| // SPDX-License-Identifier: GPL-2.0+ |
| /* |
| * Simple MTD partitioning layer |
| * |
| * Copyright © 2000 Nicolas Pitre <nico@fluxnic.net> |
| * Copyright © 2002 Thomas Gleixner <gleixner@linutronix.de> |
| * Copyright © 2000-2010 David Woodhouse <dwmw2@infradead.org> |
| * |
| */ |
| |
| #ifndef __UBOOT__ |
| #include <log.h> |
| #include <dm/devres.h> |
| #include <linux/module.h> |
| #include <linux/types.h> |
| #include <linux/kernel.h> |
| #include <linux/slab.h> |
| #include <linux/list.h> |
| #include <linux/kmod.h> |
| #endif |
| |
| #include <malloc.h> |
| #include <memalign.h> |
| #include <part.h> |
| #include <linux/bug.h> |
| #include <linux/errno.h> |
| #include <linux/compat.h> |
| #include <ubi_uboot.h> |
| |
| #include <linux/mtd/mtd.h> |
| #include <linux/mtd/partitions.h> |
| #include <linux/err.h> |
| #include <linux/sizes.h> |
| |
| #include "mtdcore.h" |
| |
| #ifndef __UBOOT__ |
| static DEFINE_MUTEX(mtd_partitions_mutex); |
| #else |
| DEFINE_MUTEX(mtd_partitions_mutex); |
| #endif |
| |
| #ifdef __UBOOT__ |
| /* from mm/util.c */ |
| |
| /** |
| * kstrdup - allocate space for and copy an existing string |
| * @s: the string to duplicate |
| * @gfp: the GFP mask used in the kmalloc() call when allocating memory |
| */ |
| char *kstrdup(const char *s, gfp_t gfp) |
| { |
| size_t len; |
| char *buf; |
| |
| if (!s) |
| return NULL; |
| |
| len = strlen(s) + 1; |
| buf = kmalloc(len, gfp); |
| if (buf) |
| memcpy(buf, s, len); |
| return buf; |
| } |
| #endif |
| |
| #define MTD_SIZE_REMAINING (~0LLU) |
| #define MTD_OFFSET_NOT_SPECIFIED (~0LLU) |
| |
| bool mtd_partitions_used(struct mtd_info *master) |
| { |
| struct mtd_info *slave; |
| |
| list_for_each_entry(slave, &master->partitions, node) { |
| if (slave->usecount) |
| return true; |
| } |
| |
| return false; |
| } |
| |
| /** |
| * mtd_parse_partition - Parse @mtdparts partition definition, fill @partition |
| * with it and update the @mtdparts string pointer. |
| * |
| * The partition name is allocated and must be freed by the caller. |
| * |
| * This function is widely inspired from part_parse (mtdparts.c). |
| * |
| * @mtdparts: String describing the partition with mtdparts command syntax |
| * @partition: MTD partition structure to fill |
| * |
| * Return: 0 on success, an error otherwise. |
| */ |
| static int mtd_parse_partition(const char **_mtdparts, |
| struct mtd_partition *partition) |
| { |
| const char *mtdparts = *_mtdparts; |
| const char *name = NULL; |
| int name_len; |
| char *buf; |
| |
| /* Ensure the partition structure is empty */ |
| memset(partition, 0, sizeof(struct mtd_partition)); |
| |
| /* Fetch the partition size */ |
| if (*mtdparts == '-') { |
| /* Assign all remaining space to this partition */ |
| partition->size = MTD_SIZE_REMAINING; |
| mtdparts++; |
| } else { |
| partition->size = ustrtoull(mtdparts, (char **)&mtdparts, 0); |
| if (partition->size < SZ_4K) { |
| printf("Minimum partition size 4kiB, %lldB requested\n", |
| partition->size); |
| return -EINVAL; |
| } |
| } |
| |
| /* Check for the offset */ |
| partition->offset = MTD_OFFSET_NOT_SPECIFIED; |
| if (*mtdparts == '@') { |
| mtdparts++; |
| partition->offset = ustrtoull(mtdparts, (char **)&mtdparts, 0); |
| } |
| |
| /* Now look for the name */ |
| if (*mtdparts == '(') { |
| name = ++mtdparts; |
| mtdparts = strchr(name, ')'); |
| if (!mtdparts) { |
| printf("No closing ')' found in partition name\n"); |
| return -EINVAL; |
| } |
| name_len = mtdparts - name + 1; |
| if ((name_len - 1) == 0) { |
| printf("Empty partition name\n"); |
| return -EINVAL; |
| } |
| mtdparts++; |
| } else { |
| /* Name will be of the form size@offset */ |
| name_len = 22; |
| } |
| |
| /* Check if the partition is read-only */ |
| if (strncmp(mtdparts, "ro", 2) == 0) { |
| partition->mask_flags |= MTD_WRITEABLE; |
| mtdparts += 2; |
| } |
| |
| /* Check for a potential next partition definition */ |
| if (*mtdparts == ',') { |
| if (partition->size == MTD_SIZE_REMAINING) { |
| printf("No partitions allowed after a fill-up\n"); |
| return -EINVAL; |
| } |
| ++mtdparts; |
| } else if ((*mtdparts == ';') || (*mtdparts == '\0')) { |
| /* NOP */ |
| } else { |
| printf("Unexpected character '%c' in mtdparts\n", *mtdparts); |
| return -EINVAL; |
| } |
| |
| /* |
| * Allocate a buffer for the name and either copy the provided name or |
| * auto-generate it with the form 'size@offset'. |
| */ |
| buf = malloc(name_len); |
| if (!buf) |
| return -ENOMEM; |
| |
| if (name) |
| strncpy(buf, name, name_len - 1); |
| else |
| snprintf(buf, name_len, "0x%08llx@0x%08llx", |
| partition->size, partition->offset); |
| |
| buf[name_len - 1] = '\0'; |
| partition->name = buf; |
| |
| *_mtdparts = mtdparts; |
| |
| return 0; |
| } |
| |
| /** |
| * mtd_parse_partitions - Create a partition array from an mtdparts definition |
| * |
| * Stateless function that takes a @parent MTD device, a string @_mtdparts |
| * describing the partitions (with the "mtdparts" command syntax) and creates |
| * the corresponding MTD partition structure array @_parts. Both the name and |
| * the structure partition itself must be freed freed, the caller may use |
| * @mtd_free_parsed_partitions() for this purpose. |
| * |
| * @parent: MTD device which contains the partitions |
| * @_mtdparts: Pointer to a string describing the partitions with "mtdparts" |
| * command syntax. |
| * @_parts: Allocated array containing the partitions, must be freed by the |
| * caller. |
| * @_nparts: Size of @_parts array. |
| * |
| * Return: 0 on success, an error otherwise. |
| */ |
| int mtd_parse_partitions(struct mtd_info *parent, const char **_mtdparts, |
| struct mtd_partition **_parts, int *_nparts) |
| { |
| struct mtd_partition partition = {}, *parts; |
| const char *mtdparts = *_mtdparts; |
| uint64_t cur_off = 0, cur_sz = 0; |
| int nparts = 0; |
| int ret, idx; |
| u64 sz; |
| |
| /* First, iterate over the partitions until we know their number */ |
| while (mtdparts[0] != '\0' && mtdparts[0] != ';') { |
| ret = mtd_parse_partition(&mtdparts, &partition); |
| if (ret) |
| return ret; |
| |
| free((char *)partition.name); |
| nparts++; |
| } |
| |
| /* Allocate an array of partitions to give back to the caller */ |
| parts = malloc(sizeof(*parts) * nparts); |
| if (!parts) { |
| printf("Not enough space to save partitions meta-data\n"); |
| return -ENOMEM; |
| } |
| |
| /* Iterate again over each partition to save the data in our array */ |
| for (idx = 0; idx < nparts; idx++) { |
| ret = mtd_parse_partition(_mtdparts, &parts[idx]); |
| if (ret) |
| return ret; |
| |
| if (parts[idx].size == MTD_SIZE_REMAINING) |
| parts[idx].size = parent->size - cur_sz; |
| cur_sz += parts[idx].size; |
| |
| sz = parts[idx].size; |
| if (sz < parent->writesize || do_div(sz, parent->writesize)) { |
| printf("Partition size must be a multiple of %d\n", |
| parent->writesize); |
| return -EINVAL; |
| } |
| |
| if (parts[idx].offset == MTD_OFFSET_NOT_SPECIFIED) |
| parts[idx].offset = cur_off; |
| cur_off += parts[idx].size; |
| |
| parts[idx].ecclayout = parent->ecclayout; |
| } |
| |
| /* Offset by one mtdparts to point to the next device if any */ |
| if (*_mtdparts[0] == ';') |
| (*_mtdparts)++; |
| |
| *_parts = parts; |
| *_nparts = nparts; |
| |
| return 0; |
| } |
| |
| /** |
| * mtd_free_parsed_partitions - Free dynamically allocated partitions |
| * |
| * Each successful call to @mtd_parse_partitions must be followed by a call to |
| * @mtd_free_parsed_partitions to free any allocated array during the parsing |
| * process. |
| * |
| * @parts: Array containing the partitions that will be freed. |
| * @nparts: Size of @parts array. |
| */ |
| void mtd_free_parsed_partitions(struct mtd_partition *parts, |
| unsigned int nparts) |
| { |
| int i; |
| |
| for (i = 0; i < nparts; i++) |
| free((char *)parts[i].name); |
| |
| free(parts); |
| } |
| |
| /* |
| * MTD methods which simply translate the effective address and pass through |
| * to the _real_ device. |
| */ |
| |
| static int part_read(struct mtd_info *mtd, loff_t from, size_t len, |
| size_t *retlen, u_char *buf) |
| { |
| struct mtd_ecc_stats stats; |
| int res; |
| |
| stats = mtd->parent->ecc_stats; |
| res = mtd->parent->_read(mtd->parent, from + mtd->offset, len, |
| retlen, buf); |
| if (unlikely(mtd_is_eccerr(res))) |
| mtd->ecc_stats.failed += |
| mtd->parent->ecc_stats.failed - stats.failed; |
| else |
| mtd->ecc_stats.corrected += |
| mtd->parent->ecc_stats.corrected - stats.corrected; |
| return res; |
| } |
| |
| #ifndef __UBOOT__ |
| static int part_point(struct mtd_info *mtd, loff_t from, size_t len, |
| size_t *retlen, void **virt, resource_size_t *phys) |
| { |
| return mtd->parent->_point(mtd->parent, from + mtd->offset, len, |
| retlen, virt, phys); |
| } |
| |
| static int part_unpoint(struct mtd_info *mtd, loff_t from, size_t len) |
| { |
| return mtd->parent->_unpoint(mtd->parent, from + mtd->offset, len); |
| } |
| #endif |
| |
| static unsigned long part_get_unmapped_area(struct mtd_info *mtd, |
| unsigned long len, |
| unsigned long offset, |
| unsigned long flags) |
| { |
| offset += mtd->offset; |
| return mtd->parent->_get_unmapped_area(mtd->parent, len, offset, flags); |
| } |
| |
| static int part_read_oob(struct mtd_info *mtd, loff_t from, |
| struct mtd_oob_ops *ops) |
| { |
| int res; |
| |
| if (from >= mtd->size) |
| return -EINVAL; |
| if (ops->datbuf && from + ops->len > mtd->size) |
| return -EINVAL; |
| |
| /* |
| * If OOB is also requested, make sure that we do not read past the end |
| * of this partition. |
| */ |
| if (ops->oobbuf) { |
| size_t len, pages; |
| |
| if (ops->mode == MTD_OPS_AUTO_OOB) |
| len = mtd->oobavail; |
| else |
| len = mtd->oobsize; |
| pages = mtd_div_by_ws(mtd->size, mtd); |
| pages -= mtd_div_by_ws(from, mtd); |
| if (ops->ooboffs + ops->ooblen > pages * len) |
| return -EINVAL; |
| } |
| |
| res = mtd->parent->_read_oob(mtd->parent, from + mtd->offset, ops); |
| if (unlikely(res)) { |
| if (mtd_is_bitflip(res)) |
| mtd->ecc_stats.corrected++; |
| if (mtd_is_eccerr(res)) |
| mtd->ecc_stats.failed++; |
| } |
| return res; |
| } |
| |
| static int part_read_user_prot_reg(struct mtd_info *mtd, loff_t from, |
| size_t len, size_t *retlen, u_char *buf) |
| { |
| return mtd->parent->_read_user_prot_reg(mtd->parent, from, len, |
| retlen, buf); |
| } |
| |
| static int part_get_user_prot_info(struct mtd_info *mtd, size_t len, |
| size_t *retlen, struct otp_info *buf) |
| { |
| return mtd->parent->_get_user_prot_info(mtd->parent, len, retlen, |
| buf); |
| } |
| |
| static int part_read_fact_prot_reg(struct mtd_info *mtd, loff_t from, |
| size_t len, size_t *retlen, u_char *buf) |
| { |
| return mtd->parent->_read_fact_prot_reg(mtd->parent, from, len, |
| retlen, buf); |
| } |
| |
| static int part_get_fact_prot_info(struct mtd_info *mtd, size_t len, |
| size_t *retlen, struct otp_info *buf) |
| { |
| return mtd->parent->_get_fact_prot_info(mtd->parent, len, retlen, |
| buf); |
| } |
| |
| static int part_write(struct mtd_info *mtd, loff_t to, size_t len, |
| size_t *retlen, const u_char *buf) |
| { |
| return mtd->parent->_write(mtd->parent, to + mtd->offset, len, |
| retlen, buf); |
| } |
| |
| static int part_panic_write(struct mtd_info *mtd, loff_t to, size_t len, |
| size_t *retlen, const u_char *buf) |
| { |
| return mtd->parent->_panic_write(mtd->parent, to + mtd->offset, len, |
| retlen, buf); |
| } |
| |
| static int part_write_oob(struct mtd_info *mtd, loff_t to, |
| struct mtd_oob_ops *ops) |
| { |
| if (to >= mtd->size) |
| return -EINVAL; |
| if (ops->datbuf && to + ops->len > mtd->size) |
| return -EINVAL; |
| return mtd->parent->_write_oob(mtd->parent, to + mtd->offset, ops); |
| } |
| |
| static int part_write_user_prot_reg(struct mtd_info *mtd, loff_t from, |
| size_t len, size_t *retlen, u_char *buf) |
| { |
| return mtd->parent->_write_user_prot_reg(mtd->parent, from, len, |
| retlen, buf); |
| } |
| |
| static int part_lock_user_prot_reg(struct mtd_info *mtd, loff_t from, |
| size_t len) |
| { |
| return mtd->parent->_lock_user_prot_reg(mtd->parent, from, len); |
| } |
| |
| #ifndef __UBOOT__ |
| static int part_writev(struct mtd_info *mtd, const struct kvec *vecs, |
| unsigned long count, loff_t to, size_t *retlen) |
| { |
| return mtd->parent->_writev(mtd->parent, vecs, count, |
| to + mtd->offset, retlen); |
| } |
| #endif |
| |
| static int part_erase(struct mtd_info *mtd, struct erase_info *instr) |
| { |
| int ret; |
| |
| instr->addr += mtd->offset; |
| |
| ret = mtd->parent->_erase(mtd->parent, instr); |
| if (ret && instr->fail_addr != MTD_FAIL_ADDR_UNKNOWN) |
| instr->fail_addr -= mtd->offset; |
| |
| instr->addr -= mtd->offset; |
| |
| return ret; |
| } |
| |
| static int part_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len) |
| { |
| return mtd->parent->_lock(mtd->parent, ofs + mtd->offset, len); |
| } |
| |
| static int part_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len) |
| { |
| return mtd->parent->_unlock(mtd->parent, ofs + mtd->offset, len); |
| } |
| |
| static int part_is_locked(struct mtd_info *mtd, loff_t ofs, uint64_t len) |
| { |
| return mtd->parent->_is_locked(mtd->parent, ofs + mtd->offset, len); |
| } |
| |
| static void part_sync(struct mtd_info *mtd) |
| { |
| mtd->parent->_sync(mtd->parent); |
| } |
| |
| #ifndef __UBOOT__ |
| static int part_suspend(struct mtd_info *mtd) |
| { |
| return mtd->parent->_suspend(mtd->parent); |
| } |
| |
| static void part_resume(struct mtd_info *mtd) |
| { |
| mtd->parent->_resume(mtd->parent); |
| } |
| #endif |
| |
| static int part_block_isreserved(struct mtd_info *mtd, loff_t ofs) |
| { |
| ofs += mtd->offset; |
| return mtd->parent->_block_isreserved(mtd->parent, ofs); |
| } |
| |
| static int part_block_isbad(struct mtd_info *mtd, loff_t ofs) |
| { |
| ofs += mtd->offset; |
| return mtd->parent->_block_isbad(mtd->parent, ofs); |
| } |
| |
| static int part_block_markbad(struct mtd_info *mtd, loff_t ofs) |
| { |
| int res; |
| |
| ofs += mtd->offset; |
| res = mtd->parent->_block_markbad(mtd->parent, ofs); |
| if (!res) |
| mtd->ecc_stats.badblocks++; |
| return res; |
| } |
| |
| static inline void free_partition(struct mtd_info *p) |
| { |
| kfree(p->name); |
| kfree(p); |
| } |
| |
| /* |
| * This function unregisters and destroy all slave MTD objects which are |
| * attached to the given master MTD object, recursively. |
| */ |
| static int do_del_mtd_partitions(struct mtd_info *master) |
| { |
| struct mtd_info *slave, *next; |
| int ret, err = 0; |
| |
| list_for_each_entry_safe(slave, next, &master->partitions, node) { |
| if (mtd_has_partitions(slave)) |
| del_mtd_partitions(slave); |
| |
| debug("Deleting %s MTD partition\n", slave->name); |
| ret = del_mtd_device(slave); |
| if (ret < 0) { |
| printf("Error when deleting partition \"%s\" (%d)\n", |
| slave->name, ret); |
| err = ret; |
| continue; |
| } |
| |
| list_del(&slave->node); |
| free_partition(slave); |
| } |
| |
| return err; |
| } |
| |
| int del_mtd_partitions(struct mtd_info *master) |
| { |
| int ret; |
| |
| debug("Deleting MTD partitions on \"%s\":\n", master->name); |
| |
| mutex_lock(&mtd_partitions_mutex); |
| ret = do_del_mtd_partitions(master); |
| mutex_unlock(&mtd_partitions_mutex); |
| |
| return ret; |
| } |
| |
| static struct mtd_info *allocate_partition(struct mtd_info *master, |
| const struct mtd_partition *part, |
| int partno, uint64_t cur_offset) |
| { |
| struct mtd_info *slave; |
| char *name; |
| |
| /* allocate the partition structure */ |
| slave = kzalloc(sizeof(*slave), GFP_KERNEL); |
| name = kstrdup(part->name, GFP_KERNEL); |
| if (!name || !slave) { |
| printk(KERN_ERR"memory allocation error while creating partitions for \"%s\"\n", |
| master->name); |
| kfree(name); |
| kfree(slave); |
| return ERR_PTR(-ENOMEM); |
| } |
| |
| /* set up the MTD object for this partition */ |
| slave->type = master->type; |
| slave->flags = master->flags & ~part->mask_flags; |
| slave->size = part->size; |
| slave->writesize = master->writesize; |
| slave->writebufsize = master->writebufsize; |
| slave->oobsize = master->oobsize; |
| slave->oobavail = master->oobavail; |
| slave->subpage_sft = master->subpage_sft; |
| |
| slave->name = name; |
| slave->owner = master->owner; |
| #ifndef __UBOOT__ |
| slave->backing_dev_info = master->backing_dev_info; |
| |
| /* NOTE: we don't arrange MTDs as a tree; it'd be error-prone |
| * to have the same data be in two different partitions. |
| */ |
| slave->dev.parent = master->dev.parent; |
| #endif |
| |
| if (master->_read) |
| slave->_read = part_read; |
| if (master->_write) |
| slave->_write = part_write; |
| |
| if (master->_panic_write) |
| slave->_panic_write = part_panic_write; |
| |
| #ifndef __UBOOT__ |
| if (master->_point && master->_unpoint) { |
| slave->_point = part_point; |
| slave->_unpoint = part_unpoint; |
| } |
| #endif |
| |
| if (master->_get_unmapped_area) |
| slave->_get_unmapped_area = part_get_unmapped_area; |
| if (master->_read_oob) |
| slave->_read_oob = part_read_oob; |
| if (master->_write_oob) |
| slave->_write_oob = part_write_oob; |
| if (master->_read_user_prot_reg) |
| slave->_read_user_prot_reg = part_read_user_prot_reg; |
| if (master->_read_fact_prot_reg) |
| slave->_read_fact_prot_reg = part_read_fact_prot_reg; |
| if (master->_write_user_prot_reg) |
| slave->_write_user_prot_reg = part_write_user_prot_reg; |
| if (master->_lock_user_prot_reg) |
| slave->_lock_user_prot_reg = part_lock_user_prot_reg; |
| if (master->_get_user_prot_info) |
| slave->_get_user_prot_info = part_get_user_prot_info; |
| if (master->_get_fact_prot_info) |
| slave->_get_fact_prot_info = part_get_fact_prot_info; |
| if (master->_sync) |
| slave->_sync = part_sync; |
| #ifndef __UBOOT__ |
| if (!partno && !master->dev.class && master->_suspend && |
| master->_resume) { |
| slave->_suspend = part_suspend; |
| slave->_resume = part_resume; |
| } |
| if (master->_writev) |
| slave->_writev = part_writev; |
| #endif |
| if (master->_lock) |
| slave->_lock = part_lock; |
| if (master->_unlock) |
| slave->_unlock = part_unlock; |
| if (master->_is_locked) |
| slave->_is_locked = part_is_locked; |
| if (master->_block_isreserved) |
| slave->_block_isreserved = part_block_isreserved; |
| if (master->_block_isbad) |
| slave->_block_isbad = part_block_isbad; |
| if (master->_block_markbad) |
| slave->_block_markbad = part_block_markbad; |
| slave->_erase = part_erase; |
| slave->parent = master; |
| slave->offset = part->offset; |
| INIT_LIST_HEAD(&slave->partitions); |
| INIT_LIST_HEAD(&slave->node); |
| |
| if (slave->offset == MTDPART_OFS_APPEND) |
| slave->offset = cur_offset; |
| if (slave->offset == MTDPART_OFS_NXTBLK) { |
| slave->offset = cur_offset; |
| if (mtd_mod_by_eb(cur_offset, master) != 0) { |
| /* Round up to next erasesize */ |
| slave->offset = (mtd_div_by_eb(cur_offset, master) + 1) * master->erasesize; |
| debug("Moving partition %d: " |
| "0x%012llx -> 0x%012llx\n", partno, |
| (unsigned long long)cur_offset, (unsigned long long)slave->offset); |
| } |
| } |
| if (slave->offset == MTDPART_OFS_RETAIN) { |
| slave->offset = cur_offset; |
| if (master->size - slave->offset >= slave->size) { |
| slave->size = master->size - slave->offset |
| - slave->size; |
| } else { |
| debug("mtd partition \"%s\" doesn't have enough space: %#llx < %#llx, disabled\n", |
| part->name, master->size - slave->offset, |
| slave->size); |
| /* register to preserve ordering */ |
| goto out_register; |
| } |
| } |
| if (slave->size == MTDPART_SIZ_FULL) |
| slave->size = master->size - slave->offset; |
| |
| debug("0x%012llx-0x%012llx : \"%s\"\n", (unsigned long long)slave->offset, |
| (unsigned long long)(slave->offset + slave->size), slave->name); |
| |
| /* let's do some sanity checks */ |
| if (slave->offset >= master->size) { |
| /* let's register it anyway to preserve ordering */ |
| slave->offset = 0; |
| slave->size = 0; |
| printk(KERN_ERR"mtd: partition \"%s\" is out of reach -- disabled\n", |
| part->name); |
| goto out_register; |
| } |
| if (slave->offset + slave->size > master->size) { |
| slave->size = master->size - slave->offset; |
| printk(KERN_WARNING"mtd: partition \"%s\" extends beyond the end of device \"%s\" -- size truncated to %#llx\n", |
| part->name, master->name, slave->size); |
| } |
| if (master->numeraseregions > 1) { |
| /* Deal with variable erase size stuff */ |
| int i, max = master->numeraseregions; |
| u64 end = slave->offset + slave->size; |
| struct mtd_erase_region_info *regions = master->eraseregions; |
| |
| /* Find the first erase regions which is part of this |
| * partition. */ |
| for (i = 0; i < max && regions[i].offset <= slave->offset; i++) |
| ; |
| /* The loop searched for the region _behind_ the first one */ |
| if (i > 0) |
| i--; |
| |
| /* Pick biggest erasesize */ |
| for (; i < max && regions[i].offset < end; i++) { |
| if (slave->erasesize < regions[i].erasesize) |
| slave->erasesize = regions[i].erasesize; |
| } |
| WARN_ON(slave->erasesize == 0); |
| } else { |
| /* Single erase size */ |
| slave->erasesize = master->erasesize; |
| } |
| |
| if ((slave->flags & MTD_WRITEABLE) && |
| mtd_mod_by_eb(slave->offset, slave)) { |
| /* Doesn't start on a boundary of major erase size */ |
| /* FIXME: Let it be writable if it is on a boundary of |
| * _minor_ erase size though */ |
| slave->flags &= ~MTD_WRITEABLE; |
| printk(KERN_WARNING"mtd: partition \"%s\" doesn't start on an erase block boundary -- force read-only\n", |
| part->name); |
| } |
| if ((slave->flags & MTD_WRITEABLE) && |
| mtd_mod_by_eb(slave->size, slave)) { |
| slave->flags &= ~MTD_WRITEABLE; |
| printk(KERN_WARNING"mtd: partition \"%s\" doesn't end on an erase block -- force read-only\n", |
| part->name); |
| } |
| |
| slave->ecclayout = master->ecclayout; |
| slave->ecc_step_size = master->ecc_step_size; |
| slave->ecc_strength = master->ecc_strength; |
| slave->bitflip_threshold = master->bitflip_threshold; |
| |
| if (master->_block_isbad) { |
| uint64_t offs = 0; |
| |
| while (offs < slave->size) { |
| if (mtd_block_isbad(master, offs + slave->offset)) |
| slave->ecc_stats.badblocks++; |
| offs += slave->erasesize; |
| } |
| } |
| |
| out_register: |
| return slave; |
| } |
| |
| #ifndef __UBOOT__ |
| int mtd_add_partition(struct mtd_info *master, const char *name, |
| long long offset, long long length) |
| { |
| struct mtd_partition part; |
| struct mtd_info *p, *new; |
| uint64_t start, end; |
| int ret = 0; |
| |
| /* the direct offset is expected */ |
| if (offset == MTDPART_OFS_APPEND || |
| offset == MTDPART_OFS_NXTBLK) |
| return -EINVAL; |
| |
| if (length == MTDPART_SIZ_FULL) |
| length = master->size - offset; |
| |
| if (length <= 0) |
| return -EINVAL; |
| |
| part.name = name; |
| part.size = length; |
| part.offset = offset; |
| part.mask_flags = 0; |
| part.ecclayout = NULL; |
| |
| new = allocate_partition(master, &part, -1, offset); |
| if (IS_ERR(new)) |
| return PTR_ERR(new); |
| |
| start = offset; |
| end = offset + length; |
| |
| mutex_lock(&mtd_partitions_mutex); |
| list_for_each_entry(p, &master->partitions, node) { |
| if (start >= p->offset && |
| (start < (p->offset + p->size))) |
| goto err_inv; |
| |
| if (end >= p->offset && |
| (end < (p->offset + p->size))) |
| goto err_inv; |
| } |
| |
| list_add_tail(&new->node, &master->partitions); |
| mutex_unlock(&mtd_partitions_mutex); |
| |
| add_mtd_device(new); |
| |
| return ret; |
| err_inv: |
| mutex_unlock(&mtd_partitions_mutex); |
| free_partition(new); |
| return -EINVAL; |
| } |
| EXPORT_SYMBOL_GPL(mtd_add_partition); |
| |
| int mtd_del_partition(struct mtd_info *master, int partno) |
| { |
| struct mtd_info *slave, *next; |
| int ret = -EINVAL; |
| |
| mutex_lock(&mtd_partitions_mutex); |
| list_for_each_entry_safe(slave, next, &master->partitions, node) |
| if (slave->index == partno) { |
| ret = del_mtd_device(slave); |
| if (ret < 0) |
| break; |
| |
| list_del(&slave->node); |
| free_partition(slave); |
| break; |
| } |
| mutex_unlock(&mtd_partitions_mutex); |
| |
| return ret; |
| } |
| EXPORT_SYMBOL_GPL(mtd_del_partition); |
| #endif |
| |
| /* |
| * This function, given a master MTD object and a partition table, creates |
| * and registers slave MTD objects which are bound to the master according to |
| * the partition definitions. |
| * |
| * We don't register the master, or expect the caller to have done so, |
| * for reasons of data integrity. |
| */ |
| |
| int add_mtd_partitions(struct mtd_info *master, |
| const struct mtd_partition *parts, |
| int nbparts) |
| { |
| struct mtd_info *slave; |
| uint64_t cur_offset = 0; |
| int i; |
| |
| debug("Creating %d MTD partitions on \"%s\":\n", nbparts, master->name); |
| |
| for (i = 0; i < nbparts; i++) { |
| slave = allocate_partition(master, parts + i, i, cur_offset); |
| if (IS_ERR(slave)) |
| return PTR_ERR(slave); |
| |
| mutex_lock(&mtd_partitions_mutex); |
| list_add_tail(&slave->node, &master->partitions); |
| mutex_unlock(&mtd_partitions_mutex); |
| |
| add_mtd_device(slave); |
| |
| cur_offset = slave->offset + slave->size; |
| } |
| |
| return 0; |
| } |
| |
| #if CONFIG_IS_ENABLED(DM) && CONFIG_IS_ENABLED(OF_CONTROL) |
| int add_mtd_partitions_of(struct mtd_info *master) |
| { |
| ofnode parts, child; |
| int i = 0; |
| |
| if (!master->dev && !ofnode_valid(master->flash_node)) |
| return 0; |
| |
| if (master->dev) |
| parts = ofnode_find_subnode(mtd_get_ofnode(master), "partitions"); |
| else |
| parts = ofnode_find_subnode(master->flash_node, "partitions"); |
| |
| if (!ofnode_valid(parts) || !ofnode_is_enabled(parts) || |
| !ofnode_device_is_compatible(parts, "fixed-partitions")) |
| return 0; |
| |
| ofnode_for_each_subnode(child, parts) { |
| struct mtd_partition part = { 0 }; |
| struct mtd_info *slave; |
| fdt_addr_t offset; |
| fdt_size_t size; |
| |
| if (!ofnode_is_enabled(child)) |
| continue; |
| |
| offset = ofnode_get_addr_size_index_notrans(child, 0, &size); |
| if (offset == FDT_ADDR_T_NONE || !size) { |
| debug("Missing partition offset/size on \"%s\" partition\n", |
| master->name); |
| continue; |
| } |
| |
| part.name = ofnode_read_string(child, "label"); |
| if (!part.name) |
| part.name = ofnode_read_string(child, "name"); |
| |
| /* |
| * .mask_flags is used to remove flags in allocate_partition(), |
| * so when "read-only" is present, we add MTD_WRITABLE to the |
| * mask, and so MTD_WRITABLE will be removed on partition |
| * allocation |
| */ |
| if (ofnode_read_bool(child, "read-only")) |
| part.mask_flags |= MTD_WRITEABLE; |
| if (ofnode_read_bool(child, "lock")) |
| part.mask_flags |= MTD_POWERUP_LOCK; |
| |
| part.offset = offset; |
| part.size = size; |
| part.ecclayout = master->ecclayout; |
| |
| slave = allocate_partition(master, &part, i++, 0); |
| if (IS_ERR(slave)) |
| return PTR_ERR(slave); |
| |
| mutex_lock(&mtd_partitions_mutex); |
| list_add_tail(&slave->node, &master->partitions); |
| mutex_unlock(&mtd_partitions_mutex); |
| |
| add_mtd_device(slave); |
| } |
| |
| return 0; |
| } |
| #endif /* CONFIG_IS_ENABLED(DM) && CONFIG_IS_ENABLED(OF_CONTROL) */ |
| |
| #ifndef __UBOOT__ |
| static DEFINE_SPINLOCK(part_parser_lock); |
| static LIST_HEAD(part_parsers); |
| |
| static struct mtd_part_parser *get_partition_parser(const char *name) |
| { |
| struct mtd_part_parser *p, *ret = NULL; |
| |
| spin_lock(&part_parser_lock); |
| |
| list_for_each_entry(p, &part_parsers, list) |
| if (!strcmp(p->name, name) && try_module_get(p->owner)) { |
| ret = p; |
| break; |
| } |
| |
| spin_unlock(&part_parser_lock); |
| |
| return ret; |
| } |
| |
| #define put_partition_parser(p) do { module_put((p)->owner); } while (0) |
| |
| void register_mtd_parser(struct mtd_part_parser *p) |
| { |
| spin_lock(&part_parser_lock); |
| list_add(&p->list, &part_parsers); |
| spin_unlock(&part_parser_lock); |
| } |
| EXPORT_SYMBOL_GPL(register_mtd_parser); |
| |
| void deregister_mtd_parser(struct mtd_part_parser *p) |
| { |
| spin_lock(&part_parser_lock); |
| list_del(&p->list); |
| spin_unlock(&part_parser_lock); |
| } |
| EXPORT_SYMBOL_GPL(deregister_mtd_parser); |
| |
| /* |
| * Do not forget to update 'parse_mtd_partitions()' kerneldoc comment if you |
| * are changing this array! |
| */ |
| static const char * const default_mtd_part_types[] = { |
| "cmdlinepart", |
| "ofpart", |
| NULL |
| }; |
| |
| /** |
| * parse_mtd_partitions - parse MTD partitions |
| * @master: the master partition (describes whole MTD device) |
| * @types: names of partition parsers to try or %NULL |
| * @pparts: array of partitions found is returned here |
| * @data: MTD partition parser-specific data |
| * |
| * This function tries to find partition on MTD device @master. It uses MTD |
| * partition parsers, specified in @types. However, if @types is %NULL, then |
| * the default list of parsers is used. The default list contains only the |
| * "cmdlinepart" and "ofpart" parsers ATM. |
| * Note: If there are more then one parser in @types, the kernel only takes the |
| * partitions parsed out by the first parser. |
| * |
| * This function may return: |
| * o a negative error code in case of failure |
| * o zero if no partitions were found |
| * o a positive number of found partitions, in which case on exit @pparts will |
| * point to an array containing this number of &struct mtd_info objects. |
| */ |
| int parse_mtd_partitions(struct mtd_info *master, const char *const *types, |
| struct mtd_partition **pparts, |
| struct mtd_part_parser_data *data) |
| { |
| struct mtd_part_parser *parser; |
| int ret = 0; |
| |
| if (!types) |
| types = default_mtd_part_types; |
| |
| for ( ; ret <= 0 && *types; types++) { |
| parser = get_partition_parser(*types); |
| if (!parser && !request_module("%s", *types)) |
| parser = get_partition_parser(*types); |
| if (!parser) |
| continue; |
| ret = (*parser->parse_fn)(master, pparts, data); |
| put_partition_parser(parser); |
| if (ret > 0) { |
| printk(KERN_NOTICE "%d %s partitions found on MTD device %s\n", |
| ret, parser->name, master->name); |
| break; |
| } |
| } |
| return ret; |
| } |
| #endif |
| |
| /* Returns the size of the entire flash chip */ |
| uint64_t mtd_get_device_size(const struct mtd_info *mtd) |
| { |
| if (mtd_is_partition(mtd)) |
| return mtd->parent->size; |
| |
| return mtd->size; |
| } |
| EXPORT_SYMBOL_GPL(mtd_get_device_size); |
| |
| static struct mtd_info *mtd_get_partition_by_index(struct mtd_info *mtd, int index) |
| { |
| struct mtd_info *part; |
| int i = 0; |
| |
| list_for_each_entry(part, &mtd->partitions, node) |
| if (i++ == index) |
| return part; |
| |
| debug("Partition with idx=%d not found on MTD device %s\n", index, mtd->name); |
| return NULL; |
| } |
| |
| static int __maybe_unused part_get_info_mtd(struct blk_desc *dev_desc, int part_idx, |
| struct disk_partition *info) |
| { |
| struct mtd_info *master = blk_desc_to_mtd(dev_desc); |
| struct mtd_info *part; |
| |
| if (!master) { |
| debug("MTD device is NULL\n"); |
| return -EINVAL; |
| } |
| |
| part = mtd_get_partition_by_index(master, part_idx); |
| if (!part) { |
| debug("Failed to find partition with idx=%d\n", part_idx); |
| return -EINVAL; |
| } |
| |
| snprintf(info->name, PART_NAME_LEN, part->name); |
| info->start = part->offset / dev_desc->blksz; |
| info->size = part->size / dev_desc->blksz; |
| info->blksz = dev_desc->blksz; |
| |
| return 0; |
| } |
| |
| static void __maybe_unused part_print_mtd(struct blk_desc *dev_desc) |
| { |
| struct mtd_info *master = blk_desc_to_mtd(dev_desc); |
| struct mtd_info *part; |
| |
| if (!master) |
| return; |
| |
| list_for_each_entry(part, &master->partitions, node) |
| printf("- 0x%012llx-0x%012llx : \"%s\"\n", |
| part->offset, part->offset + part->size, part->name); |
| } |
| |
| static int part_test_mtd(struct blk_desc *dev_desc) |
| { |
| struct mtd_info *master = blk_desc_to_mtd(dev_desc); |
| ALLOC_CACHE_ALIGN_BUFFER(unsigned char, buffer, dev_desc->blksz); |
| |
| if (!master) |
| return -1; |
| |
| if (blk_dread(dev_desc, 0, 1, (ulong *)buffer) != 1) |
| return -1; |
| |
| return 0; |
| } |
| |
| U_BOOT_PART_TYPE(mtd) = { |
| .name = "MTD", |
| .part_type = PART_TYPE_MTD, |
| .max_entries = MTD_ENTRY_NUMBERS, |
| .get_info = part_get_info_ptr(part_get_info_mtd), |
| .print = part_print_ptr(part_print_mtd), |
| .test = part_test_mtd, |
| }; |