object	*lov = cl2lov(obj);
+	struct lov_stripe_md	*lsm;
+	int			rc = 0;
+	ENTRY;
+
+	lsm = lov_lsm_addref(lov);
+	if (lsm == NULL)
+		RETURN(-ENODATA);
+
+	rc = lov_getstripe(env, cl2lov(obj), lsm, lum, size);
+	lov_lsm_put(lsm);
+	RETURN(rc);
+}
+
+static int lov_object_layout_get(const struct lu_env *env,
+				 struct cl_object *obj,
+				 struct cl_layout *cl)
+{
+	struct lov_object *lov = cl2lov(obj);
+	struct lov_stripe_md *lsm = lov_lsm_addref(lov);
+	struct lu_buf *buf = &cl->cl_buf;
+	ssize_t rc;
+	ENTRY;
+
+	if (lsm == NULL) {
+		cl->cl_size = 0;
+		cl->cl_layout_gen = CL_LAYOUT_GEN_EMPTY;
+
+		RETURN(0);
+	}
+
+	cl->cl_size = lov_comp_md_size(lsm);
+	cl->cl_layout_gen = lsm->lsm_layout_gen;
+	cl->cl_dom_comp_size = 0;
+	cl->cl_is_released = lsm->lsm_is_released;
+	if (lsm_is_composite(lsm->lsm_magic)) {
+		struct lov_stripe_md_entry *lsme = lsm->lsm_entries[0];
+
+		cl->cl_is_composite = true;
+
+		if (lsme_is_dom(lsme))
+			cl->cl_dom_comp_size = lsme->lsme_extent.e_end;
+	} else {
+		cl->cl_is_composite = false;
+	}
+
+	rc = lov_lsm_pack(lsm, buf->lb_buf, buf->lb_len);
+	lov_lsm_put(lsm);
+
+	RETURN(rc < 0 ? rc : 0);
+}
+
+static loff_t lov_object_maxbytes(struct cl_object *obj)
+{
+	struct lov_object *lov = cl2lov(obj);
+	struct lov_stripe_md *lsm = lov_lsm_addref(lov);
+	loff_t maxbytes;
+
+	if (lsm == NULL)
+		return LLONG_MAX;
+
+	maxbytes = lsm->lsm_maxbytes;
+
+	lov_lsm_put(lsm);
+
+	return maxbytes;
+}
+
+static int lov_object_flush(const struct lu_env *env, struct cl_object *obj,
+			    struct ldlm_lock *lock)
+{
+	return LOV_2DISPATCH_NOLOCK(cl2lov(obj), llo_flush, env, obj, lock);
+}
+
 static const struct cl_object_operations lov_ops = {
-        .coo_page_init = lov_page_init,
-        .coo_lock_init = lov_lock_init,
-        .coo_io_init   = lov_io_init,
-        .coo_attr_get  = lov_attr_get,
-        .coo_attr_set  = lov_attr_set,
-        .coo_conf_set  = lov_conf_set
+	.coo_page_init    = lov_page_init,
+	.coo_lock_init    = lov_lock_init,
+	.coo_io_init      = lov_io_init,
+	.coo_attr_get     = lov_attr_get,
+	.coo_attr_update  = lov_attr_update,
+	.coo_conf_set     = lov_conf_set,
+	.coo_getstripe    = lov_object_getstripe,
+	.coo_layout_get   = lov_object_layout_get,
+	.coo_maxbytes     = lov_object_maxbytes,
+	.coo_fiemap       = lov_object_fiemap,
+	.coo_object_flush = lov_object_flush
 };
 
 static const struct lu_object_operations lov_lu_obj_ops = {
@@ -974,29 +2166,6 @@ struct lov_stripe_md *lov_lsm_addref(struct lov_object *lov)
 	return lsm;
 }
 
-struct lov_stripe_md *lov_lsm_get(struct cl_object *clobj)
-{
-	struct lu_object *luobj;
-	struct lov_stripe_md *lsm = NULL;
-
-	if (clobj == NULL)
-		return NULL;
-
-	luobj = lu_object_locate(&cl_object_header(clobj)->coh_lu,
-				 &lov_device_type);
-	if (luobj != NULL)
-		lsm = lov_lsm_addref(lu2lov(luobj));
-	return lsm;
-}
-EXPORT_SYMBOL(lov_lsm_get);
-
-void lov_lsm_put(struct cl_object *unused, struct lov_stripe_md *lsm)
-{
-	if (lsm != NULL)
-		lov_free_memmd(&lsm);
-}
-EXPORT_SYMBOL(lov_lsm_put);
-
 int lov_read_and_clear_async_rc(struct cl_object *clob)
 {
 	struct lu_object *luobj;
@@ -1010,25 +2179,37 @@ int lov_read_and_clear_async_rc(struct cl_object *clob)
 
 		lov_conf_freeze(lov);
 		switch (lov->lo_type) {
-		case LLT_RAID0: {
+		case LLT_COMP: {
 			struct lov_stripe_md *lsm;
 			int i;
 
 			lsm = lov->lo_lsm;
 			LASSERT(lsm != NULL);
-			for (i = 0; i < lsm->lsm_stripe_count; i++) {
-				struct lov_oinfo *loi = lsm->lsm_oinfo[i];
+			for (i = 0; i < lsm->lsm_entry_count; i++) {
+				struct lov_stripe_md_entry *lse =
+						lsm->lsm_entries[i];
+				int j;
+
+				if (!lsme_inited(lse))
+					break;
 
-				if (lov_oinfo_is_dummy(loi))
-					continue;
+				for (j = 0; j < lse->lsme_stripe_count; j++) {
+					struct lov_oinfo *loi =
+							lse->lsme_oinfo[j];
 
-				if (loi->loi_ar.ar_rc && !rc)
-					rc = loi->loi_ar.ar_rc;
-				loi->loi_ar.ar_rc = 0;
+					if (lov_oinfo_is_dummy(loi))
+						continue;
+
+					if (loi->loi_ar.ar_rc && !rc)
+						rc = loi->loi_ar.ar_rc;
+					loi->loi_ar.ar_rc = 0;
+				}
 			}
 		}
 		case LLT_RELEASED:
 		case LLT_EMPTY:
+			/* fall through */
+		case LLT_FOREIGN:
 			break;
 		default:
 			LBUG();