/*
 * GPL HEADER START
 *
 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 only,
 * as published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful, but
 * WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * General Public License version 2 for more details (a copy is included
 * in the LICENSE file that accompanied this code).
 *
 * You should have received a copy of the GNU General Public License
 * version 2 along with this program; If not, see
 * http://www.gnu.org/licenses/gpl-2.0.html
 *
 * GPL HEADER END
 */
/*
 * Copyright (c) 2012, 2017, Intel Corporation.
 */
/*
 * lustre/osd-ldiskfs/osd_scrub.c
 *
 * Top-level entry points into osd module
 *
 * The OI scrub is used for rebuilding Object Index files when restores MDT from
 * file-level backup.
 *
 * The otable based iterator scans ldiskfs inode table to feed up layer LFSCK.
 *
 * Author: Fan Yong <yong.fan@whamcloud.com>
 */

#define DEBUG_SUBSYSTEM S_LFSCK

#include <linux/kthread.h>
#include <uapi/linux/lustre/lustre_idl.h>
#include <lustre_disk.h>
#include <dt_object.h>
#include <linux/xattr.h>
#include <lustre_scrub.h>
#include <lustre_nodemap.h>

#include "osd_internal.h"
#include "osd_oi.h"
#include "osd_scrub.h"

#define OSD_OTABLE_MAX_HASH		0x00000000ffffffffULL

static inline int osd_scrub_has_window(struct lustre_scrub *scrub,
				       struct osd_otable_cache *ooc)
{
	return scrub->os_pos_current < ooc->ooc_pos_preload + SCRUB_WINDOW_SIZE;
}

/**
 * update/insert/delete the specified OI mapping (@fid @id) according to the ops
 *
 * \retval   1, changed nothing
 * \retval   0, changed successfully
 * \retval -ve, on error
 */
int osd_scrub_refresh_mapping(struct osd_thread_info *info,
			      struct osd_device *dev,
			      const struct lu_fid *fid,
			      const struct osd_inode_id *id,
			      int ops, bool force,
			      enum oi_check_flags flags, bool *exist)
{
	handle_t *th;
	int	  rc;
	ENTRY;

	if (dev->od_scrub.os_scrub.os_file.sf_param & SP_DRYRUN && !force)
		RETURN(0);

	/* DTO_INDEX_INSERT is enough for other two ops:
	 * delete/update, but save stack. */
	th = osd_journal_start_sb(osd_sb(dev), LDISKFS_HT_MISC,
				osd_dto_credits_noquota[DTO_INDEX_INSERT]);
	if (IS_ERR(th)) {
		rc = PTR_ERR(th);
		CDEBUG(D_LFSCK, "%s: fail to start trans for scrub op %d "
		       DFID" => %u/%u: rc = %d\n", osd_name(dev), ops,
		       PFID(fid), id ? id->oii_ino : -1, id ? id->oii_gen : -1,
		       rc);
		RETURN(rc);
	}

	switch (ops) {
	case DTO_INDEX_UPDATE:
		rc = osd_oi_update(info, dev, fid, id, th, flags);
		if (unlikely(rc == -ENOENT)) {
			/* Some unlink thread may removed the OI mapping. */
			rc = 1;
		}
		break;
	case DTO_INDEX_INSERT:
		rc = osd_oi_insert(info, dev, fid, id, th, flags, exist);
		if (unlikely(rc == -EEXIST)) {
			rc = 1;
			/* XXX: There are trouble things when adding OI
			 *	mapping for IGIF object, which may cause
			 *	multiple objects to be mapped to the same
			 *	IGIF formatted FID. Consider the following
			 *	situations:
			 *
			 *	1) The MDT is upgrading from 1.8 device.
			 *	The OI scrub generates IGIF FID1 for the
			 *	OBJ1 and adds the OI mapping.
			 *
			 *	2) For some reason, the OI scrub does not
			 *	process all the IGIF objects completely.
			 *
			 *	3) The MDT is backuped and restored against
			 *	this device.
			 *
			 *	4) When the MDT mounts up, the OI scrub will
			 *	try to rebuild the OI files. For some IGIF
			 *	object, OBJ2, which was not processed by the
			 *	OI scrub before the backup/restore, and the
			 *	new generated IGIF formatted FID may be just
			 *	the FID1, the same as OBJ1.
			 *
			 *	Under such case, the OI scrub cannot know how
			 *	to generate new FID for the OBJ2.
			 *
			 *	Currently, we do nothing for that. One possible
			 *	solution is to generate new normal FID for the
			 *	conflict object.
			 *
			 *	Anyway, it is rare, only exists in theory. */
		}
		break;
	case DTO_INDEX_DELETE:
		rc = osd_oi_delete(info, dev, fid, th, flags);
		if (rc == -ENOENT) {
			/* It is normal that the unlink thread has removed the
			 * OI mapping already. */
			rc = 1;
		}
		break;
	default:
		LASSERTF(0, "Unexpected ops %d\n", ops);
		break;
	}

	ldiskfs_journal_stop(th);
	if (rc < 0)
		CDEBUG(D_LFSCK, "%s: fail to refresh OI map for scrub op %d "
		       DFID" => %u/%u: rc = %d\n", osd_name(dev), ops,
		       PFID(fid), id ? id->oii_ino : -1, id ? id->oii_gen : -1,
		       rc);

	RETURN(rc);
}

static int
osd_scrub_convert_ff(struct osd_thread_info *info, struct osd_device *dev,
		     struct inode *inode, const struct lu_fid *fid)
{
	struct filter_fid_18_23 *ff = &info->oti_ff_old;
	struct dentry *dentry = &info->oti_obj_dentry;
	struct lu_fid *tfid = &info->oti_fid;
	bool fid_18_23 = false;
	handle_t *jh;
	int size = 0;
	int rc;
	ENTRY;

	if (dev->od_scrub.os_scrub.os_file.sf_param & SP_DRYRUN)
		RETURN(0);

	if (fid_is_idif(fid) && dev->od_index_in_idif == 0) {
		struct ost_id *oi = &info->oti_ostid;

		fid_to_ostid(fid, oi);
		ostid_to_fid(tfid, oi, 0);
	} else {
		*tfid = *fid;
	}

	/* We want the LMA to fit into the 256-byte OST inode, so operate
	 * as following:
	 * 1) read old XATTR_NAME_FID and save the parent FID;
	 * 2) delete the old XATTR_NAME_FID;
	 * 3) make new LMA and add it;
	 * 4) generate new XATTR_NAME_FID with the saved parent FID and add it.
	 *
	 * Making the LMA to fit into the 256-byte OST inode can save time for
	 * normal osd_check_lma() and for other OI scrub scanning in future.
	 * So it is worth to make some slow conversion here. */
	jh = osd_journal_start_sb(osd_sb(dev), LDISKFS_HT_MISC,
				osd_dto_credits_noquota[DTO_XATTR_SET] * 3);
	if (IS_ERR(jh)) {
		rc = PTR_ERR(jh);
		CDEBUG(D_LFSCK, "%s: fail to start trans for convert ff "
		       DFID": rc = %d\n", osd_name(dev), PFID(tfid), rc);
		RETURN(rc);
	}

	/* 1) read old XATTR_NAME_FID and save the parent FID */
	rc = __osd_xattr_get(inode, dentry, XATTR_NAME_FID, ff, sizeof(*ff));
	if (rc == sizeof(*ff)) {
		/* 2) delete the old XATTR_NAME_FID */
		dquot_initialize(inode);
		rc = ll_vfs_removexattr(dentry, inode, XATTR_NAME_FID);
		if (rc)
			GOTO(stop, rc);

		fid_18_23 = true;
	} else if (rc != -ENODATA && rc < (int)sizeof(struct filter_fid_24_29)) {
		GOTO(stop, rc = -EINVAL);
	}

	/* 3) make new LMA and add it */
	rc = osd_ea_fid_set(info, inode, tfid, LMAC_FID_ON_OST, 0);
	if (fid_18_23) {
		if (rc)
			/* If failed, we should try to add the old back. */
			size = sizeof(*ff);
		else
			/* The new PFID EA will only contains ::ff_parent */
			size = sizeof(ff->ff_parent);
	}

	/* 4) generate new XATTR_NAME_FID with the saved parent FID and add it*/
	if (size > 0) {
		int rc1;

		rc1 = __osd_xattr_set(info, inode, XATTR_NAME_FID, ff, size,
				      XATTR_CREATE);
		if (rc1 != 0 && rc == 0)
			rc = rc1;
	}

	GOTO(stop, rc);

stop:
	ldiskfs_journal_stop(jh);
	if (rc < 0)
		CDEBUG(D_LFSCK, "%s: fail to convert ff "DFID": rc = %d\n",
		       osd_name(dev), PFID(tfid), rc);
	return rc;
}

static int
osd_scrub_check_update(struct osd_thread_info *info, struct osd_device *dev,
		       struct osd_idmap_cache *oic, int val)
{
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;
	struct scrub_file	     *sf     = &scrub->os_file;
	struct lu_fid		     *fid    = &oic->oic_fid;
	struct osd_inode_id	     *lid    = &oic->oic_lid;
	struct osd_inode_id	     *lid2   = &info->oti_id;
	struct osd_inconsistent_item *oii    = NULL;
	struct inode		     *inode  = NULL;
	int			      ops    = DTO_INDEX_UPDATE;
	int			      rc;
	bool			      converted = false;
	bool			      exist	= false;
	ENTRY;

	down_write(&scrub->os_rwsem);
	scrub->os_new_checked++;
	if (val < 0)
		GOTO(out, rc = val);

	if (scrub->os_in_prior)
		oii = list_entry(oic, struct osd_inconsistent_item,
				 oii_cache);

	if (lid->oii_ino < sf->sf_pos_latest_start && oii == NULL)
		GOTO(out, rc = 0);

	if (fid_is_igif(fid))
		sf->sf_items_igif++;

	if (val == SCRUB_NEXT_OSTOBJ_OLD) {
		inode = osd_iget(info, dev, lid);
		if (IS_ERR(inode)) {
			rc = PTR_ERR(inode);
			/* Someone removed the inode. */
			if (rc == -ENOENT || rc == -ESTALE)
				rc = 0;
			GOTO(out, rc);
		}

		/* The inode has been reused as EA inode, ignore it. */
		if (unlikely(osd_is_ea_inode(inode)))
			GOTO(out, rc = 0);

		sf->sf_flags |= SF_UPGRADE;
		sf->sf_internal_flags &= ~SIF_NO_HANDLE_OLD_FID;
		dev->od_check_ff = 1;
		rc = osd_scrub_convert_ff(info, dev, inode, fid);
		if (rc != 0)
			GOTO(out, rc);

		converted = true;
	}

	if ((val == SCRUB_NEXT_NOLMA) &&
	    (!scrub->os_convert_igif || OBD_FAIL_CHECK(OBD_FAIL_FID_NOLMA)))
		GOTO(out, rc = 0);

	if ((oii != NULL && oii->oii_insert) || (val == SCRUB_NEXT_NOLMA)) {
		ops = DTO_INDEX_INSERT;

		goto iget;
	}

	rc = osd_oi_lookup(info, dev, fid, lid2,
		(val == SCRUB_NEXT_OSTOBJ ||
		 val == SCRUB_NEXT_OSTOBJ_OLD) ? OI_KNOWN_ON_OST : 0);
	if (rc != 0) {
		if (rc == -ENOENT)
			ops = DTO_INDEX_INSERT;
		else if (rc != -ESTALE)
			GOTO(out, rc);

iget:
		if (inode == NULL) {
			inode = osd_iget(info, dev, lid);
			if (IS_ERR(inode)) {
				rc = PTR_ERR(inode);
				/* Someone removed the inode. */
				if (rc == -ENOENT || rc == -ESTALE)
					rc = 0;
				GOTO(out, rc);
			}

			/* The inode has been reused as EA inode, ignore it. */
			if (unlikely(osd_is_ea_inode(inode)))
				GOTO(out, rc = 0);
		}

		switch (val) {
		case SCRUB_NEXT_NOLMA:
			sf->sf_flags |= SF_UPGRADE;
			if (!(sf->sf_param & SP_DRYRUN)) {
				rc = osd_ea_fid_set(info, inode, fid, 0, 0);
				if (rc != 0)
					GOTO(out, rc);
			}

			if (!(sf->sf_flags & SF_INCONSISTENT))
				dev->od_igif_inoi = 0;
			break;
		case SCRUB_NEXT_OSTOBJ:
			sf->sf_flags |= SF_INCONSISTENT;
		case SCRUB_NEXT_OSTOBJ_OLD:
			break;
		default:
			break;
		}
	} else if (osd_id_eq(lid, lid2)) {
		if (converted)
			sf->sf_items_updated++;

		GOTO(out, rc = 0);
	} else {
		if (!scrub->os_partial_scan) {
			spin_lock(&scrub->os_lock);
			scrub->os_full_speed = 1;
			spin_unlock(&scrub->os_lock);
		}
		sf->sf_flags |= SF_INCONSISTENT;

		/* XXX: If the device is restored from file-level backup, then
		 *	some IGIFs may have been already in OI files, and some
		 *	may be not yet. Means upgrading from 1.8 may be partly
		 *	processed, but some clients may hold some immobilized
		 *	IGIFs, and use them to access related objects. Under
		 *	such case, OSD does not know whether an given IGIF has
		 *	been processed or to be processed, and it also cannot
		 *	generate local ino#/gen# directly from the immobilized
		 *	IGIF because of the backup/restore. Then force OSD to
		 *	lookup the given IGIF in OI files, and if no entry,
		 *	then ask the client to retry after upgrading completed.
		 *	No better choice. */
		dev->od_igif_inoi = 1;
	}

	rc = osd_scrub_refresh_mapping(info, dev, fid, lid, ops, false,
			(val == SCRUB_NEXT_OSTOBJ ||
			 val == SCRUB_NEXT_OSTOBJ_OLD) ? OI_KNOWN_ON_OST : 0,
			&exist);
	if (rc == 0) {
		if (scrub->os_in_prior)
			sf->sf_items_updated_prior++;
		else
			sf->sf_items_updated++;

		if (ops == DTO_INDEX_INSERT && val == 0 && !exist) {
			int idx = osd_oi_fid2idx(dev, fid);

			sf->sf_flags |= SF_RECREATED;
			if (unlikely(!ldiskfs_test_bit(idx, sf->sf_oi_bitmap)))
				ldiskfs_set_bit(idx, sf->sf_oi_bitmap);
		}
	}

	GOTO(out, rc);

out:
	if (rc < 0) {
		sf->sf_items_failed++;
		if (sf->sf_pos_first_inconsistent == 0 ||
		    sf->sf_pos_first_inconsistent > lid->oii_ino)
			sf->sf_pos_first_inconsistent = lid->oii_ino;
	} else {
		rc = 0;
	}

	/* There may be conflict unlink during the OI scrub,
	 * if happend, then remove the new added OI mapping. */
	if (ops == DTO_INDEX_INSERT && inode != NULL && !IS_ERR(inode) &&
	    unlikely(ldiskfs_test_inode_state(inode,
					      LDISKFS_STATE_LUSTRE_DESTROY)))
		osd_scrub_refresh_mapping(info, dev, fid, lid,
				DTO_INDEX_DELETE, false,
				(val == SCRUB_NEXT_OSTOBJ ||
				 val == SCRUB_NEXT_OSTOBJ_OLD) ?
				OI_KNOWN_ON_OST : 0, NULL);
	up_write(&scrub->os_rwsem);

	if (!IS_ERR(inode))
		iput(inode);

	if (oii != NULL) {
		spin_lock(&scrub->os_lock);
		if (likely(!list_empty(&oii->oii_list)))
			list_del(&oii->oii_list);
		spin_unlock(&scrub->os_lock);

		OBD_FREE_PTR(oii);
	}

	RETURN(sf->sf_param & SP_FAILOUT ? rc : 0);
}

/* iteration engine */

typedef int (*osd_iit_next_policy)(struct osd_thread_info *info,
				   struct osd_device *dev,
				   struct osd_iit_param *param,
				   struct osd_idmap_cache **oic,
				   const bool noslot);

typedef int (*osd_iit_exec_policy)(struct osd_thread_info *info,
				   struct osd_device *dev,
				   struct osd_iit_param *param,
				   struct osd_idmap_cache *oic,
				   bool *noslot, int rc);

static int osd_iit_next(struct osd_iit_param *param, __u64 *pos)
{
	__u32 offset;

again:
	param->offset = ldiskfs_find_next_bit(param->bitmap->b_data,
			LDISKFS_INODES_PER_GROUP(param->sb), param->offset);
	if (param->offset >= LDISKFS_INODES_PER_GROUP(param->sb)) {
		*pos = 1 + (param->bg+1) * LDISKFS_INODES_PER_GROUP(param->sb);
		return SCRUB_NEXT_BREAK;
	}

	offset = param->offset++;
	if (unlikely(*pos == param->gbase + offset && *pos != param->start)) {
		/* We should NOT find the same object more than once. */
		CERROR("%s: scan the same object multiple times at the pos: "
		       "group = %u, base = %u, offset = %u, start = %u\n",
		       osd_sb2name(param->sb), (__u32)param->bg, param->gbase,
		       offset, param->start);
		goto again;
	}

	*pos = param->gbase + offset;
	return 0;
}

/**
 * \retval SCRUB_NEXT_OSTOBJ_OLD: FID-on-OST
 * \retval 0: FID-on-MDT
 */
static int osd_scrub_check_local_fldb(struct osd_thread_info *info,
				      struct osd_device *dev,
				      struct lu_fid *fid)
{
	/* XXX: The initial OI scrub will scan the top level /O to generate
	 *	a small local FLDB according to the <seq>. If the given FID
	 *	is in the local FLDB, then it is FID-on-OST; otherwise it's
	 *	quite possible for FID-on-MDT. */
	if (dev->od_is_ost)
		return SCRUB_NEXT_OSTOBJ_OLD;

	return 0;
}

static int osd_scrub_get_fid(struct osd_thread_info *info,
			     struct osd_device *dev, struct inode *inode,
			     struct lu_fid *fid, bool scrub)
{
	struct lustre_mdt_attrs *lma = &info->oti_ost_attrs.loa_lma;
	bool has_lma = false;
	int rc;

	rc = osd_get_lma(info, inode, &info->oti_obj_dentry,
			 &info->oti_ost_attrs);
	if (rc == 0) {
		has_lma = true;
		if (lma->lma_compat & LMAC_NOT_IN_OI ||
		    lma->lma_incompat & LMAI_AGENT)
			return SCRUB_NEXT_CONTINUE;

		*fid = lma->lma_self_fid;
		if (!scrub)
			return 0;

		if (lma->lma_compat & LMAC_FID_ON_OST)
			return SCRUB_NEXT_OSTOBJ;

		if (fid_is_idif(fid))
			return SCRUB_NEXT_OSTOBJ_OLD;

		/* For local object. */
		if (fid_is_internal(fid))
			return 0;

		/* For external visible MDT-object with non-normal FID. */
		if (fid_is_namespace_visible(fid) && !fid_is_norm(fid))
			return 0;

		/* For the object with normal FID, it may be MDT-object,
		 * or may be 2.4 OST-object, need further distinguish.
		 * Fall through to next section. */
	}

	if (rc == -ENODATA || rc == 0) {
		rc = osd_get_idif(info, inode, &info->oti_obj_dentry, fid);
		if (rc == 0) {
			if (scrub)
				/* It is 2.3 or older OST-object. */
				rc = SCRUB_NEXT_OSTOBJ_OLD;
			return rc;
		}

		if (rc > 0) {
			if (!has_lma)
				/* It is FID-on-OST, but we do not know how
				 * to generate its FID, ignore it directly. */
				rc = SCRUB_NEXT_CONTINUE;
			else
				/* It is 2.4 or newer OST-object. */
				rc = SCRUB_NEXT_OSTOBJ_OLD;
			return rc;
		}

		if (rc != -ENODATA)
			return rc;

		if (!has_lma) {
			if (dev->od_scrub.os_scrub.os_convert_igif) {
				lu_igif_build(fid, inode->i_ino,
					      inode->i_generation);
				if (scrub)
					rc = SCRUB_NEXT_NOLMA;
				else
					rc = 0;
			} else {
				/* It may be FID-on-OST, or may be FID for
				 * non-MDT0, anyway, we do not know how to
				 * generate its FID, ignore it directly. */
				rc = SCRUB_NEXT_CONTINUE;
			}
			return rc;
		}

		/* For OI scrub case only: the object has LMA but has no ff
		 * (or ff crashed). It may be MDT-object, may be OST-object
		 * with crashed ff. The last check is local FLDB. */
		rc = osd_scrub_check_local_fldb(info, dev, fid);
	}

	return rc;
}

static int osd_iit_iget(struct osd_thread_info *info, struct osd_device *dev,
			struct lu_fid *fid, struct osd_inode_id *lid, __u32 pos,
			struct super_block *sb, bool scrub)
{
	struct inode *inode;
	int	      rc;
	ENTRY;

	/* Not handle the backend root object and agent parent object.
	 * They are neither visible to namespace nor have OI mappings. */
	if (unlikely(pos == osd_sb(dev)->s_root->d_inode->i_ino ||
		     is_remote_parent_ino(dev, pos)))
		RETURN(SCRUB_NEXT_CONTINUE);

	 /* Skip project quota inode since it is greater than s_first_ino. */
#ifdef HAVE_PROJECT_QUOTA
	if (ldiskfs_has_feature_project(sb) &&
	    pos == le32_to_cpu(LDISKFS_SB(sb)->s_es->s_prj_quota_inum))
		RETURN(SCRUB_NEXT_CONTINUE);
#endif

	osd_id_gen(lid, pos, OSD_OII_NOGEN);
	inode = osd_iget(info, dev, lid);
	if (IS_ERR(inode)) {
		rc = PTR_ERR(inode);
		/* The inode may be removed after bitmap searching, or the
		 * file is new created without inode initialized yet. */
		if (rc == -ENOENT || rc == -ESTALE)
			RETURN(SCRUB_NEXT_CONTINUE);

		CDEBUG(D_LFSCK, "%s: fail to read inode, ino# = %u: "
		       "rc = %d\n", osd_dev2name(dev), pos, rc);
		RETURN(rc);
	}

	if (dev->od_is_ost && S_ISREG(inode->i_mode) && inode->i_nlink > 1)
		dev->od_scrub.os_scrub.os_has_ml_file = 1;

	/* It is an EA inode, no OI mapping for it, skip it. */
	if (osd_is_ea_inode(inode))
		GOTO(put, rc = SCRUB_NEXT_CONTINUE);

	if (scrub &&
	    ldiskfs_test_inode_state(inode, LDISKFS_STATE_LUSTRE_NOSCRUB)) {
		/* Only skip it for the first OI scrub accessing. */
		ldiskfs_clear_inode_state(inode, LDISKFS_STATE_LUSTRE_NOSCRUB);
		GOTO(put, rc = SCRUB_NEXT_NOSCRUB);
	}

	rc = osd_scrub_get_fid(info, dev, inode, fid, scrub);

	GOTO(put, rc);

put:
	iput(inode);
	return rc;
}

static int osd_scrub_next(struct osd_thread_info *info, struct osd_device *dev,
			  struct osd_iit_param *param,
			  struct osd_idmap_cache **oic, const bool noslot)
{
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;
	struct lu_fid *fid;
	struct osd_inode_id *lid;
	int rc;

	if (OBD_FAIL_CHECK(OBD_FAIL_OSD_SCRUB_DELAY) && cfs_fail_val > 0)
		wait_var_event_timeout(
			scrub,
			!list_empty(&scrub->os_inconsistent_items) ||
			kthread_should_stop(),
			cfs_time_seconds(cfs_fail_val));

	if (OBD_FAIL_CHECK(OBD_FAIL_OSD_SCRUB_CRASH)) {
		spin_lock(&scrub->os_lock);
		scrub->os_running = 0;
		spin_unlock(&scrub->os_lock);
		return SCRUB_NEXT_CRASH;
	}

	if (OBD_FAIL_CHECK(OBD_FAIL_OSD_SCRUB_FATAL))
		return SCRUB_NEXT_FATAL;

	if (kthread_should_stop())
		return SCRUB_NEXT_EXIT;

	if (!list_empty(&scrub->os_inconsistent_items)) {
		spin_lock(&scrub->os_lock);
		if (likely(!list_empty(&scrub->os_inconsistent_items))) {
			struct osd_inconsistent_item *oii;

			oii = list_entry(scrub->os_inconsistent_items.next,
				struct osd_inconsistent_item, oii_list);

			*oic = &oii->oii_cache;
			scrub->os_in_prior = 1;
			spin_unlock(&scrub->os_lock);

			return 0;
		}
		spin_unlock(&scrub->os_lock);
	}

	if (noslot)
		return SCRUB_NEXT_WAIT;

	rc = osd_iit_next(param, &scrub->os_pos_current);
	if (rc != 0)
		return rc;

	*oic = &dev->od_scrub.os_oic;
	fid = &(*oic)->oic_fid;
	lid = &(*oic)->oic_lid;
	rc = osd_iit_iget(info, dev, fid, lid,
			  scrub->os_pos_current, param->sb, true);
	return rc;
}

static int osd_preload_next(struct osd_thread_info *info,
			    struct osd_device *dev, struct osd_iit_param *param,
			    struct osd_idmap_cache **oic, const bool noslot)
{
	struct osd_otable_cache *ooc = &dev->od_otable_it->ooi_cache;
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;
	int rc;

	if (scrub->os_running &&
	    ooc->ooc_pos_preload >= scrub->os_pos_current)
		return SCRUB_NEXT_EXIT;

	rc = osd_iit_next(param, &ooc->ooc_pos_preload);
	if (rc)
		return rc;

	rc = osd_iit_iget(info, dev,
			  &ooc->ooc_cache[ooc->ooc_producer_idx].oic_fid,
			  &ooc->ooc_cache[ooc->ooc_producer_idx].oic_lid,
			  ooc->ooc_pos_preload, param->sb, false);
	return rc;
}

static inline int
osd_scrub_wakeup(struct lustre_scrub *scrub, struct osd_otable_it *it)
{
	spin_lock(&scrub->os_lock);
	if (osd_scrub_has_window(scrub, &it->ooi_cache) ||
	    !list_empty(&scrub->os_inconsistent_items) ||
	    it->ooi_waiting || kthread_should_stop())
		scrub->os_waiting = 0;
	else
		scrub->os_waiting = 1;
	spin_unlock(&scrub->os_lock);

	return !scrub->os_waiting;
}

static int osd_scrub_exec(struct osd_thread_info *info, struct osd_device *dev,
			  struct osd_iit_param *param,
			  struct osd_idmap_cache *oic, bool *noslot, int rc)
{
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;
	struct scrub_file *sf = &scrub->os_file;
	struct osd_otable_it *it = dev->od_otable_it;
	struct osd_otable_cache *ooc = it ? &it->ooi_cache : NULL;

	switch (rc) {
	case SCRUB_NEXT_NOSCRUB:
		down_write(&scrub->os_rwsem);
		scrub->os_new_checked++;
		sf->sf_items_noscrub++;
		up_write(&scrub->os_rwsem);
	case SCRUB_NEXT_CONTINUE:
	case SCRUB_NEXT_WAIT:
		goto wait;
	}

	rc = osd_scrub_check_update(info, dev, oic, rc);
	if (rc != 0) {
		spin_lock(&scrub->os_lock);
		scrub->os_in_prior = 0;
		spin_unlock(&scrub->os_lock);
		return rc;
	}

	rc = scrub_checkpoint(info->oti_env, scrub);
	if (rc) {
		CDEBUG(D_LFSCK, "%s: fail to checkpoint, pos = %llu: "
		       "rc = %d\n", osd_scrub2name(scrub),
		       scrub->os_pos_current, rc);
		/* Continue, as long as the scrub itself can go ahead. */
	}

	if (scrub->os_in_prior) {
		spin_lock(&scrub->os_lock);
		scrub->os_in_prior = 0;
		spin_unlock(&scrub->os_lock);
		return 0;
	}

wait:
	if (it != NULL && it->ooi_waiting && ooc != NULL &&
	    ooc->ooc_pos_preload < scrub->os_pos_current) {
		spin_lock(&scrub->os_lock);
		it->ooi_waiting = 0;
		wake_up_var(scrub);
		spin_unlock(&scrub->os_lock);
	}

	if (rc == SCRUB_NEXT_CONTINUE)
		return 0;

	if (scrub->os_full_speed || !ooc || osd_scrub_has_window(scrub, ooc)) {
		*noslot = false;
		return 0;
	}

	if (it)
		wait_var_event(scrub, osd_scrub_wakeup(scrub, it));

	if (!ooc || osd_scrub_has_window(scrub, ooc))
		*noslot = false;
	else
		*noslot = true;
	return 0;
}

static int osd_preload_exec(struct osd_thread_info *info,
			    struct osd_device *dev, struct osd_iit_param *param,
			    struct osd_idmap_cache *oic, bool *noslot, int rc)
{
	struct osd_otable_cache *ooc = &dev->od_otable_it->ooi_cache;

	if (rc == 0) {
		ooc->ooc_cached_items++;
		ooc->ooc_producer_idx = (ooc->ooc_producer_idx + 1) &
					~OSD_OTABLE_IT_CACHE_MASK;
	}
	return rc > 0 ? 0 : rc;
}

#define SCRUB_IT_ALL	1
#define SCRUB_IT_CRASH	2

static void osd_scrub_join(const struct lu_env *env, struct osd_device *dev,
			   __u32 flags, bool inconsistent)
{
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;
	struct scrub_file    *sf     = &scrub->os_file;
	int		      rc;
	ENTRY;

	LASSERT(!(flags & SS_AUTO_PARTIAL));

	down_write(&scrub->os_rwsem);
	spin_lock(&scrub->os_lock);
	scrub->os_in_join = 1;
	if (flags & SS_SET_FAILOUT)
		sf->sf_param |= SP_FAILOUT;
	else if (flags & SS_CLEAR_FAILOUT)
		sf->sf_param &= ~SP_FAILOUT;

	if (flags & SS_SET_DRYRUN)
		sf->sf_param |= SP_DRYRUN;
	else if (flags & SS_CLEAR_DRYRUN)
		sf->sf_param &= ~SP_DRYRUN;

	if (flags & SS_RESET) {
		scrub_file_reset(scrub, dev->od_uuid,
				 inconsistent ? SF_INCONSISTENT : 0);
		sf->sf_status = SS_SCANNING;
	}

	if (sf->sf_flags & (SF_RECREATED | SF_INCONSISTENT | SF_UPGRADE))
		scrub->os_full_speed = 1;
	else
		scrub->os_full_speed = 0;

	if (flags & SS_AUTO_FULL) {
		sf->sf_flags |= SF_AUTO;
		scrub->os_full_speed = 1;
	}
	spin_unlock(&scrub->os_lock);

	scrub->os_new_checked = 0;
	if (sf->sf_pos_last_checkpoint != 0)
		sf->sf_pos_latest_start = sf->sf_pos_last_checkpoint + 1;
	else
		sf->sf_pos_latest_start = LDISKFS_FIRST_INO(osd_sb(dev)) + 1;

	scrub->os_pos_current = sf->sf_pos_latest_start;
	sf->sf_time_latest_start = ktime_get_real_seconds();
	sf->sf_time_last_checkpoint = sf->sf_time_latest_start;
	sf->sf_pos_last_checkpoint = sf->sf_pos_latest_start - 1;
	rc = scrub_file_store(env, scrub);

	spin_lock(&scrub->os_lock);
	scrub->os_waiting = 0;
	scrub->os_paused = 0;
	scrub->os_partial_scan = 0;
	scrub->os_in_join = 0;
	scrub->os_full_scrub = 0;
	spin_unlock(&scrub->os_lock);
	wake_up_var(scrub);
	up_write(&scrub->os_rwsem);

	CDEBUG(D_LFSCK, "%s: joined in the OI scrub with flag %u: rc = %d\n",
	       osd_scrub2name(scrub), flags, rc);

	EXIT;
}

static int osd_inode_iteration(struct osd_thread_info *info,
			       struct osd_device *dev, __u32 max, bool preload)
{
	struct lustre_scrub *scrub  = &dev->od_scrub.os_scrub;
	struct scrub_file *sf = &scrub->os_file;
	osd_iit_next_policy next;
	osd_iit_exec_policy exec;
	__u64 *pos;
	__u64 *count;
	struct osd_iit_param *param;
	__u32 limit;
	int rc;
	bool noslot = true;
	ENTRY;

	if (preload)
		goto full;

	param = &dev->od_scrub.os_iit_param;
	memset(param, 0, sizeof(*param));
	param->sb = osd_sb(dev);

	while (scrub->os_partial_scan && !scrub->os_in_join) {
		struct osd_idmap_cache *oic = NULL;

		rc = osd_scrub_next(info, dev, param, &oic, noslot);
		switch (rc) {
		case SCRUB_NEXT_EXIT:
			RETURN(0);
		case SCRUB_NEXT_CRASH:
			RETURN(SCRUB_IT_CRASH);
		case SCRUB_NEXT_FATAL:
			RETURN(-EINVAL);
		case SCRUB_NEXT_WAIT: {
			struct kstatfs *ksfs = &info->oti_ksfs;
			__u64 saved_flags;

			if (dev->od_full_scrub_ratio == OFSR_NEVER ||
			    unlikely(sf->sf_items_updated_prior == 0))
				goto wait;

			if (dev->od_full_scrub_ratio == OFSR_DIRECTLY ||
			    scrub->os_full_scrub) {
				osd_scrub_join(info->oti_env, dev,
					       SS_AUTO_FULL | SS_RESET, true);
				goto full;
			}

			rc = param->sb->s_op->statfs(param->sb->s_root, ksfs);
			if (rc == 0) {
				__u64 used = ksfs->f_files - ksfs->f_ffree;

				used = div64_u64(used, sf->sf_items_updated_prior);
				/* If we hit too much inconsistent OI
				 * mappings during the partial scan,
				 * then scan the device completely. */
				if (used < dev->od_full_scrub_ratio) {
					osd_scrub_join(info->oti_env, dev,
						SS_AUTO_FULL | SS_RESET, true);
					goto full;
				}
			}

wait:
			if (OBD_FAIL_CHECK(OBD_FAIL_OSD_SCRUB_DELAY) &&
			    cfs_fail_val > 0)
				continue;

			saved_flags = sf->sf_flags;
			sf->sf_flags &= ~(SF_RECREATED | SF_INCONSISTENT |
					  SF_UPGRADE | SF_AUTO);
			sf->sf_status = SS_COMPLETED;
			wait_var_event(
				scrub,
				kthread_should_stop() ||
				!scrub->os_partial_scan ||
				scrub->os_in_join ||
				!list_empty(&scrub->os_inconsistent_items));
			sf->sf_flags = saved_flags;
			sf->sf_status = SS_SCANNING;

			if (kthread_should_stop())
				RETURN(0);

			if (!scrub->os_partial_scan || scrub->os_in_join)
				goto full;

			continue;
		}
		default:
			LASSERTF(rc == 0, "rc = %d\n", rc);

			osd_scrub_exec(info, dev, param, oic, &noslot, rc);
			break;
		}
	}

full:
	if (!preload) {
		wait_var_event(scrub,
			       kthread_should_stop() ||
			       !scrub->os_in_join);

		if (kthread_should_stop())
			RETURN(0);
	}

	noslot = false;
	if (!preload) {
		next = osd_scrub_next;
		exec = osd_scrub_exec;
		pos = &scrub->os_pos_current;
		count = &scrub->os_new_checked;
		param->start = *pos;
		param->bg = (*pos - 1) / LDISKFS_INODES_PER_GROUP(param->sb);
		param->offset =
			(*pos - 1) % LDISKFS_INODES_PER_GROUP(param->sb);
		param->gbase =
			1 + param->bg * LDISKFS_INODES_PER_GROUP(param->sb);
	} else {
		struct osd_otable_cache *ooc = &dev->od_otable_it->ooi_cache;

		next = osd_preload_next;
		exec = osd_preload_exec;
		pos = &ooc->ooc_pos_preload;
		count = &ooc->ooc_cached_items;
		param = &dev->od_otable_it->ooi_iit_param;
	}

	rc = 0;
	limit = le32_to_cpu(LDISKFS_SB(osd_sb(dev))->s_es->s_inodes_count);
	while (*pos <= limit && *count < max) {
		struct ldiskfs_group_desc *desc;
		bool next_group = false;

		desc = ldiskfs_get_group_desc(param->sb, param->bg, NULL);
		if (!desc)
			RETURN(-EIO);

		if (desc->bg_flags & cpu_to_le16(LDISKFS_BG_INODE_UNINIT)) {
			next_group = true;
			goto next_group;
		}

		param->bitmap = ldiskfs_read_inode_bitmap(param->sb, param->bg);
		if (!param->bitmap) {
			CERROR("%s: fail to read bitmap for %u, "
			       "scrub will stop, urgent mode\n",
			       osd_scrub2name(scrub), (__u32)param->bg);
			RETURN(-EIO);
		}

		do {
			struct osd_idmap_cache *oic = NULL;

			if (param->offset +
				ldiskfs_itable_unused_count(param->sb, desc) >=
			    LDISKFS_INODES_PER_GROUP(param->sb)) {
				next_group = true;
				goto next_group;
			}

			rc = next(info, dev, param, &oic, noslot);
			switch (rc) {
			case SCRUB_NEXT_BREAK:
				next_group = true;
				goto next_group;
			case SCRUB_NEXT_EXIT:
				brelse(param->bitmap);
				RETURN(0);
			case SCRUB_NEXT_CRASH:
				brelse(param->bitmap);
				RETURN(SCRUB_IT_CRASH);
			case SCRUB_NEXT_FATAL:
				brelse(param->bitmap);
				RETURN(-EINVAL);
			}

			rc = exec(info, dev, param, oic, &noslot, rc);
		} while (!rc && *pos <= limit && *count < max);

next_group:
		if (param->bitmap) {
			brelse(param->bitmap);
			param->bitmap = NULL;
		}

		if (rc < 0)
			GOTO(out, rc);

		if (next_group) {
			param->bg++;
			param->offset = 0;
			param->gbase = 1 +
				param->bg * LDISKFS_INODES_PER_GROUP(param->sb);
			*pos = param->gbase;
			param->start = *pos;
		}
	}

	if (*pos > limit)
		RETURN(SCRUB_IT_ALL);

out:
	RETURN(rc);
}

static int osd_otable_it_preload(const struct lu_env *env,
				 struct osd_otable_it *it)
{
	struct osd_device *dev = it->ooi_dev;
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;
	struct osd_otable_cache *ooc   = &it->ooi_cache;
	int			 rc;
	ENTRY;

	rc = osd_inode_iteration(osd_oti_get(env), dev,
				 OSD_OTABLE_IT_CACHE_SIZE, true);
	if (rc == SCRUB_IT_ALL)
		it->ooi_all_cached = 1;

	if (scrub->os_waiting && osd_scrub_has_window(scrub, ooc)) {
		spin_lock(&scrub->os_lock);
		scrub->os_waiting = 0;
		wake_up_var(scrub);
		spin_unlock(&scrub->os_lock);
	}

	RETURN(rc < 0 ? rc : ooc->ooc_cached_items);
}

static int osd_scan_ml_file_main(const struct lu_env *env,
				 struct osd_device *dev);

static int osd_scrub_main(void *args)
{
	struct lu_env env;
	struct osd_device *dev = (struct osd_device *)args;
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;
	int rc, ret;
	ENTRY;

	rc = lu_env_init(&env, LCT_LOCAL | LCT_DT_THREAD);
	if (rc != 0) {
		CDEBUG(D_LFSCK, "%s: OI scrub fail to init env: rc = %d\n",
		       osd_scrub2name(scrub), rc);
		GOTO(noenv, rc);
	}

	rc = scrub_thread_prep(&env, scrub, dev->od_uuid,
			       LDISKFS_FIRST_INO(osd_sb(dev)) + 1);
	if (rc != 0) {
		CDEBUG(D_LFSCK, "%s: OI scrub fail to scrub prep: rc = %d\n",
		       osd_scrub2name(scrub), rc);
		GOTO(out, rc);
	}

	if (!scrub->os_full_speed && !scrub->os_partial_scan) {
		struct osd_otable_it *it = dev->od_otable_it;
		struct osd_otable_cache *ooc = &it->ooi_cache;

		wait_var_event(scrub,
			       it->ooi_user_ready || kthread_should_stop());
		if (kthread_should_stop())
			GOTO(post, rc = 0);

		scrub->os_pos_current = ooc->ooc_pos_preload;
	}

	CDEBUG(D_LFSCK, "%s: OI scrub start, flags = 0x%x, pos = %llu\n",
	       osd_scrub2name(scrub), scrub->os_start_flags,
	       scrub->os_pos_current);

	rc = osd_inode_iteration(osd_oti_get(&env), dev, ~0U, false);
	if (unlikely(rc == SCRUB_IT_CRASH)) {
		spin_lock(&scrub->os_lock);
		scrub->os_running = 0;
		spin_unlock(&scrub->os_lock);
		GOTO(out, rc = -EINVAL);
	}

	if (scrub->os_has_ml_file) {
		ret = osd_scan_ml_file_main(&env, dev);
		if (ret != 0)
			rc = ret;
	}

	GOTO(post, rc);

post:
	if (rc > 0) {
		dev->od_igif_inoi = 1;
		dev->od_check_ff = 0;
	}
	rc = scrub_thread_post(&env, &dev->od_scrub.os_scrub, rc);
	CDEBUG(D_LFSCK, "%s: OI scrub: stop, pos = %llu: rc = %d\n",
	       osd_scrub2name(scrub), scrub->os_pos_current, rc);

out:
	while (!list_empty(&scrub->os_inconsistent_items)) {
		struct osd_inconsistent_item *oii;

		oii = list_entry(scrub->os_inconsistent_items.next,
				 struct osd_inconsistent_item, oii_list);
		list_del_init(&oii->oii_list);
		OBD_FREE_PTR(oii);
	}

	lu_env_fini(&env);

noenv:
	spin_lock(&scrub->os_lock);
	scrub->os_running = 0;
	spin_unlock(&scrub->os_lock);
	if (xchg(&scrub->os_task, NULL) == NULL)
		/* scrub_stop() is waiting, we need to synchronize */
		wait_var_event(scrub, kthread_should_stop());
	wake_up_var(scrub);
	return rc;
}

/* initial OI scrub */

typedef int (*scandir_t)(struct osd_thread_info *, struct osd_device *,
			 struct dentry *, filldir_t filldir);

#ifdef HAVE_FILLDIR_USE_CTX
static int osd_ios_varfid_fill(struct dir_context *buf, const char *name,
			       int namelen, loff_t offset, __u64 ino,
			       unsigned d_type);
static int osd_ios_lf_fill(struct dir_context *buf, const char *name,
			   int namelen, loff_t offset, __u64 ino,
			   unsigned d_type);
static int osd_ios_dl_fill(struct dir_context *buf, const char *name,
			   int namelen, loff_t offset, __u64 ino,
			   unsigned d_type);
static int osd_ios_uld_fill(struct dir_context *buf, const char *name,
			    int namelen, loff_t offset, __u64 ino,
			    unsigned d_type);
#else
static int osd_ios_varfid_fill(void *buf, const char *name, int namelen,
			       loff_t offset, __u64 ino, unsigned d_type);
static int osd_ios_lf_fill(void *buf, const char *name, int namelen,
			   loff_t offset, __u64 ino, unsigned d_type);
static int osd_ios_dl_fill(void *buf, const char *name, int namelen,
			   loff_t offset, __u64 ino, unsigned d_type);
static int osd_ios_uld_fill(void *buf, const char *name, int namelen,
			    loff_t offset, __u64 ino, unsigned d_type);
#endif

static int
osd_ios_general_scan(struct osd_thread_info *info, struct osd_device *dev,
		     struct dentry *dentry, filldir_t filldir);
static int
osd_ios_ROOT_scan(struct osd_thread_info *info, struct osd_device *dev,
		  struct dentry *dentry, filldir_t filldir);

static int
osd_ios_OBJECTS_scan(struct osd_thread_info *info, struct osd_device *dev,
		     struct dentry *dentry, filldir_t filldir);

struct osd_lf_map {
	char		*olm_name;
	struct lu_fid	 olm_fid;
	__u16		 olm_flags;
	__u16		 olm_namelen;
	scandir_t	 olm_scandir;
	filldir_t	 olm_filldir;
};

/* Add the new introduced local files in the list in the future. */
static const struct osd_lf_map osd_lf_maps[] = {
	/* CATALOGS */
	{
		.olm_name	= CATLIST,
		.olm_fid	= {
			.f_seq	= FID_SEQ_LOCAL_FILE,
			.f_oid	= LLOG_CATALOGS_OID,
		},
		.olm_flags	= OLF_SHOW_NAME,
		.olm_namelen	= sizeof(CATLIST) - 1,
	},

	/* CONFIGS */
	{
		.olm_name	= MOUNT_CONFIGS_DIR,
		.olm_fid	= {
			.f_seq	= FID_SEQ_LOCAL_FILE,
			.f_oid	= MGS_CONFIGS_OID,
		},
		.olm_flags	= OLF_SCAN_SUBITEMS,
		.olm_namelen	= sizeof(MOUNT_CONFIGS_DIR) - 1,
		.olm_scandir	= osd_ios_general_scan,
		.olm_filldir	= osd_ios_varfid_fill,
	},

	/* NIDTBL_VERSIONS */
	{
		.olm_name	= MGS_NIDTBL_DIR,
		.olm_flags	= OLF_SCAN_SUBITEMS,
		.olm_namelen	= sizeof(MGS_NIDTBL_DIR) - 1,
		.olm_scandir	= osd_ios_general_scan,
		.olm_filldir	= osd_ios_varfid_fill,
	},

	/* PENDING */
	{
		.olm_name	= MDT_ORPHAN_DIR,
		.olm_namelen	= sizeof(MDT_ORPHAN_DIR) - 1,
	},

	/* ROOT */
	{
		.olm_name	= "ROOT",
		.olm_fid	= {
			.f_seq	= FID_SEQ_ROOT,
			.f_oid	= FID_OID_ROOT,
		},
		.olm_flags	= OLF_SCAN_SUBITEMS | OLF_HIDE_FID,
		.olm_namelen	= sizeof("ROOT") - 1,
		.olm_scandir	= osd_ios_ROOT_scan,
	},

	/* changelog_catalog */
	{
		.olm_name	= CHANGELOG_CATALOG,
		.olm_namelen	= sizeof(CHANGELOG_CATALOG) - 1,
	},

	/* changelog_users */
	{
		.olm_name	= CHANGELOG_USERS,
		.olm_namelen	= sizeof(CHANGELOG_USERS) - 1,
	},

	/* fld */
	{
		.olm_name	= "fld",
		.olm_fid	= {
			.f_seq	= FID_SEQ_LOCAL_FILE,
			.f_oid	= FLD_INDEX_OID,
		},
		.olm_flags	= OLF_SHOW_NAME,
		.olm_namelen	= sizeof("fld") - 1,
	},

	/* last_rcvd */
	{
		.olm_name	= LAST_RCVD,
		.olm_fid	= {
			.f_seq	= FID_SEQ_LOCAL_FILE,
			.f_oid	= LAST_RECV_OID,
		},
		.olm_flags	= OLF_SHOW_NAME,
		.olm_namelen	= sizeof(LAST_RCVD) - 1,
	},

	/* reply_data */
	{
		.olm_name	= REPLY_DATA,
		.olm_fid	= {
			.f_seq	= FID_SEQ_LOCAL_FILE,
			.f_oid	= REPLY_DATA_OID,
		},
		.olm_flags	= OLF_SHOW_NAME,
		.olm_namelen	= sizeof(REPLY_DATA) - 1,
	},

	/* lov_objid */
	{
		.olm_name	= LOV_OBJID,
		.olm_fid	= {
			.f_seq	= FID_SEQ_LOCAL_FILE,
			.f_oid	= MDD_LOV_OBJ_OID,
		},
		.olm_flags	= OLF_SHOW_NAME,
		.olm_namelen	= sizeof(LOV_OBJID) - 1,
	},

	/* lov_objseq */
	{
		.olm_name	= LOV_OBJSEQ,
		.olm_fid	= {
			.f_seq	= FID_SEQ_LOCAL_FILE,
			.f_oid	= MDD_LOV_OBJ_OSEQ,
		},
		.olm_flags	= OLF_SHOW_NAME,
		.olm_namelen	= sizeof(LOV_OBJSEQ) - 1,
	},

	/* quota_master */
	{
		.olm_name	= QMT_DIR,
		.olm_flags	= OLF_SCAN_SUBITEMS,
		.olm_namelen	= sizeof(QMT_DIR) - 1,
		.olm_scandir	= osd_ios_general_scan,
		.olm_filldir	= osd_ios_varfid_fill,
	},

	/* quota_slave */
	{
		.olm_name	= QSD_DIR,
		.olm_flags	= OLF_SCAN_SUBITEMS,
		.olm_namelen	= sizeof(QSD_DIR) - 1,
		.olm_scandir	= osd_ios_general_scan,
		.olm_filldir	= osd_ios_varfid_fill,
	},

	/* seq_ctl */
	{
		.olm_name	= "seq_ctl",
		.olm_fid	= {
			.f_seq	= FID_SEQ_LOCAL_FILE,
			.f_oid	= FID_SEQ_CTL_OID,
		},
		.olm_flags	= OLF_SHOW_NAME,
		.olm_namelen	= sizeof("seq_ctl") - 1,
	},

	/* seq_srv */
	{
		.olm_name	= "seq_srv",
		.olm_fid	= {
			.f_seq	= FID_SEQ_LOCAL_FILE,
			.f_oid	= FID_SEQ_SRV_OID,
		},
		.olm_flags	= OLF_SHOW_NAME,
		.olm_namelen	= sizeof("seq_srv") - 1,
	},

	/* health_check */
	{
		.olm_name	= HEALTH_CHECK,
		.olm_fid	= {
			.f_seq	= FID_SEQ_LOCAL_FILE,
			.f_oid	= OFD_HEALTH_CHECK_OID,
		},
		.olm_flags	= OLF_SHOW_NAME,
		.olm_namelen	= sizeof(HEALTH_CHECK) - 1,
	},

	/* LFSCK */
	{
		.olm_name	= LFSCK_DIR,
		.olm_flags	= OLF_SCAN_SUBITEMS,
		.olm_namelen	= sizeof(LFSCK_DIR) - 1,
		.olm_scandir	= osd_ios_general_scan,
		.olm_filldir	= osd_ios_varfid_fill,
	},

	/* lfsck_bookmark */
	{
		.olm_name	= LFSCK_BOOKMARK,
		.olm_namelen	= sizeof(LFSCK_BOOKMARK) - 1,
	},

	/* lfsck_layout */
	{
		.olm_name	= LFSCK_LAYOUT,
		.olm_namelen	= sizeof(LFSCK_LAYOUT) - 1,
	},

	/* lfsck_namespace */
	{
		.olm_name	= LFSCK_NAMESPACE,
		.olm_namelen	= sizeof(LFSCK_NAMESPACE) - 1,
	},

	/* OBJECTS, upgrade from old device */
	{
		.olm_name	= OBJECTS,
		.olm_flags	= OLF_SCAN_SUBITEMS,
		.olm_namelen	= sizeof(OBJECTS) - 1,
		.olm_scandir	= osd_ios_OBJECTS_scan,
	},

	/* lquota_v2.user, upgrade from old device */
	{
		.olm_name	= "lquota_v2.user",
		.olm_namelen	= sizeof("lquota_v2.user") - 1,
	},

	/* lquota_v2.group, upgrade from old device */
	{
		.olm_name	= "lquota_v2.group",
		.olm_namelen	= sizeof("lquota_v2.group") - 1,
	},

	/* LAST_GROUP, upgrade from old device */
	{
		.olm_name	= "LAST_GROUP",
		.olm_fid	= {
			.f_seq	= FID_SEQ_LOCAL_FILE,
			.f_oid	= OFD_LAST_GROUP_OID,
		},
		.olm_flags	= OLF_SHOW_NAME,
		.olm_namelen	= sizeof("LAST_GROUP") - 1,
	},

	/* committed batchid for cross-MDT operation */
	{
		.olm_name	= "BATCHID",
		.olm_fid	= {
			.f_seq	= FID_SEQ_LOCAL_FILE,
			.f_oid	= BATCHID_COMMITTED_OID,
		},
		.olm_flags	= OLF_SHOW_NAME,
		.olm_namelen	= sizeof("BATCHID") - 1,
	},

	/* OSP update logs update_log{_dir} use f_seq = FID_SEQ_UPDATE_LOG{_DIR}
	 * and f_oid = index for their log files.  See lu_update_log{_dir}_fid()
	 * for more details. */

	/* update_log */
	{
		.olm_name	= "update_log",
		.olm_fid	= {
			.f_seq	= FID_SEQ_UPDATE_LOG,
		},
		.olm_flags	= OLF_SHOW_NAME | OLF_IDX_IN_FID,
		.olm_namelen	= sizeof("update_log") - 1,
	},

	/* update_log_dir */
	{
		.olm_name	= "update_log_dir",
		.olm_fid	= {
			.f_seq	= FID_SEQ_UPDATE_LOG_DIR,
		},
		.olm_flags	= OLF_SHOW_NAME | OLF_SCAN_SUBITEMS |
				  OLF_IDX_IN_FID,
		.olm_namelen	= sizeof("update_log_dir") - 1,
		.olm_scandir	= osd_ios_general_scan,
		.olm_filldir	= osd_ios_uld_fill,
	},

	/* lost+found */
	{
		.olm_name	= "lost+found",
		.olm_fid	= {
			.f_seq	= FID_SEQ_LOCAL_FILE,
			.f_oid	= OSD_LPF_OID,
		},
		.olm_flags	= OLF_SCAN_SUBITEMS,
		.olm_namelen	= sizeof("lost+found") - 1,
		.olm_scandir	= osd_ios_general_scan,
		.olm_filldir	= osd_ios_lf_fill,
	},

	/* hsm_actions */
	{
		.olm_name	= HSM_ACTIONS,
	},

	/* nodemap */
	{
		.olm_name	= LUSTRE_NODEMAP_NAME,
	},

	/* index_backup */
	{
		.olm_name	= INDEX_BACKUP_DIR,
		.olm_fid	= {
			.f_seq	= FID_SEQ_LOCAL_FILE,
			.f_oid	= INDEX_BACKUP_OID,
		},
		.olm_flags	= OLF_SCAN_SUBITEMS | OLF_NOT_BACKUP,
		.olm_namelen	= sizeof(INDEX_BACKUP_DIR) - 1,
		.olm_scandir	= osd_ios_general_scan,
		.olm_filldir	= osd_ios_varfid_fill,
	},

	{
		.olm_name	= NULL
	}
};

/* Add the new introduced files under .lustre/ in the list in the future. */
static const struct osd_lf_map osd_dl_maps[] = {
	/* .lustre/fid */
	{
		.olm_name	= "fid",
		.olm_fid	= {
			.f_seq	= FID_SEQ_DOT_LUSTRE,
			.f_oid	= FID_OID_DOT_LUSTRE_OBF,
		},
		.olm_namelen	= sizeof("fid") - 1,
	},

	/* .lustre/lost+found */
	{
		.olm_name	= "lost+found",
		.olm_fid	= {
			.f_seq	= FID_SEQ_DOT_LUSTRE,
			.f_oid	= FID_OID_DOT_LUSTRE_LPF,
		},
		.olm_namelen	= sizeof("lost+found") - 1,
	},

	{
		.olm_name	= NULL
	}
};

struct osd_ios_item {
	struct list_head oii_list;
	struct dentry	*oii_dentry;
	scandir_t	 oii_scandir;
	filldir_t	 oii_filldir;
};

struct osd_ios_filldir_buf {
	/* please keep it as first member */
	struct dir_context	 ctx;
	struct osd_thread_info	*oifb_info;
	struct osd_device	*oifb_dev;
	struct dentry		*oifb_dentry;
	int			 oifb_items;
};

static int
osd_ios_new_item(struct osd_device *dev, struct dentry *dentry,
		 scandir_t scandir, filldir_t filldir)
{
	struct osd_ios_item *item;
	ENTRY;

	OBD_ALLOC_PTR(item);
	if (item == NULL)
		RETURN(-ENOMEM);

	INIT_LIST_HEAD(&item->oii_list);
	item->oii_dentry = dget(dentry);
	item->oii_scandir = scandir;
	item->oii_filldir = filldir;
	list_add_tail(&item->oii_list, &dev->od_ios_list);

	RETURN(0);
}

static bool osd_index_need_recreate(const struct lu_env *env,
				    struct osd_device *dev, struct inode *inode)
{
	struct osd_directory *iam = &osd_oti_get(env)->oti_iam;
	struct iam_container *bag = &iam->od_container;
	int rc;
	ENTRY;

	rc = iam_container_init(bag, &iam->od_descr, inode);
	if (rc)
		RETURN(true);

	rc = iam_container_setup(bag);
	iam_container_fini(bag);
	if (rc)
		RETURN(true);

	RETURN(false);
}

static void osd_ios_index_register(const struct lu_env *env,
				   struct osd_device *osd,
				   const struct lu_fid *fid,
				   struct inode *inode)
{
	struct osd_directory *iam = &osd_oti_get(env)->oti_iam;
	struct iam_container *bag = &iam->od_container;
	struct super_block *sb = osd_sb(osd);
	struct iam_descr *descr;
	__u32 keysize = 0;
	__u32 recsize = 0;
	int rc;
	ENTRY;

	/* Index must be a regular file. */
	if (!S_ISREG(inode->i_mode))
		RETURN_EXIT;

	/* Index's size must be block aligned. */
	if (inode->i_size < sb->s_blocksize ||
	    (inode->i_size & (sb->s_blocksize - 1)) != 0)
		RETURN_EXIT;

	iam_container_init(bag, &iam->od_descr, inode);
	rc = iam_container_setup(bag);
	if (rc)
		GOTO(fini, rc = 1);

	descr = bag->ic_descr;
	/* May be regular file with IAM_LFIX_ROOT_MAGIC matched
	 * coincidentally, or corrupted index object, skip it. */
	if (descr->id_ptr_size != 4)
		GOTO(fini, rc = 1);

	keysize = descr->id_key_size;
	recsize = descr->id_rec_size;
	rc = osd_index_register(osd, fid, keysize, recsize);

	GOTO(fini, rc);

fini:
	iam_container_fini(bag);
	if (!rc)
		CDEBUG(D_LFSCK, "%s: index object "DFID" (%u/%u) registered\n",
		       osd_name(osd), PFID(fid), keysize, recsize);
}

static void osd_index_restore(const struct lu_env *env, struct osd_device *dev,
			      struct lustre_index_restore_unit *liru,
			      void *buf, int bufsize)
{
	struct osd_thread_info *info = osd_oti_get(env);
	struct osd_inode_id *id = &info->oti_id;
	struct lu_fid *tgt_fid = &liru->liru_cfid;
	struct inode *bak_inode = NULL;
	struct ldiskfs_dir_entry_2 *de = NULL;
	struct buffer_head *bh = NULL;
	struct dentry *dentry;
	char *name = buf;
	struct lu_fid bak_fid;
	int rc;
	ENTRY;

	lustre_fid2lbx(name, tgt_fid, bufsize);
	dentry = osd_child_dentry_by_inode(env, dev->od_index_backup_inode,
					   name, strlen(name));
	bh = osd_ldiskfs_find_entry(dev->od_index_backup_inode,
				    &dentry->d_name, &de, NULL, NULL);
	if (IS_ERR(bh))
		GOTO(log, rc = PTR_ERR(bh));

	osd_id_gen(id, le32_to_cpu(de->inode), OSD_OII_NOGEN);
	brelse(bh);
	bak_inode = osd_iget_fid(info, dev, id, &bak_fid);
	if (IS_ERR(bak_inode))
		GOTO(log, rc = PTR_ERR(bak_inode));

	iput(bak_inode);
	/* The OI mapping for index may be invalid, since it will be
	 * re-created, not update the OI mapping, just cache it in RAM. */
	osd_id_gen(id, liru->liru_clid, OSD_OII_NOGEN);
	osd_add_oi_cache(info, dev, id, tgt_fid);
	rc = lustre_index_restore(env, &dev->od_dt_dev, &liru->liru_pfid,
				  tgt_fid, &bak_fid, liru->liru_name,
				  &dev->od_index_backup_list, &dev->od_lock,
				  buf, bufsize);
	GOTO(log, rc);

log:
	CDEBUG(D_WARNING, "%s: restore index '%s' with "DFID": rc = %d\n",
	       osd_name(dev), liru->liru_name, PFID(tgt_fid), rc);
}

/**
 * osd_ios_scan_one() - check/fix LMA FID and OI entry for one inode
 *
 * The passed \a inode's \a fid is verified against the LMA FID. If the \a fid
 * is NULL or is empty the IGIF FID is used. The FID is verified in the OI to
 * reference the inode, or fixed if it is missing or references another inode.
 */
static int
osd_ios_scan_one(struct osd_thread_info *info, struct osd_device *dev,
		 struct inode *parent, struct inode *inode,
		 const struct lu_fid *fid, const char *name,
		 int namelen, int flags)
{
	struct lustre_mdt_attrs	*lma	= &info->oti_ost_attrs.loa_lma;
	struct osd_inode_id	*id	= &info->oti_id;
	struct osd_inode_id	*id2	= &info->oti_id2;
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;
	struct scrub_file	*sf     = &scrub->os_file;
	struct lu_fid		 tfid;
	int			 rc;
	ENTRY;

	if (!inode) {
		CDEBUG(D_INODE, "%s: child '%.*s' lacks inode: rc = -2\n",
		       osd_name(dev), namelen, name);
		RETURN(-ENOENT);
	}

	rc = osd_get_lma(info, inode, &info->oti_obj_dentry,
			 &info->oti_ost_attrs);
	if (rc != 0 && rc != -ENODATA) {
		CDEBUG(D_LFSCK, "%s: fail to get lma for init OI scrub: "
		       "rc = %d\n", osd_name(dev), rc);

		RETURN(rc);
	}

	osd_id_gen(id, inode->i_ino, inode->i_generation);
	if (rc == -ENODATA) {
		if (fid == NULL || fid_is_zero(fid) || flags & OLF_HIDE_FID) {
			lu_igif_build(&tfid, inode->i_ino, inode->i_generation);
		} else {
			tfid = *fid;
			if (flags & OLF_IDX_IN_FID) {
				LASSERT(dev->od_index >= 0);

				tfid.f_oid = dev->od_index;
			}
		}
		rc = osd_ea_fid_set(info, inode, &tfid, 0, 0);
		if (rc != 0) {
			CDEBUG(D_LFSCK, "%s: fail to set LMA for init OI "
			      "scrub: rc = %d\n", osd_name(dev), rc);

			RETURN(rc);
		}
	} else {
		if (lma->lma_compat & LMAC_NOT_IN_OI)
			RETURN(0);

		tfid = lma->lma_self_fid;
		if (lma->lma_compat & LMAC_IDX_BACKUP &&
		    osd_index_need_recreate(info->oti_env, dev, inode)) {
			struct lu_fid *pfid = &info->oti_fid3;

			if (is_root_inode(parent)) {
				lu_local_obj_fid(pfid, OSD_FS_ROOT_OID);
			} else {
				rc = osd_scrub_get_fid(info, dev, parent, pfid,
						       false);
				if (rc)
					RETURN(rc);
			}

			rc = lustre_liru_new(&dev->od_index_restore_list, pfid,
					&tfid, inode->i_ino, name, namelen);

			RETURN(rc);
		}

		if (!(flags & OLF_NOT_BACKUP))
			osd_ios_index_register(info->oti_env, dev, &tfid,
					       inode);
	}

	/* Since this called from iterate_dir() the inode lock will be taken */
	rc = osd_oi_lookup(info, dev, &tfid, id2, OI_LOCKED);
	if (rc != 0) {
		if (rc != -ENOENT)
			RETURN(rc);

		rc = osd_scrub_refresh_mapping(info, dev, &tfid, id,
					       DTO_INDEX_INSERT, true, 0, NULL);
		if (rc > 0)
			rc = 0;

		RETURN(rc);
	}

	if (osd_id_eq_strict(id, id2))
		RETURN(0);

	if (!(sf->sf_flags & SF_INCONSISTENT)) {
		scrub_file_reset(scrub, dev->od_uuid, SF_INCONSISTENT);
		rc = scrub_file_store(info->oti_env, scrub);
		if (rc != 0)
			RETURN(rc);
	}

	rc = osd_scrub_refresh_mapping(info, dev, &tfid, id,
				       DTO_INDEX_UPDATE, true, 0, NULL);
	if (rc > 0)
		rc = 0;

	RETURN(rc);
}

/**
 * It scans the /lost+found, and for the OST-object (with filter_fid
 * or filter_fid_18_23), move them back to its proper /O/<seq>/d<x>.
 */
#ifdef HAVE_FILLDIR_USE_CTX
static int osd_ios_lf_fill(struct dir_context *buf,
#else
static int osd_ios_lf_fill(void *buf,
#endif
			   const char *name, int namelen,
			   loff_t offset, __u64 ino, unsigned d_type)
{
	struct osd_ios_filldir_buf *fill_buf =
		(struct osd_ios_filldir_buf *)buf;
	struct osd_thread_info     *info     = fill_buf->oifb_info;
	struct osd_device	   *dev      = fill_buf->oifb_dev;
	struct lu_fid		   *fid      = &info->oti_fid;
	struct osd_scrub	   *scrub    = &dev->od_scrub;
	struct dentry		   *parent   = fill_buf->oifb_dentry;
	struct dentry		   *child;
	struct inode		   *dir      = parent->d_inode;
	struct inode		   *inode;
	int			    rc;
	ENTRY;

	fill_buf->oifb_items++;

	/* skip any '.' started names */
	if (name[0] == '.')
		RETURN(0);

	scrub->os_lf_scanned++;
	child = osd_lookup_one_len(dev, name, parent, namelen);
	if (IS_ERR(child)) {
		rc = PTR_ERR(child);
		CDEBUG(D_LFSCK, "%s: cannot lookup child '%.*s': rc = %d\n",
		      osd_name(dev), namelen, name, rc);
		RETURN(rc);
	} else if (!child->d_inode) {
		dput(child);
		CDEBUG(D_INODE, "%s: child '%.*s' lacks inode\n",
		       osd_name(dev), namelen, name);
		RETURN(-ENOENT);
	}

	inode = child->d_inode;
	if (S_ISDIR(inode->i_mode)) {
		rc = osd_ios_new_item(dev, child, osd_ios_general_scan,
				      osd_ios_lf_fill);
		if (rc != 0)
			CDEBUG(D_LFSCK, "%s: cannot add child '%.*s': "
			      "rc = %d\n", osd_name(dev), namelen, name, rc);
		GOTO(put, rc);
	}

	if (!S_ISREG(inode->i_mode))
		GOTO(put, rc = 0);

	rc = osd_scrub_get_fid(info, dev, inode, fid, true);
	if (rc == SCRUB_NEXT_OSTOBJ || rc == SCRUB_NEXT_OSTOBJ_OLD) {
		rc = osd_obj_map_recover(info, dev, dir, child, fid);
		if (rc == 0) {
			CDEBUG(D_LFSCK, "recovered '%.*s' ["DFID"] from "
			       "/lost+found.\n", namelen, name, PFID(fid));
			scrub->os_lf_repaired++;
		} else {
			CDEBUG(D_LFSCK, "%s: cannot rename for '%.*s' "
			       DFID": rc = %d\n",
			       osd_name(dev), namelen, name, PFID(fid), rc);
		}
	}

	/* XXX: For MDT-objects, we can move them from /lost+found to namespace
	 * 	visible place, such as the /ROOT/.lustre/lost+found, then LFSCK
	 * 	can process them in furtuer. */

	GOTO(put, rc);

put:
	if (rc < 0)
		scrub->os_lf_failed++;
	dput(child);
	/* skip the failure to make the scanning to continue. */
	return 0;
}

#ifdef HAVE_FILLDIR_USE_CTX
static int osd_ios_varfid_fill(struct dir_context *buf,
#else
static int osd_ios_varfid_fill(void *buf,
#endif
			       const char *name, int namelen,
			       loff_t offset, __u64 ino, unsigned d_type)
{
	struct osd_ios_filldir_buf *fill_buf =
		(struct osd_ios_filldir_buf *)buf;
	struct osd_device	   *dev      = fill_buf->oifb_dev;
	struct dentry		   *child;
	int			    rc;
	ENTRY;

	fill_buf->oifb_items++;

	/* skip any '.' started names */
	if (name[0] == '.')
		RETURN(0);

	child = osd_lookup_one_len(dev, name, fill_buf->oifb_dentry, namelen);
	if (IS_ERR(child))
		RETURN(PTR_ERR(child));

	rc = osd_ios_scan_one(fill_buf->oifb_info, dev,
			      fill_buf->oifb_dentry->d_inode, child->d_inode,
			      NULL, name, namelen, 0);
	if (rc == 0 && S_ISDIR(child->d_inode->i_mode))
		rc = osd_ios_new_item(dev, child, osd_ios_general_scan,
				      osd_ios_varfid_fill);
	dput(child);

	RETURN(rc);
}

#ifdef HAVE_FILLDIR_USE_CTX
static int osd_ios_dl_fill(struct dir_context *buf,
#else
static int osd_ios_dl_fill(void *buf,
#endif
			   const char *name, int namelen,
			   loff_t offset, __u64 ino, unsigned d_type)
{
	struct osd_ios_filldir_buf *fill_buf =
		(struct osd_ios_filldir_buf *)buf;
	struct osd_device	   *dev      = fill_buf->oifb_dev;
	const struct osd_lf_map    *map;
	struct dentry		   *child;
	int			    rc       = 0;
	ENTRY;

	fill_buf->oifb_items++;

	/* skip any '.' started names */
	if (name[0] == '.')
		RETURN(0);

	for (map = osd_dl_maps; map->olm_name != NULL; map++) {
		if (map->olm_namelen != namelen)
			continue;

		if (strncmp(map->olm_name, name, namelen) == 0)
			break;
	}

	if (map->olm_name == NULL)
		RETURN(0);

	child = osd_lookup_one_len(dev, name, fill_buf->oifb_dentry, namelen);
	if (IS_ERR(child))
		RETURN(PTR_ERR(child));

	rc = osd_ios_scan_one(fill_buf->oifb_info, dev,
			      fill_buf->oifb_dentry->d_inode, child->d_inode,
			      &map->olm_fid, name, namelen, map->olm_flags);
	dput(child);

	RETURN(rc);
}

#ifdef HAVE_FILLDIR_USE_CTX
static int osd_ios_uld_fill(struct dir_context *buf,
#else
static int osd_ios_uld_fill(void *buf,
#endif
			    const char *name, int namelen,
			    loff_t offset, __u64 ino, unsigned d_type)
{
	struct osd_ios_filldir_buf *fill_buf =
		(struct osd_ios_filldir_buf *)buf;
	struct osd_device *dev = fill_buf->oifb_dev;
	struct dentry		   *child;
	struct lu_fid		    tfid;
	int			    rc       = 0;
	ENTRY;

	fill_buf->oifb_items++;

	/* skip any non-DFID format name */
	if (name[0] != '[')
		RETURN(0);

	child = osd_lookup_one_len(dev, name, fill_buf->oifb_dentry, namelen);
	if (IS_ERR(child))
		RETURN(PTR_ERR(child));

	/* skip the start '[' */
	sscanf(&name[1], SFID, RFID(&tfid));
	if (fid_is_sane(&tfid))
		rc = osd_ios_scan_one(fill_buf->oifb_info, fill_buf->oifb_dev,
				      fill_buf->oifb_dentry->d_inode,
				      child->d_inode, &tfid, name, namelen, 0);
	else
		rc = -EIO;
	dput(child);

	RETURN(rc);
}

#ifdef HAVE_FILLDIR_USE_CTX
static int osd_ios_root_fill(struct dir_context *buf,
#else
static int osd_ios_root_fill(void *buf,
#endif
			     const char *name, int namelen,
			     loff_t offset, __u64 ino, unsigned d_type)
{
	struct osd_ios_filldir_buf *fill_buf =
		(struct osd_ios_filldir_buf *)buf;
	struct osd_device	   *dev      = fill_buf->oifb_dev;
	const struct osd_lf_map    *map;
	struct dentry		   *child;
	int			    rc       = 0;
	ENTRY;

	fill_buf->oifb_items++;

	/* skip any '.' started names */
	if (name[0] == '.')
		RETURN(0);

	for (map = osd_lf_maps; map->olm_name != NULL; map++) {
		if (map->olm_namelen != namelen)
			continue;

		if (strncmp(map->olm_name, name, namelen) == 0)
			break;
	}

	if (map->olm_name == NULL)
		RETURN(0);

	child = osd_lookup_one_len(dev, name, fill_buf->oifb_dentry, namelen);
	if (IS_ERR(child))
		RETURN(PTR_ERR(child));
	else if (!child->d_inode)
		GOTO(out_put, rc = -ENOENT);

	if (!(map->olm_flags & OLF_NO_OI))
		rc = osd_ios_scan_one(fill_buf->oifb_info, dev,
				fill_buf->oifb_dentry->d_inode, child->d_inode,
				&map->olm_fid, name, namelen, map->olm_flags);
	if (rc == 0 && map->olm_flags & OLF_SCAN_SUBITEMS)
		rc = osd_ios_new_item(dev, child, map->olm_scandir,
				      map->olm_filldir);
out_put:
	dput(child);

	RETURN(rc);
}

static int
osd_ios_general_scan(struct osd_thread_info *info, struct osd_device *dev,
		     struct dentry *dentry, filldir_t filldir)
{
	struct osd_ios_filldir_buf buf = {
		.ctx.actor = filldir,
		.oifb_info = info,
		.oifb_dev = dev,
		.oifb_dentry = dentry
	};
	struct file *filp;
	struct path path;
	int rc;

	ENTRY;
	LASSERT(filldir);
	path.dentry = dget(dentry);
	path.mnt = mntget(dev->od_mnt);

	filp = dentry_open(&path, O_RDONLY, current_cred());
	path_put(&path);
	if (IS_ERR(filp))
		RETURN(PTR_ERR(filp));

	filp->f_mode |= FMODE_64BITHASH | FMODE_NONOTIFY;
	filp->f_flags |= O_NOATIME;
	filp->f_pos = 0;

	do {
		buf.oifb_items = 0;
		rc = iterate_dir(filp, &buf.ctx);
	} while (rc >= 0 && buf.oifb_items > 0 &&
		 filp->f_pos != LDISKFS_HTREE_EOF_64BIT);
	fput(filp);

	RETURN(rc);
}

static int
osd_ios_ROOT_scan(struct osd_thread_info *info, struct osd_device *dev,
		  struct dentry *dentry, filldir_t filldir)
{
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;
	struct scrub_file *sf = &scrub->os_file;
	struct dentry *child;
	int rc;
	ENTRY;

	/* It is existing MDT0 device. We only allow the case of object without
	 * LMA to happen on the MDT0, which is usually for old 1.8 MDT. Then we
	 * can generate IGIF mode FID for the object and related OI mapping. If
	 * it is on other MDTs, then becuase file-level backup/restore, related
	 * OI mapping may be invalid already, we do not know which is the right
	 * FID for the object. We only allow IGIF objects to reside on the MDT0.
	 *
	 * XXX: For the case of object on non-MDT0 device with neither LMA nor
	 *	"fid" xattr, then something crashed. We cannot re-generate the
	 *	FID directly, instead, the OI scrub will scan the OI structure
	 *	and try to re-generate the LMA from the OI mapping. But if the
	 *	OI mapping crashed or lost also, then we have to give up under
	 *	double failure cases.
	 */
	spin_lock(&scrub->os_lock);
	scrub->os_convert_igif = 1;
	spin_unlock(&scrub->os_lock);
	child = osd_lookup_one_len_unlocked(dev, dot_lustre_name, dentry,
					    strlen(dot_lustre_name));
	if (IS_ERR(child)) {
		if (PTR_ERR(child) != -ENOENT)
			RETURN(PTR_ERR(child));
		goto out_scrub;
	}

	/* For lustre-2.x (x <= 3), the ".lustre" has NO FID-in-LMA,
	 * so the client will get IGIF for the ".lustre" object when
	 * the MDT restart.
	 *
	 * From the OI scrub view, when the MDT upgrade to Lustre-2.4,
	 * it does not know whether there are some old clients cached
	 * the ".lustre" IGIF during the upgrading. Two choices:
	 *
	 * 1) Generate IGIF-in-LMA and IGIF-in-OI for the ".lustre".
	 *    It will allow the old connected clients to access the
	 *    ".lustre" with cached IGIF. But it will cause others
	 *    on the MDT failed to check "fid_is_dot_lustre()".
	 *
	 * 2) Use fixed FID {FID_SEQ_DOT_LUSTRE, FID_OID_DOT_LUSTRE, 0}
	 *    for ".lustre" in spite of whether there are some clients
	 *    cached the ".lustre" IGIF or not. It enables the check
	 *    "fid_is_dot_lustre()" on the MDT, although it will cause
	 *    that the old connected clients cannot access the ".lustre"
	 *    with the cached IGIF.
	 *
	 * Usually, it is rare case for the old connected clients
	 * to access the ".lustre" with cached IGIF. So we prefer
	 * to the solution 2).
	 */
	inode_lock(dentry->d_inode);
	rc = osd_ios_scan_one(info, dev, dentry->d_inode,
			      child->d_inode, &LU_DOT_LUSTRE_FID,
			      dot_lustre_name,
			      strlen(dot_lustre_name), 0);
	inode_unlock(dentry->d_inode);
	if (rc == -ENOENT) {
out_scrub:
		/* It is 1.8 MDT device. */
		if (!(sf->sf_flags & SF_UPGRADE)) {
			scrub_file_reset(scrub, dev->od_uuid,
					 SF_UPGRADE);
			sf->sf_internal_flags &= ~SIF_NO_HANDLE_OLD_FID;
			rc = scrub_file_store(info->oti_env, scrub);
		} else {
			rc = 0;
		}
	} else if (rc == 0) {
		rc = osd_ios_new_item(dev, child, osd_ios_general_scan,
				      osd_ios_dl_fill);
	}
	dput(child);

	RETURN(rc);
}

static int
osd_ios_OBJECTS_scan(struct osd_thread_info *info, struct osd_device *dev,
		     struct dentry *dentry, filldir_t filldir)
{
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;
	struct scrub_file *sf = &scrub->os_file;
	struct dentry *child;
	int rc;
	ENTRY;

	if (unlikely(sf->sf_internal_flags & SIF_NO_HANDLE_OLD_FID)) {
		sf->sf_internal_flags &= ~SIF_NO_HANDLE_OLD_FID;
		rc = scrub_file_store(info->oti_env, scrub);
		if (rc != 0)
			RETURN(rc);
	}

	child = osd_lookup_one_len_unlocked(dev, ADMIN_USR, dentry,
					    strlen(ADMIN_USR));
	if (IS_ERR(child)) {
		rc = PTR_ERR(child);
	} else {
		inode_lock(dentry->d_inode);
		rc = osd_ios_scan_one(info, dev, dentry->d_inode,
				      child->d_inode, NULL, ADMIN_USR,
				      strlen(ADMIN_USR), 0);
		inode_unlock(dentry->d_inode);
		dput(child);
	}

	if (rc != 0 && rc != -ENOENT)
		GOTO(out, rc);

	child = osd_lookup_one_len_unlocked(dev, ADMIN_GRP, dentry,
					    strlen(ADMIN_GRP));
	if (IS_ERR(child))
		GOTO(out, rc = PTR_ERR(child));

	inode_lock(dentry->d_inode);
	rc = osd_ios_scan_one(info, dev, dentry->d_inode,
			      child->d_inode, NULL, ADMIN_GRP,
			      strlen(ADMIN_GRP), 0);
	inode_unlock(dentry->d_inode);
	dput(child);
out:
	RETURN(rc == -ENOENT ? 0 : rc);
}

static void osd_initial_OI_scrub(struct osd_thread_info *info,
				 struct osd_device *dev)
{
	struct osd_ios_item	*item    = NULL;
	scandir_t		 scandir = osd_ios_general_scan;
	filldir_t		 filldir = osd_ios_root_fill;
	struct dentry		*dentry  = osd_sb(dev)->s_root;
	const struct osd_lf_map *map     = osd_lf_maps;
	ENTRY;

	/* Lookup IGIF in OI by force for initial OI scrub. */
	dev->od_igif_inoi = 1;

	while (1) {
		/* Don't take inode_lock here since scandir() callbacks
		 * can call VFS functions which may manully take the
		 * inode lock itself like iterate_dir(). Since this
		 * is the case it is best to leave the scandir()
		 * callbacks to managing the inode lock.
		 */
		scandir(info, dev, dentry, filldir);
		if (item != NULL) {
			dput(item->oii_dentry);
			OBD_FREE_PTR(item);
		}

		if (list_empty(&dev->od_ios_list))
			break;

		item = list_entry(dev->od_ios_list.next,
				  struct osd_ios_item, oii_list);
		list_del_init(&item->oii_list);

		LASSERT(item->oii_scandir != NULL);
		scandir = item->oii_scandir;
		filldir = item->oii_filldir;
		dentry = item->oii_dentry;
	}

	/* There maybe the case that the object has been removed, but its OI
	 * mapping is still in the OI file, such as the "CATALOGS" after MDT
	 * file-level backup/restore. So here cleanup the stale OI mappings. */
	while (map->olm_name != NULL) {
		struct dentry *child;

		if (fid_is_zero(&map->olm_fid)) {
			map++;
			continue;
		}

		child = osd_lookup_one_len_unlocked(dev, map->olm_name,
						    osd_sb(dev)->s_root,
						    map->olm_namelen);
		if (PTR_ERR(child) == -ENOENT ||
		    (!IS_ERR(child) && !child->d_inode))
			osd_scrub_refresh_mapping(info, dev, &map->olm_fid,
						  NULL, DTO_INDEX_DELETE,
						  true, 0, NULL);
		if (!IS_ERR(child))
			dput(child);
		map++;
	}

	if (!list_empty(&dev->od_index_restore_list)) {
		char *buf;

		OBD_ALLOC_LARGE(buf, INDEX_BACKUP_BUFSIZE);
		if (!buf)
			CERROR("%s: not enough RAM for rebuild index\n",
			       osd_name(dev));

		while (!list_empty(&dev->od_index_restore_list)) {
			struct lustre_index_restore_unit *liru;

			liru = list_entry(dev->od_index_restore_list.next,
					  struct lustre_index_restore_unit,
					  liru_link);
			list_del(&liru->liru_link);
			if (buf)
				osd_index_restore(info->oti_env, dev, liru,
						  buf, INDEX_BACKUP_BUFSIZE);
			OBD_FREE(liru, liru->liru_len);
		}

		if (buf)
			OBD_FREE_LARGE(buf, INDEX_BACKUP_BUFSIZE);
	}

	EXIT;
}

char *osd_lf_fid2name(const struct lu_fid *fid)
{
	const struct osd_lf_map *map = osd_lf_maps;

	while (map->olm_name != NULL) {
		if (!lu_fid_eq(fid, &map->olm_fid)) {
			map++;
			continue;
		}

		if (map->olm_flags & OLF_SHOW_NAME)
			return map->olm_name;
		else
			return "";
	}

	return NULL;
}

/* OI scrub start/stop */

int osd_scrub_start(const struct lu_env *env, struct osd_device *dev,
		    __u32 flags)
{
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;
	int rc;
	ENTRY;

	if (dev->od_dt_dev.dd_rdonly)
		RETURN(-EROFS);

	/* od_otable_mutex: prevent curcurrent start/stop */
	mutex_lock(&dev->od_otable_mutex);
	rc = scrub_start(osd_scrub_main, scrub, dev, flags);
	if (rc == -EALREADY) {
		rc = 0;
		if ((scrub->os_file.sf_flags & SF_AUTO ||
		     scrub->os_partial_scan) &&
		    !(flags & SS_AUTO_PARTIAL))
			osd_scrub_join(env, dev, flags, false);
	}
	mutex_unlock(&dev->od_otable_mutex);

	RETURN(rc);
}

void osd_scrub_stop(struct osd_device *dev)
{
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;

	/* od_otable_mutex: prevent curcurrent start/stop */
	mutex_lock(&dev->od_otable_mutex);
	spin_lock(&scrub->os_lock);
	scrub->os_paused = 1;
	spin_unlock(&scrub->os_lock);
	scrub_stop(scrub);
	mutex_unlock(&dev->od_otable_mutex);
}

/* OI scrub setup/cleanup */

static const char osd_scrub_name[] = "OI_scrub";

int osd_scrub_setup(const struct lu_env *env, struct osd_device *dev,
		    bool restored)
{
	struct osd_thread_info *info = osd_oti_get(env);
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;
	struct lvfs_run_ctxt *ctxt = &dev->od_scrub.os_ctxt;
	time64_t interval = scrub->os_auto_scrub_interval;
	struct scrub_file *sf = &scrub->os_file;
	struct super_block *sb = osd_sb(dev);
	struct lvfs_run_ctxt saved;
	struct file *filp;
	struct inode *inode;
	struct lu_fid *fid = &info->oti_fid;
	struct osd_inode_id *id = &info->oti_id;
	struct dt_object *obj;
	bool dirty = false;
	int rc = 0;
	ENTRY;

	memset(&dev->od_scrub, 0, sizeof(struct osd_scrub));
	OBD_SET_CTXT_MAGIC(ctxt);
	ctxt->pwdmnt = dev->od_mnt;
	ctxt->pwd = dev->od_mnt->mnt_root;

	init_rwsem(&scrub->os_rwsem);
	spin_lock_init(&scrub->os_lock);
	INIT_LIST_HEAD(&scrub->os_inconsistent_items);
	scrub->os_name = osd_name(dev);
	scrub->os_auto_scrub_interval = interval;

	push_ctxt(&saved, ctxt);
	filp = filp_open(osd_scrub_name, O_RDWR |
			 (dev->od_dt_dev.dd_rdonly ? 0 : O_CREAT), 0644);
	if (IS_ERR(filp)) {
		pop_ctxt(&saved, ctxt);
		RETURN(PTR_ERR(filp));
	}

	inode = file_inode(filp);
	ldiskfs_set_inode_flag(inode, LDISKFS_INODE_JOURNAL_DATA);
	if (!dev->od_dt_dev.dd_rdonly) {
		/* 'What the @fid is' is not imporatant, because the object
		 * has no OI mapping, and only is visible inside the OSD.*/
		lu_igif_build(fid, inode->i_ino, inode->i_generation);
		rc = osd_ea_fid_set(info, inode, fid, LMAC_NOT_IN_OI, 0);
		if (rc) {
			filp_close(filp, NULL);
			pop_ctxt(&saved, ctxt);
			RETURN(rc);
		}
	}

	osd_id_gen(id, inode->i_ino, inode->i_generation);
	osd_add_oi_cache(info, dev, id, fid);
	filp_close(filp, NULL);
	pop_ctxt(&saved, ctxt);

	obj = lu2dt(lu_object_find_slice(env, osd2lu_dev(dev), fid, NULL));
	if (IS_ERR_OR_NULL(obj))
		RETURN(obj ? PTR_ERR(obj) : -ENOENT);

#ifndef HAVE_S_UUID_AS_UUID_T
	memcpy(dev->od_uuid.b, sb->s_uuid, sizeof(dev->od_uuid));
#else
	uuid_copy(&dev->od_uuid, &sb->s_uuid);
#endif
	scrub->os_obj = obj;
	rc = scrub_file_load(env, scrub);
	if (rc == -ENOENT || rc == -EFAULT) {
		scrub_file_init(scrub, dev->od_uuid);
		/* If the "/O" dir does not exist when mount (indicated by
		 * osd_device::od_maybe_new), neither for the "/OI_scrub",
		 * then it is quite probably that the device is a new one,
		 * under such case, mark it as SIF_NO_HANDLE_OLD_FID.
		 *
		 * For the rare case that "/O" and "OI_scrub" both lost on
		 * an old device, it can be found and cleared later.
		 *
		 * For the system with "SIF_NO_HANDLE_OLD_FID", we do not
		 * need to check "filter_fid_18_23" and to convert it to
		 * "filter_fid" for each object, and all the IGIF should
		 * have their FID mapping in OI files already. */
		if (dev->od_maybe_new && rc == -ENOENT)
			sf->sf_internal_flags = SIF_NO_HANDLE_OLD_FID;
		dirty = true;
	} else if (rc < 0) {
		GOTO(cleanup_obj, rc);
	} else {
		if (!uuid_equal(&sf->sf_uuid, &dev->od_uuid)) {
			CDEBUG(D_LFSCK,
			       "%s: UUID has been changed from %pU to %pU\n",
			       osd_dev2name(dev), &sf->sf_uuid, &dev->od_uuid);
			scrub_file_reset(scrub, dev->od_uuid, SF_INCONSISTENT);
			dirty = true;
			restored = true;
		} else if (sf->sf_status == SS_SCANNING) {
			sf->sf_status = SS_CRASHED;
			dirty = true;
		}

		if ((sf->sf_oi_count & (sf->sf_oi_count - 1)) != 0) {
			LCONSOLE_WARN("%s: invalid oi count %d, set it to %d\n",
				      osd_dev2name(dev), sf->sf_oi_count,
				      osd_oi_count);
			sf->sf_oi_count = osd_oi_count;
			dirty = true;
		}
	}

	if (sf->sf_pos_last_checkpoint != 0)
		scrub->os_pos_current = sf->sf_pos_last_checkpoint + 1;
	else
		scrub->os_pos_current = LDISKFS_FIRST_INO(sb) + 1;

	if (dirty) {
		rc = scrub_file_store(env, scrub);
		if (rc)
			GOTO(cleanup_obj, rc);
	}

	/* Initialize OI files. */
	rc = osd_oi_init(info, dev, restored);
	if (rc < 0)
		GOTO(cleanup_obj, rc);

	if (!dev->od_dt_dev.dd_rdonly)
		osd_initial_OI_scrub(info, dev);

	if (sf->sf_flags & SF_UPGRADE ||
	    !(sf->sf_internal_flags & SIF_NO_HANDLE_OLD_FID ||
	      sf->sf_success_count > 0)) {
		dev->od_igif_inoi = 0;
		dev->od_check_ff = dev->od_is_ost;
	} else {
		dev->od_igif_inoi = 1;
		dev->od_check_ff = 0;
	}

	if (sf->sf_flags & SF_INCONSISTENT)
		/* The 'od_igif_inoi' will be set under the
		 * following cases:
		 * 1) new created system, or
		 * 2) restored from file-level backup, or
		 * 3) the upgrading completed.
		 *
		 * The 'od_igif_inoi' may be cleared by OI scrub
		 * later if found that the system is upgrading. */
		dev->od_igif_inoi = 1;

	if (!dev->od_dt_dev.dd_rdonly &&
	    dev->od_scrub.os_scrub.os_auto_scrub_interval != AS_NEVER &&
	    ((sf->sf_status == SS_PAUSED) ||
	     (sf->sf_status == SS_CRASHED &&
	      sf->sf_flags & (SF_RECREATED | SF_INCONSISTENT |
			      SF_UPGRADE | SF_AUTO)) ||
	     (sf->sf_status == SS_INIT &&
	      sf->sf_flags & (SF_RECREATED | SF_INCONSISTENT |
			      SF_UPGRADE))))
		rc = osd_scrub_start(env, dev, SS_AUTO_FULL);

	if (rc != 0)
		GOTO(cleanup_oi, rc);

	/* it is possible that dcache entries may keep objects after they are
	 * deleted by OSD. While it looks safe this can cause object data to
	 * stay until umount causing failures in tests calculating free space,
	 * e.g. replay-ost-single. Since those dcache entries are not used
	 * anymore let's just free them after use here */
	shrink_dcache_sb(sb);

	RETURN(0);
cleanup_oi:
	osd_oi_fini(info, dev);
cleanup_obj:
	dt_object_put_nocache(env, scrub->os_obj);
	scrub->os_obj = NULL;

	return rc;
}

void osd_scrub_cleanup(const struct lu_env *env, struct osd_device *dev)
{
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;

	LASSERT(dev->od_otable_it == NULL);

	if (scrub->os_obj != NULL) {
		osd_scrub_stop(dev);
		dt_object_put_nocache(env, scrub->os_obj);
		scrub->os_obj = NULL;
	}
}

/* object table based iteration APIs */

static struct dt_it *osd_otable_it_init(const struct lu_env *env,
				       struct dt_object *dt, __u32 attr)
{
	enum dt_otable_it_flags flags = attr >> DT_OTABLE_IT_FLAGS_SHIFT;
	enum dt_otable_it_valid valid = attr & ~DT_OTABLE_IT_FLAGS_MASK;
	struct osd_device      *dev   = osd_dev(dt->do_lu.lo_dev);
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;
	struct osd_otable_it   *it;
	__u32			start = 0;
	int			rc;
	ENTRY;

	/* od_otable_mutex: prevent curcurrent init/fini */
	mutex_lock(&dev->od_otable_mutex);
	if (dev->od_otable_it != NULL)
		GOTO(out, it = ERR_PTR(-EALREADY));

	OBD_ALLOC_PTR(it);
	if (it == NULL)
		GOTO(out, it = ERR_PTR(-ENOMEM));

	dev->od_otable_it = it;
	it->ooi_dev = dev;
	it->ooi_cache.ooc_consumer_idx = -1;
	if (flags & DOIF_OUTUSED)
		it->ooi_used_outside = 1;

	if (flags & DOIF_RESET)
		start |= SS_RESET;

	if (valid & DOIV_ERROR_HANDLE) {
		if (flags & DOIF_FAILOUT)
			start |= SS_SET_FAILOUT;
		else
			start |= SS_CLEAR_FAILOUT;
	}

	if (valid & DOIV_DRYRUN) {
		if (flags & DOIF_DRYRUN)
			start |= SS_SET_DRYRUN;
		else
			start |= SS_CLEAR_DRYRUN;
	}

	rc = scrub_start(osd_scrub_main, scrub, dev, start & ~SS_AUTO_PARTIAL);
	if (rc == -EALREADY) {
		it->ooi_cache.ooc_pos_preload = scrub->os_pos_current;
	} else  if (rc < 0) {
		dev->od_otable_it = NULL;
		OBD_FREE_PTR(it);
		it = ERR_PTR(rc);
	} else {
		/* We have to start from the begining. */
		it->ooi_cache.ooc_pos_preload =
			LDISKFS_FIRST_INO(osd_sb(dev)) + 1;
	}

	GOTO(out, it);

out:
	mutex_unlock(&dev->od_otable_mutex);
	return (struct dt_it *)it;
}

static void osd_otable_it_fini(const struct lu_env *env, struct dt_it *di)
{
	struct osd_otable_it *it  = (struct osd_otable_it *)di;
	struct osd_device    *dev = it->ooi_dev;

	/* od_otable_mutex: prevent curcurrent init/fini */
	mutex_lock(&dev->od_otable_mutex);
	scrub_stop(&dev->od_scrub.os_scrub);
	LASSERT(dev->od_otable_it == it);

	dev->od_otable_it = NULL;
	mutex_unlock(&dev->od_otable_mutex);
	OBD_FREE_PTR(it);
}

static int osd_otable_it_get(const struct lu_env *env,
			     struct dt_it *di, const struct dt_key *key)
{
	return 0;
}

static void osd_otable_it_put(const struct lu_env *env, struct dt_it *di)
{
}

static inline int
osd_otable_it_wakeup(struct lustre_scrub *scrub, struct osd_otable_it *it)
{
	spin_lock(&scrub->os_lock);
	if (it->ooi_cache.ooc_pos_preload < scrub->os_pos_current ||
	    scrub->os_waiting || !scrub->os_running)
		it->ooi_waiting = 0;
	else
		it->ooi_waiting = 1;
	spin_unlock(&scrub->os_lock);

	return !it->ooi_waiting;
}

static int osd_otable_it_next(const struct lu_env *env, struct dt_it *di)
{
	struct osd_otable_it *it = (struct osd_otable_it *)di;
	struct osd_device *dev = it->ooi_dev;
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;
	struct osd_otable_cache *ooc = &it->ooi_cache;
	int rc;
	ENTRY;

	LASSERT(it->ooi_user_ready);

again:
	if (!scrub->os_running && !it->ooi_used_outside)
		RETURN(1);

	if (ooc->ooc_cached_items > 0) {
		ooc->ooc_cached_items--;
		ooc->ooc_consumer_idx = (ooc->ooc_consumer_idx + 1) &
					~OSD_OTABLE_IT_CACHE_MASK;
		RETURN(0);
	}

	if (it->ooi_all_cached) {
		wait_var_event(scrub, !scrub->os_running);
		RETURN(1);
	}

	if (scrub->os_waiting && osd_scrub_has_window(scrub, ooc)) {
		spin_lock(&scrub->os_lock);
		scrub->os_waiting = 0;
		wake_up_var(scrub);
		spin_unlock(&scrub->os_lock);
	}

	if (it->ooi_cache.ooc_pos_preload >= scrub->os_pos_current)
		wait_var_event(scrub, osd_otable_it_wakeup(scrub, it));

	if (!scrub->os_running && !it->ooi_used_outside)
		RETURN(1);

	rc = osd_otable_it_preload(env, it);
	if (rc >= 0)
		goto again;

	RETURN(rc);
}

static struct dt_key *osd_otable_it_key(const struct lu_env *env,
					const struct dt_it *di)
{
	return NULL;
}

static int osd_otable_it_key_size(const struct lu_env *env,
				  const struct dt_it *di)
{
	return sizeof(__u64);
}

static int osd_otable_it_rec(const struct lu_env *env, const struct dt_it *di,
			     struct dt_rec *rec, __u32 attr)
{
	struct osd_otable_it    *it  = (struct osd_otable_it *)di;
	struct osd_otable_cache *ooc = &it->ooi_cache;

	*(struct lu_fid *)rec = ooc->ooc_cache[ooc->ooc_consumer_idx].oic_fid;

	/* Filter out Invald FID already. */
	LASSERTF(fid_is_sane((struct lu_fid *)rec),
		 "Invalid FID "DFID", p_idx = %d, c_idx = %d\n",
		 PFID((struct lu_fid *)rec),
		 ooc->ooc_producer_idx, ooc->ooc_consumer_idx);

	return 0;
}

static __u64 osd_otable_it_store(const struct lu_env *env,
				 const struct dt_it *di)
{
	struct osd_otable_it    *it  = (struct osd_otable_it *)di;
	struct osd_otable_cache *ooc = &it->ooi_cache;
	__u64			 hash;

	if (it->ooi_user_ready && ooc->ooc_consumer_idx != -1)
		hash = ooc->ooc_cache[ooc->ooc_consumer_idx].oic_lid.oii_ino;
	else
		hash = ooc->ooc_pos_preload;
	return hash;
}

/**
 * Set the OSD layer iteration start position as the specified hash.
 */
static int osd_otable_it_load(const struct lu_env *env,
			      const struct dt_it *di, __u64 hash)
{
	struct osd_otable_it    *it    = (struct osd_otable_it *)di;
	struct osd_device       *dev   = it->ooi_dev;
	struct osd_otable_cache *ooc   = &it->ooi_cache;
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;
	struct osd_iit_param	*param = &it->ooi_iit_param;
	int			 rc;
	ENTRY;

	/* Forbid to set iteration position after iteration started. */
	if (it->ooi_user_ready)
		RETURN(-EPERM);

	LASSERT(!scrub->os_partial_scan);

	if (hash > OSD_OTABLE_MAX_HASH)
		hash = OSD_OTABLE_MAX_HASH;

	/* The hash is the last checkpoint position,
	 * we will start from the next one. */
	ooc->ooc_pos_preload = hash + 1;
	if (ooc->ooc_pos_preload <= LDISKFS_FIRST_INO(osd_sb(dev)))
		ooc->ooc_pos_preload = LDISKFS_FIRST_INO(osd_sb(dev)) + 1;

	it->ooi_user_ready = 1;
	if (!scrub->os_full_speed)
		wake_up_var(scrub);

	memset(param, 0, sizeof(*param));
	param->sb = osd_sb(dev);
	param->start = ooc->ooc_pos_preload;
	param->bg = (ooc->ooc_pos_preload - 1) /
		    LDISKFS_INODES_PER_GROUP(param->sb);
	param->offset = (ooc->ooc_pos_preload - 1) %
			LDISKFS_INODES_PER_GROUP(param->sb);
	param->gbase = 1 + param->bg * LDISKFS_INODES_PER_GROUP(param->sb);

	/* Unplug OSD layer iteration by the first next() call. */
	rc = osd_otable_it_next(env, (struct dt_it *)it);

	RETURN(rc);
}

static int osd_otable_it_key_rec(const struct lu_env *env,
				 const struct dt_it *di, void *key_rec)
{
	return 0;
}

const struct dt_index_operations osd_otable_ops = {
	.dio_it = {
		.init     = osd_otable_it_init,
		.fini     = osd_otable_it_fini,
		.get      = osd_otable_it_get,
		.put	  = osd_otable_it_put,
		.next     = osd_otable_it_next,
		.key	  = osd_otable_it_key,
		.key_size = osd_otable_it_key_size,
		.rec      = osd_otable_it_rec,
		.store    = osd_otable_it_store,
		.load     = osd_otable_it_load,
		.key_rec  = osd_otable_it_key_rec,
	}
};

/* high priority inconsistent items list APIs */

#define SCRUB_BAD_OIMAP_DECAY_INTERVAL	60

int osd_oii_insert(struct osd_device *dev, const struct lu_fid *fid,
		   struct osd_inode_id *id, int insert)
{
	struct osd_inconsistent_item *oii;
	struct osd_scrub *oscrub = &dev->od_scrub;
	struct lustre_scrub *lscrub = &oscrub->os_scrub;
	int wakeup = 0;
	ENTRY;

	OBD_ALLOC_PTR(oii);
	if (unlikely(oii == NULL))
		RETURN(-ENOMEM);

	INIT_LIST_HEAD(&oii->oii_list);
	oii->oii_cache.oic_fid = *fid;
	oii->oii_cache.oic_lid = *id;
	oii->oii_cache.oic_dev = dev;
	oii->oii_insert = insert;

	spin_lock(&lscrub->os_lock);
	if (lscrub->os_partial_scan) {
		__u64 now = ktime_get_real_seconds();

		/* If there haven't been errors in a long time,
		 * decay old count until either the errors are
		 * gone or we reach the current interval. */
		while (unlikely(oscrub->os_bad_oimap_count > 0 &&
				oscrub->os_bad_oimap_time +
				SCRUB_BAD_OIMAP_DECAY_INTERVAL < now)) {
			oscrub->os_bad_oimap_count >>= 1;
			oscrub->os_bad_oimap_time +=
				SCRUB_BAD_OIMAP_DECAY_INTERVAL;
		}

		oscrub->os_bad_oimap_time = now;
		if (++oscrub->os_bad_oimap_count >
		    dev->od_full_scrub_threshold_rate)
			lscrub->os_full_scrub = 1;
	}

	if (!lscrub->os_running) {
		spin_unlock(&lscrub->os_lock);
		OBD_FREE_PTR(oii);
		RETURN(-EAGAIN);
	}

	if (list_empty(&lscrub->os_inconsistent_items))
		wakeup = 1;
	list_add_tail(&oii->oii_list, &lscrub->os_inconsistent_items);
	spin_unlock(&lscrub->os_lock);

	if (wakeup)
		wake_up_var(lscrub);

	RETURN(0);
}

int osd_oii_lookup(struct osd_device *dev, const struct lu_fid *fid,
		   struct osd_inode_id *id)
{
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;
	struct osd_inconsistent_item *oii;
	ENTRY;

	spin_lock(&scrub->os_lock);
	list_for_each_entry(oii, &scrub->os_inconsistent_items, oii_list) {
		if (lu_fid_eq(fid, &oii->oii_cache.oic_fid)) {
			*id = oii->oii_cache.oic_lid;
			spin_unlock(&scrub->os_lock);
			RETURN(0);
		}
	}
	spin_unlock(&scrub->os_lock);

	RETURN(-ENOENT);
}

void osd_scrub_dump(struct seq_file *m, struct osd_device *dev)
{
	struct osd_scrub *scrub = &dev->od_scrub;

	scrub_dump(m, &scrub->os_scrub);
	seq_printf(m, "lf_scanned: %llu\n"
		   "lf_%s: %llu\n"
		   "lf_failed: %llu\n",
		   scrub->os_lf_scanned,
		   scrub->os_scrub.os_file.sf_param & SP_DRYRUN ?
			"inconsistent" : "repaired",
		   scrub->os_lf_repaired,
		   scrub->os_lf_failed);
}

typedef int (*scan_dir_helper_t)(const struct lu_env *env,
				 struct osd_device *dev, struct inode *dir,
				 struct osd_it_ea *oie);

static int osd_scan_dir(const struct lu_env *env, struct osd_device *dev,
			struct inode *inode, scan_dir_helper_t cb)
{
	struct osd_it_ea *oie;
	int rc;

	ENTRY;

	oie = osd_it_dir_init(env, dev, inode, LUDA_TYPE);
	if (IS_ERR(oie))
		RETURN(PTR_ERR(oie));

	oie->oie_file->f_pos = 0;
	rc = osd_ldiskfs_it_fill(env, (struct dt_it *)oie);
	if (rc > 0)
		rc = -ENODATA;
	if (rc)
		GOTO(out, rc);

	while (oie->oie_it_dirent <= oie->oie_rd_dirent) {
		if (!name_is_dot_or_dotdot(oie->oie_dirent->oied_name,
					   oie->oie_dirent->oied_namelen))
			cb(env, dev, inode, oie);

		oie->oie_dirent = (void *)oie->oie_dirent +
				cfs_size_round(sizeof(struct osd_it_ea_dirent) +
				oie->oie_dirent->oied_namelen);

		oie->oie_it_dirent++;
		if (oie->oie_it_dirent <= oie->oie_rd_dirent)
			continue;

		if (oie->oie_file->f_pos ==
		    ldiskfs_get_htree_eof(oie->oie_file))
			break;

		rc = osd_ldiskfs_it_fill(env, (struct dt_it *)oie);
		if (rc) {
			if (rc > 0)
				rc = 0;
			break;
		}
	}

out:
	osd_it_dir_fini(env, oie, inode);
	RETURN(rc);
}

static int osd_remove_ml_file(struct osd_thread_info *info,
			      struct osd_device *dev, struct inode *dir,
			      struct inode *inode, struct osd_it_ea *oie)
{
	handle_t *th;
	struct lustre_scrub *scrub = &dev->od_scrub.os_scrub;
	struct dentry dentry;
	int rc;

	ENTRY;

	if (scrub->os_file.sf_param & SP_DRYRUN)
		RETURN(0);

	th = osd_journal_start_sb(osd_sb(dev), LDISKFS_HT_MISC,
				  osd_dto_credits_noquota[DTO_INDEX_DELETE] +
				  osd_dto_credits_noquota[DTO_ATTR_SET_BASE]);
	if (IS_ERR(th))
		RETURN(PTR_ERR(th));

	/* Should be created by the VFS layer */
	dentry.d_inode = dir;
	dentry.d_sb = dir->i_sb;
	rc = osd_obj_del_entry(info, dev, &dentry, oie->oie_dirent->oied_name,
			       oie->oie_dirent->oied_namelen, th);
	drop_nlink(inode);
	mark_inode_dirty(inode);
	ldiskfs_journal_stop(th);
	RETURN(rc);
}

static int osd_scan_ml_file(const struct lu_env *env, struct osd_device *dev,
			    struct inode *dir, struct osd_it_ea *oie)
{
	struct osd_thread_info *info = osd_oti_get(env);
	struct osd_inode_id id;
	struct inode *inode;
	struct osd_obj_seq *oseq;
	struct ost_id *ostid = &info->oti_ostid;
	struct lu_fid *fid = &oie->oie_dirent->oied_fid;
	char name[32];
	int dirn, rc = 0;

	ENTRY;

	osd_id_gen(&id, oie->oie_dirent->oied_ino, OSD_OII_NOGEN);

	if (!fid_is_sane(fid))
		inode = osd_iget_fid(info, dev, &id, fid);
	else
		inode = osd_iget(info, dev, &id);

	if (IS_ERR(inode))
		RETURN(PTR_ERR(inode));

	fid_to_ostid(fid, ostid);
	oseq = osd_seq_load(info, dev, ostid_seq(ostid));
	if (IS_ERR(oseq))
		RETURN(PTR_ERR(oseq));

	dirn = ostid_id(ostid) & (oseq->oos_subdir_count - 1);
	LASSERT(oseq->oos_dirs[dirn] != NULL);

	osd_oid_name(name, sizeof(name), fid, ostid_id(ostid));
	if (((strlen(oseq->oos_root->d_name.name) !=
	      info->oti_seq_dirent->oied_namelen) ||
	     strncmp(oseq->oos_root->d_name.name,
		     info->oti_seq_dirent->oied_name,
		     info->oti_seq_dirent->oied_namelen) != 0) ||
	    ((strlen(oseq->oos_dirs[dirn]->d_name.name) !=
	      info->oti_dir_dirent->oied_namelen) ||
	     strncmp(oseq->oos_dirs[dirn]->d_name.name,
		     info->oti_dir_dirent->oied_name,
		     info->oti_dir_dirent->oied_namelen) != 0) ||
	    ((strlen(name) != oie->oie_dirent->oied_namelen) ||
	     strncmp(oie->oie_dirent->oied_name, name,
		     oie->oie_dirent->oied_namelen) != 0)) {
		CDEBUG(D_LFSCK, "%s: the file O/%s/%s/%s is corrupted\n",
		       osd_name(dev), info->oti_seq_dirent->oied_name,
		       info->oti_dir_dirent->oied_name,
		       oie->oie_dirent->oied_name);

		rc = osd_remove_ml_file(info, dev, dir, inode, oie);
	}

	iput(inode);
	RETURN(rc);
}

static int osd_scan_ml_file_dir(const struct lu_env *env,
				struct osd_device *dev, struct inode *dir,
				struct osd_it_ea *oie)
{
	struct osd_thread_info *info = osd_oti_get(env);
	struct inode *inode;
	struct osd_inode_id id;
	int rc;

	ENTRY;

	osd_id_gen(&id, oie->oie_dirent->oied_ino, OSD_OII_NOGEN);
	inode = osd_iget(info, dev, &id);
	if (IS_ERR(inode))
		RETURN(PTR_ERR(inode));

	if (!S_ISDIR(inode->i_mode))
		GOTO(out, rc = 0);

	info->oti_dir_dirent = oie->oie_dirent;
	rc = osd_scan_dir(env, dev, inode, osd_scan_ml_file);
	info->oti_dir_dirent = NULL;

out:
	iput(inode);
	RETURN(rc);
}

static int osd_scan_ml_file_seq(const struct lu_env *env,
				struct osd_device *dev, struct inode *dir,
				struct osd_it_ea *oie)
{
	struct osd_thread_info *info = osd_oti_get(env);
	struct inode *inode;
	struct osd_inode_id id;
	int rc;

	ENTRY;

	osd_id_gen(&id, oie->oie_dirent->oied_ino, OSD_OII_NOGEN);
	inode = osd_iget(info, dev, &id);
	if (IS_ERR(inode))
		RETURN(PTR_ERR(inode));

	if (!S_ISDIR(inode->i_mode))
		GOTO(out, rc = 0);

	info->oti_seq_dirent = oie->oie_dirent;
	rc = osd_scan_dir(env, dev, inode, osd_scan_ml_file_dir);
	info->oti_seq_dirent = NULL;

out:
	iput(inode);
	RETURN(rc);
}

static int osd_scan_ml_file_main(const struct lu_env *env,
				 struct osd_device *dev)
{
	return osd_scan_dir(env, dev, dev->od_ost_map->om_root->d_inode,
			    osd_scan_ml_file_seq);
}