UBI: Add basic UBI support to U-Boot (Part 5/8)

This patch adds basic UBI (Unsorted Block Image) support to U-Boot.
It's based on the Linux UBI version and basically has a "OS"
translation wrapper that defines most Linux specific calls
(spin_lock() etc.) into no-ops. Some source code parts have been
uncommented by "#ifdef UBI_LINUX". This makes it easier to compare
this version with the Linux version and simplifies future UBI
ports/bug-fixes from the Linux version.

Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Stefan Roese <sr@denx.de>

drivers/mtd/ubi/upd.c

@@ -0,0 +1,441 @@
+/*
+ * Copyright (c) International Business Machines Corp., 2006
+ * Copyright (c) Nokia Corporation, 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * 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 for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Author: Artem Bityutskiy (Битюцкий Артём)
+ *
+ * Jan 2007: Alexander Schmidt, hacked per-volume update.
+ */
+
+/*
+ * This file contains implementation of the volume update and atomic LEB change
+ * functionality.
+ *
+ * The update operation is based on the per-volume update marker which is
+ * stored in the volume table. The update marker is set before the update
+ * starts, and removed after the update has been finished. So if the update was
+ * interrupted by an unclean re-boot or due to some other reasons, the update
+ * marker stays on the flash media and UBI finds it when it attaches the MTD
+ * device next time. If the update marker is set for a volume, the volume is
+ * treated as damaged and most I/O operations are prohibited. Only a new update
+ * operation is allowed.
+ *
+ * Note, in general it is possible to implement the update operation as a
+ * transaction with a roll-back capability.
+ */
+
+#ifdef UBI_LINUX
+#include <linux/err.h>
+#include <asm/uaccess.h>
+#include <asm/div64.h>
+#endif
+
+#include <ubi_uboot.h>
+#include "ubi.h"
+
+/**
+ * set_update_marker - set update marker.
+ * @ubi: UBI device description object
+ * @vol: volume description object
+ *
+ * This function sets the update marker flag for volume @vol. Returns zero
+ * in case of success and a negative error code in case of failure.
+ */
+static int set_update_marker(struct ubi_device *ubi, struct ubi_volume *vol)
+{
+	int err;
+	struct ubi_vtbl_record vtbl_rec;
+
+	dbg_msg("set update marker for volume %d", vol->vol_id);
+
+	if (vol->upd_marker) {
+		ubi_assert(ubi->vtbl[vol->vol_id].upd_marker);
+		dbg_msg("already set");
+		return 0;
+	}
+
+	memcpy(&vtbl_rec, &ubi->vtbl[vol->vol_id],
+	       sizeof(struct ubi_vtbl_record));
+	vtbl_rec.upd_marker = 1;
+
+	mutex_lock(&ubi->volumes_mutex);
+	err = ubi_change_vtbl_record(ubi, vol->vol_id, &vtbl_rec);
+	mutex_unlock(&ubi->volumes_mutex);
+	vol->upd_marker = 1;
+	return err;
+}
+
+/**
+ * clear_update_marker - clear update marker.
+ * @ubi: UBI device description object
+ * @vol: volume description object
+ * @bytes: new data size in bytes
+ *
+ * This function clears the update marker for volume @vol, sets new volume
+ * data size and clears the "corrupted" flag (static volumes only). Returns
+ * zero in case of success and a negative error code in case of failure.
+ */
+static int clear_update_marker(struct ubi_device *ubi, struct ubi_volume *vol,
+			       long long bytes)
+{
+	int err;
+	uint64_t tmp;
+	struct ubi_vtbl_record vtbl_rec;
+
+	dbg_msg("clear update marker for volume %d", vol->vol_id);
+
+	memcpy(&vtbl_rec, &ubi->vtbl[vol->vol_id],
+	       sizeof(struct ubi_vtbl_record));
+	ubi_assert(vol->upd_marker && vtbl_rec.upd_marker);
+	vtbl_rec.upd_marker = 0;
+
+	if (vol->vol_type == UBI_STATIC_VOLUME) {
+		vol->corrupted = 0;
+		vol->used_bytes = tmp = bytes;
+		vol->last_eb_bytes = do_div(tmp, vol->usable_leb_size);
+		vol->used_ebs = tmp;
+		if (vol->last_eb_bytes)
+			vol->used_ebs += 1;
+		else
+			vol->last_eb_bytes = vol->usable_leb_size;
+	}
+
+	mutex_lock(&ubi->volumes_mutex);
+	err = ubi_change_vtbl_record(ubi, vol->vol_id, &vtbl_rec);
+	mutex_unlock(&ubi->volumes_mutex);
+	vol->upd_marker = 0;
+	return err;
+}
+
+/**
+ * ubi_start_update - start volume update.
+ * @ubi: UBI device description object
+ * @vol: volume description object
+ * @bytes: update bytes
+ *
+ * This function starts volume update operation. If @bytes is zero, the volume
+ * is just wiped out. Returns zero in case of success and a negative error code
+ * in case of failure.
+ */
+int ubi_start_update(struct ubi_device *ubi, struct ubi_volume *vol,
+		     long long bytes)
+{
+	int i, err;
+	uint64_t tmp;
+
+	dbg_msg("start update of volume %d, %llu bytes", vol->vol_id, bytes);
+	ubi_assert(!vol->updating && !vol->changing_leb);
+	vol->updating = 1;
+
+	err = set_update_marker(ubi, vol);
+	if (err)
+		return err;
+
+	/* Before updating - wipe out the volume */
+	for (i = 0; i < vol->reserved_pebs; i++) {
+		err = ubi_eba_unmap_leb(ubi, vol, i);
+		if (err)
+			return err;
+	}
+
+	if (bytes == 0) {
+		err = clear_update_marker(ubi, vol, 0);
+		if (err)
+			return err;
+		err = ubi_wl_flush(ubi);
+		if (!err)
+			vol->updating = 0;
+	}
+
+	vol->upd_buf = vmalloc(ubi->leb_size);
+	if (!vol->upd_buf)
+		return -ENOMEM;
+
+	tmp = bytes;
+	vol->upd_ebs = !!do_div(tmp, vol->usable_leb_size);
+	vol->upd_ebs += tmp;
+	vol->upd_bytes = bytes;
+	vol->upd_received = 0;
+	return 0;
+}
+
+/**
+ * ubi_start_leb_change - start atomic LEB change.
+ * @ubi: UBI device description object
+ * @vol: volume description object
+ * @req: operation request
+ *
+ * This function starts atomic LEB change operation. Returns zero in case of
+ * success and a negative error code in case of failure.
+ */
+int ubi_start_leb_change(struct ubi_device *ubi, struct ubi_volume *vol,
+			 const struct ubi_leb_change_req *req)
+{
+	ubi_assert(!vol->updating && !vol->changing_leb);
+
+	dbg_msg("start changing LEB %d:%d, %u bytes",
+		vol->vol_id, req->lnum, req->bytes);
+	if (req->bytes == 0)
+		return ubi_eba_atomic_leb_change(ubi, vol, req->lnum, NULL, 0,
+						 req->dtype);
+
+	vol->upd_bytes = req->bytes;
+	vol->upd_received = 0;
+	vol->changing_leb = 1;
+	vol->ch_lnum = req->lnum;
+	vol->ch_dtype = req->dtype;
+
+	vol->upd_buf = vmalloc(req->bytes);
+	if (!vol->upd_buf)
+		return -ENOMEM;
+
+	return 0;
+}
+
+/**
+ * write_leb - write update data.
+ * @ubi: UBI device description object
+ * @vol: volume description object
+ * @lnum: logical eraseblock number
+ * @buf: data to write
+ * @len: data size
+ * @used_ebs: how many logical eraseblocks will this volume contain (static
+ * volumes only)
+ *
+ * This function writes update data to corresponding logical eraseblock. In
+ * case of dynamic volume, this function checks if the data contains 0xFF bytes
+ * at the end. If yes, the 0xFF bytes are cut and not written. So if the whole
+ * buffer contains only 0xFF bytes, the LEB is left unmapped.
+ *
+ * The reason why we skip the trailing 0xFF bytes in case of dynamic volume is
+ * that we want to make sure that more data may be appended to the logical
+ * eraseblock in future. Indeed, writing 0xFF bytes may have side effects and
+ * this PEB won't be writable anymore. So if one writes the file-system image
+ * to the UBI volume where 0xFFs mean free space - UBI makes sure this free
+ * space is writable after the update.
+ *
+ * We do not do this for static volumes because they are read-only. But this
+ * also cannot be done because we have to store per-LEB CRC and the correct
+ * data length.
+ *
+ * This function returns zero in case of success and a negative error code in
+ * case of failure.
+ */
+static int write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
+		     void *buf, int len, int used_ebs)
+{
+	int err;
+
+	if (vol->vol_type == UBI_DYNAMIC_VOLUME) {
+		int l = ALIGN(len, ubi->min_io_size);
+
+		memset(buf + len, 0xFF, l - len);
+		len = ubi_calc_data_len(ubi, buf, l);
+		if (len == 0) {
+			dbg_msg("all %d bytes contain 0xFF - skip", len);
+			return 0;
+		}
+
+		err = ubi_eba_write_leb(ubi, vol, lnum, buf, 0, len, UBI_UNKNOWN);
+	} else {
+		/*
+		 * When writing static volume, and this is the last logical
+		 * eraseblock, the length (@len) does not have to be aligned to
+		 * the minimal flash I/O unit. The 'ubi_eba_write_leb_st()'
+		 * function accepts exact (unaligned) length and stores it in
+		 * the VID header. And it takes care of proper alignment by
+		 * padding the buffer. Here we just make sure the padding will
+		 * contain zeros, not random trash.
+		 */
+		memset(buf + len, 0, vol->usable_leb_size - len);
+		err = ubi_eba_write_leb_st(ubi, vol, lnum, buf, len,
+					   UBI_UNKNOWN, used_ebs);
+	}
+
+	return err;
+}
+
+/**
+ * ubi_more_update_data - write more update data.
+ * @vol: volume description object
+ * @buf: write data (user-space memory buffer)
+ * @count: how much bytes to write
+ *
+ * This function writes more data to the volume which is being updated. It may
+ * be called arbitrary number of times until all the update data arriveis. This
+ * function returns %0 in case of success, number of bytes written during the
+ * last call if the whole volume update has been successfully finished, and a
+ * negative error code in case of failure.
+ */
+int ubi_more_update_data(struct ubi_device *ubi, struct ubi_volume *vol,
+			 const void __user *buf, int count)
+{
+	uint64_t tmp;
+	int lnum, offs, err = 0, len, to_write = count;
+
+	dbg_msg("write %d of %lld bytes, %lld already passed",
+		count, vol->upd_bytes, vol->upd_received);
+
+	if (ubi->ro_mode)
+		return -EROFS;
+
+	tmp = vol->upd_received;
+	offs = do_div(tmp, vol->usable_leb_size);
+	lnum = tmp;
+
+	if (vol->upd_received + count > vol->upd_bytes)
+		to_write = count = vol->upd_bytes - vol->upd_received;
+
+	/*
+	 * When updating volumes, we accumulate whole logical eraseblock of
+	 * data and write it at once.
+	 */
+	if (offs != 0) {
+		/*
+		 * This is a write to the middle of the logical eraseblock. We
+		 * copy the data to our update buffer and wait for more data or
+		 * flush it if the whole eraseblock is written or the update
+		 * is finished.
+		 */
+
+		len = vol->usable_leb_size - offs;
+		if (len > count)
+			len = count;
+
+		err = copy_from_user(vol->upd_buf + offs, buf, len);
+		if (err)
+			return -EFAULT;
+
+		if (offs + len == vol->usable_leb_size ||
+		    vol->upd_received + len == vol->upd_bytes) {
+			int flush_len = offs + len;
+
+			/*
+			 * OK, we gathered either the whole eraseblock or this
+			 * is the last chunk, it's time to flush the buffer.
+			 */
+			ubi_assert(flush_len <= vol->usable_leb_size);
+			err = write_leb(ubi, vol, lnum, vol->upd_buf, flush_len,
+					vol->upd_ebs);
+			if (err)
+				return err;
+		}
+
+		vol->upd_received += len;
+		count -= len;
+		buf += len;
+		lnum += 1;
+	}
+
+	/*
+	 * If we've got more to write, let's continue. At this point we know we
+	 * are starting from the beginning of an eraseblock.
+	 */
+	while (count) {
+		if (count > vol->usable_leb_size)
+			len = vol->usable_leb_size;
+		else
+			len = count;
+
+		err = copy_from_user(vol->upd_buf, buf, len);
+		if (err)
+			return -EFAULT;
+
+		if (len == vol->usable_leb_size ||
+		    vol->upd_received + len == vol->upd_bytes) {
+			err = write_leb(ubi, vol, lnum, vol->upd_buf,
+					len, vol->upd_ebs);
+			if (err)
+				break;
+		}
+
+		vol->upd_received += len;
+		count -= len;
+		lnum += 1;
+		buf += len;
+	}
+
+	ubi_assert(vol->upd_received <= vol->upd_bytes);
+	if (vol->upd_received == vol->upd_bytes) {
+		/* The update is finished, clear the update marker */
+		err = clear_update_marker(ubi, vol, vol->upd_bytes);
+		if (err)
+			return err;
+		err = ubi_wl_flush(ubi);
+		if (err == 0) {
+			vol->updating = 0;
+			err = to_write;
+			vfree(vol->upd_buf);
+		}
+	}
+
+	return err;
+}
+
+/**
+ * ubi_more_leb_change_data - accept more data for atomic LEB change.
+ * @vol: volume description object
+ * @buf: write data (user-space memory buffer)
+ * @count: how much bytes to write
+ *
+ * This function accepts more data to the volume which is being under the
+ * "atomic LEB change" operation. It may be called arbitrary number of times
+ * until all data arrives. This function returns %0 in case of success, number
+ * of bytes written during the last call if the whole "atomic LEB change"
+ * operation has been successfully finished, and a negative error code in case
+ * of failure.
+ */
+int ubi_more_leb_change_data(struct ubi_device *ubi, struct ubi_volume *vol,
+			     const void __user *buf, int count)
+{
+	int err;
+
+	dbg_msg("write %d of %lld bytes, %lld already passed",
+		count, vol->upd_bytes, vol->upd_received);
+
+	if (ubi->ro_mode)
+		return -EROFS;
+
+	if (vol->upd_received + count > vol->upd_bytes)
+		count = vol->upd_bytes - vol->upd_received;
+
+	err = copy_from_user(vol->upd_buf + vol->upd_received, buf, count);
+	if (err)
+		return -EFAULT;
+
+	vol->upd_received += count;
+
+	if (vol->upd_received == vol->upd_bytes) {
+		int len = ALIGN((int)vol->upd_bytes, ubi->min_io_size);
+
+		memset(vol->upd_buf + vol->upd_bytes, 0xFF, len - vol->upd_bytes);
+		len = ubi_calc_data_len(ubi, vol->upd_buf, len);
+		err = ubi_eba_atomic_leb_change(ubi, vol, vol->ch_lnum,
+						vol->upd_buf, len, UBI_UNKNOWN);
+		if (err)
+			return err;
+	}
+
+	ubi_assert(vol->upd_received <= vol->upd_bytes);
+	if (vol->upd_received == vol->upd_bytes) {
+		vol->changing_leb = 0;
+		err = count;
+		vfree(vol->upd_buf);
+	}
+
+	return err;
+}

drivers/mtd/ubi/vmt.c

@@ -0,0 +1,862 @@
+/*
+ * Copyright (c) International Business Machines Corp., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation;  either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * 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 for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Author: Artem Bityutskiy (Битюцкий Артём)
+ */
+
+/*
+ * This file contains implementation of volume creation, deletion, updating and
+ * resizing.
+ */
+
+#ifdef UBI_LINUX
+#include <linux/err.h>
+#include <asm/div64.h>
+#endif
+
+#include <ubi_uboot.h>
+#include "ubi.h"
+
+#ifdef CONFIG_MTD_UBI_DEBUG_PARANOID
+static void paranoid_check_volumes(struct ubi_device *ubi);
+#else
+#define paranoid_check_volumes(ubi)
+#endif
+
+#ifdef UBI_LINUX
+static ssize_t vol_attribute_show(struct device *dev,
+				  struct device_attribute *attr, char *buf);
+
+/* Device attributes corresponding to files in '/<sysfs>/class/ubi/ubiX_Y' */
+static struct device_attribute attr_vol_reserved_ebs =
+	__ATTR(reserved_ebs, S_IRUGO, vol_attribute_show, NULL);
+static struct device_attribute attr_vol_type =
+	__ATTR(type, S_IRUGO, vol_attribute_show, NULL);
+static struct device_attribute attr_vol_name =
+	__ATTR(name, S_IRUGO, vol_attribute_show, NULL);
+static struct device_attribute attr_vol_corrupted =
+	__ATTR(corrupted, S_IRUGO, vol_attribute_show, NULL);
+static struct device_attribute attr_vol_alignment =
+	__ATTR(alignment, S_IRUGO, vol_attribute_show, NULL);
+static struct device_attribute attr_vol_usable_eb_size =
+	__ATTR(usable_eb_size, S_IRUGO, vol_attribute_show, NULL);
+static struct device_attribute attr_vol_data_bytes =
+	__ATTR(data_bytes, S_IRUGO, vol_attribute_show, NULL);
+static struct device_attribute attr_vol_upd_marker =
+	__ATTR(upd_marker, S_IRUGO, vol_attribute_show, NULL);
+
+/*
+ * "Show" method for files in '/<sysfs>/class/ubi/ubiX_Y/'.
+ *
+ * Consider a situation:
+ * A. process 1 opens a sysfs file related to volume Y, say
+ *    /<sysfs>/class/ubi/ubiX_Y/reserved_ebs;
+ * B. process 2 removes volume Y;
+ * C. process 1 starts reading the /<sysfs>/class/ubi/ubiX_Y/reserved_ebs file;
+ *
+ * In this situation, this function will return %-ENODEV because it will find
+ * out that the volume was removed from the @ubi->volumes array.
+ */
+static ssize_t vol_attribute_show(struct device *dev,
+				  struct device_attribute *attr, char *buf)
+{
+	int ret;
+	struct ubi_volume *vol = container_of(dev, struct ubi_volume, dev);
+	struct ubi_device *ubi;
+
+	ubi = ubi_get_device(vol->ubi->ubi_num);
+	if (!ubi)
+		return -ENODEV;
+
+	spin_lock(&ubi->volumes_lock);
+	if (!ubi->volumes[vol->vol_id]) {
+		spin_unlock(&ubi->volumes_lock);
+		ubi_put_device(ubi);
+		return -ENODEV;
+	}
+	/* Take a reference to prevent volume removal */
+	vol->ref_count += 1;
+	spin_unlock(&ubi->volumes_lock);
+
+	if (attr == &attr_vol_reserved_ebs)
+		ret = sprintf(buf, "%d\n", vol->reserved_pebs);
+	else if (attr == &attr_vol_type) {
+		const char *tp;
+
+		if (vol->vol_type == UBI_DYNAMIC_VOLUME)
+			tp = "dynamic";
+		else
+			tp = "static";
+		ret = sprintf(buf, "%s\n", tp);
+	} else if (attr == &attr_vol_name)
+		ret = sprintf(buf, "%s\n", vol->name);
+	else if (attr == &attr_vol_corrupted)
+		ret = sprintf(buf, "%d\n", vol->corrupted);
+	else if (attr == &attr_vol_alignment)
+		ret = sprintf(buf, "%d\n", vol->alignment);
+	else if (attr == &attr_vol_usable_eb_size)
+		ret = sprintf(buf, "%d\n", vol->usable_leb_size);
+	else if (attr == &attr_vol_data_bytes)
+		ret = sprintf(buf, "%lld\n", vol->used_bytes);
+	else if (attr == &attr_vol_upd_marker)
+		ret = sprintf(buf, "%d\n", vol->upd_marker);
+	else
+		/* This must be a bug */
+		ret = -EINVAL;
+
+	/* We've done the operation, drop volume and UBI device references */
+	spin_lock(&ubi->volumes_lock);
+	vol->ref_count -= 1;
+	ubi_assert(vol->ref_count >= 0);
+	spin_unlock(&ubi->volumes_lock);
+	ubi_put_device(ubi);
+	return ret;
+}
+#endif
+
+/* Release method for volume devices */
+static void vol_release(struct device *dev)
+{
+	struct ubi_volume *vol = container_of(dev, struct ubi_volume, dev);
+
+	kfree(vol);
+}
+
+#ifdef UBI_LINUX
+/**
+ * volume_sysfs_init - initialize sysfs for new volume.
+ * @ubi: UBI device description object
+ * @vol: volume description object
+ *
+ * This function returns zero in case of success and a negative error code in
+ * case of failure.
+ *
+ * Note, this function does not free allocated resources in case of failure -
+ * the caller does it. This is because this would cause release() here and the
+ * caller would oops.
+ */
+static int volume_sysfs_init(struct ubi_device *ubi, struct ubi_volume *vol)
+{
+	int err;
+
+	err = device_create_file(&vol->dev, &attr_vol_reserved_ebs);
+	if (err)
+		return err;
+	err = device_create_file(&vol->dev, &attr_vol_type);
+	if (err)
+		return err;
+	err = device_create_file(&vol->dev, &attr_vol_name);
+	if (err)
+		return err;
+	err = device_create_file(&vol->dev, &attr_vol_corrupted);
+	if (err)
+		return err;
+	err = device_create_file(&vol->dev, &attr_vol_alignment);
+	if (err)
+		return err;
+	err = device_create_file(&vol->dev, &attr_vol_usable_eb_size);
+	if (err)
+		return err;
+	err = device_create_file(&vol->dev, &attr_vol_data_bytes);
+	if (err)
+		return err;
+	err = device_create_file(&vol->dev, &attr_vol_upd_marker);
+	return err;
+}
+
+/**
+ * volume_sysfs_close - close sysfs for a volume.
+ * @vol: volume description object
+ */
+static void volume_sysfs_close(struct ubi_volume *vol)
+{
+	device_remove_file(&vol->dev, &attr_vol_upd_marker);
+	device_remove_file(&vol->dev, &attr_vol_data_bytes);
+	device_remove_file(&vol->dev, &attr_vol_usable_eb_size);
+	device_remove_file(&vol->dev, &attr_vol_alignment);
+	device_remove_file(&vol->dev, &attr_vol_corrupted);
+	device_remove_file(&vol->dev, &attr_vol_name);
+	device_remove_file(&vol->dev, &attr_vol_type);
+	device_remove_file(&vol->dev, &attr_vol_reserved_ebs);
+	device_unregister(&vol->dev);
+}
+#endif
+
+/**
+ * ubi_create_volume - create volume.
+ * @ubi: UBI device description object
+ * @req: volume creation request
+ *
+ * This function creates volume described by @req. If @req->vol_id id
+ * %UBI_VOL_NUM_AUTO, this function automatically assign ID to the new volume
+ * and saves it in @req->vol_id. Returns zero in case of success and a negative
+ * error code in case of failure. Note, the caller has to have the
+ * @ubi->volumes_mutex locked.
+ */
+int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req)
+{
+	int i, err, vol_id = req->vol_id, dont_free = 0;
+	struct ubi_volume *vol;
+	struct ubi_vtbl_record vtbl_rec;
+	uint64_t bytes;
+	dev_t dev;
+
+	if (ubi->ro_mode)
+		return -EROFS;
+
+	vol = kzalloc(sizeof(struct ubi_volume), GFP_KERNEL);
+	if (!vol)
+		return -ENOMEM;
+
+	spin_lock(&ubi->volumes_lock);
+	if (vol_id == UBI_VOL_NUM_AUTO) {
+		/* Find unused volume ID */
+		dbg_msg("search for vacant volume ID");
+		for (i = 0; i < ubi->vtbl_slots; i++)
+			if (!ubi->volumes[i]) {
+				vol_id = i;
+				break;
+			}
+
+		if (vol_id == UBI_VOL_NUM_AUTO) {
+			dbg_err("out of volume IDs");
+			err = -ENFILE;
+			goto out_unlock;
+		}
+		req->vol_id = vol_id;
+	}
+
+	dbg_msg("volume ID %d, %llu bytes, type %d, name %s",
+		vol_id, (unsigned long long)req->bytes,
+		(int)req->vol_type, req->name);
+
+	/* Ensure that this volume does not exist */
+	err = -EEXIST;
+	if (ubi->volumes[vol_id]) {
+		dbg_err("volume %d already exists", vol_id);
+		goto out_unlock;
+	}
+
+	/* Ensure that the name is unique */
+	for (i = 0; i < ubi->vtbl_slots; i++)
+		if (ubi->volumes[i] &&
+		    ubi->volumes[i]->name_len == req->name_len &&
+		    !strcmp(ubi->volumes[i]->name, req->name)) {
+			dbg_err("volume \"%s\" exists (ID %d)", req->name, i);
+			goto out_unlock;
+		}
+
+        /* Calculate how many eraseblocks are requested */
+	vol->usable_leb_size = ubi->leb_size - ubi->leb_size % req->alignment;
+	bytes = req->bytes;
+	if (do_div(bytes, vol->usable_leb_size))
+		vol->reserved_pebs = 1;
+	vol->reserved_pebs += bytes;
+
+	/* Reserve physical eraseblocks */
+	if (vol->reserved_pebs > ubi->avail_pebs) {
+		dbg_err("not enough PEBs, only %d available", ubi->avail_pebs);
+		err = -ENOSPC;
+		goto out_unlock;
+	}
+	ubi->avail_pebs -= vol->reserved_pebs;
+	ubi->rsvd_pebs += vol->reserved_pebs;
+	spin_unlock(&ubi->volumes_lock);
+
+	vol->vol_id    = vol_id;
+	vol->alignment = req->alignment;
+	vol->data_pad  = ubi->leb_size % vol->alignment;
+	vol->vol_type  = req->vol_type;
+	vol->name_len  = req->name_len;
+	memcpy(vol->name, req->name, vol->name_len + 1);
+	vol->ubi = ubi;
+
+	/*
+	 * Finish all pending erases because there may be some LEBs belonging
+	 * to the same volume ID.
+	 */
+	err = ubi_wl_flush(ubi);
+	if (err)
+		goto out_acc;
+
+	vol->eba_tbl = kmalloc(vol->reserved_pebs * sizeof(int), GFP_KERNEL);
+	if (!vol->eba_tbl) {
+		err = -ENOMEM;
+		goto out_acc;
+	}
+
+	for (i = 0; i < vol->reserved_pebs; i++)
+		vol->eba_tbl[i] = UBI_LEB_UNMAPPED;
+
+	if (vol->vol_type == UBI_DYNAMIC_VOLUME) {
+		vol->used_ebs = vol->reserved_pebs;
+		vol->last_eb_bytes = vol->usable_leb_size;
+		vol->used_bytes =
+			(long long)vol->used_ebs * vol->usable_leb_size;
+	} else {
+		bytes = vol->used_bytes;
+		vol->last_eb_bytes = do_div(bytes, vol->usable_leb_size);
+		vol->used_ebs = bytes;
+		if (vol->last_eb_bytes)
+			vol->used_ebs += 1;
+		else
+			vol->last_eb_bytes = vol->usable_leb_size;
+	}
+
+	/* Register character device for the volume */
+	cdev_init(&vol->cdev, &ubi_vol_cdev_operations);
+	vol->cdev.owner = THIS_MODULE;
+	dev = MKDEV(MAJOR(ubi->cdev.dev), vol_id + 1);
+	err = cdev_add(&vol->cdev, dev, 1);
+	if (err) {
+		ubi_err("cannot add character device");
+		goto out_mapping;
+	}
+
+	err = ubi_create_gluebi(ubi, vol);
+	if (err)
+		goto out_cdev;
+
+	vol->dev.release = vol_release;
+	vol->dev.parent = &ubi->dev;
+	vol->dev.devt = dev;
+	vol->dev.class = ubi_class;
+
+	sprintf(&vol->dev.bus_id[0], "%s_%d", ubi->ubi_name, vol->vol_id);
+	err = device_register(&vol->dev);
+	if (err) {
+		ubi_err("cannot register device");
+		goto out_gluebi;
+	}
+
+	err = volume_sysfs_init(ubi, vol);
+	if (err)
+		goto out_sysfs;
+
+	/* Fill volume table record */
+	memset(&vtbl_rec, 0, sizeof(struct ubi_vtbl_record));
+	vtbl_rec.reserved_pebs = cpu_to_be32(vol->reserved_pebs);
+	vtbl_rec.alignment     = cpu_to_be32(vol->alignment);
+	vtbl_rec.data_pad      = cpu_to_be32(vol->data_pad);
+	vtbl_rec.name_len      = cpu_to_be16(vol->name_len);
+	if (vol->vol_type == UBI_DYNAMIC_VOLUME)
+		vtbl_rec.vol_type = UBI_VID_DYNAMIC;
+	else
+		vtbl_rec.vol_type = UBI_VID_STATIC;
+	memcpy(vtbl_rec.name, vol->name, vol->name_len + 1);
+
+	err = ubi_change_vtbl_record(ubi, vol_id, &vtbl_rec);
+	if (err)
+		goto out_sysfs;
+
+	spin_lock(&ubi->volumes_lock);
+	ubi->volumes[vol_id] = vol;
+	ubi->vol_count += 1;
+	spin_unlock(&ubi->volumes_lock);
+
+	paranoid_check_volumes(ubi);
+	return 0;
+
+out_sysfs:
+	/*
+	 * We have registered our device, we should not free the volume*
+	 * description object in this function in case of an error - it is
+	 * freed by the release function.
+	 *
+	 * Get device reference to prevent the release function from being
+	 * called just after sysfs has been closed.
+	 */
+	dont_free = 1;
+	get_device(&vol->dev);
+	volume_sysfs_close(vol);
+out_gluebi:
+	if (ubi_destroy_gluebi(vol))
+		dbg_err("cannot destroy gluebi for volume %d:%d",
+			ubi->ubi_num, vol_id);
+out_cdev:
+	cdev_del(&vol->cdev);
+out_mapping:
+	kfree(vol->eba_tbl);
+out_acc:
+	spin_lock(&ubi->volumes_lock);
+	ubi->rsvd_pebs -= vol->reserved_pebs;
+	ubi->avail_pebs += vol->reserved_pebs;
+out_unlock:
+	spin_unlock(&ubi->volumes_lock);
+	if (dont_free)
+		put_device(&vol->dev);
+	else
+		kfree(vol);
+	ubi_err("cannot create volume %d, error %d", vol_id, err);
+	return err;
+}
+
+/**
+ * ubi_remove_volume - remove volume.
+ * @desc: volume descriptor
+ *
+ * This function removes volume described by @desc. The volume has to be opened
+ * in "exclusive" mode. Returns zero in case of success and a negative error
+ * code in case of failure. The caller has to have the @ubi->volumes_mutex
+ * locked.
+ */
+int ubi_remove_volume(struct ubi_volume_desc *desc)
+{
+	struct ubi_volume *vol = desc->vol;
+	struct ubi_device *ubi = vol->ubi;
+	int i, err, vol_id = vol->vol_id, reserved_pebs = vol->reserved_pebs;
+
+	dbg_msg("remove UBI volume %d", vol_id);
+	ubi_assert(desc->mode == UBI_EXCLUSIVE);
+	ubi_assert(vol == ubi->volumes[vol_id]);
+
+	if (ubi->ro_mode)
+		return -EROFS;
+
+	spin_lock(&ubi->volumes_lock);
+	if (vol->ref_count > 1) {
+		/*
+		 * The volume is busy, probably someone is reading one of its
+		 * sysfs files.
+		 */
+		err = -EBUSY;
+		goto out_unlock;
+	}
+	ubi->volumes[vol_id] = NULL;
+	spin_unlock(&ubi->volumes_lock);
+
+	err = ubi_destroy_gluebi(vol);
+	if (err)
+		goto out_err;
+
+	err = ubi_change_vtbl_record(ubi, vol_id, NULL);
+	if (err)
+		goto out_err;
+
+	for (i = 0; i < vol->reserved_pebs; i++) {
+		err = ubi_eba_unmap_leb(ubi, vol, i);
+		if (err)
+			goto out_err;
+	}
+
+	kfree(vol->eba_tbl);
+	vol->eba_tbl = NULL;
+	cdev_del(&vol->cdev);
+	volume_sysfs_close(vol);
+
+	spin_lock(&ubi->volumes_lock);
+	ubi->rsvd_pebs -= reserved_pebs;
+	ubi->avail_pebs += reserved_pebs;
+	i = ubi->beb_rsvd_level - ubi->beb_rsvd_pebs;
+	if (i > 0) {
+		i = ubi->avail_pebs >= i ? i : ubi->avail_pebs;
+		ubi->avail_pebs -= i;
+		ubi->rsvd_pebs += i;
+		ubi->beb_rsvd_pebs += i;
+		if (i > 0)
+			ubi_msg("reserve more %d PEBs", i);
+	}
+	ubi->vol_count -= 1;
+	spin_unlock(&ubi->volumes_lock);
+
+	paranoid_check_volumes(ubi);
+	return 0;
+
+out_err:
+	ubi_err("cannot remove volume %d, error %d", vol_id, err);
+	spin_lock(&ubi->volumes_lock);
+	ubi->volumes[vol_id] = vol;
+out_unlock:
+	spin_unlock(&ubi->volumes_lock);
+	return err;
+}
+
+/**
+ * ubi_resize_volume - re-size volume.
+ * @desc: volume descriptor
+ * @reserved_pebs: new size in physical eraseblocks
+ *
+ * This function re-sizes the volume and returns zero in case of success, and a
+ * negative error code in case of failure. The caller has to have the
+ * @ubi->volumes_mutex locked.
+ */
+int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs)
+{
+	int i, err, pebs, *new_mapping;
+	struct ubi_volume *vol = desc->vol;
+	struct ubi_device *ubi = vol->ubi;
+	struct ubi_vtbl_record vtbl_rec;
+	int vol_id = vol->vol_id;
+
+	if (ubi->ro_mode)
+		return -EROFS;
+
+	dbg_msg("re-size volume %d to from %d to %d PEBs",
+		vol_id, vol->reserved_pebs, reserved_pebs);
+
+	if (vol->vol_type == UBI_STATIC_VOLUME &&
+	    reserved_pebs < vol->used_ebs) {
+		dbg_err("too small size %d, %d LEBs contain data",
+			reserved_pebs, vol->used_ebs);
+		return -EINVAL;
+	}
+
+	/* If the size is the same, we have nothing to do */
+	if (reserved_pebs == vol->reserved_pebs)
+		return 0;
+
+	new_mapping = kmalloc(reserved_pebs * sizeof(int), GFP_KERNEL);
+	if (!new_mapping)
+		return -ENOMEM;
+
+	for (i = 0; i < reserved_pebs; i++)
+		new_mapping[i] = UBI_LEB_UNMAPPED;
+
+	spin_lock(&ubi->volumes_lock);
+	if (vol->ref_count > 1) {
+		spin_unlock(&ubi->volumes_lock);
+		err = -EBUSY;
+		goto out_free;
+	}
+	spin_unlock(&ubi->volumes_lock);
+
+	/* Reserve physical eraseblocks */
+	pebs = reserved_pebs - vol->reserved_pebs;
+	if (pebs > 0) {
+		spin_lock(&ubi->volumes_lock);
+		if (pebs > ubi->avail_pebs) {
+			dbg_err("not enough PEBs: requested %d, available %d",
+				pebs, ubi->avail_pebs);
+			spin_unlock(&ubi->volumes_lock);
+			err = -ENOSPC;
+			goto out_free;
+		}
+		ubi->avail_pebs -= pebs;
+		ubi->rsvd_pebs += pebs;
+		for (i = 0; i < vol->reserved_pebs; i++)
+			new_mapping[i] = vol->eba_tbl[i];
+		kfree(vol->eba_tbl);
+		vol->eba_tbl = new_mapping;
+		spin_unlock(&ubi->volumes_lock);
+	}
+
+	/* Change volume table record */
+	memcpy(&vtbl_rec, &ubi->vtbl[vol_id], sizeof(struct ubi_vtbl_record));
+	vtbl_rec.reserved_pebs = cpu_to_be32(reserved_pebs);
+	err = ubi_change_vtbl_record(ubi, vol_id, &vtbl_rec);
+	if (err)
+		goto out_acc;
+
+	if (pebs < 0) {
+		for (i = 0; i < -pebs; i++) {
+			err = ubi_eba_unmap_leb(ubi, vol, reserved_pebs + i);
+			if (err)
+				goto out_acc;
+		}
+		spin_lock(&ubi->volumes_lock);
+		ubi->rsvd_pebs += pebs;
+		ubi->avail_pebs -= pebs;
+		pebs = ubi->beb_rsvd_level - ubi->beb_rsvd_pebs;
+		if (pebs > 0) {
+			pebs = ubi->avail_pebs >= pebs ? pebs : ubi->avail_pebs;
+			ubi->avail_pebs -= pebs;
+			ubi->rsvd_pebs += pebs;
+			ubi->beb_rsvd_pebs += pebs;
+			if (pebs > 0)
+				ubi_msg("reserve more %d PEBs", pebs);
+		}
+		for (i = 0; i < reserved_pebs; i++)
+			new_mapping[i] = vol->eba_tbl[i];
+		kfree(vol->eba_tbl);
+		vol->eba_tbl = new_mapping;
+		spin_unlock(&ubi->volumes_lock);
+	}
+
+	vol->reserved_pebs = reserved_pebs;
+	if (vol->vol_type == UBI_DYNAMIC_VOLUME) {
+		vol->used_ebs = reserved_pebs;
+		vol->last_eb_bytes = vol->usable_leb_size;
+		vol->used_bytes =
+			(long long)vol->used_ebs * vol->usable_leb_size;
+	}
+
+	paranoid_check_volumes(ubi);
+	return 0;
+
+out_acc:
+	if (pebs > 0) {
+		spin_lock(&ubi->volumes_lock);
+		ubi->rsvd_pebs -= pebs;
+		ubi->avail_pebs += pebs;
+		spin_unlock(&ubi->volumes_lock);
+	}
+out_free:
+	kfree(new_mapping);
+	return err;
+}
+
+/**
+ * ubi_add_volume - add volume.
+ * @ubi: UBI device description object
+ * @vol: volume description object
+ *
+ * This function adds an existing volume and initializes all its data
+ * structures. Returns zero in case of success and a negative error code in
+ * case of failure.
+ */
+int ubi_add_volume(struct ubi_device *ubi, struct ubi_volume *vol)
+{
+	int err, vol_id = vol->vol_id;
+	dev_t dev;
+
+	dbg_msg("add volume %d", vol_id);
+	ubi_dbg_dump_vol_info(vol);
+
+	/* Register character device for the volume */
+	cdev_init(&vol->cdev, &ubi_vol_cdev_operations);
+	vol->cdev.owner = THIS_MODULE;
+	dev = MKDEV(MAJOR(ubi->cdev.dev), vol->vol_id + 1);
+	err = cdev_add(&vol->cdev, dev, 1);
+	if (err) {
+		ubi_err("cannot add character device for volume %d, error %d",
+			vol_id, err);
+		return err;
+	}
+
+	err = ubi_create_gluebi(ubi, vol);
+	if (err)
+		goto out_cdev;
+
+	vol->dev.release = vol_release;
+	vol->dev.parent = &ubi->dev;
+	vol->dev.devt = dev;
+	vol->dev.class = ubi_class;
+	sprintf(&vol->dev.bus_id[0], "%s_%d", ubi->ubi_name, vol->vol_id);
+	err = device_register(&vol->dev);
+	if (err)
+		goto out_gluebi;
+
+	err = volume_sysfs_init(ubi, vol);
+	if (err) {
+		cdev_del(&vol->cdev);
+		err = ubi_destroy_gluebi(vol);
+		volume_sysfs_close(vol);
+		return err;
+	}
+
+	paranoid_check_volumes(ubi);
+	return 0;
+
+out_gluebi:
+	err = ubi_destroy_gluebi(vol);
+out_cdev:
+	cdev_del(&vol->cdev);
+	return err;
+}
+
+/**
+ * ubi_free_volume - free volume.
+ * @ubi: UBI device description object
+ * @vol: volume description object
+ *
+ * This function frees all resources for volume @vol but does not remove it.
+ * Used only when the UBI device is detached.
+ */
+void ubi_free_volume(struct ubi_device *ubi, struct ubi_volume *vol)
+{
+	int err;
+
+	dbg_msg("free volume %d", vol->vol_id);
+
+	ubi->volumes[vol->vol_id] = NULL;
+	err = ubi_destroy_gluebi(vol);
+	cdev_del(&vol->cdev);
+	volume_sysfs_close(vol);
+}
+
+#ifdef CONFIG_MTD_UBI_DEBUG_PARANOID
+
+/**
+ * paranoid_check_volume - check volume information.
+ * @ubi: UBI device description object
+ * @vol_id: volume ID
+ */
+static void paranoid_check_volume(struct ubi_device *ubi, int vol_id)
+{
+	int idx = vol_id2idx(ubi, vol_id);
+	int reserved_pebs, alignment, data_pad, vol_type, name_len, upd_marker;
+	const struct ubi_volume *vol;
+	long long n;
+	const char *name;
+
+	spin_lock(&ubi->volumes_lock);
+	reserved_pebs = be32_to_cpu(ubi->vtbl[vol_id].reserved_pebs);
+	vol = ubi->volumes[idx];
+
+	if (!vol) {
+		if (reserved_pebs) {
+			ubi_err("no volume info, but volume exists");
+			goto fail;
+		}
+		spin_unlock(&ubi->volumes_lock);
+		return;
+	}
+
+	if (vol->exclusive) {
+		/*
+		 * The volume may be being created at the moment, do not check
+		 * it (e.g., it may be in the middle of ubi_create_volume().
+		 */
+		spin_unlock(&ubi->volumes_lock);
+		return;
+	}
+
+	if (vol->reserved_pebs < 0 || vol->alignment < 0 || vol->data_pad < 0 ||
+	    vol->name_len < 0) {
+		ubi_err("negative values");
+		goto fail;
+	}
+	if (vol->alignment > ubi->leb_size || vol->alignment == 0) {
+		ubi_err("bad alignment");
+		goto fail;
+	}
+
+	n = vol->alignment & (ubi->min_io_size - 1);
+	if (vol->alignment != 1 && n) {
+		ubi_err("alignment is not multiple of min I/O unit");
+		goto fail;
+	}
+
+	n = ubi->leb_size % vol->alignment;
+	if (vol->data_pad != n) {
+		ubi_err("bad data_pad, has to be %lld", n);
+		goto fail;
+	}
+
+	if (vol->vol_type != UBI_DYNAMIC_VOLUME &&
+	    vol->vol_type != UBI_STATIC_VOLUME) {
+		ubi_err("bad vol_type");
+		goto fail;
+	}
+
+	if (vol->upd_marker && vol->corrupted) {
+		dbg_err("update marker and corrupted simultaneously");
+		goto fail;
+	}
+
+	if (vol->reserved_pebs > ubi->good_peb_count) {
+		ubi_err("too large reserved_pebs");
+		goto fail;
+	}
+
+	n = ubi->leb_size - vol->data_pad;
+	if (vol->usable_leb_size != ubi->leb_size - vol->data_pad) {
+		ubi_err("bad usable_leb_size, has to be %lld", n);
+		goto fail;
+	}
+
+	if (vol->name_len > UBI_VOL_NAME_MAX) {
+		ubi_err("too long volume name, max is %d", UBI_VOL_NAME_MAX);
+		goto fail;
+	}
+
+	if (!vol->name) {
+		ubi_err("NULL volume name");
+		goto fail;
+	}
+
+	n = strnlen(vol->name, vol->name_len + 1);
+	if (n != vol->name_len) {
+		ubi_err("bad name_len %lld", n);
+		goto fail;
+	}
+
+	n = (long long)vol->used_ebs * vol->usable_leb_size;
+	if (vol->vol_type == UBI_DYNAMIC_VOLUME) {
+		if (vol->corrupted) {
+			ubi_err("corrupted dynamic volume");
+			goto fail;
+		}
+		if (vol->used_ebs != vol->reserved_pebs) {
+			ubi_err("bad used_ebs");
+			goto fail;
+		}
+		if (vol->last_eb_bytes != vol->usable_leb_size) {
+			ubi_err("bad last_eb_bytes");
+			goto fail;
+		}
+		if (vol->used_bytes != n) {
+			ubi_err("bad used_bytes");
+			goto fail;
+		}
+	} else {
+		if (vol->used_ebs < 0 || vol->used_ebs > vol->reserved_pebs) {
+			ubi_err("bad used_ebs");
+			goto fail;
+		}
+		if (vol->last_eb_bytes < 0 ||
+		    vol->last_eb_bytes > vol->usable_leb_size) {
+			ubi_err("bad last_eb_bytes");
+			goto fail;
+		}
+		if (vol->used_bytes < 0 || vol->used_bytes > n ||
+		    vol->used_bytes < n - vol->usable_leb_size) {
+			ubi_err("bad used_bytes");
+			goto fail;
+		}
+	}
+
+	alignment  = be32_to_cpu(ubi->vtbl[vol_id].alignment);
+	data_pad   = be32_to_cpu(ubi->vtbl[vol_id].data_pad);
+	name_len   = be16_to_cpu(ubi->vtbl[vol_id].name_len);
+	upd_marker = ubi->vtbl[vol_id].upd_marker;
+	name       = &ubi->vtbl[vol_id].name[0];
+	if (ubi->vtbl[vol_id].vol_type == UBI_VID_DYNAMIC)
+		vol_type = UBI_DYNAMIC_VOLUME;
+	else
+		vol_type = UBI_STATIC_VOLUME;
+
+	if (alignment != vol->alignment || data_pad != vol->data_pad ||
+	    upd_marker != vol->upd_marker || vol_type != vol->vol_type ||
+	    name_len!= vol->name_len || strncmp(name, vol->name, name_len)) {
+		ubi_err("volume info is different");
+		goto fail;
+	}
+
+	spin_unlock(&ubi->volumes_lock);
+	return;
+
+fail:
+	ubi_err("paranoid check failed for volume %d", vol_id);
+	ubi_dbg_dump_vol_info(vol);
+	ubi_dbg_dump_vtbl_record(&ubi->vtbl[vol_id], vol_id);
+	spin_unlock(&ubi->volumes_lock);
+	BUG();
+}
+
+/**
+ * paranoid_check_volumes - check information about all volumes.
+ * @ubi: UBI device description object
+ */
+static void paranoid_check_volumes(struct ubi_device *ubi)
+{
+	int i;
+
+	for (i = 0; i < ubi->vtbl_slots; i++)
+		paranoid_check_volume(ubi, i);
+}
+#endif

drivers/mtd/ubi/vtbl.c

@@ -0,0 +1,837 @@
+/*
+ * Copyright (c) International Business Machines Corp., 2006
+ * Copyright (c) Nokia Corporation, 2006, 2007
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * 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 for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Author: Artem Bityutskiy (Битюцкий Артём)
+ */
+
+/*
+ * This file includes volume table manipulation code. The volume table is an
+ * on-flash table containing volume meta-data like name, number of reserved
+ * physical eraseblocks, type, etc. The volume table is stored in the so-called
+ * "layout volume".
+ *
+ * The layout volume is an internal volume which is organized as follows. It
+ * consists of two logical eraseblocks - LEB 0 and LEB 1. Each logical
+ * eraseblock stores one volume table copy, i.e. LEB 0 and LEB 1 duplicate each
+ * other. This redundancy guarantees robustness to unclean reboots. The volume
+ * table is basically an array of volume table records. Each record contains
+ * full information about the volume and protected by a CRC checksum.
+ *
+ * The volume table is changed, it is first changed in RAM. Then LEB 0 is
+ * erased, and the updated volume table is written back to LEB 0. Then same for
+ * LEB 1. This scheme guarantees recoverability from unclean reboots.
+ *
+ * In this UBI implementation the on-flash volume table does not contain any
+ * information about how many data static volumes contain. This information may
+ * be found from the scanning data.
+ *
+ * But it would still be beneficial to store this information in the volume
+ * table. For example, suppose we have a static volume X, and all its physical
+ * eraseblocks became bad for some reasons. Suppose we are attaching the
+ * corresponding MTD device, the scanning has found no logical eraseblocks
+ * corresponding to the volume X. According to the volume table volume X does
+ * exist. So we don't know whether it is just empty or all its physical
+ * eraseblocks went bad. So we cannot alarm the user about this corruption.
+ *
+ * The volume table also stores so-called "update marker", which is used for
+ * volume updates. Before updating the volume, the update marker is set, and
+ * after the update operation is finished, the update marker is cleared. So if
+ * the update operation was interrupted (e.g. by an unclean reboot) - the
+ * update marker is still there and we know that the volume's contents is
+ * damaged.
+ */
+
+#ifdef UBI_LINUX
+#include <linux/crc32.h>
+#include <linux/err.h>
+#include <asm/div64.h>
+#endif
+
+#include <ubi_uboot.h>
+#include "ubi.h"
+
+#ifdef CONFIG_MTD_UBI_DEBUG_PARANOID
+static void paranoid_vtbl_check(const struct ubi_device *ubi);
+#else
+#define paranoid_vtbl_check(ubi)
+#endif
+
+/* Empty volume table record */
+static struct ubi_vtbl_record empty_vtbl_record;
+
+/**
+ * ubi_change_vtbl_record - change volume table record.
+ * @ubi: UBI device description object
+ * @idx: table index to change
+ * @vtbl_rec: new volume table record
+ *
+ * This function changes volume table record @idx. If @vtbl_rec is %NULL, empty
+ * volume table record is written. The caller does not have to calculate CRC of
+ * the record as it is done by this function. Returns zero in case of success
+ * and a negative error code in case of failure.
+ */
+int ubi_change_vtbl_record(struct ubi_device *ubi, int idx,
+			   struct ubi_vtbl_record *vtbl_rec)
+{
+	int i, err;
+	uint32_t crc;
+	struct ubi_volume *layout_vol;
+
+	ubi_assert(idx >= 0 && idx < ubi->vtbl_slots);
+	layout_vol = ubi->volumes[vol_id2idx(ubi, UBI_LAYOUT_VOLUME_ID)];
+
+	if (!vtbl_rec)
+		vtbl_rec = &empty_vtbl_record;
+	else {
+		crc = crc32(UBI_CRC32_INIT, vtbl_rec, UBI_VTBL_RECORD_SIZE_CRC);
+		vtbl_rec->crc = cpu_to_be32(crc);
+	}
+
+	memcpy(&ubi->vtbl[idx], vtbl_rec, sizeof(struct ubi_vtbl_record));
+	for (i = 0; i < UBI_LAYOUT_VOLUME_EBS; i++) {
+		err = ubi_eba_unmap_leb(ubi, layout_vol, i);
+		if (err)
+			return err;
+
+		err = ubi_eba_write_leb(ubi, layout_vol, i, ubi->vtbl, 0,
+					ubi->vtbl_size, UBI_LONGTERM);
+		if (err)
+			return err;
+	}
+
+	paranoid_vtbl_check(ubi);
+	return 0;
+}
+
+/**
+ * vtbl_check - check if volume table is not corrupted and contains sensible
+ *              data.
+ * @ubi: UBI device description object
+ * @vtbl: volume table
+ *
+ * This function returns zero if @vtbl is all right, %1 if CRC is incorrect,
+ * and %-EINVAL if it contains inconsistent data.
+ */
+static int vtbl_check(const struct ubi_device *ubi,
+		      const struct ubi_vtbl_record *vtbl)
+{
+	int i, n, reserved_pebs, alignment, data_pad, vol_type, name_len;
+	int upd_marker, err;
+	uint32_t crc;
+	const char *name;
+
+	for (i = 0; i < ubi->vtbl_slots; i++) {
+		cond_resched();
+
+		reserved_pebs = be32_to_cpu(vtbl[i].reserved_pebs);
+		alignment = be32_to_cpu(vtbl[i].alignment);
+		data_pad = be32_to_cpu(vtbl[i].data_pad);
+		upd_marker = vtbl[i].upd_marker;
+		vol_type = vtbl[i].vol_type;
+		name_len = be16_to_cpu(vtbl[i].name_len);
+		name = (const char *) &vtbl[i].name[0];
+
+		crc = crc32(UBI_CRC32_INIT, &vtbl[i], UBI_VTBL_RECORD_SIZE_CRC);
+		if (be32_to_cpu(vtbl[i].crc) != crc) {
+			ubi_err("bad CRC at record %u: %#08x, not %#08x",
+				 i, crc, be32_to_cpu(vtbl[i].crc));
+			ubi_dbg_dump_vtbl_record(&vtbl[i], i);
+			return 1;
+		}
+
+		if (reserved_pebs == 0) {
+			if (memcmp(&vtbl[i], &empty_vtbl_record,
+						UBI_VTBL_RECORD_SIZE)) {
+				err = 2;
+				goto bad;
+			}
+			continue;
+		}
+
+		if (reserved_pebs < 0 || alignment < 0 || data_pad < 0 ||
+		    name_len < 0) {
+			err = 3;
+			goto bad;
+		}
+
+		if (alignment > ubi->leb_size || alignment == 0) {
+			err = 4;
+			goto bad;
+		}
+
+		n = alignment & (ubi->min_io_size - 1);
+		if (alignment != 1 && n) {
+			err = 5;
+			goto bad;
+		}
+
+		n = ubi->leb_size % alignment;
+		if (data_pad != n) {
+			dbg_err("bad data_pad, has to be %d", n);
+			err = 6;
+			goto bad;
+		}
+
+		if (vol_type != UBI_VID_DYNAMIC && vol_type != UBI_VID_STATIC) {
+			err = 7;
+			goto bad;
+		}
+
+		if (upd_marker != 0 && upd_marker != 1) {
+			err = 8;
+			goto bad;
+		}
+
+		if (reserved_pebs > ubi->good_peb_count) {
+			dbg_err("too large reserved_pebs, good PEBs %d",
+				ubi->good_peb_count);
+			err = 9;
+			goto bad;
+		}
+
+		if (name_len > UBI_VOL_NAME_MAX) {
+			err = 10;
+			goto bad;
+		}
+
+		if (name[0] == '\0') {
+			err = 11;
+			goto bad;
+		}
+
+		if (name_len != strnlen(name, name_len + 1)) {
+			err = 12;
+			goto bad;
+		}
+	}
+
+	/* Checks that all names are unique */
+	for (i = 0; i < ubi->vtbl_slots - 1; i++) {
+		for (n = i + 1; n < ubi->vtbl_slots; n++) {
+			int len1 = be16_to_cpu(vtbl[i].name_len);
+			int len2 = be16_to_cpu(vtbl[n].name_len);
+
+			if (len1 > 0 && len1 == len2 &&
+			    !strncmp((char *)vtbl[i].name, (char *)vtbl[n].name, len1)) {
+				ubi_err("volumes %d and %d have the same name"
+					" \"%s\"", i, n, vtbl[i].name);
+				ubi_dbg_dump_vtbl_record(&vtbl[i], i);
+				ubi_dbg_dump_vtbl_record(&vtbl[n], n);
+				return -EINVAL;
+			}
+		}
+	}
+
+	return 0;
+
+bad:
+	ubi_err("volume table check failed: record %d, error %d", i, err);
+	ubi_dbg_dump_vtbl_record(&vtbl[i], i);
+	return -EINVAL;
+}
+
+/**
+ * create_vtbl - create a copy of volume table.
+ * @ubi: UBI device description object
+ * @si: scanning information
+ * @copy: number of the volume table copy
+ * @vtbl: contents of the volume table
+ *
+ * This function returns zero in case of success and a negative error code in
+ * case of failure.
+ */
+static int create_vtbl(struct ubi_device *ubi, struct ubi_scan_info *si,
+		       int copy, void *vtbl)
+{
+	int err, tries = 0;
+	static struct ubi_vid_hdr *vid_hdr;
+	struct ubi_scan_volume *sv;
+	struct ubi_scan_leb *new_seb, *old_seb = NULL;
+
+	ubi_msg("create volume table (copy #%d)", copy + 1);
+
+	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_KERNEL);
+	if (!vid_hdr)
+		return -ENOMEM;
+
+	/*
+	 * Check if there is a logical eraseblock which would have to contain
+	 * this volume table copy was found during scanning. It has to be wiped
+	 * out.
+	 */
+	sv = ubi_scan_find_sv(si, UBI_LAYOUT_VOLUME_ID);
+	if (sv)
+		old_seb = ubi_scan_find_seb(sv, copy);
+
+retry:
+	new_seb = ubi_scan_get_free_peb(ubi, si);
+	if (IS_ERR(new_seb)) {
+		err = PTR_ERR(new_seb);
+		goto out_free;
+	}
+
+	vid_hdr->vol_type = UBI_VID_DYNAMIC;
+	vid_hdr->vol_id = cpu_to_be32(UBI_LAYOUT_VOLUME_ID);
+	vid_hdr->compat = UBI_LAYOUT_VOLUME_COMPAT;
+	vid_hdr->data_size = vid_hdr->used_ebs =
+			     vid_hdr->data_pad = cpu_to_be32(0);
+	vid_hdr->lnum = cpu_to_be32(copy);
+	vid_hdr->sqnum = cpu_to_be64(++si->max_sqnum);
+	vid_hdr->leb_ver = cpu_to_be32(old_seb ? old_seb->leb_ver + 1: 0);
+
+	/* The EC header is already there, write the VID header */
+	err = ubi_io_write_vid_hdr(ubi, new_seb->pnum, vid_hdr);
+	if (err)
+		goto write_error;
+
+	/* Write the layout volume contents */
+	err = ubi_io_write_data(ubi, vtbl, new_seb->pnum, 0, ubi->vtbl_size);
+	if (err)
+		goto write_error;
+
+	/*
+	 * And add it to the scanning information. Don't delete the old
+	 * @old_seb as it will be deleted and freed in 'ubi_scan_add_used()'.
+	 */
+	err = ubi_scan_add_used(ubi, si, new_seb->pnum, new_seb->ec,
+				vid_hdr, 0);
+	kfree(new_seb);
+	ubi_free_vid_hdr(ubi, vid_hdr);
+	return err;
+
+write_error:
+	if (err == -EIO && ++tries <= 5) {
+		/*
+		 * Probably this physical eraseblock went bad, try to pick
+		 * another one.
+		 */
+		list_add_tail(&new_seb->u.list, &si->corr);
+		goto retry;
+	}
+	kfree(new_seb);
+out_free:
+	ubi_free_vid_hdr(ubi, vid_hdr);
+	return err;
+
+}
+
+/**
+ * process_lvol - process the layout volume.
+ * @ubi: UBI device description object
+ * @si: scanning information
+ * @sv: layout volume scanning information
+ *
+ * This function is responsible for reading the layout volume, ensuring it is
+ * not corrupted, and recovering from corruptions if needed. Returns volume
+ * table in case of success and a negative error code in case of failure.
+ */
+static struct ubi_vtbl_record *process_lvol(struct ubi_device *ubi,
+					    struct ubi_scan_info *si,
+					    struct ubi_scan_volume *sv)
+{
+	int err;
+	struct rb_node *rb;
+	struct ubi_scan_leb *seb;
+	struct ubi_vtbl_record *leb[UBI_LAYOUT_VOLUME_EBS] = { NULL, NULL };
+	int leb_corrupted[UBI_LAYOUT_VOLUME_EBS] = {1, 1};
+
+	/*
+	 * UBI goes through the following steps when it changes the layout
+	 * volume:
+	 * a. erase LEB 0;
+	 * b. write new data to LEB 0;
+	 * c. erase LEB 1;
+	 * d. write new data to LEB 1.
+	 *
+	 * Before the change, both LEBs contain the same data.
+	 *
+	 * Due to unclean reboots, the contents of LEB 0 may be lost, but there
+	 * should LEB 1. So it is OK if LEB 0 is corrupted while LEB 1 is not.
+	 * Similarly, LEB 1 may be lost, but there should be LEB 0. And
+	 * finally, unclean reboots may result in a situation when neither LEB
+	 * 0 nor LEB 1 are corrupted, but they are different. In this case, LEB
+	 * 0 contains more recent information.
+	 *
+	 * So the plan is to first check LEB 0. Then
+	 * a. if LEB 0 is OK, it must be containing the most resent data; then
+	 *    we compare it with LEB 1, and if they are different, we copy LEB
+	 *    0 to LEB 1;
+	 * b. if LEB 0 is corrupted, but LEB 1 has to be OK, and we copy LEB 1
+	 *    to LEB 0.
+	 */
+
+	dbg_msg("check layout volume");
+
+	/* Read both LEB 0 and LEB 1 into memory */
+	ubi_rb_for_each_entry(rb, seb, &sv->root, u.rb) {
+		leb[seb->lnum] = vmalloc(ubi->vtbl_size);
+		if (!leb[seb->lnum]) {
+			err = -ENOMEM;
+			goto out_free;
+		}
+		memset(leb[seb->lnum], 0, ubi->vtbl_size);
+
+		err = ubi_io_read_data(ubi, leb[seb->lnum], seb->pnum, 0,
+				       ubi->vtbl_size);
+		if (err == UBI_IO_BITFLIPS || err == -EBADMSG)
+			/*
+			 * Scrub the PEB later. Note, -EBADMSG indicates an
+			 * uncorrectable ECC error, but we have our own CRC and
+			 * the data will be checked later. If the data is OK,
+			 * the PEB will be scrubbed (because we set
+			 * seb->scrub). If the data is not OK, the contents of
+			 * the PEB will be recovered from the second copy, and
+			 * seb->scrub will be cleared in
+			 * 'ubi_scan_add_used()'.
+			 */
+			seb->scrub = 1;
+		else if (err)
+			goto out_free;
+	}
+
+	err = -EINVAL;
+	if (leb[0]) {
+		leb_corrupted[0] = vtbl_check(ubi, leb[0]);
+		if (leb_corrupted[0] < 0)
+			goto out_free;
+	}
+
+	if (!leb_corrupted[0]) {
+		/* LEB 0 is OK */
+		if (leb[1])
+			leb_corrupted[1] = memcmp(leb[0], leb[1], ubi->vtbl_size);
+		if (leb_corrupted[1]) {
+			ubi_warn("volume table copy #2 is corrupted");
+			err = create_vtbl(ubi, si, 1, leb[0]);
+			if (err)
+				goto out_free;
+			ubi_msg("volume table was restored");
+		}
+
+		/* Both LEB 1 and LEB 2 are OK and consistent */
+		vfree(leb[1]);
+		return leb[0];
+	} else {
+		/* LEB 0 is corrupted or does not exist */
+		if (leb[1]) {
+			leb_corrupted[1] = vtbl_check(ubi, leb[1]);
+			if (leb_corrupted[1] < 0)
+				goto out_free;
+		}
+		if (leb_corrupted[1]) {
+			/* Both LEB 0 and LEB 1 are corrupted */
+			ubi_err("both volume tables are corrupted");
+			goto out_free;
+		}
+
+		ubi_warn("volume table copy #1 is corrupted");
+		err = create_vtbl(ubi, si, 0, leb[1]);
+		if (err)
+			goto out_free;
+		ubi_msg("volume table was restored");
+
+		vfree(leb[0]);
+		return leb[1];
+	}
+
+out_free:
+	vfree(leb[0]);
+	vfree(leb[1]);
+	return ERR_PTR(err);
+}
+
+/**
+ * create_empty_lvol - create empty layout volume.
+ * @ubi: UBI device description object
+ * @si: scanning information
+ *
+ * This function returns volume table contents in case of success and a
+ * negative error code in case of failure.
+ */
+static struct ubi_vtbl_record *create_empty_lvol(struct ubi_device *ubi,
+						 struct ubi_scan_info *si)
+{
+	int i;
+	struct ubi_vtbl_record *vtbl;
+
+	vtbl = vmalloc(ubi->vtbl_size);
+	if (!vtbl)
+		return ERR_PTR(-ENOMEM);
+	memset(vtbl, 0, ubi->vtbl_size);
+
+	for (i = 0; i < ubi->vtbl_slots; i++)
+		memcpy(&vtbl[i], &empty_vtbl_record, UBI_VTBL_RECORD_SIZE);
+
+	for (i = 0; i < UBI_LAYOUT_VOLUME_EBS; i++) {
+		int err;
+
+		err = create_vtbl(ubi, si, i, vtbl);
+		if (err) {
+			vfree(vtbl);
+			return ERR_PTR(err);
+		}
+	}
+
+	return vtbl;
+}
+
+/**
+ * init_volumes - initialize volume information for existing volumes.
+ * @ubi: UBI device description object
+ * @si: scanning information
+ * @vtbl: volume table
+ *
+ * This function allocates volume description objects for existing volumes.
+ * Returns zero in case of success and a negative error code in case of
+ * failure.
+ */
+static int init_volumes(struct ubi_device *ubi, const struct ubi_scan_info *si,
+			const struct ubi_vtbl_record *vtbl)
+{
+	int i, reserved_pebs = 0;
+	struct ubi_scan_volume *sv;
+	struct ubi_volume *vol;
+
+	for (i = 0; i < ubi->vtbl_slots; i++) {
+		cond_resched();
+
+		if (be32_to_cpu(vtbl[i].reserved_pebs) == 0)
+			continue; /* Empty record */
+
+		vol = kzalloc(sizeof(struct ubi_volume), GFP_KERNEL);
+		if (!vol)
+			return -ENOMEM;
+
+		vol->reserved_pebs = be32_to_cpu(vtbl[i].reserved_pebs);
+		vol->alignment = be32_to_cpu(vtbl[i].alignment);
+		vol->data_pad = be32_to_cpu(vtbl[i].data_pad);
+		vol->vol_type = vtbl[i].vol_type == UBI_VID_DYNAMIC ?
+					UBI_DYNAMIC_VOLUME : UBI_STATIC_VOLUME;
+		vol->name_len = be16_to_cpu(vtbl[i].name_len);
+		vol->usable_leb_size = ubi->leb_size - vol->data_pad;
+		memcpy(vol->name, vtbl[i].name, vol->name_len);
+		vol->name[vol->name_len] = '\0';
+		vol->vol_id = i;
+
+		if (vtbl[i].flags & UBI_VTBL_AUTORESIZE_FLG) {
+			/* Auto re-size flag may be set only for one volume */
+			if (ubi->autoresize_vol_id != -1) {
+				ubi_err("more then one auto-resize volume (%d "
+					"and %d)", ubi->autoresize_vol_id, i);
+				kfree(vol);
+				return -EINVAL;
+			}
+
+			ubi->autoresize_vol_id = i;
+		}
+
+		ubi_assert(!ubi->volumes[i]);
+		ubi->volumes[i] = vol;
+		ubi->vol_count += 1;
+		vol->ubi = ubi;
+		reserved_pebs += vol->reserved_pebs;
+
+		/*
+		 * In case of dynamic volume UBI knows nothing about how many
+		 * data is stored there. So assume the whole volume is used.
+		 */
+		if (vol->vol_type == UBI_DYNAMIC_VOLUME) {
+			vol->used_ebs = vol->reserved_pebs;
+			vol->last_eb_bytes = vol->usable_leb_size;
+			vol->used_bytes =
+				(long long)vol->used_ebs * vol->usable_leb_size;
+			continue;
+		}
+
+		/* Static volumes only */
+		sv = ubi_scan_find_sv(si, i);
+		if (!sv) {
+			/*
+			 * No eraseblocks belonging to this volume found. We
+			 * don't actually know whether this static volume is
+			 * completely corrupted or just contains no data. And
+			 * we cannot know this as long as data size is not
+			 * stored on flash. So we just assume the volume is
+			 * empty. FIXME: this should be handled.
+			 */
+			continue;
+		}
+
+		if (sv->leb_count != sv->used_ebs) {
+			/*
+			 * We found a static volume which misses several
+			 * eraseblocks. Treat it as corrupted.
+			 */
+			ubi_warn("static volume %d misses %d LEBs - corrupted",
+				 sv->vol_id, sv->used_ebs - sv->leb_count);
+			vol->corrupted = 1;
+			continue;
+		}
+
+		vol->used_ebs = sv->used_ebs;
+		vol->used_bytes =
+			(long long)(vol->used_ebs - 1) * vol->usable_leb_size;
+		vol->used_bytes += sv->last_data_size;
+		vol->last_eb_bytes = sv->last_data_size;
+	}
+
+	/* And add the layout volume */
+	vol = kzalloc(sizeof(struct ubi_volume), GFP_KERNEL);
+	if (!vol)
+		return -ENOMEM;
+
+	vol->reserved_pebs = UBI_LAYOUT_VOLUME_EBS;
+	vol->alignment = 1;
+	vol->vol_type = UBI_DYNAMIC_VOLUME;
+	vol->name_len = sizeof(UBI_LAYOUT_VOLUME_NAME) - 1;
+	memcpy(vol->name, UBI_LAYOUT_VOLUME_NAME, vol->name_len + 1);
+	vol->usable_leb_size = ubi->leb_size;
+	vol->used_ebs = vol->reserved_pebs;
+	vol->last_eb_bytes = vol->reserved_pebs;
+	vol->used_bytes =
+		(long long)vol->used_ebs * (ubi->leb_size - vol->data_pad);
+	vol->vol_id = UBI_LAYOUT_VOLUME_ID;
+	vol->ref_count = 1;
+
+	ubi_assert(!ubi->volumes[i]);
+	ubi->volumes[vol_id2idx(ubi, vol->vol_id)] = vol;
+	reserved_pebs += vol->reserved_pebs;
+	ubi->vol_count += 1;
+	vol->ubi = ubi;
+
+	if (reserved_pebs > ubi->avail_pebs)
+		ubi_err("not enough PEBs, required %d, available %d",
+			reserved_pebs, ubi->avail_pebs);
+	ubi->rsvd_pebs += reserved_pebs;
+	ubi->avail_pebs -= reserved_pebs;
+
+	return 0;
+}
+
+/**
+ * check_sv - check volume scanning information.
+ * @vol: UBI volume description object
+ * @sv: volume scanning information
+ *
+ * This function returns zero if the volume scanning information is consistent
+ * to the data read from the volume tabla, and %-EINVAL if not.
+ */
+static int check_sv(const struct ubi_volume *vol,
+		    const struct ubi_scan_volume *sv)
+{
+	int err;
+
+	if (sv->highest_lnum >= vol->reserved_pebs) {
+		err = 1;
+		goto bad;
+	}
+	if (sv->leb_count > vol->reserved_pebs) {
+		err = 2;
+		goto bad;
+	}
+	if (sv->vol_type != vol->vol_type) {
+		err = 3;
+		goto bad;
+	}
+	if (sv->used_ebs > vol->reserved_pebs) {
+		err = 4;
+		goto bad;
+	}
+	if (sv->data_pad != vol->data_pad) {
+		err = 5;
+		goto bad;
+	}
+	return 0;
+
+bad:
+	ubi_err("bad scanning information, error %d", err);
+	ubi_dbg_dump_sv(sv);
+	ubi_dbg_dump_vol_info(vol);
+	return -EINVAL;
+}
+
+/**
+ * check_scanning_info - check that scanning information.
+ * @ubi: UBI device description object
+ * @si: scanning information
+ *
+ * Even though we protect on-flash data by CRC checksums, we still don't trust
+ * the media. This function ensures that scanning information is consistent to
+ * the information read from the volume table. Returns zero if the scanning
+ * information is OK and %-EINVAL if it is not.
+ */
+static int check_scanning_info(const struct ubi_device *ubi,
+			       struct ubi_scan_info *si)
+{
+	int err, i;
+	struct ubi_scan_volume *sv;
+	struct ubi_volume *vol;
+
+	if (si->vols_found > UBI_INT_VOL_COUNT + ubi->vtbl_slots) {
+		ubi_err("scanning found %d volumes, maximum is %d + %d",
+			si->vols_found, UBI_INT_VOL_COUNT, ubi->vtbl_slots);
+		return -EINVAL;
+	}
+
+	if (si->highest_vol_id >= ubi->vtbl_slots + UBI_INT_VOL_COUNT &&
+	    si->highest_vol_id < UBI_INTERNAL_VOL_START) {
+		ubi_err("too large volume ID %d found by scanning",
+			si->highest_vol_id);
+		return -EINVAL;
+	}
+
+	for (i = 0; i < ubi->vtbl_slots + UBI_INT_VOL_COUNT; i++) {
+		cond_resched();
+
+		sv = ubi_scan_find_sv(si, i);
+		vol = ubi->volumes[i];
+		if (!vol) {
+			if (sv)
+				ubi_scan_rm_volume(si, sv);
+			continue;
+		}
+
+		if (vol->reserved_pebs == 0) {
+			ubi_assert(i < ubi->vtbl_slots);
+
+			if (!sv)
+				continue;
+
+			/*
+			 * During scanning we found a volume which does not
+			 * exist according to the information in the volume
+			 * table. This must have happened due to an unclean
+			 * reboot while the volume was being removed. Discard
+			 * these eraseblocks.
+			 */
+			ubi_msg("finish volume %d removal", sv->vol_id);
+			ubi_scan_rm_volume(si, sv);
+		} else if (sv) {
+			err = check_sv(vol, sv);
+			if (err)
+				return err;
+		}
+	}
+
+	return 0;
+}
+
+/**
+ * ubi_read_volume_table - read volume table.
+ * information.
+ * @ubi: UBI device description object
+ * @si: scanning information
+ *
+ * This function reads volume table, checks it, recover from errors if needed,
+ * or creates it if needed. Returns zero in case of success and a negative
+ * error code in case of failure.
+ */
+int ubi_read_volume_table(struct ubi_device *ubi, struct ubi_scan_info *si)
+{
+	int i, err;
+	struct ubi_scan_volume *sv;
+
+	empty_vtbl_record.crc = cpu_to_be32(0xf116c36b);
+
+	/*
+	 * The number of supported volumes is limited by the eraseblock size
+	 * and by the UBI_MAX_VOLUMES constant.
+	 */
+	ubi->vtbl_slots = ubi->leb_size / UBI_VTBL_RECORD_SIZE;
+	if (ubi->vtbl_slots > UBI_MAX_VOLUMES)
+		ubi->vtbl_slots = UBI_MAX_VOLUMES;
+
+	ubi->vtbl_size = ubi->vtbl_slots * UBI_VTBL_RECORD_SIZE;
+	ubi->vtbl_size = ALIGN(ubi->vtbl_size, ubi->min_io_size);
+
+	sv = ubi_scan_find_sv(si, UBI_LAYOUT_VOLUME_ID);
+	if (!sv) {
+		/*
+		 * No logical eraseblocks belonging to the layout volume were
+		 * found. This could mean that the flash is just empty. In
+		 * this case we create empty layout volume.
+		 *
+		 * But if flash is not empty this must be a corruption or the
+		 * MTD device just contains garbage.
+		 */
+		if (si->is_empty) {
+			ubi->vtbl = create_empty_lvol(ubi, si);
+			if (IS_ERR(ubi->vtbl))
+				return PTR_ERR(ubi->vtbl);
+		} else {
+			ubi_err("the layout volume was not found");
+			return -EINVAL;
+		}
+	} else {
+		if (sv->leb_count > UBI_LAYOUT_VOLUME_EBS) {
+			/* This must not happen with proper UBI images */
+			dbg_err("too many LEBs (%d) in layout volume",
+				sv->leb_count);
+			return -EINVAL;
+		}
+
+		ubi->vtbl = process_lvol(ubi, si, sv);
+		if (IS_ERR(ubi->vtbl))
+			return PTR_ERR(ubi->vtbl);
+	}
+
+	ubi->avail_pebs = ubi->good_peb_count;
+
+	/*
+	 * The layout volume is OK, initialize the corresponding in-RAM data
+	 * structures.
+	 */
+	err = init_volumes(ubi, si, ubi->vtbl);
+	if (err)
+		goto out_free;
+
+	/*
+	 * Get sure that the scanning information is consistent to the
+	 * information stored in the volume table.
+	 */
+	err = check_scanning_info(ubi, si);
+	if (err)
+		goto out_free;
+
+	return 0;
+
+out_free:
+	vfree(ubi->vtbl);
+	for (i = 0; i < ubi->vtbl_slots + UBI_INT_VOL_COUNT; i++)
+		if (ubi->volumes[i]) {
+			kfree(ubi->volumes[i]);
+			ubi->volumes[i] = NULL;
+		}
+	return err;
+}
+
+#ifdef CONFIG_MTD_UBI_DEBUG_PARANOID
+
+/**
+ * paranoid_vtbl_check - check volume table.
+ * @ubi: UBI device description object
+ */
+static void paranoid_vtbl_check(const struct ubi_device *ubi)
+{
+	if (vtbl_check(ubi, ubi->vtbl)) {
+		ubi_err("paranoid check failed");
+		BUG();
+	}
+}
+
+#endif /* CONFIG_MTD_UBI_DEBUG_PARANOID */

drivers/mtd/ubi/wl.c

@@ -0,0 +1,1670 @@
+/*
+ * Copyright (c) International Business Machines Corp., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * 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 for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Authors: Artem Bityutskiy (Битюцкий Артём), Thomas Gleixner
+ */
+
+/*
+ * UBI wear-leveling unit.
+ *
+ * This unit is responsible for wear-leveling. It works in terms of physical
+ * eraseblocks and erase counters and knows nothing about logical eraseblocks,
+ * volumes, etc. From this unit's perspective all physical eraseblocks are of
+ * two types - used and free. Used physical eraseblocks are those that were
+ * "get" by the 'ubi_wl_get_peb()' function, and free physical eraseblocks are
+ * those that were put by the 'ubi_wl_put_peb()' function.
+ *
+ * Physical eraseblocks returned by 'ubi_wl_get_peb()' have only erase counter
+ * header. The rest of the physical eraseblock contains only 0xFF bytes.
+ *
+ * When physical eraseblocks are returned to the WL unit by means of the
+ * 'ubi_wl_put_peb()' function, they are scheduled for erasure. The erasure is
+ * done asynchronously in context of the per-UBI device background thread,
+ * which is also managed by the WL unit.
+ *
+ * The wear-leveling is ensured by means of moving the contents of used
+ * physical eraseblocks with low erase counter to free physical eraseblocks
+ * with high erase counter.
+ *
+ * The 'ubi_wl_get_peb()' function accepts data type hints which help to pick
+ * an "optimal" physical eraseblock. For example, when it is known that the
+ * physical eraseblock will be "put" soon because it contains short-term data,
+ * the WL unit may pick a free physical eraseblock with low erase counter, and
+ * so forth.
+ *
+ * If the WL unit fails to erase a physical eraseblock, it marks it as bad.
+ *
+ * This unit is also responsible for scrubbing. If a bit-flip is detected in a
+ * physical eraseblock, it has to be moved. Technically this is the same as
+ * moving it for wear-leveling reasons.
+ *
+ * As it was said, for the UBI unit all physical eraseblocks are either "free"
+ * or "used". Free eraseblock are kept in the @wl->free RB-tree, while used
+ * eraseblocks are kept in a set of different RB-trees: @wl->used,
+ * @wl->prot.pnum, @wl->prot.aec, and @wl->scrub.
+ *
+ * Note, in this implementation, we keep a small in-RAM object for each physical
+ * eraseblock. This is surely not a scalable solution. But it appears to be good
+ * enough for moderately large flashes and it is simple. In future, one may
+ * re-work this unit and make it more scalable.
+ *
+ * At the moment this unit does not utilize the sequence number, which was
+ * introduced relatively recently. But it would be wise to do this because the
+ * sequence number of a logical eraseblock characterizes how old is it. For
+ * example, when we move a PEB with low erase counter, and we need to pick the
+ * target PEB, we pick a PEB with the highest EC if our PEB is "old" and we
+ * pick target PEB with an average EC if our PEB is not very "old". This is a
+ * room for future re-works of the WL unit.
+ *
+ * FIXME: looks too complex, should be simplified (later).
+ */
+
+#ifdef UBI_LINUX
+#include <linux/slab.h>
+#include <linux/crc32.h>
+#include <linux/freezer.h>
+#include <linux/kthread.h>
+#endif
+
+#include <ubi_uboot.h>
+#include "ubi.h"
+
+/* Number of physical eraseblocks reserved for wear-leveling purposes */
+#define WL_RESERVED_PEBS 1
+
+/*
+ * How many erase cycles are short term, unknown, and long term physical
+ * eraseblocks protected.
+ */
+#define ST_PROTECTION 16
+#define U_PROTECTION  10
+#define LT_PROTECTION 4
+
+/*
+ * Maximum difference between two erase counters. If this threshold is
+ * exceeded, the WL unit starts moving data from used physical eraseblocks with
+ * low erase counter to free physical eraseblocks with high erase counter.
+ */
+#define UBI_WL_THRESHOLD CONFIG_MTD_UBI_WL_THRESHOLD
+
+/*
+ * When a physical eraseblock is moved, the WL unit has to pick the target
+ * physical eraseblock to move to. The simplest way would be just to pick the
+ * one with the highest erase counter. But in certain workloads this could lead
+ * to an unlimited wear of one or few physical eraseblock. Indeed, imagine a
+ * situation when the picked physical eraseblock is constantly erased after the
+ * data is written to it. So, we have a constant which limits the highest erase
+ * counter of the free physical eraseblock to pick. Namely, the WL unit does
+ * not pick eraseblocks with erase counter greater then the lowest erase
+ * counter plus %WL_FREE_MAX_DIFF.
+ */
+#define WL_FREE_MAX_DIFF (2*UBI_WL_THRESHOLD)
+
+/*
+ * Maximum number of consecutive background thread failures which is enough to
+ * switch to read-only mode.
+ */
+#define WL_MAX_FAILURES 32
+
+/**
+ * struct ubi_wl_prot_entry - PEB protection entry.
+ * @rb_pnum: link in the @wl->prot.pnum RB-tree
+ * @rb_aec: link in the @wl->prot.aec RB-tree
+ * @abs_ec: the absolute erase counter value when the protection ends
+ * @e: the wear-leveling entry of the physical eraseblock under protection
+ *
+ * When the WL unit returns a physical eraseblock, the physical eraseblock is
+ * protected from being moved for some "time". For this reason, the physical
+ * eraseblock is not directly moved from the @wl->free tree to the @wl->used
+ * tree. There is one more tree in between where this physical eraseblock is
+ * temporarily stored (@wl->prot).
+ *
+ * All this protection stuff is needed because:
+ *  o we don't want to move physical eraseblocks just after we have given them
+ *    to the user; instead, we first want to let users fill them up with data;
+ *
+ *  o there is a chance that the user will put the physical eraseblock very
+ *    soon, so it makes sense not to move it for some time, but wait; this is
+ *    especially important in case of "short term" physical eraseblocks.
+ *
+ * Physical eraseblocks stay protected only for limited time. But the "time" is
+ * measured in erase cycles in this case. This is implemented with help of the
+ * absolute erase counter (@wl->abs_ec). When it reaches certain value, the
+ * physical eraseblocks are moved from the protection trees (@wl->prot.*) to
+ * the @wl->used tree.
+ *
+ * Protected physical eraseblocks are searched by physical eraseblock number
+ * (when they are put) and by the absolute erase counter (to check if it is
+ * time to move them to the @wl->used tree). So there are actually 2 RB-trees
+ * storing the protected physical eraseblocks: @wl->prot.pnum and
+ * @wl->prot.aec. They are referred to as the "protection" trees. The
+ * first one is indexed by the physical eraseblock number. The second one is
+ * indexed by the absolute erase counter. Both trees store
+ * &struct ubi_wl_prot_entry objects.
+ *
+ * Each physical eraseblock has 2 main states: free and used. The former state
+ * corresponds to the @wl->free tree. The latter state is split up on several
+ * sub-states:
+ * o the WL movement is allowed (@wl->used tree);
+ * o the WL movement is temporarily prohibited (@wl->prot.pnum and
+ * @wl->prot.aec trees);
+ * o scrubbing is needed (@wl->scrub tree).
+ *
+ * Depending on the sub-state, wear-leveling entries of the used physical
+ * eraseblocks may be kept in one of those trees.
+ */
+struct ubi_wl_prot_entry {
+	struct rb_node rb_pnum;
+	struct rb_node rb_aec;
+	unsigned long long abs_ec;
+	struct ubi_wl_entry *e;
+};
+
+/**
+ * struct ubi_work - UBI work description data structure.
+ * @list: a link in the list of pending works
+ * @func: worker function
+ * @priv: private data of the worker function
+ *
+ * @e: physical eraseblock to erase
+ * @torture: if the physical eraseblock has to be tortured
+ *
+ * The @func pointer points to the worker function. If the @cancel argument is
+ * not zero, the worker has to free the resources and exit immediately. The
+ * worker has to return zero in case of success and a negative error code in
+ * case of failure.
+ */
+struct ubi_work {
+	struct list_head list;
+	int (*func)(struct ubi_device *ubi, struct ubi_work *wrk, int cancel);
+	/* The below fields are only relevant to erasure works */
+	struct ubi_wl_entry *e;
+	int torture;
+};
+
+#ifdef CONFIG_MTD_UBI_DEBUG_PARANOID
+static int paranoid_check_ec(struct ubi_device *ubi, int pnum, int ec);
+static int paranoid_check_in_wl_tree(struct ubi_wl_entry *e,
+				     struct rb_root *root);
+#else
+#define paranoid_check_ec(ubi, pnum, ec) 0
+#define paranoid_check_in_wl_tree(e, root)
+#endif
+
+/**
+ * wl_tree_add - add a wear-leveling entry to a WL RB-tree.
+ * @e: the wear-leveling entry to add
+ * @root: the root of the tree
+ *
+ * Note, we use (erase counter, physical eraseblock number) pairs as keys in
+ * the @ubi->used and @ubi->free RB-trees.
+ */
+static void wl_tree_add(struct ubi_wl_entry *e, struct rb_root *root)
+{
+	struct rb_node **p, *parent = NULL;
+
+	p = &root->rb_node;
+	while (*p) {
+		struct ubi_wl_entry *e1;
+
+		parent = *p;
+		e1 = rb_entry(parent, struct ubi_wl_entry, rb);
+
+		if (e->ec < e1->ec)
+			p = &(*p)->rb_left;
+		else if (e->ec > e1->ec)
+			p = &(*p)->rb_right;
+		else {
+			ubi_assert(e->pnum != e1->pnum);
+			if (e->pnum < e1->pnum)
+				p = &(*p)->rb_left;
+			else
+				p = &(*p)->rb_right;
+		}
+	}
+
+	rb_link_node(&e->rb, parent, p);
+	rb_insert_color(&e->rb, root);
+}
+
+/**
+ * do_work - do one pending work.
+ * @ubi: UBI device description object
+ *
+ * This function returns zero in case of success and a negative error code in
+ * case of failure.
+ */
+static int do_work(struct ubi_device *ubi)
+{
+	int err;
+	struct ubi_work *wrk;
+
+	cond_resched();
+
+	/*
+	 * @ubi->work_sem is used to synchronize with the workers. Workers take
+	 * it in read mode, so many of them may be doing works at a time. But
+	 * the queue flush code has to be sure the whole queue of works is
+	 * done, and it takes the mutex in write mode.
+	 */
+	down_read(&ubi->work_sem);
+	spin_lock(&ubi->wl_lock);
+	if (list_empty(&ubi->works)) {
+		spin_unlock(&ubi->wl_lock);
+		up_read(&ubi->work_sem);
+		return 0;
+	}
+
+	wrk = list_entry(ubi->works.next, struct ubi_work, list);
+	list_del(&wrk->list);
+	ubi->works_count -= 1;
+	ubi_assert(ubi->works_count >= 0);
+	spin_unlock(&ubi->wl_lock);
+
+	/*
+	 * Call the worker function. Do not touch the work structure
+	 * after this call as it will have been freed or reused by that
+	 * time by the worker function.
+	 */
+	err = wrk->func(ubi, wrk, 0);
+	if (err)
+		ubi_err("work failed with error code %d", err);
+	up_read(&ubi->work_sem);
+
+	return err;
+}
+
+/**
+ * produce_free_peb - produce a free physical eraseblock.
+ * @ubi: UBI device description object
+ *
+ * This function tries to make a free PEB by means of synchronous execution of
+ * pending works. This may be needed if, for example the background thread is
+ * disabled. Returns zero in case of success and a negative error code in case
+ * of failure.
+ */
+static int produce_free_peb(struct ubi_device *ubi)
+{
+	int err;
+
+	spin_lock(&ubi->wl_lock);
+	while (!ubi->free.rb_node) {
+		spin_unlock(&ubi->wl_lock);
+
+		dbg_wl("do one work synchronously");
+		err = do_work(ubi);
+		if (err)
+			return err;
+
+		spin_lock(&ubi->wl_lock);
+	}
+	spin_unlock(&ubi->wl_lock);
+
+	return 0;
+}
+
+/**
+ * in_wl_tree - check if wear-leveling entry is present in a WL RB-tree.
+ * @e: the wear-leveling entry to check
+ * @root: the root of the tree
+ *
+ * This function returns non-zero if @e is in the @root RB-tree and zero if it
+ * is not.
+ */
+static int in_wl_tree(struct ubi_wl_entry *e, struct rb_root *root)
+{
+	struct rb_node *p;
+
+	p = root->rb_node;
+	while (p) {
+		struct ubi_wl_entry *e1;
+
+		e1 = rb_entry(p, struct ubi_wl_entry, rb);
+
+		if (e->pnum == e1->pnum) {
+			ubi_assert(e == e1);
+			return 1;
+		}
+
+		if (e->ec < e1->ec)
+			p = p->rb_left;
+		else if (e->ec > e1->ec)
+			p = p->rb_right;
+		else {
+			ubi_assert(e->pnum != e1->pnum);
+			if (e->pnum < e1->pnum)
+				p = p->rb_left;
+			else
+				p = p->rb_right;
+		}
+	}
+
+	return 0;
+}
+
+/**
+ * prot_tree_add - add physical eraseblock to protection trees.
+ * @ubi: UBI device description object
+ * @e: the physical eraseblock to add
+ * @pe: protection entry object to use
+ * @abs_ec: absolute erase counter value when this physical eraseblock has
+ * to be removed from the protection trees.
+ *
+ * @wl->lock has to be locked.
+ */
+static void prot_tree_add(struct ubi_device *ubi, struct ubi_wl_entry *e,
+			  struct ubi_wl_prot_entry *pe, int abs_ec)
+{
+	struct rb_node **p, *parent = NULL;
+	struct ubi_wl_prot_entry *pe1;
+
+	pe->e = e;
+	pe->abs_ec = ubi->abs_ec + abs_ec;
+
+	p = &ubi->prot.pnum.rb_node;
+	while (*p) {
+		parent = *p;
+		pe1 = rb_entry(parent, struct ubi_wl_prot_entry, rb_pnum);
+
+		if (e->pnum < pe1->e->pnum)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+	rb_link_node(&pe->rb_pnum, parent, p);
+	rb_insert_color(&pe->rb_pnum, &ubi->prot.pnum);
+
+	p = &ubi->prot.aec.rb_node;
+	parent = NULL;
+	while (*p) {
+		parent = *p;
+		pe1 = rb_entry(parent, struct ubi_wl_prot_entry, rb_aec);
+
+		if (pe->abs_ec < pe1->abs_ec)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+	rb_link_node(&pe->rb_aec, parent, p);
+	rb_insert_color(&pe->rb_aec, &ubi->prot.aec);
+}
+
+/**
+ * find_wl_entry - find wear-leveling entry closest to certain erase counter.
+ * @root: the RB-tree where to look for
+ * @max: highest possible erase counter
+ *
+ * This function looks for a wear leveling entry with erase counter closest to
+ * @max and less then @max.
+ */
+static struct ubi_wl_entry *find_wl_entry(struct rb_root *root, int max)
+{
+	struct rb_node *p;
+	struct ubi_wl_entry *e;
+
+	e = rb_entry(rb_first(root), struct ubi_wl_entry, rb);
+	max += e->ec;
+
+	p = root->rb_node;
+	while (p) {
+		struct ubi_wl_entry *e1;
+
+		e1 = rb_entry(p, struct ubi_wl_entry, rb);
+		if (e1->ec >= max)
+			p = p->rb_left;
+		else {
+			p = p->rb_right;
+			e = e1;
+		}
+	}
+
+	return e;
+}
+
+/**
+ * ubi_wl_get_peb - get a physical eraseblock.
+ * @ubi: UBI device description object
+ * @dtype: type of data which will be stored in this physical eraseblock
+ *
+ * This function returns a physical eraseblock in case of success and a
+ * negative error code in case of failure. Might sleep.
+ */
+int ubi_wl_get_peb(struct ubi_device *ubi, int dtype)
+{
+	int err, protect, medium_ec;
+	struct ubi_wl_entry *e, *first, *last;
+	struct ubi_wl_prot_entry *pe;
+
+	ubi_assert(dtype == UBI_LONGTERM || dtype == UBI_SHORTTERM ||
+		   dtype == UBI_UNKNOWN);
+
+	pe = kmalloc(sizeof(struct ubi_wl_prot_entry), GFP_NOFS);
+	if (!pe)
+		return -ENOMEM;
+
+retry:
+	spin_lock(&ubi->wl_lock);
+	if (!ubi->free.rb_node) {
+		if (ubi->works_count == 0) {
+			ubi_assert(list_empty(&ubi->works));
+			ubi_err("no free eraseblocks");
+			spin_unlock(&ubi->wl_lock);
+			kfree(pe);
+			return -ENOSPC;
+		}
+		spin_unlock(&ubi->wl_lock);
+
+		err = produce_free_peb(ubi);
+		if (err < 0) {
+			kfree(pe);
+			return err;
+		}
+		goto retry;
+	}
+
+	switch (dtype) {
+		case UBI_LONGTERM:
+			/*
+			 * For long term data we pick a physical eraseblock
+			 * with high erase counter. But the highest erase
+			 * counter we can pick is bounded by the the lowest
+			 * erase counter plus %WL_FREE_MAX_DIFF.
+			 */
+			e = find_wl_entry(&ubi->free, WL_FREE_MAX_DIFF);
+			protect = LT_PROTECTION;
+			break;
+		case UBI_UNKNOWN:
+			/*
+			 * For unknown data we pick a physical eraseblock with
+			 * medium erase counter. But we by no means can pick a
+			 * physical eraseblock with erase counter greater or
+			 * equivalent than the lowest erase counter plus
+			 * %WL_FREE_MAX_DIFF.
+			 */
+			first = rb_entry(rb_first(&ubi->free),
+					 struct ubi_wl_entry, rb);
+			last = rb_entry(rb_last(&ubi->free),
+					struct ubi_wl_entry, rb);
+
+			if (last->ec - first->ec < WL_FREE_MAX_DIFF)
+				e = rb_entry(ubi->free.rb_node,
+						struct ubi_wl_entry, rb);
+			else {
+				medium_ec = (first->ec + WL_FREE_MAX_DIFF)/2;
+				e = find_wl_entry(&ubi->free, medium_ec);
+			}
+			protect = U_PROTECTION;
+			break;
+		case UBI_SHORTTERM:
+			/*
+			 * For short term data we pick a physical eraseblock
+			 * with the lowest erase counter as we expect it will
+			 * be erased soon.
+			 */
+			e = rb_entry(rb_first(&ubi->free),
+				     struct ubi_wl_entry, rb);
+			protect = ST_PROTECTION;
+			break;
+		default:
+			protect = 0;
+			e = NULL;
+			BUG();
+	}
+
+	/*
+	 * Move the physical eraseblock to the protection trees where it will
+	 * be protected from being moved for some time.
+	 */
+	paranoid_check_in_wl_tree(e, &ubi->free);
+	rb_erase(&e->rb, &ubi->free);
+	prot_tree_add(ubi, e, pe, protect);
+
+	dbg_wl("PEB %d EC %d, protection %d", e->pnum, e->ec, protect);
+	spin_unlock(&ubi->wl_lock);
+
+	return e->pnum;
+}
+
+/**
+ * prot_tree_del - remove a physical eraseblock from the protection trees
+ * @ubi: UBI device description object
+ * @pnum: the physical eraseblock to remove
+ *
+ * This function returns PEB @pnum from the protection trees and returns zero
+ * in case of success and %-ENODEV if the PEB was not found in the protection
+ * trees.
+ */
+static int prot_tree_del(struct ubi_device *ubi, int pnum)
+{
+	struct rb_node *p;
+	struct ubi_wl_prot_entry *pe = NULL;
+
+	p = ubi->prot.pnum.rb_node;
+	while (p) {
+
+		pe = rb_entry(p, struct ubi_wl_prot_entry, rb_pnum);
+
+		if (pnum == pe->e->pnum)
+			goto found;
+
+		if (pnum < pe->e->pnum)
+			p = p->rb_left;
+		else
+			p = p->rb_right;
+	}
+
+	return -ENODEV;
+
+found:
+	ubi_assert(pe->e->pnum == pnum);
+	rb_erase(&pe->rb_aec, &ubi->prot.aec);
+	rb_erase(&pe->rb_pnum, &ubi->prot.pnum);
+	kfree(pe);
+	return 0;
+}
+
+/**
+ * sync_erase - synchronously erase a physical eraseblock.
+ * @ubi: UBI device description object
+ * @e: the the physical eraseblock to erase
+ * @torture: if the physical eraseblock has to be tortured
+ *
+ * This function returns zero in case of success and a negative error code in
+ * case of failure.
+ */
+static int sync_erase(struct ubi_device *ubi, struct ubi_wl_entry *e, int torture)
+{
+	int err;
+	struct ubi_ec_hdr *ec_hdr;
+	unsigned long long ec = e->ec;
+
+	dbg_wl("erase PEB %d, old EC %llu", e->pnum, ec);
+
+	err = paranoid_check_ec(ubi, e->pnum, e->ec);
+	if (err > 0)
+		return -EINVAL;
+
+	ec_hdr = kzalloc(ubi->ec_hdr_alsize, GFP_NOFS);
+	if (!ec_hdr)
+		return -ENOMEM;
+
+	err = ubi_io_sync_erase(ubi, e->pnum, torture);
+	if (err < 0)
+		goto out_free;
+
+	ec += err;
+	if (ec > UBI_MAX_ERASECOUNTER) {
+		/*
+		 * Erase counter overflow. Upgrade UBI and use 64-bit
+		 * erase counters internally.
+		 */
+		ubi_err("erase counter overflow at PEB %d, EC %llu",
+			e->pnum, ec);
+		err = -EINVAL;
+		goto out_free;
+	}
+
+	dbg_wl("erased PEB %d, new EC %llu", e->pnum, ec);
+
+	ec_hdr->ec = cpu_to_be64(ec);
+
+	err = ubi_io_write_ec_hdr(ubi, e->pnum, ec_hdr);
+	if (err)
+		goto out_free;
+
+	e->ec = ec;
+	spin_lock(&ubi->wl_lock);
+	if (e->ec > ubi->max_ec)
+		ubi->max_ec = e->ec;
+	spin_unlock(&ubi->wl_lock);
+
+out_free:
+	kfree(ec_hdr);
+	return err;
+}
+
+/**
+ * check_protection_over - check if it is time to stop protecting some
+ * physical eraseblocks.
+ * @ubi: UBI device description object
+ *
+ * This function is called after each erase operation, when the absolute erase
+ * counter is incremented, to check if some physical eraseblock  have not to be
+ * protected any longer. These physical eraseblocks are moved from the
+ * protection trees to the used tree.
+ */
+static void check_protection_over(struct ubi_device *ubi)
+{
+	struct ubi_wl_prot_entry *pe;
+
+	/*
+	 * There may be several protected physical eraseblock to remove,
+	 * process them all.
+	 */
+	while (1) {
+		spin_lock(&ubi->wl_lock);
+		if (!ubi->prot.aec.rb_node) {
+			spin_unlock(&ubi->wl_lock);
+			break;
+		}
+
+		pe = rb_entry(rb_first(&ubi->prot.aec),
+			      struct ubi_wl_prot_entry, rb_aec);
+
+		if (pe->abs_ec > ubi->abs_ec) {
+			spin_unlock(&ubi->wl_lock);
+			break;
+		}
+
+		dbg_wl("PEB %d protection over, abs_ec %llu, PEB abs_ec %llu",
+		       pe->e->pnum, ubi->abs_ec, pe->abs_ec);
+		rb_erase(&pe->rb_aec, &ubi->prot.aec);
+		rb_erase(&pe->rb_pnum, &ubi->prot.pnum);
+		wl_tree_add(pe->e, &ubi->used);
+		spin_unlock(&ubi->wl_lock);
+
+		kfree(pe);
+		cond_resched();
+	}
+}
+
+/**
+ * schedule_ubi_work - schedule a work.
+ * @ubi: UBI device description object
+ * @wrk: the work to schedule
+ *
+ * This function enqueues a work defined by @wrk to the tail of the pending
+ * works list.
+ */
+static void schedule_ubi_work(struct ubi_device *ubi, struct ubi_work *wrk)
+{
+	spin_lock(&ubi->wl_lock);
+	list_add_tail(&wrk->list, &ubi->works);
+	ubi_assert(ubi->works_count >= 0);
+	ubi->works_count += 1;
+	if (ubi->thread_enabled)
+		wake_up_process(ubi->bgt_thread);
+	spin_unlock(&ubi->wl_lock);
+}
+
+static int erase_worker(struct ubi_device *ubi, struct ubi_work *wl_wrk,
+			int cancel);
+
+/**
+ * schedule_erase - schedule an erase work.
+ * @ubi: UBI device description object
+ * @e: the WL entry of the physical eraseblock to erase
+ * @torture: if the physical eraseblock has to be tortured
+ *
+ * This function returns zero in case of success and a %-ENOMEM in case of
+ * failure.
+ */
+static int schedule_erase(struct ubi_device *ubi, struct ubi_wl_entry *e,
+			  int torture)
+{
+	struct ubi_work *wl_wrk;
+
+	dbg_wl("schedule erasure of PEB %d, EC %d, torture %d",
+	       e->pnum, e->ec, torture);
+
+	wl_wrk = kmalloc(sizeof(struct ubi_work), GFP_NOFS);
+	if (!wl_wrk)
+		return -ENOMEM;
+
+	wl_wrk->func = &erase_worker;
+	wl_wrk->e = e;
+	wl_wrk->torture = torture;
+
+	schedule_ubi_work(ubi, wl_wrk);
+	return 0;
+}
+
+/**
+ * wear_leveling_worker - wear-leveling worker function.
+ * @ubi: UBI device description object
+ * @wrk: the work object
+ * @cancel: non-zero if the worker has to free memory and exit
+ *
+ * This function copies a more worn out physical eraseblock to a less worn out
+ * one. Returns zero in case of success and a negative error code in case of
+ * failure.
+ */
+static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk,
+				int cancel)
+{
+	int err, put = 0, scrubbing = 0, protect = 0;
+	struct ubi_wl_prot_entry *uninitialized_var(pe);
+	struct ubi_wl_entry *e1, *e2;
+	struct ubi_vid_hdr *vid_hdr;
+
+	kfree(wrk);
+
+	if (cancel)
+		return 0;
+
+	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
+	if (!vid_hdr)
+		return -ENOMEM;
+
+	mutex_lock(&ubi->move_mutex);
+	spin_lock(&ubi->wl_lock);
+	ubi_assert(!ubi->move_from && !ubi->move_to);
+	ubi_assert(!ubi->move_to_put);
+
+	if (!ubi->free.rb_node ||
+	    (!ubi->used.rb_node && !ubi->scrub.rb_node)) {
+		/*
+		 * No free physical eraseblocks? Well, they must be waiting in
+		 * the queue to be erased. Cancel movement - it will be
+		 * triggered again when a free physical eraseblock appears.
+		 *
+		 * No used physical eraseblocks? They must be temporarily
+		 * protected from being moved. They will be moved to the
+		 * @ubi->used tree later and the wear-leveling will be
+		 * triggered again.
+		 */
+		dbg_wl("cancel WL, a list is empty: free %d, used %d",
+		       !ubi->free.rb_node, !ubi->used.rb_node);
+		goto out_cancel;
+	}
+
+	if (!ubi->scrub.rb_node) {
+		/*
+		 * Now pick the least worn-out used physical eraseblock and a
+		 * highly worn-out free physical eraseblock. If the erase
+		 * counters differ much enough, start wear-leveling.
+		 */
+		e1 = rb_entry(rb_first(&ubi->used), struct ubi_wl_entry, rb);
+		e2 = find_wl_entry(&ubi->free, WL_FREE_MAX_DIFF);
+
+		if (!(e2->ec - e1->ec >= UBI_WL_THRESHOLD)) {
+			dbg_wl("no WL needed: min used EC %d, max free EC %d",
+			       e1->ec, e2->ec);
+			goto out_cancel;
+		}
+		paranoid_check_in_wl_tree(e1, &ubi->used);
+		rb_erase(&e1->rb, &ubi->used);
+		dbg_wl("move PEB %d EC %d to PEB %d EC %d",
+		       e1->pnum, e1->ec, e2->pnum, e2->ec);
+	} else {
+		/* Perform scrubbing */
+		scrubbing = 1;
+		e1 = rb_entry(rb_first(&ubi->scrub), struct ubi_wl_entry, rb);
+		e2 = find_wl_entry(&ubi->free, WL_FREE_MAX_DIFF);
+		paranoid_check_in_wl_tree(e1, &ubi->scrub);
+		rb_erase(&e1->rb, &ubi->scrub);
+		dbg_wl("scrub PEB %d to PEB %d", e1->pnum, e2->pnum);
+	}
+
+	paranoid_check_in_wl_tree(e2, &ubi->free);
+	rb_erase(&e2->rb, &ubi->free);
+	ubi->move_from = e1;
+	ubi->move_to = e2;
+	spin_unlock(&ubi->wl_lock);
+
+	/*
+	 * Now we are going to copy physical eraseblock @e1->pnum to @e2->pnum.
+	 * We so far do not know which logical eraseblock our physical
+	 * eraseblock (@e1) belongs to. We have to read the volume identifier
+	 * header first.
+	 *
+	 * Note, we are protected from this PEB being unmapped and erased. The
+	 * 'ubi_wl_put_peb()' would wait for moving to be finished if the PEB
+	 * which is being moved was unmapped.
+	 */
+
+	err = ubi_io_read_vid_hdr(ubi, e1->pnum, vid_hdr, 0);
+	if (err && err != UBI_IO_BITFLIPS) {
+		if (err == UBI_IO_PEB_FREE) {
+			/*
+			 * We are trying to move PEB without a VID header. UBI
+			 * always write VID headers shortly after the PEB was
+			 * given, so we have a situation when it did not have
+			 * chance to write it down because it was preempted.
+			 * Just re-schedule the work, so that next time it will
+			 * likely have the VID header in place.
+			 */
+			dbg_wl("PEB %d has no VID header", e1->pnum);
+			goto out_not_moved;
+		}
+
+		ubi_err("error %d while reading VID header from PEB %d",
+			err, e1->pnum);
+		if (err > 0)
+			err = -EIO;
+		goto out_error;
+	}
+
+	err = ubi_eba_copy_leb(ubi, e1->pnum, e2->pnum, vid_hdr);
+	if (err) {
+
+		if (err < 0)
+			goto out_error;
+		if (err == 1)
+			goto out_not_moved;
+
+		/*
+		 * For some reason the LEB was not moved - it might be because
+		 * the volume is being deleted. We should prevent this PEB from
+		 * being selected for wear-levelling movement for some "time",
+		 * so put it to the protection tree.
+		 */
+
+		dbg_wl("cancelled moving PEB %d", e1->pnum);
+		pe = kmalloc(sizeof(struct ubi_wl_prot_entry), GFP_NOFS);
+		if (!pe) {
+			err = -ENOMEM;
+			goto out_error;
+		}
+
+		protect = 1;
+	}
+
+	ubi_free_vid_hdr(ubi, vid_hdr);
+	spin_lock(&ubi->wl_lock);
+	if (protect)
+		prot_tree_add(ubi, e1, pe, protect);
+	if (!ubi->move_to_put)
+		wl_tree_add(e2, &ubi->used);
+	else
+		put = 1;
+	ubi->move_from = ubi->move_to = NULL;
+	ubi->move_to_put = ubi->wl_scheduled = 0;
+	spin_unlock(&ubi->wl_lock);
+
+	if (put) {
+		/*
+		 * Well, the target PEB was put meanwhile, schedule it for
+		 * erasure.
+		 */
+		dbg_wl("PEB %d was put meanwhile, erase", e2->pnum);
+		err = schedule_erase(ubi, e2, 0);
+		if (err)
+			goto out_error;
+	}
+
+	if (!protect) {
+		err = schedule_erase(ubi, e1, 0);
+		if (err)
+			goto out_error;
+	}
+
+
+	dbg_wl("done");
+	mutex_unlock(&ubi->move_mutex);
+	return 0;
+
+	/*
+	 * For some reasons the LEB was not moved, might be an error, might be
+	 * something else. @e1 was not changed, so return it back. @e2 might
+	 * be changed, schedule it for erasure.
+	 */
+out_not_moved:
+	ubi_free_vid_hdr(ubi, vid_hdr);
+	spin_lock(&ubi->wl_lock);
+	if (scrubbing)
+		wl_tree_add(e1, &ubi->scrub);
+	else
+		wl_tree_add(e1, &ubi->used);
+	ubi->move_from = ubi->move_to = NULL;
+	ubi->move_to_put = ubi->wl_scheduled = 0;
+	spin_unlock(&ubi->wl_lock);
+
+	err = schedule_erase(ubi, e2, 0);
+	if (err)
+		goto out_error;
+
+	mutex_unlock(&ubi->move_mutex);
+	return 0;
+
+out_error:
+	ubi_err("error %d while moving PEB %d to PEB %d",
+		err, e1->pnum, e2->pnum);
+
+	ubi_free_vid_hdr(ubi, vid_hdr);
+	spin_lock(&ubi->wl_lock);
+	ubi->move_from = ubi->move_to = NULL;
+	ubi->move_to_put = ubi->wl_scheduled = 0;
+	spin_unlock(&ubi->wl_lock);
+
+	kmem_cache_free(ubi_wl_entry_slab, e1);
+	kmem_cache_free(ubi_wl_entry_slab, e2);
+	ubi_ro_mode(ubi);
+
+	mutex_unlock(&ubi->move_mutex);
+	return err;
+
+out_cancel:
+	ubi->wl_scheduled = 0;
+	spin_unlock(&ubi->wl_lock);
+	mutex_unlock(&ubi->move_mutex);
+	ubi_free_vid_hdr(ubi, vid_hdr);
+	return 0;
+}
+
+/**
+ * ensure_wear_leveling - schedule wear-leveling if it is needed.
+ * @ubi: UBI device description object
+ *
+ * This function checks if it is time to start wear-leveling and schedules it
+ * if yes. This function returns zero in case of success and a negative error
+ * code in case of failure.
+ */
+static int ensure_wear_leveling(struct ubi_device *ubi)
+{
+	int err = 0;
+	struct ubi_wl_entry *e1;
+	struct ubi_wl_entry *e2;
+	struct ubi_work *wrk;
+
+	spin_lock(&ubi->wl_lock);
+	if (ubi->wl_scheduled)
+		/* Wear-leveling is already in the work queue */
+		goto out_unlock;
+
+	/*
+	 * If the ubi->scrub tree is not empty, scrubbing is needed, and the
+	 * the WL worker has to be scheduled anyway.
+	 */
+	if (!ubi->scrub.rb_node) {
+		if (!ubi->used.rb_node || !ubi->free.rb_node)
+			/* No physical eraseblocks - no deal */
+			goto out_unlock;
+
+		/*
+		 * We schedule wear-leveling only if the difference between the
+		 * lowest erase counter of used physical eraseblocks and a high
+		 * erase counter of free physical eraseblocks is greater then
+		 * %UBI_WL_THRESHOLD.
+		 */
+		e1 = rb_entry(rb_first(&ubi->used), struct ubi_wl_entry, rb);
+		e2 = find_wl_entry(&ubi->free, WL_FREE_MAX_DIFF);
+
+		if (!(e2->ec - e1->ec >= UBI_WL_THRESHOLD))
+			goto out_unlock;
+		dbg_wl("schedule wear-leveling");
+	} else
+		dbg_wl("schedule scrubbing");
+
+	ubi->wl_scheduled = 1;
+	spin_unlock(&ubi->wl_lock);
+
+	wrk = kmalloc(sizeof(struct ubi_work), GFP_NOFS);
+	if (!wrk) {
+		err = -ENOMEM;
+		goto out_cancel;
+	}
+
+	wrk->func = &wear_leveling_worker;
+	schedule_ubi_work(ubi, wrk);
+	return err;
+
+out_cancel:
+	spin_lock(&ubi->wl_lock);
+	ubi->wl_scheduled = 0;
+out_unlock:
+	spin_unlock(&ubi->wl_lock);
+	return err;
+}
+
+/**
+ * erase_worker - physical eraseblock erase worker function.
+ * @ubi: UBI device description object
+ * @wl_wrk: the work object
+ * @cancel: non-zero if the worker has to free memory and exit
+ *
+ * This function erases a physical eraseblock and perform torture testing if
+ * needed. It also takes care about marking the physical eraseblock bad if
+ * needed. Returns zero in case of success and a negative error code in case of
+ * failure.
+ */
+static int erase_worker(struct ubi_device *ubi, struct ubi_work *wl_wrk,
+			int cancel)
+{
+	struct ubi_wl_entry *e = wl_wrk->e;
+	int pnum = e->pnum, err, need;
+
+	if (cancel) {
+		dbg_wl("cancel erasure of PEB %d EC %d", pnum, e->ec);
+		kfree(wl_wrk);
+		kmem_cache_free(ubi_wl_entry_slab, e);
+		return 0;
+	}
+
+	dbg_wl("erase PEB %d EC %d", pnum, e->ec);
+
+	err = sync_erase(ubi, e, wl_wrk->torture);
+	if (!err) {
+		/* Fine, we've erased it successfully */
+		kfree(wl_wrk);
+
+		spin_lock(&ubi->wl_lock);
+		ubi->abs_ec += 1;
+		wl_tree_add(e, &ubi->free);
+		spin_unlock(&ubi->wl_lock);
+
+		/*
+		 * One more erase operation has happened, take care about protected
+		 * physical eraseblocks.
+		 */
+		check_protection_over(ubi);
+
+		/* And take care about wear-leveling */
+		err = ensure_wear_leveling(ubi);
+		return err;
+	}
+
+	ubi_err("failed to erase PEB %d, error %d", pnum, err);
+	kfree(wl_wrk);
+	kmem_cache_free(ubi_wl_entry_slab, e);
+
+	if (err == -EINTR || err == -ENOMEM || err == -EAGAIN ||
+	    err == -EBUSY) {
+		int err1;
+
+		/* Re-schedule the LEB for erasure */
+		err1 = schedule_erase(ubi, e, 0);
+		if (err1) {
+			err = err1;
+			goto out_ro;
+		}
+		return err;
+	} else if (err != -EIO) {
+		/*
+		 * If this is not %-EIO, we have no idea what to do. Scheduling
+		 * this physical eraseblock for erasure again would cause
+		 * errors again and again. Well, lets switch to RO mode.
+		 */
+		goto out_ro;
+	}
+
+	/* It is %-EIO, the PEB went bad */
+
+	if (!ubi->bad_allowed) {
+		ubi_err("bad physical eraseblock %d detected", pnum);
+		goto out_ro;
+	}
+
+	spin_lock(&ubi->volumes_lock);
+	need = ubi->beb_rsvd_level - ubi->beb_rsvd_pebs + 1;
+	if (need > 0) {
+		need = ubi->avail_pebs >= need ? need : ubi->avail_pebs;
+		ubi->avail_pebs -= need;
+		ubi->rsvd_pebs += need;
+		ubi->beb_rsvd_pebs += need;
+		if (need > 0)
+			ubi_msg("reserve more %d PEBs", need);
+	}
+
+	if (ubi->beb_rsvd_pebs == 0) {
+		spin_unlock(&ubi->volumes_lock);
+		ubi_err("no reserved physical eraseblocks");
+		goto out_ro;
+	}
+
+	spin_unlock(&ubi->volumes_lock);
+	ubi_msg("mark PEB %d as bad", pnum);
+
+	err = ubi_io_mark_bad(ubi, pnum);
+	if (err)
+		goto out_ro;
+
+	spin_lock(&ubi->volumes_lock);
+	ubi->beb_rsvd_pebs -= 1;
+	ubi->bad_peb_count += 1;
+	ubi->good_peb_count -= 1;
+	ubi_calculate_reserved(ubi);
+	if (ubi->beb_rsvd_pebs == 0)
+		ubi_warn("last PEB from the reserved pool was used");
+	spin_unlock(&ubi->volumes_lock);
+
+	return err;
+
+out_ro:
+	ubi_ro_mode(ubi);
+	return err;
+}
+
+/**
+ * ubi_wl_put_peb - return a physical eraseblock to the wear-leveling unit.
+ * @ubi: UBI device description object
+ * @pnum: physical eraseblock to return
+ * @torture: if this physical eraseblock has to be tortured
+ *
+ * This function is called to return physical eraseblock @pnum to the pool of
+ * free physical eraseblocks. The @torture flag has to be set if an I/O error
+ * occurred to this @pnum and it has to be tested. This function returns zero
+ * in case of success, and a negative error code in case of failure.
+ */
+int ubi_wl_put_peb(struct ubi_device *ubi, int pnum, int torture)
+{
+	int err;
+	struct ubi_wl_entry *e;
+
+	dbg_wl("PEB %d", pnum);
+	ubi_assert(pnum >= 0);
+	ubi_assert(pnum < ubi->peb_count);
+
+retry:
+	spin_lock(&ubi->wl_lock);
+	e = ubi->lookuptbl[pnum];
+	if (e == ubi->move_from) {
+		/*
+		 * User is putting the physical eraseblock which was selected to
+		 * be moved. It will be scheduled for erasure in the
+		 * wear-leveling worker.
+		 */
+		dbg_wl("PEB %d is being moved, wait", pnum);
+		spin_unlock(&ubi->wl_lock);
+
+		/* Wait for the WL worker by taking the @ubi->move_mutex */
+		mutex_lock(&ubi->move_mutex);
+		mutex_unlock(&ubi->move_mutex);
+		goto retry;
+	} else if (e == ubi->move_to) {
+		/*
+		 * User is putting the physical eraseblock which was selected
+		 * as the target the data is moved to. It may happen if the EBA
+		 * unit already re-mapped the LEB in 'ubi_eba_copy_leb()' but
+		 * the WL unit has not put the PEB to the "used" tree yet, but
+		 * it is about to do this. So we just set a flag which will
+		 * tell the WL worker that the PEB is not needed anymore and
+		 * should be scheduled for erasure.
+		 */
+		dbg_wl("PEB %d is the target of data moving", pnum);
+		ubi_assert(!ubi->move_to_put);
+		ubi->move_to_put = 1;
+		spin_unlock(&ubi->wl_lock);
+		return 0;
+	} else {
+		if (in_wl_tree(e, &ubi->used)) {
+			paranoid_check_in_wl_tree(e, &ubi->used);
+			rb_erase(&e->rb, &ubi->used);
+		} else if (in_wl_tree(e, &ubi->scrub)) {
+			paranoid_check_in_wl_tree(e, &ubi->scrub);
+			rb_erase(&e->rb, &ubi->scrub);
+		} else {
+			err = prot_tree_del(ubi, e->pnum);
+			if (err) {
+				ubi_err("PEB %d not found", pnum);
+				ubi_ro_mode(ubi);
+				spin_unlock(&ubi->wl_lock);
+				return err;
+			}
+		}
+	}
+	spin_unlock(&ubi->wl_lock);
+
+	err = schedule_erase(ubi, e, torture);
+	if (err) {
+		spin_lock(&ubi->wl_lock);
+		wl_tree_add(e, &ubi->used);
+		spin_unlock(&ubi->wl_lock);
+	}
+
+	return err;
+}
+
+/**
+ * ubi_wl_scrub_peb - schedule a physical eraseblock for scrubbing.
+ * @ubi: UBI device description object
+ * @pnum: the physical eraseblock to schedule
+ *
+ * If a bit-flip in a physical eraseblock is detected, this physical eraseblock
+ * needs scrubbing. This function schedules a physical eraseblock for
+ * scrubbing which is done in background. This function returns zero in case of
+ * success and a negative error code in case of failure.
+ */
+int ubi_wl_scrub_peb(struct ubi_device *ubi, int pnum)
+{
+	struct ubi_wl_entry *e;
+
+	ubi_msg("schedule PEB %d for scrubbing", pnum);
+
+retry:
+	spin_lock(&ubi->wl_lock);
+	e = ubi->lookuptbl[pnum];
+	if (e == ubi->move_from || in_wl_tree(e, &ubi->scrub)) {
+		spin_unlock(&ubi->wl_lock);
+		return 0;
+	}
+
+	if (e == ubi->move_to) {
+		/*
+		 * This physical eraseblock was used to move data to. The data
+		 * was moved but the PEB was not yet inserted to the proper
+		 * tree. We should just wait a little and let the WL worker
+		 * proceed.
+		 */
+		spin_unlock(&ubi->wl_lock);
+		dbg_wl("the PEB %d is not in proper tree, retry", pnum);
+		yield();
+		goto retry;
+	}
+
+	if (in_wl_tree(e, &ubi->used)) {
+		paranoid_check_in_wl_tree(e, &ubi->used);
+		rb_erase(&e->rb, &ubi->used);
+	} else {
+		int err;
+
+		err = prot_tree_del(ubi, e->pnum);
+		if (err) {
+			ubi_err("PEB %d not found", pnum);
+			ubi_ro_mode(ubi);
+			spin_unlock(&ubi->wl_lock);
+			return err;
+		}
+	}
+
+	wl_tree_add(e, &ubi->scrub);
+	spin_unlock(&ubi->wl_lock);
+
+	/*
+	 * Technically scrubbing is the same as wear-leveling, so it is done
+	 * by the WL worker.
+	 */
+	return ensure_wear_leveling(ubi);
+}
+
+/**
+ * ubi_wl_flush - flush all pending works.
+ * @ubi: UBI device description object
+ *
+ * This function returns zero in case of success and a negative error code in
+ * case of failure.
+ */
+int ubi_wl_flush(struct ubi_device *ubi)
+{
+	int err;
+
+	/*
+	 * Erase while the pending works queue is not empty, but not more then
+	 * the number of currently pending works.
+	 */
+	dbg_wl("flush (%d pending works)", ubi->works_count);
+	while (ubi->works_count) {
+		err = do_work(ubi);
+		if (err)
+			return err;
+	}
+
+	/*
+	 * Make sure all the works which have been done in parallel are
+	 * finished.
+	 */
+	down_write(&ubi->work_sem);
+	up_write(&ubi->work_sem);
+
+	/*
+	 * And in case last was the WL worker and it cancelled the LEB
+	 * movement, flush again.
+	 */
+	while (ubi->works_count) {
+		dbg_wl("flush more (%d pending works)", ubi->works_count);
+		err = do_work(ubi);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+/**
+ * tree_destroy - destroy an RB-tree.
+ * @root: the root of the tree to destroy
+ */
+static void tree_destroy(struct rb_root *root)
+{
+	struct rb_node *rb;
+	struct ubi_wl_entry *e;
+
+	rb = root->rb_node;
+	while (rb) {
+		if (rb->rb_left)
+			rb = rb->rb_left;
+		else if (rb->rb_right)
+			rb = rb->rb_right;
+		else {
+			e = rb_entry(rb, struct ubi_wl_entry, rb);
+
+			rb = rb_parent(rb);
+			if (rb) {
+				if (rb->rb_left == &e->rb)
+					rb->rb_left = NULL;
+				else
+					rb->rb_right = NULL;
+			}
+
+			kmem_cache_free(ubi_wl_entry_slab, e);
+		}
+	}
+}
+
+/**
+ * ubi_thread - UBI background thread.
+ * @u: the UBI device description object pointer
+ */
+int ubi_thread(void *u)
+{
+	int failures = 0;
+	struct ubi_device *ubi = u;
+
+	ubi_msg("background thread \"%s\" started, PID %d",
+		ubi->bgt_name, task_pid_nr(current));
+
+	set_freezable();
+	for (;;) {
+		int err;
+
+		if (kthread_should_stop())
+			break;
+
+		if (try_to_freeze())
+			continue;
+
+		spin_lock(&ubi->wl_lock);
+		if (list_empty(&ubi->works) || ubi->ro_mode ||
+			       !ubi->thread_enabled) {
+			set_current_state(TASK_INTERRUPTIBLE);
+			spin_unlock(&ubi->wl_lock);
+			schedule();
+			continue;
+		}
+		spin_unlock(&ubi->wl_lock);
+
+		err = do_work(ubi);
+		if (err) {
+			ubi_err("%s: work failed with error code %d",
+				ubi->bgt_name, err);
+			if (failures++ > WL_MAX_FAILURES) {
+				/*
+				 * Too many failures, disable the thread and
+				 * switch to read-only mode.
+				 */
+				ubi_msg("%s: %d consecutive failures",
+					ubi->bgt_name, WL_MAX_FAILURES);
+				ubi_ro_mode(ubi);
+				break;
+			}
+		} else
+			failures = 0;
+
+		cond_resched();
+	}
+
+	dbg_wl("background thread \"%s\" is killed", ubi->bgt_name);
+	return 0;
+}
+
+/**
+ * cancel_pending - cancel all pending works.
+ * @ubi: UBI device description object
+ */
+static void cancel_pending(struct ubi_device *ubi)
+{
+	while (!list_empty(&ubi->works)) {
+		struct ubi_work *wrk;
+
+		wrk = list_entry(ubi->works.next, struct ubi_work, list);
+		list_del(&wrk->list);
+		wrk->func(ubi, wrk, 1);
+		ubi->works_count -= 1;
+		ubi_assert(ubi->works_count >= 0);
+	}
+}
+
+/**
+ * ubi_wl_init_scan - initialize the wear-leveling unit using scanning
+ * information.
+ * @ubi: UBI device description object
+ * @si: scanning information
+ *
+ * This function returns zero in case of success, and a negative error code in
+ * case of failure.
+ */
+int ubi_wl_init_scan(struct ubi_device *ubi, struct ubi_scan_info *si)
+{
+	int err;
+	struct rb_node *rb1, *rb2;
+	struct ubi_scan_volume *sv;
+	struct ubi_scan_leb *seb, *tmp;
+	struct ubi_wl_entry *e;
+
+
+	ubi->used = ubi->free = ubi->scrub = RB_ROOT;
+	ubi->prot.pnum = ubi->prot.aec = RB_ROOT;
+	spin_lock_init(&ubi->wl_lock);
+	mutex_init(&ubi->move_mutex);
+	init_rwsem(&ubi->work_sem);
+	ubi->max_ec = si->max_ec;
+	INIT_LIST_HEAD(&ubi->works);
+
+	sprintf(ubi->bgt_name, UBI_BGT_NAME_PATTERN, ubi->ubi_num);
+
+	err = -ENOMEM;
+	ubi->lookuptbl = kzalloc(ubi->peb_count * sizeof(void *), GFP_KERNEL);
+	if (!ubi->lookuptbl)
+		return err;
+
+	list_for_each_entry_safe(seb, tmp, &si->erase, u.list) {
+		cond_resched();
+
+		e = kmem_cache_alloc(ubi_wl_entry_slab, GFP_KERNEL);
+		if (!e)
+			goto out_free;
+
+		e->pnum = seb->pnum;
+		e->ec = seb->ec;
+		ubi->lookuptbl[e->pnum] = e;
+		if (schedule_erase(ubi, e, 0)) {
+			kmem_cache_free(ubi_wl_entry_slab, e);
+			goto out_free;
+		}
+	}
+
+	list_for_each_entry(seb, &si->free, u.list) {
+		cond_resched();
+
+		e = kmem_cache_alloc(ubi_wl_entry_slab, GFP_KERNEL);
+		if (!e)
+			goto out_free;
+
+		e->pnum = seb->pnum;
+		e->ec = seb->ec;
+		ubi_assert(e->ec >= 0);
+		wl_tree_add(e, &ubi->free);
+		ubi->lookuptbl[e->pnum] = e;
+	}
+
+	list_for_each_entry(seb, &si->corr, u.list) {
+		cond_resched();
+
+		e = kmem_cache_alloc(ubi_wl_entry_slab, GFP_KERNEL);
+		if (!e)
+			goto out_free;
+
+		e->pnum = seb->pnum;
+		e->ec = seb->ec;
+		ubi->lookuptbl[e->pnum] = e;
+		if (schedule_erase(ubi, e, 0)) {
+			kmem_cache_free(ubi_wl_entry_slab, e);
+			goto out_free;
+		}
+	}
+
+	ubi_rb_for_each_entry(rb1, sv, &si->volumes, rb) {
+		ubi_rb_for_each_entry(rb2, seb, &sv->root, u.rb) {
+			cond_resched();
+
+			e = kmem_cache_alloc(ubi_wl_entry_slab, GFP_KERNEL);
+			if (!e)
+				goto out_free;
+
+			e->pnum = seb->pnum;
+			e->ec = seb->ec;
+			ubi->lookuptbl[e->pnum] = e;
+			if (!seb->scrub) {
+				dbg_wl("add PEB %d EC %d to the used tree",
+				       e->pnum, e->ec);
+				wl_tree_add(e, &ubi->used);
+			} else {
+				dbg_wl("add PEB %d EC %d to the scrub tree",
+				       e->pnum, e->ec);
+				wl_tree_add(e, &ubi->scrub);
+			}
+		}
+	}
+
+	if (ubi->avail_pebs < WL_RESERVED_PEBS) {
+		ubi_err("no enough physical eraseblocks (%d, need %d)",
+			ubi->avail_pebs, WL_RESERVED_PEBS);
+		goto out_free;
+	}
+	ubi->avail_pebs -= WL_RESERVED_PEBS;
+	ubi->rsvd_pebs += WL_RESERVED_PEBS;
+
+	/* Schedule wear-leveling if needed */
+	err = ensure_wear_leveling(ubi);
+	if (err)
+		goto out_free;
+
+	return 0;
+
+out_free:
+	cancel_pending(ubi);
+	tree_destroy(&ubi->used);
+	tree_destroy(&ubi->free);
+	tree_destroy(&ubi->scrub);
+	kfree(ubi->lookuptbl);
+	return err;
+}
+
+/**
+ * protection_trees_destroy - destroy the protection RB-trees.
+ * @ubi: UBI device description object
+ */
+static void protection_trees_destroy(struct ubi_device *ubi)
+{
+	struct rb_node *rb;
+	struct ubi_wl_prot_entry *pe;
+
+	rb = ubi->prot.aec.rb_node;
+	while (rb) {
+		if (rb->rb_left)
+			rb = rb->rb_left;
+		else if (rb->rb_right)
+			rb = rb->rb_right;
+		else {
+			pe = rb_entry(rb, struct ubi_wl_prot_entry, rb_aec);
+
+			rb = rb_parent(rb);
+			if (rb) {
+				if (rb->rb_left == &pe->rb_aec)
+					rb->rb_left = NULL;
+				else
+					rb->rb_right = NULL;
+			}
+
+			kmem_cache_free(ubi_wl_entry_slab, pe->e);
+			kfree(pe);
+		}
+	}
+}
+
+/**
+ * ubi_wl_close - close the wear-leveling unit.
+ * @ubi: UBI device description object
+ */
+void ubi_wl_close(struct ubi_device *ubi)
+{
+	dbg_wl("close the UBI wear-leveling unit");
+
+	cancel_pending(ubi);
+	protection_trees_destroy(ubi);
+	tree_destroy(&ubi->used);
+	tree_destroy(&ubi->free);
+	tree_destroy(&ubi->scrub);
+	kfree(ubi->lookuptbl);
+}
+
+#ifdef CONFIG_MTD_UBI_DEBUG_PARANOID
+
+/**
+ * paranoid_check_ec - make sure that the erase counter of a physical eraseblock
+ * is correct.
+ * @ubi: UBI device description object
+ * @pnum: the physical eraseblock number to check
+ * @ec: the erase counter to check
+ *
+ * This function returns zero if the erase counter of physical eraseblock @pnum
+ * is equivalent to @ec, %1 if not, and a negative error code if an error
+ * occurred.
+ */
+static int paranoid_check_ec(struct ubi_device *ubi, int pnum, int ec)
+{
+	int err;
+	long long read_ec;
+	struct ubi_ec_hdr *ec_hdr;
+
+	ec_hdr = kzalloc(ubi->ec_hdr_alsize, GFP_NOFS);
+	if (!ec_hdr)
+		return -ENOMEM;
+
+	err = ubi_io_read_ec_hdr(ubi, pnum, ec_hdr, 0);
+	if (err && err != UBI_IO_BITFLIPS) {
+		/* The header does not have to exist */
+		err = 0;
+		goto out_free;
+	}
+
+	read_ec = be64_to_cpu(ec_hdr->ec);
+	if (ec != read_ec) {
+		ubi_err("paranoid check failed for PEB %d", pnum);
+		ubi_err("read EC is %lld, should be %d", read_ec, ec);
+		ubi_dbg_dump_stack();
+		err = 1;
+	} else
+		err = 0;
+
+out_free:
+	kfree(ec_hdr);
+	return err;
+}
+
+/**
+ * paranoid_check_in_wl_tree - make sure that a wear-leveling entry is present
+ * in a WL RB-tree.
+ * @e: the wear-leveling entry to check
+ * @root: the root of the tree
+ *
+ * This function returns zero if @e is in the @root RB-tree and %1 if it
+ * is not.
+ */
+static int paranoid_check_in_wl_tree(struct ubi_wl_entry *e,
+				     struct rb_root *root)
+{
+	if (in_wl_tree(e, root))
+		return 0;
+
+	ubi_err("paranoid check failed for PEB %d, EC %d, RB-tree %p ",
+		e->pnum, e->ec, root);
+	ubi_dbg_dump_stack();
+	return 1;
+}
+
+#endif /* CONFIG_MTD_UBI_DEBUG_PARANOID */