symbol.c revision 1.33 1 /* $NetBSD: symbol.c,v 1.33 2003/08/12 09:18:50 skrll Exp $ */
2
3 /*
4 * Copyright 1996 John D. Polstra.
5 * Copyright 1996 Matt Thomas <matt (at) 3am-software.com>
6 * Copyright 2002 Charles M. Hannum <root (at) ihack.net>
7 * All rights reserved.
8 *
9 * Redistribution and use in source and binary forms, with or without
10 * modification, are permitted provided that the following conditions
11 * are met:
12 * 1. Redistributions of source code must retain the above copyright
13 * notice, this list of conditions and the following disclaimer.
14 * 2. Redistributions in binary form must reproduce the above copyright
15 * notice, this list of conditions and the following disclaimer in the
16 * documentation and/or other materials provided with the distribution.
17 * 3. All advertising materials mentioning features or use of this software
18 * must display the following acknowledgement:
19 * This product includes software developed by John Polstra.
20 * 4. The name of the author may not be used to endorse or promote products
21 * derived from this software without specific prior written permission.
22 *
23 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
24 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
25 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
26 * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
27 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
28 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
29 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
30 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
31 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
32 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
33 */
34
35 /*
36 * Dynamic linker for ELF.
37 *
38 * John Polstra <jdp (at) polstra.com>.
39 */
40
41 #include <err.h>
42 #include <errno.h>
43 #include <fcntl.h>
44 #include <stdarg.h>
45 #include <stdio.h>
46 #include <stdlib.h>
47 #include <string.h>
48 #include <unistd.h>
49 #include <sys/types.h>
50 #include <sys/mman.h>
51 #include <dirent.h>
52
53 #include "debug.h"
54 #include "rtld.h"
55
56 static bool
57 _rtld_is_exported(const Elf_Sym *def)
58 {
59 static Elf_Addr _rtld_exports[] = {
60 (Elf_Addr)dlopen,
61 (Elf_Addr)dlclose,
62 (Elf_Addr)dlsym,
63 (Elf_Addr)dlerror,
64 (Elf_Addr)dladdr,
65 NULL
66 };
67 int i;
68
69 Elf_Addr value;
70 value = (Elf_Addr)(_rtld_objself.relocbase + def->st_value);
71
72 for (i = 0; _rtld_exports[i] != NULL; i++) {
73 if (value == _rtld_exports[i])
74 return true;
75 }
76 return false;
77 }
78
79 /*
80 * Hash function for symbol table lookup. Don't even think about changing
81 * this. It is specified by the System V ABI.
82 */
83 unsigned long
84 _rtld_elf_hash(const char *name)
85 {
86 const unsigned char *p = (const unsigned char *) name;
87 unsigned long h = 0;
88 unsigned long g;
89 unsigned long c;
90
91 for (; __predict_true((c = *p) != '\0'); p++) {
92 h <<= 4;
93 h += c;
94 if ((g = h & 0xf0000000) != 0) {
95 h ^= g;
96 h ^= g >> 24;
97 }
98 }
99 return (h);
100 }
101
102 const Elf_Sym *
103 _rtld_symlook_list(const char *name, unsigned long hash, const Objlist *objlist,
104 const Obj_Entry **defobj_out, bool in_plt)
105 {
106 const Elf_Sym *symp;
107 const Elf_Sym *def;
108 const Obj_Entry *defobj;
109 const Objlist_Entry *elm;
110
111 def = NULL;
112 defobj = NULL;
113 SIMPLEQ_FOREACH(elm, objlist, link) {
114 rdbg(("search object %p (%s)", elm->obj, elm->obj->path));
115 if ((symp = _rtld_symlook_obj(name, hash, elm->obj, in_plt))
116 != NULL) {
117 if ((def == NULL) ||
118 (ELF_ST_BIND(symp->st_info) != STB_WEAK)) {
119 def = symp;
120 defobj = elm->obj;
121 if (ELF_ST_BIND(def->st_info) != STB_WEAK)
122 break;
123 }
124 }
125 }
126 if (def != NULL)
127 *defobj_out = defobj;
128 return def;
129 }
130
131 /*
132 * Search the symbol table of a single shared object for a symbol of
133 * the given name. Returns a pointer to the symbol, or NULL if no
134 * definition was found.
135 *
136 * The symbol's hash value is passed in for efficiency reasons; that
137 * eliminates many recomputations of the hash value.
138 */
139 const Elf_Sym *
140 _rtld_symlook_obj(const char *name, unsigned long hash,
141 const Obj_Entry *obj, bool in_plt)
142 {
143 unsigned long symnum;
144
145 for (symnum = obj->buckets[hash % obj->nbuckets];
146 symnum != ELF_SYM_UNDEFINED;
147 symnum = obj->chains[symnum]) {
148 const Elf_Sym *symp;
149 const char *strp;
150
151 assert(symnum < obj->nchains);
152 symp = obj->symtab + symnum;
153 strp = obj->strtab + symp->st_name;
154 rdbg(("check %s vs %s in %p", name, strp, obj));
155 if (name[1] == strp[1] && !strcmp(name, strp)) {
156 if (symp->st_shndx != SHN_UNDEF)
157 return symp;
158 #ifndef __mips__
159 /*
160 * XXX DANGER WILL ROBINSON!
161 * If we have a function pointer in the executable's
162 * data section, it points to the executable's PLT
163 * slot, and there is NO relocation emitted. To make
164 * the function pointer comparable to function pointers
165 * in shared libraries, we must resolve data references
166 * in the libraries to point to PLT slots in the
167 * executable, if they exist.
168 */
169 else if (!in_plt && symp->st_value != 0 &&
170 ELF_ST_TYPE(symp->st_info) == STT_FUNC)
171 return symp;
172 #endif
173 else
174 return NULL;
175 }
176 }
177
178 return NULL;
179 }
180
181 /*
182 * Given a symbol number in a referencing object, find the corresponding
183 * definition of the symbol. Returns a pointer to the symbol, or NULL if
184 * no definition was found. Returns a pointer to the Obj_Entry of the
185 * defining object via the reference parameter DEFOBJ_OUT.
186 */
187 const Elf_Sym *
188 _rtld_find_symdef(unsigned long symnum, const Obj_Entry *refobj,
189 const Obj_Entry **defobj_out, bool in_plt)
190 {
191 const Elf_Sym *ref;
192 const Elf_Sym *def;
193 const Elf_Sym *symp;
194 const Obj_Entry *obj;
195 const Obj_Entry *defobj;
196 const Objlist_Entry *elm;
197 const char *name;
198 unsigned long hash;
199
200 ref = refobj->symtab + symnum;
201 name = refobj->strtab + ref->st_name;
202
203 hash = _rtld_elf_hash(name);
204 def = NULL;
205 defobj = NULL;
206
207 /* Look first in the referencing object if linked symbolically */
208 if (refobj->symbolic) {
209 symp = _rtld_symlook_obj(name, hash, refobj, in_plt);
210 if (symp != NULL) {
211 def = symp;
212 defobj = refobj;
213 }
214 }
215
216 /* Search all objects loaded at program start up. */
217 if (def == NULL || ELF_ST_BIND(def->st_info) == STB_WEAK) {
218 rdbg(("search _rtld_list_main"));
219 symp = _rtld_symlook_list(name, hash, &_rtld_list_main, &obj, in_plt);
220 if (symp != NULL &&
221 (def == NULL || ELF_ST_BIND(symp->st_info) != STB_WEAK)) {
222 def = symp;
223 defobj = obj;
224 }
225 }
226
227 /* Search all dlopened DAGs containing the referencing object. */
228 SIMPLEQ_FOREACH(elm, &refobj->dldags, link) {
229 if (def != NULL && ELF_ST_BIND(def->st_info) != STB_WEAK)
230 break;
231 rdbg(("search DAG with root %p (%s)", elm->obj, elm->obj->path));
232 symp = _rtld_symlook_list(name, hash, &elm->obj->dagmembers, &obj, in_plt);
233 if (symp != NULL &&
234 (def == NULL || ELF_ST_BIND(symp->st_info) != STB_WEAK)) {
235 def = symp;
236 defobj = obj;
237 }
238 }
239
240 /* Search all RTLD_GLOBAL objects. */
241 if (def == NULL || ELF_ST_BIND(def->st_info) == STB_WEAK) {
242 rdbg(("search _rtld_list_global"));
243 symp = _rtld_symlook_list(name, hash, &_rtld_list_global, &obj, in_plt);
244 if (symp != NULL &&
245 (def == NULL || ELF_ST_BIND(symp->st_info) != STB_WEAK)) {
246 def = symp;
247 defobj = obj;
248 }
249 }
250
251 /*
252 * Search the dynamic linker itself, and possibly resolve the
253 * symbol from there. This is how the application links to
254 * dynamic linker services such as dlopen. Only the values listed
255 * in the "_rtld_exports" array can be resolved from the dynamic linker.
256 */
257 if (def == NULL || ELF_ST_BIND(def->st_info) == STB_WEAK) {
258 symp = _rtld_symlook_obj(name, hash, &_rtld_objself, in_plt);
259 if (symp != NULL && _rtld_is_exported(symp)) {
260 def = symp;
261 defobj = &_rtld_objself;
262 }
263 }
264
265 /*
266 * If we found no definition and the reference is weak, treat the
267 * symbol as having the value zero.
268 */
269 if (def == NULL && ELF_ST_BIND(ref->st_info) == STB_WEAK) {
270 rdbg((" returning _rtld_sym_zero@_rtld_objmain"));
271 def = &_rtld_sym_zero;
272 defobj = _rtld_objmain;
273 }
274
275 if (def != NULL)
276 *defobj_out = defobj;
277 else {
278 rdbg(("lookup failed"));
279 _rtld_error("%s: Undefined %ssymbol \"%s\" (symnum = %ld)",
280 refobj->path, in_plt ? "PLT " : "", name, symnum);
281 }
282 return def;
283 }
284
285 /*
286 * Given a symbol name in a referencing object, find the corresponding
287 * definition of the symbol. Returns a pointer to the symbol, or NULL if
288 * no definition was found. Returns a pointer to the Obj_Entry of the
289 * defining object via the reference parameter DEFOBJ_OUT.
290 */
291 const Elf_Sym *
292 _rtld_symlook_default(const char *name, unsigned long hash,
293 const Obj_Entry *refobj, const Obj_Entry **defobj_out, bool in_plt)
294 {
295 const Elf_Sym *def;
296 const Elf_Sym *symp;
297 const Obj_Entry *obj;
298 const Obj_Entry *defobj;
299 const Objlist_Entry *elm;
300 def = NULL;
301 defobj = NULL;
302
303 /* Look first in the referencing object if linked symbolically. */
304 if (refobj->symbolic) {
305 symp = _rtld_symlook_obj(name, hash, refobj, in_plt);
306 if (symp != NULL) {
307 def = symp;
308 defobj = refobj;
309 }
310 }
311
312 /* Search all objects loaded at program start up. */
313 if (def == NULL || ELF_ST_BIND(def->st_info) == STB_WEAK) {
314 symp = _rtld_symlook_list(name, hash, &_rtld_list_main, &obj, in_plt);
315 if (symp != NULL &&
316 (def == NULL || ELF_ST_BIND(symp->st_info) != STB_WEAK)) {
317 def = symp;
318 defobj = obj;
319 }
320 }
321
322 /* Search all dlopened DAGs containing the referencing object. */
323 SIMPLEQ_FOREACH(elm, &refobj->dldags, link) {
324 if (def != NULL && ELF_ST_BIND(def->st_info) != STB_WEAK)
325 break;
326 symp = _rtld_symlook_list(name, hash, &elm->obj->dagmembers, &obj,
327 in_plt);
328 if (symp != NULL &&
329 (def == NULL || ELF_ST_BIND(symp->st_info) != STB_WEAK)) {
330 def = symp;
331 defobj = obj;
332 }
333 }
334
335 /* Search all DAGs whose roots are RTLD_GLOBAL objects. */
336 SIMPLEQ_FOREACH(elm, &_rtld_list_global, link) {
337 if (def != NULL && ELF_ST_BIND(def->st_info) != STB_WEAK)
338 break;
339 symp = _rtld_symlook_list(name, hash, &elm->obj->dagmembers, &obj,
340 in_plt);
341 if (symp != NULL &&
342 (def == NULL || ELF_ST_BIND(symp->st_info) != STB_WEAK)) {
343 def = symp;
344 defobj = obj;
345 }
346 }
347
348 #ifdef notyet
349 /*
350 * Search the dynamic linker itself, and possibly resolve the
351 * symbol from there. This is how the application links to
352 * dynamic linker services such as dlopen. Only the values listed
353 * in the "exports" array can be resolved from the dynamic linker.
354 */
355 if (def == NULL || ELF_ST_BIND(def->st_info) == STB_WEAK) {
356 symp = _rtld_symlook_obj(name, hash, &_rtld_objself, in_plt);
357 if (symp != NULL && is_exported(symp)) {
358 def = symp;
359 defobj = &_rtld_objself;
360 }
361 }
362 #endif
363
364 if (def != NULL)
365 *defobj_out = defobj;
366 return def;
367 }
368