/* * CDDL HEADER START * * The contents of this file are subject to the terms of the * Common Development and Distribution License (the "License"). * You may not use this file except in compliance with the License. * * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE * or http://www.opensolaris.org/os/licensing. * See the License for the specific language governing permissions * and limitations under the License. * * When distributing Covered Code, include this CDDL HEADER in each * file and include the License file at usr/src/OPENSOLARIS.LICENSE. * If applicable, add the following below this CDDL HEADER, with the * fields enclosed by brackets "[]" replaced with your own identifying * information: Portions Copyright [yyyy] [name of copyright owner] * * CDDL HEADER END */ /* * Copyright 2007 Sun Microsystems, Inc. All rights reserved. * Use is subject to license terms. */ #pragma ident "%Z%%M% %I% %E% SMI" /* * Generate a cache of section header information for an ELF * object from the information found in its program headers. * * Malicious code can remove or corrupt section headers. The * resulting program will be difficult to analyze, but is still * runnable. Hence, scribbling on the section headers or removing * them is an effective form of obfuscation. On the other hand, * program headers must be accurate or the program will not run. * Section headers derived from them will necessarily lack information * found in the originals (particularly for non-allocable sections), * but will provide essential symbol information. The focus is on * recovering information that elfdump knows how to display, and that * might be interesting in a forensic situation. * * There are some things we don't attempt to create sections for: * * plt, got * We have no way to determine the length of either of * these sections from the information available via * the program headers or dynamic section. The data in * the PLT is of little use to elfdump. The data in the * GOT might be somewhat more interesting, especially as * it pertains to relocations. However, the sizing issue * remains. * * text, data, bss * Although we could create these, there is little value * to doing so. elfdump cannot display the arbitrary * data in these sections, so this would amount to a * simple repetition of the information already displayed * in the program headers, with no additional benefit. */ #include #include #include #include #include #include #include #include #include #include <_elfdump.h> /* * Common information about the object that is needed by * all the routines in this module. */ typedef struct { const char *file; int fd; Ehdr *ehdr; Phdr *phdr; size_t phnum; } FSTATE; /* * These values uniquely identify the sections that we know * how to recover. * * Note: We write the sections to the cache array in this same order. * It simplifies this code if the dynamic, dynstr, dynsym, and ldynsym * sections occupy known slots in the cache array. Other sections reference * them by index, and if they are at a known spot, there is no need * for a fixup pass. Putting them in positions [1-4] solves this. * * The order they are in was chosen such that if any one of them exists, * all of the ones before it must also exist. This means that if the * desired section exists, it will end up in the desired index in the * cache array. * * The order of the other sections is arbitrary. I've arranged them * in roughly related groups. */ typedef enum { SINFO_T_NULL = 0, SINFO_T_DYN = 1, SINFO_T_DYNSTR = 2, SINFO_T_DYNSYM = 3, SINFO_T_LDYNSYM = 4, SINFO_T_HASH = 5, SINFO_T_SYMINFO = 6, SINFO_T_SYMSORT = 7, SINFO_T_TLSSORT = 8, SINFO_T_VERNEED = 9, SINFO_T_VERDEF = 10, SINFO_T_VERSYM = 11, SINFO_T_INTERP = 12, SINFO_T_CAP = 13, SINFO_T_UNWIND = 14, SINFO_T_MOVE = 15, SINFO_T_REL = 16, SINFO_T_RELA = 17, SINFO_T_PREINITARR = 18, SINFO_T_INITARR = 19, SINFO_T_FINIARR = 20, SINFO_T_NOTE = 21, SINFO_T_NUM = 22 /* Count of items. Must come last */ } SINFO_TYPE; /* * Table of per-section constant data used to set up the section * header cache and the various sub-parts it references. Indexed by * SINFO_T value. * * note: The sh_flags value should be either SHF_ALLOC, or 0. * get_data() sets SHF_WRITE if the program header containing the * section is writable. The other flags require information that * the program headers don't contain (i.e. SHF_STRINGS, etc) so * we don't set them. */ typedef struct { const char *name; Word sh_type; Word sh_flags; Word sh_addralign; Word sh_entsize; Elf_Type libelf_type; } SINFO_DATA; static SINFO_DATA sinfo_data[SINFO_T_NUM] = { /* SINFO_T_NULL */ { 0 }, /* SINFO_T_DYN */ { MSG_ORIG(MSG_PHDRNAM_DYN), SHT_DYNAMIC, SHF_ALLOC, M_WORD_ALIGN, sizeof (Dyn), ELF_T_DYN }, /* SINFO_T_DYNSTR */ { MSG_ORIG(MSG_PHDRNAM_DYNSTR), SHT_STRTAB, SHF_ALLOC, 1, 0, ELF_T_BYTE }, /* SINFO_T_DYNSYM */ { MSG_ORIG(MSG_PHDRNAM_DYNSYM), SHT_DYNSYM, SHF_ALLOC, M_WORD_ALIGN, sizeof (Sym), ELF_T_SYM }, /* SINFO_T_LDYNSYM */ { MSG_ORIG(MSG_PHDRNAM_LDYNSYM), SHT_SUNW_LDYNSYM, SHF_ALLOC, M_WORD_ALIGN, sizeof (Sym), ELF_T_SYM }, /* SINFO_T_HASH */ { MSG_ORIG(MSG_PHDRNAM_HASH), SHT_HASH, SHF_ALLOC, M_WORD_ALIGN, sizeof (Word), ELF_T_WORD }, /* SINFO_T_SYMINFO */ { MSG_ORIG(MSG_PHDRNAM_SYMINFO), SHT_SUNW_syminfo, SHF_ALLOC, M_WORD_ALIGN, sizeof (Syminfo), ELF_T_SYMINFO }, /* SINFO_T_SYMSORT */ { MSG_ORIG(MSG_PHDRNAM_SYMSORT), SHT_SUNW_symsort, SHF_ALLOC, M_WORD_ALIGN, sizeof (Word), ELF_T_WORD }, /* SINFO_T_TLSSORT */ { MSG_ORIG(MSG_PHDRNAM_TLSSORT), SHT_SUNW_tlssort, SHF_ALLOC, M_WORD_ALIGN, sizeof (Word), ELF_T_WORD }, /* SINFO_T_VERNEED */ { MSG_ORIG(MSG_PHDRNAM_VER), SHT_SUNW_verneed, SHF_ALLOC, M_WORD_ALIGN, 1, ELF_T_VNEED }, /* SINFO_T_VERDEF */ { MSG_ORIG(MSG_PHDRNAM_VER), SHT_SUNW_verdef, SHF_ALLOC, M_WORD_ALIGN, 1, ELF_T_VDEF }, /* SINFO_T_VERSYM */ { MSG_ORIG(MSG_PHDRNAM_VER), SHT_SUNW_versym, SHF_ALLOC, M_WORD_ALIGN, sizeof (Versym), ELF_T_HALF }, /* SINFO_T_INTERP */ { MSG_ORIG(MSG_PHDRNAM_INTERP), SHT_PROGBITS, SHF_ALLOC, 1, 0, ELF_T_BYTE }, /* SINFO_T_CAP */ { MSG_ORIG(MSG_PHDRNAM_CAP), SHT_SUNW_cap, SHF_ALLOC, sizeof (Addr), sizeof (Cap), ELF_T_CAP }, /* SINFO_T_UNWIND */ { MSG_ORIG(MSG_PHDRNAM_UNWIND), SHT_AMD64_UNWIND, SHF_ALLOC, sizeof (Addr), 0, ELF_T_BYTE }, /* SINFO_T_MOVE */ { MSG_ORIG(MSG_PHDRNAM_MOVE), SHT_SUNW_move, SHF_ALLOC, sizeof (Lword), sizeof (Move), ELF_T_MOVE }, /* SINFO_T_REL */ { MSG_ORIG(MSG_PHDRNAM_REL), SHT_REL, SHF_ALLOC, M_WORD_ALIGN, sizeof (Rel), ELF_T_REL }, /* SINFO_T_RELA */ { MSG_ORIG(MSG_PHDRNAM_RELA), SHT_RELA, SHF_ALLOC, M_WORD_ALIGN, sizeof (Rela), ELF_T_RELA }, /* SINFO_T_PREINITARR */ { MSG_ORIG(MSG_PHDRNAM_PREINITARR), SHT_PREINIT_ARRAY, SHF_ALLOC, sizeof (Addr), sizeof (Addr), ELF_T_ADDR }, /* SINFO_T_INITARR */ { MSG_ORIG(MSG_PHDRNAM_INITARR), SHT_INIT_ARRAY, SHF_ALLOC, sizeof (Addr), sizeof (Addr), ELF_T_ADDR }, /* SINFO_T_FINIARR */ { MSG_ORIG(MSG_PHDRNAM_FINIARR), SHT_FINI_ARRAY, SHF_ALLOC, sizeof (Addr), sizeof (Addr), ELF_T_ADDR }, /* SINFO_T_NOTE */ { MSG_ORIG(MSG_PHDRNAM_NOTE), SHT_NOTE, 0, M_WORD_ALIGN, 1, ELF_T_NOTE } }; /* * As we read program headers and dynamic elements, we build up * the data for our fake section headers in variables of the * SINFO type. SINFO is used to track the sections that can only * appear a fixed number of times (usually once). * * SINFO_LISTELT is used for sections that can occur an arbitrary * number of times. They are kept in a doubly linked circular * buffer. */ typedef struct { SINFO_TYPE type; /* Our type code for the section */ Addr vaddr; /* Virtual memory address */ Off offset; /* File offset of data. Ignored unless */ /* vaddr is 0. Used by program headers */ size_t size; /* # bytes in section */ size_t vercnt; /* Used by verdef and verneed to hold count */ Shdr *shdr; /* Constructed shdr */ Elf_Data *data; /* Constructed data descriptor */ } SINFO; typedef struct _sinfo_listelt { struct _sinfo_listelt *next; struct _sinfo_listelt *prev; SINFO sinfo; } SINFO_LISTELT; /* * Free dynamic memory used by SINFO structures. * * entry: * sinfo - Address of first SINFO structure to free * n - # of structures to clear * * exit: * For each SINFO struct, the section header, data descriptor, * and data buffer are freed if non-NULL. The relevant * fields are set to NULL, and the type is set to SINFO_T_NULL. */ static void sinfo_free(SINFO *sinfo, size_t n) { for (; n-- > 0; sinfo++) { if (sinfo->data != NULL) { if (sinfo->data->d_buf != NULL) free(sinfo->data->d_buf); free(sinfo->data); sinfo->data = NULL; } if (sinfo->shdr) { free(sinfo->shdr); sinfo->shdr = NULL; } sinfo->type = SINFO_T_NULL; } } /* * Allocate a new SINFO_LISTELT and put it at the end of the * doubly linked list anchored by the given list root node. * * On success, a new node has been put at the end of the circular * doubly linked list, and a pointer to the SINFO sub-structure is * returned. On failure, an error is printed, and NULL is returned. */ static SINFO * sinfo_list_alloc(FSTATE *fstate, SINFO_LISTELT *root) { SINFO_LISTELT *elt; if ((elt = malloc(sizeof (*elt))) == NULL) { int err = errno; (void) fprintf(stderr, MSG_INTL(MSG_ERR_MALLOC), fstate->file, strerror(err)); return (0); } elt->next = root; elt->prev = root->prev; root->prev = elt; elt->prev->next = elt; bzero(&elt->sinfo, sizeof (elt->sinfo)); return (&elt->sinfo); } /* * Release the memory used by the given list, restoring it to * an empty list. */ static void sinfo_list_free_all(SINFO_LISTELT *root) { SINFO_LISTELT *elt; for (elt = root->next; elt != root; elt = elt->next) sinfo_free(&elt->sinfo, 1); root->next = root->prev = root; } /* * Given a virtual address and desired size of the data to be found * at that address, look through the program headers for the PT_LOAD * segment that contains it and return the offset within the ELF file * at which it resides. * * entry: * fstate - Object state * addr - virtual address to be translated * size - Size of the data to be found at that address, in bytes * zero_bytes - NULL, or address to receive the number of data * bytes at the end of the data that are not contained * in the file, and which must be zero filled by the caller. * If zero_bytes is NULL, the file must contain all of the * desired data. If zero_bytes is not NULL, then the program * header must reserve the space for all of the data (p_memsz) * but it is acceptable for only part of the data to be in * the file (p_filesz). *zero_bytes is set to the difference * in size, and is the number of bytes the caller must * set to 0 rather than reading from the file. * phdr_ret - NULL, or address of variable to receive pointer * to program header that contains offset. * exit: * On success: If zero_bytes is non-NULL, it is updated. If phdr_ret * is non-NULL, it is updated. The file offset is returned. * * On failure, 0 is returned. Since any ELF file we can understand * must start with an ELF magic number, 0 cannot be a valid file * offset for a virtual address, and is therefore unambiguous as * a failure indication. */ static Off map_addr_to_offset(FSTATE *fstate, Addr addr, size_t size, size_t *zero_bytes, Phdr **phdr_ret) { Off offset; Addr end_addr = addr + size; size_t avail_file; Phdr *phdr = fstate->phdr; size_t phnum = fstate->phnum; for (; phnum--; phdr++) { if (phdr->p_type != PT_LOAD) continue; if ((addr >= phdr->p_vaddr) && (end_addr <= (phdr->p_vaddr + phdr->p_memsz))) { /* * Subtract segment virtual address, leaving the * offset relative to the segment (not the file). */ offset = addr - phdr->p_vaddr; avail_file = phdr->p_filesz - offset; /* * The addr/size are in bounds for this segment. * Is there enough data in the file to satisfy * the request? If zero_bytes is NULL, it must * all be in the file. Otherwise it can be * zero filled. */ if (zero_bytes == NULL) { if (size > avail_file) continue; } else { *zero_bytes = (size > avail_file) ? (size - avail_file) : 0; } if (phdr_ret != NULL) *phdr_ret = phdr; /* Add segment file offset, giving overall offset */ return (phdr->p_offset + offset); } } /* If we get here, the mapping failed */ return (0); } /* * This routine is the same thing as map_addr_to_offset(), except that * it goes the other way, mapping from offset to virtual address. * * The comments for map_addr_to_offset() are applicable if you * reverse offset and address. */ static Addr map_offset_to_addr(FSTATE *fstate, Off offset, size_t size, size_t *zero_bytes, Phdr **phdr_ret) { Off end_offset = offset + size; size_t avail_file; Phdr *phdr = fstate->phdr; size_t phnum = fstate->phnum; for (; phnum--; phdr++) { if (phdr->p_type != PT_LOAD) continue; if ((offset >= phdr->p_offset) && (end_offset <= (phdr->p_offset + phdr->p_memsz))) { /* * Subtract segment offset, leaving the * offset relative to the segment (not the file). */ offset -= phdr->p_offset; avail_file = phdr->p_filesz - offset; /* * The offset/size are in bounds for this segment. * Is there enough data in the file to satisfy * the request? If zero_bytes is NULL, it must * all be in the file. Otherwise it can be * zero filled. */ if (zero_bytes == NULL) { if (size > avail_file) continue; } else { *zero_bytes = (size > avail_file) ? (size - avail_file) : 0; } if (phdr_ret != NULL) *phdr_ret = phdr; /* Add segment virtual address, giving overall addr */ return (phdr->p_vaddr + offset); } } /* If we get here, the mapping failed */ return (0); } /* * Use elf_xlatetom() to convert the bytes in buf from their * in-file representation to their in-memory representation. * * Returns True(1) for success. On failure, an error message is printed * and False(0) is returned. */ static int xlate_data(FSTATE *fstate, void *buf, size_t nbyte, Elf_Type xlate_type) { Elf_Data data; data.d_type = xlate_type; data.d_size = nbyte; data.d_off = 0; data.d_align = 0; data.d_version = fstate->ehdr->e_version; data.d_buf = buf; if (elf_xlatetom(&data, &data, fstate->ehdr->e_ident[EI_DATA]) == NULL) { failure(fstate->file, MSG_ORIG(MSG_ELF_XLATETOM)); return (0); } return (1); } /* * Read nbytes of data into buf, starting at the specified offset * within the ELF file. * * entry: * fstate - Object state * offset - Offset within the file at which desired data resides. * buf - Buffer to receive the data * nbyte - # of bytes to read into buf * xlate_type - An ELF xlate type, specifying the type of data * being input. If xlate_type is ELF_T_BYTE, xlate is not * done. Otherwise, xlate_data() is called to convert the * data into its in-memory representation. * exit: * On success, the data has been written into buf, xlate_data() * called on it if required, and True(1) is returned. Otherwise * False(0) is returned. * * note: * This routine does not move the file pointer. */ static int read_data(FSTATE *fstate, Off offset, void *buf, size_t nbyte, Elf_Type xlate_type) { if (pread(fstate->fd, buf, nbyte, offset) != nbyte) { int err = errno; (void) fprintf(stderr, MSG_INTL(MSG_ERR_READ), fstate->file, strerror(err)); return (0); } if (xlate_type != ELF_T_BYTE) return (xlate_data(fstate, buf, nbyte, xlate_type)); return (1); } /* * Read the hash nbucket/nchain values from the start of the hash * table found at the given virtual address in the mapped ELF object. * * On success, *nbucket, and *nchain have been filled in with their * values, *total contains the number of elements in the hash table, * and this routine returns True (1). * * On failure, False (0) is returned. */ static int hash_size(FSTATE *fstate, SINFO *hash_sinfo, Word *nbucket, Word *nchain, size_t *total) { Off offset; Word buf[2]; offset = map_addr_to_offset(fstate, hash_sinfo->vaddr, sizeof (buf), NULL, NULL); if (offset == 0) return (0); if (read_data(fstate, offset, buf, sizeof (buf), ELF_T_WORD) == 0) return (0); *nbucket = buf[0]; *nchain = buf[1]; *total = 2 + *nbucket + *nchain; return (1); } /* * Read a Verdef structure at the specified file offset and return * its vd_cnt, vd_aux, and vd_next fields. */ static int read_verdef(FSTATE *fstate, Off offset, Half *cnt, Word *aux, Word *next) { Verdef verdef; if (read_data(fstate, offset, &verdef, sizeof (verdef), ELF_T_BYTE) == 0) return (0); /* xlate vd_cnt */ if (xlate_data(fstate, &verdef.vd_cnt, sizeof (verdef.vd_cnt), ELF_T_HALF) == 0) return (0); /* * xlate vd_aux and vd_next. These items are adjacent and are * both Words, so they can be handled in a single operation. */ if (xlate_data(fstate, &verdef.vd_aux, 2 * sizeof (Word), ELF_T_WORD) == 0) return (0); *cnt = verdef.vd_cnt; *aux = verdef.vd_aux; *next = verdef.vd_next; return (1); } /* * Read a Verdaux structure at the specified file offset and return * its vda_next field. */ static int read_verdaux(FSTATE *fstate, Off offset, Word *next) { Verdaux verdaux; if (read_data(fstate, offset, &verdaux, sizeof (verdaux), ELF_T_BYTE) == 0) return (0); /* xlate vda_next */ if (xlate_data(fstate, &verdaux.vda_next, sizeof (verdaux.vda_next), ELF_T_WORD) == 0) return (0); *next = verdaux.vda_next; return (1); } /* * Read a Verneed structure at the specified file offset and return * its vn_cnt, vn_aux, and vn_next fields. */ static int read_verneed(FSTATE *fstate, Off offset, Half *cnt, Word *aux, Word *next) { Verneed verneed; if (read_data(fstate, offset, &verneed, sizeof (verneed), ELF_T_BYTE) == 0) return (0); /* xlate vn_cnt */ if (xlate_data(fstate, &verneed.vn_cnt, sizeof (verneed.vn_cnt), ELF_T_HALF) == 0) return (0); /* * xlate vn_aux and vn_next. These items are adjacent and are * both Words, so they can be handled in a single operation. */ if (xlate_data(fstate, &verneed.vn_aux, 2 * sizeof (Word), ELF_T_WORD) == 0) return (0); *cnt = verneed.vn_cnt; *aux = verneed.vn_aux; *next = verneed.vn_next; return (1); } /* * Read a Vernaux structure at the specified file offset and return * its vna_next field. */ static int read_vernaux(FSTATE *fstate, Off offset, Word *next) { Vernaux vernaux; if (read_data(fstate, offset, &vernaux, sizeof (vernaux), ELF_T_BYTE) == 0) return (0); /* xlate vna_next */ if (xlate_data(fstate, &vernaux.vna_next, sizeof (vernaux.vna_next), ELF_T_WORD) == 0) return (0); *next = vernaux.vna_next; return (1); } /* * Compute the size of Verdef and Verneed sections. Both of these * sections are made up of interleaved main nodes (Verdef and Verneed) * and auxiliary blocks (Verdaux and Vernaux). These nodes refer to * each other by relative offsets. The linker has a lot of flexibility * in how it lays out these items, and we cannot assume a standard * layout. To determine the size of the section, we must read each * main node and compute the high water mark of the memory it and its * auxiliary structs access. * * Although Verdef/Verdaux and Verneed/Vernaux are different types, * their logical organization is the same. Each main block has * a cnt field that tells how many auxiliary blocks it has, an * aux field that gives the offset of the first auxiliary block, and * an offset to the next main block. Each auxiliary block contains * an offset to the next auxiliary block. By breaking the type specific * code into separate sub-functions, we can process both Verdef and * sections Verdaux from a single routine. * * entry: * fstate - Object state * sec - Section to be processed (SINFO_T_VERDEF or SINFO_T_VERNEED). * * exit: * On success, sec->size is set to the section size in bytes, and * True (1) is returned. On failure, False (0) is returned. */ static int verdefneed_size(FSTATE *fstate, SINFO *sec) { int (* read_main)(FSTATE *, Off, Half *, Word *, Word *); int (* read_aux)(FSTATE *, Off, Word *); size_t size_main, size_aux; Off offset, aux_offset; Off highwater, extent; size_t num_main = sec->vercnt; Half v_cnt; Word v_aux, v_next, va_next; /* * Set up the function pointers to the type-specific code * for fetching data from the main and auxiliary blocks. */ if (sec->type == SINFO_T_VERDEF) { read_main = read_verdef; read_aux = read_verdaux; size_main = sizeof (Verdef); size_aux = sizeof (Verdaux); } else { /* SINFO_T_VERNEED */ read_main = read_verneed; read_aux = read_vernaux; size_main = sizeof (Verneed); size_aux = sizeof (Vernaux); } /* * Map starting address to file offset. Save the starting offset * in the SINFO size field. Once we have the high water offset, we * can subtract this from it to get the size. * * Note: The size argument set here is a lower bound --- the * size of the main blocks without any auxiliary ones. It's * the best we can do until the size has been determined for real. */ offset = highwater = map_addr_to_offset(fstate, sec->vaddr, size_main * num_main, NULL, NULL); if (offset == 0) return (0); sec->size = offset; for (; num_main-- > 0; offset += v_next) { /* Does this move the high water mark up? */ extent = offset + size_main; if (extent > highwater) highwater = extent; if ((*read_main)(fstate, offset, &v_cnt, &v_aux, &v_next) == 0) return (0); /* * If there are auxiliary structures referenced, * check their position to see if it pushes * the high water mark. */ aux_offset = offset + v_aux; for (; v_cnt-- > 0; aux_offset += va_next) { extent = aux_offset + size_aux; if (extent > highwater) highwater = extent; if ((*read_aux)(fstate, aux_offset, &va_next) == 0) return (0); } } sec->size = highwater - sec->size; return (1); } /* * Allocate and fill in a fake section header, data descriptor, * and data buffer for the given section. Fill them in and read * the associated data into the buffer. * * entry: * fstate - Object state * sec - Section information * * exit: * On success, the actions described above are complete, and * True (1) is returned. * * On failure, an error is reported, all resources used by sec * are released, and sec->type is set to SINFO_T_NULL, effectively * eliminating its contents from any further use. False (0) is * returned. */ static int get_data(FSTATE *fstate, SINFO *sec) { SINFO_DATA *tinfo; size_t read_bytes, zero_bytes; Phdr *phdr = NULL; /* * If this is a NULL section, or if we've already processed * this item, then we are already done. */ if ((sec->type == SINFO_T_NULL) || (sec->shdr != NULL)) return (1); if (((sec->shdr = malloc(sizeof (*sec->shdr))) == NULL) || ((sec->data = malloc(sizeof (*sec->data))) == NULL)) { int err = errno; sinfo_free(sec, 1); (void) fprintf(stderr, MSG_INTL(MSG_ERR_MALLOC), fstate->file, strerror(err)); return (0); } tinfo = &sinfo_data[sec->type]; /* * Fill in fake section header * * sh_name should be the offset of the name in the shstrtab * section referenced by the ELF header. There is no * value to elfdump in creating shstrtab, so we set * sh_name to 0, knowing that elfdump doesn't look at it. */ sec->shdr->sh_name = 0; sec->shdr->sh_type = tinfo->sh_type; sec->shdr->sh_flags = tinfo->sh_flags; if ((tinfo->sh_flags & SHF_ALLOC) == 0) { /* * Non-allocable section: Pass the addr (which is probably * 0) and offset through without inspection. */ sec->shdr->sh_addr = sec->vaddr; sec->shdr->sh_offset = sec->offset; zero_bytes = 0; } else if (sec->vaddr == 0) { /* * Allocable section with a 0 vaddr. Figure out the * real address by mapping the offset to it using the * program headers. */ sec->shdr->sh_addr = map_offset_to_addr(fstate, sec->offset, sec->size, &zero_bytes, &phdr); sec->shdr->sh_offset = sec->offset; } else { /* * Allocable section with non-0 vaddr. Use the vaddr * to derive the offset. */ sec->shdr->sh_addr = sec->vaddr; sec->shdr->sh_offset = map_addr_to_offset(fstate, sec->vaddr, sec->size, &zero_bytes, &phdr); } if (sec->shdr->sh_offset == 0) { sinfo_free(sec, 1); return (0); } /* * If the program header has its write flags set, then set * the section write flag. */ if (phdr && ((phdr->p_flags & PF_W) != 0)) sec->shdr->sh_flags |= SHF_WRITE; sec->shdr->sh_size = sec->size; sec->shdr->sh_link = 0; sec->shdr->sh_info = 0; sec->shdr->sh_addralign = tinfo->sh_addralign; sec->shdr->sh_entsize = tinfo->sh_entsize; /* * Some sections define special meanings for sh_link and sh_info. */ switch (tinfo->sh_type) { case SHT_DYNAMIC: sec->shdr->sh_link = SINFO_T_DYNSTR; break; case SHT_DYNSYM: sec->shdr->sh_link = SINFO_T_DYNSTR; sec->shdr->sh_info = 1; /* First global symbol */ break; case SHT_SUNW_LDYNSYM: sec->shdr->sh_link = SINFO_T_DYNSTR; /* * ldynsym is all local symbols, so the index of the * first global is equivalent to the number of symbols. */ sec->shdr->sh_info = sec->shdr->sh_size / sizeof (Sym); break; case SHT_HASH: case SHT_SUNW_move: case SHT_REL: case SHT_RELA: case SHT_SUNW_versym: sec->shdr->sh_link = SINFO_T_DYNSYM; break; case SHT_SUNW_verdef: case SHT_SUNW_verneed: sec->shdr->sh_link = SINFO_T_DYNSTR; sec->shdr->sh_info = sec->vercnt; break; case SHT_SUNW_syminfo: sec->shdr->sh_link = SINFO_T_DYNSYM; sec->shdr->sh_info = SINFO_T_DYN; break; case SHT_SUNW_symsort: case SHT_SUNW_tlssort: sec->shdr->sh_link = SINFO_T_LDYNSYM; break; } /* Fill in fake Elf_Data descriptor */ sec->data->d_type = tinfo->libelf_type; sec->data->d_size = sec->size; sec->data->d_off = 0; sec->data->d_align = tinfo->sh_addralign; sec->data->d_version = fstate->ehdr->e_version; if (sec->size == 0) { sec->data->d_buf = NULL; return (1); } if ((sec->data->d_buf = malloc(sec->size)) == NULL) { int err = errno; sinfo_free(sec, 1); (void) fprintf(stderr, MSG_INTL(MSG_ERR_MALLOC), fstate->file, strerror(err)); return (0); } read_bytes = sec->size - zero_bytes; if ((read_bytes > 0) && (read_data(fstate, sec->shdr->sh_offset, sec->data->d_buf, read_bytes, ELF_T_BYTE) == 0)) { sinfo_free(sec, 1); return (0); } if (zero_bytes > 0) bzero(read_bytes + (char *)sec->data->d_buf, zero_bytes); if ((tinfo->libelf_type != ELF_T_BYTE) && (elf_xlatetom(sec->data, sec->data, fstate->ehdr->e_ident[EI_DATA]) == NULL)) { sinfo_free(sec, 1); failure(fstate->file, MSG_ORIG(MSG_ELF_XLATETOM)); return (0); } return (1); } /* * Generate a section header cache made up of information derived * from the program headers. * * entry: * file - Name of object * fd - Open file handle for object * elf - ELF descriptor * ehdr - Elf header * cache, shnum - Addresses of variables to receive resulting * cache and number of sections. * * exit: * On success, *cache and *shnum are set, and True (1) is returned. * On failure, False (0) is returned. * * note: * The cache returned by this routine must be freed using * fake_shdr_cache_free(), and not by a direct call to free(). * Otherwise, memory will leak. */ int fake_shdr_cache(const char *file, int fd, Elf *elf, Ehdr *ehdr, Cache **cache, size_t *shnum) { /* * The C language guarantees that a structure of homogeneous * items will receive exactly the same layout in a structure * as a plain array of the same type. Hence, this structure, which * gives us by-name or by-index access to the various section * info descriptors we maintain. * * We use this for sections where * - Only one instance is allowed * - We need to be able to access them easily by * name (for instance, when mining the .dynamic * section for information to build them up. * * NOTE: These fields must be in the same order as the * SINFO_T_ type codes that correspond to them. Otherwise, * they will end up in the wrong order in the cache array, * and the sh_link/sh_info fields may be wrong. */ struct { /* Note: No entry is needed for SINFO_T_NULL */ SINFO dyn; SINFO dynstr; SINFO dynsym; SINFO ldynsym; SINFO hash; SINFO syminfo; SINFO symsort; SINFO tlssort; SINFO verneed; SINFO verdef; SINFO versym; SINFO interp; SINFO cap; SINFO unwind; SINFO move; SINFO rel; SINFO rela; SINFO preinitarr; SINFO initarr; SINFO finiarr; } sec; static const size_t sinfo_n = sizeof (sec) / sizeof (sec.dyn); SINFO *secarr = (SINFO *) &sec; /* * Doubly linked circular list, used to track sections * where multiple sections of a given type can exist. * seclist is the root of the list. Its sinfo field is not * used --- it serves to anchor the root of the list, allowing * rapid access to the first and last element in the list. */ SINFO_LISTELT seclist; FSTATE fstate; size_t ndx; size_t num_sinfo, num_list_sinfo; SINFO *sinfo; SINFO_LISTELT *sinfo_list; Cache *_cache; fstate.file = file; fstate.fd = fd; fstate.ehdr = ehdr; if (elf_getphnum(elf, &fstate.phnum) == 0) { failure(file, MSG_ORIG(MSG_ELF_GETPHNUM)); return (0); } if ((fstate.phdr = elf_getphdr(elf)) == NULL) { failure(file, MSG_ORIG(MSG_ELF_GETPHDR)); return (0); } bzero(&sec, sizeof (sec)); /* Initialize "by-name" sec info */ seclist.next = seclist.prev = &seclist; /* Empty circular list */ /* * Go through the program headers and look for information * we can use to synthesize section headers. By far the most * valuable thing is a dynamic section, the contents of * which point at all sections used by ld.so.1. */ for (ndx = 0; ndx < fstate.phnum; ndx++) { /* * A program header with no file size does * not have a backing section. */ if (fstate.phdr[ndx].p_filesz == 0) continue; switch (fstate.phdr[ndx].p_type) { default: /* Header we can't use. Move on to next one */ continue; case PT_DYNAMIC: sec.dyn.type = SINFO_T_DYN; sinfo = &sec.dyn; break; case PT_INTERP: sec.interp.type = SINFO_T_INTERP; sinfo = &sec.interp; break; case PT_NOTE: if ((sinfo = sinfo_list_alloc(&fstate, &seclist)) == NULL) continue; sinfo->type = SINFO_T_NOTE; break; case PT_SUNW_UNWIND: sec.unwind.type = SINFO_T_UNWIND; sinfo = &sec.unwind; break; case PT_SUNWCAP: sec.cap.type = SINFO_T_CAP; sinfo = &sec.cap; break; } /* * Capture the position/extent information for * the header in the SINFO struct set up by the * switch statement above. */ sinfo->vaddr = fstate.phdr[ndx].p_vaddr; sinfo->offset = fstate.phdr[ndx].p_offset; sinfo->size = fstate.phdr[ndx].p_filesz; } /* * If we found a dynamic section, look through it and * gather information about the sections it references. */ if (sec.dyn.type == SINFO_T_DYN) (void) get_data(&fstate, &sec.dyn); if ((sec.dyn.type == SINFO_T_DYN) && (sec.dyn.data->d_buf != NULL)) { Dyn *dyn; for (dyn = sec.dyn.data->d_buf; dyn->d_tag != DT_NULL; dyn++) { switch (dyn->d_tag) { case DT_HASH: sec.hash.type = SINFO_T_HASH; sec.hash.vaddr = dyn->d_un.d_ptr; break; case DT_STRTAB: sec.dynstr.type = SINFO_T_DYNSTR; sec.dynstr.vaddr = dyn->d_un.d_ptr; break; case DT_SYMTAB: sec.dynsym.type = SINFO_T_DYNSYM; sec.dynsym.vaddr = dyn->d_un.d_ptr; break; case DT_RELA: sec.rela.type = SINFO_T_RELA; sec.rela.vaddr = dyn->d_un.d_ptr; break; case DT_RELASZ: sec.rela.size = dyn->d_un.d_val; break; case DT_STRSZ: sec.dynstr.size = dyn->d_un.d_val; break; case DT_REL: sec.rel.type = SINFO_T_REL; sec.rel.vaddr = dyn->d_un.d_ptr; break; case DT_RELSZ: sec.rel.size = dyn->d_un.d_val; break; case DT_INIT_ARRAY: sec.initarr.type = SINFO_T_INITARR; sec.initarr.vaddr = dyn->d_un.d_ptr; break; case DT_INIT_ARRAYSZ: sec.initarr.size = dyn->d_un.d_val; break; case DT_FINI_ARRAY: sec.finiarr.type = SINFO_T_FINIARR; sec.finiarr.vaddr = dyn->d_un.d_ptr; break; case DT_FINI_ARRAYSZ: sec.finiarr.size = dyn->d_un.d_val; break; case DT_PREINIT_ARRAY: sec.preinitarr.type = SINFO_T_PREINITARR; sec.preinitarr.vaddr = dyn->d_un.d_ptr; break; case DT_PREINIT_ARRAYSZ: sec.preinitarr.size = dyn->d_un.d_val; break; case DT_SUNW_SYMTAB: sec.ldynsym.type = SINFO_T_LDYNSYM; sec.ldynsym.vaddr = dyn->d_un.d_ptr; break; case DT_SUNW_SYMSZ: sec.ldynsym.size = dyn->d_un.d_val; break; case DT_SUNW_SYMSORT: sec.symsort.type = SINFO_T_SYMSORT; sec.symsort.vaddr = dyn->d_un.d_ptr; break; case DT_SUNW_SYMSORTSZ: sec.symsort.size = dyn->d_un.d_val; break; case DT_SUNW_TLSSORT: sec.tlssort.type = SINFO_T_TLSSORT; sec.tlssort.vaddr = dyn->d_un.d_ptr; break; case DT_SUNW_TLSSORTSZ: sec.tlssort.size = dyn->d_un.d_val; break; case DT_MOVETAB: sec.move.type = SINFO_T_MOVE; sec.move.vaddr = dyn->d_un.d_ptr; break; case DT_MOVESZ: sec.move.size = dyn->d_un.d_val; break; case DT_SYMINFO: sec.syminfo.type = SINFO_T_SYMINFO; sec.syminfo.vaddr = dyn->d_un.d_ptr; break; case DT_SYMINSZ: sec.syminfo.size = dyn->d_un.d_val; break; case DT_VERSYM: sec.versym.type = SINFO_T_VERSYM; sec.versym.vaddr = dyn->d_un.d_ptr; break; case DT_VERDEF: sec.verdef.type = SINFO_T_VERDEF; sec.verdef.vaddr = dyn->d_un.d_ptr; break; case DT_VERDEFNUM: sec.verdef.vercnt = dyn->d_un.d_val; sec.verdef.size = sizeof (Verdef) * dyn->d_un.d_val; break; case DT_VERNEED: sec.verneed.type = SINFO_T_VERNEED; sec.verneed.vaddr = dyn->d_un.d_ptr; break; case DT_VERNEEDNUM: sec.verneed.vercnt = dyn->d_un.d_val; sec.verneed.size = sizeof (Verneed) * dyn->d_un.d_val; break; } } } /* * Different sections depend on each other, and are meaningless * without them. For instance, even if a .dynsym exists, * no use can be made of it without a dynstr. These relationships * fan out: Disqualifying the .dynsym will disqualify the hash * section, and so forth. * * Disqualify sections that don't have the necessary prerequisites. */ /* Things that need the dynamic string table */ if (sec.dynstr.size == 0) sec.dynstr.type = SINFO_T_NULL; if (sec.dynstr.type != SINFO_T_DYNSTR) { sinfo_free(&sec.dyn, 1); /* Data already fetched */ sec.dynsym.type = SINFO_T_NULL; sec.dynsym.type = SINFO_T_NULL; sec.verdef.type = SINFO_T_NULL; sec.verneed.type = SINFO_T_NULL; } /* * The length of the hash section is encoded in its first two * elements (nbucket, and nchain). The length of the dynsym, * ldynsym, and versym are not given in the dynamic section, * but are known to be the same as nchain. * * If we don't have a hash table, or cannot read nbuckets and * nchain, we have to invalidate all of these. */ if (sec.hash.type == SINFO_T_HASH) { Word nbucket; Word nchain; size_t total; if (hash_size(&fstate, &sec.hash, &nbucket, &nchain, &total) == 0) { sec.hash.type = SINFO_T_NULL; } else { /* Use these counts to set sizes for related sections */ sec.hash.size = total * sizeof (Word); sec.dynsym.size = nchain * sizeof (Sym); sec.versym.size = nchain * sizeof (Versym); /* * The ldynsym size received the DT_SUNW_SYMSZ * value, which is the combined size of .dynsym * and .ldynsym. Now that we have the dynsym size, * use it to lower the ldynsym size to its real size. */ if (sec.ldynsym.size > sec.dynsym.size) sec.ldynsym.size -= sec.dynsym.size; } } /* * If the hash table is not present, or if the call to * hash_size() failed, then discard the sections that * need it to determine their length. */ if (sec.hash.type != SINFO_T_HASH) { sec.dynsym.type = SINFO_T_NULL; sec.ldynsym.type = SINFO_T_NULL; sec.versym.type = SINFO_T_NULL; } /* * The runtime linker does not receive size information for * Verdef and Verneed sections. We have to read their data * in pieces and calculate it. */ if ((sec.verdef.type == SINFO_T_VERDEF) && (verdefneed_size(&fstate, &sec.verdef) == 0)) sec.verdef.type = SINFO_T_NULL; if ((sec.verneed.type == SINFO_T_VERNEED) && (verdefneed_size(&fstate, &sec.verneed) == 0)) sec.verneed.type = SINFO_T_NULL; /* Discard any section with a zero length */ ndx = sinfo_n; for (sinfo = secarr; ndx-- > 0; sinfo++) if ((sinfo->type != SINFO_T_NULL) && (sinfo->size == 0)) sinfo->type = SINFO_T_NULL; /* Things that need the dynamic symbol table */ if (sec.dynsym.type != SINFO_T_DYNSYM) { sec.ldynsym.type = SINFO_T_NULL; sec.hash.type = SINFO_T_NULL; sec.syminfo.type = SINFO_T_NULL; sec.versym.type = SINFO_T_NULL; sec.move.type = SINFO_T_NULL; sec.rel.type = SINFO_T_NULL; sec.rela.type = SINFO_T_NULL; } /* Things that need the dynamic local symbol table */ if (sec.ldynsym.type != SINFO_T_DYNSYM) { sec.symsort.type = SINFO_T_NULL; sec.tlssort.type = SINFO_T_NULL; } /* * Look through the results and fetch the data for any sections * we have found. At the same time, count the number. */ num_sinfo = num_list_sinfo = 0; ndx = sinfo_n; for (sinfo = secarr; ndx-- > 0; sinfo++) { if ((sinfo->type != SINFO_T_NULL) && (sinfo->data == NULL)) (void) get_data(&fstate, sinfo); if (sinfo->data != NULL) num_sinfo++; } for (sinfo_list = seclist.next; sinfo_list != &seclist; sinfo_list = sinfo_list->next) { sinfo = &sinfo_list->sinfo; if ((sinfo->type != SINFO_T_NULL) && (sinfo->data == NULL)) (void) get_data(&fstate, sinfo); if (sinfo->data != NULL) num_list_sinfo++; } /* * Allocate the cache array and fill it in. The cache array * ends up taking all the dynamic memory we've allocated * to build up sec and seclist, so on success, we have nothing * left to clean up. If we can't allocate the cache array * though, we have to free up everything else. */ *shnum = num_sinfo + num_list_sinfo + 1; /* Extra for 1st NULL sec. */ if ((*cache = _cache = malloc((*shnum) * sizeof (Cache))) == NULL) { int err = errno; (void) fprintf(stderr, MSG_INTL(MSG_ERR_MALLOC), file, strerror(err)); sinfo_free(secarr, num_sinfo); sinfo_list_free_all(&seclist); return (0); } *_cache = cache_init; _cache++; ndx = 1; for (sinfo = secarr; num_sinfo > 0; sinfo++) { if (sinfo->data != NULL) { _cache->c_scn = NULL; _cache->c_shdr = sinfo->shdr; _cache->c_data = sinfo->data; _cache->c_name = (char *)sinfo_data[sinfo->type].name; _cache->c_ndx = ndx++; _cache++; num_sinfo--; } } for (sinfo_list = seclist.next; num_list_sinfo > 0; sinfo_list = sinfo_list->next) { sinfo = &sinfo_list->sinfo; if (sinfo->data != NULL) { _cache->c_scn = NULL; _cache->c_shdr = sinfo->shdr; _cache->c_data = sinfo->data; _cache->c_name = (char *)sinfo_data[sinfo->type].name; _cache->c_ndx = ndx++; _cache++; num_list_sinfo--; } } return (1); } /* * Release all the memory referenced by a cache array allocated * by fake_shdr_cache(). */ void fake_shdr_cache_free(Cache *cache, size_t shnum) { Cache *_cache; for (_cache = cache; shnum--; _cache++) { if (_cache->c_data != NULL) { if (_cache->c_data->d_buf != NULL) free(_cache->c_data->d_buf); free(_cache->c_data); } if (_cache->c_shdr) free(_cache->c_shdr); } free(cache); }