/* * 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 (c) 1988 AT&T * All Rights Reserved * * Copyright (c) 1989, 2010, Oracle and/or its affiliates. All rights reserved. * Copyright (c) 2012, Joyent, Inc. All rights reserved. */ /* * Processing of relocatable objects and shared objects. */ #define ELF_TARGET_AMD64 #define ELF_TARGET_SPARC #include #include #include #include #include #include #include #include #include #include #include <_libld.h> /* * Decide if we can link against this input file. */ static int ifl_verify(Ehdr *ehdr, Ofl_desc *ofl, Rej_desc *rej) { /* * Check the validity of the elf header information for compatibility * with this machine and our own internal elf library. */ if ((ehdr->e_machine != ld_targ.t_m.m_mach) && ((ehdr->e_machine != ld_targ.t_m.m_machplus) && ((ehdr->e_flags & ld_targ.t_m.m_flagsplus) == 0))) { rej->rej_type = SGS_REJ_MACH; rej->rej_info = (uint_t)ehdr->e_machine; return (0); } if (ehdr->e_ident[EI_DATA] != ld_targ.t_m.m_data) { rej->rej_type = SGS_REJ_DATA; rej->rej_info = (uint_t)ehdr->e_ident[EI_DATA]; return (0); } if (ehdr->e_version > ofl->ofl_dehdr->e_version) { rej->rej_type = SGS_REJ_VERSION; rej->rej_info = (uint_t)ehdr->e_version; return (0); } return (1); } /* * Check sanity of file header and allocate an infile descriptor * for the file being processed. */ static Ifl_desc * ifl_setup(const char *name, Ehdr *ehdr, Elf *elf, Word flags, Ofl_desc *ofl, Rej_desc *rej) { Ifl_desc *ifl; Rej_desc _rej = { 0 }; if (ifl_verify(ehdr, ofl, &_rej) == 0) { _rej.rej_name = name; DBG_CALL(Dbg_file_rejected(ofl->ofl_lml, &_rej, ld_targ.t_m.m_mach)); if (rej->rej_type == 0) { *rej = _rej; rej->rej_name = strdup(_rej.rej_name); } return (0); } if ((ifl = libld_calloc(1, sizeof (Ifl_desc))) == NULL) return ((Ifl_desc *)S_ERROR); ifl->ifl_name = name; ifl->ifl_ehdr = ehdr; ifl->ifl_elf = elf; ifl->ifl_flags = flags; /* * Is this file using 'extended Section Indexes'. If so, use the * e_shnum & e_shstrndx which can be found at: * * e_shnum == Shdr[0].sh_size * e_shstrndx == Shdr[0].sh_link */ if ((ehdr->e_shnum == 0) && (ehdr->e_shoff != 0)) { Elf_Scn *scn; Shdr *shdr0; if ((scn = elf_getscn(elf, 0)) == NULL) { ld_eprintf(ofl, ERR_ELF, MSG_INTL(MSG_ELF_GETSCN), name); return ((Ifl_desc *)S_ERROR); } if ((shdr0 = elf_getshdr(scn)) == NULL) { ld_eprintf(ofl, ERR_ELF, MSG_INTL(MSG_ELF_GETSHDR), name); return ((Ifl_desc *)S_ERROR); } ifl->ifl_shnum = (Word)shdr0->sh_size; if (ehdr->e_shstrndx == SHN_XINDEX) ifl->ifl_shstrndx = shdr0->sh_link; else ifl->ifl_shstrndx = ehdr->e_shstrndx; } else { ifl->ifl_shnum = ehdr->e_shnum; ifl->ifl_shstrndx = ehdr->e_shstrndx; } if ((ifl->ifl_isdesc = libld_calloc(ifl->ifl_shnum, sizeof (Is_desc *))) == NULL) return ((Ifl_desc *)S_ERROR); /* * Record this new input file on the shared object or relocatable * object input file list. */ if (ifl->ifl_ehdr->e_type == ET_DYN) { if (aplist_append(&ofl->ofl_sos, ifl, AL_CNT_OFL_LIBS) == NULL) return ((Ifl_desc *)S_ERROR); } else { if (aplist_append(&ofl->ofl_objs, ifl, AL_CNT_OFL_OBJS) == NULL) return ((Ifl_desc *)S_ERROR); } return (ifl); } /* * Process a generic section. The appropriate section information is added * to the files input descriptor list. */ static uintptr_t process_section(const char *name, Ifl_desc *ifl, Shdr *shdr, Elf_Scn *scn, Word ndx, int ident, Ofl_desc *ofl) { Is_desc *isp; /* * Create a new input section descriptor. If this is a NOBITS * section elf_getdata() will still create a data buffer (the buffer * will be null and the size will reflect the actual memory size). */ if ((isp = libld_calloc(sizeof (Is_desc), 1)) == NULL) return (S_ERROR); isp->is_shdr = shdr; isp->is_file = ifl; isp->is_name = name; isp->is_scnndx = ndx; isp->is_flags = FLG_IS_EXTERNAL; isp->is_keyident = ident; if ((isp->is_indata = elf_getdata(scn, NULL)) == NULL) { ld_eprintf(ofl, ERR_ELF, MSG_INTL(MSG_ELF_GETDATA), ifl->ifl_name); return (0); } if ((shdr->sh_flags & SHF_EXCLUDE) && ((ofl->ofl_flags & FLG_OF_RELOBJ) == 0)) { isp->is_flags |= FLG_IS_DISCARD; } /* * Add the new input section to the files input section list and * flag whether the section needs placing in an output section. This * placement is deferred until all input section processing has been * completed, as SHT_GROUP sections can provide information that will * affect how other sections within the file should be placed. */ ifl->ifl_isdesc[ndx] = isp; if (ident) { if (shdr->sh_flags & ALL_SHF_ORDER) { isp->is_flags |= FLG_IS_ORDERED; ifl->ifl_flags |= FLG_IF_ORDERED; } isp->is_flags |= FLG_IS_PLACE; } return (1); } /* * Determine the software capabilities of the object being built from the * capabilities of the input relocatable objects. One software capability * is presently recognized, and represented with the following (sys/elf.h): * * SF1_SUNW_FPKNWN use/non-use of frame pointer is known, and * SF1_SUNW_FPUSED the frame pointer is in use. * * The resolution of the present fame pointer state, and the capabilities * provided by a new input relocatable object are: * * new input relocatable object * * present | SF1_SUNW_FPKNWN | SF1_SUNW_FPKNWN | * state | SF1_SUNW_FPUSED | | * --------------------------------------------------------------------------- * SF1_SUNW_FPKNWN | SF1_SUNW_FPKNWN | SF1_SUNW_FPKNWN | SF1_SUNW_FPKNWN * SF1_SUNW_FPUSED | SF1_SUNW_FPUSED | | SF1_SUNW_FPUSED * --------------------------------------------------------------------------- * SF1_SUNW_FPKNWN | SF1_SUNW_FPKNWN | SF1_SUNW_FPKNWN | SF1_SUNW_FPKNWN * | | | * --------------------------------------------------------------------------- * | SF1_SUNW_FPKNWN | SF1_SUNW_FPKNWN | * | SF1_SUNW_FPUSED | | */ static void sf1_cap(Ofl_desc *ofl, Xword val, Ifl_desc *ifl, Is_desc *cisp) { #define FP_FLAGS (SF1_SUNW_FPKNWN | SF1_SUNW_FPUSED) Xword badval; /* * If a mapfile has established definitions to override any object * capabilities, ignore any new object capabilities. */ if (ofl->ofl_flags1 & FLG_OF1_OVSFCAP1) { DBG_CALL(Dbg_cap_val_entry(ofl->ofl_lml, DBG_STATE_IGNORED, CA_SUNW_SF_1, val, ld_targ.t_m.m_mach)); return; } #if !defined(_ELF64) if (ifl && (ifl->ifl_ehdr->e_type == ET_REL)) { /* * The SF1_SUNW_ADDR32 is only meaningful when building a 64-bit * object. Warn the user, and remove the setting, if we're * building a 32-bit object. */ if (val & SF1_SUNW_ADDR32) { ld_eprintf(ofl, ERR_WARNING, MSG_INTL(MSG_FIL_INADDR32SF1), ifl->ifl_name, EC_WORD(cisp->is_scnndx), cisp->is_name); val &= ~SF1_SUNW_ADDR32; } } #endif /* * If this object doesn't specify any capabilities, ignore it, and * leave the state as is. */ if (val == 0) return; /* * Make sure we only accept known software capabilities. Note, that * an F1_SUNW_FPUSED by itself is viewed as bad practice. */ if ((badval = (val & ~SF1_SUNW_MASK)) != 0) { ld_eprintf(ofl, ERR_WARNING, MSG_INTL(MSG_FIL_BADSF1), ifl->ifl_name, EC_WORD(cisp->is_scnndx), cisp->is_name, EC_XWORD(badval)); val &= SF1_SUNW_MASK; } if ((val & FP_FLAGS) == SF1_SUNW_FPUSED) { ld_eprintf(ofl, ERR_WARNING, MSG_INTL(MSG_FIL_BADSF1), ifl->ifl_name, EC_WORD(cisp->is_scnndx), cisp->is_name, EC_XWORD(val)); return; } /* * If the input file is not a relocatable object, then we're only here * to warn the user of any questionable capabilities. */ if (ifl->ifl_ehdr->e_type != ET_REL) { #if defined(_ELF64) /* * If we're building a 64-bit executable, and we come across a * dependency that requires a restricted address space, then * that dependencies requirement can only be satisfied if the * executable triggers the restricted address space. This is a * warning rather than a fatal error, as the possibility exists * that an appropriate dependency will be provided at runtime. * The runtime linker will refuse to use this dependency. */ if ((val & SF1_SUNW_ADDR32) && (ofl->ofl_flags & FLG_OF_EXEC) && ((ofl->ofl_ocapset.oc_sf_1.cm_val & SF1_SUNW_ADDR32) == 0)) { ld_eprintf(ofl, ERR_WARNING, MSG_INTL(MSG_FIL_EXADDR32SF1), ifl->ifl_name, EC_WORD(cisp->is_scnndx), cisp->is_name); } #endif return; } if (DBG_ENABLED) { Dbg_cap_val_entry(ofl->ofl_lml, DBG_STATE_CURRENT, CA_SUNW_SF_1, ofl->ofl_ocapset.oc_sf_1.cm_val, ld_targ.t_m.m_mach); Dbg_cap_val_entry(ofl->ofl_lml, DBG_STATE_NEW, CA_SUNW_SF_1, val, ld_targ.t_m.m_mach); } /* * Determine the resolution of the present frame pointer and the * new input relocatable objects frame pointer. */ if ((ofl->ofl_ocapset.oc_sf_1.cm_val & FP_FLAGS) == FP_FLAGS) { /* * If the new relocatable object isn't using a frame pointer, * reduce the present state to unused. */ if ((val & FP_FLAGS) != FP_FLAGS) ofl->ofl_ocapset.oc_sf_1.cm_val &= ~SF1_SUNW_FPUSED; /* * Having processed the frame pointer bits, remove them from * the value so they don't get OR'd in below. */ val &= ~FP_FLAGS; } else if ((ofl->ofl_ocapset.oc_sf_1.cm_val & SF1_SUNW_FPKNWN) == 0) { /* * If the present frame pointer state is unknown, mask it out * and allow the values from the new relocatable object * to overwrite them. */ ofl->ofl_ocapset.oc_sf_1.cm_val &= ~FP_FLAGS; } else { /* Do not take the frame pointer flags from the object */ val &= ~FP_FLAGS; } ofl->ofl_ocapset.oc_sf_1.cm_val |= val; DBG_CALL(Dbg_cap_val_entry(ofl->ofl_lml, DBG_STATE_RESOLVED, CA_SUNW_SF_1, ofl->ofl_ocapset.oc_sf_1.cm_val, ld_targ.t_m.m_mach)); #undef FP_FLAGS } /* * Determine the hardware capabilities of the object being built from the * capabilities of the input relocatable objects. There's really little to * do here, other than to offer diagnostics, hardware capabilities are simply * additive. */ static void hw_cap(Ofl_desc *ofl, Xword tag, Xword val) { elfcap_mask_t *hwcap; ofl_flag_t flags1; if (tag == CA_SUNW_HW_1) { hwcap = &ofl->ofl_ocapset.oc_hw_1.cm_val; flags1 = FLG_OF1_OVHWCAP1; } else { hwcap = &ofl->ofl_ocapset.oc_hw_2.cm_val; flags1 = FLG_OF1_OVHWCAP2; } /* * If a mapfile has established definitions to override any object * capabilities, ignore any new object capabilities. */ if (ofl->ofl_flags1 & flags1) { DBG_CALL(Dbg_cap_val_entry(ofl->ofl_lml, DBG_STATE_IGNORED, tag, val, ld_targ.t_m.m_mach)); return; } /* * If this object doesn't specify any capabilities, ignore it, and * leave the state as is. */ if (val == 0) return; if (DBG_ENABLED) { Dbg_cap_val_entry(ofl->ofl_lml, DBG_STATE_CURRENT, CA_SUNW_HW_1, ofl->ofl_ocapset.oc_hw_1.cm_val, ld_targ.t_m.m_mach); Dbg_cap_val_entry(ofl->ofl_lml, DBG_STATE_NEW, CA_SUNW_HW_1, val, ld_targ.t_m.m_mach); } *hwcap |= val; DBG_CALL(Dbg_cap_val_entry(ofl->ofl_lml, DBG_STATE_RESOLVED, tag, *hwcap, ld_targ.t_m.m_mach)); } /* * Promote a machine capability or platform capability to the output file. * Multiple instances of these names can be defined. */ static void str_cap(Ofl_desc *ofl, char *pstr, ofl_flag_t flags, Xword tag, Caplist *list) { Capstr *capstr; Aliste idx; Boolean found = FALSE; /* * If a mapfile has established definitions to override this capability, * ignore any new capability. */ if (ofl->ofl_flags1 & flags) { DBG_CALL(Dbg_cap_ptr_entry(ofl->ofl_lml, DBG_STATE_IGNORED, tag, pstr)); return; } for (ALIST_TRAVERSE(list->cl_val, idx, capstr)) { DBG_CALL(Dbg_cap_ptr_entry(ofl->ofl_lml, DBG_STATE_CURRENT, tag, capstr->cs_str)); if (strcmp(capstr->cs_str, pstr) == 0) found = TRUE; } DBG_CALL(Dbg_cap_ptr_entry(ofl->ofl_lml, DBG_STATE_NEW, tag, pstr)); if (found == FALSE) { if ((capstr = alist_append(&list->cl_val, NULL, sizeof (Capstr), AL_CNT_CAP_NAMES)) == NULL) { ofl->ofl_flags |= FLG_OF_FATAL; return; } capstr->cs_str = pstr; } if (DBG_ENABLED) { for (ALIST_TRAVERSE(list->cl_val, idx, capstr)) { DBG_CALL(Dbg_cap_ptr_entry(ofl->ofl_lml, DBG_STATE_RESOLVED, tag, capstr->cs_str)); } } } /* * Promote a capability identifier to the output file. A capability group can * only have one identifier, and thus only the first identifier seen from any * input relocatable objects is retained. An explicit user defined identifier, * rather than an an identifier fabricated by ld(1) with -z symbcap processing, * takes precedence. Note, a user may have defined an identifier via a mapfile, * in which case the mapfile identifier is retained. */ static void id_cap(Ofl_desc *ofl, char *pstr, oc_flag_t flags) { Objcapset *ocapset = &ofl->ofl_ocapset; if (ocapset->oc_id.cs_str) { DBG_CALL(Dbg_cap_ptr_entry(ofl->ofl_lml, DBG_STATE_CURRENT, CA_SUNW_ID, ocapset->oc_id.cs_str)); if ((ocapset->oc_flags & FLG_OCS_USRDEFID) || ((flags & FLG_OCS_USRDEFID) == 0)) { DBG_CALL(Dbg_cap_ptr_entry(ofl->ofl_lml, DBG_STATE_IGNORED, CA_SUNW_ID, pstr)); return; } } DBG_CALL(Dbg_cap_ptr_entry(ofl->ofl_lml, DBG_STATE_NEW, CA_SUNW_ID, pstr)); ocapset->oc_id.cs_str = pstr; ocapset->oc_flags |= flags; DBG_CALL(Dbg_cap_ptr_entry(ofl->ofl_lml, DBG_STATE_RESOLVED, CA_SUNW_ID, pstr)); } /* * Promote a capabilities group to the object capabilities. This catches a * corner case. An object capabilities file can be converted to symbol * capabilities with -z symbolcap. However, if the user has indicated that all * the symbols should be demoted, we'd be left with a symbol capabilities file, * with no associated symbols. Catch this case by promoting the symbol * capabilities back to object capabilities. */ void ld_cap_move_symtoobj(Ofl_desc *ofl) { Cap_group *cgp; Aliste idx1; for (APLIST_TRAVERSE(ofl->ofl_capgroups, idx1, cgp)) { Objcapset *scapset = &cgp->cg_set; Capstr *capstr; Aliste idx2; if (scapset->oc_id.cs_str) { if (scapset->oc_flags & FLG_OCS_USRDEFID) id_cap(ofl, scapset->oc_id.cs_str, scapset->oc_flags); } if (scapset->oc_plat.cl_val) { for (ALIST_TRAVERSE(scapset->oc_plat.cl_val, idx2, capstr)) { str_cap(ofl, capstr->cs_str, FLG_OF1_OVPLATCAP, CA_SUNW_PLAT, &ofl->ofl_ocapset.oc_plat); } } if (scapset->oc_mach.cl_val) { for (ALIST_TRAVERSE(scapset->oc_mach.cl_val, idx2, capstr)) { str_cap(ofl, capstr->cs_str, FLG_OF1_OVMACHCAP, CA_SUNW_MACH, &ofl->ofl_ocapset.oc_mach); } } if (scapset->oc_hw_2.cm_val) hw_cap(ofl, CA_SUNW_HW_2, scapset->oc_hw_2.cm_val); if (scapset->oc_hw_1.cm_val) hw_cap(ofl, CA_SUNW_HW_1, scapset->oc_hw_1.cm_val); if (scapset->oc_sf_1.cm_val) sf1_cap(ofl, scapset->oc_sf_1.cm_val, NULL, NULL); } } /* * Determine whether a capabilities group already exists that describes this * new capabilities group. * * Note, a capability group identifier, CA_SUNW_ID, isn't used as part of the * comparison. This attribute simply assigns a diagnostic name to the group, * and in the case of multiple identifiers, the first will be taken. */ static Cap_group * get_cap_group(Objcapset *ocapset, Word cnum, Ofl_desc *ofl, Is_desc *isp) { Aliste idx; Cap_group *cgp; Word ccnum = cnum; /* * If the new capabilities contains a CA_SUNW_ID, drop the count of the * number of comparable items. */ if (ocapset->oc_id.cs_str) ccnum--; /* * Traverse the existing symbols capabilities groups. */ for (APLIST_TRAVERSE(ofl->ofl_capgroups, idx, cgp)) { Word onum = cgp->cg_num; Alist *calp, *oalp; if (cgp->cg_set.oc_id.cs_str) onum--; if (onum != ccnum) continue; if (cgp->cg_set.oc_hw_1.cm_val != ocapset->oc_hw_1.cm_val) continue; if (cgp->cg_set.oc_sf_1.cm_val != ocapset->oc_sf_1.cm_val) continue; if (cgp->cg_set.oc_hw_2.cm_val != ocapset->oc_hw_2.cm_val) continue; calp = cgp->cg_set.oc_plat.cl_val; oalp = ocapset->oc_plat.cl_val; if ((calp == NULL) && oalp) continue; if (calp && ((oalp == NULL) || cap_names_match(calp, oalp))) continue; calp = cgp->cg_set.oc_mach.cl_val; oalp = ocapset->oc_mach.cl_val; if ((calp == NULL) && oalp) continue; if (calp && ((oalp == NULL) || cap_names_match(calp, oalp))) continue; /* * If a matching group is found, then this new group has * already been supplied by a previous file, and hence the * existing group can be used. Record this new input section, * from which we can also derive the input file name, on the * existing groups input sections. */ if (aplist_append(&(cgp->cg_secs), isp, AL_CNT_CAP_SECS) == NULL) return (NULL); return (cgp); } /* * If a capabilities group is not found, create a new one. */ if (((cgp = libld_calloc(sizeof (Cap_group), 1)) == NULL) || (aplist_append(&(ofl->ofl_capgroups), cgp, AL_CNT_CAP_DESCS) == NULL)) return (NULL); /* * If we're converting object capabilities to symbol capabilities and * no CA_SUNW_ID is defined, fabricate one. This identifier is appended * to all symbol names that are converted into capabilities symbols, * see ld_sym_process(). */ if ((isp->is_file->ifl_flags & FLG_IF_OTOSCAP) && (ocapset->oc_id.cs_str == NULL)) { size_t len; /* * Create an identifier using the group number together with a * default template. We allocate a buffer large enough for any * possible number of items (way more than we need). */ len = MSG_STR_CAPGROUPID_SIZE + CONV_INV_BUFSIZE; if ((ocapset->oc_id.cs_str = libld_malloc(len)) == NULL) return (NULL); (void) snprintf(ocapset->oc_id.cs_str, len, MSG_ORIG(MSG_STR_CAPGROUPID), aplist_nitems(ofl->ofl_capgroups)); cnum++; } cgp->cg_set = *ocapset; cgp->cg_num = cnum; /* * Null the callers alist's as they've effectively been transferred * to this new Cap_group. */ ocapset->oc_plat.cl_val = ocapset->oc_mach.cl_val = NULL; /* * Keep track of which input section, and hence input file, established * this group. */ if (aplist_append(&(cgp->cg_secs), isp, AL_CNT_CAP_SECS) == NULL) return (NULL); /* * Keep track of the number of symbol capabilities entries that will be * required in the output file. Each group requires a terminating * CA_SUNW_NULL. */ ofl->ofl_capsymcnt += (cnum + 1); return (cgp); } /* * Capture symbol capability family information. This data structure is focal * in maintaining all symbol capability relationships, and provides for the * eventual creation of a capabilities information section, and possibly a * capabilities chain section. * * Capabilities families are lead by a CAPINFO_SUNW_GLOB symbol. This symbol * provides the visible global symbol that is referenced by all external * callers. This symbol may have aliases. For example, a weak/global symbol * pair, such as memcpy()/_memcpy() may lead the same capabilities family. * Each family contains one or more local symbol members. These members provide * the capabilities specific functions, and are associated to a capabilities * group. For example, the capability members memcpy%sun4u and memcpy%sun4v * might be associated with the memcpy() capability family. * * This routine is called when a relocatable object that provides object * capabilities is transformed into a symbol capabilities object, using the * -z symbolcap option. * * This routine is also called to collect the SUNW_capinfo section information * of a relocatable object that contains symbol capability definitions. */ uintptr_t ld_cap_add_family(Ofl_desc *ofl, Sym_desc *lsdp, Sym_desc *csdp, Cap_group *cgp, APlist **csyms) { Cap_avlnode qcav, *cav; avl_tree_t *avlt; avl_index_t where = 0; Cap_sym *mcsp; Aliste idx; /* * Make sure the capability families have an initialized AVL tree. */ if ((avlt = ofl->ofl_capfamilies) == NULL) { if ((avlt = libld_calloc(sizeof (avl_tree_t), 1)) == NULL) return (S_ERROR); avl_create(avlt, &ld_sym_avl_comp, sizeof (Cap_avlnode), SGSOFFSETOF(Cap_avlnode, cn_symavlnode.sav_node)); ofl->ofl_capfamilies = avlt; /* * When creating a dynamic object, capability family members * are maintained in a .SUNW_capchain, the first entry of * which is the version number of the chain. */ ofl->ofl_capchaincnt = 1; } /* * Determine whether a family already exists, and if not, create one * using the lead family symbol. */ qcav.cn_symavlnode.sav_hash = (Word)elf_hash(lsdp->sd_name); qcav.cn_symavlnode.sav_name = lsdp->sd_name; if ((cav = avl_find(avlt, &qcav, &where)) == NULL) { if ((cav = libld_calloc(sizeof (Cap_avlnode), 1)) == NULL) return (S_ERROR); cav->cn_symavlnode.sav_hash = qcav.cn_symavlnode.sav_hash; cav->cn_symavlnode.sav_name = qcav.cn_symavlnode.sav_name; cav->cn_symavlnode.sav_sdp = lsdp; avl_insert(avlt, cav, where); /* * When creating a dynamic object, capability family members * are maintained in a .SUNW_capchain, each family starts with * this lead symbol, and is terminated with a 0 element. */ ofl->ofl_capchaincnt += 2; } /* * If no group information is provided then this request is to add a * lead capability symbol, or lead symbol alias. If this is the lead * symbol there's nothing more to do. Otherwise save the alias. */ if (cgp == NULL) { if ((lsdp != csdp) && (aplist_append(&cav->cn_aliases, csdp, AL_CNT_CAP_ALIASES) == NULL)) return (S_ERROR); return (0); } /* * Determine whether a member of the same group as this new member is * already defined within this family. If so, we have a multiply * defined symbol. */ for (APLIST_TRAVERSE(cav->cn_members, idx, mcsp)) { Sym_desc *msdp; if (cgp != mcsp->cs_group) continue; /* * Diagnose that a multiple symbol definition exists. */ msdp = mcsp->cs_sdp; ld_eprintf(ofl, ERR_FATAL, MSG_INTL(MSG_CAP_MULDEF), demangle(lsdp->sd_name)); ld_eprintf(ofl, ERR_NONE, MSG_INTL(MSG_CAP_MULDEFSYMS), msdp->sd_file->ifl_name, msdp->sd_name, csdp->sd_file->ifl_name, csdp->sd_name); } /* * Add this capabilities symbol member to the family. */ if (((mcsp = libld_malloc(sizeof (Cap_sym))) == NULL) || (aplist_append(&cav->cn_members, mcsp, AL_CNT_CAP_MEMS) == NULL)) return (S_ERROR); mcsp->cs_sdp = csdp; mcsp->cs_group = cgp; /* * When creating a dynamic object, capability family members are * maintained in a .SUNW_capchain. Account for this family member. */ ofl->ofl_capchaincnt++; /* * If this input file is undergoing object capabilities to symbol * capabilities conversion, then this member is a new local symbol * that has been generated from an original global symbol. Keep track * of this symbol so that the output file symbol table can be populated * with these new symbol entries. */ if (csyms && (aplist_append(csyms, mcsp, AL_CNT_CAP_SYMS) == NULL)) return (S_ERROR); return (0); } /* * Process a SHT_SUNW_cap capabilities section. */ static uintptr_t process_cap(Ofl_desc *ofl, Ifl_desc *ifl, Is_desc *cisp) { Objcapset ocapset = { 0 }; Cap_desc *cdp; Cap *data, *cdata; char *strs; Word ndx, cnum; int objcapndx, descapndx, symcapndx; int nulls, capstrs = 0; /* * Determine the capabilities data and size. */ cdata = (Cap *)cisp->is_indata->d_buf; cnum = (Word)(cisp->is_shdr->sh_size / cisp->is_shdr->sh_entsize); if ((cdata == NULL) || (cnum == 0)) return (0); DBG_CALL(Dbg_cap_sec_title(ofl->ofl_lml, ifl->ifl_name)); /* * Traverse the section to determine what capabilities groups are * available. * * A capabilities section can contain one or more, CA_SUNW_NULL * terminated groups. * * - The first group defines the object capabilities. * - Additional groups define symbol capabilities. * - Since the initial group is always reserved for object * capabilities, any object with symbol capabilities must also * have an object capabilities group. If the object has no object * capabilities, an empty object group is defined, consisting of a * CA_SUNW_NULL element in index [0]. * - If any capabilities require references to a named string, then * the section header sh_info points to the associated string * table. * - If an object contains symbol capability groups, then the * section header sh_link points to the associated capinfo table. */ objcapndx = 0; descapndx = symcapndx = -1; nulls = 0; for (ndx = 0, data = cdata; ndx < cnum; ndx++, data++) { switch (data->c_tag) { case CA_SUNW_NULL: /* * If this is the first CA_SUNW_NULL entry, and no * capabilities group has been found, then this object * does not define any object capabilities. */ if (nulls++ == 0) { if (ndx == 0) objcapndx = -1; } else if ((symcapndx == -1) && (descapndx != -1)) symcapndx = descapndx; break; case CA_SUNW_PLAT: case CA_SUNW_MACH: case CA_SUNW_ID: capstrs++; /* FALLTHROUGH */ case CA_SUNW_HW_1: case CA_SUNW_SF_1: case CA_SUNW_HW_2: /* * If this is the start of a new group, save it. */ if (descapndx == -1) descapndx = ndx; break; default: ld_eprintf(ofl, ERR_WARNING, MSG_INTL(MSG_FIL_UNKCAP), ifl->ifl_name, EC_WORD(cisp->is_scnndx), cisp->is_name, data->c_tag); } } /* * If a string capabilities entry has been found, the capabilities * section must reference the associated string table. */ if (capstrs) { Word info = cisp->is_shdr->sh_info; if ((info == 0) || (info > ifl->ifl_shnum)) { ld_eprintf(ofl, ERR_FATAL, MSG_INTL(MSG_FIL_INVSHINFO), ifl->ifl_name, EC_WORD(cisp->is_scnndx), cisp->is_name, EC_XWORD(info)); return (S_ERROR); } strs = (char *)ifl->ifl_isdesc[info]->is_indata->d_buf; } /* * The processing of capabilities groups is as follows: * * - if a relocatable object provides only object capabilities, and * the -z symbolcap option is in effect, then the object * capabilities are transformed into symbol capabilities and the * symbol capabilities are carried over to the output file. * - in all other cases, any capabilities present in an input * relocatable object are carried from the input object to the * output without any transformation or conversion. * * Capture any object capabilities that are to be carried over to the * output file. */ if ((objcapndx == 0) && ((symcapndx != -1) || ((ofl->ofl_flags & FLG_OF_OTOSCAP) == 0))) { for (ndx = 0, data = cdata; ndx < cnum; ndx++, data++) { /* * Object capabilities end at the first null. */ if (data->c_tag == CA_SUNW_NULL) break; /* * Only the object software capabilities that are * defined in a relocatable object become part of the * object software capabilities in the output file. * However, check the validity of any object software * capabilities of any dependencies. */ if (data->c_tag == CA_SUNW_SF_1) { sf1_cap(ofl, data->c_un.c_val, ifl, cisp); continue; } /* * The remaining capability types must come from a * relocatable object in order to contribute to the * output. */ if (ifl->ifl_ehdr->e_type != ET_REL) continue; switch (data->c_tag) { case CA_SUNW_HW_1: case CA_SUNW_HW_2: hw_cap(ofl, data->c_tag, data->c_un.c_val); break; case CA_SUNW_PLAT: str_cap(ofl, strs + data->c_un.c_ptr, FLG_OF1_OVPLATCAP, CA_SUNW_PLAT, &ofl->ofl_ocapset.oc_plat); break; case CA_SUNW_MACH: str_cap(ofl, strs + data->c_un.c_ptr, FLG_OF1_OVMACHCAP, CA_SUNW_MACH, &ofl->ofl_ocapset.oc_mach); break; case CA_SUNW_ID: id_cap(ofl, strs + data->c_un.c_ptr, FLG_OCS_USRDEFID); break; default: assert(0); /* Unknown capability type */ } } /* * If there are no symbol capabilities, or this objects * capabilities aren't being transformed into a symbol * capabilities, then we're done. */ if ((symcapndx == -1) && ((ofl->ofl_flags & FLG_OF_OTOSCAP) == 0)) return (1); } /* * If these capabilities don't originate from a relocatable object * there's no further processing required. */ if (ifl->ifl_ehdr->e_type != ET_REL) return (1); /* * If this object only defines an object capabilities group, and the * -z symbolcap option is in effect, then all global function symbols * and initialized global data symbols are renamed and assigned to the * transformed symbol capabilities group. */ if ((objcapndx == 0) && (symcapndx == -1) && (ofl->ofl_flags & FLG_OF_OTOSCAP)) ifl->ifl_flags |= FLG_IF_OTOSCAP; /* * Allocate a capabilities descriptor to collect the capabilities data * for this input file. Allocate a mirror of the raw capabilities data * that points to the individual symbol capabilities groups. An APlist * is used, although it will be sparsely populated, as the list provides * a convenient mechanism for traversal later. */ if (((cdp = libld_calloc(sizeof (Cap_desc), 1)) == NULL) || (aplist_append(&(cdp->ca_groups), NULL, cnum) == NULL)) return (S_ERROR); /* * Clear the allocated APlist data array, and assign the number of * items as the total number of array items. */ (void) memset(&cdp->ca_groups->apl_data[0], 0, (cnum * sizeof (void *))); cdp->ca_groups->apl_nitems = cnum; ifl->ifl_caps = cdp; /* * Traverse the capabilities data, unpacking the data into a * capabilities set. Process each capabilities set as a unique group. */ descapndx = -1; nulls = 0; for (ndx = 0, data = cdata; ndx < cnum; ndx++, data++) { Capstr *capstr; switch (data->c_tag) { case CA_SUNW_NULL: nulls++; /* * Process the capabilities group that this null entry * terminates. The capabilities group that is returned * will either point to this file's data, or to a * matching capabilities group that has already been * processed. * * Note, if this object defines object capabilities, * the first group descriptor points to these object * capabilities. It is only necessary to save this * descriptor when object capabilities are being * transformed into symbol capabilities (-z symbolcap). */ if (descapndx != -1) { if ((nulls > 1) || (ifl->ifl_flags & FLG_IF_OTOSCAP)) { APlist *alp = cdp->ca_groups; if ((alp->apl_data[descapndx] = get_cap_group(&ocapset, (ndx - descapndx), ofl, cisp)) == NULL) return (S_ERROR); } /* * Clean up the capabilities data in preparation * for processing additional groups. If the * collected capabilities strings were used to * establish a new output group, they will have * been saved in get_cap_group(). If these * descriptors still exist, then an existing * descriptor has been used to associate with * this file, and these string descriptors can * be freed. */ ocapset.oc_hw_1.cm_val = ocapset.oc_sf_1.cm_val = ocapset.oc_hw_2.cm_val = 0; if (ocapset.oc_plat.cl_val) { free((void *)ocapset.oc_plat.cl_val); ocapset.oc_plat.cl_val = NULL; } if (ocapset.oc_mach.cl_val) { free((void *)ocapset.oc_mach.cl_val); ocapset.oc_mach.cl_val = NULL; } descapndx = -1; } continue; case CA_SUNW_HW_1: ocapset.oc_hw_1.cm_val = data->c_un.c_val; DBG_CALL(Dbg_cap_val_entry(ofl->ofl_lml, DBG_STATE_ORIGINAL, CA_SUNW_HW_1, ocapset.oc_hw_1.cm_val, ld_targ.t_m.m_mach)); break; case CA_SUNW_SF_1: ocapset.oc_sf_1.cm_val = data->c_un.c_val; DBG_CALL(Dbg_cap_val_entry(ofl->ofl_lml, DBG_STATE_ORIGINAL, CA_SUNW_SF_1, ocapset.oc_sf_1.cm_val, ld_targ.t_m.m_mach)); break; case CA_SUNW_HW_2: ocapset.oc_hw_2.cm_val = data->c_un.c_val; DBG_CALL(Dbg_cap_val_entry(ofl->ofl_lml, DBG_STATE_ORIGINAL, CA_SUNW_HW_2, ocapset.oc_hw_2.cm_val, ld_targ.t_m.m_mach)); break; case CA_SUNW_PLAT: if ((capstr = alist_append(&ocapset.oc_plat.cl_val, NULL, sizeof (Capstr), AL_CNT_CAP_NAMES)) == NULL) return (S_ERROR); capstr->cs_str = strs + data->c_un.c_ptr; DBG_CALL(Dbg_cap_ptr_entry(ofl->ofl_lml, DBG_STATE_ORIGINAL, CA_SUNW_PLAT, capstr->cs_str)); break; case CA_SUNW_MACH: if ((capstr = alist_append(&ocapset.oc_mach.cl_val, NULL, sizeof (Capstr), AL_CNT_CAP_NAMES)) == NULL) return (S_ERROR); capstr->cs_str = strs + data->c_un.c_ptr; DBG_CALL(Dbg_cap_ptr_entry(ofl->ofl_lml, DBG_STATE_ORIGINAL, CA_SUNW_MACH, capstr->cs_str)); break; case CA_SUNW_ID: ocapset.oc_id.cs_str = strs + data->c_un.c_ptr; DBG_CALL(Dbg_cap_ptr_entry(ofl->ofl_lml, DBG_STATE_ORIGINAL, CA_SUNW_ID, ocapset.oc_id.cs_str)); break; } /* * Save the start of this new group. */ if (descapndx == -1) descapndx = ndx; } return (1); } /* * Capture any symbol capabilities symbols. An object file that contains symbol * capabilities has an associated .SUNW_capinfo section. This section * identifies which symbols are associated to which capabilities, together with * their associated lead symbol. Each of these symbol pairs are recorded for * processing later. */ static uintptr_t process_capinfo(Ofl_desc *ofl, Ifl_desc *ifl, Is_desc *isp) { Cap_desc *cdp = ifl->ifl_caps; Capinfo *capinfo = isp->is_indata->d_buf; Shdr *shdr = isp->is_shdr; Word cndx, capinfonum; capinfonum = (Word)(shdr->sh_size / shdr->sh_entsize); if ((cdp == NULL) || (capinfo == NULL) || (capinfonum == 0)) return (0); for (cndx = 1, capinfo++; cndx < capinfonum; cndx++, capinfo++) { Sym_desc *sdp, *lsdp; Word lndx; uchar_t gndx; if ((gndx = (uchar_t)ELF_C_GROUP(*capinfo)) == 0) continue; lndx = (Word)ELF_C_SYM(*capinfo); /* * Catch any anomalies. A capabilities symbol should be valid, * and the capabilities lead symbol should also be global. * Note, ld(1) -z symbolcap would create local capabilities * symbols, but we don't enforce this so as to give the * compilation environment a little more freedom. */ if ((sdp = ifl->ifl_oldndx[cndx]) == NULL) { ld_eprintf(ofl, ERR_WARNING, MSG_INTL(MSG_CAPINFO_INVALSYM), ifl->ifl_name, EC_WORD(isp->is_scnndx), isp->is_name, cndx, MSG_INTL(MSG_STR_UNKNOWN)); continue; } if ((lndx == 0) || (lndx >= ifl->ifl_symscnt) || ((lsdp = ifl->ifl_oldndx[lndx]) == NULL) || (ELF_ST_BIND(lsdp->sd_sym->st_info) != STB_GLOBAL)) { ld_eprintf(ofl, ERR_WARNING, MSG_INTL(MSG_CAPINFO_INVALLEAD), ifl->ifl_name, EC_WORD(isp->is_scnndx), isp->is_name, cndx, lsdp ? demangle(lsdp->sd_name) : MSG_INTL(MSG_STR_UNKNOWN), lndx); continue; } /* * Indicate that this is a capabilities symbol. */ sdp->sd_flags |= FLG_SY_CAP; /* * Save any global capability symbols. Global capability * symbols are identified with a CAPINFO_SUNW_GLOB group id. * The lead symbol for this global capability symbol is either * the symbol itself, or an alias. */ if (gndx == CAPINFO_SUNW_GLOB) { if (ld_cap_add_family(ofl, lsdp, sdp, NULL, NULL) == S_ERROR) return (S_ERROR); continue; } /* * Track the number of non-global capabilities symbols, as these * are used to size any symbol tables. If we're generating a * dynamic object, this symbol will be added to the dynamic * symbol table, therefore ensure there is space in the dynamic * string table. */ ofl->ofl_caploclcnt++; if (((ofl->ofl_flags & FLG_OF_RELOBJ) == 0) && (st_insert(ofl->ofl_dynstrtab, sdp->sd_name) == -1)) return (S_ERROR); /* * As we're tracking this local symbol as a capabilities symbol, * reduce the local symbol count to compensate. */ ofl->ofl_locscnt--; /* * Determine whether the associated lead symbol indicates * NODYNSORT. If so, remove this local entry from the * SUNW_dynsort section too. NODYNSORT tagging can only be * obtained from a mapfile symbol definition, and thus any * global definition that has this tagging has already been * instantiated and this instance resolved to it. */ if (lsdp->sd_flags & FLG_SY_NODYNSORT) { Sym *lsym = lsdp->sd_sym; uchar_t ltype = ELF_ST_TYPE(lsym->st_info); DYNSORT_COUNT(lsdp, lsym, ltype, --); lsdp->sd_flags |= FLG_SY_NODYNSORT; } /* * Track this family member, together with its associated group. */ if (ld_cap_add_family(ofl, lsdp, sdp, cdp->ca_groups->apl_data[gndx], NULL) == S_ERROR) return (S_ERROR); } return (0); } /* * Simply process the section so that we have pointers to the data for use * in later routines, however don't add the section to the output section * list as we will be creating our own replacement sections later (ie. * symtab and relocation). */ static uintptr_t /* ARGSUSED5 */ process_input(const char *name, Ifl_desc *ifl, Shdr *shdr, Elf_Scn *scn, Word ndx, int ident, Ofl_desc *ofl) { return (process_section(name, ifl, shdr, scn, ndx, ld_targ.t_id.id_null, ofl)); } /* * Keep a running count of relocation entries from input relocatable objects for * sizing relocation buckets later. If we're building an executable, save any * relocations from shared objects to determine if any copy relocation symbol * has a displacement relocation against it. */ static uintptr_t /* ARGSUSED5 */ process_reloc(const char *name, Ifl_desc *ifl, Shdr *shdr, Elf_Scn *scn, Word ndx, int ident, Ofl_desc *ofl) { if (process_section(name, ifl, shdr, scn, ndx, ld_targ.t_id.id_null, ofl) == S_ERROR) return (S_ERROR); if (ifl->ifl_ehdr->e_type == ET_REL) { if (shdr->sh_entsize && (shdr->sh_entsize <= shdr->sh_size)) /* LINTED */ ofl->ofl_relocincnt += (Word)(shdr->sh_size / shdr->sh_entsize); } else if (ofl->ofl_flags & FLG_OF_EXEC) { if (aplist_append(&ifl->ifl_relsect, ifl->ifl_isdesc[ndx], AL_CNT_IFL_RELSECS) == NULL) return (S_ERROR); } return (1); } /* * Process a string table section. A valid section contains an initial and * final null byte. */ static uintptr_t process_strtab(const char *name, Ifl_desc *ifl, Shdr *shdr, Elf_Scn *scn, Word ndx, int ident, Ofl_desc *ofl) { char *data; size_t size; Is_desc *isp; uintptr_t error; /* * Never include .stab.excl sections in any output file. * If the -s flag has been specified strip any .stab sections. */ if (((ofl->ofl_flags & FLG_OF_STRIP) && ident && (strncmp(name, MSG_ORIG(MSG_SCN_STAB), MSG_SCN_STAB_SIZE) == 0)) || (strcmp(name, MSG_ORIG(MSG_SCN_STABEXCL)) == 0) && ident) return (1); /* * If we got here to process a .shstrtab or .dynstr table, `ident' will * be null. Otherwise make sure we don't have a .strtab section as this * should not be added to the output section list either. */ if ((ident != ld_targ.t_id.id_null) && (strcmp(name, MSG_ORIG(MSG_SCN_STRTAB)) == 0)) ident = ld_targ.t_id.id_null; error = process_section(name, ifl, shdr, scn, ndx, ident, ofl); if ((error == 0) || (error == S_ERROR)) return (error); /* * String tables should start and end with a NULL byte. Note, it has * been known for the assembler to create empty string tables, so check * the size before attempting to verify the data itself. */ isp = ifl->ifl_isdesc[ndx]; size = isp->is_indata->d_size; if (size) { data = isp->is_indata->d_buf; if (data[0] != '\0' || data[size - 1] != '\0') ld_eprintf(ofl, ERR_WARNING, MSG_INTL(MSG_FIL_MALSTR), ifl->ifl_name, EC_WORD(isp->is_scnndx), name); } else isp->is_indata->d_buf = (void *)MSG_ORIG(MSG_STR_EMPTY); ifl->ifl_flags |= FLG_IF_HSTRTAB; return (1); } /* * Invalid sections produce a warning and are skipped. */ static uintptr_t /* ARGSUSED3 */ invalid_section(const char *name, Ifl_desc *ifl, Shdr *shdr, Elf_Scn *scn, Word ndx, int ident, Ofl_desc *ofl) { Conv_inv_buf_t inv_buf; ld_eprintf(ofl, ERR_WARNING, MSG_INTL(MSG_FIL_INVALSEC), ifl->ifl_name, EC_WORD(ndx), name, conv_sec_type(ifl->ifl_ehdr->e_ident[EI_OSABI], ifl->ifl_ehdr->e_machine, shdr->sh_type, 0, &inv_buf)); return (1); } /* * Compare an input section name to a given string, taking the ELF '%' * section naming convention into account. If an input section name * contains a '%' character, the '%' and all following characters are * ignored in the comparison. * * entry: * is_name - Name of input section * match_name - Name to compare to * match_len - strlen(match_name) * * exit: * Returns True (1) if the names match, and False (0) otherwise. */ inline static int is_name_cmp(const char *is_name, const char *match_name, size_t match_len) { /* * If the start of is_name is not a match for name, * the match fails. */ if (strncmp(is_name, match_name, match_len) != 0) return (0); /* * The prefix matched. The next character must be either '%', or * NULL, in order for a match to be true. */ is_name += match_len; return ((*is_name == '\0') || (*is_name == '%')); } /* * Helper routine for process_progbits() to process allocable sections. * * entry: * name, ifl, shdr, ndx, ident, ofl - As passed to process_progbits(). * is_stab_index - TRUE if section is .index. * is_flags - Additional flags to be added to the input section. * * exit: * The allocable section has been processed. *ident and *is_flags * are updated as necessary to reflect the changes. Returns TRUE * for success, FALSE for failure. */ /*ARGSUSED*/ inline static Boolean process_progbits_alloc(const char *name, Ifl_desc *ifl, Shdr *shdr, Word ndx, int *ident, Ofl_desc *ofl, Boolean is_stab_index, Word *is_flags) { Boolean done = FALSE; if (name[0] == '.') { switch (name[1]) { case 'e': if (!is_name_cmp(name, MSG_ORIG(MSG_SCN_EHFRAME), MSG_SCN_EHFRAME_SIZE)) break; *ident = ld_targ.t_id.id_unwind; *is_flags |= FLG_IS_EHFRAME; done = TRUE; /* * Historically, the section containing the logic to * unwind stack frames -- the .eh_frame section -- was * of type SHT_PROGBITS. Apparently the most * aesthetically galling aspect of this was not the * .eh_frame section's dubious purpose or its filthy * implementation, but rather its section type; with the * introduction of the AMD64 ABI, a new section header * type (SHT_AMD64_UNWIND) was introduced for (and * dedicated to) this section. When both the Sun * compilers and the GNU compilers had been modified to * generate this new section type, the linker became * much more pedantic about .eh_frame: it refused to * link an AMD64 object that contained a .eh_frame with * the legacy SHT_PROGBITS. That this was too fussy is * evidenced by searching the net for the error message * that it generated ("section type is SHT_PROGBITS: * expected SHT_AMD64_UNWIND"), which reveals a myriad * of problems, including legacy objects, hand-coded * assembly and otherwise cross-platform objects * created on other platforms (the GNU toolchain was * only modified to create the new section type on * Solaris and derivatives). We therefore always accept * a .eh_frame of SHT_PROGBITS -- regardless of * m_sht_unwind. */ break; case 'g': if (is_name_cmp(name, MSG_ORIG(MSG_SCN_GOT), MSG_SCN_GOT_SIZE)) { *ident = ld_targ.t_id.id_null; done = TRUE; break; } if ((ld_targ.t_m.m_sht_unwind == SHT_PROGBITS) && is_name_cmp(name, MSG_ORIG(MSG_SCN_GCC_X_TBL), MSG_SCN_GCC_X_TBL_SIZE)) { *ident = ld_targ.t_id.id_unwind; done = TRUE; break; } break; case 'p': if (is_name_cmp(name, MSG_ORIG(MSG_SCN_PLT), MSG_SCN_PLT_SIZE)) { *ident = ld_targ.t_id.id_null; done = TRUE; } break; } } if (!done) { if (is_stab_index) { /* * This is a work-around for x86 compilers that have * set SHF_ALLOC for the .stab.index section. * * Because of this, make sure that the .stab.index * does not end up as the last section in the text * segment. Older linkers can produce segmentation * violations when they strip (ld -s) against a * shared object whose last section in the text * segment is a .stab. */ *ident = ld_targ.t_id.id_interp; } else { *ident = ld_targ.t_id.id_data; } } return (TRUE); } /* * Process a progbits section. */ static uintptr_t process_progbits(const char *name, Ifl_desc *ifl, Shdr *shdr, Elf_Scn *scn, Word ndx, int ident, Ofl_desc *ofl) { Boolean is_stab_index = FALSE; Word is_flags = 0; uintptr_t r; /* * Never include .stab.excl sections in any output file. * If the -s flag has been specified strip any .stab sections. */ if (ident && (strncmp(name, MSG_ORIG(MSG_SCN_STAB), MSG_SCN_STAB_SIZE) == 0)) { if ((ofl->ofl_flags & FLG_OF_STRIP) || (strcmp((name + MSG_SCN_STAB_SIZE), MSG_ORIG(MSG_SCN_EXCL)) == 0)) return (1); if (strcmp((name + MSG_SCN_STAB_SIZE), MSG_ORIG(MSG_SCN_INDEX)) == 0) is_stab_index = TRUE; } if ((ofl->ofl_flags & FLG_OF_STRIP) && ident) { if ((strncmp(name, MSG_ORIG(MSG_SCN_DEBUG), MSG_SCN_DEBUG_SIZE) == 0) || (strcmp(name, MSG_ORIG(MSG_SCN_LINE)) == 0)) return (1); } /* * Update the ident to reflect the type of section we've got. * * If there is any .plt or .got section to generate we'll be creating * our own version, so don't allow any input sections of these types to * be added to the output section list (why a relocatable object would * have a .plt or .got is a mystery, but stranger things have occurred). * * If there are any unwind sections, and this is a platform that uses * SHT_PROGBITS for unwind sections, then set their ident to reflect * that. */ if (ident) { if (shdr->sh_flags & SHF_TLS) { ident = ld_targ.t_id.id_tls; } else if ((shdr->sh_flags & ~ALL_SHF_IGNORE) == (SHF_ALLOC | SHF_EXECINSTR)) { ident = ld_targ.t_id.id_text; } else if (shdr->sh_flags & SHF_ALLOC) { if (process_progbits_alloc(name, ifl, shdr, ndx, &ident, ofl, is_stab_index, &is_flags) == FALSE) return (S_ERROR); } else { ident = ld_targ.t_id.id_note; } } r = process_section(name, ifl, shdr, scn, ndx, ident, ofl); /* * On success, process_section() creates an input section descriptor. * Now that it exists, we can add any pending input section flags. */ if ((is_flags != 0) && (r == 1)) ifl->ifl_isdesc[ndx]->is_flags |= is_flags; return (r); } /* * Handles the SHT_SUNW_{DEBUG,DEBUGSTR) sections. */ static uintptr_t process_debug(const char *name, Ifl_desc *ifl, Shdr *shdr, Elf_Scn *scn, Word ndx, int ident, Ofl_desc *ofl) { /* * Debug information is discarded when the 'ld -s' flag is invoked. */ if (ofl->ofl_flags & FLG_OF_STRIP) { return (1); } return (process_progbits(name, ifl, shdr, scn, ndx, ident, ofl)); } /* * Process a nobits section. */ static uintptr_t process_nobits(const char *name, Ifl_desc *ifl, Shdr *shdr, Elf_Scn *scn, Word ndx, int ident, Ofl_desc *ofl) { if (ident) { if (shdr->sh_flags & SHF_TLS) ident = ld_targ.t_id.id_tlsbss; #if defined(_ELF64) else if ((shdr->sh_flags & SHF_AMD64_LARGE) && (ld_targ.t_m.m_mach == EM_AMD64)) ident = ld_targ.t_id.id_lbss; #endif else ident = ld_targ.t_id.id_bss; } return (process_section(name, ifl, shdr, scn, ndx, ident, ofl)); } /* * Process a SHT_*_ARRAY section. */ static uintptr_t process_array(const char *name, Ifl_desc *ifl, Shdr *shdr, Elf_Scn *scn, Word ndx, int ident, Ofl_desc *ofl) { uintptr_t error; if (ident) ident = ld_targ.t_id.id_array; error = process_section(name, ifl, shdr, scn, ndx, ident, ofl); if ((error == 0) || (error == S_ERROR)) return (error); return (1); } static uintptr_t /* ARGSUSED1 */ array_process(Is_desc *isc, Ifl_desc *ifl, Ofl_desc *ofl) { Os_desc *osp; Shdr *shdr; if ((isc == NULL) || ((osp = isc->is_osdesc) == NULL)) return (0); shdr = isc->is_shdr; if ((shdr->sh_type == SHT_FINI_ARRAY) && (ofl->ofl_osfiniarray == NULL)) ofl->ofl_osfiniarray = osp; else if ((shdr->sh_type == SHT_INIT_ARRAY) && (ofl->ofl_osinitarray == NULL)) ofl->ofl_osinitarray = osp; else if ((shdr->sh_type == SHT_PREINIT_ARRAY) && (ofl->ofl_ospreinitarray == NULL)) ofl->ofl_ospreinitarray = osp; return (1); } /* * Process a SHT_SYMTAB_SHNDX section. */ static uintptr_t process_sym_shndx(const char *name, Ifl_desc *ifl, Shdr *shdr, Elf_Scn *scn, Word ndx, int ident, Ofl_desc *ofl) { if (process_input(name, ifl, shdr, scn, ndx, ident, ofl) == S_ERROR) return (S_ERROR); /* * Have we already seen the related SYMTAB - if so verify it now. */ if (shdr->sh_link < ndx) { Is_desc *isp = ifl->ifl_isdesc[shdr->sh_link]; if ((isp == NULL) || ((isp->is_shdr->sh_type != SHT_SYMTAB) && (isp->is_shdr->sh_type != SHT_DYNSYM))) { ld_eprintf(ofl, ERR_FATAL, MSG_INTL(MSG_FIL_INVSHLINK), ifl->ifl_name, EC_WORD(ndx), name, EC_XWORD(shdr->sh_link)); return (S_ERROR); } isp->is_symshndx = ifl->ifl_isdesc[ndx]; } return (1); } /* * Final processing for SHT_SYMTAB_SHNDX section. */ static uintptr_t /* ARGSUSED2 */ sym_shndx_process(Is_desc *isc, Ifl_desc *ifl, Ofl_desc *ofl) { if (isc->is_shdr->sh_link > isc->is_scnndx) { Is_desc *isp = ifl->ifl_isdesc[isc->is_shdr->sh_link]; if ((isp == NULL) || ((isp->is_shdr->sh_type != SHT_SYMTAB) && (isp->is_shdr->sh_type != SHT_DYNSYM))) { ld_eprintf(ofl, ERR_FATAL, MSG_INTL(MSG_FIL_INVSHLINK), isc->is_file->ifl_name, EC_WORD(isc->is_scnndx), isc->is_name, EC_XWORD(isc->is_shdr->sh_link)); return (S_ERROR); } isp->is_symshndx = isc; } return (1); } /* * Process .dynamic section from a relocatable object. * * Note: That the .dynamic section is only considered interesting when * dlopen()ing a relocatable object (thus FLG_OF1_RELDYN can only get * set when libld is called from ld.so.1). */ /*ARGSUSED*/ static uintptr_t process_rel_dynamic(const char *name, Ifl_desc *ifl, Shdr *shdr, Elf_Scn *scn, Word ndx, int ident, Ofl_desc *ofl) { Dyn *dyn; Elf_Scn *strscn; Elf_Data *dp; char *str; /* * Process .dynamic sections from relocatable objects ? */ if ((ofl->ofl_flags1 & FLG_OF1_RELDYN) == 0) return (1); /* * Find the string section associated with the .dynamic section. */ if ((strscn = elf_getscn(ifl->ifl_elf, shdr->sh_link)) == NULL) { ld_eprintf(ofl, ERR_ELF, MSG_INTL(MSG_ELF_GETSCN), ifl->ifl_name); return (0); } dp = elf_getdata(strscn, NULL); str = (char *)dp->d_buf; /* * And get the .dynamic data */ dp = elf_getdata(scn, NULL); for (dyn = (Dyn *)dp->d_buf; dyn->d_tag != DT_NULL; dyn++) { Ifl_desc *difl; switch (dyn->d_tag) { case DT_NEEDED: case DT_USED: if (((difl = libld_calloc(1, sizeof (Ifl_desc))) == NULL) || (aplist_append(&ofl->ofl_sos, difl, AL_CNT_OFL_LIBS) == NULL)) return (S_ERROR); difl->ifl_name = MSG_ORIG(MSG_STR_DYNAMIC); difl->ifl_soname = str + (size_t)dyn->d_un.d_val; difl->ifl_flags = FLG_IF_NEEDSTR; break; case DT_RPATH: case DT_RUNPATH: if ((ofl->ofl_rpath = add_string(ofl->ofl_rpath, (str + (size_t)dyn->d_un.d_val))) == (const char *)S_ERROR) return (S_ERROR); break; case DT_VERSYM: /* * The Solaris ld does not put DT_VERSYM in the * dynamic section. If the object has DT_VERSYM, * then it must have been produced by the GNU ld, * and is using the GNU style of versioning. */ ifl->ifl_flags |= FLG_IF_GNUVER; break; } } return (1); } /* * Expand implicit references. Dependencies can be specified in terms of the * $ORIGIN, $MACHINE, $PLATFORM, $OSREL and $OSNAME tokens, either from their * needed name, or via a runpath. In addition runpaths may also specify the * $ISALIST token. * * Probably the most common reference to explicit dependencies (via -L) will be * sufficient to find any associated implicit dependencies, but just in case we * expand any occurrence of these known tokens here. * * Note, if any errors occur we simply return the original name. * * This code is remarkably similar to expand() in rtld/common/paths.c. */ static char *machine = NULL; static size_t machine_sz = 0; static char *platform = NULL; static size_t platform_sz = 0; static Isa_desc *isa = NULL; static Uts_desc *uts = NULL; static char * expand(const char *parent, const char *name, char **next) { char _name[PATH_MAX], *nptr, *_next; const char *optr; size_t nrem = PATH_MAX - 1; int expanded = 0, _expanded, isaflag = 0; optr = name; nptr = _name; while (*optr) { if (nrem == 0) return ((char *)name); if (*optr != '$') { *nptr++ = *optr++, nrem--; continue; } _expanded = 0; if (strncmp(optr, MSG_ORIG(MSG_STR_ORIGIN), MSG_STR_ORIGIN_SIZE) == 0) { char *eptr; /* * For $ORIGIN, expansion is really just a concatenation * of the parents directory name. For example, an * explicit dependency foo/bar/lib1.so with a dependency * on $ORIGIN/lib2.so would be expanded to * foo/bar/lib2.so. */ if ((eptr = strrchr(parent, '/')) == NULL) { *nptr++ = '.'; nrem--; } else { size_t len = eptr - parent; if (len >= nrem) return ((char *)name); (void) strncpy(nptr, parent, len); nptr = nptr + len; nrem -= len; } optr += MSG_STR_ORIGIN_SIZE; expanded = _expanded = 1; } else if (strncmp(optr, MSG_ORIG(MSG_STR_MACHINE), MSG_STR_MACHINE_SIZE) == 0) { /* * Establish the machine from sysconf - like uname -i. */ if ((machine == NULL) && (machine_sz == 0)) { char info[SYS_NMLN]; long size; size = sysinfo(SI_MACHINE, info, SYS_NMLN); if ((size != -1) && (machine = libld_malloc((size_t)size))) { (void) strcpy(machine, info); machine_sz = (size_t)size - 1; } else machine_sz = 1; } if (machine) { if (machine_sz >= nrem) return ((char *)name); (void) strncpy(nptr, machine, machine_sz); nptr = nptr + machine_sz; nrem -= machine_sz; optr += MSG_STR_MACHINE_SIZE; expanded = _expanded = 1; } } else if (strncmp(optr, MSG_ORIG(MSG_STR_PLATFORM), MSG_STR_PLATFORM_SIZE) == 0) { /* * Establish the platform from sysconf - like uname -i. */ if ((platform == NULL) && (platform_sz == 0)) { char info[SYS_NMLN]; long size; size = sysinfo(SI_PLATFORM, info, SYS_NMLN); if ((size != -1) && (platform = libld_malloc((size_t)size))) { (void) strcpy(platform, info); platform_sz = (size_t)size - 1; } else platform_sz = 1; } if (platform) { if (platform_sz >= nrem) return ((char *)name); (void) strncpy(nptr, platform, platform_sz); nptr = nptr + platform_sz; nrem -= platform_sz; optr += MSG_STR_PLATFORM_SIZE; expanded = _expanded = 1; } } else if (strncmp(optr, MSG_ORIG(MSG_STR_OSNAME), MSG_STR_OSNAME_SIZE) == 0) { /* * Establish the os name - like uname -s. */ if (uts == NULL) uts = conv_uts(); if (uts && uts->uts_osnamesz) { if (uts->uts_osnamesz >= nrem) return ((char *)name); (void) strncpy(nptr, uts->uts_osname, uts->uts_osnamesz); nptr = nptr + uts->uts_osnamesz; nrem -= uts->uts_osnamesz; optr += MSG_STR_OSNAME_SIZE; expanded = _expanded = 1; } } else if (strncmp(optr, MSG_ORIG(MSG_STR_OSREL), MSG_STR_OSREL_SIZE) == 0) { /* * Establish the os release - like uname -r. */ if (uts == NULL) uts = conv_uts(); if (uts && uts->uts_osrelsz) { if (uts->uts_osrelsz >= nrem) return ((char *)name); (void) strncpy(nptr, uts->uts_osrel, uts->uts_osrelsz); nptr = nptr + uts->uts_osrelsz; nrem -= uts->uts_osrelsz; optr += MSG_STR_OSREL_SIZE; expanded = _expanded = 1; } } else if ((strncmp(optr, MSG_ORIG(MSG_STR_ISALIST), MSG_STR_ISALIST_SIZE) == 0) && next && (isaflag++ == 0)) { /* * Establish instruction sets from sysconf. Note that * this is only meaningful from runpaths. */ if (isa == NULL) isa = conv_isalist(); if (isa && isa->isa_listsz && (nrem > isa->isa_opt->isa_namesz)) { size_t mlen, tlen, hlen = optr - name; size_t no; char *lptr; Isa_opt *opt = isa->isa_opt; (void) strncpy(nptr, opt->isa_name, opt->isa_namesz); nptr = nptr + opt->isa_namesz; nrem -= opt->isa_namesz; optr += MSG_STR_ISALIST_SIZE; expanded = _expanded = 1; tlen = strlen(optr); /* * As ISALIST expands to a number of elements, * establish a new list to return to the caller. * This will contain the present path being * processed redefined for each isalist option, * plus the original remaining list entries. */ mlen = ((hlen + tlen) * (isa->isa_optno - 1)) + isa->isa_listsz - opt->isa_namesz; if (*next) mlen += strlen(*next); if ((_next = lptr = libld_malloc(mlen)) == NULL) return (0); for (no = 1, opt++; no < isa->isa_optno; no++, opt++) { (void) strncpy(lptr, name, hlen); lptr = lptr + hlen; (void) strncpy(lptr, opt->isa_name, opt->isa_namesz); lptr = lptr + opt->isa_namesz; (void) strncpy(lptr, optr, tlen); lptr = lptr + tlen; *lptr++ = ':'; } if (*next) (void) strcpy(lptr, *next); else *--lptr = '\0'; } } /* * If no expansion occurred skip the $ and continue. */ if (_expanded == 0) *nptr++ = *optr++, nrem--; } /* * If any ISALIST processing has occurred not only do we return the * expanded node we're presently working on, but we must also update the * remaining list so that it is effectively prepended with this node * expanded to all remaining isalist options. Note that we can only * handle one ISALIST per node. For more than one ISALIST to be * processed we'd need a better algorithm than above to replace the * newly generated list. Whether we want to encourage the number of * pathname permutations this would provide is another question. So, for * now if more than one ISALIST is encountered we return the original * node untouched. */ if (isaflag) { if (isaflag == 1) *next = _next; else return ((char *)name); } *nptr = '\0'; if (expanded) { if ((nptr = libld_malloc(strlen(_name) + 1)) == NULL) return ((char *)name); (void) strcpy(nptr, _name); return (nptr); } return ((char *)name); } /* * The Solaris ld does not put DT_VERSYM in the dynamic section, but the * GNU ld does, and it is used by the runtime linker to implement their * versioning scheme. Use this fact to determine if the sharable object * was produced by the GNU ld rather than the Solaris one, and to set * FLG_IF_GNUVER if so. This needs to be done before the symbols are * processed, since the answer determines whether we interpret the * symbols versions according to Solaris or GNU rules. */ /*ARGSUSED*/ static uintptr_t process_dynamic_isgnu(const char *name, Ifl_desc *ifl, Shdr *shdr, Elf_Scn *scn, Word ndx, int ident, Ofl_desc *ofl) { Dyn *dyn; Elf_Data *dp; uintptr_t error; error = process_section(name, ifl, shdr, scn, ndx, ident, ofl); if ((error == 0) || (error == S_ERROR)) return (error); /* Get the .dynamic data */ dp = elf_getdata(scn, NULL); for (dyn = (Dyn *)dp->d_buf; dyn->d_tag != DT_NULL; dyn++) { if (dyn->d_tag == DT_VERSYM) { ifl->ifl_flags |= FLG_IF_GNUVER; break; } } return (1); } /* * Process a dynamic section. If we are processing an explicit shared object * then we need to determine if it has a recorded SONAME, if so, this name will * be recorded in the output file being generated as the NEEDED entry rather * than the shared objects filename itself. * If the mode of the link-edit indicates that no undefined symbols should * remain, then we also need to build up a list of any additional shared object * dependencies this object may have. In this case save any NEEDED entries * together with any associated run-path specifications. This information is * recorded on the `ofl_soneed' list and will be analyzed after all explicit * file processing has been completed (refer finish_libs()). */ static uintptr_t process_dynamic(Is_desc *isc, Ifl_desc *ifl, Ofl_desc *ofl) { Dyn *data, *dyn; char *str, *rpath = NULL; const char *soname, *needed; Boolean no_undef; data = (Dyn *)isc->is_indata->d_buf; str = (char *)ifl->ifl_isdesc[isc->is_shdr->sh_link]->is_indata->d_buf; /* Determine if we need to examine the runpaths and NEEDED entries */ no_undef = (ofl->ofl_flags & (FLG_OF_NOUNDEF | FLG_OF_SYMBOLIC)) || OFL_GUIDANCE(ofl, FLG_OFG_NO_DEFS); /* * First loop through the dynamic section looking for a run path. */ if (no_undef) { for (dyn = data; dyn->d_tag != DT_NULL; dyn++) { if ((dyn->d_tag != DT_RPATH) && (dyn->d_tag != DT_RUNPATH)) continue; if ((rpath = str + (size_t)dyn->d_un.d_val) == NULL) continue; break; } } /* * Now look for any needed dependencies (which may use the rpath) * or a new SONAME. */ for (dyn = data; dyn->d_tag != DT_NULL; dyn++) { if (dyn->d_tag == DT_SONAME) { if ((soname = str + (size_t)dyn->d_un.d_val) == NULL) continue; /* * Update the input file structure with this new name. */ ifl->ifl_soname = soname; } else if ((dyn->d_tag == DT_NEEDED) || (dyn->d_tag == DT_USED)) { Sdf_desc *sdf; if (!no_undef) continue; if ((needed = str + (size_t)dyn->d_un.d_val) == NULL) continue; /* * Determine if this needed entry is already recorded on * the shared object needed list, if not create a new * definition for later processing (see finish_libs()). */ needed = expand(ifl->ifl_name, needed, NULL); if ((sdf = sdf_find(needed, ofl->ofl_soneed)) == NULL) { if ((sdf = sdf_add(needed, &ofl->ofl_soneed)) == (Sdf_desc *)S_ERROR) return (S_ERROR); sdf->sdf_rfile = ifl->ifl_name; } /* * Record the runpath (Note that we take the first * runpath which is exactly what ld.so.1 would do during * its dependency processing). */ if (rpath && (sdf->sdf_rpath == NULL)) sdf->sdf_rpath = rpath; } else if (dyn->d_tag == DT_FLAGS_1) { if (dyn->d_un.d_val & (DF_1_INITFIRST | DF_1_INTERPOSE)) ifl->ifl_flags &= ~FLG_IF_LAZYLD; if (dyn->d_un.d_val & DF_1_DISPRELPND) ifl->ifl_flags |= FLG_IF_DISPPEND; if (dyn->d_un.d_val & DF_1_DISPRELDNE) ifl->ifl_flags |= FLG_IF_DISPDONE; if (dyn->d_un.d_val & DF_1_NODIRECT) ifl->ifl_flags |= FLG_IF_NODIRECT; /* * If we are building an executable, and this * dependency is tagged as an interposer, then * assume that it is required even if symbol * resolution uncovers no evident use. * * If we are building a shared object, then an * interposer dependency has no special meaning, and we * treat it as a regular dependency. By definition, all * interposers must be visible to the runtime linker * at initialization time, and cannot be added later. */ if ((dyn->d_un.d_val & DF_1_INTERPOSE) && (ofl->ofl_flags & FLG_OF_EXEC)) ifl->ifl_flags |= FLG_IF_DEPREQD; } else if ((dyn->d_tag == DT_AUDIT) && (ifl->ifl_flags & FLG_IF_NEEDED)) { /* * Record audit string as DT_DEPAUDIT. */ if ((ofl->ofl_depaudit = add_string(ofl->ofl_depaudit, (str + (size_t)dyn->d_un.d_val))) == (const char *)S_ERROR) return (S_ERROR); } else if (dyn->d_tag == DT_SUNW_RTLDINF) { /* * If this dependency has the DT_SUNW_RTLDINF .dynamic * entry, then ensure no specialized dependency * processing is in effect. This tag identifies libc, * which provides critical startup information (TLS * routines, threads initialization, etc.) that must * be exercised as part of process initialization. */ ifl->ifl_flags &= ~MSK_IF_POSFLAG1; /* * libc is not subject to the usual guidance checks * for lazy loading. It cannot be lazy loaded, libld * ignores the request, and rtld would ignore the * setting if it were present. */ ifl->ifl_flags |= FLG_IF_RTLDINF; } } /* * Perform some SONAME sanity checks. */ if (ifl->ifl_flags & FLG_IF_NEEDED) { Ifl_desc *sifl; Aliste idx; /* * Determine if anyone else will cause the same SONAME to be * used (this is either caused by two different files having the * same SONAME, or by one file SONAME actually matching another * file basename (if no SONAME is specified within a shared * library its basename will be used)). Probably rare, but some * idiot will do it. */ for (APLIST_TRAVERSE(ofl->ofl_sos, idx, sifl)) { if ((strcmp(ifl->ifl_soname, sifl->ifl_soname) == 0) && (ifl != sifl)) { const char *hint, *iflb, *siflb; /* * Determine the basename of each file. Perhaps * there are multiple copies of the same file * being brought in using different -L search * paths, and if so give an extra hint in the * error message. */ iflb = strrchr(ifl->ifl_name, '/'); if (iflb == NULL) iflb = ifl->ifl_name; else iflb++; siflb = strrchr(sifl->ifl_name, '/'); if (siflb == NULL) siflb = sifl->ifl_name; else siflb++; if (strcmp(iflb, siflb) == 0) hint = MSG_INTL(MSG_REC_CNFLTHINT); else hint = MSG_ORIG(MSG_STR_EMPTY); ld_eprintf(ofl, ERR_FATAL, MSG_INTL(MSG_REC_OBJCNFLT), sifl->ifl_name, ifl->ifl_name, sifl->ifl_soname, hint); return (0); } } /* * If the SONAME is the same as the name the user wishes to * record when building a dynamic library (refer -h option), * we also have a name clash. */ if (ofl->ofl_soname && (strcmp(ofl->ofl_soname, ifl->ifl_soname) == 0)) { ld_eprintf(ofl, ERR_FATAL, MSG_INTL(MSG_REC_OPTCNFLT), ifl->ifl_name, MSG_INTL(MSG_MARG_SONAME), ifl->ifl_soname); return (0); } } return (1); } /* * Process a progbits section from a relocatable object (ET_REL). * This is used on non-amd64 objects to recognize .eh_frame sections. */ /*ARGSUSED1*/ static uintptr_t process_progbits_final(Is_desc *isc, Ifl_desc *ifl, Ofl_desc *ofl) { if (isc->is_osdesc && (isc->is_flags & FLG_IS_EHFRAME) && (ld_unwind_register(isc->is_osdesc, ofl) == S_ERROR)) return (S_ERROR); return (1); } /* * Process a group section. */ static uintptr_t process_group(const char *name, Ifl_desc *ifl, Shdr *shdr, Elf_Scn *scn, Word ndx, int ident, Ofl_desc *ofl) { uintptr_t error; error = process_section(name, ifl, shdr, scn, ndx, ident, ofl); if ((error == 0) || (error == S_ERROR)) return (error); /* * Indicate that this input file has groups to process. Groups are * processed after all input sections have been processed. */ ifl->ifl_flags |= FLG_IS_GROUPS; return (1); } /* * Process a relocation entry. At this point all input sections from this * input file have been assigned an input section descriptor which is saved * in the `ifl_isdesc' array. */ static uintptr_t rel_process(Is_desc *isc, Ifl_desc *ifl, Ofl_desc *ofl) { Word rndx; Is_desc *risc; Os_desc *osp; Shdr *shdr = isc->is_shdr; Conv_inv_buf_t inv_buf; /* * Make sure this is a valid relocation we can handle. */ if (shdr->sh_type != ld_targ.t_m.m_rel_sht_type) { ld_eprintf(ofl, ERR_FATAL, MSG_INTL(MSG_FIL_INVALSEC), ifl->ifl_name, EC_WORD(isc->is_scnndx), isc->is_name, conv_sec_type(ifl->ifl_ehdr->e_ident[EI_OSABI], ifl->ifl_ehdr->e_machine, shdr->sh_type, 0, &inv_buf)); return (0); } /* * From the relocation section header information determine which * section needs the actual relocation. Determine which output section * this input section has been assigned to and add to its relocation * list. Note that the relocation section may be null if it is not * required (ie. .debug, .stabs, etc). */ rndx = shdr->sh_info; if (rndx >= ifl->ifl_shnum) { /* * Broken input file. */ ld_eprintf(ofl, ERR_FATAL, MSG_INTL(MSG_FIL_INVSHINFO), ifl->ifl_name, EC_WORD(isc->is_scnndx), isc->is_name, EC_XWORD(rndx)); return (0); } if (rndx == 0) { if (aplist_append(&ofl->ofl_extrarels, isc, AL_CNT_OFL_RELS) == NULL) return (S_ERROR); } else if ((risc = ifl->ifl_isdesc[rndx]) != NULL) { /* * Discard relocations if they are against a section * which has been discarded. */ if (risc->is_flags & FLG_IS_DISCARD) return (1); if ((osp = risc->is_osdesc) == NULL) { if (risc->is_shdr->sh_type == SHT_SUNW_move) { /* * This section is processed later in * process_movereloc(). */ if (aplist_append(&ofl->ofl_ismoverel, isc, AL_CNT_OFL_MOVE) == NULL) return (S_ERROR); return (1); } ld_eprintf(ofl, ERR_FATAL, MSG_INTL(MSG_FIL_INVRELOC1), ifl->ifl_name, EC_WORD(isc->is_scnndx), isc->is_name, EC_WORD(risc->is_scnndx), risc->is_name); return (0); } if (aplist_append(&osp->os_relisdescs, isc, AL_CNT_OS_RELISDESCS) == NULL) return (S_ERROR); } return (1); } /* * SHF_EXCLUDE flags is set for this section. */ static uintptr_t process_exclude(const char *name, Ifl_desc *ifl, Shdr *shdr, Elf_Scn *scn, Word ndx, Ofl_desc *ofl) { /* * Sections SHT_SYMTAB and SHT_DYNDYM, even if SHF_EXCLUDE is on, might * be needed for ld processing. These sections need to be in the * internal table. Later it will be determined whether they can be * eliminated or not. */ if (shdr->sh_type == SHT_SYMTAB || shdr->sh_type == SHT_DYNSYM) return (0); /* * Other checks */ if (shdr->sh_flags & SHF_ALLOC) { /* * A conflict, issue an warning message, and ignore the section. */ ld_eprintf(ofl, ERR_WARNING, MSG_INTL(MSG_FIL_EXCLUDE), ifl->ifl_name, EC_WORD(ndx), name); return (0); } /* * This sections is not going to the output file. */ return (process_section(name, ifl, shdr, scn, ndx, 0, ofl)); } /* * Section processing state table. `Initial' describes the required initial * procedure to be called (if any), `Final' describes the final processing * procedure (ie. things that can only be done when all required sections * have been collected). */ typedef uintptr_t (* initial_func_t)(const char *, Ifl_desc *, Shdr *, Elf_Scn *, Word, int, Ofl_desc *); static initial_func_t Initial[SHT_NUM][2] = { /* ET_REL ET_DYN */ /* SHT_NULL */ invalid_section, invalid_section, /* SHT_PROGBITS */ process_progbits, process_progbits, /* SHT_SYMTAB */ process_input, process_input, /* SHT_STRTAB */ process_strtab, process_strtab, /* SHT_RELA */ process_reloc, process_reloc, /* SHT_HASH */ invalid_section, NULL, /* SHT_DYNAMIC */ process_rel_dynamic, process_dynamic_isgnu, /* SHT_NOTE */ process_section, NULL, /* SHT_NOBITS */ process_nobits, process_nobits, /* SHT_REL */ process_reloc, process_reloc, /* SHT_SHLIB */ process_section, invalid_section, /* SHT_DYNSYM */ invalid_section, process_input, /* SHT_UNKNOWN12 */ process_progbits, process_progbits, /* SHT_UNKNOWN13 */ process_progbits, process_progbits, /* SHT_INIT_ARRAY */ process_array, NULL, /* SHT_FINI_ARRAY */ process_array, NULL, /* SHT_PREINIT_ARRAY */ process_array, NULL, /* SHT_GROUP */ process_group, invalid_section, /* SHT_SYMTAB_SHNDX */ process_sym_shndx, NULL }; typedef uintptr_t (* final_func_t)(Is_desc *, Ifl_desc *, Ofl_desc *); static final_func_t Final[SHT_NUM][2] = { /* ET_REL ET_DYN */ /* SHT_NULL */ NULL, NULL, /* SHT_PROGBITS */ process_progbits_final, NULL, /* SHT_SYMTAB */ ld_sym_process, ld_sym_process, /* SHT_STRTAB */ NULL, NULL, /* SHT_RELA */ rel_process, NULL, /* SHT_HASH */ NULL, NULL, /* SHT_DYNAMIC */ NULL, process_dynamic, /* SHT_NOTE */ NULL, NULL, /* SHT_NOBITS */ NULL, NULL, /* SHT_REL */ rel_process, NULL, /* SHT_SHLIB */ NULL, NULL, /* SHT_DYNSYM */ NULL, ld_sym_process, /* SHT_UNKNOWN12 */ NULL, NULL, /* SHT_UNKNOWN13 */ NULL, NULL, /* SHT_INIT_ARRAY */ array_process, NULL, /* SHT_FINI_ARRAY */ array_process, NULL, /* SHT_PREINIT_ARRAY */ array_process, NULL, /* SHT_GROUP */ NULL, NULL, /* SHT_SYMTAB_SHNDX */ sym_shndx_process, NULL }; #define MAXNDXSIZE 10 /* * Process an elf file. Each section is compared against the section state * table to determine whether it should be processed (saved), ignored, or * is invalid for the type of input file being processed. */ static uintptr_t process_elf(Ifl_desc *ifl, Elf *elf, Ofl_desc *ofl) { Elf_Scn *scn; Shdr *shdr; Word ndx, sndx, ordndx = 0, ordcnt = 0; char *str, *name; Word row, column; int ident; uintptr_t error; Is_desc *vdfisp, *vndisp, *vsyisp, *sifisp; Is_desc *capinfoisp, *capisp; Sdf_desc *sdf; Place_path_info path_info_buf, *path_info; /* * Path information buffer used by ld_place_section() and related * routines. This information is used to evaluate entrance criteria * with non-empty file matching lists (ec_files). */ path_info = ld_place_path_info_init(ofl, ifl, &path_info_buf); /* * First process the .shstrtab section so that later sections can * reference their name. */ ld_sup_file(ofl, ifl->ifl_name, elf_kind(elf), ifl->ifl_flags, elf); sndx = ifl->ifl_shstrndx; if ((scn = elf_getscn(elf, (size_t)sndx)) == NULL) { ld_eprintf(ofl, ERR_ELF, MSG_INTL(MSG_ELF_GETSCN), ifl->ifl_name); return (0); } if ((shdr = elf_getshdr(scn)) == NULL) { ld_eprintf(ofl, ERR_ELF, MSG_INTL(MSG_ELF_GETSHDR), ifl->ifl_name); return (0); } if ((name = elf_strptr(elf, (size_t)sndx, (size_t)shdr->sh_name)) == NULL) { ld_eprintf(ofl, ERR_ELF, MSG_INTL(MSG_ELF_STRPTR), ifl->ifl_name); return (0); } if (ld_sup_input_section(ofl, ifl, name, &shdr, sndx, scn, elf) == S_ERROR) return (S_ERROR); /* * Reset the name since the shdr->sh_name could have been changed as * part of ld_sup_input_section(). */ if ((name = elf_strptr(elf, (size_t)sndx, (size_t)shdr->sh_name)) == NULL) { ld_eprintf(ofl, ERR_ELF, MSG_INTL(MSG_ELF_STRPTR), ifl->ifl_name); return (0); } error = process_strtab(name, ifl, shdr, scn, sndx, FALSE, ofl); if ((error == 0) || (error == S_ERROR)) return (error); str = ifl->ifl_isdesc[sndx]->is_indata->d_buf; /* * Determine the state table column from the input file type. Note, * shared library sections are not added to the output section list. */ if (ifl->ifl_ehdr->e_type == ET_DYN) { column = 1; ofl->ofl_soscnt++; ident = ld_targ.t_id.id_null; } else { column = 0; ofl->ofl_objscnt++; ident = ld_targ.t_id.id_unknown; } DBG_CALL(Dbg_file_generic(ofl->ofl_lml, ifl)); ndx = 0; vdfisp = vndisp = vsyisp = sifisp = capinfoisp = capisp = NULL; scn = NULL; while (scn = elf_nextscn(elf, scn)) { ndx++; /* * As we've already processed the .shstrtab don't do it again. */ if (ndx == sndx) continue; if ((shdr = elf_getshdr(scn)) == NULL) { ld_eprintf(ofl, ERR_ELF, MSG_INTL(MSG_ELF_GETSHDR), ifl->ifl_name); return (0); } name = str + (size_t)(shdr->sh_name); if (ld_sup_input_section(ofl, ifl, name, &shdr, ndx, scn, elf) == S_ERROR) return (S_ERROR); /* * Reset the name since the shdr->sh_name could have been * changed as part of ld_sup_input_section(). */ name = str + (size_t)(shdr->sh_name); row = shdr->sh_type; /* * If the section has the SHF_EXCLUDE flag on, and we're not * generating a relocatable object, exclude the section. */ if (((shdr->sh_flags & SHF_EXCLUDE) != 0) && ((ofl->ofl_flags & FLG_OF_RELOBJ) == 0)) { if ((error = process_exclude(name, ifl, shdr, scn, ndx, ofl)) == S_ERROR) return (S_ERROR); if (error == 1) continue; } /* * If this is a standard section type process it via the * appropriate action routine. */ if (row < SHT_NUM) { if (Initial[row][column] != NULL) { if (Initial[row][column](name, ifl, shdr, scn, ndx, ident, ofl) == S_ERROR) return (S_ERROR); } } else { /* * If this section is below SHT_LOSUNW then we don't * really know what to do with it, issue a warning * message but do the basic section processing anyway. */ if (row < (Word)SHT_LOSUNW) { Conv_inv_buf_t inv_buf; ld_eprintf(ofl, ERR_WARNING, MSG_INTL(MSG_FIL_INVALSEC), ifl->ifl_name, EC_WORD(ndx), name, conv_sec_type( ifl->ifl_ehdr->e_ident[EI_OSABI], ifl->ifl_ehdr->e_machine, shdr->sh_type, 0, &inv_buf)); } /* * Handle sections greater than SHT_LOSUNW. */ switch (row) { case SHT_SUNW_dof: if (process_section(name, ifl, shdr, scn, ndx, ident, ofl) == S_ERROR) return (S_ERROR); break; case SHT_SUNW_cap: if (process_section(name, ifl, shdr, scn, ndx, ld_targ.t_id.id_null, ofl) == S_ERROR) return (S_ERROR); capisp = ifl->ifl_isdesc[ndx]; break; case SHT_SUNW_capinfo: if (process_section(name, ifl, shdr, scn, ndx, ld_targ.t_id.id_null, ofl) == S_ERROR) return (S_ERROR); capinfoisp = ifl->ifl_isdesc[ndx]; break; case SHT_SUNW_DEBUGSTR: case SHT_SUNW_DEBUG: if (process_debug(name, ifl, shdr, scn, ndx, ident, ofl) == S_ERROR) return (S_ERROR); break; case SHT_SUNW_move: if (process_section(name, ifl, shdr, scn, ndx, ld_targ.t_id.id_null, ofl) == S_ERROR) return (S_ERROR); break; case SHT_SUNW_syminfo: if (process_section(name, ifl, shdr, scn, ndx, ld_targ.t_id.id_null, ofl) == S_ERROR) return (S_ERROR); sifisp = ifl->ifl_isdesc[ndx]; break; case SHT_SUNW_ANNOTATE: if (process_progbits(name, ifl, shdr, scn, ndx, ident, ofl) == S_ERROR) return (S_ERROR); break; case SHT_SUNW_COMDAT: if (process_progbits(name, ifl, shdr, scn, ndx, ident, ofl) == S_ERROR) return (S_ERROR); ifl->ifl_isdesc[ndx]->is_flags |= FLG_IS_COMDAT; break; case SHT_SUNW_verdef: if (process_section(name, ifl, shdr, scn, ndx, ld_targ.t_id.id_null, ofl) == S_ERROR) return (S_ERROR); vdfisp = ifl->ifl_isdesc[ndx]; break; case SHT_SUNW_verneed: if (process_section(name, ifl, shdr, scn, ndx, ld_targ.t_id.id_null, ofl) == S_ERROR) return (S_ERROR); vndisp = ifl->ifl_isdesc[ndx]; break; case SHT_SUNW_versym: if (process_section(name, ifl, shdr, scn, ndx, ld_targ.t_id.id_null, ofl) == S_ERROR) return (S_ERROR); vsyisp = ifl->ifl_isdesc[ndx]; break; case SHT_SPARC_GOTDATA: /* * SHT_SPARC_GOTDATA (0x70000000) is in the * SHT_LOPROC - SHT_HIPROC range reserved * for processor-specific semantics. It is * only meaningful for sparc targets. */ if (ld_targ.t_m.m_mach != LD_TARG_BYCLASS(EM_SPARC, EM_SPARCV9)) goto do_default; if (process_section(name, ifl, shdr, scn, ndx, ld_targ.t_id.id_gotdata, ofl) == S_ERROR) return (S_ERROR); break; #if defined(_ELF64) case SHT_AMD64_UNWIND: /* * SHT_AMD64_UNWIND (0x70000001) is in the * SHT_LOPROC - SHT_HIPROC range reserved * for processor-specific semantics. It is * only meaningful for amd64 targets. */ if (ld_targ.t_m.m_mach != EM_AMD64) goto do_default; /* * Target is x86, so this really is * SHT_AMD64_UNWIND */ if (column == 0) { /* * column == ET_REL */ if (process_section(name, ifl, shdr, scn, ndx, ld_targ.t_id.id_unwind, ofl) == S_ERROR) return (S_ERROR); ifl->ifl_isdesc[ndx]->is_flags |= FLG_IS_EHFRAME; } break; #endif default: do_default: if (process_section(name, ifl, shdr, scn, ndx, ((ident == ld_targ.t_id.id_null) ? ident : ld_targ.t_id.id_user), ofl) == S_ERROR) return (S_ERROR); break; } } } /* * Now that all input sections have been analyzed, and prior to placing * any input sections to their output sections, process any groups. * Groups can contribute COMDAT items, which may get discarded as part * of placement. In addition, COMDAT names may require transformation * to indicate different output section placement. */ if (ifl->ifl_flags & FLG_IS_GROUPS) { for (ndx = 1; ndx < ifl->ifl_shnum; ndx++) { Is_desc *isp; if (((isp = ifl->ifl_isdesc[ndx]) == NULL) || (isp->is_shdr->sh_type != SHT_GROUP)) continue; if (ld_group_process(isp, ofl) == S_ERROR) return (S_ERROR); } } /* * Now that all of the input sections have been processed, place * them in the appropriate output sections. */ for (ndx = 1; ndx < ifl->ifl_shnum; ndx++) { Is_desc *isp; if (((isp = ifl->ifl_isdesc[ndx]) == NULL) || ((isp->is_flags & FLG_IS_PLACE) == 0)) continue; /* * Place all non-ordered sections within their appropriate * output section. */ if ((isp->is_flags & FLG_IS_ORDERED) == 0) { if (ld_place_section(ofl, isp, path_info, isp->is_keyident, NULL) == (Os_desc *)S_ERROR) return (S_ERROR); continue; } /* * Count the number of ordered sections and retain the first * ordered section index. This will be used to optimize the * ordered section loop that immediately follows this one. */ ordcnt++; if (ordndx == 0) ordndx = ndx; } /* * Having placed all the non-ordered sections, it is now * safe to place SHF_ORDERED/SHF_LINK_ORDER sections. */ if (ifl->ifl_flags & FLG_IF_ORDERED) { for (ndx = ordndx; ndx < ifl->ifl_shnum; ndx++) { Is_desc *isp; if (((isp = ifl->ifl_isdesc[ndx]) == NULL) || ((isp->is_flags & (FLG_IS_PLACE | FLG_IS_ORDERED)) != (FLG_IS_PLACE | FLG_IS_ORDERED))) continue; /* ld_process_ordered() calls ld_place_section() */ if (ld_process_ordered(ofl, ifl, path_info, ndx) == S_ERROR) return (S_ERROR); /* If we've done them all, stop searching */ if (--ordcnt == 0) break; } } /* * If this is a shared object explicitly specified on the command * line (as opposed to being a dependency of such an object), * determine if the user has specified a control definition. This * descriptor may specify which version definitions can be used * from this object. It may also update the dependency to USED and * supply an alternative SONAME. */ sdf = NULL; if (column && (ifl->ifl_flags & FLG_IF_NEEDED)) { const char *base; /* * Use the basename of the input file (typically this is the * compilation environment name, ie. libfoo.so). */ if ((base = strrchr(ifl->ifl_name, '/')) == NULL) base = ifl->ifl_name; else base++; if ((sdf = sdf_find(base, ofl->ofl_socntl)) != NULL) { sdf->sdf_file = ifl; ifl->ifl_sdfdesc = sdf; } } /* * Before symbol processing, process any capabilities. Capabilities * can reference a string table, which is why this processing is * carried out after the initial section processing. Capabilities, * together with -z symbolcap, can require the conversion of global * symbols to local symbols. */ if (capisp && (process_cap(ofl, ifl, capisp) == S_ERROR)) return (S_ERROR); /* * Process any version dependencies. These will establish shared object * `needed' entries in the same manner as will be generated from the * .dynamic's NEEDED entries. */ if (vndisp && ((ofl->ofl_flags & (FLG_OF_NOUNDEF | FLG_OF_SYMBOLIC)) || OFL_GUIDANCE(ofl, FLG_OFG_NO_DEFS))) if (ld_vers_need_process(vndisp, ifl, ofl) == S_ERROR) return (S_ERROR); /* * Before processing any symbol resolution or relocations process any * version sections. */ if (vsyisp) (void) ld_vers_sym_process(ofl, vsyisp, ifl); if (ifl->ifl_versym && (vdfisp || (sdf && (sdf->sdf_flags & FLG_SDF_SELECT)))) if (ld_vers_def_process(vdfisp, ifl, ofl) == S_ERROR) return (S_ERROR); /* * Having collected the appropriate sections carry out any additional * processing if necessary. */ for (ndx = 0; ndx < ifl->ifl_shnum; ndx++) { Is_desc *isp; if ((isp = ifl->ifl_isdesc[ndx]) == NULL) continue; row = isp->is_shdr->sh_type; if ((isp->is_flags & FLG_IS_DISCARD) == 0) ld_sup_section(ofl, isp->is_name, isp->is_shdr, ndx, isp->is_indata, elf); /* * If this is a SHT_SUNW_move section from a relocatable file, * keep track of the section for later processing. */ if ((row == SHT_SUNW_move) && (column == 0)) { if (aplist_append(&(ofl->ofl_ismove), isp, AL_CNT_OFL_MOVE) == NULL) return (S_ERROR); } /* * If this is a standard section type process it via the * appropriate action routine. */ if (row < SHT_NUM) { if (Final[row][column] != NULL) { if (Final[row][column](isp, ifl, ofl) == S_ERROR) return (S_ERROR); } #if defined(_ELF64) } else if ((row == SHT_AMD64_UNWIND) && (column == 0)) { Os_desc *osp = isp->is_osdesc; /* * SHT_AMD64_UNWIND (0x70000001) is in the SHT_LOPROC - * SHT_HIPROC range reserved for processor-specific * semantics, and is only meaningful for amd64 targets. * * Only process unwind contents from relocatable * objects. */ if (osp && (ld_targ.t_m.m_mach == EM_AMD64) && (ld_unwind_register(osp, ofl) == S_ERROR)) return (S_ERROR); #endif } } /* * Following symbol processing, if this relocatable object input file * provides symbol capabilities, tag the associated symbols so that * the symbols can be re-assigned to the new capabilities symbol * section that will be created for the output file. */ if (capinfoisp && (ifl->ifl_ehdr->e_type == ET_REL) && (process_capinfo(ofl, ifl, capinfoisp) == S_ERROR)) return (S_ERROR); /* * After processing any symbol resolution, and if this dependency * indicates it contains symbols that can't be directly bound to, * set the symbols appropriately. */ if (sifisp && ((ifl->ifl_flags & (FLG_IF_NEEDED | FLG_IF_NODIRECT)) == (FLG_IF_NEEDED | FLG_IF_NODIRECT))) (void) ld_sym_nodirect(sifisp, ifl, ofl); return (1); } /* * Process the current input file. There are basically three types of files * that come through here: * * - files explicitly defined on the command line (ie. foo.o or bar.so), * in this case only the `name' field is valid. * * - libraries determined from the -l command line option (ie. -lbar), * in this case the `soname' field contains the basename of the located * file. * * Any shared object specified via the above two conventions must be recorded * as a needed dependency. * * - libraries specified as dependencies of those libraries already obtained * via the command line (ie. bar.so has a DT_NEEDED entry of fred.so.1), * in this case the `soname' field contains either a full pathname (if the * needed entry contained a `/'), or the basename of the located file. * These libraries are processed to verify symbol binding but are not * recorded as dependencies of the output file being generated. * * entry: * name - File name * soname - SONAME for needed sharable library, as described above * fd - Open file descriptor * elf - Open ELF handle * flags - FLG_IF_ flags applicable to file * ofl - Output file descriptor * rej - Rejection descriptor used to record rejection reason * ifl_ret - NULL, or address of pointer to receive reference to * resulting input descriptor for file. If ifl_ret is non-NULL, * the file cannot be an archive or it will be rejected. * * exit: * If a error occurs in examining the file, S_ERROR is returned. * If the file can be examined, but is not suitable, *rej is updated, * and 0 is returned. If the file is acceptable, 1 is returned, and if * ifl_ret is non-NULL, *ifl_ret is set to contain the pointer to the * resulting input descriptor. */ uintptr_t ld_process_ifl(const char *name, const char *soname, int fd, Elf *elf, Word flags, Ofl_desc *ofl, Rej_desc *rej, Ifl_desc **ifl_ret) { Ifl_desc *ifl; Ehdr *ehdr; uintptr_t error = 0; struct stat status; Ar_desc *adp; Rej_desc _rej; /* * If this file was not extracted from an archive obtain its device * information. This will be used to determine if the file has already * been processed (rather than simply comparing filenames, the device * information provides a quicker comparison and detects linked files). */ if (fd && ((flags & FLG_IF_EXTRACT) == 0)) (void) fstat(fd, &status); else { status.st_dev = 0; status.st_ino = 0; } switch (elf_kind(elf)) { case ELF_K_AR: /* * If the caller has supplied a non-NULL ifl_ret, then * we cannot process archives, for there will be no * input file descriptor for us to return. In this case, * reject the attempt. */ if (ifl_ret != NULL) { _rej.rej_type = SGS_REJ_ARCHIVE; _rej.rej_name = name; DBG_CALL(Dbg_file_rejected(ofl->ofl_lml, &_rej, ld_targ.t_m.m_mach)); if (rej->rej_type == 0) { *rej = _rej; rej->rej_name = strdup(_rej.rej_name); } return (0); } /* * Determine if we've already come across this archive file. */ if (!(flags & FLG_IF_EXTRACT)) { Aliste idx; for (APLIST_TRAVERSE(ofl->ofl_ars, idx, adp)) { if ((adp->ad_stdev != status.st_dev) || (adp->ad_stino != status.st_ino)) continue; /* * We've seen this file before so reuse the * original archive descriptor and discard the * new elf descriptor. Note that a file * descriptor is unnecessary, as the file is * already available in memory. */ DBG_CALL(Dbg_file_reuse(ofl->ofl_lml, name, adp->ad_name)); (void) elf_end(elf); if (!ld_process_archive(name, -1, adp, ofl)) return (S_ERROR); return (1); } } /* * As we haven't processed this file before establish a new * archive descriptor. */ adp = ld_ar_setup(name, elf, ofl); if ((adp == NULL) || (adp == (Ar_desc *)S_ERROR)) return ((uintptr_t)adp); adp->ad_stdev = status.st_dev; adp->ad_stino = status.st_ino; ld_sup_file(ofl, name, ELF_K_AR, flags, elf); /* * Indicate that the ELF descriptor no longer requires a file * descriptor by reading the entire file. The file is already * read via the initial mmap(2) behind elf_begin(3elf), thus * this operation is effectively a no-op. However, a side- * effect is that the internal file descriptor, maintained in * the ELF descriptor, is set to -1. This setting will not * be compared with any file descriptor that is passed to * elf_begin(), should this archive, or one of the archive * members, be processed again from the command line or * because of a -z rescan. */ if (elf_cntl(elf, ELF_C_FDREAD) == -1) { ld_eprintf(ofl, ERR_ELF, MSG_INTL(MSG_ELF_CNTL), name); return (0); } if (!ld_process_archive(name, -1, adp, ofl)) return (S_ERROR); return (1); case ELF_K_ELF: /* * Obtain the elf header so that we can determine what type of * elf ELF_K_ELF file this is. */ if ((ehdr = elf_getehdr(elf)) == NULL) { int _class = gelf_getclass(elf); /* * This can fail for a number of reasons. Typically * the object class is incorrect (ie. user is building * 64-bit but managed to point at 32-bit libraries). * Other ELF errors can include a truncated or corrupt * file. Try to get the best error message possible. */ if (ld_targ.t_m.m_class != _class) { _rej.rej_type = SGS_REJ_CLASS; _rej.rej_info = (uint_t)_class; } else { _rej.rej_type = SGS_REJ_STR; _rej.rej_str = elf_errmsg(-1); } _rej.rej_name = name; DBG_CALL(Dbg_file_rejected(ofl->ofl_lml, &_rej, ld_targ.t_m.m_mach)); if (rej->rej_type == 0) { *rej = _rej; rej->rej_name = strdup(_rej.rej_name); } return (0); } /* * Determine if we've already come across this file. */ if (!(flags & FLG_IF_EXTRACT)) { APlist *apl; Aliste idx; if (ehdr->e_type == ET_REL) apl = ofl->ofl_objs; else apl = ofl->ofl_sos; /* * Traverse the appropriate file list and determine if * a dev/inode match is found. */ for (APLIST_TRAVERSE(apl, idx, ifl)) { /* * Ifl_desc generated via -Nneed, therefore no * actual file behind it. */ if (ifl->ifl_flags & FLG_IF_NEEDSTR) continue; if ((ifl->ifl_stino != status.st_ino) || (ifl->ifl_stdev != status.st_dev)) continue; /* * Disregard (skip) this image. */ DBG_CALL(Dbg_file_skip(ofl->ofl_lml, ifl->ifl_name, name)); (void) elf_end(elf); /* * If the file was explicitly defined on the * command line (this is always the case for * relocatable objects, and is true for shared * objects when they weren't specified via -l or * were dragged in as an implicit dependency), * then warn the user. */ if ((flags & FLG_IF_CMDLINE) || (ifl->ifl_flags & FLG_IF_CMDLINE)) { const char *errmsg; /* * Determine whether this is the same * file name as originally encountered * so as to provide the most * descriptive diagnostic. */ errmsg = (strcmp(name, ifl->ifl_name) == 0) ? MSG_INTL(MSG_FIL_MULINC_1) : MSG_INTL(MSG_FIL_MULINC_2); ld_eprintf(ofl, ERR_WARNING, errmsg, name, ifl->ifl_name); } if (ifl_ret) *ifl_ret = ifl; return (1); } } /* * At this point, we know we need the file. Establish an input * file descriptor and continue processing. */ ifl = ifl_setup(name, ehdr, elf, flags, ofl, rej); if ((ifl == NULL) || (ifl == (Ifl_desc *)S_ERROR)) return ((uintptr_t)ifl); ifl->ifl_stdev = status.st_dev; ifl->ifl_stino = status.st_ino; /* * If -zignore is in effect, mark this file as a potential * candidate (the files use isn't actually determined until * symbol resolution and relocation processing are completed). */ if (ofl->ofl_flags1 & FLG_OF1_IGNORE) ifl->ifl_flags |= FLG_IF_IGNORE; switch (ehdr->e_type) { case ET_REL: (*ld_targ.t_mr.mr_mach_eflags)(ehdr, ofl); error = process_elf(ifl, elf, ofl); break; case ET_DYN: if ((ofl->ofl_flags & FLG_OF_STATIC) || !(ofl->ofl_flags & FLG_OF_DYNLIBS)) { ld_eprintf(ofl, ERR_FATAL, MSG_INTL(MSG_FIL_SOINSTAT), name); return (0); } /* * Record any additional shared object information. * If no soname is specified (eg. this file was * derived from a explicit filename declaration on the * command line, ie. bar.so) use the pathname. * This entry may be overridden if the files dynamic * section specifies an DT_SONAME value. */ if (soname == NULL) ifl->ifl_soname = ifl->ifl_name; else ifl->ifl_soname = soname; /* * If direct bindings, lazy loading, group permissions, * or deferred dependencies need to be established, mark * this object. */ if (ofl->ofl_flags1 & FLG_OF1_ZDIRECT) ifl->ifl_flags |= FLG_IF_DIRECT; if (ofl->ofl_flags1 & FLG_OF1_LAZYLD) ifl->ifl_flags |= FLG_IF_LAZYLD; if (ofl->ofl_flags1 & FLG_OF1_GRPPRM) ifl->ifl_flags |= FLG_IF_GRPPRM; if (ofl->ofl_flags1 & FLG_OF1_DEFERRED) ifl->ifl_flags |= (FLG_IF_LAZYLD | FLG_IF_DEFERRED); error = process_elf(ifl, elf, ofl); /* * Determine whether this dependency requires a syminfo. */ if (ifl->ifl_flags & MSK_IF_SYMINFO) ofl->ofl_flags |= FLG_OF_SYMINFO; /* * Guidance: Use -z lazyload/nolazyload. * libc is exempt from this advice, because it cannot * be lazy loaded, and requests to do so are ignored. */ if (OFL_GUIDANCE(ofl, FLG_OFG_NO_LAZY) && ((ifl->ifl_flags & FLG_IF_RTLDINF) == 0)) { ld_eprintf(ofl, ERR_GUIDANCE, MSG_INTL(MSG_GUIDE_LAZYLOAD)); ofl->ofl_guideflags |= FLG_OFG_NO_LAZY; } /* * Guidance: Use -B direct/nodirect or * -z direct/nodirect. */ if (OFL_GUIDANCE(ofl, FLG_OFG_NO_DB)) { ld_eprintf(ofl, ERR_GUIDANCE, MSG_INTL(MSG_GUIDE_DIRECT)); ofl->ofl_guideflags |= FLG_OFG_NO_DB; } break; default: (void) elf_errno(); _rej.rej_type = SGS_REJ_UNKFILE; _rej.rej_name = name; DBG_CALL(Dbg_file_rejected(ofl->ofl_lml, &_rej, ld_targ.t_m.m_mach)); if (rej->rej_type == 0) { *rej = _rej; rej->rej_name = strdup(_rej.rej_name); } return (0); } break; default: (void) elf_errno(); _rej.rej_type = SGS_REJ_UNKFILE; _rej.rej_name = name; DBG_CALL(Dbg_file_rejected(ofl->ofl_lml, &_rej, ld_targ.t_m.m_mach)); if (rej->rej_type == 0) { *rej = _rej; rej->rej_name = strdup(_rej.rej_name); } return (0); } if ((error == 0) || (error == S_ERROR)) return (error); if (ifl_ret) *ifl_ret = ifl; return (1); } /* * Having successfully opened a file, set up the necessary elf structures to * process it further. This small section of processing is slightly different * from the elf initialization required to process a relocatable object from an * archive (see libs.c: ld_process_archive()). */ uintptr_t ld_process_open(const char *opath, const char *ofile, int *fd, Ofl_desc *ofl, Word flags, Rej_desc *rej, Ifl_desc **ifl_ret) { Elf *elf; const char *npath = opath; const char *nfile = ofile; if ((elf = elf_begin(*fd, ELF_C_READ, NULL)) == NULL) { ld_eprintf(ofl, ERR_ELF, MSG_INTL(MSG_ELF_BEGIN), npath); return (0); } /* * Determine whether the support library wishes to process this open. * The support library may return: * . a different ELF descriptor (in which case they should have * closed the original) * . a different file descriptor (in which case they should have * closed the original) * . a different path and file name (presumably associated with * a different file descriptor) * * A file descriptor of -1, or and ELF descriptor of zero indicates * the file should be ignored. */ ld_sup_open(ofl, &npath, &nfile, fd, flags, &elf, NULL, 0, elf_kind(elf)); if ((*fd == -1) || (elf == NULL)) return (0); return (ld_process_ifl(npath, nfile, *fd, elf, flags, ofl, rej, ifl_ret)); } /* * Having successfully mapped a file, set up the necessary elf structures to * process it further. This routine is patterned after ld_process_open() and * is only called by ld.so.1(1) to process a relocatable object. */ Ifl_desc * ld_process_mem(const char *path, const char *file, char *addr, size_t size, Ofl_desc *ofl, Rej_desc *rej) { Elf *elf; uintptr_t open_ret; Ifl_desc *ifl; if ((elf = elf_memory(addr, size)) == NULL) { ld_eprintf(ofl, ERR_ELF, MSG_INTL(MSG_ELF_MEMORY), path); return (0); } open_ret = ld_process_ifl(path, file, 0, elf, 0, ofl, rej, &ifl); if (open_ret != 1) return ((Ifl_desc *) open_ret); return (ifl); } /* * Process a required library (i.e. the dependency of a shared object). * Combine the directory and filename, check the resultant path size, and try * opening the pathname. */ static Ifl_desc * process_req_lib(Sdf_desc *sdf, const char *dir, const char *file, Ofl_desc *ofl, Rej_desc *rej) { size_t dlen, plen; int fd; char path[PATH_MAX]; const char *_dir = dir; /* * Determine the sizes of the directory and filename to insure we don't * exceed our buffer. */ if ((dlen = strlen(dir)) == 0) { _dir = MSG_ORIG(MSG_STR_DOT); dlen = 1; } dlen++; plen = dlen + strlen(file) + 1; if (plen > PATH_MAX) { ld_eprintf(ofl, ERR_FATAL, MSG_INTL(MSG_FIL_PTHTOLONG), _dir, file); return (0); } /* * Build the entire pathname and try and open the file. */ (void) strcpy(path, _dir); (void) strcat(path, MSG_ORIG(MSG_STR_SLASH)); (void) strcat(path, file); DBG_CALL(Dbg_libs_req(ofl->ofl_lml, sdf->sdf_name, sdf->sdf_rfile, path)); if ((fd = open(path, O_RDONLY)) == -1) return (0); else { uintptr_t open_ret; Ifl_desc *ifl; char *_path; if ((_path = libld_malloc(strlen(path) + 1)) == NULL) return ((Ifl_desc *)S_ERROR); (void) strcpy(_path, path); open_ret = ld_process_open(_path, &_path[dlen], &fd, ofl, 0, rej, &ifl); if (fd != -1) (void) close(fd); if (open_ret != 1) return ((Ifl_desc *)open_ret); return (ifl); } } /* * Finish any library processing. Walk the list of so's that have been listed * as "included" by shared objects we have previously processed. Examine them, * without adding them as explicit dependents of this program, in order to * complete our symbol definition process. The search path rules are: * * - use any user supplied paths, i.e. LD_LIBRARY_PATH and -L, then * * - use any RPATH defined within the parent shared object, then * * - use the default directories, i.e. LIBPATH or -YP. */ uintptr_t ld_finish_libs(Ofl_desc *ofl) { Aliste idx1; Sdf_desc *sdf; Rej_desc rej = { 0 }; /* * Make sure we are back in dynamic mode. */ ofl->ofl_flags |= FLG_OF_DYNLIBS; for (APLIST_TRAVERSE(ofl->ofl_soneed, idx1, sdf)) { Aliste idx2; char *path, *slash = NULL; int fd; Ifl_desc *ifl; char *file = (char *)sdf->sdf_name; /* * See if this file has already been processed. At the time * this implicit dependency was determined there may still have * been more explicit dependencies to process. Note, if we ever * do parse the command line three times we would be able to * do all this checking when processing the dynamic section. */ if (sdf->sdf_file) continue; for (APLIST_TRAVERSE(ofl->ofl_sos, idx2, ifl)) { if (!(ifl->ifl_flags & FLG_IF_NEEDSTR) && (strcmp(file, ifl->ifl_soname) == 0)) { sdf->sdf_file = ifl; break; } } if (sdf->sdf_file) continue; /* * If the current path name element embeds a "/", then it's to * be taken "as is", with no searching involved. Process all * "/" occurrences, so that we can deduce the base file name. */ for (path = file; *path; path++) { if (*path == '/') slash = path; } if (slash) { DBG_CALL(Dbg_libs_req(ofl->ofl_lml, sdf->sdf_name, sdf->sdf_rfile, file)); if ((fd = open(file, O_RDONLY)) == -1) { ld_eprintf(ofl, ERR_WARNING, MSG_INTL(MSG_FIL_NOTFOUND), file, sdf->sdf_rfile); } else { uintptr_t open_ret; Rej_desc _rej = { 0 }; open_ret = ld_process_open(file, ++slash, &fd, ofl, 0, &_rej, &ifl); if (fd != -1) (void) close(fd); if (open_ret == S_ERROR) return (S_ERROR); if (_rej.rej_type) { Conv_reject_desc_buf_t rej_buf; ld_eprintf(ofl, ERR_WARNING, MSG_INTL(reject[_rej.rej_type]), _rej.rej_name ? rej.rej_name : MSG_INTL(MSG_STR_UNKNOWN), conv_reject_desc(&_rej, &rej_buf, ld_targ.t_m.m_mach)); } else sdf->sdf_file = ifl; } continue; } /* * Now search for this file in any user defined directories. */ for (APLIST_TRAVERSE(ofl->ofl_ulibdirs, idx2, path)) { Rej_desc _rej = { 0 }; ifl = process_req_lib(sdf, path, file, ofl, &_rej); if (ifl == (Ifl_desc *)S_ERROR) { return (S_ERROR); } if (_rej.rej_type) { if (rej.rej_type == 0) { rej = _rej; rej.rej_name = strdup(_rej.rej_name); } } if (ifl) { sdf->sdf_file = ifl; break; } } if (sdf->sdf_file) continue; /* * Next use the local rules defined within the parent shared * object. */ if (sdf->sdf_rpath != NULL) { char *rpath, *next; rpath = libld_malloc(strlen(sdf->sdf_rpath) + 1); if (rpath == NULL) return (S_ERROR); (void) strcpy(rpath, sdf->sdf_rpath); DBG_CALL(Dbg_libs_path(ofl->ofl_lml, rpath, LA_SER_RUNPATH, sdf->sdf_rfile)); if ((path = strtok_r(rpath, MSG_ORIG(MSG_STR_COLON), &next)) != NULL) { do { Rej_desc _rej = { 0 }; path = expand(sdf->sdf_rfile, path, &next); ifl = process_req_lib(sdf, path, file, ofl, &_rej); if (ifl == (Ifl_desc *)S_ERROR) { return (S_ERROR); } if ((_rej.rej_type) && (rej.rej_type == 0)) { rej = _rej; rej.rej_name = strdup(_rej.rej_name); } if (ifl) { sdf->sdf_file = ifl; break; } } while ((path = strtok_r(NULL, MSG_ORIG(MSG_STR_COLON), &next)) != NULL); } } if (sdf->sdf_file) continue; /* * Finally try the default library search directories. */ for (APLIST_TRAVERSE(ofl->ofl_dlibdirs, idx2, path)) { Rej_desc _rej = { 0 }; ifl = process_req_lib(sdf, path, file, ofl, &rej); if (ifl == (Ifl_desc *)S_ERROR) { return (S_ERROR); } if (_rej.rej_type) { if (rej.rej_type == 0) { rej = _rej; rej.rej_name = strdup(_rej.rej_name); } } if (ifl) { sdf->sdf_file = ifl; break; } } if (sdf->sdf_file) continue; /* * If we've got this far we haven't found the shared object. * If an object was found, but was rejected for some reason, * print a diagnostic to that effect, otherwise generate a * generic "not found" diagnostic. */ if (rej.rej_type) { Conv_reject_desc_buf_t rej_buf; ld_eprintf(ofl, ERR_WARNING, MSG_INTL(reject[rej.rej_type]), rej.rej_name ? rej.rej_name : MSG_INTL(MSG_STR_UNKNOWN), conv_reject_desc(&rej, &rej_buf, ld_targ.t_m.m_mach)); } else { ld_eprintf(ofl, ERR_WARNING, MSG_INTL(MSG_FIL_NOTFOUND), file, sdf->sdf_rfile); } } /* * Finally, now that all objects have been input, make sure any version * requirements have been met. */ return (ld_vers_verify(ofl)); }