From f175083413f0f94de88def865eeb65e465ded389 Mon Sep 17 00:00:00 2001 From: Frank Haverkamp Date: Wed, 14 Jun 2006 11:53:59 +0200 Subject: UBI - Unsorted Block Images UBI (Latin: "where?") manages multiple logical volumes on a single flash device, specifically supporting NAND flash devices. UBI provides a flexible partitioning concept which still allows for wear-levelling across the whole flash device. In a sense, UBI may be compared to the Logical Volume Manager (LVM). Whereas LVM maps logical sector numbers to physical HDD sector numbers, UBI maps logical eraseblocks to physical eraseblocks. More information may be found in the UBI design documentation: ubidesign.pdf. Which can be found here: http://www.linux-mtd.infradead.org/doc/ubi.html Partitioning/Re-partitioning An UBI volume occupies a certain number of erase blocks. This is limited by a configured maximum volume size, which could also be viewed as the partition size. Each individual UBI volume's size can be changed independently of the other UBI volumes, provided that the sum of all volume sizes doesn't exceed a certain limit. UBI supports dynamic volumes and static volumes. Static volumes are read-only and their contents are protected by CRC check sums. Bad eraseblocks handling UBI transparently handles bad eraseblocks. When a physical eraseblock becomes bad, it is substituted by a good physical eraseblock, and the user does not even notice this. Scrubbing On a NAND flash bit flips can occur on any write operation, sometimes also on read. If bit flips persist on the device, at first they can still be corrected by ECC, but once they accumulate, correction will become impossible. Thus it is best to actively scrub the affected eraseblock, by first copying it to a free eraseblock and then erasing the original. The UBI layer performs this type of scrubbing under the covers, transparently to the UBI volume users. Erase Counts UBI maintains an erase count header per eraseblock. This frees higher-level layers (like file systems) from doing this and allows for centralized erase count management instead. The erase counts are used by the wear-levelling algorithm in the UBI layer. The algorithm itself is exchangeable. Booting from NAND For booting directly from NAND flash the hardware must at least be capable of fetching and executing a small portion of the NAND flash. Some NAND flash controllers have this kind of support. They usually limit the window to a few kilobytes in erase block 0. This "initial program loader" (IPL) must then contain sufficient logic to load and execute the next boot phase. Due to bad eraseblocks, which may be randomly scattered over the flash device, it is problematic to store the "secondary program loader" (SPL) statically. Also, due to bit-flips it may become corrupted over time. UBI allows to solve this problem gracefully by storing the SPL in a small static UBI volume. UBI volumes vs. static partitions UBI volumes are still very similar to static MTD partitions: * both consist of eraseblocks (logical eraseblocks in case of UBI volumes, and physical eraseblocks in case of static partitions; * both support three basic operations - read, write, erase. But UBI volumes have the following advantages over traditional static MTD partitions: * there are no eraseblock wear-leveling constraints in case of UBI volumes, so the user should not care about this; * there are no bit-flips and bad eraseblocks in case of UBI volumes. So, UBI volumes may be considered as flash devices with relaxed restrictions. Where can it be found? Documentation, kernel code and applications can be found in the MTD gits. What are the applications for? The applications help to create binary flash images for two purposes: pfi files (partial flash images) for in-system update of UBI volumes, and plain binary images, with or without OOB data in case of NAND, for a manufacturing step. Furthermore some tools are/and will be created that allow flash content analysis after a system has crashed. Who did UBI? The original ideas, where UBI is based on, were developed by Andreas Arnez, Frank Haverkamp and Thomas Gleixner. Josh W. Boyer and some others were involved too. The implementation of the kernel layer was done by Artem B. Bityutskiy. The user-space applications and tools were written by Oliver Lohmann with contributions from Frank Haverkamp, Andreas Arnez, and Artem. Joern Engel contributed a patch which modifies JFFS2 so that it can be run on a UBI volume. Thomas Gleixner did modifications to the NAND layer and also some to JFFS2 to make it work. Signed-off-by: Frank Haverkamp --- ubi-utils/src/pfi2bin/pfi2bin.c | 678 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 678 insertions(+) create mode 100644 ubi-utils/src/pfi2bin/pfi2bin.c (limited to 'ubi-utils/src/pfi2bin') diff --git a/ubi-utils/src/pfi2bin/pfi2bin.c b/ubi-utils/src/pfi2bin/pfi2bin.c new file mode 100644 index 0000000..6536c19 --- /dev/null +++ b/ubi-utils/src/pfi2bin/pfi2bin.c @@ -0,0 +1,678 @@ +/* + * 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., 675 Mass Ave, Cambridge, MA 02139, USA. + * + * Author: Oliver Lohmann + * + * Convert a PFI file (partial flash image) into a plain binary file. + * This tool can be used to prepare the data to be burned into flash + * chips in a manufacturing step where the flashes are written before + * being soldered onto the hardware. For NAND images another step is + * required to add the right OOB data to the binary image. + */ + +#include +#include +#include +#include +#include +#include +#include + +#include +#include + +#include "config.h" +#include "list.h" +#include "error.h" +#include "reader.h" +#include "peb.h" +#include "crc32.h" + +#define MAX_FNAME 255 +#define DEFAULT_ERASE_COUNT 0 /* Hmmm.... Perhaps */ +#define ERR_BUF_SIZE 1024 + +#define MIN(a,b) ((a) < (b) ? (a) : (b)) + +static uint32_t crc32_table[256]; +static char err_buf[ERR_BUF_SIZE]; + +/* + * Data used to buffer raw blocks which have to be + * located at a specific point inside the generated RAW file + */ + +typedef enum action_t { + ACT_NOTHING = 0x00000000, + ACT_RAW = 0x00000001, +} action_t; + +static const char copyright [] __attribute__((unused)) = + "Licensed Materials - Property of IBM\n" + "IBM Flexible Support Processor Licensed Material\n" + "(c) Copyright IBM Corp 2006 All Rights Reserved.\n" + "US Government Users Restricted Rights - Use, duplication\n" + "or disclosure restricted by GSA ADP Schedule Contract\n" + "with IBM Corp."; + +static error_t parse_opt (int key, char *arg, struct argp_state *state); + +const char *argp_program_version = PACKAGE_VERSION; +const char *argp_program_bug_address = PACKAGE_BUGREPORT; +static char doc[] = "\nVersion: " PACKAGE_VERSION "\n\tBuilt on " + BUILD_CPU" "BUILD_OS" at "__DATE__" "__TIME__"\n" + "\n" + "pfi2bin - a tool to convert PFI files into binary images.\n"; + +static struct argp_option options[] = { + /* COMMON */ + { name: NULL, key: 0, arg: NULL, flags: 0, + doc: "Common settings:", + group: OPTION_ARG_OPTIONAL}, + + { name: "verbose", key: 'v', arg: NULL, flags: 0, + doc: "Print more information.", + group: OPTION_ARG_OPTIONAL }, + + { name: "copyright", key: 'c', arg: NULL, flags: 0, + group: OPTION_ARG_OPTIONAL }, + + + /* INPUT */ + { name: NULL, key: 0, arg: NULL, flags: 0, + doc: "Input:", + group: 4}, + + { name: "platform", key: 'j', arg: "pdd-file", flags: 0, + doc: "PDD information which contains the card settings.", + group: 4 }, + + /* OUTPUT */ + { name: NULL, key: 0, arg: NULL, flags: 0, + doc: "Output:", + group: 5}, + + { name: "output", key: 'o', arg: "filename", flags: 0, + doc: "Outputfile, default: stdout.", + group: 5 }, + + { name: NULL, key: 0, arg: NULL, flags: 0, doc: NULL, group: 0 }, +}; + +typedef struct io { + FILE* fp_pdd; /* a FilePointer to the PDD data */ + FILE* fp_pfi; /* a FilePointer to the PFI input stream */ + FILE* fp_out; /* a FilePointer to the output stream */ +} *io_t; + +typedef struct myargs { + /* common settings */ + action_t action; + int verbose; + const char *f_in_pfi; + const char *f_in_pdd; + const char *f_out; + + /* special stuff needed to get additional arguments */ + char *arg1; + char **options; /* [STRING...] */ +} myargs; + +static struct argp argp = { + options: options, + parser: parse_opt, + args_doc: "pfifile", + doc: doc, + children: NULL, + help_filter: NULL, + argp_domain: NULL, +}; + +static error_t +parse_opt(int key, char *arg, struct argp_state *state) +{ + myargs *args = state->input; + + switch (key) { + /* common settings */ + case 'v': /* --verbose= */ + args->verbose = 1; + break; + + case 'c': /* --copyright */ + fprintf(stderr, "%s\n", copyright); + exit(0); + break; + + case 'j': /* --platform */ + args->f_in_pdd = arg; + break; + + case 'o': /* --output */ + args->f_out = arg; + break; + + case ARGP_KEY_ARG: + args->f_in_pfi = arg; + /* args->arg1 = arg; */ + args->options = &state->argv[state->next]; + state->next = state->argc; + break; + + case ARGP_KEY_END: + if (args->action == ACT_NOTHING) { + argp_usage(state); + exit(1); + } + break; + + default: + return(ARGP_ERR_UNKNOWN); + } + + return 0; +} + + +static size_t +byte_to_blk(size_t byte, size_t blk_size) +{ + return (byte % blk_size) == 0 + ? byte / blk_size + : byte / blk_size + 1; +} + + + + +/** + * @precondition IO: File stream points to first byte of RAW data. + * @postcondition IO: File stream points to first byte of next + * or EOF. + */ +static int +memorize_raw_eb(pfi_raw_t pfi_raw, pdd_data_t pdd, list_t *raw_pebs, + io_t io) +{ + int rc = 0; + uint32_t i; + + size_t read, to_read, eb_num; + size_t bytes_left; + list_t pebs = *raw_pebs; + peb_t peb = NULL; + + long old_file_pos = ftell(io->fp_pfi); + for (i = 0; i < pfi_raw->starts_size; i++) { + bytes_left = pfi_raw->data_size; + rc = fseek(io->fp_pfi, old_file_pos, SEEK_SET); + if (rc != 0) + goto err; + + eb_num = byte_to_blk(pfi_raw->starts[i], pdd->eb_size); + while (bytes_left) { + to_read = MIN(bytes_left, pdd->eb_size); + rc = peb_new(eb_num++, pdd->eb_size, &peb); + if (rc != 0) + goto err; + read = fread(peb->data, 1, to_read, io->fp_pfi); + if (read != to_read) { + rc = -EIO; + goto err; + } + pebs = append_elem(peb, pebs); + bytes_left -= read; + } + + } + *raw_pebs = pebs; + return 0; +err: + pebs = remove_all((free_func_t)&peb_free, pebs); + return rc; +} + +static int +convert_ubi_volume(pfi_ubi_t ubi, pdd_data_t pdd, list_t raw_pebs, + struct ubi_vol_tbl_record *vol_tab, + size_t *ebs_written, io_t io) +{ + int rc = 0; + uint32_t i, j; + peb_t raw_peb; + peb_t cmp_peb; + ubi_info_t u; + size_t leb_total = 0; + uint8_t vol_type; + + switch (ubi->type) { + case pfi_ubi_static: + vol_type = UBI_VID_STATIC; break; + case pfi_ubi_dynamic: + vol_type = UBI_VID_DYNAMIC; break; + default: + vol_type = UBI_VID_DYNAMIC; + } + + rc = peb_new(0, 0, &cmp_peb); + if (rc != 0) + goto err; + + long old_file_pos = ftell(io->fp_pfi); + for (i = 0; i < ubi->ids_size; i++) { + rc = fseek(io->fp_pfi, old_file_pos, SEEK_SET); + if (rc != 0) + goto err; + rc = ubigen_create(&u, ubi->ids[i], vol_type, + pdd->eb_size, DEFAULT_ERASE_COUNT, + ubi->alignment, UBI_VERSION, + pdd->vid_hdr_offset, 0, ubi->data_size, + io->fp_pfi, io->fp_out); + if (rc != 0) + goto err; + + rc = ubigen_get_leb_total(u, &leb_total); + if (rc != 0) + goto err; + + j = 0; + while(j < leb_total) { + cmp_peb->num = *ebs_written; + raw_peb = is_in((cmp_func_t)peb_cmp, cmp_peb, + raw_pebs); + if (raw_peb) { + rc = peb_write(io->fp_out, raw_peb); + } + else { + rc = ubigen_write_leb(u, NO_ERROR); + j++; + } + if (rc != 0) + goto err; + (*ebs_written)++; + } + /* memorize volume table entry */ + rc = ubigen_set_lvol_rec(u, ubi->size, + ubi->names[i], + (void*) &vol_tab[ubi->ids[i]]); + if (rc != 0) + goto err; + ubigen_destroy(&u); + } + + peb_free(&cmp_peb); + return 0; + +err: + peb_free(&cmp_peb); + ubigen_destroy(&u); + return rc; +} + + +static FILE* +my_fmemopen (void *buf, size_t size, const char *opentype) +{ + FILE* f; + + assert(strcmp(opentype, "r") == 0); + + f = tmpfile(); + fwrite(buf, 1, size, f); + rewind(f); + + return f; +} + +/** + * @brief Builds a UBI volume table from a volume entry list. + * @return 0 On success. + * else Error. + */ +static int +write_ubi_volume_table(pdd_data_t pdd, list_t raw_pebs, + struct ubi_vol_tbl_record *vol_tab, size_t vol_tab_size, + size_t *ebs_written, io_t io) +{ + int rc = 0; + ubi_info_t u; + peb_t raw_peb; + peb_t cmp_peb; + size_t leb_size, leb_total, j = 0; + uint8_t *ptr = NULL; + FILE* fp_leb = NULL; + + rc = peb_new(0, 0, &cmp_peb); + if (rc != 0) + goto err; + + /* @FIXME: Artem creates one volume with 2 LEBs. + * IMO 2 volumes would be more convenient. In order + * to get 2 reserved LEBs from ubigen, I have to + * introduce this stupid mechanism. Until no final + * decision of the VTAB structure is made... Good enough. + */ + rc = ubigen_create(&u, UBI_LAYOUT_VOL_ID, UBI_VID_DYNAMIC, + pdd->eb_size, DEFAULT_ERASE_COUNT, + 1, UBI_VERSION, + pdd->vid_hdr_offset, UBI_COMPAT_REJECT, + vol_tab_size, stdin, io->fp_out); + /* @FIXME stdin for fp_in is a hack */ + if (rc != 0) + goto err; + rc = ubigen_get_leb_size(u, &leb_size); + if (rc != 0) + goto err; + ubigen_destroy(&u); + + ptr = (uint8_t*) malloc(leb_size * sizeof(uint8_t)); + if (ptr == NULL) + goto err; + memset(ptr, 0xff, leb_size); + memcpy(ptr, vol_tab, vol_tab_size); + fp_leb = my_fmemopen(ptr, leb_size, "r"); + + rc = ubigen_create(&u, UBI_LAYOUT_VOL_ID, UBI_VID_DYNAMIC, + pdd->eb_size, DEFAULT_ERASE_COUNT, + 1, UBI_VERSION, pdd->vid_hdr_offset, + UBI_COMPAT_REJECT, leb_size * UBI_LAYOUT_VOLUME_EBS, + fp_leb, io->fp_out); + if (rc != 0) + goto err; + rc = ubigen_get_leb_total(u, &leb_total); + if (rc != 0) + goto err; + + long old_file_pos = ftell(fp_leb); + while(j < leb_total) { + rc = fseek(fp_leb, old_file_pos, SEEK_SET); + if (rc != 0) + goto err; + + cmp_peb->num = *ebs_written; + raw_peb = is_in((cmp_func_t)peb_cmp, cmp_peb, + raw_pebs); + if (raw_peb) { + rc = peb_write(io->fp_out, raw_peb); + } + else { + rc = ubigen_write_leb(u, NO_ERROR); + j++; + } + + if (rc != 0) + goto err; + (*ebs_written)++; + } + +err: + free(ptr); + peb_free(&cmp_peb); + ubigen_destroy(&u); + fclose(fp_leb); + return rc; +} + +static int +write_remaining_raw_ebs(pdd_data_t pdd, list_t raw_blocks, size_t *ebs_written, + FILE* fp_out) +{ + int rc = 0; + uint32_t j, delta; + list_t ptr; + peb_t empty_eb, peb; + + /* create an empty 0xff EB (for padding) */ + rc = peb_new(0, pdd->eb_size, &empty_eb); + + foreach(peb, ptr, raw_blocks) { + if (peb->num < *ebs_written) { + continue; /* omit blocks which + are already passed */ + } + + if (peb->num < *ebs_written) { + err_msg("eb_num: %d\n", peb->num); + err_msg("Bug: This should never happen. %d %s", + __LINE__, __FILE__); + goto err; + } + + delta = peb->num - *ebs_written; + if (((delta + *ebs_written) * pdd->eb_size) > pdd->flash_size) { + err_msg("RAW block outside of flash_size."); + goto err; + } + for (j = 0; j < delta; j++) { + rc = peb_write(fp_out, empty_eb); + if (rc != 0) + goto err; + (*ebs_written)++; + } + rc = peb_write(fp_out, peb); + if (rc != 0) + goto err; + (*ebs_written)++; + } + +err: + peb_free(&empty_eb); + return rc; +} + +static int +init_vol_tab(struct ubi_vol_tbl_record **vol_tab, size_t *vol_tab_size) +{ + uint32_t crc; + size_t i; + struct ubi_vol_tbl_record* res = NULL; + + *vol_tab_size = UBI_MAX_VOLUMES * UBI_VTBL_RECORD_SIZE; + + res = (struct ubi_vol_tbl_record*) calloc(1, *vol_tab_size); + if (vol_tab == NULL) { + return -ENOMEM; + } + + for (i = 0; i < UBI_MAX_VOLUMES; i++) { + crc = clc_crc32(crc32_table, UBI_CRC32_INIT, + &(res[i]), UBI_VTBL_RECORD_SIZE_CRC); + res[i].crc = cpu_to_ubi32(crc); + } + + *vol_tab = res; + return 0; +} + +static int +create_raw(io_t io) +{ + int rc = 0; + size_t ebs_written = 0; /* eraseblocks written already... */ + size_t vol_tab_size; + list_t ptr; + + list_t pfi_raws = mk_empty(); /* list of raw sections from a pfi */ + list_t pfi_ubis = mk_empty(); /* list of ubi sections from a pfi */ + list_t raw_pebs = mk_empty(); /* list of raw eraseblocks */ + + struct ubi_vol_tbl_record *vol_tab = NULL; + pdd_data_t pdd = NULL; + + rc = init_vol_tab (&vol_tab, &vol_tab_size); + if (rc != 0) { + err_msg("Cannot initialize volume table."); + goto err; + } + + rc = read_pdd_data(io->fp_pdd, &pdd, + err_buf, ERR_BUF_SIZE); + if (rc != 0) { + err_msg("Cannot read necessary pdd_data: %s rc: %d", + err_buf, rc); + goto err; + } + + rc = read_pfi_headers(&pfi_raws, &pfi_ubis, io->fp_pfi, + err_buf, ERR_BUF_SIZE); + if (rc != 0) { + err_msg("Cannot read pfi header: %s rc: %d", + err_buf, rc); + goto err; + } + + pfi_raw_t pfi_raw; + foreach(pfi_raw, ptr, pfi_raws) { + rc = memorize_raw_eb(pfi_raw, pdd, &raw_pebs, + io); + if (rc != 0) { + err_msg("Cannot create raw_block in mem. rc: %d\n", + rc); + goto err; + } + } + + pfi_ubi_t pfi_ubi; + foreach(pfi_ubi, ptr, pfi_ubis) { + rc = convert_ubi_volume(pfi_ubi, pdd, raw_pebs, + vol_tab, &ebs_written, io); + if (rc != 0) { + err_msg("Cannot convert UBI volume. rc: %d\n", rc); + goto err; + } + } + + rc = write_ubi_volume_table(pdd, raw_pebs, vol_tab, vol_tab_size, + &ebs_written, io); + if (rc != 0) { + err_msg("Cannot write UBI volume table. rc: %d\n", rc); + goto err; + } + + rc = write_remaining_raw_ebs(pdd, raw_pebs, &ebs_written, io->fp_out); + if (rc != 0) + goto err; + + if (io->fp_out != stdout) + info_msg("Physical eraseblocks written: %8d\n", ebs_written); +err: + free(vol_tab); + pfi_raws = remove_all((free_func_t)&free_pfi_raw, pfi_raws); + pfi_ubis = remove_all((free_func_t)&free_pfi_ubi, pfi_ubis); + raw_pebs = remove_all((free_func_t)&peb_free, raw_pebs); + free_pdd_data(&pdd); + return rc; +} + + +/* ------------------------------------------------------------------------- */ +static void +open_io_handle(myargs *args, io_t io) +{ + /* set PDD input */ + io->fp_pdd = fopen(args->f_in_pdd, "r"); + if (io->fp_pdd == NULL) { + err_sys("Cannot open: %s", args->f_in_pdd); + } + + /* set PFI input */ + io->fp_pfi = fopen(args->f_in_pfi, "r"); + if (io->fp_pfi == NULL) { + err_sys("Cannot open PFI input file: %s", args->f_in_pfi); + } + + /* set output prefix */ + if (strcmp(args->f_out,"") == 0) + io->fp_out = stdout; + else { + io->fp_out = fopen(args->f_out, "wb"); + if (io->fp_out == NULL) { + err_sys("Cannot open output file: %s", args->f_out); + } + } +} + +static void +close_io_handle(io_t io) +{ + if (fclose(io->fp_pdd) != 0) { + err_sys("Cannot close PDD file."); + } + if (fclose(io->fp_pfi) != 0) { + err_sys("Cannot close PFI file."); + } + if (io->fp_out != stdout) { + if (fclose(io->fp_out) != 0) { + err_sys("Cannot close output file."); + } + } + + io->fp_pdd = NULL; + io->fp_pfi = NULL; + io->fp_out = NULL; +} + +int +main(int argc, char *argv[]) +{ + int rc = 0; + + ubigen_init(); + init_crc32_table(crc32_table); + + struct io io = {NULL, NULL, NULL}; + myargs args = { + .action = ACT_RAW, + .verbose = 0, + + .f_in_pfi = "", + .f_in_pdd = "", + .f_out = "", + + /* arguments */ + .arg1 = NULL, + .options = NULL, + }; + + /* parse arguments */ + argp_parse(&argp, argc, argv, ARGP_IN_ORDER, 0, &args); + + if (strcmp(args.f_in_pfi, "") == 0) { + err_quit("No PFI input file specified!"); + } + + if (strcmp(args.f_in_pdd, "") == 0) { + err_quit("No PDD input file specified!"); + } + + open_io_handle(&args, &io); + + info_msg("[ Creating RAW..."); + rc = create_raw(&io); + if (rc != 0) { + err_msg("Creating RAW failed."); + goto err; + } + +err: + close_io_handle(&io); + if (rc != 0) { + remove(args.f_out); + } + + return rc; +} -- cgit v1.2.3