add idl4k kernel firmware version 1.13.0.105

kernel/drivers/mtd/ubi/Kconfig

@@ -0,0 +1,64 @@
+# drivers/mtd/ubi/Kconfig
+
+menu "UBI - Unsorted block images"
+	depends on MTD
+
+config MTD_UBI
+	tristate "Enable UBI"
+	depends on MTD
+	select CRC32
+	help
+	  UBI is a software layer above MTD layer which admits of LVM-like
+	  logical volumes on top of MTD devices, hides some complexities of
+	  flash chips like wear and bad blocks and provides some other useful
+	  capabilities. Please, consult the MTD web site for more details
+	  (www.linux-mtd.infradead.org).
+
+config MTD_UBI_WL_THRESHOLD
+	int "UBI wear-leveling threshold"
+	default 4096
+	range 2 65536
+	depends on MTD_UBI
+	help
+	  This parameter defines the maximum difference between the highest
+	  erase counter value and the lowest erase counter value of eraseblocks
+	  of UBI devices. When this threshold is exceeded, UBI starts performing
+	  wear leveling by means of moving data from eraseblock with low erase
+	  counter to eraseblocks with high erase counter.
+
+	  The default value should be OK for SLC NAND flashes, NOR flashes and
+	  other flashes which have eraseblock life-cycle 100000 or more.
+	  However, in case of MLC NAND flashes which typically have eraseblock
+	  life-cycle less then 10000, the threshold should be lessened (e.g.,
+	  to 128 or 256, although it does not have to be power of 2).
+
+config MTD_UBI_BEB_RESERVE
+	int "Percentage of reserved eraseblocks for bad eraseblocks handling"
+	default 1
+	range 0 25
+	depends on MTD_UBI
+	help
+	  If the MTD device admits of bad eraseblocks (e.g. NAND flash), UBI
+	  reserves some amount of physical eraseblocks to handle new bad
+	  eraseblocks. For example, if a flash physical eraseblock becomes bad,
+	  UBI uses these reserved physical eraseblocks to relocate the bad one.
+	  This option specifies how many physical eraseblocks will be reserved
+	  for bad eraseblock handling (percents of total number of good flash
+	  eraseblocks). If the underlying flash does not admit of bad
+	  eraseblocks (e.g. NOR flash), this value is ignored and nothing is
+	  reserved. Leave the default value if unsure.
+
+config MTD_UBI_GLUEBI
+	tristate "MTD devices emulation driver (gluebi)"
+	default n
+	depends on MTD_UBI
+	help
+	   This option enables gluebi - an additional driver which emulates MTD
+	   devices on top of UBI volumes: for each UBI volumes an MTD device is
+	   created, and all I/O to this MTD device is redirected to the UBI
+	   volume. This is handy to make MTD-oriented software (like JFFS2)
+	   work on top of UBI. Do not enable this unless you use legacy
+	   software.
+
+source "drivers/mtd/ubi/Kconfig.debug"
+endmenu

kernel/drivers/mtd/ubi/Kconfig.debug

@@ -0,0 +1,94 @@
+comment "UBI debugging options"
+	depends on MTD_UBI
+
+config MTD_UBI_DEBUG
+	bool "UBI debugging"
+	depends on SYSFS
+	depends on MTD_UBI
+	select DEBUG_FS
+	select KALLSYMS_ALL
+	help
+	  This option enables UBI debugging.
+
+config MTD_UBI_DEBUG_MSG
+	bool "UBI debugging messages"
+	depends on MTD_UBI_DEBUG
+	default n
+	help
+	  This option enables UBI debugging messages.
+
+config MTD_UBI_DEBUG_PARANOID
+	bool "Extra self-checks"
+	default n
+	depends on MTD_UBI_DEBUG
+	help
+	  This option enables extra checks in UBI code. Note this slows UBI down
+	  significantly.
+
+config MTD_UBI_DEBUG_DISABLE_BGT
+	bool "Do not enable the UBI background thread"
+	depends on MTD_UBI_DEBUG
+	default n
+	help
+	  This option switches the background thread off by default. The thread
+	  may be also be enabled/disabled via UBI sysfs.
+
+config MTD_UBI_DEBUG_EMULATE_BITFLIPS
+	bool "Emulate flash bit-flips"
+	depends on MTD_UBI_DEBUG
+	default n
+	help
+	  This option emulates bit-flips with probability 1/50, which in turn
+	  causes scrubbing. Useful for debugging and stressing UBI.
+
+config MTD_UBI_DEBUG_EMULATE_WRITE_FAILURES
+	bool "Emulate flash write failures"
+	depends on MTD_UBI_DEBUG
+	default n
+	help
+	  This option emulates write failures with probability 1/100. Useful for
+	  debugging and testing how UBI handlines errors.
+
+config MTD_UBI_DEBUG_EMULATE_ERASE_FAILURES
+	bool "Emulate flash erase failures"
+	depends on MTD_UBI_DEBUG
+	default n
+	help
+	  This option emulates erase failures with probability 1/100. Useful for
+	  debugging and testing how UBI handlines errors.
+
+menu "Additional UBI debugging messages"
+	depends on MTD_UBI_DEBUG
+
+config MTD_UBI_DEBUG_MSG_BLD
+	bool "Additional UBI initialization and build messages"
+	default n
+	depends on MTD_UBI_DEBUG
+	help
+	  This option enables detailed UBI initialization and device build
+	  debugging messages.
+
+config MTD_UBI_DEBUG_MSG_EBA
+	bool "Eraseblock association unit messages"
+	default n
+	depends on MTD_UBI_DEBUG
+	help
+	  This option enables debugging messages from the UBI eraseblock
+	  association unit.
+
+config MTD_UBI_DEBUG_MSG_WL
+	bool "Wear-leveling unit messages"
+	default n
+	depends on MTD_UBI_DEBUG
+	help
+	  This option enables debugging messages from the UBI wear-leveling
+	  unit.
+
+config MTD_UBI_DEBUG_MSG_IO
+	bool "Input/output unit messages"
+	default n
+	depends on MTD_UBI_DEBUG
+	help
+	  This option enables debugging messages from the UBI input/output unit.
+
+endmenu # UBI debugging messages

kernel/drivers/mtd/ubi/Makefile

@@ -0,0 +1,7 @@
+obj-$(CONFIG_MTD_UBI) += ubi.o
+
+ubi-y += vtbl.o vmt.o upd.o build.o cdev.o kapi.o eba.o io.o wl.o scan.o
+ubi-y += misc.o
+
+ubi-$(CONFIG_MTD_UBI_DEBUG) += debug.o
+obj-$(CONFIG_MTD_UBI_GLUEBI) += gluebi.o

kernel/drivers/mtd/ubi/debug.c

@@ -0,0 +1,231 @@
+/*
+ * 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)
+ */
+
+/*
+ * Here we keep all the UBI debugging stuff which should normally be disabled
+ * and compiled-out, but it is extremely helpful when hunting bugs or doing big
+ * changes.
+ */
+
+#ifdef CONFIG_MTD_UBI_DEBUG
+
+#include "ubi.h"
+
+/**
+ * ubi_dbg_dump_ec_hdr - dump an erase counter header.
+ * @ec_hdr: the erase counter header to dump
+ */
+void ubi_dbg_dump_ec_hdr(const struct ubi_ec_hdr *ec_hdr)
+{
+	printk(KERN_DEBUG "Erase counter header dump:\n");
+	printk(KERN_DEBUG "\tmagic          %#08x\n",
+	       be32_to_cpu(ec_hdr->magic));
+	printk(KERN_DEBUG "\tversion        %d\n", (int)ec_hdr->version);
+	printk(KERN_DEBUG "\tec             %llu\n",
+	       (long long)be64_to_cpu(ec_hdr->ec));
+	printk(KERN_DEBUG "\tvid_hdr_offset %d\n",
+	       be32_to_cpu(ec_hdr->vid_hdr_offset));
+	printk(KERN_DEBUG "\tdata_offset    %d\n",
+	       be32_to_cpu(ec_hdr->data_offset));
+	printk(KERN_DEBUG "\timage_seq      %d\n",
+	       be32_to_cpu(ec_hdr->image_seq));
+	printk(KERN_DEBUG "\thdr_crc        %#08x\n",
+	       be32_to_cpu(ec_hdr->hdr_crc));
+	printk(KERN_DEBUG "erase counter header hexdump:\n");
+	print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 1,
+		       ec_hdr, UBI_EC_HDR_SIZE, 1);
+}
+
+/**
+ * ubi_dbg_dump_vid_hdr - dump a volume identifier header.
+ * @vid_hdr: the volume identifier header to dump
+ */
+void ubi_dbg_dump_vid_hdr(const struct ubi_vid_hdr *vid_hdr)
+{
+	printk(KERN_DEBUG "Volume identifier header dump:\n");
+	printk(KERN_DEBUG "\tmagic     %08x\n", be32_to_cpu(vid_hdr->magic));
+	printk(KERN_DEBUG "\tversion   %d\n",   (int)vid_hdr->version);
+	printk(KERN_DEBUG "\tvol_type  %d\n",   (int)vid_hdr->vol_type);
+	printk(KERN_DEBUG "\tcopy_flag %d\n",   (int)vid_hdr->copy_flag);
+	printk(KERN_DEBUG "\tcompat    %d\n",   (int)vid_hdr->compat);
+	printk(KERN_DEBUG "\tvol_id    %d\n",   be32_to_cpu(vid_hdr->vol_id));
+	printk(KERN_DEBUG "\tlnum      %d\n",   be32_to_cpu(vid_hdr->lnum));
+	printk(KERN_DEBUG "\tdata_size %d\n",   be32_to_cpu(vid_hdr->data_size));
+	printk(KERN_DEBUG "\tused_ebs  %d\n",   be32_to_cpu(vid_hdr->used_ebs));
+	printk(KERN_DEBUG "\tdata_pad  %d\n",   be32_to_cpu(vid_hdr->data_pad));
+	printk(KERN_DEBUG "\tsqnum     %llu\n",
+		(unsigned long long)be64_to_cpu(vid_hdr->sqnum));
+	printk(KERN_DEBUG "\thdr_crc   %08x\n", be32_to_cpu(vid_hdr->hdr_crc));
+	printk(KERN_DEBUG "Volume identifier header hexdump:\n");
+	print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 1,
+		       vid_hdr, UBI_VID_HDR_SIZE, 1);
+}
+
+/**
+ * ubi_dbg_dump_vol_info- dump volume information.
+ * @vol: UBI volume description object
+ */
+void ubi_dbg_dump_vol_info(const struct ubi_volume *vol)
+{
+	printk(KERN_DEBUG "Volume information dump:\n");
+	printk(KERN_DEBUG "\tvol_id          %d\n", vol->vol_id);
+	printk(KERN_DEBUG "\treserved_pebs   %d\n", vol->reserved_pebs);
+	printk(KERN_DEBUG "\talignment       %d\n", vol->alignment);
+	printk(KERN_DEBUG "\tdata_pad        %d\n", vol->data_pad);
+	printk(KERN_DEBUG "\tvol_type        %d\n", vol->vol_type);
+	printk(KERN_DEBUG "\tname_len        %d\n", vol->name_len);
+	printk(KERN_DEBUG "\tusable_leb_size %d\n", vol->usable_leb_size);
+	printk(KERN_DEBUG "\tused_ebs        %d\n", vol->used_ebs);
+	printk(KERN_DEBUG "\tused_bytes      %lld\n", vol->used_bytes);
+	printk(KERN_DEBUG "\tlast_eb_bytes   %d\n", vol->last_eb_bytes);
+	printk(KERN_DEBUG "\tcorrupted       %d\n", vol->corrupted);
+	printk(KERN_DEBUG "\tupd_marker      %d\n", vol->upd_marker);
+
+	if (vol->name_len <= UBI_VOL_NAME_MAX &&
+	    strnlen(vol->name, vol->name_len + 1) == vol->name_len) {
+		printk(KERN_DEBUG "\tname            %s\n", vol->name);
+	} else {
+		printk(KERN_DEBUG "\t1st 5 characters of name: %c%c%c%c%c\n",
+		       vol->name[0], vol->name[1], vol->name[2],
+		       vol->name[3], vol->name[4]);
+	}
+}
+
+/**
+ * ubi_dbg_dump_vtbl_record - dump a &struct ubi_vtbl_record object.
+ * @r: the object to dump
+ * @idx: volume table index
+ */
+void ubi_dbg_dump_vtbl_record(const struct ubi_vtbl_record *r, int idx)
+{
+	int name_len = be16_to_cpu(r->name_len);
+
+	printk(KERN_DEBUG "Volume table record %d dump:\n", idx);
+	printk(KERN_DEBUG "\treserved_pebs   %d\n",
+	       be32_to_cpu(r->reserved_pebs));
+	printk(KERN_DEBUG "\talignment       %d\n", be32_to_cpu(r->alignment));
+	printk(KERN_DEBUG "\tdata_pad        %d\n", be32_to_cpu(r->data_pad));
+	printk(KERN_DEBUG "\tvol_type        %d\n", (int)r->vol_type);
+	printk(KERN_DEBUG "\tupd_marker      %d\n", (int)r->upd_marker);
+	printk(KERN_DEBUG "\tname_len        %d\n", name_len);
+
+	if (r->name[0] == '\0') {
+		printk(KERN_DEBUG "\tname            NULL\n");
+		return;
+	}
+
+	if (name_len <= UBI_VOL_NAME_MAX &&
+	    strnlen(&r->name[0], name_len + 1) == name_len) {
+		printk(KERN_DEBUG "\tname            %s\n", &r->name[0]);
+	} else {
+		printk(KERN_DEBUG "\t1st 5 characters of name: %c%c%c%c%c\n",
+			r->name[0], r->name[1], r->name[2], r->name[3],
+			r->name[4]);
+	}
+	printk(KERN_DEBUG "\tcrc             %#08x\n", be32_to_cpu(r->crc));
+}
+
+/**
+ * ubi_dbg_dump_sv - dump a &struct ubi_scan_volume object.
+ * @sv: the object to dump
+ */
+void ubi_dbg_dump_sv(const struct ubi_scan_volume *sv)
+{
+	printk(KERN_DEBUG "Volume scanning information dump:\n");
+	printk(KERN_DEBUG "\tvol_id         %d\n", sv->vol_id);
+	printk(KERN_DEBUG "\thighest_lnum   %d\n", sv->highest_lnum);
+	printk(KERN_DEBUG "\tleb_count      %d\n", sv->leb_count);
+	printk(KERN_DEBUG "\tcompat         %d\n", sv->compat);
+	printk(KERN_DEBUG "\tvol_type       %d\n", sv->vol_type);
+	printk(KERN_DEBUG "\tused_ebs       %d\n", sv->used_ebs);
+	printk(KERN_DEBUG "\tlast_data_size %d\n", sv->last_data_size);
+	printk(KERN_DEBUG "\tdata_pad       %d\n", sv->data_pad);
+}
+
+/**
+ * ubi_dbg_dump_seb - dump a &struct ubi_scan_leb object.
+ * @seb: the object to dump
+ * @type: object type: 0 - not corrupted, 1 - corrupted
+ */
+void ubi_dbg_dump_seb(const struct ubi_scan_leb *seb, int type)
+{
+	printk(KERN_DEBUG "eraseblock scanning information dump:\n");
+	printk(KERN_DEBUG "\tec       %d\n", seb->ec);
+	printk(KERN_DEBUG "\tpnum     %d\n", seb->pnum);
+	if (type == 0) {
+		printk(KERN_DEBUG "\tlnum     %d\n", seb->lnum);
+		printk(KERN_DEBUG "\tscrub    %d\n", seb->scrub);
+		printk(KERN_DEBUG "\tsqnum    %llu\n", seb->sqnum);
+	}
+}
+
+/**
+ * ubi_dbg_dump_mkvol_req - dump a &struct ubi_mkvol_req object.
+ * @req: the object to dump
+ */
+void ubi_dbg_dump_mkvol_req(const struct ubi_mkvol_req *req)
+{
+	char nm[17];
+
+	printk(KERN_DEBUG "Volume creation request dump:\n");
+	printk(KERN_DEBUG "\tvol_id    %d\n",   req->vol_id);
+	printk(KERN_DEBUG "\talignment %d\n",   req->alignment);
+	printk(KERN_DEBUG "\tbytes     %lld\n", (long long)req->bytes);
+	printk(KERN_DEBUG "\tvol_type  %d\n",   req->vol_type);
+	printk(KERN_DEBUG "\tname_len  %d\n",   req->name_len);
+
+	memcpy(nm, req->name, 16);
+	nm[16] = 0;
+	printk(KERN_DEBUG "\t1st 16 characters of name: %s\n", nm);
+}
+
+/**
+ * ubi_dbg_dump_flash - dump a region of flash.
+ * @ubi: UBI device description object
+ * @pnum: the physical eraseblock number to dump
+ * @offset: the starting offset within the physical eraseblock to dump
+ * @len: the length of the region to dump
+ */
+void ubi_dbg_dump_flash(struct ubi_device *ubi, int pnum, int offset, int len)
+{
+	int err;
+	size_t read;
+	void *buf;
+	loff_t addr = (loff_t)pnum * ubi->peb_size + offset;
+
+	buf = vmalloc(len);
+	if (!buf)
+		return;
+	err = ubi->mtd->read(ubi->mtd, addr, len, &read, buf);
+	if (err && err != -EUCLEAN) {
+		ubi_err("error %d while reading %d bytes from PEB %d:%d, "
+			"read %zd bytes", err, len, pnum, offset, read);
+		goto out;
+	}
+
+	dbg_msg("dumping %d bytes of data from PEB %d, offset %d",
+		len, pnum, offset);
+	print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 1, buf, len, 1);
+out:
+	vfree(buf);
+	return;
+}
+
+#endif /* CONFIG_MTD_UBI_DEBUG */

kernel/drivers/mtd/ubi/debug.h

@@ -0,0 +1,181 @@
+/*
+ * 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)
+ */
+
+#ifndef __UBI_DEBUG_H__
+#define __UBI_DEBUG_H__
+
+#ifdef CONFIG_MTD_UBI_DEBUG
+#include <linux/random.h>
+
+#define dbg_err(fmt, ...) ubi_err(fmt, ##__VA_ARGS__)
+
+#define ubi_assert(expr)  do {                                               \
+	if (unlikely(!(expr))) {                                             \
+		printk(KERN_CRIT "UBI assert failed in %s at %u (pid %d)\n", \
+		       __func__, __LINE__, current->pid);                    \
+		ubi_dbg_dump_stack();                                        \
+	}                                                                    \
+} while (0)
+
+#define dbg_msg(fmt, ...)                                    \
+	printk(KERN_DEBUG "UBI DBG (pid %d): %s: " fmt "\n", \
+	       current->pid, __func__, ##__VA_ARGS__)
+
+#define ubi_dbg_dump_stack() dump_stack()
+
+struct ubi_ec_hdr;
+struct ubi_vid_hdr;
+struct ubi_volume;
+struct ubi_vtbl_record;
+struct ubi_scan_volume;
+struct ubi_scan_leb;
+struct ubi_mkvol_req;
+
+void ubi_dbg_dump_ec_hdr(const struct ubi_ec_hdr *ec_hdr);
+void ubi_dbg_dump_vid_hdr(const struct ubi_vid_hdr *vid_hdr);
+void ubi_dbg_dump_vol_info(const struct ubi_volume *vol);
+void ubi_dbg_dump_vtbl_record(const struct ubi_vtbl_record *r, int idx);
+void ubi_dbg_dump_sv(const struct ubi_scan_volume *sv);
+void ubi_dbg_dump_seb(const struct ubi_scan_leb *seb, int type);
+void ubi_dbg_dump_mkvol_req(const struct ubi_mkvol_req *req);
+void ubi_dbg_dump_flash(struct ubi_device *ubi, int pnum, int offset, int len);
+
+#ifdef CONFIG_MTD_UBI_DEBUG_MSG
+/* General debugging messages */
+#define dbg_gen(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__)
+#else
+#define dbg_gen(fmt, ...) ({})
+#endif
+
+#ifdef CONFIG_MTD_UBI_DEBUG_MSG_EBA
+/* Messages from the eraseblock association sub-system */
+#define dbg_eba(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__)
+#else
+#define dbg_eba(fmt, ...) ({})
+#endif
+
+#ifdef CONFIG_MTD_UBI_DEBUG_MSG_WL
+/* Messages from the wear-leveling sub-system */
+#define dbg_wl(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__)
+#else
+#define dbg_wl(fmt, ...) ({})
+#endif
+
+#ifdef CONFIG_MTD_UBI_DEBUG_MSG_IO
+/* Messages from the input/output sub-system */
+#define dbg_io(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__)
+#else
+#define dbg_io(fmt, ...) ({})
+#endif
+
+#ifdef CONFIG_MTD_UBI_DEBUG_MSG_BLD
+/* Initialization and build messages */
+#define dbg_bld(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__)
+#define UBI_IO_DEBUG 1
+#else
+#define dbg_bld(fmt, ...) ({})
+#define UBI_IO_DEBUG 0
+#endif
+
+#ifdef CONFIG_MTD_UBI_DEBUG_PARANOID
+int ubi_dbg_check_all_ff(struct ubi_device *ubi, int pnum, int offset, int len);
+#else
+#define ubi_dbg_check_all_ff(ubi, pnum, offset, len) 0
+#endif
+
+#ifdef CONFIG_MTD_UBI_DEBUG_DISABLE_BGT
+#define DBG_DISABLE_BGT 1
+#else
+#define DBG_DISABLE_BGT 0
+#endif
+
+#ifdef CONFIG_MTD_UBI_DEBUG_EMULATE_BITFLIPS
+/**
+ * ubi_dbg_is_bitflip - if it is time to emulate a bit-flip.
+ *
+ * Returns non-zero if a bit-flip should be emulated, otherwise returns zero.
+ */
+static inline int ubi_dbg_is_bitflip(void)
+{
+	return !(random32() % 200);
+}
+#else
+#define ubi_dbg_is_bitflip() 0
+#endif
+
+#ifdef CONFIG_MTD_UBI_DEBUG_EMULATE_WRITE_FAILURES
+/**
+ * ubi_dbg_is_write_failure - if it is time to emulate a write failure.
+ *
+ * Returns non-zero if a write failure should be emulated, otherwise returns
+ * zero.
+ */
+static inline int ubi_dbg_is_write_failure(void)
+{
+	return !(random32() % 500);
+}
+#else
+#define ubi_dbg_is_write_failure() 0
+#endif
+
+#ifdef CONFIG_MTD_UBI_DEBUG_EMULATE_ERASE_FAILURES
+/**
+ * ubi_dbg_is_erase_failure - if its time to emulate an erase failure.
+ *
+ * Returns non-zero if an erase failure should be emulated, otherwise returns
+ * zero.
+ */
+static inline int ubi_dbg_is_erase_failure(void)
+{
+		return !(random32() % 400);
+}
+#else
+#define ubi_dbg_is_erase_failure() 0
+#endif
+
+#else
+
+#define ubi_assert(expr)                 ({})
+#define dbg_err(fmt, ...)                ({})
+#define dbg_msg(fmt, ...)                ({})
+#define dbg_gen(fmt, ...)                ({})
+#define dbg_eba(fmt, ...)                ({})
+#define dbg_wl(fmt, ...)                 ({})
+#define dbg_io(fmt, ...)                 ({})
+#define dbg_bld(fmt, ...)                ({})
+#define ubi_dbg_dump_stack()             ({})
+#define ubi_dbg_dump_ec_hdr(ec_hdr)      ({})
+#define ubi_dbg_dump_vid_hdr(vid_hdr)    ({})
+#define ubi_dbg_dump_vol_info(vol)       ({})
+#define ubi_dbg_dump_vtbl_record(r, idx) ({})
+#define ubi_dbg_dump_sv(sv)              ({})
+#define ubi_dbg_dump_seb(seb, type)      ({})
+#define ubi_dbg_dump_mkvol_req(req)      ({})
+#define ubi_dbg_dump_flash(ubi, pnum, offset, len) ({})
+
+#define UBI_IO_DEBUG               0
+#define DBG_DISABLE_BGT            0
+#define ubi_dbg_is_bitflip()       0
+#define ubi_dbg_is_write_failure() 0
+#define ubi_dbg_is_erase_failure() 0
+#define ubi_dbg_check_all_ff(ubi, pnum, offset, len) 0
+
+#endif /* !CONFIG_MTD_UBI_DEBUG */
+#endif /* !__UBI_DEBUG_H__ */

kernel/drivers/mtd/ubi/gluebi.c

@@ -0,0 +1,541 @@
+/*
+ * 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), Joern Engel
+ */
+
+/*
+ * This is a small driver which implements fake MTD devices on top of UBI
+ * volumes. This sounds strange, but it is in fact quite useful to make
+ * MTD-oriented software (including all the legacy software) work on top of
+ * UBI.
+ *
+ * Gluebi emulates MTD devices of "MTD_UBIVOLUME" type. Their minimal I/O unit
+ * size (@mtd->writesize) is equivalent to the UBI minimal I/O unit. The
+ * eraseblock size is equivalent to the logical eraseblock size of the volume.
+ */
+
+#include <linux/err.h>
+#include <linux/list.h>
+#include <linux/sched.h>
+#include <linux/math64.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/mtd/ubi.h>
+#include <linux/mtd/mtd.h>
+#include "ubi-media.h"
+
+#define err_msg(fmt, ...)                                   \
+	printk(KERN_DEBUG "gluebi (pid %d): %s: " fmt "\n", \
+	       current->pid, __func__, ##__VA_ARGS__)
+
+/**
+ * struct gluebi_device - a gluebi device description data structure.
+ * @mtd: emulated MTD device description object
+ * @refcnt: gluebi device reference count
+ * @desc: UBI volume descriptor
+ * @ubi_num: UBI device number this gluebi device works on
+ * @vol_id: ID of UBI volume this gluebi device works on
+ * @list: link in a list of gluebi devices
+ */
+struct gluebi_device {
+	struct mtd_info mtd;
+	int refcnt;
+	struct ubi_volume_desc *desc;
+	int ubi_num;
+	int vol_id;
+	struct list_head list;
+};
+
+/* List of all gluebi devices */
+static LIST_HEAD(gluebi_devices);
+static DEFINE_MUTEX(devices_mutex);
+
+/**
+ * find_gluebi_nolock - find a gluebi device.
+ * @ubi_num: UBI device number
+ * @vol_id: volume ID
+ *
+ * This function seraches for gluebi device corresponding to UBI device
+ * @ubi_num and UBI volume @vol_id. Returns the gluebi device description
+ * object in case of success and %NULL in case of failure. The caller has to
+ * have the &devices_mutex locked.
+ */
+static struct gluebi_device *find_gluebi_nolock(int ubi_num, int vol_id)
+{
+	struct gluebi_device *gluebi;
+
+	list_for_each_entry(gluebi, &gluebi_devices, list)
+		if (gluebi->ubi_num == ubi_num && gluebi->vol_id == vol_id)
+			return gluebi;
+	return NULL;
+}
+
+/**
+ * gluebi_get_device - get MTD device reference.
+ * @mtd: the MTD device description object
+ *
+ * This function is called every time the MTD device is being opened and
+ * implements the MTD get_device() operation. Returns zero in case of success
+ * and a negative error code in case of failure.
+ */
+static int gluebi_get_device(struct mtd_info *mtd)
+{
+	struct gluebi_device *gluebi;
+	int ubi_mode = UBI_READONLY;
+
+	if (!try_module_get(THIS_MODULE))
+		return -ENODEV;
+
+	if (mtd->flags & MTD_WRITEABLE)
+		ubi_mode = UBI_READWRITE;
+
+	gluebi = container_of(mtd, struct gluebi_device, mtd);
+	mutex_lock(&devices_mutex);
+	if (gluebi->refcnt > 0) {
+		/*
+		 * The MTD device is already referenced and this is just one
+		 * more reference. MTD allows many users to open the same
+		 * volume simultaneously and do not distinguish between
+		 * readers/writers/exclusive openers as UBI does. So we do not
+		 * open the UBI volume again - just increase the reference
+		 * counter and return.
+		 */
+		gluebi->refcnt += 1;
+		mutex_unlock(&devices_mutex);
+		return 0;
+	}
+
+	/*
+	 * This is the first reference to this UBI volume via the MTD device
+	 * interface. Open the corresponding volume in read-write mode.
+	 */
+	gluebi->desc = ubi_open_volume(gluebi->ubi_num, gluebi->vol_id,
+				       ubi_mode);
+	if (IS_ERR(gluebi->desc)) {
+		mutex_unlock(&devices_mutex);
+		module_put(THIS_MODULE);
+		return PTR_ERR(gluebi->desc);
+	}
+	gluebi->refcnt += 1;
+	mutex_unlock(&devices_mutex);
+	return 0;
+}
+
+/**
+ * gluebi_put_device - put MTD device reference.
+ * @mtd: the MTD device description object
+ *
+ * This function is called every time the MTD device is being put. Returns
+ * zero in case of success and a negative error code in case of failure.
+ */
+static void gluebi_put_device(struct mtd_info *mtd)
+{
+	struct gluebi_device *gluebi;
+
+	gluebi = container_of(mtd, struct gluebi_device, mtd);
+	mutex_lock(&devices_mutex);
+	gluebi->refcnt -= 1;
+	if (gluebi->refcnt == 0)
+		ubi_close_volume(gluebi->desc);
+	module_put(THIS_MODULE);
+	mutex_unlock(&devices_mutex);
+}
+
+/**
+ * gluebi_read - read operation of emulated MTD devices.
+ * @mtd: MTD device description object
+ * @from: absolute offset from where to read
+ * @len: how many bytes to read
+ * @retlen: count of read bytes is returned here
+ * @buf: buffer to store the read data
+ *
+ * This function returns zero in case of success and a negative error code in
+ * case of failure.
+ */
+static int gluebi_read(struct mtd_info *mtd, loff_t from, size_t len,
+		       size_t *retlen, unsigned char *buf)
+{
+	int err = 0, lnum, offs, total_read;
+	struct gluebi_device *gluebi;
+
+	if (len < 0 || from < 0 || from + len > mtd->size)
+		return -EINVAL;
+
+	gluebi = container_of(mtd, struct gluebi_device, mtd);
+
+	lnum = div_u64_rem(from, mtd->erasesize, &offs);
+	total_read = len;
+	while (total_read) {
+		size_t to_read = mtd->erasesize - offs;
+
+		if (to_read > total_read)
+			to_read = total_read;
+
+		err = ubi_read(gluebi->desc, lnum, buf, offs, to_read);
+		if (err)
+			break;
+
+		lnum += 1;
+		offs = 0;
+		total_read -= to_read;
+		buf += to_read;
+	}
+
+	*retlen = len - total_read;
+	return err;
+}
+
+/**
+ * gluebi_write - write operation of emulated MTD devices.
+ * @mtd: MTD device description object
+ * @to: absolute offset where to write
+ * @len: how many bytes to write
+ * @retlen: count of written bytes is returned here
+ * @buf: buffer with data to write
+ *
+ * This function returns zero in case of success and a negative error code in
+ * case of failure.
+ */
+static int gluebi_write(struct mtd_info *mtd, loff_t to, size_t len,
+			size_t *retlen, const u_char *buf)
+{
+	int err = 0, lnum, offs, total_written;
+	struct gluebi_device *gluebi;
+
+	if (len < 0 || to < 0 || len + to > mtd->size)
+		return -EINVAL;
+
+	gluebi = container_of(mtd, struct gluebi_device, mtd);
+
+	if (!(mtd->flags & MTD_WRITEABLE))
+		return -EROFS;
+
+	lnum = div_u64_rem(to, mtd->erasesize, &offs);
+
+	if (len % mtd->writesize || offs % mtd->writesize)
+		return -EINVAL;
+
+	total_written = len;
+	while (total_written) {
+		size_t to_write = mtd->erasesize - offs;
+
+		if (to_write > total_written)
+			to_write = total_written;
+
+		err = ubi_write(gluebi->desc, lnum, buf, offs, to_write);
+		if (err)
+			break;
+
+		lnum += 1;
+		offs = 0;
+		total_written -= to_write;
+		buf += to_write;
+	}
+
+	*retlen = len - total_written;
+	return err;
+}
+
+/**
+ * gluebi_erase - erase operation of emulated MTD devices.
+ * @mtd: the MTD device description object
+ * @instr: the erase operation description
+ *
+ * This function calls the erase callback when finishes. Returns zero in case
+ * of success and a negative error code in case of failure.
+ */
+static int gluebi_erase(struct mtd_info *mtd, struct erase_info *instr)
+{
+	int err, i, lnum, count;
+	struct gluebi_device *gluebi;
+
+	if (instr->addr < 0 || instr->addr > mtd->size - mtd->erasesize)
+		return -EINVAL;
+	if (instr->len < 0 || instr->addr + instr->len > mtd->size)
+		return -EINVAL;
+	if (mtd_mod_by_ws(instr->addr, mtd) || mtd_mod_by_ws(instr->len, mtd))
+		return -EINVAL;
+
+	lnum = mtd_div_by_eb(instr->addr, mtd);
+	count = mtd_div_by_eb(instr->len, mtd);
+
+	gluebi = container_of(mtd, struct gluebi_device, mtd);
+
+	if (!(mtd->flags & MTD_WRITEABLE))
+		return -EROFS;
+
+	for (i = 0; i < count - 1; i++) {
+		err = ubi_leb_unmap(gluebi->desc, lnum + i);
+		if (err)
+			goto out_err;
+	}
+	/*
+	 * MTD erase operations are synchronous, so we have to make sure the
+	 * physical eraseblock is wiped out.
+	 *
+	 * Thus, perform leb_erase instead of leb_unmap operation - leb_erase
+	 * will wait for the end of operations
+	 */
+	err = ubi_leb_erase(gluebi->desc, lnum + i);
+	if (err)
+		goto out_err;
+
+	instr->state = MTD_ERASE_DONE;
+	mtd_erase_callback(instr);
+	return 0;
+
+out_err:
+	instr->state = MTD_ERASE_FAILED;
+	instr->fail_addr = (long long)lnum * mtd->erasesize;
+	return err;
+}
+
+/**
+ * gluebi_create - create a gluebi device for an UBI volume.
+ * @di: UBI device description object
+ * @vi: UBI volume description object
+ *
+ * This function is called when a new UBI volume is created in order to create
+ * corresponding fake MTD device. Returns zero in case of success and a
+ * negative error code in case of failure.
+ */
+static int gluebi_create(struct ubi_device_info *di,
+			 struct ubi_volume_info *vi)
+{
+	struct gluebi_device *gluebi, *g;
+	struct mtd_info *mtd;
+
+	gluebi = kzalloc(sizeof(struct gluebi_device), GFP_KERNEL);
+	if (!gluebi)
+		return -ENOMEM;
+
+	mtd = &gluebi->mtd;
+	mtd->name = kmemdup(vi->name, vi->name_len + 1, GFP_KERNEL);
+	if (!mtd->name) {
+		kfree(gluebi);
+		return -ENOMEM;
+	}
+
+	gluebi->vol_id = vi->vol_id;
+	gluebi->ubi_num = vi->ubi_num;
+	mtd->type = MTD_UBIVOLUME;
+	if (!di->ro_mode)
+		mtd->flags = MTD_WRITEABLE;
+	mtd->owner      = THIS_MODULE;
+	mtd->writesize  = di->min_io_size;
+	mtd->erasesize  = vi->usable_leb_size;
+	mtd->read       = gluebi_read;
+	mtd->write      = gluebi_write;
+	mtd->erase      = gluebi_erase;
+	mtd->get_device = gluebi_get_device;
+	mtd->put_device = gluebi_put_device;
+
+	/*
+	 * In case of dynamic a volume, MTD device size is just volume size. In
+	 * case of a static volume the size is equivalent to the amount of data
+	 * bytes.
+	 */
+	if (vi->vol_type == UBI_DYNAMIC_VOLUME)
+		mtd->size = (unsigned long long)vi->usable_leb_size * vi->size;
+	else
+		mtd->size = vi->used_bytes;
+
+	/* Just a sanity check - make sure this gluebi device does not exist */
+	mutex_lock(&devices_mutex);
+	g = find_gluebi_nolock(vi->ubi_num, vi->vol_id);
+	if (g)
+		err_msg("gluebi MTD device %d form UBI device %d volume %d "
+			"already exists", g->mtd.index, vi->ubi_num,
+			vi->vol_id);
+	mutex_unlock(&devices_mutex);
+
+	if (add_mtd_device(mtd)) {
+		err_msg("cannot add MTD device");
+		kfree(mtd->name);
+		kfree(gluebi);
+		return -ENFILE;
+	}
+
+	mutex_lock(&devices_mutex);
+	list_add_tail(&gluebi->list, &gluebi_devices);
+	mutex_unlock(&devices_mutex);
+	return 0;
+}
+
+/**
+ * gluebi_remove - remove a gluebi device.
+ * @vi: UBI volume description object
+ *
+ * This function is called when an UBI volume is removed and it removes
+ * corresponding fake MTD device. Returns zero in case of success and a
+ * negative error code in case of failure.
+ */
+static int gluebi_remove(struct ubi_volume_info *vi)
+{
+	int err = 0;
+	struct mtd_info *mtd;
+	struct gluebi_device *gluebi;
+
+	mutex_lock(&devices_mutex);
+	gluebi = find_gluebi_nolock(vi->ubi_num, vi->vol_id);
+	if (!gluebi) {
+		err_msg("got remove notification for unknown UBI device %d "
+			"volume %d", vi->ubi_num, vi->vol_id);
+		err = -ENOENT;
+	} else if (gluebi->refcnt)
+		err = -EBUSY;
+	else
+		list_del(&gluebi->list);
+	mutex_unlock(&devices_mutex);
+	if (err)
+		return err;
+
+	mtd = &gluebi->mtd;
+	err = del_mtd_device(mtd);
+	if (err) {
+		err_msg("cannot remove fake MTD device %d, UBI device %d, "
+			"volume %d, error %d", mtd->index, gluebi->ubi_num,
+			gluebi->vol_id, err);
+		mutex_lock(&devices_mutex);
+		list_add_tail(&gluebi->list, &gluebi_devices);
+		mutex_unlock(&devices_mutex);
+		return err;
+	}
+
+	kfree(mtd->name);
+	kfree(gluebi);
+	return 0;
+}
+
+/**
+ * gluebi_updated - UBI volume was updated notifier.
+ * @vi: volume info structure
+ *
+ * This function is called every time an UBI volume is updated. It does nothing
+ * if te volume @vol is dynamic, and changes MTD device size if the
+ * volume is static. This is needed because static volumes cannot be read past
+ * data they contain. This function returns zero in case of success and a
+ * negative error code in case of error.
+ */
+static int gluebi_updated(struct ubi_volume_info *vi)
+{
+	struct gluebi_device *gluebi;
+
+	mutex_lock(&devices_mutex);
+	gluebi = find_gluebi_nolock(vi->ubi_num, vi->vol_id);
+	if (!gluebi) {
+		mutex_unlock(&devices_mutex);
+		err_msg("got update notification for unknown UBI device %d "
+			"volume %d", vi->ubi_num, vi->vol_id);
+		return -ENOENT;
+	}
+
+	if (vi->vol_type == UBI_STATIC_VOLUME)
+		gluebi->mtd.size = vi->used_bytes;
+	mutex_unlock(&devices_mutex);
+	return 0;
+}
+
+/**
+ * gluebi_resized - UBI volume was re-sized notifier.
+ * @vi: volume info structure
+ *
+ * This function is called every time an UBI volume is re-size. It changes the
+ * corresponding fake MTD device size. This function returns zero in case of
+ * success and a negative error code in case of error.
+ */
+static int gluebi_resized(struct ubi_volume_info *vi)
+{
+	struct gluebi_device *gluebi;
+
+	mutex_lock(&devices_mutex);
+	gluebi = find_gluebi_nolock(vi->ubi_num, vi->vol_id);
+	if (!gluebi) {
+		mutex_unlock(&devices_mutex);
+		err_msg("got update notification for unknown UBI device %d "
+			"volume %d", vi->ubi_num, vi->vol_id);
+		return -ENOENT;
+	}
+	gluebi->mtd.size = vi->used_bytes;
+	mutex_unlock(&devices_mutex);
+	return 0;
+}
+
+/**
+ * gluebi_notify - UBI notification handler.
+ * @nb: registered notifier block
+ * @l: notification type
+ * @ptr: pointer to the &struct ubi_notification object
+ */
+static int gluebi_notify(struct notifier_block *nb, unsigned long l,
+			 void *ns_ptr)
+{
+	struct ubi_notification *nt = ns_ptr;
+
+	switch (l) {
+	case UBI_VOLUME_ADDED:
+		gluebi_create(&nt->di, &nt->vi);
+		break;
+	case UBI_VOLUME_REMOVED:
+		gluebi_remove(&nt->vi);
+		break;
+	case UBI_VOLUME_RESIZED:
+		gluebi_resized(&nt->vi);
+		break;
+	case UBI_VOLUME_UPDATED:
+		gluebi_updated(&nt->vi);
+		break;
+	default:
+		break;
+	}
+	return NOTIFY_OK;
+}
+
+static struct notifier_block gluebi_notifier = {
+	.notifier_call	= gluebi_notify,
+};
+
+static int __init ubi_gluebi_init(void)
+{
+	return ubi_register_volume_notifier(&gluebi_notifier, 0);
+}
+
+static void __exit ubi_gluebi_exit(void)
+{
+	struct gluebi_device *gluebi, *g;
+
+	list_for_each_entry_safe(gluebi, g, &gluebi_devices, list) {
+		int err;
+		struct mtd_info *mtd = &gluebi->mtd;
+
+		err = del_mtd_device(mtd);
+		if (err)
+			err_msg("error %d while removing gluebi MTD device %d, "
+				"UBI device %d, volume %d - ignoring", err,
+				mtd->index, gluebi->ubi_num, gluebi->vol_id);
+		kfree(mtd->name);
+		kfree(gluebi);
+	}
+	ubi_unregister_volume_notifier(&gluebi_notifier);
+}
+
+module_init(ubi_gluebi_init);
+module_exit(ubi_gluebi_exit);
+MODULE_DESCRIPTION("MTD emulation layer over UBI volumes");
+MODULE_AUTHOR("Artem Bityutskiy, Joern Engel");
+MODULE_LICENSE("GPL");

kernel/drivers/mtd/ubi/kapi.c

@@ -0,0 +1,737 @@
+/*
+ * 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 mostly implements UBI kernel API functions */
+
+#include <linux/module.h>
+#include <linux/err.h>
+#include <asm/div64.h>
+#include "ubi.h"
+
+/**
+ * ubi_do_get_device_info - get information about UBI device.
+ * @ubi: UBI device description object
+ * @di: the information is stored here
+ *
+ * This function is the same as 'ubi_get_device_info()', but it assumes the UBI
+ * device is locked and cannot disappear.
+ */
+void ubi_do_get_device_info(struct ubi_device *ubi, struct ubi_device_info *di)
+{
+	di->ubi_num = ubi->ubi_num;
+	di->leb_size = ubi->leb_size;
+	di->min_io_size = ubi->min_io_size;
+	di->ro_mode = ubi->ro_mode;
+	di->cdev = ubi->cdev.dev;
+}
+EXPORT_SYMBOL_GPL(ubi_do_get_device_info);
+
+/**
+ * ubi_get_device_info - get information about UBI device.
+ * @ubi_num: UBI device number
+ * @di: the information is stored here
+ *
+ * This function returns %0 in case of success, %-EINVAL if the UBI device
+ * number is invalid, and %-ENODEV if there is no such UBI device.
+ */
+int ubi_get_device_info(int ubi_num, struct ubi_device_info *di)
+{
+	struct ubi_device *ubi;
+
+	if (ubi_num < 0 || ubi_num >= UBI_MAX_DEVICES)
+		return -EINVAL;
+	ubi = ubi_get_device(ubi_num);
+	if (!ubi)
+		return -ENODEV;
+	ubi_do_get_device_info(ubi, di);
+	ubi_put_device(ubi);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ubi_get_device_info);
+
+/**
+ * ubi_do_get_volume_info - get information about UBI volume.
+ * @ubi: UBI device description object
+ * @vol: volume description object
+ * @vi: the information is stored here
+ */
+void ubi_do_get_volume_info(struct ubi_device *ubi, struct ubi_volume *vol,
+			    struct ubi_volume_info *vi)
+{
+	vi->vol_id = vol->vol_id;
+	vi->ubi_num = ubi->ubi_num;
+	vi->size = vol->reserved_pebs;
+	vi->used_bytes = vol->used_bytes;
+	vi->vol_type = vol->vol_type;
+	vi->corrupted = vol->corrupted;
+	vi->upd_marker = vol->upd_marker;
+	vi->alignment = vol->alignment;
+	vi->usable_leb_size = vol->usable_leb_size;
+	vi->name_len = vol->name_len;
+	vi->name = vol->name;
+	vi->cdev = vol->cdev.dev;
+}
+
+/**
+ * ubi_get_volume_info - get information about UBI volume.
+ * @desc: volume descriptor
+ * @vi: the information is stored here
+ */
+void ubi_get_volume_info(struct ubi_volume_desc *desc,
+			 struct ubi_volume_info *vi)
+{
+	ubi_do_get_volume_info(desc->vol->ubi, desc->vol, vi);
+}
+EXPORT_SYMBOL_GPL(ubi_get_volume_info);
+
+/**
+ * ubi_open_volume - open UBI volume.
+ * @ubi_num: UBI device number
+ * @vol_id: volume ID
+ * @mode: open mode
+ *
+ * The @mode parameter specifies if the volume should be opened in read-only
+ * mode, read-write mode, or exclusive mode. The exclusive mode guarantees that
+ * nobody else will be able to open this volume. UBI allows to have many volume
+ * readers and one writer at a time.
+ *
+ * If a static volume is being opened for the first time since boot, it will be
+ * checked by this function, which means it will be fully read and the CRC
+ * checksum of each logical eraseblock will be checked.
+ *
+ * This function returns volume descriptor in case of success and a negative
+ * error code in case of failure.
+ */
+struct ubi_volume_desc *ubi_open_volume(int ubi_num, int vol_id, int mode)
+{
+	int err;
+	struct ubi_volume_desc *desc;
+	struct ubi_device *ubi;
+	struct ubi_volume *vol;
+
+	dbg_gen("open device %d, volume %d, mode %d", ubi_num, vol_id, mode);
+
+	if (ubi_num < 0 || ubi_num >= UBI_MAX_DEVICES)
+		return ERR_PTR(-EINVAL);
+
+	if (mode != UBI_READONLY && mode != UBI_READWRITE &&
+	    mode != UBI_EXCLUSIVE)
+		return ERR_PTR(-EINVAL);
+
+	/*
+	 * First of all, we have to get the UBI device to prevent its removal.
+	 */
+	ubi = ubi_get_device(ubi_num);
+	if (!ubi)
+		return ERR_PTR(-ENODEV);
+
+	if (vol_id < 0 || vol_id >= ubi->vtbl_slots) {
+		err = -EINVAL;
+		goto out_put_ubi;
+	}
+
+	desc = kmalloc(sizeof(struct ubi_volume_desc), GFP_KERNEL);
+	if (!desc) {
+		err = -ENOMEM;
+		goto out_put_ubi;
+	}
+
+	err = -ENODEV;
+	if (!try_module_get(THIS_MODULE))
+		goto out_free;
+
+	spin_lock(&ubi->volumes_lock);
+	vol = ubi->volumes[vol_id];
+	if (!vol)
+		goto out_unlock;
+
+	err = -EBUSY;
+	switch (mode) {
+	case UBI_READONLY:
+		if (vol->exclusive)
+			goto out_unlock;
+		vol->readers += 1;
+		break;
+
+	case UBI_READWRITE:
+		if (vol->exclusive || vol->writers > 0)
+			goto out_unlock;
+		vol->writers += 1;
+		break;
+
+	case UBI_EXCLUSIVE:
+		if (vol->exclusive || vol->writers || vol->readers)
+			goto out_unlock;
+		vol->exclusive = 1;
+		break;
+	}
+	get_device(&vol->dev);
+	vol->ref_count += 1;
+	spin_unlock(&ubi->volumes_lock);
+
+	desc->vol = vol;
+	desc->mode = mode;
+
+	mutex_lock(&ubi->ckvol_mutex);
+	if (!vol->checked) {
+		/* This is the first open - check the volume */
+		err = ubi_check_volume(ubi, vol_id);
+		if (err < 0) {
+			mutex_unlock(&ubi->ckvol_mutex);
+			ubi_close_volume(desc);
+			return ERR_PTR(err);
+		}
+		if (err == 1) {
+			ubi_warn("volume %d on UBI device %d is corrupted",
+				 vol_id, ubi->ubi_num);
+			vol->corrupted = 1;
+		}
+		vol->checked = 1;
+	}
+	mutex_unlock(&ubi->ckvol_mutex);
+
+	return desc;
+
+out_unlock:
+	spin_unlock(&ubi->volumes_lock);
+	module_put(THIS_MODULE);
+out_free:
+	kfree(desc);
+out_put_ubi:
+	ubi_put_device(ubi);
+	dbg_err("cannot open device %d, volume %d, error %d",
+		ubi_num, vol_id, err);
+	return ERR_PTR(err);
+}
+EXPORT_SYMBOL_GPL(ubi_open_volume);
+
+/**
+ * ubi_open_volume_nm - open UBI volume by name.
+ * @ubi_num: UBI device number
+ * @name: volume name
+ * @mode: open mode
+ *
+ * This function is similar to 'ubi_open_volume()', but opens a volume by name.
+ */
+struct ubi_volume_desc *ubi_open_volume_nm(int ubi_num, const char *name,
+					   int mode)
+{
+	int i, vol_id = -1, len;
+	struct ubi_device *ubi;
+	struct ubi_volume_desc *ret;
+
+	dbg_gen("open device %d, volume %s, mode %d", ubi_num, name, mode);
+
+	if (!name)
+		return ERR_PTR(-EINVAL);
+
+	len = strnlen(name, UBI_VOL_NAME_MAX + 1);
+	if (len > UBI_VOL_NAME_MAX)
+		return ERR_PTR(-EINVAL);
+
+	if (ubi_num < 0 || ubi_num >= UBI_MAX_DEVICES)
+		return ERR_PTR(-EINVAL);
+
+	ubi = ubi_get_device(ubi_num);
+	if (!ubi)
+		return ERR_PTR(-ENODEV);
+
+	spin_lock(&ubi->volumes_lock);
+	/* Walk all volumes of this UBI device */
+	for (i = 0; i < ubi->vtbl_slots; i++) {
+		struct ubi_volume *vol = ubi->volumes[i];
+
+		if (vol && len == vol->name_len && !strcmp(name, vol->name)) {
+			vol_id = i;
+			break;
+		}
+	}
+	spin_unlock(&ubi->volumes_lock);
+
+	if (vol_id >= 0)
+		ret = ubi_open_volume(ubi_num, vol_id, mode);
+	else
+		ret = ERR_PTR(-ENODEV);
+
+	/*
+	 * We should put the UBI device even in case of success, because
+	 * 'ubi_open_volume()' took a reference as well.
+	 */
+	ubi_put_device(ubi);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(ubi_open_volume_nm);
+
+/**
+ * ubi_close_volume - close UBI volume.
+ * @desc: volume descriptor
+ */
+void ubi_close_volume(struct ubi_volume_desc *desc)
+{
+	struct ubi_volume *vol = desc->vol;
+	struct ubi_device *ubi = vol->ubi;
+
+	dbg_gen("close device %d, volume %d, mode %d",
+		ubi->ubi_num, vol->vol_id, desc->mode);
+
+	spin_lock(&ubi->volumes_lock);
+	switch (desc->mode) {
+	case UBI_READONLY:
+		vol->readers -= 1;
+		break;
+	case UBI_READWRITE:
+		vol->writers -= 1;
+		break;
+	case UBI_EXCLUSIVE:
+		vol->exclusive = 0;
+	}
+	vol->ref_count -= 1;
+	spin_unlock(&ubi->volumes_lock);
+
+	kfree(desc);
+	put_device(&vol->dev);
+	ubi_put_device(ubi);
+	module_put(THIS_MODULE);
+}
+EXPORT_SYMBOL_GPL(ubi_close_volume);
+
+/**
+ * ubi_leb_read - read data.
+ * @desc: volume descriptor
+ * @lnum: logical eraseblock number to read from
+ * @buf: buffer where to store the read data
+ * @offset: offset within the logical eraseblock to read from
+ * @len: how many bytes to read
+ * @check: whether UBI has to check the read data's CRC or not.
+ *
+ * This function reads data from offset @offset of logical eraseblock @lnum and
+ * stores the data at @buf. When reading from static volumes, @check specifies
+ * whether the data has to be checked or not. If yes, the whole logical
+ * eraseblock will be read and its CRC checksum will be checked (i.e., the CRC
+ * checksum is per-eraseblock). So checking may substantially slow down the
+ * read speed. The @check argument is ignored for dynamic volumes.
+ *
+ * In case of success, this function returns zero. In case of failure, this
+ * function returns a negative error code.
+ *
+ * %-EBADMSG error code is returned:
+ * o for both static and dynamic volumes if MTD driver has detected a data
+ *   integrity problem (unrecoverable ECC checksum mismatch in case of NAND);
+ * o for static volumes in case of data CRC mismatch.
+ *
+ * If the volume is damaged because of an interrupted update this function just
+ * returns immediately with %-EBADF error code.
+ */
+int ubi_leb_read(struct ubi_volume_desc *desc, int lnum, char *buf, int offset,
+		 int len, int check)
+{
+	struct ubi_volume *vol = desc->vol;
+	struct ubi_device *ubi = vol->ubi;
+	int err, vol_id = vol->vol_id;
+
+	dbg_gen("read %d bytes from LEB %d:%d:%d", len, vol_id, lnum, offset);
+
+	if (vol_id < 0 || vol_id >= ubi->vtbl_slots || lnum < 0 ||
+	    lnum >= vol->used_ebs || offset < 0 || len < 0 ||
+	    offset + len > vol->usable_leb_size)
+		return -EINVAL;
+
+	if (vol->vol_type == UBI_STATIC_VOLUME) {
+		if (vol->used_ebs == 0)
+			/* Empty static UBI volume */
+			return 0;
+		if (lnum == vol->used_ebs - 1 &&
+		    offset + len > vol->last_eb_bytes)
+			return -EINVAL;
+	}
+
+	if (vol->upd_marker)
+		return -EBADF;
+	if (len == 0)
+		return 0;
+
+	err = ubi_eba_read_leb(ubi, vol, lnum, buf, offset, len, check);
+	if (err && err == -EBADMSG && vol->vol_type == UBI_STATIC_VOLUME) {
+		ubi_warn("mark volume %d as corrupted", vol_id);
+		vol->corrupted = 1;
+	}
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(ubi_leb_read);
+
+/**
+ * ubi_leb_write - write data.
+ * @desc: volume descriptor
+ * @lnum: logical eraseblock number to write to
+ * @buf: data to write
+ * @offset: offset within the logical eraseblock where to write
+ * @len: how many bytes to write
+ * @dtype: expected data type
+ *
+ * This function writes @len bytes of data from @buf to offset @offset of
+ * logical eraseblock @lnum. The @dtype argument describes expected lifetime of
+ * the data.
+ *
+ * This function takes care of physical eraseblock write failures. If write to
+ * the physical eraseblock write operation fails, the logical eraseblock is
+ * re-mapped to another physical eraseblock, the data is recovered, and the
+ * write finishes. UBI has a pool of reserved physical eraseblocks for this.
+ *
+ * If all the data were successfully written, zero is returned. If an error
+ * occurred and UBI has not been able to recover from it, this function returns
+ * a negative error code. Note, in case of an error, it is possible that
+ * something was still written to the flash media, but that may be some
+ * garbage.
+ *
+ * If the volume is damaged because of an interrupted update this function just
+ * returns immediately with %-EBADF code.
+ */
+int ubi_leb_write(struct ubi_volume_desc *desc, int lnum, const void *buf,
+		  int offset, int len, int dtype)
+{
+	struct ubi_volume *vol = desc->vol;
+	struct ubi_device *ubi = vol->ubi;
+	int vol_id = vol->vol_id;
+
+	dbg_gen("write %d bytes to LEB %d:%d:%d", len, vol_id, lnum, offset);
+
+	if (vol_id < 0 || vol_id >= ubi->vtbl_slots)
+		return -EINVAL;
+
+	if (desc->mode == UBI_READONLY || vol->vol_type == UBI_STATIC_VOLUME)
+		return -EROFS;
+
+	if (lnum < 0 || lnum >= vol->reserved_pebs || offset < 0 || len < 0 ||
+	    offset + len > vol->usable_leb_size ||
+	    offset & (ubi->min_io_size - 1) || len & (ubi->min_io_size - 1))
+		return -EINVAL;
+
+	if (dtype != UBI_LONGTERM && dtype != UBI_SHORTTERM &&
+	    dtype != UBI_UNKNOWN)
+		return -EINVAL;
+
+	if (vol->upd_marker)
+		return -EBADF;
+
+	if (len == 0)
+		return 0;
+
+	return ubi_eba_write_leb(ubi, vol, lnum, buf, offset, len, dtype);
+}
+EXPORT_SYMBOL_GPL(ubi_leb_write);
+
+/*
+ * ubi_leb_change - change logical eraseblock atomically.
+ * @desc: volume descriptor
+ * @lnum: logical eraseblock number to change
+ * @buf: data to write
+ * @len: how many bytes to write
+ * @dtype: expected data type
+ *
+ * This function changes the contents of a logical eraseblock atomically. @buf
+ * has to contain new logical eraseblock data, and @len - the length of the
+ * data, which has to be aligned. The length may be shorter then the logical
+ * eraseblock size, ant the logical eraseblock may be appended to more times
+ * later on. This function guarantees that in case of an unclean reboot the old
+ * contents is preserved. Returns zero in case of success and a negative error
+ * code in case of failure.
+ */
+int ubi_leb_change(struct ubi_volume_desc *desc, int lnum, const void *buf,
+		   int len, int dtype)
+{
+	struct ubi_volume *vol = desc->vol;
+	struct ubi_device *ubi = vol->ubi;
+	int vol_id = vol->vol_id;
+
+	dbg_gen("atomically write %d bytes to LEB %d:%d", len, vol_id, lnum);
+
+	if (vol_id < 0 || vol_id >= ubi->vtbl_slots)
+		return -EINVAL;
+
+	if (desc->mode == UBI_READONLY || vol->vol_type == UBI_STATIC_VOLUME)
+		return -EROFS;
+
+	if (lnum < 0 || lnum >= vol->reserved_pebs || len < 0 ||
+	    len > vol->usable_leb_size || len & (ubi->min_io_size - 1))
+		return -EINVAL;
+
+	if (dtype != UBI_LONGTERM && dtype != UBI_SHORTTERM &&
+	    dtype != UBI_UNKNOWN)
+		return -EINVAL;
+
+	if (vol->upd_marker)
+		return -EBADF;
+
+	if (len == 0)
+		return 0;
+
+	return ubi_eba_atomic_leb_change(ubi, vol, lnum, buf, len, dtype);
+}
+EXPORT_SYMBOL_GPL(ubi_leb_change);
+
+/**
+ * ubi_leb_erase - erase logical eraseblock.
+ * @desc: volume descriptor
+ * @lnum: logical eraseblock number
+ *
+ * This function un-maps logical eraseblock @lnum and synchronously erases the
+ * correspondent physical eraseblock. Returns zero in case of success and a
+ * negative error code in case of failure.
+ *
+ * If the volume is damaged because of an interrupted update this function just
+ * returns immediately with %-EBADF code.
+ */
+int ubi_leb_erase(struct ubi_volume_desc *desc, int lnum)
+{
+	struct ubi_volume *vol = desc->vol;
+	struct ubi_device *ubi = vol->ubi;
+	int err;
+
+	dbg_gen("erase LEB %d:%d", vol->vol_id, lnum);
+
+	if (desc->mode == UBI_READONLY || vol->vol_type == UBI_STATIC_VOLUME)
+		return -EROFS;
+
+	if (lnum < 0 || lnum >= vol->reserved_pebs)
+		return -EINVAL;
+
+	if (vol->upd_marker)
+		return -EBADF;
+
+	err = ubi_eba_unmap_leb(ubi, vol, lnum);
+	if (err)
+		return err;
+
+	return ubi_wl_flush(ubi);
+}
+EXPORT_SYMBOL_GPL(ubi_leb_erase);
+
+/**
+ * ubi_leb_unmap - un-map logical eraseblock.
+ * @desc: volume descriptor
+ * @lnum: logical eraseblock number
+ *
+ * This function un-maps logical eraseblock @lnum and schedules the
+ * corresponding physical eraseblock for erasure, so that it will eventually be
+ * physically erased in background. This operation is much faster then the
+ * erase operation.
+ *
+ * Unlike erase, the un-map operation does not guarantee that the logical
+ * eraseblock will contain all 0xFF bytes when UBI is initialized again. For
+ * example, if several logical eraseblocks are un-mapped, and an unclean reboot
+ * happens after this, the logical eraseblocks will not necessarily be
+ * un-mapped again when this MTD device is attached. They may actually be
+ * mapped to the same physical eraseblocks again. So, this function has to be
+ * used with care.
+ *
+ * In other words, when un-mapping a logical eraseblock, UBI does not store
+ * any information about this on the flash media, it just marks the logical
+ * eraseblock as "un-mapped" in RAM. If UBI is detached before the physical
+ * eraseblock is physically erased, it will be mapped again to the same logical
+ * eraseblock when the MTD device is attached again.
+ *
+ * The main and obvious use-case of this function is when the contents of a
+ * logical eraseblock has to be re-written. Then it is much more efficient to
+ * first un-map it, then write new data, rather then first erase it, then write
+ * new data. Note, once new data has been written to the logical eraseblock,
+ * UBI guarantees that the old contents has gone forever. In other words, if an
+ * unclean reboot happens after the logical eraseblock has been un-mapped and
+ * then written to, it will contain the last written data.
+ *
+ * This function returns zero in case of success and a negative error code in
+ * case of failure. If the volume is damaged because of an interrupted update
+ * this function just returns immediately with %-EBADF code.
+ */
+int ubi_leb_unmap(struct ubi_volume_desc *desc, int lnum)
+{
+	struct ubi_volume *vol = desc->vol;
+	struct ubi_device *ubi = vol->ubi;
+
+	dbg_gen("unmap LEB %d:%d", vol->vol_id, lnum);
+
+	if (desc->mode == UBI_READONLY || vol->vol_type == UBI_STATIC_VOLUME)
+		return -EROFS;
+
+	if (lnum < 0 || lnum >= vol->reserved_pebs)
+		return -EINVAL;
+
+	if (vol->upd_marker)
+		return -EBADF;
+
+	return ubi_eba_unmap_leb(ubi, vol, lnum);
+}
+EXPORT_SYMBOL_GPL(ubi_leb_unmap);
+
+/**
+ * ubi_leb_map - map logical eraseblock to a physical eraseblock.
+ * @desc: volume descriptor
+ * @lnum: logical eraseblock number
+ * @dtype: expected data type
+ *
+ * This function maps an un-mapped logical eraseblock @lnum to a physical
+ * eraseblock. This means, that after a successful invocation of this
+ * function the logical eraseblock @lnum will be empty (contain only %0xFF
+ * bytes) and be mapped to a physical eraseblock, even if an unclean reboot
+ * happens.
+ *
+ * This function returns zero in case of success, %-EBADF if the volume is
+ * damaged because of an interrupted update, %-EBADMSG if the logical
+ * eraseblock is already mapped, and other negative error codes in case of
+ * other failures.
+ */
+int ubi_leb_map(struct ubi_volume_desc *desc, int lnum, int dtype)
+{
+	struct ubi_volume *vol = desc->vol;
+	struct ubi_device *ubi = vol->ubi;
+
+	dbg_gen("unmap LEB %d:%d", vol->vol_id, lnum);
+
+	if (desc->mode == UBI_READONLY || vol->vol_type == UBI_STATIC_VOLUME)
+		return -EROFS;
+
+	if (lnum < 0 || lnum >= vol->reserved_pebs)
+		return -EINVAL;
+
+	if (dtype != UBI_LONGTERM && dtype != UBI_SHORTTERM &&
+	    dtype != UBI_UNKNOWN)
+		return -EINVAL;
+
+	if (vol->upd_marker)
+		return -EBADF;
+
+	if (vol->eba_tbl[lnum] >= 0)
+		return -EBADMSG;
+
+	return ubi_eba_write_leb(ubi, vol, lnum, NULL, 0, 0, dtype);
+}
+EXPORT_SYMBOL_GPL(ubi_leb_map);
+
+/**
+ * ubi_is_mapped - check if logical eraseblock is mapped.
+ * @desc: volume descriptor
+ * @lnum: logical eraseblock number
+ *
+ * This function checks if logical eraseblock @lnum is mapped to a physical
+ * eraseblock. If a logical eraseblock is un-mapped, this does not necessarily
+ * mean it will still be un-mapped after the UBI device is re-attached. The
+ * logical eraseblock may become mapped to the physical eraseblock it was last
+ * mapped to.
+ *
+ * This function returns %1 if the LEB is mapped, %0 if not, and a negative
+ * error code in case of failure. If the volume is damaged because of an
+ * interrupted update this function just returns immediately with %-EBADF error
+ * code.
+ */
+int ubi_is_mapped(struct ubi_volume_desc *desc, int lnum)
+{
+	struct ubi_volume *vol = desc->vol;
+
+	dbg_gen("test LEB %d:%d", vol->vol_id, lnum);
+
+	if (lnum < 0 || lnum >= vol->reserved_pebs)
+		return -EINVAL;
+
+	if (vol->upd_marker)
+		return -EBADF;
+
+	return vol->eba_tbl[lnum] >= 0;
+}
+EXPORT_SYMBOL_GPL(ubi_is_mapped);
+
+/**
+ * ubi_sync - synchronize UBI device buffers.
+ * @ubi_num: UBI device to synchronize
+ *
+ * The underlying MTD device may cache data in hardware or in software. This
+ * function ensures the caches are flushed. Returns zero in case of success and
+ * a negative error code in case of failure.
+ */
+int ubi_sync(int ubi_num)
+{
+	struct ubi_device *ubi;
+
+	ubi = ubi_get_device(ubi_num);
+	if (!ubi)
+		return -ENODEV;
+
+	if (ubi->mtd->sync)
+		ubi->mtd->sync(ubi->mtd);
+
+	ubi_put_device(ubi);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ubi_sync);
+
+BLOCKING_NOTIFIER_HEAD(ubi_notifiers);
+
+/**
+ * ubi_register_volume_notifier - register a volume notifier.
+ * @nb: the notifier description object
+ * @ignore_existing: if non-zero, do not send "added" notification for all
+ *                   already existing volumes
+ *
+ * This function registers a volume notifier, which means that
+ * 'nb->notifier_call()' will be invoked when an UBI  volume is created,
+ * removed, re-sized, re-named, or updated. The first argument of the function
+ * is the notification type. The second argument is pointer to a
+ * &struct ubi_notification object which describes the notification event.
+ * Using UBI API from the volume notifier is prohibited.
+ *
+ * This function returns zero in case of success and a negative error code
+ * in case of failure.
+ */
+int ubi_register_volume_notifier(struct notifier_block *nb,
+				 int ignore_existing)
+{
+	int err;
+
+	err = blocking_notifier_chain_register(&ubi_notifiers, nb);
+	if (err != 0)
+		return err;
+	if (ignore_existing)
+		return 0;
+
+	/*
+	 * We are going to walk all UBI devices and all volumes, and
+	 * notify the user about existing volumes by the %UBI_VOLUME_ADDED
+	 * event. We have to lock the @ubi_devices_mutex to make sure UBI
+	 * devices do not disappear.
+	 */
+	mutex_lock(&ubi_devices_mutex);
+	ubi_enumerate_volumes(nb);
+	mutex_unlock(&ubi_devices_mutex);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(ubi_register_volume_notifier);
+
+/**
+ * ubi_unregister_volume_notifier - unregister the volume notifier.
+ * @nb: the notifier description object
+ *
+ * This function unregisters volume notifier @nm and returns zero in case of
+ * success and a negative error code in case of failure.
+ */
+int ubi_unregister_volume_notifier(struct notifier_block *nb)
+{
+	return blocking_notifier_chain_unregister(&ubi_notifiers, nb);
+}
+EXPORT_SYMBOL_GPL(ubi_unregister_volume_notifier);

kernel/drivers/mtd/ubi/misc.c

@@ -0,0 +1,105 @@
+/*
+ * 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)
+ */
+
+/* Here we keep miscellaneous functions which are used all over the UBI code */
+
+#include "ubi.h"
+
+/**
+ * calc_data_len - calculate how much real data is stored in a buffer.
+ * @ubi: UBI device description object
+ * @buf: a buffer with the contents of the physical eraseblock
+ * @length: the buffer length
+ *
+ * This function calculates how much "real data" is stored in @buf and returnes
+ * the length. Continuous 0xFF bytes at the end of the buffer are not
+ * considered as "real data".
+ */
+int ubi_calc_data_len(const struct ubi_device *ubi, const void *buf,
+		      int length)
+{
+	int i;
+
+	ubi_assert(!(length & (ubi->min_io_size - 1)));
+
+	for (i = length - 1; i >= 0; i--)
+		if (((const uint8_t *)buf)[i] != 0xFF)
+			break;
+
+	/* The resulting length must be aligned to the minimum flash I/O size */
+	length = ALIGN(i + 1, ubi->min_io_size);
+	return length;
+}
+
+/**
+ * ubi_check_volume - check the contents of a static volume.
+ * @ubi: UBI device description object
+ * @vol_id: ID of the volume to check
+ *
+ * This function checks if static volume @vol_id is corrupted by fully reading
+ * it and checking data CRC. This function returns %0 if the volume is not
+ * corrupted, %1 if it is corrupted and a negative error code in case of
+ * failure. Dynamic volumes are not checked and zero is returned immediately.
+ */
+int ubi_check_volume(struct ubi_device *ubi, int vol_id)
+{
+	void *buf;
+	int err = 0, i;
+	struct ubi_volume *vol = ubi->volumes[vol_id];
+
+	if (vol->vol_type != UBI_STATIC_VOLUME)
+		return 0;
+
+	buf = vmalloc(vol->usable_leb_size);
+	if (!buf)
+		return -ENOMEM;
+
+	for (i = 0; i < vol->used_ebs; i++) {
+		int size;
+
+		if (i == vol->used_ebs - 1)
+			size = vol->last_eb_bytes;
+		else
+			size = vol->usable_leb_size;
+
+		err = ubi_eba_read_leb(ubi, vol, i, buf, 0, size, 1);
+		if (err) {
+			if (err == -EBADMSG)
+				err = 1;
+			break;
+		}
+	}
+
+	vfree(buf);
+	return err;
+}
+
+/**
+ * ubi_calculate_rsvd_pool - calculate how many PEBs must be reserved for bad
+ * eraseblock handling.
+ * @ubi: UBI device description object
+ */
+void ubi_calculate_reserved(struct ubi_device *ubi)
+{
+	ubi->beb_rsvd_level = ubi->good_peb_count/100;
+	ubi->beb_rsvd_level *= CONFIG_MTD_UBI_BEB_RESERVE;
+	if (ubi->beb_rsvd_level < MIN_RESEVED_PEBS)
+		ubi->beb_rsvd_level = MIN_RESEVED_PEBS;
+}

kernel/drivers/mtd/ubi/scan.h

@@ -0,0 +1,164 @@
+/*
+ * 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)
+ */
+
+#ifndef __UBI_SCAN_H__
+#define __UBI_SCAN_H__
+
+/* The erase counter value for this physical eraseblock is unknown */
+#define UBI_SCAN_UNKNOWN_EC (-1)
+
+/**
+ * struct ubi_scan_leb - scanning information about a physical eraseblock.
+ * @ec: erase counter (%UBI_SCAN_UNKNOWN_EC if it is unknown)
+ * @pnum: physical eraseblock number
+ * @lnum: logical eraseblock number
+ * @scrub: if this physical eraseblock needs scrubbing
+ * @sqnum: sequence number
+ * @u: unions RB-tree or @list links
+ * @u.rb: link in the per-volume RB-tree of &struct ubi_scan_leb objects
+ * @u.list: link in one of the eraseblock lists
+ *
+ * One object of this type is allocated for each physical eraseblock during
+ * scanning.
+ */
+struct ubi_scan_leb {
+	int ec;
+	int pnum;
+	int lnum;
+	int scrub;
+	unsigned long long sqnum;
+	union {
+		struct rb_node rb;
+		struct list_head list;
+	} u;
+};
+
+/**
+ * struct ubi_scan_volume - scanning information about a volume.
+ * @vol_id: volume ID
+ * @highest_lnum: highest logical eraseblock number in this volume
+ * @leb_count: number of logical eraseblocks in this volume
+ * @vol_type: volume type
+ * @used_ebs: number of used logical eraseblocks in this volume (only for
+ *            static volumes)
+ * @last_data_size: amount of data in the last logical eraseblock of this
+ *                  volume (always equivalent to the usable logical eraseblock
+ *                  size in case of dynamic volumes)
+ * @data_pad: how many bytes at the end of logical eraseblocks of this volume
+ *            are not used (due to volume alignment)
+ * @compat: compatibility flags of this volume
+ * @rb: link in the volume RB-tree
+ * @root: root of the RB-tree containing all the eraseblock belonging to this
+ *        volume (&struct ubi_scan_leb objects)
+ *
+ * One object of this type is allocated for each volume during scanning.
+ */
+struct ubi_scan_volume {
+	int vol_id;
+	int highest_lnum;
+	int leb_count;
+	int vol_type;
+	int used_ebs;
+	int last_data_size;
+	int data_pad;
+	int compat;
+	struct rb_node rb;
+	struct rb_root root;
+};
+
+/**
+ * struct ubi_scan_info - UBI scanning information.
+ * @volumes: root of the volume RB-tree
+ * @corr: list of corrupted physical eraseblocks
+ * @free: list of free physical eraseblocks
+ * @erase: list of physical eraseblocks which have to be erased
+ * @alien: list of physical eraseblocks which should not be used by UBI (e.g.,
+ *         those belonging to "preserve"-compatible internal volumes)
+ * @bad_peb_count: count of bad physical eraseblocks
+ * @vols_found: number of volumes found during scanning
+ * @highest_vol_id: highest volume ID
+ * @alien_peb_count: count of physical eraseblocks in the @alien list
+ * @is_empty: flag indicating whether the MTD device is empty or not
+ * @min_ec: lowest erase counter value
+ * @max_ec: highest erase counter value
+ * @max_sqnum: highest sequence number value
+ * @mean_ec: mean erase counter value
+ * @ec_sum: a temporary variable used when calculating @mean_ec
+ * @ec_count: a temporary variable used when calculating @mean_ec
+ * @corr_count: count of corrupted PEBs
+ *
+ * This data structure contains the result of scanning and may be used by other
+ * UBI sub-systems to build final UBI data structures, further error-recovery
+ * and so on.
+ */
+struct ubi_scan_info {
+	struct rb_root volumes;
+	struct list_head corr;
+	struct list_head free;
+	struct list_head erase;
+	struct list_head alien;
+	int bad_peb_count;
+	int vols_found;
+	int highest_vol_id;
+	int alien_peb_count;
+	int is_empty;
+	int min_ec;
+	int max_ec;
+	unsigned long long max_sqnum;
+	int mean_ec;
+	uint64_t ec_sum;
+	int ec_count;
+	int corr_count;
+};
+
+struct ubi_device;
+struct ubi_vid_hdr;
+
+/*
+ * ubi_scan_move_to_list - move a PEB from the volume tree to a list.
+ *
+ * @sv: volume scanning information
+ * @seb: scanning eraseblock infprmation
+ * @list: the list to move to
+ */
+static inline void ubi_scan_move_to_list(struct ubi_scan_volume *sv,
+					 struct ubi_scan_leb *seb,
+					 struct list_head *list)
+{
+		rb_erase(&seb->u.rb, &sv->root);
+		list_add_tail(&seb->u.list, list);
+}
+
+int ubi_scan_add_used(struct ubi_device *ubi, struct ubi_scan_info *si,
+		      int pnum, int ec, const struct ubi_vid_hdr *vid_hdr,
+		      int bitflips);
+struct ubi_scan_volume *ubi_scan_find_sv(const struct ubi_scan_info *si,
+					 int vol_id);
+struct ubi_scan_leb *ubi_scan_find_seb(const struct ubi_scan_volume *sv,
+				       int lnum);
+void ubi_scan_rm_volume(struct ubi_scan_info *si, struct ubi_scan_volume *sv);
+struct ubi_scan_leb *ubi_scan_get_free_peb(struct ubi_device *ubi,
+					   struct ubi_scan_info *si);
+int ubi_scan_erase_peb(struct ubi_device *ubi, const struct ubi_scan_info *si,
+		       int pnum, int ec);
+struct ubi_scan_info *ubi_scan(struct ubi_device *ubi);
+void ubi_scan_destroy_si(struct ubi_scan_info *si);
+
+#endif /* !__UBI_SCAN_H__ */

kernel/drivers/mtd/ubi/ubi-media.h

@@ -0,0 +1,378 @@
+/*
+ * 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
+ *          Frank Haverkamp
+ *          Oliver Lohmann
+ *          Andreas Arnez
+ */
+
+/*
+ * This file defines the layout of UBI headers and all the other UBI on-flash
+ * data structures.
+ */
+
+#ifndef __UBI_MEDIA_H__
+#define __UBI_MEDIA_H__
+
+#include <asm/byteorder.h>
+
+/* The version of UBI images supported by this implementation */
+#define UBI_VERSION 1
+
+/* The highest erase counter value supported by this implementation */
+#define UBI_MAX_ERASECOUNTER 0x7FFFFFFF
+
+/* The initial CRC32 value used when calculating CRC checksums */
+#define UBI_CRC32_INIT 0xFFFFFFFFU
+
+/* Erase counter header magic number (ASCII "UBI#") */
+#define UBI_EC_HDR_MAGIC  0x55424923
+/* Volume identifier header magic number (ASCII "UBI!") */
+#define UBI_VID_HDR_MAGIC 0x55424921
+
+/*
+ * Volume type constants used in the volume identifier header.
+ *
+ * @UBI_VID_DYNAMIC: dynamic volume
+ * @UBI_VID_STATIC: static volume
+ */
+enum {
+	UBI_VID_DYNAMIC = 1,
+	UBI_VID_STATIC  = 2
+};
+
+/*
+ * Volume flags used in the volume table record.
+ *
+ * @UBI_VTBL_AUTORESIZE_FLG: auto-resize this volume
+ *
+ * %UBI_VTBL_AUTORESIZE_FLG flag can be set only for one volume in the volume
+ * table. UBI automatically re-sizes the volume which has this flag and makes
+ * the volume to be of largest possible size. This means that if after the
+ * initialization UBI finds out that there are available physical eraseblocks
+ * present on the device, it automatically appends all of them to the volume
+ * (the physical eraseblocks reserved for bad eraseblocks handling and other
+ * reserved physical eraseblocks are not taken). So, if there is a volume with
+ * the %UBI_VTBL_AUTORESIZE_FLG flag set, the amount of available logical
+ * eraseblocks will be zero after UBI is loaded, because all of them will be
+ * reserved for this volume. Note, the %UBI_VTBL_AUTORESIZE_FLG bit is cleared
+ * after the volume had been initialized.
+ *
+ * The auto-resize feature is useful for device production purposes. For
+ * example, different NAND flash chips may have different amount of initial bad
+ * eraseblocks, depending of particular chip instance. Manufacturers of NAND
+ * chips usually guarantee that the amount of initial bad eraseblocks does not
+ * exceed certain percent, e.g. 2%. When one creates an UBI image which will be
+ * flashed to the end devices in production, he does not know the exact amount
+ * of good physical eraseblocks the NAND chip on the device will have, but this
+ * number is required to calculate the volume sized and put them to the volume
+ * table of the UBI image. In this case, one of the volumes (e.g., the one
+ * which will store the root file system) is marked as "auto-resizable", and
+ * UBI will adjust its size on the first boot if needed.
+ *
+ * Note, first UBI reserves some amount of physical eraseblocks for bad
+ * eraseblock handling, and then re-sizes the volume, not vice-versa. This
+ * means that the pool of reserved physical eraseblocks will always be present.
+ */
+enum {
+	UBI_VTBL_AUTORESIZE_FLG = 0x01,
+};
+
+/*
+ * Compatibility constants used by internal volumes.
+ *
+ * @UBI_COMPAT_DELETE: delete this internal volume before anything is written
+ *                     to the flash
+ * @UBI_COMPAT_RO: attach this device in read-only mode
+ * @UBI_COMPAT_PRESERVE: preserve this internal volume - do not touch its
+ *                       physical eraseblocks, don't allow the wear-leveling
+ *                       sub-system to move them
+ * @UBI_COMPAT_REJECT: reject this UBI image
+ */
+enum {
+	UBI_COMPAT_DELETE   = 1,
+	UBI_COMPAT_RO       = 2,
+	UBI_COMPAT_PRESERVE = 4,
+	UBI_COMPAT_REJECT   = 5
+};
+
+/* Sizes of UBI headers */
+#define UBI_EC_HDR_SIZE  sizeof(struct ubi_ec_hdr)
+#define UBI_VID_HDR_SIZE sizeof(struct ubi_vid_hdr)
+
+/* Sizes of UBI headers without the ending CRC */
+#define UBI_EC_HDR_SIZE_CRC  (UBI_EC_HDR_SIZE  - sizeof(__be32))
+#define UBI_VID_HDR_SIZE_CRC (UBI_VID_HDR_SIZE - sizeof(__be32))
+
+/**
+ * struct ubi_ec_hdr - UBI erase counter header.
+ * @magic: erase counter header magic number (%UBI_EC_HDR_MAGIC)
+ * @version: version of UBI implementation which is supposed to accept this
+ *           UBI image
+ * @padding1: reserved for future, zeroes
+ * @ec: the erase counter
+ * @vid_hdr_offset: where the VID header starts
+ * @data_offset: where the user data start
+ * @image_seq: image sequence number
+ * @padding2: reserved for future, zeroes
+ * @hdr_crc: erase counter header CRC checksum
+ *
+ * The erase counter header takes 64 bytes and has a plenty of unused space for
+ * future usage. The unused fields are zeroed. The @version field is used to
+ * indicate the version of UBI implementation which is supposed to be able to
+ * work with this UBI image. If @version is greater than the current UBI
+ * version, the image is rejected. This may be useful in future if something
+ * is changed radically. This field is duplicated in the volume identifier
+ * header.
+ *
+ * The @vid_hdr_offset and @data_offset fields contain the offset of the the
+ * volume identifier header and user data, relative to the beginning of the
+ * physical eraseblock. These values have to be the same for all physical
+ * eraseblocks.
+ *
+ * The @image_seq field is used to validate a UBI image that has been prepared
+ * for a UBI device. The @image_seq value can be any value, but it must be the
+ * same on all eraseblocks. UBI will ensure that all new erase counter headers
+ * also contain this value, and will check the value when scanning at start-up.
+ * One way to make use of @image_seq is to increase its value by one every time
+ * an image is flashed over an existing image, then, if the flashing does not
+ * complete, UBI will detect the error when scanning.
+ */
+struct ubi_ec_hdr {
+	__be32  magic;
+	__u8    version;
+	__u8    padding1[3];
+	__be64  ec; /* Warning: the current limit is 31-bit anyway! */
+	__be32  vid_hdr_offset;
+	__be32  data_offset;
+	__be32  image_seq;
+	__u8    padding2[32];
+	__be32  hdr_crc;
+} __attribute__ ((packed));
+
+/**
+ * struct ubi_vid_hdr - on-flash UBI volume identifier header.
+ * @magic: volume identifier header magic number (%UBI_VID_HDR_MAGIC)
+ * @version: UBI implementation version which is supposed to accept this UBI
+ *           image (%UBI_VERSION)
+ * @vol_type: volume type (%UBI_VID_DYNAMIC or %UBI_VID_STATIC)
+ * @copy_flag: if this logical eraseblock was copied from another physical
+ *             eraseblock (for wear-leveling reasons)
+ * @compat: compatibility of this volume (%0, %UBI_COMPAT_DELETE,
+ *          %UBI_COMPAT_IGNORE, %UBI_COMPAT_PRESERVE, or %UBI_COMPAT_REJECT)
+ * @vol_id: ID of this volume
+ * @lnum: logical eraseblock number
+ * @padding1: reserved for future, zeroes
+ * @data_size: how many bytes of data this logical eraseblock contains
+ * @used_ebs: total number of used logical eraseblocks in this volume
+ * @data_pad: how many bytes at the end of this physical eraseblock are not
+ *            used
+ * @data_crc: CRC checksum of the data stored in this logical eraseblock
+ * @padding2: reserved for future, zeroes
+ * @sqnum: sequence number
+ * @padding3: reserved for future, zeroes
+ * @hdr_crc: volume identifier header CRC checksum
+ *
+ * The @sqnum is the value of the global sequence counter at the time when this
+ * VID header was created. The global sequence counter is incremented each time
+ * UBI writes a new VID header to the flash, i.e. when it maps a logical
+ * eraseblock to a new physical eraseblock. The global sequence counter is an
+ * unsigned 64-bit integer and we assume it never overflows. The @sqnum
+ * (sequence number) is used to distinguish between older and newer versions of
+ * logical eraseblocks.
+ *
+ * There are 2 situations when there may be more than one physical eraseblock
+ * corresponding to the same logical eraseblock, i.e., having the same @vol_id
+ * and @lnum values in the volume identifier header. Suppose we have a logical
+ * eraseblock L and it is mapped to the physical eraseblock P.
+ *
+ * 1. Because UBI may erase physical eraseblocks asynchronously, the following
+ * situation is possible: L is asynchronously erased, so P is scheduled for
+ * erasure, then L is written to,i.e. mapped to another physical eraseblock P1,
+ * so P1 is written to, then an unclean reboot happens. Result - there are 2
+ * physical eraseblocks P and P1 corresponding to the same logical eraseblock
+ * L. But P1 has greater sequence number, so UBI picks P1 when it attaches the
+ * flash.
+ *
+ * 2. From time to time UBI moves logical eraseblocks to other physical
+ * eraseblocks for wear-leveling reasons. If, for example, UBI moves L from P
+ * to P1, and an unclean reboot happens before P is physically erased, there
+ * are two physical eraseblocks P and P1 corresponding to L and UBI has to
+ * select one of them when the flash is attached. The @sqnum field says which
+ * PEB is the original (obviously P will have lower @sqnum) and the copy. But
+ * it is not enough to select the physical eraseblock with the higher sequence
+ * number, because the unclean reboot could have happen in the middle of the
+ * copying process, so the data in P is corrupted. It is also not enough to
+ * just select the physical eraseblock with lower sequence number, because the
+ * data there may be old (consider a case if more data was added to P1 after
+ * the copying). Moreover, the unclean reboot may happen when the erasure of P
+ * was just started, so it result in unstable P, which is "mostly" OK, but
+ * still has unstable bits.
+ *
+ * UBI uses the @copy_flag field to indicate that this logical eraseblock is a
+ * copy. UBI also calculates data CRC when the data is moved and stores it at
+ * the @data_crc field of the copy (P1). So when UBI needs to pick one physical
+ * eraseblock of two (P or P1), the @copy_flag of the newer one (P1) is
+ * examined. If it is cleared, the situation* is simple and the newer one is
+ * picked. If it is set, the data CRC of the copy (P1) is examined. If the CRC
+ * checksum is correct, this physical eraseblock is selected (P1). Otherwise
+ * the older one (P) is selected.
+ *
+ * There are 2 sorts of volumes in UBI: user volumes and internal volumes.
+ * Internal volumes are not seen from outside and are used for various internal
+ * UBI purposes. In this implementation there is only one internal volume - the
+ * layout volume. Internal volumes are the main mechanism of UBI extensions.
+ * For example, in future one may introduce a journal internal volume. Internal
+ * volumes have their own reserved range of IDs.
+ *
+ * The @compat field is only used for internal volumes and contains the "degree
+ * of their compatibility". It is always zero for user volumes. This field
+ * provides a mechanism to introduce UBI extensions and to be still compatible
+ * with older UBI binaries. For example, if someone introduced a journal in
+ * future, he would probably use %UBI_COMPAT_DELETE compatibility for the
+ * journal volume.  And in this case, older UBI binaries, which know nothing
+ * about the journal volume, would just delete this volume and work perfectly
+ * fine. This is similar to what Ext2fs does when it is fed by an Ext3fs image
+ * - it just ignores the Ext3fs journal.
+ *
+ * The @data_crc field contains the CRC checksum of the contents of the logical
+ * eraseblock if this is a static volume. In case of dynamic volumes, it does
+ * not contain the CRC checksum as a rule. The only exception is when the
+ * data of the physical eraseblock was moved by the wear-leveling sub-system,
+ * then the wear-leveling sub-system calculates the data CRC and stores it in
+ * the @data_crc field. And of course, the @copy_flag is %in this case.
+ *
+ * The @data_size field is used only for static volumes because UBI has to know
+ * how many bytes of data are stored in this eraseblock. For dynamic volumes,
+ * this field usually contains zero. The only exception is when the data of the
+ * physical eraseblock was moved to another physical eraseblock for
+ * wear-leveling reasons. In this case, UBI calculates CRC checksum of the
+ * contents and uses both @data_crc and @data_size fields. In this case, the
+ * @data_size field contains data size.
+ *
+ * The @used_ebs field is used only for static volumes and indicates how many
+ * eraseblocks the data of the volume takes. For dynamic volumes this field is
+ * not used and always contains zero.
+ *
+ * The @data_pad is calculated when volumes are created using the alignment
+ * parameter. So, effectively, the @data_pad field reduces the size of logical
+ * eraseblocks of this volume. This is very handy when one uses block-oriented
+ * software (say, cramfs) on top of the UBI volume.
+ */
+struct ubi_vid_hdr {
+	__be32  magic;
+	__u8    version;
+	__u8    vol_type;
+	__u8    copy_flag;
+	__u8    compat;
+	__be32  vol_id;
+	__be32  lnum;
+	__u8    padding1[4];
+	__be32  data_size;
+	__be32  used_ebs;
+	__be32  data_pad;
+	__be32  data_crc;
+	__u8    padding2[4];
+	__be64  sqnum;
+	__u8    padding3[12];
+	__be32  hdr_crc;
+} __attribute__ ((packed));
+
+/* Internal UBI volumes count */
+#define UBI_INT_VOL_COUNT 1
+
+/*
+ * Starting ID of internal volumes. There is reserved room for 4096 internal
+ * volumes.
+ */
+#define UBI_INTERNAL_VOL_START (0x7FFFFFFF - 4096)
+
+/* The layout volume contains the volume table */
+
+#define UBI_LAYOUT_VOLUME_ID     UBI_INTERNAL_VOL_START
+#define UBI_LAYOUT_VOLUME_TYPE   UBI_VID_DYNAMIC
+#define UBI_LAYOUT_VOLUME_ALIGN  1
+#define UBI_LAYOUT_VOLUME_EBS    2
+#define UBI_LAYOUT_VOLUME_NAME   "layout volume"
+#define UBI_LAYOUT_VOLUME_COMPAT UBI_COMPAT_REJECT
+
+/* The maximum number of volumes per one UBI device */
+#define UBI_MAX_VOLUMES 128
+
+/* The maximum volume name length */
+#define UBI_VOL_NAME_MAX 127
+
+/* Size of the volume table record */
+#define UBI_VTBL_RECORD_SIZE sizeof(struct ubi_vtbl_record)
+
+/* Size of the volume table record without the ending CRC */
+#define UBI_VTBL_RECORD_SIZE_CRC (UBI_VTBL_RECORD_SIZE - sizeof(__be32))
+
+/**
+ * struct ubi_vtbl_record - a record in the volume table.
+ * @reserved_pebs: how many physical eraseblocks are reserved for this volume
+ * @alignment: volume alignment
+ * @data_pad: how many bytes are unused at the end of the each physical
+ * eraseblock to satisfy the requested alignment
+ * @vol_type: volume type (%UBI_DYNAMIC_VOLUME or %UBI_STATIC_VOLUME)
+ * @upd_marker: if volume update was started but not finished
+ * @name_len: volume name length
+ * @name: the volume name
+ * @flags: volume flags (%UBI_VTBL_AUTORESIZE_FLG)
+ * @padding: reserved, zeroes
+ * @crc: a CRC32 checksum of the record
+ *
+ * The volume table records are stored in the volume table, which is stored in
+ * the layout volume. The layout volume consists of 2 logical eraseblock, each
+ * of which contains a copy of the volume table (i.e., the volume table is
+ * duplicated). The volume table is an array of &struct ubi_vtbl_record
+ * objects indexed by the volume ID.
+ *
+ * If the size of the logical eraseblock is large enough to fit
+ * %UBI_MAX_VOLUMES records, the volume table contains %UBI_MAX_VOLUMES
+ * records. Otherwise, it contains as many records as it can fit (i.e., size of
+ * logical eraseblock divided by sizeof(struct ubi_vtbl_record)).
+ *
+ * The @upd_marker flag is used to implement volume update. It is set to %1
+ * before update and set to %0 after the update. So if the update operation was
+ * interrupted, UBI knows that the volume is corrupted.
+ *
+ * The @alignment field is specified when the volume is created and cannot be
+ * later changed. It may be useful, for example, when a block-oriented file
+ * system works on top of UBI. The @data_pad field is calculated using the
+ * logical eraseblock size and @alignment. The alignment must be multiple to the
+ * minimal flash I/O unit. If @alignment is 1, all the available space of
+ * the physical eraseblocks is used.
+ *
+ * Empty records contain all zeroes and the CRC checksum of those zeroes.
+ */
+struct ubi_vtbl_record {
+	__be32  reserved_pebs;
+	__be32  alignment;
+	__be32  data_pad;
+	__u8    vol_type;
+	__u8    upd_marker;
+	__be16  name_len;
+	__u8    name[UBI_VOL_NAME_MAX+1];
+	__u8    flags;
+	__u8    padding[23];
+	__be32  crc;
+} __attribute__ ((packed));
+
+#endif /* !__UBI_MEDIA_H__ */

kernel/drivers/mtd/ubi/ubi.h

@@ -0,0 +1,688 @@
+/*
+ * 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)
+ */
+
+#ifndef __UBI_UBI_H__
+#define __UBI_UBI_H__
+
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/list.h>
+#include <linux/rbtree.h>
+#include <linux/sched.h>
+#include <linux/wait.h>
+#include <linux/mutex.h>
+#include <linux/rwsem.h>
+#include <linux/spinlock.h>
+#include <linux/fs.h>
+#include <linux/cdev.h>
+#include <linux/device.h>
+#include <linux/string.h>
+#include <linux/vmalloc.h>
+#include <linux/notifier.h>
+#include <linux/mtd/mtd.h>
+#include <linux/mtd/ubi.h>
+
+#include "ubi-media.h"
+#include "scan.h"
+#include "debug.h"
+
+/* Maximum number of supported UBI devices */
+#define UBI_MAX_DEVICES 32
+
+/* UBI name used for character devices, sysfs, etc */
+#define UBI_NAME_STR "ubi"
+
+/* Normal UBI messages */
+#define ubi_msg(fmt, ...) printk(KERN_NOTICE "UBI: " fmt "\n", ##__VA_ARGS__)
+/* UBI warning messages */
+#define ubi_warn(fmt, ...) printk(KERN_WARNING "UBI warning: %s: " fmt "\n", \
+				  __func__, ##__VA_ARGS__)
+/* UBI error messages */
+#define ubi_err(fmt, ...) printk(KERN_ERR "UBI error: %s: " fmt "\n", \
+				 __func__, ##__VA_ARGS__)
+
+/* Lowest number PEBs reserved for bad PEB handling */
+#define MIN_RESEVED_PEBS 2
+
+/* Background thread name pattern */
+#define UBI_BGT_NAME_PATTERN "ubi_bgt%dd"
+
+/* This marker in the EBA table means that the LEB is um-mapped */
+#define UBI_LEB_UNMAPPED -1
+
+/*
+ * In case of errors, UBI tries to repeat the operation several times before
+ * returning error. The below constant defines how many times UBI re-tries.
+ */
+#define UBI_IO_RETRIES 3
+
+/*
+ * Length of the protection queue. The length is effectively equivalent to the
+ * number of (global) erase cycles PEBs are protected from the wear-leveling
+ * worker.
+ */
+#define UBI_PROT_QUEUE_LEN 10
+
+/*
+ * Error codes returned by the I/O sub-system.
+ *
+ * UBI_IO_PEB_EMPTY: the physical eraseblock is empty, i.e. it contains only
+ *                   %0xFF bytes
+ * UBI_IO_PEB_FREE: the physical eraseblock is free, i.e. it contains only a
+ *                  valid erase counter header, and the rest are %0xFF bytes
+ * UBI_IO_BAD_EC_HDR: the erase counter header is corrupted (bad magic or CRC)
+ * UBI_IO_BAD_VID_HDR: the volume identifier header is corrupted (bad magic or
+ *                     CRC)
+ * UBI_IO_BITFLIPS: bit-flips were detected and corrected
+ */
+enum {
+	UBI_IO_PEB_EMPTY = 1,
+	UBI_IO_PEB_FREE,
+	UBI_IO_BAD_EC_HDR,
+	UBI_IO_BAD_VID_HDR,
+	UBI_IO_BITFLIPS
+};
+
+/*
+ * Return codes of the 'ubi_eba_copy_leb()' function.
+ *
+ * MOVE_CANCEL_RACE: canceled because the volume is being deleted, the source
+ *                   PEB was put meanwhile, or there is I/O on the source PEB
+ * MOVE_SOURCE_RD_ERR: canceled because there was a read error from the source
+ *                     PEB
+ * MOVE_TARGET_RD_ERR: canceled because there was a read error from the target
+ *                     PEB
+ * MOVE_TARGET_WR_ERR: canceled because there was a write error to the target
+ *                     PEB
+ * MOVE_CANCEL_BITFLIPS: canceled because a bit-flip was detected in the
+ *                       target PEB
+ */
+enum {
+	MOVE_CANCEL_RACE = 1,
+	MOVE_SOURCE_RD_ERR,
+	MOVE_TARGET_RD_ERR,
+	MOVE_TARGET_WR_ERR,
+	MOVE_CANCEL_BITFLIPS,
+};
+
+/**
+ * struct ubi_wl_entry - wear-leveling entry.
+ * @u.rb: link in the corresponding (free/used) RB-tree
+ * @u.list: link in the protection queue
+ * @ec: erase counter
+ * @pnum: physical eraseblock number
+ *
+ * This data structure is used in the WL sub-system. Each physical eraseblock
+ * has a corresponding &struct wl_entry object which may be kept in different
+ * RB-trees. See WL sub-system for details.
+ */
+struct ubi_wl_entry {
+	union {
+		struct rb_node rb;
+		struct list_head list;
+	} u;
+	int ec;
+	int pnum;
+};
+
+/**
+ * struct ubi_ltree_entry - an entry in the lock tree.
+ * @rb: links RB-tree nodes
+ * @vol_id: volume ID of the locked logical eraseblock
+ * @lnum: locked logical eraseblock number
+ * @users: how many tasks are using this logical eraseblock or wait for it
+ * @mutex: read/write mutex to implement read/write access serialization to
+ *         the (@vol_id, @lnum) logical eraseblock
+ *
+ * This data structure is used in the EBA sub-system to implement per-LEB
+ * locking. When a logical eraseblock is being locked - corresponding
+ * &struct ubi_ltree_entry object is inserted to the lock tree (@ubi->ltree).
+ * See EBA sub-system for details.
+ */
+struct ubi_ltree_entry {
+	struct rb_node rb;
+	int vol_id;
+	int lnum;
+	int users;
+	struct rw_semaphore mutex;
+};
+
+/**
+ * struct ubi_rename_entry - volume re-name description data structure.
+ * @new_name_len: new volume name length
+ * @new_name: new volume name
+ * @remove: if not zero, this volume should be removed, not re-named
+ * @desc: descriptor of the volume
+ * @list: links re-name entries into a list
+ *
+ * This data structure is utilized in the multiple volume re-name code. Namely,
+ * UBI first creates a list of &struct ubi_rename_entry objects from the
+ * &struct ubi_rnvol_req request object, and then utilizes this list to do all
+ * the job.
+ */
+struct ubi_rename_entry {
+	int new_name_len;
+	char new_name[UBI_VOL_NAME_MAX + 1];
+	int remove;
+	struct ubi_volume_desc *desc;
+	struct list_head list;
+};
+
+struct ubi_volume_desc;
+
+/**
+ * struct ubi_volume - UBI volume description data structure.
+ * @dev: device object to make use of the the Linux device model
+ * @cdev: character device object to create character device
+ * @ubi: reference to the UBI device description object
+ * @vol_id: volume ID
+ * @ref_count: volume reference count
+ * @readers: number of users holding this volume in read-only mode
+ * @writers: number of users holding this volume in read-write mode
+ * @exclusive: whether somebody holds this volume in exclusive mode
+ *
+ * @reserved_pebs: how many physical eraseblocks are reserved for this volume
+ * @vol_type: volume type (%UBI_DYNAMIC_VOLUME or %UBI_STATIC_VOLUME)
+ * @usable_leb_size: logical eraseblock size without padding
+ * @used_ebs: how many logical eraseblocks in this volume contain data
+ * @last_eb_bytes: how many bytes are stored in the last logical eraseblock
+ * @used_bytes: how many bytes of data this volume contains
+ * @alignment: volume alignment
+ * @data_pad: how many bytes are not used at the end of physical eraseblocks to
+ *            satisfy the requested alignment
+ * @name_len: volume name length
+ * @name: volume name
+ *
+ * @upd_ebs: how many eraseblocks are expected to be updated
+ * @ch_lnum: LEB number which is being changing by the atomic LEB change
+ *           operation
+ * @ch_dtype: data persistency type which is being changing by the atomic LEB
+ *            change operation
+ * @upd_bytes: how many bytes are expected to be received for volume update or
+ *             atomic LEB change
+ * @upd_received: how many bytes were already received for volume update or
+ *                atomic LEB change
+ * @upd_buf: update buffer which is used to collect update data or data for
+ *           atomic LEB change
+ *
+ * @eba_tbl: EBA table of this volume (LEB->PEB mapping)
+ * @checked: %1 if this static volume was checked
+ * @corrupted: %1 if the volume is corrupted (static volumes only)
+ * @upd_marker: %1 if the update marker is set for this volume
+ * @updating: %1 if the volume is being updated
+ * @changing_leb: %1 if the atomic LEB change ioctl command is in progress
+ * @direct_writes: %1 if direct writes are enabled for this volume
+ *
+ * The @corrupted field indicates that the volume's contents is corrupted.
+ * Since UBI protects only static volumes, this field is not relevant to
+ * dynamic volumes - it is user's responsibility to assure their data
+ * integrity.
+ *
+ * The @upd_marker flag indicates that this volume is either being updated at
+ * the moment or is damaged because of an unclean reboot.
+ */
+struct ubi_volume {
+	struct device dev;
+	struct cdev cdev;
+	struct ubi_device *ubi;
+	int vol_id;
+	int ref_count;
+	int readers;
+	int writers;
+	int exclusive;
+
+	int reserved_pebs;
+	int vol_type;
+	int usable_leb_size;
+	int used_ebs;
+	int last_eb_bytes;
+	long long used_bytes;
+	int alignment;
+	int data_pad;
+	int name_len;
+	char name[UBI_VOL_NAME_MAX + 1];
+
+	int upd_ebs;
+	int ch_lnum;
+	int ch_dtype;
+	long long upd_bytes;
+	long long upd_received;
+	void *upd_buf;
+
+	int *eba_tbl;
+	unsigned int checked:1;
+	unsigned int corrupted:1;
+	unsigned int upd_marker:1;
+	unsigned int updating:1;
+	unsigned int changing_leb:1;
+	unsigned int direct_writes:1;
+};
+
+/**
+ * struct ubi_volume_desc - UBI volume descriptor returned when it is opened.
+ * @vol: reference to the corresponding volume description object
+ * @mode: open mode (%UBI_READONLY, %UBI_READWRITE, or %UBI_EXCLUSIVE)
+ */
+struct ubi_volume_desc {
+	struct ubi_volume *vol;
+	int mode;
+};
+
+struct ubi_wl_entry;
+
+/**
+ * struct ubi_device - UBI device description structure
+ * @dev: UBI device object to use the the Linux device model
+ * @cdev: character device object to create character device
+ * @ubi_num: UBI device number
+ * @ubi_name: UBI device name
+ * @vol_count: number of volumes in this UBI device
+ * @volumes: volumes of this UBI device
+ * @volumes_lock: protects @volumes, @rsvd_pebs, @avail_pebs, beb_rsvd_pebs,
+ *                @beb_rsvd_level, @bad_peb_count, @good_peb_count, @vol_count,
+ *                @vol->readers, @vol->writers, @vol->exclusive,
+ *                @vol->ref_count, @vol->mapping and @vol->eba_tbl.
+ * @ref_count: count of references on the UBI device
+ * @image_seq: image sequence number recorded on EC headers
+ *
+ * @rsvd_pebs: count of reserved physical eraseblocks
+ * @avail_pebs: count of available physical eraseblocks
+ * @beb_rsvd_pebs: how many physical eraseblocks are reserved for bad PEB
+ *                 handling
+ * @beb_rsvd_level: normal level of PEBs reserved for bad PEB handling
+ *
+ * @autoresize_vol_id: ID of the volume which has to be auto-resized at the end
+ *                     of UBI initialization
+ * @vtbl_slots: how many slots are available in the volume table
+ * @vtbl_size: size of the volume table in bytes
+ * @vtbl: in-RAM volume table copy
+ * @device_mutex: protects on-flash volume table and serializes volume
+ *                creation, deletion, update, re-size, re-name and set
+ *                property
+ *
+ * @max_ec: current highest erase counter value
+ * @mean_ec: current mean erase counter value
+ *
+ * @global_sqnum: global sequence number
+ * @ltree_lock: protects the lock tree and @global_sqnum
+ * @ltree: the lock tree
+ * @alc_mutex: serializes "atomic LEB change" operations
+ *
+ * @used: RB-tree of used physical eraseblocks
+ * @erroneous: RB-tree of erroneous used physical eraseblocks
+ * @free: RB-tree of free physical eraseblocks
+ * @scrub: RB-tree of physical eraseblocks which need scrubbing
+ * @pq: protection queue (contain physical eraseblocks which are temporarily
+ *      protected from the wear-leveling worker)
+ * @pq_head: protection queue head
+ * @wl_lock: protects the @used, @free, @pq, @pq_head, @lookuptbl, @move_from,
+ * 	     @move_to, @move_to_put @erase_pending, @wl_scheduled, @works,
+ * 	     @erroneous, and @erroneous_peb_count fields
+ * @move_mutex: serializes eraseblock moves
+ * @work_sem: synchronizes the WL worker with use tasks
+ * @wl_scheduled: non-zero if the wear-leveling was scheduled
+ * @lookuptbl: a table to quickly find a &struct ubi_wl_entry object for any
+ *             physical eraseblock
+ * @move_from: physical eraseblock from where the data is being moved
+ * @move_to: physical eraseblock where the data is being moved to
+ * @move_to_put: if the "to" PEB was put
+ * @works: list of pending works
+ * @works_count: count of pending works
+ * @bgt_thread: background thread description object
+ * @thread_enabled: if the background thread is enabled
+ * @bgt_name: background thread name
+ * @reboot_notifier: notifier to terminate background thread before rebooting
+ *
+ * @flash_size: underlying MTD device size (in bytes)
+ * @peb_count: count of physical eraseblocks on the MTD device
+ * @peb_size: physical eraseblock size
+ * @bad_peb_count: count of bad physical eraseblocks
+ * @good_peb_count: count of good physical eraseblocks
+ * @erroneous_peb_count: count of erroneous physical eraseblocks in @erroneous
+ * @max_erroneous: maximum allowed amount of erroneous physical eraseblocks
+ * @min_io_size: minimal input/output unit size of the underlying MTD device
+ * @hdrs_min_io_size: minimal I/O unit size used for VID and EC headers
+ * @ro_mode: if the UBI device is in read-only mode
+ * @leb_size: logical eraseblock size
+ * @leb_start: starting offset of logical eraseblocks within physical
+ *             eraseblocks
+ * @ec_hdr_alsize: size of the EC header aligned to @hdrs_min_io_size
+ * @vid_hdr_alsize: size of the VID header aligned to @hdrs_min_io_size
+ * @vid_hdr_offset: starting offset of the volume identifier header (might be
+ *                  unaligned)
+ * @vid_hdr_aloffset: starting offset of the VID header aligned to
+ * @hdrs_min_io_size
+ * @vid_hdr_shift: contains @vid_hdr_offset - @vid_hdr_aloffset
+ * @bad_allowed: whether the MTD device admits of bad physical eraseblocks or
+ *               not
+ * @nor_flash: non-zero if working on top of NOR flash
+ * @mtd: MTD device descriptor
+ *
+ * @peb_buf1: a buffer of PEB size used for different purposes
+ * @peb_buf2: another buffer of PEB size used for different purposes
+ * @buf_mutex: protects @peb_buf1 and @peb_buf2
+ * @ckvol_mutex: serializes static volume checking when opening
+ * @dbg_peb_buf: buffer of PEB size used for debugging
+ * @dbg_buf_mutex: protects @dbg_peb_buf
+ */
+struct ubi_device {
+	struct cdev cdev;
+	struct device dev;
+	int ubi_num;
+	char ubi_name[sizeof(UBI_NAME_STR)+5];
+	int vol_count;
+	struct ubi_volume *volumes[UBI_MAX_VOLUMES+UBI_INT_VOL_COUNT];
+	spinlock_t volumes_lock;
+	int ref_count;
+	int image_seq;
+
+	int rsvd_pebs;
+	int avail_pebs;
+	int beb_rsvd_pebs;
+	int beb_rsvd_level;
+
+	int autoresize_vol_id;
+	int vtbl_slots;
+	int vtbl_size;
+	struct ubi_vtbl_record *vtbl;
+	struct mutex device_mutex;
+
+	int max_ec;
+	/* Note, mean_ec is not updated run-time - should be fixed */
+	int mean_ec;
+
+	/* EBA sub-system's stuff */
+	unsigned long long global_sqnum;
+	spinlock_t ltree_lock;
+	struct rb_root ltree;
+	struct mutex alc_mutex;
+
+	/* Wear-leveling sub-system's stuff */
+	struct rb_root used;
+	struct rb_root erroneous;
+	struct rb_root free;
+	struct rb_root scrub;
+	struct list_head pq[UBI_PROT_QUEUE_LEN];
+	int pq_head;
+	spinlock_t wl_lock;
+	struct mutex move_mutex;
+	struct rw_semaphore work_sem;
+	int wl_scheduled;
+	struct ubi_wl_entry **lookuptbl;
+	struct ubi_wl_entry *move_from;
+	struct ubi_wl_entry *move_to;
+	int move_to_put;
+	struct list_head works;
+	int works_count;
+	struct task_struct *bgt_thread;
+	int thread_enabled;
+	char bgt_name[sizeof(UBI_BGT_NAME_PATTERN)+2];
+	struct notifier_block reboot_notifier;
+
+	/* I/O sub-system's stuff */
+	long long flash_size;
+	int peb_count;
+	int peb_size;
+	int bad_peb_count;
+	int good_peb_count;
+	int erroneous_peb_count;
+	int max_erroneous;
+	int min_io_size;
+	int hdrs_min_io_size;
+	int ro_mode;
+	int leb_size;
+	int leb_start;
+	int ec_hdr_alsize;
+	int vid_hdr_alsize;
+	int vid_hdr_offset;
+	int vid_hdr_aloffset;
+	int vid_hdr_shift;
+	unsigned int bad_allowed:1;
+	unsigned int nor_flash:1;
+	struct mtd_info *mtd;
+
+	void *peb_buf1;
+	void *peb_buf2;
+	struct mutex buf_mutex;
+	struct mutex ckvol_mutex;
+#ifdef CONFIG_MTD_UBI_DEBUG_PARANOID
+	void *dbg_peb_buf;
+	struct mutex dbg_buf_mutex;
+#endif
+};
+
+extern struct kmem_cache *ubi_wl_entry_slab;
+extern const struct file_operations ubi_ctrl_cdev_operations;
+extern const struct file_operations ubi_cdev_operations;
+extern const struct file_operations ubi_vol_cdev_operations;
+extern struct class *ubi_class;
+extern struct mutex ubi_devices_mutex;
+extern struct blocking_notifier_head ubi_notifiers;
+
+/* vtbl.c */
+int ubi_change_vtbl_record(struct ubi_device *ubi, int idx,
+			   struct ubi_vtbl_record *vtbl_rec);
+int ubi_vtbl_rename_volumes(struct ubi_device *ubi,
+			    struct list_head *rename_list);
+int ubi_read_volume_table(struct ubi_device *ubi, struct ubi_scan_info *si);
+
+/* vmt.c */
+int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req);
+int ubi_remove_volume(struct ubi_volume_desc *desc, int no_vtbl);
+int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs);
+int ubi_rename_volumes(struct ubi_device *ubi, struct list_head *rename_list);
+int ubi_add_volume(struct ubi_device *ubi, struct ubi_volume *vol);
+void ubi_free_volume(struct ubi_device *ubi, struct ubi_volume *vol);
+
+/* upd.c */
+int ubi_start_update(struct ubi_device *ubi, struct ubi_volume *vol,
+		     long long bytes);
+int ubi_more_update_data(struct ubi_device *ubi, struct ubi_volume *vol,
+			 const void __user *buf, int count);
+int ubi_start_leb_change(struct ubi_device *ubi, struct ubi_volume *vol,
+			 const struct ubi_leb_change_req *req);
+int ubi_more_leb_change_data(struct ubi_device *ubi, struct ubi_volume *vol,
+			     const void __user *buf, int count);
+
+/* misc.c */
+int ubi_calc_data_len(const struct ubi_device *ubi, const void *buf,
+		      int length);
+int ubi_check_volume(struct ubi_device *ubi, int vol_id);
+void ubi_calculate_reserved(struct ubi_device *ubi);
+
+/* eba.c */
+int ubi_eba_unmap_leb(struct ubi_device *ubi, struct ubi_volume *vol,
+		      int lnum);
+int ubi_eba_read_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
+		     void *buf, int offset, int len, int check);
+int ubi_eba_write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
+		      const void *buf, int offset, int len, int dtype);
+int ubi_eba_write_leb_st(struct ubi_device *ubi, struct ubi_volume *vol,
+			 int lnum, const void *buf, int len, int dtype,
+			 int used_ebs);
+int ubi_eba_atomic_leb_change(struct ubi_device *ubi, struct ubi_volume *vol,
+			      int lnum, const void *buf, int len, int dtype);
+int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to,
+		     struct ubi_vid_hdr *vid_hdr);
+int ubi_eba_init_scan(struct ubi_device *ubi, struct ubi_scan_info *si);
+
+/* wl.c */
+int ubi_wl_get_peb(struct ubi_device *ubi, int dtype);
+int ubi_wl_put_peb(struct ubi_device *ubi, int pnum, int torture);
+int ubi_wl_flush(struct ubi_device *ubi);
+int ubi_wl_scrub_peb(struct ubi_device *ubi, int pnum);
+int ubi_wl_init_scan(struct ubi_device *ubi, struct ubi_scan_info *si);
+void ubi_wl_close(struct ubi_device *ubi);
+int ubi_thread(void *u);
+
+/* io.c */
+int ubi_io_read(const struct ubi_device *ubi, void *buf, int pnum, int offset,
+		int len);
+int ubi_io_write(struct ubi_device *ubi, const void *buf, int pnum, int offset,
+		 int len);
+int ubi_io_sync_erase(struct ubi_device *ubi, int pnum, int torture);
+int ubi_io_is_bad(const struct ubi_device *ubi, int pnum);
+int ubi_io_mark_bad(const struct ubi_device *ubi, int pnum);
+int ubi_io_read_ec_hdr(struct ubi_device *ubi, int pnum,
+		       struct ubi_ec_hdr *ec_hdr, int verbose);
+int ubi_io_write_ec_hdr(struct ubi_device *ubi, int pnum,
+			struct ubi_ec_hdr *ec_hdr);
+int ubi_io_read_vid_hdr(struct ubi_device *ubi, int pnum,
+			struct ubi_vid_hdr *vid_hdr, int verbose);
+int ubi_io_write_vid_hdr(struct ubi_device *ubi, int pnum,
+			 struct ubi_vid_hdr *vid_hdr);
+
+/* build.c */
+int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset);
+int ubi_detach_mtd_dev(int ubi_num, int anyway);
+struct ubi_device *ubi_get_device(int ubi_num);
+void ubi_put_device(struct ubi_device *ubi);
+struct ubi_device *ubi_get_by_major(int major);
+int ubi_major2num(int major);
+int ubi_volume_notify(struct ubi_device *ubi, struct ubi_volume *vol,
+		      int ntype);
+int ubi_notify_all(struct ubi_device *ubi, int ntype,
+		   struct notifier_block *nb);
+int ubi_enumerate_volumes(struct notifier_block *nb);
+
+/* kapi.c */
+void ubi_do_get_device_info(struct ubi_device *ubi, struct ubi_device_info *di);
+void ubi_do_get_volume_info(struct ubi_device *ubi, struct ubi_volume *vol,
+			    struct ubi_volume_info *vi);
+
+/*
+ * ubi_rb_for_each_entry - walk an RB-tree.
+ * @rb: a pointer to type 'struct rb_node' to use as a loop counter
+ * @pos: a pointer to RB-tree entry type to use as a loop counter
+ * @root: RB-tree's root
+ * @member: the name of the 'struct rb_node' within the RB-tree entry
+ */
+#define ubi_rb_for_each_entry(rb, pos, root, member)                         \
+	for (rb = rb_first(root),                                            \
+	     pos = (rb ? container_of(rb, typeof(*pos), member) : NULL);     \
+	     rb;                                                             \
+	     rb = rb_next(rb),                                               \
+	     pos = (rb ? container_of(rb, typeof(*pos), member) : NULL))
+
+/**
+ * ubi_zalloc_vid_hdr - allocate a volume identifier header object.
+ * @ubi: UBI device description object
+ * @gfp_flags: GFP flags to allocate with
+ *
+ * This function returns a pointer to the newly allocated and zero-filled
+ * volume identifier header object in case of success and %NULL in case of
+ * failure.
+ */
+static inline struct ubi_vid_hdr *
+ubi_zalloc_vid_hdr(const struct ubi_device *ubi, gfp_t gfp_flags)
+{
+	void *vid_hdr;
+
+	vid_hdr = kzalloc(ubi->vid_hdr_alsize, gfp_flags);
+	if (!vid_hdr)
+		return NULL;
+
+	/*
+	 * VID headers may be stored at un-aligned flash offsets, so we shift
+	 * the pointer.
+	 */
+	return vid_hdr + ubi->vid_hdr_shift;
+}
+
+/**
+ * ubi_free_vid_hdr - free a volume identifier header object.
+ * @ubi: UBI device description object
+ * @vid_hdr: the object to free
+ */
+static inline void ubi_free_vid_hdr(const struct ubi_device *ubi,
+				    struct ubi_vid_hdr *vid_hdr)
+{
+	void *p = vid_hdr;
+
+	if (!p)
+		return;
+
+	kfree(p - ubi->vid_hdr_shift);
+}
+
+/*
+ * This function is equivalent to 'ubi_io_read()', but @offset is relative to
+ * the beginning of the logical eraseblock, not to the beginning of the
+ * physical eraseblock.
+ */
+static inline int ubi_io_read_data(const struct ubi_device *ubi, void *buf,
+				   int pnum, int offset, int len)
+{
+	ubi_assert(offset >= 0);
+	return ubi_io_read(ubi, buf, pnum, offset + ubi->leb_start, len);
+}
+
+/*
+ * This function is equivalent to 'ubi_io_write()', but @offset is relative to
+ * the beginning of the logical eraseblock, not to the beginning of the
+ * physical eraseblock.
+ */
+static inline int ubi_io_write_data(struct ubi_device *ubi, const void *buf,
+				    int pnum, int offset, int len)
+{
+	ubi_assert(offset >= 0);
+	return ubi_io_write(ubi, buf, pnum, offset + ubi->leb_start, len);
+}
+
+/**
+ * ubi_ro_mode - switch to read-only mode.
+ * @ubi: UBI device description object
+ */
+static inline void ubi_ro_mode(struct ubi_device *ubi)
+{
+	if (!ubi->ro_mode) {
+		ubi->ro_mode = 1;
+		ubi_warn("switch to read-only mode");
+	}
+}
+
+/**
+ * vol_id2idx - get table index by volume ID.
+ * @ubi: UBI device description object
+ * @vol_id: volume ID
+ */
+static inline int vol_id2idx(const struct ubi_device *ubi, int vol_id)
+{
+	if (vol_id >= UBI_INTERNAL_VOL_START)
+		return vol_id - UBI_INTERNAL_VOL_START + ubi->vtbl_slots;
+	else
+		return vol_id;
+}
+
+/**
+ * idx2vol_id - get volume ID by table index.
+ * @ubi: UBI device description object
+ * @idx: table index
+ */
+static inline int idx2vol_id(const struct ubi_device *ubi, int idx)
+{
+	if (idx >= ubi->vtbl_slots)
+		return idx - ubi->vtbl_slots + UBI_INTERNAL_VOL_START;
+	else
+		return idx;
+}
+
+#endif /* !__UBI_UBI_H__ */

kernel/drivers/mtd/ubi/upd.c

@@ -0,0 +1,437 @@
+/*
+ * 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.
+ */
+
+#include <linux/err.h>
+#include <linux/uaccess.h>
+#include <linux/math64.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_gen("set update marker for volume %d", vol->vol_id);
+
+	if (vol->upd_marker) {
+		ubi_assert(ubi->vtbl[vol->vol_id].upd_marker);
+		dbg_gen("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->device_mutex);
+	err = ubi_change_vtbl_record(ubi, vol->vol_id, &vtbl_rec);
+	vol->upd_marker = 1;
+	mutex_unlock(&ubi->device_mutex);
+	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;
+	struct ubi_vtbl_record vtbl_rec;
+
+	dbg_gen("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 = bytes;
+		vol->used_ebs = div_u64_rem(bytes, vol->usable_leb_size,
+					    &vol->last_eb_bytes);
+		if (vol->last_eb_bytes)
+			vol->used_ebs += 1;
+		else
+			vol->last_eb_bytes = vol->usable_leb_size;
+	}
+
+	mutex_lock(&ubi->device_mutex);
+	err = ubi_change_vtbl_record(ubi, vol->vol_id, &vtbl_rec);
+	vol->upd_marker = 0;
+	mutex_unlock(&ubi->device_mutex);
+	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;
+
+	dbg_gen("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 = ubi_wl_flush(ubi);
+		if (err)
+			return err;
+
+		err = clear_update_marker(ubi, vol, 0);
+		if (err)
+			return err;
+		vol->updating = 0;
+		return 0;
+	}
+
+	vol->upd_buf = vmalloc(ubi->leb_size);
+	if (!vol->upd_buf)
+		return -ENOMEM;
+
+	vol->upd_ebs = div_u64(bytes + vol->usable_leb_size - 1,
+			       vol->usable_leb_size);
+	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_gen("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_gen("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.
+ * @ubi: UBI device description object
+ * @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)
+{
+	int lnum, offs, err = 0, len, to_write = count;
+
+	dbg_gen("write %d of %lld bytes, %lld already passed",
+		count, vol->upd_bytes, vol->upd_received);
+
+	if (ubi->ro_mode)
+		return -EROFS;
+
+	lnum = div_u64_rem(vol->upd_received,  vol->usable_leb_size, &offs);
+	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) {
+		err = ubi_wl_flush(ubi);
+		if (err)
+			return err;
+		/* The update is finished, clear the update marker */
+		err = clear_update_marker(ubi, vol, vol->upd_bytes);
+		if (err)
+			return err;
+		vol->updating = 0;
+		err = to_write;
+		vfree(vol->upd_buf);
+	}
+
+	return err;
+}
+
+/**
+ * ubi_more_leb_change_data - accept more data for atomic LEB change.
+ * @ubi: UBI device description object
+ * @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_gen("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;
+}

kernel/drivers/mtd/ubi/vmt.c

@@ -0,0 +1,880 @@
+/*
+ * 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.
+ */
+
+#include <linux/err.h>
+#include <linux/math64.h>
+#include "ubi.h"
+
+#ifdef CONFIG_MTD_UBI_DEBUG_PARANOID
+static int paranoid_check_volumes(struct ubi_device *ubi);
+#else
+#define paranoid_check_volumes(ubi) 0
+#endif
+
+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;
+}
+
+/* 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->eba_tbl);
+	kfree(vol);
+}
+
+/**
+ * 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);
+}
+
+/**
+ * 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->device_mutex locked.
+ */
+int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req)
+{
+	int i, err, vol_id = req->vol_id, do_free = 1;
+	struct ubi_volume *vol;
+	struct ubi_vtbl_record vtbl_rec;
+	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_gen("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_gen("create device %d, volume %d, %llu bytes, type %d, name %s",
+		ubi->ubi_num, 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;
+	vol->reserved_pebs += div_u64(req->bytes + vol->usable_leb_size - 1,
+				      vol->usable_leb_size);
+
+	/* 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);
+	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 {
+		vol->used_ebs = div_u64_rem(vol->used_bytes,
+					    vol->usable_leb_size,
+					    &vol->last_eb_bytes);
+		if (vol->last_eb_bytes != 0)
+			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;
+	}
+
+	vol->dev.release = vol_release;
+	vol->dev.parent = &ubi->dev;
+	vol->dev.devt = dev;
+	vol->dev.class = ubi_class;
+
+	dev_set_name(&vol->dev, "%s_%d", ubi->ubi_name, vol->vol_id);
+	err = device_register(&vol->dev);
+	if (err) {
+		ubi_err("cannot register device");
+		goto out_cdev;
+	}
+
+	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);
+
+	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);
+
+	ubi_volume_notify(ubi, vol, UBI_VOLUME_ADDED);
+	if (paranoid_check_volumes(ubi))
+		dbg_err("check failed while creating volume %d", vol_id);
+	return err;
+
+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.
+	 */
+	do_free = 0;
+	get_device(&vol->dev);
+	volume_sysfs_close(vol);
+out_cdev:
+	cdev_del(&vol->cdev);
+out_mapping:
+	if (do_free)
+		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 (do_free)
+		kfree(vol);
+	else
+		put_device(&vol->dev);
+	ubi_err("cannot create volume %d, error %d", vol_id, err);
+	return err;
+}
+
+/**
+ * ubi_remove_volume - remove volume.
+ * @desc: volume descriptor
+ * @no_vtbl: do not change volume table if not zero
+ *
+ * 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->device_mutex
+ * locked.
+ */
+int ubi_remove_volume(struct ubi_volume_desc *desc, int no_vtbl)
+{
+	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_gen("remove device %d, volume %d", ubi->ubi_num, 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);
+
+	if (!no_vtbl) {
+		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;
+	}
+
+	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);
+
+	ubi_volume_notify(ubi, vol, UBI_VOLUME_REMOVED);
+	if (!no_vtbl && paranoid_check_volumes(ubi))
+		dbg_err("check failed while removing volume %d", vol_id);
+
+	return err;
+
+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->device_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_gen("re-size device %d, volume %d to from %d to %d PEBs",
+		ubi->ubi_num, 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;
+	}
+
+	ubi_volume_notify(ubi, vol, UBI_VOLUME_RESIZED);
+	if (paranoid_check_volumes(ubi))
+		dbg_err("check failed while re-sizing volume %d", vol_id);
+	return err;
+
+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_rename_volumes - re-name UBI volumes.
+ * @ubi: UBI device description object
+ * @rename_list: list of &struct ubi_rename_entry objects
+ *
+ * This function re-names or removes volumes specified in the re-name list.
+ * Returns zero in case of success and a negative error code in case of
+ * failure.
+ */
+int ubi_rename_volumes(struct ubi_device *ubi, struct list_head *rename_list)
+{
+	int err;
+	struct ubi_rename_entry *re;
+
+	err = ubi_vtbl_rename_volumes(ubi, rename_list);
+	if (err)
+		return err;
+
+	list_for_each_entry(re, rename_list, list) {
+		if (re->remove) {
+			err = ubi_remove_volume(re->desc, 1);
+			if (err)
+				break;
+		} else {
+			struct ubi_volume *vol = re->desc->vol;
+
+			spin_lock(&ubi->volumes_lock);
+			vol->name_len = re->new_name_len;
+			memcpy(vol->name, re->new_name, re->new_name_len + 1);
+			spin_unlock(&ubi->volumes_lock);
+			ubi_volume_notify(ubi, vol, UBI_VOLUME_RENAMED);
+		}
+	}
+
+	if (!err && paranoid_check_volumes(ubi))
+		;
+	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_gen("add volume %d", vol_id);
+
+	/* 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;
+	}
+
+	vol->dev.release = vol_release;
+	vol->dev.parent = &ubi->dev;
+	vol->dev.devt = dev;
+	vol->dev.class = ubi_class;
+	dev_set_name(&vol->dev, "%s_%d", ubi->ubi_name, vol->vol_id);
+	err = device_register(&vol->dev);
+	if (err)
+		goto out_cdev;
+
+	err = volume_sysfs_init(ubi, vol);
+	if (err) {
+		cdev_del(&vol->cdev);
+		volume_sysfs_close(vol);
+		return err;
+	}
+
+	if (paranoid_check_volumes(ubi))
+		dbg_err("check failed while adding volume %d", vol_id);
+	return err;
+
+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)
+{
+	dbg_gen("free volume %d", vol->vol_id);
+
+	ubi->volumes[vol->vol_id] = NULL;
+	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
+ *
+ * Returns zero if volume is all right and a a negative error code if not.
+ */
+static int 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 0;
+	}
+
+	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 0;
+
+fail:
+	ubi_err("paranoid check failed for volume %d", vol_id);
+	if (vol)
+		ubi_dbg_dump_vol_info(vol);
+	ubi_dbg_dump_vtbl_record(&ubi->vtbl[vol_id], vol_id);
+	dump_stack();
+	spin_unlock(&ubi->volumes_lock);
+	return -EINVAL;
+}
+
+/**
+ * paranoid_check_volumes - check information about all volumes.
+ * @ubi: UBI device description object
+ *
+ * Returns zero if volumes are all right and a a negative error code if not.
+ */
+static int paranoid_check_volumes(struct ubi_device *ubi)
+{
+	int i, err = 0;
+
+	for (i = 0; i < ubi->vtbl_slots; i++) {
+		err = paranoid_check_volume(ubi, i);
+		if (err)
+			break;
+	}
+
+	return err;
+}
+#endif

kernel/drivers/mtd/ubi/vtbl.c

@@ -0,0 +1,882 @@
+/*
+ * 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.
+ */
+
+#include <linux/crc32.h>
+#include <linux/err.h>
+#include <asm/div64.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;
+}
+
+/**
+ * ubi_vtbl_rename_volumes - rename UBI volumes in the volume table.
+ * @ubi: UBI device description object
+ * @rename_list: list of &struct ubi_rename_entry objects
+ *
+ * This function re-names multiple volumes specified in @req in the volume
+ * table. Returns zero in case of success and a negative error code in case of
+ * failure.
+ */
+int ubi_vtbl_rename_volumes(struct ubi_device *ubi,
+			    struct list_head *rename_list)
+{
+	int i, err;
+	struct ubi_rename_entry *re;
+	struct ubi_volume *layout_vol;
+
+	list_for_each_entry(re, rename_list, list) {
+		uint32_t crc;
+		struct ubi_volume *vol = re->desc->vol;
+		struct ubi_vtbl_record *vtbl_rec = &ubi->vtbl[vol->vol_id];
+
+		if (re->remove) {
+			memcpy(vtbl_rec, &empty_vtbl_record,
+			       sizeof(struct ubi_vtbl_record));
+			continue;
+		}
+
+		vtbl_rec->name_len = cpu_to_be16(re->new_name_len);
+		memcpy(vtbl_rec->name, re->new_name, re->new_name_len);
+		memset(vtbl_rec->name + re->new_name_len, 0,
+		       UBI_VOL_NAME_MAX + 1 - re->new_name_len);
+		crc = crc32(UBI_CRC32_INIT, vtbl_rec,
+			    UBI_VTBL_RECORD_SIZE_CRC);
+		vtbl_rec->crc = cpu_to_be32(crc);
+	}
+
+	layout_vol = ubi->volumes[vol_id2idx(ubi, UBI_LAYOUT_VOLUME_ID)];
+	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;
+	}
+
+	return 0;
+}
+
+/**
+ * vtbl_check - check if volume table is not corrupted and sensible.
+ * @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 = &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 %d, good PEBs %d",
+				reserved_pebs, 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(vtbl[i].name, 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);
+
+	/* 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_gen("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->upd_marker = vtbl[i].upd_marker;
+		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 than 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 the volume table.
+ * @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++) {
+		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 */